From f41c7cab8727d99a2376450d233f128f23060b98 Mon Sep 17 00:00:00 2001
From: Kalle Valo <quic_kvalo@quicinc.com>
Date: Fri, 3 May 2024 13:34:36 +0300
Subject: wifi: ath11k: ath11k_mac_op_ipv6_changed(): use list_for_each_entry()

Simplify the loop by using list_for_each_entry(). No functional changes.

Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.37

Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://msgid.link/20240411165516.4070649-3-kvalo@kernel.org
---
 drivers/net/wireless/ath/ath11k/mac.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 4f62e38ba48b..a71195344288 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -9104,7 +9104,6 @@ static void ath11k_mac_op_ipv6_changed(struct ieee80211_hw *hw,
 	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
 	struct inet6_ifaddr *ifa6;
 	struct ifacaddr6 *ifaca6;
-	struct list_head *p;
 	u32 count, scope;
 
 	ath11k_dbg(ar->ab, ATH11K_DBG_MAC, "op ipv6 changed\n");
@@ -9120,11 +9119,10 @@ static void ath11k_mac_op_ipv6_changed(struct ieee80211_hw *hw,
 	memcpy(offload->mac_addr, vif->addr, ETH_ALEN);
 
 	/* get unicast address */
-	list_for_each(p, &idev->addr_list) {
+	list_for_each_entry(ifa6, &idev->addr_list, if_list) {
 		if (count >= ATH11K_IPV6_MAX_COUNT)
 			goto generate;
 
-		ifa6 = list_entry(p, struct inet6_ifaddr, if_list);
 		if (ifa6->flags & IFA_F_DADFAILED)
 			continue;
 		scope = ipv6_addr_src_scope(&ifa6->addr);
-- 
cgit v1.2.3-58-ga151


From ecd509b6f263e4f0d86876320c36260dc28bb53b Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Fri, 3 May 2024 13:34:36 +0300
Subject: wifi: ath12k: Refactor the hardware recovery procedure

Currently, in multi-wiphy models, the recovery handler access mac80211
HW from the radio/link structure. This will be incorrect for single wiphy
model, as they will hold multiple link/radio structures. To fix this,
access mac80211 HW based on the number of hardware in the SoC/chip. This
approach makes the recovery handler compatible with both multi wiphy and
single wiphy models.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240425090307.3233434-2-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.c | 119 ++++++++++++++++++---------------
 drivers/net/wireless/ath/ath12k/mac.c  |  23 +++++--
 2 files changed, 80 insertions(+), 62 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c
index 6663f4e1792d..ae14614c3568 100644
--- a/drivers/net/wireless/ath/ath12k/core.c
+++ b/drivers/net/wireless/ath/ath12k/core.c
@@ -994,9 +994,8 @@ void ath12k_core_halt(struct ath12k *ar)
 static void ath12k_core_pre_reconfigure_recovery(struct ath12k_base *ab)
 {
 	struct ath12k *ar;
-	struct ath12k_pdev *pdev;
 	struct ath12k_hw *ah;
-	int i;
+	int i, j;
 
 	spin_lock_bh(&ab->base_lock);
 	ab->stats.fw_crash_counter++;
@@ -1006,35 +1005,34 @@ static void ath12k_core_pre_reconfigure_recovery(struct ath12k_base *ab)
 		set_bit(ATH12K_FLAG_CRASH_FLUSH, &ab->dev_flags);
 
 	for (i = 0; i < ab->num_hw; i++) {
-		if (!ab->ah[i])
+		ah = ab->ah[i];
+		if (!ah)
 			continue;
 
-		ah = ab->ah[i];
 		ieee80211_stop_queues(ah->hw);
-	}
 
-	for (i = 0; i < ab->num_radios; i++) {
-		pdev = &ab->pdevs[i];
-		ar = pdev->ar;
-		if (!ar || ar->state == ATH12K_STATE_OFF)
-			continue;
+		for (j = 0; j < ah->num_radio; j++) {
+			ar = &ah->radio[j];
+			if (ar->state == ATH12K_STATE_OFF)
+				continue;
 
-		ath12k_mac_drain_tx(ar);
-		complete(&ar->scan.started);
-		complete(&ar->scan.completed);
-		complete(&ar->scan.on_channel);
-		complete(&ar->peer_assoc_done);
-		complete(&ar->peer_delete_done);
-		complete(&ar->install_key_done);
-		complete(&ar->vdev_setup_done);
-		complete(&ar->vdev_delete_done);
-		complete(&ar->bss_survey_done);
-
-		wake_up(&ar->dp.tx_empty_waitq);
-		idr_for_each(&ar->txmgmt_idr,
-			     ath12k_mac_tx_mgmt_pending_free, ar);
-		idr_destroy(&ar->txmgmt_idr);
-		wake_up(&ar->txmgmt_empty_waitq);
+			ath12k_mac_drain_tx(ar);
+			complete(&ar->scan.started);
+			complete(&ar->scan.completed);
+			complete(&ar->scan.on_channel);
+			complete(&ar->peer_assoc_done);
+			complete(&ar->peer_delete_done);
+			complete(&ar->install_key_done);
+			complete(&ar->vdev_setup_done);
+			complete(&ar->vdev_delete_done);
+			complete(&ar->bss_survey_done);
+
+			wake_up(&ar->dp.tx_empty_waitq);
+			idr_for_each(&ar->txmgmt_idr,
+				     ath12k_mac_tx_mgmt_pending_free, ar);
+			idr_destroy(&ar->txmgmt_idr);
+			wake_up(&ar->txmgmt_empty_waitq);
+		}
 	}
 
 	wake_up(&ab->wmi_ab.tx_credits_wq);
@@ -1043,41 +1041,52 @@ static void ath12k_core_pre_reconfigure_recovery(struct ath12k_base *ab)
 
 static void ath12k_core_post_reconfigure_recovery(struct ath12k_base *ab)
 {
+	struct ath12k_hw *ah;
 	struct ath12k *ar;
-	struct ath12k_pdev *pdev;
-	int i;
+	int i, j;
+	u8 restart_count;
 
-	for (i = 0; i < ab->num_radios; i++) {
-		pdev = &ab->pdevs[i];
-		ar = pdev->ar;
-		if (!ar || ar->state == ATH12K_STATE_OFF)
+	for (i = 0; i < ab->num_hw; i++) {
+		ah = ab->ah[i];
+		if (!ah)
 			continue;
 
-		mutex_lock(&ar->conf_mutex);
-
-		switch (ar->state) {
-		case ATH12K_STATE_ON:
-			ar->state = ATH12K_STATE_RESTARTING;
-			ath12k_core_halt(ar);
-			ieee80211_restart_hw(ath12k_ar_to_hw(ar));
-			break;
-		case ATH12K_STATE_OFF:
-			ath12k_warn(ab,
-				    "cannot restart radio %d that hasn't been started\n",
-				    i);
-			break;
-		case ATH12K_STATE_RESTARTING:
-			break;
-		case ATH12K_STATE_RESTARTED:
-			ar->state = ATH12K_STATE_WEDGED;
-			fallthrough;
-		case ATH12K_STATE_WEDGED:
-			ath12k_warn(ab,
-				    "device is wedged, will not restart radio %d\n", i);
-			break;
+		for (j = 0, restart_count = 0; j < ah->num_radio; j++) {
+			ar = &ah->radio[j];
+			if (ar->state == ATH12K_STATE_OFF)
+				continue;
+
+			mutex_lock(&ar->conf_mutex);
+
+			switch (ar->state) {
+			case ATH12K_STATE_ON:
+				ar->state = ATH12K_STATE_RESTARTING;
+				ath12k_core_halt(ar);
+				restart_count++;
+				break;
+			case ATH12K_STATE_OFF:
+				ath12k_warn(ab,
+					    "cannot restart radio %d that hasn't been started\n",
+					    j);
+				break;
+			case ATH12K_STATE_RESTARTING:
+				break;
+			case ATH12K_STATE_RESTARTED:
+				ar->state = ATH12K_STATE_WEDGED;
+				fallthrough;
+			case ATH12K_STATE_WEDGED:
+				ath12k_warn(ab,
+					    "device is wedged, will not restart radio %d\n", j);
+				break;
+			}
+			mutex_unlock(&ar->conf_mutex);
 		}
-		mutex_unlock(&ar->conf_mutex);
+
+		/* Restart after all the link/radio got restart */
+		if (restart_count == ah->num_radio)
+			ieee80211_restart_hw(ah->hw);
 	}
+
 	complete(&ab->driver_recovery);
 }
 
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 805cb084484a..6d19b6003aea 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -7920,26 +7920,33 @@ ath12k_mac_op_reconfig_complete(struct ieee80211_hw *hw,
 	struct ath12k *ar;
 	struct ath12k_base *ab;
 	struct ath12k_vif *arvif;
-	int recovery_count;
+	int recovery_count, i;
 
 	if (reconfig_type != IEEE80211_RECONFIG_TYPE_RESTART)
 		return;
 
-	ar = ath12k_ah_to_ar(ah, 0);
-	ab = ar->ab;
+	for_each_ar(ah, ar, i) {
+		mutex_lock(&ar->conf_mutex);
 
-	mutex_lock(&ar->conf_mutex);
+		if (ar->state != ATH12K_STATE_RESTARTED) {
+			mutex_unlock(&ar->conf_mutex);
+			continue;
+		}
+
+		ab = ar->ab;
 
-	if (ar->state == ATH12K_STATE_RESTARTED) {
 		ath12k_warn(ar->ab, "pdev %d successfully recovered\n",
 			    ar->pdev->pdev_id);
+
 		ar->state = ATH12K_STATE_ON;
 		ieee80211_wake_queues(hw);
 
 		if (ab->is_reset) {
 			recovery_count = atomic_inc_return(&ab->recovery_count);
+
 			ath12k_dbg(ab, ATH12K_DBG_BOOT, "recovery count %d\n",
 				   recovery_count);
+
 			/* When there are multiple radios in an SOC,
 			 * the recovery has to be done for each radio
 			 */
@@ -7958,6 +7965,7 @@ ath12k_mac_op_reconfig_complete(struct ieee80211_hw *hw,
 				   arvif->key_cipher,
 				   arvif->is_up,
 				   arvif->vdev_type);
+
 			/* After trigger disconnect, then upper layer will
 			 * trigger connect again, then the PN number of
 			 * upper layer will be reset to keep up with AP
@@ -7967,13 +7975,14 @@ ath12k_mac_op_reconfig_complete(struct ieee80211_hw *hw,
 			    arvif->vdev_type == WMI_VDEV_TYPE_STA &&
 			    arvif->vdev_subtype == WMI_VDEV_SUBTYPE_NONE) {
 				ieee80211_hw_restart_disconnect(arvif->vif);
+
 				ath12k_dbg(ab, ATH12K_DBG_BOOT,
 					   "restart disconnect\n");
 			}
 		}
-	}
 
-	mutex_unlock(&ar->conf_mutex);
+		mutex_unlock(&ar->conf_mutex);
+	}
 }
 
 static void
-- 
cgit v1.2.3-58-ga151


From 9b4e5caaf5905edb7c184e8d14a9fab09f2f865b Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Fri, 3 May 2024 13:34:37 +0300
Subject: wifi: ath12k: Refactor the hardware state

Currently, in multi wiphy models, the mac80211 hardware state is maintained
within the radio/link structure. However, in single wiphy models, the
mac80211 hardware state is needed at the hardware abstraction layer
(ath12k_hw). Therefore, move the hardware state from the radio/link
structure to the hardware abstraction layer (ath12k_hw). Additionally,
update the naming convention of the state enums to enhance clarity and
consistency.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240425090307.3233434-3-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.c | 59 +++++++++++++++-----------------
 drivers/net/wireless/ath/ath12k/core.h | 22 +++++++-----
 drivers/net/wireless/ath/ath12k/mac.c  | 61 ++++++++++++++++++----------------
 drivers/net/wireless/ath/ath12k/reg.c  | 19 ++++++-----
 4 files changed, 83 insertions(+), 78 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c
index ae14614c3568..a685cfd6fd92 100644
--- a/drivers/net/wireless/ath/ath12k/core.c
+++ b/drivers/net/wireless/ath/ath12k/core.c
@@ -1006,15 +1006,13 @@ static void ath12k_core_pre_reconfigure_recovery(struct ath12k_base *ab)
 
 	for (i = 0; i < ab->num_hw; i++) {
 		ah = ab->ah[i];
-		if (!ah)
+		if (!ah || ah->state == ATH12K_HW_STATE_OFF)
 			continue;
 
 		ieee80211_stop_queues(ah->hw);
 
 		for (j = 0; j < ah->num_radio; j++) {
 			ar = &ah->radio[j];
-			if (ar->state == ATH12K_STATE_OFF)
-				continue;
 
 			ath12k_mac_drain_tx(ar);
 			complete(&ar->scan.started);
@@ -1044,47 +1042,42 @@ static void ath12k_core_post_reconfigure_recovery(struct ath12k_base *ab)
 	struct ath12k_hw *ah;
 	struct ath12k *ar;
 	int i, j;
-	u8 restart_count;
 
 	for (i = 0; i < ab->num_hw; i++) {
 		ah = ab->ah[i];
-		if (!ah)
+		if (!ah || ah->state == ATH12K_HW_STATE_OFF)
 			continue;
 
-		for (j = 0, restart_count = 0; j < ah->num_radio; j++) {
-			ar = &ah->radio[j];
-			if (ar->state == ATH12K_STATE_OFF)
-				continue;
+		switch (ah->state) {
+		case ATH12K_HW_STATE_ON:
+			ah->state = ATH12K_HW_STATE_RESTARTING;
 
-			mutex_lock(&ar->conf_mutex);
+			for (j = 0; j < ah->num_radio; j++) {
+				ar = &ah->radio[j];
 
-			switch (ar->state) {
-			case ATH12K_STATE_ON:
-				ar->state = ATH12K_STATE_RESTARTING;
+				mutex_lock(&ar->conf_mutex);
 				ath12k_core_halt(ar);
-				restart_count++;
-				break;
-			case ATH12K_STATE_OFF:
-				ath12k_warn(ab,
-					    "cannot restart radio %d that hasn't been started\n",
-					    j);
-				break;
-			case ATH12K_STATE_RESTARTING:
-				break;
-			case ATH12K_STATE_RESTARTED:
-				ar->state = ATH12K_STATE_WEDGED;
-				fallthrough;
-			case ATH12K_STATE_WEDGED:
-				ath12k_warn(ab,
-					    "device is wedged, will not restart radio %d\n", j);
-				break;
+				mutex_unlock(&ar->conf_mutex);
 			}
-			mutex_unlock(&ar->conf_mutex);
-		}
 
-		/* Restart after all the link/radio got restart */
-		if (restart_count == ah->num_radio)
+			/* Restart after all the link/radio halt */
 			ieee80211_restart_hw(ah->hw);
+			break;
+		case ATH12K_HW_STATE_OFF:
+			ath12k_warn(ab,
+				    "cannot restart hw %d that hasn't been started\n",
+				    i);
+			break;
+		case ATH12K_HW_STATE_RESTARTING:
+			break;
+		case ATH12K_HW_STATE_RESTARTED:
+			ah->state = ATH12K_HW_STATE_WEDGED;
+			fallthrough;
+		case ATH12K_HW_STATE_WEDGED:
+			ath12k_warn(ab,
+				    "device is wedged, will not restart hw %d\n", i);
+			break;
+		}
 	}
 
 	complete(&ab->driver_recovery);
diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index 47dde4401210..c4eb8b25398c 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -457,12 +457,12 @@ struct ath12k_sta {
 #define ATH12K_NUM_CHANS 100
 #define ATH12K_MAX_5G_CHAN 173
 
-enum ath12k_state {
-	ATH12K_STATE_OFF,
-	ATH12K_STATE_ON,
-	ATH12K_STATE_RESTARTING,
-	ATH12K_STATE_RESTARTED,
-	ATH12K_STATE_WEDGED,
+enum ath12k_hw_state {
+	ATH12K_HW_STATE_OFF,
+	ATH12K_HW_STATE_ON,
+	ATH12K_HW_STATE_RESTARTING,
+	ATH12K_HW_STATE_RESTARTED,
+	ATH12K_HW_STATE_WEDGED,
 	/* Add other states as required */
 };
 
@@ -511,7 +511,6 @@ struct ath12k {
 	u32 ht_cap_info;
 	u32 vht_cap_info;
 	struct ath12k_he ar_he;
-	enum ath12k_state state;
 	bool supports_6ghz;
 	struct {
 		struct completion started;
@@ -636,10 +635,12 @@ struct ath12k {
 
 struct ath12k_hw {
 	struct ieee80211_hw *hw;
+	struct ath12k_base *ab;
+	enum ath12k_hw_state state;
 	bool regd_updated;
 	bool use_6ghz_regd;
-
 	u8 num_radio;
+
 	struct ath12k radio[] __aligned(sizeof(void *));
 };
 
@@ -1037,6 +1038,11 @@ static inline struct ath12k *ath12k_ah_to_ar(struct ath12k_hw *ah, u8 hw_link_id
 	return &ah->radio[hw_link_id];
 }
 
+static inline struct ath12k_hw *ath12k_ar_to_ah(struct ath12k *ar)
+{
+	return ar->ah;
+}
+
 static inline struct ieee80211_hw *ath12k_ar_to_hw(struct ath12k *ar)
 {
 	return ar->ah->hw;
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 6d19b6003aea..0cb2e0435cc0 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -5269,6 +5269,7 @@ static void ath12k_mac_setup_sband_iftype_data(struct ath12k *ar,
 
 static int __ath12k_set_antenna(struct ath12k *ar, u32 tx_ant, u32 rx_ant)
 {
+	struct ath12k_hw *ah = ath12k_ar_to_ah(ar);
 	int ret;
 
 	lockdep_assert_held(&ar->conf_mutex);
@@ -5289,8 +5290,8 @@ static int __ath12k_set_antenna(struct ath12k *ar, u32 tx_ant, u32 rx_ant)
 	ar->cfg_tx_chainmask = tx_ant;
 	ar->cfg_rx_chainmask = rx_ant;
 
-	if (ar->state != ATH12K_STATE_ON &&
-	    ar->state != ATH12K_STATE_RESTARTED)
+	if (ah->state != ATH12K_HW_STATE_ON &&
+	    ah->state != ATH12K_HW_STATE_RESTARTED)
 		return 0;
 
 	ret = ath12k_wmi_pdev_set_param(ar, WMI_PDEV_PARAM_TX_CHAIN_MASK,
@@ -5620,22 +5621,6 @@ static int ath12k_mac_start(struct ath12k *ar)
 
 	mutex_lock(&ar->conf_mutex);
 
-	switch (ar->state) {
-	case ATH12K_STATE_OFF:
-		ar->state = ATH12K_STATE_ON;
-		break;
-	case ATH12K_STATE_RESTARTING:
-		ar->state = ATH12K_STATE_RESTARTED;
-		ath12k_mac_wait_reconfigure(ab);
-		break;
-	case ATH12K_STATE_RESTARTED:
-	case ATH12K_STATE_WEDGED:
-	case ATH12K_STATE_ON:
-		WARN_ON(1);
-		ret = -EINVAL;
-		goto err;
-	}
-
 	ret = ath12k_wmi_pdev_set_param(ar, WMI_PDEV_PARAM_PMF_QOS,
 					1, pdev->pdev_id);
 
@@ -5726,7 +5711,6 @@ static int ath12k_mac_start(struct ath12k *ar)
 
 	return 0;
 err:
-	ar->state = ATH12K_STATE_OFF;
 	mutex_unlock(&ar->conf_mutex);
 
 	return ret;
@@ -5749,9 +5733,28 @@ static int ath12k_mac_op_start(struct ieee80211_hw *hw)
 
 	ath12k_drain_tx(ah);
 
+	switch (ah->state) {
+	case ATH12K_HW_STATE_OFF:
+		ah->state = ATH12K_HW_STATE_ON;
+		break;
+	case ATH12K_HW_STATE_RESTARTING:
+		ah->state = ATH12K_HW_STATE_RESTARTED;
+		ath12k_mac_wait_reconfigure(ah->ab);
+		break;
+	case ATH12K_HW_STATE_RESTARTED:
+	case ATH12K_HW_STATE_WEDGED:
+	case ATH12K_HW_STATE_ON:
+		ah->state = ATH12K_HW_STATE_OFF;
+
+		WARN_ON(1);
+		return -EINVAL;
+	}
+
 	for_each_ar(ah, ar, i) {
 		ret = ath12k_mac_start(ar);
 		if (ret) {
+			ah->state = ATH12K_HW_STATE_OFF;
+
 			ath12k_err(ar->ab, "fail to start mac operations in pdev idx %d ret %d\n",
 				   ar->pdev_idx, ret);
 			goto fail_start;
@@ -5759,11 +5762,13 @@ static int ath12k_mac_op_start(struct ieee80211_hw *hw)
 	}
 
 	return 0;
+
 fail_start:
 	for (; i > 0; i--) {
 		ar = ath12k_ah_to_ar(ah, i - 1);
 		ath12k_mac_stop(ar);
 	}
+
 	return ret;
 }
 
@@ -5836,7 +5841,6 @@ static void ath12k_mac_stop(struct ath12k *ar)
 			   ret);
 
 	clear_bit(ATH12K_CAC_RUNNING, &ar->dev_flags);
-	ar->state = ATH12K_STATE_OFF;
 	mutex_unlock(&ar->conf_mutex);
 
 	cancel_delayed_work_sync(&ar->scan.timeout);
@@ -5865,6 +5869,8 @@ static void ath12k_mac_op_stop(struct ieee80211_hw *hw)
 
 	ath12k_drain_tx(ah);
 
+	ah->state = ATH12K_HW_STATE_OFF;
+
 	for_each_ar(ah, ar, i)
 		ath12k_mac_stop(ar);
 }
@@ -7925,22 +7931,20 @@ ath12k_mac_op_reconfig_complete(struct ieee80211_hw *hw,
 	if (reconfig_type != IEEE80211_RECONFIG_TYPE_RESTART)
 		return;
 
+	if (ah->state != ATH12K_HW_STATE_RESTARTED)
+		return;
+
+	ah->state = ATH12K_HW_STATE_ON;
+	ieee80211_wake_queues(hw);
+
 	for_each_ar(ah, ar, i) {
 		mutex_lock(&ar->conf_mutex);
 
-		if (ar->state != ATH12K_STATE_RESTARTED) {
-			mutex_unlock(&ar->conf_mutex);
-			continue;
-		}
-
 		ab = ar->ab;
 
 		ath12k_warn(ar->ab, "pdev %d successfully recovered\n",
 			    ar->pdev->pdev_id);
 
-		ar->state = ATH12K_STATE_ON;
-		ieee80211_wake_queues(hw);
-
 		if (ab->is_reset) {
 			recovery_count = atomic_inc_return(&ab->recovery_count);
 
@@ -8926,6 +8930,7 @@ static struct ath12k_hw *ath12k_mac_hw_allocate(struct ath12k_base *ab,
 
 	ah = ath12k_hw_to_ah(hw);
 	ah->hw = hw;
+	ah->ab = ab;
 	ah->num_radio = num_pdev_map;
 
 	for (i = 0; i < num_pdev_map; i++) {
diff --git a/drivers/net/wireless/ath/ath12k/reg.c b/drivers/net/wireless/ath/ath12k/reg.c
index fbf38044938c..439d61f284d8 100644
--- a/drivers/net/wireless/ath/ath12k/reg.c
+++ b/drivers/net/wireless/ath/ath12k/reg.c
@@ -206,9 +206,9 @@ static void ath12k_copy_regd(struct ieee80211_regdomain *regd_orig,
 
 int ath12k_regd_update(struct ath12k *ar, bool init)
 {
-	struct ieee80211_hw *hw = ath12k_ar_to_hw(ar);
+	struct ath12k_hw *ah = ath12k_ar_to_ah(ar);
+	struct ieee80211_hw *hw = ah->hw;
 	struct ieee80211_regdomain *regd, *regd_copy = NULL;
-	struct ath12k_hw *ah = ar->ah;
 	int ret, regd_len, pdev_id;
 	struct ath12k_base *ab;
 	int i;
@@ -286,19 +286,20 @@ int ath12k_regd_update(struct ath12k *ar, bool init)
 	if (ret)
 		goto err;
 
+	if (ah->state != ATH12K_HW_STATE_ON)
+		goto skip;
+
 	ah->regd_updated = true;
 	/* Apply the new regd to all the radios, this is expected to be received only once
 	 * since we check for ah->regd_updated and allow here only once.
 	 */
 	for_each_ar(ah, ar, i) {
-		if (ar->state == ATH12K_STATE_ON) {
-			ab = ar->ab;
-			ret = ath12k_reg_update_chan_list(ar);
-			if (ret)
-				goto err;
-		}
+		ab = ar->ab;
+		ret = ath12k_reg_update_chan_list(ar);
+		if (ret)
+			goto err;
 	}
-
+skip:
 	return 0;
 err:
 	ath12k_warn(ab, "failed to perform regd update : %d\n", ret);
-- 
cgit v1.2.3-58-ga151


From acaa84009fade2ae80922861aa9464e89c9d3ef0 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Fri, 3 May 2024 13:34:37 +0300
Subject: wifi: ath12k: Add lock to protect the hardware state

Currently, hardware state is not protected across the reconfigure
operations. However, in single wiphy models, multiple radio/links is
exposed as a MAC hardware (ieee80211_hw) through the driver hardware
abstraction (ath12k_hw) layer. In such scenario, we need to protect
hardware state across the multiple radio/link at the driver hardware
abstraction (ath12k_hw) layer. Therefore, introduce a new mutex in
the ath12k_hw layer.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240425090307.3233434-4-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.c |  4 ++++
 drivers/net/wireless/ath/ath12k/core.h |  6 ++++++
 drivers/net/wireless/ath/ath12k/mac.c  | 16 ++++++++++++++++
 3 files changed, 26 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c
index a685cfd6fd92..e9aabdb9341c 100644
--- a/drivers/net/wireless/ath/ath12k/core.c
+++ b/drivers/net/wireless/ath/ath12k/core.c
@@ -1048,6 +1048,8 @@ static void ath12k_core_post_reconfigure_recovery(struct ath12k_base *ab)
 		if (!ah || ah->state == ATH12K_HW_STATE_OFF)
 			continue;
 
+		mutex_lock(&ah->hw_mutex);
+
 		switch (ah->state) {
 		case ATH12K_HW_STATE_ON:
 			ah->state = ATH12K_HW_STATE_RESTARTING;
@@ -1078,6 +1080,8 @@ static void ath12k_core_post_reconfigure_recovery(struct ath12k_base *ab)
 				    "device is wedged, will not restart hw %d\n", i);
 			break;
 		}
+
+		mutex_unlock(&ah->hw_mutex);
 	}
 
 	complete(&ab->driver_recovery);
diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index c4eb8b25398c..d833361948b7 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -636,11 +636,17 @@ struct ath12k {
 struct ath12k_hw {
 	struct ieee80211_hw *hw;
 	struct ath12k_base *ab;
+
+	/* Protect the write operation of the hardware state ath12k_hw::state
+	 * between hardware start<=>reconfigure<=>stop transitions.
+	 */
+	struct mutex hw_mutex;
 	enum ath12k_hw_state state;
 	bool regd_updated;
 	bool use_6ghz_regd;
 	u8 num_radio;
 
+	/* Keep last */
 	struct ath12k radio[] __aligned(sizeof(void *));
 };
 
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 0cb2e0435cc0..4f07ac8c0287 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -5615,10 +5615,13 @@ static void ath12k_mac_wait_reconfigure(struct ath12k_base *ab)
 
 static int ath12k_mac_start(struct ath12k *ar)
 {
+	struct ath12k_hw *ah = ar->ah;
 	struct ath12k_base *ab = ar->ab;
 	struct ath12k_pdev *pdev = ar->pdev;
 	int ret;
 
+	lockdep_assert_held(&ah->hw_mutex);
+
 	mutex_lock(&ar->conf_mutex);
 
 	ret = ath12k_wmi_pdev_set_param(ar, WMI_PDEV_PARAM_PMF_QOS,
@@ -5733,6 +5736,8 @@ static int ath12k_mac_op_start(struct ieee80211_hw *hw)
 
 	ath12k_drain_tx(ah);
 
+	guard(mutex)(&ah->hw_mutex);
+
 	switch (ah->state) {
 	case ATH12K_HW_STATE_OFF:
 		ah->state = ATH12K_HW_STATE_ON;
@@ -5831,9 +5836,12 @@ int ath12k_mac_rfkill_enable_radio(struct ath12k *ar, bool enable)
 
 static void ath12k_mac_stop(struct ath12k *ar)
 {
+	struct ath12k_hw *ah = ar->ah;
 	struct htt_ppdu_stats_info *ppdu_stats, *tmp;
 	int ret;
 
+	lockdep_assert_held(&ah->hw_mutex);
+
 	mutex_lock(&ar->conf_mutex);
 	ret = ath12k_mac_config_mon_status_default(ar, false);
 	if (ret && (ret != -EOPNOTSUPP))
@@ -5869,10 +5877,14 @@ static void ath12k_mac_op_stop(struct ieee80211_hw *hw)
 
 	ath12k_drain_tx(ah);
 
+	mutex_lock(&ah->hw_mutex);
+
 	ah->state = ATH12K_HW_STATE_OFF;
 
 	for_each_ar(ah, ar, i)
 		ath12k_mac_stop(ar);
+
+	mutex_unlock(&ah->hw_mutex);
 }
 
 static u8
@@ -7931,6 +7943,8 @@ ath12k_mac_op_reconfig_complete(struct ieee80211_hw *hw,
 	if (reconfig_type != IEEE80211_RECONFIG_TYPE_RESTART)
 		return;
 
+	guard(mutex)(&ah->hw_mutex);
+
 	if (ah->state != ATH12K_HW_STATE_RESTARTED)
 		return;
 
@@ -8933,6 +8947,8 @@ static struct ath12k_hw *ath12k_mac_hw_allocate(struct ath12k_base *ab,
 	ah->ab = ab;
 	ah->num_radio = num_pdev_map;
 
+	mutex_init(&ah->hw_mutex);
+
 	for (i = 0; i < num_pdev_map; i++) {
 		ab = pdev_map[i].ab;
 		pdev_idx = pdev_map[i].pdev_idx;
-- 
cgit v1.2.3-58-ga151


From 8b9ea752a9d67b0b668b38e7efa34fffba8b5331 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Fri, 3 May 2024 13:34:37 +0300
Subject: wifi: ath11k: refactor CE remap & unmap

Currently the logic that handles hw_params->ce_remap is inline code,
both for doing the remap and the unmap. An upcoming change needs to do
the unmap in a second place, so refactor the unmap logic into a
separate function. And although it is only called from one place,
refactor the remap logic as well to have functional symmetry.

No functional changes, compile tested only.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240430-ce-unmap-v1-1-e468328f95d9@quicinc.com
---
 drivers/net/wireless/ath/ath11k/ahb.c | 52 +++++++++++++++++++++++------------
 1 file changed, 34 insertions(+), 18 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/ahb.c b/drivers/net/wireless/ath/ath11k/ahb.c
index ca0f17ddebba..88dfef71c2b5 100644
--- a/drivers/net/wireless/ath/ath11k/ahb.c
+++ b/drivers/net/wireless/ath/ath11k/ahb.c
@@ -954,6 +954,36 @@ static int ath11k_ahb_setup_msa_resources(struct ath11k_base *ab)
 	return 0;
 }
 
+static int ath11k_ahb_ce_remap(struct ath11k_base *ab)
+{
+	const struct ce_remap *ce_remap = ab->hw_params.ce_remap;
+	struct platform_device *pdev = ab->pdev;
+
+	if (!ce_remap) {
+		/* no separate CE register space */
+		ab->mem_ce = ab->mem;
+		return 0;
+	}
+
+	/* ce register space is moved out of wcss unlike ipq8074 or ipq6018
+	 * and the space is not contiguous, hence remapping the CE registers
+	 * to a new space for accessing them.
+	 */
+	ab->mem_ce = ioremap(ce_remap->base, ce_remap->size);
+	if (!ab->mem_ce) {
+		dev_err(&pdev->dev, "ce ioremap error\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void ath11k_ahb_ce_unmap(struct ath11k_base *ab)
+{
+	if (ab->hw_params.ce_remap)
+		iounmap(ab->mem_ce);
+}
+
 static int ath11k_ahb_fw_resources_init(struct ath11k_base *ab)
 {
 	struct ath11k_ahb *ab_ahb = ath11k_ahb_priv(ab);
@@ -1146,21 +1176,9 @@ static int ath11k_ahb_probe(struct platform_device *pdev)
 	if (ret)
 		goto err_core_free;
 
-	ab->mem_ce = ab->mem;
-
-	if (ab->hw_params.ce_remap) {
-		const struct ce_remap *ce_remap = ab->hw_params.ce_remap;
-		/* ce register space is moved out of wcss unlike ipq8074 or ipq6018
-		 * and the space is not contiguous, hence remapping the CE registers
-		 * to a new space for accessing them.
-		 */
-		ab->mem_ce = ioremap(ce_remap->base, ce_remap->size);
-		if (!ab->mem_ce) {
-			dev_err(&pdev->dev, "ce ioremap error\n");
-			ret = -ENOMEM;
-			goto err_core_free;
-		}
-	}
+	ret = ath11k_ahb_ce_remap(ab);
+	if (ret)
+		goto err_core_free;
 
 	ret = ath11k_ahb_fw_resources_init(ab);
 	if (ret)
@@ -1248,9 +1266,7 @@ static void ath11k_ahb_free_resources(struct ath11k_base *ab)
 	ath11k_ahb_release_smp2p_handle(ab);
 	ath11k_ahb_fw_resource_deinit(ab);
 	ath11k_ce_free_pipes(ab);
-
-	if (ab->hw_params.ce_remap)
-		iounmap(ab->mem_ce);
+	ath11k_ahb_ce_unmap(ab);
 
 	ath11k_core_free(ab);
 	platform_set_drvdata(pdev, NULL);
-- 
cgit v1.2.3-58-ga151


From c57d00a4d3d80a6af376bf37527c93654afadc02 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Fri, 3 May 2024 13:34:38 +0300
Subject: wifi: ath11k: unmap the CE in ath11k_ahb_probe() error path

Currently, in ath11k_ahb_probe(), if a failure occurs after
ath11k_ahb_ce_remap() is called, and if hw_params->ce_remap is
enabled, the CE register memory is not unmapped. So add a call to
ath11k_ahb_ce_unmap() in the error path.

This issue was identified by smatch/smatch_scripts/kchecker:

drivers/net/wireless/ath/ath11k/ahb.c:1218 ath11k_ahb_probe() warn: 'ab->mem_ce' from ioremap() not released on lines: 1218.

Since this is a rare error path with no simple way to test, and since
the change is trivial to review, compile tested only.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240430-ce-unmap-v1-2-e468328f95d9@quicinc.com
---
 drivers/net/wireless/ath/ath11k/ahb.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/ahb.c b/drivers/net/wireless/ath/ath11k/ahb.c
index 88dfef71c2b5..e3ff4786c714 100644
--- a/drivers/net/wireless/ath/ath11k/ahb.c
+++ b/drivers/net/wireless/ath/ath11k/ahb.c
@@ -1182,7 +1182,7 @@ static int ath11k_ahb_probe(struct platform_device *pdev)
 
 	ret = ath11k_ahb_fw_resources_init(ab);
 	if (ret)
-		goto err_core_free;
+		goto err_ce_unmap;
 
 	ret = ath11k_ahb_setup_smp2p_handle(ab);
 	if (ret)
@@ -1234,6 +1234,9 @@ err_release_smp2p_handle:
 err_fw_deinit:
 	ath11k_ahb_fw_resource_deinit(ab);
 
+err_ce_unmap:
+	ath11k_ahb_ce_unmap(ab);
+
 err_core_free:
 	ath11k_core_free(ab);
 	platform_set_drvdata(pdev, NULL);
-- 
cgit v1.2.3-58-ga151


From 5f6c3a418b401187c36536dae235beaf322dc7fc Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Fri, 3 May 2024 13:34:38 +0300
Subject: wifi: ath12k: Replace "chip" with "device" in hal Rx return buffer
 manager

Prefer the term "device" over "chip". Therefore, modify the relevant terms
in hal Rx return buffer manager to use "device" instead.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240430165811.1377182-2-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_rx.c |  4 ++--
 drivers/net/wireless/ath/ath12k/hal.h   | 18 +++++++++---------
 2 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index 75df622f25d8..fc4b95324955 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -3071,7 +3071,7 @@ static int ath12k_dp_rx_h_defrag_reo_reinject(struct ath12k *ar,
 
 	ath12k_hal_rx_buf_addr_info_set(&reo_ent_ring->buf_addr_info, link_paddr,
 					cookie,
-					HAL_RX_BUF_RBM_WBM_CHIP0_IDLE_DESC_LIST);
+					HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST);
 
 	mpdu_info = u32_encode_bits(1, RX_MPDU_DESC_INFO0_MSDU_COUNT) |
 		    u32_encode_bits(0, RX_MPDU_DESC_INFO0_FRAG_FLAG) |
@@ -3451,7 +3451,7 @@ int ath12k_dp_rx_process_err(struct ath12k_base *ab, struct napi_struct *napi,
 			       (paddr - link_desc_banks[desc_bank].paddr);
 		ath12k_hal_rx_msdu_link_info_get(link_desc_va, &num_msdus, msdu_cookies,
 						 &rbm);
-		if (rbm != HAL_RX_BUF_RBM_WBM_CHIP0_IDLE_DESC_LIST &&
+		if (rbm != HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST &&
 		    rbm != HAL_RX_BUF_RBM_SW3_BM &&
 		    rbm != ab->hw_params->hal_params->rx_buf_rbm) {
 			ab->soc_stats.invalid_rbm++;
diff --git a/drivers/net/wireless/ath/ath12k/hal.h b/drivers/net/wireless/ath/ath12k/hal.h
index dbb9205bfa10..f364b2938ef8 100644
--- a/drivers/net/wireless/ath/ath12k/hal.h
+++ b/drivers/net/wireless/ath/ath12k/hal.h
@@ -770,12 +770,12 @@ struct hal_srng_config {
  * enum hal_rx_buf_return_buf_manager - manager for returned rx buffers
  *
  * @HAL_RX_BUF_RBM_WBM_IDLE_BUF_LIST: Buffer returned to WBM idle buffer list
- * @HAL_RX_BUF_RBM_WBM_CHIP0_IDLE_DESC_LIST: Descriptor returned to WBM idle
- *	descriptor list, where the chip 0 WBM is chosen in case of a multi-chip config
- * @HAL_RX_BUF_RBM_WBM_CHIP1_IDLE_DESC_LIST: Descriptor returned to WBM idle
- *	descriptor list, where the chip 1 WBM is chosen in case of a multi-chip config
- * @HAL_RX_BUF_RBM_WBM_CHIP2_IDLE_DESC_LIST: Descriptor returned to WBM idle
- *	descriptor list, where the chip 2 WBM is chosen in case of a multi-chip config
+ * @HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST: Descriptor returned to WBM idle
+ *	descriptor list, where the device 0 WBM is chosen in case of a multi-device config
+ * @HAL_RX_BUF_RBM_WBM_DEV1_IDLE_DESC_LIST: Descriptor returned to WBM idle
+ *	descriptor list, where the device 1 WBM is chosen in case of a multi-device config
+ * @HAL_RX_BUF_RBM_WBM_DEV2_IDLE_DESC_LIST: Descriptor returned to WBM idle
+ *	descriptor list, where the device 2 WBM is chosen in case of a multi-device config
  * @HAL_RX_BUF_RBM_FW_BM: Buffer returned to FW
  * @HAL_RX_BUF_RBM_SW0_BM: For ring 0 -- returned to host
  * @HAL_RX_BUF_RBM_SW1_BM: For ring 1 -- returned to host
@@ -788,9 +788,9 @@ struct hal_srng_config {
 
 enum hal_rx_buf_return_buf_manager {
 	HAL_RX_BUF_RBM_WBM_IDLE_BUF_LIST,
-	HAL_RX_BUF_RBM_WBM_CHIP0_IDLE_DESC_LIST,
-	HAL_RX_BUF_RBM_WBM_CHIP1_IDLE_DESC_LIST,
-	HAL_RX_BUF_RBM_WBM_CHIP2_IDLE_DESC_LIST,
+	HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST,
+	HAL_RX_BUF_RBM_WBM_DEV1_IDLE_DESC_LIST,
+	HAL_RX_BUF_RBM_WBM_DEV2_IDLE_DESC_LIST,
 	HAL_RX_BUF_RBM_FW_BM,
 	HAL_RX_BUF_RBM_SW0_BM,
 	HAL_RX_BUF_RBM_SW1_BM,
-- 
cgit v1.2.3-58-ga151


From acc6afb31c3f999076d998f2d1678c0899f0bfb1 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Fri, 3 May 2024 13:34:38 +0300
Subject: wifi: ath12k: Refactor idle ring descriptor setup

Currently, the WBM idle ring descriptor setup uses implicit value
HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST for the return buffer manager
parameter. To support inter-device MLO (Multi-link operation), this
parameter needs to be configure dynamically based on the device identifier
within the MLO group. Therefore, introduce a new argument to the helper
function ath12k_hal_set_link_desc_addr().

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240430165811.1377182-3-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp.c    | 9 ++++++---
 drivers/net/wireless/ath/ath12k/dp.h    | 1 +
 drivers/net/wireless/ath/ath12k/dp_rx.c | 5 +++--
 drivers/net/wireless/ath/ath12k/hal.c   | 5 +++--
 drivers/net/wireless/ath/ath12k/hal.h   | 3 ++-
 5 files changed, 15 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp.c b/drivers/net/wireless/ath/ath12k/dp.c
index 7843c76a82c1..f41b3f9dac3d 100644
--- a/drivers/net/wireless/ath/ath12k/dp.c
+++ b/drivers/net/wireless/ath/ath12k/dp.c
@@ -616,6 +616,7 @@ static int ath12k_dp_scatter_idle_link_desc_setup(struct ath12k_base *ab,
 	int i;
 	int ret = 0;
 	u32 end_offset, cookie;
+	enum hal_rx_buf_return_buf_manager rbm = dp->idle_link_rbm;
 
 	n_entries_per_buf = HAL_WBM_IDLE_SCATTER_BUF_SIZE /
 		ath12k_hal_srng_get_entrysize(ab, HAL_WBM_IDLE_LINK);
@@ -646,7 +647,8 @@ static int ath12k_dp_scatter_idle_link_desc_setup(struct ath12k_base *ab,
 		paddr = link_desc_banks[i].paddr;
 		while (n_entries) {
 			cookie = DP_LINK_DESC_COOKIE_SET(n_entries, i);
-			ath12k_hal_set_link_desc_addr(scatter_buf, cookie, paddr);
+			ath12k_hal_set_link_desc_addr(scatter_buf, cookie,
+						      paddr, rbm);
 			n_entries--;
 			paddr += HAL_LINK_DESC_SIZE;
 			if (rem_entries) {
@@ -790,6 +792,7 @@ int ath12k_dp_link_desc_setup(struct ath12k_base *ab,
 	u32 paddr;
 	int i, ret;
 	u32 cookie;
+	enum hal_rx_buf_return_buf_manager rbm = ab->dp.idle_link_rbm;
 
 	tot_mem_sz = n_link_desc * HAL_LINK_DESC_SIZE;
 	tot_mem_sz += HAL_LINK_DESC_ALIGN;
@@ -850,8 +853,7 @@ int ath12k_dp_link_desc_setup(struct ath12k_base *ab,
 		while (n_entries &&
 		       (desc = ath12k_hal_srng_src_get_next_entry(ab, srng))) {
 			cookie = DP_LINK_DESC_COOKIE_SET(n_entries, i);
-			ath12k_hal_set_link_desc_addr(desc,
-						      cookie, paddr);
+			ath12k_hal_set_link_desc_addr(desc, cookie, paddr, rbm);
 			n_entries--;
 			paddr += HAL_LINK_DESC_SIZE;
 		}
@@ -1603,6 +1605,7 @@ int ath12k_dp_alloc(struct ath12k_base *ab)
 	spin_lock_init(&dp->reo_cmd_lock);
 
 	dp->reo_cmd_cache_flush_count = 0;
+	dp->idle_link_rbm = HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST;
 
 	ret = ath12k_wbm_idle_ring_setup(ab, &n_link_desc);
 	if (ret) {
diff --git a/drivers/net/wireless/ath/ath12k/dp.h b/drivers/net/wireless/ath/ath12k/dp.h
index 5cf0d21ef184..e9880b34a7ab 100644
--- a/drivers/net/wireless/ath/ath12k/dp.h
+++ b/drivers/net/wireless/ath/ath12k/dp.h
@@ -325,6 +325,7 @@ struct ath12k_dp {
 	u8 htt_tgt_ver_major;
 	u8 htt_tgt_ver_minor;
 	struct dp_link_desc_bank link_desc_banks[DP_LINK_DESC_BANKS_MAX];
+	enum hal_rx_buf_return_buf_manager idle_link_rbm;
 	struct dp_srng wbm_idle_ring;
 	struct dp_srng wbm_desc_rel_ring;
 	struct dp_srng tcl_cmd_ring;
diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index fc4b95324955..18030c9e735d 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -2994,6 +2994,7 @@ static int ath12k_dp_rx_h_defrag_reo_reinject(struct ath12k *ar,
 	u32 cookie, hal_rx_desc_sz, dest_ring_info0;
 	int ret;
 	struct ath12k_rx_desc_info *desc_info;
+	enum hal_rx_buf_return_buf_manager idle_link_rbm = dp->idle_link_rbm;
 	u8 dst_ind;
 
 	hal_rx_desc_sz = ab->hal.hal_desc_sz;
@@ -3071,7 +3072,7 @@ static int ath12k_dp_rx_h_defrag_reo_reinject(struct ath12k *ar,
 
 	ath12k_hal_rx_buf_addr_info_set(&reo_ent_ring->buf_addr_info, link_paddr,
 					cookie,
-					HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST);
+					idle_link_rbm);
 
 	mpdu_info = u32_encode_bits(1, RX_MPDU_DESC_INFO0_MSDU_COUNT) |
 		    u32_encode_bits(0, RX_MPDU_DESC_INFO0_FRAG_FLAG) |
@@ -3451,7 +3452,7 @@ int ath12k_dp_rx_process_err(struct ath12k_base *ab, struct napi_struct *napi,
 			       (paddr - link_desc_banks[desc_bank].paddr);
 		ath12k_hal_rx_msdu_link_info_get(link_desc_va, &num_msdus, msdu_cookies,
 						 &rbm);
-		if (rbm != HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST &&
+		if (rbm != dp->idle_link_rbm &&
 		    rbm != HAL_RX_BUF_RBM_SW3_BM &&
 		    rbm != ab->hw_params->hal_params->rx_buf_rbm) {
 			ab->soc_stats.invalid_rbm++;
diff --git a/drivers/net/wireless/ath/ath12k/hal.c b/drivers/net/wireless/ath/ath12k/hal.c
index 78310da8cfe8..ca04bfae8bdc 100644
--- a/drivers/net/wireless/ath/ath12k/hal.c
+++ b/drivers/net/wireless/ath/ath12k/hal.c
@@ -1969,14 +1969,15 @@ u32 ath12k_hal_ce_dst_status_get_length(struct hal_ce_srng_dst_status_desc *desc
 }
 
 void ath12k_hal_set_link_desc_addr(struct hal_wbm_link_desc *desc, u32 cookie,
-				   dma_addr_t paddr)
+				   dma_addr_t paddr,
+				   enum hal_rx_buf_return_buf_manager rbm)
 {
 	desc->buf_addr_info.info0 = le32_encode_bits((paddr & HAL_ADDR_LSB_REG_MASK),
 						     BUFFER_ADDR_INFO0_ADDR);
 	desc->buf_addr_info.info1 =
 			le32_encode_bits(((u64)paddr >> HAL_ADDR_MSB_REG_SHIFT),
 					 BUFFER_ADDR_INFO1_ADDR) |
-			le32_encode_bits(1, BUFFER_ADDR_INFO1_RET_BUF_MGR) |
+			le32_encode_bits(rbm, BUFFER_ADDR_INFO1_RET_BUF_MGR) |
 			le32_encode_bits(cookie, BUFFER_ADDR_INFO1_SW_COOKIE);
 }
 
diff --git a/drivers/net/wireless/ath/ath12k/hal.h b/drivers/net/wireless/ath/ath12k/hal.h
index f364b2938ef8..8a78bb9a10bc 100644
--- a/drivers/net/wireless/ath/ath12k/hal.h
+++ b/drivers/net/wireless/ath/ath12k/hal.h
@@ -1113,7 +1113,8 @@ dma_addr_t ath12k_hal_srng_get_tp_addr(struct ath12k_base *ab,
 dma_addr_t ath12k_hal_srng_get_hp_addr(struct ath12k_base *ab,
 				       struct hal_srng *srng);
 void ath12k_hal_set_link_desc_addr(struct hal_wbm_link_desc *desc, u32 cookie,
-				   dma_addr_t paddr);
+				   dma_addr_t paddr,
+				   enum hal_rx_buf_return_buf_manager rbm);
 u32 ath12k_hal_ce_get_desc_size(enum hal_ce_desc type);
 void ath12k_hal_ce_src_set_desc(struct hal_ce_srng_src_desc *desc, dma_addr_t paddr,
 				u32 len, u32 id, u8 byte_swap_data);
-- 
cgit v1.2.3-58-ga151


From b2d54ca73a5a177490f3f1ee2d23ff6ab6045b25 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Fri, 3 May 2024 13:34:38 +0300
Subject: wifi: ath12k: Introduce device index

Introduce a device index parameter in the ath12k_base structure. This
value is used as unique identifier for devices involved in Inter-device
MLO. It is also used to configure the MLO parameter within the QMI host
capability send request. While currently set to zero, future logic will
be added for inter device support to populate this parameter with
non-zero values.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240430165811.1377182-4-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.c | 10 ++++++++++
 drivers/net/wireless/ath/ath12k/core.h |  1 +
 drivers/net/wireless/ath/ath12k/qmi.c  |  4 ++--
 3 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c
index e9aabdb9341c..9482d5db71e7 100644
--- a/drivers/net/wireless/ath/ath12k/core.c
+++ b/drivers/net/wireless/ath/ath12k/core.c
@@ -1268,6 +1268,16 @@ struct ath12k_base *ath12k_core_alloc(struct device *dev, size_t priv_size,
 	ab->qmi.num_radios = U8_MAX;
 	ab->mlo_capable_flags = ATH12K_INTRA_DEVICE_MLO_SUPPORT;
 
+	/* Device index used to identify the devices in a group.
+	 *
+	 * In Intra-device MLO, only one device present in a group,
+	 * so it is always zero.
+	 *
+	 * In Inter-device MLO, Multiple device present in a group,
+	 * expect non-zero value.
+	 */
+	ab->device_id = 0;
+
 	return ab;
 
 err_free_wq:
diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index d833361948b7..3cfc65b36579 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -754,6 +754,7 @@ struct ath12k_base {
 	struct ath12k_qmi qmi;
 	struct ath12k_wmi_base wmi_ab;
 	struct completion fw_ready;
+	u8 device_id;
 	int num_radios;
 	/* HW channel counters frequency value in hertz common to all MACs */
 	u32 cc_freq_hz;
diff --git a/drivers/net/wireless/ath/ath12k/qmi.c b/drivers/net/wireless/ath/ath12k/qmi.c
index 5484112859a6..3bb730326a02 100644
--- a/drivers/net/wireless/ath/ath12k/qmi.c
+++ b/drivers/net/wireless/ath/ath12k/qmi.c
@@ -2041,7 +2041,7 @@ static void ath12k_host_cap_parse_mlo(struct ath12k_base *ab,
 	req->mlo_capable_valid = 1;
 	req->mlo_capable = 1;
 	req->mlo_chip_id_valid = 1;
-	req->mlo_chip_id = 0;
+	req->mlo_chip_id = ab->device_id;
 	req->mlo_group_id_valid = 1;
 	req->mlo_group_id = 0;
 	req->max_mlo_peer_valid = 1;
@@ -2053,7 +2053,7 @@ static void ath12k_host_cap_parse_mlo(struct ath12k_base *ab,
 	req->mlo_num_chips = 1;
 
 	info = &req->mlo_chip_info[0];
-	info->chip_id = 0;
+	info->chip_id = ab->device_id;
 	info->num_local_links = ab->qmi.num_radios;
 
 	for (i = 0; i < info->num_local_links; i++) {
-- 
cgit v1.2.3-58-ga151


From 53bf4a14f8c55e84d200f8328fde39ab6bdf1bf5 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Fri, 3 May 2024 13:34:39 +0300
Subject: wifi: ath12k: add multi device support for WBM idle ring buffer setup

Currently, inter device MLO is not supported. Therefore, the WBM idle ring
buffers choose the implicit return buffer manager (DEV0_IDLE_DESC_LIST).
However, this implicit return buffer manager design is not meeting the
requirements to support inter device MLO. In inter device MLO, multiple
devices participate. The device specific WBM idle ring buffers transmit
to multiple device REO rings. To distinguish between device specific WBM
idle buffers, the setup configuration need to choose a different return
buffer manager based on the unique identifier (device index).

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240430165811.1377182-5-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp.c | 20 +++++++++++++++++++-
 1 file changed, 19 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp.c b/drivers/net/wireless/ath/ath12k/dp.c
index f41b3f9dac3d..72ea9baaf8d7 100644
--- a/drivers/net/wireless/ath/ath12k/dp.c
+++ b/drivers/net/wireless/ath/ath12k/dp.c
@@ -1589,6 +1589,24 @@ static int ath12k_dp_reoq_lut_setup(struct ath12k_base *ab)
 	return 0;
 }
 
+static enum hal_rx_buf_return_buf_manager
+ath12k_dp_get_idle_link_rbm(struct ath12k_base *ab)
+{
+	switch (ab->device_id) {
+	case 0:
+		return HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST;
+	case 1:
+		return HAL_RX_BUF_RBM_WBM_DEV1_IDLE_DESC_LIST;
+	case 2:
+		return HAL_RX_BUF_RBM_WBM_DEV2_IDLE_DESC_LIST;
+	default:
+		ath12k_warn(ab, "invalid %d device id, so choose default rbm\n",
+			    ab->device_id);
+		WARN_ON(1);
+		return HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST;
+	}
+}
+
 int ath12k_dp_alloc(struct ath12k_base *ab)
 {
 	struct ath12k_dp *dp = &ab->dp;
@@ -1605,7 +1623,7 @@ int ath12k_dp_alloc(struct ath12k_base *ab)
 	spin_lock_init(&dp->reo_cmd_lock);
 
 	dp->reo_cmd_cache_flush_count = 0;
-	dp->idle_link_rbm = HAL_RX_BUF_RBM_WBM_DEV0_IDLE_DESC_LIST;
+	dp->idle_link_rbm = ath12k_dp_get_idle_link_rbm(ab);
 
 	ret = ath12k_wbm_idle_ring_setup(ab, &n_link_desc);
 	if (ret) {
-- 
cgit v1.2.3-58-ga151


From 5453bbd6fef4ca2fea3b4b338fc715d7135afc6f Mon Sep 17 00:00:00 2001
From: Sven Eckelmann <sven@narfation.org>
Date: Fri, 3 May 2024 13:34:39 +0300
Subject: wifi: ath12k: Don't drop tx_status in failure case

When a station idles for a long time, hostapd will try to send
a QoS Null frame to the station as "poll". NL80211_CMD_PROBE_CLIENT
is used for this purpose.
And the skb will be added to ack_status_frame - waiting for a
completion via ieee80211_report_ack_skb().

But when the peer was already removed before the tx_complete arrives,
the peer will be missing. And when using dev_kfree_skb_any (instead
of going through mac80211), the entry will stay inside
ack_status_frames thus not clean up related information in its
internal data structures. This IDR will therefore run full after
8K request were generated for such clients.
At this point, the access point will then just stall and not allow
any new clients because idr_alloc() for ack_status_frame will fail.

ieee80211_free_txskb() on the other hand will (when required) call
ieee80211_report_ack_skb() and make sure that (when required) remove
the entry from the ack_status_frame and clean up related
information in its internal data structures.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Sarika Sharma <quic_sarishar@quicinc.com>
Signed-off-by: Karthikeyan Kathirvel <quic_kathirve@quicinc.com>
Link: https://lore.kernel.org/r/20230802-ath11k-ack_status_leak-v2-1-c0af729d6229@narfation.org
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240430074313.885807-2-quic_kathirve@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_tx.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.c b/drivers/net/wireless/ath/ath12k/dp_tx.c
index 9b6d7d72f57c..81a85d5946f5 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.c
@@ -352,15 +352,15 @@ static void ath12k_dp_tx_free_txbuf(struct ath12k_base *ab,
 	u8 pdev_id = ath12k_hw_mac_id_to_pdev_id(ab->hw_params, mac_id);
 
 	skb_cb = ATH12K_SKB_CB(msdu);
+	ar = ab->pdevs[pdev_id].ar;
 
 	dma_unmap_single(ab->dev, skb_cb->paddr, msdu->len, DMA_TO_DEVICE);
 	if (skb_cb->paddr_ext_desc)
 		dma_unmap_single(ab->dev, skb_cb->paddr_ext_desc,
 				 sizeof(struct hal_tx_msdu_ext_desc), DMA_TO_DEVICE);
 
-	dev_kfree_skb_any(msdu);
+	ieee80211_free_txskb(ar->ah->hw, msdu);
 
-	ar = ab->pdevs[pdev_id].ar;
 	if (atomic_dec_and_test(&ar->dp.num_tx_pending))
 		wake_up(&ar->dp.tx_empty_waitq);
 }
@@ -448,6 +448,7 @@ static void ath12k_dp_tx_complete_msdu(struct ath12k *ar,
 				       struct hal_tx_status *ts)
 {
 	struct ath12k_base *ab = ar->ab;
+	struct ath12k_hw *ah = ar->ah;
 	struct ieee80211_tx_info *info;
 	struct ath12k_skb_cb *skb_cb;
 
@@ -466,12 +467,12 @@ static void ath12k_dp_tx_complete_msdu(struct ath12k *ar,
 	rcu_read_lock();
 
 	if (!rcu_dereference(ab->pdevs_active[ar->pdev_idx])) {
-		dev_kfree_skb_any(msdu);
+		ieee80211_free_txskb(ah->hw, msdu);
 		goto exit;
 	}
 
 	if (!skb_cb->vif) {
-		dev_kfree_skb_any(msdu);
+		ieee80211_free_txskb(ah->hw, msdu);
 		goto exit;
 	}
 
-- 
cgit v1.2.3-58-ga151


From 50971dc6694c0845fcddfe337ea39c5b723d5a92 Mon Sep 17 00:00:00 2001
From: Karthikeyan Kathirvel <quic_kathirve@quicinc.com>
Date: Fri, 3 May 2024 13:34:39 +0300
Subject: wifi: ath12k: drop failed transmitted frames from metric calculation.

In mesh node traffic, internal firmware-transmitted failures are
reported as transmitted failures in mesh metric calculation, leading
to the breakage of the mesh link.

Fix the issue by dropping the internal firmware-transmitted failures
before updating the TX completion status to mac80211, in order to
prevent false failure averaging in mesh metric calculation.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Signed-off-by: Karthikeyan Kathirvel <quic_kathirve@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240430074313.885807-3-quic_kathirve@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_tx.c    | 38 ++++++++++++++++++++++--------
 drivers/net/wireless/ath/ath12k/hal_desc.h | 22 ++++++++++++++++-
 2 files changed, 49 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.c b/drivers/net/wireless/ath/ath12k/dp_tx.c
index 81a85d5946f5..a7c7a868c14c 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.c
@@ -482,18 +482,36 @@ static void ath12k_dp_tx_complete_msdu(struct ath12k *ar,
 	/* skip tx rate update from ieee80211_status*/
 	info->status.rates[0].idx = -1;
 
-	if (ts->status == HAL_WBM_TQM_REL_REASON_FRAME_ACKED &&
-	    !(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
-		info->flags |= IEEE80211_TX_STAT_ACK;
-		info->status.ack_signal = ATH12K_DEFAULT_NOISE_FLOOR +
-					  ts->ack_rssi;
-		info->status.flags = IEEE80211_TX_STATUS_ACK_SIGNAL_VALID;
+	switch (ts->status) {
+	case HAL_WBM_TQM_REL_REASON_FRAME_ACKED:
+		if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
+			info->flags |= IEEE80211_TX_STAT_ACK;
+			info->status.ack_signal = ATH12K_DEFAULT_NOISE_FLOOR +
+						  ts->ack_rssi;
+			info->status.flags = IEEE80211_TX_STATUS_ACK_SIGNAL_VALID;
+		}
+		break;
+	case HAL_WBM_TQM_REL_REASON_CMD_REMOVE_TX:
+		if (info->flags & IEEE80211_TX_CTL_NO_ACK) {
+			info->flags |= IEEE80211_TX_STAT_NOACK_TRANSMITTED;
+			break;
+		}
+		fallthrough;
+	case HAL_WBM_TQM_REL_REASON_CMD_REMOVE_MPDU:
+	case HAL_WBM_TQM_REL_REASON_DROP_THRESHOLD:
+	case HAL_WBM_TQM_REL_REASON_CMD_REMOVE_AGED_FRAMES:
+		/* The failure status is due to internal firmware tx failure
+		 * hence drop the frame; do not update the status of frame to
+		 * the upper layer
+		 */
+		ieee80211_free_txskb(ah->hw, msdu);
+		goto exit;
+	default:
+		ath12k_dbg(ab, ATH12K_DBG_DP_TX, "tx frame is not acked status %d\n",
+			   ts->status);
+		break;
 	}
 
-	if (ts->status == HAL_WBM_TQM_REL_REASON_CMD_REMOVE_TX &&
-	    (info->flags & IEEE80211_TX_CTL_NO_ACK))
-		info->flags |= IEEE80211_TX_STAT_NOACK_TRANSMITTED;
-
 	/* NOTE: Tx rate status reporting. Tx completion status does not have
 	 * necessary information (for example nss) to build the tx rate.
 	 * Might end up reporting it out-of-band from HTT stats.
diff --git a/drivers/net/wireless/ath/ath12k/hal_desc.h b/drivers/net/wireless/ath/ath12k/hal_desc.h
index 63340256d3f6..814c02f876d6 100644
--- a/drivers/net/wireless/ath/ath12k/hal_desc.h
+++ b/drivers/net/wireless/ath/ath12k/hal_desc.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause-Clear */
 /*
  * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved.
- * Copyright (c) 2021-2022 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2021-2022, 2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 #include "core.h"
 
@@ -2048,6 +2048,19 @@ struct hal_wbm_release_ring {
  *	fw with fw_reason2.
  * @HAL_WBM_TQM_REL_REASON_CMD_REMOVE_RESEAON3: Remove command initiated by
  *	fw with fw_reason3.
+ * @HAL_WBM_TQM_REL_REASON_CMD_DISABLE_QUEUE: Remove command initiated by
+ *	fw with disable queue.
+ * @HAL_WBM_TQM_REL_REASON_CMD_TILL_NONMATCHING: Remove command initiated by
+ *	fw to remove all mpdu until 1st non-match.
+ * @HAL_WBM_TQM_REL_REASON_DROP_THRESHOLD: Dropped due to drop threshold
+ *	criteria
+ * @HAL_WBM_TQM_REL_REASON_DROP_LINK_DESC_UNAVAIL: Dropped due to link desc
+ *	not available
+ * @HAL_WBM_TQM_REL_REASON_DROP_OR_INVALID_MSDU: Dropped due drop bit set or
+ *	null flow
+ * @HAL_WBM_TQM_REL_REASON_MULTICAST_DROP: Dropped due mcast drop set for VDEV
+ * @HAL_WBM_TQM_REL_REASON_VDEV_MISMATCH_DROP: Dropped due to being set with
+ *	'TCL_drop_reason'
  */
 enum hal_wbm_tqm_rel_reason {
 	HAL_WBM_TQM_REL_REASON_FRAME_ACKED,
@@ -2058,6 +2071,13 @@ enum hal_wbm_tqm_rel_reason {
 	HAL_WBM_TQM_REL_REASON_CMD_REMOVE_RESEAON1,
 	HAL_WBM_TQM_REL_REASON_CMD_REMOVE_RESEAON2,
 	HAL_WBM_TQM_REL_REASON_CMD_REMOVE_RESEAON3,
+	HAL_WBM_TQM_REL_REASON_CMD_DISABLE_QUEUE,
+	HAL_WBM_TQM_REL_REASON_CMD_TILL_NONMATCHING,
+	HAL_WBM_TQM_REL_REASON_DROP_THRESHOLD,
+	HAL_WBM_TQM_REL_REASON_DROP_LINK_DESC_UNAVAIL,
+	HAL_WBM_TQM_REL_REASON_DROP_OR_INVALID_MSDU,
+	HAL_WBM_TQM_REL_REASON_MULTICAST_DROP,
+	HAL_WBM_TQM_REL_REASON_VDEV_MISMATCH_DROP,
 };
 
 struct hal_wbm_buffer_ring {
-- 
cgit v1.2.3-58-ga151


From 4c2b796be3a12a11ab611917fafdabc9d3862a1d Mon Sep 17 00:00:00 2001
From: Venkateswara Naralasetty <quic_vnaralas@quicinc.com>
Date: Fri, 3 May 2024 16:24:16 +0300
Subject: wifi: ath11k: skip status ring entry processing

If STATUS_BUFFER_DONE is not set for a monitor status ring entry,
we don't process the status ring until STATUS_BUFFER_DONE set
for that status ring entry.

During LMAC reset it may happen that hardware will not write
STATUS_BUFFER_DONE tlv in status buffer, in that case we end up
waiting for STATUS_BUFFER_DONE leading to backpressure on monitor
status ring.

To fix the issue, when HP (Head Pointer) + 1 entry is peeked and if DMA is not
done and if HP + 2 entry's DMA done is set, replenish HP + 1 entry and start
processing in next interrupt. If HP + 2 entry's DMA done is not set, poll onto
HP + 1 entry DMA done to be set.

Also, during monitor attach HP points to the end of the ring and TP (Tail
Pointer) points to the start of the ring.  Using ath11k_hal_srng_src_peek() may
result in processing invalid buffer for the very first interrupt. Since, HW
starts writing buffer from TP.

To avoid this issue call ath11k_hal_srng_src_next_peek() instead of
calling ath11k_hal_srng_src_peek().

Tested-on: IPQ5018 hw1.0 AHB WLAN.HK.2.6.0.1-00861-QCAHKSWPL_SILICONZ-1

Signed-off-by: Venkateswara Naralasetty <quic_vnaralas@quicinc.com>
Co-developed-by: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Signed-off-by: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240429073624.736147-1-quic_tamizhr@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dp_rx.c | 82 +++++++++++++++++++++++++++++----
 drivers/net/wireless/ath/ath11k/hal.c   | 16 ++++++-
 drivers/net/wireless/ath/ath11k/hal.h   |  2 +
 3 files changed, 90 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index afd481f5858f..8a860a17d7e4 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: BSD-3-Clause-Clear
 /*
  * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
- * Copyright (c) 2021-2023 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2021-2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #include <linux/ieee80211.h>
@@ -2990,11 +2990,52 @@ ath11k_dp_rx_mon_update_status_buf_state(struct ath11k_mon_data *pmon,
 	}
 }
 
+static enum dp_mon_status_buf_state
+ath11k_dp_rx_mon_buf_done(struct ath11k_base *ab, struct hal_srng *srng,
+			  struct dp_rxdma_ring *rx_ring)
+{
+	struct ath11k_skb_rxcb *rxcb;
+	struct hal_tlv_hdr *tlv;
+	struct sk_buff *skb;
+	void *status_desc;
+	dma_addr_t paddr;
+	u32 cookie;
+	int buf_id;
+	u8 rbm;
+
+	status_desc = ath11k_hal_srng_src_next_peek(ab, srng);
+	if (!status_desc)
+		return DP_MON_STATUS_NO_DMA;
+
+	ath11k_hal_rx_buf_addr_info_get(status_desc, &paddr, &cookie, &rbm);
+
+	buf_id = FIELD_GET(DP_RXDMA_BUF_COOKIE_BUF_ID, cookie);
+
+	spin_lock_bh(&rx_ring->idr_lock);
+	skb = idr_find(&rx_ring->bufs_idr, buf_id);
+	spin_unlock_bh(&rx_ring->idr_lock);
+
+	if (!skb)
+		return DP_MON_STATUS_NO_DMA;
+
+	rxcb = ATH11K_SKB_RXCB(skb);
+	dma_sync_single_for_cpu(ab->dev, rxcb->paddr,
+				skb->len + skb_tailroom(skb),
+				DMA_FROM_DEVICE);
+
+	tlv = (struct hal_tlv_hdr *)skb->data;
+	if (FIELD_GET(HAL_TLV_HDR_TAG, tlv->tl) != HAL_RX_STATUS_BUFFER_DONE)
+		return DP_MON_STATUS_NO_DMA;
+
+	return DP_MON_STATUS_REPLINISH;
+}
+
 static int ath11k_dp_rx_reap_mon_status_ring(struct ath11k_base *ab, int mac_id,
 					     int *budget, struct sk_buff_head *skb_list)
 {
 	struct ath11k *ar;
 	const struct ath11k_hw_hal_params *hal_params;
+	enum dp_mon_status_buf_state reap_status;
 	struct ath11k_pdev_dp *dp;
 	struct dp_rxdma_ring *rx_ring;
 	struct ath11k_mon_data *pmon;
@@ -3057,15 +3098,38 @@ static int ath11k_dp_rx_reap_mon_status_ring(struct ath11k_base *ab, int mac_id,
 				ath11k_warn(ab, "mon status DONE not set %lx, buf_id %d\n",
 					    FIELD_GET(HAL_TLV_HDR_TAG,
 						      tlv->tl), buf_id);
-				/* If done status is missing, hold onto status
-				 * ring until status is done for this status
-				 * ring buffer.
-				 * Keep HP in mon_status_ring unchanged,
-				 * and break from here.
-				 * Check status for same buffer for next time
+				/* RxDMA status done bit might not be set even
+				 * though tp is moved by HW.
 				 */
-				pmon->buf_state = DP_MON_STATUS_NO_DMA;
-				break;
+
+				/* If done status is missing:
+				 * 1. As per MAC team's suggestion,
+				 *    when HP + 1 entry is peeked and if DMA
+				 *    is not done and if HP + 2 entry's DMA done
+				 *    is set. skip HP + 1 entry and
+				 *    start processing in next interrupt.
+				 * 2. If HP + 2 entry's DMA done is not set,
+				 *    poll onto HP + 1 entry DMA done to be set.
+				 *    Check status for same buffer for next time
+				 *    dp_rx_mon_status_srng_process
+				 */
+
+				reap_status = ath11k_dp_rx_mon_buf_done(ab, srng,
+									rx_ring);
+				if (reap_status == DP_MON_STATUS_NO_DMA)
+					continue;
+
+				spin_lock_bh(&rx_ring->idr_lock);
+				idr_remove(&rx_ring->bufs_idr, buf_id);
+				spin_unlock_bh(&rx_ring->idr_lock);
+
+				dma_unmap_single(ab->dev, rxcb->paddr,
+						 skb->len + skb_tailroom(skb),
+						 DMA_FROM_DEVICE);
+
+				dev_kfree_skb_any(skb);
+				pmon->buf_state = DP_MON_STATUS_REPLINISH;
+				goto move_next;
 			}
 
 			spin_lock_bh(&rx_ring->idr_lock);
diff --git a/drivers/net/wireless/ath/ath11k/hal.c b/drivers/net/wireless/ath/ath11k/hal.c
index f3d04568c221..f02599bd1c36 100644
--- a/drivers/net/wireless/ath/ath11k/hal.c
+++ b/drivers/net/wireless/ath/ath11k/hal.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: BSD-3-Clause-Clear
 /*
  * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
- * Copyright (c) 2021-2023 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2021-2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 #include <linux/dma-mapping.h>
 #include "hal_tx.h"
@@ -796,6 +796,20 @@ u32 *ath11k_hal_srng_src_get_next_reaped(struct ath11k_base *ab,
 	return desc;
 }
 
+u32 *ath11k_hal_srng_src_next_peek(struct ath11k_base *ab, struct hal_srng *srng)
+{
+	u32 next_hp;
+
+	lockdep_assert_held(&srng->lock);
+
+	next_hp = (srng->u.src_ring.hp + srng->entry_size) % srng->ring_size;
+
+	if (next_hp != srng->u.src_ring.cached_tp)
+		return srng->ring_base_vaddr + next_hp;
+
+	return NULL;
+}
+
 u32 *ath11k_hal_srng_src_peek(struct ath11k_base *ab, struct hal_srng *srng)
 {
 	lockdep_assert_held(&srng->lock);
diff --git a/drivers/net/wireless/ath/ath11k/hal.h b/drivers/net/wireless/ath/ath11k/hal.h
index e453c137385e..dc8bbe073017 100644
--- a/drivers/net/wireless/ath/ath11k/hal.h
+++ b/drivers/net/wireless/ath/ath11k/hal.h
@@ -947,6 +947,8 @@ u32 *ath11k_hal_srng_dst_peek(struct ath11k_base *ab, struct hal_srng *srng);
 int ath11k_hal_srng_dst_num_free(struct ath11k_base *ab, struct hal_srng *srng,
 				 bool sync_hw_ptr);
 u32 *ath11k_hal_srng_src_peek(struct ath11k_base *ab, struct hal_srng *srng);
+u32 *ath11k_hal_srng_src_next_peek(struct ath11k_base *ab,
+				   struct hal_srng *srng);
 u32 *ath11k_hal_srng_src_get_next_reaped(struct ath11k_base *ab,
 					 struct hal_srng *srng);
 u32 *ath11k_hal_srng_src_reap_next(struct ath11k_base *ab,
-- 
cgit v1.2.3-58-ga151


From 10d2b4f4aa0b9a1ac55d2313d2c682add25e31c6 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Fri, 3 May 2024 03:32:56 -0700
Subject: wifi: wil6210: Do not use embedded netdev in wil6210_priv

Embedding net_device into structures prohibits the usage of flexible
arrays in the net_device structure. For more details, see the discussion
at [1].

Un-embed the net_device from struct wil6210_priv by converting it
into a pointer. Then use the leverage alloc_netdev_dummy() to allocate
the net_device object at wil_if_add(). The free of the device
occurs at wil_if_remove().

Link: https://lore.kernel.org/all/20240229225910.79e224cf@kernel.org/ [1]
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240503103304.339489-1-leitao@debian.org
---
 drivers/net/wireless/ath/wil6210/netdev.c  | 21 +++++++++++++++------
 drivers/net/wireless/ath/wil6210/wil6210.h |  2 +-
 2 files changed, 16 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/ath/wil6210/netdev.c b/drivers/net/wireless/ath/wil6210/netdev.c
index ee7d7e9c2718..d5d364683c0e 100644
--- a/drivers/net/wireless/ath/wil6210/netdev.c
+++ b/drivers/net/wireless/ath/wil6210/netdev.c
@@ -453,16 +453,21 @@ int wil_if_add(struct wil6210_priv *wil)
 		return rc;
 	}
 
-	init_dummy_netdev(&wil->napi_ndev);
+	wil->napi_ndev = alloc_netdev_dummy(0);
+	if (!wil->napi_ndev) {
+		wil_err(wil, "failed to allocate dummy netdev");
+		rc = -ENOMEM;
+		goto out_wiphy;
+	}
 	if (wil->use_enhanced_dma_hw) {
-		netif_napi_add(&wil->napi_ndev, &wil->napi_rx,
+		netif_napi_add(wil->napi_ndev, &wil->napi_rx,
 			       wil6210_netdev_poll_rx_edma);
-		netif_napi_add_tx(&wil->napi_ndev,
+		netif_napi_add_tx(wil->napi_ndev,
 				  &wil->napi_tx, wil6210_netdev_poll_tx_edma);
 	} else {
-		netif_napi_add(&wil->napi_ndev, &wil->napi_rx,
+		netif_napi_add(wil->napi_ndev, &wil->napi_rx,
 			       wil6210_netdev_poll_rx);
-		netif_napi_add_tx(&wil->napi_ndev,
+		netif_napi_add_tx(wil->napi_ndev,
 				  &wil->napi_tx, wil6210_netdev_poll_tx);
 	}
 
@@ -474,10 +479,12 @@ int wil_if_add(struct wil6210_priv *wil)
 	wiphy_unlock(wiphy);
 	rtnl_unlock();
 	if (rc < 0)
-		goto out_wiphy;
+		goto free_dummy;
 
 	return 0;
 
+free_dummy:
+	free_netdev(wil->napi_ndev);
 out_wiphy:
 	wiphy_unregister(wiphy);
 	return rc;
@@ -554,5 +561,7 @@ void wil_if_remove(struct wil6210_priv *wil)
 	netif_napi_del(&wil->napi_tx);
 	netif_napi_del(&wil->napi_rx);
 
+	free_netdev(wil->napi_ndev);
+
 	wiphy_unregister(wiphy);
 }
diff --git a/drivers/net/wireless/ath/wil6210/wil6210.h b/drivers/net/wireless/ath/wil6210/wil6210.h
index 22a6eb3e12b7..9bd1286d2857 100644
--- a/drivers/net/wireless/ath/wil6210/wil6210.h
+++ b/drivers/net/wireless/ath/wil6210/wil6210.h
@@ -983,7 +983,7 @@ struct wil6210_priv {
 	spinlock_t eap_lock; /* guarding access to eap rekey fields */
 	struct napi_struct napi_rx;
 	struct napi_struct napi_tx;
-	struct net_device napi_ndev; /* dummy net_device serving all VIFs */
+	struct net_device *napi_ndev; /* dummy net_device serving all VIFs */
 
 	/* DMA related */
 	struct wil_ring ring_rx;
-- 
cgit v1.2.3-58-ga151


From 9f74e7b306dcd7a163822ecb3b1587e9bd3f9c62 Mon Sep 17 00:00:00 2001
From: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Date: Tue, 30 Apr 2024 14:44:12 +0530
Subject: wifi: ath12k: fix calling correct function for rx monitor mode

Currently in ath12k_dp_tx_htt_monitor_mode_ring_config()
ath12k_dp_tx_htt_tx_monitor_mode_ring_config() function wrongly called twice.
Fix that by calling ath12k_dp_tx_htt_rx_monitor_mode_ring_config().

Currently monitor mode is disabled in driver so the change is compile
tested and boot sequence verified.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240430091414.2486196-2-quic_tamizhr@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_tx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.c b/drivers/net/wireless/ath/ath12k/dp_tx.c
index a7c7a868c14c..7adfdefc3bdc 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.c
@@ -1069,7 +1069,7 @@ int ath12k_dp_tx_htt_monitor_mode_ring_config(struct ath12k *ar, bool reset)
 		return ret;
 	}
 
-	ret = ath12k_dp_tx_htt_tx_monitor_mode_ring_config(ar, reset);
+	ret = ath12k_dp_tx_htt_rx_monitor_mode_ring_config(ar, reset);
 	if (ret) {
 		ath12k_err(ab, "failed to setup rx monitor filter %d\n", ret);
 		return ret;
-- 
cgit v1.2.3-58-ga151


From ed07ff6745092f36cd996658bd64d74e4331d331 Mon Sep 17 00:00:00 2001
From: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Date: Tue, 30 Apr 2024 14:44:13 +0530
Subject: wifi: ath12k: Remove unsupported tx monitor handling

Currently tx monitor is not supported in driver. So remove
the tx monitor ring configuration, allocation and handling.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240430091414.2486196-3-quic_tamizhr@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_mon.c | 40 +---------------------
 drivers/net/wireless/ath/ath12k/dp_rx.c  | 59 +-------------------------------
 drivers/net/wireless/ath/ath12k/dp_tx.c  | 42 -----------------------
 drivers/net/wireless/ath/ath12k/dp_tx.h  |  3 +-
 4 files changed, 3 insertions(+), 141 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_mon.c b/drivers/net/wireless/ath/ath12k/dp_mon.c
index 6b0b72477540..5c6749bc4039 100644
--- a/drivers/net/wireless/ath/ath12k/dp_mon.c
+++ b/drivers/net/wireless/ath/ath12k/dp_mon.c
@@ -1903,43 +1903,6 @@ ath12k_dp_mon_tx_parse_status_tlv(struct ath12k_base *ab,
 		break;
 	}
 
-	case HAL_MON_BUF_ADDR: {
-		struct dp_rxdma_mon_ring *buf_ring = &ab->dp.tx_mon_buf_ring;
-		struct dp_mon_packet_info *packet_info =
-			(struct dp_mon_packet_info *)tlv_data;
-		int buf_id = u32_get_bits(packet_info->cookie,
-					  DP_RXDMA_BUF_COOKIE_BUF_ID);
-		struct sk_buff *msdu;
-		struct dp_mon_mpdu *mon_mpdu = tx_ppdu_info->tx_mon_mpdu;
-		struct ath12k_skb_rxcb *rxcb;
-
-		spin_lock_bh(&buf_ring->idr_lock);
-		msdu = idr_remove(&buf_ring->bufs_idr, buf_id);
-		spin_unlock_bh(&buf_ring->idr_lock);
-
-		if (unlikely(!msdu)) {
-			ath12k_warn(ab, "monitor destination with invalid buf_id %d\n",
-				    buf_id);
-			return DP_MON_TX_STATUS_PPDU_NOT_DONE;
-		}
-
-		rxcb = ATH12K_SKB_RXCB(msdu);
-		dma_unmap_single(ab->dev, rxcb->paddr,
-				 msdu->len + skb_tailroom(msdu),
-				 DMA_FROM_DEVICE);
-
-		if (!mon_mpdu->head)
-			mon_mpdu->head = msdu;
-		else if (mon_mpdu->tail)
-			mon_mpdu->tail->next = msdu;
-
-		mon_mpdu->tail = msdu;
-
-		ath12k_dp_mon_buf_replenish(ab, buf_ring, 1);
-		status = DP_MON_TX_BUFFER_ADDR;
-		break;
-	}
-
 	case HAL_TX_MPDU_END:
 		list_add_tail(&tx_ppdu_info->tx_mon_mpdu->list,
 			      &tx_ppdu_info->dp_tx_mon_mpdu_list);
@@ -2088,8 +2051,7 @@ int ath12k_dp_mon_srng_process(struct ath12k *ar, int mac_id, int *budget,
 		mon_dst_ring = &pdev_dp->rxdma_mon_dst_ring[srng_id];
 		buf_ring = &dp->rxdma_mon_buf_ring;
 	} else {
-		mon_dst_ring = &pdev_dp->tx_mon_dst_ring[srng_id];
-		buf_ring = &dp->tx_mon_buf_ring;
+		return 0;
 	}
 
 	srng = &ab->hal.srng_list[mon_dst_ring->ring_id];
diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index 18030c9e735d..952a377b5491 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -422,8 +422,6 @@ static int ath12k_dp_rxdma_buf_free(struct ath12k_base *ab)
 
 	ath12k_dp_rxdma_mon_buf_ring_free(ab, &dp->rxdma_mon_buf_ring);
 
-	ath12k_dp_rxdma_mon_buf_ring_free(ab, &dp->tx_mon_buf_ring);
-
 	return 0;
 }
 
@@ -476,15 +474,6 @@ static int ath12k_dp_rxdma_buf_setup(struct ath12k_base *ab)
 				    "failed to setup HAL_RXDMA_MONITOR_BUF\n");
 			return ret;
 		}
-
-		ret = ath12k_dp_rxdma_mon_ring_buf_setup(ab,
-							 &dp->tx_mon_buf_ring,
-							 HAL_TX_MONITOR_BUF);
-		if (ret) {
-			ath12k_warn(ab,
-				    "failed to setup HAL_TX_MONITOR_BUF\n");
-			return ret;
-		}
 	}
 
 	return 0;
@@ -496,10 +485,8 @@ static void ath12k_dp_rx_pdev_srng_free(struct ath12k *ar)
 	struct ath12k_base *ab = ar->ab;
 	int i;
 
-	for (i = 0; i < ab->hw_params->num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params->num_rxmda_per_pdev; i++)
 		ath12k_dp_srng_cleanup(ab, &dp->rxdma_mon_dst_ring[i]);
-		ath12k_dp_srng_cleanup(ab, &dp->tx_mon_dst_ring[i]);
-	}
 }
 
 void ath12k_dp_rx_pdev_reo_cleanup(struct ath12k_base *ab)
@@ -554,17 +541,6 @@ static int ath12k_dp_rx_pdev_srng_alloc(struct ath12k *ar)
 				    "failed to setup HAL_RXDMA_MONITOR_DST\n");
 			return ret;
 		}
-
-		ret = ath12k_dp_srng_setup(ar->ab,
-					   &dp->tx_mon_dst_ring[i],
-					   HAL_TX_MONITOR_DST,
-					   0, mac_id + i,
-					   DP_TX_MONITOR_DEST_RING_SIZE);
-		if (ret) {
-			ath12k_warn(ar->ab,
-				    "failed to setup HAL_TX_MONITOR_DST\n");
-			return ret;
-		}
 	}
 
 	return 0;
@@ -3971,7 +3947,6 @@ void ath12k_dp_rx_free(struct ath12k_base *ab)
 		ath12k_dp_srng_cleanup(ab, &dp->rxdma_err_dst_ring[i]);
 
 	ath12k_dp_srng_cleanup(ab, &dp->rxdma_mon_buf_ring.refill_buf_ring);
-	ath12k_dp_srng_cleanup(ab, &dp->tx_mon_buf_ring.refill_buf_ring);
 
 	ath12k_dp_rxdma_buf_free(ab);
 }
@@ -4114,15 +4089,6 @@ int ath12k_dp_rx_htt_setup(struct ath12k_base *ab)
 				    ret);
 			return ret;
 		}
-
-		ring_id = dp->tx_mon_buf_ring.refill_buf_ring.ring_id;
-		ret = ath12k_dp_tx_htt_srng_setup(ab, ring_id,
-						  0, HAL_TX_MONITOR_BUF);
-		if (ret) {
-			ath12k_warn(ab, "failed to configure rxdma_mon_buf_ring %d\n",
-				    ret);
-			return ret;
-		}
 	}
 
 	ret = ab->hw_params->hw_ops->rxdma_ring_sel_config(ab);
@@ -4142,9 +4108,6 @@ int ath12k_dp_rx_alloc(struct ath12k_base *ab)
 	idr_init(&dp->rxdma_mon_buf_ring.bufs_idr);
 	spin_lock_init(&dp->rxdma_mon_buf_ring.idr_lock);
 
-	idr_init(&dp->tx_mon_buf_ring.bufs_idr);
-	spin_lock_init(&dp->tx_mon_buf_ring.idr_lock);
-
 	ret = ath12k_dp_srng_setup(ab,
 				   &dp->rx_refill_buf_ring.refill_buf_ring,
 				   HAL_RXDMA_BUF, 0, 0,
@@ -4187,15 +4150,6 @@ int ath12k_dp_rx_alloc(struct ath12k_base *ab)
 			ath12k_warn(ab, "failed to setup HAL_RXDMA_MONITOR_BUF\n");
 			return ret;
 		}
-
-		ret = ath12k_dp_srng_setup(ab,
-					   &dp->tx_mon_buf_ring.refill_buf_ring,
-					   HAL_TX_MONITOR_BUF, 0, 0,
-					   DP_TX_MONITOR_BUF_RING_SIZE);
-		if (ret) {
-			ath12k_warn(ab, "failed to setup DP_TX_MONITOR_BUF_RING_SIZE\n");
-			return ret;
-		}
 	}
 
 	ret = ath12k_dp_rxdma_buf_setup(ab);
@@ -4235,17 +4189,6 @@ int ath12k_dp_rx_pdev_alloc(struct ath12k_base *ab, int mac_id)
 				    i, ret);
 			return ret;
 		}
-
-		ring_id = dp->tx_mon_dst_ring[i].ring_id;
-		ret = ath12k_dp_tx_htt_srng_setup(ab, ring_id,
-						  mac_id + i,
-						  HAL_TX_MONITOR_DST);
-		if (ret) {
-			ath12k_warn(ab,
-				    "failed to configure tx_mon_dst_ring %d %d\n",
-				    i, ret);
-			return ret;
-		}
 	}
 out:
 	return 0;
diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.c b/drivers/net/wireless/ath/ath12k/dp_tx.c
index 7adfdefc3bdc..a22fa43c87ec 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.c
@@ -688,14 +688,6 @@ ath12k_dp_tx_get_ring_id_type(struct ath12k_base *ab,
 		*htt_ring_id = HTT_RXDMA_MONITOR_DESC_RING;
 		*htt_ring_type = HTT_SW_TO_HW_RING;
 		break;
-	case HAL_TX_MONITOR_BUF:
-		*htt_ring_id = HTT_TX_MON_HOST2MON_BUF_RING;
-		*htt_ring_type = HTT_SW_TO_HW_RING;
-		break;
-	case HAL_TX_MONITOR_DST:
-		*htt_ring_id = HTT_TX_MON_MON2HOST_DEST_RING;
-		*htt_ring_type = HTT_HW_TO_SW_RING;
-		break;
 	default:
 		ath12k_warn(ab, "Unsupported ring type in DP :%d\n", ring_type);
 		ret = -EINVAL;
@@ -1063,12 +1055,6 @@ int ath12k_dp_tx_htt_monitor_mode_ring_config(struct ath12k *ar, bool reset)
 	struct ath12k_base *ab = ar->ab;
 	int ret;
 
-	ret = ath12k_dp_tx_htt_tx_monitor_mode_ring_config(ar, reset);
-	if (ret) {
-		ath12k_err(ab, "failed to setup tx monitor filter %d\n", ret);
-		return ret;
-	}
-
 	ret = ath12k_dp_tx_htt_rx_monitor_mode_ring_config(ar, reset);
 	if (ret) {
 		ath12k_err(ab, "failed to setup rx monitor filter %d\n", ret);
@@ -1228,31 +1214,3 @@ err_free:
 	dev_kfree_skb_any(skb);
 	return ret;
 }
-
-int ath12k_dp_tx_htt_tx_monitor_mode_ring_config(struct ath12k *ar, bool reset)
-{
-	struct ath12k_base *ab = ar->ab;
-	struct ath12k_dp *dp = &ab->dp;
-	struct htt_tx_ring_tlv_filter tlv_filter = {0};
-	int ret, ring_id;
-
-	ring_id = dp->tx_mon_buf_ring.refill_buf_ring.ring_id;
-
-	/* TODO: Need to set upstream/downstream tlv filters
-	 * here
-	 */
-
-	if (ab->hw_params->rxdma1_enable) {
-		ret = ath12k_dp_tx_htt_tx_filter_setup(ar->ab, ring_id, 0,
-						       HAL_TX_MONITOR_BUF,
-						       DP_RXDMA_REFILL_RING_SIZE,
-						       &tlv_filter);
-		if (ret) {
-			ath12k_err(ab,
-				   "failed to setup filter for monitor buf %d\n", ret);
-			return ret;
-		}
-	}
-
-	return 0;
-}
diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.h b/drivers/net/wireless/ath/ath12k/dp_tx.h
index 436d77e5e9ee..7faa81faf6a7 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.h
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause-Clear */
 /*
  * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved.
- * Copyright (c) 2021-2022 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2021-2022, 2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #ifndef ATH12K_DP_TX_H
@@ -36,6 +36,5 @@ int ath12k_dp_tx_htt_tx_filter_setup(struct ath12k_base *ab, u32 ring_id,
 				     int mac_id, enum hal_ring_type ring_type,
 				     int tx_buf_size,
 				     struct htt_tx_ring_tlv_filter *htt_tlv_filter);
-int ath12k_dp_tx_htt_tx_monitor_mode_ring_config(struct ath12k *ar, bool reset);
 int ath12k_dp_tx_htt_monitor_mode_ring_config(struct ath12k *ar, bool reset);
 #endif
-- 
cgit v1.2.3-58-ga151


From 582e94ebe230afa35baef06833006b24b601ab11 Mon Sep 17 00:00:00 2001
From: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Date: Tue, 30 Apr 2024 14:44:14 +0530
Subject: wifi: ath12k: Remove unused tcl_*_ring configuration

Currently memory allocation for multiple hardware rings done
in host as these memory will be used by firmware/hardware.
But in case of tcl_cmd_ring and tcl_status_ring neither firmware
nor hardware uses host allocated memory and it uses memory
allocated by firmware itself. So avoid tcl_cmd_ring and
tcl_status_ring setup in host.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Tamizh Chelvam Raja <quic_tamizhr@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240430091414.2486196-4-quic_tamizhr@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp.c | 16 ----------------
 drivers/net/wireless/ath/ath12k/dp.h |  2 --
 2 files changed, 18 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp.c b/drivers/net/wireless/ath/ath12k/dp.c
index 72ea9baaf8d7..ca41f4431394 100644
--- a/drivers/net/wireless/ath/ath12k/dp.c
+++ b/drivers/net/wireless/ath/ath12k/dp.c
@@ -457,8 +457,6 @@ static void ath12k_dp_srng_common_cleanup(struct ath12k_base *ab)
 		ath12k_dp_srng_cleanup(ab, &dp->tx_ring[i].tcl_comp_ring);
 		ath12k_dp_srng_cleanup(ab, &dp->tx_ring[i].tcl_data_ring);
 	}
-	ath12k_dp_srng_cleanup(ab, &dp->tcl_status_ring);
-	ath12k_dp_srng_cleanup(ab, &dp->tcl_cmd_ring);
 	ath12k_dp_srng_cleanup(ab, &dp->wbm_desc_rel_ring);
 }
 
@@ -479,20 +477,6 @@ static int ath12k_dp_srng_common_setup(struct ath12k_base *ab)
 		goto err;
 	}
 
-	ret = ath12k_dp_srng_setup(ab, &dp->tcl_cmd_ring, HAL_TCL_CMD, 0, 0,
-				   DP_TCL_CMD_RING_SIZE);
-	if (ret) {
-		ath12k_warn(ab, "failed to set up tcl_cmd ring :%d\n", ret);
-		goto err;
-	}
-
-	ret = ath12k_dp_srng_setup(ab, &dp->tcl_status_ring, HAL_TCL_STATUS,
-				   0, 0, DP_TCL_STATUS_RING_SIZE);
-	if (ret) {
-		ath12k_warn(ab, "failed to set up tcl_status ring :%d\n", ret);
-		goto err;
-	}
-
 	for (i = 0; i < ab->hw_params->max_tx_ring; i++) {
 		map = ab->hw_params->hal_ops->tcl_to_wbm_rbm_map;
 		tx_comp_ring_num = map[i].wbm_ring_num;
diff --git a/drivers/net/wireless/ath/ath12k/dp.h b/drivers/net/wireless/ath/ath12k/dp.h
index e9880b34a7ab..43589c3dd4bc 100644
--- a/drivers/net/wireless/ath/ath12k/dp.h
+++ b/drivers/net/wireless/ath/ath12k/dp.h
@@ -328,8 +328,6 @@ struct ath12k_dp {
 	enum hal_rx_buf_return_buf_manager idle_link_rbm;
 	struct dp_srng wbm_idle_ring;
 	struct dp_srng wbm_desc_rel_ring;
-	struct dp_srng tcl_cmd_ring;
-	struct dp_srng tcl_status_ring;
 	struct dp_srng reo_reinject_ring;
 	struct dp_srng rx_rel_ring;
 	struct dp_srng reo_except_ring;
-- 
cgit v1.2.3-58-ga151


From 782615f280ce8ad4e3effb42d70304f4dd2d487f Mon Sep 17 00:00:00 2001
From: Kang Yang <quic_kangyang@quicinc.com>
Date: Mon, 29 Apr 2024 16:15:21 +0800
Subject: wifi: ath12k: remove unused variable monitor_flags

monitor_flags is defined in struct ath12k. Although it is changed in
some places, but it is not actually used.

So remove related code.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Kang Yang <quic_kangyang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240429081525.983-2-quic_kangyang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.h |  5 -----
 drivers/net/wireless/ath/ath12k/mac.c  | 10 +---------
 2 files changed, 1 insertion(+), 14 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index 3cfc65b36579..bb6c1b562baf 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -212,10 +212,6 @@ enum ath12k_dev_flags {
 	ATH12K_FLAG_EXT_IRQ_ENABLED,
 };
 
-enum ath12k_monitor_flags {
-	ATH12K_FLAG_MONITOR_ENABLED,
-};
-
 struct ath12k_tx_conf {
 	bool changed;
 	u16 ac;
@@ -532,7 +528,6 @@ struct ath12k {
 
 	unsigned long dev_flags;
 	unsigned int filter_flags;
-	unsigned long monitor_flags;
 	u32 min_tx_power;
 	u32 max_tx_power;
 	u32 txpower_limit_2g;
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 4f07ac8c0287..a2b5aa86ec0b 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -6510,7 +6510,6 @@ err_vdev_del:
 
 	/* Recalc txpower for remaining vdev */
 	ath12k_mac_txpower_recalc(ar);
-	clear_bit(ATH12K_FLAG_MONITOR_ENABLED, &ar->monitor_flags);
 
 	/* TODO: recal traffic pause state based on the available vdevs */
 	arvif->is_created = false;
@@ -6581,15 +6580,9 @@ static void ath12k_mac_configure_filter(struct ath12k *ar,
 	reset_flag = !(ar->filter_flags & FIF_BCN_PRBRESP_PROMISC);
 
 	ret = ath12k_dp_tx_htt_monitor_mode_ring_config(ar, reset_flag);
-	if (!ret) {
-		if (!reset_flag)
-			set_bit(ATH12K_FLAG_MONITOR_ENABLED, &ar->monitor_flags);
-		else
-			clear_bit(ATH12K_FLAG_MONITOR_ENABLED, &ar->monitor_flags);
-	} else {
+	if (ret)
 		ath12k_warn(ar->ab,
 			    "fail to set monitor filter: %d\n", ret);
-	}
 
 	ath12k_dbg(ar->ab, ATH12K_DBG_MAC,
 		   "total_flags:0x%x, reset_flag:%d\n",
@@ -8869,7 +8862,6 @@ static void ath12k_mac_setup(struct ath12k *ar)
 
 	INIT_WORK(&ar->wmi_mgmt_tx_work, ath12k_mgmt_over_wmi_tx_work);
 	skb_queue_head_init(&ar->wmi_mgmt_tx_queue);
-	clear_bit(ATH12K_FLAG_MONITOR_ENABLED, &ar->monitor_flags);
 }
 
 int ath12k_mac_register(struct ath12k_base *ab)
-- 
cgit v1.2.3-58-ga151


From 3b0989e925f38df733a03ff5a320d6841006b3f9 Mon Sep 17 00:00:00 2001
From: Kang Yang <quic_kangyang@quicinc.com>
Date: Mon, 29 Apr 2024 16:15:22 +0800
Subject: wifi: ath12k: avoid duplicated vdev stop

ath12k_mac_op_unassign_vif_chanctx() will do vdev stop in
ath12k_mac_monitor_stop(). This ath12k_mac_vdev_stop() will do vdev stop
again, then might trigger firmware crash.

So add judgement to avoid duplicated vdev stop.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Fixes: c9e4e41e71ff ("wifi: ath12k: move peer delete after vdev stop of station for WCN7850")
Signed-off-by: Kang Yang <quic_kangyang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240429081525.983-3-quic_kangyang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index a2b5aa86ec0b..cdea47faef74 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -7397,7 +7397,8 @@ ath12k_mac_op_unassign_vif_chanctx(struct ieee80211_hw *hw,
 		arvif->is_started = false;
 	}
 
-	if (arvif->vdev_type != WMI_VDEV_TYPE_STA) {
+	if (arvif->vdev_type != WMI_VDEV_TYPE_STA &&
+	    arvif->vdev_type != WMI_VDEV_TYPE_MONITOR) {
 		ath12k_bss_disassoc(ar, arvif);
 		ret = ath12k_mac_vdev_stop(arvif);
 		if (ret)
-- 
cgit v1.2.3-58-ga151


From f40aaca07bf7b52e82ca8e5ddb29190466b5dcbd Mon Sep 17 00:00:00 2001
From: Kang Yang <quic_kangyang@quicinc.com>
Date: Mon, 29 Apr 2024 16:15:23 +0800
Subject: wifi: ath12k: avoid duplicated vdev down

ath12k_mac_op_unassign_vif_chanctx() will do vdev down in function
ath12k_mac_monitor_stop(). So no need to do vdev down again.

remove the duplicated one.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Kang Yang <quic_kangyang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240429081525.983-4-quic_kangyang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index cdea47faef74..1fc727b614d6 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -7407,10 +7407,6 @@ ath12k_mac_op_unassign_vif_chanctx(struct ieee80211_hw *hw,
 	}
 	arvif->is_started = false;
 
-	if (ab->hw_params->vdev_start_delay &&
-	    arvif->vdev_type == WMI_VDEV_TYPE_MONITOR)
-		ath12k_wmi_vdev_down(ar, arvif->vdev_id);
-
 	if (arvif->vdev_type != WMI_VDEV_TYPE_MONITOR &&
 	    ar->num_started_vdevs == 1 && ar->monitor_vdev_created)
 		ath12k_mac_monitor_stop(ar);
-- 
cgit v1.2.3-58-ga151


From efd920f48d1f9c41368579d1a278e88c082cc233 Mon Sep 17 00:00:00 2001
From: Kang Yang <quic_kangyang@quicinc.com>
Date: Mon, 29 Apr 2024 16:15:25 +0800
Subject: wifi: ath12k: remove invalid peer create logic

In ath12k_mac_op_assign_vif_chanctx(), there is a logic to
create peer using ar->mac_addr for a STA vdev. This is invalid
because a STA vdev should have a peer created using AP's
MAC address. Besides, if we run into that logic, it means a peer
has already been created earlier, we should not create it again.
So remove it.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Kang Yang <quic_kangyang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240429081525.983-6-quic_kangyang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 16 ----------------
 1 file changed, 16 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 1fc727b614d6..96dc5c2e096f 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -7270,7 +7270,6 @@ ath12k_mac_op_assign_vif_chanctx(struct ieee80211_hw *hw,
 	struct ath12k_base *ab;
 	struct ath12k_vif *arvif = ath12k_vif_to_arvif(vif);
 	int ret;
-	struct ath12k_wmi_peer_create_arg param;
 
 	/* For multi radio wiphy, the vdev was not created during add_interface
 	 * create now since we have a channel ctx now to assign to a specific ar/fw
@@ -7306,21 +7305,6 @@ ath12k_mac_op_assign_vif_chanctx(struct ieee80211_hw *hw,
 		goto out;
 	}
 
-	if (ab->hw_params->vdev_start_delay &&
-	    arvif->vdev_type != WMI_VDEV_TYPE_AP &&
-	    arvif->vdev_type != WMI_VDEV_TYPE_MONITOR) {
-		param.vdev_id = arvif->vdev_id;
-		param.peer_type = WMI_PEER_TYPE_DEFAULT;
-		param.peer_addr = ar->mac_addr;
-
-		ret = ath12k_peer_create(ar, arvif, NULL, &param);
-		if (ret) {
-			ath12k_warn(ab, "failed to create peer after vdev start delay: %d",
-				    ret);
-			goto out;
-		}
-	}
-
 	if (arvif->vdev_type == WMI_VDEV_TYPE_MONITOR) {
 		ret = ath12k_mac_monitor_start(ar);
 		if (ret)
-- 
cgit v1.2.3-58-ga151


From 4c9aa94a39e6ef1a1381b1e6d841105fee5b7c96 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Wed, 24 Apr 2024 11:23:49 -0700
Subject: wifi: rtw89: Un-embed dummy device

Embedding net_device into structures prohibits the usage of flexible
arrays in the net_device structure. For more details, see the discussion
at [1].

Un-embed the net_device from the private struct by converting it
into a pointer. Then use the leverage the new alloc_netdev_dummy()
helper to allocate and initialize dummy devices.

[1] https://lore.kernel.org/all/20240229225910.79e224cf@kernel.org/

Tested with RTL8852BE, RTL8852C and RTL8922AE.

Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Tested-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240424182351.3936556-1-leitao@debian.org
---
 drivers/net/wireless/realtek/rtw89/core.c | 11 ++++++++---
 drivers/net/wireless/realtek/rtw89/core.h |  4 ++--
 drivers/net/wireless/realtek/rtw89/pci.c  |  6 +++++-
 3 files changed, 15 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index ddc390d24ec1..53997fa51ba0 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -2492,11 +2492,15 @@ void rtw89_core_napi_stop(struct rtw89_dev *rtwdev)
 }
 EXPORT_SYMBOL(rtw89_core_napi_stop);
 
-void rtw89_core_napi_init(struct rtw89_dev *rtwdev)
+int rtw89_core_napi_init(struct rtw89_dev *rtwdev)
 {
-	init_dummy_netdev(&rtwdev->netdev);
-	netif_napi_add(&rtwdev->netdev, &rtwdev->napi,
+	rtwdev->netdev = alloc_netdev_dummy(0);
+	if (!rtwdev->netdev)
+		return -ENOMEM;
+
+	netif_napi_add(rtwdev->netdev, &rtwdev->napi,
 		       rtwdev->hci.ops->napi_poll);
+	return 0;
 }
 EXPORT_SYMBOL(rtw89_core_napi_init);
 
@@ -2504,6 +2508,7 @@ void rtw89_core_napi_deinit(struct rtw89_dev *rtwdev)
 {
 	rtw89_core_napi_stop(rtwdev);
 	netif_napi_del(&rtwdev->napi);
+	free_netdev(rtwdev->netdev);
 }
 EXPORT_SYMBOL(rtw89_core_napi_deinit);
 
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 112bdd95fc6e..0648458110ce 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -5469,7 +5469,7 @@ struct rtw89_dev {
 	struct rtw89_wow_param wow;
 
 	/* napi structure */
-	struct net_device netdev;
+	struct net_device *netdev;
 	struct napi_struct napi;
 	int napi_budget_countdown;
 
@@ -6441,7 +6441,7 @@ void rtw89_core_query_rxdesc_v2(struct rtw89_dev *rtwdev,
 				u8 *data, u32 data_offset);
 void rtw89_core_napi_start(struct rtw89_dev *rtwdev);
 void rtw89_core_napi_stop(struct rtw89_dev *rtwdev);
-void rtw89_core_napi_init(struct rtw89_dev *rtwdev);
+int rtw89_core_napi_init(struct rtw89_dev *rtwdev);
 void rtw89_core_napi_deinit(struct rtw89_dev *rtwdev);
 int rtw89_core_sta_add(struct rtw89_dev *rtwdev,
 		       struct ieee80211_vif *vif,
diff --git a/drivers/net/wireless/realtek/rtw89/pci.c b/drivers/net/wireless/realtek/rtw89/pci.c
index 7b00476a5dee..f93385cc20cd 100644
--- a/drivers/net/wireless/realtek/rtw89/pci.c
+++ b/drivers/net/wireless/realtek/rtw89/pci.c
@@ -4293,7 +4293,11 @@ int rtw89_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	rtw89_pci_link_cfg(rtwdev);
 	rtw89_pci_l1ss_cfg(rtwdev);
 
-	rtw89_core_napi_init(rtwdev);
+	ret = rtw89_core_napi_init(rtwdev);
+	if (ret) {
+		rtw89_err(rtwdev, "failed to init napi\n");
+		goto err_clear_resource;
+	}
 
 	ret = rtw89_pci_request_irq(rtwdev, pdev);
 	if (ret) {
-- 
cgit v1.2.3-58-ga151


From d0bcb10e7b94861c714ab1e62c98bf8abaa37527 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Thu, 25 Apr 2024 01:29:09 -0700
Subject: wifi: rtw88: Un-embed dummy device

Embedding net_device into structures prohibits the usage of flexible
arrays in the net_device structure. For more details, see the discussion
at [1].

Un-embed the net_device from the private struct by converting it
into a pointer. Then use the leverage the new alloc_netdev_dummy()
helper to allocate and initialize dummy devices.

[1] https://lore.kernel.org/all/20240229225910.79e224cf@kernel.org/

Tested with RTL8723DE and RTL8822CE.

Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Tested-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240425082910.2824393-1-leitao@debian.org
---
 drivers/net/wireless/realtek/rtw88/pci.c | 17 +++++++++++++----
 drivers/net/wireless/realtek/rtw88/pci.h |  2 +-
 2 files changed, 14 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/pci.c b/drivers/net/wireless/realtek/rtw88/pci.c
index 7a093f3d5f74..095dcbadbeda 100644
--- a/drivers/net/wireless/realtek/rtw88/pci.c
+++ b/drivers/net/wireless/realtek/rtw88/pci.c
@@ -1682,12 +1682,16 @@ static int rtw_pci_napi_poll(struct napi_struct *napi, int budget)
 	return work_done;
 }
 
-static void rtw_pci_napi_init(struct rtw_dev *rtwdev)
+static int rtw_pci_napi_init(struct rtw_dev *rtwdev)
 {
 	struct rtw_pci *rtwpci = (struct rtw_pci *)rtwdev->priv;
 
-	init_dummy_netdev(&rtwpci->netdev);
-	netif_napi_add(&rtwpci->netdev, &rtwpci->napi, rtw_pci_napi_poll);
+	rtwpci->netdev = alloc_netdev_dummy(0);
+	if (!rtwpci->netdev)
+		return -ENOMEM;
+
+	netif_napi_add(rtwpci->netdev, &rtwpci->napi, rtw_pci_napi_poll);
+	return 0;
 }
 
 static void rtw_pci_napi_deinit(struct rtw_dev *rtwdev)
@@ -1696,6 +1700,7 @@ static void rtw_pci_napi_deinit(struct rtw_dev *rtwdev)
 
 	rtw_pci_napi_stop(rtwdev);
 	netif_napi_del(&rtwpci->napi);
+	free_netdev(rtwpci->netdev);
 }
 
 int rtw_pci_probe(struct pci_dev *pdev,
@@ -1745,7 +1750,11 @@ int rtw_pci_probe(struct pci_dev *pdev,
 		goto err_pci_declaim;
 	}
 
-	rtw_pci_napi_init(rtwdev);
+	ret = rtw_pci_napi_init(rtwdev);
+	if (ret) {
+		rtw_err(rtwdev, "failed to setup NAPI\n");
+		goto err_pci_declaim;
+	}
 
 	ret = rtw_chip_info_setup(rtwdev);
 	if (ret) {
diff --git a/drivers/net/wireless/realtek/rtw88/pci.h b/drivers/net/wireless/realtek/rtw88/pci.h
index 0c37efd8c66f..13988db1cb4c 100644
--- a/drivers/net/wireless/realtek/rtw88/pci.h
+++ b/drivers/net/wireless/realtek/rtw88/pci.h
@@ -215,7 +215,7 @@ struct rtw_pci {
 	bool running;
 
 	/* napi structure */
-	struct net_device netdev;
+	struct net_device *netdev;
 	struct napi_struct napi;
 
 	u16 rx_tag;
-- 
cgit v1.2.3-58-ga151


From a892f6ffbec7a1a25c639534bee62200418242f9 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Fri, 3 May 2024 13:53:28 +0300
Subject: wifi: rtw88: usb: Simplify rtw_usb_write_data

The skb created in this function always has the same headroom,
the chip's TX descriptor size. (pkt_info->offset is set by
rtw_usb_write_data_rsvd_page() to chip->tx_pkt_desc_sz.) Use
chip->tx_pkt_desc_sz directly.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Tested-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/2479507e-3946-492f-857e-83e54969aad2@gmail.com
---
 drivers/net/wireless/realtek/rtw88/usb.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/usb.c b/drivers/net/wireless/realtek/rtw88/usb.c
index a0188511099a..90afeefe002f 100644
--- a/drivers/net/wireless/realtek/rtw88/usb.c
+++ b/drivers/net/wireless/realtek/rtw88/usb.c
@@ -433,23 +433,21 @@ static int rtw_usb_write_data(struct rtw_dev *rtwdev,
 {
 	const struct rtw_chip_info *chip = rtwdev->chip;
 	struct sk_buff *skb;
-	unsigned int desclen, headsize, size;
+	unsigned int size;
 	u8 qsel;
 	int ret = 0;
 
 	size = pkt_info->tx_pkt_size;
 	qsel = pkt_info->qsel;
-	desclen = chip->tx_pkt_desc_sz;
-	headsize = pkt_info->offset ? pkt_info->offset : desclen;
 
-	skb = dev_alloc_skb(headsize + size);
+	skb = dev_alloc_skb(chip->tx_pkt_desc_sz + size);
 	if (unlikely(!skb))
 		return -ENOMEM;
 
-	skb_reserve(skb, headsize);
+	skb_reserve(skb, chip->tx_pkt_desc_sz);
 	skb_put_data(skb, buf, size);
-	skb_push(skb, headsize);
-	memset(skb->data, 0, headsize);
+	skb_push(skb, chip->tx_pkt_desc_sz);
+	memset(skb->data, 0, chip->tx_pkt_desc_sz);
 	rtw_tx_fill_tx_desc(pkt_info, skb);
 	rtw_tx_fill_txdesc_checksum(rtwdev, pkt_info, skb->data);
 
-- 
cgit v1.2.3-58-ga151


From 28818b4d871bc93cc4f5c7c7d7c526a6a096c09c Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Mon, 29 Apr 2024 20:57:52 +0300
Subject: wifi: rtw88: usb: Fix disconnection after beacon loss

When there is beacon loss, for example due to unrelated Bluetooth
devices transmitting music nearby, the wifi connection dies soon
after the first beacon loss message:

Apr 28 20:47:14 ideapad2 wpa_supplicant[1161]: wlp3s0f3u4:
 CTRL-EVENT-BEACON-LOSS
Apr 28 20:47:15 ideapad2 wpa_supplicant[1161]: wlp3s0f3u4:
 CTRL-EVENT-DISCONNECTED bssid=... reason=4 locally_generated=1

Apr 28 20:47:24 ideapad2 wpa_supplicant[1161]: wlp3s0f3u4:
 CTRL-EVENT-BEACON-LOSS
Apr 28 20:47:25 ideapad2 wpa_supplicant[1161]: wlp3s0f3u4:
 CTRL-EVENT-DISCONNECTED bssid=... reason=4 locally_generated=1

Apr 28 20:47:34 ideapad2 wpa_supplicant[1161]: wlp3s0f3u4:
 CTRL-EVENT-BEACON-LOSS
Apr 28 20:47:35 ideapad2 wpa_supplicant[1161]: wlp3s0f3u4:
 CTRL-EVENT-DISCONNECTED bssid=... reason=4 locally_generated=1

When the beacon loss happens, mac80211 makes rtw88 transmit a QOS
NULL frame and asks to confirm the ACK status. Even though rtw88
confirms to mac80211 that the QOS NULL was transmitted successfully,
the connection still dies. This is because rtw88 is handing the QOS
NULL back to mac80211 with skb->data pointing to the headroom (the
TX descriptor) instead of ieee80211_hdr.

Fix the disconnection by moving skb->data to the correct position
before ieee80211_tx_status_irqsafe().

The problem was observed with RTL8811AU (TP-Link Archer T2U Nano)
and the potential future rtw88_8821au driver. Also tested with
RTL8811CU (Tenda U9).

Cc: stable@vger.kernel.org
Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/ecbf0601-810d-4609-b8fc-8b0e38d2948d@gmail.com
---
 drivers/net/wireless/realtek/rtw88/usb.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw88/usb.c b/drivers/net/wireless/realtek/rtw88/usb.c
index 90afeefe002f..d204d138afe2 100644
--- a/drivers/net/wireless/realtek/rtw88/usb.c
+++ b/drivers/net/wireless/realtek/rtw88/usb.c
@@ -273,6 +273,8 @@ static void rtw_usb_write_port_tx_complete(struct urb *urb)
 		info = IEEE80211_SKB_CB(skb);
 		tx_data = rtw_usb_get_tx_data(skb);
 
+		skb_pull(skb, rtwdev->chip->tx_pkt_desc_sz);
+
 		/* enqueue to wait for tx report */
 		if (info->flags & IEEE80211_TX_CTL_REQ_TX_STATUS) {
 			rtw_tx_report_enqueue(rtwdev, skb, tx_data->sn);
-- 
cgit v1.2.3-58-ga151


From bb0b0a6b96e6de854cb1e349e17bd0e8bf421a59 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Sat, 4 May 2024 11:52:08 -0700
Subject: wifi: ath12k: initialize 'ret' in ath12k_qmi_load_file_target_mem()

smatch flagged the following issue:

drivers/net/wireless/ath/ath12k/qmi.c:2619 ath12k_qmi_load_file_target_mem() error: uninitialized symbol 'ret'.

The reality is that 'ret' is initialized in every path through
ath12k_qmi_load_file_target_mem() except one, the case where the input
'len' is 0, and hence the "while (remaining)" loop is never entered.
But to make sure this case is also handled, add an initializer to the
declaration of 'ret'.

No functional changes, compile tested only.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240504-qmi_load_file_target_mem-v1-1-069fc44c45eb@quicinc.com
---
 drivers/net/wireless/ath/ath12k/qmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/qmi.c b/drivers/net/wireless/ath/ath12k/qmi.c
index 3bb730326a02..f273cb93dedb 100644
--- a/drivers/net/wireless/ath/ath12k/qmi.c
+++ b/drivers/net/wireless/ath/ath12k/qmi.c
@@ -2538,7 +2538,7 @@ static int ath12k_qmi_load_file_target_mem(struct ath12k_base *ab,
 	struct qmi_wlanfw_bdf_download_resp_msg_v01 resp = {};
 	struct qmi_txn txn;
 	const u8 *temp = data;
-	int ret;
+	int ret = 0;
 	u32 remaining = len;
 
 	req = kzalloc(sizeof(*req), GFP_KERNEL);
-- 
cgit v1.2.3-58-ga151


From 199f149e97dc7be80e5eed4b232529c1d1aa8055 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Sat, 4 May 2024 11:52:09 -0700
Subject: wifi: ath11k: initialize 'ret' in ath11k_qmi_load_file_target_mem()

smatch flagged the following issue:

drivers/net/wireless/ath/ath11k/qmi.c:2401 ath11k_qmi_load_file_target_mem() error: uninitialized symbol 'ret'.

The reality is that 'ret' is initialized in every path through
ath11k_qmi_load_file_target_mem() except one, the case where the input
'len' is 0, and hence the "while (remaining)" loop is never entered.
But to make sure this case is also handled, add an initializer to the
declaration of 'ret'.

No functional changes, compile tested only.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240504-qmi_load_file_target_mem-v1-2-069fc44c45eb@quicinc.com
---
 drivers/net/wireless/ath/ath11k/qmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/qmi.c b/drivers/net/wireless/ath/ath11k/qmi.c
index d4a243b64f6c..aa160e6fe24f 100644
--- a/drivers/net/wireless/ath/ath11k/qmi.c
+++ b/drivers/net/wireless/ath/ath11k/qmi.c
@@ -2293,7 +2293,7 @@ static int ath11k_qmi_load_file_target_mem(struct ath11k_base *ab,
 	struct qmi_txn txn;
 	const u8 *temp = data;
 	void __iomem *bdf_addr = NULL;
-	int ret;
+	int ret = 0;
 	u32 remaining = len;
 
 	req = kzalloc(sizeof(*req), GFP_KERNEL);
-- 
cgit v1.2.3-58-ga151


From 2db7a82fd150193849d04592e205b3d178207840 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Mon, 6 May 2024 23:00:17 +0530
Subject: wifi: ath12k: avoid double SW2HW_MACID conversion

Currently, ath12k_wmi_pdev_dma_ring_cfg() fails due to an invalid pdev_id
parameter passed to the WMI_PDEV_DMA_RING_CFG_REQ_CMDID WMI command. This
invalid pdev_id is caused by a double conversion of the MAC ID. Since the
the caller of ath12k_wmi_pdev_dma_ring_cfg() already performs the MAC ID
conversion, it is unnecessary to do it again within the function. To fix
this, remove the software (SW) to hardware (HW) MAC ID conversion from
ath12k_wmi_pdev_dma_ring_cfg() to avoid this redundant conversion.

Found in code review.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240506173017.597715-1-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/wmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 7a52d2082b79..9e9f58b5cbfb 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -3808,7 +3808,7 @@ int ath12k_wmi_pdev_dma_ring_cfg(struct ath12k *ar,
 	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_DMA_RING_CFG_REQ,
 						 sizeof(*cmd));
 
-	cmd->pdev_id = cpu_to_le32(DP_SW2HW_MACID(arg->pdev_id));
+	cmd->pdev_id = cpu_to_le32(arg->pdev_id);
 	cmd->module_id = cpu_to_le32(arg->module_id);
 	cmd->base_paddr_lo = cpu_to_le32(arg->base_paddr_lo);
 	cmd->base_paddr_hi = cpu_to_le32(arg->base_paddr_hi);
-- 
cgit v1.2.3-58-ga151


From a6efa4ea06138d04b50ed5a7954a307a91a33478 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Sat, 4 May 2024 14:36:35 -0700
Subject: wifi: ath11k: fix misspelling of "dma" in num_rxmda_per_pdev

The ath11k_hw_params member num_rxmda_per_pdev contains a misspelling
of "dma" so fix it.

No functional changes, compile tested only.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240504-ath11k-num_rxmda_per_pdev-v1-1-7f97bced8495@quicinc.com
---
 drivers/net/wireless/ath/ath11k/core.c    | 20 ++++++++++----------
 drivers/net/wireless/ath/ath11k/debugfs.c |  6 +++---
 drivers/net/wireless/ath/ath11k/dp.c      | 12 ++++++------
 drivers/net/wireless/ath/ath11k/dp_rx.c   | 22 +++++++++++-----------
 drivers/net/wireless/ath/ath11k/dp_tx.c   |  6 +++---
 drivers/net/wireless/ath/ath11k/hw.h      |  4 ++--
 drivers/net/wireless/ath/ath11k/mac.c     |  2 +-
 drivers/net/wireless/ath/ath11k/reg.c     |  4 ++--
 drivers/net/wireless/ath/ath11k/wmi.c     |  2 +-
 9 files changed, 39 insertions(+), 39 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/core.c b/drivers/net/wireless/ath/ath11k/core.c
index 3cc817a3b4a4..d4ef556852c2 100644
--- a/drivers/net/wireless/ath/ath11k/core.c
+++ b/drivers/net/wireless/ath/ath11k/core.c
@@ -62,7 +62,7 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.ce_ie_addr = &ath11k_ce_ie_addr_ipq8074,
 		.single_pdev_only = false,
 		.rxdma1_enable = true,
-		.num_rxmda_per_pdev = 1,
+		.num_rxdma_per_pdev = 1,
 		.rx_mac_buf_ring = false,
 		.vdev_start_delay = false,
 		.htt_peer_map_v2 = true,
@@ -148,7 +148,7 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.ce_ie_addr = &ath11k_ce_ie_addr_ipq8074,
 		.single_pdev_only = false,
 		.rxdma1_enable = true,
-		.num_rxmda_per_pdev = 1,
+		.num_rxdma_per_pdev = 1,
 		.rx_mac_buf_ring = false,
 		.vdev_start_delay = false,
 		.htt_peer_map_v2 = true,
@@ -232,7 +232,7 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.ce_ie_addr = &ath11k_ce_ie_addr_ipq8074,
 		.single_pdev_only = true,
 		.rxdma1_enable = false,
-		.num_rxmda_per_pdev = 2,
+		.num_rxdma_per_pdev = 2,
 		.rx_mac_buf_ring = true,
 		.vdev_start_delay = true,
 		.htt_peer_map_v2 = false,
@@ -320,7 +320,7 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.svc_to_ce_map_len = 18,
 		.ce_ie_addr = &ath11k_ce_ie_addr_ipq8074,
 		.rxdma1_enable = true,
-		.num_rxmda_per_pdev = 1,
+		.num_rxdma_per_pdev = 1,
 		.rx_mac_buf_ring = false,
 		.vdev_start_delay = false,
 		.htt_peer_map_v2 = true,
@@ -404,7 +404,7 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.ce_ie_addr = &ath11k_ce_ie_addr_ipq8074,
 		.single_pdev_only = true,
 		.rxdma1_enable = false,
-		.num_rxmda_per_pdev = 2,
+		.num_rxdma_per_pdev = 2,
 		.rx_mac_buf_ring = true,
 		.vdev_start_delay = true,
 		.htt_peer_map_v2 = false,
@@ -492,7 +492,7 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.svc_to_ce_map_len = 14,
 		.single_pdev_only = true,
 		.rxdma1_enable = false,
-		.num_rxmda_per_pdev = 2,
+		.num_rxdma_per_pdev = 2,
 		.rx_mac_buf_ring = true,
 		.vdev_start_delay = true,
 		.htt_peer_map_v2 = false,
@@ -580,7 +580,7 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.ce_ie_addr = &ath11k_ce_ie_addr_ipq8074,
 		.single_pdev_only = true,
 		.rxdma1_enable = false,
-		.num_rxmda_per_pdev = 1,
+		.num_rxdma_per_pdev = 1,
 		.rx_mac_buf_ring = true,
 		.vdev_start_delay = true,
 		.htt_peer_map_v2 = false,
@@ -673,7 +673,7 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.ce_ie_addr = &ath11k_ce_ie_addr_ipq5018,
 		.ce_remap = &ath11k_ce_remap_ipq5018,
 		.rxdma1_enable = true,
-		.num_rxmda_per_pdev = RXDMA_PER_PDEV_5018,
+		.num_rxdma_per_pdev = RXDMA_PER_PDEV_5018,
 		.rx_mac_buf_ring = false,
 		.vdev_start_delay = false,
 		.htt_peer_map_v2 = true,
@@ -744,7 +744,7 @@ static const struct ath11k_hw_params ath11k_hw_params[] = {
 		.ce_ie_addr = &ath11k_ce_ie_addr_ipq8074,
 		.single_pdev_only = true,
 		.rxdma1_enable = false,
-		.num_rxmda_per_pdev = 2,
+		.num_rxdma_per_pdev = 2,
 		.rx_mac_buf_ring = true,
 		.vdev_start_delay = true,
 		.htt_peer_map_v2 = false,
@@ -1801,7 +1801,7 @@ static int ath11k_core_start(struct ath11k_base *ab)
 	}
 
 	/* put hardware to DBS mode */
-	if (ab->hw_params.single_pdev_only && ab->hw_params.num_rxmda_per_pdev > 1) {
+	if (ab->hw_params.single_pdev_only && ab->hw_params.num_rxdma_per_pdev > 1) {
 		ret = ath11k_wmi_set_hw_mode(ab, WMI_HOST_HW_MODE_DBS);
 		if (ret) {
 			ath11k_err(ab, "failed to send dbs mode: %d\n", ret);
diff --git a/drivers/net/wireless/ath/ath11k/debugfs.c b/drivers/net/wireless/ath/ath11k/debugfs.c
index 414a5ce279f7..57281a135dd7 100644
--- a/drivers/net/wireless/ath/ath11k/debugfs.c
+++ b/drivers/net/wireless/ath/ath11k/debugfs.c
@@ -668,7 +668,7 @@ static ssize_t ath11k_write_extd_rx_stats(struct file *file,
 
 	ar->debug.rx_filter = tlv_filter.rx_filter;
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		ring_id = ar->dp.rx_mon_status_refill_ring[i].refill_buf_ring.ring_id;
 		ret = ath11k_dp_tx_htt_rx_filter_setup(ar->ab, ring_id, ar->dp.mac_id,
 						       HAL_RXDMA_MONITOR_STATUS,
@@ -1112,7 +1112,7 @@ static ssize_t ath11k_write_pktlog_filter(struct file *file,
 	}
 
 	/* Clear rx filter set for monitor mode and rx status */
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		ring_id = ar->dp.rx_mon_status_refill_ring[i].refill_buf_ring.ring_id;
 		ret = ath11k_dp_tx_htt_rx_filter_setup(ar->ab, ring_id, ar->dp.mac_id,
 						       HAL_RXDMA_MONITOR_STATUS,
@@ -1171,7 +1171,7 @@ static ssize_t ath11k_write_pktlog_filter(struct file *file,
 					       HTT_RX_FP_DATA_FILTER_FLASG3;
 	}
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		ring_id = ar->dp.rx_mon_status_refill_ring[i].refill_buf_ring.ring_id;
 		ret = ath11k_dp_tx_htt_rx_filter_setup(ab, ring_id,
 						       ar->dp.mac_id + i,
diff --git a/drivers/net/wireless/ath/ath11k/dp.c b/drivers/net/wireless/ath/ath11k/dp.c
index 1a62407e5a9f..fbf666d0ecf1 100644
--- a/drivers/net/wireless/ath/ath11k/dp.c
+++ b/drivers/net/wireless/ath/ath11k/dp.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: BSD-3-Clause-Clear
 /*
  * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
- * Copyright (c) 2021-2023 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2021-2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #include <crypto/hash.h>
@@ -830,8 +830,8 @@ int ath11k_dp_service_srng(struct ath11k_base *ab,
 
 	if (ab->hw_params.ring_mask->rx_mon_status[grp_id]) {
 		for (i = 0; i < ab->num_radios; i++) {
-			for (j = 0; j < ab->hw_params.num_rxmda_per_pdev; j++) {
-				int id = i * ab->hw_params.num_rxmda_per_pdev + j;
+			for (j = 0; j < ab->hw_params.num_rxdma_per_pdev; j++) {
+				int id = i * ab->hw_params.num_rxdma_per_pdev + j;
 
 				if (ab->hw_params.ring_mask->rx_mon_status[grp_id] &
 					BIT(id)) {
@@ -853,8 +853,8 @@ int ath11k_dp_service_srng(struct ath11k_base *ab,
 		ath11k_dp_process_reo_status(ab);
 
 	for (i = 0; i < ab->num_radios; i++) {
-		for (j = 0; j < ab->hw_params.num_rxmda_per_pdev; j++) {
-			int id = i * ab->hw_params.num_rxmda_per_pdev + j;
+		for (j = 0; j < ab->hw_params.num_rxdma_per_pdev; j++) {
+			int id = i * ab->hw_params.num_rxdma_per_pdev + j;
 
 			if (ab->hw_params.ring_mask->rxdma2host[grp_id] & BIT(id)) {
 				work_done = ath11k_dp_process_rxdma_err(ab, id, budget);
@@ -913,7 +913,7 @@ void ath11k_dp_pdev_pre_alloc(struct ath11k_base *ab)
 		spin_lock_init(&dp->rx_refill_buf_ring.idr_lock);
 		atomic_set(&dp->num_tx_pending, 0);
 		init_waitqueue_head(&dp->tx_empty_waitq);
-		for (j = 0; j < ab->hw_params.num_rxmda_per_pdev; j++) {
+		for (j = 0; j < ab->hw_params.num_rxdma_per_pdev; j++) {
 			idr_init(&dp->rx_mon_status_refill_ring[j].bufs_idr);
 			spin_lock_init(&dp->rx_mon_status_refill_ring[j].idr_lock);
 		}
diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 8a860a17d7e4..198fb359a688 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -311,7 +311,7 @@ static void ath11k_dp_service_mon_ring(struct timer_list *t)
 	struct ath11k_base *ab = from_timer(ab, t, mon_reap_timer);
 	int i;
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++)
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++)
 		ath11k_dp_rx_process_mon_rings(ab, i, NULL, DP_MON_SERVICE_BUDGET);
 
 	mod_timer(&ab->mon_reap_timer, jiffies +
@@ -324,7 +324,7 @@ static int ath11k_dp_purge_mon_ring(struct ath11k_base *ab)
 	unsigned long timeout = jiffies + msecs_to_jiffies(DP_MON_PURGE_TIMEOUT_MS);
 
 	do {
-		for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++)
+		for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++)
 			reaped += ath11k_dp_rx_process_mon_rings(ab, i,
 								 NULL,
 								 DP_MON_SERVICE_BUDGET);
@@ -468,7 +468,7 @@ static int ath11k_dp_rxdma_pdev_buf_free(struct ath11k *ar)
 	rx_ring = &dp->rxdma_mon_buf_ring;
 	ath11k_dp_rxdma_buf_ring_free(ar, rx_ring);
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		rx_ring = &dp->rx_mon_status_refill_ring[i];
 		ath11k_dp_rxdma_buf_ring_free(ar, rx_ring);
 	}
@@ -506,7 +506,7 @@ static int ath11k_dp_rxdma_pdev_buf_setup(struct ath11k *ar)
 		ath11k_dp_rxdma_ring_buf_setup(ar, rx_ring, HAL_RXDMA_MONITOR_BUF);
 	}
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		rx_ring = &dp->rx_mon_status_refill_ring[i];
 		ath11k_dp_rxdma_ring_buf_setup(ar, rx_ring, HAL_RXDMA_MONITOR_STATUS);
 	}
@@ -522,7 +522,7 @@ static void ath11k_dp_rx_pdev_srng_free(struct ath11k *ar)
 
 	ath11k_dp_srng_cleanup(ab, &dp->rx_refill_buf_ring.refill_buf_ring);
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		if (ab->hw_params.rx_mac_buf_ring)
 			ath11k_dp_srng_cleanup(ab, &dp->rx_mac_buf_ring[i]);
 
@@ -585,7 +585,7 @@ static int ath11k_dp_rx_pdev_srng_alloc(struct ath11k *ar)
 	}
 
 	if (ar->ab->hw_params.rx_mac_buf_ring) {
-		for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+		for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 			ret = ath11k_dp_srng_setup(ar->ab,
 						   &dp->rx_mac_buf_ring[i],
 						   HAL_RXDMA_BUF, 1,
@@ -598,7 +598,7 @@ static int ath11k_dp_rx_pdev_srng_alloc(struct ath11k *ar)
 		}
 	}
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		ret = ath11k_dp_srng_setup(ar->ab, &dp->rxdma_err_dst_ring[i],
 					   HAL_RXDMA_DST, 0, dp->mac_id + i,
 					   DP_RXDMA_ERR_DST_RING_SIZE);
@@ -608,7 +608,7 @@ static int ath11k_dp_rx_pdev_srng_alloc(struct ath11k *ar)
 		}
 	}
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		srng = &dp->rx_mon_status_refill_ring[i].refill_buf_ring;
 		ret = ath11k_dp_srng_setup(ar->ab,
 					   srng,
@@ -4455,7 +4455,7 @@ int ath11k_dp_rx_pdev_alloc(struct ath11k_base *ab, int mac_id)
 	}
 
 	if (ab->hw_params.rx_mac_buf_ring) {
-		for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+		for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 			ring_id = dp->rx_mac_buf_ring[i].ring_id;
 			ret = ath11k_dp_tx_htt_srng_setup(ab, ring_id,
 							  mac_id + i, HAL_RXDMA_BUF);
@@ -4467,7 +4467,7 @@ int ath11k_dp_rx_pdev_alloc(struct ath11k_base *ab, int mac_id)
 		}
 	}
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		ring_id = dp->rxdma_err_dst_ring[i].ring_id;
 		ret = ath11k_dp_tx_htt_srng_setup(ab, ring_id,
 						  mac_id + i, HAL_RXDMA_DST);
@@ -4507,7 +4507,7 @@ int ath11k_dp_rx_pdev_alloc(struct ath11k_base *ab, int mac_id)
 	}
 
 config_refill_ring:
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		ring_id = dp->rx_mon_status_refill_ring[i].refill_buf_ring.ring_id;
 		ret = ath11k_dp_tx_htt_srng_setup(ab, ring_id, mac_id + i,
 						  HAL_RXDMA_MONITOR_STATUS);
diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
index 272b1c35f98d..21819b741701 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -1035,7 +1035,7 @@ int ath11k_dp_tx_htt_h2t_ppdu_stats_req(struct ath11k *ar, u32 mask)
 	int ret;
 	int i;
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		skb = ath11k_htc_alloc_skb(ab, len);
 		if (!skb)
 			return -ENOMEM;
@@ -1218,7 +1218,7 @@ int ath11k_dp_tx_htt_monitor_mode_ring_config(struct ath11k *ar, bool reset)
 						       &tlv_filter);
 	} else if (!reset) {
 		/* set in monitor mode only */
-		for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+		for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 			ring_id = dp->rx_mac_buf_ring[i].ring_id;
 			ret = ath11k_dp_tx_htt_rx_filter_setup(ar->ab, ring_id,
 							       dp->mac_id + i,
@@ -1231,7 +1231,7 @@ int ath11k_dp_tx_htt_monitor_mode_ring_config(struct ath11k *ar, bool reset)
 	if (ret)
 		return ret;
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		ring_id = dp->rx_mon_status_refill_ring[i].refill_buf_ring.ring_id;
 		if (!reset) {
 			tlv_filter.rx_filter =
diff --git a/drivers/net/wireless/ath/ath11k/hw.h b/drivers/net/wireless/ath/ath11k/hw.h
index 14ef4eb48f80..300322535766 100644
--- a/drivers/net/wireless/ath/ath11k/hw.h
+++ b/drivers/net/wireless/ath/ath11k/hw.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause-Clear */
 /*
  * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
- * Copyright (c) 2021-2023 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2021-2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #ifndef ATH11K_HW_H
@@ -167,7 +167,7 @@ struct ath11k_hw_params {
 	bool single_pdev_only;
 
 	bool rxdma1_enable;
-	int num_rxmda_per_pdev;
+	int num_rxdma_per_pdev;
 	bool rx_mac_buf_ring;
 	bool vdev_start_delay;
 	bool htt_peer_map_v2;
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index a71195344288..17b06c0cd062 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -6108,7 +6108,7 @@ static int ath11k_mac_config_mon_status_default(struct ath11k *ar, bool enable)
 			tlv_filter.rx_filter = ath11k_debugfs_rx_filter(ar);
 	}
 
-	for (i = 0; i < ab->hw_params.num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params.num_rxdma_per_pdev; i++) {
 		ring_id = ar->dp.rx_mon_status_refill_ring[i].refill_buf_ring.ring_id;
 		ret = ath11k_dp_tx_htt_rx_filter_setup(ar->ab, ring_id,
 						       ar->dp.mac_id + i,
diff --git a/drivers/net/wireless/ath/ath11k/reg.c b/drivers/net/wireless/ath/ath11k/reg.c
index 737fcd450d4b..4c3aefbbabbe 100644
--- a/drivers/net/wireless/ath/ath11k/reg.c
+++ b/drivers/net/wireless/ath/ath11k/reg.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: BSD-3-Clause-Clear
 /*
  * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
- * Copyright (c) 2021-2023 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2021-2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 #include <linux/rtnetlink.h>
 
@@ -878,7 +878,7 @@ int ath11k_reg_handle_chan_list(struct ath11k_base *ab,
 		ath11k_reg_reset_info(reg_info);
 
 		if (ab->hw_params.single_pdev_only &&
-		    pdev_idx < ab->hw_params.num_rxmda_per_pdev)
+		    pdev_idx < ab->hw_params.num_rxdma_per_pdev)
 			return 0;
 		goto fallback;
 	}
diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
index 6ff01c45f165..38f175dd1557 100644
--- a/drivers/net/wireless/ath/ath11k/wmi.c
+++ b/drivers/net/wireless/ath/ath11k/wmi.c
@@ -9082,7 +9082,7 @@ int ath11k_wmi_attach(struct ath11k_base *ab)
 	ab->wmi_ab.preferred_hw_mode = WMI_HOST_HW_MODE_MAX;
 
 	/* It's overwritten when service_ext_ready is handled */
-	if (ab->hw_params.single_pdev_only && ab->hw_params.num_rxmda_per_pdev > 1)
+	if (ab->hw_params.single_pdev_only && ab->hw_params.num_rxdma_per_pdev > 1)
 		ab->wmi_ab.preferred_hw_mode = WMI_HOST_HW_MODE_SINGLE;
 
 	/* TODO: Init remaining wmi soc resources required */
-- 
cgit v1.2.3-58-ga151


From ff5fa958d2fd45de7282a8bd3968426c947591d5 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Sat, 4 May 2024 14:45:17 -0700
Subject: wifi: ath12k: fix misspelling of "dma" in num_rxmda_per_pdev

The ath12k_hw_params member num_rxmda_per_pdev contains a misspelling
of "dma" so fix it.

No functional changes, compile tested only.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240504-ath12k_dp_rxdma_ring_sel_config_wcn7850-ret-v1-1-44d2843a2857@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp.c    | 10 +++++-----
 drivers/net/wireless/ath/ath12k/dp_rx.c | 14 +++++++-------
 drivers/net/wireless/ath/ath12k/dp_tx.c |  2 +-
 drivers/net/wireless/ath/ath12k/hw.c    |  6 +++---
 drivers/net/wireless/ath/ath12k/hw.h    |  2 +-
 drivers/net/wireless/ath/ath12k/wmi.c   |  2 +-
 6 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp.c b/drivers/net/wireless/ath/ath12k/dp.c
index ca41f4431394..412ab08e6cca 100644
--- a/drivers/net/wireless/ath/ath12k/dp.c
+++ b/drivers/net/wireless/ath/ath12k/dp.c
@@ -907,8 +907,8 @@ int ath12k_dp_service_srng(struct ath12k_base *ab,
 		monitor_mode = ATH12K_DP_RX_MONITOR_MODE;
 		ring_mask = ab->hw_params->ring_mask->rx_mon_dest[grp_id];
 		for (i = 0; i < ab->num_radios; i++) {
-			for (j = 0; j < ab->hw_params->num_rxmda_per_pdev; j++) {
-				int id = i * ab->hw_params->num_rxmda_per_pdev + j;
+			for (j = 0; j < ab->hw_params->num_rxdma_per_pdev; j++) {
+				int id = i * ab->hw_params->num_rxdma_per_pdev + j;
 
 				if (ring_mask & BIT(id)) {
 					work_done =
@@ -928,8 +928,8 @@ int ath12k_dp_service_srng(struct ath12k_base *ab,
 		monitor_mode = ATH12K_DP_TX_MONITOR_MODE;
 		ring_mask = ab->hw_params->ring_mask->tx_mon_dest[grp_id];
 		for (i = 0; i < ab->num_radios; i++) {
-			for (j = 0; j < ab->hw_params->num_rxmda_per_pdev; j++) {
-				int id = i * ab->hw_params->num_rxmda_per_pdev + j;
+			for (j = 0; j < ab->hw_params->num_rxdma_per_pdev; j++) {
+				int id = i * ab->hw_params->num_rxdma_per_pdev + j;
 
 				if (ring_mask & BIT(id)) {
 					work_done =
@@ -1017,7 +1017,7 @@ static void ath12k_dp_service_mon_ring(struct timer_list *t)
 	struct ath12k_base *ab = from_timer(ab, t, mon_reap_timer);
 	int i;
 
-	for (i = 0; i < ab->hw_params->num_rxmda_per_pdev; i++)
+	for (i = 0; i < ab->hw_params->num_rxdma_per_pdev; i++)
 		ath12k_dp_mon_process_ring(ab, i, NULL, DP_MON_SERVICE_BUDGET,
 					   ATH12K_DP_RX_MONITOR_MODE);
 
diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index 952a377b5491..e3ff14cd283f 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -485,7 +485,7 @@ static void ath12k_dp_rx_pdev_srng_free(struct ath12k *ar)
 	struct ath12k_base *ab = ar->ab;
 	int i;
 
-	for (i = 0; i < ab->hw_params->num_rxmda_per_pdev; i++)
+	for (i = 0; i < ab->hw_params->num_rxdma_per_pdev; i++)
 		ath12k_dp_srng_cleanup(ab, &dp->rxdma_mon_dst_ring[i]);
 }
 
@@ -530,7 +530,7 @@ static int ath12k_dp_rx_pdev_srng_alloc(struct ath12k *ar)
 	int ret;
 	u32 mac_id = dp->mac_id;
 
-	for (i = 0; i < ab->hw_params->num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params->num_rxdma_per_pdev; i++) {
 		ret = ath12k_dp_srng_setup(ar->ab,
 					   &dp->rxdma_mon_dst_ring[i],
 					   HAL_RXDMA_MONITOR_DST,
@@ -3938,7 +3938,7 @@ void ath12k_dp_rx_free(struct ath12k_base *ab)
 
 	ath12k_dp_srng_cleanup(ab, &dp->rx_refill_buf_ring.refill_buf_ring);
 
-	for (i = 0; i < ab->hw_params->num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params->num_rxdma_per_pdev; i++) {
 		if (ab->hw_params->rx_mac_buf_ring)
 			ath12k_dp_srng_cleanup(ab, &dp->rx_mac_buf_ring[i]);
 	}
@@ -4030,7 +4030,7 @@ int ath12k_dp_rxdma_ring_sel_config_wcn7850(struct ath12k_base *ab)
 	 * and modify the rx_desc struct
 	 */
 
-	for (i = 0; i < ab->hw_params->num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params->num_rxdma_per_pdev; i++) {
 		ring_id = dp->rx_mac_buf_ring[i].ring_id;
 		ret = ath12k_dp_tx_htt_rx_filter_setup(ab, ring_id, i,
 						       HAL_RXDMA_BUF,
@@ -4057,7 +4057,7 @@ int ath12k_dp_rx_htt_setup(struct ath12k_base *ab)
 	}
 
 	if (ab->hw_params->rx_mac_buf_ring) {
-		for (i = 0; i < ab->hw_params->num_rxmda_per_pdev; i++) {
+		for (i = 0; i < ab->hw_params->num_rxdma_per_pdev; i++) {
 			ring_id = dp->rx_mac_buf_ring[i].ring_id;
 			ret = ath12k_dp_tx_htt_srng_setup(ab, ring_id,
 							  i, HAL_RXDMA_BUF);
@@ -4118,7 +4118,7 @@ int ath12k_dp_rx_alloc(struct ath12k_base *ab)
 	}
 
 	if (ab->hw_params->rx_mac_buf_ring) {
-		for (i = 0; i < ab->hw_params->num_rxmda_per_pdev; i++) {
+		for (i = 0; i < ab->hw_params->num_rxdma_per_pdev; i++) {
 			ret = ath12k_dp_srng_setup(ab,
 						   &dp->rx_mac_buf_ring[i],
 						   HAL_RXDMA_BUF, 1,
@@ -4178,7 +4178,7 @@ int ath12k_dp_rx_pdev_alloc(struct ath12k_base *ab, int mac_id)
 		return ret;
 	}
 
-	for (i = 0; i < ab->hw_params->num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params->num_rxdma_per_pdev; i++) {
 		ring_id = dp->rxdma_mon_dst_ring[i].ring_id;
 		ret = ath12k_dp_tx_htt_srng_setup(ab, ring_id,
 						  mac_id + i,
diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.c b/drivers/net/wireless/ath/ath12k/dp_tx.c
index a22fa43c87ec..32496015cf9a 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.c
@@ -865,7 +865,7 @@ int ath12k_dp_tx_htt_h2t_ppdu_stats_req(struct ath12k *ar, u32 mask)
 	int ret;
 	int i;
 
-	for (i = 0; i < ab->hw_params->num_rxmda_per_pdev; i++) {
+	for (i = 0; i < ab->hw_params->num_rxdma_per_pdev; i++) {
 		skb = ath12k_htc_alloc_skb(ab, len);
 		if (!skb)
 			return -ENOMEM;
diff --git a/drivers/net/wireless/ath/ath12k/hw.c b/drivers/net/wireless/ath/ath12k/hw.c
index f4c827015821..06f443216488 100644
--- a/drivers/net/wireless/ath/ath12k/hw.c
+++ b/drivers/net/wireless/ath/ath12k/hw.c
@@ -884,7 +884,7 @@ static const struct ath12k_hw_params ath12k_hw_params[] = {
 		.hal_params = &ath12k_hw_hal_params_qcn9274,
 
 		.rxdma1_enable = false,
-		.num_rxmda_per_pdev = 1,
+		.num_rxdma_per_pdev = 1,
 		.num_rxdma_dst_ring = 0,
 		.rx_mac_buf_ring = false,
 		.vdev_start_delay = false,
@@ -956,7 +956,7 @@ static const struct ath12k_hw_params ath12k_hw_params[] = {
 		.hal_params = &ath12k_hw_hal_params_wcn7850,
 
 		.rxdma1_enable = false,
-		.num_rxmda_per_pdev = 2,
+		.num_rxdma_per_pdev = 2,
 		.num_rxdma_dst_ring = 1,
 		.rx_mac_buf_ring = true,
 		.vdev_start_delay = true,
@@ -1029,7 +1029,7 @@ static const struct ath12k_hw_params ath12k_hw_params[] = {
 		.hal_params = &ath12k_hw_hal_params_qcn9274,
 
 		.rxdma1_enable = false,
-		.num_rxmda_per_pdev = 1,
+		.num_rxdma_per_pdev = 1,
 		.num_rxdma_dst_ring = 0,
 		.rx_mac_buf_ring = false,
 		.vdev_start_delay = false,
diff --git a/drivers/net/wireless/ath/ath12k/hw.h b/drivers/net/wireless/ath/ath12k/hw.h
index 3f450ee93f34..2a5a9d638d93 100644
--- a/drivers/net/wireless/ath/ath12k/hw.h
+++ b/drivers/net/wireless/ath/ath12k/hw.h
@@ -174,7 +174,7 @@ struct ath12k_hw_params {
 	const struct ath12k_hw_hal_params *hal_params;
 
 	bool rxdma1_enable:1;
-	int num_rxmda_per_pdev;
+	int num_rxdma_per_pdev;
 	int num_rxdma_dst_ring;
 	bool rx_mac_buf_ring:1;
 	bool vdev_start_delay:1;
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 9e9f58b5cbfb..3004c7463be9 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -5693,7 +5693,7 @@ static int ath12k_reg_chan_list_event(struct ath12k_base *ab, struct sk_buff *sk
 		 * event. Otherwise, it goes to fallback.
 		 */
 		if (ab->hw_params->single_pdev_only &&
-		    pdev_idx < ab->hw_params->num_rxmda_per_pdev)
+		    pdev_idx < ab->hw_params->num_rxdma_per_pdev)
 			goto mem_free;
 		else
 			goto fallback;
-- 
cgit v1.2.3-58-ga151


From 3b9344740843d965e9e37fba30620b3b1c0afa4f Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Tue, 7 May 2024 13:45:03 +0300
Subject: wifi: ath12k: initialize 'ret' in
 ath12k_dp_rxdma_ring_sel_config_wcn7850()

smatch flagged the following issue:

drivers/net/wireless/ath/ath12k/dp_rx.c:4065 ath12k_dp_rxdma_ring_sel_config_wcn7850() error: uninitialized symbol 'ret'.

In ath12k_dp_rxdma_ring_sel_config_wcn7850() if it were ever the case
that ab->hw_params->num_rxdma_per_pdev was 0 then 'ret' would be
uninitialized when it is returned. This should never be the case, but
to be safe and to quiet smatch, add an initializer to the declaration
of 'ret'.

No functional changes, compile tested only.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240504-ath12k_dp_rxdma_ring_sel_config_wcn7850-ret-v1-2-44d2843a2857@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_rx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index e3ff14cd283f..37205e894afe 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -4004,7 +4004,7 @@ int ath12k_dp_rxdma_ring_sel_config_wcn7850(struct ath12k_base *ab)
 	struct ath12k_dp *dp = &ab->dp;
 	struct htt_rx_ring_tlv_filter tlv_filter = {0};
 	u32 ring_id;
-	int ret;
+	int ret = 0;
 	u32 hal_rx_desc_sz = ab->hal.hal_desc_sz;
 	int i;
 
-- 
cgit v1.2.3-58-ga151


From 71b6e321e30271beb08772871c4f76777f49e402 Mon Sep 17 00:00:00 2001
From: Marc Gonzalez <mgonzalez@freebox.fr>
Date: Mon, 29 Apr 2024 16:04:51 +0200
Subject: dt-bindings: net: wireless: ath10k: add qcom,no-msa-ready-indicator
 prop

The ath10k driver waits for an "MSA_READY" indicator
to complete initialization. If the indicator is not
received, then the device remains unusable.

cf. ath10k_qmi_driver_event_work()

Several msm8998-based devices are affected by this issue.
Oddly, it seems safe to NOT wait for the indicator, and
proceed immediately when QMI_EVENT_SERVER_ARRIVE.

Jeff Johnson wrote:

  The feedback I received was "it might be ok to change all ath10k qmi
  to skip waiting for msa_ready", and it was pointed out that ath11k
  (and ath12k) do not wait for it.

  However with so many deployed devices, "might be ok" isn't a strong
  argument for changing the default behavior.

Kalle Valo first suggested setting a bit in firmware-5.bin to trigger
work-around in the driver. However, firmware-5.bin is parsed too late.
So we are stuck with a DT property.

Signed-off-by: Marc Gonzalez <mgonzalez@freebox.fr>
Reviewed-by: Bjorn Andersson <quic_bjorande@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/54ac2295-36b4-49fc-9583-a10db8d9d5d6@freebox.fr
---
 Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml
index 5c4498b762c8..070c4c9b8643 100644
--- a/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml
+++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath10k.yaml
@@ -128,6 +128,11 @@ properties:
       Whether to skip executing an SCM call that reassigns the memory
       region ownership.
 
+  qcom,no-msa-ready-indicator:
+    type: boolean
+    description:
+      Don't wait for MSA_READY indicator to complete init.
+
   qcom,smem-states:
     $ref: /schemas/types.yaml#/definitions/phandle-array
     description: State bits used by the AP to signal the WLAN Q6.
-- 
cgit v1.2.3-58-ga151


From 6d67d18014a8130defbca1972d7f0e0a8f8f259e Mon Sep 17 00:00:00 2001
From: Marc Gonzalez <mgonzalez@freebox.fr>
Date: Mon, 29 Apr 2024 16:06:29 +0200
Subject: wifi: ath10k: do not always wait for MSA_READY indicator

The ath10k driver waits for an "MSA_READY" indicator
to complete initialization. If the indicator is not
received, then the device remains unusable.

Several msm8998-based devices are affected by this issue.
Oddly, it seems safe to NOT wait for the indicator, and
proceed immediately when QMI_EVENT_SERVER_ARRIVE.

fw_version 0x100204b2
fw_build_timestamp 2019-09-04 03:01
fw_build_id QC_IMAGE_VERSION_STRING=WLAN.HL.1.0-01202-QCAHLSWMTPLZ-1.221523.2

Jeff Johnson wrote:

  The feedback I received was "it might be ok to change all ath10k qmi
  to skip waiting for msa_ready", and it was pointed out that ath11k
  (and ath12k) do not wait for it.

  However with so many deployed devices, "might be ok" isn't a strong
  argument for changing the default behavior.

Signed-off-by: Marc Gonzalez <mgonzalez@freebox.fr>
Reviewed-by: Bjorn Andersson <quic_bjorande@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/23540303-5816-45d5-a1af-5f09d645a73b@freebox.fr
---
 drivers/net/wireless/ath/ath10k/qmi.c | 11 +++++++++++
 drivers/net/wireless/ath/ath10k/qmi.h |  1 +
 2 files changed, 12 insertions(+)

diff --git a/drivers/net/wireless/ath/ath10k/qmi.c b/drivers/net/wireless/ath/ath10k/qmi.c
index 38e939f572a9..f1f33af0170a 100644
--- a/drivers/net/wireless/ath/ath10k/qmi.c
+++ b/drivers/net/wireless/ath/ath10k/qmi.c
@@ -1040,6 +1040,10 @@ static void ath10k_qmi_driver_event_work(struct work_struct *work)
 		switch (event->type) {
 		case ATH10K_QMI_EVENT_SERVER_ARRIVE:
 			ath10k_qmi_event_server_arrive(qmi);
+			if (qmi->no_msa_ready_indicator) {
+				ath10k_info(ar, "qmi not waiting for msa_ready indicator");
+				ath10k_qmi_event_msa_ready(qmi);
+			}
 			break;
 		case ATH10K_QMI_EVENT_SERVER_EXIT:
 			ath10k_qmi_event_server_exit(qmi);
@@ -1048,6 +1052,10 @@ static void ath10k_qmi_driver_event_work(struct work_struct *work)
 			ath10k_qmi_event_fw_ready_ind(qmi);
 			break;
 		case ATH10K_QMI_EVENT_MSA_READY_IND:
+			if (qmi->no_msa_ready_indicator) {
+				ath10k_warn(ar, "qmi unexpected msa_ready indicator");
+				break;
+			}
 			ath10k_qmi_event_msa_ready(qmi);
 			break;
 		default:
@@ -1077,6 +1085,9 @@ int ath10k_qmi_init(struct ath10k *ar, u32 msa_size)
 	if (of_property_read_bool(dev->of_node, "qcom,msa-fixed-perm"))
 		qmi->msa_fixed_perm = true;
 
+	if (of_property_read_bool(dev->of_node, "qcom,no-msa-ready-indicator"))
+		qmi->no_msa_ready_indicator = true;
+
 	ret = qmi_handle_init(&qmi->qmi_hdl,
 			      WLFW_BDF_DOWNLOAD_REQ_MSG_V01_MAX_MSG_LEN,
 			      &ath10k_qmi_ops, qmi_msg_handler);
diff --git a/drivers/net/wireless/ath/ath10k/qmi.h b/drivers/net/wireless/ath/ath10k/qmi.h
index 89464239fe96..0816eb4e4a18 100644
--- a/drivers/net/wireless/ath/ath10k/qmi.h
+++ b/drivers/net/wireless/ath/ath10k/qmi.h
@@ -107,6 +107,7 @@ struct ath10k_qmi {
 	char fw_build_timestamp[MAX_TIMESTAMP_LEN + 1];
 	struct ath10k_qmi_cal_data cal_data[MAX_NUM_CAL_V01];
 	bool msa_fixed_perm;
+	bool no_msa_ready_indicator;
 	enum ath10k_qmi_state state;
 };
 
-- 
cgit v1.2.3-58-ga151


From 4f9206e8c2c15a3770e3eb0b9755a05d559fecea Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Wed, 8 May 2024 02:54:09 -0700
Subject: wifi: ath12k: allocate dummy net_device dynamically

Embedding net_device into structures prohibits the usage of flexible
arrays in the net_device structure. For more details, see the discussion
at [1].

Un-embed the net_device from struct ath12k_ext_irq_grp by converting it
into a pointer. Then use the leverage alloc_netdev_dummy() to allocate
the net_device object at ath12k_pci_ext_irq_config().

The free of the device occurs at ath12k_pci_free_ext_irq().

Link: https://lore.kernel.org/all/20240229225910.79e224cf@kernel.org/ [1]
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508095410.1923198-1-leitao@debian.org
---
 drivers/net/wireless/ath/ath12k/core.h |  2 +-
 drivers/net/wireless/ath/ath12k/pci.c  | 27 ++++++++++++++++++++++-----
 2 files changed, 23 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index bb6c1b562baf..c6b7524ce3a0 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -146,7 +146,7 @@ struct ath12k_ext_irq_grp {
 	u32 grp_id;
 	u64 timestamp;
 	struct napi_struct napi;
-	struct net_device napi_ndev;
+	struct net_device *napi_ndev;
 };
 
 struct ath12k_smbios_bdf {
diff --git a/drivers/net/wireless/ath/ath12k/pci.c b/drivers/net/wireless/ath/ath12k/pci.c
index 16af046c33d9..ac75e8e3916b 100644
--- a/drivers/net/wireless/ath/ath12k/pci.c
+++ b/drivers/net/wireless/ath/ath12k/pci.c
@@ -350,6 +350,7 @@ static void ath12k_pci_free_ext_irq(struct ath12k_base *ab)
 			free_irq(ab->irq_num[irq_grp->irqs[j]], irq_grp);
 
 		netif_napi_del(&irq_grp->napi);
+		free_netdev(irq_grp->napi_ndev);
 	}
 }
 
@@ -560,8 +561,9 @@ static irqreturn_t ath12k_pci_ext_interrupt_handler(int irq, void *arg)
 static int ath12k_pci_ext_irq_config(struct ath12k_base *ab)
 {
 	struct ath12k_pci *ab_pci = ath12k_pci_priv(ab);
-	int i, j, ret, num_vectors = 0;
+	int i, j, n, ret, num_vectors = 0;
 	u32 user_base_data = 0, base_vector = 0, base_idx;
+	struct ath12k_ext_irq_grp *irq_grp;
 
 	base_idx = ATH12K_PCI_IRQ_CE0_OFFSET + CE_COUNT_MAX;
 	ret = ath12k_pci_get_user_msi_assignment(ab, "DP",
@@ -572,13 +574,18 @@ static int ath12k_pci_ext_irq_config(struct ath12k_base *ab)
 		return ret;
 
 	for (i = 0; i < ATH12K_EXT_IRQ_GRP_NUM_MAX; i++) {
-		struct ath12k_ext_irq_grp *irq_grp = &ab->ext_irq_grp[i];
+		irq_grp = &ab->ext_irq_grp[i];
 		u32 num_irq = 0;
 
 		irq_grp->ab = ab;
 		irq_grp->grp_id = i;
-		init_dummy_netdev(&irq_grp->napi_ndev);
-		netif_napi_add(&irq_grp->napi_ndev, &irq_grp->napi,
+		irq_grp->napi_ndev = alloc_netdev_dummy(0);
+		if (!irq_grp->napi_ndev) {
+			ret = -ENOMEM;
+			goto fail_allocate;
+		}
+
+		netif_napi_add(irq_grp->napi_ndev, &irq_grp->napi,
 			       ath12k_pci_ext_grp_napi_poll);
 
 		if (ab->hw_params->ring_mask->tx[i] ||
@@ -611,13 +618,23 @@ static int ath12k_pci_ext_irq_config(struct ath12k_base *ab)
 			if (ret) {
 				ath12k_err(ab, "failed request irq %d: %d\n",
 					   vector, ret);
-				return ret;
+				goto fail_request;
 			}
 		}
 		ath12k_pci_ext_grp_disable(irq_grp);
 	}
 
 	return 0;
+
+fail_request:
+	/* i ->napi_ndev was properly allocated. Free it also */
+	i += 1;
+fail_allocate:
+	for (n = 0; n < i; n++) {
+		irq_grp = &ab->ext_irq_grp[n];
+		free_netdev(irq_grp->napi_ndev);
+	}
+	return ret;
 }
 
 static int ath12k_pci_set_irq_affinity_hint(struct ath12k_pci *ab_pci,
-- 
cgit v1.2.3-58-ga151


From cea4066588308fa932b6b03486c608efff1d761c Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Thu, 9 May 2024 17:06:43 +0800
Subject: wifi: rtw89: ser: avoid multiple deinit on same CAM

We did deinit CAM in STA iteration in VIF loop. But, the STA iteration
missed to restrict the target VIF. So, if there are multiple VIFs, we
would deinit a CAM multiple times. Now, fix it.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240509090646.35304-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/ser.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/ser.c b/drivers/net/wireless/realtek/rtw89/ser.c
index 99896d85d2f8..5fc2faa9ba5a 100644
--- a/drivers/net/wireless/realtek/rtw89/ser.c
+++ b/drivers/net/wireless/realtek/rtw89/ser.c
@@ -308,9 +308,13 @@ static void ser_reset_vif(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
 
 static void ser_sta_deinit_cam_iter(void *data, struct ieee80211_sta *sta)
 {
-	struct rtw89_vif *rtwvif = (struct rtw89_vif *)data;
-	struct rtw89_dev *rtwdev = rtwvif->rtwdev;
+	struct rtw89_vif *target_rtwvif = (struct rtw89_vif *)data;
 	struct rtw89_sta *rtwsta = (struct rtw89_sta *)sta->drv_priv;
+	struct rtw89_vif *rtwvif = rtwsta->rtwvif;
+	struct rtw89_dev *rtwdev = rtwvif->rtwdev;
+
+	if (rtwvif != target_rtwvif)
+		return;
 
 	if (rtwvif->net_type == RTW89_NET_TYPE_AP_MODE || sta->tdls)
 		rtw89_cam_deinit_addr_cam(rtwdev, &rtwsta->addr_cam);
-- 
cgit v1.2.3-58-ga151


From 0448c65af9219ee296f844748d60448f7418180f Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Thu, 9 May 2024 17:06:44 +0800
Subject: wifi: rtw89: cam: tweak relation between sec CAM and addr CAM

In original code architecture, sec CAM is a component of addr CAM.
Hence, sec CAM could not be shared. After MLO, each link has its own
addr CAM. However, one MLD PTK takes only one sec CAM but it needs to
work on multiple links' addr CAMs. So, we now manage sec CAMs in global
pool, and each of them is not bound to a single addr CAM now.

before:

+-----------------+         +--------------------+
| rtw89_{vif/sta} |   ...   | ieee80211_key_conf |
+-----------------+         +--------------------+
        ^                              |
        V                              | hw_key_idx
        |                              |
        |                              V
        |                             +--
   +----------+   addr_cam::key_idx   |  \        +---------+
   | addr_cam |<>---------------------|   |-------| sec_cam |
   +----------+                       |  /        +---------+
                                      +--

after:

+----------------------+         +--------------------+
| rtw89_{vif/sta}_link |   ...   | ieee80211_key_conf |
+----------------------+         +--------------------+
        ^                                   |
        V                                   | hw_key_idx
        |                                   |
        |                                   V
        |                                  --+                  +---------+
        |              +---------+        /  |  sec_cam_idx     | global  |
        |              | sec_cam |-------|   |----------------<>| sec_cam |
        |              +---------+        \  |                  | pool    |
        |                   ^              --+                  +---------+
        |                   |
        |               (*) |
   +----------+             |
   | addr_cam |-------------+
   +----------+

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240509090646.35304-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/cam.c   | 79 +++++++++++++++++++-----------
 drivers/net/wireless/realtek/rtw89/core.c  | 20 ++------
 drivers/net/wireless/realtek/rtw89/core.h  |  2 +-
 drivers/net/wireless/realtek/rtw89/debug.c | 16 ++++--
 4 files changed, 67 insertions(+), 50 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/cam.c b/drivers/net/wireless/realtek/rtw89/cam.c
index 1864f543a6c6..35291efbbae9 100644
--- a/drivers/net/wireless/realtek/rtw89/cam.c
+++ b/drivers/net/wireless/realtek/rtw89/cam.c
@@ -211,6 +211,46 @@ static int rtw89_cam_get_addr_cam_key_idx(struct rtw89_addr_cam_entry *addr_cam,
 	return 0;
 }
 
+static int rtw89_cam_detach_sec_cam(struct rtw89_dev *rtwdev,
+				    struct ieee80211_vif *vif,
+				    struct ieee80211_sta *sta,
+				    const struct rtw89_sec_cam_entry *sec_cam,
+				    bool inform_fw)
+{
+	struct rtw89_sta *rtwsta = sta_to_rtwsta_safe(sta);
+	struct rtw89_vif *rtwvif;
+	struct rtw89_addr_cam_entry *addr_cam;
+	unsigned int i;
+	int ret = 0;
+
+	if (!vif) {
+		rtw89_err(rtwdev, "No iface for deleting sec cam\n");
+		return -EINVAL;
+	}
+
+	rtwvif = (struct rtw89_vif *)vif->drv_priv;
+	addr_cam = rtw89_get_addr_cam_of(rtwvif, rtwsta);
+
+	for_each_set_bit(i, addr_cam->sec_cam_map, RTW89_SEC_CAM_IN_ADDR_CAM) {
+		if (addr_cam->sec_ent[i] != sec_cam->sec_cam_idx)
+			continue;
+
+		clear_bit(i, addr_cam->sec_cam_map);
+	}
+
+	if (inform_fw) {
+		ret = rtw89_chip_h2c_dctl_sec_cam(rtwdev, rtwvif, rtwsta);
+		if (ret)
+			rtw89_err(rtwdev,
+				  "failed to update dctl cam del key: %d\n", ret);
+		ret = rtw89_fw_h2c_cam(rtwdev, rtwvif, rtwsta, NULL);
+		if (ret)
+			rtw89_err(rtwdev, "failed to update cam del key: %d\n", ret);
+	}
+
+	return ret;
+}
+
 static int rtw89_cam_attach_sec_cam(struct rtw89_dev *rtwdev,
 				    struct ieee80211_vif *vif,
 				    struct ieee80211_sta *sta,
@@ -242,10 +282,8 @@ static int rtw89_cam_attach_sec_cam(struct rtw89_dev *rtwdev,
 		return ret;
 	}
 
-	key->hw_key_idx = key_idx;
 	addr_cam->sec_ent_keyid[key_idx] = key->keyidx;
 	addr_cam->sec_ent[key_idx] = sec_cam->sec_cam_idx;
-	addr_cam->sec_entries[key_idx] = sec_cam;
 	set_bit(key_idx, addr_cam->sec_cam_map);
 	ret = rtw89_chip_h2c_dctl_sec_cam(rtwdev, rtwvif, rtwsta);
 	if (ret) {
@@ -258,7 +296,6 @@ static int rtw89_cam_attach_sec_cam(struct rtw89_dev *rtwdev,
 		rtw89_err(rtwdev, "failed to update addr cam sec entry: %d\n",
 			  ret);
 		clear_bit(key_idx, addr_cam->sec_cam_map);
-		addr_cam->sec_entries[key_idx] = NULL;
 		return ret;
 	}
 
@@ -295,6 +332,9 @@ static int rtw89_cam_sec_key_install(struct rtw89_dev *rtwdev,
 		goto err_release_cam;
 	}
 
+	key->hw_key_idx = sec_cam_idx;
+	cam_info->sec_entries[sec_cam_idx] = sec_cam;
+
 	sec_cam->sec_cam_idx = sec_cam_idx;
 	sec_cam->type = hw_key_type;
 	sec_cam->len = RTW89_SEC_CAM_LEN;
@@ -316,6 +356,7 @@ static int rtw89_cam_sec_key_install(struct rtw89_dev *rtwdev,
 	return 0;
 
 err_release_cam:
+	cam_info->sec_entries[sec_cam_idx] = NULL;
 	kfree(sec_cam);
 	clear_bit(sec_cam_idx, cam_info->sec_cam_map);
 	if (ext_key)
@@ -386,42 +427,22 @@ int rtw89_cam_sec_key_del(struct rtw89_dev *rtwdev,
 			  struct ieee80211_key_conf *key,
 			  bool inform_fw)
 {
-	struct rtw89_sta *rtwsta = sta_to_rtwsta_safe(sta);
 	struct rtw89_cam_info *cam_info = &rtwdev->cam_info;
-	struct rtw89_vif *rtwvif;
-	struct rtw89_addr_cam_entry *addr_cam;
-	struct rtw89_sec_cam_entry *sec_cam;
-	u8 key_idx = key->hw_key_idx;
+	const struct rtw89_sec_cam_entry *sec_cam;
 	u8 sec_cam_idx;
-	int ret = 0;
-
-	if (!vif) {
-		rtw89_err(rtwdev, "No iface for deleting sec cam\n");
-		return -EINVAL;
-	}
+	int ret;
 
-	rtwvif = (struct rtw89_vif *)vif->drv_priv;
-	addr_cam = rtw89_get_addr_cam_of(rtwvif, rtwsta);
-	sec_cam = addr_cam->sec_entries[key_idx];
+	sec_cam_idx = key->hw_key_idx;
+	sec_cam = cam_info->sec_entries[sec_cam_idx];
 	if (!sec_cam)
 		return -EINVAL;
 
-	/* detach sec cam from addr cam */
-	clear_bit(key_idx, addr_cam->sec_cam_map);
-	addr_cam->sec_entries[key_idx] = NULL;
-	if (inform_fw) {
-		ret = rtw89_chip_h2c_dctl_sec_cam(rtwdev, rtwvif, rtwsta);
-		if (ret)
-			rtw89_err(rtwdev, "failed to update dctl cam del key: %d\n", ret);
-		ret = rtw89_fw_h2c_cam(rtwdev, rtwvif, rtwsta, NULL);
-		if (ret)
-			rtw89_err(rtwdev, "failed to update cam del key: %d\n", ret);
-	}
+	ret = rtw89_cam_detach_sec_cam(rtwdev, vif, sta, sec_cam, inform_fw);
 
 	/* clear valid bit in addr cam will disable sec cam,
 	 * so we don't need to send H2C command again
 	 */
-	sec_cam_idx = sec_cam->sec_cam_idx;
+	cam_info->sec_entries[sec_cam_idx] = NULL;
 	clear_bit(sec_cam_idx, cam_info->sec_cam_map);
 	if (sec_cam->ext_key)
 		clear_bit(sec_cam_idx + 1, cam_info->sec_cam_map);
diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 53997fa51ba0..1382bd24ba3b 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -499,31 +499,21 @@ static void
 rtw89_core_tx_update_sec_key(struct rtw89_dev *rtwdev,
 			     struct rtw89_core_tx_request *tx_req)
 {
+	struct rtw89_cam_info *cam_info = &rtwdev->cam_info;
 	const struct rtw89_chip_info *chip = rtwdev->chip;
-	struct ieee80211_vif *vif = tx_req->vif;
-	struct ieee80211_sta *sta = tx_req->sta;
+	const struct rtw89_sec_cam_entry *sec_cam;
 	struct ieee80211_tx_info *info;
 	struct ieee80211_key_conf *key;
-	struct rtw89_vif *rtwvif;
-	struct rtw89_sta *rtwsta = sta_to_rtwsta_safe(sta);
-	struct rtw89_addr_cam_entry *addr_cam;
-	struct rtw89_sec_cam_entry *sec_cam;
 	struct rtw89_tx_desc_info *desc_info = &tx_req->desc_info;
 	struct sk_buff *skb = tx_req->skb;
 	u8 sec_type = RTW89_SEC_KEY_TYPE_NONE;
+	u8 sec_cam_idx;
 	u64 pn64;
 
-	if (!vif) {
-		rtw89_warn(rtwdev, "cannot set sec key without vif\n");
-		return;
-	}
-
-	rtwvif = (struct rtw89_vif *)vif->drv_priv;
-	addr_cam = rtw89_get_addr_cam_of(rtwvif, rtwsta);
-
 	info = IEEE80211_SKB_CB(skb);
 	key = info->control.hw_key;
-	sec_cam = addr_cam->sec_entries[key->hw_key_idx];
+	sec_cam_idx = key->hw_key_idx;
+	sec_cam = cam_info->sec_entries[sec_cam_idx];
 	if (!sec_cam) {
 		rtw89_warn(rtwdev, "sec cam entry is empty\n");
 		return;
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 0648458110ce..32ae500579bb 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -3249,7 +3249,6 @@ struct rtw89_addr_cam_entry {
 	DECLARE_BITMAP(sec_cam_map, RTW89_SEC_CAM_IN_ADDR_CAM);
 	u8 sec_ent_keyid[RTW89_SEC_CAM_IN_ADDR_CAM];
 	u8 sec_ent[RTW89_SEC_CAM_IN_ADDR_CAM];
-	struct rtw89_sec_cam_entry *sec_entries[RTW89_SEC_CAM_IN_ADDR_CAM];
 };
 
 struct rtw89_bssid_cam_entry {
@@ -4436,6 +4435,7 @@ struct rtw89_cam_info {
 	DECLARE_BITMAP(sec_cam_map, RTW89_MAX_SEC_CAM_NUM);
 	DECLARE_BITMAP(ba_cam_map, RTW89_MAX_BA_CAM_NUM);
 	struct rtw89_ba_cam_entry ba_cam_entry[RTW89_MAX_BA_CAM_NUM];
+	const struct rtw89_sec_cam_entry *sec_entries[RTW89_MAX_SEC_CAM_NUM];
 };
 
 enum rtw89_sar_sources {
diff --git a/drivers/net/wireless/realtek/rtw89/debug.c b/drivers/net/wireless/realtek/rtw89/debug.c
index affffc4092ba..210f192ac9ec 100644
--- a/drivers/net/wireless/realtek/rtw89/debug.c
+++ b/drivers/net/wireless/realtek/rtw89/debug.c
@@ -3645,17 +3645,21 @@ static int rtw89_debug_priv_phy_info_get(struct seq_file *m, void *v)
 }
 
 static void rtw89_dump_addr_cam(struct seq_file *m,
+				struct rtw89_dev *rtwdev,
 				struct rtw89_addr_cam_entry *addr_cam)
 {
-	struct rtw89_sec_cam_entry *sec_entry;
+	struct rtw89_cam_info *cam_info = &rtwdev->cam_info;
+	const struct rtw89_sec_cam_entry *sec_entry;
+	u8 sec_cam_idx;
 	int i;
 
 	seq_printf(m, "\taddr_cam_idx=%u\n", addr_cam->addr_cam_idx);
 	seq_printf(m, "\t-> bssid_cam_idx=%u\n", addr_cam->bssid_cam_idx);
 	seq_printf(m, "\tsec_cam_bitmap=%*ph\n", (int)sizeof(addr_cam->sec_cam_map),
 		   addr_cam->sec_cam_map);
-	for (i = 0; i < RTW89_SEC_CAM_IN_ADDR_CAM; i++) {
-		sec_entry = addr_cam->sec_entries[i];
+	for_each_set_bit(i, addr_cam->sec_cam_map, RTW89_SEC_CAM_IN_ADDR_CAM) {
+		sec_cam_idx = addr_cam->sec_ent[i];
+		sec_entry = cam_info->sec_entries[sec_cam_idx];
 		if (!sec_entry)
 			continue;
 		seq_printf(m, "\tsec[%d]: sec_cam_idx %u", i, sec_entry->sec_cam_idx);
@@ -3694,12 +3698,13 @@ static
 void rtw89_vif_ids_get_iter(void *data, u8 *mac, struct ieee80211_vif *vif)
 {
 	struct rtw89_vif *rtwvif = (struct rtw89_vif *)vif->drv_priv;
+	struct rtw89_dev *rtwdev = rtwvif->rtwdev;
 	struct seq_file *m = (struct seq_file *)data;
 	struct rtw89_bssid_cam_entry *bssid_cam = &rtwvif->bssid_cam;
 
 	seq_printf(m, "VIF [%d] %pM\n", rtwvif->mac_id, rtwvif->mac_addr);
 	seq_printf(m, "\tbssid_cam_idx=%u\n", bssid_cam->bssid_cam_idx);
-	rtw89_dump_addr_cam(m, &rtwvif->addr_cam);
+	rtw89_dump_addr_cam(m, rtwdev, &rtwvif->addr_cam);
 	rtw89_dump_pkt_offload(m, &rtwvif->general_pkt_list, "\tpkt_ofld[GENERAL]: ");
 }
 
@@ -3726,11 +3731,12 @@ static void rtw89_dump_ba_cam(struct seq_file *m, struct rtw89_sta *rtwsta)
 static void rtw89_sta_ids_get_iter(void *data, struct ieee80211_sta *sta)
 {
 	struct rtw89_sta *rtwsta = (struct rtw89_sta *)sta->drv_priv;
+	struct rtw89_dev *rtwdev = rtwsta->rtwdev;
 	struct seq_file *m = (struct seq_file *)data;
 
 	seq_printf(m, "STA [%d] %pM %s\n", rtwsta->mac_id, sta->addr,
 		   sta->tdls ? "(TDLS)" : "");
-	rtw89_dump_addr_cam(m, &rtwsta->addr_cam);
+	rtw89_dump_addr_cam(m, rtwdev, &rtwsta->addr_cam);
 	rtw89_dump_ba_cam(m, rtwsta);
 }
 
-- 
cgit v1.2.3-58-ga151


From e9f1a901c9cc67ffb610ada8c2f4a6382ceb045a Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Thu, 9 May 2024 17:06:45 +0800
Subject: wifi: rtw89: switch to register vif_cfg_changed and link_info_changed

For impending MLO support, we drop bss_info_changed callback ahead and
switch to register new callbacks, vif_cfg_changed and link_info_changed,
which are introduced by mac80211 for multi-link point of view.

Besides, there is a place where we called ieee80211_find_sta() under
station mode. To work on both non-MLD and MLD connection, we should
call ieee80211_find_sta() with vif->cfg.ap_addr.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240509090646.35304-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/mac80211.c | 42 +++++++++++++++++----------
 1 file changed, 27 insertions(+), 15 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c
index 1ec97250e88e..41b286da3d59 100644
--- a/drivers/net/wireless/realtek/rtw89/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw89/mac80211.c
@@ -397,15 +397,14 @@ static void rtw89_conf_tx(struct rtw89_dev *rtwdev,
 }
 
 static void rtw89_station_mode_sta_assoc(struct rtw89_dev *rtwdev,
-					 struct ieee80211_vif *vif,
-					 struct ieee80211_bss_conf *conf)
+					 struct ieee80211_vif *vif)
 {
 	struct ieee80211_sta *sta;
 
 	if (vif->type != NL80211_IFTYPE_STATION)
 		return;
 
-	sta = ieee80211_find_sta(vif, conf->bssid);
+	sta = ieee80211_find_sta(vif, vif->cfg.ap_addr);
 	if (!sta) {
 		rtw89_err(rtwdev, "can't find sta to set sta_assoc state\n");
 		return;
@@ -416,10 +415,8 @@ static void rtw89_station_mode_sta_assoc(struct rtw89_dev *rtwdev,
 	rtw89_core_sta_assoc(rtwdev, vif, sta);
 }
 
-static void rtw89_ops_bss_info_changed(struct ieee80211_hw *hw,
-				       struct ieee80211_vif *vif,
-				       struct ieee80211_bss_conf *conf,
-				       u64 changed)
+static void rtw89_ops_vif_cfg_changed(struct ieee80211_hw *hw,
+				      struct ieee80211_vif *vif, u64 changed)
 {
 	struct rtw89_dev *rtwdev = hw->priv;
 	struct rtw89_vif *rtwvif = (struct rtw89_vif *)vif->drv_priv;
@@ -429,7 +426,7 @@ static void rtw89_ops_bss_info_changed(struct ieee80211_hw *hw,
 
 	if (changed & BSS_CHANGED_ASSOC) {
 		if (vif->cfg.assoc) {
-			rtw89_station_mode_sta_assoc(rtwdev, vif, conf);
+			rtw89_station_mode_sta_assoc(rtwdev, vif);
 			rtw89_phy_set_bss_color(rtwdev, vif);
 			rtw89_chip_cfg_txpwr_ul_tb_offset(rtwdev, vif);
 			rtw89_mac_port_update(rtwdev, rtwvif);
@@ -445,6 +442,26 @@ static void rtw89_ops_bss_info_changed(struct ieee80211_hw *hw,
 		}
 	}
 
+	if (changed & BSS_CHANGED_PS)
+		rtw89_recalc_lps(rtwdev);
+
+	if (changed & BSS_CHANGED_ARP_FILTER)
+		rtwvif->ip_addr = vif->cfg.arp_addr_list[0];
+
+	mutex_unlock(&rtwdev->mutex);
+}
+
+static void rtw89_ops_link_info_changed(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif,
+					struct ieee80211_bss_conf *conf,
+					u64 changed)
+{
+	struct rtw89_dev *rtwdev = hw->priv;
+	struct rtw89_vif *rtwvif = (struct rtw89_vif *)vif->drv_priv;
+
+	mutex_lock(&rtwdev->mutex);
+	rtw89_leave_ps_mode(rtwdev);
+
 	if (changed & BSS_CHANGED_BSSID) {
 		ether_addr_copy(rtwvif->bssid, conf->bssid);
 		rtw89_cam_bssid_changed(rtwdev, rtwvif);
@@ -470,12 +487,6 @@ static void rtw89_ops_bss_info_changed(struct ieee80211_hw *hw,
 	if (changed & BSS_CHANGED_CQM)
 		rtw89_fw_h2c_set_bcn_fltr_cfg(rtwdev, vif, true);
 
-	if (changed & BSS_CHANGED_PS)
-		rtw89_recalc_lps(rtwdev);
-
-	if (changed & BSS_CHANGED_ARP_FILTER)
-		rtwvif->ip_addr = vif->cfg.arp_addr_list[0];
-
 	mutex_unlock(&rtwdev->mutex);
 }
 
@@ -1143,7 +1154,8 @@ const struct ieee80211_ops rtw89_ops = {
 	.change_interface       = rtw89_ops_change_interface,
 	.remove_interface	= rtw89_ops_remove_interface,
 	.configure_filter	= rtw89_ops_configure_filter,
-	.bss_info_changed	= rtw89_ops_bss_info_changed,
+	.vif_cfg_changed	= rtw89_ops_vif_cfg_changed,
+	.link_info_changed	= rtw89_ops_link_info_changed,
 	.start_ap		= rtw89_ops_start_ap,
 	.stop_ap		= rtw89_ops_stop_ap,
 	.set_tim		= rtw89_ops_set_tim,
-- 
cgit v1.2.3-58-ga151


From 578bdd984f4562a47ec8d57742affd688688a0a3 Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Thu, 9 May 2024 17:06:46 +0800
Subject: wifi: rtw89: support mac_id number according to chip

On 802.11be chips, to consider MLO, HW doesn't design number
of support mac_id as large as before. And, it might be various
according to chip. For example, old chips support mac_id up to
128, but RTL8922a only supports mac_id up to 32.

Besides, the mac_id acquiring function will be extended when
impending MLO support.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240509090646.35304-5-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c     | 28 ++++++++++++++++++++++-----
 drivers/net/wireless/realtek/rtw89/core.h     |  3 +++
 drivers/net/wireless/realtek/rtw89/mac.c      |  7 +++----
 drivers/net/wireless/realtek/rtw89/rtw8851b.c |  1 +
 drivers/net/wireless/realtek/rtw89/rtw8852a.c |  1 +
 drivers/net/wireless/realtek/rtw89/rtw8852b.c |  1 +
 drivers/net/wireless/realtek/rtw89/rtw8852c.c |  1 +
 drivers/net/wireless/realtek/rtw89/rtw8922a.c |  1 +
 8 files changed, 34 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 1382bd24ba3b..8b6cb310eaba 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -3343,14 +3343,13 @@ int rtw89_core_sta_add(struct rtw89_dev *rtwdev,
 					 BTC_ROLE_MSTS_STA_CONN_START);
 		rtw89_chip_rfk_channel(rtwdev);
 	} else if (vif->type == NL80211_IFTYPE_AP || sta->tdls) {
-		rtwsta->mac_id = rtw89_core_acquire_bit_map(rtwdev->mac_id_map,
-							    RTW89_MAX_MAC_ID_NUM);
+		rtwsta->mac_id = rtw89_acquire_mac_id(rtwdev);
 		if (rtwsta->mac_id == RTW89_MAX_MAC_ID_NUM)
 			return -ENOSPC;
 
 		ret = rtw89_mac_set_macid_pause(rtwdev, rtwsta->mac_id, false);
 		if (ret) {
-			rtw89_core_release_bit_map(rtwdev->mac_id_map, rtwsta->mac_id);
+			rtw89_release_mac_id(rtwdev, rtwsta->mac_id);
 			rtw89_warn(rtwdev, "failed to send h2c macid pause\n");
 			return ret;
 		}
@@ -3358,7 +3357,7 @@ int rtw89_core_sta_add(struct rtw89_dev *rtwdev,
 		ret = rtw89_fw_h2c_role_maintain(rtwdev, rtwvif, rtwsta,
 						 RTW89_ROLE_CREATE);
 		if (ret) {
-			rtw89_core_release_bit_map(rtwdev->mac_id_map, rtwsta->mac_id);
+			rtw89_release_mac_id(rtwdev, rtwsta->mac_id);
 			rtw89_warn(rtwdev, "failed to send h2c role info\n");
 			return ret;
 		}
@@ -3531,7 +3530,7 @@ int rtw89_core_sta_remove(struct rtw89_dev *rtwdev,
 		rtw89_btc_ntfy_role_info(rtwdev, rtwvif, rtwsta,
 					 BTC_ROLE_MSTS_STA_DIS_CONN);
 	} else if (vif->type == NL80211_IFTYPE_AP || sta->tdls) {
-		rtw89_core_release_bit_map(rtwdev->mac_id_map, rtwsta->mac_id);
+		rtw89_release_mac_id(rtwdev, rtwsta->mac_id);
 
 		ret = rtw89_fw_h2c_role_maintain(rtwdev, rtwvif, rtwsta,
 						 RTW89_ROLE_REMOVE);
@@ -4181,6 +4180,25 @@ void rtw89_core_stop(struct rtw89_dev *rtwdev)
 	rtw89_hci_reset(rtwdev);
 }
 
+u8 rtw89_acquire_mac_id(struct rtw89_dev *rtwdev)
+{
+	const struct rtw89_chip_info *chip = rtwdev->chip;
+	u8 mac_id_num = chip->support_macid_num;
+	u8 mac_id;
+
+	mac_id = find_first_zero_bit(rtwdev->mac_id_map, mac_id_num);
+	if (mac_id == mac_id_num)
+		return RTW89_MAX_MAC_ID_NUM;
+
+	set_bit(mac_id, rtwdev->mac_id_map);
+	return mac_id;
+}
+
+void rtw89_release_mac_id(struct rtw89_dev *rtwdev, u8 mac_id)
+{
+	clear_bit(mac_id, rtwdev->mac_id_map);
+}
+
 int rtw89_core_init(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_btc *btc = &rtwdev->btc;
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 32ae500579bb..c0f2afb4888a 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -4143,6 +4143,7 @@ struct rtw89_chip_info {
 	u8 wde_qempty_acq_grpnum;
 	u8 wde_qempty_mgq_grpsel;
 	u32 rf_base_addr[2];
+	u8 support_macid_num;
 	u8 support_chanctx_num;
 	u8 support_bands;
 	u16 support_bandwidths;
@@ -6470,6 +6471,8 @@ struct rtw89_dev *rtw89_alloc_ieee80211_hw(struct device *device,
 					   u32 bus_data_size,
 					   const struct rtw89_chip_info *chip);
 void rtw89_free_ieee80211_hw(struct rtw89_dev *rtwdev);
+u8 rtw89_acquire_mac_id(struct rtw89_dev *rtwdev);
+void rtw89_release_mac_id(struct rtw89_dev *rtwdev, u8 mac_id);
 void rtw89_core_set_chip_txpwr(struct rtw89_dev *rtwdev);
 void rtw89_get_default_chandef(struct cfg80211_chan_def *chandef);
 void rtw89_get_channel_params(const struct cfg80211_chan_def *chandef,
diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index 3fe0046f6eaa..baefc30b88c4 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -4664,8 +4664,7 @@ int rtw89_mac_add_vif(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
 {
 	int ret;
 
-	rtwvif->mac_id = rtw89_core_acquire_bit_map(rtwdev->mac_id_map,
-						    RTW89_MAX_MAC_ID_NUM);
+	rtwvif->mac_id = rtw89_acquire_mac_id(rtwdev);
 	if (rtwvif->mac_id == RTW89_MAX_MAC_ID_NUM)
 		return -ENOSPC;
 
@@ -4676,7 +4675,7 @@ int rtw89_mac_add_vif(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
 	return 0;
 
 release_mac_id:
-	rtw89_core_release_bit_map(rtwdev->mac_id_map, rtwvif->mac_id);
+	rtw89_release_mac_id(rtwdev, rtwvif->mac_id);
 
 	return ret;
 }
@@ -4686,7 +4685,7 @@ int rtw89_mac_remove_vif(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
 	int ret;
 
 	ret = rtw89_mac_vif_deinit(rtwdev, rtwvif);
-	rtw89_core_release_bit_map(rtwdev->mac_id_map, rtwvif->mac_id);
+	rtw89_release_mac_id(rtwdev, rtwvif->mac_id);
 
 	return ret;
 }
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851b.c b/drivers/net/wireless/realtek/rtw89/rtw8851b.c
index 87b51823244d..78e276f321c2 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8851b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8851b.c
@@ -2447,6 +2447,7 @@ const struct rtw89_chip_info rtw8851b_chip_info = {
 	.dig_table		= NULL,
 	.dig_regs		= &rtw8851b_dig_regs,
 	.tssi_dbw_table		= NULL,
+	.support_macid_num	= RTW89_MAX_MAC_ID_NUM,
 	.support_chanctx_num	= 0,
 	.support_rnr		= false,
 	.support_bands		= BIT(NL80211_BAND_2GHZ) |
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
index e93cee1456bd..aebbceea93f8 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
@@ -2162,6 +2162,7 @@ const struct rtw89_chip_info rtw8852a_chip_info = {
 	.dig_table		= &rtw89_8852a_phy_dig_table,
 	.dig_regs		= &rtw8852a_dig_regs,
 	.tssi_dbw_table		= NULL,
+	.support_macid_num	= RTW89_MAX_MAC_ID_NUM,
 	.support_chanctx_num	= 1,
 	.support_rnr		= false,
 	.support_bands		= BIT(NL80211_BAND_2GHZ) |
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index d351096fa4b4..53ae00b5ab7c 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -2610,6 +2610,7 @@ const struct rtw89_chip_info rtw8852b_chip_info = {
 	.dig_table		= NULL,
 	.dig_regs		= &rtw8852b_dig_regs,
 	.tssi_dbw_table		= NULL,
+	.support_macid_num	= RTW89_MAX_MAC_ID_NUM,
 	.support_chanctx_num	= 0,
 	.support_rnr		= false,
 	.support_bands		= BIT(NL80211_BAND_2GHZ) |
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index 3571b41786d7..544ea1d3b577 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -2941,6 +2941,7 @@ const struct rtw89_chip_info rtw8852c_chip_info = {
 	.dig_table		= NULL,
 	.dig_regs		= &rtw8852c_dig_regs,
 	.tssi_dbw_table		= &rtw89_8852c_tssi_dbw_table,
+	.support_macid_num	= RTW89_MAX_MAC_ID_NUM,
 	.support_chanctx_num	= 2,
 	.support_rnr		= false,
 	.support_bands		= BIT(NL80211_BAND_2GHZ) |
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8922a.c b/drivers/net/wireless/realtek/rtw89/rtw8922a.c
index 3b3ea3a7c19a..cc4f251caadd 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8922a.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8922a.c
@@ -2544,6 +2544,7 @@ const struct rtw89_chip_info rtw8922a_chip_info = {
 	.dig_table		= NULL,
 	.dig_regs		= &rtw8922a_dig_regs,
 	.tssi_dbw_table		= NULL,
+	.support_macid_num	= 32,
 	.support_chanctx_num	= 2,
 	.support_rnr		= true,
 	.support_bands		= BIT(NL80211_BAND_2GHZ) |
-- 
cgit v1.2.3-58-ga151


From 02b682d54598f61cbb7dbb14d98ec1801112b878 Mon Sep 17 00:00:00 2001
From: Andrii Batyiev <batyiev@gmail.com>
Date: Thu, 9 May 2024 13:11:25 +0300
Subject: wifi: iwlegacy: do not skip frames with bad FCS

Monitor/sniffer mode benefits from all types of frames, even if FCS
check fails. But we must mark frames as such.

Tested on iwl3945 only.

Signed-off-by: Andrii Batyiev <batyiev@gmail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240509101140.32664-1-batyiev@gmail.com
---
 drivers/net/wireless/intel/iwlegacy/3945.c     | 2 +-
 drivers/net/wireless/intel/iwlegacy/4965-mac.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlegacy/3945.c b/drivers/net/wireless/intel/iwlegacy/3945.c
index a773939b8c2a..1fab7849f56d 100644
--- a/drivers/net/wireless/intel/iwlegacy/3945.c
+++ b/drivers/net/wireless/intel/iwlegacy/3945.c
@@ -566,7 +566,7 @@ il3945_hdl_rx(struct il_priv *il, struct il_rx_buf *rxb)
 	if (!(rx_end->status & RX_RES_STATUS_NO_CRC32_ERROR) ||
 	    !(rx_end->status & RX_RES_STATUS_NO_RXE_OVERFLOW)) {
 		D_RX("Bad CRC or FIFO: 0x%08X.\n", rx_end->status);
-		return;
+		rx_status.flag |= RX_FLAG_FAILED_FCS_CRC;
 	}
 
 	/* Convert 3945's rssi indicator to dBm */
diff --git a/drivers/net/wireless/intel/iwlegacy/4965-mac.c b/drivers/net/wireless/intel/iwlegacy/4965-mac.c
index 4beb7be6d51d..d018f56be966 100644
--- a/drivers/net/wireless/intel/iwlegacy/4965-mac.c
+++ b/drivers/net/wireless/intel/iwlegacy/4965-mac.c
@@ -664,7 +664,7 @@ il4965_hdl_rx(struct il_priv *il, struct il_rx_buf *rxb)
 	if (!(rx_pkt_status & RX_RES_STATUS_NO_CRC32_ERROR) ||
 	    !(rx_pkt_status & RX_RES_STATUS_NO_RXE_OVERFLOW)) {
 		D_RX("Bad CRC or FIFO: 0x%08X.\n", le32_to_cpu(rx_pkt_status));
-		return;
+		rx_status.flag |= RX_FLAG_FAILED_FCS_CRC;
 	}
 
 	/* This will be used in several places later */
-- 
cgit v1.2.3-58-ga151


From c636fa85feb450ca414a10010ed05361a73c93a6 Mon Sep 17 00:00:00 2001
From: Samasth Norway Ananda <samasth.norway.ananda@oracle.com>
Date: Thu, 9 May 2024 16:10:37 -0700
Subject: wifi: brcmsmac: LCN PHY code is used for BCM4313 2G-only device

The band_idx variable in the function wlc_lcnphy_tx_iqlo_cal() will
never be set to 1 as BCM4313 is the only device for which the LCN PHY
code is used. This is a 2G-only device.

Fixes: 5b435de0d786 ("net: wireless: add brcm80211 drivers")
Signed-off-by: Samasth Norway Ananda <samasth.norway.ananda@oracle.com>
Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240509231037.2014109-1-samasth.norway.ananda@oracle.com
---
 .../wireless/broadcom/brcm80211/brcmsmac/phy/phy_lcn.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_lcn.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_lcn.c
index aae2cf95fe95..e472591f321b 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_lcn.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_lcn.c
@@ -2567,7 +2567,6 @@ wlc_lcnphy_tx_iqlo_cal(struct brcms_phy *pi,
 
 	struct lcnphy_txgains cal_gains, temp_gains;
 	u16 hash;
-	u8 band_idx;
 	int j;
 	u16 ncorr_override[5];
 	u16 syst_coeffs[] = { 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000,
@@ -2599,6 +2598,9 @@ wlc_lcnphy_tx_iqlo_cal(struct brcms_phy *pi,
 	u16 *values_to_save;
 	struct brcms_phy_lcnphy *pi_lcn = pi->u.pi_lcnphy;
 
+	if (WARN_ON(CHSPEC_IS5G(pi->radio_chanspec)))
+		return;
+
 	values_to_save = kmalloc_array(20, sizeof(u16), GFP_ATOMIC);
 	if (NULL == values_to_save)
 		return;
@@ -2662,20 +2664,18 @@ wlc_lcnphy_tx_iqlo_cal(struct brcms_phy *pi,
 	hash = (target_gains->gm_gain << 8) |
 	       (target_gains->pga_gain << 4) | (target_gains->pad_gain);
 
-	band_idx = (CHSPEC_IS5G(pi->radio_chanspec) ? 1 : 0);
-
 	cal_gains = *target_gains;
 	memset(ncorr_override, 0, sizeof(ncorr_override));
-	for (j = 0; j < iqcal_gainparams_numgains_lcnphy[band_idx]; j++) {
-		if (hash == tbl_iqcal_gainparams_lcnphy[band_idx][j][0]) {
+	for (j = 0; j < iqcal_gainparams_numgains_lcnphy[0]; j++) {
+		if (hash == tbl_iqcal_gainparams_lcnphy[0][j][0]) {
 			cal_gains.gm_gain =
-				tbl_iqcal_gainparams_lcnphy[band_idx][j][1];
+				tbl_iqcal_gainparams_lcnphy[0][j][1];
 			cal_gains.pga_gain =
-				tbl_iqcal_gainparams_lcnphy[band_idx][j][2];
+				tbl_iqcal_gainparams_lcnphy[0][j][2];
 			cal_gains.pad_gain =
-				tbl_iqcal_gainparams_lcnphy[band_idx][j][3];
+				tbl_iqcal_gainparams_lcnphy[0][j][3];
 			memcpy(ncorr_override,
-			       &tbl_iqcal_gainparams_lcnphy[band_idx][j][3],
+			       &tbl_iqcal_gainparams_lcnphy[0][j][3],
 			       sizeof(ncorr_override));
 			break;
 		}
-- 
cgit v1.2.3-58-ga151


From 8e1debd82466a3fe711784ab37e6b54e56011267 Mon Sep 17 00:00:00 2001
From: Sebastian Gottschall <s.gottschall@dd-wrt.com>
Date: Mon, 13 May 2024 17:22:25 +0300
Subject: wifi: ath10k: add LED and GPIO controlling support for various
 chipsets

Adds LED and GPIO Control support for 988x, 9887, 9888, 99x0, 9984
based chipsets with on chipset connected led's using WMI Firmware API.
The LED device will get available named as "ath10k-phyX" at sysfs and
can be controlled with various triggers.
Adds also debugfs interface for gpio control.

Signed-off-by: Sebastian Gottschall <s.gottschall@dd-wrt.com>
Reviewed-by: Steve deRosier <derosier@cal-sierra.com>
[kvalo: major reorg and cleanup]
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
[ansuel: rebase and small cleanup]
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Tested-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20230611080505.17393-1-ansuelsmth@gmail.com
---
 drivers/net/wireless/ath/ath10k/Kconfig   |  6 +++
 drivers/net/wireless/ath/ath10k/Makefile  |  1 +
 drivers/net/wireless/ath/ath10k/core.c    | 32 +++++++++++
 drivers/net/wireless/ath/ath10k/core.h    |  8 +++
 drivers/net/wireless/ath/ath10k/hw.h      |  1 +
 drivers/net/wireless/ath/ath10k/leds.c    | 90 +++++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath10k/leds.h    | 34 ++++++++++++
 drivers/net/wireless/ath/ath10k/mac.c     |  1 +
 drivers/net/wireless/ath/ath10k/wmi-ops.h | 32 +++++++++++
 drivers/net/wireless/ath/ath10k/wmi-tlv.c |  2 +
 drivers/net/wireless/ath/ath10k/wmi.c     | 54 +++++++++++++++++++
 drivers/net/wireless/ath/ath10k/wmi.h     | 35 ++++++++++++
 12 files changed, 296 insertions(+)
 create mode 100644 drivers/net/wireless/ath/ath10k/leds.c
 create mode 100644 drivers/net/wireless/ath/ath10k/leds.h

diff --git a/drivers/net/wireless/ath/ath10k/Kconfig b/drivers/net/wireless/ath/ath10k/Kconfig
index e6ea884cafc1..76572cd2e1ba 100644
--- a/drivers/net/wireless/ath/ath10k/Kconfig
+++ b/drivers/net/wireless/ath/ath10k/Kconfig
@@ -67,6 +67,12 @@ config ATH10K_DEBUGFS
 
 	  If unsure, say Y to make it easier to debug problems.
 
+config ATH10K_LEDS
+	bool
+	depends on ATH10K
+	depends on LEDS_CLASS=y || LEDS_CLASS=MAC80211
+	default y
+
 config ATH10K_SPECTRAL
 	bool "Atheros ath10k spectral scan support"
 	depends on ATH10K_DEBUGFS
diff --git a/drivers/net/wireless/ath/ath10k/Makefile b/drivers/net/wireless/ath/ath10k/Makefile
index 142c777b287f..02bf9b629038 100644
--- a/drivers/net/wireless/ath/ath10k/Makefile
+++ b/drivers/net/wireless/ath/ath10k/Makefile
@@ -19,6 +19,7 @@ ath10k_core-$(CONFIG_ATH10K_SPECTRAL) += spectral.o
 ath10k_core-$(CONFIG_NL80211_TESTMODE) += testmode.o
 ath10k_core-$(CONFIG_ATH10K_TRACING) += trace.o
 ath10k_core-$(CONFIG_THERMAL) += thermal.o
+ath10k_core-$(CONFIG_ATH10K_LEDS) += leds.o
 ath10k_core-$(CONFIG_MAC80211_DEBUGFS) += debugfs_sta.o
 ath10k_core-$(CONFIG_PM) += wow.o
 ath10k_core-$(CONFIG_DEV_COREDUMP) += coredump.o
diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
index bdf0552cd1c3..b3294287bce1 100644
--- a/drivers/net/wireless/ath/ath10k/core.c
+++ b/drivers/net/wireless/ath/ath10k/core.c
@@ -27,6 +27,7 @@
 #include "testmode.h"
 #include "wmi-ops.h"
 #include "coredump.h"
+#include "leds.h"
 
 unsigned int ath10k_debug_mask;
 EXPORT_SYMBOL(ath10k_debug_mask);
@@ -68,6 +69,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca988x hw2.0",
 		.patch_load_addr = QCA988X_HW_2_0_PATCH_LOAD_ADDR,
 		.uart_pin = 7,
+		.led_pin = 1,
 		.cc_wraparound_type = ATH10K_HW_CC_WRAP_SHIFTED_ALL,
 		.otp_exe_param = 0,
 		.channel_counters_freq_hz = 88000,
@@ -108,6 +110,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca988x hw2.0 ubiquiti",
 		.patch_load_addr = QCA988X_HW_2_0_PATCH_LOAD_ADDR,
 		.uart_pin = 7,
+		.led_pin = 0,
 		.cc_wraparound_type = ATH10K_HW_CC_WRAP_SHIFTED_ALL,
 		.otp_exe_param = 0,
 		.channel_counters_freq_hz = 88000,
@@ -149,6 +152,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca9887 hw1.0",
 		.patch_load_addr = QCA9887_HW_1_0_PATCH_LOAD_ADDR,
 		.uart_pin = 7,
+		.led_pin = 1,
 		.cc_wraparound_type = ATH10K_HW_CC_WRAP_SHIFTED_ALL,
 		.otp_exe_param = 0,
 		.channel_counters_freq_hz = 88000,
@@ -190,6 +194,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca6174 hw3.2 sdio",
 		.patch_load_addr = QCA6174_HW_3_0_PATCH_LOAD_ADDR,
 		.uart_pin = 19,
+		.led_pin = 0,
 		.otp_exe_param = 0,
 		.channel_counters_freq_hz = 88000,
 		.max_probe_resp_desc_thres = 0,
@@ -226,6 +231,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca6164 hw2.1",
 		.patch_load_addr = QCA6174_HW_2_1_PATCH_LOAD_ADDR,
 		.uart_pin = 6,
+		.led_pin = 0,
 		.otp_exe_param = 0,
 		.channel_counters_freq_hz = 88000,
 		.max_probe_resp_desc_thres = 0,
@@ -266,6 +272,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca6174 hw2.1",
 		.patch_load_addr = QCA6174_HW_2_1_PATCH_LOAD_ADDR,
 		.uart_pin = 6,
+		.led_pin = 0,
 		.otp_exe_param = 0,
 		.channel_counters_freq_hz = 88000,
 		.max_probe_resp_desc_thres = 0,
@@ -306,6 +313,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca6174 hw3.0",
 		.patch_load_addr = QCA6174_HW_3_0_PATCH_LOAD_ADDR,
 		.uart_pin = 6,
+		.led_pin = 0,
 		.otp_exe_param = 0,
 		.channel_counters_freq_hz = 88000,
 		.max_probe_resp_desc_thres = 0,
@@ -346,6 +354,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca6174 hw3.2",
 		.patch_load_addr = QCA6174_HW_3_0_PATCH_LOAD_ADDR,
 		.uart_pin = 6,
+		.led_pin = 0,
 		.otp_exe_param = 0,
 		.channel_counters_freq_hz = 88000,
 		.max_probe_resp_desc_thres = 0,
@@ -390,6 +399,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca99x0 hw2.0",
 		.patch_load_addr = QCA99X0_HW_2_0_PATCH_LOAD_ADDR,
 		.uart_pin = 7,
+		.led_pin = 17,
 		.otp_exe_param = 0x00000700,
 		.continuous_frag_desc = true,
 		.cck_rate_map_rev2 = true,
@@ -436,6 +446,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca9984/qca9994 hw1.0",
 		.patch_load_addr = QCA9984_HW_1_0_PATCH_LOAD_ADDR,
 		.uart_pin = 7,
+		.led_pin = 17,
 		.cc_wraparound_type = ATH10K_HW_CC_WRAP_SHIFTED_EACH,
 		.otp_exe_param = 0x00000700,
 		.continuous_frag_desc = true,
@@ -488,6 +499,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca9888 hw2.0",
 		.patch_load_addr = QCA9888_HW_2_0_PATCH_LOAD_ADDR,
 		.uart_pin = 7,
+		.led_pin = 17,
 		.cc_wraparound_type = ATH10K_HW_CC_WRAP_SHIFTED_EACH,
 		.otp_exe_param = 0x00000700,
 		.continuous_frag_desc = true,
@@ -538,6 +550,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca9377 hw1.0",
 		.patch_load_addr = QCA9377_HW_1_0_PATCH_LOAD_ADDR,
 		.uart_pin = 6,
+		.led_pin = 0,
 		.otp_exe_param = 0,
 		.channel_counters_freq_hz = 88000,
 		.max_probe_resp_desc_thres = 0,
@@ -578,6 +591,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca9377 hw1.1",
 		.patch_load_addr = QCA9377_HW_1_0_PATCH_LOAD_ADDR,
 		.uart_pin = 6,
+		.led_pin = 0,
 		.otp_exe_param = 0,
 		.channel_counters_freq_hz = 88000,
 		.max_probe_resp_desc_thres = 0,
@@ -620,6 +634,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca9377 hw1.1 sdio",
 		.patch_load_addr = QCA9377_HW_1_0_PATCH_LOAD_ADDR,
 		.uart_pin = 19,
+		.led_pin = 0,
 		.otp_exe_param = 0,
 		.channel_counters_freq_hz = 88000,
 		.max_probe_resp_desc_thres = 0,
@@ -653,6 +668,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.name = "qca4019 hw1.0",
 		.patch_load_addr = QCA4019_HW_1_0_PATCH_LOAD_ADDR,
 		.uart_pin = 7,
+		.led_pin = 0,
 		.cc_wraparound_type = ATH10K_HW_CC_WRAP_SHIFTED_EACH,
 		.otp_exe_param = 0x0010000,
 		.continuous_frag_desc = true,
@@ -698,6 +714,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 		.dev_id = 0,
 		.bus = ATH10K_BUS_SNOC,
 		.name = "wcn3990 hw1.0",
+		.led_pin = 0,
 		.continuous_frag_desc = true,
 		.tx_chain_mask = 0x7,
 		.rx_chain_mask = 0x7,
@@ -3224,6 +3241,10 @@ int ath10k_core_start(struct ath10k *ar, enum ath10k_firmware_mode mode,
 		goto err_hif_stop;
 	}
 
+	status = ath10k_leds_start(ar);
+	if (status)
+		goto err_hif_stop;
+
 	return 0;
 
 err_hif_stop:
@@ -3482,9 +3503,18 @@ static void ath10k_core_register_work(struct work_struct *work)
 		goto err_spectral_destroy;
 	}
 
+	status = ath10k_leds_register(ar);
+	if (status) {
+		ath10k_err(ar, "could not register leds: %d\n",
+			   status);
+		goto err_thermal_unregister;
+	}
+
 	set_bit(ATH10K_FLAG_CORE_REGISTERED, &ar->dev_flags);
 	return;
 
+err_thermal_unregister:
+	ath10k_thermal_unregister(ar);
 err_spectral_destroy:
 	ath10k_spectral_destroy(ar);
 err_debug_destroy:
@@ -3520,6 +3550,8 @@ void ath10k_core_unregister(struct ath10k *ar)
 	if (!test_bit(ATH10K_FLAG_CORE_REGISTERED, &ar->dev_flags))
 		return;
 
+	ath10k_leds_unregister(ar);
+
 	ath10k_thermal_unregister(ar);
 	/* Stop spectral before unregistering from mac80211 to remove the
 	 * relayfs debugfs file cleanly. Otherwise the parent debugfs tree
diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h
index b00099f0b24e..446dca74f06a 100644
--- a/drivers/net/wireless/ath/ath10k/core.h
+++ b/drivers/net/wireless/ath/ath10k/core.h
@@ -15,6 +15,7 @@
 #include <linux/pci.h>
 #include <linux/uuid.h>
 #include <linux/time.h>
+#include <linux/leds.h>
 
 #include "htt.h"
 #include "htc.h"
@@ -1258,6 +1259,13 @@ struct ath10k {
 		bool utf_monitor;
 	} testmode;
 
+	struct {
+		struct gpio_led wifi_led;
+		struct led_classdev cdev;
+		char label[48];
+		u32 gpio_state_pin;
+	} leds;
+
 	struct {
 		/* protected by data_lock */
 		u32 rx_crc_err_drop;
diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h
index 48897e5eca06..442091c6dfd2 100644
--- a/drivers/net/wireless/ath/ath10k/hw.h
+++ b/drivers/net/wireless/ath/ath10k/hw.h
@@ -512,6 +512,7 @@ struct ath10k_hw_params {
 	const char *name;
 	u32 patch_load_addr;
 	int uart_pin;
+	int led_pin;
 	u32 otp_exe_param;
 
 	/* Type of hw cycle counter wraparound logic, for more info
diff --git a/drivers/net/wireless/ath/ath10k/leds.c b/drivers/net/wireless/ath/ath10k/leds.c
new file mode 100644
index 000000000000..9b1d04eb4265
--- /dev/null
+++ b/drivers/net/wireless/ath/ath10k/leds.c
@@ -0,0 +1,90 @@
+// SPDX-License-Identifier: ISC
+/*
+ * Copyright (c) 2005-2011 Atheros Communications Inc.
+ * Copyright (c) 2011-2017 Qualcomm Atheros, Inc.
+ * Copyright (c) 2018 Sebastian Gottschall <s.gottschall@dd-wrt.com>
+ * Copyright (c) 2018 The Linux Foundation. All rights reserved.
+ */
+
+#include <linux/leds.h>
+
+#include "core.h"
+#include "wmi.h"
+#include "wmi-ops.h"
+
+#include "leds.h"
+
+static int ath10k_leds_set_brightness_blocking(struct led_classdev *led_cdev,
+					       enum led_brightness brightness)
+{
+	struct ath10k *ar = container_of(led_cdev, struct ath10k,
+					 leds.cdev);
+	struct gpio_led *led = &ar->leds.wifi_led;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (ar->state != ATH10K_STATE_ON)
+		goto out;
+
+	ar->leds.gpio_state_pin = (brightness != LED_OFF) ^ led->active_low;
+	ath10k_wmi_gpio_output(ar, led->gpio, ar->leds.gpio_state_pin);
+
+out:
+	mutex_unlock(&ar->conf_mutex);
+
+	return 0;
+}
+
+int ath10k_leds_start(struct ath10k *ar)
+{
+	if (ar->hw_params.led_pin == 0)
+		/* leds not supported */
+		return 0;
+
+	/* under some circumstances, the gpio pin gets reconfigured
+	 * to default state by the firmware, so we need to
+	 * reconfigure it this behaviour has only ben seen on
+	 * QCA9984 and QCA99XX devices so far
+	 */
+	ath10k_wmi_gpio_config(ar, ar->hw_params.led_pin, 0,
+			       WMI_GPIO_PULL_NONE, WMI_GPIO_INTTYPE_DISABLE);
+	ath10k_wmi_gpio_output(ar, ar->hw_params.led_pin, 1);
+
+	return 0;
+}
+
+int ath10k_leds_register(struct ath10k *ar)
+{
+	int ret;
+
+	if (ar->hw_params.led_pin == 0)
+		/* leds not supported */
+		return 0;
+
+	snprintf(ar->leds.label, sizeof(ar->leds.label), "ath10k-%s",
+		 wiphy_name(ar->hw->wiphy));
+	ar->leds.wifi_led.active_low = 1;
+	ar->leds.wifi_led.gpio = ar->hw_params.led_pin;
+	ar->leds.wifi_led.name = ar->leds.label;
+	ar->leds.wifi_led.default_state = LEDS_GPIO_DEFSTATE_KEEP;
+
+	ar->leds.cdev.name = ar->leds.label;
+	ar->leds.cdev.brightness_set_blocking = ath10k_leds_set_brightness_blocking;
+	ar->leds.cdev.default_trigger = ar->leds.wifi_led.default_trigger;
+
+	ret = led_classdev_register(wiphy_dev(ar->hw->wiphy), &ar->leds.cdev);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+void ath10k_leds_unregister(struct ath10k *ar)
+{
+	if (ar->hw_params.led_pin == 0)
+		/* leds not supported */
+		return;
+
+	led_classdev_unregister(&ar->leds.cdev);
+}
+
diff --git a/drivers/net/wireless/ath/ath10k/leds.h b/drivers/net/wireless/ath/ath10k/leds.h
new file mode 100644
index 000000000000..56325b0875e5
--- /dev/null
+++ b/drivers/net/wireless/ath/ath10k/leds.h
@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: ISC */
+/*
+ * Copyright (c) 2005-2011 Atheros Communications Inc.
+ * Copyright (c) 2011-2017 Qualcomm Atheros, Inc.
+ * Copyright (c) 2018 Sebastian Gottschall <s.gottschall@dd-wrt.com>
+ * Copyright (c) 2018 The Linux Foundation. All rights reserved.
+ */
+
+#ifndef _LEDS_H_
+#define _LEDS_H_
+
+#include "core.h"
+
+#ifdef CONFIG_ATH10K_LEDS
+void ath10k_leds_unregister(struct ath10k *ar);
+int ath10k_leds_start(struct ath10k *ar);
+int ath10k_leds_register(struct ath10k *ar);
+#else
+static inline void ath10k_leds_unregister(struct ath10k *ar)
+{
+}
+
+static inline int ath10k_leds_start(struct ath10k *ar)
+{
+	return 0;
+}
+
+static inline int ath10k_leds_register(struct ath10k *ar)
+{
+	return 0;
+}
+
+#endif
+#endif /* _LEDS_H_ */
diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index e322b528baaf..3bf67b2ecd6d 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -25,6 +25,7 @@
 #include "wmi-tlv.h"
 #include "wmi-ops.h"
 #include "wow.h"
+#include "leds.h"
 
 /*********/
 /* Rates */
diff --git a/drivers/net/wireless/ath/ath10k/wmi-ops.h b/drivers/net/wireless/ath/ath10k/wmi-ops.h
index aa57d807491c..f3f6b5954b27 100644
--- a/drivers/net/wireless/ath/ath10k/wmi-ops.h
+++ b/drivers/net/wireless/ath/ath10k/wmi-ops.h
@@ -226,7 +226,10 @@ struct wmi_ops {
 			 const struct wmi_bb_timing_cfg_arg *arg);
 	struct sk_buff *(*gen_per_peer_per_tid_cfg)(struct ath10k *ar,
 						    const struct wmi_per_peer_per_tid_cfg_arg *arg);
+	struct sk_buff *(*gen_gpio_config)(struct ath10k *ar, u32 gpio_num,
+					   u32 input, u32 pull_type, u32 intr_mode);
 
+	struct sk_buff *(*gen_gpio_output)(struct ath10k *ar, u32 gpio_num, u32 set);
 };
 
 int ath10k_wmi_cmd_send(struct ath10k *ar, struct sk_buff *skb, u32 cmd_id);
@@ -1122,6 +1125,35 @@ ath10k_wmi_force_fw_hang(struct ath10k *ar,
 	return ath10k_wmi_cmd_send(ar, skb, ar->wmi.cmd->force_fw_hang_cmdid);
 }
 
+static inline int ath10k_wmi_gpio_config(struct ath10k *ar, u32 gpio_num,
+					 u32 input, u32 pull_type, u32 intr_mode)
+{
+	struct sk_buff *skb;
+
+	if (!ar->wmi.ops->gen_gpio_config)
+		return -EOPNOTSUPP;
+
+	skb = ar->wmi.ops->gen_gpio_config(ar, gpio_num, input, pull_type, intr_mode);
+	if (IS_ERR(skb))
+		return PTR_ERR(skb);
+
+	return ath10k_wmi_cmd_send(ar, skb, ar->wmi.cmd->gpio_config_cmdid);
+}
+
+static inline int ath10k_wmi_gpio_output(struct ath10k *ar, u32 gpio_num, u32 set)
+{
+	struct sk_buff *skb;
+
+	if (!ar->wmi.ops->gen_gpio_config)
+		return -EOPNOTSUPP;
+
+	skb = ar->wmi.ops->gen_gpio_output(ar, gpio_num, set);
+	if (IS_ERR(skb))
+		return PTR_ERR(skb);
+
+	return ath10k_wmi_cmd_send(ar, skb, ar->wmi.cmd->gpio_output_cmdid);
+}
+
 static inline int
 ath10k_wmi_dbglog_cfg(struct ath10k *ar, u64 module_enable, u32 log_level)
 {
diff --git a/drivers/net/wireless/ath/ath10k/wmi-tlv.c b/drivers/net/wireless/ath/ath10k/wmi-tlv.c
index aed97fd121ba..dbaf26d6a7a6 100644
--- a/drivers/net/wireless/ath/ath10k/wmi-tlv.c
+++ b/drivers/net/wireless/ath/ath10k/wmi-tlv.c
@@ -4606,6 +4606,8 @@ static const struct wmi_ops wmi_tlv_ops = {
 	.gen_echo = ath10k_wmi_tlv_op_gen_echo,
 	.gen_vdev_spectral_conf = ath10k_wmi_tlv_op_gen_vdev_spectral_conf,
 	.gen_vdev_spectral_enable = ath10k_wmi_tlv_op_gen_vdev_spectral_enable,
+	/* .gen_gpio_config not implemented */
+	/* .gen_gpio_output not implemented */
 };
 
 static const struct wmi_peer_flags_map wmi_tlv_peer_flags_map = {
diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c
index 80d255aaff1b..fe2344598364 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.c
+++ b/drivers/net/wireless/ath/ath10k/wmi.c
@@ -7493,6 +7493,49 @@ ath10k_wmi_op_gen_peer_set_param(struct ath10k *ar, u32 vdev_id,
 	return skb;
 }
 
+static struct sk_buff *ath10k_wmi_op_gen_gpio_config(struct ath10k *ar,
+						     u32 gpio_num, u32 input,
+						     u32 pull_type, u32 intr_mode)
+{
+	struct wmi_gpio_config_cmd *cmd;
+	struct sk_buff *skb;
+
+	skb = ath10k_wmi_alloc_skb(ar, sizeof(*cmd));
+	if (!skb)
+		return ERR_PTR(-ENOMEM);
+
+	cmd = (struct wmi_gpio_config_cmd *)skb->data;
+	cmd->pull_type = __cpu_to_le32(pull_type);
+	cmd->gpio_num = __cpu_to_le32(gpio_num);
+	cmd->input = __cpu_to_le32(input);
+	cmd->intr_mode = __cpu_to_le32(intr_mode);
+
+	ath10k_dbg(ar, ATH10K_DBG_WMI, "wmi gpio_config gpio_num 0x%08x input 0x%08x pull_type 0x%08x intr_mode 0x%08x\n",
+		   gpio_num, input, pull_type, intr_mode);
+
+	return skb;
+}
+
+static struct sk_buff *ath10k_wmi_op_gen_gpio_output(struct ath10k *ar,
+						     u32 gpio_num, u32 set)
+{
+	struct wmi_gpio_output_cmd *cmd;
+	struct sk_buff *skb;
+
+	skb = ath10k_wmi_alloc_skb(ar, sizeof(*cmd));
+	if (!skb)
+		return ERR_PTR(-ENOMEM);
+
+	cmd = (struct wmi_gpio_output_cmd *)skb->data;
+	cmd->gpio_num = __cpu_to_le32(gpio_num);
+	cmd->set = __cpu_to_le32(set);
+
+	ath10k_dbg(ar, ATH10K_DBG_WMI, "wmi gpio_output gpio_num 0x%08x set 0x%08x\n",
+		   gpio_num, set);
+
+	return skb;
+}
+
 static struct sk_buff *
 ath10k_wmi_op_gen_set_psmode(struct ath10k *ar, u32 vdev_id,
 			     enum wmi_sta_ps_mode psmode)
@@ -9157,6 +9200,9 @@ static const struct wmi_ops wmi_ops = {
 	.fw_stats_fill = ath10k_wmi_main_op_fw_stats_fill,
 	.get_vdev_subtype = ath10k_wmi_op_get_vdev_subtype,
 	.gen_echo = ath10k_wmi_op_gen_echo,
+	.gen_gpio_config = ath10k_wmi_op_gen_gpio_config,
+	.gen_gpio_output = ath10k_wmi_op_gen_gpio_output,
+
 	/* .gen_bcn_tmpl not implemented */
 	/* .gen_prb_tmpl not implemented */
 	/* .gen_p2p_go_bcn_ie not implemented */
@@ -9227,6 +9273,8 @@ static const struct wmi_ops wmi_10_1_ops = {
 	.fw_stats_fill = ath10k_wmi_10x_op_fw_stats_fill,
 	.get_vdev_subtype = ath10k_wmi_op_get_vdev_subtype,
 	.gen_echo = ath10k_wmi_op_gen_echo,
+	.gen_gpio_config = ath10k_wmi_op_gen_gpio_config,
+	.gen_gpio_output = ath10k_wmi_op_gen_gpio_output,
 	/* .gen_bcn_tmpl not implemented */
 	/* .gen_prb_tmpl not implemented */
 	/* .gen_p2p_go_bcn_ie not implemented */
@@ -9299,6 +9347,8 @@ static const struct wmi_ops wmi_10_2_ops = {
 	.gen_delba_send = ath10k_wmi_op_gen_delba_send,
 	.fw_stats_fill = ath10k_wmi_10x_op_fw_stats_fill,
 	.get_vdev_subtype = ath10k_wmi_op_get_vdev_subtype,
+	.gen_gpio_config = ath10k_wmi_op_gen_gpio_config,
+	.gen_gpio_output = ath10k_wmi_op_gen_gpio_output,
 	/* .gen_pdev_enable_adaptive_cca not implemented */
 };
 
@@ -9370,6 +9420,8 @@ static const struct wmi_ops wmi_10_2_4_ops = {
 		ath10k_wmi_op_gen_pdev_enable_adaptive_cca,
 	.get_vdev_subtype = ath10k_wmi_10_2_4_op_get_vdev_subtype,
 	.gen_bb_timing = ath10k_wmi_10_2_4_op_gen_bb_timing,
+	.gen_gpio_config = ath10k_wmi_op_gen_gpio_config,
+	.gen_gpio_output = ath10k_wmi_op_gen_gpio_output,
 	/* .gen_bcn_tmpl not implemented */
 	/* .gen_prb_tmpl not implemented */
 	/* .gen_p2p_go_bcn_ie not implemented */
@@ -9451,6 +9503,8 @@ static const struct wmi_ops wmi_10_4_ops = {
 	.gen_pdev_bss_chan_info_req = ath10k_wmi_10_2_op_gen_pdev_bss_chan_info,
 	.gen_echo = ath10k_wmi_op_gen_echo,
 	.gen_pdev_get_tpc_config = ath10k_wmi_10_2_4_op_gen_pdev_get_tpc_config,
+	.gen_gpio_config = ath10k_wmi_op_gen_gpio_config,
+	.gen_gpio_output = ath10k_wmi_op_gen_gpio_output,
 };
 
 int ath10k_wmi_attach(struct ath10k *ar)
diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h
index 2379501225a4..0faefc0a9a40 100644
--- a/drivers/net/wireless/ath/ath10k/wmi.h
+++ b/drivers/net/wireless/ath/ath10k/wmi.h
@@ -3034,6 +3034,41 @@ enum wmi_10_4_feature_mask {
 
 };
 
+/* WMI_GPIO_CONFIG_CMDID */
+enum {
+	WMI_GPIO_PULL_NONE,
+	WMI_GPIO_PULL_UP,
+	WMI_GPIO_PULL_DOWN,
+};
+
+enum {
+	WMI_GPIO_INTTYPE_DISABLE,
+	WMI_GPIO_INTTYPE_RISING_EDGE,
+	WMI_GPIO_INTTYPE_FALLING_EDGE,
+	WMI_GPIO_INTTYPE_BOTH_EDGE,
+	WMI_GPIO_INTTYPE_LEVEL_LOW,
+	WMI_GPIO_INTTYPE_LEVEL_HIGH
+};
+
+/* WMI_GPIO_CONFIG_CMDID */
+struct wmi_gpio_config_cmd {
+	__le32 gpio_num;             /* GPIO number to be setup */
+	__le32 input;                /* 0 - Output/ 1 - Input */
+	__le32 pull_type;            /* Pull type defined above */
+	__le32 intr_mode;            /* Interrupt mode defined above (Input) */
+} __packed;
+
+/* WMI_GPIO_OUTPUT_CMDID */
+struct wmi_gpio_output_cmd {
+	__le32 gpio_num;    /* GPIO number to be setup */
+	__le32 set;         /* Set the GPIO pin*/
+} __packed;
+
+/* WMI_GPIO_INPUT_EVENTID */
+struct wmi_gpio_input_event {
+	__le32 gpio_num;    /* GPIO number which changed state */
+} __packed;
+
 struct wmi_ext_resource_config_10_4_cmd {
 	/* contains enum wmi_host_platform_type */
 	__le32 host_platform_config;
-- 
cgit v1.2.3-58-ga151


From 3745d7963cc34c08ee4dda4f3c81981e277f0d96 Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Date: Wed, 8 May 2024 10:36:50 -0700
Subject: wifi: ath12k: add channel 2 into 6 GHz channel list

Add support for the 6 GHz channel 2 with center frequency
5935 MHz and operating class 136 per IEEE Std 802.11ax-2021,
Table E-4.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508173655.22191-2-quic_pradeepc@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.h | 2 +-
 drivers/net/wireless/ath/ath12k/mac.c  | 4 ++++
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index c6b7524ce3a0..6da242c0d51d 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -450,7 +450,7 @@ struct ath12k_sta {
 #define ATH12K_MIN_5G_FREQ 4150
 #define ATH12K_MIN_6G_FREQ 5925
 #define ATH12K_MAX_6G_FREQ 7115
-#define ATH12K_NUM_CHANS 100
+#define ATH12K_NUM_CHANS 101
 #define ATH12K_MAX_5G_CHAN 173
 
 enum ath12k_hw_state {
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 96dc5c2e096f..ae9f41a17b4f 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -91,6 +91,10 @@ static const struct ieee80211_channel ath12k_5ghz_channels[] = {
 };
 
 static const struct ieee80211_channel ath12k_6ghz_channels[] = {
+	/* Operating Class 136 */
+	CHAN6G(2, 5935, 0),
+
+	/* Operating Classes 131-135 */
 	CHAN6G(1, 5955, 0),
 	CHAN6G(5, 5975, 0),
 	CHAN6G(9, 5995, 0),
-- 
cgit v1.2.3-58-ga151


From c3c84a74bd797f76d7da036c9fef947d674bbc18 Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Date: Wed, 8 May 2024 10:36:51 -0700
Subject: wifi: ath12k: Correct 6 GHz frequency value in rx status

The frequency in the rx status is currently being filled
incorrectly for the 6 GHz band. The channel number received is
invalid in this case, resulting in packet drops. Fix this
issue by correcting the frequency calculation.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Signed-off-by: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508173655.22191-3-quic_pradeepc@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_rx.c |  9 ++++++---
 drivers/net/wireless/ath/ath12k/wmi.c   | 10 +++++++---
 2 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index 37205e894afe..d8998ac3a482 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -2359,8 +2359,10 @@ void ath12k_dp_rx_h_ppdu(struct ath12k *ar, struct hal_rx_desc *rx_desc,
 	channel_num = meta_data;
 	center_freq = meta_data >> 16;
 
-	if (center_freq >= 5935 && center_freq <= 7105) {
+	if (center_freq >= ATH12K_MIN_6G_FREQ &&
+	    center_freq <= ATH12K_MAX_6G_FREQ) {
 		rx_status->band = NL80211_BAND_6GHZ;
+		rx_status->freq = center_freq;
 	} else if (channel_num >= 1 && channel_num <= 14) {
 		rx_status->band = NL80211_BAND_2GHZ;
 	} else if (channel_num >= 36 && channel_num <= 173) {
@@ -2378,8 +2380,9 @@ void ath12k_dp_rx_h_ppdu(struct ath12k *ar, struct hal_rx_desc *rx_desc,
 				rx_desc, sizeof(*rx_desc));
 	}
 
-	rx_status->freq = ieee80211_channel_to_frequency(channel_num,
-							 rx_status->band);
+	if (rx_status->band != NL80211_BAND_6GHZ)
+		rx_status->freq = ieee80211_channel_to_frequency(channel_num,
+								 rx_status->band);
 
 	ath12k_dp_rx_h_rate(ar, rx_desc, rx_status);
 }
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 3004c7463be9..a592b75e5ef3 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -6022,8 +6022,10 @@ static void ath12k_mgmt_rx_event(struct ath12k_base *ab, struct sk_buff *skb)
 	if (rx_ev.status & WMI_RX_STATUS_ERR_MIC)
 		status->flag |= RX_FLAG_MMIC_ERROR;
 
-	if (rx_ev.chan_freq >= ATH12K_MIN_6G_FREQ) {
+	if (rx_ev.chan_freq >= ATH12K_MIN_6G_FREQ &&
+	    rx_ev.chan_freq <= ATH12K_MAX_6G_FREQ) {
 		status->band = NL80211_BAND_6GHZ;
+		status->freq = rx_ev.chan_freq;
 	} else if (rx_ev.channel >= 1 && rx_ev.channel <= 14) {
 		status->band = NL80211_BAND_2GHZ;
 	} else if (rx_ev.channel >= 36 && rx_ev.channel <= ATH12K_MAX_5G_CHAN) {
@@ -6044,8 +6046,10 @@ static void ath12k_mgmt_rx_event(struct ath12k_base *ab, struct sk_buff *skb)
 
 	sband = &ar->mac.sbands[status->band];
 
-	status->freq = ieee80211_channel_to_frequency(rx_ev.channel,
-						      status->band);
+	if (status->band != NL80211_BAND_6GHZ)
+		status->freq = ieee80211_channel_to_frequency(rx_ev.channel,
+							      status->band);
+
 	status->signal = rx_ev.snr + ATH12K_DEFAULT_NOISE_FLOOR;
 	status->rate_idx = ath12k_mac_bitrate_to_idx(sband, rx_ev.rate / 100);
 
-- 
cgit v1.2.3-58-ga151


From 70e3be54bbddc39123b8785408fb9aeb8f822ab6 Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Date: Wed, 8 May 2024 10:36:52 -0700
Subject: wifi: ath12k: fix survey dump collection in 6 GHz

When ath12k receives survey request, choose the 6 GHz band
when enabled. Without this, survey requests do not include
any 6 GHz band results, leading to failures in auto channel
selection.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508173655.22191-4-quic_pradeepc@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index ae9f41a17b4f..8af450cdace6 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -8031,6 +8031,17 @@ static int ath12k_mac_op_get_survey(struct ieee80211_hw *hw, int idx,
 
 	if (!sband)
 		sband = hw->wiphy->bands[NL80211_BAND_5GHZ];
+	if (sband && idx >= sband->n_channels) {
+		idx -= sband->n_channels;
+		sband = NULL;
+	}
+
+	if (!sband)
+		sband = hw->wiphy->bands[NL80211_BAND_6GHZ];
+	if (!sband || idx >= sband->n_channels) {
+		idx -= sband->n_channels;
+		sband = NULL;
+	}
 
 	if (!sband || idx >= sband->n_channels)
 		return -ENOENT;
-- 
cgit v1.2.3-58-ga151


From 112dbc6af807021ac65f46300d564163ec4710a9 Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Date: Wed, 8 May 2024 10:36:53 -0700
Subject: wifi: ath12k: add 6 GHz params in peer assoc command

Currently A-MPDU aggregation parameters are not being configured
during peer association for 6 GHz band. Hence, extract these
parameters from station's capabilities received in association
request and send to firmware. Without this, A-MPDU aggregation
is not happening in 6 GHz band.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Co-developed-by: Avula Sri Charan <quic_asrichar@quicinc.com>
Signed-off-by: Avula Sri Charan <quic_asrichar@quicinc.com>
Signed-off-by: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508173655.22191-5-quic_pradeepc@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 57 ++++++++++++++++++++++++++++++++++-
 1 file changed, 56 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 8af450cdace6..b9b9f4d17b34 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -2032,6 +2032,60 @@ static void ath12k_peer_assoc_h_he(struct ath12k *ar,
 	}
 }
 
+static void ath12k_peer_assoc_h_he_6ghz(struct ath12k *ar,
+					struct ieee80211_vif *vif,
+					struct ieee80211_sta *sta,
+					struct ath12k_wmi_peer_assoc_arg *arg)
+{
+	const struct ieee80211_sta_he_cap *he_cap = &sta->deflink.he_cap;
+	struct cfg80211_chan_def def;
+	enum nl80211_band band;
+	u8 ampdu_factor, mpdu_density;
+
+	if (WARN_ON(ath12k_mac_vif_chan(vif, &def)))
+		return;
+
+	band = def.chan->band;
+
+	if (!arg->he_flag || band != NL80211_BAND_6GHZ || !sta->deflink.he_6ghz_capa.capa)
+		return;
+
+	if (sta->deflink.bandwidth == IEEE80211_STA_RX_BW_40)
+		arg->bw_40 = true;
+
+	if (sta->deflink.bandwidth == IEEE80211_STA_RX_BW_80)
+		arg->bw_80 = true;
+
+	if (sta->deflink.bandwidth == IEEE80211_STA_RX_BW_160)
+		arg->bw_160 = true;
+
+	if (sta->deflink.bandwidth == IEEE80211_STA_RX_BW_320)
+		arg->bw_320 = true;
+
+	arg->peer_he_caps_6ghz = le16_to_cpu(sta->deflink.he_6ghz_capa.capa);
+
+	mpdu_density = u32_get_bits(arg->peer_he_caps_6ghz,
+				    IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START);
+	arg->peer_mpdu_density = ath12k_parse_mpdudensity(mpdu_density);
+
+	/* From IEEE Std 802.11ax-2021 - Section 10.12.2: An HE STA shall be capable of
+	 * receiving A-MPDU where the A-MPDU pre-EOF padding length is up to the value
+	 * indicated by the Maximum A-MPDU Length Exponent Extension field in the HE
+	 * Capabilities element and the Maximum A-MPDU Length Exponent field in HE 6 GHz
+	 * Band Capabilities element in the 6 GHz band.
+	 *
+	 * Here, we are extracting the Max A-MPDU Exponent Extension from HE caps and
+	 * factor is the Maximum A-MPDU Length Exponent from HE 6 GHZ Band capability.
+	 */
+	ampdu_factor = u8_get_bits(he_cap->he_cap_elem.mac_cap_info[3],
+				   IEEE80211_HE_MAC_CAP3_MAX_AMPDU_LEN_EXP_MASK) +
+			u32_get_bits(arg->peer_he_caps_6ghz,
+				     IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP);
+
+	arg->peer_max_mpdu = (1u << (IEEE80211_HE_6GHZ_MAX_AMPDU_FACTOR +
+				     ampdu_factor)) - 1;
+}
+
 static void ath12k_peer_assoc_h_smps(struct ieee80211_sta *sta,
 				     struct ath12k_wmi_peer_assoc_arg *arg)
 {
@@ -2504,6 +2558,7 @@ static void ath12k_peer_assoc_prepare(struct ath12k *ar,
 	ath12k_peer_assoc_h_ht(ar, vif, sta, arg);
 	ath12k_peer_assoc_h_vht(ar, vif, sta, arg);
 	ath12k_peer_assoc_h_he(ar, vif, sta, arg);
+	ath12k_peer_assoc_h_he_6ghz(ar, vif, sta, arg);
 	ath12k_peer_assoc_h_eht(ar, vif, sta, arg);
 	ath12k_peer_assoc_h_qos(ar, vif, sta, arg);
 	ath12k_peer_assoc_h_phymode(ar, vif, sta, arg);
@@ -8701,7 +8756,7 @@ static int ath12k_mac_hw_register(struct ath12k_hw *ah)
 	ieee80211_hw_set(hw, SUPPORTS_TX_FRAG);
 	ieee80211_hw_set(hw, REPORTS_LOW_ACK);
 
-	if (ht_cap & WMI_HT_CAP_ENABLED) {
+	if ((ht_cap & WMI_HT_CAP_ENABLED) || ar->supports_6ghz) {
 		ieee80211_hw_set(hw, AMPDU_AGGREGATION);
 		ieee80211_hw_set(hw, TX_AMPDU_SETUP_IN_HW);
 		ieee80211_hw_set(hw, SUPPORTS_REORDERING_BUFFER);
-- 
cgit v1.2.3-58-ga151


From f0e61dc7ecf9a709d22265a17d553fd314e3f986 Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Date: Wed, 8 May 2024 10:36:54 -0700
Subject: wifi: ath12k: refactor SMPS configuration

Move fetching SMPS value to a new function and use u16_get_bits
to extract it from capabilities. This will help in extending the
functionality when SMPS support in 6 GHz band gets added in
subsequent patches.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508173655.22191-6-quic_pradeepc@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 32 +++++++++++++++++++-------------
 1 file changed, 19 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index b9b9f4d17b34..0e9b6b7b83d7 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -2086,18 +2086,29 @@ static void ath12k_peer_assoc_h_he_6ghz(struct ath12k *ar,
 				     ampdu_factor)) - 1;
 }
 
+static int ath12k_get_smps_from_capa(const struct ieee80211_sta_ht_cap *ht_cap,
+				     int *smps)
+{
+	if (!ht_cap->ht_supported)
+		return -EOPNOTSUPP;
+
+	*smps = u16_get_bits(ht_cap->cap, IEEE80211_HT_CAP_SM_PS);
+
+	if (*smps >= ARRAY_SIZE(ath12k_smps_map))
+		return -EINVAL;
+
+	return 0;
+}
+
 static void ath12k_peer_assoc_h_smps(struct ieee80211_sta *sta,
 				     struct ath12k_wmi_peer_assoc_arg *arg)
 {
 	const struct ieee80211_sta_ht_cap *ht_cap = &sta->deflink.ht_cap;
 	int smps;
 
-	if (!ht_cap->ht_supported)
+	if (ath12k_get_smps_from_capa(ht_cap, &smps))
 		return;
 
-	smps = ht_cap->cap & IEEE80211_HT_CAP_SM_PS;
-	smps >>= IEEE80211_HT_CAP_SM_PS_SHIFT;
-
 	switch (smps) {
 	case WLAN_HT_CAP_SM_PS_STATIC:
 		arg->static_mimops_flag = true;
@@ -2571,16 +2582,11 @@ static int ath12k_setup_peer_smps(struct ath12k *ar, struct ath12k_vif *arvif,
 				  const u8 *addr,
 				  const struct ieee80211_sta_ht_cap *ht_cap)
 {
-	int smps;
+	int smps, ret = 0;
 
-	if (!ht_cap->ht_supported)
-		return 0;
-
-	smps = ht_cap->cap & IEEE80211_HT_CAP_SM_PS;
-	smps >>= IEEE80211_HT_CAP_SM_PS_SHIFT;
-
-	if (smps >= ARRAY_SIZE(ath12k_smps_map))
-		return -EINVAL;
+	ret = ath12k_get_smps_from_capa(ht_cap, &smps);
+	if (ret < 0)
+		return ret;
 
 	return ath12k_wmi_set_peer_param(ar, addr, arvif->vdev_id,
 					 WMI_PEER_MIMO_PS_STATE,
-- 
cgit v1.2.3-58-ga151


From 97b7cbb7a3cbeee87ed82af67bca881ec8881d83 Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Date: Wed, 8 May 2024 10:36:55 -0700
Subject: wifi: ath12k: support SMPS configuration for 6 GHz

Parse SMPS configuration from IEs and configure. Without this,
SMPS is not enabled for 6 GHz band. This is disabled for
WCN7850 as hardware does not support it.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508173655.22191-7-quic_pradeepc@quicinc.com
---
 drivers/net/wireless/ath/ath12k/hw.c  |  3 +++
 drivers/net/wireless/ath/ath12k/hw.h  |  1 +
 drivers/net/wireless/ath/ath12k/mac.c | 26 ++++++++++++++++++--------
 3 files changed, 22 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/hw.c b/drivers/net/wireless/ath/ath12k/hw.c
index 06f443216488..b3ba753d4a85 100644
--- a/drivers/net/wireless/ath/ath12k/hw.c
+++ b/drivers/net/wireless/ath/ath12k/hw.c
@@ -926,6 +926,7 @@ static const struct ath12k_hw_params ath12k_hw_params[] = {
 		.supports_sta_ps = false,
 
 		.acpi_guid = NULL,
+		.supports_dynamic_smps_6ghz = true,
 	},
 	{
 		.name = "wcn7850 hw2.0",
@@ -1001,6 +1002,7 @@ static const struct ath12k_hw_params ath12k_hw_params[] = {
 		.supports_sta_ps = true,
 
 		.acpi_guid = &wcn7850_uuid,
+		.supports_dynamic_smps_6ghz = false,
 	},
 	{
 		.name = "qcn9274 hw2.0",
@@ -1071,6 +1073,7 @@ static const struct ath12k_hw_params ath12k_hw_params[] = {
 		.supports_sta_ps = false,
 
 		.acpi_guid = NULL,
+		.supports_dynamic_smps_6ghz = true,
 	},
 };
 
diff --git a/drivers/net/wireless/ath/ath12k/hw.h b/drivers/net/wireless/ath/ath12k/hw.h
index 2a5a9d638d93..7422d3258548 100644
--- a/drivers/net/wireless/ath/ath12k/hw.h
+++ b/drivers/net/wireless/ath/ath12k/hw.h
@@ -215,6 +215,7 @@ struct ath12k_hw_params {
 	bool supports_sta_ps;
 
 	const guid_t *acpi_guid;
+	bool supports_dynamic_smps_6ghz;
 };
 
 struct ath12k_hw_ops {
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 0e9b6b7b83d7..ea0a9ccc8e92 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -2087,12 +2087,17 @@ static void ath12k_peer_assoc_h_he_6ghz(struct ath12k *ar,
 }
 
 static int ath12k_get_smps_from_capa(const struct ieee80211_sta_ht_cap *ht_cap,
+				     const struct ieee80211_he_6ghz_capa *he_6ghz_capa,
 				     int *smps)
 {
-	if (!ht_cap->ht_supported)
+	if (!ht_cap->ht_supported && !he_6ghz_capa->capa)
 		return -EOPNOTSUPP;
 
-	*smps = u16_get_bits(ht_cap->cap, IEEE80211_HT_CAP_SM_PS);
+	if (ht_cap->ht_supported)
+		*smps = u16_get_bits(ht_cap->cap, IEEE80211_HT_CAP_SM_PS);
+	else
+		*smps = le16_get_bits(he_6ghz_capa->capa,
+				      IEEE80211_HE_6GHZ_CAP_SM_PS);
 
 	if (*smps >= ARRAY_SIZE(ath12k_smps_map))
 		return -EINVAL;
@@ -2103,10 +2108,11 @@ static int ath12k_get_smps_from_capa(const struct ieee80211_sta_ht_cap *ht_cap,
 static void ath12k_peer_assoc_h_smps(struct ieee80211_sta *sta,
 				     struct ath12k_wmi_peer_assoc_arg *arg)
 {
+	const struct ieee80211_he_6ghz_capa *he_6ghz_capa = &sta->deflink.he_6ghz_capa;
 	const struct ieee80211_sta_ht_cap *ht_cap = &sta->deflink.ht_cap;
 	int smps;
 
-	if (ath12k_get_smps_from_capa(ht_cap, &smps))
+	if (ath12k_get_smps_from_capa(ht_cap, he_6ghz_capa, &smps))
 		return;
 
 	switch (smps) {
@@ -2580,11 +2586,12 @@ static void ath12k_peer_assoc_prepare(struct ath12k *ar,
 
 static int ath12k_setup_peer_smps(struct ath12k *ar, struct ath12k_vif *arvif,
 				  const u8 *addr,
-				  const struct ieee80211_sta_ht_cap *ht_cap)
+				  const struct ieee80211_sta_ht_cap *ht_cap,
+				  const struct ieee80211_he_6ghz_capa *he_6ghz_capa)
 {
 	int smps, ret = 0;
 
-	ret = ath12k_get_smps_from_capa(ht_cap, &smps);
+	ret = ath12k_get_smps_from_capa(ht_cap, he_6ghz_capa, &smps);
 	if (ret < 0)
 		return ret;
 
@@ -2637,7 +2644,8 @@ static void ath12k_bss_assoc(struct ath12k *ar,
 	}
 
 	ret = ath12k_setup_peer_smps(ar, arvif, bss_conf->bssid,
-				     &ap_sta->deflink.ht_cap);
+				     &ap_sta->deflink.ht_cap,
+				     &ap_sta->deflink.he_6ghz_capa);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to setup peer SMPS for vdev %d: %d\n",
 			    arvif->vdev_id, ret);
@@ -3944,7 +3952,8 @@ static int ath12k_station_assoc(struct ath12k *ar,
 		return 0;
 
 	ret = ath12k_setup_peer_smps(ar, arvif, sta->addr,
-				     &sta->deflink.ht_cap);
+				     &sta->deflink.ht_cap,
+				     &sta->deflink.he_6ghz_capa);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to setup peer SMPS for vdev %d: %d\n",
 			    arvif->vdev_id, ret);
@@ -8777,7 +8786,8 @@ static int ath12k_mac_hw_register(struct ath12k_hw *ah)
 	 * for each band for a dual band capable radio. It will be tricky to
 	 * handle it when the ht capability different for each band.
 	 */
-	if (ht_cap & WMI_HT_CAP_DYNAMIC_SMPS)
+	if (ht_cap & WMI_HT_CAP_DYNAMIC_SMPS ||
+	    (ar->supports_6ghz && ab->hw_params->supports_dynamic_smps_6ghz))
 		wiphy->features |= NL80211_FEATURE_DYNAMIC_SMPS;
 
 	wiphy->max_scan_ssids = WLAN_SCAN_PARAMS_MAX_SSID;
-- 
cgit v1.2.3-58-ga151


From 519a545cfee790024c623c7aacd165f7431773d5 Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Wed, 8 May 2024 13:29:04 -0700
Subject: wifi: ath12k: advertise driver capabilities for MBSSID and EMA

Advertise the driver support for multiple BSSID (MBSSID) and
enhanced multi-BSSID advertisements (EMA) by setting extended
capabilities.

Configure mbssid_max_interfaces and ema_max_profile_periodicity
fields in structure wiphy which are used to advertise maximum number
of interfaces and profile periodicity supported by the driver.

Add new WMI fields to configure maximum vdev count supported for
MBSSID and profile periodicity in case of EMA.

Set WMI_RSRC_CFG_FLAGS2_CALC_NEXT_DTIM_COUNT_SET flag to allow
firmware to track and update the DTIM counts for each nontransmitted
profile.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508202912.11902-2-quic_alokad@quicinc.com
---
 drivers/net/wireless/ath/ath12k/hw.h  | 1 +
 drivers/net/wireless/ath/ath12k/mac.c | 7 +++++++
 drivers/net/wireless/ath/ath12k/wmi.c | 7 ++++++-
 drivers/net/wireless/ath/ath12k/wmi.h | 3 +++
 4 files changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/hw.h b/drivers/net/wireless/ath/ath12k/hw.h
index 7422d3258548..af33bf11416b 100644
--- a/drivers/net/wireless/ath/ath12k/hw.h
+++ b/drivers/net/wireless/ath/ath12k/hw.h
@@ -80,6 +80,7 @@
 #define TARGET_RX_BATCHMODE		1
 #define TARGET_RX_PEER_METADATA_VER_V1A	2
 #define TARGET_RX_PEER_METADATA_VER_V1B	3
+#define TARGET_EMA_MAX_PROFILE_PERIOD	8
 
 #define ATH12K_HW_DEFAULT_QUEUE		0
 #define ATH12K_HW_MAX_QUEUES		4
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index ea0a9ccc8e92..9d82f289b23b 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -8574,19 +8574,23 @@ static int ath12k_mac_setup_iface_combinations(struct ath12k_hw *ah)
 
 static const u8 ath12k_if_types_ext_capa[] = {
 	[0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING,
+	[2] = WLAN_EXT_CAPA3_MULTI_BSSID_SUPPORT,
 	[7] = WLAN_EXT_CAPA8_OPMODE_NOTIF,
 };
 
 static const u8 ath12k_if_types_ext_capa_sta[] = {
 	[0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING,
+	[2] = WLAN_EXT_CAPA3_MULTI_BSSID_SUPPORT,
 	[7] = WLAN_EXT_CAPA8_OPMODE_NOTIF,
 	[9] = WLAN_EXT_CAPA10_TWT_REQUESTER_SUPPORT,
 };
 
 static const u8 ath12k_if_types_ext_capa_ap[] = {
 	[0] = WLAN_EXT_CAPA1_EXT_CHANNEL_SWITCHING,
+	[2] = WLAN_EXT_CAPA3_MULTI_BSSID_SUPPORT,
 	[7] = WLAN_EXT_CAPA8_OPMODE_NOTIF,
 	[9] = WLAN_EXT_CAPA10_TWT_RESPONDER_SUPPORT,
+	[10] = WLAN_EXT_CAPA11_EMA_SUPPORT,
 };
 
 static const struct wiphy_iftype_ext_capab ath12k_iftypes_ext_capa[] = {
@@ -8826,6 +8830,9 @@ static int ath12k_mac_hw_register(struct ath12k_hw *ah)
 	wiphy->iftype_ext_capab = ath12k_iftypes_ext_capa;
 	wiphy->num_iftype_ext_capab = ARRAY_SIZE(ath12k_iftypes_ext_capa);
 
+	wiphy->mbssid_max_interfaces = TARGET_NUM_VDEVS;
+	wiphy->ema_max_profile_periodicity = TARGET_EMA_MAX_PROFILE_PERIOD;
+
 	if (is_6ghz) {
 		wiphy_ext_feature_set(wiphy,
 				      NL80211_EXT_FEATURE_FILS_DISCOVERY);
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index a592b75e5ef3..3da16d4abb5a 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -228,6 +228,9 @@ void ath12k_wmi_init_qcn9274(struct ath12k_base *ab,
 	config->peer_map_unmap_version = 0x32;
 	config->twt_ap_pdev_count = ab->num_radios;
 	config->twt_ap_sta_count = 1000;
+	config->ema_max_vap_cnt = ab->num_radios;
+	config->ema_max_profile_period = TARGET_EMA_MAX_PROFILE_PERIOD;
+	config->beacon_tx_offload_max_vdev += config->ema_max_vap_cnt;
 
 	if (test_bit(WMI_TLV_SERVICE_PEER_METADATA_V1A_V1B_SUPPORT, ab->wmi_ab.svc_map))
 		config->dp_peer_meta_data_ver = TARGET_RX_PEER_METADATA_VER_V1B;
@@ -3475,9 +3478,11 @@ ath12k_wmi_copy_resource_config(struct ath12k_wmi_resource_config_params *wmi_cf
 	wmi_cfg->twt_ap_sta_count = cpu_to_le32(tg_cfg->twt_ap_sta_count);
 	wmi_cfg->flags2 = le32_encode_bits(tg_cfg->dp_peer_meta_data_ver,
 					   WMI_RSRC_CFG_FLAGS2_RX_PEER_METADATA_VERSION);
-
 	wmi_cfg->host_service_flags = cpu_to_le32(tg_cfg->is_reg_cc_ext_event_supported <<
 				WMI_RSRC_CFG_HOST_SVC_FLAG_REG_CC_EXT_SUPPORT_BIT);
+	wmi_cfg->ema_max_vap_cnt = cpu_to_le32(tg_cfg->ema_max_vap_cnt);
+	wmi_cfg->ema_max_profile_period = cpu_to_le32(tg_cfg->ema_max_profile_period);
+	wmi_cfg->flags2 |= cpu_to_le32(WMI_RSRC_CFG_FLAGS2_CALC_NEXT_DTIM_COUNT_SET);
 }
 
 static int ath12k_init_cmd_send(struct ath12k_wmi_pdev *wmi,
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 496866673aea..e71e6c73f249 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -2356,6 +2356,8 @@ struct ath12k_wmi_resource_config_arg {
 	u32 twt_ap_sta_count;
 	bool is_reg_cc_ext_event_supported;
 	u8  dp_peer_meta_data_ver;
+	u32 ema_max_vap_cnt;
+	u32 ema_max_profile_period;
 };
 
 struct ath12k_wmi_init_cmd_arg {
@@ -2410,6 +2412,7 @@ struct wmi_init_cmd {
 #define WMI_RSRC_CFG_HOST_SVC_FLAG_REG_CC_EXT_SUPPORT_BIT 4
 #define WMI_RSRC_CFG_FLAGS2_RX_PEER_METADATA_VERSION		GENMASK(5, 4)
 #define WMI_RSRC_CFG_FLAG1_BSS_CHANNEL_INFO_64	BIT(5)
+#define WMI_RSRC_CFG_FLAGS2_CALC_NEXT_DTIM_COUNT_SET      BIT(9)
 
 struct ath12k_wmi_resource_config_params {
 	__le32 tlv_header;
-- 
cgit v1.2.3-58-ga151


From ea68f1bba61632652f722e5979a8a853cfa55eff Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Wed, 8 May 2024 13:29:05 -0700
Subject: wifi: ath12k: configure MBSSID params in vdev create/start

Configure multiple BSSID flags and vdev_id of the transmited profile
in vdev create/start commands depending on the service bit
WMI_TLV_SERVICE_MBSS_PARAM_IN_VDEV_START_SUPPORT.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508202912.11902-3-quic_alokad@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 66 +++++++++++++++++++++++++++++++----
 drivers/net/wireless/ath/ath12k/wmi.c |  3 ++
 drivers/net/wireless/ath/ath12k/wmi.h |  7 ++++
 3 files changed, 70 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 9d82f289b23b..5967c94fe8f2 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -5984,17 +5984,59 @@ ath12k_mac_get_vdev_stats_id(struct ath12k_vif *arvif)
 	return vdev_stats_id;
 }
 
-static void ath12k_mac_setup_vdev_create_arg(struct ath12k_vif *arvif,
-					     struct ath12k_wmi_vdev_create_arg *arg)
+static int ath12k_mac_setup_vdev_params_mbssid(struct ath12k_vif *arvif,
+					       u32 *flags, u32 *tx_vdev_id)
+{
+	struct ieee80211_vif *tx_vif = arvif->vif->mbssid_tx_vif;
+	struct ath12k *ar = arvif->ar;
+	struct ath12k_vif *tx_arvif;
+
+	if (!tx_vif)
+		return 0;
+
+	tx_arvif = ath12k_vif_to_arvif(tx_vif);
+
+	if (arvif->vif->bss_conf.nontransmitted) {
+		if (ar->ah->hw->wiphy != ieee80211_vif_to_wdev(tx_vif)->wiphy)
+			return -EINVAL;
+
+		*flags = WMI_VDEV_MBSSID_FLAGS_NON_TRANSMIT_AP;
+		*tx_vdev_id = tx_arvif->vdev_id;
+	} else if (tx_arvif == arvif) {
+		*flags = WMI_VDEV_MBSSID_FLAGS_TRANSMIT_AP;
+	} else {
+		return -EINVAL;
+	}
+
+	if (arvif->vif->bss_conf.ema_ap)
+		*flags |= WMI_VDEV_MBSSID_FLAGS_EMA_MODE;
+
+	return 0;
+}
+
+static int ath12k_mac_setup_vdev_create_arg(struct ath12k_vif *arvif,
+					    struct ath12k_wmi_vdev_create_arg *arg)
 {
 	struct ath12k *ar = arvif->ar;
 	struct ath12k_pdev *pdev = ar->pdev;
+	int ret;
 
 	arg->if_id = arvif->vdev_id;
 	arg->type = arvif->vdev_type;
 	arg->subtype = arvif->vdev_subtype;
 	arg->pdev_id = pdev->pdev_id;
 
+	arg->mbssid_flags = WMI_VDEV_MBSSID_FLAGS_NON_MBSSID_AP;
+	arg->mbssid_tx_vdev_id = 0;
+	if (!test_bit(WMI_TLV_SERVICE_MBSS_PARAM_IN_VDEV_START_SUPPORT,
+		      ar->ab->wmi_ab.svc_map)) {
+		ret = ath12k_mac_setup_vdev_params_mbssid(arvif,
+							  &arg->mbssid_flags,
+							  &arg->mbssid_tx_vdev_id);
+		if (ret)
+			return ret;
+	}
+
 	if (pdev->cap.supported_bands & WMI_HOST_WLAN_2G_CAP) {
 		arg->chains[NL80211_BAND_2GHZ].tx = ar->num_tx_chains;
 		arg->chains[NL80211_BAND_2GHZ].rx = ar->num_rx_chains;
@@ -6010,6 +6052,7 @@ static void ath12k_mac_setup_vdev_create_arg(struct ath12k_vif *arvif,
 	}
 
 	arg->if_stats_id = ath12k_mac_get_vdev_stats_id(arvif);
+	return 0;
 }
 
 static u32
@@ -6191,7 +6234,12 @@ static int ath12k_mac_vdev_create(struct ath12k *ar, struct ieee80211_vif *vif)
 	for (i = 0; i < ARRAY_SIZE(vif->hw_queue); i++)
 		vif->hw_queue[i] = i % (ATH12K_HW_MAX_QUEUES - 1);
 
-	ath12k_mac_setup_vdev_create_arg(arvif, &vdev_arg);
+	ret = ath12k_mac_setup_vdev_create_arg(arvif, &vdev_arg);
+	if (ret) {
+		ath12k_warn(ab, "failed to create vdev parameters %d: %d\n",
+			    arvif->vdev_id, ret);
+		goto err;
+	}
 
 	ret = ath12k_wmi_vdev_create(ar, vif->addr, &vdev_arg);
 	if (ret) {
@@ -6933,10 +6981,16 @@ ath12k_mac_vdev_start_restart(struct ath12k_vif *arvif,
 	arg.pref_tx_streams = ar->num_tx_chains;
 	arg.pref_rx_streams = ar->num_rx_chains;
 
-	/* Fill the MBSSID flags to indicate AP is non MBSSID by default
-	 * Corresponding flags would be updated with MBSSID support.
-	 */
 	arg.mbssid_flags = WMI_VDEV_MBSSID_FLAGS_NON_MBSSID_AP;
+	arg.mbssid_tx_vdev_id = 0;
+	if (test_bit(WMI_TLV_SERVICE_MBSS_PARAM_IN_VDEV_START_SUPPORT,
+		     ar->ab->wmi_ab.svc_map)) {
+		ret = ath12k_mac_setup_vdev_params_mbssid(arvif,
+							  &arg.mbssid_flags,
+							  &arg.mbssid_tx_vdev_id);
+		if (ret)
+			return ret;
+	}
 
 	if (arvif->vdev_type == WMI_VDEV_TYPE_AP) {
 		arg.ssid = arvif->u.ap.ssid;
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 3da16d4abb5a..e77132d6aa83 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -844,6 +844,8 @@ int ath12k_wmi_vdev_create(struct ath12k *ar, u8 *macaddr,
 	cmd->vdev_subtype = cpu_to_le32(args->subtype);
 	cmd->num_cfg_txrx_streams = cpu_to_le32(WMI_NUM_SUPPORTED_BAND_MAX);
 	cmd->pdev_id = cpu_to_le32(args->pdev_id);
+	cmd->mbssid_flags = cpu_to_le32(args->mbssid_flags);
+	cmd->mbssid_tx_vdev_id = cpu_to_le32(args->mbssid_tx_vdev_id);
 	cmd->vdev_stats_id = cpu_to_le32(args->if_stats_id);
 	ether_addr_copy(cmd->vdev_macaddr.addr, macaddr);
 
@@ -1049,6 +1051,7 @@ int ath12k_wmi_vdev_start(struct ath12k *ar, struct wmi_vdev_start_req_arg *arg,
 	cmd->he_ops = cpu_to_le32(arg->he_ops);
 	cmd->punct_bitmap = cpu_to_le32(arg->punct_bitmap);
 	cmd->mbssid_flags = cpu_to_le32(arg->mbssid_flags);
+	cmd->mbssid_tx_vdev_id = cpu_to_le32(arg->mbssid_tx_vdev_id);
 
 	if (!restart) {
 		if (arg->ssid) {
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index e71e6c73f249..86a0ef09aa39 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -2154,6 +2154,7 @@ enum wmi_tlv_service {
 	WMI_TLV_SERVICE_PER_PEER_HTT_STATS_RESET = 213,
 	WMI_TLV_SERVICE_FREQINFO_IN_METADATA = 219,
 	WMI_TLV_SERVICE_EXT2_MSG = 220,
+	WMI_TLV_SERVICE_MBSS_PARAM_IN_VDEV_START_SUPPORT = 253,
 
 	WMI_MAX_EXT_SERVICE = 256,
 
@@ -2729,6 +2730,8 @@ struct ath12k_wmi_vdev_create_arg {
 	} chains[NUM_NL80211_BANDS];
 	u32 pdev_id;
 	u8 if_stats_id;
+	u32 mbssid_flags;
+	u32 mbssid_tx_vdev_id;
 };
 
 #define ATH12K_MAX_VDEV_STATS_ID	0x30
@@ -2795,6 +2798,10 @@ struct ath12k_wmi_ssid_params {
 
 enum wmi_vdev_mbssid_flags {
 	WMI_VDEV_MBSSID_FLAGS_NON_MBSSID_AP	= BIT(0),
+	WMI_VDEV_MBSSID_FLAGS_TRANSMIT_AP	= BIT(1),
+	WMI_VDEV_MBSSID_FLAGS_NON_TRANSMIT_AP	= BIT(2),
+	WMI_VDEV_MBSSID_FLAGS_EMA_MODE		= BIT(3),
+	WMI_VDEV_MBSSID_FLAGS_SCAN_MODE_VAP	= BIT(4),
 };
 
 struct wmi_vdev_start_request_cmd {
-- 
cgit v1.2.3-58-ga151


From 5fbd97f25c4c62eba921b8bc13a45f0210437787 Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Wed, 8 May 2024 13:29:06 -0700
Subject: wifi: ath12k: rename MBSSID fields in wmi_vdev_up_cmd

Rename trans_bssid to tx_vdev_bssid to make it similar to vdev_bssid.

Rename profile_num to nontx_profile_cnt, and profile_idx to
nontx_profile_idx which make it clear that these configurations
correspond to nontransmitted MBSSID profiles.

These members are currently unused in the driver.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508202912.11902-4-quic_alokad@quicinc.com
---
 drivers/net/wireless/ath/ath12k/wmi.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 86a0ef09aa39..4051ed5a0b0a 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -2768,9 +2768,9 @@ struct wmi_vdev_up_cmd {
 	__le32 vdev_id;
 	__le32 vdev_assoc_id;
 	struct ath12k_wmi_mac_addr_params vdev_bssid;
-	struct ath12k_wmi_mac_addr_params trans_bssid;
-	__le32 profile_idx;
-	__le32 profile_num;
+	struct ath12k_wmi_mac_addr_params tx_vdev_bssid;
+	__le32 nontx_profile_idx;
+	__le32 nontx_profile_cnt;
 } __packed;
 
 struct wmi_vdev_stop_cmd {
-- 
cgit v1.2.3-58-ga151


From 51c47463543ef4f1f0cf2354527baa297a754995 Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Wed, 8 May 2024 13:29:07 -0700
Subject: wifi: ath12k: create a structure for WMI vdev up parameters

Host needs to send multiple BSSID configurations to firmware for
each vdev in ath12k_wmi_vdev_up() for AP mode. This function accepts
individual input parameters hence any new argument will require changes
to all callers. Most of these will use default value (0 or NULL).
Create a structure for the function arguments and include objects with
all members initialized to zero to minimize future changes.

Tested-on : QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508202912.11902-5-quic_alokad@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 30 +++++++++++++++++++++++-------
 drivers/net/wireless/ath/ath12k/wmi.c | 10 +++++-----
 drivers/net/wireless/ath/ath12k/wmi.h |  9 +++++++--
 3 files changed, 35 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 5967c94fe8f2..c6fa68aafdd8 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -867,9 +867,12 @@ static int ath12k_mac_vdev_setup_sync(struct ath12k *ar)
 
 static int ath12k_monitor_vdev_up(struct ath12k *ar, int vdev_id)
 {
+	struct ath12k_wmi_vdev_up_params params = {};
 	int ret;
 
-	ret = ath12k_wmi_vdev_up(ar, vdev_id, 0, ar->mac_addr);
+	params.vdev_id = vdev_id;
+	params.bssid = ar->mac_addr;
+	ret = ath12k_wmi_vdev_up(ar, &params);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to put up monitor vdev %i: %d\n",
 			    vdev_id, ret);
@@ -886,6 +889,7 @@ static int ath12k_mac_monitor_vdev_start(struct ath12k *ar, int vdev_id,
 {
 	struct ieee80211_channel *channel;
 	struct wmi_vdev_start_req_arg arg = {};
+	struct ath12k_wmi_vdev_up_params params = {};
 	int ret;
 
 	lockdep_assert_held(&ar->conf_mutex);
@@ -926,7 +930,9 @@ static int ath12k_mac_monitor_vdev_start(struct ath12k *ar, int vdev_id,
 		return ret;
 	}
 
-	ret = ath12k_wmi_vdev_up(ar, vdev_id, 0, ar->mac_addr);
+	params.vdev_id = vdev_id;
+	params.bssid = ar->mac_addr;
+	ret = ath12k_wmi_vdev_up(ar, &params);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to put up monitor vdev %i: %d\n",
 			    vdev_id, ret);
@@ -1362,6 +1368,7 @@ free_bcn_skb:
 static void ath12k_control_beaconing(struct ath12k_vif *arvif,
 				     struct ieee80211_bss_conf *info)
 {
+	struct ath12k_wmi_vdev_up_params params = {};
 	struct ath12k *ar = arvif->ar;
 	int ret;
 
@@ -1389,8 +1396,10 @@ static void ath12k_control_beaconing(struct ath12k_vif *arvif,
 
 	ether_addr_copy(arvif->bssid, info->bssid);
 
-	ret = ath12k_wmi_vdev_up(arvif->ar, arvif->vdev_id, arvif->aid,
-				 arvif->bssid);
+	params.vdev_id = arvif->vdev_id;
+	params.aid = arvif->aid;
+	params.bssid = arvif->bssid;
+	ret = ath12k_wmi_vdev_up(arvif->ar, &params);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to bring up vdev %d: %i\n",
 			    arvif->vdev_id, ret);
@@ -2605,6 +2614,7 @@ static void ath12k_bss_assoc(struct ath12k *ar,
 			     struct ieee80211_bss_conf *bss_conf)
 {
 	struct ieee80211_vif *vif = arvif->vif;
+	struct ath12k_wmi_vdev_up_params params = {};
 	struct ath12k_wmi_peer_assoc_arg peer_arg;
 	struct ieee80211_sta *ap_sta;
 	struct ath12k_peer *peer;
@@ -2657,7 +2667,10 @@ static void ath12k_bss_assoc(struct ath12k *ar,
 	arvif->aid = vif->cfg.aid;
 	ether_addr_copy(arvif->bssid, bss_conf->bssid);
 
-	ret = ath12k_wmi_vdev_up(ar, arvif->vdev_id, arvif->aid, arvif->bssid);
+	params.vdev_id = arvif->vdev_id;
+	params.aid = arvif->aid;
+	params.bssid = arvif->bssid;
+	ret = ath12k_wmi_vdev_up(ar, &params);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to set vdev %d up: %d\n",
 			    arvif->vdev_id, ret);
@@ -7184,6 +7197,7 @@ ath12k_mac_update_vif_chan(struct ath12k *ar,
 			   struct ieee80211_vif_chanctx_switch *vifs,
 			   int n_vifs)
 {
+	struct ath12k_wmi_vdev_up_params params = {};
 	struct ath12k_base *ab = ar->ab;
 	struct ath12k_vif *arvif;
 	int ret;
@@ -7264,8 +7278,10 @@ ath12k_mac_update_vif_chan(struct ath12k *ar,
 			ath12k_warn(ab, "failed to update bcn tmpl during csa: %d\n",
 				    ret);
 
-		ret = ath12k_wmi_vdev_up(arvif->ar, arvif->vdev_id, arvif->aid,
-					 arvif->bssid);
+		params.vdev_id = arvif->vdev_id;
+		params.aid = arvif->aid;
+		params.bssid = arvif->bssid;
+		ret = ath12k_wmi_vdev_up(arvif->ar, &params);
 		if (ret) {
 			ath12k_warn(ab, "failed to bring vdev up %d: %d\n",
 				    arvif->vdev_id, ret);
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index e77132d6aa83..6e73194a849e 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -1103,7 +1103,7 @@ int ath12k_wmi_vdev_start(struct ath12k *ar, struct wmi_vdev_start_req_arg *arg,
 	return ret;
 }
 
-int ath12k_wmi_vdev_up(struct ath12k *ar, u32 vdev_id, u32 aid, const u8 *bssid)
+int ath12k_wmi_vdev_up(struct ath12k *ar, struct ath12k_wmi_vdev_up_params *params)
 {
 	struct ath12k_wmi_pdev *wmi = ar->wmi;
 	struct wmi_vdev_up_cmd *cmd;
@@ -1118,14 +1118,14 @@ int ath12k_wmi_vdev_up(struct ath12k *ar, u32 vdev_id, u32 aid, const u8 *bssid)
 
 	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_VDEV_UP_CMD,
 						 sizeof(*cmd));
-	cmd->vdev_id = cpu_to_le32(vdev_id);
-	cmd->vdev_assoc_id = cpu_to_le32(aid);
+	cmd->vdev_id = cpu_to_le32(params->vdev_id);
+	cmd->vdev_assoc_id = cpu_to_le32(params->aid);
 
-	ether_addr_copy(cmd->vdev_bssid.addr, bssid);
+	ether_addr_copy(cmd->vdev_bssid.addr, params->bssid);
 
 	ath12k_dbg(ar->ab, ATH12K_DBG_WMI,
 		   "WMI mgmt vdev up id 0x%x assoc id %d bssid %pM\n",
-		   vdev_id, aid, bssid);
+		   params->vdev_id, params->aid, params->bssid);
 
 	ret = ath12k_wmi_cmd_send(wmi, skb, WMI_VDEV_UP_CMDID);
 	if (ret) {
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 4051ed5a0b0a..12c6cfb943c8 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -2763,6 +2763,12 @@ struct wmi_vdev_delete_cmd {
 	__le32 vdev_id;
 } __packed;
 
+struct ath12k_wmi_vdev_up_params {
+	u32 vdev_id;
+	u32 aid;
+	const u8 *bssid;
+};
+
 struct wmi_vdev_up_cmd {
 	__le32 tlv_header;
 	__le32 vdev_id;
@@ -4893,8 +4899,7 @@ int ath12k_wmi_bcn_tmpl(struct ath12k *ar, u32 vdev_id,
 			struct ieee80211_mutable_offsets *offs,
 			struct sk_buff *bcn);
 int ath12k_wmi_vdev_down(struct ath12k *ar, u8 vdev_id);
-int ath12k_wmi_vdev_up(struct ath12k *ar, u32 vdev_id, u32 aid,
-		       const u8 *bssid);
+int ath12k_wmi_vdev_up(struct ath12k *ar, struct ath12k_wmi_vdev_up_params *params);
 int ath12k_wmi_vdev_stop(struct ath12k *ar, u8 vdev_id);
 int ath12k_wmi_vdev_start(struct ath12k *ar, struct wmi_vdev_start_req_arg *arg,
 			  bool restart);
-- 
cgit v1.2.3-58-ga151


From 84edf47bcb31746a4070bc4783546132d855ed4b Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Wed, 8 May 2024 13:29:08 -0700
Subject: wifi: ath12k: configure MBSSID parameters in AP mode

Include MBSSID parameters in WMI vdev up operation.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508202912.11902-6-quic_alokad@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 15 ++++++++++++++-
 drivers/net/wireless/ath/ath12k/wmi.c |  6 ++++++
 drivers/net/wireless/ath/ath12k/wmi.h |  3 +++
 3 files changed, 23 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index c6fa68aafdd8..3f57acc16c40 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -1399,6 +1399,11 @@ static void ath12k_control_beaconing(struct ath12k_vif *arvif,
 	params.vdev_id = arvif->vdev_id;
 	params.aid = arvif->aid;
 	params.bssid = arvif->bssid;
+	if (arvif->vif->mbssid_tx_vif) {
+		params.tx_bssid = ath12k_vif_to_arvif(arvif->vif->mbssid_tx_vif)->bssid;
+		params.nontx_profile_idx = info->bssid_index;
+		params.nontx_profile_cnt = 1 << info->bssid_indicator;
+	}
 	ret = ath12k_wmi_vdev_up(arvif->ar, &params);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to bring up vdev %d: %i\n",
@@ -7199,6 +7204,7 @@ ath12k_mac_update_vif_chan(struct ath12k *ar,
 {
 	struct ath12k_wmi_vdev_up_params params = {};
 	struct ath12k_base *ab = ar->ab;
+	struct ieee80211_vif *vif;
 	struct ath12k_vif *arvif;
 	int ret;
 	int i;
@@ -7241,7 +7247,8 @@ ath12k_mac_update_vif_chan(struct ath12k *ar,
 	/* TODO: Update ar->rx_channel */
 
 	for (i = 0; i < n_vifs; i++) {
-		arvif = ath12k_vif_to_arvif(vifs[i].vif);
+		vif = vifs[i].vif;
+		arvif = ath12k_vif_to_arvif(vif);
 
 		if (WARN_ON(!arvif->is_started))
 			continue;
@@ -7278,9 +7285,15 @@ ath12k_mac_update_vif_chan(struct ath12k *ar,
 			ath12k_warn(ab, "failed to update bcn tmpl during csa: %d\n",
 				    ret);
 
+		memset(&params, 0, sizeof(params));
 		params.vdev_id = arvif->vdev_id;
 		params.aid = arvif->aid;
 		params.bssid = arvif->bssid;
+		if (vif->mbssid_tx_vif) {
+			params.tx_bssid = ath12k_vif_to_arvif(vif->mbssid_tx_vif)->bssid;
+			params.nontx_profile_idx = vif->bss_conf.bssid_index;
+			params.nontx_profile_cnt = 1 << vif->bss_conf.bssid_indicator;
+		}
 		ret = ath12k_wmi_vdev_up(arvif->ar, &params);
 		if (ret) {
 			ath12k_warn(ab, "failed to bring vdev up %d: %d\n",
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 6e73194a849e..4d5b02caee13 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -1123,6 +1123,12 @@ int ath12k_wmi_vdev_up(struct ath12k *ar, struct ath12k_wmi_vdev_up_params *para
 
 	ether_addr_copy(cmd->vdev_bssid.addr, params->bssid);
 
+	if (params->tx_bssid) {
+		ether_addr_copy(cmd->tx_vdev_bssid.addr, params->tx_bssid);
+		cmd->nontx_profile_idx = cpu_to_le32(params->nontx_profile_idx);
+		cmd->nontx_profile_cnt = cpu_to_le32(params->nontx_profile_cnt);
+	}
+
 	ath12k_dbg(ar->ab, ATH12K_DBG_WMI,
 		   "WMI mgmt vdev up id 0x%x assoc id %d bssid %pM\n",
 		   params->vdev_id, params->aid, params->bssid);
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 12c6cfb943c8..9b0166a5a575 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -2767,6 +2767,9 @@ struct ath12k_wmi_vdev_up_params {
 	u32 vdev_id;
 	u32 aid;
 	const u8 *bssid;
+	const u8 *tx_bssid;
+	u32 nontx_profile_idx;
+	u32 nontx_profile_cnt;
 };
 
 struct wmi_vdev_up_cmd {
-- 
cgit v1.2.3-58-ga151


From 9fd28936b641ab05ab244a85991e95c5be31e1a2 Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Wed, 8 May 2024 13:29:09 -0700
Subject: wifi: ath12k: refactor arvif security parameter configuration

Configure security parameters in a separate function to make it
easier to support setting the same for nontransmitted profiles.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508202912.11902-7-quic_alokad@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 33 +++++++++++++++++++++------------
 1 file changed, 21 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 3f57acc16c40..6720ec0d7fdd 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -1299,6 +1299,26 @@ static int ath12k_mac_remove_vendor_ie(struct sk_buff *skb, unsigned int oui,
 	return 0;
 }
 
+static void ath12k_mac_set_arvif_ies(struct ath12k_vif *arvif, struct sk_buff *bcn)
+{
+	struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *)bcn->data;
+	const u8 *start, *tail;
+	u16 rem_len;
+
+	start = bcn->data + ieee80211_get_hdrlen_from_skb(bcn) + sizeof(mgmt->u.beacon);
+	tail = skb_tail_pointer(bcn);
+	rem_len = tail - start;
+
+	arvif->rsnie_present = false;
+	arvif->wpaie_present = false;
+
+	if (cfg80211_find_ie(WLAN_EID_RSN, start, rem_len))
+		arvif->rsnie_present = true;
+	if (cfg80211_find_vendor_ie(WLAN_OUI_MICROSOFT, WLAN_OUI_TYPE_MICROSOFT_WPA,
+				    start, rem_len))
+		arvif->wpaie_present = true;
+}
+
 static int ath12k_mac_setup_bcn_tmpl(struct ath12k_vif *arvif)
 {
 	struct ath12k *ar = arvif->ar;
@@ -1307,8 +1327,6 @@ static int ath12k_mac_setup_bcn_tmpl(struct ath12k_vif *arvif)
 	struct ieee80211_vif *vif = arvif->vif;
 	struct ieee80211_mutable_offsets offs = {};
 	struct sk_buff *bcn;
-	struct ieee80211_mgmt *mgmt;
-	u8 *ies;
 	int ret;
 
 	if (arvif->vdev_type != WMI_VDEV_TYPE_AP)
@@ -1320,16 +1338,7 @@ static int ath12k_mac_setup_bcn_tmpl(struct ath12k_vif *arvif)
 		return -EPERM;
 	}
 
-	ies = bcn->data + ieee80211_get_hdrlen_from_skb(bcn);
-	ies += sizeof(mgmt->u.beacon);
-
-	if (cfg80211_find_ie(WLAN_EID_RSN, ies, (skb_tail_pointer(bcn) - ies)))
-		arvif->rsnie_present = true;
-
-	if (cfg80211_find_vendor_ie(WLAN_OUI_MICROSOFT,
-				    WLAN_OUI_TYPE_MICROSOFT_WPA,
-				    ies, (skb_tail_pointer(bcn) - ies)))
-		arvif->wpaie_present = true;
+	ath12k_mac_set_arvif_ies(arvif, bcn);
 
 	if (arvif->vif->type == NL80211_IFTYPE_AP && arvif->vif->p2p) {
 		ret = ath12k_mac_setup_bcn_p2p_ie(arvif, bcn);
-- 
cgit v1.2.3-58-ga151


From 8c6faa56bfb245a165820f46b8ea90e6ae1f24f5 Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Wed, 8 May 2024 13:29:10 -0700
Subject: wifi: ath12k: add MBSSID beacon support

- A single beacon template is used for one multiple BSSID group as
only one interface from group transmits the beacons. But firmware
requires the host to send this template each time a nontransmitted
interface is brought up. Hence, modify existing logic to retrieve
the template from vif context of the transmitting interface. Also,
firmware crashes if host sends a template for a nontransmitted
interface which is already up, add a check to avoid this.
- Set rsnie_present and wpaie_present fields for nontransmitted VAPs.
- Send offset to the multiple BSSID element in the beacon template
firmware.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508202912.11902-8-quic_alokad@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 97 +++++++++++++++++++++++++++++++++--
 drivers/net/wireless/ath/ath12k/wmi.c |  1 +
 2 files changed, 94 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 6720ec0d7fdd..2cfed85990a9 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -1299,11 +1299,14 @@ static int ath12k_mac_remove_vendor_ie(struct sk_buff *skb, unsigned int oui,
 	return 0;
 }
 
-static void ath12k_mac_set_arvif_ies(struct ath12k_vif *arvif, struct sk_buff *bcn)
+static void ath12k_mac_set_arvif_ies(struct ath12k_vif *arvif, struct sk_buff *bcn,
+				     u8 bssid_index, bool *nontx_profile_found)
 {
 	struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *)bcn->data;
+	const struct element *elem, *nontx, *index, *nie;
 	const u8 *start, *tail;
 	u16 rem_len;
+	u8 i;
 
 	start = bcn->data + ieee80211_get_hdrlen_from_skb(bcn) + sizeof(mgmt->u.beacon);
 	tail = skb_tail_pointer(bcn);
@@ -1317,28 +1320,114 @@ static void ath12k_mac_set_arvif_ies(struct ath12k_vif *arvif, struct sk_buff *b
 	if (cfg80211_find_vendor_ie(WLAN_OUI_MICROSOFT, WLAN_OUI_TYPE_MICROSOFT_WPA,
 				    start, rem_len))
 		arvif->wpaie_present = true;
+
+	/* Return from here for the transmitted profile */
+	if (!bssid_index)
+		return;
+
+	/* Initial rsnie_present for the nontransmitted profile is set to be same as that
+	 * of the transmitted profile. It will be changed if security configurations are
+	 * different.
+	 */
+	*nontx_profile_found = false;
+	for_each_element_id(elem, WLAN_EID_MULTIPLE_BSSID, start, rem_len) {
+		/* Fixed minimum MBSSID element length with at least one
+		 * nontransmitted BSSID profile is 12 bytes as given below;
+		 * 1 (max BSSID indicator) +
+		 * 2 (Nontransmitted BSSID profile: Subelement ID + length) +
+		 * 4 (Nontransmitted BSSID Capabilities: tag + length + info)
+		 * 2 (Nontransmitted BSSID SSID: tag + length)
+		 * 3 (Nontransmitted BSSID Index: tag + length + BSSID index
+		 */
+		if (elem->datalen < 12 || elem->data[0] < 1)
+			continue; /* Max BSSID indicator must be >=1 */
+
+		for_each_element(nontx, elem->data + 1, elem->datalen - 1) {
+			start = nontx->data;
+
+			if (nontx->id != 0 || nontx->datalen < 4)
+				continue; /* Invalid nontransmitted profile */
+
+			if (nontx->data[0] != WLAN_EID_NON_TX_BSSID_CAP ||
+			    nontx->data[1] != 2) {
+				continue; /* Missing nontransmitted BSS capabilities */
+			}
+
+			if (nontx->data[4] != WLAN_EID_SSID)
+				continue; /* Missing SSID for nontransmitted BSS */
+
+			index = cfg80211_find_elem(WLAN_EID_MULTI_BSSID_IDX,
+						   start, nontx->datalen);
+			if (!index || index->datalen < 1 || index->data[0] == 0)
+				continue; /* Invalid MBSSID Index element */
+
+			if (index->data[0] == bssid_index) {
+				*nontx_profile_found = true;
+				if (cfg80211_find_ie(WLAN_EID_RSN,
+						     nontx->data,
+						     nontx->datalen)) {
+					arvif->rsnie_present = true;
+					return;
+				} else if (!arvif->rsnie_present) {
+					return; /* Both tx and nontx BSS are open */
+				}
+
+				nie = cfg80211_find_ext_elem(WLAN_EID_EXT_NON_INHERITANCE,
+							     nontx->data,
+							     nontx->datalen);
+				if (!nie || nie->datalen < 2)
+					return; /* Invalid non-inheritance element */
+
+				for (i = 1; i < nie->datalen - 1; i++) {
+					if (nie->data[i] == WLAN_EID_RSN) {
+						arvif->rsnie_present = false;
+						break;
+					}
+				}
+
+				return;
+			}
+		}
+	}
 }
 
 static int ath12k_mac_setup_bcn_tmpl(struct ath12k_vif *arvif)
 {
+	struct ath12k_vif *tx_arvif = arvif;
 	struct ath12k *ar = arvif->ar;
 	struct ath12k_base *ab = ar->ab;
-	struct ieee80211_hw *hw = ath12k_ar_to_hw(ar);
 	struct ieee80211_vif *vif = arvif->vif;
 	struct ieee80211_mutable_offsets offs = {};
+	bool nontx_profile_found = false;
 	struct sk_buff *bcn;
 	int ret;
 
 	if (arvif->vdev_type != WMI_VDEV_TYPE_AP)
 		return 0;
 
-	bcn = ieee80211_beacon_get_template(hw, vif, &offs, 0);
+	if (vif->mbssid_tx_vif) {
+		tx_arvif = ath12k_vif_to_arvif(vif->mbssid_tx_vif);
+		if (tx_arvif != arvif && arvif->is_up)
+			return 0;
+	}
+
+	bcn = ieee80211_beacon_get_template(ath12k_ar_to_hw(tx_arvif->ar), tx_arvif->vif,
+					    &offs, 0);
 	if (!bcn) {
 		ath12k_warn(ab, "failed to get beacon template from mac80211\n");
 		return -EPERM;
 	}
 
-	ath12k_mac_set_arvif_ies(arvif, bcn);
+	if (tx_arvif == arvif) {
+		ath12k_mac_set_arvif_ies(arvif, bcn, 0, NULL);
+	} else {
+		ath12k_mac_set_arvif_ies(arvif, bcn,
+					 arvif->vif->bss_conf.bssid_index,
+					 &nontx_profile_found);
+		if (!nontx_profile_found)
+			ath12k_warn(ab,
+				    "nontransmitted profile not found in beacon template\n");
+	}
 
 	if (arvif->vif->type == NL80211_IFTYPE_AP && arvif->vif->p2p) {
 		ret = ath12k_mac_setup_bcn_p2p_ie(arvif, bcn);
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 4d5b02caee13..57fb01d66a0d 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -1813,6 +1813,7 @@ int ath12k_wmi_bcn_tmpl(struct ath12k *ar, u32 vdev_id,
 	cmd->csa_switch_count_offset = cpu_to_le32(offs->cntdwn_counter_offs[0]);
 	cmd->ext_csa_switch_count_offset = cpu_to_le32(offs->cntdwn_counter_offs[1]);
 	cmd->buf_len = cpu_to_le32(bcn->len);
+	cmd->mbssid_ie_offset = cpu_to_le32(offs->mbssid_off);
 
 	ptr = skb->data + sizeof(*cmd);
 
-- 
cgit v1.2.3-58-ga151


From f27fbd76a98feca81688175985fbfa31bad8cd75 Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Wed, 8 May 2024 13:29:11 -0700
Subject: wifi: ath12k: add EMA beacon support

Add new function ath12k_mac_setup_bcn_tmpl_ema() which retrieves
a list of EMA (Enhanced Multi-BSSID Advertisements) beacon templates
from mac80211 and sends all to firmware. Add support to send EMA
configurations to firmware by adding a new parameter 'ema_params' in
struct wmi_bcn_tmpl_cmd.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508202912.11902-9-quic_alokad@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 55 ++++++++++++++++++++++++++++++++++-
 drivers/net/wireless/ath/ath12k/wmi.c | 13 ++++++++-
 drivers/net/wireless/ath/ath12k/wmi.h | 18 +++++++++++-
 3 files changed, 83 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 2cfed85990a9..e86af398b359 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -1391,6 +1391,56 @@ static void ath12k_mac_set_arvif_ies(struct ath12k_vif *arvif, struct sk_buff *b
 	}
 }
 
+static int ath12k_mac_setup_bcn_tmpl_ema(struct ath12k_vif *arvif)
+{
+	struct ieee80211_bss_conf *bss_conf = &arvif->vif->bss_conf;
+	struct ath12k_wmi_bcn_tmpl_ema_arg ema_args;
+	struct ieee80211_ema_beacons *beacons;
+	struct ath12k_vif *tx_arvif;
+	bool nontx_profile_found = false;
+	int ret = 0;
+	u8 i;
+
+	tx_arvif = ath12k_vif_to_arvif(arvif->vif->mbssid_tx_vif);
+	beacons = ieee80211_beacon_get_template_ema_list(ath12k_ar_to_hw(tx_arvif->ar),
+							 tx_arvif->vif, 0);
+	if (!beacons || !beacons->cnt) {
+		ath12k_warn(arvif->ar->ab,
+			    "failed to get ema beacon templates from mac80211\n");
+		return -EPERM;
+	}
+
+	if (tx_arvif == arvif)
+		ath12k_mac_set_arvif_ies(arvif, beacons->bcn[0].skb, 0, NULL);
+
+	for (i = 0; i < beacons->cnt; i++) {
+		if (tx_arvif != arvif && !nontx_profile_found)
+			ath12k_mac_set_arvif_ies(arvif, beacons->bcn[i].skb,
+						 bss_conf->bssid_index,
+						 &nontx_profile_found);
+
+		ema_args.bcn_cnt = beacons->cnt;
+		ema_args.bcn_index = i;
+		ret = ath12k_wmi_bcn_tmpl(tx_arvif->ar, tx_arvif->vdev_id,
+					  &beacons->bcn[i].offs,
+					  beacons->bcn[i].skb, &ema_args);
+		if (ret) {
+			ath12k_warn(tx_arvif->ar->ab,
+				    "failed to set ema beacon template id %i error %d\n",
+				    i, ret);
+			break;
+		}
+	}
+
+	if (tx_arvif != arvif && !nontx_profile_found)
+		ath12k_warn(arvif->ar->ab,
+			    "nontransmitted bssid index %u not found in beacon template\n",
+			    bss_conf->bssid_index);
+
+	ieee80211_beacon_free_ema_list(beacons);
+	return ret;
+}
+
 static int ath12k_mac_setup_bcn_tmpl(struct ath12k_vif *arvif)
 {
 	struct ath12k_vif *tx_arvif = arvif;
@@ -1409,6 +1459,9 @@ static int ath12k_mac_setup_bcn_tmpl(struct ath12k_vif *arvif)
 		tx_arvif = ath12k_vif_to_arvif(vif->mbssid_tx_vif);
 		if (tx_arvif != arvif && arvif->is_up)
 			return 0;
+
+		if (vif->bss_conf.ema_ap)
+			return ath12k_mac_setup_bcn_tmpl_ema(arvif);
 	}
 
 	bcn = ieee80211_beacon_get_template(ath12k_ar_to_hw(tx_arvif->ar), tx_arvif->vif,
@@ -1452,7 +1505,7 @@ static int ath12k_mac_setup_bcn_tmpl(struct ath12k_vif *arvif)
 		}
 	}
 
-	ret = ath12k_wmi_bcn_tmpl(ar, arvif->vdev_id, &offs, bcn);
+	ret = ath12k_wmi_bcn_tmpl(ar, arvif->vdev_id, &offs, bcn, NULL);
 
 	if (ret)
 		ath12k_warn(ab, "failed to submit beacon template command: %d\n",
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 57fb01d66a0d..99106b088311 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -1788,13 +1788,15 @@ int ath12k_wmi_p2p_go_bcn_ie(struct ath12k *ar, u32 vdev_id,
 
 int ath12k_wmi_bcn_tmpl(struct ath12k *ar, u32 vdev_id,
 			struct ieee80211_mutable_offsets *offs,
-			struct sk_buff *bcn)
+			struct sk_buff *bcn,
+			struct ath12k_wmi_bcn_tmpl_ema_arg *ema_args)
 {
 	struct ath12k_wmi_pdev *wmi = ar->wmi;
 	struct wmi_bcn_tmpl_cmd *cmd;
 	struct ath12k_wmi_bcn_prb_info_params *bcn_prb_info;
 	struct wmi_tlv *tlv;
 	struct sk_buff *skb;
+	u32 ema_params = 0;
 	void *ptr;
 	int ret, len;
 	size_t aligned_len = roundup(bcn->len, 4);
@@ -1814,6 +1816,15 @@ int ath12k_wmi_bcn_tmpl(struct ath12k *ar, u32 vdev_id,
 	cmd->ext_csa_switch_count_offset = cpu_to_le32(offs->cntdwn_counter_offs[1]);
 	cmd->buf_len = cpu_to_le32(bcn->len);
 	cmd->mbssid_ie_offset = cpu_to_le32(offs->mbssid_off);
+	if (ema_args) {
+		u32p_replace_bits(&ema_params, ema_args->bcn_cnt, WMI_EMA_BEACON_CNT);
+		u32p_replace_bits(&ema_params, ema_args->bcn_index, WMI_EMA_BEACON_IDX);
+		if (ema_args->bcn_index == 0)
+			u32p_replace_bits(&ema_params, 1, WMI_EMA_BEACON_FIRST);
+		if (ema_args->bcn_index + 1 == ema_args->bcn_cnt)
+			u32p_replace_bits(&ema_params, 1, WMI_EMA_BEACON_LAST);
+		cmd->ema_params = cpu_to_le32(ema_params);
+	}
 
 	ptr = skb->data + sizeof(*cmd);
 
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 9b0166a5a575..7f83c4583919 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -3533,6 +3533,16 @@ struct ath12k_wmi_p2p_noa_info {
 
 #define WMI_BEACON_TX_BUFFER_SIZE	512
 
+#define WMI_EMA_BEACON_CNT      GENMASK(7, 0)
+#define WMI_EMA_BEACON_IDX      GENMASK(15, 8)
+#define WMI_EMA_BEACON_FIRST    GENMASK(23, 16)
+#define WMI_EMA_BEACON_LAST     GENMASK(31, 24)
+
+struct ath12k_wmi_bcn_tmpl_ema_arg {
+	u8 bcn_cnt;
+	u8 bcn_index;
+};
+
 struct wmi_bcn_tmpl_cmd {
 	__le32 tlv_header;
 	__le32 vdev_id;
@@ -3543,6 +3553,11 @@ struct wmi_bcn_tmpl_cmd {
 	__le32 csa_event_bitmap;
 	__le32 mbssid_ie_offset;
 	__le32 esp_ie_offset;
+	__le32 csc_switch_count_offset;
+	__le32 csc_event_bitmap;
+	__le32 mu_edca_ie_offset;
+	__le32 feature_enable_bitmap;
+	__le32 ema_params;
 } __packed;
 
 struct wmi_p2p_go_set_beacon_ie_cmd {
@@ -4900,7 +4915,8 @@ int ath12k_wmi_p2p_go_bcn_ie(struct ath12k *ar, u32 vdev_id,
 			     const u8 *p2p_ie);
 int ath12k_wmi_bcn_tmpl(struct ath12k *ar, u32 vdev_id,
 			struct ieee80211_mutable_offsets *offs,
-			struct sk_buff *bcn);
+			struct sk_buff *bcn,
+			struct ath12k_wmi_bcn_tmpl_ema_arg *ema_args);
 int ath12k_wmi_vdev_down(struct ath12k *ar, u8 vdev_id);
 int ath12k_wmi_vdev_up(struct ath12k *ar, struct ath12k_wmi_vdev_up_params *params);
 int ath12k_wmi_vdev_stop(struct ath12k *ar, u8 vdev_id);
-- 
cgit v1.2.3-58-ga151


From c740a18f62f5bbf2c88c3160d189704cbe001255 Mon Sep 17 00:00:00 2001
From: Aloka Dixit <quic_alokad@quicinc.com>
Date: Wed, 8 May 2024 13:29:12 -0700
Subject: wifi: ath12k: skip sending vdev down for channel switch

Remove the unnecessary WMI vdev down operation as firmware
internally does so already as part of WMI vdev restart.
Repeated vdev down results in a firmware crash during channel
switch operation when multiple BSSID advertisements are enabled:
ath12k_pci 0000:06:00.0: firmware crashed: MHI_CB_SYS_ERROR
ath12k_pci 0000:06:00.0: failed to send WMI_VDEV_SET_PARAM_CMDID
ath12k_pci 0000:06:00.0: failed to set vdev 5 HE MU mode: -108 param_value 7b
ath12k_pci 0000:06:00.0: failed to set he mode vdev 5
ath12k_pci 0000:06:00.0: failed to restart vdev 5: -108
ath12k_pci 0000:06:00.0: failed to send WMI_VDEV_SET_PARAM_CMDID
ath12k_pci 0000:06:00.0: failed to set vdev 4 HE MU mode: -108 param_value 7b
ath12k_pci 0000:06:00.0: failed to set he mode vdev 4
ath12k_pci 0000:06:00.0: failed to restart vdev 4: -108
ath12k_pci 0000:06:00.0: failed to send WMI_VDEV_SET_PARAM_CMDID

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aloka Dixit <quic_alokad@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240508202912.11902-10-quic_alokad@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 29 +++--------------------------
 1 file changed, 3 insertions(+), 26 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index e86af398b359..33741ebecdf6 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -7364,9 +7364,10 @@ ath12k_mac_update_vif_chan(struct ath12k *ar,
 	lockdep_assert_held(&ar->conf_mutex);
 
 	for (i = 0; i < n_vifs; i++) {
-		arvif = ath12k_vif_to_arvif(vifs[i].vif);
+		vif = vifs[i].vif;
+		arvif = ath12k_vif_to_arvif(vif);
 
-		if (vifs[i].vif->type == NL80211_IFTYPE_MONITOR)
+		if (vif->type == NL80211_IFTYPE_MONITOR)
 			monitor_vif = true;
 
 		ath12k_dbg(ab, ATH12K_DBG_MAC,
@@ -7377,30 +7378,6 @@ ath12k_mac_update_vif_chan(struct ath12k *ar,
 			   vifs[i].old_ctx->def.width,
 			   vifs[i].new_ctx->def.width);
 
-		if (WARN_ON(!arvif->is_started))
-			continue;
-
-		if (WARN_ON(!arvif->is_up))
-			continue;
-
-		ret = ath12k_wmi_vdev_down(ar, arvif->vdev_id);
-		if (ret) {
-			ath12k_warn(ab, "failed to down vdev %d: %d\n",
-				    arvif->vdev_id, ret);
-			continue;
-		}
-	}
-
-	/* All relevant vdevs are downed and associated channel resources
-	 * should be available for the channel switch now.
-	 */
-
-	/* TODO: Update ar->rx_channel */
-
-	for (i = 0; i < n_vifs; i++) {
-		vif = vifs[i].vif;
-		arvif = ath12k_vif_to_arvif(vif);
-
 		if (WARN_ON(!arvif->is_started))
 			continue;
 
-- 
cgit v1.2.3-58-ga151


From 0ce9ab2785e4e9ca0752390d8e5ab65bd08f0e78 Mon Sep 17 00:00:00 2001
From: Nithyanantham Paramasivam <quic_nithp@quicinc.com>
Date: Fri, 10 May 2024 12:34:27 +0530
Subject: wifi: ath12k: Fix tx completion ring (WBM2SW) setup failure

We observe intermittent ping failures from the access point (AP) to
station (STA) in any mode (AP-STA or Mesh) configured. Specifically,
the transmission completion status is not received at tx completion
ring id 4 (WBM2SW ring4) for the packets transmitted via TCL DATA
ring id 3. This prevents freeing up tx descriptors and leads
to buffer exhaustion.

Currently, during initialization of the WBM2SW ring, we are directly
mapping the ring number to the ring mask to obtain the ring mask
group index. This approach is causing setup failures for WBM2SW
ring 4. Similarly, during runtime, when receiving incoming
transmission completion status, the validation of the ring number by
mapping the interrupted ring mask. This is resulting in
validation failure. Thereby preventing entry into the completion
handler ath12k_dp_tx_completion_handler().

The existing design assumed that the ring numbers would always be
sequential and could be directly mapped with the ring mask. However,
this assumption does not hold true for WBM2SW ring 4. Therefore,
modify the design such that, instead of mapping the ring number,
the ring ID is mapped with the ring mask.

According to this design:

1. During initialization of the WBM2SW ring, mapping the ring ID
to the ring mask will ensure obtaining the correct ring mask group
ID.

2. During runtime, validating the interrupted ring mask group ID
within the transmission completion group is sufficient. This
approach allows the ring ID to be derived from the interrupted ring
mask and enables entry into the completion handler.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Signed-off-by: Nithyanantham Paramasivam <quic_nithp@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240510070427.206152-1-quic_nithp@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp.c | 18 +++++++++++++-----
 drivers/net/wireless/ath/ath12k/hw.c |  2 +-
 2 files changed, 14 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp.c b/drivers/net/wireless/ath/ath12k/dp.c
index 412ab08e6cca..c04fa3eb57d6 100644
--- a/drivers/net/wireless/ath/ath12k/dp.c
+++ b/drivers/net/wireless/ath/ath12k/dp.c
@@ -132,7 +132,9 @@ static int ath12k_dp_srng_find_ring_in_mask(int ring_num, const u8 *grp_mask)
 static int ath12k_dp_srng_calculate_msi_group(struct ath12k_base *ab,
 					      enum hal_ring_type type, int ring_num)
 {
+	const struct ath12k_hal_tcl_to_wbm_rbm_map *map;
 	const u8 *grp_mask;
+	int i;
 
 	switch (type) {
 	case HAL_WBM2SW_RELEASE:
@@ -140,6 +142,14 @@ static int ath12k_dp_srng_calculate_msi_group(struct ath12k_base *ab,
 			grp_mask = &ab->hw_params->ring_mask->rx_wbm_rel[0];
 			ring_num = 0;
 		} else {
+			map = ab->hw_params->hal_ops->tcl_to_wbm_rbm_map;
+			for (i = 0; i < ab->hw_params->max_tx_ring; i++) {
+				if (ring_num == map[i].wbm_ring_num) {
+					ring_num = i;
+					break;
+				}
+			}
+
 			grp_mask = &ab->hw_params->ring_mask->tx[0];
 		}
 		break;
@@ -867,11 +877,9 @@ int ath12k_dp_service_srng(struct ath12k_base *ab,
 	enum dp_monitor_mode monitor_mode;
 	u8 ring_mask;
 
-	while (i < ab->hw_params->max_tx_ring) {
-		if (ab->hw_params->ring_mask->tx[grp_id] &
-			BIT(ab->hw_params->hal_ops->tcl_to_wbm_rbm_map[i].wbm_ring_num))
-			ath12k_dp_tx_completion_handler(ab, i);
-		i++;
+	if (ab->hw_params->ring_mask->tx[grp_id]) {
+		i = fls(ab->hw_params->ring_mask->tx[grp_id]) - 1;
+		ath12k_dp_tx_completion_handler(ab, i);
 	}
 
 	if (ab->hw_params->ring_mask->rx_err[grp_id]) {
diff --git a/drivers/net/wireless/ath/ath12k/hw.c b/drivers/net/wireless/ath/ath12k/hw.c
index b3ba753d4a85..5a1d406e6b1f 100644
--- a/drivers/net/wireless/ath/ath12k/hw.c
+++ b/drivers/net/wireless/ath/ath12k/hw.c
@@ -580,8 +580,8 @@ static const struct ath12k_hw_ring_mask ath12k_hw_ring_mask_qcn9274 = {
 static const struct ath12k_hw_ring_mask ath12k_hw_ring_mask_wcn7850 = {
 	.tx  = {
 		ATH12K_TX_RING_MASK_0,
+		ATH12K_TX_RING_MASK_1,
 		ATH12K_TX_RING_MASK_2,
-		ATH12K_TX_RING_MASK_4,
 	},
 	.rx_mon_dest = {
 	},
-- 
cgit v1.2.3-58-ga151


From 3a29442a17413e9e64f00db3b98dbf911324cb86 Mon Sep 17 00:00:00 2001
From: Kang Yang <quic_kangyang@quicinc.com>
Date: Fri, 10 May 2024 19:12:38 +0800
Subject: wifi: ath12k: remove redundant peer delete for WCN7850

ath12k_mac_op_unassign_vif_chanctx() will do peer delete for WCN7850's monitor
interface. But for host no need to do peer create/delete for monitor interface
because firmware doesn't require it, QCN9274 is the same. Also, the first
judgement is true only for WCN7850, so this logic won't affect QCN9274 at any
time.

So remove this peer delete.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Kang Yang <quic_kangyang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240510111238.1113-1-quic_kangyang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 33741ebecdf6..784964ae03ec 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -7651,11 +7651,6 @@ ath12k_mac_op_unassign_vif_chanctx(struct ieee80211_hw *hw,
 
 	WARN_ON(!arvif->is_started);
 
-	if (ab->hw_params->vdev_start_delay &&
-	    arvif->vdev_type == WMI_VDEV_TYPE_MONITOR &&
-	    ath12k_peer_find_by_addr(ab, ar->mac_addr))
-		ath12k_peer_delete(ar, arvif->vdev_id, ar->mac_addr);
-
 	if (arvif->vdev_type == WMI_VDEV_TYPE_MONITOR) {
 		ret = ath12k_mac_monitor_stop(ar);
 		if (ret) {
-- 
cgit v1.2.3-58-ga151


From 28bfc1c9c9d0e2165ee9e3b50793b8b463e52e01 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Tue, 14 May 2024 12:04:11 +0530
Subject: wifi: ath12k: remove duplicate definition of MAX_RADIOS

Remove the duplicate definition of MAX_RADIOS from the core header file.

No functional changes, compile tested only.

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240514063412.3483723-2-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index 6da242c0d51d..7d20b09c52e6 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -180,8 +180,6 @@ struct ath12k_he {
 	u32 heop_param;
 };
 
-#define MAX_RADIOS 3
-
 enum {
 	WMI_HOST_TP_SCALE_MAX   = 0,
 	WMI_HOST_TP_SCALE_50    = 1,
-- 
cgit v1.2.3-58-ga151


From 7678f8bfd74c324bd10499cf1d05e5720d9449b2 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Tue, 14 May 2024 12:04:12 +0530
Subject: wifi: ath12k: use correct MAX_RADIOS

The current value of the MAX_RADIOS definition is 3. This is incorrect
because no device supports more than 2 radios.

An incorrect MAX_RADIOS value can lead to:
 - unnecessary memory allocation for pdev related entities.
 - invalid pdev id validation.
 - Misconceptions for code readers.

Therefore, modify the MAX_RADIOS definition as 2.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240514063412.3483723-3-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/wmi.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 7f83c4583919..c2b86e187a03 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -4804,7 +4804,7 @@ struct wmi_probe_tmpl_cmd {
 	__le32 buf_len;
 } __packed;
 
-#define MAX_RADIOS 3
+#define MAX_RADIOS 2
 
 #define WMI_SERVICE_READY_TIMEOUT_HZ (5 * HZ)
 #define WMI_SEND_TIMEOUT_HZ (3 * HZ)
-- 
cgit v1.2.3-58-ga151


From ed61863debb2df0856524b5c61d2a7f7731c2e83 Mon Sep 17 00:00:00 2001
From: Lingbo Kong <quic_lingbok@quicinc.com>
Date: Thu, 16 May 2024 11:30:44 +0300
Subject: wifi: ath12k: fix ack signal strength calculation

Currently, the calculation of ack signal strength is incorrect.

This is because before calculating the ack signal strength, ath12k need
to determine whether the hardware and firmware support db2dbm. If the
hardware and firmware support db2dbm, do not need to add noise floor,
otherwise, need to add noise floor.

Besides, the value of ack_rssi passed by firmware to ath12k should be a
signed number, so change its type to s8.

This patch affects QCN9274 and WCN7850 because they have the same issue.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Lingbo Kong <quic_lingbok@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240511030917.65811-1-quic_lingbok@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_tx.c  | 16 ++++++++++++----
 drivers/net/wireless/ath/ath12k/dp_tx.h  |  2 +-
 drivers/net/wireless/ath/ath12k/hal_tx.h |  4 ++--
 3 files changed, 15 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.c b/drivers/net/wireless/ath/ath12k/dp_tx.c
index 32496015cf9a..f57bc8d69715 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.c
@@ -393,8 +393,12 @@ ath12k_dp_tx_htt_tx_complete_buf(struct ath12k_base *ab,
 	if (ts->acked) {
 		if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 			info->flags |= IEEE80211_TX_STAT_ACK;
-			info->status.ack_signal = ATH12K_DEFAULT_NOISE_FLOOR +
-						  ts->ack_rssi;
+			info->status.ack_signal = ts->ack_rssi;
+
+			if (!test_bit(WMI_TLV_SERVICE_HW_DB2DBM_CONVERSION_SUPPORT,
+				      ab->wmi_ab.svc_map))
+				info->status.ack_signal += ATH12K_DEFAULT_NOISE_FLOOR;
+
 			info->status.flags = IEEE80211_TX_STATUS_ACK_SIGNAL_VALID;
 		} else {
 			info->flags |= IEEE80211_TX_STAT_NOACK_TRANSMITTED;
@@ -486,8 +490,12 @@ static void ath12k_dp_tx_complete_msdu(struct ath12k *ar,
 	case HAL_WBM_TQM_REL_REASON_FRAME_ACKED:
 		if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 			info->flags |= IEEE80211_TX_STAT_ACK;
-			info->status.ack_signal = ATH12K_DEFAULT_NOISE_FLOOR +
-						  ts->ack_rssi;
+			info->status.ack_signal = ts->ack_rssi;
+
+			if (!test_bit(WMI_TLV_SERVICE_HW_DB2DBM_CONVERSION_SUPPORT,
+				      ab->wmi_ab.svc_map))
+				info->status.ack_signal += ATH12K_DEFAULT_NOISE_FLOOR;
+
 			info->status.flags = IEEE80211_TX_STATUS_ACK_SIGNAL_VALID;
 		}
 		break;
diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.h b/drivers/net/wireless/ath/ath12k/dp_tx.h
index 7faa81faf6a7..55ff8cc721e3 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.h
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.h
@@ -12,7 +12,7 @@
 
 struct ath12k_dp_htt_wbm_tx_status {
 	bool acked;
-	int ack_rssi;
+	s8 ack_rssi;
 };
 
 int ath12k_dp_tx_htt_h2t_ver_req_msg(struct ath12k_base *ab);
diff --git a/drivers/net/wireless/ath/ath12k/hal_tx.h b/drivers/net/wireless/ath/ath12k/hal_tx.h
index 7c837094a6f7..3cf5973771d7 100644
--- a/drivers/net/wireless/ath/ath12k/hal_tx.h
+++ b/drivers/net/wireless/ath/ath12k/hal_tx.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause-Clear */
 /*
  * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved.
- * Copyright (c) 2021-2022 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2021-2022, 2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #ifndef ATH12K_HAL_TX_H
@@ -57,7 +57,7 @@ struct hal_tx_info {
 struct hal_tx_status {
 	enum hal_wbm_rel_src_module buf_rel_source;
 	enum hal_wbm_tqm_rel_reason status;
-	u8 ack_rssi;
+	s8 ack_rssi;
 	u32 flags; /* %HAL_TX_STATUS_FLAGS_ */
 	u32 ppdu_id;
 	u8 try_cnt;
-- 
cgit v1.2.3-58-ga151


From 6fe46d5c0a84471e06fa5b22d7f4253ebf98b071 Mon Sep 17 00:00:00 2001
From: Alexis Lothoré <alexis.lothore@bootlin.com>
Date: Tue, 14 May 2024 17:51:19 +0200
Subject: wifi: wilc1000: set net device registration as last step during
 interface creation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

net device registration is currently done in wilc_netdev_ifc_init but
other initialization operations are still done after this registration.
Since net device is assumed to be usable right after registration, it
should be the very last step of initialization.

Move netdev registration at the very end of wilc_netdev_ifc_init to let
this function completely initialize netdevice before registering it.

Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240514-mac_addr_at_probe-v2-1-afef09f1cd10@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/netdev.c | 31 ++++++++++++------------
 1 file changed, 16 insertions(+), 15 deletions(-)

diff --git a/drivers/net/wireless/microchip/wilc1000/netdev.c b/drivers/net/wireless/microchip/wilc1000/netdev.c
index 73f56f7b002b..acc9b9a64552 100644
--- a/drivers/net/wireless/microchip/wilc1000/netdev.c
+++ b/drivers/net/wireless/microchip/wilc1000/netdev.c
@@ -965,16 +965,6 @@ struct wilc_vif *wilc_netdev_ifc_init(struct wilc *wl, const char *name,
 	vif->priv.wdev.iftype = type;
 	vif->priv.dev = ndev;
 
-	if (rtnl_locked)
-		ret = cfg80211_register_netdevice(ndev);
-	else
-		ret = register_netdev(ndev);
-
-	if (ret) {
-		ret = -EFAULT;
-		goto error;
-	}
-
 	ndev->needs_free_netdev = true;
 	vif->iftype = vif_type;
 	vif->idx = wilc_get_available_idx(wl);
@@ -985,13 +975,24 @@ struct wilc_vif *wilc_netdev_ifc_init(struct wilc *wl, const char *name,
 	mutex_unlock(&wl->vif_mutex);
 	synchronize_rcu();
 
-	return vif;
-
-error:
 	if (rtnl_locked)
-		cfg80211_unregister_netdevice(ndev);
+		ret = cfg80211_register_netdevice(ndev);
 	else
-		unregister_netdev(ndev);
+		ret = register_netdev(ndev);
+
+	if (ret) {
+		ret = -EFAULT;
+		goto error_remove_vif;
+	}
+
+	return vif;
+
+error_remove_vif:
+	mutex_lock(&wl->vif_mutex);
+	list_del_rcu(&vif->list);
+	wl->vif_num -= 1;
+	mutex_unlock(&wl->vif_mutex);
+	synchronize_rcu();
 	free_netdev(ndev);
 	return ERR_PTR(ret);
 }
-- 
cgit v1.2.3-58-ga151


From ec99908906e9e322b3d0a34822be44a4659452de Mon Sep 17 00:00:00 2001
From: Alexis Lothoré <alexis.lothore@bootlin.com>
Date: Tue, 14 May 2024 17:51:20 +0200
Subject: wifi: wilc1000: register net device only after bus being fully
 initialized
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

SDIO/SPI probes functions automatically add a default wlan interface on top
of registered wiphy, through wilc_cfg80211_init which in turn calls
wilc_netdev_ifc_init. However, bus is still not fully initialized when we
register corresponding net device (for example we still miss some private
driver data pointers), which for example makes it impossible to
retrieve MAC address from chip (which is supposed to be set on net device
before its registration) before registering net device. More generally, net
device registration should not be done until driver has fully initialized
everything and is ready to handle any operation  on the net device.

Prevent net device from being registered so early by doing it at the end of
probe functions. Apply this logic to both sdio and spi buses.

Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240514-mac_addr_at_probe-v2-2-afef09f1cd10@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/cfg80211.c | 10 ----------
 drivers/net/wireless/microchip/wilc1000/netdev.c   |  1 +
 drivers/net/wireless/microchip/wilc1000/sdio.c     | 14 ++++++++++++--
 drivers/net/wireless/microchip/wilc1000/spi.c      | 11 +++++++++--
 4 files changed, 22 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/microchip/wilc1000/cfg80211.c b/drivers/net/wireless/microchip/wilc1000/cfg80211.c
index 7d9fb9f2d527..f716981f62ad 100644
--- a/drivers/net/wireless/microchip/wilc1000/cfg80211.c
+++ b/drivers/net/wireless/microchip/wilc1000/cfg80211.c
@@ -1773,7 +1773,6 @@ int wilc_cfg80211_init(struct wilc **wilc, struct device *dev, int io_type,
 		       const struct wilc_hif_func *ops)
 {
 	struct wilc *wl;
-	struct wilc_vif *vif;
 	int ret, i;
 
 	wl = wilc_create_wiphy(dev);
@@ -1802,18 +1801,9 @@ int wilc_cfg80211_init(struct wilc **wilc, struct device *dev, int io_type,
 		ret = -ENOMEM;
 		goto free_cfg;
 	}
-	vif = wilc_netdev_ifc_init(wl, "wlan%d", WILC_STATION_MODE,
-				   NL80211_IFTYPE_STATION, false);
-	if (IS_ERR(vif)) {
-		ret = PTR_ERR(vif);
-		goto free_hq;
-	}
 
 	return 0;
 
-free_hq:
-	destroy_workqueue(wl->hif_workqueue);
-
 free_cfg:
 	wilc_wlan_cfg_deinit(wl);
 
diff --git a/drivers/net/wireless/microchip/wilc1000/netdev.c b/drivers/net/wireless/microchip/wilc1000/netdev.c
index acc9b9a64552..166165ee7712 100644
--- a/drivers/net/wireless/microchip/wilc1000/netdev.c
+++ b/drivers/net/wireless/microchip/wilc1000/netdev.c
@@ -996,6 +996,7 @@ error_remove_vif:
 	free_netdev(ndev);
 	return ERR_PTR(ret);
 }
+EXPORT_SYMBOL_GPL(wilc_netdev_ifc_init);
 
 MODULE_DESCRIPTION("Atmel WILC1000 core wireless driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/wireless/microchip/wilc1000/sdio.c b/drivers/net/wireless/microchip/wilc1000/sdio.c
index 52a770c5e76f..a841dad08410 100644
--- a/drivers/net/wireless/microchip/wilc1000/sdio.c
+++ b/drivers/net/wireless/microchip/wilc1000/sdio.c
@@ -136,9 +136,11 @@ out:
 static int wilc_sdio_probe(struct sdio_func *func,
 			   const struct sdio_device_id *id)
 {
+	struct wilc_sdio *sdio_priv;
+	struct wilc_vif *vif;
 	struct wilc *wilc;
 	int ret;
-	struct wilc_sdio *sdio_priv;
+
 
 	sdio_priv = kzalloc(sizeof(*sdio_priv), GFP_KERNEL);
 	if (!sdio_priv)
@@ -176,9 +178,17 @@ static int wilc_sdio_probe(struct sdio_func *func,
 	}
 	clk_prepare_enable(wilc->rtc_clk);
 
+	vif = wilc_netdev_ifc_init(wilc, "wlan%d", WILC_STATION_MODE,
+				   NL80211_IFTYPE_STATION, false);
+	if (IS_ERR(vif)) {
+		ret = PTR_ERR(vif);
+		goto clk_disable_unprepare;
+	}
+
 	dev_info(&func->dev, "Driver Initializing success\n");
 	return 0;
-
+clk_disable_unprepare:
+	clk_disable_unprepare(wilc->rtc_clk);
 dispose_irq:
 	irq_dispose_mapping(wilc->dev_irq_num);
 	wilc_netdev_cleanup(wilc);
diff --git a/drivers/net/wireless/microchip/wilc1000/spi.c b/drivers/net/wireless/microchip/wilc1000/spi.c
index 61c3572ce321..add0e70a09ea 100644
--- a/drivers/net/wireless/microchip/wilc1000/spi.c
+++ b/drivers/net/wireless/microchip/wilc1000/spi.c
@@ -206,9 +206,10 @@ static void wilc_wlan_power(struct wilc *wilc, bool on)
 
 static int wilc_bus_probe(struct spi_device *spi)
 {
-	int ret;
-	struct wilc *wilc;
 	struct wilc_spi *spi_priv;
+	struct wilc_vif *vif;
+	struct wilc *wilc;
+	int ret;
 
 	spi_priv = kzalloc(sizeof(*spi_priv), GFP_KERNEL);
 	if (!spi_priv)
@@ -250,6 +251,12 @@ static int wilc_bus_probe(struct spi_device *spi)
 		goto power_down;
 
 	wilc_wlan_power(wilc, false);
+	vif = wilc_netdev_ifc_init(wilc, "wlan%d", WILC_STATION_MODE,
+				   NL80211_IFTYPE_STATION, false);
+	if (IS_ERR(vif)) {
+		ret = PTR_ERR(vif);
+		goto power_down;
+	}
 	return 0;
 
 power_down:
-- 
cgit v1.2.3-58-ga151


From 5f1191ed386cf5d21eb82195d252b9ff9549db09 Mon Sep 17 00:00:00 2001
From: Alexis Lothoré <alexis.lothore@bootlin.com>
Date: Tue, 14 May 2024 17:51:21 +0200
Subject: wifi: wilc1000: set wilc_set_mac_address parameter as const
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Any attempt to provide a const mac address to wilc_set_mac_address results
in the following warning:

warning: passing argument 2 of 'wilc_set_mac_address' discards 'const'
qualifier from pointer target type [-Wdiscarded-qualifiers]
[...]
drivers/net/wireless/microchip/wilc1000/hif.h:170:52: note: expected 'u8 *'
{aka 'unsigned char *'} but argument is of type 'const unsigned char *'a
int wilc_set_mac_address(struct wilc_vif *vif, u8 *mac_addr);

Instead of using an explicit cast each time we need provide a MAC address,
set the function parameter as const

Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240514-mac_addr_at_probe-v2-3-afef09f1cd10@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/hif.c    | 4 ++--
 drivers/net/wireless/microchip/wilc1000/hif.h    | 2 +-
 drivers/net/wireless/microchip/wilc1000/netdev.c | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/microchip/wilc1000/hif.c b/drivers/net/wireless/microchip/wilc1000/hif.c
index 919de6ffb821..c6892b7e190f 100644
--- a/drivers/net/wireless/microchip/wilc1000/hif.c
+++ b/drivers/net/wireless/microchip/wilc1000/hif.c
@@ -1293,7 +1293,7 @@ int wilc_get_mac_address(struct wilc_vif *vif, u8 *mac_addr)
 	return result;
 }
 
-int wilc_set_mac_address(struct wilc_vif *vif, u8 *mac_addr)
+int wilc_set_mac_address(struct wilc_vif *vif, const u8 *mac_addr)
 {
 	struct wid wid;
 	int result;
@@ -1301,7 +1301,7 @@ int wilc_set_mac_address(struct wilc_vif *vif, u8 *mac_addr)
 	wid.id = WID_MAC_ADDR;
 	wid.type = WID_STR;
 	wid.size = ETH_ALEN;
-	wid.val = mac_addr;
+	wid.val = (u8 *)mac_addr;
 
 	result = wilc_send_config_pkt(vif, WILC_SET_CFG, &wid, 1);
 	if (result)
diff --git a/drivers/net/wireless/microchip/wilc1000/hif.h b/drivers/net/wireless/microchip/wilc1000/hif.h
index 0d380586b1d9..96eeaf31d237 100644
--- a/drivers/net/wireless/microchip/wilc1000/hif.h
+++ b/drivers/net/wireless/microchip/wilc1000/hif.h
@@ -167,7 +167,7 @@ int wilc_add_rx_gtk(struct wilc_vif *vif, const u8 *rx_gtk, u8 gtk_key_len,
 		    u8 cipher_mode);
 int wilc_set_pmkid_info(struct wilc_vif *vif, struct wilc_pmkid_attr *pmkid);
 int wilc_get_mac_address(struct wilc_vif *vif, u8 *mac_addr);
-int wilc_set_mac_address(struct wilc_vif *vif, u8 *mac_addr);
+int wilc_set_mac_address(struct wilc_vif *vif, const u8 *mac_addr);
 int wilc_set_join_req(struct wilc_vif *vif, u8 *bssid, const u8 *ies,
 		      size_t ies_len);
 int wilc_disconnect(struct wilc_vif *vif);
diff --git a/drivers/net/wireless/microchip/wilc1000/netdev.c b/drivers/net/wireless/microchip/wilc1000/netdev.c
index 166165ee7712..2ee3d9827826 100644
--- a/drivers/net/wireless/microchip/wilc1000/netdev.c
+++ b/drivers/net/wireless/microchip/wilc1000/netdev.c
@@ -678,7 +678,7 @@ static int wilc_set_mac_addr(struct net_device *dev, void *p)
 	}
 	rcu_read_unlock();
 
-	result = wilc_set_mac_address(vif, (u8 *)addr->sa_data);
+	result = wilc_set_mac_address(vif, addr->sa_data);
 	if (result)
 		return result;
 
-- 
cgit v1.2.3-58-ga151


From 59cf9277c1fe6f8a72af83bf4a1d9b3f22124750 Mon Sep 17 00:00:00 2001
From: Alexis Lothoré <alexis.lothore@bootlin.com>
Date: Tue, 14 May 2024 17:51:22 +0200
Subject: wifi: wilc1000: add function to read mac address from eFuse
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

wilc driver currently reads and sets mac address by firmware calls. It
means that we can not access mac address if no interface has been brought
up (so firmware is up and running). Another way to get mac address is to
read it directly from eFUSE.

Add a function helper to read the mac address written in eFuse, without
firmware assistance

Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240514-mac_addr_at_probe-v2-4-afef09f1cd10@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/fw.h     | 13 +++++++
 drivers/net/wireless/microchip/wilc1000/netdev.h |  2 +
 drivers/net/wireless/microchip/wilc1000/wlan.c   | 49 ++++++++++++++++++++++++
 drivers/net/wireless/microchip/wilc1000/wlan.h   |  1 +
 4 files changed, 65 insertions(+)

diff --git a/drivers/net/wireless/microchip/wilc1000/fw.h b/drivers/net/wireless/microchip/wilc1000/fw.h
index 5c5cac4aab02..7a930e89614c 100644
--- a/drivers/net/wireless/microchip/wilc1000/fw.h
+++ b/drivers/net/wireless/microchip/wilc1000/fw.h
@@ -13,6 +13,12 @@
 #define WILC_MAX_RATES_SUPPORTED		12
 #define WILC_MAX_NUM_PMKIDS			16
 #define WILC_MAX_NUM_SCANNED_CH			14
+#define WILC_NVMEM_MAX_NUM_BANK			6
+#define WILC_NVMEM_BANK_BASE			0x30000000
+#define WILC_NVMEM_LOW_BANK_OFFSET		0x102c
+#define WILC_NVMEM_HIGH_BANK_OFFSET		0x1380
+#define WILC_NVMEM_IS_BANK_USED			BIT(31)
+#define WILC_NVMEM_IS_BANK_INVALID		BIT(30)
 
 struct wilc_assoc_resp {
 	__le16 capab_info;
@@ -127,4 +133,11 @@ struct wilc_external_auth_param {
 	__le32 key_mgmt_suites;
 	__le16 status;
 } __packed;
+
+static inline u32 get_bank_offset_from_bank_index(unsigned int i)
+{
+	return (((i) < 2) ? WILC_NVMEM_LOW_BANK_OFFSET + ((i) * 32) :
+		WILC_NVMEM_HIGH_BANK_OFFSET + ((i) - 2) * 16);
+}
+
 #endif
diff --git a/drivers/net/wireless/microchip/wilc1000/netdev.h b/drivers/net/wireless/microchip/wilc1000/netdev.h
index eecee3973d6a..20ba030022a1 100644
--- a/drivers/net/wireless/microchip/wilc1000/netdev.h
+++ b/drivers/net/wireless/microchip/wilc1000/netdev.h
@@ -14,6 +14,7 @@
 #include <linux/if_arp.h>
 #include <linux/gpio/consumer.h>
 #include <linux/rculist.h>
+#include <uapi/linux/if_ether.h>
 
 #include "hif.h"
 #include "wlan.h"
@@ -278,6 +279,7 @@ struct wilc {
 	struct ieee80211_rate bitrates[ARRAY_SIZE(wilc_bitrates)];
 	struct ieee80211_supported_band band;
 	u32 cipher_suites[ARRAY_SIZE(wilc_cipher_suites)];
+	u8 nv_mac_address[ETH_ALEN];
 };
 
 struct wilc_wfi_mon_priv {
diff --git a/drivers/net/wireless/microchip/wilc1000/wlan.c b/drivers/net/wireless/microchip/wilc1000/wlan.c
index 37c32d17856e..b3fee8c8b7ad 100644
--- a/drivers/net/wireless/microchip/wilc1000/wlan.c
+++ b/drivers/net/wireless/microchip/wilc1000/wlan.c
@@ -1472,6 +1472,55 @@ u32 wilc_get_chipid(struct wilc *wilc, bool update)
 	return wilc->chipid;
 }
 
+int wilc_load_mac_from_nv(struct wilc *wl)
+{
+	int ret = -EINVAL;
+	unsigned int i;
+
+	acquire_bus(wl, WILC_BUS_ACQUIRE_AND_WAKEUP);
+
+	for (i = 0; i < WILC_NVMEM_MAX_NUM_BANK; i++) {
+		int bank_offset = get_bank_offset_from_bank_index(i);
+		u32 reg1, reg2;
+		u8 invalid;
+		u8 used;
+
+		ret = wl->hif_func->hif_read_reg(wl,
+						 WILC_NVMEM_BANK_BASE + bank_offset,
+						 &reg1);
+		if (ret) {
+			pr_err("Can not read address %d lower part", i);
+			break;
+		}
+		ret = wl->hif_func->hif_read_reg(wl,
+						 WILC_NVMEM_BANK_BASE + bank_offset + 4,
+						 &reg2);
+		if (ret) {
+			pr_err("Can not read address %d upper part", i);
+			break;
+		}
+
+		used = FIELD_GET(WILC_NVMEM_IS_BANK_USED, reg1);
+		invalid = FIELD_GET(WILC_NVMEM_IS_BANK_INVALID, reg1);
+		if (!used || invalid)
+			continue;
+
+		wl->nv_mac_address[0] = FIELD_GET(GENMASK(23, 16), reg1);
+		wl->nv_mac_address[1] = FIELD_GET(GENMASK(15, 8), reg1);
+		wl->nv_mac_address[2] = FIELD_GET(GENMASK(7, 0), reg1);
+		wl->nv_mac_address[3] = FIELD_GET(GENMASK(31, 24), reg2);
+		wl->nv_mac_address[4] = FIELD_GET(GENMASK(23, 16), reg2);
+		wl->nv_mac_address[5] = FIELD_GET(GENMASK(15, 8), reg2);
+
+		ret = 0;
+		break;
+	}
+
+	release_bus(wl, WILC_BUS_RELEASE_ALLOW_SLEEP);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(wilc_load_mac_from_nv);
+
 int wilc_wlan_init(struct net_device *dev)
 {
 	int ret = 0;
diff --git a/drivers/net/wireless/microchip/wilc1000/wlan.h b/drivers/net/wireless/microchip/wilc1000/wlan.h
index 54643d8fef04..d72a0a81bbda 100644
--- a/drivers/net/wireless/microchip/wilc1000/wlan.h
+++ b/drivers/net/wireless/microchip/wilc1000/wlan.h
@@ -445,4 +445,5 @@ int wilc_send_config_pkt(struct wilc_vif *vif, u8 mode, struct wid *wids,
 			 u32 count);
 int wilc_wlan_init(struct net_device *dev);
 u32 wilc_get_chipid(struct wilc *wilc, bool update);
+int wilc_load_mac_from_nv(struct wilc *wilc);
 #endif
-- 
cgit v1.2.3-58-ga151


From a5d6b1d453a149e70452d5e69320db04afb520e2 Mon Sep 17 00:00:00 2001
From: Alexis Lothoré <alexis.lothore@bootlin.com>
Date: Tue, 14 May 2024 17:51:23 +0200
Subject: wifi: wilc1000: make sdio deinit function really deinit the sdio card
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In order to be able to read raw registers (eg the nv mac address) in
wilc1000 during probe before the firmware is loaded and running, we need to
run the basic sdio functions initialization, but then we also need to
properly deinitialize those right after, to preserve the current driver
behavior (keeping the chip idle/unconfigured until the corresponding
interface is brought up). Calling wilc_sdio_deinit in its current form is
not enough because it merely resets an internal flag.

Implement a deinit sequence which symmetrically reset all steps performed
in wilc_sdio_init (only for parts activating/deactivating features, for the
sake of simplicity, let's ignore blocks size configuration reset)

Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240514-mac_addr_at_probe-v2-5-afef09f1cd10@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/sdio.c | 45 ++++++++++++++++++++++++++
 1 file changed, 45 insertions(+)

diff --git a/drivers/net/wireless/microchip/wilc1000/sdio.c b/drivers/net/wireless/microchip/wilc1000/sdio.c
index a841dad08410..04d6565df2cb 100644
--- a/drivers/net/wireless/microchip/wilc1000/sdio.c
+++ b/drivers/net/wireless/microchip/wilc1000/sdio.c
@@ -627,7 +627,52 @@ static int wilc_sdio_read(struct wilc *wilc, u32 addr, u8 *buf, u32 size)
 
 static int wilc_sdio_deinit(struct wilc *wilc)
 {
+	struct sdio_func *func = dev_to_sdio_func(wilc->dev);
 	struct wilc_sdio *sdio_priv = wilc->bus_data;
+	struct sdio_cmd52 cmd;
+	int ret;
+
+	cmd.read_write = 1;
+	cmd.function = 0;
+	cmd.raw = 1;
+
+	/* Disable all functions interrupts */
+	cmd.address = SDIO_CCCR_IENx;
+	cmd.data = 0;
+	ret = wilc_sdio_cmd52(wilc, &cmd);
+	if (ret) {
+		dev_err(&func->dev, "Failed to disable functions interrupts\n");
+		return ret;
+	}
+
+	/* Disable all functions */
+	cmd.address = SDIO_CCCR_IOEx;
+	cmd.data = 0;
+	ret = wilc_sdio_cmd52(wilc, &cmd);
+	if (ret) {
+		dev_err(&func->dev,
+			"Failed to reset all functions\n");
+		return ret;
+	}
+
+	/* Disable CSA */
+	cmd.read_write = 0;
+	cmd.address = SDIO_FBR_BASE(1);
+	ret = wilc_sdio_cmd52(wilc, &cmd);
+	if (ret) {
+		dev_err(&func->dev,
+			"Failed to read CSA for function 1\n");
+		return ret;
+	}
+	cmd.read_write = 1;
+	cmd.address = SDIO_FBR_BASE(1);
+	cmd.data &= ~SDIO_FBR_ENABLE_CSA;
+	ret = wilc_sdio_cmd52(wilc, &cmd);
+	if (ret) {
+		dev_err(&func->dev,
+			"Failed to disable CSA for function 1\n");
+		return ret;
+	}
 
 	sdio_priv->isinit = false;
 	return 0;
-- 
cgit v1.2.3-58-ga151


From 83d9b54ee5d4c12d00f2cbf71d827e31994c5a6c Mon Sep 17 00:00:00 2001
From: Ajay Singh <ajay.kathat@microchip.com>
Date: Tue, 14 May 2024 17:51:24 +0200
Subject: wifi: wilc1000: read MAC address from fuse at probe
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The default netdev interface exposed by WILC1000 is registered at probe,
but the chip mac address is not known until ndo_open, which will load and
start chip firmware and then retrieve stored MAC address from it. As a
consequence, the interface has uninitialized value (00:00:00:00:00) until a
user brings up the interface.

Fix MAC address at probe by setting the following steps:
- at probe, read MAC address directly from fuse
- whenever a new netdevice is created, apply saved mac address (which can
  be a user-provided address, or the eFuse Mac address if no address has
  been passed by user)
- whenever an interface is brought up for the first time (and so the
  firmware is loaded and started), enforce netdevice mac address to the
  chip (in case user has changed it)

Reported-by: Heiko Thiery <heiko.thiery@gmail.com>
Closes: https://lore.kernel.org/netdev/CAEyMn7aV-B4OEhHR4Ad0LM3sKCz1-nDqSb9uZNmRWR-hMZ=z+A@mail.gmail.com/T/
Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Co-developed-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Tested-by: Heiko Thiery <heiko.thiery@gmail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240514-mac_addr_at_probe-v2-6-afef09f1cd10@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/netdev.c | 38 ++++++++++++++----------
 drivers/net/wireless/microchip/wilc1000/sdio.c   | 14 +++++++++
 drivers/net/wireless/microchip/wilc1000/spi.c    |  6 ++++
 3 files changed, 42 insertions(+), 16 deletions(-)

diff --git a/drivers/net/wireless/microchip/wilc1000/netdev.c b/drivers/net/wireless/microchip/wilc1000/netdev.c
index 2ee3d9827826..4e2698528a49 100644
--- a/drivers/net/wireless/microchip/wilc1000/netdev.c
+++ b/drivers/net/wireless/microchip/wilc1000/netdev.c
@@ -588,7 +588,6 @@ static int wilc_mac_open(struct net_device *ndev)
 	struct wilc *wl = vif->wilc;
 	int ret = 0;
 	struct mgmt_frame_regs mgmt_regs = {};
-	u8 addr[ETH_ALEN] __aligned(2);
 
 	if (!wl || !wl->dev) {
 		netdev_err(ndev, "device not ready\n");
@@ -607,25 +606,19 @@ static int wilc_mac_open(struct net_device *ndev)
 		return ret;
 	}
 
-	wilc_set_operation_mode(vif, wilc_get_vif_idx(vif), vif->iftype,
-				vif->idx);
-
-	if (is_valid_ether_addr(ndev->dev_addr)) {
-		ether_addr_copy(addr, ndev->dev_addr);
-		wilc_set_mac_address(vif, addr);
-	} else {
-		wilc_get_mac_address(vif, addr);
-		eth_hw_addr_set(ndev, addr);
-	}
 	netdev_dbg(ndev, "Mac address: %pM\n", ndev->dev_addr);
-
-	if (!is_valid_ether_addr(ndev->dev_addr)) {
-		netdev_err(ndev, "Wrong MAC address\n");
+	ret = wilc_set_mac_address(vif, ndev->dev_addr);
+	if (ret) {
+		netdev_err(ndev, "Failed to enforce MAC address in chip");
 		wilc_deinit_host_int(ndev);
-		wilc_wlan_deinitialize(ndev);
-		return -EINVAL;
+		if (!wl->open_ifcs)
+			wilc_wlan_deinitialize(ndev);
+		return ret;
 	}
 
+	wilc_set_operation_mode(vif, wilc_get_vif_idx(vif), vif->iftype,
+				vif->idx);
+
 	mgmt_regs.interface_stypes = vif->mgmt_reg_stypes;
 	/* so we detect a change */
 	vif->mgmt_reg_stypes = 0;
@@ -941,6 +934,7 @@ struct wilc_vif *wilc_netdev_ifc_init(struct wilc *wl, const char *name,
 				      int vif_type, enum nl80211_iftype type,
 				      bool rtnl_locked)
 {
+	u8 mac_address[ETH_ALEN];
 	struct net_device *ndev;
 	struct wilc_vif *vif;
 	int ret;
@@ -969,6 +963,18 @@ struct wilc_vif *wilc_netdev_ifc_init(struct wilc *wl, const char *name,
 	vif->iftype = vif_type;
 	vif->idx = wilc_get_available_idx(wl);
 	vif->mac_opened = 0;
+
+	memcpy(mac_address, wl->nv_mac_address, ETH_ALEN);
+	/* WILC firmware uses locally administered MAC address for the
+	 * second virtual interface (bit 1 of first byte set), but
+	 * since it is possibly not loaded/running yet, reproduce this behavior
+	 * in the driver during interface creation.
+	 */
+	if (vif->idx)
+		mac_address[0] |= 0x2;
+
+	eth_hw_addr_set(vif->ndev, mac_address);
+
 	mutex_lock(&wl->vif_mutex);
 	list_add_tail_rcu(&vif->list, &wl->vif_list);
 	wl->vif_num += 1;
diff --git a/drivers/net/wireless/microchip/wilc1000/sdio.c b/drivers/net/wireless/microchip/wilc1000/sdio.c
index 04d6565df2cb..e6e20c86b791 100644
--- a/drivers/net/wireless/microchip/wilc1000/sdio.c
+++ b/drivers/net/wireless/microchip/wilc1000/sdio.c
@@ -24,6 +24,9 @@ MODULE_DEVICE_TABLE(sdio, wilc_sdio_ids);
 
 #define WILC_SDIO_BLOCK_SIZE 512
 
+static int wilc_sdio_init(struct wilc *wilc, bool resume);
+static int wilc_sdio_deinit(struct wilc *wilc);
+
 struct wilc_sdio {
 	bool irq_gpio;
 	u32 block_size;
@@ -178,6 +181,16 @@ static int wilc_sdio_probe(struct sdio_func *func,
 	}
 	clk_prepare_enable(wilc->rtc_clk);
 
+	wilc_sdio_init(wilc, false);
+
+	ret = wilc_load_mac_from_nv(wilc);
+	if (ret) {
+		pr_err("Can not retrieve MAC address from chip\n");
+		goto clk_disable_unprepare;
+	}
+
+	wilc_sdio_deinit(wilc);
+
 	vif = wilc_netdev_ifc_init(wilc, "wlan%d", WILC_STATION_MODE,
 				   NL80211_IFTYPE_STATION, false);
 	if (IS_ERR(vif)) {
@@ -187,6 +200,7 @@ static int wilc_sdio_probe(struct sdio_func *func,
 
 	dev_info(&func->dev, "Driver Initializing success\n");
 	return 0;
+
 clk_disable_unprepare:
 	clk_disable_unprepare(wilc->rtc_clk);
 dispose_irq:
diff --git a/drivers/net/wireless/microchip/wilc1000/spi.c b/drivers/net/wireless/microchip/wilc1000/spi.c
index add0e70a09ea..5ff940c53ad9 100644
--- a/drivers/net/wireless/microchip/wilc1000/spi.c
+++ b/drivers/net/wireless/microchip/wilc1000/spi.c
@@ -250,6 +250,12 @@ static int wilc_bus_probe(struct spi_device *spi)
 	if (ret)
 		goto power_down;
 
+	ret = wilc_load_mac_from_nv(wilc);
+	if (ret) {
+		pr_err("Can not retrieve MAC address from chip\n");
+		goto power_down;
+	}
+
 	wilc_wlan_power(wilc, false);
 	vif = wilc_netdev_ifc_init(wilc, "wlan%d", WILC_STATION_MODE,
 				   NL80211_IFTYPE_STATION, false);
-- 
cgit v1.2.3-58-ga151


From a17b9f590f6ec2b9f1b12b1db3bf1d181de6b272 Mon Sep 17 00:00:00 2001
From: Rafael Beims <rafael.beims@toradex.com>
Date: Fri, 10 May 2024 13:04:58 +0200
Subject: wifi: mwifiex: Fix interface type change

When changing the interface type we also need to update the bss_num, the
driver private data is searched based on a unique (bss_type, bss_num)
tuple, therefore every time bss_type changes, bss_num must also change.

This fixes for example an issue in which, after the mode changed, a
wireless scan on the changed interface would not finish, leading to
repeated -EBUSY messages to userspace when other scan requests were
sent.

Fixes: c606008b7062 ("mwifiex: Properly initialize private structure on interface type changes")
Cc: stable@vger.kernel.org
Signed-off-by: Rafael Beims <rafael.beims@toradex.com>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240510110458.15475-1-francesco@dolcini.it
---
 drivers/net/wireless/marvell/mwifiex/cfg80211.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
index b909a7665e9c..155eb0fab12a 100644
--- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c
+++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c
@@ -926,6 +926,8 @@ mwifiex_init_new_priv_params(struct mwifiex_private *priv,
 		return -EOPNOTSUPP;
 	}
 
+	priv->bss_num = mwifiex_get_unused_bss_num(adapter, priv->bss_type);
+
 	spin_lock_irqsave(&adapter->main_proc_lock, flags);
 	adapter->main_locked = false;
 	spin_unlock_irqrestore(&adapter->main_proc_lock, flags);
-- 
cgit v1.2.3-58-ga151


From 94133cf24bb33889aac267a7f0e3e6a08b8a8e5a Mon Sep 17 00:00:00 2001
From: Mykyta Yatsenko <yatsenko@meta.com>
Date: Tue, 14 May 2024 14:12:21 +0100
Subject: bpftool: Introduce btf c dump sorting

Sort bpftool c dump output; aiming to simplify vmlinux.h diffing and
forcing more natural type definitions ordering.

Definitions are sorted first by their BTF kind ranks, then by their base
type name and by their own name.

Type ranks

Assign ranks to btf kinds (defined in function btf_type_rank) to set
next order:
1. Anonymous enums/enums64
2. Named enums/enums64
3. Trivial types typedefs (ints, then floats)
4. Structs/Unions
5. Function prototypes
6. Forward declarations

Type rank is set to maximum for unnamed reference types, structs and
unions to avoid emitting those types early. They will be emitted as
part of the type chain starting with named type.

Lexicographical ordering

Each type is assigned a sort_name and own_name.
sort_name is the resolved name of the final base type for reference
types (typedef, pointer, array etc). Sorting by sort_name allows to
group typedefs of the same base type. sort_name for non-reference type
is the same as own_name. own_name is a direct name of particular type,
is used as final sorting step.

Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Tested-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Quentin Monnet <qmo@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240514131221.20585-1-yatsenko@meta.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/Documentation/bpftool-btf.rst |   6 +-
 tools/bpf/bpftool/bash-completion/bpftool       |   3 +
 tools/bpf/bpftool/btf.c                         | 138 +++++++++++++++++++++++-
 3 files changed, 140 insertions(+), 7 deletions(-)

diff --git a/tools/bpf/bpftool/Documentation/bpftool-btf.rst b/tools/bpf/bpftool/Documentation/bpftool-btf.rst
index eaba24320fb2..3f6bca03ad2e 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-btf.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-btf.rst
@@ -28,7 +28,7 @@ BTF COMMANDS
 | **bpftool** **btf help**
 |
 | *BTF_SRC* := { **id** *BTF_ID* | **prog** *PROG* | **map** *MAP* [{**key** | **value** | **kv** | **all**}] | **file** *FILE* }
-| *FORMAT* := { **raw** | **c** }
+| *FORMAT* := { **raw** | **c** [**unsorted**] }
 | *MAP* := { **id** *MAP_ID* | **pinned** *FILE* }
 | *PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* | **name** *PROG_NAME* }
 
@@ -63,7 +63,9 @@ bpftool btf dump *BTF_SRC*
     pahole.
 
     **format** option can be used to override default (raw) output format. Raw
-    (**raw**) or C-syntax (**c**) output formats are supported.
+    (**raw**) or C-syntax (**c**) output formats are supported. With C-style
+    formatting, the output is sorted by default. Use the **unsorted** option
+    to avoid sorting the output.
 
 bpftool btf help
     Print short help message.
diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
index 04afe2ac2228..be99d49b8714 100644
--- a/tools/bpf/bpftool/bash-completion/bpftool
+++ b/tools/bpf/bpftool/bash-completion/bpftool
@@ -930,6 +930,9 @@ _bpftool()
                         format)
                             COMPREPLY=( $( compgen -W "c raw" -- "$cur" ) )
                             ;;
+                        c)
+                            COMPREPLY=( $( compgen -W "unsorted" -- "$cur" ) )
+                            ;;
                         *)
                             # emit extra options
                             case ${words[3]} in
diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c
index 91fcb75babe3..af047dedde38 100644
--- a/tools/bpf/bpftool/btf.c
+++ b/tools/bpf/bpftool/btf.c
@@ -43,6 +43,13 @@ static const char * const btf_kind_str[NR_BTF_KINDS] = {
 	[BTF_KIND_ENUM64]	= "ENUM64",
 };
 
+struct sort_datum {
+	int index;
+	int type_rank;
+	const char *sort_name;
+	const char *own_name;
+};
+
 static const char *btf_int_enc_str(__u8 encoding)
 {
 	switch (encoding) {
@@ -460,9 +467,122 @@ static void __printf(2, 0) btf_dump_printf(void *ctx,
 	vfprintf(stdout, fmt, args);
 }
 
+static int btf_type_rank(const struct btf *btf, __u32 index, bool has_name)
+{
+	const struct btf_type *t = btf__type_by_id(btf, index);
+	const int kind = btf_kind(t);
+	const int max_rank = 10;
+
+	if (t->name_off)
+		has_name = true;
+
+	switch (kind) {
+	case BTF_KIND_ENUM:
+	case BTF_KIND_ENUM64:
+		return has_name ? 1 : 0;
+	case BTF_KIND_INT:
+	case BTF_KIND_FLOAT:
+		return 2;
+	case BTF_KIND_STRUCT:
+	case BTF_KIND_UNION:
+		return has_name ? 3 : max_rank;
+	case BTF_KIND_FUNC_PROTO:
+		return has_name ? 4 : max_rank;
+	case BTF_KIND_ARRAY:
+		if (has_name)
+			return btf_type_rank(btf, btf_array(t)->type, has_name);
+		return max_rank;
+	case BTF_KIND_TYPE_TAG:
+	case BTF_KIND_CONST:
+	case BTF_KIND_PTR:
+	case BTF_KIND_VOLATILE:
+	case BTF_KIND_RESTRICT:
+	case BTF_KIND_TYPEDEF:
+	case BTF_KIND_DECL_TAG:
+		if (has_name)
+			return btf_type_rank(btf, t->type, has_name);
+		return max_rank;
+	default:
+		return max_rank;
+	}
+}
+
+static const char *btf_type_sort_name(const struct btf *btf, __u32 index, bool from_ref)
+{
+	const struct btf_type *t = btf__type_by_id(btf, index);
+
+	switch (btf_kind(t)) {
+	case BTF_KIND_ENUM:
+	case BTF_KIND_ENUM64: {
+		int name_off = t->name_off;
+
+		/* Use name of the first element for anonymous enums if allowed */
+		if (!from_ref && !t->name_off && btf_vlen(t))
+			name_off = btf_enum(t)->name_off;
+
+		return btf__name_by_offset(btf, name_off);
+	}
+	case BTF_KIND_ARRAY:
+		return btf_type_sort_name(btf, btf_array(t)->type, true);
+	case BTF_KIND_TYPE_TAG:
+	case BTF_KIND_CONST:
+	case BTF_KIND_PTR:
+	case BTF_KIND_VOLATILE:
+	case BTF_KIND_RESTRICT:
+	case BTF_KIND_TYPEDEF:
+	case BTF_KIND_DECL_TAG:
+		return btf_type_sort_name(btf, t->type, true);
+	default:
+		return btf__name_by_offset(btf, t->name_off);
+	}
+	return NULL;
+}
+
+static int btf_type_compare(const void *left, const void *right)
+{
+	const struct sort_datum *d1 = (const struct sort_datum *)left;
+	const struct sort_datum *d2 = (const struct sort_datum *)right;
+	int r;
+
+	if (d1->type_rank != d2->type_rank)
+		return d1->type_rank < d2->type_rank ? -1 : 1;
+
+	r = strcmp(d1->sort_name, d2->sort_name);
+	if (r)
+		return r;
+
+	return strcmp(d1->own_name, d2->own_name);
+}
+
+static struct sort_datum *sort_btf_c(const struct btf *btf)
+{
+	struct sort_datum *datums;
+	int n;
+
+	n = btf__type_cnt(btf);
+	datums = malloc(sizeof(struct sort_datum) * n);
+	if (!datums)
+		return NULL;
+
+	for (int i = 0; i < n; ++i) {
+		struct sort_datum *d = datums + i;
+		const struct btf_type *t = btf__type_by_id(btf, i);
+
+		d->index = i;
+		d->type_rank = btf_type_rank(btf, i, false);
+		d->sort_name = btf_type_sort_name(btf, i, false);
+		d->own_name = btf__name_by_offset(btf, t->name_off);
+	}
+
+	qsort(datums, n, sizeof(struct sort_datum), btf_type_compare);
+
+	return datums;
+}
+
 static int dump_btf_c(const struct btf *btf,
-		      __u32 *root_type_ids, int root_type_cnt)
+		      __u32 *root_type_ids, int root_type_cnt, bool sort_dump)
 {
+	struct sort_datum *datums = NULL;
 	struct btf_dump *d;
 	int err = 0, i;
 
@@ -486,8 +606,12 @@ static int dump_btf_c(const struct btf *btf,
 	} else {
 		int cnt = btf__type_cnt(btf);
 
+		if (sort_dump)
+			datums = sort_btf_c(btf);
 		for (i = 1; i < cnt; i++) {
-			err = btf_dump__dump_type(d, i);
+			int idx = datums ? datums[i].index : i;
+
+			err = btf_dump__dump_type(d, idx);
 			if (err)
 				goto done;
 		}
@@ -500,6 +624,7 @@ static int dump_btf_c(const struct btf *btf,
 	printf("#endif /* __VMLINUX_H__ */\n");
 
 done:
+	free(datums);
 	btf_dump__free(d);
 	return err;
 }
@@ -549,10 +674,10 @@ static bool btf_is_kernel_module(__u32 btf_id)
 
 static int do_dump(int argc, char **argv)
 {
+	bool dump_c = false, sort_dump_c = true;
 	struct btf *btf = NULL, *base = NULL;
 	__u32 root_type_ids[2];
 	int root_type_cnt = 0;
-	bool dump_c = false;
 	__u32 btf_id = -1;
 	const char *src;
 	int fd = -1;
@@ -663,6 +788,9 @@ static int do_dump(int argc, char **argv)
 				goto done;
 			}
 			NEXT_ARG();
+		} else if (is_prefix(*argv, "unsorted")) {
+			sort_dump_c = false;
+			NEXT_ARG();
 		} else {
 			p_err("unrecognized option: '%s'", *argv);
 			err = -EINVAL;
@@ -691,7 +819,7 @@ static int do_dump(int argc, char **argv)
 			err = -ENOTSUP;
 			goto done;
 		}
-		err = dump_btf_c(btf, root_type_ids, root_type_cnt);
+		err = dump_btf_c(btf, root_type_ids, root_type_cnt, sort_dump_c);
 	} else {
 		err = dump_btf_raw(btf, root_type_ids, root_type_cnt);
 	}
@@ -1063,7 +1191,7 @@ static int do_help(int argc, char **argv)
 		"       %1$s %2$s help\n"
 		"\n"
 		"       BTF_SRC := { id BTF_ID | prog PROG | map MAP [{key | value | kv | all}] | file FILE }\n"
-		"       FORMAT  := { raw | c }\n"
+		"       FORMAT  := { raw | c [unsorted] }\n"
 		"       " HELP_SPEC_MAP "\n"
 		"       " HELP_SPEC_PROGRAM "\n"
 		"       " HELP_SPEC_OPTIONS " |\n"
-- 
cgit v1.2.3-58-ga151


From e7b64f9d3f5b10186038201e0b91f734cbd7fc3d Mon Sep 17 00:00:00 2001
From: Artem Savkov <asavkov@redhat.com>
Date: Mon, 13 May 2024 13:26:58 +0200
Subject: bpftool: Fix make dependencies for vmlinux.h

With pre-generated vmlinux.h there is no dependency on neither vmlinux
nor bootstrap bpftool. Define dependencies separately for both modes.
This avoids needless rebuilds in some corner cases.

Suggested-by: Jan Stancek <jstancek@redhat.com>
Signed-off-by: Artem Savkov <asavkov@redhat.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Quentin Monnet <qmo@kernel.org>
Link: https://lore.kernel.org/bpf/20240513112658.43691-1-asavkov@redhat.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/Makefile | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/bpf/bpftool/Makefile b/tools/bpf/bpftool/Makefile
index dfa4f1bebbb3..ba927379eb20 100644
--- a/tools/bpf/bpftool/Makefile
+++ b/tools/bpf/bpftool/Makefile
@@ -204,10 +204,11 @@ ifeq ($(feature-clang-bpf-co-re),1)
 
 BUILD_BPF_SKELS := 1
 
-$(OUTPUT)vmlinux.h: $(VMLINUX_BTF) $(BPFTOOL_BOOTSTRAP)
 ifeq ($(VMLINUX_H),)
+$(OUTPUT)vmlinux.h: $(VMLINUX_BTF) $(BPFTOOL_BOOTSTRAP)
 	$(QUIET_GEN)$(BPFTOOL_BOOTSTRAP) btf dump file $< format c > $@
 else
+$(OUTPUT)vmlinux.h: $(VMLINUX_H)
 	$(Q)cp "$(VMLINUX_H)" $@
 endif
 
-- 
cgit v1.2.3-58-ga151


From 34021caef79f76e70ac31247d321ecd0683c4939 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Tue, 14 May 2024 17:27:16 +0100
Subject: kbuild, bpf: Use test-ge check for v1.25-only pahole

There is no need to set the pahole v1.25-only flags in an
"ifeq" version clause; we are already in a <= v1.25 branch
of "ifeq", so that combined with a "test-ge" v1.25 ensures the
flags will be applied for v1.25 only.

Suggested-by: Masahiro Yamada <masahiroy@kernel.org>
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240514162716.2448265-1-alan.maguire@oracle.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 scripts/Makefile.btf | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/scripts/Makefile.btf b/scripts/Makefile.btf
index 2d6e5ed9081e..bca8a8f26ea4 100644
--- a/scripts/Makefile.btf
+++ b/scripts/Makefile.btf
@@ -14,9 +14,7 @@ pahole-flags-$(call test-ge, $(pahole-ver), 121)	+= --btf_gen_floats
 
 pahole-flags-$(call test-ge, $(pahole-ver), 122)	+= -j
 
-ifeq ($(pahole-ver), 125)
-pahole-flags-y	+= --skip_encoding_btf_inconsistent_proto --btf_gen_optimized
-endif
+pahole-flags-$(call test-ge, $(pahole-ver), 125)	+= --skip_encoding_btf_inconsistent_proto --btf_gen_optimized
 
 else
 
-- 
cgit v1.2.3-58-ga151


From 6f130e4d4a5f7174f98300376f3994817ad7e21c Mon Sep 17 00:00:00 2001
From: Mohammad Shehar Yaar Tausif <sheharyaar48@gmail.com>
Date: Thu, 16 May 2024 12:54:11 +0530
Subject: bpf: Fix order of args in call to bpf_map_kvcalloc

The original function call passed size of smap->bucket before the number of
buckets which raises the error 'calloc-transposed-args' on compilation.

Signed-off-by: Mohammad Shehar Yaar Tausif <sheharyaar48@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240516072411.42016-1-sheharyaar48@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/bpf_local_storage.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c
index 976cb258a0ed..c938dea5ddbf 100644
--- a/kernel/bpf/bpf_local_storage.c
+++ b/kernel/bpf/bpf_local_storage.c
@@ -782,8 +782,8 @@ bpf_local_storage_map_alloc(union bpf_attr *attr,
 	nbuckets = max_t(u32, 2, nbuckets);
 	smap->bucket_log = ilog2(nbuckets);
 
-	smap->buckets = bpf_map_kvcalloc(&smap->map, sizeof(*smap->buckets),
-					 nbuckets, GFP_USER | __GFP_NOWARN);
+	smap->buckets = bpf_map_kvcalloc(&smap->map, nbuckets,
+					 sizeof(*smap->buckets), GFP_USER | __GFP_NOWARN);
 	if (!smap->buckets) {
 		err = -ENOMEM;
 		goto free_smap;
-- 
cgit v1.2.3-58-ga151


From 2785ea9673a7305abeea87111849a4e04b0f4626 Mon Sep 17 00:00:00 2001
From: Dmitry Antipov <dmantipov@yandex.ru>
Date: Wed, 15 May 2024 18:28:24 +0300
Subject: wifi: rt2x00: remove unused delayed work data from link description

Remove unused 'agc_work' and 'vco_work' members of 'struct link'.
This follows commit d96324703ffa ("rt2x00: merge agc and vco works
with link tuner") where AGC and VCO calibration code was completely
redesigned. Compile tested only.

Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
Acked-by: Stanislaw Gruszka <stf_xl@wp.pl>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240515152824.143537-1-dmantipov@yandex.ru
---
 drivers/net/wireless/ralink/rt2x00/rt2x00.h | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00.h b/drivers/net/wireless/ralink/rt2x00/rt2x00.h
index 82af01448a0a..bb648f95dfdd 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/ralink/rt2x00/rt2x00.h
@@ -335,16 +335,6 @@ struct link {
 	struct delayed_work watchdog_work;
 	unsigned int watchdog_interval;
 	unsigned int watchdog;
-
-	/*
-	 * Work structure for scheduling periodic AGC adjustments.
-	 */
-	struct delayed_work agc_work;
-
-	/*
-	 * Work structure for scheduling periodic VCO calibration.
-	 */
-	struct delayed_work vco_work;
 };
 
 enum rt2x00_delayed_flags {
-- 
cgit v1.2.3-58-ga151


From 26dd8ccdba4d492d00e35c410bc4314e043bdf73 Mon Sep 17 00:00:00 2001
From: Muna Sinada <quic_msinada@quicinc.com>
Date: Fri, 17 May 2024 10:00:30 +0300
Subject: wifi: ath12k: dynamic VLAN support

Add support for dynamic VLAN. VLAN group traffic is encapsulated and
encrypted in mac80211 and driver needs to set flags for the VLAN group
traffic to skip hardware encapsulation and encryption.

VLAN group traffic utilizes ext MSDU and HTT Metadata to set
encapsulation type to RAW and encryption type to OPEN in order to
inform firmware to skip hardware encapsulation and encryption.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Muna Sinada <quic_msinada@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240515192830.995013-1-quic_msinada@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_tx.c    | 60 ++++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/hal_desc.h | 25 +++++++++++++
 drivers/net/wireless/ath/ath12k/hw.c       |  6 ++-
 3 files changed, 89 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.c b/drivers/net/wireless/ath/ath12k/dp_tx.c
index f57bc8d69715..c4cfa7cf7cb9 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.c
@@ -124,6 +124,44 @@ static void ath12k_hal_tx_cmd_ext_desc_setup(struct ath12k_base *ab,
 						 HAL_TX_MSDU_EXT_INFO1_ENCRYPT_TYPE);
 }
 
+#define HTT_META_DATA_ALIGNMENT 0x8
+
+static void *ath12k_dp_metadata_align_skb(struct sk_buff *skb, u8 tail_len)
+{
+	struct sk_buff *tail;
+	void *metadata;
+
+	if (unlikely(skb_cow_data(skb, tail_len, &tail) < 0))
+		return NULL;
+
+	metadata = pskb_put(skb, tail, tail_len);
+	memset(metadata, 0, tail_len);
+	return metadata;
+}
+
+/* Preparing HTT Metadata when utilized with ext MSDU */
+static int ath12k_dp_prepare_htt_metadata(struct sk_buff *skb)
+{
+	struct hal_tx_msdu_metadata *desc_ext;
+	u8 htt_desc_size;
+	/* Size rounded of multiple of 8 bytes */
+	u8 htt_desc_size_aligned;
+
+	htt_desc_size = sizeof(struct hal_tx_msdu_metadata);
+	htt_desc_size_aligned = ALIGN(htt_desc_size, HTT_META_DATA_ALIGNMENT);
+
+	desc_ext = ath12k_dp_metadata_align_skb(skb, htt_desc_size_aligned);
+	if (!desc_ext)
+		return -ENOMEM;
+
+	desc_ext->info0 = le32_encode_bits(1, HAL_TX_MSDU_METADATA_INFO0_ENCRYPT_FLAG) |
+			  le32_encode_bits(0, HAL_TX_MSDU_METADATA_INFO0_ENCRYPT_TYPE) |
+			  le32_encode_bits(1,
+					   HAL_TX_MSDU_METADATA_INFO0_HOST_TX_DESC_POOL);
+
+	return 0;
+}
+
 int ath12k_dp_tx(struct ath12k *ar, struct ath12k_vif *arvif,
 		 struct sk_buff *skb)
 {
@@ -145,6 +183,7 @@ int ath12k_dp_tx(struct ath12k *ar, struct ath12k_vif *arvif,
 	u8 ring_selector, ring_map = 0;
 	bool tcl_ring_retry;
 	bool msdu_ext_desc = false;
+	bool add_htt_metadata = false;
 
 	if (test_bit(ATH12K_FLAG_CRASH_FLUSH, &ar->ab->dev_flags))
 		return -ESHUTDOWN;
@@ -248,6 +287,18 @@ tcl_ring_sel:
 		goto fail_remove_tx_buf;
 	}
 
+	if (!test_bit(ATH12K_FLAG_HW_CRYPTO_DISABLED, &ar->ab->dev_flags) &&
+	    !(skb_cb->flags & ATH12K_SKB_HW_80211_ENCAP) &&
+	    !(skb_cb->flags & ATH12K_SKB_CIPHER_SET) &&
+	    ieee80211_has_protected(hdr->frame_control)) {
+		/* Add metadata for sw encrypted vlan group traffic */
+		add_htt_metadata = true;
+		msdu_ext_desc = true;
+		ti.flags0 |= u32_encode_bits(1, HAL_TCL_DATA_CMD_INFO2_TO_FW);
+		ti.encap_type = HAL_TCL_ENCAP_TYPE_RAW;
+		ti.encrypt_type = HAL_ENCRYPT_TYPE_OPEN;
+	}
+
 	tx_desc->skb = skb;
 	tx_desc->mac_id = ar->pdev_idx;
 	ti.desc_id = tx_desc->desc_id;
@@ -269,6 +320,15 @@ tcl_ring_sel:
 		msg = (struct hal_tx_msdu_ext_desc *)skb_ext_desc->data;
 		ath12k_hal_tx_cmd_ext_desc_setup(ab, msg, &ti);
 
+		if (add_htt_metadata) {
+			ret = ath12k_dp_prepare_htt_metadata(skb_ext_desc);
+			if (ret < 0) {
+				ath12k_dbg(ab, ATH12K_DBG_DP_TX,
+					   "Failed to add HTT meta data, dropping packet\n");
+				goto fail_unmap_dma;
+			}
+		}
+
 		ti.paddr = dma_map_single(ab->dev, skb_ext_desc->data,
 					  skb_ext_desc->len, DMA_TO_DEVICE);
 		ret = dma_mapping_error(ab->dev, ti.paddr);
diff --git a/drivers/net/wireless/ath/ath12k/hal_desc.h b/drivers/net/wireless/ath/ath12k/hal_desc.h
index 814c02f876d6..02b7db06b24e 100644
--- a/drivers/net/wireless/ath/ath12k/hal_desc.h
+++ b/drivers/net/wireless/ath/ath12k/hal_desc.h
@@ -2984,4 +2984,29 @@ struct hal_mon_dest_desc {
  *	updated by SRNG.
  */
 
+#define HAL_TX_MSDU_METADATA_INFO0_ENCRYPT_FLAG		BIT(8)
+#define HAL_TX_MSDU_METADATA_INFO0_ENCRYPT_TYPE		GENMASK(16, 15)
+#define HAL_TX_MSDU_METADATA_INFO0_HOST_TX_DESC_POOL	BIT(31)
+
+struct hal_tx_msdu_metadata {
+	__le32 info0;
+	__le32 rsvd0[6];
+} __packed;
+
+/* hal_tx_msdu_metadata
+ * valid_encrypt_type
+ *		if set, encrypt type is valid
+ * encrypt_type
+ *		0 = NO_ENCRYPT,
+ *		1 = ENCRYPT,
+ *		2 ~ 3 - Reserved
+ * host_tx_desc_pool
+ *		If set, Firmware allocates tx_descriptors
+ *		in WAL_BUFFERID_TX_HOST_DATA_EXP,instead
+ *		of WAL_BUFFERID_TX_TCL_DATA_EXP.
+ *		Use cases:
+ *		Any time firmware uses TQM-BYPASS for Data
+ *		TID, firmware expect host to set this bit.
+ */
+
 #endif /* ATH12K_HAL_DESC_H */
diff --git a/drivers/net/wireless/ath/ath12k/hw.c b/drivers/net/wireless/ath/ath12k/hw.c
index 5a1d406e6b1f..b45f929a0204 100644
--- a/drivers/net/wireless/ath/ath12k/hw.c
+++ b/drivers/net/wireless/ath/ath12k/hw.c
@@ -891,7 +891,8 @@ static const struct ath12k_hw_params ath12k_hw_params[] = {
 
 		.interface_modes = BIT(NL80211_IFTYPE_STATION) |
 					BIT(NL80211_IFTYPE_AP) |
-					BIT(NL80211_IFTYPE_MESH_POINT),
+					BIT(NL80211_IFTYPE_MESH_POINT) |
+					BIT(NL80211_IFTYPE_AP_VLAN),
 		.supports_monitor = false,
 
 		.idle_ps = false,
@@ -1038,7 +1039,8 @@ static const struct ath12k_hw_params ath12k_hw_params[] = {
 
 		.interface_modes = BIT(NL80211_IFTYPE_STATION) |
 					BIT(NL80211_IFTYPE_AP) |
-					BIT(NL80211_IFTYPE_MESH_POINT),
+					BIT(NL80211_IFTYPE_MESH_POINT) |
+					BIT(NL80211_IFTYPE_AP_VLAN),
 		.supports_monitor = false,
 
 		.idle_ps = false,
-- 
cgit v1.2.3-58-ga151


From 33370412eced2dc7f81f4324e109d69319cafd82 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Mon, 20 May 2024 14:58:28 +0300
Subject: wifi: ath12k: fix Smatch warnings on ath12k_core_suspend()

Smatch is throwing below warning:

Commit 692921ead832 ("wifi: ath12k: flush all packets before
suspend") leads to the following Smatch static checker warning:

	drivers/net/wireless/ath/ath12k/core.c:58 ath12k_core_suspend()
	warn: sleeping in atomic context

and also gives the reason:

drivers/net/wireless/ath/ath12k/core.c
    48         int ret, i;
    49
    50         if (!ab->hw_params->supports_suspend)
    51                 return -EOPNOTSUPP;
    52
    53         rcu_read_lock();
               ^^^^^^^^^^^^^^^
Disables preemption.

    54         for (i = 0; i < ab->num_radios; i++) {
    55                 ar = ath12k_mac_get_ar_by_pdev_id(ab, i);
    56                 if (!ar)
    57                         continue;
--> 58                 ret = ath12k_mac_wait_tx_complete(ar);
                                        ^^^^^^^
Sleeping in atomic context.

    59                 if (ret) {
    60                         ath12k_warn(ab, "failed to wait tx complete: %d\n", ret);
    61                         rcu_read_unlock();
    62                         return ret;
    63                 }
    64         }
    65         rcu_read_unlock();

But it is weird that no warning on this in run time even with
CONFIG_DEBUG_ATOMIC_SLEEP=y. With some debug it is found that this is
because: when system goes to suspend, ath12k_mac_op_stop() gets called
where then in ath12k_mac_stop() ab->pdevs_active[ar->pdev_idx] is cleared.
This results in ath12k_mac_get_ar_by_pdev_id() always returning a NULL ar,
and thereby ath12k_mac_wait_tx_complete() never gets a chance to run.

Fix it by retrieving ar directly from ab->pdevs[].ar instead of using
ath12k_mac_get_ar_by_pdev_id(). Since ab->pdevs[].ar is set at boot time
and won't get cleared when suspend, ath12k_mac_wait_tx_complete() won't
be skipped. In addition, with ath12k_mac_get_ar_by_pdev_id() removed,
rcu_read_lock()/unlock() are not needed any more, so remove them. This
also fixes the warning above.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Fixes: 692921ead832 ("wifi: ath12k: flush all packets before suspend")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/ath12k/7a96ca11-80b5-4751-8cfc-fa637f3aa63a@moroto.mountain/
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240511095045.9623-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c
index 9482d5db71e7..e7f628e935e4 100644
--- a/drivers/net/wireless/ath/ath12k/core.c
+++ b/drivers/net/wireless/ath/ath12k/core.c
@@ -50,19 +50,16 @@ int ath12k_core_suspend(struct ath12k_base *ab)
 	if (!ab->hw_params->supports_suspend)
 		return -EOPNOTSUPP;
 
-	rcu_read_lock();
 	for (i = 0; i < ab->num_radios; i++) {
-		ar = ath12k_mac_get_ar_by_pdev_id(ab, i);
+		ar = ab->pdevs[i].ar;
 		if (!ar)
 			continue;
 		ret = ath12k_mac_wait_tx_complete(ar);
 		if (ret) {
 			ath12k_warn(ab, "failed to wait tx complete: %d\n", ret);
-			rcu_read_unlock();
 			return ret;
 		}
 	}
-	rcu_read_unlock();
 
 	/* PM framework skips suspend_late/resume_early callbacks
 	 * if other devices report errors in their suspend callbacks.
-- 
cgit v1.2.3-58-ga151


From f4aba3471cfb9ccf69b476463f19b4c50fef6b14 Mon Sep 17 00:00:00 2001
From: Ivan Babrou <ivan@cloudflare.com>
Date: Mon, 20 May 2024 15:51:49 -0700
Subject: bpftool: Un-const bpf_func_info to fix it for llvm 17 and newer

LLVM 17 started treating const structs as constants:

* https://github.com/llvm/llvm-project/commit/0b2d5b967d98

Combined with pointer laundering via ptr_to_u64, which takes a const ptr,
but in reality treats the underlying memory as mutable, this makes clang
always pass zero to btf__type_by_id, which breaks full name resolution.

Disassembly before (LLVM 16) and after (LLVM 17):

    -    8b 75 cc                 mov    -0x34(%rbp),%esi
    -    e8 47 8d 02 00           call   3f5b0 <btf__type_by_id>
    +    31 f6                    xor    %esi,%esi
    +    e8 a9 8c 02 00           call   3f510 <btf__type_by_id>

It's a bigger project to fix this properly (and a question whether LLVM
itself should detect this), but for right now let's just fix bpftool.

For more information, see this thread in bpf mailing list:

* https://lore.kernel.org/bpf/CABWYdi0ymezpYsQsPv7qzpx2fWuTkoD1-wG1eT-9x-TSREFrQg@mail.gmail.com/T/

Fixes: b662000aff84 ("bpftool: Adding support for BTF program names")
Signed-off-by: Ivan Babrou <ivan@cloudflare.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Nick Desaulniers <ndesaulniers@google.com>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/bpf/20240520225149.5517-1-ivan@cloudflare.com
---
 tools/bpf/bpftool/common.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/bpf/bpftool/common.c b/tools/bpf/bpftool/common.c
index 958e92acca8e..9b75639434b8 100644
--- a/tools/bpf/bpftool/common.c
+++ b/tools/bpf/bpftool/common.c
@@ -410,7 +410,7 @@ void get_prog_full_name(const struct bpf_prog_info *prog_info, int prog_fd,
 {
 	const char *prog_name = prog_info->name;
 	const struct btf_type *func_type;
-	const struct bpf_func_info finfo = {};
+	struct bpf_func_info finfo = {};
 	struct bpf_prog_info info = {};
 	__u32 info_len = sizeof(info);
 	struct btf *prog_btf = NULL;
-- 
cgit v1.2.3-58-ga151


From 1b0215a3633a4c54ed7ec3af93e7a782dda8d965 Mon Sep 17 00:00:00 2001
From: Ying Zhang <yingzhang098@163.com>
Date: Tue, 21 May 2024 16:17:02 +0000
Subject: bpf: Remove unused variable "prev_state"

The variable "prev_state" is not used for any actual operations

v2: Fix commit message and description.

Signed-off-by: Ying Zhang <yingzhang098@163.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/bpf/20240521161702.4339-1-yingzhang098@163.com
---
 samples/bpf/cpustat_kern.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/samples/bpf/cpustat_kern.c b/samples/bpf/cpustat_kern.c
index 944f13fe164a..7ec7143e2757 100644
--- a/samples/bpf/cpustat_kern.c
+++ b/samples/bpf/cpustat_kern.c
@@ -211,7 +211,7 @@ int bpf_prog1(struct cpu_args *ctx)
 SEC("tracepoint/power/cpu_frequency")
 int bpf_prog2(struct cpu_args *ctx)
 {
-	u64 *pts, *cstate, *pstate, prev_state, cur_ts, delta;
+	u64 *pts, *cstate, *pstate, cur_ts, delta;
 	u32 key, cpu, pstate_idx;
 	u64 *val;
 
@@ -232,7 +232,6 @@ int bpf_prog2(struct cpu_args *ctx)
 	if (!cstate)
 		return 0;
 
-	prev_state = *pstate;
 	*pstate = ctx->state;
 
 	if (!*pts) {
-- 
cgit v1.2.3-58-ga151


From 6c8d7598dfed759bf1d9d0322b4c2b42eb7252d8 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Fri, 17 May 2024 14:21:46 +0800
Subject: selftests/bpf: Fix prog numbers in test_sockmap

bpf_prog5 and bpf_prog7 are removed from progs/test_sockmap_kern.h in
commit d79a32129b21 ("bpf: Selftests, remove prints from sockmap tests"),
now there are only 9 progs in it, not 11:

	SEC("sk_skb1")
	int bpf_prog1(struct __sk_buff *skb)
	SEC("sk_skb2")
	int bpf_prog2(struct __sk_buff *skb)
	SEC("sk_skb3")
	int bpf_prog3(struct __sk_buff *skb)
	SEC("sockops")
	int bpf_sockmap(struct bpf_sock_ops *skops)
	SEC("sk_msg1")
	int bpf_prog4(struct sk_msg_md *msg)
	SEC("sk_msg2")
	int bpf_prog6(struct sk_msg_md *msg)
	SEC("sk_msg3")
	int bpf_prog8(struct sk_msg_md *msg)
	SEC("sk_msg4")
	int bpf_prog9(struct sk_msg_md *msg)
	SEC("sk_msg5")
	int bpf_prog10(struct sk_msg_md *msg)

This patch updates the array sizes of prog_fd[], prog_attach_type[] and
prog_type[] from 11 to 9 accordingly.

Fixes: d79a32129b21 ("bpf: Selftests, remove prints from sockmap tests")
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/9c10d9f974f07fcb354a43a8eca67acb2fafc587.1715926605.git.tanggeliang@kylinos.cn
---
 tools/testing/selftests/bpf/test_sockmap.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index 92752f5eeded..4499b3cfc3a6 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -63,7 +63,7 @@ int passed;
 int failed;
 int map_fd[9];
 struct bpf_map *maps[9];
-int prog_fd[11];
+int prog_fd[9];
 
 int txmsg_pass;
 int txmsg_redir;
@@ -1793,8 +1793,6 @@ int prog_attach_type[] = {
 	BPF_SK_MSG_VERDICT,
 	BPF_SK_MSG_VERDICT,
 	BPF_SK_MSG_VERDICT,
-	BPF_SK_MSG_VERDICT,
-	BPF_SK_MSG_VERDICT,
 };
 
 int prog_type[] = {
@@ -1807,8 +1805,6 @@ int prog_type[] = {
 	BPF_PROG_TYPE_SK_MSG,
 	BPF_PROG_TYPE_SK_MSG,
 	BPF_PROG_TYPE_SK_MSG,
-	BPF_PROG_TYPE_SK_MSG,
-	BPF_PROG_TYPE_SK_MSG,
 };
 
 static int populate_progs(char *bpf_file)
-- 
cgit v1.2.3-58-ga151


From ece4b296904167336d0aaab26bd7122018835202 Mon Sep 17 00:00:00 2001
From: Brad Cowie <brad@faucet.nz>
Date: Wed, 22 May 2024 17:07:11 +1200
Subject: net: netfilter: Make ct zone opts configurable for bpf ct helpers

Add ct zone id and direction to bpf_ct_opts so that arbitrary ct zones
can be used for xdp/tc bpf ct helper functions bpf_{xdp,skb}_ct_alloc
and bpf_{xdp,skb}_ct_lookup.

Signed-off-by: Brad Cowie <brad@faucet.nz>
Link: https://lore.kernel.org/r/20240522050712.732558-1-brad@faucet.nz
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 net/netfilter/nf_conntrack_bpf.c | 68 ++++++++++++++++++++++++++++++++--------
 1 file changed, 55 insertions(+), 13 deletions(-)

diff --git a/net/netfilter/nf_conntrack_bpf.c b/net/netfilter/nf_conntrack_bpf.c
index d2492d050fe6..4a136fc3a9c0 100644
--- a/net/netfilter/nf_conntrack_bpf.c
+++ b/net/netfilter/nf_conntrack_bpf.c
@@ -32,7 +32,9 @@
  *		   -EINVAL - Passed NULL for bpf_tuple pointer
  *		   -EINVAL - opts->reserved is not 0
  *		   -EINVAL - netns_id is less than -1
- *		   -EINVAL - opts__sz isn't NF_BPF_CT_OPTS_SZ (12)
+ *		   -EINVAL - opts__sz isn't NF_BPF_CT_OPTS_SZ (16) or 12
+ *		   -EINVAL - opts->ct_zone_id set when
+			     opts__sz isn't NF_BPF_CT_OPTS_SZ (16)
  *		   -EPROTO - l4proto isn't one of IPPROTO_TCP or IPPROTO_UDP
  *		   -ENONET - No network namespace found for netns_id
  *		   -ENOENT - Conntrack lookup could not find entry for tuple
@@ -42,6 +44,8 @@
  *		 Values:
  *		   IPPROTO_TCP, IPPROTO_UDP
  * @dir:       - connection tracking tuple direction.
+ * @ct_zone_id - connection tracking zone id.
+ * @ct_zone_dir - connection tracking zone direction.
  * @reserved   - Reserved member, will be reused for more options in future
  *		 Values:
  *		   0
@@ -51,11 +55,13 @@ struct bpf_ct_opts {
 	s32 error;
 	u8 l4proto;
 	u8 dir;
-	u8 reserved[2];
+	u16 ct_zone_id;
+	u8 ct_zone_dir;
+	u8 reserved[3];
 };
 
 enum {
-	NF_BPF_CT_OPTS_SZ = 12,
+	NF_BPF_CT_OPTS_SZ = 16,
 };
 
 static int bpf_nf_ct_tuple_parse(struct bpf_sock_tuple *bpf_tuple,
@@ -104,12 +110,21 @@ __bpf_nf_ct_alloc_entry(struct net *net, struct bpf_sock_tuple *bpf_tuple,
 			u32 timeout)
 {
 	struct nf_conntrack_tuple otuple, rtuple;
+	struct nf_conntrack_zone ct_zone;
 	struct nf_conn *ct;
 	int err;
 
-	if (!opts || !bpf_tuple || opts->reserved[0] || opts->reserved[1] ||
-	    opts_len != NF_BPF_CT_OPTS_SZ)
+	if (!opts || !bpf_tuple)
 		return ERR_PTR(-EINVAL);
+	if (!(opts_len == NF_BPF_CT_OPTS_SZ || opts_len == 12))
+		return ERR_PTR(-EINVAL);
+	if (opts_len == NF_BPF_CT_OPTS_SZ) {
+		if (opts->reserved[0] || opts->reserved[1] || opts->reserved[2])
+			return ERR_PTR(-EINVAL);
+	} else {
+		if (opts->ct_zone_id)
+			return ERR_PTR(-EINVAL);
+	}
 
 	if (unlikely(opts->netns_id < BPF_F_CURRENT_NETNS))
 		return ERR_PTR(-EINVAL);
@@ -130,7 +145,16 @@ __bpf_nf_ct_alloc_entry(struct net *net, struct bpf_sock_tuple *bpf_tuple,
 			return ERR_PTR(-ENONET);
 	}
 
-	ct = nf_conntrack_alloc(net, &nf_ct_zone_dflt, &otuple, &rtuple,
+	if (opts_len == NF_BPF_CT_OPTS_SZ) {
+		if (opts->ct_zone_dir == 0)
+			opts->ct_zone_dir = NF_CT_DEFAULT_ZONE_DIR;
+		nf_ct_zone_init(&ct_zone,
+				opts->ct_zone_id, opts->ct_zone_dir, 0);
+	} else {
+		ct_zone = nf_ct_zone_dflt;
+	}
+
+	ct = nf_conntrack_alloc(net, &ct_zone, &otuple, &rtuple,
 				GFP_ATOMIC);
 	if (IS_ERR(ct))
 		goto out;
@@ -152,12 +176,21 @@ static struct nf_conn *__bpf_nf_ct_lookup(struct net *net,
 {
 	struct nf_conntrack_tuple_hash *hash;
 	struct nf_conntrack_tuple tuple;
+	struct nf_conntrack_zone ct_zone;
 	struct nf_conn *ct;
 	int err;
 
-	if (!opts || !bpf_tuple || opts->reserved[0] || opts->reserved[1] ||
-	    opts_len != NF_BPF_CT_OPTS_SZ)
+	if (!opts || !bpf_tuple)
 		return ERR_PTR(-EINVAL);
+	if (!(opts_len == NF_BPF_CT_OPTS_SZ || opts_len == 12))
+		return ERR_PTR(-EINVAL);
+	if (opts_len == NF_BPF_CT_OPTS_SZ) {
+		if (opts->reserved[0] || opts->reserved[1] || opts->reserved[2])
+			return ERR_PTR(-EINVAL);
+	} else {
+		if (opts->ct_zone_id)
+			return ERR_PTR(-EINVAL);
+	}
 	if (unlikely(opts->l4proto != IPPROTO_TCP && opts->l4proto != IPPROTO_UDP))
 		return ERR_PTR(-EPROTO);
 	if (unlikely(opts->netns_id < BPF_F_CURRENT_NETNS))
@@ -174,7 +207,16 @@ static struct nf_conn *__bpf_nf_ct_lookup(struct net *net,
 			return ERR_PTR(-ENONET);
 	}
 
-	hash = nf_conntrack_find_get(net, &nf_ct_zone_dflt, &tuple);
+	if (opts_len == NF_BPF_CT_OPTS_SZ) {
+		if (opts->ct_zone_dir == 0)
+			opts->ct_zone_dir = NF_CT_DEFAULT_ZONE_DIR;
+		nf_ct_zone_init(&ct_zone,
+				opts->ct_zone_id, opts->ct_zone_dir, 0);
+	} else {
+		ct_zone = nf_ct_zone_dflt;
+	}
+
+	hash = nf_conntrack_find_get(net, &ct_zone, &tuple);
 	if (opts->netns_id >= 0)
 		put_net(net);
 	if (!hash)
@@ -245,7 +287,7 @@ __bpf_kfunc_start_defs();
  * @opts	- Additional options for allocation (documented above)
  *		    Cannot be NULL
  * @opts__sz	- Length of the bpf_ct_opts structure
- *		    Must be NF_BPF_CT_OPTS_SZ (12)
+ *		    Must be NF_BPF_CT_OPTS_SZ (16) or 12
  */
 __bpf_kfunc struct nf_conn___init *
 bpf_xdp_ct_alloc(struct xdp_md *xdp_ctx, struct bpf_sock_tuple *bpf_tuple,
@@ -279,7 +321,7 @@ bpf_xdp_ct_alloc(struct xdp_md *xdp_ctx, struct bpf_sock_tuple *bpf_tuple,
  * @opts	- Additional options for lookup (documented above)
  *		    Cannot be NULL
  * @opts__sz	- Length of the bpf_ct_opts structure
- *		    Must be NF_BPF_CT_OPTS_SZ (12)
+ *		    Must be NF_BPF_CT_OPTS_SZ (16) or 12
  */
 __bpf_kfunc struct nf_conn *
 bpf_xdp_ct_lookup(struct xdp_md *xdp_ctx, struct bpf_sock_tuple *bpf_tuple,
@@ -312,7 +354,7 @@ bpf_xdp_ct_lookup(struct xdp_md *xdp_ctx, struct bpf_sock_tuple *bpf_tuple,
  * @opts	- Additional options for allocation (documented above)
  *		    Cannot be NULL
  * @opts__sz	- Length of the bpf_ct_opts structure
- *		    Must be NF_BPF_CT_OPTS_SZ (12)
+ *		    Must be NF_BPF_CT_OPTS_SZ (16) or 12
  */
 __bpf_kfunc struct nf_conn___init *
 bpf_skb_ct_alloc(struct __sk_buff *skb_ctx, struct bpf_sock_tuple *bpf_tuple,
@@ -347,7 +389,7 @@ bpf_skb_ct_alloc(struct __sk_buff *skb_ctx, struct bpf_sock_tuple *bpf_tuple,
  * @opts	- Additional options for lookup (documented above)
  *		    Cannot be NULL
  * @opts__sz	- Length of the bpf_ct_opts structure
- *		    Must be NF_BPF_CT_OPTS_SZ (12)
+ *		    Must be NF_BPF_CT_OPTS_SZ (16) or 12
  */
 __bpf_kfunc struct nf_conn *
 bpf_skb_ct_lookup(struct __sk_buff *skb_ctx, struct bpf_sock_tuple *bpf_tuple,
-- 
cgit v1.2.3-58-ga151


From a87f34e742d279d54d529e4bc4763fdaab32a466 Mon Sep 17 00:00:00 2001
From: Brad Cowie <brad@faucet.nz>
Date: Wed, 22 May 2024 17:07:12 +1200
Subject: selftests/bpf: Update tests for new ct zone opts for nf_conntrack
 kfuncs

Add test for allocating and looking up ct entry in a
non-default ct zone with kfuncs bpf_{xdp,skb}_ct_alloc
and bpf_{xdp,skb}_ct_lookup.

Add negative tests for looking up ct entry in a different
ct zone to where it was allocated and with a different
direction.

Update reserved test for old struct definition to test for
ct_zone_id being set when opts size isn't NF_BPF_CT_OPTS_SZ (16).

Signed-off-by: Brad Cowie <brad@faucet.nz>
Link: https://lore.kernel.org/r/20240522050712.732558-2-brad@faucet.nz
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/config              |   1 +
 tools/testing/selftests/bpf/prog_tests/bpf_nf.c |   7 ++
 tools/testing/selftests/bpf/progs/test_bpf_nf.c | 108 ++++++++++++++++++++++++
 3 files changed, 116 insertions(+)

diff --git a/tools/testing/selftests/bpf/config b/tools/testing/selftests/bpf/config
index eeabd798bc3a..2fb16da78dce 100644
--- a/tools/testing/selftests/bpf/config
+++ b/tools/testing/selftests/bpf/config
@@ -80,6 +80,7 @@ CONFIG_NETFILTER_XT_TARGET_CT=y
 CONFIG_NETKIT=y
 CONFIG_NF_CONNTRACK=y
 CONFIG_NF_CONNTRACK_MARK=y
+CONFIG_NF_CONNTRACK_ZONES=y
 CONFIG_NF_DEFRAG_IPV4=y
 CONFIG_NF_DEFRAG_IPV6=y
 CONFIG_NF_NAT=y
diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_nf.c b/tools/testing/selftests/bpf/prog_tests/bpf_nf.c
index b30ff6b3b81a..a4a1f93878d4 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_nf.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_nf.c
@@ -104,6 +104,7 @@ static void test_bpf_nf_ct(int mode)
 
 	ASSERT_EQ(skel->bss->test_einval_bpf_tuple, -EINVAL, "Test EINVAL for NULL bpf_tuple");
 	ASSERT_EQ(skel->bss->test_einval_reserved, -EINVAL, "Test EINVAL for reserved not set to 0");
+	ASSERT_EQ(skel->bss->test_einval_reserved_new, -EINVAL, "Test EINVAL for reserved in new struct not set to 0");
 	ASSERT_EQ(skel->bss->test_einval_netns_id, -EINVAL, "Test EINVAL for netns_id < -1");
 	ASSERT_EQ(skel->bss->test_einval_len_opts, -EINVAL, "Test EINVAL for len__opts != NF_BPF_CT_OPTS_SZ");
 	ASSERT_EQ(skel->bss->test_eproto_l4proto, -EPROTO, "Test EPROTO for l4proto != TCP or UDP");
@@ -122,6 +123,12 @@ static void test_bpf_nf_ct(int mode)
 	ASSERT_EQ(skel->bss->test_exist_lookup_mark, 43, "Test existing connection lookup ctmark");
 	ASSERT_EQ(skel->data->test_snat_addr, 0, "Test for source natting");
 	ASSERT_EQ(skel->data->test_dnat_addr, 0, "Test for destination natting");
+	ASSERT_EQ(skel->data->test_ct_zone_id_alloc_entry, 0, "Test for alloc new entry in specified ct zone");
+	ASSERT_EQ(skel->data->test_ct_zone_id_insert_entry, 0, "Test for insert new entry in specified ct zone");
+	ASSERT_EQ(skel->data->test_ct_zone_id_succ_lookup, 0, "Test for successful lookup in specified ct_zone");
+	ASSERT_EQ(skel->bss->test_ct_zone_dir_enoent_lookup, -ENOENT, "Test ENOENT for lookup with wrong ct zone dir");
+	ASSERT_EQ(skel->bss->test_ct_zone_id_enoent_lookup, -ENOENT, "Test ENOENT for lookup in wrong ct zone");
+
 end:
 	if (client_fd != -1)
 		close(client_fd);
diff --git a/tools/testing/selftests/bpf/progs/test_bpf_nf.c b/tools/testing/selftests/bpf/progs/test_bpf_nf.c
index 77ad8adf68da..0289d8ce2b80 100644
--- a/tools/testing/selftests/bpf/progs/test_bpf_nf.c
+++ b/tools/testing/selftests/bpf/progs/test_bpf_nf.c
@@ -9,10 +9,14 @@
 #define EINVAL 22
 #define ENOENT 2
 
+#define NF_CT_ZONE_DIR_ORIG (1 << IP_CT_DIR_ORIGINAL)
+#define NF_CT_ZONE_DIR_REPL (1 << IP_CT_DIR_REPLY)
+
 extern unsigned long CONFIG_HZ __kconfig;
 
 int test_einval_bpf_tuple = 0;
 int test_einval_reserved = 0;
+int test_einval_reserved_new = 0;
 int test_einval_netns_id = 0;
 int test_einval_len_opts = 0;
 int test_eproto_l4proto = 0;
@@ -22,6 +26,11 @@ int test_eafnosupport = 0;
 int test_alloc_entry = -EINVAL;
 int test_insert_entry = -EAFNOSUPPORT;
 int test_succ_lookup = -ENOENT;
+int test_ct_zone_id_alloc_entry = -EINVAL;
+int test_ct_zone_id_insert_entry = -EAFNOSUPPORT;
+int test_ct_zone_id_succ_lookup = -ENOENT;
+int test_ct_zone_dir_enoent_lookup = 0;
+int test_ct_zone_id_enoent_lookup = 0;
 u32 test_delta_timeout = 0;
 u32 test_status = 0;
 u32 test_insert_lookup_mark = 0;
@@ -45,6 +54,17 @@ struct bpf_ct_opts___local {
 	s32 netns_id;
 	s32 error;
 	u8 l4proto;
+	u8 dir;
+	u8 reserved[2];
+};
+
+struct bpf_ct_opts___new {
+	s32 netns_id;
+	s32 error;
+	u8 l4proto;
+	u8 dir;
+	u16 ct_zone_id;
+	u8 ct_zone_dir;
 	u8 reserved[3];
 } __attribute__((preserve_access_index));
 
@@ -220,10 +240,97 @@ nf_ct_test(struct nf_conn *(*lookup_fn)(void *, struct bpf_sock_tuple *, u32,
 	}
 }
 
+static __always_inline void
+nf_ct_opts_new_test(struct nf_conn *(*lookup_fn)(void *, struct bpf_sock_tuple *, u32,
+						 struct bpf_ct_opts___new *, u32),
+		    struct nf_conn *(*alloc_fn)(void *, struct bpf_sock_tuple *, u32,
+						struct bpf_ct_opts___new *, u32),
+		    void *ctx)
+{
+	struct bpf_ct_opts___new opts_def = { .l4proto = IPPROTO_TCP, .netns_id = -1 };
+	struct bpf_sock_tuple bpf_tuple;
+	struct nf_conn *ct;
+
+	__builtin_memset(&bpf_tuple, 0, sizeof(bpf_tuple.ipv4));
+
+	opts_def.reserved[0] = 1;
+	ct = lookup_fn(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def,
+		       sizeof(opts_def));
+	opts_def.reserved[0] = 0;
+	if (ct)
+		bpf_ct_release(ct);
+	else
+		test_einval_reserved_new = opts_def.error;
+
+	bpf_tuple.ipv4.saddr = bpf_get_prandom_u32(); /* src IP */
+	bpf_tuple.ipv4.daddr = bpf_get_prandom_u32(); /* dst IP */
+	bpf_tuple.ipv4.sport = bpf_get_prandom_u32(); /* src port */
+	bpf_tuple.ipv4.dport = bpf_get_prandom_u32(); /* dst port */
+
+	/* use non-default ct zone */
+	opts_def.ct_zone_id = 10;
+	opts_def.ct_zone_dir = NF_CT_ZONE_DIR_ORIG;
+	ct = alloc_fn(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4), &opts_def,
+		      sizeof(opts_def));
+	if (ct) {
+		__u16 sport = bpf_get_prandom_u32();
+		__u16 dport = bpf_get_prandom_u32();
+		union nf_inet_addr saddr = {};
+		union nf_inet_addr daddr = {};
+		struct nf_conn *ct_ins;
+
+		bpf_ct_set_timeout(ct, 10000);
+
+		/* snat */
+		saddr.ip = bpf_get_prandom_u32();
+		bpf_ct_set_nat_info(ct, &saddr, sport, NF_NAT_MANIP_SRC___local);
+		/* dnat */
+		daddr.ip = bpf_get_prandom_u32();
+		bpf_ct_set_nat_info(ct, &daddr, dport, NF_NAT_MANIP_DST___local);
+
+		ct_ins = bpf_ct_insert_entry(ct);
+		if (ct_ins) {
+			struct nf_conn *ct_lk;
+
+			/* entry should exist in same ct zone we inserted it */
+			ct_lk = lookup_fn(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4),
+					  &opts_def, sizeof(opts_def));
+			if (ct_lk) {
+				bpf_ct_release(ct_lk);
+				test_ct_zone_id_succ_lookup = 0;
+			}
+
+			/* entry should not exist with wrong direction */
+			opts_def.ct_zone_dir = NF_CT_ZONE_DIR_REPL;
+			ct_lk = lookup_fn(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4),
+					  &opts_def, sizeof(opts_def));
+			opts_def.ct_zone_dir = NF_CT_ZONE_DIR_ORIG;
+			if (ct_lk)
+				bpf_ct_release(ct_lk);
+			else
+				test_ct_zone_dir_enoent_lookup = opts_def.error;
+
+			/* entry should not exist in default ct zone */
+			opts_def.ct_zone_id = 0;
+			ct_lk = lookup_fn(ctx, &bpf_tuple, sizeof(bpf_tuple.ipv4),
+					  &opts_def, sizeof(opts_def));
+			if (ct_lk)
+				bpf_ct_release(ct_lk);
+			else
+				test_ct_zone_id_enoent_lookup = opts_def.error;
+
+			bpf_ct_release(ct_ins);
+			test_ct_zone_id_insert_entry = 0;
+		}
+		test_ct_zone_id_alloc_entry = 0;
+	}
+}
+
 SEC("xdp")
 int nf_xdp_ct_test(struct xdp_md *ctx)
 {
 	nf_ct_test((void *)bpf_xdp_ct_lookup, (void *)bpf_xdp_ct_alloc, ctx);
+	nf_ct_opts_new_test((void *)bpf_xdp_ct_lookup, (void *)bpf_xdp_ct_alloc, ctx);
 	return 0;
 }
 
@@ -231,6 +338,7 @@ SEC("tc")
 int nf_skb_ct_test(struct __sk_buff *ctx)
 {
 	nf_ct_test((void *)bpf_skb_ct_lookup, (void *)bpf_skb_ct_alloc, ctx);
+	nf_ct_opts_new_test((void *)bpf_skb_ct_lookup, (void *)bpf_skb_ct_alloc, ctx);
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From 669b692247d4516e252c898c0e7366a09d84d1be Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Fri, 17 May 2024 09:33:50 +0800
Subject: wifi: rtw89: fix HW scan not aborting properly

There is a length limit on the commands we send to firmware, so
dividing to two commands is sometimes required when scanning.
When aborting scan, we should not send second scan command to
firmware after the first one is finished. This could cause some
unexpected errors when we cannot receive firmware events
(e.g. in suspend).

Another case is scan happens before suspending, ieee80211_do_stop() is
called to abort scan and driver indicate scan completion by
ieee80211_scan_completed(), which queues event to scan work. But scan work
might be late to execute after ieee80211_do_stop(). To correct this, driver
indicates ieee80211_scan_completed() before returning, so that
ieee80211_do_stop() can flush scan work properly.

Fixes: bcbefbd032df ("wifi: rtw89: add wait/completion for abort scan")
Cc: stable@vger.kernel.org
Co-developed-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240517013350.11278-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c  | 9 ++++++++-
 drivers/net/wireless/realtek/rtw89/mac.c | 5 ++++-
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 044a5b90c7f4..f5cd06c35982 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -6245,7 +6245,14 @@ void rtw89_hw_scan_abort(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif)
 
 	ret = rtw89_hw_scan_offload(rtwdev, vif, false);
 	if (ret)
-		rtw89_hw_scan_complete(rtwdev, vif, true);
+		rtw89_warn(rtwdev, "rtw89_hw_scan_offload failed ret %d\n", ret);
+
+	/* Indicate ieee80211_scan_completed() before returning, which is safe
+	 * because scan abort command always waits for completion of
+	 * RTW89_SCAN_END_SCAN_NOTIFY, so that ieee80211_stop() can flush scan
+	 * work properly.
+	 */
+	rtw89_hw_scan_complete(rtwdev, vif, true);
 }
 
 static bool rtw89_is_any_vif_connected_or_connecting(struct rtw89_dev *rtwdev)
diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index baefc30b88c4..824ece03d92d 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -4756,6 +4756,9 @@ rtw89_mac_c2h_scanofld_rsp(struct rtw89_dev *rtwdev, struct sk_buff *skb,
 		}
 		return;
 	case RTW89_SCAN_END_SCAN_NOTIFY:
+		if (rtwdev->scan_info.abort)
+			return;
+
 		if (rtwvif && rtwvif->scan_req &&
 		    last_chan < rtwvif->scan_req->n_channels) {
 			ret = rtw89_hw_scan_offload(rtwdev, vif, true);
@@ -4764,7 +4767,7 @@ rtw89_mac_c2h_scanofld_rsp(struct rtw89_dev *rtwdev, struct sk_buff *skb,
 				rtw89_warn(rtwdev, "HW scan failed: %d\n", ret);
 			}
 		} else {
-			rtw89_hw_scan_complete(rtwdev, vif, rtwdev->scan_info.abort);
+			rtw89_hw_scan_complete(rtwdev, vif, false);
 		}
 		break;
 	case RTW89_SCAN_ENTER_OP_NOTIFY:
-- 
cgit v1.2.3-58-ga151


From 2c1fc7c24cd47396580c5a7b238673da618aeedd Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 17 May 2024 09:35:43 +0800
Subject: wifi: rtw89: 8852b: restore setting for RFE type 5 after device
 resume

The RFE type 5 set SPS analog parameters only once at probe stage, but the
setting is missing after suspend/resume, so remove restriction and set the
value when card power on/off.

Fixes: 3ef60f44830a ("wifi: rtw89: 8852b: update hardware parameters for RFE type 5")
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240517013543.11533-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852b.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index 53ae00b5ab7c..517ad1a763fa 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -403,6 +403,8 @@ static int rtw8852b_pwr_on_func(struct rtw89_dev *rtwdev)
 	u32 val32;
 	u32 ret;
 
+	rtw8852b_pwr_sps_ana(rtwdev);
+
 	rtw89_write32_clr(rtwdev, R_AX_SYS_PW_CTRL, B_AX_AFSM_WLSUS_EN |
 						    B_AX_AFSM_PCIE_SUS_EN);
 	rtw89_write32_set(rtwdev, R_AX_SYS_PW_CTRL, B_AX_DIS_WLBT_PDNSUSEN_SOPC);
@@ -530,9 +532,7 @@ static int rtw8852b_pwr_off_func(struct rtw89_dev *rtwdev)
 	u32 val32;
 	u32 ret;
 
-	/* Only do once during probe stage after reading efuse */
-	if (!test_bit(RTW89_FLAG_PROBE_DONE, rtwdev->flags))
-		rtw8852b_pwr_sps_ana(rtwdev);
+	rtw8852b_pwr_sps_ana(rtwdev);
 
 	ret = rtw89_mac_write_xtal_si(rtwdev, XTAL_SI_ANAPAR_WL, XTAL_SI_RFC2RF,
 				      XTAL_SI_RFC2RF);
-- 
cgit v1.2.3-58-ga151


From d5b96a4a31593a609acd564cfa54f269461a33f0 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 17 May 2024 09:36:39 +0800
Subject: wifi: rtw89: correct hardware value of nominal packet padding for
 WiFi 7 chips

The hardware value of 3 for WiFi 7 chips is padding 16 or 20us selected
by hardware automatically. Correct the value to resolve interoperability
problem. Also define enum to be more readable.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240517013639.11643-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.h |  7 +++++++
 drivers/net/wireless/realtek/rtw89/fw.c   | 12 ++++++------
 2 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index c0f2afb4888a..523ea9f5e6b1 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -884,6 +884,13 @@ enum rtw89_ps_mode {
 #define RTW89_BYR_BW_NUM (RTW89_CHANNEL_WIDTH_320 + 1)
 #define RTW89_PPE_BW_NUM (RTW89_CHANNEL_WIDTH_320 + 1)
 
+enum rtw89_pe_duration {
+	RTW89_PE_DURATION_0 = 0,
+	RTW89_PE_DURATION_8 = 1,
+	RTW89_PE_DURATION_16 = 2,
+	RTW89_PE_DURATION_16_20 = 3,
+};
+
 enum rtw89_ru_bandwidth {
 	RTW89_RU26 = 0,
 	RTW89_RU52 = 1,
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index f5cd06c35982..1ea2b662fc89 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -2752,11 +2752,11 @@ static void __get_sta_he_pkt_padding(struct rtw89_dev *rtwdev,
 		ppe8 = (ppe >> sh) & IEEE80211_PPE_THRES_NSS_MASK;
 
 		if (ppe16 != 7 && ppe8 == 7)
-			pads[i] = 2;
+			pads[i] = RTW89_PE_DURATION_16;
 		else if (ppe8 != 7)
-			pads[i] = 1;
+			pads[i] = RTW89_PE_DURATION_8;
 		else
-			pads[i] = 0;
+			pads[i] = RTW89_PE_DURATION_0;
 	}
 }
 
@@ -2889,11 +2889,11 @@ static void __get_sta_eht_pkt_padding(struct rtw89_dev *rtwdev,
 		ppe8 = (ppe >> sh) & IEEE80211_PPE_THRES_NSS_MASK;
 
 		if (ppe16 != 7 && ppe8 == 7)
-			pads[i] = 2;
+			pads[i] = RTW89_PE_DURATION_16_20;
 		else if (ppe8 != 7)
-			pads[i] = 1;
+			pads[i] = RTW89_PE_DURATION_8;
 		else
-			pads[i] = 0;
+			pads[i] = RTW89_PE_DURATION_0;
 	}
 }
 
-- 
cgit v1.2.3-58-ga151


From 96f887a612e4cda89efc3f54bc10c1997e3ab0e9 Mon Sep 17 00:00:00 2001
From: Hagar Hemdan <hagarhem@amazon.com>
Date: Sat, 18 May 2024 13:04:39 +0000
Subject: net: esp: cleanup esp_output_tail_tcp() in case of unsupported
 ESPINTCP

xmit() functions should consume skb or return error codes in error
paths.
When the configuration "CONFIG_INET_ESPINTCP" is not set, the
implementation of the function "esp_output_tail_tcp" violates this rule.
The function frees the skb and returns the error code.
This change removes the kfree_skb from both functions, for both
esp4 and esp6.
WARN_ON is added because esp_output_tail_tcp() should never be called if
CONFIG_INET_ESPINTCP is not set.

This bug was discovered and resolved using Coverity Static Analysis
Security Testing (SAST) by Synopsys, Inc.

Fixes: e27cca96cd68 ("xfrm: add espintcp (RFC 8229)")
Signed-off-by: Hagar Hemdan <hagarhem@amazon.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/ipv4/esp4.c | 3 +--
 net/ipv6/esp6.c | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index 3968d3f98e08..619a4df7be1e 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -239,8 +239,7 @@ static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb)
 #else
 static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb)
 {
-	kfree_skb(skb);
-
+	WARN_ON(1);
 	return -EOPNOTSUPP;
 }
 #endif
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index 34a9a5b9ed00..3920e8aa1031 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -256,8 +256,7 @@ static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb)
 #else
 static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb)
 {
-	kfree_skb(skb);
-
+	WARN_ON(1);
 	return -EOPNOTSUPP;
 }
 #endif
-- 
cgit v1.2.3-58-ga151


From 5716146295183651cfd89ae2ea46c5ef0d31faa8 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 15 May 2024 09:38:35 +0200
Subject: wifi: regulatory: remove extra documentation

The struct member country_ie_checksum doesn't exist, so
don't document it.

Link: https://msgid.link/20240515093852.ebcc9673558b.Ie0b58c1249c6375c60859fa6474d7cdd8862b065@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/regulatory.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/include/net/regulatory.h b/include/net/regulatory.h
index ebf9e028d1ef..a103f4c8cf75 100644
--- a/include/net/regulatory.h
+++ b/include/net/regulatory.h
@@ -71,8 +71,6 @@ enum environment_cap {
  *	CRDA and can be used by other regulatory requests. When a
  *	the last request is not yet processed we must yield until it
  *	is processed before processing any new requests.
- * @country_ie_checksum: checksum of the last processed and accepted
- *	country IE
  * @country_ie_env: lets us know if the AP is telling us we are outdoor,
  *	indoor, or if it doesn't matter
  * @list: used to insert into the reg_requests_list linked list
-- 
cgit v1.2.3-58-ga151


From 3bb8dce41c023e39ec4d79a83742403b5064a276 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 15 May 2024 09:38:36 +0200
Subject: wifi: ieee80211: add missing doc short descriptions

Some structures erroneously don't have a short description,
add the missing descriptions.

Link: https://msgid.link/20240515093852.16f4355e918e.I940276a4fb006ada68ab1a3e6077e3229fff0f14@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/linux/ieee80211.h | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index de2dce743ee2..713266ce48a7 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -1101,7 +1101,7 @@ enum ieee80211_vht_opmode_bits {
 };
 
 /**
- * enum ieee80211_s1g_chanwidth
+ * enum ieee80211_s1g_chanwidth - S1G channel widths
  * These are defined in IEEE802.11-2016ah Table 10-20
  * as BSS Channel Width
  *
@@ -4145,7 +4145,7 @@ enum ieee80211_idle_options {
 };
 
 /**
- * struct ieee80211_bss_max_idle_period_ie
+ * struct ieee80211_bss_max_idle_period_ie - BSS max idle period element struct
  *
  * This structure refers to "BSS Max idle period element"
  *
@@ -4180,7 +4180,7 @@ enum ieee80211_sa_query_action {
 };
 
 /**
- * struct ieee80211_bssid_index
+ * struct ieee80211_bssid_index - multiple BSSID index element structure
  *
  * This structure refers to "Multiple BSSID-index element"
  *
@@ -4195,7 +4195,8 @@ struct ieee80211_bssid_index {
 };
 
 /**
- * struct ieee80211_multiple_bssid_configuration
+ * struct ieee80211_multiple_bssid_configuration - multiple BSSID configuration
+ *	element structure
  *
  * This structure refers to "Multiple BSSID Configuration element"
  *
-- 
cgit v1.2.3-58-ga151


From 1c26f09b20bafc92e0de80e84942be77d4d3c61e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 15 May 2024 09:38:37 +0200
Subject: wifi: radiotap: document ieee80211_get_radiotap_len() return value

Document the return value of ieee80211_get_radiotap_len() in
the proper kernel-doc format.

Link: https://msgid.link/20240515093852.143aadfdb094.I8795ec1e8cfd7106d58325fb514bae92625fb45c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/ieee80211_radiotap.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/net/ieee80211_radiotap.h b/include/net/ieee80211_radiotap.h
index 925bac726a92..91762faecc13 100644
--- a/include/net/ieee80211_radiotap.h
+++ b/include/net/ieee80211_radiotap.h
@@ -582,6 +582,7 @@ enum ieee80211_radiotap_eht_usig_tb {
 /**
  * ieee80211_get_radiotap_len - get radiotap header length
  * @data: pointer to the header
+ * Return: the radiotap header length
  */
 static inline u16 ieee80211_get_radiotap_len(const char *data)
 {
-- 
cgit v1.2.3-58-ga151


From 54856871298c9a8717450351699a8e6fe706101c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 15 May 2024 09:38:38 +0200
Subject: wifi: ieee80211: remove ieee80211_next_tbtt_present()

This is actually completely equivalent to the other function
ieee80211_is_s1g_short_beacon(), but open-codes the logic.
Implement the necessary logic in ieee80211_is_s1g_short_beacon()
and remove ieee80211_next_tbtt_present().

Link: https://msgid.link/20240515093852.774ced74dea8.I152525b4cff6e6a25be6c48fe6a4b89f17bab8a9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/linux/ieee80211.h | 20 +++++---------------
 1 file changed, 5 insertions(+), 15 deletions(-)

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 713266ce48a7..72b351abb4f6 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -604,25 +604,15 @@ static inline bool ieee80211_is_s1g_beacon(__le16 fc)
 }
 
 /**
- * ieee80211_next_tbtt_present - check if IEEE80211_FTYPE_EXT &&
- * IEEE80211_STYPE_S1G_BEACON && IEEE80211_S1G_BCN_NEXT_TBTT
- * @fc: frame control bytes in little-endian byteorder
- */
-static inline bool ieee80211_next_tbtt_present(__le16 fc)
-{
-	return (fc & cpu_to_le16(IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE)) ==
-	       cpu_to_le16(IEEE80211_FTYPE_EXT | IEEE80211_STYPE_S1G_BEACON) &&
-	       fc & cpu_to_le16(IEEE80211_S1G_BCN_NEXT_TBTT);
-}
-
-/**
- * ieee80211_is_s1g_short_beacon - check if next tbtt present bit is set. Only
- * true for S1G beacons when they're short.
+ * ieee80211_is_s1g_short_beacon - check if frame is an S1G short beacon
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is an S1G short beacon,
+ *	i.e. it is an S1G beacon with 'next TBTT' flag set
  */
 static inline bool ieee80211_is_s1g_short_beacon(__le16 fc)
 {
-	return ieee80211_is_s1g_beacon(fc) && ieee80211_next_tbtt_present(fc);
+	return ieee80211_is_s1g_beacon(fc) &&
+		(fc & cpu_to_le16(IEEE80211_S1G_BCN_NEXT_TBTT));
 }
 
 /**
-- 
cgit v1.2.3-58-ga151


From 3c75e99c7036a87f159dbd526060554afb3482f5 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 15 May 2024 09:38:39 +0200
Subject: wifi: ieee80211: document function return values

These are all missing, as pointed out when running kernel-doc.
Add return value documentation and fix some small things while
at it.

Link: https://msgid.link/20240515093852.1cd5ad8f354d.Idc16e9767fa42de80b659c32efc58aea38c26996@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/linux/ieee80211.h | 106 ++++++++++++++++++++++++++++++++++++----------
 1 file changed, 84 insertions(+), 22 deletions(-)

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 72b351abb4f6..595f83783f0e 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -373,6 +373,7 @@ struct ieee80211_trigger {
 /**
  * ieee80211_has_tods - check if IEEE80211_FCTL_TODS is set
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame has to-DS set
  */
 static inline bool ieee80211_has_tods(__le16 fc)
 {
@@ -382,6 +383,7 @@ static inline bool ieee80211_has_tods(__le16 fc)
 /**
  * ieee80211_has_fromds - check if IEEE80211_FCTL_FROMDS is set
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame has from-DS set
  */
 static inline bool ieee80211_has_fromds(__le16 fc)
 {
@@ -391,6 +393,7 @@ static inline bool ieee80211_has_fromds(__le16 fc)
 /**
  * ieee80211_has_a4 - check if IEEE80211_FCTL_TODS and IEEE80211_FCTL_FROMDS are set
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not it's a 4-address frame (from-DS and to-DS set)
  */
 static inline bool ieee80211_has_a4(__le16 fc)
 {
@@ -401,6 +404,7 @@ static inline bool ieee80211_has_a4(__le16 fc)
 /**
  * ieee80211_has_morefrags - check if IEEE80211_FCTL_MOREFRAGS is set
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame has more fragments (more frags bit set)
  */
 static inline bool ieee80211_has_morefrags(__le16 fc)
 {
@@ -410,6 +414,7 @@ static inline bool ieee80211_has_morefrags(__le16 fc)
 /**
  * ieee80211_has_retry - check if IEEE80211_FCTL_RETRY is set
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the retry flag is set
  */
 static inline bool ieee80211_has_retry(__le16 fc)
 {
@@ -419,6 +424,7 @@ static inline bool ieee80211_has_retry(__le16 fc)
 /**
  * ieee80211_has_pm - check if IEEE80211_FCTL_PM is set
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the power management flag is set
  */
 static inline bool ieee80211_has_pm(__le16 fc)
 {
@@ -428,6 +434,7 @@ static inline bool ieee80211_has_pm(__le16 fc)
 /**
  * ieee80211_has_moredata - check if IEEE80211_FCTL_MOREDATA is set
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the more data flag is set
  */
 static inline bool ieee80211_has_moredata(__le16 fc)
 {
@@ -437,6 +444,7 @@ static inline bool ieee80211_has_moredata(__le16 fc)
 /**
  * ieee80211_has_protected - check if IEEE80211_FCTL_PROTECTED is set
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the protected flag is set
  */
 static inline bool ieee80211_has_protected(__le16 fc)
 {
@@ -446,6 +454,7 @@ static inline bool ieee80211_has_protected(__le16 fc)
 /**
  * ieee80211_has_order - check if IEEE80211_FCTL_ORDER is set
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the order flag is set
  */
 static inline bool ieee80211_has_order(__le16 fc)
 {
@@ -455,6 +464,7 @@ static inline bool ieee80211_has_order(__le16 fc)
 /**
  * ieee80211_is_mgmt - check if type is IEEE80211_FTYPE_MGMT
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame type is management
  */
 static inline bool ieee80211_is_mgmt(__le16 fc)
 {
@@ -465,6 +475,7 @@ static inline bool ieee80211_is_mgmt(__le16 fc)
 /**
  * ieee80211_is_ctl - check if type is IEEE80211_FTYPE_CTL
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame type is control
  */
 static inline bool ieee80211_is_ctl(__le16 fc)
 {
@@ -475,6 +486,7 @@ static inline bool ieee80211_is_ctl(__le16 fc)
 /**
  * ieee80211_is_data - check if type is IEEE80211_FTYPE_DATA
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a data frame
  */
 static inline bool ieee80211_is_data(__le16 fc)
 {
@@ -485,6 +497,7 @@ static inline bool ieee80211_is_data(__le16 fc)
 /**
  * ieee80211_is_ext - check if type is IEEE80211_FTYPE_EXT
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame type is extended
  */
 static inline bool ieee80211_is_ext(__le16 fc)
 {
@@ -496,6 +509,7 @@ static inline bool ieee80211_is_ext(__le16 fc)
 /**
  * ieee80211_is_data_qos - check if type is IEEE80211_FTYPE_DATA and IEEE80211_STYPE_QOS_DATA is set
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a QoS data frame
  */
 static inline bool ieee80211_is_data_qos(__le16 fc)
 {
@@ -510,6 +524,8 @@ static inline bool ieee80211_is_data_qos(__le16 fc)
 /**
  * ieee80211_is_data_present - check if type is IEEE80211_FTYPE_DATA and has data
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a QoS data frame that has data
+ *	(i.e. is not null data)
  */
 static inline bool ieee80211_is_data_present(__le16 fc)
 {
@@ -524,6 +540,7 @@ static inline bool ieee80211_is_data_present(__le16 fc)
 /**
  * ieee80211_is_assoc_req - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ASSOC_REQ
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is an association request
  */
 static inline bool ieee80211_is_assoc_req(__le16 fc)
 {
@@ -534,6 +551,7 @@ static inline bool ieee80211_is_assoc_req(__le16 fc)
 /**
  * ieee80211_is_assoc_resp - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ASSOC_RESP
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is an association response
  */
 static inline bool ieee80211_is_assoc_resp(__le16 fc)
 {
@@ -544,6 +562,7 @@ static inline bool ieee80211_is_assoc_resp(__le16 fc)
 /**
  * ieee80211_is_reassoc_req - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_REASSOC_REQ
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a reassociation request
  */
 static inline bool ieee80211_is_reassoc_req(__le16 fc)
 {
@@ -554,6 +573,7 @@ static inline bool ieee80211_is_reassoc_req(__le16 fc)
 /**
  * ieee80211_is_reassoc_resp - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_REASSOC_RESP
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a reassociation response
  */
 static inline bool ieee80211_is_reassoc_resp(__le16 fc)
 {
@@ -564,6 +584,7 @@ static inline bool ieee80211_is_reassoc_resp(__le16 fc)
 /**
  * ieee80211_is_probe_req - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_PROBE_REQ
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a probe request
  */
 static inline bool ieee80211_is_probe_req(__le16 fc)
 {
@@ -574,6 +595,7 @@ static inline bool ieee80211_is_probe_req(__le16 fc)
 /**
  * ieee80211_is_probe_resp - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_PROBE_RESP
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a probe response
  */
 static inline bool ieee80211_is_probe_resp(__le16 fc)
 {
@@ -584,6 +606,7 @@ static inline bool ieee80211_is_probe_resp(__le16 fc)
 /**
  * ieee80211_is_beacon - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_BEACON
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a (regular, not S1G) beacon
  */
 static inline bool ieee80211_is_beacon(__le16 fc)
 {
@@ -595,6 +618,7 @@ static inline bool ieee80211_is_beacon(__le16 fc)
  * ieee80211_is_s1g_beacon - check if IEEE80211_FTYPE_EXT &&
  * IEEE80211_STYPE_S1G_BEACON
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is an S1G beacon
  */
 static inline bool ieee80211_is_s1g_beacon(__le16 fc)
 {
@@ -618,6 +642,7 @@ static inline bool ieee80211_is_s1g_short_beacon(__le16 fc)
 /**
  * ieee80211_is_atim - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ATIM
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is an ATIM frame
  */
 static inline bool ieee80211_is_atim(__le16 fc)
 {
@@ -628,6 +653,7 @@ static inline bool ieee80211_is_atim(__le16 fc)
 /**
  * ieee80211_is_disassoc - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_DISASSOC
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a disassociation frame
  */
 static inline bool ieee80211_is_disassoc(__le16 fc)
 {
@@ -638,6 +664,7 @@ static inline bool ieee80211_is_disassoc(__le16 fc)
 /**
  * ieee80211_is_auth - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_AUTH
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is an authentication frame
  */
 static inline bool ieee80211_is_auth(__le16 fc)
 {
@@ -648,6 +675,7 @@ static inline bool ieee80211_is_auth(__le16 fc)
 /**
  * ieee80211_is_deauth - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_DEAUTH
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a deauthentication frame
  */
 static inline bool ieee80211_is_deauth(__le16 fc)
 {
@@ -658,6 +686,7 @@ static inline bool ieee80211_is_deauth(__le16 fc)
 /**
  * ieee80211_is_action - check if IEEE80211_FTYPE_MGMT && IEEE80211_STYPE_ACTION
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is an action frame
  */
 static inline bool ieee80211_is_action(__le16 fc)
 {
@@ -668,6 +697,7 @@ static inline bool ieee80211_is_action(__le16 fc)
 /**
  * ieee80211_is_back_req - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_BACK_REQ
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a block-ACK request frame
  */
 static inline bool ieee80211_is_back_req(__le16 fc)
 {
@@ -678,6 +708,7 @@ static inline bool ieee80211_is_back_req(__le16 fc)
 /**
  * ieee80211_is_back - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_BACK
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a block-ACK frame
  */
 static inline bool ieee80211_is_back(__le16 fc)
 {
@@ -688,6 +719,7 @@ static inline bool ieee80211_is_back(__le16 fc)
 /**
  * ieee80211_is_pspoll - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_PSPOLL
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a PS-poll frame
  */
 static inline bool ieee80211_is_pspoll(__le16 fc)
 {
@@ -698,6 +730,7 @@ static inline bool ieee80211_is_pspoll(__le16 fc)
 /**
  * ieee80211_is_rts - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_RTS
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is an RTS frame
  */
 static inline bool ieee80211_is_rts(__le16 fc)
 {
@@ -708,6 +741,7 @@ static inline bool ieee80211_is_rts(__le16 fc)
 /**
  * ieee80211_is_cts - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_CTS
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a CTS frame
  */
 static inline bool ieee80211_is_cts(__le16 fc)
 {
@@ -718,6 +752,7 @@ static inline bool ieee80211_is_cts(__le16 fc)
 /**
  * ieee80211_is_ack - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_ACK
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is an ACK frame
  */
 static inline bool ieee80211_is_ack(__le16 fc)
 {
@@ -728,6 +763,7 @@ static inline bool ieee80211_is_ack(__le16 fc)
 /**
  * ieee80211_is_cfend - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_CFEND
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a CF-end frame
  */
 static inline bool ieee80211_is_cfend(__le16 fc)
 {
@@ -738,6 +774,7 @@ static inline bool ieee80211_is_cfend(__le16 fc)
 /**
  * ieee80211_is_cfendack - check if IEEE80211_FTYPE_CTL && IEEE80211_STYPE_CFENDACK
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a CF-end-ack frame
  */
 static inline bool ieee80211_is_cfendack(__le16 fc)
 {
@@ -748,6 +785,7 @@ static inline bool ieee80211_is_cfendack(__le16 fc)
 /**
  * ieee80211_is_nullfunc - check if frame is a regular (non-QoS) nullfunc frame
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a nullfunc frame
  */
 static inline bool ieee80211_is_nullfunc(__le16 fc)
 {
@@ -758,6 +796,7 @@ static inline bool ieee80211_is_nullfunc(__le16 fc)
 /**
  * ieee80211_is_qos_nullfunc - check if frame is a QoS nullfunc frame
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a QoS nullfunc frame
  */
 static inline bool ieee80211_is_qos_nullfunc(__le16 fc)
 {
@@ -768,6 +807,7 @@ static inline bool ieee80211_is_qos_nullfunc(__le16 fc)
 /**
  * ieee80211_is_trigger - check if frame is trigger frame
  * @fc: frame control field in little-endian byteorder
+ * Return: whether or not the frame is a trigger frame
  */
 static inline bool ieee80211_is_trigger(__le16 fc)
 {
@@ -778,6 +818,7 @@ static inline bool ieee80211_is_trigger(__le16 fc)
 /**
  * ieee80211_is_any_nullfunc - check if frame is regular or QoS nullfunc frame
  * @fc: frame control bytes in little-endian byteorder
+ * Return: whether or not the frame is a nullfunc or QoS nullfunc frame
  */
 static inline bool ieee80211_is_any_nullfunc(__le16 fc)
 {
@@ -787,6 +828,8 @@ static inline bool ieee80211_is_any_nullfunc(__le16 fc)
 /**
  * ieee80211_is_first_frag - check if IEEE80211_SCTL_FRAG is not set
  * @seq_ctrl: frame sequence control bytes in little-endian byteorder
+ * Return: whether or not the frame is the first fragment (also true if
+ *	it's not fragmented at all)
  */
 static inline bool ieee80211_is_first_frag(__le16 seq_ctrl)
 {
@@ -796,6 +839,7 @@ static inline bool ieee80211_is_first_frag(__le16 seq_ctrl)
 /**
  * ieee80211_is_frag - check if a frame is a fragment
  * @hdr: 802.11 header of the frame
+ * Return: whether or not the frame is a fragment
  */
 static inline bool ieee80211_is_frag(struct ieee80211_hdr *hdr)
 {
@@ -2349,6 +2393,8 @@ struct ieee80211_eht_operation_info {
  * @max_vht_nss: current maximum NSS as advertised by the STA in
  *	operating mode notification, can be 0 in which case the
  *	capability data will be used to derive this (from MCS support)
+ * Return: The maximum NSS that can be used for the given bandwidth/MCS
+ *	combination
  *
  * Due to the VHT Extended NSS Bandwidth Support, the maximum NSS can
  * vary for a given BW/MCS. This function parses the data.
@@ -4317,6 +4363,7 @@ struct ieee80211_he_6ghz_capa {
 /**
  * ieee80211_get_qos_ctl - get pointer to qos control bytes
  * @hdr: the frame
+ * Return: a pointer to the QoS control field in the frame header
  *
  * The qos ctrl bytes come after the frame_control, duration, seq_num
  * and 3 or 4 addresses of length ETH_ALEN. Checks frame_control to choose
@@ -4339,6 +4386,7 @@ static inline u8 *ieee80211_get_qos_ctl(struct ieee80211_hdr *hdr)
 /**
  * ieee80211_get_tid - get qos TID
  * @hdr: the frame
+ * Return: the TID from the QoS control field
  */
 static inline u8 ieee80211_get_tid(struct ieee80211_hdr *hdr)
 {
@@ -4350,6 +4398,7 @@ static inline u8 ieee80211_get_tid(struct ieee80211_hdr *hdr)
 /**
  * ieee80211_get_SA - get pointer to SA
  * @hdr: the frame
+ * Return: a pointer to the source address (SA)
  *
  * Given an 802.11 frame, this function returns the offset
  * to the source address (SA). It does not verify that the
@@ -4369,6 +4418,7 @@ static inline u8 *ieee80211_get_SA(struct ieee80211_hdr *hdr)
 /**
  * ieee80211_get_DA - get pointer to DA
  * @hdr: the frame
+ * Return: a pointer to the destination address (DA)
  *
  * Given an 802.11 frame, this function returns the offset
  * to the destination address (DA). It does not verify that
@@ -4387,6 +4437,7 @@ static inline u8 *ieee80211_get_DA(struct ieee80211_hdr *hdr)
 /**
  * ieee80211_is_bufferable_mmpdu - check if frame is bufferable MMPDU
  * @skb: the skb to check, starting with the 802.11 header
+ * Return: whether or not the MMPDU is bufferable
  */
 static inline bool ieee80211_is_bufferable_mmpdu(struct sk_buff *skb)
 {
@@ -4425,6 +4476,7 @@ static inline bool ieee80211_is_bufferable_mmpdu(struct sk_buff *skb)
 /**
  * _ieee80211_is_robust_mgmt_frame - check if frame is a robust management frame
  * @hdr: the frame (buffer must include at least the first octet of payload)
+ * Return: whether or not the frame is a robust management frame
  */
 static inline bool _ieee80211_is_robust_mgmt_frame(struct ieee80211_hdr *hdr)
 {
@@ -4461,6 +4513,7 @@ static inline bool _ieee80211_is_robust_mgmt_frame(struct ieee80211_hdr *hdr)
 /**
  * ieee80211_is_robust_mgmt_frame - check if skb contains a robust mgmt frame
  * @skb: the skb containing the frame, length will be checked
+ * Return: whether or not the frame is a robust management frame
  */
 static inline bool ieee80211_is_robust_mgmt_frame(struct sk_buff *skb)
 {
@@ -4473,6 +4526,7 @@ static inline bool ieee80211_is_robust_mgmt_frame(struct sk_buff *skb)
  * ieee80211_is_public_action - check if frame is a public action frame
  * @hdr: the frame
  * @len: length of the frame
+ * Return: whether or not the frame is a public action frame
  */
 static inline bool ieee80211_is_public_action(struct ieee80211_hdr *hdr,
 					      size_t len)
@@ -4518,8 +4572,9 @@ ieee80211_is_protected_dual_of_public_action(struct sk_buff *skb)
 
 /**
  * _ieee80211_is_group_privacy_action - check if frame is a group addressed
- * privacy action frame
+ *	privacy action frame
  * @hdr: the frame
+ * Return: whether or not the frame is a group addressed privacy action frame
  */
 static inline bool _ieee80211_is_group_privacy_action(struct ieee80211_hdr *hdr)
 {
@@ -4535,8 +4590,9 @@ static inline bool _ieee80211_is_group_privacy_action(struct ieee80211_hdr *hdr)
 
 /**
  * ieee80211_is_group_privacy_action - check if frame is a group addressed
- * privacy action frame
+ *	privacy action frame
  * @skb: the skb containing the frame, length will be checked
+ * Return: whether or not the frame is a group addressed privacy action frame
  */
 static inline bool ieee80211_is_group_privacy_action(struct sk_buff *skb)
 {
@@ -4548,6 +4604,7 @@ static inline bool ieee80211_is_group_privacy_action(struct sk_buff *skb)
 /**
  * ieee80211_tu_to_usec - convert time units (TU) to microseconds
  * @tu: the TUs
+ * Return: the time value converted to microseconds
  */
 static inline unsigned long ieee80211_tu_to_usec(unsigned long tu)
 {
@@ -4559,6 +4616,7 @@ static inline unsigned long ieee80211_tu_to_usec(unsigned long tu)
  * @tim: the TIM IE
  * @tim_len: length of the TIM IE
  * @aid: the AID to look for
+ * Return: whether or not traffic is indicated in the TIM for the given AID
  */
 static inline bool ieee80211_check_tim(const struct ieee80211_tim_ie *tim,
 				       u8 tim_len, u16 aid)
@@ -4585,8 +4643,10 @@ static inline bool ieee80211_check_tim(const struct ieee80211_tim_ie *tim,
 }
 
 /**
- * ieee80211_get_tdls_action - get tdls packet action (or -1, if not tdls packet)
+ * ieee80211_get_tdls_action - get TDLS action code
  * @skb: the skb containing the frame, length will not be checked
+ * Return: the TDLS action code, or -1 if it's not an encapsulated TDLS action
+ *	frame
  *
  * This function assumes the frame is a data frame, and that the network header
  * is in the correct place.
@@ -4626,6 +4686,7 @@ static inline int ieee80211_get_tdls_action(struct sk_buff *skb)
 /**
  * ieee80211_action_contains_tpc - checks if the frame contains TPC element
  * @skb: the skb containing the frame, length will be checked
+ * Return: %true if the frame contains a TPC element, %false otherwise
  *
  * This function checks if it's either TPC report action frame or Link
  * Measurement report action frame as defined in IEEE Std. 802.11-2012 8.5.2.5
@@ -4743,6 +4804,7 @@ struct element {
  * @element: element pointer after for_each_element() or friends
  * @data: same data pointer as passed to for_each_element() or friends
  * @datalen: same data length as passed to for_each_element() or friends
+ * Return: %true if all elements were iterated, %false otherwise; see notes
  *
  * This function returns %true if all the data was parsed or considered
  * while walking the elements. Only use this if your for_each_element()
@@ -4946,6 +5008,7 @@ struct ieee80211_mle_tdls_common_info {
  * ieee80211_mle_common_size - check multi-link element common size
  * @data: multi-link element, must already be checked for size using
  *	ieee80211_mle_size_ok()
+ * Return: the size of the multi-link element's "common" subfield 
  */
 static inline u8 ieee80211_mle_common_size(const u8 *data)
 {
@@ -4978,11 +5041,10 @@ static inline u8 ieee80211_mle_common_size(const u8 *data)
 /**
  * ieee80211_mle_get_link_id - returns the link ID
  * @data: the basic multi link element
+ * Return: the link ID, or -1 if not present
  *
  * The element is assumed to be of the correct type (BASIC) and big enough,
  * this must be checked using ieee80211_mle_type_ok().
- *
- * If the BSS link ID can't be found, -1 will be returned
  */
 static inline int ieee80211_mle_get_link_id(const u8 *data)
 {
@@ -5002,12 +5064,10 @@ static inline int ieee80211_mle_get_link_id(const u8 *data)
 /**
  * ieee80211_mle_get_bss_param_ch_cnt - returns the BSS parameter change count
  * @data: pointer to the basic multi link element
+ * Return: the BSS Parameter Change Count field value, or -1 if not present
  *
  * The element is assumed to be of the correct type (BASIC) and big enough,
  * this must be checked using ieee80211_mle_type_ok().
- *
- * If the BSS parameter change count value can't be found (the presence bit
- * for it is clear), -1 will be returned.
  */
 static inline int
 ieee80211_mle_get_bss_param_ch_cnt(const u8 *data)
@@ -5030,13 +5090,13 @@ ieee80211_mle_get_bss_param_ch_cnt(const u8 *data)
 
 /**
  * ieee80211_mle_get_eml_med_sync_delay - returns the medium sync delay
- * @data: pointer to the multi link EHT IE
+ * @data: pointer to the multi-link element
+ * Return: the medium synchronization delay field value from the multi-link
+ *	element, or the default value (%IEEE80211_MED_SYNC_DELAY_DEFAULT)
+ *	if not present
  *
  * The element is assumed to be of the correct type (BASIC) and big enough,
  * this must be checked using ieee80211_mle_type_ok().
- *
- * If the medium synchronization is not present, then the default value is
- * returned.
  */
 static inline u16 ieee80211_mle_get_eml_med_sync_delay(const u8 *data)
 {
@@ -5060,12 +5120,12 @@ static inline u16 ieee80211_mle_get_eml_med_sync_delay(const u8 *data)
 
 /**
  * ieee80211_mle_get_eml_cap - returns the EML capability
- * @data: pointer to the multi link EHT IE
+ * @data: pointer to the multi-link element
+ * Return: the EML capability field value from the multi-link element,
+ *	or 0 if not present
  *
  * The element is assumed to be of the correct type (BASIC) and big enough,
  * this must be checked using ieee80211_mle_type_ok().
- *
- * If the EML capability is not present, 0 will be returned.
  */
 static inline u16 ieee80211_mle_get_eml_cap(const u8 *data)
 {
@@ -5091,13 +5151,12 @@ static inline u16 ieee80211_mle_get_eml_cap(const u8 *data)
 
 /**
  * ieee80211_mle_get_mld_capa_op - returns the MLD capabilities and operations.
- * @data: pointer to the multi link EHT IE
+ * @data: pointer to the multi-link element
+ * Return: the MLD capabilities and operations field value from the multi-link
+ *	element, or 0 if not present
  *
  * The element is assumed to be of the correct type (BASIC) and big enough,
  * this must be checked using ieee80211_mle_type_ok().
- *
- * If the MLD capabilities and operations field is not present, 0 will be
- * returned.
  */
 static inline u16 ieee80211_mle_get_mld_capa_op(const u8 *data)
 {
@@ -5128,12 +5187,11 @@ static inline u16 ieee80211_mle_get_mld_capa_op(const u8 *data)
 
 /**
  * ieee80211_mle_get_mld_id - returns the MLD ID
- * @data: pointer to the multi link element
+ * @data: pointer to the multi-link element
+ * Return: The MLD ID in the given multi-link element, or 0 if not present
  *
  * The element is assumed to be of the correct type (BASIC) and big enough,
  * this must be checked using ieee80211_mle_type_ok().
- *
- * If the MLD ID is not present, 0 will be returned.
  */
 static inline u8 ieee80211_mle_get_mld_id(const u8 *data)
 {
@@ -5168,6 +5226,7 @@ static inline u8 ieee80211_mle_get_mld_id(const u8 *data)
  * ieee80211_mle_size_ok - validate multi-link element size
  * @data: pointer to the element data
  * @len: length of the containing element
+ * Return: whether or not the multi-link element size is OK
  */
 static inline bool ieee80211_mle_size_ok(const u8 *data, size_t len)
 {
@@ -5237,6 +5296,7 @@ static inline bool ieee80211_mle_size_ok(const u8 *data, size_t len)
  * @data: pointer to the element data
  * @type: expected type of the element
  * @len: length of the containing element
+ * Return: whether or not the multi-link element type matches and size is OK
  */
 static inline bool ieee80211_mle_type_ok(const u8 *data, u8 type, size_t len)
 {
@@ -5280,6 +5340,7 @@ struct ieee80211_mle_per_sta_profile {
  *	profile size
  * @data: pointer to the sub element data
  * @len: length of the containing sub element
+ * Return: %true if the STA profile is large enough, %false otherwise
  */
 static inline bool ieee80211_mle_basic_sta_prof_size_ok(const u8 *data,
 							size_t len)
@@ -5364,6 +5425,7 @@ ieee80211_mle_basic_sta_prof_bss_param_ch_cnt(const struct ieee80211_mle_per_sta
  *	element sta profile size.
  * @data: pointer to the sub element data
  * @len: length of the containing sub element
+ * Return: %true if the STA profile is large enough, %false otherwise
  */
 static inline bool ieee80211_mle_reconf_sta_prof_size_ok(const u8 *data,
 							 size_t len)
-- 
cgit v1.2.3-58-ga151


From 9d222c12834d80849afbad7b42822b8ffbbc025f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 15 May 2024 09:38:40 +0200
Subject: wifi: ieee80211: document two FTM related functions

Add some documentation to ieee80211_is_timing_measurement()
and ieee80211_is_ftm().

Link: https://msgid.link/20240515093852.229aa69e972c.Ifae6762a698e79cd5a49a055fe4c32330e826200@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/linux/ieee80211.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 595f83783f0e..fb50a99daa93 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -4731,6 +4731,11 @@ static inline bool ieee80211_action_contains_tpc(struct sk_buff *skb)
 	return true;
 }
 
+/**
+ * ieee80211_is_timing_measurement - check if frame is timing measurement response
+ * @skb: the SKB to check
+ * Return: whether or not the frame is a valid timing measurement response
+ */
 static inline bool ieee80211_is_timing_measurement(struct sk_buff *skb)
 {
 	struct ieee80211_mgmt *mgmt = (void *)skb->data;
@@ -4750,6 +4755,11 @@ static inline bool ieee80211_is_timing_measurement(struct sk_buff *skb)
 	return false;
 }
 
+/**
+ * ieee80211_is_ftm - check if frame is FTM response
+ * @skb: the SKB to check
+ * Return: whether or not the frame is a valid FTM response action frame
+ */
 static inline bool ieee80211_is_ftm(struct sk_buff *skb)
 {
 	struct ieee80211_mgmt *mgmt = (void *)skb->data;
-- 
cgit v1.2.3-58-ga151


From 23daf1b4c91db9b26f8425cc7039cf96d22ccbfe Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 15 May 2024 14:16:00 +0200
Subject: wifi: nl80211: disallow setting special AP channel widths

Setting the AP channel width is meant for use with the normal
20/40/... MHz channel width progression, and switching around
in S1G or narrow channels isn't supported. Disallow that.

Reported-by: syzbot+bc0f5b92cc7091f45fb6@syzkaller.appspotmail.com
Link: https://msgid.link/20240515141600.d4a9590bfe32.I19a32d60097e81b527eafe6b0924f6c5fbb2dc45@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/nl80211.c | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 3c0bca4238d3..93c313149f57 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -3415,6 +3415,33 @@ static int __nl80211_set_channel(struct cfg80211_registered_device *rdev,
 			if (chandef.chan != cur_chan)
 				return -EBUSY;
 
+			/* only allow this for regular channel widths */
+			switch (wdev->links[link_id].ap.chandef.width) {
+			case NL80211_CHAN_WIDTH_20_NOHT:
+			case NL80211_CHAN_WIDTH_20:
+			case NL80211_CHAN_WIDTH_40:
+			case NL80211_CHAN_WIDTH_80:
+			case NL80211_CHAN_WIDTH_80P80:
+			case NL80211_CHAN_WIDTH_160:
+			case NL80211_CHAN_WIDTH_320:
+				break;
+			default:
+				return -EINVAL;
+			}
+
+			switch (chandef.width) {
+			case NL80211_CHAN_WIDTH_20_NOHT:
+			case NL80211_CHAN_WIDTH_20:
+			case NL80211_CHAN_WIDTH_40:
+			case NL80211_CHAN_WIDTH_80:
+			case NL80211_CHAN_WIDTH_80P80:
+			case NL80211_CHAN_WIDTH_160:
+			case NL80211_CHAN_WIDTH_320:
+				break;
+			default:
+				return -EINVAL;
+			}
+
 			result = rdev_set_ap_chanwidth(rdev, dev, link_id,
 						       &chandef);
 			if (result)
-- 
cgit v1.2.3-58-ga151


From 2449db1f2186e6cd8410631d0ee9e951289fe5ae Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:00:02 +0200
Subject: wifi: cfg80211: sort trace events again

They were meant to be split into ops and APIs, but some
ops were added in the wrong place. Fix that.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506210002.0b3a86a5d8d7.I5591c03223bdb95597e181de63a2eded424de34c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/trace.h | 138 +++++++++++++++++++++++++--------------------------
 1 file changed, 69 insertions(+), 69 deletions(-)

diff --git a/net/wireless/trace.h b/net/wireless/trace.h
index 9bf987519811..b76e3b21051a 100644
--- a/net/wireless/trace.h
+++ b/net/wireless/trace.h
@@ -2889,6 +2889,75 @@ DEFINE_EVENT(wiphy_wdev_link_evt, rdev_del_intf_link,
 	TP_ARGS(wiphy, wdev, link_id)
 );
 
+TRACE_EVENT(rdev_del_link_station,
+	TP_PROTO(struct wiphy *wiphy, struct net_device *netdev,
+		 struct link_station_del_parameters *params),
+	TP_ARGS(wiphy, netdev, params),
+	TP_STRUCT__entry(
+		WIPHY_ENTRY
+		NETDEV_ENTRY
+		__array(u8, mld_mac, 6)
+		__field(u32, link_id)
+	),
+	TP_fast_assign(
+		WIPHY_ASSIGN;
+		NETDEV_ASSIGN;
+		memset(__entry->mld_mac, 0, 6);
+		if (params->mld_mac)
+			memcpy(__entry->mld_mac, params->mld_mac, 6);
+		__entry->link_id = params->link_id;
+	),
+	TP_printk(WIPHY_PR_FMT ", " NETDEV_PR_FMT ", station mac: %pM"
+		  ", link id: %u",
+		  WIPHY_PR_ARG, NETDEV_PR_ARG, __entry->mld_mac,
+		  __entry->link_id)
+);
+
+TRACE_EVENT(rdev_set_hw_timestamp,
+	TP_PROTO(struct wiphy *wiphy, struct net_device *netdev,
+		 struct cfg80211_set_hw_timestamp *hwts),
+
+	TP_ARGS(wiphy, netdev, hwts),
+
+	TP_STRUCT__entry(
+		WIPHY_ENTRY
+		NETDEV_ENTRY
+		MAC_ENTRY(macaddr)
+		__field(bool, enable)
+	),
+
+	TP_fast_assign(
+		WIPHY_ASSIGN;
+		NETDEV_ASSIGN;
+		MAC_ASSIGN(macaddr, hwts->macaddr);
+		__entry->enable = hwts->enable;
+	),
+
+	TP_printk(WIPHY_PR_FMT ", " NETDEV_PR_FMT ", mac %pM, enable: %u",
+		  WIPHY_PR_ARG, NETDEV_PR_ARG, __entry->macaddr,
+		  __entry->enable)
+);
+
+TRACE_EVENT(rdev_set_ttlm,
+	TP_PROTO(struct wiphy *wiphy, struct net_device *netdev,
+		 struct cfg80211_ttlm_params *params),
+	TP_ARGS(wiphy, netdev, params),
+	TP_STRUCT__entry(
+		WIPHY_ENTRY
+		NETDEV_ENTRY
+		__array(u8, dlink, sizeof(u16) * 8)
+		__array(u8, ulink, sizeof(u16) * 8)
+	),
+	TP_fast_assign(
+		WIPHY_ASSIGN;
+		NETDEV_ASSIGN;
+		memcpy(__entry->dlink, params->dlink, sizeof(params->dlink));
+		memcpy(__entry->ulink, params->ulink, sizeof(params->ulink));
+	),
+	TP_printk(WIPHY_PR_FMT ", " NETDEV_PR_FMT,
+		  WIPHY_PR_ARG, NETDEV_PR_ARG)
+);
+
 /*************************************************************
  *	     cfg80211 exported functions traces		     *
  *************************************************************/
@@ -3923,55 +3992,6 @@ DEFINE_EVENT(link_station_add_mod, rdev_mod_link_station,
 	TP_ARGS(wiphy, netdev, params)
 );
 
-TRACE_EVENT(rdev_del_link_station,
-	TP_PROTO(struct wiphy *wiphy, struct net_device *netdev,
-		 struct link_station_del_parameters *params),
-	TP_ARGS(wiphy, netdev, params),
-	TP_STRUCT__entry(
-		WIPHY_ENTRY
-		NETDEV_ENTRY
-		__array(u8, mld_mac, 6)
-		__field(u32, link_id)
-	),
-	TP_fast_assign(
-		WIPHY_ASSIGN;
-		NETDEV_ASSIGN;
-		memset(__entry->mld_mac, 0, 6);
-		if (params->mld_mac)
-			memcpy(__entry->mld_mac, params->mld_mac, 6);
-		__entry->link_id = params->link_id;
-	),
-	TP_printk(WIPHY_PR_FMT ", " NETDEV_PR_FMT ", station mac: %pM"
-		  ", link id: %u",
-		  WIPHY_PR_ARG, NETDEV_PR_ARG, __entry->mld_mac,
-		  __entry->link_id)
-);
-
-TRACE_EVENT(rdev_set_hw_timestamp,
-	TP_PROTO(struct wiphy *wiphy, struct net_device *netdev,
-		 struct cfg80211_set_hw_timestamp *hwts),
-
-	TP_ARGS(wiphy, netdev, hwts),
-
-	TP_STRUCT__entry(
-		WIPHY_ENTRY
-		NETDEV_ENTRY
-		MAC_ENTRY(macaddr)
-		__field(bool, enable)
-	),
-
-	TP_fast_assign(
-		WIPHY_ASSIGN;
-		NETDEV_ASSIGN;
-		MAC_ASSIGN(macaddr, hwts->macaddr);
-		__entry->enable = hwts->enable;
-	),
-
-	TP_printk(WIPHY_PR_FMT ", " NETDEV_PR_FMT ", mac %pM, enable: %u",
-		  WIPHY_PR_ARG, NETDEV_PR_ARG, __entry->macaddr,
-		  __entry->enable)
-);
-
 TRACE_EVENT(cfg80211_links_removed,
 	TP_PROTO(struct net_device *netdev, u16 link_mask),
 	TP_ARGS(netdev, link_mask),
@@ -3987,26 +4007,6 @@ TRACE_EVENT(cfg80211_links_removed,
 		  __entry->link_mask)
 );
 
-TRACE_EVENT(rdev_set_ttlm,
-	TP_PROTO(struct wiphy *wiphy, struct net_device *netdev,
-		 struct cfg80211_ttlm_params *params),
-	TP_ARGS(wiphy, netdev, params),
-	TP_STRUCT__entry(
-		WIPHY_ENTRY
-		NETDEV_ENTRY
-		__array(u8, dlink, sizeof(u16) * 8)
-		__array(u8, ulink, sizeof(u16) * 8)
-	),
-	TP_fast_assign(
-		WIPHY_ASSIGN;
-		NETDEV_ASSIGN;
-		memcpy(__entry->dlink, params->dlink, sizeof(params->dlink));
-		memcpy(__entry->ulink, params->ulink, sizeof(params->ulink));
-	),
-	TP_printk(WIPHY_PR_FMT ", " NETDEV_PR_FMT,
-		  WIPHY_PR_ARG, NETDEV_PR_ARG)
-);
-
 #endif /* !__RDEV_OPS_TRACE || TRACE_HEADER_MULTI_READ */
 
 #undef TRACE_INCLUDE_PATH
-- 
cgit v1.2.3-58-ga151


From eb745c7c854e200d7fd6ba0944691a1aaf882d4d Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:00:03 +0200
Subject: wifi: cfg80211: add tracing for wiphy work

Add trace events to trace when wiphy works are queued (or
delayed ones scheduled), and other APIs are called. Also
add an event when the worker starts, before acquiring the
mutex, to be able to see potential delays due to locking.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Benjamin Berg <benjamin.berg@intel.com>
Link: https://msgid.link/20240506210002.bf1840a1d22d.I4abba048c1c4017345640219cf1384a0b2288dd3@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/core.c  | 12 +++++++++
 net/wireless/trace.h | 74 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 86 insertions(+)

diff --git a/net/wireless/core.c b/net/wireless/core.c
index 3fb1b637352a..61f7cd8a8e9c 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -421,6 +421,8 @@ static void cfg80211_wiphy_work(struct work_struct *work)
 
 	rdev = container_of(work, struct cfg80211_registered_device, wiphy_work);
 
+	trace_wiphy_work_worker_start(&rdev->wiphy);
+
 	wiphy_lock(&rdev->wiphy);
 	if (rdev->suspended)
 		goto out;
@@ -434,6 +436,7 @@ static void cfg80211_wiphy_work(struct work_struct *work)
 			schedule_work(work);
 		spin_unlock_irq(&rdev->wiphy_work_lock);
 
+		trace_wiphy_work_run(&rdev->wiphy, wk);
 		wk->func(&rdev->wiphy, wk);
 	} else {
 		spin_unlock_irq(&rdev->wiphy_work_lock);
@@ -1066,6 +1069,7 @@ void cfg80211_process_wiphy_works(struct cfg80211_registered_device *rdev,
 		list_del_init(&wk->entry);
 		spin_unlock_irqrestore(&rdev->wiphy_work_lock, flags);
 
+		trace_wiphy_work_run(&rdev->wiphy, wk);
 		wk->func(&rdev->wiphy, wk);
 
 		spin_lock_irqsave(&rdev->wiphy_work_lock, flags);
@@ -1610,6 +1614,8 @@ void wiphy_work_queue(struct wiphy *wiphy, struct wiphy_work *work)
 	struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
 	unsigned long flags;
 
+	trace_wiphy_work_queue(wiphy, work);
+
 	spin_lock_irqsave(&rdev->wiphy_work_lock, flags);
 	if (list_empty(&work->entry))
 		list_add_tail(&work->entry, &rdev->wiphy_work_list);
@@ -1626,6 +1632,8 @@ void wiphy_work_cancel(struct wiphy *wiphy, struct wiphy_work *work)
 
 	lockdep_assert_held(&wiphy->mtx);
 
+	trace_wiphy_work_cancel(wiphy, work);
+
 	spin_lock_irqsave(&rdev->wiphy_work_lock, flags);
 	if (!list_empty(&work->entry))
 		list_del_init(&work->entry);
@@ -1639,6 +1647,8 @@ void wiphy_work_flush(struct wiphy *wiphy, struct wiphy_work *work)
 	unsigned long flags;
 	bool run;
 
+	trace_wiphy_work_flush(wiphy, work);
+
 	spin_lock_irqsave(&rdev->wiphy_work_lock, flags);
 	run = !work || !list_empty(&work->entry);
 	spin_unlock_irqrestore(&rdev->wiphy_work_lock, flags);
@@ -1660,6 +1670,8 @@ void wiphy_delayed_work_queue(struct wiphy *wiphy,
 			      struct wiphy_delayed_work *dwork,
 			      unsigned long delay)
 {
+	trace_wiphy_delayed_work_queue(wiphy, &dwork->work, delay);
+
 	if (!delay) {
 		del_timer(&dwork->timer);
 		wiphy_work_queue(wiphy, &dwork->work);
diff --git a/net/wireless/trace.h b/net/wireless/trace.h
index b76e3b21051a..14cfa0aba93a 100644
--- a/net/wireless/trace.h
+++ b/net/wireless/trace.h
@@ -242,6 +242,80 @@
 		}						\
 	} while (0)
 
+/*************************************************************
+ *			wiphy work traces		     *
+ *************************************************************/
+
+DECLARE_EVENT_CLASS(wiphy_work_event,
+	TP_PROTO(struct wiphy *wiphy, struct wiphy_work *work),
+	TP_ARGS(wiphy, work),
+	TP_STRUCT__entry(
+		WIPHY_ENTRY
+		__field(void *, instance)
+		__field(void *, func)
+	),
+	TP_fast_assign(
+		WIPHY_ASSIGN;
+		__entry->instance = work;
+		__entry->func = work ? work->func : NULL;
+	),
+	TP_printk(WIPHY_PR_FMT " instance=%p func=%pS",
+		  WIPHY_PR_ARG, __entry->instance, __entry->func)
+);
+
+DEFINE_EVENT(wiphy_work_event, wiphy_work_queue,
+	TP_PROTO(struct wiphy *wiphy, struct wiphy_work *work),
+	TP_ARGS(wiphy, work)
+);
+
+DEFINE_EVENT(wiphy_work_event, wiphy_work_run,
+	TP_PROTO(struct wiphy *wiphy, struct wiphy_work *work),
+	TP_ARGS(wiphy, work)
+);
+
+DEFINE_EVENT(wiphy_work_event, wiphy_work_cancel,
+	TP_PROTO(struct wiphy *wiphy, struct wiphy_work *work),
+	TP_ARGS(wiphy, work)
+);
+
+DEFINE_EVENT(wiphy_work_event, wiphy_work_flush,
+	TP_PROTO(struct wiphy *wiphy, struct wiphy_work *work),
+	TP_ARGS(wiphy, work)
+);
+
+TRACE_EVENT(wiphy_delayed_work_queue,
+	TP_PROTO(struct wiphy *wiphy, struct wiphy_work *work,
+		 unsigned long delay),
+	TP_ARGS(wiphy, work, delay),
+	TP_STRUCT__entry(
+		WIPHY_ENTRY
+		__field(void *, instance)
+		__field(void *, func)
+		__field(unsigned long, delay)
+	),
+	TP_fast_assign(
+		WIPHY_ASSIGN;
+		__entry->instance = work;
+		__entry->func = work->func;
+		__entry->delay = delay;
+	),
+	TP_printk(WIPHY_PR_FMT " instance=%p func=%pS delay=%ld",
+		  WIPHY_PR_ARG, __entry->instance, __entry->func,
+		  __entry->delay)
+);
+
+TRACE_EVENT(wiphy_work_worker_start,
+	TP_PROTO(struct wiphy *wiphy),
+	TP_ARGS(wiphy),
+	TP_STRUCT__entry(
+		WIPHY_ENTRY
+	),
+	TP_fast_assign(
+		WIPHY_ASSIGN;
+	),
+	TP_printk(WIPHY_PR_FMT, WIPHY_PR_ARG)
+);
+
 /*************************************************************
  *			rdev->ops traces		     *
  *************************************************************/
-- 
cgit v1.2.3-58-ga151


From 7aa5c8b4f9de463abc0407da38c9ffccdbc757d1 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:14:22 +0200
Subject: wifi: mac80211: remove outdated comments

These comments are no longer correct, it's a wiphy work now
so it will go away immediately when canceled.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506211422.68bc10efbd8a.If80f43f4c8b9db1f5266f70d93a805f8c7463fe2@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c  | 4 ----
 net/mac80211/util.c | 4 ----
 2 files changed, 8 deletions(-)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index b08e5d7687e3..88af2b8037cb 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -3482,10 +3482,6 @@ static void ieee80211_end_cac(struct wiphy *wiphy,
 	lockdep_assert_wiphy(local->hw.wiphy);
 
 	list_for_each_entry(sdata, &local->interfaces, list) {
-		/* it might be waiting for the local->mtx, but then
-		 * by the time it gets it, sdata->wdev.cac_started
-		 * will no longer be true
-		 */
 		wiphy_delayed_work_cancel(wiphy,
 					  &sdata->deflink.dfs_cac_timer_work);
 
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 0b893e958959..6139c930b572 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -3456,10 +3456,6 @@ void ieee80211_dfs_cac_cancel(struct ieee80211_local *local)
 	lockdep_assert_wiphy(local->hw.wiphy);
 
 	list_for_each_entry(sdata, &local->interfaces, list) {
-		/* it might be waiting for the local->mtx, but then
-		 * by the time it gets it, sdata->wdev.cac_started
-		 * will no longer be true
-		 */
 		wiphy_delayed_work_cancel(local->hw.wiphy,
 					  &sdata->deflink.dfs_cac_timer_work);
 
-- 
cgit v1.2.3-58-ga151


From 5c24e83f6895a01a8d79a79d3ec567bd0ec13b56 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:19:34 +0200
Subject: wifi: mac80211: remove extra link STA functions

There's no need to have a lockdep assert and then call
another function, just move everything into one place.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506211934.9759564a25f4.I88d43aa459d15c1d6230152e76b7757c2cdd6085@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c | 49 +++++++++++++------------------------------------
 1 file changed, 13 insertions(+), 36 deletions(-)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 88af2b8037cb..3673c7f6b21a 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -4974,13 +4974,17 @@ static void ieee80211_del_intf_link(struct wiphy *wiphy,
 	ieee80211_vif_set_links(sdata, wdev->valid_links, 0);
 }
 
-static int sta_add_link_station(struct ieee80211_local *local,
-				struct ieee80211_sub_if_data *sdata,
-				struct link_station_parameters *params)
+static int
+ieee80211_add_link_station(struct wiphy *wiphy, struct net_device *dev,
+			   struct link_station_parameters *params)
 {
+	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+	struct ieee80211_local *local = wiphy_priv(wiphy);
 	struct sta_info *sta;
 	int ret;
 
+	lockdep_assert_wiphy(local->hw.wiphy);
+
 	sta = sta_info_get_bss(sdata, params->mld_mac);
 	if (!sta)
 		return -ENOENT;
@@ -5006,23 +5010,15 @@ static int sta_add_link_station(struct ieee80211_local *local,
 }
 
 static int
-ieee80211_add_link_station(struct wiphy *wiphy, struct net_device *dev,
+ieee80211_mod_link_station(struct wiphy *wiphy, struct net_device *dev,
 			   struct link_station_parameters *params)
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
 	struct ieee80211_local *local = wiphy_priv(wiphy);
-
-	lockdep_assert_wiphy(sdata->local->hw.wiphy);
-
-	return sta_add_link_station(local, sdata, params);
-}
-
-static int sta_mod_link_station(struct ieee80211_local *local,
-				struct ieee80211_sub_if_data *sdata,
-				struct link_station_parameters *params)
-{
 	struct sta_info *sta;
 
+	lockdep_assert_wiphy(local->hw.wiphy);
+
 	sta = sta_info_get_bss(sdata, params->mld_mac);
 	if (!sta)
 		return -ENOENT;
@@ -5034,22 +5030,14 @@ static int sta_mod_link_station(struct ieee80211_local *local,
 }
 
 static int
-ieee80211_mod_link_station(struct wiphy *wiphy, struct net_device *dev,
-			   struct link_station_parameters *params)
+ieee80211_del_link_station(struct wiphy *wiphy, struct net_device *dev,
+			   struct link_station_del_parameters *params)
 {
 	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-	struct ieee80211_local *local = wiphy_priv(wiphy);
+	struct sta_info *sta;
 
 	lockdep_assert_wiphy(sdata->local->hw.wiphy);
 
-	return sta_mod_link_station(local, sdata, params);
-}
-
-static int sta_del_link_station(struct ieee80211_sub_if_data *sdata,
-				struct link_station_del_parameters *params)
-{
-	struct sta_info *sta;
-
 	sta = sta_info_get_bss(sdata, params->mld_mac);
 	if (!sta)
 		return -ENOENT;
@@ -5066,17 +5054,6 @@ static int sta_del_link_station(struct ieee80211_sub_if_data *sdata,
 	return 0;
 }
 
-static int
-ieee80211_del_link_station(struct wiphy *wiphy, struct net_device *dev,
-			   struct link_station_del_parameters *params)
-{
-	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
-
-	lockdep_assert_wiphy(sdata->local->hw.wiphy);
-
-	return sta_del_link_station(sdata, params);
-}
-
 static int ieee80211_set_hw_timestamp(struct wiphy *wiphy,
 				      struct net_device *dev,
 				      struct cfg80211_set_hw_timestamp *hwts)
-- 
cgit v1.2.3-58-ga151


From e3bae9b22873338d1a7989a2568c5aca2c44bde7 Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Mon, 6 May 2024 21:37:53 +0200
Subject: wifi: mac80211: update 6 GHz AP power type before association

6 GHz AP power type details are required to set proper tx power
used to send frames.

Update AP power type received in beacon while preparing
for connection instead of after association so the frames
before association can use the correct tx power.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Link: https://msgid.link/20240506214536.310434f55f76.I6aca291ee06265e3f63e0f9024ba19a850b53a33@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 55 ++++++++++++++++++++++++++---------------------------
 1 file changed, 27 insertions(+), 28 deletions(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index a5f2d3cfe60d..e410a43dc681 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -4445,40 +4445,12 @@ static bool ieee80211_assoc_config_link(struct ieee80211_link_data *link,
 	if (elems->he_operation &&
 	    link->u.mgd.conn.mode >= IEEE80211_CONN_MODE_HE &&
 	    elems->he_cap) {
-		const struct ieee80211_he_6ghz_oper *he_6ghz_oper;
-
 		ieee80211_he_cap_ie_to_sta_he_cap(sdata, sband,
 						  elems->he_cap,
 						  elems->he_cap_len,
 						  elems->he_6ghz_capa,
 						  link_sta);
 
-		he_6ghz_oper = ieee80211_he_6ghz_oper(elems->he_operation);
-
-		if (is_6ghz && he_6ghz_oper) {
-			switch (u8_get_bits(he_6ghz_oper->control,
-					    IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) {
-			case IEEE80211_6GHZ_CTRL_REG_LPI_AP:
-			case IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP:
-				bss_conf->power_type = IEEE80211_REG_LPI_AP;
-				break;
-			case IEEE80211_6GHZ_CTRL_REG_SP_AP:
-			case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP:
-				bss_conf->power_type = IEEE80211_REG_SP_AP;
-				break;
-			case IEEE80211_6GHZ_CTRL_REG_VLP_AP:
-				bss_conf->power_type = IEEE80211_REG_VLP_AP;
-				break;
-			default:
-				bss_conf->power_type = IEEE80211_REG_UNSET_AP;
-				break;
-			}
-		} else if (is_6ghz) {
-			link_info(link,
-				  "HE 6 GHz operation missing (on %d MHz), expect issues\n",
-				  bss_conf->chanreq.oper.chan->center_freq);
-		}
-
 		bss_conf->he_support = link_sta->pub->he_cap.has_he;
 		if (elems->rsnx && elems->rsnx_len &&
 		    (elems->rsnx[0] & WLAN_RSNX_CAPA_PROTECTED_TWT) &&
@@ -5020,6 +4992,23 @@ ieee80211_determine_our_sta_mode_assoc(struct ieee80211_sub_if_data *sdata,
 			       conn->bw_limit, tmp.bw_limit);
 }
 
+static enum ieee80211_ap_reg_power
+ieee80211_ap_power_type(u8 control)
+{
+	switch (u8_get_bits(control, IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) {
+	case IEEE80211_6GHZ_CTRL_REG_LPI_AP:
+	case IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP:
+		return IEEE80211_REG_LPI_AP;
+	case IEEE80211_6GHZ_CTRL_REG_SP_AP:
+	case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP:
+		return IEEE80211_REG_SP_AP;
+	case IEEE80211_6GHZ_CTRL_REG_VLP_AP:
+		return IEEE80211_REG_VLP_AP;
+	default:
+		return IEEE80211_REG_UNSET_AP;
+	}
+}
+
 static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata,
 				  struct ieee80211_link_data *link,
 				  int link_id,
@@ -5052,6 +5041,7 @@ static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata,
 	}
 
 	if (link && is_6ghz && conn->mode >= IEEE80211_CONN_MODE_HE) {
+		const struct ieee80211_he_6ghz_oper *he_6ghz_oper;
 		struct ieee80211_bss_conf *bss_conf;
 		u8 j = 0;
 
@@ -5072,6 +5062,15 @@ static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata,
 			       elems->tx_pwr_env_len[i]);
 			j++;
 		}
+
+		he_6ghz_oper = ieee80211_he_6ghz_oper(elems->he_operation);
+		if (he_6ghz_oper)
+			bss_conf->power_type =
+				ieee80211_ap_power_type(he_6ghz_oper->control);
+		else
+			link_info(link,
+				  "HE 6 GHz operation missing (on %d MHz), expect issues\n",
+				  cbss->channel->center_freq);
 	}
 	rcu_read_unlock();
 	/* the element data was RCU protected so no longer valid anyway */
-- 
cgit v1.2.3-58-ga151


From 8592fd7ccc95a7cf222985e6297041463e8b4e9d Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:37:54 +0200
Subject: wifi: ieee80211/ath11k: remove IEEE80211_MAX_NUM_PWR_LEVEL

The define IEEE80211_MAX_NUM_PWR_LEVEL doesn't make much
sense. Yes, that table has a maximum value of 8, and the
table will actually remain that way, but EHT introduced
a way to encode more levels for 320 MHz channels.

Remove IEEE80211_MAX_NUM_PWR_LEVEL and, for ath11k being
the only user, add ATH11K_NUM_PWR_LEVELS, where it makes
sense since it cannot support 320 MHz channels.

Acked-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240506214536.9818e5471055.Icece7e47e963d6b68e0d97ba13c102b37fbaa689@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/ath/ath11k/core.h | 9 ++++++---
 drivers/net/wireless/ath/ath11k/mac.c  | 2 +-
 include/linux/ieee80211.h              | 5 -----
 3 files changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/core.h b/drivers/net/wireless/ath/ath11k/core.h
index 205f40ee6b66..df24f0e409af 100644
--- a/drivers/net/wireless/ath/ath11k/core.h
+++ b/drivers/net/wireless/ath/ath11k/core.h
@@ -330,6 +330,9 @@ struct ath11k_chan_power_info {
 	s8 tx_power;
 };
 
+/* ath11k only deals with 160 MHz, so 8 subchannels */
+#define ATH11K_NUM_PWR_LEVELS	8
+
 /**
  * struct ath11k_reg_tpc_power_info - regulatory TPC power info
  * @is_psd_power: is PSD power or not
@@ -346,10 +349,10 @@ struct ath11k_reg_tpc_power_info {
 	u8 eirp_power;
 	enum wmi_reg_6ghz_ap_type ap_power_type;
 	u8 num_pwr_levels;
-	u8 reg_max[IEEE80211_MAX_NUM_PWR_LEVEL];
+	u8 reg_max[ATH11K_NUM_PWR_LEVELS];
 	u8 ap_constraint_power;
-	s8 tpe[IEEE80211_MAX_NUM_PWR_LEVEL];
-	struct ath11k_chan_power_info chan_power_info[IEEE80211_MAX_NUM_PWR_LEVEL];
+	s8 tpe[ATH11K_NUM_PWR_LEVELS];
+	struct ath11k_chan_power_info chan_power_info[ATH11K_NUM_PWR_LEVELS];
 };
 
 struct ath11k_vif {
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 4f62e38ba48b..59e5e8085742 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -7688,7 +7688,7 @@ void ath11k_mac_fill_reg_tpc_info(struct ath11k *ar,
 	struct ieee80211_channel *chan, *temp_chan;
 	u8 pwr_lvl_idx, num_pwr_levels, pwr_reduction;
 	bool is_psd_power = false, is_tpe_present = false;
-	s8 max_tx_power[IEEE80211_MAX_NUM_PWR_LEVEL],
+	s8 max_tx_power[ATH11K_NUM_PWR_LEVELS],
 		psd_power, tx_power;
 	s8 eirp_power = 0;
 	u16 start_freq, center_freq;
diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index fb50a99daa93..1c3a683a3ee2 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -2833,11 +2833,6 @@ struct ieee80211_he_6ghz_oper {
  * So it it totally max 8 Transmit Power Envelope element.
  */
 #define IEEE80211_TPE_MAX_IE_COUNT	8
-/*
- * In "Table 9-277—Meaning of Maximum Transmit Power Count subfield"
- * of "IEEE Std 802.11ax™‐2021", the max power level is 8.
- */
-#define IEEE80211_MAX_NUM_PWR_LEVEL	8
 
 #define IEEE80211_TPE_MAX_POWER_COUNT	8
 
-- 
cgit v1.2.3-58-ga151


From caa12b3dd44996f63b8dd0d842c68866d4ead307 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:37:55 +0200
Subject: wifi: ath11k: fix TPE power levels

The chanctx can be shared, but it seems like the TPE values for
this vif should only reflect the channel that the vif uses, which
is captured in bss_conf->chanreq.oper rather than the chanctx.
Use that when calculating defaults, a similar issue is where the
data from the AP is processed, but I'm moving that to mac80211.

Acked-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240506214536.d99fcf5856b9.I359bb401ea74bf82dc981a86a7ca3fa44d169d2e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/ath/ath11k/mac.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 59e5e8085742..73876760afe9 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -7701,7 +7701,8 @@ void ath11k_mac_fill_reg_tpc_info(struct ath11k *ar,
 		is_tpe_present = true;
 		num_pwr_levels = arvif->reg_tpc_info.num_pwr_levels;
 	} else {
-		num_pwr_levels = ath11k_mac_get_num_pwr_levels(&ctx->def);
+		num_pwr_levels =
+			ath11k_mac_get_num_pwr_levels(&bss_conf->chanreq.oper);
 	}
 
 	for (pwr_lvl_idx = 0; pwr_lvl_idx < num_pwr_levels; pwr_lvl_idx++) {
-- 
cgit v1.2.3-58-ga151


From 39dc8b8ea387ce7f4fe2d2d6d550ed52aa9aa040 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:37:56 +0200
Subject: wifi: mac80211: pass parsed TPE data to drivers

Instead of passing the full TPE elements, in all their glory
and mixed up data formats for HE backward compatibility, parse
them fully into the right values, and pass that to the drivers.

Also introduce proper validation already in mac80211, so that
drivers don't need to do it, and parse the EHT portions.

The code now passes the values in the right order according to
the channel used by an interface, which could also be a subset
of the data advertised by the AP, if we couldn't connect with
the full bandwidth (for whatever reason.)

Also add kunit tests for the more complicated bits of it.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Acked-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240506214536.2aa839969b60.I265b28209e0b29772b2f125f7f83de44a4da877b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/ath/ath11k/mac.c | 143 ++++-------------
 include/linux/ieee80211.h             | 102 ++++++++++--
 include/net/mac80211.h                |  41 ++++-
 net/mac80211/ieee80211_i.h            |  13 +-
 net/mac80211/mlme.c                   | 230 +++++++++++++++++++++++----
 net/mac80211/parse.c                  |  92 +++++++++--
 net/mac80211/tests/Makefile           |   2 +-
 net/mac80211/tests/tpe.c              | 284 ++++++++++++++++++++++++++++++++++
 net/mac80211/util.c                   |  25 +++
 9 files changed, 761 insertions(+), 171 deletions(-)
 create mode 100644 net/mac80211/tests/tpe.c

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 73876760afe9..d4225bf30e51 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -7507,32 +7507,6 @@ static int ath11k_mac_stop_vdev_early(struct ieee80211_hw *hw,
 	return 0;
 }
 
-static u8 ath11k_mac_get_tpe_count(u8 txpwr_intrprt, u8 txpwr_cnt)
-{
-	switch (txpwr_intrprt) {
-	/* Refer "Table 9-276-Meaning of Maximum Transmit Power Count subfield
-	 * if the Maximum Transmit Power Interpretation subfield is 0 or 2" of
-	 * "IEEE Std 802.11ax 2021".
-	 */
-	case IEEE80211_TPE_LOCAL_EIRP:
-	case IEEE80211_TPE_REG_CLIENT_EIRP:
-		txpwr_cnt = txpwr_cnt <= 3 ? txpwr_cnt : 3;
-		txpwr_cnt = txpwr_cnt + 1;
-		break;
-	/* Refer "Table 9-277-Meaning of Maximum Transmit Power Count subfield
-	 * if Maximum Transmit Power Interpretation subfield is 1 or 3" of
-	 * "IEEE Std 802.11ax 2021".
-	 */
-	case IEEE80211_TPE_LOCAL_EIRP_PSD:
-	case IEEE80211_TPE_REG_CLIENT_EIRP_PSD:
-		txpwr_cnt = txpwr_cnt <= 4 ? txpwr_cnt : 4;
-		txpwr_cnt = txpwr_cnt ? (BIT(txpwr_cnt - 1)) : 1;
-		break;
-	}
-
-	return txpwr_cnt;
-}
-
 static u8 ath11k_mac_get_num_pwr_levels(struct cfg80211_chan_def *chan_def)
 {
 	if (chan_def->chan->flags & IEEE80211_CHAN_PSD) {
@@ -7859,33 +7833,23 @@ static void ath11k_mac_parse_tx_pwr_env(struct ath11k *ar,
 	struct ath11k_base *ab = ar->ab;
 	struct ath11k_vif *arvif = ath11k_vif_to_arvif(vif);
 	struct ieee80211_bss_conf *bss_conf = &vif->bss_conf;
-	struct ieee80211_tx_pwr_env *single_tpe;
+	struct ieee80211_parsed_tpe_eirp *non_psd = NULL;
+	struct ieee80211_parsed_tpe_psd *psd = NULL;
 	enum wmi_reg_6ghz_client_type client_type;
 	struct cur_regulatory_info *reg_info;
+	u8 local_tpe_count, reg_tpe_count;
+	bool use_local_tpe;
 	int i;
-	u8 pwr_count, pwr_interpret, pwr_category;
-	u8 psd_index = 0, non_psd_index = 0, local_tpe_count = 0, reg_tpe_count = 0;
-	bool use_local_tpe, non_psd_set = false, psd_set = false;
 
 	reg_info = &ab->reg_info_store[ar->pdev_idx];
 	client_type = reg_info->client_type;
 
-	for (i = 0; i < bss_conf->tx_pwr_env_num; i++) {
-		single_tpe = &bss_conf->tx_pwr_env[i];
-		pwr_category = u8_get_bits(single_tpe->tx_power_info,
-					   IEEE80211_TX_PWR_ENV_INFO_CATEGORY);
-		pwr_interpret = u8_get_bits(single_tpe->tx_power_info,
-					    IEEE80211_TX_PWR_ENV_INFO_INTERPRET);
-
-		if (pwr_category == client_type) {
-			if (pwr_interpret == IEEE80211_TPE_LOCAL_EIRP ||
-			    pwr_interpret == IEEE80211_TPE_LOCAL_EIRP_PSD)
-				local_tpe_count++;
-			else if (pwr_interpret == IEEE80211_TPE_REG_CLIENT_EIRP ||
-				 pwr_interpret == IEEE80211_TPE_REG_CLIENT_EIRP_PSD)
-				reg_tpe_count++;
-		}
-	}
+	local_tpe_count =
+		bss_conf->tpe.max_local[client_type].valid +
+		bss_conf->tpe.psd_local[client_type].valid;
+	reg_tpe_count =
+		bss_conf->tpe.max_reg_client[client_type].valid +
+		bss_conf->tpe.psd_reg_client[client_type].valid;
 
 	if (!reg_tpe_count && !local_tpe_count) {
 		ath11k_warn(ab,
@@ -7898,83 +7862,44 @@ static void ath11k_mac_parse_tx_pwr_env(struct ath11k *ar,
 		use_local_tpe = false;
 	}
 
-	for (i = 0; i < bss_conf->tx_pwr_env_num; i++) {
-		single_tpe = &bss_conf->tx_pwr_env[i];
-		pwr_category = u8_get_bits(single_tpe->tx_power_info,
-					   IEEE80211_TX_PWR_ENV_INFO_CATEGORY);
-		pwr_interpret = u8_get_bits(single_tpe->tx_power_info,
-					    IEEE80211_TX_PWR_ENV_INFO_INTERPRET);
-
-		if (pwr_category != client_type)
-			continue;
-
-		/* get local transmit power envelope */
-		if (use_local_tpe) {
-			if (pwr_interpret == IEEE80211_TPE_LOCAL_EIRP) {
-				non_psd_index = i;
-				non_psd_set = true;
-			} else if (pwr_interpret == IEEE80211_TPE_LOCAL_EIRP_PSD) {
-				psd_index = i;
-				psd_set = true;
-			}
-		/* get regulatory transmit power envelope */
-		} else {
-			if (pwr_interpret == IEEE80211_TPE_REG_CLIENT_EIRP) {
-				non_psd_index = i;
-				non_psd_set = true;
-			} else if (pwr_interpret == IEEE80211_TPE_REG_CLIENT_EIRP_PSD) {
-				psd_index = i;
-				psd_set = true;
-			}
-		}
+	if (use_local_tpe) {
+		psd = &bss_conf->tpe.psd_local[client_type];
+		if (!psd->valid)
+			psd = NULL;
+		non_psd = &bss_conf->tpe.max_local[client_type];
+		if (!non_psd->valid)
+			non_psd = NULL;
+	} else {
+		psd = &bss_conf->tpe.psd_reg_client[client_type];
+		if (!psd->valid)
+			psd = NULL;
+		non_psd = &bss_conf->tpe.max_reg_client[client_type];
+		if (!non_psd->valid)
+			non_psd = NULL;
 	}
 
-	if (non_psd_set && !psd_set) {
-		single_tpe = &bss_conf->tx_pwr_env[non_psd_index];
-		pwr_count = u8_get_bits(single_tpe->tx_power_info,
-					IEEE80211_TX_PWR_ENV_INFO_COUNT);
-		pwr_interpret = u8_get_bits(single_tpe->tx_power_info,
-					    IEEE80211_TX_PWR_ENV_INFO_INTERPRET);
+	if (non_psd && !psd) {
 		arvif->reg_tpc_info.is_psd_power = false;
 		arvif->reg_tpc_info.eirp_power = 0;
 
-		arvif->reg_tpc_info.num_pwr_levels =
-			ath11k_mac_get_tpe_count(pwr_interpret, pwr_count);
+		arvif->reg_tpc_info.num_pwr_levels = non_psd->count;
 
 		for (i = 0; i < arvif->reg_tpc_info.num_pwr_levels; i++) {
 			ath11k_dbg(ab, ATH11K_DBG_MAC,
 				   "non PSD power[%d] : %d\n",
-				   i, single_tpe->tx_power[i]);
-			arvif->reg_tpc_info.tpe[i] = single_tpe->tx_power[i] / 2;
+				   i, non_psd->power[i]);
+			arvif->reg_tpc_info.tpe[i] = non_psd->power[i] / 2;
 		}
 	}
 
-	if (psd_set) {
-		single_tpe = &bss_conf->tx_pwr_env[psd_index];
-		pwr_count = u8_get_bits(single_tpe->tx_power_info,
-					IEEE80211_TX_PWR_ENV_INFO_COUNT);
-		pwr_interpret = u8_get_bits(single_tpe->tx_power_info,
-					    IEEE80211_TX_PWR_ENV_INFO_INTERPRET);
-		arvif->reg_tpc_info.is_psd_power = true;
+	if (psd) {
+		arvif->reg_tpc_info.num_pwr_levels = psd->count;
 
-		if (pwr_count == 0) {
+		for (i = 0; i < arvif->reg_tpc_info.num_pwr_levels; i++) {
 			ath11k_dbg(ab, ATH11K_DBG_MAC,
-				   "TPE PSD power : %d\n", single_tpe->tx_power[0]);
-			arvif->reg_tpc_info.num_pwr_levels =
-				ath11k_mac_get_num_pwr_levels(&ctx->def);
-
-			for (i = 0; i < arvif->reg_tpc_info.num_pwr_levels; i++)
-				arvif->reg_tpc_info.tpe[i] = single_tpe->tx_power[0] / 2;
-		} else {
-			arvif->reg_tpc_info.num_pwr_levels =
-				ath11k_mac_get_tpe_count(pwr_interpret, pwr_count);
-
-			for (i = 0; i < arvif->reg_tpc_info.num_pwr_levels; i++) {
-				ath11k_dbg(ab, ATH11K_DBG_MAC,
-					   "TPE PSD power[%d] : %d\n",
-					   i, single_tpe->tx_power[i]);
-				arvif->reg_tpc_info.tpe[i] = single_tpe->tx_power[i] / 2;
-			}
+				   "TPE PSD power[%d] : %d\n",
+				   i, psd->power[i]);
+			arvif->reg_tpc_info.tpe[i] = psd->power[i] / 2;
 		}
 	}
 }
diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 1c3a683a3ee2..769008a51809 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -2825,17 +2825,6 @@ struct ieee80211_he_6ghz_oper {
 	u8 minrate;
 } __packed;
 
-/*
- * In "9.4.2.161 Transmit Power Envelope element" of "IEEE Std 802.11ax-2021",
- * it show four types in "Table 9-275a-Maximum Transmit Power Interpretation
- * subfield encoding", and two category for each type in "Table E-12-Regulatory
- * Info subfield encoding in the United States".
- * So it it totally max 8 Transmit Power Envelope element.
- */
-#define IEEE80211_TPE_MAX_IE_COUNT	8
-
-#define IEEE80211_TPE_MAX_POWER_COUNT	8
-
 /* transmit power interpretation type of transmit power envelope element */
 enum ieee80211_tx_power_intrpt_type {
 	IEEE80211_TPE_LOCAL_EIRP,
@@ -2844,24 +2833,107 @@ enum ieee80211_tx_power_intrpt_type {
 	IEEE80211_TPE_REG_CLIENT_EIRP_PSD,
 };
 
+/* category type of transmit power envelope element */
+enum ieee80211_tx_power_category_6ghz {
+	IEEE80211_TPE_CAT_6GHZ_DEFAULT = 0,
+	IEEE80211_TPE_CAT_6GHZ_SUBORDINATE = 1,
+};
+
+/*
+ * For IEEE80211_TPE_LOCAL_EIRP / IEEE80211_TPE_REG_CLIENT_EIRP,
+ * setting to 63.5 dBm means no constraint.
+ */
+#define IEEE80211_TPE_MAX_TX_PWR_NO_CONSTRAINT	127
+
+/*
+ * For IEEE80211_TPE_LOCAL_EIRP_PSD / IEEE80211_TPE_REG_CLIENT_EIRP_PSD,
+ * setting to 127 indicates no PSD limit for the 20 MHz channel.
+ */
+#define IEEE80211_TPE_PSD_NO_LIMIT		127
+
 /**
  * struct ieee80211_tx_pwr_env - Transmit Power Envelope
- * @tx_power_info: Transmit Power Information field
- * @tx_power: Maximum Transmit Power field
+ * @info: Transmit Power Information field
+ * @variable: Maximum Transmit Power field
  *
  * This structure represents the payload of the "Transmit Power
  * Envelope element" as described in IEEE Std 802.11ax-2021 section
  * 9.4.2.161
  */
 struct ieee80211_tx_pwr_env {
-	u8 tx_power_info;
-	s8 tx_power[IEEE80211_TPE_MAX_POWER_COUNT];
+	u8 info;
+	u8 variable[];
 } __packed;
 
 #define IEEE80211_TX_PWR_ENV_INFO_COUNT 0x7
 #define IEEE80211_TX_PWR_ENV_INFO_INTERPRET 0x38
 #define IEEE80211_TX_PWR_ENV_INFO_CATEGORY 0xC0
 
+#define IEEE80211_TX_PWR_ENV_EXT_COUNT	0xF
+
+static inline bool ieee80211_valid_tpe_element(const u8 *data, u8 len)
+{
+	const struct ieee80211_tx_pwr_env *env = (const void *)data;
+	u8 count, interpret, category;
+	u8 needed = sizeof(*env);
+	u8 N; /* also called N in the spec */
+
+	if (len < needed)
+		return false;
+
+	count = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_COUNT);
+	interpret = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_INTERPRET);
+	category = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_CATEGORY);
+
+	switch (category) {
+	case IEEE80211_TPE_CAT_6GHZ_DEFAULT:
+	case IEEE80211_TPE_CAT_6GHZ_SUBORDINATE:
+		break;
+	default:
+		return false;
+	}
+
+	switch (interpret) {
+	case IEEE80211_TPE_LOCAL_EIRP:
+	case IEEE80211_TPE_REG_CLIENT_EIRP:
+		if (count > 3)
+			return false;
+
+		/* count == 0 encodes 1 value for 20 MHz, etc. */
+		needed += count + 1;
+
+		if (len < needed)
+			return false;
+
+		/* there can be extension fields not accounted for in 'count' */
+
+		return true;
+	case IEEE80211_TPE_LOCAL_EIRP_PSD:
+	case IEEE80211_TPE_REG_CLIENT_EIRP_PSD:
+		if (count > 4)
+			return false;
+
+		N = count ? 1 << (count - 1) : 1;
+		needed += N;
+
+		if (len < needed)
+			return false;
+
+		if (len > needed) {
+			u8 K = u8_get_bits(env->variable[N],
+					   IEEE80211_TX_PWR_ENV_EXT_COUNT);
+
+			needed += 1 + K;
+			if (len < needed)
+				return false;
+		}
+
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * ieee80211_he_oper_size - calculate 802.11ax HE Operations IE size
  * @he_oper_ie: byte data of the He Operations IE, stating from the byte
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index cafc664ee531..a4efbfb8d796 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -550,6 +550,39 @@ struct ieee80211_fils_discovery {
 	u32 max_interval;
 };
 
+#define IEEE80211_TPE_EIRP_ENTRIES_320MHZ	5
+struct ieee80211_parsed_tpe_eirp {
+	bool valid;
+	s8 power[IEEE80211_TPE_EIRP_ENTRIES_320MHZ];
+	u8 count;
+};
+
+#define IEEE80211_TPE_PSD_ENTRIES_320MHZ	16
+struct ieee80211_parsed_tpe_psd {
+	bool valid;
+	s8 power[IEEE80211_TPE_PSD_ENTRIES_320MHZ];
+	u8 count, n;
+};
+
+/**
+ * struct ieee80211_parsed_tpe - parsed transmit power envelope information
+ * @max_local: maximum local EIRP, one value for 20, 40, 80, 160, 320 MHz each
+ *	(indexed by TX power category)
+ * @max_reg_client: maximum regulatory client EIRP, one value for 20, 40, 80,
+ *	160, 320 MHz each
+ *	(indexed by TX power category)
+ * @psd_local: maximum local power spectral density, one value for each 20 MHz
+ *	subchannel per bss_conf's chanreq.oper
+ *	(indexed by TX power category)
+ * @psd_reg_client: maximum regulatory power spectral density, one value for
+ *	each 20 MHz subchannel per bss_conf's chanreq.oper
+ *	(indexed by TX power category)
+ */
+struct ieee80211_parsed_tpe {
+	struct ieee80211_parsed_tpe_eirp max_local[2], max_reg_client[2];
+	struct ieee80211_parsed_tpe_psd psd_local[2], psd_reg_client[2];
+};
+
 /**
  * struct ieee80211_bss_conf - holds the BSS's changing parameters
  *
@@ -662,8 +695,7 @@ struct ieee80211_fils_discovery {
  * @beacon_tx_rate: The configured beacon transmit rate that needs to be passed
  *	to driver when rate control is offloaded to firmware.
  * @power_type: power type of BSS for 6 GHz
- * @tx_pwr_env: transmit power envelope array of BSS.
- * @tx_pwr_env_num: number of @tx_pwr_env.
+ * @tpe: transmit power envelope information
  * @pwr_reduction: power constraint of BSS.
  * @eht_support: does this BSS support EHT
  * @csa_active: marks whether a channel switch is going on.
@@ -766,8 +798,9 @@ struct ieee80211_bss_conf {
 	u32 unsol_bcast_probe_resp_interval;
 	struct cfg80211_bitrate_mask beacon_tx_rate;
 	enum ieee80211_ap_reg_power power_type;
-	struct ieee80211_tx_pwr_env tx_pwr_env[IEEE80211_TPE_MAX_IE_COUNT];
-	u8 tx_pwr_env_num;
+
+	struct ieee80211_parsed_tpe tpe;
+
 	u8 pwr_reduction;
 	bool eht_support;
 
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index eb62b7d4b4f7..7d541a2355f6 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -26,6 +26,7 @@
 #include <linux/idr.h>
 #include <linux/rhashtable.h>
 #include <linux/rbtree.h>
+#include <kunit/visibility.h>
 #include <net/ieee80211_radiotap.h>
 #include <net/cfg80211.h>
 #include <net/mac80211.h>
@@ -1708,7 +1709,6 @@ struct ieee802_11_elems {
 	const struct ieee80211_he_spr *he_spr;
 	const struct ieee80211_mu_edca_param_set *mu_edca_param_set;
 	const struct ieee80211_he_6ghz_capa *he_6ghz_capa;
-	const struct ieee80211_tx_pwr_env *tx_pwr_env[IEEE80211_TPE_MAX_IE_COUNT];
 	const u8 *uora_element;
 	const u8 *mesh_id;
 	const u8 *peering;
@@ -1746,6 +1746,9 @@ struct ieee802_11_elems {
 	const struct ieee80211_bandwidth_indication *bandwidth_indication;
 	const struct ieee80211_ttlm_elem *ttlm[IEEE80211_TTLM_MAX_CNT];
 
+	/* not the order in the psd values is per element, not per chandef */
+	struct ieee80211_parsed_tpe tpe;
+
 	/* length of them, respectively */
 	u8 ext_capab_len;
 	u8 ssid_len;
@@ -1764,8 +1767,6 @@ struct ieee802_11_elems {
 	u8 perr_len;
 	u8 country_elem_len;
 	u8 bssid_index_len;
-	u8 tx_pwr_env_len[IEEE80211_TPE_MAX_IE_COUNT];
-	u8 tx_pwr_env_num;
 	u8 eht_cap_len;
 
 	/* mult-link element can be de-fragmented and thus u8 is not sufficient */
@@ -2243,6 +2244,7 @@ int ieee80211_frame_duration(enum nl80211_band band, size_t len,
 void ieee80211_regulatory_limit_wmm_params(struct ieee80211_sub_if_data *sdata,
 					   struct ieee80211_tx_queue_params *qparam,
 					   int ac);
+void ieee80211_clear_tpe(struct ieee80211_parsed_tpe *tpe);
 void ieee80211_set_wmm_default(struct ieee80211_link_data *link,
 			       bool bss_notify, bool enable_qos);
 void ieee80211_xmit(struct ieee80211_sub_if_data *sdata,
@@ -2681,6 +2683,11 @@ void ieee80211_remove_wbrf(struct ieee80211_local *local, struct cfg80211_chan_d
 #define VISIBLE_IF_MAC80211_KUNIT
 ieee80211_rx_result
 ieee80211_drop_unencrypted_mgmt(struct ieee80211_rx_data *rx);
+int ieee80211_calc_chandef_subchan_offset(const struct cfg80211_chan_def *ap,
+					  u8 n_partial_subchans);
+void ieee80211_rearrange_tpe_psd(struct ieee80211_parsed_tpe_psd *psd,
+				 const struct cfg80211_chan_def *ap,
+				 const struct cfg80211_chan_def *used);
 #else
 #define EXPORT_SYMBOL_IF_MAC80211_KUNIT(sym)
 #define VISIBLE_IF_MAC80211_KUNIT static
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index e410a43dc681..ed9851faac05 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -606,11 +606,195 @@ static bool ieee80211_chandef_usable(struct ieee80211_sub_if_data *sdata,
 	return true;
 }
 
+static int ieee80211_chandef_num_subchans(const struct cfg80211_chan_def *c)
+{
+	if (c->width == NL80211_CHAN_WIDTH_80P80)
+		return 4 + 4;
+
+	return nl80211_chan_width_to_mhz(c->width) / 20;
+}
+
+static int ieee80211_chandef_num_widths(const struct cfg80211_chan_def *c)
+{
+	switch (c->width) {
+	case NL80211_CHAN_WIDTH_20:
+	case NL80211_CHAN_WIDTH_20_NOHT:
+		return 1;
+	case NL80211_CHAN_WIDTH_40:
+		return 2;
+	case NL80211_CHAN_WIDTH_80P80:
+	case NL80211_CHAN_WIDTH_80:
+		return 3;
+	case NL80211_CHAN_WIDTH_160:
+		return 4;
+	case NL80211_CHAN_WIDTH_320:
+		return 5;
+	default:
+		WARN_ON(1);
+		return 0;
+	}
+}
+
+VISIBLE_IF_MAC80211_KUNIT int
+ieee80211_calc_chandef_subchan_offset(const struct cfg80211_chan_def *ap,
+				      u8 n_partial_subchans)
+{
+	int n = ieee80211_chandef_num_subchans(ap);
+	struct cfg80211_chan_def tmp = *ap;
+	int offset = 0;
+
+	/*
+	 * Given a chandef (in this context, it's the AP's) and a number
+	 * of subchannels that we want to look at ('n_partial_subchans'),
+	 * calculate the offset in number of subchannels between the full
+	 * and the subset with the desired width.
+	 */
+
+	/* same number of subchannels means no offset, obviously */
+	if (n == n_partial_subchans)
+		return 0;
+
+	/* don't WARN - misconfigured APs could cause this if their N > width */
+	if (n < n_partial_subchans)
+		return 0;
+
+	while (ieee80211_chandef_num_subchans(&tmp) > n_partial_subchans) {
+		u32 prev = tmp.center_freq1;
+
+		ieee80211_chandef_downgrade(&tmp, NULL);
+
+		/*
+		 * if center_freq moved up, half the original channels
+		 * are gone now but were below, so increase offset
+		 */
+		if (prev < tmp.center_freq1)
+			offset += ieee80211_chandef_num_subchans(&tmp);
+	}
+
+	/*
+	 * 80+80 with secondary 80 below primary - four subchannels for it
+	 * (we cannot downgrade *to* 80+80, so no need to consider 'tmp')
+	 */
+	if (ap->width == NL80211_CHAN_WIDTH_80P80 &&
+	    ap->center_freq2 < ap->center_freq1)
+		offset += 4;
+
+	return offset;
+}
+EXPORT_SYMBOL_IF_MAC80211_KUNIT(ieee80211_calc_chandef_subchan_offset);
+
+VISIBLE_IF_MAC80211_KUNIT void
+ieee80211_rearrange_tpe_psd(struct ieee80211_parsed_tpe_psd *psd,
+			    const struct cfg80211_chan_def *ap,
+			    const struct cfg80211_chan_def *used)
+{
+	u8 needed = ieee80211_chandef_num_subchans(used);
+	u8 have = ieee80211_chandef_num_subchans(ap);
+	u8 tmp[IEEE80211_TPE_PSD_ENTRIES_320MHZ];
+	u8 offset;
+
+	if (!psd->valid)
+		return;
+
+	/* if N is zero, all defaults were used, no point in rearranging */
+	if (!psd->n)
+		goto out;
+
+	BUILD_BUG_ON(sizeof(tmp) != sizeof(psd->power));
+
+	/*
+	 * This assumes that 'N' is consistent with the HE channel, as
+	 * it should be (otherwise the AP is broken).
+	 *
+	 * In psd->power we have values in the order 0..N, 0..K, where
+	 * N+K should cover the entire channel per 'ap', but even if it
+	 * doesn't then we've pre-filled 'unlimited' as defaults.
+	 *
+	 * But this is all the wrong order, we want to have them in the
+	 * order of the 'used' channel.
+	 *
+	 * So for example, we could have a 320 MHz EHT AP, which has the
+	 * HE channel as 80 MHz (e.g. due to puncturing, which doesn't
+	 * seem to be considered for the TPE), as follows:
+	 *
+	 * EHT  320:   |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |
+	 * HE    80:                           |  |  |  |  |
+	 * used 160:                           |  |  |  |  |  |  |  |  |
+	 *
+	 * N entries:                          |--|--|--|--|
+	 * K entries:  |--|--|--|--|--|--|--|--|           |--|--|--|--|
+	 * power idx:   4  5  6  7  8  9  10 11 0  1  2  3  12 13 14 15
+	 * full chan:   0  1  2  3  4  5  6  7  8  9  10 11 12 13 14 15
+	 * used chan:                           0  1  2  3  4  5  6  7
+	 *
+	 * The idx in the power array ('power idx') is like this since it
+	 * comes directly from the element's N and K entries in their
+	 * element order, and those are this way for HE compatibility.
+	 *
+	 * Rearrange them as desired here, first by putting them into the
+	 * 'full chan' order, and then selecting the necessary subset for
+	 * the 'used chan'.
+	 */
+
+	/* first reorder according to AP channel */
+	offset = ieee80211_calc_chandef_subchan_offset(ap, psd->n);
+	for (int i = 0; i < have; i++) {
+		if (i < offset)
+			tmp[i] = psd->power[i + psd->n];
+		else if (i < offset + psd->n)
+			tmp[i] = psd->power[i - offset];
+		else
+			tmp[i] = psd->power[i];
+	}
+
+	/*
+	 * and then select the subset for the used channel
+	 * (set everything to defaults first in case a driver is confused)
+	 */
+	memset(psd->power, IEEE80211_TPE_PSD_NO_LIMIT, sizeof(psd->power));
+	offset = ieee80211_calc_chandef_subchan_offset(ap, needed);
+	for (int i = 0; i < needed; i++)
+		psd->power[i] = tmp[offset + i];
+
+out:
+	/* limit, but don't lie if there are defaults in the data */
+	if (needed < psd->count)
+		psd->count = needed;
+}
+EXPORT_SYMBOL_IF_MAC80211_KUNIT(ieee80211_rearrange_tpe_psd);
+
+static void ieee80211_rearrange_tpe(struct ieee80211_parsed_tpe *tpe,
+				    const struct cfg80211_chan_def *ap,
+				    const struct cfg80211_chan_def *used)
+{
+	/* ignore this completely for narrow/invalid channels */
+	if (!ieee80211_chandef_num_subchans(ap) ||
+	    !ieee80211_chandef_num_subchans(used)) {
+		ieee80211_clear_tpe(tpe);
+		return;
+	}
+
+	for (int i = 0; i < 2; i++) {
+		int needed_pwr_count;
+
+		ieee80211_rearrange_tpe_psd(&tpe->psd_local[i], ap, used);
+		ieee80211_rearrange_tpe_psd(&tpe->psd_reg_client[i], ap, used);
+
+		/* limit this to the widths we actually need */
+		needed_pwr_count = ieee80211_chandef_num_widths(used);
+		if (needed_pwr_count < tpe->max_local[i].count)
+			tpe->max_local[i].count = needed_pwr_count;
+		if (needed_pwr_count < tpe->max_reg_client[i].count)
+			tpe->max_reg_client[i].count = needed_pwr_count;
+	}
+}
+
 static struct ieee802_11_elems *
 ieee80211_determine_chan_mode(struct ieee80211_sub_if_data *sdata,
 			      struct ieee80211_conn_settings *conn,
 			      struct cfg80211_bss *cbss, int link_id,
-			      struct ieee80211_chan_req *chanreq)
+			      struct ieee80211_chan_req *chanreq,
+			      struct cfg80211_chan_def *ap_chandef)
 {
 	const struct cfg80211_bss_ies *ies = rcu_dereference(cbss->ies);
 	struct ieee80211_bss *bss = (void *)cbss->priv;
@@ -623,7 +807,6 @@ ieee80211_determine_chan_mode(struct ieee80211_sub_if_data *sdata,
 	};
 	struct ieee802_11_elems *elems;
 	struct ieee80211_supported_band *sband;
-	struct cfg80211_chan_def ap_chandef;
 	enum ieee80211_conn_mode ap_mode;
 	int ret;
 
@@ -634,7 +817,7 @@ again:
 		return ERR_PTR(-ENOMEM);
 
 	ap_mode = ieee80211_determine_ap_chan(sdata, channel, bss->vht_cap_info,
-					      elems, false, conn, &ap_chandef);
+					      elems, false, conn, ap_chandef);
 
 	/* this should be impossible since parsing depends on our mode */
 	if (WARN_ON(ap_mode > conn->mode)) {
@@ -701,12 +884,12 @@ again:
 		break;
 	}
 
-	chanreq->oper = ap_chandef;
+	chanreq->oper = *ap_chandef;
 
 	/* wider-bandwidth OFDMA is only done in EHT */
 	if (conn->mode >= IEEE80211_CONN_MODE_EHT &&
 	    !(sdata->vif.driver_flags & IEEE80211_VIF_IGNORE_OFDMA_WIDER_BW))
-		chanreq->ap = ap_chandef;
+		chanreq->ap = *ap_chandef;
 	else
 		chanreq->ap.chan = NULL;
 
@@ -738,7 +921,7 @@ again:
 				       IEEE80211_CONN_BW_LIMIT_160);
 	}
 
-	if (chanreq->oper.width != ap_chandef.width || ap_mode != conn->mode)
+	if (chanreq->oper.width != ap_chandef->width || ap_mode != conn->mode)
 		sdata_info(sdata,
 			   "regulatory prevented using AP config, downgraded\n");
 
@@ -3275,9 +3458,7 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata,
 
 	sdata->vif.bss_conf.power_type = IEEE80211_REG_UNSET_AP;
 	sdata->vif.bss_conf.pwr_reduction = 0;
-	sdata->vif.bss_conf.tx_pwr_env_num = 0;
-	memset(sdata->vif.bss_conf.tx_pwr_env, 0,
-	       sizeof(sdata->vif.bss_conf.tx_pwr_env));
+	ieee80211_clear_tpe(&sdata->vif.bss_conf.tpe);
 
 	sdata->vif.cfg.eml_cap = 0;
 	sdata->vif.cfg.eml_med_sync_delay = 0;
@@ -5018,15 +5199,15 @@ static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata,
 	struct ieee80211_local *local = sdata->local;
 	bool is_6ghz = cbss->channel->band == NL80211_BAND_6GHZ;
 	struct ieee80211_chan_req chanreq = {};
+	struct cfg80211_chan_def ap_chandef;
 	struct ieee802_11_elems *elems;
 	int ret;
-	u32 i;
 
 	lockdep_assert_wiphy(local->hw.wiphy);
 
 	rcu_read_lock();
 	elems = ieee80211_determine_chan_mode(sdata, conn, cbss, link_id,
-					      &chanreq);
+					      &chanreq, &ap_chandef);
 
 	if (IS_ERR(elems)) {
 		rcu_read_unlock();
@@ -5042,35 +5223,22 @@ static int ieee80211_prep_channel(struct ieee80211_sub_if_data *sdata,
 
 	if (link && is_6ghz && conn->mode >= IEEE80211_CONN_MODE_HE) {
 		const struct ieee80211_he_6ghz_oper *he_6ghz_oper;
-		struct ieee80211_bss_conf *bss_conf;
-		u8 j = 0;
-
-		bss_conf = link->conf;
 
 		if (elems->pwr_constr_elem)
-			bss_conf->pwr_reduction = *elems->pwr_constr_elem;
-
-		BUILD_BUG_ON(ARRAY_SIZE(bss_conf->tx_pwr_env) !=
-			     ARRAY_SIZE(elems->tx_pwr_env));
-
-		for (i = 0; i < elems->tx_pwr_env_num; i++) {
-			if (elems->tx_pwr_env_len[i] > sizeof(bss_conf->tx_pwr_env[j]))
-				continue;
-
-			bss_conf->tx_pwr_env_num++;
-			memcpy(&bss_conf->tx_pwr_env[j], elems->tx_pwr_env[i],
-			       elems->tx_pwr_env_len[i]);
-			j++;
-		}
+			link->conf->pwr_reduction = *elems->pwr_constr_elem;
 
 		he_6ghz_oper = ieee80211_he_6ghz_oper(elems->he_operation);
 		if (he_6ghz_oper)
-			bss_conf->power_type =
+			link->conf->power_type =
 				ieee80211_ap_power_type(he_6ghz_oper->control);
 		else
 			link_info(link,
 				  "HE 6 GHz operation missing (on %d MHz), expect issues\n",
 				  cbss->channel->center_freq);
+
+		link->conf->tpe = elems->tpe;
+		ieee80211_rearrange_tpe(&link->conf->tpe, &ap_chandef,
+					&chanreq.oper);
 	}
 	rcu_read_unlock();
 	/* the element data was RCU protected so no longer valid anyway */
@@ -7558,6 +7726,8 @@ void ieee80211_mgd_setup_link(struct ieee80211_link_data *link)
 	wiphy_delayed_work_init(&link->u.mgd.chswitch_work,
 				ieee80211_chswitch_work);
 
+	ieee80211_clear_tpe(&link->conf->tpe);
+
 	if (sdata->u.mgd.assoc_data)
 		ether_addr_copy(link->conf->addr,
 				sdata->u.mgd.assoc_data->link[link_id].addr);
diff --git a/net/mac80211/parse.c b/net/mac80211/parse.c
index 55e5497f8978..6efeb977f8e5 100644
--- a/net/mac80211/parse.c
+++ b/net/mac80211/parse.c
@@ -187,6 +187,84 @@ ieee80211_parse_extension_element(u32 *crc,
 		*crc = crc32_be(*crc, (void *)elem, elem->datalen + 2);
 }
 
+static void ieee80211_parse_tpe(struct ieee80211_parsed_tpe *tpe,
+				const u8 *data, u8 len)
+{
+	const struct ieee80211_tx_pwr_env *env = (const void *)data;
+	u8 count, interpret, category;
+	u8 *out, N, *cnt_out = NULL, *N_out = NULL;
+
+	if (!ieee80211_valid_tpe_element(data, len))
+		return;
+
+	count = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_COUNT);
+	interpret = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_INTERPRET);
+	category = u8_get_bits(env->info, IEEE80211_TX_PWR_ENV_INFO_CATEGORY);
+
+	switch (interpret) {
+	case IEEE80211_TPE_LOCAL_EIRP:
+		out = tpe->max_local[category].power;
+		cnt_out = &tpe->max_local[category].count;
+		tpe->max_local[category].valid = true;
+		break;
+	case IEEE80211_TPE_REG_CLIENT_EIRP:
+		out = tpe->max_reg_client[category].power;
+		cnt_out = &tpe->max_reg_client[category].count;
+		tpe->max_reg_client[category].valid = true;
+		break;
+	case IEEE80211_TPE_LOCAL_EIRP_PSD:
+		out = tpe->psd_local[category].power;
+		cnt_out = &tpe->psd_local[category].count;
+		N_out = &tpe->psd_local[category].n;
+		tpe->psd_local[category].valid = true;
+		break;
+	case IEEE80211_TPE_REG_CLIENT_EIRP_PSD:
+		out = tpe->psd_reg_client[category].power;
+		cnt_out = &tpe->psd_reg_client[category].count;
+		N_out = &tpe->psd_reg_client[category].n;
+		tpe->psd_reg_client[category].valid = true;
+		break;
+	}
+
+	switch (interpret) {
+	case IEEE80211_TPE_LOCAL_EIRP:
+	case IEEE80211_TPE_REG_CLIENT_EIRP:
+		/* count was validated <= 3, plus 320 MHz */
+		BUILD_BUG_ON(IEEE80211_TPE_EIRP_ENTRIES_320MHZ < 5);
+		memcpy(out, env->variable, count + 1);
+		*cnt_out = count + 1;
+		/* separately take 320 MHz if present */
+		if (count == 3 && len > sizeof(*env) + count + 1) {
+			out[4] = env->variable[count + 2];
+			*cnt_out = 5;
+		}
+		break;
+	case IEEE80211_TPE_LOCAL_EIRP_PSD:
+	case IEEE80211_TPE_REG_CLIENT_EIRP_PSD:
+		if (!count) {
+			memset(out, env->variable[0],
+			       IEEE80211_TPE_PSD_ENTRIES_320MHZ);
+			*cnt_out = IEEE80211_TPE_PSD_ENTRIES_320MHZ;
+			break;
+		}
+
+		N = 1 << (count - 1);
+		memcpy(out, env->variable, N);
+		*cnt_out = N;
+		*N_out = N;
+
+		if (len > sizeof(*env) + N) {
+			int K = u8_get_bits(env->variable[N],
+					    IEEE80211_TX_PWR_ENV_EXT_COUNT);
+
+			K = min(K, IEEE80211_TPE_PSD_ENTRIES_320MHZ - N);
+			memcpy(out + N, env->variable + N + 1, K);
+			(*cnt_out) += K;
+		}
+		break;
+	}
+}
+
 static u32
 _ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params,
 			     struct ieee80211_elems_parse *elems_parse,
@@ -593,16 +671,9 @@ _ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params,
 			elems->rsnx_len = elen;
 			break;
 		case WLAN_EID_TX_POWER_ENVELOPE:
-			if (elen < 1 ||
-			    elen > sizeof(struct ieee80211_tx_pwr_env))
-				break;
-
-			if (elems->tx_pwr_env_num >= ARRAY_SIZE(elems->tx_pwr_env))
+			if (params->mode < IEEE80211_CONN_MODE_HE)
 				break;
-
-			elems->tx_pwr_env[elems->tx_pwr_env_num] = (void *)pos;
-			elems->tx_pwr_env_len[elems->tx_pwr_env_num] = elen;
-			elems->tx_pwr_env_num++;
+			ieee80211_parse_tpe(&elems->tpe, pos, elen);
 			break;
 		case WLAN_EID_EXTENSION:
 			ieee80211_parse_extension_element(calc_crc ?
@@ -889,6 +960,9 @@ ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params)
 	elems->ie_start = params->start;
 	elems->total_len = params->len;
 
+	/* set all TPE entries to unlimited (but invalid) */
+	ieee80211_clear_tpe(&elems->tpe);
+
 	nontransmitted_profile = elems_parse->scratch_pos;
 	nontransmitted_profile_len =
 		ieee802_11_find_bssid_profile(params->start, params->len,
diff --git a/net/mac80211/tests/Makefile b/net/mac80211/tests/Makefile
index 4fdaf3feaca3..511dfa226699 100644
--- a/net/mac80211/tests/Makefile
+++ b/net/mac80211/tests/Makefile
@@ -1,3 +1,3 @@
-mac80211-tests-y += module.o elems.o mfp.o
+mac80211-tests-y += module.o elems.o mfp.o tpe.o
 
 obj-$(CONFIG_MAC80211_KUNIT_TEST) += mac80211-tests.o
diff --git a/net/mac80211/tests/tpe.c b/net/mac80211/tests/tpe.c
new file mode 100644
index 000000000000..dd63303a2985
--- /dev/null
+++ b/net/mac80211/tests/tpe.c
@@ -0,0 +1,284 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * KUnit tests for TPE element handling
+ *
+ * Copyright (C) 2024 Intel Corporation
+ */
+#include <kunit/test.h>
+#include "../ieee80211_i.h"
+
+MODULE_IMPORT_NS(EXPORTED_FOR_KUNIT_TESTING);
+
+static struct ieee80211_channel chan6g_1 = {
+	.band = NL80211_BAND_6GHZ,
+	.center_freq = 5955,
+};
+
+static struct ieee80211_channel chan6g_33 = {
+	.band = NL80211_BAND_6GHZ,
+	.center_freq = 6115,
+};
+
+static struct ieee80211_channel chan6g_61 = {
+	.band = NL80211_BAND_6GHZ,
+	.center_freq = 6255,
+};
+
+static const struct subchan_test_case {
+	const char *desc;
+	struct cfg80211_chan_def c;
+	u8 n;
+	int expect;
+} subchan_offset_cases[] = {
+	{
+		.desc = "identical 20 MHz",
+		.c.width = NL80211_CHAN_WIDTH_20,
+		.c.chan = &chan6g_1,
+		.c.center_freq1 = 5955,
+		.n = 1,
+		.expect = 0,
+	},
+	{
+		.desc = "identical 40 MHz",
+		.c.width = NL80211_CHAN_WIDTH_40,
+		.c.chan = &chan6g_1,
+		.c.center_freq1 = 5965,
+		.n = 2,
+		.expect = 0,
+	},
+	{
+		.desc = "identical 80+80 MHz",
+		/* not really is valid? doesn't matter for the test */
+		.c.width = NL80211_CHAN_WIDTH_80P80,
+		.c.chan = &chan6g_1,
+		.c.center_freq1 = 5985,
+		.c.center_freq2 = 6225,
+		.n = 16,
+		.expect = 0,
+	},
+	{
+		.desc = "identical 320 MHz",
+		.c.width = NL80211_CHAN_WIDTH_320,
+		.c.chan = &chan6g_1,
+		.c.center_freq1 = 6105,
+		.n = 16,
+		.expect = 0,
+	},
+	{
+		.desc = "lower 160 MHz of 320 MHz",
+		.c.width = NL80211_CHAN_WIDTH_320,
+		.c.chan = &chan6g_1,
+		.c.center_freq1 = 6105,
+		.n = 8,
+		.expect = 0,
+	},
+	{
+		.desc = "upper 160 MHz of 320 MHz",
+		.c.width = NL80211_CHAN_WIDTH_320,
+		.c.chan = &chan6g_61,
+		.c.center_freq1 = 6105,
+		.n = 8,
+		.expect = 8,
+	},
+	{
+		.desc = "upper 160 MHz of 320 MHz, go to 40",
+		.c.width = NL80211_CHAN_WIDTH_320,
+		.c.chan = &chan6g_61,
+		.c.center_freq1 = 6105,
+		.n = 2,
+		.expect = 8 + 4 + 2,
+	},
+	{
+		.desc = "secondary 80 above primary in 80+80 MHz",
+		/* not really is valid? doesn't matter for the test */
+		.c.width = NL80211_CHAN_WIDTH_80P80,
+		.c.chan = &chan6g_1,
+		.c.center_freq1 = 5985,
+		.c.center_freq2 = 6225,
+		.n = 4,
+		.expect = 0,
+	},
+	{
+		.desc = "secondary 80 below primary in 80+80 MHz",
+		/* not really is valid? doesn't matter for the test */
+		.c.width = NL80211_CHAN_WIDTH_80P80,
+		.c.chan = &chan6g_61,
+		.c.center_freq1 = 6225,
+		.c.center_freq2 = 5985,
+		.n = 4,
+		.expect = 4,
+	},
+	{
+		.desc = "secondary 80 below primary in 80+80 MHz, go to 20",
+		/* not really is valid? doesn't matter for the test */
+		.c.width = NL80211_CHAN_WIDTH_80P80,
+		.c.chan = &chan6g_61,
+		.c.center_freq1 = 6225,
+		.c.center_freq2 = 5985,
+		.n = 1,
+		.expect = 7,
+	},
+};
+
+KUNIT_ARRAY_PARAM_DESC(subchan_offset, subchan_offset_cases, desc);
+
+static void subchan_offset(struct kunit *test)
+{
+	const struct subchan_test_case *params = test->param_value;
+	int offset;
+
+	KUNIT_ASSERT_EQ(test, cfg80211_chandef_valid(&params->c), true);
+
+	offset = ieee80211_calc_chandef_subchan_offset(&params->c, params->n);
+
+	KUNIT_EXPECT_EQ(test, params->expect, offset);
+}
+
+static const struct psd_reorder_test_case {
+	const char *desc;
+	struct cfg80211_chan_def ap, used;
+	struct ieee80211_parsed_tpe_psd psd, out;
+} psd_reorder_cases[] = {
+	{
+		.desc = "no changes, 320 MHz",
+
+		.ap.width = NL80211_CHAN_WIDTH_320,
+		.ap.chan = &chan6g_1,
+		.ap.center_freq1 = 6105,
+
+		.used.width = NL80211_CHAN_WIDTH_320,
+		.used.chan = &chan6g_1,
+		.used.center_freq1 = 6105,
+
+		.psd.valid = true,
+		.psd.count = 16,
+		.psd.n = 8,
+		.psd.power = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 },
+
+		.out.valid = true,
+		.out.count = 16,
+		.out.n = 8,
+		.out.power = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 },
+	},
+	{
+		.desc = "no changes, 320 MHz, 160 MHz used, n=0",
+
+		.ap.width = NL80211_CHAN_WIDTH_320,
+		.ap.chan = &chan6g_1,
+		.ap.center_freq1 = 6105,
+
+		.used.width = NL80211_CHAN_WIDTH_160,
+		.used.chan = &chan6g_1,
+		.used.center_freq1 = 6025,
+
+		.psd.valid = true,
+		.psd.count = 16,
+		.psd.n = 0,
+		.psd.power = { 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, },
+
+		.out.valid = true,
+		.out.count = 8,
+		.out.n = 0,
+		.out.power = { 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, },
+	},
+	{
+		.desc = "320 MHz, HE is 80, used 160, all lower",
+
+		.ap.width = NL80211_CHAN_WIDTH_320,
+		.ap.chan = &chan6g_1,
+		.ap.center_freq1 = 6105,
+
+		.used.width = NL80211_CHAN_WIDTH_160,
+		.used.chan = &chan6g_1,
+		.used.center_freq1 = 6025,
+
+		.psd.valid = true,
+		.psd.count = 16,
+		.psd.n = 4,
+		.psd.power = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 },
+
+		.out.valid = true,
+		.out.count = 8,
+		.out.n = 4,
+		.out.power = { 0, 1, 2, 3, 4, 5, 6, 7, 127, 127, 127, 127, 127, 127, 127, 127},
+	},
+	{
+		.desc = "320 MHz, HE is 80, used 160, all upper",
+		/*
+		 * EHT: | | | | | | | | | | | | | | | | |
+		 * HE:                          | | | | |
+		 * used:                | | | | | | | | |
+		 */
+
+		.ap.width = NL80211_CHAN_WIDTH_320,
+		.ap.chan = &chan6g_61,
+		.ap.center_freq1 = 6105,
+
+		.used.width = NL80211_CHAN_WIDTH_160,
+		.used.chan = &chan6g_61,
+		.used.center_freq1 = 6185,
+
+		.psd.valid = true,
+		.psd.count = 16,
+		.psd.n = 4,
+		.psd.power = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 },
+
+		.out.valid = true,
+		.out.count = 8,
+		.out.n = 4,
+		.out.power = { 12, 13, 14, 15, 0, 1, 2, 3, 127, 127, 127, 127, 127, 127, 127, 127},
+	},
+	{
+		.desc = "320 MHz, HE is 80, used 160, split",
+		/*
+		 * EHT: | | | | | | | | | | | | | | | | |
+		 * HE:                  | | | | |
+		 * used:                | | | | | | | | |
+		 */
+
+		.ap.width = NL80211_CHAN_WIDTH_320,
+		.ap.chan = &chan6g_33,
+		.ap.center_freq1 = 6105,
+
+		.used.width = NL80211_CHAN_WIDTH_160,
+		.used.chan = &chan6g_33,
+		.used.center_freq1 = 6185,
+
+		.psd.valid = true,
+		.psd.count = 16,
+		.psd.n = 4,
+		.psd.power = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 },
+
+		.out.valid = true,
+		.out.count = 8,
+		.out.n = 4,
+		.out.power = { 0, 1, 2, 3, 12, 13, 14, 15, 127, 127, 127, 127, 127, 127, 127, 127},
+	},
+};
+
+KUNIT_ARRAY_PARAM_DESC(psd_reorder, psd_reorder_cases, desc);
+
+static void psd_reorder(struct kunit *test)
+{
+	const struct psd_reorder_test_case *params = test->param_value;
+	struct ieee80211_parsed_tpe_psd tmp = params->psd;
+
+	KUNIT_ASSERT_EQ(test, cfg80211_chandef_valid(&params->ap), true);
+	KUNIT_ASSERT_EQ(test, cfg80211_chandef_valid(&params->used), true);
+
+	ieee80211_rearrange_tpe_psd(&tmp, &params->ap, &params->used);
+	KUNIT_EXPECT_MEMEQ(test, &tmp, &params->out, sizeof(tmp));
+}
+
+static struct kunit_case tpe_test_cases[] = {
+	KUNIT_CASE_PARAM(subchan_offset, subchan_offset_gen_params),
+	KUNIT_CASE_PARAM(psd_reorder, psd_reorder_gen_params),
+	{}
+};
+
+static struct kunit_suite tpe = {
+	.name = "mac80211-tpe",
+	.test_cases = tpe_test_cases,
+};
+
+kunit_test_suite(tpe);
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 6139c930b572..183de2075fb9 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -4334,3 +4334,28 @@ ieee80211_min_bw_limit_from_chandef(struct cfg80211_chan_def *chandef)
 		return IEEE80211_CONN_BW_LIMIT_20;
 	}
 }
+
+void ieee80211_clear_tpe(struct ieee80211_parsed_tpe *tpe)
+{
+	for (int i = 0; i < 2; i++) {
+		tpe->max_local[i].valid = false;
+		memset(tpe->max_local[i].power,
+		       IEEE80211_TPE_MAX_TX_PWR_NO_CONSTRAINT,
+		       sizeof(tpe->max_local[i].power));
+
+		tpe->max_reg_client[i].valid = false;
+		memset(tpe->max_reg_client[i].power,
+		       IEEE80211_TPE_MAX_TX_PWR_NO_CONSTRAINT,
+		       sizeof(tpe->max_reg_client[i].power));
+
+		tpe->psd_local[i].valid = false;
+		memset(tpe->psd_local[i].power,
+		       IEEE80211_TPE_PSD_NO_LIMIT,
+		       sizeof(tpe->psd_local[i].power));
+
+		tpe->psd_reg_client[i].valid = false;
+		memset(tpe->psd_reg_client[i].power,
+		       IEEE80211_TPE_PSD_NO_LIMIT,
+		       sizeof(tpe->psd_reg_client[i].power));
+	}
+}
-- 
cgit v1.2.3-58-ga151


From 5a009b42e0418d30b3ffaff2f46c534cd79b3f23 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:37:57 +0200
Subject: wifi: mac80211: track changes in AP's TPE

If the TPE (transmit power envelope) is changed, detect and
report that to the driver.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506214536.103dda923f45.I990877e409ab8eade9ed7c172272e0cae57256cf@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h |  2 ++
 net/mac80211/mlme.c    | 16 ++++++++++++++--
 2 files changed, 16 insertions(+), 2 deletions(-)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index a4efbfb8d796..a59eacfe0480 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -362,6 +362,7 @@ struct ieee80211_vif_chanctx_switch {
  *	status changed.
  * @BSS_CHANGED_MLD_VALID_LINKS: MLD valid links status changed.
  * @BSS_CHANGED_MLD_TTLM: negotiated TID to link mapping was changed
+ * @BSS_CHANGED_TPE: transmit power envelope changed
  */
 enum ieee80211_bss_change {
 	BSS_CHANGED_ASSOC		= 1<<0,
@@ -398,6 +399,7 @@ enum ieee80211_bss_change {
 	BSS_CHANGED_UNSOL_BCAST_PROBE_RESP = 1<<31,
 	BSS_CHANGED_MLD_VALID_LINKS	= BIT_ULL(33),
 	BSS_CHANGED_MLD_TTLM		= BIT_ULL(34),
+	BSS_CHANGED_TPE			= BIT_ULL(35),
 
 	/* when adding here, make sure to change ieee80211_reconfig */
 };
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index ed9851faac05..08c0999746fb 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -973,6 +973,7 @@ static int ieee80211_config_bw(struct ieee80211_link_data *link,
 	struct ieee80211_channel *channel = link->conf->chanreq.oper.chan;
 	struct ieee80211_sub_if_data *sdata = link->sdata;
 	struct ieee80211_chan_req chanreq = {};
+	struct cfg80211_chan_def ap_chandef;
 	enum ieee80211_conn_mode ap_mode;
 	u32 vht_cap_info = 0;
 	u16 ht_opmode;
@@ -988,7 +989,7 @@ static int ieee80211_config_bw(struct ieee80211_link_data *link,
 
 	ap_mode = ieee80211_determine_ap_chan(sdata, channel, vht_cap_info,
 					      elems, true, &link->u.mgd.conn,
-					      &chanreq.ap);
+					      &ap_chandef);
 
 	if (ap_mode != link->u.mgd.conn.mode) {
 		link_info(link,
@@ -998,7 +999,8 @@ static int ieee80211_config_bw(struct ieee80211_link_data *link,
 		return -EINVAL;
 	}
 
-	chanreq.oper = chanreq.ap;
+	chanreq.ap = ap_chandef;
+	chanreq.oper = ap_chandef;
 	if (link->u.mgd.conn.mode < IEEE80211_CONN_MODE_EHT ||
 	    sdata->vif.driver_flags & IEEE80211_VIF_IGNORE_OFDMA_WIDER_BW)
 		chanreq.ap.chan = NULL;
@@ -1026,6 +1028,16 @@ static int ieee80211_config_bw(struct ieee80211_link_data *link,
 			ieee80211_min_bw_limit_from_chandef(&chanreq.oper))
 		ieee80211_chandef_downgrade(&chanreq.oper, NULL);
 
+	if (ap_chandef.chan->band == NL80211_BAND_6GHZ &&
+	    link->u.mgd.conn.mode >= IEEE80211_CONN_MODE_HE) {
+		ieee80211_rearrange_tpe(&elems->tpe, &ap_chandef,
+					&chanreq.oper);
+		if (memcmp(&link->conf->tpe, &elems->tpe, sizeof(elems->tpe))) {
+			link->conf->tpe = elems->tpe;
+			*changed |= BSS_CHANGED_TPE;
+		}
+	}
+
 	if (ieee80211_chanreq_identical(&chanreq, &link->conf->chanreq))
 		return 0;
 
-- 
cgit v1.2.3-58-ga151


From 32fca7e3820cc915d1d47e33b13da8124208b04c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 10 May 2024 17:06:28 +0300
Subject: wifi: iwlwifi: dvm: clean up rs_get_rate() logic

The logic here is rather convoluted - we cannot get here with
lq_sta being NULL as mac80211 will (no longer) call us like
that, and since I removed the rate_control_send_low() call in
this function there's no longer any point in setting priv_sta
to NULL either.

So the only thing that remains to check is if we have actually
initialized our lq_sta->drv pointer, and exit if we didn't in
which case we'll use the data mac80211 already set up for the
low rate usage.

Reviewed-by: Luciano Coelho <luciano.coelho@intel.com>
Link: https://msgid.link/20240510170500.a4cdb41825eb.Id202bcc967c32829f70ab1412f8893b6eb7f78e2@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/dvm/rs.c | 21 ++++++++-------------
 1 file changed, 8 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/rs.c b/drivers/net/wireless/intel/iwlwifi/dvm/rs.c
index f4a6f76cf193..8879e668ef0d 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/rs.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/rs.c
@@ -2673,20 +2673,16 @@ static void rs_get_rate(void *priv_r, struct ieee80211_sta *sta, void *priv_sta,
 	IWL_DEBUG_RATE_LIMIT(priv, "rate scale calculate new rate for skb\n");
 
 	/* Get max rate if user set max rate */
-	if (lq_sta) {
-		lq_sta->max_rate_idx = fls(txrc->rate_idx_mask) - 1;
-		if ((sband->band == NL80211_BAND_5GHZ) &&
-		    (lq_sta->max_rate_idx != -1))
-			lq_sta->max_rate_idx += IWL_FIRST_OFDM_RATE;
-		if ((lq_sta->max_rate_idx < 0) ||
-		    (lq_sta->max_rate_idx >= IWL_RATE_COUNT))
-			lq_sta->max_rate_idx = -1;
-	}
+	lq_sta->max_rate_idx = fls(txrc->rate_idx_mask) - 1;
+	if (sband->band == NL80211_BAND_5GHZ && lq_sta->max_rate_idx != -1)
+		lq_sta->max_rate_idx += IWL_FIRST_OFDM_RATE;
+	if (lq_sta->max_rate_idx < 0 || lq_sta->max_rate_idx >= IWL_RATE_COUNT)
+		lq_sta->max_rate_idx = -1;
 
-	/* Treat uninitialized rate scaling data same as non-existing. */
-	if (lq_sta && !lq_sta->drv) {
+	if (!lq_sta->drv) {
 		IWL_DEBUG_RATE(priv, "Rate scaling not initialized yet.\n");
-		priv_sta = NULL;
+		/* mac80211 already set up the data for using low rates */
+		return;
 	}
 
 	rate_idx  = lq_sta->last_txrate_idx;
@@ -2756,7 +2752,6 @@ void iwl_rs_rate_init(struct iwl_priv *priv, struct ieee80211_sta *sta, u8 sta_i
 	lq_sta = &sta_priv->lq_sta;
 	sband = hw->wiphy->bands[conf->chandef.chan->band];
 
-
 	lq_sta->lq.sta_id = sta_id;
 
 	for (j = 0; j < LQ_SIZE; j++)
-- 
cgit v1.2.3-58-ga151


From 7aea71803777cddf17bbb1d4e7cd1c45df0dd47a Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Fri, 10 May 2024 17:06:30 +0300
Subject: wifi: iwlwifi: mvm: simplify the uAPSD coexistence limitation code

uAPSD can't be enabled on BSS or P2P Client if there is
another active interface (e.g. associated client). Allow
the code that apply those limitations to run on BSS as well.
Also forbid uAPSD if we have an IBSS or NAN interface in
the system.
Since we are now forbidding uAPSD in any concurrency
scenario, we don't need to check the number of PHY
contexts.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240510170500.4ac26534adf7.Ib892020177e86603b2be3a105f8717b9a15951d3@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/power.c | 55 +++++++++++++-------------
 1 file changed, 27 insertions(+), 28 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index 568f53c56199..72f4480ae7f5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -211,19 +211,29 @@ static void iwl_mvm_power_configure_uapsd(struct iwl_mvm *mvm,
 		IWL_MVM_PS_HEAVY_RX_THLD_PERCENT;
 }
 
-static void iwl_mvm_p2p_standalone_iterator(void *_data, u8 *mac,
-					    struct ieee80211_vif *vif)
+struct iwl_allow_uapsd_iface_iterator_data {
+	struct ieee80211_vif *exclude_vif;
+	bool allow_uapsd;
+};
+
+static void iwl_mvm_allow_uapsd_iterator(void *_data, u8 *mac,
+					 struct ieee80211_vif *vif)
 {
-	bool *is_p2p_standalone = _data;
+	struct iwl_allow_uapsd_iface_iterator_data *data = _data;
 
-	switch (ieee80211_vif_type_p2p(vif)) {
-	case NL80211_IFTYPE_P2P_GO:
+	/* exclude the given vif */
+	if (vif == data->exclude_vif)
+		return;
+
+	switch (vif->type) {
 	case NL80211_IFTYPE_AP:
-		*is_p2p_standalone = false;
+	case NL80211_IFTYPE_ADHOC:
+	case NL80211_IFTYPE_NAN:
+		data->allow_uapsd = false;
 		break;
 	case NL80211_IFTYPE_STATION:
 		if (vif->cfg.assoc)
-			*is_p2p_standalone = false;
+			data->allow_uapsd = false;
 		break;
 
 	default:
@@ -235,6 +245,10 @@ static bool iwl_mvm_power_allow_uapsd(struct iwl_mvm *mvm,
 				       struct ieee80211_vif *vif)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_allow_uapsd_iface_iterator_data data = {
+		.exclude_vif = vif,
+		.allow_uapsd = true,
+	};
 
 	if (ether_addr_equal(mvmvif->uapsd_misbehaving_ap_addr,
 			     vif->cfg.ap_addr))
@@ -249,30 +263,15 @@ static bool iwl_mvm_power_allow_uapsd(struct iwl_mvm *mvm,
 	    IEEE80211_P2P_OPPPS_ENABLE_BIT))
 		return false;
 
-	/*
-	 * Avoid using uAPSD if client is in DCM -
-	 * low latency issue in Miracast
-	 */
-	if (iwl_mvm_phy_ctx_count(mvm) >= 2)
+	if (vif->p2p && !iwl_mvm_is_p2p_scm_uapsd_supported(mvm))
 		return false;
 
-	if (vif->p2p) {
-		/* Allow U-APSD only if p2p is stand alone */
-		bool is_p2p_standalone = true;
-
-		if (!iwl_mvm_is_p2p_scm_uapsd_supported(mvm))
-			return false;
-
-		ieee80211_iterate_active_interfaces_atomic(mvm->hw,
-					IEEE80211_IFACE_ITER_NORMAL,
-					iwl_mvm_p2p_standalone_iterator,
-					&is_p2p_standalone);
-
-		if (!is_p2p_standalone)
-			return false;
-	}
+	ieee80211_iterate_active_interfaces_atomic(mvm->hw,
+				IEEE80211_IFACE_ITER_NORMAL,
+				iwl_mvm_allow_uapsd_iterator,
+				&data);
 
-	return true;
+	return data.allow_uapsd;
 }
 
 static bool iwl_mvm_power_is_radar(struct ieee80211_vif *vif)
-- 
cgit v1.2.3-58-ga151


From 42884d2c2f512d7123962ab0425c782316f3805f Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Fri, 10 May 2024 17:06:31 +0300
Subject: wifi: iwlwifi: mvm: allow UAPSD when in SCM

Allow UAPSD when P2P and BSS interfaces share the same channel.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Reviewed-by: Luciano Coelho <luciano.coelho@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240510170500.1ec19d432497.Id6de5337c878fa70e85bfcf6f4e0e34ce60756fb@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/power.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index 72f4480ae7f5..cc7176b0be4f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -212,7 +212,7 @@ static void iwl_mvm_power_configure_uapsd(struct iwl_mvm *mvm,
 }
 
 struct iwl_allow_uapsd_iface_iterator_data {
-	struct ieee80211_vif *exclude_vif;
+	struct ieee80211_vif *current_vif;
 	bool allow_uapsd;
 };
 
@@ -220,9 +220,12 @@ static void iwl_mvm_allow_uapsd_iterator(void *_data, u8 *mac,
 					 struct ieee80211_vif *vif)
 {
 	struct iwl_allow_uapsd_iface_iterator_data *data = _data;
+	struct iwl_mvm_vif *other_mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_vif *curr_mvmvif =
+		iwl_mvm_vif_from_mac80211(data->current_vif);
 
 	/* exclude the given vif */
-	if (vif == data->exclude_vif)
+	if (vif == data->current_vif)
 		return;
 
 	switch (vif->type) {
@@ -232,7 +235,12 @@ static void iwl_mvm_allow_uapsd_iterator(void *_data, u8 *mac,
 		data->allow_uapsd = false;
 		break;
 	case NL80211_IFTYPE_STATION:
-		if (vif->cfg.assoc)
+		/* allow UAPSD if P2P interface and BSS station interface share
+		 * the same channel.
+		 */
+		if (vif->cfg.assoc && other_mvmvif->deflink.phy_ctxt &&
+		    curr_mvmvif->deflink.phy_ctxt &&
+		    other_mvmvif->deflink.phy_ctxt->id != curr_mvmvif->deflink.phy_ctxt->id)
 			data->allow_uapsd = false;
 		break;
 
@@ -246,7 +254,7 @@ static bool iwl_mvm_power_allow_uapsd(struct iwl_mvm *mvm,
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_allow_uapsd_iface_iterator_data data = {
-		.exclude_vif = vif,
+		.current_vif = vif,
 		.allow_uapsd = true,
 	};
 
-- 
cgit v1.2.3-58-ga151


From 8a29b006dc39484624f841c1e1c30a64a523bbc8 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Fri, 10 May 2024 17:06:32 +0300
Subject: wifi: iwlwifi: mvm: debugfs: add entry for setting maximum TXOP time

Add an entry for setting the maximum TXOP time in microseconds.
The configured value can be read from the same entry.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Reviewed-by: Gregory Greenman <gregory.greenman@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240510170500.b8d17a9c1731.I57c33574a61edd68bd0ec1aa7009f31111fd7efe@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/mvm/debugfs-vif.c   | 38 ++++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  3 ++
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c     |  3 ++
 3 files changed, 44 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
index 17c97dfbc62a..25f07e00db42 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs-vif.c
@@ -692,6 +692,42 @@ static ssize_t iwl_dbgfs_quota_min_read(struct file *file,
 	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
 }
 
+static ssize_t iwl_dbgfs_max_tx_op_write(struct ieee80211_vif *vif, char *buf,
+					 size_t count, loff_t *ppos)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm *mvm = mvmvif->mvm;
+	u16 value;
+	int ret;
+
+	ret = kstrtou16(buf, 0, &value);
+	if (ret)
+		return ret;
+
+	mutex_lock(&mvm->mutex);
+	mvmvif->max_tx_op = value;
+	mutex_unlock(&mvm->mutex);
+
+	return count;
+}
+
+static ssize_t iwl_dbgfs_max_tx_op_read(struct file *file,
+					char __user *user_buf,
+					size_t count, loff_t *ppos)
+{
+	struct ieee80211_vif *vif = file->private_data;
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm *mvm = mvmvif->mvm;
+	char buf[10];
+	int len;
+
+	mutex_lock(&mvm->mutex);
+	len = scnprintf(buf, sizeof(buf), "%hu\n", mvmvif->max_tx_op);
+	mutex_unlock(&mvm->mutex);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
+}
+
 static ssize_t iwl_dbgfs_int_mlo_scan_write(struct ieee80211_vif *vif,
 					    char *buf, size_t count,
 					    loff_t *ppos)
@@ -801,6 +837,7 @@ MVM_DEBUGFS_READ_WRITE_FILE_OPS(uapsd_misbehaving, 20);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(rx_phyinfo, 10);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(quota_min, 32);
 MVM_DEBUGFS_READ_FILE_OPS(os_device_timediff);
+MVM_DEBUGFS_READ_WRITE_FILE_OPS(max_tx_op, 10);
 MVM_DEBUGFS_WRITE_FILE_OPS(int_mlo_scan, 32);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(esr_disable_reason, 32);
 
@@ -830,6 +867,7 @@ void iwl_mvm_vif_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 	MVM_DEBUGFS_ADD_FILE_VIF(rx_phyinfo, mvmvif->dbgfs_dir, 0600);
 	MVM_DEBUGFS_ADD_FILE_VIF(quota_min, mvmvif->dbgfs_dir, 0600);
 	MVM_DEBUGFS_ADD_FILE_VIF(os_device_timediff, mvmvif->dbgfs_dir, 0400);
+	MVM_DEBUGFS_ADD_FILE_VIF(max_tx_op, mvmvif->dbgfs_dir, 0600);
 	debugfs_create_bool("ftm_unprotected", 0200, mvmvif->dbgfs_dir,
 			    &mvmvif->ftm_unprotected);
 	MVM_DEBUGFS_ADD_FILE_VIF(int_mlo_scan, mvmvif->dbgfs_dir, 0200);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 1f58c727fa63..ca1c5944da67 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -426,6 +426,7 @@ struct iwl_mvm_esr_exit {
  * @csa_bcn_pending: indicates that we are waiting for a beacon on a new channel
  * @csa_blocks_tx: CSA is blocking TX
  * @features: hw features active for this vif
+ * @max_tx_op: max TXOP in usecs for all ACs, zero for no limit.
  * @ap_beacon_time: AP beacon time for synchronisation (on older FW)
  * @bf_enabled: indicates if beacon filtering is enabled
  * @ba_enabled: indicated if beacon abort is enabled
@@ -538,6 +539,8 @@ struct iwl_mvm_vif {
 		struct ieee80211_key_conf __rcu *keys[2];
 	} bcn_prot;
 
+	u16 max_tx_op;
+
 	u16 link_selection_res;
 	u8 link_selection_primary;
 	u8 primary_link;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index 3ba62fb2c85e..884fd56e7426 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -609,6 +609,7 @@ void iwl_mvm_rs_fw_rate_init(struct iwl_mvm *mvm,
 				cpu_to_le16(max_amsdu_len) : 0,
 	};
 	unsigned int link_id = link_conf->link_id;
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(mvmsta->vif);
 	int cmd_ver;
 	int ret;
 
@@ -654,6 +655,8 @@ void iwl_mvm_rs_fw_rate_init(struct iwl_mvm *mvm,
 	 */
 	sta->deflink.agg.max_amsdu_len = max_amsdu_len;
 
+	cfg_cmd.max_tx_op = cpu_to_le16(mvmvif->max_tx_op);
+
 	cmd_ver = iwl_fw_lookup_cmd_ver(mvm->fw, cmd_id, 0);
 	IWL_DEBUG_RATE(mvm, "TLC CONFIG CMD, sta_id=%d, max_ch_width=%d, mode=%d\n",
 		       cfg_cmd.sta_id, cfg_cmd.max_ch_width, cfg_cmd.mode);
-- 
cgit v1.2.3-58-ga151


From 7c789b91f538b3a31b1512b46873174fd3960fc3 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Fri, 10 May 2024 17:06:34 +0300
Subject: wifi: iwlwifi: mvm: define link_sta in the relevant scope

The link_sta pointer is now only used inside this if, so define it inside
its scope.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240510170500.e0a1dd285d9a.Id5ff16fb98af0c5f533aed6b5ec3cc856401e488@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index d78af2928152..8bbef4454399 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -2032,7 +2032,6 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 	u32 len;
 	u32 pkt_len = iwl_rx_packet_payload_len(pkt);
 	struct ieee80211_sta *sta = NULL;
-	struct ieee80211_link_sta *link_sta = NULL;
 	struct sk_buff *skb;
 	u8 crypt_len = 0;
 	u8 sta_id = le32_get_bits(desc->status, IWL_RX_MPDU_STATUS_STA_ID);
@@ -2185,6 +2184,8 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 
 	if (desc->status & cpu_to_le32(IWL_RX_MPDU_STATUS_SRC_STA_FOUND)) {
 		if (!WARN_ON_ONCE(sta_id >= mvm->fw->ucode_capa.num_stations)) {
+			struct ieee80211_link_sta *link_sta;
+
 			sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]);
 			if (IS_ERR(sta))
 				sta = NULL;
-- 
cgit v1.2.3-58-ga151


From bb3261943fd8b4b0d21c9f609eb42b6c948e4578 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Fri, 10 May 2024 17:06:36 +0300
Subject: wifi: iwlwifi: mvm: set A-MSDU size on the correct link

The existing code sets the max A-MSDU size to the deflink, set it
on the given link instead.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240510170500.4de5c1a031ca.I063bc1508e5068b32e321b0f268bc51dba0c0747@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index 884fd56e7426..031b5e70d181 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -653,7 +653,7 @@ void iwl_mvm_rs_fw_rate_init(struct iwl_mvm *mvm,
 	 * since TLC offload works with one mode we can assume
 	 * that only vht/ht is used and also set it as station max amsdu
 	 */
-	sta->deflink.agg.max_amsdu_len = max_amsdu_len;
+	link_sta->agg.max_amsdu_len = max_amsdu_len;
 
 	cfg_cmd.max_tx_op = cpu_to_le16(mvmvif->max_tx_op);
 
-- 
cgit v1.2.3-58-ga151


From b79b56091d2456491f4471bea202453f9755a42d Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Fri, 10 May 2024 17:06:37 +0300
Subject: wifi: iwlwifi: mvm: call ieee80211_sta_recalc_aggregates on A-MSDU
 size update

The driver needs to call this function when the A-MSDU size changes.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240510170500.632ee6612a79.Ice3a536a4cbfe60d0edfa231fcb79ee7ab9495dd@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
index 031b5e70d181..05715e5af6ab 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs-fw.c
@@ -514,6 +514,8 @@ void iwl_mvm_tlc_update_notif(struct iwl_mvm *mvm,
 				link_sta->agg.max_tid_amsdu_len[i] = 1;
 		}
 
+		ieee80211_sta_recalc_aggregates(sta);
+
 		IWL_DEBUG_RATE(mvm,
 			       "AMSDU update. AMSDU size: %d, AMSDU selected size: %d, AMSDU TID bitmap 0x%X\n",
 			       le32_to_cpu(notif->amsdu_size), size,
@@ -654,6 +656,7 @@ void iwl_mvm_rs_fw_rate_init(struct iwl_mvm *mvm,
 	 * that only vht/ht is used and also set it as station max amsdu
 	 */
 	link_sta->agg.max_amsdu_len = max_amsdu_len;
+	ieee80211_sta_recalc_aggregates(sta);
 
 	cfg_cmd.max_tx_op = cpu_to_le16(mvmvif->max_tx_op);
 
-- 
cgit v1.2.3-58-ga151


From d49c5be9f7fe4864f698c15bf332524f2be3ee69 Mon Sep 17 00:00:00 2001
From: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Date: Fri, 10 May 2024 17:06:38 +0300
Subject: wifi: iwlwifi: mvm: Don't set NO_HT40+/- flags on 6 GHz band

These flags are not needed on 6 GHz channels, and anyway they were set
incorrectly.
This in turn resulted in alternating channel flags, preventing reg domain
rules to be merged together, so cfg80211 couldn't even send them to the
user space, as the regulatory domain was too large.
Fix it.

Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com>
Reviewed-by: Luciano Coelho <luciano.coelho@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240510170500.b0d51b2229f8.I092e21cde43320ffc2eff17f5748ff9c87c87fcf@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index 149903f52567..cf5c9500af8c 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -1625,11 +1625,15 @@ static u32 iwl_nvm_get_regdom_bw_flags(const u16 *nvm_chan,
 			flags &= ~NL80211_RRF_NO_HT40PLUS;
 		if (nvm_chan[ch_idx] >= FIRST_2GHZ_HT_MINUS)
 			flags &= ~NL80211_RRF_NO_HT40MINUS;
-	} else if (nvm_flags & NVM_CHANNEL_40MHZ) {
+	} else if (ch_idx < NUM_2GHZ_CHANNELS + NUM_5GHZ_CHANNELS &&
+		   nvm_flags & NVM_CHANNEL_40MHZ) {
 		if ((ch_idx - NUM_2GHZ_CHANNELS) % 2 == 0)
 			flags &= ~NL80211_RRF_NO_HT40PLUS;
 		else
 			flags &= ~NL80211_RRF_NO_HT40MINUS;
+	} else if (nvm_flags & NVM_CHANNEL_40MHZ) {
+		flags &= ~NL80211_RRF_NO_HT40PLUS;
+		flags &= ~NL80211_RRF_NO_HT40MINUS;
 	}
 
 	if (!(nvm_flags & NVM_CHANNEL_80MHZ))
-- 
cgit v1.2.3-58-ga151


From 37733bffda3285d18bd1d72c14b3a1cf39c56a5e Mon Sep 17 00:00:00 2001
From: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Date: Fri, 10 May 2024 17:06:40 +0300
Subject: wifi: iwlwifi: remove fw_running op

fw_running assumes that memory can be retrieved only after alive.
This assumption is no longer true as we support dump before alive.
To avoid invalid access to the NIC, check that STATUS_DEVICE_ENABLED
bit in trans status is set before dumping instead of the prior check.

Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com>
Reviewed-by: Luciano Coelho <luciano.coelho@intel.com>
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240510170500.ca07138cedeb.I090e31d3eaeb4ba19f5f84aba997ccd36927e9ac@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/debugfs.c | 3 +--
 drivers/net/wireless/intel/iwlwifi/fw/runtime.h | 1 -
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c    | 6 ------
 3 files changed, 1 insertion(+), 9 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c b/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c
index 751a125a1566..893b21fcaf87 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/debugfs.c
@@ -230,8 +230,7 @@ static ssize_t iwl_dbgfs_send_hcmd_write(struct iwl_fw_runtime *fwrt, char *buf,
 		.data = { NULL, },
 	};
 
-	if (fwrt->ops && fwrt->ops->fw_running &&
-	    !fwrt->ops->fw_running(fwrt->ops_ctx))
+	if (!iwl_trans_fw_running(fwrt->trans))
 		return -EIO;
 
 	if (count < header_size + 1 || count > 1024 * 4)
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
index 9122f9a1260a..d201440066ea 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
@@ -19,7 +19,6 @@
 struct iwl_fw_runtime_ops {
 	void (*dump_start)(void *ctx);
 	void (*dump_end)(void *ctx);
-	bool (*fw_running)(void *ctx);
 	int (*send_hcmd)(void *ctx, struct iwl_host_cmd *host_cmd);
 	bool (*d3_debug_enable)(void *ctx);
 };
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 53283d052e18..f19577474f4b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -770,11 +770,6 @@ static void iwl_mvm_fwrt_dump_end(void *ctx)
 	mutex_unlock(&mvm->mutex);
 }
 
-static bool iwl_mvm_fwrt_fw_running(void *ctx)
-{
-	return iwl_mvm_firmware_running(ctx);
-}
-
 static int iwl_mvm_fwrt_send_hcmd(void *ctx, struct iwl_host_cmd *host_cmd)
 {
 	struct iwl_mvm *mvm = (struct iwl_mvm *)ctx;
@@ -795,7 +790,6 @@ static bool iwl_mvm_d3_debug_enable(void *ctx)
 static const struct iwl_fw_runtime_ops iwl_mvm_fwrt_ops = {
 	.dump_start = iwl_mvm_fwrt_dump_start,
 	.dump_end = iwl_mvm_fwrt_dump_end,
-	.fw_running = iwl_mvm_fwrt_fw_running,
 	.send_hcmd = iwl_mvm_fwrt_send_hcmd,
 	.d3_debug_enable = iwl_mvm_d3_debug_enable,
 };
-- 
cgit v1.2.3-58-ga151


From 797af7ac20283d10818e9fc21b405dcaf6902b69 Mon Sep 17 00:00:00 2001
From: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Date: Fri, 10 May 2024 17:06:41 +0300
Subject: wifi: iwlwifi: remove redundant reading from NVM file

The driver reads xtal_calib from NVM file, but actually never uses it.
This is only used in dvm driver.

Signed-off-by: Chaya Rachel Ivgi <chaya.rachel.ivgi@intel.com>
Reviewed-by: Luciano Coelho <luciano.coelho@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240510170500.c4d1a4571049.I7d7b73dccb793e220f023e0d049b082b043ca95e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index cf5c9500af8c..5132964661c8 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -38,16 +38,13 @@ enum nvm_offsets {
 	N_HW_ADDRS = 3,
 	NVM_CHANNELS = 0x1E0 - NVM_SW_SECTION,
 
-	/* NVM calibration section offset (in words) definitions */
-	NVM_CALIB_SECTION = 0x2B8,
-	XTAL_CALIB = 0x316 - NVM_CALIB_SECTION,
-
 	/* NVM REGULATORY -Section offset (in words) definitions */
 	NVM_CHANNELS_SDP = 0,
 };
 
 enum ext_nvm_offsets {
 	/* NVM HW-Section offset (in words) definitions */
+
 	MAC_ADDRESS_OVERRIDE_EXT_NVM = 1,
 
 	/* NVM SW-Section offset (in words) definitions */
@@ -1574,9 +1571,6 @@ iwl_parse_nvm_data(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 			     &regulatory[NVM_CHANNELS_SDP] :
 			     &nvm_sw[NVM_CHANNELS];
 
-		/* in family 8000 Xtal calibration values moved to OTP */
-		data->xtal_calib[0] = *(nvm_calib + XTAL_CALIB);
-		data->xtal_calib[1] = *(nvm_calib + XTAL_CALIB + 1);
 		lar_enabled = true;
 	} else {
 		u16 lar_offset = data->nvm_version < 0xE39 ?
-- 
cgit v1.2.3-58-ga151


From 6a32ebd534ff31e93d6ef3ee5640b97efff8185d Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Sun, 12 May 2024 07:30:55 +0300
Subject: wifi: iwlwifi: mvm: don't always set antenna in beacon template cmd

This needs to be set only if the FW is not capable of selecting an
antenna on its own.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512072733.73d790911268.I158dc1dbc9c2e199c69d8213e5496d2de72da28b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index 5a06f887769a..f6baf7135efd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -1010,12 +1010,13 @@ static void iwl_mvm_mac_ctxt_set_tx(struct iwl_mvm *mvm,
 	tx->tx_flags = cpu_to_le32(tx_flags);
 
 	if (!fw_has_capa(&mvm->fw->ucode_capa,
-			 IWL_UCODE_TLV_CAPA_BEACON_ANT_SELECTION))
+			 IWL_UCODE_TLV_CAPA_BEACON_ANT_SELECTION)) {
 		iwl_mvm_toggle_tx_ant(mvm, &mvm->mgmt_last_antenna_idx);
 
-	tx->rate_n_flags =
-		cpu_to_le32(BIT(mvm->mgmt_last_antenna_idx) <<
-			    RATE_MCS_ANT_POS);
+		tx->rate_n_flags =
+			cpu_to_le32(BIT(mvm->mgmt_last_antenna_idx) <<
+				    RATE_MCS_ANT_POS);
+	}
 
 	rate = iwl_mvm_mac_ctxt_get_beacon_rate(mvm, info, vif);
 
-- 
cgit v1.2.3-58-ga151


From ca88709d10b1c21aca2bf58fa528723a81322b03 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Sun, 12 May 2024 07:30:56 +0300
Subject: wifi: iwlwifi: mvm: Use the SMPS cfg of the correct link

The exiting code is checking the SMPS mode of deflink.
Use the SMPS mode of the link_sta function parameter.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512072733.f3e2834e78f5.Ie24b431a6ad0dbf977afbbd4d2116bfe2426a0d5@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index b7a461dba41e..3b2875f2cab7 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -470,7 +470,7 @@ static int iwl_mvm_mld_cfg_sta(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		break;
 	}
 
-	switch (sta->deflink.smps_mode) {
+	switch (link_sta->smps_mode) {
 	case IEEE80211_SMPS_AUTOMATIC:
 	case IEEE80211_SMPS_NUM_MODES:
 		WARN_ON(1);
-- 
cgit v1.2.3-58-ga151


From bc4867aabfead3d8eb64827208f59bc1d517e213 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Sun, 12 May 2024 07:30:57 +0300
Subject: wifi: iwlwifi: mvm: add a of print of a few commands

This is needed for debugging.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512072733.65fc46e2ffbc.I19d4ef2cc06bfee7d9644a376d39399f0f6eaa15@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index f19577474f4b..c3cd5ab11aec 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -572,6 +572,7 @@ static const struct iwl_hcmd_names iwl_mvm_legacy_names[] = {
 	HCMD_NAME(D0I3_END_CMD),
 	HCMD_NAME(LTR_CONFIG),
 	HCMD_NAME(LDBG_CONFIG_CMD),
+	HCMD_NAME(DEBUG_LOG_MSG),
 };
 
 /* Please keep this array *SORTED* by hex value.
@@ -579,6 +580,7 @@ static const struct iwl_hcmd_names iwl_mvm_legacy_names[] = {
  */
 static const struct iwl_hcmd_names iwl_mvm_system_names[] = {
 	HCMD_NAME(SHARED_MEM_CFG_CMD),
+	HCMD_NAME(SOC_CONFIGURATION_CMD),
 	HCMD_NAME(INIT_EXTENDED_CFG_CMD),
 	HCMD_NAME(FW_ERROR_RECOVERY_CMD),
 	HCMD_NAME(RFI_CONFIG_CMD),
@@ -593,8 +595,10 @@ static const struct iwl_hcmd_names iwl_mvm_system_names[] = {
  * Access is done through binary search
  */
 static const struct iwl_hcmd_names iwl_mvm_mac_conf_names[] = {
+	HCMD_NAME(LOW_LATENCY_CMD),
 	HCMD_NAME(CHANNEL_SWITCH_TIME_EVENT_CMD),
 	HCMD_NAME(SESSION_PROTECTION_CMD),
+	HCMD_NAME(CANCEL_CHANNEL_SWITCH_CMD),
 	HCMD_NAME(MAC_CONFIG_CMD),
 	HCMD_NAME(LINK_CONFIG_CMD),
 	HCMD_NAME(STA_CONFIG_CMD),
@@ -603,7 +607,10 @@ static const struct iwl_hcmd_names iwl_mvm_mac_conf_names[] = {
 	HCMD_NAME(STA_DISABLE_TX_CMD),
 	HCMD_NAME(ROC_CMD),
 	HCMD_NAME(ROC_NOTIF),
+	HCMD_NAME(CHANNEL_SWITCH_ERROR_NOTIF),
+	HCMD_NAME(MISSED_VAP_NOTIF),
 	HCMD_NAME(SESSION_PROTECTION_NOTIF),
+	HCMD_NAME(PROBE_RESPONSE_DATA_NOTIF),
 	HCMD_NAME(CHANNEL_SWITCH_START_NOTIF),
 };
 
@@ -627,6 +634,8 @@ static const struct iwl_hcmd_names iwl_mvm_data_path_names[] = {
 	HCMD_NAME(DQA_ENABLE_CMD),
 	HCMD_NAME(UPDATE_MU_GROUPS_CMD),
 	HCMD_NAME(TRIGGER_RX_QUEUES_NOTIF_CMD),
+	HCMD_NAME(WNM_PLATFORM_PTM_REQUEST_CMD),
+	HCMD_NAME(WNM_80211V_TIMING_MEASUREMENT_CONFIG_CMD),
 	HCMD_NAME(STA_HE_CTXT_CMD),
 	HCMD_NAME(RLC_CONFIG_CMD),
 	HCMD_NAME(RFH_QUEUE_CONFIG_CMD),
@@ -650,6 +659,21 @@ static const struct iwl_hcmd_names iwl_mvm_statistics_names[] = {
 	HCMD_NAME(STATISTICS_OPER_PART1_NOTIF),
 };
 
+/* Please keep this array *SORTED* by hex value.
+ * Access is done through binary search
+ */
+static const struct iwl_hcmd_names iwl_mvm_debug_names[] = {
+	HCMD_NAME(LMAC_RD_WR),
+	HCMD_NAME(UMAC_RD_WR),
+	HCMD_NAME(HOST_EVENT_CFG),
+	HCMD_NAME(DBGC_SUSPEND_RESUME),
+	HCMD_NAME(BUFFER_ALLOCATION),
+	HCMD_NAME(GET_TAS_STATUS),
+	HCMD_NAME(FW_DUMP_COMPLETE_CMD),
+	HCMD_NAME(FW_CLEAR_BUFFER),
+	HCMD_NAME(MFU_ASSERT_DUMP_NTF),
+};
+
 /* Please keep this array *SORTED* by hex value.
  * Access is done through binary search
  */
@@ -705,6 +729,7 @@ static const struct iwl_hcmd_arr iwl_mvm_groups[] = {
 	[PROT_OFFLOAD_GROUP] = HCMD_ARR(iwl_mvm_prot_offload_names),
 	[REGULATORY_AND_NVM_GROUP] =
 		HCMD_ARR(iwl_mvm_regulatory_and_nvm_names),
+	[DEBUG_GROUP] = HCMD_ARR(iwl_mvm_debug_names),
 	[STATISTICS_GROUP] = HCMD_ARR(iwl_mvm_statistics_names),
 };
 
-- 
cgit v1.2.3-58-ga151


From e213eb95a49724a8168e6ef48a62fa06a785fa41 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Sun, 12 May 2024 07:30:58 +0300
Subject: wifi: iwlwifi: bump FW API to 91 for BZ/SC devices

Start supporting API version 91 for new devices.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512072733.27c00099e8b7.I99b4f24bba073414dd04a6e04a359c7fbba52990@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/bz.c | 2 +-
 drivers/net/wireless/intel/iwlwifi/cfg/sc.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c
index bc98b87cf2a1..45905e57e084 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c
@@ -10,7 +10,7 @@
 #include "fw/api/txq.h"
 
 /* Highest firmware API version supported */
-#define IWL_BZ_UCODE_API_MAX	90
+#define IWL_BZ_UCODE_API_MAX	91
 
 /* Lowest firmware API version supported */
 #define IWL_BZ_UCODE_API_MIN	80
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
index 9b79279fd76c..653279e08927 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
@@ -10,7 +10,7 @@
 #include "fw/api/txq.h"
 
 /* Highest firmware API version supported */
-#define IWL_SC_UCODE_API_MAX	90
+#define IWL_SC_UCODE_API_MAX	91
 
 /* Lowest firmware API version supported */
 #define IWL_SC_UCODE_API_MIN	82
-- 
cgit v1.2.3-58-ga151


From 472a15d319b8560da94c7c6e94ead2fea491dd1c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 12 May 2024 07:30:59 +0300
Subject: wifi: iwlwifi: mvm: mark bad no-data RX as having bad PLCP

Just like with any other frame, mac80211 will do sanity checks
on no-data RX (representing e.g. sounding PPDUs), and if e.g.
the NSS is wrong, it will warn. This isn't a good idea if we
already know the frame wasn't received well, e.g. has bad PLCP.
Unless the firmware reports "no error", set the bad PLCP flag
to skip checks in mac80211.

Also, since we're now extracting two different values from the
info field, use le32_get_bits() for both.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512072733.e6adcb9a6ece.Ic14c2e8ed5e80d48af78b2f04e9f08beeb62d68e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 8bbef4454399..2f9e2254fb46 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -2361,7 +2361,6 @@ void iwl_mvm_rx_monitor_no_data(struct iwl_mvm *mvm, struct napi_struct *napi,
 	struct iwl_rx_packet *pkt = rxb_addr(rxb);
 	struct iwl_rx_no_data_ver_3 *desc = (void *)pkt->data;
 	u32 rssi;
-	u32 info_type;
 	struct ieee80211_sta *sta = NULL;
 	struct sk_buff *skb;
 	struct iwl_mvm_rx_phy_data phy_data;
@@ -2374,7 +2373,6 @@ void iwl_mvm_rx_monitor_no_data(struct iwl_mvm *mvm, struct napi_struct *napi,
 		return;
 
 	rssi = le32_to_cpu(desc->rssi);
-	info_type = le32_to_cpu(desc->info) & RX_NO_DATA_INFO_TYPE_MSK;
 	phy_data.d0 = desc->phy_info[0];
 	phy_data.d1 = desc->phy_info[1];
 	phy_data.phy_info = IWL_RX_MPDU_PHY_TSF_OVERLOAD;
@@ -2426,7 +2424,12 @@ void iwl_mvm_rx_monitor_no_data(struct iwl_mvm *mvm, struct napi_struct *napi,
 	/* 0-length PSDU */
 	rx_status->flag |= RX_FLAG_NO_PSDU;
 
-	switch (info_type) {
+	/* mark as failed PLCP on any errors to skip checks in mac80211 */
+	if (le32_get_bits(desc->info, RX_NO_DATA_INFO_ERR_MSK) !=
+	    RX_NO_DATA_INFO_ERR_NONE)
+		rx_status->flag |= RX_FLAG_FAILED_PLCP_CRC;
+
+	switch (le32_get_bits(desc->info, RX_NO_DATA_INFO_TYPE_MSK)) {
 	case RX_NO_DATA_INFO_TYPE_NDP:
 		rx_status->zero_length_psdu_type =
 			IEEE80211_RADIOTAP_ZERO_LEN_PSDU_SOUNDING;
-- 
cgit v1.2.3-58-ga151


From 9c93c0b44be36fd5267fb79ae33453f989fbe909 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 12 May 2024 07:31:00 +0300
Subject: wifi: iwlwifi: mei: unify iwl_mei_set_power_limit() prototype

The two versions of iwl_mei_set_power_limit() are a bit different,
and while really the const isn't all that necessary on the inline,
it's still better to have it be the same.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512072733.854427753c92.I557716085cb1f6a35d1f97640388fa421f42a56a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h b/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h
index 1f3c885aeb65..ee48b86674a6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h
+++ b/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 /*
- * Copyright (C) 2021-2023 Intel Corporation
+ * Copyright (C) 2021-2024 Intel Corporation
  */
 
 #ifndef __iwl_mei_h__
@@ -488,7 +488,7 @@ static inline void iwl_mei_set_nic_info(const u8 *mac_address, const u8 *nvm_add
 static inline void iwl_mei_set_country_code(u16 mcc)
 {}
 
-static inline void iwl_mei_set_power_limit(__le16 *power_limit)
+static inline void iwl_mei_set_power_limit(const __le16 *power_limit)
 {}
 
 static inline int iwl_mei_register(void *priv,
-- 
cgit v1.2.3-58-ga151


From 8d59995067d9a6024fc63afa1434e0f5196727de Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Sun, 12 May 2024 07:31:01 +0300
Subject: wifi: iwlwifi: fw: api: Add new timepoint for scan failure

Add new time point to get the details of scan failure
While on it added other missing enum to match with FW.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Reviewed-by: Pagadala Yesu Anjaneyulu <pagadala.yesu.anjaneyulu@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512072733.6781e76990c4.Iadfb3f6dd63a98fc35019772266b5ebddc5b5270@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h b/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h
index 47c914de2992..6cfe8a779cc7 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h
@@ -470,6 +470,10 @@ enum iwl_fw_ini_region_device_memory_subtype {
  * @IWL_FW_INI_TIME_POINT_EAPOL_FAILED: EAPOL failed
  * @IWL_FW_INI_TIME_POINT_FAKE_TX: fake Tx
  * @IWL_FW_INI_TIME_POINT_DEASSOC: de association
+ * @IWL_FW_INI_TIME_POINT_PRESET_OVERRIDE_EXT_REQ: request to override preset
+ * @IWL_FW_INI_TIME_POINT_PRESET_OVERRIDE_START: start handling override preset
+ *	request
+ * @IWL_FW_INI_TIME_SCAN_FAILURE: failed scan channel list
  * @IWL_FW_INI_TIME_POINT_NUM: number of time points
  */
 enum iwl_fw_ini_time_point {
@@ -500,6 +504,9 @@ enum iwl_fw_ini_time_point {
 	IWL_FW_INI_TIME_POINT_EAPOL_FAILED,
 	IWL_FW_INI_TIME_POINT_FAKE_TX,
 	IWL_FW_INI_TIME_POINT_DEASSOC,
+	IWL_FW_INI_TIME_POINT_PRESET_OVERRIDE_EXT_REQ,
+	IWL_FW_INI_TIME_POINT_PRESET_OVERRIDE_START,
+	IWL_FW_INI_TIME_SCAN_FAILURE,
 	IWL_FW_INI_TIME_POINT_NUM,
 }; /* FW_TLV_DEBUG_TIME_POINT_API_E */
 
-- 
cgit v1.2.3-58-ga151


From 6ccfc7c49b6d99e3b857afb064e56a004b0a5e03 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 12 May 2024 07:31:03 +0300
Subject: wifi: iwlwifi: fw: avoid bad FW config on RXQ DMA failure

In practice, iwl_trans_get_rxq_dma_data() will not fail,
it only can do that if called with the wrong arguments.
But it does have an error value and doesn't initialize
the argument when it fails, so don't use the value then.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512072733.faf3a1b87589.I77c950173cb914676fbd28483e5ef420bb9f1bc9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/init.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/init.c b/drivers/net/wireless/intel/iwlwifi/fw/init.c
index 135bd48bfe9f..d8b083be5b6b 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/init.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/init.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
  * Copyright (C) 2017 Intel Deutschland GmbH
- * Copyright (C) 2019-2021 Intel Corporation
+ * Copyright (C) 2019-2021, 2024 Intel Corporation
  */
 #include "iwl-drv.h"
 #include "runtime.h"
@@ -135,7 +135,9 @@ int iwl_configure_rxq(struct iwl_fw_runtime *fwrt)
 		struct iwl_trans_rxq_dma_data data;
 
 		cmd->data[i].q_num = i + 1;
-		iwl_trans_get_rxq_dma_data(fwrt->trans, i + 1, &data);
+		ret = iwl_trans_get_rxq_dma_data(fwrt->trans, i + 1, &data);
+		if (ret)
+			goto out;
 
 		cmd->data[i].fr_bd_cb = cpu_to_le64(data.fr_bd_cb);
 		cmd->data[i].urbd_stts_wrptr =
@@ -149,6 +151,7 @@ int iwl_configure_rxq(struct iwl_fw_runtime *fwrt)
 
 	ret = iwl_trans_send_cmd(fwrt->trans, &hcmd);
 
+out:
 	kfree(cmd);
 
 	if (ret)
-- 
cgit v1.2.3-58-ga151


From 45eeee1f300d2b0b1ad8570b4ef6e2a23932f1a5 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Sun, 12 May 2024 07:31:05 +0300
Subject: wifi: iwlwifi: always print the firmware version in hex

All our firmware versions should now be printed in hex.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512072733.cae4c562b2c3.I5043383bd0e81f872c55046afd2c6560f482c8e0@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-drv.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
index 33654f228ee8..a392a5f1847f 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
@@ -982,16 +982,10 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv,
 			minor = le32_to_cpup(ptr++);
 			local_comp = le32_to_cpup(ptr);
 
-			if (major >= 35)
-				snprintf(drv->fw.fw_version,
-					 sizeof(drv->fw.fw_version),
-					"%u.%08x.%u %s", major, minor,
-					local_comp, iwl_reduced_fw_name(drv));
-			else
-				snprintf(drv->fw.fw_version,
-					 sizeof(drv->fw.fw_version),
-					"%u.%u.%u %s", major, minor,
-					local_comp, iwl_reduced_fw_name(drv));
+			snprintf(drv->fw.fw_version,
+				 sizeof(drv->fw.fw_version),
+				 "%u.%08x.%u %s", major, minor,
+				 local_comp, iwl_reduced_fw_name(drv));
 			break;
 			}
 		case IWL_UCODE_TLV_FW_DBG_DEST: {
-- 
cgit v1.2.3-58-ga151


From ed93faf01797d83b3347168adae93e6eb5fc1d9c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 12 May 2024 07:31:07 +0300
Subject: wifi: iwlwifi: mvm: don't track used links separately

We track which link is using which FW link ID, so there
really isn't a need to separately track which link IDs
are in use. Remove that code and check the table when
looking for a new link ID to use.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512072733.1a67d8af815f.Ie642c12dce3ab55c688abd9a25918569e83e558a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c   |  2 --
 drivers/net/wireless/intel/iwlwifi/mvm/link.c | 22 +++++-----------------
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h  |  1 -
 3 files changed, 5 insertions(+), 20 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index e7f5978ef2d7..8198451d70a2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -1499,8 +1499,6 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 	for (i = 0; i < IWL_MVM_FW_MAX_LINK_ID + 1; i++)
 		RCU_INIT_POINTER(mvm->link_id_to_link_conf[i], NULL);
 
-	memset(&mvm->fw_link_ids_map, 0, sizeof(mvm->fw_link_ids_map));
-
 	mvm->tdls_cs.peer.sta_id = IWL_MVM_INVALID_STA;
 
 	/* reset quota debouncing buffer - 0xff will yield invalid data */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 6ec9a8e21a34..14dab617c0bd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -50,26 +50,15 @@ static void iwl_mvm_print_esr_state(struct iwl_mvm *mvm, u32 mask)
 static u32 iwl_mvm_get_free_fw_link_id(struct iwl_mvm *mvm,
 				       struct iwl_mvm_vif *mvm_vif)
 {
-	u32 link_id;
+	u32 i;
 
 	lockdep_assert_held(&mvm->mutex);
 
-	link_id = ffz(mvm->fw_link_ids_map);
+	for (i = 0; i < ARRAY_SIZE(mvm->link_id_to_link_conf); i++)
+		if (!rcu_access_pointer(mvm->link_id_to_link_conf[i]))
+			return i;
 
-	/* this case can happen if there're deactivated but not removed links */
-	if (link_id > IWL_MVM_FW_MAX_LINK_ID)
-		return IWL_MVM_FW_LINK_ID_INVALID;
-
-	mvm->fw_link_ids_map |= BIT(link_id);
-	return link_id;
-}
-
-static void iwl_mvm_release_fw_link_id(struct iwl_mvm *mvm, u32 link_id)
-{
-	lockdep_assert_held(&mvm->mutex);
-
-	if (!WARN_ON(link_id > IWL_MVM_FW_MAX_LINK_ID))
-		mvm->fw_link_ids_map &= ~BIT(link_id);
+	return IWL_MVM_FW_LINK_ID_INVALID;
 }
 
 static int iwl_mvm_link_cmd_send(struct iwl_mvm *mvm,
@@ -380,7 +369,6 @@ int iwl_mvm_unset_link_mapping(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	RCU_INIT_POINTER(mvm->link_id_to_link_conf[link_info->fw_link_id],
 			 NULL);
-	iwl_mvm_release_fw_link_id(mvm, link_info->fw_link_id);
 	return 0;
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index ca1c5944da67..d732780aae5f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1043,7 +1043,6 @@ struct iwl_mvm {
 	struct iwl_rx_phy_info last_phy_info;
 	struct ieee80211_sta __rcu *fw_id_to_mac_id[IWL_MVM_STATION_COUNT_MAX];
 	struct ieee80211_link_sta __rcu *fw_id_to_link_sta[IWL_MVM_STATION_COUNT_MAX];
-	unsigned long fw_link_ids_map;
 	u8 rx_ba_sessions;
 
 	/* configured by mac80211 */
-- 
cgit v1.2.3-58-ga151


From 157c8a448354518f9a1e8d758094fc80afeb8568 Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Sun, 12 May 2024 07:31:08 +0300
Subject: wifi: iwlwifi: mvm: use vif P2P type helper

Use ieee80211_vif_type_p2p instead of checking for IFTYPE_AP in
combination with vif->p2p.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512072733.15a00a812c1b.I5ffee795d960c9beda46a934f5b4c70acde112f9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index a7ec172eeade..695b5ba3d023 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -208,7 +208,7 @@ static void iwl_mvm_scan_iterator(void *_data, u8 *mac,
 
 	curr_mvmvif = iwl_mvm_vif_from_mac80211(data->current_vif);
 
-	if (vif->type == NL80211_IFTYPE_AP && vif->p2p &&
+	if (ieee80211_vif_type_p2p(vif) == NL80211_IFTYPE_P2P_GO &&
 	    mvmvif->deflink.phy_ctxt && curr_mvmvif->deflink.phy_ctxt &&
 	    mvmvif->deflink.phy_ctxt->id != curr_mvmvif->deflink.phy_ctxt->id)
 		data->is_dcm_with_p2p_go = true;
@@ -2875,7 +2875,7 @@ static void iwl_mvm_scan_respect_p2p_go_iter(void *_data, u8 *mac,
 	if (vif == data->current_vif)
 		return;
 
-	if (vif->type == NL80211_IFTYPE_AP && vif->p2p) {
+	if (ieee80211_vif_type_p2p(vif) == NL80211_IFTYPE_P2P_GO) {
 		u32 link_id;
 
 		for (link_id = 0;
-- 
cgit v1.2.3-58-ga151


From 9574c7592c5080a3e1f3fdfe658bd13083a04cb5 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Sun, 12 May 2024 07:31:10 +0300
Subject: wifi: iwlwifi: mvm: disable dynamic EMLSR when AUTO_EML is false

When AUTO_EML is set to false, this change ignores the exit criteria,
for testing purposes. Currently, if AUTO_EML is disabled, the
driver will not select a link or enter EMLSR, but will still exit if one
of the criteria is not fulfilled.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240512072733.10a0b5da6ec2.I46fd578a3ef6cdbf14fdc4dfa97b4be008fe68e3@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/link.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 14dab617c0bd..fd8a21891d26 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -966,6 +966,9 @@ void iwl_mvm_exit_esr(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	lockdep_assert_held(&mvm->mutex);
 
+	if (!IWL_MVM_AUTO_EML_ENABLE)
+		return;
+
 	/* Nothing to do */
 	if (!mvmvif->esr_active)
 		return;
@@ -1013,6 +1016,9 @@ void iwl_mvm_block_esr(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	lockdep_assert_held(&mvm->mutex);
 
+	if (!IWL_MVM_AUTO_EML_ENABLE)
+		return;
+
 	/* This should be called only with disable reasons */
 	if (WARN_ON(!(reason & IWL_MVM_BLOCK_ESR_REASONS)))
 		return;
@@ -1099,6 +1105,9 @@ void iwl_mvm_unblock_esr(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 	lockdep_assert_held(&mvm->mutex);
 
+	if (!IWL_MVM_AUTO_EML_ENABLE)
+		return;
+
 	/* This should be called only with disable reasons */
 	if (WARN_ON(!(reason & IWL_MVM_BLOCK_ESR_REASONS)))
 		return;
-- 
cgit v1.2.3-58-ga151


From f79968eb417574f84a06c22510cbeb414e89a809 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Sun, 12 May 2024 07:58:57 +0300
Subject: wifi: iwlwifi: mvm: leave a print in the logs when we call fw_nmi()

When we crash the firmware, we need to know why we decided to do so.
Almost all the callsites of iwl_force_nmi() print something in the logs
that explain why the driver decided to crash the firmware.
Debugfs doesn't print anything and it is then hard to understand why the
firmware has crashed.
Add a simple print in the debugfs hook to ease the debug.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512075822.e2558de222dc.Idd81777c47264e6f557b086625895c1dc2f667f3@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 79f4ac8cbc72..96c9382b98fb 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -1396,6 +1396,8 @@ static ssize_t iwl_dbgfs_fw_nmi_write(struct iwl_mvm *mvm, char *buf,
 	if (!iwl_mvm_firmware_running(mvm))
 		return -EIO;
 
+	IWL_ERR(mvm, "Triggering an NMI from debugfs\n");
+
 	if (count == 6 && !strcmp(buf, "nolog\n"))
 		set_bit(IWL_MVM_STATUS_SUPPRESS_ERROR_LOG_ONCE, &mvm->status);
 
-- 
cgit v1.2.3-58-ga151


From bdaae563e097a40e942d66a15cb3a5537f156dd5 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 12 May 2024 15:24:56 +0300
Subject: wifi: iwlwifi: tracing: fix condition to allocate buf1

There's no need to allocate buf1 if we're not going to
write anything to it, and the condition for writing is
(correctly) "hdr_len > 0 && !iwl_trace_data()". Also
allocate the event space only under the same condition.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512152312.a11e6490d053.Ie090beea70d2f271bb22c9b0e287dbd342fb2a96@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
index e656bf6bc003..b9aa04cf1d83 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
@@ -4,7 +4,7 @@
  * Copyright(c) 2009 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2015 Intel Mobile Communications GmbH
  * Copyright(c) 2016 - 2017 Intel Deutschland GmbH
- * Copyright(c) 2018, 2023  Intel Corporation
+ * Copyright(c) 2018, 2023-2024  Intel Corporation
  *****************************************************************************/
 
 #if !defined(__IWLWIFI_DEVICE_TRACE_IWLWIFI) || defined(TRACE_HEADER_MULTI_READ)
@@ -88,8 +88,8 @@ TRACE_EVENT(iwlwifi_dev_tx,
 		 * for the possible padding).
 		 */
 		__dynamic_array(u8, buf0, buf0_len)
-		__dynamic_array(u8, buf1, hdr_len > 0 && iwl_trace_data(skb) ?
-						0 : skb->len - hdr_len)
+		__dynamic_array(u8, buf1, hdr_len > 0 && !iwl_trace_data(skb) ?
+						skb->len - hdr_len : 0)
 	),
 	TP_fast_assign(
 		DEV_ASSIGN;
-- 
cgit v1.2.3-58-ga151


From 9d42be9b181b825d7401a8dd10a198b087783e0b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 12 May 2024 15:24:57 +0300
Subject: wifi: iwlwifi: simplify TX tracing

There's no need to calculate again whether data should
be included or be handled externally, just check if any
space for it was already reserved.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512152312.a3d3b4796460.I85bd3029baee24ebf0be04db7d6bf01834090869@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-devtrace-data.h    | 4 ++--
 drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-data.h b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-data.h
index 2c280a2fe3df..0d4a0896a2c5 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-data.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-data.h
@@ -3,7 +3,7 @@
  *
  * Copyright(c) 2009 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2015        Intel Deutschland GmbH
- * Copyright(c) 2018 - 2019, 2023 Intel Corporation
+ * Copyright(c) 2018 - 2019, 2023-2024 Intel Corporation
  *****************************************************************************/
 
 #if !defined(__IWLWIFI_DEVICE_TRACE_DATA) || defined(TRACE_HEADER_MULTI_READ)
@@ -28,7 +28,7 @@ TRACE_EVENT(iwlwifi_dev_tx_tb,
 	TP_fast_assign(
 		DEV_ASSIGN;
 		__entry->phys = phys;
-		if (iwl_trace_data(skb))
+		if (__get_dynamic_array_len(data))
 			memcpy(__get_dynamic_array(data), data_src, data_len);
 	),
 	TP_printk("[%s] TX frame data", __get_str(dev))
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
index b9aa04cf1d83..ead72c3d33bd 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-iwlwifi.h
@@ -99,7 +99,7 @@ TRACE_EVENT(iwlwifi_dev_tx,
 			__entry->framelen += skb->len - hdr_len;
 		memcpy(__get_dynamic_array(tfd), tfd, tfdlen);
 		memcpy(__get_dynamic_array(buf0), buf0, buf0_len);
-		if (hdr_len > 0 && !iwl_trace_data(skb))
+		if (__get_dynamic_array_len(buf1))
 			skb_copy_bits(skb, hdr_len,
 				      __get_dynamic_array(buf1),
 				      skb->len - hdr_len);
-- 
cgit v1.2.3-58-ga151


From 3f3126515fbee0e364f8218f6171c091d8e2555e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Sun, 12 May 2024 15:24:58 +0300
Subject: wifi: iwlwifi: mvm: add mvm-specific guard

Add and use an mvm-specific guard for the mvm mutex.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512152312.7391425b3af2.I1b109f56753dd3f35602dd2c5cf6e1e35aee313d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 225 ++++++---------------
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  |  88 +++-----
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |   3 +
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c       |  14 +-
 drivers/net/wireless/intel/iwlwifi/mvm/tdls.c      |  34 ++--
 drivers/net/wireless/intel/iwlwifi/mvm/tt.c        |  60 ++----
 drivers/net/wireless/intel/iwlwifi/mvm/utils.c     |   7 +-
 7 files changed, 136 insertions(+), 295 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 486a6b8f3c97..f2f9e2620ff6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1258,7 +1258,7 @@ static void iwl_mvm_restart_complete(struct iwl_mvm *mvm)
 {
 	int ret;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	clear_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status);
 
@@ -1274,8 +1274,6 @@ static void iwl_mvm_restart_complete(struct iwl_mvm *mvm)
 	 * of packets the FW sent out, so we must reconnect.
 	 */
 	iwl_mvm_teardown_tdls_peers(mvm);
-
-	mutex_unlock(&mvm->mutex);
 }
 
 void iwl_mvm_mac_reconfig_complete(struct ieee80211_hw *hw,
@@ -1621,9 +1619,8 @@ static void iwl_mvm_prevent_esr_done_wk(struct wiphy *wiphy,
 	struct iwl_mvm *mvm = mvmvif->mvm;
 	struct ieee80211_vif *vif = iwl_mvm_get_bss_vif(mvm);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	iwl_mvm_unblock_esr(mvm, vif, IWL_MVM_ESR_BLOCKED_PREVENTION);
-	mutex_unlock(&mvm->mutex);
 }
 
 static void iwl_mvm_mlo_int_scan_wk(struct wiphy *wiphy, struct wiphy_work *wk)
@@ -1633,11 +1630,8 @@ static void iwl_mvm_mlo_int_scan_wk(struct wiphy *wiphy, struct wiphy_work *wk)
 	struct ieee80211_vif *vif =
 		container_of((void *)mvmvif, struct ieee80211_vif, drv_priv);
 
-	mutex_lock(&mvmvif->mvm->mutex);
-
+	guard(mvm)(mvmvif->mvm);
 	iwl_mvm_int_mlo_scan(mvmvif->mvm, vif);
-
-	mutex_unlock(&mvmvif->mvm->mutex);
 }
 
 static void iwl_mvm_unblock_esr_tpt(struct wiphy *wiphy, struct wiphy_work *wk)
@@ -1647,9 +1641,8 @@ static void iwl_mvm_unblock_esr_tpt(struct wiphy *wiphy, struct wiphy_work *wk)
 	struct iwl_mvm *mvm = mvmvif->mvm;
 	struct ieee80211_vif *vif = iwl_mvm_get_bss_vif(mvm);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	iwl_mvm_unblock_esr(mvm, vif, IWL_MVM_ESR_BLOCKED_TPT);
-	mutex_unlock(&mvm->mutex);
 }
 
 void iwl_mvm_mac_init_mvmvif(struct iwl_mvm *mvm, struct iwl_mvm_vif *mvmvif)
@@ -2039,7 +2032,7 @@ void iwl_mvm_configure_filter(struct ieee80211_hw *hw,
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mcast_filter_cmd *cmd = (void *)(unsigned long)multicast;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	/* replace previous configuration */
 	kfree(mvm->mcast_filter_cmd);
@@ -2056,7 +2049,6 @@ void iwl_mvm_configure_filter(struct ieee80211_hw *hw,
 
 	iwl_mvm_recalc_multicast(mvm);
 out:
-	mutex_unlock(&mvm->mutex);
 	*total_flags = 0;
 }
 
@@ -2076,9 +2068,8 @@ static void iwl_mvm_config_iface_filter(struct ieee80211_hw *hw,
 	    !vif->p2p)
 		return;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
-	mutex_unlock(&mvm->mutex);
 }
 
 int iwl_mvm_update_mu_groups(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
@@ -3123,7 +3114,7 @@ static void iwl_mvm_stop_ap_ibss(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	iwl_mvm_stop_ap_ibss_common(mvm, vif);
 
@@ -3153,8 +3144,6 @@ static void iwl_mvm_stop_ap_ibss(struct ieee80211_hw *hw,
 	iwl_mvm_power_update_mac(mvm);
 
 	iwl_mvm_mac_ctxt_remove(mvm, vif);
-
-	mutex_unlock(&mvm->mutex);
 }
 
 static void iwl_mvm_stop_ap(struct ieee80211_hw *hw,
@@ -3209,7 +3198,7 @@ static void iwl_mvm_bss_info_changed(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	if (changes & BSS_CHANGED_IDLE && !vif->cfg.idle)
 		iwl_mvm_scan_stop(mvm, IWL_MVM_SCAN_SCHED, true);
@@ -3236,25 +3225,19 @@ static void iwl_mvm_bss_info_changed(struct ieee80211_hw *hw,
 				bss_conf->txpower);
 		iwl_mvm_set_tx_power(mvm, vif, bss_conf->txpower);
 	}
-
-	mutex_unlock(&mvm->mutex);
 }
 
 int iwl_mvm_mac_hw_scan(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 			struct ieee80211_scan_request *hw_req)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	int ret;
 
 	if (hw_req->req.n_channels == 0 ||
 	    hw_req->req.n_channels > mvm->fw->ucode_capa.n_scan_channels)
 		return -EINVAL;
 
-	mutex_lock(&mvm->mutex);
-	ret = iwl_mvm_reg_scan_start(mvm, vif, &hw_req->req, &hw_req->ies);
-	mutex_unlock(&mvm->mutex);
-
-	return ret;
+	guard(mvm)(mvm);
+	return iwl_mvm_reg_scan_start(mvm, vif, &hw_req->req, &hw_req->ies);
 }
 
 void iwl_mvm_mac_cancel_hw_scan(struct ieee80211_hw *hw,
@@ -3262,7 +3245,7 @@ void iwl_mvm_mac_cancel_hw_scan(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	/* Due to a race condition, it's possible that mac80211 asks
 	 * us to stop a hw_scan when it's already stopped.  This can
@@ -3273,8 +3256,6 @@ void iwl_mvm_mac_cancel_hw_scan(struct ieee80211_hw *hw,
 	*/
 	if (mvm->scan_status & IWL_MVM_SCAN_REGULAR)
 		iwl_mvm_scan_stop(mvm, IWL_MVM_SCAN_REGULAR, true);
-
-	mutex_unlock(&mvm->mutex);
 }
 
 void
@@ -3443,7 +3424,7 @@ void iwl_mvm_sta_pre_rcu_remove(struct ieee80211_hw *hw,
 	 * Since there's mvm->mutex here, no need to have RCU lock for
 	 * mvm_sta->link access.
 	 */
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	for (link_id = 0; link_id < ARRAY_SIZE(mvm_sta->link); link_id++) {
 		struct iwl_mvm_link_sta *link_sta;
 		u32 sta_id;
@@ -3460,7 +3441,6 @@ void iwl_mvm_sta_pre_rcu_remove(struct ieee80211_hw *hw,
 			RCU_INIT_POINTER(mvm->fw_id_to_link_sta[sta_id], NULL);
 		}
 	}
-	mutex_unlock(&mvm->mutex);
 }
 
 static void iwl_mvm_check_uapsd(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
@@ -4207,12 +4187,8 @@ static int iwl_mvm_mac_conf_tx(struct ieee80211_hw *hw,
 	 * The exception is P2P_DEVICE interface which needs immediate update.
 	 */
 	if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
-		int ret;
-
-		mutex_lock(&mvm->mutex);
-		ret = iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
-		mutex_unlock(&mvm->mutex);
-		return ret;
+		guard(mvm)(mvm);
+		return iwl_mvm_mac_ctxt_changed(mvm, vif, false, NULL);
 	}
 	return 0;
 }
@@ -4223,9 +4199,8 @@ void iwl_mvm_mac_mgd_prepare_tx(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	iwl_mvm_protect_assoc(mvm, vif, info->duration, info->link_id);
-	mutex_unlock(&mvm->mutex);
 }
 
 void iwl_mvm_mac_mgd_complete_tx(struct ieee80211_hw *hw,
@@ -4238,9 +4213,8 @@ void iwl_mvm_mac_mgd_complete_tx(struct ieee80211_hw *hw,
 	if (info->success)
 		return;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	iwl_mvm_stop_session_protection(mvm, vif);
-	mutex_unlock(&mvm->mutex);
 }
 
 int iwl_mvm_mac_sched_scan_start(struct ieee80211_hw *hw,
@@ -4250,20 +4224,12 @@ int iwl_mvm_mac_sched_scan_start(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	int ret;
-
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
-	if (!vif->cfg.idle) {
-		ret = -EBUSY;
-		goto out;
-	}
-
-	ret = iwl_mvm_sched_scan_start(mvm, vif, req, ies, IWL_MVM_SCAN_SCHED);
+	if (!vif->cfg.idle)
+		return -EBUSY;
 
-out:
-	mutex_unlock(&mvm->mutex);
-	return ret;
+	return iwl_mvm_sched_scan_start(mvm, vif, req, ies, IWL_MVM_SCAN_SCHED);
 }
 
 int iwl_mvm_mac_sched_scan_stop(struct ieee80211_hw *hw,
@@ -4541,13 +4507,9 @@ int iwl_mvm_mac_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 			struct ieee80211_key_conf *key)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	int ret;
 
-	mutex_lock(&mvm->mutex);
-	ret = __iwl_mvm_mac_set_key(hw, cmd, vif, sta, key);
-	mutex_unlock(&mvm->mutex);
-
-	return ret;
+	guard(mvm)(mvm);
+	return __iwl_mvm_mac_set_key(hw, cmd, vif, sta, key);
 }
 
 void iwl_mvm_mac_update_tkip_key(struct ieee80211_hw *hw,
@@ -4838,7 +4800,7 @@ int iwl_mvm_roc_common(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	if (ret)
 		return ret;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_STATION:
@@ -4848,30 +4810,25 @@ int iwl_mvm_roc_common(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		ret = ops->add_aux_sta_for_hs20(mvm, lmac_id);
 		if (!ret)
 			ret = iwl_mvm_roc_station(mvm, channel, vif, duration);
-		goto out_unlock;
+		return ret;
 	case NL80211_IFTYPE_P2P_DEVICE:
 		/* handle below */
 		break;
 	default:
 		IWL_ERR(mvm, "ROC: Invalid vif type=%u\n", vif->type);
-		ret = -EINVAL;
-		goto out_unlock;
+		return -EINVAL;
 	}
 
 
 	ret = iwl_mvm_p2p_find_phy_ctxt(mvm, vif, channel);
 	if (ret)
-		goto out_unlock;
+		return ret;
 
 	ret = ops->link(mvm, vif);
 	if (ret)
-		goto out_unlock;
+		return ret;
 
-	ret = iwl_mvm_start_p2p_roc(mvm, vif, duration, type);
-out_unlock:
-	mutex_unlock(&mvm->mutex);
-	IWL_DEBUG_MAC80211(mvm, "leave\n");
-	return ret;
+	return iwl_mvm_start_p2p_roc(mvm, vif, duration, type);
 }
 
 int iwl_mvm_cancel_roc(struct ieee80211_hw *hw,
@@ -4952,13 +4909,9 @@ int iwl_mvm_add_chanctx(struct ieee80211_hw *hw,
 			struct ieee80211_chanctx_conf *ctx)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	int ret;
 
-	mutex_lock(&mvm->mutex);
-	ret = __iwl_mvm_add_chanctx(mvm, ctx);
-	mutex_unlock(&mvm->mutex);
-
-	return ret;
+	guard(mvm)(mvm);
+	return __iwl_mvm_add_chanctx(mvm, ctx);
 }
 
 static void __iwl_mvm_remove_chanctx(struct iwl_mvm *mvm,
@@ -4977,9 +4930,8 @@ void iwl_mvm_remove_chanctx(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	__iwl_mvm_remove_chanctx(mvm, ctx);
-	mutex_unlock(&mvm->mutex);
 }
 
 void iwl_mvm_change_chanctx(struct ieee80211_hw *hw,
@@ -4999,26 +4951,23 @@ void iwl_mvm_change_chanctx(struct ieee80211_hw *hw,
 		      phy_ctxt->ref, changed))
 		return;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	/* we are only changing the min_width, may be a noop */
 	if (changed == IEEE80211_CHANCTX_CHANGE_MIN_WIDTH) {
 		if (phy_ctxt->width == def->width)
-			goto out_unlock;
+			return;
 
 		/* we are just toggling between 20_NOHT and 20 */
 		if (phy_ctxt->width <= NL80211_CHAN_WIDTH_20 &&
 		    def->width <= NL80211_CHAN_WIDTH_20)
-			goto out_unlock;
+			return;
 	}
 
 	iwl_mvm_bt_coex_vif_change(mvm);
 	iwl_mvm_phy_ctxt_changed(mvm, phy_ctxt, def, &ctx->ap,
 				 ctx->rx_chains_static,
 				 ctx->rx_chains_dynamic);
-
-out_unlock:
-	mutex_unlock(&mvm->mutex);
 }
 
 /*
@@ -5157,13 +5106,9 @@ static int iwl_mvm_assign_vif_chanctx(struct ieee80211_hw *hw,
 				      struct ieee80211_chanctx_conf *ctx)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	int ret;
 
-	mutex_lock(&mvm->mutex);
-	ret = __iwl_mvm_assign_vif_chanctx(mvm, vif, link_conf, ctx, false);
-	mutex_unlock(&mvm->mutex);
-
-	return ret;
+	guard(mvm)(mvm);
+	return __iwl_mvm_assign_vif_chanctx(mvm, vif, link_conf, ctx, false);
 }
 
 /*
@@ -5251,9 +5196,8 @@ static void iwl_mvm_unassign_vif_chanctx(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	__iwl_mvm_unassign_vif_chanctx(mvm, vif, link_conf, ctx, false);
-	mutex_unlock(&mvm->mutex);
 }
 
 static int
@@ -5263,7 +5207,7 @@ iwl_mvm_switch_vif_chanctx_swap(struct iwl_mvm *mvm,
 {
 	int ret;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	ops->__unassign_vif_chanctx(mvm, vifs[0].vif, vifs[0].link_conf,
 				    vifs[0].old_ctx, true);
 	__iwl_mvm_remove_chanctx(mvm, vifs[0].old_ctx);
@@ -5286,7 +5230,7 @@ iwl_mvm_switch_vif_chanctx_swap(struct iwl_mvm *mvm,
 	if (iwl_mvm_phy_ctx_count(mvm) > 1)
 		iwl_mvm_teardown_tdls_peers(mvm);
 
-	goto out;
+	return 0;
 
 out_remove:
 	__iwl_mvm_remove_chanctx(mvm, vifs[0].new_ctx);
@@ -5303,15 +5247,11 @@ out_reassign:
 		goto out_restart;
 	}
 
-	goto out;
+	return ret;
 
 out_restart:
 	/* things keep failing, better restart the hw */
 	iwl_mvm_nic_restart(mvm, false);
-
-out:
-	mutex_unlock(&mvm->mutex);
-
 	return ret;
 }
 
@@ -5322,7 +5262,7 @@ iwl_mvm_switch_vif_chanctx_reassign(struct iwl_mvm *mvm,
 {
 	int ret;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	ops->__unassign_vif_chanctx(mvm, vifs[0].vif, vifs[0].link_conf,
 				    vifs[0].old_ctx, true);
 
@@ -5334,7 +5274,7 @@ iwl_mvm_switch_vif_chanctx_reassign(struct iwl_mvm *mvm,
 		goto out_reassign;
 	}
 
-	goto out;
+	return 0;
 
 out_reassign:
 	if (ops->__assign_vif_chanctx(mvm, vifs[0].vif, vifs[0].link_conf,
@@ -5343,15 +5283,11 @@ out_reassign:
 		goto out_restart;
 	}
 
-	goto out;
+	return ret;
 
 out_restart:
 	/* things keep failing, better restart the hw */
 	iwl_mvm_nic_restart(mvm, false);
-
-out:
-	mutex_unlock(&mvm->mutex);
-
 	return ret;
 }
 
@@ -5478,13 +5414,9 @@ int iwl_mvm_mac_testmode_cmd(struct ieee80211_hw *hw,
 			     void *data, int len)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	int err;
 
-	mutex_lock(&mvm->mutex);
-	err = __iwl_mvm_mac_testmode_cmd(mvm, vif, data, len);
-	mutex_unlock(&mvm->mutex);
-
-	return err;
+	guard(mvm)(mvm);
+	return __iwl_mvm_mac_testmode_cmd(mvm, vif, data, len);
 }
 #endif
 
@@ -5701,13 +5633,9 @@ static int iwl_mvm_mac_pre_channel_switch(struct ieee80211_hw *hw,
 					  struct ieee80211_channel_switch *chsw)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	int ret;
 
-	mutex_lock(&mvm->mutex);
-	ret = iwl_mvm_pre_channel_switch(mvm, vif, chsw);
-	mutex_unlock(&mvm->mutex);
-
-	return ret;
+	guard(mvm)(mvm);
+	return iwl_mvm_pre_channel_switch(mvm, vif, chsw);
 }
 
 void iwl_mvm_channel_switch_rx_beacon(struct ieee80211_hw *hw,
@@ -5759,16 +5687,14 @@ void iwl_mvm_channel_switch_rx_beacon(struct ieee80211_hw *hw,
 	}
 	mvmvif->csa_count = chsw->count;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	if (mvmvif->csa_failed)
-		goto out_unlock;
+		return;
 
 	WARN_ON(iwl_mvm_send_cmd_pdu(mvm,
 				     WIDE_ID(MAC_CONF_GROUP,
 					     CHANNEL_SWITCH_TIME_EVENT_CMD),
 				     0, sizeof(cmd), &cmd));
-out_unlock:
-	mutex_unlock(&mvm->mutex);
 }
 
 static void iwl_mvm_flush_no_vif(struct iwl_mvm *mvm, u32 queues, bool drop)
@@ -5777,17 +5703,16 @@ static void iwl_mvm_flush_no_vif(struct iwl_mvm *mvm, u32 queues, bool drop)
 
 	if (!iwl_mvm_has_new_tx_api(mvm)) {
 		if (drop) {
-			mutex_lock(&mvm->mutex);
+			guard(mvm)(mvm);
 			iwl_mvm_flush_tx_path(mvm,
 				iwl_mvm_flushable_queues(mvm) & queues);
-			mutex_unlock(&mvm->mutex);
 		} else {
 			iwl_trans_wait_tx_queues_empty(mvm->trans, queues);
 		}
 		return;
 	}
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	for (i = 0; i < mvm->fw->ucode_capa.num_stations; i++) {
 		struct ieee80211_sta *sta;
 
@@ -5802,7 +5727,6 @@ static void iwl_mvm_flush_no_vif(struct iwl_mvm *mvm, u32 queues, bool drop)
 			iwl_mvm_wait_sta_queues_empty(mvm,
 					iwl_mvm_sta_from_mac80211(sta));
 	}
-	mutex_unlock(&mvm->mutex);
 }
 
 void iwl_mvm_mac_flush(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
@@ -5885,7 +5809,7 @@ void iwl_mvm_mac_flush_sta(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	struct ieee80211_link_sta *link_sta;
 	int link_id;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	for_each_sta_active_link(vif, sta, link_sta, link_id) {
 		mvm_link_sta = rcu_dereference_protected(mvmsta->link[link_id],
 							 lockdep_is_held(&mvm->mutex));
@@ -5896,7 +5820,6 @@ void iwl_mvm_mac_flush_sta(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 				      mvmsta->tfd_queue_msk))
 			IWL_ERR(mvm, "flush request fail\n");
 	}
-	mutex_unlock(&mvm->mutex);
 }
 
 static int iwl_mvm_mac_get_acs_survey(struct iwl_mvm *mvm, int idx,
@@ -5962,7 +5885,6 @@ int iwl_mvm_mac_get_survey(struct ieee80211_hw *hw, int idx,
 			   struct survey_info *survey)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	int ret = 0;
 	u8 cmd_ver = iwl_fw_lookup_cmd_ver(mvm->fw,
 					   WIDE_ID(SYSTEM_GROUP,
 						   SYSTEM_STATISTICS_CMD),
@@ -5982,12 +5904,13 @@ int iwl_mvm_mac_get_survey(struct ieee80211_hw *hw, int idx,
 	if (idx > 0)
 		return iwl_mvm_mac_get_acs_survey(mvm, idx - 1, survey);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	if (iwl_mvm_firmware_running(mvm)) {
-		ret = iwl_mvm_request_statistics(mvm, false);
+		int ret = iwl_mvm_request_statistics(mvm, false);
+
 		if (ret)
-			goto out;
+			return ret;
 	}
 
 	survey->filled = SURVEY_INFO_TIME_RX |
@@ -6003,7 +5926,7 @@ int iwl_mvm_mac_get_survey(struct ieee80211_hw *hw, int idx,
 
 	/* the new fw api doesn't support the following fields */
 	if (cmd_ver != IWL_FW_CMD_VER_UNKNOWN)
-		goto out;
+		return 0;
 
 	survey->filled |= SURVEY_INFO_TIME |
 			  SURVEY_INFO_TIME_SCAN;
@@ -6015,9 +5938,7 @@ int iwl_mvm_mac_get_survey(struct ieee80211_hw *hw, int idx,
 			    mvm->radio_stats.on_time_scan;
 	do_div(survey->time_scan, USEC_PER_MSEC);
 
- out:
-	mutex_unlock(&mvm->mutex);
-	return ret;
+	return 0;
 }
 
 static void iwl_mvm_set_sta_rate(u32 rate_n_flags, struct rate_info *rinfo)
@@ -6184,13 +6105,13 @@ void iwl_mvm_mac_sta_statistics(struct ieee80211_hw *hw,
 	if (!vif->cfg.assoc)
 		return;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	if (mvmvif->deflink.ap_sta_id != mvmsta->deflink.sta_id)
-		goto unlock;
+		return;
 
 	if (iwl_mvm_request_statistics(mvm, false))
-		goto unlock;
+		return;
 
 	sinfo->rx_beacon = 0;
 	for_each_mvm_vif_valid_link(mvmvif, i)
@@ -6204,8 +6125,6 @@ void iwl_mvm_mac_sta_statistics(struct ieee80211_hw *hw,
 			mvmvif->deflink.beacon_stats.avg_signal;
 		sinfo->filled |= BIT_ULL(NL80211_STA_INFO_BEACON_SIGNAL_AVG);
 	}
- unlock:
-	mutex_unlock(&mvm->mutex);
 }
 
 static void iwl_mvm_event_mlme_callback_ini(struct iwl_mvm *mvm,
@@ -6393,9 +6312,8 @@ void iwl_mvm_sync_rx_queues(struct ieee80211_hw *hw)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	iwl_mvm_sync_rx_queues_internal(mvm, IWL_MVM_RXQ_EMPTY, true, NULL, 0);
-	mutex_unlock(&mvm->mutex);
 }
 
 int
@@ -6431,13 +6349,9 @@ int iwl_mvm_start_pmsr(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		       struct cfg80211_pmsr_request *request)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	int ret;
 
-	mutex_lock(&mvm->mutex);
-	ret = iwl_mvm_ftm_start(mvm, vif, request);
-	mutex_unlock(&mvm->mutex);
-
-	return ret;
+	guard(mvm)(mvm);
+	return iwl_mvm_ftm_start(mvm, vif, request);
 }
 
 void iwl_mvm_abort_pmsr(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
@@ -6445,9 +6359,8 @@ void iwl_mvm_abort_pmsr(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	iwl_mvm_ftm_abort(mvm, request);
-	mutex_unlock(&mvm->mutex);
 }
 
 static bool iwl_mvm_can_hw_csum(struct sk_buff *skb)
@@ -6482,7 +6395,6 @@ int iwl_mvm_set_hw_timestamp(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	u32 protocols = 0;
-	int ret;
 
 	/* HW timestamping is only supported for a specific station */
 	if (!hwts->macaddr)
@@ -6492,11 +6404,8 @@ int iwl_mvm_set_hw_timestamp(struct ieee80211_hw *hw,
 		protocols =
 			IWL_TIME_SYNC_PROTOCOL_TM | IWL_TIME_SYNC_PROTOCOL_FTM;
 
-	mutex_lock(&mvm->mutex);
-	ret = iwl_mvm_time_sync_config(mvm, hwts->macaddr, protocols);
-	mutex_unlock(&mvm->mutex);
-
-	return ret;
+	guard(mvm)(mvm);
+	return iwl_mvm_time_sync_config(mvm, hwts->macaddr, protocols);
 }
 
 const struct ieee80211_ops iwl_mvm_hw_ops = {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 0a3b7284eedd..f69f7df40a6d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -12,7 +12,7 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 	int ret;
 	int i;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	iwl_mvm_mac_init_mvmvif(mvm, mvmvif);
 
@@ -32,7 +32,7 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 	/* Allocate resources for the MAC context, and add it to the fw  */
 	ret = iwl_mvm_mac_ctxt_init(mvm, vif);
 	if (ret)
-		goto out_unlock;
+		return ret;
 
 	rcu_assign_pointer(mvm->vif_id_to_mac[mvmvif->id], vif);
 
@@ -46,7 +46,7 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 
 	ret = iwl_mvm_mld_mac_ctxt_add(mvm, vif);
 	if (ret)
-		goto out_unlock;
+		return ret;
 
 	/* beacon filtering */
 	ret = iwl_mvm_disable_beacon_filter(mvm, vif);
@@ -97,7 +97,7 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
 	if (vif->p2p || iwl_fw_lookup_cmd_ver(mvm->fw, PHY_CONTEXT_CMD, 1) < 5)
 		vif->driver_flags |= IEEE80211_VIF_IGNORE_OFDMA_WIDER_BW;
 
-	goto out_unlock;
+	return 0;
 
  out_free_bf:
 	if (mvm->bf_allowed_vif == mvmvif) {
@@ -108,9 +108,6 @@ static int iwl_mvm_mld_mac_add_interface(struct ieee80211_hw *hw,
  out_remove_mac:
 	mvmvif->link[0] = NULL;
 	iwl_mvm_mld_mac_ctxt_remove(mvm, vif);
- out_unlock:
-	mutex_unlock(&mvm->mutex);
-
 	return ret;
 }
 
@@ -127,7 +124,7 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 	      vif->type == NL80211_IFTYPE_ADHOC))
 		iwl_mvm_tcm_rm_vif(mvm, vif);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	if (vif == mvm->csme_vif) {
 		iwl_mei_set_netdev(NULL);
@@ -190,8 +187,6 @@ static void iwl_mvm_mld_mac_remove_interface(struct ieee80211_hw *hw,
 		mvm->monitor_on = false;
 		__clear_bit(IEEE80211_HW_RX_INCLUDES_FCS, mvm->hw->flags);
 	}
-
-	mutex_unlock(&mvm->mutex);
 }
 
 static unsigned int iwl_mvm_mld_count_active_links(struct iwl_mvm_vif *mvmvif)
@@ -389,10 +384,11 @@ static int iwl_mvm_mld_assign_vif_chanctx(struct ieee80211_hw *hw,
 					  struct ieee80211_chanctx_conf *ctx)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	int ret;
 
 	/* update EMLSR mode */
 	if (ieee80211_vif_type_p2p(vif) != NL80211_IFTYPE_STATION) {
+		int ret;
+
 		ret = iwl_mvm_esr_non_bss_link(mvm, vif, link_conf->link_id,
 					       true);
 		/*
@@ -403,11 +399,8 @@ static int iwl_mvm_mld_assign_vif_chanctx(struct ieee80211_hw *hw,
 			return ret;
 	}
 
-	mutex_lock(&mvm->mutex);
-	ret = __iwl_mvm_mld_assign_vif_chanctx(mvm, vif, link_conf, ctx, false);
-	mutex_unlock(&mvm->mutex);
-
-	return ret;
+	guard(mvm)(mvm);
+	return __iwl_mvm_mld_assign_vif_chanctx(mvm, vif, link_conf, ctx, false);
 }
 
 static int iwl_mvm_esr_mode_inactive(struct iwl_mvm *mvm,
@@ -586,7 +579,7 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	int ret;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	if (vif->type == NL80211_IFTYPE_AP)
 		iwl_mvm_send_ap_tx_power_constraint_cmd(mvm, vif, link_conf);
@@ -594,7 +587,7 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 	/* Send the beacon template */
 	ret = iwl_mvm_mac_ctxt_beacon_changed(mvm, vif, link_conf);
 	if (ret)
-		goto out_unlock;
+		return ret;
 
 	/* the link should be already activated when assigning chan context */
 	ret = iwl_mvm_link_changed(mvm, vif, link_conf,
@@ -602,11 +595,11 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 				   ~LINK_CONTEXT_MODIFY_ACTIVE,
 				   true);
 	if (ret)
-		goto out_unlock;
+		return ret;
 
 	ret = iwl_mvm_mld_add_mcast_sta(mvm, vif, link_conf);
 	if (ret)
-		goto out_unlock;
+		return ret;
 
 	/* Send the bcast station. At this stage the TBTT and DTIM time
 	 * events are added and applied to the scheduler
@@ -630,7 +623,7 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 
 	iwl_mvm_ftm_restart_responder(mvm, vif, link_conf);
 
-	goto out_unlock;
+	return 0;
 
 out_failed:
 	iwl_mvm_power_update_mac(mvm);
@@ -638,8 +631,6 @@ out_failed:
 	iwl_mvm_mld_rm_bcast_sta(mvm, vif, link_conf);
 out_rm_mcast:
 	iwl_mvm_mld_rm_mcast_sta(mvm, vif, link_conf);
-out_unlock:
-	mutex_unlock(&mvm->mutex);
 	return ret;
 }
 
@@ -662,7 +653,7 @@ static void iwl_mvm_mld_stop_ap_ibss(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	iwl_mvm_stop_ap_ibss_common(mvm, vif);
 
@@ -676,7 +667,6 @@ static void iwl_mvm_mld_stop_ap_ibss(struct ieee80211_hw *hw,
 	iwl_mvm_mld_rm_mcast_sta(mvm, vif, link_conf);
 
 	iwl_mvm_power_update_mac(mvm);
-	mutex_unlock(&mvm->mutex);
 }
 
 static void iwl_mvm_mld_stop_ap(struct ieee80211_hw *hw,
@@ -979,7 +969,7 @@ static void iwl_mvm_mld_link_info_changed(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	switch (vif->type) {
 	case NL80211_IFTYPE_STATION:
@@ -1005,8 +995,6 @@ static void iwl_mvm_mld_link_info_changed(struct ieee80211_hw *hw,
 				link_conf->txpower);
 		iwl_mvm_set_tx_power(mvm, vif, link_conf->txpower);
 	}
-
-	mutex_unlock(&mvm->mutex);
 }
 
 static void iwl_mvm_mld_vif_cfg_changed(struct ieee80211_hw *hw,
@@ -1015,15 +1003,13 @@ static void iwl_mvm_mld_vif_cfg_changed(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	if (changes & BSS_CHANGED_IDLE && !vif->cfg.idle)
 		iwl_mvm_scan_stop(mvm, IWL_MVM_SCAN_SCHED, true);
 
 	if (vif->type == NL80211_IFTYPE_STATION)
 		iwl_mvm_mld_vif_cfg_changed_station(mvm, vif, changes);
-
-	mutex_unlock(&mvm->mutex);
 }
 
 static int
@@ -1056,9 +1042,8 @@ static void iwl_mvm_mld_config_iface_filter(struct ieee80211_hw *hw,
 	    !vif->p2p)
 		return;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	iwl_mvm_mld_mac_ctxt_changed(mvm, vif, false);
-	mutex_unlock(&mvm->mutex);
 }
 
 static int
@@ -1080,14 +1065,10 @@ iwl_mvm_mld_mac_conf_tx(struct ieee80211_hw *hw,
 	 * The exception is P2P_DEVICE interface which needs immediate update.
 	 */
 	if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
-		int ret;
-
-		mutex_lock(&mvm->mutex);
-		ret = iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
-					   LINK_CONTEXT_MODIFY_QOS_PARAMS,
-					   true);
-		mutex_unlock(&mvm->mutex);
-		return ret;
+		guard(mvm)(mvm);
+		return iwl_mvm_link_changed(mvm, vif, &vif->bss_conf,
+					    LINK_CONTEXT_MODIFY_QOS_PARAMS,
+					    true);
 	}
 	return 0;
 }
@@ -1232,13 +1213,9 @@ iwl_mvm_mld_change_sta_links(struct ieee80211_hw *hw,
 			     u16 old_links, u16 new_links)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	int ret;
-
-	mutex_lock(&mvm->mutex);
-	ret = iwl_mvm_mld_update_sta_links(mvm, vif, sta, old_links, new_links);
-	mutex_unlock(&mvm->mutex);
 
-	return ret;
+	guard(mvm)(mvm);
+	return iwl_mvm_mld_update_sta_links(mvm, vif, sta, old_links, new_links);
 }
 
 bool iwl_mvm_vif_has_esr_cap(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
@@ -1266,26 +1243,19 @@ static bool iwl_mvm_mld_can_activate_links(struct ieee80211_hw *hw,
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	int n_links = hweight16(desired_links);
-	bool ret = true;
 
 	if (n_links <= 1)
 		return true;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	/* Check if HW supports the wanted number of links */
-	if (n_links > iwl_mvm_max_active_links(mvm, vif)) {
-		ret = false;
-		goto unlock;
-	}
+	if (n_links > iwl_mvm_max_active_links(mvm, vif))
+		return false;
 
 	/* If it is an eSR device, check that we can enter eSR */
-	ret = iwl_mvm_is_esr_supported(mvm->fwrt.trans) &&
-	      iwl_mvm_vif_has_esr_cap(mvm, vif);
-
-unlock:
-	mutex_unlock(&mvm->mutex);
-	return ret;
+	return iwl_mvm_is_esr_supported(mvm->fwrt.trans) &&
+	       iwl_mvm_vif_has_esr_cap(mvm, vif);
 }
 
 static enum ieee80211_neg_ttlm_res
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index d732780aae5f..dd4599ca1f45 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -9,6 +9,7 @@
 
 #include <linux/list.h>
 #include <linux/spinlock.h>
+#include <linux/cleanup.h>
 #include <linux/leds.h>
 #include <linux/in6.h>
 
@@ -1332,6 +1333,8 @@ struct iwl_mvm {
 #define IWL_MAC80211_GET_MVM(_hw)			\
 	IWL_OP_MODE_GET_MVM((struct iwl_op_mode *)((_hw)->priv))
 
+DEFINE_GUARD(mvm, struct iwl_mvm *, mutex_lock(&_T->mutex), mutex_unlock(&_T->mutex))
+
 /**
  * enum iwl_mvm_status - MVM status bits
  * @IWL_MVM_STATUS_HW_RFKILL: HW RF-kill is asserted
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index c3cd5ab11aec..160bb619ba9b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -765,20 +765,18 @@ static void iwl_mvm_tx_unblock_dwork(struct work_struct *work)
 	struct ieee80211_vif *tx_blocked_vif;
 	struct iwl_mvm_vif *mvmvif;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	tx_blocked_vif =
 		rcu_dereference_protected(mvm->csa_tx_blocked_vif,
 					  lockdep_is_held(&mvm->mutex));
 
 	if (!tx_blocked_vif)
-		goto unlock;
+		return;
 
 	mvmvif = iwl_mvm_vif_from_mac80211(tx_blocked_vif);
 	iwl_mvm_modify_all_sta_disable_tx(mvm, mvmvif, false);
 	RCU_INIT_POINTER(mvm->csa_tx_blocked_vif, NULL);
-unlock:
-	mutex_unlock(&mvm->mutex);
 }
 
 static void iwl_mvm_fwrt_dump_start(void *ctx)
@@ -798,13 +796,9 @@ static void iwl_mvm_fwrt_dump_end(void *ctx)
 static int iwl_mvm_fwrt_send_hcmd(void *ctx, struct iwl_host_cmd *host_cmd)
 {
 	struct iwl_mvm *mvm = (struct iwl_mvm *)ctx;
-	int ret;
 
-	mutex_lock(&mvm->mutex);
-	ret = iwl_mvm_send_cmd(mvm, host_cmd);
-	mutex_unlock(&mvm->mutex);
-
-	return ret;
+	guard(mvm)(mvm);
+	return iwl_mvm_send_cmd(mvm, host_cmd);
 }
 
 static bool iwl_mvm_d3_debug_enable(void *ctx)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tdls.c b/drivers/net/wireless/intel/iwlwifi/mvm/tdls.c
index e7d5f4ebeb25..3d25ff5cd7e8 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tdls.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tdls.c
@@ -2,7 +2,7 @@
 /*
  * Copyright (C) 2014 Intel Mobile Communications GmbH
  * Copyright (C) 2017 Intel Deutschland GmbH
- * Copyright (C) 2018-2020, 2022-2023 Intel Corporation
+ * Copyright (C) 2018-2020, 2022-2024 Intel Corporation
  */
 #include <linux/etherdevice.h>
 #include "mvm.h"
@@ -151,7 +151,7 @@ void iwl_mvm_mac_mgd_protect_tdls_discover(struct ieee80211_hw *hw,
 	u32 duration = 2 * vif->bss_conf.dtim_period * vif->bss_conf.beacon_int;
 
 	/* Protect the session to hear the TDLS setup response on the channel */
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 	if (fw_has_capa(&mvm->fw->ucode_capa,
 			IWL_UCODE_TLV_CAPA_SESSION_PROT_CMD))
 		iwl_mvm_schedule_session_protection(mvm, vif, duration,
@@ -159,7 +159,6 @@ void iwl_mvm_mac_mgd_protect_tdls_discover(struct ieee80211_hw *hw,
 	else
 		iwl_mvm_protect_session(mvm, vif, duration,
 					duration, 100, true);
-	mutex_unlock(&mvm->mutex);
 }
 
 static const char *
@@ -460,21 +459,21 @@ void iwl_mvm_tdls_ch_switch_work(struct work_struct *work)
 	int ret;
 
 	mvm = container_of(work, struct iwl_mvm, tdls_cs.dwork.work);
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	/* called after an active channel switch has finished or timed-out */
 	iwl_mvm_tdls_update_cs_state(mvm, IWL_MVM_TDLS_SW_IDLE);
 
 	/* station might be gone, in that case do nothing */
 	if (mvm->tdls_cs.peer.sta_id == IWL_MVM_INVALID_STA)
-		goto out;
+		return;
 
 	sta = rcu_dereference_protected(
 				mvm->fw_id_to_mac_id[mvm->tdls_cs.peer.sta_id],
 				lockdep_is_held(&mvm->mutex));
 	/* the station may not be here, but if it is, it must be a TDLS peer */
 	if (!sta || IS_ERR(sta) || WARN_ON(!sta->tdls))
-		goto out;
+		return;
 
 	mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	vif = mvmsta->vif;
@@ -493,8 +492,6 @@ void iwl_mvm_tdls_ch_switch_work(struct work_struct *work)
 	/* retry after a DTIM if we failed sending now */
 	delay = TU_TO_MS(vif->bss_conf.dtim_period * vif->bss_conf.beacon_int);
 	schedule_delayed_work(&mvm->tdls_cs.dwork, msecs_to_jiffies(delay));
-out:
-	mutex_unlock(&mvm->mutex);
 }
 
 int
@@ -509,7 +506,7 @@ iwl_mvm_tdls_channel_switch(struct ieee80211_hw *hw,
 	unsigned int delay;
 	int ret;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	IWL_DEBUG_TDLS(mvm, "TDLS channel switch with %pM ch %d width %d\n",
 		       sta->addr, chandef->chan->center_freq, chandef->width);
@@ -519,8 +516,7 @@ iwl_mvm_tdls_channel_switch(struct ieee80211_hw *hw,
 		IWL_DEBUG_TDLS(mvm,
 			       "Existing peer. Can't start switch with %pM\n",
 			       sta->addr);
-		ret = -EBUSY;
-		goto out;
+		return -EBUSY;
 	}
 
 	ret = iwl_mvm_tdls_config_channel_switch(mvm, vif,
@@ -529,17 +525,15 @@ iwl_mvm_tdls_channel_switch(struct ieee80211_hw *hw,
 						 oper_class, chandef, 0, 0, 0,
 						 tmpl_skb, ch_sw_tm_ie);
 	if (ret)
-		goto out;
+		return ret;
 
 	/*
 	 * Mark the peer as "in tdls switch" for this vif. We only allow a
 	 * single such peer per vif.
 	 */
 	mvm->tdls_cs.peer.skb = skb_copy(tmpl_skb, GFP_KERNEL);
-	if (!mvm->tdls_cs.peer.skb) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (!mvm->tdls_cs.peer.skb)
+		return -ENOMEM;
 
 	mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	mvm->tdls_cs.peer.sta_id = mvmsta->deflink.sta_id;
@@ -556,10 +550,7 @@ iwl_mvm_tdls_channel_switch(struct ieee80211_hw *hw,
 			     vif->bss_conf.beacon_int);
 	mod_delayed_work(system_wq, &mvm->tdls_cs.dwork,
 			 msecs_to_jiffies(delay));
-
-out:
-	mutex_unlock(&mvm->mutex);
-	return ret;
+	return 0;
 }
 
 void iwl_mvm_tdls_cancel_channel_switch(struct ieee80211_hw *hw,
@@ -626,7 +617,7 @@ iwl_mvm_tdls_recv_channel_switch(struct ieee80211_hw *hw,
 		params->action_code == WLAN_TDLS_CHANNEL_SWITCH_REQUEST ?
 		"REQ" : "RESP";
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	IWL_DEBUG_TDLS(mvm,
 		       "Received TDLS ch switch action %s from %pM status %d\n",
@@ -670,5 +661,4 @@ retry:
 		1024 / 1000;
 	mod_delayed_work(system_wq, &mvm->tdls_cs.dwork,
 			 msecs_to_jiffies(delay));
-	mutex_unlock(&mvm->mutex);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
index 61a4638d1be2..45b431ffbc97 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2012-2014, 2019-2022 Intel Corporation
+ * Copyright (C) 2012-2014, 2019-2022, 2024 Intel Corporation
  * Copyright (C) 2013-2014 Intel Mobile Communications GmbH
  * Copyright (C) 2015-2016 Intel Deutschland GmbH
  */
@@ -618,48 +618,35 @@ static int iwl_mvm_tzone_get_temp(struct thermal_zone_device *device,
 	int ret;
 	int temp;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	if (!iwl_mvm_firmware_running(mvm) ||
-	    mvm->fwrt.cur_fw_img != IWL_UCODE_REGULAR) {
-		ret = -ENODATA;
-		goto out;
-	}
+	    mvm->fwrt.cur_fw_img != IWL_UCODE_REGULAR)
+		return -ENODATA;
 
 	ret = iwl_mvm_get_temp(mvm, &temp);
 	if (ret)
-		goto out;
+		return ret;
 
 	*temperature = temp * 1000;
-
-out:
-	mutex_unlock(&mvm->mutex);
-	return ret;
+	return 0;
 }
 
 static int iwl_mvm_tzone_set_trip_temp(struct thermal_zone_device *device,
 				       int trip, int temp)
 {
 	struct iwl_mvm *mvm = thermal_zone_device_priv(device);
-	int ret;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	if (!iwl_mvm_firmware_running(mvm) ||
-	    mvm->fwrt.cur_fw_img != IWL_UCODE_REGULAR) {
-		ret = -EIO;
-		goto out;
-	}
+	    mvm->fwrt.cur_fw_img != IWL_UCODE_REGULAR)
+		return -EIO;
 
-	if ((temp / 1000) > S16_MAX) {
-		ret = -EINVAL;
-		goto out;
-	}
+	if ((temp / 1000) > S16_MAX)
+		return -EINVAL;
 
-	ret = iwl_mvm_send_temp_report_ths_cmd(mvm);
-out:
-	mutex_unlock(&mvm->mutex);
-	return ret;
+	return iwl_mvm_send_temp_report_ths_cmd(mvm);
 }
 
 static  struct thermal_zone_device_ops tzone_ops = {
@@ -733,27 +720,18 @@ static int iwl_mvm_tcool_set_cur_state(struct thermal_cooling_device *cdev,
 				       unsigned long new_state)
 {
 	struct iwl_mvm *mvm = (struct iwl_mvm *)(cdev->devdata);
-	int ret;
 
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	if (!iwl_mvm_firmware_running(mvm) ||
-	    mvm->fwrt.cur_fw_img != IWL_UCODE_REGULAR) {
-		ret = -EIO;
-		goto unlock;
-	}
+	    mvm->fwrt.cur_fw_img != IWL_UCODE_REGULAR)
+		return -EIO;
 
-	if (new_state >= ARRAY_SIZE(iwl_mvm_cdev_budgets)) {
-		ret = -EINVAL;
-		goto unlock;
-	}
-
-	ret = iwl_mvm_ctdp_command(mvm, CTDP_CMD_OPERATION_START,
-				   new_state);
+	if (new_state >= ARRAY_SIZE(iwl_mvm_cdev_budgets))
+		return -EINVAL;
 
-unlock:
-	mutex_unlock(&mvm->mutex);
-	return ret;
+	return iwl_mvm_ctdp_command(mvm, CTDP_CMD_OPERATION_START,
+				    new_state);
 }
 
 static const struct thermal_cooling_device_ops tcooling_ops = {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
index 47283a358ffd..0e5fa8374103 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c
@@ -892,7 +892,7 @@ static void iwl_mvm_tcm_iter(void *_data, u8 *mac, struct ieee80211_vif *vif)
 
 static void iwl_mvm_tcm_results(struct iwl_mvm *mvm)
 {
-	mutex_lock(&mvm->mutex);
+	guard(mvm)(mvm);
 
 	ieee80211_iterate_active_interfaces(
 		mvm->hw, IEEE80211_IFACE_ITER_NORMAL,
@@ -900,8 +900,6 @@ static void iwl_mvm_tcm_results(struct iwl_mvm *mvm)
 
 	if (fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_UMAC_SCAN))
 		iwl_mvm_config_scan(mvm);
-
-	mutex_unlock(&mvm->mutex);
 }
 
 static void iwl_mvm_tcm_uapsd_nonagg_detected_wk(struct work_struct *wk)
@@ -1130,10 +1128,9 @@ void iwl_mvm_recalc_tcm(struct iwl_mvm *mvm)
 	spin_unlock(&mvm->tcm.lock);
 
 	if (handle_uapsd && iwl_mvm_has_new_rx_api(mvm)) {
-		mutex_lock(&mvm->mutex);
+		guard(mvm)(mvm);
 		if (iwl_mvm_request_statistics(mvm, true))
 			handle_uapsd = false;
-		mutex_unlock(&mvm->mutex);
 	}
 
 	spin_lock(&mvm->tcm.lock);
-- 
cgit v1.2.3-58-ga151


From 46144103ace2863e26f4e911aa45200753b7dbca Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Sun, 12 May 2024 15:24:59 +0300
Subject: wifi: iwlwifi: mvm: don't skip link selection

If we exit EMLSR due to a IWL_MVM_ESR_EXIT*, a MLO scan followed by a
link selection is scheduled with a delay of 30 seconds.
If during that 30 seconds EMLSR was blocked and unblocked
(IWL_MVM_ESR_BLOCKED*), we would still want to get the needed data from
the MLO scan and select link accordingly, and not return immediately to
EMLSR.

Fixes: 2f33561ea8f9 ("wifi: iwlwifi: mvm: trigger link selection after exiting EMLSR")
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240512152312.caab27a8dd8f.I63f67e213d5e05416f71513a8d914917d59aa44f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/link.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index fd8a21891d26..0745229c8c37 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -1076,6 +1076,13 @@ static void iwl_mvm_esr_unblocked(struct iwl_mvm *mvm,
 
 	IWL_DEBUG_INFO(mvm, "EMLSR is unblocked\n");
 
+	/* We exited due to an EXIT reason, so MLO scan was scheduled already */
+	if (mvmvif->last_esr_exit.reason &&
+	    !(mvmvif->last_esr_exit.reason & IWL_MVM_BLOCK_ESR_REASONS)) {
+		IWL_DEBUG_INFO(mvm, "Wait for MLO scan\n");
+		return;
+	}
+
 	/*
 	 * If EMLSR was blocked for more than 30 seconds, or the last link
 	 * selection decided to not enter EMLSR, trigger a new scan.
-- 
cgit v1.2.3-58-ga151


From b476564b90668ea957db2411fc946753cdc36d8d Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Sun, 12 May 2024 15:25:02 +0300
Subject: wifi: iwlwifi: kill iwl-eeprom-read

This is used by dvm only, move to dvm.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512152312.72f4bf256c8d.I7433bda9b0fc9eece5210db2cb90c2f03973f5ec@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/Makefile        |   2 +-
 drivers/net/wireless/intel/iwlwifi/dvm/Makefile    |   2 +-
 drivers/net/wireless/intel/iwlwifi/dvm/agn.h       |   4 +-
 drivers/net/wireless/intel/iwlwifi/dvm/eeprom.c    | 393 ++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/dvm/main.c      |   4 +-
 .../net/wireless/intel/iwlwifi/iwl-eeprom-read.c   | 394 ---------------------
 .../net/wireless/intel/iwlwifi/iwl-eeprom-read.h   |  12 -
 drivers/net/wireless/intel/iwlwifi/mvm/nvm.c       |   3 +-
 8 files changed, 401 insertions(+), 413 deletions(-)
 create mode 100644 drivers/net/wireless/intel/iwlwifi/dvm/eeprom.c
 delete mode 100644 drivers/net/wireless/intel/iwlwifi/iwl-eeprom-read.c
 delete mode 100644 drivers/net/wireless/intel/iwlwifi/iwl-eeprom-read.h

diff --git a/drivers/net/wireless/intel/iwlwifi/Makefile b/drivers/net/wireless/intel/iwlwifi/Makefile
index 8bb94a4c12cd..2c2c18cef9ff 100644
--- a/drivers/net/wireless/intel/iwlwifi/Makefile
+++ b/drivers/net/wireless/intel/iwlwifi/Makefile
@@ -4,7 +4,7 @@ obj-$(CONFIG_IWLWIFI)	+= iwlwifi.o
 iwlwifi-objs		+= iwl-io.o
 iwlwifi-objs		+= iwl-drv.o
 iwlwifi-objs		+= iwl-debug.o
-iwlwifi-objs		+= iwl-eeprom-read.o iwl-eeprom-parse.o
+iwlwifi-objs		+= iwl-eeprom-parse.o
 iwlwifi-objs		+= iwl-phy-db.o iwl-nvm-parse.o
 iwlwifi-objs		+= pcie/drv.o pcie/rx.o pcie/tx.o pcie/trans.o
 iwlwifi-objs		+= pcie/ctxt-info.o pcie/ctxt-info-gen3.o
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/Makefile b/drivers/net/wireless/intel/iwlwifi/dvm/Makefile
index 0486b17d7c41..2d0b6d287af4 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/Makefile
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/Makefile
@@ -2,7 +2,7 @@
 # DVM
 obj-$(CONFIG_IWLDVM)	+= iwldvm.o
 iwldvm-objs		+= main.o rs.o mac80211.o ucode.o tx.o
-iwldvm-objs		+= lib.o calib.o tt.o sta.o rx.o
+iwldvm-objs		+= lib.o calib.o tt.o sta.o rx.o eeprom.o
 
 iwldvm-objs		+= power.o
 iwldvm-objs		+= scan.o
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/agn.h b/drivers/net/wireless/intel/iwlwifi/dvm/agn.h
index fefaa414272b..47d21413ba59 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/agn.h
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/agn.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2005-2014, 2021 Intel Corporation
+ * Copyright (C) 2005-2014, 2021, 2024 Intel Corporation
  */
 #ifndef __iwl_agn_h__
 #define __iwl_agn_h__
@@ -385,6 +385,8 @@ static inline void iwl_dvm_set_pmi(struct iwl_priv *priv, bool state)
 	iwl_trans_set_pmi(priv->trans, state);
 }
 
+int iwl_read_eeprom(struct iwl_trans *trans, u8 **eeprom, size_t *eeprom_size);
+
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 void iwl_dbgfs_register(struct iwl_priv *priv, struct dentry *dbgfs_dir);
 #else
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/eeprom.c b/drivers/net/wireless/intel/iwlwifi/dvm/eeprom.c
new file mode 100644
index 000000000000..3972775fc133
--- /dev/null
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/eeprom.c
@@ -0,0 +1,393 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/*
+ * Copyright (C) 2005-2014, 2018-2019, 2021, 2024 Intel Corporation
+ */
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/export.h>
+
+#include "iwl-drv.h"
+#include "iwl-debug.h"
+#include "iwl-io.h"
+#include "iwl-prph.h"
+#include "iwl-csr.h"
+#include "agn.h"
+
+/*
+ * EEPROM access time values:
+ *
+ * Driver initiates EEPROM read by writing byte address << 1 to CSR_EEPROM_REG.
+ * Driver then polls CSR_EEPROM_REG for CSR_EEPROM_REG_READ_VALID_MSK (0x1).
+ * When polling, wait 10 uSec between polling loops, up to a maximum 5000 uSec.
+ * Driver reads 16-bit value from bits 31-16 of CSR_EEPROM_REG.
+ */
+#define IWL_EEPROM_ACCESS_TIMEOUT	5000 /* uSec */
+
+/*
+ * The device's EEPROM semaphore prevents conflicts between driver and uCode
+ * when accessing the EEPROM; each access is a series of pulses to/from the
+ * EEPROM chip, not a single event, so even reads could conflict if they
+ * weren't arbitrated by the semaphore.
+ */
+#define IWL_EEPROM_SEM_TIMEOUT		10   /* microseconds */
+#define IWL_EEPROM_SEM_RETRY_LIMIT	1000 /* number of attempts (not time) */
+
+
+static int iwl_eeprom_acquire_semaphore(struct iwl_trans *trans)
+{
+	u16 count;
+	int ret;
+
+	for (count = 0; count < IWL_EEPROM_SEM_RETRY_LIMIT; count++) {
+		/* Request semaphore */
+		iwl_set_bit(trans, CSR_HW_IF_CONFIG_REG,
+			    CSR_HW_IF_CONFIG_REG_BIT_EEPROM_OWN_SEM);
+
+		/* See if we got it */
+		ret = iwl_poll_bit(trans, CSR_HW_IF_CONFIG_REG,
+				CSR_HW_IF_CONFIG_REG_BIT_EEPROM_OWN_SEM,
+				CSR_HW_IF_CONFIG_REG_BIT_EEPROM_OWN_SEM,
+				IWL_EEPROM_SEM_TIMEOUT);
+		if (ret >= 0) {
+			IWL_DEBUG_EEPROM(trans->dev,
+					 "Acquired semaphore after %d tries.\n",
+					 count+1);
+			return ret;
+		}
+	}
+
+	return ret;
+}
+
+static void iwl_eeprom_release_semaphore(struct iwl_trans *trans)
+{
+	iwl_clear_bit(trans, CSR_HW_IF_CONFIG_REG,
+		      CSR_HW_IF_CONFIG_REG_BIT_EEPROM_OWN_SEM);
+}
+
+static int iwl_eeprom_verify_signature(struct iwl_trans *trans, bool nvm_is_otp)
+{
+	u32 gp = iwl_read32(trans, CSR_EEPROM_GP) & CSR_EEPROM_GP_VALID_MSK;
+
+	IWL_DEBUG_EEPROM(trans->dev, "EEPROM signature=0x%08x\n", gp);
+
+	switch (gp) {
+	case CSR_EEPROM_GP_BAD_SIG_EEP_GOOD_SIG_OTP:
+		if (!nvm_is_otp) {
+			IWL_ERR(trans, "EEPROM with bad signature: 0x%08x\n",
+				gp);
+			return -ENOENT;
+		}
+		return 0;
+	case CSR_EEPROM_GP_GOOD_SIG_EEP_LESS_THAN_4K:
+	case CSR_EEPROM_GP_GOOD_SIG_EEP_MORE_THAN_4K:
+		if (nvm_is_otp) {
+			IWL_ERR(trans, "OTP with bad signature: 0x%08x\n", gp);
+			return -ENOENT;
+		}
+		return 0;
+	case CSR_EEPROM_GP_BAD_SIGNATURE_BOTH_EEP_AND_OTP:
+	default:
+		IWL_ERR(trans,
+			"bad EEPROM/OTP signature, type=%s, EEPROM_GP=0x%08x\n",
+			nvm_is_otp ? "OTP" : "EEPROM", gp);
+		return -ENOENT;
+	}
+}
+
+/******************************************************************************
+ *
+ * OTP related functions
+ *
+******************************************************************************/
+
+static void iwl_set_otp_access_absolute(struct iwl_trans *trans)
+{
+	iwl_read32(trans, CSR_OTP_GP_REG);
+
+	iwl_clear_bit(trans, CSR_OTP_GP_REG,
+		      CSR_OTP_GP_REG_OTP_ACCESS_MODE);
+}
+
+static int iwl_nvm_is_otp(struct iwl_trans *trans)
+{
+	u32 otpgp;
+
+	/* OTP only valid for CP/PP and after */
+	switch (trans->hw_rev & CSR_HW_REV_TYPE_MSK) {
+	case CSR_HW_REV_TYPE_NONE:
+		IWL_ERR(trans, "Unknown hardware type\n");
+		return -EIO;
+	case CSR_HW_REV_TYPE_5300:
+	case CSR_HW_REV_TYPE_5350:
+	case CSR_HW_REV_TYPE_5100:
+	case CSR_HW_REV_TYPE_5150:
+		return 0;
+	default:
+		otpgp = iwl_read32(trans, CSR_OTP_GP_REG);
+		if (otpgp & CSR_OTP_GP_REG_DEVICE_SELECT)
+			return 1;
+		return 0;
+	}
+}
+
+static int iwl_init_otp_access(struct iwl_trans *trans)
+{
+	int ret;
+
+	ret = iwl_finish_nic_init(trans);
+	if (ret)
+		return ret;
+
+	iwl_set_bits_prph(trans, APMG_PS_CTRL_REG,
+			  APMG_PS_CTRL_VAL_RESET_REQ);
+	udelay(5);
+	iwl_clear_bits_prph(trans, APMG_PS_CTRL_REG,
+			    APMG_PS_CTRL_VAL_RESET_REQ);
+
+	/*
+	 * CSR auto clock gate disable bit -
+	 * this is only applicable for HW with OTP shadow RAM
+	 */
+	if (trans->trans_cfg->base_params->shadow_ram_support)
+		iwl_set_bit(trans, CSR_DBG_LINK_PWR_MGMT_REG,
+			    CSR_RESET_LINK_PWR_MGMT_DISABLED);
+
+	return 0;
+}
+
+static int iwl_read_otp_word(struct iwl_trans *trans, u16 addr,
+			     __le16 *eeprom_data)
+{
+	int ret = 0;
+	u32 r;
+	u32 otpgp;
+
+	iwl_write32(trans, CSR_EEPROM_REG,
+		    CSR_EEPROM_REG_MSK_ADDR & (addr << 1));
+	ret = iwl_poll_bit(trans, CSR_EEPROM_REG,
+				 CSR_EEPROM_REG_READ_VALID_MSK,
+				 CSR_EEPROM_REG_READ_VALID_MSK,
+				 IWL_EEPROM_ACCESS_TIMEOUT);
+	if (ret < 0) {
+		IWL_ERR(trans, "Time out reading OTP[%d]\n", addr);
+		return ret;
+	}
+	r = iwl_read32(trans, CSR_EEPROM_REG);
+	/* check for ECC errors: */
+	otpgp = iwl_read32(trans, CSR_OTP_GP_REG);
+	if (otpgp & CSR_OTP_GP_REG_ECC_UNCORR_STATUS_MSK) {
+		/* stop in this case */
+		/* set the uncorrectable OTP ECC bit for acknowledgment */
+		iwl_set_bit(trans, CSR_OTP_GP_REG,
+			    CSR_OTP_GP_REG_ECC_UNCORR_STATUS_MSK);
+		IWL_ERR(trans, "Uncorrectable OTP ECC error, abort OTP read\n");
+		return -EINVAL;
+	}
+	if (otpgp & CSR_OTP_GP_REG_ECC_CORR_STATUS_MSK) {
+		/* continue in this case */
+		/* set the correctable OTP ECC bit for acknowledgment */
+		iwl_set_bit(trans, CSR_OTP_GP_REG,
+			    CSR_OTP_GP_REG_ECC_CORR_STATUS_MSK);
+		IWL_ERR(trans, "Correctable OTP ECC error, continue read\n");
+	}
+	*eeprom_data = cpu_to_le16(r >> 16);
+	return 0;
+}
+
+/*
+ * iwl_is_otp_empty: check for empty OTP
+ */
+static bool iwl_is_otp_empty(struct iwl_trans *trans)
+{
+	u16 next_link_addr = 0;
+	__le16 link_value;
+	bool is_empty = false;
+
+	/* locate the beginning of OTP link list */
+	if (!iwl_read_otp_word(trans, next_link_addr, &link_value)) {
+		if (!link_value) {
+			IWL_ERR(trans, "OTP is empty\n");
+			is_empty = true;
+		}
+	} else {
+		IWL_ERR(trans, "Unable to read first block of OTP list.\n");
+		is_empty = true;
+	}
+
+	return is_empty;
+}
+
+
+/*
+ * iwl_find_otp_image: find EEPROM image in OTP
+ *   finding the OTP block that contains the EEPROM image.
+ *   the last valid block on the link list (the block _before_ the last block)
+ *   is the block we should read and used to configure the device.
+ *   If all the available OTP blocks are full, the last block will be the block
+ *   we should read and used to configure the device.
+ *   only perform this operation if shadow RAM is disabled
+ */
+static int iwl_find_otp_image(struct iwl_trans *trans,
+					u16 *validblockaddr)
+{
+	u16 next_link_addr = 0, valid_addr;
+	__le16 link_value = 0;
+	int usedblocks = 0;
+
+	/* set addressing mode to absolute to traverse the link list */
+	iwl_set_otp_access_absolute(trans);
+
+	/* checking for empty OTP or error */
+	if (iwl_is_otp_empty(trans))
+		return -EINVAL;
+
+	/*
+	 * start traverse link list
+	 * until reach the max number of OTP blocks
+	 * different devices have different number of OTP blocks
+	 */
+	do {
+		/* save current valid block address
+		 * check for more block on the link list
+		 */
+		valid_addr = next_link_addr;
+		next_link_addr = le16_to_cpu(link_value) * sizeof(u16);
+		IWL_DEBUG_EEPROM(trans->dev, "OTP blocks %d addr 0x%x\n",
+				 usedblocks, next_link_addr);
+		if (iwl_read_otp_word(trans, next_link_addr, &link_value))
+			return -EINVAL;
+		if (!link_value) {
+			/*
+			 * reach the end of link list, return success and
+			 * set address point to the starting address
+			 * of the image
+			 */
+			*validblockaddr = valid_addr;
+			/* skip first 2 bytes (link list pointer) */
+			*validblockaddr += 2;
+			return 0;
+		}
+		/* more in the link list, continue */
+		usedblocks++;
+	} while (usedblocks <= trans->trans_cfg->base_params->max_ll_items);
+
+	/* OTP has no valid blocks */
+	IWL_DEBUG_EEPROM(trans->dev, "OTP has no valid blocks\n");
+	return -EINVAL;
+}
+
+/*
+ * iwl_read_eeprom - read EEPROM contents
+ *
+ * Load the EEPROM contents from adapter and return it
+ * and its size.
+ *
+ * NOTE:  This routine uses the non-debug IO access functions.
+ */
+int iwl_read_eeprom(struct iwl_trans *trans, u8 **eeprom, size_t *eeprom_size)
+{
+	__le16 *e;
+	u32 gp = iwl_read32(trans, CSR_EEPROM_GP);
+	int sz;
+	int ret;
+	u16 addr;
+	u16 validblockaddr = 0;
+	u16 cache_addr = 0;
+	int nvm_is_otp;
+
+	if (!eeprom || !eeprom_size)
+		return -EINVAL;
+
+	nvm_is_otp = iwl_nvm_is_otp(trans);
+	if (nvm_is_otp < 0)
+		return nvm_is_otp;
+
+	sz = trans->trans_cfg->base_params->eeprom_size;
+	IWL_DEBUG_EEPROM(trans->dev, "NVM size = %d\n", sz);
+
+	e = kmalloc(sz, GFP_KERNEL);
+	if (!e)
+		return -ENOMEM;
+
+	ret = iwl_eeprom_verify_signature(trans, nvm_is_otp);
+	if (ret < 0) {
+		IWL_ERR(trans, "EEPROM not found, EEPROM_GP=0x%08x\n", gp);
+		goto err_free;
+	}
+
+	/* Make sure driver (instead of uCode) is allowed to read EEPROM */
+	ret = iwl_eeprom_acquire_semaphore(trans);
+	if (ret < 0) {
+		IWL_ERR(trans, "Failed to acquire EEPROM semaphore.\n");
+		goto err_free;
+	}
+
+	if (nvm_is_otp) {
+		ret = iwl_init_otp_access(trans);
+		if (ret) {
+			IWL_ERR(trans, "Failed to initialize OTP access.\n");
+			goto err_unlock;
+		}
+
+		iwl_write32(trans, CSR_EEPROM_GP,
+			    iwl_read32(trans, CSR_EEPROM_GP) &
+			    ~CSR_EEPROM_GP_IF_OWNER_MSK);
+
+		iwl_set_bit(trans, CSR_OTP_GP_REG,
+			    CSR_OTP_GP_REG_ECC_CORR_STATUS_MSK |
+			    CSR_OTP_GP_REG_ECC_UNCORR_STATUS_MSK);
+		/* traversing the linked list if no shadow ram supported */
+		if (!trans->trans_cfg->base_params->shadow_ram_support) {
+			ret = iwl_find_otp_image(trans, &validblockaddr);
+			if (ret)
+				goto err_unlock;
+		}
+		for (addr = validblockaddr; addr < validblockaddr + sz;
+		     addr += sizeof(u16)) {
+			__le16 eeprom_data;
+
+			ret = iwl_read_otp_word(trans, addr, &eeprom_data);
+			if (ret)
+				goto err_unlock;
+			e[cache_addr / 2] = eeprom_data;
+			cache_addr += sizeof(u16);
+		}
+	} else {
+		/* eeprom is an array of 16bit values */
+		for (addr = 0; addr < sz; addr += sizeof(u16)) {
+			u32 r;
+
+			iwl_write32(trans, CSR_EEPROM_REG,
+				    CSR_EEPROM_REG_MSK_ADDR & (addr << 1));
+
+			ret = iwl_poll_bit(trans, CSR_EEPROM_REG,
+					   CSR_EEPROM_REG_READ_VALID_MSK,
+					   CSR_EEPROM_REG_READ_VALID_MSK,
+					   IWL_EEPROM_ACCESS_TIMEOUT);
+			if (ret < 0) {
+				IWL_ERR(trans,
+					"Time out reading EEPROM[%d]\n", addr);
+				goto err_unlock;
+			}
+			r = iwl_read32(trans, CSR_EEPROM_REG);
+			e[addr / 2] = cpu_to_le16(r >> 16);
+		}
+	}
+
+	IWL_DEBUG_EEPROM(trans->dev, "NVM Type: %s\n",
+			 nvm_is_otp ? "OTP" : "EEPROM");
+
+	iwl_eeprom_release_semaphore(trans);
+
+	*eeprom_size = sz;
+	*eeprom = (u8 *)e;
+	return 0;
+
+ err_unlock:
+	iwl_eeprom_release_semaphore(trans);
+ err_free:
+	kfree(e);
+
+	return ret;
+}
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/main.c b/drivers/net/wireless/intel/iwlwifi/dvm/main.c
index 8774dd7b921e..7bb3011e12e3 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/main.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/main.c
@@ -1,7 +1,8 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /******************************************************************************
  *
- * Copyright(c) 2003 - 2014, 2018 - 2022  Intel Corporation. All rights reserved.
+ * Copyright(c) 2003 - 2014, 2018 - 2022 Intel Corporation. All rights reserved.
+ * Copyright(c) 2024 Intel Corporation. All rights reserved.
  * Copyright(c) 2015 Intel Deutschland GmbH
  *
  * Portions of this file are derived from the ipw3945 project, as well
@@ -25,7 +26,6 @@
 
 #include <asm/div64.h>
 
-#include "iwl-eeprom-read.h"
 #include "iwl-eeprom-parse.h"
 #include "iwl-io.h"
 #include "iwl-trans.h"
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-read.c b/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-read.c
deleted file mode 100644
index 5f386bb1a353..000000000000
--- a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-read.c
+++ /dev/null
@@ -1,394 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
-/*
- * Copyright (C) 2005-2014, 2018-2019, 2021 Intel Corporation
- */
-#include <linux/types.h>
-#include <linux/slab.h>
-#include <linux/export.h>
-
-#include "iwl-drv.h"
-#include "iwl-debug.h"
-#include "iwl-eeprom-read.h"
-#include "iwl-io.h"
-#include "iwl-prph.h"
-#include "iwl-csr.h"
-
-/*
- * EEPROM access time values:
- *
- * Driver initiates EEPROM read by writing byte address << 1 to CSR_EEPROM_REG.
- * Driver then polls CSR_EEPROM_REG for CSR_EEPROM_REG_READ_VALID_MSK (0x1).
- * When polling, wait 10 uSec between polling loops, up to a maximum 5000 uSec.
- * Driver reads 16-bit value from bits 31-16 of CSR_EEPROM_REG.
- */
-#define IWL_EEPROM_ACCESS_TIMEOUT	5000 /* uSec */
-
-/*
- * The device's EEPROM semaphore prevents conflicts between driver and uCode
- * when accessing the EEPROM; each access is a series of pulses to/from the
- * EEPROM chip, not a single event, so even reads could conflict if they
- * weren't arbitrated by the semaphore.
- */
-#define IWL_EEPROM_SEM_TIMEOUT		10   /* microseconds */
-#define IWL_EEPROM_SEM_RETRY_LIMIT	1000 /* number of attempts (not time) */
-
-
-static int iwl_eeprom_acquire_semaphore(struct iwl_trans *trans)
-{
-	u16 count;
-	int ret;
-
-	for (count = 0; count < IWL_EEPROM_SEM_RETRY_LIMIT; count++) {
-		/* Request semaphore */
-		iwl_set_bit(trans, CSR_HW_IF_CONFIG_REG,
-			    CSR_HW_IF_CONFIG_REG_BIT_EEPROM_OWN_SEM);
-
-		/* See if we got it */
-		ret = iwl_poll_bit(trans, CSR_HW_IF_CONFIG_REG,
-				CSR_HW_IF_CONFIG_REG_BIT_EEPROM_OWN_SEM,
-				CSR_HW_IF_CONFIG_REG_BIT_EEPROM_OWN_SEM,
-				IWL_EEPROM_SEM_TIMEOUT);
-		if (ret >= 0) {
-			IWL_DEBUG_EEPROM(trans->dev,
-					 "Acquired semaphore after %d tries.\n",
-					 count+1);
-			return ret;
-		}
-	}
-
-	return ret;
-}
-
-static void iwl_eeprom_release_semaphore(struct iwl_trans *trans)
-{
-	iwl_clear_bit(trans, CSR_HW_IF_CONFIG_REG,
-		      CSR_HW_IF_CONFIG_REG_BIT_EEPROM_OWN_SEM);
-}
-
-static int iwl_eeprom_verify_signature(struct iwl_trans *trans, bool nvm_is_otp)
-{
-	u32 gp = iwl_read32(trans, CSR_EEPROM_GP) & CSR_EEPROM_GP_VALID_MSK;
-
-	IWL_DEBUG_EEPROM(trans->dev, "EEPROM signature=0x%08x\n", gp);
-
-	switch (gp) {
-	case CSR_EEPROM_GP_BAD_SIG_EEP_GOOD_SIG_OTP:
-		if (!nvm_is_otp) {
-			IWL_ERR(trans, "EEPROM with bad signature: 0x%08x\n",
-				gp);
-			return -ENOENT;
-		}
-		return 0;
-	case CSR_EEPROM_GP_GOOD_SIG_EEP_LESS_THAN_4K:
-	case CSR_EEPROM_GP_GOOD_SIG_EEP_MORE_THAN_4K:
-		if (nvm_is_otp) {
-			IWL_ERR(trans, "OTP with bad signature: 0x%08x\n", gp);
-			return -ENOENT;
-		}
-		return 0;
-	case CSR_EEPROM_GP_BAD_SIGNATURE_BOTH_EEP_AND_OTP:
-	default:
-		IWL_ERR(trans,
-			"bad EEPROM/OTP signature, type=%s, EEPROM_GP=0x%08x\n",
-			nvm_is_otp ? "OTP" : "EEPROM", gp);
-		return -ENOENT;
-	}
-}
-
-/******************************************************************************
- *
- * OTP related functions
- *
-******************************************************************************/
-
-static void iwl_set_otp_access_absolute(struct iwl_trans *trans)
-{
-	iwl_read32(trans, CSR_OTP_GP_REG);
-
-	iwl_clear_bit(trans, CSR_OTP_GP_REG,
-		      CSR_OTP_GP_REG_OTP_ACCESS_MODE);
-}
-
-static int iwl_nvm_is_otp(struct iwl_trans *trans)
-{
-	u32 otpgp;
-
-	/* OTP only valid for CP/PP and after */
-	switch (trans->hw_rev & CSR_HW_REV_TYPE_MSK) {
-	case CSR_HW_REV_TYPE_NONE:
-		IWL_ERR(trans, "Unknown hardware type\n");
-		return -EIO;
-	case CSR_HW_REV_TYPE_5300:
-	case CSR_HW_REV_TYPE_5350:
-	case CSR_HW_REV_TYPE_5100:
-	case CSR_HW_REV_TYPE_5150:
-		return 0;
-	default:
-		otpgp = iwl_read32(trans, CSR_OTP_GP_REG);
-		if (otpgp & CSR_OTP_GP_REG_DEVICE_SELECT)
-			return 1;
-		return 0;
-	}
-}
-
-static int iwl_init_otp_access(struct iwl_trans *trans)
-{
-	int ret;
-
-	ret = iwl_finish_nic_init(trans);
-	if (ret)
-		return ret;
-
-	iwl_set_bits_prph(trans, APMG_PS_CTRL_REG,
-			  APMG_PS_CTRL_VAL_RESET_REQ);
-	udelay(5);
-	iwl_clear_bits_prph(trans, APMG_PS_CTRL_REG,
-			    APMG_PS_CTRL_VAL_RESET_REQ);
-
-	/*
-	 * CSR auto clock gate disable bit -
-	 * this is only applicable for HW with OTP shadow RAM
-	 */
-	if (trans->trans_cfg->base_params->shadow_ram_support)
-		iwl_set_bit(trans, CSR_DBG_LINK_PWR_MGMT_REG,
-			    CSR_RESET_LINK_PWR_MGMT_DISABLED);
-
-	return 0;
-}
-
-static int iwl_read_otp_word(struct iwl_trans *trans, u16 addr,
-			     __le16 *eeprom_data)
-{
-	int ret = 0;
-	u32 r;
-	u32 otpgp;
-
-	iwl_write32(trans, CSR_EEPROM_REG,
-		    CSR_EEPROM_REG_MSK_ADDR & (addr << 1));
-	ret = iwl_poll_bit(trans, CSR_EEPROM_REG,
-				 CSR_EEPROM_REG_READ_VALID_MSK,
-				 CSR_EEPROM_REG_READ_VALID_MSK,
-				 IWL_EEPROM_ACCESS_TIMEOUT);
-	if (ret < 0) {
-		IWL_ERR(trans, "Time out reading OTP[%d]\n", addr);
-		return ret;
-	}
-	r = iwl_read32(trans, CSR_EEPROM_REG);
-	/* check for ECC errors: */
-	otpgp = iwl_read32(trans, CSR_OTP_GP_REG);
-	if (otpgp & CSR_OTP_GP_REG_ECC_UNCORR_STATUS_MSK) {
-		/* stop in this case */
-		/* set the uncorrectable OTP ECC bit for acknowledgment */
-		iwl_set_bit(trans, CSR_OTP_GP_REG,
-			    CSR_OTP_GP_REG_ECC_UNCORR_STATUS_MSK);
-		IWL_ERR(trans, "Uncorrectable OTP ECC error, abort OTP read\n");
-		return -EINVAL;
-	}
-	if (otpgp & CSR_OTP_GP_REG_ECC_CORR_STATUS_MSK) {
-		/* continue in this case */
-		/* set the correctable OTP ECC bit for acknowledgment */
-		iwl_set_bit(trans, CSR_OTP_GP_REG,
-			    CSR_OTP_GP_REG_ECC_CORR_STATUS_MSK);
-		IWL_ERR(trans, "Correctable OTP ECC error, continue read\n");
-	}
-	*eeprom_data = cpu_to_le16(r >> 16);
-	return 0;
-}
-
-/*
- * iwl_is_otp_empty: check for empty OTP
- */
-static bool iwl_is_otp_empty(struct iwl_trans *trans)
-{
-	u16 next_link_addr = 0;
-	__le16 link_value;
-	bool is_empty = false;
-
-	/* locate the beginning of OTP link list */
-	if (!iwl_read_otp_word(trans, next_link_addr, &link_value)) {
-		if (!link_value) {
-			IWL_ERR(trans, "OTP is empty\n");
-			is_empty = true;
-		}
-	} else {
-		IWL_ERR(trans, "Unable to read first block of OTP list.\n");
-		is_empty = true;
-	}
-
-	return is_empty;
-}
-
-
-/*
- * iwl_find_otp_image: find EEPROM image in OTP
- *   finding the OTP block that contains the EEPROM image.
- *   the last valid block on the link list (the block _before_ the last block)
- *   is the block we should read and used to configure the device.
- *   If all the available OTP blocks are full, the last block will be the block
- *   we should read and used to configure the device.
- *   only perform this operation if shadow RAM is disabled
- */
-static int iwl_find_otp_image(struct iwl_trans *trans,
-					u16 *validblockaddr)
-{
-	u16 next_link_addr = 0, valid_addr;
-	__le16 link_value = 0;
-	int usedblocks = 0;
-
-	/* set addressing mode to absolute to traverse the link list */
-	iwl_set_otp_access_absolute(trans);
-
-	/* checking for empty OTP or error */
-	if (iwl_is_otp_empty(trans))
-		return -EINVAL;
-
-	/*
-	 * start traverse link list
-	 * until reach the max number of OTP blocks
-	 * different devices have different number of OTP blocks
-	 */
-	do {
-		/* save current valid block address
-		 * check for more block on the link list
-		 */
-		valid_addr = next_link_addr;
-		next_link_addr = le16_to_cpu(link_value) * sizeof(u16);
-		IWL_DEBUG_EEPROM(trans->dev, "OTP blocks %d addr 0x%x\n",
-				 usedblocks, next_link_addr);
-		if (iwl_read_otp_word(trans, next_link_addr, &link_value))
-			return -EINVAL;
-		if (!link_value) {
-			/*
-			 * reach the end of link list, return success and
-			 * set address point to the starting address
-			 * of the image
-			 */
-			*validblockaddr = valid_addr;
-			/* skip first 2 bytes (link list pointer) */
-			*validblockaddr += 2;
-			return 0;
-		}
-		/* more in the link list, continue */
-		usedblocks++;
-	} while (usedblocks <= trans->trans_cfg->base_params->max_ll_items);
-
-	/* OTP has no valid blocks */
-	IWL_DEBUG_EEPROM(trans->dev, "OTP has no valid blocks\n");
-	return -EINVAL;
-}
-
-/*
- * iwl_read_eeprom - read EEPROM contents
- *
- * Load the EEPROM contents from adapter and return it
- * and its size.
- *
- * NOTE:  This routine uses the non-debug IO access functions.
- */
-int iwl_read_eeprom(struct iwl_trans *trans, u8 **eeprom, size_t *eeprom_size)
-{
-	__le16 *e;
-	u32 gp = iwl_read32(trans, CSR_EEPROM_GP);
-	int sz;
-	int ret;
-	u16 addr;
-	u16 validblockaddr = 0;
-	u16 cache_addr = 0;
-	int nvm_is_otp;
-
-	if (!eeprom || !eeprom_size)
-		return -EINVAL;
-
-	nvm_is_otp = iwl_nvm_is_otp(trans);
-	if (nvm_is_otp < 0)
-		return nvm_is_otp;
-
-	sz = trans->trans_cfg->base_params->eeprom_size;
-	IWL_DEBUG_EEPROM(trans->dev, "NVM size = %d\n", sz);
-
-	e = kmalloc(sz, GFP_KERNEL);
-	if (!e)
-		return -ENOMEM;
-
-	ret = iwl_eeprom_verify_signature(trans, nvm_is_otp);
-	if (ret < 0) {
-		IWL_ERR(trans, "EEPROM not found, EEPROM_GP=0x%08x\n", gp);
-		goto err_free;
-	}
-
-	/* Make sure driver (instead of uCode) is allowed to read EEPROM */
-	ret = iwl_eeprom_acquire_semaphore(trans);
-	if (ret < 0) {
-		IWL_ERR(trans, "Failed to acquire EEPROM semaphore.\n");
-		goto err_free;
-	}
-
-	if (nvm_is_otp) {
-		ret = iwl_init_otp_access(trans);
-		if (ret) {
-			IWL_ERR(trans, "Failed to initialize OTP access.\n");
-			goto err_unlock;
-		}
-
-		iwl_write32(trans, CSR_EEPROM_GP,
-			    iwl_read32(trans, CSR_EEPROM_GP) &
-			    ~CSR_EEPROM_GP_IF_OWNER_MSK);
-
-		iwl_set_bit(trans, CSR_OTP_GP_REG,
-			    CSR_OTP_GP_REG_ECC_CORR_STATUS_MSK |
-			    CSR_OTP_GP_REG_ECC_UNCORR_STATUS_MSK);
-		/* traversing the linked list if no shadow ram supported */
-		if (!trans->trans_cfg->base_params->shadow_ram_support) {
-			ret = iwl_find_otp_image(trans, &validblockaddr);
-			if (ret)
-				goto err_unlock;
-		}
-		for (addr = validblockaddr; addr < validblockaddr + sz;
-		     addr += sizeof(u16)) {
-			__le16 eeprom_data;
-
-			ret = iwl_read_otp_word(trans, addr, &eeprom_data);
-			if (ret)
-				goto err_unlock;
-			e[cache_addr / 2] = eeprom_data;
-			cache_addr += sizeof(u16);
-		}
-	} else {
-		/* eeprom is an array of 16bit values */
-		for (addr = 0; addr < sz; addr += sizeof(u16)) {
-			u32 r;
-
-			iwl_write32(trans, CSR_EEPROM_REG,
-				    CSR_EEPROM_REG_MSK_ADDR & (addr << 1));
-
-			ret = iwl_poll_bit(trans, CSR_EEPROM_REG,
-					   CSR_EEPROM_REG_READ_VALID_MSK,
-					   CSR_EEPROM_REG_READ_VALID_MSK,
-					   IWL_EEPROM_ACCESS_TIMEOUT);
-			if (ret < 0) {
-				IWL_ERR(trans,
-					"Time out reading EEPROM[%d]\n", addr);
-				goto err_unlock;
-			}
-			r = iwl_read32(trans, CSR_EEPROM_REG);
-			e[addr / 2] = cpu_to_le16(r >> 16);
-		}
-	}
-
-	IWL_DEBUG_EEPROM(trans->dev, "NVM Type: %s\n",
-			 nvm_is_otp ? "OTP" : "EEPROM");
-
-	iwl_eeprom_release_semaphore(trans);
-
-	*eeprom_size = sz;
-	*eeprom = (u8 *)e;
-	return 0;
-
- err_unlock:
-	iwl_eeprom_release_semaphore(trans);
- err_free:
-	kfree(e);
-
-	return ret;
-}
-IWL_EXPORT_SYMBOL(iwl_read_eeprom);
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-read.h b/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-read.h
deleted file mode 100644
index 63b8e6c6659b..000000000000
--- a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-read.h
+++ /dev/null
@@ -1,12 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
-/*
- * Copyright (C) 2005-2014 Intel Corporation
- */
-#ifndef __iwl_eeprom_h__
-#define __iwl_eeprom_h__
-
-#include "iwl-trans.h"
-
-int iwl_read_eeprom(struct iwl_trans *trans, u8 **eeprom, size_t *eeprom_size);
-
-#endif  /* __iwl_eeprom_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
index ae8177222881..1e949b48188e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2012-2014, 2018-2019, 2021-2023 Intel Corporation
+ * Copyright (C) 2012-2014, 2018-2019, 2021-2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -10,7 +10,6 @@
 #include "iwl-csr.h"
 #include "mvm.h"
 #include "iwl-eeprom-parse.h"
-#include "iwl-eeprom-read.h"
 #include "iwl-nvm-parse.h"
 #include "iwl-prph.h"
 #include "fw/acpi.h"
-- 
cgit v1.2.3-58-ga151


From 6584b9d0aa3234bd8af419e46de1a9648cd34116 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Sun, 12 May 2024 15:25:03 +0300
Subject: wifi: iwlwifi: move code from iwl-eeprom-parse to dvm

Move code that is DVM only to dvm.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512152312.9a1b6ef116e0.I217a513f544d5288a7760d265f51419e81abfd9d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/Makefile        |   2 +-
 drivers/net/wireless/intel/iwlwifi/dvm/agn.h       |  17 +
 drivers/net/wireless/intel/iwlwifi/dvm/dev.h       |   2 +-
 drivers/net/wireless/intel/iwlwifi/dvm/devices.c   |   2 +-
 drivers/net/wireless/intel/iwlwifi/dvm/eeprom.c    | 757 ++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/dvm/main.c      |   2 +-
 drivers/net/wireless/intel/iwlwifi/fw/runtime.h    |   2 +-
 .../net/wireless/intel/iwlwifi/iwl-eeprom-parse.c  | 882 ---------------------
 .../net/wireless/intel/iwlwifi/iwl-eeprom-parse.h  |  88 --
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h |   2 +-
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-utils.c | 118 +++
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-utils.h |  71 ++
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/nvm.c       |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c       |   2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c        |   2 +-
 17 files changed, 974 insertions(+), 981 deletions(-)
 delete mode 100644 drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.c
 delete mode 100644 drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h
 create mode 100644 drivers/net/wireless/intel/iwlwifi/iwl-nvm-utils.c
 create mode 100644 drivers/net/wireless/intel/iwlwifi/iwl-nvm-utils.h

diff --git a/drivers/net/wireless/intel/iwlwifi/Makefile b/drivers/net/wireless/intel/iwlwifi/Makefile
index 2c2c18cef9ff..93cce88f2e27 100644
--- a/drivers/net/wireless/intel/iwlwifi/Makefile
+++ b/drivers/net/wireless/intel/iwlwifi/Makefile
@@ -4,7 +4,7 @@ obj-$(CONFIG_IWLWIFI)	+= iwlwifi.o
 iwlwifi-objs		+= iwl-io.o
 iwlwifi-objs		+= iwl-drv.o
 iwlwifi-objs		+= iwl-debug.o
-iwlwifi-objs		+= iwl-eeprom-parse.o
+iwlwifi-objs		+= iwl-nvm-utils.o
 iwlwifi-objs		+= iwl-phy-db.o iwl-nvm-parse.o
 iwlwifi-objs		+= pcie/drv.o pcie/rx.o pcie/tx.o pcie/trans.o
 iwlwifi-objs		+= pcie/ctxt-info.o pcie/ctxt-info-gen3.o
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/agn.h b/drivers/net/wireless/intel/iwlwifi/dvm/agn.h
index 47d21413ba59..a13add556a7b 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/agn.h
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/agn.h
@@ -385,6 +385,23 @@ static inline void iwl_dvm_set_pmi(struct iwl_priv *priv, bool state)
 	iwl_trans_set_pmi(priv->trans, state);
 }
 
+/**
+ * iwl_parse_eeprom_data - parse EEPROM data and return values
+ *
+ * @trans: ransport we're parsing for, for debug only
+ * @cfg: device configuration for parsing and overrides
+ * @eeprom: the EEPROM data
+ * @eeprom_size: length of the EEPROM data
+ *
+ * This function parses all EEPROM values we need and then
+ * returns a (newly allocated) struct containing all the
+ * relevant values for driver use. The struct must be freed
+ * later with iwl_free_nvm_data().
+ */
+struct iwl_nvm_data *
+iwl_parse_eeprom_data(struct iwl_trans *trans, const struct iwl_cfg *cfg,
+		      const u8 *eeprom, size_t eeprom_size);
+
 int iwl_read_eeprom(struct iwl_trans *trans, u8 **eeprom, size_t *eeprom_size);
 
 #ifdef CONFIG_IWLWIFI_DEBUGFS
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/dev.h b/drivers/net/wireless/intel/iwlwifi/dvm/dev.h
index 25283e4b849f..4ac8b862ad41 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/dev.h
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/dev.h
@@ -19,7 +19,7 @@
 #include <linux/mutex.h>
 
 #include "fw/img.h"
-#include "iwl-eeprom-parse.h"
+#include "iwl-nvm-utils.h"
 #include "iwl-csr.h"
 #include "iwl-debug.h"
 #include "iwl-agn-hw.h"
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/devices.c b/drivers/net/wireless/intel/iwlwifi/dvm/devices.c
index 39e40901fa46..48a8349680fc 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/devices.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/devices.c
@@ -12,7 +12,7 @@
  */
 #include "iwl-io.h"
 #include "iwl-prph.h"
-#include "iwl-eeprom-parse.h"
+#include "iwl-nvm-utils.h"
 
 #include "agn.h"
 #include "dev.h"
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/eeprom.c b/drivers/net/wireless/intel/iwlwifi/dvm/eeprom.c
index 3972775fc133..931aa3f5798d 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/eeprom.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/eeprom.c
@@ -13,6 +13,641 @@
 #include "iwl-csr.h"
 #include "agn.h"
 
+/* EEPROM offset definitions */
+
+/* indirect access definitions */
+#define ADDRESS_MSK                 0x0000FFFF
+#define INDIRECT_TYPE_MSK           0x000F0000
+#define INDIRECT_HOST               0x00010000
+#define INDIRECT_GENERAL            0x00020000
+#define INDIRECT_REGULATORY         0x00030000
+#define INDIRECT_CALIBRATION        0x00040000
+#define INDIRECT_PROCESS_ADJST      0x00050000
+#define INDIRECT_OTHERS             0x00060000
+#define INDIRECT_TXP_LIMIT          0x00070000
+#define INDIRECT_TXP_LIMIT_SIZE     0x00080000
+#define INDIRECT_ADDRESS            0x00100000
+
+/* corresponding link offsets in EEPROM */
+#define EEPROM_LINK_HOST             (2*0x64)
+#define EEPROM_LINK_GENERAL          (2*0x65)
+#define EEPROM_LINK_REGULATORY       (2*0x66)
+#define EEPROM_LINK_CALIBRATION      (2*0x67)
+#define EEPROM_LINK_PROCESS_ADJST    (2*0x68)
+#define EEPROM_LINK_OTHERS           (2*0x69)
+#define EEPROM_LINK_TXP_LIMIT        (2*0x6a)
+#define EEPROM_LINK_TXP_LIMIT_SIZE   (2*0x6b)
+
+/* General */
+#define EEPROM_DEVICE_ID                    (2*0x08)	/* 2 bytes */
+#define EEPROM_SUBSYSTEM_ID		    (2*0x0A)	/* 2 bytes */
+#define EEPROM_MAC_ADDRESS                  (2*0x15)	/* 6  bytes */
+#define EEPROM_BOARD_REVISION               (2*0x35)	/* 2  bytes */
+#define EEPROM_BOARD_PBA_NUMBER             (2*0x3B+1)	/* 9  bytes */
+#define EEPROM_VERSION                      (2*0x44)	/* 2  bytes */
+#define EEPROM_SKU_CAP                      (2*0x45)	/* 2  bytes */
+#define EEPROM_OEM_MODE                     (2*0x46)	/* 2  bytes */
+#define EEPROM_RADIO_CONFIG                 (2*0x48)	/* 2  bytes */
+#define EEPROM_NUM_MAC_ADDRESS              (2*0x4C)	/* 2  bytes */
+
+/* calibration */
+struct iwl_eeprom_calib_hdr {
+	u8 version;
+	u8 pa_type;
+	__le16 voltage;
+} __packed;
+
+#define EEPROM_CALIB_ALL	(INDIRECT_ADDRESS | INDIRECT_CALIBRATION)
+#define EEPROM_XTAL		((2*0x128) | EEPROM_CALIB_ALL)
+
+/* temperature */
+#define EEPROM_KELVIN_TEMPERATURE	((2*0x12A) | EEPROM_CALIB_ALL)
+#define EEPROM_RAW_TEMPERATURE		((2*0x12B) | EEPROM_CALIB_ALL)
+
+/* SKU Capabilities (actual values from EEPROM definition) */
+enum eeprom_sku_bits {
+	EEPROM_SKU_CAP_BAND_24GHZ	= BIT(4),
+	EEPROM_SKU_CAP_BAND_52GHZ	= BIT(5),
+	EEPROM_SKU_CAP_11N_ENABLE	= BIT(6),
+	EEPROM_SKU_CAP_AMT_ENABLE	= BIT(7),
+	EEPROM_SKU_CAP_IPAN_ENABLE	= BIT(8)
+};
+
+/* radio config bits (actual values from EEPROM definition) */
+#define EEPROM_RF_CFG_TYPE_MSK(x)   (x & 0x3)         /* bits 0-1   */
+#define EEPROM_RF_CFG_STEP_MSK(x)   ((x >> 2)  & 0x3) /* bits 2-3   */
+#define EEPROM_RF_CFG_DASH_MSK(x)   ((x >> 4)  & 0x3) /* bits 4-5   */
+#define EEPROM_RF_CFG_PNUM_MSK(x)   ((x >> 6)  & 0x3) /* bits 6-7   */
+#define EEPROM_RF_CFG_TX_ANT_MSK(x) ((x >> 8)  & 0xF) /* bits 8-11  */
+#define EEPROM_RF_CFG_RX_ANT_MSK(x) ((x >> 12) & 0xF) /* bits 12-15 */
+
+/*
+ * EEPROM bands
+ * These are the channel numbers from each band in the order
+ * that they are stored in the EEPROM band information. Note
+ * that EEPROM bands aren't the same as mac80211 bands, and
+ * there are even special "ht40 bands" in the EEPROM.
+ */
+static const u8 iwl_eeprom_band_1[14] = { /* 2.4 GHz */
+	1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14
+};
+
+static const u8 iwl_eeprom_band_2[] = {	/* 4915-5080MHz */
+	183, 184, 185, 187, 188, 189, 192, 196, 7, 8, 11, 12, 16
+};
+
+static const u8 iwl_eeprom_band_3[] = {	/* 5170-5320MHz */
+	34, 36, 38, 40, 42, 44, 46, 48, 52, 56, 60, 64
+};
+
+static const u8 iwl_eeprom_band_4[] = {	/* 5500-5700MHz */
+	100, 104, 108, 112, 116, 120, 124, 128, 132, 136, 140
+};
+
+static const u8 iwl_eeprom_band_5[] = {	/* 5725-5825MHz */
+	145, 149, 153, 157, 161, 165
+};
+
+static const u8 iwl_eeprom_band_6[] = {	/* 2.4 ht40 channel */
+	1, 2, 3, 4, 5, 6, 7
+};
+
+static const u8 iwl_eeprom_band_7[] = {	/* 5.2 ht40 channel */
+	36, 44, 52, 60, 100, 108, 116, 124, 132, 149, 157
+};
+
+#define IWL_NUM_CHANNELS	(ARRAY_SIZE(iwl_eeprom_band_1) + \
+				 ARRAY_SIZE(iwl_eeprom_band_2) + \
+				 ARRAY_SIZE(iwl_eeprom_band_3) + \
+				 ARRAY_SIZE(iwl_eeprom_band_4) + \
+				 ARRAY_SIZE(iwl_eeprom_band_5))
+
+/* rate data (static) */
+static struct ieee80211_rate iwl_cfg80211_rates[] = {
+	{ .bitrate = 1 * 10, .hw_value = 0, .hw_value_short = 0, },
+	{ .bitrate = 2 * 10, .hw_value = 1, .hw_value_short = 1,
+	  .flags = IEEE80211_RATE_SHORT_PREAMBLE, },
+	{ .bitrate = 5.5 * 10, .hw_value = 2, .hw_value_short = 2,
+	  .flags = IEEE80211_RATE_SHORT_PREAMBLE, },
+	{ .bitrate = 11 * 10, .hw_value = 3, .hw_value_short = 3,
+	  .flags = IEEE80211_RATE_SHORT_PREAMBLE, },
+	{ .bitrate = 6 * 10, .hw_value = 4, .hw_value_short = 4, },
+	{ .bitrate = 9 * 10, .hw_value = 5, .hw_value_short = 5, },
+	{ .bitrate = 12 * 10, .hw_value = 6, .hw_value_short = 6, },
+	{ .bitrate = 18 * 10, .hw_value = 7, .hw_value_short = 7, },
+	{ .bitrate = 24 * 10, .hw_value = 8, .hw_value_short = 8, },
+	{ .bitrate = 36 * 10, .hw_value = 9, .hw_value_short = 9, },
+	{ .bitrate = 48 * 10, .hw_value = 10, .hw_value_short = 10, },
+	{ .bitrate = 54 * 10, .hw_value = 11, .hw_value_short = 11, },
+};
+#define RATES_24_OFFS	0
+#define N_RATES_24	ARRAY_SIZE(iwl_cfg80211_rates)
+#define RATES_52_OFFS	4
+#define N_RATES_52	(N_RATES_24 - RATES_52_OFFS)
+
+/* EEPROM reading functions */
+
+static u16 iwl_eeprom_query16(const u8 *eeprom, size_t eeprom_size, int offset)
+{
+	if (WARN_ON(offset + sizeof(u16) > eeprom_size))
+		return 0;
+	return le16_to_cpup((__le16 *)(eeprom + offset));
+}
+
+static u32 eeprom_indirect_address(const u8 *eeprom, size_t eeprom_size,
+				   u32 address)
+{
+	u16 offset = 0;
+
+	if ((address & INDIRECT_ADDRESS) == 0)
+		return address;
+
+	switch (address & INDIRECT_TYPE_MSK) {
+	case INDIRECT_HOST:
+		offset = iwl_eeprom_query16(eeprom, eeprom_size,
+					    EEPROM_LINK_HOST);
+		break;
+	case INDIRECT_GENERAL:
+		offset = iwl_eeprom_query16(eeprom, eeprom_size,
+					    EEPROM_LINK_GENERAL);
+		break;
+	case INDIRECT_REGULATORY:
+		offset = iwl_eeprom_query16(eeprom, eeprom_size,
+					    EEPROM_LINK_REGULATORY);
+		break;
+	case INDIRECT_TXP_LIMIT:
+		offset = iwl_eeprom_query16(eeprom, eeprom_size,
+					    EEPROM_LINK_TXP_LIMIT);
+		break;
+	case INDIRECT_TXP_LIMIT_SIZE:
+		offset = iwl_eeprom_query16(eeprom, eeprom_size,
+					    EEPROM_LINK_TXP_LIMIT_SIZE);
+		break;
+	case INDIRECT_CALIBRATION:
+		offset = iwl_eeprom_query16(eeprom, eeprom_size,
+					    EEPROM_LINK_CALIBRATION);
+		break;
+	case INDIRECT_PROCESS_ADJST:
+		offset = iwl_eeprom_query16(eeprom, eeprom_size,
+					    EEPROM_LINK_PROCESS_ADJST);
+		break;
+	case INDIRECT_OTHERS:
+		offset = iwl_eeprom_query16(eeprom, eeprom_size,
+					    EEPROM_LINK_OTHERS);
+		break;
+	default:
+		WARN_ON(1);
+		break;
+	}
+
+	/* translate the offset from words to byte */
+	return (address & ADDRESS_MSK) + (offset << 1);
+}
+
+static const u8 *iwl_eeprom_query_addr(const u8 *eeprom, size_t eeprom_size,
+				       u32 offset)
+{
+	u32 address = eeprom_indirect_address(eeprom, eeprom_size, offset);
+
+	if (WARN_ON(address >= eeprom_size))
+		return NULL;
+
+	return &eeprom[address];
+}
+
+static int iwl_eeprom_read_calib(const u8 *eeprom, size_t eeprom_size,
+				 struct iwl_nvm_data *data)
+{
+	struct iwl_eeprom_calib_hdr *hdr;
+
+	hdr = (void *)iwl_eeprom_query_addr(eeprom, eeprom_size,
+					    EEPROM_CALIB_ALL);
+	if (!hdr)
+		return -ENODATA;
+	data->calib_version = hdr->version;
+	data->calib_voltage = hdr->voltage;
+
+	return 0;
+}
+
+/**
+ * enum iwl_eeprom_channel_flags - channel flags in EEPROM
+ * @EEPROM_CHANNEL_VALID: channel is usable for this SKU/geo
+ * @EEPROM_CHANNEL_IBSS: usable as an IBSS channel
+ * @EEPROM_CHANNEL_ACTIVE: active scanning allowed
+ * @EEPROM_CHANNEL_RADAR: radar detection required
+ * @EEPROM_CHANNEL_WIDE: 20 MHz channel okay (?)
+ * @EEPROM_CHANNEL_DFS: dynamic freq selection candidate
+ */
+enum iwl_eeprom_channel_flags {
+	EEPROM_CHANNEL_VALID = BIT(0),
+	EEPROM_CHANNEL_IBSS = BIT(1),
+	EEPROM_CHANNEL_ACTIVE = BIT(3),
+	EEPROM_CHANNEL_RADAR = BIT(4),
+	EEPROM_CHANNEL_WIDE = BIT(5),
+	EEPROM_CHANNEL_DFS = BIT(7),
+};
+
+/**
+ * struct iwl_eeprom_channel - EEPROM channel data
+ * @flags: %EEPROM_CHANNEL_* flags
+ * @max_power_avg: max power (in dBm) on this channel, at most 31 dBm
+ */
+struct iwl_eeprom_channel {
+	u8 flags;
+	s8 max_power_avg;
+} __packed;
+
+enum iwl_eeprom_enhanced_txpwr_flags {
+	IWL_EEPROM_ENH_TXP_FL_VALID = BIT(0),
+	IWL_EEPROM_ENH_TXP_FL_BAND_52G = BIT(1),
+	IWL_EEPROM_ENH_TXP_FL_OFDM = BIT(2),
+	IWL_EEPROM_ENH_TXP_FL_40MHZ = BIT(3),
+	IWL_EEPROM_ENH_TXP_FL_HT_AP = BIT(4),
+	IWL_EEPROM_ENH_TXP_FL_RES1 = BIT(5),
+	IWL_EEPROM_ENH_TXP_FL_RES2 = BIT(6),
+	IWL_EEPROM_ENH_TXP_FL_COMMON_TYPE = BIT(7),
+};
+
+/**
+ * struct iwl_eeprom_enhanced_txpwr - enhanced regulatory TX power limits
+ * @flags: entry flags
+ * @channel: channel number
+ * @chain_a_max: chain a max power in 1/2 dBm
+ * @chain_b_max: chain b max power in 1/2 dBm
+ * @chain_c_max: chain c max power in 1/2 dBm
+ * @delta_20_in_40: 20-in-40 deltas (hi/lo)
+ * @mimo2_max: mimo2 max power in 1/2 dBm
+ * @mimo3_max: mimo3 max power in 1/2 dBm
+ *
+ * This structure presents the enhanced regulatory tx power limit layout
+ * in an EEPROM image.
+ */
+struct iwl_eeprom_enhanced_txpwr {
+	u8 flags;
+	u8 channel;
+	s8 chain_a_max;
+	s8 chain_b_max;
+	s8 chain_c_max;
+	u8 delta_20_in_40;
+	s8 mimo2_max;
+	s8 mimo3_max;
+} __packed;
+
+static s8 iwl_get_max_txpwr_half_dbm(const struct iwl_nvm_data *data,
+				     struct iwl_eeprom_enhanced_txpwr *txp)
+{
+	s8 result = 0; /* (.5 dBm) */
+
+	/* Take the highest tx power from any valid chains */
+	if (data->valid_tx_ant & ANT_A && txp->chain_a_max > result)
+		result = txp->chain_a_max;
+
+	if (data->valid_tx_ant & ANT_B && txp->chain_b_max > result)
+		result = txp->chain_b_max;
+
+	if (data->valid_tx_ant & ANT_C && txp->chain_c_max > result)
+		result = txp->chain_c_max;
+
+	if ((data->valid_tx_ant == ANT_AB ||
+	     data->valid_tx_ant == ANT_BC ||
+	     data->valid_tx_ant == ANT_AC) && txp->mimo2_max > result)
+		result = txp->mimo2_max;
+
+	if (data->valid_tx_ant == ANT_ABC && txp->mimo3_max > result)
+		result = txp->mimo3_max;
+
+	return result;
+}
+
+#define EEPROM_TXP_OFFS	(0x00 | INDIRECT_ADDRESS | INDIRECT_TXP_LIMIT)
+#define EEPROM_TXP_ENTRY_LEN sizeof(struct iwl_eeprom_enhanced_txpwr)
+#define EEPROM_TXP_SZ_OFFS (0x00 | INDIRECT_ADDRESS | INDIRECT_TXP_LIMIT_SIZE)
+
+#define TXP_CHECK_AND_PRINT(x) \
+	((txp->flags & IWL_EEPROM_ENH_TXP_FL_##x) ? # x " " : "")
+
+static void
+iwl_eeprom_enh_txp_read_element(struct iwl_nvm_data *data,
+				struct iwl_eeprom_enhanced_txpwr *txp,
+				int n_channels, s8 max_txpower_avg)
+{
+	int ch_idx;
+	enum nl80211_band band;
+
+	band = txp->flags & IWL_EEPROM_ENH_TXP_FL_BAND_52G ?
+		NL80211_BAND_5GHZ : NL80211_BAND_2GHZ;
+
+	for (ch_idx = 0; ch_idx < n_channels; ch_idx++) {
+		struct ieee80211_channel *chan = &data->channels[ch_idx];
+
+		/* update matching channel or from common data only */
+		if (txp->channel != 0 && chan->hw_value != txp->channel)
+			continue;
+
+		/* update matching band only */
+		if (band != chan->band)
+			continue;
+
+		if (chan->max_power < max_txpower_avg &&
+		    !(txp->flags & IWL_EEPROM_ENH_TXP_FL_40MHZ))
+			chan->max_power = max_txpower_avg;
+	}
+}
+
+static void iwl_eeprom_enhanced_txpower(struct device *dev,
+					struct iwl_nvm_data *data,
+					const u8 *eeprom, size_t eeprom_size,
+					int n_channels)
+{
+	struct iwl_eeprom_enhanced_txpwr *txp_array, *txp;
+	int idx, entries;
+	__le16 *txp_len;
+	s8 max_txp_avg_halfdbm;
+
+	BUILD_BUG_ON(sizeof(struct iwl_eeprom_enhanced_txpwr) != 8);
+
+	/* the length is in 16-bit words, but we want entries */
+	txp_len = (__le16 *)iwl_eeprom_query_addr(eeprom, eeprom_size,
+						  EEPROM_TXP_SZ_OFFS);
+	entries = le16_to_cpup(txp_len) * 2 / EEPROM_TXP_ENTRY_LEN;
+
+	txp_array = (void *)iwl_eeprom_query_addr(eeprom, eeprom_size,
+						  EEPROM_TXP_OFFS);
+
+	for (idx = 0; idx < entries; idx++) {
+		txp = &txp_array[idx];
+		/* skip invalid entries */
+		if (!(txp->flags & IWL_EEPROM_ENH_TXP_FL_VALID))
+			continue;
+
+		IWL_DEBUG_EEPROM(dev, "%s %d:\t %s%s%s%s%s%s%s%s (0x%02x)\n",
+				 (txp->channel && (txp->flags &
+					IWL_EEPROM_ENH_TXP_FL_COMMON_TYPE)) ?
+					"Common " : (txp->channel) ?
+					"Channel" : "Common",
+				 (txp->channel),
+				 TXP_CHECK_AND_PRINT(VALID),
+				 TXP_CHECK_AND_PRINT(BAND_52G),
+				 TXP_CHECK_AND_PRINT(OFDM),
+				 TXP_CHECK_AND_PRINT(40MHZ),
+				 TXP_CHECK_AND_PRINT(HT_AP),
+				 TXP_CHECK_AND_PRINT(RES1),
+				 TXP_CHECK_AND_PRINT(RES2),
+				 TXP_CHECK_AND_PRINT(COMMON_TYPE),
+				 txp->flags);
+		IWL_DEBUG_EEPROM(dev,
+				 "\t\t chain_A: %d chain_B: %d chain_C: %d\n",
+				 txp->chain_a_max, txp->chain_b_max,
+				 txp->chain_c_max);
+		IWL_DEBUG_EEPROM(dev,
+				 "\t\t MIMO2: %d MIMO3: %d High 20_on_40: 0x%02x Low 20_on_40: 0x%02x\n",
+				 txp->mimo2_max, txp->mimo3_max,
+				 ((txp->delta_20_in_40 & 0xf0) >> 4),
+				 (txp->delta_20_in_40 & 0x0f));
+
+		max_txp_avg_halfdbm = iwl_get_max_txpwr_half_dbm(data, txp);
+
+		iwl_eeprom_enh_txp_read_element(data, txp, n_channels,
+				DIV_ROUND_UP(max_txp_avg_halfdbm, 2));
+
+		if (max_txp_avg_halfdbm > data->max_tx_pwr_half_dbm)
+			data->max_tx_pwr_half_dbm = max_txp_avg_halfdbm;
+	}
+}
+
+static void iwl_init_band_reference(const struct iwl_cfg *cfg,
+				    const u8 *eeprom, size_t eeprom_size,
+				    int eeprom_band, int *eeprom_ch_count,
+				    const struct iwl_eeprom_channel **ch_info,
+				    const u8 **eeprom_ch_array)
+{
+	u32 offset = cfg->eeprom_params->regulatory_bands[eeprom_band - 1];
+
+	offset |= INDIRECT_ADDRESS | INDIRECT_REGULATORY;
+
+	*ch_info = (void *)iwl_eeprom_query_addr(eeprom, eeprom_size, offset);
+
+	switch (eeprom_band) {
+	case 1:		/* 2.4GHz band */
+		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_1);
+		*eeprom_ch_array = iwl_eeprom_band_1;
+		break;
+	case 2:		/* 4.9GHz band */
+		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_2);
+		*eeprom_ch_array = iwl_eeprom_band_2;
+		break;
+	case 3:		/* 5.2GHz band */
+		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_3);
+		*eeprom_ch_array = iwl_eeprom_band_3;
+		break;
+	case 4:		/* 5.5GHz band */
+		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_4);
+		*eeprom_ch_array = iwl_eeprom_band_4;
+		break;
+	case 5:		/* 5.7GHz band */
+		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_5);
+		*eeprom_ch_array = iwl_eeprom_band_5;
+		break;
+	case 6:		/* 2.4GHz ht40 channels */
+		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_6);
+		*eeprom_ch_array = iwl_eeprom_band_6;
+		break;
+	case 7:		/* 5 GHz ht40 channels */
+		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_7);
+		*eeprom_ch_array = iwl_eeprom_band_7;
+		break;
+	default:
+		*eeprom_ch_count = 0;
+		*eeprom_ch_array = NULL;
+		WARN_ON(1);
+	}
+}
+
+#define CHECK_AND_PRINT(x) \
+	((eeprom_ch->flags & EEPROM_CHANNEL_##x) ? # x " " : "")
+
+static void iwl_mod_ht40_chan_info(struct device *dev,
+				   struct iwl_nvm_data *data, int n_channels,
+				   enum nl80211_band band, u16 channel,
+				   const struct iwl_eeprom_channel *eeprom_ch,
+				   u8 clear_ht40_extension_channel)
+{
+	struct ieee80211_channel *chan = NULL;
+	int i;
+
+	for (i = 0; i < n_channels; i++) {
+		if (data->channels[i].band != band)
+			continue;
+		if (data->channels[i].hw_value != channel)
+			continue;
+		chan = &data->channels[i];
+		break;
+	}
+
+	if (!chan)
+		return;
+
+	IWL_DEBUG_EEPROM(dev,
+			 "HT40 Ch. %d [%sGHz] %s%s%s%s%s(0x%02x %ddBm): Ad-Hoc %ssupported\n",
+			 channel,
+			 band == NL80211_BAND_5GHZ ? "5.2" : "2.4",
+			 CHECK_AND_PRINT(IBSS),
+			 CHECK_AND_PRINT(ACTIVE),
+			 CHECK_AND_PRINT(RADAR),
+			 CHECK_AND_PRINT(WIDE),
+			 CHECK_AND_PRINT(DFS),
+			 eeprom_ch->flags,
+			 eeprom_ch->max_power_avg,
+			 ((eeprom_ch->flags & EEPROM_CHANNEL_IBSS) &&
+			  !(eeprom_ch->flags & EEPROM_CHANNEL_RADAR)) ? ""
+								      : "not ");
+
+	if (eeprom_ch->flags & EEPROM_CHANNEL_VALID)
+		chan->flags &= ~clear_ht40_extension_channel;
+}
+
+#define CHECK_AND_PRINT_I(x)	\
+	((eeprom_ch_info[ch_idx].flags & EEPROM_CHANNEL_##x) ? # x " " : "")
+
+static int iwl_init_channel_map(struct device *dev, const struct iwl_cfg *cfg,
+				struct iwl_nvm_data *data,
+				const u8 *eeprom, size_t eeprom_size)
+{
+	int band, ch_idx;
+	const struct iwl_eeprom_channel *eeprom_ch_info;
+	const u8 *eeprom_ch_array;
+	int eeprom_ch_count;
+	int n_channels = 0;
+
+	/*
+	 * Loop through the 5 EEPROM bands and add them to the parse list
+	 */
+	for (band = 1; band <= 5; band++) {
+		struct ieee80211_channel *channel;
+
+		iwl_init_band_reference(cfg, eeprom, eeprom_size, band,
+					&eeprom_ch_count, &eeprom_ch_info,
+					&eeprom_ch_array);
+
+		/* Loop through each band adding each of the channels */
+		for (ch_idx = 0; ch_idx < eeprom_ch_count; ch_idx++) {
+			const struct iwl_eeprom_channel *eeprom_ch;
+
+			eeprom_ch = &eeprom_ch_info[ch_idx];
+
+			if (!(eeprom_ch->flags & EEPROM_CHANNEL_VALID)) {
+				IWL_DEBUG_EEPROM(dev,
+						 "Ch. %d Flags %x [%sGHz] - No traffic\n",
+						 eeprom_ch_array[ch_idx],
+						 eeprom_ch_info[ch_idx].flags,
+						 (band != 1) ? "5.2" : "2.4");
+				continue;
+			}
+
+			channel = &data->channels[n_channels];
+			n_channels++;
+
+			channel->hw_value = eeprom_ch_array[ch_idx];
+			channel->band = (band == 1) ? NL80211_BAND_2GHZ
+						    : NL80211_BAND_5GHZ;
+			channel->center_freq =
+				ieee80211_channel_to_frequency(
+					channel->hw_value, channel->band);
+
+			/* set no-HT40, will enable as appropriate later */
+			channel->flags = IEEE80211_CHAN_NO_HT40;
+
+			if (!(eeprom_ch->flags & EEPROM_CHANNEL_IBSS))
+				channel->flags |= IEEE80211_CHAN_NO_IR;
+
+			if (!(eeprom_ch->flags & EEPROM_CHANNEL_ACTIVE))
+				channel->flags |= IEEE80211_CHAN_NO_IR;
+
+			if (eeprom_ch->flags & EEPROM_CHANNEL_RADAR)
+				channel->flags |= IEEE80211_CHAN_RADAR;
+
+			/* Initialize regulatory-based run-time data */
+			channel->max_power =
+				eeprom_ch_info[ch_idx].max_power_avg;
+			IWL_DEBUG_EEPROM(dev,
+					 "Ch. %d [%sGHz] %s%s%s%s%s%s(0x%02x %ddBm): Ad-Hoc %ssupported\n",
+					 channel->hw_value,
+					 (band != 1) ? "5.2" : "2.4",
+					 CHECK_AND_PRINT_I(VALID),
+					 CHECK_AND_PRINT_I(IBSS),
+					 CHECK_AND_PRINT_I(ACTIVE),
+					 CHECK_AND_PRINT_I(RADAR),
+					 CHECK_AND_PRINT_I(WIDE),
+					 CHECK_AND_PRINT_I(DFS),
+					 eeprom_ch_info[ch_idx].flags,
+					 eeprom_ch_info[ch_idx].max_power_avg,
+					 ((eeprom_ch_info[ch_idx].flags &
+							EEPROM_CHANNEL_IBSS) &&
+					  !(eeprom_ch_info[ch_idx].flags &
+							EEPROM_CHANNEL_RADAR))
+						? "" : "not ");
+		}
+	}
+
+	if (cfg->eeprom_params->enhanced_txpower) {
+		/*
+		 * for newer device (6000 series and up)
+		 * EEPROM contain enhanced tx power information
+		 * driver need to process addition information
+		 * to determine the max channel tx power limits
+		 */
+		iwl_eeprom_enhanced_txpower(dev, data, eeprom, eeprom_size,
+					    n_channels);
+	} else {
+		/* All others use data from channel map */
+		int i;
+
+		data->max_tx_pwr_half_dbm = -128;
+
+		for (i = 0; i < n_channels; i++)
+			data->max_tx_pwr_half_dbm =
+				max_t(s8, data->max_tx_pwr_half_dbm,
+				      data->channels[i].max_power * 2);
+	}
+
+	/* Check if we do have HT40 channels */
+	if (cfg->eeprom_params->regulatory_bands[5] ==
+				EEPROM_REGULATORY_BAND_NO_HT40 &&
+	    cfg->eeprom_params->regulatory_bands[6] ==
+				EEPROM_REGULATORY_BAND_NO_HT40)
+		return n_channels;
+
+	/* Two additional EEPROM bands for 2.4 and 5 GHz HT40 channels */
+	for (band = 6; band <= 7; band++) {
+		enum nl80211_band ieeeband;
+
+		iwl_init_band_reference(cfg, eeprom, eeprom_size, band,
+					&eeprom_ch_count, &eeprom_ch_info,
+					&eeprom_ch_array);
+
+		/* EEPROM band 6 is 2.4, band 7 is 5 GHz */
+		ieeeband = (band == 6) ? NL80211_BAND_2GHZ
+				       : NL80211_BAND_5GHZ;
+
+		/* Loop through each band adding each of the channels */
+		for (ch_idx = 0; ch_idx < eeprom_ch_count; ch_idx++) {
+			/* Set up driver's info for lower half */
+			iwl_mod_ht40_chan_info(dev, data, n_channels, ieeeband,
+					       eeprom_ch_array[ch_idx],
+					       &eeprom_ch_info[ch_idx],
+					       IEEE80211_CHAN_NO_HT40PLUS);
+
+			/* Set up driver's info for upper half */
+			iwl_mod_ht40_chan_info(dev, data, n_channels, ieeeband,
+					       eeprom_ch_array[ch_idx] + 4,
+					       &eeprom_ch_info[ch_idx],
+					       IEEE80211_CHAN_NO_HT40MINUS);
+		}
+	}
+
+	return n_channels;
+}
 /*
  * EEPROM access time values:
  *
@@ -391,3 +1026,125 @@ int iwl_read_eeprom(struct iwl_trans *trans, u8 **eeprom, size_t *eeprom_size)
 
 	return ret;
 }
+
+static void iwl_init_sbands(struct iwl_trans *trans, const struct iwl_cfg *cfg,
+			    struct iwl_nvm_data *data,
+			    const u8 *eeprom, size_t eeprom_size)
+{
+	struct device *dev = trans->dev;
+	int n_channels = iwl_init_channel_map(dev, cfg, data,
+					      eeprom, eeprom_size);
+	int n_used = 0;
+	struct ieee80211_supported_band *sband;
+
+	sband = &data->bands[NL80211_BAND_2GHZ];
+	sband->band = NL80211_BAND_2GHZ;
+	sband->bitrates = &iwl_cfg80211_rates[RATES_24_OFFS];
+	sband->n_bitrates = N_RATES_24;
+	n_used += iwl_init_sband_channels(data, sband, n_channels,
+					  NL80211_BAND_2GHZ);
+	iwl_init_ht_hw_capab(trans, data, &sband->ht_cap, NL80211_BAND_2GHZ,
+			     data->valid_tx_ant, data->valid_rx_ant);
+
+	sband = &data->bands[NL80211_BAND_5GHZ];
+	sband->band = NL80211_BAND_5GHZ;
+	sband->bitrates = &iwl_cfg80211_rates[RATES_52_OFFS];
+	sband->n_bitrates = N_RATES_52;
+	n_used += iwl_init_sband_channels(data, sband, n_channels,
+					  NL80211_BAND_5GHZ);
+	iwl_init_ht_hw_capab(trans, data, &sband->ht_cap, NL80211_BAND_5GHZ,
+			     data->valid_tx_ant, data->valid_rx_ant);
+
+	if (n_channels != n_used)
+		IWL_ERR_DEV(dev, "EEPROM: used only %d of %d channels\n",
+			    n_used, n_channels);
+}
+
+/* EEPROM data functions */
+struct iwl_nvm_data *
+iwl_parse_eeprom_data(struct iwl_trans *trans, const struct iwl_cfg *cfg,
+		      const u8 *eeprom, size_t eeprom_size)
+{
+	struct iwl_nvm_data *data;
+	struct device *dev = trans->dev;
+	const void *tmp;
+	u16 radio_cfg, sku;
+
+	if (WARN_ON(!cfg || !cfg->eeprom_params))
+		return NULL;
+
+	data = kzalloc(struct_size(data, channels, IWL_NUM_CHANNELS),
+		       GFP_KERNEL);
+	if (!data)
+		return NULL;
+
+	/* get MAC address(es) */
+	tmp = iwl_eeprom_query_addr(eeprom, eeprom_size, EEPROM_MAC_ADDRESS);
+	if (!tmp)
+		goto err_free;
+	memcpy(data->hw_addr, tmp, ETH_ALEN);
+	data->n_hw_addrs = iwl_eeprom_query16(eeprom, eeprom_size,
+					      EEPROM_NUM_MAC_ADDRESS);
+
+	if (iwl_eeprom_read_calib(eeprom, eeprom_size, data))
+		goto err_free;
+
+	tmp = iwl_eeprom_query_addr(eeprom, eeprom_size, EEPROM_XTAL);
+	if (!tmp)
+		goto err_free;
+	memcpy(data->xtal_calib, tmp, sizeof(data->xtal_calib));
+
+	tmp = iwl_eeprom_query_addr(eeprom, eeprom_size,
+				    EEPROM_RAW_TEMPERATURE);
+	if (!tmp)
+		goto err_free;
+	data->raw_temperature = *(__le16 *)tmp;
+
+	tmp = iwl_eeprom_query_addr(eeprom, eeprom_size,
+				    EEPROM_KELVIN_TEMPERATURE);
+	if (!tmp)
+		goto err_free;
+	data->kelvin_temperature = *(__le16 *)tmp;
+	data->kelvin_voltage = *((__le16 *)tmp + 1);
+
+	radio_cfg =
+		iwl_eeprom_query16(eeprom, eeprom_size, EEPROM_RADIO_CONFIG);
+	data->radio_cfg_dash = EEPROM_RF_CFG_DASH_MSK(radio_cfg);
+	data->radio_cfg_pnum = EEPROM_RF_CFG_PNUM_MSK(radio_cfg);
+	data->radio_cfg_step = EEPROM_RF_CFG_STEP_MSK(radio_cfg);
+	data->radio_cfg_type = EEPROM_RF_CFG_TYPE_MSK(radio_cfg);
+	data->valid_rx_ant = EEPROM_RF_CFG_RX_ANT_MSK(radio_cfg);
+	data->valid_tx_ant = EEPROM_RF_CFG_TX_ANT_MSK(radio_cfg);
+
+	sku = iwl_eeprom_query16(eeprom, eeprom_size,
+				 EEPROM_SKU_CAP);
+	data->sku_cap_11n_enable = sku & EEPROM_SKU_CAP_11N_ENABLE;
+	data->sku_cap_amt_enable = sku & EEPROM_SKU_CAP_AMT_ENABLE;
+	data->sku_cap_band_24ghz_enable = sku & EEPROM_SKU_CAP_BAND_24GHZ;
+	data->sku_cap_band_52ghz_enable = sku & EEPROM_SKU_CAP_BAND_52GHZ;
+	data->sku_cap_ipan_enable = sku & EEPROM_SKU_CAP_IPAN_ENABLE;
+	if (iwlwifi_mod_params.disable_11n & IWL_DISABLE_HT_ALL)
+		data->sku_cap_11n_enable = false;
+
+	data->nvm_version = iwl_eeprom_query16(eeprom, eeprom_size,
+					       EEPROM_VERSION);
+
+	/* check overrides (some devices have wrong EEPROM) */
+	if (cfg->valid_tx_ant)
+		data->valid_tx_ant = cfg->valid_tx_ant;
+	if (cfg->valid_rx_ant)
+		data->valid_rx_ant = cfg->valid_rx_ant;
+
+	if (!data->valid_tx_ant || !data->valid_rx_ant) {
+		IWL_ERR_DEV(dev, "invalid antennas (0x%x, 0x%x)\n",
+			    data->valid_tx_ant, data->valid_rx_ant);
+		goto err_free;
+	}
+
+	iwl_init_sbands(trans, cfg, data, eeprom, eeprom_size);
+
+	return data;
+ err_free:
+	kfree(data);
+	return NULL;
+}
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/main.c b/drivers/net/wireless/intel/iwlwifi/dvm/main.c
index 7bb3011e12e3..65b7c68e5ca7 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/main.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/main.c
@@ -26,7 +26,7 @@
 
 #include <asm/div64.h>
 
-#include "iwl-eeprom-parse.h"
+#include "iwl-nvm-utils.h"
 #include "iwl-io.h"
 #include "iwl-trans.h"
 #include "iwl-op-mode.h"
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
index d201440066ea..3fcc08b3c0db 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
@@ -12,7 +12,7 @@
 #include "fw/api/debug.h"
 #include "fw/api/paging.h"
 #include "fw/api/power.h"
-#include "iwl-eeprom-parse.h"
+#include "iwl-nvm-utils.h"
 #include "fw/acpi.h"
 #include "fw/regulatory.h"
 
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.c
deleted file mode 100644
index 2b290fab1ef2..000000000000
--- a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.c
+++ /dev/null
@@ -1,882 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
-/*
- * Copyright (C) 2005-2014, 2018-2021, 2023 Intel Corporation
- * Copyright (C) 2015 Intel Mobile Communications GmbH
- */
-#include <linux/types.h>
-#include <linux/slab.h>
-#include <linux/export.h>
-#include "iwl-drv.h"
-#include "iwl-modparams.h"
-#include "iwl-eeprom-parse.h"
-
-#if IS_ENABLED(CONFIG_IWLDVM)
-/* EEPROM offset definitions */
-
-/* indirect access definitions */
-#define ADDRESS_MSK                 0x0000FFFF
-#define INDIRECT_TYPE_MSK           0x000F0000
-#define INDIRECT_HOST               0x00010000
-#define INDIRECT_GENERAL            0x00020000
-#define INDIRECT_REGULATORY         0x00030000
-#define INDIRECT_CALIBRATION        0x00040000
-#define INDIRECT_PROCESS_ADJST      0x00050000
-#define INDIRECT_OTHERS             0x00060000
-#define INDIRECT_TXP_LIMIT          0x00070000
-#define INDIRECT_TXP_LIMIT_SIZE     0x00080000
-#define INDIRECT_ADDRESS            0x00100000
-
-/* corresponding link offsets in EEPROM */
-#define EEPROM_LINK_HOST             (2*0x64)
-#define EEPROM_LINK_GENERAL          (2*0x65)
-#define EEPROM_LINK_REGULATORY       (2*0x66)
-#define EEPROM_LINK_CALIBRATION      (2*0x67)
-#define EEPROM_LINK_PROCESS_ADJST    (2*0x68)
-#define EEPROM_LINK_OTHERS           (2*0x69)
-#define EEPROM_LINK_TXP_LIMIT        (2*0x6a)
-#define EEPROM_LINK_TXP_LIMIT_SIZE   (2*0x6b)
-
-/* General */
-#define EEPROM_DEVICE_ID                    (2*0x08)	/* 2 bytes */
-#define EEPROM_SUBSYSTEM_ID		    (2*0x0A)	/* 2 bytes */
-#define EEPROM_MAC_ADDRESS                  (2*0x15)	/* 6  bytes */
-#define EEPROM_BOARD_REVISION               (2*0x35)	/* 2  bytes */
-#define EEPROM_BOARD_PBA_NUMBER             (2*0x3B+1)	/* 9  bytes */
-#define EEPROM_VERSION                      (2*0x44)	/* 2  bytes */
-#define EEPROM_SKU_CAP                      (2*0x45)	/* 2  bytes */
-#define EEPROM_OEM_MODE                     (2*0x46)	/* 2  bytes */
-#define EEPROM_RADIO_CONFIG                 (2*0x48)	/* 2  bytes */
-#define EEPROM_NUM_MAC_ADDRESS              (2*0x4C)	/* 2  bytes */
-
-/* calibration */
-struct iwl_eeprom_calib_hdr {
-	u8 version;
-	u8 pa_type;
-	__le16 voltage;
-} __packed;
-
-#define EEPROM_CALIB_ALL	(INDIRECT_ADDRESS | INDIRECT_CALIBRATION)
-#define EEPROM_XTAL		((2*0x128) | EEPROM_CALIB_ALL)
-
-/* temperature */
-#define EEPROM_KELVIN_TEMPERATURE	((2*0x12A) | EEPROM_CALIB_ALL)
-#define EEPROM_RAW_TEMPERATURE		((2*0x12B) | EEPROM_CALIB_ALL)
-
-/* SKU Capabilities (actual values from EEPROM definition) */
-enum eeprom_sku_bits {
-	EEPROM_SKU_CAP_BAND_24GHZ	= BIT(4),
-	EEPROM_SKU_CAP_BAND_52GHZ	= BIT(5),
-	EEPROM_SKU_CAP_11N_ENABLE	= BIT(6),
-	EEPROM_SKU_CAP_AMT_ENABLE	= BIT(7),
-	EEPROM_SKU_CAP_IPAN_ENABLE	= BIT(8)
-};
-
-/* radio config bits (actual values from EEPROM definition) */
-#define EEPROM_RF_CFG_TYPE_MSK(x)   (x & 0x3)         /* bits 0-1   */
-#define EEPROM_RF_CFG_STEP_MSK(x)   ((x >> 2)  & 0x3) /* bits 2-3   */
-#define EEPROM_RF_CFG_DASH_MSK(x)   ((x >> 4)  & 0x3) /* bits 4-5   */
-#define EEPROM_RF_CFG_PNUM_MSK(x)   ((x >> 6)  & 0x3) /* bits 6-7   */
-#define EEPROM_RF_CFG_TX_ANT_MSK(x) ((x >> 8)  & 0xF) /* bits 8-11  */
-#define EEPROM_RF_CFG_RX_ANT_MSK(x) ((x >> 12) & 0xF) /* bits 12-15 */
-
-
-/*
- * EEPROM bands
- * These are the channel numbers from each band in the order
- * that they are stored in the EEPROM band information. Note
- * that EEPROM bands aren't the same as mac80211 bands, and
- * there are even special "ht40 bands" in the EEPROM.
- */
-static const u8 iwl_eeprom_band_1[14] = { /* 2.4 GHz */
-	1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14
-};
-
-static const u8 iwl_eeprom_band_2[] = {	/* 4915-5080MHz */
-	183, 184, 185, 187, 188, 189, 192, 196, 7, 8, 11, 12, 16
-};
-
-static const u8 iwl_eeprom_band_3[] = {	/* 5170-5320MHz */
-	34, 36, 38, 40, 42, 44, 46, 48, 52, 56, 60, 64
-};
-
-static const u8 iwl_eeprom_band_4[] = {	/* 5500-5700MHz */
-	100, 104, 108, 112, 116, 120, 124, 128, 132, 136, 140
-};
-
-static const u8 iwl_eeprom_band_5[] = {	/* 5725-5825MHz */
-	145, 149, 153, 157, 161, 165
-};
-
-static const u8 iwl_eeprom_band_6[] = {	/* 2.4 ht40 channel */
-	1, 2, 3, 4, 5, 6, 7
-};
-
-static const u8 iwl_eeprom_band_7[] = {	/* 5.2 ht40 channel */
-	36, 44, 52, 60, 100, 108, 116, 124, 132, 149, 157
-};
-
-#define IWL_NUM_CHANNELS	(ARRAY_SIZE(iwl_eeprom_band_1) + \
-				 ARRAY_SIZE(iwl_eeprom_band_2) + \
-				 ARRAY_SIZE(iwl_eeprom_band_3) + \
-				 ARRAY_SIZE(iwl_eeprom_band_4) + \
-				 ARRAY_SIZE(iwl_eeprom_band_5))
-
-/* rate data (static) */
-static struct ieee80211_rate iwl_cfg80211_rates[] = {
-	{ .bitrate = 1 * 10, .hw_value = 0, .hw_value_short = 0, },
-	{ .bitrate = 2 * 10, .hw_value = 1, .hw_value_short = 1,
-	  .flags = IEEE80211_RATE_SHORT_PREAMBLE, },
-	{ .bitrate = 5.5 * 10, .hw_value = 2, .hw_value_short = 2,
-	  .flags = IEEE80211_RATE_SHORT_PREAMBLE, },
-	{ .bitrate = 11 * 10, .hw_value = 3, .hw_value_short = 3,
-	  .flags = IEEE80211_RATE_SHORT_PREAMBLE, },
-	{ .bitrate = 6 * 10, .hw_value = 4, .hw_value_short = 4, },
-	{ .bitrate = 9 * 10, .hw_value = 5, .hw_value_short = 5, },
-	{ .bitrate = 12 * 10, .hw_value = 6, .hw_value_short = 6, },
-	{ .bitrate = 18 * 10, .hw_value = 7, .hw_value_short = 7, },
-	{ .bitrate = 24 * 10, .hw_value = 8, .hw_value_short = 8, },
-	{ .bitrate = 36 * 10, .hw_value = 9, .hw_value_short = 9, },
-	{ .bitrate = 48 * 10, .hw_value = 10, .hw_value_short = 10, },
-	{ .bitrate = 54 * 10, .hw_value = 11, .hw_value_short = 11, },
-};
-#define RATES_24_OFFS	0
-#define N_RATES_24	ARRAY_SIZE(iwl_cfg80211_rates)
-#define RATES_52_OFFS	4
-#define N_RATES_52	(N_RATES_24 - RATES_52_OFFS)
-
-/* EEPROM reading functions */
-
-static u16 iwl_eeprom_query16(const u8 *eeprom, size_t eeprom_size, int offset)
-{
-	if (WARN_ON(offset + sizeof(u16) > eeprom_size))
-		return 0;
-	return le16_to_cpup((__le16 *)(eeprom + offset));
-}
-
-static u32 eeprom_indirect_address(const u8 *eeprom, size_t eeprom_size,
-				   u32 address)
-{
-	u16 offset = 0;
-
-	if ((address & INDIRECT_ADDRESS) == 0)
-		return address;
-
-	switch (address & INDIRECT_TYPE_MSK) {
-	case INDIRECT_HOST:
-		offset = iwl_eeprom_query16(eeprom, eeprom_size,
-					    EEPROM_LINK_HOST);
-		break;
-	case INDIRECT_GENERAL:
-		offset = iwl_eeprom_query16(eeprom, eeprom_size,
-					    EEPROM_LINK_GENERAL);
-		break;
-	case INDIRECT_REGULATORY:
-		offset = iwl_eeprom_query16(eeprom, eeprom_size,
-					    EEPROM_LINK_REGULATORY);
-		break;
-	case INDIRECT_TXP_LIMIT:
-		offset = iwl_eeprom_query16(eeprom, eeprom_size,
-					    EEPROM_LINK_TXP_LIMIT);
-		break;
-	case INDIRECT_TXP_LIMIT_SIZE:
-		offset = iwl_eeprom_query16(eeprom, eeprom_size,
-					    EEPROM_LINK_TXP_LIMIT_SIZE);
-		break;
-	case INDIRECT_CALIBRATION:
-		offset = iwl_eeprom_query16(eeprom, eeprom_size,
-					    EEPROM_LINK_CALIBRATION);
-		break;
-	case INDIRECT_PROCESS_ADJST:
-		offset = iwl_eeprom_query16(eeprom, eeprom_size,
-					    EEPROM_LINK_PROCESS_ADJST);
-		break;
-	case INDIRECT_OTHERS:
-		offset = iwl_eeprom_query16(eeprom, eeprom_size,
-					    EEPROM_LINK_OTHERS);
-		break;
-	default:
-		WARN_ON(1);
-		break;
-	}
-
-	/* translate the offset from words to byte */
-	return (address & ADDRESS_MSK) + (offset << 1);
-}
-
-static const u8 *iwl_eeprom_query_addr(const u8 *eeprom, size_t eeprom_size,
-				       u32 offset)
-{
-	u32 address = eeprom_indirect_address(eeprom, eeprom_size, offset);
-
-	if (WARN_ON(address >= eeprom_size))
-		return NULL;
-
-	return &eeprom[address];
-}
-
-static int iwl_eeprom_read_calib(const u8 *eeprom, size_t eeprom_size,
-				 struct iwl_nvm_data *data)
-{
-	struct iwl_eeprom_calib_hdr *hdr;
-
-	hdr = (void *)iwl_eeprom_query_addr(eeprom, eeprom_size,
-					    EEPROM_CALIB_ALL);
-	if (!hdr)
-		return -ENODATA;
-	data->calib_version = hdr->version;
-	data->calib_voltage = hdr->voltage;
-
-	return 0;
-}
-
-/**
- * enum iwl_eeprom_channel_flags - channel flags in EEPROM
- * @EEPROM_CHANNEL_VALID: channel is usable for this SKU/geo
- * @EEPROM_CHANNEL_IBSS: usable as an IBSS channel
- * @EEPROM_CHANNEL_ACTIVE: active scanning allowed
- * @EEPROM_CHANNEL_RADAR: radar detection required
- * @EEPROM_CHANNEL_WIDE: 20 MHz channel okay (?)
- * @EEPROM_CHANNEL_DFS: dynamic freq selection candidate
- */
-enum iwl_eeprom_channel_flags {
-	EEPROM_CHANNEL_VALID = BIT(0),
-	EEPROM_CHANNEL_IBSS = BIT(1),
-	EEPROM_CHANNEL_ACTIVE = BIT(3),
-	EEPROM_CHANNEL_RADAR = BIT(4),
-	EEPROM_CHANNEL_WIDE = BIT(5),
-	EEPROM_CHANNEL_DFS = BIT(7),
-};
-
-/**
- * struct iwl_eeprom_channel - EEPROM channel data
- * @flags: %EEPROM_CHANNEL_* flags
- * @max_power_avg: max power (in dBm) on this channel, at most 31 dBm
- */
-struct iwl_eeprom_channel {
-	u8 flags;
-	s8 max_power_avg;
-} __packed;
-
-
-enum iwl_eeprom_enhanced_txpwr_flags {
-	IWL_EEPROM_ENH_TXP_FL_VALID = BIT(0),
-	IWL_EEPROM_ENH_TXP_FL_BAND_52G = BIT(1),
-	IWL_EEPROM_ENH_TXP_FL_OFDM = BIT(2),
-	IWL_EEPROM_ENH_TXP_FL_40MHZ = BIT(3),
-	IWL_EEPROM_ENH_TXP_FL_HT_AP = BIT(4),
-	IWL_EEPROM_ENH_TXP_FL_RES1 = BIT(5),
-	IWL_EEPROM_ENH_TXP_FL_RES2 = BIT(6),
-	IWL_EEPROM_ENH_TXP_FL_COMMON_TYPE = BIT(7),
-};
-
-/**
- * struct iwl_eeprom_enhanced_txpwr - enhanced regulatory TX power limits
- * @flags: entry flags
- * @channel: channel number
- * @chain_a_max: chain a max power in 1/2 dBm
- * @chain_b_max: chain b max power in 1/2 dBm
- * @chain_c_max: chain c max power in 1/2 dBm
- * @delta_20_in_40: 20-in-40 deltas (hi/lo)
- * @mimo2_max: mimo2 max power in 1/2 dBm
- * @mimo3_max: mimo3 max power in 1/2 dBm
- *
- * This structure presents the enhanced regulatory tx power limit layout
- * in an EEPROM image.
- */
-struct iwl_eeprom_enhanced_txpwr {
-	u8 flags;
-	u8 channel;
-	s8 chain_a_max;
-	s8 chain_b_max;
-	s8 chain_c_max;
-	u8 delta_20_in_40;
-	s8 mimo2_max;
-	s8 mimo3_max;
-} __packed;
-
-static s8 iwl_get_max_txpwr_half_dbm(const struct iwl_nvm_data *data,
-				     struct iwl_eeprom_enhanced_txpwr *txp)
-{
-	s8 result = 0; /* (.5 dBm) */
-
-	/* Take the highest tx power from any valid chains */
-	if (data->valid_tx_ant & ANT_A && txp->chain_a_max > result)
-		result = txp->chain_a_max;
-
-	if (data->valid_tx_ant & ANT_B && txp->chain_b_max > result)
-		result = txp->chain_b_max;
-
-	if (data->valid_tx_ant & ANT_C && txp->chain_c_max > result)
-		result = txp->chain_c_max;
-
-	if ((data->valid_tx_ant == ANT_AB ||
-	     data->valid_tx_ant == ANT_BC ||
-	     data->valid_tx_ant == ANT_AC) && txp->mimo2_max > result)
-		result = txp->mimo2_max;
-
-	if (data->valid_tx_ant == ANT_ABC && txp->mimo3_max > result)
-		result = txp->mimo3_max;
-
-	return result;
-}
-
-#define EEPROM_TXP_OFFS	(0x00 | INDIRECT_ADDRESS | INDIRECT_TXP_LIMIT)
-#define EEPROM_TXP_ENTRY_LEN sizeof(struct iwl_eeprom_enhanced_txpwr)
-#define EEPROM_TXP_SZ_OFFS (0x00 | INDIRECT_ADDRESS | INDIRECT_TXP_LIMIT_SIZE)
-
-#define TXP_CHECK_AND_PRINT(x) \
-	((txp->flags & IWL_EEPROM_ENH_TXP_FL_##x) ? # x " " : "")
-
-static void
-iwl_eeprom_enh_txp_read_element(struct iwl_nvm_data *data,
-				struct iwl_eeprom_enhanced_txpwr *txp,
-				int n_channels, s8 max_txpower_avg)
-{
-	int ch_idx;
-	enum nl80211_band band;
-
-	band = txp->flags & IWL_EEPROM_ENH_TXP_FL_BAND_52G ?
-		NL80211_BAND_5GHZ : NL80211_BAND_2GHZ;
-
-	for (ch_idx = 0; ch_idx < n_channels; ch_idx++) {
-		struct ieee80211_channel *chan = &data->channels[ch_idx];
-
-		/* update matching channel or from common data only */
-		if (txp->channel != 0 && chan->hw_value != txp->channel)
-			continue;
-
-		/* update matching band only */
-		if (band != chan->band)
-			continue;
-
-		if (chan->max_power < max_txpower_avg &&
-		    !(txp->flags & IWL_EEPROM_ENH_TXP_FL_40MHZ))
-			chan->max_power = max_txpower_avg;
-	}
-}
-
-static void iwl_eeprom_enhanced_txpower(struct device *dev,
-					struct iwl_nvm_data *data,
-					const u8 *eeprom, size_t eeprom_size,
-					int n_channels)
-{
-	struct iwl_eeprom_enhanced_txpwr *txp_array, *txp;
-	int idx, entries;
-	__le16 *txp_len;
-	s8 max_txp_avg_halfdbm;
-
-	BUILD_BUG_ON(sizeof(struct iwl_eeprom_enhanced_txpwr) != 8);
-
-	/* the length is in 16-bit words, but we want entries */
-	txp_len = (__le16 *)iwl_eeprom_query_addr(eeprom, eeprom_size,
-						  EEPROM_TXP_SZ_OFFS);
-	entries = le16_to_cpup(txp_len) * 2 / EEPROM_TXP_ENTRY_LEN;
-
-	txp_array = (void *)iwl_eeprom_query_addr(eeprom, eeprom_size,
-						  EEPROM_TXP_OFFS);
-
-	for (idx = 0; idx < entries; idx++) {
-		txp = &txp_array[idx];
-		/* skip invalid entries */
-		if (!(txp->flags & IWL_EEPROM_ENH_TXP_FL_VALID))
-			continue;
-
-		IWL_DEBUG_EEPROM(dev, "%s %d:\t %s%s%s%s%s%s%s%s (0x%02x)\n",
-				 (txp->channel && (txp->flags &
-					IWL_EEPROM_ENH_TXP_FL_COMMON_TYPE)) ?
-					"Common " : (txp->channel) ?
-					"Channel" : "Common",
-				 (txp->channel),
-				 TXP_CHECK_AND_PRINT(VALID),
-				 TXP_CHECK_AND_PRINT(BAND_52G),
-				 TXP_CHECK_AND_PRINT(OFDM),
-				 TXP_CHECK_AND_PRINT(40MHZ),
-				 TXP_CHECK_AND_PRINT(HT_AP),
-				 TXP_CHECK_AND_PRINT(RES1),
-				 TXP_CHECK_AND_PRINT(RES2),
-				 TXP_CHECK_AND_PRINT(COMMON_TYPE),
-				 txp->flags);
-		IWL_DEBUG_EEPROM(dev,
-				 "\t\t chain_A: %d chain_B: %d chain_C: %d\n",
-				 txp->chain_a_max, txp->chain_b_max,
-				 txp->chain_c_max);
-		IWL_DEBUG_EEPROM(dev,
-				 "\t\t MIMO2: %d MIMO3: %d High 20_on_40: 0x%02x Low 20_on_40: 0x%02x\n",
-				 txp->mimo2_max, txp->mimo3_max,
-				 ((txp->delta_20_in_40 & 0xf0) >> 4),
-				 (txp->delta_20_in_40 & 0x0f));
-
-		max_txp_avg_halfdbm = iwl_get_max_txpwr_half_dbm(data, txp);
-
-		iwl_eeprom_enh_txp_read_element(data, txp, n_channels,
-				DIV_ROUND_UP(max_txp_avg_halfdbm, 2));
-
-		if (max_txp_avg_halfdbm > data->max_tx_pwr_half_dbm)
-			data->max_tx_pwr_half_dbm = max_txp_avg_halfdbm;
-	}
-}
-
-static void iwl_init_band_reference(const struct iwl_cfg *cfg,
-				    const u8 *eeprom, size_t eeprom_size,
-				    int eeprom_band, int *eeprom_ch_count,
-				    const struct iwl_eeprom_channel **ch_info,
-				    const u8 **eeprom_ch_array)
-{
-	u32 offset = cfg->eeprom_params->regulatory_bands[eeprom_band - 1];
-
-	offset |= INDIRECT_ADDRESS | INDIRECT_REGULATORY;
-
-	*ch_info = (void *)iwl_eeprom_query_addr(eeprom, eeprom_size, offset);
-
-	switch (eeprom_band) {
-	case 1:		/* 2.4GHz band */
-		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_1);
-		*eeprom_ch_array = iwl_eeprom_band_1;
-		break;
-	case 2:		/* 4.9GHz band */
-		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_2);
-		*eeprom_ch_array = iwl_eeprom_band_2;
-		break;
-	case 3:		/* 5.2GHz band */
-		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_3);
-		*eeprom_ch_array = iwl_eeprom_band_3;
-		break;
-	case 4:		/* 5.5GHz band */
-		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_4);
-		*eeprom_ch_array = iwl_eeprom_band_4;
-		break;
-	case 5:		/* 5.7GHz band */
-		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_5);
-		*eeprom_ch_array = iwl_eeprom_band_5;
-		break;
-	case 6:		/* 2.4GHz ht40 channels */
-		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_6);
-		*eeprom_ch_array = iwl_eeprom_band_6;
-		break;
-	case 7:		/* 5 GHz ht40 channels */
-		*eeprom_ch_count = ARRAY_SIZE(iwl_eeprom_band_7);
-		*eeprom_ch_array = iwl_eeprom_band_7;
-		break;
-	default:
-		*eeprom_ch_count = 0;
-		*eeprom_ch_array = NULL;
-		WARN_ON(1);
-	}
-}
-
-#define CHECK_AND_PRINT(x) \
-	((eeprom_ch->flags & EEPROM_CHANNEL_##x) ? # x " " : "")
-
-static void iwl_mod_ht40_chan_info(struct device *dev,
-				   struct iwl_nvm_data *data, int n_channels,
-				   enum nl80211_band band, u16 channel,
-				   const struct iwl_eeprom_channel *eeprom_ch,
-				   u8 clear_ht40_extension_channel)
-{
-	struct ieee80211_channel *chan = NULL;
-	int i;
-
-	for (i = 0; i < n_channels; i++) {
-		if (data->channels[i].band != band)
-			continue;
-		if (data->channels[i].hw_value != channel)
-			continue;
-		chan = &data->channels[i];
-		break;
-	}
-
-	if (!chan)
-		return;
-
-	IWL_DEBUG_EEPROM(dev,
-			 "HT40 Ch. %d [%sGHz] %s%s%s%s%s(0x%02x %ddBm): Ad-Hoc %ssupported\n",
-			 channel,
-			 band == NL80211_BAND_5GHZ ? "5.2" : "2.4",
-			 CHECK_AND_PRINT(IBSS),
-			 CHECK_AND_PRINT(ACTIVE),
-			 CHECK_AND_PRINT(RADAR),
-			 CHECK_AND_PRINT(WIDE),
-			 CHECK_AND_PRINT(DFS),
-			 eeprom_ch->flags,
-			 eeprom_ch->max_power_avg,
-			 ((eeprom_ch->flags & EEPROM_CHANNEL_IBSS) &&
-			  !(eeprom_ch->flags & EEPROM_CHANNEL_RADAR)) ? ""
-								      : "not ");
-
-	if (eeprom_ch->flags & EEPROM_CHANNEL_VALID)
-		chan->flags &= ~clear_ht40_extension_channel;
-}
-
-#define CHECK_AND_PRINT_I(x)	\
-	((eeprom_ch_info[ch_idx].flags & EEPROM_CHANNEL_##x) ? # x " " : "")
-
-static int iwl_init_channel_map(struct device *dev, const struct iwl_cfg *cfg,
-				struct iwl_nvm_data *data,
-				const u8 *eeprom, size_t eeprom_size)
-{
-	int band, ch_idx;
-	const struct iwl_eeprom_channel *eeprom_ch_info;
-	const u8 *eeprom_ch_array;
-	int eeprom_ch_count;
-	int n_channels = 0;
-
-	/*
-	 * Loop through the 5 EEPROM bands and add them to the parse list
-	 */
-	for (band = 1; band <= 5; band++) {
-		struct ieee80211_channel *channel;
-
-		iwl_init_band_reference(cfg, eeprom, eeprom_size, band,
-					&eeprom_ch_count, &eeprom_ch_info,
-					&eeprom_ch_array);
-
-		/* Loop through each band adding each of the channels */
-		for (ch_idx = 0; ch_idx < eeprom_ch_count; ch_idx++) {
-			const struct iwl_eeprom_channel *eeprom_ch;
-
-			eeprom_ch = &eeprom_ch_info[ch_idx];
-
-			if (!(eeprom_ch->flags & EEPROM_CHANNEL_VALID)) {
-				IWL_DEBUG_EEPROM(dev,
-						 "Ch. %d Flags %x [%sGHz] - No traffic\n",
-						 eeprom_ch_array[ch_idx],
-						 eeprom_ch_info[ch_idx].flags,
-						 (band != 1) ? "5.2" : "2.4");
-				continue;
-			}
-
-			channel = &data->channels[n_channels];
-			n_channels++;
-
-			channel->hw_value = eeprom_ch_array[ch_idx];
-			channel->band = (band == 1) ? NL80211_BAND_2GHZ
-						    : NL80211_BAND_5GHZ;
-			channel->center_freq =
-				ieee80211_channel_to_frequency(
-					channel->hw_value, channel->band);
-
-			/* set no-HT40, will enable as appropriate later */
-			channel->flags = IEEE80211_CHAN_NO_HT40;
-
-			if (!(eeprom_ch->flags & EEPROM_CHANNEL_IBSS))
-				channel->flags |= IEEE80211_CHAN_NO_IR;
-
-			if (!(eeprom_ch->flags & EEPROM_CHANNEL_ACTIVE))
-				channel->flags |= IEEE80211_CHAN_NO_IR;
-
-			if (eeprom_ch->flags & EEPROM_CHANNEL_RADAR)
-				channel->flags |= IEEE80211_CHAN_RADAR;
-
-			/* Initialize regulatory-based run-time data */
-			channel->max_power =
-				eeprom_ch_info[ch_idx].max_power_avg;
-			IWL_DEBUG_EEPROM(dev,
-					 "Ch. %d [%sGHz] %s%s%s%s%s%s(0x%02x %ddBm): Ad-Hoc %ssupported\n",
-					 channel->hw_value,
-					 (band != 1) ? "5.2" : "2.4",
-					 CHECK_AND_PRINT_I(VALID),
-					 CHECK_AND_PRINT_I(IBSS),
-					 CHECK_AND_PRINT_I(ACTIVE),
-					 CHECK_AND_PRINT_I(RADAR),
-					 CHECK_AND_PRINT_I(WIDE),
-					 CHECK_AND_PRINT_I(DFS),
-					 eeprom_ch_info[ch_idx].flags,
-					 eeprom_ch_info[ch_idx].max_power_avg,
-					 ((eeprom_ch_info[ch_idx].flags &
-							EEPROM_CHANNEL_IBSS) &&
-					  !(eeprom_ch_info[ch_idx].flags &
-							EEPROM_CHANNEL_RADAR))
-						? "" : "not ");
-		}
-	}
-
-	if (cfg->eeprom_params->enhanced_txpower) {
-		/*
-		 * for newer device (6000 series and up)
-		 * EEPROM contain enhanced tx power information
-		 * driver need to process addition information
-		 * to determine the max channel tx power limits
-		 */
-		iwl_eeprom_enhanced_txpower(dev, data, eeprom, eeprom_size,
-					    n_channels);
-	} else {
-		/* All others use data from channel map */
-		int i;
-
-		data->max_tx_pwr_half_dbm = -128;
-
-		for (i = 0; i < n_channels; i++)
-			data->max_tx_pwr_half_dbm =
-				max_t(s8, data->max_tx_pwr_half_dbm,
-				      data->channels[i].max_power * 2);
-	}
-
-	/* Check if we do have HT40 channels */
-	if (cfg->eeprom_params->regulatory_bands[5] ==
-				EEPROM_REGULATORY_BAND_NO_HT40 &&
-	    cfg->eeprom_params->regulatory_bands[6] ==
-				EEPROM_REGULATORY_BAND_NO_HT40)
-		return n_channels;
-
-	/* Two additional EEPROM bands for 2.4 and 5 GHz HT40 channels */
-	for (band = 6; band <= 7; band++) {
-		enum nl80211_band ieeeband;
-
-		iwl_init_band_reference(cfg, eeprom, eeprom_size, band,
-					&eeprom_ch_count, &eeprom_ch_info,
-					&eeprom_ch_array);
-
-		/* EEPROM band 6 is 2.4, band 7 is 5 GHz */
-		ieeeband = (band == 6) ? NL80211_BAND_2GHZ
-				       : NL80211_BAND_5GHZ;
-
-		/* Loop through each band adding each of the channels */
-		for (ch_idx = 0; ch_idx < eeprom_ch_count; ch_idx++) {
-			/* Set up driver's info for lower half */
-			iwl_mod_ht40_chan_info(dev, data, n_channels, ieeeband,
-					       eeprom_ch_array[ch_idx],
-					       &eeprom_ch_info[ch_idx],
-					       IEEE80211_CHAN_NO_HT40PLUS);
-
-			/* Set up driver's info for upper half */
-			iwl_mod_ht40_chan_info(dev, data, n_channels, ieeeband,
-					       eeprom_ch_array[ch_idx] + 4,
-					       &eeprom_ch_info[ch_idx],
-					       IEEE80211_CHAN_NO_HT40MINUS);
-		}
-	}
-
-	return n_channels;
-}
-#endif
-
-int iwl_init_sband_channels(struct iwl_nvm_data *data,
-			    struct ieee80211_supported_band *sband,
-			    int n_channels, enum nl80211_band band)
-{
-	struct ieee80211_channel *chan = &data->channels[0];
-	int n = 0, idx = 0;
-
-	while (idx < n_channels && chan->band != band)
-		chan = &data->channels[++idx];
-
-	sband->channels = &data->channels[idx];
-
-	while (idx < n_channels && chan->band == band) {
-		chan = &data->channels[++idx];
-		n++;
-	}
-
-	sband->n_channels = n;
-
-	return n;
-}
-
-#define MAX_BIT_RATE_40_MHZ	150 /* Mbps */
-#define MAX_BIT_RATE_20_MHZ	72 /* Mbps */
-
-void iwl_init_ht_hw_capab(struct iwl_trans *trans,
-			  struct iwl_nvm_data *data,
-			  struct ieee80211_sta_ht_cap *ht_info,
-			  enum nl80211_band band,
-			  u8 tx_chains, u8 rx_chains)
-{
-	const struct iwl_cfg *cfg = trans->cfg;
-	int max_bit_rate = 0;
-
-	tx_chains = hweight8(tx_chains);
-	if (cfg->rx_with_siso_diversity)
-		rx_chains = 1;
-	else
-		rx_chains = hweight8(rx_chains);
-
-	if (!(data->sku_cap_11n_enable) ||
-	    (iwlwifi_mod_params.disable_11n & IWL_DISABLE_HT_ALL) ||
-	    !cfg->ht_params) {
-		ht_info->ht_supported = false;
-		return;
-	}
-
-	if (data->sku_cap_mimo_disabled)
-		rx_chains = 1;
-
-	ht_info->ht_supported = true;
-	ht_info->cap = IEEE80211_HT_CAP_DSSSCCK40;
-
-	if (cfg->ht_params->stbc) {
-		ht_info->cap |= (1 << IEEE80211_HT_CAP_RX_STBC_SHIFT);
-
-		if (tx_chains > 1)
-			ht_info->cap |= IEEE80211_HT_CAP_TX_STBC;
-	}
-
-	if (cfg->ht_params->ldpc)
-		ht_info->cap |= IEEE80211_HT_CAP_LDPC_CODING;
-
-	if (trans->trans_cfg->mq_rx_supported ||
-	    iwlwifi_mod_params.amsdu_size >= IWL_AMSDU_8K)
-		ht_info->cap |= IEEE80211_HT_CAP_MAX_AMSDU;
-
-	ht_info->ampdu_factor = IEEE80211_HT_MAX_AMPDU_64K;
-	ht_info->ampdu_density = IEEE80211_HT_MPDU_DENSITY_4;
-
-	ht_info->mcs.rx_mask[0] = 0xFF;
-	ht_info->mcs.rx_mask[1] = 0x00;
-	ht_info->mcs.rx_mask[2] = 0x00;
-
-	if (rx_chains >= 2)
-		ht_info->mcs.rx_mask[1] = 0xFF;
-	if (rx_chains >= 3)
-		ht_info->mcs.rx_mask[2] = 0xFF;
-
-	if (cfg->ht_params->ht_greenfield_support)
-		ht_info->cap |= IEEE80211_HT_CAP_GRN_FLD;
-	ht_info->cap |= IEEE80211_HT_CAP_SGI_20;
-
-	max_bit_rate = MAX_BIT_RATE_20_MHZ;
-
-	if (cfg->ht_params->ht40_bands & BIT(band)) {
-		ht_info->cap |= IEEE80211_HT_CAP_SUP_WIDTH_20_40;
-		ht_info->cap |= IEEE80211_HT_CAP_SGI_40;
-		max_bit_rate = MAX_BIT_RATE_40_MHZ;
-	}
-
-	/* Highest supported Rx data rate */
-	max_bit_rate *= rx_chains;
-	WARN_ON(max_bit_rate & ~IEEE80211_HT_MCS_RX_HIGHEST_MASK);
-	ht_info->mcs.rx_highest = cpu_to_le16(max_bit_rate);
-
-	/* Tx MCS capabilities */
-	ht_info->mcs.tx_params = IEEE80211_HT_MCS_TX_DEFINED;
-	if (tx_chains != rx_chains) {
-		ht_info->mcs.tx_params |= IEEE80211_HT_MCS_TX_RX_DIFF;
-		ht_info->mcs.tx_params |= ((tx_chains - 1) <<
-				IEEE80211_HT_MCS_TX_MAX_STREAMS_SHIFT);
-	}
-}
-
-#if IS_ENABLED(CONFIG_IWLDVM)
-static void iwl_init_sbands(struct iwl_trans *trans, const struct iwl_cfg *cfg,
-			    struct iwl_nvm_data *data,
-			    const u8 *eeprom, size_t eeprom_size)
-{
-	struct device *dev = trans->dev;
-	int n_channels = iwl_init_channel_map(dev, cfg, data,
-					      eeprom, eeprom_size);
-	int n_used = 0;
-	struct ieee80211_supported_band *sband;
-
-	sband = &data->bands[NL80211_BAND_2GHZ];
-	sband->band = NL80211_BAND_2GHZ;
-	sband->bitrates = &iwl_cfg80211_rates[RATES_24_OFFS];
-	sband->n_bitrates = N_RATES_24;
-	n_used += iwl_init_sband_channels(data, sband, n_channels,
-					  NL80211_BAND_2GHZ);
-	iwl_init_ht_hw_capab(trans, data, &sband->ht_cap, NL80211_BAND_2GHZ,
-			     data->valid_tx_ant, data->valid_rx_ant);
-
-	sband = &data->bands[NL80211_BAND_5GHZ];
-	sband->band = NL80211_BAND_5GHZ;
-	sband->bitrates = &iwl_cfg80211_rates[RATES_52_OFFS];
-	sband->n_bitrates = N_RATES_52;
-	n_used += iwl_init_sband_channels(data, sband, n_channels,
-					  NL80211_BAND_5GHZ);
-	iwl_init_ht_hw_capab(trans, data, &sband->ht_cap, NL80211_BAND_5GHZ,
-			     data->valid_tx_ant, data->valid_rx_ant);
-
-	if (n_channels != n_used)
-		IWL_ERR_DEV(dev, "EEPROM: used only %d of %d channels\n",
-			    n_used, n_channels);
-}
-
-/* EEPROM data functions */
-
-struct iwl_nvm_data *
-iwl_parse_eeprom_data(struct iwl_trans *trans, const struct iwl_cfg *cfg,
-		      const u8 *eeprom, size_t eeprom_size)
-{
-	struct iwl_nvm_data *data;
-	struct device *dev = trans->dev;
-	const void *tmp;
-	u16 radio_cfg, sku;
-
-	if (WARN_ON(!cfg || !cfg->eeprom_params))
-		return NULL;
-
-	data = kzalloc(struct_size(data, channels, IWL_NUM_CHANNELS),
-		       GFP_KERNEL);
-	if (!data)
-		return NULL;
-
-	/* get MAC address(es) */
-	tmp = iwl_eeprom_query_addr(eeprom, eeprom_size, EEPROM_MAC_ADDRESS);
-	if (!tmp)
-		goto err_free;
-	memcpy(data->hw_addr, tmp, ETH_ALEN);
-	data->n_hw_addrs = iwl_eeprom_query16(eeprom, eeprom_size,
-					      EEPROM_NUM_MAC_ADDRESS);
-
-	if (iwl_eeprom_read_calib(eeprom, eeprom_size, data))
-		goto err_free;
-
-	tmp = iwl_eeprom_query_addr(eeprom, eeprom_size, EEPROM_XTAL);
-	if (!tmp)
-		goto err_free;
-	memcpy(data->xtal_calib, tmp, sizeof(data->xtal_calib));
-
-	tmp = iwl_eeprom_query_addr(eeprom, eeprom_size,
-				    EEPROM_RAW_TEMPERATURE);
-	if (!tmp)
-		goto err_free;
-	data->raw_temperature = *(__le16 *)tmp;
-
-	tmp = iwl_eeprom_query_addr(eeprom, eeprom_size,
-				    EEPROM_KELVIN_TEMPERATURE);
-	if (!tmp)
-		goto err_free;
-	data->kelvin_temperature = *(__le16 *)tmp;
-	data->kelvin_voltage = *((__le16 *)tmp + 1);
-
-	radio_cfg = iwl_eeprom_query16(eeprom, eeprom_size,
-					     EEPROM_RADIO_CONFIG);
-	data->radio_cfg_dash = EEPROM_RF_CFG_DASH_MSK(radio_cfg);
-	data->radio_cfg_pnum = EEPROM_RF_CFG_PNUM_MSK(radio_cfg);
-	data->radio_cfg_step = EEPROM_RF_CFG_STEP_MSK(radio_cfg);
-	data->radio_cfg_type = EEPROM_RF_CFG_TYPE_MSK(radio_cfg);
-	data->valid_rx_ant = EEPROM_RF_CFG_RX_ANT_MSK(radio_cfg);
-	data->valid_tx_ant = EEPROM_RF_CFG_TX_ANT_MSK(radio_cfg);
-
-	sku = iwl_eeprom_query16(eeprom, eeprom_size,
-				 EEPROM_SKU_CAP);
-	data->sku_cap_11n_enable = sku & EEPROM_SKU_CAP_11N_ENABLE;
-	data->sku_cap_amt_enable = sku & EEPROM_SKU_CAP_AMT_ENABLE;
-	data->sku_cap_band_24ghz_enable = sku & EEPROM_SKU_CAP_BAND_24GHZ;
-	data->sku_cap_band_52ghz_enable = sku & EEPROM_SKU_CAP_BAND_52GHZ;
-	data->sku_cap_ipan_enable = sku & EEPROM_SKU_CAP_IPAN_ENABLE;
-	if (iwlwifi_mod_params.disable_11n & IWL_DISABLE_HT_ALL)
-		data->sku_cap_11n_enable = false;
-
-	data->nvm_version = iwl_eeprom_query16(eeprom, eeprom_size,
-					       EEPROM_VERSION);
-
-	/* check overrides (some devices have wrong EEPROM) */
-	if (cfg->valid_tx_ant)
-		data->valid_tx_ant = cfg->valid_tx_ant;
-	if (cfg->valid_rx_ant)
-		data->valid_rx_ant = cfg->valid_rx_ant;
-
-	if (!data->valid_tx_ant || !data->valid_rx_ant) {
-		IWL_ERR_DEV(dev, "invalid antennas (0x%x, 0x%x)\n",
-			    data->valid_tx_ant, data->valid_rx_ant);
-		goto err_free;
-	}
-
-	iwl_init_sbands(trans, cfg, data, eeprom, eeprom_size);
-
-	return data;
- err_free:
-	kfree(data);
-	return NULL;
-}
-IWL_EXPORT_SYMBOL(iwl_parse_eeprom_data);
-#endif
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h b/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h
deleted file mode 100644
index 34a178a2eb5d..000000000000
--- a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h
+++ /dev/null
@@ -1,88 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
-/*
- * Copyright (C) 2005-2014, 2018, 2020-2023 Intel Corporation
- * Copyright (C) 2015 Intel Mobile Communications GmbH
- */
-#ifndef __iwl_eeprom_parse_h__
-#define __iwl_eeprom_parse_h__
-
-#include <linux/types.h>
-#include <linux/if_ether.h>
-#include <net/cfg80211.h>
-#include "iwl-trans.h"
-
-struct iwl_nvm_data {
-	int n_hw_addrs;
-	u8 hw_addr[ETH_ALEN];
-
-	u8 calib_version;
-	__le16 calib_voltage;
-
-	__le16 raw_temperature;
-	__le16 kelvin_temperature;
-	__le16 kelvin_voltage;
-	__le16 xtal_calib[2];
-
-	bool sku_cap_band_24ghz_enable;
-	bool sku_cap_band_52ghz_enable;
-	bool sku_cap_11n_enable;
-	bool sku_cap_11ac_enable;
-	bool sku_cap_11ax_enable;
-	bool sku_cap_amt_enable;
-	bool sku_cap_ipan_enable;
-	bool sku_cap_mimo_disabled;
-	bool sku_cap_11be_enable;
-
-	u16 radio_cfg_type;
-	u8 radio_cfg_step;
-	u8 radio_cfg_dash;
-	u8 radio_cfg_pnum;
-	u8 valid_tx_ant, valid_rx_ant;
-
-	u32 nvm_version;
-	s8 max_tx_pwr_half_dbm;
-
-	bool lar_enabled;
-	bool vht160_supported;
-	struct ieee80211_supported_band bands[NUM_NL80211_BANDS];
-
-	/*
-	 * iftype data for low (2.4 GHz) high (5 GHz) and uhb (6 GHz) bands
-	 */
-	struct {
-		struct ieee80211_sband_iftype_data low[2];
-		struct ieee80211_sband_iftype_data high[2];
-		struct ieee80211_sband_iftype_data uhb[2];
-	} iftd;
-
-	struct ieee80211_channel channels[];
-};
-
-/**
- * iwl_parse_eeprom_data - parse EEPROM data and return values
- *
- * @trans: ransport we're parsing for, for debug only
- * @cfg: device configuration for parsing and overrides
- * @eeprom: the EEPROM data
- * @eeprom_size: length of the EEPROM data
- *
- * This function parses all EEPROM values we need and then
- * returns a (newly allocated) struct containing all the
- * relevant values for driver use. The struct must be freed
- * later with iwl_free_nvm_data().
- */
-struct iwl_nvm_data *
-iwl_parse_eeprom_data(struct iwl_trans *trans, const struct iwl_cfg *cfg,
-		      const u8 *eeprom, size_t eeprom_size);
-
-int iwl_init_sband_channels(struct iwl_nvm_data *data,
-			    struct ieee80211_supported_band *sband,
-			    int n_channels, enum nl80211_band band);
-
-void iwl_init_ht_hw_capab(struct iwl_trans *trans,
-			  struct iwl_nvm_data *data,
-			  struct ieee80211_sta_ht_cap *ht_info,
-			  enum nl80211_band band,
-			  u8 tx_chains, u8 rx_chains);
-
-#endif /* __iwl_eeprom_parse_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
index fd9c3bed9407..5a1b5ab62de1 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
@@ -7,7 +7,7 @@
 #define __iwl_nvm_parse_h__
 
 #include <net/cfg80211.h>
-#include "iwl-eeprom-parse.h"
+#include "iwl-nvm-utils.h"
 #include "mei/iwl-mei.h"
 
 /**
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-utils.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-utils.c
new file mode 100644
index 000000000000..b3c25acd3691
--- /dev/null
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-utils.c
@@ -0,0 +1,118 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/*
+ * Copyright (C) 2005-2014, 2018-2021, 2023 Intel Corporation
+ * Copyright (C) 2015 Intel Mobile Communications GmbH
+ */
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/export.h>
+#include "iwl-drv.h"
+#include "iwl-modparams.h"
+#include "iwl-nvm-utils.h"
+
+int iwl_init_sband_channels(struct iwl_nvm_data *data,
+			    struct ieee80211_supported_band *sband,
+			    int n_channels, enum nl80211_band band)
+{
+	struct ieee80211_channel *chan = &data->channels[0];
+	int n = 0, idx = 0;
+
+	while (idx < n_channels && chan->band != band)
+		chan = &data->channels[++idx];
+
+	sband->channels = &data->channels[idx];
+
+	while (idx < n_channels && chan->band == band) {
+		chan = &data->channels[++idx];
+		n++;
+	}
+
+	sband->n_channels = n;
+
+	return n;
+}
+IWL_EXPORT_SYMBOL(iwl_init_sband_channels);
+
+#define MAX_BIT_RATE_40_MHZ	150 /* Mbps */
+#define MAX_BIT_RATE_20_MHZ	72 /* Mbps */
+
+void iwl_init_ht_hw_capab(struct iwl_trans *trans,
+			  struct iwl_nvm_data *data,
+			  struct ieee80211_sta_ht_cap *ht_info,
+			  enum nl80211_band band,
+			  u8 tx_chains, u8 rx_chains)
+{
+	const struct iwl_cfg *cfg = trans->cfg;
+	int max_bit_rate = 0;
+
+	tx_chains = hweight8(tx_chains);
+	if (cfg->rx_with_siso_diversity)
+		rx_chains = 1;
+	else
+		rx_chains = hweight8(rx_chains);
+
+	if (!(data->sku_cap_11n_enable) ||
+	    (iwlwifi_mod_params.disable_11n & IWL_DISABLE_HT_ALL) ||
+	    !cfg->ht_params) {
+		ht_info->ht_supported = false;
+		return;
+	}
+
+	if (data->sku_cap_mimo_disabled)
+		rx_chains = 1;
+
+	ht_info->ht_supported = true;
+	ht_info->cap = IEEE80211_HT_CAP_DSSSCCK40;
+
+	if (cfg->ht_params->stbc) {
+		ht_info->cap |= (1 << IEEE80211_HT_CAP_RX_STBC_SHIFT);
+
+		if (tx_chains > 1)
+			ht_info->cap |= IEEE80211_HT_CAP_TX_STBC;
+	}
+
+	if (cfg->ht_params->ldpc)
+		ht_info->cap |= IEEE80211_HT_CAP_LDPC_CODING;
+
+	if (trans->trans_cfg->mq_rx_supported ||
+	    iwlwifi_mod_params.amsdu_size >= IWL_AMSDU_8K)
+		ht_info->cap |= IEEE80211_HT_CAP_MAX_AMSDU;
+
+	ht_info->ampdu_factor = IEEE80211_HT_MAX_AMPDU_64K;
+	ht_info->ampdu_density = IEEE80211_HT_MPDU_DENSITY_4;
+
+	ht_info->mcs.rx_mask[0] = 0xFF;
+	ht_info->mcs.rx_mask[1] = 0x00;
+	ht_info->mcs.rx_mask[2] = 0x00;
+
+	if (rx_chains >= 2)
+		ht_info->mcs.rx_mask[1] = 0xFF;
+	if (rx_chains >= 3)
+		ht_info->mcs.rx_mask[2] = 0xFF;
+
+	if (cfg->ht_params->ht_greenfield_support)
+		ht_info->cap |= IEEE80211_HT_CAP_GRN_FLD;
+	ht_info->cap |= IEEE80211_HT_CAP_SGI_20;
+
+	max_bit_rate = MAX_BIT_RATE_20_MHZ;
+
+	if (cfg->ht_params->ht40_bands & BIT(band)) {
+		ht_info->cap |= IEEE80211_HT_CAP_SUP_WIDTH_20_40;
+		ht_info->cap |= IEEE80211_HT_CAP_SGI_40;
+		max_bit_rate = MAX_BIT_RATE_40_MHZ;
+	}
+
+	/* Highest supported Rx data rate */
+	max_bit_rate *= rx_chains;
+	WARN_ON(max_bit_rate & ~IEEE80211_HT_MCS_RX_HIGHEST_MASK);
+	ht_info->mcs.rx_highest = cpu_to_le16(max_bit_rate);
+
+	/* Tx MCS capabilities */
+	ht_info->mcs.tx_params = IEEE80211_HT_MCS_TX_DEFINED;
+	if (tx_chains != rx_chains) {
+		ht_info->mcs.tx_params |= IEEE80211_HT_MCS_TX_RX_DIFF;
+		ht_info->mcs.tx_params |= ((tx_chains - 1) <<
+				IEEE80211_HT_MCS_TX_MAX_STREAMS_SHIFT);
+	}
+}
+IWL_EXPORT_SYMBOL(iwl_init_ht_hw_capab);
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-utils.h b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-utils.h
new file mode 100644
index 000000000000..ac0a29a1c31f
--- /dev/null
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-utils.h
@@ -0,0 +1,71 @@
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+/*
+ * Copyright (C) 2005-2014, 2018, 2020-2023 Intel Corporation
+ * Copyright (C) 2015 Intel Mobile Communications GmbH
+ */
+#ifndef __iwl_eeprom_parse_h__
+#define __iwl_eeprom_parse_h__
+
+#include <linux/types.h>
+#include <linux/if_ether.h>
+#include <net/cfg80211.h>
+#include "iwl-trans.h"
+
+struct iwl_nvm_data {
+	int n_hw_addrs;
+	u8 hw_addr[ETH_ALEN];
+
+	u8 calib_version;
+	__le16 calib_voltage;
+
+	__le16 raw_temperature;
+	__le16 kelvin_temperature;
+	__le16 kelvin_voltage;
+	__le16 xtal_calib[2];
+
+	bool sku_cap_band_24ghz_enable;
+	bool sku_cap_band_52ghz_enable;
+	bool sku_cap_11n_enable;
+	bool sku_cap_11ac_enable;
+	bool sku_cap_11ax_enable;
+	bool sku_cap_amt_enable;
+	bool sku_cap_ipan_enable;
+	bool sku_cap_mimo_disabled;
+	bool sku_cap_11be_enable;
+
+	u16 radio_cfg_type;
+	u8 radio_cfg_step;
+	u8 radio_cfg_dash;
+	u8 radio_cfg_pnum;
+	u8 valid_tx_ant, valid_rx_ant;
+
+	u32 nvm_version;
+	s8 max_tx_pwr_half_dbm;
+
+	bool lar_enabled;
+	bool vht160_supported;
+	struct ieee80211_supported_band bands[NUM_NL80211_BANDS];
+
+	/*
+	 * iftype data for low (2.4 GHz) high (5 GHz) and uhb (6 GHz) bands
+	 */
+	struct {
+		struct ieee80211_sband_iftype_data low[2];
+		struct ieee80211_sband_iftype_data high[2];
+		struct ieee80211_sband_iftype_data uhb[2];
+	} iftd;
+
+	struct ieee80211_channel channels[];
+};
+
+int iwl_init_sband_channels(struct iwl_nvm_data *data,
+			    struct ieee80211_supported_band *sband,
+			    int n_channels, enum nl80211_band band);
+
+void iwl_init_ht_hw_capab(struct iwl_trans *trans,
+			  struct iwl_nvm_data *data,
+			  struct ieee80211_sta_ht_cap *ht_info,
+			  enum nl80211_band band,
+			  u8 tx_chains, u8 rx_chains);
+
+#endif /* __iwl_eeprom_parse_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index f2f9e2620ff6..8ec699d9f316 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -22,7 +22,7 @@
 #include "mvm.h"
 #include "sta.h"
 #include "time-event.h"
-#include "iwl-eeprom-parse.h"
+#include "iwl-nvm-utils.h"
 #include "iwl-phy-db.h"
 #include "testmode.h"
 #include "fw/error-dump.h"
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index dd4599ca1f45..770a7098b0f9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -24,7 +24,7 @@
 #include "iwl-op-mode.h"
 #include "iwl-trans.h"
 #include "fw/notif-wait.h"
-#include "iwl-eeprom-parse.h"
+#include "iwl-nvm-utils.h"
 #include "fw/file.h"
 #include "iwl-config.h"
 #include "sta.h"
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
index 1e949b48188e..836ca22597bc 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c
@@ -9,7 +9,7 @@
 #include "iwl-trans.h"
 #include "iwl-csr.h"
 #include "mvm.h"
-#include "iwl-eeprom-parse.h"
+#include "iwl-nvm-utils.h"
 #include "iwl-nvm-parse.h"
 #include "iwl-prph.h"
 #include "fw/acpi.h"
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 160bb619ba9b..516decc647dd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -18,7 +18,7 @@
 #include "iwl-modparams.h"
 #include "mvm.h"
 #include "iwl-phy-db.h"
-#include "iwl-eeprom-parse.h"
+#include "iwl-nvm-utils.h"
 #include "iwl-csr.h"
 #include "iwl-io.h"
 #include "iwl-prph.h"
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 1d695ece93e9..599550f311de 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -12,7 +12,7 @@
 #include <net/ipv6.h>
 
 #include "iwl-trans.h"
-#include "iwl-eeprom-parse.h"
+#include "iwl-nvm-utils.h"
 #include "mvm.h"
 #include "sta.h"
 #include "time-sync.h"
-- 
cgit v1.2.3-58-ga151


From 48443a4a1c97dd8b6614833af928bc15aeff2e47 Mon Sep 17 00:00:00 2001
From: Anjaneyulu <pagadala.yesu.anjaneyulu@intel.com>
Date: Sun, 12 May 2024 15:25:05 +0300
Subject: wifi: iwlwifi: Add support for LARI_CONFIG_CHANGE_CMD v11

Add support for enable/disable 11be for china and South Korea by
reading DSM function 12 from UEFI or ACPI and sending it to the FW.

Signed-off-by: Anjaneyulu <pagadala.yesu.anjaneyulu@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240512152312.037613d03c87.I4e5ee63c0dbdd4bd6c48daee7b0c88462b702423@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/acpi.c       |  1 +
 .../net/wireless/intel/iwlwifi/fw/api/nvm-reg.h    | 52 ++++++++++++++++++++--
 drivers/net/wireless/intel/iwlwifi/fw/regulatory.c | 15 ++++++-
 drivers/net/wireless/intel/iwlwifi/fw/regulatory.h |  3 +-
 4 files changed, 65 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c
index fa339791223b..79774c8c7ff4 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c
@@ -27,6 +27,7 @@ static const size_t acpi_dsm_size[DSM_FUNC_NUM_FUNCS] = {
 	[DSM_FUNC_FORCE_DISABLE_CHANNELS] =	sizeof(u32),
 	[DSM_FUNC_ENERGY_DETECTION_THRESHOLD] =	sizeof(u32),
 	[DSM_FUNC_RFI_CONFIG] =			sizeof(u32),
+	[DSM_FUNC_ENABLE_11BE] =		sizeof(u32),
 };
 
 static int iwl_acpi_get_handle(struct device *dev, acpi_string method,
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h
index a08497a04733..644c8df780bc 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h
@@ -23,7 +23,8 @@ enum iwl_regulatory_and_nvm_subcmd_ids {
 	 *	&struct iwl_lari_config_change_cmd_v4,
 	 *	&struct iwl_lari_config_change_cmd_v5,
 	 *	&struct iwl_lari_config_change_cmd_v6,
-	 *	&struct iwl_lari_config_change_cmd_v7 or
+	 *	&struct iwl_lari_config_change_cmd_v7,
+	 *	&struct iwl_lari_config_change_cmd_v10 or
 	 *	&struct iwl_lari_config_change_cmd
 	 */
 	LARI_CONFIG_CHANGE = 0x1,
@@ -648,7 +649,7 @@ struct iwl_lari_config_change_cmd_v7 {
 /* LARI_CHANGE_CONF_CMD_S_VER_9 */
 
 /**
- * struct iwl_lari_config_change_cmd - change LARI configuration
+ * struct iwl_lari_config_change_cmd_v10 - change LARI configuration
  * @config_bitmap: Bitmap of the config commands. Each bit will trigger a
  *	different predefined FW config operation.
  * @oem_uhb_allow_bitmap: Bitmap of UHB enabled MCC sets.
@@ -674,7 +675,7 @@ struct iwl_lari_config_change_cmd_v7 {
  *	bit1: enable 320Mhz in South Korea.
  *	bit 2 - 31: reserved.
  */
-struct iwl_lari_config_change_cmd {
+struct iwl_lari_config_change_cmd_v10 {
 	__le32 config_bitmap;
 	__le32 oem_uhb_allow_bitmap;
 	__le32 oem_11ax_allow_bitmap;
@@ -686,6 +687,51 @@ struct iwl_lari_config_change_cmd {
 } __packed;
 /* LARI_CHANGE_CONF_CMD_S_VER_10 */
 
+/**
+ * struct iwl_lari_config_change_cmd - change LARI configuration
+ * @config_bitmap: Bitmap of the config commands. Each bit will trigger a
+ *	different predefined FW config operation.
+ * @oem_uhb_allow_bitmap: Bitmap of UHB enabled MCC sets.
+ * @oem_11ax_allow_bitmap: Bitmap of 11ax allowed MCCs. There are two bits
+ *	per country, one to indicate whether to override and the other to
+ *	indicate the value to use.
+ * @oem_unii4_allow_bitmap: Bitmap of unii4 allowed MCCs.There are two bits
+ *	per country, one to indicate whether to override and the other to
+ *	indicate allow/disallow unii4 channels.
+ *	For LARI cmd version 11 - bits 0:5 are supported.
+ * @chan_state_active_bitmap: Bitmap to enable different bands per country
+ *	or region.
+ *	Each bit represents a country or region, and a band to activate
+ *	according to the BIOS definitions.
+ *	For LARI cmd version 11 - bits 0:4 are supported.
+ * @force_disable_channels_bitmap: Bitmap of disabled bands/channels.
+ *	Each bit represents a set of channels in a specific band that should be
+ *	disabled
+ * @edt_bitmap: Bitmap of energy detection threshold table.
+ *	Disable/enable the EDT optimization method for different band.
+ * @oem_320mhz_allow_bitmap: 320Mhz bandwidth enablement bitmap per MCC.
+ *	bit0: enable 320Mhz in Japan.
+ *	bit1: enable 320Mhz in South Korea.
+ *	bit 2 - 31: reserved.
+ * @oem_11be_allow_bitmap: Bitmap of 11be allowed MCCs. No need to mask out the
+ *	unsupported bits
+ *	bit0: enable 11be in China(CB/CN).
+ *	bit1: enable 11be in South Korea.
+ *	bit 2 - 31: reserved.
+ */
+struct iwl_lari_config_change_cmd {
+	__le32 config_bitmap;
+	__le32 oem_uhb_allow_bitmap;
+	__le32 oem_11ax_allow_bitmap;
+	__le32 oem_unii4_allow_bitmap;
+	__le32 chan_state_active_bitmap;
+	__le32 force_disable_channels_bitmap;
+	__le32 edt_bitmap;
+	__le32 oem_320mhz_allow_bitmap;
+	__le32 oem_11be_allow_bitmap;
+} __packed;
+/* LARI_CHANGE_CONF_CMD_S_VER_11 */
+
 /* Activate UNII-1 (5.2GHz) for World Wide */
 #define ACTIVATE_5G2_IN_WW_MASK	BIT(4)
 
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c b/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c
index b9bb3636e88f..5228b837a9ef 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c
@@ -497,9 +497,12 @@ static size_t iwl_get_lari_config_cmd_size(u8 cmd_ver)
 	size_t cmd_size;
 
 	switch (cmd_ver) {
-	case 10:
+	case 11:
 		cmd_size = sizeof(struct iwl_lari_config_change_cmd);
 		break;
+	case 10:
+		cmd_size = sizeof(struct iwl_lari_config_change_cmd_v10);
+		break;
 	case 9:
 	case 8:
 	case 7:
@@ -580,6 +583,10 @@ int iwl_fill_lari_config(struct iwl_fw_runtime *fwrt,
 	if (!ret)
 		cmd->oem_320mhz_allow_bitmap = cpu_to_le32(value);
 
+	ret = iwl_bios_get_dsm(fwrt, DSM_FUNC_ENABLE_11BE, &value);
+	if (!ret)
+		cmd->oem_11be_allow_bitmap = cpu_to_le32(value);
+
 	if (cmd->config_bitmap ||
 	    cmd->oem_uhb_allow_bitmap ||
 	    cmd->oem_11ax_allow_bitmap ||
@@ -587,7 +594,8 @@ int iwl_fill_lari_config(struct iwl_fw_runtime *fwrt,
 	    cmd->chan_state_active_bitmap ||
 	    cmd->force_disable_channels_bitmap ||
 	    cmd->edt_bitmap ||
-	    cmd->oem_320mhz_allow_bitmap) {
+	    cmd->oem_320mhz_allow_bitmap ||
+	    cmd->oem_11be_allow_bitmap) {
 		IWL_DEBUG_RADIO(fwrt,
 				"sending LARI_CONFIG_CHANGE, config_bitmap=0x%x, oem_11ax_allow_bitmap=0x%x\n",
 				le32_to_cpu(cmd->config_bitmap),
@@ -605,6 +613,9 @@ int iwl_fill_lari_config(struct iwl_fw_runtime *fwrt,
 				"sending LARI_CONFIG_CHANGE, edt_bitmap=0x%x, oem_320mhz_allow_bitmap=0x%x\n",
 				le32_to_cpu(cmd->edt_bitmap),
 				le32_to_cpu(cmd->oem_320mhz_allow_bitmap));
+		IWL_DEBUG_RADIO(fwrt,
+				"sending LARI_CONFIG_CHANGE, oem_11be_allow_bitmap=0x%x\n",
+				le32_to_cpu(cmd->oem_11be_allow_bitmap));
 	} else {
 		return 1;
 	}
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/regulatory.h b/drivers/net/wireless/intel/iwlwifi/fw/regulatory.h
index 633c9ad9af84..e2c056f483c1 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/regulatory.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/regulatory.h
@@ -115,7 +115,8 @@ enum iwl_dsm_funcs {
 	DSM_FUNC_FORCE_DISABLE_CHANNELS = 9,
 	DSM_FUNC_ENERGY_DETECTION_THRESHOLD = 10,
 	DSM_FUNC_RFI_CONFIG = 11,
-	DSM_FUNC_NUM_FUNCS = 12,
+	DSM_FUNC_ENABLE_11BE = 12,
+	DSM_FUNC_NUM_FUNCS = 13,
 };
 
 enum iwl_dsm_values_srd {
-- 
cgit v1.2.3-58-ga151


From ce9e660ef32e87441bf59b04f67a24113e82546a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:13:12 +0200
Subject: wifi: mac80211: move radar detect work to sdata

At some point we thought perhaps this could be per link, but
really that didn't happen, and it's confusing. Radar detection
still uses the deflink to allocate the channel, but the work
need not be there. Move it back.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506211311.43bd82c6da04.Ib39bec3aa198d137385f081e7e1910dcbde3aa1b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c         | 6 +++---
 net/mac80211/ieee80211_i.h | 3 ++-
 net/mac80211/iface.c       | 4 +++-
 net/mac80211/link.c        | 2 --
 net/mac80211/mlme.c        | 9 ++++-----
 net/mac80211/util.c        | 2 +-
 6 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 3673c7f6b21a..0ed7e8bbf5c0 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1666,7 +1666,7 @@ static int ieee80211_stop_ap(struct wiphy *wiphy, struct net_device *dev,
 
 	if (sdata->wdev.cac_started) {
 		chandef = link_conf->chanreq.oper;
-		wiphy_delayed_work_cancel(wiphy, &link->dfs_cac_timer_work);
+		wiphy_delayed_work_cancel(wiphy, &sdata->dfs_cac_timer_work);
 		cfg80211_cac_event(sdata->dev, &chandef,
 				   NL80211_RADAR_CAC_ABORTED,
 				   GFP_KERNEL);
@@ -3466,7 +3466,7 @@ static int ieee80211_start_radar_detection(struct wiphy *wiphy,
 	if (err)
 		goto out_unlock;
 
-	wiphy_delayed_work_queue(wiphy, &sdata->deflink.dfs_cac_timer_work,
+	wiphy_delayed_work_queue(wiphy, &sdata->dfs_cac_timer_work,
 				 msecs_to_jiffies(cac_time_ms));
 
  out_unlock:
@@ -3483,7 +3483,7 @@ static void ieee80211_end_cac(struct wiphy *wiphy,
 
 	list_for_each_entry(sdata, &local->interfaces, list) {
 		wiphy_delayed_work_cancel(wiphy,
-					  &sdata->deflink.dfs_cac_timer_work);
+					  &sdata->dfs_cac_timer_work);
 
 		if (sdata->wdev.cac_started) {
 			ieee80211_link_release_channel(&sdata->deflink);
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 7d541a2355f6..4f4ef6710e8d 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -1060,7 +1060,6 @@ struct ieee80211_link_data {
 	int ap_power_level; /* in dBm */
 
 	bool radar_required;
-	struct wiphy_delayed_work dfs_cac_timer_work;
 
 	union {
 		struct ieee80211_link_data_managed mgd;
@@ -1159,6 +1158,8 @@ struct ieee80211_sub_if_data {
 	struct ieee80211_link_data deflink;
 	struct ieee80211_link_data __rcu *link[IEEE80211_MLD_MAX_NUM_LINKS];
 
+	struct wiphy_delayed_work dfs_cac_timer_work;
+
 	/* for ieee80211_set_active_links_async() */
 	struct wiphy_work activate_links_work;
 	u16 desired_active_links;
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index dc42902e2693..1ef4b6dc3730 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -554,7 +554,7 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, bool going_do
 	wiphy_work_cancel(local->hw.wiphy,
 			  &sdata->deflink.color_change_finalize_work);
 	wiphy_delayed_work_cancel(local->hw.wiphy,
-				  &sdata->deflink.dfs_cac_timer_work);
+				  &sdata->dfs_cac_timer_work);
 
 	if (sdata->wdev.cac_started) {
 		chandef = sdata->vif.bss_conf.chanreq.oper;
@@ -1746,6 +1746,8 @@ static void ieee80211_setup_sdata(struct ieee80211_sub_if_data *sdata,
 	wiphy_work_init(&sdata->work, ieee80211_iface_work);
 	wiphy_work_init(&sdata->activate_links_work,
 			ieee80211_activate_links_work);
+	wiphy_delayed_work_init(&sdata->dfs_cac_timer_work,
+				ieee80211_dfs_cac_timer_work);
 
 	switch (type) {
 	case NL80211_IFTYPE_P2P_GO:
diff --git a/net/mac80211/link.c b/net/mac80211/link.c
index af0321408a97..f57282488222 100644
--- a/net/mac80211/link.c
+++ b/net/mac80211/link.c
@@ -45,8 +45,6 @@ void ieee80211_link_init(struct ieee80211_sub_if_data *sdata,
 			  ieee80211_color_collision_detection_work);
 	INIT_LIST_HEAD(&link->assigned_chanctx_list);
 	INIT_LIST_HEAD(&link->reserved_chanctx_list);
-	wiphy_delayed_work_init(&link->dfs_cac_timer_work,
-				ieee80211_dfs_cac_timer_work);
 
 	if (!deflink) {
 		switch (sdata->vif.type) {
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 08c0999746fb..a2b8bfc1fe0e 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2797,16 +2797,15 @@ void ieee80211_dynamic_ps_timer(struct timer_list *t)
 
 void ieee80211_dfs_cac_timer_work(struct wiphy *wiphy, struct wiphy_work *work)
 {
-	struct ieee80211_link_data *link =
-		container_of(work, struct ieee80211_link_data,
+	struct ieee80211_sub_if_data *sdata =
+		container_of(work, struct ieee80211_sub_if_data,
 			     dfs_cac_timer_work.work);
-	struct cfg80211_chan_def chandef = link->conf->chanreq.oper;
-	struct ieee80211_sub_if_data *sdata = link->sdata;
+	struct cfg80211_chan_def chandef = sdata->vif.bss_conf.chanreq.oper;
 
 	lockdep_assert_wiphy(sdata->local->hw.wiphy);
 
 	if (sdata->wdev.cac_started) {
-		ieee80211_link_release_channel(link);
+		ieee80211_link_release_channel(&sdata->deflink);
 		cfg80211_cac_event(sdata->dev, &chandef,
 				   NL80211_RADAR_CAC_FINISHED,
 				   GFP_KERNEL);
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 183de2075fb9..43625ca87d5e 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -3457,7 +3457,7 @@ void ieee80211_dfs_cac_cancel(struct ieee80211_local *local)
 
 	list_for_each_entry(sdata, &local->interfaces, list) {
 		wiphy_delayed_work_cancel(local->hw.wiphy,
-					  &sdata->deflink.dfs_cac_timer_work);
+					  &sdata->dfs_cac_timer_work);
 
 		if (sdata->wdev.cac_started) {
 			chandef = sdata->vif.bss_conf.chanreq.oper;
-- 
cgit v1.2.3-58-ga151


From 2d33ecf5d0148671c74e68e18755b9411a7ba923 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:11:59 +0200
Subject: wifi: cfg80211: restrict operation during radar detection

Just like it's not currently possible to start radar
detection while already operating, it shouldn't be
possible to start operating while radar detection is
running. Fix that.

Also, improve the check whether operating (carrier
might not be up if e.g. attempting to join IBSS).

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506211158.ae8dca3d0d6c.I7c70a66a5fbdbc63a78fee8a34f31d1995491bc3@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/ibss.c    |  5 ++++-
 net/wireless/mesh.c    |  5 ++++-
 net/wireless/nl80211.c | 21 +++++++++++++++------
 3 files changed, 23 insertions(+), 8 deletions(-)

diff --git a/net/wireless/ibss.c b/net/wireless/ibss.c
index 9f02ee5f08be..34e5acff3935 100644
--- a/net/wireless/ibss.c
+++ b/net/wireless/ibss.c
@@ -3,7 +3,7 @@
  * Some IBSS support code for cfg80211.
  *
  * Copyright 2009	Johannes Berg <johannes@sipsolutions.net>
- * Copyright (C) 2020-2023 Intel Corporation
+ * Copyright (C) 2020-2024 Intel Corporation
  */
 
 #include <linux/etherdevice.h>
@@ -94,6 +94,9 @@ int __cfg80211_join_ibss(struct cfg80211_registered_device *rdev,
 
 	lockdep_assert_held(&rdev->wiphy.mtx);
 
+	if (wdev->cac_started)
+		return -EBUSY;
+
 	if (wdev->u.ibss.ssid_len)
 		return -EALREADY;
 
diff --git a/net/wireless/mesh.c b/net/wireless/mesh.c
index 83306979fbe2..aaca65b66af4 100644
--- a/net/wireless/mesh.c
+++ b/net/wireless/mesh.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /*
  * Portions
- * Copyright (C) 2022-2023 Intel Corporation
+ * Copyright (C) 2022-2024 Intel Corporation
  */
 #include <linux/ieee80211.h>
 #include <linux/export.h>
@@ -127,6 +127,9 @@ int __cfg80211_join_mesh(struct cfg80211_registered_device *rdev,
 	if (!rdev->ops->join_mesh)
 		return -EOPNOTSUPP;
 
+	if (wdev->cac_started)
+		return -EBUSY;
+
 	if (!setup->chandef.chan) {
 		/* if no channel explicitly given, use preset channel */
 		setup->chandef = wdev->u.mesh.preset_chandef;
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 93c313149f57..6ba988a6f5a2 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -5964,6 +5964,9 @@ static int nl80211_start_ap(struct sk_buff *skb, struct genl_info *info)
 	if (!rdev->ops->start_ap)
 		return -EOPNOTSUPP;
 
+	if (wdev->cac_started)
+		return -EBUSY;
+
 	if (wdev->links[link_id].ap.beacon_interval)
 		return -EALREADY;
 
@@ -9956,6 +9959,17 @@ static int nl80211_start_radar_detection(struct sk_buff *skb,
 
 	flush_delayed_work(&rdev->dfs_update_channels_wk);
 
+	switch (wdev->iftype) {
+	case NL80211_IFTYPE_AP:
+	case NL80211_IFTYPE_P2P_GO:
+	case NL80211_IFTYPE_MESH_POINT:
+	case NL80211_IFTYPE_ADHOC:
+		break;
+	default:
+		/* caution - see cfg80211_beaconing_iface_active() below */
+		return -EINVAL;
+	}
+
 	wiphy_lock(wiphy);
 
 	dfs_region = reg_get_dfs_region(wiphy);
@@ -9986,12 +10000,7 @@ static int nl80211_start_radar_detection(struct sk_buff *skb,
 		goto unlock;
 	}
 
-	if (netif_carrier_ok(dev)) {
-		err = -EBUSY;
-		goto unlock;
-	}
-
-	if (wdev->cac_started) {
+	if (cfg80211_beaconing_iface_active(wdev) || wdev->cac_started) {
 		err = -EBUSY;
 		goto unlock;
 	}
-- 
cgit v1.2.3-58-ga151


From 7ef8f6821d16623a3f6ea279542f9f47ece9ff1e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 15 Apr 2024 11:27:11 +0300
Subject: wifi: mac80211: mlme: handle cross-link CSA

If we see a channel switch announcement on one link for
another, handle that case and start the CSA. The driver
can react to this in whatever way it needs. The stack
will have the ability to track it via the RNR/MLE in the
reporting link's beacon if it sees it for inactive links
and adjust everything accordingly.

Note that currently the timings for the CSA aren't set,
the values are only used by the Intel drivers, and they
don't need this for newer devices that support MLO, so
I've left it out for now.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240415112355.4d34b6a31be7.Ie8453979f5805873a8411c99346bcc3810cd6476@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/driver-ops.h  |  12 ++
 net/mac80211/ieee80211_i.h |   1 +
 net/mac80211/link.c        |  12 ++
 net/mac80211/mlme.c        | 470 ++++++++++++++++++++++++++++++++++++---------
 4 files changed, 403 insertions(+), 92 deletions(-)

diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h
index 5d078c0a2323..d4e73d3630e0 100644
--- a/net/mac80211/driver-ops.h
+++ b/net/mac80211/driver-ops.h
@@ -1150,6 +1150,9 @@ drv_pre_channel_switch(struct ieee80211_sub_if_data *sdata,
 	if (!check_sdata_in_driver(sdata))
 		return -EIO;
 
+	if (!ieee80211_vif_link_active(&sdata->vif, ch_switch->link_id))
+		return 0;
+
 	trace_drv_pre_channel_switch(local, sdata, ch_switch);
 	if (local->ops->pre_channel_switch)
 		ret = local->ops->pre_channel_switch(&local->hw, &sdata->vif,
@@ -1171,6 +1174,9 @@ drv_post_channel_switch(struct ieee80211_link_data *link)
 	if (!check_sdata_in_driver(sdata))
 		return -EIO;
 
+	if (!ieee80211_vif_link_active(&sdata->vif, link->link_id))
+		return 0;
+
 	trace_drv_post_channel_switch(local, sdata);
 	if (local->ops->post_channel_switch)
 		ret = local->ops->post_channel_switch(&local->hw, &sdata->vif,
@@ -1191,6 +1197,9 @@ drv_abort_channel_switch(struct ieee80211_link_data *link)
 	if (!check_sdata_in_driver(sdata))
 		return;
 
+	if (!ieee80211_vif_link_active(&sdata->vif, link->link_id))
+		return;
+
 	trace_drv_abort_channel_switch(local, sdata);
 
 	if (local->ops->abort_channel_switch)
@@ -1210,6 +1219,9 @@ drv_channel_switch_rx_beacon(struct ieee80211_sub_if_data *sdata,
 	if (!check_sdata_in_driver(sdata))
 		return;
 
+	if (!ieee80211_vif_link_active(&sdata->vif, ch_switch->link_id))
+		return;
+
 	trace_drv_channel_switch_rx_beacon(local, sdata, ch_switch);
 	if (local->ops->channel_switch_rx_beacon)
 		local->ops->channel_switch_rx_beacon(&local->hw, &sdata->vif,
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 4f4ef6710e8d..4ce34e95902e 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -978,6 +978,7 @@ struct ieee80211_link_data_managed {
 	bool csa_waiting_bcn;
 	bool csa_ignored_same_chan;
 	bool csa_blocked_tx;
+	unsigned long csa_time;
 	struct wiphy_delayed_work chswitch_work;
 
 	struct wiphy_work request_smps_work;
diff --git a/net/mac80211/link.c b/net/mac80211/link.c
index f57282488222..508759cc9979 100644
--- a/net/mac80211/link.c
+++ b/net/mac80211/link.c
@@ -357,6 +357,18 @@ static int _ieee80211_set_active_links(struct ieee80211_sub_if_data *sdata,
 		ieee80211_teardown_tdls_peers(link);
 
 		__ieee80211_link_release_channel(link, true);
+
+		/*
+		 * If CSA is (still) active while the link is deactivated,
+		 * just schedule the channel switch work for the time we
+		 * had previously calculated, and we'll take the process
+		 * from there.
+		 */
+		if (link->conf->csa_active)
+			wiphy_delayed_work_queue(local->hw.wiphy,
+						 &link->u.mgd.chswitch_work,
+						 link->u.mgd.csa_time -
+						 jiffies);
 	}
 
 	list_for_each_entry(sta, &local->sta_list, list) {
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index a2b8bfc1fe0e..aedd1ce1fb8d 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2079,6 +2079,18 @@ static void ieee80211_chswitch_work(struct wiphy *wiphy,
 	if (!link->conf->csa_active)
 		return;
 
+	/*
+	 * If the link isn't active (now), we cannot wait for beacons, won't
+	 * have a reserved chanctx, etc. Just switch over the chandef and
+	 * update cfg80211 directly.
+	 */
+	if (!ieee80211_vif_link_active(&sdata->vif, link->link_id)) {
+		link->conf->chanreq = link->csa_chanreq;
+		cfg80211_ch_switch_notify(sdata->dev, &link->csa_chanreq.oper,
+					  link->link_id);
+		return;
+	}
+
 	/*
 	 * using reservation isn't immediate as it may be deferred until later
 	 * with multi-vif. once reservation is complete it will re-schedule the
@@ -2097,9 +2109,9 @@ static void ieee80211_chswitch_work(struct wiphy *wiphy,
 
 		ret = ieee80211_link_use_reserved_context(link);
 		if (ret) {
-			sdata_info(sdata,
-				   "failed to use reserved channel context, disconnecting (err=%d)\n",
-				   ret);
+			link_info(link,
+				  "failed to use reserved channel context, disconnecting (err=%d)\n",
+				  ret);
 			wiphy_work_queue(sdata->local->hw.wiphy,
 					 &ifmgd->csa_connection_drop_work);
 		}
@@ -2108,8 +2120,8 @@ static void ieee80211_chswitch_work(struct wiphy *wiphy,
 
 	if (!ieee80211_chanreq_identical(&link->conf->chanreq,
 					 &link->csa_chanreq)) {
-		sdata_info(sdata,
-			   "failed to finalize channel switch, disconnecting\n");
+		link_info(link,
+			  "failed to finalize channel switch, disconnecting\n");
 		wiphy_work_queue(sdata->local->hw.wiphy,
 				 &ifmgd->csa_connection_drop_work);
 		return;
@@ -2144,14 +2156,14 @@ static void ieee80211_chswitch_post_beacon(struct ieee80211_link_data *link)
 
 	ret = drv_post_channel_switch(link);
 	if (ret) {
-		sdata_info(sdata,
-			   "driver post channel switch failed, disconnecting\n");
+		link_info(link,
+			  "driver post channel switch failed, disconnecting\n");
 		wiphy_work_queue(sdata->local->hw.wiphy,
 				 &ifmgd->csa_connection_drop_work);
 		return;
 	}
 
-	cfg80211_ch_switch_notify(sdata->dev, &link->reserved.oper,
+	cfg80211_ch_switch_notify(sdata->dev, &link->conf->chanreq.oper,
 				  link->link_id);
 }
 
@@ -2166,7 +2178,8 @@ void ieee80211_chswitch_done(struct ieee80211_vif *vif, bool success,
 
 	if (!success) {
 		sdata_info(sdata,
-			   "driver channel switch failed, disconnecting\n");
+			   "driver channel switch failed (link %d), disconnecting\n",
+			   link_id);
 		wiphy_work_queue(sdata->local->hw.wiphy,
 				 &sdata->u.mgd.csa_connection_drop_work);
 	} else {
@@ -2211,69 +2224,221 @@ ieee80211_sta_abort_chanswitch(struct ieee80211_link_data *link)
 	drv_abort_channel_switch(link);
 }
 
+struct sta_csa_rnr_iter_data {
+	struct ieee80211_link_data *link;
+	struct ieee80211_channel *chan;
+	u8 mld_id;
+};
+
+static enum cfg80211_rnr_iter_ret
+ieee80211_sta_csa_rnr_iter(void *_data, u8 type,
+			   const struct ieee80211_neighbor_ap_info *info,
+			   const u8 *tbtt_info, u8 tbtt_info_len)
+{
+	struct sta_csa_rnr_iter_data *data = _data;
+	struct ieee80211_link_data *link = data->link;
+	struct ieee80211_sub_if_data *sdata = link->sdata;
+	struct ieee80211_if_managed *ifmgd = &sdata->u.mgd;
+	const struct ieee80211_tbtt_info_ge_11 *ti;
+	enum nl80211_band band;
+	unsigned int center_freq;
+	int link_id;
+
+	if (type != IEEE80211_TBTT_INFO_TYPE_TBTT)
+		return RNR_ITER_CONTINUE;
+
+	if (tbtt_info_len < sizeof(*ti))
+		return RNR_ITER_CONTINUE;
+
+	ti = (const void *)tbtt_info;
+
+	if (ti->mld_params.mld_id != data->mld_id)
+		return RNR_ITER_CONTINUE;
+
+	link_id = le16_get_bits(ti->mld_params.params,
+				IEEE80211_RNR_MLD_PARAMS_LINK_ID);
+	if (link_id != data->link->link_id)
+		return RNR_ITER_CONTINUE;
+
+	/* we found the entry for our link! */
+
+	/* this AP is confused, it had this right before ... just disconnect */
+	if (!ieee80211_operating_class_to_band(info->op_class, &band)) {
+		link_info(link,
+			  "AP now has invalid operating class in RNR, disconnect\n");
+		wiphy_work_queue(sdata->local->hw.wiphy,
+				 &ifmgd->csa_connection_drop_work);
+		return RNR_ITER_BREAK;
+	}
+
+	center_freq = ieee80211_channel_to_frequency(info->channel, band);
+	data->chan = ieee80211_get_channel(sdata->local->hw.wiphy, center_freq);
+
+	return RNR_ITER_BREAK;
+}
+
+static void
+ieee80211_sta_other_link_csa_disappeared(struct ieee80211_link_data *link,
+					 struct ieee802_11_elems *elems)
+{
+	struct ieee80211_sub_if_data *sdata = link->sdata;
+	struct ieee80211_if_managed *ifmgd = &sdata->u.mgd;
+	struct sta_csa_rnr_iter_data data = {
+		.link = link,
+	};
+
+	/*
+	 * If we get here, we see a beacon from another link without
+	 * CSA still being reported for it, so now we have to check
+	 * if the CSA was aborted or completed. This may not even be
+	 * perfectly possible if the CSA was only done for changing
+	 * the puncturing, but in that case if the link in inactive
+	 * we don't really care, and if it's an active link (or when
+	 * it's activated later) we'll get a beacon and adjust.
+	 */
+
+	if (WARN_ON(!elems->ml_basic))
+		return;
+
+	data.mld_id = ieee80211_mle_get_mld_id((const void *)elems->ml_basic);
+
+	/*
+	 * So in order to do this, iterate the RNR element(s) and see
+	 * what channel is reported now.
+	 */
+	cfg80211_iter_rnr(elems->ie_start, elems->total_len,
+			  ieee80211_sta_csa_rnr_iter, &data);
+
+	if (!data.chan) {
+		link_info(link,
+			  "couldn't find (valid) channel in RNR for CSA, disconnect\n");
+		wiphy_work_queue(sdata->local->hw.wiphy,
+				 &ifmgd->csa_connection_drop_work);
+		return;
+	}
+
+	/*
+	 * If it doesn't match the CSA, then assume it aborted. This
+	 * may erroneously detect that it was _not_ aborted when it
+	 * was in fact aborted, but only changed the bandwidth or the
+	 * puncturing configuration, but we don't have enough data to
+	 * detect that.
+	 */
+	if (data.chan != link->csa_chanreq.oper.chan)
+		ieee80211_sta_abort_chanswitch(link);
+}
+
+enum ieee80211_csa_source {
+	IEEE80211_CSA_SOURCE_BEACON,
+	IEEE80211_CSA_SOURCE_OTHER_LINK,
+	IEEE80211_CSA_SOURCE_ACTION,
+};
+
 static void
 ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 				 u64 timestamp, u32 device_timestamp,
-				 struct ieee802_11_elems *elems,
-				 bool beacon)
+				 struct ieee802_11_elems *full_elems,
+				 struct ieee802_11_elems *csa_elems,
+				 enum ieee80211_csa_source source)
 {
 	struct ieee80211_sub_if_data *sdata = link->sdata;
 	struct ieee80211_local *local = sdata->local;
 	struct ieee80211_if_managed *ifmgd = &sdata->u.mgd;
-	struct cfg80211_bss *cbss = link->conf->bss;
+	struct ieee80211_chanctx *chanctx = NULL;
 	struct ieee80211_chanctx_conf *conf;
-	struct ieee80211_chanctx *chanctx;
-	enum nl80211_band current_band;
-	struct ieee80211_csa_ie csa_ie;
+	struct ieee80211_csa_ie csa_ie = {};
 	struct ieee80211_channel_switch ch_switch = {
 		.link_id = link->link_id,
+		.timestamp = timestamp,
+		.device_timestamp = device_timestamp,
 	};
-	struct ieee80211_bss *bss;
-	unsigned long timeout;
+	unsigned long now;
 	int res;
 
 	lockdep_assert_wiphy(local->hw.wiphy);
 
-	if (!cbss)
-		return;
+	if (csa_elems) {
+		struct cfg80211_bss *cbss = link->conf->bss;
+		enum nl80211_band current_band;
+		struct ieee80211_bss *bss;
 
-	current_band = cbss->channel->band;
-	bss = (void *)cbss->priv;
-	res = ieee80211_parse_ch_switch_ie(sdata, elems, current_band,
-					   bss->vht_cap_info,
-					   &link->u.mgd.conn,
-					   link->u.mgd.bssid, &csa_ie);
+		if (WARN_ON(!cbss))
+			return;
 
-	if (!res) {
-		ch_switch.timestamp = timestamp;
-		ch_switch.device_timestamp = device_timestamp;
-		ch_switch.block_tx = csa_ie.mode;
-		ch_switch.chandef = csa_ie.chanreq.oper;
-		ch_switch.count = csa_ie.count;
-		ch_switch.delay = csa_ie.max_switch_time;
+		current_band = cbss->channel->band;
+		bss = (void *)cbss->priv;
+
+		res = ieee80211_parse_ch_switch_ie(sdata, csa_elems,
+						   current_band,
+						   bss->vht_cap_info,
+						   &link->u.mgd.conn,
+						   link->u.mgd.bssid, &csa_ie);
+		if (res == 0) {
+			ch_switch.block_tx = csa_ie.mode;
+			ch_switch.chandef = csa_ie.chanreq.oper;
+			ch_switch.count = csa_ie.count;
+			ch_switch.delay = csa_ie.max_switch_time;
+		}
+	} else {
+		/*
+		 * If there was no per-STA profile for this link, we
+		 * get called with csa_elems == NULL. This of course means
+		 * there are no CSA elements, so set res=1 indicating
+		 * no more CSA.
+		 */
+		res = 1;
 	}
 
 	if (res < 0)
 		goto drop_connection;
 
 	if (link->conf->csa_active) {
-		/* already processing - disregard action frames */
-		if (!beacon)
+		switch (source) {
+		case IEEE80211_CSA_SOURCE_ACTION:
+			/* already processing - disregard action frames */
 			return;
+		case IEEE80211_CSA_SOURCE_BEACON:
+			if (link->u.mgd.csa_waiting_bcn) {
+				ieee80211_chswitch_post_beacon(link);
+				/*
+				 * If the CSA is still present after the switch
+				 * we need to consider it as a new CSA (possibly
+				 * to self). This happens by not returning here
+				 * so we'll get to the check below.
+				 */
+			} else if (res) {
+				ieee80211_sta_abort_chanswitch(link);
+				return;
+			} else {
+				drv_channel_switch_rx_beacon(sdata, &ch_switch);
+				return;
+			}
+			break;
+		case IEEE80211_CSA_SOURCE_OTHER_LINK:
+			/* active link: we want to see the beacon to continue */
+			if (ieee80211_vif_link_active(&sdata->vif,
+						      link->link_id))
+				return;
 
-		if (link->u.mgd.csa_waiting_bcn) {
-			ieee80211_chswitch_post_beacon(link);
-			/*
-			 * If the CSA IE is still present in the beacon after
-			 * the switch, we need to consider it as a new CSA
-			 * (possibly to self) - this happens by not returning
-			 * here so we'll get to the check below.
-			 */
-		} else if (res) {
-			ieee80211_sta_abort_chanswitch(link);
-			return;
-		} else {
-			drv_channel_switch_rx_beacon(sdata, &ch_switch);
+			/* switch work ran, so just complete the process */
+			if (link->u.mgd.csa_waiting_bcn) {
+				ieee80211_chswitch_post_beacon(link);
+				/*
+				 * If the CSA is still present after the switch
+				 * we need to consider it as a new CSA (possibly
+				 * to self). This happens by not returning here
+				 * so we'll get to the check below.
+				 */
+				break;
+			}
+
+			/* link still has CSA but we already know, do nothing */
+			if (!res)
+				return;
+
+			/* check in the RNR if the CSA aborted */
+			ieee80211_sta_other_link_csa_disappeared(link,
+								 full_elems);
 			return;
 		}
 	}
@@ -2284,40 +2449,38 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 
 	if (link->conf->chanreq.oper.chan->band !=
 	    csa_ie.chanreq.oper.chan->band) {
-		sdata_info(sdata,
-			   "AP %pM switches to different band (%d MHz, width:%d, CF1/2: %d/%d MHz), disconnecting\n",
-			   link->u.mgd.bssid,
-			   csa_ie.chanreq.oper.chan->center_freq,
-			   csa_ie.chanreq.oper.width,
-			   csa_ie.chanreq.oper.center_freq1,
-			   csa_ie.chanreq.oper.center_freq2);
+		link_info(link,
+			  "AP %pM switches to different band (%d MHz, width:%d, CF1/2: %d/%d MHz), disconnecting\n",
+			  link->u.mgd.bssid,
+			  csa_ie.chanreq.oper.chan->center_freq,
+			  csa_ie.chanreq.oper.width,
+			  csa_ie.chanreq.oper.center_freq1,
+			  csa_ie.chanreq.oper.center_freq2);
 		goto drop_connection;
 	}
 
 	if (!cfg80211_chandef_usable(local->hw.wiphy, &csa_ie.chanreq.oper,
 				     IEEE80211_CHAN_DISABLED)) {
-		sdata_info(sdata,
-			   "AP %pM switches to unsupported channel "
-			   "(%d.%03d MHz, width:%d, CF1/2: %d.%03d/%d MHz), "
-			   "disconnecting\n",
-			   link->u.mgd.bssid,
-			   csa_ie.chanreq.oper.chan->center_freq,
-			   csa_ie.chanreq.oper.chan->freq_offset,
-			   csa_ie.chanreq.oper.width,
-			   csa_ie.chanreq.oper.center_freq1,
-			   csa_ie.chanreq.oper.freq1_offset,
-			   csa_ie.chanreq.oper.center_freq2);
+		link_info(link,
+			  "AP %pM switches to unsupported channel (%d.%03d MHz, width:%d, CF1/2: %d.%03d/%d MHz), disconnecting\n",
+			  link->u.mgd.bssid,
+			  csa_ie.chanreq.oper.chan->center_freq,
+			  csa_ie.chanreq.oper.chan->freq_offset,
+			  csa_ie.chanreq.oper.width,
+			  csa_ie.chanreq.oper.center_freq1,
+			  csa_ie.chanreq.oper.freq1_offset,
+			  csa_ie.chanreq.oper.center_freq2);
 		goto drop_connection;
 	}
 
 	if (cfg80211_chandef_identical(&csa_ie.chanreq.oper,
 				       &link->conf->chanreq.oper) &&
-	    (!csa_ie.mode || !beacon)) {
+	    (!csa_ie.mode || source != IEEE80211_CSA_SOURCE_BEACON)) {
 		if (link->u.mgd.csa_ignored_same_chan)
 			return;
-		sdata_info(sdata,
-			   "AP %pM tries to chanswitch to same channel, ignore\n",
-			   link->u.mgd.bssid);
+		link_info(link,
+			  "AP %pM tries to chanswitch to same channel, ignore\n",
+			  link->u.mgd.bssid);
 		link->u.mgd.csa_ignored_same_chan = true;
 		return;
 	}
@@ -2333,33 +2496,36 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 
 	conf = rcu_dereference_protected(link->conf->chanctx_conf,
 					 lockdep_is_held(&local->hw.wiphy->mtx));
-	if (!conf) {
-		sdata_info(sdata,
-			   "no channel context assigned to vif?, disconnecting\n");
+	if (ieee80211_vif_link_active(&sdata->vif, link->link_id) && !conf) {
+		link_info(link,
+			  "no channel context assigned to vif?, disconnecting\n");
 		goto drop_connection;
 	}
 
-	chanctx = container_of(conf, struct ieee80211_chanctx, conf);
+	if (conf)
+		chanctx = container_of(conf, struct ieee80211_chanctx, conf);
 
 	if (!ieee80211_hw_check(&local->hw, CHANCTX_STA_CSA)) {
-		sdata_info(sdata,
-			   "driver doesn't support chan-switch with channel contexts\n");
+		link_info(link,
+			  "driver doesn't support chan-switch with channel contexts\n");
 		goto drop_connection;
 	}
 
 	if (drv_pre_channel_switch(sdata, &ch_switch)) {
-		sdata_info(sdata,
-			   "preparing for channel switch failed, disconnecting\n");
+		link_info(link,
+			  "preparing for channel switch failed, disconnecting\n");
 		goto drop_connection;
 	}
 
-	res = ieee80211_link_reserve_chanctx(link, &csa_ie.chanreq,
-					     chanctx->mode, false);
-	if (res) {
-		sdata_info(sdata,
-			   "failed to reserve channel context for channel switch, disconnecting (err=%d)\n",
-			   res);
-		goto drop_connection;
+	if (chanctx) {
+		res = ieee80211_link_reserve_chanctx(link, &csa_ie.chanreq,
+						     chanctx->mode, false);
+		if (res) {
+			link_info(link,
+				  "failed to reserve channel context for channel switch, disconnecting (err=%d)\n",
+				  res);
+			goto drop_connection;
+		}
 	}
 
 	link->conf->csa_active = true;
@@ -2379,18 +2545,28 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 					  link->link_id, csa_ie.count,
 					  csa_ie.mode);
 
-	if (local->ops->channel_switch) {
-		/* use driver's channel switch callback */
+	/* we may have to handle timeout for deactivated link in software */
+	now = jiffies;
+	link->u.mgd.csa_time = now +
+			       TU_TO_JIFFIES((max_t(int, csa_ie.count, 1) - 1) *
+					     link->conf->beacon_int);
+
+	if (ieee80211_vif_link_active(&sdata->vif, link->link_id) &&
+	    local->ops->channel_switch) {
+		/*
+		 * Use driver's channel switch callback, the driver will
+		 * later call ieee80211_chswitch_done(). It may deactivate
+		 * the link as well, we handle that elsewhere and queue
+		 * the chswitch_work for the calculated time then.
+		 */
 		drv_channel_switch(local, sdata, &ch_switch);
 		return;
 	}
 
 	/* channel switch handled in software */
-	timeout = TU_TO_JIFFIES((max_t(int, csa_ie.count, 1) - 1) *
-				cbss->beacon_interval);
 	wiphy_delayed_work_queue(local->hw.wiphy,
 				 &link->u.mgd.chswitch_work,
-				 timeout);
+				 link->u.mgd.csa_time - now);
 	return;
  drop_connection:
 	/*
@@ -6328,6 +6504,110 @@ static void ieee80211_process_adv_ttlm(struct ieee80211_sub_if_data *sdata,
 	}
 }
 
+static void
+ieee80211_mgd_check_cross_link_csa(struct ieee80211_sub_if_data *sdata,
+				   int reporting_link_id,
+				   struct ieee802_11_elems *elems)
+{
+	const struct element *sta_profiles[IEEE80211_MLD_MAX_NUM_LINKS] = {};
+	ssize_t sta_profiles_len[IEEE80211_MLD_MAX_NUM_LINKS] = {};
+	const struct element *sub;
+	const u8 *subelems;
+	size_t subelems_len;
+	u8 common_size;
+	int link_id;
+
+	if (!ieee80211_mle_size_ok((u8 *)elems->ml_basic, elems->ml_basic_len))
+		return;
+
+	common_size = ieee80211_mle_common_size((u8 *)elems->ml_basic);
+	subelems = (u8 *)elems->ml_basic + common_size;
+	subelems_len = elems->ml_basic_len - common_size;
+
+	for_each_element_id(sub, IEEE80211_MLE_SUBELEM_PER_STA_PROFILE,
+			    subelems, subelems_len) {
+		struct ieee80211_mle_per_sta_profile *prof = (void *)sub->data;
+		struct ieee80211_link_data *link;
+		ssize_t len;
+
+		if (!ieee80211_mle_basic_sta_prof_size_ok(sub->data,
+							  sub->datalen))
+			continue;
+
+		link_id = le16_get_bits(prof->control,
+					IEEE80211_MLE_STA_CONTROL_LINK_ID);
+		/* need a valid link ID, but also not our own, both AP bugs */
+		if (link_id == reporting_link_id ||
+		    link_id >= IEEE80211_MLD_MAX_NUM_LINKS)
+			continue;
+
+		link = sdata_dereference(sdata->link[link_id], sdata);
+		if (!link)
+			continue;
+
+		len = cfg80211_defragment_element(sub, subelems, subelems_len,
+						  NULL, 0,
+						  IEEE80211_MLE_SUBELEM_FRAGMENT);
+		if (WARN_ON(len < 0))
+			continue;
+
+		sta_profiles[link_id] = sub;
+		sta_profiles_len[link_id] = len;
+	}
+
+	for (link_id = 0; link_id < IEEE80211_MLD_MAX_NUM_LINKS; link_id++) {
+		struct ieee80211_mle_per_sta_profile *prof;
+		struct ieee802_11_elems *prof_elems;
+		struct ieee80211_link_data *link;
+		ssize_t len;
+
+		if (link_id == reporting_link_id)
+			continue;
+
+		link = sdata_dereference(sdata->link[link_id], sdata);
+		if (!link)
+			continue;
+
+		if (!sta_profiles[link_id]) {
+			prof_elems = NULL;
+			goto handle;
+		}
+
+		/* we can defragment in-place, won't use the buffer again */
+		len = cfg80211_defragment_element(sta_profiles[link_id],
+						  subelems, subelems_len,
+						  (void *)sta_profiles[link_id],
+						  sta_profiles_len[link_id],
+						  IEEE80211_MLE_SUBELEM_FRAGMENT);
+		if (WARN_ON(len != sta_profiles_len[link_id]))
+			continue;
+
+		prof = (void *)sta_profiles[link_id];
+		prof_elems = ieee802_11_parse_elems(prof->variable +
+						    (prof->sta_info_len - 1),
+						    len -
+						    (prof->sta_info_len - 1),
+						    false, NULL);
+
+		/* memory allocation failed - let's hope that's transient */
+		if (!prof_elems)
+			continue;
+
+handle:
+		/*
+		 * FIXME: the timings here are obviously incorrect,
+		 * but only older Intel drivers seem to care, and
+		 * those don't have MLO. If you really need this,
+		 * the problem is having to calculate it with the
+		 * TSF offset etc. The device_timestamp is still
+		 * correct, of course.
+		 */
+		ieee80211_sta_process_chanswitch(link, 0, 0, elems, prof_elems,
+						 IEEE80211_CSA_SOURCE_OTHER_LINK);
+		kfree(prof_elems);
+	}
+}
+
 static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 				     struct ieee80211_hdr *hdr, size_t len,
 				     struct ieee80211_rx_status *rx_status)
@@ -6552,7 +6832,11 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 
 	ieee80211_sta_process_chanswitch(link, rx_status->mactime,
 					 rx_status->device_timestamp,
-					 elems, true);
+					 elems, elems,
+					 IEEE80211_CSA_SOURCE_BEACON);
+
+	/* note that after this elems->ml_basic can no longer be used fully */
+	ieee80211_mgd_check_cross_link_csa(sdata, rx_status->link_id, elems);
 
 	if (!link->u.mgd.disable_wmm_tracking &&
 	    ieee80211_sta_wmm_params(local, link, elems->wmm_param,
@@ -7148,7 +7432,8 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata,
 				ieee80211_sta_process_chanswitch(link,
 								 rx_status->mactime,
 								 rx_status->device_timestamp,
-								 elems, false);
+								 elems, elems,
+								 IEEE80211_CSA_SOURCE_ACTION);
 			kfree(elems);
 		} else if (mgmt->u.action.category == WLAN_CATEGORY_PUBLIC) {
 			struct ieee802_11_elems *elems;
@@ -7176,7 +7461,8 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata,
 				ieee80211_sta_process_chanswitch(link,
 								 rx_status->mactime,
 								 rx_status->device_timestamp,
-								 elems, false);
+								 elems, elems,
+								 IEEE80211_CSA_SOURCE_ACTION);
 			}
 
 			kfree(elems);
-- 
cgit v1.2.3-58-ga151


From 344d18cec23142b3921e068a90f58e8fc5f7637a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:54:49 +0200
Subject: wifi: mac80211: collect some CSA data into sub-structs

Collect the CSA data in ieee80211_link_data_managed and
ieee80211_link_data into a csa sub-struct to clean up a
bit and make adding new things more obvious.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506215543.29f954b1f576.I9a683a9647c33d4dd3011aade6677982428c1082@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c         | 16 +++++------
 net/mac80211/chan.c        |  4 +--
 net/mac80211/ibss.c        |  6 ++--
 net/mac80211/ieee80211_i.h | 18 +++++++-----
 net/mac80211/iface.c       |  4 +--
 net/mac80211/link.c        |  8 +++---
 net/mac80211/mlme.c        | 70 +++++++++++++++++++++++-----------------------
 7 files changed, 65 insertions(+), 61 deletions(-)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 0ed7e8bbf5c0..3319c9360507 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -3633,10 +3633,10 @@ void ieee80211_csa_finish(struct ieee80211_vif *vif, unsigned int link_id)
 				continue;
 
 			wiphy_work_queue(iter->local->hw.wiphy,
-					 &iter->deflink.csa_finalize_work);
+					 &iter->deflink.csa.finalize_work);
 		}
 	}
-	wiphy_work_queue(local->hw.wiphy, &link_data->csa_finalize_work);
+	wiphy_work_queue(local->hw.wiphy, &link_data->csa.finalize_work);
 
 	rcu_read_unlock();
 }
@@ -3723,7 +3723,7 @@ static int __ieee80211_csa_finalize(struct ieee80211_link_data *link_data)
 	}
 
 	if (!cfg80211_chandef_identical(&link_conf->chanreq.oper,
-					&link_data->csa_chanreq.oper))
+					&link_data->csa.chanreq.oper))
 		return -EINVAL;
 
 	link_conf->csa_active = false;
@@ -3744,7 +3744,7 @@ static int __ieee80211_csa_finalize(struct ieee80211_link_data *link_data)
 	if (err)
 		return err;
 
-	cfg80211_ch_switch_notify(sdata->dev, &link_data->csa_chanreq.oper,
+	cfg80211_ch_switch_notify(sdata->dev, &link_data->csa.chanreq.oper,
 				  link_data->link_id);
 
 	return 0;
@@ -3765,7 +3765,7 @@ static void ieee80211_csa_finalize(struct ieee80211_link_data *link_data)
 void ieee80211_csa_finalize_work(struct wiphy *wiphy, struct wiphy_work *work)
 {
 	struct ieee80211_link_data *link =
-		container_of(work, struct ieee80211_link_data, csa_finalize_work);
+		container_of(work, struct ieee80211_link_data, csa.finalize_work);
 	struct ieee80211_sub_if_data *sdata = link->sdata;
 	struct ieee80211_local *local = sdata->local;
 
@@ -4012,7 +4012,7 @@ __ieee80211_channel_switch(struct wiphy *wiphy, struct net_device *dev,
 		goto out;
 	}
 
-	link_data->csa_chanreq = chanreq; 
+	link_data->csa.chanreq = chanreq;
 	link_conf->csa_active = true;
 
 	if (params->block_tx &&
@@ -4023,12 +4023,12 @@ __ieee80211_channel_switch(struct wiphy *wiphy, struct net_device *dev,
 	}
 
 	cfg80211_ch_switch_started_notify(sdata->dev,
-					  &link_data->csa_chanreq.oper, 0,
+					  &link_data->csa.chanreq.oper, 0,
 					  params->count, params->block_tx);
 
 	if (changed) {
 		ieee80211_link_info_change_notify(sdata, link_data, changed);
-		drv_channel_switch_beacon(sdata, &link_data->csa_chanreq.oper);
+		drv_channel_switch_beacon(sdata, &link_data->csa.chanreq.oper);
 	} else {
 		/* if the beacon didn't change, we can finalize immediately */
 		ieee80211_csa_finalize(link_data);
diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c
index 380695fdc32f..ec16d7676088 100644
--- a/net/mac80211/chan.c
+++ b/net/mac80211/chan.c
@@ -1162,11 +1162,11 @@ ieee80211_link_chanctx_reservation_complete(struct ieee80211_link_data *link)
 	case NL80211_IFTYPE_MESH_POINT:
 	case NL80211_IFTYPE_OCB:
 		wiphy_work_queue(sdata->local->hw.wiphy,
-				 &link->csa_finalize_work);
+				 &link->csa.finalize_work);
 		break;
 	case NL80211_IFTYPE_STATION:
 		wiphy_delayed_work_queue(sdata->local->hw.wiphy,
-					 &link->u.mgd.chswitch_work, 0);
+					 &link->u.mgd.csa.switch_work, 0);
 		break;
 	case NL80211_IFTYPE_UNSPECIFIED:
 	case NL80211_IFTYPE_AP_VLAN:
diff --git a/net/mac80211/ibss.c b/net/mac80211/ibss.c
index 7ace5cdc6c26..bf338f3d4dd3 100644
--- a/net/mac80211/ibss.c
+++ b/net/mac80211/ibss.c
@@ -9,7 +9,7 @@
  * Copyright 2009, Johannes Berg <johannes@sipsolutions.net>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
  * Copyright(c) 2016 Intel Deutschland GmbH
- * Copyright(c) 2018-2023 Intel Corporation
+ * Copyright(c) 2018-2024 Intel Corporation
  */
 
 #include <linux/delay.h>
@@ -533,12 +533,12 @@ int ieee80211_ibss_finish_csa(struct ieee80211_sub_if_data *sdata, u64 *changed)
 					IEEE80211_PRIVACY(ifibss->privacy));
 		/* XXX: should not really modify cfg80211 data */
 		if (cbss) {
-			cbss->channel = sdata->deflink.csa_chanreq.oper.chan;
+			cbss->channel = sdata->deflink.csa.chanreq.oper.chan;
 			cfg80211_put_bss(sdata->local->hw.wiphy, cbss);
 		}
 	}
 
-	ifibss->chandef = sdata->deflink.csa_chanreq.oper;
+	ifibss->chandef = sdata->deflink.csa.chanreq.oper;
 
 	/* generate the beacon */
 	return ieee80211_ibss_csa_beacon(sdata, NULL, changed);
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 4ce34e95902e..cba1c2c3d9f1 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -975,11 +975,13 @@ struct ieee80211_link_data_managed {
 	bool disable_wmm_tracking;
 	bool operating_11g_mode;
 
-	bool csa_waiting_bcn;
-	bool csa_ignored_same_chan;
-	bool csa_blocked_tx;
-	unsigned long csa_time;
-	struct wiphy_delayed_work chswitch_work;
+	struct {
+		struct wiphy_delayed_work switch_work;
+		unsigned long time;
+		bool waiting_bcn;
+		bool ignored_same_chan;
+		bool blocked_tx;
+	} csa;
 
 	struct wiphy_work request_smps_work;
 	/* used to reconfigure hardware SM PS */
@@ -1038,11 +1040,13 @@ struct ieee80211_link_data {
 	struct ieee80211_key __rcu *default_mgmt_key;
 	struct ieee80211_key __rcu *default_beacon_key;
 
-	struct wiphy_work csa_finalize_work;
 
 	bool operating_11g_mode;
 
-	struct ieee80211_chan_req csa_chanreq;
+	struct {
+		struct wiphy_work finalize_work;
+		struct ieee80211_chan_req chanreq;
+	} csa;
 
 	struct wiphy_work color_change_finalize_work;
 	struct delayed_work color_collision_detect_work;
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 1ef4b6dc3730..d1a49ee4a194 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -543,14 +543,14 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, bool going_do
 
 	sdata->vif.bss_conf.csa_active = false;
 	if (sdata->vif.type == NL80211_IFTYPE_STATION)
-		sdata->deflink.u.mgd.csa_waiting_bcn = false;
+		sdata->deflink.u.mgd.csa.waiting_bcn = false;
 	if (sdata->csa_blocked_queues) {
 		ieee80211_wake_vif_queues(local, sdata,
 					  IEEE80211_QUEUE_STOP_REASON_CSA);
 		sdata->csa_blocked_queues = false;
 	}
 
-	wiphy_work_cancel(local->hw.wiphy, &sdata->deflink.csa_finalize_work);
+	wiphy_work_cancel(local->hw.wiphy, &sdata->deflink.csa.finalize_work);
 	wiphy_work_cancel(local->hw.wiphy,
 			  &sdata->deflink.color_change_finalize_work);
 	wiphy_delayed_work_cancel(local->hw.wiphy,
diff --git a/net/mac80211/link.c b/net/mac80211/link.c
index 508759cc9979..2e6e92defbca 100644
--- a/net/mac80211/link.c
+++ b/net/mac80211/link.c
@@ -37,7 +37,7 @@ void ieee80211_link_init(struct ieee80211_sub_if_data *sdata,
 	link_conf->link_id = link_id;
 	link_conf->vif = &sdata->vif;
 
-	wiphy_work_init(&link->csa_finalize_work,
+	wiphy_work_init(&link->csa.finalize_work,
 			ieee80211_csa_finalize_work);
 	wiphy_work_init(&link->color_change_finalize_work,
 			ieee80211_color_change_finalize_work);
@@ -72,7 +72,7 @@ void ieee80211_link_stop(struct ieee80211_link_data *link)
 
 	cancel_delayed_work_sync(&link->color_collision_detect_work);
 	wiphy_work_cancel(link->sdata->local->hw.wiphy,
-			  &link->csa_finalize_work);
+			  &link->csa.finalize_work);
 	ieee80211_link_release_channel(link);
 }
 
@@ -366,8 +366,8 @@ static int _ieee80211_set_active_links(struct ieee80211_sub_if_data *sdata,
 		 */
 		if (link->conf->csa_active)
 			wiphy_delayed_work_queue(local->hw.wiphy,
-						 &link->u.mgd.chswitch_work,
-						 link->u.mgd.csa_time -
+						 &link->u.mgd.csa.switch_work,
+						 link->u.mgd.csa.time -
 						 jiffies);
 	}
 
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index aedd1ce1fb8d..75d53257dd97 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2057,12 +2057,12 @@ void ieee80211_send_4addr_nullfunc(struct ieee80211_local *local,
 }
 
 /* spectrum management related things */
-static void ieee80211_chswitch_work(struct wiphy *wiphy,
-				    struct wiphy_work *work)
+static void ieee80211_csa_switch_work(struct wiphy *wiphy,
+				      struct wiphy_work *work)
 {
 	struct ieee80211_link_data *link =
 		container_of(work, struct ieee80211_link_data,
-			     u.mgd.chswitch_work.work);
+			     u.mgd.csa.switch_work.work);
 	struct ieee80211_sub_if_data *sdata = link->sdata;
 	struct ieee80211_local *local = sdata->local;
 	struct ieee80211_if_managed *ifmgd = &sdata->u.mgd;
@@ -2085,8 +2085,8 @@ static void ieee80211_chswitch_work(struct wiphy *wiphy,
 	 * update cfg80211 directly.
 	 */
 	if (!ieee80211_vif_link_active(&sdata->vif, link->link_id)) {
-		link->conf->chanreq = link->csa_chanreq;
-		cfg80211_ch_switch_notify(sdata->dev, &link->csa_chanreq.oper,
+		link->conf->chanreq = link->csa.chanreq;
+		cfg80211_ch_switch_notify(sdata->dev, &link->csa.chanreq.oper,
 					  link->link_id);
 		return;
 	}
@@ -2119,7 +2119,7 @@ static void ieee80211_chswitch_work(struct wiphy *wiphy,
 	}
 
 	if (!ieee80211_chanreq_identical(&link->conf->chanreq,
-					 &link->csa_chanreq)) {
+					 &link->csa.chanreq)) {
 		link_info(link,
 			  "failed to finalize channel switch, disconnecting\n");
 		wiphy_work_queue(sdata->local->hw.wiphy,
@@ -2127,7 +2127,7 @@ static void ieee80211_chswitch_work(struct wiphy *wiphy,
 		return;
 	}
 
-	link->u.mgd.csa_waiting_bcn = true;
+	link->u.mgd.csa.waiting_bcn = true;
 
 	ieee80211_sta_reset_beacon_monitor(sdata);
 	ieee80211_sta_reset_conn_monitor(sdata);
@@ -2151,8 +2151,8 @@ static void ieee80211_chswitch_post_beacon(struct ieee80211_link_data *link)
 	}
 
 	link->conf->csa_active = false;
-	link->u.mgd.csa_blocked_tx = false;
-	link->u.mgd.csa_waiting_bcn = false;
+	link->u.mgd.csa.blocked_tx = false;
+	link->u.mgd.csa.waiting_bcn = false;
 
 	ret = drv_post_channel_switch(link);
 	if (ret) {
@@ -2192,7 +2192,7 @@ void ieee80211_chswitch_done(struct ieee80211_vif *vif, bool success,
 		}
 
 		wiphy_delayed_work_queue(sdata->local->hw.wiphy,
-					 &link->u.mgd.chswitch_work, 0);
+					 &link->u.mgd.csa.switch_work, 0);
 	}
 
 	rcu_read_unlock();
@@ -2219,7 +2219,7 @@ ieee80211_sta_abort_chanswitch(struct ieee80211_link_data *link)
 	}
 
 	link->conf->csa_active = false;
-	link->u.mgd.csa_blocked_tx = false;
+	link->u.mgd.csa.blocked_tx = false;
 
 	drv_abort_channel_switch(link);
 }
@@ -2324,7 +2324,7 @@ ieee80211_sta_other_link_csa_disappeared(struct ieee80211_link_data *link,
 	 * puncturing configuration, but we don't have enough data to
 	 * detect that.
 	 */
-	if (data.chan != link->csa_chanreq.oper.chan)
+	if (data.chan != link->csa.chanreq.oper.chan)
 		ieee80211_sta_abort_chanswitch(link);
 }
 
@@ -2398,7 +2398,7 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 			/* already processing - disregard action frames */
 			return;
 		case IEEE80211_CSA_SOURCE_BEACON:
-			if (link->u.mgd.csa_waiting_bcn) {
+			if (link->u.mgd.csa.waiting_bcn) {
 				ieee80211_chswitch_post_beacon(link);
 				/*
 				 * If the CSA is still present after the switch
@@ -2421,7 +2421,7 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 				return;
 
 			/* switch work ran, so just complete the process */
-			if (link->u.mgd.csa_waiting_bcn) {
+			if (link->u.mgd.csa.waiting_bcn) {
 				ieee80211_chswitch_post_beacon(link);
 				/*
 				 * If the CSA is still present after the switch
@@ -2476,12 +2476,12 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 	if (cfg80211_chandef_identical(&csa_ie.chanreq.oper,
 				       &link->conf->chanreq.oper) &&
 	    (!csa_ie.mode || source != IEEE80211_CSA_SOURCE_BEACON)) {
-		if (link->u.mgd.csa_ignored_same_chan)
+		if (link->u.mgd.csa.ignored_same_chan)
 			return;
 		link_info(link,
 			  "AP %pM tries to chanswitch to same channel, ignore\n",
 			  link->u.mgd.bssid);
-		link->u.mgd.csa_ignored_same_chan = true;
+		link->u.mgd.csa.ignored_same_chan = true;
 		return;
 	}
 
@@ -2529,10 +2529,10 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 	}
 
 	link->conf->csa_active = true;
-	link->csa_chanreq = csa_ie.chanreq;
-	link->u.mgd.csa_ignored_same_chan = false;
+	link->csa.chanreq = csa_ie.chanreq;
+	link->u.mgd.csa.ignored_same_chan = false;
 	link->u.mgd.beacon_crc_valid = false;
-	link->u.mgd.csa_blocked_tx = csa_ie.mode;
+	link->u.mgd.csa.blocked_tx = csa_ie.mode;
 
 	if (csa_ie.mode &&
 	    !ieee80211_hw_check(&local->hw, HANDLES_QUIET_CSA)) {
@@ -2547,7 +2547,7 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 
 	/* we may have to handle timeout for deactivated link in software */
 	now = jiffies;
-	link->u.mgd.csa_time = now +
+	link->u.mgd.csa.time = now +
 			       TU_TO_JIFFIES((max_t(int, csa_ie.count, 1) - 1) *
 					     link->conf->beacon_int);
 
@@ -2557,7 +2557,7 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 		 * Use driver's channel switch callback, the driver will
 		 * later call ieee80211_chswitch_done(). It may deactivate
 		 * the link as well, we handle that elsewhere and queue
-		 * the chswitch_work for the calculated time then.
+		 * the csa.switch_work for the calculated time then.
 		 */
 		drv_channel_switch(local, sdata, &ch_switch);
 		return;
@@ -2565,8 +2565,8 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 
 	/* channel switch handled in software */
 	wiphy_delayed_work_queue(local->hw.wiphy,
-				 &link->u.mgd.chswitch_work,
-				 link->u.mgd.csa_time - now);
+				 &link->u.mgd.csa.switch_work,
+				 link->u.mgd.csa.time - now);
 	return;
  drop_connection:
 	/*
@@ -2577,7 +2577,7 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 	 * reset when the disconnection worker runs.
 	 */
 	link->conf->csa_active = true;
-	link->u.mgd.csa_blocked_tx = csa_ie.mode;
+	link->u.mgd.csa.blocked_tx = csa_ie.mode;
 	sdata->csa_blocked_queues =
 		csa_ie.mode && !ieee80211_hw_check(&local->hw, HANDLES_QUIET_CSA);
 
@@ -3630,9 +3630,9 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata,
 	}
 
 	sdata->vif.bss_conf.csa_active = false;
-	sdata->deflink.u.mgd.csa_blocked_tx = false;
-	sdata->deflink.u.mgd.csa_waiting_bcn = false;
-	sdata->deflink.u.mgd.csa_ignored_same_chan = false;
+	sdata->deflink.u.mgd.csa.blocked_tx = false;
+	sdata->deflink.u.mgd.csa.waiting_bcn = false;
+	sdata->deflink.u.mgd.csa.ignored_same_chan = false;
 	if (sdata->csa_blocked_queues) {
 		ieee80211_wake_vif_queues(local, sdata,
 					  IEEE80211_QUEUE_STOP_REASON_CSA);
@@ -3960,7 +3960,7 @@ static void __ieee80211_disconnect(struct ieee80211_sub_if_data *sdata)
 		if (WARN_ON_ONCE(!link))
 			continue;
 
-		if (link->u.mgd.csa_blocked_tx)
+		if (link->u.mgd.csa.blocked_tx)
 			continue;
 
 		tx = true;
@@ -3997,8 +3997,8 @@ static void __ieee80211_disconnect(struct ieee80211_sub_if_data *sdata)
 			       tx, frame_buf);
 	/* the other links will be destroyed */
 	sdata->vif.bss_conf.csa_active = false;
-	sdata->deflink.u.mgd.csa_waiting_bcn = false;
-	sdata->deflink.u.mgd.csa_blocked_tx = false;
+	sdata->deflink.u.mgd.csa.waiting_bcn = false;
+	sdata->deflink.u.mgd.csa.blocked_tx = false;
 	if (sdata->csa_blocked_queues) {
 		ieee80211_wake_vif_queues(local, sdata,
 					  IEEE80211_QUEUE_STOP_REASON_CSA);
@@ -7785,7 +7785,7 @@ static void ieee80211_sta_bcn_mon_timer(struct timer_list *t)
 		return;
 
 	if (sdata->vif.bss_conf.csa_active &&
-	    !sdata->deflink.u.mgd.csa_waiting_bcn)
+	    !sdata->deflink.u.mgd.csa.waiting_bcn)
 		return;
 
 	if (sdata->vif.driver_flags & IEEE80211_VIF_BEACON_FILTER)
@@ -7809,7 +7809,7 @@ static void ieee80211_sta_conn_mon_timer(struct timer_list *t)
 		return;
 
 	if (sdata->vif.bss_conf.csa_active &&
-	    !sdata->deflink.u.mgd.csa_waiting_bcn)
+	    !sdata->deflink.u.mgd.csa.waiting_bcn)
 		return;
 
 	sta = sta_info_get(sdata, sdata->vif.cfg.ap_addr);
@@ -8020,8 +8020,8 @@ void ieee80211_mgd_setup_link(struct ieee80211_link_data *link)
 	else
 		link->u.mgd.req_smps = IEEE80211_SMPS_OFF;
 
-	wiphy_delayed_work_init(&link->u.mgd.chswitch_work,
-				ieee80211_chswitch_work);
+	wiphy_delayed_work_init(&link->u.mgd.csa.switch_work,
+				ieee80211_csa_switch_work);
 
 	ieee80211_clear_tpe(&link->conf->tpe);
 
@@ -9152,7 +9152,7 @@ void ieee80211_mgd_stop_link(struct ieee80211_link_data *link)
 	wiphy_work_cancel(link->sdata->local->hw.wiphy,
 			  &link->u.mgd.recalc_smps);
 	wiphy_delayed_work_cancel(link->sdata->local->hw.wiphy,
-				  &link->u.mgd.chswitch_work);
+				  &link->u.mgd.csa.switch_work);
 }
 
 void ieee80211_mgd_stop(struct ieee80211_sub_if_data *sdata)
-- 
cgit v1.2.3-58-ga151


From f81747a9ad2e773ee72d881717e859a9f7a01c76 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:54:50 +0200
Subject: wifi: mac80211: handle wider bandwidth OFDMA during CSA

During channel switch, track the AP configuration in the
chanreq, so that wider bandwidth OFDMA is taken into
account correctly, since multiple channel contexts may
be needed due to sharing not being possible due to
wider bandwidth OFDMA.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506215543.b2c5a72dac1b.I69f65cb2e75d4a49a174b1aede68bf8ff0a3cab3@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c      | 8 ++++++--
 net/mac80211/spectmgmt.c | 5 ++++-
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 75d53257dd97..1d0b414619a4 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2517,8 +2517,13 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 		goto drop_connection;
 	}
 
+	link->csa.chanreq = csa_ie.chanreq;
+	if (link->u.mgd.conn.mode < IEEE80211_CONN_MODE_EHT ||
+	    sdata->vif.driver_flags & IEEE80211_VIF_IGNORE_OFDMA_WIDER_BW)
+		link->csa.chanreq.ap.chan = NULL;
+
 	if (chanctx) {
-		res = ieee80211_link_reserve_chanctx(link, &csa_ie.chanreq,
+		res = ieee80211_link_reserve_chanctx(link, &link->csa.chanreq,
 						     chanctx->mode, false);
 		if (res) {
 			link_info(link,
@@ -2529,7 +2534,6 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 	}
 
 	link->conf->csa_active = true;
-	link->csa.chanreq = csa_ie.chanreq;
 	link->u.mgd.csa.ignored_same_chan = false;
 	link->u.mgd.beacon_crc_valid = false;
 	link->u.mgd.csa.blocked_tx = csa_ie.mode;
diff --git a/net/mac80211/spectmgmt.c b/net/mac80211/spectmgmt.c
index b2de4c6fb808..df96d3db1c0e 100644
--- a/net/mac80211/spectmgmt.c
+++ b/net/mac80211/spectmgmt.c
@@ -9,7 +9,7 @@
  * Copyright 2007, Michael Wu <flamingice@sourmilk.net>
  * Copyright 2007-2008, Intel Corporation
  * Copyright 2008, Johannes Berg <johannes@sipsolutions.net>
- * Copyright (C) 2018, 2020, 2022-2023 Intel Corporation
+ * Copyright (C) 2018, 2020, 2022-2024 Intel Corporation
  */
 
 #include <linux/ieee80211.h>
@@ -366,6 +366,9 @@ int ieee80211_parse_ch_switch_ie(struct ieee80211_sub_if_data *sdata,
 		validate_chandef_by_ht_vht_oper(sdata, conn, vht_cap_info,
 						&new_chandef);
 
+	/* capture the AP chandef before (potential) downgrading */
+	csa_ie->chanreq.ap = new_chandef;
+
 	/* if data is there validate the bandwidth & use it */
 	if (new_chandef.chan) {
 		if (conn->bw_limit < IEEE80211_CONN_BW_LIMIT_320 &&
-- 
cgit v1.2.3-58-ga151


From 4540568136fefe0fcc366ded125cc56b0d9e1724 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:54:51 +0200
Subject: wifi: mac80211: handle TPE element during CSA

Handle the transmit power envelope (TPE) element during
channel switch, applying it when the channel switch is
done.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506215543.486c33157d18.Idf971ad801b6961c177bdf42cc323fd1a4ca8165@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/ieee80211_i.h |  3 +++
 net/mac80211/mlme.c        | 18 ++++++++++++++++++
 net/mac80211/parse.c       |  8 ++++++++
 3 files changed, 29 insertions(+)

diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index cba1c2c3d9f1..76965d64a0fa 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -977,6 +977,8 @@ struct ieee80211_link_data_managed {
 
 	struct {
 		struct wiphy_delayed_work switch_work;
+		struct cfg80211_chan_def ap_chandef;
+		struct ieee80211_parsed_tpe tpe;
 		unsigned long time;
 		bool waiting_bcn;
 		bool ignored_same_chan;
@@ -1754,6 +1756,7 @@ struct ieee802_11_elems {
 
 	/* not the order in the psd values is per element, not per chandef */
 	struct ieee80211_parsed_tpe tpe;
+	struct ieee80211_parsed_tpe csa_tpe;
 
 	/* length of them, respectively */
 	u8 ext_capab_len;
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 1d0b414619a4..18793f2fb91c 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2129,6 +2129,20 @@ static void ieee80211_csa_switch_work(struct wiphy *wiphy,
 
 	link->u.mgd.csa.waiting_bcn = true;
 
+	/* apply new TPE restrictions immediately on the new channel */
+	if (link->u.mgd.csa.ap_chandef.chan->band == NL80211_BAND_6GHZ &&
+	    link->u.mgd.conn.mode >= IEEE80211_CONN_MODE_HE) {
+		ieee80211_rearrange_tpe(&link->u.mgd.csa.tpe,
+					&link->u.mgd.csa.ap_chandef,
+					&link->conf->chanreq.oper);
+		if (memcmp(&link->conf->tpe, &link->u.mgd.csa.tpe,
+			   sizeof(link->u.mgd.csa.tpe))) {
+			link->conf->tpe = link->u.mgd.csa.tpe;
+			ieee80211_link_info_change_notify(sdata, link,
+							  BSS_CHANGED_TPE);
+		}
+	}
+
 	ieee80211_sta_reset_beacon_monitor(sdata);
 	ieee80211_sta_reset_conn_monitor(sdata);
 }
@@ -2379,6 +2393,8 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 			ch_switch.count = csa_ie.count;
 			ch_switch.delay = csa_ie.max_switch_time;
 		}
+
+		link->u.mgd.csa.tpe = csa_elems->csa_tpe;
 	} else {
 		/*
 		 * If there was no per-STA profile for this link, we
@@ -2517,6 +2533,8 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 		goto drop_connection;
 	}
 
+	link->u.mgd.csa.ap_chandef = csa_ie.chanreq.ap;
+
 	link->csa.chanreq = csa_ie.chanreq;
 	if (link->u.mgd.conn.mode < IEEE80211_CONN_MODE_EHT ||
 	    sdata->vif.driver_flags & IEEE80211_VIF_IGNORE_OFDMA_WIDER_BW)
diff --git a/net/mac80211/parse.c b/net/mac80211/parse.c
index 6efeb977f8e5..5c5c21ecb2b7 100644
--- a/net/mac80211/parse.c
+++ b/net/mac80211/parse.c
@@ -607,6 +607,13 @@ _ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params,
 					elem_parse_failed =
 						IEEE80211_PARSE_ERR_BAD_ELEM_SIZE;
 			}
+
+			subelem = cfg80211_find_ext_elem(WLAN_EID_TX_POWER_ENVELOPE,
+							 pos, elen);
+			if (subelem)
+				ieee80211_parse_tpe(&elems->csa_tpe,
+						    subelem->data + 1,
+						    subelem->datalen - 1);
 			break;
 		case WLAN_EID_COUNTRY:
 			elems->country_elem = pos;
@@ -962,6 +969,7 @@ ieee802_11_parse_elems_full(struct ieee80211_elems_parse_params *params)
 
 	/* set all TPE entries to unlimited (but invalid) */
 	ieee80211_clear_tpe(&elems->tpe);
+	ieee80211_clear_tpe(&elems->csa_tpe);
 
 	nontransmitted_profile = elems_parse->scratch_pos;
 	nontransmitted_profile_len =
-- 
cgit v1.2.3-58-ga151


From 9f472520f6f7b8ed3cf82d800b75c91c4c3d4ef1 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:54:52 +0200
Subject: wifi: mac80211: refactor chanreq.ap setting

There are now three places setting up chanreq.ap which always
depends on the mode (EHT being used or not) and override flag.
Refactor that code into a common function with a comment, to
make that clearer.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506215543.5cd6a209e58a.I3be318959d9e2df5dccd2d0938c3d2fcc6688030@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 43 +++++++++++++++++++++++++++++--------------
 1 file changed, 29 insertions(+), 14 deletions(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 18793f2fb91c..a974a75df266 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -789,6 +789,29 @@ static void ieee80211_rearrange_tpe(struct ieee80211_parsed_tpe *tpe,
 	}
 }
 
+/*
+ * The AP part of the channel request is used to distinguish settings
+ * to the device used for wider bandwidth OFDMA. This is used in the
+ * channel context code to assign two channel contexts even if they're
+ * both for the same channel, if the AP bandwidths are incompatible.
+ * If not EHT (or driver override) then ap.chan == NULL indicates that
+ * there's no wider BW OFDMA used.
+ */
+static void ieee80211_set_chanreq_ap(struct ieee80211_sub_if_data *sdata,
+				     struct ieee80211_chan_req *chanreq,
+				     struct ieee80211_conn_settings *conn,
+				     struct cfg80211_chan_def *ap_chandef)
+{
+	chanreq->ap.chan = NULL;
+
+	if (conn->mode < IEEE80211_CONN_MODE_EHT)
+		return;
+	if (sdata->vif.driver_flags & IEEE80211_VIF_IGNORE_OFDMA_WIDER_BW)
+		return;
+
+	chanreq->ap = *ap_chandef;
+}
+
 static struct ieee802_11_elems *
 ieee80211_determine_chan_mode(struct ieee80211_sub_if_data *sdata,
 			      struct ieee80211_conn_settings *conn,
@@ -886,12 +909,7 @@ again:
 
 	chanreq->oper = *ap_chandef;
 
-	/* wider-bandwidth OFDMA is only done in EHT */
-	if (conn->mode >= IEEE80211_CONN_MODE_EHT &&
-	    !(sdata->vif.driver_flags & IEEE80211_VIF_IGNORE_OFDMA_WIDER_BW))
-		chanreq->ap = *ap_chandef;
-	else
-		chanreq->ap.chan = NULL;
+	ieee80211_set_chanreq_ap(sdata, chanreq, conn, ap_chandef);
 
 	while (!ieee80211_chandef_usable(sdata, &chanreq->oper,
 					 IEEE80211_CHAN_DISABLED)) {
@@ -999,11 +1017,9 @@ static int ieee80211_config_bw(struct ieee80211_link_data *link,
 		return -EINVAL;
 	}
 
-	chanreq.ap = ap_chandef;
 	chanreq.oper = ap_chandef;
-	if (link->u.mgd.conn.mode < IEEE80211_CONN_MODE_EHT ||
-	    sdata->vif.driver_flags & IEEE80211_VIF_IGNORE_OFDMA_WIDER_BW)
-		chanreq.ap.chan = NULL;
+	ieee80211_set_chanreq_ap(sdata, &chanreq, &link->u.mgd.conn,
+				 &ap_chandef);
 
 	/*
 	 * if HT operation mode changed store the new one -
@@ -2535,10 +2551,9 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 
 	link->u.mgd.csa.ap_chandef = csa_ie.chanreq.ap;
 
-	link->csa.chanreq = csa_ie.chanreq;
-	if (link->u.mgd.conn.mode < IEEE80211_CONN_MODE_EHT ||
-	    sdata->vif.driver_flags & IEEE80211_VIF_IGNORE_OFDMA_WIDER_BW)
-		link->csa.chanreq.ap.chan = NULL;
+	link->csa.chanreq.oper = csa_ie.chanreq.oper;
+	ieee80211_set_chanreq_ap(sdata, &link->csa.chanreq, &link->u.mgd.conn,
+				 &csa_ie.chanreq.ap);
 
 	if (chanctx) {
 		res = ieee80211_link_reserve_chanctx(link, &link->csa.chanreq,
-- 
cgit v1.2.3-58-ga151


From f9a0757a4b2f5df4376963c25a3d7d7aeba78444 Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Date: Wed, 15 May 2024 11:13:18 -0700
Subject: wifi: mac80211: Add EHT UL MU-MIMO flag in ieee80211_bss_conf

Add flag for Full Bandwidth UL MU-MIMO for EHT. This is utilized
to pass EHT MU-MIMO configurations from user space to driver
in AP mode.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Link: https://msgid.link/20240515181327.12855-2-quic_pradeepc@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h | 4 ++++
 net/mac80211/cfg.c     | 5 +++++
 2 files changed, 9 insertions(+)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index a59eacfe0480..ecfa65ade226 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -735,6 +735,9 @@ struct ieee80211_parsed_tpe {
  *	beamformee
  * @eht_mu_beamformer: in AP-mode, does this BSS enable operation as an EHT MU
  *	beamformer
+ * @eht_80mhz_full_bw_ul_mumimo: in AP-mode, does this BSS support the
+ *	reception of an EHT TB PPDU on an RU that spans the entire PPDU
+ *	bandwidth
  */
 struct ieee80211_bss_conf {
 	struct ieee80211_vif *vif;
@@ -828,6 +831,7 @@ struct ieee80211_bss_conf {
 	bool eht_su_beamformer;
 	bool eht_su_beamformee;
 	bool eht_mu_beamformer;
+	bool eht_80mhz_full_bw_ul_mumimo;
 };
 
 /**
diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 3319c9360507..62119e957cd8 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1379,6 +1379,11 @@ static int ieee80211_start_ap(struct wiphy *wiphy, struct net_device *dev,
 				(IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_80MHZ |
 				 IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_160MHZ |
 				 IEEE80211_EHT_PHY_CAP7_MU_BEAMFORMER_320MHZ);
+		link_conf->eht_80mhz_full_bw_ul_mumimo =
+			params->eht_cap->fixed.phy_cap_info[7] &
+				(IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_80MHZ |
+				 IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_160MHZ |
+				 IEEE80211_EHT_PHY_CAP7_NON_OFDMA_UL_MU_MIMO_320MHZ);
 	} else {
 		link_conf->eht_su_beamformer = false;
 		link_conf->eht_su_beamformee = false;
-- 
cgit v1.2.3-58-ga151


From aa4ec06c455d0200eea0a4361cc58eb5b8bf68c4 Mon Sep 17 00:00:00 2001
From: Dmitry Antipov <dmantipov@yandex.ru>
Date: Fri, 17 May 2024 18:33:31 +0300
Subject: wifi: cfg80211: use __counted_by where appropriate

Annotate 'sub_specs' of 'struct cfg80211_sar_specs', 'channels'
of 'struct cfg80211_sched_scan_request', 'channels' of 'struct
cfg80211_wowlan_nd_match', and 'matches' of 'struct
cfg80211_wowlan_nd_info' with '__counted_by' attribute. Briefly
tested with clang 18.1.1 and CONFIG_UBSAN_BOUNDS running iwlwifi.

Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
Link: https://msgid.link/20240517153332.18271-1-dmantipov@yandex.ru
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index cbf1664dc569..d79180bec7a1 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -2200,7 +2200,7 @@ struct cfg80211_sar_sub_specs {
 struct cfg80211_sar_specs {
 	enum nl80211_sar_type type;
 	u32 num_sub_specs;
-	struct cfg80211_sar_sub_specs sub_specs[];
+	struct cfg80211_sar_sub_specs sub_specs[] __counted_by(num_sub_specs);
 };
 
 
@@ -2838,7 +2838,7 @@ struct cfg80211_sched_scan_request {
 	struct list_head list;
 
 	/* keep last */
-	struct ieee80211_channel *channels[];
+	struct ieee80211_channel *channels[] __counted_by(n_channels);
 };
 
 /**
@@ -3582,7 +3582,7 @@ struct cfg80211_coalesce {
 struct cfg80211_wowlan_nd_match {
 	struct cfg80211_ssid ssid;
 	int n_channels;
-	u32 channels[];
+	u32 channels[] __counted_by(n_channels);
 };
 
 /**
@@ -3596,7 +3596,7 @@ struct cfg80211_wowlan_nd_match {
  */
 struct cfg80211_wowlan_nd_info {
 	int n_matches;
-	struct cfg80211_wowlan_nd_match *matches[];
+	struct cfg80211_wowlan_nd_match *matches[] __counted_by(n_matches);
 };
 
 /**
-- 
cgit v1.2.3-58-ga151


From 2fe0a605d083b884490ee4de02be071b5b4291b1 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:18:52 +0200
Subject: wifi: mac80211: fix TTLM teardown work

The worker calculates the wrong sdata pointer, so if it ever
runs, it'll crash. Fix that.

Fixes: a17a58ad2ff2 ("wifi: mac80211: add support for tearing down negotiated TTLM")
Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506211853.e6471800c76d.I8b7c2d6984c89a11cd33d1a610e9645fa965f6e1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index a974a75df266..a29c4a4deb0b 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -7333,7 +7333,7 @@ static void ieee80211_teardown_ttlm_work(struct wiphy *wiphy,
 	u16 new_dormant_links;
 	struct ieee80211_sub_if_data *sdata =
 		container_of(work, struct ieee80211_sub_if_data,
-			     u.mgd.neg_ttlm_timeout_work.work);
+			     u.mgd.teardown_ttlm_work);
 
 	if (!sdata->vif.neg_ttlm.valid)
 		return;
-- 
cgit v1.2.3-58-ga151


From 53b739fd46462dc40fd18390d76f2ee05c18ea3a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:10:33 +0200
Subject: wifi: mac80211: cancel multi-link reconf work on disconnect

This work shouldn't run after we're disconnecting. Cancel it earlier
(and then don't cancel it in stop later.)

Fixes: 8eb8dd2ffbbb ("wifi: mac80211: Support link removal using Reconfiguration ML element")
Reviewed-by: Ilan Peer <ilan.peer@intel.com>
Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506211034.ac754794279f.Ib9fbb1dab50c6b67f6de9be09a6c452ce89bbd50@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index a29c4a4deb0b..ab6b379d522b 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -3694,6 +3694,11 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata,
 
 	wiphy_delayed_work_cancel(sdata->local->hw.wiphy,
 				  &ifmgd->neg_ttlm_timeout_work);
+
+	sdata->u.mgd.removed_links = 0;
+	wiphy_delayed_work_cancel(sdata->local->hw.wiphy,
+				  &sdata->u.mgd.ml_reconf_work);
+
 	ieee80211_vif_set_links(sdata, 0, 0);
 
 	ifmgd->mcast_seq_last = IEEE80211_SN_MODULO;
@@ -9211,8 +9216,6 @@ void ieee80211_mgd_stop(struct ieee80211_sub_if_data *sdata)
 			  &ifmgd->teardown_ttlm_work);
 	wiphy_delayed_work_cancel(sdata->local->hw.wiphy,
 				  &ifmgd->tdls_peer_del_work);
-	wiphy_delayed_work_cancel(sdata->local->hw.wiphy,
-				  &ifmgd->ml_reconf_work);
 	wiphy_delayed_work_cancel(sdata->local->hw.wiphy, &ifmgd->ttlm_work);
 	wiphy_delayed_work_cancel(sdata->local->hw.wiphy,
 				  &ifmgd->neg_ttlm_timeout_work);
-- 
cgit v1.2.3-58-ga151


From 3567bd6dcd1467d2ad0f597be94114c6f9c62680 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:10:34 +0200
Subject: wifi: mac80211: cancel TTLM teardown work earlier

It shouldn't be possible to run this after disconnecting, so
cancel the work earlier.

Fixes: a17a58ad2ff2 ("wifi: mac80211: add support for tearing down negotiated TTLM")
Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506211034.096a10ccebec.I5584a21c27eb9b3e87b9e26380b627114b32ccba@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index ab6b379d522b..e0b44f47747b 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -3699,6 +3699,9 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata,
 	wiphy_delayed_work_cancel(sdata->local->hw.wiphy,
 				  &sdata->u.mgd.ml_reconf_work);
 
+	wiphy_work_cancel(sdata->local->hw.wiphy,
+			  &ifmgd->teardown_ttlm_work);
+
 	ieee80211_vif_set_links(sdata, 0, 0);
 
 	ifmgd->mcast_seq_last = IEEE80211_SN_MODULO;
@@ -9212,8 +9215,6 @@ void ieee80211_mgd_stop(struct ieee80211_sub_if_data *sdata)
 			  &ifmgd->beacon_connection_loss_work);
 	wiphy_work_cancel(sdata->local->hw.wiphy,
 			  &ifmgd->csa_connection_drop_work);
-	wiphy_work_cancel(sdata->local->hw.wiphy,
-			  &ifmgd->teardown_ttlm_work);
 	wiphy_delayed_work_cancel(sdata->local->hw.wiphy,
 				  &ifmgd->tdls_peer_del_work);
 	wiphy_delayed_work_cancel(sdata->local->hw.wiphy, &ifmgd->ttlm_work);
-- 
cgit v1.2.3-58-ga151


From 0d22026f3287ef5af80d983db150ff260e7117e8 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:10:35 +0200
Subject: wifi: mac80211: don't stop TTLM works again

There's no need to stop works that have already been
stopped during disconnect, so don't.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506211034.f8434be19f56.I021afadc538508da3bc8f95c89f424ca62b94bef@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index e0b44f47747b..09451a6f74b8 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -9217,9 +9217,6 @@ void ieee80211_mgd_stop(struct ieee80211_sub_if_data *sdata)
 			  &ifmgd->csa_connection_drop_work);
 	wiphy_delayed_work_cancel(sdata->local->hw.wiphy,
 				  &ifmgd->tdls_peer_del_work);
-	wiphy_delayed_work_cancel(sdata->local->hw.wiphy, &ifmgd->ttlm_work);
-	wiphy_delayed_work_cancel(sdata->local->hw.wiphy,
-				  &ifmgd->neg_ttlm_timeout_work);
 
 	if (ifmgd->assoc_data)
 		ieee80211_destroy_assoc_data(sdata, ASSOC_TIMEOUT);
-- 
cgit v1.2.3-58-ga151


From 609c12a2af046c3674af2d5c7978b025718de5e8 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 6 May 2024 21:18:59 +0200
Subject: wifi: mac80211: reset negotiated TTLM on disconnect

The negotiated TTLM data must be reset on disconnect, otherwise
it may end up getting reused on another connection. Fix that.

Fixes: 8f500fbc6c65 ("wifi: mac80211: process and save negotiated TID to Link mapping request")
Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240506211858.04142e8fe01c.Ia144457e086ebd8ddcfa31bdf5ff210b4b351c22@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 09451a6f74b8..ef3280fafbe9 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -3692,6 +3692,7 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata,
 	       sizeof(sdata->u.mgd.ttlm_info));
 	wiphy_delayed_work_cancel(sdata->local->hw.wiphy, &ifmgd->ttlm_work);
 
+	memset(&sdata->vif.neg_ttlm, 0, sizeof(sdata->vif.neg_ttlm));
 	wiphy_delayed_work_cancel(sdata->local->hw.wiphy,
 				  &ifmgd->neg_ttlm_timeout_work);
 
-- 
cgit v1.2.3-58-ga151


From a92fd2d9327ba877f29753eec15b93072ac300b0 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 10 May 2024 11:26:02 +0200
Subject: wifi: mac80211: send DelBA with correct BSSID

In MLO, the deflink BSSID is clearly invalid. Since we fill
the addresses as MLD addresses and translate later, use the
AP address here instead.

This fixes an issue that happens with HW restart, where the
DelBA frame is transmitted, but not processed correctly due
to the wrong BSSID (or even just discarded entirely).  As a
result, the BA sessions are kept alive; however, as other
state is reset during HW restart, this then fails (reorder,
etc.) and data doesn't go through until new BA sessions are
established.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240510112601.f4e1effdea29.I98e81f22166b68d4b6211191bcaaf8531b324a77@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/ht.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mac80211/ht.c b/net/mac80211/ht.c
index d7e8cf8e48b7..79caeb485fd5 100644
--- a/net/mac80211/ht.c
+++ b/net/mac80211/ht.c
@@ -475,7 +475,7 @@ void ieee80211_send_delba(struct ieee80211_sub_if_data *sdata,
 	    sdata->vif.type == NL80211_IFTYPE_MESH_POINT)
 		memcpy(mgmt->bssid, sdata->vif.addr, ETH_ALEN);
 	else if (sdata->vif.type == NL80211_IFTYPE_STATION)
-		memcpy(mgmt->bssid, sdata->deflink.u.mgd.bssid, ETH_ALEN);
+		memcpy(mgmt->bssid, sdata->vif.cfg.ap_addr, ETH_ALEN);
 	else if (sdata->vif.type == NL80211_IFTYPE_ADHOC)
 		memcpy(mgmt->bssid, sdata->u.ibss.bssid, ETH_ALEN);
 
-- 
cgit v1.2.3-58-ga151


From 4d25ca2d6801cfcf26f7f39c561611ba5be99bf8 Mon Sep 17 00:00:00 2001
From: Abhishek Chauhan <quic_abchauha@quicinc.com>
Date: Thu, 9 May 2024 14:18:32 -0700
Subject: net: Rename mono_delivery_time to tstamp_type for scalabilty

mono_delivery_time was added to check if skb->tstamp has delivery
time in mono clock base (i.e. EDT) otherwise skb->tstamp has
timestamp in ingress and delivery_time at egress.

Renaming the bitfield from mono_delivery_time to tstamp_type is for
extensibilty for other timestamps such as userspace timestamp
(i.e. SO_TXTIME) set via sock opts.

As we are renaming the mono_delivery_time to tstamp_type, it makes
sense to start assigning tstamp_type based on enum defined
in this commit.

Earlier we used bool arg flag to check if the tstamp is mono in
function skb_set_delivery_time, Now the signature of the functions
accepts tstamp_type to distinguish between mono and real time.

Also skb_set_delivery_type_by_clockid is a new function which accepts
clockid to determine the tstamp_type.

In future tstamp_type:1 can be extended to support userspace timestamp
by increasing the bitfield.

Signed-off-by: Abhishek Chauhan <quic_abchauha@quicinc.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20240509211834.3235191-2-quic_abchauha@quicinc.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/skbuff.h                     | 52 ++++++++++++++++++++++--------
 include/net/inet_frag.h                    |  4 +--
 net/bridge/netfilter/nf_conntrack_bridge.c |  6 ++--
 net/core/dev.c                             |  2 +-
 net/core/filter.c                          | 10 +++---
 net/ieee802154/6lowpan/reassembly.c        |  2 +-
 net/ipv4/inet_fragment.c                   |  2 +-
 net/ipv4/ip_fragment.c                     |  2 +-
 net/ipv4/ip_output.c                       |  9 +++---
 net/ipv4/tcp_output.c                      | 14 ++++----
 net/ipv6/ip6_output.c                      |  6 ++--
 net/ipv6/netfilter.c                       |  6 ++--
 net/ipv6/netfilter/nf_conntrack_reasm.c    |  2 +-
 net/ipv6/reassembly.c                      |  2 +-
 net/ipv6/tcp_ipv6.c                        |  2 +-
 net/sched/act_bpf.c                        |  4 +--
 net/sched/cls_bpf.c                        |  4 +--
 17 files changed, 78 insertions(+), 51 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index c0b97c93a6de..3a721cc3b644 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -706,6 +706,11 @@ typedef unsigned int sk_buff_data_t;
 typedef unsigned char *sk_buff_data_t;
 #endif
 
+enum skb_tstamp_type {
+	SKB_CLOCK_REALTIME,
+	SKB_CLOCK_MONOTONIC,
+};
+
 /**
  * DOC: Basic sk_buff geometry
  *
@@ -823,10 +828,8 @@ typedef unsigned char *sk_buff_data_t;
  *	@dst_pending_confirm: need to confirm neighbour
  *	@decrypted: Decrypted SKB
  *	@slow_gro: state present at GRO time, slower prepare step required
- *	@mono_delivery_time: When set, skb->tstamp has the
- *		delivery_time in mono clock base (i.e. EDT).  Otherwise, the
- *		skb->tstamp has the (rcv) timestamp at ingress and
- *		delivery_time at egress.
+ *	@tstamp_type: When set, skb->tstamp has the
+ *		delivery_time clock base of skb->tstamp.
  *	@napi_id: id of the NAPI struct this skb came from
  *	@sender_cpu: (aka @napi_id) source CPU in XPS
  *	@alloc_cpu: CPU which did the skb allocation.
@@ -954,7 +957,7 @@ struct sk_buff {
 	/* private: */
 	__u8			__mono_tc_offset[0];
 	/* public: */
-	__u8			mono_delivery_time:1;	/* See SKB_MONO_DELIVERY_TIME_MASK */
+	__u8			tstamp_type:1;	/* See skb_tstamp_type */
 #ifdef CONFIG_NET_XGRESS
 	__u8			tc_at_ingress:1;	/* See TC_AT_INGRESS_MASK */
 	__u8			tc_skip_classify:1;
@@ -4183,7 +4186,7 @@ static inline void skb_get_new_timestampns(const struct sk_buff *skb,
 static inline void __net_timestamp(struct sk_buff *skb)
 {
 	skb->tstamp = ktime_get_real();
-	skb->mono_delivery_time = 0;
+	skb->tstamp_type = SKB_CLOCK_REALTIME;
 }
 
 static inline ktime_t net_timedelta(ktime_t t)
@@ -4192,10 +4195,33 @@ static inline ktime_t net_timedelta(ktime_t t)
 }
 
 static inline void skb_set_delivery_time(struct sk_buff *skb, ktime_t kt,
-					 bool mono)
+					 u8 tstamp_type)
 {
 	skb->tstamp = kt;
-	skb->mono_delivery_time = kt && mono;
+
+	if (kt)
+		skb->tstamp_type = tstamp_type;
+	else
+		skb->tstamp_type = SKB_CLOCK_REALTIME;
+}
+
+static inline void skb_set_delivery_type_by_clockid(struct sk_buff *skb,
+						    ktime_t kt, clockid_t clockid)
+{
+	u8 tstamp_type = SKB_CLOCK_REALTIME;
+
+	switch (clockid) {
+	case CLOCK_REALTIME:
+		break;
+	case CLOCK_MONOTONIC:
+		tstamp_type = SKB_CLOCK_MONOTONIC;
+		break;
+	default:
+		WARN_ON_ONCE(1);
+		kt = 0;
+	}
+
+	skb_set_delivery_time(skb, kt, tstamp_type);
 }
 
 DECLARE_STATIC_KEY_FALSE(netstamp_needed_key);
@@ -4205,8 +4231,8 @@ DECLARE_STATIC_KEY_FALSE(netstamp_needed_key);
  */
 static inline void skb_clear_delivery_time(struct sk_buff *skb)
 {
-	if (skb->mono_delivery_time) {
-		skb->mono_delivery_time = 0;
+	if (skb->tstamp_type) {
+		skb->tstamp_type = SKB_CLOCK_REALTIME;
 		if (static_branch_unlikely(&netstamp_needed_key))
 			skb->tstamp = ktime_get_real();
 		else
@@ -4216,7 +4242,7 @@ static inline void skb_clear_delivery_time(struct sk_buff *skb)
 
 static inline void skb_clear_tstamp(struct sk_buff *skb)
 {
-	if (skb->mono_delivery_time)
+	if (skb->tstamp_type)
 		return;
 
 	skb->tstamp = 0;
@@ -4224,7 +4250,7 @@ static inline void skb_clear_tstamp(struct sk_buff *skb)
 
 static inline ktime_t skb_tstamp(const struct sk_buff *skb)
 {
-	if (skb->mono_delivery_time)
+	if (skb->tstamp_type)
 		return 0;
 
 	return skb->tstamp;
@@ -4232,7 +4258,7 @@ static inline ktime_t skb_tstamp(const struct sk_buff *skb)
 
 static inline ktime_t skb_tstamp_cond(const struct sk_buff *skb, bool cond)
 {
-	if (!skb->mono_delivery_time && skb->tstamp)
+	if (skb->tstamp_type != SKB_CLOCK_MONOTONIC && skb->tstamp)
 		return skb->tstamp;
 
 	if (static_branch_unlikely(&netstamp_needed_key) || cond)
diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h
index 153960663ce4..5af6eb14c5db 100644
--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -76,7 +76,7 @@ struct frag_v6_compare_key {
  * @stamp: timestamp of the last received fragment
  * @len: total length of the original datagram
  * @meat: length of received fragments so far
- * @mono_delivery_time: stamp has a mono delivery time (EDT)
+ * @tstamp_type: stamp has a mono delivery time (EDT)
  * @flags: fragment queue flags
  * @max_size: maximum received fragment size
  * @fqdir: pointer to struct fqdir
@@ -97,7 +97,7 @@ struct inet_frag_queue {
 	ktime_t			stamp;
 	int			len;
 	int			meat;
-	u8			mono_delivery_time;
+	u8			tstamp_type;
 	__u8			flags;
 	u16			max_size;
 	struct fqdir		*fqdir;
diff --git a/net/bridge/netfilter/nf_conntrack_bridge.c b/net/bridge/netfilter/nf_conntrack_bridge.c
index c3c51b9a6826..816bb0fde718 100644
--- a/net/bridge/netfilter/nf_conntrack_bridge.c
+++ b/net/bridge/netfilter/nf_conntrack_bridge.c
@@ -32,7 +32,7 @@ static int nf_br_ip_fragment(struct net *net, struct sock *sk,
 					   struct sk_buff *))
 {
 	int frag_max_size = BR_INPUT_SKB_CB(skb)->frag_max_size;
-	bool mono_delivery_time = skb->mono_delivery_time;
+	u8 tstamp_type = skb->tstamp_type;
 	unsigned int hlen, ll_rs, mtu;
 	ktime_t tstamp = skb->tstamp;
 	struct ip_frag_state state;
@@ -82,7 +82,7 @@ static int nf_br_ip_fragment(struct net *net, struct sock *sk,
 			if (iter.frag)
 				ip_fraglist_prepare(skb, &iter);
 
-			skb_set_delivery_time(skb, tstamp, mono_delivery_time);
+			skb_set_delivery_time(skb, tstamp, tstamp_type);
 			err = output(net, sk, data, skb);
 			if (err || !iter.frag)
 				break;
@@ -113,7 +113,7 @@ slow_path:
 			goto blackhole;
 		}
 
-		skb_set_delivery_time(skb2, tstamp, mono_delivery_time);
+		skb_set_delivery_time(skb2, tstamp, tstamp_type);
 		err = output(net, sk, data, skb2);
 		if (err)
 			goto blackhole;
diff --git a/net/core/dev.c b/net/core/dev.c
index e1bb6d7856d9..85fe8138f3e4 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2160,7 +2160,7 @@ EXPORT_SYMBOL(net_disable_timestamp);
 static inline void net_timestamp_set(struct sk_buff *skb)
 {
 	skb->tstamp = 0;
-	skb->mono_delivery_time = 0;
+	skb->tstamp_type = SKB_CLOCK_REALTIME;
 	if (static_branch_unlikely(&netstamp_needed_key))
 		skb->tstamp = ktime_get_real();
 }
diff --git a/net/core/filter.c b/net/core/filter.c
index 2510464692af..a3781a796da4 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -7730,13 +7730,13 @@ BPF_CALL_3(bpf_skb_set_tstamp, struct sk_buff *, skb,
 		if (!tstamp)
 			return -EINVAL;
 		skb->tstamp = tstamp;
-		skb->mono_delivery_time = 1;
+		skb->tstamp_type = SKB_CLOCK_MONOTONIC;
 		break;
 	case BPF_SKB_TSTAMP_UNSPEC:
 		if (tstamp)
 			return -EINVAL;
 		skb->tstamp = 0;
-		skb->mono_delivery_time = 0;
+		skb->tstamp_type = SKB_CLOCK_REALTIME;
 		break;
 	default:
 		return -EINVAL;
@@ -9443,7 +9443,7 @@ static struct bpf_insn *bpf_convert_tstamp_read(const struct bpf_prog *prog,
 					TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK);
 		*insn++ = BPF_JMP32_IMM(BPF_JNE, tmp_reg,
 					TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK, 2);
-		/* skb->tc_at_ingress && skb->mono_delivery_time,
+		/* skb->tc_at_ingress && skb->tstamp_type,
 		 * read 0 as the (rcv) timestamp.
 		 */
 		*insn++ = BPF_MOV64_IMM(value_reg, 0);
@@ -9468,7 +9468,7 @@ static struct bpf_insn *bpf_convert_tstamp_write(const struct bpf_prog *prog,
 	 * the bpf prog is aware the tstamp could have delivery time.
 	 * Thus, write skb->tstamp as is if tstamp_type_access is true.
 	 * Otherwise, writing at ingress will have to clear the
-	 * mono_delivery_time bit also.
+	 * skb->tstamp_type bit also.
 	 */
 	if (!prog->tstamp_type_access) {
 		__u8 tmp_reg = BPF_REG_AX;
@@ -9478,7 +9478,7 @@ static struct bpf_insn *bpf_convert_tstamp_write(const struct bpf_prog *prog,
 		*insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg, TC_AT_INGRESS_MASK, 1);
 		/* goto <store> */
 		*insn++ = BPF_JMP_A(2);
-		/* <clear>: mono_delivery_time */
+		/* <clear>: skb->tstamp_type */
 		*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg, ~SKB_MONO_DELIVERY_TIME_MASK);
 		*insn++ = BPF_STX_MEM(BPF_B, skb_reg, tmp_reg, SKB_BF_MONO_TC_OFFSET);
 	}
diff --git a/net/ieee802154/6lowpan/reassembly.c b/net/ieee802154/6lowpan/reassembly.c
index 56ef873828f4..867d637d86f0 100644
--- a/net/ieee802154/6lowpan/reassembly.c
+++ b/net/ieee802154/6lowpan/reassembly.c
@@ -130,7 +130,7 @@ static int lowpan_frag_queue(struct lowpan_frag_queue *fq,
 		goto err;
 
 	fq->q.stamp = skb->tstamp;
-	fq->q.mono_delivery_time = skb->mono_delivery_time;
+	fq->q.tstamp_type = skb->tstamp_type;
 	if (frag_type == LOWPAN_DISPATCH_FRAG1)
 		fq->q.flags |= INET_FRAG_FIRST_IN;
 
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index faaec92a46ac..d179a2c84222 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -619,7 +619,7 @@ void inet_frag_reasm_finish(struct inet_frag_queue *q, struct sk_buff *head,
 	skb_mark_not_on_list(head);
 	head->prev = NULL;
 	head->tstamp = q->stamp;
-	head->mono_delivery_time = q->mono_delivery_time;
+	head->tstamp_type = q->tstamp_type;
 
 	if (sk)
 		refcount_add(sum_truesize - head_truesize, &sk->sk_wmem_alloc);
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 08e2c92e25ab..a92664a5ef2e 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -355,7 +355,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
 		qp->iif = dev->ifindex;
 
 	qp->q.stamp = skb->tstamp;
-	qp->q.mono_delivery_time = skb->mono_delivery_time;
+	qp->q.tstamp_type = skb->tstamp_type;
 	qp->q.meat += skb->len;
 	qp->ecn |= ecn;
 	add_frag_mem_limit(qp->q.fqdir, skb->truesize);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 9500031a1f55..fe86cadfa85b 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -764,7 +764,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
 {
 	struct iphdr *iph;
 	struct sk_buff *skb2;
-	bool mono_delivery_time = skb->mono_delivery_time;
+	u8 tstamp_type = skb->tstamp_type;
 	struct rtable *rt = skb_rtable(skb);
 	unsigned int mtu, hlen, ll_rs;
 	struct ip_fraglist_iter iter;
@@ -856,7 +856,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
 				}
 			}
 
-			skb_set_delivery_time(skb, tstamp, mono_delivery_time);
+			skb_set_delivery_time(skb, tstamp, tstamp_type);
 			err = output(net, sk, skb);
 
 			if (!err)
@@ -912,7 +912,7 @@ slow_path:
 		/*
 		 *	Put this fragment into the sending queue.
 		 */
-		skb_set_delivery_time(skb2, tstamp, mono_delivery_time);
+		skb_set_delivery_time(skb2, tstamp, tstamp_type);
 		err = output(net, sk, skb2);
 		if (err)
 			goto fail;
@@ -1649,7 +1649,8 @@ void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb,
 			  arg->csumoffset) = csum_fold(csum_add(nskb->csum,
 								arg->csum));
 		nskb->ip_summed = CHECKSUM_NONE;
-		nskb->mono_delivery_time = !!transmit_time;
+		if (transmit_time)
+			nskb->tstamp_type = SKB_CLOCK_MONOTONIC;
 		if (txhash)
 			skb_set_hash(nskb, txhash, PKT_HASH_TYPE_L4);
 		ip_push_pending_frames(sk, &fl4);
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 95caf8aaa8be..d44371cfa6ec 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -1301,7 +1301,7 @@ static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb,
 	tp = tcp_sk(sk);
 	prior_wstamp = tp->tcp_wstamp_ns;
 	tp->tcp_wstamp_ns = max(tp->tcp_wstamp_ns, tp->tcp_clock_cache);
-	skb_set_delivery_time(skb, tp->tcp_wstamp_ns, true);
+	skb_set_delivery_time(skb, tp->tcp_wstamp_ns, SKB_CLOCK_MONOTONIC);
 	if (clone_it) {
 		oskb = skb;
 
@@ -1655,7 +1655,7 @@ int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue,
 
 	skb_split(skb, buff, len);
 
-	skb_set_delivery_time(buff, skb->tstamp, true);
+	skb_set_delivery_time(buff, skb->tstamp, SKB_CLOCK_MONOTONIC);
 	tcp_fragment_tstamp(skb, buff);
 
 	old_factor = tcp_skb_pcount(skb);
@@ -2764,7 +2764,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle,
 		if (unlikely(tp->repair) && tp->repair_queue == TCP_SEND_QUEUE) {
 			/* "skb_mstamp_ns" is used as a start point for the retransmit timer */
 			tp->tcp_wstamp_ns = tp->tcp_clock_cache;
-			skb_set_delivery_time(skb, tp->tcp_wstamp_ns, true);
+			skb_set_delivery_time(skb, tp->tcp_wstamp_ns, SKB_CLOCK_MONOTONIC);
 			list_move_tail(&skb->tcp_tsorted_anchor, &tp->tsorted_sent_queue);
 			tcp_init_tso_segs(skb, mss_now);
 			goto repair; /* Skip network transmission */
@@ -3752,11 +3752,11 @@ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst,
 #ifdef CONFIG_SYN_COOKIES
 	if (unlikely(synack_type == TCP_SYNACK_COOKIE && ireq->tstamp_ok))
 		skb_set_delivery_time(skb, cookie_init_timestamp(req, now),
-				      true);
+				      SKB_CLOCK_MONOTONIC);
 	else
 #endif
 	{
-		skb_set_delivery_time(skb, now, true);
+		skb_set_delivery_time(skb, now, SKB_CLOCK_MONOTONIC);
 		if (!tcp_rsk(req)->snt_synack) /* Timestamp first SYNACK */
 			tcp_rsk(req)->snt_synack = tcp_skb_timestamp_us(skb);
 	}
@@ -3843,7 +3843,7 @@ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst,
 	bpf_skops_write_hdr_opt((struct sock *)sk, skb, req, syn_skb,
 				synack_type, &opts);
 
-	skb_set_delivery_time(skb, now, true);
+	skb_set_delivery_time(skb, now, SKB_CLOCK_MONOTONIC);
 	tcp_add_tx_delay(skb, tp);
 
 	return skb;
@@ -4027,7 +4027,7 @@ static int tcp_send_syn_data(struct sock *sk, struct sk_buff *syn)
 
 	err = tcp_transmit_skb(sk, syn_data, 1, sk->sk_allocation);
 
-	skb_set_delivery_time(syn, syn_data->skb_mstamp_ns, true);
+	skb_set_delivery_time(syn, syn_data->skb_mstamp_ns, SKB_CLOCK_MONOTONIC);
 
 	/* Now full SYN+DATA was cloned and sent (or not),
 	 * remove the SYN from the original skb (syn_data)
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 27d8725445e3..1ab0f23d37bf 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -859,7 +859,7 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
 	struct rt6_info *rt = dst_rt6_info(skb_dst(skb));
 	struct ipv6_pinfo *np = skb->sk && !dev_recursion_level() ?
 				inet6_sk(skb->sk) : NULL;
-	bool mono_delivery_time = skb->mono_delivery_time;
+	u8 tstamp_type = skb->tstamp_type;
 	struct ip6_frag_state state;
 	unsigned int mtu, hlen, nexthdr_offset;
 	ktime_t tstamp = skb->tstamp;
@@ -955,7 +955,7 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
 			if (iter.frag)
 				ip6_fraglist_prepare(skb, &iter);
 
-			skb_set_delivery_time(skb, tstamp, mono_delivery_time);
+			skb_set_delivery_time(skb, tstamp, tstamp_type);
 			err = output(net, sk, skb);
 			if (!err)
 				IP6_INC_STATS(net, ip6_dst_idev(&rt->dst),
@@ -1016,7 +1016,7 @@ slow_path:
 		/*
 		 *	Put this fragment into the sending queue.
 		 */
-		skb_set_delivery_time(frag, tstamp, mono_delivery_time);
+		skb_set_delivery_time(frag, tstamp, tstamp_type);
 		err = output(net, sk, frag);
 		if (err)
 			goto fail;
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index 53d255838e6a..e0c2347b4dc6 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -126,7 +126,7 @@ int br_ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
 				  struct sk_buff *))
 {
 	int frag_max_size = BR_INPUT_SKB_CB(skb)->frag_max_size;
-	bool mono_delivery_time = skb->mono_delivery_time;
+	u8 tstamp_type = skb->tstamp_type;
 	ktime_t tstamp = skb->tstamp;
 	struct ip6_frag_state state;
 	u8 *prevhdr, nexthdr = 0;
@@ -192,7 +192,7 @@ int br_ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
 			if (iter.frag)
 				ip6_fraglist_prepare(skb, &iter);
 
-			skb_set_delivery_time(skb, tstamp, mono_delivery_time);
+			skb_set_delivery_time(skb, tstamp, tstamp_type);
 			err = output(net, sk, data, skb);
 			if (err || !iter.frag)
 				break;
@@ -225,7 +225,7 @@ slow_path:
 			goto blackhole;
 		}
 
-		skb_set_delivery_time(skb2, tstamp, mono_delivery_time);
+		skb_set_delivery_time(skb2, tstamp, tstamp_type);
 		err = output(net, sk, data, skb2);
 		if (err)
 			goto blackhole;
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index 5e1b50c6a44d..6f0844c9315d 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -263,7 +263,7 @@ static int nf_ct_frag6_queue(struct frag_queue *fq, struct sk_buff *skb,
 		fq->iif = dev->ifindex;
 
 	fq->q.stamp = skb->tstamp;
-	fq->q.mono_delivery_time = skb->mono_delivery_time;
+	fq->q.tstamp_type = skb->tstamp_type;
 	fq->q.meat += skb->len;
 	fq->ecn |= ecn;
 	if (payload_len > fq->q.max_size)
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 327caca64257..a48be617a8ab 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -198,7 +198,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb,
 		fq->iif = dev->ifindex;
 
 	fq->q.stamp = skb->tstamp;
-	fq->q.mono_delivery_time = skb->mono_delivery_time;
+	fq->q.tstamp_type = skb->tstamp_type;
 	fq->q.meat += skb->len;
 	fq->ecn |= ecn;
 	add_frag_mem_limit(fq->q.fqdir, skb->truesize);
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 4c3605485b68..8333005c5c2e 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -975,7 +975,7 @@ static void tcp_v6_send_response(const struct sock *sk, struct sk_buff *skb, u32
 			mark = inet_twsk(sk)->tw_mark;
 		else
 			mark = READ_ONCE(sk->sk_mark);
-		skb_set_delivery_time(buff, tcp_transmit_time(sk), true);
+		skb_set_delivery_time(buff, tcp_transmit_time(sk), SKB_CLOCK_MONOTONIC);
 	}
 	if (txhash) {
 		/* autoflowlabel/skb_get_hash_flowi6 rely on buff->hash */
diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c
index 0e3cf11ae5fc..396b576390d0 100644
--- a/net/sched/act_bpf.c
+++ b/net/sched/act_bpf.c
@@ -54,8 +54,8 @@ TC_INDIRECT_SCOPE int tcf_bpf_act(struct sk_buff *skb,
 		bpf_compute_data_pointers(skb);
 		filter_res = bpf_prog_run(filter, skb);
 	}
-	if (unlikely(!skb->tstamp && skb->mono_delivery_time))
-		skb->mono_delivery_time = 0;
+	if (unlikely(!skb->tstamp && skb->tstamp_type))
+		skb->tstamp_type = SKB_CLOCK_REALTIME;
 	if (skb_sk_is_prefetched(skb) && filter_res != TC_ACT_OK)
 		skb_orphan(skb);
 
diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c
index 5e83e890f6a4..1941ebec23ff 100644
--- a/net/sched/cls_bpf.c
+++ b/net/sched/cls_bpf.c
@@ -104,8 +104,8 @@ TC_INDIRECT_SCOPE int cls_bpf_classify(struct sk_buff *skb,
 			bpf_compute_data_pointers(skb);
 			filter_res = bpf_prog_run(prog->filter, skb);
 		}
-		if (unlikely(!skb->tstamp && skb->mono_delivery_time))
-			skb->mono_delivery_time = 0;
+		if (unlikely(!skb->tstamp && skb->tstamp_type))
+			skb->tstamp_type = SKB_CLOCK_REALTIME;
 
 		if (prog->exts_integrated) {
 			res->class   = 0;
-- 
cgit v1.2.3-58-ga151


From 1693c5db6ab8262e6f5263f9d211855959aa5acd Mon Sep 17 00:00:00 2001
From: Abhishek Chauhan <quic_abchauha@quicinc.com>
Date: Thu, 9 May 2024 14:18:33 -0700
Subject: net: Add additional bit to support clockid_t timestamp type

tstamp_type is now set based on actual clockid_t compressed
into 2 bits.

To make the design scalable for future needs this commit bring in
the change to extend the tstamp_type:1 to tstamp_type:2 to support
other clockid_t timestamp.

We now support CLOCK_TAI as part of tstamp_type as part of this
commit with existing support CLOCK_MONOTONIC and CLOCK_REALTIME.

Signed-off-by: Abhishek Chauhan <quic_abchauha@quicinc.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20240509211834.3235191-3-quic_abchauha@quicinc.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/skbuff.h         | 18 +++++++++++------
 include/uapi/linux/bpf.h       | 15 +++++++++-----
 net/core/filter.c              | 46 ++++++++++++++++++++++++------------------
 net/ipv4/ip_output.c           |  5 ++++-
 net/ipv4/raw.c                 |  2 +-
 net/ipv4/tcp_ipv4.c            |  2 ++
 net/ipv6/ip6_output.c          |  5 ++++-
 net/ipv6/raw.c                 |  2 +-
 net/ipv6/tcp_ipv6.c            | 10 +++++++--
 net/packet/af_packet.c         |  7 +++----
 tools/include/uapi/linux/bpf.h | 15 +++++++++-----
 11 files changed, 81 insertions(+), 46 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 3a721cc3b644..1e5c97daaa37 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -709,6 +709,8 @@ typedef unsigned char *sk_buff_data_t;
 enum skb_tstamp_type {
 	SKB_CLOCK_REALTIME,
 	SKB_CLOCK_MONOTONIC,
+	SKB_CLOCK_TAI,
+	__SKB_CLOCK_MAX = SKB_CLOCK_TAI,
 };
 
 /**
@@ -957,7 +959,7 @@ struct sk_buff {
 	/* private: */
 	__u8			__mono_tc_offset[0];
 	/* public: */
-	__u8			tstamp_type:1;	/* See skb_tstamp_type */
+	__u8			tstamp_type:2;	/* See skb_tstamp_type */
 #ifdef CONFIG_NET_XGRESS
 	__u8			tc_at_ingress:1;	/* See TC_AT_INGRESS_MASK */
 	__u8			tc_skip_classify:1;
@@ -1087,15 +1089,16 @@ struct sk_buff {
 #endif
 #define PKT_TYPE_OFFSET		offsetof(struct sk_buff, __pkt_type_offset)
 
-/* if you move tc_at_ingress or mono_delivery_time
+/* if you move tc_at_ingress or tstamp_type
  * around, you also must adapt these constants.
  */
 #ifdef __BIG_ENDIAN_BITFIELD
-#define SKB_MONO_DELIVERY_TIME_MASK	(1 << 7)
-#define TC_AT_INGRESS_MASK		(1 << 6)
+#define SKB_TSTAMP_TYPE_MASK		(3 << 6)
+#define SKB_TSTAMP_TYPE_RSHIFT		(6)
+#define TC_AT_INGRESS_MASK		(1 << 5)
 #else
-#define SKB_MONO_DELIVERY_TIME_MASK	(1 << 0)
-#define TC_AT_INGRESS_MASK		(1 << 1)
+#define SKB_TSTAMP_TYPE_MASK		(3)
+#define TC_AT_INGRESS_MASK		(1 << 2)
 #endif
 #define SKB_BF_MONO_TC_OFFSET		offsetof(struct sk_buff, __mono_tc_offset)
 
@@ -4216,6 +4219,9 @@ static inline void skb_set_delivery_type_by_clockid(struct sk_buff *skb,
 	case CLOCK_MONOTONIC:
 		tstamp_type = SKB_CLOCK_MONOTONIC;
 		break;
+	case CLOCK_TAI:
+		tstamp_type = SKB_CLOCK_TAI;
+		break;
 	default:
 		WARN_ON_ONCE(1);
 		kt = 0;
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 90706a47f6ff..25ea393cf084 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -6207,12 +6207,17 @@ union {					\
 	__u64 :64;			\
 } __attribute__((aligned(8)))
 
+/* The enum used in skb->tstamp_type. It specifies the clock type
+ * of the time stored in the skb->tstamp.
+ */
 enum {
-	BPF_SKB_TSTAMP_UNSPEC,
-	BPF_SKB_TSTAMP_DELIVERY_MONO,	/* tstamp has mono delivery time */
-	/* For any BPF_SKB_TSTAMP_* that the bpf prog cannot handle,
-	 * the bpf prog should handle it like BPF_SKB_TSTAMP_UNSPEC
-	 * and try to deduce it by ingress, egress or skb->sk->sk_clockid.
+	BPF_SKB_TSTAMP_UNSPEC = 0,		/* DEPRECATED */
+	BPF_SKB_TSTAMP_DELIVERY_MONO = 1,	/* DEPRECATED */
+	BPF_SKB_CLOCK_REALTIME = 0,
+	BPF_SKB_CLOCK_MONOTONIC = 1,
+	BPF_SKB_CLOCK_TAI = 2,
+	/* For any future BPF_SKB_CLOCK_* that the bpf prog cannot handle,
+	 * the bpf prog can try to deduce it by ingress/egress/skb->sk->sk_clockid.
 	 */
 };
 
diff --git a/net/core/filter.c b/net/core/filter.c
index a3781a796da4..c6edfe9f41bc 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -7726,17 +7726,21 @@ BPF_CALL_3(bpf_skb_set_tstamp, struct sk_buff *, skb,
 		return -EOPNOTSUPP;
 
 	switch (tstamp_type) {
-	case BPF_SKB_TSTAMP_DELIVERY_MONO:
+	case BPF_SKB_CLOCK_REALTIME:
+		skb->tstamp = tstamp;
+		skb->tstamp_type = SKB_CLOCK_REALTIME;
+		break;
+	case BPF_SKB_CLOCK_MONOTONIC:
 		if (!tstamp)
 			return -EINVAL;
 		skb->tstamp = tstamp;
 		skb->tstamp_type = SKB_CLOCK_MONOTONIC;
 		break;
-	case BPF_SKB_TSTAMP_UNSPEC:
-		if (tstamp)
+	case BPF_SKB_CLOCK_TAI:
+		if (!tstamp)
 			return -EINVAL;
-		skb->tstamp = 0;
-		skb->tstamp_type = SKB_CLOCK_REALTIME;
+		skb->tstamp = tstamp;
+		skb->tstamp_type = SKB_CLOCK_TAI;
 		break;
 	default:
 		return -EINVAL;
@@ -9387,16 +9391,17 @@ static struct bpf_insn *bpf_convert_tstamp_type_read(const struct bpf_insn *si,
 {
 	__u8 value_reg = si->dst_reg;
 	__u8 skb_reg = si->src_reg;
-	/* AX is needed because src_reg and dst_reg could be the same */
-	__u8 tmp_reg = BPF_REG_AX;
-
-	*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg,
-			      SKB_BF_MONO_TC_OFFSET);
-	*insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg,
-				SKB_MONO_DELIVERY_TIME_MASK, 2);
-	*insn++ = BPF_MOV32_IMM(value_reg, BPF_SKB_TSTAMP_UNSPEC);
-	*insn++ = BPF_JMP_A(1);
-	*insn++ = BPF_MOV32_IMM(value_reg, BPF_SKB_TSTAMP_DELIVERY_MONO);
+	BUILD_BUG_ON(__SKB_CLOCK_MAX != (int)BPF_SKB_CLOCK_TAI);
+	BUILD_BUG_ON(SKB_CLOCK_REALTIME != (int)BPF_SKB_CLOCK_REALTIME);
+	BUILD_BUG_ON(SKB_CLOCK_MONOTONIC != (int)BPF_SKB_CLOCK_MONOTONIC);
+	BUILD_BUG_ON(SKB_CLOCK_TAI != (int)BPF_SKB_CLOCK_TAI);
+	*insn++ = BPF_LDX_MEM(BPF_B, value_reg, skb_reg, SKB_BF_MONO_TC_OFFSET);
+	*insn++ = BPF_ALU32_IMM(BPF_AND, value_reg, SKB_TSTAMP_TYPE_MASK);
+#ifdef __BIG_ENDIAN_BITFIELD
+	*insn++ = BPF_ALU32_IMM(BPF_RSH, value_reg, SKB_TSTAMP_TYPE_RSHIFT);
+#else
+	BUILD_BUG_ON(!(SKB_TSTAMP_TYPE_MASK & 0x1));
+#endif
 
 	return insn;
 }
@@ -9439,10 +9444,11 @@ static struct bpf_insn *bpf_convert_tstamp_read(const struct bpf_prog *prog,
 		__u8 tmp_reg = BPF_REG_AX;
 
 		*insn++ = BPF_LDX_MEM(BPF_B, tmp_reg, skb_reg, SKB_BF_MONO_TC_OFFSET);
-		*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg,
-					TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK);
-		*insn++ = BPF_JMP32_IMM(BPF_JNE, tmp_reg,
-					TC_AT_INGRESS_MASK | SKB_MONO_DELIVERY_TIME_MASK, 2);
+		/* check if ingress mask bits is set */
+		*insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg, TC_AT_INGRESS_MASK, 1);
+		*insn++ = BPF_JMP_A(4);
+		*insn++ = BPF_JMP32_IMM(BPF_JSET, tmp_reg, SKB_TSTAMP_TYPE_MASK, 1);
+		*insn++ = BPF_JMP_A(2);
 		/* skb->tc_at_ingress && skb->tstamp_type,
 		 * read 0 as the (rcv) timestamp.
 		 */
@@ -9479,7 +9485,7 @@ static struct bpf_insn *bpf_convert_tstamp_write(const struct bpf_prog *prog,
 		/* goto <store> */
 		*insn++ = BPF_JMP_A(2);
 		/* <clear>: skb->tstamp_type */
-		*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg, ~SKB_MONO_DELIVERY_TIME_MASK);
+		*insn++ = BPF_ALU32_IMM(BPF_AND, tmp_reg, ~SKB_TSTAMP_TYPE_MASK);
 		*insn++ = BPF_STX_MEM(BPF_B, skb_reg, tmp_reg, SKB_BF_MONO_TC_OFFSET);
 	}
 #endif
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index fe86cadfa85b..b90d0f78ac80 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -1457,7 +1457,10 @@ struct sk_buff *__ip_make_skb(struct sock *sk,
 
 	skb->priority = (cork->tos != -1) ? cork->priority: READ_ONCE(sk->sk_priority);
 	skb->mark = cork->mark;
-	skb->tstamp = cork->transmit_time;
+	if (sk_is_tcp(sk))
+		skb_set_delivery_time(skb, cork->transmit_time, SKB_CLOCK_MONOTONIC);
+	else
+		skb_set_delivery_type_by_clockid(skb, cork->transmit_time, sk->sk_clockid);
 	/*
 	 * Steal rt from cork.dst to avoid a pair of atomic_inc/atomic_dec
 	 * on dst refcount
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 4cb43401e0e0..1a0953650356 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -360,7 +360,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4,
 	skb->protocol = htons(ETH_P_IP);
 	skb->priority = READ_ONCE(sk->sk_priority);
 	skb->mark = sockc->mark;
-	skb->tstamp = sockc->transmit_time;
+	skb_set_delivery_type_by_clockid(skb, sockc->transmit_time, sk->sk_clockid);
 	skb_dst_set(skb, &rt->dst);
 	*rtp = NULL;
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 30ef0c8f5e92..8f70b8d1d1e5 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -3625,6 +3625,8 @@ void __init tcp_v4_init(void)
 		 */
 		inet_sk(sk)->pmtudisc = IP_PMTUDISC_DO;
 
+		sk->sk_clockid = CLOCK_MONOTONIC;
+
 		per_cpu(ipv4_tcp_sk, cpu) = sk;
 	}
 	if (register_pernet_subsys(&tcp_sk_ops))
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 1ab0f23d37bf..e7a19df3125e 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1924,7 +1924,10 @@ struct sk_buff *__ip6_make_skb(struct sock *sk,
 
 	skb->priority = READ_ONCE(sk->sk_priority);
 	skb->mark = cork->base.mark;
-	skb->tstamp = cork->base.transmit_time;
+	if (sk_is_tcp(sk))
+		skb_set_delivery_time(skb, cork->base.transmit_time, SKB_CLOCK_MONOTONIC);
+	else
+		skb_set_delivery_type_by_clockid(skb, cork->base.transmit_time, sk->sk_clockid);
 
 	ip6_cork_steal_dst(skb, cork);
 	IP6_INC_STATS(net, rt->rt6i_idev, IPSTATS_MIB_OUTREQUESTS);
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 2eedf255600b..f838366e8256 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -621,7 +621,7 @@ static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length,
 	skb->protocol = htons(ETH_P_IPV6);
 	skb->priority = READ_ONCE(sk->sk_priority);
 	skb->mark = sockc->mark;
-	skb->tstamp = sockc->transmit_time;
+	skb_set_delivery_type_by_clockid(skb, sockc->transmit_time, sk->sk_clockid);
 
 	skb_put(skb, length);
 	skb_reset_network_header(skb);
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 8333005c5c2e..750aa681779c 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2387,8 +2387,14 @@ static struct inet_protosw tcpv6_protosw = {
 
 static int __net_init tcpv6_net_init(struct net *net)
 {
-	return inet_ctl_sock_create(&net->ipv6.tcp_sk, PF_INET6,
-				    SOCK_RAW, IPPROTO_TCP, net);
+	int res;
+
+	res = inet_ctl_sock_create(&net->ipv6.tcp_sk, PF_INET6,
+				   SOCK_RAW, IPPROTO_TCP, net);
+	if (!res)
+		net->ipv6.tcp_sk->sk_clockid = CLOCK_MONOTONIC;
+
+	return res;
 }
 
 static void __net_exit tcpv6_net_exit(struct net *net)
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index ea3ebc160e25..fce390887591 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2056,8 +2056,7 @@ retry:
 	skb->dev = dev;
 	skb->priority = READ_ONCE(sk->sk_priority);
 	skb->mark = READ_ONCE(sk->sk_mark);
-	skb->tstamp = sockc.transmit_time;
-
+	skb_set_delivery_type_by_clockid(skb, sockc.transmit_time, sk->sk_clockid);
 	skb_setup_tx_timestamp(skb, sockc.tsflags);
 
 	if (unlikely(extra_len == 4))
@@ -2584,7 +2583,7 @@ static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb,
 	skb->dev = dev;
 	skb->priority = READ_ONCE(po->sk.sk_priority);
 	skb->mark = READ_ONCE(po->sk.sk_mark);
-	skb->tstamp = sockc->transmit_time;
+	skb_set_delivery_type_by_clockid(skb, sockc->transmit_time, po->sk.sk_clockid);
 	skb_setup_tx_timestamp(skb, sockc->tsflags);
 	skb_zcopy_set_nouarg(skb, ph.raw);
 
@@ -3062,7 +3061,7 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 	skb->dev = dev;
 	skb->priority = READ_ONCE(sk->sk_priority);
 	skb->mark = sockc.mark;
-	skb->tstamp = sockc.transmit_time;
+	skb_set_delivery_type_by_clockid(skb, sockc.transmit_time, sk->sk_clockid);
 
 	if (unlikely(extra_len == 4))
 		skb->no_fcs = 1;
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 90706a47f6ff..25ea393cf084 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -6207,12 +6207,17 @@ union {					\
 	__u64 :64;			\
 } __attribute__((aligned(8)))
 
+/* The enum used in skb->tstamp_type. It specifies the clock type
+ * of the time stored in the skb->tstamp.
+ */
 enum {
-	BPF_SKB_TSTAMP_UNSPEC,
-	BPF_SKB_TSTAMP_DELIVERY_MONO,	/* tstamp has mono delivery time */
-	/* For any BPF_SKB_TSTAMP_* that the bpf prog cannot handle,
-	 * the bpf prog should handle it like BPF_SKB_TSTAMP_UNSPEC
-	 * and try to deduce it by ingress, egress or skb->sk->sk_clockid.
+	BPF_SKB_TSTAMP_UNSPEC = 0,		/* DEPRECATED */
+	BPF_SKB_TSTAMP_DELIVERY_MONO = 1,	/* DEPRECATED */
+	BPF_SKB_CLOCK_REALTIME = 0,
+	BPF_SKB_CLOCK_MONOTONIC = 1,
+	BPF_SKB_CLOCK_TAI = 2,
+	/* For any future BPF_SKB_CLOCK_* that the bpf prog cannot handle,
+	 * the bpf prog can try to deduce it by ingress/egress/skb->sk->sk_clockid.
 	 */
 };
 
-- 
cgit v1.2.3-58-ga151


From c34e3ab2a76e6a55a64e0d56acc5607062c2bad9 Mon Sep 17 00:00:00 2001
From: Abhishek Chauhan <quic_abchauha@quicinc.com>
Date: Thu, 9 May 2024 14:18:34 -0700
Subject: selftests/bpf: Handle forwarding of UDP CLOCK_TAI packets

With changes in the design to forward CLOCK_TAI in the skbuff
framework,  existing selftest framework needs modification
to handle forwarding of UDP packets with CLOCK_TAI as clockid.

Signed-off-by: Abhishek Chauhan <quic_abchauha@quicinc.com>
Reviewed-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20240509211834.3235191-4-quic_abchauha@quicinc.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 .../testing/selftests/bpf/prog_tests/ctx_rewrite.c | 10 +++---
 .../testing/selftests/bpf/prog_tests/tc_redirect.c |  3 --
 tools/testing/selftests/bpf/progs/test_tc_dtime.c  | 39 ++++++++++------------
 3 files changed, 23 insertions(+), 29 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c b/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c
index 3b7c57fe55a5..08b6391f2f56 100644
--- a/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c
+++ b/tools/testing/selftests/bpf/prog_tests/ctx_rewrite.c
@@ -69,15 +69,17 @@ static struct test_case test_cases[] = {
 	{
 		N(SCHED_CLS, struct __sk_buff, tstamp),
 		.read  = "r11 = *(u8 *)($ctx + sk_buff::__mono_tc_offset);"
-			 "w11 &= 3;"
-			 "if w11 != 0x3 goto pc+2;"
+			 "if w11 & 0x4 goto pc+1;"
+			 "goto pc+4;"
+			 "if w11 & 0x3 goto pc+1;"
+			 "goto pc+2;"
 			 "$dst = 0;"
 			 "goto pc+1;"
 			 "$dst = *(u64 *)($ctx + sk_buff::tstamp);",
 		.write = "r11 = *(u8 *)($ctx + sk_buff::__mono_tc_offset);"
-			 "if w11 & 0x2 goto pc+1;"
+			 "if w11 & 0x4 goto pc+1;"
 			 "goto pc+2;"
-			 "w11 &= -2;"
+			 "w11 &= -4;"
 			 "*(u8 *)($ctx + sk_buff::__mono_tc_offset) = r11;"
 			 "*(u64 *)($ctx + sk_buff::tstamp) = $src;",
 	},
diff --git a/tools/testing/selftests/bpf/prog_tests/tc_redirect.c b/tools/testing/selftests/bpf/prog_tests/tc_redirect.c
index b1073d36d77a..327d51f59142 100644
--- a/tools/testing/selftests/bpf/prog_tests/tc_redirect.c
+++ b/tools/testing/selftests/bpf/prog_tests/tc_redirect.c
@@ -890,9 +890,6 @@ static void test_udp_dtime(struct test_tc_dtime *skel, int family, bool bpf_fwd)
 
 	ASSERT_EQ(dtimes[INGRESS_FWDNS_P100], 0,
 		  dtime_cnt_str(t, INGRESS_FWDNS_P100));
-	/* non mono delivery time is not forwarded */
-	ASSERT_EQ(dtimes[INGRESS_FWDNS_P101], 0,
-		  dtime_cnt_str(t, INGRESS_FWDNS_P101));
 	for (i = EGRESS_FWDNS_P100; i < SET_DTIME; i++)
 		ASSERT_GT(dtimes[i], 0, dtime_cnt_str(t, i));
 
diff --git a/tools/testing/selftests/bpf/progs/test_tc_dtime.c b/tools/testing/selftests/bpf/progs/test_tc_dtime.c
index 74ec09f040b7..ca8e8734d901 100644
--- a/tools/testing/selftests/bpf/progs/test_tc_dtime.c
+++ b/tools/testing/selftests/bpf/progs/test_tc_dtime.c
@@ -222,17 +222,21 @@ int egress_host(struct __sk_buff *skb)
 		return TC_ACT_OK;
 
 	if (skb_proto(skb_type) == IPPROTO_TCP) {
-		if (skb->tstamp_type == BPF_SKB_TSTAMP_DELIVERY_MONO &&
+		if (skb->tstamp_type == BPF_SKB_CLOCK_MONOTONIC &&
 		    skb->tstamp)
 			inc_dtimes(EGRESS_ENDHOST);
 		else
 			inc_errs(EGRESS_ENDHOST);
-	} else {
-		if (skb->tstamp_type == BPF_SKB_TSTAMP_UNSPEC &&
+	} else if (skb_proto(skb_type) == IPPROTO_UDP) {
+		if (skb->tstamp_type == BPF_SKB_CLOCK_TAI &&
 		    skb->tstamp)
 			inc_dtimes(EGRESS_ENDHOST);
 		else
 			inc_errs(EGRESS_ENDHOST);
+	} else {
+		if (skb->tstamp_type == BPF_SKB_CLOCK_REALTIME &&
+		    skb->tstamp)
+			inc_errs(EGRESS_ENDHOST);
 	}
 
 	skb->tstamp = EGRESS_ENDHOST_MAGIC;
@@ -252,7 +256,7 @@ int ingress_host(struct __sk_buff *skb)
 	if (!skb_type)
 		return TC_ACT_OK;
 
-	if (skb->tstamp_type == BPF_SKB_TSTAMP_DELIVERY_MONO &&
+	if (skb->tstamp_type == BPF_SKB_CLOCK_MONOTONIC &&
 	    skb->tstamp == EGRESS_FWDNS_MAGIC)
 		inc_dtimes(INGRESS_ENDHOST);
 	else
@@ -315,7 +319,6 @@ int egress_fwdns_prio100(struct __sk_buff *skb)
 SEC("tc")
 int ingress_fwdns_prio101(struct __sk_buff *skb)
 {
-	__u64 expected_dtime = EGRESS_ENDHOST_MAGIC;
 	int skb_type;
 
 	skb_type = skb_get_type(skb);
@@ -323,29 +326,24 @@ int ingress_fwdns_prio101(struct __sk_buff *skb)
 		/* Should have handled in prio100 */
 		return TC_ACT_SHOT;
 
-	if (skb_proto(skb_type) == IPPROTO_UDP)
-		expected_dtime = 0;
-
 	if (skb->tstamp_type) {
 		if (fwdns_clear_dtime() ||
-		    skb->tstamp_type != BPF_SKB_TSTAMP_DELIVERY_MONO ||
-		    skb->tstamp != expected_dtime)
+		    (skb->tstamp_type != BPF_SKB_CLOCK_MONOTONIC &&
+		    skb->tstamp_type != BPF_SKB_CLOCK_TAI) ||
+		    skb->tstamp != EGRESS_ENDHOST_MAGIC)
 			inc_errs(INGRESS_FWDNS_P101);
 		else
 			inc_dtimes(INGRESS_FWDNS_P101);
 	} else {
-		if (!fwdns_clear_dtime() && expected_dtime)
+		if (!fwdns_clear_dtime())
 			inc_errs(INGRESS_FWDNS_P101);
 	}
 
-	if (skb->tstamp_type == BPF_SKB_TSTAMP_DELIVERY_MONO) {
+	if (skb->tstamp_type == BPF_SKB_CLOCK_MONOTONIC) {
 		skb->tstamp = INGRESS_FWDNS_MAGIC;
 	} else {
 		if (bpf_skb_set_tstamp(skb, INGRESS_FWDNS_MAGIC,
-				       BPF_SKB_TSTAMP_DELIVERY_MONO))
-			inc_errs(SET_DTIME);
-		if (!bpf_skb_set_tstamp(skb, INGRESS_FWDNS_MAGIC,
-					BPF_SKB_TSTAMP_UNSPEC))
+				       BPF_SKB_CLOCK_MONOTONIC))
 			inc_errs(SET_DTIME);
 	}
 
@@ -370,7 +368,7 @@ int egress_fwdns_prio101(struct __sk_buff *skb)
 
 	if (skb->tstamp_type) {
 		if (fwdns_clear_dtime() ||
-		    skb->tstamp_type != BPF_SKB_TSTAMP_DELIVERY_MONO ||
+		    skb->tstamp_type != BPF_SKB_CLOCK_MONOTONIC ||
 		    skb->tstamp != INGRESS_FWDNS_MAGIC)
 			inc_errs(EGRESS_FWDNS_P101);
 		else
@@ -380,14 +378,11 @@ int egress_fwdns_prio101(struct __sk_buff *skb)
 			inc_errs(EGRESS_FWDNS_P101);
 	}
 
-	if (skb->tstamp_type == BPF_SKB_TSTAMP_DELIVERY_MONO) {
+	if (skb->tstamp_type == BPF_SKB_CLOCK_MONOTONIC) {
 		skb->tstamp = EGRESS_FWDNS_MAGIC;
 	} else {
 		if (bpf_skb_set_tstamp(skb, EGRESS_FWDNS_MAGIC,
-				       BPF_SKB_TSTAMP_DELIVERY_MONO))
-			inc_errs(SET_DTIME);
-		if (!bpf_skb_set_tstamp(skb, INGRESS_FWDNS_MAGIC,
-					BPF_SKB_TSTAMP_UNSPEC))
+				       BPF_SKB_CLOCK_MONOTONIC))
 			inc_errs(SET_DTIME);
 	}
 
-- 
cgit v1.2.3-58-ga151


From c12603e76ef666ce5c51a9d6faf155c9e3de7601 Mon Sep 17 00:00:00 2001
From: Xiao Wang <xiao.w.wang@intel.com>
Date: Thu, 16 May 2024 17:04:30 +0800
Subject: riscv, bpf: Optimize zextw insn with Zba extension

The Zba extension provides add.uw insn which can be used to implement
zext.w with rs2 set as ZERO.

Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Pu Lehui <pulehui@huawei.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Pu Lehui <pulehui@huawei.com>
Link: https://lore.kernel.org/bpf/20240516090430.493122-1-xiao.w.wang@intel.com
---
 arch/riscv/Kconfig       | 12 ++++++++++++
 arch/riscv/net/bpf_jit.h | 18 ++++++++++++++++++
 2 files changed, 30 insertions(+)

diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
index 9e87287942dc..6b8f1059594d 100644
--- a/arch/riscv/Kconfig
+++ b/arch/riscv/Kconfig
@@ -595,6 +595,18 @@ config TOOLCHAIN_HAS_VECTOR_CRYPTO
 	def_bool $(as-instr, .option arch$(comma) +v$(comma) +zvkb)
 	depends on AS_HAS_OPTION_ARCH
 
+config RISCV_ISA_ZBA
+	bool "Zba extension support for bit manipulation instructions"
+	default y
+	help
+	   Add support for enabling optimisations in the kernel when the Zba
+	   extension is detected at boot.
+
+	   The Zba extension provides instructions to accelerate the generation
+	   of addresses that index into arrays of basic data types.
+
+	   If you don't know what to do here, say Y.
+
 config RISCV_ISA_ZBB
 	bool "Zbb extension support for bit manipulation instructions"
 	depends on TOOLCHAIN_HAS_ZBB
diff --git a/arch/riscv/net/bpf_jit.h b/arch/riscv/net/bpf_jit.h
index fdbf88ca8b70..97041b58237a 100644
--- a/arch/riscv/net/bpf_jit.h
+++ b/arch/riscv/net/bpf_jit.h
@@ -18,6 +18,11 @@ static inline bool rvc_enabled(void)
 	return IS_ENABLED(CONFIG_RISCV_ISA_C);
 }
 
+static inline bool rvzba_enabled(void)
+{
+	return IS_ENABLED(CONFIG_RISCV_ISA_ZBA) && riscv_has_extension_likely(RISCV_ISA_EXT_ZBA);
+}
+
 static inline bool rvzbb_enabled(void)
 {
 	return IS_ENABLED(CONFIG_RISCV_ISA_ZBB) && riscv_has_extension_likely(RISCV_ISA_EXT_ZBB);
@@ -939,6 +944,14 @@ static inline u16 rvc_sdsp(u32 imm9, u8 rs2)
 	return rv_css_insn(0x7, imm, rs2, 0x2);
 }
 
+/* RV64-only ZBA instructions. */
+
+static inline u32 rvzba_zextw(u8 rd, u8 rs1)
+{
+	/* add.uw rd, rs1, ZERO */
+	return rv_r_insn(0x04, RV_REG_ZERO, rs1, 0, rd, 0x3b);
+}
+
 #endif /* __riscv_xlen == 64 */
 
 /* Helper functions that emit RVC instructions when possible. */
@@ -1161,6 +1174,11 @@ static inline void emit_zexth(u8 rd, u8 rs, struct rv_jit_context *ctx)
 
 static inline void emit_zextw(u8 rd, u8 rs, struct rv_jit_context *ctx)
 {
+	if (rvzba_enabled()) {
+		emit(rvzba_zextw(rd, rs), ctx);
+		return;
+	}
+
 	emit_slli(rd, rs, 32, ctx);
 	emit_srli(rd, rd, 32, ctx);
 }
-- 
cgit v1.2.3-58-ga151


From e944fc8152744a41dc62e720995538e48b053bb9 Mon Sep 17 00:00:00 2001
From: Xiao Wang <xiao.w.wang@intel.com>
Date: Thu, 23 May 2024 11:18:35 +0800
Subject: riscv, bpf: Use STACK_ALIGN macro for size rounding up

Use the macro STACK_ALIGN that is defined in asm/processor.h for stack size
rounding up, just like bpf_jit_comp32.c does.

Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Pu Lehui <pulehui@huawei.com>
Link: https://lore.kernel.org/bpf/20240523031835.3977713-1-xiao.w.wang@intel.com
---
 arch/riscv/net/bpf_jit_comp64.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c
index 79a001d5533e..c21a0ff23415 100644
--- a/arch/riscv/net/bpf_jit_comp64.c
+++ b/arch/riscv/net/bpf_jit_comp64.c
@@ -868,7 +868,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 	stack_size += 8;
 	sreg_off = stack_size;
 
-	stack_size = round_up(stack_size, 16);
+	stack_size = round_up(stack_size, STACK_ALIGN);
 
 	if (!is_struct_ops) {
 		/* For the trampoline called from function entry,
@@ -1960,7 +1960,7 @@ void bpf_jit_build_prologue(struct rv_jit_context *ctx, bool is_subprog)
 {
 	int i, stack_adjust = 0, store_offset, bpf_stack_adjust;
 
-	bpf_stack_adjust = round_up(ctx->prog->aux->stack_depth, 16);
+	bpf_stack_adjust = round_up(ctx->prog->aux->stack_depth, STACK_ALIGN);
 	if (bpf_stack_adjust)
 		mark_fp(ctx);
 
@@ -1982,7 +1982,7 @@ void bpf_jit_build_prologue(struct rv_jit_context *ctx, bool is_subprog)
 	if (ctx->arena_vm_start)
 		stack_adjust += 8;
 
-	stack_adjust = round_up(stack_adjust, 16);
+	stack_adjust = round_up(stack_adjust, STACK_ALIGN);
 	stack_adjust += bpf_stack_adjust;
 
 	store_offset = stack_adjust - 8;
-- 
cgit v1.2.3-58-ga151


From 99fa63d9ca60c4c1cc843fde205e4bc6e86b218f Mon Sep 17 00:00:00 2001
From: Xiao Wang <xiao.w.wang@intel.com>
Date: Sun, 19 May 2024 13:05:07 +0800
Subject: riscv, bpf: Try RVC for reg move within BPF_CMPXCHG JIT
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We could try to emit compressed insn for reg move operation during CMPXCHG
JIT, the instruction compression has no impact on the jump offsets of
following forward and backward jump instructions.

Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Björn Töpel <bjorn@kernel.org>
Link: https://lore.kernel.org/bpf/20240519050507.2217791-1-xiao.w.wang@intel.com
---
 arch/riscv/net/bpf_jit_comp64.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c
index c21a0ff23415..2bc4c14ea59a 100644
--- a/arch/riscv/net/bpf_jit_comp64.c
+++ b/arch/riscv/net/bpf_jit_comp64.c
@@ -537,8 +537,10 @@ static void emit_atomic(u8 rd, u8 rs, s16 off, s32 imm, bool is64,
 	/* r0 = atomic_cmpxchg(dst_reg + off16, r0, src_reg); */
 	case BPF_CMPXCHG:
 		r0 = bpf_to_rv_reg(BPF_REG_0, ctx);
-		emit(is64 ? rv_addi(RV_REG_T2, r0, 0) :
-		     rv_addiw(RV_REG_T2, r0, 0), ctx);
+		if (is64)
+			emit_mv(RV_REG_T2, r0, ctx);
+		else
+			emit_addiw(RV_REG_T2, r0, 0, ctx);
 		emit(is64 ? rv_lr_d(r0, 0, rd, 0, 0) :
 		     rv_lr_w(r0, 0, rd, 0, 0), ctx);
 		jmp_offset = ninsns_rvoff(8);
-- 
cgit v1.2.3-58-ga151


From 2c1713a8f1c94033a6e00aae4693ab03e8a3b9f1 Mon Sep 17 00:00:00 2001
From: Thomas Weißschuh <linux@weissschuh.net>
Date: Sat, 18 May 2024 16:58:47 +0200
Subject: bpf: constify member bpf_sysctl_kern:: Table
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The sysctl core is preparing to only expose instances of struct ctl_table
as "const". This will also affect the ctl_table argument of sysctl handlers,
for which bpf_sysctl_kern::table is also used.

As the function prototype of all sysctl handlers throughout the tree
needs to stay consistent that change will be done in one commit.

To reduce the size of that final commit, switch this utility type which
is not bound by "typedef proc_handler" to "const struct ctl_table".

No functional change.

Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Joel Granados <j.granados@samsung.com>
Link: https://lore.kernel.org/bpf/20240518-sysctl-const-handler-bpf-v1-1-f0d7186743c1@weissschuh.net
---
 include/linux/filter.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/filter.h b/include/linux/filter.h
index 0f12cf01070e..b02aea291b7e 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -1406,7 +1406,7 @@ struct bpf_sock_ops_kern {
 
 struct bpf_sysctl_kern {
 	struct ctl_table_header *head;
-	struct ctl_table *table;
+	const struct ctl_table *table;
 	void *cur_val;
 	size_t cur_len;
 	void *new_val;
-- 
cgit v1.2.3-58-ga151


From 4f82a8d5c27748a3b4318af99627f4df248133e8 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Tue, 21 May 2024 11:08:10 +0300
Subject: wifi: ath12k: refactor rx descriptor CMEM configuration

Currently, data path Rx descriptor placed in the CMEM memory respective
to the Rx offset. This Rx descriptor CMEM placement is not meeting the
requirements to support inter device MLO. In inter device MLO, multiple
devices participate in a group. The device specific Rx descriptor buffers
transmit to multiple device REO rings. To distinguish between device
specific Rx descriptor buffers, the CMEM setup configuration need to
choose a different placement based on the unique identifier (device index).

Below are the CMEM configuration placement representation:

Intra-Device MLO scenario:

        Device 0                                Device 1

     +-------------+                         +-------------+
     |             |                         |             |
     |     Tx      |                         |     Tx      |
     |             |                         |             |
     +-------------+                         +-------------+
     |             |                         |             |
     |     Rx      |                         |     Rx      |
     |             |                         |             |
     +-------------+                         +-------------+

Inter-Device MLO scenario:

        Device 0                                Device 1

     +-------------+                         +-------------+
     |             |                         |             |
     |     Tx      |                         |     Tx      |
     |             |                         |             |
     +-------------+                         +-------------+
     |             |                         |             |
     |     Rx      |                         |   Reserved  |
     |  Device 0   |                         |             |
     |             |                         |             |
     +-------------+                         +-------------+
     |             |                         |     Rx      |
     |   Reserved  |                         |   Device 1  |
     |             |                         |             |
     +-------------+                         +-------------+

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 HW2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240516000807.1704913-2-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp.c | 12 +++++++++---
 drivers/net/wireless/ath/ath12k/dp.h |  1 +
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp.c b/drivers/net/wireless/ath/ath12k/dp.c
index c04fa3eb57d6..61aa78d8bd8c 100644
--- a/drivers/net/wireless/ath/ath12k/dp.c
+++ b/drivers/net/wireless/ath/ath12k/dp.c
@@ -1349,13 +1349,14 @@ static inline void *ath12k_dp_cc_get_desc_addr_ptr(struct ath12k_base *ab,
 struct ath12k_rx_desc_info *ath12k_dp_get_rx_desc(struct ath12k_base *ab,
 						  u32 cookie)
 {
+	struct ath12k_dp *dp = &ab->dp;
 	struct ath12k_rx_desc_info **desc_addr_ptr;
 	u16 start_ppt_idx, end_ppt_idx, ppt_idx, spt_idx;
 
 	ppt_idx = u32_get_bits(cookie, ATH12K_DP_CC_COOKIE_PPT);
 	spt_idx = u32_get_bits(cookie, ATH12K_DP_CC_COOKIE_SPT);
 
-	start_ppt_idx = ATH12K_RX_SPT_PAGE_OFFSET;
+	start_ppt_idx = dp->rx_ppt_base + ATH12K_RX_SPT_PAGE_OFFSET;
 	end_ppt_idx = start_ppt_idx + ATH12K_NUM_RX_SPT_PAGES;
 
 	if (ppt_idx < start_ppt_idx ||
@@ -1363,6 +1364,7 @@ struct ath12k_rx_desc_info *ath12k_dp_get_rx_desc(struct ath12k_base *ab,
 	    spt_idx > ATH12K_MAX_SPT_ENTRIES)
 		return NULL;
 
+	ppt_idx = ppt_idx - dp->rx_ppt_base;
 	desc_addr_ptr = ath12k_dp_cc_get_desc_addr_ptr(ab, ppt_idx, spt_idx);
 
 	return *desc_addr_ptr;
@@ -1397,7 +1399,7 @@ static int ath12k_dp_cc_desc_init(struct ath12k_base *ab)
 	struct ath12k_rx_desc_info *rx_descs, **rx_desc_addr;
 	struct ath12k_tx_desc_info *tx_descs, **tx_desc_addr;
 	u32 i, j, pool_id, tx_spt_page;
-	u32 ppt_idx;
+	u32 ppt_idx, cookie_ppt_idx;
 
 	spin_lock_bh(&dp->rx_desc_lock);
 
@@ -1412,10 +1414,11 @@ static int ath12k_dp_cc_desc_init(struct ath12k_base *ab)
 		}
 
 		ppt_idx = ATH12K_RX_SPT_PAGE_OFFSET + i;
+		cookie_ppt_idx = dp->rx_ppt_base + ppt_idx;
 		dp->spt_info->rxbaddr[i] = &rx_descs[0];
 
 		for (j = 0; j < ATH12K_MAX_SPT_ENTRIES; j++) {
-			rx_descs[j].cookie = ath12k_dp_cc_cookie_gen(ppt_idx, j);
+			rx_descs[j].cookie = ath12k_dp_cc_cookie_gen(cookie_ppt_idx, j);
 			rx_descs[j].magic = ATH12K_DP_RX_DESC_MAGIC;
 			list_add_tail(&rx_descs[j].list, &dp->rx_desc_free_list);
 
@@ -1476,6 +1479,7 @@ static int ath12k_dp_cmem_init(struct ath12k_base *ab,
 		end = start + ATH12K_NUM_TX_SPT_PAGES;
 		break;
 	case ATH12K_DP_RX_DESC:
+		cmem_base += ATH12K_PPT_ADDR_OFFSET(dp->rx_ppt_base);
 		start = ATH12K_RX_SPT_PAGE_OFFSET;
 		end = start + ATH12K_NUM_RX_SPT_PAGES;
 		break;
@@ -1518,6 +1522,8 @@ static int ath12k_dp_cc_init(struct ath12k_base *ab)
 		return -ENOMEM;
 	}
 
+	dp->rx_ppt_base = ab->device_id * ATH12K_NUM_RX_SPT_PAGES;
+
 	for (i = 0; i < dp->num_spt_pages; i++) {
 		dp->spt_info[i].vaddr = dma_alloc_coherent(ab->dev,
 							   ATH12K_PAGE_SIZE,
diff --git a/drivers/net/wireless/ath/ath12k/dp.h b/drivers/net/wireless/ath/ath12k/dp.h
index 43589c3dd4bc..742094545089 100644
--- a/drivers/net/wireless/ath/ath12k/dp.h
+++ b/drivers/net/wireless/ath/ath12k/dp.h
@@ -350,6 +350,7 @@ struct ath12k_dp {
 	struct ath12k_hp_update_timer tx_ring_timer[DP_TCL_NUM_RING_MAX];
 	struct ath12k_spt_info *spt_info;
 	u32 num_spt_pages;
+	u32 rx_ppt_base;
 	struct list_head rx_desc_free_list;
 	/* protects the free desc list */
 	spinlock_t rx_desc_lock;
-- 
cgit v1.2.3-58-ga151


From 35864b62fa129196af4641b8bef50ac4226a22ed Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Tue, 21 May 2024 11:08:10 +0300
Subject: wifi: ath12k: improve the rx descriptor error information

The ath12k_dp_get_rx_desc() failure log currently contains the same
information across multiple Rx data paths and lacks sufficient detail for
debugging purposes. To address this, change the ath12k_dp_get_rx_desc()
failure log to include cookie information along with Rx path details.
This will provide more specific data for debugging purposes.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 HW2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240516000807.1704913-3-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_rx.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index d8998ac3a482..000df6edf6b4 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -2628,7 +2628,8 @@ try_again:
 		if (!desc_info) {
 			desc_info = ath12k_dp_get_rx_desc(ab, cookie);
 			if (!desc_info) {
-				ath12k_warn(ab, "Invalid cookie in manual desc retrieval");
+				ath12k_warn(ab, "Invalid cookie in manual descriptor retrieval: 0x%x\n",
+					    cookie);
 				continue;
 			}
 		}
@@ -3326,7 +3327,8 @@ ath12k_dp_process_rx_err_buf(struct ath12k *ar, struct hal_reo_dest_ring *desc,
 	if (!desc_info) {
 		desc_info = ath12k_dp_get_rx_desc(ab, cookie);
 		if (!desc_info) {
-			ath12k_warn(ab, "Invalid cookie in manual desc retrieval");
+			ath12k_warn(ab, "Invalid cookie in DP rx error descriptor retrieval: 0x%x\n",
+				    cookie);
 			return -EINVAL;
 		}
 	}
@@ -3745,7 +3747,8 @@ int ath12k_dp_rx_process_wbm_err(struct ath12k_base *ab,
 		if (!desc_info) {
 			desc_info = ath12k_dp_get_rx_desc(ab, err_info.cookie);
 			if (!desc_info) {
-				ath12k_warn(ab, "Invalid cookie in manual desc retrieval");
+				ath12k_warn(ab, "Invalid cookie in DP WBM rx error descriptor retrieval: 0x%x\n",
+					    err_info.cookie);
 				continue;
 			}
 		}
-- 
cgit v1.2.3-58-ga151


From b363614c0c80f9e2f3ffe8e77270e783a0fb8ab5 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Tue, 21 May 2024 11:08:10 +0300
Subject: wifi: ath11k: refactor setting country code logic

ath11k_wmi_send_set_current_country_cmd() is called in several places
and all of them are just simply repeating the same logic.

Refactor to make code clean.

Compile tested only.

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240520024148.5472-2-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath11k/core.c | 19 ++++++++-----------
 drivers/net/wireless/ath/ath11k/mac.c  | 13 +++----------
 drivers/net/wireless/ath/ath11k/reg.c  | 14 ++++++++++----
 drivers/net/wireless/ath/ath11k/reg.h  |  4 ++--
 4 files changed, 23 insertions(+), 27 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/core.c b/drivers/net/wireless/ath/ath11k/core.c
index d4ef556852c2..748e3ad2bec3 100644
--- a/drivers/net/wireless/ath/ath11k/core.c
+++ b/drivers/net/wireless/ath/ath11k/core.c
@@ -1978,23 +1978,20 @@ static void ath11k_update_11d(struct work_struct *work)
 	struct ath11k_base *ab = container_of(work, struct ath11k_base, update_11d_work);
 	struct ath11k *ar;
 	struct ath11k_pdev *pdev;
-	struct wmi_set_current_country_params set_current_param = {};
 	int ret, i;
 
-	spin_lock_bh(&ab->base_lock);
-	memcpy(&set_current_param.alpha2, &ab->new_alpha2, 2);
-	spin_unlock_bh(&ab->base_lock);
-
-	ath11k_dbg(ab, ATH11K_DBG_WMI, "update 11d new cc %c%c\n",
-		   set_current_param.alpha2[0],
-		   set_current_param.alpha2[1]);
-
 	for (i = 0; i < ab->num_radios; i++) {
 		pdev = &ab->pdevs[i];
 		ar = pdev->ar;
 
-		memcpy(&ar->alpha2, &set_current_param.alpha2, 2);
-		ret = ath11k_wmi_send_set_current_country_cmd(ar, &set_current_param);
+		spin_lock_bh(&ab->base_lock);
+		memcpy(&ar->alpha2, &ab->new_alpha2, 2);
+		spin_unlock_bh(&ab->base_lock);
+
+		ath11k_dbg(ab, ATH11K_DBG_WMI, "update 11d new cc %c%c for pdev %d\n",
+			   ar->alpha2[0], ar->alpha2[1], i);
+
+		ret = ath11k_reg_set_cc(ar);
 		if (ret)
 			ath11k_warn(ar->ab,
 				    "pdev id %d failed set current country code: %d\n",
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 17b06c0cd062..84415187416d 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -8864,12 +8864,8 @@ ath11k_mac_op_reconfig_complete(struct ieee80211_hw *hw,
 		ieee80211_wake_queues(ar->hw);
 
 		if (ar->ab->hw_params.current_cc_support &&
-		    ar->alpha2[0] != 0 && ar->alpha2[1] != 0) {
-			struct wmi_set_current_country_params set_current_param = {};
-
-			memcpy(&set_current_param.alpha2, ar->alpha2, 2);
-			ath11k_wmi_send_set_current_country_cmd(ar, &set_current_param);
-		}
+		    ar->alpha2[0] != 0 && ar->alpha2[1] != 0)
+			ath11k_reg_set_cc(ar);
 
 		if (ab->is_reset) {
 			recovery_count = atomic_inc_return(&ab->recovery_count);
@@ -10311,11 +10307,8 @@ static int __ath11k_mac_register(struct ath11k *ar)
 	}
 
 	if (ab->hw_params.current_cc_support && ab->new_alpha2[0]) {
-		struct wmi_set_current_country_params set_current_param = {};
-
-		memcpy(&set_current_param.alpha2, ab->new_alpha2, 2);
 		memcpy(&ar->alpha2, ab->new_alpha2, 2);
-		ret = ath11k_wmi_send_set_current_country_cmd(ar, &set_current_param);
+		ret = ath11k_reg_set_cc(ar);
 		if (ret)
 			ath11k_warn(ar->ab,
 				    "failed set cc code for mac register: %d\n", ret);
diff --git a/drivers/net/wireless/ath/ath11k/reg.c b/drivers/net/wireless/ath/ath11k/reg.c
index 4c3aefbbabbe..b0f289784dd3 100644
--- a/drivers/net/wireless/ath/ath11k/reg.c
+++ b/drivers/net/wireless/ath/ath11k/reg.c
@@ -49,7 +49,6 @@ ath11k_reg_notifier(struct wiphy *wiphy, struct regulatory_request *request)
 {
 	struct ieee80211_hw *hw = wiphy_to_ieee80211_hw(wiphy);
 	struct wmi_init_country_params init_country_param;
-	struct wmi_set_current_country_params set_current_param = {};
 	struct ath11k *ar = hw->priv;
 	int ret;
 
@@ -83,9 +82,8 @@ ath11k_reg_notifier(struct wiphy *wiphy, struct regulatory_request *request)
 	 * reg info
 	 */
 	if (ar->ab->hw_params.current_cc_support) {
-		memcpy(&set_current_param.alpha2, request->alpha2, 2);
-		memcpy(&ar->alpha2, &set_current_param.alpha2, 2);
-		ret = ath11k_wmi_send_set_current_country_cmd(ar, &set_current_param);
+		memcpy(&ar->alpha2, request->alpha2, 2);
+		ret = ath11k_reg_set_cc(ar);
 		if (ret)
 			ath11k_warn(ar->ab,
 				    "failed set current country code: %d\n", ret);
@@ -1017,3 +1015,11 @@ void ath11k_reg_free(struct ath11k_base *ab)
 		kfree(ab->new_regd[i]);
 	}
 }
+
+int ath11k_reg_set_cc(struct ath11k *ar)
+{
+	struct wmi_set_current_country_params set_current_param = {};
+
+	memcpy(&set_current_param.alpha2, ar->alpha2, 2);
+	return ath11k_wmi_send_set_current_country_cmd(ar, &set_current_param);
+}
diff --git a/drivers/net/wireless/ath/ath11k/reg.h b/drivers/net/wireless/ath/ath11k/reg.h
index 64edb794260a..263ea9061948 100644
--- a/drivers/net/wireless/ath/ath11k/reg.h
+++ b/drivers/net/wireless/ath/ath11k/reg.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause-Clear */
 /*
  * Copyright (c) 2019 The Linux Foundation. All rights reserved.
- * Copyright (c) 2022-2023 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2022-2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #ifndef ATH11K_REG_H
@@ -45,5 +45,5 @@ ath11k_reg_ap_pwr_convert(enum ieee80211_ap_reg_power power_type);
 int ath11k_reg_handle_chan_list(struct ath11k_base *ab,
 				struct cur_regulatory_info *reg_info,
 				enum ieee80211_ap_reg_power power_type);
-
+int ath11k_reg_set_cc(struct ath11k *ar);
 #endif
-- 
cgit v1.2.3-58-ga151


From 7f0343b7b8710436c1e6355c71782d32ada47e0c Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Tue, 21 May 2024 11:08:11 +0300
Subject: wifi: ath11k: restore country code during resume

We got report that regdomain is not correct after
return from hibernation:

Before hibernation:
% iw reg get
[...]
phy#0 (self-managed)
country CH: DFS-ETSI
        (2402 - 2482 @ 40), (N/A, 20), (N/A)
        (5170 - 5250 @ 80), (N/A, 23), (N/A), NO-OUTDOOR, AUTO-BW
        (5250 - 5330 @ 80), (N/A, 23), (0 ms), NO-OUTDOOR, DFS, AUTO-BW
        (5490 - 5590 @ 80), (N/A, 30), (0 ms), DFS, AUTO-BW
        (5590 - 5650 @ 40), (N/A, 30), (600000 ms), DFS, AUTO-BW
        (5650 - 5730 @ 80), (N/A, 30), (0 ms), DFS, AUTO-BW
        (5735 - 5875 @ 80), (N/A, 14), (N/A), AUTO-BW

After hibernation:
% iw reg get
[...]
phy#0 (self-managed)
country na: DFS-UNSET
        (2402 - 2472 @ 40), (N/A, 20), (N/A)
        (2457 - 2482 @ 20), (N/A, 20), (N/A), PASSIVE-SCAN
        (5170 - 5330 @ 160), (N/A, 20), (N/A), AUTO-BW, PASSIVE-SCAN
        (5490 - 5730 @ 160), (N/A, 20), (N/A), AUTO-BW, PASSIVE-SCAN
        (5735 - 5895 @ 160), (N/A, 20), (N/A), AUTO-BW, PASSIVE-SCAN
        (5945 - 7125 @ 160), (N/A, 30), (N/A), AUTO-BW, PASSIVE-SCAN

The reason is, during resume, firmware is reinitialized but host does
not send current country code to firmware. So default reg rules with
country code set to 'na' is uploaded to host, as shown above.

Fix it by restoring country code to firmware during resume.

Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.30

Fixes: 166a490f59ac ("wifi: ath11k: support hibernation")
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240520024148.5472-3-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath11k/core.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/net/wireless/ath/ath11k/core.c b/drivers/net/wireless/ath/ath11k/core.c
index 748e3ad2bec3..3772e64d4976 100644
--- a/drivers/net/wireless/ath/ath11k/core.c
+++ b/drivers/net/wireless/ath/ath11k/core.c
@@ -1009,6 +1009,16 @@ int ath11k_core_resume(struct ath11k_base *ab)
 		return -ETIMEDOUT;
 	}
 
+	if (ab->hw_params.current_cc_support &&
+	    ar->alpha2[0] != 0 && ar->alpha2[1] != 0) {
+		ret = ath11k_reg_set_cc(ar);
+		if (ret) {
+			ath11k_warn(ab, "failed to set country code during resume: %d\n",
+				    ret);
+			return ret;
+		}
+	}
+
 	ret = ath11k_dp_rx_pktlog_start(ab);
 	if (ret)
 		ath11k_warn(ab, "failed to start rx pktlog during resume: %d\n",
-- 
cgit v1.2.3-58-ga151


From 33322e3ef07409278a18c6919c448e369d66a18e Mon Sep 17 00:00:00 2001
From: P Praneesh <quic_ppranees@quicinc.com>
Date: Mon, 20 May 2024 12:30:43 +0530
Subject: wifi: ath12k: change DMA direction while mapping reinjected packets
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For fragmented packets, ath12k reassembles each fragment as a normal
packet and then reinjects it into HW ring. In this case, the DMA
direction should be DMA_TO_DEVICE, not DMA_FROM_DEVICE. Otherwise,
an invalid payload may be reinjected into the HW and
subsequently delivered to the host.

Given that arbitrary memory can be allocated to the skb buffer,
knowledge about the data contained in the reinjected buffer is lacking.
Consequently, there’s a risk of private information being leaked.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00209-QCAHKSWPL_SILICONZ-1

Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Co-developed-by: Baochen Qiang <quic_bqiang@quicinc.com>
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Signed-off-by: P Praneesh <quic_ppranees@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240520070045.631029-2-quic_ppranees@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_rx.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index 000df6edf6b4..ffd3f003d003 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -3008,7 +3008,7 @@ static int ath12k_dp_rx_h_defrag_reo_reinject(struct ath12k *ar,
 
 	buf_paddr = dma_map_single(ab->dev, defrag_skb->data,
 				   defrag_skb->len + skb_tailroom(defrag_skb),
-				   DMA_FROM_DEVICE);
+				   DMA_TO_DEVICE);
 	if (dma_mapping_error(ab->dev, buf_paddr))
 		return -ENOMEM;
 
@@ -3094,7 +3094,7 @@ err_free_desc:
 	spin_unlock_bh(&dp->rx_desc_lock);
 err_unmap_dma:
 	dma_unmap_single(ab->dev, buf_paddr, defrag_skb->len + skb_tailroom(defrag_skb),
-			 DMA_FROM_DEVICE);
+			 DMA_TO_DEVICE);
 	return ret;
 }
 
-- 
cgit v1.2.3-58-ga151


From 073f9f249eecd64ab9d59c91c4a23cfdcc02afe4 Mon Sep 17 00:00:00 2001
From: P Praneesh <quic_ppranees@quicinc.com>
Date: Mon, 20 May 2024 12:30:44 +0530
Subject: wifi: ath12k: fix invalid memory access while processing fragmented
 packets

The monitor ring and the reo reinject ring share the same ring mask index.
When the driver receives an interrupt for the reo reinject ring, the
monitor ring is also processed, leading to invalid memory access. Since
monitor support is not yet enabled in ath12k, the ring mask for the monitor
ring should be removed.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00209-QCAHKSWPL_SILICONZ-1

Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Signed-off-by: P Praneesh <quic_ppranees@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240520070045.631029-3-quic_ppranees@quicinc.com
---
 drivers/net/wireless/ath/ath12k/hw.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/hw.c b/drivers/net/wireless/ath/ath12k/hw.c
index b45f929a0204..2e11ea763574 100644
--- a/drivers/net/wireless/ath/ath12k/hw.c
+++ b/drivers/net/wireless/ath/ath12k/hw.c
@@ -544,9 +544,6 @@ static const struct ath12k_hw_ring_mask ath12k_hw_ring_mask_qcn9274 = {
 	},
 	.rx_mon_dest = {
 		0, 0, 0,
-		ATH12K_RX_MON_RING_MASK_0,
-		ATH12K_RX_MON_RING_MASK_1,
-		ATH12K_RX_MON_RING_MASK_2,
 	},
 	.rx = {
 		0, 0, 0, 0,
@@ -572,8 +569,7 @@ static const struct ath12k_hw_ring_mask ath12k_hw_ring_mask_qcn9274 = {
 		ATH12K_HOST2RXDMA_RING_MASK_0,
 	},
 	.tx_mon_dest = {
-		ATH12K_TX_MON_RING_MASK_0,
-		ATH12K_TX_MON_RING_MASK_1,
+		0, 0, 0,
 	},
 };
 
-- 
cgit v1.2.3-58-ga151


From a57ab7cced454f69b8ee8aa5f5019ea8de4674da Mon Sep 17 00:00:00 2001
From: P Praneesh <quic_ppranees@quicinc.com>
Date: Mon, 20 May 2024 12:30:45 +0530
Subject: wifi: ath12k: fix firmware crash during reo reinject
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When handling fragmented packets, the ath12k driver reassembles each
fragment into a normal packet and then reinjects it into the HW ring.
However, a firmware crash occurs during this reinjection process.
The issue arises because the driver populates peer metadata in
reo_ent_ring->queue_addr_lo, while the firmware expects the physical
address obtained from the corresponding peer’s queue descriptor. Fix it
by filling peer's queue descriptor's physical address in queue_addr_lo.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00209-QCAHKSWPL_SILICONZ-1

Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Signed-off-by: P Praneesh <quic_ppranees@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240520070045.631029-4-quic_ppranees@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_rx.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index ffd3f003d003..a52b7b273540 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -2971,7 +2971,7 @@ static int ath12k_dp_rx_h_defrag_reo_reinject(struct ath12k *ar,
 	struct hal_srng *srng;
 	dma_addr_t link_paddr, buf_paddr;
 	u32 desc_bank, msdu_info, msdu_ext_info, mpdu_info;
-	u32 cookie, hal_rx_desc_sz, dest_ring_info0;
+	u32 cookie, hal_rx_desc_sz, dest_ring_info0, queue_addr_hi;
 	int ret;
 	struct ath12k_rx_desc_info *desc_info;
 	enum hal_rx_buf_return_buf_manager idle_link_rbm = dp->idle_link_rbm;
@@ -3064,13 +3064,11 @@ static int ath12k_dp_rx_h_defrag_reo_reinject(struct ath12k *ar,
 	reo_ent_ring->rx_mpdu_info.peer_meta_data =
 		reo_dest_ring->rx_mpdu_info.peer_meta_data;
 
-	/* Firmware expects physical address to be filled in queue_addr_lo in
-	 * the MLO scenario and in case of non MLO peer meta data needs to be
-	 * filled.
-	 * TODO: Need to handle for MLO scenario.
-	 */
-	reo_ent_ring->queue_addr_lo = reo_dest_ring->rx_mpdu_info.peer_meta_data;
-	reo_ent_ring->info0 = le32_encode_bits(dst_ind,
+	reo_ent_ring->queue_addr_lo = cpu_to_le32(lower_32_bits(rx_tid->paddr));
+	queue_addr_hi = upper_32_bits(rx_tid->paddr);
+	reo_ent_ring->info0 = le32_encode_bits(queue_addr_hi,
+					       HAL_REO_ENTR_RING_INFO0_QUEUE_ADDR_HI) |
+			      le32_encode_bits(dst_ind,
 					       HAL_REO_ENTR_RING_INFO0_DEST_IND);
 
 	reo_ent_ring->info1 = le32_encode_bits(rx_tid->cur_sn,
-- 
cgit v1.2.3-58-ga151


From 4652072e7b9d643edc9ebb04e3e2c021461b7af0 Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler1968@googlemail.com>
Date: Fri, 17 May 2024 08:34:45 -0700
Subject: bpf, docs: Move sentence about returning R0 to abi.rst

As discussed at LSF/MM/BPF, the sentence about using R0 for returning
values from calls is part of the calling convention and belongs in
abi.rst.  Any further additions or clarifications to this text are left
for future patches on abi.rst.  The current patch is simply to unblock
progression of instruction-set.rst to a standard.

In contrast, the restriction of register numbers to the range 0-10
is untouched, left in the instruction-set.rst definition of the
src_reg and dst_reg fields.

Signed-off-by: Dave Thaler <dthaler1968@googlemail.com>
Link: https://lore.kernel.org/r/20240517153445.3914-1-dthaler1968@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/standardization/abi.rst             | 3 +++
 Documentation/bpf/standardization/instruction-set.rst | 3 ---
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/Documentation/bpf/standardization/abi.rst b/Documentation/bpf/standardization/abi.rst
index 0c2e10eeb89a..41514137cb7b 100644
--- a/Documentation/bpf/standardization/abi.rst
+++ b/Documentation/bpf/standardization/abi.rst
@@ -23,3 +23,6 @@ The BPF calling convention is defined as:
 
 R0 - R5 are scratch registers and BPF programs needs to spill/fill them if
 necessary across calls.
+
+The BPF program needs to store the return value into register R0 before doing an
+``EXIT``.
diff --git a/Documentation/bpf/standardization/instruction-set.rst b/Documentation/bpf/standardization/instruction-set.rst
index 00c93eb42613..e2f6c0a4a559 100644
--- a/Documentation/bpf/standardization/instruction-set.rst
+++ b/Documentation/bpf/standardization/instruction-set.rst
@@ -476,9 +476,6 @@ the jump instruction.  Thus 'PC += 1' skips execution of the next
 instruction if it's a basic instruction or results in undefined behavior
 if the next instruction is a 128-bit wide instruction.
 
-The BPF program needs to store the return value into register R0 before doing an
-``EXIT``.
-
 Example:
 
 ``{JSGE, X, JMP32}`` means::
-- 
cgit v1.2.3-58-ga151


From a985fdca5e7e665d58dc40c92a67c8b67b6291db Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler1968@googlemail.com>
Date: Fri, 17 May 2024 09:58:55 -0700
Subject: bpf, docs: Use RFC 2119 language for ISA requirements

Per IETF convention and discussion at LSF/MM/BPF, use MUST etc.
keywords as requested by IETF Area Director review.  Also as
requested, indicate that documenting BTF is out of scope of this
document and will be covered by a separate IETF specification.

Added paragraph about the terminology that is required IETF boilerplate
and must be worded exactly as such.

Signed-off-by: Dave Thaler <dthaler1968@googlemail.com>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20240517165855.4688-1-dthaler1968@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../bpf/standardization/instruction-set.rst        | 24 ++++++++++++++--------
 1 file changed, 16 insertions(+), 8 deletions(-)

diff --git a/Documentation/bpf/standardization/instruction-set.rst b/Documentation/bpf/standardization/instruction-set.rst
index e2f6c0a4a559..02e2490433c6 100644
--- a/Documentation/bpf/standardization/instruction-set.rst
+++ b/Documentation/bpf/standardization/instruction-set.rst
@@ -14,6 +14,13 @@ set architecture (ISA).
 Documentation conventions
 =========================
 
+The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
+"SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
+"OPTIONAL" in this document are to be interpreted as described in
+BCP 14 `<https://www.rfc-editor.org/info/rfc2119>`_
+`RFC8174 <https://www.rfc-editor.org/info/rfc8174>`_
+when, and only when, they appear in all capitals, as shown here.
+
 For brevity and consistency, this document refers to families
 of types using a shorthand syntax and refers to several expository,
 mnemonic functions when describing the semantics of instructions.
@@ -106,9 +113,9 @@ Conformance groups
 
 An implementation does not need to support all instructions specified in this
 document (e.g., deprecated instructions).  Instead, a number of conformance
-groups are specified.  An implementation must support the base32 conformance
-group and may support additional conformance groups, where supporting a
-conformance group means it must support all instructions in that conformance
+groups are specified.  An implementation MUST support the base32 conformance
+group and MAY support additional conformance groups, where supporting a
+conformance group means it MUST support all instructions in that conformance
 group.
 
 The use of named conformance groups enables interoperability between a runtime
@@ -209,7 +216,7 @@ For example::
   07     1       0        00 00  11 22 33 44  r1 += 0x11223344 // big
 
 Note that most instructions do not use all of the fields.
-Unused fields shall be cleared to zero.
+Unused fields SHALL be cleared to zero.
 
 Wide instruction encoding
 --------------------------
@@ -374,7 +381,7 @@ interpreted as a 64-bit signed value.
 Note that there are varying definitions of the signed modulo operation
 when the dividend or divisor are negative, where implementations often
 vary by language such that Python, Ruby, etc.  differ from C, Go, Java,
-etc. This specification requires that signed modulo use truncated division
+etc. This specification requires that signed modulo MUST use truncated division
 (where -13 % 3 == -1) as implemented in C, Go, etc.::
 
    a % n = a - n * trunc(a / n)
@@ -404,7 +411,7 @@ only and do not use a separate source register or immediate value.
 For ``ALU``, the 1-bit source operand field in the opcode is used to
 select what byte order the operation converts from or to. For
 ``ALU64``, the 1-bit source operand field in the opcode is reserved
-and must be set to 0.
+and MUST be set to 0.
 
 =====  ========  =====  =================================================
 class  source    value  description
@@ -512,7 +519,8 @@ for each program type, but static IDs are unique across all program types.
 
 Platforms that support the BPF Type Format (BTF) support identifying
 a helper function by a BTF ID encoded in the 'imm' field, where the BTF ID
-identifies the helper name and type.
+identifies the helper name and type.  Further documentation of BTF
+is outside the scope of this document and is left for future work.
 
 Program-local functions
 ~~~~~~~~~~~~~~~~~~~~~~~
@@ -722,5 +730,5 @@ carried over from classic BPF. These instructions used an instruction
 class of ``LD``, a size modifier of ``W``, ``H``, or ``B``, and a
 mode modifier of ``ABS`` or ``IND``.  The 'dst_reg' and 'offset' fields were
 set to zero, and 'src_reg' was set to zero for ``ABS``.  However, these
-instructions are deprecated and should no longer be used.  All legacy packet
+instructions are deprecated and SHOULD no longer be used.  All legacy packet
 access instructions belong to the "packet" conformance group.
-- 
cgit v1.2.3-58-ga151


From 4e1215d9a1903fc9e976aa8903674d050c7af5ff Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler1968@googlemail.com>
Date: Mon, 20 May 2024 14:52:55 -0700
Subject: bpf, docs: clarify sign extension of 64-bit use of 32-bit imm

imm is defined as a 32-bit signed integer.

{MOV, K, ALU64} says it does "dst = src" (where src is 'imm') and it
does do dst = (s64)imm, which in that sense does sign extend imm. The MOVSX
instruction is explained as sign extending, so added the example of
{MOV, K, ALU64} to make this more clear.

{JLE, K, JMP} says it does "PC += offset if dst <= src" (where src is 'imm',
and the comparison is unsigned). This was apparently ambiguous to some
readers as to whether the comparison was "dst <= (u64)(u32)imm" or
"dst <= (u64)(s64)imm" so added an example to make this more clear.

v1 -> v2: Address comments from Yonghong

Signed-off-by: Dave Thaler <dthaler1968@googlemail.com>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/r/20240520215255.10595-1-dthaler1968@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/standardization/instruction-set.rst | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/Documentation/bpf/standardization/instruction-set.rst b/Documentation/bpf/standardization/instruction-set.rst
index 02e2490433c6..6b11d1109f12 100644
--- a/Documentation/bpf/standardization/instruction-set.rst
+++ b/Documentation/bpf/standardization/instruction-set.rst
@@ -393,6 +393,19 @@ The ``MOVSX`` instruction does a move operation with sign extension.
 operands into 64-bit operands.  Unlike other arithmetic instructions,
 ``MOVSX`` is only defined for register source operands (``X``).
 
+``{MOV, K, ALU64}`` means::
+
+  dst = (s64)imm
+
+``{MOV, X, ALU}`` means::
+
+  dst = (u32)src
+
+``{MOVSX, X, ALU}`` with 'offset' 8 means::
+
+  dst = (u32)(s32)(s8)src
+
+
 The ``NEG`` instruction is only defined when the source bit is clear
 (``K``).
 
@@ -491,6 +504,10 @@ Example:
 
 where 's>=' indicates a signed '>=' comparison.
 
+``{JLE, K, JMP}`` means::
+
+  if dst <= (u64)(s64)imm goto +offset
+
 ``{JA, K, JMP32}`` means::
 
   gotol +imm
-- 
cgit v1.2.3-58-ga151


From 6a6d8b6f00ade597e0030669fae3fdf57cfba33b Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler1968@googlemail.com>
Date: Fri, 24 May 2024 09:46:18 -0700
Subject: bpf, docs: Add table captions

As suggested by Ines Robles in his IETF GENART review at
https://datatracker.ietf.org/doc/review-ietf-bpf-isa-02-genart-lc-robles-2024-05-16/

Signed-off-by: Dave Thaler <dthaler1968@gmail.com>
Link: https://lore.kernel.org/r/20240524164618.18894-1-dthaler1968@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../bpf/standardization/instruction-set.rst        | 184 ++++++++++++---------
 1 file changed, 102 insertions(+), 82 deletions(-)

diff --git a/Documentation/bpf/standardization/instruction-set.rst b/Documentation/bpf/standardization/instruction-set.rst
index 6b11d1109f12..1602454fe7cf 100644
--- a/Documentation/bpf/standardization/instruction-set.rst
+++ b/Documentation/bpf/standardization/instruction-set.rst
@@ -32,7 +32,7 @@ Types
 This document refers to integer types with the notation `SN` to specify
 a type's signedness (`S`) and bit width (`N`), respectively.
 
-.. table:: Meaning of signedness notation.
+.. table:: Meaning of signedness notation
 
   ==== =========
   S    Meaning
@@ -41,7 +41,7 @@ a type's signedness (`S`) and bit width (`N`), respectively.
   s    signed
   ==== =========
 
-.. table:: Meaning of bit-width notation.
+.. table:: Meaning of bit-width notation
 
   ===== =========
   N     Bit width
@@ -263,18 +263,20 @@ Instruction classes
 
 The three least significant bits of the 'opcode' field store the instruction class:
 
-=====  =====  ===============================  ===================================
-class  value  description                      reference
-=====  =====  ===============================  ===================================
-LD     0x0    non-standard load operations     `Load and store instructions`_
-LDX    0x1    load into register operations    `Load and store instructions`_
-ST     0x2    store from immediate operations  `Load and store instructions`_
-STX    0x3    store from register operations   `Load and store instructions`_
-ALU    0x4    32-bit arithmetic operations     `Arithmetic and jump instructions`_
-JMP    0x5    64-bit jump operations           `Arithmetic and jump instructions`_
-JMP32  0x6    32-bit jump operations           `Arithmetic and jump instructions`_
-ALU64  0x7    64-bit arithmetic operations     `Arithmetic and jump instructions`_
-=====  =====  ===============================  ===================================
+.. table:: Instruction class
+
+  =====  =====  ===============================  ===================================
+  class  value  description                      reference
+  =====  =====  ===============================  ===================================
+  LD     0x0    non-standard load operations     `Load and store instructions`_
+  LDX    0x1    load into register operations    `Load and store instructions`_
+  ST     0x2    store from immediate operations  `Load and store instructions`_
+  STX    0x3    store from register operations   `Load and store instructions`_
+  ALU    0x4    32-bit arithmetic operations     `Arithmetic and jump instructions`_
+  JMP    0x5    64-bit jump operations           `Arithmetic and jump instructions`_
+  JMP32  0x6    32-bit jump operations           `Arithmetic and jump instructions`_
+  ALU64  0x7    64-bit arithmetic operations     `Arithmetic and jump instructions`_
+  =====  =====  ===============================  ===================================
 
 Arithmetic and jump instructions
 ================================
@@ -292,6 +294,8 @@ For arithmetic and jump instructions (``ALU``, ``ALU64``, ``JMP`` and
 **s (source)**
   the source operand location, which unless otherwise specified is one of:
 
+  .. table:: Source operand location
+
   ======  =====  ==============================================
   source  value  description
   ======  =====  ==============================================
@@ -312,27 +316,29 @@ The 'code' field encodes the operation as below, where 'src' refers to the
 the source operand and 'dst' refers to the value of the destination
 register.
 
-=====  =====  =======  ==========================================================
-name   code   offset   description
-=====  =====  =======  ==========================================================
-ADD    0x0    0        dst += src
-SUB    0x1    0        dst -= src
-MUL    0x2    0        dst \*= src
-DIV    0x3    0        dst = (src != 0) ? (dst / src) : 0
-SDIV   0x3    1        dst = (src != 0) ? (dst s/ src) : 0
-OR     0x4    0        dst \|= src
-AND    0x5    0        dst &= src
-LSH    0x6    0        dst <<= (src & mask)
-RSH    0x7    0        dst >>= (src & mask)
-NEG    0x8    0        dst = -dst
-MOD    0x9    0        dst = (src != 0) ? (dst % src) : dst
-SMOD   0x9    1        dst = (src != 0) ? (dst s% src) : dst
-XOR    0xa    0        dst ^= src
-MOV    0xb    0        dst = src
-MOVSX  0xb    8/16/32  dst = (s8,s16,s32)src
-ARSH   0xc    0        :term:`sign extending<Sign Extend>` dst >>= (src & mask)
-END    0xd    0        byte swap operations (see `Byte swap instructions`_ below)
-=====  =====  =======  ==========================================================
+.. table:: Arithmetic instructions
+
+  =====  =====  =======  ==========================================================
+  name   code   offset   description
+  =====  =====  =======  ==========================================================
+  ADD    0x0    0        dst += src
+  SUB    0x1    0        dst -= src
+  MUL    0x2    0        dst \*= src
+  DIV    0x3    0        dst = (src != 0) ? (dst / src) : 0
+  SDIV   0x3    1        dst = (src != 0) ? (dst s/ src) : 0
+  OR     0x4    0        dst \|= src
+  AND    0x5    0        dst &= src
+  LSH    0x6    0        dst <<= (src & mask)
+  RSH    0x7    0        dst >>= (src & mask)
+  NEG    0x8    0        dst = -dst
+  MOD    0x9    0        dst = (src != 0) ? (dst % src) : dst
+  SMOD   0x9    1        dst = (src != 0) ? (dst s% src) : dst
+  XOR    0xa    0        dst ^= src
+  MOV    0xb    0        dst = src
+  MOVSX  0xb    8/16/32  dst = (s8,s16,s32)src
+  ARSH   0xc    0        :term:`sign extending<Sign Extend>` dst >>= (src & mask)
+  END    0xd    0        byte swap operations (see `Byte swap instructions`_ below)
+  =====  =====  =======  ==========================================================
 
 Underflow and overflow are allowed during arithmetic operations, meaning
 the 64-bit or 32-bit value will wrap. If BPF program execution would
@@ -426,13 +432,15 @@ select what byte order the operation converts from or to. For
 ``ALU64``, the 1-bit source operand field in the opcode is reserved
 and MUST be set to 0.
 
-=====  ========  =====  =================================================
-class  source    value  description
-=====  ========  =====  =================================================
-ALU    TO_LE     0      convert between host byte order and little endian
-ALU    TO_BE     1      convert between host byte order and big endian
-ALU64  Reserved  0      do byte swap unconditionally
-=====  ========  =====  =================================================
+.. table:: Byte swap instructions
+
+  =====  ========  =====  =================================================
+  class  source    value  description
+  =====  ========  =====  =================================================
+  ALU    TO_LE     0      convert between host byte order and little endian
+  ALU    TO_BE     1      convert between host byte order and big endian
+  ALU64  Reserved  0      do byte swap unconditionally
+  =====  ========  =====  =================================================
 
 The 'imm' field encodes the width of the swap operations.  The following widths
 are supported: 16, 32 and 64.  Width 64 operations belong to the base64
@@ -468,27 +476,29 @@ otherwise identical operations, and indicates the base64 conformance
 group unless otherwise specified.
 The 'code' field encodes the operation as below:
 
-========  =====  =======  =================================  ===================================================
-code      value  src_reg  description                        notes
-========  =====  =======  =================================  ===================================================
-JA        0x0    0x0      PC += offset                       {JA, K, JMP} only
-JA        0x0    0x0      PC += imm                          {JA, K, JMP32} only
-JEQ       0x1    any      PC += offset if dst == src
-JGT       0x2    any      PC += offset if dst > src          unsigned
-JGE       0x3    any      PC += offset if dst >= src         unsigned
-JSET      0x4    any      PC += offset if dst & src
-JNE       0x5    any      PC += offset if dst != src
-JSGT      0x6    any      PC += offset if dst > src          signed
-JSGE      0x7    any      PC += offset if dst >= src         signed
-CALL      0x8    0x0      call helper function by static ID  {CALL, K, JMP} only, see `Helper functions`_
-CALL      0x8    0x1      call PC += imm                     {CALL, K, JMP} only, see `Program-local functions`_
-CALL      0x8    0x2      call helper function by BTF ID     {CALL, K, JMP} only, see `Helper functions`_
-EXIT      0x9    0x0      return                             {CALL, K, JMP} only
-JLT       0xa    any      PC += offset if dst < src          unsigned
-JLE       0xb    any      PC += offset if dst <= src         unsigned
-JSLT      0xc    any      PC += offset if dst < src          signed
-JSLE      0xd    any      PC += offset if dst <= src         signed
-========  =====  =======  =================================  ===================================================
+.. table:: Jump instructions
+
+  ========  =====  =======  =================================  ===================================================
+  code      value  src_reg  description                        notes
+  ========  =====  =======  =================================  ===================================================
+  JA        0x0    0x0      PC += offset                       {JA, K, JMP} only
+  JA        0x0    0x0      PC += imm                          {JA, K, JMP32} only
+  JEQ       0x1    any      PC += offset if dst == src
+  JGT       0x2    any      PC += offset if dst > src          unsigned
+  JGE       0x3    any      PC += offset if dst >= src         unsigned
+  JSET      0x4    any      PC += offset if dst & src
+  JNE       0x5    any      PC += offset if dst != src
+  JSGT      0x6    any      PC += offset if dst > src          signed
+  JSGE      0x7    any      PC += offset if dst >= src         signed
+  CALL      0x8    0x0      call helper function by static ID  {CALL, K, JMP} only, see `Helper functions`_
+  CALL      0x8    0x1      call PC += imm                     {CALL, K, JMP} only, see `Program-local functions`_
+  CALL      0x8    0x2      call helper function by BTF ID     {CALL, K, JMP} only, see `Helper functions`_
+  EXIT      0x9    0x0      return                             {CALL, K, JMP} only
+  JLT       0xa    any      PC += offset if dst < src          unsigned
+  JLE       0xb    any      PC += offset if dst <= src         unsigned
+  JSLT      0xc    any      PC += offset if dst < src          signed
+  JSLE      0xd    any      PC += offset if dst <= src         signed
+  ========  =====  =======  =================================  ===================================================
 
 where 'PC' denotes the program counter, and the offset to increment by
 is in units of 64-bit instructions relative to the instruction following
@@ -559,6 +569,8 @@ For load and store instructions (``LD``, ``LDX``, ``ST``, and ``STX``), the
 **mode**
   The mode modifier is one of:
 
+  .. table:: Mode modifier
+
     =============  =====  ====================================  =============
     mode modifier  value  description                           reference
     =============  =====  ====================================  =============
@@ -573,6 +585,8 @@ For load and store instructions (``LD``, ``LDX``, ``ST``, and ``STX``), the
 **sz (size)**
   The size modifier is one of:
 
+  .. table:: Size modifier
+
     ====  =====  =====================
     size  value  description
     ====  =====  =====================
@@ -641,14 +655,16 @@ The 'imm' field is used to encode the actual atomic operation.
 Simple atomic operation use a subset of the values defined to encode
 arithmetic operations in the 'imm' field to encode the atomic operation:
 
-========  =====  ===========
-imm       value  description
-========  =====  ===========
-ADD       0x00   atomic add
-OR        0x40   atomic or
-AND       0x50   atomic and
-XOR       0xa0   atomic xor
-========  =====  ===========
+.. table:: Simple atomic operations
+
+  ========  =====  ===========
+  imm       value  description
+  ========  =====  ===========
+  ADD       0x00   atomic add
+  OR        0x40   atomic or
+  AND       0x50   atomic and
+  XOR       0xa0   atomic xor
+  ========  =====  ===========
 
 
 ``{ATOMIC, W, STX}`` with 'imm' = ADD means::
@@ -662,6 +678,8 @@ XOR       0xa0   atomic xor
 In addition to the simple atomic operations, there also is a modifier and
 two complex atomic operations:
 
+.. table:: Complex atomic operations
+
 ===========  ================  ===========================
 imm          value             description
 ===========  ================  ===========================
@@ -695,17 +713,19 @@ The following table defines a set of ``{IMM, DW, LD}`` instructions
 with opcode subtypes in the 'src_reg' field, using new terms such as "map"
 defined further below:
 
-=======  =========================================  ===========  ==============
-src_reg  pseudocode                                 imm type     dst type
-=======  =========================================  ===========  ==============
-0x0      dst = (next_imm << 32) | imm               integer      integer
-0x1      dst = map_by_fd(imm)                       map fd       map
-0x2      dst = map_val(map_by_fd(imm)) + next_imm   map fd       data address
-0x3      dst = var_addr(imm)                        variable id  data address
-0x4      dst = code_addr(imm)                       integer      code address
-0x5      dst = map_by_idx(imm)                      map index    map
-0x6      dst = map_val(map_by_idx(imm)) + next_imm  map index    data address
-=======  =========================================  ===========  ==============
+.. table:: 64-bit immediate instructions
+
+  =======  =========================================  ===========  ==============
+  src_reg  pseudocode                                 imm type     dst type
+  =======  =========================================  ===========  ==============
+  0x0      dst = (next_imm << 32) | imm               integer      integer
+  0x1      dst = map_by_fd(imm)                       map fd       map
+  0x2      dst = map_val(map_by_fd(imm)) + next_imm   map fd       data address
+  0x3      dst = var_addr(imm)                        variable id  data address
+  0x4      dst = code_addr(imm)                       integer      code address
+  0x5      dst = map_by_idx(imm)                      map index    map
+  0x6      dst = map_val(map_by_idx(imm)) + next_imm  map index    data address
+  =======  =========================================  ===========  ==============
 
 where
 
-- 
cgit v1.2.3-58-ga151


From f980f13e4eb299abba6692365315196e1ba6fd2c Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler1968@googlemail.com>
Date: Sat, 25 May 2024 08:33:32 -0700
Subject: bpf, docs: Clarify call local offset

In the Jump instructions section it explains that the offset is
"relative to the instruction following the jump instruction".
But the program-local section confusingly said "referenced by
offset from the call instruction, similar to JA".

This patch updates that sentence with consistent wording, saying
it's relative to the instruction following the call instruction.

Signed-off-by: Dave Thaler <dthaler1968@gmail.com>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20240525153332.21355-1-dthaler1968@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 Documentation/bpf/standardization/instruction-set.rst | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/Documentation/bpf/standardization/instruction-set.rst b/Documentation/bpf/standardization/instruction-set.rst
index 1602454fe7cf..08f614b10a56 100644
--- a/Documentation/bpf/standardization/instruction-set.rst
+++ b/Documentation/bpf/standardization/instruction-set.rst
@@ -552,9 +552,10 @@ is outside the scope of this document and is left for future work.
 Program-local functions
 ~~~~~~~~~~~~~~~~~~~~~~~
 Program-local functions are functions exposed by the same BPF program as the
-caller, and are referenced by offset from the call instruction, similar to
-``JA``.  The offset is encoded in the 'imm' field of the call instruction.
-An ``EXIT`` within the program-local function will return to the caller.
+caller, and are referenced by offset from the instruction following the call
+instruction, similar to ``JA``.  The offset is encoded in the 'imm' field of
+the call instruction. An ``EXIT`` within the program-local function will
+return to the caller.
 
 Load and store instructions
 ===========================
-- 
cgit v1.2.3-58-ga151


From e245ef8a0b06fa5a18ca3bdcf82430166419283e Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler1968@googlemail.com>
Date: Sat, 25 May 2024 23:18:15 -0700
Subject: bpf, docs: Fix instruction.rst indentation

The table captions patch corrected indented most tables to work with
the table directive for adding a caption but missed two of them.

Signed-off-by: Dave Thaler <dthaler1968@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20240526061815.22497-1-dthaler1968@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../bpf/standardization/instruction-set.rst        | 26 +++++++++++-----------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/Documentation/bpf/standardization/instruction-set.rst b/Documentation/bpf/standardization/instruction-set.rst
index 08f614b10a56..8d19810504b8 100644
--- a/Documentation/bpf/standardization/instruction-set.rst
+++ b/Documentation/bpf/standardization/instruction-set.rst
@@ -296,12 +296,12 @@ For arithmetic and jump instructions (``ALU``, ``ALU64``, ``JMP`` and
 
   .. table:: Source operand location
 
-  ======  =====  ==============================================
-  source  value  description
-  ======  =====  ==============================================
-  K       0      use 32-bit 'imm' value as source operand
-  X       1      use 'src_reg' register value as source operand
-  ======  =====  ==============================================
+    ======  =====  ==============================================
+    source  value  description
+    ======  =====  ==============================================
+    K       0      use 32-bit 'imm' value as source operand
+    X       1      use 'src_reg' register value as source operand
+    ======  =====  ==============================================
 
 **instruction class**
   the instruction class (see `Instruction classes`_)
@@ -681,13 +681,13 @@ two complex atomic operations:
 
 .. table:: Complex atomic operations
 
-===========  ================  ===========================
-imm          value             description
-===========  ================  ===========================
-FETCH        0x01              modifier: return old value
-XCHG         0xe0 | FETCH      atomic exchange
-CMPXCHG      0xf0 | FETCH      atomic compare and exchange
-===========  ================  ===========================
+  ===========  ================  ===========================
+  imm          value             description
+  ===========  ================  ===========================
+  FETCH        0x01              modifier: return old value
+  XCHG         0xe0 | FETCH      atomic exchange
+  CMPXCHG      0xf0 | FETCH      atomic compare and exchange
+  ===========  ================  ===========================
 
 The ``FETCH`` modifier is optional for simple atomic operations, and
 always set for the complex atomic operations.  If the ``FETCH`` flag
-- 
cgit v1.2.3-58-ga151


From 9e305a6f01ad539dc134b8017521495378d8e00e Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Tue, 21 May 2024 12:01:39 +0800
Subject: wifi: rtw89: 8852c: correct logic and restore PCI PHY EQ after device
 resume

PCI PHY EQ value is missing after card off/on, so update the value after
device resume. The original commit only updates once at probe stage, which
could lead problem after suspend/resume.

The logic should be read a value from one register and write to another
register with a mask to avoid affecting unrelated bits.

Fixes: a78d33a1286c ("wifi: rtw89: 8852c: disable PCI PHY EQ to improve compatibility")
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240521040139.20311-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/pci.c | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/pci.c b/drivers/net/wireless/realtek/rtw89/pci.c
index f93385cc20cd..dbbb2f6500e2 100644
--- a/drivers/net/wireless/realtek/rtw89/pci.c
+++ b/drivers/net/wireless/realtek/rtw89/pci.c
@@ -2330,21 +2330,20 @@ static void rtw89_pci_disable_eq(struct rtw89_dev *rtwdev)
 	u32 backup_aspm;
 	u32 phy_offset;
 	u16 oobs_val;
-	u16 val16;
 	int ret;
 
 	if (rtwdev->chip->chip_id != RTL8852C)
 		return;
 
-	backup_aspm = rtw89_read32(rtwdev, R_AX_PCIE_MIX_CFG_V1);
-	rtw89_write32_clr(rtwdev, R_AX_PCIE_MIX_CFG_V1, B_AX_ASPM_CTRL_MASK);
-
 	g1_oobs = rtw89_read16_mask(rtwdev, R_RAC_DIRECT_OFFSET_G1 +
 					    RAC_ANA09 * RAC_MULT, BAC_OOBS_SEL);
 	g2_oobs = rtw89_read16_mask(rtwdev, R_RAC_DIRECT_OFFSET_G2 +
 					    RAC_ANA09 * RAC_MULT, BAC_OOBS_SEL);
 	if (g1_oobs && g2_oobs)
-		goto out;
+		return;
+
+	backup_aspm = rtw89_read32(rtwdev, R_AX_PCIE_MIX_CFG_V1);
+	rtw89_write32_clr(rtwdev, R_AX_PCIE_MIX_CFG_V1, B_AX_ASPM_CTRL_MASK);
 
 	ret = rtw89_pci_get_phy_offset_by_link_speed(rtwdev, &phy_offset);
 	if (ret)
@@ -2354,15 +2353,16 @@ static void rtw89_pci_disable_eq(struct rtw89_dev *rtwdev)
 	rtw89_write16(rtwdev, phy_offset + RAC_ANA10 * RAC_MULT, ADDR_SEL_PINOUT_DIS_VAL);
 	rtw89_write16_set(rtwdev, phy_offset + RAC_ANA19 * RAC_MULT, B_PCIE_BIT_RD_SEL);
 
-	val16 = rtw89_read16_mask(rtwdev, phy_offset + RAC_ANA1F * RAC_MULT,
-				  OOBS_LEVEL_MASK);
-	oobs_val = u16_encode_bits(val16, OOBS_SEN_MASK);
+	oobs_val = rtw89_read16_mask(rtwdev, phy_offset + RAC_ANA1F * RAC_MULT,
+				     OOBS_LEVEL_MASK);
 
-	rtw89_write16(rtwdev, R_RAC_DIRECT_OFFSET_G1 + RAC_ANA03 * RAC_MULT, oobs_val);
+	rtw89_write16_mask(rtwdev, R_RAC_DIRECT_OFFSET_G1 + RAC_ANA03 * RAC_MULT,
+			   OOBS_SEN_MASK, oobs_val);
 	rtw89_write16_set(rtwdev, R_RAC_DIRECT_OFFSET_G1 + RAC_ANA09 * RAC_MULT,
 			  BAC_OOBS_SEL);
 
-	rtw89_write16(rtwdev, R_RAC_DIRECT_OFFSET_G2 + RAC_ANA03 * RAC_MULT, oobs_val);
+	rtw89_write16_mask(rtwdev, R_RAC_DIRECT_OFFSET_G2 + RAC_ANA03 * RAC_MULT,
+			   OOBS_SEN_MASK, oobs_val);
 	rtw89_write16_set(rtwdev, R_RAC_DIRECT_OFFSET_G2 + RAC_ANA09 * RAC_MULT,
 			  BAC_OOBS_SEL);
 
@@ -2783,7 +2783,6 @@ static int rtw89_pci_ops_mac_pre_init_ax(struct rtw89_dev *rtwdev)
 	const struct rtw89_pci_info *info = rtwdev->pci_info;
 	int ret;
 
-	rtw89_pci_disable_eq(rtwdev);
 	rtw89_pci_ber(rtwdev);
 	rtw89_pci_rxdma_prefth(rtwdev);
 	rtw89_pci_l1off_pwroff(rtwdev);
@@ -4155,6 +4154,7 @@ static int __maybe_unused rtw89_pci_resume(struct device *dev)
 				  B_AX_SEL_REQ_ENTR_L1);
 	}
 	rtw89_pci_l2_hci_ldo(rtwdev);
+	rtw89_pci_disable_eq(rtwdev);
 	rtw89_pci_filter_out(rtwdev);
 	rtw89_pci_link_cfg(rtwdev);
 	rtw89_pci_l1ss_cfg(rtwdev);
@@ -4289,6 +4289,7 @@ int rtw89_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		goto err_clear_resource;
 	}
 
+	rtw89_pci_disable_eq(rtwdev);
 	rtw89_pci_filter_out(rtwdev);
 	rtw89_pci_link_cfg(rtwdev);
 	rtw89_pci_l1ss_cfg(rtwdev);
-- 
cgit v1.2.3-58-ga151


From f12fc2065647f8032fef2b43ad17843fc82ef456 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Wed, 22 May 2024 10:38:29 +0800
Subject: wifi: rtw89: fill STBC and LDPC capabilities to TX descriptor

Fill STBC and LDPC fields to get better performance when transmitting under
noisy environment. Old chips like 8852AE use firmware command to enable
these two capabilities, but newer chips must use TX descriptor.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240522023830.9798-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 12 ++++++++++--
 drivers/net/wireless/realtek/rtw89/core.h |  2 ++
 drivers/net/wireless/realtek/rtw89/txrx.h |  2 ++
 3 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 8b6cb310eaba..36d31608157f 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -813,6 +813,8 @@ rtw89_core_tx_update_data_info(struct rtw89_dev *rtwdev,
 	desc_info->mac_id = rtw89_core_tx_get_mac_id(rtwdev, tx_req);
 	desc_info->port = desc_info->hiq ? rtwvif->port : 0;
 	desc_info->er_cap = rtwsta ? rtwsta->er_cap : false;
+	desc_info->stbc = rtwsta ? rtwsta->ra.stbc_cap : false;
+	desc_info->ldpc = rtwsta ? rtwsta->ra.ldpc_cap : false;
 
 	/* enable wd_info for AMPDU */
 	desc_info->en_wd_info = true;
@@ -1127,6 +1129,8 @@ static __le32 rtw89_build_txwd_info0(struct rtw89_tx_desc_info *desc_info)
 {
 	u32 dword = FIELD_PREP(RTW89_TXWD_INFO0_USE_RATE, desc_info->use_rate) |
 		    FIELD_PREP(RTW89_TXWD_INFO0_DATA_RATE, desc_info->data_rate) |
+		    FIELD_PREP(RTW89_TXWD_INFO0_DATA_STBC, desc_info->stbc) |
+		    FIELD_PREP(RTW89_TXWD_INFO0_DATA_LDPC, desc_info->ldpc) |
 		    FIELD_PREP(RTW89_TXWD_INFO0_DISDATAFB, desc_info->dis_data_fb) |
 		    FIELD_PREP(RTW89_TXWD_INFO0_MULTIPORT_ID, desc_info->port);
 
@@ -1135,7 +1139,9 @@ static __le32 rtw89_build_txwd_info0(struct rtw89_tx_desc_info *desc_info)
 
 static __le32 rtw89_build_txwd_info0_v1(struct rtw89_tx_desc_info *desc_info)
 {
-	u32 dword = FIELD_PREP(RTW89_TXWD_INFO0_DISDATAFB, desc_info->dis_data_fb) |
+	u32 dword = FIELD_PREP(RTW89_TXWD_INFO0_DATA_STBC, desc_info->stbc) |
+		    FIELD_PREP(RTW89_TXWD_INFO0_DATA_LDPC, desc_info->ldpc) |
+		    FIELD_PREP(RTW89_TXWD_INFO0_DISDATAFB, desc_info->dis_data_fb) |
 		    FIELD_PREP(RTW89_TXWD_INFO0_MULTIPORT_ID, desc_info->port) |
 		    FIELD_PREP(RTW89_TXWD_INFO0_DATA_ER, desc_info->er_cap) |
 		    FIELD_PREP(RTW89_TXWD_INFO0_DATA_BW_ER, 0);
@@ -1301,7 +1307,9 @@ static __le32 rtw89_build_txwd_body7_v2(struct rtw89_tx_desc_info *desc_info)
 
 static __le32 rtw89_build_txwd_info0_v2(struct rtw89_tx_desc_info *desc_info)
 {
-	u32 dword = FIELD_PREP(BE_TXD_INFO0_DISDATAFB, desc_info->dis_data_fb) |
+	u32 dword = FIELD_PREP(BE_TXD_INFO0_DATA_STBC, desc_info->stbc) |
+		    FIELD_PREP(BE_TXD_INFO0_DATA_LDPC, desc_info->ldpc) |
+		    FIELD_PREP(BE_TXD_INFO0_DISDATAFB, desc_info->dis_data_fb) |
 		    FIELD_PREP(BE_TXD_INFO0_MULTIPORT_ID, desc_info->port);
 
 	return cpu_to_le32(dword);
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 523ea9f5e6b1..e953c99b9a47 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -1136,6 +1136,8 @@ struct rtw89_tx_desc_info {
 	bool hiq;
 	u8 port;
 	bool er_cap;
+	bool stbc;
+	bool ldpc;
 };
 
 struct rtw89_core_tx_request {
diff --git a/drivers/net/wireless/realtek/rtw89/txrx.h b/drivers/net/wireless/realtek/rtw89/txrx.h
index c467a80ffa88..11049a50de2c 100644
--- a/drivers/net/wireless/realtek/rtw89/txrx.h
+++ b/drivers/net/wireless/realtek/rtw89/txrx.h
@@ -113,6 +113,8 @@ static inline u8 rtw89_get_data_nss(struct rtw89_dev *rtwdev, u16 hw_rate)
 #define RTW89_TXWD_INFO0_GI_LTF GENMASK(27, 25)
 #define RTW89_TXWD_INFO0_DATA_RATE GENMASK(24, 16)
 #define RTW89_TXWD_INFO0_DATA_ER BIT(15)
+#define RTW89_TXWD_INFO0_DATA_STBC BIT(12)
+#define RTW89_TXWD_INFO0_DATA_LDPC BIT(11)
 #define RTW89_TXWD_INFO0_DISDATAFB BIT(10)
 #define RTW89_TXWD_INFO0_DATA_BW_ER BIT(8)
 #define RTW89_TXWD_INFO0_MULTIPORT_ID GENMASK(6, 4)
-- 
cgit v1.2.3-58-ga151


From 85e02a0aee586d031c3804de488e67daf8e16f9f Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Wed, 22 May 2024 10:38:30 +0800
Subject: wifi: rtw89: add LDPC and STBC to rx_status and radiotap known fields
 for monitor mode

Add LDPC and STBC to rx_status by hardware PPDU status, and set them as
known fields of radiotap, so we can check these capabilities in monitor
mode.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240522023830.9798-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 36 ++++++++++++++++++++++++++++++-
 drivers/net/wireless/realtek/rtw89/core.h |  2 ++
 drivers/net/wireless/realtek/rtw89/txrx.h |  2 ++
 3 files changed, 39 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 36d31608157f..3e99b63a7995 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -1557,6 +1557,12 @@ static void rtw89_core_parse_phy_status_ie01(struct rtw89_dev *rtwdev,
 	u32 t;
 
 	phy_ppdu->chan_idx = le32_get_bits(ie->w0, RTW89_PHY_STS_IE01_W0_CH_IDX);
+
+	if (rtwdev->hw->conf.flags & IEEE80211_CONF_MONITOR) {
+		phy_ppdu->ldpc = le32_get_bits(ie->w2, RTW89_PHY_STS_IE01_W2_LDPC);
+		phy_ppdu->stbc = le32_get_bits(ie->w2, RTW89_PHY_STS_IE01_W2_STBC);
+	}
+
 	if (phy_ppdu->rate < RTW89_HW_RATE_OFDM6)
 		return;
 
@@ -1982,6 +1988,23 @@ static void rtw89_core_hw_to_sband_rate(struct ieee80211_rx_status *rx_status)
 	rx_status->rate_idx -= 4;
 }
 
+static
+void rtw89_core_update_rx_status_by_ppdu(struct rtw89_dev *rtwdev,
+					 struct ieee80211_rx_status *rx_status,
+					 struct rtw89_rx_phy_ppdu *phy_ppdu)
+{
+	if (!(rtwdev->hw->conf.flags & IEEE80211_CONF_MONITOR))
+		return;
+
+	if (!phy_ppdu)
+		return;
+
+	if (phy_ppdu->ldpc)
+		rx_status->enc_flags |= RX_ENC_FLAG_LDPC;
+	if (phy_ppdu->stbc)
+		rx_status->enc_flags |= u8_encode_bits(1, RX_ENC_FLAG_STBC_MASK);
+}
+
 static const u8 rx_status_bw_to_radiotap_eht_usig[] = {
 	[RATE_INFO_BW_20] = IEEE80211_RADIOTAP_EHT_USIG_COMMON_BW_20MHZ,
 	[RATE_INFO_BW_5] = U8_MAX,
@@ -2025,10 +2048,14 @@ static void rtw89_core_update_radiotap_eht(struct rtw89_dev *rtwdev,
 
 	eht->user_info[0] =
 		cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_MCS_KNOWN |
-			    IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_KNOWN_O);
+			    IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_KNOWN_O |
+			    IEEE80211_RADIOTAP_EHT_USER_INFO_CODING_KNOWN);
 	eht->user_info[0] |=
 		le32_encode_bits(rx_status->rate_idx, IEEE80211_RADIOTAP_EHT_USER_INFO_MCS) |
 		le32_encode_bits(rx_status->nss, IEEE80211_RADIOTAP_EHT_USER_INFO_NSS_O);
+	if (rx_status->enc_flags & RX_ENC_FLAG_LDPC)
+		eht->user_info[0] |=
+			cpu_to_le32(IEEE80211_RADIOTAP_EHT_USER_INFO_CODING);
 
 	/* U-SIG */
 	tlv = (void *)tlv + sizeof(*tlv) + ALIGN(eht_len, 4);
@@ -2054,6 +2081,8 @@ static void rtw89_core_update_radiotap(struct rtw89_dev *rtwdev,
 {
 	static const struct ieee80211_radiotap_he known_he = {
 		.data1 = cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA1_DATA_MCS_KNOWN |
+				     IEEE80211_RADIOTAP_HE_DATA1_CODING_KNOWN |
+				     IEEE80211_RADIOTAP_HE_DATA1_STBC_KNOWN |
 				     IEEE80211_RADIOTAP_HE_DATA1_BW_RU_ALLOC_KNOWN),
 		.data2 = cpu_to_le16(IEEE80211_RADIOTAP_HE_DATA2_GI_KNOWN),
 	};
@@ -2085,6 +2114,7 @@ static void rtw89_core_rx_to_mac80211(struct rtw89_dev *rtwdev,
 
 	rtw89_core_hw_to_sband_rate(rx_status);
 	rtw89_core_rx_stats(rtwdev, phy_ppdu, desc_info, skb_ppdu);
+	rtw89_core_update_rx_status_by_ppdu(rtwdev, rx_status, phy_ppdu);
 	rtw89_core_update_radiotap(rtwdev, skb_ppdu, rx_status);
 	/* In low power mode, it does RX in thread context. */
 	local_bh_disable();
@@ -4496,6 +4526,10 @@ static int rtw89_core_register_hw(struct rtw89_dev *rtwdev)
 	hw->max_tx_aggregation_subframes = RTW89_MAX_TX_AGG_NUM;
 	hw->uapsd_max_sp_len = IEEE80211_WMM_IE_STA_QOSINFO_SP_ALL;
 
+	hw->radiotap_mcs_details |= IEEE80211_RADIOTAP_MCS_HAVE_FEC |
+				    IEEE80211_RADIOTAP_MCS_HAVE_STBC;
+	hw->radiotap_vht_details |= IEEE80211_RADIOTAP_VHT_KNOWN_STBC;
+
 	ieee80211_hw_set(hw, SIGNAL_DBM);
 	ieee80211_hw_set(hw, HAS_RATE_CONTROL);
 	ieee80211_hw_set(hw, MFP_CAPABLE);
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index e953c99b9a47..68221f0b156e 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -793,6 +793,8 @@ struct rtw89_rx_phy_ppdu {
 		u8 evm_max;
 		u8 evm_min;
 	} ofdm;
+	bool ldpc;
+	bool stbc;
 	bool to_self;
 	bool valid;
 };
diff --git a/drivers/net/wireless/realtek/rtw89/txrx.h b/drivers/net/wireless/realtek/rtw89/txrx.h
index 11049a50de2c..3882938c0893 100644
--- a/drivers/net/wireless/realtek/rtw89/txrx.h
+++ b/drivers/net/wireless/realtek/rtw89/txrx.h
@@ -558,6 +558,8 @@ struct rtw89_phy_sts_ie0 {
 #define RTW89_PHY_STS_IE01_W2_AVG_SNR GENMASK(5, 0)
 #define RTW89_PHY_STS_IE01_W2_EVM_MAX GENMASK(15, 8)
 #define RTW89_PHY_STS_IE01_W2_EVM_MIN GENMASK(23, 16)
+#define RTW89_PHY_STS_IE01_W2_LDPC BIT(28)
+#define RTW89_PHY_STS_IE01_W2_STBC BIT(30)
 
 enum rtw89_tx_channel {
 	RTW89_TXCH_ACH0	= 0,
-- 
cgit v1.2.3-58-ga151


From e88c1714b2c7b230c799a00973bdeb374c765c06 Mon Sep 17 00:00:00 2001
From: Po-Hao Huang <phhuang@realtek.com>
Date: Wed, 22 May 2024 10:45:31 +0800
Subject: wifi: rtw89: Fix P2P behavior for WiFi 7 chips

Previously we used CCK rate when scanning for P2P on WiFi 7 chips.
Fix this by explicitly setting the rate to OFDM 6Mbps.

Signed-off-by: Po-Hao Huang <phhuang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240522024531.11401-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c | 10 ++++++++++
 drivers/net/wireless/realtek/rtw89/fw.h |  4 ++++
 2 files changed, 14 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 1ea2b662fc89..be39a8468d32 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -4850,6 +4850,7 @@ int rtw89_fw_h2c_scan_offload_be(struct rtw89_dev *rtwdev,
 {
 	struct rtw89_hw_scan_info *scan_info = &rtwdev->scan_info;
 	struct rtw89_wait_info *wait = &rtwdev->mac.fw_ofld_wait;
+	struct cfg80211_scan_request *req = rtwvif->scan_req;
 	struct rtw89_h2c_scanofld_be_macc_role *macc_role;
 	struct rtw89_chan *op = &scan_info->op_chan;
 	struct rtw89_h2c_scanofld_be_opch *opch;
@@ -4923,6 +4924,15 @@ int rtw89_fw_h2c_scan_offload_be(struct rtw89_dev *rtwdev,
 				   RTW89_H2C_SCANOFLD_BE_W6_CHAN_PROHIB_LOW);
 	h2c->w7 = le32_encode_bits(option->prohib_chan >> 32,
 				   RTW89_H2C_SCANOFLD_BE_W7_CHAN_PROHIB_HIGH);
+	if (req->no_cck) {
+		h2c->w0 |= le32_encode_bits(true, RTW89_H2C_SCANOFLD_BE_W0_PROBE_WITH_RATE);
+		h2c->w8 = le32_encode_bits(RTW89_HW_RATE_OFDM6,
+					   RTW89_H2C_SCANOFLD_BE_W8_PROBE_RATE_2GHZ) |
+			  le32_encode_bits(RTW89_HW_RATE_OFDM6,
+					   RTW89_H2C_SCANOFLD_BE_W8_PROBE_RATE_5GHZ) |
+			  le32_encode_bits(RTW89_HW_RATE_OFDM6,
+					   RTW89_H2C_SCANOFLD_BE_W8_PROBE_RATE_6GHZ);
+	}
 	ptr += sizeof(*h2c);
 
 	for (i = 0; i < option->num_macc_role; i++) {
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index 4151c9d566bd..01fea0b004d3 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -2722,6 +2722,7 @@ struct rtw89_h2c_scanofld_be {
 #define RTW89_H2C_SCANOFLD_BE_W0_MACID GENMASK(23, 8)
 #define RTW89_H2C_SCANOFLD_BE_W0_PORT GENMASK(26, 24)
 #define RTW89_H2C_SCANOFLD_BE_W0_BAND GENMASK(28, 27)
+#define RTW89_H2C_SCANOFLD_BE_W0_PROBE_WITH_RATE BIT(29)
 #define RTW89_H2C_SCANOFLD_BE_W1_NUM_MACC_ROLE GENMASK(7, 0)
 #define RTW89_H2C_SCANOFLD_BE_W1_NUM_OP GENMASK(15, 8)
 #define RTW89_H2C_SCANOFLD_BE_W1_NORM_PD GENMASK(31, 16)
@@ -2738,6 +2739,9 @@ struct rtw89_h2c_scanofld_be {
 #define RTW89_H2C_SCANOFLD_BE_W5_MLO_MODE GENMASK(31, 0)
 #define RTW89_H2C_SCANOFLD_BE_W6_CHAN_PROHIB_LOW GENMASK(31, 0)
 #define RTW89_H2C_SCANOFLD_BE_W7_CHAN_PROHIB_HIGH GENMASK(31, 0)
+#define RTW89_H2C_SCANOFLD_BE_W8_PROBE_RATE_2GHZ GENMASK(7, 0)
+#define RTW89_H2C_SCANOFLD_BE_W8_PROBE_RATE_5GHZ GENMASK(15, 8)
+#define RTW89_H2C_SCANOFLD_BE_W8_PROBE_RATE_6GHZ GENMASK(23, 16)
 
 static inline void RTW89_SET_FWCMD_P2P_MACID(void *cmd, u32 val)
 {
-- 
cgit v1.2.3-58-ga151


From 5233a55a5254ea38dcdd8d836a0f9ee886c3df51 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Thu, 23 May 2024 16:59:22 +0100
Subject: mISDN: remove unused struct 'bf_ctx'

'bf_ctx' appears unused since the original
commit 960366cf8dbb ("Add mISDN DSP").

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://lore.kernel.org/r/20240523155922.67329-1-linux@treblig.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/isdn/mISDN/dsp_blowfish.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/isdn/mISDN/dsp_blowfish.c b/drivers/isdn/mISDN/dsp_blowfish.c
index 0aa572f3858d..0e77c282c862 100644
--- a/drivers/isdn/mISDN/dsp_blowfish.c
+++ b/drivers/isdn/mISDN/dsp_blowfish.c
@@ -73,11 +73,6 @@
  * crypto-api for faster implementation
  */
 
-struct bf_ctx {
-	u32 p[18];
-	u32 s[1024];
-};
-
 static const u32 bf_pbox[16 + 2] = {
 	0x243f6a88, 0x85a308d3, 0x13198a2e, 0x03707344,
 	0xa4093822, 0x299f31d0, 0x082efa98, 0xec4e6c89,
-- 
cgit v1.2.3-58-ga151


From d9cbd8343b010016fcaabc361c37720dcafddcbe Mon Sep 17 00:00:00 2001
From: yunshui <jiangyunshui@kylinos.cn>
Date: Thu, 23 May 2024 11:35:20 +0800
Subject: bpf, net: Use DEV_STAT_INC()

syzbot/KCSAN reported that races happen when multiple CPUs updating
dev->stats.tx_error concurrently. Adopt SMP safe DEV_STATS_INC() to
update the dev->stats fields.

Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: yunshui <jiangyunshui@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240523033520.4029314-1-jiangyunshui@kylinos.cn
---
 net/core/filter.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/core/filter.c b/net/core/filter.c
index c6edfe9f41bc..7c46ecba3b01 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -2274,12 +2274,12 @@ static int __bpf_redirect_neigh_v6(struct sk_buff *skb, struct net_device *dev,
 
 	err = bpf_out_neigh_v6(net, skb, dev, nh);
 	if (unlikely(net_xmit_eval(err)))
-		dev->stats.tx_errors++;
+		DEV_STATS_INC(dev, tx_errors);
 	else
 		ret = NET_XMIT_SUCCESS;
 	goto out_xmit;
 out_drop:
-	dev->stats.tx_errors++;
+	DEV_STATS_INC(dev, tx_errors);
 	kfree_skb(skb);
 out_xmit:
 	return ret;
@@ -2380,12 +2380,12 @@ static int __bpf_redirect_neigh_v4(struct sk_buff *skb, struct net_device *dev,
 
 	err = bpf_out_neigh_v4(net, skb, dev, nh);
 	if (unlikely(net_xmit_eval(err)))
-		dev->stats.tx_errors++;
+		DEV_STATS_INC(dev, tx_errors);
 	else
 		ret = NET_XMIT_SUCCESS;
 	goto out_xmit;
 out_drop:
-	dev->stats.tx_errors++;
+	DEV_STATS_INC(dev, tx_errors);
 	kfree_skb(skb);
 out_xmit:
 	return ret;
-- 
cgit v1.2.3-58-ga151


From c65b6521115ed478471bf8c4a3687059fcea01aa Mon Sep 17 00:00:00 2001
From: Gou Hao <gouhao@uniontech.com>
Date: Sun, 26 May 2024 22:57:17 +0800
Subject: net/core: remove redundant sk_callback_lock initialization

sk_callback_lock has already been initialized in sk_init_common().

Signed-off-by: Gou Hao <gouhao@uniontech.com>
Reviewed-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240526145718.9542-1-gouhao@uniontech.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/sock.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/core/sock.c b/net/core/sock.c
index 8629f9aecf91..67b10954e0cf 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -3460,7 +3460,6 @@ void sock_init_data_uid(struct socket *sock, struct sock *sk, kuid_t uid)
 	}
 	sk->sk_uid	=	uid;
 
-	rwlock_init(&sk->sk_callback_lock);
 	if (sk->sk_kern_sock)
 		lockdep_set_class_and_name(
 			&sk->sk_callback_lock,
-- 
cgit v1.2.3-58-ga151


From de31e96cf423848a1d541142446801430c12da3c Mon Sep 17 00:00:00 2001
From: Gou Hao <gouhao@uniontech.com>
Date: Sun, 26 May 2024 22:57:18 +0800
Subject: net/core: move the lockdep-init of sk_callback_lock to
 sk_init_common()

In commit cdfbabfb2f0c ("net: Work around lockdep limitation in
sockets that use sockets"), it introduces 'af_kern_callback_keys'
to lockdep-init of sk_callback_lock according to 'sk_kern_sock',
it modifies sock_init_data() only, and sk_clone_lock() calls
sk_init_common() to initialize sk_callback_lock too, so the
lockdep-init of sk_callback_lock should be moved to sk_init_common().

Signed-off-by: Gou Hao <gouhao@uniontech.com>
Link: https://lore.kernel.org/r/20240526145718.9542-2-gouhao@uniontech.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/sock.c | 18 ++++++------------
 1 file changed, 6 insertions(+), 12 deletions(-)

diff --git a/net/core/sock.c b/net/core/sock.c
index 67b10954e0cf..521e6373d4f7 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -2262,7 +2262,12 @@ static void sk_init_common(struct sock *sk)
 	lockdep_set_class_and_name(&sk->sk_error_queue.lock,
 			af_elock_keys + sk->sk_family,
 			af_family_elock_key_strings[sk->sk_family]);
-	lockdep_set_class_and_name(&sk->sk_callback_lock,
+	if (sk->sk_kern_sock)
+		lockdep_set_class_and_name(&sk->sk_callback_lock,
+			af_kern_callback_keys + sk->sk_family,
+			af_family_kern_clock_key_strings[sk->sk_family]);
+	else
+		lockdep_set_class_and_name(&sk->sk_callback_lock,
 			af_callback_keys + sk->sk_family,
 			af_family_clock_key_strings[sk->sk_family]);
 }
@@ -3460,17 +3465,6 @@ void sock_init_data_uid(struct socket *sock, struct sock *sk, kuid_t uid)
 	}
 	sk->sk_uid	=	uid;
 
-	if (sk->sk_kern_sock)
-		lockdep_set_class_and_name(
-			&sk->sk_callback_lock,
-			af_kern_callback_keys + sk->sk_family,
-			af_family_kern_clock_key_strings[sk->sk_family]);
-	else
-		lockdep_set_class_and_name(
-			&sk->sk_callback_lock,
-			af_callback_keys + sk->sk_family,
-			af_family_clock_key_strings[sk->sk_family]);
-
 	sk->sk_state_change	=	sock_def_wakeup;
 	sk->sk_data_ready	=	sock_def_readable;
 	sk->sk_write_space	=	sock_def_write_space;
-- 
cgit v1.2.3-58-ga151


From b2ff2698508ffc31c2dcbce58785d8b9dc847e1b Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Sun, 26 May 2024 18:24:25 +0100
Subject: net: ethernet: starfire: remove unused structs

'short_rx_done_desc' and 'basic_rx_done_desc' are unused since
commit fdecea66687d ("  [netdrvr starfire] Add GPL'd firmware, remove
compat code").

Remove them.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/adaptec/starfire.c | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/drivers/net/ethernet/adaptec/starfire.c b/drivers/net/ethernet/adaptec/starfire.c
index 857361c74f5d..e1b8794b14c9 100644
--- a/drivers/net/ethernet/adaptec/starfire.c
+++ b/drivers/net/ethernet/adaptec/starfire.c
@@ -441,14 +441,6 @@ enum rx_desc_bits {
 };
 
 /* Completion queue entry. */
-struct short_rx_done_desc {
-	__le32 status;			/* Low 16 bits is length. */
-};
-struct basic_rx_done_desc {
-	__le32 status;			/* Low 16 bits is length. */
-	__le16 vlanid;
-	__le16 status2;
-};
 struct csum_rx_done_desc {
 	__le32 status;			/* Low 16 bits is length. */
 	__le16 csum;			/* Partial checksum */
-- 
cgit v1.2.3-58-ga151


From a09892f6e281397fc8c4ffd512c8a0fd0a81dd32 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Sun, 26 May 2024 18:24:26 +0100
Subject: net: ethernet: liquidio: remove unused structs

'niclist' and 'oct_link_status_resp' are unused since the original
commit f21fb3ed364b ("Add support of Cavium Liquidio ethernet
adapters").

Remove them.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/cavium/liquidio/lio_main.c    | 6 ------
 drivers/net/ethernet/cavium/liquidio/octeon_droq.c | 5 -----
 2 files changed, 11 deletions(-)

diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index 34f02a8ec2ca..1d79f6eaa41f 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -92,12 +92,6 @@ static int octeon_console_debug_enabled(u32 console)
 /* time to wait for possible in-flight requests in milliseconds */
 #define WAIT_INFLIGHT_REQUEST	msecs_to_jiffies(1000)
 
-struct oct_link_status_resp {
-	u64 rh;
-	struct oct_link_info link_info;
-	u64 status;
-};
-
 struct oct_timestamp_resp {
 	u64 rh;
 	u64 timestamp;
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
index 0d6ee30affb9..eef12fdd246d 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
@@ -30,11 +30,6 @@
 #include "cn23xx_pf_device.h"
 #include "cn23xx_vf_device.h"
 
-struct niclist {
-	struct list_head list;
-	void *ptr;
-};
-
 struct __dispatch {
 	struct list_head list;
 	struct octeon_recv_info *rinfo;
-- 
cgit v1.2.3-58-ga151


From ef7f9febb33d22e687226211c67aa50158a0205b Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Sun, 26 May 2024 18:24:27 +0100
Subject: net: ethernet: mlx4: remove unused struct 'mlx4_port_config'

'mlx4_port_config was added by
commit ab9c17a009ee ("mlx4_core: Modify driver initialization flow to
accommodate SRIOV for Ethernet")
but remained unused.

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/mellanox/mlx4/main.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index 98688e4dbec5..febeadfdd5a5 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -169,12 +169,6 @@ module_param_array(port_type_array, int, &arr_argc, 0444);
 MODULE_PARM_DESC(port_type_array, "Array of port types: HW_DEFAULT (0) is default "
 				"1 for IB, 2 for Ethernet");
 
-struct mlx4_port_config {
-	struct list_head list;
-	enum mlx4_port_type port_type[MLX4_MAX_PORTS + 1];
-	struct pci_dev *pdev;
-};
-
 static atomic_t pf_loading = ATOMIC_INIT(0);
 
 static int mlx4_devlink_ierr_reset_get(struct devlink *devlink, u32 id,
-- 
cgit v1.2.3-58-ga151


From 18ae4c093cd23b741607fd35ba449f80fbd9f322 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Sun, 26 May 2024 18:24:28 +0100
Subject: net: ethernet: 8390: ne2k-pci: remove unused struct 'ne2k_pci_card'

'ne2k_pci_card' is unused since 2.3.99-pre3 in March 2000.

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/8390/ne2k-pci.c | 11 -----------
 1 file changed, 11 deletions(-)

diff --git a/drivers/net/ethernet/8390/ne2k-pci.c b/drivers/net/ethernet/8390/ne2k-pci.c
index 65f56a98c0a0..1a34da07c0db 100644
--- a/drivers/net/ethernet/8390/ne2k-pci.c
+++ b/drivers/net/ethernet/8390/ne2k-pci.c
@@ -186,17 +186,6 @@ static void ne2k_pci_block_output(struct net_device *dev, const int count,
 static const struct ethtool_ops ne2k_pci_ethtool_ops;
 
 
-
-/* There is no room in the standard 8390 structure for extra info we need,
- * so we build a meta/outer-wrapper structure..
- */
-struct ne2k_pci_card {
-	struct net_device *dev;
-	struct pci_dev *pci_dev;
-};
-
-
-
 /* NEx000-clone boards have a Station Address (SA) PROM (SAPROM) in the packet
  * buffer memory space.  By-the-spec NE2000 clones have 0x57,0x57 in bytes
  * 0x0e,0x0f of the SAPROM, while other supposed NE2000 clones must be
-- 
cgit v1.2.3-58-ga151


From c30ff5f3aec3f77e13cfd7373390639bfdcffba7 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Sun, 26 May 2024 21:59:22 +0100
Subject: net: usb: remove unused structs 'usb_context'

Both lan78xx and smsc75xx have a 'usb_context'
struct which is unused, since their original commits.

Remove them.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://lore.kernel.org/r/20240526205922.176578-1-linux@treblig.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/usb/lan78xx.c  | 5 -----
 drivers/net/usb/smsc75xx.c | 5 -----
 2 files changed, 10 deletions(-)

diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index 5a2c38b63012..7a5cc49ebec6 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -380,11 +380,6 @@ struct skb_data {		/* skb->cb is one of these */
 	int num_of_packet;
 };
 
-struct usb_context {
-	struct usb_ctrlrequest req;
-	struct lan78xx_net *dev;
-};
-
 #define EVENT_TX_HALT			0
 #define EVENT_RX_HALT			1
 #define EVENT_RX_MEMORY			2
diff --git a/drivers/net/usb/smsc75xx.c b/drivers/net/usb/smsc75xx.c
index 0726e18bee6f..78c821349f48 100644
--- a/drivers/net/usb/smsc75xx.c
+++ b/drivers/net/usb/smsc75xx.c
@@ -61,11 +61,6 @@ struct smsc75xx_priv {
 	u8 suspend_flags;
 };
 
-struct usb_context {
-	struct usb_ctrlrequest req;
-	struct usbnet *dev;
-};
-
 static bool turbo_mode = true;
 module_param(turbo_mode, bool, 0644);
 MODULE_PARM_DESC(turbo_mode, "Enable multiple frames per Rx transaction");
-- 
cgit v1.2.3-58-ga151


From af69d862a8baa600b018f66aca8174b23c166a5f Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Thu, 23 May 2024 08:18:05 -0700
Subject: wifi: ath12k: Fix devmem address prefix when logging

Currently when ath12k QMI logging is enabled, messages such as the
following can be logged:

ath12k_pci 0000:03:00.0: devmem [0] start ox113000 size 20480

Replace ox% with 0x% to get a proper hex address prefix:

ath12k_pci 0000:03:00.0: devmem [0] start 0x113000 size 20480

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240523-ox-v1-1-112ae7350059@quicinc.com
---
 drivers/net/wireless/ath/ath12k/qmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/qmi.c b/drivers/net/wireless/ath/ath12k/qmi.c
index f273cb93dedb..b93ce9f87f61 100644
--- a/drivers/net/wireless/ath/ath12k/qmi.c
+++ b/drivers/net/wireless/ath/ath12k/qmi.c
@@ -2503,7 +2503,7 @@ static int ath12k_qmi_request_target_cap(struct ath12k_base *ab)
 			ab->qmi.dev_mem[i].size =
 				resp.dev_mem[i].size;
 			ath12k_dbg(ab, ATH12K_DBG_QMI,
-				   "devmem [%d] start ox%llx size %llu\n", i,
+				   "devmem [%d] start 0x%llx size %llu\n", i,
 				   ab->qmi.dev_mem[i].start,
 				   ab->qmi.dev_mem[i].size);
 		}
-- 
cgit v1.2.3-58-ga151


From 5714e25f1d1875b300fb337dadfaa75324c1161a Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Fri, 24 May 2024 10:15:58 +0800
Subject: wifi: ath11k: fix wrong definition of CE ring's base address

Base address of CE ring is defined as u32, currently this works
because coherent DMA mask configured as 32 bit:

	#define ATH11K_PCI_COHERENT_DMA_MASK	32

However this mask could be changed once firmware bugs are fixed
to fully support 36 bit DMA addressing. So to protect against any
future changes to the DMA mask, change the type of the fields that
are dependent upon it.

This is found during code review. Compile tested only.

Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240524021558.34452-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath11k/ce.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/ce.h b/drivers/net/wireless/ath/ath11k/ce.h
index 69946fc70077..bcde2fcf02cf 100644
--- a/drivers/net/wireless/ath/ath11k/ce.h
+++ b/drivers/net/wireless/ath/ath11k/ce.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause-Clear */
 /*
  * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
- * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2022, 2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #ifndef ATH11K_CE_H
@@ -146,7 +146,7 @@ struct ath11k_ce_ring {
 	/* Host address space */
 	void *base_addr_owner_space_unaligned;
 	/* CE address space */
-	u32 base_addr_ce_space_unaligned;
+	dma_addr_t base_addr_ce_space_unaligned;
 
 	/* Actual start of descriptors.
 	 * Aligned to descriptor-size boundary.
@@ -156,7 +156,7 @@ struct ath11k_ce_ring {
 	void *base_addr_owner_space;
 
 	/* CE address space */
-	u32 base_addr_ce_space;
+	dma_addr_t base_addr_ce_space;
 
 	/* HAL ring id */
 	u32 hal_ring_id;
-- 
cgit v1.2.3-58-ga151


From 0a993772e0f0934d730c0d451622c80e03a40ab1 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Fri, 24 May 2024 10:36:42 +0800
Subject: wifi: ath12k: fix race due to setting ATH12K_FLAG_EXT_IRQ_ENABLED too
 early

Commit 5082b3e3027e ("wifi: ath11k: fix race due to setting
ATH11K_FLAG_EXT_IRQ_ENABLED too early") fixes a race in ath11k
driver. Since ath12k shares the same logic as ath11k, currently
the race also exists in ath12k: in ath12k_pci_ext_irq_enable(),
ATH12K_FLAG_EXT_IRQ_ENABLED is set before NAPI is enabled.
In cases where only one MSI vector is allocated, this results
in a race condition: after ATH12K_FLAG_EXT_IRQ_ENABLED is set
but before NAPI enabled, CE interrupt breaks in. Since IRQ is
shared by CE and data path, ath12k_pci_ext_interrupt_handler()
is also called where we call disable_irq_nosync() to disable
IRQ. Then napi_schedule() is called but it does nothing because
NAPI is not enabled at that time, meaning that
ath12k_pci_ext_grp_napi_poll() will never run, so we have
no chance to call enable_irq() to enable IRQ back. Since IRQ
is shared, all interrupts are disabled and we would finally
get no response from target.

So port ath11k fix here, this is done by setting
ATH12K_FLAG_EXT_IRQ_ENABLED after all NAPI and IRQ work are
done. With the fix, we are sure that by the time
ATH12K_FLAG_EXT_IRQ_ENABLED is set, NAPI is enabled.

Note that the fix above also introduce some side effects:
if ath12k_pci_ext_interrupt_handler() breaks in after NAPI
enabled but before ATH12K_FLAG_EXT_IRQ_ENABLED set, nothing
will be done by the handler this time, the work will be
postponed till the next time the IRQ fires.

This is found during code review.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240524023642.37030-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/pci.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/pci.c b/drivers/net/wireless/ath/ath12k/pci.c
index ac75e8e3916b..11b95d037051 100644
--- a/drivers/net/wireless/ath/ath12k/pci.c
+++ b/drivers/net/wireless/ath/ath12k/pci.c
@@ -1107,14 +1107,14 @@ void ath12k_pci_ext_irq_enable(struct ath12k_base *ab)
 {
 	int i;
 
-	set_bit(ATH12K_FLAG_EXT_IRQ_ENABLED, &ab->dev_flags);
-
 	for (i = 0; i < ATH12K_EXT_IRQ_GRP_NUM_MAX; i++) {
 		struct ath12k_ext_irq_grp *irq_grp = &ab->ext_irq_grp[i];
 
 		napi_enable(&irq_grp->napi);
 		ath12k_pci_ext_grp_enable(irq_grp);
 	}
+
+	set_bit(ATH12K_FLAG_EXT_IRQ_ENABLED, &ab->dev_flags);
 }
 
 void ath12k_pci_ext_irq_disable(struct ath12k_base *ab)
-- 
cgit v1.2.3-58-ga151


From 0ae570703754858a77cc42b3c9fff42e9f084608 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Fri, 24 May 2024 10:40:21 +0800
Subject: wifi: ath12k: fix wrong definition of CE ring's base address

Base address of CE ring is defined as u32, currently this works
because DMA mask configured as 32 bit:

	#define ATH12K_PCI_DMA_MASK	32

However this mask could be changed once firmware bugs are fixed
to fully support 36 bit DMA addressing. So to protect against any
future changes to the DMA mask, change the type of the fields that
are dependent upon it.

This is found during code review. Compile tested only.

Fixes: d889913205cf ("wifi: ath12k: driver for Qualcomm Wi-Fi 7 devices")
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240524024021.37711-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/ce.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/ce.h b/drivers/net/wireless/ath/ath12k/ce.h
index 79af3b6159f1..857bc5f9e946 100644
--- a/drivers/net/wireless/ath/ath12k/ce.h
+++ b/drivers/net/wireless/ath/ath12k/ce.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause-Clear */
 /*
  * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved.
- * Copyright (c) 2021-2022 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2021-2022, 2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #ifndef ATH12K_CE_H
@@ -119,7 +119,7 @@ struct ath12k_ce_ring {
 	/* Host address space */
 	void *base_addr_owner_space_unaligned;
 	/* CE address space */
-	u32 base_addr_ce_space_unaligned;
+	dma_addr_t base_addr_ce_space_unaligned;
 
 	/* Actual start of descriptors.
 	 * Aligned to descriptor-size boundary.
@@ -129,7 +129,7 @@ struct ath12k_ce_ring {
 	void *base_addr_owner_space;
 
 	/* CE address space */
-	u32 base_addr_ce_space;
+	dma_addr_t base_addr_ce_space;
 
 	/* HAL ring id */
 	u32 hal_ring_id;
-- 
cgit v1.2.3-58-ga151


From eb4e7726279a344c82e3c23be396bcfd0a4d5669 Mon Sep 17 00:00:00 2001
From: Mykyta Yatsenko <yatsenko@meta.com>
Date: Fri, 24 May 2024 14:18:40 +0100
Subject: libbpf: Configure log verbosity with env variable

Configure logging verbosity by setting LIBBPF_LOG_LEVEL environment
variable, which is applied only to default logger. Once user set their
custom logging callback, it is up to them to handle filtering.

Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240524131840.114289-1-yatsenko@meta.com
---
 Documentation/bpf/libbpf/libbpf_overview.rst |  8 ++++++++
 tools/lib/bpf/libbpf.c                       | 25 ++++++++++++++++++++++++-
 tools/lib/bpf/libbpf.h                       |  5 ++++-
 3 files changed, 36 insertions(+), 2 deletions(-)

diff --git a/Documentation/bpf/libbpf/libbpf_overview.rst b/Documentation/bpf/libbpf/libbpf_overview.rst
index f36a2d4ffea2..f4d22f0c62b0 100644
--- a/Documentation/bpf/libbpf/libbpf_overview.rst
+++ b/Documentation/bpf/libbpf/libbpf_overview.rst
@@ -219,6 +219,14 @@ compilation and skeleton generation. Using Libbpf-rs will make building user
 space part of the BPF application easier. Note that the BPF program themselves
 must still be written in plain C.
 
+libbpf logging
+==============
+
+By default, libbpf logs informational and warning messages to stderr. The
+verbosity of these messages can be controlled by setting the environment
+variable LIBBPF_LOG_LEVEL to either warn, info, or debug. A custom log
+callback can be set using ``libbpf_set_print()``.
+
 Additional Documentation
 ========================
 
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 5401f2df463d..d1627a2ca30b 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -229,7 +229,30 @@ static const char * const prog_type_name[] = {
 static int __base_pr(enum libbpf_print_level level, const char *format,
 		     va_list args)
 {
-	if (level == LIBBPF_DEBUG)
+	const char *env_var = "LIBBPF_LOG_LEVEL";
+	static enum libbpf_print_level min_level = LIBBPF_INFO;
+	static bool initialized;
+
+	if (!initialized) {
+		char *verbosity;
+
+		initialized = true;
+		verbosity = getenv(env_var);
+		if (verbosity) {
+			if (strcasecmp(verbosity, "warn") == 0)
+				min_level = LIBBPF_WARN;
+			else if (strcasecmp(verbosity, "debug") == 0)
+				min_level = LIBBPF_DEBUG;
+			else if (strcasecmp(verbosity, "info") == 0)
+				min_level = LIBBPF_INFO;
+			else
+				fprintf(stderr, "libbpf: unrecognized '%s' envvar value: '%s', should be one of 'warn', 'debug', or 'info'.\n",
+					env_var, verbosity);
+		}
+	}
+
+	/* if too verbose, skip logging  */
+	if (level > min_level)
 		return 0;
 
 	return vfprintf(stderr, format, args);
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index c3f77d9260fe..26e4e35528c5 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -98,7 +98,10 @@ typedef int (*libbpf_print_fn_t)(enum libbpf_print_level level,
 
 /**
  * @brief **libbpf_set_print()** sets user-provided log callback function to
- * be used for libbpf warnings and informational messages.
+ * be used for libbpf warnings and informational messages. If the user callback
+ * is not set, messages are logged to stderr by default. The verbosity of these
+ * messages can be controlled by setting the environment variable
+ * LIBBPF_LOG_LEVEL to either warn, info, or debug.
  * @param fn The log print function. If NULL, libbpf won't print anything.
  * @return Pointer to old print function.
  *
-- 
cgit v1.2.3-58-ga151


From ed31adf6874db172e3212ac1ebaf701ed6190650 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Sat, 25 May 2024 20:08:15 +0800
Subject: selftests/bpf: Drop struct post_socket_opts

It's not possible to have one generic/common "struct post_socket_opts"
for all tests. It's better to have the individual test define its own
callback opts struct.

So this patch drops struct post_socket_opts, and changes the second
parameter of post_socket_cb as "void *" type.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/f8bda41c7cb9cb6979b2779f89fb3a684234304f.1716638248.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/network_helpers.c               | 2 +-
 tools/testing/selftests/bpf/network_helpers.h               | 4 +---
 tools/testing/selftests/bpf/prog_tests/sockopt_inherit.c    | 2 +-
 tools/testing/selftests/bpf/test_tcp_check_syncookie_user.c | 4 ++--
 4 files changed, 5 insertions(+), 7 deletions(-)

diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index 35250e6cde7f..4d776b78929c 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -133,7 +133,7 @@ int start_server(int family, int type, const char *addr_str, __u16 port,
 	return __start_server(type, (struct sockaddr *)&addr, addrlen, &opts);
 }
 
-static int reuseport_cb(int fd, const struct post_socket_opts *opts)
+static int reuseport_cb(int fd, void *opts)
 {
 	int on = 1;
 
diff --git a/tools/testing/selftests/bpf/network_helpers.h b/tools/testing/selftests/bpf/network_helpers.h
index 883c7ea9d8d5..40011e0f584b 100644
--- a/tools/testing/selftests/bpf/network_helpers.h
+++ b/tools/testing/selftests/bpf/network_helpers.h
@@ -21,8 +21,6 @@ typedef __u16 __sum16;
 #define VIP_NUM 5
 #define MAGIC_BYTES 123
 
-struct post_socket_opts {};
-
 struct network_helper_opts {
 	const char *cc;
 	int timeout_ms;
@@ -30,7 +28,7 @@ struct network_helper_opts {
 	bool noconnect;
 	int type;
 	int proto;
-	int (*post_socket_cb)(int fd, const struct post_socket_opts *opts);
+	int (*post_socket_cb)(int fd, void *opts);
 };
 
 /* ipv4 test vector */
diff --git a/tools/testing/selftests/bpf/prog_tests/sockopt_inherit.c b/tools/testing/selftests/bpf/prog_tests/sockopt_inherit.c
index 1d3a20f01b60..7cd8be2780ca 100644
--- a/tools/testing/selftests/bpf/prog_tests/sockopt_inherit.c
+++ b/tools/testing/selftests/bpf/prog_tests/sockopt_inherit.c
@@ -70,7 +70,7 @@ static void *server_thread(void *arg)
 	return (void *)(long)err;
 }
 
-static int custom_cb(int fd, const struct post_socket_opts *opts)
+static int custom_cb(int fd, void *opts)
 {
 	char buf;
 	int err;
diff --git a/tools/testing/selftests/bpf/test_tcp_check_syncookie_user.c b/tools/testing/selftests/bpf/test_tcp_check_syncookie_user.c
index 7b5fc98838cd..aebc58c24dc5 100644
--- a/tools/testing/selftests/bpf/test_tcp_check_syncookie_user.c
+++ b/tools/testing/selftests/bpf/test_tcp_check_syncookie_user.c
@@ -139,14 +139,14 @@ out:
 	return ret;
 }
 
-static int v6only_true(int fd, const struct post_socket_opts *opts)
+static int v6only_true(int fd, void *opts)
 {
 	int mode = true;
 
 	return setsockopt(fd, IPPROTO_IPV6, IPV6_V6ONLY, &mode, sizeof(mode));
 }
 
-static int v6only_false(int fd, const struct post_socket_opts *opts)
+static int v6only_false(int fd, void *opts)
 {
 	int mode = false;
 
-- 
cgit v1.2.3-58-ga151


From 6f802cb8988e8e41f2fdb74ac949d3a0ef9a9594 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Sat, 25 May 2024 20:08:16 +0800
Subject: selftests/bpf: Add start_server_str helper

It's a tech debt that start_server() does not take the "opts" argument.
It's pretty handy to have start_server() as a helper that takes string
address.

So this patch creates a new helper start_server_str(). Then start_server()
can be a wrapper of it.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/606e6cfd7e1aff8bc51ede49862eed0802e52170.1716638248.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/network_helpers.c | 22 ++++++++++++++++------
 tools/testing/selftests/bpf/network_helpers.h |  2 ++
 2 files changed, 18 insertions(+), 6 deletions(-)

diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index 4d776b78929c..0e8266f439e4 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -118,19 +118,29 @@ error_close:
 	return -1;
 }
 
-int start_server(int family, int type, const char *addr_str, __u16 port,
-		 int timeout_ms)
+int start_server_str(int family, int type, const char *addr_str, __u16 port,
+		     const struct network_helper_opts *opts)
 {
-	struct network_helper_opts opts = {
-		.timeout_ms	= timeout_ms,
-	};
 	struct sockaddr_storage addr;
 	socklen_t addrlen;
 
+	if (!opts)
+		opts = &default_opts;
+
 	if (make_sockaddr(family, addr_str, port, &addr, &addrlen))
 		return -1;
 
-	return __start_server(type, (struct sockaddr *)&addr, addrlen, &opts);
+	return __start_server(type, (struct sockaddr *)&addr, addrlen, opts);
+}
+
+int start_server(int family, int type, const char *addr_str, __u16 port,
+		 int timeout_ms)
+{
+	struct network_helper_opts opts = {
+		.timeout_ms	= timeout_ms,
+	};
+
+	return start_server_str(family, type, addr_str, port, &opts);
 }
 
 static int reuseport_cb(int fd, void *opts)
diff --git a/tools/testing/selftests/bpf/network_helpers.h b/tools/testing/selftests/bpf/network_helpers.h
index 40011e0f584b..4e3e6afe7d3a 100644
--- a/tools/testing/selftests/bpf/network_helpers.h
+++ b/tools/testing/selftests/bpf/network_helpers.h
@@ -48,6 +48,8 @@ struct ipv6_packet {
 extern struct ipv6_packet pkt_v6;
 
 int settimeo(int fd, int timeout_ms);
+int start_server_str(int family, int type, const char *addr_str, __u16 port,
+		     const struct network_helper_opts *opts);
 int start_server(int family, int type, const char *addr, __u16 port,
 		 int timeout_ms);
 int *start_reuseport_server(int family, int type, const char *addr_str,
-- 
cgit v1.2.3-58-ga151


From e078255abd53ac44c9133fd98d51645dbd196123 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Sat, 25 May 2024 20:08:17 +0800
Subject: selftests/bpf: Use post_socket_cb in connect_to_fd_opts

Since the post_socket_cb() callback is added in struct network_helper_opts,
it's make sense to use it not only in __start_server(), but also in
connect_to_fd_opts(). Then it can be used to set TCP_CONGESTION sockopt.

Add a "void *" type member cb_opts into struct network_helper_opts, and add
a new struct named cb_opts in prog_tests/bpf_tcp_ca.c, then cc can be moved
into struct cb_opts from network_helper_opts. Define a new callback cc_cb()
to set TCP_CONGESTION sockopt, and set it to post_socket_cb pointer of opts.
Define a new cb_opts cubic, set it to cb_opts of opts. Pass this opts to
connect_to_fd_opts() in test_dctcp_fallback().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/b512bb8d8f6854c9ea5c409b69d1bf37c6f272c6.1716638248.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/network_helpers.c       |  5 ++---
 tools/testing/selftests/bpf/network_helpers.h       |  2 +-
 tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c | 17 ++++++++++++++++-
 3 files changed, 19 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index 0e8266f439e4..8502917b6c7b 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -348,9 +348,8 @@ int connect_to_fd_opts(int server_fd, const struct network_helper_opts *opts)
 	if (settimeo(fd, opts->timeout_ms))
 		goto error_close;
 
-	if (opts->cc && opts->cc[0] &&
-	    setsockopt(fd, SOL_TCP, TCP_CONGESTION, opts->cc,
-		       strlen(opts->cc) + 1))
+	if (opts->post_socket_cb &&
+	    opts->post_socket_cb(fd, opts->cb_opts))
 		goto error_close;
 
 	if (!opts->noconnect)
diff --git a/tools/testing/selftests/bpf/network_helpers.h b/tools/testing/selftests/bpf/network_helpers.h
index 4e3e6afe7d3a..11eea8e2e4f1 100644
--- a/tools/testing/selftests/bpf/network_helpers.h
+++ b/tools/testing/selftests/bpf/network_helpers.h
@@ -22,13 +22,13 @@ typedef __u16 __sum16;
 #define MAGIC_BYTES 123
 
 struct network_helper_opts {
-	const char *cc;
 	int timeout_ms;
 	bool must_fail;
 	bool noconnect;
 	int type;
 	int proto;
 	int (*post_socket_cb)(int fd, void *opts);
+	void *cb_opts;
 };
 
 /* ipv4 test vector */
diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index 0aca02532794..6f034059888a 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -23,6 +23,10 @@
 static const unsigned int total_bytes = 10 * 1024 * 1024;
 static int expected_stg = 0xeB9F;
 
+struct cb_opts {
+	const char *cc;
+};
+
 static int settcpca(int fd, const char *tcp_ca)
 {
 	int err;
@@ -81,6 +85,13 @@ done:
 	close(fd);
 }
 
+static int cc_cb(int fd, void *opts)
+{
+	struct cb_opts *cb_opts = (struct cb_opts *)opts;
+
+	return settcpca(fd, cb_opts->cc);
+}
+
 static void test_cubic(void)
 {
 	struct bpf_cubic *cubic_skel;
@@ -172,10 +183,13 @@ static void test_dctcp_fallback(void)
 {
 	int err, lfd = -1, cli_fd = -1, srv_fd = -1;
 	struct network_helper_opts opts = {
-		.cc = "cubic",
+		.post_socket_cb	= cc_cb,
 	};
 	struct bpf_dctcp *dctcp_skel;
 	struct bpf_link *link = NULL;
+	struct cb_opts cubic = {
+		.cc = "cubic",
+	};
 	char srv_cc[16];
 	socklen_t cc_len = sizeof(srv_cc);
 
@@ -195,6 +209,7 @@ static void test_dctcp_fallback(void)
 	    !ASSERT_OK(settcpca(lfd, "bpf_dctcp"), "lfd=>bpf_dctcp"))
 		goto done;
 
+	opts.cb_opts = &cubic;
 	cli_fd = connect_to_fd_opts(lfd, &opts);
 	if (!ASSERT_GE(cli_fd, 0, "cli_fd"))
 		goto done;
-- 
cgit v1.2.3-58-ga151


From 79b330c57debe6b15f441e999bb62042afd5b08e Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Sat, 25 May 2024 20:08:18 +0800
Subject: selftests/bpf: Use post_socket_cb in start_server_str

This patch uses start_server_str() helper in test_dctcp_fallback() in
bpf_tcp_ca.c, instead of using start_server() and settcpca(). For
support opts in start_server_str() helper, opts->cb_opts needs to be
passed to post_socket_cb() in __start_server().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/414c749321fa150435f7fe8e12c80fec8b447c78.1716638248.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/network_helpers.c       | 3 ++-
 tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c | 9 ++++++---
 2 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index 8502917b6c7b..e20caef06aae 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -94,7 +94,8 @@ static int __start_server(int type, const struct sockaddr *addr, socklen_t addrl
 	if (settimeo(fd, opts->timeout_ms))
 		goto error_close;
 
-	if (opts->post_socket_cb && opts->post_socket_cb(fd, NULL)) {
+	if (opts->post_socket_cb &&
+	    opts->post_socket_cb(fd, opts->cb_opts)) {
 		log_err("Failed to call post_socket_cb");
 		goto error_close;
 	}
diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index 6f034059888a..7a523c7a0c6a 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -187,6 +187,9 @@ static void test_dctcp_fallback(void)
 	};
 	struct bpf_dctcp *dctcp_skel;
 	struct bpf_link *link = NULL;
+	struct cb_opts dctcp = {
+		.cc = "bpf_dctcp",
+	};
 	struct cb_opts cubic = {
 		.cc = "cubic",
 	};
@@ -204,9 +207,9 @@ static void test_dctcp_fallback(void)
 	if (!ASSERT_OK_PTR(link, "dctcp link"))
 		goto done;
 
-	lfd = start_server(AF_INET6, SOCK_STREAM, "::1", 0, 0);
-	if (!ASSERT_GE(lfd, 0, "lfd") ||
-	    !ASSERT_OK(settcpca(lfd, "bpf_dctcp"), "lfd=>bpf_dctcp"))
+	opts.cb_opts = &dctcp;
+	lfd = start_server_str(AF_INET6, SOCK_STREAM, "::1", 0, &opts);
+	if (!ASSERT_GE(lfd, 0, "lfd"))
 		goto done;
 
 	opts.cb_opts = &cubic;
-- 
cgit v1.2.3-58-ga151


From ed61271af5230cef9b9329bb1eacc1b1a9800d07 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Sat, 25 May 2024 20:08:19 +0800
Subject: selftests/bpf: Use start_server_str in do_test in bpf_tcp_ca

This patch uses new helper start_server_str() in do_test() in bpf_tcp_ca.c
to accept a struct network_helper_opts argument instead of using
start_server() and settcpca(). Then change the type of the first paramenter
of do_test() into a struct network_helper_opts one.

Define its own cb_opts and opts for each test, set its own cc name into
cb_opts.cc, and cc_cb() into post_socket_cb callback, then pass it to
do_test().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/6e1b6555e3284e77c8aa60668c61a66c5f99aa37.1716638248.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 .../testing/selftests/bpf/prog_tests/bpf_tcp_ca.c  | 66 ++++++++++++++++++----
 1 file changed, 55 insertions(+), 11 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index 7a523c7a0c6a..ebc7d4616880 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -38,12 +38,14 @@ static int settcpca(int fd, const char *tcp_ca)
 	return 0;
 }
 
-static void do_test(const char *tcp_ca, const struct bpf_map *sk_stg_map)
+static void do_test(const struct network_helper_opts *opts,
+		    const struct bpf_map *sk_stg_map)
 {
+	struct cb_opts *cb_opts = (struct cb_opts *)opts->cb_opts;
 	int lfd = -1, fd = -1;
 	int err;
 
-	lfd = start_server(AF_INET6, SOCK_STREAM, NULL, 0, 0);
+	lfd = start_server_str(AF_INET6, SOCK_STREAM, NULL, 0, opts);
 	if (!ASSERT_NEQ(lfd, -1, "socket"))
 		return;
 
@@ -53,7 +55,7 @@ static void do_test(const char *tcp_ca, const struct bpf_map *sk_stg_map)
 		return;
 	}
 
-	if (settcpca(lfd, tcp_ca) || settcpca(fd, tcp_ca))
+	if (settcpca(fd, cb_opts->cc))
 		goto done;
 
 	if (sk_stg_map) {
@@ -94,6 +96,13 @@ static int cc_cb(int fd, void *opts)
 
 static void test_cubic(void)
 {
+	struct cb_opts cb_opts = {
+		.cc = "bpf_cubic",
+	};
+	struct network_helper_opts opts = {
+		.post_socket_cb	= cc_cb,
+		.cb_opts	= &cb_opts,
+	};
 	struct bpf_cubic *cubic_skel;
 	struct bpf_link *link;
 
@@ -107,7 +116,7 @@ static void test_cubic(void)
 		return;
 	}
 
-	do_test("bpf_cubic", NULL);
+	do_test(&opts, NULL);
 
 	ASSERT_EQ(cubic_skel->bss->bpf_cubic_acked_called, 1, "pkts_acked called");
 
@@ -117,6 +126,13 @@ static void test_cubic(void)
 
 static void test_dctcp(void)
 {
+	struct cb_opts cb_opts = {
+		.cc = "bpf_dctcp",
+	};
+	struct network_helper_opts opts = {
+		.post_socket_cb	= cc_cb,
+		.cb_opts	= &cb_opts,
+	};
 	struct bpf_dctcp *dctcp_skel;
 	struct bpf_link *link;
 
@@ -130,7 +146,7 @@ static void test_dctcp(void)
 		return;
 	}
 
-	do_test("bpf_dctcp", dctcp_skel->maps.sk_stg_map);
+	do_test(&opts, dctcp_skel->maps.sk_stg_map);
 	ASSERT_EQ(dctcp_skel->bss->stg_result, expected_stg, "stg_result");
 
 	bpf_link__destroy(link);
@@ -315,6 +331,13 @@ static void test_unsupp_cong_op(void)
 
 static void test_update_ca(void)
 {
+	struct cb_opts cb_opts = {
+		.cc = "tcp_ca_update",
+	};
+	struct network_helper_opts opts = {
+		.post_socket_cb	= cc_cb,
+		.cb_opts	= &cb_opts,
+	};
 	struct tcp_ca_update *skel;
 	struct bpf_link *link;
 	int saved_ca1_cnt;
@@ -327,14 +350,14 @@ static void test_update_ca(void)
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
 	ASSERT_OK_PTR(link, "attach_struct_ops");
 
-	do_test("tcp_ca_update", NULL);
+	do_test(&opts, NULL);
 	saved_ca1_cnt = skel->bss->ca1_cnt;
 	ASSERT_GT(saved_ca1_cnt, 0, "ca1_ca1_cnt");
 
 	err = bpf_link__update_map(link, skel->maps.ca_update_2);
 	ASSERT_OK(err, "update_map");
 
-	do_test("tcp_ca_update", NULL);
+	do_test(&opts, NULL);
 	ASSERT_EQ(skel->bss->ca1_cnt, saved_ca1_cnt, "ca2_ca1_cnt");
 	ASSERT_GT(skel->bss->ca2_cnt, 0, "ca2_ca2_cnt");
 
@@ -344,6 +367,13 @@ static void test_update_ca(void)
 
 static void test_update_wrong(void)
 {
+	struct cb_opts cb_opts = {
+		.cc = "tcp_ca_update",
+	};
+	struct network_helper_opts opts = {
+		.post_socket_cb	= cc_cb,
+		.cb_opts	= &cb_opts,
+	};
 	struct tcp_ca_update *skel;
 	struct bpf_link *link;
 	int saved_ca1_cnt;
@@ -356,14 +386,14 @@ static void test_update_wrong(void)
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
 	ASSERT_OK_PTR(link, "attach_struct_ops");
 
-	do_test("tcp_ca_update", NULL);
+	do_test(&opts, NULL);
 	saved_ca1_cnt = skel->bss->ca1_cnt;
 	ASSERT_GT(saved_ca1_cnt, 0, "ca1_ca1_cnt");
 
 	err = bpf_link__update_map(link, skel->maps.ca_wrong);
 	ASSERT_ERR(err, "update_map");
 
-	do_test("tcp_ca_update", NULL);
+	do_test(&opts, NULL);
 	ASSERT_GT(skel->bss->ca1_cnt, saved_ca1_cnt, "ca2_ca1_cnt");
 
 	bpf_link__destroy(link);
@@ -372,6 +402,13 @@ static void test_update_wrong(void)
 
 static void test_mixed_links(void)
 {
+	struct cb_opts cb_opts = {
+		.cc = "tcp_ca_update",
+	};
+	struct network_helper_opts opts = {
+		.post_socket_cb	= cc_cb,
+		.cb_opts	= &cb_opts,
+	};
 	struct tcp_ca_update *skel;
 	struct bpf_link *link, *link_nl;
 	int err;
@@ -386,7 +423,7 @@ static void test_mixed_links(void)
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
 	ASSERT_OK_PTR(link, "attach_struct_ops");
 
-	do_test("tcp_ca_update", NULL);
+	do_test(&opts, NULL);
 	ASSERT_GT(skel->bss->ca1_cnt, 0, "ca1_ca1_cnt");
 
 	err = bpf_link__update_map(link, skel->maps.ca_no_link);
@@ -473,6 +510,13 @@ static void test_tcp_ca_kfunc(void)
 
 static void test_cc_cubic(void)
 {
+	struct cb_opts cb_opts = {
+		.cc = "bpf_cc_cubic",
+	};
+	struct network_helper_opts opts = {
+		.post_socket_cb	= cc_cb,
+		.cb_opts	= &cb_opts,
+	};
 	struct bpf_cc_cubic *cc_cubic_skel;
 	struct bpf_link *link;
 
@@ -486,7 +530,7 @@ static void test_cc_cubic(void)
 		return;
 	}
 
-	do_test("bpf_cc_cubic", NULL);
+	do_test(&opts, NULL);
 
 	bpf_link__destroy(link);
 	bpf_cc_cubic__destroy(cc_cubic_skel);
-- 
cgit v1.2.3-58-ga151


From 6994520a332887f1688464f250c9ec8002a89a8e Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 27 May 2024 21:16:56 +0200
Subject: r8169: disable interrupt source RxOverflow

Vendor driver calls this bit RxDescUnavail. All we do in the interrupt
handler in this case is scheduling NAPI. If we should be out of
RX descriptors, then NAPI is scheduled anyway. Therefore remove this
interrupt source. Tested on RTL8168h.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Sunil Goutham <sgoutham@marvell.com>
Link: https://lore.kernel.org/r/9b2054b2-0548-4f48-bf91-b646572093b4@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/realtek/r8169_main.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 7b9e04884575..5c4296daf8f7 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -5086,12 +5086,10 @@ static void rtl_set_irq_mask(struct rtl8169_private *tp)
 	tp->irq_mask = RxOK | RxErr | TxOK | TxErr | LinkChg;
 
 	if (tp->mac_version <= RTL_GIGA_MAC_VER_06)
-		tp->irq_mask |= SYSErr | RxOverflow | RxFIFOOver;
+		tp->irq_mask |= SYSErr | RxFIFOOver;
 	else if (tp->mac_version == RTL_GIGA_MAC_VER_11)
 		/* special workaround needed */
 		tp->irq_mask |= RxFIFOOver;
-	else
-		tp->irq_mask |= RxOverflow;
 }
 
 static int rtl_alloc_irq(struct rtl8169_private *tp)
-- 
cgit v1.2.3-58-ga151


From 982300c115d229565d7af8e8b38aa1ee7bb1f5bd Mon Sep 17 00:00:00 2001
From: Heiner Kallweit <hkallweit1@gmail.com>
Date: Mon, 27 May 2024 21:20:16 +0200
Subject: r8169: remove detection of chip version 11 (early RTL8168b)

This early RTL8168b version was the first PCIe chip version, and it's
quite quirky. Last sign of life is from more than 15 yrs ago.
Let's remove detection of this chip version, we'll see whether anybody
complains. If not, support for this chip version can be removed a few
kernel versions later.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Link: https://lore.kernel.org/r/875cdcf4-843c-420a-ad5d-417447b68572@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/realtek/r8169_main.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 5c4296daf8f7..9246ea2118ff 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -2274,7 +2274,9 @@ static enum mac_version rtl8169_get_mac_version(u16 xid, bool gmii)
 
 		/* 8168B family. */
 		{ 0x7c8, 0x380,	RTL_GIGA_MAC_VER_17 },
-		{ 0x7c8, 0x300,	RTL_GIGA_MAC_VER_11 },
+		/* This one is very old and rare, let's see if anybody complains.
+		 * { 0x7c8, 0x300,	RTL_GIGA_MAC_VER_11 },
+		 */
 
 		/* 8101 family. */
 		{ 0x7c8, 0x448,	RTL_GIGA_MAC_VER_39 },
-- 
cgit v1.2.3-58-ga151


From 874aa96d78c791565781bb38570971f614d800ff Mon Sep 17 00:00:00 2001
From: Thomas Weißschuh <linux@weissschuh.net>
Date: Mon, 27 May 2024 19:04:19 +0200
Subject: net/neighbour: constify ctl_table arguments of utility function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The sysctl core is preparing to only expose instances of
struct ctl_table as "const".
This will also affect the ctl_table argument of sysctl handlers.

As the function prototype of all sysctl handlers throughout the tree
needs to stay consistent that change will be done in one commit.

To reduce the size of that final commit, switch utility functions which
are not bound by "typedef proc_handler" to "const struct ctl_table".

No functional change.

Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
Link: https://lore.kernel.org/r/20240527-sysctl-const-handler-net-v1-1-16523767d0b2@weissschuh.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/neighbour.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 45fd88405b6b..277751375b0a 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -3578,7 +3578,7 @@ static void neigh_copy_dflt_parms(struct net *net, struct neigh_parms *p,
 	rcu_read_unlock();
 }
 
-static void neigh_proc_update(struct ctl_table *ctl, int write)
+static void neigh_proc_update(const struct ctl_table *ctl, int write)
 {
 	struct net_device *dev = ctl->extra1;
 	struct neigh_parms *p = ctl->extra2;
-- 
cgit v1.2.3-58-ga151


From 551814313f113206800b025d89d4af99b2db13a2 Mon Sep 17 00:00:00 2001
From: Thomas Weißschuh <linux@weissschuh.net>
Date: Mon, 27 May 2024 19:04:20 +0200
Subject: net/ipv4/sysctl: constify ctl_table arguments of utility functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The sysctl core is preparing to only expose instances of
struct ctl_table as "const".
This will also affect the ctl_table argument of sysctl handlers.

As the function prototype of all sysctl handlers throughout the tree
needs to stay consistent that change will be done in one commit.

To reduce the size of that final commit, switch utility functions which
are not bound by "typedef proc_handler" to "const struct ctl_table".

No functional change.

Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
Link: https://lore.kernel.org/r/20240527-sysctl-const-handler-net-v1-2-16523767d0b2@weissschuh.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/sysctl_net_ipv4.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 162a0a3b6ba5..d7892f34a15b 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -130,7 +130,8 @@ static int ipv4_privileged_ports(struct ctl_table *table, int write,
 	return ret;
 }
 
-static void inet_get_ping_group_range_table(struct ctl_table *table, kgid_t *low, kgid_t *high)
+static void inet_get_ping_group_range_table(const struct ctl_table *table,
+					    kgid_t *low, kgid_t *high)
 {
 	kgid_t *data = table->data;
 	struct net *net =
@@ -145,7 +146,8 @@ static void inet_get_ping_group_range_table(struct ctl_table *table, kgid_t *low
 }
 
 /* Update system visible IP port range */
-static void set_ping_group_range(struct ctl_table *table, kgid_t low, kgid_t high)
+static void set_ping_group_range(const struct ctl_table *table,
+				 kgid_t low, kgid_t high)
 {
 	kgid_t *data = table->data;
 	struct net *net =
-- 
cgit v1.2.3-58-ga151


From c55eb03765f413d43ee96e2c4eeac27fde8e63d8 Mon Sep 17 00:00:00 2001
From: Thomas Weißschuh <linux@weissschuh.net>
Date: Mon, 27 May 2024 19:04:21 +0200
Subject: net/ipv6/addrconf: constify ctl_table arguments of utility functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The sysctl core is preparing to only expose instances of
struct ctl_table as "const".
This will also affect the ctl_table argument of sysctl handlers.

As the function prototype of all sysctl handlers throughout the tree
needs to stay consistent that change will be done in one commit.

To reduce the size of that final commit, switch utility functions which
are not bound by "typedef proc_handler" to "const struct ctl_table".

No functional change.

Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
Link: https://lore.kernel.org/r/20240527-sysctl-const-handler-net-v1-3-16523767d0b2@weissschuh.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv6/addrconf.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 5c424a0e7232..1e69756d53d9 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -863,7 +863,7 @@ static void addrconf_forward_change(struct net *net, __s32 newf)
 	}
 }
 
-static int addrconf_fixup_forwarding(struct ctl_table *table, int *p, int newf)
+static int addrconf_fixup_forwarding(const struct ctl_table *table, int *p, int newf)
 {
 	struct net *net;
 	int old;
@@ -931,7 +931,7 @@ static void addrconf_linkdown_change(struct net *net, __s32 newf)
 	}
 }
 
-static int addrconf_fixup_linkdown(struct ctl_table *table, int *p, int newf)
+static int addrconf_fixup_linkdown(const struct ctl_table *table, int *p, int newf)
 {
 	struct net *net;
 	int old;
@@ -6378,7 +6378,7 @@ static void addrconf_disable_change(struct net *net, __s32 newf)
 	}
 }
 
-static int addrconf_disable_ipv6(struct ctl_table *table, int *p, int newf)
+static int addrconf_disable_ipv6(const struct ctl_table *table, int *p, int newf)
 {
 	struct net *net = (struct net *)table->extra2;
 	int old;
@@ -6669,7 +6669,7 @@ void addrconf_disable_policy_idev(struct inet6_dev *idev, int val)
 }
 
 static
-int addrconf_disable_policy(struct ctl_table *ctl, int *valp, int val)
+int addrconf_disable_policy(const struct ctl_table *ctl, int *valp, int val)
 {
 	struct net *net = (struct net *)ctl->extra2;
 	struct inet6_dev *idev;
-- 
cgit v1.2.3-58-ga151


From 7a20cd1e71d87907dbba8887cd434a2195699c85 Mon Sep 17 00:00:00 2001
From: Thomas Weißschuh <linux@weissschuh.net>
Date: Mon, 27 May 2024 19:04:22 +0200
Subject: net/ipv6/ndisc: constify ctl_table arguments of utility function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The sysctl core is preparing to only expose instances of
struct ctl_table as "const".
This will also affect the ctl_table argument of sysctl handlers.

As the function prototype of all sysctl handlers throughout the tree
needs to stay consistent that change will be done in one commit.

To reduce the size of that final commit, switch utility functions which
are not bound by "typedef proc_handler" to "const struct ctl_table".

No functional change.

Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
Link: https://lore.kernel.org/r/20240527-sysctl-const-handler-net-v1-4-16523767d0b2@weissschuh.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv6/ndisc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index d914b23256ce..254b192c5705 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1936,7 +1936,7 @@ static struct notifier_block ndisc_netdev_notifier = {
 };
 
 #ifdef CONFIG_SYSCTL
-static void ndisc_warn_deprecated_sysctl(struct ctl_table *ctl,
+static void ndisc_warn_deprecated_sysctl(const struct ctl_table *ctl,
 					 const char *func, const char *dev_name)
 {
 	static char warncomm[TASK_COMM_LEN];
-- 
cgit v1.2.3-58-ga151


From 0a9f788fdde4af08547455fcdc21e21b822218f2 Mon Sep 17 00:00:00 2001
From: Thomas Weißschuh <linux@weissschuh.net>
Date: Mon, 27 May 2024 19:04:23 +0200
Subject: ipvs: constify ctl_table arguments of utility functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The sysctl core is preparing to only expose instances of
struct ctl_table as "const".
This will also affect the ctl_table argument of sysctl handlers.

As the function prototype of all sysctl handlers throughout the tree
needs to stay consistent that change will be done in one commit.

To reduce the size of that final commit, switch utility functions which
are not bound by "typedef proc_handler" to "const struct ctl_table".

No functional change.

Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
Link: https://lore.kernel.org/r/20240527-sysctl-const-handler-net-v1-5-16523767d0b2@weissschuh.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/netfilter/ipvs/ip_vs_ctl.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index b6d0dcf3a5c3..78a1cc72dc38 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -1924,7 +1924,8 @@ proc_do_sync_ports(struct ctl_table *table, int write,
 	return rc;
 }
 
-static int ipvs_proc_est_cpumask_set(struct ctl_table *table, void *buffer)
+static int ipvs_proc_est_cpumask_set(const struct ctl_table *table,
+				     void *buffer)
 {
 	struct netns_ipvs *ipvs = table->extra2;
 	cpumask_var_t *valp = table->data;
@@ -1962,8 +1963,8 @@ out:
 	return ret;
 }
 
-static int ipvs_proc_est_cpumask_get(struct ctl_table *table, void *buffer,
-				     size_t size)
+static int ipvs_proc_est_cpumask_get(const struct ctl_table *table,
+				     void *buffer, size_t size)
 {
 	struct netns_ipvs *ipvs = table->extra2;
 	cpumask_var_t *valp = table->data;
-- 
cgit v1.2.3-58-ga151


From 0e49e940d1bc47e5f8937aa981337debaae69bd8 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Mon, 27 May 2024 19:05:56 +0300
Subject: wifi: iwlwifi: mvm: add an option to use ptp clock for rx timestamp

Add a debugfs option to use the ptp clock time for rx device
timestamps. This can be useful to e.g. synchronize multiple NICs or
reporting the timestamp in the system clock instead of gp2.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Reviewed-by: Gregory Greenman <gregory.greenman@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.b197b4f3f9ed.I578ee916d5f517a2e3a89e890a8ea24065946427@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c | 3 +++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h     | 3 +++
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c    | 8 ++++++++
 3 files changed, 14 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 96c9382b98fb..879640a7d46a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -2441,6 +2441,9 @@ void iwl_mvm_dbgfs_register(struct iwl_mvm *mvm)
 	debugfs_create_file("mem", 0600, mvm->debugfs_dir, mvm,
 			    &iwl_dbgfs_mem_ops);
 
+	debugfs_create_bool("rx_ts_ptp", 0600, mvm->debugfs_dir,
+			    &mvm->rx_ts_ptp);
+
 	/*
 	 * Create a symlink with mac80211. It will be removed when mac80211
 	 * exists (before the opmode exists which removes the target.)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 770a7098b0f9..879fa1065bec 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1309,6 +1309,9 @@ struct iwl_mvm {
 	struct iwl_phy_specific_cfg phy_filters;
 #endif
 
+	/* report rx timestamp in ptp clock time */
+	bool rx_ts_ptp;
+
 	unsigned long last_6ghz_passive_scan_jiffies;
 	unsigned long last_reset_or_resume_time_jiffies;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 2f9e2254fb46..5af169f3d708 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -1954,6 +1954,14 @@ static void iwl_mvm_rx_fill_status(struct iwl_mvm *mvm,
 	iwl_mvm_decode_lsig(skb, phy_data);
 
 	rx_status->device_timestamp = phy_data->gp2_on_air_rise;
+
+	if (mvm->rx_ts_ptp && mvm->monitor_on) {
+		u64 adj_time =
+			iwl_mvm_ptp_get_adj_time(mvm, phy_data->gp2_on_air_rise * NSEC_PER_USEC);
+
+		rx_status->device_timestamp = div64_u64(adj_time, NSEC_PER_USEC);
+	}
+
 	rx_status->freq = ieee80211_channel_to_frequency(phy_data->channel,
 							 rx_status->band);
 	iwl_mvm_get_signal_strength(mvm, rx_status, rate_n_flags,
-- 
cgit v1.2.3-58-ga151


From 94854648b5b0774447dace5e8802c641b24da4af Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 27 May 2024 19:05:57 +0300
Subject: wifi: iwlwifi: mvm: report 64-bit radiotap timestamp

Use the new RX_FLAG_MACTIME_IS_RTAP_TS64 in mac80211 to report
a 64-bit timestamp in the mactime for radiotap only, in case we
report the synchronized PTP clock timestamp (otherwise we really
only have 32 bits anyway).

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Reviewed-by: Gregory Greenman <gregory.greenman@intel.com>
Reviewed-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.4fc44356e7df.Ic8642d35f1090a415e09299a1bd409f51dfb6351@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 5af169f3d708..ff8d801b4e65 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -1959,7 +1959,9 @@ static void iwl_mvm_rx_fill_status(struct iwl_mvm *mvm,
 		u64 adj_time =
 			iwl_mvm_ptp_get_adj_time(mvm, phy_data->gp2_on_air_rise * NSEC_PER_USEC);
 
-		rx_status->device_timestamp = div64_u64(adj_time, NSEC_PER_USEC);
+		rx_status->mactime = div64_u64(adj_time, NSEC_PER_USEC);
+		rx_status->flag |= RX_FLAG_MACTIME_IS_RTAP_TS64;
+		rx_status->flag &= ~RX_FLAG_MACTIME;
 	}
 
 	rx_status->freq = ieee80211_channel_to_frequency(phy_data->channel,
-- 
cgit v1.2.3-58-ga151


From e1b44c8f7ba40801365dc8689f0cd88acd2ee901 Mon Sep 17 00:00:00 2001
From: Avri Altman <avri.altman@intel.com>
Date: Mon, 27 May 2024 19:05:58 +0300
Subject: wifi: iwlwifi: mvm: Enable p2p low latency

Enable p2p low latency by default.

Signed-off-by: Avri Altman <avri.altman@intel.com>
Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.21b431bdfc26.I018e330d1e6a76e967b25e9542c2260f0f6ed2a8@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/constants.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
index 3cbeaddf4358..eda6634b4fb5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
@@ -23,7 +23,7 @@
 #define IWL_MVM_WOWLAN_PS_RX_DATA_TIMEOUT	(10 * USEC_PER_MSEC)
 #define IWL_MVM_SHORT_PS_TX_DATA_TIMEOUT	(2 * 1024) /* defined in TU */
 #define IWL_MVM_SHORT_PS_RX_DATA_TIMEOUT	(40 * 1024) /* defined in TU */
-#define IWL_MVM_P2P_LOWLATENCY_PS_ENABLE	0
+#define IWL_MVM_P2P_LOWLATENCY_PS_ENABLE	1
 #define IWL_MVM_UAPSD_RX_DATA_TIMEOUT		(50 * USEC_PER_MSEC)
 #define IWL_MVM_UAPSD_TX_DATA_TIMEOUT		(50 * USEC_PER_MSEC)
 #define IWL_MVM_UAPSD_QUEUES		(IEEE80211_WMM_IE_STA_QOSINFO_AC_VO |\
-- 
cgit v1.2.3-58-ga151


From a423e37ecc3ede97621815da4923fbca5497c4a1 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 27 May 2024 19:05:59 +0300
Subject: wifi: iwlwifi: mvm: remove IWL_MVM_USE_NSSN_SYNC

This is not used and not needed anymore.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.c5edfb967e35.I34a3d54a1dc90d9b2937b0bcbbc508fe3cd41773@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/constants.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
index eda6634b4fb5..c991eaeade15 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
@@ -114,7 +114,6 @@
 #define IWL_MVM_D3_DEBUG			false
 #define IWL_MVM_USE_TWT				true
 #define IWL_MVM_AMPDU_CONSEC_DROPS_DELBA	20
-#define IWL_MVM_USE_NSSN_SYNC			0
 #define IWL_MVM_FTM_INITIATOR_ENABLE_SMOOTH     false
 #define IWL_MVM_FTM_INITIATOR_SMOOTH_ALPHA      40
 /*  20016 pSec is 6 meter RTT, meaning 3 meter range */
-- 
cgit v1.2.3-58-ga151


From fb82e76a3b3b23c4a51fe254fa78497bfa42007c Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 27 May 2024 19:06:00 +0300
Subject: wifi: iwlwifi: mvm: move a constant to constants.h

Move IWL_MVM_MIN_BEACON_INTERVAL_TU to constants.h

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.395f5b24ff82.Id78a5a404c3ae15cbbc950b1f1d7e500a8b92a43@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/constants.h | 1 +
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 2 --
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
index c991eaeade15..d840653cc907 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
@@ -123,6 +123,7 @@
 #define IWL_MVM_DISABLE_AP_FILS			false
 #define IWL_MVM_6GHZ_PASSIVE_SCAN_TIMEOUT       3000 /* in seconds */
 #define IWL_MVM_6GHZ_PASSIVE_SCAN_ASSOC_TIMEOUT 60   /* in seconds */
+#define IWL_MVM_MIN_BEACON_INTERVAL_TU		16
 #define IWL_MVM_AUTO_EML_ENABLE                 true
 #define IWL_MVM_MISSED_BEACONS_EXIT_ESR_THRESH	7
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 8ec699d9f316..28540fd4a8da 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -3716,8 +3716,6 @@ static void iwl_mvm_rs_rate_init_all_links(struct iwl_mvm *mvm,
 	}
 }
 
-#define IWL_MVM_MIN_BEACON_INTERVAL_TU 16
-
 static bool iwl_mvm_vif_conf_from_sta(struct iwl_mvm *mvm,
 				      struct ieee80211_vif *vif,
 				      struct ieee80211_sta *sta)
-- 
cgit v1.2.3-58-ga151


From 6b82f4e119d1e27632cb350a6165c02a8ec9db07 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 27 May 2024 19:06:01 +0300
Subject: wifi: iwlwifi: mvm: handle TPE advertised by AP

6 GHz BSS SP client shall respect TX power limits advertised
by the AP in TPE elements, send the data to the firmware using
the AP_TX_POWER_CONSTRAINTS_CMD command, so do that.

Co-developed-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.32f1e2e1447c.I58ac91c38585362aa42bb4a8a59c7d88e67bc40b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 11 +++++
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  | 55 ++++++++++++++++++----
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  6 +++
 3 files changed, 63 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 28540fd4a8da..7b030995a6a4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -2731,6 +2731,13 @@ iwl_mvm_bss_info_changed_station_common(struct iwl_mvm *mvm,
 
 	if (changes & BSS_CHANGED_BANDWIDTH)
 		iwl_mvm_update_link_smps(vif, link_conf);
+
+	if (changes & BSS_CHANGED_TPE) {
+		IWL_DEBUG_CALIB(mvm, "Changing TPE\n");
+		iwl_mvm_send_ap_tx_power_constraint_cmd(mvm, vif,
+							link_conf,
+							false);
+	}
 }
 
 static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
@@ -5085,6 +5092,10 @@ static int __iwl_mvm_assign_vif_chanctx(struct iwl_mvm *mvm,
 		}
 
 		iwl_mvm_update_quotas(mvm, false, NULL);
+
+		iwl_mvm_send_ap_tx_power_constraint_cmd(mvm, vif,
+							link_conf,
+							false);
 	}
 
 	goto out;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index f69f7df40a6d..3a1504e95b0f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -347,6 +347,11 @@ __iwl_mvm_mld_assign_vif_chanctx(struct iwl_mvm *mvm,
 		rcu_read_unlock();
 	}
 
+	if (vif->type == NL80211_IFTYPE_STATION)
+		iwl_mvm_send_ap_tx_power_constraint_cmd(mvm, vif,
+							link_conf,
+							false);
+
 	/* then activate */
 	ret = iwl_mvm_link_changed(mvm, vif, link_conf,
 				   LINK_CONTEXT_MODIFY_ACTIVE |
@@ -526,9 +531,37 @@ static void iwl_mvm_mld_unassign_vif_chanctx(struct ieee80211_hw *hw,
 }
 
 static void
+iwl_mvm_tpe_sta_cmd_data(struct iwl_txpower_constraints_cmd *cmd,
+			 const struct ieee80211_bss_conf *bss_info)
+{
+	u8 i;
+
+	/*
+	 * NOTE: the 0 here is IEEE80211_TPE_CAT_6GHZ_DEFAULT,
+	 * we fully ignore IEEE80211_TPE_CAT_6GHZ_SUBORDINATE
+	 */
+
+	BUILD_BUG_ON(ARRAY_SIZE(cmd->psd_pwr) !=
+		     ARRAY_SIZE(bss_info->tpe.psd_local[0].power));
+
+	/* if not valid, mac80211 puts default (max value) */
+	for (i = 0; i < ARRAY_SIZE(cmd->psd_pwr); i++)
+		cmd->psd_pwr[i] = min(bss_info->tpe.psd_local[0].power[i],
+				      bss_info->tpe.psd_reg_client[0].power[i]);
+
+	BUILD_BUG_ON(ARRAY_SIZE(cmd->eirp_pwr) !=
+		     ARRAY_SIZE(bss_info->tpe.max_local[0].power));
+
+	for (i = 0; i < ARRAY_SIZE(cmd->eirp_pwr); i++)
+		cmd->eirp_pwr[i] = min(bss_info->tpe.max_local[0].power[i],
+				       bss_info->tpe.max_reg_client[0].power[i]);
+}
+
+void
 iwl_mvm_send_ap_tx_power_constraint_cmd(struct iwl_mvm *mvm,
 					struct ieee80211_vif *vif,
-					struct ieee80211_bss_conf *bss_conf)
+					struct ieee80211_bss_conf *bss_conf,
+					bool is_ap)
 {
 	struct iwl_txpower_constraints_cmd cmd = {};
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
@@ -548,19 +581,22 @@ iwl_mvm_send_ap_tx_power_constraint_cmd(struct iwl_mvm *mvm,
 	    link_info->fw_link_id == IWL_MVM_FW_LINK_ID_INVALID)
 		return;
 
-	if (bss_conf->chanreq.oper.chan->band != NL80211_BAND_6GHZ ||
-	    bss_conf->chanreq.oper.chan->flags &
-		    IEEE80211_CHAN_NO_6GHZ_VLP_CLIENT)
+	if (bss_conf->chanreq.oper.chan->band != NL80211_BAND_6GHZ)
 		return;
 
 	cmd.link_id = cpu_to_le16(link_info->fw_link_id);
-	/*
-	 * Currently supporting VLP Soft AP only.
-	 */
-	cmd.ap_type = cpu_to_le16(IWL_6GHZ_AP_TYPE_VLP);
 	memset(cmd.psd_pwr, DEFAULT_TPE_TX_POWER, sizeof(cmd.psd_pwr));
 	memset(cmd.eirp_pwr, DEFAULT_TPE_TX_POWER, sizeof(cmd.eirp_pwr));
 
+	if (is_ap) {
+		cmd.ap_type = cpu_to_le16(IWL_6GHZ_AP_TYPE_VLP);
+	} else if (bss_conf->power_type == IEEE80211_REG_UNSET_AP) {
+		return;
+	} else {
+		cmd.ap_type = cpu_to_le16(bss_conf->power_type - 1);
+		iwl_mvm_tpe_sta_cmd_data(&cmd, bss_conf);
+	}
+
 	ret = iwl_mvm_send_cmd_pdu(mvm,
 				   WIDE_ID(PHY_OPS_GROUP,
 					   AP_TX_POWER_CONSTRAINTS_CMD),
@@ -582,7 +618,8 @@ static int iwl_mvm_mld_start_ap_ibss(struct ieee80211_hw *hw,
 	guard(mvm)(mvm);
 
 	if (vif->type == NL80211_IFTYPE_AP)
-		iwl_mvm_send_ap_tx_power_constraint_cmd(mvm, vif, link_conf);
+		iwl_mvm_send_ap_tx_power_constraint_cmd(mvm, vif,
+							link_conf, true);
 
 	/* Send the beacon template */
 	ret = iwl_mvm_mac_ctxt_beacon_changed(mvm, vif, link_conf);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 879fa1065bec..82080d920d59 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2961,4 +2961,10 @@ iwl_mvm_bt_coex_calculate_esr_mode(struct iwl_mvm *mvm,
 				   bool primary);
 int iwl_mvm_esr_non_bss_link(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			     unsigned int link_id, bool active);
+
+void
+iwl_mvm_send_ap_tx_power_constraint_cmd(struct iwl_mvm *mvm,
+					struct ieee80211_vif *vif,
+					struct ieee80211_bss_conf *bss_conf,
+					bool is_ap);
 #endif /* __IWL_MVM_H__ */
-- 
cgit v1.2.3-58-ga151


From 8a039ef63170860ddcccd615543ea000c10b444f Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Mon, 27 May 2024 19:06:02 +0300
Subject: wifi: iwlwifi: mvm: initiator: move setting target flags into a
 function

Move setting the target flags into a dedicated function to support
different versions of the target struct.
This is done as preparation for moving to the new range request
version.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.9d22b61ce589.I7dbe596b4f677638d9a48c3f39b0826a9e35bea4@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/mvm/ftm-initiator.c | 49 +++++++++++++---------
 1 file changed, 30 insertions(+), 19 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
index 72a3d71f46f0..98b1bbfc84bc 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
@@ -434,44 +434,55 @@ iwl_mvm_ftm_put_target_v2(struct iwl_mvm *mvm,
 #define FTM_PUT_FLAG(flag)	(target->initiator_ap_flags |= \
 				 cpu_to_le32(IWL_INITIATOR_AP_FLAGS_##flag))
 
+#define FTM_SET_FLAG(flag)	(*flags |= \
+				 cpu_to_le32(IWL_INITIATOR_AP_FLAGS_##flag))
+
 static void
-iwl_mvm_ftm_put_target_common(struct iwl_mvm *mvm,
-			      struct cfg80211_pmsr_request_peer *peer,
-			      struct iwl_tof_range_req_ap_entry_v6 *target)
+iwl_mvm_ftm_set_target_flags(struct iwl_mvm *mvm,
+			     struct cfg80211_pmsr_request_peer *peer,
+			     __le32 *flags)
 {
-	memcpy(target->bssid, peer->addr, ETH_ALEN);
-	target->burst_period =
-		cpu_to_le16(peer->ftm.burst_period);
-	target->samples_per_burst = peer->ftm.ftms_per_burst;
-	target->num_of_bursts = peer->ftm.num_bursts_exp;
-	target->ftmr_max_retries = peer->ftm.ftmr_retries;
-	target->initiator_ap_flags = cpu_to_le32(0);
+	*flags = cpu_to_le32(0);
 
 	if (peer->ftm.asap)
-		FTM_PUT_FLAG(ASAP);
+		FTM_SET_FLAG(ASAP);
 
 	if (peer->ftm.request_lci)
-		FTM_PUT_FLAG(LCI_REQUEST);
+		FTM_SET_FLAG(LCI_REQUEST);
 
 	if (peer->ftm.request_civicloc)
-		FTM_PUT_FLAG(CIVIC_REQUEST);
+		FTM_SET_FLAG(CIVIC_REQUEST);
 
 	if (IWL_MVM_FTM_INITIATOR_DYNACK)
-		FTM_PUT_FLAG(DYN_ACK);
+		FTM_SET_FLAG(DYN_ACK);
 
 	if (IWL_MVM_FTM_INITIATOR_ALGO == IWL_TOF_ALGO_TYPE_LINEAR_REG)
-		FTM_PUT_FLAG(ALGO_LR);
+		FTM_SET_FLAG(ALGO_LR);
 	else if (IWL_MVM_FTM_INITIATOR_ALGO == IWL_TOF_ALGO_TYPE_FFT)
-		FTM_PUT_FLAG(ALGO_FFT);
+		FTM_SET_FLAG(ALGO_FFT);
 
 	if (peer->ftm.trigger_based)
-		FTM_PUT_FLAG(TB);
+		FTM_SET_FLAG(TB);
 	else if (peer->ftm.non_trigger_based)
-		FTM_PUT_FLAG(NON_TB);
+		FTM_SET_FLAG(NON_TB);
 
 	if ((peer->ftm.trigger_based || peer->ftm.non_trigger_based) &&
 	    peer->ftm.lmr_feedback)
-		FTM_PUT_FLAG(LMR_FEEDBACK);
+		FTM_SET_FLAG(LMR_FEEDBACK);
+}
+
+static void
+iwl_mvm_ftm_put_target_common(struct iwl_mvm *mvm,
+			      struct cfg80211_pmsr_request_peer *peer,
+			      struct iwl_tof_range_req_ap_entry_v6 *target)
+{
+	memcpy(target->bssid, peer->addr, ETH_ALEN);
+	target->burst_period =
+		cpu_to_le16(peer->ftm.burst_period);
+	target->samples_per_burst = peer->ftm.ftms_per_burst;
+	target->num_of_bursts = peer->ftm.num_bursts_exp;
+	target->ftmr_max_retries = peer->ftm.ftmr_retries;
+	iwl_mvm_ftm_set_target_flags(mvm, peer, &target->initiator_ap_flags);
 }
 
 static int
-- 
cgit v1.2.3-58-ga151


From 94df6cbd4c4aea633a9b4bbfab835dcfee21578b Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 27 May 2024 19:06:03 +0300
Subject: wifi: iwlwifi: mvm: Remove debug related code

This code is intended for internal testing. Remove it.

Fixes: 30ce039094b5 ("wifi: iwlwifi: mvm: Don't allow EMLSR when the RSSI is low")
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.c3b89828efb2.I7ded6348d2fb0cb7e103c4eee033451924c3461d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/tests/links.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tests/links.c b/drivers/net/wireless/intel/iwlwifi/mvm/tests/links.c
index f49e3c98b1ba..6a680b5d03e0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tests/links.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tests/links.c
@@ -393,9 +393,6 @@ static void test_valid_link_pair(struct kunit *test)
 	chandef_a.width = params->cw_a ?: NL80211_CHAN_WIDTH_20;
 	chandef_b.width = params->cw_b ?: NL80211_CHAN_WIDTH_20;
 
-#ifdef CONFIG_IWLWIFI_SUPPORT_DEBUG_OVERRIDES
-	trans->dbg_cfg = default_dbg_config;
-#endif
 	mvm.trans = trans;
 
 	mvm.last_bt_notif.wifi_loss_low_rssi = params->bt;
-- 
cgit v1.2.3-58-ga151


From 16ec82d3473c10ea7435432a0ed01bf9125b9ca5 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Mon, 27 May 2024 19:06:04 +0300
Subject: wifi: iwlwifi: mvm: initiator: move setting the sta_id into a
 function

Move setting the target's sta_id (and related flags) into a function
to support different versions of the target struct.
This is done as preparation for moving to the new range request
version.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.18e3a6a6f1cb.I85e3ee607b3947448532bc16730f8898a11c92b8@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/mvm/ftm-initiator.c | 55 ++++++++++++----------
 1 file changed, 30 insertions(+), 25 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
index 98b1bbfc84bc..8596e6f7d1dc 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
@@ -431,9 +431,6 @@ iwl_mvm_ftm_put_target_v2(struct iwl_mvm *mvm,
 	return 0;
 }
 
-#define FTM_PUT_FLAG(flag)	(target->initiator_ap_flags |= \
-				 cpu_to_le32(IWL_INITIATOR_AP_FLAGS_##flag))
-
 #define FTM_SET_FLAG(flag)	(*flags |= \
 				 cpu_to_le32(IWL_INITIATOR_AP_FLAGS_##flag))
 
@@ -525,21 +522,10 @@ iwl_mvm_ftm_put_target_v4(struct iwl_mvm *mvm,
 	return 0;
 }
 
-static int
-iwl_mvm_ftm_put_target(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-		       struct cfg80211_pmsr_request_peer *peer,
-		       struct iwl_tof_range_req_ap_entry_v6 *target)
+static int iwl_mvm_ftm_set_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			       struct cfg80211_pmsr_request_peer *peer,
+			       u8 *sta_id, __le32 *flags)
 {
-	int ret;
-
-	ret = iwl_mvm_ftm_target_chandef_v2(mvm, peer, &target->channel_num,
-					    &target->format_bw,
-					    &target->ctrl_ch_position);
-	if (ret)
-		return ret;
-
-	iwl_mvm_ftm_put_target_common(mvm, peer, target);
-
 	if (vif->cfg.assoc) {
 		struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 		struct ieee80211_sta *sta;
@@ -551,8 +537,8 @@ iwl_mvm_ftm_put_target(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			if (memcmp(peer->addr, link_conf->bssid, ETH_ALEN))
 				continue;
 
-			target->sta_id = mvmvif->link[link_id]->ap_sta_id;
-			sta = rcu_dereference(mvm->fw_id_to_mac_id[target->sta_id]);
+			*sta_id = mvmvif->link[link_id]->ap_sta_id;
+			sta = rcu_dereference(mvm->fw_id_to_mac_id[*sta_id]);
 			if (WARN_ON_ONCE(IS_ERR_OR_NULL(sta))) {
 				rcu_read_unlock();
 				return PTR_ERR_OR_ZERO(sta);
@@ -560,23 +546,42 @@ iwl_mvm_ftm_put_target(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 
 			if (sta->mfp && (peer->ftm.trigger_based ||
 					 peer->ftm.non_trigger_based))
-				FTM_PUT_FLAG(PMF);
+				FTM_SET_FLAG(PMF);
 			break;
 		}
 		rcu_read_unlock();
 
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 		if (mvmvif->ftm_unprotected) {
-			target->sta_id = IWL_MVM_INVALID_STA;
-			target->initiator_ap_flags &=
-				~cpu_to_le32(IWL_INITIATOR_AP_FLAGS_PMF);
+			*sta_id = IWL_MVM_INVALID_STA;
+			*flags &= ~cpu_to_le32(IWL_INITIATOR_AP_FLAGS_PMF);
 		}
-
 #endif
 	} else {
-		target->sta_id = IWL_MVM_INVALID_STA;
+		*sta_id = IWL_MVM_INVALID_STA;
 	}
 
+	return 0;
+}
+
+static int
+iwl_mvm_ftm_put_target(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+		       struct cfg80211_pmsr_request_peer *peer,
+		       struct iwl_tof_range_req_ap_entry_v6 *target)
+{
+	int ret;
+
+	ret = iwl_mvm_ftm_target_chandef_v2(mvm, peer, &target->channel_num,
+					    &target->format_bw,
+					    &target->ctrl_ch_position);
+	if (ret)
+		return ret;
+
+	iwl_mvm_ftm_put_target_common(mvm, peer, target);
+
+	iwl_mvm_ftm_set_sta(mvm, vif, peer, &target->sta_id,
+			    &target->initiator_ap_flags);
+
 	/*
 	 * TODO: Beacon interval is currently unknown, so use the common value
 	 * of 100 TUs.
-- 
cgit v1.2.3-58-ga151


From 626be4bf99f6250cd66da5d311a72ad7455c5a64 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Mon, 27 May 2024 19:06:05 +0300
Subject: wifi: iwlwifi: mvm: modify iwl_mvm_ftm_set_secured_ranging()
 parameters

Modify iwl_mvm_ftm_set_secured_ranging() parameters to support
multiple versions of the target struct.
This is done as preparation for moving to the new range request
version.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.56d6aee320b3.I5a52fa93cd791d0229b392a20f076b7cebb110cd@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/mvm/ftm-initiator.c | 54 ++++++++++++++--------
 1 file changed, 34 insertions(+), 20 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
index 8596e6f7d1dc..f5bf99f7fc2e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
@@ -40,6 +40,12 @@ struct iwl_mvm_ftm_pasn_entry {
 	u32 flags;
 };
 
+struct iwl_mvm_ftm_iter_data {
+	u8 *cipher;
+	u8 *bssid;
+	u8 *tk;
+};
+
 int iwl_mvm_ftm_add_pasn_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 			     u8 *addr, u32 cipher, u8 *tk, u32 tk_len,
 			     u8 *hltk, u32 hltk_len)
@@ -719,7 +725,7 @@ static void iter(struct ieee80211_hw *hw,
 		 struct ieee80211_key_conf *key,
 		 void *data)
 {
-	struct iwl_tof_range_req_ap_entry_v6 *target = data;
+	struct iwl_mvm_ftm_iter_data *target = data;
 
 	if (!sta || memcmp(sta->addr, target->bssid, ETH_ALEN))
 		return;
@@ -730,16 +736,16 @@ static void iter(struct ieee80211_hw *hw,
 		return;
 
 	memcpy(target->tk, key->key, key->keylen);
-	target->cipher = iwl_mvm_cipher_to_location_cipher(key->cipher);
-	WARN_ON(target->cipher == IWL_LOCATION_CIPHER_INVALID);
+	*target->cipher = iwl_mvm_cipher_to_location_cipher(key->cipher);
+	WARN_ON(*target->cipher == IWL_LOCATION_CIPHER_INVALID);
 }
 
 static void
 iwl_mvm_ftm_set_secured_ranging(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
-				struct iwl_tof_range_req_ap_entry_v7 *target)
+				u8 *bssid, u8 *cipher, u8 *hltk, u8 *tk,
+				u8 *rx_pn, u8 *tx_pn, __le32 *flags)
 {
 	struct iwl_mvm_ftm_pasn_entry *entry;
-	u32 flags = le32_to_cpu(target->initiator_ap_flags);
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
@@ -747,35 +753,39 @@ iwl_mvm_ftm_set_secured_ranging(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		return;
 #endif
 
-	if (!(flags & (IWL_INITIATOR_AP_FLAGS_NON_TB |
+	if (!(le32_to_cpu(*flags) & (IWL_INITIATOR_AP_FLAGS_NON_TB |
 		       IWL_INITIATOR_AP_FLAGS_TB)))
 		return;
 
 	lockdep_assert_held(&mvm->mutex);
 
 	list_for_each_entry(entry, &mvm->ftm_initiator.pasn_list, list) {
-		if (memcmp(entry->addr, target->bssid, sizeof(entry->addr)))
+		if (memcmp(entry->addr, bssid, sizeof(entry->addr)))
 			continue;
 
-		target->cipher = entry->cipher;
+		*cipher = entry->cipher;
 
 		if (entry->flags & IWL_MVM_PASN_FLAG_HAS_HLTK)
-			memcpy(target->hltk, entry->hltk, sizeof(target->hltk));
+			memcpy(hltk, entry->hltk, sizeof(entry->hltk));
 		else
-			memset(target->hltk, 0, sizeof(target->hltk));
+			memset(hltk, 0, sizeof(entry->hltk));
 
 		if (vif->cfg.assoc &&
-		    !memcmp(vif->bss_conf.bssid, target->bssid,
-			    sizeof(target->bssid)))
-			ieee80211_iter_keys(mvm->hw, vif, iter, target);
-		else
-			memcpy(target->tk, entry->tk, sizeof(target->tk));
+		    !memcmp(vif->bss_conf.bssid, bssid, ETH_ALEN)) {
+			struct iwl_mvm_ftm_iter_data target;
+
+			target.cipher = cipher;
+			target.bssid = bssid;
+			target.tk = tk;
+			ieee80211_iter_keys(mvm->hw, vif, iter, &target);
+		} else {
+			memcpy(tk, entry->tk, sizeof(entry->tk));
+		}
 
-		memcpy(target->rx_pn, entry->rx_pn, sizeof(target->rx_pn));
-		memcpy(target->tx_pn, entry->tx_pn, sizeof(target->tx_pn));
+		memcpy(rx_pn, entry->rx_pn, sizeof(entry->rx_pn));
+		memcpy(tx_pn, entry->tx_pn, sizeof(entry->tx_pn));
 
-		target->initiator_ap_flags |=
-			cpu_to_le32(IWL_INITIATOR_AP_FLAGS_SECURED);
+		FTM_SET_FLAG(SECURED);
 		return;
 	}
 }
@@ -789,7 +799,11 @@ iwl_mvm_ftm_put_target_v7(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	if (err)
 		return err;
 
-	iwl_mvm_ftm_set_secured_ranging(mvm, vif, target);
+	iwl_mvm_ftm_set_secured_ranging(mvm, vif, target->bssid,
+					&target->cipher, target->hltk,
+					target->tk, target->rx_pn,
+					target->tx_pn,
+					&target->initiator_ap_flags);
 	return err;
 }
 
-- 
cgit v1.2.3-58-ga151


From d29fc6af1ab845fd0e9ab1e2e6060ba6146bacbb Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Mon, 27 May 2024 19:06:06 +0300
Subject: wifi: iwlwifi: mvm: add support for version 14 of the range request
 command

This version adds a testing option to send an incorrect SAC in the
first NDPA.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.8f36ccb1cc74.I9b76461634f308e75dc3cd016d0b0bce812c6e9d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/location.h   | 105 +++++++++++++++++++++
 drivers/net/wireless/intel/iwlwifi/mvm/constants.h |   1 +
 .../net/wireless/intel/iwlwifi/mvm/ftm-initiator.c | 102 ++++++++++++++++++++
 3 files changed, 208 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
index 25530a29317e..c20bb289a623 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
@@ -2,6 +2,7 @@
 /*
  * Copyright (C) 2015-2017 Intel Deutschland GmbH
  * Copyright (C) 2018-2022 Intel Corporation
+ * Copyright (C) 2024 Intel Corporation
  */
 #ifndef __iwl_fw_api_location_h__
 #define __iwl_fw_api_location_h__
@@ -561,6 +562,8 @@ struct iwl_tof_range_req_ap_entry_v2 {
  *	the responder asked for LMR feedback although the initiator did not set
  *	the LMR feedback bit in the FTM request. If not set, the initiator will
  *	continue with the session and will provide the LMR feedback.
+ * @IWL_INITIATOR_AP_FLAGS_TEST_INCORRECT_SAC: send an incorrect SAC in the
+ *	first NDP exchange. This is used for testing.
  */
 enum iwl_initiator_ap_flags {
 	IWL_INITIATOR_AP_FLAGS_ASAP = BIT(1),
@@ -577,6 +580,7 @@ enum iwl_initiator_ap_flags {
 	IWL_INITIATOR_AP_FLAGS_USE_CALIB = BIT(13),
 	IWL_INITIATOR_AP_FLAGS_PMF = BIT(14),
 	IWL_INITIATOR_AP_FLAGS_TERMINATE_ON_LMR_FEEDBACK = BIT(15),
+	IWL_INITIATOR_AP_FLAGS_TEST_INCORRECT_SAC = BIT(16),
 };
 
 /**
@@ -797,6 +801,7 @@ struct iwl_tof_range_req_ap_entry_v7 {
 } __packed; /* LOCATION_RANGE_REQ_AP_ENTRY_CMD_API_S_VER_7 */
 
 #define IWL_LOCATION_MAX_STS_POS	3
+#define IWL_LOCATION_TOTAL_LTF_POS	6
 
 /**
  * struct iwl_tof_range_req_ap_entry_v8 - AP configuration parameters
@@ -953,6 +958,78 @@ struct iwl_tof_range_req_ap_entry_v9 {
 	__le16 min_time_between_msr;
 } __packed; /* LOCATION_RANGE_REQ_AP_ENTRY_CMD_API_S_VER_9 */
 
+/**
+ * struct iwl_tof_range_req_ap_entry_v10 - AP configuration parameters
+ * @initiator_ap_flags: see &enum iwl_initiator_ap_flags.
+ * @band: 0 for 5.2 GHz, 1 for 2.4 GHz, 2 for 6GHz
+ * @channel_num: AP Channel number
+ * @format_bw: bits 0 - 3: &enum iwl_location_frame_format.
+ *             bits 4 - 7: &enum iwl_location_bw.
+ * @ctrl_ch_position: Coding of the control channel position relative to the
+ *	center frequency, see iwl_mvm_get_ctrl_pos().
+ * @bssid: AP's BSSID
+ * @burst_period: For EDCA based ranging: Recommended value to be sent to the
+ *	AP. Measurement periodicity In units of 100ms. ignored if
+ *	num_of_bursts_exp = 0.
+ *	For non trigger based NDP ranging, the maximum time between
+ *	measurements in units of milliseconds.
+ * @samples_per_burst: the number of FTMs pairs in single Burst (1-31);
+ * @num_of_bursts: Recommended value to be sent to the AP. 2s Exponent of
+ *	the number of measurement iterations (min 2^0 = 1, max 2^14)
+ * @sta_id: the station id of the AP. Only relevant when associated to the AP,
+ *	otherwise should be set to &IWL_MVM_INVALID_STA.
+ * @cipher: pairwise cipher suite for secured measurement.
+ *          &enum iwl_location_cipher.
+ * @hltk: HLTK to be used for secured 11az measurement
+ * @tk: TK to be used for secured 11az measurement
+ * @calib: An array of calibration values per FTM rx bandwidth.
+ *         If &IWL_INITIATOR_AP_FLAGS_USE_CALIB is set, the fw will use the
+ *         calibration value that corresponds to the rx bandwidth of the FTM
+ *         frame.
+ * @beacon_interval: beacon interval of the AP in TUs. Only required if
+ *	&IWL_INITIATOR_AP_FLAGS_TB is set.
+ * @rx_pn: the next expected PN for protected management frames Rx. LE byte
+ *	order. Only valid if &IWL_INITIATOR_AP_FLAGS_SECURED is set and sta_id
+ *	is set to &IWL_MVM_INVALID_STA.
+ * @tx_pn: the next PN to use for protected management frames Tx. LE byte
+ *	order. Only valid if &IWL_INITIATOR_AP_FLAGS_SECURED is set and sta_id
+ *	is set to &IWL_MVM_INVALID_STA.
+ * @r2i_ndp_params: parameters for R2I NDP ranging negotiation.
+ *      bits 0 - 2: max LTF repetitions
+ *      bits 3 - 5: max number of spatial streams
+ *      bits 6 - 7: max total LTFs. One of
+ *		&enum ieee80211_range_params_max_total_ltf.
+ * @i2r_ndp_params: parameters for I2R NDP ranging negotiation.
+ *      bits 0 - 2: max LTF repetitions
+ *      bits 3 - 5: max number of spatial streams (supported values are < 2)
+ *      bits 6 - 7: max total LTFs. One of
+ *		&enum ieee80211_range_params_max_total_ltf.
+ * @min_time_between_msr: For non trigger based NDP ranging, the minimum time
+ *	between measurements in units of milliseconds
+ */
+struct iwl_tof_range_req_ap_entry_v10 {
+	__le32 initiator_ap_flags;
+	u8 band;
+	u8 channel_num;
+	u8 format_bw;
+	u8 ctrl_ch_position;
+	u8 bssid[ETH_ALEN];
+	__le16 burst_period;
+	u8 samples_per_burst;
+	u8 num_of_bursts;
+	u8 sta_id;
+	u8 cipher;
+	u8 hltk[HLTK_11AZ_LEN];
+	u8 tk[TK_11AZ_LEN];
+	__le16 calib[IWL_TOF_BW_NUM];
+	__le16 beacon_interval;
+	u8 rx_pn[IEEE80211_CCMP_PN_LEN];
+	u8 tx_pn[IEEE80211_CCMP_PN_LEN];
+	u8 r2i_ndp_params;
+	u8 i2r_ndp_params;
+	__le16 min_time_between_msr;
+} __packed; /* LOCATION_RANGE_REQ_AP_ENTRY_CMD_API_S_VER_9 */
+
 /**
  * enum iwl_tof_response_mode
  * @IWL_MVM_TOF_RESPONSE_ASAP: report each AP measurement separately as soon as
@@ -1230,6 +1307,34 @@ struct iwl_tof_range_req_cmd_v13 {
 	struct iwl_tof_range_req_ap_entry_v9 ap[IWL_MVM_TOF_MAX_APS];
 } __packed; /* LOCATION_RANGE_REQ_CMD_API_S_VER_13 */
 
+/**
+ * struct iwl_tof_range_req_cmd_v14 - start measurement cmd
+ * @initiator_flags: see flags @ iwl_tof_initiator_flags
+ * @request_id: A Token incremented per request. The same Token will be
+ *		sent back in the range response
+ * @num_of_ap: Number of APs to measure (error if > IWL_MVM_TOF_MAX_APS)
+ * @range_req_bssid: ranging request BSSID
+ * @macaddr_mask: Bits set to 0 shall be copied from the MAC address template.
+ *		  Bits set to 1 shall be randomized by the UMAC
+ * @macaddr_template: MAC address template to use for non-randomized bits
+ * @req_timeout_ms: Requested timeout of the response in units of milliseconds.
+ *	This is the session time for completing the measurement.
+ * @tsf_mac_id: report the measurement start time for each ap in terms of the
+ *	TSF of this mac id. 0xff to disable TSF reporting.
+ * @ap: per-AP request data, see &struct iwl_tof_range_req_ap_entry_v10.
+ */
+struct iwl_tof_range_req_cmd_v14 {
+	__le32 initiator_flags;
+	u8 request_id;
+	u8 num_of_ap;
+	u8 range_req_bssid[ETH_ALEN];
+	u8 macaddr_mask[ETH_ALEN];
+	u8 macaddr_template[ETH_ALEN];
+	__le32 req_timeout_ms;
+	__le32 tsf_mac_id;
+	struct iwl_tof_range_req_ap_entry_v10 ap[IWL_MVM_TOF_MAX_APS];
+} __packed; /* LOCATION_RANGE_REQ_CMD_API_S_VER_13 */
+
 /*
  * enum iwl_tof_range_request_status - status of the sent request
  * @IWL_TOF_RANGE_REQUEST_STATUS_SUCCESSFUL - FW successfully received the
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
index d840653cc907..1abe9e9fd46b 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
@@ -100,6 +100,7 @@
 #define IWL_MVM_FTM_INITIATOR_ALGO		IWL_TOF_ALGO_TYPE_MAX_LIKE
 #define IWL_MVM_FTM_INITIATOR_DYNACK		true
 #define IWL_MVM_FTM_LMR_FEEDBACK_TERMINATE	false
+#define IWL_MVM_FTM_TEST_INCORRECT_SAC	        false
 #define IWL_MVM_FTM_R2I_MAX_REP			7
 #define IWL_MVM_FTM_I2R_MAX_REP			7
 #define IWL_MVM_FTM_R2I_MAX_STS			1
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
index f5bf99f7fc2e..00d447e40cc6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
@@ -950,6 +950,105 @@ static int iwl_mvm_ftm_start_v13(struct iwl_mvm *mvm,
 	return iwl_mvm_ftm_send_cmd(mvm, &hcmd);
 }
 
+static int
+iwl_mvm_ftm_put_target_v10(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
+			   struct cfg80211_pmsr_request_peer *peer,
+			   struct iwl_tof_range_req_ap_entry_v10 *target)
+{
+	u32 i2r_max_sts, flags;
+	int ret;
+
+	ret = iwl_mvm_ftm_target_chandef_v2(mvm, peer, &target->channel_num,
+					    &target->format_bw,
+					    &target->ctrl_ch_position);
+	if (ret)
+		return ret;
+
+	memcpy(target->bssid, peer->addr, ETH_ALEN);
+	target->burst_period =
+		cpu_to_le16(peer->ftm.burst_period);
+	target->samples_per_burst = peer->ftm.ftms_per_burst;
+	target->num_of_bursts = peer->ftm.num_bursts_exp;
+	iwl_mvm_ftm_set_target_flags(mvm, peer, &target->initiator_ap_flags);
+	iwl_mvm_ftm_set_sta(mvm, vif, peer, &target->sta_id,
+			    &target->initiator_ap_flags);
+	iwl_mvm_ftm_set_secured_ranging(mvm, vif, target->bssid,
+					&target->cipher, target->hltk,
+					target->tk, target->rx_pn,
+					target->tx_pn,
+					&target->initiator_ap_flags);
+
+	i2r_max_sts = IWL_MVM_FTM_I2R_MAX_STS > 1 ? 1 :
+		IWL_MVM_FTM_I2R_MAX_STS;
+
+	target->r2i_ndp_params = IWL_MVM_FTM_R2I_MAX_REP |
+		(IWL_MVM_FTM_R2I_MAX_STS << IWL_LOCATION_MAX_STS_POS) |
+		(IWL_MVM_FTM_R2I_MAX_TOTAL_LTF << IWL_LOCATION_TOTAL_LTF_POS);
+	target->i2r_ndp_params = IWL_MVM_FTM_I2R_MAX_REP |
+		(i2r_max_sts << IWL_LOCATION_MAX_STS_POS) |
+		(IWL_MVM_FTM_I2R_MAX_TOTAL_LTF << IWL_LOCATION_TOTAL_LTF_POS);
+
+	if (peer->ftm.non_trigger_based) {
+		target->min_time_between_msr =
+			cpu_to_le16(IWL_MVM_FTM_NON_TB_MIN_TIME_BETWEEN_MSR);
+		target->burst_period =
+			cpu_to_le16(IWL_MVM_FTM_NON_TB_MAX_TIME_BETWEEN_MSR);
+	} else {
+		target->min_time_between_msr = cpu_to_le16(0);
+	}
+
+	target->band =
+		iwl_mvm_phy_band_from_nl80211(peer->chandef.chan->band);
+
+	/*
+	 * TODO: Beacon interval is currently unknown, so use the common value
+	 * of 100 TUs.
+	 */
+	target->beacon_interval = cpu_to_le16(100);
+
+	/*
+	 * If secure LTF is turned off, replace the flag with PMF only
+	 */
+	flags = le32_to_cpu(target->initiator_ap_flags);
+	if (flags & IWL_INITIATOR_AP_FLAGS_SECURED) {
+		if (!IWL_MVM_FTM_INITIATOR_SECURE_LTF)
+			flags &= ~IWL_INITIATOR_AP_FLAGS_SECURED;
+
+		flags |= IWL_INITIATOR_AP_FLAGS_PMF;
+		target->initiator_ap_flags = cpu_to_le32(flags);
+	}
+
+	return 0;
+}
+
+static int iwl_mvm_ftm_start_v14(struct iwl_mvm *mvm,
+				 struct ieee80211_vif *vif,
+				 struct cfg80211_pmsr_request *req)
+{
+	struct iwl_tof_range_req_cmd_v14 cmd;
+	struct iwl_host_cmd hcmd = {
+		.id = WIDE_ID(LOCATION_GROUP, TOF_RANGE_REQ_CMD),
+		.dataflags[0] = IWL_HCMD_DFL_DUP,
+		.data[0] = &cmd,
+		.len[0] = sizeof(cmd),
+	};
+	u8 i;
+	int err;
+
+	iwl_mvm_ftm_cmd_common(mvm, vif, (void *)&cmd, req);
+
+	for (i = 0; i < cmd.num_of_ap; i++) {
+		struct cfg80211_pmsr_request_peer *peer = &req->peers[i];
+		struct iwl_tof_range_req_ap_entry_v10 *target = &cmd.ap[i];
+
+		err = iwl_mvm_ftm_put_target_v10(mvm, vif, peer, target);
+		if (err)
+			return err;
+	}
+
+	return iwl_mvm_ftm_send_cmd(mvm, &hcmd);
+}
+
 int iwl_mvm_ftm_start(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		      struct cfg80211_pmsr_request *req)
 {
@@ -968,6 +1067,9 @@ int iwl_mvm_ftm_start(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 						   IWL_FW_CMD_VER_UNKNOWN);
 
 		switch (cmd_ver) {
+		case 14:
+			err = iwl_mvm_ftm_start_v14(mvm, vif, req);
+			break;
 		case 13:
 			err = iwl_mvm_ftm_start_v13(mvm, vif, req);
 			break;
-- 
cgit v1.2.3-58-ga151


From f1c9ba4403dd5b0abe7a9c4bbb55e09f1ee98bd0 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Mon, 27 May 2024 19:06:07 +0300
Subject: wifi: iwlwifi: mvm: add support for version 10 of the responder
 config command

This version adds the band to the responder config command.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.af09129ebd74.I9356e2504a4c19961d4856494416ae49b36bfe62@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/location.h   | 52 ++++++++++++++++++++++
 .../net/wireless/intel/iwlwifi/mvm/ftm-responder.c | 15 ++++---
 2 files changed, 62 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
index c20bb289a623..7c158a8dfed0 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
@@ -391,6 +391,58 @@ struct iwl_tof_responder_config_cmd_v9 {
 	__le16 max_time_between_msr;
 } __packed; /* TOF_RESPONDER_CONFIG_CMD_API_S_VER_8 */
 
+/**
+ * struct iwl_tof_responder_config_cmd - ToF AP mode
+ * @cmd_valid_fields: &iwl_tof_responder_cmd_valid_field
+ * @responder_cfg_flags: &iwl_tof_responder_cfg_flags
+ * @format_bw: bits 0 - 3: &enum iwl_location_frame_format.
+ *             bits 4 - 7: &enum iwl_location_bw.
+ * @bss_color: current AP bss_color
+ * @channel_num: current AP Channel
+ * @ctrl_ch_position: coding of the control channel position relative to
+ *	the center frequency, see iwl_mvm_get_ctrl_pos()
+ * @sta_id: index of the AP STA when in AP mode
+ * @band: current AP band
+ * @toa_offset: Artificial addition [pSec] for the ToA - to be used for debug
+ *	purposes, simulating station movement by adding various values
+ *	to this field
+ * @common_calib: XVT: common calibration value
+ * @specific_calib: XVT: specific calibration value
+ * @bssid: Current AP BSSID
+ * @r2i_ndp_params: parameters for R2I NDP.
+ *	bits 0 - 2: max number of LTF repetitions
+ *	bits 3 - 5: max number of spatial streams (supported values are < 2)
+ *	bits 6 - 7: max number of total LTFs see
+ *	&enum ieee80211_range_params_max_total_ltf
+ * @i2r_ndp_params: parameters for I2R NDP.
+ *	bits 0 - 2: max number of LTF repetitions
+ *	bits 3 - 5: max number of spatial streams
+ *	bits 6 - 7: max number of total LTFs see
+ *	&enum ieee80211_range_params_max_total_ltf
+ * @min_time_between_msr: for non trigger based NDP ranging, minimum time
+ *	between measurements in milliseconds.
+ * @max_time_between_msr: for non trigger based NDP ranging, maximum time
+ *	between measurements in milliseconds.
+ */
+struct iwl_tof_responder_config_cmd {
+	__le32 cmd_valid_fields;
+	__le32 responder_cfg_flags;
+	u8 format_bw;
+	u8 bss_color;
+	u8 channel_num;
+	u8 ctrl_ch_position;
+	u8 sta_id;
+	u8 band;
+	__le16 toa_offset;
+	__le16 common_calib;
+	__le16 specific_calib;
+	u8 bssid[ETH_ALEN];
+	u8 r2i_ndp_params;
+	u8 i2r_ndp_params;
+	__le16 min_time_between_msr;
+	__le16 max_time_between_msr;
+} __packed; /* TOF_RESPONDER_CONFIG_CMD_API_S_VER_10 */
+
 #define IWL_LCI_CIVIC_IE_MAX_SIZE	400
 
 /**
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
index 8e760300a1ab..e4caa362f597 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-responder.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
  * Copyright (C) 2015-2017 Intel Deutschland GmbH
- * Copyright (C) 2018-2023 Intel Corporation
+ * Copyright (C) 2018-2024 Intel Corporation
  */
 #include <net/cfg80211.h>
 #include <linux/etherdevice.h>
@@ -88,7 +88,7 @@ static int iwl_mvm_ftm_responder_set_bw_v2(struct cfg80211_chan_def *chandef,
 
 static void
 iwl_mvm_ftm_responder_set_ndp(struct iwl_mvm *mvm,
-			      struct iwl_tof_responder_config_cmd_v9 *cmd)
+			      struct iwl_tof_responder_config_cmd *cmd)
 {
 	/* Up to 2 R2I STS are allowed on the responder */
 	u32 r2i_max_sts = IWL_MVM_FTM_R2I_MAX_STS < 2 ?
@@ -117,7 +117,7 @@ iwl_mvm_ftm_responder_cmd(struct iwl_mvm *mvm,
 	 * field interpretation is different), so the same struct can be use
 	 * for all cases.
 	 */
-	struct iwl_tof_responder_config_cmd_v9 cmd = {
+	struct iwl_tof_responder_config_cmd cmd = {
 		.channel_num = chandef->chan->hw_value,
 		.cmd_valid_fields =
 			cpu_to_le32(IWL_TOF_RESPONDER_CMD_VALID_CHAN_INFO |
@@ -131,8 +131,13 @@ iwl_mvm_ftm_responder_cmd(struct iwl_mvm *mvm,
 
 	lockdep_assert_held(&mvm->mutex);
 
+	if (cmd_ver == 10) {
+		cmd.band =
+			iwl_mvm_phy_band_from_nl80211(chandef->chan->band);
+	}
+
 	/* Use a default of bss_color=1 for now */
-	if (cmd_ver == 9) {
+	if (cmd_ver >= 9) {
 		cmd.cmd_valid_fields |=
 			cpu_to_le32(IWL_TOF_RESPONDER_CMD_VALID_BSS_COLOR |
 				    IWL_TOF_RESPONDER_CMD_VALID_MIN_MAX_TIME_BETWEEN_MSR);
@@ -148,7 +153,7 @@ iwl_mvm_ftm_responder_cmd(struct iwl_mvm *mvm,
 	}
 
 	if (cmd_ver >= 8)
-		iwl_mvm_ftm_responder_set_ndp(mvm, &cmd);
+		iwl_mvm_ftm_responder_set_ndp(mvm, (void *)&cmd);
 
 	if (cmd_ver >= 7)
 		err = iwl_mvm_ftm_responder_set_bw_v2(chandef, &cmd.format_bw,
-- 
cgit v1.2.3-58-ga151


From d9b7531fe98a0c2a0387270557cf0a514c28dd28 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 27 May 2024 19:06:08 +0300
Subject: wifi: iwlwifi: mvm: use only beacon BSS load for active links

For active links, don't take information that may have been
received in probe responses, as those are not protected. For
inactive links, there may not be a choice.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.6947dc462fa9.I2076961211d6785c8a15b4308e0e87a413148222@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/link.c        | 14 ++++++++++++--
 drivers/net/wireless/intel/iwlwifi/mvm/tests/links.c |  1 +
 2 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 0745229c8c37..02a475a1f1b8 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -492,17 +492,27 @@ iwl_mvm_get_puncturing_factor(const struct ieee80211_bss_conf *link_conf)
 static unsigned int
 iwl_mvm_get_chan_load(struct ieee80211_bss_conf *link_conf)
 {
+	struct ieee80211_vif *vif = link_conf->vif;
 	struct iwl_mvm_vif_link_info *mvm_link =
 		iwl_mvm_vif_from_mac80211(link_conf->vif)->link[link_conf->link_id];
 	const struct element *bss_load_elem;
 	const struct ieee80211_bss_load_elem *bss_load;
 	enum nl80211_band band = link_conf->chanreq.oper.chan->band;
+	const struct cfg80211_bss_ies *ies;
 	unsigned int chan_load;
 	u32 chan_load_by_us;
 
 	rcu_read_lock();
-	bss_load_elem = ieee80211_bss_get_elem(link_conf->bss,
-					       WLAN_EID_QBSS_LOAD);
+	if (ieee80211_vif_link_active(vif, link_conf->link_id))
+		ies = rcu_dereference(link_conf->bss->beacon_ies);
+	else
+		ies = rcu_dereference(link_conf->bss->ies);
+
+	if (ies)
+		bss_load_elem = cfg80211_find_elem(WLAN_EID_QBSS_LOAD,
+						   ies->data, ies->len);
+	else
+		bss_load_elem = NULL;
 
 	/* If there isn't BSS Load element, take the defaults */
 	if (!bss_load_elem ||
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tests/links.c b/drivers/net/wireless/intel/iwlwifi/mvm/tests/links.c
index 6a680b5d03e0..47b8e7b64ead 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tests/links.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tests/links.c
@@ -208,6 +208,7 @@ static void setup_link_conf(struct kunit *test)
 	bss_load->channel_util = params->channel_util;
 
 	rcu_assign_pointer(bss.ies, ies);
+	rcu_assign_pointer(bss.beacon_ies, ies);
 }
 
 static void test_link_grading(struct kunit *test)
-- 
cgit v1.2.3-58-ga151


From 6958c4becdf89c2877edd901458ae18ecd06169f Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 27 May 2024 19:06:09 +0300
Subject: wifi: iwlwifi: mvm: add debug data for MPDU counting

It is hard to debug issues of EMLSR entry/exit due to low throughput.
Add debug data.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240527190228.77ef3c2654dc.I1796a3995da2a49dd5102d33766af1ad416dd60b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c | 2 ++
 drivers/net/wireless/intel/iwlwifi/mvm/rx.c           | 3 +++
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c          | 3 +++
 3 files changed, 8 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index 3a1504e95b0f..d972fcbc2cf7 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -224,6 +224,8 @@ static void iwl_mvm_restart_mpdu_count(struct iwl_mvm *mvm,
 		mvmsta->mpdu_counters[q].window_start = jiffies;
 		spin_unlock_bh(&mvmsta->mpdu_counters[q].lock);
 	}
+
+	IWL_DEBUG_STATS(mvm, "MPDU counters are cleared\n");
 }
 
 static int iwl_mvm_esr_mode_active(struct iwl_mvm *mvm,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
index 4fa8066a89b6..b7185ddcca87 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c
@@ -1010,6 +1010,9 @@ static void iwl_mvm_update_esr_mode_tpt(struct iwl_mvm *mvm)
 		spin_unlock_bh(&mvmsta->mpdu_counters[q].lock);
 	}
 
+	IWL_DEBUG_STATS(mvm, "total Tx MPDUs: %ld. total Rx MPDUs: %ld\n",
+			total_tx, total_rx);
+
 	/* If we don't have enough MPDUs - exit EMLSR */
 	if (total_tx < IWL_MVM_ENTER_ESR_TPT_THRESH &&
 	    total_rx < IWL_MVM_ENTER_ESR_TPT_THRESH) {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 20d4968d692a..276bcccb7a38 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -4429,6 +4429,7 @@ void iwl_mvm_count_mpdu(struct iwl_mvm_sta *mvm_sta, u8 fw_sta_id, u32 count,
 			bool tx, int queue)
 {
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(mvm_sta->vif);
+	struct iwl_mvm *mvm = mvmvif->mvm;
 	struct iwl_mvm_tpt_counter *queue_counter;
 	struct iwl_mvm_mpdu_counter *link_counter;
 	u32 total_mpdus = 0;
@@ -4465,6 +4466,8 @@ void iwl_mvm_count_mpdu(struct iwl_mvm_sta *mvm_sta, u8 fw_sta_id, u32 count,
 		memset(queue_counter->per_link, 0,
 		       sizeof(queue_counter->per_link));
 		queue_counter->window_start = jiffies;
+
+		IWL_DEBUG_STATS(mvm, "MPDU counters are cleared\n");
 	}
 
 	for (int i = 0; i < IWL_MVM_FW_MAX_LINK_ID; i++)
-- 
cgit v1.2.3-58-ga151


From d9196023b2c2528af96eeae9aee28f25ea112b43 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Mon, 27 May 2024 19:06:10 +0300
Subject: wifi: iwlwifi: mvm: show full firmware ID in debugfs

The firmware prefix is unused today, but it might still be
useful to have some information. Since the prefix will get
the intel/ directory in some cases, replace the prefix by
the full FW id which has all the information.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240527190228.527a1c72996d.If7588b854149d51605031fc9a70a650534351ef4@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index 879640a7d46a..2b2b5b76b5ec 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -754,7 +754,6 @@ static ssize_t iwl_dbgfs_fw_ver_read(struct file *file, char __user *user_buf,
 	struct iwl_mvm *mvm = file->private_data;
 	char *buff, *pos, *endpos;
 	static const size_t bufsz = 1024;
-	char _fw_name_pre[FW_NAME_PRE_BUFSIZE];
 	int ret;
 
 	buff = kmalloc(bufsz, GFP_KERNEL);
@@ -764,8 +763,8 @@ static ssize_t iwl_dbgfs_fw_ver_read(struct file *file, char __user *user_buf,
 	pos = buff;
 	endpos = pos + bufsz;
 
-	pos += scnprintf(pos, endpos - pos, "FW prefix: %s\n",
-			 iwl_drv_get_fwname_pre(mvm->trans, _fw_name_pre));
+	pos += scnprintf(pos, endpos - pos, "FW id: %s\n",
+			 mvm->fwrt.fw->fw_version);
 	pos += scnprintf(pos, endpos - pos, "FW: %s\n",
 			 mvm->fwrt.fw->human_readable);
 	pos += scnprintf(pos, endpos - pos, "Device: %s\n",
-- 
cgit v1.2.3-58-ga151


From 05ce6e9b059d8a45b97848bceccd495a98509e4a Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 27 May 2024 19:06:11 +0300
Subject: wifi: iwlwifi: mvm: declare band variable in the scope

band is not used outside of the for loop, declare it inside it.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240527190228.b634d184980a.I9d9e71125f9a77ba4f9a33a7a2ff1e8e78e17767@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index 5132964661c8..712dabb30acc 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -1730,7 +1730,6 @@ iwl_parse_nvm_mcc_info(struct device *dev, const struct iwl_cfg *cfg,
 	const u16 *nvm_chan;
 	struct ieee80211_regdomain *regd, *copy_rd;
 	struct ieee80211_reg_rule *rule;
-	enum nl80211_band band;
 	int center_freq, prev_center_freq = 0;
 	int valid_rules = 0;
 	bool new_rule;
@@ -1774,8 +1773,10 @@ iwl_parse_nvm_mcc_info(struct device *dev, const struct iwl_cfg *cfg,
 	reg_capa = iwl_get_reg_capa(cap, resp_ver);
 
 	for (ch_idx = 0; ch_idx < num_of_ch; ch_idx++) {
+		enum nl80211_band band =
+			iwl_nl80211_band_from_channel_idx(ch_idx);
+
 		ch_flags = (u16)__le32_to_cpup(channels + ch_idx);
-		band = iwl_nl80211_band_from_channel_idx(ch_idx);
 		center_freq = ieee80211_channel_to_frequency(nvm_chan[ch_idx],
 							     band);
 		new_rule = false;
-- 
cgit v1.2.3-58-ga151


From db680c60a89ccb36e11fa561be2583764fc8a21e Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 27 May 2024 19:06:12 +0300
Subject: wifi: iwlwifi: mvm: fix a wrong comment

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240527190228.2105ffcf747d.Ic838959b812b6cb4cbb856e8c0bcaad2f46ac71b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
index 3b2875f2cab7..5e17559c7bd4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-sta.c
@@ -241,7 +241,7 @@ int iwl_mvm_mld_add_bcast_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 				       IWL_MAX_TID_COUNT, &wdg_timeout);
 }
 
-/* Allocate a new station entry for the broadcast station to the given vif,
+/* Allocate a new station entry for the multicast station to the given vif,
  * and send it to the FW.
  * Note that each AP/GO mac should have its own multicast station.
  */
-- 
cgit v1.2.3-58-ga151


From e506f8e755158560c151b1768a954e3c6c28bb34 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 27 May 2024 19:06:13 +0300
Subject: wifi: iwlwifi: remove redundant prints

Upon start, the different opmodes are currently printing:
1. HW rev, which is already print by iwlwifi
2. The HW name (e.g. "Intel(R) Wi-Fi 7 BE201 320MHz")

cleanup things such as the hw rev won't be printed again, the HW name will
be printed by iwlwifi instead of each opmode.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Link: https://msgid.link/20240527190228.5c589c7abed0.I3590c4ee0ee99d1b207852c32d25d326afb327dd@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c  | 3 ---
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 2 ++
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 516decc647dd..ed9487062093 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -1450,9 +1450,6 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 		goto out_free;
 	}
 
-	IWL_INFO(mvm, "Detected %s, REV=0x%X\n",
-		 mvm->trans->name, mvm->trans->hw_rev);
-
 	if (iwlwifi_mod_params.nvm_file)
 		mvm->nvm_file_name = iwlwifi_mod_params.nvm_file;
 	else
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index fed2754be680..92e48cff49de 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -1476,6 +1476,8 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (!iwl_trans->name)
 		iwl_trans->name = iwl_trans->cfg->name;
 
+	IWL_INFO(iwl_trans, "Detected %s\n", iwl_trans->name);
+
 	if (iwl_trans->trans_cfg->mq_rx_supported) {
 		if (WARN_ON(!iwl_trans->cfg->num_rbds)) {
 			ret = -EINVAL;
-- 
cgit v1.2.3-58-ga151


From 5860c6f59220a15bf7b6cfc674a6cbb6dc861df8 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 27 May 2024 19:06:14 +0300
Subject: wifi: iwlwifi: move amsdu_size parsing to iwlwifi

The code that is parsing the amsdu_size module parameter and mapping it
to the corresponding Rx buffer size is common to all opmodes.
Move it into a function in iwlwifi, as preparation to a new op mode we
are working on.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240527190228.a3430d32923d.Iab3c22ef0df685f72f22dafc47021f0dc7bd6fa5@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-modparams.h | 21 ++++++++++++++++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c       | 19 +------------------
 2 files changed, 21 insertions(+), 19 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h
index 1cf26ab4f488..21eabfc3ffc8 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2005-2014, 2018-2022 Intel Corporation
+ * Copyright (C) 2005-2014, 2018-2022, 2024 Intel Corporation
  */
 #ifndef __iwl_modparams_h__
 #define __iwl_modparams_h__
@@ -106,4 +106,23 @@ static inline bool iwl_enable_tx_ampdu(void)
 	return true;
 }
 
+/* Verify amsdu_size module parameter and convert it to a rxb size */
+static inline enum iwl_amsdu_size
+iwl_amsdu_size_to_rxb_size(void)
+{
+	switch (iwlwifi_mod_params.amsdu_size) {
+	case IWL_AMSDU_8K:
+		return IWL_AMSDU_8K;
+	case IWL_AMSDU_12K:
+		return IWL_AMSDU_12K;
+	default:
+		pr_err("%s: Unsupported amsdu_size: %d\n", KBUILD_MODNAME,
+		       iwlwifi_mod_params.amsdu_size);
+		fallthrough;
+	case IWL_AMSDU_DEF:
+	case IWL_AMSDU_4K:
+		return IWL_AMSDU_4K;
+	}
+}
+
 #endif /* #__iwl_modparams_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index ed9487062093..85178fe1b7ea 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -1373,24 +1373,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	trans_cfg.no_reclaim_cmds = no_reclaim_cmds;
 	trans_cfg.n_no_reclaim_cmds = ARRAY_SIZE(no_reclaim_cmds);
 
-	switch (iwlwifi_mod_params.amsdu_size) {
-	case IWL_AMSDU_DEF:
-		trans_cfg.rx_buf_size = IWL_AMSDU_4K;
-		break;
-	case IWL_AMSDU_4K:
-		trans_cfg.rx_buf_size = IWL_AMSDU_4K;
-		break;
-	case IWL_AMSDU_8K:
-		trans_cfg.rx_buf_size = IWL_AMSDU_8K;
-		break;
-	case IWL_AMSDU_12K:
-		trans_cfg.rx_buf_size = IWL_AMSDU_12K;
-		break;
-	default:
-		pr_err("%s: Unsupported amsdu_size: %d\n", KBUILD_MODNAME,
-		       iwlwifi_mod_params.amsdu_size);
-		trans_cfg.rx_buf_size = IWL_AMSDU_4K;
-	}
+	trans_cfg.rx_buf_size = iwl_amsdu_size_to_rxb_size();
 
 	trans->wide_cmd_header = true;
 	trans_cfg.bc_table_dword =
-- 
cgit v1.2.3-58-ga151


From fca7018d4b47504b9b23e847fd54b1f85f2371c6 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Mon, 27 May 2024 19:06:15 +0300
Subject: wifi: iwlwifi: move Bz and Gl iwl_dev_info entries

The iwl_dev_info entries for these devices were incorrectly positioned
within the array, out of chronological order. Move them arround.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Link: https://msgid.link/20240527190228.a75e9a1eb86c.I003980f8ca60a96019657e396e848a07df6b92a4@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 52 +++++++++++++--------------
 1 file changed, 26 insertions(+), 26 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index 92e48cff49de..d6da25e24818 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -997,32 +997,6 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = {
 		      IWL_CFG_160, IWL_CFG_ANY, IWL_CFG_CDB,
 		      iwlax411_2ax_cfg_so_gf4_a0, iwl_ax411_name),
 
-/* Bz */
-/* FIXME: need to change the naming according to the actual CRF */
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
-		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
-		      iwl_cfg_bz, iwl_fm_name),
-
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_BZ_W, IWL_CFG_ANY,
-		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
-		      iwl_cfg_bz, iwl_fm_name),
-
-/* Ga (Gl) */
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_320, IWL_CFG_ANY, IWL_CFG_NO_CDB,
-		      iwl_cfg_gl, iwl_gl_name),
-	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
-		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
-		      IWL_CFG_NO_320, IWL_CFG_ANY, IWL_CFG_NO_CDB,
-		      iwl_cfg_gl, iwl_mtp_name),
-
 /* SoF with JF2 */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SOF, IWL_CFG_ANY,
@@ -1103,6 +1077,32 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct iwl_dev_info iwl_dev_info_table[] = {
 		      IWL_CFG_NO_160, IWL_CFG_CORES_BT, IWL_CFG_NO_CDB,
 		      iwlax210_2ax_cfg_so_jf_b0, iwl9462_name),
 
+/* Bz */
+/* FIXME: need to change the naming according to the actual CRF */
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_BZ, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      iwl_cfg_bz, iwl_fm_name),
+
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_BZ_W, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_ANY, IWL_CFG_ANY, IWL_CFG_ANY,
+		      iwl_cfg_bz, iwl_fm_name),
+
+/* Ga (Gl) */
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_320, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      iwl_cfg_gl, iwl_gl_name),
+	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_MAC_TYPE_GL, IWL_CFG_ANY,
+		      IWL_CFG_RF_TYPE_FM, IWL_CFG_ANY, IWL_CFG_ANY,
+		      IWL_CFG_NO_320, IWL_CFG_ANY, IWL_CFG_NO_CDB,
+		      iwl_cfg_gl, iwl_mtp_name),
+
 /* Sc */
 	_IWL_DEV_INFO(IWL_CFG_ANY, IWL_CFG_ANY,
 		      IWL_CFG_MAC_TYPE_SC, IWL_CFG_ANY,
-- 
cgit v1.2.3-58-ga151


From a0ca76e5b7d550fcd74753d5fdaaf23f1a9bfdb4 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 23 May 2024 12:11:40 +0200
Subject: wifi: mac80211: check ieee80211_bss_info_change_notify() against MLD

It's not valid to call ieee80211_bss_info_change_notify() with
an sdata that's an MLD, remove the FIXME comment (it's not true)
and add a warning.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240523121140.97a589b13d24.I61988788d81fb3cf97a490dfd3167f67a141d1fd@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 4eaea0a9975b..40fbf397ce74 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -337,6 +337,8 @@ void ieee80211_bss_info_change_notify(struct ieee80211_sub_if_data *sdata,
 
 	might_sleep();
 
+	WARN_ON_ONCE(ieee80211_vif_is_mld(&sdata->vif));
+
 	if (!changed || sdata->vif.type == NL80211_IFTYPE_AP_VLAN)
 		return;
 
@@ -369,7 +371,6 @@ void ieee80211_bss_info_change_notify(struct ieee80211_sub_if_data *sdata,
 	if (changed & ~BSS_CHANGED_VIF_CFG_FLAGS) {
 		u64 ch = changed & ~BSS_CHANGED_VIF_CFG_FLAGS;
 
-		/* FIXME: should be for each link */
 		trace_drv_link_info_changed(local, sdata, &sdata->vif.bss_conf,
 					    changed);
 		if (local->ops->link_info_changed)
-- 
cgit v1.2.3-58-ga151


From 6322e0e75a01d30b914fa87470ad00d8db844f72 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 23 May 2024 12:03:52 +0200
Subject: wifi: mac80211: handle HW restart during ROC

If we have a HW restart in the middle of a ROC period,
then there are two cases:
 - if it's a software ROC, we really don't need to do
   anything, since the ROC work will still be queued
   and will run later, albeit with the interruption
   due to the restart;
 - if it's a hardware ROC, then it may have begun or
   not, if it did begin already we can only remove it
   and tell userspace about that.

In both cases, this fixes the warning that would appear
in ieee80211_start_next_roc() in this case.

In the case of some drivers such as iwlwifi, the part of
restarting is never going to happen since the driver will
cancel the ROC, but flushing the work to ensure nothing
is pending here will also result in no longer being able
to trigger the warning in this case.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240523120352.f1924b5411ea.Ifc02a45a5ce23868dc7e428bad8d0e6996dd10f4@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/ieee80211_i.h |  1 +
 net/mac80211/offchannel.c  | 35 ++++++++++++++++++++++++++++++++++-
 net/mac80211/util.c        |  3 +--
 3 files changed, 36 insertions(+), 3 deletions(-)

diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 76965d64a0fa..3fd7b1adbfab 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -1978,6 +1978,7 @@ void ieee80211_offchannel_stop_vifs(struct ieee80211_local *local);
 void ieee80211_offchannel_return(struct ieee80211_local *local);
 void ieee80211_roc_setup(struct ieee80211_local *local);
 void ieee80211_start_next_roc(struct ieee80211_local *local);
+void ieee80211_reconfig_roc(struct ieee80211_local *local);
 void ieee80211_roc_purge(struct ieee80211_local *local,
 			 struct ieee80211_sub_if_data *sdata);
 int ieee80211_remain_on_channel(struct wiphy *wiphy, struct wireless_dev *wdev,
diff --git a/net/mac80211/offchannel.c b/net/mac80211/offchannel.c
index 65e1e9e971fd..28d03196ef75 100644
--- a/net/mac80211/offchannel.c
+++ b/net/mac80211/offchannel.c
@@ -8,7 +8,7 @@
  * Copyright 2006-2007	Jiri Benc <jbenc@suse.cz>
  * Copyright 2007, Michael Wu <flamingice@sourmilk.net>
  * Copyright 2009	Johannes Berg <johannes@sipsolutions.net>
- * Copyright (C) 2019, 2022-2023 Intel Corporation
+ * Copyright (C) 2019, 2022-2024 Intel Corporation
  */
 #include <linux/export.h>
 #include <net/mac80211.h>
@@ -413,6 +413,39 @@ void ieee80211_start_next_roc(struct ieee80211_local *local)
 	}
 }
 
+void ieee80211_reconfig_roc(struct ieee80211_local *local)
+{
+	struct ieee80211_roc_work *roc, *tmp;
+
+	/*
+	 * In the software implementation can just continue with the
+	 * interruption due to reconfig, roc_work is still queued if
+	 * needed.
+	 */
+	if (!local->ops->remain_on_channel)
+		return;
+
+	/* flush work so nothing from the driver is still pending */
+	wiphy_work_flush(local->hw.wiphy, &local->hw_roc_start);
+	wiphy_work_flush(local->hw.wiphy, &local->hw_roc_done);
+
+	list_for_each_entry_safe(roc, tmp, &local->roc_list, list) {
+		if (!roc->started)
+			break;
+
+		if (!roc->hw_begun) {
+			/* it didn't start in HW yet, so we can restart it */
+			roc->started = false;
+			continue;
+		}
+
+		/* otherwise destroy it and tell userspace */
+		ieee80211_roc_notify_destroy(roc);
+	}
+
+	ieee80211_start_next_roc(local);
+}
+
 static void __ieee80211_roc_work(struct ieee80211_local *local)
 {
 	struct ieee80211_roc_work *roc;
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 43625ca87d5e..927f752a0209 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -2175,8 +2175,7 @@ int ieee80211_reconfig(struct ieee80211_local *local)
 		local->in_reconfig = false;
 		barrier();
 
-		/* Restart deferred ROCs */
-		ieee80211_start_next_roc(local);
+		ieee80211_reconfig_roc(local);
 
 		/* Requeue all works */
 		list_for_each_entry(sdata, &local->interfaces, list)
-- 
cgit v1.2.3-58-ga151


From 8526f8c877baf3f9e678b31fd7d1066b776775cc Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 23 May 2024 12:02:13 +0200
Subject: wifi: nl80211: clean up coalescing rule handling

There's no need to allocate a tiny struct and then
an array again, just allocate the two together and
use __counted_by(). Also unify the freeing.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240523120213.48a40cfb96f9.Ia02bf8f8fefbf533c64c5fa26175848d4a3a7899@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h |  2 +-
 net/wireless/core.c    |  3 ++-
 net/wireless/nl80211.c | 47 ++++++++++++++++-------------------------------
 net/wireless/nl80211.h |  4 ++--
 4 files changed, 21 insertions(+), 35 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index d79180bec7a1..5da9bb0ac6a4 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -3566,8 +3566,8 @@ struct cfg80211_coalesce_rules {
  * @n_rules: number of rules
  */
 struct cfg80211_coalesce {
-	struct cfg80211_coalesce_rules *rules;
 	int n_rules;
+	struct cfg80211_coalesce_rules rules[] __counted_by(n_rules);
 };
 
 /**
diff --git a/net/wireless/core.c b/net/wireless/core.c
index 61f7cd8a8e9c..7c35349b9596 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -1145,7 +1145,8 @@ void wiphy_unregister(struct wiphy *wiphy)
 	flush_work(&rdev->background_cac_abort_wk);
 
 	cfg80211_rdev_free_wowlan(rdev);
-	cfg80211_rdev_free_coalesce(rdev);
+	cfg80211_free_coalesce(rdev->coalesce);
+	rdev->coalesce = NULL;
 }
 EXPORT_SYMBOL(wiphy_unregister);
 
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 6ba988a6f5a2..8ff5f79d446a 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -13897,9 +13897,8 @@ nla_put_failure:
 	return -ENOBUFS;
 }
 
-void cfg80211_rdev_free_coalesce(struct cfg80211_registered_device *rdev)
+void cfg80211_free_coalesce(struct cfg80211_coalesce *coalesce)
 {
-	struct cfg80211_coalesce *coalesce = rdev->coalesce;
 	int i, j;
 	struct cfg80211_coalesce_rules *rule;
 
@@ -13908,13 +13907,13 @@ void cfg80211_rdev_free_coalesce(struct cfg80211_registered_device *rdev)
 
 	for (i = 0; i < coalesce->n_rules; i++) {
 		rule = &coalesce->rules[i];
+		if (!rule)
+			continue;
 		for (j = 0; j < rule->n_patterns; j++)
 			kfree(rule->patterns[j].mask);
 		kfree(rule->patterns);
 	}
-	kfree(coalesce->rules);
 	kfree(coalesce);
-	rdev->coalesce = NULL;
 }
 
 static int nl80211_parse_coalesce_rule(struct cfg80211_registered_device *rdev,
@@ -14012,17 +14011,16 @@ static int nl80211_set_coalesce(struct sk_buff *skb, struct genl_info *info)
 {
 	struct cfg80211_registered_device *rdev = info->user_ptr[0];
 	const struct wiphy_coalesce_support *coalesce = rdev->wiphy.coalesce;
-	struct cfg80211_coalesce new_coalesce = {};
-	struct cfg80211_coalesce *n_coalesce;
-	int err, rem_rule, n_rules = 0, i, j;
+	struct cfg80211_coalesce *new_coalesce;
+	int err, rem_rule, n_rules = 0, i;
 	struct nlattr *rule;
-	struct cfg80211_coalesce_rules *tmp_rule;
 
 	if (!rdev->wiphy.coalesce || !rdev->ops->set_coalesce)
 		return -EOPNOTSUPP;
 
 	if (!info->attrs[NL80211_ATTR_COALESCE_RULE]) {
-		cfg80211_rdev_free_coalesce(rdev);
+		cfg80211_free_coalesce(rdev->coalesce);
+		rdev->coalesce = NULL;
 		rdev_set_coalesce(rdev, NULL);
 		return 0;
 	}
@@ -14033,47 +14031,34 @@ static int nl80211_set_coalesce(struct sk_buff *skb, struct genl_info *info)
 	if (n_rules > coalesce->n_rules)
 		return -EINVAL;
 
-	new_coalesce.rules = kcalloc(n_rules, sizeof(new_coalesce.rules[0]),
-				     GFP_KERNEL);
-	if (!new_coalesce.rules)
+	new_coalesce = kzalloc(struct_size(new_coalesce, rules, n_rules),
+			       GFP_KERNEL);
+	if (!new_coalesce)
 		return -ENOMEM;
 
-	new_coalesce.n_rules = n_rules;
+	new_coalesce->n_rules = n_rules;
 	i = 0;
 
 	nla_for_each_nested(rule, info->attrs[NL80211_ATTR_COALESCE_RULE],
 			    rem_rule) {
 		err = nl80211_parse_coalesce_rule(rdev, rule,
-						  &new_coalesce.rules[i]);
+						  &new_coalesce->rules[i]);
 		if (err)
 			goto error;
 
 		i++;
 	}
 
-	err = rdev_set_coalesce(rdev, &new_coalesce);
+	err = rdev_set_coalesce(rdev, new_coalesce);
 	if (err)
 		goto error;
 
-	n_coalesce = kmemdup(&new_coalesce, sizeof(new_coalesce), GFP_KERNEL);
-	if (!n_coalesce) {
-		err = -ENOMEM;
-		goto error;
-	}
-	cfg80211_rdev_free_coalesce(rdev);
-	rdev->coalesce = n_coalesce;
+	cfg80211_free_coalesce(rdev->coalesce);
+	rdev->coalesce = new_coalesce;
 
 	return 0;
 error:
-	for (i = 0; i < new_coalesce.n_rules; i++) {
-		tmp_rule = &new_coalesce.rules[i];
-		if (!tmp_rule)
-			continue;
-		for (j = 0; j < tmp_rule->n_patterns; j++)
-			kfree(tmp_rule->patterns[j].mask);
-		kfree(tmp_rule->patterns);
-	}
-	kfree(new_coalesce.rules);
+	cfg80211_free_coalesce(new_coalesce);
 
 	return err;
 }
diff --git a/net/wireless/nl80211.h b/net/wireless/nl80211.h
index 6376f3a87f8a..ffaab9a92e5b 100644
--- a/net/wireless/nl80211.h
+++ b/net/wireless/nl80211.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: GPL-2.0 */
 /*
  * Portions of this file
- * Copyright (C) 2018, 2020-2022 Intel Corporation
+ * Copyright (C) 2018, 2020-2024 Intel Corporation
  */
 #ifndef __NET_WIRELESS_NL80211_H
 #define __NET_WIRELESS_NL80211_H
@@ -119,7 +119,7 @@ nl80211_radar_notify(struct cfg80211_registered_device *rdev,
 
 void nl80211_send_ap_stopped(struct wireless_dev *wdev, unsigned int link_id);
 
-void cfg80211_rdev_free_coalesce(struct cfg80211_registered_device *rdev);
+void cfg80211_free_coalesce(struct cfg80211_coalesce *coalesce);
 
 /* peer measurement */
 int nl80211_pmsr_start(struct sk_buff *skb, struct genl_info *info);
-- 
cgit v1.2.3-58-ga151


From 2942dfab630444d46aaa37fb7d629b620abbf6ba Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@linaro.org>
Date: Mon, 27 May 2024 21:26:44 +0200
Subject: net: ethernet: cortina: Restore TSO support

An earlier commit deleted the TSO support in the Cortina Gemini
driver because the driver was confusing gso_size and MTU,
probably because what the Linux kernel calls "gso_size" was
called "MTU" in the datasheet.

Restore the functionality properly reading the gso_size from
the skbuff.

Tested with iperf3, running a server on a different machine
and client on the device with the cortina gemini ethernet:

Connecting to host 192.168.1.2, port 5201
60008000.ethernet-port eth0: segment offloading mss = 05ea len=1c8a
60008000.ethernet-port eth0: segment offloading mss = 05ea len=1c8a
60008000.ethernet-port eth0: segment offloading mss = 05ea len=27da
60008000.ethernet-port eth0: segment offloading mss = 05ea len=0b92
60008000.ethernet-port eth0: segment offloading mss = 05ea len=2bda
(...)

(The hardware MSS 0x05ea here includes the ethernet headers.)

If I disable all segment offloading on the receiving host and
dump packets using tcpdump -xx like this:

ethtool -K enp2s0 gro off gso off tso off
tcpdump -xx -i enp2s0 host 192.168.1.136

I get segmented packages such as this when running iperf3:

23:16:54.024139 IP OpenWrt.lan.59168 > Fecusia.targus-getdata1:
Flags [.], seq 1486:2934, ack 1, win 4198,
options [nop,nop,TS val 3886192908 ecr 3601341877], length 1448
0x0000:  fc34 9701 a0c6 14d6 4da8 3c4f 0800 4500
0x0010:  05dc 16a0 4000 4006 9aa1 c0a8 0188 c0a8
0x0020:  0102 e720 1451 ff25 9822 4c52 29cf 8010
0x0030:  1066 ac8c 0000 0101 080a e7a2 990c d6a8
(...)
0x05c0:  5e49 e109 fe8c 4617 5e18 7a82 7eae d647
0x05d0:  e8ee ae64 dc88 c897 3f8a 07a4 3a33 6b1b
0x05e0:  3501 a30f 2758 cc44 4b4a

Several such packets often follow after each other verifying
the segmentation into 0x05a8 (1448) byte packages also on the
reveiving end. As can be seen, the ethernet frames are
0x05ea (1514) in size.

Performance with iperf3 before this patch: ~15.5 Mbit/s
Performance with iperf3 after this patch: ~175 Mbit/s

This was running a 60 second test (twice) the best measurement
was 179 Mbit/s.

For comparison if I run iperf3 with UDP I get around 1.05 Mbit/s
both before and after this patch.

While this is a gigabit ethernet interface, the CPU is a cheap
D-Link DIR-685 router (based on the ARMv5 Faraday FA526 at
~50 MHz), and the software is not supposed to drive traffic,
as the device has a DSA chip, so this kind of numbers can be
expected.

Fixes: ac631873c9e7 ("net: ethernet: cortina: Drop TSO support")
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/cortina/gemini.c | 23 +++++++++++++++++++----
 1 file changed, 19 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c
index 5f0c9e1771db..7ebd61a3a49b 100644
--- a/drivers/net/ethernet/cortina/gemini.c
+++ b/drivers/net/ethernet/cortina/gemini.c
@@ -79,7 +79,8 @@ MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
 #define GMAC0_IRQ4_8 (GMAC0_MIB_INT_BIT | GMAC0_RX_OVERRUN_INT_BIT)
 
 #define GMAC_OFFLOAD_FEATURES (NETIF_F_SG | NETIF_F_IP_CSUM | \
-			       NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM)
+			       NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM | \
+			       NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6)
 
 /**
  * struct gmac_queue_page - page buffer per-page info
@@ -1148,13 +1149,25 @@ static int gmac_map_tx_bufs(struct net_device *netdev, struct sk_buff *skb,
 	skb_frag_t *skb_frag;
 	dma_addr_t mapping;
 	void *buffer;
+	u16 mss;
 	int ret;
 
-	/* TODO: implement proper TSO using MTU in word3 */
 	word1 = skb->len;
 	word3 = SOF_BIT;
 
-	if (skb->len >= ETH_FRAME_LEN) {
+	mss = skb_shinfo(skb)->gso_size;
+	if (mss) {
+		/* This means we are dealing with TCP and skb->len is the
+		 * sum total of all the segments. The TSO will deal with
+		 * chopping this up for us.
+		 */
+		/* The accelerator needs the full frame size here */
+		mss += skb_tcp_all_headers(skb);
+		netdev_dbg(netdev, "segment offloading mss = %04x len=%04x\n",
+			   mss, skb->len);
+		word1 |= TSS_MTU_ENABLE_BIT;
+		word3 |= mss;
+	} else if (skb->len >= ETH_FRAME_LEN) {
 		/* Hardware offloaded checksumming isn't working on frames
 		 * bigger than 1514 bytes. A hypothesis about this is that the
 		 * checksum buffer is only 1518 bytes, so when the frames get
@@ -1169,7 +1182,9 @@ static int gmac_map_tx_bufs(struct net_device *netdev, struct sk_buff *skb,
 				return ret;
 		}
 		word1 |= TSS_BYPASS_BIT;
-	} else if (skb->ip_summed == CHECKSUM_PARTIAL) {
+	}
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
 		int tcp = 0;
 
 		/* We do not switch off the checksumming on non TCP/UDP
-- 
cgit v1.2.3-58-ga151


From c1d96671088f03e7a685ce1fd3d745af85193141 Mon Sep 17 00:00:00 2001
From: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Date: Tue, 28 May 2024 11:50:07 +0530
Subject: dt-bindings: net: xilinx_gmii2rgmii: Add clock support

Add "clocks" bindings for the input clock.

Signed-off-by: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/devicetree/bindings/net/xlnx,gmii-to-rgmii.yaml | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/xlnx,gmii-to-rgmii.yaml b/Documentation/devicetree/bindings/net/xlnx,gmii-to-rgmii.yaml
index 0f781dac6717..eb803ddd13e0 100644
--- a/Documentation/devicetree/bindings/net/xlnx,gmii-to-rgmii.yaml
+++ b/Documentation/devicetree/bindings/net/xlnx,gmii-to-rgmii.yaml
@@ -31,6 +31,10 @@ properties:
   phy-handle:
     $ref: ethernet-controller.yaml#/properties/phy-handle
 
+  clocks:
+    items:
+      - description: 200/375 MHz free-running clock is used as input clock.
+
 required:
   - compatible
   - reg
@@ -51,5 +55,6 @@ examples:
             compatible = "xlnx,gmii-to-rgmii-1.0";
             reg = <8>;
             phy-handle = <&phy>;
+            clocks = <&dummy>;
         };
     };
-- 
cgit v1.2.3-58-ga151


From daab0ac53e774f14ea78a1849358dbefb0d3efd2 Mon Sep 17 00:00:00 2001
From: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Date: Tue, 28 May 2024 11:50:08 +0530
Subject: net: phy: xilinx-gmii2rgmii: Adopt clock support

Add clock support to the gmii_to_rgmii IP.
Make clk optional to keep DTB backward compatibility.

Signed-off-by: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/xilinx_gmii2rgmii.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/phy/xilinx_gmii2rgmii.c b/drivers/net/phy/xilinx_gmii2rgmii.c
index 7b1bc5fcef9b..7c51daecf18e 100644
--- a/drivers/net/phy/xilinx_gmii2rgmii.c
+++ b/drivers/net/phy/xilinx_gmii2rgmii.c
@@ -15,6 +15,7 @@
 #include <linux/mii.h>
 #include <linux/mdio.h>
 #include <linux/phy.h>
+#include <linux/clk.h>
 #include <linux/of_mdio.h>
 
 #define XILINX_GMII2RGMII_REG		0x10
@@ -85,11 +86,17 @@ static int xgmiitorgmii_probe(struct mdio_device *mdiodev)
 	struct device *dev = &mdiodev->dev;
 	struct device_node *np = dev->of_node, *phy_node;
 	struct gmii2rgmii *priv;
+	struct clk *clkin;
 
 	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
 	if (!priv)
 		return -ENOMEM;
 
+	clkin = devm_clk_get_optional_enabled(dev, NULL);
+	if (IS_ERR(clkin))
+		return dev_err_probe(dev, PTR_ERR(clkin),
+					"Failed to get and enable clock from Device Tree\n");
+
 	phy_node = of_parse_phandle(np, "phy-handle", 0);
 	if (!phy_node) {
 		dev_err(dev, "Couldn't parse phy-handle\n");
-- 
cgit v1.2.3-58-ga151


From 4665415975b0827e9646cab91c61d02a6b364d59 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Fri, 17 May 2024 10:30:33 +0800
Subject: bpf: Add bits iterator

Add three new kfuncs for the bits iterator:
- bpf_iter_bits_new
  Initialize a new bits iterator for a given memory area. Due to the
  limitation of bpf memalloc, the max number of words (8-byte units) that
  can be iterated over is limited to (4096 / 8).
- bpf_iter_bits_next
  Get the next bit in a bpf_iter_bits
- bpf_iter_bits_destroy
  Destroy a bpf_iter_bits

The bits iterator facilitates the iteration of the bits of a memory area,
such as cpumask. It can be used in any context and on any address.

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240517023034.48138-2-laoar.shao@gmail.com
---
 kernel/bpf/helpers.c | 119 +++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 119 insertions(+)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 2a69a9a36c0f..6f1abcb4b084 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2744,6 +2744,122 @@ __bpf_kfunc void bpf_preempt_enable(void)
 	preempt_enable();
 }
 
+struct bpf_iter_bits {
+	__u64 __opaque[2];
+} __aligned(8);
+
+struct bpf_iter_bits_kern {
+	union {
+		unsigned long *bits;
+		unsigned long bits_copy;
+	};
+	u32 nr_bits;
+	int bit;
+} __aligned(8);
+
+/**
+ * bpf_iter_bits_new() - Initialize a new bits iterator for a given memory area
+ * @it: The new bpf_iter_bits to be created
+ * @unsafe_ptr__ign: A pointer pointing to a memory area to be iterated over
+ * @nr_words: The size of the specified memory area, measured in 8-byte units.
+ * Due to the limitation of memalloc, it can't be greater than 512.
+ *
+ * This function initializes a new bpf_iter_bits structure for iterating over
+ * a memory area which is specified by the @unsafe_ptr__ign and @nr_words. It
+ * copies the data of the memory area to the newly created bpf_iter_bits @it for
+ * subsequent iteration operations.
+ *
+ * On success, 0 is returned. On failure, ERR is returned.
+ */
+__bpf_kfunc int
+bpf_iter_bits_new(struct bpf_iter_bits *it, const u64 *unsafe_ptr__ign, u32 nr_words)
+{
+	struct bpf_iter_bits_kern *kit = (void *)it;
+	u32 nr_bytes = nr_words * sizeof(u64);
+	u32 nr_bits = BYTES_TO_BITS(nr_bytes);
+	int err;
+
+	BUILD_BUG_ON(sizeof(struct bpf_iter_bits_kern) != sizeof(struct bpf_iter_bits));
+	BUILD_BUG_ON(__alignof__(struct bpf_iter_bits_kern) !=
+		     __alignof__(struct bpf_iter_bits));
+
+	kit->nr_bits = 0;
+	kit->bits_copy = 0;
+	kit->bit = -1;
+
+	if (!unsafe_ptr__ign || !nr_words)
+		return -EINVAL;
+
+	/* Optimization for u64 mask */
+	if (nr_bits == 64) {
+		err = bpf_probe_read_kernel_common(&kit->bits_copy, nr_bytes, unsafe_ptr__ign);
+		if (err)
+			return -EFAULT;
+
+		kit->nr_bits = nr_bits;
+		return 0;
+	}
+
+	/* Fallback to memalloc */
+	kit->bits = bpf_mem_alloc(&bpf_global_ma, nr_bytes);
+	if (!kit->bits)
+		return -ENOMEM;
+
+	err = bpf_probe_read_kernel_common(kit->bits, nr_bytes, unsafe_ptr__ign);
+	if (err) {
+		bpf_mem_free(&bpf_global_ma, kit->bits);
+		return err;
+	}
+
+	kit->nr_bits = nr_bits;
+	return 0;
+}
+
+/**
+ * bpf_iter_bits_next() - Get the next bit in a bpf_iter_bits
+ * @it: The bpf_iter_bits to be checked
+ *
+ * This function returns a pointer to a number representing the value of the
+ * next bit in the bits.
+ *
+ * If there are no further bits available, it returns NULL.
+ */
+__bpf_kfunc int *bpf_iter_bits_next(struct bpf_iter_bits *it)
+{
+	struct bpf_iter_bits_kern *kit = (void *)it;
+	u32 nr_bits = kit->nr_bits;
+	const unsigned long *bits;
+	int bit;
+
+	if (nr_bits == 0)
+		return NULL;
+
+	bits = nr_bits == 64 ? &kit->bits_copy : kit->bits;
+	bit = find_next_bit(bits, nr_bits, kit->bit + 1);
+	if (bit >= nr_bits) {
+		kit->nr_bits = 0;
+		return NULL;
+	}
+
+	kit->bit = bit;
+	return &kit->bit;
+}
+
+/**
+ * bpf_iter_bits_destroy() - Destroy a bpf_iter_bits
+ * @it: The bpf_iter_bits to be destroyed
+ *
+ * Destroy the resource associated with the bpf_iter_bits.
+ */
+__bpf_kfunc void bpf_iter_bits_destroy(struct bpf_iter_bits *it)
+{
+	struct bpf_iter_bits_kern *kit = (void *)it;
+
+	if (kit->nr_bits <= 64)
+		return;
+	bpf_mem_free(&bpf_global_ma, kit->bits);
+}
+
 __bpf_kfunc_end_defs();
 
 BTF_KFUNCS_START(generic_btf_ids)
@@ -2826,6 +2942,9 @@ BTF_ID_FLAGS(func, bpf_wq_set_callback_impl)
 BTF_ID_FLAGS(func, bpf_wq_start)
 BTF_ID_FLAGS(func, bpf_preempt_disable)
 BTF_ID_FLAGS(func, bpf_preempt_enable)
+BTF_ID_FLAGS(func, bpf_iter_bits_new, KF_ITER_NEW)
+BTF_ID_FLAGS(func, bpf_iter_bits_next, KF_ITER_NEXT | KF_RET_NULL)
+BTF_ID_FLAGS(func, bpf_iter_bits_destroy, KF_ITER_DESTROY)
 BTF_KFUNCS_END(common_btf_ids)
 
 static const struct btf_kfunc_id_set common_kfunc_set = {
-- 
cgit v1.2.3-58-ga151


From 6ba7acdb93b4ecb554d5838fca3f5f0fcf9fff14 Mon Sep 17 00:00:00 2001
From: Yafang Shao <laoar.shao@gmail.com>
Date: Fri, 17 May 2024 10:30:34 +0800
Subject: selftests/bpf: Add selftest for bits iter

Add test cases for the bits iter:

- Positive cases
  - Bit mask representing a single word (8-byte unit)
  - Bit mask representing data spanning more than one word
  - The index of the set bit

- Nagative cases
  - bpf_iter_bits_destroy() is required after calling
    bpf_iter_bits_new()
  - bpf_iter_bits_destroy() can only destroy an initialized iter
  - bpf_iter_bits_next() must use an initialized iter
  - Bit mask representing zero words
  - Bit mask representing fewer words than expected
  - Case for ENOMEM
  - Case for NULL pointer

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240517023034.48138-3-laoar.shao@gmail.com
---
 tools/testing/selftests/bpf/prog_tests/verifier.c  |   2 +
 .../selftests/bpf/progs/verifier_bits_iter.c       | 153 +++++++++++++++++++++
 2 files changed, 155 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/progs/verifier_bits_iter.c

diff --git a/tools/testing/selftests/bpf/prog_tests/verifier.c b/tools/testing/selftests/bpf/prog_tests/verifier.c
index c60db8beeb73..8743340b5bf6 100644
--- a/tools/testing/selftests/bpf/prog_tests/verifier.c
+++ b/tools/testing/selftests/bpf/prog_tests/verifier.c
@@ -85,6 +85,7 @@
 #include "verifier_xadd.skel.h"
 #include "verifier_xdp.skel.h"
 #include "verifier_xdp_direct_packet_access.skel.h"
+#include "verifier_bits_iter.skel.h"
 
 #define MAX_ENTRIES 11
 
@@ -200,6 +201,7 @@ void test_verifier_var_off(void)              { RUN(verifier_var_off); }
 void test_verifier_xadd(void)                 { RUN(verifier_xadd); }
 void test_verifier_xdp(void)                  { RUN(verifier_xdp); }
 void test_verifier_xdp_direct_packet_access(void) { RUN(verifier_xdp_direct_packet_access); }
+void test_verifier_bits_iter(void) { RUN(verifier_bits_iter); }
 
 static int init_test_val_map(struct bpf_object *obj, char *map_name)
 {
diff --git a/tools/testing/selftests/bpf/progs/verifier_bits_iter.c b/tools/testing/selftests/bpf/progs/verifier_bits_iter.c
new file mode 100644
index 000000000000..716113c2bce2
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/verifier_bits_iter.c
@@ -0,0 +1,153 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2024 Yafang Shao <laoar.shao@gmail.com> */
+
+#include "vmlinux.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
+
+#include "bpf_misc.h"
+#include "task_kfunc_common.h"
+
+char _license[] SEC("license") = "GPL";
+
+int bpf_iter_bits_new(struct bpf_iter_bits *it, const u64 *unsafe_ptr__ign,
+		      u32 nr_bits) __ksym __weak;
+int *bpf_iter_bits_next(struct bpf_iter_bits *it) __ksym __weak;
+void bpf_iter_bits_destroy(struct bpf_iter_bits *it) __ksym __weak;
+
+SEC("iter.s/cgroup")
+__description("bits iter without destroy")
+__failure __msg("Unreleased reference")
+int BPF_PROG(no_destroy, struct bpf_iter_meta *meta, struct cgroup *cgrp)
+{
+	struct bpf_iter_bits it;
+	u64 data = 1;
+
+	bpf_iter_bits_new(&it, &data, 1);
+	bpf_iter_bits_next(&it);
+	return 0;
+}
+
+SEC("iter/cgroup")
+__description("uninitialized iter in ->next()")
+__failure __msg("expected an initialized iter_bits as arg #1")
+int BPF_PROG(next_uninit, struct bpf_iter_meta *meta, struct cgroup *cgrp)
+{
+	struct bpf_iter_bits *it = NULL;
+
+	bpf_iter_bits_next(it);
+	return 0;
+}
+
+SEC("iter/cgroup")
+__description("uninitialized iter in ->destroy()")
+__failure __msg("expected an initialized iter_bits as arg #1")
+int BPF_PROG(destroy_uninit, struct bpf_iter_meta *meta, struct cgroup *cgrp)
+{
+	struct bpf_iter_bits it = {};
+
+	bpf_iter_bits_destroy(&it);
+	return 0;
+}
+
+SEC("syscall")
+__description("null pointer")
+__success __retval(0)
+int null_pointer(void)
+{
+	int nr = 0;
+	int *bit;
+
+	bpf_for_each(bits, bit, NULL, 1)
+		nr++;
+	return nr;
+}
+
+SEC("syscall")
+__description("bits copy")
+__success __retval(10)
+int bits_copy(void)
+{
+	u64 data = 0xf7310UL; /* 4 + 3 + 2 + 1 + 0*/
+	int nr = 0;
+	int *bit;
+
+	bpf_for_each(bits, bit, &data, 1)
+		nr++;
+	return nr;
+}
+
+SEC("syscall")
+__description("bits memalloc")
+__success __retval(64)
+int bits_memalloc(void)
+{
+	u64 data[2];
+	int nr = 0;
+	int *bit;
+
+	__builtin_memset(&data, 0xf0, sizeof(data)); /* 4 * 16 */
+	bpf_for_each(bits, bit, &data[0], sizeof(data) / sizeof(u64))
+		nr++;
+	return nr;
+}
+
+SEC("syscall")
+__description("bit index")
+__success __retval(8)
+int bit_index(void)
+{
+	u64 data = 0x100;
+	int bit_idx = 0;
+	int *bit;
+
+	bpf_for_each(bits, bit, &data, 1) {
+		if (*bit == 0)
+			continue;
+		bit_idx = *bit;
+	}
+	return bit_idx;
+}
+
+SEC("syscall")
+__description("bits nomem")
+__success __retval(0)
+int bits_nomem(void)
+{
+	u64 data[4];
+	int nr = 0;
+	int *bit;
+
+	__builtin_memset(&data, 0xff, sizeof(data));
+	bpf_for_each(bits, bit, &data[0], 513) /* Be greater than 512 */
+		nr++;
+	return nr;
+}
+
+SEC("syscall")
+__description("fewer words")
+__success __retval(1)
+int fewer_words(void)
+{
+	u64 data[2] = {0x1, 0xff};
+	int nr = 0;
+	int *bit;
+
+	bpf_for_each(bits, bit, &data[0], 1)
+		nr++;
+	return nr;
+}
+
+SEC("syscall")
+__description("zero words")
+__success __retval(0)
+int zero_words(void)
+{
+	u64 data[2] = {0x1, 0xff};
+	int nr = 0;
+	int *bit;
+
+	bpf_for_each(bits, bit, &data[0], 0)
+		nr++;
+	return nr;
+}
-- 
cgit v1.2.3-58-ga151


From 126913479e88e7e0114efe380d2437cc70d9dc05 Mon Sep 17 00:00:00 2001
From: Alexander Sverdlin <alexander.sverdlin@siemens.com>
Date: Tue, 28 May 2024 09:31:13 +0200
Subject: net: dsa: lan9303: imply SMSC_PHY

Both LAN9303 and LAN9354 have internal PHYs on both external ports.
Therefore a configuration without SMSC PHY support is non-practical at
least and leads to:

LAN9303_MDIO 8000f00.mdio:00: Found LAN9303 rev. 1
mdio_bus 8000f00.mdio:00: deferred probe pending: (reason unknown)

Signed-off-by: Alexander Sverdlin <alexander.sverdlin@siemens.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20240528073147.3604083-1-alexander.sverdlin@siemens.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig
index 3092b391031a..8508b5145bc1 100644
--- a/drivers/net/dsa/Kconfig
+++ b/drivers/net/dsa/Kconfig
@@ -102,6 +102,7 @@ config NET_DSA_SMSC_LAN9303
 	tristate
 	select NET_DSA_TAG_LAN9303
 	select REGMAP
+	imply SMSC_PHY
 	help
 	  This enables support for the Microchip LAN9303/LAN9354 3 port ethernet
 	  switch chips.
-- 
cgit v1.2.3-58-ga151


From 544a74c32bcc2eff5f93b9f638f687170e7274e9 Mon Sep 17 00:00:00 2001
From: Jonas Karlman <jonas@kwiboo.se>
Date: Tue, 28 May 2024 09:37:48 +0000
Subject: dt-bindings: net: rockchip-dwmac: Fix rockchip,rk3308-gmac compatible

Schema validation using rockchip,rk3308-gmac compatible fails with:

  ethernet@ff4e0000: compatible: ['rockchip,rk3308-gmac'] does not contain items matching the given schema
        from schema $id: http://devicetree.org/schemas/net/rockchip-dwmac.yaml#
  ethernet@ff4e0000: Unevaluated properties are not allowed ('interrupt-names', 'interrupts', 'phy-mode',
                     'reg', 'reset-names', 'resets', 'snps,reset-active-low', 'snps,reset-delays-us',
                     'snps,reset-gpio' were unexpected)
        from schema $id: http://devicetree.org/schemas/net/rockchip-dwmac.yaml#

Add rockchip,rk3308-gmac to snps,dwmac.yaml to fix DT schema validation.

Fixes: 2cc8c910f515 ("dt-bindings: net: rockchip-dwmac: add rk3308 gmac compatible")
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20240528093751.3690231-1-jonas@kwiboo.se
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/snps,dwmac.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
index 21cc27e75f50..3bab4e1f3fbf 100644
--- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
@@ -76,6 +76,7 @@ properties:
         - rockchip,rk3128-gmac
         - rockchip,rk3228-gmac
         - rockchip,rk3288-gmac
+        - rockchip,rk3308-gmac
         - rockchip,rk3328-gmac
         - rockchip,rk3366-gmac
         - rockchip,rk3368-gmac
-- 
cgit v1.2.3-58-ga151


From c3390677f6258748a91bf37b9bb21eab89f63b42 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Tue, 28 May 2024 01:42:24 -0700
Subject: netconsole: Do not shutdown dynamic configuration if cmdline is
 invalid

If a user provides an invalid netconsole configuration during boot time
(e.g., specifying an invalid ethX interface), netconsole will be
entirely disabled. Consequently, the user won't be able to create new
entries in /sys/kernel/config/netconsole/ as that directory does not
exist.

Apart from misconfiguration, another issue arises when ethX is loaded as
a module and the netconsole= line in the command line points to ethX,
resulting in an obvious failure. This renders netconsole unusable, as
/sys/kernel/config/netconsole/ will never appear. This is more annoying
since users reconfigure (or just toggle) the configuratin later (see
commit 5fbd6cdbe304b ("netconsole: Attach cmdline target to dynamic
target"))

Create /sys/kernel/config/netconsole/ even if the command line arguments
are invalid, so, users can create dynamic entries in netconsole.

Reported-by: Aijay Adams <aijay@meta.com>
Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240528084225.3215853-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/netconsole.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index d7070dd4fe73..ab8a0623b1a1 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -1262,6 +1262,8 @@ static int __init init_netconsole(void)
 		while ((target_config = strsep(&input, ";"))) {
 			nt = alloc_param_target(target_config, count);
 			if (IS_ERR(nt)) {
+				if (IS_ENABLED(CONFIG_NETCONSOLE_DYNAMIC))
+					continue;
 				err = PTR_ERR(nt);
 				goto fail;
 			}
-- 
cgit v1.2.3-58-ga151


From 5e514f1cba090e1c8fff03e92a175eccfe46305f Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 28 May 2024 12:52:50 +0000
Subject: tcp: add tcp_done_with_error() helper

tcp_reset() ends with a sequence that is carefuly ordered.

We need to fix [e]poll bugs in the following patches,
it makes sense to use a common helper.

Suggested-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Link: https://lore.kernel.org/r/20240528125253.1966136-2-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/tcp.h    |  1 +
 net/ipv4/tcp.c       |  2 +-
 net/ipv4/tcp_input.c | 32 +++++++++++++++++++++-----------
 3 files changed, 23 insertions(+), 12 deletions(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 060e95b331a2..32815a40dea1 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -677,6 +677,7 @@ void tcp_skb_collapse_tstamp(struct sk_buff *skb,
 /* tcp_input.c */
 void tcp_rearm_rto(struct sock *sk);
 void tcp_synack_rtt_meas(struct sock *sk, struct request_sock *req);
+void tcp_done_with_error(struct sock *sk, int err);
 void tcp_reset(struct sock *sk, struct sk_buff *skb);
 void tcp_fin(struct sock *sk);
 void tcp_check_space(struct sock *sk);
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 681b54e1f3a6..2a8f8d8676ff 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -598,7 +598,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
 		 */
 		mask |= EPOLLOUT | EPOLLWRNORM;
 	}
-	/* This barrier is coupled with smp_wmb() in tcp_reset() */
+	/* This barrier is coupled with smp_wmb() in tcp_done_with_error() */
 	smp_rmb();
 	if (READ_ONCE(sk->sk_err) ||
 	    !skb_queue_empty_lockless(&sk->sk_error_queue))
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 9c04a9c8be9d..5aadf64e554d 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4436,9 +4436,26 @@ static enum skb_drop_reason tcp_sequence(const struct tcp_sock *tp,
 	return SKB_NOT_DROPPED_YET;
 }
 
+
+void tcp_done_with_error(struct sock *sk, int err)
+{
+	/* This barrier is coupled with smp_rmb() in tcp_poll() */
+	WRITE_ONCE(sk->sk_err, err);
+	smp_wmb();
+
+	tcp_write_queue_purge(sk);
+	tcp_done(sk);
+
+	if (!sock_flag(sk, SOCK_DEAD))
+		sk_error_report(sk);
+}
+EXPORT_SYMBOL(tcp_done_with_error);
+
 /* When we get a reset we do this. */
 void tcp_reset(struct sock *sk, struct sk_buff *skb)
 {
+	int err;
+
 	trace_tcp_receive_reset(sk);
 
 	/* mptcp can't tell us to ignore reset pkts,
@@ -4450,24 +4467,17 @@ void tcp_reset(struct sock *sk, struct sk_buff *skb)
 	/* We want the right error as BSD sees it (and indeed as we do). */
 	switch (sk->sk_state) {
 	case TCP_SYN_SENT:
-		WRITE_ONCE(sk->sk_err, ECONNREFUSED);
+		err = ECONNREFUSED;
 		break;
 	case TCP_CLOSE_WAIT:
-		WRITE_ONCE(sk->sk_err, EPIPE);
+		err = EPIPE;
 		break;
 	case TCP_CLOSE:
 		return;
 	default:
-		WRITE_ONCE(sk->sk_err, ECONNRESET);
+		err = ECONNRESET;
 	}
-	/* This barrier is coupled with smp_rmb() in tcp_poll() */
-	smp_wmb();
-
-	tcp_write_queue_purge(sk);
-	tcp_done(sk);
-
-	if (!sock_flag(sk, SOCK_DEAD))
-		sk_error_report(sk);
+	tcp_done_with_error(sk, err);
 }
 
 /*
-- 
cgit v1.2.3-58-ga151


From 853c3bd7b7917670224c9fe5245bd045cac411dd Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 28 May 2024 12:52:51 +0000
Subject: tcp: fix race in tcp_write_err()

I noticed flakes in a packetdrill test, expecting an epoll_wait()
to return EPOLLERR | EPOLLHUP on a failed connect() attempt,
after multiple SYN retransmits. It sometimes return EPOLLERR only.

The issue is that tcp_write_err():
 1) writes an error in sk->sk_err,
 2) calls sk_error_report(),
 3) then calls tcp_done().

tcp_done() is writing SHUTDOWN_MASK into sk->sk_shutdown,
among other things.

Problem is that the awaken user thread (from 2) sk_error_report())
might call tcp_poll() before tcp_done() has written sk->sk_shutdown.

tcp_poll() only sees a non zero sk->sk_err and returns EPOLLERR.

This patch fixes the issue by making sure to call sk_error_report()
after tcp_done().

tcp_write_err() also lacks an smp_wmb().

We can reuse tcp_done_with_error() to factor out the details,
as Neal suggested.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Link: https://lore.kernel.org/r/20240528125253.1966136-3-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/tcp_timer.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 83fe7f62f7f1..3e8604ae7d06 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -74,11 +74,7 @@ u32 tcp_clamp_probe0_to_user_timeout(const struct sock *sk, u32 when)
 
 static void tcp_write_err(struct sock *sk)
 {
-	WRITE_ONCE(sk->sk_err, READ_ONCE(sk->sk_err_soft) ? : ETIMEDOUT);
-	sk_error_report(sk);
-
-	tcp_write_queue_purge(sk);
-	tcp_done(sk);
+	tcp_done_with_error(sk, READ_ONCE(sk->sk_err_soft) ? : ETIMEDOUT);
 	__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONTIMEOUT);
 }
 
-- 
cgit v1.2.3-58-ga151


From 5ce4645c23cf5f048eb8e9ce49e514bababdee85 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 28 May 2024 12:52:52 +0000
Subject: tcp: fix races in tcp_abort()

tcp_abort() has the same issue than the one fixed in the prior patch
in tcp_write_err().

In order to get consistent results from tcp_poll(), we must call
sk_error_report() after tcp_done().

We can use tcp_done_with_error() to centralize this logic.

Fixes: c1e64e298b8c ("net: diag: Support destroying TCP sockets.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Link: https://lore.kernel.org/r/20240528125253.1966136-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/tcp.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 2a8f8d8676ff..5fa68e7f6ddb 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -4576,14 +4576,10 @@ int tcp_abort(struct sock *sk, int err)
 	bh_lock_sock(sk);
 
 	if (!sock_flag(sk, SOCK_DEAD)) {
-		WRITE_ONCE(sk->sk_err, err);
-		/* This barrier is coupled with smp_rmb() in tcp_poll() */
-		smp_wmb();
-		sk_error_report(sk);
 		if (tcp_need_reset(sk->sk_state))
 			tcp_send_active_reset(sk, GFP_ATOMIC,
 					      SK_RST_REASON_NOT_SPECIFIED);
-		tcp_done(sk);
+		tcp_done_with_error(sk, err);
 	}
 
 	bh_unlock_sock(sk);
-- 
cgit v1.2.3-58-ga151


From fde6f897f2a184546bf5516ac736523ef24dc6a7 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 28 May 2024 12:52:53 +0000
Subject: tcp: fix races in tcp_v[46]_err()

These functions have races when they:

1) Write sk->sk_err
2) call sk_error_report(sk)
3) call tcp_done(sk)

As described in prior patches in this series:

An smp_wmb() is missing.
We should call tcp_done() before sk_error_report(sk)
to have consistent tcp_poll() results on SMP hosts.

Use tcp_done_with_error() where we centralized the
correct sequence.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Link: https://lore.kernel.org/r/20240528125253.1966136-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/tcp_ipv4.c | 11 +++--------
 net/ipv6/tcp_ipv6.c | 10 +++-------
 2 files changed, 6 insertions(+), 15 deletions(-)

diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 8f70b8d1d1e5..041c7eda9abe 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -611,15 +611,10 @@ int tcp_v4_err(struct sk_buff *skb, u32 info)
 
 		ip_icmp_error(sk, skb, err, th->dest, info, (u8 *)th);
 
-		if (!sock_owned_by_user(sk)) {
-			WRITE_ONCE(sk->sk_err, err);
-
-			sk_error_report(sk);
-
-			tcp_done(sk);
-		} else {
+		if (!sock_owned_by_user(sk))
+			tcp_done_with_error(sk, err);
+		else
 			WRITE_ONCE(sk->sk_err_soft, err);
-		}
 		goto out;
 	}
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 750aa681779c..1ac7502e1bf5 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -490,14 +490,10 @@ static int tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 
 		ipv6_icmp_error(sk, skb, err, th->dest, ntohl(info), (u8 *)th);
 
-		if (!sock_owned_by_user(sk)) {
-			WRITE_ONCE(sk->sk_err, err);
-			sk_error_report(sk);		/* Wake people up to see the error (see connect in sock.c) */
-
-			tcp_done(sk);
-		} else {
+		if (!sock_owned_by_user(sk))
+			tcp_done_with_error(sk, err);
+		else
 			WRITE_ONCE(sk->sk_err_soft, err);
-		}
 		goto out;
 	case TCP_LISTEN:
 		break;
-- 
cgit v1.2.3-58-ga151


From 78269025e192ec8122ddd87a1ec2805598d8a1ab Mon Sep 17 00:00:00 2001
From: Alexander Sverdlin <alexander.sverdlin@siemens.com>
Date: Tue, 28 May 2024 09:59:49 +0200
Subject: net: ethernet: ti: am65-cpsw-nuss: rename phy_node -> port_np

Rename phy_node to port_np to better reflect what it actually is,
because the new phylink API takes netdev node (or DSA port node),
and resolves the phandle internally.

Signed-off-by: Alexander Sverdlin <alexander.sverdlin@siemens.com>
Link: https://lore.kernel.org/r/20240528075954.3608118-2-alexander.sverdlin@siemens.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 6 +++---
 drivers/net/ethernet/ti/am65-cpsw-nuss.h | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index 4e50b3792888..eaadf8f09c40 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -896,7 +896,7 @@ static int am65_cpsw_nuss_ndo_slave_open(struct net_device *ndev)
 	/* mac_sl should be configured via phy-link interface */
 	am65_cpsw_sl_ctl_reset(port);
 
-	ret = phylink_of_phy_connect(port->slave.phylink, port->slave.phy_node, 0);
+	ret = phylink_of_phy_connect(port->slave.phylink, port->slave.port_np, 0);
 	if (ret)
 		goto error_cleanup;
 
@@ -2611,7 +2611,7 @@ static int am65_cpsw_nuss_init_slave_ports(struct am65_cpsw_common *common)
 				of_property_read_bool(port_np, "ti,mac-only");
 
 		/* get phy/link info */
-		port->slave.phy_node = port_np;
+		port->slave.port_np = port_np;
 		ret = of_get_phy_mode(port_np, &port->slave.phy_if);
 		if (ret) {
 			dev_err(dev, "%pOF read phy-mode err %d\n",
@@ -2760,7 +2760,7 @@ am65_cpsw_nuss_init_port_ndev(struct am65_cpsw_common *common, u32 port_idx)
 	}
 
 	phylink = phylink_create(&port->slave.phylink_config,
-				 of_node_to_fwnode(port->slave.phy_node),
+				 of_node_to_fwnode(port->slave.port_np),
 				 port->slave.phy_if,
 				 &am65_cpsw_phylink_mac_ops);
 	if (IS_ERR(phylink))
diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.h b/drivers/net/ethernet/ti/am65-cpsw-nuss.h
index d8ce88dc9c89..e2ce2be320bd 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.h
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.h
@@ -30,7 +30,7 @@ struct am65_cpts;
 struct am65_cpsw_slave_data {
 	bool				mac_only;
 	struct cpsw_sl			*mac_sl;
-	struct device_node		*phy_node;
+	struct device_node		*port_np;
 	phy_interface_t			phy_if;
 	struct phy			*ifphy;
 	struct phy			*serdes_phy;
-- 
cgit v1.2.3-58-ga151


From 29c71bf2a05a11f0d139590d37d92547477d5eb2 Mon Sep 17 00:00:00 2001
From: Alexander Sverdlin <alexander.sverdlin@siemens.com>
Date: Tue, 28 May 2024 09:59:50 +0200
Subject: net: ethernet: ti: am65-cpsw-nuss: populate netdev of_node

So that of_find_net_device_by_node() can find cpsw-nuss ports and other DSA
switches can be stacked downstream.

Signed-off-by: Alexander Sverdlin <alexander.sverdlin@siemens.com>
Link: https://lore.kernel.org/r/20240528075954.3608118-3-alexander.sverdlin@siemens.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/am65-cpsw-nuss.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index eaadf8f09c40..e6f87ac394fe 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -2703,6 +2703,7 @@ am65_cpsw_nuss_init_port_ndev(struct am65_cpsw_common *common, u32 port_idx)
 	mutex_init(&ndev_priv->mm_lock);
 	port->qos.link_speed = SPEED_UNKNOWN;
 	SET_NETDEV_DEV(port->ndev, dev);
+	port->ndev->dev.of_node = port->slave.port_np;
 
 	eth_hw_addr_set(port->ndev, port->slave.mac_addr);
 
-- 
cgit v1.2.3-58-ga151


From c697f515b6390b17e4a54a6f93aedf27133929c5 Mon Sep 17 00:00:00 2001
From: Donald Hunter <donald.hunter@gmail.com>
Date: Tue, 28 May 2024 15:06:49 +0100
Subject: doc: netlink: Fix generated .rst for multi-line docs

Fix the newline replacement in ynl-gen-rst.py to put spaces between
concatenated lines. This fixes the broken doc string formatting.

See the dpll docs for an example of broken concatenation:

https://docs.kernel.org/6.9/networking/netlink_spec/dpll.html#lock-status

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://lore.kernel.org/r/20240528140652.9445-2-donald.hunter@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/ynl-gen-rst.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/net/ynl/ynl-gen-rst.py b/tools/net/ynl/ynl-gen-rst.py
index 657e881d2ea4..5c7465d6befa 100755
--- a/tools/net/ynl/ynl-gen-rst.py
+++ b/tools/net/ynl/ynl-gen-rst.py
@@ -49,7 +49,7 @@ def inline(text: str) -> str:
 def sanitize(text: str) -> str:
     """Remove newlines and multiple spaces"""
     # This is useful for some fields that are spread across multiple lines
-    return str(text).replace("\n", "").strip()
+    return str(text).replace("\n", " ").strip()
 
 
 def rst_fields(key: str, value: str, level: int = 0) -> str:
-- 
cgit v1.2.3-58-ga151


From ebf9004136c76b7b62fe628a4bc88b3e894b4b95 Mon Sep 17 00:00:00 2001
From: Donald Hunter <donald.hunter@gmail.com>
Date: Tue, 28 May 2024 15:06:50 +0100
Subject: doc: netlink: Don't 'sanitize' op docstrings in generated .rst

The doc strings for do/dump ops are emitted as toplevel .rst constructs
so they can be multi-line. Pass multi-line text straight through to the
.rst to retain any simple formatting from the .yaml

This fixes e.g. list formatting for the pin-get docs in dpll.yaml:

https://docs.kernel.org/6.9/networking/netlink_spec/dpll.html#pin-get

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://lore.kernel.org/r/20240528140652.9445-3-donald.hunter@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/dpll.yaml | 1 +
 tools/net/ynl/ynl-gen-rst.py          | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/Documentation/netlink/specs/dpll.yaml b/Documentation/netlink/specs/dpll.yaml
index 95b0eb1486bf..94132d30e0e0 100644
--- a/Documentation/netlink/specs/dpll.yaml
+++ b/Documentation/netlink/specs/dpll.yaml
@@ -479,6 +479,7 @@ operations:
       name: pin-get
       doc: |
         Get list of pins and its attributes.
+
         - dump request without any attributes given - list all the pins in the
           system
         - dump request with target dpll - list all the pins registered with
diff --git a/tools/net/ynl/ynl-gen-rst.py b/tools/net/ynl/ynl-gen-rst.py
index 5c7465d6befa..1096a71d7867 100755
--- a/tools/net/ynl/ynl-gen-rst.py
+++ b/tools/net/ynl/ynl-gen-rst.py
@@ -178,7 +178,7 @@ def parse_operations(operations: List[Dict[str, Any]], namespace: str) -> str:
 
     for operation in operations:
         lines.append(rst_section(namespace, 'operation', operation["name"]))
-        lines.append(rst_paragraph(sanitize(operation["doc"])) + "\n")
+        lines.append(rst_paragraph(operation["doc"]) + "\n")
 
         for key in operation.keys():
             if key in preprocessed:
-- 
cgit v1.2.3-58-ga151


From cb7351ac17862cf8d4e00831ce4aa27f53bda01c Mon Sep 17 00:00:00 2001
From: Donald Hunter <donald.hunter@gmail.com>
Date: Tue, 28 May 2024 15:06:51 +0100
Subject: doc: netlink: Fix formatting of op flags in generated .rst

Generate op flags as an inline list instead of a stringified python
value.

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://lore.kernel.org/r/20240528140652.9445-4-donald.hunter@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/ynl-gen-rst.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tools/net/ynl/ynl-gen-rst.py b/tools/net/ynl/ynl-gen-rst.py
index 1096a71d7867..a957725b20dc 100755
--- a/tools/net/ynl/ynl-gen-rst.py
+++ b/tools/net/ynl/ynl-gen-rst.py
@@ -172,7 +172,7 @@ def parse_do_attributes(attrs: Dict[str, Any], level: int = 0) -> str:
 
 def parse_operations(operations: List[Dict[str, Any]], namespace: str) -> str:
     """Parse operations block"""
-    preprocessed = ["name", "doc", "title", "do", "dump"]
+    preprocessed = ["name", "doc", "title", "do", "dump", "flags"]
     linkable = ["fixed-header", "attribute-set"]
     lines = []
 
@@ -188,6 +188,8 @@ def parse_operations(operations: List[Dict[str, Any]], namespace: str) -> str:
             if key in linkable:
                 value = rst_ref(namespace, key, value)
             lines.append(rst_fields(key, value, 0))
+        if 'flags' in operation:
+            lines.append(rst_fields('flags', rst_list_inline(operation['flags'])))
 
         if "do" in operation:
             lines.append(rst_paragraph(":do:", 0))
-- 
cgit v1.2.3-58-ga151


From 9104feed4c6454b9a720e7e11047be7e5cd83487 Mon Sep 17 00:00:00 2001
From: Donald Hunter <donald.hunter@gmail.com>
Date: Tue, 28 May 2024 15:06:52 +0100
Subject: doc: netlink: Fix op pre and post fields in generated .rst

The generated .rst has pre and post headings without any values, e.g.
here:

https://docs.kernel.org/6.9/networking/netlink_spec/dpll.html#device-id-get

Emit keys and values in the generated .rst

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://lore.kernel.org/r/20240528140652.9445-5-donald.hunter@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/ynl-gen-rst.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/tools/net/ynl/ynl-gen-rst.py b/tools/net/ynl/ynl-gen-rst.py
index a957725b20dc..6c56d0d726b4 100755
--- a/tools/net/ynl/ynl-gen-rst.py
+++ b/tools/net/ynl/ynl-gen-rst.py
@@ -156,7 +156,10 @@ def parse_do(do_dict: Dict[str, Any], level: int = 0) -> str:
     lines = []
     for key in do_dict.keys():
         lines.append(rst_paragraph(bold(key), level + 1))
-        lines.append(parse_do_attributes(do_dict[key], level + 1) + "\n")
+        if key in ['request', 'reply']:
+            lines.append(parse_do_attributes(do_dict[key], level + 1) + "\n")
+        else:
+            lines.append(headroom(level + 2) + do_dict[key] + "\n")
 
     return "\n".join(lines)
 
-- 
cgit v1.2.3-58-ga151


From bbb31b7ae14594aa2a7e74923ee38f312404ad66 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 28 May 2024 16:05:09 +0100
Subject: net: dsa: remove mac_prepare()/mac_finish() shims

No DSA driver makes use of the mac_prepare()/mac_finish() shimmed
operations anymore, so we can remove these.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://lore.kernel.org/r/E1sByNx-00ELW1-Vp@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/dsa.h |  6 ------
 net/dsa/dsa.c     |  2 --
 net/dsa/port.c    | 32 --------------------------------
 3 files changed, 40 deletions(-)

diff --git a/include/net/dsa.h b/include/net/dsa.h
index b60e7e410aba..f9ae3ca66b6f 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -882,15 +882,9 @@ struct dsa_switch_ops {
 	struct phylink_pcs *(*phylink_mac_select_pcs)(struct dsa_switch *ds,
 						      int port,
 						      phy_interface_t iface);
-	int	(*phylink_mac_prepare)(struct dsa_switch *ds, int port,
-				       unsigned int mode,
-				       phy_interface_t interface);
 	void	(*phylink_mac_config)(struct dsa_switch *ds, int port,
 				      unsigned int mode,
 				      const struct phylink_link_state *state);
-	int	(*phylink_mac_finish)(struct dsa_switch *ds, int port,
-				      unsigned int mode,
-				      phy_interface_t interface);
 	void	(*phylink_mac_link_down)(struct dsa_switch *ds, int port,
 					 unsigned int mode,
 					 phy_interface_t interface);
diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index 12521a7d4048..668c729946ea 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -1507,9 +1507,7 @@ static int dsa_switch_probe(struct dsa_switch *ds)
 
 	if (ds->phylink_mac_ops) {
 		if (ds->ops->phylink_mac_select_pcs ||
-		    ds->ops->phylink_mac_prepare ||
 		    ds->ops->phylink_mac_config ||
-		    ds->ops->phylink_mac_finish ||
 		    ds->ops->phylink_mac_link_down ||
 		    ds->ops->phylink_mac_link_up)
 			return -EINVAL;
diff --git a/net/dsa/port.c b/net/dsa/port.c
index 9a249d4ac3a5..e23db9507546 100644
--- a/net/dsa/port.c
+++ b/net/dsa/port.c
@@ -1549,21 +1549,6 @@ dsa_port_phylink_mac_select_pcs(struct phylink_config *config,
 	return pcs;
 }
 
-static int dsa_port_phylink_mac_prepare(struct phylink_config *config,
-					unsigned int mode,
-					phy_interface_t interface)
-{
-	struct dsa_port *dp = dsa_phylink_to_port(config);
-	struct dsa_switch *ds = dp->ds;
-	int err = 0;
-
-	if (ds->ops->phylink_mac_prepare)
-		err = ds->ops->phylink_mac_prepare(ds, dp->index, mode,
-						   interface);
-
-	return err;
-}
-
 static void dsa_port_phylink_mac_config(struct phylink_config *config,
 					unsigned int mode,
 					const struct phylink_link_state *state)
@@ -1577,21 +1562,6 @@ static void dsa_port_phylink_mac_config(struct phylink_config *config,
 	ds->ops->phylink_mac_config(ds, dp->index, mode, state);
 }
 
-static int dsa_port_phylink_mac_finish(struct phylink_config *config,
-				       unsigned int mode,
-				       phy_interface_t interface)
-{
-	struct dsa_port *dp = dsa_phylink_to_port(config);
-	struct dsa_switch *ds = dp->ds;
-	int err = 0;
-
-	if (ds->ops->phylink_mac_finish)
-		err = ds->ops->phylink_mac_finish(ds, dp->index, mode,
-						  interface);
-
-	return err;
-}
-
 static void dsa_port_phylink_mac_link_down(struct phylink_config *config,
 					   unsigned int mode,
 					   phy_interface_t interface)
@@ -1624,9 +1594,7 @@ static void dsa_port_phylink_mac_link_up(struct phylink_config *config,
 
 static const struct phylink_mac_ops dsa_port_phylink_mac_ops = {
 	.mac_select_pcs = dsa_port_phylink_mac_select_pcs,
-	.mac_prepare = dsa_port_phylink_mac_prepare,
 	.mac_config = dsa_port_phylink_mac_config,
-	.mac_finish = dsa_port_phylink_mac_finish,
 	.mac_link_down = dsa_port_phylink_mac_link_down,
 	.mac_link_up = dsa_port_phylink_mac_link_up,
 };
-- 
cgit v1.2.3-58-ga151


From ef0e51dccdc43fd849720589c3214d97b2fd453f Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 28 May 2024 16:15:42 +0100
Subject: net: dsa: felix: provide own phylink MAC operations

Convert felix to provide its own phylink MAC operations, thus
avoiding the shim layer in DSA's port.c.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://lore.kernel.org/r/E1sByYA-00EM0y-Jn@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/ocelot/felix.c           | 54 +++++++++++++++++++++-----------
 drivers/net/dsa/ocelot/felix.h           |  1 +
 drivers/net/dsa/ocelot/felix_vsc9959.c   |  1 +
 drivers/net/dsa/ocelot/ocelot_ext.c      |  1 +
 drivers/net/dsa/ocelot/seville_vsc9953.c |  1 +
 5 files changed, 40 insertions(+), 18 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index 61e95487732d..3aa66bf9eafc 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -1050,24 +1050,32 @@ static void felix_phylink_get_caps(struct dsa_switch *ds, int port,
 		  config->supported_interfaces);
 }
 
-static void felix_phylink_mac_config(struct dsa_switch *ds, int port,
+static void felix_phylink_mac_config(struct phylink_config *config,
 				     unsigned int mode,
 				     const struct phylink_link_state *state)
 {
-	struct ocelot *ocelot = ds->priv;
-	struct felix *felix = ocelot_to_felix(ocelot);
+	struct dsa_port *dp = dsa_phylink_to_port(config);
+	struct ocelot *ocelot = dp->ds->priv;
+	int port = dp->index;
+	struct felix *felix;
+
+	felix = ocelot_to_felix(ocelot);
 
 	if (felix->info->phylink_mac_config)
 		felix->info->phylink_mac_config(ocelot, port, mode, state);
 }
 
-static struct phylink_pcs *felix_phylink_mac_select_pcs(struct dsa_switch *ds,
-							int port,
-							phy_interface_t iface)
+static struct phylink_pcs *
+felix_phylink_mac_select_pcs(struct phylink_config *config,
+			     phy_interface_t iface)
 {
-	struct ocelot *ocelot = ds->priv;
-	struct felix *felix = ocelot_to_felix(ocelot);
+	struct dsa_port *dp = dsa_phylink_to_port(config);
+	struct ocelot *ocelot = dp->ds->priv;
 	struct phylink_pcs *pcs = NULL;
+	int port = dp->index;
+	struct felix *felix;
+
+	felix = ocelot_to_felix(ocelot);
 
 	if (felix->pcs && felix->pcs[port])
 		pcs = felix->pcs[port];
@@ -1075,11 +1083,13 @@ static struct phylink_pcs *felix_phylink_mac_select_pcs(struct dsa_switch *ds,
 	return pcs;
 }
 
-static void felix_phylink_mac_link_down(struct dsa_switch *ds, int port,
+static void felix_phylink_mac_link_down(struct phylink_config *config,
 					unsigned int link_an_mode,
 					phy_interface_t interface)
 {
-	struct ocelot *ocelot = ds->priv;
+	struct dsa_port *dp = dsa_phylink_to_port(config);
+	struct ocelot *ocelot = dp->ds->priv;
+	int port = dp->index;
 	struct felix *felix;
 
 	felix = ocelot_to_felix(ocelot);
@@ -1088,15 +1098,19 @@ static void felix_phylink_mac_link_down(struct dsa_switch *ds, int port,
 				     felix->info->quirks);
 }
 
-static void felix_phylink_mac_link_up(struct dsa_switch *ds, int port,
+static void felix_phylink_mac_link_up(struct phylink_config *config,
+				      struct phy_device *phydev,
 				      unsigned int link_an_mode,
 				      phy_interface_t interface,
-				      struct phy_device *phydev,
 				      int speed, int duplex,
 				      bool tx_pause, bool rx_pause)
 {
-	struct ocelot *ocelot = ds->priv;
-	struct felix *felix = ocelot_to_felix(ocelot);
+	struct dsa_port *dp = dsa_phylink_to_port(config);
+	struct ocelot *ocelot = dp->ds->priv;
+	int port = dp->index;
+	struct felix *felix;
+
+	felix = ocelot_to_felix(ocelot);
 
 	ocelot_phylink_mac_link_up(ocelot, port, phydev, link_an_mode,
 				   interface, speed, duplex, tx_pause, rx_pause,
@@ -2083,6 +2097,14 @@ static void felix_get_mm_stats(struct dsa_switch *ds, int port,
 	ocelot_port_get_mm_stats(ocelot, port, stats);
 }
 
+const struct phylink_mac_ops felix_phylink_mac_ops = {
+	.mac_select_pcs		= felix_phylink_mac_select_pcs,
+	.mac_config		= felix_phylink_mac_config,
+	.mac_link_down		= felix_phylink_mac_link_down,
+	.mac_link_up		= felix_phylink_mac_link_up,
+};
+EXPORT_SYMBOL_GPL(felix_phylink_mac_ops);
+
 const struct dsa_switch_ops felix_switch_ops = {
 	.get_tag_protocol		= felix_get_tag_protocol,
 	.change_tag_protocol		= felix_change_tag_protocol,
@@ -2104,10 +2126,6 @@ const struct dsa_switch_ops felix_switch_ops = {
 	.get_sset_count			= felix_get_sset_count,
 	.get_ts_info			= felix_get_ts_info,
 	.phylink_get_caps		= felix_phylink_get_caps,
-	.phylink_mac_config		= felix_phylink_mac_config,
-	.phylink_mac_select_pcs		= felix_phylink_mac_select_pcs,
-	.phylink_mac_link_down		= felix_phylink_mac_link_down,
-	.phylink_mac_link_up		= felix_phylink_mac_link_up,
 	.port_enable			= felix_port_enable,
 	.port_fast_age			= felix_port_fast_age,
 	.port_fdb_dump			= felix_fdb_dump,
diff --git a/drivers/net/dsa/ocelot/felix.h b/drivers/net/dsa/ocelot/felix.h
index dbf5872fe367..4d3489aaa659 100644
--- a/drivers/net/dsa/ocelot/felix.h
+++ b/drivers/net/dsa/ocelot/felix.h
@@ -82,6 +82,7 @@ struct felix_tag_proto_ops {
 			      struct netlink_ext_ack *extack);
 };
 
+extern const struct phylink_mac_ops felix_phylink_mac_ops;
 extern const struct dsa_switch_ops felix_switch_ops;
 
 /* DSA glue / front-end for struct ocelot */
diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index 85952d841f28..d4799a908abc 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -2717,6 +2717,7 @@ static int felix_pci_probe(struct pci_dev *pdev,
 	ds->num_ports = felix->info->num_ports;
 	ds->num_tx_queues = felix->info->num_tx_queues;
 	ds->ops = &felix_switch_ops;
+	ds->phylink_mac_ops = &felix_phylink_mac_ops;
 	ds->priv = ocelot;
 	felix->ds = ds;
 	felix->tag_proto = DSA_TAG_PROTO_OCELOT;
diff --git a/drivers/net/dsa/ocelot/ocelot_ext.c b/drivers/net/dsa/ocelot/ocelot_ext.c
index 22187d831c4b..a8927dc7aca4 100644
--- a/drivers/net/dsa/ocelot/ocelot_ext.c
+++ b/drivers/net/dsa/ocelot/ocelot_ext.c
@@ -96,6 +96,7 @@ static int ocelot_ext_probe(struct platform_device *pdev)
 	ds->num_tx_queues = felix->info->num_tx_queues;
 
 	ds->ops = &felix_switch_ops;
+	ds->phylink_mac_ops = &felix_phylink_mac_ops;
 	ds->priv = ocelot;
 	felix->ds = ds;
 	felix->tag_proto = DSA_TAG_PROTO_OCELOT;
diff --git a/drivers/net/dsa/ocelot/seville_vsc9953.c b/drivers/net/dsa/ocelot/seville_vsc9953.c
index 049930da0521..5ac8897e232b 100644
--- a/drivers/net/dsa/ocelot/seville_vsc9953.c
+++ b/drivers/net/dsa/ocelot/seville_vsc9953.c
@@ -1009,6 +1009,7 @@ static int seville_probe(struct platform_device *pdev)
 	ds->dev = &pdev->dev;
 	ds->num_ports = felix->info->num_ports;
 	ds->ops = &felix_switch_ops;
+	ds->phylink_mac_ops = &felix_phylink_mac_ops;
 	ds->priv = ocelot;
 	felix->ds = ds;
 	felix->tag_proto = DSA_TAG_PROTO_SEVILLE;
-- 
cgit v1.2.3-58-ga151


From c53a46b16ce2605181688ea6af5f6c8d7fb3c9c1 Mon Sep 17 00:00:00 2001
From: Thorsten Blum <thorsten.blum@toblux.com>
Date: Tue, 28 May 2024 18:00:37 +0200
Subject: net: smc91x: Remove commented out code

Remove commented out code

Signed-off-by: Thorsten Blum <thorsten.blum@toblux.com>
Link: https://lore.kernel.org/r/20240528160036.404946-2-thorsten.blum@toblux.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/smsc/smc91x.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/ethernet/smsc/smc91x.c b/drivers/net/ethernet/smsc/smc91x.c
index 78ff3af7911a..907498848028 100644
--- a/drivers/net/ethernet/smsc/smc91x.c
+++ b/drivers/net/ethernet/smsc/smc91x.c
@@ -1574,12 +1574,8 @@ smc_ethtool_set_link_ksettings(struct net_device *dev,
 		    (cmd->base.port != PORT_TP && cmd->base.port != PORT_AUI))
 			return -EINVAL;
 
-//		lp->port = cmd->base.port;
 		lp->ctl_rfduplx = cmd->base.duplex == DUPLEX_FULL;
 
-//		if (netif_running(dev))
-//			smc_set_port(dev);
-
 		ret = 0;
 	}
 
-- 
cgit v1.2.3-58-ga151


From b21faf0edefc26df604fb783c53992e61503df33 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:43:37 +0300
Subject: wifi: rtlwifi: rtl8192d: Use "rtl92d" prefix

Some functions moved from rtl8192de still use the "rtl92de" prefix.
Rename them.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/abe3624d-f986-4a24-966c-67defa962e3f@gmail.com
---
 .../wireless/realtek/rtlwifi/rtl8192d/hw_common.c  | 94 +++++++++++-----------
 .../wireless/realtek/rtlwifi/rtl8192d/hw_common.h  | 28 +++----
 .../wireless/realtek/rtlwifi/rtl8192d/trx_common.c | 92 ++++++++++-----------
 .../wireless/realtek/rtlwifi/rtl8192d/trx_common.h | 16 ++--
 .../net/wireless/realtek/rtlwifi/rtl8192de/hw.c    | 18 ++---
 .../net/wireless/realtek/rtlwifi/rtl8192de/sw.c    | 20 ++---
 .../net/wireless/realtek/rtlwifi/rtl8192de/trx.c   |  2 +-
 7 files changed, 135 insertions(+), 135 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.c
index 6570d5e168e9..97e0d9c01e0a 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.c
@@ -14,7 +14,7 @@
 #include "hw_common.h"
 #include "phy_common.h"
 
-void rtl92de_stop_tx_beacon(struct ieee80211_hw *hw)
+void rtl92d_stop_tx_beacon(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	u8 tmp1byte;
@@ -27,9 +27,9 @@ void rtl92de_stop_tx_beacon(struct ieee80211_hw *hw)
 	tmp1byte &= ~(BIT(0));
 	rtl_write_byte(rtlpriv, REG_TBTT_PROHIBIT + 2, tmp1byte);
 }
-EXPORT_SYMBOL_GPL(rtl92de_stop_tx_beacon);
+EXPORT_SYMBOL_GPL(rtl92d_stop_tx_beacon);
 
-void rtl92de_resume_tx_beacon(struct ieee80211_hw *hw)
+void rtl92d_resume_tx_beacon(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	u8 tmp1byte;
@@ -42,7 +42,7 @@ void rtl92de_resume_tx_beacon(struct ieee80211_hw *hw)
 	tmp1byte |= BIT(0);
 	rtl_write_byte(rtlpriv, REG_TBTT_PROHIBIT + 2, tmp1byte);
 }
-EXPORT_SYMBOL_GPL(rtl92de_resume_tx_beacon);
+EXPORT_SYMBOL_GPL(rtl92d_resume_tx_beacon);
 
 void rtl92d_get_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val)
 {
@@ -285,7 +285,7 @@ void rtl92d_set_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val)
 }
 EXPORT_SYMBOL_GPL(rtl92d_set_hw_reg);
 
-bool rtl92de_llt_write(struct ieee80211_hw *hw, u32 address, u32 data)
+bool rtl92d_llt_write(struct ieee80211_hw *hw, u32 address, u32 data)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	bool status = true;
@@ -307,9 +307,9 @@ bool rtl92de_llt_write(struct ieee80211_hw *hw, u32 address, u32 data)
 	} while (++count);
 	return status;
 }
-EXPORT_SYMBOL_GPL(rtl92de_llt_write);
+EXPORT_SYMBOL_GPL(rtl92d_llt_write);
 
-void rtl92de_enable_hw_security_config(struct ieee80211_hw *hw)
+void rtl92d_enable_hw_security_config(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	u8 sec_reg_value;
@@ -334,16 +334,16 @@ void rtl92de_enable_hw_security_config(struct ieee80211_hw *hw)
 		"The SECR-value %x\n", sec_reg_value);
 	rtlpriv->cfg->ops->set_hw_reg(hw, HW_VAR_WPA_CONFIG, &sec_reg_value);
 }
-EXPORT_SYMBOL_GPL(rtl92de_enable_hw_security_config);
+EXPORT_SYMBOL_GPL(rtl92d_enable_hw_security_config);
 
 /* don't set REG_EDCA_BE_PARAM here because
  * mac80211 will send pkt when scan
  */
-void rtl92de_set_qos(struct ieee80211_hw *hw, int aci)
+void rtl92d_set_qos(struct ieee80211_hw *hw, int aci)
 {
 	rtl92d_dm_init_edca_turbo(hw);
 }
-EXPORT_SYMBOL_GPL(rtl92de_set_qos);
+EXPORT_SYMBOL_GPL(rtl92d_set_qos);
 
 static enum version_8192d _rtl92d_read_chip_version(struct ieee80211_hw *hw)
 {
@@ -362,8 +362,8 @@ static enum version_8192d _rtl92d_read_chip_version(struct ieee80211_hw *hw)
 	return version;
 }
 
-static void _rtl92de_readpowervalue_fromprom(struct txpower_info *pwrinfo,
-					     u8 *efuse, bool autoloadfail)
+static void _rtl92d_readpowervalue_fromprom(struct txpower_info *pwrinfo,
+					    u8 *efuse, bool autoloadfail)
 {
 	u32 rfpath, eeaddr, group, offset, offset1, offset2;
 	u8 i, val8;
@@ -500,8 +500,8 @@ static void _rtl92de_readpowervalue_fromprom(struct txpower_info *pwrinfo,
 	}
 }
 
-static void _rtl92de_read_txpower_info(struct ieee80211_hw *hw,
-				       bool autoload_fail, u8 *hwinfo)
+static void _rtl92d_read_txpower_info(struct ieee80211_hw *hw,
+				      bool autoload_fail, u8 *hwinfo)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw));
@@ -509,7 +509,7 @@ static void _rtl92de_read_txpower_info(struct ieee80211_hw *hw,
 	u8 tempval[2], i, pwr, diff;
 	u32 ch, rfpath, group;
 
-	_rtl92de_readpowervalue_fromprom(&pwrinfo, hwinfo, autoload_fail);
+	_rtl92d_readpowervalue_fromprom(&pwrinfo, hwinfo, autoload_fail);
 	if (!autoload_fail) {
 		/* bit0~2 */
 		rtlefuse->eeprom_regulatory = (hwinfo[EEPROM_RF_OPT1] & 0x7);
@@ -613,8 +613,8 @@ static void _rtl92de_read_txpower_info(struct ieee80211_hw *hw,
 	}
 }
 
-static void _rtl92de_read_macphymode_from_prom(struct ieee80211_hw *hw,
-					       u8 *content)
+static void _rtl92d_read_macphymode_from_prom(struct ieee80211_hw *hw,
+					      u8 *content)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw));
@@ -636,15 +636,15 @@ static void _rtl92de_read_macphymode_from_prom(struct ieee80211_hw *hw,
 	}
 }
 
-static void _rtl92de_read_macphymode_and_bandtype(struct ieee80211_hw *hw,
-						  u8 *content)
+static void _rtl92d_read_macphymode_and_bandtype(struct ieee80211_hw *hw,
+						 u8 *content)
 {
-	_rtl92de_read_macphymode_from_prom(hw, content);
+	_rtl92d_read_macphymode_from_prom(hw, content);
 	rtl92d_phy_config_macphymode(hw);
 	rtl92d_phy_config_macphymode_info(hw);
 }
 
-static void _rtl92de_efuse_update_chip_version(struct ieee80211_hw *hw)
+static void _rtl92d_efuse_update_chip_version(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	enum version_8192d chipver = rtlpriv->rtlhal.version;
@@ -676,7 +676,7 @@ static void _rtl92de_efuse_update_chip_version(struct ieee80211_hw *hw)
 	rtlpriv->rtlhal.version = chipver;
 }
 
-static void _rtl92de_read_adapter_info(struct ieee80211_hw *hw)
+static void _rtl92d_read_adapter_info(struct ieee80211_hw *hw)
 {
 	static const int params_pci[] = {
 		RTL8190_EEPROM_ID, EEPROM_VID, EEPROM_DID,
@@ -706,8 +706,8 @@ static void _rtl92de_read_adapter_info(struct ieee80211_hw *hw)
 	if (rtl_get_hwinfo(hw, rtlpriv, HWSET_MAX_SIZE, hwinfo, params))
 		goto exit;
 
-	_rtl92de_efuse_update_chip_version(hw);
-	_rtl92de_read_macphymode_and_bandtype(hw, hwinfo);
+	_rtl92d_efuse_update_chip_version(hw);
+	_rtl92d_read_macphymode_and_bandtype(hw, hwinfo);
 
 	/* Read Permanent MAC address for 2nd interface */
 	if (rtlhal->interfaceindex != 0)
@@ -717,7 +717,7 @@ static void _rtl92de_read_adapter_info(struct ieee80211_hw *hw)
 	rtlpriv->cfg->ops->set_hw_reg(hw, HW_VAR_ETHER_ADDR,
 				      rtlefuse->dev_addr);
 	rtl_dbg(rtlpriv, COMP_INIT, DBG_DMESG, "%pM\n", rtlefuse->dev_addr);
-	_rtl92de_read_txpower_info(hw, rtlefuse->autoload_failflag, hwinfo);
+	_rtl92d_read_txpower_info(hw, rtlefuse->autoload_failflag, hwinfo);
 
 	/* Read Channel Plan */
 	switch (rtlhal->bandset) {
@@ -739,7 +739,7 @@ exit:
 	kfree(hwinfo);
 }
 
-void rtl92de_read_eeprom_info(struct ieee80211_hw *hw)
+void rtl92d_read_eeprom_info(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw));
@@ -760,15 +760,15 @@ void rtl92de_read_eeprom_info(struct ieee80211_hw *hw)
 		rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD, "Autoload OK\n");
 
 		rtlefuse->autoload_failflag = false;
-		_rtl92de_read_adapter_info(hw);
+		_rtl92d_read_adapter_info(hw);
 	} else {
 		pr_err("Autoload ERR!!\n");
 	}
 }
-EXPORT_SYMBOL_GPL(rtl92de_read_eeprom_info);
+EXPORT_SYMBOL_GPL(rtl92d_read_eeprom_info);
 
-static void rtl92de_update_hal_rate_table(struct ieee80211_hw *hw,
-					  struct ieee80211_sta *sta)
+static void rtl92d_update_hal_rate_table(struct ieee80211_hw *hw,
+					 struct ieee80211_sta *sta)
 {
 	struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw));
 	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
@@ -851,9 +851,9 @@ static void rtl92de_update_hal_rate_table(struct ieee80211_hw *hw,
 		rtl_read_dword(rtlpriv, REG_ARFR0));
 }
 
-static void rtl92de_update_hal_rate_mask(struct ieee80211_hw *hw,
-					 struct ieee80211_sta *sta,
-					 u8 rssi_level, bool update_bw)
+static void rtl92d_update_hal_rate_mask(struct ieee80211_hw *hw,
+					struct ieee80211_sta *sta,
+					u8 rssi_level, bool update_bw)
 {
 	struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw));
 	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
@@ -1009,20 +1009,20 @@ static void rtl92de_update_hal_rate_mask(struct ieee80211_hw *hw,
 		sta_entry->ratr_index = ratr_index;
 }
 
-void rtl92de_update_hal_rate_tbl(struct ieee80211_hw *hw,
-				 struct ieee80211_sta *sta,
-				 u8 rssi_level, bool update_bw)
+void rtl92d_update_hal_rate_tbl(struct ieee80211_hw *hw,
+				struct ieee80211_sta *sta,
+				u8 rssi_level, bool update_bw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 
 	if (rtlpriv->dm.useramask)
-		rtl92de_update_hal_rate_mask(hw, sta, rssi_level, update_bw);
+		rtl92d_update_hal_rate_mask(hw, sta, rssi_level, update_bw);
 	else
-		rtl92de_update_hal_rate_table(hw, sta);
+		rtl92d_update_hal_rate_table(hw, sta);
 }
-EXPORT_SYMBOL_GPL(rtl92de_update_hal_rate_tbl);
+EXPORT_SYMBOL_GPL(rtl92d_update_hal_rate_tbl);
 
-void rtl92de_update_channel_access_setting(struct ieee80211_hw *hw)
+void rtl92d_update_channel_access_setting(struct ieee80211_hw *hw)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
@@ -1036,9 +1036,9 @@ void rtl92de_update_channel_access_setting(struct ieee80211_hw *hw)
 		sifs_timer = 0x1010;
 	rtlpriv->cfg->ops->set_hw_reg(hw, HW_VAR_SIFS, (u8 *)&sifs_timer);
 }
-EXPORT_SYMBOL_GPL(rtl92de_update_channel_access_setting);
+EXPORT_SYMBOL_GPL(rtl92d_update_channel_access_setting);
 
-bool rtl92de_gpio_radio_on_off_checking(struct ieee80211_hw *hw, u8 *valid)
+bool rtl92d_gpio_radio_on_off_checking(struct ieee80211_hw *hw, u8 *valid)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_ps_ctl *ppsc = rtl_psc(rtl_priv(hw));
@@ -1093,11 +1093,11 @@ bool rtl92de_gpio_radio_on_off_checking(struct ieee80211_hw *hw, u8 *valid)
 	*valid = 1;
 	return !ppsc->hwradiooff;
 }
-EXPORT_SYMBOL_GPL(rtl92de_gpio_radio_on_off_checking);
+EXPORT_SYMBOL_GPL(rtl92d_gpio_radio_on_off_checking);
 
-void rtl92de_set_key(struct ieee80211_hw *hw, u32 key_index,
-		     u8 *p_macaddr, bool is_group, u8 enc_algo,
-		     bool is_wepkey, bool clear_all)
+void rtl92d_set_key(struct ieee80211_hw *hw, u32 key_index,
+		    u8 *p_macaddr, bool is_group, u8 enc_algo,
+		    bool is_wepkey, bool clear_all)
 {
 	static const u8 cam_const_addr[4][6] = {
 		{0x00, 0x00, 0x00, 0x00, 0x00, 0x00},
@@ -1222,4 +1222,4 @@ void rtl92de_set_key(struct ieee80211_hw *hw, u32 key_index,
 		}
 	}
 }
-EXPORT_SYMBOL_GPL(rtl92de_set_key);
+EXPORT_SYMBOL_GPL(rtl92d_set_key);
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.h
index 2c07f5cc5766..4da1bab15f36 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192d/hw_common.h
@@ -4,21 +4,21 @@
 #ifndef __RTL92D_HW_COMMON_H__
 #define __RTL92D_HW_COMMON_H__
 
-void rtl92de_stop_tx_beacon(struct ieee80211_hw *hw);
-void rtl92de_resume_tx_beacon(struct ieee80211_hw *hw);
+void rtl92d_stop_tx_beacon(struct ieee80211_hw *hw);
+void rtl92d_resume_tx_beacon(struct ieee80211_hw *hw);
 void rtl92d_get_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val);
 void rtl92d_set_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val);
-bool rtl92de_llt_write(struct ieee80211_hw *hw, u32 address, u32 data);
-void rtl92de_enable_hw_security_config(struct ieee80211_hw *hw);
-void rtl92de_set_qos(struct ieee80211_hw *hw, int aci);
-void rtl92de_read_eeprom_info(struct ieee80211_hw *hw);
-void rtl92de_update_hal_rate_tbl(struct ieee80211_hw *hw,
-				 struct ieee80211_sta *sta,
-				 u8 rssi_level, bool update_bw);
-void rtl92de_update_channel_access_setting(struct ieee80211_hw *hw);
-bool rtl92de_gpio_radio_on_off_checking(struct ieee80211_hw *hw, u8 *valid);
-void rtl92de_set_key(struct ieee80211_hw *hw, u32 key_index,
-		     u8 *p_macaddr, bool is_group, u8 enc_algo,
-		     bool is_wepkey, bool clear_all);
+bool rtl92d_llt_write(struct ieee80211_hw *hw, u32 address, u32 data);
+void rtl92d_enable_hw_security_config(struct ieee80211_hw *hw);
+void rtl92d_set_qos(struct ieee80211_hw *hw, int aci);
+void rtl92d_read_eeprom_info(struct ieee80211_hw *hw);
+void rtl92d_update_hal_rate_tbl(struct ieee80211_hw *hw,
+				struct ieee80211_sta *sta,
+				u8 rssi_level, bool update_bw);
+void rtl92d_update_channel_access_setting(struct ieee80211_hw *hw);
+bool rtl92d_gpio_radio_on_off_checking(struct ieee80211_hw *hw, u8 *valid);
+void rtl92d_set_key(struct ieee80211_hw *hw, u32 key_index,
+		    u8 *p_macaddr, bool is_group, u8 enc_algo,
+		    bool is_wepkey, bool clear_all);
 
 #endif
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192d/trx_common.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192d/trx_common.c
index 72d2b7426d82..9f9a34492030 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192d/trx_common.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192d/trx_common.c
@@ -7,8 +7,8 @@
 #include "def.h"
 #include "trx_common.h"
 
-static long _rtl92de_translate_todbm(struct ieee80211_hw *hw,
-				     u8 signal_strength_index)
+static long _rtl92d_translate_todbm(struct ieee80211_hw *hw,
+				    u8 signal_strength_index)
 {
 	long signal_power;
 
@@ -17,13 +17,13 @@ static long _rtl92de_translate_todbm(struct ieee80211_hw *hw,
 	return signal_power;
 }
 
-static void _rtl92de_query_rxphystatus(struct ieee80211_hw *hw,
-				       struct rtl_stats *pstats,
-				       __le32 *pdesc,
-				       struct rx_fwinfo_92d *p_drvinfo,
-				       bool packet_match_bssid,
-				       bool packet_toself,
-				       bool packet_beacon)
+static void _rtl92d_query_rxphystatus(struct ieee80211_hw *hw,
+				      struct rtl_stats *pstats,
+				      __le32 *pdesc,
+				      struct rx_fwinfo_92d *p_drvinfo,
+				      bool packet_match_bssid,
+				      bool packet_toself,
+				      bool packet_beacon)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_phy *rtlphy = &rtlpriv->phy;
@@ -203,8 +203,8 @@ static void rtl92d_loop_over_paths(struct ieee80211_hw *hw,
 	}
 }
 
-static void _rtl92de_process_ui_rssi(struct ieee80211_hw *hw,
-				     struct rtl_stats *pstats)
+static void _rtl92d_process_ui_rssi(struct ieee80211_hw *hw,
+				    struct rtl_stats *pstats)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rt_smooth_data *ui_rssi;
@@ -226,15 +226,15 @@ static void _rtl92de_process_ui_rssi(struct ieee80211_hw *hw,
 	if (ui_rssi->index >= PHY_RSSI_SLID_WIN_MAX)
 		ui_rssi->index = 0;
 	tmpval = ui_rssi->total_val / ui_rssi->total_num;
-	rtlpriv->stats.signal_strength = _rtl92de_translate_todbm(hw, (u8)tmpval);
+	rtlpriv->stats.signal_strength = _rtl92d_translate_todbm(hw, (u8)tmpval);
 	pstats->rssi = rtlpriv->stats.signal_strength;
 
 	if (!pstats->is_cck && pstats->packet_toself)
 		rtl92d_loop_over_paths(hw, pstats);
 }
 
-static void _rtl92de_update_rxsignalstatistics(struct ieee80211_hw *hw,
-					       struct rtl_stats *pstats)
+static void _rtl92d_update_rxsignalstatistics(struct ieee80211_hw *hw,
+					      struct rtl_stats *pstats)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	int weighting = 0;
@@ -249,8 +249,8 @@ static void _rtl92de_update_rxsignalstatistics(struct ieee80211_hw *hw,
 		5 + pstats->recvsignalpower + weighting) / 6;
 }
 
-static void _rtl92de_process_pwdb(struct ieee80211_hw *hw,
-				  struct rtl_stats *pstats)
+static void _rtl92d_process_pwdb(struct ieee80211_hw *hw,
+				 struct rtl_stats *pstats)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
@@ -276,7 +276,7 @@ static void _rtl92de_process_pwdb(struct ieee80211_hw *hw,
 			      (pstats->rx_pwdb_all)) / (RX_SMOOTH_FACTOR);
 		}
 		rtlpriv->dm.undec_sm_pwdb = undec_sm_pwdb;
-		_rtl92de_update_rxsignalstatistics(hw, pstats);
+		_rtl92d_update_rxsignalstatistics(hw, pstats);
 	}
 }
 
@@ -301,8 +301,8 @@ static void rtl92d_loop_over_streams(struct ieee80211_hw *hw,
 	}
 }
 
-static void _rtl92de_process_ui_link_quality(struct ieee80211_hw *hw,
-					     struct rtl_stats *pstats)
+static void _rtl92d_process_ui_link_quality(struct ieee80211_hw *hw,
+					    struct rtl_stats *pstats)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rt_smooth_data *ui_link_quality;
@@ -330,24 +330,24 @@ static void _rtl92de_process_ui_link_quality(struct ieee80211_hw *hw,
 	rtl92d_loop_over_streams(hw, pstats);
 }
 
-static void _rtl92de_process_phyinfo(struct ieee80211_hw *hw,
-				     u8 *buffer,
-				     struct rtl_stats *pcurrent_stats)
+static void _rtl92d_process_phyinfo(struct ieee80211_hw *hw,
+				    u8 *buffer,
+				    struct rtl_stats *pcurrent_stats)
 {
 	if (!pcurrent_stats->packet_matchbssid &&
 	    !pcurrent_stats->packet_beacon)
 		return;
 
-	_rtl92de_process_ui_rssi(hw, pcurrent_stats);
-	_rtl92de_process_pwdb(hw, pcurrent_stats);
-	_rtl92de_process_ui_link_quality(hw, pcurrent_stats);
+	_rtl92d_process_ui_rssi(hw, pcurrent_stats);
+	_rtl92d_process_pwdb(hw, pcurrent_stats);
+	_rtl92d_process_ui_link_quality(hw, pcurrent_stats);
 }
 
-static void _rtl92de_translate_rx_signal_stuff(struct ieee80211_hw *hw,
-					       struct sk_buff *skb,
-					       struct rtl_stats *pstats,
-					       __le32 *pdesc,
-					       struct rx_fwinfo_92d *p_drvinfo)
+static void _rtl92d_translate_rx_signal_stuff(struct ieee80211_hw *hw,
+					      struct sk_buff *skb,
+					      struct rtl_stats *pstats,
+					      __le32 *pdesc,
+					      struct rx_fwinfo_92d *p_drvinfo)
 {
 	struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw));
 	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
@@ -375,15 +375,15 @@ static void _rtl92de_translate_rx_signal_stuff(struct ieee80211_hw *hw,
 	packet_toself = packet_matchbssid &&
 			ether_addr_equal(praddr, rtlefuse->dev_addr);
 	packet_beacon = ieee80211_is_beacon(fc);
-	_rtl92de_query_rxphystatus(hw, pstats, pdesc, p_drvinfo,
-				   packet_matchbssid, packet_toself,
-				   packet_beacon);
-	_rtl92de_process_phyinfo(hw, tmp_buf, pstats);
+	_rtl92d_query_rxphystatus(hw, pstats, pdesc, p_drvinfo,
+				  packet_matchbssid, packet_toself,
+				  packet_beacon);
+	_rtl92d_process_phyinfo(hw, tmp_buf, pstats);
 }
 
-bool rtl92de_rx_query_desc(struct ieee80211_hw *hw, struct rtl_stats *stats,
-			   struct ieee80211_rx_status *rx_status,
-			   u8 *pdesc8, struct sk_buff *skb)
+bool rtl92d_rx_query_desc(struct ieee80211_hw *hw, struct rtl_stats *stats,
+			  struct ieee80211_rx_status *rx_status,
+			  u8 *pdesc8, struct sk_buff *skb)
 {
 	__le32 *pdesc = (__le32 *)pdesc8;
 	struct rx_fwinfo_92d *p_drvinfo;
@@ -423,17 +423,17 @@ bool rtl92de_rx_query_desc(struct ieee80211_hw *hw, struct rtl_stats *stats,
 	if (phystatus) {
 		p_drvinfo = (struct rx_fwinfo_92d *)(skb->data +
 						     stats->rx_bufshift);
-		_rtl92de_translate_rx_signal_stuff(hw, skb, stats, pdesc,
-						   p_drvinfo);
+		_rtl92d_translate_rx_signal_stuff(hw, skb, stats, pdesc,
+						  p_drvinfo);
 	}
 	/*rx_status->qual = stats->signal; */
 	rx_status->signal = stats->recvsignalpower + 10;
 	return true;
 }
-EXPORT_SYMBOL_GPL(rtl92de_rx_query_desc);
+EXPORT_SYMBOL_GPL(rtl92d_rx_query_desc);
 
-void rtl92de_set_desc(struct ieee80211_hw *hw, u8 *pdesc8, bool istx,
-		      u8 desc_name, u8 *val)
+void rtl92d_set_desc(struct ieee80211_hw *hw, u8 *pdesc8, bool istx,
+		     u8 desc_name, u8 *val)
 {
 	__le32 *pdesc = (__le32 *)pdesc8;
 
@@ -473,10 +473,10 @@ void rtl92de_set_desc(struct ieee80211_hw *hw, u8 *pdesc8, bool istx,
 		}
 	}
 }
-EXPORT_SYMBOL_GPL(rtl92de_set_desc);
+EXPORT_SYMBOL_GPL(rtl92d_set_desc);
 
-u64 rtl92de_get_desc(struct ieee80211_hw *hw,
-		     u8 *p_desc8, bool istx, u8 desc_name)
+u64 rtl92d_get_desc(struct ieee80211_hw *hw,
+		    u8 *p_desc8, bool istx, u8 desc_name)
 {
 	__le32 *p_desc = (__le32 *)p_desc8;
 	u32 ret = 0;
@@ -513,4 +513,4 @@ u64 rtl92de_get_desc(struct ieee80211_hw *hw,
 	}
 	return ret;
 }
-EXPORT_SYMBOL_GPL(rtl92de_get_desc);
+EXPORT_SYMBOL_GPL(rtl92d_get_desc);
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192d/trx_common.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192d/trx_common.h
index 87d956d771eb..528182b1eba6 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192d/trx_common.h
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192d/trx_common.h
@@ -393,13 +393,13 @@ struct rx_fwinfo_92d {
 #endif
 } __packed;
 
-bool rtl92de_rx_query_desc(struct ieee80211_hw *hw,
-			   struct rtl_stats *stats,
-			   struct ieee80211_rx_status *rx_status,
-			   u8 *pdesc, struct sk_buff *skb);
-void rtl92de_set_desc(struct ieee80211_hw *hw, u8 *pdesc, bool istx,
-		      u8 desc_name, u8 *val);
-u64 rtl92de_get_desc(struct ieee80211_hw *hw,
-		     u8 *p_desc, bool istx, u8 desc_name);
+bool rtl92d_rx_query_desc(struct ieee80211_hw *hw,
+			  struct rtl_stats *stats,
+			  struct ieee80211_rx_status *rx_status,
+			  u8 *pdesc, struct sk_buff *skb);
+void rtl92d_set_desc(struct ieee80211_hw *hw, u8 *pdesc, bool istx,
+		     u8 desc_name, u8 *val);
+u64 rtl92d_get_desc(struct ieee80211_hw *hw,
+		    u8 *p_desc, bool istx, u8 desc_name);
 
 #endif
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c
index 73b81e60cfa9..03f4314bdb2e 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c
@@ -181,7 +181,7 @@ void rtl92de_set_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val)
 		u8 btype_ibss = val[0];
 
 		if (btype_ibss)
-			rtl92de_stop_tx_beacon(hw);
+			rtl92d_stop_tx_beacon(hw);
 		_rtl92de_set_bcn_ctrl_reg(hw, 0, BIT(3));
 		rtl_write_dword(rtlpriv, REG_TSFTR,
 				(u32) (mac->tsf & 0xffffffff));
@@ -189,7 +189,7 @@ void rtl92de_set_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val)
 				(u32) ((mac->tsf >> 32) & 0xffffffff));
 		_rtl92de_set_bcn_ctrl_reg(hw, BIT(3), 0);
 		if (btype_ibss)
-			rtl92de_resume_tx_beacon(hw);
+			rtl92d_resume_tx_beacon(hw);
 
 		break;
 	}
@@ -295,13 +295,13 @@ static bool _rtl92de_llt_table_init(struct ieee80211_hw *hw)
 
 	/* 18.  LLT_table_init(Adapter);  */
 	for (i = 0; i < (txpktbuf_bndy - 1); i++) {
-		status = rtl92de_llt_write(hw, i, i + 1);
+		status = rtl92d_llt_write(hw, i, i + 1);
 		if (!status)
 			return status;
 	}
 
 	/* end of list */
-	status = rtl92de_llt_write(hw, (txpktbuf_bndy - 1), 0xFF);
+	status = rtl92d_llt_write(hw, (txpktbuf_bndy - 1), 0xFF);
 	if (!status)
 		return status;
 
@@ -310,13 +310,13 @@ static bool _rtl92de_llt_table_init(struct ieee80211_hw *hw)
 	/* config this MAC as two MAC transfer. */
 	/* Otherwise used as local loopback buffer.  */
 	for (i = txpktbuf_bndy; i < maxpage; i++) {
-		status = rtl92de_llt_write(hw, i, (i + 1));
+		status = rtl92d_llt_write(hw, i, (i + 1));
 		if (!status)
 			return status;
 	}
 
 	/* Let last entry point to the start entry of ring buffer */
-	status = rtl92de_llt_write(hw, maxpage, txpktbuf_bndy);
+	status = rtl92d_llt_write(hw, maxpage, txpktbuf_bndy);
 	if (!status)
 		return status;
 
@@ -688,7 +688,7 @@ int rtl92de_hw_init(struct ieee80211_hw *hw)
 
 	/* reset hw sec */
 	rtl_cam_reset_all_entry(hw);
-	rtl92de_enable_hw_security_config(hw);
+	rtl92d_enable_hw_security_config(hw);
 
 	/* Read EEPROM TX power index and PHY_REG_PG.txt to capture correct */
 	/* TX power index for different rate set. */
@@ -742,11 +742,11 @@ static int _rtl92de_set_media_status(struct ieee80211_hw *hw,
 
 	if (type == NL80211_IFTYPE_UNSPECIFIED ||
 	    type == NL80211_IFTYPE_STATION) {
-		rtl92de_stop_tx_beacon(hw);
+		rtl92d_stop_tx_beacon(hw);
 		_rtl92de_enable_bcn_sub_func(hw);
 	} else if (type == NL80211_IFTYPE_ADHOC ||
 		type == NL80211_IFTYPE_AP) {
-		rtl92de_resume_tx_beacon(hw);
+		rtl92d_resume_tx_beacon(hw);
 		_rtl92de_disable_bcn_sub_func(hw);
 	} else {
 		rtl_dbg(rtlpriv, COMP_ERR, DBG_WARNING,
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/sw.c
index 5f6311c2aac4..f5ce4889523e 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/sw.c
@@ -187,7 +187,7 @@ static void rtl92d_deinit_sw_vars(struct ieee80211_hw *hw)
 static struct rtl_hal_ops rtl8192de_hal_ops = {
 	.init_sw_vars = rtl92d_init_sw_vars,
 	.deinit_sw_vars = rtl92d_deinit_sw_vars,
-	.read_eeprom_info = rtl92de_read_eeprom_info,
+	.read_eeprom_info = rtl92d_read_eeprom_info,
 	.interrupt_recognized = rtl92de_interrupt_recognized,
 	.hw_init = rtl92de_hw_init,
 	.hw_disable = rtl92de_card_disable,
@@ -197,30 +197,30 @@ static struct rtl_hal_ops rtl8192de_hal_ops = {
 	.disable_interrupt = rtl92de_disable_interrupt,
 	.set_network_type = rtl92de_set_network_type,
 	.set_chk_bssid = rtl92de_set_check_bssid,
-	.set_qos = rtl92de_set_qos,
+	.set_qos = rtl92d_set_qos,
 	.set_bcn_reg = rtl92de_set_beacon_related_registers,
 	.set_bcn_intv = rtl92de_set_beacon_interval,
 	.update_interrupt_mask = rtl92de_update_interrupt_mask,
 	.get_hw_reg = rtl92de_get_hw_reg,
 	.set_hw_reg = rtl92de_set_hw_reg,
-	.update_rate_tbl = rtl92de_update_hal_rate_tbl,
+	.update_rate_tbl = rtl92d_update_hal_rate_tbl,
 	.fill_tx_desc = rtl92de_tx_fill_desc,
 	.fill_tx_cmddesc = rtl92de_tx_fill_cmddesc,
-	.query_rx_desc = rtl92de_rx_query_desc,
-	.set_channel_access = rtl92de_update_channel_access_setting,
-	.radio_onoff_checking = rtl92de_gpio_radio_on_off_checking,
+	.query_rx_desc = rtl92d_rx_query_desc,
+	.set_channel_access = rtl92d_update_channel_access_setting,
+	.radio_onoff_checking = rtl92d_gpio_radio_on_off_checking,
 	.set_bw_mode = rtl92d_phy_set_bw_mode,
 	.switch_channel = rtl92d_phy_sw_chnl,
 	.dm_watchdog = rtl92de_dm_watchdog,
 	.scan_operation_backup = rtl_phy_scan_operation_backup,
 	.set_rf_power_state = rtl92d_phy_set_rf_power_state,
 	.led_control = rtl92de_led_control,
-	.set_desc = rtl92de_set_desc,
-	.get_desc = rtl92de_get_desc,
+	.set_desc = rtl92d_set_desc,
+	.get_desc = rtl92d_get_desc,
 	.is_tx_desc_closed = rtl92de_is_tx_desc_closed,
 	.tx_polling = rtl92de_tx_polling,
-	.enable_hw_sec = rtl92de_enable_hw_security_config,
-	.set_key = rtl92de_set_key,
+	.enable_hw_sec = rtl92d_enable_hw_security_config,
+	.set_key = rtl92d_set_key,
 	.get_bbreg = rtl92d_phy_query_bb_reg,
 	.set_bbreg = rtl92d_phy_set_bb_reg,
 	.get_rfreg = rtl92d_phy_query_rf_reg,
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
index 2b9b352f7783..91bf399c9ef1 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c
@@ -292,7 +292,7 @@ bool rtl92de_is_tx_desc_closed(struct ieee80211_hw *hw,
 	struct rtl_pci *rtlpci = rtl_pcidev(rtl_pcipriv(hw));
 	struct rtl8192_tx_ring *ring = &rtlpci->tx_ring[hw_queue];
 	u8 *entry = (u8 *)(&ring->desc[ring->idx]);
-	u8 own = (u8)rtl92de_get_desc(hw, entry, true, HW_DESC_OWN);
+	u8 own = (u8)rtl92d_get_desc(hw, entry, true, HW_DESC_OWN);
 
 	/* a beacon packet will only use the first
 	 * descriptor by defaut, and the own bit may not
-- 
cgit v1.2.3-58-ga151


From ed4117567419a9f7e9bfc32855525ce56b98899a Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:44:09 +0300
Subject: wifi: rtlwifi: Add rtl8192du/table.{c,h}

These contain the MAC, BB, RF, and AGC initialisation tables for
RTL8192DU.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/5ea548b9-5ec6-4358-aa5d-94071f2341cb@gmail.com
---
 .../net/wireless/realtek/rtlwifi/rtl8192du/table.c | 1675 ++++++++++++++++++++
 .../net/wireless/realtek/rtlwifi/rtl8192du/table.h |   29 +
 2 files changed, 1704 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/table.c
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/table.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/table.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/table.c
new file mode 100644
index 000000000000..036701433d85
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/table.c
@@ -0,0 +1,1675 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#include <linux/types.h>
+
+#include "table.h"
+
+const u32 rtl8192du_phy_reg_2tarray[PHY_REG_2T_ARRAYLENGTH] = {
+	0x800, 0x80040002,
+	0x804, 0x00000003,
+	0x808, 0x0000fc00,
+	0x80c, 0x0000000a,
+	0x810, 0x10001331,
+	0x814, 0x020c3d10,
+	0x818, 0x02200385,
+	0x81c, 0x00000000,
+	0x820, 0x01000100,
+	0x824, 0x00390004,
+	0x828, 0x01000100,
+	0x82c, 0x00390004,
+	0x830, 0x27272727,
+	0x834, 0x27272727,
+	0x838, 0x27272727,
+	0x83c, 0x27272727,
+	0x840, 0x00010000,
+	0x844, 0x00010000,
+	0x848, 0x27272727,
+	0x84c, 0x27272727,
+	0x850, 0x00000000,
+	0x854, 0x00000000,
+	0x858, 0x569a569a,
+	0x85c, 0x0c1b25a4,
+	0x860, 0x66e60250,
+	0x864, 0x061f0150,
+	0x868, 0x27272727,
+	0x86c, 0x272b2b2b,
+	0x870, 0x07000700,
+	0x874, 0x22188000,
+	0x878, 0x08080808,
+	0x87c, 0x0001fff8,
+	0x880, 0xc0083070,
+	0x884, 0x00000cd5,
+	0x888, 0x00000000,
+	0x88c, 0xcc0000c0,
+	0x890, 0x00000800,
+	0x894, 0xfffffffe,
+	0x898, 0x40302010,
+	0x89c, 0x00706050,
+	0x900, 0x00000000,
+	0x904, 0x00000023,
+	0x908, 0x00000000,
+	0x90c, 0x81121313,
+	0xa00, 0x00d047c8,
+	0xa04, 0x80ff000c,
+	0xa08, 0x8c8a8300,
+	0xa0c, 0x2e68120f,
+	0xa10, 0x9500bb78,
+	0xa14, 0x11144028,
+	0xa18, 0x00881117,
+	0xa1c, 0x89140f00,
+	0xa20, 0x1a1b0000,
+	0xa24, 0x090e1317,
+	0xa28, 0x00000204,
+	0xa2c, 0x00d30000,
+	0xa70, 0x101fff00,
+	0xa74, 0x00000007,
+	0xc00, 0x40071d40,
+	0xc04, 0x03a05633,
+	0xc08, 0x001000e4,
+	0xc0c, 0x6c6c6c6c,
+	0xc10, 0x08800000,
+	0xc14, 0x40000100,
+	0xc18, 0x08800000,
+	0xc1c, 0x40000100,
+	0xc20, 0x00000000,
+	0xc24, 0x00000000,
+	0xc28, 0x00000000,
+	0xc2c, 0x00000000,
+	0xc30, 0x69e9ac44,
+	0xc34, 0x469652af,
+	0xc38, 0x49795994,
+	0xc3c, 0x0a979718,
+	0xc40, 0x1f7c403f,
+	0xc44, 0x000100b7,
+	0xc48, 0xec020107,
+	0xc4c, 0x007f037f,
+	0xc50, 0x69543420,
+	0xc54, 0x43bc009e,
+	0xc58, 0x69543420,
+	0xc5c, 0x433c00a8,
+	0xc60, 0x00000000,
+	0xc64, 0x7112848b,
+	0xc68, 0x47c00bff,
+	0xc6c, 0x00000036,
+	0xc70, 0x2c7f000d,
+	0xc74, 0x258610db,
+	0xc78, 0x0000001f,
+	0xc7c, 0x40b95612,
+	0xc80, 0x40000100,
+	0xc84, 0x20f60000,
+	0xc88, 0x40000100,
+	0xc8c, 0xa0e40000,
+	0xc90, 0x00121820,
+	0xc94, 0x00000007,
+	0xc98, 0x00121820,
+	0xc9c, 0x00007f7f,
+	0xca0, 0x00000000,
+	0xca4, 0x00000080,
+	0xca8, 0x00000000,
+	0xcac, 0x00000000,
+	0xcb0, 0x00000000,
+	0xcb4, 0x00000000,
+	0xcb8, 0x00000000,
+	0xcbc, 0x28000000,
+	0xcc0, 0x00000000,
+	0xcc4, 0x00000000,
+	0xcc8, 0x00000000,
+	0xccc, 0x00000000,
+	0xcd0, 0x00000000,
+	0xcd4, 0x00000000,
+	0xcd8, 0x64b11e20,
+	0xcdc, 0xe0767533,
+	0xce0, 0x00222222,
+	0xce4, 0x00000000,
+	0xce8, 0x37644302,
+	0xcec, 0x2f97d40c,
+	0xd00, 0x00080740,
+	0xd04, 0x00020403,
+	0xd08, 0x0000907f,
+	0xd0c, 0x20010201,
+	0xd10, 0xa0633333,
+	0xd14, 0x3333bc43,
+	0xd18, 0x7a8f5b6b,
+	0xd2c, 0xcc979975,
+	0xd30, 0x00000000,
+	0xd34, 0x80608404,
+	0xd38, 0x00000000,
+	0xd3c, 0x00027353,
+	0xd40, 0x00000000,
+	0xd44, 0x00000000,
+	0xd48, 0x00000000,
+	0xd4c, 0x00000000,
+	0xd50, 0x6437140a,
+	0xd54, 0x00000000,
+	0xd58, 0x00000000,
+	0xd5c, 0x30032064,
+	0xd60, 0x4653de68,
+	0xd64, 0x04518a3c,
+	0xd68, 0x00002101,
+	0xd6c, 0x2a201c16,
+	0xd70, 0x1812362e,
+	0xd74, 0x322c2220,
+	0xd78, 0x000e3c24,
+	0xe00, 0x2a2a2a2a,
+	0xe04, 0x2a2a2a2a,
+	0xe08, 0x03902a2a,
+	0xe10, 0x2a2a2a2a,
+	0xe14, 0x2a2a2a2a,
+	0xe18, 0x2a2a2a2a,
+	0xe1c, 0x2a2a2a2a,
+	0xe28, 0x00000000,
+	0xe30, 0x1000dc1f,
+	0xe34, 0x10008c1f,
+	0xe38, 0x02140102,
+	0xe3c, 0x681604c2,
+	0xe40, 0x01007c00,
+	0xe44, 0x01004800,
+	0xe48, 0xfb000000,
+	0xe4c, 0x000028d1,
+	0xe50, 0x1000dc1f,
+	0xe54, 0x10008c1f,
+	0xe58, 0x02140102,
+	0xe5c, 0x28160d05,
+	0xe60, 0x00000010,
+	0xe68, 0x001b25a4,
+	0xe6c, 0x63db25a4,
+	0xe70, 0x63db25a4,
+	0xe74, 0x0c126da4,
+	0xe78, 0x0c126da4,
+	0xe7c, 0x0c126da4,
+	0xe80, 0x0c126da4,
+	0xe84, 0x63db25a4,
+	0xe88, 0x0c126da4,
+	0xe8c, 0x63db25a4,
+	0xed0, 0x63db25a4,
+	0xed4, 0x63db25a4,
+	0xed8, 0x63db25a4,
+	0xedc, 0x001b25a4,
+	0xee0, 0x001b25a4,
+	0xeec, 0x6fdb25a4,
+	0xf14, 0x00000003,
+	0xf1c, 0x00000064,
+	0xf4c, 0x00000004,
+	0xf00, 0x00000300,
+};
+
+const u32 rtl8192du_phy_reg_array_pg[PHY_REG_ARRAY_PG_LENGTH] = {
+	0xe00, 0xffffffff, 0x07090c0c,
+	0xe04, 0xffffffff, 0x01020405,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x0b0c0c0e,
+	0xe14, 0xffffffff, 0x01030506,
+	0xe18, 0xffffffff, 0x0b0c0d0e,
+	0xe1c, 0xffffffff, 0x01030509,
+	0x830, 0xffffffff, 0x07090c0c,
+	0x834, 0xffffffff, 0x01020405,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x0b0c0c0e,
+	0x848, 0xffffffff, 0x01030506,
+	0x84c, 0xffffffff, 0x0b0c0d0e,
+	0x868, 0xffffffff, 0x01030509,
+	0xe00, 0xffffffff, 0x00000000,
+	0xe04, 0xffffffff, 0x00000000,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x00000000,
+	0xe14, 0xffffffff, 0x00000000,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x00000000,
+	0x834, 0xffffffff, 0x00000000,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x00000000,
+	0x848, 0xffffffff, 0x00000000,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+	0xe00, 0xffffffff, 0x04040404,
+	0xe04, 0xffffffff, 0x00020204,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x06060606,
+	0xe14, 0xffffffff, 0x00020406,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x04040404,
+	0x834, 0xffffffff, 0x00020204,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x06060606,
+	0x848, 0xffffffff, 0x00020406,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+	0xe00, 0xffffffff, 0x00000000,
+	0xe04, 0xffffffff, 0x00000000,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x00000000,
+	0xe14, 0xffffffff, 0x00000000,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x00000000,
+	0x834, 0xffffffff, 0x00000000,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x00000000,
+	0x848, 0xffffffff, 0x00000000,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+	0xe00, 0xffffffff, 0x00000000,
+	0xe04, 0xffffffff, 0x00000000,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x00000000,
+	0xe14, 0xffffffff, 0x00000000,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x00000000,
+	0x834, 0xffffffff, 0x00000000,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x00000000,
+	0x848, 0xffffffff, 0x00000000,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+	0xe00, 0xffffffff, 0x04040404,
+	0xe04, 0xffffffff, 0x00020204,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x00000000,
+	0xe14, 0xffffffff, 0x00000000,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x04040404,
+	0x834, 0xffffffff, 0x00020204,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x00000000,
+	0x848, 0xffffffff, 0x00000000,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+	0xe00, 0xffffffff, 0x00000000,
+	0xe04, 0xffffffff, 0x00000000,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x00000000,
+	0xe14, 0xffffffff, 0x00000000,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x00000000,
+	0x834, 0xffffffff, 0x00000000,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x00000000,
+	0x848, 0xffffffff, 0x00000000,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+	0xe00, 0xffffffff, 0x04040404,
+	0xe04, 0xffffffff, 0x00020204,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x08080808,
+	0xe14, 0xffffffff, 0x00040408,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x04040404,
+	0x834, 0xffffffff, 0x00020204,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x08080808,
+	0x848, 0xffffffff, 0x00040408,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+	0xe00, 0xffffffff, 0x04040404,
+	0xe04, 0xffffffff, 0x00020204,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x08080808,
+	0xe14, 0xffffffff, 0x00040408,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x04040404,
+	0x834, 0xffffffff, 0x00020204,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x08080808,
+	0x848, 0xffffffff, 0x00040408,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+	0xe00, 0xffffffff, 0x04040404,
+	0xe04, 0xffffffff, 0x00020204,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x08080808,
+	0xe14, 0xffffffff, 0x00040408,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x04040404,
+	0x834, 0xffffffff, 0x00020204,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x08080808,
+	0x848, 0xffffffff, 0x00040408,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+	0xe00, 0xffffffff, 0x04040404,
+	0xe04, 0xffffffff, 0x00020204,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x08080808,
+	0xe14, 0xffffffff, 0x00040408,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x04040404,
+	0x834, 0xffffffff, 0x00020204,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x08080808,
+	0x848, 0xffffffff, 0x00040408,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+	0xe00, 0xffffffff, 0x04040404,
+	0xe04, 0xffffffff, 0x00020204,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x08080808,
+	0xe14, 0xffffffff, 0x00040408,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x04040404,
+	0x834, 0xffffffff, 0x00020204,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x08080808,
+	0x848, 0xffffffff, 0x00040408,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+	0xe00, 0xffffffff, 0x04040404,
+	0xe04, 0xffffffff, 0x00020204,
+	0xe08, 0x0000ff00, 0x00000000,
+	0x86c, 0xffffff00, 0x00000000,
+	0xe10, 0xffffffff, 0x08080808,
+	0xe14, 0xffffffff, 0x00040408,
+	0xe18, 0xffffffff, 0x00000000,
+	0xe1c, 0xffffffff, 0x00000000,
+	0x830, 0xffffffff, 0x04040404,
+	0x834, 0xffffffff, 0x00020204,
+	0x838, 0xffffff00, 0x00000000,
+	0x86c, 0x000000ff, 0x00000000,
+	0x83c, 0xffffffff, 0x08080808,
+	0x848, 0xffffffff, 0x00040408,
+	0x84c, 0xffffffff, 0x00000000,
+	0x868, 0xffffffff, 0x00000000,
+};
+
+const u32 rtl8192du_radioa_2tarray[RADIOA_2T_ARRAYLENGTH] = {
+	0x000, 0x00030000,
+	0x001, 0x00030000,
+	0x002, 0x00000000,
+	0x003, 0x00018c63,
+	0x004, 0x00018c63,
+	0x008, 0x00084000,
+	0x00b, 0x0001c000,
+	0x00e, 0x00018c67,
+	0x00f, 0x00000851,
+	0x014, 0x00021440,
+	0x018, 0x00017524,
+	0x019, 0x00000000,
+	0x01d, 0x000a1290,
+	0x023, 0x00001558,
+	0x01a, 0x00030a99,
+	0x01b, 0x00040b00,
+	0x01c, 0x000fc339,
+	0x03a, 0x000a57eb,
+	0x03b, 0x00020000,
+	0x03c, 0x000ff454,
+	0x020, 0x0000aa52,
+	0x021, 0x00054000,
+	0x040, 0x0000aa52,
+	0x041, 0x00014000,
+	0x025, 0x000803be,
+	0x026, 0x000fc638,
+	0x027, 0x00077c18,
+	0x028, 0x000de471,
+	0x029, 0x000d7110,
+	0x02a, 0x0008cb04,
+	0x02b, 0x0004128b,
+	0x02c, 0x00001840,
+	0x043, 0x0002444f,
+	0x044, 0x0001adb0,
+	0x045, 0x00056467,
+	0x046, 0x0008992c,
+	0x047, 0x0000452c,
+	0x048, 0x000f9c43,
+	0x049, 0x00002e0c,
+	0x04a, 0x000546eb,
+	0x04b, 0x0008966c,
+	0x04c, 0x0000dde9,
+	0x018, 0x00007401,
+	0x000, 0x00070000,
+	0x012, 0x000dc000,
+	0x012, 0x00090000,
+	0x012, 0x00051000,
+	0x012, 0x00012000,
+	0x013, 0x000287b7,
+	0x013, 0x000247ab,
+	0x013, 0x0002079f,
+	0x013, 0x0001c793,
+	0x013, 0x0001839b,
+	0x013, 0x00014392,
+	0x013, 0x0001019a,
+	0x013, 0x0000c191,
+	0x013, 0x00008194,
+	0x013, 0x000040a0,
+	0x013, 0x00000018,
+	0x015, 0x0000f424,
+	0x015, 0x0004f424,
+	0x015, 0x0008f424,
+	0x016, 0x000e1330,
+	0x016, 0x000a1330,
+	0x016, 0x00061330,
+	0x016, 0x00021330,
+	0x018, 0x00017524,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bc,
+	0x013, 0x000247b0,
+	0x013, 0x000203b4,
+	0x013, 0x0001c3a8,
+	0x013, 0x000181b4,
+	0x013, 0x000141a8,
+	0x013, 0x000100b4,
+	0x013, 0x0000c0a8,
+	0x013, 0x0000b030,
+	0x013, 0x00004024,
+	0x013, 0x00000018,
+	0x015, 0x0000f4c3,
+	0x015, 0x0004f4c3,
+	0x015, 0x0008f4c3,
+	0x016, 0x000e085f,
+	0x016, 0x000a085f,
+	0x016, 0x0006085f,
+	0x016, 0x0002085f,
+	0x018, 0x00037524,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bc,
+	0x013, 0x000247b0,
+	0x013, 0x000203b4,
+	0x013, 0x0001c3a8,
+	0x013, 0x000181b4,
+	0x013, 0x000141a8,
+	0x013, 0x000100b4,
+	0x013, 0x0000c0a8,
+	0x013, 0x0000b030,
+	0x013, 0x00004024,
+	0x013, 0x00000018,
+	0x015, 0x0000f4c3,
+	0x015, 0x0004f4c3,
+	0x015, 0x0008f4c3,
+	0x016, 0x000e085f,
+	0x016, 0x000a085f,
+	0x016, 0x0006085f,
+	0x016, 0x0002085f,
+	0x018, 0x00057568,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bc,
+	0x013, 0x000247b0,
+	0x013, 0x000203b4,
+	0x013, 0x0001c3a8,
+	0x013, 0x000181b4,
+	0x013, 0x000141a8,
+	0x013, 0x000100b4,
+	0x013, 0x0000c0a8,
+	0x013, 0x0000b030,
+	0x013, 0x00004024,
+	0x013, 0x00000018,
+	0x015, 0x0000f4c3,
+	0x015, 0x0004f4c3,
+	0x015, 0x0008f4c3,
+	0x016, 0x000e085f,
+	0x016, 0x000a085f,
+	0x016, 0x0006085f,
+	0x016, 0x0002085f,
+	0x030, 0x0004470f,
+	0x031, 0x00044ff0,
+	0x032, 0x00000070,
+	0x033, 0x000dd480,
+	0x034, 0x000ffac0,
+	0x035, 0x000b80c0,
+	0x036, 0x00077000,
+	0x037, 0x00064ff2,
+	0x038, 0x000e7661,
+	0x039, 0x00000e90,
+	0x000, 0x00030000,
+	0x018, 0x0000f401,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x01e, 0x00088009,
+	0x01f, 0x00080003,
+	0x0fe, 0x00000000,
+	0x01e, 0x00088001,
+	0x01f, 0x00080000,
+	0x0fe, 0x00000000,
+	0x018, 0x00097524,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x02b, 0x00041289,
+	0x0fe, 0x00000000,
+	0x02d, 0x0006aaaa,
+	0x02e, 0x000b4d01,
+	0x02d, 0x00080000,
+	0x02e, 0x00004d02,
+	0x02d, 0x00095555,
+	0x02e, 0x00054d03,
+	0x02d, 0x000aaaaa,
+	0x02e, 0x000b4d04,
+	0x02d, 0x000c0000,
+	0x02e, 0x00004d05,
+	0x02d, 0x000d5555,
+	0x02e, 0x00054d06,
+	0x02d, 0x000eaaaa,
+	0x02e, 0x000b4d07,
+	0x02d, 0x00000000,
+	0x02e, 0x00005108,
+	0x02d, 0x00015555,
+	0x02e, 0x00055109,
+	0x02d, 0x0002aaaa,
+	0x02e, 0x000b510a,
+	0x02d, 0x00040000,
+	0x02e, 0x0000510b,
+	0x02d, 0x00055555,
+	0x02e, 0x0005510c,
+};
+
+const u32 rtl8192du_radiob_2tarray[RADIOB_2T_ARRAYLENGTH] = {
+	0x000, 0x00030000,
+	0x001, 0x00030000,
+	0x002, 0x00000000,
+	0x003, 0x00018c63,
+	0x004, 0x00018c63,
+	0x008, 0x00084000,
+	0x00b, 0x0001c000,
+	0x00e, 0x00018c67,
+	0x00f, 0x00000851,
+	0x014, 0x00021440,
+	0x018, 0x00007401,
+	0x019, 0x00000060,
+	0x01d, 0x000a1290,
+	0x023, 0x00001558,
+	0x01a, 0x00030a99,
+	0x01b, 0x00040b00,
+	0x01c, 0x000fc339,
+	0x03a, 0x000a57eb,
+	0x03b, 0x00020000,
+	0x03c, 0x000ff454,
+	0x020, 0x0000aa52,
+	0x021, 0x00054000,
+	0x040, 0x0000aa52,
+	0x041, 0x00014000,
+	0x025, 0x000803be,
+	0x026, 0x000fc638,
+	0x027, 0x00077c18,
+	0x028, 0x000d1c31,
+	0x029, 0x000d7110,
+	0x02a, 0x000aeb04,
+	0x02b, 0x0004128b,
+	0x02c, 0x00001840,
+	0x043, 0x0002444f,
+	0x044, 0x0001adb0,
+	0x045, 0x00056467,
+	0x046, 0x0008992c,
+	0x047, 0x0000452c,
+	0x048, 0x000f9c43,
+	0x049, 0x00002e0c,
+	0x04a, 0x000546eb,
+	0x04b, 0x0008966c,
+	0x04c, 0x0000dde9,
+	0x018, 0x00007401,
+	0x000, 0x00070000,
+	0x012, 0x000dc000,
+	0x012, 0x00090000,
+	0x012, 0x00051000,
+	0x012, 0x00012000,
+	0x013, 0x000287b7,
+	0x013, 0x000247ab,
+	0x013, 0x0002079f,
+	0x013, 0x0001c793,
+	0x013, 0x0001839b,
+	0x013, 0x00014392,
+	0x013, 0x0001019a,
+	0x013, 0x0000c191,
+	0x013, 0x00008194,
+	0x013, 0x000040a0,
+	0x013, 0x00000018,
+	0x015, 0x0000f424,
+	0x015, 0x0004f424,
+	0x015, 0x0008f424,
+	0x016, 0x000e1330,
+	0x016, 0x000a1330,
+	0x016, 0x00061330,
+	0x016, 0x00021330,
+	0x018, 0x00017524,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bc,
+	0x013, 0x000247b0,
+	0x013, 0x000203b4,
+	0x013, 0x0001c3a8,
+	0x013, 0x000181b4,
+	0x013, 0x000141a8,
+	0x013, 0x000100b4,
+	0x013, 0x0000c0a8,
+	0x013, 0x0000b030,
+	0x013, 0x00004024,
+	0x013, 0x00000018,
+	0x015, 0x0000f4c3,
+	0x015, 0x0004f4c3,
+	0x015, 0x0008f4c3,
+	0x016, 0x000e085f,
+	0x016, 0x000a085f,
+	0x016, 0x0006085f,
+	0x016, 0x0002085f,
+	0x018, 0x00037524,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bc,
+	0x013, 0x000247b0,
+	0x013, 0x000203b4,
+	0x013, 0x0001c3a8,
+	0x013, 0x000181b4,
+	0x013, 0x000141a8,
+	0x013, 0x000100b4,
+	0x013, 0x0000c0a8,
+	0x013, 0x0000b030,
+	0x013, 0x00004024,
+	0x013, 0x00000018,
+	0x015, 0x0000f4c3,
+	0x015, 0x0004f4c3,
+	0x015, 0x0008f4c3,
+	0x016, 0x000e085f,
+	0x016, 0x000a085f,
+	0x016, 0x0006085f,
+	0x016, 0x0002085f,
+	0x018, 0x00057524,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bc,
+	0x013, 0x000247b0,
+	0x013, 0x000203b4,
+	0x013, 0x0001c3a8,
+	0x013, 0x000181b4,
+	0x013, 0x000141a8,
+	0x013, 0x000100b4,
+	0x013, 0x0000c0a8,
+	0x013, 0x0000b030,
+	0x013, 0x00004024,
+	0x013, 0x00000018,
+	0x015, 0x0000f4c3,
+	0x015, 0x0004f4c3,
+	0x015, 0x0008f4c3,
+	0x016, 0x000e085f,
+	0x016, 0x000a085f,
+	0x016, 0x0006085f,
+	0x016, 0x0002085f,
+	0x030, 0x0004470f,
+	0x031, 0x00044ff0,
+	0x032, 0x00000070,
+	0x033, 0x000dd480,
+	0x034, 0x000ffac0,
+	0x035, 0x000b80c0,
+	0x036, 0x00077000,
+	0x037, 0x00064ff2,
+	0x038, 0x000e7661,
+	0x039, 0x00000e90,
+	0x000, 0x00030000,
+	0x018, 0x0000f401,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x01e, 0x00088009,
+	0x01f, 0x00080003,
+	0x0fe, 0x00000000,
+	0x01e, 0x00088001,
+	0x01f, 0x00080000,
+	0x0fe, 0x00000000,
+	0x018, 0x00087401,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x02b, 0x00041289,
+	0x0fe, 0x00000000,
+	0x02d, 0x00066666,
+	0x02e, 0x00064001,
+	0x02d, 0x00091111,
+	0x02e, 0x00014002,
+	0x02d, 0x000bbbbb,
+	0x02e, 0x000b4003,
+	0x02d, 0x000e6666,
+	0x02e, 0x00064004,
+	0x02d, 0x00088888,
+	0x02e, 0x00084005,
+	0x02d, 0x0009dddd,
+	0x02e, 0x000d4006,
+	0x02d, 0x000b3333,
+	0x02e, 0x00034007,
+	0x02d, 0x00048888,
+	0x02e, 0x00084408,
+	0x02d, 0x000bbbbb,
+	0x02e, 0x000b4409,
+	0x02d, 0x000e6666,
+	0x02e, 0x0006440a,
+	0x02d, 0x00011111,
+	0x02e, 0x0001480b,
+	0x02d, 0x0003bbbb,
+	0x02e, 0x000b480c,
+	0x02d, 0x00066666,
+	0x02e, 0x0006480d,
+	0x02d, 0x000ccccc,
+	0x02e, 0x000c480e,
+};
+
+const u32 rtl8192du_radioa_2t_int_paarray[RADIOA_2T_INT_PA_ARRAYLENGTH] = {
+	0x000, 0x00030000,
+	0x001, 0x00030000,
+	0x002, 0x00000000,
+	0x003, 0x00018c63,
+	0x004, 0x00018c63,
+	0x008, 0x00084000,
+	0x00b, 0x0001c000,
+	0x00e, 0x00018c67,
+	0x00f, 0x00000851,
+	0x014, 0x00021440,
+	0x018, 0x00017524,
+	0x019, 0x00000000,
+	0x01d, 0x000a1290,
+	0x023, 0x00001558,
+	0x01a, 0x00030a99,
+	0x01b, 0x00040b00,
+	0x01c, 0x000fc339,
+	0x03a, 0x000a57eb,
+	0x03b, 0x00020000,
+	0x03c, 0x000ff455,
+	0x020, 0x0000aa52,
+	0x021, 0x00054000,
+	0x040, 0x0000aa52,
+	0x041, 0x00014000,
+	0x025, 0x000803be,
+	0x026, 0x000fc638,
+	0x027, 0x00077c18,
+	0x028, 0x000de471,
+	0x029, 0x000d7110,
+	0x02a, 0x0008eb04,
+	0x02b, 0x0004128b,
+	0x02c, 0x00001840,
+	0x043, 0x0002444f,
+	0x044, 0x0001adb0,
+	0x045, 0x00056467,
+	0x046, 0x0008992c,
+	0x047, 0x0000452c,
+	0x048, 0x000c0443,
+	0x049, 0x00000730,
+	0x04a, 0x00050f0f,
+	0x04b, 0x000896ef,
+	0x04c, 0x0000ddee,
+	0x018, 0x00007401,
+	0x000, 0x00070000,
+	0x012, 0x000dc000,
+	0x012, 0x00090000,
+	0x012, 0x00051000,
+	0x012, 0x00012000,
+	0x013, 0x000287b7,
+	0x013, 0x000247ab,
+	0x013, 0x0002079f,
+	0x013, 0x0001c793,
+	0x013, 0x0001839b,
+	0x013, 0x00014392,
+	0x013, 0x0001019a,
+	0x013, 0x0000c191,
+	0x013, 0x00008194,
+	0x013, 0x000040a0,
+	0x013, 0x00000018,
+	0x015, 0x0000f424,
+	0x015, 0x0004f424,
+	0x015, 0x0008f424,
+	0x016, 0x000e1330,
+	0x016, 0x000a1330,
+	0x016, 0x00061330,
+	0x016, 0x00021330,
+	0x018, 0x00017524,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bf,
+	0x013, 0x000247b3,
+	0x013, 0x000207a7,
+	0x013, 0x0001c79b,
+	0x013, 0x0001839f,
+	0x013, 0x00014393,
+	0x013, 0x00010399,
+	0x013, 0x0000c38d,
+	0x013, 0x00008199,
+	0x013, 0x0000418d,
+	0x013, 0x00000099,
+	0x015, 0x0000f495,
+	0x015, 0x0004f495,
+	0x015, 0x0008f495,
+	0x016, 0x000e1874,
+	0x016, 0x000a1874,
+	0x016, 0x00061874,
+	0x016, 0x00021874,
+	0x018, 0x00037564,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bf,
+	0x013, 0x000247b3,
+	0x013, 0x000207a7,
+	0x013, 0x0001c79b,
+	0x013, 0x0001839f,
+	0x013, 0x00014393,
+	0x013, 0x00010399,
+	0x013, 0x0000c38d,
+	0x013, 0x00008199,
+	0x013, 0x0000418d,
+	0x013, 0x00000099,
+	0x015, 0x0000f495,
+	0x015, 0x0004f495,
+	0x015, 0x0008f495,
+	0x016, 0x000e1874,
+	0x016, 0x000a1874,
+	0x016, 0x00061874,
+	0x016, 0x00021874,
+	0x018, 0x00057595,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bf,
+	0x013, 0x000247b3,
+	0x013, 0x000207a7,
+	0x013, 0x0001c79b,
+	0x013, 0x0001839f,
+	0x013, 0x00014393,
+	0x013, 0x00010399,
+	0x013, 0x0000c38d,
+	0x013, 0x00008199,
+	0x013, 0x0000418d,
+	0x013, 0x00000099,
+	0x015, 0x0000f495,
+	0x015, 0x0004f495,
+	0x015, 0x0008f495,
+	0x016, 0x000e1874,
+	0x016, 0x000a1874,
+	0x016, 0x00061874,
+	0x016, 0x00021874,
+	0x030, 0x0004470f,
+	0x031, 0x00044ff0,
+	0x032, 0x00000070,
+	0x033, 0x000dd480,
+	0x034, 0x000ffac0,
+	0x035, 0x000b80c0,
+	0x036, 0x00077000,
+	0x037, 0x00064ff2,
+	0x038, 0x000e7661,
+	0x039, 0x00000e90,
+	0x000, 0x00030000,
+	0x018, 0x0000f401,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x01e, 0x00088009,
+	0x01f, 0x00080003,
+	0x0fe, 0x00000000,
+	0x01e, 0x00088001,
+	0x01f, 0x00080000,
+	0x0fe, 0x00000000,
+	0x018, 0x00097524,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x02b, 0x00041289,
+	0x0fe, 0x00000000,
+	0x02d, 0x0006aaaa,
+	0x02e, 0x000b4d01,
+	0x02d, 0x00080000,
+	0x02e, 0x00004d02,
+	0x02d, 0x00095555,
+	0x02e, 0x00054d03,
+	0x02d, 0x000aaaaa,
+	0x02e, 0x000b4d04,
+	0x02d, 0x000c0000,
+	0x02e, 0x00004d05,
+	0x02d, 0x000d5555,
+	0x02e, 0x00054d06,
+	0x02d, 0x000eaaaa,
+	0x02e, 0x000b4d07,
+	0x02d, 0x00000000,
+	0x02e, 0x00005108,
+	0x02d, 0x00015555,
+	0x02e, 0x00055109,
+	0x02d, 0x0002aaaa,
+	0x02e, 0x000b510a,
+	0x02d, 0x00040000,
+	0x02e, 0x0000510b,
+	0x02d, 0x00055555,
+	0x02e, 0x0005510c,
+};
+
+const u32 rtl8192du_radiob_2t_int_paarray[RADIOB_2T_INT_PA_ARRAYLENGTH] = {
+	0x000, 0x00030000,
+	0x001, 0x00030000,
+	0x002, 0x00000000,
+	0x003, 0x00018c63,
+	0x004, 0x00018c63,
+	0x008, 0x00084000,
+	0x00b, 0x0001c000,
+	0x00e, 0x00018c67,
+	0x00f, 0x00000851,
+	0x014, 0x00021440,
+	0x018, 0x00007401,
+	0x019, 0x00000060,
+	0x01d, 0x000a1290,
+	0x023, 0x00001558,
+	0x01a, 0x00030a99,
+	0x01b, 0x00040b00,
+	0x01c, 0x000fc339,
+	0x03a, 0x000a57eb,
+	0x03b, 0x00020000,
+	0x03c, 0x000ff455,
+	0x020, 0x0000aa52,
+	0x021, 0x00054000,
+	0x040, 0x0000aa52,
+	0x041, 0x00014000,
+	0x025, 0x000803be,
+	0x026, 0x000fc638,
+	0x027, 0x00077c18,
+	0x028, 0x000d1c31,
+	0x029, 0x000d7110,
+	0x02a, 0x000aeb04,
+	0x02b, 0x0004128b,
+	0x02c, 0x00001840,
+	0x043, 0x0002444f,
+	0x044, 0x0001adb0,
+	0x045, 0x00056467,
+	0x046, 0x0008992c,
+	0x047, 0x0000452c,
+	0x048, 0x000c0443,
+	0x049, 0x00000730,
+	0x04a, 0x00050f0f,
+	0x04b, 0x000896ef,
+	0x04c, 0x0000ddee,
+	0x018, 0x00007401,
+	0x000, 0x00070000,
+	0x012, 0x000dc000,
+	0x012, 0x00090000,
+	0x012, 0x00051000,
+	0x012, 0x00012000,
+	0x013, 0x000287b7,
+	0x013, 0x000247ab,
+	0x013, 0x0002079f,
+	0x013, 0x0001c793,
+	0x013, 0x0001839b,
+	0x013, 0x00014392,
+	0x013, 0x0001019a,
+	0x013, 0x0000c191,
+	0x013, 0x00008194,
+	0x013, 0x000040a0,
+	0x013, 0x00000018,
+	0x015, 0x0000f424,
+	0x015, 0x0004f424,
+	0x015, 0x0008f424,
+	0x016, 0x000e1330,
+	0x016, 0x000a1330,
+	0x016, 0x00061330,
+	0x016, 0x00021330,
+	0x018, 0x00017524,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bf,
+	0x013, 0x000247b3,
+	0x013, 0x000207a7,
+	0x013, 0x0001c79b,
+	0x013, 0x0001839f,
+	0x013, 0x00014393,
+	0x013, 0x00010399,
+	0x013, 0x0000c38d,
+	0x013, 0x00008199,
+	0x013, 0x0000418d,
+	0x013, 0x00000099,
+	0x015, 0x0000f495,
+	0x015, 0x0004f495,
+	0x015, 0x0008f495,
+	0x016, 0x000e1874,
+	0x016, 0x000a1874,
+	0x016, 0x00061874,
+	0x016, 0x00021874,
+	0x018, 0x00037564,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bf,
+	0x013, 0x000247b3,
+	0x013, 0x000207a7,
+	0x013, 0x0001c79b,
+	0x013, 0x0001839f,
+	0x013, 0x00014393,
+	0x013, 0x00010399,
+	0x013, 0x0000c38d,
+	0x013, 0x00008199,
+	0x013, 0x0000418d,
+	0x013, 0x00000099,
+	0x015, 0x0000f495,
+	0x015, 0x0004f495,
+	0x015, 0x0008f495,
+	0x016, 0x000e1874,
+	0x016, 0x000a1874,
+	0x016, 0x00061874,
+	0x016, 0x00021874,
+	0x018, 0x00057595,
+	0x000, 0x00070000,
+	0x012, 0x000cf000,
+	0x012, 0x000bc000,
+	0x012, 0x00078000,
+	0x012, 0x00000000,
+	0x013, 0x000287bf,
+	0x013, 0x000247b3,
+	0x013, 0x000207a7,
+	0x013, 0x0001c79b,
+	0x013, 0x0001839f,
+	0x013, 0x00014393,
+	0x013, 0x00010399,
+	0x013, 0x0000c38d,
+	0x013, 0x00008199,
+	0x013, 0x0000418d,
+	0x013, 0x00000099,
+	0x015, 0x0000f495,
+	0x015, 0x0004f495,
+	0x015, 0x0008f495,
+	0x016, 0x000e1874,
+	0x016, 0x000a1874,
+	0x016, 0x00061874,
+	0x016, 0x00021874,
+	0x030, 0x0004470f,
+	0x031, 0x00044ff0,
+	0x032, 0x00000070,
+	0x033, 0x000dd480,
+	0x034, 0x000ffac0,
+	0x035, 0x000b80c0,
+	0x036, 0x00077000,
+	0x037, 0x00064ff2,
+	0x038, 0x000e7661,
+	0x039, 0x00000e90,
+	0x000, 0x00030000,
+	0x018, 0x0000f401,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x01e, 0x00088009,
+	0x01f, 0x00080003,
+	0x0fe, 0x00000000,
+	0x01e, 0x00088001,
+	0x01f, 0x00080000,
+	0x0fe, 0x00000000,
+	0x018, 0x00087401,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x0fe, 0x00000000,
+	0x02b, 0x00041289,
+	0x0fe, 0x00000000,
+	0x02d, 0x00066666,
+	0x02e, 0x00064001,
+	0x02d, 0x00091111,
+	0x02e, 0x00014002,
+	0x02d, 0x000bbbbb,
+	0x02e, 0x000b4003,
+	0x02d, 0x000e6666,
+	0x02e, 0x00064004,
+	0x02d, 0x00088888,
+	0x02e, 0x00084005,
+	0x02d, 0x0009dddd,
+	0x02e, 0x000d4006,
+	0x02d, 0x000b3333,
+	0x02e, 0x00034007,
+	0x02d, 0x00048888,
+	0x02e, 0x00084408,
+	0x02d, 0x000bbbbb,
+	0x02e, 0x000b4409,
+	0x02d, 0x000e6666,
+	0x02e, 0x0006440a,
+	0x02d, 0x00011111,
+	0x02e, 0x0001480b,
+	0x02d, 0x0003bbbb,
+	0x02e, 0x000b480c,
+	0x02d, 0x00066666,
+	0x02e, 0x0006480d,
+	0x02d, 0x000ccccc,
+	0x02e, 0x000c480e,
+};
+
+const u32 rtl8192du_mac_2tarray[MAC_2T_ARRAYLENGTH] = {
+	0x420, 0x00000080,
+	0x423, 0x00000000,
+	0x430, 0x00000000,
+	0x431, 0x00000000,
+	0x432, 0x00000000,
+	0x433, 0x00000001,
+	0x434, 0x00000004,
+	0x435, 0x00000005,
+	0x436, 0x00000006,
+	0x437, 0x00000007,
+	0x438, 0x00000000,
+	0x439, 0x00000000,
+	0x43a, 0x00000000,
+	0x43b, 0x00000001,
+	0x43c, 0x00000004,
+	0x43d, 0x00000005,
+	0x43e, 0x00000006,
+	0x43f, 0x00000007,
+	0x440, 0x00000050,
+	0x441, 0x00000001,
+	0x442, 0x00000000,
+	0x444, 0x00000015,
+	0x445, 0x000000f0,
+	0x446, 0x0000000f,
+	0x447, 0x00000000,
+	0x462, 0x00000008,
+	0x463, 0x00000003,
+	0x4c8, 0x000000ff,
+	0x4c9, 0x00000008,
+	0x4cc, 0x000000ff,
+	0x4cd, 0x000000ff,
+	0x4ce, 0x00000001,
+	0x500, 0x00000026,
+	0x501, 0x000000a2,
+	0x502, 0x0000002f,
+	0x503, 0x00000000,
+	0x504, 0x00000028,
+	0x505, 0x000000a3,
+	0x506, 0x0000005e,
+	0x507, 0x00000000,
+	0x508, 0x0000002b,
+	0x509, 0x000000a4,
+	0x50a, 0x0000005e,
+	0x50b, 0x00000000,
+	0x50c, 0x0000004f,
+	0x50d, 0x000000a4,
+	0x50e, 0x00000000,
+	0x50f, 0x00000000,
+	0x512, 0x0000001c,
+	0x514, 0x0000000a,
+	0x515, 0x00000010,
+	0x516, 0x0000000a,
+	0x517, 0x00000010,
+	0x51a, 0x00000016,
+	0x524, 0x0000000f,
+	0x525, 0x0000004f,
+	0x546, 0x00000040,
+	0x547, 0x00000000,
+	0x550, 0x00000010,
+	0x551, 0x00000010,
+	0x559, 0x00000002,
+	0x55a, 0x00000002,
+	0x55d, 0x000000ff,
+	0x605, 0x00000080,
+	0x608, 0x0000000e,
+	0x609, 0x0000002a,
+	0x652, 0x00000020,
+	0x63c, 0x0000000a,
+	0x63d, 0x0000000a,
+	0x63e, 0x0000000e,
+	0x63f, 0x0000000e,
+	0x66e, 0x00000005,
+	0x700, 0x00000021,
+	0x701, 0x00000043,
+	0x702, 0x00000065,
+	0x703, 0x00000087,
+	0x708, 0x00000021,
+	0x709, 0x00000043,
+	0x70a, 0x00000065,
+	0x70b, 0x00000087,
+	0x024, 0x0000000d,
+	0x025, 0x00000080,
+	0x026, 0x00000011,
+	0x027, 0x00000000,
+	0x028, 0x00000083,
+	0x029, 0x000000db,
+	0x02a, 0x000000ff,
+	0x02b, 0x00000000,
+	0x014, 0x00000055,
+	0x015, 0x000000a9,
+	0x016, 0x0000008b,
+	0x017, 0x00000008,
+	0x010, 0x00000003,
+	0x011, 0x0000002b,
+	0x012, 0x00000002,
+	0x013, 0x00000049,
+};
+
+const u32 rtl8192du_agctab_array[AGCTAB_ARRAYLENGTH] = {
+	0xc78, 0x7b000001,
+	0xc78, 0x7b010001,
+	0xc78, 0x7b020001,
+	0xc78, 0x7b030001,
+	0xc78, 0x7b040001,
+	0xc78, 0x7b050001,
+	0xc78, 0x7b060001,
+	0xc78, 0x7a070001,
+	0xc78, 0x79080001,
+	0xc78, 0x78090001,
+	0xc78, 0x770a0001,
+	0xc78, 0x760b0001,
+	0xc78, 0x750c0001,
+	0xc78, 0x740d0001,
+	0xc78, 0x730e0001,
+	0xc78, 0x720f0001,
+	0xc78, 0x71100001,
+	0xc78, 0x70110001,
+	0xc78, 0x6f120001,
+	0xc78, 0x6e130001,
+	0xc78, 0x6d140001,
+	0xc78, 0x6c150001,
+	0xc78, 0x6b160001,
+	0xc78, 0x6a170001,
+	0xc78, 0x69180001,
+	0xc78, 0x68190001,
+	0xc78, 0x671a0001,
+	0xc78, 0x661b0001,
+	0xc78, 0x651c0001,
+	0xc78, 0x641d0001,
+	0xc78, 0x631e0001,
+	0xc78, 0x621f0001,
+	0xc78, 0x61200001,
+	0xc78, 0x60210001,
+	0xc78, 0x49220001,
+	0xc78, 0x48230001,
+	0xc78, 0x47240001,
+	0xc78, 0x46250001,
+	0xc78, 0x45260001,
+	0xc78, 0x44270001,
+	0xc78, 0x43280001,
+	0xc78, 0x42290001,
+	0xc78, 0x412a0001,
+	0xc78, 0x402b0001,
+	0xc78, 0x262c0001,
+	0xc78, 0x252d0001,
+	0xc78, 0x242e0001,
+	0xc78, 0x232f0001,
+	0xc78, 0x22300001,
+	0xc78, 0x21310001,
+	0xc78, 0x20320001,
+	0xc78, 0x06330001,
+	0xc78, 0x05340001,
+	0xc78, 0x04350001,
+	0xc78, 0x03360001,
+	0xc78, 0x02370001,
+	0xc78, 0x01380001,
+	0xc78, 0x00390001,
+	0xc78, 0x003a0001,
+	0xc78, 0x003b0001,
+	0xc78, 0x003c0001,
+	0xc78, 0x003d0001,
+	0xc78, 0x003e0001,
+	0xc78, 0x003f0001,
+	0xc78, 0x7b400001,
+	0xc78, 0x7b410001,
+	0xc78, 0x7a420001,
+	0xc78, 0x79430001,
+	0xc78, 0x78440001,
+	0xc78, 0x77450001,
+	0xc78, 0x76460001,
+	0xc78, 0x75470001,
+	0xc78, 0x74480001,
+	0xc78, 0x73490001,
+	0xc78, 0x724a0001,
+	0xc78, 0x714b0001,
+	0xc78, 0x704c0001,
+	0xc78, 0x6f4d0001,
+	0xc78, 0x6e4e0001,
+	0xc78, 0x6d4f0001,
+	0xc78, 0x6c500001,
+	0xc78, 0x6b510001,
+	0xc78, 0x6a520001,
+	0xc78, 0x69530001,
+	0xc78, 0x68540001,
+	0xc78, 0x67550001,
+	0xc78, 0x66560001,
+	0xc78, 0x65570001,
+	0xc78, 0x64580001,
+	0xc78, 0x63590001,
+	0xc78, 0x625a0001,
+	0xc78, 0x615b0001,
+	0xc78, 0x605c0001,
+	0xc78, 0x485d0001,
+	0xc78, 0x475e0001,
+	0xc78, 0x465f0001,
+	0xc78, 0x45600001,
+	0xc78, 0x44610001,
+	0xc78, 0x43620001,
+	0xc78, 0x42630001,
+	0xc78, 0x41640001,
+	0xc78, 0x40650001,
+	0xc78, 0x27660001,
+	0xc78, 0x26670001,
+	0xc78, 0x25680001,
+	0xc78, 0x24690001,
+	0xc78, 0x236a0001,
+	0xc78, 0x226b0001,
+	0xc78, 0x216c0001,
+	0xc78, 0x206d0001,
+	0xc78, 0x206e0001,
+	0xc78, 0x206f0001,
+	0xc78, 0x20700001,
+	0xc78, 0x20710001,
+	0xc78, 0x20720001,
+	0xc78, 0x20730001,
+	0xc78, 0x20740001,
+	0xc78, 0x20750001,
+	0xc78, 0x20760001,
+	0xc78, 0x20770001,
+	0xc78, 0x20780001,
+	0xc78, 0x20790001,
+	0xc78, 0x207a0001,
+	0xc78, 0x207b0001,
+	0xc78, 0x207c0001,
+	0xc78, 0x207d0001,
+	0xc78, 0x207e0001,
+	0xc78, 0x207f0001,
+	0xc78, 0x38000002,
+	0xc78, 0x38010002,
+	0xc78, 0x38020002,
+	0xc78, 0x38030002,
+	0xc78, 0x38040002,
+	0xc78, 0x38050002,
+	0xc78, 0x38060002,
+	0xc78, 0x38070002,
+	0xc78, 0x38080002,
+	0xc78, 0x3c090002,
+	0xc78, 0x3e0a0002,
+	0xc78, 0x400b0002,
+	0xc78, 0x440c0002,
+	0xc78, 0x480d0002,
+	0xc78, 0x4c0e0002,
+	0xc78, 0x500f0002,
+	0xc78, 0x52100002,
+	0xc78, 0x56110002,
+	0xc78, 0x5a120002,
+	0xc78, 0x5e130002,
+	0xc78, 0x60140002,
+	0xc78, 0x60150002,
+	0xc78, 0x60160002,
+	0xc78, 0x62170002,
+	0xc78, 0x62180002,
+	0xc78, 0x62190002,
+	0xc78, 0x621a0002,
+	0xc78, 0x621b0002,
+	0xc78, 0x621c0002,
+	0xc78, 0x621d0002,
+	0xc78, 0x621e0002,
+	0xc78, 0x621f0002,
+	0xc78, 0x32000044,
+	0xc78, 0x32010044,
+	0xc78, 0x32020044,
+	0xc78, 0x32030044,
+	0xc78, 0x32040044,
+	0xc78, 0x32050044,
+	0xc78, 0x32060044,
+	0xc78, 0x34070044,
+	0xc78, 0x35080044,
+	0xc78, 0x36090044,
+	0xc78, 0x370a0044,
+	0xc78, 0x380b0044,
+	0xc78, 0x390c0044,
+	0xc78, 0x3a0d0044,
+	0xc78, 0x3e0e0044,
+	0xc78, 0x420f0044,
+	0xc78, 0x44100044,
+	0xc78, 0x46110044,
+	0xc78, 0x4a120044,
+	0xc78, 0x4e130044,
+	0xc78, 0x50140044,
+	0xc78, 0x55150044,
+	0xc78, 0x5a160044,
+	0xc78, 0x5e170044,
+	0xc78, 0x64180044,
+	0xc78, 0x6e190044,
+	0xc78, 0x6e1a0044,
+	0xc78, 0x6e1b0044,
+	0xc78, 0x6e1c0044,
+	0xc78, 0x6e1d0044,
+	0xc78, 0x6e1e0044,
+	0xc78, 0x6e1f0044,
+	0xc78, 0x6e1f0000,
+};
+
+const u32 rtl8192du_agctab_5garray[AGCTAB_5G_ARRAYLENGTH] = {
+	0xc78, 0x7b000001,
+	0xc78, 0x7b010001,
+	0xc78, 0x7a020001,
+	0xc78, 0x79030001,
+	0xc78, 0x78040001,
+	0xc78, 0x77050001,
+	0xc78, 0x76060001,
+	0xc78, 0x75070001,
+	0xc78, 0x74080001,
+	0xc78, 0x73090001,
+	0xc78, 0x720a0001,
+	0xc78, 0x710b0001,
+	0xc78, 0x700c0001,
+	0xc78, 0x6f0d0001,
+	0xc78, 0x6e0e0001,
+	0xc78, 0x6d0f0001,
+	0xc78, 0x6c100001,
+	0xc78, 0x6b110001,
+	0xc78, 0x6a120001,
+	0xc78, 0x69130001,
+	0xc78, 0x68140001,
+	0xc78, 0x67150001,
+	0xc78, 0x66160001,
+	0xc78, 0x65170001,
+	0xc78, 0x64180001,
+	0xc78, 0x63190001,
+	0xc78, 0x621a0001,
+	0xc78, 0x611b0001,
+	0xc78, 0x601c0001,
+	0xc78, 0x481d0001,
+	0xc78, 0x471e0001,
+	0xc78, 0x461f0001,
+	0xc78, 0x45200001,
+	0xc78, 0x44210001,
+	0xc78, 0x43220001,
+	0xc78, 0x42230001,
+	0xc78, 0x41240001,
+	0xc78, 0x40250001,
+	0xc78, 0x27260001,
+	0xc78, 0x26270001,
+	0xc78, 0x25280001,
+	0xc78, 0x24290001,
+	0xc78, 0x232a0001,
+	0xc78, 0x222b0001,
+	0xc78, 0x212c0001,
+	0xc78, 0x202d0001,
+	0xc78, 0x202e0001,
+	0xc78, 0x202f0001,
+	0xc78, 0x20300001,
+	0xc78, 0x20310001,
+	0xc78, 0x20320001,
+	0xc78, 0x20330001,
+	0xc78, 0x20340001,
+	0xc78, 0x20350001,
+	0xc78, 0x20360001,
+	0xc78, 0x20370001,
+	0xc78, 0x20380001,
+	0xc78, 0x20390001,
+	0xc78, 0x203a0001,
+	0xc78, 0x203b0001,
+	0xc78, 0x203c0001,
+	0xc78, 0x203d0001,
+	0xc78, 0x203e0001,
+	0xc78, 0x203f0001,
+	0xc78, 0x32000044,
+	0xc78, 0x32010044,
+	0xc78, 0x32020044,
+	0xc78, 0x32030044,
+	0xc78, 0x32040044,
+	0xc78, 0x32050044,
+	0xc78, 0x32060044,
+	0xc78, 0x34070044,
+	0xc78, 0x35080044,
+	0xc78, 0x36090044,
+	0xc78, 0x370a0044,
+	0xc78, 0x380b0044,
+	0xc78, 0x390c0044,
+	0xc78, 0x3a0d0044,
+	0xc78, 0x3e0e0044,
+	0xc78, 0x420f0044,
+	0xc78, 0x44100044,
+	0xc78, 0x46110044,
+	0xc78, 0x4a120044,
+	0xc78, 0x4e130044,
+	0xc78, 0x50140044,
+	0xc78, 0x55150044,
+	0xc78, 0x5a160044,
+	0xc78, 0x5e170044,
+	0xc78, 0x64180044,
+	0xc78, 0x6e190044,
+	0xc78, 0x6e1a0044,
+	0xc78, 0x6e1b0044,
+	0xc78, 0x6e1c0044,
+	0xc78, 0x6e1d0044,
+	0xc78, 0x6e1e0044,
+	0xc78, 0x6e1f0044,
+	0xc78, 0x6e1f0000,
+};
+
+const u32 rtl8192du_agctab_2garray[AGCTAB_2G_ARRAYLENGTH] = {
+	0xc78, 0x7b000001,
+	0xc78, 0x7b010001,
+	0xc78, 0x7b020001,
+	0xc78, 0x7b030001,
+	0xc78, 0x7b040001,
+	0xc78, 0x7b050001,
+	0xc78, 0x7b060001,
+	0xc78, 0x7a070001,
+	0xc78, 0x79080001,
+	0xc78, 0x78090001,
+	0xc78, 0x770a0001,
+	0xc78, 0x760b0001,
+	0xc78, 0x750c0001,
+	0xc78, 0x740d0001,
+	0xc78, 0x730e0001,
+	0xc78, 0x720f0001,
+	0xc78, 0x71100001,
+	0xc78, 0x70110001,
+	0xc78, 0x6f120001,
+	0xc78, 0x6e130001,
+	0xc78, 0x6d140001,
+	0xc78, 0x6c150001,
+	0xc78, 0x6b160001,
+	0xc78, 0x6a170001,
+	0xc78, 0x69180001,
+	0xc78, 0x68190001,
+	0xc78, 0x671a0001,
+	0xc78, 0x661b0001,
+	0xc78, 0x651c0001,
+	0xc78, 0x641d0001,
+	0xc78, 0x631e0001,
+	0xc78, 0x621f0001,
+	0xc78, 0x61200001,
+	0xc78, 0x60210001,
+	0xc78, 0x49220001,
+	0xc78, 0x48230001,
+	0xc78, 0x47240001,
+	0xc78, 0x46250001,
+	0xc78, 0x45260001,
+	0xc78, 0x44270001,
+	0xc78, 0x43280001,
+	0xc78, 0x42290001,
+	0xc78, 0x412a0001,
+	0xc78, 0x402b0001,
+	0xc78, 0x262c0001,
+	0xc78, 0x252d0001,
+	0xc78, 0x242e0001,
+	0xc78, 0x232f0001,
+	0xc78, 0x22300001,
+	0xc78, 0x21310001,
+	0xc78, 0x20320001,
+	0xc78, 0x06330001,
+	0xc78, 0x05340001,
+	0xc78, 0x04350001,
+	0xc78, 0x03360001,
+	0xc78, 0x02370001,
+	0xc78, 0x01380001,
+	0xc78, 0x00390001,
+	0xc78, 0x003a0001,
+	0xc78, 0x003b0001,
+	0xc78, 0x003c0001,
+	0xc78, 0x003d0001,
+	0xc78, 0x003e0001,
+	0xc78, 0x003f0001,
+	0xc78, 0x38000002,
+	0xc78, 0x38010002,
+	0xc78, 0x38020002,
+	0xc78, 0x38030002,
+	0xc78, 0x38040002,
+	0xc78, 0x38050002,
+	0xc78, 0x38060002,
+	0xc78, 0x38070002,
+	0xc78, 0x38080002,
+	0xc78, 0x3c090002,
+	0xc78, 0x3e0a0002,
+	0xc78, 0x400b0002,
+	0xc78, 0x440c0002,
+	0xc78, 0x480d0002,
+	0xc78, 0x4c0e0002,
+	0xc78, 0x500f0002,
+	0xc78, 0x52100002,
+	0xc78, 0x56110002,
+	0xc78, 0x5a120002,
+	0xc78, 0x5e130002,
+	0xc78, 0x60140002,
+	0xc78, 0x60150002,
+	0xc78, 0x60160002,
+	0xc78, 0x62170002,
+	0xc78, 0x62180002,
+	0xc78, 0x62190002,
+	0xc78, 0x621a0002,
+	0xc78, 0x621b0002,
+	0xc78, 0x621c0002,
+	0xc78, 0x621d0002,
+	0xc78, 0x621e0002,
+	0xc78, 0x621f0002,
+	0xc78, 0x6e1f0000,
+};
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/table.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/table.h
new file mode 100644
index 000000000000..b809ba511320
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/table.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#ifndef __RTL92DU_TABLE_H__
+#define __RTL92DU_TABLE_H__
+
+#define PHY_REG_2T_ARRAYLENGTH 372
+#define PHY_REG_ARRAY_PG_LENGTH 624
+#define RADIOA_2T_ARRAYLENGTH 378
+#define RADIOB_2T_ARRAYLENGTH 384
+#define RADIOA_2T_INT_PA_ARRAYLENGTH 378
+#define RADIOB_2T_INT_PA_ARRAYLENGTH 384
+#define MAC_2T_ARRAYLENGTH 192
+#define AGCTAB_ARRAYLENGTH 386
+#define AGCTAB_5G_ARRAYLENGTH 194
+#define AGCTAB_2G_ARRAYLENGTH 194
+
+extern const u32 rtl8192du_phy_reg_2tarray[PHY_REG_2T_ARRAYLENGTH];
+extern const u32 rtl8192du_phy_reg_array_pg[PHY_REG_ARRAY_PG_LENGTH];
+extern const u32 rtl8192du_radioa_2tarray[RADIOA_2T_ARRAYLENGTH];
+extern const u32 rtl8192du_radiob_2tarray[RADIOB_2T_ARRAYLENGTH];
+extern const u32 rtl8192du_radioa_2t_int_paarray[RADIOA_2T_INT_PA_ARRAYLENGTH];
+extern const u32 rtl8192du_radiob_2t_int_paarray[RADIOB_2T_INT_PA_ARRAYLENGTH];
+extern const u32 rtl8192du_mac_2tarray[MAC_2T_ARRAYLENGTH];
+extern const u32 rtl8192du_agctab_array[AGCTAB_ARRAYLENGTH];
+extern const u32 rtl8192du_agctab_5garray[AGCTAB_5G_ARRAYLENGTH];
+extern const u32 rtl8192du_agctab_2garray[AGCTAB_2G_ARRAYLENGTH];
+
+#endif
-- 
cgit v1.2.3-58-ga151


From dc71f4e142df928747dc5b493faa2c841a400d88 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:44:39 +0300
Subject: wifi: rtlwifi: Add new members to struct rtl_priv for RTL8192DU

These are needed for the dual MAC version of RTL8192DU.

The two mutexes are used to avoid concurrent access to the hardware
from the two USB interfaces.

The two arrays are filled by one interface during LC calibration and
accessed by the other interface during channel switching.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/d1bf581b-de41-4d09-bf56-768c2fc29dd3@gmail.com
---
 drivers/net/wireless/realtek/rtlwifi/wifi.h | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtlwifi/wifi.h b/drivers/net/wireless/realtek/rtlwifi/wifi.h
index 442419568734..6a8212fd36af 100644
--- a/drivers/net/wireless/realtek/rtlwifi/wifi.h
+++ b/drivers/net/wireless/realtek/rtlwifi/wifi.h
@@ -2746,6 +2746,12 @@ struct rtl_priv {
 	 */
 	bool use_new_trx_flow;
 
+	/* For dual MAC RTL8192DU, things shared by the 2 USB interfaces */
+	u32 *curveindex_2g;
+	u32 *curveindex_5g;
+	struct mutex *mutex_for_power_on_off; /* for power on/off */
+	struct mutex *mutex_for_hw_init; /* for hardware init */
+
 #ifdef CONFIG_PM
 	struct wiphy_wowlan_support wowlan;
 #endif
-- 
cgit v1.2.3-58-ga151


From e769c67105d35da254a62892111e7dcad93ff3a6 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:45:31 +0300
Subject: wifi: rtlwifi: Add rtl8192du/hw.{c,h}

These contain mostly hardware init/deinit routines for RTL8192DU.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/282afe18-f37d-45f7-9275-68c39d1fde09@gmail.com
---
 .../net/wireless/realtek/rtlwifi/rtl8192du/hw.c    | 1212 ++++++++++++++++++++
 .../net/wireless/realtek/rtlwifi/rtl8192du/hw.h    |   22 +
 2 files changed, 1234 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/hw.c
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/hw.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/hw.c
new file mode 100644
index 000000000000..700c6e2bcad1
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/hw.c
@@ -0,0 +1,1212 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#include "../wifi.h"
+#include "../cam.h"
+#include "../usb.h"
+#include "../rtl8192d/reg.h"
+#include "../rtl8192d/def.h"
+#include "../rtl8192d/dm_common.h"
+#include "../rtl8192d/fw_common.h"
+#include "../rtl8192d/hw_common.h"
+#include "../rtl8192d/phy_common.h"
+#include "phy.h"
+#include "dm.h"
+#include "fw.h"
+#include "hw.h"
+#include "trx.h"
+
+static void _rtl92du_set_bcn_ctrl_reg(struct ieee80211_hw *hw,
+				      u8 set_bits, u8 clear_bits)
+{
+	struct rtl_usb *rtlusb = rtl_usbdev(rtl_usbpriv(hw));
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+
+	rtlusb->reg_bcn_ctrl_val |= set_bits;
+	rtlusb->reg_bcn_ctrl_val &= ~clear_bits;
+	rtl_write_byte(rtlpriv, REG_BCN_CTRL, (u8)rtlusb->reg_bcn_ctrl_val);
+}
+
+static void _rtl92du_enable_bcn_sub_func(struct ieee80211_hw *hw)
+{
+	_rtl92du_set_bcn_ctrl_reg(hw, 0, BIT(1));
+}
+
+static void _rtl92du_disable_bcn_sub_func(struct ieee80211_hw *hw)
+{
+	_rtl92du_set_bcn_ctrl_reg(hw, BIT(1), 0);
+}
+
+void rtl92du_get_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val)
+{
+	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
+
+	switch (variable) {
+	case HW_VAR_RCR:
+		*((u32 *)val) = mac->rx_conf;
+		break;
+	default:
+		rtl92d_get_hw_reg(hw, variable, val);
+		break;
+	}
+}
+
+void rtl92du_set_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_mac *mac = rtl_mac(rtlpriv);
+
+	switch (variable) {
+	case HW_VAR_AC_PARAM:
+		rtl92d_dm_init_edca_turbo(hw);
+		break;
+	case HW_VAR_ACM_CTRL: {
+		u8 e_aci = *val;
+		union aci_aifsn *p_aci_aifsn =
+		    (union aci_aifsn *)(&mac->ac[0].aifs);
+		u8 acm = p_aci_aifsn->f.acm;
+		u8 acm_ctrl = rtl_read_byte(rtlpriv, REG_ACMHWCTRL);
+
+		if (acm) {
+			switch (e_aci) {
+			case AC0_BE:
+				acm_ctrl |= ACMHW_BEQEN;
+				break;
+			case AC2_VI:
+				acm_ctrl |= ACMHW_VIQEN;
+				break;
+			case AC3_VO:
+				acm_ctrl |= ACMHW_VOQEN;
+				break;
+			default:
+				rtl_dbg(rtlpriv, COMP_ERR, DBG_WARNING,
+					"HW_VAR_ACM_CTRL acm set failed: eACI is %d\n",
+					acm);
+				break;
+			}
+		} else {
+			switch (e_aci) {
+			case AC0_BE:
+				acm_ctrl &= (~ACMHW_BEQEN);
+				break;
+			case AC2_VI:
+				acm_ctrl &= (~ACMHW_VIQEN);
+				break;
+			case AC3_VO:
+				acm_ctrl &= (~ACMHW_VOQEN);
+				break;
+			default:
+				pr_err("%s:%d switch case %#x not processed\n",
+				       __func__, __LINE__, e_aci);
+				break;
+			}
+		}
+		rtl_dbg(rtlpriv, COMP_QOS, DBG_TRACE,
+			"SetHwReg8190pci(): [HW_VAR_ACM_CTRL] Write 0x%X\n",
+			acm_ctrl);
+		rtl_write_byte(rtlpriv, REG_ACMHWCTRL, acm_ctrl);
+		break;
+	}
+	case HW_VAR_RCR:
+		mac->rx_conf = ((u32 *)val)[0];
+		rtl_write_dword(rtlpriv, REG_RCR, mac->rx_conf);
+		break;
+	case HW_VAR_H2C_FW_JOINBSSRPT: {
+		u8 tmp_regcr, tmp_reg422;
+		bool recover = false;
+		u8 mstatus = *val;
+
+		if (mstatus == RT_MEDIA_CONNECT) {
+			rtlpriv->cfg->ops->set_hw_reg(hw,
+						      HW_VAR_AID, NULL);
+			tmp_regcr = rtl_read_byte(rtlpriv, REG_CR + 1);
+			rtl_write_byte(rtlpriv, REG_CR + 1,
+				       tmp_regcr | ENSWBCN);
+			_rtl92du_set_bcn_ctrl_reg(hw, 0, EN_BCN_FUNCTION);
+			_rtl92du_set_bcn_ctrl_reg(hw, DIS_TSF_UDT, 0);
+			tmp_reg422 = rtl_read_byte(rtlpriv,
+						   REG_FWHW_TXQ_CTRL + 2);
+			if (tmp_reg422 & (EN_BCNQ_DL >> 16))
+				recover = true;
+			rtl_write_byte(rtlpriv, REG_FWHW_TXQ_CTRL + 2,
+				       tmp_reg422 & ~(EN_BCNQ_DL >> 16));
+
+			/* We don't implement FW LPS so this is not needed. */
+			/* rtl92d_set_fw_rsvdpagepkt(hw, 0); */
+
+			_rtl92du_set_bcn_ctrl_reg(hw, EN_BCN_FUNCTION, 0);
+			_rtl92du_set_bcn_ctrl_reg(hw, 0, DIS_TSF_UDT);
+			if (recover)
+				rtl_write_byte(rtlpriv,
+					       REG_FWHW_TXQ_CTRL + 2,
+					       tmp_reg422);
+			rtl_write_byte(rtlpriv, REG_CR + 1,
+				       tmp_regcr & ~ENSWBCN);
+		}
+		rtl92d_set_fw_joinbss_report_cmd(hw, (*val));
+		break;
+	}
+	case HW_VAR_CORRECT_TSF: {
+		u8 btype_ibss = val[0];
+
+		if (btype_ibss)
+			rtl92d_stop_tx_beacon(hw);
+		_rtl92du_set_bcn_ctrl_reg(hw, 0, EN_BCN_FUNCTION);
+		rtl_write_dword(rtlpriv, REG_TSFTR,
+				(u32)(mac->tsf & 0xffffffff));
+		rtl_write_dword(rtlpriv, REG_TSFTR + 4,
+				(u32)((mac->tsf >> 32) & 0xffffffff));
+		_rtl92du_set_bcn_ctrl_reg(hw, EN_BCN_FUNCTION, 0);
+		if (btype_ibss)
+			rtl92d_resume_tx_beacon(hw);
+
+		break;
+	}
+	case HW_VAR_KEEP_ALIVE:
+		/* Avoid "switch case not processed" error. RTL8192DU doesn't
+		 * need to do anything here, maybe.
+		 */
+		break;
+	default:
+		rtl92d_set_hw_reg(hw, variable, val);
+		break;
+	}
+}
+
+static void _rtl92du_init_queue_reserved_page(struct ieee80211_hw *hw,
+					      u8 out_ep_num,
+					      u8 queue_sel)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	u32 txqpagenum, txqpageunit;
+	u32 txqremainingpage;
+	u32 numhq = 0;
+	u32 numlq = 0;
+	u32 numnq = 0;
+	u32 numpubq;
+	u32 value32;
+
+	if (rtlhal->macphymode != SINGLEMAC_SINGLEPHY) {
+		numpubq = NORMAL_PAGE_NUM_PUBQ_92D_DUAL_MAC;
+		txqpagenum = TX_TOTAL_PAGE_NUMBER_92D_DUAL_MAC - numpubq;
+	} else {
+		numpubq = TEST_PAGE_NUM_PUBQ_92DU;
+		txqpagenum = TX_TOTAL_PAGE_NUMBER_92DU - numpubq;
+	}
+
+	if (rtlhal->macphymode != SINGLEMAC_SINGLEPHY && out_ep_num == 3) {
+		numhq = NORMAL_PAGE_NUM_HPQ_92D_DUAL_MAC;
+		numlq = NORMAL_PAGE_NUM_LPQ_92D_DUAL_MAC;
+		numnq = NORMAL_PAGE_NUM_NORMALQ_92D_DUAL_MAC;
+	} else {
+		txqpageunit = txqpagenum / out_ep_num;
+		txqremainingpage = txqpagenum % out_ep_num;
+
+		if (queue_sel & TX_SELE_HQ)
+			numhq = txqpageunit;
+		if (queue_sel & TX_SELE_LQ)
+			numlq = txqpageunit;
+		if (queue_sel & TX_SELE_NQ)
+			numnq = txqpageunit;
+
+		/* HIGH priority queue always present in the
+		 * configuration of 2 or 3 out-ep. Remainder pages
+		 * assigned to High queue
+		 */
+		if (out_ep_num > 1 && txqremainingpage)
+			numhq += txqremainingpage;
+	}
+
+	/* NOTE: This step done before writing REG_RQPN. */
+	rtl_write_byte(rtlpriv, REG_RQPN_NPQ, (u8)numnq);
+
+	/* TX DMA */
+	u32p_replace_bits(&value32, numhq, HPQ_MASK);
+	u32p_replace_bits(&value32, numlq, LPQ_MASK);
+	u32p_replace_bits(&value32, numpubq, PUBQ_MASK);
+	value32 |= LD_RQPN;
+	rtl_write_dword(rtlpriv, REG_RQPN, value32);
+}
+
+static void _rtl92du_init_tx_buffer_boundary(struct ieee80211_hw *hw,
+					     u8 txpktbuf_bndy)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+
+	rtl_write_byte(rtlpriv, REG_TXPKTBUF_BCNQ_BDNY, txpktbuf_bndy);
+	rtl_write_byte(rtlpriv, REG_TXPKTBUF_MGQ_BDNY, txpktbuf_bndy);
+
+	rtl_write_byte(rtlpriv, REG_TXPKTBUF_WMAC_LBK_BF_HD, txpktbuf_bndy);
+
+	/* TXRKTBUG_PG_BNDY */
+	rtl_write_byte(rtlpriv, REG_TRXFF_BNDY, txpktbuf_bndy);
+
+	/* Beacon Head for TXDMA */
+	rtl_write_byte(rtlpriv, REG_TDECTRL + 1, txpktbuf_bndy);
+}
+
+static bool _rtl92du_llt_table_init(struct ieee80211_hw *hw, u8 txpktbuf_bndy)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	unsigned short i;
+	bool status;
+	u8 maxpage;
+
+	if (rtlpriv->rtlhal.macphymode == SINGLEMAC_SINGLEPHY)
+		maxpage = 255;
+	else
+		maxpage = 127;
+
+	for (i = 0; i < (txpktbuf_bndy - 1); i++) {
+		status = rtl92d_llt_write(hw, i, i + 1);
+		if (!status)
+			return status;
+	}
+
+	/* end of list */
+	status = rtl92d_llt_write(hw, txpktbuf_bndy - 1, 0xFF);
+	if (!status)
+		return status;
+
+	/* Make the other pages as ring buffer
+	 * This ring buffer is used as beacon buffer if we
+	 * config this MAC as two MAC transfer.
+	 * Otherwise used as local loopback buffer.
+	 */
+	for (i = txpktbuf_bndy; i < maxpage; i++) {
+		status = rtl92d_llt_write(hw, i, i + 1);
+		if (!status)
+			return status;
+	}
+
+	/* Let last entry point to the start entry of ring buffer */
+	status = rtl92d_llt_write(hw, maxpage, txpktbuf_bndy);
+	if (!status)
+		return status;
+
+	return true;
+}
+
+static void _rtl92du_init_chipn_reg_priority(struct ieee80211_hw *hw, u16 beq,
+					     u16 bkq, u16 viq, u16 voq,
+					     u16 mgtq, u16 hiq)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u16 value16;
+
+	value16 = rtl_read_word(rtlpriv, REG_TRXDMA_CTRL) & 0x7;
+	u16p_replace_bits(&value16, beq, TXDMA_BEQ_MAP);
+	u16p_replace_bits(&value16, bkq, TXDMA_BKQ_MAP);
+	u16p_replace_bits(&value16, viq, TXDMA_VIQ_MAP);
+	u16p_replace_bits(&value16, voq, TXDMA_VOQ_MAP);
+	u16p_replace_bits(&value16, mgtq, TXDMA_MGQ_MAP);
+	u16p_replace_bits(&value16, hiq, TXDMA_HIQ_MAP);
+	rtl_write_word(rtlpriv,  REG_TRXDMA_CTRL, value16);
+}
+
+static void _rtl92du_init_chipn_one_out_ep_priority(struct ieee80211_hw *hw,
+						    u8 queue_sel)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u16 value;
+
+	switch (queue_sel) {
+	case TX_SELE_HQ:
+		value = QUEUE_HIGH;
+		break;
+	case TX_SELE_LQ:
+		value = QUEUE_LOW;
+		break;
+	case TX_SELE_NQ:
+		value = QUEUE_NORMAL;
+		break;
+	default:
+		WARN_ON(1); /* Shall not reach here! */
+		return;
+	}
+	_rtl92du_init_chipn_reg_priority(hw, value, value, value, value,
+					 value, value);
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+		"Tx queue select: 0x%02x\n", queue_sel);
+}
+
+static void _rtl92du_init_chipn_two_out_ep_priority(struct ieee80211_hw *hw,
+						    u8 queue_sel)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u16 beq, bkq, viq, voq, mgtq, hiq;
+	u16 valuehi, valuelow;
+
+	switch (queue_sel) {
+	default:
+		WARN_ON(1);
+		fallthrough;
+	case (TX_SELE_HQ | TX_SELE_LQ):
+		valuehi = QUEUE_HIGH;
+		valuelow = QUEUE_LOW;
+		break;
+	case (TX_SELE_NQ | TX_SELE_LQ):
+		valuehi = QUEUE_NORMAL;
+		valuelow = QUEUE_LOW;
+		break;
+	case (TX_SELE_HQ | TX_SELE_NQ):
+		valuehi = QUEUE_HIGH;
+		valuelow = QUEUE_NORMAL;
+		break;
+	}
+
+	beq = valuelow;
+	bkq = valuelow;
+	viq = valuehi;
+	voq = valuehi;
+	mgtq = valuehi;
+	hiq = valuehi;
+
+	_rtl92du_init_chipn_reg_priority(hw, beq, bkq, viq, voq, mgtq, hiq);
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+		"Tx queue select: 0x%02x\n", queue_sel);
+}
+
+static void _rtl92du_init_chipn_three_out_ep_priority(struct ieee80211_hw *hw,
+						      u8 queue_sel)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u16 beq, bkq, viq, voq, mgtq, hiq;
+
+	beq = QUEUE_LOW;
+	bkq = QUEUE_LOW;
+	viq = QUEUE_NORMAL;
+	voq = QUEUE_HIGH;
+	mgtq = QUEUE_HIGH;
+	hiq = QUEUE_HIGH;
+
+	_rtl92du_init_chipn_reg_priority(hw, beq, bkq, viq, voq, mgtq, hiq);
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+		"Tx queue select: 0x%02x\n", queue_sel);
+}
+
+static void _rtl92du_init_queue_priority(struct ieee80211_hw *hw,
+					 u8 out_ep_num,
+					 u8 queue_sel)
+{
+	switch (out_ep_num) {
+	case 1:
+		_rtl92du_init_chipn_one_out_ep_priority(hw, queue_sel);
+		break;
+	case 2:
+		_rtl92du_init_chipn_two_out_ep_priority(hw, queue_sel);
+		break;
+	case 3:
+		_rtl92du_init_chipn_three_out_ep_priority(hw, queue_sel);
+		break;
+	default:
+		WARN_ON(1); /* Shall not reach here! */
+		break;
+	}
+}
+
+static void _rtl92du_init_wmac_setting(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_mac *mac = rtl_mac(rtlpriv);
+
+	mac->rx_conf = RCR_APM | RCR_AM | RCR_AB | RCR_ADF | RCR_APP_ICV |
+		       RCR_AMF | RCR_HTC_LOC_CTRL | RCR_APP_MIC |
+		       RCR_APP_PHYST_RXFF | RCR_APPFCS;
+
+	rtl_write_dword(rtlpriv, REG_RCR, mac->rx_conf);
+
+	/* Set Multicast Address. */
+	rtl_write_dword(rtlpriv, REG_MAR, 0xffffffff);
+	rtl_write_dword(rtlpriv, REG_MAR + 4, 0xffffffff);
+}
+
+static void _rtl92du_init_adaptive_ctrl(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u32 val32;
+
+	val32 = rtl_read_dword(rtlpriv, REG_RRSR);
+	val32 &= ~0xfffff;
+	if (rtlpriv->rtlhal.current_bandtype == BAND_ON_5G)
+		val32 |= 0xffff0; /* No CCK */
+	else
+		val32 |= 0xffff1;
+	rtl_write_dword(rtlpriv, REG_RRSR, val32);
+
+	/* Set Spec SIFS (used in NAV) */
+	rtl_write_word(rtlpriv, REG_SPEC_SIFS, 0x1010);
+
+	/* Retry limit 0x30 */
+	rtl_write_word(rtlpriv, REG_RL, 0x3030);
+}
+
+static void _rtl92du_init_edca(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u16 val16;
+
+	/* Disable EDCCA count down, to reduce collison and retry */
+	val16 = rtl_read_word(rtlpriv, REG_RD_CTRL);
+	val16 |= DIS_EDCA_CNT_DWN;
+	rtl_write_word(rtlpriv, REG_RD_CTRL, val16);
+
+	/* CCK SIFS shall always be 10us. */
+	rtl_write_word(rtlpriv, REG_SIFS_CTX, 0x0a0a);
+	/* Set SIFS for OFDM */
+	rtl_write_word(rtlpriv, REG_SIFS_TRX, 0x1010);
+
+	rtl_write_word(rtlpriv, REG_PROT_MODE_CTRL, 0x0204);
+
+	rtl_write_dword(rtlpriv, REG_BAR_MODE_CTRL, 0x014004);
+
+	/* TXOP */
+	rtl_write_dword(rtlpriv, REG_EDCA_BE_PARAM, 0x005EA42B);
+	rtl_write_dword(rtlpriv, REG_EDCA_BK_PARAM, 0x0000A44F);
+	rtl_write_dword(rtlpriv, REG_EDCA_VI_PARAM, 0x005EA324);
+	rtl_write_dword(rtlpriv, REG_EDCA_VO_PARAM, 0x002FA226);
+
+	rtl_write_byte(rtlpriv, REG_PIFS, 0x1C);
+
+	rtl_write_byte(rtlpriv, REG_AGGR_BREAK_TIME, 0x16);
+
+	rtl_write_word(rtlpriv, REG_NAV_PROT_LEN, 0x0040);
+
+	rtl_write_byte(rtlpriv, REG_BCNDMATIM, 0x2);
+	rtl_write_byte(rtlpriv, REG_ATIMWND, 0x2);
+}
+
+static void _rtl92du_init_retry_function(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u8 val8;
+
+	val8 = rtl_read_byte(rtlpriv, REG_FWHW_TXQ_CTRL);
+	val8 |= EN_AMPDU_RTY_NEW;
+	rtl_write_byte(rtlpriv, REG_FWHW_TXQ_CTRL, val8);
+
+	rtl_write_byte(rtlpriv, REG_ACKTO, 0x40);
+}
+
+static void _rtl92du_init_operation_mode(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+
+	rtl_write_byte(rtlpriv, REG_BWOPMODE, BW_OPMODE_20MHZ);
+
+	switch (rtlpriv->phy.rf_type) {
+	case RF_1T2R:
+	case RF_1T1R:
+		rtlhal->minspace_cfg = (MAX_MSS_DENSITY_1T << 3);
+		break;
+	case RF_2T2R:
+	case RF_2T2R_GREEN:
+		rtlhal->minspace_cfg = (MAX_MSS_DENSITY_2T << 3);
+		break;
+	}
+	rtl_write_byte(rtlpriv, REG_AMPDU_MIN_SPACE, rtlhal->minspace_cfg);
+}
+
+static void _rtl92du_init_beacon_parameters(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+
+	rtl_write_word(rtlpriv, REG_BCN_CTRL, 0x1010);
+
+	rtl_write_word(rtlpriv, REG_TBTT_PROHIBIT, 0x3c02);
+	rtl_write_byte(rtlpriv, REG_DRVERLYINT, 0x05);
+	rtl_write_byte(rtlpriv, REG_BCNDMATIM, 0x03);
+
+	rtl_write_word(rtlpriv, REG_BCNTCFG, 0x660f);
+}
+
+static void _rtl92du_init_ampdu_aggregation(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+
+	/* Aggregation threshold */
+	if (rtlhal->macphymode == DUALMAC_DUALPHY)
+		rtl_write_dword(rtlpriv, REG_AGGLEN_LMT, 0x66525541);
+	else if (rtlhal->macphymode == DUALMAC_SINGLEPHY)
+		rtl_write_dword(rtlpriv, REG_AGGLEN_LMT, 0x44444441);
+	else
+		rtl_write_dword(rtlpriv, REG_AGGLEN_LMT, 0x88728841);
+
+	rtl_write_byte(rtlpriv, REG_AGGR_BREAK_TIME, 0x16);
+}
+
+static bool _rtl92du_init_power_on(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	unsigned short wordtmp;
+	unsigned char bytetmp;
+	u16 retry = 0;
+
+	do {
+		if (rtl_read_byte(rtlpriv, REG_APS_FSMCO) & PFM_ALDN)
+			break;
+
+		if (retry++ > 1000)
+			return false;
+	} while (true);
+
+	/* Unlock ISO/CLK/Power control register */
+	rtl_write_byte(rtlpriv, REG_RSV_CTRL, 0x00);
+
+	/* SPS0_CTRL 0x11[7:0] = 0x2b  enable SPS into PWM mode */
+	rtl_write_byte(rtlpriv, REG_SPS0_CTRL, 0x2b);
+
+	msleep(1);
+
+	bytetmp = rtl_read_byte(rtlpriv, REG_LDOV12D_CTRL);
+	if ((bytetmp & LDV12_EN) == 0) {
+		bytetmp |= LDV12_EN;
+		rtl_write_byte(rtlpriv, REG_LDOV12D_CTRL, bytetmp);
+
+		msleep(1);
+
+		bytetmp = rtl_read_byte(rtlpriv, REG_SYS_ISO_CTRL);
+		bytetmp &= ~ISO_MD2PP;
+		rtl_write_byte(rtlpriv, REG_SYS_ISO_CTRL, bytetmp);
+	}
+
+	/* Auto enable WLAN */
+	wordtmp = rtl_read_word(rtlpriv, REG_APS_FSMCO);
+	wordtmp |= APFM_ONMAC;
+	rtl_write_word(rtlpriv, REG_APS_FSMCO, wordtmp);
+
+	wordtmp = rtl_read_word(rtlpriv, REG_APS_FSMCO);
+	retry = 0;
+	while ((wordtmp & APFM_ONMAC) && retry < 1000) {
+		retry++;
+		wordtmp = rtl_read_word(rtlpriv, REG_APS_FSMCO);
+	}
+
+	/* Release RF digital isolation */
+	wordtmp = rtl_read_word(rtlpriv, REG_SYS_ISO_CTRL);
+	wordtmp &= ~ISO_DIOR;
+	rtl_write_word(rtlpriv, REG_SYS_ISO_CTRL, wordtmp);
+
+	/* Enable MAC DMA/WMAC/SCHEDULE/SEC block */
+	wordtmp = rtl_read_word(rtlpriv, REG_CR);
+	wordtmp |= HCI_TXDMA_EN | HCI_RXDMA_EN | TXDMA_EN | RXDMA_EN |
+		   PROTOCOL_EN | SCHEDULE_EN | MACTXEN | MACRXEN | ENSEC;
+	rtl_write_word(rtlpriv, REG_CR, wordtmp);
+
+	return true;
+}
+
+static bool _rtl92du_init_mac(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u8 val8;
+
+	rtl_write_byte(rtlpriv, REG_RSV_CTRL, 0x00);
+
+	val8 = rtl_read_byte(rtlpriv, REG_SYS_FUNC_EN + 1);
+	val8 &= ~(FEN_MREGEN >> 8);
+	rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN + 1, val8);
+
+	/* For s3/s4 may reset mac, Reg0xf8 may be set to 0,
+	 * so reset macphy control reg here.
+	 */
+	rtl92d_phy_config_macphymode(hw);
+
+	rtl92du_phy_set_poweron(hw);
+
+	if (!_rtl92du_init_power_on(hw)) {
+		pr_err("Failed to init power on!\n");
+		return false;
+	}
+
+	rtl92d_phy_config_maccoexist_rfpage(hw);
+
+	return true;
+}
+
+int rtl92du_hw_init(struct ieee80211_hw *hw)
+{
+	struct rtl_usb_priv *usb_priv = rtl_usbpriv(hw);
+	struct rtl_usb *rtlusb = rtl_usbdev(usb_priv);
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_ps_ctl *ppsc = rtl_psc(rtlpriv);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	struct rtl_mac *mac = rtl_mac(rtlpriv);
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	u8 val8, txpktbuf_bndy;
+	int err, i;
+	u32 val32;
+	u16 val16;
+
+	mutex_lock(rtlpriv->mutex_for_hw_init);
+
+	/* we should do iqk after disable/enable */
+	rtl92d_phy_reset_iqk_result(hw);
+
+	if (!_rtl92du_init_mac(hw)) {
+		pr_err("Init MAC failed\n");
+		mutex_unlock(rtlpriv->mutex_for_hw_init);
+		return 1;
+	}
+
+	if (rtlhal->macphymode == SINGLEMAC_SINGLEPHY)
+		txpktbuf_bndy = 249;
+	else
+		txpktbuf_bndy = 123;
+
+	if (!_rtl92du_llt_table_init(hw, txpktbuf_bndy)) {
+		pr_err("Init LLT failed\n");
+		mutex_unlock(rtlpriv->mutex_for_hw_init);
+		return 1;
+	}
+
+	err = rtl92du_download_fw(hw);
+
+	/* return fail only when part number check fail */
+	if (err && rtl_read_byte(rtlpriv, 0x1c5) == 0xe0) {
+		rtl_dbg(rtlpriv, COMP_ERR, DBG_WARNING,
+			"Failed to download FW. Init HW without FW..\n");
+		mutex_unlock(rtlpriv->mutex_for_hw_init);
+		return 1;
+	}
+	rtlhal->last_hmeboxnum = 0;
+	rtlpriv->psc.fw_current_inpsmode = false;
+
+	rtl92du_phy_mac_config(hw);
+
+	/* Set reserved page for each queue */
+	_rtl92du_init_queue_reserved_page(hw, rtlusb->out_ep_nums,
+					  rtlusb->out_queue_sel);
+
+	_rtl92du_init_tx_buffer_boundary(hw, txpktbuf_bndy);
+
+	_rtl92du_init_queue_priority(hw, rtlusb->out_ep_nums,
+				     rtlusb->out_queue_sel);
+
+	/* Set Tx/Rx page size (Tx must be 128 Bytes,
+	 * Rx can be 64, 128, 256, 512, 1024 bytes)
+	 */
+	rtl_write_byte(rtlpriv, REG_PBP, 0x11);
+
+	/* Get Rx PHY status in order to report RSSI and others. */
+	rtl_write_byte(rtlpriv, REG_RX_DRVINFO_SZ, 0x4);
+
+	rtl_write_dword(rtlpriv, REG_HISR, 0xffffffff);
+	rtl_write_dword(rtlpriv, REG_HIMR, 0xffffffff);
+
+	val8 = rtl_read_byte(rtlpriv, MSR);
+	val8 &= ~MSR_MASK;
+	val8 |= MSR_INFRA;
+	rtl_write_byte(rtlpriv, MSR, val8);
+
+	_rtl92du_init_wmac_setting(hw);
+	_rtl92du_init_adaptive_ctrl(hw);
+	_rtl92du_init_edca(hw);
+
+	rtl_write_dword(rtlpriv, REG_DARFRC, 0x00000000);
+	rtl_write_dword(rtlpriv, REG_DARFRC + 4, 0x10080404);
+	rtl_write_dword(rtlpriv, REG_RARFRC, 0x04030201);
+	rtl_write_dword(rtlpriv, REG_RARFRC + 4, 0x08070605);
+
+	_rtl92du_init_retry_function(hw);
+	/* _InitUsbAggregationSetting(padapter); no aggregation for now */
+	_rtl92du_init_operation_mode(hw);
+	_rtl92du_init_beacon_parameters(hw);
+	_rtl92du_init_ampdu_aggregation(hw);
+
+	rtl_write_byte(rtlpriv, REG_BCN_MAX_ERR, 0xff);
+
+	/* unit: 256us. 256ms */
+	rtl_write_word(rtlpriv, REG_PKT_VO_VI_LIFE_TIME, 0x0400);
+	rtl_write_word(rtlpriv, REG_PKT_BE_BK_LIFE_TIME, 0x0400);
+
+	/* Hardware-controlled blinking. */
+	rtl_write_word(rtlpriv, REG_LEDCFG0, 0x8282);
+	rtl_write_byte(rtlpriv, REG_LEDCFG2, 0x82);
+
+	val32 = rtl_read_dword(rtlpriv, REG_TXDMA_OFFSET_CHK);
+	val32 |= DROP_DATA_EN;
+	rtl_write_dword(rtlpriv, REG_TXDMA_OFFSET_CHK, val32);
+
+	if (mac->rdg_en) {
+		rtl_write_byte(rtlpriv, REG_RD_CTRL, 0xff);
+		rtl_write_word(rtlpriv, REG_RD_NAV_NXT, 0x200);
+		rtl_write_byte(rtlpriv, REG_RD_RESP_PKT_TH, 0x05);
+	}
+
+	for (i = 0; i < 4; i++)
+		rtl_write_dword(rtlpriv, REG_ARFR0 + i * 4, 0x1f8ffff0);
+
+	if (rtlhal->macphymode == SINGLEMAC_SINGLEPHY) {
+		if (rtlusb->out_ep_nums == 2)
+			rtl_write_dword(rtlpriv, REG_FAST_EDCA_CTRL, 0x03066666);
+		else
+			rtl_write_word(rtlpriv, REG_FAST_EDCA_CTRL, 0x8888);
+	} else {
+		rtl_write_word(rtlpriv, REG_FAST_EDCA_CTRL, 0x5555);
+	}
+
+	val8 = rtl_read_byte(rtlpriv, 0x605);
+	val8 |= 0xf0;
+	rtl_write_byte(rtlpriv, 0x605, val8);
+
+	rtl_write_byte(rtlpriv, REG_RXTSF_OFFSET_CCK, 0x30);
+	rtl_write_byte(rtlpriv, REG_RXTSF_OFFSET_OFDM, 0x30);
+	rtl_write_byte(rtlpriv, 0x606, 0x30);
+
+	/* temp for high queue and mgnt Queue corrupt in time; it may
+	 * cause hang when sw beacon use high_Q, other frame use mgnt_Q;
+	 * or, sw beacon use mgnt_Q, other frame use high_Q;
+	 */
+	rtl_write_byte(rtlpriv, REG_DIS_TXREQ_CLR, 0x10);
+	val16 = rtl_read_word(rtlpriv, REG_RD_CTRL);
+	val16 |= BIT(12);
+	rtl_write_word(rtlpriv, REG_RD_CTRL, val16);
+
+	rtl_write_byte(rtlpriv, REG_TXPAUSE, 0);
+
+	/* usb suspend idle time count for bitfile0927 */
+	val8 = rtl_read_byte(rtlpriv, 0xfe56);
+	val8 |= BIT(0) | BIT(1);
+	rtl_write_byte(rtlpriv, 0xfe56, val8);
+
+	if (rtlhal->earlymode_enable) {
+		rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+			"EarlyMode Enabled!!!\n");
+
+		val8 = rtl_read_byte(rtlpriv, REG_EARLY_MODE_CONTROL);
+		val8 |= 0x1f;
+		rtl_write_byte(rtlpriv, REG_EARLY_MODE_CONTROL, val8);
+
+		rtl_write_byte(rtlpriv, REG_EARLY_MODE_CONTROL + 3, 0x80);
+
+		val8 = rtl_read_byte(rtlpriv, 0x605);
+		val8 |= 0x40;
+		rtl_write_byte(rtlpriv, 0x605, val8);
+	} else {
+		rtl_write_byte(rtlpriv, REG_EARLY_MODE_CONTROL, 0);
+	}
+
+	rtl92du_phy_bb_config(hw);
+
+	rtlphy->rf_mode = RF_OP_BY_SW_3WIRE;
+	/* set before initialize RF */
+	rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER4, 0x00f00000, 0xf);
+
+	/* config RF */
+	rtl92du_phy_rf_config(hw);
+
+	/* set default value after initialize RF */
+	rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER4, 0x00f00000, 0);
+
+	/* After load BB, RF params, we need to do more for 92D. */
+	rtl92du_update_bbrf_configuration(hw);
+
+	rtlphy->rfreg_chnlval[0] =
+		rtl_get_rfreg(hw, RF90_PATH_A, RF_CHNLBW, RFREG_OFFSET_MASK);
+	rtlphy->rfreg_chnlval[1] =
+		rtl_get_rfreg(hw, RF90_PATH_B, RF_CHNLBW, RFREG_OFFSET_MASK);
+
+	/*---- Set CCK and OFDM Block "ON"----*/
+	if (rtlhal->current_bandtype == BAND_ON_2_4G)
+		rtl_set_bbreg(hw, RFPGA0_RFMOD, BCCKEN, 0x1);
+	rtl_set_bbreg(hw, RFPGA0_RFMOD, BOFDMEN, 0x1);
+
+	/* reset hw sec */
+	rtl_cam_reset_all_entry(hw);
+	rtl92d_enable_hw_security_config(hw);
+
+	rtl_write_byte(rtlpriv, REG_HWSEQ_CTRL, 0xFF);
+
+	/* schmitt trigger, improve tx evm for 92du */
+	val8 = rtl_read_byte(rtlpriv, REG_AFE_XTAL_CTRL);
+	val8 |= BIT(1);
+	rtl_write_byte(rtlpriv, REG_AFE_XTAL_CTRL, val8);
+
+	/* Disable bar */
+	rtl_write_dword(rtlpriv, REG_BAR_MODE_CTRL, 0xffff);
+
+	/* Nav limit */
+	rtl_write_byte(rtlpriv, REG_NAV_CTRL + 2, 0);
+	rtl_write_byte(rtlpriv, ROFDM0_XATXAFE + 3, 0x50);
+
+	/* Read EEPROM TX power index and PHY_REG_PG.txt to capture correct
+	 * TX power index for different rate set.
+	 */
+	rtl92d_phy_get_hw_reg_originalvalue(hw);
+
+	ppsc->rfpwr_state = ERFON;
+
+	/* do IQK for 2.4G for better scan result */
+	if (rtlhal->current_bandtype == BAND_ON_2_4G)
+		rtl92du_phy_iq_calibrate(hw);
+
+	rtl92du_phy_lc_calibrate(hw, IS_92D_SINGLEPHY(rtlhal->version));
+
+	rtl92du_phy_init_pa_bias(hw);
+
+	mutex_unlock(rtlpriv->mutex_for_hw_init);
+
+	rtl92du_dm_init(hw);
+
+	/* For 2 PORT TSF SYNC */
+	rtl_write_word(rtlpriv, REG_BCN_CTRL, 0x1818);
+	rtlusb->reg_bcn_ctrl_val = 0x18;
+
+	udelay(500);
+
+	if (rtlhal->macphymode != DUALMAC_DUALPHY) {
+		rtl_write_dword(rtlpriv, RFPGA1_TXINFO,
+				rtl_read_dword(rtlpriv, RFPGA1_TXINFO) & ~BIT(30));
+
+		rtl_write_dword(rtlpriv, RFPGA0_TXGAINSTAGE,
+				rtl_read_dword(rtlpriv, RFPGA0_TXGAINSTAGE) & ~BIT(31));
+
+		rtl_write_dword(rtlpriv, ROFDM0_XBTXAFE, 0xa0e40000);
+	}
+
+	val32 = rtl_read_dword(rtlpriv, REG_FWHW_TXQ_CTRL);
+	val32 |= BIT(12);
+	rtl_write_dword(rtlpriv, REG_FWHW_TXQ_CTRL, val32);
+
+	return err;
+}
+
+static int _rtl92du_set_media_status(struct ieee80211_hw *hw,
+				     enum nl80211_iftype type)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	enum led_ctl_mode ledaction = LED_CTL_NO_LINK;
+	u8 bt_msr = rtl_read_byte(rtlpriv, MSR);
+
+	bt_msr &= 0xfc;
+
+	if (type == NL80211_IFTYPE_UNSPECIFIED ||
+	    type == NL80211_IFTYPE_STATION) {
+		rtl92d_stop_tx_beacon(hw);
+		_rtl92du_enable_bcn_sub_func(hw);
+	} else if (type == NL80211_IFTYPE_ADHOC ||
+		   type == NL80211_IFTYPE_AP) {
+		rtl92d_resume_tx_beacon(hw);
+		_rtl92du_disable_bcn_sub_func(hw);
+	} else {
+		rtl_dbg(rtlpriv, COMP_ERR, DBG_WARNING,
+			"Set HW_VAR_MEDIA_STATUS: No such media status(%x)\n",
+			type);
+	}
+
+	switch (type) {
+	case NL80211_IFTYPE_UNSPECIFIED:
+		bt_msr |= MSR_NOLINK;
+		ledaction = LED_CTL_LINK;
+		rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE,
+			"Set Network type to NO LINK!\n");
+		break;
+	case NL80211_IFTYPE_ADHOC:
+		bt_msr |= MSR_ADHOC;
+		rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE,
+			"Set Network type to Ad Hoc!\n");
+		break;
+	case NL80211_IFTYPE_STATION:
+		bt_msr |= MSR_INFRA;
+		ledaction = LED_CTL_LINK;
+		rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE,
+			"Set Network type to STA!\n");
+		break;
+	case NL80211_IFTYPE_AP:
+		bt_msr |= MSR_AP;
+		rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE,
+			"Set Network type to AP!\n");
+		break;
+	default:
+		pr_err("Network type %d not supported!\n", type);
+		return 1;
+	}
+	rtl_write_byte(rtlpriv, MSR, bt_msr);
+
+	rtlpriv->cfg->ops->led_control(hw, ledaction);
+
+	if ((bt_msr & MSR_MASK) == MSR_AP)
+		rtl_write_byte(rtlpriv, REG_BCNTCFG + 1, 0x00);
+	else
+		rtl_write_byte(rtlpriv, REG_BCNTCFG + 1, 0x66);
+
+	return 0;
+}
+
+void rtl92du_set_check_bssid(struct ieee80211_hw *hw, bool check_bssid)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u32 reg_rcr;
+
+	if (rtlpriv->psc.rfpwr_state != ERFON)
+		return;
+
+	rtlpriv->cfg->ops->get_hw_reg(hw, HW_VAR_RCR, (u8 *)(&reg_rcr));
+
+	if (check_bssid) {
+		reg_rcr |= RCR_CBSSID_DATA | RCR_CBSSID_BCN;
+		rtlpriv->cfg->ops->set_hw_reg(hw, HW_VAR_RCR, (u8 *)&reg_rcr);
+		_rtl92du_set_bcn_ctrl_reg(hw, 0, DIS_TSF_UDT);
+	} else if (!check_bssid) {
+		reg_rcr &= ~(RCR_CBSSID_DATA | RCR_CBSSID_BCN);
+		_rtl92du_set_bcn_ctrl_reg(hw, DIS_TSF_UDT, 0);
+		rtlpriv->cfg->ops->set_hw_reg(hw, HW_VAR_RCR, (u8 *)&reg_rcr);
+	}
+}
+
+int rtl92du_set_network_type(struct ieee80211_hw *hw, enum nl80211_iftype type)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+
+	if (_rtl92du_set_media_status(hw, type))
+		return -EOPNOTSUPP;
+
+	/* check bssid */
+	if (rtlpriv->mac80211.link_state == MAC80211_LINKED) {
+		if (type != NL80211_IFTYPE_AP)
+			rtl92du_set_check_bssid(hw, true);
+	} else {
+		rtl92du_set_check_bssid(hw, false);
+	}
+
+	return 0;
+}
+
+/* do iqk or reload iqk */
+/* windows just rtl92d_phy_reload_iqk_setting in set channel,
+ * but it's very strict for time sequence so we add
+ * rtl92d_phy_reload_iqk_setting here
+ */
+void rtl92du_linked_set_reg(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	u8 channel = rtlphy->current_channel;
+	u8 indexforchannel;
+
+	indexforchannel = rtl92d_get_rightchnlplace_for_iqk(channel);
+	if (!rtlphy->iqk_matrix[indexforchannel].iqk_done) {
+		rtl_dbg(rtlpriv, COMP_SCAN | COMP_INIT, DBG_DMESG,
+			"Do IQK for channel:%d\n", channel);
+		rtl92du_phy_iq_calibrate(hw);
+	}
+}
+
+void rtl92du_enable_interrupt(struct ieee80211_hw *hw)
+{
+	/* Nothing to do. */
+}
+
+void rtl92du_disable_interrupt(struct ieee80211_hw *hw)
+{
+	/* Nothing to do. */
+}
+
+static void _rtl92du_poweroff_adapter(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u8 retry = 100;
+	u8 u1b_tmp;
+	u16 val16;
+	u32 val32;
+
+	rtl_write_byte(rtlpriv, REG_LDOA15_CTRL, 0x04);
+
+	rtl_write_byte(rtlpriv, REG_RF_CTRL, 0x00);
+
+	/* IF fw in RAM code, do reset */
+	if (rtl_read_byte(rtlpriv, REG_MCUFWDL) & MCUFWDL_RDY) {
+		rtl_write_byte(rtlpriv, REG_FSIMR, 0);
+
+		/* We need to disable other HRCV INT to influence 8051 reset. */
+		rtl_write_byte(rtlpriv, REG_FWIMR, 0x20);
+
+		/* Close mask to prevent incorrect FW write operation. */
+		rtl_write_byte(rtlpriv, REG_FTIMR, 0);
+
+		rtl_write_byte(rtlpriv, REG_MCUFWDL, 0);
+
+		/* Set (REG_HMETFR + 3) to 0x20 is reset 8051 */
+		rtl_write_byte(rtlpriv, REG_HMETFR + 3, 0x20);
+		val16 = rtl_read_word(rtlpriv, REG_SYS_FUNC_EN);
+		while (val16 & FEN_CPUEN) {
+			retry--;
+			if (retry == 0)
+				break;
+			udelay(50);
+			val16 = rtl_read_word(rtlpriv, REG_SYS_FUNC_EN);
+		}
+
+		if (retry == 0) {
+			rtl_write_byte(rtlpriv, REG_FWIMR, 0);
+
+			/* if 8051 reset fail, reset MAC directly. */
+			rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN + 1, 0x50);
+
+			mdelay(10);
+		}
+	}
+
+	/* reset MCU, MAC register, DCORE */
+	rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN + 1, 0x54);
+
+	/* reset MCU ready status */
+	rtl_write_byte(rtlpriv, REG_MCUFWDL, 0x00);
+
+	/* Pull GPIO PIN to balance level and LED control */
+
+	/* Disable GPIO[7:0] */
+	rtl_write_word(rtlpriv, REG_GPIO_PIN_CTRL + 2, 0x0000);
+	val32 = rtl_read_dword(rtlpriv, REG_GPIO_PIN_CTRL);
+	u32p_replace_bits(&val32, val32 & 0xff, 0x0000ff00);
+	u32p_replace_bits(&val32, 0xff, 0x00ff0000);
+	rtl_write_dword(rtlpriv, REG_GPIO_PIN_CTRL, val32);
+
+	/* Disable GPIO[10:8] */
+	rtl_write_byte(rtlpriv, REG_MAC_PINMUX_CFG, 0);
+	val16 = rtl_read_word(rtlpriv, REG_GPIO_IO_SEL);
+	u16p_replace_bits(&val16, val16 & 0xf, 0x00f0);
+	u16p_replace_bits(&val16, 0xf, 0x0780);
+	rtl_write_word(rtlpriv, REG_GPIO_IO_SEL, val16);
+
+	/* Disable LED 0, 1, and 2 */
+	rtl_write_word(rtlpriv, REG_LEDCFG0, 0x8888);
+	rtl_write_byte(rtlpriv, REG_LEDCFG2, 0x88);
+
+	/* Disable analog sequence */
+
+	/* enter PFM mode */
+	rtl_write_byte(rtlpriv, REG_SPS0_CTRL, 0x23);
+
+	rtl_write_word(rtlpriv, REG_APS_FSMCO,
+		       APDM_HOST | AFSM_HSUS | PFM_ALDN);
+
+	/* lock ISO/CLK/Power control register */
+	rtl_write_byte(rtlpriv, REG_RSV_CTRL, 0x0e);
+
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+		"In PowerOff,reg0x%x=%X\n",
+		REG_SPS0_CTRL, rtl_read_byte(rtlpriv, REG_SPS0_CTRL));
+
+	/* 0x17[7] 1b': power off in process  0b' : power off over */
+	if (rtlpriv->rtlhal.macphymode != SINGLEMAC_SINGLEPHY) {
+		mutex_lock(rtlpriv->mutex_for_power_on_off);
+		u1b_tmp = rtl_read_byte(rtlpriv, REG_POWER_OFF_IN_PROCESS);
+		u1b_tmp &= ~BIT(7);
+		rtl_write_byte(rtlpriv, REG_POWER_OFF_IN_PROCESS, u1b_tmp);
+		mutex_unlock(rtlpriv->mutex_for_power_on_off);
+	}
+
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD, "<=======\n");
+}
+
+void rtl92du_card_disable(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_ps_ctl *ppsc = rtl_psc(rtl_priv(hw));
+	struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw));
+	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
+	enum nl80211_iftype opmode;
+	u32 val32;
+	u16 val16;
+	u8 val8;
+
+	mac->link_state = MAC80211_NOLINK;
+	opmode = NL80211_IFTYPE_UNSPECIFIED;
+	_rtl92du_set_media_status(hw, opmode);
+
+	RT_SET_PS_LEVEL(ppsc, RT_RF_OFF_LEVL_HALT_NIC);
+	/* Power sequence for each MAC. */
+	/* a. stop tx DMA  */
+	/* b. close RF */
+	/* c. clear rx buf */
+	/* d. stop rx DMA */
+	/* e. reset MAC */
+
+	val16 = rtl_read_word(rtlpriv, REG_GPIO_MUXCFG);
+	val16 &= ~BIT(12);
+	rtl_write_word(rtlpriv, REG_GPIO_MUXCFG, val16);
+
+	rtl_write_byte(rtlpriv, REG_TXPAUSE, 0xff);
+	udelay(500);
+	rtl_write_byte(rtlpriv, REG_CR, 0);
+
+	/* RF OFF sequence */
+	rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER4, 0x00f00000, 0xf);
+	rtl_set_rfreg(hw, RF90_PATH_A, RF_AC, RFREG_OFFSET_MASK, 0x00);
+
+	rtl_write_byte(rtlpriv, REG_APSD_CTRL, 0x40);
+
+	val8 = FEN_USBD | FEN_USBA | FEN_BB_GLB_RSTN;
+	rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN, val8);
+
+	/* Mac0 can not do Global reset. Mac1 can do. */
+	if (rtlhal->macphymode == SINGLEMAC_SINGLEPHY ||
+	    rtlhal->interfaceindex == 1) {
+		/* before BB reset should do clock gated */
+		val32 = rtl_read_dword(rtlpriv, RFPGA0_XCD_RFPARAMETER);
+		val32 |= BIT(31);
+		rtl_write_dword(rtlpriv, RFPGA0_XCD_RFPARAMETER, val32);
+
+		val8 &= ~FEN_BB_GLB_RSTN;
+		rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN, val8);
+	}
+
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD, "==> Do power off.......\n");
+	if (!rtl92du_phy_check_poweroff(hw))
+		return;
+
+	_rtl92du_poweroff_adapter(hw);
+}
+
+void rtl92du_set_beacon_related_registers(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_mac *mac = rtl_mac(rtlpriv);
+	u16 bcn_interval, atim_window;
+
+	bcn_interval = mac->beacon_interval;
+	atim_window = 2;
+	rtl92du_disable_interrupt(hw);
+	rtl_write_word(rtlpriv, REG_ATIMWND, atim_window);
+	rtl_write_word(rtlpriv, REG_BCN_INTERVAL, bcn_interval);
+	rtl_write_word(rtlpriv, REG_BCNTCFG, 0x660f);
+	rtl_write_byte(rtlpriv, REG_RXTSF_OFFSET_CCK, 0x20);
+	if (rtlpriv->rtlhal.current_bandtype == BAND_ON_5G)
+		rtl_write_byte(rtlpriv, REG_RXTSF_OFFSET_OFDM, 0x30);
+	else
+		rtl_write_byte(rtlpriv, REG_RXTSF_OFFSET_OFDM, 0x20);
+	rtl_write_byte(rtlpriv, 0x606, 0x30);
+}
+
+void rtl92du_set_beacon_interval(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
+	u16 bcn_interval = mac->beacon_interval;
+
+	rtl_dbg(rtlpriv, COMP_BEACON, DBG_DMESG,
+		"beacon_interval:%d\n", bcn_interval);
+	rtl92du_disable_interrupt(hw);
+	rtl_write_word(rtlpriv, REG_BCN_INTERVAL, bcn_interval);
+	rtl92du_enable_interrupt(hw);
+}
+
+void rtl92du_update_interrupt_mask(struct ieee80211_hw *hw,
+				   u32 add_msr, u32 rm_msr)
+{
+	/* Nothing to do here. */
+}
+
+void rtl92du_read_chip_version(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+
+	/* Chip version reading is done in rtl92d_read_eeprom_info. */
+
+	rtlpriv->rtlhal.hw_type = HARDWARE_TYPE_RTL8192DU;
+}
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/hw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/hw.h
new file mode 100644
index 000000000000..80ed00c90c16
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/hw.h
@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#ifndef __RTL92DU_HW_H__
+#define __RTL92DU_HW_H__
+
+void rtl92du_get_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val);
+void rtl92du_set_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val);
+void rtl92du_read_chip_version(struct ieee80211_hw *hw);
+int rtl92du_hw_init(struct ieee80211_hw *hw);
+void rtl92du_card_disable(struct ieee80211_hw *hw);
+void rtl92du_enable_interrupt(struct ieee80211_hw *hw);
+void rtl92du_disable_interrupt(struct ieee80211_hw *hw);
+int rtl92du_set_network_type(struct ieee80211_hw *hw, enum nl80211_iftype type);
+void rtl92du_set_check_bssid(struct ieee80211_hw *hw, bool check_bssid);
+void rtl92du_set_beacon_related_registers(struct ieee80211_hw *hw);
+void rtl92du_set_beacon_interval(struct ieee80211_hw *hw);
+void rtl92du_update_interrupt_mask(struct ieee80211_hw *hw,
+				   u32 add_msr, u32 rm_msr);
+void rtl92du_linked_set_reg(struct ieee80211_hw *hw);
+
+#endif
-- 
cgit v1.2.3-58-ga151


From af46caf0515281a30e119b53afb3083fcc5110a9 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:46:02 +0300
Subject: wifi: rtlwifi: Add rtl8192du/phy.{c,h}

These contain mostly the calibration and channel switching routines
for RTL8192DU.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/299a5ab7-46bf-410b-bbeb-e252eeeeb34c@gmail.com
---
 .../net/wireless/realtek/rtlwifi/rtl8192du/phy.c   | 3123 ++++++++++++++++++++
 .../net/wireless/realtek/rtlwifi/rtl8192du/phy.h   |   32 +
 2 files changed, 3155 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/phy.c
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/phy.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/phy.c
new file mode 100644
index 000000000000..289ec71ce3e5
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/phy.c
@@ -0,0 +1,3123 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#include "../wifi.h"
+#include "../ps.h"
+#include "../core.h"
+#include "../efuse.h"
+#include "../usb.h"
+#include "../rtl8192d/reg.h"
+#include "../rtl8192d/def.h"
+#include "../rtl8192d/phy_common.h"
+#include "../rtl8192d/rf_common.h"
+#include "phy.h"
+#include "rf.h"
+#include "table.h"
+
+#define MAX_RF_IMR_INDEX			12
+#define MAX_RF_IMR_INDEX_NORMAL			13
+#define RF_REG_NUM_FOR_C_CUT_5G			6
+#define RF_REG_NUM_FOR_C_CUT_5G_INTERNALPA	7
+#define RF_REG_NUM_FOR_C_CUT_2G			5
+#define RF_CHNL_NUM_5G				19
+#define RF_CHNL_NUM_5G_40M			17
+#define CV_CURVE_CNT				64
+
+static const u32 rf_reg_for_5g_swchnl_normal[MAX_RF_IMR_INDEX_NORMAL] = {
+	0, 0x2f, 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x0
+};
+
+static const u8 rf_reg_for_c_cut_5g[RF_REG_NUM_FOR_C_CUT_5G] = {
+	RF_SYN_G1, RF_SYN_G2, RF_SYN_G3, RF_SYN_G4, RF_SYN_G5, RF_SYN_G6
+};
+
+static const u8 rf_reg_for_c_cut_2g[RF_REG_NUM_FOR_C_CUT_2G] = {
+	RF_SYN_G1, RF_SYN_G2, RF_SYN_G3, RF_SYN_G7, RF_SYN_G8
+};
+
+static const u8 rf_for_c_cut_5g_internal_pa[RF_REG_NUM_FOR_C_CUT_5G_INTERNALPA] = {
+	0x0B, 0x48, 0x49, 0x4B, 0x03, 0x04, 0x0E
+};
+
+static const u32 rf_reg_mask_for_c_cut_2g[RF_REG_NUM_FOR_C_CUT_2G] = {
+	BIT(19) | BIT(18) | BIT(17) | BIT(14) | BIT(1),
+	BIT(10) | BIT(9),
+	BIT(18) | BIT(17) | BIT(16) | BIT(1),
+	BIT(2) | BIT(1),
+	BIT(15) | BIT(14) | BIT(13) | BIT(12) | BIT(11)
+};
+
+static const u8 rf_chnl_5g[RF_CHNL_NUM_5G] = {
+	36, 40, 44, 48, 52, 56, 60, 64, 100, 104, 108,
+	112, 116, 120, 124, 128, 132, 136, 140
+};
+
+static const u8 rf_chnl_5g_40m[RF_CHNL_NUM_5G_40M] = {
+	38, 42, 46, 50, 54, 58, 62, 102, 106, 110, 114,
+	118, 122, 126, 130, 134, 138
+};
+
+static const u32 rf_reg_pram_c_5g[5][RF_REG_NUM_FOR_C_CUT_5G] = {
+	{0xE43BE, 0xFC638, 0x77C0A, 0xDE471, 0xd7110, 0x8EB04},
+	{0xE43BE, 0xFC078, 0xF7C1A, 0xE0C71, 0xD7550, 0xAEB04},
+	{0xE43BF, 0xFF038, 0xF7C0A, 0xDE471, 0xE5550, 0xAEB04},
+	{0xE43BF, 0xFF079, 0xF7C1A, 0xDE471, 0xE5550, 0xAEB04},
+	{0xE43BF, 0xFF038, 0xF7C1A, 0xDE471, 0xd7550, 0xAEB04}
+};
+
+static const u32 rf_reg_param_for_c_cut_2g[3][RF_REG_NUM_FOR_C_CUT_2G] = {
+	{0x643BC, 0xFC038, 0x77C1A, 0x41289, 0x01840},
+	{0x643BC, 0xFC038, 0x07C1A, 0x41289, 0x01840},
+	{0x243BC, 0xFC438, 0x07C1A, 0x4128B, 0x0FC41}
+};
+
+static const u32 rf_syn_g4_for_c_cut_2g = 0xD1C31 & 0x7FF;
+
+static const u32 rf_pram_c_5g_int_pa[3][RF_REG_NUM_FOR_C_CUT_5G_INTERNALPA] = {
+	{0x01a00, 0x40443, 0x00eb5, 0x89bec, 0x94a12, 0x94a12, 0x94a12},
+	{0x01800, 0xc0443, 0x00730, 0x896ee, 0x94a52, 0x94a52, 0x94a52},
+	{0x01800, 0xc0443, 0x00730, 0x896ee, 0x94a12, 0x94a12, 0x94a12}
+};
+
+/* [patha+b][reg] */
+static const u32 rf_imr_param_normal[3][MAX_RF_IMR_INDEX_NORMAL] = {
+	/* channels 1-14. */
+	{
+		0x70000, 0x00ff0, 0x4400f, 0x00ff0, 0x0, 0x0, 0x0,
+		0x0, 0x0, 0x64888, 0xe266c, 0x00090, 0x22fff
+	},
+	/* channels 36-64 */
+	{
+		0x70000, 0x22880, 0x4470f, 0x55880, 0x00070, 0x88000,
+		0x0, 0x88080, 0x70000, 0x64a82, 0xe466c, 0x00090,
+		0x32c9a
+	},
+	/* channels 100-165 */
+	{
+		0x70000, 0x44880, 0x4477f, 0x77880, 0x00070, 0x88000,
+		0x0, 0x880b0, 0x0, 0x64b82, 0xe466c, 0x00090, 0x32c9a
+	}
+};
+
+static const u32 targetchnl_5g[TARGET_CHNL_NUM_5G] = {
+	25141, 25116, 25091, 25066, 25041,
+	25016, 24991, 24966, 24941, 24917,
+	24892, 24867, 24843, 24818, 24794,
+	24770, 24765, 24721, 24697, 24672,
+	24648, 24624, 24600, 24576, 24552,
+	24528, 24504, 24480, 24457, 24433,
+	24409, 24385, 24362, 24338, 24315,
+	24291, 24268, 24245, 24221, 24198,
+	24175, 24151, 24128, 24105, 24082,
+	24059, 24036, 24013, 23990, 23967,
+	23945, 23922, 23899, 23876, 23854,
+	23831, 23809, 23786, 23764, 23741,
+	23719, 23697, 23674, 23652, 23630,
+	23608, 23586, 23564, 23541, 23519,
+	23498, 23476, 23454, 23432, 23410,
+	23388, 23367, 23345, 23323, 23302,
+	23280, 23259, 23237, 23216, 23194,
+	23173, 23152, 23130, 23109, 23088,
+	23067, 23046, 23025, 23003, 22982,
+	22962, 22941, 22920, 22899, 22878,
+	22857, 22837, 22816, 22795, 22775,
+	22754, 22733, 22713, 22692, 22672,
+	22652, 22631, 22611, 22591, 22570,
+	22550, 22530, 22510, 22490, 22469,
+	22449, 22429, 22409, 22390, 22370,
+	22350, 22336, 22310, 22290, 22271,
+	22251, 22231, 22212, 22192, 22173,
+	22153, 22134, 22114, 22095, 22075,
+	22056, 22037, 22017, 21998, 21979,
+	21960, 21941, 21921, 21902, 21883,
+	21864, 21845, 21826, 21807, 21789,
+	21770, 21751, 21732, 21713, 21695,
+	21676, 21657, 21639, 21620, 21602,
+	21583, 21565, 21546, 21528, 21509,
+	21491, 21473, 21454, 21436, 21418,
+	21400, 21381, 21363, 21345, 21327,
+	21309, 21291, 21273, 21255, 21237,
+	21219, 21201, 21183, 21166, 21148,
+	21130, 21112, 21095, 21077, 21059,
+	21042, 21024, 21007, 20989, 20972,
+	25679, 25653, 25627, 25601, 25575,
+	25549, 25523, 25497, 25471, 25446,
+	25420, 25394, 25369, 25343, 25318,
+	25292, 25267, 25242, 25216, 25191,
+	25166
+};
+
+/* channel 1~14 */
+static const u32 targetchnl_2g[TARGET_CHNL_NUM_2G] = {
+	26084, 26030, 25976, 25923, 25869, 25816, 25764,
+	25711, 25658, 25606, 25554, 25502, 25451, 25328
+};
+
+u32 rtl92du_phy_query_bb_reg(struct ieee80211_hw *hw, u32 regaddr, u32 bitmask)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	u32 returnvalue, originalvalue, bitshift;
+
+	rtl_dbg(rtlpriv, COMP_RF, DBG_TRACE, "regaddr(%#x), bitmask(%#x)\n",
+		regaddr, bitmask);
+
+	if (rtlhal->during_mac1init_radioa)
+		regaddr |= MAC1_ACCESS_PHY0;
+	else if (rtlhal->during_mac0init_radiob)
+		regaddr |= MAC0_ACCESS_PHY1;
+
+	originalvalue = rtl_read_dword(rtlpriv, regaddr);
+	bitshift = calculate_bit_shift(bitmask);
+	returnvalue = (originalvalue & bitmask) >> bitshift;
+	rtl_dbg(rtlpriv, COMP_RF, DBG_TRACE,
+		"BBR MASK=0x%x Addr[0x%x]=0x%x\n",
+		bitmask, regaddr, originalvalue);
+	return returnvalue;
+}
+
+void rtl92du_phy_set_bb_reg(struct ieee80211_hw *hw,
+			    u32 regaddr, u32 bitmask, u32 data)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	u32 originalvalue, bitshift;
+
+	rtl_dbg(rtlpriv, COMP_RF, DBG_TRACE,
+		"regaddr(%#x), bitmask(%#x), data(%#x)\n",
+		regaddr, bitmask, data);
+
+	if (rtlhal->during_mac1init_radioa)
+		regaddr |= MAC1_ACCESS_PHY0;
+	else if (rtlhal->during_mac0init_radiob)
+		regaddr |= MAC0_ACCESS_PHY1;
+
+	if (bitmask != MASKDWORD) {
+		originalvalue = rtl_read_dword(rtlpriv, regaddr);
+		bitshift = calculate_bit_shift(bitmask);
+		data = (originalvalue & (~bitmask)) |
+			((data << bitshift) & bitmask);
+	}
+
+	rtl_write_dword(rtlpriv, regaddr, data);
+	rtl_dbg(rtlpriv, COMP_RF, DBG_TRACE,
+		"regaddr(%#x), bitmask(%#x), data(%#x)\n",
+		regaddr, bitmask, data);
+}
+
+/* To avoid miswrite Reg0x800 for 92D */
+static void rtl92du_phy_set_bb_reg_1byte(struct ieee80211_hw *hw,
+					 u32 regaddr, u32 bitmask, u32 data)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u32 originalvalue, bitshift, offset;
+	u8 value;
+
+	/* BitMask only support bit0~bit7 or bit8~bit15, bit16~bit23,
+	 * bit24~bit31, should be in 1 byte scale;
+	 */
+	bitshift = calculate_bit_shift(bitmask);
+	offset = bitshift / 8;
+
+	originalvalue = rtl_read_dword(rtlpriv, regaddr);
+	data = (originalvalue & (~bitmask)) | ((data << bitshift) & bitmask);
+
+	value = data >> (8 * offset);
+
+	rtl_write_byte(rtlpriv, regaddr + offset, value);
+}
+
+bool rtl92du_phy_mac_config(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u32 arraylength;
+	const u32 *ptrarray;
+	u32 i;
+
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE, "Read Rtl819XMACPHY_Array\n");
+
+	arraylength = MAC_2T_ARRAYLENGTH;
+	ptrarray = rtl8192du_mac_2tarray;
+
+	for (i = 0; i < arraylength; i = i + 2)
+		rtl_write_byte(rtlpriv, ptrarray[i], (u8)ptrarray[i + 1]);
+
+	if (rtlpriv->rtlhal.macphymode == SINGLEMAC_SINGLEPHY) {
+		/* improve 2-stream TX EVM */
+		/* rtl_write_byte(rtlpriv, 0x14,0x71); */
+		/* AMPDU aggregation number 9 */
+		/* rtl_write_word(rtlpriv, REG_MAX_AGGR_NUM, MAX_AGGR_NUM); */
+		rtl_write_byte(rtlpriv, REG_MAX_AGGR_NUM, 0x0B);
+	} else {
+		/* 92D need to test to decide the num. */
+		rtl_write_byte(rtlpriv, REG_MAX_AGGR_NUM, 0x07);
+	}
+
+	return true;
+}
+
+static bool _rtl92du_phy_config_bb(struct ieee80211_hw *hw, u8 configtype)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	u16 phy_reg_arraylen, agctab_arraylen = 0;
+	const u32 *agctab_array_table = NULL;
+	const u32 *phy_regarray_table;
+	int i;
+
+	/* Normal chip, Mac0 use AGC_TAB.txt for 2G and 5G band. */
+	if (rtlhal->interfaceindex == 0) {
+		agctab_arraylen = AGCTAB_ARRAYLENGTH;
+		agctab_array_table = rtl8192du_agctab_array;
+		rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+			" ===> phy:MAC0, Rtl819XAGCTAB_Array\n");
+	} else {
+		if (rtlhal->current_bandtype == BAND_ON_2_4G) {
+			agctab_arraylen = AGCTAB_2G_ARRAYLENGTH;
+			agctab_array_table = rtl8192du_agctab_2garray;
+			rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+				" ===> phy:MAC1, Rtl819XAGCTAB_2GArray\n");
+		} else {
+			agctab_arraylen = AGCTAB_5G_ARRAYLENGTH;
+			agctab_array_table = rtl8192du_agctab_5garray;
+			rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+				" ===> phy:MAC1, Rtl819XAGCTAB_5GArray\n");
+		}
+	}
+	phy_reg_arraylen = PHY_REG_2T_ARRAYLENGTH;
+	phy_regarray_table = rtl8192du_phy_reg_2tarray;
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+		" ===> phy:Rtl819XPHY_REG_Array_PG\n");
+
+	if (configtype == BASEBAND_CONFIG_PHY_REG) {
+		for (i = 0; i < phy_reg_arraylen; i = i + 2) {
+			rtl_addr_delay(phy_regarray_table[i]);
+			rtl_set_bbreg(hw, phy_regarray_table[i], MASKDWORD,
+				      phy_regarray_table[i + 1]);
+			udelay(1);
+			rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE,
+				"The phy_regarray_table[0] is %x Rtl819XPHY_REGArray[1] is %x\n",
+				phy_regarray_table[i],
+				phy_regarray_table[i + 1]);
+		}
+	} else if (configtype == BASEBAND_CONFIG_AGC_TAB) {
+		for (i = 0; i < agctab_arraylen; i = i + 2) {
+			rtl_set_bbreg(hw, agctab_array_table[i],
+				      MASKDWORD, agctab_array_table[i + 1]);
+
+			/* Add 1us delay between BB/RF register setting. */
+			udelay(1);
+
+			rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE,
+				"AGC table %u %u\n",
+				agctab_array_table[i],
+				agctab_array_table[i + 1]);
+		}
+		rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+			"Normal Chip, loaded AGC table\n");
+	}
+	return true;
+}
+
+static bool _rtl92du_phy_config_bb_pg(struct ieee80211_hw *hw, u8 configtype)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	const u32 *phy_regarray_table_pg;
+	u16 phy_regarray_pg_len;
+	int i;
+
+	phy_regarray_pg_len = PHY_REG_ARRAY_PG_LENGTH;
+	phy_regarray_table_pg = rtl8192du_phy_reg_array_pg;
+
+	if (configtype == BASEBAND_CONFIG_PHY_REG) {
+		for (i = 0; i < phy_regarray_pg_len; i = i + 3) {
+			rtl_addr_delay(phy_regarray_table_pg[i]);
+			rtl92d_store_pwrindex_diffrate_offset(hw,
+				phy_regarray_table_pg[i],
+				phy_regarray_table_pg[i + 1],
+				phy_regarray_table_pg[i + 2]);
+		}
+	} else {
+		rtl_dbg(rtlpriv, COMP_SEND, DBG_TRACE,
+			"configtype != BaseBand_Config_PHY_REG\n");
+	}
+	return true;
+}
+
+static bool _rtl92du_phy_bb_config(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_efuse *rtlefuse = rtl_efuse(rtlpriv);
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	bool ret;
+
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE, "==>\n");
+	ret = _rtl92du_phy_config_bb(hw, BASEBAND_CONFIG_PHY_REG);
+	if (!ret) {
+		pr_err("Write BB Reg Fail!!\n");
+		return false;
+	}
+
+	if (!rtlefuse->autoload_failflag) {
+		rtlphy->pwrgroup_cnt = 0;
+		ret = _rtl92du_phy_config_bb_pg(hw, BASEBAND_CONFIG_PHY_REG);
+	}
+	if (!ret) {
+		pr_err("BB_PG Reg Fail!!\n");
+		return false;
+	}
+
+	ret = _rtl92du_phy_config_bb(hw, BASEBAND_CONFIG_AGC_TAB);
+	if (!ret) {
+		pr_err("AGC Table Fail\n");
+		return false;
+	}
+
+	rtlphy->cck_high_power = (bool)rtl_get_bbreg(hw,
+						     RFPGA0_XA_HSSIPARAMETER2,
+						     0x200);
+
+	return true;
+}
+
+bool rtl92du_phy_bb_config(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	bool rtstatus;
+	u32 regvaldw;
+	u16 regval;
+	u8 value;
+
+	rtl92d_phy_init_bb_rf_register_definition(hw);
+
+	regval = rtl_read_word(rtlpriv, REG_SYS_FUNC_EN);
+	rtl_write_word(rtlpriv, REG_SYS_FUNC_EN,
+		       regval | BIT(13) | BIT(0) | BIT(1));
+
+	rtl_write_byte(rtlpriv, REG_AFE_PLL_CTRL, 0x83);
+	rtl_write_byte(rtlpriv, REG_AFE_PLL_CTRL + 1, 0xdb);
+
+	/* 0x1f bit7 bit6 represent for mac0/mac1 driver ready */
+	value = rtl_read_byte(rtlpriv, REG_RF_CTRL);
+	rtl_write_byte(rtlpriv, REG_RF_CTRL, value | RF_EN | RF_RSTB |
+		RF_SDMRSTB);
+
+	value = FEN_BB_GLB_RSTN | FEN_BBRSTB;
+	if (rtlhal->interface == INTF_PCI)
+		value |= FEN_PPLL | FEN_PCIEA | FEN_DIO_PCIE;
+	else if (rtlhal->interface == INTF_USB)
+		value |= FEN_USBA | FEN_USBD;
+	rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN, value);
+
+	regvaldw = rtl_read_dword(rtlpriv, RFPGA0_XCD_RFPARAMETER);
+	regvaldw &= ~BIT(31);
+	rtl_write_dword(rtlpriv, RFPGA0_XCD_RFPARAMETER, regvaldw);
+
+	/* To Fix MAC loopback mode fail. */
+	rtl_write_byte(rtlpriv, REG_LDOHCI12_CTRL, 0x0f);
+	rtl_write_byte(rtlpriv, 0x15, 0xe9);
+
+	rtl_write_byte(rtlpriv, REG_AFE_XTAL_CTRL + 1, 0x80);
+	if (!(IS_92D_SINGLEPHY(rtlpriv->rtlhal.version)) &&
+	    rtlhal->interface == INTF_PCI) {
+		regvaldw = rtl_read_dword(rtlpriv, REG_LEDCFG0);
+		rtl_write_dword(rtlpriv, REG_LEDCFG0, regvaldw | BIT(23));
+	}
+
+	rtstatus = _rtl92du_phy_bb_config(hw);
+
+	/* Crystal calibration */
+	rtl_set_bbreg(hw, REG_AFE_XTAL_CTRL, 0xf0,
+		      rtlpriv->efuse.crystalcap & 0x0f);
+	rtl_set_bbreg(hw, REG_AFE_PLL_CTRL, 0xf0000000,
+		      (rtlpriv->efuse.crystalcap & 0xf0) >> 4);
+
+	return rtstatus;
+}
+
+bool rtl92du_phy_rf_config(struct ieee80211_hw *hw)
+{
+	return rtl92du_phy_rf6052_config(hw);
+}
+
+bool rtl92du_phy_config_rf_with_headerfile(struct ieee80211_hw *hw,
+					   enum rf_content content,
+					   enum radio_path rfpath)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u16 radioa_arraylen, radiob_arraylen;
+	const u32 *radioa_array_table;
+	const u32 *radiob_array_table;
+	int i;
+
+	radioa_arraylen = RADIOA_2T_ARRAYLENGTH;
+	radioa_array_table = rtl8192du_radioa_2tarray;
+	radiob_arraylen = RADIOB_2T_ARRAYLENGTH;
+	radiob_array_table = rtl8192du_radiob_2tarray;
+	if (rtlpriv->efuse.internal_pa_5g[0]) {
+		radioa_arraylen = RADIOA_2T_INT_PA_ARRAYLENGTH;
+		radioa_array_table = rtl8192du_radioa_2t_int_paarray;
+	}
+	if (rtlpriv->efuse.internal_pa_5g[1]) {
+		radiob_arraylen = RADIOB_2T_INT_PA_ARRAYLENGTH;
+		radiob_array_table = rtl8192du_radiob_2t_int_paarray;
+	}
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+		"PHY_ConfigRFWithHeaderFile() Radio_A:Rtl819XRadioA_1TArray\n");
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+		"PHY_ConfigRFWithHeaderFile() Radio_B:Rtl819XRadioB_1TArray\n");
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE, "Radio No %x\n", rfpath);
+
+	/* this only happens when DMDP, mac0 start on 2.4G,
+	 * mac1 start on 5G, mac 0 has to set phy0 & phy1
+	 * pathA or mac1 has to set phy0 & phy1 pathA
+	 */
+	if (content == radiob_txt && rfpath == RF90_PATH_A) {
+		rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+			" ===> althougth Path A, we load radiob.txt\n");
+		radioa_arraylen = radiob_arraylen;
+		radioa_array_table = radiob_array_table;
+	}
+
+	switch (rfpath) {
+	case RF90_PATH_A:
+		for (i = 0; i < radioa_arraylen; i = i + 2) {
+			rtl_rfreg_delay(hw, rfpath, radioa_array_table[i],
+					RFREG_OFFSET_MASK,
+					radioa_array_table[i + 1]);
+		}
+		break;
+	case RF90_PATH_B:
+		for (i = 0; i < radiob_arraylen; i = i + 2) {
+			rtl_rfreg_delay(hw, rfpath, radiob_array_table[i],
+					RFREG_OFFSET_MASK,
+					radiob_array_table[i + 1]);
+		}
+		break;
+	case RF90_PATH_C:
+	case RF90_PATH_D:
+		pr_err("switch case %#x not processed\n", rfpath);
+		break;
+	}
+
+	return true;
+}
+
+void rtl92du_phy_set_bw_mode(struct ieee80211_hw *hw,
+			     enum nl80211_channel_type ch_type)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	struct rtl_mac *mac = rtl_mac(rtlpriv);
+	u8 reg_bw_opmode;
+	u8 reg_prsr_rsc;
+
+	if (rtlphy->set_bwmode_inprogress)
+		return;
+
+	if ((is_hal_stop(rtlhal)) || (RT_CANNOT_IO(hw))) {
+		rtl_dbg(rtlpriv, COMP_ERR, DBG_WARNING,
+			"FALSE driver sleep or unload\n");
+		return;
+	}
+
+	rtlphy->set_bwmode_inprogress = true;
+
+	rtl_dbg(rtlpriv, COMP_SCAN, DBG_TRACE, "Switch to %s bandwidth\n",
+		rtlphy->current_chan_bw == HT_CHANNEL_WIDTH_20 ?
+		"20MHz" : "40MHz");
+
+	reg_bw_opmode = rtl_read_byte(rtlpriv, REG_BWOPMODE);
+	reg_prsr_rsc = rtl_read_byte(rtlpriv, REG_RRSR + 2);
+
+	switch (rtlphy->current_chan_bw) {
+	case HT_CHANNEL_WIDTH_20:
+		reg_bw_opmode |= BW_OPMODE_20MHZ;
+		rtl_write_byte(rtlpriv, REG_BWOPMODE, reg_bw_opmode);
+		break;
+	case HT_CHANNEL_WIDTH_20_40:
+		reg_bw_opmode &= ~BW_OPMODE_20MHZ;
+		rtl_write_byte(rtlpriv, REG_BWOPMODE, reg_bw_opmode);
+
+		reg_prsr_rsc = (reg_prsr_rsc & 0x90) |
+			       (mac->cur_40_prime_sc << 5);
+		rtl_write_byte(rtlpriv, REG_RRSR + 2, reg_prsr_rsc);
+		break;
+	default:
+		pr_err("unknown bandwidth: %#X\n",
+		       rtlphy->current_chan_bw);
+		break;
+	}
+
+	switch (rtlphy->current_chan_bw) {
+	case HT_CHANNEL_WIDTH_20:
+		rtl92du_phy_set_bb_reg_1byte(hw, RFPGA0_RFMOD, BRFMOD, 0x0);
+		rtl_set_bbreg(hw, RFPGA1_RFMOD, BRFMOD, 0x0);
+		/* SET BIT10 BIT11  for receive cck */
+		rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER2, BIT(10) | BIT(11), 3);
+		break;
+	case HT_CHANNEL_WIDTH_20_40:
+		rtl92du_phy_set_bb_reg_1byte(hw, RFPGA0_RFMOD, BRFMOD, 0x1);
+		rtl_set_bbreg(hw, RFPGA1_RFMOD, BRFMOD, 0x1);
+		/* Set Control channel to upper or lower.
+		 * These settings are required only for 40MHz
+		 */
+		if (rtlhal->current_bandtype == BAND_ON_2_4G)
+			rtl_set_bbreg(hw, RCCK0_SYSTEM, BCCKSIDEBAND,
+				      mac->cur_40_prime_sc >> 1);
+		rtl_set_bbreg(hw, ROFDM1_LSTF, 0xC00, mac->cur_40_prime_sc);
+		/* SET BIT10 BIT11  for receive cck */
+		rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER2,
+			      BIT(10) | BIT(11), 0);
+		rtl_set_bbreg(hw, 0x818, BIT(26) | BIT(27),
+			      mac->cur_40_prime_sc ==
+			      HAL_PRIME_CHNL_OFFSET_LOWER ? 2 : 1);
+		break;
+	default:
+		pr_err("unknown bandwidth: %#X\n",
+		       rtlphy->current_chan_bw);
+		break;
+	}
+
+	rtl92d_phy_rf6052_set_bandwidth(hw, rtlphy->current_chan_bw);
+
+	rtlphy->set_bwmode_inprogress = false;
+	rtl_dbg(rtlpriv, COMP_SCAN, DBG_TRACE, "<==\n");
+}
+
+static void _rtl92du_phy_stop_trx_before_changeband(struct ieee80211_hw *hw)
+{
+	rtl92du_phy_set_bb_reg_1byte(hw, RFPGA0_RFMOD, BCCKEN | BOFDMEN, 0);
+	rtl_set_bbreg(hw, ROFDM0_TRXPATHENABLE, MASKBYTE0, 0x00);
+	rtl_set_bbreg(hw, ROFDM1_TRXPATHENABLE, BDWORD, 0x0);
+}
+
+static void rtl92du_phy_switch_wirelessband(struct ieee80211_hw *hw, u8 band)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	u16 basic_rates;
+	u32 reg_mac;
+	u8 value8;
+
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD, "==>\n");
+	rtlhal->bandset = band;
+	rtlhal->current_bandtype = band;
+	if (IS_92D_SINGLEPHY(rtlhal->version))
+		rtlhal->bandset = BAND_ON_BOTH;
+
+	/* stop RX/Tx */
+	_rtl92du_phy_stop_trx_before_changeband(hw);
+
+	/* reconfig BB/RF according to wireless mode */
+	if (rtlhal->current_bandtype == BAND_ON_2_4G)
+		/* BB & RF Config */
+		rtl_dbg(rtlpriv, COMP_CMD, DBG_DMESG, "====>2.4G\n");
+	else
+		/* 5G band */
+		rtl_dbg(rtlpriv, COMP_CMD, DBG_DMESG, "====>5G\n");
+
+	if (rtlhal->interfaceindex == 1)
+		_rtl92du_phy_config_bb(hw, BASEBAND_CONFIG_AGC_TAB);
+
+	rtl92du_update_bbrf_configuration(hw);
+
+	basic_rates = RRSR_6M | RRSR_12M | RRSR_24M;
+	if (rtlhal->current_bandtype == BAND_ON_2_4G)
+		basic_rates |= RRSR_1M | RRSR_2M | RRSR_5_5M | RRSR_11M;
+	rtlpriv->cfg->ops->set_hw_reg(hw, HW_VAR_BASIC_RATE,
+				      (u8 *)&basic_rates);
+
+	rtl92du_phy_set_bb_reg_1byte(hw, RFPGA0_RFMOD, BCCKEN | BOFDMEN, 0x3);
+
+	/* 20M BW. */
+	/* rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER2, BIT(10), 1); */
+	rtlhal->reloadtxpowerindex = true;
+
+	reg_mac = rtlhal->interfaceindex == 0 ? REG_MAC0 : REG_MAC1;
+
+	/* notice fw know band status  0x81[1]/0x53[1] = 0: 5G, 1: 2G */
+	if (rtlhal->current_bandtype == BAND_ON_2_4G) {
+		value8 = rtl_read_byte(rtlpriv,	reg_mac);
+		value8 |= BIT(1);
+		rtl_write_byte(rtlpriv, reg_mac, value8);
+	} else {
+		value8 = rtl_read_byte(rtlpriv, reg_mac);
+		value8 &= ~BIT(1);
+		rtl_write_byte(rtlpriv, reg_mac, value8);
+	}
+	mdelay(1);
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD, "<==Switch Band OK\n");
+}
+
+static void _rtl92du_phy_reload_imr_setting(struct ieee80211_hw *hw,
+					    u8 channel, u8 rfpath)
+{
+	struct rtl_usb *rtlusb = rtl_usbdev(rtl_usbpriv(hw));
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u8 group, i;
+
+	if (rtlusb->udev->speed != USB_SPEED_HIGH)
+		return;
+
+	rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "====>path %d\n", rfpath);
+	if (rtlpriv->rtlhal.current_bandtype == BAND_ON_5G) {
+		rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "====>5G\n");
+		rtl92du_phy_set_bb_reg_1byte(hw, RFPGA0_RFMOD,
+					     BOFDMEN | BCCKEN, 0);
+		rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER4, 0x00f00000, 0xf);
+
+		/* fc area 0xd2c */
+		if (channel >= 149)
+			rtl_set_bbreg(hw, ROFDM1_CFOTRACKING, BIT(13) |
+				      BIT(14), 2);
+		else
+			rtl_set_bbreg(hw, ROFDM1_CFOTRACKING, BIT(13) |
+				      BIT(14), 1);
+
+		/* leave 0 for channel1-14. */
+		group = channel <= 64 ? 1 : 2;
+		for (i = 0; i < MAX_RF_IMR_INDEX_NORMAL; i++)
+			rtl_set_rfreg(hw, (enum radio_path)rfpath,
+				      rf_reg_for_5g_swchnl_normal[i],
+				      RFREG_OFFSET_MASK,
+				      rf_imr_param_normal[group][i]);
+
+		rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER4, 0x00f00000, 0);
+		rtl92du_phy_set_bb_reg_1byte(hw, RFPGA0_RFMOD,
+					     BOFDMEN | BCCKEN, 3);
+	} else {
+		/* G band. */
+		rtl_dbg(rtlpriv, COMP_SCAN, DBG_LOUD,
+			"Load RF IMR parameters for G band. IMR already setting %d\n",
+			rtlpriv->rtlhal.load_imrandiqk_setting_for2g);
+		rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "====>2.4G\n");
+
+		if (!rtlpriv->rtlhal.load_imrandiqk_setting_for2g) {
+			rtl_dbg(rtlpriv, COMP_SCAN, DBG_LOUD,
+				"Load RF IMR parameters for G band. %d\n",
+				rfpath);
+			rtl92du_phy_set_bb_reg_1byte(hw, RFPGA0_RFMOD,
+						     BOFDMEN | BCCKEN, 0);
+			rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER4,
+				      0x00f00000, 0xf);
+
+			for (i = 0; i < MAX_RF_IMR_INDEX_NORMAL; i++) {
+				rtl_set_rfreg(hw, (enum radio_path)rfpath,
+					      rf_reg_for_5g_swchnl_normal[i],
+					      RFREG_OFFSET_MASK,
+					      rf_imr_param_normal[0][i]);
+			}
+
+			rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER4,
+				      0x00f00000, 0);
+			rtl92du_phy_set_bb_reg_1byte(hw, RFPGA0_RFMOD,
+						     BOFDMEN | BCCKEN, 3);
+		}
+	}
+	rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "<====\n");
+}
+
+static void _rtl92du_phy_switch_rf_setting(struct ieee80211_hw *hw, u8 channel)
+{
+	struct rtl_usb *rtlusb = rtl_usbdev(rtl_usbpriv(hw));
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	u8 path = rtlhal->current_bandtype == BAND_ON_5G ? RF90_PATH_A
+							 : RF90_PATH_B;
+	u32 u4regvalue, mask = 0x1C000, value = 0, u4tmp, u4tmp2;
+	bool need_pwr_down = false, internal_pa = false;
+	u32 regb30 = rtl_get_bbreg(hw, 0xb30, BIT(27));
+	u8 index = 0, i, rfpath;
+
+	if (rtlusb->udev->speed != USB_SPEED_HIGH)
+		return;
+
+	rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "====>\n");
+	/* config path A for 5G */
+	if (rtlhal->current_bandtype == BAND_ON_5G) {
+		rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "====>5G\n");
+		u4tmp = rtlpriv->curveindex_5g[channel - 1];
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"ver 1 set RF-A, 5G, 0x28 = 0x%x !!\n", u4tmp);
+
+		for (i = 0; i < RF_CHNL_NUM_5G; i++) {
+			if (channel == rf_chnl_5g[i] && channel <= 140)
+				index = 0;
+		}
+		for (i = 0; i < RF_CHNL_NUM_5G_40M; i++) {
+			if (channel == rf_chnl_5g_40m[i] && channel <= 140)
+				index = 1;
+		}
+		if (channel == 149 || channel == 155 || channel == 161)
+			index = 2;
+		else if (channel == 151 || channel == 153 || channel == 163 ||
+			 channel == 165)
+			index = 3;
+		else if (channel == 157 || channel == 159)
+			index = 4;
+
+		if (rtlhal->macphymode == DUALMAC_DUALPHY &&
+		    rtlhal->interfaceindex == 1) {
+			need_pwr_down = rtl92du_phy_enable_anotherphy(hw, false);
+			rtlhal->during_mac1init_radioa = true;
+			/* asume no this case */
+			if (need_pwr_down)
+				rtl92d_phy_enable_rf_env(hw, path,
+							 &u4regvalue);
+		}
+
+		/* DMDP, if band = 5G, Mac0 need to set PHY1 when regB30[27]=1 */
+		if (regb30 && rtlhal->interfaceindex == 0) {
+			need_pwr_down = rtl92du_phy_enable_anotherphy(hw, true);
+			rtlhal->during_mac0init_radiob = true;
+			if (need_pwr_down)
+				rtl92d_phy_enable_rf_env(hw, path,
+							 &u4regvalue);
+		}
+
+		for (i = 0; i < RF_REG_NUM_FOR_C_CUT_5G; i++) {
+			if (i == 0 && rtlhal->macphymode == DUALMAC_DUALPHY) {
+				rtl_set_rfreg(hw, (enum radio_path)path,
+					      rf_reg_for_c_cut_5g[i],
+					      RFREG_OFFSET_MASK, 0xE439D);
+			} else if (rf_reg_for_c_cut_5g[i] == RF_SYN_G4) {
+				u4tmp2 = (rf_reg_pram_c_5g[index][i] &
+				     0x7FF) | (u4tmp << 11);
+				if (channel == 36)
+					u4tmp2 &= ~(BIT(7) | BIT(6));
+				rtl_set_rfreg(hw, (enum radio_path)path,
+					      rf_reg_for_c_cut_5g[i],
+					      RFREG_OFFSET_MASK, u4tmp2);
+			} else {
+				rtl_set_rfreg(hw, (enum radio_path)path,
+					      rf_reg_for_c_cut_5g[i],
+					      RFREG_OFFSET_MASK,
+					      rf_reg_pram_c_5g[index][i]);
+			}
+			rtl_dbg(rtlpriv, COMP_RF, DBG_TRACE,
+				"offset 0x%x value 0x%x path %d index %d readback 0x%x\n",
+				rf_reg_for_c_cut_5g[i],
+				rf_reg_pram_c_5g[index][i],
+				path, index,
+				rtl_get_rfreg(hw, (enum radio_path)path,
+					      rf_reg_for_c_cut_5g[i],
+					      RFREG_OFFSET_MASK));
+		}
+		if (rtlhal->macphymode == DUALMAC_DUALPHY &&
+		    rtlhal->interfaceindex == 1) {
+			if (need_pwr_down)
+				rtl92d_phy_restore_rf_env(hw, path, &u4regvalue);
+
+			rtl92du_phy_powerdown_anotherphy(hw, false);
+		}
+
+		if (regb30 && rtlhal->interfaceindex == 0) {
+			if (need_pwr_down)
+				rtl92d_phy_restore_rf_env(hw, path, &u4regvalue);
+
+			rtl92du_phy_powerdown_anotherphy(hw, true);
+		}
+
+		if (channel < 149)
+			value = 0x07;
+		else if (channel >= 149)
+			value = 0x02;
+		if (channel >= 36 && channel <= 64)
+			index = 0;
+		else if (channel >= 100 && channel <= 140)
+			index = 1;
+		else
+			index = 2;
+
+		for (rfpath = RF90_PATH_A; rfpath < rtlphy->num_total_rfpath;
+			rfpath++) {
+			if (rtlhal->macphymode == DUALMAC_DUALPHY &&
+			    rtlhal->interfaceindex == 1) /* MAC 1 5G */
+				internal_pa = rtlpriv->efuse.internal_pa_5g[1];
+			else
+				internal_pa =
+					 rtlpriv->efuse.internal_pa_5g[rfpath];
+
+			if (internal_pa) {
+				for (i = 0;
+				     i < RF_REG_NUM_FOR_C_CUT_5G_INTERNALPA;
+				     i++) {
+					if (rf_for_c_cut_5g_internal_pa[i] == 0x03 &&
+					    channel >= 36 && channel <= 64)
+						rtl_set_rfreg(hw, rfpath,
+							rf_for_c_cut_5g_internal_pa[i],
+							RFREG_OFFSET_MASK,
+							0x7bdef);
+					else
+						rtl_set_rfreg(hw, rfpath,
+							rf_for_c_cut_5g_internal_pa[i],
+							RFREG_OFFSET_MASK,
+							rf_pram_c_5g_int_pa[index][i]);
+					rtl_dbg(rtlpriv, COMP_RF, DBG_LOUD,
+						"offset 0x%x value 0x%x path %d index %d\n",
+						rf_for_c_cut_5g_internal_pa[i],
+						rf_pram_c_5g_int_pa[index][i],
+						rfpath, index);
+				}
+			} else {
+				rtl_set_rfreg(hw, (enum radio_path)rfpath, RF_TXPA_AG,
+					      mask, value);
+			}
+		}
+	} else if (rtlhal->current_bandtype == BAND_ON_2_4G) {
+		rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "====>2.4G\n");
+		u4tmp = rtlpriv->curveindex_2g[channel - 1];
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"ver 3 set RF-B, 2G, 0x28 = 0x%x !!\n", u4tmp);
+
+		if (channel == 1 || channel == 2 || channel == 4 ||
+		    channel == 9 || channel == 10 || channel == 11 ||
+		    channel == 12)
+			index = 0;
+		else if (channel == 3 || channel == 13 || channel == 14)
+			index = 1;
+		else if (channel >= 5 && channel <= 8)
+			index = 2;
+
+		if (rtlhal->macphymode == DUALMAC_DUALPHY) {
+			path = RF90_PATH_A;
+			if (rtlhal->interfaceindex == 0) {
+				need_pwr_down =
+					 rtl92du_phy_enable_anotherphy(hw, true);
+				rtlhal->during_mac0init_radiob = true;
+
+				if (need_pwr_down)
+					rtl92d_phy_enable_rf_env(hw, path,
+								 &u4regvalue);
+			}
+
+			/* DMDP, if band = 2G, MAC1 need to set PHY0 when regB30[27]=1 */
+			if (regb30 && rtlhal->interfaceindex == 1) {
+				need_pwr_down =
+					 rtl92du_phy_enable_anotherphy(hw, false);
+				rtlhal->during_mac1init_radioa = true;
+
+				if (need_pwr_down)
+					rtl92d_phy_enable_rf_env(hw, path,
+								 &u4regvalue);
+			}
+		}
+
+		for (i = 0; i < RF_REG_NUM_FOR_C_CUT_2G; i++) {
+			if (rf_reg_for_c_cut_2g[i] == RF_SYN_G7)
+				rtl_set_rfreg(hw, (enum radio_path)path,
+					      rf_reg_for_c_cut_2g[i],
+					      RFREG_OFFSET_MASK,
+					      rf_reg_param_for_c_cut_2g[index][i] |
+					      BIT(17));
+			else
+				rtl_set_rfreg(hw, (enum radio_path)path,
+					      rf_reg_for_c_cut_2g[i],
+					      RFREG_OFFSET_MASK,
+					      rf_reg_param_for_c_cut_2g
+					      [index][i]);
+
+			rtl_dbg(rtlpriv, COMP_RF, DBG_TRACE,
+				"offset 0x%x value 0x%x mak 0x%x path %d index %d readback 0x%x\n",
+				rf_reg_for_c_cut_2g[i],
+				rf_reg_param_for_c_cut_2g[index][i],
+				rf_reg_mask_for_c_cut_2g[i], path, index,
+				rtl_get_rfreg(hw, (enum radio_path)path,
+					      rf_reg_for_c_cut_2g[i],
+					      RFREG_OFFSET_MASK));
+		}
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"cosa ver 3 set RF-B, 2G, 0x28 = 0x%x !!\n",
+			rf_syn_g4_for_c_cut_2g | (u4tmp << 11));
+
+		rtl_set_rfreg(hw, (enum radio_path)path, RF_SYN_G4,
+			      RFREG_OFFSET_MASK,
+			      rf_syn_g4_for_c_cut_2g | (u4tmp << 11));
+
+		if (rtlhal->macphymode == DUALMAC_DUALPHY &&
+		    rtlhal->interfaceindex == 0) {
+			if (need_pwr_down)
+				rtl92d_phy_restore_rf_env(hw, path, &u4regvalue);
+
+			rtl92du_phy_powerdown_anotherphy(hw, true);
+		}
+
+		if (regb30 && rtlhal->interfaceindex == 1) {
+			if (need_pwr_down)
+				rtl92d_phy_restore_rf_env(hw, path, &u4regvalue);
+
+			rtl92du_phy_powerdown_anotherphy(hw, false);
+		}
+	}
+	rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "<====\n");
+}
+
+/* bit0 = 1 => Tx OK, bit1 = 1 => Rx OK */
+static u8 _rtl92du_phy_patha_iqk(struct ieee80211_hw *hw, bool configpathb)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw));
+	u32 regeac, rege94, rege9c, regea4;
+	u8 result = 0;
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "Path-A IQK setting!\n");
+
+	if (rtlhal->interfaceindex == 0) {
+		rtl_set_bbreg(hw, RTX_IQK_TONE_A, MASKDWORD, 0x10008c1f);
+		rtl_set_bbreg(hw, RRX_IQK_TONE_A, MASKDWORD, 0x10008c1f);
+	} else {
+		rtl_set_bbreg(hw, RTX_IQK_TONE_A, MASKDWORD, 0x10008c22);
+		rtl_set_bbreg(hw, RRX_IQK_TONE_A, MASKDWORD, 0x10008c22);
+	}
+	rtl_set_bbreg(hw, RTX_IQK_PI_A, MASKDWORD, 0x82140102);
+	rtl_set_bbreg(hw, RRX_IQK_PI_A, MASKDWORD,
+		      configpathb ? 0x28160202 : 0x28160502);
+	/* path-B IQK setting */
+	if (configpathb) {
+		rtl_set_bbreg(hw, RTX_IQK_TONE_B, MASKDWORD, 0x10008c22);
+		rtl_set_bbreg(hw, RRX_IQK_TONE_B, MASKDWORD, 0x10008c22);
+		rtl_set_bbreg(hw, RTX_IQK_PI_B, MASKDWORD, 0x82140102);
+		rtl_set_bbreg(hw, RRX_IQK_PI_B, MASKDWORD, 0x28160206);
+	}
+
+	/* LO calibration setting */
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "LO calibration setting!\n");
+	rtl_set_bbreg(hw, RIQK_AGC_RSP, MASKDWORD, 0x00462911);
+
+	/* One shot, path A LOK & IQK */
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "One shot, path A LOK & IQK!\n");
+	rtl_set_bbreg(hw, RIQK_AGC_PTS, MASKDWORD, 0xf9000000);
+	rtl_set_bbreg(hw, RIQK_AGC_PTS, MASKDWORD, 0xf8000000);
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"Delay %d ms for One shot, path A LOK & IQK\n",
+		IQK_DELAY_TIME);
+	mdelay(IQK_DELAY_TIME);
+
+	/* Check failed */
+	regeac = rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_A_2, MASKDWORD);
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xeac = 0x%x\n", regeac);
+	rege94 = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_A, MASKDWORD);
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xe94 = 0x%x\n", rege94);
+	rege9c = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_A, MASKDWORD);
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xe9c = 0x%x\n", rege9c);
+	regea4 = rtl_get_bbreg(hw, RRX_POWER_BEFORE_IQK_A_2, MASKDWORD);
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xea4 = 0x%x\n", regea4);
+
+	if (!(regeac & BIT(28)) &&
+	    (((rege94 & 0x03FF0000) >> 16) != 0x142) &&
+	    (((rege9c & 0x03FF0000) >> 16) != 0x42))
+		result |= 0x01;
+	else /* if Tx not OK, ignore Rx */
+		return result;
+
+	/* if Tx is OK, check whether Rx is OK */
+	if (!(regeac & BIT(27)) &&
+	    (((regea4 & 0x03FF0000) >> 16) != 0x132) &&
+	    (((regeac & 0x03FF0000) >> 16) != 0x36))
+		result |= 0x02;
+	else
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "Path A Rx IQK fail!!\n");
+
+	return result;
+}
+
+/* bit0 = 1 => Tx OK, bit1 = 1 => Rx OK */
+static u8 _rtl92du_phy_patha_iqk_5g_normal(struct ieee80211_hw *hw,
+					   bool configpathb)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	u32 TXOKBIT = BIT(28), RXOKBIT = BIT(27);
+	u32 regeac, rege94, rege9c, regea4;
+	u8 timeout = 20, timecount = 0;
+	u8 retrycount = 2;
+	u8 result = 0;
+	u8 i;
+
+	if (rtlhal->interfaceindex == 1) { /* PHY1 */
+		TXOKBIT = BIT(31);
+		RXOKBIT = BIT(30);
+	}
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "Path-A IQK setting!\n");
+	rtl_set_bbreg(hw, RTX_IQK_TONE_A, MASKDWORD, 0x18008c1f);
+	rtl_set_bbreg(hw, RRX_IQK_TONE_A, MASKDWORD, 0x18008c1f);
+	rtl_set_bbreg(hw, RTX_IQK_PI_A, MASKDWORD, 0x82140307);
+	rtl_set_bbreg(hw, RRX_IQK_PI_A, MASKDWORD, 0x68160960);
+	/* path-B IQK setting */
+	if (configpathb) {
+		rtl_set_bbreg(hw, RTX_IQK_TONE_B, MASKDWORD, 0x18008c2f);
+		rtl_set_bbreg(hw, RRX_IQK_TONE_B, MASKDWORD, 0x18008c2f);
+		rtl_set_bbreg(hw, RTX_IQK_PI_B, MASKDWORD, 0x82110000);
+		rtl_set_bbreg(hw, RRX_IQK_PI_B, MASKDWORD, 0x68110000);
+	}
+
+	/* LO calibration setting */
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "LO calibration setting!\n");
+	rtl_set_bbreg(hw, RIQK_AGC_RSP, MASKDWORD, 0x00462911);
+
+	/* path-A PA on */
+	rtl_set_bbreg(hw, RFPGA0_XAB_RFINTERFACESW, MASKDWORD, 0x07000f60);
+	rtl_set_bbreg(hw, RFPGA0_XA_RFINTERFACEOE, MASKDWORD, 0x66e60e30);
+
+	for (i = 0; i < retrycount; i++) {
+		/* One shot, path A LOK & IQK */
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"One shot, path A LOK & IQK!\n");
+		rtl_set_bbreg(hw, RIQK_AGC_PTS, MASKDWORD, 0xf9000000);
+		rtl_set_bbreg(hw, RIQK_AGC_PTS, MASKDWORD, 0xf8000000);
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"Delay %d ms for One shot, path A LOK & IQK.\n",
+			IQK_DELAY_TIME);
+		mdelay(IQK_DELAY_TIME * 10);
+
+		while (timecount < timeout &&
+		       rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_A_2, BIT(26)) == 0) {
+			udelay(IQK_DELAY_TIME * 1000 * 2);
+			timecount++;
+		}
+
+		timecount = 0;
+		while (timecount < timeout &&
+		       rtl_get_bbreg(hw, RRX_POWER_BEFORE_IQK_A_2, MASK_IQK_RESULT) == 0) {
+			udelay(IQK_DELAY_TIME * 1000 * 2);
+			timecount++;
+		}
+
+		/* Check failed */
+		regeac = rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_A_2, MASKDWORD);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xeac = 0x%x\n", regeac);
+		rege94 = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_A, MASKDWORD);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xe94 = 0x%x\n", rege94);
+		rege9c = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_A, MASKDWORD);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xe9c = 0x%x\n", rege9c);
+		regea4 = rtl_get_bbreg(hw, RRX_POWER_BEFORE_IQK_A_2, MASKDWORD);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xea4 = 0x%x\n", regea4);
+
+		if (!(regeac & TXOKBIT) &&
+		    (((rege94 & 0x03FF0000) >> 16) != 0x142)) {
+			result |= 0x01;
+		} else { /* if Tx not OK, ignore Rx */
+			RTPRINT(rtlpriv, FINIT, INIT_IQK,
+				"Path A Tx IQK fail!!\n");
+			continue;
+		}
+
+		/* if Tx is OK, check whether Rx is OK */
+		if (!(regeac & RXOKBIT) &&
+		    (((regea4 & 0x03FF0000) >> 16) != 0x132)) {
+			result |= 0x02;
+			break;
+		}
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "Path A Rx IQK fail!!\n");
+	}
+
+	/* path A PA off */
+	rtl_set_bbreg(hw, RFPGA0_XAB_RFINTERFACESW, MASKDWORD,
+		      rtlphy->iqk_bb_backup[0]);
+	rtl_set_bbreg(hw, RFPGA0_XA_RFINTERFACEOE, MASKDWORD,
+		      rtlphy->iqk_bb_backup[1]);
+
+	if (!(result & 0x01)) /* Tx IQK fail */
+		rtl_set_bbreg(hw, RTX_IQK_TONE_A, MASKDWORD, 0x19008c00);
+
+	if (!(result & 0x02)) { /* Rx IQK fail */
+		rtl_set_bbreg(hw, ROFDM0_XARXIQIMBALANCE, MASKDWORD, 0x40000100);
+		rtl_set_bbreg(hw, RRX_IQK_TONE_A, MASKDWORD, 0x19008c00);
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"Path A Rx IQK fail!! 0xe34 = %#x\n",
+			rtl_get_bbreg(hw, RRX_IQK_TONE_A, MASKDWORD));
+	}
+
+	return result;
+}
+
+/* bit0 = 1 => Tx OK, bit1 = 1 => Rx OK */
+static u8 _rtl92du_phy_pathb_iqk(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u32 regeac, regeb4, regebc, regec4, regecc;
+	u8 result = 0;
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "One shot, path B LOK & IQK!\n");
+	rtl_set_bbreg(hw, RIQK_AGC_CONT, MASKDWORD, 0x00000002);
+	rtl_set_bbreg(hw, RIQK_AGC_CONT, MASKDWORD, 0x00000000);
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"Delay %d ms for One shot, path B LOK & IQK\n", IQK_DELAY_TIME);
+	mdelay(IQK_DELAY_TIME);
+
+	/* Check failed */
+	regeac = rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_A_2, MASKDWORD);
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xeac = 0x%x\n", regeac);
+	regeb4 = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_B, MASKDWORD);
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xeb4 = 0x%x\n", regeb4);
+	regebc = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_B, MASKDWORD);
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xebc = 0x%x\n", regebc);
+	regec4 = rtl_get_bbreg(hw, RRX_POWER_BEFORE_IQK_B_2, MASKDWORD);
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xec4 = 0x%x\n", regec4);
+	regecc = rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_B_2, MASKDWORD);
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xecc = 0x%x\n", regecc);
+
+	if (!(regeac & BIT(31)) &&
+	    (((regeb4 & 0x03FF0000) >> 16) != 0x142) &&
+	    (((regebc & 0x03FF0000) >> 16) != 0x42))
+		result |= 0x01;
+	else
+		return result;
+
+	if (!(regeac & BIT(30)) &&
+	    (((regec4 & 0x03FF0000) >> 16) != 0x132) &&
+	    (((regecc & 0x03FF0000) >> 16) != 0x36))
+		result |= 0x02;
+	else
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "Path B Rx IQK fail!!\n");
+
+	return result;
+}
+
+/* bit0 = 1 => Tx OK, bit1 = 1 => Rx OK */
+static u8 _rtl92du_phy_pathb_iqk_5g_normal(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	u32 regeac, regeb4, regebc, regec4, regecc;
+	u8 timeout = 20, timecount = 0;
+	u8 retrycount = 2;
+	u8 result = 0;
+	u8 i;
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "Path-B IQK setting!\n");
+	rtl_set_bbreg(hw, RTX_IQK_TONE_A, MASKDWORD, 0x18008c1f);
+	rtl_set_bbreg(hw, RRX_IQK_TONE_A, MASKDWORD, 0x18008c1f);
+	rtl_set_bbreg(hw, RTX_IQK_PI_A, MASKDWORD, 0x82110000);
+	rtl_set_bbreg(hw, RRX_IQK_PI_A, MASKDWORD, 0x68110000);
+
+	/* path-B IQK setting */
+	rtl_set_bbreg(hw, RTX_IQK_TONE_B, MASKDWORD, 0x18008c2f);
+	rtl_set_bbreg(hw, RRX_IQK_TONE_B, MASKDWORD, 0x18008c2f);
+	rtl_set_bbreg(hw, RTX_IQK_PI_B, MASKDWORD, 0x82140307);
+	rtl_set_bbreg(hw, RRX_IQK_PI_B, MASKDWORD, 0x68160960);
+
+	/* LO calibration setting */
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "LO calibration setting!\n");
+	rtl_set_bbreg(hw, RIQK_AGC_RSP, MASKDWORD, 0x00462911);
+
+	/* path-B PA on */
+	rtl_set_bbreg(hw, RFPGA0_XAB_RFINTERFACESW, MASKDWORD, 0x0f600700);
+	rtl_set_bbreg(hw, RFPGA0_XB_RFINTERFACEOE, MASKDWORD, 0x061f0d30);
+
+	for (i = 0; i < retrycount; i++) {
+		/* One shot, path B LOK & IQK */
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"One shot, path A LOK & IQK!\n");
+		rtl_set_bbreg(hw, RIQK_AGC_PTS, MASKDWORD, 0xfa000000);
+		rtl_set_bbreg(hw, RIQK_AGC_PTS, MASKDWORD, 0xf8000000);
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"Delay %d ms for One shot, path B LOK & IQK.\n", 10);
+		mdelay(IQK_DELAY_TIME * 10);
+
+		while (timecount < timeout &&
+		       rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_A_2, BIT(29)) == 0) {
+			udelay(IQK_DELAY_TIME * 1000 * 2);
+			timecount++;
+		}
+
+		timecount = 0;
+		while (timecount < timeout &&
+		       rtl_get_bbreg(hw, RRX_POWER_BEFORE_IQK_B_2, MASK_IQK_RESULT) == 0) {
+			udelay(IQK_DELAY_TIME * 1000 * 2);
+			timecount++;
+		}
+
+		/* Check failed */
+		regeac = rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_A_2, MASKDWORD);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xeac = 0x%x\n", regeac);
+		regeb4 = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_B, MASKDWORD);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xeb4 = 0x%x\n", regeb4);
+		regebc = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_B, MASKDWORD);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xebc = 0x%x\n", regebc);
+		regec4 = rtl_get_bbreg(hw, RRX_POWER_BEFORE_IQK_B_2, MASKDWORD);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xec4 = 0x%x\n", regec4);
+		regecc = rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_B_2, MASKDWORD);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "0xecc = 0x%x\n", regecc);
+
+		if (!(regeac & BIT(31)) &&
+		    (((regeb4 & 0x03FF0000) >> 16) != 0x142))
+			result |= 0x01;
+		else
+			continue;
+
+		if (!(regeac & BIT(30)) &&
+		    (((regec4 & 0x03FF0000) >> 16) != 0x132)) {
+			result |= 0x02;
+			break;
+		}
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "Path B Rx IQK fail!!\n");
+	}
+
+	/* path B PA off */
+	rtl_set_bbreg(hw, RFPGA0_XAB_RFINTERFACESW, MASKDWORD,
+		      rtlphy->iqk_bb_backup[0]);
+	rtl_set_bbreg(hw, RFPGA0_XB_RFINTERFACEOE, MASKDWORD,
+		      rtlphy->iqk_bb_backup[2]);
+
+	if (!(result & 0x01))
+		rtl_set_bbreg(hw, RTX_IQK_TONE_B, MASKDWORD, 0x19008c00);
+
+	if (!(result & 0x02)) {
+		rtl_set_bbreg(hw, ROFDM0_XBRXIQIMBALANCE, MASKDWORD, 0x40000100);
+		rtl_set_bbreg(hw, RRX_IQK_TONE_B, MASKDWORD, 0x19008c00);
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"Path B Rx IQK fail!! 0xe54 = %#x\n",
+			rtl_get_bbreg(hw, RRX_IQK_TONE_B, MASKDWORD));
+	}
+
+	return result;
+}
+
+static void _rtl92du_phy_reload_adda_registers(struct ieee80211_hw *hw,
+					       const u32 *adda_reg,
+					       u32 *adda_backup, u32 regnum)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u32 i;
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"Reload ADDA power saving parameters !\n");
+	for (i = 0; i < regnum; i++) {
+		/* path-A/B BB to initial gain */
+		if (adda_reg[i] == ROFDM0_XAAGCCORE1 ||
+		    adda_reg[i] == ROFDM0_XBAGCCORE1)
+			rtl_set_bbreg(hw, adda_reg[i], MASKDWORD, 0x50);
+
+		rtl_set_bbreg(hw, adda_reg[i], MASKDWORD, adda_backup[i]);
+	}
+}
+
+static void _rtl92du_phy_reload_mac_registers(struct ieee80211_hw *hw,
+					      const u32 *macreg, u32 *macbackup)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u32 i;
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK, "Reload MAC parameters !\n");
+	for (i = 0; i < (IQK_MAC_REG_NUM - 1); i++)
+		rtl_write_byte(rtlpriv, macreg[i], (u8)macbackup[i]);
+	rtl_write_dword(rtlpriv, macreg[i], macbackup[i]);
+}
+
+static void _rtl92du_phy_patha_standby(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK, "Path-A standby mode!\n");
+
+	rtl_set_bbreg(hw, RFPGA0_IQK, MASKH3BYTES, 0x0);
+	rtl_set_bbreg(hw, RFPGA0_XA_LSSIPARAMETER, MASKDWORD, 0x00010000);
+	rtl_set_bbreg(hw, RFPGA0_IQK, MASKH3BYTES, 0x808000);
+}
+
+static void _rtl92du_phy_pimode_switch(struct ieee80211_hw *hw, bool pi_mode)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u32 mode;
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"BB Switch to %s mode!\n", pi_mode ? "PI" : "SI");
+	mode = pi_mode ? 0x01000100 : 0x01000000;
+	rtl_set_bbreg(hw, RFPGA0_XA_HSSIPARAMETER1, MASKDWORD, mode);
+	rtl_set_bbreg(hw, RFPGA0_XB_HSSIPARAMETER1, MASKDWORD, mode);
+}
+
+static void _rtl92du_phy_iq_calibrate(struct ieee80211_hw *hw, long result[][8],
+				      u8 t, bool is2t)
+{
+	static const u32 adda_reg[IQK_ADDA_REG_NUM] = {
+		RFPGA0_XCD_SWITCHCONTROL, RBLUE_TOOTH, RRX_WAIT_CCA,
+		RTX_CCK_RFON, RTX_CCK_BBON, RTX_OFDM_RFON, RTX_OFDM_BBON,
+		RTX_TO_RX, RTX_TO_TX, RRX_CCK, RRX_OFDM, RRX_WAIT_RIFS,
+		RRX_TO_RX, RSTANDBY, RSLEEP, RPMPD_ANAEN
+	};
+	static const u32 iqk_mac_reg[IQK_MAC_REG_NUM] = {
+		REG_TXPAUSE, REG_BCN_CTRL, REG_BCN_CTRL_1, REG_GPIO_MUXCFG
+	};
+	static const u32 iqk_bb_reg[IQK_BB_REG_NUM] = {
+		RFPGA0_XAB_RFINTERFACESW, RFPGA0_XA_RFINTERFACEOE,
+		RFPGA0_XB_RFINTERFACEOE, ROFDM0_TRMUXPAR,
+		RFPGA0_XCD_RFINTERFACESW, ROFDM0_TRXPATHENABLE,
+		RFPGA0_RFMOD, RFPGA0_ANALOGPARAMETER4,
+		ROFDM0_XAAGCCORE1, ROFDM0_XBAGCCORE1
+	};
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	const u32 retrycount = 2;
+	u8 patha_ok, pathb_ok;
+	u32 bbvalue;
+	u32 i;
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "IQK for 2.4G :Start!!!\n");
+	if (t == 0) {
+		bbvalue = rtl_get_bbreg(hw, RFPGA0_RFMOD, MASKDWORD);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "==>0x%08x\n", bbvalue);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "IQ Calibration for %s\n",
+			is2t ? "2T2R" : "1T1R");
+
+		/*  Save ADDA parameters, turn Path A ADDA on */
+		rtl92d_phy_save_adda_registers(hw, adda_reg,
+					       rtlphy->adda_backup,
+					       IQK_ADDA_REG_NUM);
+		rtl92d_phy_save_mac_registers(hw, iqk_mac_reg,
+					      rtlphy->iqk_mac_backup);
+		rtl92d_phy_save_adda_registers(hw, iqk_bb_reg,
+					       rtlphy->iqk_bb_backup,
+					       IQK_BB_REG_NUM);
+	}
+	rtl92d_phy_path_adda_on(hw, adda_reg, true, is2t);
+
+	rtl_set_bbreg(hw, RPDP_ANTA, MASKDWORD, 0x01017038);
+
+	if (t == 0)
+		rtlphy->rfpi_enable = (u8)rtl_get_bbreg(hw,
+				RFPGA0_XA_HSSIPARAMETER1, BIT(8));
+
+	/*  Switch BB to PI mode to do IQ Calibration. */
+	if (!rtlphy->rfpi_enable)
+		_rtl92du_phy_pimode_switch(hw, true);
+
+	rtl92du_phy_set_bb_reg_1byte(hw, RFPGA0_RFMOD, BCCKEN, 0x00);
+	rtl_set_bbreg(hw, ROFDM0_TRXPATHENABLE, MASKDWORD, 0x03a05600);
+	rtl_set_bbreg(hw, ROFDM0_TRMUXPAR, MASKDWORD, 0x000800e4);
+	rtl_set_bbreg(hw, RFPGA0_XCD_RFINTERFACESW, MASKDWORD, 0x22204000);
+	rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER4, 0xf00000, 0x0f);
+	if (is2t) {
+		rtl_set_bbreg(hw, RFPGA0_XA_LSSIPARAMETER, MASKDWORD,
+			      0x00010000);
+		rtl_set_bbreg(hw, RFPGA0_XB_LSSIPARAMETER, MASKDWORD,
+			      0x00010000);
+	}
+
+	/* MAC settings */
+	rtl92d_phy_mac_setting_calibration(hw, iqk_mac_reg,
+					   rtlphy->iqk_mac_backup);
+
+	/* Page B init */
+	rtl_set_bbreg(hw, RCONFIG_ANTA, MASKDWORD, 0x0f600000);
+	if (is2t)
+		rtl_set_bbreg(hw, RCONFIG_ANTB, MASKDWORD, 0x0f600000);
+
+	/* IQ calibration setting */
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "IQK setting!\n");
+	rtl_set_bbreg(hw, RFPGA0_IQK, MASKH3BYTES, 0x808000);
+	rtl_set_bbreg(hw, RTX_IQK, MASKDWORD, 0x01007c00);
+	rtl_set_bbreg(hw, RRX_IQK, MASKDWORD, 0x01004800);
+
+	for (i = 0; i < retrycount; i++) {
+		patha_ok = _rtl92du_phy_patha_iqk(hw, is2t);
+		if (patha_ok == 0x03) {
+			RTPRINT(rtlpriv, FINIT, INIT_IQK,
+				"Path A IQK Success!!\n");
+			result[t][0] = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_A,
+						     MASK_IQK_RESULT);
+			result[t][1] = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_A,
+						     MASK_IQK_RESULT);
+			result[t][2] = rtl_get_bbreg(hw, RRX_POWER_BEFORE_IQK_A_2,
+						     MASK_IQK_RESULT);
+			result[t][3] = rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_A_2,
+						     MASK_IQK_RESULT);
+			break;
+		} else if (i == (retrycount - 1) && patha_ok == 0x01) {
+			/* Tx IQK OK */
+			RTPRINT(rtlpriv, FINIT, INIT_IQK,
+				"Path A IQK Only  Tx Success!!\n");
+
+			result[t][0] = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_A,
+						     MASK_IQK_RESULT);
+			result[t][1] = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_A,
+						     MASK_IQK_RESULT);
+		}
+	}
+	if (patha_ok == 0x00)
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "Path A IQK failed!!\n");
+
+	if (is2t) {
+		_rtl92du_phy_patha_standby(hw);
+		/* Turn Path B ADDA on */
+		rtl92d_phy_path_adda_on(hw, adda_reg, false, is2t);
+
+		for (i = 0; i < retrycount; i++) {
+			pathb_ok = _rtl92du_phy_pathb_iqk(hw);
+			if (pathb_ok == 0x03) {
+				RTPRINT(rtlpriv, FINIT, INIT_IQK,
+					"Path B IQK Success!!\n");
+				result[t][4] = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_B,
+							     MASK_IQK_RESULT);
+				result[t][5] = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_B,
+							     MASK_IQK_RESULT);
+				result[t][6] = rtl_get_bbreg(hw, RRX_POWER_BEFORE_IQK_B_2,
+							     MASK_IQK_RESULT);
+				result[t][7] = rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_B_2,
+							     MASK_IQK_RESULT);
+				break;
+			} else if (i == (retrycount - 1) && pathb_ok == 0x01) {
+				/* Tx IQK OK */
+				RTPRINT(rtlpriv, FINIT, INIT_IQK,
+					"Path B Only Tx IQK Success!!\n");
+				result[t][4] = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_B,
+							     MASK_IQK_RESULT);
+				result[t][5] = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_B,
+							     MASK_IQK_RESULT);
+			}
+		}
+		if (pathb_ok == 0x00)
+			RTPRINT(rtlpriv, FINIT, INIT_IQK,
+				"Path B IQK failed!!\n");
+	}
+
+	/* Back to BB mode, load original value */
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"IQK:Back to BB mode, load original value!\n");
+
+	rtl_set_bbreg(hw, RFPGA0_IQK, MASKH3BYTES, 0x000000);
+
+	if (t != 0) {
+		/* Switch back BB to SI mode after finish IQ Calibration. */
+		if (!rtlphy->rfpi_enable)
+			_rtl92du_phy_pimode_switch(hw, false);
+
+		/* Reload ADDA power saving parameters */
+		_rtl92du_phy_reload_adda_registers(hw, adda_reg,
+						   rtlphy->adda_backup,
+						   IQK_ADDA_REG_NUM);
+
+		/* Reload MAC parameters */
+		_rtl92du_phy_reload_mac_registers(hw, iqk_mac_reg,
+						  rtlphy->iqk_mac_backup);
+
+		if (is2t)
+			_rtl92du_phy_reload_adda_registers(hw, iqk_bb_reg,
+							   rtlphy->iqk_bb_backup,
+							   IQK_BB_REG_NUM);
+		else
+			_rtl92du_phy_reload_adda_registers(hw, iqk_bb_reg,
+							   rtlphy->iqk_bb_backup,
+							   IQK_BB_REG_NUM - 1);
+
+		/* load 0xe30 IQC default value */
+		rtl_set_bbreg(hw, RTX_IQK_TONE_A, MASKDWORD, 0x01008c00);
+		rtl_set_bbreg(hw, RRX_IQK_TONE_A, MASKDWORD, 0x01008c00);
+	}
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "<==\n");
+}
+
+static void _rtl92du_phy_iq_calibrate_5g_normal(struct ieee80211_hw *hw,
+						long result[][8], u8 t)
+{
+	static const u32 adda_reg[IQK_ADDA_REG_NUM] = {
+		RFPGA0_XCD_SWITCHCONTROL, RBLUE_TOOTH, RRX_WAIT_CCA,
+		RTX_CCK_RFON, RTX_CCK_BBON, RTX_OFDM_RFON, RTX_OFDM_BBON,
+		RTX_TO_RX, RTX_TO_TX, RRX_CCK, RRX_OFDM, RRX_WAIT_RIFS,
+		RRX_TO_RX, RSTANDBY, RSLEEP, RPMPD_ANAEN
+	};
+	static const u32 iqk_mac_reg[IQK_MAC_REG_NUM] = {
+		REG_TXPAUSE, REG_BCN_CTRL, REG_BCN_CTRL_1, REG_GPIO_MUXCFG
+	};
+	static const u32 iqk_bb_reg[IQK_BB_REG_NUM] = {
+		RFPGA0_XAB_RFINTERFACESW, RFPGA0_XA_RFINTERFACEOE,
+		RFPGA0_XB_RFINTERFACEOE, ROFDM0_TRMUXPAR,
+		RFPGA0_XCD_RFINTERFACESW, ROFDM0_TRXPATHENABLE,
+		RFPGA0_RFMOD, RFPGA0_ANALOGPARAMETER4,
+		ROFDM0_XAAGCCORE1, ROFDM0_XBAGCCORE1
+	};
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	bool is2t = IS_92D_SINGLEPHY(rtlhal->version);
+	u8 patha_ok, pathb_ok;
+	bool rf_path_div;
+	u32 bbvalue;
+
+	/* Note: IQ calibration must be performed after loading
+	 * PHY_REG.txt , and radio_a, radio_b.txt
+	 */
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "IQK for 5G NORMAL:Start!!!\n");
+
+	mdelay(IQK_DELAY_TIME * 20);
+
+	if (t == 0) {
+		bbvalue = rtl_get_bbreg(hw, RFPGA0_RFMOD, MASKDWORD);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "==>0x%08x\n", bbvalue);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "IQ Calibration for %s\n",
+			is2t ? "2T2R" : "1T1R");
+
+		/* Save ADDA parameters, turn Path A ADDA on */
+		rtl92d_phy_save_adda_registers(hw, adda_reg,
+					       rtlphy->adda_backup,
+					       IQK_ADDA_REG_NUM);
+		rtl92d_phy_save_mac_registers(hw, iqk_mac_reg,
+					      rtlphy->iqk_mac_backup);
+		if (is2t)
+			rtl92d_phy_save_adda_registers(hw, iqk_bb_reg,
+						       rtlphy->iqk_bb_backup,
+						       IQK_BB_REG_NUM);
+		else
+			rtl92d_phy_save_adda_registers(hw, iqk_bb_reg,
+						       rtlphy->iqk_bb_backup,
+						       IQK_BB_REG_NUM - 1);
+	}
+
+	rf_path_div = rtl_get_bbreg(hw, 0xb30, BIT(27));
+	rtl92d_phy_path_adda_on(hw, adda_reg, !rf_path_div, is2t);
+
+	if (t == 0)
+		rtlphy->rfpi_enable = rtl_get_bbreg(hw,
+						    RFPGA0_XA_HSSIPARAMETER1,
+						    BIT(8));
+
+	/*  Switch BB to PI mode to do IQ Calibration. */
+	if (!rtlphy->rfpi_enable)
+		_rtl92du_phy_pimode_switch(hw, true);
+
+	/* MAC settings */
+	rtl92d_phy_mac_setting_calibration(hw, iqk_mac_reg,
+					   rtlphy->iqk_mac_backup);
+
+	rtl92du_phy_set_bb_reg_1byte(hw, RFPGA0_RFMOD, BCCKEN, 0x00);
+	rtl_set_bbreg(hw, ROFDM0_TRXPATHENABLE, MASKDWORD, 0x03a05600);
+	rtl_set_bbreg(hw, ROFDM0_TRMUXPAR, MASKDWORD, 0x000800e4);
+	rtl_set_bbreg(hw, RFPGA0_XCD_RFINTERFACESW, MASKDWORD, 0x22208000);
+	rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER4, 0xf00000, 0x0f);
+
+	/* Page A AP setting for IQK */
+	rtl_set_bbreg(hw, RPDP_ANTA, MASKDWORD, 0);
+	rtl_set_bbreg(hw, RCONFIG_ANTA, MASKDWORD, 0x20000000);
+	if (is2t) {
+		/* Page B AP setting for IQK */
+		rtl_set_bbreg(hw, RPDP_ANTB, MASKDWORD, 0);
+		rtl_set_bbreg(hw, RCONFIG_ANTB, MASKDWORD, 0x20000000);
+	}
+
+	/* IQ calibration setting  */
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "IQK setting!\n");
+	rtl_set_bbreg(hw, RFPGA0_IQK, MASKH3BYTES, 0x808000);
+	rtl_set_bbreg(hw, RTX_IQK, MASKDWORD, 0x10007c00);
+	rtl_set_bbreg(hw, RRX_IQK, MASKDWORD, 0x01004800);
+
+	patha_ok = _rtl92du_phy_patha_iqk_5g_normal(hw, is2t);
+	if (patha_ok == 0x03) {
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "Path A IQK Success!!\n");
+		result[t][0] = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_A,
+					     MASK_IQK_RESULT);
+		result[t][1] = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_A,
+					     MASK_IQK_RESULT);
+		result[t][2] = rtl_get_bbreg(hw, RRX_POWER_BEFORE_IQK_A_2,
+					     MASK_IQK_RESULT);
+		result[t][3] = rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_A_2,
+					     MASK_IQK_RESULT);
+	} else if (patha_ok == 0x01) {	/* Tx IQK OK */
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"Path A IQK Only  Tx Success!!\n");
+
+		result[t][0] = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_A,
+					     MASK_IQK_RESULT);
+		result[t][1] = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_A,
+					     MASK_IQK_RESULT);
+	} else {
+		rtl_set_bbreg(hw, RFPGA0_IQK, MASKH3BYTES, 0x000000);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "0xe70 = %#x\n",
+			rtl_get_bbreg(hw, RRX_WAIT_CCA, MASKDWORD));
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "RF path A 0x0 = %#x\n",
+			rtl_get_rfreg(hw, RF90_PATH_A, RF_AC, RFREG_OFFSET_MASK));
+		rtl_set_bbreg(hw, RFPGA0_IQK, MASKH3BYTES, 0x808000);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "Path A IQK Fail!!\n");
+	}
+
+	if (is2t) {
+		/* _rtl92d_phy_patha_standby(hw); */
+		/* Turn Path B ADDA on  */
+		rtl92d_phy_path_adda_on(hw, adda_reg, false, is2t);
+
+		pathb_ok = _rtl92du_phy_pathb_iqk_5g_normal(hw);
+		if (pathb_ok == 0x03) {
+			RTPRINT(rtlpriv, FINIT, INIT_IQK,
+				"Path B IQK Success!!\n");
+			result[t][4] = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_B,
+						     MASK_IQK_RESULT);
+			result[t][5] = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_B,
+						     MASK_IQK_RESULT);
+			result[t][6] = rtl_get_bbreg(hw, RRX_POWER_BEFORE_IQK_B_2,
+						     MASK_IQK_RESULT);
+			result[t][7] = rtl_get_bbreg(hw, RRX_POWER_AFTER_IQK_B_2,
+						     MASK_IQK_RESULT);
+		} else if (pathb_ok == 0x01) { /* Tx IQK OK */
+			RTPRINT(rtlpriv, FINIT, INIT_IQK,
+				"Path B Only Tx IQK Success!!\n");
+			result[t][4] = rtl_get_bbreg(hw, RTX_POWER_BEFORE_IQK_B,
+						     MASK_IQK_RESULT);
+			result[t][5] = rtl_get_bbreg(hw, RTX_POWER_AFTER_IQK_B,
+						     MASK_IQK_RESULT);
+		} else {
+			RTPRINT(rtlpriv, FINIT, INIT_IQK,
+				"Path B IQK failed!!\n");
+		}
+	}
+
+	/* Back to BB mode, load original value */
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"IQK:Back to BB mode, load original value!\n");
+	rtl_set_bbreg(hw, RFPGA0_IQK, MASKH3BYTES, 0);
+
+	if (is2t)
+		_rtl92du_phy_reload_adda_registers(hw, iqk_bb_reg,
+						   rtlphy->iqk_bb_backup,
+						   IQK_BB_REG_NUM);
+	else
+		_rtl92du_phy_reload_adda_registers(hw, iqk_bb_reg,
+						   rtlphy->iqk_bb_backup,
+						   IQK_BB_REG_NUM - 1);
+
+	/* path A IQ path to DP block */
+	rtl_set_bbreg(hw, RPDP_ANTA, MASKDWORD, 0x010170b8);
+	if (is2t) /* path B IQ path to DP block */
+		rtl_set_bbreg(hw, RPDP_ANTB, MASKDWORD, 0x010170b8);
+
+	/* Reload MAC parameters */
+	_rtl92du_phy_reload_mac_registers(hw, iqk_mac_reg,
+					  rtlphy->iqk_mac_backup);
+
+	/* Switch back BB to SI mode after finish IQ Calibration. */
+	if (!rtlphy->rfpi_enable)
+		_rtl92du_phy_pimode_switch(hw, false);
+
+	/* Reload ADDA power saving parameters */
+	_rtl92du_phy_reload_adda_registers(hw, adda_reg,
+					   rtlphy->adda_backup,
+					   IQK_ADDA_REG_NUM);
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "<==\n");
+}
+
+static bool _rtl92du_phy_simularity_compare(struct ieee80211_hw *hw,
+					    long result[][8], u8 c1, u8 c2)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	u32 i, j, diff, sim_bitmap, bound, u4temp = 0;
+	u8 final_candidate[2] = {0xFF, 0xFF};	/* for path A and path B */
+	bool is2t = IS_92D_SINGLEPHY(rtlhal->version);
+	bool bresult = true;
+
+	if (is2t)
+		bound = 8;
+	else
+		bound = 4;
+
+	sim_bitmap = 0;
+
+	for (i = 0; i < bound; i++) {
+		diff = abs_diff(result[c1][i], result[c2][i]);
+
+		if (diff > MAX_TOLERANCE_92D) {
+			if ((i == 2 || i == 6) && !sim_bitmap) {
+				if (result[c1][i] + result[c1][i + 1] == 0)
+					final_candidate[(i / 4)] = c2;
+				else if (result[c2][i] + result[c2][i + 1] == 0)
+					final_candidate[(i / 4)] = c1;
+				else
+					sim_bitmap = sim_bitmap | (1 << i);
+			} else {
+				sim_bitmap = sim_bitmap | (1 << i);
+			}
+		}
+	}
+
+	if (sim_bitmap == 0) {
+		for (i = 0; i < (bound / 4); i++) {
+			if (final_candidate[i] != 0xFF) {
+				for (j = i * 4; j < (i + 1) * 4 - 2; j++)
+					result[3][j] =
+						 result[final_candidate[i]][j];
+				bresult = false;
+			}
+		}
+
+		for (i = 0; i < bound; i++)
+			u4temp += result[c1][i] + result[c2][i];
+
+		if (u4temp == 0) /* IQK fail for c1 & c2 */
+			bresult = false;
+
+		return bresult;
+	}
+
+	if (!(sim_bitmap & 0x0F)) { /* path A OK */
+		for (i = 0; i < 4; i++)
+			result[3][i] = result[c1][i];
+	} else if (!(sim_bitmap & 0x03)) { /* path A, Tx OK */
+		for (i = 0; i < 2; i++)
+			result[3][i] = result[c1][i];
+	}
+
+	if (!(sim_bitmap & 0xF0) && is2t) { /* path B OK */
+		for (i = 4; i < 8; i++)
+			result[3][i] = result[c1][i];
+	} else if (!(sim_bitmap & 0x30)) { /* path B, Tx OK */
+		for (i = 4; i < 6; i++)
+			result[3][i] = result[c1][i];
+	}
+
+	return false;
+}
+
+static void _rtl92du_phy_patha_fill_iqk_matrix_5g_normal(struct ieee80211_hw *hw,
+							 bool iqk_ok,
+							 long result[][8],
+							 u8 final_candidate,
+							 bool txonly)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	u32 val_x, reg;
+	int val_y;
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"Path A IQ Calibration %s !\n", iqk_ok ? "Success" : "Failed");
+	if (iqk_ok && final_candidate != 0xFF) {
+		val_x = result[final_candidate][0];
+		if ((val_x & 0x00000200) != 0)
+			val_x = val_x | 0xFFFFFC00;
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "X = 0x%x\n", val_x);
+		rtl_set_bbreg(hw, RTX_IQK_TONE_A, 0x3FF0000, val_x);
+		rtl_set_bbreg(hw, ROFDM0_ECCATHRESHOLD, BIT(24), 0);
+
+		val_y = result[final_candidate][1];
+		if ((val_y & 0x00000200) != 0)
+			val_y = val_y | 0xFFFFFC00;
+
+		/* path B IQK result + 3 */
+		if (rtlhal->current_bandtype == BAND_ON_5G)
+			val_y += 3;
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "Y = 0x%x\n", val_y);
+
+		rtl_set_bbreg(hw, RTX_IQK_TONE_A, 0x3FF, val_y);
+		rtl_set_bbreg(hw, ROFDM0_ECCATHRESHOLD, BIT(26), 0);
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "0xe30 = 0x%x\n",
+			rtl_get_bbreg(hw, RTX_IQK_TONE_A, MASKDWORD));
+
+		if (txonly) {
+			RTPRINT(rtlpriv, FINIT, INIT_IQK, "only Tx OK\n");
+			return;
+		}
+
+		reg = result[final_candidate][2];
+		rtl_set_bbreg(hw, ROFDM0_XARXIQIMBALANCE, 0x3FF, reg);
+		reg = result[final_candidate][3] & 0x3F;
+		rtl_set_bbreg(hw, ROFDM0_XARXIQIMBALANCE, 0xFC00, reg);
+		reg = (result[final_candidate][3] >> 6) & 0xF;
+		rtl_set_bbreg(hw, ROFDM0_RXIQEXTANTA, 0xF0000000, reg);
+	} else {
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"%s: Tx/Rx fail restore default value\n", __func__);
+
+		rtl_set_bbreg(hw, RTX_IQK_TONE_A, MASKDWORD, 0x19008c00);
+		rtl_set_bbreg(hw, ROFDM0_XARXIQIMBALANCE, MASKDWORD, 0x40000100);
+		rtl_set_bbreg(hw, RRX_IQK_TONE_A, MASKDWORD, 0x19008c00);
+	}
+}
+
+static void _rtl92du_phy_patha_fill_iqk_matrix(struct ieee80211_hw *hw,
+					       bool iqk_ok, long result[][8],
+					       u8 final_candidate, bool txonly)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	u32 oldval_0, val_x, tx0_a, reg;
+	long val_y, tx0_c;
+	bool is2t = IS_92D_SINGLEPHY(rtlhal->version) ||
+		    rtlhal->macphymode == DUALMAC_DUALPHY;
+
+	if (rtlhal->current_bandtype == BAND_ON_5G) {
+		_rtl92du_phy_patha_fill_iqk_matrix_5g_normal(hw, iqk_ok, result,
+							     final_candidate,
+							     txonly);
+		return;
+	}
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"Path A IQ Calibration %s !\n", iqk_ok ? "Success" : "Failed");
+	if (final_candidate == 0xFF || !iqk_ok)
+		return;
+
+	/* OFDM0_D */
+	oldval_0 = rtl_get_bbreg(hw, ROFDM0_XATXIQIMBALANCE, 0xffc00000);
+
+	val_x = result[final_candidate][0];
+	if ((val_x & 0x00000200) != 0)
+		val_x = val_x | 0xFFFFFC00;
+
+	tx0_a = (val_x * oldval_0) >> 8;
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"X = 0x%x, tx0_a = 0x%x, oldval_0 0x%x\n",
+		val_x, tx0_a, oldval_0);
+	rtl_set_bbreg(hw, ROFDM0_XATXIQIMBALANCE, 0x3FF, tx0_a);
+	rtl_set_bbreg(hw, ROFDM0_ECCATHRESHOLD, BIT(24),
+		      ((val_x * oldval_0 >> 7) & 0x1));
+
+	val_y = result[final_candidate][1];
+	if ((val_y & 0x00000200) != 0)
+		val_y = val_y | 0xFFFFFC00;
+
+	/* path B IQK result + 3 */
+	if (rtlhal->interfaceindex == 1 &&
+	    rtlhal->current_bandtype == BAND_ON_5G)
+		val_y += 3;
+
+	tx0_c = (val_y * oldval_0) >> 8;
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"Y = 0x%lx, tx0_c = 0x%lx\n",
+		val_y, tx0_c);
+
+	rtl_set_bbreg(hw, ROFDM0_XCTXAFE, 0xF0000000, (tx0_c & 0x3C0) >> 6);
+	rtl_set_bbreg(hw, ROFDM0_XATXIQIMBALANCE, 0x003F0000, tx0_c & 0x3F);
+	if (is2t)
+		rtl_set_bbreg(hw, ROFDM0_ECCATHRESHOLD, BIT(26),
+			      (val_y * oldval_0 >> 7) & 0x1);
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK, "0xC80 = 0x%x\n",
+		rtl_get_bbreg(hw, ROFDM0_XATXIQIMBALANCE,
+			      MASKDWORD));
+
+	if (txonly) {
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "only Tx OK\n");
+		return;
+	}
+
+	reg = result[final_candidate][2];
+	rtl_set_bbreg(hw, ROFDM0_XARXIQIMBALANCE, 0x3FF, reg);
+	reg = result[final_candidate][3] & 0x3F;
+	rtl_set_bbreg(hw, ROFDM0_XARXIQIMBALANCE, 0xFC00, reg);
+	reg = (result[final_candidate][3] >> 6) & 0xF;
+	rtl_set_bbreg(hw, ROFDM0_RXIQEXTANTA, 0xF0000000, reg);
+}
+
+static void _rtl92du_phy_pathb_fill_iqk_matrix_5g_normal(struct ieee80211_hw *hw,
+							 bool iqk_ok,
+							 long result[][8],
+							 u8 final_candidate,
+							 bool txonly)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	u32 val_x, reg;
+	int val_y;
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"Path B IQ Calibration %s !\n", iqk_ok ? "Success" : "Failed");
+	if (iqk_ok && final_candidate != 0xFF) {
+		val_x = result[final_candidate][4];
+		if ((val_x & 0x00000200) != 0)
+			val_x = val_x | 0xFFFFFC00;
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "X = 0x%x\n", val_x);
+		rtl_set_bbreg(hw, RTX_IQK_TONE_B, 0x3FF0000, val_x);
+		rtl_set_bbreg(hw, ROFDM0_ECCATHRESHOLD, BIT(28), 0);
+
+		val_y = result[final_candidate][5];
+		if ((val_y & 0x00000200) != 0)
+			val_y = val_y | 0xFFFFFC00;
+
+		/* path B IQK result + 3 */
+		if (rtlhal->current_bandtype == BAND_ON_5G)
+			val_y += 3;
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "Y = 0x%x\n", val_y);
+
+		rtl_set_bbreg(hw, RTX_IQK_TONE_B, 0x3FF, val_y);
+		rtl_set_bbreg(hw, ROFDM0_ECCATHRESHOLD, BIT(30), 0);
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK, "0xe50 = 0x%x\n",
+			rtl_get_bbreg(hw, RTX_IQK_TONE_B, MASKDWORD));
+
+		if (txonly) {
+			RTPRINT(rtlpriv, FINIT, INIT_IQK, "only Tx OK\n");
+			return;
+		}
+
+		reg = result[final_candidate][6];
+		rtl_set_bbreg(hw, ROFDM0_XBRXIQIMBALANCE, 0x3FF, reg);
+		reg = result[final_candidate][7] & 0x3F;
+		rtl_set_bbreg(hw, ROFDM0_XBRXIQIMBALANCE, 0xFC00, reg);
+		reg = (result[final_candidate][7] >> 6) & 0xF;
+		rtl_set_bbreg(hw, ROFDM0_AGCRSSITABLE, 0x0000F000, reg);
+	} else {
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"%s: Tx/Rx fail restore default value\n", __func__);
+
+		rtl_set_bbreg(hw, RTX_IQK_TONE_B, MASKDWORD, 0x19008c00);
+		rtl_set_bbreg(hw, ROFDM0_XBRXIQIMBALANCE, MASKDWORD, 0x40000100);
+		rtl_set_bbreg(hw, RRX_IQK_TONE_B, MASKDWORD, 0x19008c00);
+	}
+}
+
+static void _rtl92du_phy_pathb_fill_iqk_matrix(struct ieee80211_hw *hw,
+					       bool iqk_ok, long result[][8],
+					       u8 final_candidate, bool txonly)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	u32 oldval_1, val_x, tx1_a, reg;
+	long val_y, tx1_c;
+
+	if (rtlhal->current_bandtype == BAND_ON_5G) {
+		_rtl92du_phy_pathb_fill_iqk_matrix_5g_normal(hw, iqk_ok, result,
+							     final_candidate,
+							     txonly);
+		return;
+	}
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK, "Path B IQ Calibration %s !\n",
+		iqk_ok ? "Success" : "Failed");
+
+	if (final_candidate == 0xFF || !iqk_ok)
+		return;
+
+	oldval_1 = rtl_get_bbreg(hw, ROFDM0_XBTXIQIMBALANCE, 0xffc00000);
+
+	val_x = result[final_candidate][4];
+	if ((val_x & 0x00000200) != 0)
+		val_x = val_x | 0xFFFFFC00;
+
+	tx1_a = (val_x * oldval_1) >> 8;
+	RTPRINT(rtlpriv, FINIT, INIT_IQK, "X = 0x%x, tx1_a = 0x%x\n",
+		val_x, tx1_a);
+	rtl_set_bbreg(hw, ROFDM0_XBTXIQIMBALANCE, 0x3FF, tx1_a);
+	rtl_set_bbreg(hw, ROFDM0_ECCATHRESHOLD, BIT(28),
+		      (val_x * oldval_1 >> 7) & 0x1);
+
+	val_y = result[final_candidate][5];
+	if ((val_y & 0x00000200) != 0)
+		val_y = val_y | 0xFFFFFC00;
+
+	if (rtlhal->current_bandtype == BAND_ON_5G)
+		val_y += 3;
+
+	tx1_c = (val_y * oldval_1) >> 8;
+	RTPRINT(rtlpriv, FINIT, INIT_IQK, "Y = 0x%lx, tx1_c = 0x%lx\n",
+		val_y, tx1_c);
+
+	rtl_set_bbreg(hw, ROFDM0_XDTXAFE, 0xF0000000, (tx1_c & 0x3C0) >> 6);
+	rtl_set_bbreg(hw, ROFDM0_XBTXIQIMBALANCE, 0x003F0000, tx1_c & 0x3F);
+	rtl_set_bbreg(hw, ROFDM0_ECCATHRESHOLD, BIT(30),
+		      (val_y * oldval_1 >> 7) & 0x1);
+
+	if (txonly)
+		return;
+
+	reg = result[final_candidate][6];
+	rtl_set_bbreg(hw, ROFDM0_XBRXIQIMBALANCE, 0x3FF, reg);
+	reg = result[final_candidate][7] & 0x3F;
+	rtl_set_bbreg(hw, ROFDM0_XBRXIQIMBALANCE, 0xFC00, reg);
+	reg = (result[final_candidate][7] >> 6) & 0xF;
+	rtl_set_bbreg(hw, ROFDM0_AGCRSSITABLE, 0x0000F000, reg);
+}
+
+void rtl92du_phy_iq_calibrate(struct ieee80211_hw *hw)
+{
+	long rege94, rege9c, regea4, regeac, regeb4;
+	bool is12simular, is13simular, is23simular;
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	long regebc, regec4, regecc, regtmp = 0;
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	u8 i, final_candidate, indexforchannel;
+	bool patha_ok, pathb_ok;
+	long result[4][8] = {};
+
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"IQK:Start!!!channel %d\n", rtlphy->current_channel);
+
+	final_candidate = 0xff;
+	patha_ok = false;
+	pathb_ok = false;
+	is12simular = false;
+	is23simular = false;
+	is13simular = false;
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"IQK !!!currentband %d\n", rtlhal->current_bandtype);
+
+	for (i = 0; i < 3; i++) {
+		if (rtlhal->current_bandtype == BAND_ON_5G) {
+			_rtl92du_phy_iq_calibrate_5g_normal(hw, result, i);
+		} else if (rtlhal->current_bandtype == BAND_ON_2_4G) {
+			if (IS_92D_SINGLEPHY(rtlhal->version))
+				_rtl92du_phy_iq_calibrate(hw, result, i, true);
+			else
+				_rtl92du_phy_iq_calibrate(hw, result, i, false);
+		}
+
+		if (i == 1) {
+			is12simular = _rtl92du_phy_simularity_compare(hw, result,
+								      0, 1);
+			if (is12simular) {
+				final_candidate = 0;
+				break;
+			}
+		}
+
+		if (i == 2) {
+			is13simular = _rtl92du_phy_simularity_compare(hw, result,
+								      0, 2);
+			if (is13simular) {
+				final_candidate = 0;
+				break;
+			}
+
+			is23simular = _rtl92du_phy_simularity_compare(hw, result,
+								      1, 2);
+			if (is23simular) {
+				final_candidate = 1;
+			} else {
+				for (i = 0; i < 8; i++)
+					regtmp += result[3][i];
+
+				if (regtmp != 0)
+					final_candidate = 3;
+				else
+					final_candidate = 0xFF;
+			}
+		}
+	}
+
+	for (i = 0; i < 4; i++) {
+		rege94 = result[i][0];
+		rege9c = result[i][1];
+		regea4 = result[i][2];
+		regeac = result[i][3];
+		regeb4 = result[i][4];
+		regebc = result[i][5];
+		regec4 = result[i][6];
+		regecc = result[i][7];
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"IQK: rege94=%lx rege9c=%lx regea4=%lx regeac=%lx regeb4=%lx regebc=%lx regec4=%lx regecc=%lx\n",
+			rege94, rege9c, regea4, regeac, regeb4, regebc, regec4,
+			regecc);
+	}
+
+	if (final_candidate != 0xff) {
+		rege94 = result[final_candidate][0];
+		rtlphy->reg_e94 = rege94;
+		rege9c = result[final_candidate][1];
+		rtlphy->reg_e9c = rege9c;
+		regea4 = result[final_candidate][2];
+		regeac = result[final_candidate][3];
+		regeb4 = result[final_candidate][4];
+		rtlphy->reg_eb4 = regeb4;
+		regebc = result[final_candidate][5];
+		rtlphy->reg_ebc = regebc;
+		regec4 = result[final_candidate][6];
+		regecc = result[final_candidate][7];
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"IQK: final_candidate is %x\n", final_candidate);
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"IQK: rege94=%lx rege9c=%lx regea4=%lx regeac=%lx regeb4=%lx regebc=%lx regec4=%lx regecc=%lx\n",
+			rege94, rege9c, regea4, regeac, regeb4, regebc, regec4,
+			regecc);
+
+		patha_ok = true;
+		pathb_ok = true;
+	} else {
+		rtlphy->reg_e94 = 0x100;
+		rtlphy->reg_eb4 = 0x100; /* X default value */
+		rtlphy->reg_e9c = 0x0;
+		rtlphy->reg_ebc = 0x0;   /* Y default value */
+	}
+	if (rege94 != 0 /*&& regea4 != 0*/)
+		_rtl92du_phy_patha_fill_iqk_matrix(hw, patha_ok, result,
+						   final_candidate,
+						   regea4 == 0);
+	if (IS_92D_SINGLEPHY(rtlhal->version) &&
+	    regeb4 != 0 /*&& regec4 != 0*/)
+		_rtl92du_phy_pathb_fill_iqk_matrix(hw, pathb_ok, result,
+						   final_candidate,
+						   regec4 == 0);
+
+	if (final_candidate != 0xFF) {
+		indexforchannel =
+			rtl92d_get_rightchnlplace_for_iqk(rtlphy->current_channel);
+
+		for (i = 0; i < IQK_MATRIX_REG_NUM; i++)
+			rtlphy->iqk_matrix[indexforchannel].value[0][i] =
+				result[final_candidate][i];
+
+		rtlphy->iqk_matrix[indexforchannel].iqk_done = true;
+
+		rtl_dbg(rtlpriv, COMP_SCAN | COMP_MLME, DBG_LOUD,
+			"IQK OK indexforchannel %d\n", indexforchannel);
+	}
+}
+
+void rtl92du_phy_reload_iqk_setting(struct ieee80211_hw *hw, u8 channel)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	struct rtl_mac *mac = rtl_mac(rtlpriv);
+	u8 indexforchannel;
+	bool need_iqk;
+
+	rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "channel %d\n", channel);
+	/*------Do IQK for normal chip and test chip 5G band------- */
+
+	indexforchannel = rtl92d_get_rightchnlplace_for_iqk(channel);
+	rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "indexforchannel %d done %d\n",
+		indexforchannel,
+		rtlphy->iqk_matrix[indexforchannel].iqk_done);
+
+	/* We need to do IQK if we're about to connect to a network on 5 GHz.
+	 * On 5 GHz a channel switch outside of scanning happens only before
+	 * connecting.
+	 */
+	need_iqk = !mac->act_scanning;
+
+	if (!rtlphy->iqk_matrix[indexforchannel].iqk_done && need_iqk) {
+		rtl_dbg(rtlpriv, COMP_SCAN | COMP_INIT, DBG_LOUD,
+			"Do IQK Matrix reg for channel:%d....\n", channel);
+		rtl92du_phy_iq_calibrate(hw);
+		return;
+	}
+
+	/* Just load the value. */
+	/* 2G band just load once. */
+	if ((!rtlhal->load_imrandiqk_setting_for2g && indexforchannel == 0) ||
+	    indexforchannel > 0) {
+		rtl_dbg(rtlpriv, COMP_SCAN, DBG_LOUD,
+			"Just Read IQK Matrix reg for channel:%d....\n",
+			channel);
+
+		if (rtlphy->iqk_matrix[indexforchannel].value[0][0] != 0)
+			_rtl92du_phy_patha_fill_iqk_matrix(hw, true,
+				rtlphy->iqk_matrix[indexforchannel].value, 0,
+				rtlphy->iqk_matrix[indexforchannel].value[0][2] == 0);
+
+		if (IS_92D_SINGLEPHY(rtlhal->version) &&
+		    rtlphy->iqk_matrix[indexforchannel].value[0][4] != 0)
+			_rtl92du_phy_pathb_fill_iqk_matrix(hw, true,
+				rtlphy->iqk_matrix[indexforchannel].value, 0,
+				rtlphy->iqk_matrix[indexforchannel].value[0][6] == 0);
+	}
+	rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "<====\n");
+}
+
+static void _rtl92du_phy_reload_lck_setting(struct ieee80211_hw *hw, u8 channel)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	u8 erfpath = rtlhal->current_bandtype == BAND_ON_5G ? RF90_PATH_A :
+		IS_92D_SINGLEPHY(rtlhal->version) ? RF90_PATH_B : RF90_PATH_A;
+	bool bneed_powerdown_radio = false;
+	u32 u4tmp, u4regvalue;
+
+	rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "path %d\n", erfpath);
+	RTPRINT(rtlpriv, FINIT, INIT_IQK, "band type = %d\n",
+		rtlpriv->rtlhal.current_bandtype);
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "channel = %d\n", channel);
+
+	if (rtlpriv->rtlhal.current_bandtype == BAND_ON_5G) {/* Path-A for 5G */
+		u4tmp = rtlpriv->curveindex_5g[channel - 1];
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"ver 1 set RF-A, 5G,	0x28 = 0x%x !!\n", u4tmp);
+
+		if (rtlpriv->rtlhal.macphymode == DUALMAC_DUALPHY &&
+		    rtlpriv->rtlhal.interfaceindex == 1) {
+			bneed_powerdown_radio =
+				rtl92du_phy_enable_anotherphy(hw, false);
+			rtlpriv->rtlhal.during_mac1init_radioa = true;
+			/* asume no this case */
+			if (bneed_powerdown_radio)
+				rtl92d_phy_enable_rf_env(hw, erfpath,
+							 &u4regvalue);
+		}
+
+		rtl_set_rfreg(hw, erfpath, RF_SYN_G4, 0x3f800, u4tmp);
+
+		if (bneed_powerdown_radio) {
+			rtl92d_phy_restore_rf_env(hw, erfpath, &u4regvalue);
+			rtl92du_phy_powerdown_anotherphy(hw, false);
+		}
+	} else if (rtlpriv->rtlhal.current_bandtype == BAND_ON_2_4G) {
+		u4tmp = rtlpriv->curveindex_2g[channel - 1];
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"ver 3 set RF-B, 2G, 0x28 = 0x%x !!\n", u4tmp);
+
+		if (rtlpriv->rtlhal.macphymode == DUALMAC_DUALPHY &&
+		    rtlpriv->rtlhal.interfaceindex == 0) {
+			bneed_powerdown_radio =
+				rtl92du_phy_enable_anotherphy(hw, true);
+			rtlpriv->rtlhal.during_mac0init_radiob = true;
+			if (bneed_powerdown_radio)
+				rtl92d_phy_enable_rf_env(hw, erfpath,
+							 &u4regvalue);
+		}
+
+		rtl_set_rfreg(hw, erfpath, RF_SYN_G4, 0x3f800, u4tmp);
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"ver 3 set RF-B, 2G, 0x28 = 0x%x !!\n",
+			rtl_get_rfreg(hw,  erfpath, RF_SYN_G4, 0x3f800));
+
+		if (bneed_powerdown_radio) {
+			rtl92d_phy_restore_rf_env(hw, erfpath, &u4regvalue);
+			rtl92du_phy_powerdown_anotherphy(hw, true);
+		}
+	}
+	rtl_dbg(rtlpriv, COMP_CMD, DBG_LOUD, "<====\n");
+}
+
+static void _rtl92du_phy_lc_calibrate_sw(struct ieee80211_hw *hw, bool is2t)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	u32 curvecount_val[CV_CURVE_CNT * 2];
+	u16 timeout = 800, timecount = 0;
+	u32 u4tmp, offset, rf_syn_g4[2];
+	u8 tmpreg, index, rf_mode[2];
+	u8 path = is2t ? 2 : 1;
+	u8 i;
+
+	/* Check continuous TX and Packet TX */
+	tmpreg = rtl_read_byte(rtlpriv, 0xd03);
+	if ((tmpreg & 0x70) != 0)
+		/* if Deal with contisuous TX case, disable all continuous TX */
+		rtl_write_byte(rtlpriv, 0xd03, tmpreg & 0x8F);
+	else
+		/* if Deal with Packet TX case, block all queues */
+		rtl_write_byte(rtlpriv, REG_TXPAUSE, 0xFF);
+
+	rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER4, 0xF00000, 0x0F);
+
+	for (index = 0; index < path; index++) {
+		/* 1. Read original RF mode */
+		offset = index == 0 ? ROFDM0_XAAGCCORE1 : ROFDM0_XBAGCCORE1;
+		rf_mode[index] = rtl_read_byte(rtlpriv, offset);
+
+		/* 2. Set RF mode = standby mode */
+		rtl_set_rfreg(hw, (enum radio_path)index, RF_AC,
+			      RFREG_OFFSET_MASK, 0x010000);
+
+		rf_syn_g4[index] = rtl_get_rfreg(hw, index, RF_SYN_G4,
+						 RFREG_OFFSET_MASK);
+		rtl_set_rfreg(hw, index, RF_SYN_G4, 0x700, 0x7);
+
+		/* switch CV-curve control by LC-calibration */
+		rtl_set_rfreg(hw, (enum radio_path)index, RF_SYN_G7,
+			      BIT(17), 0x0);
+
+		/* 4. Set LC calibration begin */
+		rtl_set_rfreg(hw, (enum radio_path)index, RF_CHNLBW,
+			      0x08000, 0x01);
+	}
+
+	for (index = 0; index < path; index++) {
+		u4tmp = rtl_get_rfreg(hw, (enum radio_path)index, RF_SYN_G6,
+				      RFREG_OFFSET_MASK);
+
+		while ((!(u4tmp & BIT(11))) && timecount <= timeout) {
+			mdelay(50);
+			timecount += 50;
+			u4tmp = rtl_get_rfreg(hw, (enum radio_path)index,
+					      RF_SYN_G6, RFREG_OFFSET_MASK);
+		}
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,
+			"PHY_LCK finish delay for %d ms=2\n", timecount);
+	}
+
+	if ((tmpreg & 0x70) != 0)
+		rtl_write_byte(rtlpriv, 0xd03, tmpreg);
+	else /* Deal with Packet TX case */
+		rtl_write_byte(rtlpriv, REG_TXPAUSE, 0x00);
+
+	rtl_set_bbreg(hw, RFPGA0_ANALOGPARAMETER4, 0xF00000, 0x00);
+
+	for (index = 0; index < path; index++) {
+		rtl_get_rfreg(hw, index, RF_SYN_G4, RFREG_OFFSET_MASK);
+
+		if (index == 0 && rtlhal->interfaceindex == 0) {
+			RTPRINT(rtlpriv, FINIT, INIT_IQK,
+				"path-A / 5G LCK\n");
+		} else {
+			RTPRINT(rtlpriv, FINIT, INIT_IQK,
+				"path-B / 2.4G LCK\n");
+		}
+
+		memset(curvecount_val, 0, sizeof(curvecount_val));
+
+		/* Set LC calibration off */
+		rtl_set_rfreg(hw, (enum radio_path)index, RF_CHNLBW,
+			      0x08000, 0x0);
+
+		RTPRINT(rtlpriv, FINIT, INIT_IQK,  "set RF 0x18[15] = 0\n");
+
+		/* save Curve-counting number */
+		for (i = 0; i < CV_CURVE_CNT; i++) {
+			u32 readval = 0, readval2 = 0;
+
+			rtl_set_rfreg(hw, (enum radio_path)index, 0x3F,
+				      0x7f, i);
+
+			rtl_set_rfreg(hw, (enum radio_path)index, 0x4D,
+				      RFREG_OFFSET_MASK, 0x0);
+
+			readval = rtl_get_rfreg(hw, (enum radio_path)index,
+						0x4F, RFREG_OFFSET_MASK);
+			curvecount_val[2 * i + 1] = (readval & 0xfffe0) >> 5;
+
+			/* reg 0x4f [4:0] */
+			/* reg 0x50 [19:10] */
+			readval2 = rtl_get_rfreg(hw, (enum radio_path)index,
+						 0x50, 0xffc00);
+			curvecount_val[2 * i] = (((readval & 0x1F) << 10) |
+						 readval2);
+		}
+
+		if (index == 0 && rtlhal->interfaceindex == 0)
+			rtl92d_phy_calc_curvindex(hw, targetchnl_5g,
+						  curvecount_val,
+						  true, rtlpriv->curveindex_5g);
+		else
+			rtl92d_phy_calc_curvindex(hw, targetchnl_2g,
+						  curvecount_val,
+						  false, rtlpriv->curveindex_2g);
+
+		/* switch CV-curve control mode */
+		rtl_set_rfreg(hw, (enum radio_path)index, RF_SYN_G7,
+			      BIT(17), 0x1);
+	}
+
+	/* Restore original situation  */
+	for (index = 0; index < path; index++) {
+		rtl_set_rfreg(hw, index, RF_SYN_G4, RFREG_OFFSET_MASK,
+			      rf_syn_g4[index]);
+
+		offset = index == 0 ? ROFDM0_XAAGCCORE1 : ROFDM0_XBAGCCORE1;
+		rtl_write_byte(rtlpriv, offset, 0x50);
+		rtl_write_byte(rtlpriv, offset, rf_mode[index]);
+	}
+
+	_rtl92du_phy_reload_lck_setting(hw, rtlpriv->phy.current_channel);
+}
+
+void rtl92du_phy_lc_calibrate(struct ieee80211_hw *hw, bool is2t)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	u32 timeout = 2000, timecount = 0;
+
+	while (rtlpriv->mac80211.act_scanning && timecount < timeout) {
+		udelay(50);
+		timecount += 50;
+	}
+
+	rtlphy->lck_inprogress = true;
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,
+		"LCK:Start!!! currentband %x delay %d ms\n",
+		rtlhal->current_bandtype, timecount);
+
+	_rtl92du_phy_lc_calibrate_sw(hw, is2t);
+
+	rtlphy->lck_inprogress = false;
+	RTPRINT(rtlpriv, FINIT, INIT_IQK,  "LCK:Finish!!!\n");
+}
+
+void rtl92du_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta)
+{
+	/* Nothing to do. */
+}
+
+u8 rtl92du_phy_sw_chnl(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	u8 num_total_rfpath = rtlphy->num_total_rfpath;
+	u8 channel = rtlphy->current_channel;
+	u32 timeout = 1000, timecount = 0;
+	u32 ret_value;
+	u8 rfpath;
+
+	if (rtlphy->sw_chnl_inprogress)
+		return 0;
+	if (rtlphy->set_bwmode_inprogress)
+		return 0;
+
+	if ((is_hal_stop(rtlhal)) || (RT_CANNOT_IO(hw))) {
+		rtl_dbg(rtlpriv, COMP_CHAN, DBG_LOUD,
+			"sw_chnl_inprogress false driver sleep or unload\n");
+		return 0;
+	}
+
+	while (rtlphy->lck_inprogress && timecount < timeout) {
+		mdelay(50);
+		timecount += 50;
+	}
+
+	if (rtlhal->macphymode == SINGLEMAC_SINGLEPHY &&
+	    rtlhal->bandset == BAND_ON_BOTH) {
+		ret_value = rtl_get_bbreg(hw, RFPGA0_XAB_RFPARAMETER,
+					  MASKDWORD);
+		if (rtlphy->current_channel > 14 && !(ret_value & BIT(0)))
+			rtl92du_phy_switch_wirelessband(hw, BAND_ON_5G);
+		else if (rtlphy->current_channel <= 14 && (ret_value & BIT(0)))
+			rtl92du_phy_switch_wirelessband(hw, BAND_ON_2_4G);
+	}
+
+	switch (rtlhal->current_bandtype) {
+	case BAND_ON_5G:
+		/* Get first channel error when change between
+		 * 5G and 2.4G band.
+		 */
+		if (WARN_ONCE(channel <= 14, "rtl8192du: 5G but channel<=14\n"))
+			return 0;
+		break;
+	case BAND_ON_2_4G:
+		/* Get first channel error when change between
+		 * 5G and 2.4G band.
+		 */
+		if (WARN_ONCE(channel > 14, "rtl8192du: 2G but channel>14\n"))
+			return 0;
+		break;
+	default:
+		WARN_ONCE(true, "rtl8192du: Invalid WirelessMode(%#x)!!\n",
+			  rtlpriv->mac80211.mode);
+		break;
+	}
+
+	rtlphy->sw_chnl_inprogress = true;
+
+	rtl_dbg(rtlpriv, COMP_SCAN, DBG_TRACE,
+		"switch to channel%d\n", rtlphy->current_channel);
+
+	rtl92d_phy_set_txpower_level(hw, channel);
+
+	for (rfpath = 0; rfpath < num_total_rfpath; rfpath++) {
+		u32p_replace_bits(&rtlphy->rfreg_chnlval[rfpath],
+				  channel, 0xff);
+
+		if (rtlpriv->rtlhal.current_bandtype == BAND_ON_5G) {
+			if (channel > 99)
+				rtlphy->rfreg_chnlval[rfpath] |= (BIT(18));
+			else
+				rtlphy->rfreg_chnlval[rfpath] &= ~BIT(18);
+			rtlphy->rfreg_chnlval[rfpath] |= (BIT(16) | BIT(8));
+		} else {
+			rtlphy->rfreg_chnlval[rfpath] &=
+				~(BIT(8) | BIT(16) | BIT(18));
+		}
+		rtl_set_rfreg(hw, rfpath, RF_CHNLBW, RFREG_OFFSET_MASK,
+			      rtlphy->rfreg_chnlval[rfpath]);
+
+		_rtl92du_phy_reload_imr_setting(hw, channel, rfpath);
+	}
+
+	_rtl92du_phy_switch_rf_setting(hw, channel);
+
+	/* do IQK when all parameters are ready */
+	rtl92du_phy_reload_iqk_setting(hw, channel);
+
+	rtl_dbg(rtlpriv, COMP_SCAN, DBG_TRACE, "<==\n");
+	rtlphy->sw_chnl_inprogress = false;
+	return 1;
+}
+
+static void _rtl92du_phy_set_rfon(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+
+	/* a.  SYS_CLKR 0x08[11] = 1  restore MAC clock */
+	/* b.  SPS_CTRL 0x11[7:0] = 0x2b */
+	if (rtlpriv->rtlhal.macphymode == SINGLEMAC_SINGLEPHY)
+		rtl_write_byte(rtlpriv, REG_SPS0_CTRL, 0x2b);
+
+	/* c.  For PCIE: SYS_FUNC_EN 0x02[7:0] = 0xE3 enable BB TRX function */
+	rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN, 0xE3);
+
+	/* RF_ON_EXCEP(d~g): */
+	/* d.  APSD_CTRL 0x600[7:0] = 0x00 */
+	rtl_write_byte(rtlpriv, REG_APSD_CTRL, 0x00);
+
+	/* e.  SYS_FUNC_EN 0x02[7:0] = 0xE2  reset BB TRX function again */
+	/* f.  SYS_FUNC_EN 0x02[7:0] = 0xE3  enable BB TRX function*/
+	rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN, 0xE2);
+	rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN, 0xE3);
+
+	/* g.   txpause 0x522[7:0] = 0x00  enable mac tx queue */
+	rtl_write_byte(rtlpriv, REG_TXPAUSE, 0x00);
+}
+
+static void _rtl92du_phy_set_rfsleep(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	u32 u4btmp;
+	u8 retry = 5;
+
+	/* a.   TXPAUSE 0x522[7:0] = 0xFF  Pause MAC TX queue  */
+	rtl_write_byte(rtlpriv, REG_TXPAUSE, 0xFF);
+
+	/* b.   RF path 0 offset 0x00 = 0x00  disable RF  */
+	rtl_set_rfreg(hw, RF90_PATH_A, 0x00, RFREG_OFFSET_MASK, 0x00);
+
+	/* c.   APSD_CTRL 0x600[7:0] = 0x40 */
+	rtl_write_byte(rtlpriv, REG_APSD_CTRL, 0x40);
+
+	/* d. APSD_CTRL 0x600[7:0] = 0x00
+	 * APSD_CTRL 0x600[7:0] = 0x00
+	 * RF path 0 offset 0x00 = 0x00
+	 * APSD_CTRL 0x600[7:0] = 0x40
+	 */
+	u4btmp = rtl_get_rfreg(hw, RF90_PATH_A, 0, RFREG_OFFSET_MASK);
+	while (u4btmp != 0 && retry > 0) {
+		rtl_write_byte(rtlpriv, REG_APSD_CTRL, 0x0);
+		rtl_set_rfreg(hw, RF90_PATH_A, 0x00, RFREG_OFFSET_MASK, 0x00);
+		rtl_write_byte(rtlpriv, REG_APSD_CTRL, 0x40);
+		u4btmp = rtl_get_rfreg(hw, RF90_PATH_A, 0, RFREG_OFFSET_MASK);
+		retry--;
+	}
+	if (retry == 0) {
+		/* Jump out the LPS turn off sequence to RF_ON_EXCEP */
+		rtl_write_byte(rtlpriv, REG_APSD_CTRL, 0x00);
+
+		rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN, 0xE2);
+		rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN, 0xE3);
+		rtl_write_byte(rtlpriv, REG_TXPAUSE, 0x00);
+		rtl_dbg(rtlpriv, COMP_POWER, DBG_LOUD,
+			"Fail !!! Switch RF timeout\n");
+		return;
+	}
+
+	/* e.   For PCIE: SYS_FUNC_EN 0x02[7:0] = 0xE2 reset BB TRX function */
+	rtl_write_byte(rtlpriv, REG_SYS_FUNC_EN, 0xE2);
+
+	/* f.   SPS_CTRL 0x11[7:0] = 0x22 */
+	if (rtlpriv->rtlhal.macphymode == SINGLEMAC_SINGLEPHY)
+		rtl_write_byte(rtlpriv, REG_SPS0_CTRL, 0x22);
+}
+
+bool rtl92du_phy_set_rf_power_state(struct ieee80211_hw *hw,
+				    enum rf_pwrstate rfpwr_state)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_ps_ctl *ppsc = rtl_psc(rtlpriv);
+	struct rtl_mac *mac = rtl_mac(rtlpriv);
+	bool bresult = true;
+
+	if (rfpwr_state == ppsc->rfpwr_state)
+		return false;
+
+	switch (rfpwr_state) {
+	case ERFON:
+		if (ppsc->rfpwr_state == ERFOFF &&
+		    RT_IN_PS_LEVEL(ppsc, RT_RF_OFF_LEVL_HALT_NIC)) {
+			u32 initializecount = 0;
+			bool rtstatus;
+
+			do {
+				initializecount++;
+				rtl_dbg(rtlpriv, COMP_RF, DBG_DMESG,
+					"IPS Set eRf nic enable\n");
+				rtstatus = rtl_ps_enable_nic(hw);
+			} while (!rtstatus && (initializecount < 10));
+
+			RT_CLEAR_PS_LEVEL(ppsc,
+					  RT_RF_OFF_LEVL_HALT_NIC);
+		} else {
+			rtl_dbg(rtlpriv, COMP_POWER, DBG_DMESG,
+				"awake, slept:%d ms state_inap:%x\n",
+				jiffies_to_msecs(jiffies -
+						 ppsc->last_sleep_jiffies),
+				 rtlpriv->psc.state_inap);
+			ppsc->last_awake_jiffies = jiffies;
+			_rtl92du_phy_set_rfon(hw);
+		}
+
+		if (mac->link_state == MAC80211_LINKED)
+			rtlpriv->cfg->ops->led_control(hw, LED_CTL_LINK);
+		else
+			rtlpriv->cfg->ops->led_control(hw, LED_CTL_NO_LINK);
+		break;
+	case ERFOFF:
+		if (ppsc->reg_rfps_level & RT_RF_OFF_LEVL_HALT_NIC) {
+			rtl_dbg(rtlpriv, COMP_RF, DBG_DMESG,
+				"IPS Set eRf nic disable\n");
+			rtl_ps_disable_nic(hw);
+			RT_SET_PS_LEVEL(ppsc, RT_RF_OFF_LEVL_HALT_NIC);
+		} else {
+			if (ppsc->rfoff_reason == RF_CHANGE_BY_IPS)
+				rtlpriv->cfg->ops->led_control(hw, LED_CTL_NO_LINK);
+			else
+				rtlpriv->cfg->ops->led_control(hw, LED_CTL_POWER_OFF);
+		}
+		break;
+	case ERFSLEEP:
+		if (ppsc->rfpwr_state == ERFOFF)
+			return false;
+
+		rtl_dbg(rtlpriv, COMP_POWER, DBG_DMESG,
+			"sleep awakened:%d ms state_inap:%x\n",
+			jiffies_to_msecs(jiffies -
+					 ppsc->last_awake_jiffies),
+			rtlpriv->psc.state_inap);
+		ppsc->last_sleep_jiffies = jiffies;
+		_rtl92du_phy_set_rfsleep(hw);
+		break;
+	default:
+		pr_err("switch case %#x not processed\n",
+		       rfpwr_state);
+		return false;
+	}
+
+	if (bresult)
+		ppsc->rfpwr_state = rfpwr_state;
+
+	return bresult;
+}
+
+void rtl92du_phy_set_poweron(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw));
+	u32 mac_reg = (rtlhal->interfaceindex == 0 ? REG_MAC0 : REG_MAC1);
+	u8 value8;
+	u16 i;
+
+	/* notice fw know band status  0x81[1]/0x53[1] = 0: 5G, 1: 2G */
+	if (rtlhal->current_bandtype == BAND_ON_2_4G) {
+		value8 = rtl_read_byte(rtlpriv, mac_reg);
+		value8 |= BIT(1);
+		rtl_write_byte(rtlpriv, mac_reg, value8);
+	} else {
+		value8 = rtl_read_byte(rtlpriv, mac_reg);
+		value8 &= ~BIT(1);
+		rtl_write_byte(rtlpriv, mac_reg, value8);
+	}
+
+	if (rtlhal->macphymode == SINGLEMAC_SINGLEPHY) {
+		value8 = rtl_read_byte(rtlpriv, REG_MAC0);
+		rtl_write_byte(rtlpriv, REG_MAC0, value8 | MAC0_ON);
+	} else {
+		mutex_lock(rtlpriv->mutex_for_power_on_off);
+		if (rtlhal->interfaceindex == 0) {
+			value8 = rtl_read_byte(rtlpriv, REG_MAC0);
+			rtl_write_byte(rtlpriv, REG_MAC0, value8 | MAC0_ON);
+		} else {
+			value8 = rtl_read_byte(rtlpriv, REG_MAC1);
+			rtl_write_byte(rtlpriv, REG_MAC1, value8 | MAC1_ON);
+		}
+		value8 = rtl_read_byte(rtlpriv, REG_POWER_OFF_IN_PROCESS);
+		mutex_unlock(rtlpriv->mutex_for_power_on_off);
+
+		for (i = 0; i < 200; i++) {
+			if ((value8 & BIT(7)) == 0)
+				break;
+
+			udelay(500);
+			mutex_lock(rtlpriv->mutex_for_power_on_off);
+			value8 = rtl_read_byte(rtlpriv,
+					       REG_POWER_OFF_IN_PROCESS);
+			mutex_unlock(rtlpriv->mutex_for_power_on_off);
+		}
+		if (i == 200)
+			WARN_ONCE(true, "rtl8192du: Another mac power off over time\n");
+	}
+}
+
+void rtl92du_update_bbrf_configuration(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_efuse *rtlefuse = rtl_efuse(rtlpriv);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	u8 rfpath, i;
+
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD, "==>\n");
+	/* r_select_5G for path_A/B 0 for 2.4G, 1 for 5G */
+	if (rtlhal->current_bandtype == BAND_ON_2_4G) {
+		/* r_select_5G for path_A/B, 0x878 */
+		rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER, BIT(0), 0x0);
+		rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER, BIT(15), 0x0);
+		if (rtlhal->macphymode != DUALMAC_DUALPHY) {
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER, BIT(16), 0x0);
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER, BIT(31), 0x0);
+		}
+
+		/* rssi_table_select: index 0 for 2.4G. 1~3 for 5G, 0xc78 */
+		rtl_set_bbreg(hw, ROFDM0_AGCRSSITABLE, BIT(6) | BIT(7), 0x0);
+
+		/* fc_area  0xd2c */
+		rtl_set_bbreg(hw, ROFDM1_CFOTRACKING, BIT(14) | BIT(13), 0x0);
+
+		/* 5G LAN ON */
+		rtl_set_bbreg(hw, 0xB30, 0x00F00000, 0xa);
+
+		/* TX BB gain shift*1, Just for testchip, 0xc80, 0xc88 */
+		rtl_set_bbreg(hw, ROFDM0_XATXIQIMBALANCE, MASKDWORD, 0x40000100);
+		rtl_set_bbreg(hw, ROFDM0_XBTXIQIMBALANCE, MASKDWORD, 0x40000100);
+		if (rtlhal->macphymode == DUALMAC_DUALPHY) {
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFINTERFACESW,
+				      BIT(10) | BIT(6) | BIT(5),
+				      ((rtlefuse->eeprom_c9 & BIT(3)) >> 3) |
+				      (rtlefuse->eeprom_c9 & BIT(1)) |
+				      ((rtlefuse->eeprom_cc & BIT(1)) << 4));
+			rtl_set_bbreg(hw, RFPGA0_XA_RFINTERFACEOE,
+				      BIT(10) | BIT(6) | BIT(5),
+				      ((rtlefuse->eeprom_c9 & BIT(2)) >> 2) |
+				      ((rtlefuse->eeprom_c9 & BIT(0)) << 1) |
+				      ((rtlefuse->eeprom_cc & BIT(0)) << 5));
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER, BIT(15), 0);
+
+			rtl_set_bbreg(hw, RPDP_ANTA, MASKDWORD, 0x01017038);
+			rtl_set_bbreg(hw, RCONFIG_ANTA, MASKDWORD, 0x0f600000);
+		} else {
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFINTERFACESW,
+				      BIT(26) | BIT(22) | BIT(21) | BIT(10) |
+				      BIT(6) | BIT(5),
+				      ((rtlefuse->eeprom_c9 & BIT(3)) >> 3) |
+				      (rtlefuse->eeprom_c9 & BIT(1)) |
+				      ((rtlefuse->eeprom_cc & BIT(1)) << 4) |
+				      ((rtlefuse->eeprom_c9 & BIT(7)) << 9) |
+				      ((rtlefuse->eeprom_c9 & BIT(5)) << 12) |
+				      ((rtlefuse->eeprom_cc & BIT(3)) << 18));
+			rtl_set_bbreg(hw, RFPGA0_XA_RFINTERFACEOE,
+				      BIT(10) | BIT(6) | BIT(5),
+				      ((rtlefuse->eeprom_c9 & BIT(2)) >> 2) |
+				      ((rtlefuse->eeprom_c9 & BIT(0)) << 1) |
+				      ((rtlefuse->eeprom_cc & BIT(0)) << 5));
+			rtl_set_bbreg(hw, RFPGA0_XB_RFINTERFACEOE,
+				      BIT(10) | BIT(6) | BIT(5),
+				      ((rtlefuse->eeprom_c9 & BIT(6)) >> 6) |
+				      ((rtlefuse->eeprom_c9 & BIT(4)) >> 3) |
+				      ((rtlefuse->eeprom_cc & BIT(2)) << 3));
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER,
+				      BIT(31) | BIT(15), 0);
+
+			rtl_set_bbreg(hw, RPDP_ANTA, MASKDWORD, 0x01017038);
+			rtl_set_bbreg(hw, RPDP_ANTB, MASKDWORD, 0x01017038);
+			rtl_set_bbreg(hw, RCONFIG_ANTA, MASKDWORD, 0x0f600000);
+			rtl_set_bbreg(hw, RCONFIG_ANTB, MASKDWORD, 0x0f600000);
+		}
+		/* 1.5V_LDO */
+	} else {
+		/* r_select_5G for path_A/B */
+		rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER, BIT(0), 0x1);
+		rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER, BIT(15), 0x1);
+		if (rtlhal->macphymode != DUALMAC_DUALPHY) {
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER, BIT(16), 0x1);
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER, BIT(31), 0x1);
+		}
+
+		/* rssi_table_select: index 0 for 2.4G. 1~3 for 5G */
+		rtl_set_bbreg(hw, ROFDM0_AGCRSSITABLE, BIT(6) | BIT(7), 0x1);
+
+		/* fc_area */
+		rtl_set_bbreg(hw, ROFDM1_CFOTRACKING, BIT(14) | BIT(13), 0x1);
+
+		/* 5G LAN ON */
+		rtl_set_bbreg(hw, 0xB30, 0x00F00000, 0x0);
+
+		/* TX BB gain shift, Just for testchip, 0xc80, 0xc88 */
+		if (rtlefuse->internal_pa_5g[rtlhal->interfaceindex])
+			rtl_set_bbreg(hw, ROFDM0_XATXIQIMBALANCE, MASKDWORD,
+				      0x2d4000b5);
+		else
+			rtl_set_bbreg(hw, ROFDM0_XATXIQIMBALANCE, MASKDWORD,
+				      0x20000080);
+
+		if (rtlhal->macphymode != DUALMAC_DUALPHY) {
+			if (rtlefuse->internal_pa_5g[1])
+				rtl_set_bbreg(hw, ROFDM0_XBTXIQIMBALANCE,
+					      MASKDWORD, 0x2d4000b5);
+			else
+				rtl_set_bbreg(hw, ROFDM0_XBTXIQIMBALANCE,
+					      MASKDWORD, 0x20000080);
+		}
+
+		rtl_set_bbreg(hw, 0xB30, BIT(27), 0);
+
+		if (rtlhal->macphymode == DUALMAC_DUALPHY) {
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFINTERFACESW,
+				      BIT(10) | BIT(6) | BIT(5),
+				      (rtlefuse->eeprom_cc & BIT(5)));
+			rtl_set_bbreg(hw, RFPGA0_XA_RFINTERFACEOE, BIT(10),
+				      ((rtlefuse->eeprom_cc & BIT(4)) >> 4));
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER, BIT(15),
+				      (rtlefuse->eeprom_cc & BIT(4)) >> 4);
+
+			rtl_set_bbreg(hw, RPDP_ANTA, MASKDWORD, 0x01017098);
+			rtl_set_bbreg(hw, RCONFIG_ANTA, MASKDWORD, 0x20000000);
+		} else {
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFINTERFACESW,
+				      BIT(26) | BIT(22) | BIT(21) | BIT(10) |
+				      BIT(6) | BIT(5),
+				      (rtlefuse->eeprom_cc & BIT(5)) |
+				      ((rtlefuse->eeprom_cc & BIT(7)) << 14));
+			rtl_set_bbreg(hw, RFPGA0_XA_RFINTERFACEOE, BIT(10),
+				      ((rtlefuse->eeprom_cc & BIT(4)) >> 4));
+			rtl_set_bbreg(hw, RFPGA0_XB_RFINTERFACEOE, BIT(10),
+				      ((rtlefuse->eeprom_cc & BIT(6)) >> 6));
+			rtl_set_bbreg(hw, RFPGA0_XAB_RFPARAMETER,
+				      BIT(31) | BIT(15),
+				      ((rtlefuse->eeprom_cc & BIT(4)) >> 4) |
+				      ((rtlefuse->eeprom_cc & BIT(6)) << 10));
+
+			rtl_set_bbreg(hw, RPDP_ANTA, MASKDWORD, 0x01017098);
+			rtl_set_bbreg(hw, RPDP_ANTB, MASKDWORD, 0x01017098);
+			rtl_set_bbreg(hw, RCONFIG_ANTA, MASKDWORD, 0x20000000);
+			rtl_set_bbreg(hw, RCONFIG_ANTB, MASKDWORD, 0x20000000);
+		}
+	}
+
+	/* update IQK related settings */
+	rtl_set_bbreg(hw, ROFDM0_XARXIQIMBALANCE, MASKDWORD, 0x40000100);
+	rtl_set_bbreg(hw, ROFDM0_XBRXIQIMBALANCE, MASKDWORD, 0x40000100);
+	rtl_set_bbreg(hw, ROFDM0_XCTXAFE, 0xF0000000, 0x00);
+	rtl_set_bbreg(hw, ROFDM0_ECCATHRESHOLD, BIT(30) | BIT(28) |
+		      BIT(26) | BIT(24), 0x00);
+	rtl_set_bbreg(hw, ROFDM0_XDTXAFE, 0xF0000000, 0x00);
+	rtl_set_bbreg(hw, ROFDM0_RXIQEXTANTA, 0xF0000000, 0x00);
+	rtl_set_bbreg(hw, ROFDM0_AGCRSSITABLE, 0x0000F000, 0x00);
+
+	/* Update RF */
+	for (rfpath = RF90_PATH_A; rfpath < rtlphy->num_total_rfpath;
+	     rfpath++) {
+		if (rtlhal->current_bandtype == BAND_ON_2_4G) {
+			/* MOD_AG for RF path_A 0x18 BIT8,BIT16 */
+			rtl_set_rfreg(hw, rfpath, RF_CHNLBW, BIT(8) | BIT(16) |
+				      BIT(18) | 0xff, 1);
+
+			/* RF0x0b[16:14] =3b'111 */
+			rtl_set_rfreg(hw, (enum radio_path)rfpath, 0x0B,
+				      0x1c000, 0x07);
+		} else {
+			/* MOD_AG for RF path_A 0x18 BIT8,BIT16 */
+			rtl_set_rfreg(hw, RF90_PATH_A, RF_CHNLBW, RFREG_OFFSET_MASK,
+				      0x97524);
+		}
+
+		/* Set right channel on RF reg0x18 for another mac. */
+		if (rtlhal->interfaceindex == 0 && rtlhal->bandset == BAND_ON_2_4G) {
+			/* Set MAC1 default channel if MAC1 not up. */
+			if (!(rtl_read_byte(rtlpriv, REG_MAC1) & MAC1_ON)) {
+				rtl92du_phy_enable_anotherphy(hw, true);
+				rtlhal->during_mac0init_radiob = true;
+				rtl_set_rfreg(hw, RF90_PATH_A, RF_CHNLBW,
+					      RFREG_OFFSET_MASK, 0x97524);
+				rtl92du_phy_powerdown_anotherphy(hw, true);
+			}
+		} else if (rtlhal->interfaceindex == 1 && rtlhal->bandset == BAND_ON_5G) {
+			/* Set MAC0 default channel */
+			if (!(rtl_read_byte(rtlpriv, REG_MAC0) & MAC0_ON)) {
+				rtl92du_phy_enable_anotherphy(hw, false);
+				rtlhal->during_mac1init_radioa = true;
+				rtl_set_rfreg(hw, RF90_PATH_A, RF_CHNLBW,
+					      RFREG_OFFSET_MASK, 0x87401);
+				rtl92du_phy_powerdown_anotherphy(hw, false);
+			}
+		}
+	}
+
+	/* Update for all band. */
+	/* DMDP */
+	if (rtlphy->rf_type == RF_1T1R) {
+		/* Use antenna 0, 0xc04, 0xd04 */
+		rtl_set_bbreg(hw, ROFDM0_TRXPATHENABLE, MASKBYTE0, 0x11);
+		rtl_set_bbreg(hw, ROFDM1_TRXPATHENABLE, BDWORD, 0x1);
+
+		/* enable ad/da clock1 for dual-phy reg0x888 */
+		if (rtlhal->interfaceindex == 0) {
+			rtl_set_bbreg(hw, RFPGA0_ADDALLOCKEN, BIT(12) |
+				      BIT(13), 0x3);
+		} else if (rtl92du_phy_enable_anotherphy(hw, false)) {
+			rtlhal->during_mac1init_radioa = true;
+			rtl_set_bbreg(hw, RFPGA0_ADDALLOCKEN,
+				      BIT(12) | BIT(13), 0x3);
+			rtl92du_phy_powerdown_anotherphy(hw, false);
+		}
+
+		rtl_set_bbreg(hw, ROFDM1_LSTF, BIT(19) | BIT(20), 0x0);
+	} else {
+		/* Single PHY */
+		/* Use antenna 0 & 1, 0xc04, 0xd04 */
+		rtl_set_bbreg(hw, ROFDM0_TRXPATHENABLE, MASKBYTE0, 0x33);
+		rtl_set_bbreg(hw, ROFDM1_TRXPATHENABLE, BDWORD, 0x3);
+		/* disable ad/da clock1,0x888 */
+		rtl_set_bbreg(hw, RFPGA0_ADDALLOCKEN, BIT(12) | BIT(13), 0);
+
+		rtl_set_bbreg(hw, ROFDM1_LSTF, BIT(19) | BIT(20), 0x1);
+	}
+
+	for (rfpath = RF90_PATH_A; rfpath < rtlphy->num_total_rfpath;
+	     rfpath++) {
+		rtlphy->rfreg_chnlval[rfpath] = rtl_get_rfreg(hw, rfpath,
+							      RF_CHNLBW,
+							      RFREG_OFFSET_MASK);
+		rtlphy->reg_rf3c[rfpath] = rtl_get_rfreg(hw, rfpath, 0x3C,
+							 RFREG_OFFSET_MASK);
+	}
+
+	for (i = 0; i < 2; i++)
+		rtl_dbg(rtlpriv, COMP_RF, DBG_LOUD, "RF 0x18 = 0x%x\n",
+			rtlphy->rfreg_chnlval[i]);
+
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD, "<==\n");
+}
+
+bool rtl92du_phy_check_poweroff(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw));
+	u8 u1btmp;
+
+	if (rtlhal->macphymode == SINGLEMAC_SINGLEPHY) {
+		u1btmp = rtl_read_byte(rtlpriv, REG_MAC0);
+		rtl_write_byte(rtlpriv, REG_MAC0, u1btmp & ~MAC0_ON);
+		return true;
+	}
+
+	mutex_lock(rtlpriv->mutex_for_power_on_off);
+	if (rtlhal->interfaceindex == 0) {
+		u1btmp = rtl_read_byte(rtlpriv, REG_MAC0);
+		rtl_write_byte(rtlpriv, REG_MAC0, u1btmp & ~MAC0_ON);
+		u1btmp = rtl_read_byte(rtlpriv, REG_MAC1);
+		u1btmp &= MAC1_ON;
+	} else {
+		u1btmp = rtl_read_byte(rtlpriv, REG_MAC1);
+		rtl_write_byte(rtlpriv, REG_MAC1, u1btmp & ~MAC1_ON);
+		u1btmp = rtl_read_byte(rtlpriv, REG_MAC0);
+		u1btmp &= MAC0_ON;
+	}
+	if (u1btmp) {
+		mutex_unlock(rtlpriv->mutex_for_power_on_off);
+		return false;
+	}
+	u1btmp = rtl_read_byte(rtlpriv, REG_POWER_OFF_IN_PROCESS);
+	u1btmp |= BIT(7);
+	rtl_write_byte(rtlpriv, REG_POWER_OFF_IN_PROCESS, u1btmp);
+	mutex_unlock(rtlpriv->mutex_for_power_on_off);
+
+	return true;
+}
+
+void rtl92du_phy_init_pa_bias(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	bool is_single_mac = rtlhal->macphymode == SINGLEMAC_SINGLEPHY;
+	enum radio_path rf_path;
+	u8 val8;
+
+	read_efuse_byte(hw, 0x3FA, &val8);
+
+	rtl_dbg(rtlpriv, COMP_RF, DBG_TRACE, "%s: 0x3FA %#x\n",
+		__func__, val8);
+
+	if (!(val8 & BIT(0)) && (is_single_mac || rtlhal->interfaceindex == 0)) {
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_CHNLBW, RFREG_OFFSET_MASK, 0x07401);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_AC, RFREG_OFFSET_MASK, 0x70000);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x0F425);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x4F425);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x8F425);
+
+		/* Back to RX Mode */
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_AC, RFREG_OFFSET_MASK, 0x30000);
+
+		rtl_dbg(rtlpriv, COMP_RF, DBG_TRACE, "2G PA BIAS path A\n");
+	}
+
+	if (!(val8 & BIT(1)) && (is_single_mac || rtlhal->interfaceindex == 1)) {
+		rf_path = rtlhal->interfaceindex == 1 ? RF90_PATH_A : RF90_PATH_B;
+
+		rtl_set_rfreg(hw, rf_path, RF_CHNLBW, RFREG_OFFSET_MASK, 0x07401);
+		rtl_set_rfreg(hw, rf_path, RF_AC, RFREG_OFFSET_MASK, 0x70000);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x0F425);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x4F425);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x8F425);
+
+		/* Back to RX Mode */
+		rtl_set_rfreg(hw, rf_path, RF_AC, RFREG_OFFSET_MASK, 0x30000);
+
+		rtl_dbg(rtlpriv, COMP_RF, DBG_TRACE, "2G PA BIAS path B\n");
+	}
+
+	if (!(val8 & BIT(2)) && (is_single_mac || rtlhal->interfaceindex == 0)) {
+		/* 5GL_channel */
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_CHNLBW, RFREG_OFFSET_MASK, 0x17524);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_AC, RFREG_OFFSET_MASK, 0x70000);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x0F496);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x4F496);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x8F496);
+
+		/* 5GM_channel */
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_CHNLBW, RFREG_OFFSET_MASK, 0x37564);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_AC, RFREG_OFFSET_MASK, 0x70000);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x0F496);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x4F496);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x8F496);
+
+		/* 5GH_channel */
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_CHNLBW, RFREG_OFFSET_MASK, 0x57595);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_AC, RFREG_OFFSET_MASK, 0x70000);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x0F496);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x4F496);
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_IPA, RFREG_OFFSET_MASK, 0x8F496);
+
+		/* Back to RX Mode */
+		rtl_set_rfreg(hw, RF90_PATH_A, RF_AC, RFREG_OFFSET_MASK, 0x30000);
+
+		rtl_dbg(rtlpriv, COMP_RF, DBG_TRACE, "5G PA BIAS path A\n");
+	}
+
+	if (!(val8 & BIT(3)) && (is_single_mac || rtlhal->interfaceindex == 1)) {
+		rf_path = rtlhal->interfaceindex == 1 ? RF90_PATH_A : RF90_PATH_B;
+
+		/* 5GL_channel */
+		rtl_set_rfreg(hw, rf_path, RF_CHNLBW, RFREG_OFFSET_MASK, 0x17524);
+		rtl_set_rfreg(hw, rf_path, RF_AC, RFREG_OFFSET_MASK, 0x70000);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x0F496);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x4F496);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x8F496);
+
+		/* 5GM_channel */
+		rtl_set_rfreg(hw, rf_path, RF_CHNLBW, RFREG_OFFSET_MASK, 0x37564);
+		rtl_set_rfreg(hw, rf_path, RF_AC, RFREG_OFFSET_MASK, 0x70000);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x0F496);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x4F496);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x8F496);
+
+		/* 5GH_channel */
+		rtl_set_rfreg(hw, rf_path, RF_CHNLBW, RFREG_OFFSET_MASK, 0x57595);
+		rtl_set_rfreg(hw, rf_path, RF_AC, RFREG_OFFSET_MASK, 0x70000);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x0F496);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x4F496);
+		rtl_set_rfreg(hw, rf_path, RF_IPA, RFREG_OFFSET_MASK, 0x8F496);
+
+		/* Back to RX Mode */
+		rtl_set_rfreg(hw, rf_path, RF_AC, RFREG_OFFSET_MASK, 0x30000);
+
+		rtl_dbg(rtlpriv, COMP_RF, DBG_TRACE, "5G PA BIAS path B\n");
+	}
+}
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/phy.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/phy.h
new file mode 100644
index 000000000000..090a6203db7e
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/phy.h
@@ -0,0 +1,32 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#ifndef __RTL92DU_PHY_H__
+#define __RTL92DU_PHY_H__
+
+u32 rtl92du_phy_query_bb_reg(struct ieee80211_hw *hw,
+			     u32 regaddr, u32 bitmask);
+void rtl92du_phy_set_bb_reg(struct ieee80211_hw *hw,
+			    u32 regaddr, u32 bitmask, u32 data);
+bool rtl92du_phy_mac_config(struct ieee80211_hw *hw);
+bool rtl92du_phy_bb_config(struct ieee80211_hw *hw);
+bool rtl92du_phy_rf_config(struct ieee80211_hw *hw);
+void rtl92du_phy_set_bw_mode(struct ieee80211_hw *hw,
+			     enum nl80211_channel_type ch_type);
+u8 rtl92du_phy_sw_chnl(struct ieee80211_hw *hw);
+bool rtl92du_phy_config_rf_with_headerfile(struct ieee80211_hw *hw,
+					   enum rf_content content,
+					   enum radio_path rfpath);
+bool rtl92du_phy_set_rf_power_state(struct ieee80211_hw *hw,
+				    enum rf_pwrstate rfpwr_state);
+
+void rtl92du_phy_set_poweron(struct ieee80211_hw *hw);
+bool rtl92du_phy_check_poweroff(struct ieee80211_hw *hw);
+void rtl92du_phy_lc_calibrate(struct ieee80211_hw *hw, bool is2t);
+void rtl92du_update_bbrf_configuration(struct ieee80211_hw *hw);
+void rtl92du_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta);
+void rtl92du_phy_iq_calibrate(struct ieee80211_hw *hw);
+void rtl92du_phy_reload_iqk_setting(struct ieee80211_hw *hw, u8 channel);
+void rtl92du_phy_init_pa_bias(struct ieee80211_hw *hw);
+
+#endif
-- 
cgit v1.2.3-58-ga151


From 8321424134a400a5e3eb39f9acca6bc6946ff447 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:46:26 +0300
Subject: wifi: rtlwifi: Add rtl8192du/trx.{c,h}

These contain routines related to sending frames to the chip.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/bd2c24a0-a1fe-493a-8079-d83fda81a0d5@gmail.com
---
 .../net/wireless/realtek/rtlwifi/rtl8192du/trx.c   | 372 +++++++++++++++++++++
 .../net/wireless/realtek/rtlwifi/rtl8192du/trx.h   |  60 ++++
 2 files changed, 432 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/trx.c
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/trx.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/trx.c
new file mode 100644
index 000000000000..743ce0cfffe6
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/trx.c
@@ -0,0 +1,372 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#include "../wifi.h"
+#include "../base.h"
+#include "../usb.h"
+#include "../rtl8192d/reg.h"
+#include "../rtl8192d/def.h"
+#include "../rtl8192d/trx_common.h"
+#include "trx.h"
+
+void rtl92du_tx_cleanup(struct ieee80211_hw *hw, struct sk_buff *skb)
+{
+}
+
+int rtl92du_tx_post_hdl(struct ieee80211_hw *hw, struct urb *urb,
+			struct sk_buff *skb)
+{
+	return 0;
+}
+
+struct sk_buff *rtl92du_tx_aggregate_hdl(struct ieee80211_hw *hw,
+					 struct sk_buff_head *list)
+{
+	return skb_dequeue(list);
+}
+
+static enum rtl_desc_qsel _rtl92du_hwq_to_descq(u16 queue_index)
+{
+	switch (queue_index) {
+	case RTL_TXQ_BCN:
+		return QSLT_BEACON;
+	case RTL_TXQ_MGT:
+		return QSLT_MGNT;
+	case RTL_TXQ_VO:
+		return QSLT_VO;
+	case RTL_TXQ_VI:
+		return QSLT_VI;
+	case RTL_TXQ_BK:
+		return QSLT_BK;
+	default:
+	case RTL_TXQ_BE:
+		return QSLT_BE;
+	}
+}
+
+/* For HW recovery information */
+static void _rtl92du_tx_desc_checksum(__le32 *txdesc)
+{
+	__le16 *ptr = (__le16 *)txdesc;
+	u16 checksum = 0;
+	u32 index;
+
+	/* Clear first */
+	set_tx_desc_tx_desc_checksum(txdesc, 0);
+	for (index = 0; index < 16; index++)
+		checksum = checksum ^ le16_to_cpu(*(ptr + index));
+	set_tx_desc_tx_desc_checksum(txdesc, checksum);
+}
+
+void rtl92du_tx_fill_desc(struct ieee80211_hw *hw,
+			  struct ieee80211_hdr *hdr, u8 *pdesc_tx,
+			  u8 *pbd_desc_tx, struct ieee80211_tx_info *info,
+			  struct ieee80211_sta *sta,
+			  struct sk_buff *skb,
+			  u8 queue_index,
+			  struct rtl_tcb_desc *tcb_desc)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_ps_ctl *ppsc = rtl_psc(rtlpriv);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	struct rtl_mac *mac = rtl_mac(rtlpriv);
+	struct rtl_sta_info *sta_entry;
+	__le16 fc = hdr->frame_control;
+	u8 agg_state = RTL_AGG_STOP;
+	u16 pktlen = skb->len;
+	u32 rts_en, hw_rts_en;
+	u8 ampdu_density = 0;
+	u16 seq_number;
+	__le32 *txdesc;
+	u8 rate_flag;
+	u8 tid;
+
+	rtl_get_tcb_desc(hw, info, sta, skb, tcb_desc);
+
+	txdesc = (__le32 *)skb_push(skb, RTL_TX_HEADER_SIZE);
+	memset(txdesc, 0, RTL_TX_HEADER_SIZE);
+
+	set_tx_desc_pkt_size(txdesc, pktlen);
+	set_tx_desc_linip(txdesc, 0);
+	set_tx_desc_pkt_offset(txdesc, RTL_DUMMY_OFFSET);
+	set_tx_desc_offset(txdesc, RTL_TX_HEADER_SIZE);
+	/* 5G have no CCK rate */
+	if (rtlhal->current_bandtype == BAND_ON_5G)
+		if (tcb_desc->hw_rate < DESC_RATE6M)
+			tcb_desc->hw_rate = DESC_RATE6M;
+
+	set_tx_desc_tx_rate(txdesc, tcb_desc->hw_rate);
+	if (tcb_desc->use_shortgi || tcb_desc->use_shortpreamble)
+		set_tx_desc_data_shortgi(txdesc, 1);
+
+	if (rtlhal->macphymode == DUALMAC_DUALPHY &&
+	    tcb_desc->hw_rate == DESC_RATEMCS7)
+		set_tx_desc_data_shortgi(txdesc, 1);
+
+	if (sta) {
+		sta_entry = (struct rtl_sta_info *)sta->drv_priv;
+		tid = ieee80211_get_tid(hdr);
+		agg_state = sta_entry->tids[tid].agg.agg_state;
+		ampdu_density = sta->deflink.ht_cap.ampdu_density;
+	}
+
+	if (agg_state == RTL_AGG_OPERATIONAL &&
+	    info->flags & IEEE80211_TX_CTL_AMPDU) {
+		set_tx_desc_agg_enable(txdesc, 1);
+		set_tx_desc_max_agg_num(txdesc, 0x14);
+		set_tx_desc_ampdu_density(txdesc, ampdu_density);
+		tcb_desc->rts_enable = 1;
+		tcb_desc->rts_rate = DESC_RATE24M;
+	} else {
+		set_tx_desc_agg_break(txdesc, 1);
+	}
+	seq_number = (le16_to_cpu(hdr->seq_ctrl) & IEEE80211_SCTL_SEQ) >> 4;
+	set_tx_desc_seq(txdesc, seq_number);
+
+	rts_en = tcb_desc->rts_enable && !tcb_desc->cts_enable;
+	hw_rts_en = tcb_desc->rts_enable || tcb_desc->cts_enable;
+	set_tx_desc_rts_enable(txdesc, rts_en);
+	set_tx_desc_hw_rts_enable(txdesc, hw_rts_en);
+	set_tx_desc_cts2self(txdesc, tcb_desc->cts_enable);
+	set_tx_desc_rts_stbc(txdesc, tcb_desc->rts_stbc);
+	/* 5G have no CCK rate */
+	if (rtlhal->current_bandtype == BAND_ON_5G)
+		if (tcb_desc->rts_rate < DESC_RATE6M)
+			tcb_desc->rts_rate = DESC_RATE6M;
+	set_tx_desc_rts_rate(txdesc, tcb_desc->rts_rate);
+	set_tx_desc_rts_bw(txdesc, 0);
+	set_tx_desc_rts_sc(txdesc, tcb_desc->rts_sc);
+	set_tx_desc_rts_short(txdesc, tcb_desc->rts_use_shortpreamble);
+
+	rate_flag = info->control.rates[0].flags;
+	if (mac->bw_40) {
+		if (rate_flag & IEEE80211_TX_RC_DUP_DATA) {
+			set_tx_desc_data_bw(txdesc, 1);
+			set_tx_desc_tx_sub_carrier(txdesc, 3);
+		} else if (rate_flag & IEEE80211_TX_RC_40_MHZ_WIDTH) {
+			set_tx_desc_data_bw(txdesc, 1);
+			set_tx_desc_tx_sub_carrier(txdesc, mac->cur_40_prime_sc);
+		} else {
+			set_tx_desc_data_bw(txdesc, 0);
+			set_tx_desc_tx_sub_carrier(txdesc, 0);
+		}
+	} else {
+		set_tx_desc_data_bw(txdesc, 0);
+		set_tx_desc_tx_sub_carrier(txdesc, 0);
+	}
+
+	if (info->control.hw_key) {
+		struct ieee80211_key_conf *keyconf = info->control.hw_key;
+
+		switch (keyconf->cipher) {
+		case WLAN_CIPHER_SUITE_WEP40:
+		case WLAN_CIPHER_SUITE_WEP104:
+		case WLAN_CIPHER_SUITE_TKIP:
+			set_tx_desc_sec_type(txdesc, 0x1);
+			break;
+		case WLAN_CIPHER_SUITE_CCMP:
+			set_tx_desc_sec_type(txdesc, 0x3);
+			break;
+		default:
+			set_tx_desc_sec_type(txdesc, 0x0);
+			break;
+		}
+	}
+
+	set_tx_desc_pkt_id(txdesc, 0);
+	set_tx_desc_queue_sel(txdesc, _rtl92du_hwq_to_descq(queue_index));
+	set_tx_desc_data_rate_fb_limit(txdesc, 0x1F);
+	set_tx_desc_rts_rate_fb_limit(txdesc, 0xF);
+	set_tx_desc_disable_fb(txdesc, 0);
+	set_tx_desc_use_rate(txdesc, tcb_desc->use_driver_rate);
+
+	if (ieee80211_is_data_qos(fc)) {
+		if (mac->rdg_en) {
+			rtl_dbg(rtlpriv, COMP_SEND, DBG_TRACE,
+				"Enable RDG function\n");
+			set_tx_desc_rdg_enable(txdesc, 1);
+			set_tx_desc_htc(txdesc, 1);
+		}
+		set_tx_desc_qos(txdesc, 1);
+	}
+
+	if (rtlpriv->dm.useramask) {
+		set_tx_desc_rate_id(txdesc, tcb_desc->ratr_index);
+		set_tx_desc_macid(txdesc, tcb_desc->mac_id);
+	} else {
+		set_tx_desc_rate_id(txdesc, 0xC + tcb_desc->ratr_index);
+		set_tx_desc_macid(txdesc, tcb_desc->ratr_index);
+	}
+
+	if (!ieee80211_is_data_qos(fc) && ppsc->leisure_ps &&
+	    ppsc->fwctrl_lps) {
+		set_tx_desc_hwseq_en(txdesc, 1);
+		set_tx_desc_pkt_id(txdesc, 8);
+	}
+
+	if (ieee80211_has_morefrags(fc))
+		set_tx_desc_more_frag(txdesc, 1);
+	if (is_multicast_ether_addr(ieee80211_get_DA(hdr)) ||
+	    is_broadcast_ether_addr(ieee80211_get_DA(hdr)))
+		set_tx_desc_bmc(txdesc, 1);
+
+	set_tx_desc_own(txdesc, 1);
+	set_tx_desc_last_seg(txdesc, 1);
+	set_tx_desc_first_seg(txdesc, 1);
+	_rtl92du_tx_desc_checksum(txdesc);
+
+	rtl_dbg(rtlpriv, COMP_SEND, DBG_TRACE, "==>\n");
+}
+
+static void _rtl92du_config_out_ep(struct ieee80211_hw *hw, u8 num_out_pipe)
+{
+	struct rtl_usb *rtlusb = rtl_usbdev(rtl_usbpriv(hw));
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	u16 ep_cfg;
+
+	rtlusb->out_queue_sel = 0;
+	rtlusb->out_ep_nums = 0;
+
+	if (rtlhal->interfaceindex == 0)
+		ep_cfg = rtl_read_word(rtlpriv, REG_USB_Queue_Select_MAC0);
+	else
+		ep_cfg = rtl_read_word(rtlpriv, REG_USB_Queue_Select_MAC1);
+
+	if (ep_cfg & 0x00f) {
+		rtlusb->out_queue_sel |= TX_SELE_HQ;
+		rtlusb->out_ep_nums++;
+	}
+	if (ep_cfg & 0x0f0) {
+		rtlusb->out_queue_sel |= TX_SELE_NQ;
+		rtlusb->out_ep_nums++;
+	}
+	if (ep_cfg & 0xf00) {
+		rtlusb->out_queue_sel |= TX_SELE_LQ;
+		rtlusb->out_ep_nums++;
+	}
+
+	switch (num_out_pipe) {
+	case 3:
+		rtlusb->out_queue_sel = TX_SELE_HQ | TX_SELE_NQ | TX_SELE_LQ;
+		rtlusb->out_ep_nums = 3;
+		break;
+	case 2:
+		rtlusb->out_queue_sel = TX_SELE_HQ | TX_SELE_NQ;
+		rtlusb->out_ep_nums = 2;
+		break;
+	case 1:
+		rtlusb->out_queue_sel = TX_SELE_HQ;
+		rtlusb->out_ep_nums = 1;
+		break;
+	default:
+		break;
+	}
+}
+
+static void _rtl92du_one_out_ep_mapping(struct rtl_usb *rtlusb,
+					struct rtl_ep_map *ep_map)
+{
+	ep_map->ep_mapping[RTL_TXQ_BE] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_BK] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_VI] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_VO] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_MGT] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_BCN] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_HI] = rtlusb->out_eps[0];
+}
+
+static void _rtl92du_two_out_ep_mapping(struct rtl_usb *rtlusb,
+					struct rtl_ep_map *ep_map)
+{
+	ep_map->ep_mapping[RTL_TXQ_BE] = rtlusb->out_eps[1];
+	ep_map->ep_mapping[RTL_TXQ_BK] = rtlusb->out_eps[1];
+	ep_map->ep_mapping[RTL_TXQ_VI] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_VO] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_MGT] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_BCN] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_HI] = rtlusb->out_eps[0];
+}
+
+static void _rtl92du_three_out_ep_mapping(struct rtl_usb *rtlusb,
+					  struct rtl_ep_map *ep_map)
+{
+	ep_map->ep_mapping[RTL_TXQ_BE] = rtlusb->out_eps[2];
+	ep_map->ep_mapping[RTL_TXQ_BK] = rtlusb->out_eps[2];
+	ep_map->ep_mapping[RTL_TXQ_VI] = rtlusb->out_eps[1];
+	ep_map->ep_mapping[RTL_TXQ_VO] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_MGT] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_BCN] = rtlusb->out_eps[0];
+	ep_map->ep_mapping[RTL_TXQ_HI] = rtlusb->out_eps[0];
+}
+
+static int _rtl92du_out_ep_mapping(struct ieee80211_hw *hw)
+{
+	struct rtl_usb *rtlusb = rtl_usbdev(rtl_usbpriv(hw));
+	struct rtl_ep_map *ep_map = &rtlusb->ep_map;
+
+	switch (rtlusb->out_ep_nums) {
+	case 1:
+		_rtl92du_one_out_ep_mapping(rtlusb, ep_map);
+		break;
+	case 2:
+		_rtl92du_two_out_ep_mapping(rtlusb, ep_map);
+		break;
+	case 3:
+		_rtl92du_three_out_ep_mapping(rtlusb, ep_map);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int rtl92du_endpoint_mapping(struct ieee80211_hw *hw)
+{
+	struct rtl_usb *rtlusb = rtl_usbdev(rtl_usbpriv(hw));
+
+	_rtl92du_config_out_ep(hw, rtlusb->out_ep_nums);
+
+	/* Normal chip with one IN and one OUT doesn't have interrupt IN EP. */
+	if (rtlusb->out_ep_nums == 1 && rtlusb->in_ep_nums != 1)
+		return -EINVAL;
+
+	return _rtl92du_out_ep_mapping(hw);
+}
+
+u16 rtl92du_mq_to_hwq(__le16 fc, u16 mac80211_queue_index)
+{
+	u16 hw_queue_index;
+
+	if (unlikely(ieee80211_is_beacon(fc))) {
+		hw_queue_index = RTL_TXQ_BCN;
+		goto out;
+	}
+	if (ieee80211_is_mgmt(fc)) {
+		hw_queue_index = RTL_TXQ_MGT;
+		goto out;
+	}
+
+	switch (mac80211_queue_index) {
+	case 0:
+		hw_queue_index = RTL_TXQ_VO;
+		break;
+	case 1:
+		hw_queue_index = RTL_TXQ_VI;
+		break;
+	case 2:
+		hw_queue_index = RTL_TXQ_BE;
+		break;
+	case 3:
+		hw_queue_index = RTL_TXQ_BK;
+		break;
+	default:
+		hw_queue_index = RTL_TXQ_BE;
+		WARN_ONCE(true, "rtl8192du: QSLT_BE queue, skb_queue:%d\n",
+			  mac80211_queue_index);
+		break;
+	}
+out:
+	return hw_queue_index;
+}
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/trx.h
new file mode 100644
index 000000000000..8c3d24622fa7
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/trx.h
@@ -0,0 +1,60 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#ifndef __RTL92DU_TRX_H__
+#define __RTL92DU_TRX_H__
+
+#define TX_SELE_HQ				BIT(0)	/* High Queue */
+#define TX_SELE_LQ				BIT(1)	/* Low Queue */
+#define TX_SELE_NQ				BIT(2)	/* Normal Queue */
+
+#define TX_TOTAL_PAGE_NUMBER_92DU			0xF8
+#define TEST_PAGE_NUM_PUBQ_92DU				0x89
+#define TX_TOTAL_PAGE_NUMBER_92D_DUAL_MAC		0x7A
+#define NORMAL_PAGE_NUM_PUBQ_92D_DUAL_MAC		0x5A
+#define NORMAL_PAGE_NUM_HPQ_92D_DUAL_MAC		0x10
+#define NORMAL_PAGE_NUM_LPQ_92D_DUAL_MAC		0x10
+#define NORMAL_PAGE_NUM_NORMALQ_92D_DUAL_MAC		0
+
+#define WMM_NORMAL_TX_TOTAL_PAGE_NUMBER			0xF5
+
+#define WMM_NORMAL_PAGE_NUM_PUBQ_92D			0x65
+#define WMM_NORMAL_PAGE_NUM_HPQ_92D			0x30
+#define WMM_NORMAL_PAGE_NUM_LPQ_92D			0x30
+#define WMM_NORMAL_PAGE_NUM_NPQ_92D			0x30
+
+#define WMM_NORMAL_PAGE_NUM_PUBQ_92D_DUAL_MAC		0x32
+#define WMM_NORMAL_PAGE_NUM_HPQ_92D_DUAL_MAC		0x18
+#define WMM_NORMAL_PAGE_NUM_LPQ_92D_DUAL_MAC		0x18
+#define WMM_NORMAL_PAGE_NUM_NPQ_92D_DUAL_MAC		0x18
+
+static inline void set_tx_desc_bmc(__le32 *__txdesc, u32 __value)
+{
+	le32p_replace_bits(__txdesc, __value, BIT(24));
+}
+
+static inline void set_tx_desc_agg_break(__le32 *__txdesc, u32 __value)
+{
+	le32p_replace_bits((__txdesc + 1), __value, BIT(6));
+}
+
+static inline void set_tx_desc_tx_desc_checksum(__le32 *__txdesc, u32 __value)
+{
+	le32p_replace_bits((__txdesc + 7), __value, GENMASK(15, 0));
+}
+
+void rtl92du_tx_fill_desc(struct ieee80211_hw *hw,
+			  struct ieee80211_hdr *hdr, u8 *pdesc,
+			  u8 *pbd_desc_tx, struct ieee80211_tx_info *info,
+			  struct ieee80211_sta *sta,
+			  struct sk_buff *skb, u8 hw_queue,
+			  struct rtl_tcb_desc *ptcb_desc);
+int rtl92du_endpoint_mapping(struct ieee80211_hw *hw);
+u16 rtl92du_mq_to_hwq(__le16 fc, u16 mac80211_queue_index);
+struct sk_buff *rtl92du_tx_aggregate_hdl(struct ieee80211_hw *hw,
+					 struct sk_buff_head *list);
+void rtl92du_tx_cleanup(struct ieee80211_hw *hw, struct sk_buff  *skb);
+int rtl92du_tx_post_hdl(struct ieee80211_hw *hw, struct urb *urb,
+			struct sk_buff *skb);
+
+#endif
-- 
cgit v1.2.3-58-ga151


From c0e4ae46ffa20fa12f1748fffcd4c95590da80d4 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:46:51 +0300
Subject: wifi: rtlwifi: Add rtl8192du/rf.{c,h}

These contain one RF configuration function and some functions related
to dual MAC operation.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/36258d72-11fd-4f70-84c6-782972f706ed@gmail.com
---
 .../net/wireless/realtek/rtlwifi/rtl8192du/rf.c    | 240 +++++++++++++++++++++
 .../net/wireless/realtek/rtlwifi/rtl8192du/rf.h    |  11 +
 2 files changed, 251 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/rf.c
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/rf.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/rf.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/rf.c
new file mode 100644
index 000000000000..044dd65eafd0
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/rf.c
@@ -0,0 +1,240 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#include "../wifi.h"
+#include "../rtl8192d/reg.h"
+#include "../rtl8192d/phy_common.h"
+#include "phy.h"
+#include "rf.h"
+
+bool rtl92du_phy_enable_anotherphy(struct ieee80211_hw *hw, bool bmac0)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	u8 mac_on_bit = bmac0 ? MAC1_ON : MAC0_ON;
+	u8 mac_reg = bmac0 ? REG_MAC1 : REG_MAC0;
+	bool bresult = true; /* true: need to enable BB/RF power */
+	u32 maskforphyset = 0;
+	u16 val16;
+	u8 u1btmp;
+
+	rtlhal->during_mac0init_radiob = false;
+	rtlhal->during_mac1init_radioa = false;
+	rtl_dbg(rtlpriv, COMP_RF, DBG_LOUD, "===>\n");
+
+	/* MAC0 Need PHY1 load radio_b.txt . Driver use DBI to write. */
+	u1btmp = rtl_read_byte(rtlpriv, mac_reg);
+	if (!(u1btmp & mac_on_bit)) {
+		rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD, "enable BB & RF\n");
+		/* Enable BB and RF power */
+
+		maskforphyset = bmac0 ? MAC0_ACCESS_PHY1 : MAC1_ACCESS_PHY0;
+
+		val16 = rtl_read_word(rtlpriv, REG_SYS_FUNC_EN | maskforphyset);
+		val16 &= 0xfffc;
+		rtl_write_word(rtlpriv, REG_SYS_FUNC_EN | maskforphyset, val16);
+
+		val16 = rtl_read_word(rtlpriv, REG_SYS_FUNC_EN | maskforphyset);
+		val16 |= BIT(13) | BIT(0) | BIT(1);
+		rtl_write_word(rtlpriv, REG_SYS_FUNC_EN | maskforphyset, val16);
+	} else {
+		/* We think if MAC1 is ON,then radio_a.txt
+		 * and radio_b.txt has been load.
+		 */
+		bresult = false;
+	}
+	rtl_dbg(rtlpriv, COMP_RF, DBG_LOUD, "<===\n");
+	return bresult;
+}
+
+void rtl92du_phy_powerdown_anotherphy(struct ieee80211_hw *hw, bool bmac0)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	u8 mac_on_bit = bmac0 ? MAC1_ON : MAC0_ON;
+	u8 mac_reg = bmac0 ? REG_MAC1 : REG_MAC0;
+	u32 maskforphyset = 0;
+	u8 u1btmp;
+
+	rtlhal->during_mac0init_radiob = false;
+	rtlhal->during_mac1init_radioa = false;
+	rtl_dbg(rtlpriv, COMP_RF, DBG_LOUD, "====>\n");
+
+	/* check MAC0 enable or not again now, if
+	 * enabled, not power down radio A.
+	 */
+	u1btmp = rtl_read_byte(rtlpriv, mac_reg);
+	if (!(u1btmp & mac_on_bit)) {
+		rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD, "power down\n");
+		/* power down RF radio A according to YuNan's advice. */
+		maskforphyset = bmac0 ? MAC0_ACCESS_PHY1 : MAC1_ACCESS_PHY0;
+		rtl_write_dword(rtlpriv, RFPGA0_XA_LSSIPARAMETER | maskforphyset,
+				0x00000000);
+	}
+	rtl_dbg(rtlpriv, COMP_RF, DBG_LOUD, "<====\n");
+}
+
+bool rtl92du_phy_rf6052_config(struct ieee80211_hw *hw)
+{
+	bool mac1_initradioa_first = false, mac0_initradiob_first = false;
+	bool need_pwrdown_radioa = false, need_pwrdown_radiob = false;
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = &rtlpriv->rtlhal;
+	struct rtl_phy *rtlphy = &rtlpriv->phy;
+	struct bb_reg_def *pphyreg;
+	bool true_bpath = false;
+	bool rtstatus = true;
+	u32 u4_regvalue = 0;
+	u8 rfpath;
+
+	if (rtlphy->rf_type == RF_1T1R)
+		rtlphy->num_total_rfpath = 1;
+	else
+		rtlphy->num_total_rfpath = 2;
+
+	/* Single phy mode: use radio_a radio_b config path_A path_B
+	 * separately by MAC0, and MAC1 needn't configure RF;
+	 * Dual PHY mode: MAC0 use radio_a config 1st phy path_A,
+	 * MAC1 use radio_b config 2nd PHY path_A.
+	 * DMDP, MAC0 on G band, MAC1 on A band.
+	 */
+	if (rtlhal->macphymode == DUALMAC_DUALPHY) {
+		if (rtlhal->current_bandtype == BAND_ON_2_4G &&
+		    rtlhal->interfaceindex == 0) {
+			/* MAC0 needs PHY1 load radio_b.txt. */
+			if (rtl92du_phy_enable_anotherphy(hw, true)) {
+				rtlphy->num_total_rfpath = 2;
+				mac0_initradiob_first = true;
+			} else {
+				/* We think if MAC1 is ON,then radio_a.txt and
+				 * radio_b.txt has been load.
+				 */
+				return rtstatus;
+			}
+		} else if (rtlhal->current_bandtype == BAND_ON_5G &&
+			   rtlhal->interfaceindex == 1) {
+			/* MAC1 needs PHY0 load radio_a.txt. */
+			if (rtl92du_phy_enable_anotherphy(hw, false)) {
+				rtlphy->num_total_rfpath = 2;
+				mac1_initradioa_first = true;
+			} else {
+				/* We think if MAC0 is ON, then radio_a.txt and
+				 * radio_b.txt has been load.
+				 */
+				return rtstatus;
+			}
+		} else if (rtlhal->interfaceindex == 1) {
+			/* MAC0 enabled, only init radia B.   */
+			true_bpath = true;
+		}
+	}
+
+	for (rfpath = 0; rfpath < rtlphy->num_total_rfpath; rfpath++) {
+		/* Mac1 use PHY0 write */
+		if (mac1_initradioa_first) {
+			if (rfpath == RF90_PATH_A) {
+				rtlhal->during_mac1init_radioa = true;
+				need_pwrdown_radioa = true;
+			} else if (rfpath == RF90_PATH_B) {
+				rtlhal->during_mac1init_radioa = false;
+				mac1_initradioa_first = false;
+				rfpath = RF90_PATH_A;
+				true_bpath = true;
+				rtlphy->num_total_rfpath = 1;
+			}
+		} else if (mac0_initradiob_first) {
+			/* Mac0 use PHY1 write */
+			if (rfpath == RF90_PATH_A)
+				rtlhal->during_mac0init_radiob = false;
+			if (rfpath == RF90_PATH_B) {
+				rtlhal->during_mac0init_radiob = true;
+				mac0_initradiob_first = false;
+				need_pwrdown_radiob = true;
+				rfpath = RF90_PATH_A;
+				true_bpath = true;
+				rtlphy->num_total_rfpath = 1;
+			}
+		}
+
+		pphyreg = &rtlphy->phyreg_def[rfpath];
+
+		switch (rfpath) {
+		case RF90_PATH_A:
+		case RF90_PATH_C:
+			u4_regvalue = rtl_get_bbreg(hw, pphyreg->rfintfs,
+						    BRFSI_RFENV);
+			break;
+		case RF90_PATH_B:
+		case RF90_PATH_D:
+			u4_regvalue = rtl_get_bbreg(hw, pphyreg->rfintfs,
+						    BRFSI_RFENV << 16);
+			break;
+		}
+
+		rtl_set_bbreg(hw, pphyreg->rfintfe, BRFSI_RFENV << 16, 0x1);
+		udelay(1);
+		rtl_set_bbreg(hw, pphyreg->rfintfo, BRFSI_RFENV, 0x1);
+		udelay(1);
+
+		/* Set bit number of Address and Data for RF register */
+		rtl_set_bbreg(hw, pphyreg->rfhssi_para2,
+			      B3WIREADDRESSLENGTH, 0x0);
+		udelay(1);
+		rtl_set_bbreg(hw, pphyreg->rfhssi_para2, B3WIREDATALENGTH, 0x0);
+		udelay(1);
+
+		switch (rfpath) {
+		case RF90_PATH_A:
+			if (true_bpath)
+				rtstatus = rtl92du_phy_config_rf_with_headerfile(
+						hw, radiob_txt,
+						(enum radio_path)rfpath);
+			else
+				rtstatus = rtl92du_phy_config_rf_with_headerfile(
+						hw, radioa_txt,
+						(enum radio_path)rfpath);
+			break;
+		case RF90_PATH_B:
+			rtstatus =
+			    rtl92du_phy_config_rf_with_headerfile(hw, radiob_txt,
+						(enum radio_path)rfpath);
+			break;
+		case RF90_PATH_C:
+			break;
+		case RF90_PATH_D:
+			break;
+		}
+
+		switch (rfpath) {
+		case RF90_PATH_A:
+		case RF90_PATH_C:
+			rtl_set_bbreg(hw, pphyreg->rfintfs, BRFSI_RFENV,
+				      u4_regvalue);
+			break;
+		case RF90_PATH_B:
+		case RF90_PATH_D:
+			rtl_set_bbreg(hw, pphyreg->rfintfs, BRFSI_RFENV << 16,
+				      u4_regvalue);
+			break;
+		}
+
+		if (!rtstatus) {
+			rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE,
+				"Radio[%d] Fail!!\n", rfpath);
+			return rtstatus;
+		}
+	}
+
+	/* check MAC0 enable or not again, if enabled,
+	 * not power down radio A.
+	 * check MAC1 enable or not again, if enabled,
+	 * not power down radio B.
+	 */
+	if (need_pwrdown_radioa)
+		rtl92du_phy_powerdown_anotherphy(hw, false);
+	else if (need_pwrdown_radiob)
+		rtl92du_phy_powerdown_anotherphy(hw, true);
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_TRACE, "<---\n");
+
+	return rtstatus;
+}
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/rf.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/rf.h
new file mode 100644
index 000000000000..4a92cbdd00c0
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/rf.h
@@ -0,0 +1,11 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#ifndef __RTL92DU_RF_H__
+#define __RTL92DU_RF_H__
+
+bool rtl92du_phy_rf6052_config(struct ieee80211_hw *hw);
+bool rtl92du_phy_enable_anotherphy(struct ieee80211_hw *hw, bool bmac0);
+void rtl92du_phy_powerdown_anotherphy(struct ieee80211_hw *hw, bool bmac0);
+
+#endif
-- 
cgit v1.2.3-58-ga151


From 91f43aee0b1a5e8ba0ea7573e7473546b5cc7292 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:47:19 +0300
Subject: wifi: rtlwifi: Add rtl8192du/fw.{c,h} and rtl8192du/led.{c,h}

fw.c contains a function for loading the firmware.
led.c contains a function for controlling the LED.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/a52b8907-267c-4811-90ca-5eb852a2aa09@gmail.com
---
 .../net/wireless/realtek/rtlwifi/rtl8192du/fw.c    | 63 ++++++++++++++++++++++
 .../net/wireless/realtek/rtlwifi/rtl8192du/fw.h    |  9 ++++
 .../net/wireless/realtek/rtlwifi/rtl8192du/led.c   | 10 ++++
 .../net/wireless/realtek/rtlwifi/rtl8192du/led.h   |  9 ++++
 4 files changed, 91 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/fw.c
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/fw.h
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/led.c
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/led.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/fw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/fw.c
new file mode 100644
index 000000000000..f74e4e84fe39
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/fw.c
@@ -0,0 +1,63 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#include "../wifi.h"
+#include "../rtl8192d/reg.h"
+#include "../rtl8192d/def.h"
+#include "../rtl8192d/fw_common.h"
+#include "fw.h"
+
+int rtl92du_download_fw(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	enum version_8192d version = rtlhal->version;
+	u8 *pfwheader;
+	u8 *pfwdata;
+	u32 fwsize;
+	int err;
+
+	if (rtlpriv->max_fw_size == 0 || !rtlhal->pfirmware)
+		return 1;
+
+	fwsize = rtlhal->fwsize;
+	pfwheader = rtlhal->pfirmware;
+	pfwdata = rtlhal->pfirmware;
+	rtlhal->fw_version = (u16)GET_FIRMWARE_HDR_VERSION(pfwheader);
+	rtlhal->fw_subversion = (u16)GET_FIRMWARE_HDR_SUB_VER(pfwheader);
+
+	rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+		"FirmwareVersion(%d), FirmwareSubVersion(%d), Signature(%#x)\n",
+		rtlhal->fw_version, rtlhal->fw_subversion,
+		GET_FIRMWARE_HDR_SIGNATURE(pfwheader));
+
+	if (IS_FW_HEADER_EXIST(pfwheader)) {
+		rtl_dbg(rtlpriv, COMP_INIT, DBG_LOUD,
+			"Shift 32 bytes for FW header!!\n");
+		pfwdata = pfwdata + 32;
+		fwsize = fwsize - 32;
+	}
+
+	if (rtl92d_is_fw_downloaded(rtlpriv))
+		goto exit;
+
+	/* If 8051 is running in RAM code, driver should
+	 * inform Fw to reset by itself, or it will cause
+	 * download Fw fail.
+	 */
+	if (rtl_read_byte(rtlpriv, REG_MCUFWDL) & BIT(7)) {
+		rtl92d_firmware_selfreset(hw);
+		rtl_write_byte(rtlpriv, REG_MCUFWDL, 0x00);
+	}
+
+	rtl92d_enable_fw_download(hw, true);
+	rtl92d_write_fw(hw, version, pfwdata, fwsize);
+	rtl92d_enable_fw_download(hw, false);
+
+	err = rtl92d_fw_free_to_go(hw);
+	if (err)
+		pr_err("fw is not ready to run!\n");
+exit:
+	err = rtl92d_fw_init(hw);
+	return err;
+}
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/fw.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/fw.h
new file mode 100644
index 000000000000..7904bfbda4ba
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/fw.h
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#ifndef __RTL92DU_FW_H__
+#define __RTL92DU_FW_H__
+
+int rtl92du_download_fw(struct ieee80211_hw *hw);
+
+#endif
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/led.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/led.c
new file mode 100644
index 000000000000..6c12dfbd6367
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/led.c
@@ -0,0 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#include "../wifi.h"
+#include "led.h"
+
+void rtl92du_led_control(struct ieee80211_hw *hw, enum led_ctl_mode ledaction)
+{
+	/* The hardware has control. */
+}
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/led.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/led.h
new file mode 100644
index 000000000000..d7ebc8afcc7b
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/led.h
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#ifndef __RTL92DU_LED_H__
+#define __RTL92DU_LED_H__
+
+void rtl92du_led_control(struct ieee80211_hw *hw, enum led_ctl_mode ledaction);
+
+#endif
-- 
cgit v1.2.3-58-ga151


From 3f8b94dbff7ed9cbf07b397b08387da459085c6c Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:47:40 +0300
Subject: wifi: rtlwifi: Add rtl8192du/dm.{c,h}

These contain functions related to the dynamic mechanism, which runs
every two seconds to adjust to changes in the environment.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/a010ed6d-efb0-456d-bd4a-dfc6da23e7aa@gmail.com
---
 .../net/wireless/realtek/rtlwifi/rtl8192du/dm.c    | 120 +++++++++++++++++++++
 .../net/wireless/realtek/rtlwifi/rtl8192du/dm.h    |  10 ++
 2 files changed, 130 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/dm.c
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/dm.h

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/dm.c
new file mode 100644
index 000000000000..dd57707a9184
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/dm.c
@@ -0,0 +1,120 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#include "../wifi.h"
+#include "../core.h"
+#include "../rtl8192d/reg.h"
+#include "../rtl8192d/def.h"
+#include "../rtl8192d/dm_common.h"
+#include "../rtl8192d/fw_common.h"
+#include "dm.h"
+
+static void rtl92du_dm_init_1r_cca(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct ps_t *dm_pstable = &rtlpriv->dm_pstable;
+
+	dm_pstable->pre_ccastate = CCA_MAX;
+	dm_pstable->cur_ccasate = CCA_MAX;
+}
+
+static void rtl92du_dm_1r_cca(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct ps_t *dm_pstable = &rtlpriv->dm_pstable;
+	struct rtl_hal *rtlhal = rtl_hal(rtlpriv);
+	int pwdb = rtlpriv->dm_digtable.min_undec_pwdb_for_dm;
+
+	if (rtlhal->macphymode != SINGLEMAC_SINGLEPHY ||
+	    rtlhal->current_bandtype != BAND_ON_5G)
+		return;
+
+	if (pwdb != 0) {
+		if (dm_pstable->pre_ccastate == CCA_2R ||
+		    dm_pstable->pre_ccastate == CCA_MAX)
+			dm_pstable->cur_ccasate = (pwdb >= 35) ? CCA_1R : CCA_2R;
+		else
+			dm_pstable->cur_ccasate = (pwdb <= 30) ? CCA_2R : CCA_1R;
+	} else {
+		dm_pstable->cur_ccasate = CCA_MAX;
+	}
+
+	if (dm_pstable->pre_ccastate == dm_pstable->cur_ccasate)
+		return;
+
+	rtl_dbg(rtlpriv, COMP_BB_POWERSAVING, DBG_TRACE,
+		"Old CCA state: %d new CCA state: %d\n",
+		dm_pstable->pre_ccastate, dm_pstable->cur_ccasate);
+
+	if (dm_pstable->cur_ccasate == CCA_1R) {
+		if (rtlpriv->phy.rf_type == RF_2T2R)
+			rtl_set_bbreg(hw, ROFDM0_TRXPATHENABLE, MASKBYTE0, 0x13);
+		else /* Is this branch reachable? */
+			rtl_set_bbreg(hw, ROFDM0_TRXPATHENABLE, MASKBYTE0, 0x23);
+	} else { /* CCA_2R or CCA_MAX */
+		rtl_set_bbreg(hw, ROFDM0_TRXPATHENABLE, MASKBYTE0, 0x33);
+	}
+}
+
+static void rtl92du_dm_pwdb_monitor(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	const u32 max_macid = 32;
+	u32 temp;
+
+	/* AP & ADHOC & MESH will return tmp */
+	if (rtlpriv->mac80211.opmode != NL80211_IFTYPE_STATION)
+		return;
+
+	/* Indicate Rx signal strength to FW. */
+	if (rtlpriv->dm.useramask) {
+		temp = rtlpriv->dm.undec_sm_pwdb << 16;
+		temp |= max_macid << 8;
+
+		rtl92d_fill_h2c_cmd(hw, H2C_RSSI_REPORT, 3, (u8 *)(&temp));
+	} else {
+		rtl_write_byte(rtlpriv, 0x4fe, (u8)rtlpriv->dm.undec_sm_pwdb);
+	}
+}
+
+void rtl92du_dm_init(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+
+	rtlpriv->dm.dm_type = DM_TYPE_BYDRIVER;
+	rtl_dm_diginit(hw, 0x20);
+	rtlpriv->dm_digtable.rx_gain_max = DM_DIG_FA_UPPER;
+	rtlpriv->dm_digtable.rx_gain_min = DM_DIG_FA_LOWER;
+	rtl92d_dm_init_edca_turbo(hw);
+	rtl92du_dm_init_1r_cca(hw);
+	rtl92d_dm_init_rate_adaptive_mask(hw);
+	rtl92d_dm_initialize_txpower_tracking(hw);
+}
+
+void rtl92du_dm_watchdog(struct ieee80211_hw *hw)
+{
+	struct rtl_ps_ctl *ppsc = rtl_psc(rtl_priv(hw));
+	bool fw_current_inpsmode = false;
+	bool fwps_awake = true;
+
+	/* 1. RF is OFF. (No need to do DM.)
+	 * 2. Fw is under power saving mode for FwLPS.
+	 *    (Prevent from SW/FW I/O racing.)
+	 * 3. IPS workitem is scheduled. (Prevent from IPS sequence
+	 *    to be swapped with DM.
+	 * 4. RFChangeInProgress is TRUE.
+	 *    (Prevent from broken by IPS/HW/SW Rf off.)
+	 */
+
+	if (ppsc->rfpwr_state != ERFON || fw_current_inpsmode ||
+	    !fwps_awake || ppsc->rfchange_inprogress)
+		return;
+
+	rtl92du_dm_pwdb_monitor(hw);
+	rtl92d_dm_false_alarm_counter_statistics(hw);
+	rtl92d_dm_find_minimum_rssi(hw);
+	rtl92d_dm_dig(hw);
+	rtl92d_dm_check_txpower_tracking_thermal_meter(hw);
+	rtl92d_dm_check_edca_turbo(hw);
+	rtl92du_dm_1r_cca(hw);
+}
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/dm.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/dm.h
new file mode 100644
index 000000000000..2f283bf1e4d8
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/dm.h
@@ -0,0 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#ifndef __RTL92DU_DM_H__
+#define __RTL92DU_DM_H__
+
+void rtl92du_dm_init(struct ieee80211_hw *hw);
+void rtl92du_dm_watchdog(struct ieee80211_hw *hw);
+
+#endif
-- 
cgit v1.2.3-58-ga151


From 1d8b5003a55ccab64e865d2f34ab1fd7a4cfe3bb Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:48:02 +0300
Subject: wifi: rtlwifi: Constify rtl_hal_cfg.{ops,usb_interface_cfg} and
 rtl_priv.cfg

This allows the drivers to declare the structs rtl_hal_cfg, rtl_hal_ops,
and rtl_hal_usbint_cfg as const.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/1b29994f-3d07-4297-875d-57c3a87a1ec6@gmail.com
---
 drivers/net/wireless/realtek/rtlwifi/base.c         | 2 +-
 drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c | 3 +--
 drivers/net/wireless/realtek/rtlwifi/usb.c          | 2 +-
 drivers/net/wireless/realtek/rtlwifi/usb.h          | 2 +-
 drivers/net/wireless/realtek/rtlwifi/wifi.h         | 6 +++---
 5 files changed, 7 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/base.c b/drivers/net/wireless/realtek/rtlwifi/base.c
index 1a8d715b7c07..aab4605de9c4 100644
--- a/drivers/net/wireless/realtek/rtlwifi/base.c
+++ b/drivers/net/wireless/realtek/rtlwifi/base.c
@@ -2272,7 +2272,7 @@ static void rtl_c2h_content_parsing(struct ieee80211_hw *hw,
 				    struct sk_buff *skb)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
-	struct rtl_hal_ops *hal_ops = rtlpriv->cfg->ops;
+	const struct rtl_hal_ops *hal_ops = rtlpriv->cfg->ops;
 	const struct rtl_btc_ops *btc_ops = rtlpriv->btcoexist.btc_ops;
 	u8 cmd_id, cmd_len;
 	u8 *cmd_buf = NULL;
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c
index 48be7e346efc..c9b9e2bc90cc 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/sw.c
@@ -53,8 +53,6 @@ static int rtl92cu_init_sw_vars(struct ieee80211_hw *hw)
 	} else {
 		fw_name = "rtlwifi/rtl8192cufw_TMSC.bin";
 	}
-	/* provide name of alternative file */
-	rtlpriv->cfg->alt_fw_name = "rtlwifi/rtl8192cufw.bin";
 	pr_info("Loading firmware %s\n", fw_name);
 	rtlpriv->max_fw_size = 0x4000;
 	err = request_firmware_nowait(THIS_MODULE, 1,
@@ -160,6 +158,7 @@ static struct rtl_hal_usbint_cfg rtl92cu_interface_cfg = {
 
 static struct rtl_hal_cfg rtl92cu_hal_cfg = {
 	.name = "rtl92c_usb",
+	.alt_fw_name = "rtlwifi/rtl8192cufw.bin",
 	.ops = &rtl8192cu_hal_ops,
 	.mod_params = &rtl92cu_mod_params,
 	.usb_interface_cfg = &rtl92cu_interface_cfg,
diff --git a/drivers/net/wireless/realtek/rtlwifi/usb.c b/drivers/net/wireless/realtek/rtlwifi/usb.c
index 2ea72d9e3957..b6d300bec1e9 100644
--- a/drivers/net/wireless/realtek/rtlwifi/usb.c
+++ b/drivers/net/wireless/realtek/rtlwifi/usb.c
@@ -937,7 +937,7 @@ static const struct rtl_intf_ops rtl_usb_ops = {
 
 int rtl_usb_probe(struct usb_interface *intf,
 		  const struct usb_device_id *id,
-		  struct rtl_hal_cfg *rtl_hal_cfg)
+		  const struct rtl_hal_cfg *rtl_hal_cfg)
 {
 	int err;
 	struct ieee80211_hw *hw = NULL;
diff --git a/drivers/net/wireless/realtek/rtlwifi/usb.h b/drivers/net/wireless/realtek/rtlwifi/usb.h
index 12529afc0510..b66d6f9ae564 100644
--- a/drivers/net/wireless/realtek/rtlwifi/usb.h
+++ b/drivers/net/wireless/realtek/rtlwifi/usb.h
@@ -136,7 +136,7 @@ struct rtl_usb_priv {
 
 int rtl_usb_probe(struct usb_interface *intf,
 		  const struct usb_device_id *id,
-		  struct rtl_hal_cfg *rtl92cu_hal_cfg);
+		  const struct rtl_hal_cfg *rtl92cu_hal_cfg);
 void rtl_usb_disconnect(struct usb_interface *intf);
 int rtl_usb_suspend(struct usb_interface *pusb_intf, pm_message_t message);
 int rtl_usb_resume(struct usb_interface *pusb_intf);
diff --git a/drivers/net/wireless/realtek/rtlwifi/wifi.h b/drivers/net/wireless/realtek/rtlwifi/wifi.h
index 6a8212fd36af..ae6e351bc83c 100644
--- a/drivers/net/wireless/realtek/rtlwifi/wifi.h
+++ b/drivers/net/wireless/realtek/rtlwifi/wifi.h
@@ -2356,9 +2356,9 @@ struct rtl_hal_cfg {
 	bool write_readback;
 	char *name;
 	char *alt_fw_name;
-	struct rtl_hal_ops *ops;
+	const struct rtl_hal_ops *ops;
 	struct rtl_mod_params *mod_params;
-	struct rtl_hal_usbint_cfg *usb_interface_cfg;
+	const struct rtl_hal_usbint_cfg *usb_interface_cfg;
 	enum rtl_spec_ver spec_ver;
 
 	/*this map used for some registers or vars
@@ -2707,7 +2707,7 @@ struct rtl_priv {
 	/* hal_cfg : for diff cards
 	 * intf_ops : for diff interrface usb/pcie
 	 */
-	struct rtl_hal_cfg *cfg;
+	const struct rtl_hal_cfg *cfg;
 	const struct rtl_intf_ops *intf_ops;
 
 	/* this var will be set by set_bit,
-- 
cgit v1.2.3-58-ga151


From b5dc8873b6ffebd331e46814e537179eac8b6aa4 Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:48:32 +0300
Subject: wifi: rtlwifi: Add rtl8192du/sw.c

This contains the new module's entry point.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/1fdc83ac-4a6f-4645-8837-6e38683b6c84@gmail.com
---
 .../net/wireless/realtek/rtlwifi/rtl8192du/sw.c    | 395 +++++++++++++++++++++
 1 file changed, 395 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/sw.c

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/sw.c
new file mode 100644
index 000000000000..d069a81ac617
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/sw.c
@@ -0,0 +1,395 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2024  Realtek Corporation.*/
+
+#include "../wifi.h"
+#include "../core.h"
+#include "../usb.h"
+#include "../base.h"
+#include "../rtl8192d/reg.h"
+#include "../rtl8192d/def.h"
+#include "../rtl8192d/fw_common.h"
+#include "../rtl8192d/hw_common.h"
+#include "../rtl8192d/phy_common.h"
+#include "../rtl8192d/trx_common.h"
+#include "phy.h"
+#include "dm.h"
+#include "hw.h"
+#include "trx.h"
+#include "led.h"
+
+#include <linux/module.h>
+
+static struct usb_interface *rtl92du_get_other_intf(struct ieee80211_hw *hw)
+{
+	struct usb_interface *intf;
+	struct usb_device *udev;
+	u8 other_interfaceindex;
+
+	/* See SET_IEEE80211_DEV(hw, &intf->dev); in usb.c */
+	intf = container_of_const(wiphy_dev(hw->wiphy), struct usb_interface, dev);
+
+	if (intf->altsetting[0].desc.bInterfaceNumber == 0)
+		other_interfaceindex = 1;
+	else
+		other_interfaceindex = 0;
+
+	udev = interface_to_usbdev(intf);
+
+	return usb_ifnum_to_if(udev, other_interfaceindex);
+}
+
+static int rtl92du_init_shared_data(struct ieee80211_hw *hw)
+{
+	struct usb_interface *other_intf = rtl92du_get_other_intf(hw);
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	struct rtl_priv *other_rtlpriv = NULL;
+	struct ieee80211_hw *other_hw = NULL;
+
+	if (other_intf)
+		other_hw = usb_get_intfdata(other_intf);
+
+	if (other_hw) {
+		/* The other interface was already probed. */
+		other_rtlpriv = rtl_priv(other_hw);
+		rtlpriv->curveindex_2g = other_rtlpriv->curveindex_2g;
+		rtlpriv->curveindex_5g = other_rtlpriv->curveindex_5g;
+		rtlpriv->mutex_for_power_on_off = other_rtlpriv->mutex_for_power_on_off;
+		rtlpriv->mutex_for_hw_init = other_rtlpriv->mutex_for_hw_init;
+
+		if (!rtlpriv->curveindex_2g || !rtlpriv->curveindex_5g ||
+		    !rtlpriv->mutex_for_power_on_off || !rtlpriv->mutex_for_hw_init)
+			return -ENOMEM;
+
+		return 0;
+	}
+
+	/* The other interface doesn't exist or was not probed yet. */
+	rtlpriv->curveindex_2g = kcalloc(TARGET_CHNL_NUM_2G,
+					 sizeof(*rtlpriv->curveindex_2g),
+					 GFP_KERNEL);
+	rtlpriv->curveindex_5g = kcalloc(TARGET_CHNL_NUM_5G,
+					 sizeof(*rtlpriv->curveindex_5g),
+					 GFP_KERNEL);
+	rtlpriv->mutex_for_power_on_off =
+		kzalloc(sizeof(*rtlpriv->mutex_for_power_on_off), GFP_KERNEL);
+	rtlpriv->mutex_for_hw_init =
+		kzalloc(sizeof(*rtlpriv->mutex_for_hw_init), GFP_KERNEL);
+
+	if (!rtlpriv->curveindex_2g || !rtlpriv->curveindex_5g ||
+	    !rtlpriv->mutex_for_power_on_off || !rtlpriv->mutex_for_hw_init) {
+		kfree(rtlpriv->curveindex_2g);
+		kfree(rtlpriv->curveindex_5g);
+		kfree(rtlpriv->mutex_for_power_on_off);
+		kfree(rtlpriv->mutex_for_hw_init);
+		rtlpriv->curveindex_2g = NULL;
+		rtlpriv->curveindex_5g = NULL;
+		rtlpriv->mutex_for_power_on_off = NULL;
+		rtlpriv->mutex_for_hw_init = NULL;
+		return -ENOMEM;
+	}
+
+	mutex_init(rtlpriv->mutex_for_power_on_off);
+	mutex_init(rtlpriv->mutex_for_hw_init);
+
+	return 0;
+}
+
+static void rtl92du_deinit_shared_data(struct ieee80211_hw *hw)
+{
+	struct usb_interface *other_intf = rtl92du_get_other_intf(hw);
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+
+	if (!other_intf || !usb_get_intfdata(other_intf)) {
+		/* The other interface doesn't exist or was already disconnected. */
+		kfree(rtlpriv->curveindex_2g);
+		kfree(rtlpriv->curveindex_5g);
+		if (rtlpriv->mutex_for_power_on_off)
+			mutex_destroy(rtlpriv->mutex_for_power_on_off);
+		if (rtlpriv->mutex_for_hw_init)
+			mutex_destroy(rtlpriv->mutex_for_hw_init);
+		kfree(rtlpriv->mutex_for_power_on_off);
+		kfree(rtlpriv->mutex_for_hw_init);
+	}
+}
+
+static int rtl92du_init_sw_vars(struct ieee80211_hw *hw)
+{
+	const char *fw_name = "rtlwifi/rtl8192dufw.bin";
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+	int err;
+
+	err = rtl92du_init_shared_data(hw);
+	if (err)
+		return err;
+
+	rtlpriv->dm.dm_initialgain_enable = true;
+	rtlpriv->dm.dm_flag = 0;
+	rtlpriv->dm.disable_framebursting = false;
+	rtlpriv->dm.thermalvalue = 0;
+	rtlpriv->dm.useramask = true;
+
+	/* dual mac */
+	if (rtlpriv->rtlhal.current_bandtype == BAND_ON_5G)
+		rtlpriv->phy.current_channel = 36;
+	else
+		rtlpriv->phy.current_channel = 1;
+
+	if (rtlpriv->rtlhal.macphymode != SINGLEMAC_SINGLEPHY)
+		rtlpriv->rtlhal.disable_amsdu_8k = true;
+
+	/* for LPS & IPS */
+	rtlpriv->psc.inactiveps = rtlpriv->cfg->mod_params->inactiveps;
+	rtlpriv->psc.swctrl_lps = rtlpriv->cfg->mod_params->swctrl_lps;
+	rtlpriv->psc.fwctrl_lps = rtlpriv->cfg->mod_params->fwctrl_lps;
+
+	/* for early mode */
+	rtlpriv->rtlhal.earlymode_enable = false;
+
+	/* for firmware buf */
+	rtlpriv->rtlhal.pfirmware = kmalloc(0x8000, GFP_KERNEL);
+	if (!rtlpriv->rtlhal.pfirmware)
+		return -ENOMEM;
+
+	rtlpriv->max_fw_size = 0x8000;
+	pr_info("Driver for Realtek RTL8192DU WLAN interface\n");
+	pr_info("Loading firmware file %s\n", fw_name);
+
+	/* request fw */
+	err = request_firmware_nowait(THIS_MODULE, 1, fw_name,
+				      rtlpriv->io.dev, GFP_KERNEL, hw,
+				      rtl_fw_cb);
+	if (err) {
+		pr_err("Failed to request firmware!\n");
+		kfree(rtlpriv->rtlhal.pfirmware);
+		rtlpriv->rtlhal.pfirmware = NULL;
+		return err;
+	}
+
+	return 0;
+}
+
+static void rtl92du_deinit_sw_vars(struct ieee80211_hw *hw)
+{
+	struct rtl_priv *rtlpriv = rtl_priv(hw);
+
+	kfree(rtlpriv->rtlhal.pfirmware);
+	rtlpriv->rtlhal.pfirmware = NULL;
+
+	rtl92du_deinit_shared_data(hw);
+}
+
+static const struct rtl_hal_ops rtl8192du_hal_ops = {
+	.init_sw_vars = rtl92du_init_sw_vars,
+	.deinit_sw_vars = rtl92du_deinit_sw_vars,
+	.read_chip_version = rtl92du_read_chip_version,
+	.read_eeprom_info = rtl92d_read_eeprom_info,
+	.hw_init = rtl92du_hw_init,
+	.hw_disable = rtl92du_card_disable,
+	.enable_interrupt = rtl92du_enable_interrupt,
+	.disable_interrupt = rtl92du_disable_interrupt,
+	.set_network_type = rtl92du_set_network_type,
+	.set_chk_bssid = rtl92du_set_check_bssid,
+	.set_qos = rtl92d_set_qos,
+	.set_bcn_reg = rtl92du_set_beacon_related_registers,
+	.set_bcn_intv = rtl92du_set_beacon_interval,
+	.update_interrupt_mask = rtl92du_update_interrupt_mask,
+	.get_hw_reg = rtl92du_get_hw_reg,
+	.set_hw_reg = rtl92du_set_hw_reg,
+	.update_rate_tbl = rtl92d_update_hal_rate_tbl,
+	.fill_tx_desc = rtl92du_tx_fill_desc,
+	.query_rx_desc = rtl92d_rx_query_desc,
+	.set_channel_access = rtl92d_update_channel_access_setting,
+	.radio_onoff_checking = rtl92d_gpio_radio_on_off_checking,
+	.set_bw_mode = rtl92du_phy_set_bw_mode,
+	.switch_channel = rtl92du_phy_sw_chnl,
+	.dm_watchdog = rtl92du_dm_watchdog,
+	.scan_operation_backup = rtl_phy_scan_operation_backup,
+	.set_rf_power_state = rtl92du_phy_set_rf_power_state,
+	.led_control = rtl92du_led_control,
+	.set_desc = rtl92d_set_desc,
+	.get_desc = rtl92d_get_desc,
+	.enable_hw_sec = rtl92d_enable_hw_security_config,
+	.set_key = rtl92d_set_key,
+	.get_bbreg = rtl92du_phy_query_bb_reg,
+	.set_bbreg = rtl92du_phy_set_bb_reg,
+	.get_rfreg = rtl92d_phy_query_rf_reg,
+	.set_rfreg = rtl92d_phy_set_rf_reg,
+	.linked_set_reg = rtl92du_linked_set_reg,
+	.fill_h2c_cmd = rtl92d_fill_h2c_cmd,
+	.get_btc_status = rtl_btc_status_false,
+	.phy_iq_calibrate = rtl92du_phy_iq_calibrate,
+	.phy_lc_calibrate = rtl92du_phy_lc_calibrate,
+};
+
+static struct rtl_mod_params rtl92du_mod_params = {
+	.sw_crypto = false,
+	.inactiveps = false,
+	.swctrl_lps = false,
+	.debug_level = 0,
+	.debug_mask = 0,
+};
+
+static const struct rtl_hal_usbint_cfg rtl92du_interface_cfg = {
+	/* rx */
+	.rx_urb_num = 8,
+	.rx_max_size = 15360,
+	.usb_rx_hdl = NULL,
+	.usb_rx_segregate_hdl = NULL,
+	/* tx */
+	.usb_tx_cleanup = rtl92du_tx_cleanup,
+	.usb_tx_post_hdl = rtl92du_tx_post_hdl,
+	.usb_tx_aggregate_hdl = rtl92du_tx_aggregate_hdl,
+	.usb_endpoint_mapping = rtl92du_endpoint_mapping,
+	.usb_mq_to_hwq = rtl92du_mq_to_hwq,
+};
+
+static const struct rtl_hal_cfg rtl92du_hal_cfg = {
+	.name = "rtl8192du",
+	.ops = &rtl8192du_hal_ops,
+	.mod_params = &rtl92du_mod_params,
+	.usb_interface_cfg = &rtl92du_interface_cfg,
+
+	.maps[SYS_ISO_CTRL] = REG_SYS_ISO_CTRL,
+	.maps[SYS_FUNC_EN] = REG_SYS_FUNC_EN,
+	.maps[SYS_CLK] = REG_SYS_CLKR,
+	.maps[MAC_RCR_AM] = RCR_AM,
+	.maps[MAC_RCR_AB] = RCR_AB,
+	.maps[MAC_RCR_ACRC32] = RCR_ACRC32,
+	.maps[MAC_RCR_ACF] = RCR_ACF,
+	.maps[MAC_RCR_AAP] = RCR_AAP,
+
+	.maps[EFUSE_TEST] = REG_EFUSE_TEST,
+	.maps[EFUSE_ACCESS] = REG_EFUSE_ACCESS,
+	.maps[EFUSE_CTRL] = REG_EFUSE_CTRL,
+	.maps[EFUSE_CLK] = 0,	/* just for 92se */
+	.maps[EFUSE_CLK_CTRL] = REG_EFUSE_CTRL,
+	.maps[EFUSE_PWC_EV12V] = PWC_EV12V,
+	.maps[EFUSE_FEN_ELDR] = FEN_ELDR,
+	.maps[EFUSE_LOADER_CLK_EN] = 0,
+	.maps[EFUSE_ANA8M] = 0,	/* just for 92se */
+	.maps[EFUSE_HWSET_MAX_SIZE] = HWSET_MAX_SIZE,
+	.maps[EFUSE_MAX_SECTION_MAP] = EFUSE_MAX_SECTION,
+	.maps[EFUSE_REAL_CONTENT_SIZE] = EFUSE_REAL_CONTENT_LEN,
+
+	.maps[RWCAM] = REG_CAMCMD,
+	.maps[WCAMI] = REG_CAMWRITE,
+	.maps[RCAMO] = REG_CAMREAD,
+	.maps[CAMDBG] = REG_CAMDBG,
+	.maps[SECR] = REG_SECCFG,
+	.maps[SEC_CAM_NONE] = CAM_NONE,
+	.maps[SEC_CAM_WEP40] = CAM_WEP40,
+	.maps[SEC_CAM_TKIP] = CAM_TKIP,
+	.maps[SEC_CAM_AES] = CAM_AES,
+	.maps[SEC_CAM_WEP104] = CAM_WEP104,
+
+	.maps[RTL_IMR_BCNDMAINT6] = IMR_BCNDMAINT6,
+	.maps[RTL_IMR_BCNDMAINT5] = IMR_BCNDMAINT5,
+	.maps[RTL_IMR_BCNDMAINT4] = IMR_BCNDMAINT4,
+	.maps[RTL_IMR_BCNDMAINT3] = IMR_BCNDMAINT3,
+	.maps[RTL_IMR_BCNDMAINT2] = IMR_BCNDMAINT2,
+	.maps[RTL_IMR_BCNDMAINT1] = IMR_BCNDMAINT1,
+	.maps[RTL_IMR_BCNDOK8] = IMR_BCNDOK8,
+	.maps[RTL_IMR_BCNDOK7] = IMR_BCNDOK7,
+	.maps[RTL_IMR_BCNDOK6] = IMR_BCNDOK6,
+	.maps[RTL_IMR_BCNDOK5] = IMR_BCNDOK5,
+	.maps[RTL_IMR_BCNDOK4] = IMR_BCNDOK4,
+	.maps[RTL_IMR_BCNDOK3] = IMR_BCNDOK3,
+	.maps[RTL_IMR_BCNDOK2] = IMR_BCNDOK2,
+	.maps[RTL_IMR_BCNDOK1] = IMR_BCNDOK1,
+	.maps[RTL_IMR_TIMEOUT2] = IMR_TIMEOUT2,
+	.maps[RTL_IMR_TIMEOUT1] = IMR_TIMEOUT1,
+
+	.maps[RTL_IMR_TXFOVW] = IMR_TXFOVW,
+	.maps[RTL_IMR_PSTIMEOUT] = IMR_PSTIMEOUT,
+	.maps[RTL_IMR_BCNINT] = IMR_BCNINT,
+	.maps[RTL_IMR_RXFOVW] = IMR_RXFOVW,
+	.maps[RTL_IMR_RDU] = IMR_RDU,
+	.maps[RTL_IMR_ATIMEND] = IMR_ATIMEND,
+	.maps[RTL_IMR_BDOK] = IMR_BDOK,
+	.maps[RTL_IMR_MGNTDOK] = IMR_MGNTDOK,
+	.maps[RTL_IMR_TBDER] = IMR_TBDER,
+	.maps[RTL_IMR_HIGHDOK] = IMR_HIGHDOK,
+	.maps[RTL_IMR_TBDOK] = IMR_TBDOK,
+	.maps[RTL_IMR_BKDOK] = IMR_BKDOK,
+	.maps[RTL_IMR_BEDOK] = IMR_BEDOK,
+	.maps[RTL_IMR_VIDOK] = IMR_VIDOK,
+	.maps[RTL_IMR_VODOK] = IMR_VODOK,
+	.maps[RTL_IMR_ROK] = IMR_ROK,
+	.maps[RTL_IBSS_INT_MASKS] = (IMR_BCNINT | IMR_TBDOK | IMR_TBDER),
+
+	.maps[RTL_RC_CCK_RATE1M] = DESC_RATE1M,
+	.maps[RTL_RC_CCK_RATE2M] = DESC_RATE2M,
+	.maps[RTL_RC_CCK_RATE5_5M] = DESC_RATE5_5M,
+	.maps[RTL_RC_CCK_RATE11M] = DESC_RATE11M,
+	.maps[RTL_RC_OFDM_RATE6M] = DESC_RATE6M,
+	.maps[RTL_RC_OFDM_RATE9M] = DESC_RATE9M,
+	.maps[RTL_RC_OFDM_RATE12M] = DESC_RATE12M,
+	.maps[RTL_RC_OFDM_RATE18M] = DESC_RATE18M,
+	.maps[RTL_RC_OFDM_RATE24M] = DESC_RATE24M,
+	.maps[RTL_RC_OFDM_RATE36M] = DESC_RATE36M,
+	.maps[RTL_RC_OFDM_RATE48M] = DESC_RATE48M,
+	.maps[RTL_RC_OFDM_RATE54M] = DESC_RATE54M,
+
+	.maps[RTL_RC_HT_RATEMCS7] = DESC_RATEMCS7,
+	.maps[RTL_RC_HT_RATEMCS15] = DESC_RATEMCS15,
+};
+
+module_param_named(swenc, rtl92du_mod_params.sw_crypto, bool, 0444);
+module_param_named(debug_level, rtl92du_mod_params.debug_level, int, 0644);
+module_param_named(ips, rtl92du_mod_params.inactiveps, bool, 0444);
+module_param_named(swlps, rtl92du_mod_params.swctrl_lps, bool, 0444);
+module_param_named(debug_mask, rtl92du_mod_params.debug_mask, ullong, 0644);
+MODULE_PARM_DESC(swenc, "Set to 1 for software crypto (default 0)\n");
+MODULE_PARM_DESC(ips, "Set to 0 to not use link power save (default 0)\n");
+MODULE_PARM_DESC(swlps, "Set to 1 to use SW control power save (default 0)\n");
+MODULE_PARM_DESC(debug_level, "Set debug level (0-5) (default 0)");
+MODULE_PARM_DESC(debug_mask, "Set debug mask (default 0)");
+
+#define USB_VENDOR_ID_REALTEK		0x0bda
+
+static const struct usb_device_id rtl8192d_usb_ids[] = {
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x8193, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x8194, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x8111, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x0193, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0x8171, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(USB_VENDOR_ID_REALTEK, 0xe194, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x2019, 0xab2c, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x2019, 0xab2d, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x2019, 0x4903, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x2019, 0x4904, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x07b8, 0x8193, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x20f4, 0x664b, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x04dd, 0x954f, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x04dd, 0x96a6, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x050d, 0x110a, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x050d, 0x1105, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x050d, 0x120a, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x1668, 0x8102, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x0930, 0x0a0a, rtl92du_hal_cfg)},
+	{RTL_USB_DEVICE(0x2001, 0x330c, rtl92du_hal_cfg)},
+	{}
+};
+
+MODULE_DEVICE_TABLE(usb, rtl8192d_usb_ids);
+
+static int rtl8192du_probe(struct usb_interface *intf,
+			   const struct usb_device_id *id)
+{
+	return rtl_usb_probe(intf, id, &rtl92du_hal_cfg);
+}
+
+static struct usb_driver rtl8192du_driver = {
+	.name = "rtl8192du",
+	.probe = rtl8192du_probe,
+	.disconnect = rtl_usb_disconnect,
+	.id_table = rtl8192d_usb_ids,
+	.disable_hub_initiated_lpm = 1,
+};
+
+module_usb_driver(rtl8192du_driver);
+
+MODULE_AUTHOR("Bitterblue Smith	<rtl8821cerfe2@gmail.com>");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Realtek 8192DU 802.11n Dual Mac USB wireless");
+MODULE_FIRMWARE("rtlwifi/rtl8192dufw.bin");
-- 
cgit v1.2.3-58-ga151


From 59ea089dcba3b51769280522fd62696d4d436cbc Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Thu, 23 May 2024 17:49:00 +0300
Subject: wifi: rtlwifi: Enable the new rtl8192du driver

The RTL8192DU is an older Wifi 4 dual band chip. It comes in two
flavours: single MAC single PHY (like most Realtek Wifi 4 USB devices),
and dual MAC dual PHY.

The single MAC single PHY version is 2T2R and can work either in the
2.4 GHz band or the 5 GHz band.

The dual MAC dual PHY version has two USB interfaces and appears to the
system as two separate 1T1R Wifi devices, one working in the 2.4 GHz
band, the other in the 5 GHz band.

This was tested only with a single MAC single PHY device, mostly in
station mode. The speeds in the 2.4 GHz band with 20 MHz channel width
are similar to the out-of-tree driver: 85/51 megabits/second.

Stefan Lippers-Hollmann tested the speed in the 5 GHz band with 40 MHz
channel width: 173/99 megabits/second.

It was also tested briefly in AP mode. It's emitting beacons and my
phone can connect to it.

Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Tested-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/5f2da7ee-876a-42fc-8fec-ec5386fa8c26@gmail.com
---
 drivers/net/wireless/realtek/rtlwifi/Kconfig            | 12 ++++++++++++
 drivers/net/wireless/realtek/rtlwifi/Makefile           |  1 +
 drivers/net/wireless/realtek/rtlwifi/rtl8192du/Makefile | 13 +++++++++++++
 3 files changed, 26 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtlwifi/rtl8192du/Makefile

diff --git a/drivers/net/wireless/realtek/rtlwifi/Kconfig b/drivers/net/wireless/realtek/rtlwifi/Kconfig
index cfe63f7b28d9..1e66c1bf7c8b 100644
--- a/drivers/net/wireless/realtek/rtlwifi/Kconfig
+++ b/drivers/net/wireless/realtek/rtlwifi/Kconfig
@@ -119,6 +119,18 @@ config RTL8192CU
 
 	If you choose to build it as a module, it will be called rtl8192cu
 
+config RTL8192DU
+	tristate "Realtek RTL8192DU USB Wireless Network Adapter"
+	depends on USB
+	select RTLWIFI
+	select RTLWIFI_USB
+	select RTL8192D_COMMON
+	help
+	This is the driver for Realtek RTL8192DU 802.11n USB
+	wireless network adapters.
+
+	If you choose to build it as a module, it will be called rtl8192du
+
 config RTLWIFI
 	tristate
 	select FW_LOADER
diff --git a/drivers/net/wireless/realtek/rtlwifi/Makefile b/drivers/net/wireless/realtek/rtlwifi/Makefile
index 423981b148df..9cf32277c7f1 100644
--- a/drivers/net/wireless/realtek/rtlwifi/Makefile
+++ b/drivers/net/wireless/realtek/rtlwifi/Makefile
@@ -25,6 +25,7 @@ obj-$(CONFIG_RTL8192CU)		+= rtl8192cu/
 obj-$(CONFIG_RTL8192SE)		+= rtl8192se/
 obj-$(CONFIG_RTL8192D_COMMON)	+= rtl8192d/
 obj-$(CONFIG_RTL8192DE)		+= rtl8192de/
+obj-$(CONFIG_RTL8192DU)		+= rtl8192du/
 obj-$(CONFIG_RTL8723AE)		+= rtl8723ae/
 obj-$(CONFIG_RTL8723BE)		+= rtl8723be/
 obj-$(CONFIG_RTL8188EE)		+= rtl8188ee/
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192du/Makefile b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/Makefile
new file mode 100644
index 000000000000..569bfd3d5030
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192du/Makefile
@@ -0,0 +1,13 @@
+# SPDX-License-Identifier: GPL-2.0
+rtl8192du-objs :=		\
+		dm.o		\
+		fw.o		\
+		hw.o		\
+		led.o		\
+		phy.o		\
+		rf.o		\
+		sw.o		\
+		table.o		\
+		trx.o
+
+obj-$(CONFIG_RTL8192DU) += rtl8192du.o
-- 
cgit v1.2.3-58-ga151


From 9c4fde42cce05719120cf892a44b76ff61d908c7 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 24 May 2024 08:32:48 +0800
Subject: wifi: rtlwifi: handle return value of usb init TX/RX

Handle error code to cause failed to USB probe result from unexpected
USB EP number, otherwise when USB disconnect skb_dequeue() an uninitialized
skb list and cause warnings below.

usb 2-1: USB disconnect, device number 76
INFO: trying to register non-static key.
The code is fine but needs lockdep annotation, or maybe
you didn't initialize this object before use?
turning off the locking correctness validator.
CPU: 0 PID: 54060 Comm: kworker/0:1 Not tainted 6.9.0-rc7 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.2-0-gea1b7a073390-prebuilt.qemu.org 04/01/2014
Workqueue: usb_hub_wq hub_event
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0x116/0x1b0 lib/dump_stack.c:114
 assign_lock_key kernel/locking/lockdep.c:976 [inline]
 register_lock_class+0xc18/0xfa0 kernel/locking/lockdep.c:1289
 __lock_acquire+0x108/0x3bc0 kernel/locking/lockdep.c:5014
 lock_acquire kernel/locking/lockdep.c:5754 [inline]
 lock_acquire+0x1b0/0x550 kernel/locking/lockdep.c:5719
 __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
 _raw_spin_lock_irqsave+0x3d/0x60 kernel/locking/spinlock.c:162
 skb_dequeue+0x20/0x180 net/core/skbuff.c:3846
 rtl_usb_cleanup drivers/net/wireless/realtek/rtlwifi/usb.c:706 [inline]
 rtl_usb_deinit drivers/net/wireless/realtek/rtlwifi/usb.c:721 [inline]
 rtl_usb_disconnect+0x4a4/0x850 drivers/net/wireless/realtek/rtlwifi/usb.c:1051
 usb_unbind_interface+0x1e8/0x980 drivers/usb/core/driver.c:461
 device_remove drivers/base/dd.c:568 [inline]
 device_remove+0x122/0x170 drivers/base/dd.c:560
 __device_release_driver drivers/base/dd.c:1270 [inline]
 device_release_driver_internal+0x443/0x620 drivers/base/dd.c:1293
 bus_remove_device+0x22f/0x420 drivers/base/bus.c:574
 device_del+0x395/0x9f0 drivers/base/core.c:3909
 usb_disable_device+0x360/0x7b0 drivers/usb/core/message.c:1418
 usb_disconnect+0x2db/0x930 drivers/usb/core/hub.c:2305
 hub_port_connect drivers/usb/core/hub.c:5362 [inline]
 hub_port_connect_change drivers/usb/core/hub.c:5662 [inline]
 port_event drivers/usb/core/hub.c:5822 [inline]
 hub_event+0x1e39/0x4ce0 drivers/usb/core/hub.c:5904
 process_one_work+0x97b/0x1a90 kernel/workqueue.c:3267
 process_scheduled_works kernel/workqueue.c:3348 [inline]
 worker_thread+0x680/0xf00 kernel/workqueue.c:3429
 kthread+0x2c7/0x3b0 kernel/kthread.c:388
 ret_from_fork+0x45/0x80 arch/x86/kernel/process.c:147
 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244
 </TASK>

Reported-by: Shichao Lai <shichaorai@gmail.com>
Closes: https://lore.kernel.org/linux-wireless/CAEk6kZuuezkH1dVRJf3EAVZK-83=OpTz62qCugkpTkswj8JF6w@mail.gmail.com/T/#u
Tested-by: Shichao Lai <shichaorai@gmail.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240524003248.5952-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtlwifi/usb.c | 34 +++++++++++++++++++++++-------
 1 file changed, 26 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/usb.c b/drivers/net/wireless/realtek/rtlwifi/usb.c
index b6d300bec1e9..d37a017b2b81 100644
--- a/drivers/net/wireless/realtek/rtlwifi/usb.c
+++ b/drivers/net/wireless/realtek/rtlwifi/usb.c
@@ -23,6 +23,8 @@ MODULE_DESCRIPTION("USB basic driver for rtlwifi");
 
 #define MAX_USBCTRL_VENDORREQ_TIMES		10
 
+static void _rtl_usb_cleanup_tx(struct ieee80211_hw *hw);
+
 static void _usbctrl_vendorreq_sync(struct usb_device *udev, u8 reqtype,
 				   u16 value, void *pdata, u16 len)
 {
@@ -285,9 +287,23 @@ static int _rtl_usb_init(struct ieee80211_hw *hw)
 	}
 	/* usb endpoint mapping */
 	err = rtlpriv->cfg->usb_interface_cfg->usb_endpoint_mapping(hw);
-	rtlusb->usb_mq_to_hwq =  rtlpriv->cfg->usb_interface_cfg->usb_mq_to_hwq;
-	_rtl_usb_init_tx(hw);
-	_rtl_usb_init_rx(hw);
+	if (err)
+		return err;
+
+	rtlusb->usb_mq_to_hwq = rtlpriv->cfg->usb_interface_cfg->usb_mq_to_hwq;
+
+	err = _rtl_usb_init_tx(hw);
+	if (err)
+		return err;
+
+	err = _rtl_usb_init_rx(hw);
+	if (err)
+		goto err_out;
+
+	return 0;
+
+err_out:
+	_rtl_usb_cleanup_tx(hw);
 	return err;
 }
 
@@ -691,17 +707,13 @@ static int rtl_usb_start(struct ieee80211_hw *hw)
 }
 
 /*=======================  tx =========================================*/
-static void rtl_usb_cleanup(struct ieee80211_hw *hw)
+static void _rtl_usb_cleanup_tx(struct ieee80211_hw *hw)
 {
 	u32 i;
 	struct sk_buff *_skb;
 	struct rtl_usb *rtlusb = rtl_usbdev(rtl_usbpriv(hw));
 	struct ieee80211_tx_info *txinfo;
 
-	/* clean up rx stuff. */
-	_rtl_usb_cleanup_rx(hw);
-
-	/* clean up tx stuff */
 	for (i = 0; i < RTL_USB_MAX_EP_NUM; i++) {
 		while ((_skb = skb_dequeue(&rtlusb->tx_skb_queue[i]))) {
 			rtlusb->usb_tx_cleanup(hw, _skb);
@@ -715,6 +727,12 @@ static void rtl_usb_cleanup(struct ieee80211_hw *hw)
 	usb_kill_anchored_urbs(&rtlusb->tx_submitted);
 }
 
+static void rtl_usb_cleanup(struct ieee80211_hw *hw)
+{
+	_rtl_usb_cleanup_rx(hw);
+	_rtl_usb_cleanup_tx(hw);
+}
+
 /* We may add some struct into struct rtl_usb later. Do deinit here.  */
 static void rtl_usb_deinit(struct ieee80211_hw *hw)
 {
-- 
cgit v1.2.3-58-ga151


From 487f7323f39a9d45a31c59e909de4a626f02d34f Mon Sep 17 00:00:00 2001
From: MD Danish Anwar <danishanwar@ti.com>
Date: Tue, 28 May 2024 17:07:32 +0530
Subject: net: ti: icssg-prueth: Add helper functions to configure FDB

Introduce helper functions to configure firmware FDB tables, VLAN tables
and Port VLAN ID settings to aid adding Switch mode support.

Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/ti/icssg/icssg_config.c | 170 +++++++++++++++++++++++++++
 drivers/net/ethernet/ti/icssg/icssg_config.h |  19 +++
 drivers/net/ethernet/ti/icssg/icssg_prueth.h |  12 ++
 3 files changed, 201 insertions(+)

diff --git a/drivers/net/ethernet/ti/icssg/icssg_config.c b/drivers/net/ethernet/ti/icssg/icssg_config.c
index 15f2235bf90f..2213374d4d45 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_config.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_config.c
@@ -477,3 +477,173 @@ void icssg_config_set_speed(struct prueth_emac *emac)
 
 	writeb(fw_speed, emac->dram.va + PORT_LINK_SPEED_OFFSET);
 }
+
+int icssg_send_fdb_msg(struct prueth_emac *emac, struct mgmt_cmd *cmd,
+		       struct mgmt_cmd_rsp *rsp)
+{
+	struct prueth *prueth = emac->prueth;
+	int slice = prueth_emac_slice(emac);
+	int addr, ret;
+
+	addr = icssg_queue_pop(prueth, slice == 0 ?
+			       ICSSG_CMD_POP_SLICE0 : ICSSG_CMD_POP_SLICE1);
+	if (addr < 0)
+		return addr;
+
+	/* First 4 bytes have FW owned buffer linking info which should
+	 * not be touched
+	 */
+	memcpy_toio(prueth->shram.va + addr + 4, cmd, sizeof(*cmd));
+	icssg_queue_push(prueth, slice == 0 ?
+			 ICSSG_CMD_PUSH_SLICE0 : ICSSG_CMD_PUSH_SLICE1, addr);
+	ret = read_poll_timeout(icssg_queue_pop, addr, addr >= 0,
+				2000, 20000000, false, prueth, slice == 0 ?
+				ICSSG_RSP_POP_SLICE0 : ICSSG_RSP_POP_SLICE1);
+	if (ret) {
+		netdev_err(emac->ndev, "Timedout sending HWQ message\n");
+		return ret;
+	}
+
+	memcpy_fromio(rsp, prueth->shram.va + addr, sizeof(*rsp));
+	/* Return buffer back for to pool */
+	icssg_queue_push(prueth, slice == 0 ?
+			 ICSSG_RSP_PUSH_SLICE0 : ICSSG_RSP_PUSH_SLICE1, addr);
+
+	return 0;
+}
+
+static void icssg_fdb_setup(struct prueth_emac *emac, struct mgmt_cmd *fdb_cmd,
+			    const unsigned char *addr, u8 fid, int cmd)
+{
+	int slice = prueth_emac_slice(emac);
+	u8 mac_fid[ETH_ALEN + 2];
+	u16 fdb_slot;
+
+	ether_addr_copy(mac_fid, addr);
+
+	/* 1-1 VID-FID mapping is already setup */
+	mac_fid[ETH_ALEN] = fid;
+	mac_fid[ETH_ALEN + 1] = 0;
+
+	fdb_slot = bitrev32(crc32_le(0, mac_fid, 8)) & PRUETH_SWITCH_FDB_MASK;
+
+	fdb_cmd->header = ICSSG_FW_MGMT_CMD_HEADER;
+	fdb_cmd->type   = ICSSG_FW_MGMT_FDB_CMD_TYPE;
+	fdb_cmd->seqnum = ++(emac->prueth->icssg_hwcmdseq);
+	fdb_cmd->param  = cmd;
+	fdb_cmd->param |= (slice << 4);
+
+	memcpy(&fdb_cmd->cmd_args[0], addr, 4);
+	memcpy(&fdb_cmd->cmd_args[1], &addr[4], 2);
+	fdb_cmd->cmd_args[2] = fdb_slot;
+
+	netdev_dbg(emac->ndev, "MAC %pM slot %X FID %X\n", addr, fdb_slot, fid);
+}
+
+int icssg_fdb_add_del(struct prueth_emac *emac, const unsigned char *addr,
+		      u8 vid, u8 fid_c2, bool add)
+{
+	struct mgmt_cmd_rsp fdb_cmd_rsp = { 0 };
+	struct mgmt_cmd fdb_cmd = { 0 };
+	u8 fid = vid;
+	int ret;
+
+	icssg_fdb_setup(emac, &fdb_cmd, addr, fid, add ? ICSS_CMD_ADD_FDB : ICSS_CMD_DEL_FDB);
+
+	fid_c2 |= ICSSG_FDB_ENTRY_VALID;
+	fdb_cmd.cmd_args[1] |= ((fid << 16) | (fid_c2 << 24));
+
+	ret = icssg_send_fdb_msg(emac, &fdb_cmd, &fdb_cmd_rsp);
+	if (ret)
+		return ret;
+
+	WARN_ON(fdb_cmd.seqnum != fdb_cmd_rsp.seqnum);
+	if (fdb_cmd_rsp.status == 1)
+		return 0;
+
+	return -EINVAL;
+}
+
+int icssg_fdb_lookup(struct prueth_emac *emac, const unsigned char *addr,
+		     u8 vid)
+{
+	struct mgmt_cmd_rsp fdb_cmd_rsp = { 0 };
+	struct mgmt_cmd fdb_cmd = { 0 };
+	struct prueth_fdb_slot *slot;
+	u8 fid = vid;
+	int ret, i;
+
+	icssg_fdb_setup(emac, &fdb_cmd, addr, fid, ICSS_CMD_GET_FDB_SLOT);
+
+	fdb_cmd.cmd_args[1] |= fid << 16;
+
+	ret = icssg_send_fdb_msg(emac, &fdb_cmd, &fdb_cmd_rsp);
+	if (ret)
+		return ret;
+
+	WARN_ON(fdb_cmd.seqnum != fdb_cmd_rsp.seqnum);
+
+	slot = (struct prueth_fdb_slot __force *)(emac->dram.va + FDB_CMD_BUFFER);
+	for (i = 0; i < 4; i++) {
+		if (ether_addr_equal(addr, slot->mac) && vid == slot->fid)
+			return (slot->fid_c2 & ~ICSSG_FDB_ENTRY_VALID);
+		slot++;
+	}
+
+	return 0;
+}
+
+void icssg_vtbl_modify(struct prueth_emac *emac, u8 vid, u8 port_mask,
+		       u8 untag_mask, bool add)
+{
+	struct prueth *prueth = emac->prueth;
+	struct prueth_vlan_tbl *tbl;
+	u8 fid_c1;
+
+	tbl = prueth->vlan_tbl;
+	fid_c1 = tbl[vid].fid_c1;
+
+	/* FID_C1: bit0..2 port membership mask,
+	 * bit3..5 tagging mask for each port
+	 * bit6 Stream VID (not handled currently)
+	 * bit7 MC flood (not handled currently)
+	 */
+	if (add) {
+		fid_c1 |= (port_mask | port_mask << 3);
+		fid_c1 &= ~(untag_mask << 3);
+	} else {
+		fid_c1 &= ~(port_mask | port_mask << 3);
+	}
+
+	tbl[vid].fid_c1 = fid_c1;
+}
+
+u16 icssg_get_pvid(struct prueth_emac *emac)
+{
+	struct prueth *prueth = emac->prueth;
+	u32 pvid;
+
+	if (emac->port_id == PRUETH_PORT_MII0)
+		pvid = readl(prueth->shram.va + EMAC_ICSSG_SWITCH_PORT1_DEFAULT_VLAN_OFFSET);
+	else
+		pvid = readl(prueth->shram.va + EMAC_ICSSG_SWITCH_PORT2_DEFAULT_VLAN_OFFSET);
+
+	pvid = pvid >> 24;
+
+	return pvid;
+}
+
+void icssg_set_pvid(struct prueth *prueth, u8 vid, u8 port)
+{
+	u32 pvid;
+
+	/* only 256 VLANs are supported */
+	pvid = (u32 __force)cpu_to_be32((ETH_P_8021Q << 16) | (vid & 0xff));
+
+	if (port == PRUETH_PORT_MII0)
+		writel(pvid, prueth->shram.va + EMAC_ICSSG_SWITCH_PORT1_DEFAULT_VLAN_OFFSET);
+	else if (port == PRUETH_PORT_MII1)
+		writel(pvid, prueth->shram.va + EMAC_ICSSG_SWITCH_PORT2_DEFAULT_VLAN_OFFSET);
+	else
+		writel(pvid, prueth->shram.va + EMAC_ICSSG_SWITCH_PORT0_DEFAULT_VLAN_OFFSET);
+}
diff --git a/drivers/net/ethernet/ti/icssg/icssg_config.h b/drivers/net/ethernet/ti/icssg/icssg_config.h
index cf2ea4bd22a2..4a9721aa6057 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_config.h
+++ b/drivers/net/ethernet/ti/icssg/icssg_config.h
@@ -35,6 +35,8 @@ struct icssg_flow_cfg {
 	(2 * (PRUETH_EMAC_BUF_POOL_SIZE * PRUETH_NUM_BUF_POOLS + \
 	 PRUETH_EMAC_RX_CTX_BUF_SIZE * 2))
 
+#define PRUETH_SWITCH_FDB_MASK ((SIZE_OF_FDB / NUMBER_OF_FDB_BUCKET_ENTRIES) - 1)
+
 struct icssg_rxq_ctx {
 	__le32 start[3];
 	__le32 end;
@@ -202,6 +204,23 @@ struct icssg_setclock_desc {
 #define ICSSG_TS_PUSH_SLICE0	40
 #define ICSSG_TS_PUSH_SLICE1	41
 
+struct mgmt_cmd {
+	u8 param;
+	u8 seqnum;
+	u8 type;
+	u8 header;
+	u32 cmd_args[3];
+};
+
+struct mgmt_cmd_rsp {
+	u32 reserved;
+	u8 status;
+	u8 seqnum;
+	u8 type;
+	u8 header;
+	u32 cmd_args[3];
+};
+
 /* FDB FID_C2 flag definitions */
 /* Indicates host port membership.*/
 #define ICSSG_FDB_ENTRY_P0_MEMBERSHIP         BIT(0)
diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.h b/drivers/net/ethernet/ti/icssg/icssg_prueth.h
index a78c5eb75fb8..82bdad9702c3 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_prueth.h
+++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.h
@@ -232,6 +232,7 @@ struct icssg_firmwares {
  * @emacs_initialized: num of EMACs/ext ports that are up/running
  * @iep0: pointer to IEP0 device
  * @iep1: pointer to IEP1 device
+ * @vlan_tbl: VLAN-FID table pointer
  */
 struct prueth {
 	struct device *dev;
@@ -256,6 +257,7 @@ struct prueth {
 	int emacs_initialized;
 	struct icss_iep *iep0;
 	struct icss_iep *iep1;
+	struct prueth_vlan_tbl *vlan_tbl;
 };
 
 struct emac_tx_ts_response {
@@ -313,6 +315,16 @@ int icssg_queue_pop(struct prueth *prueth, u8 queue);
 void icssg_queue_push(struct prueth *prueth, int queue, u16 addr);
 u32 icssg_queue_level(struct prueth *prueth, int queue);
 
+int icssg_send_fdb_msg(struct prueth_emac *emac, struct mgmt_cmd *cmd,
+		       struct mgmt_cmd_rsp *rsp);
+int icssg_fdb_add_del(struct prueth_emac *emac,  const unsigned char *addr,
+		      u8 vid, u8 fid_c2, bool add);
+int icssg_fdb_lookup(struct prueth_emac *emac, const unsigned char *addr,
+		     u8 vid);
+void icssg_vtbl_modify(struct prueth_emac *emac, u8 vid, u8 port_mask,
+		       u8 untag_mask, bool add);
+u16 icssg_get_pvid(struct prueth_emac *emac);
+void icssg_set_pvid(struct prueth *prueth, u8 vid, u8 port);
 #define prueth_napi_to_tx_chn(pnapi) \
 	container_of(pnapi, struct prueth_tx_chn, napi_tx)
 
-- 
cgit v1.2.3-58-ga151


From 972383aecf43c3664750496d5395e669c35ba06e Mon Sep 17 00:00:00 2001
From: MD Danish Anwar <danishanwar@ti.com>
Date: Tue, 28 May 2024 17:07:33 +0530
Subject: net: ti: icssg-switch: Add switchdev based driver for ethernet switch
 support

ICSSG can operating in switch mode with 2 ext port and 1 host port with
VLAN/FDB/MDB and STP offloading. Add switchdev based driver to
support the same.

Driver itself will be integrated with icssg_prueth in future commits

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/ti/icssg/icssg_prueth.c    |  12 +
 drivers/net/ethernet/ti/icssg/icssg_prueth.h    |  24 ++
 drivers/net/ethernet/ti/icssg/icssg_switchdev.c | 477 ++++++++++++++++++++++++
 drivers/net/ethernet/ti/icssg/icssg_switchdev.h |  13 +
 4 files changed, 526 insertions(+)
 create mode 100644 drivers/net/ethernet/ti/icssg/icssg_switchdev.c
 create mode 100644 drivers/net/ethernet/ti/icssg/icssg_switchdev.h

diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
index 1ea3fbd5e954..1db67a8107cc 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
@@ -30,6 +30,7 @@
 
 #include "icssg_prueth.h"
 #include "icssg_mii_rt.h"
+#include "icssg_switchdev.h"
 #include "../k3-cppi-desc-pool.h"
 
 #define PRUETH_MODULE_DESCRIPTION "PRUSS ICSSG Ethernet driver"
@@ -833,6 +834,17 @@ free_ndev:
 	return ret;
 }
 
+bool prueth_dev_check(const struct net_device *ndev)
+{
+	if (ndev->netdev_ops == &emac_netdev_ops && netif_running(ndev)) {
+		struct prueth_emac *emac = netdev_priv(ndev);
+
+		return emac->prueth->is_switch_mode;
+	}
+
+	return false;
+}
+
 static int prueth_probe(struct platform_device *pdev)
 {
 	struct device_node *eth_node, *eth_ports_node;
diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.h b/drivers/net/ethernet/ti/icssg/icssg_prueth.h
index 82bdad9702c3..5eeeccb73665 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_prueth.h
+++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.h
@@ -186,6 +186,9 @@ struct prueth_emac {
 
 	struct pruss_mem_region dram;
 
+	bool offload_fwd_mark;
+	int port_vlan;
+
 	struct delayed_work stats_work;
 	u64 stats[ICSSG_NUM_STATS];
 
@@ -198,10 +201,12 @@ struct prueth_emac {
  * struct prueth_pdata - PRUeth platform data
  * @fdqring_mode: Free desc queue mode
  * @quirk_10m_link_issue: 10M link detect errata
+ * @switch_mode: switch firmware support
  */
 struct prueth_pdata {
 	enum k3_ring_mode fdqring_mode;
 	u32	quirk_10m_link_issue:1;
+	u32	switch_mode:1;
 };
 
 struct icssg_firmwares {
@@ -233,6 +238,15 @@ struct icssg_firmwares {
  * @iep0: pointer to IEP0 device
  * @iep1: pointer to IEP1 device
  * @vlan_tbl: VLAN-FID table pointer
+ * @hw_bridge_dev: pointer to HW bridge net device
+ * @br_members: bitmask of bridge member ports
+ * @prueth_netdevice_nb: netdevice notifier block
+ * @prueth_switchdev_nb: switchdev notifier block
+ * @prueth_switchdev_bl_nb: switchdev blocking notifier block
+ * @is_switch_mode: flag to indicate if device is in Switch mode
+ * @is_switchmode_supported: indicates platform support for switch mode
+ * @switch_id: ID for mapping switch ports to bridge
+ * @default_vlan: Default VLAN for host
  */
 struct prueth {
 	struct device *dev;
@@ -258,6 +272,16 @@ struct prueth {
 	struct icss_iep *iep0;
 	struct icss_iep *iep1;
 	struct prueth_vlan_tbl *vlan_tbl;
+
+	struct net_device *hw_bridge_dev;
+	u8 br_members;
+	struct notifier_block prueth_netdevice_nb;
+	struct notifier_block prueth_switchdev_nb;
+	struct notifier_block prueth_switchdev_bl_nb;
+	bool is_switch_mode;
+	bool is_switchmode_supported;
+	unsigned char switch_id[MAX_PHYS_ITEM_ID_LEN];
+	int default_vlan;
 };
 
 struct emac_tx_ts_response {
diff --git a/drivers/net/ethernet/ti/icssg/icssg_switchdev.c b/drivers/net/ethernet/ti/icssg/icssg_switchdev.c
new file mode 100644
index 000000000000..fceb8bb7d34e
--- /dev/null
+++ b/drivers/net/ethernet/ti/icssg/icssg_switchdev.c
@@ -0,0 +1,477 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/* Texas Instruments K3 ICSSG Ethernet Switchdev Driver
+ *
+ * Copyright (C) 2021 Texas Instruments Incorporated - https://www.ti.com/
+ *
+ */
+
+#include <linux/etherdevice.h>
+#include <linux/if_bridge.h>
+#include <linux/netdevice.h>
+#include <linux/workqueue.h>
+#include <net/switchdev.h>
+
+#include "icssg_prueth.h"
+#include "icssg_switchdev.h"
+#include "icssg_mii_rt.h"
+
+struct prueth_switchdev_event_work {
+	struct work_struct work;
+	struct switchdev_notifier_fdb_info fdb_info;
+	struct prueth_emac *emac;
+	unsigned long event;
+};
+
+static int prueth_switchdev_stp_state_set(struct prueth_emac *emac,
+					  u8 state)
+{
+	enum icssg_port_state_cmd emac_state;
+	int ret = 0;
+
+	switch (state) {
+	case BR_STATE_FORWARDING:
+		emac_state = ICSSG_EMAC_PORT_FORWARD;
+		break;
+	case BR_STATE_DISABLED:
+		emac_state = ICSSG_EMAC_PORT_DISABLE;
+		break;
+	case BR_STATE_LISTENING:
+	case BR_STATE_BLOCKING:
+		emac_state = ICSSG_EMAC_PORT_BLOCK;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	emac_set_port_state(emac, emac_state);
+	netdev_dbg(emac->ndev, "STP state: %u\n", emac_state);
+
+	return ret;
+}
+
+static int prueth_switchdev_attr_br_flags_set(struct prueth_emac *emac,
+					      struct net_device *orig_dev,
+					      struct switchdev_brport_flags brport_flags)
+{
+	enum icssg_port_state_cmd emac_state;
+
+	if (brport_flags.mask & BR_MCAST_FLOOD)
+		emac_state = ICSSG_EMAC_PORT_MC_FLOODING_ENABLE;
+	else
+		emac_state = ICSSG_EMAC_PORT_MC_FLOODING_DISABLE;
+
+	netdev_dbg(emac->ndev, "BR_MCAST_FLOOD: %d port %u\n",
+		   emac_state, emac->port_id);
+
+	emac_set_port_state(emac, emac_state);
+
+	return 0;
+}
+
+static int prueth_switchdev_attr_br_flags_pre_set(struct net_device *netdev,
+						  struct switchdev_brport_flags brport_flags)
+{
+	if (brport_flags.mask & ~(BR_LEARNING | BR_MCAST_FLOOD))
+		return -EINVAL;
+
+	return 0;
+}
+
+static int prueth_switchdev_attr_set(struct net_device *ndev, const void *ctx,
+				     const struct switchdev_attr *attr,
+				     struct netlink_ext_ack *extack)
+{
+	struct prueth_emac *emac = netdev_priv(ndev);
+	int ret;
+
+	netdev_dbg(ndev, "attr: id %u port: %u\n", attr->id, emac->port_id);
+
+	switch (attr->id) {
+	case SWITCHDEV_ATTR_ID_PORT_PRE_BRIDGE_FLAGS:
+		ret = prueth_switchdev_attr_br_flags_pre_set(ndev,
+							     attr->u.brport_flags);
+		break;
+	case SWITCHDEV_ATTR_ID_PORT_STP_STATE:
+		ret = prueth_switchdev_stp_state_set(emac,
+						     attr->u.stp_state);
+		netdev_dbg(ndev, "stp state: %u\n", attr->u.stp_state);
+		break;
+	case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS:
+		ret = prueth_switchdev_attr_br_flags_set(emac, attr->orig_dev,
+							 attr->u.brport_flags);
+		break;
+	default:
+		ret = -EOPNOTSUPP;
+		break;
+	}
+
+	return ret;
+}
+
+static void prueth_switchdev_fdb_offload_notify(struct net_device *ndev,
+						struct switchdev_notifier_fdb_info *rcv)
+{
+	struct switchdev_notifier_fdb_info info;
+
+	memset(&info, 0, sizeof(info));
+	info.addr = rcv->addr;
+	info.vid = rcv->vid;
+	info.offloaded = true;
+	call_switchdev_notifiers(SWITCHDEV_FDB_OFFLOADED,
+				 ndev, &info.info, NULL);
+}
+
+static void prueth_switchdev_event_work(struct work_struct *work)
+{
+	struct prueth_switchdev_event_work *switchdev_work =
+		container_of(work, struct prueth_switchdev_event_work, work);
+	struct prueth_emac *emac = switchdev_work->emac;
+	struct switchdev_notifier_fdb_info *fdb;
+	int port_id = emac->port_id;
+	int ret;
+
+	rtnl_lock();
+	switch (switchdev_work->event) {
+	case SWITCHDEV_FDB_ADD_TO_DEVICE:
+		fdb = &switchdev_work->fdb_info;
+
+		netdev_dbg(emac->ndev, "prueth_fdb_add: MACID = %pM vid = %u flags = %u %u -- port %d\n",
+			   fdb->addr, fdb->vid, fdb->added_by_user,
+			   fdb->offloaded, port_id);
+
+		if (!fdb->added_by_user)
+			break;
+		if (!ether_addr_equal(emac->mac_addr, fdb->addr))
+			break;
+
+		ret = icssg_fdb_add_del(emac, fdb->addr, fdb->vid,
+					BIT(port_id), true);
+		if (!ret)
+			prueth_switchdev_fdb_offload_notify(emac->ndev, fdb);
+		break;
+	case SWITCHDEV_FDB_DEL_TO_DEVICE:
+		fdb = &switchdev_work->fdb_info;
+
+		netdev_dbg(emac->ndev, "prueth_fdb_del: MACID = %pM vid = %u flags = %u %u -- port %d\n",
+			   fdb->addr, fdb->vid, fdb->added_by_user,
+			   fdb->offloaded, port_id);
+
+		if (!fdb->added_by_user)
+			break;
+		if (!ether_addr_equal(emac->mac_addr, fdb->addr))
+			break;
+		icssg_fdb_add_del(emac, fdb->addr, fdb->vid,
+				  BIT(port_id), false);
+		break;
+	default:
+		break;
+	}
+	rtnl_unlock();
+
+	kfree(switchdev_work->fdb_info.addr);
+	kfree(switchdev_work);
+	dev_put(emac->ndev);
+}
+
+static int prueth_switchdev_event(struct notifier_block *unused,
+				  unsigned long event, void *ptr)
+{
+	struct net_device *ndev = switchdev_notifier_info_to_dev(ptr);
+	struct prueth_switchdev_event_work *switchdev_work;
+	struct switchdev_notifier_fdb_info *fdb_info = ptr;
+	struct prueth_emac *emac = netdev_priv(ndev);
+	int err;
+
+	if (!prueth_dev_check(ndev))
+		return NOTIFY_DONE;
+
+	if (event == SWITCHDEV_PORT_ATTR_SET) {
+		err = switchdev_handle_port_attr_set(ndev, ptr,
+						     prueth_dev_check,
+						     prueth_switchdev_attr_set);
+		return notifier_from_errno(err);
+	}
+
+	switchdev_work = kzalloc(sizeof(*switchdev_work), GFP_ATOMIC);
+	if (WARN_ON(!switchdev_work))
+		return NOTIFY_BAD;
+
+	INIT_WORK(&switchdev_work->work, prueth_switchdev_event_work);
+	switchdev_work->emac = emac;
+	switchdev_work->event = event;
+
+	switch (event) {
+	case SWITCHDEV_FDB_ADD_TO_DEVICE:
+	case SWITCHDEV_FDB_DEL_TO_DEVICE:
+		memcpy(&switchdev_work->fdb_info, ptr,
+		       sizeof(switchdev_work->fdb_info));
+		switchdev_work->fdb_info.addr = kzalloc(ETH_ALEN, GFP_ATOMIC);
+		if (!switchdev_work->fdb_info.addr)
+			goto err_addr_alloc;
+		ether_addr_copy((u8 *)switchdev_work->fdb_info.addr,
+				fdb_info->addr);
+		dev_hold(ndev);
+		break;
+	default:
+		kfree(switchdev_work);
+		return NOTIFY_DONE;
+	}
+
+	queue_work(system_long_wq, &switchdev_work->work);
+
+	return NOTIFY_DONE;
+
+err_addr_alloc:
+	kfree(switchdev_work);
+	return NOTIFY_BAD;
+}
+
+static int prueth_switchdev_vlan_add(struct prueth_emac *emac, bool untag, bool pvid,
+				     u8 vid, struct net_device *orig_dev)
+{
+	bool cpu_port = netif_is_bridge_master(orig_dev);
+	int untag_mask = 0;
+	int port_mask;
+	int ret = 0;
+
+	if (cpu_port)
+		port_mask = BIT(PRUETH_PORT_HOST);
+	else
+		port_mask = BIT(emac->port_id);
+
+	if (untag)
+		untag_mask = port_mask;
+
+	icssg_vtbl_modify(emac, vid, port_mask, untag_mask, true);
+
+	netdev_dbg(emac->ndev, "VID add vid:%u port_mask:%X untag_mask %X PVID %d\n",
+		   vid, port_mask, untag_mask, pvid);
+
+	if (!pvid)
+		return ret;
+
+	icssg_set_pvid(emac->prueth, vid, emac->port_id);
+
+	return ret;
+}
+
+static int prueth_switchdev_vlan_del(struct prueth_emac *emac, u16 vid,
+				     struct net_device *orig_dev)
+{
+	bool cpu_port = netif_is_bridge_master(orig_dev);
+	int port_mask;
+	int ret = 0;
+
+	if (cpu_port)
+		port_mask = BIT(PRUETH_PORT_HOST);
+	else
+		port_mask = BIT(emac->port_id);
+
+	icssg_vtbl_modify(emac, vid, port_mask, 0, false);
+
+	if (cpu_port)
+		icssg_fdb_add_del(emac, emac->mac_addr, vid,
+				  BIT(PRUETH_PORT_HOST), false);
+
+	if (vid == icssg_get_pvid(emac))
+		icssg_set_pvid(emac->prueth, 0, emac->port_id);
+
+	netdev_dbg(emac->ndev, "VID del vid:%u port_mask:%X\n",
+		   vid, port_mask);
+
+	return ret;
+}
+
+static int prueth_switchdev_vlans_add(struct prueth_emac *emac,
+				      const struct switchdev_obj_port_vlan *vlan)
+{
+	bool untag = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED;
+	struct net_device *orig_dev = vlan->obj.orig_dev;
+	bool cpu_port = netif_is_bridge_master(orig_dev);
+	bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID;
+
+	netdev_dbg(emac->ndev, "VID add vid:%u flags:%X\n",
+		   vlan->vid, vlan->flags);
+
+	if (cpu_port && !(vlan->flags & BRIDGE_VLAN_INFO_BRENTRY))
+		return 0;
+
+	if (vlan->vid > 0xff)
+		return 0;
+
+	return prueth_switchdev_vlan_add(emac, untag, pvid, vlan->vid,
+					 orig_dev);
+}
+
+static int prueth_switchdev_vlans_del(struct prueth_emac *emac,
+				      const struct switchdev_obj_port_vlan *vlan)
+{
+	if (vlan->vid > 0xff)
+		return 0;
+
+	return prueth_switchdev_vlan_del(emac, vlan->vid,
+					 vlan->obj.orig_dev);
+}
+
+static int prueth_switchdev_mdb_add(struct prueth_emac *emac,
+				    struct switchdev_obj_port_mdb *mdb)
+{
+	struct net_device *orig_dev = mdb->obj.orig_dev;
+	u8 port_mask, fid_c2;
+	bool cpu_port;
+	int err;
+
+	cpu_port = netif_is_bridge_master(orig_dev);
+
+	if (cpu_port)
+		port_mask = BIT(PRUETH_PORT_HOST);
+	else
+		port_mask = BIT(emac->port_id);
+
+	fid_c2 = icssg_fdb_lookup(emac, mdb->addr, mdb->vid);
+
+	err = icssg_fdb_add_del(emac, mdb->addr, mdb->vid, fid_c2 | port_mask, true);
+	netdev_dbg(emac->ndev, "MDB add vid %u:%pM  ports: %X\n",
+		   mdb->vid, mdb->addr, port_mask);
+
+	return err;
+}
+
+static int prueth_switchdev_mdb_del(struct prueth_emac *emac,
+				    struct switchdev_obj_port_mdb *mdb)
+{
+	struct net_device *orig_dev = mdb->obj.orig_dev;
+	int del_mask, ret, fid_c2;
+	bool cpu_port;
+
+	cpu_port = netif_is_bridge_master(orig_dev);
+
+	if (cpu_port)
+		del_mask = BIT(PRUETH_PORT_HOST);
+	else
+		del_mask = BIT(emac->port_id);
+
+	fid_c2 = icssg_fdb_lookup(emac, mdb->addr, mdb->vid);
+
+	if (fid_c2 & ~del_mask)
+		ret = icssg_fdb_add_del(emac, mdb->addr, mdb->vid, fid_c2 & ~del_mask, true);
+	else
+		ret = icssg_fdb_add_del(emac, mdb->addr, mdb->vid, 0, false);
+
+	netdev_dbg(emac->ndev, "MDB del vid %u:%pM  ports: %X\n",
+		   mdb->vid, mdb->addr, del_mask);
+
+	return ret;
+}
+
+static int prueth_switchdev_obj_add(struct net_device *ndev, const void *ctx,
+				    const struct switchdev_obj *obj,
+				    struct netlink_ext_ack *extack)
+{
+	struct switchdev_obj_port_vlan *vlan = SWITCHDEV_OBJ_PORT_VLAN(obj);
+	struct switchdev_obj_port_mdb *mdb = SWITCHDEV_OBJ_PORT_MDB(obj);
+	struct prueth_emac *emac = netdev_priv(ndev);
+	int err = 0;
+
+	netdev_dbg(ndev, "obj_add: id %u port: %u\n", obj->id, emac->port_id);
+
+	switch (obj->id) {
+	case SWITCHDEV_OBJ_ID_PORT_VLAN:
+		err = prueth_switchdev_vlans_add(emac, vlan);
+		break;
+	case SWITCHDEV_OBJ_ID_PORT_MDB:
+	case SWITCHDEV_OBJ_ID_HOST_MDB:
+		err = prueth_switchdev_mdb_add(emac, mdb);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int prueth_switchdev_obj_del(struct net_device *ndev, const void *ctx,
+				    const struct switchdev_obj *obj)
+{
+	struct switchdev_obj_port_vlan *vlan = SWITCHDEV_OBJ_PORT_VLAN(obj);
+	struct switchdev_obj_port_mdb *mdb = SWITCHDEV_OBJ_PORT_MDB(obj);
+	struct prueth_emac *emac = netdev_priv(ndev);
+	int err = 0;
+
+	netdev_dbg(ndev, "obj_del: id %u port: %u\n", obj->id, emac->port_id);
+
+	switch (obj->id) {
+	case SWITCHDEV_OBJ_ID_PORT_VLAN:
+		err = prueth_switchdev_vlans_del(emac, vlan);
+		break;
+	case SWITCHDEV_OBJ_ID_PORT_MDB:
+	case SWITCHDEV_OBJ_ID_HOST_MDB:
+		err = prueth_switchdev_mdb_del(emac, mdb);
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int prueth_switchdev_blocking_event(struct notifier_block *unused,
+					   unsigned long event, void *ptr)
+{
+	struct net_device *dev = switchdev_notifier_info_to_dev(ptr);
+	int err;
+
+	switch (event) {
+	case SWITCHDEV_PORT_OBJ_ADD:
+		err = switchdev_handle_port_obj_add(dev, ptr,
+						    prueth_dev_check,
+						    prueth_switchdev_obj_add);
+		return notifier_from_errno(err);
+	case SWITCHDEV_PORT_OBJ_DEL:
+		err = switchdev_handle_port_obj_del(dev, ptr,
+						    prueth_dev_check,
+						    prueth_switchdev_obj_del);
+		return notifier_from_errno(err);
+	case SWITCHDEV_PORT_ATTR_SET:
+		err = switchdev_handle_port_attr_set(dev, ptr,
+						     prueth_dev_check,
+						     prueth_switchdev_attr_set);
+		return notifier_from_errno(err);
+	default:
+		break;
+	}
+
+	return NOTIFY_DONE;
+}
+
+int prueth_switchdev_register_notifiers(struct prueth *prueth)
+{
+	int ret = 0;
+
+	prueth->prueth_switchdev_nb.notifier_call = &prueth_switchdev_event;
+	ret = register_switchdev_notifier(&prueth->prueth_switchdev_nb);
+	if (ret) {
+		dev_err(prueth->dev, "register switchdev notifier fail ret:%d\n",
+			ret);
+		return ret;
+	}
+
+	prueth->prueth_switchdev_bl_nb.notifier_call = &prueth_switchdev_blocking_event;
+	ret = register_switchdev_blocking_notifier(&prueth->prueth_switchdev_bl_nb);
+	if (ret) {
+		dev_err(prueth->dev, "register switchdev blocking notifier ret:%d\n",
+			ret);
+		unregister_switchdev_notifier(&prueth->prueth_switchdev_nb);
+	}
+
+	return ret;
+}
+
+void prueth_switchdev_unregister_notifiers(struct prueth *prueth)
+{
+	unregister_switchdev_blocking_notifier(&prueth->prueth_switchdev_bl_nb);
+	unregister_switchdev_notifier(&prueth->prueth_switchdev_nb);
+}
diff --git a/drivers/net/ethernet/ti/icssg/icssg_switchdev.h b/drivers/net/ethernet/ti/icssg/icssg_switchdev.h
new file mode 100644
index 000000000000..0e64e7760a00
--- /dev/null
+++ b/drivers/net/ethernet/ti/icssg/icssg_switchdev.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (C) 2021 Texas Instruments Incorporated - https://www.ti.com/
+ */
+#ifndef __NET_TI_ICSSG_SWITCHDEV_H
+#define __NET_TI_ICSSG_SWITCHDEV_H
+
+#include "icssg_prueth.h"
+
+int prueth_switchdev_register_notifiers(struct prueth *prueth);
+void prueth_switchdev_unregister_notifiers(struct prueth *prueth);
+bool prueth_dev_check(const struct net_device *ndev);
+
+#endif /* __NET_TI_ICSSG_SWITCHDEV_H */
-- 
cgit v1.2.3-58-ga151


From abd5576b9c57fbf1ac2421463900061f4d67d96f Mon Sep 17 00:00:00 2001
From: MD Danish Anwar <danishanwar@ti.com>
Date: Tue, 28 May 2024 17:07:34 +0530
Subject: net: ti: icssg-prueth: Add support for ICSSG switch firmware

Add support for ICSSG switch firmware using existing Dual EMAC driver
with switchdev.

Limitations:
VLAN offloading is limited to 0-256 IDs.
MDB/FDB static entries are limited to 511 entries and different FDBs can
hash to same bucket and thus may not completely offloaded

Example assuming ETH1 and ETH2 as ICSSG2 interfaces:

Switch to ICSSG Switch mode:
 ip link add name br0 type bridge
 ip link set dev eth1 master br0
 ip link set dev eth2 master br0
 ip link set dev br0 up
 bridge vlan add dev br0 vid 1 pvid untagged self

Going back to Dual EMAC mode:

 ip link set dev br0 down
 ip link set dev eth1 nomaster
 ip link set dev eth2 nomaster
 ip link del name br0 type bridge

By default, Dual EMAC firmware is loaded, and can be changed to switch
mode by above steps

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/ti/Kconfig                  |   1 +
 drivers/net/ethernet/ti/Makefile                 |   3 +-
 drivers/net/ethernet/ti/icssg/icssg_classifier.c |   2 +-
 drivers/net/ethernet/ti/icssg/icssg_common.c     |   2 +
 drivers/net/ethernet/ti/icssg/icssg_config.c     | 152 +++++++++++++--
 drivers/net/ethernet/ti/icssg/icssg_config.h     |   7 +
 drivers/net/ethernet/ti/icssg/icssg_prueth.c     | 238 ++++++++++++++++++++++-
 7 files changed, 386 insertions(+), 19 deletions(-)

diff --git a/drivers/net/ethernet/ti/Kconfig b/drivers/net/ethernet/ti/Kconfig
index 1729eb0e0b41..f160a3b71499 100644
--- a/drivers/net/ethernet/ti/Kconfig
+++ b/drivers/net/ethernet/ti/Kconfig
@@ -204,6 +204,7 @@ config TI_ICSSG_PRUETH_SR1
 	select TI_ICSS_IEP
 	select TI_K3_CPPI_DESC_POOL
 	depends on PRU_REMOTEPROC
+	depends on NET_SWITCHDEV
 	depends on ARCH_K3 && OF && TI_K3_UDMA_GLUE_LAYER
 	help
 	  Support dual Gigabit Ethernet ports over the ICSSG PRU Subsystem.
diff --git a/drivers/net/ethernet/ti/Makefile b/drivers/net/ethernet/ti/Makefile
index 6e086b4c0384..59cd20a38267 100644
--- a/drivers/net/ethernet/ti/Makefile
+++ b/drivers/net/ethernet/ti/Makefile
@@ -39,7 +39,8 @@ icssg-prueth-y := icssg/icssg_prueth.o \
 		  icssg/icssg_config.o \
 		  icssg/icssg_mii_cfg.o \
 		  icssg/icssg_stats.o \
-		  icssg/icssg_ethtool.o
+		  icssg/icssg_ethtool.o \
+		  icssg/icssg_switchdev.o
 obj-$(CONFIG_TI_ICSSG_PRUETH_SR1) += icssg-prueth-sr1.o
 icssg-prueth-sr1-y := icssg/icssg_prueth_sr1.o \
 		      icssg/icssg_common.o \
diff --git a/drivers/net/ethernet/ti/icssg/icssg_classifier.c b/drivers/net/ethernet/ti/icssg/icssg_classifier.c
index 79ba47bb3602..8dee737639b6 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_classifier.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_classifier.c
@@ -455,7 +455,7 @@ void icssg_ft1_set_mac_addr(struct regmap *miig_rt, int slice, u8 *mac_addr)
 {
 	const u8 mask_addr[] = { 0, 0, 0, 0, 0, 0, };
 
-	rx_class_ft1_set_start_len(miig_rt, slice, 0, 6);
+	rx_class_ft1_set_start_len(miig_rt, slice, 6, 6);
 	rx_class_ft1_set_da(miig_rt, slice, 0, mac_addr);
 	rx_class_ft1_set_da_mask(miig_rt, slice, 0, mask_addr);
 	rx_class_ft1_cfg_set_type(miig_rt, slice, 0, FT1_CFG_TYPE_EQ);
diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c
index 088ab8076db4..873126dfc173 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_common.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_common.c
@@ -581,6 +581,8 @@ static int emac_rx_packet(struct prueth_emac *emac, u32 flow_id)
 	} else {
 		/* send the filled skb up the n/w stack */
 		skb_put(skb, pkt_len);
+		if (emac->prueth->is_switch_mode)
+			skb->offload_fwd_mark = emac->offload_fwd_mark;
 		skb->protocol = eth_type_trans(skb, ndev);
 		napi_gro_receive(&emac->napi_rx, skb);
 		ndev->stats.rx_bytes += pkt_len;
diff --git a/drivers/net/ethernet/ti/icssg/icssg_config.c b/drivers/net/ethernet/ti/icssg/icssg_config.c
index 2213374d4d45..9444e56b7672 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_config.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_config.c
@@ -107,28 +107,49 @@ static const struct map hwq_map[2][ICSSG_NUM_OTHER_QUEUES] = {
 	},
 };
 
+static void icssg_config_mii_init_switch(struct prueth_emac *emac)
+{
+	struct prueth *prueth = emac->prueth;
+	int mii = prueth_emac_slice(emac);
+	u32 txcfg_reg, pcnt_reg, txcfg;
+	struct regmap *mii_rt;
+
+	mii_rt = prueth->mii_rt;
+
+	txcfg_reg = (mii == ICSS_MII0) ? PRUSS_MII_RT_TXCFG0 :
+				       PRUSS_MII_RT_TXCFG1;
+	pcnt_reg = (mii == ICSS_MII0) ? PRUSS_MII_RT_RX_PCNT0 :
+				       PRUSS_MII_RT_RX_PCNT1;
+
+	txcfg = PRUSS_MII_RT_TXCFG_TX_ENABLE |
+		PRUSS_MII_RT_TXCFG_TX_AUTO_PREAMBLE |
+		PRUSS_MII_RT_TXCFG_TX_IPG_WIRE_CLK_EN;
+
+	if (emac->phy_if == PHY_INTERFACE_MODE_MII && mii == ICSS_MII1)
+		txcfg |= PRUSS_MII_RT_TXCFG_TX_MUX_SEL;
+	else if (emac->phy_if != PHY_INTERFACE_MODE_MII && mii == ICSS_MII0)
+		txcfg |= PRUSS_MII_RT_TXCFG_TX_MUX_SEL;
+
+	regmap_write(mii_rt, txcfg_reg, txcfg);
+	regmap_write(mii_rt, pcnt_reg, 0x1);
+}
+
 static void icssg_config_mii_init(struct prueth_emac *emac)
 {
-	u32 rxcfg, txcfg, rxcfg_reg, txcfg_reg, pcnt_reg;
 	struct prueth *prueth = emac->prueth;
 	int slice = prueth_emac_slice(emac);
+	u32 txcfg, txcfg_reg, pcnt_reg;
 	struct regmap *mii_rt;
 
 	mii_rt = prueth->mii_rt;
 
-	rxcfg_reg = (slice == ICSS_MII0) ? PRUSS_MII_RT_RXCFG0 :
-				       PRUSS_MII_RT_RXCFG1;
 	txcfg_reg = (slice == ICSS_MII0) ? PRUSS_MII_RT_TXCFG0 :
 				       PRUSS_MII_RT_TXCFG1;
 	pcnt_reg = (slice == ICSS_MII0) ? PRUSS_MII_RT_RX_PCNT0 :
 				       PRUSS_MII_RT_RX_PCNT1;
 
-	rxcfg = MII_RXCFG_DEFAULT;
 	txcfg = MII_TXCFG_DEFAULT;
 
-	if (slice == ICSS_MII1)
-		rxcfg |= PRUSS_MII_RT_RXCFG_RX_MUX_SEL;
-
 	/* In MII mode TX lines swapped inside ICSSG, so TX_MUX_SEL cfg need
 	 * to be swapped also comparing to RGMII mode.
 	 */
@@ -137,7 +158,6 @@ static void icssg_config_mii_init(struct prueth_emac *emac)
 	else if (emac->phy_if != PHY_INTERFACE_MODE_MII && slice == ICSS_MII1)
 		txcfg |= PRUSS_MII_RT_TXCFG_TX_MUX_SEL;
 
-	regmap_write(mii_rt, rxcfg_reg, rxcfg);
 	regmap_write(mii_rt, txcfg_reg, txcfg);
 	regmap_write(mii_rt, pcnt_reg, 0x1);
 }
@@ -257,6 +277,66 @@ static int emac_r30_is_done(struct prueth_emac *emac)
 	return 1;
 }
 
+static int prueth_switch_buffer_setup(struct prueth_emac *emac)
+{
+	struct icssg_buffer_pool_cfg __iomem *bpool_cfg;
+	struct icssg_rxq_ctx __iomem *rxq_ctx;
+	struct prueth *prueth = emac->prueth;
+	int slice = prueth_emac_slice(emac);
+	u32 addr;
+	int i;
+
+	addr = lower_32_bits(prueth->msmcram.pa);
+	if (slice)
+		addr += PRUETH_NUM_BUF_POOLS * PRUETH_EMAC_BUF_POOL_SIZE;
+
+	if (addr % SZ_64K) {
+		dev_warn(prueth->dev, "buffer pool needs to be 64KB aligned\n");
+		return -EINVAL;
+	}
+
+	bpool_cfg = emac->dram.va + BUFFER_POOL_0_ADDR_OFFSET;
+	/* workaround for f/w bug. bpool 0 needs to be initialized */
+	for (i = 0; i <  PRUETH_NUM_BUF_POOLS; i++) {
+		writel(addr, &bpool_cfg[i].addr);
+		writel(PRUETH_EMAC_BUF_POOL_SIZE, &bpool_cfg[i].len);
+		addr += PRUETH_EMAC_BUF_POOL_SIZE;
+	}
+
+	if (!slice)
+		addr += PRUETH_NUM_BUF_POOLS * PRUETH_EMAC_BUF_POOL_SIZE;
+	else
+		addr += PRUETH_SW_NUM_BUF_POOLS_HOST * PRUETH_SW_BUF_POOL_SIZE_HOST;
+
+	for (i = PRUETH_NUM_BUF_POOLS;
+	     i < 2 * PRUETH_SW_NUM_BUF_POOLS_HOST + PRUETH_NUM_BUF_POOLS;
+	     i++) {
+		/* The driver only uses first 4 queues per PRU so only initialize them */
+		if (i % PRUETH_SW_NUM_BUF_POOLS_HOST < PRUETH_SW_NUM_BUF_POOLS_PER_PRU) {
+			writel(addr, &bpool_cfg[i].addr);
+			writel(PRUETH_SW_BUF_POOL_SIZE_HOST, &bpool_cfg[i].len);
+			addr += PRUETH_SW_BUF_POOL_SIZE_HOST;
+		} else {
+			writel(0, &bpool_cfg[i].addr);
+			writel(0, &bpool_cfg[i].len);
+		}
+	}
+
+	if (!slice)
+		addr += PRUETH_SW_NUM_BUF_POOLS_HOST * PRUETH_SW_BUF_POOL_SIZE_HOST;
+	else
+		addr += PRUETH_EMAC_RX_CTX_BUF_SIZE;
+
+	rxq_ctx = emac->dram.va + HOST_RX_Q_PRE_CONTEXT_OFFSET;
+	for (i = 0; i < 3; i++)
+		writel(addr, &rxq_ctx->start[i]);
+
+	addr += PRUETH_EMAC_RX_CTX_BUF_SIZE;
+	writel(addr - SZ_2K, &rxq_ctx->end);
+
+	return 0;
+}
+
 static int prueth_emac_buffer_setup(struct prueth_emac *emac)
 {
 	struct icssg_buffer_pool_cfg __iomem *bpool_cfg;
@@ -321,25 +401,63 @@ static void icssg_init_emac_mode(struct prueth *prueth)
 	/* When the device is configured as a bridge and it is being brought
 	 * back to the emac mode, the host mac address has to be set as 0.
 	 */
+	u32 addr = prueth->shram.pa + EMAC_ICSSG_SWITCH_DEFAULT_VLAN_TABLE_OFFSET;
+	int i;
 	u8 mac[ETH_ALEN] = { 0 };
 
 	if (prueth->emacs_initialized)
 		return;
 
-	regmap_update_bits(prueth->miig_rt, FDB_GEN_CFG1,
-			   SMEM_VLAN_OFFSET_MASK, 0);
-	regmap_write(prueth->miig_rt, FDB_GEN_CFG2, 0);
+	/* Set VLAN TABLE address base */
+	regmap_update_bits(prueth->miig_rt, FDB_GEN_CFG1, SMEM_VLAN_OFFSET_MASK,
+			   addr <<  SMEM_VLAN_OFFSET);
+	/* Set enable VLAN aware mode, and FDBs for all PRUs */
+	regmap_write(prueth->miig_rt, FDB_GEN_CFG2, (FDB_PRU0_EN | FDB_PRU1_EN | FDB_HOST_EN));
+	prueth->vlan_tbl = (struct prueth_vlan_tbl __force *)(prueth->shram.va +
+			    EMAC_ICSSG_SWITCH_DEFAULT_VLAN_TABLE_OFFSET);
+	for (i = 0; i < SZ_4K - 1; i++) {
+		prueth->vlan_tbl[i].fid = i;
+		prueth->vlan_tbl[i].fid_c1 = 0;
+	}
 	/* Clear host MAC address */
 	icssg_class_set_host_mac_addr(prueth->miig_rt, mac);
 }
 
+static void icssg_init_switch_mode(struct prueth *prueth)
+{
+	u32 addr = prueth->shram.pa + EMAC_ICSSG_SWITCH_DEFAULT_VLAN_TABLE_OFFSET;
+	int i;
+
+	if (prueth->emacs_initialized)
+		return;
+
+	/* Set VLAN TABLE address base */
+	regmap_update_bits(prueth->miig_rt, FDB_GEN_CFG1, SMEM_VLAN_OFFSET_MASK,
+			   addr <<  SMEM_VLAN_OFFSET);
+	/* Set enable VLAN aware mode, and FDBs for all PRUs */
+	regmap_write(prueth->miig_rt, FDB_GEN_CFG2, FDB_EN_ALL);
+	prueth->vlan_tbl = (struct prueth_vlan_tbl __force *)(prueth->shram.va +
+			    EMAC_ICSSG_SWITCH_DEFAULT_VLAN_TABLE_OFFSET);
+	for (i = 0; i < SZ_4K - 1; i++) {
+		prueth->vlan_tbl[i].fid = i;
+		prueth->vlan_tbl[i].fid_c1 = 0;
+	}
+
+	if (prueth->hw_bridge_dev)
+		icssg_class_set_host_mac_addr(prueth->miig_rt, prueth->hw_bridge_dev->dev_addr);
+	icssg_set_pvid(prueth, prueth->default_vlan, PRUETH_PORT_HOST);
+}
+
 int icssg_config(struct prueth *prueth, struct prueth_emac *emac, int slice)
 {
 	void __iomem *config = emac->dram.va + ICSSG_CONFIG_OFFSET;
 	struct icssg_flow_cfg __iomem *flow_cfg;
 	int ret;
 
-	icssg_init_emac_mode(prueth);
+	if (prueth->is_switch_mode)
+		icssg_init_switch_mode(prueth);
+	else
+		icssg_init_emac_mode(prueth);
 
 	memset_io(config, 0, TAS_GATE_MASK_LIST0);
 	icssg_miig_queues_init(prueth, slice);
@@ -353,7 +471,10 @@ int icssg_config(struct prueth *prueth, struct prueth_emac *emac, int slice)
 	regmap_update_bits(prueth->miig_rt, ICSSG_CFG_OFFSET,
 			   ICSSG_CFG_DEFAULT, ICSSG_CFG_DEFAULT);
 	icssg_miig_set_interface_mode(prueth->miig_rt, slice, emac->phy_if);
-	icssg_config_mii_init(emac);
+	if (prueth->is_switch_mode)
+		icssg_config_mii_init_switch(emac);
+	else
+		icssg_config_mii_init(emac);
 	icssg_config_ipg(emac);
 	icssg_update_rgmii_cfg(prueth->miig_rt, emac);
 
@@ -376,7 +497,10 @@ int icssg_config(struct prueth *prueth, struct prueth_emac *emac, int slice)
 	writeb(0, config + SPL_PKT_DEFAULT_PRIORITY);
 	writeb(0, config + QUEUE_NUM_UNTAGGED);
 
-	ret = prueth_emac_buffer_setup(emac);
+	if (prueth->is_switch_mode)
+		ret = prueth_switch_buffer_setup(emac);
+	else
+		ret = prueth_emac_buffer_setup(emac);
 	if (ret)
 		return ret;
 
diff --git a/drivers/net/ethernet/ti/icssg/icssg_config.h b/drivers/net/ethernet/ti/icssg/icssg_config.h
index 4a9721aa6057..1ac60283923b 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_config.h
+++ b/drivers/net/ethernet/ti/icssg/icssg_config.h
@@ -35,6 +35,13 @@ struct icssg_flow_cfg {
 	(2 * (PRUETH_EMAC_BUF_POOL_SIZE * PRUETH_NUM_BUF_POOLS + \
 	 PRUETH_EMAC_RX_CTX_BUF_SIZE * 2))
 
+#define PRUETH_SW_BUF_POOL_SIZE_HOST	SZ_4K
+#define PRUETH_SW_NUM_BUF_POOLS_HOST	8
+#define PRUETH_SW_NUM_BUF_POOLS_PER_PRU 4
+#define MSMC_RAM_SIZE_SWITCH_MODE \
+	(MSMC_RAM_SIZE + \
+	(2 * PRUETH_SW_BUF_POOL_SIZE_HOST * PRUETH_SW_NUM_BUF_POOLS_HOST))
+
 #define PRUETH_SWITCH_FDB_MASK ((SIZE_OF_FDB / NUMBER_OF_FDB_BUCKET_ENTRIES) - 1)
 
 struct icssg_rxq_ctx {
diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
index 1db67a8107cc..6e65aa0977d4 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
@@ -27,6 +27,7 @@
 #include <linux/remoteproc/pruss.h>
 #include <linux/regmap.h>
 #include <linux/remoteproc.h>
+#include <net/switchdev.h>
 
 #include "icssg_prueth.h"
 #include "icssg_mii_rt.h"
@@ -35,6 +36,10 @@
 
 #define PRUETH_MODULE_DESCRIPTION "PRUSS ICSSG Ethernet driver"
 
+#define DEFAULT_VID		1
+#define DEFAULT_PORT_MASK	1
+#define DEFAULT_UNTAG_MASK	1
+
 /* CTRLMMR_ICSSG_RGMII_CTRL register bits */
 #define ICSSG_CTRL_RGMII_ID_MODE                BIT(24)
 
@@ -113,6 +118,19 @@ static irqreturn_t prueth_tx_ts_irq(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
+static struct icssg_firmwares icssg_switch_firmwares[] = {
+	{
+		.pru = "ti-pruss/am65x-sr2-pru0-prusw-fw.elf",
+		.rtu = "ti-pruss/am65x-sr2-rtu0-prusw-fw.elf",
+		.txpru = "ti-pruss/am65x-sr2-txpru0-prusw-fw.elf",
+	},
+	{
+		.pru = "ti-pruss/am65x-sr2-pru1-prusw-fw.elf",
+		.rtu = "ti-pruss/am65x-sr2-rtu1-prusw-fw.elf",
+		.txpru = "ti-pruss/am65x-sr2-txpru1-prusw-fw.elf",
+	}
+};
+
 static struct icssg_firmwares icssg_emac_firmwares[] = {
 	{
 		.pru = "ti-pruss/am65x-sr2-pru0-prueth-fw.elf",
@@ -132,7 +150,10 @@ static int prueth_emac_start(struct prueth *prueth, struct prueth_emac *emac)
 	struct device *dev = prueth->dev;
 	int slice, ret;
 
-	firmwares = icssg_emac_firmwares;
+	if (prueth->is_switch_mode)
+		firmwares = icssg_switch_firmwares;
+	else
+		firmwares = icssg_emac_firmwares;
 
 	slice = prueth_emac_slice(emac);
 	if (slice < 0) {
@@ -446,9 +467,8 @@ static int emac_ndo_open(struct net_device *ndev)
 	ether_addr_copy(emac->mac_addr, ndev->dev_addr);
 
 	icssg_class_set_mac_addr(prueth->miig_rt, slice, emac->mac_addr);
-	icssg_ft1_set_mac_addr(prueth->miig_rt, slice, emac->mac_addr);
-
 	icssg_class_default(prueth->miig_rt, slice, 0, false);
+	icssg_ft1_set_mac_addr(prueth->miig_rt, slice, emac->mac_addr);
 
 	/* Notify the stack of the actual queue counts. */
 	ret = netif_set_real_num_tx_queues(ndev, num_data_chn);
@@ -845,6 +865,203 @@ bool prueth_dev_check(const struct net_device *ndev)
 	return false;
 }
 
+static void prueth_offload_fwd_mark_update(struct prueth *prueth)
+{
+	int set_val = 0;
+	int i;
+
+	if (prueth->br_members == (BIT(PRUETH_PORT_MII0) | BIT(PRUETH_PORT_MII1)))
+		set_val = 1;
+
+	dev_dbg(prueth->dev, "set offload_fwd_mark %d\n", set_val);
+
+	for (i = PRUETH_MAC0; i < PRUETH_NUM_MACS; i++) {
+		struct prueth_emac *emac = prueth->emac[i];
+
+		if (!emac || !emac->ndev)
+			continue;
+
+		emac->offload_fwd_mark = set_val;
+	}
+}
+
+static void prueth_emac_restart(struct prueth *prueth)
+{
+	struct prueth_emac *emac0 = prueth->emac[PRUETH_MAC0];
+	struct prueth_emac *emac1 = prueth->emac[PRUETH_MAC1];
+
+	/* Detach the net_device for both PRUeth ports*/
+	if (netif_running(emac0->ndev))
+		netif_device_detach(emac0->ndev);
+	if (netif_running(emac1->ndev))
+		netif_device_detach(emac1->ndev);
+
+	/* Disable both PRUeth ports */
+	emac_set_port_state(emac0, ICSSG_EMAC_PORT_DISABLE);
+	emac_set_port_state(emac1, ICSSG_EMAC_PORT_DISABLE);
+
+	/* Stop both pru cores for both PRUeth ports*/
+	prueth_emac_stop(emac0);
+	prueth->emacs_initialized--;
+	prueth_emac_stop(emac1);
+	prueth->emacs_initialized--;
+
+	/* Start both pru cores for both PRUeth ports */
+	prueth_emac_start(prueth, emac0);
+	prueth->emacs_initialized++;
+	prueth_emac_start(prueth, emac1);
+	prueth->emacs_initialized++;
+
+	/* Enable forwarding for both PRUeth ports */
+	emac_set_port_state(emac0, ICSSG_EMAC_PORT_FORWARD);
+	emac_set_port_state(emac1, ICSSG_EMAC_PORT_FORWARD);
+
+	/* Attache net_device for both PRUeth ports */
+	netif_device_attach(emac0->ndev);
+	netif_device_attach(emac1->ndev);
+}
+
+static void icssg_enable_switch_mode(struct prueth *prueth)
+{
+	struct prueth_emac *emac;
+	int mac;
+
+	prueth_emac_restart(prueth);
+
+	for (mac = PRUETH_MAC0; mac < PRUETH_NUM_MACS; mac++) {
+		emac = prueth->emac[mac];
+		if (netif_running(emac->ndev)) {
+			icssg_fdb_add_del(emac, eth_stp_addr, prueth->default_vlan,
+					  ICSSG_FDB_ENTRY_P0_MEMBERSHIP |
+					  ICSSG_FDB_ENTRY_P1_MEMBERSHIP |
+					  ICSSG_FDB_ENTRY_P2_MEMBERSHIP |
+					  ICSSG_FDB_ENTRY_BLOCK,
+					  true);
+			icssg_vtbl_modify(emac, emac->port_vlan | DEFAULT_VID,
+					  BIT(emac->port_id) | DEFAULT_PORT_MASK,
+					  BIT(emac->port_id) | DEFAULT_UNTAG_MASK,
+					  true);
+			icssg_set_pvid(prueth, emac->port_vlan, emac->port_id);
+			emac_set_port_state(emac, ICSSG_EMAC_PORT_VLAN_AWARE_ENABLE);
+		}
+	}
+}
+
+static int prueth_netdevice_port_link(struct net_device *ndev,
+				      struct net_device *br_ndev,
+				      struct netlink_ext_ack *extack)
+{
+	struct prueth_emac *emac = netdev_priv(ndev);
+	struct prueth *prueth = emac->prueth;
+	int err;
+
+	if (!prueth->br_members) {
+		prueth->hw_bridge_dev = br_ndev;
+	} else {
+		/* This is adding the port to a second bridge, this is
+		 * unsupported
+		 */
+		if (prueth->hw_bridge_dev != br_ndev)
+			return -EOPNOTSUPP;
+	}
+
+	err = switchdev_bridge_port_offload(br_ndev, ndev, emac,
+					    &prueth->prueth_switchdev_nb,
+					    &prueth->prueth_switchdev_bl_nb,
+					    false, extack);
+	if (err)
+		return err;
+
+	prueth->br_members |= BIT(emac->port_id);
+
+	if (!prueth->is_switch_mode) {
+		if (prueth->br_members & BIT(PRUETH_PORT_MII0) &&
+		    prueth->br_members & BIT(PRUETH_PORT_MII1)) {
+			prueth->is_switch_mode = true;
+			prueth->default_vlan = 1;
+			emac->port_vlan = prueth->default_vlan;
+			icssg_enable_switch_mode(prueth);
+		}
+	}
+
+	prueth_offload_fwd_mark_update(prueth);
+
+	return NOTIFY_DONE;
+}
+
+static void prueth_netdevice_port_unlink(struct net_device *ndev)
+{
+	struct prueth_emac *emac = netdev_priv(ndev);
+	struct prueth *prueth = emac->prueth;
+
+	prueth->br_members &= ~BIT(emac->port_id);
+
+	if (prueth->is_switch_mode) {
+		prueth->is_switch_mode = false;
+		emac->port_vlan = 0;
+		prueth_emac_restart(prueth);
+	}
+
+	prueth_offload_fwd_mark_update(prueth);
+
+	if (!prueth->br_members)
+		prueth->hw_bridge_dev = NULL;
+}
+
+/* netdev notifier */
+static int prueth_netdevice_event(struct notifier_block *unused,
+				  unsigned long event, void *ptr)
+{
+	struct netlink_ext_ack *extack = netdev_notifier_info_to_extack(ptr);
+	struct net_device *ndev = netdev_notifier_info_to_dev(ptr);
+	struct netdev_notifier_changeupper_info *info;
+	int ret = NOTIFY_DONE;
+
+	if (ndev->netdev_ops != &emac_netdev_ops)
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case NETDEV_CHANGEUPPER:
+		info = ptr;
+
+		if (netif_is_bridge_master(info->upper_dev)) {
+			if (info->linking)
+				ret = prueth_netdevice_port_link(ndev, info->upper_dev, extack);
+			else
+				prueth_netdevice_port_unlink(ndev);
+		}
+		break;
+	default:
+		return NOTIFY_DONE;
+	}
+
+	return notifier_from_errno(ret);
+}
+
+static int prueth_register_notifiers(struct prueth *prueth)
+{
+	int ret = 0;
+
+	prueth->prueth_netdevice_nb.notifier_call = &prueth_netdevice_event;
+	ret = register_netdevice_notifier(&prueth->prueth_netdevice_nb);
+	if (ret) {
+		dev_err(prueth->dev, "can't register netdevice notifier\n");
+		return ret;
+	}
+
+	ret = prueth_switchdev_register_notifiers(prueth);
+	if (ret)
+		unregister_netdevice_notifier(&prueth->prueth_netdevice_nb);
+
+	return ret;
+}
+
+static void prueth_unregister_notifiers(struct prueth *prueth)
+{
+	prueth_switchdev_unregister_notifiers(prueth);
+	unregister_netdevice_notifier(&prueth->prueth_netdevice_nb);
+}
+
 static int prueth_probe(struct platform_device *pdev)
 {
 	struct device_node *eth_node, *eth_ports_node;
@@ -972,6 +1189,9 @@ static int prueth_probe(struct platform_device *pdev)
 	}
 
 	msmc_ram_size = MSMC_RAM_SIZE;
+	prueth->is_switchmode_supported = prueth->pdata.switch_mode;
+	if (prueth->is_switchmode_supported)
+		msmc_ram_size = MSMC_RAM_SIZE_SWITCH_MODE;
 
 	/* NOTE: FW bug needs buffer base to be 64KB aligned */
 	prueth->msmcram.va =
@@ -1077,6 +1297,14 @@ static int prueth_probe(struct platform_device *pdev)
 		phy_attached_info(prueth->emac[PRUETH_MAC1]->ndev->phydev);
 	}
 
+	if (prueth->is_switchmode_supported) {
+		ret = prueth_register_notifiers(prueth);
+		if (ret)
+			goto netdev_unregister;
+
+		sprintf(prueth->switch_id, "%s", dev_name(dev));
+	}
+
 	dev_info(dev, "TI PRU ethernet driver initialized: %s EMAC mode\n",
 		 (!eth0_node || !eth1_node) ? "single" : "dual");
 
@@ -1146,6 +1374,8 @@ static void prueth_remove(struct platform_device *pdev)
 	struct device_node *eth_node;
 	int i;
 
+	prueth_unregister_notifiers(prueth);
+
 	for (i = 0; i < PRUETH_NUM_MACS; i++) {
 		if (!prueth->registered_netdevs[i])
 			continue;
@@ -1187,10 +1417,12 @@ static void prueth_remove(struct platform_device *pdev)
 static const struct prueth_pdata am654_icssg_pdata = {
 	.fdqring_mode = K3_RINGACC_RING_MODE_MESSAGE,
 	.quirk_10m_link_issue = 1,
+	.switch_mode = 1,
 };
 
 static const struct prueth_pdata am64x_icssg_pdata = {
 	.fdqring_mode = K3_RINGACC_RING_MODE_RING,
+	.switch_mode = 1,
 };
 
 static const struct of_device_id prueth_dt_match[] = {
-- 
cgit v1.2.3-58-ga151


From 3d60041543189438cd1b03a1fa40ff6681c77970 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Sun, 26 May 2024 20:42:26 +0800
Subject: wifi: ath12k: fix memory leak in ath12k_dp_rx_peer_frag_setup()

Currently the resource allocated by crypto_alloc_shash() is not
freed in case ath12k_peer_find() fails, resulting in memory leak.

Add crypto_free_shash() to fix it.

This is found during code review, compile tested only.

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240526124226.24661-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_rx.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index a52b7b273540..cb1f308f096b 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -2742,6 +2742,7 @@ int ath12k_dp_rx_peer_frag_setup(struct ath12k *ar, const u8 *peer_mac, int vdev
 	peer = ath12k_peer_find(ab, vdev_id, peer_mac);
 	if (!peer) {
 		spin_unlock_bh(&ab->base_lock);
+		crypto_free_shash(tfm);
 		ath12k_warn(ab, "failed to find the peer to set up fragment info\n");
 		return -ENOENT;
 	}
-- 
cgit v1.2.3-58-ga151


From 43e934360d1d8fd3230d0c1ffa40f0d49deb4cf2 Mon Sep 17 00:00:00 2001
From: "Rob Herring (Arm)" <robh@kernel.org>
Date: Tue, 28 May 2024 08:46:09 -0500
Subject: dt-bindings: net: wireless: ath11k: Drop "qcom,ipq8074-wcss-pil" from
 example

Convention for examples is to only show what's covered by the binding,
so drop the provider "qcom,ipq8074-wcss-pil". It is also not documented
by a schema which caused a warning.

Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240528134610.4075204-1-robh@kernel.org
---
 Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml
index a2d55bf4c7a5..ff5763dc66a8 100644
--- a/Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml
+++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k.yaml
@@ -265,15 +265,6 @@ allOf:
 
 examples:
   - |
-
-    q6v5_wcss: remoteproc@cd00000 {
-        compatible = "qcom,ipq8074-wcss-pil";
-        reg = <0xcd00000 0x4040>,
-              <0x4ab000 0x20>;
-        reg-names = "qdsp6",
-                    "rmb";
-    };
-
     wifi0: wifi@c000000 {
         compatible = "qcom,ipq8074-wifi";
         reg = <0xc000000 0x2000000>;
-- 
cgit v1.2.3-58-ga151


From 8233d2716570b66ab45b6d24e577ee6612002dc4 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Wed, 29 May 2024 09:53:57 +0800
Subject: wifi: ath12k: do not process consecutive RDDM event

Currently we do reset for each RDDM event from MHI, however there are
cases, see below log, that we get two or more consecutive events, and
it is pointless to do reset for the subsequent ones. What's more, it
makes reset process more likely to fail.

[  103.289864] mhi mhi0: System error detected
[  103.289871] ath12k_pci 0000:03:00.0: mhi notify status reason MHI_CB_EE_RDDM
[  103.293144] mhi mhi0: System error detected
[  103.293150] ath12k_pci 0000:03:00.0: mhi notify status reason MHI_CB_EE_RDDM

Add a check to avoid reset again and again. This is done by tracking previous
MHI status: if we receive a new RDDM event while the previous event is
also the same, we treat it as duplicate and ignore it, because normally
we should at least receive a MHI_CB_EE_MISSION_MODE event between them.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240529015357.9846-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mhi.c | 11 +++++++++++
 drivers/net/wireless/ath/ath12k/pci.h |  1 +
 2 files changed, 12 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/mhi.c b/drivers/net/wireless/ath/ath12k/mhi.c
index fef2f7622033..df96b0f91f54 100644
--- a/drivers/net/wireless/ath/ath12k/mhi.c
+++ b/drivers/net/wireless/ath/ath12k/mhi.c
@@ -16,6 +16,7 @@
 #define MHI_TIMEOUT_DEFAULT_MS	90000
 #define OTP_INVALID_BOARD_ID	0xFFFF
 #define OTP_VALID_DUALMAC_BOARD_ID_MASK		0x1000
+#define MHI_CB_INVALID	0xff
 
 static const struct mhi_channel_config ath12k_mhi_channels_qcn9274[] = {
 	{
@@ -268,6 +269,7 @@ static void ath12k_mhi_op_status_cb(struct mhi_controller *mhi_cntrl,
 				    enum mhi_callback cb)
 {
 	struct ath12k_base *ab = dev_get_drvdata(mhi_cntrl->cntrl_dev);
+	struct ath12k_pci *ab_pci = ath12k_pci_priv(ab);
 
 	ath12k_dbg(ab, ATH12K_DBG_BOOT, "mhi notify status reason %s\n",
 		   ath12k_mhi_op_callback_to_str(cb));
@@ -277,12 +279,20 @@ static void ath12k_mhi_op_status_cb(struct mhi_controller *mhi_cntrl,
 		ath12k_warn(ab, "firmware crashed: MHI_CB_SYS_ERROR\n");
 		break;
 	case MHI_CB_EE_RDDM:
+		if (ab_pci->mhi_pre_cb == MHI_CB_EE_RDDM) {
+			ath12k_dbg(ab, ATH12K_DBG_BOOT,
+				   "do not queue again for consecutive RDDM event\n");
+			break;
+		}
+
 		if (!(test_bit(ATH12K_FLAG_UNREGISTERING, &ab->dev_flags)))
 			queue_work(ab->workqueue_aux, &ab->reset_work);
 		break;
 	default:
 		break;
 	}
+
+	ab_pci->mhi_pre_cb = cb;
 }
 
 static int ath12k_mhi_op_read_reg(struct mhi_controller *mhi_cntrl,
@@ -313,6 +323,7 @@ int ath12k_mhi_register(struct ath12k_pci *ab_pci)
 	if (!mhi_ctrl)
 		return -ENOMEM;
 
+	ab_pci->mhi_pre_cb = MHI_CB_INVALID;
 	ab_pci->mhi_ctrl = mhi_ctrl;
 	mhi_ctrl->cntrl_dev = ab->dev;
 	mhi_ctrl->regs = ab->mem;
diff --git a/drivers/net/wireless/ath/ath12k/pci.h b/drivers/net/wireless/ath/ath12k/pci.h
index 6186a78038cf..31584a7ad80e 100644
--- a/drivers/net/wireless/ath/ath12k/pci.h
+++ b/drivers/net/wireless/ath/ath12k/pci.h
@@ -104,6 +104,7 @@ struct ath12k_pci {
 	struct mhi_controller *mhi_ctrl;
 	const struct ath12k_msi_config *msi_config;
 	unsigned long mhi_state;
+	enum mhi_callback mhi_pre_cb;
 	u32 register_window;
 
 	/* protects register_window above */
-- 
cgit v1.2.3-58-ga151


From 46253c4ae96162a840ad65c1394de63796d7798a Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Wed, 22 May 2024 10:09:36 +0200
Subject: selftests/bpf: use section names understood by libbpf in test_sockmap

libbpf can deduce program type and attach type from the ELF section name.
We don't need to pass it out-of-band if we switch to libbpf convention [1].

[1] https://docs.kernel.org/bpf/libbpf/program_types.html

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240522080936.2475833-1-jakub@cloudflare.com
---
 .../selftests/bpf/progs/test_sockmap_kern.h        | 17 ++++++------
 tools/testing/selftests/bpf/test_sockmap.c         | 31 ----------------------
 2 files changed, 9 insertions(+), 39 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/test_sockmap_kern.h b/tools/testing/selftests/bpf/progs/test_sockmap_kern.h
index 99d2ea9fb658..3dff0813730b 100644
--- a/tools/testing/selftests/bpf/progs/test_sockmap_kern.h
+++ b/tools/testing/selftests/bpf/progs/test_sockmap_kern.h
@@ -92,7 +92,7 @@ struct {
 	__uint(value_size, sizeof(int));
 } tls_sock_map SEC(".maps");
 
-SEC("sk_skb1")
+SEC("sk_skb/stream_parser")
 int bpf_prog1(struct __sk_buff *skb)
 {
 	int *f, two = 2;
@@ -104,7 +104,7 @@ int bpf_prog1(struct __sk_buff *skb)
 	return skb->len;
 }
 
-SEC("sk_skb2")
+SEC("sk_skb/stream_verdict")
 int bpf_prog2(struct __sk_buff *skb)
 {
 	__u32 lport = skb->local_port;
@@ -151,7 +151,7 @@ static inline void bpf_write_pass(struct __sk_buff *skb, int offset)
 		memcpy(c + offset, "PASS", 4);
 }
 
-SEC("sk_skb3")
+SEC("sk_skb/stream_verdict")
 int bpf_prog3(struct __sk_buff *skb)
 {
 	int err, *f, ret = SK_PASS;
@@ -233,7 +233,7 @@ int bpf_sockmap(struct bpf_sock_ops *skops)
 	return 0;
 }
 
-SEC("sk_msg1")
+SEC("sk_msg")
 int bpf_prog4(struct sk_msg_md *msg)
 {
 	int *bytes, zero = 0, one = 1, two = 2, three = 3, four = 4, five = 5;
@@ -263,7 +263,7 @@ int bpf_prog4(struct sk_msg_md *msg)
 	return SK_PASS;
 }
 
-SEC("sk_msg2")
+SEC("sk_msg")
 int bpf_prog6(struct sk_msg_md *msg)
 {
 	int zero = 0, one = 1, two = 2, three = 3, four = 4, five = 5, key = 0;
@@ -308,7 +308,7 @@ int bpf_prog6(struct sk_msg_md *msg)
 #endif
 }
 
-SEC("sk_msg3")
+SEC("sk_msg")
 int bpf_prog8(struct sk_msg_md *msg)
 {
 	void *data_end = (void *)(long) msg->data_end;
@@ -329,7 +329,8 @@ int bpf_prog8(struct sk_msg_md *msg)
 
 	return SK_PASS;
 }
-SEC("sk_msg4")
+
+SEC("sk_msg")
 int bpf_prog9(struct sk_msg_md *msg)
 {
 	void *data_end = (void *)(long) msg->data_end;
@@ -347,7 +348,7 @@ int bpf_prog9(struct sk_msg_md *msg)
 	return SK_PASS;
 }
 
-SEC("sk_msg5")
+SEC("sk_msg")
 int bpf_prog10(struct sk_msg_md *msg)
 {
 	int *bytes, *start, *end, *start_push, *end_push, *start_pop, *pop;
diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index 4499b3cfc3a6..ddc6a9cef36f 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -1783,30 +1783,6 @@ char *map_names[] = {
 	"tls_sock_map",
 };
 
-int prog_attach_type[] = {
-	BPF_SK_SKB_STREAM_PARSER,
-	BPF_SK_SKB_STREAM_VERDICT,
-	BPF_SK_SKB_STREAM_VERDICT,
-	BPF_CGROUP_SOCK_OPS,
-	BPF_SK_MSG_VERDICT,
-	BPF_SK_MSG_VERDICT,
-	BPF_SK_MSG_VERDICT,
-	BPF_SK_MSG_VERDICT,
-	BPF_SK_MSG_VERDICT,
-};
-
-int prog_type[] = {
-	BPF_PROG_TYPE_SK_SKB,
-	BPF_PROG_TYPE_SK_SKB,
-	BPF_PROG_TYPE_SK_SKB,
-	BPF_PROG_TYPE_SOCK_OPS,
-	BPF_PROG_TYPE_SK_MSG,
-	BPF_PROG_TYPE_SK_MSG,
-	BPF_PROG_TYPE_SK_MSG,
-	BPF_PROG_TYPE_SK_MSG,
-	BPF_PROG_TYPE_SK_MSG,
-};
-
 static int populate_progs(char *bpf_file)
 {
 	struct bpf_program *prog;
@@ -1825,13 +1801,6 @@ static int populate_progs(char *bpf_file)
 		return -1;
 	}
 
-	bpf_object__for_each_program(prog, obj) {
-		bpf_program__set_type(prog, prog_type[i]);
-		bpf_program__set_expected_attach_type(prog,
-						      prog_attach_type[i]);
-		i++;
-	}
-
 	i = bpf_object__load(obj);
 	i = 0;
 	bpf_object__for_each_program(prog, obj) {
-- 
cgit v1.2.3-58-ga151


From 73287fe228721b05690e671adbcccc6cf5435be6 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Wed, 29 May 2024 23:59:39 -0700
Subject: bpf: pass bpf_struct_ops_link to callbacks in bpf_struct_ops.

Pass an additional pointer of bpf_struct_ops_link to callback function reg,
unreg, and update provided by subsystems defined in bpf_struct_ops. A
bpf_struct_ops_map can be registered for multiple links. Passing a pointer
of bpf_struct_ops_link helps subsystems to distinguish them.

This pointer will be used in the later patches to let the subsystem
initiate a detachment on a link that was registered to it previously.

Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240530065946.979330-2-thinker.li@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/bpf.h                                           |  6 +++---
 kernel/bpf/bpf_struct_ops.c                                   | 10 +++++-----
 net/bpf/bpf_dummy_struct_ops.c                                |  4 ++--
 net/ipv4/bpf_tcp_ca.c                                         |  6 +++---
 tools/testing/selftests/bpf/bpf_test_no_cfi/bpf_test_no_cfi.c |  4 ++--
 tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c         |  6 +++---
 6 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 5e694a308081..19f8836382fc 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1730,9 +1730,9 @@ struct bpf_struct_ops {
 	int (*init_member)(const struct btf_type *t,
 			   const struct btf_member *member,
 			   void *kdata, const void *udata);
-	int (*reg)(void *kdata);
-	void (*unreg)(void *kdata);
-	int (*update)(void *kdata, void *old_kdata);
+	int (*reg)(void *kdata, struct bpf_link *link);
+	void (*unreg)(void *kdata, struct bpf_link *link);
+	int (*update)(void *kdata, void *old_kdata, struct bpf_link *link);
 	int (*validate)(void *kdata);
 	void *cfi_stubs;
 	struct module *owner;
diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index 86c7884abaf8..1542dded7489 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -757,7 +757,7 @@ static long bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 		goto unlock;
 	}
 
-	err = st_ops->reg(kdata);
+	err = st_ops->reg(kdata, NULL);
 	if (likely(!err)) {
 		/* This refcnt increment on the map here after
 		 * 'st_ops->reg()' is secure since the state of the
@@ -805,7 +805,7 @@ static long bpf_struct_ops_map_delete_elem(struct bpf_map *map, void *key)
 			     BPF_STRUCT_OPS_STATE_TOBEFREE);
 	switch (prev_state) {
 	case BPF_STRUCT_OPS_STATE_INUSE:
-		st_map->st_ops_desc->st_ops->unreg(&st_map->kvalue.data);
+		st_map->st_ops_desc->st_ops->unreg(&st_map->kvalue.data, NULL);
 		bpf_map_put(map);
 		return 0;
 	case BPF_STRUCT_OPS_STATE_TOBEFREE:
@@ -1060,7 +1060,7 @@ static void bpf_struct_ops_map_link_dealloc(struct bpf_link *link)
 		/* st_link->map can be NULL if
 		 * bpf_struct_ops_link_create() fails to register.
 		 */
-		st_map->st_ops_desc->st_ops->unreg(&st_map->kvalue.data);
+		st_map->st_ops_desc->st_ops->unreg(&st_map->kvalue.data, link);
 		bpf_map_put(&st_map->map);
 	}
 	kfree(st_link);
@@ -1125,7 +1125,7 @@ static int bpf_struct_ops_map_link_update(struct bpf_link *link, struct bpf_map
 		goto err_out;
 	}
 
-	err = st_map->st_ops_desc->st_ops->update(st_map->kvalue.data, old_st_map->kvalue.data);
+	err = st_map->st_ops_desc->st_ops->update(st_map->kvalue.data, old_st_map->kvalue.data, link);
 	if (err)
 		goto err_out;
 
@@ -1176,7 +1176,7 @@ int bpf_struct_ops_link_create(union bpf_attr *attr)
 	if (err)
 		goto err_out;
 
-	err = st_map->st_ops_desc->st_ops->reg(st_map->kvalue.data);
+	err = st_map->st_ops_desc->st_ops->reg(st_map->kvalue.data, &link->link);
 	if (err) {
 		bpf_link_cleanup(&link_primer);
 		link = NULL;
diff --git a/net/bpf/bpf_dummy_struct_ops.c b/net/bpf/bpf_dummy_struct_ops.c
index 891cdf61c65a..3ea52b05adfb 100644
--- a/net/bpf/bpf_dummy_struct_ops.c
+++ b/net/bpf/bpf_dummy_struct_ops.c
@@ -272,12 +272,12 @@ static int bpf_dummy_init_member(const struct btf_type *t,
 	return -EOPNOTSUPP;
 }
 
-static int bpf_dummy_reg(void *kdata)
+static int bpf_dummy_reg(void *kdata, struct bpf_link *link)
 {
 	return -EOPNOTSUPP;
 }
 
-static void bpf_dummy_unreg(void *kdata)
+static void bpf_dummy_unreg(void *kdata, struct bpf_link *link)
 {
 }
 
diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c
index 18227757ec0c..3f88d0961e5b 100644
--- a/net/ipv4/bpf_tcp_ca.c
+++ b/net/ipv4/bpf_tcp_ca.c
@@ -260,17 +260,17 @@ static int bpf_tcp_ca_check_member(const struct btf_type *t,
 	return 0;
 }
 
-static int bpf_tcp_ca_reg(void *kdata)
+static int bpf_tcp_ca_reg(void *kdata, struct bpf_link *link)
 {
 	return tcp_register_congestion_control(kdata);
 }
 
-static void bpf_tcp_ca_unreg(void *kdata)
+static void bpf_tcp_ca_unreg(void *kdata, struct bpf_link *link)
 {
 	tcp_unregister_congestion_control(kdata);
 }
 
-static int bpf_tcp_ca_update(void *kdata, void *old_kdata)
+static int bpf_tcp_ca_update(void *kdata, void *old_kdata, struct bpf_link *link)
 {
 	return tcp_update_congestion_control(kdata, old_kdata);
 }
diff --git a/tools/testing/selftests/bpf/bpf_test_no_cfi/bpf_test_no_cfi.c b/tools/testing/selftests/bpf/bpf_test_no_cfi/bpf_test_no_cfi.c
index b1dd889d5d7d..948eb3962732 100644
--- a/tools/testing/selftests/bpf/bpf_test_no_cfi/bpf_test_no_cfi.c
+++ b/tools/testing/selftests/bpf/bpf_test_no_cfi/bpf_test_no_cfi.c
@@ -22,12 +22,12 @@ static int dummy_init_member(const struct btf_type *t,
 	return 0;
 }
 
-static int dummy_reg(void *kdata)
+static int dummy_reg(void *kdata, struct bpf_link *link)
 {
 	return 0;
 }
 
-static void dummy_unreg(void *kdata)
+static void dummy_unreg(void *kdata, struct bpf_link *link)
 {
 }
 
diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index 2a18bd320e92..0a09732cde4b 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -820,7 +820,7 @@ static const struct bpf_verifier_ops bpf_testmod_verifier_ops = {
 	.is_valid_access = bpf_testmod_ops_is_valid_access,
 };
 
-static int bpf_dummy_reg(void *kdata)
+static int bpf_dummy_reg(void *kdata, struct bpf_link *link)
 {
 	struct bpf_testmod_ops *ops = kdata;
 
@@ -835,7 +835,7 @@ static int bpf_dummy_reg(void *kdata)
 	return 0;
 }
 
-static void bpf_dummy_unreg(void *kdata)
+static void bpf_dummy_unreg(void *kdata, struct bpf_link *link)
 {
 }
 
@@ -871,7 +871,7 @@ struct bpf_struct_ops bpf_bpf_testmod_ops = {
 	.owner = THIS_MODULE,
 };
 
-static int bpf_dummy_reg2(void *kdata)
+static int bpf_dummy_reg2(void *kdata, struct bpf_link *link)
 {
 	struct bpf_testmod_ops2 *ops = kdata;
 
-- 
cgit v1.2.3-58-ga151


From 6fb2544ea1493f52e50b753604791c01bd2cf897 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Wed, 29 May 2024 23:59:40 -0700
Subject: bpf: enable detaching links of struct_ops objects.

Implement the detach callback in bpf_link_ops for struct_ops so that user
programs can detach a struct_ops link. The subsystems that struct_ops
objects are registered to can also use this callback to detach the links
being passed to them.

Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240530065946.979330-3-thinker.li@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 kernel/bpf/bpf_struct_ops.c | 48 ++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 43 insertions(+), 5 deletions(-)

diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index 1542dded7489..58314b1fc39c 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -1057,9 +1057,6 @@ static void bpf_struct_ops_map_link_dealloc(struct bpf_link *link)
 	st_map = (struct bpf_struct_ops_map *)
 		rcu_dereference_protected(st_link->map, true);
 	if (st_map) {
-		/* st_link->map can be NULL if
-		 * bpf_struct_ops_link_create() fails to register.
-		 */
 		st_map->st_ops_desc->st_ops->unreg(&st_map->kvalue.data, link);
 		bpf_map_put(&st_map->map);
 	}
@@ -1075,7 +1072,8 @@ static void bpf_struct_ops_map_link_show_fdinfo(const struct bpf_link *link,
 	st_link = container_of(link, struct bpf_struct_ops_link, link);
 	rcu_read_lock();
 	map = rcu_dereference(st_link->map);
-	seq_printf(seq, "map_id:\t%d\n", map->id);
+	if (map)
+		seq_printf(seq, "map_id:\t%d\n", map->id);
 	rcu_read_unlock();
 }
 
@@ -1088,7 +1086,8 @@ static int bpf_struct_ops_map_link_fill_link_info(const struct bpf_link *link,
 	st_link = container_of(link, struct bpf_struct_ops_link, link);
 	rcu_read_lock();
 	map = rcu_dereference(st_link->map);
-	info->struct_ops.map_id = map->id;
+	if (map)
+		info->struct_ops.map_id = map->id;
 	rcu_read_unlock();
 	return 0;
 }
@@ -1113,6 +1112,10 @@ static int bpf_struct_ops_map_link_update(struct bpf_link *link, struct bpf_map
 	mutex_lock(&update_mutex);
 
 	old_map = rcu_dereference_protected(st_link->map, lockdep_is_held(&update_mutex));
+	if (!old_map) {
+		err = -ENOLINK;
+		goto err_out;
+	}
 	if (expected_old_map && old_map != expected_old_map) {
 		err = -EPERM;
 		goto err_out;
@@ -1139,8 +1142,37 @@ err_out:
 	return err;
 }
 
+static int bpf_struct_ops_map_link_detach(struct bpf_link *link)
+{
+	struct bpf_struct_ops_link *st_link = container_of(link, struct bpf_struct_ops_link, link);
+	struct bpf_struct_ops_map *st_map;
+	struct bpf_map *map;
+
+	mutex_lock(&update_mutex);
+
+	map = rcu_dereference_protected(st_link->map, lockdep_is_held(&update_mutex));
+	if (!map) {
+		mutex_unlock(&update_mutex);
+		return 0;
+	}
+	st_map = container_of(map, struct bpf_struct_ops_map, map);
+
+	st_map->st_ops_desc->st_ops->unreg(&st_map->kvalue.data, link);
+
+	RCU_INIT_POINTER(st_link->map, NULL);
+	/* Pair with bpf_map_get() in bpf_struct_ops_link_create() or
+	 * bpf_map_inc() in bpf_struct_ops_map_link_update().
+	 */
+	bpf_map_put(&st_map->map);
+
+	mutex_unlock(&update_mutex);
+
+	return 0;
+}
+
 static const struct bpf_link_ops bpf_struct_ops_map_lops = {
 	.dealloc = bpf_struct_ops_map_link_dealloc,
+	.detach = bpf_struct_ops_map_link_detach,
 	.show_fdinfo = bpf_struct_ops_map_link_show_fdinfo,
 	.fill_link_info = bpf_struct_ops_map_link_fill_link_info,
 	.update_map = bpf_struct_ops_map_link_update,
@@ -1176,13 +1208,19 @@ int bpf_struct_ops_link_create(union bpf_attr *attr)
 	if (err)
 		goto err_out;
 
+	/* Hold the update_mutex such that the subsystem cannot
+	 * do link->ops->detach() before the link is fully initialized.
+	 */
+	mutex_lock(&update_mutex);
 	err = st_map->st_ops_desc->st_ops->reg(st_map->kvalue.data, &link->link);
 	if (err) {
+		mutex_unlock(&update_mutex);
 		bpf_link_cleanup(&link_primer);
 		link = NULL;
 		goto err_out;
 	}
 	RCU_INIT_POINTER(link->map, map);
+	mutex_unlock(&update_mutex);
 
 	return bpf_link_settle(&link_primer);
 
-- 
cgit v1.2.3-58-ga151


From 1adddc97aa44c8783f9f0276ea70854d56f9f6df Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Wed, 29 May 2024 23:59:41 -0700
Subject: bpf: support epoll from bpf struct_ops links.

Add epoll support to bpf struct_ops links to trigger EPOLLHUP event upon
detachment.

This patch implements the "poll" of the "struct file_operations" for BPF
links and introduces a new "poll" operator in the "struct bpf_link_ops". By
implementing "poll" of "struct bpf_link_ops" for the links of struct_ops,
the file descriptor of a struct_ops link can be added to an epoll file
descriptor to receive EPOLLHUP events.

Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240530065946.979330-4-thinker.li@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/bpf.h         |  1 +
 kernel/bpf/bpf_struct_ops.c | 17 +++++++++++++++++
 kernel/bpf/syscall.c        | 31 ++++++++++++++++++++++++++-----
 3 files changed, 44 insertions(+), 5 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 19f8836382fc..5eb61120e4f5 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1612,6 +1612,7 @@ struct bpf_link_ops {
 			      struct bpf_link_info *info);
 	int (*update_map)(struct bpf_link *link, struct bpf_map *new_map,
 			  struct bpf_map *old_map);
+	__poll_t (*poll)(struct file *file, struct poll_table_struct *pts);
 };
 
 struct bpf_tramp_link {
diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index 58314b1fc39c..a2cf31b14be4 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -12,6 +12,7 @@
 #include <linux/mutex.h>
 #include <linux/btf_ids.h>
 #include <linux/rcupdate_wait.h>
+#include <linux/poll.h>
 
 struct bpf_struct_ops_value {
 	struct bpf_struct_ops_common_value common;
@@ -56,6 +57,7 @@ struct bpf_struct_ops_map {
 struct bpf_struct_ops_link {
 	struct bpf_link link;
 	struct bpf_map __rcu *map;
+	wait_queue_head_t wait_hup;
 };
 
 static DEFINE_MUTEX(update_mutex);
@@ -1167,15 +1169,28 @@ static int bpf_struct_ops_map_link_detach(struct bpf_link *link)
 
 	mutex_unlock(&update_mutex);
 
+	wake_up_interruptible_poll(&st_link->wait_hup, EPOLLHUP);
+
 	return 0;
 }
 
+static __poll_t bpf_struct_ops_map_link_poll(struct file *file,
+					     struct poll_table_struct *pts)
+{
+	struct bpf_struct_ops_link *st_link = file->private_data;
+
+	poll_wait(file, &st_link->wait_hup, pts);
+
+	return rcu_access_pointer(st_link->map) ? 0 : EPOLLHUP;
+}
+
 static const struct bpf_link_ops bpf_struct_ops_map_lops = {
 	.dealloc = bpf_struct_ops_map_link_dealloc,
 	.detach = bpf_struct_ops_map_link_detach,
 	.show_fdinfo = bpf_struct_ops_map_link_show_fdinfo,
 	.fill_link_info = bpf_struct_ops_map_link_fill_link_info,
 	.update_map = bpf_struct_ops_map_link_update,
+	.poll = bpf_struct_ops_map_link_poll,
 };
 
 int bpf_struct_ops_link_create(union bpf_attr *attr)
@@ -1208,6 +1223,8 @@ int bpf_struct_ops_link_create(union bpf_attr *attr)
 	if (err)
 		goto err_out;
 
+	init_waitqueue_head(&link->wait_hup);
+
 	/* Hold the update_mutex such that the subsystem cannot
 	 * do link->ops->detach() before the link is fully initialized.
 	 */
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 2222c3ff88e7..81efa1944942 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -3150,6 +3150,13 @@ static void bpf_link_show_fdinfo(struct seq_file *m, struct file *filp)
 }
 #endif
 
+static __poll_t bpf_link_poll(struct file *file, struct poll_table_struct *pts)
+{
+	struct bpf_link *link = file->private_data;
+
+	return link->ops->poll(file, pts);
+}
+
 static const struct file_operations bpf_link_fops = {
 #ifdef CONFIG_PROC_FS
 	.show_fdinfo	= bpf_link_show_fdinfo,
@@ -3159,6 +3166,16 @@ static const struct file_operations bpf_link_fops = {
 	.write		= bpf_dummy_write,
 };
 
+static const struct file_operations bpf_link_fops_poll = {
+#ifdef CONFIG_PROC_FS
+	.show_fdinfo	= bpf_link_show_fdinfo,
+#endif
+	.release	= bpf_link_release,
+	.read		= bpf_dummy_read,
+	.write		= bpf_dummy_write,
+	.poll		= bpf_link_poll,
+};
+
 static int bpf_link_alloc_id(struct bpf_link *link)
 {
 	int id;
@@ -3201,7 +3218,9 @@ int bpf_link_prime(struct bpf_link *link, struct bpf_link_primer *primer)
 		return id;
 	}
 
-	file = anon_inode_getfile("bpf_link", &bpf_link_fops, link, O_CLOEXEC);
+	file = anon_inode_getfile("bpf_link",
+				  link->ops->poll ? &bpf_link_fops_poll : &bpf_link_fops,
+				  link, O_CLOEXEC);
 	if (IS_ERR(file)) {
 		bpf_link_free_id(id);
 		put_unused_fd(fd);
@@ -3229,7 +3248,9 @@ int bpf_link_settle(struct bpf_link_primer *primer)
 
 int bpf_link_new_fd(struct bpf_link *link)
 {
-	return anon_inode_getfd("bpf-link", &bpf_link_fops, link, O_CLOEXEC);
+	return anon_inode_getfd("bpf-link",
+				link->ops->poll ? &bpf_link_fops_poll : &bpf_link_fops,
+				link, O_CLOEXEC);
 }
 
 struct bpf_link *bpf_link_get_from_fd(u32 ufd)
@@ -3239,7 +3260,7 @@ struct bpf_link *bpf_link_get_from_fd(u32 ufd)
 
 	if (!f.file)
 		return ERR_PTR(-EBADF);
-	if (f.file->f_op != &bpf_link_fops) {
+	if (f.file->f_op != &bpf_link_fops && f.file->f_op != &bpf_link_fops_poll) {
 		fdput(f);
 		return ERR_PTR(-EINVAL);
 	}
@@ -4971,7 +4992,7 @@ static int bpf_obj_get_info_by_fd(const union bpf_attr *attr,
 					     uattr);
 	else if (f.file->f_op == &btf_fops)
 		err = bpf_btf_get_info_by_fd(f.file, f.file->private_data, attr, uattr);
-	else if (f.file->f_op == &bpf_link_fops)
+	else if (f.file->f_op == &bpf_link_fops || f.file->f_op == &bpf_link_fops_poll)
 		err = bpf_link_get_info_by_fd(f.file, f.file->private_data,
 					      attr, uattr);
 	else
@@ -5106,7 +5127,7 @@ static int bpf_task_fd_query(const union bpf_attr *attr,
 	if (!file)
 		return -EBADF;
 
-	if (file->f_op == &bpf_link_fops) {
+	if (file->f_op == &bpf_link_fops || file->f_op == &bpf_link_fops_poll) {
 		struct bpf_link *link = file->private_data;
 
 		if (link->ops == &bpf_raw_tp_link_lops) {
-- 
cgit v1.2.3-58-ga151


From 67c3e8353f45c27800eecc46e00e8272f063f7d1 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Wed, 29 May 2024 23:59:42 -0700
Subject: bpf: export bpf_link_inc_not_zero.

bpf_link_inc_not_zero() will be used by kernel modules.  We will use it in
bpf_testmod.c later.

Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240530065946.979330-5-thinker.li@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 include/linux/bpf.h  | 6 ++++++
 kernel/bpf/syscall.c | 3 ++-
 2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 5eb61120e4f5..a834f4b761bc 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -2334,6 +2334,7 @@ int bpf_link_prime(struct bpf_link *link, struct bpf_link_primer *primer);
 int bpf_link_settle(struct bpf_link_primer *primer);
 void bpf_link_cleanup(struct bpf_link_primer *primer);
 void bpf_link_inc(struct bpf_link *link);
+struct bpf_link *bpf_link_inc_not_zero(struct bpf_link *link);
 void bpf_link_put(struct bpf_link *link);
 int bpf_link_new_fd(struct bpf_link *link);
 struct bpf_link *bpf_link_get_from_fd(u32 ufd);
@@ -2705,6 +2706,11 @@ static inline void bpf_link_inc(struct bpf_link *link)
 {
 }
 
+static inline struct bpf_link *bpf_link_inc_not_zero(struct bpf_link *link)
+{
+	return NULL;
+}
+
 static inline void bpf_link_put(struct bpf_link *link)
 {
 }
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 81efa1944942..5070fa20d05c 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -5437,10 +5437,11 @@ static int link_detach(union bpf_attr *attr)
 	return ret;
 }
 
-static struct bpf_link *bpf_link_inc_not_zero(struct bpf_link *link)
+struct bpf_link *bpf_link_inc_not_zero(struct bpf_link *link)
 {
 	return atomic64_fetch_add_unless(&link->refcnt, 1, 0) ? link : ERR_PTR(-ENOENT);
 }
+EXPORT_SYMBOL(bpf_link_inc_not_zero);
 
 struct bpf_link *bpf_link_by_id(u32 id)
 {
-- 
cgit v1.2.3-58-ga151


From 1a4b858b6a045828de1b536cfab7819c50864ed6 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Wed, 29 May 2024 23:59:43 -0700
Subject: selftests/bpf: test struct_ops with epoll

Verify whether a user space program is informed through epoll with EPOLLHUP
when a struct_ops object is detached.

The BPF code in selftests/bpf/progs/struct_ops_module.c has become
complex. Therefore, struct_ops_detach.c has been added to segregate the BPF
code for detachment tests from the BPF code for other tests based on the
recommendation of Andrii Nakryiko.

Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240530065946.979330-6-thinker.li@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 .../bpf/prog_tests/test_struct_ops_module.c        | 57 ++++++++++++++++++++++
 .../selftests/bpf/progs/struct_ops_detach.c        | 10 ++++
 2 files changed, 67 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/progs/struct_ops_detach.c

diff --git a/tools/testing/selftests/bpf/prog_tests/test_struct_ops_module.c b/tools/testing/selftests/bpf/prog_tests/test_struct_ops_module.c
index 29e183a80f49..bbcf12696a6b 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_struct_ops_module.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_struct_ops_module.c
@@ -3,9 +3,12 @@
 #include <test_progs.h>
 #include <time.h>
 
+#include <sys/epoll.h>
+
 #include "struct_ops_module.skel.h"
 #include "struct_ops_nulled_out_cb.skel.h"
 #include "struct_ops_forgotten_cb.skel.h"
+#include "struct_ops_detach.skel.h"
 
 static void check_map_info(struct bpf_map_info *info)
 {
@@ -242,6 +245,58 @@ cleanup:
 	struct_ops_forgotten_cb__destroy(skel);
 }
 
+/* Detach a link from a user space program */
+static void test_detach_link(void)
+{
+	struct epoll_event ev, events[2];
+	struct struct_ops_detach *skel;
+	struct bpf_link *link = NULL;
+	int fd, epollfd = -1, nfds;
+	int err;
+
+	skel = struct_ops_detach__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "struct_ops_detach__open_and_load"))
+		return;
+
+	link = bpf_map__attach_struct_ops(skel->maps.testmod_do_detach);
+	if (!ASSERT_OK_PTR(link, "attach_struct_ops"))
+		goto cleanup;
+
+	fd = bpf_link__fd(link);
+	if (!ASSERT_GE(fd, 0, "link_fd"))
+		goto cleanup;
+
+	epollfd = epoll_create1(0);
+	if (!ASSERT_GE(epollfd, 0, "epoll_create1"))
+		goto cleanup;
+
+	ev.events = EPOLLHUP;
+	ev.data.fd = fd;
+	err = epoll_ctl(epollfd, EPOLL_CTL_ADD, fd, &ev);
+	if (!ASSERT_OK(err, "epoll_ctl"))
+		goto cleanup;
+
+	err = bpf_link__detach(link);
+	if (!ASSERT_OK(err, "detach_link"))
+		goto cleanup;
+
+	/* Wait for EPOLLHUP */
+	nfds = epoll_wait(epollfd, events, 2, 500);
+	if (!ASSERT_EQ(nfds, 1, "epoll_wait"))
+		goto cleanup;
+
+	if (!ASSERT_EQ(events[0].data.fd, fd, "epoll_wait_fd"))
+		goto cleanup;
+	if (!ASSERT_TRUE(events[0].events & EPOLLHUP, "events[0].events"))
+		goto cleanup;
+
+cleanup:
+	if (epollfd >= 0)
+		close(epollfd);
+	bpf_link__destroy(link);
+	struct_ops_detach__destroy(skel);
+}
+
 void serial_test_struct_ops_module(void)
 {
 	if (test__start_subtest("struct_ops_load"))
@@ -254,5 +309,7 @@ void serial_test_struct_ops_module(void)
 		test_struct_ops_nulled_out_cb();
 	if (test__start_subtest("struct_ops_forgotten_cb"))
 		test_struct_ops_forgotten_cb();
+	if (test__start_subtest("test_detach_link"))
+		test_detach_link();
 }
 
diff --git a/tools/testing/selftests/bpf/progs/struct_ops_detach.c b/tools/testing/selftests/bpf/progs/struct_ops_detach.c
new file mode 100644
index 000000000000..56b787a89876
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/struct_ops_detach.c
@@ -0,0 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2024 Meta Platforms, Inc. and affiliates. */
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+#include "../bpf_testmod/bpf_testmod.h"
+
+char _license[] SEC("license") = "GPL";
+
+SEC(".struct_ops.link")
+struct bpf_testmod_ops testmod_do_detach;
-- 
cgit v1.2.3-58-ga151


From d14c1fac0c9722c4ec79589921c9e798601ca9d5 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Wed, 29 May 2024 23:59:46 -0700
Subject: bpftool: Change pid_iter.bpf.c to comply with the change of
 bpf_link_fops.

To support epoll, a new instance of file_operations, bpf_link_fops_poll,
has been added for links that support epoll. The pid_iter.bpf.c checks
f_ops for links and other BPF objects. The check should fail for struct_ops
links without this patch.

Acked-by: Quentin Monnet <qmo@kernel.org>
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240530065946.979330-9-thinker.li@gmail.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/bpf/bpftool/skeleton/pid_iter.bpf.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/tools/bpf/bpftool/skeleton/pid_iter.bpf.c b/tools/bpf/bpftool/skeleton/pid_iter.bpf.c
index 7bdbcac3cf62..948dde25034e 100644
--- a/tools/bpf/bpftool/skeleton/pid_iter.bpf.c
+++ b/tools/bpf/bpftool/skeleton/pid_iter.bpf.c
@@ -29,6 +29,7 @@ enum bpf_link_type___local {
 };
 
 extern const void bpf_link_fops __ksym;
+extern const void bpf_link_fops_poll __ksym __weak;
 extern const void bpf_map_fops __ksym;
 extern const void bpf_prog_fops __ksym;
 extern const void btf_fops __ksym;
@@ -84,7 +85,11 @@ int iter(struct bpf_iter__task_file *ctx)
 		fops = &btf_fops;
 		break;
 	case BPF_OBJ_LINK:
-		fops = &bpf_link_fops;
+		if (&bpf_link_fops_poll &&
+		    file->f_op == &bpf_link_fops_poll)
+			fops = &bpf_link_fops_poll;
+		else
+			fops = &bpf_link_fops;
 		break;
 	default:
 		return 0;
-- 
cgit v1.2.3-58-ga151


From 6166bb0cacb62c388e4ed06167511a70591cc59a Mon Sep 17 00:00:00 2001
From: Joe Damato <jdamato@fastly.com>
Date: Tue, 28 May 2024 18:11:36 +0000
Subject: net/mlx4: Track RX allocation failures in a stat

mlx4_en_alloc_frags currently returns -ENOMEM when mlx4_alloc_page
fails but does not increment a stat field when this occurs.

A new field called alloc_fail has been added to struct mlx4_en_rx_ring
which is now incremented in mlx4_en_rx_ring when -ENOMEM occurs.

Signed-off-by: Joe Damato <jdamato@fastly.com>
Tested-by: Martin Karsten <mkarsten@uwaterloo.ca>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528181139.515070-2-jdamato@fastly.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 1 +
 drivers/net/ethernet/mellanox/mlx4/en_rx.c     | 4 +++-
 drivers/net/ethernet/mellanox/mlx4/mlx4_en.h   | 1 +
 3 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 4c089cfa027a..4d2f8c458346 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -2073,6 +2073,7 @@ static void mlx4_en_clear_stats(struct net_device *dev)
 		priv->rx_ring[i]->csum_ok = 0;
 		priv->rx_ring[i]->csum_none = 0;
 		priv->rx_ring[i]->csum_complete = 0;
+		priv->rx_ring[i]->alloc_fail = 0;
 	}
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
index 8328df8645d5..15c57e9517e9 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
@@ -82,8 +82,10 @@ static int mlx4_en_alloc_frags(struct mlx4_en_priv *priv,
 
 	for (i = 0; i < priv->num_frags; i++, frags++) {
 		if (!frags->page) {
-			if (mlx4_alloc_page(priv, frags, gfp))
+			if (mlx4_alloc_page(priv, frags, gfp)) {
+				ring->alloc_fail++;
 				return -ENOMEM;
+			}
 			ring->rx_alloc_pages++;
 		}
 		rx_desc->data[i].addr = cpu_to_be64(frags->dma +
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index efe3f97b874f..cd70df22724b 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -355,6 +355,7 @@ struct mlx4_en_rx_ring {
 	unsigned long xdp_tx;
 	unsigned long xdp_tx_full;
 	unsigned long dropped;
+	unsigned long alloc_fail;
 	int hwtstamp_rx_filter;
 	cpumask_var_t affinity_mask;
 	struct xdp_rxq_info xdp_rxq;
-- 
cgit v1.2.3-58-ga151


From 64b62146ba9ee78c67e079b2ff09b106d8d6fd57 Mon Sep 17 00:00:00 2001
From: Joe Damato <jdamato@fastly.com>
Date: Tue, 28 May 2024 18:11:37 +0000
Subject: net/mlx4: link NAPI instances to queues and IRQs

Make mlx4 compatible with the newly added netlink queue GET APIs.

Signed-off-by: Joe Damato <jdamato@fastly.com>
Tested-by: Martin Karsten <mkarsten@uwaterloo.ca>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528181139.515070-3-jdamato@fastly.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx4/en_cq.c   | 14 ++++++++++++++
 drivers/net/ethernet/mellanox/mlx4/mlx4_en.h |  1 +
 2 files changed, 15 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_cq.c b/drivers/net/ethernet/mellanox/mlx4/en_cq.c
index 1184ac5751e1..461cc2c79c71 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_cq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_cq.c
@@ -126,6 +126,7 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq,
 		cq_idx = cq_idx % priv->rx_ring_num;
 		rx_cq = priv->rx_cq[cq_idx];
 		cq->vector = rx_cq->vector;
+		irq = mlx4_eq_get_irq(mdev->dev, cq->vector);
 	}
 
 	if (cq->type == RX)
@@ -142,18 +143,23 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq,
 	if (err)
 		goto free_eq;
 
+	cq->cq_idx = cq_idx;
 	cq->mcq.event = mlx4_en_cq_event;
 
 	switch (cq->type) {
 	case TX:
 		cq->mcq.comp = mlx4_en_tx_irq;
 		netif_napi_add_tx(cq->dev, &cq->napi, mlx4_en_poll_tx_cq);
+		netif_napi_set_irq(&cq->napi, irq);
 		napi_enable(&cq->napi);
+		netif_queue_set_napi(cq->dev, cq_idx, NETDEV_QUEUE_TYPE_TX, &cq->napi);
 		break;
 	case RX:
 		cq->mcq.comp = mlx4_en_rx_irq;
 		netif_napi_add(cq->dev, &cq->napi, mlx4_en_poll_rx_cq);
+		netif_napi_set_irq(&cq->napi, irq);
 		napi_enable(&cq->napi);
+		netif_queue_set_napi(cq->dev, cq_idx, NETDEV_QUEUE_TYPE_RX, &cq->napi);
 		break;
 	case TX_XDP:
 		/* nothing regarding napi, it's shared with rx ring */
@@ -189,6 +195,14 @@ void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq **pcq)
 void mlx4_en_deactivate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq)
 {
 	if (cq->type != TX_XDP) {
+		enum netdev_queue_type qtype;
+
+		if (cq->type == RX)
+			qtype = NETDEV_QUEUE_TYPE_RX;
+		else
+			qtype = NETDEV_QUEUE_TYPE_TX;
+
+		netif_queue_set_napi(cq->dev, cq->cq_idx, qtype, NULL);
 		napi_disable(&cq->napi);
 		netif_napi_del(&cq->napi);
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
index cd70df22724b..28b70dcc652e 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h
@@ -380,6 +380,7 @@ struct mlx4_en_cq {
 #define MLX4_EN_OPCODE_ERROR	0x1e
 
 	const struct cpumask *aff_mask;
+	int cq_idx;
 };
 
 struct mlx4_en_port_profile {
-- 
cgit v1.2.3-58-ga151


From a5602c6edf7cbf5d69a32f089fb3938f42a3ff03 Mon Sep 17 00:00:00 2001
From: Joe Damato <jdamato@fastly.com>
Date: Tue, 28 May 2024 18:11:38 +0000
Subject: net/mlx4: support per-queue statistics via netlink

Make mlx4 compatible with the newly added netlink queue stats API.

Signed-off-by: Joe Damato <jdamato@fastly.com>
Tested-by: Martin Karsten <mkarsten@uwaterloo.ca>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528181139.515070-4-jdamato@fastly.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 73 ++++++++++++++++++++++++++
 1 file changed, 73 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 4d2f8c458346..281b34af0bb4 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -43,6 +43,7 @@
 #include <net/vxlan.h>
 #include <net/devlink.h>
 #include <net/rps.h>
+#include <net/netdev_queues.h>
 
 #include <linux/mlx4/driver.h>
 #include <linux/mlx4/device.h>
@@ -3100,6 +3101,77 @@ void mlx4_en_set_stats_bitmap(struct mlx4_dev *dev,
 	last_i += NUM_PHY_STATS;
 }
 
+static void mlx4_get_queue_stats_rx(struct net_device *dev, int i,
+				    struct netdev_queue_stats_rx *stats)
+{
+	struct mlx4_en_priv *priv = netdev_priv(dev);
+	const struct mlx4_en_rx_ring *ring;
+
+	spin_lock_bh(&priv->stats_lock);
+
+	if (!priv->port_up || mlx4_is_master(priv->mdev->dev))
+		goto out_unlock;
+
+	ring = priv->rx_ring[i];
+	stats->packets = READ_ONCE(ring->packets);
+	stats->bytes   = READ_ONCE(ring->bytes);
+	stats->alloc_fail = READ_ONCE(ring->alloc_fail);
+
+out_unlock:
+	spin_unlock_bh(&priv->stats_lock);
+}
+
+static void mlx4_get_queue_stats_tx(struct net_device *dev, int i,
+				    struct netdev_queue_stats_tx *stats)
+{
+	struct mlx4_en_priv *priv = netdev_priv(dev);
+	const struct mlx4_en_tx_ring *ring;
+
+	spin_lock_bh(&priv->stats_lock);
+
+	if (!priv->port_up || mlx4_is_master(priv->mdev->dev))
+		goto out_unlock;
+
+	ring = priv->tx_ring[TX][i];
+	stats->packets = READ_ONCE(ring->packets);
+	stats->bytes   = READ_ONCE(ring->bytes);
+
+out_unlock:
+	spin_unlock_bh(&priv->stats_lock);
+}
+
+static void mlx4_get_base_stats(struct net_device *dev,
+				struct netdev_queue_stats_rx *rx,
+				struct netdev_queue_stats_tx *tx)
+{
+	struct mlx4_en_priv *priv = netdev_priv(dev);
+
+	spin_lock_bh(&priv->stats_lock);
+
+	if (!priv->port_up || mlx4_is_master(priv->mdev->dev))
+		goto out_unlock;
+
+	if (priv->rx_ring_num) {
+		rx->packets = 0;
+		rx->bytes = 0;
+		rx->alloc_fail = 0;
+	}
+
+	if (priv->tx_ring_num[TX]) {
+		tx->packets = 0;
+		tx->bytes = 0;
+	}
+
+out_unlock:
+	spin_unlock_bh(&priv->stats_lock);
+}
+
+static const struct netdev_stat_ops mlx4_stat_ops = {
+	.get_queue_stats_rx     = mlx4_get_queue_stats_rx,
+	.get_queue_stats_tx     = mlx4_get_queue_stats_tx,
+	.get_base_stats         = mlx4_get_base_stats,
+};
+
 int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 			struct mlx4_en_port_profile *prof)
 {
@@ -3263,6 +3335,7 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 	netif_set_real_num_tx_queues(dev, priv->tx_ring_num[TX]);
 	netif_set_real_num_rx_queues(dev, priv->rx_ring_num);
 
+	dev->stat_ops = &mlx4_stat_ops;
 	dev->ethtool_ops = &mlx4_en_ethtool_ops;
 
 	/*
-- 
cgit v1.2.3-58-ga151


From 727c94c9539aa8865cdbf6a783da6a6585f1fec2 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Tue, 28 May 2024 17:25:05 +0200
Subject: ethernet: octeontx2: avoid linking objects into multiple modules

Each object file contains information about which module it gets linked
into, so linking the same file into multiple modules now causes a warning:

scripts/Makefile.build:254: drivers/net/ethernet/marvell/octeontx2/nic/Makefile: otx2_devlink.o is added to multiple modules: rvu_nicpf rvu_nicvf

Change the way that octeontx2 ethernet is built by moving the common
file into a separate module with exported symbols instead.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528152527.2148092-1-arnd@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/marvell/octeontx2/nic/Makefile       | 14 ++++++++------
 drivers/net/ethernet/marvell/octeontx2/nic/otx2_dcbnl.c   | 11 +++++++++++
 drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c |  6 ++++++
 3 files changed, 25 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
index 5664f768cb0c..e4c5dc46dd42 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
@@ -3,16 +3,18 @@
 # Makefile for Marvell's RVU Ethernet device drivers
 #
 
-obj-$(CONFIG_OCTEONTX2_PF) += rvu_nicpf.o otx2_ptp.o
-obj-$(CONFIG_OCTEONTX2_VF) += rvu_nicvf.o otx2_ptp.o
+obj-$(CONFIG_OCTEONTX2_PF) += rvu_nicpf.o otx2_ptp.o otx2_devlink.o
+obj-$(CONFIG_OCTEONTX2_VF) += rvu_nicvf.o otx2_ptp.o otx2_devlink.o
 
 rvu_nicpf-y := otx2_pf.o otx2_common.o otx2_txrx.o otx2_ethtool.o \
                otx2_flows.o otx2_tc.o cn10k.o otx2_dmac_flt.o \
-               otx2_devlink.o qos_sq.o qos.o
-rvu_nicvf-y := otx2_vf.o otx2_devlink.o
+               qos_sq.o qos.o
+rvu_nicvf-y := otx2_vf.o
 
-rvu_nicpf-$(CONFIG_DCB) += otx2_dcbnl.o
-rvu_nicvf-$(CONFIG_DCB) += otx2_dcbnl.o
+ifdef CONFIG_DCB
+obj-$(CONFIG_OCTEONTX2_PF) += otx2_dcbnl.o
+obj-$(CONFIG_OCTEONTX2_VF) += otx2_dcbnl.o
+endif
 rvu_nicpf-$(CONFIG_MACSEC) += cn10k_macsec.o
 
 ccflags-y += -I$(srctree)/drivers/net/ethernet/marvell/octeontx2/af
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_dcbnl.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_dcbnl.c
index 28fb643d2917..0d7e611d9a05 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_dcbnl.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_dcbnl.c
@@ -54,6 +54,7 @@ int otx2_pfc_txschq_config(struct otx2_nic *pfvf)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(otx2_pfc_txschq_config);
 
 static int otx2_pfc_txschq_alloc_one(struct otx2_nic *pfvf, u8 prio)
 {
@@ -122,6 +123,7 @@ int otx2_pfc_txschq_alloc(struct otx2_nic *pfvf)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(otx2_pfc_txschq_alloc);
 
 static int otx2_pfc_txschq_stop_one(struct otx2_nic *pfvf, u8 prio)
 {
@@ -260,6 +262,7 @@ update_sq_smq_map:
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(otx2_pfc_txschq_update);
 
 int otx2_pfc_txschq_stop(struct otx2_nic *pfvf)
 {
@@ -282,6 +285,7 @@ int otx2_pfc_txschq_stop(struct otx2_nic *pfvf)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(otx2_pfc_txschq_stop);
 
 int otx2_config_priority_flow_ctrl(struct otx2_nic *pfvf)
 {
@@ -321,6 +325,7 @@ unlock:
 	mutex_unlock(&pfvf->mbox.lock);
 	return err;
 }
+EXPORT_SYMBOL_GPL(otx2_config_priority_flow_ctrl);
 
 void otx2_update_bpid_in_rqctx(struct otx2_nic *pfvf, int vlan_prio, int qidx,
 			       bool pfc_enable)
@@ -385,6 +390,7 @@ out:
 			 "Updating BPIDs in CQ and Aura contexts of RQ%d failed with err %d\n",
 			 qidx, err);
 }
+EXPORT_SYMBOL_GPL(otx2_update_bpid_in_rqctx);
 
 static int otx2_dcbnl_ieee_getpfc(struct net_device *dev, struct ieee_pfc *pfc)
 {
@@ -472,3 +478,8 @@ int otx2_dcbnl_set_ops(struct net_device *dev)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(otx2_dcbnl_set_ops);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Marvell RVU dcbnl");
+MODULE_AUTHOR("Sunil Goutham <sgoutham@marvell.com>");
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c
index 99ddf31269d9..440f574d1195 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c
@@ -113,6 +113,7 @@ err_dl:
 	devlink_free(dl);
 	return err;
 }
+EXPORT_SYMBOL_GPL(otx2_register_dl);
 
 void otx2_unregister_dl(struct otx2_nic *pfvf)
 {
@@ -124,3 +125,8 @@ void otx2_unregister_dl(struct otx2_nic *pfvf)
 				  ARRAY_SIZE(otx2_dl_params));
 	devlink_free(dl);
 }
+EXPORT_SYMBOL_GPL(otx2_unregister_dl);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Marvell RVU PF/VF Netdev Devlink");
+MODULE_AUTHOR("Sunil Goutham <sgoutham@marvell.com>");
-- 
cgit v1.2.3-58-ga151


From e3bbb994a7e0ead1e028c5a8ec5f8526b4a9c5c9 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Tue, 28 May 2024 18:15:25 +0200
Subject: hns3: avoid linking objects into multiple modules

Each object file contains information about which module it gets linked
into, so linking the same file into multiple modules now causes a warning:

scripts/Makefile.build:254: drivers/net/ethernet/hisilicon/hns3/Makefile: hns3_common/hclge_comm_cmd.o is added to multiple modules: hclge hclgevf
scripts/Makefile.build:254: drivers/net/ethernet/hisilicon/hns3/Makefile: hns3_common/hclge_comm_rss.o is added to multiple modules: hclge hclgevf
scripts/Makefile.build:254: drivers/net/ethernet/hisilicon/hns3/Makefile: hns3_common/hclge_comm_tqp_stats.o is added to multiple modules: hclge hclgevf

Change the way that hns3 is built by moving the three common files into a
separate module with exported symbols instead.

Fixes: 5f20be4e90e6 ("net: hns3: refactor hns3 makefile to support hns3_common module")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528161603.2443125-1-arnd@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/hisilicon/hns3/Makefile               | 11 +++++------
 .../ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c   | 11 +++++++++++
 .../ethernet/hisilicon/hns3/hns3_common/hclge_comm_rss.c   | 14 ++++++++++++++
 .../hisilicon/hns3/hns3_common/hclge_comm_tqp_stats.c      |  5 +++++
 4 files changed, 35 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns3/Makefile b/drivers/net/ethernet/hisilicon/hns3/Makefile
index 8e9293e57bfd..e8af26da1fc1 100644
--- a/drivers/net/ethernet/hisilicon/hns3/Makefile
+++ b/drivers/net/ethernet/hisilicon/hns3/Makefile
@@ -15,15 +15,14 @@ hns3-objs = hns3_enet.o hns3_ethtool.o hns3_debugfs.o
 
 hns3-$(CONFIG_HNS3_DCB) += hns3_dcbnl.o
 
-obj-$(CONFIG_HNS3_HCLGEVF) += hclgevf.o
+obj-$(CONFIG_HNS3_HCLGEVF) += hclgevf.o hclge-common.o
 
-hclgevf-objs = hns3vf/hclgevf_main.o hns3vf/hclgevf_mbx.o  hns3vf/hclgevf_devlink.o hns3vf/hclgevf_regs.o \
-		hns3_common/hclge_comm_cmd.o hns3_common/hclge_comm_rss.o hns3_common/hclge_comm_tqp_stats.o
+hclge-common-objs += hns3_common/hclge_comm_cmd.o hns3_common/hclge_comm_rss.o hns3_common/hclge_comm_tqp_stats.o
 
-obj-$(CONFIG_HNS3_HCLGE) += hclge.o
+hclgevf-objs = hns3vf/hclgevf_main.o hns3vf/hclgevf_mbx.o  hns3vf/hclgevf_devlink.o hns3vf/hclgevf_regs.o
+
+obj-$(CONFIG_HNS3_HCLGE) += hclge.o hclge-common.o
 hclge-objs = hns3pf/hclge_main.o hns3pf/hclge_mdio.o hns3pf/hclge_tm.o hns3pf/hclge_regs.o \
 		hns3pf/hclge_mbx.o hns3pf/hclge_err.o  hns3pf/hclge_debugfs.o hns3pf/hclge_ptp.o hns3pf/hclge_devlink.o \
-		hns3_common/hclge_comm_cmd.o hns3_common/hclge_comm_rss.o hns3_common/hclge_comm_tqp_stats.o
-
 
 hclge-$(CONFIG_HNS3_DCB) += hns3pf/hclge_dcb.o
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c
index ea40b594dbac..4ad4e8ab2f1f 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c
@@ -48,6 +48,7 @@ void hclge_comm_cmd_reuse_desc(struct hclge_desc *desc, bool is_read)
 	else
 		desc->flag &= cpu_to_le16(~HCLGE_COMM_CMD_FLAG_WR);
 }
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_reuse_desc);
 
 static void hclge_comm_set_default_capability(struct hnae3_ae_dev *ae_dev,
 					      bool is_pf)
@@ -72,6 +73,7 @@ void hclge_comm_cmd_setup_basic_desc(struct hclge_desc *desc,
 	if (is_read)
 		desc->flag |= cpu_to_le16(HCLGE_COMM_CMD_FLAG_WR);
 }
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_setup_basic_desc);
 
 int hclge_comm_firmware_compat_config(struct hnae3_ae_dev *ae_dev,
 				      struct hclge_comm_hw *hw, bool en)
@@ -517,6 +519,7 @@ int hclge_comm_cmd_send(struct hclge_comm_hw *hw, struct hclge_desc *desc,
 
 	return ret;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_send);
 
 static void hclge_comm_cmd_uninit_regs(struct hclge_comm_hw *hw)
 {
@@ -553,6 +556,7 @@ void hclge_comm_cmd_uninit(struct hnae3_ae_dev *ae_dev,
 	hclge_comm_free_cmd_desc(&cmdq->csq);
 	hclge_comm_free_cmd_desc(&cmdq->crq);
 }
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_uninit);
 
 int hclge_comm_cmd_queue_init(struct pci_dev *pdev, struct hclge_comm_hw *hw)
 {
@@ -591,6 +595,7 @@ err_csq:
 	hclge_comm_free_cmd_desc(&hw->cmq.csq);
 	return ret;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_queue_init);
 
 void hclge_comm_cmd_init_ops(struct hclge_comm_hw *hw,
 			     const struct hclge_comm_cmq_ops *ops)
@@ -602,6 +607,7 @@ void hclge_comm_cmd_init_ops(struct hclge_comm_hw *hw,
 		cmdq->ops.trace_cmd_get = ops->trace_cmd_get;
 	}
 }
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_init_ops);
 
 int hclge_comm_cmd_init(struct hnae3_ae_dev *ae_dev, struct hclge_comm_hw *hw,
 			u32 *fw_version, bool is_pf,
@@ -672,3 +678,8 @@ err_cmd_init:
 
 	return ret;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_cmd_init);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("HNS3: Hisilicon Ethernet PF/VF Common Library");
+MODULE_AUTHOR("Huawei Tech. Co., Ltd.");
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_rss.c b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_rss.c
index b4ae2160aff4..4e2bb6556b1c 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_rss.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_rss.c
@@ -62,6 +62,7 @@ int hclge_comm_rss_init_cfg(struct hnae3_handle *nic,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_rss_init_cfg);
 
 void hclge_comm_get_rss_tc_info(u16 rss_size, u8 hw_tc_map, u16 *tc_offset,
 				u16 *tc_valid, u16 *tc_size)
@@ -78,6 +79,7 @@ void hclge_comm_get_rss_tc_info(u16 rss_size, u8 hw_tc_map, u16 *tc_offset,
 		tc_offset[i] = (hw_tc_map & BIT(i)) ? rss_size * i : 0;
 	}
 }
+EXPORT_SYMBOL_GPL(hclge_comm_get_rss_tc_info);
 
 int hclge_comm_set_rss_tc_mode(struct hclge_comm_hw *hw, u16 *tc_offset,
 			       u16 *tc_valid, u16 *tc_size)
@@ -113,6 +115,7 @@ int hclge_comm_set_rss_tc_mode(struct hclge_comm_hw *hw, u16 *tc_offset,
 
 	return ret;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_tc_mode);
 
 int hclge_comm_set_rss_hash_key(struct hclge_comm_rss_cfg *rss_cfg,
 				struct hclge_comm_hw *hw, const u8 *key,
@@ -143,6 +146,7 @@ int hclge_comm_set_rss_hash_key(struct hclge_comm_rss_cfg *rss_cfg,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_hash_key);
 
 int hclge_comm_set_rss_tuple(struct hnae3_ae_dev *ae_dev,
 			     struct hclge_comm_hw *hw,
@@ -185,11 +189,13 @@ int hclge_comm_set_rss_tuple(struct hnae3_ae_dev *ae_dev,
 	rss_cfg->rss_tuple_sets.ipv6_fragment_en = req->ipv6_fragment_en;
 	return 0;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_tuple);
 
 u32 hclge_comm_get_rss_key_size(struct hnae3_handle *handle)
 {
 	return HCLGE_COMM_RSS_KEY_SIZE;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_get_rss_key_size);
 
 int hclge_comm_parse_rss_hfunc(struct hclge_comm_rss_cfg *rss_cfg,
 			       const u8 hfunc, u8 *hash_algo)
@@ -217,6 +223,7 @@ void hclge_comm_rss_indir_init_cfg(struct hnae3_ae_dev *ae_dev,
 	for (i = 0; i < ae_dev->dev_specs.rss_ind_tbl_size; i++)
 		rss_cfg->rss_indirection_tbl[i] = i % rss_cfg->rss_size;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_rss_indir_init_cfg);
 
 int hclge_comm_get_rss_tuple(struct hclge_comm_rss_cfg *rss_cfg, int flow_type,
 			     u8 *tuple_sets)
@@ -250,6 +257,7 @@ int hclge_comm_get_rss_tuple(struct hclge_comm_rss_cfg *rss_cfg, int flow_type,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_get_rss_tuple);
 
 static void
 hclge_comm_append_rss_msb_info(struct hclge_comm_rss_ind_tbl_cmd *req,
@@ -304,6 +312,7 @@ int hclge_comm_set_rss_indir_table(struct hnae3_ae_dev *ae_dev,
 	}
 	return 0;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_indir_table);
 
 int hclge_comm_set_rss_input_tuple(struct hclge_comm_hw *hw,
 				   struct hclge_comm_rss_cfg *rss_cfg)
@@ -332,6 +341,7 @@ int hclge_comm_set_rss_input_tuple(struct hclge_comm_hw *hw,
 			"failed to configure rss input, ret = %d.\n", ret);
 	return ret;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_input_tuple);
 
 void hclge_comm_get_rss_hash_info(struct hclge_comm_rss_cfg *rss_cfg, u8 *key,
 				  u8 *hfunc)
@@ -355,6 +365,7 @@ void hclge_comm_get_rss_hash_info(struct hclge_comm_rss_cfg *rss_cfg, u8 *key,
 	if (key)
 		memcpy(key, rss_cfg->rss_hash_key, HCLGE_COMM_RSS_KEY_SIZE);
 }
+EXPORT_SYMBOL_GPL(hclge_comm_get_rss_hash_info);
 
 void hclge_comm_get_rss_indir_tbl(struct hclge_comm_rss_cfg *rss_cfg,
 				  u32 *indir, u16 rss_ind_tbl_size)
@@ -367,6 +378,7 @@ void hclge_comm_get_rss_indir_tbl(struct hclge_comm_rss_cfg *rss_cfg,
 	for (i = 0; i < rss_ind_tbl_size; i++)
 		indir[i] = rss_cfg->rss_indirection_tbl[i];
 }
+EXPORT_SYMBOL_GPL(hclge_comm_get_rss_indir_tbl);
 
 int hclge_comm_set_rss_algo_key(struct hclge_comm_hw *hw, const u8 hfunc,
 				const u8 *key)
@@ -408,6 +420,7 @@ int hclge_comm_set_rss_algo_key(struct hclge_comm_hw *hw, const u8 hfunc,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_set_rss_algo_key);
 
 static u8 hclge_comm_get_rss_hash_bits(struct ethtool_rxnfc *nfc)
 {
@@ -502,3 +515,4 @@ u64 hclge_comm_convert_rss_tuple(u8 tuple_sets)
 
 	return tuple_data;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_convert_rss_tuple);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_tqp_stats.c b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_tqp_stats.c
index 618f66d9586b..2b31188ff555 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_tqp_stats.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_tqp_stats.c
@@ -26,6 +26,7 @@ u64 *hclge_comm_tqps_get_stats(struct hnae3_handle *handle, u64 *data)
 
 	return buff;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_tqps_get_stats);
 
 int hclge_comm_tqps_get_sset_count(struct hnae3_handle *handle)
 {
@@ -33,6 +34,7 @@ int hclge_comm_tqps_get_sset_count(struct hnae3_handle *handle)
 
 	return kinfo->num_tqps * HCLGE_COMM_QUEUE_PAIR_SIZE;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_tqps_get_sset_count);
 
 u8 *hclge_comm_tqps_get_strings(struct hnae3_handle *handle, u8 *data)
 {
@@ -56,6 +58,7 @@ u8 *hclge_comm_tqps_get_strings(struct hnae3_handle *handle, u8 *data)
 
 	return buff;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_tqps_get_strings);
 
 int hclge_comm_tqps_update_stats(struct hnae3_handle *handle,
 				 struct hclge_comm_hw *hw)
@@ -99,6 +102,7 @@ int hclge_comm_tqps_update_stats(struct hnae3_handle *handle,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(hclge_comm_tqps_update_stats);
 
 void hclge_comm_reset_tqp_stats(struct hnae3_handle *handle)
 {
@@ -113,3 +117,4 @@ void hclge_comm_reset_tqp_stats(struct hnae3_handle *handle)
 		memset(&tqp->tqp_stats, 0, sizeof(tqp->tqp_stats));
 	}
 }
+EXPORT_SYMBOL_GPL(hclge_comm_reset_tqp_stats);
-- 
cgit v1.2.3-58-ga151


From 3eb76e71b16e8ba5277bf97617aef51f5e64dbe4 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Tue, 28 May 2024 17:02:53 -0700
Subject: ionic: fix potential irq name truncation

Address a warning about potential string truncation based on the
string buffer sizes.  We can add some hints to the string format
specifier to set limits on the resulting possible string to
squelch the complaints.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240529000259.25775-2-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_lif.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
index 24870da3f484..12fda3b860b9 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
@@ -242,7 +242,7 @@ static int ionic_request_irq(struct ionic_lif *lif, struct ionic_qcq *qcq)
 		name = dev_name(dev);
 
 	snprintf(intr->name, sizeof(intr->name),
-		 "%s-%s-%s", IONIC_DRV_NAME, name, q->name);
+		 "%.5s-%.16s-%.8s", IONIC_DRV_NAME, name, q->name);
 
 	return devm_request_irq(dev, intr->vector, ionic_isr,
 				0, intr->name, &qcq->napi);
-- 
cgit v1.2.3-58-ga151


From 8097a2f3d21a672c772ca1919f253e611f2f0bad Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Tue, 28 May 2024 17:02:54 -0700
Subject: ionic: Reset LIF device while restarting LIF

Recovery from broken states can be hard.  If the LIF reset in
the fw_down path didn't work because the PCI link was broken,
the FW won't be in the right state for proper restart.  We can
fire another LIF reset in the fw_up path to be sure things
are clean on restart.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240529000259.25775-3-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_lif.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
index 12fda3b860b9..101cbc088853 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
@@ -3388,6 +3388,7 @@ static void ionic_lif_handle_fw_up(struct ionic_lif *lif)
 	 * just need to reanimate it.
 	 */
 	ionic_init_devinfo(ionic);
+	ionic_reset(ionic);
 	err = ionic_identify(ionic);
 	if (err)
 		goto err_out;
-- 
cgit v1.2.3-58-ga151


From 4dde9588c54d1d86a415a5ed7dc6d8a605fe11ce Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@amd.com>
Date: Tue, 28 May 2024 17:02:55 -0700
Subject: ionic: Pass ionic_txq_desc to ionic_tx_tso_post

Pass the ionic_txq_desc instead of re-referencing it from the q->txq
array since the caller to ionic_tx_tso_post will always have the
current ionic_txq_desc pointer already.

Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240529000259.25775-4-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_txrx.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
index 5dba6d2d633c..c6aa8fb743be 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
@@ -1357,7 +1357,7 @@ static int ionic_tx_tcp_pseudo_csum(struct sk_buff *skb)
 }
 
 static void ionic_tx_tso_post(struct net_device *netdev, struct ionic_queue *q,
-			      struct ionic_tx_desc_info *desc_info,
+			      struct ionic_txq_desc *desc,
 			      struct sk_buff *skb,
 			      dma_addr_t addr, u8 nsge, u16 len,
 			      unsigned int hdrlen, unsigned int mss,
@@ -1365,7 +1365,6 @@ static void ionic_tx_tso_post(struct net_device *netdev, struct ionic_queue *q,
 			      u16 vlan_tci, bool has_vlan,
 			      bool start, bool done)
 {
-	struct ionic_txq_desc *desc = &q->txq[q->head_idx];
 	u8 flags = 0;
 	u64 cmd;
 
@@ -1503,10 +1502,9 @@ static int ionic_tx_tso(struct net_device *netdev, struct ionic_queue *q,
 		seg_rem = min(tso_rem, mss);
 		done = (tso_rem == 0);
 		/* post descriptor */
-		ionic_tx_tso_post(netdev, q, desc_info, skb,
-				  desc_addr, desc_nsge, desc_len,
-				  hdrlen, mss, outer_csum, vlan_tci, has_vlan,
-				  start, done);
+		ionic_tx_tso_post(netdev, q, desc, skb, desc_addr, desc_nsge,
+				  desc_len, hdrlen, mss, outer_csum, vlan_tci,
+				  has_vlan, start, done);
 		start = false;
 		/* Buffer information is stored with the first tso descriptor */
 		desc_info = &q->tx_info[q->head_idx];
-- 
cgit v1.2.3-58-ga151


From d9c04209990be6c03a354b124e012293273e05d5 Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@amd.com>
Date: Tue, 28 May 2024 17:02:56 -0700
Subject: ionic: Mark error paths in the data path as unlikely

As the title states, mark unlikely error paths in the data path as
unlikely.

Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240529000259.25775-5-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_txrx.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
index c6aa8fb743be..14aa3844b699 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
@@ -582,7 +582,7 @@ static bool ionic_run_xdp(struct ionic_rx_stats *stats,
 					   buf_info->page_offset,
 					   true);
 		__netif_tx_unlock(nq);
-		if (err) {
+		if (unlikely(err)) {
 			netdev_dbg(netdev, "tx ionic_xdp_post_frame err %d\n", err);
 			goto out_xdp_abort;
 		}
@@ -597,7 +597,7 @@ static bool ionic_run_xdp(struct ionic_rx_stats *stats,
 			       IONIC_PAGE_SIZE, DMA_FROM_DEVICE);
 
 		err = xdp_do_redirect(netdev, &xdp_buf, xdp_prog);
-		if (err) {
+		if (unlikely(err)) {
 			netdev_dbg(netdev, "xdp_do_redirect err %d\n", err);
 			goto out_xdp_abort;
 		}
@@ -1058,7 +1058,7 @@ static dma_addr_t ionic_tx_map_single(struct ionic_queue *q,
 	dma_addr_t dma_addr;
 
 	dma_addr = dma_map_single(dev, data, len, DMA_TO_DEVICE);
-	if (dma_mapping_error(dev, dma_addr)) {
+	if (unlikely(dma_mapping_error(dev, dma_addr))) {
 		net_warn_ratelimited("%s: DMA single map failed on %s!\n",
 				     dev_name(dev), q->name);
 		q_to_tx_stats(q)->dma_map_err++;
@@ -1075,7 +1075,7 @@ static dma_addr_t ionic_tx_map_frag(struct ionic_queue *q,
 	dma_addr_t dma_addr;
 
 	dma_addr = skb_frag_dma_map(dev, frag, offset, len, DMA_TO_DEVICE);
-	if (dma_mapping_error(dev, dma_addr)) {
+	if (unlikely(dma_mapping_error(dev, dma_addr))) {
 		net_warn_ratelimited("%s: DMA frag map failed on %s!\n",
 				     dev_name(dev), q->name);
 		q_to_tx_stats(q)->dma_map_err++;
@@ -1316,7 +1316,7 @@ static int ionic_tx_tcp_inner_pseudo_csum(struct sk_buff *skb)
 	int err;
 
 	err = skb_cow_head(skb, 0);
-	if (err)
+	if (unlikely(err))
 		return err;
 
 	if (skb->protocol == cpu_to_be16(ETH_P_IP)) {
@@ -1340,7 +1340,7 @@ static int ionic_tx_tcp_pseudo_csum(struct sk_buff *skb)
 	int err;
 
 	err = skb_cow_head(skb, 0);
-	if (err)
+	if (unlikely(err))
 		return err;
 
 	if (skb->protocol == cpu_to_be16(ETH_P_IP)) {
@@ -1444,7 +1444,7 @@ static int ionic_tx_tso(struct net_device *netdev, struct ionic_queue *q,
 		err = ionic_tx_tcp_inner_pseudo_csum(skb);
 	else
 		err = ionic_tx_tcp_pseudo_csum(skb);
-	if (err) {
+	if (unlikely(err)) {
 		/* clean up mapping from ionic_tx_map_skb */
 		ionic_tx_desc_unmap_bufs(q, desc_info);
 		return err;
@@ -1729,7 +1729,7 @@ static int ionic_tx_descs_needed(struct ionic_queue *q, struct sk_buff *skb)
 linearize:
 	if (too_many_frags) {
 		err = skb_linearize(skb);
-		if (err)
+		if (unlikely(err))
 			return err;
 		q_to_tx_stats(q)->linearize++;
 	}
@@ -1763,7 +1763,7 @@ static netdev_tx_t ionic_start_hwstamp_xmit(struct sk_buff *skb,
 	else
 		err = ionic_tx(netdev, q, skb);
 
-	if (err)
+	if (unlikely(err))
 		goto err_out_drop;
 
 	return NETDEV_TX_OK;
@@ -1809,7 +1809,7 @@ netdev_tx_t ionic_start_xmit(struct sk_buff *skb, struct net_device *netdev)
 	else
 		err = ionic_tx(netdev, q, skb);
 
-	if (err)
+	if (unlikely(err))
 		goto err_out_drop;
 
 	return NETDEV_TX_OK;
-- 
cgit v1.2.3-58-ga151


From fc53d46524480a732a233bc9d6cdf6155205d9eb Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@amd.com>
Date: Tue, 28 May 2024 17:02:57 -0700
Subject: ionic: Use netdev_name() function instead of netdev->name

There is no reason not to use netdev_name() in these places, so do just
as the title states.

Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Link: https://lore.kernel.org/r/20240529000259.25775-6-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_debugfs.c | 2 +-
 drivers/net/ethernet/pensando/ionic/ionic_lif.c     | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c b/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c
index c3ae11a48024..59e5a9f21105 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_debugfs.c
@@ -220,7 +220,7 @@ static int netdev_show(struct seq_file *seq, void *v)
 {
 	struct net_device *netdev = seq->private;
 
-	seq_printf(seq, "%s\n", netdev->name);
+	seq_printf(seq, "%s\n", netdev_name(netdev));
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
index 101cbc088853..23e1f6638b38 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
@@ -237,7 +237,7 @@ static int ionic_request_irq(struct ionic_lif *lif, struct ionic_qcq *qcq)
 	const char *name;
 
 	if (lif->registered)
-		name = lif->netdev->name;
+		name = netdev_name(lif->netdev);
 	else
 		name = dev_name(dev);
 
@@ -3732,7 +3732,7 @@ static void ionic_lif_set_netdev_info(struct ionic_lif *lif)
 		},
 	};
 
-	strscpy(ctx.cmd.lif_setattr.name, lif->netdev->name,
+	strscpy(ctx.cmd.lif_setattr.name, netdev_name(lif->netdev),
 		sizeof(ctx.cmd.lif_setattr.name));
 
 	ionic_adminq_post_wait(lif, &ctx);
-- 
cgit v1.2.3-58-ga151


From 488da00479d50ec5970b3f84e9fad2c5fda2d78c Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Tue, 28 May 2024 17:02:58 -0700
Subject: ionic: only sync frag_len in first buffer of xdp

We don't want to try to sync more length than might be
in the first frag of an Rx skb, so make sure to use
the frag_len rather than the full len.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240529000259.25775-7-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_txrx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
index 14aa3844b699..c3a6c4af52f1 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
@@ -502,7 +502,7 @@ static bool ionic_run_xdp(struct ionic_rx_stats *stats,
 			 XDP_PACKET_HEADROOM, frag_len, false);
 
 	dma_sync_single_range_for_cpu(rxq->dev, ionic_rx_buf_pa(buf_info),
-				      XDP_PACKET_HEADROOM, len,
+				      XDP_PACKET_HEADROOM, frag_len,
 				      DMA_FROM_DEVICE);
 
 	prefetchw(&xdp_buf.data_hard_start);
-- 
cgit v1.2.3-58-ga151


From a54e2a36b68c30822d3571b969e24aa429b3d1a5 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Tue, 28 May 2024 17:02:59 -0700
Subject: ionic: fix up ionic_if.h kernel-doc issues

All the changes here are whitespace and comments, no code
or definitions changed.  Not all issues were addressed, but
it is much better than it was.  Mostly fixed was a lot
of "Excess union member" and ''rsvd' not described' warnings.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240529000259.25775-8-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_if.h | 237 +++++++++++++++++++------
 1 file changed, 181 insertions(+), 56 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_if.h b/drivers/net/ethernet/pensando/ionic/ionic_if.h
index 9a1825edf0d0..9c85c0706c6e 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_if.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_if.h
@@ -71,7 +71,7 @@ enum ionic_cmd_opcode {
 	IONIC_CMD_FW_CONTROL_V1		        = 255,
 };
 
-/**
+/*
  * enum ionic_status_code - Device command return codes
  */
 enum ionic_status_code {
@@ -112,6 +112,7 @@ enum ionic_notifyq_opcode {
 /**
  * struct ionic_admin_cmd - General admin command format
  * @opcode:     Opcode for the command
+ * @rsvd:       reserved byte(s)
  * @lif_index:  LIF index
  * @cmd_data:   Opcode-specific command bytes
  */
@@ -125,6 +126,7 @@ struct ionic_admin_cmd {
 /**
  * struct ionic_admin_comp - General admin command completion format
  * @status:     Status of the command (enum ionic_status_code)
+ * @rsvd:       reserved byte(s)
  * @comp_index: Index in the descriptor ring for which this is the completion
  * @cmd_data:   Command-specific bytes
  * @color:      Color bit (Always 0 for commands issued to the
@@ -147,6 +149,7 @@ static inline u8 color_match(u8 color, u8 done_color)
 /**
  * struct ionic_nop_cmd - NOP command
  * @opcode: opcode
+ * @rsvd:   reserved byte(s)
  */
 struct ionic_nop_cmd {
 	u8 opcode;
@@ -156,6 +159,7 @@ struct ionic_nop_cmd {
 /**
  * struct ionic_nop_comp - NOP command completion
  * @status: Status of the command (enum ionic_status_code)
+ * @rsvd:   reserved byte(s)
  */
 struct ionic_nop_comp {
 	u8 status;
@@ -166,6 +170,7 @@ struct ionic_nop_comp {
  * struct ionic_dev_init_cmd - Device init command
  * @opcode:    opcode
  * @type:      Device type
+ * @rsvd:      reserved byte(s)
  */
 struct ionic_dev_init_cmd {
 	u8     opcode;
@@ -176,6 +181,7 @@ struct ionic_dev_init_cmd {
 /**
  * struct ionic_dev_init_comp - Device init command completion
  * @status: Status of the command (enum ionic_status_code)
+ * @rsvd:   reserved byte(s)
  */
 struct ionic_dev_init_comp {
 	u8 status;
@@ -185,6 +191,7 @@ struct ionic_dev_init_comp {
 /**
  * struct ionic_dev_reset_cmd - Device reset command
  * @opcode: opcode
+ * @rsvd:   reserved byte(s)
  */
 struct ionic_dev_reset_cmd {
 	u8 opcode;
@@ -194,6 +201,7 @@ struct ionic_dev_reset_cmd {
 /**
  * struct ionic_dev_reset_comp - Reset command completion
  * @status: Status of the command (enum ionic_status_code)
+ * @rsvd:   reserved byte(s)
  */
 struct ionic_dev_reset_comp {
 	u8 status;
@@ -207,6 +215,7 @@ struct ionic_dev_reset_comp {
  * struct ionic_dev_identify_cmd - Driver/device identify command
  * @opcode:  opcode
  * @ver:     Highest version of identify supported by driver
+ * @rsvd:    reserved byte(s)
  */
 struct ionic_dev_identify_cmd {
 	u8 opcode;
@@ -218,6 +227,7 @@ struct ionic_dev_identify_cmd {
  * struct ionic_dev_identify_comp - Driver/device identify command completion
  * @status: Status of the command (enum ionic_status_code)
  * @ver:    Version of identify returned by device
+ * @rsvd:   reserved byte(s)
  */
 struct ionic_dev_identify_comp {
 	u8 status;
@@ -242,6 +252,7 @@ enum ionic_os_type {
  * @kernel_ver:       Kernel version, numeric format
  * @kernel_ver_str:   Kernel version, string format
  * @driver_ver_str:   Driver version, string format
+ * @words:            word access to struct contents
  */
 union ionic_drv_identity {
 	struct {
@@ -267,7 +278,9 @@ enum ionic_dev_capability {
  * union ionic_dev_identity - device identity information
  * @version:          Version of device identify
  * @type:             Identify type (0 for now)
+ * @rsvd:             reserved byte(s)
  * @nports:           Number of ports provisioned
+ * @rsvd2:            reserved byte(s)
  * @nlifs:            Number of LIFs provisioned
  * @nintrs:           Number of interrupts provisioned
  * @ndbpgs_per_lif:   Number of doorbell pages per LIF
@@ -284,6 +297,7 @@ enum ionic_dev_capability {
  * @hwstamp_mult:     Hardware tick to nanosecond multiplier.
  * @hwstamp_shift:    Hardware tick to nanosecond divisor (power of two).
  * @capabilities:     Device capabilities
+ * @words:            word access to struct contents
  */
 union ionic_dev_identity {
 	struct {
@@ -317,6 +331,7 @@ enum ionic_lif_type {
  * @opcode:  opcode
  * @type:    LIF type (enum ionic_lif_type)
  * @ver:     Version of identify returned by device
+ * @rsvd:    reserved byte(s)
  */
 struct ionic_lif_identify_cmd {
 	u8 opcode;
@@ -329,6 +344,7 @@ struct ionic_lif_identify_cmd {
  * struct ionic_lif_identify_comp - LIF identify command completion
  * @status:  Status of the command (enum ionic_status_code)
  * @ver:     Version of identify returned by device
+ * @rsvd2:   reserved byte(s)
  */
 struct ionic_lif_identify_comp {
 	u8 status;
@@ -416,7 +432,7 @@ enum ionic_txq_feature {
 };
 
 /**
- * struct ionic_hwstamp_bits - Hardware timestamp decoding bits
+ * enum ionic_hwstamp_bits - Hardware timestamp decoding bits
  * @IONIC_HWSTAMP_INVALID:          Invalid hardware timestamp value
  * @IONIC_HWSTAMP_CQ_NEGOFFSET:     Timestamp field negative offset
  *                                  from the base cq descriptor.
@@ -429,6 +445,7 @@ enum ionic_hwstamp_bits {
 /**
  * struct ionic_lif_logical_qtype - Descriptor of logical to HW queue type
  * @qtype:          Hardware Queue Type
+ * @rsvd:           reserved byte(s)
  * @qid_count:      Number of Queue IDs of the logical type
  * @qid_base:       Minimum Queue ID of the logical type
  */
@@ -454,12 +471,14 @@ enum ionic_lif_state {
 /**
  * union ionic_lif_config - LIF configuration
  * @state:          LIF state (enum ionic_lif_state)
+ * @rsvd:           reserved byte(s)
  * @name:           LIF name
  * @mtu:            MTU
  * @mac:            Station MAC address
  * @vlan:           Default Vlan ID
  * @features:       Features (enum ionic_eth_hw_features)
  * @queue_count:    Queue counts per queue-type
+ * @words:          word access to struct contents
  */
 union ionic_lif_config {
 	struct {
@@ -481,33 +500,39 @@ union ionic_lif_config {
  * @capabilities:        LIF capabilities
  *
  * @eth:                    Ethernet identify structure
- *     @version:            Ethernet identify structure version
- *     @max_ucast_filters:  Number of perfect unicast addresses supported
- *     @max_mcast_filters:  Number of perfect multicast addresses supported
- *     @min_frame_size:     Minimum size of frames to be sent
- *     @max_frame_size:     Maximum size of frames to be sent
- *     @hwstamp_tx_modes:   Bitmask of BIT_ULL(enum ionic_txstamp_mode)
- *     @hwstamp_rx_filters: Bitmask of enum ionic_pkt_class
- *     @config:             LIF config struct with features, mtu, mac, q counts
+ *	@eth.version:            Ethernet identify structure version
+ *	@eth.rsvd:               reserved byte(s)
+ *	@eth.max_ucast_filters:  Number of perfect unicast addresses supported
+ *	@eth.max_mcast_filters:  Number of perfect multicast addresses supported
+ *	@eth.min_frame_size:     Minimum size of frames to be sent
+ *	@eth.max_frame_size:     Maximum size of frames to be sent
+ *	@eth.rsvd2:              reserved byte(s)
+ *	@eth.hwstamp_tx_modes:   Bitmask of BIT_ULL(enum ionic_txstamp_mode)
+ *	@eth.hwstamp_rx_filters: Bitmask of enum ionic_pkt_class
+ *	@eth.rsvd3:              reserved byte(s)
+ *	@eth.config:             LIF config struct with features, mtu, mac, q counts
  *
  * @rdma:                RDMA identify structure
- *     @version:         RDMA version of opcodes and queue descriptors
- *     @qp_opcodes:      Number of RDMA queue pair opcodes supported
- *     @admin_opcodes:   Number of RDMA admin opcodes supported
- *     @npts_per_lif:    Page table size per LIF
- *     @nmrs_per_lif:    Number of memory regions per LIF
- *     @nahs_per_lif:    Number of address handles per LIF
- *     @max_stride:      Max work request stride
- *     @cl_stride:       Cache line stride
- *     @pte_stride:      Page table entry stride
- *     @rrq_stride:      Remote RQ work request stride
- *     @rsq_stride:      Remote SQ work request stride
- *     @dcqcn_profiles:  Number of DCQCN profiles
- *     @aq_qtype:        RDMA Admin Qtype
- *     @sq_qtype:        RDMA Send Qtype
- *     @rq_qtype:        RDMA Receive Qtype
- *     @cq_qtype:        RDMA Completion Qtype
- *     @eq_qtype:        RDMA Event Qtype
+ *	@rdma.version:         RDMA version of opcodes and queue descriptors
+ *	@rdma.qp_opcodes:      Number of RDMA queue pair opcodes supported
+ *	@rdma.admin_opcodes:   Number of RDMA admin opcodes supported
+ *	@rdma.rsvd:            reserved byte(s)
+ *	@rdma.npts_per_lif:    Page table size per LIF
+ *	@rdma.nmrs_per_lif:    Number of memory regions per LIF
+ *	@rdma.nahs_per_lif:    Number of address handles per LIF
+ *	@rdma.max_stride:      Max work request stride
+ *	@rdma.cl_stride:       Cache line stride
+ *	@rdma.pte_stride:      Page table entry stride
+ *	@rdma.rrq_stride:      Remote RQ work request stride
+ *	@rdma.rsq_stride:      Remote SQ work request stride
+ *	@rdma.dcqcn_profiles:  Number of DCQCN profiles
+ *	@rdma.rsvd_dimensions: reserved byte(s)
+ *	@rdma.aq_qtype:        RDMA Admin Qtype
+ *	@rdma.sq_qtype:        RDMA Send Qtype
+ *	@rdma.rq_qtype:        RDMA Receive Qtype
+ *	@rdma.cq_qtype:        RDMA Completion Qtype
+ *	@rdma.eq_qtype:        RDMA Event Qtype
+ * @words:               word access to struct contents
  */
 union ionic_lif_identity {
 	struct {
@@ -558,7 +583,9 @@ union ionic_lif_identity {
  * @opcode:       Opcode
  * @type:         LIF type (enum ionic_lif_type)
  * @index:        LIF index
+ * @rsvd:         reserved byte(s)
  * @info_pa:      Destination address for LIF info (struct ionic_lif_info)
+ * @rsvd2:        reserved byte(s)
  */
 struct ionic_lif_init_cmd {
 	u8     opcode;
@@ -572,7 +599,9 @@ struct ionic_lif_init_cmd {
 /**
  * struct ionic_lif_init_comp - LIF init command completion
  * @status:	Status of the command (enum ionic_status_code)
+ * @rsvd:	reserved byte(s)
  * @hw_index:	Hardware index of the initialized LIF
+ * @rsvd2:	reserved byte(s)
  */
 struct ionic_lif_init_comp {
 	u8 status;
@@ -584,9 +613,11 @@ struct ionic_lif_init_comp {
 /**
  * struct ionic_q_identify_cmd - queue identify command
  * @opcode:     opcode
+ * @rsvd:       reserved byte(s)
  * @lif_type:   LIF type (enum ionic_lif_type)
  * @type:       Logical queue type (enum ionic_logical_qtype)
  * @ver:        Highest queue type version that the driver supports
+ * @rsvd2:      reserved byte(s)
  */
 struct ionic_q_identify_cmd {
 	u8     opcode;
@@ -600,8 +631,10 @@ struct ionic_q_identify_cmd {
 /**
  * struct ionic_q_identify_comp - queue identify command completion
  * @status:     Status of the command (enum ionic_status_code)
+ * @rsvd:       reserved byte(s)
  * @comp_index: Index in the descriptor ring for which this is the completion
  * @ver:        Queue type version that can be used with FW
+ * @rsvd2:      reserved byte(s)
  */
 struct ionic_q_identify_comp {
 	u8     status;
@@ -615,12 +648,14 @@ struct ionic_q_identify_comp {
  * union ionic_q_identity - queue identity information
  *     @version:        Queue type version that can be used with FW
  *     @supported:      Bitfield of queue versions, first bit = ver 0
+ *     @rsvd:           reserved byte(s)
  *     @features:       Queue features (enum ionic_q_feature, etc)
  *     @desc_sz:        Descriptor size
  *     @comp_sz:        Completion descriptor size
  *     @sg_desc_sz:     Scatter/Gather descriptor size
  *     @max_sg_elems:   Maximum number of Scatter/Gather elements
  *     @sg_desc_stride: Number of Scatter/Gather elements per descriptor
+ *     @words:          word access to struct contents
  */
 union ionic_q_identity {
 	struct {
@@ -640,8 +675,10 @@ union ionic_q_identity {
 /**
  * struct ionic_q_init_cmd - Queue init command
  * @opcode:       opcode
+ * @rsvd:         reserved byte(s)
  * @type:         Logical queue type
  * @ver:          Queue type version
+ * @rsvd1:        reserved byte(s)
  * @lif_index:    LIF index
  * @index:        (LIF, qtype) relative admin queue index
  * @intr_index:   Interrupt control register index, or Event queue index
@@ -667,6 +704,7 @@ union ionic_q_identity {
  * @ring_base:    Queue ring base address
  * @cq_ring_base: Completion queue ring base address
  * @sg_ring_base: Scatter/Gather ring base address
+ * @rsvd2:        reserved byte(s)
  * @features:     Mask of queue features to enable, if not in the flags above.
  */
 struct ionic_q_init_cmd {
@@ -698,9 +736,11 @@ struct ionic_q_init_cmd {
 /**
  * struct ionic_q_init_comp - Queue init command completion
  * @status:     Status of the command (enum ionic_status_code)
+ * @rsvd:       reserved byte(s)
  * @comp_index: Index in the descriptor ring for which this is the completion
  * @hw_index:   Hardware Queue ID
  * @hw_type:    Hardware Queue type
+ * @rsvd2:      reserved byte(s)
  * @color:      Color
  */
 struct ionic_q_init_comp {
@@ -800,7 +840,7 @@ enum ionic_txq_desc_opcode {
  *                      will set CWR flag in the first segment if
  *                      CWR is set in the template header, and
  *                      clear CWR in remaining segments.
- * @flags:
+ *    flags:
  *                vlan:
  *                    Insert an L2 VLAN header using @vlan_tci
  *                encap:
@@ -813,13 +853,14 @@ enum ionic_txq_desc_opcode {
  *                    TSO start
  *                tso_eot:
  *                    TSO end
- * @num_sg_elems: Number of scatter-gather elements in SG
+ *    num_sg_elems: Number of scatter-gather elements in SG
  *                descriptor
- * @addr:         First data buffer's DMA address
+ *    addr:       First data buffer's DMA address
  *                (Subsequent data buffers are on txq_sg_desc)
  * @len:          First data buffer's length, in bytes
  * @vlan_tci:     VLAN tag to insert in the packet (if requested
  *                by @V-bit).  Includes .1p and .1q tags
+ * @hword0:       half word padding
  * @hdr_len:      Length of packet headers, including
  *                encapsulating outer header, if applicable
  *                Valid for opcodes IONIC_TXQ_DESC_OPCODE_CALC_CSUM and
@@ -830,10 +871,12 @@ enum ionic_txq_desc_opcode {
  *                IONIC_TXQ_DESC_OPCODE_TSO, @hdr_len is up to
  *                inner-most L4 payload, so inclusive of
  *                inner-most L4 header.
+ * @hword1:       half word padding
  * @mss:          Desired MSS value for TSO; only applicable for
  *                IONIC_TXQ_DESC_OPCODE_TSO
  * @csum_start:   Offset from packet to first byte checked in L4 checksum
  * @csum_offset:  Offset from csum_start to L4 checksum field
+ * @hword2:       half word padding
  */
 struct ionic_txq_desc {
 	__le64  cmd;
@@ -901,6 +944,7 @@ static inline void decode_txq_desc_cmd(u64 cmd, u8 *opcode, u8 *flags,
  * struct ionic_txq_sg_elem - Transmit scatter-gather (SG) descriptor element
  * @addr:      DMA address of SG element data buffer
  * @len:       Length of SG element data buffer, in bytes
+ * @rsvd:      reserved byte(s)
  */
 struct ionic_txq_sg_elem {
 	__le64 addr;
@@ -927,7 +971,9 @@ struct ionic_txq_sg_desc_v1 {
 /**
  * struct ionic_txq_comp - Ethernet transmit queue completion descriptor
  * @status:     Status of the command (enum ionic_status_code)
+ * @rsvd:       reserved byte(s)
  * @comp_index: Index in the descriptor ring for which this is the completion
+ * @rsvd2:      reserved byte(s)
  * @color:      Color bit
  */
 struct ionic_txq_comp {
@@ -953,6 +999,7 @@ enum ionic_rxq_desc_opcode {
  *                      receive, including actual bytes received,
  *                      are recorded in Rx completion descriptor.
  *
+ * @rsvd:         reserved byte(s)
  * @len:          Data buffer's length, in bytes
  * @addr:         Data buffer's DMA address
  */
@@ -967,6 +1014,7 @@ struct ionic_rxq_desc {
  * struct ionic_rxq_sg_elem - Receive scatter-gather (SG) descriptor element
  * @addr:      DMA address of SG element data buffer
  * @len:       Length of SG element data buffer, in bytes
+ * @rsvd:      reserved byte(s)
  */
 struct ionic_rxq_sg_elem {
 	__le64 addr;
@@ -1170,6 +1218,7 @@ enum ionic_pkt_class {
  * @lif_index:  LIF index
  * @index:      Queue index
  * @oper:       Operation (enum ionic_q_control_oper)
+ * @rsvd:       reserved byte(s)
  */
 struct ionic_q_control_cmd {
 	u8     opcode;
@@ -1182,7 +1231,7 @@ struct ionic_q_control_cmd {
 
 typedef struct ionic_admin_comp ionic_q_control_comp;
 
-enum q_control_oper {
+enum ionic_q_control_oper {
 	IONIC_Q_DISABLE		= 0,
 	IONIC_Q_ENABLE		= 1,
 	IONIC_Q_HANG_RESET	= 2,
@@ -1216,7 +1265,7 @@ enum ionic_xcvr_state {
 	IONIC_XCVR_STATE_SPROM_READ_ERR	 = 4,
 };
 
-/**
+/*
  * enum ionic_xcvr_pid - Supported link modes
  */
 enum ionic_xcvr_pid {
@@ -1351,6 +1400,7 @@ struct ionic_xcvr_status {
  * @fec_type:           fec type (enum ionic_port_fec_type)
  * @pause_type:         pause type (enum ionic_port_pause_type)
  * @loopback_mode:      loopback mode (enum ionic_port_loopback_mode)
+ * @words:              word access to struct contents
  */
 union ionic_port_config {
 	struct {
@@ -1382,6 +1432,7 @@ union ionic_port_config {
  * @speed:              link speed (in Mbps)
  * @link_down_count:    number of times link went from up to down
  * @fec_type:           fec type (enum ionic_port_fec_type)
+ * @rsvd:               reserved byte(s)
  * @xcvr:               transceiver status
  */
 struct ionic_port_status {
@@ -1399,6 +1450,7 @@ struct ionic_port_status {
  * @opcode:     opcode
  * @index:      port index
  * @ver:        Highest version of identify supported by driver
+ * @rsvd:       reserved byte(s)
  */
 struct ionic_port_identify_cmd {
 	u8 opcode;
@@ -1411,6 +1463,7 @@ struct ionic_port_identify_cmd {
  * struct ionic_port_identify_comp - Port identify command completion
  * @status: Status of the command (enum ionic_status_code)
  * @ver:    Version of identify returned by device
+ * @rsvd:   reserved byte(s)
  */
 struct ionic_port_identify_comp {
 	u8 status;
@@ -1422,7 +1475,9 @@ struct ionic_port_identify_comp {
  * struct ionic_port_init_cmd - Port initialization command
  * @opcode:     opcode
  * @index:      port index
+ * @rsvd:       reserved byte(s)
  * @info_pa:    destination address for port info (struct ionic_port_info)
+ * @rsvd2:      reserved byte(s)
  */
 struct ionic_port_init_cmd {
 	u8     opcode;
@@ -1435,6 +1490,7 @@ struct ionic_port_init_cmd {
 /**
  * struct ionic_port_init_comp - Port initialization command completion
  * @status: Status of the command (enum ionic_status_code)
+ * @rsvd:   reserved byte(s)
  */
 struct ionic_port_init_comp {
 	u8 status;
@@ -1445,6 +1501,7 @@ struct ionic_port_init_comp {
  * struct ionic_port_reset_cmd - Port reset command
  * @opcode:     opcode
  * @index:      port index
+ * @rsvd:       reserved byte(s)
  */
 struct ionic_port_reset_cmd {
 	u8 opcode;
@@ -1455,6 +1512,7 @@ struct ionic_port_reset_cmd {
 /**
  * struct ionic_port_reset_comp - Port reset command completion
  * @status: Status of the command (enum ionic_status_code)
+ * @rsvd:   reserved byte(s)
  */
 struct ionic_port_reset_comp {
 	u8 status;
@@ -1510,6 +1568,7 @@ enum ionic_port_attr {
  * @opcode:         Opcode
  * @index:          Port index
  * @attr:           Attribute type (enum ionic_port_attr)
+ * @rsvd:           reserved byte(s)
  * @state:          Port state
  * @speed:          Port speed
  * @mtu:            Port MTU
@@ -1518,6 +1577,7 @@ enum ionic_port_attr {
  * @pause_type:     Port pause type setting
  * @loopback_mode:  Port loopback mode
  * @stats_ctl:      Port stats setting
+ * @rsvd2:          reserved byte(s)
  */
 struct ionic_port_setattr_cmd {
 	u8     opcode;
@@ -1540,6 +1600,7 @@ struct ionic_port_setattr_cmd {
 /**
  * struct ionic_port_setattr_comp - Port set attr command completion
  * @status:     Status of the command (enum ionic_status_code)
+ * @rsvd:       reserved byte(s)
  * @color:      Color bit
  */
 struct ionic_port_setattr_comp {
@@ -1553,6 +1614,7 @@ struct ionic_port_setattr_comp {
  * @opcode:     Opcode
  * @index:      port index
  * @attr:       Attribute type (enum ionic_port_attr)
+ * @rsvd:       reserved byte(s)
  */
 struct ionic_port_getattr_cmd {
 	u8     opcode;
@@ -1564,6 +1626,7 @@ struct ionic_port_getattr_cmd {
 /**
  * struct ionic_port_getattr_comp - Port get attr command completion
  * @status:         Status of the command (enum ionic_status_code)
+ * @rsvd:           reserved byte(s)
  * @state:          Port state
  * @speed:          Port speed
  * @mtu:            Port MTU
@@ -1571,6 +1634,7 @@ struct ionic_port_getattr_cmd {
  * @fec_type:       Port FEC type setting
  * @pause_type:     Port pause type setting
  * @loopback_mode:  Port loopback mode
+ * @rsvd2:          reserved byte(s)
  * @color:          Color bit
  */
 struct ionic_port_getattr_comp {
@@ -1593,9 +1657,11 @@ struct ionic_port_getattr_comp {
  * struct ionic_lif_status - LIF status register
  * @eid:             most recent NotifyQ event id
  * @port_num:        port the LIF is connected to
+ * @rsvd:            reserved byte(s)
  * @link_status:     port status (enum ionic_port_oper_status)
  * @link_speed:      speed of link in Mbps
  * @link_down_count: number of times link went from up to down
+ * @rsvd2:           reserved byte(s)
  */
 struct ionic_lif_status {
 	__le64 eid;
@@ -1610,7 +1676,9 @@ struct ionic_lif_status {
 /**
  * struct ionic_lif_reset_cmd - LIF reset command
  * @opcode:    opcode
+ * @rsvd:      reserved byte(s)
  * @index:     LIF index
+ * @rsvd2:     reserved byte(s)
  */
 struct ionic_lif_reset_cmd {
 	u8     opcode;
@@ -1643,9 +1711,11 @@ enum ionic_dev_attr {
  * struct ionic_dev_setattr_cmd - Set Device attributes on the NIC
  * @opcode:     Opcode
  * @attr:       Attribute type (enum ionic_dev_attr)
+ * @rsvd:       reserved byte(s)
  * @state:      Device state (enum ionic_dev_state)
  * @name:       The bus info, e.g. PCI slot-device-function, 0 terminated
  * @features:   Device features
+ * @rsvd2:      reserved byte(s)
  */
 struct ionic_dev_setattr_cmd {
 	u8     opcode;
@@ -1662,7 +1732,9 @@ struct ionic_dev_setattr_cmd {
 /**
  * struct ionic_dev_setattr_comp - Device set attr command completion
  * @status:     Status of the command (enum ionic_status_code)
+ * @rsvd:       reserved byte(s)
  * @features:   Device features
+ * @rsvd2:      reserved byte(s)
  * @color:      Color bit
  */
 struct ionic_dev_setattr_comp {
@@ -1679,6 +1751,7 @@ struct ionic_dev_setattr_comp {
  * struct ionic_dev_getattr_cmd - Get Device attributes from the NIC
  * @opcode:     opcode
  * @attr:       Attribute type (enum ionic_dev_attr)
+ * @rsvd:       reserved byte(s)
  */
 struct ionic_dev_getattr_cmd {
 	u8     opcode;
@@ -1687,9 +1760,11 @@ struct ionic_dev_getattr_cmd {
 };
 
 /**
- * struct ionic_dev_setattr_comp - Device set attr command completion
+ * struct ionic_dev_getattr_comp - Device set attr command completion
  * @status:     Status of the command (enum ionic_status_code)
+ * @rsvd:       reserved byte(s)
  * @features:   Device features
+ * @rsvd2:      reserved byte(s)
  * @color:      Color bit
  */
 struct ionic_dev_getattr_comp {
@@ -1702,7 +1777,7 @@ struct ionic_dev_getattr_comp {
 	u8     color;
 };
 
-/**
+/*
  * RSS parameters
  */
 #define IONIC_RSS_HASH_KEY_SIZE		40
@@ -1726,6 +1801,7 @@ enum ionic_rss_hash_types {
  * @IONIC_LIF_ATTR_RSS:         LIF RSS attribute
  * @IONIC_LIF_ATTR_STATS_CTRL:  LIF statistics control attribute
  * @IONIC_LIF_ATTR_TXSTAMP:     LIF TX timestamping mode
+ * @IONIC_LIF_ATTR_MAX:         maximum attribute value
  */
 enum ionic_lif_attr {
 	IONIC_LIF_ATTR_STATE        = 0,
@@ -1736,6 +1812,7 @@ enum ionic_lif_attr {
 	IONIC_LIF_ATTR_RSS          = 5,
 	IONIC_LIF_ATTR_STATS_CTRL   = 6,
 	IONIC_LIF_ATTR_TXSTAMP      = 7,
+	IONIC_LIF_ATTR_MAX          = 255,
 };
 
 /**
@@ -1749,11 +1826,13 @@ enum ionic_lif_attr {
  * @mac:        Station mac
  * @features:   Features (enum ionic_eth_hw_features)
  * @rss:        RSS properties
- *              @types:     The hash types to enable (see rss_hash_types)
- *              @key:       The hash secret key
- *              @addr:      Address for the indirection table shared memory
+ *	@rss.types:     The hash types to enable (see rss_hash_types)
+ *	@rss.key:       The hash secret key
+ *	@rss.rsvd:      reserved byte(s)
+ *	@rss.addr:      Address for the indirection table shared memory
  * @stats_ctl:  stats control commands (enum ionic_stats_ctl_cmd)
- * @txstamp:    TX Timestamping Mode (enum ionic_txstamp_mode)
+ * @txstamp_mode:    TX Timestamping Mode (enum ionic_txstamp_mode)
+ * @rsvd:        reserved byte(s)
  */
 struct ionic_lif_setattr_cmd {
 	u8     opcode;
@@ -1772,7 +1851,7 @@ struct ionic_lif_setattr_cmd {
 			__le64 addr;
 		} rss;
 		u8      stats_ctl;
-		__le16 txstamp_mode;
+		__le16  txstamp_mode;
 		u8      rsvd[60];
 	} __packed;
 };
@@ -1780,8 +1859,10 @@ struct ionic_lif_setattr_cmd {
 /**
  * struct ionic_lif_setattr_comp - LIF set attr command completion
  * @status:     Status of the command (enum ionic_status_code)
+ * @rsvd:       reserved byte(s)
  * @comp_index: Index in the descriptor ring for which this is the completion
  * @features:   features (enum ionic_eth_hw_features)
+ * @rsvd2:      reserved byte(s)
  * @color:      Color bit
  */
 struct ionic_lif_setattr_comp {
@@ -1800,6 +1881,7 @@ struct ionic_lif_setattr_comp {
  * @opcode:     Opcode
  * @attr:       Attribute type (enum ionic_lif_attr)
  * @index:      LIF index
+ * @rsvd:       reserved byte(s)
  */
 struct ionic_lif_getattr_cmd {
 	u8     opcode;
@@ -1811,13 +1893,14 @@ struct ionic_lif_getattr_cmd {
 /**
  * struct ionic_lif_getattr_comp - LIF get attr command completion
  * @status:     Status of the command (enum ionic_status_code)
+ * @rsvd:       reserved byte(s)
  * @comp_index: Index in the descriptor ring for which this is the completion
  * @state:      LIF state (enum ionic_lif_state)
- * @name:       The netdev name string, 0 terminated
  * @mtu:        Mtu
  * @mac:        Station mac
  * @features:   Features (enum ionic_eth_hw_features)
- * @txstamp:    TX Timestamping Mode (enum ionic_txstamp_mode)
+ * @txstamp_mode:    TX Timestamping Mode (enum ionic_txstamp_mode)
+ * @rsvd2:      reserved byte(s)
  * @color:      Color bit
  */
 struct ionic_lif_getattr_comp {
@@ -1838,12 +1921,15 @@ struct ionic_lif_getattr_comp {
 /**
  * struct ionic_lif_setphc_cmd - Set LIF PTP Hardware Clock
  * @opcode:     Opcode
+ * @rsvd1:      reserved byte(s)
  * @lif_index:  LIF index
+ * @rsvd2:      reserved byte(s)
  * @tick:       Hardware stamp tick of an instant in time.
  * @nsec:       Nanosecond stamp of the same instant.
  * @frac:       Fractional nanoseconds at the same instant.
  * @mult:       Cycle to nanosecond multiplier.
  * @shift:      Cycle to nanosecond divisor (power of two).
+ * @rsvd3:      reserved byte(s)
  */
 struct ionic_lif_setphc_cmd {
 	u8	opcode;
@@ -1870,6 +1956,7 @@ enum ionic_rx_mode {
 /**
  * struct ionic_rx_mode_set_cmd - Set LIF's Rx mode command
  * @opcode:     opcode
+ * @rsvd:       reserved byte(s)
  * @lif_index:  LIF index
  * @rx_mode:    Rx mode flags:
  *                  IONIC_RX_MODE_F_UNICAST: Accept known unicast packets
@@ -1878,6 +1965,7 @@ enum ionic_rx_mode {
  *                  IONIC_RX_MODE_F_PROMISC: Accept any packets
  *                  IONIC_RX_MODE_F_ALLMULTI: Accept any multicast packets
  *                  IONIC_RX_MODE_F_RDMA_SNIFFER: Sniff RDMA packets
+ * @rsvd2:      reserved byte(s)
  */
 struct ionic_rx_mode_set_cmd {
 	u8     opcode;
@@ -1904,13 +1992,14 @@ enum ionic_rx_filter_match_type {
  * @qid:        Queue ID
  * @match:      Rx filter match type (see IONIC_RX_FILTER_MATCH_xxx)
  * @vlan:       VLAN filter
- *              @vlan:  VLAN ID
+ *	@vlan.vlan:  VLAN ID
  * @mac:        MAC filter
- *              @addr:  MAC address (network-byte order)
+ *	@mac.addr:  MAC address (network-byte order)
  * @mac_vlan:   MACVLAN filter
- *              @vlan:  VLAN ID
- *              @addr:  MAC address (network-byte order)
+ *	@mac_vlan.vlan:  VLAN ID
+ *	@mac_vlan.addr:  MAC address (network-byte order)
  * @pkt_class:  Packet classification filter
+ * @rsvd:       reserved byte(s)
  */
 struct ionic_rx_filter_add_cmd {
 	u8     opcode;
@@ -1937,8 +2026,10 @@ struct ionic_rx_filter_add_cmd {
 /**
  * struct ionic_rx_filter_add_comp - Add LIF Rx filter command completion
  * @status:     Status of the command (enum ionic_status_code)
+ * @rsvd:       reserved byte(s)
  * @comp_index: Index in the descriptor ring for which this is the completion
  * @filter_id:  Filter ID
+ * @rsvd2:      reserved byte(s)
  * @color:      Color bit
  */
 struct ionic_rx_filter_add_comp {
@@ -1953,8 +2044,10 @@ struct ionic_rx_filter_add_comp {
 /**
  * struct ionic_rx_filter_del_cmd - Delete LIF Rx filter command
  * @opcode:     opcode
+ * @rsvd:       reserved byte(s)
  * @lif_index:  LIF index
  * @filter_id:  Filter ID
+ * @rsvd2:      reserved byte(s)
  */
 struct ionic_rx_filter_del_cmd {
 	u8     opcode;
@@ -2000,6 +2093,7 @@ enum ionic_vf_link_status {
  *	@trust:		enable VF trust
  *	@linkstate:	set link up or down
  *	@stats_pa:	set DMA address for VF stats
+ *	@pad:           reserved byte(s)
  */
 struct ionic_vf_setattr_cmd {
 	u8     opcode;
@@ -2031,6 +2125,7 @@ struct ionic_vf_setattr_comp {
  * @opcode:     Opcode
  * @attr:       Attribute type (enum ionic_vf_attr)
  * @vf_index:   VF index
+ * @rsvd:       reserved byte(s)
  */
 struct ionic_vf_getattr_cmd {
 	u8     opcode;
@@ -2064,8 +2159,8 @@ enum ionic_vf_ctrl_opcode {
 /**
  * struct ionic_vf_ctrl_cmd - VF control command
  * @opcode:         Opcode for the command
- * @vf_index:       VF Index. It is unused if op START_ALL is used.
  * @ctrl_opcode:    VF control operation type
+ * @vf_index:       VF Index. It is unused if op START_ALL is used.
  */
 struct ionic_vf_ctrl_cmd {
 	u8	opcode;
@@ -2089,7 +2184,7 @@ struct ionic_vf_ctrl_comp {
  * struct ionic_qos_identify_cmd - QoS identify command
  * @opcode:  opcode
  * @ver:     Highest version of identify supported by driver
- *
+ * @rsvd:    reserved byte(s)
  */
 struct ionic_qos_identify_cmd {
 	u8 opcode;
@@ -2101,6 +2196,7 @@ struct ionic_qos_identify_cmd {
  * struct ionic_qos_identify_comp - QoS identify command completion
  * @status: Status of the command (enum ionic_status_code)
  * @ver:    Version of identify returned by device
+ * @rsvd:   reserved byte(s)
  */
 struct ionic_qos_identify_comp {
 	u8 status;
@@ -2118,7 +2214,7 @@ struct ionic_qos_identify_comp {
 #define IONIC_QOS_ALL_PCP		0xFF
 #define IONIC_DSCP_BLOCK_SIZE		8
 
-/**
+/*
  * enum ionic_qos_class
  */
 enum ionic_qos_class {
@@ -2174,6 +2270,7 @@ enum ionic_qos_sched_type {
  * @dot1q_pcp:		Dot1q pcp value
  * @ndscp:		Number of valid dscp values in the ip_dscp field
  * @ip_dscp:		IP dscp values
+ * @words:		word access to struct contents
  */
 union ionic_qos_config {
 	struct {
@@ -2219,8 +2316,9 @@ union ionic_qos_config {
  * union ionic_qos_identity - QoS identity structure
  * @version:	Version of the identify structure
  * @type:	QoS system type
- * @nclasses:	Number of usable QoS classes
+ * @rsvd:	reserved byte(s)
  * @config:	Current configuration of classes
+ * @words:	word access to struct contents
  */
 union ionic_qos_identity {
 	struct {
@@ -2236,7 +2334,9 @@ union ionic_qos_identity {
  * struct ionic_qos_init_cmd - QoS config init command
  * @opcode:	Opcode
  * @group:	QoS class id
+ * @rsvd:	reserved byte(s)
  * @info_pa:	destination address for qos info
+ * @rsvd1:	reserved byte(s)
  */
 struct ionic_qos_init_cmd {
 	u8     opcode;
@@ -2252,6 +2352,7 @@ typedef struct ionic_admin_comp ionic_qos_init_comp;
  * struct ionic_qos_reset_cmd - QoS config reset command
  * @opcode:	Opcode
  * @group:	QoS class id
+ * @rsvd:	reserved byte(s)
  */
 struct ionic_qos_reset_cmd {
 	u8    opcode;
@@ -2260,8 +2361,10 @@ struct ionic_qos_reset_cmd {
 };
 
 /**
- * struct ionic_qos_clear_port_stats_cmd - Qos config reset command
+ * struct ionic_qos_clear_stats_cmd - Qos config reset command
  * @opcode:	Opcode
+ * @group_bitmap: bitmap of groups to be cleared
+ * @rsvd:	reserved byte(s)
  */
 struct ionic_qos_clear_stats_cmd {
 	u8    opcode;
@@ -2274,6 +2377,7 @@ typedef struct ionic_admin_comp ionic_qos_reset_comp;
 /**
  * struct ionic_fw_download_cmd - Firmware download command
  * @opcode:	opcode
+ * @rsvd:	reserved byte(s)
  * @addr:	dma address of the firmware buffer
  * @offset:	offset of the firmware buffer within the full image
  * @length:	number of valid bytes in the firmware buffer
@@ -2297,6 +2401,7 @@ typedef struct ionic_admin_comp ionic_fw_download_comp;
  * @IONIC_FW_INSTALL_STATUS:	Firmware installation status
  * @IONIC_FW_ACTIVATE_ASYNC:	Activate firmware asynchronously
  * @IONIC_FW_ACTIVATE_STATUS:	Firmware activate status
+ * @IONIC_FW_UPDATE_CLEANUP:	Clean up after an interrupted fw update
  */
 enum ionic_fw_control_oper {
 	IONIC_FW_RESET			= 0,
@@ -2312,8 +2417,10 @@ enum ionic_fw_control_oper {
 /**
  * struct ionic_fw_control_cmd - Firmware control command
  * @opcode:    opcode
+ * @rsvd:      reserved byte(s)
  * @oper:      firmware control operation (enum ionic_fw_control_oper)
  * @slot:      slot to activate
+ * @rsvd1:     reserved byte(s)
  */
 struct ionic_fw_control_cmd {
 	u8  opcode;
@@ -2326,8 +2433,10 @@ struct ionic_fw_control_cmd {
 /**
  * struct ionic_fw_control_comp - Firmware control copletion
  * @status:     Status of the command (enum ionic_status_code)
+ * @rsvd:       reserved byte(s)
  * @comp_index: Index in the descriptor ring for which this is the completion
  * @slot:       Slot where the firmware was installed
+ * @rsvd1:      reserved byte(s)
  * @color:      Color bit
  */
 struct ionic_fw_control_comp {
@@ -2346,7 +2455,9 @@ struct ionic_fw_control_comp {
 /**
  * struct ionic_rdma_reset_cmd - Reset RDMA LIF cmd
  * @opcode:        opcode
+ * @rsvd:          reserved byte(s)
  * @lif_index:     LIF index
+ * @rsvd2:         reserved byte(s)
  *
  * There is no RDMA specific dev command completion struct.  Completion uses
  * the common struct ionic_admin_comp.  Only the status is indicated.
@@ -2362,6 +2473,7 @@ struct ionic_rdma_reset_cmd {
 /**
  * struct ionic_rdma_queue_cmd - Create RDMA Queue command
  * @opcode:        opcode, 52, 53
+ * @rsvd:          reserved byte(s)
  * @lif_index:     LIF index
  * @qid_ver:       (qid | (RDMA version << 24))
  * @cid:           intr, eq_id, or cq_id
@@ -2369,6 +2481,7 @@ struct ionic_rdma_reset_cmd {
  * @depth_log2:    log base two of queue depth
  * @stride_log2:   log base two of queue stride
  * @dma_addr:      address of the queue memory
+ * @rsvd2:         reserved byte(s)
  *
  * The same command struct is used to create an RDMA event queue, completion
  * queue, or RDMA admin queue.  The cid is an interrupt number for an event
@@ -2425,6 +2538,7 @@ struct ionic_notifyq_event {
  * @ecode:		event code = IONIC_EVENT_LINK_CHANGE
  * @link_status:	link up/down, with error bits (enum ionic_port_status)
  * @link_speed:		speed of the network link
+ * @rsvd:		reserved byte(s)
  *
  * Sent when the network link state changes between UP and DOWN
  */
@@ -2442,6 +2556,7 @@ struct ionic_link_change_event {
  * @ecode:		event code = IONIC_EVENT_RESET
  * @reset_code:		reset type
  * @state:		0=pending, 1=complete, 2=error
+ * @rsvd:		reserved byte(s)
  *
  * Sent when the NIC or some subsystem is going to be or
  * has been reset.
@@ -2458,6 +2573,7 @@ struct ionic_reset_event {
  * struct ionic_heartbeat_event - Sent periodically by NIC to indicate health
  * @eid:	event number
  * @ecode:	event code = IONIC_EVENT_HEARTBEAT
+ * @rsvd:	reserved byte(s)
  */
 struct ionic_heartbeat_event {
 	__le64 eid;
@@ -2481,6 +2597,7 @@ struct ionic_log_event {
  * struct ionic_xcvr_event - Transceiver change event
  * @eid:	event number
  * @ecode:	event code = IONIC_EVENT_XCVR
+ * @rsvd:	reserved byte(s)
  */
 struct ionic_xcvr_event {
 	__le64 eid;
@@ -2488,7 +2605,7 @@ struct ionic_xcvr_event {
 	u8     rsvd[54];
 };
 
-/**
+/*
  * struct ionic_port_stats - Port statistics structure
  */
 struct ionic_port_stats {
@@ -2646,8 +2763,7 @@ enum ionic_oflow_drop_stats {
 	IONIC_OFLOW_DROP_MAX,
 };
 
-/**
- * struct port_pb_stats - packet buffers system stats
+/* struct ionic_port_pb_stats - packet buffers system stats
  * uses ionic_pb_buffer_drop_stats for drop_counts[]
  */
 struct ionic_port_pb_stats {
@@ -2681,7 +2797,9 @@ struct ionic_port_pb_stats {
  * @pause_type:     supported pause types
  * @loopback_mode:  supported loopback mode
  * @speeds:         supported speeds
+ * @rsvd2:          reserved byte(s)
  * @config:         current port configuration
+ * @words:          word access to struct contents
  */
 union ionic_port_identity {
 	struct {
@@ -2707,7 +2825,8 @@ union ionic_port_identity {
  * @status:          Port status data
  * @stats:           Port statistics data
  * @mgmt_stats:      Port management statistics data
- * @port_pb_drop_stats:   uplink pb drop stats
+ * @rsvd:            reserved byte(s)
+ * @pb_stats:        uplink pb drop stats
  */
 struct ionic_port_info {
 	union ionic_port_config config;
@@ -2721,7 +2840,7 @@ struct ionic_port_info {
 	struct ionic_port_pb_stats  pb_stats;
 };
 
-/**
+/*
  * struct ionic_lif_stats - LIF statistics structure
  */
 struct ionic_lif_stats {
@@ -2983,8 +3102,10 @@ struct ionic_hwstamp_regs {
  *			bit 4-7 - 4 bit generation number, changes on fw restart
  * @fw_heartbeat:    Firmware heartbeat counter
  * @serial_num:      Serial number
+ * @rsvd_pad1024:    reserved byte(s)
  * @fw_version:      Firmware version
- * @hwstamp_regs:    Hardware current timestamp registers
+ * @hwstamp:         Hardware current timestamp registers
+ * @words:           word access to struct contents
  */
 union ionic_dev_info_regs {
 #define IONIC_DEVINFO_FWVERS_BUFLEN 32
@@ -3014,7 +3135,9 @@ union ionic_dev_info_regs {
  * @done:            Done indicator, bit 0 == 1 when command is complete
  * @cmd:             Opcode-specific command bytes
  * @comp:            Opcode-specific response bytes
+ * @rsvd:            reserved byte(s)
  * @data:            Opcode-specific side-data
+ * @words:           word access to struct contents
  */
 union ionic_dev_cmd_regs {
 	struct {
@@ -3032,6 +3155,7 @@ union ionic_dev_cmd_regs {
  * union ionic_dev_regs - Device register format for bar 0 page 0
  * @info:            Device info registers
  * @devcmd:          Device command registers
+ * @words:           word access to struct contents
  */
 union ionic_dev_regs {
 	struct {
@@ -3098,6 +3222,7 @@ union ionic_adminq_comp {
  *              interrupts when armed.
  * @qid_lo:  Queue destination for the producer index and flags (low bits)
  * @qid_hi:  Queue destination for the producer index and flags (high bits)
+ * @rsvd2:   reserved byte(s)
  */
 struct ionic_doorbell {
 	__le16 p_index;
-- 
cgit v1.2.3-58-ga151


From 57e3c5af2befaf54cbae326fe800c148852e67a1 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 28 May 2024 19:33:22 -0700
Subject: net: fjes: correct TRACE_INCLUDE_PATH
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A comment in define_trace.h clearly states:

 TRACE_INCLUDE_PATH if the path is something other than core kernel
                                                           vvvvvvvvvvvvvv
 include/trace then this macro can define the path to use. Note, the path
 is relative to define_trace.h, not the file including it. Full path names
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 for out of tree modules must be used.

fjes uses path relative to itself. Which (somehow) works most of
the time. Except when the kernel tree is "nested" in another
kernel tree, and ../drivers/net/fjes actually exists. In which
case build will use the header file from the wrong directory.

I've been trying to figure out why net NIPA builder is constantly
failing for the last 5 days, with:

include/trace/../../../drivers/net/fjes/fjes_trace.h:88:17: error: ‘__assign_str’ undeclared (first use in this function)
   88 |                 __assign_str(err, err);
      |                 ^~~~~~~~~~~~

when the line in the tree clearly has only one "err". NIPA does
indeed have "nested" trees, because it uses git work-trees and
the tree on the "outside" is not very up to date.

Link: https://lore.kernel.org/r/20240529023322.3467755-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/fjes/fjes_trace.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/fjes/fjes_trace.h b/drivers/net/fjes/fjes_trace.h
index 166ef015262b..37c6071cb333 100644
--- a/drivers/net/fjes/fjes_trace.h
+++ b/drivers/net/fjes/fjes_trace.h
@@ -358,7 +358,7 @@ TRACE_EVENT(fjes_stop_req_irq_post,
 
 #undef TRACE_INCLUDE_PATH
 #undef TRACE_INCLUDE_FILE
-#define TRACE_INCLUDE_PATH ../../../drivers/net/fjes
+#define TRACE_INCLUDE_PATH ../../drivers/net/fjes
 #define TRACE_INCLUDE_FILE fjes_trace
 
 /* This part must be outside protection */
-- 
cgit v1.2.3-58-ga151


From a79d8fe2ff8e78e549dc86cc853a61b029404871 Mon Sep 17 00:00:00 2001
From: Hangbin Liu <liuhangbin@gmail.com>
Date: Wed, 29 May 2024 12:09:08 +0800
Subject: ipv6: sr: restruct ifdefines

There are too many ifdef in IPv6 segment routing code that may cause logic
problems. like commit 160e9d275218 ("ipv6: sr: fix invalid unregister error
path"). To avoid this, the init functions are redefined for both cases. The
code could be more clear after all fidefs are removed.

Suggested-by: Simon Horman <horms@kernel.org>
Suggested-by: David Ahern <dsahern@kernel.org>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240529040908.3472952-1-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/seg6.h      |  7 +++++++
 include/net/seg6_hmac.h |  7 +++++++
 net/ipv6/seg6.c         | 33 +++++----------------------------
 3 files changed, 19 insertions(+), 28 deletions(-)

diff --git a/include/net/seg6.h b/include/net/seg6.h
index af668f17b398..82b3fbbcbb93 100644
--- a/include/net/seg6.h
+++ b/include/net/seg6.h
@@ -52,10 +52,17 @@ static inline struct seg6_pernet_data *seg6_pernet(struct net *net)
 
 extern int seg6_init(void);
 extern void seg6_exit(void);
+#ifdef CONFIG_IPV6_SEG6_LWTUNNEL
 extern int seg6_iptunnel_init(void);
 extern void seg6_iptunnel_exit(void);
 extern int seg6_local_init(void);
 extern void seg6_local_exit(void);
+#else
+static inline int seg6_iptunnel_init(void) { return 0; }
+static inline void seg6_iptunnel_exit(void) {}
+static inline int seg6_local_init(void) { return 0; }
+static inline void seg6_local_exit(void) {}
+#endif
 
 extern bool seg6_validate_srh(struct ipv6_sr_hdr *srh, int len, bool reduced);
 extern struct ipv6_sr_hdr *seg6_get_srh(struct sk_buff *skb, int flags);
diff --git a/include/net/seg6_hmac.h b/include/net/seg6_hmac.h
index 2b5d2ee5613e..24f733b3e3fe 100644
--- a/include/net/seg6_hmac.h
+++ b/include/net/seg6_hmac.h
@@ -49,9 +49,16 @@ extern int seg6_hmac_info_del(struct net *net, u32 key);
 extern int seg6_push_hmac(struct net *net, struct in6_addr *saddr,
 			  struct ipv6_sr_hdr *srh);
 extern bool seg6_hmac_validate_skb(struct sk_buff *skb);
+#ifdef CONFIG_IPV6_SEG6_HMAC
 extern int seg6_hmac_init(void);
 extern void seg6_hmac_exit(void);
 extern int seg6_hmac_net_init(struct net *net);
 extern void seg6_hmac_net_exit(struct net *net);
+#else
+static inline int seg6_hmac_init(void) { return 0; }
+static inline void seg6_hmac_exit(void) {}
+static inline int seg6_hmac_net_init(struct net *net) { return 0; }
+static inline void seg6_hmac_net_exit(struct net *net) {}
+#endif
 
 #endif
diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c
index a31521e270f7..180da19c148c 100644
--- a/net/ipv6/seg6.c
+++ b/net/ipv6/seg6.c
@@ -21,9 +21,7 @@
 #include <net/genetlink.h>
 #include <linux/seg6.h>
 #include <linux/seg6_genl.h>
-#ifdef CONFIG_IPV6_SEG6_HMAC
 #include <net/seg6_hmac.h>
-#endif
 
 bool seg6_validate_srh(struct ipv6_sr_hdr *srh, int len, bool reduced)
 {
@@ -437,13 +435,11 @@ static int __net_init seg6_net_init(struct net *net)
 
 	net->ipv6.seg6_data = sdata;
 
-#ifdef CONFIG_IPV6_SEG6_HMAC
 	if (seg6_hmac_net_init(net)) {
 		kfree(rcu_dereference_raw(sdata->tun_src));
 		kfree(sdata);
 		return -ENOMEM;
 	}
-#endif
 
 	return 0;
 }
@@ -452,9 +448,7 @@ static void __net_exit seg6_net_exit(struct net *net)
 {
 	struct seg6_pernet_data *sdata = seg6_pernet(net);
 
-#ifdef CONFIG_IPV6_SEG6_HMAC
 	seg6_hmac_net_exit(net);
-#endif
 
 	kfree(rcu_dereference_raw(sdata->tun_src));
 	kfree(sdata);
@@ -520,41 +514,28 @@ int __init seg6_init(void)
 	if (err)
 		goto out_unregister_pernet;
 
-#ifdef CONFIG_IPV6_SEG6_LWTUNNEL
 	err = seg6_iptunnel_init();
 	if (err)
 		goto out_unregister_genl;
 
 	err = seg6_local_init();
-	if (err) {
-		seg6_iptunnel_exit();
-		goto out_unregister_genl;
-	}
-#endif
+	if (err)
+		goto out_unregister_iptun;
 
-#ifdef CONFIG_IPV6_SEG6_HMAC
 	err = seg6_hmac_init();
 	if (err)
-		goto out_unregister_iptun;
-#endif
+		goto out_unregister_seg6;
 
 	pr_info("Segment Routing with IPv6\n");
 
 out:
 	return err;
-#ifdef CONFIG_IPV6_SEG6_HMAC
-out_unregister_iptun:
-#ifdef CONFIG_IPV6_SEG6_LWTUNNEL
+out_unregister_seg6:
 	seg6_local_exit();
+out_unregister_iptun:
 	seg6_iptunnel_exit();
-#endif
-#endif
-#ifdef CONFIG_IPV6_SEG6_LWTUNNEL
 out_unregister_genl:
-#endif
-#if IS_ENABLED(CONFIG_IPV6_SEG6_LWTUNNEL) || IS_ENABLED(CONFIG_IPV6_SEG6_HMAC)
 	genl_unregister_family(&seg6_genl_family);
-#endif
 out_unregister_pernet:
 	unregister_pernet_subsys(&ip6_segments_ops);
 	goto out;
@@ -562,13 +543,9 @@ out_unregister_pernet:
 
 void seg6_exit(void)
 {
-#ifdef CONFIG_IPV6_SEG6_HMAC
 	seg6_hmac_exit();
-#endif
-#ifdef CONFIG_IPV6_SEG6_LWTUNNEL
 	seg6_local_exit();
 	seg6_iptunnel_exit();
-#endif
 	genl_unregister_family(&seg6_genl_family);
 	unregister_pernet_subsys(&ip6_segments_ops);
 }
-- 
cgit v1.2.3-58-ga151


From 482b3c3ba757f8fb7368f208b411d2bad3004702 Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Wed, 29 May 2024 09:40:34 +0100
Subject: net: stmmac: Drop TBI/RTBI PCS flags

First of all the flags are never set by any of the driver parts. If nobody
have them set then the respective statements will always have the same
result. Thus the statements can be simplified or even dropped with no risk
to break things.

Secondly shall any of the TBI or RTBI flag is set the MDIO-bus
registration will be bypassed. Why? It really seems weird. It's perfectly
fine to have a TBI/RTBI-capable PHY configured over the MDIO bus
interface.

Based on the notes above the TBI/RTBI PCS flags can be freely dropped thus
simplifying the driver code.

Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Serge Semin <fancer.lancer@gmail.com>
Link: https://lore.kernel.org/r/E1sCErK-00EOPf-EP@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/common.h      |  2 --
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 35 ++++++-----------------
 2 files changed, 9 insertions(+), 28 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index 9cd62b2110a1..cd36ff4da68c 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -271,8 +271,6 @@ struct stmmac_safety_stats {
 /* PCS defines */
 #define STMMAC_PCS_RGMII	(1 << 0)
 #define STMMAC_PCS_SGMII	(1 << 1)
-#define STMMAC_PCS_TBI		(1 << 2)
-#define STMMAC_PCS_RTBI		(1 << 3)
 
 #define SF_DMA_MODE 1		/* DMA STORE-AND-FORWARD Operation Mode */
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index b3afc7cb7d72..3ab93f89be90 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -471,13 +471,6 @@ bool stmmac_eee_init(struct stmmac_priv *priv)
 {
 	int eee_tw_timer = priv->eee_tw_timer;
 
-	/* Using PCS we cannot dial with the phy registers at this stage
-	 * so we do not support extra feature like EEE.
-	 */
-	if (priv->hw->pcs == STMMAC_PCS_TBI ||
-	    priv->hw->pcs == STMMAC_PCS_RTBI)
-		return false;
-
 	/* Check if MAC core supports the EEE feature. */
 	if (!priv->dma_cap.eee)
 		return false;
@@ -3953,9 +3946,7 @@ static int __stmmac_open(struct net_device *dev,
 	if (ret < 0)
 		return ret;
 
-	if (priv->hw->pcs != STMMAC_PCS_TBI &&
-	    priv->hw->pcs != STMMAC_PCS_RTBI &&
-	    (!priv->hw->xpcs ||
+	if ((!priv->hw->xpcs ||
 	     xpcs_get_an_mode(priv->hw->xpcs, mode) != DW_AN_C73)) {
 		ret = stmmac_init_phy(dev);
 		if (ret) {
@@ -7739,16 +7730,12 @@ int stmmac_dvr_probe(struct device *device,
 	if (!pm_runtime_enabled(device))
 		pm_runtime_enable(device);
 
-	if (priv->hw->pcs != STMMAC_PCS_TBI &&
-	    priv->hw->pcs != STMMAC_PCS_RTBI) {
-		/* MDIO bus Registration */
-		ret = stmmac_mdio_register(ndev);
-		if (ret < 0) {
-			dev_err_probe(priv->device, ret,
-				      "%s: MDIO bus (id: %d) registration failed\n",
-				      __func__, priv->plat->bus_id);
-			goto error_mdio_register;
-		}
+	ret = stmmac_mdio_register(ndev);
+	if (ret < 0) {
+		dev_err_probe(priv->device, ret,
+			      "MDIO bus (id: %d) registration failed\n",
+			      priv->plat->bus_id);
+		goto error_mdio_register;
 	}
 
 	if (priv->plat->speed_mode_2500)
@@ -7790,9 +7777,7 @@ error_netdev_register:
 error_phy_setup:
 	stmmac_pcs_clean(ndev);
 error_pcs_setup:
-	if (priv->hw->pcs != STMMAC_PCS_TBI &&
-	    priv->hw->pcs != STMMAC_PCS_RTBI)
-		stmmac_mdio_unregister(ndev);
+	stmmac_mdio_unregister(ndev);
 error_mdio_register:
 	stmmac_napi_del(ndev);
 error_hw_init:
@@ -7833,10 +7818,8 @@ void stmmac_dvr_remove(struct device *dev)
 	reset_control_assert(priv->plat->stmmac_ahb_rst);
 
 	stmmac_pcs_clean(ndev);
+	stmmac_mdio_unregister(ndev);
 
-	if (priv->hw->pcs != STMMAC_PCS_TBI &&
-	    priv->hw->pcs != STMMAC_PCS_RTBI)
-		stmmac_mdio_unregister(ndev);
 	destroy_workqueue(priv->wq);
 	mutex_destroy(&priv->lock);
 	bitmap_free(priv->af_xdp_zc_qps);
-- 
cgit v1.2.3-58-ga151


From 4af90c0f484491a1bdccc7c3bfe9b7411868b139 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 29 May 2024 09:40:39 +0100
Subject: net: stmmac: dwxgmac2: remove useless NULL pointer initialisations

Remove useless NULL pointer initialisations for "PCS" methods from the
dwxgmac2 code.

Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Serge Semin <fancer.lancer@gmail.com>
Link: https://lore.kernel.org/r/E1sCErP-00EOPl-IT@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
index f8e7775bb633..6a987cf598e4 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_core.c
@@ -1554,9 +1554,6 @@ const struct stmmac_ops dwxgmac210_ops = {
 	.reset_eee_mode = dwxgmac2_reset_eee_mode,
 	.set_eee_timer = dwxgmac2_set_eee_timer,
 	.set_eee_pls = dwxgmac2_set_eee_pls,
-	.pcs_ctrl_ane = NULL,
-	.pcs_rane = NULL,
-	.pcs_get_adv_lp = NULL,
 	.debug = NULL,
 	.set_filter = dwxgmac2_set_filter,
 	.safety_feat_config = dwxgmac3_safety_feat_config,
@@ -1614,9 +1611,6 @@ const struct stmmac_ops dwxlgmac2_ops = {
 	.reset_eee_mode = dwxgmac2_reset_eee_mode,
 	.set_eee_timer = dwxgmac2_set_eee_timer,
 	.set_eee_pls = dwxgmac2_set_eee_pls,
-	.pcs_ctrl_ane = NULL,
-	.pcs_rane = NULL,
-	.pcs_get_adv_lp = NULL,
 	.debug = NULL,
 	.set_filter = dwxgmac2_set_filter,
 	.safety_feat_config = dwxgmac3_safety_feat_config,
-- 
cgit v1.2.3-58-ga151


From 3277407e5e8217cb7550453fe3184d0de511f377 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 29 May 2024 09:40:44 +0100
Subject: net: stmmac: remove pcs_rane() method

The pcs_rane() method is not called, so lets just remove this
redundant code.

Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Serge Semin <fancer.lancer@gmail.com>
Link: https://lore.kernel.org/r/E1sCErU-00EOPr-MC@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c |  6 ------
 drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c    |  8 --------
 drivers/net/ethernet/stmicro/stmmac/hwif.h           |  3 ---
 drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h     | 17 -----------------
 4 files changed, 34 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
index 8555299443f4..d0c7c2320d8d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
@@ -404,11 +404,6 @@ static void dwmac1000_ctrl_ane(void __iomem *ioaddr, bool ane, bool srgmi_ral,
 	dwmac_ctrl_ane(ioaddr, GMAC_PCS_BASE, ane, srgmi_ral, loopback);
 }
 
-static void dwmac1000_rane(void __iomem *ioaddr, bool restart)
-{
-	dwmac_rane(ioaddr, GMAC_PCS_BASE, restart);
-}
-
 static void dwmac1000_get_adv_lp(void __iomem *ioaddr, struct rgmii_adv *adv)
 {
 	dwmac_get_adv_lp(ioaddr, GMAC_PCS_BASE, adv);
@@ -519,7 +514,6 @@ const struct stmmac_ops dwmac1000_ops = {
 	.set_eee_pls = dwmac1000_set_eee_pls,
 	.debug = dwmac1000_debug,
 	.pcs_ctrl_ane = dwmac1000_ctrl_ane,
-	.pcs_rane = dwmac1000_rane,
 	.pcs_get_adv_lp = dwmac1000_get_adv_lp,
 	.set_mac_loopback = dwmac1000_set_mac_loopback,
 };
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
index b25774d69195..dbd9f93b2460 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c
@@ -758,11 +758,6 @@ static void dwmac4_ctrl_ane(void __iomem *ioaddr, bool ane, bool srgmi_ral,
 	dwmac_ctrl_ane(ioaddr, GMAC_PCS_BASE, ane, srgmi_ral, loopback);
 }
 
-static void dwmac4_rane(void __iomem *ioaddr, bool restart)
-{
-	dwmac_rane(ioaddr, GMAC_PCS_BASE, restart);
-}
-
 static void dwmac4_get_adv_lp(void __iomem *ioaddr, struct rgmii_adv *adv)
 {
 	dwmac_get_adv_lp(ioaddr, GMAC_PCS_BASE, adv);
@@ -1215,7 +1210,6 @@ const struct stmmac_ops dwmac4_ops = {
 	.set_eee_timer = dwmac4_set_eee_timer,
 	.set_eee_pls = dwmac4_set_eee_pls,
 	.pcs_ctrl_ane = dwmac4_ctrl_ane,
-	.pcs_rane = dwmac4_rane,
 	.pcs_get_adv_lp = dwmac4_get_adv_lp,
 	.debug = dwmac4_debug,
 	.set_filter = dwmac4_set_filter,
@@ -1260,7 +1254,6 @@ const struct stmmac_ops dwmac410_ops = {
 	.set_eee_timer = dwmac4_set_eee_timer,
 	.set_eee_pls = dwmac4_set_eee_pls,
 	.pcs_ctrl_ane = dwmac4_ctrl_ane,
-	.pcs_rane = dwmac4_rane,
 	.pcs_get_adv_lp = dwmac4_get_adv_lp,
 	.debug = dwmac4_debug,
 	.set_filter = dwmac4_set_filter,
@@ -1309,7 +1302,6 @@ const struct stmmac_ops dwmac510_ops = {
 	.set_eee_timer = dwmac4_set_eee_timer,
 	.set_eee_pls = dwmac4_set_eee_pls,
 	.pcs_ctrl_ane = dwmac4_ctrl_ane,
-	.pcs_rane = dwmac4_rane,
 	.pcs_get_adv_lp = dwmac4_get_adv_lp,
 	.debug = dwmac4_debug,
 	.set_filter = dwmac4_set_filter,
diff --git a/drivers/net/ethernet/stmicro/stmmac/hwif.h b/drivers/net/ethernet/stmicro/stmmac/hwif.h
index 90384db228b5..97934ccba5b1 100644
--- a/drivers/net/ethernet/stmicro/stmmac/hwif.h
+++ b/drivers/net/ethernet/stmicro/stmmac/hwif.h
@@ -370,7 +370,6 @@ struct stmmac_ops {
 	/* PCS calls */
 	void (*pcs_ctrl_ane)(void __iomem *ioaddr, bool ane, bool srgmi_ral,
 			     bool loopback);
-	void (*pcs_rane)(void __iomem *ioaddr, bool restart);
 	void (*pcs_get_adv_lp)(void __iomem *ioaddr, struct rgmii_adv *adv);
 	/* Safety Features */
 	int (*safety_feat_config)(void __iomem *ioaddr, unsigned int asp,
@@ -484,8 +483,6 @@ struct stmmac_ops {
 	stmmac_do_void_callback(__priv, mac, debug, __priv, __args)
 #define stmmac_pcs_ctrl_ane(__priv, __args...) \
 	stmmac_do_void_callback(__priv, mac, pcs_ctrl_ane, __args)
-#define stmmac_pcs_rane(__priv, __args...) \
-	stmmac_do_void_callback(__priv, mac, pcs_rane, __priv, __args)
 #define stmmac_pcs_get_adv_lp(__priv, __args...) \
 	stmmac_do_void_callback(__priv, mac, pcs_get_adv_lp, __args)
 #define stmmac_safety_feat_config(__priv, __args...) \
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h
index 13a30e6df4c1..1bdf87b237c4 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h
@@ -74,23 +74,6 @@ static inline void dwmac_pcs_isr(void __iomem *ioaddr, u32 reg,
 	}
 }
 
-/**
- * dwmac_rane - To restart ANE
- * @ioaddr: IO registers pointer
- * @reg: Base address of the AN Control Register.
- * @restart: to restart ANE
- * Description: this is to just restart the Auto-Negotiation.
- */
-static inline void dwmac_rane(void __iomem *ioaddr, u32 reg, bool restart)
-{
-	u32 value = readl(ioaddr + GMAC_AN_CTRL(reg));
-
-	if (restart)
-		value |= GMAC_AN_CTRL_RAN;
-
-	writel(value, ioaddr + GMAC_AN_CTRL(reg));
-}
-
 /**
  * dwmac_ctrl_ane - To program the AN Control Register.
  * @ioaddr: IO registers pointer
-- 
cgit v1.2.3-58-ga151


From aee04f4bb44d47205f4db9e84384116bb213160e Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 29 May 2024 09:40:49 +0100
Subject: net: stmmac: remove unnecessary netif_carrier_off()

It is incorrect to call netif_carrier_off(), or in fact any driver
teardown, before unregister_netdev() has been called.

unregister_netdev() unpublishes the network device from userspace, and
takes the interface down if it was up prior to returning. Therefore,
once the call has returned, we are guaranteed that .ndo_stop() will
have been called for an interface that was up. Phylink will take the
carrier down via phylink_stop(), making any manipulation of the carrier
in the remove path unnecessary.

In the stmmac_release() path, the netif_carrier_off() call follows the
call to phylink_stop(), so this call is redundant.

Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Serge Semin <fancer.lancer@gmail.com>
Link: https://lore.kernel.org/r/E1sCErZ-00EOPx-PF@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 3ab93f89be90..ca19b232431a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -4088,8 +4088,6 @@ static int stmmac_release(struct net_device *dev)
 	if (priv->plat->serdes_powerdown)
 		priv->plat->serdes_powerdown(dev, priv->plat->bsp_priv);
 
-	netif_carrier_off(dev);
-
 	stmmac_release_ptp(priv);
 
 	pm_runtime_put(priv->device);
@@ -7806,7 +7804,6 @@ void stmmac_dvr_remove(struct device *dev)
 
 	stmmac_stop_all_dma(priv);
 	stmmac_mac_set(priv, priv->ioaddr, false);
-	netif_carrier_off(ndev);
 	unregister_netdev(ndev);
 
 #ifdef CONFIG_DEBUG_FS
-- 
cgit v1.2.3-58-ga151


From cd56ff75ccfc37aab4b35b17b84d3ba08f33b933 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 29 May 2024 09:40:54 +0100
Subject: net: stmmac: include linux/io.h rather than asm/io.h

Include linux/io.h instead of asm/io.h since linux/ includes are
preferred.

Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Serge Semin <fancer.lancer@gmail.com>
Link: https://lore.kernel.org/r/E1sCEre-00EOQ3-SR@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c | 2 +-
 drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c  | 2 +-
 drivers/net/ethernet/stmicro/stmmac/dwmac100_core.c  | 2 +-
 drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c   | 2 +-
 drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 2 +-
 5 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
index d0c7c2320d8d..d413d76a8936 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c
@@ -15,7 +15,7 @@
 #include <linux/crc32.h>
 #include <linux/slab.h>
 #include <linux/ethtool.h>
-#include <asm/io.h>
+#include <linux/io.h>
 #include "stmmac.h"
 #include "stmmac_pcs.h"
 #include "dwmac1000.h"
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
index daf79cdbd3ec..adccdd816ea9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_dma.c
@@ -12,7 +12,7 @@
   Author: Giuseppe Cavallaro <peppe.cavallaro@st.com>
 *******************************************************************************/
 
-#include <asm/io.h>
+#include <linux/io.h>
 #include "dwmac1000.h"
 #include "dwmac_dma.h"
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac100_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac100_core.c
index 7667d103cd0e..14e847c0e1a9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac100_core.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac100_core.c
@@ -15,7 +15,7 @@
 *******************************************************************************/
 
 #include <linux/crc32.h>
-#include <asm/io.h>
+#include <linux/io.h>
 #include "stmmac.h"
 #include "dwmac100.h"
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
index dea270f60cc3..b402fb54f613 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac100_dma.c
@@ -14,7 +14,7 @@
   Author: Giuseppe Cavallaro <peppe.cavallaro@st.com>
 *******************************************************************************/
 
-#include <asm/io.h>
+#include <linux/io.h>
 #include "dwmac100.h"
 #include "dwmac_dma.h"
 
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
index 542e2633a6f5..18468c0228f0 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
@@ -11,10 +11,10 @@
 #include <linux/etherdevice.h>
 #include <linux/ethtool.h>
 #include <linux/interrupt.h>
+#include <linux/io.h>
 #include <linux/mii.h>
 #include <linux/phylink.h>
 #include <linux/net_tstamp.h>
-#include <asm/io.h>
 
 #include "stmmac.h"
 #include "dwmac_dma.h"
-- 
cgit v1.2.3-58-ga151


From 7efc706577046a8dd2a11ba7942e3a52a30c8b68 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 29 May 2024 09:40:59 +0100
Subject: net: stmmac: ethqos: clean up setting serdes speed

There are four repititions of the same sequence of code, three of which
are identical. Pull these out into a separate function to improve
readability.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Reviewed-by: Serge Semin <fancer.lancer@gmail.com>
Link: https://lore.kernel.org/r/E1sCErj-00EOQ9-Vh@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c    | 24 +++++++++++-----------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
index e254b21fdb59..d9eed415b0b7 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
@@ -605,6 +605,14 @@ static int ethqos_configure_rgmii(struct qcom_ethqos *ethqos)
 	return 0;
 }
 
+static void ethqos_set_serdes_speed(struct qcom_ethqos *ethqos, int speed)
+{
+	if (ethqos->serdes_speed != speed) {
+		phy_set_speed(ethqos->serdes_phy, speed);
+		ethqos->serdes_speed = speed;
+	}
+}
+
 /* On interface toggle MAC registers gets reset.
  * Configure MAC block for SGMII on ethernet phy link up
  */
@@ -622,9 +630,7 @@ static int ethqos_configure_sgmii(struct qcom_ethqos *ethqos)
 		rgmii_updatel(ethqos, RGMII_CONFIG2_RGMII_CLK_SEL_CFG,
 			      RGMII_CONFIG2_RGMII_CLK_SEL_CFG,
 			      RGMII_IO_MACRO_CONFIG2);
-		if (ethqos->serdes_speed != SPEED_2500)
-			phy_set_speed(ethqos->serdes_phy, SPEED_2500);
-		ethqos->serdes_speed = SPEED_2500;
+		ethqos_set_serdes_speed(ethqos, SPEED_2500);
 		stmmac_pcs_ctrl_ane(priv, priv->ioaddr, 0, 0, 0);
 		break;
 	case SPEED_1000:
@@ -632,16 +638,12 @@ static int ethqos_configure_sgmii(struct qcom_ethqos *ethqos)
 		rgmii_updatel(ethqos, RGMII_CONFIG2_RGMII_CLK_SEL_CFG,
 			      RGMII_CONFIG2_RGMII_CLK_SEL_CFG,
 			      RGMII_IO_MACRO_CONFIG2);
-		if (ethqos->serdes_speed != SPEED_1000)
-			phy_set_speed(ethqos->serdes_phy, SPEED_1000);
-		ethqos->serdes_speed = SPEED_1000;
+		ethqos_set_serdes_speed(ethqos, SPEED_1000);
 		stmmac_pcs_ctrl_ane(priv, priv->ioaddr, 1, 0, 0);
 		break;
 	case SPEED_100:
 		val |= ETHQOS_MAC_CTRL_PORT_SEL | ETHQOS_MAC_CTRL_SPEED_MODE;
-		if (ethqos->serdes_speed != SPEED_1000)
-			phy_set_speed(ethqos->serdes_phy, SPEED_1000);
-		ethqos->serdes_speed = SPEED_1000;
+		ethqos_set_serdes_speed(ethqos, SPEED_1000);
 		stmmac_pcs_ctrl_ane(priv, priv->ioaddr, 1, 0, 0);
 		break;
 	case SPEED_10:
@@ -651,9 +653,7 @@ static int ethqos_configure_sgmii(struct qcom_ethqos *ethqos)
 			      FIELD_PREP(RGMII_CONFIG_SGMII_CLK_DVDR,
 					 SGMII_10M_RX_CLK_DVDR),
 			      RGMII_IO_MACRO_CONFIG);
-		if (ethqos->serdes_speed != SPEED_1000)
-			phy_set_speed(ethqos->serdes_phy, ethqos->speed);
-		ethqos->serdes_speed = SPEED_1000;
+		ethqos_set_serdes_speed(ethqos, SPEED_1000);
 		stmmac_pcs_ctrl_ane(priv, priv->ioaddr, 1, 0, 0);
 		break;
 	}
-- 
cgit v1.2.3-58-ga151


From 2f19a795e1f9c722b182f9a3d3dedd80f90a0715 Mon Sep 17 00:00:00 2001
From: MD Danish Anwar <danishanwar@ti.com>
Date: Wed, 29 May 2024 17:22:25 +0530
Subject: dt-bindings: net: ti: icssg_prueth: Add documentation for PA_STATS
 support

Add documentation for ti,pa-stats property which is syscon regmap for
PA_STATS registers. This will be used to dump statistics maintained by
ICSSG firmware.

Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20240529115225.630535-1-danishanwar@ti.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/ti,icssg-prueth.yaml | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/ti,icssg-prueth.yaml b/Documentation/devicetree/bindings/net/ti,icssg-prueth.yaml
index e253fa786092..c296e5711848 100644
--- a/Documentation/devicetree/bindings/net/ti,icssg-prueth.yaml
+++ b/Documentation/devicetree/bindings/net/ti,icssg-prueth.yaml
@@ -55,6 +55,14 @@ properties:
     description:
       phandle to MII_RT module's syscon regmap
 
+  ti,pa-stats:
+    $ref: /schemas/types.yaml#/definitions/phandle
+    description:
+      phandle to PA_STATS module's syscon regmap. PA_STATS is a set of
+      registers where different statistics related to ICSSG, are dumped by
+      ICSSG firmware. PA_STATS module's syscon regmap will help the device to
+      access/read/write those statistics.
+
   ti,iep:
     $ref: /schemas/types.yaml#/definitions/phandle-array
     maxItems: 2
@@ -194,6 +202,7 @@ examples:
                     "tx1-0", "tx1-1", "tx1-2", "tx1-3",
                     "rx0", "rx1";
         ti,mii-g-rt = <&icssg2_mii_g_rt>;
+        ti,pa-stats = <&icssg2_pa_stats>;
         ti,iep = <&icssg2_iep0>, <&icssg2_iep1>;
         interrupt-parent = <&icssg2_intc>;
         interrupts = <24 0 2>, <25 1 3>;
-- 
cgit v1.2.3-58-ga151


From 75518b0dc9d602bdd0e0ee4e04806dff53cb761a Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 29 May 2024 14:29:19 +0100
Subject: net: phylink: rearrange phylink_parse_mode()

Of the two users of phylink_config->ovr_an_inband, both manually check
for a fixed link before setting this flag (or clearing it if they find
a fixed link.) This is unnecessary complication.

Rearrange phylink_parse_mode() a little so we can change how
phylink_config->ovr_an_inband works. This will allow the flag to be
tested before checking for the fixed link properties in the next patch.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Link: https://lore.kernel.org/r/E1sCJMl-00Ecqp-K0@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/phylink.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 994471fad833..5abd12713598 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -899,12 +899,15 @@ static int phylink_parse_mode(struct phylink *pl,
 			return -EINVAL;
 		}
 
+		pl->cfg_link_an_mode = MLO_AN_INBAND;
+	}
+
+	if (pl->cfg_link_an_mode == MLO_AN_INBAND) {
 		linkmode_zero(pl->supported);
 		phylink_set(pl->supported, MII);
 		phylink_set(pl->supported, Autoneg);
 		phylink_set(pl->supported, Asym_Pause);
 		phylink_set(pl->supported, Pause);
-		pl->cfg_link_an_mode = MLO_AN_INBAND;
 
 		switch (pl->link_config.interface) {
 		case PHY_INTERFACE_MODE_SGMII:
-- 
cgit v1.2.3-58-ga151


From fea49f065c1cff71a86e7629094f912531edf035 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 29 May 2024 14:29:24 +0100
Subject: net: phylink: move test for ovr_an_inband

Of the two users of phylink_config->ovr_an_inband, both manually check
for a fixed link before setting this flag (or clearing it if they find
a fixed link.) This is unnecessary complication.

Test ovr_an_inband before checking for the fixed-link properties, which
will allow ovr_an_inband to be overriden by a fixed link specification.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Link: https://lore.kernel.org/r/E1sCJMq-00Ecqv-P8@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/phylink.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 5abd12713598..c81f1c1ee675 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -885,14 +885,16 @@ static int phylink_parse_mode(struct phylink *pl,
 	const char *managed;
 	unsigned long caps;
 
+	if (pl->config->ovr_an_inband)
+		pl->cfg_link_an_mode = MLO_AN_INBAND;
+
 	dn = fwnode_get_named_child_node(fwnode, "fixed-link");
 	if (dn || fwnode_property_present(fwnode, "fixed-link"))
 		pl->cfg_link_an_mode = MLO_AN_FIXED;
 	fwnode_handle_put(dn);
 
 	if ((fwnode_property_read_string(fwnode, "managed", &managed) == 0 &&
-	     strcmp(managed, "in-band-status") == 0) ||
-	    pl->config->ovr_an_inband) {
+	     strcmp(managed, "in-band-status") == 0)) {
 		if (pl->cfg_link_an_mode == MLO_AN_FIXED) {
 			phylink_err(pl,
 				    "can't use both fixed-link and in-band-status\n");
-- 
cgit v1.2.3-58-ga151


From 02d00dc73d8d0613f82063ed53d67912a422c21e Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 29 May 2024 14:29:29 +0100
Subject: net: phylink: rename ovr_an_inband to default_an_inband

Since ovr_an_inband no longer overrides every MLO_AN_xxx mode, rename
it to reflect what it now does - it changes the default mode from
MLO_AN_PHY to MLO_AN_INBAND. Fix up the two users of this.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Link: https://lore.kernel.org/r/E1sCJMv-00Ecr1-Sk@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/fman/fman_memac.c  | 2 +-
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +-
 drivers/net/phy/phylink.c                         | 2 +-
 include/linux/phylink.h                           | 5 +++--
 4 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/freescale/fman/fman_memac.c b/drivers/net/ethernet/freescale/fman/fman_memac.c
index 92b8f4ab26f1..9c44a3581950 100644
--- a/drivers/net/ethernet/freescale/fman/fman_memac.c
+++ b/drivers/net/ethernet/freescale/fman/fman_memac.c
@@ -1232,7 +1232,7 @@ int memac_initialization(struct mac_device *mac_dev,
 	    !of_property_read_bool(mac_node, "managed") &&
 	    mac_dev->phy_if != PHY_INTERFACE_MODE_MII &&
 	    !phy_interface_mode_is_rgmii(mac_dev->phy_if))
-		mac_dev->phylink_config.ovr_an_inband = true;
+		mac_dev->phylink_config.default_an_inband = true;
 	of_node_put(fixed);
 
 	err = memac_init(mac_dev->fman_mac);
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index ca19b232431a..488b2fd2349c 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1221,7 +1221,7 @@ static int stmmac_phy_setup(struct stmmac_priv *priv)
 
 	mdio_bus_data = priv->plat->mdio_bus_data;
 	if (mdio_bus_data)
-		priv->phylink_config.ovr_an_inband =
+		priv->phylink_config.default_an_inband =
 			mdio_bus_data->xpcs_an_inband;
 
 	/* Set the platform/firmware specified interface mode. Note, phylink
diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index c81f1c1ee675..02427378acfd 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -885,7 +885,7 @@ static int phylink_parse_mode(struct phylink *pl,
 	const char *managed;
 	unsigned long caps;
 
-	if (pl->config->ovr_an_inband)
+	if (pl->config->default_an_inband)
 		pl->cfg_link_an_mode = MLO_AN_INBAND;
 
 	dn = fwnode_get_named_child_node(fwnode, "fixed-link");
diff --git a/include/linux/phylink.h b/include/linux/phylink.h
index 5ea6b2ad2396..a30a692acc32 100644
--- a/include/linux/phylink.h
+++ b/include/linux/phylink.h
@@ -141,7 +141,8 @@ enum phylink_op_type {
  * @mac_requires_rxc: if true, the MAC always requires a receive clock from PHY.
  *                    The PHY driver should start the clock signal as soon as
  *                    possible and avoid stopping it during suspend events.
- * @ovr_an_inband: if true, override PCS to MLO_AN_INBAND
+ * @default_an_inband: if true, defaults to MLO_AN_INBAND rather than
+ *		       MLO_AN_PHY. A fixed-link specification will override.
  * @get_fixed_state: callback to execute to determine the fixed link state,
  *		     if MAC link is at %MLO_AN_FIXED mode.
  * @supported_interfaces: bitmap describing which PHY_INTERFACE_MODE_xxx
@@ -154,7 +155,7 @@ struct phylink_config {
 	bool poll_fixed_state;
 	bool mac_managed_pm;
 	bool mac_requires_rxc;
-	bool ovr_an_inband;
+	bool default_an_inband;
 	void (*get_fixed_state)(struct phylink_config *config,
 				struct phylink_link_state *state);
 	DECLARE_PHY_INTERFACE_MASK(supported_interfaces);
-- 
cgit v1.2.3-58-ga151


From 5e332954e7609acc9a265fcd9df53a3102206c8d Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 29 May 2024 14:29:35 +0100
Subject: net: fman_memac: remove the now unnecessary checking for fixed-link

Since default_an_inband can be overriden by a fixed-link specification,
there is no need for memac to be checking for this before setting
default_an_inband. Remove this code and update the comment.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Sean Anderson <sean.anderson@seco.com>
Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Link: https://lore.kernel.org/r/E1sCJN1-00Ecr7-02@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/fman/fman_memac.c | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/freescale/fman/fman_memac.c b/drivers/net/ethernet/freescale/fman/fman_memac.c
index 9c44a3581950..796e6f4e583d 100644
--- a/drivers/net/ethernet/freescale/fman/fman_memac.c
+++ b/drivers/net/ethernet/freescale/fman/fman_memac.c
@@ -1066,7 +1066,6 @@ int memac_initialization(struct mac_device *mac_dev,
 			 struct fman_mac_params *params)
 {
 	int			 err;
-	struct device_node      *fixed;
 	struct phylink_pcs	*pcs;
 	struct fman_mac		*memac;
 	unsigned long		 capabilities;
@@ -1222,18 +1221,15 @@ int memac_initialization(struct mac_device *mac_dev,
 		memac->rgmii_no_half_duplex = true;
 
 	/* Most boards should use MLO_AN_INBAND, but existing boards don't have
-	 * a managed property. Default to MLO_AN_INBAND if nothing else is
-	 * specified. We need to be careful and not enable this if we have a
-	 * fixed link or if we are using MII or RGMII, since those
-	 * configurations modes don't use in-band autonegotiation.
+	 * a managed property. Default to MLO_AN_INBAND rather than MLO_AN_PHY.
+	 * Phylink will allow this to be overriden by a fixed link. We need to
+	 * be careful and not enable this if we are using MII or RGMII, since
+	 * those configurations modes don't use in-band autonegotiation.
 	 */
-	fixed = of_get_child_by_name(mac_node, "fixed-link");
-	if (!fixed && !of_property_read_bool(mac_node, "fixed-link") &&
-	    !of_property_read_bool(mac_node, "managed") &&
+	if (!of_property_read_bool(mac_node, "managed") &&
 	    mac_dev->phy_if != PHY_INTERFACE_MODE_MII &&
 	    !phy_interface_mode_is_rgmii(mac_dev->phy_if))
 		mac_dev->phylink_config.default_an_inband = true;
-	of_node_put(fixed);
 
 	err = memac_init(mac_dev->fman_mac);
 	if (err < 0)
-- 
cgit v1.2.3-58-ga151


From 83f55b01dd903040d6ab64f4f9d78a27391a5916 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 29 May 2024 14:29:40 +0100
Subject: net: stmmac: rename xpcs_an_inband to default_an_inband

Rename xpcs_an_inband to default_an_inband to reflect the change in
phylink and its changed functionality.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Link: https://lore.kernel.org/r/E1sCJN6-00EcrD-43@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 8 ++++----
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +-
 include/linux/stmmac.h                            | 2 +-
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
index 60283543ffc8..5e96146b8bd9 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
@@ -248,7 +248,7 @@ static void intel_speed_mode_2500(struct net_device *ndev, void *intel_data)
 		dev_info(priv->device, "Link Speed Mode: 2.5Gbps\n");
 		priv->plat->max_speed = 2500;
 		priv->plat->phy_interface = PHY_INTERFACE_MODE_2500BASEX;
-		priv->plat->mdio_bus_data->xpcs_an_inband = false;
+		priv->plat->mdio_bus_data->default_an_inband = false;
 	} else {
 		priv->plat->max_speed = 1000;
 	}
@@ -586,16 +586,16 @@ static int intel_mgbe_common_data(struct pci_dev *pdev,
 	if (plat->phy_interface == PHY_INTERFACE_MODE_SGMII ||
 	    plat->phy_interface == PHY_INTERFACE_MODE_1000BASEX) {
 		plat->mdio_bus_data->has_xpcs = true;
-		plat->mdio_bus_data->xpcs_an_inband = true;
+		plat->mdio_bus_data->default_an_inband = true;
 	}
 
-	/* For fixed-link setup, we clear xpcs_an_inband */
+	/* For fixed-link setup, we clear default_an_inband */
 	if (fwnode) {
 		struct fwnode_handle *fixed_node;
 
 		fixed_node = fwnode_get_named_child_node(fwnode, "fixed-link");
 		if (fixed_node)
-			plat->mdio_bus_data->xpcs_an_inband = false;
+			plat->mdio_bus_data->default_an_inband = false;
 
 		fwnode_handle_put(fixed_node);
 	}
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 488b2fd2349c..bbedf2a8c60f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1222,7 +1222,7 @@ static int stmmac_phy_setup(struct stmmac_priv *priv)
 	mdio_bus_data = priv->plat->mdio_bus_data;
 	if (mdio_bus_data)
 		priv->phylink_config.default_an_inband =
-			mdio_bus_data->xpcs_an_inband;
+			mdio_bus_data->default_an_inband;
 
 	/* Set the platform/firmware specified interface mode. Note, phylink
 	 * deals with the PHY interface mode, not the MAC interface mode.
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index f92c195c76ed..8f0f156d50d3 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -83,7 +83,7 @@ struct stmmac_priv;
 struct stmmac_mdio_bus_data {
 	unsigned int phy_mask;
 	unsigned int has_xpcs;
-	unsigned int xpcs_an_inband;
+	unsigned int default_an_inband;
 	int *irqs;
 	int probed_phy_irq;
 	bool needs_reset;
-- 
cgit v1.2.3-58-ga151


From ab77c7aa9388ccc82102c9e448112ab478bba80c Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Wed, 29 May 2024 14:29:45 +0100
Subject: net: stmmac: dwmac-intel: remove checking for fixed link

With the new default_an_inband functionality in phylink, there is no
need to check for a fixed link when this flag is set, since a fixed
link will now override default_an_inband.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Halaney <ahalaney@redhat.com>
Link: https://lore.kernel.org/r/E1sCJNB-00EcrJ-7L@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 11 -----------
 1 file changed, 11 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
index 5e96146b8bd9..56649edb18cd 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
@@ -589,17 +589,6 @@ static int intel_mgbe_common_data(struct pci_dev *pdev,
 		plat->mdio_bus_data->default_an_inband = true;
 	}
 
-	/* For fixed-link setup, we clear default_an_inband */
-	if (fwnode) {
-		struct fwnode_handle *fixed_node;
-
-		fixed_node = fwnode_get_named_child_node(fwnode, "fixed-link");
-		if (fixed_node)
-			plat->mdio_bus_data->default_an_inband = false;
-
-		fwnode_handle_put(fixed_node);
-	}
-
 	/* Ensure mdio bus scan skips intel serdes and pcs-xpcs */
 	plat->mdio_bus_data->phy_mask = 1 << INTEL_MGBE_ADHOC_ADDR;
 	plat->mdio_bus_data->phy_mask |= 1 << INTEL_MGBE_XPCS_ADDR;
-- 
cgit v1.2.3-58-ga151


From ccf23c916ca35239a924ec8649cc88b1ef25d3d9 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 29 May 2024 12:20:31 -0700
Subject: tools: ynl: make the attr and msg helpers more C++ friendly

Folks working on a C++ codegen would like to reuse the attribute
helpers directly. Add the few necessary casts, it's not too ugly.

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Link: https://lore.kernel.org/r/20240529192031.3785761-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/lib/ynl-priv.h | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/tools/net/ynl/lib/ynl-priv.h b/tools/net/ynl/lib/ynl-priv.h
index 6cf890080dc0..80791c34730c 100644
--- a/tools/net/ynl/lib/ynl-priv.h
+++ b/tools/net/ynl/lib/ynl-priv.h
@@ -79,7 +79,7 @@ static inline void *ynl_dump_obj_next(void *obj)
 	struct ynl_dump_list_type *list;
 
 	uptr -= offsetof(struct ynl_dump_list_type, data);
-	list = (void *)uptr;
+	list = (struct ynl_dump_list_type *)uptr;
 	uptr = (unsigned long)list->next;
 	uptr += offsetof(struct ynl_dump_list_type, data);
 
@@ -139,7 +139,7 @@ int ynl_error_parse(struct ynl_parse_arg *yarg, const char *msg);
 
 static inline struct nlmsghdr *ynl_nlmsg_put_header(void *buf)
 {
-	struct nlmsghdr *nlh = buf;
+	struct nlmsghdr *nlh = (struct nlmsghdr *)buf;
 
 	memset(nlh, 0, sizeof(*nlh));
 	nlh->nlmsg_len = NLMSG_HDRLEN;
@@ -196,7 +196,7 @@ static inline void *ynl_attr_data(const struct nlattr *attr)
 
 static inline void *ynl_attr_data_end(const struct nlattr *attr)
 {
-	return ynl_attr_data(attr) + ynl_attr_data_len(attr);
+	return (char *)ynl_attr_data(attr) + ynl_attr_data_len(attr);
 }
 
 #define ynl_attr_for_each(attr, nlh, fixed_hdr_sz)			\
@@ -228,7 +228,7 @@ ynl_attr_next(const void *end, const struct nlattr *prev)
 {
 	struct nlattr *attr;
 
-	attr = (void *)((char *)prev + NLA_ALIGN(prev->nla_len));
+	attr = (struct nlattr *)((char *)prev + NLA_ALIGN(prev->nla_len));
 	return ynl_attr_if_good(end, attr);
 }
 
@@ -237,8 +237,8 @@ ynl_attr_first(const void *start, size_t len, size_t skip)
 {
 	struct nlattr *attr;
 
-	attr = (void *)((char *)start + NLMSG_ALIGN(skip));
-	return ynl_attr_if_good(start + len, attr);
+	attr = (struct nlattr *)((char *)start + NLMSG_ALIGN(skip));
+	return ynl_attr_if_good((char *)start + len, attr);
 }
 
 static inline bool
@@ -262,9 +262,9 @@ ynl_attr_nest_start(struct nlmsghdr *nlh, unsigned int attr_type)
 	struct nlattr *attr;
 
 	if (__ynl_attr_put_overflow(nlh, 0))
-		return ynl_nlmsg_end_addr(nlh) - NLA_HDRLEN;
+		return (struct nlattr *)ynl_nlmsg_end_addr(nlh) - 1;
 
-	attr = ynl_nlmsg_end_addr(nlh);
+	attr = (struct nlattr *)ynl_nlmsg_end_addr(nlh);
 	attr->nla_type = attr_type | NLA_F_NESTED;
 	nlh->nlmsg_len += NLA_HDRLEN;
 
@@ -286,7 +286,7 @@ ynl_attr_put(struct nlmsghdr *nlh, unsigned int attr_type,
 	if (__ynl_attr_put_overflow(nlh, size))
 		return;
 
-	attr = ynl_nlmsg_end_addr(nlh);
+	attr = (struct nlattr *)ynl_nlmsg_end_addr(nlh);
 	attr->nla_type = attr_type;
 	attr->nla_len = NLA_HDRLEN + size;
 
@@ -305,10 +305,10 @@ ynl_attr_put_str(struct nlmsghdr *nlh, unsigned int attr_type, const char *str)
 	if (__ynl_attr_put_overflow(nlh, len))
 		return;
 
-	attr = ynl_nlmsg_end_addr(nlh);
+	attr = (struct nlattr *)ynl_nlmsg_end_addr(nlh);
 	attr->nla_type = attr_type;
 
-	strcpy(ynl_attr_data(attr), str);
+	strcpy((char *)ynl_attr_data(attr), str);
 	attr->nla_len = NLA_HDRLEN + NLA_ALIGN(len);
 
 	nlh->nlmsg_len += NLMSG_ALIGN(attr->nla_len);
-- 
cgit v1.2.3-58-ga151


From 531876c80004ecff7bfdbd8ba6c6b48835ef5e22 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Wed, 29 May 2024 15:32:39 -0700
Subject: libbpf: keep FD_CLOEXEC flag when dup()'ing FD

Make sure to preserve and/or enforce FD_CLOEXEC flag on duped FDs.
Use dup3() with O_CLOEXEC flag for that.

Without this fix libbpf effectively clears FD_CLOEXEC flag on each of BPF
map/prog FD, which is definitely not the right or expected behavior.

Reported-by: Lennart Poettering <lennart@poettering.net>
Fixes: bc308d011ab8 ("libbpf: call dup2() syscall directly")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20240529223239.504241-1-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/lib/bpf/libbpf_internal.h | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index a0dcfb82e455..7e7e686008c6 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -597,13 +597,9 @@ static inline int ensure_good_fd(int fd)
 	return fd;
 }
 
-static inline int sys_dup2(int oldfd, int newfd)
+static inline int sys_dup3(int oldfd, int newfd, int flags)
 {
-#ifdef __NR_dup2
-	return syscall(__NR_dup2, oldfd, newfd);
-#else
-	return syscall(__NR_dup3, oldfd, newfd, 0);
-#endif
+	return syscall(__NR_dup3, oldfd, newfd, flags);
 }
 
 /* Point *fixed_fd* to the same file that *tmp_fd* points to.
@@ -614,7 +610,7 @@ static inline int reuse_fd(int fixed_fd, int tmp_fd)
 {
 	int err;
 
-	err = sys_dup2(tmp_fd, fixed_fd);
+	err = sys_dup3(tmp_fd, fixed_fd, O_CLOEXEC);
 	err = err < 0 ? -errno : 0;
 	close(tmp_fd); /* clean up temporary FD */
 	return err;
-- 
cgit v1.2.3-58-ga151


From dcb77f854ae086bf32596fae02c88665393a13e8 Mon Sep 17 00:00:00 2001
From: Erick Archer <erick.archer@outlook.com>
Date: Mon, 27 May 2024 19:34:25 +0200
Subject: wifi: brcm80211: use sizeof(*pointer) instead of sizeof(type)

It is preferred to use sizeof(*pointer) instead of sizeof(type)
due to the type of the variable can change and one needs not
change the former (unlike the latter). This patch has no effect
on runtime behavior.

At the same time remove some redundant NULL initializations.

Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Erick Archer <erick.archer@outlook.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/AS8PR02MB7237FF1C2E880D1231684D708BF02@AS8PR02MB7237.eurprd02.prod.outlook.com
---
 .../wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c  |  4 ++--
 .../wireless/broadcom/brcm80211/brcmfmac/btcoex.c  |  4 ++--
 .../wireless/broadcom/brcm80211/brcmfmac/sdio.c    |  2 +-
 .../net/wireless/broadcom/brcm80211/brcmfmac/usb.c |  6 +++---
 .../wireless/broadcom/brcm80211/brcmsmac/aiutils.c |  2 +-
 .../wireless/broadcom/brcm80211/brcmsmac/ampdu.c   |  2 +-
 .../wireless/broadcom/brcm80211/brcmsmac/antsel.c  |  2 +-
 .../wireless/broadcom/brcm80211/brcmsmac/channel.c |  2 +-
 .../net/wireless/broadcom/brcm80211/brcmsmac/dma.c |  2 +-
 .../broadcom/brcm80211/brcmsmac/mac80211_if.c      |  2 +-
 .../wireless/broadcom/brcm80211/brcmsmac/main.c    | 23 +++++++++++-----------
 .../broadcom/brcm80211/brcmsmac/phy/phy_cmn.c      |  4 ++--
 .../broadcom/brcm80211/brcmsmac/phy/phy_lcn.c      |  6 +++---
 .../broadcom/brcm80211/brcmsmac/phy_shim.c         |  4 ++--
 14 files changed, 32 insertions(+), 33 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
index 00679a990e3d..32f6b9d419d4 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
@@ -1061,10 +1061,10 @@ static int brcmf_ops_sdio_probe(struct sdio_func *func,
 	if (func->num != 2)
 		return -ENODEV;
 
-	bus_if = kzalloc(sizeof(struct brcmf_bus), GFP_KERNEL);
+	bus_if = kzalloc(sizeof(*bus_if), GFP_KERNEL);
 	if (!bus_if)
 		return -ENOMEM;
-	sdiodev = kzalloc(sizeof(struct brcmf_sdio_dev), GFP_KERNEL);
+	sdiodev = kzalloc(sizeof(*sdiodev), GFP_KERNEL);
 	if (!sdiodev) {
 		kfree(bus_if);
 		return -ENOMEM;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/btcoex.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/btcoex.c
index 7ea2631b8069..0c3d119d1219 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/btcoex.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/btcoex.c
@@ -358,10 +358,10 @@ idle:
  */
 int brcmf_btcoex_attach(struct brcmf_cfg80211_info *cfg)
 {
-	struct brcmf_btcoex_info *btci = NULL;
+	struct brcmf_btcoex_info *btci;
 	brcmf_dbg(TRACE, "enter\n");
 
-	btci = kmalloc(sizeof(struct brcmf_btcoex_info), GFP_KERNEL);
+	btci = kmalloc(sizeof(*btci), GFP_KERNEL);
 	if (!btci)
 		return -ENOMEM;
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
index 6b38d9de71af..1461dc453ac2 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
@@ -4450,7 +4450,7 @@ struct brcmf_sdio *brcmf_sdio_probe(struct brcmf_sdio_dev *sdiodev)
 	brcmf_dbg(TRACE, "Enter\n");
 
 	/* Allocate private bus interface state */
-	bus = kzalloc(sizeof(struct brcmf_sdio), GFP_ATOMIC);
+	bus = kzalloc(sizeof(*bus), GFP_ATOMIC);
 	if (!bus)
 		goto fail;
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
index 9a105e6debe1..8afbf529c745 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/usb.c
@@ -1236,8 +1236,8 @@ brcmf_usb_prepare_fw_request(struct brcmf_usbdev_info *devinfo)
 static int brcmf_usb_probe_cb(struct brcmf_usbdev_info *devinfo,
 			      enum brcmf_fwvendor fwvid)
 {
-	struct brcmf_bus *bus = NULL;
-	struct brcmf_usbdev *bus_pub = NULL;
+	struct brcmf_bus *bus;
+	struct brcmf_usbdev *bus_pub;
 	struct device *dev = devinfo->dev;
 	struct brcmf_fw_request *fwreq;
 	int ret;
@@ -1247,7 +1247,7 @@ static int brcmf_usb_probe_cb(struct brcmf_usbdev_info *devinfo,
 	if (!bus_pub)
 		return -ENODEV;
 
-	bus = kzalloc(sizeof(struct brcmf_bus), GFP_ATOMIC);
+	bus = kzalloc(sizeof(*bus), GFP_ATOMIC);
 	if (!bus) {
 		ret = -ENOMEM;
 		goto fail;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/aiutils.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/aiutils.c
index 2084b506a450..50d817485cf9 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/aiutils.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/aiutils.c
@@ -512,7 +512,7 @@ ai_attach(struct bcma_bus *pbus)
 	struct si_info *sii;
 
 	/* alloc struct si_info */
-	sii = kzalloc(sizeof(struct si_info), GFP_ATOMIC);
+	sii = kzalloc(sizeof(*sii), GFP_ATOMIC);
 	if (sii == NULL)
 		return NULL;
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/ampdu.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/ampdu.c
index c3376f887114..33d17b779201 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/ampdu.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/ampdu.c
@@ -219,7 +219,7 @@ struct ampdu_info *brcms_c_ampdu_attach(struct brcms_c_info *wlc)
 	struct ampdu_info *ampdu;
 	int i;
 
-	ampdu = kzalloc(sizeof(struct ampdu_info), GFP_ATOMIC);
+	ampdu = kzalloc(sizeof(*ampdu), GFP_ATOMIC);
 	if (!ampdu)
 		return NULL;
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/antsel.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/antsel.c
index 54c616919590..f411bc6d795d 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/antsel.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/antsel.c
@@ -111,7 +111,7 @@ struct antsel_info *brcms_c_antsel_attach(struct brcms_c_info *wlc)
 	struct antsel_info *asi;
 	struct ssb_sprom *sprom = &wlc->hw->d11core->bus->sprom;
 
-	asi = kzalloc(sizeof(struct antsel_info), GFP_ATOMIC);
+	asi = kzalloc(sizeof(*asi), GFP_ATOMIC);
 	if (!asi)
 		return NULL;
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/channel.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/channel.c
index f6962e558d7c..d1b9a18d0374 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/channel.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/channel.c
@@ -331,7 +331,7 @@ struct brcms_cm_info *brcms_c_channel_mgr_attach(struct brcms_c_info *wlc)
 	const char *ccode = sprom->alpha2;
 	int ccode_len = sizeof(sprom->alpha2);
 
-	wlc_cm = kzalloc(sizeof(struct brcms_cm_info), GFP_ATOMIC);
+	wlc_cm = kzalloc(sizeof(*wlc_cm), GFP_ATOMIC);
 	if (wlc_cm == NULL)
 		return NULL;
 	wlc_cm->pub = pub;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/dma.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/dma.c
index 3d5c1ef8f7f2..bd480239368a 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/dma.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/dma.c
@@ -558,7 +558,7 @@ struct dma_pub *dma_attach(char *name, struct brcms_c_info *wlc,
 	struct si_info *sii = container_of(sih, struct si_info, pub);
 
 	/* allocate private info structure */
-	di = kzalloc(sizeof(struct dma_info), GFP_ATOMIC);
+	di = kzalloc(sizeof(*di), GFP_ATOMIC);
 	if (di == NULL)
 		return NULL;
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
index 92860dc0a92e..860ef9c11c46 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
@@ -1496,7 +1496,7 @@ struct brcms_timer *brcms_init_timer(struct brcms_info *wl,
 {
 	struct brcms_timer *t;
 
-	t = kzalloc(sizeof(struct brcms_timer), GFP_ATOMIC);
+	t = kzalloc(sizeof(*t), GFP_ATOMIC);
 	if (!t)
 		return NULL;
 
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c
index 34460b5815d0..d3c74440381f 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c
@@ -463,11 +463,11 @@ static struct brcms_bss_cfg *brcms_c_bsscfg_malloc(uint unit)
 {
 	struct brcms_bss_cfg *cfg;
 
-	cfg = kzalloc(sizeof(struct brcms_bss_cfg), GFP_ATOMIC);
+	cfg = kzalloc(sizeof(*cfg), GFP_ATOMIC);
 	if (cfg == NULL)
 		goto fail;
 
-	cfg->current_bss = kzalloc(sizeof(struct brcms_bss_info), GFP_ATOMIC);
+	cfg->current_bss = kzalloc(sizeof(*cfg->current_bss), GFP_ATOMIC);
 	if (cfg->current_bss == NULL)
 		goto fail;
 
@@ -483,14 +483,14 @@ brcms_c_attach_malloc(uint unit, uint *err, uint devid)
 {
 	struct brcms_c_info *wlc;
 
-	wlc = kzalloc(sizeof(struct brcms_c_info), GFP_ATOMIC);
+	wlc = kzalloc(sizeof(*wlc), GFP_ATOMIC);
 	if (wlc == NULL) {
 		*err = 1002;
 		goto fail;
 	}
 
 	/* allocate struct brcms_c_pub state structure */
-	wlc->pub = kzalloc(sizeof(struct brcms_pub), GFP_ATOMIC);
+	wlc->pub = kzalloc(sizeof(*wlc->pub), GFP_ATOMIC);
 	if (wlc->pub == NULL) {
 		*err = 1003;
 		goto fail;
@@ -499,7 +499,7 @@ brcms_c_attach_malloc(uint unit, uint *err, uint devid)
 
 	/* allocate struct brcms_hardware state structure */
 
-	wlc->hw = kzalloc(sizeof(struct brcms_hardware), GFP_ATOMIC);
+	wlc->hw = kzalloc(sizeof(*wlc->hw), GFP_ATOMIC);
 	if (wlc->hw == NULL) {
 		*err = 1005;
 		goto fail;
@@ -528,7 +528,7 @@ brcms_c_attach_malloc(uint unit, uint *err, uint devid)
 		goto fail;
 	}
 
-	wlc->default_bss = kzalloc(sizeof(struct brcms_bss_info), GFP_ATOMIC);
+	wlc->default_bss = kzalloc(sizeof(*wlc->default_bss), GFP_ATOMIC);
 	if (wlc->default_bss == NULL) {
 		*err = 1010;
 		goto fail;
@@ -540,21 +540,20 @@ brcms_c_attach_malloc(uint unit, uint *err, uint devid)
 		goto fail;
 	}
 
-	wlc->protection = kzalloc(sizeof(struct brcms_protection),
-				  GFP_ATOMIC);
+	wlc->protection = kzalloc(sizeof(*wlc->protection), GFP_ATOMIC);
 	if (wlc->protection == NULL) {
 		*err = 1016;
 		goto fail;
 	}
 
-	wlc->stf = kzalloc(sizeof(struct brcms_stf), GFP_ATOMIC);
+	wlc->stf = kzalloc(sizeof(*wlc->stf), GFP_ATOMIC);
 	if (wlc->stf == NULL) {
 		*err = 1017;
 		goto fail;
 	}
 
 	wlc->bandstate[0] =
-		kcalloc(MAXBANDS, sizeof(struct brcms_band), GFP_ATOMIC);
+		kcalloc(MAXBANDS, sizeof(*wlc->bandstate[0]), GFP_ATOMIC);
 	if (wlc->bandstate[0] == NULL) {
 		*err = 1025;
 		goto fail;
@@ -567,14 +566,14 @@ brcms_c_attach_malloc(uint unit, uint *err, uint devid)
 				+ (sizeof(struct brcms_band)*i));
 	}
 
-	wlc->corestate = kzalloc(sizeof(struct brcms_core), GFP_ATOMIC);
+	wlc->corestate = kzalloc(sizeof(*wlc->corestate), GFP_ATOMIC);
 	if (wlc->corestate == NULL) {
 		*err = 1026;
 		goto fail;
 	}
 
 	wlc->corestate->macstat_snapshot =
-		kzalloc(sizeof(struct macstat), GFP_ATOMIC);
+		kzalloc(sizeof(*wlc->corestate->macstat_snapshot), GFP_ATOMIC);
 	if (wlc->corestate->macstat_snapshot == NULL) {
 		*err = 1027;
 		goto fail;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_cmn.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_cmn.c
index a27d6f0b8819..c3d7aa570b4e 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_cmn.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_cmn.c
@@ -355,7 +355,7 @@ struct shared_phy *wlc_phy_shared_attach(struct shared_phy_params *shp)
 {
 	struct shared_phy *sh;
 
-	sh = kzalloc(sizeof(struct shared_phy), GFP_ATOMIC);
+	sh = kzalloc(sizeof(*sh), GFP_ATOMIC);
 	if (sh == NULL)
 		return NULL;
 
@@ -442,7 +442,7 @@ wlc_phy_attach(struct shared_phy *sh, struct bcma_device *d11core,
 		return &pi->pubpi_ro;
 	}
 
-	pi = kzalloc(sizeof(struct brcms_phy), GFP_ATOMIC);
+	pi = kzalloc(sizeof(*pi), GFP_ATOMIC);
 	if (pi == NULL)
 		return NULL;
 	pi->wiphy = wiphy;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_lcn.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_lcn.c
index e472591f321b..d0faba240561 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_lcn.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_lcn.c
@@ -4968,11 +4968,11 @@ bool wlc_phy_attach_lcnphy(struct brcms_phy *pi)
 {
 	struct brcms_phy_lcnphy *pi_lcn;
 
-	pi->u.pi_lcnphy = kzalloc(sizeof(struct brcms_phy_lcnphy), GFP_ATOMIC);
-	if (pi->u.pi_lcnphy == NULL)
+	pi_lcn = kzalloc(sizeof(*pi_lcn), GFP_ATOMIC);
+	if (!pi_lcn)
 		return false;
 
-	pi_lcn = pi->u.pi_lcnphy;
+	pi->u.pi_lcnphy = pi_lcn;
 
 	if (0 == (pi->sh->boardflags & BFL_NOPA)) {
 		pi->hwpwrctrl = true;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy_shim.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy_shim.c
index b72381791536..8b852581c4e4 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy_shim.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy_shim.c
@@ -38,9 +38,9 @@ struct phy_shim_info {
 struct phy_shim_info *wlc_phy_shim_attach(struct brcms_hardware *wlc_hw,
 					  struct brcms_info *wl,
 					  struct brcms_c_info *wlc) {
-	struct phy_shim_info *physhim = NULL;
+	struct phy_shim_info *physhim;
 
-	physhim = kzalloc(sizeof(struct phy_shim_info), GFP_ATOMIC);
+	physhim = kzalloc(sizeof(*physhim), GFP_ATOMIC);
 	if (!physhim)
 		return NULL;
 
-- 
cgit v1.2.3-58-ga151


From 5bcd9a0a59953b5ff55ac226f903397319bf7f40 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Mon, 27 May 2024 00:45:53 +0100
Subject: wifi: brcm80211: remove unused structs

'brcmf_pcie_core_info' was added in
commit 9e37f045d5e7 ("brcmfmac: Adding PCIe bus layer support.")
but never used.

'brcms_c_bit_desc' last use was removed in
commit cdf4352f5c59 ("brcmsmac: Improve tx trace and debug support").

Remove them.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240526234553.286773-1-linux@treblig.org
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c | 5 -----
 drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c | 6 ------
 2 files changed, 11 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index 06698a714b52..ce482a3877e9 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -313,11 +313,6 @@ struct brcmf_pcie_shared_info {
 	u8 version;
 };
 
-struct brcmf_pcie_core_info {
-	u32 base;
-	u32 wrapbase;
-};
-
 #define BRCMF_OTP_MAX_PARAM_LEN 16
 
 struct brcmf_otp_params {
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c
index d3c74440381f..2738d4d6c60a 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c
@@ -234,12 +234,6 @@
 /* max # tx status to process in wlc_txstatus() */
 #define TXSBND				8
 
-/* brcmu_format_flags() bit description structure */
-struct brcms_c_bit_desc {
-	u32 bit;
-	const char *name;
-};
-
 /*
  * The following table lists the buffer memory allocated to xmt fifos in HW.
  * the size is in units of 256bytes(one block), total size is HW dependent
-- 
cgit v1.2.3-58-ga151


From 69c8b998717c03adeada070882512ebeae11d71f Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 29 May 2024 09:29:22 -0700
Subject: net: qstat: extend kdoc about get_base_stats

mlx5 has a dedicated queue for PTP packets. Clarify that
this sort of queues can also be accounted towards the base.

Reviewed-by: Joe Damato <jdamato@fastly.com>
Link: https://lore.kernel.org/r/20240529162922.3690698-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/netdev_queues.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/net/netdev_queues.h b/include/net/netdev_queues.h
index a8a7e48dfa6c..5ca019d294ca 100644
--- a/include/net/netdev_queues.h
+++ b/include/net/netdev_queues.h
@@ -62,6 +62,8 @@ struct netdev_queue_stats_tx {
  * statistics will not generally add up to the total number of events for
  * the device. The @get_base_stats callback allows filling in the delta
  * between events for currently live queues and overall device history.
+ * @get_base_stats can also be used to report any miscellaneous packets
+ * transferred outside of the main set of queues used by the networking stack.
  * When the statistics for the entire device are queried, first @get_base_stats
  * is issued to collect the delta, and then a series of per-queue callbacks.
  * Only statistics which are set in @get_base_stats will be reported
-- 
cgit v1.2.3-58-ga151


From 6d9e9c36e1a6407818ec8ef9785ae47150d7ddc3 Mon Sep 17 00:00:00 2001
From: Thorsten Blum <thorsten.blum@toblux.com>
Date: Wed, 29 May 2024 16:39:02 +0200
Subject: net: smc91x: Fix pointer types

Use void __iomem pointers as parameters for mcf_insw() and mcf_outsw()
to align with the parameter types of readw() and writew() to fix the
following warnings reported by kernel test robot:

drivers/net/ethernet/smsc/smc91x.c:590:9: sparse: warning: incorrect type in argument 1 (different address spaces)
drivers/net/ethernet/smsc/smc91x.c:590:9: sparse:    expected void *a
drivers/net/ethernet/smsc/smc91x.c:590:9: sparse:    got void [noderef] __iomem *
drivers/net/ethernet/smsc/smc91x.c:590:9: sparse: warning: incorrect type in argument 1 (different address spaces)
drivers/net/ethernet/smsc/smc91x.c:590:9: sparse:    expected void *a
drivers/net/ethernet/smsc/smc91x.c:590:9: sparse:    got void [noderef] __iomem *
drivers/net/ethernet/smsc/smc91x.c:590:9: sparse: warning: incorrect type in argument 1 (different address spaces)
drivers/net/ethernet/smsc/smc91x.c:590:9: sparse:    expected void *a
drivers/net/ethernet/smsc/smc91x.c:590:9: sparse:    got void [noderef] __iomem *
drivers/net/ethernet/smsc/smc91x.c:483:17: sparse: warning: incorrect type in argument 1 (different address spaces)
drivers/net/ethernet/smsc/smc91x.c:483:17: sparse:    expected void *a
drivers/net/ethernet/smsc/smc91x.c:483:17: sparse:    got void [noderef] __iomem *

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202405160853.3qyaSj8w-lkp@intel.com/
Acked-by: Nicolas Pitre <nico@fluxnic.net>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Thorsten Blum <thorsten.blum@toblux.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20240529143859.108201-4-thorsten.blum@toblux.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/smsc/smc91x.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/smsc/smc91x.h b/drivers/net/ethernet/smsc/smc91x.h
index 45ef5ac0788a..38aa4374e813 100644
--- a/drivers/net/ethernet/smsc/smc91x.h
+++ b/drivers/net/ethernet/smsc/smc91x.h
@@ -142,14 +142,14 @@ static inline void _SMC_outw_align4(u16 val, void __iomem *ioaddr, int reg,
 #define SMC_CAN_USE_32BIT	0
 #define SMC_NOWAIT		1
 
-static inline void mcf_insw(void *a, unsigned char *p, int l)
+static inline void mcf_insw(void __iomem *a, unsigned char *p, int l)
 {
 	u16 *wp = (u16 *) p;
 	while (l-- > 0)
 		*wp++ = readw(a);
 }
 
-static inline void mcf_outsw(void *a, unsigned char *p, int l)
+static inline void mcf_outsw(void __iomem *a, unsigned char *p, int l)
 {
 	u16 *wp = (u16 *) p;
 	while (l-- > 0)
-- 
cgit v1.2.3-58-ga151


From ccf45c92d746e8eac2cb1e60243d38a479e8dc0a Mon Sep 17 00:00:00 2001
From: "Matthieu Baerts (NGI0)" <matttbe@kernel.org>
Date: Thu, 30 May 2024 16:07:30 +0200
Subject: doc: mptcp: add missing 'available_schedulers' entry

This sysctl knob has been added recently, but the documentation has not
been updated.

This knob is used to show the available schedulers choices that are
registered, similar to 'net.ipv4.tcp_available_congestion_control'.

Fixes: 73c900aa3660 ("mptcp: add net.mptcp.available_schedulers")
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://lore.kernel.org/r/20240530-upstream-net-20240520-mptcp-doc-v3-1-e94cdd9f2673@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/mptcp-sysctl.rst | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/networking/mptcp-sysctl.rst b/Documentation/networking/mptcp-sysctl.rst
index 69975ce25a02..102a45e7bfa8 100644
--- a/Documentation/networking/mptcp-sysctl.rst
+++ b/Documentation/networking/mptcp-sysctl.rst
@@ -93,3 +93,7 @@ scheduler - STRING
 	sysctl.
 
 	Default: "default"
+
+available_schedulers - STRING
+	Shows the available schedulers choices that are registered. More packet
+	schedulers may be available, but not loaded.
-- 
cgit v1.2.3-58-ga151


From a32c6966b23d987520f7c5ac12d153eaff425553 Mon Sep 17 00:00:00 2001
From: "Matthieu Baerts (NGI0)" <matttbe@kernel.org>
Date: Thu, 30 May 2024 16:07:31 +0200
Subject: doc: mptcp: alphabetical order

Similar to what is done in other 'sysctl' pages: it looks clearer from a
readability perspective.

This might cause troubles in the short/mid-term with the backports, but
by not putting new entries at the end, this can help to reduce conflicts
in case of backports in the long term. We don't change the information
here too often, so it looks OK to do that.

Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://lore.kernel.org/r/20240530-upstream-net-20240520-mptcp-doc-v3-2-e94cdd9f2673@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/mptcp-sysctl.rst | 74 +++++++++++++++----------------
 1 file changed, 37 insertions(+), 37 deletions(-)

diff --git a/Documentation/networking/mptcp-sysctl.rst b/Documentation/networking/mptcp-sysctl.rst
index 102a45e7bfa8..fd514bba8c43 100644
--- a/Documentation/networking/mptcp-sysctl.rst
+++ b/Documentation/networking/mptcp-sysctl.rst
@@ -7,14 +7,6 @@ MPTCP Sysfs variables
 /proc/sys/net/mptcp/* Variables
 ===============================
 
-enabled - BOOLEAN
-	Control whether MPTCP sockets can be created.
-
-	MPTCP sockets can be created if the value is 1. This is a
-	per-namespace sysctl.
-
-	Default: 1 (enabled)
-
 add_addr_timeout - INTEGER (seconds)
 	Set the timeout after which an ADD_ADDR control message will be
 	resent to an MPTCP peer that has not acknowledged a previous
@@ -25,16 +17,22 @@ add_addr_timeout - INTEGER (seconds)
 
 	Default: 120
 
-close_timeout - INTEGER (seconds)
-	Set the make-after-break timeout: in absence of any close or
-	shutdown syscall, MPTCP sockets will maintain the status
-	unchanged for such time, after the last subflow removal, before
-	moving to TCP_CLOSE.
+allow_join_initial_addr_port - BOOLEAN
+	Allow peers to send join requests to the IP address and port number used
+	by the initial subflow if the value is 1. This controls a flag that is
+	sent to the peer at connection time, and whether such join requests are
+	accepted or denied.
 
-	The default value matches TCP_TIMEWAIT_LEN. This is a per-namespace
-	sysctl.
+	Joins to addresses advertised with ADD_ADDR are not affected by this
+	value.
 
-	Default: 60
+	This is a per-namespace sysctl.
+
+	Default: 1
+
+available_schedulers - STRING
+	Shows the available schedulers choices that are registered. More packet
+	schedulers may be available, but not loaded.
 
 checksum_enabled - BOOLEAN
 	Control whether DSS checksum can be enabled.
@@ -44,18 +42,24 @@ checksum_enabled - BOOLEAN
 
 	Default: 0
 
-allow_join_initial_addr_port - BOOLEAN
-	Allow peers to send join requests to the IP address and port number used
-	by the initial subflow if the value is 1. This controls a flag that is
-	sent to the peer at connection time, and whether such join requests are
-	accepted or denied.
+close_timeout - INTEGER (seconds)
+	Set the make-after-break timeout: in absence of any close or
+	shutdown syscall, MPTCP sockets will maintain the status
+	unchanged for such time, after the last subflow removal, before
+	moving to TCP_CLOSE.
 
-	Joins to addresses advertised with ADD_ADDR are not affected by this
-	value.
+	The default value matches TCP_TIMEWAIT_LEN. This is a per-namespace
+	sysctl.
 
-	This is a per-namespace sysctl.
+	Default: 60
 
-	Default: 1
+enabled - BOOLEAN
+	Control whether MPTCP sockets can be created.
+
+	MPTCP sockets can be created if the value is 1. This is a
+	per-namespace sysctl.
+
+	Default: 1 (enabled)
 
 pm_type - INTEGER
 	Set the default path manager type to use for each new MPTCP
@@ -74,6 +78,14 @@ pm_type - INTEGER
 
 	Default: 0
 
+scheduler - STRING
+	Select the scheduler of your choice.
+
+	Support for selection of different schedulers. This is a per-namespace
+	sysctl.
+
+	Default: "default"
+
 stale_loss_cnt - INTEGER
 	The number of MPTCP-level retransmission intervals with no traffic and
 	pending outstanding data on a given subflow required to declare it stale.
@@ -85,15 +97,3 @@ stale_loss_cnt - INTEGER
 	This is a per-namespace sysctl.
 
 	Default: 4
-
-scheduler - STRING
-	Select the scheduler of your choice.
-
-	Support for selection of different schedulers. This is a per-namespace
-	sysctl.
-
-	Default: "default"
-
-available_schedulers - STRING
-	Shows the available schedulers choices that are registered. More packet
-	schedulers may be available, but not loaded.
-- 
cgit v1.2.3-58-ga151


From c049275f24de70edd43e521f5a549c9e63165906 Mon Sep 17 00:00:00 2001
From: "Matthieu Baerts (NGI0)" <matttbe@kernel.org>
Date: Thu, 30 May 2024 16:07:32 +0200
Subject: doc: new 'mptcp' page in 'networking'

A general documentation about MPTCP was missing since its introduction
in v5.6.

Most of what is there comes from our recently updated mptcp.dev website,
with additional links to resources from the kernel documentation.

This is a first version, mainly targeting app developers and users.

Link: https://www.mptcp.dev
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://lore.kernel.org/r/20240530-upstream-net-20240520-mptcp-doc-v3-3-e94cdd9f2673@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/index.rst |   1 +
 Documentation/networking/mptcp.rst | 156 +++++++++++++++++++++++++++++++++++++
 MAINTAINERS                        |   2 +-
 3 files changed, 158 insertions(+), 1 deletion(-)
 create mode 100644 Documentation/networking/mptcp.rst

diff --git a/Documentation/networking/index.rst b/Documentation/networking/index.rst
index 7664c0bfe461..a6443851a142 100644
--- a/Documentation/networking/index.rst
+++ b/Documentation/networking/index.rst
@@ -72,6 +72,7 @@ Contents:
    mac80211-injection
    mctp
    mpls-sysctl
+   mptcp
    mptcp-sysctl
    multiqueue
    multi-pf-netdev
diff --git a/Documentation/networking/mptcp.rst b/Documentation/networking/mptcp.rst
new file mode 100644
index 000000000000..17f2bab61164
--- /dev/null
+++ b/Documentation/networking/mptcp.rst
@@ -0,0 +1,156 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+=====================
+Multipath TCP (MPTCP)
+=====================
+
+Introduction
+============
+
+Multipath TCP or MPTCP is an extension to the standard TCP and is described in
+`RFC 8684 (MPTCPv1) <https://www.rfc-editor.org/rfc/rfc8684.html>`_. It allows a
+device to make use of multiple interfaces at once to send and receive TCP
+packets over a single MPTCP connection. MPTCP can aggregate the bandwidth of
+multiple interfaces or prefer the one with the lowest latency. It also allows a
+fail-over if one path is down, and the traffic is seamlessly reinjected on other
+paths.
+
+For more details about Multipath TCP in the Linux kernel, please see the
+official website: `mptcp.dev <https://www.mptcp.dev>`_.
+
+
+Use cases
+=========
+
+Thanks to MPTCP, being able to use multiple paths in parallel or simultaneously
+brings new use-cases, compared to TCP:
+
+- Seamless handovers: switching from one path to another while preserving
+  established connections, e.g. to be used in mobility use-cases, like on
+  smartphones.
+- Best network selection: using the "best" available path depending on some
+  conditions, e.g. latency, losses, cost, bandwidth, etc.
+- Network aggregation: using multiple paths at the same time to have a higher
+  throughput, e.g. to combine fixed and mobile networks to send files faster.
+
+
+Concepts
+========
+
+Technically, when a new socket is created with the ``IPPROTO_MPTCP`` protocol
+(Linux-specific), a *subflow* (or *path*) is created. This *subflow* consists of
+a regular TCP connection that is used to transmit data through one interface.
+Additional *subflows* can be negotiated later between the hosts. For the remote
+host to be able to detect the use of MPTCP, a new field is added to the TCP
+*option* field of the underlying TCP *subflow*. This field contains, amongst
+other things, a ``MP_CAPABLE`` option that tells the other host to use MPTCP if
+it is supported. If the remote host or any middlebox in between does not support
+it, the returned ``SYN+ACK`` packet will not contain MPTCP options in the TCP
+*option* field. In that case, the connection will be "downgraded" to plain TCP,
+and it will continue with a single path.
+
+This behavior is made possible by two internal components: the path manager, and
+the packet scheduler.
+
+Path Manager
+------------
+
+The Path Manager is in charge of *subflows*, from creation to deletion, and also
+address announcements. Typically, it is the client side that initiates subflows,
+and the server side that announces additional addresses via the ``ADD_ADDR`` and
+``REMOVE_ADDR`` options.
+
+Path managers are controlled by the ``net.mptcp.pm_type`` sysctl knob -- see
+mptcp-sysctl.rst. There are two types: the in-kernel one (type ``0``) where the
+same rules are applied for all the connections (see: ``ip mptcp``) ; and the
+userspace one (type ``1``), controlled by a userspace daemon (i.e. `mptcpd
+<https://mptcpd.mptcp.dev/>`_) where different rules can be applied for each
+connection. The path managers can be controlled via a Netlink API; see
+netlink_spec/mptcp_pm.rst.
+
+To be able to use multiple IP addresses on a host to create multiple *subflows*
+(paths), the default in-kernel MPTCP path-manager needs to know which IP
+addresses can be used. This can be configured with ``ip mptcp endpoint`` for
+example.
+
+Packet Scheduler
+----------------
+
+The Packet Scheduler is in charge of selecting which available *subflow(s)* to
+use to send the next data packet. It can decide to maximize the use of the
+available bandwidth, only to pick the path with the lower latency, or any other
+policy depending on the configuration.
+
+Packet schedulers are controlled by the ``net.mptcp.scheduler`` sysctl knob --
+see mptcp-sysctl.rst.
+
+
+Sockets API
+===========
+
+Creating MPTCP sockets
+----------------------
+
+On Linux, MPTCP can be used by selecting MPTCP instead of TCP when creating the
+``socket``:
+
+.. code-block:: C
+
+    int sd = socket(AF_INET(6), SOCK_STREAM, IPPROTO_MPTCP);
+
+Note that ``IPPROTO_MPTCP`` is defined as ``262``.
+
+If MPTCP is not supported, ``errno`` will be set to:
+
+- ``EINVAL``: (*Invalid argument*): MPTCP is not available, on kernels < 5.6.
+- ``EPROTONOSUPPORT`` (*Protocol not supported*): MPTCP has not been compiled,
+  on kernels >= v5.6.
+- ``ENOPROTOOPT`` (*Protocol not available*): MPTCP has been disabled using
+  ``net.mptcp.enabled`` sysctl knob; see mptcp-sysctl.rst.
+
+MPTCP is then opt-in: applications need to explicitly request it. Note that
+applications can be forced to use MPTCP with different techniques, e.g.
+``LD_PRELOAD`` (see ``mptcpize``), eBPF (see ``mptcpify``), SystemTAP,
+``GODEBUG`` (``GODEBUG=multipathtcp=1``), etc.
+
+Switching to ``IPPROTO_MPTCP`` instead of ``IPPROTO_TCP`` should be as
+transparent as possible for the userspace applications.
+
+Socket options
+--------------
+
+MPTCP supports most socket options handled by TCP. It is possible some less
+common options are not supported, but contributions are welcome.
+
+Generally, the same value is propagated to all subflows, including the ones
+created after the calls to ``setsockopt()``. eBPF can be used to set different
+values per subflow.
+
+There are some MPTCP specific socket options at the ``SOL_MPTCP`` (284) level to
+retrieve info. They fill the ``optval`` buffer of the ``getsockopt()`` system
+call:
+
+- ``MPTCP_INFO``: Uses ``struct mptcp_info``.
+- ``MPTCP_TCPINFO``: Uses ``struct mptcp_subflow_data``, followed by an array of
+  ``struct tcp_info``.
+- ``MPTCP_SUBFLOW_ADDRS``: Uses ``struct mptcp_subflow_data``, followed by an
+  array of ``mptcp_subflow_addrs``.
+- ``MPTCP_FULL_INFO``: Uses ``struct mptcp_full_info``, with one pointer to an
+  array of ``struct mptcp_subflow_info`` (including the
+  ``struct mptcp_subflow_addrs``), and one pointer to an array of
+  ``struct tcp_info``, followed by the content of ``struct mptcp_info``.
+
+Note that at the TCP level, ``TCP_IS_MPTCP`` socket option can be used to know
+if MPTCP is currently being used: the value will be set to 1 if it is.
+
+
+Design choices
+==============
+
+A new socket type has been added for MPTCP for the userspace-facing socket. The
+kernel is in charge of creating subflow sockets: they are TCP sockets where the
+behavior is modified using TCP-ULP.
+
+MPTCP listen sockets will create "plain" *accepted* TCP sockets if the
+connection request from the client didn't ask for MPTCP, making the performance
+impact minimal when MPTCP is enabled by default.
diff --git a/MAINTAINERS b/MAINTAINERS
index ba231d2f3a4e..7538152be2f1 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -15753,7 +15753,7 @@ B:	https://github.com/multipath-tcp/mptcp_net-next/issues
 T:	git https://github.com/multipath-tcp/mptcp_net-next.git export-net
 T:	git https://github.com/multipath-tcp/mptcp_net-next.git export
 F:	Documentation/netlink/specs/mptcp_pm.yaml
-F:	Documentation/networking/mptcp-sysctl.rst
+F:	Documentation/networking/mptcp*.rst
 F:	include/net/mptcp.h
 F:	include/trace/events/mptcp.h
 F:	include/uapi/linux/mptcp*.h
-- 
cgit v1.2.3-58-ga151


From 73451e9aaa24e8e44cb91e5fd6b59bf53e069c1f Mon Sep 17 00:00:00 2001
From: Abhishek Chauhan <quic_abchauha@quicinc.com>
Date: Wed, 29 May 2024 11:31:30 -0700
Subject: net: validate SO_TXTIME clockid coming from userspace

Currently there are no strict checks while setting SO_TXTIME
from userspace. With the recent development in skb->tstamp_type
clockid with unsupported clocks results in warn_on_once, which causes
unnecessary aborts in some systems which enables panic on warns.

Add validation in setsockopt to support only CLOCK_REALTIME,
CLOCK_MONOTONIC and CLOCK_TAI to be set from userspace.

Link: https://lore.kernel.org/netdev/bc037db4-58bb-4861-ac31-a361a93841d3@linux.dev/
Link: https://lore.kernel.org/lkml/6bdba7b6-fd22-4ea5-a356-12268674def1@quicinc.com/
Fixes: 1693c5db6ab8 ("net: Add additional bit to support clockid_t timestamp type")
Reported-by: syzbot+d7b227731ec589e7f4f0@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=d7b227731ec589e7f4f0
Reported-by: syzbot+30a35a2e9c5067cc43fa@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=30a35a2e9c5067cc43fa
Signed-off-by: Abhishek Chauhan <quic_abchauha@quicinc.com>
Acked-by: Martin KaFai Lau <martin.lau@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/20240529183130.1717083-1-quic_abchauha@quicinc.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/sock.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/net/core/sock.c b/net/core/sock.c
index 521e6373d4f7..69baddcfbd8c 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1083,6 +1083,17 @@ bool sockopt_capable(int cap)
 }
 EXPORT_SYMBOL(sockopt_capable);
 
+static int sockopt_validate_clockid(__kernel_clockid_t value)
+{
+	switch (value) {
+	case CLOCK_REALTIME:
+	case CLOCK_MONOTONIC:
+	case CLOCK_TAI:
+		return 0;
+	}
+	return -EINVAL;
+}
+
 /*
  *	This is meant for all protocols to use and covers goings on
  *	at the socket level. Everything here is generic.
@@ -1497,6 +1508,11 @@ set_sndbuf:
 			ret = -EPERM;
 			break;
 		}
+
+		ret = sockopt_validate_clockid(sk_txtime.clockid);
+		if (ret)
+			break;
+
 		sock_valbool_flag(sk, SOCK_TXTIME, true);
 		sk->sk_clockid = sk_txtime.clockid;
 		sk->sk_txtime_deadline_mode =
-- 
cgit v1.2.3-58-ga151


From d551d075b043821880b8afc0010ef70d050716d0 Mon Sep 17 00:00:00 2001
From: Karol Kolacinski <karol.kolacinski@intel.com>
Date: Tue, 28 May 2024 16:03:51 -0700
Subject: ice: Introduce ice_ptp_hw struct

Create new ice_ptp_hw struct and use it for all HW and PTP-related
fields from struct ice_hw.
Replace definitions with struct fields, which values are set accordingly
to a specific device.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528-next-2024-05-28-ptp-refactors-v1-1-c082739bb6f6@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_common.c |  24 +++++
 drivers/net/ethernet/intel/ice/ice_common.h |   1 +
 drivers/net/ethernet/intel/ice/ice_ptp.c    |  22 ++---
 drivers/net/ethernet/intel/ice/ice_ptp_hw.c | 134 ++++++++++++++++------------
 drivers/net/ethernet/intel/ice/ice_ptp_hw.h |   4 +-
 drivers/net/ethernet/intel/ice/ice_type.h   |  17 ++--
 6 files changed, 126 insertions(+), 76 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 24716a3b494c..61eb6af37f8e 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -239,6 +239,30 @@ bool ice_is_e810t(struct ice_hw *hw)
 	return false;
 }
 
+/**
+ * ice_is_e822 - Check if a device is E822 family device
+ * @hw: pointer to the hardware structure
+ *
+ * Return: true if the device is E822 based, false if not.
+ */
+bool ice_is_e822(struct ice_hw *hw)
+{
+	switch (hw->device_id) {
+	case ICE_DEV_ID_E822C_BACKPLANE:
+	case ICE_DEV_ID_E822C_QSFP:
+	case ICE_DEV_ID_E822C_SFP:
+	case ICE_DEV_ID_E822C_10G_BASE_T:
+	case ICE_DEV_ID_E822C_SGMII:
+	case ICE_DEV_ID_E822L_BACKPLANE:
+	case ICE_DEV_ID_E822L_SFP:
+	case ICE_DEV_ID_E822L_10G_BASE_T:
+	case ICE_DEV_ID_E822L_SGMII:
+		return true;
+	default:
+		return false;
+	}
+}
+
 /**
  * ice_is_e823
  * @hw: pointer to the hardware structure
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index ffb22c7ce28b..70f57340eb0d 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -249,6 +249,7 @@ void
 ice_stat_update32(struct ice_hw *hw, u32 reg, bool prev_stat_loaded,
 		  u64 *prev_stat, u64 *cur_stat);
 bool ice_is_e810t(struct ice_hw *hw);
+bool ice_is_e822(struct ice_hw *hw);
 bool ice_is_e823(struct ice_hw *hw);
 bool ice_is_e825c(struct ice_hw *hw);
 int
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index 0f17fc1181d2..cca9d09b2d61 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -813,7 +813,7 @@ static enum ice_tx_tstamp_work ice_ptp_tx_tstamp_owner(struct ice_pf *pf)
 	}
 	mutex_unlock(&pf->ptp.ports_owner.lock);
 
-	for (i = 0; i < ICE_MAX_QUAD; i++) {
+	for (i = 0; i < ICE_GET_QUAD_NUM(pf->hw.ptp.num_lports); i++) {
 		u64 tstamp_ready;
 		int err;
 
@@ -1027,7 +1027,7 @@ ice_ptp_release_tx_tracker(struct ice_pf *pf, struct ice_ptp_tx *tx)
 static int
 ice_ptp_init_tx_e82x(struct ice_pf *pf, struct ice_ptp_tx *tx, u8 port)
 {
-	tx->block = port / ICE_PORTS_PER_QUAD;
+	tx->block = ICE_GET_QUAD_NUM(port);
 	tx->offset = (port % ICE_PORTS_PER_QUAD) * INDEX_PER_PORT_E82X;
 	tx->len = INDEX_PER_PORT_E82X;
 	tx->has_ready_bitmap = 1;
@@ -1229,8 +1229,8 @@ static u64 ice_base_incval(struct ice_pf *pf)
  */
 static int ice_ptp_check_tx_fifo(struct ice_ptp_port *port)
 {
-	int quad = port->port_num / ICE_PORTS_PER_QUAD;
 	int offs = port->port_num % ICE_PORTS_PER_QUAD;
+	int quad = ICE_GET_QUAD_NUM(port->port_num);
 	struct ice_pf *pf;
 	struct ice_hw *hw;
 	u32 val, phy_sts;
@@ -1429,7 +1429,7 @@ void ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup)
 	if (pf->ptp.state != ICE_PTP_READY)
 		return;
 
-	if (WARN_ON_ONCE(port >= ICE_NUM_EXTERNAL_PORTS))
+	if (WARN_ON_ONCE(port >= hw->ptp.num_lports))
 		return;
 
 	ptp_port = &pf->ptp.port;
@@ -1439,7 +1439,7 @@ void ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup)
 	/* Update cached link status for this port immediately */
 	ptp_port->link_up = linkup;
 
-	switch (hw->phy_model) {
+	switch (hw->ptp.phy_model) {
 	case ICE_PHY_E810:
 		/* Do not reconfigure E810 PHY */
 		return;
@@ -1468,7 +1468,7 @@ static int ice_ptp_cfg_phy_interrupt(struct ice_pf *pf, bool ena, u32 threshold)
 
 	ice_ptp_reset_ts_memory(hw);
 
-	for (quad = 0; quad < ICE_MAX_QUAD; quad++) {
+	for (quad = 0; quad < ICE_GET_QUAD_NUM(hw->ptp.num_lports); quad++) {
 		err = ice_read_quad_reg_e82x(hw, quad, Q_REG_TX_MEM_GBL_CFG,
 					     &val);
 		if (err)
@@ -1953,7 +1953,7 @@ ice_ptp_settime64(struct ptp_clock_info *info, const struct timespec64 *ts)
 	ice_ptp_enable_all_clkout(pf);
 
 	/* Recalibrate and re-enable timestamp blocks for E822/E823 */
-	if (hw->phy_model == ICE_PHY_E82X)
+	if (hw->ptp.phy_model == ICE_PHY_E82X)
 		ice_ptp_restart_all_phy(pf);
 exit:
 	if (err) {
@@ -2578,7 +2578,7 @@ static void ice_ptp_maybe_trigger_tx_interrupt(struct ice_pf *pf)
 	if (!ice_pf_src_tmr_owned(pf))
 		return;
 
-	for (i = 0; i < ICE_MAX_QUAD; i++) {
+	for (i = 0; i < ICE_GET_QUAD_NUM(hw->ptp.num_lports); i++) {
 		u64 tstamp_ready;
 		int err;
 
@@ -3076,7 +3076,7 @@ static int ice_ptp_init_port(struct ice_pf *pf, struct ice_ptp_port *ptp_port)
 
 	mutex_init(&ptp_port->ps_lock);
 
-	switch (hw->phy_model) {
+	switch (hw->ptp.phy_model) {
 	case ICE_PHY_E810:
 		return ice_ptp_init_tx_e810(pf, &ptp_port->tx);
 	case ICE_PHY_E82X:
@@ -3171,7 +3171,7 @@ static void ice_ptp_remove_auxbus_device(struct ice_pf *pf)
  */
 static void ice_ptp_init_tx_interrupt_mode(struct ice_pf *pf)
 {
-	switch (pf->hw.phy_model) {
+	switch (pf->hw.ptp.phy_model) {
 	case ICE_PHY_E82X:
 		/* E822 based PHY has the clock owner process the interrupt
 		 * for all ports.
@@ -3207,7 +3207,7 @@ void ice_ptp_init(struct ice_pf *pf)
 
 	ptp->state = ICE_PTP_INITIALIZING;
 
-	ice_ptp_init_phy_model(hw);
+	ice_ptp_init_hw(hw);
 
 	ice_ptp_init_tx_interrupt_mode(pf);
 
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index 2b9423a173bb..22fca17a5a3c 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -288,18 +288,21 @@ static void ice_ptp_exec_tmr_cmd(struct ice_hw *hw)
 
 /**
  * ice_fill_phy_msg_e82x - Fill message data for a PHY register access
+ * @hw: pointer to the HW struct
  * @msg: the PHY message buffer to fill in
  * @port: the port to access
  * @offset: the register offset
  */
-static void
-ice_fill_phy_msg_e82x(struct ice_sbq_msg_input *msg, u8 port, u16 offset)
+static void ice_fill_phy_msg_e82x(struct ice_hw *hw,
+				  struct ice_sbq_msg_input *msg, u8 port,
+				  u16 offset)
 {
 	int phy_port, phy, quadtype;
 
-	phy_port = port % ICE_PORTS_PER_PHY_E82X;
-	phy = port / ICE_PORTS_PER_PHY_E82X;
-	quadtype = (port / ICE_PORTS_PER_QUAD) % ICE_QUADS_PER_PHY_E82X;
+	phy_port = port % hw->ptp.ports_per_phy;
+	phy = port / hw->ptp.ports_per_phy;
+	quadtype = ICE_GET_QUAD_NUM(port) %
+		   ICE_GET_QUAD_NUM(hw->ptp.ports_per_phy);
 
 	if (quadtype == 0) {
 		msg->msg_addr_low = P_Q0_L(P_0_BASE + offset, phy_port);
@@ -430,7 +433,7 @@ ice_read_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 offset, u32 *val)
 	struct ice_sbq_msg_input msg = {0};
 	int err;
 
-	ice_fill_phy_msg_e82x(&msg, port, offset);
+	ice_fill_phy_msg_e82x(hw, &msg, port, offset);
 	msg.opcode = ice_sbq_msg_rd;
 
 	err = ice_sbq_rw_reg(hw, &msg);
@@ -507,7 +510,7 @@ ice_write_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 offset, u32 val)
 	struct ice_sbq_msg_input msg = {0};
 	int err;
 
-	ice_fill_phy_msg_e82x(&msg, port, offset);
+	ice_fill_phy_msg_e82x(hw, &msg, port, offset);
 	msg.opcode = ice_sbq_msg_wr;
 	msg.data = val;
 
@@ -617,24 +620,30 @@ ice_write_64b_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 low_addr, u64 val)
 
 /**
  * ice_fill_quad_msg_e82x - Fill message data for quad register access
+ * @hw: pointer to the HW struct
  * @msg: the PHY message buffer to fill in
  * @quad: the quad to access
  * @offset: the register offset
  *
  * Fill a message buffer for accessing a register in a quad shared between
  * multiple PHYs.
+ *
+ * Return:
+ * * %0       - OK
+ * * %-EINVAL - invalid quad number
  */
-static int
-ice_fill_quad_msg_e82x(struct ice_sbq_msg_input *msg, u8 quad, u16 offset)
+static int ice_fill_quad_msg_e82x(struct ice_hw *hw,
+				  struct ice_sbq_msg_input *msg, u8 quad,
+				  u16 offset)
 {
 	u32 addr;
 
-	if (quad >= ICE_MAX_QUAD)
+	if (quad >= ICE_GET_QUAD_NUM(hw->ptp.num_lports))
 		return -EINVAL;
 
 	msg->dest_dev = rmn_0;
 
-	if ((quad % ICE_QUADS_PER_PHY_E82X) == 0)
+	if (!(quad % ICE_GET_QUAD_NUM(hw->ptp.ports_per_phy)))
 		addr = Q_0_BASE + offset;
 	else
 		addr = Q_1_BASE + offset;
@@ -661,7 +670,7 @@ ice_read_quad_reg_e82x(struct ice_hw *hw, u8 quad, u16 offset, u32 *val)
 	struct ice_sbq_msg_input msg = {0};
 	int err;
 
-	err = ice_fill_quad_msg_e82x(&msg, quad, offset);
+	err = ice_fill_quad_msg_e82x(hw, &msg, quad, offset);
 	if (err)
 		return err;
 
@@ -695,7 +704,7 @@ ice_write_quad_reg_e82x(struct ice_hw *hw, u8 quad, u16 offset, u32 val)
 	struct ice_sbq_msg_input msg = {0};
 	int err;
 
-	err = ice_fill_quad_msg_e82x(&msg, quad, offset);
+	err = ice_fill_quad_msg_e82x(hw, &msg, quad, offset);
 	if (err)
 		return err;
 
@@ -816,7 +825,7 @@ static void ice_ptp_reset_ts_memory_e82x(struct ice_hw *hw)
 {
 	unsigned int quad;
 
-	for (quad = 0; quad < ICE_MAX_QUAD; quad++)
+	for (quad = 0; quad < ICE_GET_QUAD_NUM(hw->ptp.num_lports); quad++)
 		ice_ptp_reset_ts_memory_quad_e82x(hw, quad);
 }
 
@@ -1113,7 +1122,7 @@ static int ice_ptp_set_vernier_wl(struct ice_hw *hw)
 {
 	u8 port;
 
-	for (port = 0; port < ICE_NUM_EXTERNAL_PORTS; port++) {
+	for (port = 0; port < hw->ptp.num_lports; port++) {
 		int err;
 
 		err = ice_write_phy_reg_e82x(hw, port, P_REG_WL,
@@ -1178,7 +1187,7 @@ ice_ptp_prep_phy_time_e82x(struct ice_hw *hw, u32 time)
 	 */
 	phy_time = (u64)time << 32;
 
-	for (port = 0; port < ICE_NUM_EXTERNAL_PORTS; port++) {
+	for (port = 0; port < hw->ptp.num_lports; port++) {
 		/* Tx case */
 		err = ice_write_64b_phy_reg_e82x(hw, port,
 						 P_REG_TX_TIMER_INC_PRE_L,
@@ -1281,7 +1290,7 @@ ice_ptp_prep_phy_adj_e82x(struct ice_hw *hw, s32 adj)
 	else
 		cycles = -(((s64)-adj) << 32);
 
-	for (port = 0; port < ICE_NUM_EXTERNAL_PORTS; port++) {
+	for (port = 0; port < hw->ptp.num_lports; port++) {
 		int err;
 
 		err = ice_ptp_prep_port_adj_e82x(hw, port, cycles);
@@ -1307,7 +1316,7 @@ ice_ptp_prep_phy_incval_e82x(struct ice_hw *hw, u64 incval)
 	int err;
 	u8 port;
 
-	for (port = 0; port < ICE_NUM_EXTERNAL_PORTS; port++) {
+	for (port = 0; port < hw->ptp.num_lports; port++) {
 		err = ice_write_40b_phy_reg_e82x(hw, port, P_REG_TIMETUS_L,
 						 incval);
 		if (err)
@@ -1463,7 +1472,7 @@ ice_ptp_one_port_cmd(struct ice_hw *hw, u8 configured_port,
 {
 	u8 port;
 
-	for (port = 0; port < ICE_NUM_EXTERNAL_PORTS; port++) {
+	for (port = 0; port < hw->ptp.num_lports; port++) {
 		enum ice_ptp_tmr_cmd cmd;
 		int err;
 
@@ -1493,7 +1502,7 @@ ice_ptp_port_cmd_e82x(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd)
 {
 	u8 port;
 
-	for (port = 0; port < ICE_NUM_EXTERNAL_PORTS; port++) {
+	for (port = 0; port < hw->ptp.num_lports; port++) {
 		int err;
 
 		err = ice_ptp_write_port_cmd_e82x(hw, port, cmd);
@@ -1606,7 +1615,7 @@ static void ice_phy_cfg_lane_e82x(struct ice_hw *hw, u8 port)
 		return;
 	}
 
-	quad = port / ICE_PORTS_PER_QUAD;
+	quad = ICE_GET_QUAD_NUM(port);
 
 	err = ice_read_quad_reg_e82x(hw, quad, Q_REG_TX_MEM_GBL_CFG, &val);
 	if (err) {
@@ -2636,6 +2645,17 @@ ice_get_phy_tx_tstamp_ready_e82x(struct ice_hw *hw, u8 quad, u64 *tstamp_ready)
 	return 0;
 }
 
+/**
+ * ice_ptp_init_phy_e82x - initialize PHY parameters
+ * @ptp: pointer to the PTP HW struct
+ */
+static void ice_ptp_init_phy_e82x(struct ice_ptp_hw *ptp)
+{
+	ptp->phy_model = ICE_PHY_E82X;
+	ptp->num_lports = 8;
+	ptp->ports_per_phy = 8;
+}
+
 /* E810 functions
  *
  * The following functions operate on the E810 series devices which use
@@ -2863,17 +2883,21 @@ static int ice_clear_phy_tstamp_e810(struct ice_hw *hw, u8 lport, u8 idx)
 }
 
 /**
- * ice_ptp_init_phy_e810 - Enable PTP function on the external PHY
+ * ice_ptp_init_phc_e810 - Perform E810 specific PHC initialization
  * @hw: pointer to HW struct
  *
- * Enable the timesync PTP functionality for the external PHY connected to
- * this function.
+ * Perform E810-specific PTP hardware clock initialization steps.
+ *
+ * Return: 0 on success, other error codes when failed to initialize TimeSync
  */
-int ice_ptp_init_phy_e810(struct ice_hw *hw)
+static int ice_ptp_init_phc_e810(struct ice_hw *hw)
 {
 	u8 tmr_idx;
 	int err;
 
+	/* Ensure synchronization delay is zero */
+	wr32(hw, GLTSYN_SYNC_DLAY, 0);
+
 	tmr_idx = hw->func_caps.ts_func_info.tmr_index_owned;
 	err = ice_write_phy_reg_e810(hw, ETH_GLTSYN_ENA(tmr_idx),
 				     GLTSYN_ENA_TSYN_ENA_M);
@@ -2884,21 +2908,6 @@ int ice_ptp_init_phy_e810(struct ice_hw *hw)
 	return err;
 }
 
-/**
- * ice_ptp_init_phc_e810 - Perform E810 specific PHC initialization
- * @hw: pointer to HW struct
- *
- * Perform E810-specific PTP hardware clock initialization steps.
- */
-static int ice_ptp_init_phc_e810(struct ice_hw *hw)
-{
-	/* Ensure synchronization delay is zero */
-	wr32(hw, GLTSYN_SYNC_DLAY, 0);
-
-	/* Initialize the PHY */
-	return ice_ptp_init_phy_e810(hw);
-}
-
 /**
  * ice_ptp_prep_phy_time_e810 - Prepare PHY port with initial time
  * @hw: Board private structure
@@ -3242,6 +3251,17 @@ int ice_read_pca9575_reg_e810t(struct ice_hw *hw, u8 offset, u8 *data)
 	return ice_aq_read_i2c(hw, link_topo, 0, addr, 1, data, NULL);
 }
 
+/**
+ * ice_ptp_init_phy_e810 - initialize PHY parameters
+ * @ptp: pointer to the PTP HW struct
+ */
+static void ice_ptp_init_phy_e810(struct ice_ptp_hw *ptp)
+{
+	ptp->phy_model = ICE_PHY_E810;
+	ptp->num_lports = 8;
+	ptp->ports_per_phy = 4;
+}
+
 /* Device agnostic functions
  *
  * The following functions implement shared behavior common to both E822 and
@@ -3299,18 +3319,22 @@ void ice_ptp_unlock(struct ice_hw *hw)
 }
 
 /**
- * ice_ptp_init_phy_model - Initialize hw->phy_model based on device type
+ * ice_ptp_init_hw - Initialize hw based on device type
  * @hw: pointer to the HW structure
  *
- * Determine the PHY model for the device, and initialize hw->phy_model
+ * Determine the PHY model for the device, and initialize hw
  * for use by other functions.
  */
-void ice_ptp_init_phy_model(struct ice_hw *hw)
+void ice_ptp_init_hw(struct ice_hw *hw)
 {
-	if (ice_is_e810(hw))
-		hw->phy_model = ICE_PHY_E810;
+	struct ice_ptp_hw *ptp = &hw->ptp;
+
+	if (ice_is_e822(hw) || ice_is_e823(hw))
+		ice_ptp_init_phy_e82x(ptp);
+	else if (ice_is_e810(hw))
+		ice_ptp_init_phy_e810(ptp);
 	else
-		hw->phy_model = ICE_PHY_E82X;
+		ptp->phy_model = ICE_PHY_UNSUP;
 }
 
 /**
@@ -3331,7 +3355,7 @@ static int ice_ptp_tmr_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd)
 	ice_ptp_src_cmd(hw, cmd);
 
 	/* Next, prepare the ports */
-	switch (hw->phy_model) {
+	switch (hw->ptp.phy_model) {
 	case ICE_PHY_E810:
 		err = ice_ptp_port_cmd_e810(hw, cmd);
 		break;
@@ -3383,7 +3407,7 @@ int ice_ptp_init_time(struct ice_hw *hw, u64 time)
 
 	/* PHY timers */
 	/* Fill Rx and Tx ports and send msg to PHY */
-	switch (hw->phy_model) {
+	switch (hw->ptp.phy_model) {
 	case ICE_PHY_E810:
 		err = ice_ptp_prep_phy_time_e810(hw, time & 0xFFFFFFFF);
 		break;
@@ -3425,7 +3449,7 @@ int ice_ptp_write_incval(struct ice_hw *hw, u64 incval)
 	wr32(hw, GLTSYN_SHADJ_L(tmr_idx), lower_32_bits(incval));
 	wr32(hw, GLTSYN_SHADJ_H(tmr_idx), upper_32_bits(incval));
 
-	switch (hw->phy_model) {
+	switch (hw->ptp.phy_model) {
 	case ICE_PHY_E810:
 		err = ice_ptp_prep_phy_incval_e810(hw, incval);
 		break;
@@ -3491,7 +3515,7 @@ int ice_ptp_adj_clock(struct ice_hw *hw, s32 adj)
 	wr32(hw, GLTSYN_SHADJ_L(tmr_idx), 0);
 	wr32(hw, GLTSYN_SHADJ_H(tmr_idx), adj);
 
-	switch (hw->phy_model) {
+	switch (hw->ptp.phy_model) {
 	case ICE_PHY_E810:
 		err = ice_ptp_prep_phy_adj_e810(hw, adj);
 		break;
@@ -3521,7 +3545,7 @@ int ice_ptp_adj_clock(struct ice_hw *hw, s32 adj)
  */
 int ice_read_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx, u64 *tstamp)
 {
-	switch (hw->phy_model) {
+	switch (hw->ptp.phy_model) {
 	case ICE_PHY_E810:
 		return ice_read_phy_tstamp_e810(hw, block, idx, tstamp);
 	case ICE_PHY_E82X:
@@ -3549,7 +3573,7 @@ int ice_read_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx, u64 *tstamp)
  */
 int ice_clear_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx)
 {
-	switch (hw->phy_model) {
+	switch (hw->ptp.phy_model) {
 	case ICE_PHY_E810:
 		return ice_clear_phy_tstamp_e810(hw, block, idx);
 	case ICE_PHY_E82X:
@@ -3610,7 +3634,7 @@ static int ice_get_pf_c827_idx(struct ice_hw *hw, u8 *idx)
  */
 void ice_ptp_reset_ts_memory(struct ice_hw *hw)
 {
-	switch (hw->phy_model) {
+	switch (hw->ptp.phy_model) {
 	case ICE_PHY_E82X:
 		ice_ptp_reset_ts_memory_e82x(hw);
 		break;
@@ -3636,7 +3660,7 @@ int ice_ptp_init_phc(struct ice_hw *hw)
 	/* Clear event err indications for auxiliary pins */
 	(void)rd32(hw, GLTSYN_STAT(src_idx));
 
-	switch (hw->phy_model) {
+	switch (hw->ptp.phy_model) {
 	case ICE_PHY_E810:
 		return ice_ptp_init_phc_e810(hw);
 	case ICE_PHY_E82X:
@@ -3659,7 +3683,7 @@ int ice_ptp_init_phc(struct ice_hw *hw)
  */
 int ice_get_phy_tx_tstamp_ready(struct ice_hw *hw, u8 block, u64 *tstamp_ready)
 {
-	switch (hw->phy_model) {
+	switch (hw->ptp.phy_model) {
 	case ICE_PHY_E810:
 		return ice_get_phy_tx_tstamp_ready_e810(hw, block,
 							tstamp_ready);
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
index 1f3e03124430..3dce09af0d78 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
@@ -212,6 +212,7 @@ int ice_read_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx, u64 *tstamp);
 int ice_clear_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx);
 void ice_ptp_reset_ts_memory(struct ice_hw *hw);
 int ice_ptp_init_phc(struct ice_hw *hw);
+void ice_ptp_init_hw(struct ice_hw *hw);
 int ice_get_phy_tx_tstamp_ready(struct ice_hw *hw, u8 block, u64 *tstamp_ready);
 
 /* E822 family functions */
@@ -266,7 +267,6 @@ int ice_phy_cfg_tx_offset_e82x(struct ice_hw *hw, u8 port);
 int ice_phy_cfg_rx_offset_e82x(struct ice_hw *hw, u8 port);
 
 /* E810 family functions */
-int ice_ptp_init_phy_e810(struct ice_hw *hw);
 int ice_read_sma_ctrl_e810t(struct ice_hw *hw, u8 *data);
 int ice_write_sma_ctrl_e810t(struct ice_hw *hw, u8 data);
 int ice_read_pca9575_reg_e810t(struct ice_hw *hw, u8 offset, u8 *data);
@@ -280,8 +280,6 @@ int ice_get_cgu_state(struct ice_hw *hw, u8 dpll_idx,
 		      u8 *ref_state, u8 *eec_mode, s64 *phase_offset,
 		      enum dpll_lock_status *dpll_state);
 int ice_get_cgu_rclk_pin_info(struct ice_hw *hw, u8 *base_idx, u8 *pin_num);
-
-void ice_ptp_init_phy_model(struct ice_hw *hw);
 int ice_cgu_get_output_pin_state_caps(struct ice_hw *hw, u8 pin_id,
 				      unsigned long *caps);
 
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index f0796a93f428..2d2af80a65a9 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -818,6 +818,9 @@ struct ice_mbx_data {
 	u16 async_watermark_val;
 };
 
+#define ICE_PORTS_PER_QUAD	4
+#define ICE_GET_QUAD_NUM(port) ((port) / ICE_PORTS_PER_QUAD)
+
 /* PHY model */
 enum ice_phy_model {
 	ICE_PHY_UNSUP = -1,
@@ -825,6 +828,12 @@ enum ice_phy_model {
 	ICE_PHY_E82X,
 };
 
+struct ice_ptp_hw {
+	enum ice_phy_model phy_model;
+	u8 num_lports;
+	u8 ports_per_phy;
+};
+
 /* Port hardware description */
 struct ice_hw {
 	u8 __iomem *hw_addr;
@@ -846,7 +855,6 @@ struct ice_hw {
 	u8 revision_id;
 
 	u8 pf_id;		/* device profile info */
-	enum ice_phy_model phy_model;
 
 	u16 max_burst_size;	/* driver sets this value */
 
@@ -909,12 +917,7 @@ struct ice_hw {
 	/* INTRL granularity in 1 us */
 	u8 intrl_gran;
 
-#define ICE_MAX_QUAD			2
-#define ICE_QUADS_PER_PHY_E82X		2
-#define ICE_PORTS_PER_PHY_E82X		8
-#define ICE_PORTS_PER_QUAD		4
-#define ICE_PORTS_PER_PHY_E810		4
-#define ICE_NUM_EXTERNAL_PORTS		(ICE_MAX_QUAD * ICE_PORTS_PER_QUAD)
+	struct ice_ptp_hw ptp;
 
 	/* Active package version (currently active) */
 	struct ice_pkg_ver active_pkg_ver;
-- 
cgit v1.2.3-58-ga151


From 579a2302bd16277918f862a1dd9f3f904264fa1a Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Tue, 28 May 2024 16:03:52 -0700
Subject: ice: Introduce helper to get tmr_cmd_reg values

Multiple places in the driver code need to convert enum ice_ptp_tmr_cmd
values into register bits for both the main timer and the PHY port
timers. The main MAC register has one bit scheme for timer commands,
while the PHY commands use a different scheme.

The E810 and E830 devices use the same scheme for port commands as used
for the main timer. However, E822 and ETH56G hardware has a separate
scheme used by the PHY.

Introduce helper functions to convert the timer command enumeration into
the register values, reducing some code duplication, and making it
easier to later refactor the individual port write commands.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528-next-2024-05-28-ptp-refactors-v1-2-c082739bb6f6@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_ptp_hw.c | 140 +++++++++++++++++-----------
 drivers/net/ethernet/intel/ice/ice_ptp_hw.h |   2 +-
 2 files changed, 89 insertions(+), 53 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index 22fca17a5a3c..43aa83bc54c2 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -227,40 +227,114 @@ static u64 ice_ptp_read_src_incval(struct ice_hw *hw)
 }
 
 /**
- * ice_ptp_src_cmd - Prepare source timer for a timer command
- * @hw: pointer to HW structure
+ * ice_ptp_tmr_cmd_to_src_reg - Convert to source timer command value
+ * @hw: pointer to HW struct
  * @cmd: Timer command
  *
- * Prepare the source timer for an upcoming timer sync command.
+ * Return: the source timer command register value for the given PTP timer
+ * command.
  */
-void ice_ptp_src_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd)
+static u32 ice_ptp_tmr_cmd_to_src_reg(struct ice_hw *hw,
+				      enum ice_ptp_tmr_cmd cmd)
 {
-	u32 cmd_val;
-	u8 tmr_idx;
+	u32 cmd_val, tmr_idx;
+
+	switch (cmd) {
+	case ICE_PTP_INIT_TIME:
+		cmd_val = GLTSYN_CMD_INIT_TIME;
+		break;
+	case ICE_PTP_INIT_INCVAL:
+		cmd_val = GLTSYN_CMD_INIT_INCVAL;
+		break;
+	case ICE_PTP_ADJ_TIME:
+		cmd_val = GLTSYN_CMD_ADJ_TIME;
+		break;
+	case ICE_PTP_ADJ_TIME_AT_TIME:
+		cmd_val = GLTSYN_CMD_ADJ_INIT_TIME;
+		break;
+	case ICE_PTP_NOP:
+	case ICE_PTP_READ_TIME:
+		cmd_val = GLTSYN_CMD_READ_TIME;
+		break;
+	default:
+		dev_warn(ice_hw_to_dev(hw),
+			 "Ignoring unrecognized timer command %u\n", cmd);
+		cmd_val = 0;
+	}
 
 	tmr_idx = ice_get_ptp_src_clock_index(hw);
-	cmd_val = tmr_idx << SEL_CPK_SRC;
+
+	return tmr_idx << SEL_CPK_SRC | cmd_val;
+}
+
+/**
+ * ice_ptp_tmr_cmd_to_port_reg- Convert to port timer command value
+ * @hw: pointer to HW struct
+ * @cmd: Timer command
+ *
+ * Note that some hardware families use a different command register value for
+ * the PHY ports, while other hardware families use the same register values
+ * as the source timer.
+ *
+ * Return: the PHY port timer command register value for the given PTP timer
+ * command.
+ */
+static u32 ice_ptp_tmr_cmd_to_port_reg(struct ice_hw *hw,
+				       enum ice_ptp_tmr_cmd cmd)
+{
+	u32 cmd_val, tmr_idx;
+
+	/* Certain hardware families share the same register values for the
+	 * port register and source timer register.
+	 */
+	switch (hw->ptp.phy_model) {
+	case ICE_PHY_E810:
+		return ice_ptp_tmr_cmd_to_src_reg(hw, cmd) & TS_CMD_MASK_E810;
+	default:
+		break;
+	}
 
 	switch (cmd) {
 	case ICE_PTP_INIT_TIME:
-		cmd_val |= GLTSYN_CMD_INIT_TIME;
+		cmd_val = PHY_CMD_INIT_TIME;
 		break;
 	case ICE_PTP_INIT_INCVAL:
-		cmd_val |= GLTSYN_CMD_INIT_INCVAL;
+		cmd_val = PHY_CMD_INIT_INCVAL;
 		break;
 	case ICE_PTP_ADJ_TIME:
-		cmd_val |= GLTSYN_CMD_ADJ_TIME;
+		cmd_val = PHY_CMD_ADJ_TIME;
 		break;
 	case ICE_PTP_ADJ_TIME_AT_TIME:
-		cmd_val |= GLTSYN_CMD_ADJ_INIT_TIME;
+		cmd_val = PHY_CMD_ADJ_TIME_AT_TIME;
 		break;
 	case ICE_PTP_READ_TIME:
-		cmd_val |= GLTSYN_CMD_READ_TIME;
+		cmd_val = PHY_CMD_READ_TIME;
 		break;
 	case ICE_PTP_NOP:
+		cmd_val = 0;
 		break;
+	default:
+		dev_warn(ice_hw_to_dev(hw),
+			 "Ignoring unrecognized timer command %u\n", cmd);
+		cmd_val = 0;
 	}
 
+	tmr_idx = ice_get_ptp_src_clock_index(hw);
+
+	return tmr_idx << SEL_PHY_SRC | cmd_val;
+}
+
+/**
+ * ice_ptp_src_cmd - Prepare source timer for a timer command
+ * @hw: pointer to HW structure
+ * @cmd: Timer command
+ *
+ * Prepare the source timer for an upcoming timer sync command.
+ */
+void ice_ptp_src_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd)
+{
+	u32 cmd_val = ice_ptp_tmr_cmd_to_src_reg(hw, cmd);
+
 	wr32(hw, GLTSYN_CMD, cmd_val);
 }
 
@@ -3029,47 +3103,9 @@ static int ice_ptp_prep_phy_incval_e810(struct ice_hw *hw, u64 incval)
  */
 static int ice_ptp_port_cmd_e810(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd)
 {
-	u32 cmd_val, val;
-	int err;
+	u32 val = ice_ptp_tmr_cmd_to_port_reg(hw, cmd);
 
-	switch (cmd) {
-	case ICE_PTP_INIT_TIME:
-		cmd_val = GLTSYN_CMD_INIT_TIME;
-		break;
-	case ICE_PTP_INIT_INCVAL:
-		cmd_val = GLTSYN_CMD_INIT_INCVAL;
-		break;
-	case ICE_PTP_ADJ_TIME:
-		cmd_val = GLTSYN_CMD_ADJ_TIME;
-		break;
-	case ICE_PTP_READ_TIME:
-		cmd_val = GLTSYN_CMD_READ_TIME;
-		break;
-	case ICE_PTP_ADJ_TIME_AT_TIME:
-		cmd_val = GLTSYN_CMD_ADJ_INIT_TIME;
-		break;
-	case ICE_PTP_NOP:
-		return 0;
-	}
-
-	/* Read, modify, write */
-	err = ice_read_phy_reg_e810(hw, ETH_GLTSYN_CMD, &val);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to read GLTSYN_CMD, err %d\n", err);
-		return err;
-	}
-
-	/* Modify necessary bits only and perform write */
-	val &= ~TS_CMD_MASK_E810;
-	val |= cmd_val;
-
-	err = ice_write_phy_reg_e810(hw, ETH_GLTSYN_CMD, val);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to write back GLTSYN_CMD, err %d\n", err);
-		return err;
-	}
-
-	return 0;
+	return ice_write_phy_reg_e810(hw, E810_ETH_GLTSYN_CMD, val);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
index 3dce09af0d78..6246de3bacf3 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
@@ -485,7 +485,7 @@ int ice_cgu_get_output_pin_state_caps(struct ice_hw *hw, u8 pin_id,
 #define ETH_GLTSYN_SHADJ_H(_i)		(0x0300037C + ((_i) * 32))
 
 /* E810 timer command register */
-#define ETH_GLTSYN_CMD			0x03000344
+#define E810_ETH_GLTSYN_CMD		0x03000344
 
 /* Source timer incval macros */
 #define INCVAL_HIGH_M			0xFF
-- 
cgit v1.2.3-58-ga151


From 0d80bbe254198ab20982d5c508c55518d32c033b Mon Sep 17 00:00:00 2001
From: Sergey Temerkhanov <sergey.temerkhanov@intel.com>
Date: Tue, 28 May 2024 16:03:53 -0700
Subject: ice: Implement Tx interrupt enablement functions

Introduce functions enabling/disabling Tx TS interrupts
for the E822 and ETH56G PHYs

Signed-off-by: Sergey Temerkhanov <sergey.temerkhanov@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528-next-2024-05-28-ptp-refactors-v1-3-c082739bb6f6@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_ptp.c    | 66 +++++++++++++++--------------
 drivers/net/ethernet/intel/ice/ice_ptp_hw.c | 31 ++++++++++++++
 drivers/net/ethernet/intel/ice/ice_ptp_hw.h |  4 +-
 3 files changed, 66 insertions(+), 35 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index cca9d09b2d61..412555194c97 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -1457,42 +1457,46 @@ void ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup)
  * @ena: bool value to enable or disable interrupt
  * @threshold: Minimum number of packets at which intr is triggered
  *
- * Utility function to enable or disable Tx timestamp interrupt and threshold
+ * Utility function to configure all the PHY interrupt settings, including
+ * whether the PHY interrupt is enabled, and what threshold to use. Also
+ * configures The E82X timestamp owner to react to interrupts from all PHYs.
+ *
+ * Return: 0 on success, -EOPNOTSUPP when PHY model incorrect, other error codes
+ * when failed to configure PHY interrupt for E82X
  */
 static int ice_ptp_cfg_phy_interrupt(struct ice_pf *pf, bool ena, u32 threshold)
 {
+	struct device *dev = ice_pf_to_dev(pf);
 	struct ice_hw *hw = &pf->hw;
-	int err = 0;
-	int quad;
-	u32 val;
 
 	ice_ptp_reset_ts_memory(hw);
 
-	for (quad = 0; quad < ICE_GET_QUAD_NUM(hw->ptp.num_lports); quad++) {
-		err = ice_read_quad_reg_e82x(hw, quad, Q_REG_TX_MEM_GBL_CFG,
-					     &val);
-		if (err)
-			break;
-
-		if (ena) {
-			val |= Q_REG_TX_MEM_GBL_CFG_INTR_ENA_M;
-			val &= ~Q_REG_TX_MEM_GBL_CFG_INTR_THR_M;
-			val |= FIELD_PREP(Q_REG_TX_MEM_GBL_CFG_INTR_THR_M,
-					  threshold);
-		} else {
-			val &= ~Q_REG_TX_MEM_GBL_CFG_INTR_ENA_M;
+	switch (hw->ptp.phy_model) {
+	case ICE_PHY_E82X: {
+		int quad;
+
+		for (quad = 0; quad < ICE_GET_QUAD_NUM(hw->ptp.num_lports);
+		     quad++) {
+			int err;
+
+			err = ice_phy_cfg_intr_e82x(hw, quad, ena, threshold);
+			if (err) {
+				dev_err(dev, "Failed to configure PHY interrupt for quad %d, err %d\n",
+					quad, err);
+				return err;
+			}
 		}
 
-		err = ice_write_quad_reg_e82x(hw, quad, Q_REG_TX_MEM_GBL_CFG,
-					      val);
-		if (err)
-			break;
+		return 0;
+	}
+	case ICE_PHY_E810:
+		return 0;
+	case ICE_PHY_UNSUP:
+	default:
+		dev_warn(dev, "%s: Unexpected PHY model %d\n", __func__,
+			 hw->ptp.phy_model);
+		return -EOPNOTSUPP;
 	}
-
-	if (err)
-		dev_err(ice_pf_to_dev(pf), "PTP failed in intr ena, err %d\n",
-			err);
-	return err;
 }
 
 /**
@@ -3010,12 +3014,10 @@ static int ice_ptp_init_owner(struct ice_pf *pf)
 	/* Release the global hardware lock */
 	ice_ptp_unlock(hw);
 
-	if (!ice_is_e810(hw)) {
-		/* Enable quad interrupts */
-		err = ice_ptp_cfg_phy_interrupt(pf, true, 1);
-		if (err)
-			goto err_exit;
-	}
+	/* Configure PHY interrupt settings */
+	err = ice_ptp_cfg_phy_interrupt(pf, true, 1);
+	if (err)
+		goto err_exit;
 
 	/* Ensure we have a clock device */
 	err = ice_ptp_create_clock(pf);
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index 43aa83bc54c2..0a4026c8a3ba 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -2719,6 +2719,37 @@ ice_get_phy_tx_tstamp_ready_e82x(struct ice_hw *hw, u8 quad, u64 *tstamp_ready)
 	return 0;
 }
 
+/**
+ * ice_phy_cfg_intr_e82x - Configure TX timestamp interrupt
+ * @hw: pointer to the HW struct
+ * @quad: the timestamp quad
+ * @ena: enable or disable interrupt
+ * @threshold: interrupt threshold
+ *
+ * Configure TX timestamp interrupt for the specified quad
+ *
+ * Return: 0 on success, other error codes when failed to read/write quad
+ */
+
+int ice_phy_cfg_intr_e82x(struct ice_hw *hw, u8 quad, bool ena, u8 threshold)
+{
+	int err;
+	u32 val;
+
+	err = ice_read_quad_reg_e82x(hw, quad, Q_REG_TX_MEM_GBL_CFG, &val);
+	if (err)
+		return err;
+
+	val &= ~Q_REG_TX_MEM_GBL_CFG_INTR_ENA_M;
+	if (ena) {
+		val |= Q_REG_TX_MEM_GBL_CFG_INTR_ENA_M;
+		val &= ~Q_REG_TX_MEM_GBL_CFG_INTR_THR_M;
+		val |= FIELD_PREP(Q_REG_TX_MEM_GBL_CFG_INTR_THR_M, threshold);
+	}
+
+	return ice_write_quad_reg_e82x(hw, quad, Q_REG_TX_MEM_GBL_CFG, val);
+}
+
 /**
  * ice_ptp_init_phy_e82x - initialize PHY parameters
  * @ptp: pointer to the PTP HW struct
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
index 6246de3bacf3..5645b20a9f87 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
@@ -265,6 +265,7 @@ int ice_stop_phy_timer_e82x(struct ice_hw *hw, u8 port, bool soft_reset);
 int ice_start_phy_timer_e82x(struct ice_hw *hw, u8 port);
 int ice_phy_cfg_tx_offset_e82x(struct ice_hw *hw, u8 port);
 int ice_phy_cfg_rx_offset_e82x(struct ice_hw *hw, u8 port);
+int ice_phy_cfg_intr_e82x(struct ice_hw *hw, u8 quad, bool ena, u8 threshold);
 
 /* E810 family functions */
 int ice_read_sma_ctrl_e810t(struct ice_hw *hw, u8 *data);
@@ -342,11 +343,8 @@ int ice_cgu_get_output_pin_state_caps(struct ice_hw *hw, u8 pin_id,
 #define Q_REG_TX_MEM_GBL_CFG		0xC08
 #define Q_REG_TX_MEM_GBL_CFG_LANE_TYPE_S	0
 #define Q_REG_TX_MEM_GBL_CFG_LANE_TYPE_M	BIT(0)
-#define Q_REG_TX_MEM_GBL_CFG_TX_TYPE_S	1
 #define Q_REG_TX_MEM_GBL_CFG_TX_TYPE_M	ICE_M(0xFF, 1)
-#define Q_REG_TX_MEM_GBL_CFG_INTR_THR_S	9
 #define Q_REG_TX_MEM_GBL_CFG_INTR_THR_M ICE_M(0x3F, 9)
-#define Q_REG_TX_MEM_GBL_CFG_INTR_ENA_S	15
 #define Q_REG_TX_MEM_GBL_CFG_INTR_ENA_M	BIT(15)
 
 /* Tx Timestamp data registers */
-- 
cgit v1.2.3-58-ga151


From c199b31a043c4fe6500b13c486fd06430f464b40 Mon Sep 17 00:00:00 2001
From: Karol Kolacinski <karol.kolacinski@intel.com>
Date: Tue, 28 May 2024 16:03:54 -0700
Subject: ice: Add PHY OFFSET_READY register clearing

Add a possibility to mark all transmitted/received timestamps as invalid
by clearing PHY OFFSET_READY registers.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528-next-2024-05-28-ptp-refactors-v1-4-c082739bb6f6@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_ptp.c    | 11 ++++++----
 drivers/net/ethernet/intel/ice/ice_ptp_hw.c | 34 +++++++++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_ptp_hw.h |  1 +
 3 files changed, 42 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index 412555194c97..4ed2213247f7 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -1933,11 +1933,14 @@ ice_ptp_settime64(struct ptp_clock_info *info, const struct timespec64 *ts)
 	struct ice_hw *hw = &pf->hw;
 	int err;
 
-	/* For Vernier mode, we need to recalibrate after new settime
-	 * Start with disabling timestamp block
+	/* For Vernier mode on E82X, we need to recalibrate after new settime.
+	 * Start with marking timestamps as invalid.
 	 */
-	if (pf->ptp.port.link_up)
-		ice_ptp_port_phy_stop(&pf->ptp.port);
+	if (hw->ptp.phy_model == ICE_PHY_E82X) {
+		err = ice_ptp_clear_phy_offset_ready_e82x(hw);
+		if (err)
+			dev_warn(ice_pf_to_dev(pf), "Failed to mark timestamps as invalid before settime\n");
+	}
 
 	if (!ice_ptp_lock(hw)) {
 		err = -EBUSY;
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index 0a4026c8a3ba..25b3544c4862 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -2409,6 +2409,40 @@ int ice_phy_cfg_rx_offset_e82x(struct ice_hw *hw, u8 port)
 	return 0;
 }
 
+/**
+ * ice_ptp_clear_phy_offset_ready_e82x - Clear PHY TX_/RX_OFFSET_READY registers
+ * @hw: pointer to the HW struct
+ *
+ * Clear PHY TX_/RX_OFFSET_READY registers, effectively marking all transmitted
+ * and received timestamps as invalid.
+ *
+ * Return: 0 on success, other error codes when failed to write to PHY
+ */
+int ice_ptp_clear_phy_offset_ready_e82x(struct ice_hw *hw)
+{
+	u8 port;
+
+	for (port = 0; port < hw->ptp.num_lports; port++) {
+		int err;
+
+		err = ice_write_phy_reg_e82x(hw, port, P_REG_TX_OR, 0);
+		if (err) {
+			dev_warn(ice_hw_to_dev(hw),
+				 "Failed to clear PHY TX_OFFSET_READY register\n");
+			return err;
+		}
+
+		err = ice_write_phy_reg_e82x(hw, port, P_REG_RX_OR, 0);
+		if (err) {
+			dev_warn(ice_hw_to_dev(hw),
+				 "Failed to clear PHY RX_OFFSET_READY register\n");
+			return err;
+		}
+	}
+
+	return 0;
+}
+
 /**
  * ice_read_phy_and_phc_time_e82x - Simultaneously capture PHC and PHY time
  * @hw: pointer to the HW struct
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
index 5645b20a9f87..5223e17d2806 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
@@ -208,6 +208,7 @@ int ice_ptp_init_time(struct ice_hw *hw, u64 time);
 int ice_ptp_write_incval(struct ice_hw *hw, u64 incval);
 int ice_ptp_write_incval_locked(struct ice_hw *hw, u64 incval);
 int ice_ptp_adj_clock(struct ice_hw *hw, s32 adj);
+int ice_ptp_clear_phy_offset_ready_e82x(struct ice_hw *hw);
 int ice_read_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx, u64 *tstamp);
 int ice_clear_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx);
 void ice_ptp_reset_ts_memory(struct ice_hw *hw);
-- 
cgit v1.2.3-58-ga151


From 39ecb29df855ce46f70456996c518789445f2a46 Mon Sep 17 00:00:00 2001
From: Sergey Temerkhanov <sergey.temerkhanov@intel.com>
Date: Tue, 28 May 2024 16:03:55 -0700
Subject: ice: Move CGU block

Move CGU block to the beginning of ice_ptp_hw.c

Signed-off-by: Sergey Temerkhanov <sergey.temerkhanov@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528-next-2024-05-28-ptp-refactors-v1-5-c082739bb6f6@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_ptp_hw.c | 1409 ++++++++++++++-------------
 drivers/net/ethernet/intel/ice/ice_ptp_hw.h |    2 +-
 2 files changed, 707 insertions(+), 704 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index 25b3544c4862..7224bce7aa28 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -227,563 +227,575 @@ static u64 ice_ptp_read_src_incval(struct ice_hw *hw)
 }
 
 /**
- * ice_ptp_tmr_cmd_to_src_reg - Convert to source timer command value
- * @hw: pointer to HW struct
- * @cmd: Timer command
+ * ice_read_cgu_reg_e82x - Read a CGU register
+ * @hw: pointer to the HW struct
+ * @addr: Register address to read
+ * @val: storage for register value read
  *
- * Return: the source timer command register value for the given PTP timer
- * command.
+ * Read the contents of a register of the Clock Generation Unit. Only
+ * applicable to E822 devices.
+ *
+ * Return: 0 on success, other error codes when failed to read from CGU
  */
-static u32 ice_ptp_tmr_cmd_to_src_reg(struct ice_hw *hw,
-				      enum ice_ptp_tmr_cmd cmd)
+static int ice_read_cgu_reg_e82x(struct ice_hw *hw, u32 addr, u32 *val)
 {
-	u32 cmd_val, tmr_idx;
+	struct ice_sbq_msg_input cgu_msg = {
+		.opcode = ice_sbq_msg_rd,
+		.dest_dev = cgu,
+		.msg_addr_low = addr
+	};
+	int err;
 
-	switch (cmd) {
-	case ICE_PTP_INIT_TIME:
-		cmd_val = GLTSYN_CMD_INIT_TIME;
-		break;
-	case ICE_PTP_INIT_INCVAL:
-		cmd_val = GLTSYN_CMD_INIT_INCVAL;
-		break;
-	case ICE_PTP_ADJ_TIME:
-		cmd_val = GLTSYN_CMD_ADJ_TIME;
-		break;
-	case ICE_PTP_ADJ_TIME_AT_TIME:
-		cmd_val = GLTSYN_CMD_ADJ_INIT_TIME;
-		break;
-	case ICE_PTP_NOP:
-	case ICE_PTP_READ_TIME:
-		cmd_val = GLTSYN_CMD_READ_TIME;
-		break;
-	default:
-		dev_warn(ice_hw_to_dev(hw),
-			 "Ignoring unrecognized timer command %u\n", cmd);
-		cmd_val = 0;
+	err = ice_sbq_rw_reg(hw, &cgu_msg);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read CGU register 0x%04x, err %d\n",
+			  addr, err);
+		return err;
 	}
 
-	tmr_idx = ice_get_ptp_src_clock_index(hw);
+	*val = cgu_msg.data;
 
-	return tmr_idx << SEL_CPK_SRC | cmd_val;
+	return 0;
 }
 
 /**
- * ice_ptp_tmr_cmd_to_port_reg- Convert to port timer command value
- * @hw: pointer to HW struct
- * @cmd: Timer command
+ * ice_write_cgu_reg_e82x - Write a CGU register
+ * @hw: pointer to the HW struct
+ * @addr: Register address to write
+ * @val: value to write into the register
  *
- * Note that some hardware families use a different command register value for
- * the PHY ports, while other hardware families use the same register values
- * as the source timer.
+ * Write the specified value to a register of the Clock Generation Unit. Only
+ * applicable to E822 devices.
  *
- * Return: the PHY port timer command register value for the given PTP timer
- * command.
+ * Return: 0 on success, other error codes when failed to write to CGU
  */
-static u32 ice_ptp_tmr_cmd_to_port_reg(struct ice_hw *hw,
-				       enum ice_ptp_tmr_cmd cmd)
+static int ice_write_cgu_reg_e82x(struct ice_hw *hw, u32 addr, u32 val)
 {
-	u32 cmd_val, tmr_idx;
-
-	/* Certain hardware families share the same register values for the
-	 * port register and source timer register.
-	 */
-	switch (hw->ptp.phy_model) {
-	case ICE_PHY_E810:
-		return ice_ptp_tmr_cmd_to_src_reg(hw, cmd) & TS_CMD_MASK_E810;
-	default:
-		break;
-	}
+	struct ice_sbq_msg_input cgu_msg = {
+		.opcode = ice_sbq_msg_wr,
+		.dest_dev = cgu,
+		.msg_addr_low = addr,
+		.data = val
+	};
+	int err;
 
-	switch (cmd) {
-	case ICE_PTP_INIT_TIME:
-		cmd_val = PHY_CMD_INIT_TIME;
-		break;
-	case ICE_PTP_INIT_INCVAL:
-		cmd_val = PHY_CMD_INIT_INCVAL;
-		break;
-	case ICE_PTP_ADJ_TIME:
-		cmd_val = PHY_CMD_ADJ_TIME;
-		break;
-	case ICE_PTP_ADJ_TIME_AT_TIME:
-		cmd_val = PHY_CMD_ADJ_TIME_AT_TIME;
-		break;
-	case ICE_PTP_READ_TIME:
-		cmd_val = PHY_CMD_READ_TIME;
-		break;
-	case ICE_PTP_NOP:
-		cmd_val = 0;
-		break;
-	default:
-		dev_warn(ice_hw_to_dev(hw),
-			 "Ignoring unrecognized timer command %u\n", cmd);
-		cmd_val = 0;
+	err = ice_sbq_rw_reg(hw, &cgu_msg);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write CGU register 0x%04x, err %d\n",
+			  addr, err);
+		return err;
 	}
 
-	tmr_idx = ice_get_ptp_src_clock_index(hw);
-
-	return tmr_idx << SEL_PHY_SRC | cmd_val;
+	return err;
 }
 
 /**
- * ice_ptp_src_cmd - Prepare source timer for a timer command
- * @hw: pointer to HW structure
- * @cmd: Timer command
+ * ice_clk_freq_str - Convert time_ref_freq to string
+ * @clk_freq: Clock frequency
  *
- * Prepare the source timer for an upcoming timer sync command.
+ * Return: specified TIME_REF clock frequency converted to a string
  */
-void ice_ptp_src_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd)
+static const char *ice_clk_freq_str(enum ice_time_ref_freq clk_freq)
 {
-	u32 cmd_val = ice_ptp_tmr_cmd_to_src_reg(hw, cmd);
-
-	wr32(hw, GLTSYN_CMD, cmd_val);
+	switch (clk_freq) {
+	case ICE_TIME_REF_FREQ_25_000:
+		return "25 MHz";
+	case ICE_TIME_REF_FREQ_122_880:
+		return "122.88 MHz";
+	case ICE_TIME_REF_FREQ_125_000:
+		return "125 MHz";
+	case ICE_TIME_REF_FREQ_153_600:
+		return "153.6 MHz";
+	case ICE_TIME_REF_FREQ_156_250:
+		return "156.25 MHz";
+	case ICE_TIME_REF_FREQ_245_760:
+		return "245.76 MHz";
+	default:
+		return "Unknown";
+	}
 }
 
 /**
- * ice_ptp_exec_tmr_cmd - Execute all prepared timer commands
- * @hw: pointer to HW struct
+ * ice_clk_src_str - Convert time_ref_src to string
+ * @clk_src: Clock source
  *
- * Write the SYNC_EXEC_CMD bit to the GLTSYN_CMD_SYNC register, and flush the
- * write immediately. This triggers the hardware to begin executing all of the
- * source and PHY timer commands synchronously.
+ * Return: specified clock source converted to its string name
  */
-static void ice_ptp_exec_tmr_cmd(struct ice_hw *hw)
+static const char *ice_clk_src_str(enum ice_clk_src clk_src)
 {
-	struct ice_pf *pf = container_of(hw, struct ice_pf, hw);
-
-	guard(spinlock)(&pf->adapter->ptp_gltsyn_time_lock);
-	wr32(hw, GLTSYN_CMD_SYNC, SYNC_EXEC_CMD);
-	ice_flush(hw);
+	switch (clk_src) {
+	case ICE_CLK_SRC_TCX0:
+		return "TCX0";
+	case ICE_CLK_SRC_TIME_REF:
+		return "TIME_REF";
+	default:
+		return "Unknown";
+	}
 }
 
-/* E822 family functions
- *
- * The following functions operate on the E822 family of devices.
- */
-
 /**
- * ice_fill_phy_msg_e82x - Fill message data for a PHY register access
+ * ice_cfg_cgu_pll_e82x - Configure the Clock Generation Unit
  * @hw: pointer to the HW struct
- * @msg: the PHY message buffer to fill in
- * @port: the port to access
- * @offset: the register offset
+ * @clk_freq: Clock frequency to program
+ * @clk_src: Clock source to select (TIME_REF, or TCX0)
+ *
+ * Configure the Clock Generation Unit with the desired clock frequency and
+ * time reference, enabling the PLL which drives the PTP hardware clock.
+ *
+ * Return:
+ * * %0       - success
+ * * %-EINVAL - input parameters are incorrect
+ * * %-EBUSY  - failed to lock TS PLL
+ * * %other   - CGU read/write failure
  */
-static void ice_fill_phy_msg_e82x(struct ice_hw *hw,
-				  struct ice_sbq_msg_input *msg, u8 port,
-				  u16 offset)
+static int ice_cfg_cgu_pll_e82x(struct ice_hw *hw,
+				enum ice_time_ref_freq clk_freq,
+				enum ice_clk_src clk_src)
 {
-	int phy_port, phy, quadtype;
-
-	phy_port = port % hw->ptp.ports_per_phy;
-	phy = port / hw->ptp.ports_per_phy;
-	quadtype = ICE_GET_QUAD_NUM(port) %
-		   ICE_GET_QUAD_NUM(hw->ptp.ports_per_phy);
+	union tspll_ro_bwm_lf bwm_lf;
+	union nac_cgu_dword19 dw19;
+	union nac_cgu_dword22 dw22;
+	union nac_cgu_dword24 dw24;
+	union nac_cgu_dword9 dw9;
+	int err;
 
-	if (quadtype == 0) {
-		msg->msg_addr_low = P_Q0_L(P_0_BASE + offset, phy_port);
-		msg->msg_addr_high = P_Q0_H(P_0_BASE + offset, phy_port);
-	} else {
-		msg->msg_addr_low = P_Q1_L(P_4_BASE + offset, phy_port);
-		msg->msg_addr_high = P_Q1_H(P_4_BASE + offset, phy_port);
+	if (clk_freq >= NUM_ICE_TIME_REF_FREQ) {
+		dev_warn(ice_hw_to_dev(hw), "Invalid TIME_REF frequency %u\n",
+			 clk_freq);
+		return -EINVAL;
 	}
 
-	if (phy == 0)
-		msg->dest_dev = rmn_0;
-	else if (phy == 1)
-		msg->dest_dev = rmn_1;
-	else
-		msg->dest_dev = rmn_2;
-}
-
-/**
- * ice_is_64b_phy_reg_e82x - Check if this is a 64bit PHY register
- * @low_addr: the low address to check
- * @high_addr: on return, contains the high address of the 64bit register
- *
- * Checks if the provided low address is one of the known 64bit PHY values
- * represented as two 32bit registers. If it is, return the appropriate high
- * register offset to use.
- */
-static bool ice_is_64b_phy_reg_e82x(u16 low_addr, u16 *high_addr)
-{
-	switch (low_addr) {
-	case P_REG_PAR_PCS_TX_OFFSET_L:
-		*high_addr = P_REG_PAR_PCS_TX_OFFSET_U;
-		return true;
-	case P_REG_PAR_PCS_RX_OFFSET_L:
-		*high_addr = P_REG_PAR_PCS_RX_OFFSET_U;
-		return true;
-	case P_REG_PAR_TX_TIME_L:
-		*high_addr = P_REG_PAR_TX_TIME_U;
-		return true;
-	case P_REG_PAR_RX_TIME_L:
-		*high_addr = P_REG_PAR_RX_TIME_U;
-		return true;
-	case P_REG_TOTAL_TX_OFFSET_L:
-		*high_addr = P_REG_TOTAL_TX_OFFSET_U;
-		return true;
-	case P_REG_TOTAL_RX_OFFSET_L:
-		*high_addr = P_REG_TOTAL_RX_OFFSET_U;
-		return true;
-	case P_REG_UIX66_10G_40G_L:
-		*high_addr = P_REG_UIX66_10G_40G_U;
-		return true;
-	case P_REG_UIX66_25G_100G_L:
-		*high_addr = P_REG_UIX66_25G_100G_U;
-		return true;
-	case P_REG_TX_CAPTURE_L:
-		*high_addr = P_REG_TX_CAPTURE_U;
-		return true;
-	case P_REG_RX_CAPTURE_L:
-		*high_addr = P_REG_RX_CAPTURE_U;
-		return true;
-	case P_REG_TX_TIMER_INC_PRE_L:
-		*high_addr = P_REG_TX_TIMER_INC_PRE_U;
-		return true;
-	case P_REG_RX_TIMER_INC_PRE_L:
-		*high_addr = P_REG_RX_TIMER_INC_PRE_U;
-		return true;
-	default:
-		return false;
+	if (clk_src >= NUM_ICE_CLK_SRC) {
+		dev_warn(ice_hw_to_dev(hw), "Invalid clock source %u\n",
+			 clk_src);
+		return -EINVAL;
 	}
-}
 
-/**
- * ice_is_40b_phy_reg_e82x - Check if this is a 40bit PHY register
- * @low_addr: the low address to check
- * @high_addr: on return, contains the high address of the 40bit value
- *
- * Checks if the provided low address is one of the known 40bit PHY values
- * split into two registers with the lower 8 bits in the low register and the
- * upper 32 bits in the high register. If it is, return the appropriate high
- * register offset to use.
- */
-static bool ice_is_40b_phy_reg_e82x(u16 low_addr, u16 *high_addr)
-{
-	switch (low_addr) {
-	case P_REG_TIMETUS_L:
-		*high_addr = P_REG_TIMETUS_U;
-		return true;
-	case P_REG_PAR_RX_TUS_L:
-		*high_addr = P_REG_PAR_RX_TUS_U;
-		return true;
-	case P_REG_PAR_TX_TUS_L:
-		*high_addr = P_REG_PAR_TX_TUS_U;
-		return true;
-	case P_REG_PCS_RX_TUS_L:
-		*high_addr = P_REG_PCS_RX_TUS_U;
-		return true;
-	case P_REG_PCS_TX_TUS_L:
-		*high_addr = P_REG_PCS_TX_TUS_U;
-		return true;
-	case P_REG_DESK_PAR_RX_TUS_L:
-		*high_addr = P_REG_DESK_PAR_RX_TUS_U;
-		return true;
-	case P_REG_DESK_PAR_TX_TUS_L:
-		*high_addr = P_REG_DESK_PAR_TX_TUS_U;
-		return true;
-	case P_REG_DESK_PCS_RX_TUS_L:
-		*high_addr = P_REG_DESK_PCS_RX_TUS_U;
-		return true;
-	case P_REG_DESK_PCS_TX_TUS_L:
-		*high_addr = P_REG_DESK_PCS_TX_TUS_U;
-		return true;
-	default:
-		return false;
+	if (clk_src == ICE_CLK_SRC_TCX0 &&
+	    clk_freq != ICE_TIME_REF_FREQ_25_000) {
+		dev_warn(ice_hw_to_dev(hw),
+			 "TCX0 only supports 25 MHz frequency\n");
+		return -EINVAL;
 	}
-}
 
-/**
- * ice_read_phy_reg_e82x - Read a PHY register
- * @hw: pointer to the HW struct
- * @port: PHY port to read from
- * @offset: PHY register offset to read
- * @val: on return, the contents read from the PHY
- *
- * Read a PHY register for the given port over the device sideband queue.
- */
-static int
-ice_read_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 offset, u32 *val)
-{
-	struct ice_sbq_msg_input msg = {0};
-	int err;
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD9, &dw9.val);
+	if (err)
+		return err;
 
-	ice_fill_phy_msg_e82x(hw, &msg, port, offset);
-	msg.opcode = ice_sbq_msg_rd;
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD24, &dw24.val);
+	if (err)
+		return err;
 
-	err = ice_sbq_rw_reg(hw, &msg);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
-			  err);
+	err = ice_read_cgu_reg_e82x(hw, TSPLL_RO_BWM_LF, &bwm_lf.val);
+	if (err)
 		return err;
+
+	/* Log the current clock configuration */
+	ice_debug(hw, ICE_DBG_PTP, "Current CGU configuration -- %s, clk_src %s, clk_freq %s, PLL %s\n",
+		  dw24.field.ts_pll_enable ? "enabled" : "disabled",
+		  ice_clk_src_str(dw24.field.time_ref_sel),
+		  ice_clk_freq_str(dw9.field.time_ref_freq_sel),
+		  bwm_lf.field.plllock_true_lock_cri ? "locked" : "unlocked");
+
+	/* Disable the PLL before changing the clock source or frequency */
+	if (dw24.field.ts_pll_enable) {
+		dw24.field.ts_pll_enable = 0;
+
+		err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD24, dw24.val);
+		if (err)
+			return err;
 	}
 
-	*val = msg.data;
+	/* Set the frequency */
+	dw9.field.time_ref_freq_sel = clk_freq;
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD9, dw9.val);
+	if (err)
+		return err;
 
-	return 0;
-}
+	/* Configure the TS PLL feedback divisor */
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD19, &dw19.val);
+	if (err)
+		return err;
 
-/**
- * ice_read_64b_phy_reg_e82x - Read a 64bit value from PHY registers
- * @hw: pointer to the HW struct
- * @port: PHY port to read from
- * @low_addr: offset of the lower register to read from
- * @val: on return, the contents of the 64bit value from the PHY registers
- *
- * Reads the two registers associated with a 64bit value and returns it in the
- * val pointer. The offset always specifies the lower register offset to use.
- * The high offset is looked up. This function only operates on registers
- * known to be two parts of a 64bit value.
- */
-static int
-ice_read_64b_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 low_addr, u64 *val)
-{
-	u32 low, high;
-	u16 high_addr;
-	int err;
+	dw19.field.tspll_fbdiv_intgr = e822_cgu_params[clk_freq].feedback_div;
+	dw19.field.tspll_ndivratio = 1;
 
-	/* Only operate on registers known to be split into two 32bit
-	 * registers.
-	 */
-	if (!ice_is_64b_phy_reg_e82x(low_addr, &high_addr)) {
-		ice_debug(hw, ICE_DBG_PTP, "Invalid 64b register addr 0x%08x\n",
-			  low_addr);
-		return -EINVAL;
-	}
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD19, dw19.val);
+	if (err)
+		return err;
 
-	err = ice_read_phy_reg_e82x(hw, port, low_addr, &low);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to read from low register 0x%08x\n, err %d",
-			  low_addr, err);
+	/* Configure the TS PLL post divisor */
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD22, &dw22.val);
+	if (err)
 		return err;
-	}
 
-	err = ice_read_phy_reg_e82x(hw, port, high_addr, &high);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to read from high register 0x%08x\n, err %d",
-			  high_addr, err);
+	dw22.field.time1588clk_div = e822_cgu_params[clk_freq].post_pll_div;
+	dw22.field.time1588clk_sel_div2 = 0;
+
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD22, dw22.val);
+	if (err)
+		return err;
+
+	/* Configure the TS PLL pre divisor and clock source */
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD24, &dw24.val);
+	if (err)
+		return err;
+
+	dw24.field.ref1588_ck_div = e822_cgu_params[clk_freq].refclk_pre_div;
+	dw24.field.tspll_fbdiv_frac = e822_cgu_params[clk_freq].frac_n_div;
+	dw24.field.time_ref_sel = clk_src;
+
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD24, dw24.val);
+	if (err)
 		return err;
+
+	/* Finally, enable the PLL */
+	dw24.field.ts_pll_enable = 1;
+
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD24, dw24.val);
+	if (err)
+		return err;
+
+	/* Wait to verify if the PLL locks */
+	usleep_range(1000, 5000);
+
+	err = ice_read_cgu_reg_e82x(hw, TSPLL_RO_BWM_LF, &bwm_lf.val);
+	if (err)
+		return err;
+
+	if (!bwm_lf.field.plllock_true_lock_cri) {
+		dev_warn(ice_hw_to_dev(hw), "CGU PLL failed to lock\n");
+		return -EBUSY;
 	}
 
-	*val = (u64)high << 32 | low;
+	/* Log the current clock configuration */
+	ice_debug(hw, ICE_DBG_PTP, "New CGU configuration -- %s, clk_src %s, clk_freq %s, PLL %s\n",
+		  dw24.field.ts_pll_enable ? "enabled" : "disabled",
+		  ice_clk_src_str(dw24.field.time_ref_sel),
+		  ice_clk_freq_str(dw9.field.time_ref_freq_sel),
+		  bwm_lf.field.plllock_true_lock_cri ? "locked" : "unlocked");
 
 	return 0;
 }
 
 /**
- * ice_write_phy_reg_e82x - Write a PHY register
- * @hw: pointer to the HW struct
- * @port: PHY port to write to
- * @offset: PHY register offset to write
- * @val: The value to write to the register
+ * ice_init_cgu_e82x - Initialize CGU with settings from firmware
+ * @hw: pointer to the HW structure
  *
- * Write a PHY register for the given port over the device sideband queue.
+ * Initialize the Clock Generation Unit of the E822 device.
+ *
+ * Return: 0 on success, other error codes when failed to read/write/cfg CGU
  */
-static int
-ice_write_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 offset, u32 val)
+static int ice_init_cgu_e82x(struct ice_hw *hw)
 {
-	struct ice_sbq_msg_input msg = {0};
+	struct ice_ts_func_info *ts_info = &hw->func_caps.ts_func_info;
+	union tspll_cntr_bist_settings cntr_bist;
 	int err;
 
-	ice_fill_phy_msg_e82x(hw, &msg, port, offset);
-	msg.opcode = ice_sbq_msg_wr;
-	msg.data = val;
+	err = ice_read_cgu_reg_e82x(hw, TSPLL_CNTR_BIST_SETTINGS,
+				    &cntr_bist.val);
+	if (err)
+		return err;
 
-	err = ice_sbq_rw_reg(hw, &msg);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
-			  err);
+	/* Disable sticky lock detection so lock err reported is accurate */
+	cntr_bist.field.i_plllock_sel_0 = 0;
+	cntr_bist.field.i_plllock_sel_1 = 0;
+
+	err = ice_write_cgu_reg_e82x(hw, TSPLL_CNTR_BIST_SETTINGS,
+				     cntr_bist.val);
+	if (err)
 		return err;
+
+	/* Configure the CGU PLL using the parameters from the function
+	 * capabilities.
+	 */
+	return ice_cfg_cgu_pll_e82x(hw, ts_info->time_ref,
+				   (enum ice_clk_src)ts_info->clk_src);
+}
+
+/**
+ * ice_ptp_tmr_cmd_to_src_reg - Convert to source timer command value
+ * @hw: pointer to HW struct
+ * @cmd: Timer command
+ *
+ * Return: the source timer command register value for the given PTP timer
+ * command.
+ */
+static u32 ice_ptp_tmr_cmd_to_src_reg(struct ice_hw *hw,
+				      enum ice_ptp_tmr_cmd cmd)
+{
+	u32 cmd_val, tmr_idx;
+
+	switch (cmd) {
+	case ICE_PTP_INIT_TIME:
+		cmd_val = GLTSYN_CMD_INIT_TIME;
+		break;
+	case ICE_PTP_INIT_INCVAL:
+		cmd_val = GLTSYN_CMD_INIT_INCVAL;
+		break;
+	case ICE_PTP_ADJ_TIME:
+		cmd_val = GLTSYN_CMD_ADJ_TIME;
+		break;
+	case ICE_PTP_ADJ_TIME_AT_TIME:
+		cmd_val = GLTSYN_CMD_ADJ_INIT_TIME;
+		break;
+	case ICE_PTP_NOP:
+	case ICE_PTP_READ_TIME:
+		cmd_val = GLTSYN_CMD_READ_TIME;
+		break;
+	default:
+		dev_warn(ice_hw_to_dev(hw),
+			 "Ignoring unrecognized timer command %u\n", cmd);
+		cmd_val = 0;
 	}
 
-	return 0;
+	tmr_idx = ice_get_ptp_src_clock_index(hw);
+
+	return tmr_idx << SEL_CPK_SRC | cmd_val;
 }
 
 /**
- * ice_write_40b_phy_reg_e82x - Write a 40b value to the PHY
- * @hw: pointer to the HW struct
- * @port: port to write to
- * @low_addr: offset of the low register
- * @val: 40b value to write
+ * ice_ptp_tmr_cmd_to_port_reg- Convert to port timer command value
+ * @hw: pointer to HW struct
+ * @cmd: Timer command
  *
- * Write the provided 40b value to the two associated registers by splitting
- * it up into two chunks, the lower 8 bits and the upper 32 bits.
+ * Note that some hardware families use a different command register value for
+ * the PHY ports, while other hardware families use the same register values
+ * as the source timer.
+ *
+ * Return: the PHY port timer command register value for the given PTP timer
+ * command.
  */
-static int
-ice_write_40b_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 low_addr, u64 val)
+static u32 ice_ptp_tmr_cmd_to_port_reg(struct ice_hw *hw,
+				       enum ice_ptp_tmr_cmd cmd)
 {
-	u32 low, high;
-	u16 high_addr;
-	int err;
+	u32 cmd_val, tmr_idx;
 
-	/* Only operate on registers known to be split into a lower 8 bit
-	 * register and an upper 32 bit register.
+	/* Certain hardware families share the same register values for the
+	 * port register and source timer register.
 	 */
-	if (!ice_is_40b_phy_reg_e82x(low_addr, &high_addr)) {
-		ice_debug(hw, ICE_DBG_PTP, "Invalid 40b register addr 0x%08x\n",
-			  low_addr);
-		return -EINVAL;
+	switch (hw->ptp.phy_model) {
+	case ICE_PHY_E810:
+		return ice_ptp_tmr_cmd_to_src_reg(hw, cmd) & TS_CMD_MASK_E810;
+	default:
+		break;
 	}
 
-	low = (u32)(val & P_REG_40B_LOW_M);
-	high = (u32)(val >> P_REG_40B_HIGH_S);
-
-	err = ice_write_phy_reg_e82x(hw, port, low_addr, low);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to write to low register 0x%08x\n, err %d",
-			  low_addr, err);
-		return err;
+	switch (cmd) {
+	case ICE_PTP_INIT_TIME:
+		cmd_val = PHY_CMD_INIT_TIME;
+		break;
+	case ICE_PTP_INIT_INCVAL:
+		cmd_val = PHY_CMD_INIT_INCVAL;
+		break;
+	case ICE_PTP_ADJ_TIME:
+		cmd_val = PHY_CMD_ADJ_TIME;
+		break;
+	case ICE_PTP_ADJ_TIME_AT_TIME:
+		cmd_val = PHY_CMD_ADJ_TIME_AT_TIME;
+		break;
+	case ICE_PTP_READ_TIME:
+		cmd_val = PHY_CMD_READ_TIME;
+		break;
+	case ICE_PTP_NOP:
+		cmd_val = 0;
+		break;
+	default:
+		dev_warn(ice_hw_to_dev(hw),
+			 "Ignoring unrecognized timer command %u\n", cmd);
+		cmd_val = 0;
 	}
 
-	err = ice_write_phy_reg_e82x(hw, port, high_addr, high);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to write to high register 0x%08x\n, err %d",
-			  high_addr, err);
-		return err;
-	}
+	tmr_idx = ice_get_ptp_src_clock_index(hw);
 
-	return 0;
+	return tmr_idx << SEL_PHY_SRC | cmd_val;
 }
 
 /**
- * ice_write_64b_phy_reg_e82x - Write a 64bit value to PHY registers
- * @hw: pointer to the HW struct
- * @port: PHY port to read from
- * @low_addr: offset of the lower register to read from
- * @val: the contents of the 64bit value to write to PHY
+ * ice_ptp_src_cmd - Prepare source timer for a timer command
+ * @hw: pointer to HW structure
+ * @cmd: Timer command
  *
- * Write the 64bit value to the two associated 32bit PHY registers. The offset
- * is always specified as the lower register, and the high address is looked
- * up. This function only operates on registers known to be two parts of
- * a 64bit value.
+ * Prepare the source timer for an upcoming timer sync command.
  */
-static int
-ice_write_64b_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 low_addr, u64 val)
+void ice_ptp_src_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd)
 {
-	u32 low, high;
-	u16 high_addr;
-	int err;
-
-	/* Only operate on registers known to be split into two 32bit
-	 * registers.
-	 */
-	if (!ice_is_64b_phy_reg_e82x(low_addr, &high_addr)) {
-		ice_debug(hw, ICE_DBG_PTP, "Invalid 64b register addr 0x%08x\n",
-			  low_addr);
-		return -EINVAL;
-	}
-
-	low = lower_32_bits(val);
-	high = upper_32_bits(val);
+	u32 cmd_val = ice_ptp_tmr_cmd_to_src_reg(hw, cmd);
 
-	err = ice_write_phy_reg_e82x(hw, port, low_addr, low);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to write to low register 0x%08x\n, err %d",
-			  low_addr, err);
-		return err;
-	}
+	wr32(hw, GLTSYN_CMD, cmd_val);
+}
 
-	err = ice_write_phy_reg_e82x(hw, port, high_addr, high);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to write to high register 0x%08x\n, err %d",
-			  high_addr, err);
-		return err;
-	}
+/**
+ * ice_ptp_exec_tmr_cmd - Execute all prepared timer commands
+ * @hw: pointer to HW struct
+ *
+ * Write the SYNC_EXEC_CMD bit to the GLTSYN_CMD_SYNC register, and flush the
+ * write immediately. This triggers the hardware to begin executing all of the
+ * source and PHY timer commands synchronously.
+ */
+static void ice_ptp_exec_tmr_cmd(struct ice_hw *hw)
+{
+	struct ice_pf *pf = container_of(hw, struct ice_pf, hw);
 
-	return 0;
+	guard(spinlock)(&pf->adapter->ptp_gltsyn_time_lock);
+	wr32(hw, GLTSYN_CMD_SYNC, SYNC_EXEC_CMD);
+	ice_flush(hw);
 }
 
+/* E822 family functions
+ *
+ * The following functions operate on the E822 family of devices.
+ */
+
 /**
- * ice_fill_quad_msg_e82x - Fill message data for quad register access
+ * ice_fill_phy_msg_e82x - Fill message data for a PHY register access
  * @hw: pointer to the HW struct
  * @msg: the PHY message buffer to fill in
- * @quad: the quad to access
+ * @port: the port to access
  * @offset: the register offset
- *
- * Fill a message buffer for accessing a register in a quad shared between
- * multiple PHYs.
- *
- * Return:
- * * %0       - OK
- * * %-EINVAL - invalid quad number
  */
-static int ice_fill_quad_msg_e82x(struct ice_hw *hw,
-				  struct ice_sbq_msg_input *msg, u8 quad,
+static void ice_fill_phy_msg_e82x(struct ice_hw *hw,
+				  struct ice_sbq_msg_input *msg, u8 port,
 				  u16 offset)
 {
-	u32 addr;
+	int phy_port, phy, quadtype;
 
-	if (quad >= ICE_GET_QUAD_NUM(hw->ptp.num_lports))
-		return -EINVAL;
+	phy_port = port % hw->ptp.ports_per_phy;
+	phy = port / hw->ptp.ports_per_phy;
+	quadtype = ICE_GET_QUAD_NUM(port) %
+		   ICE_GET_QUAD_NUM(hw->ptp.ports_per_phy);
 
-	msg->dest_dev = rmn_0;
+	if (quadtype == 0) {
+		msg->msg_addr_low = P_Q0_L(P_0_BASE + offset, phy_port);
+		msg->msg_addr_high = P_Q0_H(P_0_BASE + offset, phy_port);
+	} else {
+		msg->msg_addr_low = P_Q1_L(P_4_BASE + offset, phy_port);
+		msg->msg_addr_high = P_Q1_H(P_4_BASE + offset, phy_port);
+	}
 
-	if (!(quad % ICE_GET_QUAD_NUM(hw->ptp.ports_per_phy)))
-		addr = Q_0_BASE + offset;
+	if (phy == 0)
+		msg->dest_dev = rmn_0;
+	else if (phy == 1)
+		msg->dest_dev = rmn_1;
 	else
-		addr = Q_1_BASE + offset;
-
-	msg->msg_addr_low = lower_16_bits(addr);
-	msg->msg_addr_high = upper_16_bits(addr);
-
-	return 0;
+		msg->dest_dev = rmn_2;
 }
 
 /**
- * ice_read_quad_reg_e82x - Read a PHY quad register
- * @hw: pointer to the HW struct
- * @quad: quad to read from
- * @offset: quad register offset to read
- * @val: on return, the contents read from the quad
+ * ice_is_64b_phy_reg_e82x - Check if this is a 64bit PHY register
+ * @low_addr: the low address to check
+ * @high_addr: on return, contains the high address of the 64bit register
  *
- * Read a quad register over the device sideband queue. Quad registers are
- * shared between multiple PHYs.
+ * Checks if the provided low address is one of the known 64bit PHY values
+ * represented as two 32bit registers. If it is, return the appropriate high
+ * register offset to use.
  */
-int
-ice_read_quad_reg_e82x(struct ice_hw *hw, u8 quad, u16 offset, u32 *val)
+static bool ice_is_64b_phy_reg_e82x(u16 low_addr, u16 *high_addr)
 {
-	struct ice_sbq_msg_input msg = {0};
-	int err;
-
-	err = ice_fill_quad_msg_e82x(hw, &msg, quad, offset);
-	if (err)
-		return err;
-
-	msg.opcode = ice_sbq_msg_rd;
-
-	err = ice_sbq_rw_reg(hw, &msg);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
-			  err);
-		return err;
+	switch (low_addr) {
+	case P_REG_PAR_PCS_TX_OFFSET_L:
+		*high_addr = P_REG_PAR_PCS_TX_OFFSET_U;
+		return true;
+	case P_REG_PAR_PCS_RX_OFFSET_L:
+		*high_addr = P_REG_PAR_PCS_RX_OFFSET_U;
+		return true;
+	case P_REG_PAR_TX_TIME_L:
+		*high_addr = P_REG_PAR_TX_TIME_U;
+		return true;
+	case P_REG_PAR_RX_TIME_L:
+		*high_addr = P_REG_PAR_RX_TIME_U;
+		return true;
+	case P_REG_TOTAL_TX_OFFSET_L:
+		*high_addr = P_REG_TOTAL_TX_OFFSET_U;
+		return true;
+	case P_REG_TOTAL_RX_OFFSET_L:
+		*high_addr = P_REG_TOTAL_RX_OFFSET_U;
+		return true;
+	case P_REG_UIX66_10G_40G_L:
+		*high_addr = P_REG_UIX66_10G_40G_U;
+		return true;
+	case P_REG_UIX66_25G_100G_L:
+		*high_addr = P_REG_UIX66_25G_100G_U;
+		return true;
+	case P_REG_TX_CAPTURE_L:
+		*high_addr = P_REG_TX_CAPTURE_U;
+		return true;
+	case P_REG_RX_CAPTURE_L:
+		*high_addr = P_REG_RX_CAPTURE_U;
+		return true;
+	case P_REG_TX_TIMER_INC_PRE_L:
+		*high_addr = P_REG_TX_TIMER_INC_PRE_U;
+		return true;
+	case P_REG_RX_TIMER_INC_PRE_L:
+		*high_addr = P_REG_RX_TIMER_INC_PRE_U;
+		return true;
+	default:
+		return false;
 	}
+}
 
-	*val = msg.data;
-
-	return 0;
+/**
+ * ice_is_40b_phy_reg_e82x - Check if this is a 40bit PHY register
+ * @low_addr: the low address to check
+ * @high_addr: on return, contains the high address of the 40bit value
+ *
+ * Checks if the provided low address is one of the known 40bit PHY values
+ * split into two registers with the lower 8 bits in the low register and the
+ * upper 32 bits in the high register. If it is, return the appropriate high
+ * register offset to use.
+ */
+static bool ice_is_40b_phy_reg_e82x(u16 low_addr, u16 *high_addr)
+{
+	switch (low_addr) {
+	case P_REG_TIMETUS_L:
+		*high_addr = P_REG_TIMETUS_U;
+		return true;
+	case P_REG_PAR_RX_TUS_L:
+		*high_addr = P_REG_PAR_RX_TUS_U;
+		return true;
+	case P_REG_PAR_TX_TUS_L:
+		*high_addr = P_REG_PAR_TX_TUS_U;
+		return true;
+	case P_REG_PCS_RX_TUS_L:
+		*high_addr = P_REG_PCS_RX_TUS_U;
+		return true;
+	case P_REG_PCS_TX_TUS_L:
+		*high_addr = P_REG_PCS_TX_TUS_U;
+		return true;
+	case P_REG_DESK_PAR_RX_TUS_L:
+		*high_addr = P_REG_DESK_PAR_RX_TUS_U;
+		return true;
+	case P_REG_DESK_PAR_TX_TUS_L:
+		*high_addr = P_REG_DESK_PAR_TX_TUS_U;
+		return true;
+	case P_REG_DESK_PCS_RX_TUS_L:
+		*high_addr = P_REG_DESK_PCS_RX_TUS_U;
+		return true;
+	case P_REG_DESK_PCS_TX_TUS_L:
+		*high_addr = P_REG_DESK_PCS_TX_TUS_U;
+		return true;
+	default:
+		return false;
+	}
 }
 
 /**
- * ice_write_quad_reg_e82x - Write a PHY quad register
+ * ice_read_phy_reg_e82x - Read a PHY register
  * @hw: pointer to the HW struct
- * @quad: quad to write to
- * @offset: quad register offset to write
- * @val: The value to write to the register
+ * @port: PHY port to read from
+ * @offset: PHY register offset to read
+ * @val: on return, the contents read from the PHY
  *
- * Write a quad register over the device sideband queue. Quad registers are
- * shared between multiple PHYs.
+ * Read a PHY register for the given port over the device sideband queue.
  */
-int
-ice_write_quad_reg_e82x(struct ice_hw *hw, u8 quad, u16 offset, u32 val)
+static int
+ice_read_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 offset, u32 *val)
 {
 	struct ice_sbq_msg_input msg = {0};
 	int err;
 
-	err = ice_fill_quad_msg_e82x(hw, &msg, quad, offset);
-	if (err)
-		return err;
-
-	msg.opcode = ice_sbq_msg_wr;
-	msg.data = val;
+	ice_fill_phy_msg_e82x(hw, &msg, port, offset);
+	msg.opcode = ice_sbq_msg_rd;
 
 	err = ice_sbq_rw_reg(hw, &msg);
 	if (err) {
@@ -792,85 +804,81 @@ ice_write_quad_reg_e82x(struct ice_hw *hw, u8 quad, u16 offset, u32 val)
 		return err;
 	}
 
+	*val = msg.data;
+
 	return 0;
 }
 
 /**
- * ice_read_phy_tstamp_e82x - Read a PHY timestamp out of the quad block
+ * ice_read_64b_phy_reg_e82x - Read a 64bit value from PHY registers
  * @hw: pointer to the HW struct
- * @quad: the quad to read from
- * @idx: the timestamp index to read
- * @tstamp: on return, the 40bit timestamp value
+ * @port: PHY port to read from
+ * @low_addr: offset of the lower register to read from
+ * @val: on return, the contents of the 64bit value from the PHY registers
  *
- * Read a 40bit timestamp value out of the two associated registers in the
- * quad memory block that is shared between the internal PHYs of the E822
- * family of devices.
+ * Reads the two registers associated with a 64bit value and returns it in the
+ * val pointer. The offset always specifies the lower register offset to use.
+ * The high offset is looked up. This function only operates on registers
+ * known to be two parts of a 64bit value.
  */
 static int
-ice_read_phy_tstamp_e82x(struct ice_hw *hw, u8 quad, u8 idx, u64 *tstamp)
+ice_read_64b_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 low_addr, u64 *val)
 {
-	u16 lo_addr, hi_addr;
-	u32 lo, hi;
+	u32 low, high;
+	u16 high_addr;
 	int err;
 
-	lo_addr = (u16)TS_L(Q_REG_TX_MEMORY_BANK_START, idx);
-	hi_addr = (u16)TS_H(Q_REG_TX_MEMORY_BANK_START, idx);
+	/* Only operate on registers known to be split into two 32bit
+	 * registers.
+	 */
+	if (!ice_is_64b_phy_reg_e82x(low_addr, &high_addr)) {
+		ice_debug(hw, ICE_DBG_PTP, "Invalid 64b register addr 0x%08x\n",
+			  low_addr);
+		return -EINVAL;
+	}
 
-	err = ice_read_quad_reg_e82x(hw, quad, lo_addr, &lo);
+	err = ice_read_phy_reg_e82x(hw, port, low_addr, &low);
 	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to read low PTP timestamp register, err %d\n",
-			  err);
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read from low register 0x%08x\n, err %d",
+			  low_addr, err);
 		return err;
 	}
 
-	err = ice_read_quad_reg_e82x(hw, quad, hi_addr, &hi);
+	err = ice_read_phy_reg_e82x(hw, port, high_addr, &high);
 	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to read high PTP timestamp register, err %d\n",
-			  err);
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read from high register 0x%08x\n, err %d",
+			  high_addr, err);
 		return err;
 	}
 
-	/* For E822 based internal PHYs, the timestamp is reported with the
-	 * lower 8 bits in the low register, and the upper 32 bits in the high
-	 * register.
-	 */
-	*tstamp = ((u64)hi) << TS_PHY_HIGH_S | ((u64)lo & TS_PHY_LOW_M);
+	*val = (u64)high << 32 | low;
 
 	return 0;
 }
 
 /**
- * ice_clear_phy_tstamp_e82x - Clear a timestamp from the quad block
+ * ice_write_phy_reg_e82x - Write a PHY register
  * @hw: pointer to the HW struct
- * @quad: the quad to read from
- * @idx: the timestamp index to reset
- *
- * Read the timestamp out of the quad to clear its timestamp status bit from
- * the PHY quad block that is shared between the internal PHYs of the E822
- * devices.
- *
- * Note that unlike E810, software cannot directly write to the quad memory
- * bank registers. E822 relies on the ice_get_phy_tx_tstamp_ready() function
- * to determine which timestamps are valid. Reading a timestamp auto-clears
- * the valid bit.
- *
- * To directly clear the contents of the timestamp block entirely, discarding
- * all timestamp data at once, software should instead use
- * ice_ptp_reset_ts_memory_quad_e82x().
+ * @port: PHY port to write to
+ * @offset: PHY register offset to write
+ * @val: The value to write to the register
  *
- * This function should only be called on an idx whose bit is set according to
- * ice_get_phy_tx_tstamp_ready().
+ * Write a PHY register for the given port over the device sideband queue.
  */
 static int
-ice_clear_phy_tstamp_e82x(struct ice_hw *hw, u8 quad, u8 idx)
+ice_write_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 offset, u32 val)
 {
-	u64 unused_tstamp;
+	struct ice_sbq_msg_input msg = {0};
 	int err;
 
-	err = ice_read_phy_tstamp_e82x(hw, quad, idx, &unused_tstamp);
+	ice_fill_phy_msg_e82x(hw, &msg, port, offset);
+	msg.opcode = ice_sbq_msg_wr;
+	msg.data = val;
+
+	err = ice_sbq_rw_reg(hw, &msg);
 	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to read the timestamp register for quad %u, idx %u, err %d\n",
-			  quad, idx, err);
+		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
+			  err);
 		return err;
 	}
 
@@ -878,312 +886,307 @@ ice_clear_phy_tstamp_e82x(struct ice_hw *hw, u8 quad, u8 idx)
 }
 
 /**
- * ice_ptp_reset_ts_memory_quad_e82x - Clear all timestamps from the quad block
- * @hw: pointer to the HW struct
- * @quad: the quad to read from
- *
- * Clear all timestamps from the PHY quad block that is shared between the
- * internal PHYs on the E822 devices.
- */
-void ice_ptp_reset_ts_memory_quad_e82x(struct ice_hw *hw, u8 quad)
-{
-	ice_write_quad_reg_e82x(hw, quad, Q_REG_TS_CTRL, Q_REG_TS_CTRL_M);
-	ice_write_quad_reg_e82x(hw, quad, Q_REG_TS_CTRL, ~(u32)Q_REG_TS_CTRL_M);
-}
-
-/**
- * ice_ptp_reset_ts_memory_e82x - Clear all timestamps from all quad blocks
- * @hw: pointer to the HW struct
- */
-static void ice_ptp_reset_ts_memory_e82x(struct ice_hw *hw)
-{
-	unsigned int quad;
-
-	for (quad = 0; quad < ICE_GET_QUAD_NUM(hw->ptp.num_lports); quad++)
-		ice_ptp_reset_ts_memory_quad_e82x(hw, quad);
-}
-
-/**
- * ice_read_cgu_reg_e82x - Read a CGU register
+ * ice_write_40b_phy_reg_e82x - Write a 40b value to the PHY
  * @hw: pointer to the HW struct
- * @addr: Register address to read
- * @val: storage for register value read
+ * @port: port to write to
+ * @low_addr: offset of the low register
+ * @val: 40b value to write
  *
- * Read the contents of a register of the Clock Generation Unit. Only
- * applicable to E822 devices.
+ * Write the provided 40b value to the two associated registers by splitting
+ * it up into two chunks, the lower 8 bits and the upper 32 bits.
  */
 static int
-ice_read_cgu_reg_e82x(struct ice_hw *hw, u32 addr, u32 *val)
+ice_write_40b_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 low_addr, u64 val)
 {
-	struct ice_sbq_msg_input cgu_msg;
+	u32 low, high;
+	u16 high_addr;
 	int err;
 
-	cgu_msg.opcode = ice_sbq_msg_rd;
-	cgu_msg.dest_dev = cgu;
-	cgu_msg.msg_addr_low = addr;
-	cgu_msg.msg_addr_high = 0x0;
+	/* Only operate on registers known to be split into a lower 8 bit
+	 * register and an upper 32 bit register.
+	 */
+	if (!ice_is_40b_phy_reg_e82x(low_addr, &high_addr)) {
+		ice_debug(hw, ICE_DBG_PTP, "Invalid 40b register addr 0x%08x\n",
+			  low_addr);
+		return -EINVAL;
+	}
+	low = FIELD_GET(P_REG_40B_LOW_M, val);
+	high = (u32)(val >> P_REG_40B_HIGH_S);
 
-	err = ice_sbq_rw_reg(hw, &cgu_msg);
+	err = ice_write_phy_reg_e82x(hw, port, low_addr, low);
 	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to read CGU register 0x%04x, err %d\n",
-			  addr, err);
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write to low register 0x%08x\n, err %d",
+			  low_addr, err);
 		return err;
 	}
 
-	*val = cgu_msg.data;
+	err = ice_write_phy_reg_e82x(hw, port, high_addr, high);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write to high register 0x%08x\n, err %d",
+			  high_addr, err);
+		return err;
+	}
 
-	return err;
+	return 0;
 }
 
 /**
- * ice_write_cgu_reg_e82x - Write a CGU register
+ * ice_write_64b_phy_reg_e82x - Write a 64bit value to PHY registers
  * @hw: pointer to the HW struct
- * @addr: Register address to write
- * @val: value to write into the register
+ * @port: PHY port to read from
+ * @low_addr: offset of the lower register to read from
+ * @val: the contents of the 64bit value to write to PHY
  *
- * Write the specified value to a register of the Clock Generation Unit. Only
- * applicable to E822 devices.
+ * Write the 64bit value to the two associated 32bit PHY registers. The offset
+ * is always specified as the lower register, and the high address is looked
+ * up. This function only operates on registers known to be two parts of
+ * a 64bit value.
  */
 static int
-ice_write_cgu_reg_e82x(struct ice_hw *hw, u32 addr, u32 val)
+ice_write_64b_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 low_addr, u64 val)
 {
-	struct ice_sbq_msg_input cgu_msg;
+	u32 low, high;
+	u16 high_addr;
 	int err;
 
-	cgu_msg.opcode = ice_sbq_msg_wr;
-	cgu_msg.dest_dev = cgu;
-	cgu_msg.msg_addr_low = addr;
-	cgu_msg.msg_addr_high = 0x0;
-	cgu_msg.data = val;
+	/* Only operate on registers known to be split into two 32bit
+	 * registers.
+	 */
+	if (!ice_is_64b_phy_reg_e82x(low_addr, &high_addr)) {
+		ice_debug(hw, ICE_DBG_PTP, "Invalid 64b register addr 0x%08x\n",
+			  low_addr);
+		return -EINVAL;
+	}
+
+	low = lower_32_bits(val);
+	high = upper_32_bits(val);
 
-	err = ice_sbq_rw_reg(hw, &cgu_msg);
+	err = ice_write_phy_reg_e82x(hw, port, low_addr, low);
 	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to write CGU register 0x%04x, err %d\n",
-			  addr, err);
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write to low register 0x%08x\n, err %d",
+			  low_addr, err);
 		return err;
 	}
 
-	return err;
-}
-
-/**
- * ice_clk_freq_str - Convert time_ref_freq to string
- * @clk_freq: Clock frequency
- *
- * Convert the specified TIME_REF clock frequency to a string.
- */
-static const char *ice_clk_freq_str(u8 clk_freq)
-{
-	switch ((enum ice_time_ref_freq)clk_freq) {
-	case ICE_TIME_REF_FREQ_25_000:
-		return "25 MHz";
-	case ICE_TIME_REF_FREQ_122_880:
-		return "122.88 MHz";
-	case ICE_TIME_REF_FREQ_125_000:
-		return "125 MHz";
-	case ICE_TIME_REF_FREQ_153_600:
-		return "153.6 MHz";
-	case ICE_TIME_REF_FREQ_156_250:
-		return "156.25 MHz";
-	case ICE_TIME_REF_FREQ_245_760:
-		return "245.76 MHz";
-	default:
-		return "Unknown";
+	err = ice_write_phy_reg_e82x(hw, port, high_addr, high);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write to high register 0x%08x\n, err %d",
+			  high_addr, err);
+		return err;
 	}
+
+	return 0;
 }
 
 /**
- * ice_clk_src_str - Convert time_ref_src to string
- * @clk_src: Clock source
+ * ice_fill_quad_msg_e82x - Fill message data for quad register access
+ * @hw: pointer to the HW struct
+ * @msg: the PHY message buffer to fill in
+ * @quad: the quad to access
+ * @offset: the register offset
+ *
+ * Fill a message buffer for accessing a register in a quad shared between
+ * multiple PHYs.
  *
- * Convert the specified clock source to its string name.
+ * Return:
+ * * %0       - OK
+ * * %-EINVAL - invalid quad number
  */
-static const char *ice_clk_src_str(u8 clk_src)
+static int ice_fill_quad_msg_e82x(struct ice_hw *hw,
+				  struct ice_sbq_msg_input *msg, u8 quad,
+				  u16 offset)
 {
-	switch ((enum ice_clk_src)clk_src) {
-	case ICE_CLK_SRC_TCX0:
-		return "TCX0";
-	case ICE_CLK_SRC_TIME_REF:
-		return "TIME_REF";
-	default:
-		return "Unknown";
-	}
+	u32 addr;
+
+	if (quad >= ICE_GET_QUAD_NUM(hw->ptp.num_lports))
+		return -EINVAL;
+
+	msg->dest_dev = rmn_0;
+
+	if (!(quad % ICE_GET_QUAD_NUM(hw->ptp.ports_per_phy)))
+		addr = Q_0_BASE + offset;
+	else
+		addr = Q_1_BASE + offset;
+
+	msg->msg_addr_low = lower_16_bits(addr);
+	msg->msg_addr_high = upper_16_bits(addr);
+
+	return 0;
 }
 
 /**
- * ice_cfg_cgu_pll_e82x - Configure the Clock Generation Unit
+ * ice_read_quad_reg_e82x - Read a PHY quad register
  * @hw: pointer to the HW struct
- * @clk_freq: Clock frequency to program
- * @clk_src: Clock source to select (TIME_REF, or TCX0)
+ * @quad: quad to read from
+ * @offset: quad register offset to read
+ * @val: on return, the contents read from the quad
  *
- * Configure the Clock Generation Unit with the desired clock frequency and
- * time reference, enabling the PLL which drives the PTP hardware clock.
+ * Read a quad register over the device sideband queue. Quad registers are
+ * shared between multiple PHYs.
  */
-static int
-ice_cfg_cgu_pll_e82x(struct ice_hw *hw, enum ice_time_ref_freq clk_freq,
-		     enum ice_clk_src clk_src)
+int
+ice_read_quad_reg_e82x(struct ice_hw *hw, u8 quad, u16 offset, u32 *val)
 {
-	union tspll_ro_bwm_lf bwm_lf;
-	union nac_cgu_dword19 dw19;
-	union nac_cgu_dword22 dw22;
-	union nac_cgu_dword24 dw24;
-	union nac_cgu_dword9 dw9;
+	struct ice_sbq_msg_input msg = {0};
 	int err;
 
-	if (clk_freq >= NUM_ICE_TIME_REF_FREQ) {
-		dev_warn(ice_hw_to_dev(hw), "Invalid TIME_REF frequency %u\n",
-			 clk_freq);
-		return -EINVAL;
-	}
-
-	if (clk_src >= NUM_ICE_CLK_SRC) {
-		dev_warn(ice_hw_to_dev(hw), "Invalid clock source %u\n",
-			 clk_src);
-		return -EINVAL;
-	}
-
-	if (clk_src == ICE_CLK_SRC_TCX0 &&
-	    clk_freq != ICE_TIME_REF_FREQ_25_000) {
-		dev_warn(ice_hw_to_dev(hw),
-			 "TCX0 only supports 25 MHz frequency\n");
-		return -EINVAL;
-	}
-
-	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD9, &dw9.val);
+	err = ice_fill_quad_msg_e82x(hw, &msg, quad, offset);
 	if (err)
 		return err;
 
-	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD24, &dw24.val);
-	if (err)
-		return err;
+	msg.opcode = ice_sbq_msg_rd;
 
-	err = ice_read_cgu_reg_e82x(hw, TSPLL_RO_BWM_LF, &bwm_lf.val);
-	if (err)
+	err = ice_sbq_rw_reg(hw, &msg);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
+			  err);
 		return err;
-
-	/* Log the current clock configuration */
-	ice_debug(hw, ICE_DBG_PTP, "Current CGU configuration -- %s, clk_src %s, clk_freq %s, PLL %s\n",
-		  dw24.field.ts_pll_enable ? "enabled" : "disabled",
-		  ice_clk_src_str(dw24.field.time_ref_sel),
-		  ice_clk_freq_str(dw9.field.time_ref_freq_sel),
-		  bwm_lf.field.plllock_true_lock_cri ? "locked" : "unlocked");
-
-	/* Disable the PLL before changing the clock source or frequency */
-	if (dw24.field.ts_pll_enable) {
-		dw24.field.ts_pll_enable = 0;
-
-		err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD24, dw24.val);
-		if (err)
-			return err;
 	}
 
-	/* Set the frequency */
-	dw9.field.time_ref_freq_sel = clk_freq;
-	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD9, dw9.val);
-	if (err)
-		return err;
-
-	/* Configure the TS PLL feedback divisor */
-	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD19, &dw19.val);
-	if (err)
-		return err;
+	*val = msg.data;
 
-	dw19.field.tspll_fbdiv_intgr = e822_cgu_params[clk_freq].feedback_div;
-	dw19.field.tspll_ndivratio = 1;
+	return 0;
+}
 
-	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD19, dw19.val);
-	if (err)
-		return err;
+/**
+ * ice_write_quad_reg_e82x - Write a PHY quad register
+ * @hw: pointer to the HW struct
+ * @quad: quad to write to
+ * @offset: quad register offset to write
+ * @val: The value to write to the register
+ *
+ * Write a quad register over the device sideband queue. Quad registers are
+ * shared between multiple PHYs.
+ */
+int
+ice_write_quad_reg_e82x(struct ice_hw *hw, u8 quad, u16 offset, u32 val)
+{
+	struct ice_sbq_msg_input msg = {0};
+	int err;
 
-	/* Configure the TS PLL post divisor */
-	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD22, &dw22.val);
+	err = ice_fill_quad_msg_e82x(hw, &msg, quad, offset);
 	if (err)
 		return err;
 
-	dw22.field.time1588clk_div = e822_cgu_params[clk_freq].post_pll_div;
-	dw22.field.time1588clk_sel_div2 = 0;
-
-	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD22, dw22.val);
-	if (err)
-		return err;
+	msg.opcode = ice_sbq_msg_wr;
+	msg.data = val;
 
-	/* Configure the TS PLL pre divisor and clock source */
-	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD24, &dw24.val);
-	if (err)
+	err = ice_sbq_rw_reg(hw, &msg);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
+			  err);
 		return err;
+	}
 
-	dw24.field.ref1588_ck_div = e822_cgu_params[clk_freq].refclk_pre_div;
-	dw24.field.tspll_fbdiv_frac = e822_cgu_params[clk_freq].frac_n_div;
-	dw24.field.time_ref_sel = clk_src;
+	return 0;
+}
 
-	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD24, dw24.val);
-	if (err)
-		return err;
+/**
+ * ice_read_phy_tstamp_e82x - Read a PHY timestamp out of the quad block
+ * @hw: pointer to the HW struct
+ * @quad: the quad to read from
+ * @idx: the timestamp index to read
+ * @tstamp: on return, the 40bit timestamp value
+ *
+ * Read a 40bit timestamp value out of the two associated registers in the
+ * quad memory block that is shared between the internal PHYs of the E822
+ * family of devices.
+ */
+static int
+ice_read_phy_tstamp_e82x(struct ice_hw *hw, u8 quad, u8 idx, u64 *tstamp)
+{
+	u16 lo_addr, hi_addr;
+	u32 lo, hi;
+	int err;
 
-	/* Finally, enable the PLL */
-	dw24.field.ts_pll_enable = 1;
+	lo_addr = (u16)TS_L(Q_REG_TX_MEMORY_BANK_START, idx);
+	hi_addr = (u16)TS_H(Q_REG_TX_MEMORY_BANK_START, idx);
 
-	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD24, dw24.val);
-	if (err)
+	err = ice_read_quad_reg_e82x(hw, quad, lo_addr, &lo);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read low PTP timestamp register, err %d\n",
+			  err);
 		return err;
+	}
 
-	/* Wait to verify if the PLL locks */
-	usleep_range(1000, 5000);
-
-	err = ice_read_cgu_reg_e82x(hw, TSPLL_RO_BWM_LF, &bwm_lf.val);
-	if (err)
+	err = ice_read_quad_reg_e82x(hw, quad, hi_addr, &hi);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read high PTP timestamp register, err %d\n",
+			  err);
 		return err;
-
-	if (!bwm_lf.field.plllock_true_lock_cri) {
-		dev_warn(ice_hw_to_dev(hw), "CGU PLL failed to lock\n");
-		return -EBUSY;
 	}
 
-	/* Log the current clock configuration */
-	ice_debug(hw, ICE_DBG_PTP, "New CGU configuration -- %s, clk_src %s, clk_freq %s, PLL %s\n",
-		  dw24.field.ts_pll_enable ? "enabled" : "disabled",
-		  ice_clk_src_str(dw24.field.time_ref_sel),
-		  ice_clk_freq_str(dw9.field.time_ref_freq_sel),
-		  bwm_lf.field.plllock_true_lock_cri ? "locked" : "unlocked");
+	/* For E822 based internal PHYs, the timestamp is reported with the
+	 * lower 8 bits in the low register, and the upper 32 bits in the high
+	 * register.
+	 */
+	*tstamp = FIELD_PREP(TS_PHY_HIGH_M, hi) | FIELD_PREP(TS_PHY_LOW_M, lo);
 
 	return 0;
 }
 
 /**
- * ice_init_cgu_e82x - Initialize CGU with settings from firmware
- * @hw: pointer to the HW structure
+ * ice_clear_phy_tstamp_e82x - Clear a timestamp from the quad block
+ * @hw: pointer to the HW struct
+ * @quad: the quad to read from
+ * @idx: the timestamp index to reset
  *
- * Initialize the Clock Generation Unit of the E822 device.
+ * Read the timestamp out of the quad to clear its timestamp status bit from
+ * the PHY quad block that is shared between the internal PHYs of the E822
+ * devices.
+ *
+ * Note that unlike E810, software cannot directly write to the quad memory
+ * bank registers. E822 relies on the ice_get_phy_tx_tstamp_ready() function
+ * to determine which timestamps are valid. Reading a timestamp auto-clears
+ * the valid bit.
+ *
+ * To directly clear the contents of the timestamp block entirely, discarding
+ * all timestamp data at once, software should instead use
+ * ice_ptp_reset_ts_memory_quad_e82x().
+ *
+ * This function should only be called on an idx whose bit is set according to
+ * ice_get_phy_tx_tstamp_ready().
  */
-static int ice_init_cgu_e82x(struct ice_hw *hw)
+static int
+ice_clear_phy_tstamp_e82x(struct ice_hw *hw, u8 quad, u8 idx)
 {
-	struct ice_ts_func_info *ts_info = &hw->func_caps.ts_func_info;
-	union tspll_cntr_bist_settings cntr_bist;
+	u64 unused_tstamp;
 	int err;
 
-	err = ice_read_cgu_reg_e82x(hw, TSPLL_CNTR_BIST_SETTINGS,
-				    &cntr_bist.val);
-	if (err)
+	err = ice_read_phy_tstamp_e82x(hw, quad, idx, &unused_tstamp);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read the timestamp register for quad %u, idx %u, err %d\n",
+			  quad, idx, err);
 		return err;
+	}
 
-	/* Disable sticky lock detection so lock err reported is accurate */
-	cntr_bist.field.i_plllock_sel_0 = 0;
-	cntr_bist.field.i_plllock_sel_1 = 0;
+	return 0;
+}
 
-	err = ice_write_cgu_reg_e82x(hw, TSPLL_CNTR_BIST_SETTINGS,
-				     cntr_bist.val);
-	if (err)
-		return err;
+/**
+ * ice_ptp_reset_ts_memory_quad_e82x - Clear all timestamps from the quad block
+ * @hw: pointer to the HW struct
+ * @quad: the quad to read from
+ *
+ * Clear all timestamps from the PHY quad block that is shared between the
+ * internal PHYs on the E822 devices.
+ */
+void ice_ptp_reset_ts_memory_quad_e82x(struct ice_hw *hw, u8 quad)
+{
+	ice_write_quad_reg_e82x(hw, quad, Q_REG_TS_CTRL, Q_REG_TS_CTRL_M);
+	ice_write_quad_reg_e82x(hw, quad, Q_REG_TS_CTRL, ~(u32)Q_REG_TS_CTRL_M);
+}
 
-	/* Configure the CGU PLL using the parameters from the function
-	 * capabilities.
-	 */
-	err = ice_cfg_cgu_pll_e82x(hw, ts_info->time_ref,
-				   (enum ice_clk_src)ts_info->clk_src);
-	if (err)
-		return err;
+/**
+ * ice_ptp_reset_ts_memory_e82x - Clear all timestamps from all quad blocks
+ * @hw: pointer to the HW struct
+ */
+static void ice_ptp_reset_ts_memory_e82x(struct ice_hw *hw)
+{
+	unsigned int quad;
 
-	return 0;
+	for (quad = 0; quad < ICE_GET_QUAD_NUM(hw->ptp.num_lports); quad++)
+		ice_ptp_reset_ts_memory_quad_e82x(hw, quad);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
index 5223e17d2806..2d8ba9c2251d 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
@@ -377,7 +377,7 @@ int ice_cgu_get_output_pin_state_caps(struct ice_hw *hw, u8 pin_id,
 #define P_REG_TIMETUS_L			0x410
 #define P_REG_TIMETUS_U			0x414
 
-#define P_REG_40B_LOW_M			0xFF
+#define P_REG_40B_LOW_M			GENMASK(7, 0)
 #define P_REG_40B_HIGH_S		8
 
 /* PHY window length registers */
-- 
cgit v1.2.3-58-ga151


From 1f374d57c39386520586539641cafc999d0f3ef5 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Tue, 28 May 2024 16:03:56 -0700
Subject: ice: Introduce ice_get_base_incval() helper

Add a new helper for getting base clock increment value for specific HW.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528-next-2024-05-28-ptp-refactors-v1-6-c082739bb6f6@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_ptp.c    |  9 +--------
 drivers/net/ethernet/intel/ice/ice_ptp_hw.h | 18 ++++++++++++++++++
 2 files changed, 19 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index 4ed2213247f7..923011c9609a 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -7,8 +7,6 @@
 
 #define E810_OUT_PROP_DELAY_NS 1
 
-#define UNKNOWN_INCVAL_E82X 0x100000000ULL
-
 static const struct ptp_pin_desc ice_pin_desc_e810t[] = {
 	/* name    idx   func         chan */
 	{ "GNSS",  GNSS, PTP_PF_EXTTS, 0, { 0, } },
@@ -1210,12 +1208,7 @@ static u64 ice_base_incval(struct ice_pf *pf)
 	struct ice_hw *hw = &pf->hw;
 	u64 incval;
 
-	if (ice_is_e810(hw))
-		incval = ICE_PTP_NOMINAL_INCVAL_E810;
-	else if (ice_e82x_time_ref(hw) < NUM_ICE_TIME_REF_FREQ)
-		incval = ice_e82x_nominal_incval(ice_e82x_time_ref(hw));
-	else
-		incval = UNKNOWN_INCVAL_E82X;
+	incval = ice_get_base_incval(hw);
 
 	dev_dbg(ice_pf_to_dev(pf), "PTP: using base increment value of 0x%016llx\n",
 		incval);
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
index 2d8ba9c2251d..3fe12a1694fc 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
@@ -285,6 +285,24 @@ int ice_get_cgu_rclk_pin_info(struct ice_hw *hw, u8 *base_idx, u8 *pin_num);
 int ice_cgu_get_output_pin_state_caps(struct ice_hw *hw, u8 pin_id,
 				      unsigned long *caps);
 
+/**
+ * ice_get_base_incval - Get base clock increment value
+ * @hw: pointer to the HW struct
+ *
+ * Return: base clock increment value for supported PHYs, 0 otherwise
+ */
+static inline u64 ice_get_base_incval(struct ice_hw *hw)
+{
+	switch (hw->ptp.phy_model) {
+	case ICE_PHY_E810:
+		return ICE_PTP_NOMINAL_INCVAL_E810;
+	case ICE_PHY_E82X:
+		return ice_e82x_nominal_incval(ice_e82x_time_ref(hw));
+	default:
+		return 0;
+	}
+}
+
 #define PFTSYN_SEM_BYTES	4
 
 #define ICE_PTP_CLOCK_INDEX_0	0x00
-- 
cgit v1.2.3-58-ga151


From 7cab44f1c35f5fb01930e592aee9cc460f17903d Mon Sep 17 00:00:00 2001
From: Sergey Temerkhanov <sergey.temerkhanov@intel.com>
Date: Tue, 28 May 2024 16:03:57 -0700
Subject: ice: Introduce ETH56G PHY model for E825C products

E825C products feature a new PHY model - ETH56G.

Introduces all necessary PHY definitions, functions etc. for ETH56G PHY,
analogous to E82X and E810 ones with addition of a few HW-specific
functionalities for ETH56G like one-step timestamping.

It ensures correct PTP initialization and operation for E825C products.

Co-developed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Co-developed-by: Michal Michalik <michal.michalik@intel.com>
Signed-off-by: Michal Michalik <michal.michalik@intel.com>
Signed-off-by: Sergey Temerkhanov <sergey.temerkhanov@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Co-developed-by: Karol Kolacinski <karol.kolacinski@intel.com>
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528-next-2024-05-28-ptp-refactors-v1-7-c082739bb6f6@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_common.c     |   10 +-
 drivers/net/ethernet/intel/ice/ice_common.h     |    1 +
 drivers/net/ethernet/intel/ice/ice_ptp.c        |  101 +-
 drivers/net/ethernet/intel/ice/ice_ptp.h        |    1 +
 drivers/net/ethernet/intel/ice/ice_ptp_consts.h |  315 ++++
 drivers/net/ethernet/intel/ice/ice_ptp_hw.c     | 2069 +++++++++++++++++++++--
 drivers/net/ethernet/intel/ice/ice_ptp_hw.h     |  240 ++-
 drivers/net/ethernet/intel/ice/ice_sbq_cmd.h    |   10 +-
 drivers/net/ethernet/intel/ice/ice_type.h       |   20 +-
 9 files changed, 2613 insertions(+), 154 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 61eb6af37f8e..c37042d9bdd9 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -3067,11 +3067,13 @@ bool ice_is_100m_speed_supported(struct ice_hw *hw)
  * Note: In the structure of [phy_type_low, phy_type_high], there should
  * be one bit set, as this function will convert one PHY type to its
  * speed.
- * If no bit gets set, ICE_AQ_LINK_SPEED_UNKNOWN will be returned
- * If more than one bit gets set, ICE_AQ_LINK_SPEED_UNKNOWN will be returned
+ *
+ * Return:
+ * * PHY speed for recognized PHY type
+ * * If no bit gets set, ICE_AQ_LINK_SPEED_UNKNOWN will be returned
+ * * If more than one bit gets set, ICE_AQ_LINK_SPEED_UNKNOWN will be returned
  */
-static u16
-ice_get_link_speed_based_on_phy_type(u64 phy_type_low, u64 phy_type_high)
+u16 ice_get_link_speed_based_on_phy_type(u64 phy_type_low, u64 phy_type_high)
 {
 	u16 speed_phy_type_high = ICE_AQ_LINK_SPEED_UNKNOWN;
 	u16 speed_phy_type_low = ICE_AQ_LINK_SPEED_UNKNOWN;
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index 70f57340eb0d..86cc1df469dd 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -262,6 +262,7 @@ int
 ice_aq_get_gpio(struct ice_hw *hw, u16 gpio_ctrl_handle, u8 pin_idx,
 		bool *value, struct ice_sq_cd *cd);
 bool ice_is_100m_speed_supported(struct ice_hw *hw);
+u16 ice_get_link_speed_based_on_phy_type(u64 phy_type_low, u64 phy_type_high);
 int
 ice_aq_set_lldp_mib(struct ice_hw *hw, u8 mib_type, void *buf, u16 buf_size,
 		    struct ice_sq_cd *cd);
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index 923011c9609a..3af0f4a2c3be 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -1011,6 +1011,28 @@ ice_ptp_release_tx_tracker(struct ice_pf *pf, struct ice_ptp_tx *tx)
 	tx->len = 0;
 }
 
+/**
+ * ice_ptp_init_tx_eth56g - Initialize tracking for Tx timestamps
+ * @pf: Board private structure
+ * @tx: the Tx tracking structure to initialize
+ * @port: the port this structure tracks
+ *
+ * Initialize the Tx timestamp tracker for this port. ETH56G PHYs
+ * have independent memory blocks for all ports.
+ *
+ * Return: 0 for success, -ENOMEM when failed to allocate Tx tracker
+ */
+static int ice_ptp_init_tx_eth56g(struct ice_pf *pf, struct ice_ptp_tx *tx,
+				  u8 port)
+{
+	tx->block = port;
+	tx->offset = 0;
+	tx->len = INDEX_PER_PORT_ETH56G;
+	tx->has_ready_bitmap = 1;
+
+	return ice_ptp_alloc_tx_tracker(tx);
+}
+
 /**
  * ice_ptp_init_tx_e82x - Initialize tracking for Tx timestamps
  * @pf: Board private structure
@@ -1341,10 +1363,19 @@ ice_ptp_port_phy_stop(struct ice_ptp_port *ptp_port)
 
 	mutex_lock(&ptp_port->ps_lock);
 
-	kthread_cancel_delayed_work_sync(&ptp_port->ov_work);
+	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		err = ice_stop_phy_timer_eth56g(hw, port, true);
+		break;
+	case ICE_PHY_E82X:
+		kthread_cancel_delayed_work_sync(&ptp_port->ov_work);
 
-	err = ice_stop_phy_timer_e82x(hw, port, true);
-	if (err)
+		err = ice_stop_phy_timer_e82x(hw, port, true);
+		break;
+	default:
+		err = -ENODEV;
+	}
+	if (err && err != -EBUSY)
 		dev_err(ice_pf_to_dev(pf), "PTP failed to set PHY port %d down, err %d\n",
 			port, err);
 
@@ -1378,27 +1409,39 @@ ice_ptp_port_phy_restart(struct ice_ptp_port *ptp_port)
 
 	mutex_lock(&ptp_port->ps_lock);
 
-	kthread_cancel_delayed_work_sync(&ptp_port->ov_work);
+	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		err = ice_start_phy_timer_eth56g(hw, port);
+		break;
+	case ICE_PHY_E82X:
+		/* Start the PHY timer in Vernier mode */
+		kthread_cancel_delayed_work_sync(&ptp_port->ov_work);
 
-	/* temporarily disable Tx timestamps while calibrating PHY offset */
-	spin_lock_irqsave(&ptp_port->tx.lock, flags);
-	ptp_port->tx.calibrating = true;
-	spin_unlock_irqrestore(&ptp_port->tx.lock, flags);
-	ptp_port->tx_fifo_busy_cnt = 0;
+		/* temporarily disable Tx timestamps while calibrating
+		 * PHY offset
+		 */
+		spin_lock_irqsave(&ptp_port->tx.lock, flags);
+		ptp_port->tx.calibrating = true;
+		spin_unlock_irqrestore(&ptp_port->tx.lock, flags);
+		ptp_port->tx_fifo_busy_cnt = 0;
 
-	/* Start the PHY timer in Vernier mode */
-	err = ice_start_phy_timer_e82x(hw, port);
-	if (err)
-		goto out_unlock;
+		/* Start the PHY timer in Vernier mode */
+		err = ice_start_phy_timer_e82x(hw, port);
+		if (err)
+			break;
 
-	/* Enable Tx timestamps right away */
-	spin_lock_irqsave(&ptp_port->tx.lock, flags);
-	ptp_port->tx.calibrating = false;
-	spin_unlock_irqrestore(&ptp_port->tx.lock, flags);
+		/* Enable Tx timestamps right away */
+		spin_lock_irqsave(&ptp_port->tx.lock, flags);
+		ptp_port->tx.calibrating = false;
+		spin_unlock_irqrestore(&ptp_port->tx.lock, flags);
 
-	kthread_queue_delayed_work(pf->ptp.kworker, &ptp_port->ov_work, 0);
+		kthread_queue_delayed_work(pf->ptp.kworker, &ptp_port->ov_work,
+					   0);
+		break;
+	default:
+		err = -ENODEV;
+	}
 
-out_unlock:
 	if (err)
 		dev_err(ice_pf_to_dev(pf), "PTP failed to set PHY port %d up, err %d\n",
 			port, err);
@@ -1436,6 +1479,7 @@ void ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup)
 	case ICE_PHY_E810:
 		/* Do not reconfigure E810 PHY */
 		return;
+	case ICE_PHY_ETH56G:
 	case ICE_PHY_E82X:
 		ice_ptp_port_phy_restart(ptp_port);
 		return;
@@ -1465,6 +1509,22 @@ static int ice_ptp_cfg_phy_interrupt(struct ice_pf *pf, bool ena, u32 threshold)
 	ice_ptp_reset_ts_memory(hw);
 
 	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G: {
+		int port;
+
+		for (port = 0; port < hw->ptp.num_lports; port++) {
+			int err;
+
+			err = ice_phy_cfg_intr_eth56g(hw, port, ena, threshold);
+			if (err) {
+				dev_err(dev, "Failed to configure PHY interrupt for port %d, err %d\n",
+					port, err);
+				return err;
+			}
+		}
+
+		return 0;
+	}
 	case ICE_PHY_E82X: {
 		int quad;
 
@@ -3075,6 +3135,9 @@ static int ice_ptp_init_port(struct ice_pf *pf, struct ice_ptp_port *ptp_port)
 	mutex_init(&ptp_port->ps_lock);
 
 	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		return ice_ptp_init_tx_eth56g(pf, &ptp_port->tx,
+					      ptp_port->port_num);
 	case ICE_PHY_E810:
 		return ice_ptp_init_tx_e810(pf, &ptp_port->tx);
 	case ICE_PHY_E82X:
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.h b/drivers/net/ethernet/intel/ice/ice_ptp.h
index 3af20025043a..e0c23aaedc12 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.h
@@ -153,6 +153,7 @@ struct ice_ptp_tx {
 #define INDEX_PER_QUAD			64
 #define INDEX_PER_PORT_E82X		16
 #define INDEX_PER_PORT_E810		64
+#define INDEX_PER_PORT_ETH56G		64
 
 /**
  * struct ice_ptp_port - data used to initialize an external port for PTP
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_consts.h b/drivers/net/ethernet/intel/ice/ice_ptp_consts.h
index 2c4dab0c48ab..ef180936f60c 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_consts.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_consts.h
@@ -9,6 +9,321 @@
  */
 /* Constants defined for the PTP 1588 clock hardware. */
 
+const struct ice_phy_reg_info_eth56g eth56g_phy_res[NUM_ETH56G_PHY_RES] = {
+	/* ETH56G_PHY_REG_PTP */
+	{
+		/* base_addr */
+		{
+			0x092000,
+			0x126000,
+			0x1BA000,
+			0x24E000,
+			0x2E2000,
+		},
+		/* step */
+		0x98,
+	},
+	/* ETH56G_PHY_MEM_PTP */
+	{
+		/* base_addr */
+		{
+			0x093000,
+			0x127000,
+			0x1BB000,
+			0x24F000,
+			0x2E3000,
+		},
+		/* step */
+		0x200,
+	},
+	/* ETH56G_PHY_REG_XPCS */
+	{
+		/* base_addr */
+		{
+			0x000000,
+			0x009400,
+			0x128000,
+			0x1BC000,
+			0x250000,
+		},
+		/* step */
+		0x21000,
+	},
+	/* ETH56G_PHY_REG_MAC */
+	{
+		/* base_addr */
+		{
+			0x085000,
+			0x119000,
+			0x1AD000,
+			0x241000,
+			0x2D5000,
+		},
+		/* step */
+		0x1000,
+	},
+	/* ETH56G_PHY_REG_GPCS */
+	{
+		/* base_addr */
+		{
+			0x084000,
+			0x118000,
+			0x1AC000,
+			0x240000,
+			0x2D4000,
+		},
+		/* step */
+		0x400,
+	},
+};
+
+const
+struct ice_eth56g_mac_reg_cfg eth56g_mac_cfg[NUM_ICE_ETH56G_LNK_SPD] = {
+	[ICE_ETH56G_LNK_SPD_1G] = {
+		.tx_mode = { .def = 6, },
+		.rx_mode = { .def = 6, },
+		.blks_per_clk = 1,
+		.blktime = 0x4000, /* 32 */
+		.tx_offset = {
+			.serdes = 0x6666, /* 51.2 */
+			.no_fec = 0xd066, /* 104.2 */
+			.sfd = 0x3000, /* 24 */
+			.onestep = 0x30000 /* 384 */
+		},
+		.rx_offset = {
+			.serdes = 0xffffc59a, /* -29.2 */
+			.no_fec = 0xffff0a80, /* -122.75 */
+			.sfd = 0x2c00, /* 22 */
+			.bs_ds = 0x19a /* 0.8 */
+			/* Dynamic bitslip 0 equals to 10 */
+		}
+	},
+	[ICE_ETH56G_LNK_SPD_2_5G] = {
+		.tx_mode = { .def = 6, },
+		.rx_mode = { .def = 6, },
+		.blks_per_clk = 1,
+		.blktime = 0x199a, /* 12.8 */
+		.tx_offset = {
+			.serdes = 0x28f6, /* 20.48 */
+			.no_fec = 0x53b8, /* 41.86 */
+			.sfd = 0x1333, /* 9.6 */
+			.onestep = 0x13333 /* 153.6 */
+		},
+		.rx_offset = {
+			.serdes = 0xffffe8a4, /* -11.68 */
+			.no_fec = 0xffff9a76, /* -50.77 */
+			.sfd = 0xf33, /* 7.6 */
+			.bs_ds = 0xa4 /* 0.32 */
+		}
+	},
+	[ICE_ETH56G_LNK_SPD_10G] = {
+		.tx_mode = { .def = 1, },
+		.rx_mode = { .def = 1, },
+		.blks_per_clk = 1,
+		.blktime = 0x666, /* 3.2 */
+		.tx_offset = {
+			.serdes = 0x234c, /* 17.6484848 */
+			.no_fec = 0x8e80, /* 71.25 */
+			.fc = 0xb4a4, /* 90.32 */
+			.sfd = 0x4a4, /* 2.32 */
+			.onestep = 0x4ccd /* 38.4 */
+		},
+		.rx_offset = {
+			.serdes = 0xffffeb27, /* -10.42424 */
+			.no_fec = 0xffffcccd, /* -25.6 */
+			.fc = 0xfffe0014, /* -255.96 */
+			.sfd = 0x4a4, /* 2.32 */
+			.bs_ds = 0x32 /* 0.0969697 */
+		}
+	},
+	[ICE_ETH56G_LNK_SPD_25G] = {
+		.tx_mode = {
+			.def = 1,
+			.rs = 4
+		},
+		.tx_mk_dly = 4,
+		.tx_cw_dly = {
+			.def = 1,
+			.onestep = 6
+		},
+		.rx_mode = {
+			.def = 1,
+			.rs = 4
+		},
+		.rx_mk_dly = {
+			.def = 1,
+			.rs = 1
+		},
+		.rx_cw_dly = {
+			.def = 1,
+			.rs = 1
+		},
+		.blks_per_clk = 1,
+		.blktime = 0x28f, /* 1.28 */
+		.mktime = 0x147b, /* 10.24, only if RS-FEC enabled */
+		.tx_offset = {
+			.serdes = 0xe1e, /* 7.0593939 */
+			.no_fec = 0x3857, /* 28.17 */
+			.fc = 0x48c3, /* 36.38 */
+			.rs = 0x8100, /* 64.5 */
+			.sfd = 0x1dc, /* 0.93 */
+			.onestep = 0x1eb8 /* 15.36 */
+		},
+		.rx_offset = {
+			.serdes = 0xfffff7a9, /* -4.1697 */
+			.no_fec = 0xffffe71a, /* -12.45 */
+			.fc = 0xfffe894d, /* -187.35 */
+			.rs = 0xfffff8cd, /* -3.6 */
+			.sfd = 0x1dc, /* 0.93 */
+			.bs_ds = 0x14 /* 0.0387879, RS-FEC 0 */
+		}
+	},
+	[ICE_ETH56G_LNK_SPD_40G] = {
+		.tx_mode = { .def = 3 },
+		.tx_mk_dly = 4,
+		.tx_cw_dly = {
+			.def = 1,
+			.onestep = 6
+		},
+		.rx_mode = { .def = 4 },
+		.rx_mk_dly = { .def = 1 },
+		.rx_cw_dly = { .def = 1 },
+		.blktime = 0x333, /* 1.6 */
+		.mktime = 0xccd, /* 6.4 */
+		.tx_offset = {
+			.serdes = 0x234c, /* 17.6484848 */
+			.no_fec = 0x5a8a, /* 45.27 */
+			.fc = 0x81b8, /* 64.86 */
+			.sfd = 0x4a4, /* 2.32 */
+			.onestep = 0x1333 /* 9.6 */
+		},
+		.rx_offset = {
+			.serdes = 0xffffeb27, /* -10.42424 */
+			.no_fec = 0xfffff594, /* -5.21 */
+			.fc = 0xfffe3080, /* -231.75 */
+			.sfd = 0x4a4, /* 2.32 */
+			.bs_ds = 0xccd /* 6.4 */
+		}
+	},
+	[ICE_ETH56G_LNK_SPD_50G] = {
+		.tx_mode = { .def = 5 },
+		.tx_mk_dly = 4,
+		.tx_cw_dly = {
+			.def = 1,
+			.onestep = 6
+		},
+		.rx_mode = { .def = 5 },
+		.rx_mk_dly = { .def = 1 },
+		.rx_cw_dly = { .def = 1 },
+		.blktime = 0x28f, /* 1.28 */
+		.mktime = 0xa3d, /* 5.12 */
+		.tx_offset = {
+			.serdes = 0x13ba, /* 9.86353 */
+			.rs = 0x5400, /* 42 */
+			.sfd = 0xe6, /* 0.45 */
+			.onestep = 0xf5c /* 7.68 */
+		},
+		.rx_offset = {
+			.serdes = 0xfffff7e8, /* -4.04706 */
+			.rs = 0xfffff994, /* -3.21 */
+			.sfd = 0xe6 /* 0.45 */
+		}
+	},
+	[ICE_ETH56G_LNK_SPD_50G2] = {
+		.tx_mode = {
+			.def = 3,
+			.rs = 2
+		},
+		.tx_mk_dly = 4,
+		.tx_cw_dly = {
+			.def = 1,
+			.onestep = 6
+		},
+		.rx_mode = {
+			.def = 4,
+			.rs = 1
+		},
+		.rx_mk_dly = { .def = 1 },
+		.rx_cw_dly = { .def = 1 },
+		.blktime = 0x28f, /* 1.28 */
+		.mktime = 0xa3d, /* 5.12 */
+		.tx_offset = {
+			.serdes = 0xe1e, /* 7.0593939 */
+			.no_fec = 0x3d33, /* 30.6 */
+			.rs = 0x5057, /* 40.17 */
+			.sfd = 0x1dc, /* 0.93 */
+			.onestep = 0xf5c /* 7.68 */
+		},
+		.rx_offset = {
+			.serdes = 0xfffff7a9, /* -4.1697 */
+			.no_fec = 0xfffff8cd, /* -3.6 */
+			.rs = 0xfffff21a, /* -6.95 */
+			.sfd = 0x1dc, /* 0.93 */
+			.bs_ds = 0xa3d /* 5.12, RS-FEC 0x633 (3.1) */
+		}
+	},
+	[ICE_ETH56G_LNK_SPD_100G] = {
+		.tx_mode = {
+			.def = 3,
+			.rs = 2
+		},
+		.tx_mk_dly = 10,
+		.tx_cw_dly = {
+			.def = 3,
+			.onestep = 6
+		},
+		.rx_mode = {
+			.def = 4,
+			.rs = 1
+		},
+		.rx_mk_dly = { .def = 5 },
+		.rx_cw_dly = { .def = 5 },
+		.blks_per_clk = 1,
+		.blktime = 0x148, /* 0.64 */
+		.mktime = 0x199a, /* 12.8 */
+		.tx_offset = {
+			.serdes = 0xe1e, /* 7.0593939 */
+			.no_fec = 0x67ec, /* 51.96 */
+			.rs = 0x44fb, /* 34.49 */
+			.sfd = 0x1dc, /* 0.93 */
+			.onestep = 0xf5c /* 7.68 */
+		},
+		.rx_offset = {
+			.serdes = 0xfffff7a9, /* -4.1697 */
+			.no_fec = 0xfffff5a9, /* -5.17 */
+			.rs = 0xfffff6e6, /* -4.55 */
+			.sfd = 0x1dc, /* 0.93 */
+			.bs_ds = 0x199a /* 12.8, RS-FEC 0x31b (1.552) */
+		}
+	},
+	[ICE_ETH56G_LNK_SPD_100G2] = {
+		.tx_mode = { .def = 5 },
+		.tx_mk_dly = 10,
+		.tx_cw_dly = {
+			.def = 3,
+			.onestep = 6
+		},
+		.rx_mode = { .def = 5 },
+		.rx_mk_dly = { .def = 5 },
+		.rx_cw_dly = { .def = 5 },
+		.blks_per_clk = 1,
+		.blktime = 0x148, /* 0.64 */
+		.mktime = 0x199a, /* 12.8 */
+		.tx_offset = {
+			.serdes = 0x13ba, /* 9.86353 */
+			.rs = 0x460a, /* 35.02 */
+			.sfd = 0xe6, /* 0.45 */
+			.onestep = 0xf5c /* 7.68 */
+		},
+		.rx_offset = {
+			.serdes = 0xfffff7e8, /* -4.04706 */
+			.rs = 0xfffff548, /* -5.36 */
+			.sfd = 0xe6, /* 0.45 */
+			.bs_ds = 0x303 /* 1.506 */
+		}
+	}
+};
+
 /* struct ice_time_ref_info_e82x
  *
  * E822 hardware can use different sources as the reference for the PTP
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index 7224bce7aa28..1b333b1c2738 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -2,6 +2,7 @@
 /* Copyright (C) 2021, Intel Corporation. */
 
 #include <linux/delay.h>
+#include <linux/iopoll.h>
 #include "ice_common.h"
 #include "ice_ptp_hw.h"
 #include "ice_ptp_consts.h"
@@ -642,6 +643,1810 @@ static void ice_ptp_exec_tmr_cmd(struct ice_hw *hw)
 	ice_flush(hw);
 }
 
+/* 56G PHY device functions
+ *
+ * The following functions operate on devices with the ETH 56G PHY.
+ */
+
+/**
+ * ice_write_phy_eth56g - Write a PHY port register
+ * @hw: pointer to the HW struct
+ * @phy_idx: PHY index
+ * @addr: PHY register address
+ * @val: Value to write
+ *
+ * Return: 0 on success, other error codes when failed to write to PHY
+ */
+static int ice_write_phy_eth56g(struct ice_hw *hw, u8 phy_idx, u32 addr,
+				u32 val)
+{
+	struct ice_sbq_msg_input phy_msg;
+	int err;
+
+	phy_msg.opcode = ice_sbq_msg_wr;
+
+	phy_msg.msg_addr_low = lower_16_bits(addr);
+	phy_msg.msg_addr_high = upper_16_bits(addr);
+
+	phy_msg.data = val;
+	phy_msg.dest_dev = hw->ptp.phy.eth56g.phy_addr[phy_idx];
+
+	err = ice_sbq_rw_reg(hw, &phy_msg);
+
+	if (err)
+		ice_debug(hw, ICE_DBG_PTP, "PTP failed to send msg to phy %d\n",
+			  err);
+
+	return err;
+}
+
+/**
+ * ice_read_phy_eth56g - Read a PHY port register
+ * @hw: pointer to the HW struct
+ * @phy_idx: PHY index
+ * @addr: PHY register address
+ * @val: Value to write
+ *
+ * Return: 0 on success, other error codes when failed to read from PHY
+ */
+static int ice_read_phy_eth56g(struct ice_hw *hw, u8 phy_idx, u32 addr,
+			       u32 *val)
+{
+	struct ice_sbq_msg_input phy_msg;
+	int err;
+
+	phy_msg.opcode = ice_sbq_msg_rd;
+
+	phy_msg.msg_addr_low = lower_16_bits(addr);
+	phy_msg.msg_addr_high = upper_16_bits(addr);
+
+	phy_msg.data = 0;
+	phy_msg.dest_dev = hw->ptp.phy.eth56g.phy_addr[phy_idx];
+
+	err = ice_sbq_rw_reg(hw, &phy_msg);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "PTP failed to send msg to phy %d\n",
+			  err);
+		return err;
+	}
+
+	*val = phy_msg.data;
+
+	return 0;
+}
+
+/**
+ * ice_phy_res_address_eth56g - Calculate a PHY port register address
+ * @port: Port number to be written
+ * @res_type: resource type (register/memory)
+ * @offset: Offset from PHY port register base
+ * @addr: The result address
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - invalid port number or resource type
+ */
+static int ice_phy_res_address_eth56g(u8 port, enum eth56g_res_type res_type,
+				      u32 offset, u32 *addr)
+{
+	u8 lane = port % ICE_PORTS_PER_QUAD;
+	u8 phy = ICE_GET_QUAD_NUM(port);
+
+	if (res_type >= NUM_ETH56G_PHY_RES)
+		return -EINVAL;
+
+	*addr = eth56g_phy_res[res_type].base[phy] +
+		lane * eth56g_phy_res[res_type].step + offset;
+	return 0;
+}
+
+/**
+ * ice_write_port_eth56g - Write a PHY port register
+ * @hw: pointer to the HW struct
+ * @offset: PHY register offset
+ * @port: Port number
+ * @val: Value to write
+ * @res_type: resource type (register/memory)
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - invalid port number or resource type
+ * * %other  - failed to write to PHY
+ */
+static int ice_write_port_eth56g(struct ice_hw *hw, u8 port, u32 offset,
+				 u32 val, enum eth56g_res_type res_type)
+{
+	u8 phy_port = port % hw->ptp.ports_per_phy;
+	u8 phy_idx = port / hw->ptp.ports_per_phy;
+	u32 addr;
+	int err;
+
+	if (port >= hw->ptp.num_lports)
+		return -EINVAL;
+
+	err = ice_phy_res_address_eth56g(phy_port, res_type, offset, &addr);
+	if (err)
+		return err;
+
+	return ice_write_phy_eth56g(hw, phy_idx, addr, val);
+}
+
+/**
+ * ice_read_port_eth56g - Read a PHY port register
+ * @hw: pointer to the HW struct
+ * @offset: PHY register offset
+ * @port: Port number
+ * @val: Value to write
+ * @res_type: resource type (register/memory)
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - invalid port number or resource type
+ * * %other  - failed to read from PHY
+ */
+static int ice_read_port_eth56g(struct ice_hw *hw, u8 port, u32 offset,
+				u32 *val, enum eth56g_res_type res_type)
+{
+	u8 phy_port = port % hw->ptp.ports_per_phy;
+	u8 phy_idx = port / hw->ptp.ports_per_phy;
+	u32 addr;
+	int err;
+
+	if (port >= hw->ptp.num_lports)
+		return -EINVAL;
+
+	err = ice_phy_res_address_eth56g(phy_port, res_type, offset, &addr);
+	if (err)
+		return err;
+
+	return ice_read_phy_eth56g(hw, phy_idx, addr, val);
+}
+
+/**
+ * ice_write_ptp_reg_eth56g - Write a PHY port register
+ * @hw: pointer to the HW struct
+ * @port: Port number to be written
+ * @offset: Offset from PHY port register base
+ * @val: Value to write
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - invalid port number or resource type
+ * * %other  - failed to write to PHY
+ */
+static int ice_write_ptp_reg_eth56g(struct ice_hw *hw, u8 port, u16 offset,
+				    u32 val)
+{
+	return ice_write_port_eth56g(hw, port, offset, val, ETH56G_PHY_REG_PTP);
+}
+
+/**
+ * ice_write_mac_reg_eth56g - Write a MAC PHY port register
+ * parameter
+ * @hw: pointer to the HW struct
+ * @port: Port number to be written
+ * @offset: Offset from PHY port register base
+ * @val: Value to write
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - invalid port number or resource type
+ * * %other  - failed to write to PHY
+ */
+static int ice_write_mac_reg_eth56g(struct ice_hw *hw, u8 port, u32 offset,
+				    u32 val)
+{
+	return ice_write_port_eth56g(hw, port, offset, val, ETH56G_PHY_REG_MAC);
+}
+
+/**
+ * ice_write_xpcs_reg_eth56g - Write a PHY port register
+ * @hw: pointer to the HW struct
+ * @port: Port number to be written
+ * @offset: Offset from PHY port register base
+ * @val: Value to write
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - invalid port number or resource type
+ * * %other  - failed to write to PHY
+ */
+static int ice_write_xpcs_reg_eth56g(struct ice_hw *hw, u8 port, u32 offset,
+				     u32 val)
+{
+	return ice_write_port_eth56g(hw, port, offset, val,
+				     ETH56G_PHY_REG_XPCS);
+}
+
+/**
+ * ice_read_ptp_reg_eth56g - Read a PHY port register
+ * @hw: pointer to the HW struct
+ * @port: Port number to be read
+ * @offset: Offset from PHY port register base
+ * @val: Pointer to the value to read (out param)
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - invalid port number or resource type
+ * * %other  - failed to read from PHY
+ */
+static int ice_read_ptp_reg_eth56g(struct ice_hw *hw, u8 port, u16 offset,
+				   u32 *val)
+{
+	return ice_read_port_eth56g(hw, port, offset, val, ETH56G_PHY_REG_PTP);
+}
+
+/**
+ * ice_read_mac_reg_eth56g - Read a PHY port register
+ * @hw: pointer to the HW struct
+ * @port: Port number to be read
+ * @offset: Offset from PHY port register base
+ * @val: Pointer to the value to read (out param)
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - invalid port number or resource type
+ * * %other  - failed to read from PHY
+ */
+static int ice_read_mac_reg_eth56g(struct ice_hw *hw, u8 port, u16 offset,
+				   u32 *val)
+{
+	return ice_read_port_eth56g(hw, port, offset, val, ETH56G_PHY_REG_MAC);
+}
+
+/**
+ * ice_read_gpcs_reg_eth56g - Read a PHY port register
+ * @hw: pointer to the HW struct
+ * @port: Port number to be read
+ * @offset: Offset from PHY port register base
+ * @val: Pointer to the value to read (out param)
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - invalid port number or resource type
+ * * %other  - failed to read from PHY
+ */
+static int ice_read_gpcs_reg_eth56g(struct ice_hw *hw, u8 port, u16 offset,
+				    u32 *val)
+{
+	return ice_read_port_eth56g(hw, port, offset, val, ETH56G_PHY_REG_GPCS);
+}
+
+/**
+ * ice_read_port_mem_eth56g - Read a PHY port memory location
+ * @hw: pointer to the HW struct
+ * @port: Port number to be read
+ * @offset: Offset from PHY port register base
+ * @val: Pointer to the value to read (out param)
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - invalid port number or resource type
+ * * %other  - failed to read from PHY
+ */
+static int ice_read_port_mem_eth56g(struct ice_hw *hw, u8 port, u16 offset,
+				    u32 *val)
+{
+	return ice_read_port_eth56g(hw, port, offset, val, ETH56G_PHY_MEM_PTP);
+}
+
+/**
+ * ice_write_port_mem_eth56g - Write a PHY port memory location
+ * @hw: pointer to the HW struct
+ * @port: Port number to be read
+ * @offset: Offset from PHY port register base
+ * @val: Pointer to the value to read (out param)
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - invalid port number or resource type
+ * * %other  - failed to write to PHY
+ */
+static int ice_write_port_mem_eth56g(struct ice_hw *hw, u8 port, u16 offset,
+				     u32 val)
+{
+	return ice_write_port_eth56g(hw, port, offset, val, ETH56G_PHY_MEM_PTP);
+}
+
+/**
+ * ice_is_64b_phy_reg_eth56g - Check if this is a 64bit PHY register
+ * @low_addr: the low address to check
+ * @high_addr: on return, contains the high address of the 64bit register
+ *
+ * Write the appropriate high register offset to use.
+ *
+ * Return: true if the provided low address is one of the known 64bit PHY values
+ * represented as two 32bit registers, false otherwise.
+ */
+static bool ice_is_64b_phy_reg_eth56g(u16 low_addr, u16 *high_addr)
+{
+	switch (low_addr) {
+	case PHY_REG_TX_TIMER_INC_PRE_L:
+		*high_addr = PHY_REG_TX_TIMER_INC_PRE_U;
+		return true;
+	case PHY_REG_RX_TIMER_INC_PRE_L:
+		*high_addr = PHY_REG_RX_TIMER_INC_PRE_U;
+		return true;
+	case PHY_REG_TX_CAPTURE_L:
+		*high_addr = PHY_REG_TX_CAPTURE_U;
+		return true;
+	case PHY_REG_RX_CAPTURE_L:
+		*high_addr = PHY_REG_RX_CAPTURE_U;
+		return true;
+	case PHY_REG_TOTAL_TX_OFFSET_L:
+		*high_addr = PHY_REG_TOTAL_TX_OFFSET_U;
+		return true;
+	case PHY_REG_TOTAL_RX_OFFSET_L:
+		*high_addr = PHY_REG_TOTAL_RX_OFFSET_U;
+		return true;
+	case PHY_REG_TX_MEMORY_STATUS_L:
+		*high_addr = PHY_REG_TX_MEMORY_STATUS_U;
+		return true;
+	default:
+		return false;
+	}
+}
+
+/**
+ * ice_is_40b_phy_reg_eth56g - Check if this is a 40bit PHY register
+ * @low_addr: the low address to check
+ * @high_addr: on return, contains the high address of the 40bit value
+ *
+ * Write the appropriate high register offset to use.
+ *
+ * Return: true if the provided low address is one of the known 40bit PHY
+ * values split into two registers with the lower 8 bits in the low register and
+ * the upper 32 bits in the high register, false otherwise.
+ */
+static bool ice_is_40b_phy_reg_eth56g(u16 low_addr, u16 *high_addr)
+{
+	switch (low_addr) {
+	case PHY_REG_TIMETUS_L:
+		*high_addr = PHY_REG_TIMETUS_U;
+		return true;
+	case PHY_PCS_REF_TUS_L:
+		*high_addr = PHY_PCS_REF_TUS_U;
+		return true;
+	case PHY_PCS_REF_INC_L:
+		*high_addr = PHY_PCS_REF_INC_U;
+		return true;
+	default:
+		return false;
+	}
+}
+
+/**
+ * ice_read_64b_phy_reg_eth56g - Read a 64bit value from PHY registers
+ * @hw: pointer to the HW struct
+ * @port: PHY port to read from
+ * @low_addr: offset of the lower register to read from
+ * @val: on return, the contents of the 64bit value from the PHY registers
+ * @res_type: resource type
+ *
+ * Check if the caller has specified a known 40 bit register offset and read
+ * the two registers associated with a 40bit value and return it in the val
+ * pointer.
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - not a 64 bit register
+ * * %other  - failed to read from PHY
+ */
+static int ice_read_64b_phy_reg_eth56g(struct ice_hw *hw, u8 port, u16 low_addr,
+				       u64 *val, enum eth56g_res_type res_type)
+{
+	u16 high_addr;
+	u32 lo, hi;
+	int err;
+
+	if (!ice_is_64b_phy_reg_eth56g(low_addr, &high_addr))
+		return -EINVAL;
+
+	err = ice_read_port_eth56g(hw, port, low_addr, &lo, res_type);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read from low register %#08x\n, err %d",
+			  low_addr, err);
+		return err;
+	}
+
+	err = ice_read_port_eth56g(hw, port, high_addr, &hi, res_type);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read from high register %#08x\n, err %d",
+			  high_addr, err);
+		return err;
+	}
+
+	*val = ((u64)hi << 32) | lo;
+
+	return 0;
+}
+
+/**
+ * ice_read_64b_ptp_reg_eth56g - Read a 64bit value from PHY registers
+ * @hw: pointer to the HW struct
+ * @port: PHY port to read from
+ * @low_addr: offset of the lower register to read from
+ * @val: on return, the contents of the 64bit value from the PHY registers
+ *
+ * Check if the caller has specified a known 40 bit register offset and read
+ * the two registers associated with a 40bit value and return it in the val
+ * pointer.
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - not a 64 bit register
+ * * %other  - failed to read from PHY
+ */
+static int ice_read_64b_ptp_reg_eth56g(struct ice_hw *hw, u8 port, u16 low_addr,
+				       u64 *val)
+{
+	return ice_read_64b_phy_reg_eth56g(hw, port, low_addr, val,
+					   ETH56G_PHY_REG_PTP);
+}
+
+/**
+ * ice_write_40b_phy_reg_eth56g - Write a 40b value to the PHY
+ * @hw: pointer to the HW struct
+ * @port: port to write to
+ * @low_addr: offset of the low register
+ * @val: 40b value to write
+ * @res_type: resource type
+ *
+ * Check if the caller has specified a known 40 bit register offset and write
+ * provided 40b value to the two associated registers by splitting it up into
+ * two chunks, the lower 8 bits and the upper 32 bits.
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - not a 40 bit register
+ * * %other  - failed to write to PHY
+ */
+static int ice_write_40b_phy_reg_eth56g(struct ice_hw *hw, u8 port,
+					u16 low_addr, u64 val,
+					enum eth56g_res_type res_type)
+{
+	u16 high_addr;
+	u32 lo, hi;
+	int err;
+
+	if (!ice_is_40b_phy_reg_eth56g(low_addr, &high_addr))
+		return -EINVAL;
+
+	lo = FIELD_GET(P_REG_40B_LOW_M, val);
+	hi = (u32)(val >> P_REG_40B_HIGH_S);
+
+	err = ice_write_port_eth56g(hw, port, low_addr, lo, res_type);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write to low register 0x%08x\n, err %d",
+			  low_addr, err);
+		return err;
+	}
+
+	err = ice_write_port_eth56g(hw, port, high_addr, hi, res_type);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write to high register 0x%08x\n, err %d",
+			  high_addr, err);
+		return err;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_write_40b_ptp_reg_eth56g - Write a 40b value to the PHY
+ * @hw: pointer to the HW struct
+ * @port: port to write to
+ * @low_addr: offset of the low register
+ * @val: 40b value to write
+ *
+ * Check if the caller has specified a known 40 bit register offset and write
+ * provided 40b value to the two associated registers by splitting it up into
+ * two chunks, the lower 8 bits and the upper 32 bits.
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - not a 40 bit register
+ * * %other  - failed to write to PHY
+ */
+static int ice_write_40b_ptp_reg_eth56g(struct ice_hw *hw, u8 port,
+					u16 low_addr, u64 val)
+{
+	return ice_write_40b_phy_reg_eth56g(hw, port, low_addr, val,
+					    ETH56G_PHY_REG_PTP);
+}
+
+/**
+ * ice_write_64b_phy_reg_eth56g - Write a 64bit value to PHY registers
+ * @hw: pointer to the HW struct
+ * @port: PHY port to read from
+ * @low_addr: offset of the lower register to read from
+ * @val: the contents of the 64bit value to write to PHY
+ * @res_type: resource type
+ *
+ * Check if the caller has specified a known 64 bit register offset and write
+ * the 64bit value to the two associated 32bit PHY registers.
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - not a 64 bit register
+ * * %other  - failed to write to PHY
+ */
+static int ice_write_64b_phy_reg_eth56g(struct ice_hw *hw, u8 port,
+					u16 low_addr, u64 val,
+					enum eth56g_res_type res_type)
+{
+	u16 high_addr;
+	u32 lo, hi;
+	int err;
+
+	if (!ice_is_64b_phy_reg_eth56g(low_addr, &high_addr))
+		return -EINVAL;
+
+	lo = lower_32_bits(val);
+	hi = upper_32_bits(val);
+
+	err = ice_write_port_eth56g(hw, port, low_addr, lo, res_type);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write to low register 0x%08x\n, err %d",
+			  low_addr, err);
+		return err;
+	}
+
+	err = ice_write_port_eth56g(hw, port, high_addr, hi, res_type);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write to high register 0x%08x\n, err %d",
+			  high_addr, err);
+		return err;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_write_64b_ptp_reg_eth56g - Write a 64bit value to PHY registers
+ * @hw: pointer to the HW struct
+ * @port: PHY port to read from
+ * @low_addr: offset of the lower register to read from
+ * @val: the contents of the 64bit value to write to PHY
+ *
+ * Check if the caller has specified a known 64 bit register offset and write
+ * the 64bit value to the two associated 32bit PHY registers.
+ *
+ * Return:
+ * * %0      - success
+ * * %EINVAL - not a 64 bit register
+ * * %other  - failed to write to PHY
+ */
+static int ice_write_64b_ptp_reg_eth56g(struct ice_hw *hw, u8 port,
+					u16 low_addr, u64 val)
+{
+	return ice_write_64b_phy_reg_eth56g(hw, port, low_addr, val,
+					    ETH56G_PHY_REG_PTP);
+}
+
+/**
+ * ice_read_ptp_tstamp_eth56g - Read a PHY timestamp out of the port memory
+ * @hw: pointer to the HW struct
+ * @port: the port to read from
+ * @idx: the timestamp index to read
+ * @tstamp: on return, the 40bit timestamp value
+ *
+ * Read a 40bit timestamp value out of the two associated entries in the
+ * port memory block of the internal PHYs of the 56G devices.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to read from PHY
+ */
+static int ice_read_ptp_tstamp_eth56g(struct ice_hw *hw, u8 port, u8 idx,
+				      u64 *tstamp)
+{
+	u16 lo_addr, hi_addr;
+	u32 lo, hi;
+	int err;
+
+	lo_addr = (u16)PHY_TSTAMP_L(idx);
+	hi_addr = (u16)PHY_TSTAMP_U(idx);
+
+	err = ice_read_port_mem_eth56g(hw, port, lo_addr, &lo);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read low PTP timestamp register, err %d\n",
+			  err);
+		return err;
+	}
+
+	err = ice_read_port_mem_eth56g(hw, port, hi_addr, &hi);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read high PTP timestamp register, err %d\n",
+			  err);
+		return err;
+	}
+
+	/* For 56G based internal PHYs, the timestamp is reported with the
+	 * lower 8 bits in the low register, and the upper 32 bits in the high
+	 * register.
+	 */
+	*tstamp = ((u64)hi) << TS_PHY_HIGH_S | ((u64)lo & TS_PHY_LOW_M);
+
+	return 0;
+}
+
+/**
+ * ice_clear_ptp_tstamp_eth56g - Clear a timestamp from the quad block
+ * @hw: pointer to the HW struct
+ * @port: the quad to read from
+ * @idx: the timestamp index to reset
+ *
+ * Read and then forcibly clear the timestamp index to ensure the valid bit is
+ * cleared and the timestamp status bit is reset in the PHY port memory of
+ * internal PHYs of the 56G devices.
+ *
+ * To directly clear the contents of the timestamp block entirely, discarding
+ * all timestamp data at once, software should instead use
+ * ice_ptp_reset_ts_memory_quad_eth56g().
+ *
+ * This function should only be called on an idx whose bit is set according to
+ * ice_get_phy_tx_tstamp_ready().
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write to PHY
+ */
+static int ice_clear_ptp_tstamp_eth56g(struct ice_hw *hw, u8 port, u8 idx)
+{
+	u64 unused_tstamp;
+	u16 lo_addr;
+	int err;
+
+	/* Read the timestamp register to ensure the timestamp status bit is
+	 * cleared.
+	 */
+	err = ice_read_ptp_tstamp_eth56g(hw, port, idx, &unused_tstamp);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read the PHY timestamp register for port %u, idx %u, err %d\n",
+			  port, idx, err);
+	}
+
+	lo_addr = (u16)PHY_TSTAMP_L(idx);
+
+	err = ice_write_port_mem_eth56g(hw, port, lo_addr, 0);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to clear low PTP timestamp register for port %u, idx %u, err %d\n",
+			  port, idx, err);
+		return err;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_ptp_reset_ts_memory_eth56g - Clear all timestamps from the port block
+ * @hw: pointer to the HW struct
+ */
+static void ice_ptp_reset_ts_memory_eth56g(struct ice_hw *hw)
+{
+	unsigned int port;
+
+	for (port = 0; port < hw->ptp.num_lports; port++) {
+		ice_write_ptp_reg_eth56g(hw, port, PHY_REG_TX_MEMORY_STATUS_L,
+					 0);
+		ice_write_ptp_reg_eth56g(hw, port, PHY_REG_TX_MEMORY_STATUS_U,
+					 0);
+	}
+}
+
+/**
+ * ice_ptp_prep_port_time_eth56g - Prepare one PHY port with initial time
+ * @hw: pointer to the HW struct
+ * @port: port number
+ * @time: time to initialize the PHY port clocks to
+ *
+ * Write a new initial time value into registers of a specific PHY port.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write to PHY
+ */
+static int ice_ptp_prep_port_time_eth56g(struct ice_hw *hw, u8 port,
+					 u64 time)
+{
+	int err;
+
+	/* Tx case */
+	err = ice_write_64b_ptp_reg_eth56g(hw, port, PHY_REG_TX_TIMER_INC_PRE_L,
+					   time);
+	if (err)
+		return err;
+
+	/* Rx case */
+	return ice_write_64b_ptp_reg_eth56g(hw, port,
+					    PHY_REG_RX_TIMER_INC_PRE_L, time);
+}
+
+/**
+ * ice_ptp_prep_phy_time_eth56g - Prepare PHY port with initial time
+ * @hw: pointer to the HW struct
+ * @time: Time to initialize the PHY port clocks to
+ *
+ * Program the PHY port registers with a new initial time value. The port
+ * clock will be initialized once the driver issues an ICE_PTP_INIT_TIME sync
+ * command. The time value is the upper 32 bits of the PHY timer, usually in
+ * units of nominal nanoseconds.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write to PHY
+ */
+static int ice_ptp_prep_phy_time_eth56g(struct ice_hw *hw, u32 time)
+{
+	u64 phy_time;
+	u8 port;
+
+	/* The time represents the upper 32 bits of the PHY timer, so we need
+	 * to shift to account for this when programming.
+	 */
+	phy_time = (u64)time << 32;
+
+	for (port = 0; port < hw->ptp.num_lports; port++) {
+		int err;
+
+		err = ice_ptp_prep_port_time_eth56g(hw, port, phy_time);
+		if (err) {
+			ice_debug(hw, ICE_DBG_PTP, "Failed to write init time for port %u, err %d\n",
+				  port, err);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * ice_ptp_prep_port_adj_eth56g - Prepare a single port for time adjust
+ * @hw: pointer to HW struct
+ * @port: Port number to be programmed
+ * @time: time in cycles to adjust the port clocks
+ *
+ * Program the port for an atomic adjustment by writing the Tx and Rx timer
+ * registers. The atomic adjustment won't be completed until the driver issues
+ * an ICE_PTP_ADJ_TIME command.
+ *
+ * Note that time is not in units of nanoseconds. It is in clock time
+ * including the lower sub-nanosecond portion of the port timer.
+ *
+ * Negative adjustments are supported using 2s complement arithmetic.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write to PHY
+ */
+static int ice_ptp_prep_port_adj_eth56g(struct ice_hw *hw, u8 port, s64 time)
+{
+	u32 l_time, u_time;
+	int err;
+
+	l_time = lower_32_bits(time);
+	u_time = upper_32_bits(time);
+
+	/* Tx case */
+	err = ice_write_ptp_reg_eth56g(hw, port, PHY_REG_TX_TIMER_INC_PRE_L,
+				       l_time);
+	if (err)
+		goto exit_err;
+
+	err = ice_write_ptp_reg_eth56g(hw, port, PHY_REG_TX_TIMER_INC_PRE_U,
+				       u_time);
+	if (err)
+		goto exit_err;
+
+	/* Rx case */
+	err = ice_write_ptp_reg_eth56g(hw, port, PHY_REG_RX_TIMER_INC_PRE_L,
+				       l_time);
+	if (err)
+		goto exit_err;
+
+	err = ice_write_ptp_reg_eth56g(hw, port, PHY_REG_RX_TIMER_INC_PRE_U,
+				       u_time);
+	if (err)
+		goto exit_err;
+
+	return 0;
+
+exit_err:
+	ice_debug(hw, ICE_DBG_PTP, "Failed to write time adjust for port %u, err %d\n",
+		  port, err);
+	return err;
+}
+
+/**
+ * ice_ptp_prep_phy_adj_eth56g - Prep PHY ports for a time adjustment
+ * @hw: pointer to HW struct
+ * @adj: adjustment in nanoseconds
+ *
+ * Prepare the PHY ports for an atomic time adjustment by programming the PHY
+ * Tx and Rx port registers. The actual adjustment is completed by issuing an
+ * ICE_PTP_ADJ_TIME or ICE_PTP_ADJ_TIME_AT_TIME sync command.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write to PHY
+ */
+static int ice_ptp_prep_phy_adj_eth56g(struct ice_hw *hw, s32 adj)
+{
+	s64 cycles;
+	u8 port;
+
+	/* The port clock supports adjustment of the sub-nanosecond portion of
+	 * the clock (lowest 32 bits). We shift the provided adjustment in
+	 * nanoseconds by 32 to calculate the appropriate adjustment to program
+	 * into the PHY ports.
+	 */
+	cycles = (s64)adj << 32;
+
+	for (port = 0; port < hw->ptp.num_lports; port++) {
+		int err;
+
+		err = ice_ptp_prep_port_adj_eth56g(hw, port, cycles);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_ptp_prep_phy_incval_eth56g - Prepare PHY ports for time adjustment
+ * @hw: pointer to HW struct
+ * @incval: new increment value to prepare
+ *
+ * Prepare each of the PHY ports for a new increment value by programming the
+ * port's TIMETUS registers. The new increment value will be updated after
+ * issuing an ICE_PTP_INIT_INCVAL command.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write to PHY
+ */
+static int ice_ptp_prep_phy_incval_eth56g(struct ice_hw *hw, u64 incval)
+{
+	u8 port;
+
+	for (port = 0; port < hw->ptp.num_lports; port++) {
+		int err;
+
+		err = ice_write_40b_ptp_reg_eth56g(hw, port, PHY_REG_TIMETUS_L,
+						   incval);
+		if (err) {
+			ice_debug(hw, ICE_DBG_PTP, "Failed to write incval for port %u, err %d\n",
+				  port, err);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * ice_ptp_read_port_capture_eth56g - Read a port's local time capture
+ * @hw: pointer to HW struct
+ * @port: Port number to read
+ * @tx_ts: on return, the Tx port time capture
+ * @rx_ts: on return, the Rx port time capture
+ *
+ * Read the port's Tx and Rx local time capture values.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to read from PHY
+ */
+static int ice_ptp_read_port_capture_eth56g(struct ice_hw *hw, u8 port,
+					    u64 *tx_ts, u64 *rx_ts)
+{
+	int err;
+
+	/* Tx case */
+	err = ice_read_64b_ptp_reg_eth56g(hw, port, PHY_REG_TX_CAPTURE_L,
+					  tx_ts);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read REG_TX_CAPTURE, err %d\n",
+			  err);
+		return err;
+	}
+
+	ice_debug(hw, ICE_DBG_PTP, "tx_init = %#016llx\n", *tx_ts);
+
+	/* Rx case */
+	err = ice_read_64b_ptp_reg_eth56g(hw, port, PHY_REG_RX_CAPTURE_L,
+					  rx_ts);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read RX_CAPTURE, err %d\n",
+			  err);
+		return err;
+	}
+
+	ice_debug(hw, ICE_DBG_PTP, "rx_init = %#016llx\n", *rx_ts);
+
+	return 0;
+}
+
+/**
+ * ice_ptp_write_port_cmd_eth56g - Prepare a single PHY port for a timer command
+ * @hw: pointer to HW struct
+ * @port: Port to which cmd has to be sent
+ * @cmd: Command to be sent to the port
+ *
+ * Prepare the requested port for an upcoming timer sync command.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write to PHY
+ */
+static int ice_ptp_write_port_cmd_eth56g(struct ice_hw *hw, u8 port,
+					 enum ice_ptp_tmr_cmd cmd)
+{
+	u32 val = ice_ptp_tmr_cmd_to_port_reg(hw, cmd);
+	int err;
+
+	/* Tx case */
+	err = ice_write_ptp_reg_eth56g(hw, port, PHY_REG_TX_TMR_CMD, val);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write back TX_TMR_CMD, err %d\n",
+			  err);
+		return err;
+	}
+
+	/* Rx case */
+	err = ice_write_ptp_reg_eth56g(hw, port, PHY_REG_RX_TMR_CMD, val);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write back RX_TMR_CMD, err %d\n",
+			  err);
+		return err;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_phy_get_speed_eth56g - Get link speed based on PHY link type
+ * @li: pointer to link information struct
+ *
+ * Return: simplified ETH56G PHY speed
+ */
+static enum ice_eth56g_link_spd
+ice_phy_get_speed_eth56g(struct ice_link_status *li)
+{
+	u16 speed = ice_get_link_speed_based_on_phy_type(li->phy_type_low,
+							 li->phy_type_high);
+
+	switch (speed) {
+	case ICE_AQ_LINK_SPEED_1000MB:
+		return ICE_ETH56G_LNK_SPD_1G;
+	case ICE_AQ_LINK_SPEED_2500MB:
+		return ICE_ETH56G_LNK_SPD_2_5G;
+	case ICE_AQ_LINK_SPEED_10GB:
+		return ICE_ETH56G_LNK_SPD_10G;
+	case ICE_AQ_LINK_SPEED_25GB:
+		return ICE_ETH56G_LNK_SPD_25G;
+	case ICE_AQ_LINK_SPEED_40GB:
+		return ICE_ETH56G_LNK_SPD_40G;
+	case ICE_AQ_LINK_SPEED_50GB:
+		switch (li->phy_type_low) {
+		case ICE_PHY_TYPE_LOW_50GBASE_SR:
+		case ICE_PHY_TYPE_LOW_50GBASE_FR:
+		case ICE_PHY_TYPE_LOW_50GBASE_LR:
+		case ICE_PHY_TYPE_LOW_50GBASE_KR_PAM4:
+		case ICE_PHY_TYPE_LOW_50G_AUI1_AOC_ACC:
+		case ICE_PHY_TYPE_LOW_50G_AUI1:
+			return ICE_ETH56G_LNK_SPD_50G;
+		default:
+			return ICE_ETH56G_LNK_SPD_50G2;
+		}
+	case ICE_AQ_LINK_SPEED_100GB:
+		if (li->phy_type_high ||
+		    li->phy_type_low == ICE_PHY_TYPE_LOW_100GBASE_SR2)
+			return ICE_ETH56G_LNK_SPD_100G2;
+		else
+			return ICE_ETH56G_LNK_SPD_100G;
+	default:
+		return ICE_ETH56G_LNK_SPD_1G;
+	}
+}
+
+/**
+ * ice_phy_cfg_parpcs_eth56g - Configure TUs per PAR/PCS clock cycle
+ * @hw: pointer to the HW struct
+ * @port: port to configure
+ *
+ * Configure the number of TUs for the PAR and PCS clocks used as part of the
+ * timestamp calibration process.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - PHY read/write failed
+ */
+static int ice_phy_cfg_parpcs_eth56g(struct ice_hw *hw, u8 port)
+{
+	u8 port_blk = port & ~(ICE_PORTS_PER_QUAD - 1);
+	u32 val;
+	int err;
+
+	err = ice_write_xpcs_reg_eth56g(hw, port, PHY_VENDOR_TXLANE_THRESH,
+					ICE_ETH56G_NOMINAL_THRESH4);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read VENDOR_TXLANE_THRESH, status: %d",
+			  err);
+		return err;
+	}
+
+	switch (ice_phy_get_speed_eth56g(&hw->port_info->phy.link_info)) {
+	case ICE_ETH56G_LNK_SPD_1G:
+	case ICE_ETH56G_LNK_SPD_2_5G:
+		err = ice_read_ptp_reg_eth56g(hw, port_blk,
+					      PHY_GPCS_CONFIG_REG0, &val);
+		if (err) {
+			ice_debug(hw, ICE_DBG_PTP, "Failed to read PHY_GPCS_CONFIG_REG0, status: %d",
+				  err);
+			return err;
+		}
+
+		val &= ~PHY_GPCS_CONFIG_REG0_TX_THR_M;
+		val |= FIELD_PREP(PHY_GPCS_CONFIG_REG0_TX_THR_M,
+				  ICE_ETH56G_NOMINAL_TX_THRESH);
+
+		err = ice_write_ptp_reg_eth56g(hw, port_blk,
+					       PHY_GPCS_CONFIG_REG0, val);
+		if (err) {
+			ice_debug(hw, ICE_DBG_PTP, "Failed to write PHY_GPCS_CONFIG_REG0, status: %d",
+				  err);
+			return err;
+		}
+		break;
+	default:
+		break;
+	}
+
+	err = ice_write_40b_ptp_reg_eth56g(hw, port, PHY_PCS_REF_TUS_L,
+					   ICE_ETH56G_NOMINAL_PCS_REF_TUS);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write PHY_PCS_REF_TUS, status: %d",
+			  err);
+		return err;
+	}
+
+	err = ice_write_40b_ptp_reg_eth56g(hw, port, PHY_PCS_REF_INC_L,
+					   ICE_ETH56G_NOMINAL_PCS_REF_INC);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to write PHY_PCS_REF_INC, status: %d",
+			  err);
+		return err;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_phy_cfg_ptp_1step_eth56g - Configure 1-step PTP settings
+ * @hw: Pointer to the HW struct
+ * @port: Port to configure
+ *
+ * Return:
+ * * %0     - success
+ * * %other - PHY read/write failed
+ */
+int ice_phy_cfg_ptp_1step_eth56g(struct ice_hw *hw, u8 port)
+{
+	u8 port_blk = port & ~(ICE_PORTS_PER_QUAD - 1);
+	u8 blk_port = port & (ICE_PORTS_PER_QUAD - 1);
+	bool enable, sfd_ena;
+	u32 val, peer_delay;
+	int err;
+
+	enable = hw->ptp.phy.eth56g.onestep_ena;
+	peer_delay = hw->ptp.phy.eth56g.peer_delay;
+	sfd_ena = hw->ptp.phy.eth56g.sfd_ena;
+
+	/* PHY_PTP_1STEP_CONFIG */
+	err = ice_read_ptp_reg_eth56g(hw, port_blk, PHY_PTP_1STEP_CONFIG, &val);
+	if (err)
+		return err;
+
+	if (enable)
+		val |= blk_port;
+	else
+		val &= ~blk_port;
+
+	val &= ~(PHY_PTP_1STEP_T1S_UP64_M | PHY_PTP_1STEP_T1S_DELTA_M);
+
+	err = ice_write_ptp_reg_eth56g(hw, port_blk, PHY_PTP_1STEP_CONFIG, val);
+	if (err)
+		return err;
+
+	/* PHY_PTP_1STEP_PEER_DELAY */
+	val = FIELD_PREP(PHY_PTP_1STEP_PD_DELAY_M, peer_delay);
+	if (peer_delay)
+		val |= PHY_PTP_1STEP_PD_ADD_PD_M;
+	val |= PHY_PTP_1STEP_PD_DLY_V_M;
+	err = ice_write_ptp_reg_eth56g(hw, port_blk,
+				       PHY_PTP_1STEP_PEER_DELAY(blk_port), val);
+	if (err)
+		return err;
+
+	val &= ~PHY_PTP_1STEP_PD_DLY_V_M;
+	err = ice_write_ptp_reg_eth56g(hw, port_blk,
+				       PHY_PTP_1STEP_PEER_DELAY(blk_port), val);
+	if (err)
+		return err;
+
+	/* PHY_MAC_XIF_MODE */
+	err = ice_read_mac_reg_eth56g(hw, port, PHY_MAC_XIF_MODE, &val);
+	if (err)
+		return err;
+
+	val &= ~(PHY_MAC_XIF_1STEP_ENA_M | PHY_MAC_XIF_TS_BIN_MODE_M |
+		 PHY_MAC_XIF_TS_SFD_ENA_M | PHY_MAC_XIF_GMII_TS_SEL_M);
+
+	switch (ice_phy_get_speed_eth56g(&hw->port_info->phy.link_info)) {
+	case ICE_ETH56G_LNK_SPD_1G:
+	case ICE_ETH56G_LNK_SPD_2_5G:
+		val |= PHY_MAC_XIF_GMII_TS_SEL_M;
+		break;
+	default:
+		break;
+	}
+
+	val |= FIELD_PREP(PHY_MAC_XIF_1STEP_ENA_M, enable) |
+	       FIELD_PREP(PHY_MAC_XIF_TS_BIN_MODE_M, enable) |
+	       FIELD_PREP(PHY_MAC_XIF_TS_SFD_ENA_M, sfd_ena);
+
+	return ice_write_mac_reg_eth56g(hw, port, PHY_MAC_XIF_MODE, val);
+}
+
+/**
+ * mul_u32_u32_fx_q9 - Multiply two u32 fixed point Q9 values
+ * @a: multiplier value
+ * @b: multiplicand value
+ *
+ * Return: result of multiplication
+ */
+static u32 mul_u32_u32_fx_q9(u32 a, u32 b)
+{
+	return (u32)(((u64)a * b) >> ICE_ETH56G_MAC_CFG_FRAC_W);
+}
+
+/**
+ * add_u32_u32_fx - Add two u32 fixed point values and discard overflow
+ * @a: first value
+ * @b: second value
+ *
+ * Return: result of addition
+ */
+static u32 add_u32_u32_fx(u32 a, u32 b)
+{
+	return lower_32_bits(((u64)a + b));
+}
+
+/**
+ * ice_ptp_calc_bitslip_eth56g - Calculate bitslip value
+ * @hw: pointer to the HW struct
+ * @port: port to configure
+ * @bs: bitslip multiplier
+ * @fc: FC-FEC enabled
+ * @rs: RS-FEC enabled
+ * @spd: link speed
+ *
+ * Return: calculated bitslip value
+ */
+static u32 ice_ptp_calc_bitslip_eth56g(struct ice_hw *hw, u8 port, u32 bs,
+				       bool fc, bool rs,
+				       enum ice_eth56g_link_spd spd)
+{
+	u8 port_offset = port & (ICE_PORTS_PER_QUAD - 1);
+	u8 port_blk = port & ~(ICE_PORTS_PER_QUAD - 1);
+	u32 bitslip;
+	int err;
+
+	if (!bs || rs)
+		return 0;
+
+	if (spd == ICE_ETH56G_LNK_SPD_1G || spd == ICE_ETH56G_LNK_SPD_2_5G)
+		err = ice_read_gpcs_reg_eth56g(hw, port, PHY_GPCS_BITSLIP,
+					       &bitslip);
+	else
+		err = ice_read_ptp_reg_eth56g(hw, port_blk,
+					      PHY_REG_SD_BIT_SLIP(port_offset),
+					      &bitslip);
+	if (err)
+		return 0;
+
+	if (spd == ICE_ETH56G_LNK_SPD_1G && !bitslip) {
+		/* Bitslip register value of 0 corresponds to 10 so substitute
+		 * it for calculations
+		 */
+		bitslip = 10;
+	} else if (spd == ICE_ETH56G_LNK_SPD_10G ||
+		   spd == ICE_ETH56G_LNK_SPD_25G) {
+		if (fc)
+			bitslip = bitslip * 2 + 32;
+		else
+			bitslip = (u32)((s32)bitslip * -1 + 20);
+	}
+
+	bitslip <<= ICE_ETH56G_MAC_CFG_FRAC_W;
+	return mul_u32_u32_fx_q9(bitslip, bs);
+}
+
+/**
+ * ice_ptp_calc_deskew_eth56g - Calculate deskew value
+ * @hw: pointer to the HW struct
+ * @port: port to configure
+ * @ds: deskew multiplier
+ * @rs: RS-FEC enabled
+ * @spd: link speed
+ *
+ * Return: calculated deskew value
+ */
+static u32 ice_ptp_calc_deskew_eth56g(struct ice_hw *hw, u8 port, u32 ds,
+				      bool rs, enum ice_eth56g_link_spd spd)
+{
+	u32 deskew_i, deskew_f;
+	int err;
+
+	if (!ds)
+		return 0;
+
+	read_poll_timeout(ice_read_ptp_reg_eth56g, err,
+			  FIELD_GET(PHY_REG_DESKEW_0_VALID, deskew_i), 500,
+			  50 * USEC_PER_MSEC, false, hw, port, PHY_REG_DESKEW_0,
+			  &deskew_i);
+	if (err)
+		return err;
+
+	deskew_f = FIELD_GET(PHY_REG_DESKEW_0_RLEVEL_FRAC, deskew_i);
+	deskew_i = FIELD_GET(PHY_REG_DESKEW_0_RLEVEL, deskew_i);
+
+	if (rs && spd == ICE_ETH56G_LNK_SPD_50G2)
+		ds = 0x633; /* 3.1 */
+	else if (rs && spd == ICE_ETH56G_LNK_SPD_100G)
+		ds = 0x31b; /* 1.552 */
+
+	deskew_i = FIELD_PREP(ICE_ETH56G_MAC_CFG_RX_OFFSET_INT, deskew_i);
+	/* Shift 3 fractional bits to the end of the integer part */
+	deskew_f <<= ICE_ETH56G_MAC_CFG_FRAC_W - PHY_REG_DESKEW_0_RLEVEL_FRAC_W;
+	return mul_u32_u32_fx_q9(deskew_i | deskew_f, ds);
+}
+
+/**
+ * ice_phy_set_offsets_eth56g - Set Tx/Rx offset values
+ * @hw: pointer to the HW struct
+ * @port: port to configure
+ * @spd: link speed
+ * @cfg: structure to store output values
+ * @fc: FC-FEC enabled
+ * @rs: RS-FEC enabled
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write to PHY
+ */
+static int ice_phy_set_offsets_eth56g(struct ice_hw *hw, u8 port,
+				      enum ice_eth56g_link_spd spd,
+				      const struct ice_eth56g_mac_reg_cfg *cfg,
+				      bool fc, bool rs)
+{
+	u32 rx_offset, tx_offset, bs_ds;
+	bool onestep, sfd;
+
+	onestep = hw->ptp.phy.eth56g.onestep_ena;
+	sfd = hw->ptp.phy.eth56g.sfd_ena;
+	bs_ds = cfg->rx_offset.bs_ds;
+
+	if (fc)
+		rx_offset = cfg->rx_offset.fc;
+	else if (rs)
+		rx_offset = cfg->rx_offset.rs;
+	else
+		rx_offset = cfg->rx_offset.no_fec;
+
+	rx_offset = add_u32_u32_fx(rx_offset, cfg->rx_offset.serdes);
+	if (sfd)
+		rx_offset = add_u32_u32_fx(rx_offset, cfg->rx_offset.sfd);
+
+	if (spd < ICE_ETH56G_LNK_SPD_40G)
+		bs_ds = ice_ptp_calc_bitslip_eth56g(hw, port, bs_ds, fc, rs,
+						    spd);
+	else
+		bs_ds = ice_ptp_calc_deskew_eth56g(hw, port, bs_ds, rs, spd);
+	rx_offset = add_u32_u32_fx(rx_offset, bs_ds);
+	rx_offset &= ICE_ETH56G_MAC_CFG_RX_OFFSET_INT |
+		     ICE_ETH56G_MAC_CFG_RX_OFFSET_FRAC;
+
+	if (fc)
+		tx_offset = cfg->tx_offset.fc;
+	else if (rs)
+		tx_offset = cfg->tx_offset.rs;
+	else
+		tx_offset = cfg->tx_offset.no_fec;
+	tx_offset += cfg->tx_offset.serdes + cfg->tx_offset.sfd * sfd +
+		     cfg->tx_offset.onestep * onestep;
+
+	ice_write_mac_reg_eth56g(hw, port, PHY_MAC_RX_OFFSET, rx_offset);
+	return ice_write_mac_reg_eth56g(hw, port, PHY_MAC_TX_OFFSET, tx_offset);
+}
+
+/**
+ * ice_phy_cfg_mac_eth56g - Configure MAC for PTP
+ * @hw: Pointer to the HW struct
+ * @port: Port to configure
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write to PHY
+ */
+static int ice_phy_cfg_mac_eth56g(struct ice_hw *hw, u8 port)
+{
+	const struct ice_eth56g_mac_reg_cfg *cfg;
+	enum ice_eth56g_link_spd spd;
+	struct ice_link_status *li;
+	bool fc = false;
+	bool rs = false;
+	bool onestep;
+	u32 val;
+	int err;
+
+	onestep = hw->ptp.phy.eth56g.onestep_ena;
+	li = &hw->port_info->phy.link_info;
+	spd = ice_phy_get_speed_eth56g(li);
+	if (!!(li->an_info & ICE_AQ_FEC_EN)) {
+		if (spd == ICE_ETH56G_LNK_SPD_10G) {
+			fc = true;
+		} else {
+			fc = !!(li->fec_info & ICE_AQ_LINK_25G_KR_FEC_EN);
+			rs = !!(li->fec_info & ~ICE_AQ_LINK_25G_KR_FEC_EN);
+		}
+	}
+	cfg = &eth56g_mac_cfg[spd];
+
+	err = ice_write_mac_reg_eth56g(hw, port, PHY_MAC_RX_MODULO, 0);
+	if (err)
+		return err;
+
+	err = ice_write_mac_reg_eth56g(hw, port, PHY_MAC_TX_MODULO, 0);
+	if (err)
+		return err;
+
+	val = FIELD_PREP(PHY_MAC_TSU_CFG_TX_MODE_M,
+			 cfg->tx_mode.def + rs * cfg->tx_mode.rs) |
+	      FIELD_PREP(PHY_MAC_TSU_CFG_TX_MII_MK_DLY_M, cfg->tx_mk_dly) |
+	      FIELD_PREP(PHY_MAC_TSU_CFG_TX_MII_CW_DLY_M,
+			 cfg->tx_cw_dly.def +
+			 onestep * cfg->tx_cw_dly.onestep) |
+	      FIELD_PREP(PHY_MAC_TSU_CFG_RX_MODE_M,
+			 cfg->rx_mode.def + rs * cfg->rx_mode.rs) |
+	      FIELD_PREP(PHY_MAC_TSU_CFG_RX_MII_MK_DLY_M,
+			 cfg->rx_mk_dly.def + rs * cfg->rx_mk_dly.rs) |
+	      FIELD_PREP(PHY_MAC_TSU_CFG_RX_MII_CW_DLY_M,
+			 cfg->rx_cw_dly.def + rs * cfg->rx_cw_dly.rs) |
+	      FIELD_PREP(PHY_MAC_TSU_CFG_BLKS_PER_CLK_M, cfg->blks_per_clk);
+	err = ice_write_mac_reg_eth56g(hw, port, PHY_MAC_TSU_CONFIG, val);
+	if (err)
+		return err;
+
+	err = ice_write_mac_reg_eth56g(hw, port, PHY_MAC_BLOCKTIME,
+				       cfg->blktime);
+	if (err)
+		return err;
+
+	err = ice_phy_set_offsets_eth56g(hw, port, spd, cfg, fc, rs);
+	if (err)
+		return err;
+
+	if (spd == ICE_ETH56G_LNK_SPD_25G && !rs)
+		val = 0;
+	else
+		val = cfg->mktime;
+
+	return ice_write_mac_reg_eth56g(hw, port, PHY_MAC_MARKERTIME, val);
+}
+
+/**
+ * ice_phy_cfg_intr_eth56g - Configure TX timestamp interrupt
+ * @hw: pointer to the HW struct
+ * @port: the timestamp port
+ * @ena: enable or disable interrupt
+ * @threshold: interrupt threshold
+ *
+ * Configure TX timestamp interrupt for the specified port
+ *
+ * Return:
+ * * %0     - success
+ * * %other - PHY read/write failed
+ */
+int ice_phy_cfg_intr_eth56g(struct ice_hw *hw, u8 port, bool ena, u8 threshold)
+{
+	int err;
+	u32 val;
+
+	err = ice_read_ptp_reg_eth56g(hw, port, PHY_REG_TS_INT_CONFIG, &val);
+	if (err)
+		return err;
+
+	if (ena) {
+		val |= PHY_TS_INT_CONFIG_ENA_M;
+		val &= ~PHY_TS_INT_CONFIG_THRESHOLD_M;
+		val |= FIELD_PREP(PHY_TS_INT_CONFIG_THRESHOLD_M, threshold);
+	} else {
+		val &= ~PHY_TS_INT_CONFIG_ENA_M;
+	}
+
+	return ice_write_ptp_reg_eth56g(hw, port, PHY_REG_TS_INT_CONFIG, val);
+}
+
+/**
+ * ice_read_phy_and_phc_time_eth56g - Simultaneously capture PHC and PHY time
+ * @hw: pointer to the HW struct
+ * @port: the PHY port to read
+ * @phy_time: on return, the 64bit PHY timer value
+ * @phc_time: on return, the lower 64bits of PHC time
+ *
+ * Issue a ICE_PTP_READ_TIME timer command to simultaneously capture the PHY
+ * and PHC timer values.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - PHY read/write failed
+ */
+static int ice_read_phy_and_phc_time_eth56g(struct ice_hw *hw, u8 port,
+					    u64 *phy_time, u64 *phc_time)
+{
+	u64 tx_time, rx_time;
+	u32 zo, lo;
+	u8 tmr_idx;
+	int err;
+
+	tmr_idx = ice_get_ptp_src_clock_index(hw);
+
+	/* Prepare the PHC timer for a ICE_PTP_READ_TIME capture command */
+	ice_ptp_src_cmd(hw, ICE_PTP_READ_TIME);
+
+	/* Prepare the PHY timer for a ICE_PTP_READ_TIME capture command */
+	err = ice_ptp_one_port_cmd(hw, port, ICE_PTP_READ_TIME);
+	if (err)
+		return err;
+
+	/* Issue the sync to start the ICE_PTP_READ_TIME capture */
+	ice_ptp_exec_tmr_cmd(hw);
+
+	/* Read the captured PHC time from the shadow time registers */
+	zo = rd32(hw, GLTSYN_SHTIME_0(tmr_idx));
+	lo = rd32(hw, GLTSYN_SHTIME_L(tmr_idx));
+	*phc_time = (u64)lo << 32 | zo;
+
+	/* Read the captured PHY time from the PHY shadow registers */
+	err = ice_ptp_read_port_capture_eth56g(hw, port, &tx_time, &rx_time);
+	if (err)
+		return err;
+
+	/* If the PHY Tx and Rx timers don't match, log a warning message.
+	 * Note that this should not happen in normal circumstances since the
+	 * driver always programs them together.
+	 */
+	if (tx_time != rx_time)
+		dev_warn(ice_hw_to_dev(hw), "PHY port %u Tx and Rx timers do not match, tx_time 0x%016llX, rx_time 0x%016llX\n",
+			 port, tx_time, rx_time);
+
+	*phy_time = tx_time;
+
+	return 0;
+}
+
+/**
+ * ice_sync_phy_timer_eth56g - Synchronize the PHY timer with PHC timer
+ * @hw: pointer to the HW struct
+ * @port: the PHY port to synchronize
+ *
+ * Perform an adjustment to ensure that the PHY and PHC timers are in sync.
+ * This is done by issuing a ICE_PTP_READ_TIME command which triggers a
+ * simultaneous read of the PHY timer and PHC timer. Then we use the
+ * difference to calculate an appropriate 2s complement addition to add
+ * to the PHY timer in order to ensure it reads the same value as the
+ * primary PHC timer.
+ *
+ * Return:
+ * * %0     - success
+ * * %-EBUSY- failed to acquire PTP semaphore
+ * * %other - PHY read/write failed
+ */
+static int ice_sync_phy_timer_eth56g(struct ice_hw *hw, u8 port)
+{
+	u64 phc_time, phy_time, difference;
+	int err;
+
+	if (!ice_ptp_lock(hw)) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to acquire PTP semaphore\n");
+		return -EBUSY;
+	}
+
+	err = ice_read_phy_and_phc_time_eth56g(hw, port, &phy_time, &phc_time);
+	if (err)
+		goto err_unlock;
+
+	/* Calculate the amount required to add to the port time in order for
+	 * it to match the PHC time.
+	 *
+	 * Note that the port adjustment is done using 2s complement
+	 * arithmetic. This is convenient since it means that we can simply
+	 * calculate the difference between the PHC time and the port time,
+	 * and it will be interpreted correctly.
+	 */
+
+	ice_ptp_src_cmd(hw, ICE_PTP_NOP);
+	difference = phc_time - phy_time;
+
+	err = ice_ptp_prep_port_adj_eth56g(hw, port, (s64)difference);
+	if (err)
+		goto err_unlock;
+
+	err = ice_ptp_one_port_cmd(hw, port, ICE_PTP_ADJ_TIME);
+	if (err)
+		goto err_unlock;
+
+	/* Issue the sync to activate the time adjustment */
+	ice_ptp_exec_tmr_cmd(hw);
+
+	/* Re-capture the timer values to flush the command registers and
+	 * verify that the time was properly adjusted.
+	 */
+	err = ice_read_phy_and_phc_time_eth56g(hw, port, &phy_time, &phc_time);
+	if (err)
+		goto err_unlock;
+
+	dev_info(ice_hw_to_dev(hw),
+		 "Port %u PHY time synced to PHC: 0x%016llX, 0x%016llX\n",
+		 port, phy_time, phc_time);
+
+err_unlock:
+	ice_ptp_unlock(hw);
+	return err;
+}
+
+/**
+ * ice_stop_phy_timer_eth56g - Stop the PHY clock timer
+ * @hw: pointer to the HW struct
+ * @port: the PHY port to stop
+ * @soft_reset: if true, hold the SOFT_RESET bit of PHY_REG_PS
+ *
+ * Stop the clock of a PHY port. This must be done as part of the flow to
+ * re-calibrate Tx and Rx timestamping offsets whenever the clock time is
+ * initialized or when link speed changes.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write to PHY
+ */
+int ice_stop_phy_timer_eth56g(struct ice_hw *hw, u8 port, bool soft_reset)
+{
+	int err;
+
+	err = ice_write_ptp_reg_eth56g(hw, port, PHY_REG_TX_OFFSET_READY, 0);
+	if (err)
+		return err;
+
+	err = ice_write_ptp_reg_eth56g(hw, port, PHY_REG_RX_OFFSET_READY, 0);
+	if (err)
+		return err;
+
+	ice_debug(hw, ICE_DBG_PTP, "Disabled clock on PHY port %u\n", port);
+
+	return 0;
+}
+
+/**
+ * ice_start_phy_timer_eth56g - Start the PHY clock timer
+ * @hw: pointer to the HW struct
+ * @port: the PHY port to start
+ *
+ * Start the clock of a PHY port. This must be done as part of the flow to
+ * re-calibrate Tx and Rx timestamping offsets whenever the clock time is
+ * initialized or when link speed changes.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - PHY read/write failed
+ */
+int ice_start_phy_timer_eth56g(struct ice_hw *hw, u8 port)
+{
+	u32 lo, hi;
+	u64 incval;
+	u8 tmr_idx;
+	int err;
+
+	tmr_idx = ice_get_ptp_src_clock_index(hw);
+
+	err = ice_stop_phy_timer_eth56g(hw, port, false);
+	if (err)
+		return err;
+
+	ice_ptp_src_cmd(hw, ICE_PTP_NOP);
+
+	err = ice_phy_cfg_parpcs_eth56g(hw, port);
+	if (err)
+		return err;
+
+	err = ice_phy_cfg_ptp_1step_eth56g(hw, port);
+	if (err)
+		return err;
+
+	err = ice_phy_cfg_mac_eth56g(hw, port);
+	if (err)
+		return err;
+
+	lo = rd32(hw, GLTSYN_INCVAL_L(tmr_idx));
+	hi = rd32(hw, GLTSYN_INCVAL_H(tmr_idx));
+	incval = (u64)hi << 32 | lo;
+
+	err = ice_write_40b_ptp_reg_eth56g(hw, port, PHY_REG_TIMETUS_L, incval);
+	if (err)
+		return err;
+
+	err = ice_ptp_one_port_cmd(hw, port, ICE_PTP_INIT_INCVAL);
+	if (err)
+		return err;
+
+	ice_ptp_exec_tmr_cmd(hw);
+
+	err = ice_sync_phy_timer_eth56g(hw, port);
+	if (err)
+		return err;
+
+	err = ice_write_ptp_reg_eth56g(hw, port, PHY_REG_TX_OFFSET_READY, 1);
+	if (err)
+		return err;
+
+	err = ice_write_ptp_reg_eth56g(hw, port, PHY_REG_RX_OFFSET_READY, 1);
+	if (err)
+		return err;
+
+	ice_debug(hw, ICE_DBG_PTP, "Enabled clock on PHY port %u\n", port);
+
+	return 0;
+}
+
+/**
+ * ice_sb_access_ena_eth56g - Enable SB devices (PHY and others) access
+ * @hw: pointer to HW struct
+ * @enable: Enable or disable access
+ *
+ * Enable sideband devices (PHY and others) access.
+ */
+static void ice_sb_access_ena_eth56g(struct ice_hw *hw, bool enable)
+{
+	u32 val = rd32(hw, PF_SB_REM_DEV_CTL);
+
+	if (enable)
+		val |= BIT(eth56g_phy_0) | BIT(cgu) | BIT(eth56g_phy_1);
+	else
+		val &= ~(BIT(eth56g_phy_0) | BIT(cgu) | BIT(eth56g_phy_1));
+
+	wr32(hw, PF_SB_REM_DEV_CTL, val);
+}
+
+/**
+ * ice_ptp_init_phc_eth56g - Perform E82X specific PHC initialization
+ * @hw: pointer to HW struct
+ *
+ * Perform PHC initialization steps specific to E82X devices.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to initialize CGU
+ */
+static int ice_ptp_init_phc_eth56g(struct ice_hw *hw)
+{
+	ice_sb_access_ena_eth56g(hw, true);
+	/* Initialize the Clock Generation Unit */
+	return ice_init_cgu_e82x(hw);
+}
+
+/**
+ * ice_ptp_read_tx_hwtstamp_status_eth56g - Get TX timestamp status
+ * @hw: pointer to the HW struct
+ * @ts_status: the timestamp mask pointer
+ *
+ * Read the PHY Tx timestamp status mask indicating which ports have Tx
+ * timestamps available.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to read from PHY
+ */
+int ice_ptp_read_tx_hwtstamp_status_eth56g(struct ice_hw *hw, u32 *ts_status)
+{
+	const struct ice_eth56g_params *params = &hw->ptp.phy.eth56g;
+	u8 phy, mask;
+	u32 status;
+
+	mask = (1 << hw->ptp.ports_per_phy) - 1;
+	*ts_status = 0;
+
+	for (phy = 0; phy < params->num_phys; phy++) {
+		int err;
+
+		err = ice_read_phy_eth56g(hw, phy, PHY_PTP_INT_STATUS, &status);
+		if (err)
+			return err;
+
+		*ts_status |= (status & mask) << (phy * hw->ptp.ports_per_phy);
+	}
+
+	ice_debug(hw, ICE_DBG_PTP, "PHY interrupt err: %x\n", *ts_status);
+
+	return 0;
+}
+
+/**
+ * ice_get_phy_tx_tstamp_ready_eth56g - Read the Tx memory status register
+ * @hw: pointer to the HW struct
+ * @port: the PHY port to read from
+ * @tstamp_ready: contents of the Tx memory status register
+ *
+ * Read the PHY_REG_TX_MEMORY_STATUS register indicating which timestamps in
+ * the PHY are ready. A set bit means the corresponding timestamp is valid and
+ * ready to be captured from the PHY timestamp block.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to read from PHY
+ */
+static int ice_get_phy_tx_tstamp_ready_eth56g(struct ice_hw *hw, u8 port,
+					      u64 *tstamp_ready)
+{
+	int err;
+
+	err = ice_read_64b_ptp_reg_eth56g(hw, port, PHY_REG_TX_MEMORY_STATUS_L,
+					  tstamp_ready);
+	if (err) {
+		ice_debug(hw, ICE_DBG_PTP, "Failed to read TX_MEMORY_STATUS for port %u, err %d\n",
+			  port, err);
+		return err;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_ptp_init_phy_e825c - initialize PHY parameters
+ * @hw: pointer to the HW struct
+ */
+static void ice_ptp_init_phy_e825c(struct ice_hw *hw)
+{
+	struct ice_ptp_hw *ptp = &hw->ptp;
+	struct ice_eth56g_params *params;
+	u8 phy;
+
+	ptp->phy_model = ICE_PHY_ETH56G;
+	params = &ptp->phy.eth56g;
+	params->onestep_ena = false;
+	params->peer_delay = 0;
+	params->sfd_ena = false;
+	params->phy_addr[0] = eth56g_phy_0;
+	params->phy_addr[1] = eth56g_phy_1;
+	params->num_phys = 2;
+	ptp->ports_per_phy = 4;
+	ptp->num_lports = params->num_phys * ptp->ports_per_phy;
+
+	ice_sb_access_ena_eth56g(hw, true);
+	for (phy = 0; phy < params->num_phys; phy++) {
+		u32 phy_rev;
+		int err;
+
+		err = ice_read_phy_eth56g(hw, phy, PHY_REG_REVISION, &phy_rev);
+		if (err || phy_rev != PHY_REVISION_ETH56G) {
+			ptp->phy_model = ICE_PHY_UNSUP;
+			return;
+		}
+	}
+}
+
 /* E822 family functions
  *
  * The following functions operate on the E822 family of devices.
@@ -1223,15 +3028,14 @@ static int ice_ptp_set_vernier_wl(struct ice_hw *hw)
 static int ice_ptp_init_phc_e82x(struct ice_hw *hw)
 {
 	int err;
-	u32 regval;
+	u32 val;
 
 	/* Enable reading switch and PHY registers over the sideband queue */
 #define PF_SB_REM_DEV_CTL_SWITCH_READ BIT(1)
 #define PF_SB_REM_DEV_CTL_PHY0 BIT(2)
-	regval = rd32(hw, PF_SB_REM_DEV_CTL);
-	regval |= (PF_SB_REM_DEV_CTL_SWITCH_READ |
-		   PF_SB_REM_DEV_CTL_PHY0);
-	wr32(hw, PF_SB_REM_DEV_CTL, regval);
+	val = rd32(hw, PF_SB_REM_DEV_CTL);
+	val |= (PF_SB_REM_DEV_CTL_SWITCH_READ | PF_SB_REM_DEV_CTL_PHY0);
+	wr32(hw, PF_SB_REM_DEV_CTL, val);
 
 	/* Initialize the Clock Generation Unit */
 	err = ice_init_cgu_e82x(hw);
@@ -1458,51 +3262,20 @@ ice_ptp_read_port_capture(struct ice_hw *hw, u8 port, u64 *tx_ts, u64 *rx_ts)
  *
  * Prepare the requested port for an upcoming timer sync command.
  *
- * Do not use this function directly. If you want to configure exactly one
- * port, use ice_ptp_one_port_cmd() instead.
+ * Note there is no equivalent of this operation on E810, as that device
+ * always handles all external PHYs internally.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write to PHY
  */
 static int ice_ptp_write_port_cmd_e82x(struct ice_hw *hw, u8 port,
 				       enum ice_ptp_tmr_cmd cmd)
 {
-	u32 cmd_val, val;
-	u8 tmr_idx;
+	u32 val = ice_ptp_tmr_cmd_to_port_reg(hw, cmd);
 	int err;
 
-	tmr_idx = ice_get_ptp_src_clock_index(hw);
-	cmd_val = tmr_idx << SEL_PHY_SRC;
-	switch (cmd) {
-	case ICE_PTP_INIT_TIME:
-		cmd_val |= PHY_CMD_INIT_TIME;
-		break;
-	case ICE_PTP_INIT_INCVAL:
-		cmd_val |= PHY_CMD_INIT_INCVAL;
-		break;
-	case ICE_PTP_ADJ_TIME:
-		cmd_val |= PHY_CMD_ADJ_TIME;
-		break;
-	case ICE_PTP_READ_TIME:
-		cmd_val |= PHY_CMD_READ_TIME;
-		break;
-	case ICE_PTP_ADJ_TIME_AT_TIME:
-		cmd_val |= PHY_CMD_ADJ_TIME_AT_TIME;
-		break;
-	case ICE_PTP_NOP:
-		break;
-	}
-
 	/* Tx case */
-	/* Read, modify, write */
-	err = ice_read_phy_reg_e82x(hw, port, P_REG_TX_TMR_CMD, &val);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to read TX_TMR_CMD, err %d\n",
-			  err);
-		return err;
-	}
-
-	/* Modify necessary bits only and perform write */
-	val &= ~TS_CMD_MASK;
-	val |= cmd_val;
-
 	err = ice_write_phy_reg_e82x(hw, port, P_REG_TX_TMR_CMD, val);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "Failed to write back TX_TMR_CMD, err %d\n",
@@ -1511,19 +3284,8 @@ static int ice_ptp_write_port_cmd_e82x(struct ice_hw *hw, u8 port,
 	}
 
 	/* Rx case */
-	/* Read, modify, write */
-	err = ice_read_phy_reg_e82x(hw, port, P_REG_RX_TMR_CMD, &val);
-	if (err) {
-		ice_debug(hw, ICE_DBG_PTP, "Failed to read RX_TMR_CMD, err %d\n",
-			  err);
-		return err;
-	}
-
-	/* Modify necessary bits only and perform write */
-	val &= ~TS_CMD_MASK;
-	val |= cmd_val;
-
-	err = ice_write_phy_reg_e82x(hw, port, P_REG_RX_TMR_CMD, val);
+	err = ice_write_phy_reg_e82x(hw, port, P_REG_RX_TMR_CMD,
+				     val | TS_CMD_RX_TYPE);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "Failed to write back RX_TMR_CMD, err %d\n",
 			  err);
@@ -1533,63 +3295,6 @@ static int ice_ptp_write_port_cmd_e82x(struct ice_hw *hw, u8 port,
 	return 0;
 }
 
-/**
- * ice_ptp_one_port_cmd - Prepare one port for a timer command
- * @hw: pointer to the HW struct
- * @configured_port: the port to configure with configured_cmd
- * @configured_cmd: timer command to prepare on the configured_port
- *
- * Prepare the configured_port for the configured_cmd, and prepare all other
- * ports for ICE_PTP_NOP. This causes the configured_port to execute the
- * desired command while all other ports perform no operation.
- */
-static int
-ice_ptp_one_port_cmd(struct ice_hw *hw, u8 configured_port,
-		     enum ice_ptp_tmr_cmd configured_cmd)
-{
-	u8 port;
-
-	for (port = 0; port < hw->ptp.num_lports; port++) {
-		enum ice_ptp_tmr_cmd cmd;
-		int err;
-
-		if (port == configured_port)
-			cmd = configured_cmd;
-		else
-			cmd = ICE_PTP_NOP;
-
-		err = ice_ptp_write_port_cmd_e82x(hw, port, cmd);
-		if (err)
-			return err;
-	}
-
-	return 0;
-}
-
-/**
- * ice_ptp_port_cmd_e82x - Prepare all ports for a timer command
- * @hw: pointer to the HW struct
- * @cmd: timer command to prepare
- *
- * Prepare all ports connected to this device for an upcoming timer sync
- * command.
- */
-static int
-ice_ptp_port_cmd_e82x(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd)
-{
-	u8 port;
-
-	for (port = 0; port < hw->ptp.num_lports; port++) {
-		int err;
-
-		err = ice_ptp_write_port_cmd_e82x(hw, port, cmd);
-		if (err)
-			return err;
-	}
-
-	return 0;
-}
-
 /* E822 Vernier calibration functions
  *
  * The following functions are used as part of the vernier calibration of
@@ -3437,10 +5142,114 @@ void ice_ptp_init_hw(struct ice_hw *hw)
 		ice_ptp_init_phy_e82x(ptp);
 	else if (ice_is_e810(hw))
 		ice_ptp_init_phy_e810(ptp);
+	else if (ice_is_e825c(hw))
+		ice_ptp_init_phy_e825c(hw);
 	else
 		ptp->phy_model = ICE_PHY_UNSUP;
 }
 
+/**
+ * ice_ptp_write_port_cmd - Prepare a single PHY port for a timer command
+ * @hw: pointer to HW struct
+ * @port: Port to which cmd has to be sent
+ * @cmd: Command to be sent to the port
+ *
+ * Prepare one port for the upcoming timer sync command. Do not use this for
+ * programming only a single port, instead use ice_ptp_one_port_cmd() to
+ * ensure non-modified ports get properly initialized to ICE_PTP_NOP.
+ *
+ * Return:
+ * * %0     - success
+ *  %-EBUSY - PHY type not supported
+ * * %other - failed to write port command
+ */
+static int ice_ptp_write_port_cmd(struct ice_hw *hw, u8 port,
+				  enum ice_ptp_tmr_cmd cmd)
+{
+	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		return ice_ptp_write_port_cmd_eth56g(hw, port, cmd);
+	case ICE_PHY_E82X:
+		return ice_ptp_write_port_cmd_e82x(hw, port, cmd);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+/**
+ * ice_ptp_one_port_cmd - Program one PHY port for a timer command
+ * @hw: pointer to HW struct
+ * @configured_port: the port that should execute the command
+ * @configured_cmd: the command to be executed on the configured port
+ *
+ * Prepare one port for executing a timer command, while preparing all other
+ * ports to ICE_PTP_NOP. This allows executing a command on a single port
+ * while ensuring all other ports do not execute stale commands.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write port command
+ */
+int ice_ptp_one_port_cmd(struct ice_hw *hw, u8 configured_port,
+			 enum ice_ptp_tmr_cmd configured_cmd)
+{
+	u32 port;
+
+	for (port = 0; port < hw->ptp.num_lports; port++) {
+		int err;
+
+		/* Program the configured port with the configured command,
+		 * program all other ports with ICE_PTP_NOP.
+		 */
+		if (port == configured_port)
+			err = ice_ptp_write_port_cmd(hw, port, configured_cmd);
+		else
+			err = ice_ptp_write_port_cmd(hw, port, ICE_PTP_NOP);
+
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_ptp_port_cmd - Prepare PHY ports for a timer sync command
+ * @hw: pointer to HW struct
+ * @cmd: the timer command to setup
+ *
+ * Prepare all PHY ports on this device for the requested timer command. For
+ * some families this can be done in one shot, but for other families each
+ * port must be configured individually.
+ *
+ * Return:
+ * * %0     - success
+ * * %other - failed to write port command
+ */
+static int ice_ptp_port_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd)
+{
+	u32 port;
+
+	/* PHY models which can program all ports simultaneously */
+	switch (hw->ptp.phy_model) {
+	case ICE_PHY_E810:
+		return ice_ptp_port_cmd_e810(hw, cmd);
+	default:
+		break;
+	}
+
+	/* PHY models which require programming each port separately */
+	for (port = 0; port < hw->ptp.num_lports; port++) {
+		int err;
+
+		err = ice_ptp_write_port_cmd(hw, port, cmd);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
 /**
  * ice_ptp_tmr_cmd - Prepare and trigger a timer sync command
  * @hw: pointer to HW struct
@@ -3459,17 +5268,7 @@ static int ice_ptp_tmr_cmd(struct ice_hw *hw, enum ice_ptp_tmr_cmd cmd)
 	ice_ptp_src_cmd(hw, cmd);
 
 	/* Next, prepare the ports */
-	switch (hw->ptp.phy_model) {
-	case ICE_PHY_E810:
-		err = ice_ptp_port_cmd_e810(hw, cmd);
-		break;
-	case ICE_PHY_E82X:
-		err = ice_ptp_port_cmd_e82x(hw, cmd);
-		break;
-	default:
-		err = -EOPNOTSUPP;
-	}
-
+	err = ice_ptp_port_cmd(hw, cmd);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "Failed to prepare PHY ports for timer command %u, err %d\n",
 			  cmd, err);
@@ -3512,6 +5311,10 @@ int ice_ptp_init_time(struct ice_hw *hw, u64 time)
 	/* PHY timers */
 	/* Fill Rx and Tx ports and send msg to PHY */
 	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		err = ice_ptp_prep_phy_time_eth56g(hw,
+						   (u32)(time & 0xFFFFFFFF));
+		break;
 	case ICE_PHY_E810:
 		err = ice_ptp_prep_phy_time_e810(hw, time & 0xFFFFFFFF);
 		break;
@@ -3554,6 +5357,9 @@ int ice_ptp_write_incval(struct ice_hw *hw, u64 incval)
 	wr32(hw, GLTSYN_SHADJ_H(tmr_idx), upper_32_bits(incval));
 
 	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		err = ice_ptp_prep_phy_incval_eth56g(hw, incval);
+		break;
 	case ICE_PHY_E810:
 		err = ice_ptp_prep_phy_incval_e810(hw, incval);
 		break;
@@ -3620,6 +5426,9 @@ int ice_ptp_adj_clock(struct ice_hw *hw, s32 adj)
 	wr32(hw, GLTSYN_SHADJ_H(tmr_idx), adj);
 
 	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		err = ice_ptp_prep_phy_adj_eth56g(hw, adj);
+		break;
 	case ICE_PHY_E810:
 		err = ice_ptp_prep_phy_adj_e810(hw, adj);
 		break;
@@ -3650,6 +5459,8 @@ int ice_ptp_adj_clock(struct ice_hw *hw, s32 adj)
 int ice_read_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx, u64 *tstamp)
 {
 	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		return ice_read_ptp_tstamp_eth56g(hw, block, idx, tstamp);
 	case ICE_PHY_E810:
 		return ice_read_phy_tstamp_e810(hw, block, idx, tstamp);
 	case ICE_PHY_E82X:
@@ -3678,6 +5489,8 @@ int ice_read_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx, u64 *tstamp)
 int ice_clear_phy_tstamp(struct ice_hw *hw, u8 block, u8 idx)
 {
 	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		return ice_clear_ptp_tstamp_eth56g(hw, block, idx);
 	case ICE_PHY_E810:
 		return ice_clear_phy_tstamp_e810(hw, block, idx);
 	case ICE_PHY_E82X:
@@ -3739,6 +5552,9 @@ static int ice_get_pf_c827_idx(struct ice_hw *hw, u8 *idx)
 void ice_ptp_reset_ts_memory(struct ice_hw *hw)
 {
 	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		ice_ptp_reset_ts_memory_eth56g(hw);
+		break;
 	case ICE_PHY_E82X:
 		ice_ptp_reset_ts_memory_e82x(hw);
 		break;
@@ -3765,6 +5581,8 @@ int ice_ptp_init_phc(struct ice_hw *hw)
 	(void)rd32(hw, GLTSYN_STAT(src_idx));
 
 	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		return ice_ptp_init_phc_eth56g(hw);
 	case ICE_PHY_E810:
 		return ice_ptp_init_phc_e810(hw);
 	case ICE_PHY_E82X:
@@ -3788,6 +5606,9 @@ int ice_ptp_init_phc(struct ice_hw *hw)
 int ice_get_phy_tx_tstamp_ready(struct ice_hw *hw, u8 block, u64 *tstamp_ready)
 {
 	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		return ice_get_phy_tx_tstamp_ready_eth56g(hw, block,
+							  tstamp_ready);
 	case ICE_PHY_E810:
 		return ice_get_phy_tx_tstamp_ready_e810(hw, block,
 							tstamp_ready);
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
index 3fe12a1694fc..184573509e62 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
@@ -41,6 +41,41 @@ enum ice_ptp_fec_mode {
 	ICE_PTP_FEC_MODE_RS_FEC
 };
 
+enum eth56g_res_type {
+	ETH56G_PHY_REG_PTP,
+	ETH56G_PHY_MEM_PTP,
+	ETH56G_PHY_REG_XPCS,
+	ETH56G_PHY_REG_MAC,
+	ETH56G_PHY_REG_GPCS,
+	NUM_ETH56G_PHY_RES
+};
+
+enum ice_eth56g_link_spd {
+	ICE_ETH56G_LNK_SPD_1G,
+	ICE_ETH56G_LNK_SPD_2_5G,
+	ICE_ETH56G_LNK_SPD_10G,
+	ICE_ETH56G_LNK_SPD_25G,
+	ICE_ETH56G_LNK_SPD_40G,
+	ICE_ETH56G_LNK_SPD_50G,
+	ICE_ETH56G_LNK_SPD_50G2,
+	ICE_ETH56G_LNK_SPD_100G,
+	ICE_ETH56G_LNK_SPD_100G2,
+	NUM_ICE_ETH56G_LNK_SPD /* Must be last */
+};
+
+/**
+ * struct ice_phy_reg_info_eth56g - ETH56G PHY register parameters
+ * @base: base address for each PHY block
+ * @step: step between PHY lanes
+ *
+ * Characteristic information for the various PHY register parameters in the
+ * ETH56G devices
+ */
+struct ice_phy_reg_info_eth56g {
+	u32 base[NUM_ETH56G_PHY_RES];
+	u32 step;
+};
+
 /**
  * struct ice_time_ref_info_e82x
  * @pll_freq: Frequency of PLL that drives timer ticks in Hz
@@ -94,6 +129,73 @@ struct ice_vernier_info_e82x {
 	u32 rx_fixed_delay;
 };
 
+#define ICE_ETH56G_MAC_CFG_RX_OFFSET_INT	GENMASK(19, 9)
+#define ICE_ETH56G_MAC_CFG_RX_OFFSET_FRAC	GENMASK(8, 0)
+#define ICE_ETH56G_MAC_CFG_FRAC_W		9
+/**
+ * struct ice_eth56g_mac_reg_cfg - MAC config values for specific PTP registers
+ * @tx_mode: Tx timestamp compensation mode
+ * @tx_mk_dly: Tx timestamp marker start strobe delay
+ * @tx_cw_dly: Tx timestamp codeword start strobe delay
+ * @rx_mode: Rx timestamp compensation mode
+ * @rx_mk_dly: Rx timestamp marker start strobe delay
+ * @rx_cw_dly: Rx timestamp codeword start strobe delay
+ * @blks_per_clk: number of blocks transferred per clock cycle
+ * @blktime: block time, fixed point
+ * @mktime: marker time, fixed point
+ * @tx_offset: total Tx offset, fixed point
+ * @rx_offset: total Rx offset, contains value for bitslip/deskew, fixed point
+ *
+ * All fixed point registers except Rx offset are 23 bit unsigned ints with
+ * a 9 bit fractional.
+ * Rx offset is 11 bit unsigned int with a 9 bit fractional.
+ */
+struct ice_eth56g_mac_reg_cfg {
+	struct {
+		u8 def;
+		u8 rs;
+	} tx_mode;
+	u8 tx_mk_dly;
+	struct {
+		u8 def;
+		u8 onestep;
+	} tx_cw_dly;
+	struct {
+		u8 def;
+		u8 rs;
+	} rx_mode;
+	struct {
+		u8 def;
+		u8 rs;
+	} rx_mk_dly;
+	struct {
+		u8 def;
+		u8 rs;
+	} rx_cw_dly;
+	u8 blks_per_clk;
+	u16 blktime;
+	u16 mktime;
+	struct {
+		u32 serdes;
+		u32 no_fec;
+		u32 fc;
+		u32 rs;
+		u32 sfd;
+		u32 onestep;
+	} tx_offset;
+	struct {
+		u32 serdes;
+		u32 no_fec;
+		u32 fc;
+		u32 rs;
+		u32 sfd;
+		u32 bs_ds;
+	} rx_offset;
+};
+
+extern
+const struct ice_eth56g_mac_reg_cfg eth56g_mac_cfg[NUM_ICE_ETH56G_LNK_SPD];
+
 /**
  * struct ice_cgu_pll_params_e82x
  * @refclk_pre_div: Reference clock pre-divisor
@@ -188,6 +290,9 @@ ice_cgu_pll_params_e82x e822_cgu_params[NUM_ICE_TIME_REF_FREQ];
 #define E810C_QSFP_C827_0_HANDLE 2
 #define E810C_QSFP_C827_1_HANDLE 3
 
+/* Table of constants related to possible ETH56G PHY resources */
+extern const struct ice_phy_reg_info_eth56g eth56g_phy_res[NUM_ETH56G_PHY_RES];
+
 /* Table of constants related to possible TIME_REF sources */
 extern const struct ice_time_ref_info_e82x e822_time_ref[NUM_ICE_TIME_REF_FREQ];
 
@@ -197,7 +302,9 @@ extern const struct ice_vernier_info_e82x e822_vernier[NUM_ICE_PTP_LNK_SPD];
 /* Increment value to generate nanoseconds in the GLTSYN_TIME_L register for
  * the E810 devices. Based off of a PLL with an 812.5 MHz frequency.
  */
-#define ICE_PTP_NOMINAL_INCVAL_E810 0x13b13b13bULL
+#define ICE_E810_PLL_FREQ		812500000
+#define ICE_PTP_NOMINAL_INCVAL_E810	0x13b13b13bULL
+#define E810_OUT_PROP_DELAY_NS 1
 
 /* Device agnostic functions */
 u8 ice_get_ptp_src_clock_index(struct ice_hw *hw);
@@ -215,6 +322,8 @@ void ice_ptp_reset_ts_memory(struct ice_hw *hw);
 int ice_ptp_init_phc(struct ice_hw *hw);
 void ice_ptp_init_hw(struct ice_hw *hw);
 int ice_get_phy_tx_tstamp_ready(struct ice_hw *hw, u8 block, u64 *tstamp_ready);
+int ice_ptp_one_port_cmd(struct ice_hw *hw, u8 configured_port,
+			 enum ice_ptp_tmr_cmd configured_cmd);
 
 /* E822 family functions */
 int ice_read_quad_reg_e82x(struct ice_hw *hw, u8 quad, u16 offset, u32 *val);
@@ -285,6 +394,21 @@ int ice_get_cgu_rclk_pin_info(struct ice_hw *hw, u8 *base_idx, u8 *pin_num);
 int ice_cgu_get_output_pin_state_caps(struct ice_hw *hw, u8 pin_id,
 				      unsigned long *caps);
 
+/* ETH56G family functions */
+int ice_ptp_read_tx_hwtstamp_status_eth56g(struct ice_hw *hw, u32 *ts_status);
+int ice_stop_phy_timer_eth56g(struct ice_hw *hw, u8 port, bool soft_reset);
+int ice_start_phy_timer_eth56g(struct ice_hw *hw, u8 port);
+int ice_phy_cfg_tx_offset_eth56g(struct ice_hw *hw, u8 port);
+int ice_phy_cfg_rx_offset_eth56g(struct ice_hw *hw, u8 port);
+int ice_phy_cfg_intr_eth56g(struct ice_hw *hw, u8 port, bool ena, u8 threshold);
+int ice_phy_cfg_ptp_1step_eth56g(struct ice_hw *hw, u8 port);
+
+#define ICE_ETH56G_NOMINAL_INCVAL	0x140000000ULL
+#define ICE_ETH56G_NOMINAL_PCS_REF_TUS	0x100000000ULL
+#define ICE_ETH56G_NOMINAL_PCS_REF_INC	0x300000000ULL
+#define ICE_ETH56G_NOMINAL_THRESH4	0x7777
+#define ICE_ETH56G_NOMINAL_TX_THRESH	0x6
+
 /**
  * ice_get_base_incval - Get base clock increment value
  * @hw: pointer to the HW struct
@@ -294,6 +418,8 @@ int ice_cgu_get_output_pin_state_caps(struct ice_hw *hw, u8 pin_id,
 static inline u64 ice_get_base_incval(struct ice_hw *hw)
 {
 	switch (hw->ptp.phy_model) {
+	case ICE_PHY_ETH56G:
+		return ICE_ETH56G_NOMINAL_INCVAL;
 	case ICE_PHY_E810:
 		return ICE_PTP_NOMINAL_INCVAL_E810;
 	case ICE_PHY_E82X:
@@ -330,6 +456,7 @@ static inline u64 ice_get_base_incval(struct ice_hw *hw)
 #define TS_CMD_MASK_E810		0xFF
 #define TS_CMD_MASK			0xF
 #define SYNC_EXEC_CMD			0x3
+#define TS_CMD_RX_TYPE			ICE_M(0x18, 0x4)
 
 /* Macros to derive port low and high addresses on both quads */
 #define P_Q0_L(a, p) ((((a) + (0x2000 * (p)))) & 0xFFFF)
@@ -564,4 +691,115 @@ static inline u64 ice_get_base_incval(struct ice_hw *hw)
 /* E810T PCA9575 IO controller pin control */
 #define ICE_E810T_P0_GNSS_PRSNT_N	BIT(4)
 
+/* ETH56G PHY register addresses */
+/* Timestamp PHY incval registers */
+#define PHY_REG_TIMETUS_L		0x8
+#define PHY_REG_TIMETUS_U		0xC
+
+/* Timestamp PCS registers */
+#define PHY_PCS_REF_TUS_L		0x18
+#define PHY_PCS_REF_TUS_U		0x1C
+
+/* Timestamp PCS ref incval registers */
+#define PHY_PCS_REF_INC_L		0x20
+#define PHY_PCS_REF_INC_U		0x24
+
+/* Timestamp init registers */
+#define PHY_REG_RX_TIMER_INC_PRE_L	0x64
+#define PHY_REG_RX_TIMER_INC_PRE_U	0x68
+#define PHY_REG_TX_TIMER_INC_PRE_L	0x44
+#define PHY_REG_TX_TIMER_INC_PRE_U	0x48
+
+/* Timestamp match and adjust target registers */
+#define PHY_REG_RX_TIMER_CNT_ADJ_L	0x6C
+#define PHY_REG_RX_TIMER_CNT_ADJ_U	0x70
+#define PHY_REG_TX_TIMER_CNT_ADJ_L	0x4C
+#define PHY_REG_TX_TIMER_CNT_ADJ_U	0x50
+
+/* Timestamp command registers */
+#define PHY_REG_TX_TMR_CMD		0x40
+#define PHY_REG_RX_TMR_CMD		0x60
+
+/* Phy offset ready registers */
+#define PHY_REG_TX_OFFSET_READY		0x54
+#define PHY_REG_RX_OFFSET_READY		0x74
+
+/* Phy total offset registers */
+#define PHY_REG_TOTAL_TX_OFFSET_L	0x38
+#define PHY_REG_TOTAL_TX_OFFSET_U	0x3C
+#define PHY_REG_TOTAL_RX_OFFSET_L	0x58
+#define PHY_REG_TOTAL_RX_OFFSET_U	0x5C
+
+/* Timestamp capture registers */
+#define PHY_REG_TX_CAPTURE_L		0x78
+#define PHY_REG_TX_CAPTURE_U		0x7C
+#define PHY_REG_RX_CAPTURE_L		0x8C
+#define PHY_REG_RX_CAPTURE_U		0x90
+
+/* Memory status registers */
+#define PHY_REG_TX_MEMORY_STATUS_L	0x80
+#define PHY_REG_TX_MEMORY_STATUS_U	0x84
+
+/* Interrupt config register */
+#define PHY_REG_TS_INT_CONFIG		0x88
+
+/* XIF mode config register */
+#define PHY_MAC_XIF_MODE		0x24
+#define PHY_MAC_XIF_1STEP_ENA_M		ICE_M(0x1, 5)
+#define PHY_MAC_XIF_TS_BIN_MODE_M	ICE_M(0x1, 11)
+#define PHY_MAC_XIF_TS_SFD_ENA_M	ICE_M(0x1, 20)
+#define PHY_MAC_XIF_GMII_TS_SEL_M	ICE_M(0x1, 21)
+
+/* GPCS config register */
+#define PHY_GPCS_CONFIG_REG0		0x268
+#define PHY_GPCS_CONFIG_REG0_TX_THR_M	ICE_M(0xF, 24)
+#define PHY_GPCS_BITSLIP		0x5C
+
+#define PHY_TS_INT_CONFIG_THRESHOLD_M	ICE_M(0x3F, 0)
+#define PHY_TS_INT_CONFIG_ENA_M		BIT(6)
+
+/* 1-step PTP config */
+#define PHY_PTP_1STEP_CONFIG		0x270
+#define PHY_PTP_1STEP_T1S_UP64_M	ICE_M(0xF, 4)
+#define PHY_PTP_1STEP_T1S_DELTA_M	ICE_M(0xF, 8)
+#define PHY_PTP_1STEP_PEER_DELAY(_port)	(0x274 + 4 * (_port))
+#define PHY_PTP_1STEP_PD_ADD_PD_M	ICE_M(0x1, 0)
+#define PHY_PTP_1STEP_PD_DELAY_M	ICE_M(0x3fffffff, 1)
+#define PHY_PTP_1STEP_PD_DLY_V_M	ICE_M(0x1, 31)
+
+/* Macros to derive offsets for TimeStampLow and TimeStampHigh */
+#define PHY_TSTAMP_L(x) (((x) * 8) + 0)
+#define PHY_TSTAMP_U(x) (((x) * 8) + 4)
+
+#define PHY_REG_REVISION		0x85000
+
+#define PHY_REG_DESKEW_0		0x94
+#define PHY_REG_DESKEW_0_RLEVEL		GENMASK(6, 0)
+#define PHY_REG_DESKEW_0_RLEVEL_FRAC	GENMASK(9, 7)
+#define PHY_REG_DESKEW_0_RLEVEL_FRAC_W	3
+#define PHY_REG_DESKEW_0_VALID		GENMASK(10, 10)
+
+#define PHY_REG_GPCS_BITSLIP		0x5C
+#define PHY_REG_SD_BIT_SLIP(_port_offset)	(0x29C + 4 * (_port_offset))
+#define PHY_REVISION_ETH56G		0x10200
+#define PHY_VENDOR_TXLANE_THRESH	0x2000C
+
+#define PHY_MAC_TSU_CONFIG		0x40
+#define PHY_MAC_TSU_CFG_RX_MODE_M	ICE_M(0x7, 0)
+#define PHY_MAC_TSU_CFG_RX_MII_CW_DLY_M	ICE_M(0x7, 4)
+#define PHY_MAC_TSU_CFG_RX_MII_MK_DLY_M	ICE_M(0x7, 8)
+#define PHY_MAC_TSU_CFG_TX_MODE_M	ICE_M(0x7, 12)
+#define PHY_MAC_TSU_CFG_TX_MII_CW_DLY_M	ICE_M(0x1F, 16)
+#define PHY_MAC_TSU_CFG_TX_MII_MK_DLY_M	ICE_M(0x1F, 21)
+#define PHY_MAC_TSU_CFG_BLKS_PER_CLK_M	ICE_M(0x1, 28)
+#define PHY_MAC_RX_MODULO		0x44
+#define PHY_MAC_RX_OFFSET		0x48
+#define PHY_MAC_RX_OFFSET_M		ICE_M(0xFFFFFF, 0)
+#define PHY_MAC_TX_MODULO		0x4C
+#define PHY_MAC_BLOCKTIME		0x50
+#define PHY_MAC_MARKERTIME		0x54
+#define PHY_MAC_TX_OFFSET		0x58
+
+#define PHY_PTP_INT_STATUS		0x7FD140
+
 #endif /* _ICE_PTP_HW_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_sbq_cmd.h b/drivers/net/ethernet/intel/ice/ice_sbq_cmd.h
index ead75fe2bcda..3b0054faf70c 100644
--- a/drivers/net/ethernet/intel/ice/ice_sbq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_sbq_cmd.h
@@ -47,10 +47,12 @@ struct ice_sbq_evt_desc {
 };
 
 enum ice_sbq_msg_dev {
-	rmn_0	= 0x02,
-	rmn_1	= 0x03,
-	rmn_2	= 0x04,
-	cgu	= 0x06
+	eth56g_phy_0	= 0x02,
+	rmn_0		= 0x02,
+	rmn_1		= 0x03,
+	rmn_2		= 0x04,
+	cgu		= 0x06,
+	eth56g_phy_1	= 0x0D,
 };
 
 enum ice_sbq_msg_opcode {
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index 2d2af80a65a9..f039fc665455 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -322,7 +322,9 @@ enum ice_time_ref_freq {
 	ICE_TIME_REF_FREQ_156_250	= 4,
 	ICE_TIME_REF_FREQ_245_760	= 5,
 
-	NUM_ICE_TIME_REF_FREQ
+	NUM_ICE_TIME_REF_FREQ,
+
+	ICE_TIME_REF_FREQ_INVALID	= -1,
 };
 
 /* Clock source specification */
@@ -821,15 +823,29 @@ struct ice_mbx_data {
 #define ICE_PORTS_PER_QUAD	4
 #define ICE_GET_QUAD_NUM(port) ((port) / ICE_PORTS_PER_QUAD)
 
+struct ice_eth56g_params {
+	u8 num_phys;
+	u8 phy_addr[2];
+	bool onestep_ena;
+	bool sfd_ena;
+	u32 peer_delay;
+};
+
+union ice_phy_params {
+	struct ice_eth56g_params eth56g;
+};
+
 /* PHY model */
 enum ice_phy_model {
 	ICE_PHY_UNSUP = -1,
-	ICE_PHY_E810  = 1,
+	ICE_PHY_E810 = 1,
 	ICE_PHY_E82X,
+	ICE_PHY_ETH56G,
 };
 
 struct ice_ptp_hw {
 	enum ice_phy_model phy_model;
+	union ice_phy_params phy;
 	u8 num_lports;
 	u8 ports_per_phy;
 };
-- 
cgit v1.2.3-58-ga151


From b390ecc2e375f8973c512a5041a69555069f502a Mon Sep 17 00:00:00 2001
From: Karol Kolacinski <karol.kolacinski@intel.com>
Date: Tue, 28 May 2024 16:03:58 -0700
Subject: ice: Change CGU regs struct to anonymous

Simplify the code by using anonymous struct in CGU registers instead of
naming each structure 'field'.

Suggested-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528-next-2024-05-28-ptp-refactors-v1-8-c082739bb6f6@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_cgu_regs.h | 12 ++++----
 drivers/net/ethernet/intel/ice/ice_ptp_hw.c   | 44 +++++++++++++--------------
 2 files changed, 28 insertions(+), 28 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_cgu_regs.h b/drivers/net/ethernet/intel/ice/ice_cgu_regs.h
index 57abd52386d0..36aeb10eefb7 100644
--- a/drivers/net/ethernet/intel/ice/ice_cgu_regs.h
+++ b/drivers/net/ethernet/intel/ice/ice_cgu_regs.h
@@ -23,7 +23,7 @@ union nac_cgu_dword9 {
 		u32 clk_synce0_amp : 2;
 		u32 one_pps_out_amp : 2;
 		u32 misc24 : 12;
-	} field;
+	};
 	u32 val;
 };
 
@@ -39,7 +39,7 @@ union nac_cgu_dword19 {
 		u32 japll_ndivratio : 4;
 		u32 japll_iref_ndivratio : 3;
 		u32 misc27 : 1;
-	} field;
+	};
 	u32 val;
 };
 
@@ -63,7 +63,7 @@ union nac_cgu_dword22 {
 		u32 fdpllclk_sel_div2 : 1;
 		u32 time1588clk_sel_div2 : 1;
 		u32 misc3 : 1;
-	} field;
+	};
 	u32 val;
 };
 
@@ -77,7 +77,7 @@ union nac_cgu_dword24 {
 		u32 ext_synce_sel : 1;
 		u32 ref1588_ck_div : 4;
 		u32 time_ref_sel : 1;
-	} field;
+	};
 	u32 val;
 };
 
@@ -92,7 +92,7 @@ union tspll_cntr_bist_settings {
 		u32 i_plllock_cnt_6_0 : 7;
 		u32 i_plllock_cnt_10_7 : 4;
 		u32 reserved200 : 4;
-	} field;
+	};
 	u32 val;
 };
 
@@ -109,7 +109,7 @@ union tspll_ro_bwm_lf {
 		u32 afcdone_cri : 1;
 		u32 feedfwrdgain_cal_cri_7_0 : 8;
 		u32 m2fbdivmod_cri_7_0 : 8;
-	} field;
+	};
 	u32 val;
 };
 
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index 1b333b1c2738..691ea20811e7 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -393,14 +393,14 @@ static int ice_cfg_cgu_pll_e82x(struct ice_hw *hw,
 
 	/* Log the current clock configuration */
 	ice_debug(hw, ICE_DBG_PTP, "Current CGU configuration -- %s, clk_src %s, clk_freq %s, PLL %s\n",
-		  dw24.field.ts_pll_enable ? "enabled" : "disabled",
-		  ice_clk_src_str(dw24.field.time_ref_sel),
-		  ice_clk_freq_str(dw9.field.time_ref_freq_sel),
-		  bwm_lf.field.plllock_true_lock_cri ? "locked" : "unlocked");
+		  dw24.ts_pll_enable ? "enabled" : "disabled",
+		  ice_clk_src_str(dw24.time_ref_sel),
+		  ice_clk_freq_str(dw9.time_ref_freq_sel),
+		  bwm_lf.plllock_true_lock_cri ? "locked" : "unlocked");
 
 	/* Disable the PLL before changing the clock source or frequency */
-	if (dw24.field.ts_pll_enable) {
-		dw24.field.ts_pll_enable = 0;
+	if (dw24.ts_pll_enable) {
+		dw24.ts_pll_enable = 0;
 
 		err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD24, dw24.val);
 		if (err)
@@ -408,7 +408,7 @@ static int ice_cfg_cgu_pll_e82x(struct ice_hw *hw,
 	}
 
 	/* Set the frequency */
-	dw9.field.time_ref_freq_sel = clk_freq;
+	dw9.time_ref_freq_sel = clk_freq;
 	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD9, dw9.val);
 	if (err)
 		return err;
@@ -418,8 +418,8 @@ static int ice_cfg_cgu_pll_e82x(struct ice_hw *hw,
 	if (err)
 		return err;
 
-	dw19.field.tspll_fbdiv_intgr = e822_cgu_params[clk_freq].feedback_div;
-	dw19.field.tspll_ndivratio = 1;
+	dw19.tspll_fbdiv_intgr = e822_cgu_params[clk_freq].feedback_div;
+	dw19.tspll_ndivratio = 1;
 
 	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD19, dw19.val);
 	if (err)
@@ -430,8 +430,8 @@ static int ice_cfg_cgu_pll_e82x(struct ice_hw *hw,
 	if (err)
 		return err;
 
-	dw22.field.time1588clk_div = e822_cgu_params[clk_freq].post_pll_div;
-	dw22.field.time1588clk_sel_div2 = 0;
+	dw22.time1588clk_div = e822_cgu_params[clk_freq].post_pll_div;
+	dw22.time1588clk_sel_div2 = 0;
 
 	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD22, dw22.val);
 	if (err)
@@ -442,16 +442,16 @@ static int ice_cfg_cgu_pll_e82x(struct ice_hw *hw,
 	if (err)
 		return err;
 
-	dw24.field.ref1588_ck_div = e822_cgu_params[clk_freq].refclk_pre_div;
-	dw24.field.tspll_fbdiv_frac = e822_cgu_params[clk_freq].frac_n_div;
-	dw24.field.time_ref_sel = clk_src;
+	dw24.ref1588_ck_div = e822_cgu_params[clk_freq].refclk_pre_div;
+	dw24.tspll_fbdiv_frac = e822_cgu_params[clk_freq].frac_n_div;
+	dw24.time_ref_sel = clk_src;
 
 	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD24, dw24.val);
 	if (err)
 		return err;
 
 	/* Finally, enable the PLL */
-	dw24.field.ts_pll_enable = 1;
+	dw24.ts_pll_enable = 1;
 
 	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD24, dw24.val);
 	if (err)
@@ -464,17 +464,17 @@ static int ice_cfg_cgu_pll_e82x(struct ice_hw *hw,
 	if (err)
 		return err;
 
-	if (!bwm_lf.field.plllock_true_lock_cri) {
+	if (!bwm_lf.plllock_true_lock_cri) {
 		dev_warn(ice_hw_to_dev(hw), "CGU PLL failed to lock\n");
 		return -EBUSY;
 	}
 
 	/* Log the current clock configuration */
 	ice_debug(hw, ICE_DBG_PTP, "New CGU configuration -- %s, clk_src %s, clk_freq %s, PLL %s\n",
-		  dw24.field.ts_pll_enable ? "enabled" : "disabled",
-		  ice_clk_src_str(dw24.field.time_ref_sel),
-		  ice_clk_freq_str(dw9.field.time_ref_freq_sel),
-		  bwm_lf.field.plllock_true_lock_cri ? "locked" : "unlocked");
+		  dw24.ts_pll_enable ? "enabled" : "disabled",
+		  ice_clk_src_str(dw24.time_ref_sel),
+		  ice_clk_freq_str(dw9.time_ref_freq_sel),
+		  bwm_lf.plllock_true_lock_cri ? "locked" : "unlocked");
 
 	return 0;
 }
@@ -499,8 +499,8 @@ static int ice_init_cgu_e82x(struct ice_hw *hw)
 		return err;
 
 	/* Disable sticky lock detection so lock err reported is accurate */
-	cntr_bist.field.i_plllock_sel_0 = 0;
-	cntr_bist.field.i_plllock_sel_1 = 0;
+	cntr_bist.i_plllock_sel_0 = 0;
+	cntr_bist.i_plllock_sel_1 = 0;
 
 	err = ice_write_cgu_reg_e82x(hw, TSPLL_CNTR_BIST_SETTINGS,
 				     cntr_bist.val);
-- 
cgit v1.2.3-58-ga151


From 713dcad2a8c74de23d593a73bec4c22365ebc83f Mon Sep 17 00:00:00 2001
From: Michal Michalik <michal.michalik@intel.com>
Date: Tue, 28 May 2024 16:03:59 -0700
Subject: ice: Add support for E825-C TS PLL handling

The CGU layout of E825-C is a little different than E822/E823. Add
support the new hardware adding relevant functions.

Signed-off-by: Michal Michalik <michal.michalik@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528-next-2024-05-28-ptp-refactors-v1-9-c082739bb6f6@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_cgu_regs.h   |  65 ++++++
 drivers/net/ethernet/intel/ice/ice_common.c     |   9 +-
 drivers/net/ethernet/intel/ice/ice_ptp_consts.h |  87 ++++++++
 drivers/net/ethernet/intel/ice/ice_ptp_hw.c     | 261 ++++++++++++++++++++++--
 drivers/net/ethernet/intel/ice/ice_ptp_hw.h     |  24 ++-
 drivers/net/ethernet/intel/ice/ice_type.h       |   2 +-
 6 files changed, 429 insertions(+), 19 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_cgu_regs.h b/drivers/net/ethernet/intel/ice/ice_cgu_regs.h
index 36aeb10eefb7..10d9d74f3545 100644
--- a/drivers/net/ethernet/intel/ice/ice_cgu_regs.h
+++ b/drivers/net/ethernet/intel/ice/ice_cgu_regs.h
@@ -27,6 +27,17 @@ union nac_cgu_dword9 {
 	u32 val;
 };
 
+#define NAC_CGU_DWORD16_E825C 0x40
+union nac_cgu_dword16_e825c {
+	struct {
+		u32 synce_remndr : 6;
+		u32 synce_phlmt_en : 1;
+		u32 misc13 : 17;
+		u32 tspll_ck_refclkfreq : 8;
+	};
+	u32 val;
+};
+
 #define NAC_CGU_DWORD19 0x4c
 union nac_cgu_dword19 {
 	struct {
@@ -67,6 +78,22 @@ union nac_cgu_dword22 {
 	u32 val;
 };
 
+#define NAC_CGU_DWORD23_E825C 0x5C
+union nac_cgu_dword23_e825c {
+	struct {
+		u32 cgupll_fbdiv_intgr : 10;
+		u32 ux56pll_fbdiv_intgr : 10;
+		u32 misc20 : 4;
+		u32 ts_pll_enable : 1;
+		u32 time_sync_tspll_align_sel : 1;
+		u32 ext_synce_sel : 1;
+		u32 ref1588_ck_div : 4;
+		u32 time_ref_sel : 1;
+
+	};
+	u32 val;
+};
+
 #define NAC_CGU_DWORD24 0x60
 union nac_cgu_dword24 {
 	struct {
@@ -113,4 +140,42 @@ union tspll_ro_bwm_lf {
 	u32 val;
 };
 
+#define TSPLL_RO_LOCK_E825C 0x3f0
+union tspll_ro_lock_e825c {
+	struct {
+		u32 bw_freqov_high_cri_7_0 : 8;
+		u32 bw_freqov_high_cri_9_8 : 2;
+		u32 reserved455 : 1;
+		u32 plllock_gain_tran_cri : 1;
+		u32 plllock_true_lock_cri : 1;
+		u32 pllunlock_flag_cri : 1;
+		u32 afcerr_cri : 1;
+		u32 afcdone_cri : 1;
+		u32 feedfwrdgain_cal_cri_7_0 : 8;
+		u32 reserved462 : 8;
+	};
+	u32 val;
+};
+
+#define TSPLL_BW_TDC_E825C 0x31c
+union tspll_bw_tdc_e825c {
+	struct {
+		u32 i_tdc_offset_lock_1_0 : 2;
+		u32 i_bbthresh1_2_0 : 3;
+		u32 i_bbthresh2_2_0 : 3;
+		u32 i_tdcsel_1_0 : 2;
+		u32 i_tdcovccorr_en_h : 1;
+		u32 i_divretimeren : 1;
+		u32 i_bw_ampmeas_window : 1;
+		u32 i_bw_lowerbound_2_0 : 3;
+		u32 i_bw_upperbound_2_0 : 3;
+		u32 i_bw_mode_1_0 : 2;
+		u32 i_ft_mode_sel_2_0 : 3;
+		u32 i_bwphase_4_0 : 5;
+		u32 i_plllock_sel_1_0 : 2;
+		u32 i_afc_divratio : 1;
+	};
+	u32 val;
+};
+
 #endif /* _ICE_CGU_REGS_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index c37042d9bdd9..4dbd3fca5527 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -2314,8 +2314,13 @@ ice_parse_1588_func_caps(struct ice_hw *hw, struct ice_hw_func_caps *func_p,
 	info->tmr_index_owned = ((number & ICE_TS_TMR_IDX_OWND_M) != 0);
 	info->tmr_index_assoc = ((number & ICE_TS_TMR_IDX_ASSOC_M) != 0);
 
-	info->clk_freq = FIELD_GET(ICE_TS_CLK_FREQ_M, number);
-	info->clk_src = ((number & ICE_TS_CLK_SRC_M) != 0);
+	if (!ice_is_e825c(hw)) {
+		info->clk_freq = FIELD_GET(ICE_TS_CLK_FREQ_M, number);
+		info->clk_src = ((number & ICE_TS_CLK_SRC_M) != 0);
+	} else {
+		info->clk_freq = ICE_TIME_REF_FREQ_156_250;
+		info->clk_src = ICE_CLK_SRC_TCXO;
+	}
 
 	if (info->clk_freq < NUM_ICE_TIME_REF_FREQ) {
 		info->time_ref = (enum ice_time_ref_freq)info->clk_freq;
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_consts.h b/drivers/net/ethernet/intel/ice/ice_ptp_consts.h
index ef180936f60c..e6980b94a6c1 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_consts.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_consts.h
@@ -470,6 +470,93 @@ const struct ice_cgu_pll_params_e82x e822_cgu_params[NUM_ICE_TIME_REF_FREQ] = {
 	},
 };
 
+const
+struct ice_cgu_pll_params_e825c e825c_cgu_params[NUM_ICE_TIME_REF_FREQ] = {
+	/* ICE_TIME_REF_FREQ_25_000 -> 25 MHz */
+	{
+		/* tspll_ck_refclkfreq */
+		0x19,
+		/* tspll_ndivratio */
+		1,
+		/* tspll_fbdiv_intgr */
+		320,
+		/* tspll_fbdiv_frac */
+		0,
+		/* ref1588_ck_div */
+		0,
+	},
+
+	/* ICE_TIME_REF_FREQ_122_880 -> 122.88 MHz */
+	{
+		/* tspll_ck_refclkfreq */
+		0x29,
+		/* tspll_ndivratio */
+		3,
+		/* tspll_fbdiv_intgr */
+		195,
+		/* tspll_fbdiv_frac */
+		1342177280UL,
+		/* ref1588_ck_div */
+		0,
+	},
+
+	/* ICE_TIME_REF_FREQ_125_000 -> 125 MHz */
+	{
+		/* tspll_ck_refclkfreq */
+		0x3E,
+		/* tspll_ndivratio */
+		2,
+		/* tspll_fbdiv_intgr */
+		128,
+		/* tspll_fbdiv_frac */
+		0,
+		/* ref1588_ck_div */
+		0,
+	},
+
+	/* ICE_TIME_REF_FREQ_153_600 -> 153.6 MHz */
+	{
+		/* tspll_ck_refclkfreq */
+		0x33,
+		/* tspll_ndivratio */
+		3,
+		/* tspll_fbdiv_intgr */
+		156,
+		/* tspll_fbdiv_frac */
+		1073741824UL,
+		/* ref1588_ck_div */
+		0,
+	},
+
+	/* ICE_TIME_REF_FREQ_156_250 -> 156.25 MHz */
+	{
+		/* tspll_ck_refclkfreq */
+		0x1F,
+		/* tspll_ndivratio */
+		5,
+		/* tspll_fbdiv_intgr */
+		256,
+		/* tspll_fbdiv_frac */
+		0,
+		/* ref1588_ck_div */
+		0,
+	},
+
+	/* ICE_TIME_REF_FREQ_245_760 -> 245.76 MHz */
+	{
+		/* tspll_ck_refclkfreq */
+		0x52,
+		/* tspll_ndivratio */
+		3,
+		/* tspll_fbdiv_intgr */
+		97,
+		/* tspll_fbdiv_frac */
+		2818572288UL,
+		/* ref1588_ck_div */
+		0,
+	},
+};
+
 /* struct ice_vernier_info_e82x
  *
  * E822 hardware calibrates the delay of the timestamp indication from the
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index 691ea20811e7..2c41921f76e8 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -325,8 +325,8 @@ static const char *ice_clk_freq_str(enum ice_time_ref_freq clk_freq)
 static const char *ice_clk_src_str(enum ice_clk_src clk_src)
 {
 	switch (clk_src) {
-	case ICE_CLK_SRC_TCX0:
-		return "TCX0";
+	case ICE_CLK_SRC_TCXO:
+		return "TCXO";
 	case ICE_CLK_SRC_TIME_REF:
 		return "TIME_REF";
 	default:
@@ -338,7 +338,7 @@ static const char *ice_clk_src_str(enum ice_clk_src clk_src)
  * ice_cfg_cgu_pll_e82x - Configure the Clock Generation Unit
  * @hw: pointer to the HW struct
  * @clk_freq: Clock frequency to program
- * @clk_src: Clock source to select (TIME_REF, or TCX0)
+ * @clk_src: Clock source to select (TIME_REF, or TCXO)
  *
  * Configure the Clock Generation Unit with the desired clock frequency and
  * time reference, enabling the PLL which drives the PTP hardware clock.
@@ -372,10 +372,10 @@ static int ice_cfg_cgu_pll_e82x(struct ice_hw *hw,
 		return -EINVAL;
 	}
 
-	if (clk_src == ICE_CLK_SRC_TCX0 &&
+	if (clk_src == ICE_CLK_SRC_TCXO &&
 	    clk_freq != ICE_TIME_REF_FREQ_25_000) {
 		dev_warn(ice_hw_to_dev(hw),
-			 "TCX0 only supports 25 MHz frequency\n");
+			 "TCXO only supports 25 MHz frequency\n");
 		return -EINVAL;
 	}
 
@@ -480,16 +480,198 @@ static int ice_cfg_cgu_pll_e82x(struct ice_hw *hw,
 }
 
 /**
- * ice_init_cgu_e82x - Initialize CGU with settings from firmware
- * @hw: pointer to the HW structure
+ * ice_cfg_cgu_pll_e825c - Configure the Clock Generation Unit for E825-C
+ * @hw: pointer to the HW struct
+ * @clk_freq: Clock frequency to program
+ * @clk_src: Clock source to select (TIME_REF, or TCXO)
  *
- * Initialize the Clock Generation Unit of the E822 device.
+ * Configure the Clock Generation Unit with the desired clock frequency and
+ * time reference, enabling the PLL which drives the PTP hardware clock.
  *
- * Return: 0 on success, other error codes when failed to read/write/cfg CGU
+ * Return:
+ * * %0       - success
+ * * %-EINVAL - input parameters are incorrect
+ * * %-EBUSY  - failed to lock TS PLL
+ * * %other   - CGU read/write failure
  */
-static int ice_init_cgu_e82x(struct ice_hw *hw)
+static int ice_cfg_cgu_pll_e825c(struct ice_hw *hw,
+				 enum ice_time_ref_freq clk_freq,
+				 enum ice_clk_src clk_src)
+{
+	union tspll_ro_lock_e825c ro_lock;
+	union nac_cgu_dword16_e825c dw16;
+	union nac_cgu_dword23_e825c dw23;
+	union nac_cgu_dword19 dw19;
+	union nac_cgu_dword22 dw22;
+	union nac_cgu_dword24 dw24;
+	union nac_cgu_dword9 dw9;
+	int err;
+
+	if (clk_freq >= NUM_ICE_TIME_REF_FREQ) {
+		dev_warn(ice_hw_to_dev(hw), "Invalid TIME_REF frequency %u\n",
+			 clk_freq);
+		return -EINVAL;
+	}
+
+	if (clk_src >= NUM_ICE_CLK_SRC) {
+		dev_warn(ice_hw_to_dev(hw), "Invalid clock source %u\n",
+			 clk_src);
+		return -EINVAL;
+	}
+
+	if (clk_src == ICE_CLK_SRC_TCXO &&
+	    clk_freq != ICE_TIME_REF_FREQ_156_250) {
+		dev_warn(ice_hw_to_dev(hw),
+			 "TCXO only supports 156.25 MHz frequency\n");
+		return -EINVAL;
+	}
+
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD9, &dw9.val);
+	if (err)
+		return err;
+
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD24, &dw24.val);
+	if (err)
+		return err;
+
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD16_E825C, &dw16.val);
+	if (err)
+		return err;
+
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD23_E825C, &dw23.val);
+	if (err)
+		return err;
+
+	err = ice_read_cgu_reg_e82x(hw, TSPLL_RO_LOCK_E825C, &ro_lock.val);
+	if (err)
+		return err;
+
+	/* Log the current clock configuration */
+	ice_debug(hw, ICE_DBG_PTP, "Current CGU configuration -- %s, clk_src %s, clk_freq %s, PLL %s\n",
+		  dw24.ts_pll_enable ? "enabled" : "disabled",
+		  ice_clk_src_str(dw23.time_ref_sel),
+		  ice_clk_freq_str(dw9.time_ref_freq_sel),
+		  ro_lock.plllock_true_lock_cri ? "locked" : "unlocked");
+
+	/* Disable the PLL before changing the clock source or frequency */
+	if (dw23.ts_pll_enable) {
+		dw23.ts_pll_enable = 0;
+
+		err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD23_E825C,
+					     dw23.val);
+		if (err)
+			return err;
+	}
+
+	/* Set the frequency */
+	dw9.time_ref_freq_sel = clk_freq;
+
+	/* Enable the correct receiver */
+	if (clk_src == ICE_CLK_SRC_TCXO) {
+		dw9.time_ref_en = 0;
+		dw9.clk_eref0_en = 1;
+	} else {
+		dw9.time_ref_en = 1;
+		dw9.clk_eref0_en = 0;
+	}
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD9, dw9.val);
+	if (err)
+		return err;
+
+	/* Choose the referenced frequency */
+	dw16.tspll_ck_refclkfreq =
+	e825c_cgu_params[clk_freq].tspll_ck_refclkfreq;
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD16_E825C, dw16.val);
+	if (err)
+		return err;
+
+	/* Configure the TS PLL feedback divisor */
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD19, &dw19.val);
+	if (err)
+		return err;
+
+	dw19.tspll_fbdiv_intgr =
+		e825c_cgu_params[clk_freq].tspll_fbdiv_intgr;
+	dw19.tspll_ndivratio =
+		e825c_cgu_params[clk_freq].tspll_ndivratio;
+
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD19, dw19.val);
+	if (err)
+		return err;
+
+	/* Configure the TS PLL post divisor */
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD22, &dw22.val);
+	if (err)
+		return err;
+
+	/* These two are constant for E825C */
+	dw22.time1588clk_div = 5;
+	dw22.time1588clk_sel_div2 = 0;
+
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD22, dw22.val);
+	if (err)
+		return err;
+
+	/* Configure the TS PLL pre divisor and clock source */
+	err = ice_read_cgu_reg_e82x(hw, NAC_CGU_DWORD23_E825C, &dw23.val);
+	if (err)
+		return err;
+
+	dw23.ref1588_ck_div =
+		e825c_cgu_params[clk_freq].ref1588_ck_div;
+	dw23.time_ref_sel = clk_src;
+
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD23_E825C, dw23.val);
+	if (err)
+		return err;
+
+	dw24.tspll_fbdiv_frac =
+		e825c_cgu_params[clk_freq].tspll_fbdiv_frac;
+
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD24, dw24.val);
+	if (err)
+		return err;
+
+	/* Finally, enable the PLL */
+	dw23.ts_pll_enable = 1;
+
+	err = ice_write_cgu_reg_e82x(hw, NAC_CGU_DWORD23_E825C, dw23.val);
+	if (err)
+		return err;
+
+	/* Wait to verify if the PLL locks */
+	usleep_range(1000, 5000);
+
+	err = ice_read_cgu_reg_e82x(hw, TSPLL_RO_LOCK_E825C, &ro_lock.val);
+	if (err)
+		return err;
+
+	if (!ro_lock.plllock_true_lock_cri) {
+		dev_warn(ice_hw_to_dev(hw), "CGU PLL failed to lock\n");
+		return -EBUSY;
+	}
+
+	/* Log the current clock configuration */
+	ice_debug(hw, ICE_DBG_PTP, "New CGU configuration -- %s, clk_src %s, clk_freq %s, PLL %s\n",
+		  dw24.ts_pll_enable ? "enabled" : "disabled",
+		  ice_clk_src_str(dw23.time_ref_sel),
+		  ice_clk_freq_str(dw9.time_ref_freq_sel),
+		  ro_lock.plllock_true_lock_cri ? "locked" : "unlocked");
+
+	return 0;
+}
+
+/**
+ * ice_cfg_cgu_pll_dis_sticky_bits_e82x - disable TS PLL sticky bits
+ * @hw: pointer to the HW struct
+ *
+ * Configure the Clock Generation Unit TS PLL sticky bits so they don't latch on
+ * losing TS PLL lock, but always show current state.
+ *
+ * Return: 0 on success, other error codes when failed to read/write CGU
+ */
+static int ice_cfg_cgu_pll_dis_sticky_bits_e82x(struct ice_hw *hw)
 {
-	struct ice_ts_func_info *ts_info = &hw->func_caps.ts_func_info;
 	union tspll_cntr_bist_settings cntr_bist;
 	int err;
 
@@ -502,16 +684,65 @@ static int ice_init_cgu_e82x(struct ice_hw *hw)
 	cntr_bist.i_plllock_sel_0 = 0;
 	cntr_bist.i_plllock_sel_1 = 0;
 
-	err = ice_write_cgu_reg_e82x(hw, TSPLL_CNTR_BIST_SETTINGS,
-				     cntr_bist.val);
+	return ice_write_cgu_reg_e82x(hw, TSPLL_CNTR_BIST_SETTINGS,
+				      cntr_bist.val);
+}
+
+/**
+ * ice_cfg_cgu_pll_dis_sticky_bits_e825c - disable TS PLL sticky bits for E825-C
+ * @hw: pointer to the HW struct
+ *
+ * Configure the Clock Generation Unit TS PLL sticky bits so they don't latch on
+ * losing TS PLL lock, but always show current state.
+ *
+ * Return: 0 on success, other error codes when failed to read/write CGU
+ */
+static int ice_cfg_cgu_pll_dis_sticky_bits_e825c(struct ice_hw *hw)
+{
+	union tspll_bw_tdc_e825c bw_tdc;
+	int err;
+
+	err = ice_read_cgu_reg_e82x(hw, TSPLL_BW_TDC_E825C, &bw_tdc.val);
+	if (err)
+		return err;
+
+	bw_tdc.i_plllock_sel_1_0 = 0;
+
+	return ice_write_cgu_reg_e82x(hw, TSPLL_BW_TDC_E825C, bw_tdc.val);
+}
+
+/**
+ * ice_init_cgu_e82x - Initialize CGU with settings from firmware
+ * @hw: pointer to the HW structure
+ *
+ * Initialize the Clock Generation Unit of the E822 device.
+ *
+ * Return: 0 on success, other error codes when failed to read/write/cfg CGU
+ */
+static int ice_init_cgu_e82x(struct ice_hw *hw)
+{
+	struct ice_ts_func_info *ts_info = &hw->func_caps.ts_func_info;
+	int err;
+
+	/* Disable sticky lock detection so lock err reported is accurate */
+	if (ice_is_e825c(hw))
+		err = ice_cfg_cgu_pll_dis_sticky_bits_e825c(hw);
+	else
+		err = ice_cfg_cgu_pll_dis_sticky_bits_e82x(hw);
 	if (err)
 		return err;
 
 	/* Configure the CGU PLL using the parameters from the function
 	 * capabilities.
 	 */
-	return ice_cfg_cgu_pll_e82x(hw, ts_info->time_ref,
-				   (enum ice_clk_src)ts_info->clk_src);
+	if (ice_is_e825c(hw))
+		err = ice_cfg_cgu_pll_e825c(hw, ts_info->time_ref,
+					    (enum ice_clk_src)ts_info->clk_src);
+	else
+		err = ice_cfg_cgu_pll_e82x(hw, ts_info->time_ref,
+					   (enum ice_clk_src)ts_info->clk_src);
+
+	return err;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
index 184573509e62..0852a34ade91 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.h
@@ -197,7 +197,7 @@ extern
 const struct ice_eth56g_mac_reg_cfg eth56g_mac_cfg[NUM_ICE_ETH56G_LNK_SPD];
 
 /**
- * struct ice_cgu_pll_params_e82x
+ * struct ice_cgu_pll_params_e82x - E82X CGU parameters
  * @refclk_pre_div: Reference clock pre-divisor
  * @feedback_div: Feedback divisor
  * @frac_n_div: Fractional divisor
@@ -287,6 +287,28 @@ struct ice_cgu_pin_desc {
 extern const struct
 ice_cgu_pll_params_e82x e822_cgu_params[NUM_ICE_TIME_REF_FREQ];
 
+/**
+ * struct ice_cgu_pll_params_e825c - E825C CGU parameters
+ * @tspll_ck_refclkfreq: tspll_ck_refclkfreq selection
+ * @tspll_ndivratio: ndiv ratio that goes directly to the pll
+ * @tspll_fbdiv_intgr: TS PLL integer feedback divide
+ * @tspll_fbdiv_frac:  TS PLL fractional feedback divide
+ * @ref1588_ck_div: clock divider for tspll ref
+ *
+ * Clock Generation Unit parameters used to program the PLL based on the
+ * selected TIME_REF/TCXO frequency.
+ */
+struct ice_cgu_pll_params_e825c {
+	u32 tspll_ck_refclkfreq;
+	u32 tspll_ndivratio;
+	u32 tspll_fbdiv_intgr;
+	u32 tspll_fbdiv_frac;
+	u32 ref1588_ck_div;
+};
+
+extern const struct
+ice_cgu_pll_params_e825c e825c_cgu_params[NUM_ICE_TIME_REF_FREQ];
+
 #define E810C_QSFP_C827_0_HANDLE 2
 #define E810C_QSFP_C827_1_HANDLE 3
 
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index f039fc665455..39d8fc08d651 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -329,7 +329,7 @@ enum ice_time_ref_freq {
 
 /* Clock source specification */
 enum ice_clk_src {
-	ICE_CLK_SRC_TCX0	= 0, /* Temperature compensated oscillator  */
+	ICE_CLK_SRC_TCXO	= 0, /* Temperature compensated oscillator */
 	ICE_CLK_SRC_TIME_REF	= 1, /* Use TIME_REF reference clock */
 
 	NUM_ICE_CLK_SRC
-- 
cgit v1.2.3-58-ga151


From 5f847eede63889cbada9a23ea7c8f68659e3e918 Mon Sep 17 00:00:00 2001
From: Grzegorz Nitka <grzegorz.nitka@intel.com>
Date: Tue, 28 May 2024 16:04:00 -0700
Subject: ice: Add NAC Topology device capability parser

Add new device capability ICE_AQC_CAPS_NAC_TOPOLOGY which allows to
determine the mode of operation (1 or 2 NAC).
Define a new structure to store data from new capability and
corresponding parser code.

Co-developed-by: Prathisna Padmasanan <prathisna.padmasanan@intel.com>
Signed-off-by: Prathisna Padmasanan <prathisna.padmasanan@intel.com>
Signed-off-by: Grzegorz Nitka <grzegorz.nitka@intel.com>
Reviewed-by: Pawel Kaminski <pawel.kaminski@intel.com>
Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528-next-2024-05-28-ptp-refactors-v1-10-c082739bb6f6@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_adminq_cmd.h |  1 +
 drivers/net/ethernet/intel/ice/ice_common.c     | 31 +++++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_type.h       | 10 ++++++++
 3 files changed, 42 insertions(+)

diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index e76c388b9905..621a2ca7093e 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -122,6 +122,7 @@ struct ice_aqc_list_caps_elem {
 #define ICE_AQC_CAPS_POST_UPDATE_RESET_RESTRICT		0x0077
 #define ICE_AQC_CAPS_NVM_MGMT				0x0080
 #define ICE_AQC_CAPS_TX_SCHED_TOPO_COMP_MODE		0x0085
+#define ICE_AQC_CAPS_NAC_TOPOLOGY			0x0087
 #define ICE_AQC_CAPS_FW_LAG_SUPPORT			0x0092
 #define ICE_AQC_BIT_ROCEV2_LAG				0x01
 #define ICE_AQC_BIT_SRIOV_LAG				0x02
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 4dbd3fca5527..9ae61cd8923e 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -2593,6 +2593,34 @@ ice_parse_sensor_reading_cap(struct ice_hw *hw, struct ice_hw_dev_caps *dev_p,
 		  dev_p->supported_sensors);
 }
 
+/**
+ * ice_parse_nac_topo_dev_caps - Parse ICE_AQC_CAPS_NAC_TOPOLOGY cap
+ * @hw: pointer to the HW struct
+ * @dev_p: pointer to device capabilities structure
+ * @cap: capability element to parse
+ *
+ * Parse ICE_AQC_CAPS_NAC_TOPOLOGY for device capabilities.
+ */
+static void ice_parse_nac_topo_dev_caps(struct ice_hw *hw,
+					struct ice_hw_dev_caps *dev_p,
+					struct ice_aqc_list_caps_elem *cap)
+{
+	dev_p->nac_topo.mode = le32_to_cpu(cap->number);
+	dev_p->nac_topo.id = le32_to_cpu(cap->phys_id) & ICE_NAC_TOPO_ID_M;
+
+	dev_info(ice_hw_to_dev(hw),
+		 "PF is configured in %s mode with IP instance ID %d\n",
+		 (dev_p->nac_topo.mode & ICE_NAC_TOPO_PRIMARY_M) ?
+		 "primary" : "secondary", dev_p->nac_topo.id);
+
+	ice_debug(hw, ICE_DBG_INIT, "dev caps: nac topology is_primary = %d\n",
+		  !!(dev_p->nac_topo.mode & ICE_NAC_TOPO_PRIMARY_M));
+	ice_debug(hw, ICE_DBG_INIT, "dev caps: nac topology is_dual = %d\n",
+		  !!(dev_p->nac_topo.mode & ICE_NAC_TOPO_DUAL_M));
+	ice_debug(hw, ICE_DBG_INIT, "dev caps: nac topology id = %d\n",
+		  dev_p->nac_topo.id);
+}
+
 /**
  * ice_parse_dev_caps - Parse device capabilities
  * @hw: pointer to the HW struct
@@ -2644,6 +2672,9 @@ ice_parse_dev_caps(struct ice_hw *hw, struct ice_hw_dev_caps *dev_p,
 		case ICE_AQC_CAPS_SENSOR_READING:
 			ice_parse_sensor_reading_cap(hw, dev_p, &cap_resp[i]);
 			break;
+		case ICE_AQC_CAPS_NAC_TOPOLOGY:
+			ice_parse_nac_topo_dev_caps(hw, dev_p, &cap_resp[i]);
+			break;
 		default:
 			/* Don't list common capabilities as unknown */
 			if (!found)
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index 39d8fc08d651..841860784e3c 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -374,6 +374,15 @@ struct ice_ts_dev_info {
 	u8 ts_ll_int_read;
 };
 
+#define ICE_NAC_TOPO_PRIMARY_M	BIT(0)
+#define ICE_NAC_TOPO_DUAL_M	BIT(1)
+#define ICE_NAC_TOPO_ID_M	GENMASK(0xF, 0)
+
+struct ice_nac_topology {
+	u32 mode;
+	u8 id;
+};
+
 /* Function specific capabilities */
 struct ice_hw_func_caps {
 	struct ice_hw_common_caps common_cap;
@@ -395,6 +404,7 @@ struct ice_hw_dev_caps {
 	u32 num_flow_director_fltr;	/* Number of FD filters available */
 	struct ice_ts_dev_info ts_dev_info;
 	u32 num_funcs;
+	struct ice_nac_topology nac_topo;
 	/* bitmap of supported sensors
 	 * bit 0 - internal temperature sensor
 	 * bit 31:1 - Reserved
-- 
cgit v1.2.3-58-ga151


From 4409ea1726cb9ce2769808873f8fbe33378c201b Mon Sep 17 00:00:00 2001
From: Grzegorz Nitka <grzegorz.nitka@intel.com>
Date: Tue, 28 May 2024 16:04:01 -0700
Subject: ice: Adjust PTP init for 2x50G E825C devices

>From FW/HW perspective, 2 port topology in E825C devices requires
merging of 2 port mapping internally and breakout mapping externally.
As a consequence, it requires different port numbering from PTP code
perspective.
For that topology, pf_id can not be used to index PTP ports. Even if
the 2nd port is identified as port with pf_id = 1, all PHY operations
need to be performed as it was port 2. Thus, special mapping is needed
for the 2nd port.
This change adds detection of 2x50G topology and applies 'custom'
mapping on the 2nd port.

Signed-off-by: Grzegorz Nitka <grzegorz.nitka@intel.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240528-next-2024-05-28-ptp-refactors-v1-11-c082739bb6f6@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_hw_autogen.h |  4 ++++
 drivers/net/ethernet/intel/ice/ice_ptp.c        |  5 +++++
 drivers/net/ethernet/intel/ice/ice_ptp_hw.c     | 22 ++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_type.h       |  9 +++++++++
 4 files changed, 40 insertions(+)

diff --git a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
index cfac1d432c15..91cbae1eec89 100644
--- a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
+++ b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
@@ -157,6 +157,8 @@
 #define GLGEN_RTRIG_CORER_M			BIT(0)
 #define GLGEN_RTRIG_GLOBR_M			BIT(1)
 #define GLGEN_STAT				0x000B612C
+#define GLGEN_SWITCH_MODE_CONFIG		0x000B81E0
+#define GLGEN_SWITCH_MODE_CONFIG_25X4_QUAD_M	BIT(2)
 #define GLGEN_VFLRSTAT(_i)			(0x00093A04 + ((_i) * 4))
 #define PFGEN_CTRL				0x00091000
 #define PFGEN_CTRL_PFSWR_M			BIT(0)
@@ -177,6 +179,8 @@
 #define GLINT_CTL_ITR_GRAN_50_M			ICE_M(0xF, 24)
 #define GLINT_CTL_ITR_GRAN_25_S			28
 #define GLINT_CTL_ITR_GRAN_25_M			ICE_M(0xF, 28)
+#define GLGEN_MAC_LINK_TOPO			0x000B81DC
+#define GLGEN_MAC_LINK_TOPO_LINK_TOPO_M		GENMASK(1, 0)
 #define GLINT_DYN_CTL(_INT)			(0x00160000 + ((_INT) * 4))
 #define GLINT_DYN_CTL_INTENA_M			BIT(0)
 #define GLINT_DYN_CTL_CLEARPBA_M		BIT(1)
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index 3af0f4a2c3be..adbb9cffe20c 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -1469,6 +1469,8 @@ void ice_ptp_link_change(struct ice_pf *pf, u8 port, bool linkup)
 		return;
 
 	ptp_port = &pf->ptp.port;
+	if (ice_is_e825c(hw) && hw->ptp.is_2x50g_muxed_topo)
+		port *= 2;
 	if (WARN_ON_ONCE(ptp_port->port_num != port))
 		return;
 
@@ -3282,6 +3284,9 @@ void ice_ptp_init(struct ice_pf *pf)
 	}
 
 	ptp->port.port_num = hw->pf_id;
+	if (ice_is_e825c(hw) && hw->ptp.is_2x50g_muxed_topo)
+		ptp->port.port_num = hw->pf_id * 2;
+
 	err = ice_ptp_init_port(pf, &ptp->port);
 	if (err)
 		goto err;
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index 2c41921f76e8..1e9a4ccd0ea2 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -2644,6 +2644,26 @@ static int ice_get_phy_tx_tstamp_ready_eth56g(struct ice_hw *hw, u8 port,
 	return 0;
 }
 
+/**
+ * ice_is_muxed_topo - detect breakout 2x50G topology for E825C
+ * @hw: pointer to the HW struct
+ *
+ * Return: true if it's 2x50 breakout topology, false otherwise
+ */
+static bool ice_is_muxed_topo(struct ice_hw *hw)
+{
+	u8 link_topo;
+	bool mux;
+	u32 val;
+
+	val = rd32(hw, GLGEN_SWITCH_MODE_CONFIG);
+	mux = FIELD_GET(GLGEN_SWITCH_MODE_CONFIG_25X4_QUAD_M, val);
+	val = rd32(hw, GLGEN_MAC_LINK_TOPO);
+	link_topo = FIELD_GET(GLGEN_MAC_LINK_TOPO_LINK_TOPO_M, val);
+
+	return (mux && link_topo == ICE_LINK_TOPO_UP_TO_2_LINKS);
+}
+
 /**
  * ice_ptp_init_phy_e825c - initialize PHY parameters
  * @hw: pointer to the HW struct
@@ -2676,6 +2696,8 @@ static void ice_ptp_init_phy_e825c(struct ice_hw *hw)
 			return;
 		}
 	}
+
+	ptp->is_2x50g_muxed_topo = ice_is_muxed_topo(hw);
 }
 
 /* E822 family functions
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index 841860784e3c..5f0da6850b03 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -853,11 +853,20 @@ enum ice_phy_model {
 	ICE_PHY_ETH56G,
 };
 
+/* Global Link Topology */
+enum ice_global_link_topo {
+	ICE_LINK_TOPO_UP_TO_2_LINKS,
+	ICE_LINK_TOPO_UP_TO_4_LINKS,
+	ICE_LINK_TOPO_UP_TO_8_LINKS,
+	ICE_LINK_TOPO_RESERVED,
+};
+
 struct ice_ptp_hw {
 	enum ice_phy_model phy_model;
 	union ice_phy_params phy;
 	u8 num_lports;
 	u8 ports_per_phy;
+	bool is_2x50g_muxed_topo;
 };
 
 /* Port hardware description */
-- 
cgit v1.2.3-58-ga151


From 165f87691a898aea113923224c269bae14749ebc Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadfed@meta.com>
Date: Thu, 30 May 2024 13:47:51 -0700
Subject: bnxt_en: add timestamping statistics support

The ethtool_ts_stats structure was introduced earlier this year. Now
it's time to support this group of counters in more drivers.
This patch adds support to bnxt driver.

Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Link: https://lore.kernel.org/r/20240530204751.99636-1-vadfed@meta.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         | 18 +++++++++++++-----
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 14 ++++++++++++++
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c     |  8 ++++++++
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h     |  8 ++++++++
 4 files changed, 43 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index c437ca1c0fd3..6d9faa78e391 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -512,8 +512,11 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) {
 		struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
 
-		if (ptp && ptp->tx_tstamp_en && !skb_is_gso(skb) &&
-		    atomic_dec_if_positive(&ptp->tx_avail) >= 0) {
+		if (ptp && ptp->tx_tstamp_en && !skb_is_gso(skb)) {
+			if (!atomic_dec_if_positive(&ptp->tx_avail)) {
+				atomic64_inc(&ptp->stats.ts_err);
+				goto tx_no_ts;
+			}
 			if (!bnxt_ptp_parse(skb, &ptp->tx_seqid,
 					    &ptp->tx_hdr_off)) {
 				if (vlan_tag_flags)
@@ -526,6 +529,7 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		}
 	}
 
+tx_no_ts:
 	if (unlikely(skb->no_fcs))
 		lflags |= cpu_to_le32(TX_BD_FLAGS_NO_CRC);
 
@@ -732,8 +736,10 @@ tx_done:
 	return NETDEV_TX_OK;
 
 tx_dma_error:
-	if (BNXT_TX_PTP_IS_SET(lflags))
+	if (BNXT_TX_PTP_IS_SET(lflags)) {
+		atomic64_inc(&bp->ptp_cfg->stats.ts_err);
 		atomic_inc(&bp->ptp_cfg->tx_avail);
+	}
 
 	last_frag = i;
 
@@ -812,10 +818,12 @@ static void __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
 		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) {
 			if (BNXT_CHIP_P5(bp)) {
 				/* PTP worker takes ownership of the skb */
-				if (!bnxt_get_tx_ts_p5(bp, skb))
+				if (!bnxt_get_tx_ts_p5(bp, skb)) {
 					skb = NULL;
-				else
+				} else {
+					atomic64_inc(&bp->ptp_cfg->stats.ts_err);
 					atomic_inc(&bp->ptp_cfg->tx_avail);
+				}
 			}
 		}
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 8763f8a01457..bf157f6cc042 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -5233,6 +5233,19 @@ static void bnxt_get_rmon_stats(struct net_device *dev,
 	*ranges = bnxt_rmon_ranges;
 }
 
+static void bnxt_get_ptp_stats(struct net_device *dev,
+			       struct ethtool_ts_stats *ts_stats)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+
+	if (ptp) {
+		ts_stats->pkts = ptp->stats.ts_pkts;
+		ts_stats->lost = ptp->stats.ts_lost;
+		ts_stats->err = atomic64_read(&ptp->stats.ts_err);
+	}
+}
+
 static void bnxt_get_link_ext_stats(struct net_device *dev,
 				    struct ethtool_link_ext_stats *stats)
 {
@@ -5316,4 +5329,5 @@ const struct ethtool_ops bnxt_ethtool_ops = {
 	.get_eth_mac_stats	= bnxt_get_eth_mac_stats,
 	.get_eth_ctrl_stats	= bnxt_get_eth_ctrl_stats,
 	.get_rmon_stats		= bnxt_get_rmon_stats,
+	.get_ts_stats		= bnxt_get_ptp_stats,
 };
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
index e661ab154d6b..a14d46b9bfdf 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
@@ -696,11 +696,13 @@ static void bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
 		spin_unlock_bh(&ptp->ptp_lock);
 		timestamp.hwtstamp = ns_to_ktime(ns);
 		skb_tstamp_tx(ptp->tx_skb, &timestamp);
+		ptp->stats.ts_pkts++;
 	} else {
 		if (!time_after_eq(jiffies, ptp->abs_txts_tmo)) {
 			ptp->txts_pending = true;
 			return;
 		}
+		ptp->stats.ts_lost++;
 		netdev_warn_once(bp->dev,
 				 "TS query for TX timer failed rc = %x\n", rc);
 	}
@@ -979,6 +981,11 @@ int bnxt_ptp_init(struct bnxt *bp, bool phc_cfg)
 		rc = err;
 		goto out;
 	}
+
+	ptp->stats.ts_pkts = 0;
+	ptp->stats.ts_lost = 0;
+	atomic64_set(&ptp->stats.ts_err, 0);
+
 	if (BNXT_CHIP_P5(bp)) {
 		spin_lock_bh(&ptp->ptp_lock);
 		bnxt_refclk_read(bp, NULL, &ptp->current_time);
@@ -1013,5 +1020,6 @@ void bnxt_ptp_clear(struct bnxt *bp)
 		dev_kfree_skb_any(ptp->tx_skb);
 		ptp->tx_skb = NULL;
 	}
+
 	bnxt_unmap_ptp_regs(bp);
 }
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
index 2c3415c8fc03..8c30b428a428 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
@@ -79,6 +79,12 @@ struct bnxt_pps {
 	struct pps_pin pins[BNXT_MAX_TSIO_PINS];
 };
 
+struct bnxt_ptp_stats {
+	u64		ts_pkts;
+	u64		ts_lost;
+	atomic64_t	ts_err;
+};
+
 struct bnxt_ptp_cfg {
 	struct ptp_clock_info	ptp_info;
 	struct ptp_clock	*ptp_clock;
@@ -125,6 +131,8 @@ struct bnxt_ptp_cfg {
 	u32			refclk_mapped_regs[2];
 	u32			txts_tmo;
 	unsigned long		abs_txts_tmo;
+
+	struct bnxt_ptp_stats	stats;
 };
 
 #if BITS_PER_LONG == 32
-- 
cgit v1.2.3-58-ga151


From 19249c0724f2048ab68179eac69004947b07d431 Mon Sep 17 00:00:00 2001
From: Matteo Croce <teknoraver@meta.com>
Date: Fri, 31 May 2024 01:27:21 +0200
Subject: net: make net.core.{r,w}mem_{default,max} namespaced

The following sysctl are global and can't be read from a netns:

net.core.rmem_default
net.core.rmem_max
net.core.wmem_default
net.core.wmem_max

Make the following sysctl parameters available readonly from within a
network namespace, allowing a container to read them.

Signed-off-by: Matteo Croce <teknoraver@meta.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Link: https://lore.kernel.org/r/20240530232722.45255-2-technoboy85@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/sysctl_net_core.c | 75 ++++++++++++++++++++++++++--------------------
 1 file changed, 42 insertions(+), 33 deletions(-)

diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index c9fb9ad87485..2079000691e2 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -382,38 +382,6 @@ proc_dolongvec_minmax_bpf_restricted(struct ctl_table *table, int write,
 #endif
 
 static struct ctl_table net_core_table[] = {
-	{
-		.procname	= "wmem_max",
-		.data		= &sysctl_wmem_max,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &min_sndbuf,
-	},
-	{
-		.procname	= "rmem_max",
-		.data		= &sysctl_rmem_max,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &min_rcvbuf,
-	},
-	{
-		.procname	= "wmem_default",
-		.data		= &sysctl_wmem_default,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &min_sndbuf,
-	},
-	{
-		.procname	= "rmem_default",
-		.data		= &sysctl_rmem_default,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &min_rcvbuf,
-	},
 	{
 		.procname	= "mem_pcpu_rsv",
 		.data		= &net_hotdata.sysctl_mem_pcpu_rsv,
@@ -697,6 +665,41 @@ static struct ctl_table netns_core_table[] = {
 		.extra2		= SYSCTL_ONE,
 		.proc_handler	= proc_dou8vec_minmax,
 	},
+	/* sysctl_core_net_init() will set the values after this
+	 * to readonly in network namespaces
+	 */
+	{
+		.procname	= "wmem_max",
+		.data		= &sysctl_wmem_max,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_sndbuf,
+	},
+	{
+		.procname	= "rmem_max",
+		.data		= &sysctl_rmem_max,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_rcvbuf,
+	},
+	{
+		.procname	= "wmem_default",
+		.data		= &sysctl_wmem_default,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_sndbuf,
+	},
+	{
+		.procname	= "rmem_default",
+		.data		= &sysctl_rmem_default,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_rcvbuf,
+	},
 };
 
 static int __init fb_tunnels_only_for_init_net_sysctl_setup(char *str)
@@ -724,8 +727,14 @@ static __net_init int sysctl_core_net_init(struct net *net)
 		if (tbl == NULL)
 			goto err_dup;
 
-		for (i = 0; i < table_size; ++i)
+		for (i = 0; i < table_size; ++i) {
+			if (tbl[i].data == &sysctl_wmem_max)
+				break;
+
 			tbl[i].data += (char *)net - (char *)&init_net;
+		}
+		for (; i < table_size; ++i)
+			tbl[i].mode &= ~0222;
 	}
 
 	net->core.sysctl_hdr = register_net_sysctl_sz(net, "net/core", tbl, table_size);
-- 
cgit v1.2.3-58-ga151


From 5b5233fb81bfecbfb7502178a9cf6790dde04a2c Mon Sep 17 00:00:00 2001
From: Matteo Croce <teknoraver@meta.com>
Date: Fri, 31 May 2024 01:27:22 +0200
Subject: selftests: net: tests net.core.{r,w}mem_{default,max} sysctls in a
 netns

Add a selftest which checks that the sysctl is present in a netns,
that the value is read from the init one, and that it's readonly.

Signed-off-by: Matteo Croce <teknoraver@meta.com>
Link: https://lore.kernel.org/r/20240530232722.45255-3-technoboy85@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/Makefile        |  1 +
 tools/testing/selftests/net/netns-sysctl.sh | 40 +++++++++++++++++++++++++++++
 2 files changed, 41 insertions(+)
 create mode 100755 tools/testing/selftests/net/netns-sysctl.sh

diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index bd01e4a0be2c..6da63d1831c1 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -53,6 +53,7 @@ TEST_PROGS += bind_bhash.sh
 TEST_PROGS += ip_local_port_range.sh
 TEST_PROGS += rps_default_mask.sh
 TEST_PROGS += big_tcp.sh
+TEST_PROGS += netns-sysctl.sh
 TEST_PROGS_EXTENDED := toeplitz_client.sh toeplitz.sh
 TEST_GEN_FILES =  socket nettest
 TEST_GEN_FILES += psock_fanout psock_tpacket msg_zerocopy reuseport_addr_any
diff --git a/tools/testing/selftests/net/netns-sysctl.sh b/tools/testing/selftests/net/netns-sysctl.sh
new file mode 100755
index 000000000000..45c34a3b9aae
--- /dev/null
+++ b/tools/testing/selftests/net/netns-sysctl.sh
@@ -0,0 +1,40 @@
+#!/bin/bash -e
+# SPDX-License-Identifier: GPL-2.0
+#
+# This test checks that the network buffer sysctls are present
+# in a network namespaces, and that they are readonly.
+
+source lib.sh
+
+cleanup() {
+    cleanup_ns $test_ns
+}
+
+trap cleanup EXIT
+
+fail() {
+	echo "ERROR: $*" >&2
+	exit 1
+}
+
+setup_ns test_ns
+
+for sc in {r,w}mem_{default,max}; do
+	# check that this is writable in a netns
+	[ -w "/proc/sys/net/core/$sc" ] ||
+		fail "$sc isn't writable in the init netns!"
+
+	# change the value in the host netns
+	sysctl -qw "net.core.$sc=300000" ||
+		fail "Can't write $sc in init netns!"
+
+	# check that the value is read from the init netns
+	[ "$(ip netns exec $test_ns sysctl -n "net.core.$sc")" -eq 300000 ] ||
+		fail "Value for $sc mismatch!"
+
+	# check that this isn't writable in a netns
+	ip netns exec $test_ns [ -w "/proc/sys/net/core/$sc" ] &&
+		fail "$sc is writable in a netns!"
+done
+
+echo 'Test passed OK'
-- 
cgit v1.2.3-58-ga151


From a967d3cee86e8e8eb15c9e12f818c59b36416001 Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@linaro.org>
Date: Wed, 29 May 2024 16:00:00 +0200
Subject: net: ethernet: cortina: Rename adjust link callback

The callback passed to of_phy_get_and_connect() in the
Cortina Gemini driver is called "gmac_speed_set" which is
archaic, rename it to "gmac_adjust_link" following the
pattern of most other drivers.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20240529-gemini-phylib-fixes-v4-1-16487ca4c2fe@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cortina/gemini.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c
index 7ebd61a3a49b..8de68a257611 100644
--- a/drivers/net/ethernet/cortina/gemini.c
+++ b/drivers/net/ethernet/cortina/gemini.c
@@ -288,7 +288,7 @@ static void gmac_set_flow_control(struct net_device *netdev, bool tx, bool rx)
 	spin_unlock_irqrestore(&port->config_lock, flags);
 }
 
-static void gmac_speed_set(struct net_device *netdev)
+static void gmac_adjust_link(struct net_device *netdev)
 {
 	struct gemini_ethernet_port *port = netdev_priv(netdev);
 	struct phy_device *phydev = netdev->phydev;
@@ -367,7 +367,7 @@ static int gmac_setup_phy(struct net_device *netdev)
 
 	phy = of_phy_get_and_connect(netdev,
 				     dev->of_node,
-				     gmac_speed_set);
+				     gmac_adjust_link);
 	if (!phy)
 		return -ENODEV;
 	netdev->phydev = phy;
-- 
cgit v1.2.3-58-ga151


From 15c22101db71369c022e25001a339f91cd3f8ddc Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@linaro.org>
Date: Wed, 29 May 2024 16:00:01 +0200
Subject: net: ethernet: cortina: Use negotiated TX/RX pause

Instead of directly poking into registers of the PHY, use
the existing function to query phylib about this directly.

Suggested-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20240529-gemini-phylib-fixes-v4-2-16487ca4c2fe@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cortina/gemini.c | 15 +++++----------
 1 file changed, 5 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c
index 8de68a257611..a37d21ceb131 100644
--- a/drivers/net/ethernet/cortina/gemini.c
+++ b/drivers/net/ethernet/cortina/gemini.c
@@ -293,8 +293,8 @@ static void gmac_adjust_link(struct net_device *netdev)
 	struct gemini_ethernet_port *port = netdev_priv(netdev);
 	struct phy_device *phydev = netdev->phydev;
 	union gmac_status status, old_status;
-	int pause_tx = 0;
-	int pause_rx = 0;
+	bool pause_tx = false;
+	bool pause_rx = false;
 
 	status.bits32 = readl(port->gmac_base + GMAC_STATUS);
 	old_status.bits32 = status.bits32;
@@ -329,14 +329,9 @@ static void gmac_adjust_link(struct net_device *netdev)
 	}
 
 	if (phydev->duplex == DUPLEX_FULL) {
-		u16 lcladv = phy_read(phydev, MII_ADVERTISE);
-		u16 rmtadv = phy_read(phydev, MII_LPA);
-		u8 cap = mii_resolve_flowctrl_fdx(lcladv, rmtadv);
-
-		if (cap & FLOW_CTRL_RX)
-			pause_rx = 1;
-		if (cap & FLOW_CTRL_TX)
-			pause_tx = 1;
+		phy_get_pause(phydev, &pause_tx, &pause_rx);
+		netdev_dbg(netdev, "set negotiated pause params pause TX = %s, pause RX = %s\n",
+			   pause_tx ? "ON" : "OFF", pause_rx ? "ON" : "OFF");
 	}
 
 	gmac_set_flow_control(netdev, pause_tx, pause_rx);
-- 
cgit v1.2.3-58-ga151


From dbdb0918da671bde4e9c4c3e974a5e15358dfa5c Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@linaro.org>
Date: Wed, 29 May 2024 16:00:02 +0200
Subject: net: ethernet: cortina: Implement .set_pauseparam()

The Cortina Gemini ethernet can very well set up TX or RX
pausing, so add this functionality to the driver in a
.set_pauseparam() callback. Essentially just call down to
phylib and let phylib deal with this, .adjust_link()
will respect the setting from phylib.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20240529-gemini-phylib-fixes-v4-3-16487ca4c2fe@linaro.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cortina/gemini.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c
index a37d21ceb131..73e1c71c5092 100644
--- a/drivers/net/ethernet/cortina/gemini.c
+++ b/drivers/net/ethernet/cortina/gemini.c
@@ -2126,6 +2126,19 @@ static void gmac_get_pauseparam(struct net_device *netdev,
 	pparam->autoneg = true;
 }
 
+static int gmac_set_pauseparam(struct net_device *netdev,
+			       struct ethtool_pauseparam *pparam)
+{
+	struct phy_device *phydev = netdev->phydev;
+
+	if (!pparam->autoneg)
+		return -EOPNOTSUPP;
+
+	phy_set_asym_pause(phydev, pparam->rx_pause, pparam->tx_pause);
+
+	return 0;
+}
+
 static void gmac_get_ringparam(struct net_device *netdev,
 			       struct ethtool_ringparam *rp,
 			       struct kernel_ethtool_ringparam *kernel_rp,
@@ -2246,6 +2259,7 @@ static const struct ethtool_ops gmac_351x_ethtool_ops = {
 	.set_link_ksettings = gmac_set_ksettings,
 	.nway_reset	= gmac_nway_reset,
 	.get_pauseparam	= gmac_get_pauseparam,
+	.set_pauseparam = gmac_set_pauseparam,
 	.get_ringparam	= gmac_get_ringparam,
 	.set_ringparam	= gmac_set_ringparam,
 	.get_coalesce	= gmac_get_coalesce,
-- 
cgit v1.2.3-58-ga151


From 5160b129f65fc6e1a4aa282d44f824e12aa800ee Mon Sep 17 00:00:00 2001
From: Rengarajan S <rengarajan.s@microchip.com>
Date: Wed, 29 May 2024 19:32:55 +0530
Subject: lan78xx: Enable 125 MHz CLK configuration for LAN7801 if NO EEPROM is
 detected

The 125MHz and 25MHz clock configurations are enabled in the initialization
regardless of EEPROM (125MHz is needed for RGMII 1000Mbps operation). After
a lite reset (lan78xx_reset), these contents go back to defaults(all 0, so
no 125MHz or 25MHz clock).

Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Rengarajan S <rengarajan.s@microchip.com>
Link: https://lore.kernel.org/r/20240529140256.1849764-2-rengarajan.s@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/usb/lan78xx.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index 7a5cc49ebec6..2472216564e7 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -2941,6 +2941,8 @@ static int lan78xx_reset(struct lan78xx_net *dev)
 		return ret;
 
 	buf |= HW_CFG_MEF_;
+	buf |= HW_CFG_CLK125_EN_;
+	buf |= HW_CFG_REFCLK25_EN_;
 
 	ret = lan78xx_write_reg(dev, HW_CFG, buf);
 	if (ret < 0)
-- 
cgit v1.2.3-58-ga151


From 799f532de13601f91ecb9cc6169d45d6e6933b0f Mon Sep 17 00:00:00 2001
From: Rengarajan S <rengarajan.s@microchip.com>
Date: Wed, 29 May 2024 19:32:56 +0530
Subject: lan78xx: Enable Auto Speed and Auto Duplex configuration for LAN7801
 if NO EEPROM is detected

Enabled ASD/ADD configuration for LAN7801 in the absence of EEPROM.
After the lite reset these contents go back to defaults where ASD/
ADD is disabled. The check is already available for LAN7800.

Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Rengarajan S <rengarajan.s@microchip.com>
Link: https://lore.kernel.org/r/20240529140256.1849764-3-rengarajan.s@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/usb/lan78xx.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index 2472216564e7..8adf77e3557e 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -3031,8 +3031,11 @@ static int lan78xx_reset(struct lan78xx_net *dev)
 		return ret;
 
 	/* LAN7801 only has RGMII mode */
-	if (dev->chipid == ID_REV_CHIP_ID_7801_)
+	if (dev->chipid == ID_REV_CHIP_ID_7801_) {
 		buf &= ~MAC_CR_GMII_EN_;
+		/* Enable Auto Duplex and Auto speed */
+		buf |= MAC_CR_AUTO_DUPLEX_ | MAC_CR_AUTO_SPEED_;
+	}
 
 	if (dev->chipid == ID_REV_CHIP_ID_7800_ ||
 	    dev->chipid == ID_REV_CHIP_ID_7850_) {
-- 
cgit v1.2.3-58-ga151


From b5c089880723b2c18531c40e445235bd646a51d1 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Wed, 29 May 2024 07:46:48 -0700
Subject: af_unix: Remove dead code in unix_stream_read_generic().

When splice() support was added in commit 2b514574f7e8 ("net:
af_unix: implement splice for stream af_unix sockets"), we had
to release unix_sk(sk)->readlock (current iolock) before calling
splice_to_pipe().

Due to the unlock, commit 73ed5d25dce0 ("af-unix: fix use-after-free
with concurrent readers while splicing") added a safeguard in
unix_stream_read_generic(); we had to bump the skb refcount before
calling ->recv_actor() and then check if the skb was consumed by a
concurrent reader.

However, the pipe side locking was refactored, and since commit
25869262ef7a ("skb_splice_bits(): get rid of callback"), we can
call splice_to_pipe() without releasing unix_sk(sk)->iolock.

Now, the skb is always alive after the ->recv_actor() callback,
so let's remove the unnecessary drop_skb logic.

This is mostly the revert of commit 73ed5d25dce0 ("af-unix: fix
use-after-free with concurrent readers while splicing").

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://lore.kernel.org/r/20240529144648.68591-1-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/unix/af_unix.c | 21 ++-------------------
 1 file changed, 2 insertions(+), 19 deletions(-)

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 25b49efc0926..861793b489f6 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -654,8 +654,8 @@ static void unix_release_sock(struct sock *sk, int embrion)
 	while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) {
 		if (state == TCP_LISTEN)
 			unix_release_sock(skb->sk, 1);
+
 		/* passed fds are erased in the kfree_skb hook	      */
-		UNIXCB(skb).consumed = skb->len;
 		kfree_skb(skb);
 	}
 
@@ -2704,9 +2704,8 @@ static int unix_stream_read_generic(struct unix_stream_read_state *state,
 	skip = max(sk_peek_offset(sk, flags), 0);
 
 	do {
-		int chunk;
-		bool drop_skb;
 		struct sk_buff *skb, *last;
+		int chunk;
 
 redo:
 		unix_state_lock(sk);
@@ -2802,11 +2801,7 @@ unlock:
 		}
 
 		chunk = min_t(unsigned int, unix_skb_len(skb) - skip, size);
-		skb_get(skb);
 		chunk = state->recv_actor(skb, skip, chunk, state);
-		drop_skb = !unix_skb_len(skb);
-		/* skb is only safe to use if !drop_skb */
-		consume_skb(skb);
 		if (chunk < 0) {
 			if (copied == 0)
 				copied = -EFAULT;
@@ -2815,18 +2810,6 @@ unlock:
 		copied += chunk;
 		size -= chunk;
 
-		if (drop_skb) {
-			/* the skb was touched by a concurrent reader;
-			 * we should not expect anything from this skb
-			 * anymore and assume it invalid - we can be
-			 * sure it was dropped from the socket queue
-			 *
-			 * let's report a short read
-			 */
-			err = 0;
-			break;
-		}
-
 		/* Mark read part of skb as used */
 		if (!(flags & MSG_PEEK)) {
 			UNIXCB(skb).consumed += chunk;
-- 
cgit v1.2.3-58-ga151


From 9373ed280651af79e0adfc29797c26ab47bb9f34 Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Wed, 29 May 2024 10:02:44 +0800
Subject: wifi: rtw89: chan: Use swap() in rtw89_swap_sub_entity()

Use existing swap() function instead of keeping duplicate source code.

./drivers/net/wireless/realtek/rtw89/chan.c:2336:32-33: WARNING opportunity for swap().

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=9174
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240529020244.129027-1-jiapeng.chong@linux.alibaba.com
---
 drivers/net/wireless/realtek/rtw89/chan.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/chan.c b/drivers/net/wireless/realtek/rtw89/chan.c
index 051a3cad6101..3b1997223cc5 100644
--- a/drivers/net/wireless/realtek/rtw89/chan.c
+++ b/drivers/net/wireless/realtek/rtw89/chan.c
@@ -2322,7 +2322,6 @@ static void rtw89_swap_sub_entity(struct rtw89_dev *rtwdev,
 				  enum rtw89_sub_entity_idx idx2)
 {
 	struct rtw89_hal *hal = &rtwdev->hal;
-	struct rtw89_sub_entity tmp;
 	struct rtw89_vif *rtwvif;
 	u8 cur;
 
@@ -2332,9 +2331,7 @@ static void rtw89_swap_sub_entity(struct rtw89_dev *rtwdev,
 	hal->sub[idx1].cfg->idx = idx2;
 	hal->sub[idx2].cfg->idx = idx1;
 
-	tmp = hal->sub[idx1];
-	hal->sub[idx1] = hal->sub[idx2];
-	hal->sub[idx2] = tmp;
+	swap(hal->sub[idx1], hal->sub[idx2]);
 
 	rtw89_for_each_rtwvif(rtwdev, rtwvif) {
 		if (!rtwvif->chanctx_assigned)
-- 
cgit v1.2.3-58-ga151


From 3ac14b9dfbd345e891d48d89f6c2fa519848f0f4 Mon Sep 17 00:00:00 2001
From: Guangguan Wang <guangguan.wang@linux.alibaba.com>
Date: Mon, 3 Jun 2024 11:00:18 +0800
Subject: net/smc: set rmb's SG_MAX_SINGLE_ALLOC limitation only when
 CONFIG_ARCH_NO_SG_CHAIN is defined

SG_MAX_SINGLE_ALLOC is used to limit maximum number of entries that
will be allocated in one piece of scatterlist. When the entries of
scatterlist exceeds SG_MAX_SINGLE_ALLOC, sg chain will be used. From
commit 7c703e54cc71 ("arch: switch the default on ARCH_HAS_SG_CHAIN"),
we can know that the macro CONFIG_ARCH_NO_SG_CHAIN is used to identify
whether sg chain is supported. So, SMC-R's rmb buffer should be limited
by SG_MAX_SINGLE_ALLOC only when the macro CONFIG_ARCH_NO_SG_CHAIN is
defined.

Fixes: a3fe3d01bd0d ("net/smc: introduce sg-logic for RMBs")
Signed-off-by: Guangguan Wang <guangguan.wang@linux.alibaba.com>
Co-developed-by: Wen Gu <guwen@linux.alibaba.com>
Signed-off-by: Wen Gu <guwen@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/smc/smc_core.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c
index fafdb97adfad..acca3b1a068f 100644
--- a/net/smc/smc_core.c
+++ b/net/smc/smc_core.c
@@ -2015,7 +2015,6 @@ out:
  */
 static u8 smc_compress_bufsize(int size, bool is_smcd, bool is_rmb)
 {
-	const unsigned int max_scat = SG_MAX_SINGLE_ALLOC * PAGE_SIZE;
 	u8 compressed;
 
 	if (size <= SMC_BUF_MIN_SIZE)
@@ -2025,9 +2024,11 @@ static u8 smc_compress_bufsize(int size, bool is_smcd, bool is_rmb)
 	compressed = min_t(u8, ilog2(size) + 1,
 			   is_smcd ? SMCD_DMBE_SIZES : SMCR_RMBE_SIZES);
 
+#ifdef CONFIG_ARCH_NO_SG_CHAIN
 	if (!is_smcd && is_rmb)
 		/* RMBs are backed by & limited to max size of scatterlists */
-		compressed = min_t(u8, compressed, ilog2(max_scat >> 14));
+		compressed = min_t(u8, compressed, ilog2((SG_MAX_SINGLE_ALLOC * PAGE_SIZE) >> 14));
+#endif
 
 	return compressed;
 }
-- 
cgit v1.2.3-58-ga151


From 2f4b101c542e102c680d136fc6dca081e3fbda4a Mon Sep 17 00:00:00 2001
From: Guangguan Wang <guangguan.wang@linux.alibaba.com>
Date: Mon, 3 Jun 2024 11:00:19 +0800
Subject: net/smc: change SMCR_RMBE_SIZES from 5 to 15

SMCR_RMBE_SIZES is the upper boundary of SMC-R's snd_buf and rcv_buf.
The maximum bytes of snd_buf and rcv_buf can be calculated by 2^SMCR_
RMBE_SIZES * 16KB. SMCR_RMBE_SIZES = 5 means the upper boundary is 512KB.
TCP's snd_buf and rcv_buf max size is configured by net.ipv4.tcp_w/rmem[2]
whose default value is 4MB or 6MB, is much larger than SMC-R's upper
boundary.

In some scenarios, such as Recommendation System, the communication
pattern is mainly large size send/recv, where the size of snd_buf and
rcv_buf greatly affects performance. Due to the upper boundary
disadvantage, SMC-R performs poor than TCP in those scenarios. So it
is time to enlarge the upper boundary size of SMC-R's snd_buf and rcv_buf,
so that the SMC-R's snd_buf and rcv_buf can be configured to larger size
for performance gain in such scenarios.

The SMC-R rcv_buf's size will be transferred to peer by the field
rmbe_size in clc accept and confirm message. The length of the field
rmbe_size is four bits, which means the maximum value of SMCR_RMBE_SIZES
is 15. In case of frequently adjusting the value of SMCR_RMBE_SIZES
in different scenarios, set the value of SMCR_RMBE_SIZES to the maximum
value 15, which means the upper boundary of SMC-R's snd_buf and rcv_buf
is 512MB. As the real memory usage is determined by the value of
net.smc.w/rmem, not by the upper boundary, set the value of SMCR_RMBE_SIZES
to the maximum value has no side affects.

Signed-off-by: Guangguan Wang <guangguan.wang@linux.alibaba.com>
Co-developed-by: Wen Gu <guwen@linux.alibaba.com>
Signed-off-by: Wen Gu <guwen@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/smc/smc_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c
index acca3b1a068f..3b95828d9976 100644
--- a/net/smc/smc_core.c
+++ b/net/smc/smc_core.c
@@ -2006,7 +2006,7 @@ out:
 }
 
 #define SMCD_DMBE_SIZES		6 /* 0 -> 16KB, 1 -> 32KB, .. 6 -> 1MB */
-#define SMCR_RMBE_SIZES		5 /* 0 -> 16KB, 1 -> 32KB, .. 5 -> 512KB */
+#define SMCR_RMBE_SIZES		15 /* 0 -> 16KB, 1 -> 32KB, .. 15 -> 512MB */
 
 /* convert the RMB size into the compressed notation (minimum 16K, see
  * SMCD/R_DMBE_SIZES.
-- 
cgit v1.2.3-58-ga151


From 454cfffe8dc112f1ed3966923f5dccdafa3bebf2 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 30 May 2024 19:33:26 +0300
Subject: net: dsa: ocelot: use devres in ocelot_ext_probe()

Russell King suggested that felix_vsc9959, seville_vsc9953 and
ocelot_ext have a large portion of duplicated init and teardown code,
which could be made common [1]. The teardown code could even be
simplified away if we made use of devres, something which is used here
and there in the felix driver, just not very consistently.

[1] https://lore.kernel.org/all/Zh1GvcOTXqb7CpQt@shell.armlinux.org.uk/

Prepare the ground in the ocelot_ext driver, by allocating the data
structures using devres and deleting the kfree() calls. This also
deletes the "Failed to allocate ..." message, since memory allocation
errors are extremely loud anyway, and it's hard to miss them.

Suggested-by: "Russell King (Oracle)" <linux@armlinux.org.uk>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Colin Foster <colin.foster@in-advantage.com>
Tested-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/ocelot_ext.c | 24 +++++-------------------
 1 file changed, 5 insertions(+), 19 deletions(-)

diff --git a/drivers/net/dsa/ocelot/ocelot_ext.c b/drivers/net/dsa/ocelot/ocelot_ext.c
index a8927dc7aca4..c893f3ee238b 100644
--- a/drivers/net/dsa/ocelot/ocelot_ext.c
+++ b/drivers/net/dsa/ocelot/ocelot_ext.c
@@ -71,7 +71,7 @@ static int ocelot_ext_probe(struct platform_device *pdev)
 	struct felix *felix;
 	int err;
 
-	felix = kzalloc(sizeof(*felix), GFP_KERNEL);
+	felix = devm_kzalloc(dev, sizeof(*felix), GFP_KERNEL);
 	if (!felix)
 		return -ENOMEM;
 
@@ -84,12 +84,9 @@ static int ocelot_ext_probe(struct platform_device *pdev)
 
 	felix->info = &vsc7512_info;
 
-	ds = kzalloc(sizeof(*ds), GFP_KERNEL);
-	if (!ds) {
-		err = -ENOMEM;
-		dev_err_probe(dev, err, "Failed to allocate DSA switch\n");
-		goto err_free_felix;
-	}
+	ds = devm_kzalloc(dev, sizeof(*ds), GFP_KERNEL);
+	if (!ds)
+		return -ENOMEM;
 
 	ds->dev = dev;
 	ds->num_ports = felix->info->num_ports;
@@ -102,17 +99,9 @@ static int ocelot_ext_probe(struct platform_device *pdev)
 	felix->tag_proto = DSA_TAG_PROTO_OCELOT;
 
 	err = dsa_register_switch(ds);
-	if (err) {
+	if (err)
 		dev_err_probe(dev, err, "Failed to register DSA switch\n");
-		goto err_free_ds;
-	}
-
-	return 0;
 
-err_free_ds:
-	kfree(ds);
-err_free_felix:
-	kfree(felix);
 	return err;
 }
 
@@ -124,9 +113,6 @@ static void ocelot_ext_remove(struct platform_device *pdev)
 		return;
 
 	dsa_unregister_switch(felix->ds);
-
-	kfree(felix->ds);
-	kfree(felix);
 }
 
 static void ocelot_ext_shutdown(struct platform_device *pdev)
-- 
cgit v1.2.3-58-ga151


From 90ee9a5b49cea302b666217068e52e0bb325304d Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 30 May 2024 19:33:27 +0300
Subject: net: dsa: ocelot: use devres in seville_probe()

Russell King suggested that felix_vsc9959, seville_vsc9953 and
ocelot_ext have a large portion of duplicated init and teardown code,
which could be made common [1]. The teardown code could even be
simplified away if we made use of devres, something which is used here
and there in the felix driver, just not very consistently.

[1] https://lore.kernel.org/all/Zh1GvcOTXqb7CpQt@shell.armlinux.org.uk/

Prepare the ground in the seville_vsc9953 driver, by allocating the data
structures using devres and deleting the kfree() calls. This also
deletes the "Failed to allocate ..." message, since memory allocation
errors are extremely loud anyway, and it's hard to miss them.

Suggested-by: "Russell King (Oracle)" <linux@armlinux.org.uk>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/seville_vsc9953.c | 44 ++++++++++----------------------
 1 file changed, 13 insertions(+), 31 deletions(-)

diff --git a/drivers/net/dsa/ocelot/seville_vsc9953.c b/drivers/net/dsa/ocelot/seville_vsc9953.c
index 5ac8897e232b..e63247d3dfdb 100644
--- a/drivers/net/dsa/ocelot/seville_vsc9953.c
+++ b/drivers/net/dsa/ocelot/seville_vsc9953.c
@@ -971,42 +971,36 @@ static const struct felix_info seville_info_vsc9953 = {
 
 static int seville_probe(struct platform_device *pdev)
 {
+	struct device *dev = &pdev->dev;
 	struct dsa_switch *ds;
 	struct ocelot *ocelot;
 	struct resource *res;
 	struct felix *felix;
 	int err;
 
-	felix = kzalloc(sizeof(struct felix), GFP_KERNEL);
-	if (!felix) {
-		err = -ENOMEM;
-		dev_err(&pdev->dev, "Failed to allocate driver memory\n");
-		goto err_alloc_felix;
-	}
+	felix = devm_kzalloc(dev, sizeof(struct felix), GFP_KERNEL);
+	if (!felix)
+		return -ENOMEM;
 
 	platform_set_drvdata(pdev, felix);
 
 	ocelot = &felix->ocelot;
-	ocelot->dev = &pdev->dev;
+	ocelot->dev = dev;
 	ocelot->num_flooding_pgids = 1;
 	felix->info = &seville_info_vsc9953;
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (!res) {
-		err = -EINVAL;
-		dev_err(&pdev->dev, "Invalid resource\n");
-		goto err_alloc_felix;
+		dev_err(dev, "Invalid resource\n");
+		return -EINVAL;
 	}
 	felix->switch_base = res->start;
 
-	ds = kzalloc(sizeof(struct dsa_switch), GFP_KERNEL);
-	if (!ds) {
-		err = -ENOMEM;
-		dev_err(&pdev->dev, "Failed to allocate DSA switch\n");
-		goto err_alloc_ds;
-	}
+	ds = devm_kzalloc(dev, sizeof(struct dsa_switch), GFP_KERNEL);
+	if (!ds)
+		return -ENOMEM;
 
-	ds->dev = &pdev->dev;
+	ds->dev = dev;
 	ds->num_ports = felix->info->num_ports;
 	ds->ops = &felix_switch_ops;
 	ds->phylink_mac_ops = &felix_phylink_mac_ops;
@@ -1015,18 +1009,9 @@ static int seville_probe(struct platform_device *pdev)
 	felix->tag_proto = DSA_TAG_PROTO_SEVILLE;
 
 	err = dsa_register_switch(ds);
-	if (err) {
-		dev_err(&pdev->dev, "Failed to register DSA switch: %d\n", err);
-		goto err_register_ds;
-	}
+	if (err)
+		dev_err(dev, "Failed to register DSA switch: %d\n", err);
 
-	return 0;
-
-err_register_ds:
-	kfree(ds);
-err_alloc_ds:
-err_alloc_felix:
-	kfree(felix);
 	return err;
 }
 
@@ -1038,9 +1023,6 @@ static void seville_remove(struct platform_device *pdev)
 		return;
 
 	dsa_unregister_switch(felix->ds);
-
-	kfree(felix->ds);
-	kfree(felix);
 }
 
 static void seville_shutdown(struct platform_device *pdev)
-- 
cgit v1.2.3-58-ga151


From cc711c523da7280c3bb01ec726596154f5af9798 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 30 May 2024 19:33:28 +0300
Subject: net: dsa: ocelot: delete open coded status = "disabled" parsing

Since commit 6fffbc7ae137 ("PCI: Honor firmware's device disabled
status"), PCI device drivers with OF bindings no longer need this check.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix_vsc9959.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index d4799a908abc..eabb55da0982 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -2668,11 +2668,6 @@ static int felix_pci_probe(struct pci_dev *pdev,
 	struct felix *felix;
 	int err;
 
-	if (pdev->dev.of_node && !of_device_is_available(pdev->dev.of_node)) {
-		dev_info(&pdev->dev, "device is disabled, skipping\n");
-		return -ENODEV;
-	}
-
 	err = pci_enable_device(pdev);
 	if (err) {
 		dev_err(&pdev->dev, "device enable failed\n");
-- 
cgit v1.2.3-58-ga151


From 4510bbd38cbef7e6fc1a786f3e0213ba07fe0a00 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 30 May 2024 19:33:29 +0300
Subject: net: dsa: ocelot: consistently use devres in felix_pci_probe()

Russell King suggested that felix_vsc9959, seville_vsc9953 and
ocelot_ext have a large portion of duplicated init and teardown code,
which could be made common [1]. The teardown code could even be
simplified away if we made use of devres, something which is used here
and there in the felix driver, just not very consistently.

[1] https://lore.kernel.org/all/Zh1GvcOTXqb7CpQt@shell.armlinux.org.uk/

Prepare the ground in the felix_vsc9959 driver, by allocating the data
structures using devres and deleting the kfree() calls. This also
deletes the "Failed to allocate ..." message, since memory allocation
errors are extremely loud anyway, and it's hard to miss them.

Suggested-by: "Russell King (Oracle)" <linux@armlinux.org.uk>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix_vsc9959.c | 40 +++++++++++++---------------------
 1 file changed, 15 insertions(+), 25 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index eabb55da0982..34155a0ffd7e 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -2663,6 +2663,7 @@ static irqreturn_t felix_irq_handler(int irq, void *data)
 static int felix_pci_probe(struct pci_dev *pdev,
 			   const struct pci_device_id *id)
 {
+	struct device *dev = &pdev->dev;
 	struct dsa_switch *ds;
 	struct ocelot *ocelot;
 	struct felix *felix;
@@ -2670,45 +2671,43 @@ static int felix_pci_probe(struct pci_dev *pdev,
 
 	err = pci_enable_device(pdev);
 	if (err) {
-		dev_err(&pdev->dev, "device enable failed\n");
-		goto err_pci_enable;
+		dev_err(dev, "device enable failed: %pe\n", ERR_PTR(err));
+		return err;
 	}
 
-	felix = kzalloc(sizeof(struct felix), GFP_KERNEL);
+	felix = devm_kzalloc(dev, sizeof(struct felix), GFP_KERNEL);
 	if (!felix) {
 		err = -ENOMEM;
-		dev_err(&pdev->dev, "Failed to allocate driver memory\n");
-		goto err_alloc_felix;
+		goto out_disable;
 	}
 
 	pci_set_drvdata(pdev, felix);
 	ocelot = &felix->ocelot;
-	ocelot->dev = &pdev->dev;
+	ocelot->dev = dev;
 	ocelot->num_flooding_pgids = OCELOT_NUM_TC;
 	felix->info = &felix_info_vsc9959;
 	felix->switch_base = pci_resource_start(pdev, VSC9959_SWITCH_PCI_BAR);
 
 	pci_set_master(pdev);
 
-	err = devm_request_threaded_irq(&pdev->dev, pdev->irq, NULL,
+	err = devm_request_threaded_irq(dev, pdev->irq, NULL,
 					&felix_irq_handler, IRQF_ONESHOT,
 					"felix-intb", ocelot);
 	if (err) {
-		dev_err(&pdev->dev, "Failed to request irq\n");
-		goto err_alloc_irq;
+		dev_err(dev, "Failed to request irq: %pe\n", ERR_PTR(err));
+		goto out_disable;
 	}
 
 	ocelot->ptp = 1;
 	ocelot->mm_supported = true;
 
-	ds = kzalloc(sizeof(struct dsa_switch), GFP_KERNEL);
+	ds = devm_kzalloc(dev, sizeof(struct dsa_switch), GFP_KERNEL);
 	if (!ds) {
 		err = -ENOMEM;
-		dev_err(&pdev->dev, "Failed to allocate DSA switch\n");
-		goto err_alloc_ds;
+		goto out_disable;
 	}
 
-	ds->dev = &pdev->dev;
+	ds->dev = dev;
 	ds->num_ports = felix->info->num_ports;
 	ds->num_tx_queues = felix->info->num_tx_queues;
 	ds->ops = &felix_switch_ops;
@@ -2719,20 +2718,14 @@ static int felix_pci_probe(struct pci_dev *pdev,
 
 	err = dsa_register_switch(ds);
 	if (err) {
-		dev_err_probe(&pdev->dev, err, "Failed to register DSA switch\n");
-		goto err_register_ds;
+		dev_err_probe(dev, err, "Failed to register DSA switch\n");
+		goto out_disable;
 	}
 
 	return 0;
 
-err_register_ds:
-	kfree(ds);
-err_alloc_ds:
-err_alloc_irq:
-	kfree(felix);
-err_alloc_felix:
+out_disable:
 	pci_disable_device(pdev);
-err_pci_enable:
 	return err;
 }
 
@@ -2745,9 +2738,6 @@ static void felix_pci_remove(struct pci_dev *pdev)
 
 	dsa_unregister_switch(felix->ds);
 
-	kfree(felix->ds);
-	kfree(felix);
-
 	pci_disable_device(pdev);
 }
 
-- 
cgit v1.2.3-58-ga151


From 0367a1775933aab80a99ebc91a14efbcaaa45a00 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 30 May 2024 19:33:30 +0300
Subject: net: dsa: ocelot: move devm_request_threaded_irq() to felix_setup()

The current placement of devm_request_threaded_irq() is inconvenient.
It is between the allocation of the "felix" structure and
dsa_register_switch(), both of which we'd like to refactor into a
function that's common for all switches. But the IRQ is specific to
felix_vsc9959.

A closer inspection of the felix_irq_handler() code suggests that
it does things that depend on the data structures having been fully
initialized. For example, ocelot_get_txtstamp() takes
&port->tx_skbs.lock, which has only been initialized in
ocelot_init_port() which has not run yet.

It is not one of those IRQF_SHARED IRQs, so CONFIG_DEBUG_SHIRQ_FIXME
shouldn't apply here, and thus, it doesn't really matter, because in
practice, the IRQ will not be triggered so early. Nonetheless, it is a
good practice for the driver to be prepared for it to fire as soon as it
is requested.

Create a new felix->info method for running custom code for vsc9959 from
within felix_setup(), and move the request_irq() call there. The
ocelot_ext should have an IRQ as well, so this should be a step in the
right direction for that model (VSC7512) as well.

Some minor changes are made while moving the code. Casts from void *
aren't necessary, so drop them, and rename felix_irq_handler() to the
more specific vsc9959_irq_handler().

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix.c         |  9 +++++++
 drivers/net/dsa/ocelot/felix.h         |  1 +
 drivers/net/dsa/ocelot/felix_vsc9959.c | 44 ++++++++++++++++++----------------
 3 files changed, 33 insertions(+), 21 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index 3aa66bf9eafc..09c0800b18ab 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -1597,6 +1597,15 @@ static int felix_setup(struct dsa_switch *ds)
 		felix_port_qos_map_init(ocelot, dp->index);
 	}
 
+	if (felix->info->request_irq) {
+		err = felix->info->request_irq(ocelot);
+		if (err) {
+			dev_err(ocelot->dev, "Failed to request IRQ: %pe\n",
+				ERR_PTR(err));
+			goto out_deinit_ports;
+		}
+	}
+
 	err = ocelot_devlink_sb_register(ocelot);
 	if (err)
 		goto out_deinit_ports;
diff --git a/drivers/net/dsa/ocelot/felix.h b/drivers/net/dsa/ocelot/felix.h
index 4d3489aaa659..e67a25f6f816 100644
--- a/drivers/net/dsa/ocelot/felix.h
+++ b/drivers/net/dsa/ocelot/felix.h
@@ -64,6 +64,7 @@ struct felix_info {
 				      const struct phylink_link_state *state);
 	int	(*configure_serdes)(struct ocelot *ocelot, int port,
 				    struct device_node *portnp);
+	int	(*request_irq)(struct ocelot *ocelot);
 };
 
 /* Methods for initializing the hardware resources specific to a tagging
diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index 34155a0ffd7e..20563abd617f 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -2605,6 +2605,28 @@ set:
 	}
 }
 
+/* The INTB interrupt is shared between for PTP TX timestamp availability
+ * notification and MAC Merge status change on each port.
+ */
+static irqreturn_t vsc9959_irq_handler(int irq, void *data)
+{
+	struct ocelot *ocelot = data;
+
+	ocelot_get_txtstamp(ocelot);
+	ocelot_mm_irq(ocelot);
+
+	return IRQ_HANDLED;
+}
+
+static int vsc9959_request_irq(struct ocelot *ocelot)
+{
+	struct pci_dev *pdev = to_pci_dev(ocelot->dev);
+
+	return devm_request_threaded_irq(ocelot->dev, pdev->irq, NULL,
+					 &vsc9959_irq_handler, IRQF_ONESHOT,
+					 "felix-intb", ocelot);
+}
+
 static const struct ocelot_ops vsc9959_ops = {
 	.reset			= vsc9959_reset,
 	.wm_enc			= vsc9959_wm_enc,
@@ -2645,21 +2667,9 @@ static const struct felix_info felix_info_vsc9959 = {
 	.port_modes		= vsc9959_port_modes,
 	.port_setup_tc		= vsc9959_port_setup_tc,
 	.port_sched_speed_set	= vsc9959_sched_speed_set,
+	.request_irq		= vsc9959_request_irq,
 };
 
-/* The INTB interrupt is shared between for PTP TX timestamp availability
- * notification and MAC Merge status change on each port.
- */
-static irqreturn_t felix_irq_handler(int irq, void *data)
-{
-	struct ocelot *ocelot = (struct ocelot *)data;
-
-	ocelot_get_txtstamp(ocelot);
-	ocelot_mm_irq(ocelot);
-
-	return IRQ_HANDLED;
-}
-
 static int felix_pci_probe(struct pci_dev *pdev,
 			   const struct pci_device_id *id)
 {
@@ -2690,14 +2700,6 @@ static int felix_pci_probe(struct pci_dev *pdev,
 
 	pci_set_master(pdev);
 
-	err = devm_request_threaded_irq(dev, pdev->irq, NULL,
-					&felix_irq_handler, IRQF_ONESHOT,
-					"felix-intb", ocelot);
-	if (err) {
-		dev_err(dev, "Failed to request irq: %pe\n", ERR_PTR(err));
-		goto out_disable;
-	}
-
 	ocelot->ptp = 1;
 	ocelot->mm_supported = true;
 
-- 
cgit v1.2.3-58-ga151


From 4ca54dd96eca57adea961c0e6cca28bf76168429 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 30 May 2024 19:33:31 +0300
Subject: net: dsa: ocelot: use ds->num_tx_queues = OCELOT_NUM_TC for all
 models

Russell King points out that seville_vsc9953 populates
felix->info->num_tx_queues = 8, but this doesn't make it all the way
into ds->num_tx_queues (which is how the user interface netdev queues
get allocated) [1].

[1]: https://lore.kernel.org/all/20240415160150.yejcazpjqvn7vhxu@skbuf/

When num_tx_queues=0 for seville, this is implicitly converted to 1 by
dsa_user_create(), and this is good enough for basic operation for a
switch port. The tc qdisc offload layer works with netdev TX queues,
so for QoS offload we need to pretend we have multiple TX queues. The
VSC9953, like ocelot_ext, doesn't export QoS offload, so it doesn't
really matter. But we can definitely set num_tx_queues=8 for all
switches.

The felix->info->num_tx_queues construct itself seems unnecessary.
It was introduced by commit de143c0e274b ("net: dsa: felix: Configure
Time-Aware Scheduler via taprio offload") at a time when vsc9959
(LS1028A) was the only switch supported by the driver.

8 traffic classes, and 1 queue per traffic class, is a common
architectural feature of all switches in the family. So they could
all just set OCELOT_NUM_TC and be fine.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Colin Foster <colin.foster@in-advantage.com>
Tested-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix.h           | 1 -
 drivers/net/dsa/ocelot/felix_vsc9959.c   | 4 ++--
 drivers/net/dsa/ocelot/ocelot_ext.c      | 3 +--
 drivers/net/dsa/ocelot/seville_vsc9953.c | 3 ++-
 4 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix.h b/drivers/net/dsa/ocelot/felix.h
index e67a25f6f816..e0bfea10ff52 100644
--- a/drivers/net/dsa/ocelot/felix.h
+++ b/drivers/net/dsa/ocelot/felix.h
@@ -32,7 +32,6 @@ struct felix_info {
 	const u32			*port_modes;
 	int				num_mact_rows;
 	int				num_ports;
-	int				num_tx_queues;
 	struct vcap_props		*vcap;
 	u16				vcap_pol_base;
 	u16				vcap_pol_max;
diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index 20563abd617f..ec8b124e8f61 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -2658,7 +2658,6 @@ static const struct felix_info felix_info_vsc9959 = {
 	.vcap_pol_max2		= 0,
 	.num_mact_rows		= 2048,
 	.num_ports		= VSC9959_NUM_PORTS,
-	.num_tx_queues		= OCELOT_NUM_TC,
 	.quirks			= FELIX_MAC_QUIRKS,
 	.quirk_no_xtr_irq	= true,
 	.ptp_caps		= &vsc9959_ptp_caps,
@@ -2711,7 +2710,8 @@ static int felix_pci_probe(struct pci_dev *pdev,
 
 	ds->dev = dev;
 	ds->num_ports = felix->info->num_ports;
-	ds->num_tx_queues = felix->info->num_tx_queues;
+	ds->num_tx_queues = OCELOT_NUM_TC;
+
 	ds->ops = &felix_switch_ops;
 	ds->phylink_mac_ops = &felix_phylink_mac_ops;
 	ds->priv = ocelot;
diff --git a/drivers/net/dsa/ocelot/ocelot_ext.c b/drivers/net/dsa/ocelot/ocelot_ext.c
index c893f3ee238b..9cd24f77dc49 100644
--- a/drivers/net/dsa/ocelot/ocelot_ext.c
+++ b/drivers/net/dsa/ocelot/ocelot_ext.c
@@ -57,7 +57,6 @@ static const struct felix_info vsc7512_info = {
 	.vcap				= vsc7514_vcap_props,
 	.num_mact_rows			= 1024,
 	.num_ports			= VSC7514_NUM_PORTS,
-	.num_tx_queues			= OCELOT_NUM_TC,
 	.port_modes			= vsc7512_port_modes,
 	.phylink_mac_config		= ocelot_phylink_mac_config,
 	.configure_serdes		= ocelot_port_configure_serdes,
@@ -90,7 +89,7 @@ static int ocelot_ext_probe(struct platform_device *pdev)
 
 	ds->dev = dev;
 	ds->num_ports = felix->info->num_ports;
-	ds->num_tx_queues = felix->info->num_tx_queues;
+	ds->num_tx_queues = OCELOT_NUM_TC;
 
 	ds->ops = &felix_switch_ops;
 	ds->phylink_mac_ops = &felix_phylink_mac_ops;
diff --git a/drivers/net/dsa/ocelot/seville_vsc9953.c b/drivers/net/dsa/ocelot/seville_vsc9953.c
index e63247d3dfdb..83bd0e1ee692 100644
--- a/drivers/net/dsa/ocelot/seville_vsc9953.c
+++ b/drivers/net/dsa/ocelot/seville_vsc9953.c
@@ -963,7 +963,6 @@ static const struct felix_info seville_info_vsc9953 = {
 	.quirks			= FELIX_MAC_QUIRKS,
 	.num_mact_rows		= 2048,
 	.num_ports		= VSC9953_NUM_PORTS,
-	.num_tx_queues		= OCELOT_NUM_TC,
 	.mdio_bus_alloc		= vsc9953_mdio_bus_alloc,
 	.mdio_bus_free		= vsc9953_mdio_bus_free,
 	.port_modes		= vsc9953_port_modes,
@@ -1002,6 +1001,8 @@ static int seville_probe(struct platform_device *pdev)
 
 	ds->dev = dev;
 	ds->num_ports = felix->info->num_ports;
+	ds->num_tx_queues = OCELOT_NUM_TC;
+
 	ds->ops = &felix_switch_ops;
 	ds->phylink_mac_ops = &felix_phylink_mac_ops;
 	ds->priv = ocelot;
-- 
cgit v1.2.3-58-ga151


From efdbee7d07916d994bc0ff0235d00d46fa991b61 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 30 May 2024 19:33:32 +0300
Subject: net: dsa: ocelot: common probing code

Russell King suggested that felix_vsc9959, seville_vsc9953 and
ocelot_ext have a large portion of duplicated init code, which could be
made common [1].

[1]: https://lore.kernel.org/all/Zh1GvcOTXqb7CpQt@shell.armlinux.org.uk/

Here, we take the following common steps:
- "felix" and "ds" structure allocation
- "felix", "ocelot" and "ds" basic structure initialization
- dsa_register_switch() call

and we make a common function out of them.

For every driver except felix_vsc9959, this is also the entire probing
procedure. For felix_vsc9959, we also need to do some PCI-specific
stuff, which can easily be reordered to be done before, and unwound on
failure.

We also have to convert the bus-specific platform_set_drvdata() and
pci_set_drvdata() calls into dev_set_drvdata(). But this should have no
impact on the behavior.

Suggested-by: "Russell King (Oracle)" <linux@armlinux.org.uk>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Tested-by: Colin Foster <colin.foster@in-advantage.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix.c           | 47 ++++++++++++++++++++++++++++++++
 drivers/net/dsa/ocelot/felix.h           |  5 ++++
 drivers/net/dsa/ocelot/felix_vsc9959.c   | 44 ++++--------------------------
 drivers/net/dsa/ocelot/ocelot_ext.c      | 40 ++-------------------------
 drivers/net/dsa/ocelot/seville_vsc9953.c | 38 ++------------------------
 5 files changed, 63 insertions(+), 111 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index 09c0800b18ab..accf737f7b69 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -2195,6 +2195,53 @@ const struct dsa_switch_ops felix_switch_ops = {
 };
 EXPORT_SYMBOL_GPL(felix_switch_ops);
 
+int felix_register_switch(struct device *dev, resource_size_t switch_base,
+			  int num_flooding_pgids, bool ptp,
+			  bool mm_supported,
+			  enum dsa_tag_protocol init_tag_proto,
+			  const struct felix_info *info)
+{
+	struct dsa_switch *ds;
+	struct ocelot *ocelot;
+	struct felix *felix;
+	int err;
+
+	felix = devm_kzalloc(dev, sizeof(*felix), GFP_KERNEL);
+	if (!felix)
+		return -ENOMEM;
+
+	ds = devm_kzalloc(dev, sizeof(*ds), GFP_KERNEL);
+	if (!ds)
+		return -ENOMEM;
+
+	dev_set_drvdata(dev, felix);
+
+	ocelot = &felix->ocelot;
+	ocelot->dev = dev;
+	ocelot->num_flooding_pgids = num_flooding_pgids;
+	ocelot->ptp = ptp;
+	ocelot->mm_supported = mm_supported;
+
+	felix->info = info;
+	felix->switch_base = switch_base;
+	felix->ds = ds;
+	felix->tag_proto = init_tag_proto;
+
+	ds->dev = dev;
+	ds->num_ports = info->num_ports;
+	ds->num_tx_queues = OCELOT_NUM_TC;
+	ds->ops = &felix_switch_ops;
+	ds->phylink_mac_ops = &felix_phylink_mac_ops;
+	ds->priv = ocelot;
+
+	err = dsa_register_switch(ds);
+	if (err)
+		dev_err_probe(dev, err, "Failed to register DSA switch\n");
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(felix_register_switch);
+
 struct net_device *felix_port_to_netdev(struct ocelot *ocelot, int port)
 {
 	struct felix *felix = ocelot_to_felix(ocelot);
diff --git a/drivers/net/dsa/ocelot/felix.h b/drivers/net/dsa/ocelot/felix.h
index e0bfea10ff52..85b4f8616003 100644
--- a/drivers/net/dsa/ocelot/felix.h
+++ b/drivers/net/dsa/ocelot/felix.h
@@ -100,6 +100,11 @@ struct felix {
 	unsigned long			host_flood_mc_mask;
 };
 
+int felix_register_switch(struct device *dev, resource_size_t switch_base,
+			  int num_flooding_pgids, bool ptp,
+			  bool mm_supported,
+			  enum dsa_tag_protocol init_tag_proto,
+			  const struct felix_info *info);
 struct net_device *felix_port_to_netdev(struct ocelot *ocelot, int port);
 int felix_netdev_to_port(struct net_device *dev);
 
diff --git a/drivers/net/dsa/ocelot/felix_vsc9959.c b/drivers/net/dsa/ocelot/felix_vsc9959.c
index ec8b124e8f61..ba37a566da39 100644
--- a/drivers/net/dsa/ocelot/felix_vsc9959.c
+++ b/drivers/net/dsa/ocelot/felix_vsc9959.c
@@ -2673,9 +2673,7 @@ static int felix_pci_probe(struct pci_dev *pdev,
 			   const struct pci_device_id *id)
 {
 	struct device *dev = &pdev->dev;
-	struct dsa_switch *ds;
-	struct ocelot *ocelot;
-	struct felix *felix;
+	resource_size_t switch_base;
 	int err;
 
 	err = pci_enable_device(pdev);
@@ -2684,45 +2682,15 @@ static int felix_pci_probe(struct pci_dev *pdev,
 		return err;
 	}
 
-	felix = devm_kzalloc(dev, sizeof(struct felix), GFP_KERNEL);
-	if (!felix) {
-		err = -ENOMEM;
-		goto out_disable;
-	}
-
-	pci_set_drvdata(pdev, felix);
-	ocelot = &felix->ocelot;
-	ocelot->dev = dev;
-	ocelot->num_flooding_pgids = OCELOT_NUM_TC;
-	felix->info = &felix_info_vsc9959;
-	felix->switch_base = pci_resource_start(pdev, VSC9959_SWITCH_PCI_BAR);
-
 	pci_set_master(pdev);
 
-	ocelot->ptp = 1;
-	ocelot->mm_supported = true;
+	switch_base = pci_resource_start(pdev, VSC9959_SWITCH_PCI_BAR);
 
-	ds = devm_kzalloc(dev, sizeof(struct dsa_switch), GFP_KERNEL);
-	if (!ds) {
-		err = -ENOMEM;
+	err = felix_register_switch(dev, switch_base, OCELOT_NUM_TC,
+				    true, true, DSA_TAG_PROTO_OCELOT,
+				    &felix_info_vsc9959);
+	if (err)
 		goto out_disable;
-	}
-
-	ds->dev = dev;
-	ds->num_ports = felix->info->num_ports;
-	ds->num_tx_queues = OCELOT_NUM_TC;
-
-	ds->ops = &felix_switch_ops;
-	ds->phylink_mac_ops = &felix_phylink_mac_ops;
-	ds->priv = ocelot;
-	felix->ds = ds;
-	felix->tag_proto = DSA_TAG_PROTO_OCELOT;
-
-	err = dsa_register_switch(ds);
-	if (err) {
-		dev_err_probe(dev, err, "Failed to register DSA switch\n");
-		goto out_disable;
-	}
 
 	return 0;
 
diff --git a/drivers/net/dsa/ocelot/ocelot_ext.c b/drivers/net/dsa/ocelot/ocelot_ext.c
index 9cd24f77dc49..5632a7248cd4 100644
--- a/drivers/net/dsa/ocelot/ocelot_ext.c
+++ b/drivers/net/dsa/ocelot/ocelot_ext.c
@@ -64,44 +64,8 @@ static const struct felix_info vsc7512_info = {
 
 static int ocelot_ext_probe(struct platform_device *pdev)
 {
-	struct device *dev = &pdev->dev;
-	struct dsa_switch *ds;
-	struct ocelot *ocelot;
-	struct felix *felix;
-	int err;
-
-	felix = devm_kzalloc(dev, sizeof(*felix), GFP_KERNEL);
-	if (!felix)
-		return -ENOMEM;
-
-	dev_set_drvdata(dev, felix);
-
-	ocelot = &felix->ocelot;
-	ocelot->dev = dev;
-
-	ocelot->num_flooding_pgids = 1;
-
-	felix->info = &vsc7512_info;
-
-	ds = devm_kzalloc(dev, sizeof(*ds), GFP_KERNEL);
-	if (!ds)
-		return -ENOMEM;
-
-	ds->dev = dev;
-	ds->num_ports = felix->info->num_ports;
-	ds->num_tx_queues = OCELOT_NUM_TC;
-
-	ds->ops = &felix_switch_ops;
-	ds->phylink_mac_ops = &felix_phylink_mac_ops;
-	ds->priv = ocelot;
-	felix->ds = ds;
-	felix->tag_proto = DSA_TAG_PROTO_OCELOT;
-
-	err = dsa_register_switch(ds);
-	if (err)
-		dev_err_probe(dev, err, "Failed to register DSA switch\n");
-
-	return err;
+	return felix_register_switch(&pdev->dev, 0, 1, false, false,
+				     DSA_TAG_PROTO_OCELOT, &vsc7512_info);
 }
 
 static void ocelot_ext_remove(struct platform_device *pdev)
diff --git a/drivers/net/dsa/ocelot/seville_vsc9953.c b/drivers/net/dsa/ocelot/seville_vsc9953.c
index 83bd0e1ee692..70782649c395 100644
--- a/drivers/net/dsa/ocelot/seville_vsc9953.c
+++ b/drivers/net/dsa/ocelot/seville_vsc9953.c
@@ -971,49 +971,17 @@ static const struct felix_info seville_info_vsc9953 = {
 static int seville_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
-	struct dsa_switch *ds;
-	struct ocelot *ocelot;
 	struct resource *res;
-	struct felix *felix;
-	int err;
-
-	felix = devm_kzalloc(dev, sizeof(struct felix), GFP_KERNEL);
-	if (!felix)
-		return -ENOMEM;
-
-	platform_set_drvdata(pdev, felix);
-
-	ocelot = &felix->ocelot;
-	ocelot->dev = dev;
-	ocelot->num_flooding_pgids = 1;
-	felix->info = &seville_info_vsc9953;
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (!res) {
 		dev_err(dev, "Invalid resource\n");
 		return -EINVAL;
 	}
-	felix->switch_base = res->start;
-
-	ds = devm_kzalloc(dev, sizeof(struct dsa_switch), GFP_KERNEL);
-	if (!ds)
-		return -ENOMEM;
-
-	ds->dev = dev;
-	ds->num_ports = felix->info->num_ports;
-	ds->num_tx_queues = OCELOT_NUM_TC;
-
-	ds->ops = &felix_switch_ops;
-	ds->phylink_mac_ops = &felix_phylink_mac_ops;
-	ds->priv = ocelot;
-	felix->ds = ds;
-	felix->tag_proto = DSA_TAG_PROTO_SEVILLE;
-
-	err = dsa_register_switch(ds);
-	if (err)
-		dev_err(dev, "Failed to register DSA switch: %d\n", err);
 
-	return err;
+	return felix_register_switch(dev, res->start, 1, false, false,
+				     DSA_TAG_PROTO_SEVILLE,
+				     &seville_info_vsc9953);
 }
 
 static void seville_remove(struct platform_device *pdev)
-- 
cgit v1.2.3-58-ga151


From a4303941c6f399816eac0a56c627008a41c9336d Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Thu, 30 May 2024 19:33:33 +0300
Subject: net: dsa: ocelot: unexport felix_phylink_mac_ops and felix_switch_ops

Now that the common felix_register_switch() from the umbrella driver
is the only entity that accesses these data structures, we can remove
them from the list of the exported symbols.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Sai Krishna <saikrishnag@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/ocelot/felix.c | 6 ++----
 drivers/net/dsa/ocelot/felix.h | 3 ---
 2 files changed, 2 insertions(+), 7 deletions(-)

diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index accf737f7b69..d12c4e85baa7 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -2106,15 +2106,14 @@ static void felix_get_mm_stats(struct dsa_switch *ds, int port,
 	ocelot_port_get_mm_stats(ocelot, port, stats);
 }
 
-const struct phylink_mac_ops felix_phylink_mac_ops = {
+static const struct phylink_mac_ops felix_phylink_mac_ops = {
 	.mac_select_pcs		= felix_phylink_mac_select_pcs,
 	.mac_config		= felix_phylink_mac_config,
 	.mac_link_down		= felix_phylink_mac_link_down,
 	.mac_link_up		= felix_phylink_mac_link_up,
 };
-EXPORT_SYMBOL_GPL(felix_phylink_mac_ops);
 
-const struct dsa_switch_ops felix_switch_ops = {
+static const struct dsa_switch_ops felix_switch_ops = {
 	.get_tag_protocol		= felix_get_tag_protocol,
 	.change_tag_protocol		= felix_change_tag_protocol,
 	.connect_tag_protocol		= felix_connect_tag_protocol,
@@ -2193,7 +2192,6 @@ const struct dsa_switch_ops felix_switch_ops = {
 	.port_set_host_flood		= felix_port_set_host_flood,
 	.port_change_conduit		= felix_port_change_conduit,
 };
-EXPORT_SYMBOL_GPL(felix_switch_ops);
 
 int felix_register_switch(struct device *dev, resource_size_t switch_base,
 			  int num_flooding_pgids, bool ptp,
diff --git a/drivers/net/dsa/ocelot/felix.h b/drivers/net/dsa/ocelot/felix.h
index 85b4f8616003..211991f494e3 100644
--- a/drivers/net/dsa/ocelot/felix.h
+++ b/drivers/net/dsa/ocelot/felix.h
@@ -82,9 +82,6 @@ struct felix_tag_proto_ops {
 			      struct netlink_ext_ack *extack);
 };
 
-extern const struct phylink_mac_ops felix_phylink_mac_ops;
-extern const struct dsa_switch_ops felix_switch_ops;
-
 /* DSA glue / front-end for struct ocelot */
 struct felix {
 	struct dsa_switch		*ds;
-- 
cgit v1.2.3-58-ga151


From 809055628bce824b7fe18331abb65e44d02b0ecf Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Thu, 30 May 2024 18:56:52 +0300
Subject: wifi: ath12k: add panic handler

Currently for ath12k PCI devices, firmware could be running in one of
several execution environments, e.g., PBL, SBL and mission mode etc. Among
which PBL is the only stage where PCIe link negotiation could happen. So
normally firmware runs in PBL in order to be enumerated during system reboot.

However it might not work in kernel crash scenario: ath12k target is not
found after warm reboot from kernel crash. This is because when kernel crashes,
ath12k host does nothing to firmware. And during warm reboot, WLAN power
is sustained. So firmware is likely to keep running in mission mode throughout
the bootup process. As a result PCIe link is not established and thus target
not enumerated.

So add a handler in panic notification list for ath12k. When kernel crashes,
this handler gets called and tries to reset target to PBL state. Then PCIe
link negotiation could happen and target gets enumerated.

This change applies to all PCI devices including WCN7850 and QCN9274.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240529021533.10861-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.c | 29 +++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/core.h |  3 +++
 drivers/net/wireless/ath/ath12k/hif.h  |  9 +++++++++
 drivers/net/wireless/ath/ath12k/pci.c  |  8 ++++++++
 4 files changed, 49 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c
index e7f628e935e4..4c3eab4686c2 100644
--- a/drivers/net/wireless/ath/ath12k/core.c
+++ b/drivers/net/wireless/ath/ath12k/core.c
@@ -1188,6 +1188,29 @@ int ath12k_core_pre_init(struct ath12k_base *ab)
 	return 0;
 }
 
+static int ath12k_core_panic_handler(struct notifier_block *nb,
+				     unsigned long action, void *data)
+{
+	struct ath12k_base *ab = container_of(nb, struct ath12k_base,
+					      panic_nb);
+
+	return ath12k_hif_panic_handler(ab);
+}
+
+static int ath12k_core_panic_notifier_register(struct ath12k_base *ab)
+{
+	ab->panic_nb.notifier_call = ath12k_core_panic_handler;
+
+	return atomic_notifier_chain_register(&panic_notifier_list,
+					      &ab->panic_nb);
+}
+
+static void ath12k_core_panic_notifier_unregister(struct ath12k_base *ab)
+{
+	atomic_notifier_chain_unregister(&panic_notifier_list,
+					 &ab->panic_nb);
+}
+
 int ath12k_core_init(struct ath12k_base *ab)
 {
 	int ret;
@@ -1198,11 +1221,17 @@ int ath12k_core_init(struct ath12k_base *ab)
 		return ret;
 	}
 
+	ret = ath12k_core_panic_notifier_register(ab);
+	if (ret)
+		ath12k_warn(ab, "failed to register panic handler: %d\n", ret);
+
 	return 0;
 }
 
 void ath12k_core_deinit(struct ath12k_base *ab)
 {
+	ath12k_core_panic_notifier_unregister(ab);
+
 	mutex_lock(&ab->core_lock);
 
 	ath12k_core_pdev_destroy(ab);
diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index 7d20b09c52e6..44ca30b67fa2 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -14,6 +14,7 @@
 #include <linux/dmi.h>
 #include <linux/ctype.h>
 #include <linux/firmware.h>
+#include <linux/panic_notifier.h>
 #include "qmi.h"
 #include "htc.h"
 #include "wmi.h"
@@ -924,6 +925,8 @@ struct ath12k_base {
 
 #endif /* CONFIG_ACPI */
 
+	struct notifier_block panic_nb;
+
 	/* must be last */
 	u8 drv_priv[] __aligned(sizeof(void *));
 };
diff --git a/drivers/net/wireless/ath/ath12k/hif.h b/drivers/net/wireless/ath/ath12k/hif.h
index 7f0926fe751d..0e53ec269fa4 100644
--- a/drivers/net/wireless/ath/ath12k/hif.h
+++ b/drivers/net/wireless/ath/ath12k/hif.h
@@ -30,6 +30,7 @@ struct ath12k_hif_ops {
 	void (*ce_irq_enable)(struct ath12k_base *ab);
 	void (*ce_irq_disable)(struct ath12k_base *ab);
 	void (*get_ce_msi_idx)(struct ath12k_base *ab, u32 ce_id, u32 *msi_idx);
+	int (*panic_handler)(struct ath12k_base *ab);
 };
 
 static inline int ath12k_hif_map_service_to_pipe(struct ath12k_base *ab, u16 service_id,
@@ -147,4 +148,12 @@ static inline void ath12k_hif_power_down(struct ath12k_base *ab, bool is_suspend
 	ab->hif.ops->power_down(ab, is_suspend);
 }
 
+static inline int ath12k_hif_panic_handler(struct ath12k_base *ab)
+{
+	if (!ab->hif.ops->panic_handler)
+		return NOTIFY_DONE;
+
+	return ab->hif.ops->panic_handler(ab);
+}
+
 #endif /* ATH12K_HIF_H */
diff --git a/drivers/net/wireless/ath/ath12k/pci.c b/drivers/net/wireless/ath/ath12k/pci.c
index 11b95d037051..876c029f58f6 100644
--- a/drivers/net/wireless/ath/ath12k/pci.c
+++ b/drivers/net/wireless/ath/ath12k/pci.c
@@ -1302,6 +1302,13 @@ void ath12k_pci_power_down(struct ath12k_base *ab, bool is_suspend)
 	ath12k_pci_sw_reset(ab_pci->ab, false);
 }
 
+static int ath12k_pci_panic_handler(struct ath12k_base *ab)
+{
+	ath12k_pci_sw_reset(ab, false);
+
+	return NOTIFY_OK;
+}
+
 static const struct ath12k_hif_ops ath12k_pci_hif_ops = {
 	.start = ath12k_pci_start,
 	.stop = ath12k_pci_stop,
@@ -1319,6 +1326,7 @@ static const struct ath12k_hif_ops ath12k_pci_hif_ops = {
 	.ce_irq_enable = ath12k_pci_hif_ce_irq_enable,
 	.ce_irq_disable = ath12k_pci_hif_ce_irq_disable,
 	.get_ce_msi_idx = ath12k_pci_get_ce_msi_idx,
+	.panic_handler = ath12k_pci_panic_handler,
 };
 
 static
-- 
cgit v1.2.3-58-ga151


From 22767241e64427bbcffea2e2d51531ab467bc4ef Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Thu, 30 May 2024 18:56:52 +0300
Subject: wifi: ath12k: add hw_link_id in ath12k_pdev

Currently, hw_link_id is sent in WMI service ready event but it
is not parsed anywhere.

But, in future, for multi-link operation, this parameter would be
needed by many WMI commands such as WMI beacon template
(WMI_BCN_TMPL_CMDID), WMI vdev start for Multi-link virtual AP
interfaces (WMI_VDEV_START_REQUEST_CMDID), WMI peer assoc command
(WMI_PEER_ASSOC_CMDID) for Multi-link peer and so on.

Hence, add changes to parse and store the hw_link_id received in
WMI service ready event in ath12k_pdev structure.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Signed-off-by: Harshitha Prem <quic_hprem@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240529054955.4105240-1-quic_hprem@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.h | 1 +
 drivers/net/wireless/ath/ath12k/mac.c  | 2 +-
 drivers/net/wireless/ath/ath12k/wmi.c  | 1 +
 3 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index 44ca30b67fa2..d03326a68a9f 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -690,6 +690,7 @@ struct mlo_timestamp {
 struct ath12k_pdev {
 	struct ath12k *ar;
 	u32 pdev_id;
+	u32 hw_link_id;
 	struct ath12k_pdev_cap cap;
 	u8 mac_addr[ETH_ALEN];
 	struct mlo_timestamp timestamp;
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 784964ae03ec..509c02bffdae 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -9229,7 +9229,7 @@ static struct ath12k_hw *ath12k_mac_hw_allocate(struct ath12k_base *ab,
 		ar = ath12k_ah_to_ar(ah, i);
 		ar->ah = ah;
 		ar->ab = ab;
-		ar->hw_link_id = i;
+		ar->hw_link_id = pdev->hw_link_id;
 		ar->pdev = pdev;
 		ar->pdev_idx = pdev_idx;
 		pdev->ar = ar;
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 99106b088311..d6e1d1398cdb 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -500,6 +500,7 @@ ath12k_pull_mac_phy_cap_svc_ready_ext(struct ath12k_wmi_pdev *wmi_handle,
 	mac_caps = wmi_mac_phy_caps + phy_idx;
 
 	pdev->pdev_id = ath12k_wmi_mac_phy_get_pdev_id(mac_caps);
+	pdev->hw_link_id = ath12k_wmi_mac_phy_get_hw_link_id(mac_caps);
 	pdev_cap->supported_bands |= le32_to_cpu(mac_caps->supported_bands);
 	pdev_cap->ampdu_density = le32_to_cpu(mac_caps->ampdu_density);
 
-- 
cgit v1.2.3-58-ga151


From 96a27ee76f0e95b56f94b6902da7a5ebef372612 Mon Sep 17 00:00:00 2001
From: Xiao Wang <xiao.w.wang@intel.com>
Date: Fri, 24 May 2024 15:55:43 +0800
Subject: riscv, bpf: Introduce shift add helper with Zba optimization
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Zba extension is very useful for generating addresses that index into array
of basic data types. This patch introduces sh2add and sh3add helpers for
RV32 and RV64 respectively, to accelerate addressing for array of unsigned
long data.

Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Björn Töpel <bjorn@kernel.org>
Link: https://lore.kernel.org/bpf/20240524075543.4050464-3-xiao.w.wang@intel.com
---
 arch/riscv/net/bpf_jit.h        | 33 +++++++++++++++++++++++++++++++++
 arch/riscv/net/bpf_jit_comp32.c |  3 +--
 arch/riscv/net/bpf_jit_comp64.c |  9 +++------
 3 files changed, 37 insertions(+), 8 deletions(-)

diff --git a/arch/riscv/net/bpf_jit.h b/arch/riscv/net/bpf_jit.h
index 97041b58237a..1d1c78d4cff1 100644
--- a/arch/riscv/net/bpf_jit.h
+++ b/arch/riscv/net/bpf_jit.h
@@ -742,6 +742,17 @@ static inline u16 rvc_swsp(u32 imm8, u8 rs2)
 	return rv_css_insn(0x6, imm, rs2, 0x2);
 }
 
+/* RVZBA instructions. */
+static inline u32 rvzba_sh2add(u8 rd, u8 rs1, u8 rs2)
+{
+	return rv_r_insn(0x10, rs2, rs1, 0x4, rd, 0x33);
+}
+
+static inline u32 rvzba_sh3add(u8 rd, u8 rs1, u8 rs2)
+{
+	return rv_r_insn(0x10, rs2, rs1, 0x6, rd, 0x33);
+}
+
 /* RVZBB instructions. */
 static inline u32 rvzbb_sextb(u8 rd, u8 rs1)
 {
@@ -1095,6 +1106,28 @@ static inline void emit_sw(u8 rs1, s32 off, u8 rs2, struct rv_jit_context *ctx)
 		emit(rv_sw(rs1, off, rs2), ctx);
 }
 
+static inline void emit_sh2add(u8 rd, u8 rs1, u8 rs2, struct rv_jit_context *ctx)
+{
+	if (rvzba_enabled()) {
+		emit(rvzba_sh2add(rd, rs1, rs2), ctx);
+		return;
+	}
+
+	emit_slli(rd, rs1, 2, ctx);
+	emit_add(rd, rd, rs2, ctx);
+}
+
+static inline void emit_sh3add(u8 rd, u8 rs1, u8 rs2, struct rv_jit_context *ctx)
+{
+	if (rvzba_enabled()) {
+		emit(rvzba_sh3add(rd, rs1, rs2), ctx);
+		return;
+	}
+
+	emit_slli(rd, rs1, 3, ctx);
+	emit_add(rd, rd, rs2, ctx);
+}
+
 /* RV64-only helper functions. */
 #if __riscv_xlen == 64
 
diff --git a/arch/riscv/net/bpf_jit_comp32.c b/arch/riscv/net/bpf_jit_comp32.c
index f5ba73bb153d..592dd86fbf81 100644
--- a/arch/riscv/net/bpf_jit_comp32.c
+++ b/arch/riscv/net/bpf_jit_comp32.c
@@ -811,8 +811,7 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx)
 	 * if (!prog)
 	 *   goto out;
 	 */
-	emit(rv_slli(RV_REG_T0, lo(idx_reg), 2), ctx);
-	emit(rv_add(RV_REG_T0, RV_REG_T0, lo(arr_reg)), ctx);
+	emit_sh2add(RV_REG_T0, lo(idx_reg), lo(arr_reg), ctx);
 	off = offsetof(struct bpf_array, ptrs);
 	if (is_12b_check(off, insn))
 		return -1;
diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c
index 2bc4c14ea59a..d5cebb0b0afe 100644
--- a/arch/riscv/net/bpf_jit_comp64.c
+++ b/arch/riscv/net/bpf_jit_comp64.c
@@ -380,8 +380,7 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx)
 	 * if (!prog)
 	 *     goto out;
 	 */
-	emit_slli(RV_REG_T2, RV_REG_A2, 3, ctx);
-	emit_add(RV_REG_T2, RV_REG_T2, RV_REG_A1, ctx);
+	emit_sh3add(RV_REG_T2, RV_REG_A2, RV_REG_A1, ctx);
 	off = offsetof(struct bpf_array, ptrs);
 	if (is_12b_check(off, insn))
 		return -1;
@@ -1099,12 +1098,10 @@ int bpf_jit_emit_insn(const struct bpf_insn *insn, struct rv_jit_context *ctx,
 			/* Load current CPU number in T1 */
 			emit_ld(RV_REG_T1, offsetof(struct thread_info, cpu),
 				RV_REG_TP, ctx);
-			/* << 3 because offsets are 8 bytes */
-			emit_slli(RV_REG_T1, RV_REG_T1, 3, ctx);
 			/* Load address of __per_cpu_offset array in T2 */
 			emit_addr(RV_REG_T2, (u64)&__per_cpu_offset, extra_pass, ctx);
-			/* Add offset of current CPU to  __per_cpu_offset */
-			emit_add(RV_REG_T1, RV_REG_T2, RV_REG_T1, ctx);
+			/* Get address of __per_cpu_offset[cpu] in T1 */
+			emit_sh3add(RV_REG_T1, RV_REG_T1, RV_REG_T2, ctx);
 			/* Load __per_cpu_offset[cpu] in T1 */
 			emit_ld(RV_REG_T1, 0, RV_REG_T1, ctx);
 			/* Add the offset to Rd */
-- 
cgit v1.2.3-58-ga151


From dfa7c9ffa607235119e029b70ced72f29059f8f3 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Mon, 3 Jun 2024 00:41:10 +0100
Subject: selftests/bpf: Remove unused struct 'scale_test_def'

'scale_test_def' is unused since commit 3762a39ce85f ("selftests/bpf: Split out
bpf_verif_scale selftests into multiple tests"). Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240602234112.225107-2-linux@treblig.org
---
 tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c b/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c
index 4c6ada5b270b..73f669014b69 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_verif_scale.c
@@ -45,12 +45,6 @@ err_out:
 	return err;
 }
 
-struct scale_test_def {
-	const char *file;
-	enum bpf_prog_type attach_type;
-	bool fails;
-};
-
 static void scale_test(const char *file,
 		       enum bpf_prog_type attach_type,
 		       bool should_fail)
-- 
cgit v1.2.3-58-ga151


From 3f67639d8e582c89c79549c619b22a00dd330e4e Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Mon, 3 Jun 2024 00:41:11 +0100
Subject: selftests/bpf: Remove unused 'key_t' structs

'key_t' is unused in a couple of files since the original commit 60dd49ea6539
("selftests/bpf: Add test for bpf array map iterators"). Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240602234112.225107-3-linux@treblig.org
---
 tools/testing/selftests/bpf/progs/bpf_iter_bpf_array_map.c        | 6 ------
 tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_array_map.c | 6 ------
 2 files changed, 12 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_bpf_array_map.c b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_array_map.c
index c5969ca6f26b..564835ba7d51 100644
--- a/tools/testing/selftests/bpf/progs/bpf_iter_bpf_array_map.c
+++ b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_array_map.c
@@ -6,12 +6,6 @@
 
 char _license[] SEC("license") = "GPL";
 
-struct key_t {
-	int a;
-	int b;
-	int c;
-};
-
 struct {
 	__uint(type, BPF_MAP_TYPE_ARRAY);
 	__uint(max_entries, 3);
diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_array_map.c b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_array_map.c
index 85fa710fad90..9f0e0705b2bf 100644
--- a/tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_array_map.c
+++ b/tools/testing/selftests/bpf/progs/bpf_iter_bpf_percpu_array_map.c
@@ -6,12 +6,6 @@
 
 char _license[] SEC("license") = "GPL";
 
-struct key_t {
-	int a;
-	int b;
-	int c;
-};
-
 struct {
 	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
 	__uint(max_entries, 3);
-- 
cgit v1.2.3-58-ga151


From a450d36b05fa225b071ce9fbf522544caea06594 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Mon, 3 Jun 2024 00:41:12 +0100
Subject: selftests/bpf: Remove unused struct 'libcap'

'libcap' is unused since commit b1c2768a82b9 ("bpf: selftests: Remove libcap
usage from test_verifier"). Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240602234112.225107-4-linux@treblig.org
---
 tools/testing/selftests/bpf/test_verifier.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_verifier.c b/tools/testing/selftests/bpf/test_verifier.c
index df04bda1c927..610392dfc4fb 100644
--- a/tools/testing/selftests/bpf/test_verifier.c
+++ b/tools/testing/selftests/bpf/test_verifier.c
@@ -1237,11 +1237,6 @@ static void do_test_fixup(struct bpf_test *test, enum bpf_prog_type prog_type,
 	fixup_prog_kfuncs(prog, fd_array, test->fixup_kfunc_btf_id);
 }
 
-struct libcap {
-	struct __user_cap_header_struct hdr;
-	struct __user_cap_data_struct data[2];
-};
-
 static int set_admin(bool admin)
 {
 	int err;
-- 
cgit v1.2.3-58-ga151


From ce5249b91e34d81255c00950d415ebd4c3cae8d4 Mon Sep 17 00:00:00 2001
From: Swan Beaujard <beaujardswan@gmail.com>
Date: Mon, 3 Jun 2024 00:58:12 +0200
Subject: bpftool: Fix typo in MAX_NUM_METRICS macro name

Correct typo in bpftool profiler and change all instances of 'MATRICS' to
'METRICS' in the profiler.bpf.c file.

Signed-off-by: Swan Beaujard <beaujardswan@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Quentin Monnet <qmo@kernel.org>
Link: https://lore.kernel.org/bpf/20240602225812.81171-1-beaujardswan@gmail.com
---
 tools/bpf/bpftool/skeleton/profiler.bpf.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/tools/bpf/bpftool/skeleton/profiler.bpf.c b/tools/bpf/bpftool/skeleton/profiler.bpf.c
index 2f80edc682f1..f48c783cb9f7 100644
--- a/tools/bpf/bpftool/skeleton/profiler.bpf.c
+++ b/tools/bpf/bpftool/skeleton/profiler.bpf.c
@@ -40,17 +40,17 @@ struct {
 
 const volatile __u32 num_cpu = 1;
 const volatile __u32 num_metric = 1;
-#define MAX_NUM_MATRICS 4
+#define MAX_NUM_METRICS 4
 
 SEC("fentry/XXX")
 int BPF_PROG(fentry_XXX)
 {
-	struct bpf_perf_event_value___local *ptrs[MAX_NUM_MATRICS];
+	struct bpf_perf_event_value___local *ptrs[MAX_NUM_METRICS];
 	u32 key = bpf_get_smp_processor_id();
 	u32 i;
 
 	/* look up before reading, to reduce error */
-	for (i = 0; i < num_metric && i < MAX_NUM_MATRICS; i++) {
+	for (i = 0; i < num_metric && i < MAX_NUM_METRICS; i++) {
 		u32 flag = i;
 
 		ptrs[i] = bpf_map_lookup_elem(&fentry_readings, &flag);
@@ -58,7 +58,7 @@ int BPF_PROG(fentry_XXX)
 			return 0;
 	}
 
-	for (i = 0; i < num_metric && i < MAX_NUM_MATRICS; i++) {
+	for (i = 0; i < num_metric && i < MAX_NUM_METRICS; i++) {
 		struct bpf_perf_event_value___local reading;
 		int err;
 
@@ -99,14 +99,14 @@ fexit_update_maps(u32 id, struct bpf_perf_event_value___local *after)
 SEC("fexit/XXX")
 int BPF_PROG(fexit_XXX)
 {
-	struct bpf_perf_event_value___local readings[MAX_NUM_MATRICS];
+	struct bpf_perf_event_value___local readings[MAX_NUM_METRICS];
 	u32 cpu = bpf_get_smp_processor_id();
 	u32 i, zero = 0;
 	int err;
 	u64 *count;
 
 	/* read all events before updating the maps, to reduce error */
-	for (i = 0; i < num_metric && i < MAX_NUM_MATRICS; i++) {
+	for (i = 0; i < num_metric && i < MAX_NUM_METRICS; i++) {
 		err = bpf_perf_event_read_value(&events, cpu + i * num_cpu,
 						(void *)(readings + i),
 						sizeof(*readings));
@@ -116,7 +116,7 @@ int BPF_PROG(fexit_XXX)
 	count = bpf_map_lookup_elem(&counts, &zero);
 	if (count) {
 		*count += 1;
-		for (i = 0; i < num_metric && i < MAX_NUM_MATRICS; i++)
+		for (i = 0; i < num_metric && i < MAX_NUM_METRICS; i++)
 			fexit_update_maps(i, &readings[i]);
 	}
 	return 0;
-- 
cgit v1.2.3-58-ga151


From ec1249d3278183d419276b9a7fe73591cd3dd505 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Fri, 31 May 2024 09:28:43 -0700
Subject: test_bpf: Add missing MODULE_DESCRIPTION()

make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in lib/test_bpf.o

Add the missing invocation of the MODULE_DESCRIPTION() macro.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240531-md-lib-test_bpf-v1-1-868e4bd2f9ed@quicinc.com
---
 lib/test_bpf.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/lib/test_bpf.c b/lib/test_bpf.c
index 207ff87194db..ce5716c3999a 100644
--- a/lib/test_bpf.c
+++ b/lib/test_bpf.c
@@ -15706,4 +15706,5 @@ static void __exit test_bpf_exit(void)
 module_init(test_bpf_init);
 module_exit(test_bpf_exit);
 
+MODULE_DESCRIPTION("Testsuite for BPF interpreter and BPF JIT compiler");
 MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-58-ga151


From d95ba15b97847f4ae520db83bd98b61d50fb3068 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 23 May 2024 14:49:57 +0800
Subject: selftests/bpf: Fix tx_prog_fd values in test_sockmap

The values of tx_prog_fd in run_options() should not be 0, so set it as -1
in else branch, and test it using "if (tx_prog_fd > 0)" condition, not
"if (tx_prog_fd)" or "if (tx_prog_fd >= 0)".

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/08b20ffc544324d40939efeae93800772a91a58e.1716446893.git.tanggeliang@kylinos.cn
---
 tools/testing/selftests/bpf/test_sockmap.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index ddc6a9cef36f..892a690c4e7b 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -1027,9 +1027,9 @@ run:
 	else if (txmsg_drop)
 		tx_prog_fd = prog_fd[8];
 	else
-		tx_prog_fd = 0;
+		tx_prog_fd = -1;
 
-	if (tx_prog_fd) {
+	if (tx_prog_fd > 0) {
 		int redir_fd, i = 0;
 
 		err = bpf_prog_attach(tx_prog_fd,
@@ -1285,7 +1285,7 @@ out:
 	bpf_prog_detach2(prog_fd[0], map_fd[8], BPF_SK_SKB_STREAM_PARSER);
 	bpf_prog_detach2(prog_fd[2], map_fd[8], BPF_SK_SKB_STREAM_VERDICT);
 
-	if (tx_prog_fd >= 0)
+	if (tx_prog_fd > 0)
 		bpf_prog_detach2(tx_prog_fd, map_fd[1], BPF_SK_MSG_VERDICT);
 
 	for (i = 0; i < 8; i++) {
-- 
cgit v1.2.3-58-ga151


From a9f0ea175948c21640ae1cc145e679db7fc45fa6 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 23 May 2024 14:49:58 +0800
Subject: selftests/bpf: Drop duplicate definition of i in test_sockmap

There's already a definition of i in run_options() at the beginning, no
need to define a new one in "if (tx_prog_fd > 0)" block.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/8d690682330a59361562bca75d6903253d16f312.1716446893.git.tanggeliang@kylinos.cn
---
 tools/testing/selftests/bpf/test_sockmap.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index 892a690c4e7b..6d724fea59fb 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -1030,7 +1030,7 @@ run:
 		tx_prog_fd = -1;
 
 	if (tx_prog_fd > 0) {
-		int redir_fd, i = 0;
+		int redir_fd;
 
 		err = bpf_prog_attach(tx_prog_fd,
 				      map_fd[1], BPF_SK_MSG_VERDICT, 0);
@@ -1041,6 +1041,7 @@ run:
 			goto out;
 		}
 
+		i = 0;
 		err = bpf_map_update_elem(map_fd[1], &i, &c1, BPF_ANY);
 		if (err) {
 			fprintf(stderr,
-- 
cgit v1.2.3-58-ga151


From 3f32a115f61d31049e3e91d469bca849f712a979 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 23 May 2024 14:49:59 +0800
Subject: selftests/bpf: Use bpf_link attachments in test_sockmap

Switch attachments to bpf_link using bpf_program__attach_sockmap() instead
of bpf_prog_attach().

This patch adds a new array progs[] to replace prog_fd[] array, set in
populate_progs() for each program in bpf object.

And another new array links[] to save the attached bpf_link. It is
initalized as NULL in populate_progs, set as the return valuses of
bpf_program__attach_sockmap(), and detached by bpf_link__detach().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/32cf8376a810e2e9c719f8e4cfb97132ed2d1f9c.1716446893.git.tanggeliang@kylinos.cn
---
 tools/testing/selftests/bpf/test_sockmap.c | 59 ++++++++++++++++--------------
 1 file changed, 31 insertions(+), 28 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index 6d724fea59fb..81e7a57c727c 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -64,6 +64,8 @@ int failed;
 int map_fd[9];
 struct bpf_map *maps[9];
 int prog_fd[9];
+struct bpf_program *progs[9];
+struct bpf_link *links[9];
 
 int txmsg_pass;
 int txmsg_redir;
@@ -960,43 +962,39 @@ static int run_options(struct sockmap_options *options, int cg_fd,  int test)
 
 	/* Attach programs to sockmap */
 	if (!txmsg_omit_skb_parser) {
-		err = bpf_prog_attach(prog_fd[0], map_fd[0],
-				      BPF_SK_SKB_STREAM_PARSER, 0);
-		if (err) {
+		links[0] = bpf_program__attach_sockmap(progs[0], map_fd[0]);
+		if (!links[0]) {
 			fprintf(stderr,
-				"ERROR: bpf_prog_attach (sockmap %i->%i): %d (%s)\n",
-				prog_fd[0], map_fd[0], err, strerror(errno));
-			return err;
+				"ERROR: bpf_program__attach_sockmap (sockmap %i->%i): (%s)\n",
+				bpf_program__fd(progs[0]), map_fd[0], strerror(errno));
+			return -1;
 		}
 	}
 
-	err = bpf_prog_attach(prog_fd[1], map_fd[0],
-				BPF_SK_SKB_STREAM_VERDICT, 0);
-	if (err) {
-		fprintf(stderr, "ERROR: bpf_prog_attach (sockmap): %d (%s)\n",
-			err, strerror(errno));
-		return err;
+	links[1] = bpf_program__attach_sockmap(progs[1], map_fd[0]);
+	if (!links[1]) {
+		fprintf(stderr, "ERROR: bpf_program__attach_sockmap (sockmap): (%s)\n",
+			strerror(errno));
+		return -1;
 	}
 
 	/* Attach programs to TLS sockmap */
 	if (txmsg_ktls_skb) {
 		if (!txmsg_omit_skb_parser) {
-			err = bpf_prog_attach(prog_fd[0], map_fd[8],
-					      BPF_SK_SKB_STREAM_PARSER, 0);
-			if (err) {
+			links[2] = bpf_program__attach_sockmap(progs[0], map_fd[8]);
+			if (!links[2]) {
 				fprintf(stderr,
-					"ERROR: bpf_prog_attach (TLS sockmap %i->%i): %d (%s)\n",
-					prog_fd[0], map_fd[8], err, strerror(errno));
-				return err;
+					"ERROR: bpf_program__attach_sockmap (TLS sockmap %i->%i): (%s)\n",
+					bpf_program__fd(progs[0]), map_fd[8], strerror(errno));
+				return -1;
 			}
 		}
 
-		err = bpf_prog_attach(prog_fd[2], map_fd[8],
-				      BPF_SK_SKB_STREAM_VERDICT, 0);
-		if (err) {
-			fprintf(stderr, "ERROR: bpf_prog_attach (TLS sockmap): %d (%s)\n",
-				err, strerror(errno));
-			return err;
+		links[3] = bpf_program__attach_sockmap(progs[2], map_fd[8]);
+		if (!links[3]) {
+			fprintf(stderr, "ERROR: bpf_program__attach_sockmap (TLS sockmap): (%s)\n",
+				strerror(errno));
+			return -1;
 		}
 	}
 
@@ -1281,10 +1279,11 @@ run:
 out:
 	/* Detatch and zero all the maps */
 	bpf_prog_detach2(prog_fd[3], cg_fd, BPF_CGROUP_SOCK_OPS);
-	bpf_prog_detach2(prog_fd[0], map_fd[0], BPF_SK_SKB_STREAM_PARSER);
-	bpf_prog_detach2(prog_fd[1], map_fd[0], BPF_SK_SKB_STREAM_VERDICT);
-	bpf_prog_detach2(prog_fd[0], map_fd[8], BPF_SK_SKB_STREAM_PARSER);
-	bpf_prog_detach2(prog_fd[2], map_fd[8], BPF_SK_SKB_STREAM_VERDICT);
+
+	for (i = 0; i < ARRAY_SIZE(links); i++) {
+		if (links[i])
+			bpf_link__detach(links[i]);
+	}
 
 	if (tx_prog_fd > 0)
 		bpf_prog_detach2(tx_prog_fd, map_fd[1], BPF_SK_MSG_VERDICT);
@@ -1805,6 +1804,7 @@ static int populate_progs(char *bpf_file)
 	i = bpf_object__load(obj);
 	i = 0;
 	bpf_object__for_each_program(prog, obj) {
+		progs[i] = prog;
 		prog_fd[i] = bpf_program__fd(prog);
 		i++;
 	}
@@ -1819,6 +1819,9 @@ static int populate_progs(char *bpf_file)
 		}
 	}
 
+	for (i = 0; i < ARRAY_SIZE(links); i++)
+		links[i] = NULL;
+
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From 24bb90a42633ea47256d4f13289dd3181236e028 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 23 May 2024 14:50:00 +0800
Subject: selftests/bpf: Replace tx_prog_fd with tx_prog in test_sockmap

bpf_program__attach_sockmap() needs to take a parameter of type bpf_program
instead of an fd, so tx_prog_fd becomes useless. This patch uses a pointer
tx_prog to point to an item in progs[] array.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/23b37f932c547dd1ebfe154bbc0b0e957be21ee6.1716446893.git.tanggeliang@kylinos.cn
---
 tools/testing/selftests/bpf/test_sockmap.c | 30 ++++++++++++++----------------
 1 file changed, 14 insertions(+), 16 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index 81e7a57c727c..c185c4242abb 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -954,7 +954,8 @@ enum {
 
 static int run_options(struct sockmap_options *options, int cg_fd,  int test)
 {
-	int i, key, next_key, err, tx_prog_fd = -1, zero = 0;
+	int i, key, next_key, err, zero = 0;
+	struct bpf_program *tx_prog;
 
 	/* If base test skip BPF setup */
 	if (test == BASE || test == BASE_SENDPAGE)
@@ -1015,27 +1016,27 @@ run:
 
 	/* Attach txmsg program to sockmap */
 	if (txmsg_pass)
-		tx_prog_fd = prog_fd[4];
+		tx_prog = progs[4];
 	else if (txmsg_redir)
-		tx_prog_fd = prog_fd[5];
+		tx_prog = progs[5];
 	else if (txmsg_apply)
-		tx_prog_fd = prog_fd[6];
+		tx_prog = progs[6];
 	else if (txmsg_cork)
-		tx_prog_fd = prog_fd[7];
+		tx_prog = progs[7];
 	else if (txmsg_drop)
-		tx_prog_fd = prog_fd[8];
+		tx_prog = progs[8];
 	else
-		tx_prog_fd = -1;
+		tx_prog = NULL;
 
-	if (tx_prog_fd > 0) {
+	if (tx_prog) {
 		int redir_fd;
 
-		err = bpf_prog_attach(tx_prog_fd,
-				      map_fd[1], BPF_SK_MSG_VERDICT, 0);
-		if (err) {
+		links[4] = bpf_program__attach_sockmap(tx_prog, map_fd[1]);
+		if (!links[4]) {
 			fprintf(stderr,
-				"ERROR: bpf_prog_attach (txmsg): %d (%s)\n",
-				err, strerror(errno));
+				"ERROR: bpf_program__attach_sockmap (txmsg): (%s)\n",
+				strerror(errno));
+			err = -1;
 			goto out;
 		}
 
@@ -1285,9 +1286,6 @@ out:
 			bpf_link__detach(links[i]);
 	}
 
-	if (tx_prog_fd > 0)
-		bpf_prog_detach2(tx_prog_fd, map_fd[1], BPF_SK_MSG_VERDICT);
-
 	for (i = 0; i < 8; i++) {
 		key = next_key = 0;
 		bpf_map_update_elem(map_fd[i], &key, &zero, BPF_ANY);
-- 
cgit v1.2.3-58-ga151


From 467a0c79b5514d7301ae679770380679a8e32668 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 23 May 2024 14:50:01 +0800
Subject: selftests/bpf: Drop prog_fd array in test_sockmap

The program fds can be got by using bpf_program__fd(progs[]), then
prog_fd becomes useless. This patch drops it.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/9a6335e4d8dbab23c0d8906074457ceddd61e74b.1716446893.git.tanggeliang@kylinos.cn
---
 tools/testing/selftests/bpf/test_sockmap.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index c185c4242abb..64b38a1a1b60 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -63,7 +63,6 @@ int passed;
 int failed;
 int map_fd[9];
 struct bpf_map *maps[9];
-int prog_fd[9];
 struct bpf_program *progs[9];
 struct bpf_link *links[9];
 
@@ -1000,7 +999,7 @@ static int run_options(struct sockmap_options *options, int cg_fd,  int test)
 	}
 
 	/* Attach to cgroups */
-	err = bpf_prog_attach(prog_fd[3], cg_fd, BPF_CGROUP_SOCK_OPS, 0);
+	err = bpf_prog_attach(bpf_program__fd(progs[3]), cg_fd, BPF_CGROUP_SOCK_OPS, 0);
 	if (err) {
 		fprintf(stderr, "ERROR: bpf_prog_attach (groups): %d (%s)\n",
 			err, strerror(errno));
@@ -1279,7 +1278,7 @@ run:
 		fprintf(stderr, "unknown test\n");
 out:
 	/* Detatch and zero all the maps */
-	bpf_prog_detach2(prog_fd[3], cg_fd, BPF_CGROUP_SOCK_OPS);
+	bpf_prog_detach2(bpf_program__fd(progs[3]), cg_fd, BPF_CGROUP_SOCK_OPS);
 
 	for (i = 0; i < ARRAY_SIZE(links); i++) {
 		if (links[i])
@@ -1803,7 +1802,6 @@ static int populate_progs(char *bpf_file)
 	i = 0;
 	bpf_object__for_each_program(prog, obj) {
 		progs[i] = prog;
-		prog_fd[i] = bpf_program__fd(prog);
 		i++;
 	}
 
-- 
cgit v1.2.3-58-ga151


From dcb681b659f2a0a546752730c9daa92dc6120d52 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 23 May 2024 14:50:02 +0800
Subject: selftests/bpf: Fix size of map_fd in test_sockmap

The array size of map_fd[] is 9, not 8. This patch changes it as a more
general form: ARRAY_SIZE(map_fd).

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/0972529ee01ebf8a8fd2b310bdec90831c94be77.1716446893.git.tanggeliang@kylinos.cn
---
 tools/testing/selftests/bpf/test_sockmap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index 64b38a1a1b60..e29f8203bd4f 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -1285,7 +1285,7 @@ out:
 			bpf_link__detach(links[i]);
 	}
 
-	for (i = 0; i < 8; i++) {
+	for (i = 0; i < ARRAY_SIZE(map_fd); i++) {
 		key = next_key = 0;
 		bpf_map_update_elem(map_fd[i], &key, &zero, BPF_ANY);
 		while (bpf_map_get_next_key(map_fd[i], &key, &next_key) == 0) {
-- 
cgit v1.2.3-58-ga151


From de1b5ea789dc28066cc8dc634b6825bd6148f38b Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 23 May 2024 14:50:03 +0800
Subject: selftests/bpf: Check length of recv in test_sockmap

The value of recv in msg_loop may be negative, like EWOULDBLOCK, so it's
necessary to check if it is positive before accumulating it to bytes_recvd.

Fixes: 16962b2404ac ("bpf: sockmap, add selftests")
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/5172563f7c7b2a2e953cef02e89fc34664a7b190.1716446893.git.tanggeliang@kylinos.cn
---
 tools/testing/selftests/bpf/test_sockmap.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index e29f8203bd4f..9cba4ec844a5 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -681,7 +681,8 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 				}
 			}
 
-			s->bytes_recvd += recv;
+			if (recv > 0)
+				s->bytes_recvd += recv;
 
 			if (opt->check_recved_len && s->bytes_recvd > total_bytes) {
 				errno = EMSGSIZE;
-- 
cgit v1.2.3-58-ga151


From 49784c7979321c49a8055f5c588d24c34a8c55fc Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 23 May 2024 14:50:04 +0800
Subject: selftests/bpf: Drop duplicate bpf_map_lookup_elem in test_sockmap

bpf_map_lookup_elem is invoked in bpf_prog3() already, no need to invoke
it again. This patch drops it.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/ea8458462b876ee445173e3effb535fd126137ed.1716446893.git.tanggeliang@kylinos.cn
---
 tools/testing/selftests/bpf/progs/test_sockmap_kern.h | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/test_sockmap_kern.h b/tools/testing/selftests/bpf/progs/test_sockmap_kern.h
index 3dff0813730b..f48f85f1bd70 100644
--- a/tools/testing/selftests/bpf/progs/test_sockmap_kern.h
+++ b/tools/testing/selftests/bpf/progs/test_sockmap_kern.h
@@ -177,9 +177,6 @@ int bpf_prog3(struct __sk_buff *skb)
 		return bpf_sk_redirect_hash(skb, &tls_sock_map, &ret, flags);
 #endif
 	}
-	f = bpf_map_lookup_elem(&sock_skb_opts, &one);
-	if (f && *f)
-		ret = SK_DROP;
 	err = bpf_skb_adjust_room(skb, 4, 0, 0);
 	if (err)
 		return SK_DROP;
-- 
cgit v1.2.3-58-ga151


From 4933b066fefbee4f1d2d708de53c4ab7f09026ad Mon Sep 17 00:00:00 2001
From: Douglas Anderson <dianders@chromium.org>
Date: Thu, 30 May 2024 16:43:08 -0700
Subject: r8152: If inaccessible at resume time, issue a reset

If we happened to get a USB transfer error during the transition to
suspend then the usb_queue_reset_device() that r8152_control_msg()
calls will get dropped on the floor. This is because
usb_lock_device_for_reset() (which usb_queue_reset_device() uses)
silently fails if it's called when a device is suspended or if too
much time passes.

Let's resolve this by resetting the device ourselves in r8152's
resume() function.

NOTE: due to timing, it's _possible_ that we could end up with two USB
resets: the one queued previously and the one called from the resume()
patch. This didn't happen in test cases I ran, though it's conceivably
possible. We can't easily know if this happened since
usb_queue_reset_device() can just silently drop the reset request. In
any case, it's not expected that this is a problem since the two
resets can't run at the same time (because of the device lock) and it
should be OK to reset the device twice. If somehow the double-reset
causes problems we could prevent resets from being queued up while
suspend is running.

Signed-off-by: Douglas Anderson <dianders@chromium.org>
Acked-by: Hayes Wang <hayeswang@realtek.com>
Link: https://lore.kernel.org/r/66590f22.170a0220.8b5ad.1750@mx.google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/usb/r8152.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 19df1cd9f072..6a3f4b2114ee 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -8554,6 +8554,19 @@ static int rtl8152_system_resume(struct r8152 *tp)
 		usb_submit_urb(tp->intr_urb, GFP_NOIO);
 	}
 
+	/* If the device is RTL8152_INACCESSIBLE here then we should do a
+	 * reset. This is important because the usb_lock_device_for_reset()
+	 * that happens as a result of usb_queue_reset_device() will silently
+	 * fail if the device was suspended or if too much time passed.
+	 *
+	 * NOTE: The device is locked here so we can directly do the reset.
+	 * We don't need usb_lock_device_for_reset() because that's just a
+	 * wrapper over device_lock() and device_resume() (which calls us)
+	 * does that for us.
+	 */
+	if (test_bit(RTL8152_INACCESSIBLE, &tp->flags))
+		usb_reset_device(tp->udev);
+
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From 8c1d92a740c002f36a919a55689d0ef83b61fd1c Mon Sep 17 00:00:00 2001
From: Douglas Anderson <dianders@chromium.org>
Date: Thu, 30 May 2024 16:43:09 -0700
Subject: r8152: Wake up the system if the we need a reset

If we get to the end of the r8152's suspend() routine and we find that
the USB device is INACCESSIBLE then it means that some of our
preparation for suspend didn't take place. We need a USB reset to get
ourselves back in a consistent state so we can try again and that
can't happen during system suspend. Call pm_wakeup_event() to wake the
system up in this case.

Signed-off-by: Douglas Anderson <dianders@chromium.org>
Acked-by: Hayes Wang <hayeswang@realtek.com>
Link: https://lore.kernel.org/r/66590f25.170a0220.8b5ad.1752@mx.google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/usb/r8152.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 6a3f4b2114ee..09fe70bc45d4 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -8647,6 +8647,13 @@ static int rtl8152_system_suspend(struct r8152 *tp)
 		tasklet_enable(&tp->tx_tl);
 	}
 
+	/* If we're inaccessible here then some of the work that we did to
+	 * get the adapter ready for suspend didn't work. Queue up a wakeup
+	 * event so we can try again.
+	 */
+	if (test_bit(RTL8152_INACCESSIBLE, &tp->flags))
+		pm_wakeup_event(&tp->udev->dev, 0);
+
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From c95a3be45ad22ee8925d6d1ab531d5ba98216311 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Thu, 23 May 2024 10:41:54 -0700
Subject: bpf: Remove unnecessary checks on the offset of btf_field.

reg_find_field_offset() always return a btf_field with a matching offset
value. Checking the offset of the returned btf_field is unnecessary.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240523174202.461236-2-thinker.li@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 77da1f438bec..57c0c255bf4c 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -11640,7 +11640,7 @@ __process_kf_arg_ptr_to_graph_node(struct bpf_verifier_env *env,
 
 	node_off = reg->off + reg->var_off.value;
 	field = reg_find_field_offset(reg, node_off, node_field_type);
-	if (!field || field->offset != node_off) {
+	if (!field) {
 		verbose(env, "%s not found at offset=%u\n", node_type_name, node_off);
 		return -EINVAL;
 	}
-- 
cgit v1.2.3-58-ga151


From 482f7133791e894b94a57ab3251e03d4c98ea42b Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Thu, 23 May 2024 10:41:55 -0700
Subject: bpf: Remove unnecessary call to btf_field_type_size().

field->size has been initialized by bpf_parse_fields() with the value
returned by btf_field_type_size(). Use it instead of calling
btf_field_type_size() again.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240523174202.461236-3-thinker.li@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/btf.c      | 2 +-
 kernel/bpf/verifier.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 821063660d9f..226138bd139a 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -6693,7 +6693,7 @@ int btf_struct_access(struct bpf_verifier_log *log,
 		for (i = 0; i < rec->cnt; i++) {
 			struct btf_field *field = &rec->fields[i];
 			u32 offset = field->offset;
-			if (off < offset + btf_field_type_size(field->type) && offset < off + size) {
+			if (off < offset + field->size && offset < off + size) {
 				bpf_log(log,
 					"direct access to %s is disallowed\n",
 					btf_field_type_name(field->type));
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 57c0c255bf4c..81a3d2ced78d 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -5448,7 +5448,7 @@ static int check_map_access(struct bpf_verifier_env *env, u32 regno,
 		 * this program. To check that [x1, x2) overlaps with [y1, y2),
 		 * it is sufficient to check x1 < y2 && y1 < x2.
 		 */
-		if (reg->smin_value + off < p + btf_field_type_size(field->type) &&
+		if (reg->smin_value + off < p + field->size &&
 		    p < reg->umax_value + off + size) {
 			switch (field->type) {
 			case BPF_KPTR_UNREF:
-- 
cgit v1.2.3-58-ga151


From a7db0d4f872a869feb7c0201c0fa736c309192d5 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Thu, 23 May 2024 10:41:56 -0700
Subject: bpf: refactor btf_find_struct_field() and btf_find_datasec_var().

Move common code of the two functions to btf_find_field_one().

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240523174202.461236-4-thinker.li@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/btf.c | 180 ++++++++++++++++++++++++-------------------------------
 1 file changed, 79 insertions(+), 101 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 226138bd139a..2ce61c3a7e28 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3494,72 +3494,95 @@ end:
 
 #undef field_mask_test_name
 
+static int btf_find_field_one(const struct btf *btf,
+			      const struct btf_type *var,
+			      const struct btf_type *var_type,
+			      int var_idx,
+			      u32 off, u32 expected_size,
+			      u32 field_mask, u32 *seen_mask,
+			      struct btf_field_info *info, int info_cnt)
+{
+	int ret, align, sz, field_type;
+	struct btf_field_info tmp;
+
+	field_type = btf_get_field_type(__btf_name_by_offset(btf, var_type->name_off),
+					field_mask, seen_mask, &align, &sz);
+	if (field_type == 0)
+		return 0;
+	if (field_type < 0)
+		return field_type;
+
+	if (expected_size && expected_size != sz)
+		return 0;
+	if (off % align)
+		return 0;
+
+	switch (field_type) {
+	case BPF_SPIN_LOCK:
+	case BPF_TIMER:
+	case BPF_WORKQUEUE:
+	case BPF_LIST_NODE:
+	case BPF_RB_NODE:
+	case BPF_REFCOUNT:
+		ret = btf_find_struct(btf, var_type, off, sz, field_type,
+				      info_cnt ? &info[0] : &tmp);
+		if (ret < 0)
+			return ret;
+		break;
+	case BPF_KPTR_UNREF:
+	case BPF_KPTR_REF:
+	case BPF_KPTR_PERCPU:
+		ret = btf_find_kptr(btf, var_type, off, sz,
+				    info_cnt ? &info[0] : &tmp);
+		if (ret < 0)
+			return ret;
+		break;
+	case BPF_LIST_HEAD:
+	case BPF_RB_ROOT:
+		ret = btf_find_graph_root(btf, var, var_type,
+					  var_idx, off, sz,
+					  info_cnt ? &info[0] : &tmp,
+					  field_type);
+		if (ret < 0)
+			return ret;
+		break;
+	default:
+		return -EFAULT;
+	}
+
+	if (ret == BTF_FIELD_IGNORE)
+		return 0;
+	if (!info_cnt)
+		return -E2BIG;
+
+	return 1;
+}
+
 static int btf_find_struct_field(const struct btf *btf,
 				 const struct btf_type *t, u32 field_mask,
 				 struct btf_field_info *info, int info_cnt)
 {
-	int ret, idx = 0, align, sz, field_type;
+	int ret, idx = 0;
 	const struct btf_member *member;
-	struct btf_field_info tmp;
 	u32 i, off, seen_mask = 0;
 
 	for_each_member(i, t, member) {
 		const struct btf_type *member_type = btf_type_by_id(btf,
 								    member->type);
 
-		field_type = btf_get_field_type(__btf_name_by_offset(btf, member_type->name_off),
-						field_mask, &seen_mask, &align, &sz);
-		if (field_type == 0)
-			continue;
-		if (field_type < 0)
-			return field_type;
-
 		off = __btf_member_bit_offset(t, member);
 		if (off % 8)
 			/* valid C code cannot generate such BTF */
 			return -EINVAL;
 		off /= 8;
-		if (off % align)
-			continue;
-
-		switch (field_type) {
-		case BPF_SPIN_LOCK:
-		case BPF_TIMER:
-		case BPF_WORKQUEUE:
-		case BPF_LIST_NODE:
-		case BPF_RB_NODE:
-		case BPF_REFCOUNT:
-			ret = btf_find_struct(btf, member_type, off, sz, field_type,
-					      idx < info_cnt ? &info[idx] : &tmp);
-			if (ret < 0)
-				return ret;
-			break;
-		case BPF_KPTR_UNREF:
-		case BPF_KPTR_REF:
-		case BPF_KPTR_PERCPU:
-			ret = btf_find_kptr(btf, member_type, off, sz,
-					    idx < info_cnt ? &info[idx] : &tmp);
-			if (ret < 0)
-				return ret;
-			break;
-		case BPF_LIST_HEAD:
-		case BPF_RB_ROOT:
-			ret = btf_find_graph_root(btf, t, member_type,
-						  i, off, sz,
-						  idx < info_cnt ? &info[idx] : &tmp,
-						  field_type);
-			if (ret < 0)
-				return ret;
-			break;
-		default:
-			return -EFAULT;
-		}
 
-		if (ret == BTF_FIELD_IGNORE)
-			continue;
-		if (idx >= info_cnt)
-			return -E2BIG;
-		++idx;
+		ret = btf_find_field_one(btf, t, member_type, i,
+					 off, 0,
+					 field_mask, &seen_mask,
+					 &info[idx], info_cnt - idx);
+		if (ret < 0)
+			return ret;
+		idx += ret;
 	}
 	return idx;
 }
@@ -3568,66 +3591,21 @@ static int btf_find_datasec_var(const struct btf *btf, const struct btf_type *t,
 				u32 field_mask, struct btf_field_info *info,
 				int info_cnt)
 {
-	int ret, idx = 0, align, sz, field_type;
+	int ret, idx = 0;
 	const struct btf_var_secinfo *vsi;
-	struct btf_field_info tmp;
 	u32 i, off, seen_mask = 0;
 
 	for_each_vsi(i, t, vsi) {
 		const struct btf_type *var = btf_type_by_id(btf, vsi->type);
 		const struct btf_type *var_type = btf_type_by_id(btf, var->type);
 
-		field_type = btf_get_field_type(__btf_name_by_offset(btf, var_type->name_off),
-						field_mask, &seen_mask, &align, &sz);
-		if (field_type == 0)
-			continue;
-		if (field_type < 0)
-			return field_type;
-
 		off = vsi->offset;
-		if (vsi->size != sz)
-			continue;
-		if (off % align)
-			continue;
-
-		switch (field_type) {
-		case BPF_SPIN_LOCK:
-		case BPF_TIMER:
-		case BPF_WORKQUEUE:
-		case BPF_LIST_NODE:
-		case BPF_RB_NODE:
-		case BPF_REFCOUNT:
-			ret = btf_find_struct(btf, var_type, off, sz, field_type,
-					      idx < info_cnt ? &info[idx] : &tmp);
-			if (ret < 0)
-				return ret;
-			break;
-		case BPF_KPTR_UNREF:
-		case BPF_KPTR_REF:
-		case BPF_KPTR_PERCPU:
-			ret = btf_find_kptr(btf, var_type, off, sz,
-					    idx < info_cnt ? &info[idx] : &tmp);
-			if (ret < 0)
-				return ret;
-			break;
-		case BPF_LIST_HEAD:
-		case BPF_RB_ROOT:
-			ret = btf_find_graph_root(btf, var, var_type,
-						  -1, off, sz,
-						  idx < info_cnt ? &info[idx] : &tmp,
-						  field_type);
-			if (ret < 0)
-				return ret;
-			break;
-		default:
-			return -EFAULT;
-		}
-
-		if (ret == BTF_FIELD_IGNORE)
-			continue;
-		if (idx >= info_cnt)
-			return -E2BIG;
-		++idx;
+		ret = btf_find_field_one(btf, var, var_type, -1, off, vsi->size,
+					 field_mask, &seen_mask,
+					 &info[idx], info_cnt - idx);
+		if (ret < 0)
+			return ret;
+		idx += ret;
 	}
 	return idx;
 }
-- 
cgit v1.2.3-58-ga151


From 994796c0256c4001633488fd24c3d54691949f8d Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Thu, 23 May 2024 10:41:57 -0700
Subject: bpf: create repeated fields for arrays.

The verifier uses field information for certain special types, such as
kptr, rbtree root, and list head. These types are treated
differently. However, we did not previously support these types in
arrays. This update examines arrays and duplicates field information the
same number of times as the length of the array if the element type is one
of the special types.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240523174202.461236-5-thinker.li@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/btf.c | 62 ++++++++++++++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 58 insertions(+), 4 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 2ce61c3a7e28..4fefa27d5aea 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3494,6 +3494,41 @@ end:
 
 #undef field_mask_test_name
 
+/* Repeat a field for a specified number of times.
+ *
+ * Copy and repeat the first field for repeat_cnt
+ * times. The field is repeated by adding the offset of each field
+ * with
+ *   (i + 1) * elem_size
+ * where i is the repeat index and elem_size is the size of an element.
+ */
+static int btf_repeat_field(struct btf_field_info *info,
+			    u32 repeat_cnt, u32 elem_size)
+{
+	u32 i;
+	u32 cur;
+
+	/* Ensure not repeating fields that should not be repeated. */
+	switch (info[0].type) {
+	case BPF_KPTR_UNREF:
+	case BPF_KPTR_REF:
+	case BPF_KPTR_PERCPU:
+	case BPF_LIST_HEAD:
+	case BPF_RB_ROOT:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	cur = 1;
+	for (i = 0; i < repeat_cnt; i++) {
+		memcpy(&info[cur], &info[0], sizeof(info[0]));
+		info[cur++].off += (i + 1) * elem_size;
+	}
+
+	return 0;
+}
+
 static int btf_find_field_one(const struct btf *btf,
 			      const struct btf_type *var,
 			      const struct btf_type *var_type,
@@ -3504,6 +3539,21 @@ static int btf_find_field_one(const struct btf *btf,
 {
 	int ret, align, sz, field_type;
 	struct btf_field_info tmp;
+	const struct btf_array *array;
+	u32 i, nelems = 1;
+
+	/* Walk into array types to find the element type and the number of
+	 * elements in the (flattened) array.
+	 */
+	for (i = 0; i < MAX_RESOLVE_DEPTH && btf_type_is_array(var_type); i++) {
+		array = btf_array(var_type);
+		nelems *= array->nelems;
+		var_type = btf_type_by_id(btf, array->type);
+	}
+	if (i == MAX_RESOLVE_DEPTH)
+		return -E2BIG;
+	if (nelems == 0)
+		return 0;
 
 	field_type = btf_get_field_type(__btf_name_by_offset(btf, var_type->name_off),
 					field_mask, seen_mask, &align, &sz);
@@ -3512,7 +3562,7 @@ static int btf_find_field_one(const struct btf *btf,
 	if (field_type < 0)
 		return field_type;
 
-	if (expected_size && expected_size != sz)
+	if (expected_size && expected_size != sz * nelems)
 		return 0;
 	if (off % align)
 		return 0;
@@ -3552,10 +3602,14 @@ static int btf_find_field_one(const struct btf *btf,
 
 	if (ret == BTF_FIELD_IGNORE)
 		return 0;
-	if (!info_cnt)
+	if (nelems > info_cnt)
 		return -E2BIG;
-
-	return 1;
+	if (nelems > 1) {
+		ret = btf_repeat_field(info, nelems - 1, sz);
+		if (ret < 0)
+			return ret;
+	}
+	return nelems;
 }
 
 static int btf_find_struct_field(const struct btf *btf,
-- 
cgit v1.2.3-58-ga151


From 64e8ee814819f21beeeda00d4119221443d77992 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Thu, 23 May 2024 10:41:58 -0700
Subject: bpf: look into the types of the fields of a struct type recursively.

The verifier has field information for specific special types, such as
kptr, rbtree root, and list head. These types are handled
differently. However, we did not previously examine the types of fields of
a struct type variable. Field information records were not generated for
the kptrs, rbtree roots, and linked_list heads that are not located at the
outermost struct type of a variable.

For example,

  struct A {
    struct task_struct __kptr * task;
  };

  struct B {
    struct A mem_a;
  }

  struct B var_b;

It did not examine "struct A" so as not to generate field information for
the kptr in "struct A" for "var_b".

This patch enables BPF programs to define fields of these special types in
a struct type other than the direct type of a variable or in a struct type
that is the type of a field in the value type of a map.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240523174202.461236-6-thinker.li@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/btf.c | 100 ++++++++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 77 insertions(+), 23 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 4fefa27d5aea..5e2b231a9af4 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3442,10 +3442,12 @@ btf_find_graph_root(const struct btf *btf, const struct btf_type *pt,
 		goto end;						\
 	}
 
-static int btf_get_field_type(const char *name, u32 field_mask, u32 *seen_mask,
+static int btf_get_field_type(const struct btf *btf, const struct btf_type *var_type,
+			      u32 field_mask, u32 *seen_mask,
 			      int *align, int *sz)
 {
 	int type = 0;
+	const char *name = __btf_name_by_offset(btf, var_type->name_off);
 
 	if (field_mask & BPF_SPIN_LOCK) {
 		if (!strcmp(name, "bpf_spin_lock")) {
@@ -3481,7 +3483,7 @@ static int btf_get_field_type(const char *name, u32 field_mask, u32 *seen_mask,
 	field_mask_test_name(BPF_REFCOUNT,  "bpf_refcount");
 
 	/* Only return BPF_KPTR when all other types with matchable names fail */
-	if (field_mask & BPF_KPTR) {
+	if (field_mask & BPF_KPTR && !__btf_type_is_struct(var_type)) {
 		type = BPF_KPTR_REF;
 		goto end;
 	}
@@ -3494,41 +3496,83 @@ end:
 
 #undef field_mask_test_name
 
-/* Repeat a field for a specified number of times.
+/* Repeat a number of fields for a specified number of times.
  *
- * Copy and repeat the first field for repeat_cnt
- * times. The field is repeated by adding the offset of each field
- * with
+ * Copy the fields starting from the first field and repeat them for
+ * repeat_cnt times. The fields are repeated by adding the offset of each
+ * field with
  *   (i + 1) * elem_size
  * where i is the repeat index and elem_size is the size of an element.
  */
-static int btf_repeat_field(struct btf_field_info *info,
-			    u32 repeat_cnt, u32 elem_size)
+static int btf_repeat_fields(struct btf_field_info *info,
+			     u32 field_cnt, u32 repeat_cnt, u32 elem_size)
 {
-	u32 i;
+	u32 i, j;
 	u32 cur;
 
 	/* Ensure not repeating fields that should not be repeated. */
-	switch (info[0].type) {
-	case BPF_KPTR_UNREF:
-	case BPF_KPTR_REF:
-	case BPF_KPTR_PERCPU:
-	case BPF_LIST_HEAD:
-	case BPF_RB_ROOT:
-		break;
-	default:
-		return -EINVAL;
+	for (i = 0; i < field_cnt; i++) {
+		switch (info[i].type) {
+		case BPF_KPTR_UNREF:
+		case BPF_KPTR_REF:
+		case BPF_KPTR_PERCPU:
+		case BPF_LIST_HEAD:
+		case BPF_RB_ROOT:
+			break;
+		default:
+			return -EINVAL;
+		}
 	}
 
-	cur = 1;
+	cur = field_cnt;
 	for (i = 0; i < repeat_cnt; i++) {
-		memcpy(&info[cur], &info[0], sizeof(info[0]));
-		info[cur++].off += (i + 1) * elem_size;
+		memcpy(&info[cur], &info[0], field_cnt * sizeof(info[0]));
+		for (j = 0; j < field_cnt; j++)
+			info[cur++].off += (i + 1) * elem_size;
 	}
 
 	return 0;
 }
 
+static int btf_find_struct_field(const struct btf *btf,
+				 const struct btf_type *t, u32 field_mask,
+				 struct btf_field_info *info, int info_cnt);
+
+/* Find special fields in the struct type of a field.
+ *
+ * This function is used to find fields of special types that is not a
+ * global variable or a direct field of a struct type. It also handles the
+ * repetition if it is the element type of an array.
+ */
+static int btf_find_nested_struct(const struct btf *btf, const struct btf_type *t,
+				  u32 off, u32 nelems,
+				  u32 field_mask, struct btf_field_info *info,
+				  int info_cnt)
+{
+	int ret, err, i;
+
+	ret = btf_find_struct_field(btf, t, field_mask, info, info_cnt);
+
+	if (ret <= 0)
+		return ret;
+
+	/* Shift the offsets of the nested struct fields to the offsets
+	 * related to the container.
+	 */
+	for (i = 0; i < ret; i++)
+		info[i].off += off;
+
+	if (nelems > 1) {
+		err = btf_repeat_fields(info, ret, nelems - 1, t->size);
+		if (err == 0)
+			ret *= nelems;
+		else
+			ret = err;
+	}
+
+	return ret;
+}
+
 static int btf_find_field_one(const struct btf *btf,
 			      const struct btf_type *var,
 			      const struct btf_type *var_type,
@@ -3555,8 +3599,18 @@ static int btf_find_field_one(const struct btf *btf,
 	if (nelems == 0)
 		return 0;
 
-	field_type = btf_get_field_type(__btf_name_by_offset(btf, var_type->name_off),
+	field_type = btf_get_field_type(btf, var_type,
 					field_mask, seen_mask, &align, &sz);
+	/* Look into variables of struct types */
+	if (!field_type && __btf_type_is_struct(var_type)) {
+		sz = var_type->size;
+		if (expected_size && expected_size != sz * nelems)
+			return 0;
+		ret = btf_find_nested_struct(btf, var_type, off, nelems, field_mask,
+					     &info[0], info_cnt);
+		return ret;
+	}
+
 	if (field_type == 0)
 		return 0;
 	if (field_type < 0)
@@ -3605,7 +3659,7 @@ static int btf_find_field_one(const struct btf *btf,
 	if (nelems > info_cnt)
 		return -E2BIG;
 	if (nelems > 1) {
-		ret = btf_repeat_field(info, nelems - 1, sz);
+		ret = btf_repeat_fields(info, 1, nelems - 1, sz);
 		if (ret < 0)
 			return ret;
 	}
-- 
cgit v1.2.3-58-ga151


From f19caf57d80f4432acea61d858d45ce194444389 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Thu, 23 May 2024 10:41:59 -0700
Subject: bpf: limit the number of levels of a nested struct type.

Limit the number of levels looking into struct types to avoid running out
of stack space.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240523174202.461236-7-thinker.li@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/btf.c | 30 +++++++++++++++++++-----------
 1 file changed, 19 insertions(+), 11 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 5e2b231a9af4..7928d920056f 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3536,7 +3536,8 @@ static int btf_repeat_fields(struct btf_field_info *info,
 
 static int btf_find_struct_field(const struct btf *btf,
 				 const struct btf_type *t, u32 field_mask,
-				 struct btf_field_info *info, int info_cnt);
+				 struct btf_field_info *info, int info_cnt,
+				 u32 level);
 
 /* Find special fields in the struct type of a field.
  *
@@ -3547,11 +3548,15 @@ static int btf_find_struct_field(const struct btf *btf,
 static int btf_find_nested_struct(const struct btf *btf, const struct btf_type *t,
 				  u32 off, u32 nelems,
 				  u32 field_mask, struct btf_field_info *info,
-				  int info_cnt)
+				  int info_cnt, u32 level)
 {
 	int ret, err, i;
 
-	ret = btf_find_struct_field(btf, t, field_mask, info, info_cnt);
+	level++;
+	if (level >= MAX_RESOLVE_DEPTH)
+		return -E2BIG;
+
+	ret = btf_find_struct_field(btf, t, field_mask, info, info_cnt, level);
 
 	if (ret <= 0)
 		return ret;
@@ -3579,7 +3584,8 @@ static int btf_find_field_one(const struct btf *btf,
 			      int var_idx,
 			      u32 off, u32 expected_size,
 			      u32 field_mask, u32 *seen_mask,
-			      struct btf_field_info *info, int info_cnt)
+			      struct btf_field_info *info, int info_cnt,
+			      u32 level)
 {
 	int ret, align, sz, field_type;
 	struct btf_field_info tmp;
@@ -3607,7 +3613,7 @@ static int btf_find_field_one(const struct btf *btf,
 		if (expected_size && expected_size != sz * nelems)
 			return 0;
 		ret = btf_find_nested_struct(btf, var_type, off, nelems, field_mask,
-					     &info[0], info_cnt);
+					     &info[0], info_cnt, level);
 		return ret;
 	}
 
@@ -3668,7 +3674,8 @@ static int btf_find_field_one(const struct btf *btf,
 
 static int btf_find_struct_field(const struct btf *btf,
 				 const struct btf_type *t, u32 field_mask,
-				 struct btf_field_info *info, int info_cnt)
+				 struct btf_field_info *info, int info_cnt,
+				 u32 level)
 {
 	int ret, idx = 0;
 	const struct btf_member *member;
@@ -3687,7 +3694,7 @@ static int btf_find_struct_field(const struct btf *btf,
 		ret = btf_find_field_one(btf, t, member_type, i,
 					 off, 0,
 					 field_mask, &seen_mask,
-					 &info[idx], info_cnt - idx);
+					 &info[idx], info_cnt - idx, level);
 		if (ret < 0)
 			return ret;
 		idx += ret;
@@ -3697,7 +3704,7 @@ static int btf_find_struct_field(const struct btf *btf,
 
 static int btf_find_datasec_var(const struct btf *btf, const struct btf_type *t,
 				u32 field_mask, struct btf_field_info *info,
-				int info_cnt)
+				int info_cnt, u32 level)
 {
 	int ret, idx = 0;
 	const struct btf_var_secinfo *vsi;
@@ -3710,7 +3717,8 @@ static int btf_find_datasec_var(const struct btf *btf, const struct btf_type *t,
 		off = vsi->offset;
 		ret = btf_find_field_one(btf, var, var_type, -1, off, vsi->size,
 					 field_mask, &seen_mask,
-					 &info[idx], info_cnt - idx);
+					 &info[idx], info_cnt - idx,
+					 level);
 		if (ret < 0)
 			return ret;
 		idx += ret;
@@ -3723,9 +3731,9 @@ static int btf_find_field(const struct btf *btf, const struct btf_type *t,
 			  int info_cnt)
 {
 	if (__btf_type_is_struct(t))
-		return btf_find_struct_field(btf, t, field_mask, info, info_cnt);
+		return btf_find_struct_field(btf, t, field_mask, info, info_cnt, 0);
 	else if (btf_type_is_datasec(t))
-		return btf_find_datasec_var(btf, t, field_mask, info, info_cnt);
+		return btf_find_datasec_var(btf, t, field_mask, info, info_cnt, 0);
 	return -EINVAL;
 }
 
-- 
cgit v1.2.3-58-ga151


From c4c6c3b785a0b1426add15d078da61f899abeaac Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Thu, 23 May 2024 10:42:00 -0700
Subject: selftests/bpf: Test kptr arrays and kptrs in nested struct fields.

Make sure that BPF programs can declare global kptr arrays and kptr fields
in struct types that is the type of a global variable or the type of a
nested descendant field in a global variable.

An array with only one element is special case, that it treats the element
like a non-array kptr field. Nested arrays are also tested to ensure they
are handled properly.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240523174202.461236-8-thinker.li@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/cpumask.c   |   5 +
 .../testing/selftests/bpf/progs/cpumask_success.c  | 171 +++++++++++++++++++++
 2 files changed, 176 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/cpumask.c b/tools/testing/selftests/bpf/prog_tests/cpumask.c
index ecf89df78109..2570bd4b0cb2 100644
--- a/tools/testing/selftests/bpf/prog_tests/cpumask.c
+++ b/tools/testing/selftests/bpf/prog_tests/cpumask.c
@@ -18,6 +18,11 @@ static const char * const cpumask_success_testcases[] = {
 	"test_insert_leave",
 	"test_insert_remove_release",
 	"test_global_mask_rcu",
+	"test_global_mask_array_one_rcu",
+	"test_global_mask_array_rcu",
+	"test_global_mask_array_l2_rcu",
+	"test_global_mask_nested_rcu",
+	"test_global_mask_nested_deep_rcu",
 	"test_cpumask_weight",
 };
 
diff --git a/tools/testing/selftests/bpf/progs/cpumask_success.c b/tools/testing/selftests/bpf/progs/cpumask_success.c
index 7a1e64c6c065..fd8106831c32 100644
--- a/tools/testing/selftests/bpf/progs/cpumask_success.c
+++ b/tools/testing/selftests/bpf/progs/cpumask_success.c
@@ -12,6 +12,31 @@ char _license[] SEC("license") = "GPL";
 
 int pid, nr_cpus;
 
+struct kptr_nested {
+	struct bpf_cpumask __kptr * mask;
+};
+
+struct kptr_nested_pair {
+	struct bpf_cpumask __kptr * mask_1;
+	struct bpf_cpumask __kptr * mask_2;
+};
+
+struct kptr_nested_mid {
+	int dummy;
+	struct kptr_nested m;
+};
+
+struct kptr_nested_deep {
+	struct kptr_nested_mid ptrs[2];
+	struct kptr_nested_pair ptr_pairs[3];
+};
+
+private(MASK) static struct bpf_cpumask __kptr * global_mask_array[2];
+private(MASK) static struct bpf_cpumask __kptr * global_mask_array_l2[2][1];
+private(MASK) static struct bpf_cpumask __kptr * global_mask_array_one[1];
+private(MASK) static struct kptr_nested global_mask_nested[2];
+private(MASK_DEEP) static struct kptr_nested_deep global_mask_nested_deep;
+
 static bool is_test_task(void)
 {
 	int cur_pid = bpf_get_current_pid_tgid() >> 32;
@@ -460,6 +485,152 @@ int BPF_PROG(test_global_mask_rcu, struct task_struct *task, u64 clone_flags)
 	return 0;
 }
 
+SEC("tp_btf/task_newtask")
+int BPF_PROG(test_global_mask_array_one_rcu, struct task_struct *task, u64 clone_flags)
+{
+	struct bpf_cpumask *local, *prev;
+
+	if (!is_test_task())
+		return 0;
+
+	/* Kptr arrays with one element are special cased, being treated
+	 * just like a single pointer.
+	 */
+
+	local = create_cpumask();
+	if (!local)
+		return 0;
+
+	prev = bpf_kptr_xchg(&global_mask_array_one[0], local);
+	if (prev) {
+		bpf_cpumask_release(prev);
+		err = 3;
+		return 0;
+	}
+
+	bpf_rcu_read_lock();
+	local = global_mask_array_one[0];
+	if (!local) {
+		err = 4;
+		bpf_rcu_read_unlock();
+		return 0;
+	}
+
+	bpf_rcu_read_unlock();
+
+	return 0;
+}
+
+static int _global_mask_array_rcu(struct bpf_cpumask **mask0,
+				  struct bpf_cpumask **mask1)
+{
+	struct bpf_cpumask *local;
+
+	if (!is_test_task())
+		return 0;
+
+	/* Check if two kptrs in the array work and independently */
+
+	local = create_cpumask();
+	if (!local)
+		return 0;
+
+	bpf_rcu_read_lock();
+
+	local = bpf_kptr_xchg(mask0, local);
+	if (local) {
+		err = 1;
+		goto err_exit;
+	}
+
+	/* [<mask 0>, NULL] */
+	if (!*mask0 || *mask1) {
+		err = 2;
+		goto err_exit;
+	}
+
+	local = create_cpumask();
+	if (!local) {
+		err = 9;
+		goto err_exit;
+	}
+
+	local = bpf_kptr_xchg(mask1, local);
+	if (local) {
+		err = 10;
+		goto err_exit;
+	}
+
+	/* [<mask 0>, <mask 1>] */
+	if (!*mask0 || !*mask1 || *mask0 == *mask1) {
+		err = 11;
+		goto err_exit;
+	}
+
+err_exit:
+	if (local)
+		bpf_cpumask_release(local);
+	bpf_rcu_read_unlock();
+	return 0;
+}
+
+SEC("tp_btf/task_newtask")
+int BPF_PROG(test_global_mask_array_rcu, struct task_struct *task, u64 clone_flags)
+{
+	return _global_mask_array_rcu(&global_mask_array[0], &global_mask_array[1]);
+}
+
+SEC("tp_btf/task_newtask")
+int BPF_PROG(test_global_mask_array_l2_rcu, struct task_struct *task, u64 clone_flags)
+{
+	return _global_mask_array_rcu(&global_mask_array_l2[0][0], &global_mask_array_l2[1][0]);
+}
+
+SEC("tp_btf/task_newtask")
+int BPF_PROG(test_global_mask_nested_rcu, struct task_struct *task, u64 clone_flags)
+{
+	return _global_mask_array_rcu(&global_mask_nested[0].mask, &global_mask_nested[1].mask);
+}
+
+/* Ensure that the field->offset has been correctly advanced from one
+ * nested struct or array sub-tree to another. In the case of
+ * kptr_nested_deep, it comprises two sub-trees: ktpr_1 and kptr_2.  By
+ * calling bpf_kptr_xchg() on every single kptr in both nested sub-trees,
+ * the verifier should reject the program if the field->offset of any kptr
+ * is incorrect.
+ *
+ * For instance, if we have 10 kptrs in a nested struct and a program that
+ * accesses each kptr individually with bpf_kptr_xchg(), the compiler
+ * should emit instructions to access 10 different offsets if it works
+ * correctly. If the field->offset values of any pair of them are
+ * incorrectly the same, the number of unique offsets in btf_record for
+ * this nested struct should be less than 10. The verifier should fail to
+ * discover some of the offsets emitted by the compiler.
+ *
+ * Even if the field->offset values of kptrs are not duplicated, the
+ * verifier should fail to find a btf_field for the instruction accessing a
+ * kptr if the corresponding field->offset is pointing to a random
+ * incorrect offset.
+ */
+SEC("tp_btf/task_newtask")
+int BPF_PROG(test_global_mask_nested_deep_rcu, struct task_struct *task, u64 clone_flags)
+{
+	int r, i;
+
+	r = _global_mask_array_rcu(&global_mask_nested_deep.ptrs[0].m.mask,
+				   &global_mask_nested_deep.ptrs[1].m.mask);
+	if (r)
+		return r;
+
+	for (i = 0; i < 3; i++) {
+		r = _global_mask_array_rcu(&global_mask_nested_deep.ptr_pairs[i].mask_1,
+					   &global_mask_nested_deep.ptr_pairs[i].mask_2);
+		if (r)
+			return r;
+	}
+	return 0;
+}
+
 SEC("tp_btf/task_newtask")
 int BPF_PROG(test_cpumask_weight, struct task_struct *task, u64 clone_flags)
 {
-- 
cgit v1.2.3-58-ga151


From d55c765a9b2d54b53ef86a62d6209e2e5eb62585 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Thu, 23 May 2024 10:42:01 -0700
Subject: selftests/bpf: Test global bpf_rb_root arrays and fields in nested
 struct types.

Make sure global arrays of bpf_rb_root and fields of bpf_rb_root in nested
struct types work correctly.

Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240523174202.461236-9-thinker.li@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/rbtree.c | 47 +++++++++++++++
 tools/testing/selftests/bpf/progs/rbtree.c      | 77 +++++++++++++++++++++++++
 2 files changed, 124 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/rbtree.c b/tools/testing/selftests/bpf/prog_tests/rbtree.c
index e9300c96607d..9818f06c97c5 100644
--- a/tools/testing/selftests/bpf/prog_tests/rbtree.c
+++ b/tools/testing/selftests/bpf/prog_tests/rbtree.c
@@ -31,6 +31,28 @@ static void test_rbtree_add_nodes(void)
 	rbtree__destroy(skel);
 }
 
+static void test_rbtree_add_nodes_nested(void)
+{
+	LIBBPF_OPTS(bpf_test_run_opts, opts,
+		    .data_in = &pkt_v4,
+		    .data_size_in = sizeof(pkt_v4),
+		    .repeat = 1,
+	);
+	struct rbtree *skel;
+	int ret;
+
+	skel = rbtree__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "rbtree__open_and_load"))
+		return;
+
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.rbtree_add_nodes_nested), &opts);
+	ASSERT_OK(ret, "rbtree_add_nodes_nested run");
+	ASSERT_OK(opts.retval, "rbtree_add_nodes_nested retval");
+	ASSERT_EQ(skel->data->less_callback_ran, 1, "rbtree_add_nodes_nested less_callback_ran");
+
+	rbtree__destroy(skel);
+}
+
 static void test_rbtree_add_and_remove(void)
 {
 	LIBBPF_OPTS(bpf_test_run_opts, opts,
@@ -53,6 +75,27 @@ static void test_rbtree_add_and_remove(void)
 	rbtree__destroy(skel);
 }
 
+static void test_rbtree_add_and_remove_array(void)
+{
+	LIBBPF_OPTS(bpf_test_run_opts, opts,
+		    .data_in = &pkt_v4,
+		    .data_size_in = sizeof(pkt_v4),
+		    .repeat = 1,
+	);
+	struct rbtree *skel;
+	int ret;
+
+	skel = rbtree__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "rbtree__open_and_load"))
+		return;
+
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.rbtree_add_and_remove_array), &opts);
+	ASSERT_OK(ret, "rbtree_add_and_remove_array");
+	ASSERT_OK(opts.retval, "rbtree_add_and_remove_array retval");
+
+	rbtree__destroy(skel);
+}
+
 static void test_rbtree_first_and_remove(void)
 {
 	LIBBPF_OPTS(bpf_test_run_opts, opts,
@@ -104,8 +147,12 @@ void test_rbtree_success(void)
 {
 	if (test__start_subtest("rbtree_add_nodes"))
 		test_rbtree_add_nodes();
+	if (test__start_subtest("rbtree_add_nodes_nested"))
+		test_rbtree_add_nodes_nested();
 	if (test__start_subtest("rbtree_add_and_remove"))
 		test_rbtree_add_and_remove();
+	if (test__start_subtest("rbtree_add_and_remove_array"))
+		test_rbtree_add_and_remove_array();
 	if (test__start_subtest("rbtree_first_and_remove"))
 		test_rbtree_first_and_remove();
 	if (test__start_subtest("rbtree_api_release_aliasing"))
diff --git a/tools/testing/selftests/bpf/progs/rbtree.c b/tools/testing/selftests/bpf/progs/rbtree.c
index b09f4fffe57c..a3620c15c136 100644
--- a/tools/testing/selftests/bpf/progs/rbtree.c
+++ b/tools/testing/selftests/bpf/progs/rbtree.c
@@ -13,6 +13,15 @@ struct node_data {
 	struct bpf_rb_node node;
 };
 
+struct root_nested_inner {
+	struct bpf_spin_lock glock;
+	struct bpf_rb_root root __contains(node_data, node);
+};
+
+struct root_nested {
+	struct root_nested_inner inner;
+};
+
 long less_callback_ran = -1;
 long removed_key = -1;
 long first_data[2] = {-1, -1};
@@ -20,6 +29,9 @@ long first_data[2] = {-1, -1};
 #define private(name) SEC(".data." #name) __hidden __attribute__((aligned(8)))
 private(A) struct bpf_spin_lock glock;
 private(A) struct bpf_rb_root groot __contains(node_data, node);
+private(A) struct bpf_rb_root groot_array[2] __contains(node_data, node);
+private(A) struct bpf_rb_root groot_array_one[1] __contains(node_data, node);
+private(B) struct root_nested groot_nested;
 
 static bool less(struct bpf_rb_node *a, const struct bpf_rb_node *b)
 {
@@ -71,6 +83,12 @@ long rbtree_add_nodes(void *ctx)
 	return __add_three(&groot, &glock);
 }
 
+SEC("tc")
+long rbtree_add_nodes_nested(void *ctx)
+{
+	return __add_three(&groot_nested.inner.root, &groot_nested.inner.glock);
+}
+
 SEC("tc")
 long rbtree_add_and_remove(void *ctx)
 {
@@ -109,6 +127,65 @@ err_out:
 	return 1;
 }
 
+SEC("tc")
+long rbtree_add_and_remove_array(void *ctx)
+{
+	struct bpf_rb_node *res1 = NULL, *res2 = NULL, *res3 = NULL;
+	struct node_data *nodes[3][2] = {{NULL, NULL}, {NULL, NULL}, {NULL, NULL}};
+	struct node_data *n;
+	long k1 = -1, k2 = -1, k3 = -1;
+	int i, j;
+
+	for (i = 0; i < 3; i++) {
+		for (j = 0; j < 2; j++) {
+			nodes[i][j] = bpf_obj_new(typeof(*nodes[i][j]));
+			if (!nodes[i][j])
+				goto err_out;
+			nodes[i][j]->key = i * 2 + j;
+		}
+	}
+
+	bpf_spin_lock(&glock);
+	for (i = 0; i < 2; i++)
+		for (j = 0; j < 2; j++)
+			bpf_rbtree_add(&groot_array[i], &nodes[i][j]->node, less);
+	for (j = 0; j < 2; j++)
+		bpf_rbtree_add(&groot_array_one[0], &nodes[2][j]->node, less);
+	res1 = bpf_rbtree_remove(&groot_array[0], &nodes[0][0]->node);
+	res2 = bpf_rbtree_remove(&groot_array[1], &nodes[1][0]->node);
+	res3 = bpf_rbtree_remove(&groot_array_one[0], &nodes[2][0]->node);
+	bpf_spin_unlock(&glock);
+
+	if (res1) {
+		n = container_of(res1, struct node_data, node);
+		k1 = n->key;
+		bpf_obj_drop(n);
+	}
+	if (res2) {
+		n = container_of(res2, struct node_data, node);
+		k2 = n->key;
+		bpf_obj_drop(n);
+	}
+	if (res3) {
+		n = container_of(res3, struct node_data, node);
+		k3 = n->key;
+		bpf_obj_drop(n);
+	}
+	if (k1 != 0 || k2 != 2 || k3 != 4)
+		return 2;
+
+	return 0;
+
+err_out:
+	for (i = 0; i < 3; i++) {
+		for (j = 0; j < 2; j++) {
+			if (nodes[i][j])
+				bpf_obj_drop(nodes[i][j]);
+		}
+	}
+	return 1;
+}
+
 SEC("tc")
 long rbtree_first_and_remove(void *ctx)
 {
-- 
cgit v1.2.3-58-ga151


From 43d50ffb1f7e32865cdd343224659614d8b558b9 Mon Sep 17 00:00:00 2001
From: Kui-Feng Lee <thinker.li@gmail.com>
Date: Thu, 23 May 2024 10:42:02 -0700
Subject: selftests/bpf: Test global bpf_list_head arrays.

Make sure global arrays of bpf_list_heads and fields of bpf_list_heads in
nested struct types work correctly.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Link: https://lore.kernel.org/r/20240523174202.461236-10-thinker.li@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../testing/selftests/bpf/prog_tests/linked_list.c | 12 +++++++
 tools/testing/selftests/bpf/progs/linked_list.c    | 42 ++++++++++++++++++++++
 2 files changed, 54 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/linked_list.c b/tools/testing/selftests/bpf/prog_tests/linked_list.c
index 2fb89de63bd2..77d07e0a4a55 100644
--- a/tools/testing/selftests/bpf/prog_tests/linked_list.c
+++ b/tools/testing/selftests/bpf/prog_tests/linked_list.c
@@ -183,6 +183,18 @@ static void test_linked_list_success(int mode, bool leave_in_map)
 	if (!leave_in_map)
 		clear_fields(skel->maps.bss_A);
 
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.global_list_push_pop_nested), &opts);
+	ASSERT_OK(ret, "global_list_push_pop_nested");
+	ASSERT_OK(opts.retval, "global_list_push_pop_nested retval");
+	if (!leave_in_map)
+		clear_fields(skel->maps.bss_A);
+
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.global_list_array_push_pop), &opts);
+	ASSERT_OK(ret, "global_list_array_push_pop");
+	ASSERT_OK(opts.retval, "global_list_array_push_pop retval");
+	if (!leave_in_map)
+		clear_fields(skel->maps.bss_A);
+
 	if (mode == PUSH_POP)
 		goto end;
 
diff --git a/tools/testing/selftests/bpf/progs/linked_list.c b/tools/testing/selftests/bpf/progs/linked_list.c
index 26205ca80679..f69bf3e30321 100644
--- a/tools/testing/selftests/bpf/progs/linked_list.c
+++ b/tools/testing/selftests/bpf/progs/linked_list.c
@@ -11,6 +11,22 @@
 
 #include "linked_list.h"
 
+struct head_nested_inner {
+	struct bpf_spin_lock lock;
+	struct bpf_list_head head __contains(foo, node2);
+};
+
+struct head_nested {
+	int dummy;
+	struct head_nested_inner inner;
+};
+
+private(C) struct bpf_spin_lock glock_c;
+private(C) struct bpf_list_head ghead_array[2] __contains(foo, node2);
+private(C) struct bpf_list_head ghead_array_one[1] __contains(foo, node2);
+
+private(D) struct head_nested ghead_nested;
+
 static __always_inline
 int list_push_pop(struct bpf_spin_lock *lock, struct bpf_list_head *head, bool leave_in_map)
 {
@@ -309,6 +325,32 @@ int global_list_push_pop(void *ctx)
 	return test_list_push_pop(&glock, &ghead);
 }
 
+SEC("tc")
+int global_list_push_pop_nested(void *ctx)
+{
+	return test_list_push_pop(&ghead_nested.inner.lock, &ghead_nested.inner.head);
+}
+
+SEC("tc")
+int global_list_array_push_pop(void *ctx)
+{
+	int r;
+
+	r = test_list_push_pop(&glock_c, &ghead_array[0]);
+	if (r)
+		return r;
+
+	r = test_list_push_pop(&glock_c, &ghead_array[1]);
+	if (r)
+		return r;
+
+	/* Arrays with only one element is a special case, being treated
+	 * just like a bpf_list_head variable by the verifier, not an
+	 * array.
+	 */
+	return test_list_push_pop(&glock_c, &ghead_array_one[0]);
+}
+
 SEC("tc")
 int map_list_push_pop_multiple(void *ctx)
 {
-- 
cgit v1.2.3-58-ga151


From 4fdb6b6063f07d959a1c52a2ee580afc4da34e2d Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 29 May 2024 09:25:27 -0700
Subject: net: count drops due to missing qdisc as dev->tx_drops

Catching and debugging missing qdiscs is pretty tricky. When qdisc
is deleted we replace it with a noop qdisc, which silently drops
all the packets. Since the noop qdisc has a single static instance
we can't count drops at the qdisc level. Count them as dev->tx_drops.

  ip netns add red
  ip link add type veth peer netns red
  ip            link set dev veth0 up
  ip -netns red link set dev veth0 up
  ip            a a dev veth0 10.0.0.1/24
  ip -netns red a a dev veth0 10.0.0.2/24
  ping -c 2 10.0.0.2
  #  2 packets transmitted, 2 received, 0% packet loss, time 1031ms
  ip -s link show dev veth0
  #  TX:  bytes packets errors dropped carrier collsns
  #        1314      17      0       0       0       0

  tc qdisc replace dev veth0 root handle 1234: mq
  tc qdisc replace dev veth0 parent 1234:1 pfifo
  tc qdisc del dev veth0 parent 1234:1
  ping -c 2 10.0.0.2
  #  2 packets transmitted, 0 received, 100% packet loss, time 1034ms
  ip -s link show dev veth0
  #  TX:  bytes packets errors dropped carrier collsns
  #        1314      17      0       3       0       0

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20240529162527.3688979-1-kuba@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/sched/sch_generic.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 2a637a17061b..1417f1991452 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -633,6 +633,7 @@ EXPORT_SYMBOL_GPL(netif_carrier_event);
 static int noop_enqueue(struct sk_buff *skb, struct Qdisc *qdisc,
 			struct sk_buff **to_free)
 {
+	dev_core_stats_tx_dropped_inc(skb->dev);
 	__qdisc_drop(skb, to_free);
 	return NET_XMIT_CN;
 }
-- 
cgit v1.2.3-58-ga151


From 668b6a2ef832a878494cc1b12a881c8ec0494b25 Mon Sep 17 00:00:00 2001
From: Davide Caratti <dcaratti@redhat.com>
Date: Thu, 30 May 2024 19:08:34 +0200
Subject: flow_dissector: add support for tunnel control flags

Dissect [no]csum, [no]dontfrag, [no]oam, [no]crit flags from skb metadata.
This is a prerequisite for matching these control flags using TC flower.

Suggested-by: Ilya Maximets <i.maximets@ovn.org>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/flow_dissector.h |  9 +++++++++
 include/net/ip_tunnels.h     | 12 ++++++++++++
 net/core/flow_dissector.c    | 16 +++++++++++++++-
 3 files changed, 36 insertions(+), 1 deletion(-)

diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h
index 9ab376d1a677..99626475c3f4 100644
--- a/include/net/flow_dissector.h
+++ b/include/net/flow_dissector.h
@@ -329,6 +329,14 @@ struct flow_dissector_key_cfm {
 #define FLOW_DIS_CFM_MDL_MASK GENMASK(7, 5)
 #define FLOW_DIS_CFM_MDL_MAX 7
 
+/**
+ * struct flow_dissector_key_enc_flags: tunnel metadata control flags
+ * @flags: tunnel control flags
+ */
+struct flow_dissector_key_enc_flags {
+	u32 flags;
+};
+
 enum flow_dissector_key_id {
 	FLOW_DISSECTOR_KEY_CONTROL, /* struct flow_dissector_key_control */
 	FLOW_DISSECTOR_KEY_BASIC, /* struct flow_dissector_key_basic */
@@ -363,6 +371,7 @@ enum flow_dissector_key_id {
 	FLOW_DISSECTOR_KEY_L2TPV3, /* struct flow_dissector_key_l2tpv3 */
 	FLOW_DISSECTOR_KEY_CFM, /* struct flow_dissector_key_cfm */
 	FLOW_DISSECTOR_KEY_IPSEC, /* struct flow_dissector_key_ipsec */
+	FLOW_DISSECTOR_KEY_ENC_FLAGS, /* struct flow_dissector_key_enc_flags */
 
 	FLOW_DISSECTOR_KEY_MAX,
 };
diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index 9a6a08ec7713..5a530d4fb02c 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -247,6 +247,18 @@ static inline bool ip_tunnel_is_options_present(const unsigned long *flags)
 	return ip_tunnel_flags_intersect(flags, present);
 }
 
+static inline void ip_tunnel_set_encflags_present(unsigned long *flags)
+{
+	IP_TUNNEL_DECLARE_FLAGS(present) = { };
+
+	__set_bit(IP_TUNNEL_CSUM_BIT, present);
+	__set_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, present);
+	__set_bit(IP_TUNNEL_OAM_BIT, present);
+	__set_bit(IP_TUNNEL_CRIT_OPT_BIT, present);
+
+	ip_tunnel_flags_or(flags, flags, present);
+}
+
 static inline bool ip_tunnel_flags_is_be16_compat(const unsigned long *flags)
 {
 	IP_TUNNEL_DECLARE_FLAGS(supp) = { };
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index f82e9a7d3b37..59fe46077b3c 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -382,7 +382,9 @@ skb_flow_dissect_tunnel_info(const struct sk_buff *skb,
 	    !dissector_uses_key(flow_dissector,
 				FLOW_DISSECTOR_KEY_ENC_IP) &&
 	    !dissector_uses_key(flow_dissector,
-				FLOW_DISSECTOR_KEY_ENC_OPTS))
+				FLOW_DISSECTOR_KEY_ENC_OPTS) &&
+	    !dissector_uses_key(flow_dissector,
+				FLOW_DISSECTOR_KEY_ENC_FLAGS))
 		return;
 
 	info = skb_tunnel_info(skb);
@@ -475,6 +477,18 @@ skb_flow_dissect_tunnel_info(const struct sk_buff *skb,
 				    IP_TUNNEL_GENEVE_OPT_BIT);
 		enc_opt->dst_opt_type = val < __IP_TUNNEL_FLAG_NUM ? val : 0;
 	}
+
+	if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_ENC_FLAGS)) {
+		struct flow_dissector_key_enc_flags *enc_flags;
+		IP_TUNNEL_DECLARE_FLAGS(flags) = {};
+
+		enc_flags = skb_flow_dissector_target(flow_dissector,
+						      FLOW_DISSECTOR_KEY_ENC_FLAGS,
+						      target_container);
+		ip_tunnel_set_encflags_present(flags);
+		ip_tunnel_flags_and(flags, flags, info->key.tun_flags);
+		enc_flags->flags = bitmap_read(flags, IP_TUNNEL_CSUM_BIT, 32);
+	}
 }
 EXPORT_SYMBOL(skb_flow_dissect_tunnel_info);
 
-- 
cgit v1.2.3-58-ga151


From 1d17568e74dedbcb54d36af0662a15128297d681 Mon Sep 17 00:00:00 2001
From: Davide Caratti <dcaratti@redhat.com>
Date: Thu, 30 May 2024 19:08:35 +0200
Subject: net/sched: cls_flower: add support for matching tunnel control flags

extend cls_flower to match TUNNEL_FLAGS_PRESENT bits in tunnel metadata.

Suggested-by: Ilya Maximets <i.maximets@ovn.org>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/uapi/linux/pkt_cls.h |  3 +++
 net/sched/cls_flower.c       | 56 +++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 58 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h
index 229fc925ec3a..b6d38f5fd7c0 100644
--- a/include/uapi/linux/pkt_cls.h
+++ b/include/uapi/linux/pkt_cls.h
@@ -554,6 +554,9 @@ enum {
 	TCA_FLOWER_KEY_SPI,		/* be32 */
 	TCA_FLOWER_KEY_SPI_MASK,	/* be32 */
 
+	TCA_FLOWER_KEY_ENC_FLAGS,	/* u32 */
+	TCA_FLOWER_KEY_ENC_FLAGS_MASK,	/* u32 */
+
 	__TCA_FLOWER_MAX,
 };
 
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index fd9a6f20b60b..eef570c577ac 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -41,6 +41,12 @@
 #define TCA_FLOWER_KEY_CT_FLAGS_MASK \
 		(TCA_FLOWER_KEY_CT_FLAGS_MAX - 1)
 
+#define TUNNEL_FLAGS_PRESENT (\
+	_BITUL(IP_TUNNEL_CSUM_BIT) |		\
+	_BITUL(IP_TUNNEL_DONT_FRAGMENT_BIT) |	\
+	_BITUL(IP_TUNNEL_OAM_BIT) |		\
+	_BITUL(IP_TUNNEL_CRIT_OPT_BIT))
+
 struct fl_flow_key {
 	struct flow_dissector_key_meta meta;
 	struct flow_dissector_key_control control;
@@ -75,6 +81,7 @@ struct fl_flow_key {
 	struct flow_dissector_key_l2tpv3 l2tpv3;
 	struct flow_dissector_key_ipsec ipsec;
 	struct flow_dissector_key_cfm cfm;
+	struct flow_dissector_key_enc_flags enc_flags;
 } __aligned(BITS_PER_LONG / 8); /* Ensure that we can do comparisons as longs. */
 
 struct fl_flow_mask_range {
@@ -732,6 +739,10 @@ static const struct nla_policy fl_policy[TCA_FLOWER_MAX + 1] = {
 	[TCA_FLOWER_KEY_SPI_MASK]	= { .type = NLA_U32 },
 	[TCA_FLOWER_L2_MISS]		= NLA_POLICY_MAX(NLA_U8, 1),
 	[TCA_FLOWER_KEY_CFM]		= { .type = NLA_NESTED },
+	[TCA_FLOWER_KEY_ENC_FLAGS]	= NLA_POLICY_MASK(NLA_U32,
+							  TUNNEL_FLAGS_PRESENT),
+	[TCA_FLOWER_KEY_ENC_FLAGS_MASK]	= NLA_POLICY_MASK(NLA_U32,
+							  TUNNEL_FLAGS_PRESENT),
 };
 
 static const struct nla_policy
@@ -1825,6 +1836,21 @@ static int fl_set_key_cfm(struct nlattr **tb,
 	return 0;
 }
 
+static int fl_set_key_enc_flags(struct nlattr **tb, u32 *flags_key,
+				u32 *flags_mask, struct netlink_ext_ack *extack)
+{
+	/* mask is mandatory for flags */
+	if (NL_REQ_ATTR_CHECK(extack, NULL, tb, TCA_FLOWER_KEY_ENC_FLAGS_MASK)) {
+		NL_SET_ERR_MSG(extack, "missing enc_flags mask");
+		return -EINVAL;
+	}
+
+	*flags_key = nla_get_u32(tb[TCA_FLOWER_KEY_ENC_FLAGS]);
+	*flags_mask = nla_get_u32(tb[TCA_FLOWER_KEY_ENC_FLAGS_MASK]);
+
+	return 0;
+}
+
 static int fl_set_key(struct net *net, struct nlattr **tb,
 		      struct fl_flow_key *key, struct fl_flow_key *mask,
 		      struct netlink_ext_ack *extack)
@@ -2059,9 +2085,16 @@ static int fl_set_key(struct net *net, struct nlattr **tb,
 	if (ret)
 		return ret;
 
-	if (tb[TCA_FLOWER_KEY_FLAGS])
+	if (tb[TCA_FLOWER_KEY_FLAGS]) {
 		ret = fl_set_key_flags(tb, &key->control.flags,
 				       &mask->control.flags, extack);
+		if (ret)
+			return ret;
+	}
+
+	if (tb[TCA_FLOWER_KEY_ENC_FLAGS])
+		ret = fl_set_key_enc_flags(tb, &key->enc_flags.flags,
+					   &mask->enc_flags.flags, extack);
 
 	return ret;
 }
@@ -2175,6 +2208,8 @@ static void fl_init_dissector(struct flow_dissector *dissector,
 			     FLOW_DISSECTOR_KEY_IPSEC, ipsec);
 	FL_KEY_SET_IF_MASKED(mask, keys, cnt,
 			     FLOW_DISSECTOR_KEY_CFM, cfm);
+	FL_KEY_SET_IF_MASKED(mask, keys, cnt,
+			     FLOW_DISSECTOR_KEY_ENC_FLAGS, enc_flags);
 
 	skb_flow_dissector_init(dissector, keys, cnt);
 }
@@ -3291,6 +3326,22 @@ err_cfm_opts:
 	return err;
 }
 
+static int fl_dump_key_enc_flags(struct sk_buff *skb,
+				 struct flow_dissector_key_enc_flags *key,
+				 struct flow_dissector_key_enc_flags *mask)
+{
+	if (!memchr_inv(mask, 0, sizeof(*mask)))
+		return 0;
+
+	if (nla_put_u32(skb, TCA_FLOWER_KEY_ENC_FLAGS, key->flags))
+		return -EMSGSIZE;
+
+	if (nla_put_u32(skb, TCA_FLOWER_KEY_ENC_FLAGS_MASK, mask->flags))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
 static int fl_dump_key_options(struct sk_buff *skb, int enc_opt_type,
 			       struct flow_dissector_key_enc_opts *enc_opts)
 {
@@ -3592,6 +3643,9 @@ static int fl_dump_key(struct sk_buff *skb, struct net *net,
 	if (fl_dump_key_cfm(skb, &key->cfm, &mask->cfm))
 		goto nla_put_failure;
 
+	if (fl_dump_key_enc_flags(skb, &key->enc_flags, &mask->enc_flags))
+		goto nla_put_failure;
+
 	return 0;
 
 nla_put_failure:
-- 
cgit v1.2.3-58-ga151


From 071115301838c6c265065dd5d6bf43a9a987a550 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 30 May 2024 16:36:14 -0700
Subject: tcp: wrap mptcp and decrypted checks into tcp_skb_can_collapse_rx()

tcp_skb_can_collapse() checks for conditions which don't make
sense on input. Because of this we ended up sprinkling a few
pairs of mptcp_skb_can_collapse() and skb_cmp_decrypted() calls
on the input path. Group them in a new helper. This should make
it less likely that someone will check mptcp and not decrypted
or vice versa when adding new code.

This implicitly adds a decrypted check early in tcp_collapse().
AFAIU this will very slightly increase our ability to collapse
packets under memory pressure, not a real bug.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/tcp.h    |  7 +++++++
 net/ipv4/tcp_input.c | 11 +++--------
 net/ipv4/tcp_ipv4.c  |  3 +--
 3 files changed, 11 insertions(+), 10 deletions(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 32815a40dea1..32741856da01 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1071,6 +1071,13 @@ static inline bool tcp_skb_can_collapse(const struct sk_buff *to,
 		      skb_pure_zcopy_same(to, from));
 }
 
+static inline bool tcp_skb_can_collapse_rx(const struct sk_buff *to,
+					   const struct sk_buff *from)
+{
+	return likely(mptcp_skb_can_collapse(to, from) &&
+		      !skb_cmp_decrypted(to, from));
+}
+
 /* Events passed to congestion control interface */
 enum tcp_ca_event {
 	CA_EVENT_TX_START,	/* first transmit when no packets in flight */
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 5aadf64e554d..212b6fd0caf7 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4813,10 +4813,7 @@ static bool tcp_try_coalesce(struct sock *sk,
 	if (TCP_SKB_CB(from)->seq != TCP_SKB_CB(to)->end_seq)
 		return false;
 
-	if (!mptcp_skb_can_collapse(to, from))
-		return false;
-
-	if (skb_cmp_decrypted(from, to))
+	if (!tcp_skb_can_collapse_rx(to, from))
 		return false;
 
 	if (!skb_try_coalesce(to, from, fragstolen, &delta))
@@ -5372,7 +5369,7 @@ restart:
 			break;
 		}
 
-		if (n && n != tail && mptcp_skb_can_collapse(skb, n) &&
+		if (n && n != tail && tcp_skb_can_collapse_rx(skb, n) &&
 		    TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(n)->seq) {
 			end_of_skbs = false;
 			break;
@@ -5423,11 +5420,9 @@ restart:
 				skb = tcp_collapse_one(sk, skb, list, root);
 				if (!skb ||
 				    skb == tail ||
-				    !mptcp_skb_can_collapse(nskb, skb) ||
+				    !tcp_skb_can_collapse_rx(nskb, skb) ||
 				    (TCP_SKB_CB(skb)->tcp_flags & (TCPHDR_SYN | TCPHDR_FIN)))
 					goto end;
-				if (skb_cmp_decrypted(skb, nskb))
-					goto end;
 			}
 		}
 	}
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 59d5b064f233..04044605cadf 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2044,8 +2044,7 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb,
 	      TCP_SKB_CB(skb)->tcp_flags) & TCPHDR_ACK) ||
 	    ((TCP_SKB_CB(tail)->tcp_flags ^
 	      TCP_SKB_CB(skb)->tcp_flags) & (TCPHDR_ECE | TCPHDR_CWR)) ||
-	    !mptcp_skb_can_collapse(tail, skb) ||
-	    skb_cmp_decrypted(tail, skb) ||
+	    !tcp_skb_can_collapse_rx(tail, skb) ||
 	    thtail->doff != th->doff ||
 	    memcmp(thtail + 1, th + 1, hdrlen - sizeof(*th)))
 		goto no_coalesce;
-- 
cgit v1.2.3-58-ga151


From 1be68a87ab333af37b02ad928a724a722a5a8203 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 30 May 2024 16:36:15 -0700
Subject: tcp: add a helper for setting EOR on tail skb

TLS (and hopefully soon PSP will) use EOR to prevent skbs
with different decrypted state from getting merged, without
adding new tests to the skb handling. In both cases once
the connection switches to an "encrypted" state, all subsequent
skbs will be encrypted, so a single "EOR fence" is sufficient
to prevent mixing.

Add a helper for setting the EOR bit, to make this arrangement
more explicit.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/tcp.h    |  9 +++++++++
 net/tls/tls_device.c | 11 ++---------
 2 files changed, 11 insertions(+), 9 deletions(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 32741856da01..08c3b99501cf 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1066,6 +1066,7 @@ static inline bool tcp_skb_can_collapse_to(const struct sk_buff *skb)
 static inline bool tcp_skb_can_collapse(const struct sk_buff *to,
 					const struct sk_buff *from)
 {
+	/* skb_cmp_decrypted() not needed, use tcp_write_collapse_fence() */
 	return likely(tcp_skb_can_collapse_to(to) &&
 		      mptcp_skb_can_collapse(to, from) &&
 		      skb_pure_zcopy_same(to, from));
@@ -2102,6 +2103,14 @@ static inline void tcp_rtx_queue_unlink_and_free(struct sk_buff *skb, struct soc
 	tcp_wmem_free_skb(sk, skb);
 }
 
+static inline void tcp_write_collapse_fence(struct sock *sk)
+{
+	struct sk_buff *skb = tcp_write_queue_tail(sk);
+
+	if (skb)
+		TCP_SKB_CB(skb)->eor = 1;
+}
+
 static inline void tcp_push_pending_frames(struct sock *sk)
 {
 	if (tcp_send_head(sk)) {
diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c
index ab6e694f7bc2..dc063c2c7950 100644
--- a/net/tls/tls_device.c
+++ b/net/tls/tls_device.c
@@ -231,14 +231,10 @@ static void tls_device_resync_tx(struct sock *sk, struct tls_context *tls_ctx,
 				 u32 seq)
 {
 	struct net_device *netdev;
-	struct sk_buff *skb;
 	int err = 0;
 	u8 *rcd_sn;
 
-	skb = tcp_write_queue_tail(sk);
-	if (skb)
-		TCP_SKB_CB(skb)->eor = 1;
-
+	tcp_write_collapse_fence(sk);
 	rcd_sn = tls_ctx->tx.rec_seq;
 
 	trace_tls_device_tx_resync_send(sk, seq, rcd_sn);
@@ -1067,7 +1063,6 @@ int tls_set_device_offload(struct sock *sk)
 	struct tls_prot_info *prot;
 	struct net_device *netdev;
 	struct tls_context *ctx;
-	struct sk_buff *skb;
 	char *iv, *rec_seq;
 	int rc;
 
@@ -1138,9 +1133,7 @@ int tls_set_device_offload(struct sock *sk)
 	 * SKBs where only part of the payload needs to be encrypted.
 	 * So mark the last skb in the write queue as end of record.
 	 */
-	skb = tcp_write_queue_tail(sk);
-	if (skb)
-		TCP_SKB_CB(skb)->eor = 1;
+	tcp_write_collapse_fence(sk);
 
 	/* Avoid offloading if the device is down
 	 * We don't want to offload new flows after
-- 
cgit v1.2.3-58-ga151


From 99b8add01f98a8ecf498de1467e8bf13dbf02daa Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 30 May 2024 16:36:16 -0700
Subject: net: skb: add compatibility warnings to skb_shift()

According to current semantics we should never try to shift data
between skbs which differ on decrypted or pp_recycle status.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/skbuff.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 466999a7515e..c8ac79851cd6 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -4139,6 +4139,9 @@ int skb_shift(struct sk_buff *tgt, struct sk_buff *skb, int shiftlen)
 	if (skb_zcopy(tgt) || skb_zcopy(skb))
 		return 0;
 
+	DEBUG_NET_WARN_ON_ONCE(tgt->pp_recycle != skb->pp_recycle);
+	DEBUG_NET_WARN_ON_ONCE(skb_cmp_decrypted(tgt, skb));
+
 	todo = shiftlen;
 	from = 0;
 	to = skb_shinfo(tgt)->nr_frags;
-- 
cgit v1.2.3-58-ga151


From 8c3fdff2171c834df5fa5ff353b94ada2e5376ca Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Fri, 31 May 2024 04:15:49 -0700
Subject: openvswitch: Move stats allocation to core

With commit 34d21de99cea9 ("net: Move {l,t,d}stats allocation to core and
convert veth & vrf"), stats allocation could be done on net core instead
of this driver.

With this new approach, the driver doesn't have to bother with error
handling (allocation failure checking, making sure free happens in the
right spot, etc). This is core responsibility now.

Move openvswitch driver to leverage the core allocation.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240531111552.3209198-1-leitao@debian.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/openvswitch/vport-internal_dev.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c
index 74c88a6baa43..7daba6ac6912 100644
--- a/net/openvswitch/vport-internal_dev.c
+++ b/net/openvswitch/vport-internal_dev.c
@@ -140,11 +140,7 @@ static struct vport *internal_dev_create(const struct vport_parms *parms)
 		err = -ENOMEM;
 		goto error_free_vport;
 	}
-	vport->dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
-	if (!vport->dev->tstats) {
-		err = -ENOMEM;
-		goto error_free_netdev;
-	}
+	dev->pcpu_stat_type = NETDEV_PCPU_STAT_TSTATS;
 
 	dev_net_set(vport->dev, ovs_dp_get_net(vport->dp));
 	dev->ifindex = parms->desired_ifindex;
@@ -169,8 +165,6 @@ static struct vport *internal_dev_create(const struct vport_parms *parms)
 
 error_unlock:
 	rtnl_unlock();
-	free_percpu(dev->tstats);
-error_free_netdev:
 	free_netdev(dev);
 error_free_vport:
 	ovs_vport_free(vport);
@@ -186,7 +180,6 @@ static void internal_dev_destroy(struct vport *vport)
 
 	/* unregister_netdevice() waits for an RCU grace period. */
 	unregister_netdevice(vport->dev);
-	free_percpu(vport->dev->tstats);
 	rtnl_unlock();
 }
 
-- 
cgit v1.2.3-58-ga151


From 2b438c5774cc491a4aa8bb1ec4f49c1a0760a173 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Fri, 31 May 2024 04:15:50 -0700
Subject: openvswitch: Remove generic .ndo_get_stats64

Commit 3e2f544dd8a33 ("net: get stats64 if device if driver is
configured") moved the callback to dev_get_tstats64() to net core, so,
unless the driver is doing some custom stats collection, it does not
need to set .ndo_get_stats64.

Since this driver is now relying in NETDEV_PCPU_STAT_TSTATS, then, it
doesn't need to set the dev_get_tstats64() generic .ndo_get_stats64
function pointer.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Subbaraya Sundeep <sbhatta@marvell.com>
Link: https://lore.kernel.org/r/20240531111552.3209198-2-leitao@debian.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/openvswitch/vport-internal_dev.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c
index 7daba6ac6912..4b33133cbdff 100644
--- a/net/openvswitch/vport-internal_dev.c
+++ b/net/openvswitch/vport-internal_dev.c
@@ -85,7 +85,6 @@ static const struct net_device_ops internal_dev_netdev_ops = {
 	.ndo_stop = internal_dev_stop,
 	.ndo_start_xmit = internal_dev_xmit,
 	.ndo_set_mac_address = eth_mac_addr,
-	.ndo_get_stats64 = dev_get_tstats64,
 };
 
 static struct rtnl_link_ops internal_dev_link_ops __read_mostly = {
-- 
cgit v1.2.3-58-ga151


From a6ba5125f10bd7307e775e585ad21a8f7eda1b59 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 4 Jun 2024 10:54:57 -0700
Subject: Revert "ethernet: octeontx2: avoid linking objects into multiple
 modules"

This reverts commit 727c94c9539aa8865cdbf6a783da6a6585f1fec2.

Stephen reports that this commit causes a circular module dependency
for him. Revert, and we'll try to address the problem, again.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Link: https://lore.kernel.org/all/20240531152223.25591c8e@canb.auug.org.au
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/marvell/octeontx2/nic/Makefile       | 14 ++++++--------
 drivers/net/ethernet/marvell/octeontx2/nic/otx2_dcbnl.c   | 11 -----------
 drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c |  6 ------
 3 files changed, 6 insertions(+), 25 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
index e4c5dc46dd42..5664f768cb0c 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
@@ -3,18 +3,16 @@
 # Makefile for Marvell's RVU Ethernet device drivers
 #
 
-obj-$(CONFIG_OCTEONTX2_PF) += rvu_nicpf.o otx2_ptp.o otx2_devlink.o
-obj-$(CONFIG_OCTEONTX2_VF) += rvu_nicvf.o otx2_ptp.o otx2_devlink.o
+obj-$(CONFIG_OCTEONTX2_PF) += rvu_nicpf.o otx2_ptp.o
+obj-$(CONFIG_OCTEONTX2_VF) += rvu_nicvf.o otx2_ptp.o
 
 rvu_nicpf-y := otx2_pf.o otx2_common.o otx2_txrx.o otx2_ethtool.o \
                otx2_flows.o otx2_tc.o cn10k.o otx2_dmac_flt.o \
-               qos_sq.o qos.o
-rvu_nicvf-y := otx2_vf.o
+               otx2_devlink.o qos_sq.o qos.o
+rvu_nicvf-y := otx2_vf.o otx2_devlink.o
 
-ifdef CONFIG_DCB
-obj-$(CONFIG_OCTEONTX2_PF) += otx2_dcbnl.o
-obj-$(CONFIG_OCTEONTX2_VF) += otx2_dcbnl.o
-endif
+rvu_nicpf-$(CONFIG_DCB) += otx2_dcbnl.o
+rvu_nicvf-$(CONFIG_DCB) += otx2_dcbnl.o
 rvu_nicpf-$(CONFIG_MACSEC) += cn10k_macsec.o
 
 ccflags-y += -I$(srctree)/drivers/net/ethernet/marvell/octeontx2/af
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_dcbnl.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_dcbnl.c
index 0d7e611d9a05..28fb643d2917 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_dcbnl.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_dcbnl.c
@@ -54,7 +54,6 @@ int otx2_pfc_txschq_config(struct otx2_nic *pfvf)
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(otx2_pfc_txschq_config);
 
 static int otx2_pfc_txschq_alloc_one(struct otx2_nic *pfvf, u8 prio)
 {
@@ -123,7 +122,6 @@ int otx2_pfc_txschq_alloc(struct otx2_nic *pfvf)
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(otx2_pfc_txschq_alloc);
 
 static int otx2_pfc_txschq_stop_one(struct otx2_nic *pfvf, u8 prio)
 {
@@ -262,7 +260,6 @@ update_sq_smq_map:
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(otx2_pfc_txschq_update);
 
 int otx2_pfc_txschq_stop(struct otx2_nic *pfvf)
 {
@@ -285,7 +282,6 @@ int otx2_pfc_txschq_stop(struct otx2_nic *pfvf)
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(otx2_pfc_txschq_stop);
 
 int otx2_config_priority_flow_ctrl(struct otx2_nic *pfvf)
 {
@@ -325,7 +321,6 @@ unlock:
 	mutex_unlock(&pfvf->mbox.lock);
 	return err;
 }
-EXPORT_SYMBOL_GPL(otx2_config_priority_flow_ctrl);
 
 void otx2_update_bpid_in_rqctx(struct otx2_nic *pfvf, int vlan_prio, int qidx,
 			       bool pfc_enable)
@@ -390,7 +385,6 @@ out:
 			 "Updating BPIDs in CQ and Aura contexts of RQ%d failed with err %d\n",
 			 qidx, err);
 }
-EXPORT_SYMBOL_GPL(otx2_update_bpid_in_rqctx);
 
 static int otx2_dcbnl_ieee_getpfc(struct net_device *dev, struct ieee_pfc *pfc)
 {
@@ -478,8 +472,3 @@ int otx2_dcbnl_set_ops(struct net_device *dev)
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(otx2_dcbnl_set_ops);
-
-MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION("Marvell RVU dcbnl");
-MODULE_AUTHOR("Sunil Goutham <sgoutham@marvell.com>");
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c
index 440f574d1195..99ddf31269d9 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c
@@ -113,7 +113,6 @@ err_dl:
 	devlink_free(dl);
 	return err;
 }
-EXPORT_SYMBOL_GPL(otx2_register_dl);
 
 void otx2_unregister_dl(struct otx2_nic *pfvf)
 {
@@ -125,8 +124,3 @@ void otx2_unregister_dl(struct otx2_nic *pfvf)
 				  ARRAY_SIZE(otx2_dl_params));
 	devlink_free(dl);
 }
-EXPORT_SYMBOL_GPL(otx2_unregister_dl);
-
-MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION("Marvell RVU PF/VF Netdev Devlink");
-MODULE_AUTHOR("Sunil Goutham <sgoutham@marvell.com>");
-- 
cgit v1.2.3-58-ga151


From 61ce0ea7591fef2eb6e89ce40ffcc24fda4dbbc5 Mon Sep 17 00:00:00 2001
From: Song Liu <song@kernel.org>
Date: Tue, 4 Jun 2024 00:07:00 -0700
Subject: selftests/bpf: Fix bpf_cookie and find_vma in nested VM

bpf_cookie and find_vma are flaky in nested VMs, which is used by some CI
systems. It turns out these failures are caused by unreliable perf event
in nested VM. Fix these by:

  1. Use PERF_COUNT_SW_CPU_CLOCK in find_vma;
  2. Increase sample_freq in bpf_cookie.

Signed-off-by: Song Liu <song@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240604070700.3032142-1-song@kernel.org
---
 tools/testing/selftests/bpf/prog_tests/bpf_cookie.c | 2 +-
 tools/testing/selftests/bpf/prog_tests/find_vma.c   | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c b/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c
index 4407ea428e77..070c52c312e5 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c
@@ -451,7 +451,7 @@ static void pe_subtest(struct test_bpf_cookie *skel)
 	attr.type = PERF_TYPE_SOFTWARE;
 	attr.config = PERF_COUNT_SW_CPU_CLOCK;
 	attr.freq = 1;
-	attr.sample_freq = 1000;
+	attr.sample_freq = 10000;
 	pfd = syscall(__NR_perf_event_open, &attr, -1, 0, -1, PERF_FLAG_FD_CLOEXEC);
 	if (!ASSERT_GE(pfd, 0, "perf_fd"))
 		goto cleanup;
diff --git a/tools/testing/selftests/bpf/prog_tests/find_vma.c b/tools/testing/selftests/bpf/prog_tests/find_vma.c
index 5165b38f0e59..f7619e0ade10 100644
--- a/tools/testing/selftests/bpf/prog_tests/find_vma.c
+++ b/tools/testing/selftests/bpf/prog_tests/find_vma.c
@@ -29,8 +29,8 @@ static int open_pe(void)
 
 	/* create perf event */
 	attr.size = sizeof(attr);
-	attr.type = PERF_TYPE_HARDWARE;
-	attr.config = PERF_COUNT_HW_CPU_CYCLES;
+	attr.type = PERF_TYPE_SOFTWARE;
+	attr.config = PERF_COUNT_SW_CPU_CLOCK;
 	attr.freq = 1;
 	attr.sample_freq = 1000;
 	pfd = syscall(__NR_perf_event_open, &attr, 0, -1, -1, PERF_FLAG_FD_CLOEXEC);
-- 
cgit v1.2.3-58-ga151


From 898ac74c5b5f8b551a377d6a60ca4e30023ac9d2 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yonghong.song@linux.dev>
Date: Tue, 4 Jun 2024 11:00:34 -0700
Subject: selftests/bpf: Ignore .llvm.<hash> suffix in kallsyms_find()

I hit the following failure when running selftests with
internal backported upstream kernel:
  test_ksyms:PASS:kallsyms_fopen 0 nsec
  test_ksyms:FAIL:ksym_find symbol 'bpf_link_fops' not found
  #123     ksyms:FAIL

In /proc/kallsyms, we have
  $ cat /proc/kallsyms | grep bpf_link_fops
  ffffffff829f0cb0 d bpf_link_fops.llvm.12608678492448798416
The CONFIG_LTO_CLANG_THIN is enabled in the kernel which is responsible
for bpf_link_fops.llvm.12608678492448798416 symbol name.

In prog_tests/ksyms.c we have
  kallsyms_find("bpf_link_fops", &link_fops_addr)
and kallsyms_find() compares "bpf_link_fops" with symbols
in /proc/kallsyms in order to find the entry. With
bpf_link_fops.llvm.<hash> in /proc/kallsyms, the kallsyms_find()
failed.

To fix the issue, in kallsyms_find(), if a symbol has suffix
.llvm.<hash>, that suffix will be ignored for comparison.
This fixed the test failure.

Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240604180034.1356016-1-yonghong.song@linux.dev
---
 tools/testing/selftests/bpf/trace_helpers.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/trace_helpers.c b/tools/testing/selftests/bpf/trace_helpers.c
index 70e29f316fe7..465d196c7165 100644
--- a/tools/testing/selftests/bpf/trace_helpers.c
+++ b/tools/testing/selftests/bpf/trace_helpers.c
@@ -211,7 +211,7 @@ long ksym_get_addr(const char *name)
  */
 int kallsyms_find(const char *sym, unsigned long long *addr)
 {
-	char type, name[500];
+	char type, name[500], *match;
 	unsigned long long value;
 	int err = 0;
 	FILE *f;
@@ -221,6 +221,17 @@ int kallsyms_find(const char *sym, unsigned long long *addr)
 		return -EINVAL;
 
 	while (fscanf(f, "%llx %c %499s%*[^\n]\n", &value, &type, name) > 0) {
+		/* If CONFIG_LTO_CLANG_THIN is enabled, static variable/function
+		 * symbols could be promoted to global due to cross-file inlining.
+		 * For such cases, clang compiler will add .llvm.<hash> suffix
+		 * to those symbols to avoid potential naming conflict.
+		 * Let us ignore .llvm.<hash> suffix during symbol comparison.
+		 */
+		if (type == 'd') {
+			match = strstr(name, ".llvm.");
+			if (match)
+				*match = '\0';
+		}
 		if (strcmp(name, sym) == 0) {
 			*addr = value;
 			goto out;
-- 
cgit v1.2.3-58-ga151


From 3374136f3137ecdb34be228c715db6d6e0b97476 Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@linaro.org>
Date: Thu, 30 May 2024 23:53:07 +0200
Subject: dt-bindings: dsa: Rewrite Vitesse VSC73xx in schema

This rewrites the Vitesse VSC73xx DSA switches DT binding in
schema.

It was a bit tricky since I needed to come up with some way
of applying the SPI properties only on SPI devices and not
platform devices, but I figured something out that works.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../bindings/net/dsa/vitesse,vsc73xx.txt           | 129 ----------------
 .../bindings/net/dsa/vitesse,vsc73xx.yaml          | 162 +++++++++++++++++++++
 2 files changed, 162 insertions(+), 129 deletions(-)
 delete mode 100644 Documentation/devicetree/bindings/net/dsa/vitesse,vsc73xx.txt
 create mode 100644 Documentation/devicetree/bindings/net/dsa/vitesse,vsc73xx.yaml

diff --git a/Documentation/devicetree/bindings/net/dsa/vitesse,vsc73xx.txt b/Documentation/devicetree/bindings/net/dsa/vitesse,vsc73xx.txt
deleted file mode 100644
index 258bef483673..000000000000
--- a/Documentation/devicetree/bindings/net/dsa/vitesse,vsc73xx.txt
+++ /dev/null
@@ -1,129 +0,0 @@
-Vitesse VSC73xx Switches
-========================
-
-This defines device tree bindings for the Vitesse VSC73xx switch chips.
-The Vitesse company has been acquired by Microsemi and Microsemi has
-been acquired Microchip but retains this vendor branding.
-
-The currently supported switch chips are:
-Vitesse VSC7385 SparX-G5 5+1-port Integrated Gigabit Ethernet Switch
-Vitesse VSC7388 SparX-G8 8-port Integrated Gigabit Ethernet Switch
-Vitesse VSC7395 SparX-G5e 5+1-port Integrated Gigabit Ethernet Switch
-Vitesse VSC7398 SparX-G8e 8-port Integrated Gigabit Ethernet Switch
-
-This switch could have two different management interface.
-
-If SPI interface is used, the device tree node is an SPI device so it must
-reside inside a SPI bus device tree node, see spi/spi-bus.txt
-
-When the chip is connected to a parallel memory bus and work in memory-mapped
-I/O mode, a platform device is used to represent the vsc73xx. In this case it
-must reside inside a platform bus device tree node.
-
-Required properties:
-
-- compatible: must be exactly one of:
-	"vitesse,vsc7385"
-	"vitesse,vsc7388"
-	"vitesse,vsc7395"
-	"vitesse,vsc7398"
-- gpio-controller: indicates that this switch is also a GPIO controller,
-  see gpio/gpio.txt
-- #gpio-cells: this must be set to <2> and indicates that we are a twocell
-  GPIO controller, see gpio/gpio.txt
-
-Optional properties:
-
-- reset-gpios: a handle to a GPIO line that can issue reset of the chip.
-  It should be tagged as active low.
-
-Required subnodes:
-
-See net/dsa/dsa.txt for a list of additional required and optional properties
-and subnodes of DSA switches.
-
-Examples:
-
-SPI:
-switch@0 {
-	compatible = "vitesse,vsc7395";
-	reg = <0>;
-	/* Specified for 2.5 MHz or below */
-	spi-max-frequency = <2500000>;
-	gpio-controller;
-	#gpio-cells = <2>;
-
-	ports {
-		#address-cells = <1>;
-		#size-cells = <0>;
-
-		port@0 {
-			reg = <0>;
-			label = "lan1";
-		};
-		port@1 {
-			reg = <1>;
-			label = "lan2";
-		};
-		port@2 {
-			reg = <2>;
-			label = "lan3";
-		};
-		port@3 {
-			reg = <3>;
-			label = "lan4";
-		};
-		vsc: port@6 {
-			reg = <6>;
-			ethernet = <&gmac1>;
-			phy-mode = "rgmii";
-			fixed-link {
-				speed = <1000>;
-				full-duplex;
-				pause;
-			};
-		};
-	};
-};
-
-Platform:
-switch@2,0 {
-	#address-cells = <1>;
-	#size-cells = <1>;
-	compatible = "vitesse,vsc7385";
-	reg = <0x2 0x0 0x20000>;
-	reset-gpios = <&gpio0 12 GPIO_ACTIVE_LOW>;
-
-	ports {
-		#address-cells = <1>;
-		#size-cells = <0>;
-
-		port@0 {
-			reg = <0>;
-			label = "lan1";
-		};
-		port@1 {
-			reg = <1>;
-			label = "lan2";
-		};
-		port@2 {
-			reg = <2>;
-			label = "lan3";
-		};
-		port@3 {
-			reg = <3>;
-			label = "lan4";
-		};
-		vsc: port@6 {
-			reg = <6>;
-			ethernet = <&enet0>;
-			phy-mode = "rgmii";
-			fixed-link {
-				speed = <1000>;
-				full-duplex;
-				pause;
-			};
-		};
-	};
-
-};
diff --git a/Documentation/devicetree/bindings/net/dsa/vitesse,vsc73xx.yaml b/Documentation/devicetree/bindings/net/dsa/vitesse,vsc73xx.yaml
new file mode 100644
index 000000000000..b99d7a694b70
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/dsa/vitesse,vsc73xx.yaml
@@ -0,0 +1,162 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/dsa/vitesse,vsc73xx.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Vitesse VSC73xx DSA Switches
+
+maintainers:
+  - Linus Walleij <linus.walleij@linaro.org>
+
+description:
+  The Vitesse DSA Switches were produced in the early-to-mid 2000s.
+
+  The Vitesse company has been acquired by Microsemi and Microsemi has
+  been acquired Microchip but the new owner retains this vendor branding.
+
+  The currently supported switch chips are
+  Vitesse VSC7385 SparX-G5 5+1-port Integrated Gigabit Ethernet Switch
+  Vitesse VSC7388 SparX-G8 8-port Integrated Gigabit Ethernet Switch
+  Vitesse VSC7395 SparX-G5e 5+1-port Integrated Gigabit Ethernet Switch
+  Vitesse VSC7398 SparX-G8e 8-port Integrated Gigabit Ethernet Switch
+
+  This switch can use one of two different management interfaces.
+
+  If SPI interface is used, the device tree node is an SPI device so it must
+  reside inside a SPI bus device tree node, see spi/spi-bus.txt
+
+  When the chip is connected to a parallel memory bus and work in memory-mapped
+  I/O mode, a platform device is used to represent the vsc73xx. In this case it
+  must reside inside a platform bus device tree node.
+
+properties:
+  compatible:
+    enum:
+      - vitesse,vsc7385
+      - vitesse,vsc7388
+      - vitesse,vsc7395
+      - vitesse,vsc7398
+
+  reg:
+    maxItems: 1
+
+  gpio-controller: true
+  "#gpio-cells":
+    const: 2
+
+  reset-gpios:
+    description: GPIO to be used to reset the whole device
+    maxItems: 1
+
+allOf:
+  - $ref: dsa.yaml#/$defs/ethernet-ports
+
+# This checks if reg is a chipselect so the device is on an SPI
+# bus, the if-clause will fail if reg is a tuple such as for a
+# platform device.
+if:
+  properties:
+    reg:
+      minimum: 0
+      maximum: 256
+then:
+  $ref: /schemas/spi/spi-peripheral-props.yaml#
+
+required:
+  - compatible
+  - reg
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    spi {
+      #address-cells = <1>;
+      #size-cells = <0>;
+
+      ethernet-switch@0 {
+        compatible = "vitesse,vsc7395";
+        reg = <0>;
+        spi-max-frequency = <2500000>;
+        gpio-controller;
+        #gpio-cells = <2>;
+
+        ethernet-ports {
+          #address-cells = <1>;
+          #size-cells = <0>;
+
+          ethernet-port@0 {
+            reg = <0>;
+            label = "lan1";
+          };
+          ethernet-port@1 {
+            reg = <1>;
+            label = "lan2";
+          };
+          ethernet-port@2 {
+            reg = <2>;
+            label = "lan3";
+          };
+          ethernet-port@3 {
+            reg = <3>;
+            label = "lan4";
+          };
+          ethernet-port@6 {
+            reg = <6>;
+            ethernet = <&gmac1>;
+            phy-mode = "rgmii";
+            fixed-link {
+              speed = <1000>;
+              full-duplex;
+              pause;
+            };
+          };
+        };
+      };
+    };
+
+    bus {
+      #address-cells = <1>;
+      #size-cells = <1>;
+
+      ethernet-switch@10000000 {
+        compatible = "vitesse,vsc7385";
+        reg = <0x10000000 0x20000>;
+        reset-gpios = <&gpio0 12 GPIO_ACTIVE_LOW>;
+
+        ethernet-ports {
+          #address-cells = <1>;
+          #size-cells = <0>;
+
+          ethernet-port@0 {
+            reg = <0>;
+            label = "lan1";
+          };
+          ethernet-port@1 {
+            reg = <1>;
+            label = "lan2";
+          };
+          ethernet-port@2 {
+            reg = <2>;
+            label = "lan3";
+          };
+          ethernet-port@3 {
+            reg = <3>;
+            label = "lan4";
+          };
+          ethernet-port@6 {
+            reg = <6>;
+            ethernet = <&enet0>;
+            phy-mode = "rgmii";
+            fixed-link {
+              speed = <1000>;
+              full-duplex;
+              pause;
+            };
+          };
+        };
+      };
+    };
-- 
cgit v1.2.3-58-ga151


From cdbdb3c62af5bc13e6e8a9310470c71c33340d26 Mon Sep 17 00:00:00 2001
From: Chen Hanxiao <chenhx.fnst@fujitsu.com>
Date: Fri, 31 May 2024 16:54:02 +0800
Subject: net: bridge: fix an inconsistent indentation

Smatch complains:
net/bridge/br_netlink_tunnel.c:
   318 br_process_vlan_tunnel_info() warn: inconsistent indenting

Fix it with a proper indenting

Signed-off-by: Chen Hanxiao <chenhx.fnst@fujitsu.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_netlink_tunnel.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/bridge/br_netlink_tunnel.c b/net/bridge/br_netlink_tunnel.c
index 17abf092f7ca..71a12da30004 100644
--- a/net/bridge/br_netlink_tunnel.c
+++ b/net/bridge/br_netlink_tunnel.c
@@ -315,8 +315,8 @@ int br_process_vlan_tunnel_info(const struct net_bridge *br,
 
 			if (curr_change)
 				*changed = curr_change;
-			 __vlan_tunnel_handle_range(p, &v_start, &v_end, v,
-						    curr_change);
+			__vlan_tunnel_handle_range(p, &v_start, &v_end, v,
+						   curr_change);
 		}
 		if (v_start && v_end)
 			br_vlan_notify(br, p, v_start->vid, v_end->vid,
-- 
cgit v1.2.3-58-ga151


From 61e2bbafb00e4b9a5de45e6448a7b6b818658576 Mon Sep 17 00:00:00 2001
From: Jason Xing <kernelxing@tencent.com>
Date: Fri, 31 May 2024 23:46:34 +0800
Subject: net: remove NULL-pointer net parameter in ip_metrics_convert

When I was doing some experiments, I found that when using the first
parameter, namely, struct net, in ip_metrics_convert() always triggers NULL
pointer crash. Then I digged into this part, realizing that we can remove
this one due to its uselessness.

Signed-off-by: Jason Xing <kernelxing@tencent.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ip.h         |  3 +--
 include/net/tcp.h        |  2 +-
 net/ipv4/fib_semantics.c |  5 ++---
 net/ipv4/metrics.c       |  8 ++++----
 net/ipv4/tcp_cong.c      | 11 +++++------
 net/ipv6/route.c         |  2 +-
 6 files changed, 14 insertions(+), 17 deletions(-)

diff --git a/include/net/ip.h b/include/net/ip.h
index 6d735e00d3f3..c5606cadb1a5 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -506,8 +506,7 @@ static inline unsigned int ip_skb_dst_mtu(struct sock *sk,
 	return mtu - lwtunnel_headroom(skb_dst(skb)->lwtstate, mtu);
 }
 
-struct dst_metrics *ip_fib_metrics_init(struct net *net, struct nlattr *fc_mx,
-					int fc_mx_len,
+struct dst_metrics *ip_fib_metrics_init(struct nlattr *fc_mx, int fc_mx_len,
 					struct netlink_ext_ack *extack);
 static inline void ip_fib_metrics_put(struct dst_metrics *fib_metrics)
 {
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 08c3b99501cf..a70fc39090fe 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1224,7 +1224,7 @@ extern struct tcp_congestion_ops tcp_reno;
 
 struct tcp_congestion_ops *tcp_ca_find(const char *name);
 struct tcp_congestion_ops *tcp_ca_find_key(u32 key);
-u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca);
+u32 tcp_ca_get_key_by_name(const char *name, bool *ecn_ca);
 #ifdef CONFIG_INET
 char *tcp_ca_get_name_by_key(u32 key, char *buffer);
 #else
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index f669da98d11d..7b6b042208bd 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -1030,7 +1030,7 @@ bool fib_metrics_match(struct fib_config *cfg, struct fib_info *fi)
 			bool ecn_ca = false;
 
 			nla_strscpy(tmp, nla, sizeof(tmp));
-			val = tcp_ca_get_key_by_name(fi->fib_net, tmp, &ecn_ca);
+			val = tcp_ca_get_key_by_name(tmp, &ecn_ca);
 		} else {
 			if (nla_len(nla) != sizeof(u32))
 				return false;
@@ -1459,8 +1459,7 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
 	fi = kzalloc(struct_size(fi, fib_nh, nhs), GFP_KERNEL);
 	if (!fi)
 		goto failure;
-	fi->fib_metrics = ip_fib_metrics_init(fi->fib_net, cfg->fc_mx,
-					      cfg->fc_mx_len, extack);
+	fi->fib_metrics = ip_fib_metrics_init(cfg->fc_mx, cfg->fc_mx_len, extack);
 	if (IS_ERR(fi->fib_metrics)) {
 		err = PTR_ERR(fi->fib_metrics);
 		kfree(fi);
diff --git a/net/ipv4/metrics.c b/net/ipv4/metrics.c
index 0e3ee1532848..8ddac1f595ed 100644
--- a/net/ipv4/metrics.c
+++ b/net/ipv4/metrics.c
@@ -7,7 +7,7 @@
 #include <net/net_namespace.h>
 #include <net/tcp.h>
 
-static int ip_metrics_convert(struct net *net, struct nlattr *fc_mx,
+static int ip_metrics_convert(struct nlattr *fc_mx,
 			      int fc_mx_len, u32 *metrics,
 			      struct netlink_ext_ack *extack)
 {
@@ -31,7 +31,7 @@ static int ip_metrics_convert(struct net *net, struct nlattr *fc_mx,
 			char tmp[TCP_CA_NAME_MAX];
 
 			nla_strscpy(tmp, nla, sizeof(tmp));
-			val = tcp_ca_get_key_by_name(net, tmp, &ecn_ca);
+			val = tcp_ca_get_key_by_name(tmp, &ecn_ca);
 			if (val == TCP_CA_UNSPEC) {
 				NL_SET_ERR_MSG(extack, "Unknown tcp congestion algorithm");
 				return -EINVAL;
@@ -63,7 +63,7 @@ static int ip_metrics_convert(struct net *net, struct nlattr *fc_mx,
 	return 0;
 }
 
-struct dst_metrics *ip_fib_metrics_init(struct net *net, struct nlattr *fc_mx,
+struct dst_metrics *ip_fib_metrics_init(struct nlattr *fc_mx,
 					int fc_mx_len,
 					struct netlink_ext_ack *extack)
 {
@@ -77,7 +77,7 @@ struct dst_metrics *ip_fib_metrics_init(struct net *net, struct nlattr *fc_mx,
 	if (unlikely(!fib_metrics))
 		return ERR_PTR(-ENOMEM);
 
-	err = ip_metrics_convert(net, fc_mx, fc_mx_len, fib_metrics->metrics,
+	err = ip_metrics_convert(fc_mx, fc_mx_len, fib_metrics->metrics,
 				 extack);
 	if (!err) {
 		refcount_set(&fib_metrics->refcnt, 1);
diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
index 28ffcfbeef14..48617d99abb0 100644
--- a/net/ipv4/tcp_cong.c
+++ b/net/ipv4/tcp_cong.c
@@ -46,8 +46,7 @@ void tcp_set_ca_state(struct sock *sk, const u8 ca_state)
 }
 
 /* Must be called with rcu lock held */
-static struct tcp_congestion_ops *tcp_ca_find_autoload(struct net *net,
-						       const char *name)
+static struct tcp_congestion_ops *tcp_ca_find_autoload(const char *name)
 {
 	struct tcp_congestion_ops *ca = tcp_ca_find(name);
 
@@ -178,7 +177,7 @@ int tcp_update_congestion_control(struct tcp_congestion_ops *ca, struct tcp_cong
 	return ret;
 }
 
-u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca)
+u32 tcp_ca_get_key_by_name(const char *name, bool *ecn_ca)
 {
 	const struct tcp_congestion_ops *ca;
 	u32 key = TCP_CA_UNSPEC;
@@ -186,7 +185,7 @@ u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca)
 	might_sleep();
 
 	rcu_read_lock();
-	ca = tcp_ca_find_autoload(net, name);
+	ca = tcp_ca_find_autoload(name);
 	if (ca) {
 		key = ca->key;
 		*ecn_ca = ca->flags & TCP_CONG_NEEDS_ECN;
@@ -283,7 +282,7 @@ int tcp_set_default_congestion_control(struct net *net, const char *name)
 	int ret;
 
 	rcu_read_lock();
-	ca = tcp_ca_find_autoload(net, name);
+	ca = tcp_ca_find_autoload(name);
 	if (!ca) {
 		ret = -ENOENT;
 	} else if (!bpf_try_module_get(ca, ca->owner)) {
@@ -421,7 +420,7 @@ int tcp_set_congestion_control(struct sock *sk, const char *name, bool load,
 	if (!load)
 		ca = tcp_ca_find(name);
 	else
-		ca = tcp_ca_find_autoload(sock_net(sk), name);
+		ca = tcp_ca_find_autoload(name);
 
 	/* No change asking for existing value */
 	if (ca == icsk->icsk_ca_ops) {
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index a504b88ec06b..12f647d0fec0 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -3761,7 +3761,7 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
 	if (!rt)
 		goto out;
 
-	rt->fib6_metrics = ip_fib_metrics_init(net, cfg->fc_mx, cfg->fc_mx_len,
+	rt->fib6_metrics = ip_fib_metrics_init(cfg->fc_mx, cfg->fc_mx_len,
 					       extack);
 	if (IS_ERR(rt->fib6_metrics)) {
 		err = PTR_ERR(rt->fib6_metrics);
-- 
cgit v1.2.3-58-ga151


From 6f49c3fb563c0a95a838216eaf7d9b02ece44bf5 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Sat, 1 Jun 2024 00:29:17 +0100
Subject: net: caif: remove unused structs

'cfpktq' has been unused since
commit 73d6ac633c6c ("caif: code cleanup").

'caif_packet_funcs' is declared but never defined.

Remove both of them.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/caif/caif_layer.h | 2 --
 net/caif/cfpkt_skbuff.c       | 7 -------
 2 files changed, 9 deletions(-)

diff --git a/include/net/caif/caif_layer.h b/include/net/caif/caif_layer.h
index 51f7bb42a936..0f45d875905f 100644
--- a/include/net/caif/caif_layer.h
+++ b/include/net/caif/caif_layer.h
@@ -11,9 +11,7 @@
 
 struct cflayer;
 struct cfpkt;
-struct cfpktq;
 struct caif_payload_info;
-struct caif_packet_funcs;
 
 #define CAIF_LAYER_NAME_SZ 16
 
diff --git a/net/caif/cfpkt_skbuff.c b/net/caif/cfpkt_skbuff.c
index 7796414d47e5..2ae8cfa3df88 100644
--- a/net/caif/cfpkt_skbuff.c
+++ b/net/caif/cfpkt_skbuff.c
@@ -21,13 +21,6 @@ do {					   \
 	pr_warn(errmsg);		   \
 } while (0)
 
-struct cfpktq {
-	struct sk_buff_head head;
-	atomic_t count;
-	/* Lock protects count updates */
-	spinlock_t lock;
-};
-
 /*
  * net/caif/ is generic and does not
  * understand SKB, so we do this typecast
-- 
cgit v1.2.3-58-ga151


From a23b0034e93453cecf435637dfdb14560eeda6c8 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Sat, 1 Jun 2024 00:30:06 +0100
Subject: net: ethtool: remove unused struct 'cable_test_tdr_req_info'

'cable_test_tdr_req_info' is unused since the original
commit f2bc8ad31a7f ("net: ethtool: Allow PHY cable test TDR data to
configured").

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/cabletest.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/net/ethtool/cabletest.c b/net/ethtool/cabletest.c
index 06a151165c31..f6f136ec7ddf 100644
--- a/net/ethtool/cabletest.c
+++ b/net/ethtool/cabletest.c
@@ -207,10 +207,6 @@ err:
 }
 EXPORT_SYMBOL_GPL(ethnl_cable_test_fault_length);
 
-struct cable_test_tdr_req_info {
-	struct ethnl_req_info		base;
-};
-
 static const struct nla_policy cable_test_tdr_act_cfg_policy[] = {
 	[ETHTOOL_A_CABLE_TEST_TDR_CFG_FIRST]	= { .type = NLA_U32 },
 	[ETHTOOL_A_CABLE_TEST_TDR_CFG_LAST]	= { .type = NLA_U32 },
-- 
cgit v1.2.3-58-ga151


From c11d5dbbe73fa7b450aaa77bb18df86a9714b422 Mon Sep 17 00:00:00 2001
From: Christian Marangi <ansuelsmth@gmail.com>
Date: Sat, 1 Jun 2024 01:35:02 +0200
Subject: net: phy: aquantia: move priv and hw stat to header

In preparation for LEDs support, move priv and hw stat to header to
reference priv struct also in other .c outside aquantia.main

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/aquantia/aquantia.h      | 38 ++++++++++++++++++++++++++++++++
 drivers/net/phy/aquantia/aquantia_main.c | 37 -------------------------------
 2 files changed, 38 insertions(+), 37 deletions(-)

diff --git a/drivers/net/phy/aquantia/aquantia.h b/drivers/net/phy/aquantia/aquantia.h
index 1c19ae74ad2b..c79b33d95628 100644
--- a/drivers/net/phy/aquantia/aquantia.h
+++ b/drivers/net/phy/aquantia/aquantia.h
@@ -87,6 +87,18 @@
 #define VEND1_GLOBAL_RSVD_STAT9_MODE		GENMASK(7, 0)
 #define VEND1_GLOBAL_RSVD_STAT9_1000BT2		0x23
 
+/* MDIO_MMD_C22EXT */
+#define MDIO_C22EXT_STAT_SGMII_RX_GOOD_FRAMES		0xd292
+#define MDIO_C22EXT_STAT_SGMII_RX_BAD_FRAMES		0xd294
+#define MDIO_C22EXT_STAT_SGMII_RX_FALSE_CARRIER		0xd297
+#define MDIO_C22EXT_STAT_SGMII_TX_GOOD_FRAMES		0xd313
+#define MDIO_C22EXT_STAT_SGMII_TX_BAD_FRAMES		0xd315
+#define MDIO_C22EXT_STAT_SGMII_TX_FALSE_CARRIER		0xd317
+#define MDIO_C22EXT_STAT_SGMII_TX_COLLISIONS		0xd318
+#define MDIO_C22EXT_STAT_SGMII_TX_LINE_COLLISIONS	0xd319
+#define MDIO_C22EXT_STAT_SGMII_TX_FRAME_ALIGN_ERR	0xd31a
+#define MDIO_C22EXT_STAT_SGMII_TX_RUNT_FRAMES		0xd31b
+
 #define VEND1_GLOBAL_INT_STD_STATUS		0xfc00
 #define VEND1_GLOBAL_INT_VEND_STATUS		0xfc01
 
@@ -113,6 +125,32 @@
 #define VEND1_GLOBAL_INT_VEND_MASK_GLOBAL2	BIT(1)
 #define VEND1_GLOBAL_INT_VEND_MASK_GLOBAL3	BIT(0)
 
+struct aqr107_hw_stat {
+	const char *name;
+	int reg;
+	int size;
+};
+
+#define SGMII_STAT(n, r, s) { n, MDIO_C22EXT_STAT_SGMII_ ## r, s }
+static const struct aqr107_hw_stat aqr107_hw_stats[] = {
+	SGMII_STAT("sgmii_rx_good_frames",	    RX_GOOD_FRAMES,	26),
+	SGMII_STAT("sgmii_rx_bad_frames",	    RX_BAD_FRAMES,	26),
+	SGMII_STAT("sgmii_rx_false_carrier_events", RX_FALSE_CARRIER,	 8),
+	SGMII_STAT("sgmii_tx_good_frames",	    TX_GOOD_FRAMES,	26),
+	SGMII_STAT("sgmii_tx_bad_frames",	    TX_BAD_FRAMES,	26),
+	SGMII_STAT("sgmii_tx_false_carrier_events", TX_FALSE_CARRIER,	 8),
+	SGMII_STAT("sgmii_tx_collisions",	    TX_COLLISIONS,	 8),
+	SGMII_STAT("sgmii_tx_line_collisions",	    TX_LINE_COLLISIONS,	 8),
+	SGMII_STAT("sgmii_tx_frame_alignment_err",  TX_FRAME_ALIGN_ERR,	16),
+	SGMII_STAT("sgmii_tx_runt_frames",	    TX_RUNT_FRAMES,	22),
+};
+
+#define AQR107_SGMII_STAT_SZ ARRAY_SIZE(aqr107_hw_stats)
+
+struct aqr107_priv {
+	u64 sgmii_stats[AQR107_SGMII_STAT_SZ];
+};
+
 #if IS_REACHABLE(CONFIG_HWMON)
 int aqr_hwmon_probe(struct phy_device *phydev);
 #else
diff --git a/drivers/net/phy/aquantia/aquantia_main.c b/drivers/net/phy/aquantia/aquantia_main.c
index d34cdec47636..252123d12efb 100644
--- a/drivers/net/phy/aquantia/aquantia_main.c
+++ b/drivers/net/phy/aquantia/aquantia_main.c
@@ -84,49 +84,12 @@
 #define MDIO_AN_RX_VEND_STAT3			0xe832
 #define MDIO_AN_RX_VEND_STAT3_AFR		BIT(0)
 
-/* MDIO_MMD_C22EXT */
-#define MDIO_C22EXT_STAT_SGMII_RX_GOOD_FRAMES		0xd292
-#define MDIO_C22EXT_STAT_SGMII_RX_BAD_FRAMES		0xd294
-#define MDIO_C22EXT_STAT_SGMII_RX_FALSE_CARRIER		0xd297
-#define MDIO_C22EXT_STAT_SGMII_TX_GOOD_FRAMES		0xd313
-#define MDIO_C22EXT_STAT_SGMII_TX_BAD_FRAMES		0xd315
-#define MDIO_C22EXT_STAT_SGMII_TX_FALSE_CARRIER		0xd317
-#define MDIO_C22EXT_STAT_SGMII_TX_COLLISIONS		0xd318
-#define MDIO_C22EXT_STAT_SGMII_TX_LINE_COLLISIONS	0xd319
-#define MDIO_C22EXT_STAT_SGMII_TX_FRAME_ALIGN_ERR	0xd31a
-#define MDIO_C22EXT_STAT_SGMII_TX_RUNT_FRAMES		0xd31b
-
 /* Sleep and timeout for checking if the Processor-Intensive
  * MDIO operation is finished
  */
 #define AQR107_OP_IN_PROG_SLEEP		1000
 #define AQR107_OP_IN_PROG_TIMEOUT	100000
 
-struct aqr107_hw_stat {
-	const char *name;
-	int reg;
-	int size;
-};
-
-#define SGMII_STAT(n, r, s) { n, MDIO_C22EXT_STAT_SGMII_ ## r, s }
-static const struct aqr107_hw_stat aqr107_hw_stats[] = {
-	SGMII_STAT("sgmii_rx_good_frames",	    RX_GOOD_FRAMES,	26),
-	SGMII_STAT("sgmii_rx_bad_frames",	    RX_BAD_FRAMES,	26),
-	SGMII_STAT("sgmii_rx_false_carrier_events", RX_FALSE_CARRIER,	 8),
-	SGMII_STAT("sgmii_tx_good_frames",	    TX_GOOD_FRAMES,	26),
-	SGMII_STAT("sgmii_tx_bad_frames",	    TX_BAD_FRAMES,	26),
-	SGMII_STAT("sgmii_tx_false_carrier_events", TX_FALSE_CARRIER,	 8),
-	SGMII_STAT("sgmii_tx_collisions",	    TX_COLLISIONS,	 8),
-	SGMII_STAT("sgmii_tx_line_collisions",	    TX_LINE_COLLISIONS,	 8),
-	SGMII_STAT("sgmii_tx_frame_alignment_err",  TX_FRAME_ALIGN_ERR,	16),
-	SGMII_STAT("sgmii_tx_runt_frames",	    TX_RUNT_FRAMES,	22),
-};
-#define AQR107_SGMII_STAT_SZ ARRAY_SIZE(aqr107_hw_stats)
-
-struct aqr107_priv {
-	u64 sgmii_stats[AQR107_SGMII_STAT_SZ];
-};
-
 static int aqr107_get_sset_count(struct phy_device *phydev)
 {
 	return AQR107_SGMII_STAT_SZ;
-- 
cgit v1.2.3-58-ga151


From 61578f67937881abf54c8bd258eb913312dbe4c1 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Sat, 1 Jun 2024 01:35:03 +0200
Subject: net: phy: aquantia: add support for PHY LEDs

Aquantia Ethernet PHYs got 3 LED output pins which are typically used
to indicate link status and activity.
Add a minimal LED controller driver supporting the most common uses
with the 'netdev' trigger as well as software-driven forced control of
the LEDs.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
[ rework indentation, fix checkpatch error and improve some functions ]
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/aquantia/Makefile        |   2 +-
 drivers/net/phy/aquantia/aquantia.h      |  40 +++++++++
 drivers/net/phy/aquantia/aquantia_leds.c | 150 +++++++++++++++++++++++++++++++
 drivers/net/phy/aquantia/aquantia_main.c |  63 ++++++++++++-
 4 files changed, 252 insertions(+), 3 deletions(-)
 create mode 100644 drivers/net/phy/aquantia/aquantia_leds.c

diff --git a/drivers/net/phy/aquantia/Makefile b/drivers/net/phy/aquantia/Makefile
index aa77fb63c8ec..c6c4d494ee2a 100644
--- a/drivers/net/phy/aquantia/Makefile
+++ b/drivers/net/phy/aquantia/Makefile
@@ -1,5 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
-aquantia-objs			+= aquantia_main.o aquantia_firmware.o
+aquantia-objs			+= aquantia_main.o aquantia_firmware.o aquantia_leds.o
 ifdef CONFIG_HWMON
 aquantia-objs			+= aquantia_hwmon.o
 endif
diff --git a/drivers/net/phy/aquantia/aquantia.h b/drivers/net/phy/aquantia/aquantia.h
index c79b33d95628..c0e1fd9d7152 100644
--- a/drivers/net/phy/aquantia/aquantia.h
+++ b/drivers/net/phy/aquantia/aquantia.h
@@ -63,6 +63,28 @@
 #define VEND1_GLOBAL_CONTROL2_UP_RUN_STALL_OVD	BIT(6)
 #define VEND1_GLOBAL_CONTROL2_UP_RUN_STALL	BIT(0)
 
+#define VEND1_GLOBAL_LED_PROV			0xc430
+#define AQR_LED_PROV(x)				(VEND1_GLOBAL_LED_PROV + (x))
+#define VEND1_GLOBAL_LED_PROV_LINK2500		BIT(14)
+#define VEND1_GLOBAL_LED_PROV_LINK5000		BIT(15)
+#define VEND1_GLOBAL_LED_PROV_FORCE_ON		BIT(8)
+#define VEND1_GLOBAL_LED_PROV_LINK10000		BIT(7)
+#define VEND1_GLOBAL_LED_PROV_LINK1000		BIT(6)
+#define VEND1_GLOBAL_LED_PROV_LINK100		BIT(5)
+#define VEND1_GLOBAL_LED_PROV_RX_ACT		BIT(3)
+#define VEND1_GLOBAL_LED_PROV_TX_ACT		BIT(2)
+#define VEND1_GLOBAL_LED_PROV_ACT_STRETCH	GENMASK(0, 1)
+
+#define VEND1_GLOBAL_LED_PROV_LINK_MASK		(VEND1_GLOBAL_LED_PROV_LINK100 | \
+						 VEND1_GLOBAL_LED_PROV_LINK1000 | \
+						 VEND1_GLOBAL_LED_PROV_LINK10000 | \
+						 VEND1_GLOBAL_LED_PROV_LINK5000 | \
+						 VEND1_GLOBAL_LED_PROV_LINK2500)
+
+#define VEND1_GLOBAL_LED_DRIVE			0xc438
+#define VEND1_GLOBAL_LED_DRIVE_VDD		BIT(1)
+#define AQR_LED_DRIVE(x)			(VEND1_GLOBAL_LED_DRIVE + (x))
+
 #define VEND1_THERMAL_PROV_HIGH_TEMP_FAIL	0xc421
 #define VEND1_THERMAL_PROV_LOW_TEMP_FAIL	0xc422
 #define VEND1_THERMAL_PROV_HIGH_TEMP_WARN	0xc423
@@ -125,6 +147,8 @@
 #define VEND1_GLOBAL_INT_VEND_MASK_GLOBAL2	BIT(1)
 #define VEND1_GLOBAL_INT_VEND_MASK_GLOBAL3	BIT(0)
 
+#define AQR_MAX_LEDS				3
+
 struct aqr107_hw_stat {
 	const char *name;
 	int reg;
@@ -149,6 +173,7 @@ static const struct aqr107_hw_stat aqr107_hw_stats[] = {
 
 struct aqr107_priv {
 	u64 sgmii_stats[AQR107_SGMII_STAT_SZ];
+	unsigned long leds_active_low;
 };
 
 #if IS_REACHABLE(CONFIG_HWMON)
@@ -158,3 +183,18 @@ static inline int aqr_hwmon_probe(struct phy_device *phydev) { return 0; }
 #endif
 
 int aqr_firmware_load(struct phy_device *phydev);
+
+int aqr_phy_led_blink_set(struct phy_device *phydev, u8 index,
+			  unsigned long *delay_on,
+			  unsigned long *delay_off);
+int aqr_phy_led_brightness_set(struct phy_device *phydev,
+			       u8 index, enum led_brightness value);
+int aqr_phy_led_hw_is_supported(struct phy_device *phydev, u8 index,
+				unsigned long rules);
+int aqr_phy_led_hw_control_get(struct phy_device *phydev, u8 index,
+			       unsigned long *rules);
+int aqr_phy_led_hw_control_set(struct phy_device *phydev, u8 index,
+			       unsigned long rules);
+int aqr_phy_led_active_low_set(struct phy_device *phydev, int index, bool enable);
+int aqr_phy_led_polarity_set(struct phy_device *phydev, int index,
+			     unsigned long modes);
diff --git a/drivers/net/phy/aquantia/aquantia_leds.c b/drivers/net/phy/aquantia/aquantia_leds.c
new file mode 100644
index 000000000000..0516ac02c3f8
--- /dev/null
+++ b/drivers/net/phy/aquantia/aquantia_leds.c
@@ -0,0 +1,150 @@
+// SPDX-License-Identifier: GPL-2.0
+/* LED driver for Aquantia PHY
+ *
+ * Author: Daniel Golle <daniel@makrotopia.org>
+ */
+
+#include <linux/phy.h>
+
+#include "aquantia.h"
+
+int aqr_phy_led_brightness_set(struct phy_device *phydev,
+			       u8 index, enum led_brightness value)
+{
+	if (index >= AQR_MAX_LEDS)
+		return -EINVAL;
+
+	return phy_modify_mmd(phydev, MDIO_MMD_VEND1, AQR_LED_PROV(index),
+			      VEND1_GLOBAL_LED_PROV_LINK_MASK |
+			      VEND1_GLOBAL_LED_PROV_FORCE_ON |
+			      VEND1_GLOBAL_LED_PROV_RX_ACT |
+			      VEND1_GLOBAL_LED_PROV_TX_ACT,
+			      value ? VEND1_GLOBAL_LED_PROV_FORCE_ON : 0);
+}
+
+static const unsigned long supported_triggers = (BIT(TRIGGER_NETDEV_LINK) |
+						 BIT(TRIGGER_NETDEV_LINK_100) |
+						 BIT(TRIGGER_NETDEV_LINK_1000) |
+						 BIT(TRIGGER_NETDEV_LINK_2500) |
+						 BIT(TRIGGER_NETDEV_LINK_5000) |
+						 BIT(TRIGGER_NETDEV_LINK_10000)  |
+						 BIT(TRIGGER_NETDEV_RX) |
+						 BIT(TRIGGER_NETDEV_TX));
+
+int aqr_phy_led_hw_is_supported(struct phy_device *phydev, u8 index,
+				unsigned long rules)
+{
+	if (index >= AQR_MAX_LEDS)
+		return -EINVAL;
+
+	/* All combinations of the supported triggers are allowed */
+	if (rules & ~supported_triggers)
+		return -EOPNOTSUPP;
+
+	return 0;
+}
+
+int aqr_phy_led_hw_control_get(struct phy_device *phydev, u8 index,
+			       unsigned long *rules)
+{
+	int val;
+
+	if (index >= AQR_MAX_LEDS)
+		return -EINVAL;
+
+	val = phy_read_mmd(phydev, MDIO_MMD_VEND1, AQR_LED_PROV(index));
+	if (val < 0)
+		return val;
+
+	*rules = 0;
+	if (val & VEND1_GLOBAL_LED_PROV_LINK100)
+		*rules |= BIT(TRIGGER_NETDEV_LINK_100);
+
+	if (val & VEND1_GLOBAL_LED_PROV_LINK1000)
+		*rules |= BIT(TRIGGER_NETDEV_LINK_1000);
+
+	if (val & VEND1_GLOBAL_LED_PROV_LINK2500)
+		*rules |= BIT(TRIGGER_NETDEV_LINK_2500);
+
+	if (val & VEND1_GLOBAL_LED_PROV_LINK5000)
+		*rules |= BIT(TRIGGER_NETDEV_LINK_5000);
+
+	if (val & VEND1_GLOBAL_LED_PROV_LINK10000)
+		*rules |= BIT(TRIGGER_NETDEV_LINK_10000);
+
+	if (val & VEND1_GLOBAL_LED_PROV_RX_ACT)
+		*rules |= BIT(TRIGGER_NETDEV_RX);
+
+	if (val & VEND1_GLOBAL_LED_PROV_TX_ACT)
+		*rules |= BIT(TRIGGER_NETDEV_TX);
+
+	return 0;
+}
+
+int aqr_phy_led_hw_control_set(struct phy_device *phydev, u8 index,
+			       unsigned long rules)
+{
+	u16 val = 0;
+
+	if (index >= AQR_MAX_LEDS)
+		return -EINVAL;
+
+	if (rules & (BIT(TRIGGER_NETDEV_LINK_100) | BIT(TRIGGER_NETDEV_LINK)))
+		val |= VEND1_GLOBAL_LED_PROV_LINK100;
+
+	if (rules & (BIT(TRIGGER_NETDEV_LINK_1000) | BIT(TRIGGER_NETDEV_LINK)))
+		val |= VEND1_GLOBAL_LED_PROV_LINK1000;
+
+	if (rules & (BIT(TRIGGER_NETDEV_LINK_2500) | BIT(TRIGGER_NETDEV_LINK)))
+		val |= VEND1_GLOBAL_LED_PROV_LINK2500;
+
+	if (rules & (BIT(TRIGGER_NETDEV_LINK_5000) | BIT(TRIGGER_NETDEV_LINK)))
+		val |= VEND1_GLOBAL_LED_PROV_LINK5000;
+
+	if (rules & (BIT(TRIGGER_NETDEV_LINK_10000) | BIT(TRIGGER_NETDEV_LINK)))
+		val |= VEND1_GLOBAL_LED_PROV_LINK10000;
+
+	if (rules & BIT(TRIGGER_NETDEV_RX))
+		val |= VEND1_GLOBAL_LED_PROV_RX_ACT;
+
+	if (rules & BIT(TRIGGER_NETDEV_TX))
+		val |= VEND1_GLOBAL_LED_PROV_TX_ACT;
+
+	return phy_modify_mmd(phydev, MDIO_MMD_VEND1, AQR_LED_PROV(index),
+			      VEND1_GLOBAL_LED_PROV_LINK_MASK |
+			      VEND1_GLOBAL_LED_PROV_FORCE_ON |
+			      VEND1_GLOBAL_LED_PROV_RX_ACT |
+			      VEND1_GLOBAL_LED_PROV_TX_ACT, val);
+}
+
+int aqr_phy_led_active_low_set(struct phy_device *phydev, int index, bool enable)
+{
+	return phy_modify_mmd(phydev, MDIO_MMD_VEND1, AQR_LED_DRIVE(index),
+			      VEND1_GLOBAL_LED_DRIVE_VDD, enable);
+}
+
+int aqr_phy_led_polarity_set(struct phy_device *phydev, int index, unsigned long modes)
+{
+	struct aqr107_priv *priv = phydev->priv;
+	bool active_low = false;
+	u32 mode;
+
+	if (index >= AQR_MAX_LEDS)
+		return -EINVAL;
+
+	for_each_set_bit(mode, &modes, __PHY_LED_MODES_NUM) {
+		switch (mode) {
+		case PHY_LED_ACTIVE_LOW:
+			active_low = true;
+			break;
+		default:
+			return -EINVAL;
+		}
+	}
+
+	/* Save LED driver vdd state to restore on SW reset */
+	if (active_low)
+		priv->leds_active_low |= BIT(index);
+
+	return aqr_phy_led_active_low_set(phydev, index, active_low);
+}
diff --git a/drivers/net/phy/aquantia/aquantia_main.c b/drivers/net/phy/aquantia/aquantia_main.c
index 252123d12efb..6c14355744b7 100644
--- a/drivers/net/phy/aquantia/aquantia_main.c
+++ b/drivers/net/phy/aquantia/aquantia_main.c
@@ -475,7 +475,9 @@ static void aqr107_chip_info(struct phy_device *phydev)
 
 static int aqr107_config_init(struct phy_device *phydev)
 {
-	int ret;
+	struct aqr107_priv *priv = phydev->priv;
+	u32 led_active_low;
+	int ret, index = 0;
 
 	/* Check that the PHY interface type is compatible */
 	if (phydev->interface != PHY_INTERFACE_MODE_SGMII &&
@@ -496,7 +498,19 @@ static int aqr107_config_init(struct phy_device *phydev)
 	if (!ret)
 		aqr107_chip_info(phydev);
 
-	return aqr107_set_downshift(phydev, MDIO_AN_VEND_PROV_DOWNSHIFT_DFLT);
+	ret = aqr107_set_downshift(phydev, MDIO_AN_VEND_PROV_DOWNSHIFT_DFLT);
+	if (ret)
+		return ret;
+
+	/* Restore LED polarity state after reset */
+	for_each_set_bit(led_active_low, &priv->leds_active_low, AQR_MAX_LEDS) {
+		ret = aqr_phy_led_active_low_set(phydev, index, led_active_low);
+		if (ret)
+			return ret;
+		index++;
+	}
+
+	return 0;
 }
 
 static int aqcs109_config_init(struct phy_device *phydev)
@@ -786,6 +800,11 @@ static struct phy_driver aqr_driver[] = {
 	.get_strings	= aqr107_get_strings,
 	.get_stats	= aqr107_get_stats,
 	.link_change_notify = aqr107_link_change_notify,
+	.led_brightness_set = aqr_phy_led_brightness_set,
+	.led_hw_is_supported = aqr_phy_led_hw_is_supported,
+	.led_hw_control_set = aqr_phy_led_hw_control_set,
+	.led_hw_control_get = aqr_phy_led_hw_control_get,
+	.led_polarity_set = aqr_phy_led_polarity_set,
 },
 {
 	PHY_ID_MATCH_MODEL(PHY_ID_AQCS109),
@@ -805,6 +824,11 @@ static struct phy_driver aqr_driver[] = {
 	.get_strings	= aqr107_get_strings,
 	.get_stats	= aqr107_get_stats,
 	.link_change_notify = aqr107_link_change_notify,
+	.led_brightness_set = aqr_phy_led_brightness_set,
+	.led_hw_is_supported = aqr_phy_led_hw_is_supported,
+	.led_hw_control_set = aqr_phy_led_hw_control_set,
+	.led_hw_control_get = aqr_phy_led_hw_control_get,
+	.led_polarity_set = aqr_phy_led_polarity_set,
 },
 {
 	PHY_ID_MATCH_MODEL(PHY_ID_AQR111),
@@ -824,6 +848,11 @@ static struct phy_driver aqr_driver[] = {
 	.get_strings	= aqr107_get_strings,
 	.get_stats	= aqr107_get_stats,
 	.link_change_notify = aqr107_link_change_notify,
+	.led_brightness_set = aqr_phy_led_brightness_set,
+	.led_hw_is_supported = aqr_phy_led_hw_is_supported,
+	.led_hw_control_set = aqr_phy_led_hw_control_set,
+	.led_hw_control_get = aqr_phy_led_hw_control_get,
+	.led_polarity_set = aqr_phy_led_polarity_set,
 },
 {
 	PHY_ID_MATCH_MODEL(PHY_ID_AQR111B0),
@@ -843,6 +872,11 @@ static struct phy_driver aqr_driver[] = {
 	.get_strings	= aqr107_get_strings,
 	.get_stats	= aqr107_get_stats,
 	.link_change_notify = aqr107_link_change_notify,
+	.led_brightness_set = aqr_phy_led_brightness_set,
+	.led_hw_is_supported = aqr_phy_led_hw_is_supported,
+	.led_hw_control_set = aqr_phy_led_hw_control_set,
+	.led_hw_control_get = aqr_phy_led_hw_control_get,
+	.led_polarity_set = aqr_phy_led_polarity_set,
 },
 {
 	PHY_ID_MATCH_MODEL(PHY_ID_AQR405),
@@ -869,6 +903,11 @@ static struct phy_driver aqr_driver[] = {
 	.get_strings	= aqr107_get_strings,
 	.get_stats	= aqr107_get_stats,
 	.link_change_notify = aqr107_link_change_notify,
+	.led_brightness_set = aqr_phy_led_brightness_set,
+	.led_hw_is_supported = aqr_phy_led_hw_is_supported,
+	.led_hw_control_set = aqr_phy_led_hw_control_set,
+	.led_hw_control_get = aqr_phy_led_hw_control_get,
+	.led_polarity_set = aqr_phy_led_polarity_set,
 },
 {
 	PHY_ID_MATCH_MODEL(PHY_ID_AQR412),
@@ -906,6 +945,11 @@ static struct phy_driver aqr_driver[] = {
 	.get_strings    = aqr107_get_strings,
 	.get_stats      = aqr107_get_stats,
 	.link_change_notify = aqr107_link_change_notify,
+	.led_brightness_set = aqr_phy_led_brightness_set,
+	.led_hw_is_supported = aqr_phy_led_hw_is_supported,
+	.led_hw_control_set = aqr_phy_led_hw_control_set,
+	.led_hw_control_get = aqr_phy_led_hw_control_get,
+	.led_polarity_set = aqr_phy_led_polarity_set,
 },
 {
 	PHY_ID_MATCH_MODEL(PHY_ID_AQR113C),
@@ -925,6 +969,11 @@ static struct phy_driver aqr_driver[] = {
 	.get_strings    = aqr107_get_strings,
 	.get_stats      = aqr107_get_stats,
 	.link_change_notify = aqr107_link_change_notify,
+	.led_brightness_set = aqr_phy_led_brightness_set,
+	.led_hw_is_supported = aqr_phy_led_hw_is_supported,
+	.led_hw_control_set = aqr_phy_led_hw_control_set,
+	.led_hw_control_get = aqr_phy_led_hw_control_get,
+	.led_polarity_set = aqr_phy_led_polarity_set,
 },
 {
 	PHY_ID_MATCH_MODEL(PHY_ID_AQR114C),
@@ -944,6 +993,11 @@ static struct phy_driver aqr_driver[] = {
 	.get_strings    = aqr107_get_strings,
 	.get_stats      = aqr107_get_stats,
 	.link_change_notify = aqr107_link_change_notify,
+	.led_brightness_set = aqr_phy_led_brightness_set,
+	.led_hw_is_supported = aqr_phy_led_hw_is_supported,
+	.led_hw_control_set = aqr_phy_led_hw_control_set,
+	.led_hw_control_get = aqr_phy_led_hw_control_get,
+	.led_polarity_set = aqr_phy_led_polarity_set,
 },
 {
 	PHY_ID_MATCH_MODEL(PHY_ID_AQR813),
@@ -963,6 +1017,11 @@ static struct phy_driver aqr_driver[] = {
 	.get_strings	= aqr107_get_strings,
 	.get_stats	= aqr107_get_stats,
 	.link_change_notify = aqr107_link_change_notify,
+	.led_brightness_set = aqr_phy_led_brightness_set,
+	.led_hw_is_supported = aqr_phy_led_hw_is_supported,
+	.led_hw_control_set = aqr_phy_led_hw_control_set,
+	.led_hw_control_get = aqr_phy_led_hw_control_get,
+	.led_polarity_set = aqr_phy_led_polarity_set,
 },
 };
 
-- 
cgit v1.2.3-58-ga151


From 82dc29b9737edf2d13561ebcf6212c0b88c41129 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sun, 2 Jun 2024 16:18:52 +0200
Subject: devlink: Constify the 'table_ops' parameter of
 devl_dpipe_table_register()

"struct devlink_dpipe_table_ops" only contains some function pointers.

Update "struct devlink_dpipe_table" and the 'table_ops' parameter of
devl_dpipe_table_register() so that structures in drivers can be
constified.

Constifying these structures will move some data to a read-only section, so
increase overall security.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/devlink.h | 4 ++--
 net/devlink/dpipe.c   | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/net/devlink.h b/include/net/devlink.h
index 35eb0f884386..db5eff6cb60f 100644
--- a/include/net/devlink.h
+++ b/include/net/devlink.h
@@ -352,7 +352,7 @@ struct devlink_dpipe_table {
 	bool resource_valid;
 	u64 resource_id;
 	u64 resource_units;
-	struct devlink_dpipe_table_ops *table_ops;
+	const struct devlink_dpipe_table_ops *table_ops;
 	struct rcu_head rcu;
 };
 
@@ -1751,7 +1751,7 @@ void devl_sb_unregister(struct devlink *devlink, unsigned int sb_index);
 void devlink_sb_unregister(struct devlink *devlink, unsigned int sb_index);
 int devl_dpipe_table_register(struct devlink *devlink,
 			      const char *table_name,
-			      struct devlink_dpipe_table_ops *table_ops,
+			      const struct devlink_dpipe_table_ops *table_ops,
 			      void *priv, bool counter_control_extern);
 void devl_dpipe_table_unregister(struct devlink *devlink,
 				 const char *table_name);
diff --git a/net/devlink/dpipe.c b/net/devlink/dpipe.c
index a72a9292efc5..55009b377447 100644
--- a/net/devlink/dpipe.c
+++ b/net/devlink/dpipe.c
@@ -839,7 +839,7 @@ EXPORT_SYMBOL_GPL(devlink_dpipe_table_counter_enabled);
  */
 int devl_dpipe_table_register(struct devlink *devlink,
 			      const char *table_name,
-			      struct devlink_dpipe_table_ops *table_ops,
+			      const struct devlink_dpipe_table_ops *table_ops,
 			      void *priv, bool counter_control_extern)
 {
 	struct devlink_dpipe_table *table;
-- 
cgit v1.2.3-58-ga151


From b072aa7899189127e26b2daf8c38933c994f97d3 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sun, 2 Jun 2024 16:18:53 +0200
Subject: mlxsw: spectrum_router: Constify struct devlink_dpipe_table_ops

'struct devlink_dpipe_table_ops' are not modified in this driver.

Constifying these structures moves some data to a read-only section, so
increase overall security.

On a x86_64, with allmodconfig:
Before:
======
   text	   data	    bss	    dec	    hex	filename
  15557	    712	      0	  16269	   3f8d	drivers/net/ethernet/mellanox/mlxsw/spectrum_dpipe.o

After:
=====
   text	   data	    bss	    dec	    hex	filename
  15789	    488	      0	  16277	   3f95	drivers/net/ethernet/mellanox/mlxsw/spectrum_dpipe.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_dpipe.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dpipe.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dpipe.c
index ca80af06465f..fa6eddd27ecf 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_dpipe.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_dpipe.c
@@ -283,7 +283,7 @@ static u64 mlxsw_sp_dpipe_table_erif_size_get(void *priv)
 	return MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_RIFS);
 }
 
-static struct devlink_dpipe_table_ops mlxsw_sp_erif_ops = {
+static const struct devlink_dpipe_table_ops mlxsw_sp_erif_ops = {
 	.matches_dump = mlxsw_sp_dpipe_table_erif_matches_dump,
 	.actions_dump = mlxsw_sp_dpipe_table_erif_actions_dump,
 	.entries_dump = mlxsw_sp_dpipe_table_erif_entries_dump,
@@ -734,7 +734,7 @@ static u64 mlxsw_sp_dpipe_table_host4_size_get(void *priv)
 	return mlxsw_sp_dpipe_table_host_size_get(mlxsw_sp, AF_INET);
 }
 
-static struct devlink_dpipe_table_ops mlxsw_sp_host4_ops = {
+static const struct devlink_dpipe_table_ops mlxsw_sp_host4_ops = {
 	.matches_dump = mlxsw_sp_dpipe_table_host4_matches_dump,
 	.actions_dump = mlxsw_sp_dpipe_table_host_actions_dump,
 	.entries_dump = mlxsw_sp_dpipe_table_host4_entries_dump,
@@ -811,7 +811,7 @@ static u64 mlxsw_sp_dpipe_table_host6_size_get(void *priv)
 	return mlxsw_sp_dpipe_table_host_size_get(mlxsw_sp, AF_INET6);
 }
 
-static struct devlink_dpipe_table_ops mlxsw_sp_host6_ops = {
+static const struct devlink_dpipe_table_ops mlxsw_sp_host6_ops = {
 	.matches_dump = mlxsw_sp_dpipe_table_host6_matches_dump,
 	.actions_dump = mlxsw_sp_dpipe_table_host_actions_dump,
 	.entries_dump = mlxsw_sp_dpipe_table_host6_entries_dump,
@@ -1230,7 +1230,7 @@ mlxsw_sp_dpipe_table_adj_size_get(void *priv)
 	return size;
 }
 
-static struct devlink_dpipe_table_ops mlxsw_sp_dpipe_table_adj_ops = {
+static const struct devlink_dpipe_table_ops mlxsw_sp_dpipe_table_adj_ops = {
 	.matches_dump = mlxsw_sp_dpipe_table_adj_matches_dump,
 	.actions_dump = mlxsw_sp_dpipe_table_adj_actions_dump,
 	.entries_dump = mlxsw_sp_dpipe_table_adj_entries_dump,
-- 
cgit v1.2.3-58-ga151


From b907194a5d5bcd3b01d28d095f6a4d9fcd9c5354 Mon Sep 17 00:00:00 2001
From: Anshumali Gaur <agaur@marvell.com>
Date: Mon, 3 Jun 2024 16:52:48 +0530
Subject: octeontx2-af: Add debugfs support to dump NIX TM topology

This patch adds support to dump NIX transmit queue topology.
There are multiple levels of scheduling/shaping supported by
NIX and a packet traverses through multiple levels before sending
the packet out. At each level, there are set of scheduling/shaping
rules applied to a packet flow.

Each packet traverses through multiple levels
SQ->SMQ->TL4->TL3->TL2->TL1 and these levels are mapped in a parent-child
relationship.

This patch dumps the debug information related to all TM Levels in
the following way.

Example:
$ echo <nixlf> > /sys/kernel/debug/octeontx2/nix/tm_tree
$ cat /sys/kernel/debug/octeontx2/nix/tm_tree

A more desriptive set of registers at each level can be dumped
in the following way.

Example:
$ echo <nixlf> > /sys/kernel/debug/octeontx2/nix/tm_topo
$ cat /sys/kernel/debug/octeontx2/nix/tm_topo

Signed-off-by: Anshumali Gaur <agaur@marvell.com>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/marvell/octeontx2/af/rvu.h    |   1 +
 .../ethernet/marvell/octeontx2/af/rvu_debugfs.c    | 365 +++++++++++++++++++++
 .../net/ethernet/marvell/octeontx2/af/rvu_reg.h    |   7 +
 3 files changed, 373 insertions(+)

diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
index 35834687e40f..3063a84a45ef 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
@@ -76,6 +76,7 @@ struct rvu_debugfs {
 	struct dump_ctx nix_cq_ctx;
 	struct dump_ctx nix_rq_ctx;
 	struct dump_ctx nix_sq_ctx;
+	struct dump_ctx nix_tm_ctx;
 	struct cpt_ctx cpt_ctx[MAX_CPT_BLKS];
 	int npa_qsize_id;
 	int nix_qsize_id;
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c
index 881d704644fb..4a4ef5bd9e0b 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c
@@ -1603,6 +1603,367 @@ static void print_nix_cn10k_sq_ctx(struct seq_file *m,
 		   (u64)sq_ctx->dropped_pkts);
 }
 
+static void print_tm_tree(struct seq_file *m,
+			  struct nix_aq_enq_rsp *rsp, u64 sq)
+{
+	struct nix_sq_ctx_s *sq_ctx = &rsp->sq;
+	struct nix_hw *nix_hw = m->private;
+	struct rvu *rvu = nix_hw->rvu;
+	u16 p1, p2, p3, p4, schq;
+	int blkaddr;
+	u64 cfg;
+
+	blkaddr = nix_hw->blkaddr;
+	schq = sq_ctx->smq;
+
+	cfg = rvu_read64(rvu, blkaddr, NIX_AF_MDQX_PARENT(schq));
+	p1 = FIELD_GET(NIX_AF_MDQ_PARENT_MASK, cfg);
+
+	cfg = rvu_read64(rvu, blkaddr, NIX_AF_TL4X_PARENT(p1));
+	p2 = FIELD_GET(NIX_AF_TL4_PARENT_MASK, cfg);
+
+	cfg = rvu_read64(rvu, blkaddr, NIX_AF_TL3X_PARENT(p2));
+	p3 = FIELD_GET(NIX_AF_TL3_PARENT_MASK, cfg);
+
+	cfg = rvu_read64(rvu, blkaddr, NIX_AF_TL2X_PARENT(p3));
+	p4 = FIELD_GET(NIX_AF_TL2_PARENT_MASK, cfg);
+	seq_printf(m,
+		   "SQ(%llu) -> SMQ(%u) -> TL4(%u) -> TL3(%u) -> TL2(%u) -> TL1(%u)\n",
+		   sq, schq, p1, p2, p3, p4);
+}
+
+/*dumps given tm_tree registers*/
+static int rvu_dbg_nix_tm_tree_display(struct seq_file *m, void *unused)
+{
+	int qidx, nixlf, rc, id, max_id = 0;
+	struct nix_hw *nix_hw = m->private;
+	struct rvu *rvu = nix_hw->rvu;
+	struct nix_aq_enq_req aq_req;
+	struct nix_aq_enq_rsp rsp;
+	struct rvu_pfvf *pfvf;
+	u16 pcifunc;
+
+	nixlf = rvu->rvu_dbg.nix_tm_ctx.lf;
+	id = rvu->rvu_dbg.nix_tm_ctx.id;
+
+	if (!rvu_dbg_is_valid_lf(rvu, nix_hw->blkaddr, nixlf, &pcifunc))
+		return -EINVAL;
+
+	pfvf = rvu_get_pfvf(rvu, pcifunc);
+	max_id = pfvf->sq_ctx->qsize;
+
+	memset(&aq_req, 0, sizeof(struct nix_aq_enq_req));
+	aq_req.hdr.pcifunc = pcifunc;
+	aq_req.ctype = NIX_AQ_CTYPE_SQ;
+	aq_req.op = NIX_AQ_INSTOP_READ;
+	seq_printf(m, "pcifunc is 0x%x\n", pcifunc);
+	for (qidx = id; qidx < max_id; qidx++) {
+		aq_req.qidx = qidx;
+
+		/* Skip SQ's if not initialized */
+		if (!test_bit(qidx, pfvf->sq_bmap))
+			continue;
+
+		rc = rvu_mbox_handler_nix_aq_enq(rvu, &aq_req, &rsp);
+
+		if (rc) {
+			seq_printf(m, "Failed to read SQ(%d) context\n",
+				   aq_req.qidx);
+			continue;
+		}
+		print_tm_tree(m, &rsp, aq_req.qidx);
+	}
+	return 0;
+}
+
+static ssize_t rvu_dbg_nix_tm_tree_write(struct file *filp,
+					 const char __user *buffer,
+					 size_t count, loff_t *ppos)
+{
+	struct seq_file *m = filp->private_data;
+	struct nix_hw *nix_hw = m->private;
+	struct rvu *rvu = nix_hw->rvu;
+	struct rvu_pfvf *pfvf;
+	u16 pcifunc;
+	u64 nixlf;
+	int ret;
+
+	ret = kstrtoull_from_user(buffer, count, 10, &nixlf);
+	if (ret)
+		return ret;
+
+	if (!rvu_dbg_is_valid_lf(rvu, nix_hw->blkaddr, nixlf, &pcifunc))
+		return -EINVAL;
+
+	pfvf = rvu_get_pfvf(rvu, pcifunc);
+	if (!pfvf->sq_ctx) {
+		dev_warn(rvu->dev, "SQ context is not initialized\n");
+		return -EINVAL;
+	}
+
+	rvu->rvu_dbg.nix_tm_ctx.lf = nixlf;
+	return count;
+}
+
+RVU_DEBUG_SEQ_FOPS(nix_tm_tree, nix_tm_tree_display, nix_tm_tree_write);
+
+static void print_tm_topo(struct seq_file *m, u64 schq, u32 lvl)
+{
+	struct nix_hw *nix_hw = m->private;
+	struct rvu *rvu = nix_hw->rvu;
+	int blkaddr, link, link_level;
+	struct rvu_hwinfo *hw;
+
+	hw = rvu->hw;
+	blkaddr = nix_hw->blkaddr;
+	if (lvl == NIX_TXSCH_LVL_MDQ) {
+		seq_printf(m, "NIX_AF_SMQ[%llu]_CFG =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_SMQX_CFG(schq)));
+		seq_printf(m, "NIX_AF_SMQ[%llu]_STATUS =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_SMQX_STATUS(schq)));
+		seq_printf(m, "NIX_AF_MDQ[%llu]_OUT_MD_COUNT =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_MDQX_OUT_MD_COUNT(schq)));
+		seq_printf(m, "NIX_AF_MDQ[%llu]_SCHEDULE =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_MDQX_SCHEDULE(schq)));
+		seq_printf(m, "NIX_AF_MDQ[%llu]_SHAPE =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_MDQX_SHAPE(schq)));
+		seq_printf(m, "NIX_AF_MDQ[%llu]_CIR =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_MDQX_CIR(schq)));
+		seq_printf(m, "NIX_AF_MDQ[%llu]_PIR =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_MDQX_PIR(schq)));
+		seq_printf(m, "NIX_AF_MDQ[%llu]_SW_XOFF =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_MDQX_SW_XOFF(schq)));
+		seq_printf(m, "NIX_AF_MDQ[%llu]_PARENT =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_MDQX_PARENT(schq)));
+		seq_puts(m, "\n");
+	}
+
+	if (lvl == NIX_TXSCH_LVL_TL4) {
+		seq_printf(m, "NIX_AF_TL4[%llu]_SDP_LINK_CFG =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL4X_SDP_LINK_CFG(schq)));
+		seq_printf(m, "NIX_AF_TL4[%llu]_SCHEDULE =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL4X_SCHEDULE(schq)));
+		seq_printf(m, "NIX_AF_TL4[%llu]_SHAPE =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL4X_SHAPE(schq)));
+		seq_printf(m, "NIX_AF_TL4[%llu]_CIR =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL4X_CIR(schq)));
+		seq_printf(m, "NIX_AF_TL4[%llu]_PIR =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL4X_PIR(schq)));
+		seq_printf(m, "NIX_AF_TL4[%llu]_SW_XOFF =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL4X_SW_XOFF(schq)));
+		seq_printf(m, "NIX_AF_TL4[%llu]_TOPOLOGY =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL4X_TOPOLOGY(schq)));
+		seq_printf(m, "NIX_AF_TL4[%llu]_PARENT =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL4X_PARENT(schq)));
+		seq_printf(m, "NIX_AF_TL4[%llu]_MD_DEBUG0 =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL4X_MD_DEBUG0(schq)));
+		seq_printf(m, "NIX_AF_TL4[%llu]_MD_DEBUG1 =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL4X_MD_DEBUG1(schq)));
+		seq_puts(m, "\n");
+	}
+
+	if (lvl == NIX_TXSCH_LVL_TL3) {
+		seq_printf(m, "NIX_AF_TL3[%llu]_SCHEDULE =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL3X_SCHEDULE(schq)));
+		seq_printf(m, "NIX_AF_TL3[%llu]_SHAPE =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL3X_SHAPE(schq)));
+		seq_printf(m, "NIX_AF_TL3[%llu]_CIR =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL3X_CIR(schq)));
+		seq_printf(m, "NIX_AF_TL3[%llu]_PIR =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL3X_PIR(schq)));
+		seq_printf(m, "NIX_AF_TL3[%llu]_SW_XOFF =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL3X_SW_XOFF(schq)));
+		seq_printf(m, "NIX_AF_TL3[%llu]_TOPOLOGY =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL3X_TOPOLOGY(schq)));
+		seq_printf(m, "NIX_AF_TL3[%llu]_PARENT =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL3X_PARENT(schq)));
+		seq_printf(m, "NIX_AF_TL3[%llu]_MD_DEBUG0 =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL3X_MD_DEBUG0(schq)));
+		seq_printf(m, "NIX_AF_TL3[%llu]_MD_DEBUG1 =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL3X_MD_DEBUG1(schq)));
+
+		link_level = rvu_read64(rvu, blkaddr, NIX_AF_PSE_CHANNEL_LEVEL)
+				& 0x01 ? NIX_TXSCH_LVL_TL3 : NIX_TXSCH_LVL_TL2;
+		if (lvl == link_level) {
+			seq_printf(m,
+				   "NIX_AF_TL3_TL2[%llu]_BP_STATUS =0x%llx\n",
+				   schq, rvu_read64(rvu, blkaddr,
+				   NIX_AF_TL3_TL2X_BP_STATUS(schq)));
+			for (link = 0; link < hw->cgx_links; link++)
+				seq_printf(m,
+					   "NIX_AF_TL3_TL2[%llu]_LINK[%d]_CFG =0x%llx\n",
+					   schq, link,
+					   rvu_read64(rvu, blkaddr,
+						      NIX_AF_TL3_TL2X_LINKX_CFG(schq, link)));
+		}
+		seq_puts(m, "\n");
+	}
+
+	if (lvl == NIX_TXSCH_LVL_TL2) {
+		seq_printf(m, "NIX_AF_TL2[%llu]_SHAPE =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL2X_SHAPE(schq)));
+		seq_printf(m, "NIX_AF_TL2[%llu]_CIR =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL2X_CIR(schq)));
+		seq_printf(m, "NIX_AF_TL2[%llu]_PIR =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL2X_PIR(schq)));
+		seq_printf(m, "NIX_AF_TL2[%llu]_SW_XOFF =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL2X_SW_XOFF(schq)));
+		seq_printf(m, "NIX_AF_TL2[%llu]_TOPOLOGY =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL2X_TOPOLOGY(schq)));
+		seq_printf(m, "NIX_AF_TL2[%llu]_PARENT =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL2X_PARENT(schq)));
+		seq_printf(m, "NIX_AF_TL2[%llu]_MD_DEBUG0 =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL2X_MD_DEBUG0(schq)));
+		seq_printf(m, "NIX_AF_TL2[%llu]_MD_DEBUG1 =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL2X_MD_DEBUG1(schq)));
+
+		link_level = rvu_read64(rvu, blkaddr, NIX_AF_PSE_CHANNEL_LEVEL)
+				& 0x01 ? NIX_TXSCH_LVL_TL3 : NIX_TXSCH_LVL_TL2;
+		if (lvl == link_level) {
+			seq_printf(m,
+				   "NIX_AF_TL3_TL2[%llu]_BP_STATUS =0x%llx\n",
+				   schq, rvu_read64(rvu, blkaddr,
+				   NIX_AF_TL3_TL2X_BP_STATUS(schq)));
+			for (link = 0; link < hw->cgx_links; link++)
+				seq_printf(m,
+					   "NIX_AF_TL3_TL2[%llu]_LINK[%d]_CFG =0x%llx\n",
+					   schq, link, rvu_read64(rvu, blkaddr,
+					   NIX_AF_TL3_TL2X_LINKX_CFG(schq, link)));
+		}
+		seq_puts(m, "\n");
+	}
+
+	if (lvl == NIX_TXSCH_LVL_TL1) {
+		seq_printf(m, "NIX_AF_TX_LINK[%llu]_NORM_CREDIT =0x%llx\n",
+			   schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TX_LINKX_NORM_CREDIT(schq)));
+		seq_printf(m, "NIX_AF_TX_LINK[%llu]_HW_XOFF =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TX_LINKX_HW_XOFF(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_SCHEDULE =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_SCHEDULE(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_SHAPE =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL1X_SHAPE(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_CIR =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL1X_CIR(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_SW_XOFF =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr, NIX_AF_TL1X_SW_XOFF(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_TOPOLOGY =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_TOPOLOGY(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_MD_DEBUG0 =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_MD_DEBUG0(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_MD_DEBUG1 =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_MD_DEBUG1(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_DROPPED_PACKETS =0x%llx\n",
+			   schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_DROPPED_PACKETS(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_DROPPED_BYTES =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_DROPPED_BYTES(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_RED_PACKETS =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_RED_PACKETS(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_RED_BYTES =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_RED_BYTES(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_YELLOW_PACKETS =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_YELLOW_PACKETS(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_YELLOW_BYTES =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_YELLOW_BYTES(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_GREEN_PACKETS =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_GREEN_PACKETS(schq)));
+		seq_printf(m, "NIX_AF_TL1[%llu]_GREEN_BYTES =0x%llx\n", schq,
+			   rvu_read64(rvu, blkaddr,
+				      NIX_AF_TL1X_GREEN_BYTES(schq)));
+		seq_puts(m, "\n");
+	}
+}
+
+/*dumps given tm_topo registers*/
+static int rvu_dbg_nix_tm_topo_display(struct seq_file *m, void *unused)
+{
+	struct nix_hw *nix_hw = m->private;
+	struct rvu *rvu = nix_hw->rvu;
+	struct nix_aq_enq_req aq_req;
+	struct nix_txsch *txsch;
+	int nixlf, lvl, schq;
+	u16 pcifunc;
+
+	nixlf = rvu->rvu_dbg.nix_tm_ctx.lf;
+
+	if (!rvu_dbg_is_valid_lf(rvu, nix_hw->blkaddr, nixlf, &pcifunc))
+		return -EINVAL;
+
+	memset(&aq_req, 0, sizeof(struct nix_aq_enq_req));
+	aq_req.hdr.pcifunc = pcifunc;
+	aq_req.ctype = NIX_AQ_CTYPE_SQ;
+	aq_req.op = NIX_AQ_INSTOP_READ;
+	seq_printf(m, "pcifunc is 0x%x\n", pcifunc);
+
+	for (lvl = 0; lvl < NIX_TXSCH_LVL_CNT; lvl++) {
+		txsch = &nix_hw->txsch[lvl];
+		for (schq = 0; schq < txsch->schq.max; schq++) {
+			if (TXSCH_MAP_FUNC(txsch->pfvf_map[schq]) == pcifunc)
+				print_tm_topo(m, schq, lvl);
+		}
+	}
+	return 0;
+}
+
+static ssize_t rvu_dbg_nix_tm_topo_write(struct file *filp,
+					 const char __user *buffer,
+					 size_t count, loff_t *ppos)
+{
+	struct seq_file *m = filp->private_data;
+	struct nix_hw *nix_hw = m->private;
+	struct rvu *rvu = nix_hw->rvu;
+	struct rvu_pfvf *pfvf;
+	u16 pcifunc;
+	u64 nixlf;
+	int ret;
+
+	ret = kstrtoull_from_user(buffer, count, 10, &nixlf);
+	if (ret)
+		return ret;
+
+	if (!rvu_dbg_is_valid_lf(rvu, nix_hw->blkaddr, nixlf, &pcifunc))
+		return -EINVAL;
+
+	pfvf = rvu_get_pfvf(rvu, pcifunc);
+	if (!pfvf->sq_ctx) {
+		dev_warn(rvu->dev, "SQ context is not initialized\n");
+		return -EINVAL;
+	}
+
+	rvu->rvu_dbg.nix_tm_ctx.lf = nixlf;
+	return count;
+}
+
+RVU_DEBUG_SEQ_FOPS(nix_tm_topo, nix_tm_topo_display, nix_tm_topo_write);
+
 /* Dumps given nix_sq's context */
 static void print_nix_sq_ctx(struct seq_file *m, struct nix_aq_enq_rsp *rsp)
 {
@@ -2349,6 +2710,10 @@ static void rvu_dbg_nix_init(struct rvu *rvu, int blkaddr)
 		nix_hw = &rvu->hw->nix[1];
 	}
 
+	debugfs_create_file("tm_tree", 0600, rvu->rvu_dbg.nix, nix_hw,
+			    &rvu_dbg_nix_tm_tree_fops);
+	debugfs_create_file("tm_topo", 0600, rvu->rvu_dbg.nix, nix_hw,
+			    &rvu_dbg_nix_tm_topo_fops);
 	debugfs_create_file("sq_ctx", 0600, rvu->rvu_dbg.nix, nix_hw,
 			    &rvu_dbg_nix_sq_ctx_fops);
 	debugfs_create_file("rq_ctx", 0600, rvu->rvu_dbg.nix, nix_hw,
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h
index 086f05c0376f..5ec92654e7ad 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h
@@ -429,6 +429,8 @@
 #define NIX_AF_RX_ACTIVE_CYCLES_PCX(a)	(0x4800 | (a) << 16)
 #define NIX_AF_LINKX_CFG(a)		(0x4010 | (a) << 17)
 #define NIX_AF_MDQX_IN_MD_COUNT(a)	(0x14e0 | (a) << 16)
+#define NIX_AF_SMQX_STATUS(a)		(0x730 | (a) << 16)
+#define NIX_AF_MDQX_OUT_MD_COUNT(a)	(0xdb0 | (a) << 16)
 
 #define NIX_PRIV_AF_INT_CFG		(0x8000000)
 #define NIX_PRIV_LFX_CFG		(0x8000010)
@@ -442,6 +444,11 @@
 #define NIX_CONST_MAX_BPIDS		GENMASK_ULL(23, 12)
 #define NIX_CONST_SDP_CHANS		GENMASK_ULL(11, 0)
 
+#define NIX_AF_MDQ_PARENT_MASK         GENMASK_ULL(24, 16)
+#define NIX_AF_TL4_PARENT_MASK         GENMASK_ULL(23, 16)
+#define NIX_AF_TL3_PARENT_MASK         GENMASK_ULL(23, 16)
+#define NIX_AF_TL2_PARENT_MASK         GENMASK_ULL(20, 16)
+
 /* SSO */
 #define SSO_AF_CONST			(0x1000)
 #define SSO_AF_CONST1			(0x1008)
-- 
cgit v1.2.3-58-ga151


From 69e0b33a7fce4d96649b9fa32e56b696921aa48e Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Mon, 3 Jun 2024 15:51:06 +0000
Subject: tcp: annotate data-races around tw->tw_ts_recent and
 tw->tw_ts_recent_stamp

These fields can be read and written locklessly, add annotations
around these minor races.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_ipv4.c      | 12 +++++++-----
 net/ipv4/tcp_minisocks.c | 22 ++++++++++++++--------
 net/ipv6/tcp_ipv6.c      |  6 +++---
 3 files changed, 24 insertions(+), 16 deletions(-)

diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 04044605cadf..3ef4b274c24b 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -114,6 +114,7 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp)
 	const struct inet_timewait_sock *tw = inet_twsk(sktw);
 	const struct tcp_timewait_sock *tcptw = tcp_twsk(sktw);
 	struct tcp_sock *tp = tcp_sk(sk);
+	int ts_recent_stamp;
 
 	if (reuse == 2) {
 		/* Still does not detect *everything* that goes through
@@ -152,9 +153,10 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp)
 	   If TW bucket has been already destroyed we fall back to VJ's scheme
 	   and use initial timestamp retrieved from peer table.
 	 */
-	if (tcptw->tw_ts_recent_stamp &&
+	ts_recent_stamp = READ_ONCE(tcptw->tw_ts_recent_stamp);
+	if (ts_recent_stamp &&
 	    (!twp || (reuse && time_after32(ktime_get_seconds(),
-					    tcptw->tw_ts_recent_stamp)))) {
+					    ts_recent_stamp)))) {
 		/* inet_twsk_hashdance() sets sk_refcnt after putting twsk
 		 * and releasing the bucket lock.
 		 */
@@ -178,8 +180,8 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp)
 			if (!seq)
 				seq = 1;
 			WRITE_ONCE(tp->write_seq, seq);
-			tp->rx_opt.ts_recent	   = tcptw->tw_ts_recent;
-			tp->rx_opt.ts_recent_stamp = tcptw->tw_ts_recent_stamp;
+			tp->rx_opt.ts_recent	   = READ_ONCE(tcptw->tw_ts_recent);
+			tp->rx_opt.ts_recent_stamp = ts_recent_stamp;
 		}
 
 		return 1;
@@ -1064,7 +1066,7 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb)
 			tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
 			tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale,
 			tcp_tw_tsval(tcptw),
-			tcptw->tw_ts_recent,
+			READ_ONCE(tcptw->tw_ts_recent),
 			tw->tw_bound_dev_if, &key,
 			tw->tw_transparent ? IP_REPLY_ARG_NOSRCCHECK : 0,
 			tw->tw_tos,
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 538c06f95918..4c894e540730 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -101,16 +101,18 @@ tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb,
 	struct tcp_options_received tmp_opt;
 	struct tcp_timewait_sock *tcptw = tcp_twsk((struct sock *)tw);
 	bool paws_reject = false;
+	int ts_recent_stamp;
 
 	tmp_opt.saw_tstamp = 0;
-	if (th->doff > (sizeof(*th) >> 2) && tcptw->tw_ts_recent_stamp) {
+	ts_recent_stamp = READ_ONCE(tcptw->tw_ts_recent_stamp);
+	if (th->doff > (sizeof(*th) >> 2) && ts_recent_stamp) {
 		tcp_parse_options(twsk_net(tw), skb, &tmp_opt, 0, NULL);
 
 		if (tmp_opt.saw_tstamp) {
 			if (tmp_opt.rcv_tsecr)
 				tmp_opt.rcv_tsecr -= tcptw->tw_ts_offset;
-			tmp_opt.ts_recent	= tcptw->tw_ts_recent;
-			tmp_opt.ts_recent_stamp	= tcptw->tw_ts_recent_stamp;
+			tmp_opt.ts_recent	= READ_ONCE(tcptw->tw_ts_recent);
+			tmp_opt.ts_recent_stamp	= ts_recent_stamp;
 			paws_reject = tcp_paws_reject(&tmp_opt, th->rst);
 		}
 	}
@@ -152,8 +154,10 @@ tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb,
 		twsk_rcv_nxt_update(tcptw, TCP_SKB_CB(skb)->end_seq);
 
 		if (tmp_opt.saw_tstamp) {
-			tcptw->tw_ts_recent_stamp = ktime_get_seconds();
-			tcptw->tw_ts_recent	  = tmp_opt.rcv_tsval;
+			WRITE_ONCE(tcptw->tw_ts_recent_stamp,
+				  ktime_get_seconds());
+			WRITE_ONCE(tcptw->tw_ts_recent,
+				   tmp_opt.rcv_tsval);
 		}
 
 		inet_twsk_reschedule(tw, TCP_TIMEWAIT_LEN);
@@ -197,8 +201,10 @@ kill:
 		}
 
 		if (tmp_opt.saw_tstamp) {
-			tcptw->tw_ts_recent	  = tmp_opt.rcv_tsval;
-			tcptw->tw_ts_recent_stamp = ktime_get_seconds();
+			WRITE_ONCE(tcptw->tw_ts_recent,
+				   tmp_opt.rcv_tsval);
+			WRITE_ONCE(tcptw->tw_ts_recent_stamp,
+				   ktime_get_seconds());
 		}
 
 		inet_twsk_put(tw);
@@ -225,7 +231,7 @@ kill:
 	if (th->syn && !th->rst && !th->ack && !paws_reject &&
 	    (after(TCP_SKB_CB(skb)->seq, tcptw->tw_rcv_nxt) ||
 	     (tmp_opt.saw_tstamp &&
-	      (s32)(tcptw->tw_ts_recent - tmp_opt.rcv_tsval) < 0))) {
+	      (s32)(READ_ONCE(tcptw->tw_ts_recent) - tmp_opt.rcv_tsval) < 0))) {
 		u32 isn = tcptw->tw_snd_nxt + 65535 + 2;
 		if (isn == 0)
 			isn++;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 5fe4365de89d..f4777c03cdd7 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1196,9 +1196,9 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb)
 	tcp_v6_send_ack(sk, skb, tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
 			tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale,
 			tcp_tw_tsval(tcptw),
-			tcptw->tw_ts_recent, tw->tw_bound_dev_if, &key,
-			tw->tw_tclass, cpu_to_be32(tw->tw_flowlabel), tw->tw_priority,
-			tw->tw_txhash);
+			READ_ONCE(tcptw->tw_ts_recent), tw->tw_bound_dev_if,
+			&key, tw->tw_tclass, cpu_to_be32(tw->tw_flowlabel),
+			tw->tw_priority, tw->tw_txhash);
 
 #ifdef CONFIG_TCP_AO
 out:
-- 
cgit v1.2.3-58-ga151


From 512bd0f9f926a05c724a9fd72bc4e14213845e01 Mon Sep 17 00:00:00 2001
From: Kevin Yang <yyd@google.com>
Date: Mon, 3 Jun 2024 21:30:53 +0000
Subject: tcp: derive delack_max with tcp_rto_min helper

Rto_min now has multiple sources, ordered by preprecedence high to
low: ip route option rto_min, icsk->icsk_rto_min.

When derive delack_max from rto_min, we should not only use ip
route option, but should use tcp_rto_min helper to get the correct
rto_min.

Signed-off-by: Kevin Yang <yyd@google.com>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_output.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index f97e098f18a5..090fb0c24599 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -4163,16 +4163,9 @@ EXPORT_SYMBOL(tcp_connect);
 
 u32 tcp_delack_max(const struct sock *sk)
 {
-	const struct dst_entry *dst = __sk_dst_get(sk);
-	u32 delack_max = inet_csk(sk)->icsk_delack_max;
-
-	if (dst && dst_metric_locked(dst, RTAX_RTO_MIN)) {
-		u32 rto_min = dst_metric_rtt(dst, RTAX_RTO_MIN);
-		u32 delack_from_rto_min = max_t(int, 1, rto_min - 1);
+	u32 delack_from_rto_min = max(tcp_rto_min(sk), 2) - 1;
 
-		delack_max = min_t(u32, delack_max, delack_from_rto_min);
-	}
-	return delack_max;
+	return min(inet_csk(sk)->icsk_delack_max, delack_from_rto_min);
 }
 
 /* Send out a delayed ack, the caller does the policy checking
-- 
cgit v1.2.3-58-ga151


From f086edef71be7174a16c1ed67ac65a085cda28b1 Mon Sep 17 00:00:00 2001
From: Kevin Yang <yyd@google.com>
Date: Mon, 3 Jun 2024 21:30:54 +0000
Subject: tcp: add sysctl_tcp_rto_min_us

Adding a sysctl knob to allow user to specify a default
rto_min at socket init time, other than using the hard
coded 200ms default rto_min.

Note that the rto_min route option has the highest precedence
for configuring this setting, followed by the TCP_BPF_RTO_MIN
socket option, followed by the tcp_rto_min_us sysctl.

Signed-off-by: Kevin Yang <yyd@google.com>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/ip-sysctl.rst | 13 +++++++++++++
 include/net/netns/ipv4.h               |  1 +
 net/ipv4/sysctl_net_ipv4.c             |  8 ++++++++
 net/ipv4/tcp.c                         |  4 +++-
 net/ipv4/tcp_ipv4.c                    |  1 +
 5 files changed, 26 insertions(+), 1 deletion(-)

diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst
index bd50df6a5a42..6e99eccdb837 100644
--- a/Documentation/networking/ip-sysctl.rst
+++ b/Documentation/networking/ip-sysctl.rst
@@ -1196,6 +1196,19 @@ tcp_pingpong_thresh - INTEGER
 
 	Default: 1
 
+tcp_rto_min_us - INTEGER
+	Minimal TCP retransmission timeout (in microseconds). Note that the
+	rto_min route option has the highest precedence for configuring this
+	setting, followed by the TCP_BPF_RTO_MIN socket option, followed by
+	this tcp_rto_min_us sysctl.
+
+	The recommended practice is to use a value less or equal to 200000
+	microseconds.
+
+	Possible Values: 1 - INT_MAX
+
+	Default: 200000
+
 UDP variables
 =============
 
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index c356c458b340..a91bb971f901 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -170,6 +170,7 @@ struct netns_ipv4 {
 	u8 sysctl_tcp_sack;
 	u8 sysctl_tcp_window_scaling;
 	u8 sysctl_tcp_timestamps;
+	int sysctl_tcp_rto_min_us;
 	u8 sysctl_tcp_recovery;
 	u8 sysctl_tcp_thin_linear_timeouts;
 	u8 sysctl_tcp_slow_start_after_idle;
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index d7892f34a15b..bb64c0ef092d 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -1503,6 +1503,14 @@ static struct ctl_table ipv4_net_table[] = {
 		.proc_handler	= proc_dou8vec_minmax,
 		.extra1		= SYSCTL_ONE,
 	},
+	{
+		.procname	= "tcp_rto_min_us",
+		.data		= &init_net.ipv4.sysctl_tcp_rto_min_us,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= SYSCTL_ONE,
+	},
 };
 
 static __net_init int ipv4_sysctl_init_net(struct net *net)
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 5fa68e7f6ddb..fa43aaacd92b 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -420,6 +420,7 @@ void tcp_init_sock(struct sock *sk)
 {
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
+	int rto_min_us;
 
 	tp->out_of_order_queue = RB_ROOT;
 	sk->tcp_rtx_queue = RB_ROOT;
@@ -428,7 +429,8 @@ void tcp_init_sock(struct sock *sk)
 	INIT_LIST_HEAD(&tp->tsorted_sent_queue);
 
 	icsk->icsk_rto = TCP_TIMEOUT_INIT;
-	icsk->icsk_rto_min = TCP_RTO_MIN;
+	rto_min_us = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rto_min_us);
+	icsk->icsk_rto_min = usecs_to_jiffies(rto_min_us);
 	icsk->icsk_delack_max = TCP_DELACK_MAX;
 	tp->mdev_us = jiffies_to_usecs(TCP_TIMEOUT_INIT);
 	minmax_reset(&tp->rtt_min, tcp_jiffies32, ~0U);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 3ef4b274c24b..3613e08ca794 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -3502,6 +3502,7 @@ static int __net_init tcp_sk_init(struct net *net)
 	net->ipv4.sysctl_tcp_shrink_window = 0;
 
 	net->ipv4.sysctl_tcp_pingpong_thresh = 1;
+	net->ipv4.sysctl_tcp_rto_min_us = jiffies_to_usecs(TCP_RTO_MIN);
 
 	return 0;
 }
-- 
cgit v1.2.3-58-ga151


From 68153bb2fffbe59804370e514482f95c4b2053ff Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Tue, 4 Jun 2024 17:16:25 -0700
Subject: libbpf: Add BTF field iterator

Implement iterator-based type ID and string offset BTF field iterator.
This is used extensively in BTF-handling code and BPF linker code for
various sanity checks, rewriting IDs/offsets, etc. Currently this is
implemented as visitor pattern calling custom callbacks, which makes the
logic (especially in simple cases) unnecessarily obscure and harder to
follow.

Having equivalent functionality using iterator pattern makes for simpler
to understand and maintain code. As we add more code for BTF processing
logic in libbpf, it's best to switch to iterator pattern before adding
more callback-based code.

The idea for iterator-based implementation is to record offsets of
necessary fields within fixed btf_type parts (which should be iterated
just once), and, for kinds that have multiple members (based on vlen
field), record where in each member necessary fields are located.

Generic iteration code then just keeps track of last offset that was
returned and handles N members correctly. Return type is just u32
pointer, where NULL is returned when all relevant fields were already
iterated.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240605001629.4061937-2-andrii@kernel.org
---
 tools/lib/bpf/btf.c             | 162 ++++++++++++++++++++++++++++++++++++++++
 tools/lib/bpf/libbpf_internal.h |  24 ++++++
 2 files changed, 186 insertions(+)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 2d0840ef599a..50ff8b6eaf36 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -5133,6 +5133,168 @@ int btf_type_visit_str_offs(struct btf_type *t, str_off_visit_fn visit, void *ct
 	return 0;
 }
 
+int btf_field_iter_init(struct btf_field_iter *it, struct btf_type *t, enum btf_field_iter_kind iter_kind)
+{
+	it->p = NULL;
+	it->m_idx = -1;
+	it->off_idx = 0;
+	it->vlen = 0;
+
+	switch (iter_kind) {
+	case BTF_FIELD_ITER_IDS:
+		switch (btf_kind(t)) {
+		case BTF_KIND_UNKN:
+		case BTF_KIND_INT:
+		case BTF_KIND_FLOAT:
+		case BTF_KIND_ENUM:
+		case BTF_KIND_ENUM64:
+			it->desc = (struct btf_field_desc) {};
+			break;
+		case BTF_KIND_FWD:
+		case BTF_KIND_CONST:
+		case BTF_KIND_VOLATILE:
+		case BTF_KIND_RESTRICT:
+		case BTF_KIND_PTR:
+		case BTF_KIND_TYPEDEF:
+		case BTF_KIND_FUNC:
+		case BTF_KIND_VAR:
+		case BTF_KIND_DECL_TAG:
+		case BTF_KIND_TYPE_TAG:
+			it->desc = (struct btf_field_desc) { 1, {offsetof(struct btf_type, type)} };
+			break;
+		case BTF_KIND_ARRAY:
+			it->desc = (struct btf_field_desc) {
+				2, {sizeof(struct btf_type) + offsetof(struct btf_array, type),
+				    sizeof(struct btf_type) + offsetof(struct btf_array, index_type)}
+			};
+			break;
+		case BTF_KIND_STRUCT:
+		case BTF_KIND_UNION:
+			it->desc = (struct btf_field_desc) {
+				0, {},
+				sizeof(struct btf_member),
+				1, {offsetof(struct btf_member, type)}
+			};
+			break;
+		case BTF_KIND_FUNC_PROTO:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, type)},
+				sizeof(struct btf_param),
+				1, {offsetof(struct btf_param, type)}
+			};
+			break;
+		case BTF_KIND_DATASEC:
+			it->desc = (struct btf_field_desc) {
+				0, {},
+				sizeof(struct btf_var_secinfo),
+				1, {offsetof(struct btf_var_secinfo, type)}
+			};
+			break;
+		default:
+			return -EINVAL;
+		}
+		break;
+	case BTF_FIELD_ITER_STRS:
+		switch (btf_kind(t)) {
+		case BTF_KIND_UNKN:
+			it->desc = (struct btf_field_desc) {};
+			break;
+		case BTF_KIND_INT:
+		case BTF_KIND_FLOAT:
+		case BTF_KIND_FWD:
+		case BTF_KIND_ARRAY:
+		case BTF_KIND_CONST:
+		case BTF_KIND_VOLATILE:
+		case BTF_KIND_RESTRICT:
+		case BTF_KIND_PTR:
+		case BTF_KIND_TYPEDEF:
+		case BTF_KIND_FUNC:
+		case BTF_KIND_VAR:
+		case BTF_KIND_DECL_TAG:
+		case BTF_KIND_TYPE_TAG:
+		case BTF_KIND_DATASEC:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, name_off)}
+			};
+			break;
+		case BTF_KIND_ENUM:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, name_off)},
+				sizeof(struct btf_enum),
+				1, {offsetof(struct btf_enum, name_off)}
+			};
+			break;
+		case BTF_KIND_ENUM64:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, name_off)},
+				sizeof(struct btf_enum64),
+				1, {offsetof(struct btf_enum64, name_off)}
+			};
+			break;
+		case BTF_KIND_STRUCT:
+		case BTF_KIND_UNION:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, name_off)},
+				sizeof(struct btf_member),
+				1, {offsetof(struct btf_member, name_off)}
+			};
+			break;
+		case BTF_KIND_FUNC_PROTO:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, name_off)},
+				sizeof(struct btf_param),
+				1, {offsetof(struct btf_param, name_off)}
+			};
+			break;
+		default:
+			return -EINVAL;
+		}
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (it->desc.m_sz)
+		it->vlen = btf_vlen(t);
+
+	it->p = t;
+	return 0;
+}
+
+__u32 *btf_field_iter_next(struct btf_field_iter *it)
+{
+	if (!it->p)
+		return NULL;
+
+	if (it->m_idx < 0) {
+		if (it->off_idx < it->desc.t_cnt)
+			return it->p + it->desc.t_offs[it->off_idx++];
+		/* move to per-member iteration */
+		it->m_idx = 0;
+		it->p += sizeof(struct btf_type);
+		it->off_idx = 0;
+	}
+
+	/* if type doesn't have members, stop */
+	if (it->desc.m_sz == 0) {
+		it->p = NULL;
+		return NULL;
+	}
+
+	if (it->off_idx >= it->desc.m_cnt) {
+		/* exhausted this member's fields, go to the next member */
+		it->m_idx++;
+		it->p += it->desc.m_sz;
+		it->off_idx = 0;
+	}
+
+	if (it->m_idx < it->vlen)
+		return it->p + it->desc.m_offs[it->off_idx++];
+
+	it->p = NULL;
+	return NULL;
+}
+
 int btf_ext_visit_type_ids(struct btf_ext *btf_ext, type_id_visit_fn visit, void *ctx)
 {
 	const struct btf_ext_info *seg;
diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index 7e7e686008c6..80f3d346db33 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -508,6 +508,30 @@ struct bpf_line_info_min {
 	__u32	line_col;
 };
 
+enum btf_field_iter_kind {
+	BTF_FIELD_ITER_IDS,
+	BTF_FIELD_ITER_STRS,
+};
+
+struct btf_field_desc {
+	/* once-per-type offsets */
+	int t_cnt, t_offs[2];
+	/* member struct size, or zero, if no members */
+	int m_sz;
+	/* repeated per-member offsets */
+	int m_cnt, m_offs[1];
+};
+
+struct btf_field_iter {
+	struct btf_field_desc desc;
+	void *p;
+	int m_idx;
+	int off_idx;
+	int vlen;
+};
+
+int btf_field_iter_init(struct btf_field_iter *it, struct btf_type *t, enum btf_field_iter_kind iter_kind);
+__u32 *btf_field_iter_next(struct btf_field_iter *it);
 
 typedef int (*type_id_visit_fn)(__u32 *type_id, void *ctx);
 typedef int (*str_off_visit_fn)(__u32 *str_off, void *ctx);
-- 
cgit v1.2.3-58-ga151


From 2bce2c1cb2f0acbf619737a10575f99df0c43984 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Tue, 4 Jun 2024 17:16:26 -0700
Subject: libbpf: Make use of BTF field iterator in BPF linker code

Switch all BPF linker code dealing with iterating BTF type ID and string
offset fields to new btf_field_iter facilities.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240605001629.4061937-3-andrii@kernel.org
---
 tools/lib/bpf/btf.c             |  4 +--
 tools/lib/bpf/libbpf_internal.h |  4 +--
 tools/lib/bpf/linker.c          | 58 +++++++++++++++++++++++++----------------
 3 files changed, 40 insertions(+), 26 deletions(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 50ff8b6eaf36..d72260ac26a5 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -5267,7 +5267,7 @@ __u32 *btf_field_iter_next(struct btf_field_iter *it)
 		return NULL;
 
 	if (it->m_idx < 0) {
-		if (it->off_idx < it->desc.t_cnt)
+		if (it->off_idx < it->desc.t_off_cnt)
 			return it->p + it->desc.t_offs[it->off_idx++];
 		/* move to per-member iteration */
 		it->m_idx = 0;
@@ -5281,7 +5281,7 @@ __u32 *btf_field_iter_next(struct btf_field_iter *it)
 		return NULL;
 	}
 
-	if (it->off_idx >= it->desc.m_cnt) {
+	if (it->off_idx >= it->desc.m_off_cnt) {
 		/* exhausted this member's fields, go to the next member */
 		it->m_idx++;
 		it->p += it->desc.m_sz;
diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index 80f3d346db33..96c0b0993f8b 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -515,11 +515,11 @@ enum btf_field_iter_kind {
 
 struct btf_field_desc {
 	/* once-per-type offsets */
-	int t_cnt, t_offs[2];
+	int t_off_cnt, t_offs[2];
 	/* member struct size, or zero, if no members */
 	int m_sz;
 	/* repeated per-member offsets */
-	int m_cnt, m_offs[1];
+	int m_off_cnt, m_offs[1];
 };
 
 struct btf_field_iter {
diff --git a/tools/lib/bpf/linker.c b/tools/lib/bpf/linker.c
index 0d4be829551b..fa11a671da3e 100644
--- a/tools/lib/bpf/linker.c
+++ b/tools/lib/bpf/linker.c
@@ -957,19 +957,33 @@ static int check_btf_str_off(__u32 *str_off, void *ctx)
 static int linker_sanity_check_btf(struct src_obj *obj)
 {
 	struct btf_type *t;
-	int i, n, err = 0;
+	int i, n, err;
 
 	if (!obj->btf)
 		return 0;
 
 	n = btf__type_cnt(obj->btf);
 	for (i = 1; i < n; i++) {
+		struct btf_field_iter it;
+		__u32 *type_id, *str_off;
+
 		t = btf_type_by_id(obj->btf, i);
 
-		err = err ?: btf_type_visit_type_ids(t, check_btf_type_id, obj->btf);
-		err = err ?: btf_type_visit_str_offs(t, check_btf_str_off, obj->btf);
+		err = btf_field_iter_init(&it, t, BTF_FIELD_ITER_IDS);
 		if (err)
 			return err;
+		while ((type_id = btf_field_iter_next(&it))) {
+			if (*type_id >= n)
+				return -EINVAL;
+		}
+
+		err = btf_field_iter_init(&it, t, BTF_FIELD_ITER_STRS);
+		if (err)
+			return err;
+		while ((str_off = btf_field_iter_next(&it))) {
+			if (!btf__str_by_offset(obj->btf, *str_off))
+				return -EINVAL;
+		}
 	}
 
 	return 0;
@@ -2234,26 +2248,10 @@ static int linker_fixup_btf(struct src_obj *obj)
 	return 0;
 }
 
-static int remap_type_id(__u32 *type_id, void *ctx)
-{
-	int *id_map = ctx;
-	int new_id = id_map[*type_id];
-
-	/* Error out if the type wasn't remapped. Ignore VOID which stays VOID. */
-	if (new_id == 0 && *type_id != 0) {
-		pr_warn("failed to find new ID mapping for original BTF type ID %u\n", *type_id);
-		return -EINVAL;
-	}
-
-	*type_id = id_map[*type_id];
-
-	return 0;
-}
-
 static int linker_append_btf(struct bpf_linker *linker, struct src_obj *obj)
 {
 	const struct btf_type *t;
-	int i, j, n, start_id, id;
+	int i, j, n, start_id, id, err;
 	const char *name;
 
 	if (!obj->btf)
@@ -2324,9 +2322,25 @@ static int linker_append_btf(struct bpf_linker *linker, struct src_obj *obj)
 	n = btf__type_cnt(linker->btf);
 	for (i = start_id; i < n; i++) {
 		struct btf_type *dst_t = btf_type_by_id(linker->btf, i);
+		struct btf_field_iter it;
+		__u32 *type_id;
 
-		if (btf_type_visit_type_ids(dst_t, remap_type_id, obj->btf_type_map))
-			return -EINVAL;
+		err = btf_field_iter_init(&it, dst_t, BTF_FIELD_ITER_IDS);
+		if (err)
+			return err;
+
+		while ((type_id = btf_field_iter_next(&it))) {
+			int new_id = obj->btf_type_map[*type_id];
+
+			/* Error out if the type wasn't remapped. Ignore VOID which stays VOID. */
+			if (new_id == 0 && *type_id != 0) {
+				pr_warn("failed to find new ID mapping for original BTF type ID %u\n",
+					*type_id);
+				return -EINVAL;
+			}
+
+			*type_id = obj->btf_type_map[*type_id];
+		}
 	}
 
 	/* Rewrite VAR/FUNC underlying types (i.e., FUNC's FUNC_PROTO and VAR's
-- 
cgit v1.2.3-58-ga151


From c2641123696b572a3b059e1b45777317ba9f9086 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Tue, 4 Jun 2024 17:16:27 -0700
Subject: libbpf: Make use of BTF field iterator in BTF handling code

Use new BTF field iterator logic to replace all the callback-based
visitor calls. There is still a .BTF.ext callback-based visitor APIs
that should be converted, which will happens as a follow up.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240605001629.4061937-4-andrii@kernel.org
---
 tools/lib/bpf/btf.c | 76 +++++++++++++++++++++++++++++++++++++----------------
 1 file changed, 54 insertions(+), 22 deletions(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index d72260ac26a5..0190fd819f58 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -1739,9 +1739,8 @@ struct btf_pipe {
 	struct hashmap *str_off_map; /* map string offsets from src to dst */
 };
 
-static int btf_rewrite_str(__u32 *str_off, void *ctx)
+static int btf_rewrite_str(struct btf_pipe *p, __u32 *str_off)
 {
-	struct btf_pipe *p = ctx;
 	long mapped_off;
 	int off, err;
 
@@ -1774,7 +1773,9 @@ static int btf_rewrite_str(__u32 *str_off, void *ctx)
 int btf__add_type(struct btf *btf, const struct btf *src_btf, const struct btf_type *src_type)
 {
 	struct btf_pipe p = { .src = src_btf, .dst = btf };
+	struct btf_field_iter it;
 	struct btf_type *t;
+	__u32 *str_off;
 	int sz, err;
 
 	sz = btf_type_size(src_type);
@@ -1791,26 +1792,17 @@ int btf__add_type(struct btf *btf, const struct btf *src_btf, const struct btf_t
 
 	memcpy(t, src_type, sz);
 
-	err = btf_type_visit_str_offs(t, btf_rewrite_str, &p);
+	err = btf_field_iter_init(&it, t, BTF_FIELD_ITER_STRS);
 	if (err)
 		return libbpf_err(err);
 
-	return btf_commit_type(btf, sz);
-}
-
-static int btf_rewrite_type_ids(__u32 *type_id, void *ctx)
-{
-	struct btf *btf = ctx;
-
-	if (!*type_id) /* nothing to do for VOID references */
-		return 0;
+	while ((str_off = btf_field_iter_next(&it))) {
+		err = btf_rewrite_str(&p, str_off);
+		if (err)
+			return libbpf_err(err);
+	}
 
-	/* we haven't updated btf's type count yet, so
-	 * btf->start_id + btf->nr_types - 1 is the type ID offset we should
-	 * add to all newly added BTF types
-	 */
-	*type_id += btf->start_id + btf->nr_types - 1;
-	return 0;
+	return btf_commit_type(btf, sz);
 }
 
 static size_t btf_dedup_identity_hash_fn(long key, void *ctx);
@@ -1858,6 +1850,9 @@ int btf__add_btf(struct btf *btf, const struct btf *src_btf)
 	memcpy(t, src_btf->types_data, data_sz);
 
 	for (i = 0; i < cnt; i++) {
+		struct btf_field_iter it;
+		__u32 *type_id, *str_off;
+
 		sz = btf_type_size(t);
 		if (sz < 0) {
 			/* unlikely, has to be corrupted src_btf */
@@ -1869,15 +1864,31 @@ int btf__add_btf(struct btf *btf, const struct btf *src_btf)
 		*off = t - btf->types_data;
 
 		/* add, dedup, and remap strings referenced by this BTF type */
-		err = btf_type_visit_str_offs(t, btf_rewrite_str, &p);
+		err = btf_field_iter_init(&it, t, BTF_FIELD_ITER_STRS);
 		if (err)
 			goto err_out;
+		while ((str_off = btf_field_iter_next(&it))) {
+			err = btf_rewrite_str(&p, str_off);
+			if (err)
+				goto err_out;
+		}
 
 		/* remap all type IDs referenced from this BTF type */
-		err = btf_type_visit_type_ids(t, btf_rewrite_type_ids, btf);
+		err = btf_field_iter_init(&it, t, BTF_FIELD_ITER_IDS);
 		if (err)
 			goto err_out;
 
+		while ((type_id = btf_field_iter_next(&it))) {
+			if (!*type_id) /* nothing to do for VOID references */
+				continue;
+
+			/* we haven't updated btf's type count yet, so
+			 * btf->start_id + btf->nr_types - 1 is the type ID offset we should
+			 * add to all newly added BTF types
+			 */
+			*type_id += btf->start_id + btf->nr_types - 1;
+		}
+
 		/* go to next type data and type offset index entry */
 		t += sz;
 		off++;
@@ -3453,11 +3464,19 @@ static int btf_for_each_str_off(struct btf_dedup *d, str_off_visit_fn fn, void *
 	int i, r;
 
 	for (i = 0; i < d->btf->nr_types; i++) {
+		struct btf_field_iter it;
 		struct btf_type *t = btf_type_by_id(d->btf, d->btf->start_id + i);
+		__u32 *str_off;
 
-		r = btf_type_visit_str_offs(t, fn, ctx);
+		r = btf_field_iter_init(&it, t, BTF_FIELD_ITER_STRS);
 		if (r)
 			return r;
+
+		while ((str_off = btf_field_iter_next(&it))) {
+			r = fn(str_off, ctx);
+			if (r)
+				return r;
+		}
 	}
 
 	if (!d->btf_ext)
@@ -4919,10 +4938,23 @@ static int btf_dedup_remap_types(struct btf_dedup *d)
 
 	for (i = 0; i < d->btf->nr_types; i++) {
 		struct btf_type *t = btf_type_by_id(d->btf, d->btf->start_id + i);
+		struct btf_field_iter it;
+		__u32 *type_id;
 
-		r = btf_type_visit_type_ids(t, btf_dedup_remap_type_id, d);
+		r = btf_field_iter_init(&it, t, BTF_FIELD_ITER_IDS);
 		if (r)
 			return r;
+
+		while ((type_id = btf_field_iter_next(&it))) {
+			__u32 resolved_id, new_id;
+
+			resolved_id = resolve_type_id(d, *type_id);
+			new_id = d->hypot_map[resolved_id];
+			if (new_id > BTF_MAX_NR_TYPES)
+				return -EINVAL;
+
+			*type_id = new_id;
+		}
 	}
 
 	if (!d->btf_ext)
-- 
cgit v1.2.3-58-ga151


From e1a8630291fde2a0edac2955e3df48587dac9906 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Tue, 4 Jun 2024 17:16:28 -0700
Subject: bpftool: Use BTF field iterator in btfgen

Switch bpftool's code which is using libbpf-internal
btf_type_visit_type_ids() helper to new btf_field_iter functionality.

This makes bpftool code simpler, but also unblocks removing libbpf's
btf_type_visit_type_ids() helper completely.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Reviewed-by: Quentin Monnet <qmo@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240605001629.4061937-5-andrii@kernel.org
---
 tools/bpf/bpftool/gen.c | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index b3979ddc0189..d244a7de387e 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -2379,15 +2379,6 @@ out:
 	return err;
 }
 
-static int btfgen_remap_id(__u32 *type_id, void *ctx)
-{
-	unsigned int *ids = ctx;
-
-	*type_id = ids[*type_id];
-
-	return 0;
-}
-
 /* Generate BTF from relocation information previously recorded */
 static struct btf *btfgen_get_btf(struct btfgen_info *info)
 {
@@ -2467,10 +2458,15 @@ static struct btf *btfgen_get_btf(struct btfgen_info *info)
 	/* second pass: fix up type ids */
 	for (i = 1; i < btf__type_cnt(btf_new); i++) {
 		struct btf_type *btf_type = (struct btf_type *) btf__type_by_id(btf_new, i);
+		struct btf_field_iter it;
+		__u32 *type_id;
 
-		err = btf_type_visit_type_ids(btf_type, btfgen_remap_id, ids);
+		err = btf_field_iter_init(&it, btf_type, BTF_FIELD_ITER_IDS);
 		if (err)
 			goto err_out;
+
+		while ((type_id = btf_field_iter_next(&it)))
+			*type_id = ids[*type_id];
 	}
 
 	free(ids);
-- 
cgit v1.2.3-58-ga151


From 072088704433f75dacf9e33179dd7a81f0a238d4 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Tue, 4 Jun 2024 17:16:29 -0700
Subject: libbpf: Remove callback-based type/string BTF field visitor helpers

Now that all libbpf/bpftool code switched to btf_field_iter, remove
btf_type_visit_type_ids() and btf_type_visit_str_offs() callback-based
helpers as not needed anymore.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240605001629.4061937-6-andrii@kernel.org
---
 tools/lib/bpf/btf.c             | 130 ----------------------------------------
 tools/lib/bpf/libbpf_internal.h |   2 -
 2 files changed, 132 deletions(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 0190fd819f58..775ca55a541c 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -5035,136 +5035,6 @@ struct btf *btf__load_module_btf(const char *module_name, struct btf *vmlinux_bt
 	return btf__parse_split(path, vmlinux_btf);
 }
 
-int btf_type_visit_type_ids(struct btf_type *t, type_id_visit_fn visit, void *ctx)
-{
-	int i, n, err;
-
-	switch (btf_kind(t)) {
-	case BTF_KIND_INT:
-	case BTF_KIND_FLOAT:
-	case BTF_KIND_ENUM:
-	case BTF_KIND_ENUM64:
-		return 0;
-
-	case BTF_KIND_FWD:
-	case BTF_KIND_CONST:
-	case BTF_KIND_VOLATILE:
-	case BTF_KIND_RESTRICT:
-	case BTF_KIND_PTR:
-	case BTF_KIND_TYPEDEF:
-	case BTF_KIND_FUNC:
-	case BTF_KIND_VAR:
-	case BTF_KIND_DECL_TAG:
-	case BTF_KIND_TYPE_TAG:
-		return visit(&t->type, ctx);
-
-	case BTF_KIND_ARRAY: {
-		struct btf_array *a = btf_array(t);
-
-		err = visit(&a->type, ctx);
-		err = err ?: visit(&a->index_type, ctx);
-		return err;
-	}
-
-	case BTF_KIND_STRUCT:
-	case BTF_KIND_UNION: {
-		struct btf_member *m = btf_members(t);
-
-		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
-			err = visit(&m->type, ctx);
-			if (err)
-				return err;
-		}
-		return 0;
-	}
-
-	case BTF_KIND_FUNC_PROTO: {
-		struct btf_param *m = btf_params(t);
-
-		err = visit(&t->type, ctx);
-		if (err)
-			return err;
-		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
-			err = visit(&m->type, ctx);
-			if (err)
-				return err;
-		}
-		return 0;
-	}
-
-	case BTF_KIND_DATASEC: {
-		struct btf_var_secinfo *m = btf_var_secinfos(t);
-
-		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
-			err = visit(&m->type, ctx);
-			if (err)
-				return err;
-		}
-		return 0;
-	}
-
-	default:
-		return -EINVAL;
-	}
-}
-
-int btf_type_visit_str_offs(struct btf_type *t, str_off_visit_fn visit, void *ctx)
-{
-	int i, n, err;
-
-	err = visit(&t->name_off, ctx);
-	if (err)
-		return err;
-
-	switch (btf_kind(t)) {
-	case BTF_KIND_STRUCT:
-	case BTF_KIND_UNION: {
-		struct btf_member *m = btf_members(t);
-
-		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
-			err = visit(&m->name_off, ctx);
-			if (err)
-				return err;
-		}
-		break;
-	}
-	case BTF_KIND_ENUM: {
-		struct btf_enum *m = btf_enum(t);
-
-		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
-			err = visit(&m->name_off, ctx);
-			if (err)
-				return err;
-		}
-		break;
-	}
-	case BTF_KIND_ENUM64: {
-		struct btf_enum64 *m = btf_enum64(t);
-
-		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
-			err = visit(&m->name_off, ctx);
-			if (err)
-				return err;
-		}
-		break;
-	}
-	case BTF_KIND_FUNC_PROTO: {
-		struct btf_param *m = btf_params(t);
-
-		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
-			err = visit(&m->name_off, ctx);
-			if (err)
-				return err;
-		}
-		break;
-	}
-	default:
-		break;
-	}
-
-	return 0;
-}
-
 int btf_field_iter_init(struct btf_field_iter *it, struct btf_type *t, enum btf_field_iter_kind iter_kind)
 {
 	it->p = NULL;
diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index 96c0b0993f8b..e2f06609c624 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -535,8 +535,6 @@ __u32 *btf_field_iter_next(struct btf_field_iter *it);
 
 typedef int (*type_id_visit_fn)(__u32 *type_id, void *ctx);
 typedef int (*str_off_visit_fn)(__u32 *str_off, void *ctx);
-int btf_type_visit_type_ids(struct btf_type *t, type_id_visit_fn visit, void *ctx);
-int btf_type_visit_str_offs(struct btf_type *t, str_off_visit_fn visit, void *ctx);
 int btf_ext_visit_type_ids(struct btf_ext *btf_ext, type_id_visit_fn visit, void *ctx);
 int btf_ext_visit_str_offs(struct btf_ext *btf_ext, str_off_visit_fn visit, void *ctx);
 __s32 btf__find_by_name_kind_own(const struct btf *btf, const char *type_name,
-- 
cgit v1.2.3-58-ga151


From c790275b5edf5d8280ae520bda7c1f37da460c00 Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadfed@meta.com>
Date: Tue, 4 Jun 2024 02:19:39 -0700
Subject: bnxt_en: fix atomic counter for ptp packets

atomic_dec_if_positive returns new value regardless if it is updated or
not. The commit in fixes changed the behavior of the condition to one
that differs from original code. Restore original condition to properly
maintain atomic counter.

Fixes: 165f87691a89 ("bnxt_en: add timestamping statistics support")
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240604091939.785535-1-vadfed@meta.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 6d9faa78e391..7dc00c0d8992 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -513,7 +513,7 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
 
 		if (ptp && ptp->tx_tstamp_en && !skb_is_gso(skb)) {
-			if (!atomic_dec_if_positive(&ptp->tx_avail)) {
+			if (atomic_dec_if_positive(&ptp->tx_avail) < 0) {
 				atomic64_inc(&ptp->stats.ts_err);
 				goto tx_no_ts;
 			}
-- 
cgit v1.2.3-58-ga151


From 1467713eb2244a54c99028a2eacd44f2d1badaf7 Mon Sep 17 00:00:00 2001
From: David Christensen <drc@linux.ibm.com>
Date: Mon, 3 Jun 2024 17:27:41 -0400
Subject: ionic: advertise 52-bit addressing limitation for MSI-X

Current ionic devices only support 52 internal physical address
lines. This is sufficient for x86_64 systems which have similar
limitations but does not apply to all other architectures,
notably IBM POWER (ppc64). To ensure that MSI/MSI-X vectors are
not set outside the physical address limits of the NIC, set the
no_64bit_msi value of the pci_dev structure during device probe.

Signed-off-by: David Christensen <drc@linux.ibm.com>
Reviewed-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240603212747.1079134-1-drc@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
index 6ba8d4aca0a0..a7146d50f814 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
@@ -326,6 +326,11 @@ static int ionic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_out;
 	}
 
+#ifdef CONFIG_PPC64
+	/* Ensure MSI/MSI-X interrupts lie within addressable physical memory */
+	pdev->no_64bit_msi = 1;
+#endif
+
 	err = ionic_setup_one(ionic);
 	if (err)
 		goto err_out;
-- 
cgit v1.2.3-58-ga151


From 4978478a688861207503f3baabcc9c1a5ff43cce Mon Sep 17 00:00:00 2001
From: Ronak Doshi <ronak.doshi@broadcom.com>
Date: Fri, 31 May 2024 12:30:46 -0700
Subject: vmxnet3: prepare for version 9 changes

vmxnet3 is currently at version 7 and this patch initiates the
preparation to accommodate changes for up to version 9. Introduced
utility macros for vmxnet3 version 9 comparison and update Copyright
information.

Signed-off-by: Ronak Doshi <ronak.doshi@broadcom.com>
Acked-by: Guolin Yang <guolin.yang@broadcom.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240531193050.4132-2-ronak.doshi@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/vmxnet3/Makefile          | 2 +-
 drivers/net/vmxnet3/vmxnet3_defs.h    | 2 +-
 drivers/net/vmxnet3/vmxnet3_drv.c     | 2 +-
 drivers/net/vmxnet3/vmxnet3_ethtool.c | 2 +-
 drivers/net/vmxnet3/vmxnet3_int.h     | 8 +++++++-
 5 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/drivers/net/vmxnet3/Makefile b/drivers/net/vmxnet3/Makefile
index f82870c10205..59ef494ce2e0 100644
--- a/drivers/net/vmxnet3/Makefile
+++ b/drivers/net/vmxnet3/Makefile
@@ -2,7 +2,7 @@
 #
 # Linux driver for VMware's vmxnet3 ethernet NIC.
 #
-# Copyright (C) 2007-2022, VMware, Inc. All Rights Reserved.
+# Copyright (C) 2007-2024, VMware, Inc. All Rights Reserved.
 #
 # This program is free software; you can redistribute it and/or modify it
 # under the terms of the GNU General Public License as published by the
diff --git a/drivers/net/vmxnet3/vmxnet3_defs.h b/drivers/net/vmxnet3/vmxnet3_defs.h
index 41d6767283a6..67387bb7aa24 100644
--- a/drivers/net/vmxnet3/vmxnet3_defs.h
+++ b/drivers/net/vmxnet3/vmxnet3_defs.h
@@ -1,7 +1,7 @@
 /*
  * Linux driver for VMware's vmxnet3 ethernet NIC.
  *
- * Copyright (C) 2008-2022, VMware, Inc. All Rights Reserved.
+ * Copyright (C) 2008-2024, VMware, Inc. All Rights Reserved.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index 89ca6e75fcc6..b3f3136cc8be 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -1,7 +1,7 @@
 /*
  * Linux driver for VMware's vmxnet3 ethernet NIC.
  *
- * Copyright (C) 2008-2022, VMware, Inc. All Rights Reserved.
+ * Copyright (C) 2008-2024, VMware, Inc. All Rights Reserved.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c
index 7e8008d5378a..471f91c4204a 100644
--- a/drivers/net/vmxnet3/vmxnet3_ethtool.c
+++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c
@@ -1,7 +1,7 @@
 /*
  * Linux driver for VMware's vmxnet3 ethernet NIC.
  *
- * Copyright (C) 2008-2022, VMware, Inc. All Rights Reserved.
+ * Copyright (C) 2008-2024, VMware, Inc. All Rights Reserved.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h
index 915aaf18c409..2926dfe8514f 100644
--- a/drivers/net/vmxnet3/vmxnet3_int.h
+++ b/drivers/net/vmxnet3/vmxnet3_int.h
@@ -1,7 +1,7 @@
 /*
  * Linux driver for VMware's vmxnet3 ethernet NIC.
  *
- * Copyright (C) 2008-2022, VMware, Inc. All Rights Reserved.
+ * Copyright (C) 2008-2024, VMware, Inc. All Rights Reserved.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
@@ -84,6 +84,8 @@
 	#define VMXNET3_RSS
 #endif
 
+#define VMXNET3_REV_9		8	/* Vmxnet3 Rev. 9 */
+#define VMXNET3_REV_8		7	/* Vmxnet3 Rev. 8 */
 #define VMXNET3_REV_7		6	/* Vmxnet3 Rev. 7 */
 #define VMXNET3_REV_6		5	/* Vmxnet3 Rev. 6 */
 #define VMXNET3_REV_5		4	/* Vmxnet3 Rev. 5 */
@@ -463,6 +465,10 @@ struct vmxnet3_adapter {
 	(adapter->version >= VMXNET3_REV_6 + 1)
 #define VMXNET3_VERSION_GE_7(adapter) \
 	(adapter->version >= VMXNET3_REV_7 + 1)
+#define VMXNET3_VERSION_GE_8(adapter) \
+	(adapter->version >= VMXNET3_REV_8 + 1)
+#define VMXNET3_VERSION_GE_9(adapter) \
+	(adapter->version >= VMXNET3_REV_9 + 1)
 
 /* must be a multiple of VMXNET3_RING_SIZE_ALIGN */
 #define VMXNET3_DEF_TX_RING_SIZE    512
-- 
cgit v1.2.3-58-ga151


From 4c22fad70256245202f33d6f166bfb47bb634bc3 Mon Sep 17 00:00:00 2001
From: Ronak Doshi <ronak.doshi@broadcom.com>
Date: Fri, 31 May 2024 12:30:47 -0700
Subject: vmxnet3: add latency measurement support in vmxnet3

This patch enhances vmxnet3 to support latency measurement.
This support will help to track the latency in packet processing
between guest virtual nic driver and host. For this purpose, we
introduce a new timestamp ring in vmxnet3 which will be per Tx/Rx
queue. This ring will be used to carry timestamp of the packets
which will be used to calculate the latency.

User can enable latency measurement using realtime knob in vnic
setting in VCenter.

Signed-off-by: Ronak Doshi <ronak.doshi@broadcom.com>
Acked-by: Guolin Yang <guolin.yang@broadcom.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240531193050.4132-3-ronak.doshi@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/vmxnet3/vmxnet3_defs.h |  55 +++++++++++++-
 drivers/net/vmxnet3/vmxnet3_drv.c  | 150 ++++++++++++++++++++++++++++++++++++-
 drivers/net/vmxnet3/vmxnet3_int.h  |  20 +++++
 3 files changed, 220 insertions(+), 5 deletions(-)

diff --git a/drivers/net/vmxnet3/vmxnet3_defs.h b/drivers/net/vmxnet3/vmxnet3_defs.h
index 67387bb7aa24..dcf1cf8e7a86 100644
--- a/drivers/net/vmxnet3/vmxnet3_defs.h
+++ b/drivers/net/vmxnet3/vmxnet3_defs.h
@@ -80,6 +80,8 @@ enum {
 #define VMXNET3_IO_TYPE(addr)           ((addr) >> 24)
 #define VMXNET3_IO_REG(addr)            ((addr) & 0xFFFFFF)
 
+#define VMXNET3_PMC_PSEUDO_TSC  0x10003
+
 enum {
 	VMXNET3_CMD_FIRST_SET = 0xCAFE0000,
 	VMXNET3_CMD_ACTIVATE_DEV = VMXNET3_CMD_FIRST_SET,
@@ -123,6 +125,7 @@ enum {
 	VMXNET3_CMD_GET_RESERVED4,
 	VMXNET3_CMD_GET_MAX_CAPABILITIES,
 	VMXNET3_CMD_GET_DCR0_REG,
+	VMXNET3_CMD_GET_TSRING_DESC_SIZE,
 };
 
 /*
@@ -254,6 +257,24 @@ struct Vmxnet3_RxDesc {
 
 #define VMXNET3_RCD_HDR_INNER_SHIFT  13
 
+struct Vmxnet3TSInfo {
+	u64  tsData:56;
+	u64  tsType:4;
+	u64  tsi:1;      //bit to indicate to set ts
+	u64  pad:3;
+	u64  pad2;
+};
+
+struct Vmxnet3_TxTSDesc {
+	struct Vmxnet3TSInfo ts;
+	u64    pad[14];
+};
+
+struct Vmxnet3_RxTSDesc {
+	struct Vmxnet3TSInfo ts;
+	u64    pad[14];
+};
+
 struct Vmxnet3_RxCompDesc {
 #ifdef __BIG_ENDIAN_BITFIELD
 	u32		ext2:1;
@@ -427,6 +448,13 @@ union Vmxnet3_GenericDesc {
 #define VMXNET3_RXDATA_DESC_SIZE_ALIGN 64
 #define VMXNET3_RXDATA_DESC_SIZE_MASK  (VMXNET3_RXDATA_DESC_SIZE_ALIGN - 1)
 
+/* Rx TS Ring buffer size must be a multiple of 64 bytes */
+#define VMXNET3_RXTS_DESC_SIZE_ALIGN 64
+#define VMXNET3_RXTS_DESC_SIZE_MASK  (VMXNET3_RXTS_DESC_SIZE_ALIGN - 1)
+/* Tx TS Ring buffer size must be a multiple of 64 bytes */
+#define VMXNET3_TXTS_DESC_SIZE_ALIGN 64
+#define VMXNET3_TXTS_DESC_SIZE_MASK  (VMXNET3_TXTS_DESC_SIZE_ALIGN - 1)
+
 /* Max ring size */
 #define VMXNET3_TX_RING_MAX_SIZE   4096
 #define VMXNET3_TC_RING_MAX_SIZE   4096
@@ -439,6 +467,9 @@ union Vmxnet3_GenericDesc {
 
 #define VMXNET3_RXDATA_DESC_MAX_SIZE 2048
 
+#define VMXNET3_TXTS_DESC_MAX_SIZE   256
+#define VMXNET3_RXTS_DESC_MAX_SIZE   256
+
 /* a list of reasons for queue stop */
 
 enum {
@@ -546,6 +577,24 @@ struct Vmxnet3_RxQueueConf {
 };
 
 
+struct Vmxnet3_LatencyConf {
+	u16 sampleRate;
+	u16 pad;
+};
+
+struct Vmxnet3_TxQueueTSConf {
+	__le64  txTSRingBasePA;
+	__le16  txTSRingDescSize; /* size of tx timestamp ring buffer */
+	u16     pad;
+	struct Vmxnet3_LatencyConf latencyConf;
+};
+
+struct Vmxnet3_RxQueueTSConf {
+	__le64  rxTSRingBasePA;
+	__le16  rxTSRingDescSize; /* size of rx timestamp ring buffer */
+	u16     pad[3];
+};
+
 enum vmxnet3_intr_mask_mode {
 	VMXNET3_IMM_AUTO   = 0,
 	VMXNET3_IMM_ACTIVE = 1,
@@ -679,7 +728,8 @@ struct Vmxnet3_TxQueueDesc {
 	/* Driver read after a GET command */
 	struct Vmxnet3_QueueStatus		status;
 	struct UPT1_TxStats			stats;
-	u8					_pad[88]; /* 128 aligned */
+	struct Vmxnet3_TxQueueTSConf            tsConf;
+	u8					_pad[72]; /* 128 aligned */
 };
 
 
@@ -689,7 +739,8 @@ struct Vmxnet3_RxQueueDesc {
 	/* Driver read after a GET commad */
 	struct Vmxnet3_QueueStatus		status;
 	struct UPT1_RxStats			stats;
-	u8				      __pad[88]; /* 128 aligned */
+	struct Vmxnet3_RxQueueTSConf            tsConf;
+	u8				      __pad[72]; /* 128 aligned */
 };
 
 struct Vmxnet3_SetPolling {
diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index b3f3136cc8be..71677115628b 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -143,6 +143,32 @@ vmxnet3_tq_stop(struct vmxnet3_tx_queue *tq, struct vmxnet3_adapter *adapter)
 	netif_stop_subqueue(adapter->netdev, (tq - adapter->tx_queue));
 }
 
+static u64
+vmxnet3_get_cycles(int pmc)
+{
+#ifdef CONFIG_X86
+	return native_read_pmc(pmc);
+#else
+	return 0;
+#endif
+}
+
+static bool
+vmxnet3_apply_timestamp(struct vmxnet3_tx_queue *tq, u16 rate)
+{
+#ifdef CONFIG_X86
+	if (rate > 0) {
+		if (tq->tsPktCount == 1) {
+			if (rate != 1)
+				tq->tsPktCount = rate;
+			return true;
+		}
+		tq->tsPktCount--;
+	}
+#endif
+	return false;
+}
+
 /* Check if capability is supported by UPT device or
  * UPT is even requested
  */
@@ -498,6 +524,12 @@ vmxnet3_tq_destroy(struct vmxnet3_tx_queue *tq,
 				  tq->data_ring.base, tq->data_ring.basePA);
 		tq->data_ring.base = NULL;
 	}
+	if (tq->ts_ring.base) {
+		dma_free_coherent(&adapter->pdev->dev,
+				  tq->tx_ring.size * tq->tx_ts_desc_size,
+				  tq->ts_ring.base, tq->ts_ring.basePA);
+		tq->ts_ring.base = NULL;
+	}
 	if (tq->comp_ring.base) {
 		dma_free_coherent(&adapter->pdev->dev, tq->comp_ring.size *
 				  sizeof(struct Vmxnet3_TxCompDesc),
@@ -535,6 +567,10 @@ vmxnet3_tq_init(struct vmxnet3_tx_queue *tq,
 	memset(tq->data_ring.base, 0,
 	       tq->data_ring.size * tq->txdata_desc_size);
 
+	if (tq->ts_ring.base)
+		memset(tq->ts_ring.base, 0,
+		       tq->tx_ring.size * tq->tx_ts_desc_size);
+
 	/* reset the tx comp ring contents to 0 and reset comp ring states */
 	memset(tq->comp_ring.base, 0, tq->comp_ring.size *
 	       sizeof(struct Vmxnet3_TxCompDesc));
@@ -573,6 +609,18 @@ vmxnet3_tq_create(struct vmxnet3_tx_queue *tq,
 		goto err;
 	}
 
+	if (tq->tx_ts_desc_size != 0) {
+		tq->ts_ring.base = dma_alloc_coherent(&adapter->pdev->dev,
+						      tq->tx_ring.size * tq->tx_ts_desc_size,
+						      &tq->ts_ring.basePA, GFP_KERNEL);
+		if (!tq->ts_ring.base) {
+			netdev_err(adapter->netdev, "failed to allocate tx ts ring\n");
+			tq->tx_ts_desc_size = 0;
+		}
+	} else {
+		tq->ts_ring.base = NULL;
+	}
+
 	tq->comp_ring.base = dma_alloc_coherent(&adapter->pdev->dev,
 			tq->comp_ring.size * sizeof(struct Vmxnet3_TxCompDesc),
 			&tq->comp_ring.basePA, GFP_KERNEL);
@@ -861,6 +909,11 @@ vmxnet3_map_pkt(struct sk_buff *skb, struct vmxnet3_tx_ctx *ctx,
 	/* set the last buf_info for the pkt */
 	tbi->skb = skb;
 	tbi->sop_idx = ctx->sop_txd - tq->tx_ring.base;
+	if (tq->tx_ts_desc_size != 0) {
+		ctx->ts_txd = (struct Vmxnet3_TxTSDesc *)((u8 *)tq->ts_ring.base +
+							  tbi->sop_idx * tq->tx_ts_desc_size);
+		ctx->ts_txd->ts.tsi = 0;
+	}
 
 	return 0;
 }
@@ -968,7 +1021,7 @@ vmxnet3_parse_hdr(struct sk_buff *skb, struct vmxnet3_tx_queue *tq,
 					       skb_headlen(skb));
 		}
 
-		if (skb->len <= VMXNET3_HDR_COPY_SIZE)
+		if (skb->len <= tq->txdata_desc_size)
 			ctx->copy_size = skb->len;
 
 		/* make sure headers are accessible directly */
@@ -1259,6 +1312,14 @@ vmxnet3_tq_xmit(struct sk_buff *skb, struct vmxnet3_tx_queue *tq,
 		gdesc->txd.tci = skb_vlan_tag_get(skb);
 	}
 
+	if (tq->tx_ts_desc_size != 0 &&
+	    adapter->latencyConf->sampleRate != 0) {
+		if (vmxnet3_apply_timestamp(tq, adapter->latencyConf->sampleRate)) {
+			ctx.ts_txd->ts.tsData = vmxnet3_get_cycles(VMXNET3_PMC_PSEUDO_TSC);
+			ctx.ts_txd->ts.tsi = 1;
+		}
+	}
+
 	/* Ensure that the write to (&gdesc->txd)->gen will be observed after
 	 * all other writes to &gdesc->txd.
 	 */
@@ -1608,6 +1669,15 @@ skip_xdp:
 			skip_page_frags = false;
 			ctx->skb = rbi->skb;
 
+			if (rq->rx_ts_desc_size != 0 && rcd->ext2) {
+				struct Vmxnet3_RxTSDesc *ts_rxd;
+
+				ts_rxd = (struct Vmxnet3_RxTSDesc *)((u8 *)rq->ts_ring.base +
+								     idx * rq->rx_ts_desc_size);
+				ts_rxd->ts.tsData = vmxnet3_get_cycles(VMXNET3_PMC_PSEUDO_TSC);
+				ts_rxd->ts.tsi = 1;
+			}
+
 			rxDataRingUsed =
 				VMXNET3_RX_DATA_RING(adapter, rcd->rqID);
 			len = rxDataRingUsed ? rcd->len : rbi->len;
@@ -2007,6 +2077,13 @@ static void vmxnet3_rq_destroy(struct vmxnet3_rx_queue *rq,
 		rq->data_ring.base = NULL;
 	}
 
+	if (rq->ts_ring.base) {
+		dma_free_coherent(&adapter->pdev->dev,
+				  rq->rx_ring[0].size * rq->rx_ts_desc_size,
+				  rq->ts_ring.base, rq->ts_ring.basePA);
+		rq->ts_ring.base = NULL;
+	}
+
 	if (rq->comp_ring.base) {
 		dma_free_coherent(&adapter->pdev->dev, rq->comp_ring.size
 				  * sizeof(struct Vmxnet3_RxCompDesc),
@@ -2090,6 +2167,10 @@ vmxnet3_rq_init(struct vmxnet3_rx_queue *rq,
 	}
 	vmxnet3_rq_alloc_rx_buf(rq, 1, rq->rx_ring[1].size - 1, adapter);
 
+	if (rq->ts_ring.base)
+		memset(rq->ts_ring.base, 0,
+		       rq->rx_ring[0].size * rq->rx_ts_desc_size);
+
 	/* reset the comp ring */
 	rq->comp_ring.next2proc = 0;
 	memset(rq->comp_ring.base, 0, rq->comp_ring.size *
@@ -2160,6 +2241,21 @@ vmxnet3_rq_create(struct vmxnet3_rx_queue *rq, struct vmxnet3_adapter *adapter)
 		rq->data_ring.desc_size = 0;
 	}
 
+	if (rq->rx_ts_desc_size != 0) {
+		sz = rq->rx_ring[0].size * rq->rx_ts_desc_size;
+		rq->ts_ring.base =
+			dma_alloc_coherent(&adapter->pdev->dev, sz,
+					   &rq->ts_ring.basePA,
+					   GFP_KERNEL);
+		if (!rq->ts_ring.base) {
+			netdev_err(adapter->netdev,
+				   "rx ts ring will be disabled\n");
+			rq->rx_ts_desc_size = 0;
+		}
+	} else {
+		rq->ts_ring.base = NULL;
+	}
+
 	sz = rq->comp_ring.size * sizeof(struct Vmxnet3_RxCompDesc);
 	rq->comp_ring.base = dma_alloc_coherent(&adapter->pdev->dev, sz,
 						&rq->comp_ring.basePA,
@@ -2759,6 +2855,8 @@ vmxnet3_setup_driver_shared(struct vmxnet3_adapter *adapter)
 	struct Vmxnet3_DSDevReadExt *devReadExt = &shared->devReadExt;
 	struct Vmxnet3_TxQueueConf *tqc;
 	struct Vmxnet3_RxQueueConf *rqc;
+	struct Vmxnet3_TxQueueTSConf *tqtsc;
+	struct Vmxnet3_RxQueueTSConf *rqtsc;
 	int i;
 
 	memset(shared, 0, sizeof(*shared));
@@ -2815,6 +2913,11 @@ vmxnet3_setup_driver_shared(struct vmxnet3_adapter *adapter)
 		tqc->compRingSize   = cpu_to_le32(tq->comp_ring.size);
 		tqc->ddLen          = cpu_to_le32(0);
 		tqc->intrIdx        = tq->comp_ring.intr_idx;
+		if (VMXNET3_VERSION_GE_9(adapter)) {
+			tqtsc = &adapter->tqd_start[i].tsConf;
+			tqtsc->txTSRingBasePA = cpu_to_le64(tq->ts_ring.basePA);
+			tqtsc->txTSRingDescSize = cpu_to_le16(tq->tx_ts_desc_size);
+		}
 	}
 
 	/* rx queue settings */
@@ -2837,6 +2940,11 @@ vmxnet3_setup_driver_shared(struct vmxnet3_adapter *adapter)
 			rqc->rxDataRingDescSize =
 				cpu_to_le16(rq->data_ring.desc_size);
 		}
+		if (VMXNET3_VERSION_GE_9(adapter)) {
+			rqtsc = &adapter->rqd_start[i].tsConf;
+			rqtsc->rxTSRingBasePA = cpu_to_le64(rq->ts_ring.basePA);
+			rqtsc->rxTSRingDescSize = cpu_to_le16(rq->rx_ts_desc_size);
+		}
 	}
 
 #ifdef VMXNET3_RSS
@@ -3299,6 +3407,8 @@ vmxnet3_create_queues(struct vmxnet3_adapter *adapter, u32 tx_ring_size,
 		tq->stopped = true;
 		tq->adapter = adapter;
 		tq->qid = i;
+		tq->tx_ts_desc_size = adapter->tx_ts_desc_size;
+		tq->tsPktCount = 1;
 		err = vmxnet3_tq_create(tq, adapter);
 		/*
 		 * Too late to change num_tx_queues. We cannot do away with
@@ -3320,6 +3430,7 @@ vmxnet3_create_queues(struct vmxnet3_adapter *adapter, u32 tx_ring_size,
 		rq->shared = &adapter->rqd_start[i].ctrl;
 		rq->adapter = adapter;
 		rq->data_ring.desc_size = rxdata_desc_size;
+		rq->rx_ts_desc_size = adapter->rx_ts_desc_size;
 		err = vmxnet3_rq_create(rq, adapter);
 		if (err) {
 			if (i == 0) {
@@ -3361,14 +3472,15 @@ vmxnet3_open(struct net_device *netdev)
 	if (VMXNET3_VERSION_GE_3(adapter)) {
 		unsigned long flags;
 		u16 txdata_desc_size;
+		u32 ret;
 
 		spin_lock_irqsave(&adapter->cmd_lock, flags);
 		VMXNET3_WRITE_BAR1_REG(adapter, VMXNET3_REG_CMD,
 				       VMXNET3_CMD_GET_TXDATA_DESC_SIZE);
-		txdata_desc_size = VMXNET3_READ_BAR1_REG(adapter,
-							 VMXNET3_REG_CMD);
+		ret = VMXNET3_READ_BAR1_REG(adapter, VMXNET3_REG_CMD);
 		spin_unlock_irqrestore(&adapter->cmd_lock, flags);
 
+		txdata_desc_size = ret & 0xffff;
 		if ((txdata_desc_size < VMXNET3_TXDATA_DESC_MIN_SIZE) ||
 		    (txdata_desc_size > VMXNET3_TXDATA_DESC_MAX_SIZE) ||
 		    (txdata_desc_size & VMXNET3_TXDATA_DESC_SIZE_MASK)) {
@@ -3377,10 +3489,40 @@ vmxnet3_open(struct net_device *netdev)
 		} else {
 			adapter->txdata_desc_size = txdata_desc_size;
 		}
+		if (VMXNET3_VERSION_GE_9(adapter))
+			adapter->rxdata_desc_size = (ret >> 16) & 0xffff;
 	} else {
 		adapter->txdata_desc_size = sizeof(struct Vmxnet3_TxDataDesc);
 	}
 
+	if (VMXNET3_VERSION_GE_9(adapter)) {
+		unsigned long flags;
+		u16 tx_ts_desc_size = 0;
+		u16 rx_ts_desc_size = 0;
+		u32 ret;
+
+		spin_lock_irqsave(&adapter->cmd_lock, flags);
+		VMXNET3_WRITE_BAR1_REG(adapter, VMXNET3_REG_CMD,
+				       VMXNET3_CMD_GET_TSRING_DESC_SIZE);
+		ret = VMXNET3_READ_BAR1_REG(adapter, VMXNET3_REG_CMD);
+		spin_unlock_irqrestore(&adapter->cmd_lock, flags);
+		if (ret > 0) {
+			tx_ts_desc_size = (ret & 0xff);
+			rx_ts_desc_size = ((ret >> 16) & 0xff);
+		}
+		if (tx_ts_desc_size > VMXNET3_TXTS_DESC_MAX_SIZE ||
+		    tx_ts_desc_size & VMXNET3_TXTS_DESC_SIZE_MASK)
+			tx_ts_desc_size = 0;
+		if (rx_ts_desc_size > VMXNET3_RXTS_DESC_MAX_SIZE ||
+		    rx_ts_desc_size & VMXNET3_RXTS_DESC_SIZE_MASK)
+			rx_ts_desc_size = 0;
+		adapter->tx_ts_desc_size = tx_ts_desc_size;
+		adapter->rx_ts_desc_size = rx_ts_desc_size;
+	} else {
+		adapter->tx_ts_desc_size = 0;
+		adapter->rx_ts_desc_size = 0;
+	}
+
 	err = vmxnet3_create_queues(adapter,
 				    adapter->tx_ring_size,
 				    adapter->rx_ring_size,
@@ -3992,6 +4134,8 @@ vmxnet3_probe_device(struct pci_dev *pdev,
 	}
 	adapter->rqd_start = (struct Vmxnet3_RxQueueDesc *)(adapter->tqd_start +
 							    adapter->num_tx_queues);
+	if (VMXNET3_VERSION_GE_9(adapter))
+		adapter->latencyConf = &adapter->tqd_start->tsConf.latencyConf;
 
 	adapter->pm_conf = dma_alloc_coherent(&adapter->pdev->dev,
 					      sizeof(struct Vmxnet3_PMConf),
diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h
index 2926dfe8514f..68358e71526c 100644
--- a/drivers/net/vmxnet3/vmxnet3_int.h
+++ b/drivers/net/vmxnet3/vmxnet3_int.h
@@ -193,6 +193,11 @@ struct vmxnet3_tx_data_ring {
 	dma_addr_t          basePA;
 };
 
+struct vmxnet3_tx_ts_ring {
+	struct Vmxnet3_TxTSDesc *base;
+	dma_addr_t          basePA;
+};
+
 #define VMXNET3_MAP_NONE	0
 #define VMXNET3_MAP_SINGLE	BIT(0)
 #define VMXNET3_MAP_PAGE	BIT(1)
@@ -245,6 +250,7 @@ struct vmxnet3_tx_ctx {
 	u32 copy_size;       /* # of bytes copied into the data ring */
 	union Vmxnet3_GenericDesc *sop_txd;
 	union Vmxnet3_GenericDesc *eop_txd;
+	struct Vmxnet3_TxTSDesc *ts_txd;
 };
 
 struct vmxnet3_tx_queue {
@@ -254,6 +260,7 @@ struct vmxnet3_tx_queue {
 	struct vmxnet3_cmd_ring         tx_ring;
 	struct vmxnet3_tx_buf_info      *buf_info;
 	struct vmxnet3_tx_data_ring     data_ring;
+	struct vmxnet3_tx_ts_ring       ts_ring;
 	struct vmxnet3_comp_ring        comp_ring;
 	struct Vmxnet3_TxQueueCtrl      *shared;
 	struct vmxnet3_tq_driver_stats  stats;
@@ -262,6 +269,8 @@ struct vmxnet3_tx_queue {
 						    * stopped */
 	int				qid;
 	u16				txdata_desc_size;
+	u16                             tx_ts_desc_size;
+	u16                             tsPktCount;
 } ____cacheline_aligned;
 
 enum vmxnet3_rx_buf_type {
@@ -309,6 +318,11 @@ struct vmxnet3_rx_data_ring {
 	u16 desc_size;
 };
 
+struct vmxnet3_rx_ts_ring {
+	struct Vmxnet3_RxTSDesc *base;
+	dma_addr_t basePA;
+};
+
 struct vmxnet3_rx_queue {
 	char			name[IFNAMSIZ + 8]; /* To identify interrupt */
 	struct vmxnet3_adapter	  *adapter;
@@ -316,6 +330,7 @@ struct vmxnet3_rx_queue {
 	struct vmxnet3_cmd_ring   rx_ring[2];
 	struct vmxnet3_rx_data_ring data_ring;
 	struct vmxnet3_comp_ring  comp_ring;
+	struct vmxnet3_rx_ts_ring ts_ring;
 	struct vmxnet3_rx_ctx     rx_ctx;
 	u32 qid;            /* rqID in RCD for buffer from 1st ring */
 	u32 qid2;           /* rqID in RCD for buffer from 2nd ring */
@@ -325,6 +340,7 @@ struct vmxnet3_rx_queue {
 	struct vmxnet3_rq_driver_stats  stats;
 	struct page_pool *page_pool;
 	struct xdp_rxq_info xdp_rxq;
+	u16                             rx_ts_desc_size;
 } ____cacheline_aligned;
 
 #define VMXNET3_DEVICE_MAX_TX_QUEUES 32
@@ -434,6 +450,10 @@ struct vmxnet3_adapter {
 	u16    rx_prod_offset;
 	u16    rx_prod2_offset;
 	struct bpf_prog __rcu *xdp_bpf_prog;
+	struct Vmxnet3_LatencyConf *latencyConf;
+	/* Size of buffer in the ts ring */
+	u16     tx_ts_desc_size;
+	u16     rx_ts_desc_size;
 };
 
 #define VMXNET3_WRITE_BAR0_REG(adapter, reg, val)  \
-- 
cgit v1.2.3-58-ga151


From 2e5010fd0c43cb09310bd68efff3d534729a1022 Mon Sep 17 00:00:00 2001
From: Ronak Doshi <ronak.doshi@broadcom.com>
Date: Fri, 31 May 2024 12:30:48 -0700
Subject: vmxnet3: add command to allow disabling of offloads

This patch adds a new command to disable certain offloads. This
allows user to specify, using VM configuration, if certain offloads
need to be disabled.

Signed-off-by: Ronak Doshi <ronak.doshi@broadcom.com>
Acked-by: Guolin Yang <guolin.yang@broadcom.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240531193050.4132-4-ronak.doshi@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/vmxnet3/vmxnet3_defs.h |  4 ++++
 drivers/net/vmxnet3/vmxnet3_drv.c  | 19 +++++++++++++++++++
 drivers/net/vmxnet3/vmxnet3_int.h  |  1 +
 3 files changed, 24 insertions(+)

diff --git a/drivers/net/vmxnet3/vmxnet3_defs.h b/drivers/net/vmxnet3/vmxnet3_defs.h
index dcf1cf8e7a86..5c5148768039 100644
--- a/drivers/net/vmxnet3/vmxnet3_defs.h
+++ b/drivers/net/vmxnet3/vmxnet3_defs.h
@@ -126,6 +126,7 @@ enum {
 	VMXNET3_CMD_GET_MAX_CAPABILITIES,
 	VMXNET3_CMD_GET_DCR0_REG,
 	VMXNET3_CMD_GET_TSRING_DESC_SIZE,
+	VMXNET3_CMD_GET_DISABLED_OFFLOADS,
 };
 
 /*
@@ -912,4 +913,7 @@ struct Vmxnet3_DriverShared {
 /* when new capability is introduced, update VMXNET3_CAP_MAX */
 #define VMXNET3_CAP_MAX                            VMXNET3_CAP_VERSION_7_MAX
 
+#define VMXNET3_OFFLOAD_TSO         BIT(0)
+#define VMXNET3_OFFLOAD_LRO         BIT(1)
+
 #endif /* _VMXNET3_DEFS_H_ */
diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index 71677115628b..e26d223a659f 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -3645,6 +3645,15 @@ static void
 vmxnet3_declare_features(struct vmxnet3_adapter *adapter)
 {
 	struct net_device *netdev = adapter->netdev;
+	unsigned long flags;
+
+	if (VMXNET3_VERSION_GE_9(adapter)) {
+		spin_lock_irqsave(&adapter->cmd_lock, flags);
+		VMXNET3_WRITE_BAR1_REG(adapter, VMXNET3_REG_CMD,
+				       VMXNET3_CMD_GET_DISABLED_OFFLOADS);
+		adapter->disabledOffloads = VMXNET3_READ_BAR1_REG(adapter, VMXNET3_REG_CMD);
+		spin_unlock_irqrestore(&adapter->cmd_lock, flags);
+	}
 
 	netdev->hw_features = NETIF_F_SG | NETIF_F_RXCSUM |
 		NETIF_F_HW_CSUM | NETIF_F_HW_VLAN_CTAG_TX |
@@ -3662,6 +3671,16 @@ vmxnet3_declare_features(struct vmxnet3_adapter *adapter)
 			NETIF_F_GSO_UDP_TUNNEL_CSUM;
 	}
 
+	if (adapter->disabledOffloads & VMXNET3_OFFLOAD_TSO) {
+		netdev->hw_features &= ~(NETIF_F_TSO | NETIF_F_TSO6);
+		netdev->hw_enc_features &= ~(NETIF_F_TSO | NETIF_F_TSO6);
+	}
+
+	if (adapter->disabledOffloads & VMXNET3_OFFLOAD_LRO) {
+		netdev->hw_features &= ~(NETIF_F_LRO);
+		netdev->hw_enc_features &= ~(NETIF_F_LRO);
+	}
+
 	if (VMXNET3_VERSION_GE_7(adapter)) {
 		unsigned long flags;
 
diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h
index 68358e71526c..31e8db568db2 100644
--- a/drivers/net/vmxnet3/vmxnet3_int.h
+++ b/drivers/net/vmxnet3/vmxnet3_int.h
@@ -454,6 +454,7 @@ struct vmxnet3_adapter {
 	/* Size of buffer in the ts ring */
 	u16     tx_ts_desc_size;
 	u16     rx_ts_desc_size;
+	u32     disabledOffloads;
 };
 
 #define VMXNET3_WRITE_BAR0_REG(adapter, reg, val)  \
-- 
cgit v1.2.3-58-ga151


From 63587234d42ad59391cac3619bbb3c9a78c5afaa Mon Sep 17 00:00:00 2001
From: Ronak Doshi <ronak.doshi@broadcom.com>
Date: Fri, 31 May 2024 12:30:49 -0700
Subject: vmxnet3: update to version 9

With all vmxnet3 version 9 changes incorporated in the vmxnet3 driver,
the driver can configure emulation to run at vmxnet3 version 9, provided
the emulation advertises support for version 9.

Signed-off-by: Ronak Doshi <ronak.doshi@broadcom.com>
Acked-by: Guolin Yang <guolin.yang@broadcom.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240531193050.4132-5-ronak.doshi@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/vmxnet3/vmxnet3_drv.c | 46 ++++++++-------------------------------
 drivers/net/vmxnet3/vmxnet3_int.h |  4 ++--
 2 files changed, 11 insertions(+), 39 deletions(-)

diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index e26d223a659f..b16449671b8b 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -3951,7 +3951,7 @@ vmxnet3_probe_device(struct pci_dev *pdev,
 	struct net_device *netdev;
 	struct vmxnet3_adapter *adapter;
 	u8 mac[ETH_ALEN];
-	int size;
+	int size, i;
 	int num_tx_queues;
 	int num_rx_queues;
 	int queues;
@@ -4018,42 +4018,14 @@ vmxnet3_probe_device(struct pci_dev *pdev,
 		goto err_alloc_pci;
 
 	ver = VMXNET3_READ_BAR1_REG(adapter, VMXNET3_REG_VRRS);
-	if (ver & (1 << VMXNET3_REV_7)) {
-		VMXNET3_WRITE_BAR1_REG(adapter,
-				       VMXNET3_REG_VRRS,
-				       1 << VMXNET3_REV_7);
-		adapter->version = VMXNET3_REV_7 + 1;
-	} else if (ver & (1 << VMXNET3_REV_6)) {
-		VMXNET3_WRITE_BAR1_REG(adapter,
-				       VMXNET3_REG_VRRS,
-				       1 << VMXNET3_REV_6);
-		adapter->version = VMXNET3_REV_6 + 1;
-	} else if (ver & (1 << VMXNET3_REV_5)) {
-		VMXNET3_WRITE_BAR1_REG(adapter,
-				       VMXNET3_REG_VRRS,
-				       1 << VMXNET3_REV_5);
-		adapter->version = VMXNET3_REV_5 + 1;
-	} else if (ver & (1 << VMXNET3_REV_4)) {
-		VMXNET3_WRITE_BAR1_REG(adapter,
-				       VMXNET3_REG_VRRS,
-				       1 << VMXNET3_REV_4);
-		adapter->version = VMXNET3_REV_4 + 1;
-	} else if (ver & (1 << VMXNET3_REV_3)) {
-		VMXNET3_WRITE_BAR1_REG(adapter,
-				       VMXNET3_REG_VRRS,
-				       1 << VMXNET3_REV_3);
-		adapter->version = VMXNET3_REV_3 + 1;
-	} else if (ver & (1 << VMXNET3_REV_2)) {
-		VMXNET3_WRITE_BAR1_REG(adapter,
-				       VMXNET3_REG_VRRS,
-				       1 << VMXNET3_REV_2);
-		adapter->version = VMXNET3_REV_2 + 1;
-	} else if (ver & (1 << VMXNET3_REV_1)) {
-		VMXNET3_WRITE_BAR1_REG(adapter,
-				       VMXNET3_REG_VRRS,
-				       1 << VMXNET3_REV_1);
-		adapter->version = VMXNET3_REV_1 + 1;
-	} else {
+	for (i = VMXNET3_REV_9; i >= VMXNET3_REV_1; i--) {
+		if (ver & (1 << i)) {
+			VMXNET3_WRITE_BAR1_REG(adapter, VMXNET3_REG_VRRS, 1 << i);
+			adapter->version = i + 1;
+			break;
+		}
+	}
+	if (i < VMXNET3_REV_1) {
 		dev_err(&pdev->dev,
 			"Incompatible h/w version (0x%x) for adapter\n", ver);
 		err = -EBUSY;
diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h
index 31e8db568db2..9f24d66dbb27 100644
--- a/drivers/net/vmxnet3/vmxnet3_int.h
+++ b/drivers/net/vmxnet3/vmxnet3_int.h
@@ -72,12 +72,12 @@
 /*
  * Version numbers
  */
-#define VMXNET3_DRIVER_VERSION_STRING   "1.7.0.0-k"
+#define VMXNET3_DRIVER_VERSION_STRING   "1.9.0.0-k"
 
 /* Each byte of this 32-bit integer encodes a version number in
  * VMXNET3_DRIVER_VERSION_STRING.
  */
-#define VMXNET3_DRIVER_VERSION_NUM      0x01070000
+#define VMXNET3_DRIVER_VERSION_NUM      0x01090000
 
 #if defined(CONFIG_PCI_MSI)
 	/* RSS only makes sense if MSI-X is supported. */
-- 
cgit v1.2.3-58-ga151


From 955edd872baf29740b714ffa093886918fc9a47b Mon Sep 17 00:00:00 2001
From: Lukasz Majewski <lukma@denx.de>
Date: Mon, 3 Jun 2024 11:33:22 +0200
Subject: selftests: hsr: Extend the hsr_redbox.sh test to use fixed MAC
 addresses

Fixed MAC addresses help with debugging as last four bytes identify the
network namespace.

Moreover, it allows to mimic the real life setup with for example bridge
having the same MAC address on each port.

Signed-off-by: Lukasz Majewski <lukma@denx.de>
Link: https://lore.kernel.org/r/20240603093322.3150030-2-lukma@denx.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/hsr/hsr_redbox.sh | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/tools/testing/selftests/net/hsr/hsr_redbox.sh b/tools/testing/selftests/net/hsr/hsr_redbox.sh
index 1f36785347c0..998103502d5d 100755
--- a/tools/testing/selftests/net/hsr/hsr_redbox.sh
+++ b/tools/testing/selftests/net/hsr/hsr_redbox.sh
@@ -96,6 +96,21 @@ setup_hsr_interfaces()
 	ip -n "${ns4}" link set ns4eth1 up
 	ip -n "${ns5}" link set ns5eth1 up
 
+	ip -net "$ns1" link set address 00:11:22:00:01:01 dev ns1eth1
+	ip -net "$ns1" link set address 00:11:22:00:01:02 dev ns1eth2
+
+	ip -net "$ns2" link set address 00:11:22:00:02:01 dev ns2eth1
+	ip -net "$ns2" link set address 00:11:22:00:02:02 dev ns2eth2
+	ip -net "$ns2" link set address 00:11:22:00:02:03 dev ns2eth3
+
+	ip -net "$ns3" link set address 00:11:22:00:03:11 dev ns3eth1
+	ip -net "$ns3" link set address 00:11:22:00:03:11 dev ns3eth2
+	ip -net "$ns3" link set address 00:11:22:00:03:11 dev ns3eth3
+	ip -net "$ns3" link set address 00:11:22:00:03:11 dev ns3br1
+
+	ip -net "$ns4" link set address 00:11:22:00:04:01 dev ns4eth1
+	ip -net "$ns5" link set address 00:11:22:00:05:01 dev ns5eth1
+
 	ip -net "${ns1}" link add name hsr1 type hsr slave1 ns1eth1 slave2 ns1eth2 supervision 45 version ${HSRv} proto 0
 	ip -net "${ns2}" link add name hsr2 type hsr slave1 ns2eth1 slave2 ns2eth2 interlink ns2eth3 supervision 45 version ${HSRv} proto 0
 
-- 
cgit v1.2.3-58-ga151


From ed20142ed68c2b8819120508bc029e84d13cfe63 Mon Sep 17 00:00:00 2001
From: Lukasz Majewski <lukma@denx.de>
Date: Mon, 3 Jun 2024 11:33:21 +0200
Subject: selftests: hsr: Extend the hsr_ping.sh test to use fixed MAC
 addresses

Fixed MAC addresses help with debugging as last four bytes identify the
network namespace.

Signed-off-by: Lukasz Majewski <lukma@denx.de>
Link: https://lore.kernel.org/r/20240603093322.3150030-1-lukma@denx.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/hsr/hsr_ping.sh | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/tools/testing/selftests/net/hsr/hsr_ping.sh b/tools/testing/selftests/net/hsr/hsr_ping.sh
index 3684b813b0f6..f5d207fc770a 100755
--- a/tools/testing/selftests/net/hsr/hsr_ping.sh
+++ b/tools/testing/selftests/net/hsr/hsr_ping.sh
@@ -152,6 +152,15 @@ setup_hsr_interfaces()
 	ip -net "$ns3" addr add 100.64.0.3/24 dev hsr3
 	ip -net "$ns3" addr add dead:beef:1::3/64 dev hsr3 nodad
 
+	ip -net "$ns1" link set address 00:11:22:00:01:01 dev ns1eth1
+	ip -net "$ns1" link set address 00:11:22:00:01:02 dev ns1eth2
+
+	ip -net "$ns2" link set address 00:11:22:00:02:01 dev ns2eth1
+	ip -net "$ns2" link set address 00:11:22:00:02:02 dev ns2eth2
+
+	ip -net "$ns3" link set address 00:11:22:00:03:01 dev ns3eth1
+	ip -net "$ns3" link set address 00:11:22:00:03:02 dev ns3eth2
+
 	# All Links up
 	ip -net "$ns1" link set ns1eth1 up
 	ip -net "$ns1" link set ns1eth2 up
-- 
cgit v1.2.3-58-ga151


From 4e92d247418c36ed35e9d98b153637828f9882a3 Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:06 +0300
Subject: net/mlx5e: SHAMPO, Use net_prefetch API

Let the SHAMPO functions use the net-specific prefetch API,
similar to all other usages.

Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-2-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index b5333da20e8a..369d101bf03c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -2212,8 +2212,8 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 	if (likely(frag_size <= BIT(MLX5E_SHAMPO_LOG_MAX_HEADER_ENTRY_SIZE))) {
 		/* build SKB around header */
 		dma_sync_single_range_for_cpu(rq->pdev, head->addr, 0, frag_size, rq->buff.map_dir);
-		prefetchw(hdr);
-		prefetch(data);
+		net_prefetchw(hdr);
+		net_prefetch(data);
 		skb = mlx5e_build_linear_skb(rq, hdr, frag_size, rx_headroom, head_size, 0);
 
 		if (unlikely(!skb))
@@ -2230,7 +2230,7 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 			return NULL;
 		}
 
-		prefetchw(skb->data);
+		net_prefetchw(skb->data);
 		mlx5e_copy_skb_header(rq, skb, head->frag_page->page, head->addr,
 				      head_offset + rx_headroom,
 				      rx_headroom, head_size);
-- 
cgit v1.2.3-58-ga151


From 70bd03b89f20b9bbe51a7f73c4950565a17a45f7 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:07 +0300
Subject: net/mlx5e: SHAMPO, Fix incorrect page release

Under the following conditions:
1) No skb created yet
2) header_size == 0 (no SHAMPO header)
3) header_index + 1 % MLX5E_SHAMPO_WQ_HEADER_PER_PAGE == 0 (this is the
   last page fragment of a SHAMPO header page)

a new skb is formed with a page that is NOT a SHAMPO header page (it
is a regular data page). Further down in the same function
(mlx5e_handle_rx_cqe_mpwrq_shampo()), a SHAMPO header page from
header_index is released. This is wrong and it leads to SHAMPO header
pages being released more than once.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-3-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 369d101bf03c..1ddfa00f923f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -2369,7 +2369,8 @@ static void mlx5e_handle_rx_cqe_mpwrq_shampo(struct mlx5e_rq *rq, struct mlx5_cq
 	if (flush)
 		mlx5e_shampo_flush_skb(rq, cqe, match);
 free_hd_entry:
-	mlx5e_free_rx_shampo_hd_entry(rq, header_index);
+	if (likely(head_size))
+		mlx5e_free_rx_shampo_hd_entry(rq, header_index);
 mpwrq_cqe_out:
 	if (likely(wi->consumed_strides < rq->mpwqe.num_strides))
 		return;
-- 
cgit v1.2.3-58-ga151


From fba8334721e266f92079632598e46e5f89082f30 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:08 +0300
Subject: net/mlx5e: SHAMPO, Fix invalid WQ linked list unlink

When all the strides in a WQE have been consumed, the WQE is unlinked
from the WQ linked list (mlx5_wq_ll_pop()). For SHAMPO, it is possible
to receive CQEs with 0 consumed strides for the same WQE even after the
WQE is fully consumed and unlinked. This triggers an additional unlink
for the same wqe which corrupts the linked list.

Fix this scenario by accepting 0 sized consumed strides without
unlinking the WQE again.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-4-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 1ddfa00f923f..b3ef0dd23729 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -2375,6 +2375,9 @@ mpwrq_cqe_out:
 	if (likely(wi->consumed_strides < rq->mpwqe.num_strides))
 		return;
 
+	if (unlikely(!cstrides))
+		return;
+
 	wq  = &rq->mpwqe.wq;
 	wqe = mlx5_wq_ll_get_wqe(wq, wqe_id);
 	mlx5_wq_ll_pop(wq, cqe->wqe_id, &wqe->next.next_wqe_index);
-- 
cgit v1.2.3-58-ga151


From a64bbd8c286fa67319df73ce842e3f6c7b08b6f6 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:09 +0300
Subject: net/mlx5e: SHAMPO, Fix FCS config when HW GRO on

For the following scenario:

ethtool --features eth3 rx-gro-hw on
ethtool --features eth3 rx-fcs on
ethtool --features eth3 rx-fcs off

... there is a firmware error because the driver enables HW GRO first
while FCS is still enabled.

This patch fixes this by swapping the order of HW GRO and FCS for this
specific case. Take LRO into consideration as well for consistency.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-5-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index c53c99dde558..d0808dbe69d3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4259,13 +4259,19 @@ int mlx5e_set_features(struct net_device *netdev, netdev_features_t features)
 #define MLX5E_HANDLE_FEATURE(feature, handler) \
 	mlx5e_handle_feature(netdev, &oper_features, feature, handler)
 
-	err |= MLX5E_HANDLE_FEATURE(NETIF_F_LRO, set_feature_lro);
-	err |= MLX5E_HANDLE_FEATURE(NETIF_F_GRO_HW, set_feature_hw_gro);
+	if (features & (NETIF_F_GRO_HW | NETIF_F_LRO)) {
+		err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXFCS, set_feature_rx_fcs);
+		err |= MLX5E_HANDLE_FEATURE(NETIF_F_LRO, set_feature_lro);
+		err |= MLX5E_HANDLE_FEATURE(NETIF_F_GRO_HW, set_feature_hw_gro);
+	} else {
+		err |= MLX5E_HANDLE_FEATURE(NETIF_F_LRO, set_feature_lro);
+		err |= MLX5E_HANDLE_FEATURE(NETIF_F_GRO_HW, set_feature_hw_gro);
+		err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXFCS, set_feature_rx_fcs);
+	}
 	err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_VLAN_CTAG_FILTER,
 				    set_feature_cvlan_filter);
 	err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_TC, set_feature_hw_tc);
 	err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXALL, set_feature_rx_all);
-	err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXFCS, set_feature_rx_fcs);
 	err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_VLAN_CTAG_RX, set_feature_rx_vlan);
 #ifdef CONFIG_MLX5_EN_ARFS
 	err |= MLX5E_HANDLE_FEATURE(NETIF_F_NTUPLE, set_feature_arfs);
-- 
cgit v1.2.3-58-ga151


From 083dbb54c4809bcd56c5c09360ee8c7760a55d71 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:10 +0300
Subject: net/mlx5e: SHAMPO, Disable gso_size for non GRO packets

When HW GRO is enabled, forwarding of packets is broken due to gso_size
being set incorrectly on non GRO packets.

Non GRO packets have a skb GRO count of 1. mlx5 always sets gso_size on
the skb, even for non GRO packets. It leans on the fact that gso_size is
normally reset in napi_gro_complete(). But this happens only for packets
from GRO'able protocols (TCP/UDP) that have a gro_receive() handler.

The problematic scenarios are:

1) Non GRO protocol packets are received, validate_xmit_skb() will drop
   them (see EPROTONOSUPPORT in skb_mac_gso_segment()). The fix for
   this case would be to not set gso_size at all for SHAMPO packets with
   header size 0.

2) Packets from a GRO'ed protocol (TCP) are received but immediately
   flushed because they are not GRO'able (TCP SYN for example).
   mlx5e_shampo_update_hdr(), which updates the remaining GRO state on
   the skb, is not called because skb GRO count is 1. The fix here would
   be to always call mlx5e_shampo_update_hdr(), regardless of skb GRO
   count. But this call is expensive

The unified fix for both cases is to reset gso_size before calling
napi_gro_receive(). It is a change that is more effective (no call to
mlx5e_shampo_update_hdr() necessary) and simple (smallest code
footprint).

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-6-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index b3ef0dd23729..a13fa760f948 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -2267,6 +2267,8 @@ mlx5e_shampo_flush_skb(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe, bool match)
 		mlx5e_shampo_align_fragment(skb, rq->mpwqe.log_stride_sz);
 	if (NAPI_GRO_CB(skb)->count > 1)
 		mlx5e_shampo_update_hdr(rq, cqe, match);
+	else
+		skb_shinfo(skb)->gso_size = 0;
 	napi_gro_receive(rq->cq.napi, skb);
 	rq->hw_gro_data->skb = NULL;
 }
-- 
cgit v1.2.3-58-ga151


From e839ac9a89cb3bf1aa1652676fa3d6c79810e55d Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:11 +0300
Subject: net/mlx5e: SHAMPO, Simplify header page release in teardown

The function that releases SHAMPO header pages (mlx5e_shampo_dealloc_hd)
has some complicated logic that comes from the fact that it is called
twice during teardown:
1) To release the posted header pages that didn't get any completions.
2) To release all remaining header pages.

This flow is not necessary: all header pages can be released from the
driver side in one go. Furthermore, the above flow is buggy. Taking the
8 headers per page example:
1) Release fragments 5-7. Page will be released.
2) Release remaining fragments 0-4. The bits in the header will indicate
   that the page needs releasing. But this is incorrect: page was
   released in step 1.

This patch releases all header pages in one go. This simplifies the
header page cleanup function. For consistency, the datapath header
page release API (mlx5e_free_rx_shampo_hd_entry()) is used.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-7-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h      |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 12 +----
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c   | 61 ++++++-----------------
 3 files changed, 17 insertions(+), 58 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index e85fb71bf0b4..ff326601d4a4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -1014,7 +1014,7 @@ void mlx5e_build_ptys2ethtool_map(void);
 bool mlx5e_check_fragmented_striding_rq_cap(struct mlx5_core_dev *mdev, u8 page_shift,
 					    enum mlx5e_mpwrq_umr_mode umr_mode);
 
-void mlx5e_shampo_dealloc_hd(struct mlx5e_rq *rq, u16 len, u16 start, bool close);
+void mlx5e_shampo_dealloc_hd(struct mlx5e_rq *rq);
 void mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats);
 void mlx5e_fold_sw_stats64(struct mlx5e_priv *priv, struct rtnl_link_stats64 *s);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index d0808dbe69d3..d21a87ddc934 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1208,15 +1208,6 @@ void mlx5e_free_rx_missing_descs(struct mlx5e_rq *rq)
 		head = mlx5_wq_ll_get_wqe_next_ix(wq, head);
 	}
 
-	if (test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state)) {
-		u16 len;
-
-		len = (rq->mpwqe.shampo->pi - rq->mpwqe.shampo->ci) &
-		      (rq->mpwqe.shampo->hd_per_wq - 1);
-		mlx5e_shampo_dealloc_hd(rq, len, rq->mpwqe.shampo->ci, false);
-		rq->mpwqe.shampo->pi = rq->mpwqe.shampo->ci;
-	}
-
 	rq->mpwqe.actual_wq_head = wq->head;
 	rq->mpwqe.umr_in_progress = 0;
 	rq->mpwqe.umr_completed = 0;
@@ -1244,8 +1235,7 @@ void mlx5e_free_rx_descs(struct mlx5e_rq *rq)
 		}
 
 		if (test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state))
-			mlx5e_shampo_dealloc_hd(rq, rq->mpwqe.shampo->hd_per_wq,
-						0, true);
+			mlx5e_shampo_dealloc_hd(rq);
 	} else {
 		struct mlx5_wq_cyc *wq = &rq->wqe.wq;
 		u16 missing = mlx5_wq_cyc_missing(wq);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index a13fa760f948..bb59ee0b1567 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -839,44 +839,28 @@ err:
 	return err;
 }
 
-/* This function is responsible to dealloc SHAMPO header buffer.
- * close == true specifies that we are in the middle of closing RQ operation so
- * we go over all the entries and if they are not in use we free them,
- * otherwise we only go over a specific range inside the header buffer that are
- * not in use.
- */
-void mlx5e_shampo_dealloc_hd(struct mlx5e_rq *rq, u16 len, u16 start, bool close)
+static void
+mlx5e_free_rx_shampo_hd_entry(struct mlx5e_rq *rq, u16 header_index)
 {
 	struct mlx5e_shampo_hd *shampo = rq->mpwqe.shampo;
-	struct mlx5e_frag_page *deleted_page = NULL;
-	int hd_per_wq = shampo->hd_per_wq;
-	struct mlx5e_dma_info *hd_info;
-	int i, index = start;
-
-	for (i = 0; i < len; i++, index++) {
-		if (index == hd_per_wq)
-			index = 0;
-
-		if (close && !test_bit(index, shampo->bitmap))
-			continue;
+	u64 addr = shampo->info[header_index].addr;
 
-		hd_info = &shampo->info[index];
-		hd_info->addr = ALIGN_DOWN(hd_info->addr, PAGE_SIZE);
-		if (hd_info->frag_page && hd_info->frag_page != deleted_page) {
-			deleted_page = hd_info->frag_page;
-			mlx5e_page_release_fragmented(rq, hd_info->frag_page);
-		}
+	if (((header_index + 1) & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1)) == 0) {
+		struct mlx5e_dma_info *dma_info = &shampo->info[header_index];
 
-		hd_info->frag_page = NULL;
+		dma_info->addr = ALIGN_DOWN(addr, PAGE_SIZE);
+		mlx5e_page_release_fragmented(rq, dma_info->frag_page);
 	}
+	clear_bit(header_index, shampo->bitmap);
+}
 
-	if (start + len > hd_per_wq) {
-		len -= hd_per_wq - start;
-		bitmap_clear(shampo->bitmap, start, hd_per_wq - start);
-		start = 0;
-	}
+void mlx5e_shampo_dealloc_hd(struct mlx5e_rq *rq)
+{
+	struct mlx5e_shampo_hd *shampo = rq->mpwqe.shampo;
+	int i;
 
-	bitmap_clear(shampo->bitmap, start, len);
+	for_each_set_bit(i, shampo->bitmap, rq->mpwqe.shampo->hd_per_wq)
+		mlx5e_free_rx_shampo_hd_entry(rq, i);
 }
 
 static void mlx5e_dealloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
@@ -2281,21 +2265,6 @@ mlx5e_hw_gro_skb_has_enough_space(struct sk_buff *skb, u16 data_bcnt)
 	return PAGE_SIZE * nr_frags + data_bcnt <= GRO_LEGACY_MAX_SIZE;
 }
 
-static void
-mlx5e_free_rx_shampo_hd_entry(struct mlx5e_rq *rq, u16 header_index)
-{
-	struct mlx5e_shampo_hd *shampo = rq->mpwqe.shampo;
-	u64 addr = shampo->info[header_index].addr;
-
-	if (((header_index + 1) & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1)) == 0) {
-		struct mlx5e_dma_info *dma_info = &shampo->info[header_index];
-
-		dma_info->addr = ALIGN_DOWN(addr, PAGE_SIZE);
-		mlx5e_page_release_fragmented(rq, dma_info->frag_page);
-	}
-	bitmap_clear(shampo->bitmap, header_index, 1);
-}
-
 static void mlx5e_handle_rx_cqe_mpwrq_shampo(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 {
 	u16 data_bcnt		= mpwrq_get_cqe_byte_cnt(cqe) - cqe->shampo.header_size;
-- 
cgit v1.2.3-58-ga151


From d34d7d1973c4d1125c430a9612d97bfc10cf6382 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:12 +0300
Subject: net/mlx5e: SHAMPO, Specialize mlx5e_fill_skb_data()

mlx5e_fill_skb_data() used to have multiple callers. But after the XDP
multibuf refactoring from commit 2cb0e27d43b4 ("net/mlx5e: RX, Prepare
non-linear striding RQ for XDP multi-buffer support") the SHAMPO code
path is the only caller.

Take advantage of this and specialize the function:
- Drop the redundant check.
- Assume that data_bcnt is > 0. This is needed in a downstream patch.

Rename the function as well to make things clear.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Suggested-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-8-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 25 +++++++++++--------------
 1 file changed, 11 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index bb59ee0b1567..1e3a5b2afeae 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -1948,21 +1948,16 @@ const struct mlx5e_rx_handlers mlx5e_rx_handlers_rep = {
 #endif
 
 static void
-mlx5e_fill_skb_data(struct sk_buff *skb, struct mlx5e_rq *rq,
-		    struct mlx5e_frag_page *frag_page,
-		    u32 data_bcnt, u32 data_offset)
+mlx5e_shampo_fill_skb_data(struct sk_buff *skb, struct mlx5e_rq *rq,
+			   struct mlx5e_frag_page *frag_page,
+			   u32 data_bcnt, u32 data_offset)
 {
 	net_prefetchw(skb->data);
 
-	while (data_bcnt) {
+	do {
 		/* Non-linear mode, hence non-XSK, which always uses PAGE_SIZE. */
 		u32 pg_consumed_bytes = min_t(u32, PAGE_SIZE - data_offset, data_bcnt);
-		unsigned int truesize;
-
-		if (test_bit(MLX5E_RQ_STATE_SHAMPO, &rq->state))
-			truesize = pg_consumed_bytes;
-		else
-			truesize = ALIGN(pg_consumed_bytes, BIT(rq->mpwqe.log_stride_sz));
+		unsigned int truesize = pg_consumed_bytes;
 
 		frag_page->frags++;
 		mlx5e_add_skb_frag(rq, skb, frag_page->page, data_offset,
@@ -1971,7 +1966,7 @@ mlx5e_fill_skb_data(struct sk_buff *skb, struct mlx5e_rq *rq,
 		data_bcnt -= pg_consumed_bytes;
 		data_offset = 0;
 		frag_page++;
-	}
+	} while (data_bcnt);
 }
 
 static struct sk_buff *
@@ -2330,10 +2325,12 @@ static void mlx5e_handle_rx_cqe_mpwrq_shampo(struct mlx5e_rq *rq, struct mlx5_cq
 	}
 
 	if (likely(head_size)) {
-		struct mlx5e_frag_page *frag_page;
+		if (data_bcnt) {
+			struct mlx5e_frag_page *frag_page;
 
-		frag_page = &wi->alloc_units.frag_pages[page_idx];
-		mlx5e_fill_skb_data(*skb, rq, frag_page, data_bcnt, data_offset);
+			frag_page = &wi->alloc_units.frag_pages[page_idx];
+			mlx5e_shampo_fill_skb_data(*skb, rq, frag_page, data_bcnt, data_offset);
+		}
 	}
 
 	mlx5e_shampo_complete_rx_cqe(rq, cqe, cqe_bcnt, *skb);
-- 
cgit v1.2.3-58-ga151


From f5a699e00f045856b55a60ebf57e367fca389910 Mon Sep 17 00:00:00 2001
From: Yoray Zack <yorayz@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:13 +0300
Subject: net/mlx5e: SHAMPO, Skipping on duplicate flush of the same SHAMPO SKB

SHAMPO SKB can be flushed in mlx5e_shampo_complete_rx_cqe().
If the SKB was flushed, rq->hw_gro_data->skb was also set to NULL.

We can skip on flushing the SKB in mlx5e_shampo_flush_skb
if rq->hw_gro_data->skb == NULL.

Signed-off-by: Yoray Zack <yorayz@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-9-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 1e3a5b2afeae..3f76c33aada0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -2334,7 +2334,7 @@ static void mlx5e_handle_rx_cqe_mpwrq_shampo(struct mlx5e_rq *rq, struct mlx5_cq
 	}
 
 	mlx5e_shampo_complete_rx_cqe(rq, cqe, cqe_bcnt, *skb);
-	if (flush)
+	if (flush && rq->hw_gro_data->skb)
 		mlx5e_shampo_flush_skb(rq, cqe, match);
 free_hd_entry:
 	if (likely(head_size))
-- 
cgit v1.2.3-58-ga151


From 8f9eb8bb5c5af846a8b1729bd7778d08ca852379 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:14 +0300
Subject: net/mlx5e: SHAMPO, Make GRO counters more precise

Don't count non GRO packets. A non GRO packet is a packet with
a GRO cb count of 1.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-10-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../device_drivers/ethernet/mellanox/mlx5/counters.rst      | 10 ++++++----
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c             | 13 ++++++++-----
 2 files changed, 14 insertions(+), 9 deletions(-)

diff --git a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
index fed821ef9b09..7ed010dbe469 100644
--- a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
+++ b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
@@ -189,17 +189,19 @@ the software port.
 
    * - `rx[i]_gro_packets`
      - Number of received packets processed using hardware-accelerated GRO. The
-       number of hardware GRO offloaded packets received on ring i.
+       number of hardware GRO offloaded packets received on ring i. Only true GRO
+       packets are counted: only packets that are in an SKB with a GRO count > 1.
      - Acceleration
 
    * - `rx[i]_gro_bytes`
      - Number of received bytes processed using hardware-accelerated GRO. The
-       number of hardware GRO offloaded bytes received on ring i.
+       number of hardware GRO offloaded bytes received on ring i. Only true GRO
+       packets are counted: only packets that are in an SKB with a GRO count > 1.
      - Acceleration
 
    * - `rx[i]_gro_skbs`
-     - The number of receive SKBs constructed while performing
-       hardware-accelerated GRO.
+     - The number of GRO SKBs constructed from hardware-accelerated GRO. Only SKBs
+       with a GRO count > 1 are counted.
      - Informative
 
    * - `rx[i]_gro_match_packets`
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 3f76c33aada0..79b486d5475d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -1596,9 +1596,7 @@ static void mlx5e_shampo_complete_rx_cqe(struct mlx5e_rq *rq,
 	struct mlx5e_rq_stats *stats = rq->stats;
 
 	stats->packets++;
-	stats->gro_packets++;
 	stats->bytes += cqe_bcnt;
-	stats->gro_bytes += cqe_bcnt;
 	if (NAPI_GRO_CB(skb)->count != 1)
 		return;
 	mlx5e_build_rx_skb(cqe, cqe_bcnt, rq, skb);
@@ -2240,14 +2238,19 @@ mlx5e_shampo_flush_skb(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe, bool match)
 {
 	struct sk_buff *skb = rq->hw_gro_data->skb;
 	struct mlx5e_rq_stats *stats = rq->stats;
+	u16 gro_count = NAPI_GRO_CB(skb)->count;
 
-	stats->gro_skbs++;
 	if (likely(skb_shinfo(skb)->nr_frags))
 		mlx5e_shampo_align_fragment(skb, rq->mpwqe.log_stride_sz);
-	if (NAPI_GRO_CB(skb)->count > 1)
+	if (gro_count > 1) {
+		stats->gro_skbs++;
+		stats->gro_packets += gro_count;
+		stats->gro_bytes += skb->data_len + skb_headlen(skb) * gro_count;
+
 		mlx5e_shampo_update_hdr(rq, cqe, match);
-	else
+	} else {
 		skb_shinfo(skb)->gso_size = 0;
+	}
 	napi_gro_receive(rq->cq.napi, skb);
 	rq->hw_gro_data->skb = NULL;
 }
-- 
cgit v1.2.3-58-ga151


From 16f448d47a86825a421fc1211336454c42c5254f Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:15 +0300
Subject: net/mlx5e: SHAMPO, Drop rx_gro_match_packets counter

After modifying rx_gro_packets to be more accurate, the
rx_gro_match_packets counter is redundant.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-11-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../networking/device_drivers/ethernet/mellanox/mlx5/counters.rst    | 5 -----
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c                      | 2 --
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c                   | 3 ---
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.h                   | 2 --
 4 files changed, 12 deletions(-)

diff --git a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
index 7ed010dbe469..18638a8e7c73 100644
--- a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
+++ b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
@@ -204,11 +204,6 @@ the software port.
        with a GRO count > 1 are counted.
      - Informative
 
-   * - `rx[i]_gro_match_packets`
-     - Number of received packets processed using hardware-accelerated GRO that
-       met the flow table match criteria.
-     - Informative
-
    * - `rx[i]_gro_large_hds`
      - Number of receive packets using hardware-accelerated GRO that have large
        headers that require additional memory to be allocated.
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 79b486d5475d..7ab7215843b6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -2296,8 +2296,6 @@ static void mlx5e_handle_rx_cqe_mpwrq_shampo(struct mlx5e_rq *rq, struct mlx5_cq
 		goto mpwrq_cqe_out;
 	}
 
-	stats->gro_match_packets += match;
-
 	if (*skb && (!match || !(mlx5e_hw_gro_skb_has_enough_space(*skb, data_bcnt)))) {
 		match = false;
 		mlx5e_shampo_flush_skb(rq, cqe, match);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index e1ed214e8651..a3c79da1525b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -141,7 +141,6 @@ static const struct counter_desc sw_stats_desc[] = {
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_gro_packets) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_gro_bytes) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_gro_skbs) },
-	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_gro_match_packets) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_gro_large_hds) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_ecn_mark) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_removed_vlan_packets) },
@@ -343,7 +342,6 @@ static void mlx5e_stats_grp_sw_update_stats_rq_stats(struct mlx5e_sw_stats *s,
 	s->rx_gro_packets             += rq_stats->gro_packets;
 	s->rx_gro_bytes               += rq_stats->gro_bytes;
 	s->rx_gro_skbs                += rq_stats->gro_skbs;
-	s->rx_gro_match_packets       += rq_stats->gro_match_packets;
 	s->rx_gro_large_hds           += rq_stats->gro_large_hds;
 	s->rx_ecn_mark                += rq_stats->ecn_mark;
 	s->rx_removed_vlan_packets    += rq_stats->removed_vlan_packets;
@@ -2057,7 +2055,6 @@ static const struct counter_desc rq_stats_desc[] = {
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, gro_packets) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, gro_bytes) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, gro_skbs) },
-	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, gro_match_packets) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, gro_large_hds) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, ecn_mark) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, removed_vlan_packets) },
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 650732288616..25daae526caa 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -153,7 +153,6 @@ struct mlx5e_sw_stats {
 	u64 rx_gro_packets;
 	u64 rx_gro_bytes;
 	u64 rx_gro_skbs;
-	u64 rx_gro_match_packets;
 	u64 rx_gro_large_hds;
 	u64 rx_mcast_packets;
 	u64 rx_ecn_mark;
@@ -352,7 +351,6 @@ struct mlx5e_rq_stats {
 	u64 gro_packets;
 	u64 gro_bytes;
 	u64 gro_skbs;
-	u64 gro_match_packets;
 	u64 gro_large_hds;
 	u64 mcast_packets;
 	u64 ecn_mark;
-- 
cgit v1.2.3-58-ga151


From e95c5b9e89126703de8212ed8e50b185a1f970bb Mon Sep 17 00:00:00 2001
From: Tariq Toukan <tariqt@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:16 +0300
Subject: net/mlx5e: SHAMPO, Add header-only ethtool counters for header data
 split

Count the number of header-only packets and bytes from SHAMPO.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-12-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../device_drivers/ethernet/mellanox/mlx5/counters.rst           | 9 +++++++++
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c                  | 3 +++
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c               | 4 ++++
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.h               | 4 ++++
 4 files changed, 20 insertions(+)

diff --git a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
index 18638a8e7c73..3bd72577af9a 100644
--- a/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
+++ b/Documentation/networking/device_drivers/ethernet/mellanox/mlx5/counters.rst
@@ -209,6 +209,15 @@ the software port.
        headers that require additional memory to be allocated.
      - Informative
 
+   * - `rx[i]_hds_nodata_packets`
+     - Number of header only packets in header/data split mode [#accel]_.
+     - Informative
+
+   * - `rx[i]_hds_nodata_bytes`
+     - Number of bytes for header only packets in header/data split mode
+       [#accel]_.
+     - Informative
+
    * - `rx[i]_lro_packets`
      - The number of LRO packets received on ring i [#accel]_.
      - Acceleration
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 7ab7215843b6..3af4f70de334 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -2331,6 +2331,9 @@ static void mlx5e_handle_rx_cqe_mpwrq_shampo(struct mlx5e_rq *rq, struct mlx5_cq
 
 			frag_page = &wi->alloc_units.frag_pages[page_idx];
 			mlx5e_shampo_fill_skb_data(*skb, rq, frag_page, data_bcnt, data_offset);
+		} else {
+			stats->hds_nodata_packets++;
+			stats->hds_nodata_bytes += head_size;
 		}
 	}
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index a3c79da1525b..db1cac68292f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -343,6 +343,8 @@ static void mlx5e_stats_grp_sw_update_stats_rq_stats(struct mlx5e_sw_stats *s,
 	s->rx_gro_bytes               += rq_stats->gro_bytes;
 	s->rx_gro_skbs                += rq_stats->gro_skbs;
 	s->rx_gro_large_hds           += rq_stats->gro_large_hds;
+	s->rx_hds_nodata_packets      += rq_stats->hds_nodata_packets;
+	s->rx_hds_nodata_bytes        += rq_stats->hds_nodata_bytes;
 	s->rx_ecn_mark                += rq_stats->ecn_mark;
 	s->rx_removed_vlan_packets    += rq_stats->removed_vlan_packets;
 	s->rx_csum_none               += rq_stats->csum_none;
@@ -2056,6 +2058,8 @@ static const struct counter_desc rq_stats_desc[] = {
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, gro_bytes) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, gro_skbs) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, gro_large_hds) },
+	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, hds_nodata_packets) },
+	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, hds_nodata_bytes) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, ecn_mark) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, removed_vlan_packets) },
 	{ MLX5E_DECLARE_RX_STAT(struct mlx5e_rq_stats, wqe_err) },
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 25daae526caa..4c5858c1dd82 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -154,6 +154,8 @@ struct mlx5e_sw_stats {
 	u64 rx_gro_bytes;
 	u64 rx_gro_skbs;
 	u64 rx_gro_large_hds;
+	u64 rx_hds_nodata_packets;
+	u64 rx_hds_nodata_bytes;
 	u64 rx_mcast_packets;
 	u64 rx_ecn_mark;
 	u64 rx_removed_vlan_packets;
@@ -352,6 +354,8 @@ struct mlx5e_rq_stats {
 	u64 gro_bytes;
 	u64 gro_skbs;
 	u64 gro_large_hds;
+	u64 hds_nodata_packets;
+	u64 hds_nodata_bytes;
 	u64 mcast_packets;
 	u64 ecn_mark;
 	u64 removed_vlan_packets;
-- 
cgit v1.2.3-58-ga151


From 758191c9ea7bcc45dd99398a538ae4ab27c4029e Mon Sep 17 00:00:00 2001
From: Yoray Zack <yorayz@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:17 +0300
Subject: net/mlx5e: SHAMPO, Use KSMs instead of KLMs

KSM Mkey is KLM Mkey with a fixed buffer size. Due to this fact,
it is a faster mechanism than KLM.

SHAMPO feature used KLMs Mkeys for memory mappings of its headers buffer.
As it used KLMs with the same buffer size for each entry,
we can use KSMs instead.

This commit changes the Mkeys that map the SHAMPO headers buffer
from KLMs to KSMs.

Signed-off-by: Yoray Zack <yorayz@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-13-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h       | 20 +------
 .../net/ethernet/mellanox/mlx5/core/en/params.c    | 12 ++--
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h  | 19 +++++++
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  | 21 ++++---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c    | 65 +++++++++++-----------
 include/linux/mlx5/device.h                        |  1 +
 6 files changed, 71 insertions(+), 67 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index ff326601d4a4..bec784d25d7b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -80,6 +80,7 @@ struct page_pool;
 				 SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
 
 #define MLX5E_RX_MAX_HEAD (256)
+#define MLX5E_SHAMPO_LOG_HEADER_ENTRY_SIZE (8)
 #define MLX5E_SHAMPO_LOG_MAX_HEADER_ENTRY_SIZE (9)
 #define MLX5E_SHAMPO_WQ_HEADER_PER_PAGE (PAGE_SIZE >> MLX5E_SHAMPO_LOG_MAX_HEADER_ENTRY_SIZE)
 #define MLX5E_SHAMPO_WQ_BASE_HEAD_ENTRY_SIZE (64)
@@ -146,25 +147,6 @@ struct page_pool;
 #define MLX5E_TX_XSK_POLL_BUDGET       64
 #define MLX5E_SQ_RECOVER_MIN_INTERVAL  500 /* msecs */
 
-#define MLX5E_KLM_UMR_WQE_SZ(sgl_len)\
-	(sizeof(struct mlx5e_umr_wqe) +\
-	(sizeof(struct mlx5_klm) * (sgl_len)))
-
-#define MLX5E_KLM_UMR_WQEBBS(klm_entries) \
-	(DIV_ROUND_UP(MLX5E_KLM_UMR_WQE_SZ(klm_entries), MLX5_SEND_WQE_BB))
-
-#define MLX5E_KLM_UMR_DS_CNT(klm_entries)\
-	(DIV_ROUND_UP(MLX5E_KLM_UMR_WQE_SZ(klm_entries), MLX5_SEND_WQE_DS))
-
-#define MLX5E_KLM_MAX_ENTRIES_PER_WQE(wqe_size)\
-	(((wqe_size) - sizeof(struct mlx5e_umr_wqe)) / sizeof(struct mlx5_klm))
-
-#define MLX5E_KLM_ENTRIES_PER_WQE(wqe_size)\
-	ALIGN_DOWN(MLX5E_KLM_MAX_ENTRIES_PER_WQE(wqe_size), MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT)
-
-#define MLX5E_MAX_KLM_PER_WQE(mdev) \
-	MLX5E_KLM_ENTRIES_PER_WQE(MLX5_SEND_WQE_BB * mlx5e_get_max_sq_aligned_wqebbs(mdev))
-
 #define mlx5e_state_dereference(priv, p) \
 	rcu_dereference_protected((p), lockdep_is_held(&(priv)->state_lock))
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
index ec819dfc98be..6c9ccccca81e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c
@@ -1071,18 +1071,18 @@ static u32 mlx5e_shampo_icosq_sz(struct mlx5_core_dev *mdev,
 				 struct mlx5e_params *params,
 				 struct mlx5e_rq_param *rq_param)
 {
-	int max_num_of_umr_per_wqe, max_hd_per_wqe, max_klm_per_umr, rest;
+	int max_num_of_umr_per_wqe, max_hd_per_wqe, max_ksm_per_umr, rest;
 	void *wqc = MLX5_ADDR_OF(rqc, rq_param->rqc, wq);
 	int wq_size = BIT(MLX5_GET(wq, wqc, log_wq_sz));
 	u32 wqebbs;
 
-	max_klm_per_umr = MLX5E_MAX_KLM_PER_WQE(mdev);
+	max_ksm_per_umr = MLX5E_MAX_KSM_PER_WQE(mdev);
 	max_hd_per_wqe = mlx5e_shampo_hd_per_wqe(mdev, params, rq_param);
-	max_num_of_umr_per_wqe = max_hd_per_wqe / max_klm_per_umr;
-	rest = max_hd_per_wqe % max_klm_per_umr;
-	wqebbs = MLX5E_KLM_UMR_WQEBBS(max_klm_per_umr) * max_num_of_umr_per_wqe;
+	max_num_of_umr_per_wqe = max_hd_per_wqe / max_ksm_per_umr;
+	rest = max_hd_per_wqe % max_ksm_per_umr;
+	wqebbs = MLX5E_KSM_UMR_WQEBBS(max_ksm_per_umr) * max_num_of_umr_per_wqe;
 	if (rest)
-		wqebbs += MLX5E_KLM_UMR_WQEBBS(rest);
+		wqebbs += MLX5E_KSM_UMR_WQEBBS(rest);
 	wqebbs *= wq_size;
 	return wqebbs;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index 879d698b6119..d1f0f868d494 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -34,6 +34,25 @@
 
 #define MLX5E_RX_ERR_CQE(cqe) (get_cqe_opcode(cqe) != MLX5_CQE_RESP_SEND)
 
+#define MLX5E_KSM_UMR_WQE_SZ(sgl_len)\
+	(sizeof(struct mlx5e_umr_wqe) +\
+	(sizeof(struct mlx5_ksm) * (sgl_len)))
+
+#define MLX5E_KSM_UMR_WQEBBS(ksm_entries) \
+	(DIV_ROUND_UP(MLX5E_KSM_UMR_WQE_SZ(ksm_entries), MLX5_SEND_WQE_BB))
+
+#define MLX5E_KSM_UMR_DS_CNT(ksm_entries)\
+	(DIV_ROUND_UP(MLX5E_KSM_UMR_WQE_SZ(ksm_entries), MLX5_SEND_WQE_DS))
+
+#define MLX5E_KSM_MAX_ENTRIES_PER_WQE(wqe_size)\
+	(((wqe_size) - sizeof(struct mlx5e_umr_wqe)) / sizeof(struct mlx5_ksm))
+
+#define MLX5E_KSM_ENTRIES_PER_WQE(wqe_size)\
+	ALIGN_DOWN(MLX5E_KSM_MAX_ENTRIES_PER_WQE(wqe_size), MLX5_UMR_KSM_NUM_ENTRIES_ALIGNMENT)
+
+#define MLX5E_MAX_KSM_PER_WQE(mdev) \
+	MLX5E_KSM_ENTRIES_PER_WQE(MLX5_SEND_WQE_BB * mlx5e_get_max_sq_aligned_wqebbs(mdev))
+
 static inline
 ktime_t mlx5e_cqe_ts_to_ns(cqe_ts_to_ns func, struct mlx5_clock *clock, u64 cqe_ts)
 {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index d21a87ddc934..2a3e0de51f0e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -504,8 +504,8 @@ static int mlx5e_create_umr_mkey(struct mlx5_core_dev *mdev,
 	return err;
 }
 
-static int mlx5e_create_umr_klm_mkey(struct mlx5_core_dev *mdev,
-				     u64 nentries,
+static int mlx5e_create_umr_ksm_mkey(struct mlx5_core_dev *mdev,
+				     u64 nentries, u8 log_entry_size,
 				     u32 *umr_mkey)
 {
 	int inlen;
@@ -525,12 +525,13 @@ static int mlx5e_create_umr_klm_mkey(struct mlx5_core_dev *mdev,
 	MLX5_SET(mkc, mkc, umr_en, 1);
 	MLX5_SET(mkc, mkc, lw, 1);
 	MLX5_SET(mkc, mkc, lr, 1);
-	MLX5_SET(mkc, mkc, access_mode_1_0, MLX5_MKC_ACCESS_MODE_KLMS);
+	MLX5_SET(mkc, mkc, access_mode_1_0, MLX5_MKC_ACCESS_MODE_KSM);
 	mlx5e_mkey_set_relaxed_ordering(mdev, mkc);
 	MLX5_SET(mkc, mkc, qpn, 0xffffff);
 	MLX5_SET(mkc, mkc, pd, mdev->mlx5e_res.hw_objs.pdn);
 	MLX5_SET(mkc, mkc, translations_octword_size, nentries);
-	MLX5_SET(mkc, mkc, length64, 1);
+	MLX5_SET(mkc, mkc, log_page_size, log_entry_size);
+	MLX5_SET64(mkc, mkc, len, nentries << log_entry_size);
 	err = mlx5_core_create_mkey(mdev, umr_mkey, in, inlen);
 
 	kvfree(in);
@@ -565,14 +566,16 @@ static int mlx5e_create_rq_umr_mkey(struct mlx5_core_dev *mdev, struct mlx5e_rq
 static int mlx5e_create_rq_hd_umr_mkey(struct mlx5_core_dev *mdev,
 				       struct mlx5e_rq *rq)
 {
-	u32 max_klm_size = BIT(MLX5_CAP_GEN(mdev, log_max_klm_list_size));
+	u32 max_ksm_size = BIT(MLX5_CAP_GEN(mdev, log_max_klm_list_size));
 
-	if (max_klm_size < rq->mpwqe.shampo->hd_per_wq) {
-		mlx5_core_err(mdev, "max klm list size 0x%x is smaller than shampo header buffer list size 0x%x\n",
-			      max_klm_size, rq->mpwqe.shampo->hd_per_wq);
+	if (max_ksm_size < rq->mpwqe.shampo->hd_per_wq) {
+		mlx5_core_err(mdev, "max ksm list size 0x%x is smaller than shampo header buffer list size 0x%x\n",
+			      max_ksm_size, rq->mpwqe.shampo->hd_per_wq);
 		return -EINVAL;
 	}
-	return mlx5e_create_umr_klm_mkey(mdev, rq->mpwqe.shampo->hd_per_wq,
+
+	return mlx5e_create_umr_ksm_mkey(mdev, rq->mpwqe.shampo->hd_per_wq,
+					 MLX5E_SHAMPO_LOG_HEADER_ENTRY_SIZE,
 					 &rq->mpwqe.shampo->mkey);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 3af4f70de334..f1fbf60d0356 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -619,25 +619,25 @@ static int bitmap_find_window(unsigned long *bitmap, int len,
 	return min(len, count);
 }
 
-static void build_klm_umr(struct mlx5e_icosq *sq, struct mlx5e_umr_wqe *umr_wqe,
-			  __be32 key, u16 offset, u16 klm_len, u16 wqe_bbs)
+static void build_ksm_umr(struct mlx5e_icosq *sq, struct mlx5e_umr_wqe *umr_wqe,
+			  __be32 key, u16 offset, u16 ksm_len)
 {
-	memset(umr_wqe, 0, offsetof(struct mlx5e_umr_wqe, inline_klms));
+	memset(umr_wqe, 0, offsetof(struct mlx5e_umr_wqe, inline_ksms));
 	umr_wqe->ctrl.opmod_idx_opcode =
 		cpu_to_be32((sq->pc << MLX5_WQE_CTRL_WQE_INDEX_SHIFT) |
 			     MLX5_OPCODE_UMR);
 	umr_wqe->ctrl.umr_mkey = key;
 	umr_wqe->ctrl.qpn_ds = cpu_to_be32((sq->sqn << MLX5_WQE_CTRL_QPN_SHIFT)
-					    | MLX5E_KLM_UMR_DS_CNT(klm_len));
+					    | MLX5E_KSM_UMR_DS_CNT(ksm_len));
 	umr_wqe->uctrl.flags = MLX5_UMR_TRANSLATION_OFFSET_EN | MLX5_UMR_INLINE;
 	umr_wqe->uctrl.xlt_offset = cpu_to_be16(offset);
-	umr_wqe->uctrl.xlt_octowords = cpu_to_be16(klm_len);
+	umr_wqe->uctrl.xlt_octowords = cpu_to_be16(ksm_len);
 	umr_wqe->uctrl.mkey_mask     = cpu_to_be64(MLX5_MKEY_MASK_FREE);
 }
 
 static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq,
 				     struct mlx5e_icosq *sq,
-				     u16 klm_entries, u16 index)
+				     u16 ksm_entries, u16 index)
 {
 	struct mlx5e_shampo_hd *shampo = rq->mpwqe.shampo;
 	u16 entries, pi, header_offset, err, wqe_bbs, new_entries;
@@ -650,20 +650,20 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq,
 	int headroom, i;
 
 	headroom = rq->buff.headroom;
-	new_entries = klm_entries - (shampo->pi & (MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT - 1));
-	entries = ALIGN(klm_entries, MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT);
-	wqe_bbs = MLX5E_KLM_UMR_WQEBBS(entries);
+	new_entries = ksm_entries - (shampo->pi & (MLX5_UMR_KSM_NUM_ENTRIES_ALIGNMENT - 1));
+	entries = ALIGN(ksm_entries, MLX5_UMR_KSM_NUM_ENTRIES_ALIGNMENT);
+	wqe_bbs = MLX5E_KSM_UMR_WQEBBS(entries);
 	pi = mlx5e_icosq_get_next_pi(sq, wqe_bbs);
 	umr_wqe = mlx5_wq_cyc_get_wqe(&sq->wq, pi);
-	build_klm_umr(sq, umr_wqe, shampo->key, index, entries, wqe_bbs);
+	build_ksm_umr(sq, umr_wqe, shampo->key, index, entries);
 
 	frag_page = &shampo->pages[page_index];
 
 	for (i = 0; i < entries; i++, index++) {
 		dma_info = &shampo->info[index];
-		if (i >= klm_entries || (index < shampo->pi && shampo->pi - index <
-					 MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT))
-			goto update_klm;
+		if (i >= ksm_entries || (index < shampo->pi && shampo->pi - index <
+					 MLX5_UMR_KSM_NUM_ENTRIES_ALIGNMENT))
+			goto update_ksm;
 		header_offset = (index & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1)) <<
 			MLX5E_SHAMPO_LOG_MAX_HEADER_ENTRY_SIZE;
 		if (!(header_offset & (PAGE_SIZE - 1))) {
@@ -683,12 +683,11 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq,
 			dma_info->frag_page = frag_page;
 		}
 
-update_klm:
-		umr_wqe->inline_klms[i].bcount =
-			cpu_to_be32(MLX5E_RX_MAX_HEAD);
-		umr_wqe->inline_klms[i].key    = cpu_to_be32(lkey);
-		umr_wqe->inline_klms[i].va     =
-			cpu_to_be64(dma_info->addr + headroom);
+update_ksm:
+		umr_wqe->inline_ksms[i] = (struct mlx5_ksm) {
+			.key = cpu_to_be32(lkey),
+			.va  = cpu_to_be64(dma_info->addr + headroom),
+		};
 	}
 
 	sq->db.wqe_info[pi] = (struct mlx5e_icosq_wqe_info) {
@@ -720,37 +719,37 @@ err_unmap:
 static int mlx5e_alloc_rx_hd_mpwqe(struct mlx5e_rq *rq)
 {
 	struct mlx5e_shampo_hd *shampo = rq->mpwqe.shampo;
-	u16 klm_entries, num_wqe, index, entries_before;
+	u16 ksm_entries, num_wqe, index, entries_before;
 	struct mlx5e_icosq *sq = rq->icosq;
-	int i, err, max_klm_entries, len;
+	int i, err, max_ksm_entries, len;
 
-	max_klm_entries = MLX5E_MAX_KLM_PER_WQE(rq->mdev);
-	klm_entries = bitmap_find_window(shampo->bitmap,
+	max_ksm_entries = MLX5E_MAX_KSM_PER_WQE(rq->mdev);
+	ksm_entries = bitmap_find_window(shampo->bitmap,
 					 shampo->hd_per_wqe,
 					 shampo->hd_per_wq, shampo->pi);
-	if (!klm_entries)
+	if (!ksm_entries)
 		return 0;
 
-	klm_entries += (shampo->pi & (MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT - 1));
-	index = ALIGN_DOWN(shampo->pi, MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT);
+	ksm_entries += (shampo->pi & (MLX5_UMR_KSM_NUM_ENTRIES_ALIGNMENT - 1));
+	index = ALIGN_DOWN(shampo->pi, MLX5_UMR_KSM_NUM_ENTRIES_ALIGNMENT);
 	entries_before = shampo->hd_per_wq - index;
 
-	if (unlikely(entries_before < klm_entries))
-		num_wqe = DIV_ROUND_UP(entries_before, max_klm_entries) +
-			  DIV_ROUND_UP(klm_entries - entries_before, max_klm_entries);
+	if (unlikely(entries_before < ksm_entries))
+		num_wqe = DIV_ROUND_UP(entries_before, max_ksm_entries) +
+			  DIV_ROUND_UP(ksm_entries - entries_before, max_ksm_entries);
 	else
-		num_wqe = DIV_ROUND_UP(klm_entries, max_klm_entries);
+		num_wqe = DIV_ROUND_UP(ksm_entries, max_ksm_entries);
 
 	for (i = 0; i < num_wqe; i++) {
-		len = (klm_entries > max_klm_entries) ? max_klm_entries :
-							klm_entries;
+		len = (ksm_entries > max_ksm_entries) ? max_ksm_entries :
+							ksm_entries;
 		if (unlikely(index + len > shampo->hd_per_wq))
 			len = shampo->hd_per_wq - index;
 		err = mlx5e_build_shampo_hd_umr(rq, sq, len, index);
 		if (unlikely(err))
 			return err;
 		index = (index + len) & (rq->mpwqe.shampo->hd_per_wq - 1);
-		klm_entries -= len;
+		ksm_entries -= len;
 	}
 
 	return 0;
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index d7bb31d9a446..da09bfaa7b81 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -294,6 +294,7 @@ enum {
 #define MLX5_UMR_FLEX_ALIGNMENT 0x40
 #define MLX5_UMR_MTT_NUM_ENTRIES_ALIGNMENT (MLX5_UMR_FLEX_ALIGNMENT / sizeof(struct mlx5_mtt))
 #define MLX5_UMR_KLM_NUM_ENTRIES_ALIGNMENT (MLX5_UMR_FLEX_ALIGNMENT / sizeof(struct mlx5_klm))
+#define MLX5_UMR_KSM_NUM_ENTRIES_ALIGNMENT (MLX5_UMR_FLEX_ALIGNMENT / sizeof(struct mlx5_ksm))
 
 #define MLX5_USER_INDEX_LEN (MLX5_FLD_SZ_BYTES(qpc, user_index) * 8)
 
-- 
cgit v1.2.3-58-ga151


From 99be56171fa9ffea494dfe3d4a7f6e7e51630c2e Mon Sep 17 00:00:00 2001
From: Yoray Zack <yorayz@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:18 +0300
Subject: net/mlx5e: SHAMPO, Re-enable HW-GRO

Add back HW-GRO to the reported features.

As the current implementation of HW-GRO uses KSMs with a
specific fixed buffer size (256B) to map its headers buffer,
we reported the feature only if the NIC is supporting KSM and
the minimum value for buffer size is below the requested one.

iperf3 bandwidth comparison:
+---------+--------+--------+-----------+
| streams | SW GRO | HW GRO | Unit      |
|---------+--------+--------+-----------|
| 1       | 36     | 42     | Gbits/sec |
| 4       | 34     | 39     | Gbits/sec |
| 8       | 31     | 35     | Gbits/sec |
+---------+--------+--------+-----------+

A downstream patch will add skb fragment coalescing which will improve
performance considerably.

Benchmark details:
VM based setup
CPU: Intel(R) Xeon(R) Platinum 8380 CPU, 24 cores
NIC: ConnectX-7 100GbE
iperf3 and irq running on same CPU over a single receive queue

Signed-off-by: Yoray Zack <yorayz@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-14-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 26 +++++++++++++++++++++++
 include/linux/mlx5/mlx5_ifc.h                     | 16 +++++++++-----
 2 files changed, 37 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 2a3e0de51f0e..44a64d062e42 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -74,6 +74,27 @@
 #include "lib/devcom.h"
 #include "lib/sd.h"
 
+static bool mlx5e_hw_gro_supported(struct mlx5_core_dev *mdev)
+{
+	if (!MLX5_CAP_GEN(mdev, shampo))
+		return false;
+
+	/* Our HW-GRO implementation relies on "KSM Mkey" for
+	 * SHAMPO headers buffer mapping
+	 */
+	if (!MLX5_CAP_GEN(mdev, fixed_buffer_size))
+		return false;
+
+	if (!MLX5_CAP_GEN_2(mdev, min_mkey_log_entity_size_fixed_buffer_valid))
+		return false;
+
+	if (MLX5_CAP_GEN_2(mdev, min_mkey_log_entity_size_fixed_buffer) >
+	    MLX5E_SHAMPO_LOG_HEADER_ENTRY_SIZE)
+		return false;
+
+	return true;
+}
+
 bool mlx5e_check_fragmented_striding_rq_cap(struct mlx5_core_dev *mdev, u8 page_shift,
 					    enum mlx5e_mpwrq_umr_mode umr_mode)
 {
@@ -5331,6 +5352,11 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
 	netdev->hw_features      |= NETIF_F_HW_VLAN_CTAG_FILTER;
 	netdev->hw_features      |= NETIF_F_HW_VLAN_STAG_TX;
 
+	if (mlx5e_hw_gro_supported(mdev) &&
+	    mlx5e_check_fragmented_striding_rq_cap(mdev, PAGE_SHIFT,
+						   MLX5E_MPWRQ_UMR_MODE_ALIGNED))
+		netdev->hw_features    |= NETIF_F_GRO_HW;
+
 	if (mlx5e_tunnel_any_tx_proto_supported(mdev)) {
 		netdev->hw_enc_features |= NETIF_F_HW_CSUM;
 		netdev->hw_enc_features |= NETIF_F_TSO;
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 5df52e15f7d6..17acd0f3ca8e 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -1526,8 +1526,7 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         ts_cqe_to_dest_cqn[0x1];
 	u8         reserved_at_b3[0x6];
 	u8         go_back_n[0x1];
-	u8         shampo[0x1];
-	u8         reserved_at_bb[0x5];
+	u8         reserved_at_ba[0x6];
 
 	u8         max_sgl_for_optimized_performance[0x8];
 	u8         log_max_cq_sz[0x8];
@@ -1744,7 +1743,9 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         reserved_at_280[0x10];
 	u8         max_wqe_sz_sq[0x10];
 
-	u8         reserved_at_2a0[0x10];
+	u8         reserved_at_2a0[0xb];
+	u8         shampo[0x1];
+	u8         reserved_at_2ac[0x4];
 	u8         max_wqe_sz_rq[0x10];
 
 	u8         max_flow_counter_31_16[0x10];
@@ -2017,7 +2018,8 @@ struct mlx5_ifc_cmd_hca_cap_2_bits {
 	u8	   reserved_at_250[0x10];
 
 	u8	   reserved_at_260[0x120];
-	u8	   reserved_at_380[0x10];
+	u8	   reserved_at_380[0xb];
+	u8	   min_mkey_log_entity_size_fixed_buffer[0x5];
 	u8	   ec_vf_vport_base[0x10];
 
 	u8	   reserved_at_3a0[0x10];
@@ -2029,7 +2031,11 @@ struct mlx5_ifc_cmd_hca_cap_2_bits {
 	u8	   pcc_ifa2[0x1];
 	u8	   reserved_at_3f1[0xf];
 
-	u8	   reserved_at_400[0x400];
+	u8	   reserved_at_400[0x1];
+	u8	   min_mkey_log_entity_size_fixed_buffer_valid[0x1];
+	u8	   reserved_at_402[0x1e];
+
+	u8	   reserved_at_420[0x3e0];
 };
 
 enum mlx5_ifc_flow_destination_type {
-- 
cgit v1.2.3-58-ga151


From 14ae2fd12be8c5089e43fee8a21cd8631699b97a Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Tue, 4 Jun 2024 00:22:19 +0300
Subject: net/mlx5e: SHAMPO, Coalesce skb fragments to page size

When doing hardware GRO (SHAMPO), the driver puts each data payload of a
packet from the wire into one skb fragment. TCP Zero-Copy expects page
sized skb fragments to be able to do it's page-flipping magic. With the
current way of arranging fragments by the driver, only specific MTUs
(page sized multiple + header size) will yield such page sized fragments
in a high percentage.

This change improves payload arrangement in the skb for hardware GRO by
coalescing payloads into a single skb fragment when possible.

To demonstrate the fix, running tcp_mmap with a MTU of 1500 yields:
- Before:  0 % bytes mmap'ed
- After : 81 % bytes mmap'ed

More importantly, coalescing considerably improves the HW GRO performance.
Here are the results for a iperf3 bandwidth benchmark:
+---------+--------+--------+------------------------+-----------+
| streams | SW GRO | HW GRO | HW GRO with coalescing | Unit      |
|---------+--------+--------+------------------------+-----------|
| 1       | 36     | 42     | 57                     | Gbits/sec |
| 4       | 34     | 39     | 50                     | Gbits/sec |
| 8       | 31     | 35     | 43                     | Gbits/sec |
+---------+--------+--------+------------------------+-----------+

Benchmark details:
VM based setup
CPU: Intel(R) Xeon(R) Platinum 8380 CPU, 24 cores
NIC: ConnectX-7 100GbE
iperf3 and irq running on same CPU over a single receive queue

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240603212219.1037656-15-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index f1fbf60d0356..43f018567faf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -523,15 +523,23 @@ mlx5e_add_skb_shared_info_frag(struct mlx5e_rq *rq, struct skb_shared_info *sinf
 
 static inline void
 mlx5e_add_skb_frag(struct mlx5e_rq *rq, struct sk_buff *skb,
-		   struct page *page, u32 frag_offset, u32 len,
+		   struct mlx5e_frag_page *frag_page,
+		   u32 frag_offset, u32 len,
 		   unsigned int truesize)
 {
-	dma_addr_t addr = page_pool_get_dma_addr(page);
+	dma_addr_t addr = page_pool_get_dma_addr(frag_page->page);
+	u8 next_frag = skb_shinfo(skb)->nr_frags;
 
 	dma_sync_single_for_cpu(rq->pdev, addr + frag_offset, len,
 				rq->buff.map_dir);
-	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
-			page, frag_offset, len, truesize);
+
+	if (skb_can_coalesce(skb, next_frag, frag_page->page, frag_offset)) {
+		skb_coalesce_rx_frag(skb, next_frag - 1, len, truesize);
+	} else {
+		frag_page->frags++;
+		skb_add_rx_frag(skb, next_frag, frag_page->page,
+				frag_offset, len, truesize);
+	}
 }
 
 static inline void
@@ -1956,8 +1964,7 @@ mlx5e_shampo_fill_skb_data(struct sk_buff *skb, struct mlx5e_rq *rq,
 		u32 pg_consumed_bytes = min_t(u32, PAGE_SIZE - data_offset, data_bcnt);
 		unsigned int truesize = pg_consumed_bytes;
 
-		frag_page->frags++;
-		mlx5e_add_skb_frag(rq, skb, frag_page->page, data_offset,
+		mlx5e_add_skb_frag(rq, skb, frag_page, data_offset,
 				   pg_consumed_bytes, truesize);
 
 		data_bcnt -= pg_consumed_bytes;
-- 
cgit v1.2.3-58-ga151


From 118e640af30c1355cfea1eeaa2960dc9331ee3db Mon Sep 17 00:00:00 2001
From: Paul Barker <paul.barker.ct@bp.renesas.com>
Date: Tue, 4 Jun 2024 08:28:19 +0100
Subject: net: ravb: Simplify poll & receive functions

We don't need to pass the work budget to ravb_rx() by reference, it's
cleaner to pass this by value and return the amount of work done. This
allows us to simplify the ravb_poll() function and use the common
`work_done` variable name seen in other network drivers for consistency
and ease of understanding.

This is a pure refactor and should not affect behaviour.

Signed-off-by: Paul Barker <paul.barker.ct@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/renesas/ravb.h      |  2 +-
 drivers/net/ethernet/renesas/ravb_main.c | 27 ++++++++++++---------------
 2 files changed, 13 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/renesas/ravb.h b/drivers/net/ethernet/renesas/ravb.h
index b48935ec7e28..71de2a7aa27c 100644
--- a/drivers/net/ethernet/renesas/ravb.h
+++ b/drivers/net/ethernet/renesas/ravb.h
@@ -1039,7 +1039,7 @@ struct ravb_ptp {
 };
 
 struct ravb_hw_info {
-	bool (*receive)(struct net_device *ndev, int *quota, int q);
+	int (*receive)(struct net_device *ndev, int budget, int q);
 	void (*set_rate)(struct net_device *ndev);
 	int (*set_feature)(struct net_device *ndev, netdev_features_t features);
 	int (*dmac_init)(struct net_device *ndev);
diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 4d100283c30f..193ad05383a8 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -759,7 +759,7 @@ static struct sk_buff *ravb_get_skb_gbeth(struct net_device *ndev, int entry,
 }
 
 /* Packet receive function for Gigabit Ethernet */
-static bool ravb_rx_gbeth(struct net_device *ndev, int *quota, int q)
+static int ravb_rx_gbeth(struct net_device *ndev, int budget, int q)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
 	const struct ravb_hw_info *info = priv->info;
@@ -781,7 +781,7 @@ static bool ravb_rx_gbeth(struct net_device *ndev, int *quota, int q)
 	for (i = 0; i < limit; i++, priv->cur_rx[q]++) {
 		entry = priv->cur_rx[q] % priv->num_rx_ring[q];
 		desc = &priv->rx_ring[q].desc[entry];
-		if (rx_packets == *quota || desc->die_dt == DT_FEMPTY)
+		if (rx_packets == budget || desc->die_dt == DT_FEMPTY)
 			break;
 
 		/* Descriptor type must be checked before all other reads */
@@ -882,12 +882,11 @@ static bool ravb_rx_gbeth(struct net_device *ndev, int *quota, int q)
 	}
 
 	stats->rx_packets += rx_packets;
-	*quota -= rx_packets;
-	return *quota == 0;
+	return rx_packets;
 }
 
 /* Packet receive function for Ethernet AVB */
-static bool ravb_rx_rcar(struct net_device *ndev, int *quota, int q)
+static int ravb_rx_rcar(struct net_device *ndev, int budget, int q)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
 	const struct ravb_hw_info *info = priv->info;
@@ -906,7 +905,7 @@ static bool ravb_rx_rcar(struct net_device *ndev, int *quota, int q)
 	for (i = 0; i < limit; i++, priv->cur_rx[q]++) {
 		entry = priv->cur_rx[q] % priv->num_rx_ring[q];
 		desc = &priv->rx_ring[q].ex_desc[entry];
-		if (rx_packets == *quota || desc->die_dt == DT_FEMPTY)
+		if (rx_packets == budget || desc->die_dt == DT_FEMPTY)
 			break;
 
 		/* Descriptor type must be checked before all other reads */
@@ -992,17 +991,16 @@ static bool ravb_rx_rcar(struct net_device *ndev, int *quota, int q)
 	}
 
 	stats->rx_packets += rx_packets;
-	*quota -= rx_packets;
-	return *quota == 0;
+	return rx_packets;
 }
 
 /* Packet receive function for Ethernet AVB */
-static bool ravb_rx(struct net_device *ndev, int *quota, int q)
+static int ravb_rx(struct net_device *ndev, int budget, int q)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
 	const struct ravb_hw_info *info = priv->info;
 
-	return info->receive(ndev, quota, q);
+	return info->receive(ndev, budget, q);
 }
 
 static void ravb_rcv_snd_disable(struct net_device *ndev)
@@ -1319,13 +1317,12 @@ static int ravb_poll(struct napi_struct *napi, int budget)
 	unsigned long flags;
 	int q = napi - priv->napi;
 	int mask = BIT(q);
-	int quota = budget;
-	bool unmask;
+	int work_done;
 
 	/* Processing RX Descriptor Ring */
 	/* Clear RX interrupt */
 	ravb_write(ndev, ~(mask | RIS0_RESERVED), RIS0);
-	unmask = !ravb_rx(ndev, &quota, q);
+	work_done = ravb_rx(ndev, budget, q);
 
 	/* Processing TX Descriptor Ring */
 	spin_lock_irqsave(&priv->lock, flags);
@@ -1344,7 +1341,7 @@ static int ravb_poll(struct napi_struct *napi, int budget)
 	if (priv->rx_fifo_errors != ndev->stats.rx_fifo_errors)
 		ndev->stats.rx_fifo_errors = priv->rx_fifo_errors;
 
-	if (!unmask)
+	if (work_done == budget)
 		goto out;
 
 	napi_complete(napi);
@@ -1361,7 +1358,7 @@ static int ravb_poll(struct napi_struct *napi, int budget)
 	spin_unlock_irqrestore(&priv->lock, flags);
 
 out:
-	return budget - quota;
+	return work_done;
 }
 
 static void ravb_set_duplex_gbeth(struct net_device *ndev)
-- 
cgit v1.2.3-58-ga151


From b0e0e20dc60e9e37b7e5bc71f0c912f66ad75529 Mon Sep 17 00:00:00 2001
From: Paul Barker <paul.barker.ct@bp.renesas.com>
Date: Tue, 4 Jun 2024 08:28:20 +0100
Subject: net: ravb: Align poll function with NAPI docs

Align ravb_poll() with the documentation in
`Documentation/networking/kapi.rst` and
`Documentation/networking/napi.rst`.

The documentation says that we should prefer napi_complete_done() over
napi_complete(), and using the former allows us to properly support busy
polling. We should ensure that napi_complete_done() is only called if
the work budget has not been exhausted, and we should only re-arm
interrupts if it returns true.

Signed-off-by: Paul Barker <paul.barker.ct@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/renesas/ravb_main.c | 26 +++++++++++---------------
 1 file changed, 11 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 193ad05383a8..472aa80002be 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -1341,23 +1341,19 @@ static int ravb_poll(struct napi_struct *napi, int budget)
 	if (priv->rx_fifo_errors != ndev->stats.rx_fifo_errors)
 		ndev->stats.rx_fifo_errors = priv->rx_fifo_errors;
 
-	if (work_done == budget)
-		goto out;
-
-	napi_complete(napi);
-
-	/* Re-enable RX/TX interrupts */
-	spin_lock_irqsave(&priv->lock, flags);
-	if (!info->irq_en_dis) {
-		ravb_modify(ndev, RIC0, mask, mask);
-		ravb_modify(ndev, TIC,  mask, mask);
-	} else {
-		ravb_write(ndev, mask, RIE0);
-		ravb_write(ndev, mask, TIE);
+	if (work_done < budget && napi_complete_done(napi, work_done)) {
+		/* Re-enable RX/TX interrupts */
+		spin_lock_irqsave(&priv->lock, flags);
+		if (!info->irq_en_dis) {
+			ravb_modify(ndev, RIC0, mask, mask);
+			ravb_modify(ndev, TIC,  mask, mask);
+		} else {
+			ravb_write(ndev, mask, RIE0);
+			ravb_write(ndev, mask, TIE);
+		}
+		spin_unlock_irqrestore(&priv->lock, flags);
 	}
-	spin_unlock_irqrestore(&priv->lock, flags);
 
-out:
 	return work_done;
 }
 
-- 
cgit v1.2.3-58-ga151


From 37a01c12e9e89fe2545657d7b42ca2de9c780c45 Mon Sep 17 00:00:00 2001
From: Paul Barker <paul.barker.ct@bp.renesas.com>
Date: Tue, 4 Jun 2024 08:28:21 +0100
Subject: net: ravb: Refactor RX ring refill

To reduce code duplication, we add a new RX ring refill function which
can handle both the initial RX ring population (which was split between
ravb_ring_init() and ravb_ring_format()) and the RX ring refill after
polling (in ravb_rx()).

Signed-off-by: Paul Barker <paul.barker.ct@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/renesas/ravb_main.c | 150 ++++++++++++-------------------
 1 file changed, 57 insertions(+), 93 deletions(-)

diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 472aa80002be..63d03929e3f5 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -317,35 +317,43 @@ static void ravb_ring_free(struct net_device *ndev, int q)
 	priv->tx_skb[q] = NULL;
 }
 
-static void ravb_rx_ring_format(struct net_device *ndev, int q)
+static u32
+ravb_rx_ring_refill(struct net_device *ndev, int q, u32 count, gfp_t gfp_mask)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
+	const struct ravb_hw_info *info = priv->info;
 	struct ravb_rx_desc *rx_desc;
-	unsigned int rx_ring_size;
 	dma_addr_t dma_addr;
-	unsigned int i;
+	u32 i, entry;
 
-	rx_ring_size = priv->info->rx_desc_size * priv->num_rx_ring[q];
-	memset(priv->rx_ring[q].raw, 0, rx_ring_size);
-	/* Build RX ring buffer */
-	for (i = 0; i < priv->num_rx_ring[q]; i++) {
-		/* RX descriptor */
-		rx_desc = ravb_rx_get_desc(priv, q, i);
-		rx_desc->ds_cc = cpu_to_le16(priv->info->rx_max_desc_use);
-		dma_addr = dma_map_single(ndev->dev.parent, priv->rx_skb[q][i]->data,
-					  priv->info->rx_max_frame_size,
-					  DMA_FROM_DEVICE);
-		/* We just set the data size to 0 for a failed mapping which
-		 * should prevent DMA from happening...
-		 */
-		if (dma_mapping_error(ndev->dev.parent, dma_addr))
-			rx_desc->ds_cc = cpu_to_le16(0);
-		rx_desc->dptr = cpu_to_le32(dma_addr);
+	for (i = 0; i < count; i++) {
+		entry = (priv->dirty_rx[q] + i) % priv->num_rx_ring[q];
+		rx_desc = ravb_rx_get_desc(priv, q, entry);
+		rx_desc->ds_cc = cpu_to_le16(info->rx_max_desc_use);
+
+		if (!priv->rx_skb[q][entry]) {
+			priv->rx_skb[q][entry] = ravb_alloc_skb(ndev, info,
+								gfp_mask);
+			if (!priv->rx_skb[q][entry])
+				break;
+			dma_addr = dma_map_single(ndev->dev.parent,
+						  priv->rx_skb[q][entry]->data,
+						  priv->info->rx_max_frame_size,
+						  DMA_FROM_DEVICE);
+			skb_checksum_none_assert(priv->rx_skb[q][entry]);
+			/* We just set the data size to 0 for a failed mapping
+			 * which should prevent DMA from happening...
+			 */
+			if (dma_mapping_error(ndev->dev.parent, dma_addr))
+				rx_desc->ds_cc = cpu_to_le16(0);
+			rx_desc->dptr = cpu_to_le32(dma_addr);
+		}
+		/* Descriptor type must be set after all the above writes */
+		dma_wmb();
 		rx_desc->die_dt = DT_FEMPTY;
 	}
-	rx_desc = ravb_rx_get_desc(priv, q, i);
-	rx_desc->dptr = cpu_to_le32((u32)priv->rx_desc_dma[q]);
-	rx_desc->die_dt = DT_LINKFIX; /* type */
+
+	return i;
 }
 
 /* Format skb and descriptor buffer for Ethernet AVB */
@@ -353,6 +361,7 @@ static void ravb_ring_format(struct net_device *ndev, int q)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
 	unsigned int num_tx_desc = priv->num_tx_desc;
+	struct ravb_rx_desc *rx_desc;
 	struct ravb_tx_desc *tx_desc;
 	struct ravb_desc *desc;
 	unsigned int tx_ring_size = sizeof(*tx_desc) * priv->num_tx_ring[q] *
@@ -364,7 +373,13 @@ static void ravb_ring_format(struct net_device *ndev, int q)
 	priv->dirty_rx[q] = 0;
 	priv->dirty_tx[q] = 0;
 
-	ravb_rx_ring_format(ndev, q);
+	/* Regular RX descriptors have already been initialized by
+	 * ravb_rx_ring_refill(), we just need to initialize the final link
+	 * descriptor.
+	 */
+	rx_desc = ravb_rx_get_desc(priv, q, priv->num_rx_ring[q]);
+	rx_desc->dptr = cpu_to_le32((u32)priv->rx_desc_dma[q]);
+	rx_desc->die_dt = DT_LINKFIX; /* type */
 
 	memset(priv->tx_ring[q], 0, tx_ring_size);
 	/* Build TX ring buffer */
@@ -408,11 +423,9 @@ static void *ravb_alloc_rx_desc(struct net_device *ndev, int q)
 static int ravb_ring_init(struct net_device *ndev, int q)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
-	const struct ravb_hw_info *info = priv->info;
 	unsigned int num_tx_desc = priv->num_tx_desc;
 	unsigned int ring_size;
-	struct sk_buff *skb;
-	unsigned int i;
+	u32 num_filled;
 
 	/* Allocate RX and TX skb rings */
 	priv->rx_skb[q] = kcalloc(priv->num_rx_ring[q],
@@ -422,12 +435,18 @@ static int ravb_ring_init(struct net_device *ndev, int q)
 	if (!priv->rx_skb[q] || !priv->tx_skb[q])
 		goto error;
 
-	for (i = 0; i < priv->num_rx_ring[q]; i++) {
-		skb = ravb_alloc_skb(ndev, info, GFP_KERNEL);
-		if (!skb)
-			goto error;
-		priv->rx_skb[q][i] = skb;
-	}
+	/* Allocate all RX descriptors. */
+	if (!ravb_alloc_rx_desc(ndev, q))
+		goto error;
+
+	/* Populate RX ring buffer. */
+	priv->dirty_rx[q] = 0;
+	ring_size = priv->info->rx_desc_size * priv->num_rx_ring[q];
+	memset(priv->rx_ring[q].raw, 0, ring_size);
+	num_filled = ravb_rx_ring_refill(ndev, q, priv->num_rx_ring[q],
+					 GFP_KERNEL);
+	if (num_filled != priv->num_rx_ring[q])
+		goto error;
 
 	if (num_tx_desc > 1) {
 		/* Allocate rings for the aligned buffers */
@@ -437,12 +456,6 @@ static int ravb_ring_init(struct net_device *ndev, int q)
 			goto error;
 	}
 
-	/* Allocate all RX descriptors. */
-	if (!ravb_alloc_rx_desc(ndev, q))
-		goto error;
-
-	priv->dirty_rx[q] = 0;
-
 	/* Allocate all TX descriptors. */
 	ring_size = sizeof(struct ravb_tx_desc) *
 		    (priv->num_tx_ring[q] * num_tx_desc + 1);
@@ -762,11 +775,9 @@ static struct sk_buff *ravb_get_skb_gbeth(struct net_device *ndev, int entry,
 static int ravb_rx_gbeth(struct net_device *ndev, int budget, int q)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
-	const struct ravb_hw_info *info = priv->info;
 	struct net_device_stats *stats;
 	struct ravb_rx_desc *desc;
 	struct sk_buff *skb;
-	dma_addr_t dma_addr;
 	int rx_packets = 0;
 	u8  desc_status;
 	u16 desc_len;
@@ -854,32 +865,9 @@ static int ravb_rx_gbeth(struct net_device *ndev, int budget, int q)
 	}
 
 	/* Refill the RX ring buffers. */
-	for (; priv->cur_rx[q] - priv->dirty_rx[q] > 0; priv->dirty_rx[q]++) {
-		entry = priv->dirty_rx[q] % priv->num_rx_ring[q];
-		desc = &priv->rx_ring[q].desc[entry];
-		desc->ds_cc = cpu_to_le16(priv->info->rx_max_desc_use);
-
-		if (!priv->rx_skb[q][entry]) {
-			skb = ravb_alloc_skb(ndev, info, GFP_ATOMIC);
-			if (!skb)
-				break;
-			dma_addr = dma_map_single(ndev->dev.parent,
-						  skb->data,
-						  priv->info->rx_max_frame_size,
-						  DMA_FROM_DEVICE);
-			skb_checksum_none_assert(skb);
-			/* We just set the data size to 0 for a failed mapping
-			 * which should prevent DMA  from happening...
-			 */
-			if (dma_mapping_error(ndev->dev.parent, dma_addr))
-				desc->ds_cc = cpu_to_le16(0);
-			desc->dptr = cpu_to_le32(dma_addr);
-			priv->rx_skb[q][entry] = skb;
-		}
-		/* Descriptor type must be set after all the above writes */
-		dma_wmb();
-		desc->die_dt = DT_FEMPTY;
-	}
+	priv->dirty_rx[q] += ravb_rx_ring_refill(ndev, q,
+						 priv->cur_rx[q] - priv->dirty_rx[q],
+						 GFP_ATOMIC);
 
 	stats->rx_packets += rx_packets;
 	return rx_packets;
@@ -889,12 +877,10 @@ static int ravb_rx_gbeth(struct net_device *ndev, int budget, int q)
 static int ravb_rx_rcar(struct net_device *ndev, int budget, int q)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
-	const struct ravb_hw_info *info = priv->info;
 	struct net_device_stats *stats = &priv->stats[q];
 	struct ravb_ex_rx_desc *desc;
 	unsigned int limit, i;
 	struct sk_buff *skb;
-	dma_addr_t dma_addr;
 	struct timespec64 ts;
 	int rx_packets = 0;
 	u8  desc_status;
@@ -964,31 +950,9 @@ static int ravb_rx_rcar(struct net_device *ndev, int budget, int q)
 	}
 
 	/* Refill the RX ring buffers. */
-	for (; priv->cur_rx[q] - priv->dirty_rx[q] > 0; priv->dirty_rx[q]++) {
-		entry = priv->dirty_rx[q] % priv->num_rx_ring[q];
-		desc = &priv->rx_ring[q].ex_desc[entry];
-		desc->ds_cc = cpu_to_le16(priv->info->rx_max_desc_use);
-
-		if (!priv->rx_skb[q][entry]) {
-			skb = ravb_alloc_skb(ndev, info, GFP_ATOMIC);
-			if (!skb)
-				break;	/* Better luck next round. */
-			dma_addr = dma_map_single(ndev->dev.parent, skb->data,
-						  priv->info->rx_max_frame_size,
-						  DMA_FROM_DEVICE);
-			skb_checksum_none_assert(skb);
-			/* We just set the data size to 0 for a failed mapping
-			 * which should prevent DMA  from happening...
-			 */
-			if (dma_mapping_error(ndev->dev.parent, dma_addr))
-				desc->ds_cc = cpu_to_le16(0);
-			desc->dptr = cpu_to_le32(dma_addr);
-			priv->rx_skb[q][entry] = skb;
-		}
-		/* Descriptor type must be set after all the above writes */
-		dma_wmb();
-		desc->die_dt = DT_FEMPTY;
-	}
+	priv->dirty_rx[q] += ravb_rx_ring_refill(ndev, q,
+						 priv->cur_rx[q] - priv->dirty_rx[q],
+						 GFP_ATOMIC);
 
 	stats->rx_packets += rx_packets;
 	return rx_packets;
-- 
cgit v1.2.3-58-ga151


From 3ee43f09cb2cd7643d5a4bfbc78a9f75f6a563af Mon Sep 17 00:00:00 2001
From: Paul Barker <paul.barker.ct@bp.renesas.com>
Date: Tue, 4 Jun 2024 08:28:22 +0100
Subject: net: ravb: Refactor GbEth RX code path

We can reduce code duplication in ravb_rx_gbeth().

Signed-off-by: Paul Barker <paul.barker.ct@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/renesas/ravb_main.c | 59 +++++++++++++++++---------------
 1 file changed, 32 insertions(+), 27 deletions(-)

diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 63d03929e3f5..e8c3f48b213d 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -819,47 +819,52 @@ static int ravb_rx_gbeth(struct net_device *ndev, int budget, int q)
 				stats->rx_missed_errors++;
 		} else {
 			die_dt = desc->die_dt & 0xF0;
+			skb = ravb_get_skb_gbeth(ndev, entry, desc);
 			switch (die_dt) {
 			case DT_FSINGLE:
-				skb = ravb_get_skb_gbeth(ndev, entry, desc);
-				skb_put(skb, desc_len);
-				skb->protocol = eth_type_trans(skb, ndev);
-				if (ndev->features & NETIF_F_RXCSUM)
-					ravb_rx_csum_gbeth(skb);
-				napi_gro_receive(&priv->napi[q], skb);
-				rx_packets++;
-				stats->rx_bytes += desc_len;
-				break;
 			case DT_FSTART:
-				priv->rx_1st_skb = ravb_get_skb_gbeth(ndev, entry, desc);
-				skb_put(priv->rx_1st_skb, desc_len);
+				/* Start of packet: Set initial data length. */
+				skb_put(skb, desc_len);
+
+				/* Save this skb if the packet spans multiple
+				 * descriptors.
+				 */
+				if (die_dt == DT_FSTART)
+					priv->rx_1st_skb = skb;
 				break;
+
 			case DT_FMID:
-				skb = ravb_get_skb_gbeth(ndev, entry, desc);
-				skb_copy_to_linear_data_offset(priv->rx_1st_skb,
-							       priv->rx_1st_skb->len,
-							       skb->data,
-							       desc_len);
-				skb_put(priv->rx_1st_skb, desc_len);
-				dev_kfree_skb(skb);
-				break;
 			case DT_FEND:
-				skb = ravb_get_skb_gbeth(ndev, entry, desc);
+				/* Continuing a packet: Move data into the saved
+				 * skb.
+				 */
 				skb_copy_to_linear_data_offset(priv->rx_1st_skb,
 							       priv->rx_1st_skb->len,
 							       skb->data,
 							       desc_len);
 				skb_put(priv->rx_1st_skb, desc_len);
 				dev_kfree_skb(skb);
-				priv->rx_1st_skb->protocol =
-					eth_type_trans(priv->rx_1st_skb, ndev);
+
+				/* Set skb to point at the whole packet so that
+				 * we only need one code path for finishing a
+				 * packet.
+				 */
+				skb = priv->rx_1st_skb;
+			}
+
+			switch (die_dt) {
+			case DT_FSINGLE:
+			case DT_FEND:
+				/* Finishing a packet: Determine protocol &
+				 * checksum, hand off to NAPI and update our
+				 * stats.
+				 */
+				skb->protocol = eth_type_trans(skb, ndev);
 				if (ndev->features & NETIF_F_RXCSUM)
-					ravb_rx_csum_gbeth(priv->rx_1st_skb);
-				stats->rx_bytes += priv->rx_1st_skb->len;
-				napi_gro_receive(&priv->napi[q],
-						 priv->rx_1st_skb);
+					ravb_rx_csum_gbeth(skb);
+				stats->rx_bytes += skb->len;
+				napi_gro_receive(&priv->napi[q], skb);
 				rx_packets++;
-				break;
 			}
 		}
 	}
-- 
cgit v1.2.3-58-ga151


From 7b39c1814ce3bcdf95d026bbb27322218840a27d Mon Sep 17 00:00:00 2001
From: Paul Barker <paul.barker.ct@bp.renesas.com>
Date: Tue, 4 Jun 2024 08:28:23 +0100
Subject: net: ravb: Enable SW IRQ Coalescing for GbEth

Software IRQ Coalescing is required to improve network stack performance
in the RZ/G2L SoC family and the RZ/G3S SoC, i.e. the SoCs which use the
GbEth IP.

This patch gives the following improvements during testing with iperf3:

  * RZ/G2L:
    * TCP RX: same bandwidth with -6% CPU load (76% -> 71%)
    * UDP RX: same bandwidth with -10% CPU load (99% -> 89%)

  * RZ/G2UL:
    * UDP RX: +4200% bandwidth (1.23Mbps -> 53Mbps)

  * RZ/G3S:
    * UDP RX: +425% bandwidth (1.23Mbps -> 6.46Mbps)

The improvement of UDP RX bandwidth for the single core SoCs (RZ/G2UL &
RZ/G3S) is particularly critical.

Signed-off-by: Paul Barker <paul.barker.ct@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/renesas/ravb.h      | 1 +
 drivers/net/ethernet/renesas/ravb_main.c | 4 ++++
 2 files changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/renesas/ravb.h b/drivers/net/ethernet/renesas/ravb.h
index 71de2a7aa27c..6a7aa7dd17e6 100644
--- a/drivers/net/ethernet/renesas/ravb.h
+++ b/drivers/net/ethernet/renesas/ravb.h
@@ -1054,6 +1054,7 @@ struct ravb_hw_info {
 	u32 rx_max_desc_use;
 	u32 rx_desc_size;
 	unsigned aligned_tx: 1;
+	unsigned coalesce_irqs:1;	/* Needs software IRQ coalescing */
 
 	/* hardware features */
 	unsigned internal_delay:1;	/* AVB-DMAC has internal delays */
diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index e8c3f48b213d..2c24cb0271af 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -2667,6 +2667,7 @@ static const struct ravb_hw_info gbeth_hw_info = {
 	.rx_max_desc_use = 4080,
 	.rx_desc_size = sizeof(struct ravb_rx_desc),
 	.aligned_tx = 1,
+	.coalesce_irqs = 1,
 	.tx_counters = 1,
 	.carrier_counters = 1,
 	.half_duplex = 1,
@@ -2943,6 +2944,9 @@ static int ravb_probe(struct platform_device *pdev)
 	if (info->nc_queues)
 		netif_napi_add(ndev, &priv->napi[RAVB_NC], ravb_poll);
 
+	if (info->coalesce_irqs)
+		netdev_sw_irq_coalesce_default_on(ndev);
+
 	/* Network device register */
 	error = register_netdev(ndev);
 	if (error)
-- 
cgit v1.2.3-58-ga151


From 65c482bc226ab25a7884ee6fd1cc09bb08cbd1be Mon Sep 17 00:00:00 2001
From: Paul Barker <paul.barker.ct@bp.renesas.com>
Date: Tue, 4 Jun 2024 08:28:24 +0100
Subject: net: ravb: Use NAPI threaded mode on 1-core CPUs with GbEth IP

NAPI Threaded mode (along with the previously enabled SW IRQ Coalescing)
is required to improve network stack performance for single core SoCs
using the GbEth IP (currently the RZ/G2L SoC family and the RZ/G3S SoC).

This patch gives the following improvements during testing with iperf3.

  * RZ/G2UL:
    * TCP TX: +32% bandwidth (638Mbps -> 841Mbps)
    * TXP RX: +8.8% bandwidth (667Mbps -> 726Mbps)
    * UDP RX: +104% bandwidth (53Mbps -> 108Mbps)

  * RZ/G3S:
    * TCP TX: 29% bandwidth (529Mbps -> 681Mbps)
    * UDP RX: +1290% bandwidth (6.46Mbps -> 90Mbps)

  * RZ/Five:
    * UDP RX: Test no longer crashes (0 -> 20 Mbps)

This patch gives the following reductions in performance in the same
testing:

  * RZ/G2UL:
    * UDP TX: -7.5% bandwidth (594Mbps -> 549Mbps)

  * RZ/G3S:
    * UDP TX: -5% bandwidth (625Mbps -> 594Mbps)

These losses are considered acceptable given the benefits shown above.
If UDP TX bandwidth must be maximised for a particular use case, NAPI
threaded mode can be disabled at runtime via sysfs writes.

The improvement of UDP RX bandwidth for the single core SoCs (RZ/G2UL &
RZ/G3S) is particularly critical.

Signed-off-by: Paul Barker <paul.barker.ct@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/renesas/ravb_main.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 2c24cb0271af..c89bb38492b4 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -2944,8 +2944,11 @@ static int ravb_probe(struct platform_device *pdev)
 	if (info->nc_queues)
 		netif_napi_add(ndev, &priv->napi[RAVB_NC], ravb_poll);
 
-	if (info->coalesce_irqs)
+	if (info->coalesce_irqs) {
 		netdev_sw_irq_coalesce_default_on(ndev);
+		if (num_present_cpus() == 1)
+			dev_set_threaded(ndev, true);
+	}
 
 	/* Network device register */
 	error = register_netdev(ndev);
-- 
cgit v1.2.3-58-ga151


From 966726324b7b14009216fda33b47e0bc003944c6 Mon Sep 17 00:00:00 2001
From: Paul Barker <paul.barker.ct@bp.renesas.com>
Date: Tue, 4 Jun 2024 08:28:25 +0100
Subject: net: ravb: Allocate RX buffers via page pool

This patch makes multiple changes that can't be separated:

  1) Allocate plain RX buffers via a page pool instead of allocating
     SKBs, then use build_skb() when a packet is received.
  2) For GbEth IP, reduce the RX buffer size to 2kB.
  3) For GbEth IP, merge packets which span more than one RX descriptor
     as SKB fragments instead of copying data.

Implementing (1) without (2) would require the use of an order-1 page
pool (instead of an order-0 page pool split into page fragments) for
GbEth.

Implementing (2) without (3) would leave us no space to re-assemble
packets which span more than one RX descriptor.

Implementing (3) without (1) would not be possible as the network stack
expects to use put_page() or page_pool_put_page() to free SKB fragments
after an SKB is consumed.

RX checksum offload support is adjusted to handle both linear and
nonlinear (fragmented) packets.

This patch gives the following improvements during testing with iperf3.

  * RZ/G2L:
    * TCP RX: same bandwidth at -43% CPU load (70% -> 40%)
    * UDP RX: same bandwidth at -17% CPU load (88% -> 74%)

  * RZ/G2UL:
    * TCP RX: +30% bandwidth (726Mbps -> 941Mbps)
    * UDP RX: +417% bandwidth (108Mbps -> 558Mbps)

  * RZ/G3S:
    * TCP RX: +64% bandwidth (562Mbps -> 920Mbps)
    * UDP RX: +420% bandwidth (90Mbps -> 468Mbps)

  * RZ/Five:
    * TCP RX: +217% bandwidth (145Mbps -> 459Mbps)
    * UDP RX: +470% bandwidth (20Mbps -> 114Mbps)

There is no significant impact on bandwidth or CPU load in testing on
RZ/G2H or R-Car M3N.

Signed-off-by: Paul Barker <paul.barker.ct@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/renesas/ravb.h      |  11 +-
 drivers/net/ethernet/renesas/ravb_main.c | 260 ++++++++++++++++++++-----------
 2 files changed, 175 insertions(+), 96 deletions(-)

diff --git a/drivers/net/ethernet/renesas/ravb.h b/drivers/net/ethernet/renesas/ravb.h
index 6a7aa7dd17e6..6b2444d31fcc 100644
--- a/drivers/net/ethernet/renesas/ravb.h
+++ b/drivers/net/ethernet/renesas/ravb.h
@@ -19,6 +19,7 @@
 #include <linux/phy.h>
 #include <linux/platform_device.h>
 #include <linux/ptp_clock_kernel.h>
+#include <net/page_pool/types.h>
 
 #define BE_TX_RING_SIZE	64	/* TX ring size for Best Effort */
 #define BE_RX_RING_SIZE	1024	/* RX ring size for Best Effort */
@@ -1051,7 +1052,7 @@ struct ravb_hw_info {
 	int stats_len;
 	u32 tccr_mask;
 	u32 rx_max_frame_size;
-	u32 rx_max_desc_use;
+	u32 rx_buffer_size;
 	u32 rx_desc_size;
 	unsigned aligned_tx: 1;
 	unsigned coalesce_irqs:1;	/* Needs software IRQ coalescing */
@@ -1071,6 +1072,11 @@ struct ravb_hw_info {
 	unsigned half_duplex:1;		/* E-MAC supports half duplex mode */
 };
 
+struct ravb_rx_buffer {
+	struct page *page;
+	unsigned int offset;
+};
+
 struct ravb_private {
 	struct net_device *ndev;
 	struct platform_device *pdev;
@@ -1094,7 +1100,8 @@ struct ravb_private {
 	struct ravb_tx_desc *tx_ring[NUM_TX_QUEUE];
 	void *tx_align[NUM_TX_QUEUE];
 	struct sk_buff *rx_1st_skb;
-	struct sk_buff **rx_skb[NUM_RX_QUEUE];
+	struct page_pool *rx_pool[NUM_RX_QUEUE];
+	struct ravb_rx_buffer *rx_buffers[NUM_RX_QUEUE];
 	struct sk_buff **tx_skb[NUM_TX_QUEUE];
 	u32 rx_over_errors;
 	u32 rx_fifo_errors;
diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index c89bb38492b4..c1546b916e4e 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -30,6 +30,7 @@
 #include <linux/reset.h>
 #include <linux/math64.h>
 #include <net/ip.h>
+#include <net/page_pool/helpers.h>
 
 #include "ravb.h"
 
@@ -113,25 +114,6 @@ static void ravb_set_rate_rcar(struct net_device *ndev)
 	}
 }
 
-static struct sk_buff *
-ravb_alloc_skb(struct net_device *ndev, const struct ravb_hw_info *info,
-	       gfp_t gfp_mask)
-{
-	struct sk_buff *skb;
-	u32 reserve;
-
-	skb = __netdev_alloc_skb(ndev, info->rx_max_frame_size + RAVB_ALIGN - 1,
-				 gfp_mask);
-	if (!skb)
-		return NULL;
-
-	reserve = (unsigned long)skb->data & (RAVB_ALIGN - 1);
-	if (reserve)
-		skb_reserve(skb, RAVB_ALIGN - reserve);
-
-	return skb;
-}
-
 /* Get MAC address from the MAC address registers
  *
  * Ethernet AVB device doesn't have ROM for MAC address.
@@ -257,21 +239,10 @@ static void ravb_rx_ring_free(struct net_device *ndev, int q)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
 	unsigned int ring_size;
-	unsigned int i;
 
 	if (!priv->rx_ring[q].raw)
 		return;
 
-	for (i = 0; i < priv->num_rx_ring[q]; i++) {
-		struct ravb_rx_desc *desc = ravb_rx_get_desc(priv, q, i);
-
-		if (!dma_mapping_error(ndev->dev.parent,
-				       le32_to_cpu(desc->dptr)))
-			dma_unmap_single(ndev->dev.parent,
-					 le32_to_cpu(desc->dptr),
-					 priv->info->rx_max_frame_size,
-					 DMA_FROM_DEVICE);
-	}
 	ring_size = priv->info->rx_desc_size * (priv->num_rx_ring[q] + 1);
 	dma_free_coherent(ndev->dev.parent, ring_size, priv->rx_ring[q].raw,
 			  priv->rx_desc_dma[q]);
@@ -298,13 +269,16 @@ static void ravb_ring_free(struct net_device *ndev, int q)
 		priv->tx_ring[q] = NULL;
 	}
 
-	/* Free RX skb ringbuffer */
-	if (priv->rx_skb[q]) {
-		for (i = 0; i < priv->num_rx_ring[q]; i++)
-			dev_kfree_skb(priv->rx_skb[q][i]);
+	/* Free RX buffers */
+	for (i = 0; i < priv->num_rx_ring[q]; i++) {
+		if (priv->rx_buffers[q][i].page)
+			page_pool_put_page(priv->rx_pool[q],
+					   priv->rx_buffers[q][i].page,
+					   0, true);
 	}
-	kfree(priv->rx_skb[q]);
-	priv->rx_skb[q] = NULL;
+	kfree(priv->rx_buffers[q]);
+	priv->rx_buffers[q] = NULL;
+	page_pool_destroy(priv->rx_pool[q]);
 
 	/* Free aligned TX buffers */
 	kfree(priv->tx_align[q]);
@@ -317,36 +291,57 @@ static void ravb_ring_free(struct net_device *ndev, int q)
 	priv->tx_skb[q] = NULL;
 }
 
+static int
+ravb_alloc_rx_buffer(struct net_device *ndev, int q, u32 entry, gfp_t gfp_mask,
+		     struct ravb_rx_desc *rx_desc)
+{
+	struct ravb_private *priv = netdev_priv(ndev);
+	const struct ravb_hw_info *info = priv->info;
+	struct ravb_rx_buffer *rx_buff;
+	dma_addr_t dma_addr;
+	unsigned int size;
+
+	rx_buff = &priv->rx_buffers[q][entry];
+	size = info->rx_buffer_size;
+	rx_buff->page = page_pool_alloc(priv->rx_pool[q], &rx_buff->offset,
+					&size, gfp_mask);
+	if (unlikely(!rx_buff->page)) {
+		/* We just set the data size to 0 for a failed mapping which
+		 * should prevent DMA from happening...
+		 */
+		rx_desc->ds_cc = cpu_to_le16(0);
+		return -ENOMEM;
+	}
+
+	dma_addr = page_pool_get_dma_addr(rx_buff->page) + rx_buff->offset;
+	dma_sync_single_for_device(ndev->dev.parent, dma_addr,
+				   info->rx_buffer_size, DMA_FROM_DEVICE);
+	rx_desc->dptr = cpu_to_le32(dma_addr);
+
+	/* The end of the RX buffer is used to store skb shared data, so we need
+	 * to ensure that the hardware leaves enough space for this.
+	 */
+	rx_desc->ds_cc = cpu_to_le16(info->rx_buffer_size -
+				     SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) -
+				     ETH_FCS_LEN + sizeof(__sum16));
+	return 0;
+}
+
 static u32
 ravb_rx_ring_refill(struct net_device *ndev, int q, u32 count, gfp_t gfp_mask)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
-	const struct ravb_hw_info *info = priv->info;
 	struct ravb_rx_desc *rx_desc;
-	dma_addr_t dma_addr;
 	u32 i, entry;
 
 	for (i = 0; i < count; i++) {
 		entry = (priv->dirty_rx[q] + i) % priv->num_rx_ring[q];
 		rx_desc = ravb_rx_get_desc(priv, q, entry);
-		rx_desc->ds_cc = cpu_to_le16(info->rx_max_desc_use);
 
-		if (!priv->rx_skb[q][entry]) {
-			priv->rx_skb[q][entry] = ravb_alloc_skb(ndev, info,
-								gfp_mask);
-			if (!priv->rx_skb[q][entry])
+		if (!priv->rx_buffers[q][entry].page) {
+			if (unlikely(ravb_alloc_rx_buffer(ndev, q, entry,
+							  gfp_mask, rx_desc)))
 				break;
-			dma_addr = dma_map_single(ndev->dev.parent,
-						  priv->rx_skb[q][entry]->data,
-						  priv->info->rx_max_frame_size,
-						  DMA_FROM_DEVICE);
-			skb_checksum_none_assert(priv->rx_skb[q][entry]);
-			/* We just set the data size to 0 for a failed mapping
-			 * which should prevent DMA from happening...
-			 */
-			if (dma_mapping_error(ndev->dev.parent, dma_addr))
-				rx_desc->ds_cc = cpu_to_le16(0);
-			rx_desc->dptr = cpu_to_le32(dma_addr);
 		}
 		/* Descriptor type must be set after all the above writes */
 		dma_wmb();
@@ -424,15 +419,32 @@ static int ravb_ring_init(struct net_device *ndev, int q)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
 	unsigned int num_tx_desc = priv->num_tx_desc;
+	struct page_pool_params params = {
+		.order = 0,
+		.flags = PP_FLAG_DMA_MAP,
+		.pool_size = priv->num_rx_ring[q],
+		.nid = NUMA_NO_NODE,
+		.dev = ndev->dev.parent,
+		.dma_dir = DMA_FROM_DEVICE,
+	};
 	unsigned int ring_size;
 	u32 num_filled;
 
-	/* Allocate RX and TX skb rings */
-	priv->rx_skb[q] = kcalloc(priv->num_rx_ring[q],
-				  sizeof(*priv->rx_skb[q]), GFP_KERNEL);
+	/* Allocate RX page pool and buffers */
+	priv->rx_pool[q] = page_pool_create(&params);
+	if (IS_ERR(priv->rx_pool[q]))
+		goto error;
+
+	/* Allocate RX buffers */
+	priv->rx_buffers[q] = kcalloc(priv->num_rx_ring[q],
+				      sizeof(*priv->rx_buffers[q]), GFP_KERNEL);
+	if (!priv->rx_buffers[q])
+		goto error;
+
+	/* Allocate TX skb rings */
 	priv->tx_skb[q] = kcalloc(priv->num_tx_ring[q],
 				  sizeof(*priv->tx_skb[q]), GFP_KERNEL);
-	if (!priv->rx_skb[q] || !priv->tx_skb[q])
+	if (!priv->tx_skb[q])
 		goto error;
 
 	/* Allocate all RX descriptors. */
@@ -719,7 +731,9 @@ static void ravb_get_tx_tstamp(struct net_device *ndev)
 
 static void ravb_rx_csum_gbeth(struct sk_buff *skb)
 {
+	struct skb_shared_info *shinfo = skb_shinfo(skb);
 	__wsum csum_ip_hdr, csum_proto;
+	skb_frag_t *last_frag;
 	u8 *hw_csum;
 
 	/* The hardware checksum status is contained in sizeof(__sum16) * 2 = 4
@@ -729,12 +743,24 @@ static void ravb_rx_csum_gbeth(struct sk_buff *skb)
 	if (unlikely(skb->len < sizeof(__sum16) * 2))
 		return;
 
-	hw_csum = skb_tail_pointer(skb) - sizeof(__sum16);
+	if (skb_is_nonlinear(skb)) {
+		last_frag = &shinfo->frags[shinfo->nr_frags - 1];
+		hw_csum = skb_frag_address(last_frag) +
+			  skb_frag_size(last_frag);
+	} else {
+		hw_csum = skb_tail_pointer(skb);
+	}
+
+	hw_csum -= sizeof(__sum16);
 	csum_proto = csum_unfold((__force __sum16)get_unaligned_le16(hw_csum));
 
 	hw_csum -= sizeof(__sum16);
 	csum_ip_hdr = csum_unfold((__force __sum16)get_unaligned_le16(hw_csum));
-	skb_trim(skb, skb->len - 2 * sizeof(__sum16));
+
+	if (skb_is_nonlinear(skb))
+		skb_frag_size_sub(last_frag, 2 * sizeof(__sum16));
+	else
+		skb_trim(skb, skb->len - 2 * sizeof(__sum16));
 
 	/* TODO: IPV6 Rx checksum */
 	if (skb->protocol == htons(ETH_P_IP) && !csum_ip_hdr && !csum_proto)
@@ -756,25 +782,11 @@ static void ravb_rx_csum(struct sk_buff *skb)
 	skb_trim(skb, skb->len - sizeof(__sum16));
 }
 
-static struct sk_buff *ravb_get_skb_gbeth(struct net_device *ndev, int entry,
-					  struct ravb_rx_desc *desc)
-{
-	struct ravb_private *priv = netdev_priv(ndev);
-	struct sk_buff *skb;
-
-	skb = priv->rx_skb[RAVB_BE][entry];
-	priv->rx_skb[RAVB_BE][entry] = NULL;
-	dma_unmap_single(ndev->dev.parent, le32_to_cpu(desc->dptr),
-			 ALIGN(priv->info->rx_max_frame_size, 16),
-			 DMA_FROM_DEVICE);
-
-	return skb;
-}
-
 /* Packet receive function for Gigabit Ethernet */
 static int ravb_rx_gbeth(struct net_device *ndev, int budget, int q)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
+	const struct ravb_hw_info *info = priv->info;
 	struct net_device_stats *stats;
 	struct ravb_rx_desc *desc;
 	struct sk_buff *skb;
@@ -818,12 +830,30 @@ static int ravb_rx_gbeth(struct net_device *ndev, int budget, int q)
 			if (desc_status & MSC_CEEF)
 				stats->rx_missed_errors++;
 		} else {
+			struct ravb_rx_buffer *rx_buff;
+			void *rx_addr;
+
+			rx_buff = &priv->rx_buffers[q][entry];
+			rx_addr = page_address(rx_buff->page) + rx_buff->offset;
 			die_dt = desc->die_dt & 0xF0;
-			skb = ravb_get_skb_gbeth(ndev, entry, desc);
+			dma_sync_single_for_cpu(ndev->dev.parent,
+						le32_to_cpu(desc->dptr),
+						desc_len, DMA_FROM_DEVICE);
+
 			switch (die_dt) {
 			case DT_FSINGLE:
 			case DT_FSTART:
 				/* Start of packet: Set initial data length. */
+				skb = napi_build_skb(rx_addr,
+						     info->rx_buffer_size);
+				if (unlikely(!skb)) {
+					stats->rx_errors++;
+					page_pool_put_page(priv->rx_pool[q],
+							   rx_buff->page, 0,
+							   true);
+					goto refill;
+				}
+				skb_mark_for_recycle(skb);
 				skb_put(skb, desc_len);
 
 				/* Save this skb if the packet spans multiple
@@ -835,15 +865,30 @@ static int ravb_rx_gbeth(struct net_device *ndev, int budget, int q)
 
 			case DT_FMID:
 			case DT_FEND:
-				/* Continuing a packet: Move data into the saved
-				 * skb.
+				/* Continuing a packet: Add this buffer as an RX
+				 * frag.
+				 */
+
+				/* rx_1st_skb will be NULL if napi_build_skb()
+				 * failed for the first descriptor of a
+				 * multi-descriptor packet.
 				 */
-				skb_copy_to_linear_data_offset(priv->rx_1st_skb,
-							       priv->rx_1st_skb->len,
-							       skb->data,
-							       desc_len);
-				skb_put(priv->rx_1st_skb, desc_len);
-				dev_kfree_skb(skb);
+				if (unlikely(!priv->rx_1st_skb)) {
+					stats->rx_errors++;
+					page_pool_put_page(priv->rx_pool[q],
+							   rx_buff->page, 0,
+							   true);
+
+					/* We may find a DT_FSINGLE or DT_FSTART
+					 * descriptor in the queue which we can
+					 * process, so don't give up yet.
+					 */
+					continue;
+				}
+				skb_add_rx_frag(priv->rx_1st_skb,
+						skb_shinfo(priv->rx_1st_skb)->nr_frags,
+						rx_buff->page, rx_buff->offset,
+						desc_len, info->rx_buffer_size);
 
 				/* Set skb to point at the whole packet so that
 				 * we only need one code path for finishing a
@@ -865,10 +910,19 @@ static int ravb_rx_gbeth(struct net_device *ndev, int budget, int q)
 				stats->rx_bytes += skb->len;
 				napi_gro_receive(&priv->napi[q], skb);
 				rx_packets++;
+
+				/* Clear rx_1st_skb so that it will only be
+				 * non-NULL when valid.
+				 */
+				priv->rx_1st_skb = NULL;
 			}
+
+			/* Mark this RX buffer as consumed. */
+			rx_buff->page = NULL;
 		}
 	}
 
+refill:
 	/* Refill the RX ring buffers. */
 	priv->dirty_rx[q] += ravb_rx_ring_refill(ndev, q,
 						 priv->cur_rx[q] - priv->dirty_rx[q],
@@ -882,6 +936,7 @@ static int ravb_rx_gbeth(struct net_device *ndev, int budget, int q)
 static int ravb_rx_rcar(struct net_device *ndev, int budget, int q)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
+	const struct ravb_hw_info *info = priv->info;
 	struct net_device_stats *stats = &priv->stats[q];
 	struct ravb_ex_rx_desc *desc;
 	unsigned int limit, i;
@@ -924,12 +979,23 @@ static int ravb_rx_rcar(struct net_device *ndev, int budget, int q)
 				stats->rx_missed_errors++;
 		} else {
 			u32 get_ts = priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE;
-
-			skb = priv->rx_skb[q][entry];
-			priv->rx_skb[q][entry] = NULL;
-			dma_unmap_single(ndev->dev.parent, le32_to_cpu(desc->dptr),
-					 priv->info->rx_max_frame_size,
-					 DMA_FROM_DEVICE);
+			struct ravb_rx_buffer *rx_buff;
+			void *rx_addr;
+
+			rx_buff = &priv->rx_buffers[q][entry];
+			rx_addr = page_address(rx_buff->page) + rx_buff->offset;
+			dma_sync_single_for_cpu(ndev->dev.parent,
+						le32_to_cpu(desc->dptr),
+						pkt_len, DMA_FROM_DEVICE);
+
+			skb = napi_build_skb(rx_addr, info->rx_buffer_size);
+			if (unlikely(!skb)) {
+				stats->rx_errors++;
+				page_pool_put_page(priv->rx_pool[q],
+						   rx_buff->page, 0, true);
+				break;
+			}
+			skb_mark_for_recycle(skb);
 			get_ts &= (q == RAVB_NC) ?
 					RAVB_RXTSTAMP_TYPE_V2_L2_EVENT :
 					~RAVB_RXTSTAMP_TYPE_V2_L2_EVENT;
@@ -951,6 +1017,9 @@ static int ravb_rx_rcar(struct net_device *ndev, int budget, int q)
 			napi_gro_receive(&priv->napi[q], skb);
 			rx_packets++;
 			stats->rx_bytes += pkt_len;
+
+			/* Mark this RX buffer as consumed. */
+			rx_buff->page = NULL;
 		}
 	}
 
@@ -2596,7 +2665,8 @@ static const struct ravb_hw_info ravb_gen3_hw_info = {
 	.stats_len = ARRAY_SIZE(ravb_gstrings_stats),
 	.tccr_mask = TCCR_TSRQ0 | TCCR_TSRQ1 | TCCR_TSRQ2 | TCCR_TSRQ3,
 	.rx_max_frame_size = SZ_2K,
-	.rx_max_desc_use = SZ_2K - ETH_FCS_LEN + sizeof(__sum16),
+	.rx_buffer_size = SZ_2K +
+			  SKB_DATA_ALIGN(sizeof(struct skb_shared_info)),
 	.rx_desc_size = sizeof(struct ravb_ex_rx_desc),
 	.internal_delay = 1,
 	.tx_counters = 1,
@@ -2620,7 +2690,8 @@ static const struct ravb_hw_info ravb_gen2_hw_info = {
 	.stats_len = ARRAY_SIZE(ravb_gstrings_stats),
 	.tccr_mask = TCCR_TSRQ0 | TCCR_TSRQ1 | TCCR_TSRQ2 | TCCR_TSRQ3,
 	.rx_max_frame_size = SZ_2K,
-	.rx_max_desc_use = SZ_2K - ETH_FCS_LEN + sizeof(__sum16),
+	.rx_buffer_size = SZ_2K +
+			  SKB_DATA_ALIGN(sizeof(struct skb_shared_info)),
 	.rx_desc_size = sizeof(struct ravb_ex_rx_desc),
 	.aligned_tx = 1,
 	.gptp = 1,
@@ -2641,7 +2712,8 @@ static const struct ravb_hw_info ravb_rzv2m_hw_info = {
 	.stats_len = ARRAY_SIZE(ravb_gstrings_stats),
 	.tccr_mask = TCCR_TSRQ0 | TCCR_TSRQ1 | TCCR_TSRQ2 | TCCR_TSRQ3,
 	.rx_max_frame_size = SZ_2K,
-	.rx_max_desc_use = SZ_2K - ETH_FCS_LEN + sizeof(__sum16),
+	.rx_buffer_size = SZ_2K +
+			  SKB_DATA_ALIGN(sizeof(struct skb_shared_info)),
 	.rx_desc_size = sizeof(struct ravb_ex_rx_desc),
 	.multi_irqs = 1,
 	.err_mgmt_irqs = 1,
@@ -2664,7 +2736,7 @@ static const struct ravb_hw_info gbeth_hw_info = {
 	.stats_len = ARRAY_SIZE(ravb_gstrings_stats_gbeth),
 	.tccr_mask = TCCR_TSRQ0,
 	.rx_max_frame_size = SZ_8K,
-	.rx_max_desc_use = 4080,
+	.rx_buffer_size = SZ_2K,
 	.rx_desc_size = sizeof(struct ravb_rx_desc),
 	.aligned_tx = 1,
 	.coalesce_irqs = 1,
-- 
cgit v1.2.3-58-ga151


From b4cb4a1391dcdc640c4ade003aaf0ee19cc8d509 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 4 Jun 2024 11:16:03 +0000
Subject: net: use unrcu_pointer() helper
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Toke mentioned unrcu_pointer() existence, allowing
to remove some of the ugly casts we have when using
xchg() for rcu protected pointers.

Also make inet_rcv_compat const.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Toke Høiland-Jørgensen <toke@redhat.com>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/r/20240604111603.45871-1-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/sock.h       | 2 +-
 net/core/gen_estimator.c | 2 +-
 net/core/sock_diag.c     | 8 +++-----
 net/ipv4/cipso_ipv4.c    | 2 +-
 net/ipv4/tcp.c           | 2 +-
 net/ipv4/tcp_fastopen.c  | 7 ++++---
 net/ipv4/udp.c           | 2 +-
 net/ipv6/af_inet6.c      | 2 +-
 net/ipv6/ip6_fib.c       | 2 +-
 net/ipv6/ipv6_sockglue.c | 3 +--
 net/ipv6/route.c         | 6 +++---
 net/sched/act_api.c      | 2 +-
 12 files changed, 19 insertions(+), 21 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index 953c8dc4e259..b30ea0c342a6 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -2095,7 +2095,7 @@ sk_dst_set(struct sock *sk, struct dst_entry *dst)
 
 	sk_tx_queue_clear(sk);
 	WRITE_ONCE(sk->sk_dst_pending_confirm, 0);
-	old_dst = xchg((__force struct dst_entry **)&sk->sk_dst_cache, dst);
+	old_dst = unrcu_pointer(xchg(&sk->sk_dst_cache, RCU_INITIALIZER(dst)));
 	dst_release(old_dst);
 }
 
diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c
index fae9c4694186..412816076b8b 100644
--- a/net/core/gen_estimator.c
+++ b/net/core/gen_estimator.c
@@ -206,7 +206,7 @@ void gen_kill_estimator(struct net_rate_estimator __rcu **rate_est)
 {
 	struct net_rate_estimator *est;
 
-	est = xchg((__force struct net_rate_estimator **)rate_est, NULL);
+	est = unrcu_pointer(xchg(rate_est, NULL));
 	if (est) {
 		timer_shutdown_sync(&est->timer);
 		kfree_rcu(est, rcu);
diff --git a/net/core/sock_diag.c b/net/core/sock_diag.c
index 654122838025..a08eed9b9142 100644
--- a/net/core/sock_diag.c
+++ b/net/core/sock_diag.c
@@ -18,7 +18,7 @@
 
 static const struct sock_diag_handler __rcu *sock_diag_handlers[AF_MAX];
 
-static struct sock_diag_inet_compat __rcu *inet_rcv_compat;
+static const struct sock_diag_inet_compat __rcu *inet_rcv_compat;
 
 static struct workqueue_struct *broadcast_wq;
 
@@ -187,8 +187,7 @@ void sock_diag_broadcast_destroy(struct sock *sk)
 
 void sock_diag_register_inet_compat(const struct sock_diag_inet_compat *ptr)
 {
-	xchg((__force const struct sock_diag_inet_compat **)&inet_rcv_compat,
-	     ptr);
+	xchg(&inet_rcv_compat, RCU_INITIALIZER(ptr));
 }
 EXPORT_SYMBOL_GPL(sock_diag_register_inet_compat);
 
@@ -196,8 +195,7 @@ void sock_diag_unregister_inet_compat(const struct sock_diag_inet_compat *ptr)
 {
 	const struct sock_diag_inet_compat *old;
 
-	old = xchg((__force const struct sock_diag_inet_compat **)&inet_rcv_compat,
-		   NULL);
+	old = unrcu_pointer(xchg(&inet_rcv_compat, NULL));
 	WARN_ON_ONCE(old != ptr);
 }
 EXPORT_SYMBOL_GPL(sock_diag_unregister_inet_compat);
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index dd6d46015058..3a95c0f13ce3 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -1953,7 +1953,7 @@ int cipso_v4_req_setattr(struct request_sock *req,
 	buf = NULL;
 
 	req_inet = inet_rsk(req);
-	opt = xchg((__force struct ip_options_rcu **)&req_inet->ireq_opt, opt);
+	opt = unrcu_pointer(xchg(&req_inet->ireq_opt, RCU_INITIALIZER(opt)));
 	if (opt)
 		kfree_rcu(opt, rcu);
 
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index fa43aaacd92b..f727bc8d82a6 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3081,7 +3081,7 @@ int tcp_disconnect(struct sock *sk, int flags)
 	icsk->icsk_ack.rcv_mss = TCP_MIN_MSS;
 	memset(&tp->rx_opt, 0, sizeof(tp->rx_opt));
 	__sk_dst_reset(sk);
-	dst_release(xchg((__force struct dst_entry **)&sk->sk_rx_dst, NULL));
+	dst_release(unrcu_pointer(xchg(&sk->sk_rx_dst, NULL)));
 	tcp_saved_syn_free(tp);
 	tp->compressed_ack = 0;
 	tp->segs_in = 0;
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c
index 8ed54e7334a9..0f523cbfe329 100644
--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -49,7 +49,7 @@ void tcp_fastopen_ctx_destroy(struct net *net)
 {
 	struct tcp_fastopen_context *ctxt;
 
-	ctxt = xchg((__force struct tcp_fastopen_context **)&net->ipv4.tcp_fastopen_ctx, NULL);
+	ctxt = unrcu_pointer(xchg(&net->ipv4.tcp_fastopen_ctx, NULL));
 
 	if (ctxt)
 		call_rcu(&ctxt->rcu, tcp_fastopen_ctx_free);
@@ -80,9 +80,10 @@ int tcp_fastopen_reset_cipher(struct net *net, struct sock *sk,
 
 	if (sk) {
 		q = &inet_csk(sk)->icsk_accept_queue.fastopenq;
-		octx = xchg((__force struct tcp_fastopen_context **)&q->ctx, ctx);
+		octx = unrcu_pointer(xchg(&q->ctx, RCU_INITIALIZER(ctx)));
 	} else {
-		octx = xchg((__force struct tcp_fastopen_context **)&net->ipv4.tcp_fastopen_ctx, ctx);
+		octx = unrcu_pointer(xchg(&net->ipv4.tcp_fastopen_ctx,
+					  RCU_INITIALIZER(ctx)));
 	}
 
 	if (octx)
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 189c9113fe9a..c9ca6d285347 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -2230,7 +2230,7 @@ bool udp_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst)
 	struct dst_entry *old;
 
 	if (dst_hold_safe(dst)) {
-		old = xchg((__force struct dst_entry **)&sk->sk_rx_dst, dst);
+		old = unrcu_pointer(xchg(&sk->sk_rx_dst, RCU_INITIALIZER(dst)));
 		dst_release(old);
 		return old != dst;
 	}
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 8041dc181bd4..e03fb9a1dbeb 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -509,7 +509,7 @@ void inet6_cleanup_sock(struct sock *sk)
 
 	/* Free tx options */
 
-	opt = xchg((__force struct ipv6_txoptions **)&np->opt, NULL);
+	opt = unrcu_pointer(xchg(&np->opt, NULL));
 	if (opt) {
 		atomic_sub(opt->tot_len, &sk->sk_omem_alloc);
 		txopt_put(opt);
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 31d77885bcae..15f9abe50656 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -984,7 +984,7 @@ static void __fib6_drop_pcpu_from(struct fib6_nh *fib6_nh,
 		if (pcpu_rt && rcu_access_pointer(pcpu_rt->from) == match) {
 			struct fib6_info *from;
 
-			from = xchg((__force struct fib6_info **)&pcpu_rt->from, NULL);
+			from = unrcu_pointer(xchg(&pcpu_rt->from, NULL));
 			fib6_info_release(from);
 		}
 	}
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index d4c28ec1bc51..cd342d5015c6 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -111,8 +111,7 @@ struct ipv6_txoptions *ipv6_update_options(struct sock *sk,
 			icsk->icsk_sync_mss(sk, icsk->icsk_pmtu_cookie);
 		}
 	}
-	opt = xchg((__force struct ipv6_txoptions **)&inet6_sk(sk)->opt,
-		   opt);
+	opt = unrcu_pointer(xchg(&inet6_sk(sk)->opt, RCU_INITIALIZER(opt)));
 	sk_dst_reset(sk);
 
 	return opt;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 12f647d0fec0..ae6fbdf8995d 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -368,7 +368,7 @@ static void ip6_dst_destroy(struct dst_entry *dst)
 		in6_dev_put(idev);
 	}
 
-	from = xchg((__force struct fib6_info **)&rt->from, NULL);
+	from = unrcu_pointer(xchg(&rt->from, NULL));
 	fib6_info_release(from);
 }
 
@@ -1437,7 +1437,7 @@ static struct rt6_info *rt6_make_pcpu_route(struct net *net,
 	if (res->f6i->fib6_destroying) {
 		struct fib6_info *from;
 
-		from = xchg((__force struct fib6_info **)&pcpu_rt->from, NULL);
+		from = unrcu_pointer(xchg(&pcpu_rt->from, NULL));
 		fib6_info_release(from);
 	}
 
@@ -1466,7 +1466,7 @@ static void rt6_remove_exception(struct rt6_exception_bucket *bucket,
 	/* purge completely the exception to allow releasing the held resources:
 	 * some [sk] cache may keep the dst around for unlimited time
 	 */
-	from = xchg((__force struct fib6_info **)&rt6_ex->rt6i->from, NULL);
+	from = unrcu_pointer(xchg(&rt6_ex->rt6i->from, NULL));
 	fib6_info_release(from);
 	dst_dev_put(&rt6_ex->rt6i->dst);
 
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 9ee622fb1160..7458b3154426 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -62,7 +62,7 @@ static void tcf_set_action_cookie(struct tc_cookie __rcu **old_cookie,
 {
 	struct tc_cookie *old;
 
-	old = xchg((__force struct tc_cookie **)old_cookie, new_cookie);
+	old = unrcu_pointer(xchg(old_cookie, RCU_INITIALIZER(new_cookie)));
 	if (old)
 		call_rcu(&old->rcu, tcf_free_cookie_rcu);
 }
-- 
cgit v1.2.3-58-ga151


From 98aa546af5e4f180439c82da403cc77926d3953b Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 4 Jun 2024 16:51:50 +0000
Subject: inet: remove (struct uncached_list)->quarantine

This list is used to tranfert dst that are handled by
rt_flush_dev() and rt6_uncached_list_flush_dev() out
of the per-cpu lists.

But quarantine list is not used later.

If we simply use list_del_init(&rt->dst.rt_uncached),
this also removes the dst from per-cpu list.

This patch also makes the future calls to rt_del_uncached_list()
and rt6_uncached_list_del() faster, because no spinlock
acquisition is needed anymore.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240604165150.726382-1-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/ipv4/route.c | 4 +---
 net/ipv6/route.c | 5 +----
 2 files changed, 2 insertions(+), 7 deletions(-)

diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index b3073d1c8f8f..cb0bdf34ed50 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1481,7 +1481,6 @@ static bool rt_cache_route(struct fib_nh_common *nhc, struct rtable *rt)
 struct uncached_list {
 	spinlock_t		lock;
 	struct list_head	head;
-	struct list_head	quarantine;
 };
 
 static DEFINE_PER_CPU_ALIGNED(struct uncached_list, rt_uncached_list);
@@ -1532,7 +1531,7 @@ void rt_flush_dev(struct net_device *dev)
 			rt->dst.dev = blackhole_netdev;
 			netdev_ref_replace(dev, blackhole_netdev,
 					   &rt->dst.dev_tracker, GFP_ATOMIC);
-			list_move(&rt->dst.rt_uncached, &ul->quarantine);
+			list_del_init(&rt->dst.rt_uncached);
 		}
 		spin_unlock_bh(&ul->lock);
 	}
@@ -3661,7 +3660,6 @@ int __init ip_rt_init(void)
 		struct uncached_list *ul = &per_cpu(rt_uncached_list, cpu);
 
 		INIT_LIST_HEAD(&ul->head);
-		INIT_LIST_HEAD(&ul->quarantine);
 		spin_lock_init(&ul->lock);
 	}
 #ifdef CONFIG_IP_ROUTE_CLASSID
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index ae6fbdf8995d..ec5b59b221ef 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -131,7 +131,6 @@ static struct fib6_info *rt6_get_route_info(struct net *net,
 struct uncached_list {
 	spinlock_t		lock;
 	struct list_head	head;
-	struct list_head	quarantine;
 };
 
 static DEFINE_PER_CPU_ALIGNED(struct uncached_list, rt6_uncached_list);
@@ -189,8 +188,7 @@ static void rt6_uncached_list_flush_dev(struct net_device *dev)
 				handled = true;
 			}
 			if (handled)
-				list_move(&rt->dst.rt_uncached,
-					  &ul->quarantine);
+				list_del_init(&rt->dst.rt_uncached);
 		}
 		spin_unlock_bh(&ul->lock);
 	}
@@ -6755,7 +6753,6 @@ int __init ip6_route_init(void)
 		struct uncached_list *ul = per_cpu_ptr(&rt6_uncached_list, cpu);
 
 		INIT_LIST_HEAD(&ul->head);
-		INIT_LIST_HEAD(&ul->quarantine);
 		spin_lock_init(&ul->lock);
 	}
 
-- 
cgit v1.2.3-58-ga151


From 9b6a30febddf4857cb4c367f5294bb6118340e30 Mon Sep 17 00:00:00 2001
From: Jason Xing <kernelxing@tencent.com>
Date: Wed, 5 Jun 2024 10:29:32 +0800
Subject: net: allow rps/rfs related configs to be switched

After John Sperbeck reported a compile error if the CONFIG_RFS_ACCEL
is off, I found that I cannot easily enable/disable the config
because of lack of the prompt when using 'make menuconfig'. Therefore,
I decided to change rps/rfc related configs altogether.

Signed-off-by: Jason Xing <kernelxing@tencent.com>
Link: https://lore.kernel.org/r/20240605022932.33703-1-kerneljasonxing@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/Kconfig | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/net/Kconfig b/net/Kconfig
index f0a8692496ff..9fe65fa26e48 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -290,15 +290,21 @@ config MAX_SKB_FRAGS
 	  If unsure, say 17.
 
 config RPS
-	bool
+	bool "Receive packet steering"
 	depends on SMP && SYSFS
 	default y
+	help
+	  Software receive side packet steering (RPS) distributes the
+	  load of received packet processing across multiple CPUs.
 
 config RFS_ACCEL
-	bool
+	bool "Hardware acceleration of RFS"
 	depends on RPS
 	select CPU_RMAP
 	default y
+	help
+	  Allowing drivers for multiqueue hardware with flow filter tables to
+	  accelerate RFS.
 
 config SOCK_RX_QUEUE_MAPPING
 	bool
@@ -351,7 +357,7 @@ config BPF_STREAM_PARSER
 	  BPF_MAP_TYPE_SOCKMAP.
 
 config NET_FLOW_LIMIT
-	bool
+	bool "Net flow limit"
 	depends on RPS
 	default y
 	help
-- 
cgit v1.2.3-58-ga151


From 5f0d0649c83f72399c19b18591ea1413ca94c015 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Wed, 5 Jun 2024 09:15:40 +0200
Subject: mptcp: use mptcp_win_from_space helper

The MPTCP dedicated win_from_space helper mptcp_win_from_space() is defined
in protocol.h, use it in mptcp_rcv_space_adjust() instead of using the TCP
one. Here scaling_ratio is the same as msk->scaling_ratio.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/protocol.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 7d44196ec5b6..546c80c6702a 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -2046,7 +2046,7 @@ static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied)
 		if (rcvbuf > sk->sk_rcvbuf) {
 			u32 window_clamp;
 
-			window_clamp = __tcp_win_from_space(scaling_ratio, rcvbuf);
+			window_clamp = mptcp_win_from_space(sk, rcvbuf);
 			WRITE_ONCE(sk->sk_rcvbuf, rcvbuf);
 
 			/* Make subflows follow along.  If we do not do this, we
-- 
cgit v1.2.3-58-ga151


From 5cdedad62eaba22c45b9c45c8199bacd461afd87 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Wed, 5 Jun 2024 09:15:41 +0200
Subject: mptcp: add mptcp_space_from_win helper

As a wrapper of __tcp_space_from_win(), this patch adds a MPTCP dedicated
space_from_win helper mptcp_space_from_win() in protocol.h to paired with
mptcp_win_from_space().

Use it instead of __tcp_space_from_win() in both mptcp_rcv_space_adjust()
and mptcp_set_rcvlowat().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/protocol.c | 2 +-
 net/mptcp/protocol.h | 5 +++++
 net/mptcp/sockopt.c  | 2 +-
 3 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 546c80c6702a..7ce11bee3b79 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -2040,7 +2040,7 @@ static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied)
 		do_div(grow, msk->rcvq_space.space);
 		rcvwin += (grow << 1);
 
-		rcvbuf = min_t(u64, __tcp_space_from_win(scaling_ratio, rcvwin),
+		rcvbuf = min_t(u64, mptcp_space_from_win(sk, rcvwin),
 			       READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2]));
 
 		if (rcvbuf > sk->sk_rcvbuf) {
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 7aa47e2dd52b..b11a4e50d52b 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -386,6 +386,11 @@ static inline int mptcp_win_from_space(const struct sock *sk, int space)
 	return __tcp_win_from_space(mptcp_sk(sk)->scaling_ratio, space);
 }
 
+static inline int mptcp_space_from_win(const struct sock *sk, int win)
+{
+	return __tcp_space_from_win(mptcp_sk(sk)->scaling_ratio, win);
+}
+
 static inline int __mptcp_space(const struct sock *sk)
 {
 	return mptcp_win_from_space(sk, READ_ONCE(sk->sk_rcvbuf) - __mptcp_rmem(sk));
diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c
index f9a4fb17b5b7..2026a9a36f80 100644
--- a/net/mptcp/sockopt.c
+++ b/net/mptcp/sockopt.c
@@ -1579,7 +1579,7 @@ int mptcp_set_rcvlowat(struct sock *sk, int val)
 	if (sk->sk_userlocks & SOCK_RCVBUF_LOCK)
 		return 0;
 
-	space = __tcp_space_from_win(mptcp_sk(sk)->scaling_ratio, val);
+	space = mptcp_space_from_win(sk, val);
 	if (space <= sk->sk_rcvbuf)
 		return 0;
 
-- 
cgit v1.2.3-58-ga151


From 92f74c1e05b044b51398d6d4a85e659e4384f2cb Mon Sep 17 00:00:00 2001
From: Davide Caratti <dcaratti@redhat.com>
Date: Wed, 5 Jun 2024 09:15:42 +0200
Subject: mptcp: refer to 'MPTCP' socket in comments

We used to call it 'master' socket at the early stages of MPTCP
development, but the correct wording is 'MPTCP' socket opposed to 'TCP
subflows': convert the last 3 comments to use a more appropriate term.

Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/mptcp/protocol.c | 4 ++--
 net/mptcp/subflow.c  | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 7ce11bee3b79..ead0bf63cf95 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -2202,7 +2202,7 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 		if (skb_queue_empty(&msk->receive_queue) && __mptcp_move_skbs(msk))
 			continue;
 
-		/* only the master socket status is relevant here. The exit
+		/* only the MPTCP socket status is relevant here. The exit
 		 * conditions mirror closely tcp_recvmsg()
 		 */
 		if (copied >= target)
@@ -3521,7 +3521,7 @@ void mptcp_subflow_process_delegated(struct sock *ssk, long status)
 static int mptcp_hash(struct sock *sk)
 {
 	/* should never be called,
-	 * we hash the TCP subflows not the master socket
+	 * we hash the TCP subflows not the MPTCP socket
 	 */
 	WARN_ON_ONCE(1);
 	return 0;
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 612c38570a64..39e2cbdf3801 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -1719,7 +1719,7 @@ int mptcp_subflow_create_socket(struct sock *sk, unsigned short family,
 	mptcp_sockopt_sync_locked(mptcp_sk(sk), sf->sk);
 	release_sock(sf->sk);
 
-	/* the newly created socket really belongs to the owning MPTCP master
+	/* the newly created socket really belongs to the owning MPTCP
 	 * socket, even if for additional subflows the allocation is performed
 	 * by a kernel workqueue. Adjust inode references, so that the
 	 * procfs/diag interfaces really show this one belonging to the correct
-- 
cgit v1.2.3-58-ga151


From c34506406dd5cfb352f8c53bb6a1b9535c0905dd Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 5 Jun 2024 07:15:51 +0000
Subject: tcp: small changes in reqsk_put() and reqsk_free()

In reqsk_free(), use DEBUG_NET_WARN_ON_ONCE()
instead of WARN_ON_ONCE() for a condition which never fired.

In reqsk_put() directly call __reqsk_free(), there is no
point checking rsk_refcnt again right after a transition to zero.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/request_sock.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/net/request_sock.h b/include/net/request_sock.h
index ebcb8896bffc..a8f82216c628 100644
--- a/include/net/request_sock.h
+++ b/include/net/request_sock.h
@@ -172,14 +172,14 @@ static inline void __reqsk_free(struct request_sock *req)
 
 static inline void reqsk_free(struct request_sock *req)
 {
-	WARN_ON_ONCE(refcount_read(&req->rsk_refcnt) != 0);
+	DEBUG_NET_WARN_ON_ONCE(refcount_read(&req->rsk_refcnt) != 0);
 	__reqsk_free(req);
 }
 
 static inline void reqsk_put(struct request_sock *req)
 {
 	if (refcount_dec_and_test(&req->rsk_refcnt))
-		reqsk_free(req);
+		__reqsk_free(req);
 }
 
 /*
-- 
cgit v1.2.3-58-ga151


From adbe695a9765fb704d2ac0d3e284f28bcc8b5bf3 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 5 Jun 2024 07:15:52 +0000
Subject: tcp: move inet_reqsk_alloc() close to inet_reqsk_clone()

inet_reqsk_alloc() does not belong to tcp_input.c,
move it to inet_connection_sock.c instead.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/ipv4/inet_connection_sock.c | 25 +++++++++++++++++++++++++
 net/ipv4/tcp_input.c            | 25 -------------------------
 2 files changed, 25 insertions(+), 25 deletions(-)

diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index d81f74ce0f02..a9d2e6308910 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -911,6 +911,31 @@ int inet_rtx_syn_ack(const struct sock *parent, struct request_sock *req)
 }
 EXPORT_SYMBOL(inet_rtx_syn_ack);
 
+struct request_sock *inet_reqsk_alloc(const struct request_sock_ops *ops,
+				      struct sock *sk_listener,
+				      bool attach_listener)
+{
+	struct request_sock *req = reqsk_alloc(ops, sk_listener,
+					       attach_listener);
+
+	if (req) {
+		struct inet_request_sock *ireq = inet_rsk(req);
+
+		ireq->ireq_opt = NULL;
+#if IS_ENABLED(CONFIG_IPV6)
+		ireq->pktopts = NULL;
+#endif
+		atomic64_set(&ireq->ir_cookie, 0);
+		ireq->ireq_state = TCP_NEW_SYN_RECV;
+		write_pnet(&ireq->ireq_net, sock_net(sk_listener));
+		ireq->ireq_family = sk_listener->sk_family;
+		req->timeout = TCP_TIMEOUT_INIT;
+	}
+
+	return req;
+}
+EXPORT_SYMBOL(inet_reqsk_alloc);
+
 static struct request_sock *inet_reqsk_clone(struct request_sock *req,
 					     struct sock *sk)
 {
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 212b6fd0caf7..eb187450e4d7 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -6986,31 +6986,6 @@ static void tcp_openreq_init(struct request_sock *req,
 #endif
 }
 
-struct request_sock *inet_reqsk_alloc(const struct request_sock_ops *ops,
-				      struct sock *sk_listener,
-				      bool attach_listener)
-{
-	struct request_sock *req = reqsk_alloc(ops, sk_listener,
-					       attach_listener);
-
-	if (req) {
-		struct inet_request_sock *ireq = inet_rsk(req);
-
-		ireq->ireq_opt = NULL;
-#if IS_ENABLED(CONFIG_IPV6)
-		ireq->pktopts = NULL;
-#endif
-		atomic64_set(&ireq->ir_cookie, 0);
-		ireq->ireq_state = TCP_NEW_SYN_RECV;
-		write_pnet(&ireq->ireq_net, sock_net(sk_listener));
-		ireq->ireq_family = sk_listener->sk_family;
-		req->timeout = TCP_TIMEOUT_INIT;
-	}
-
-	return req;
-}
-EXPORT_SYMBOL(inet_reqsk_alloc);
-
 /*
  * Return true if a syncookie should be sent
  */
-- 
cgit v1.2.3-58-ga151


From 6971d21672827a701c5ea180891b7ea6cf06f6a7 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 5 Jun 2024 07:15:53 +0000
Subject: tcp: move reqsk_alloc() to inet_connection_sock.c

reqsk_alloc() has a single caller, no need to expose it
in include/net/request_sock.h.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/request_sock.h      | 33 ---------------------------------
 net/ipv4/inet_connection_sock.c | 33 +++++++++++++++++++++++++++++++++
 2 files changed, 33 insertions(+), 33 deletions(-)

diff --git a/include/net/request_sock.h b/include/net/request_sock.h
index a8f82216c628..b07b1cd14e9f 100644
--- a/include/net/request_sock.h
+++ b/include/net/request_sock.h
@@ -128,39 +128,6 @@ static inline struct sock *skb_steal_sock(struct sk_buff *skb,
 	return sk;
 }
 
-static inline struct request_sock *
-reqsk_alloc_noprof(const struct request_sock_ops *ops, struct sock *sk_listener,
-	    bool attach_listener)
-{
-	struct request_sock *req;
-
-	req = kmem_cache_alloc_noprof(ops->slab, GFP_ATOMIC | __GFP_NOWARN);
-	if (!req)
-		return NULL;
-	req->rsk_listener = NULL;
-	if (attach_listener) {
-		if (unlikely(!refcount_inc_not_zero(&sk_listener->sk_refcnt))) {
-			kmem_cache_free(ops->slab, req);
-			return NULL;
-		}
-		req->rsk_listener = sk_listener;
-	}
-	req->rsk_ops = ops;
-	req_to_sk(req)->sk_prot = sk_listener->sk_prot;
-	sk_node_init(&req_to_sk(req)->sk_node);
-	sk_tx_queue_clear(req_to_sk(req));
-	req->saved_syn = NULL;
-	req->syncookie = 0;
-	req->timeout = 0;
-	req->num_timeout = 0;
-	req->num_retrans = 0;
-	req->sk = NULL;
-	refcount_set(&req->rsk_refcnt, 0);
-
-	return req;
-}
-#define reqsk_alloc(...)	alloc_hooks(reqsk_alloc_noprof(__VA_ARGS__))
-
 static inline void __reqsk_free(struct request_sock *req)
 {
 	req->rsk_ops->destructor(req);
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index a9d2e6308910..7ced569778ab 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -911,6 +911,39 @@ int inet_rtx_syn_ack(const struct sock *parent, struct request_sock *req)
 }
 EXPORT_SYMBOL(inet_rtx_syn_ack);
 
+static struct request_sock *
+reqsk_alloc_noprof(const struct request_sock_ops *ops, struct sock *sk_listener,
+		   bool attach_listener)
+{
+	struct request_sock *req;
+
+	req = kmem_cache_alloc_noprof(ops->slab, GFP_ATOMIC | __GFP_NOWARN);
+	if (!req)
+		return NULL;
+	req->rsk_listener = NULL;
+	if (attach_listener) {
+		if (unlikely(!refcount_inc_not_zero(&sk_listener->sk_refcnt))) {
+			kmem_cache_free(ops->slab, req);
+			return NULL;
+		}
+		req->rsk_listener = sk_listener;
+	}
+	req->rsk_ops = ops;
+	req_to_sk(req)->sk_prot = sk_listener->sk_prot;
+	sk_node_init(&req_to_sk(req)->sk_node);
+	sk_tx_queue_clear(req_to_sk(req));
+	req->saved_syn = NULL;
+	req->syncookie = 0;
+	req->timeout = 0;
+	req->num_timeout = 0;
+	req->num_retrans = 0;
+	req->sk = NULL;
+	refcount_set(&req->rsk_refcnt, 0);
+
+	return req;
+}
+#define reqsk_alloc(...)	alloc_hooks(reqsk_alloc_noprof(__VA_ARGS__))
+
 struct request_sock *inet_reqsk_alloc(const struct request_sock_ops *ops,
 				      struct sock *sk_listener,
 				      bool attach_listener)
-- 
cgit v1.2.3-58-ga151


From 7015843afcaf68c132784c89528dfddc0005e483 Mon Sep 17 00:00:00 2001
From: Yonghong Song <yonghong.song@linux.dev>
Date: Wed, 5 Jun 2024 13:12:03 -0700
Subject: selftests/bpf: Fix send_signal test with nested CONFIG_PARAVIRT

Alexei reported that send_signal test may fail with nested CONFIG_PARAVIRT
configs. In this particular case, the base VM is AMD with 166 cpus, and I
run selftests with regular qemu on top of that and indeed send_signal test
failed. I also tried with an Intel box with 80 cpus and there is no issue.

The main qemu command line includes:

  -enable-kvm -smp 16 -cpu host

The failure log looks like:

  $ ./test_progs -t send_signal
  [   48.501588] watchdog: BUG: soft lockup - CPU#9 stuck for 26s! [test_progs:2225]
  [   48.503622] Modules linked in: bpf_testmod(O)
  [   48.503622] CPU: 9 PID: 2225 Comm: test_progs Tainted: G           O       6.9.0-08561-g2c1713a8f1c9-dirty #69
  [   48.507629] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.15.0-0-g2dd4b9b3f840-prebuilt.qemu.org 04/01/2014
  [   48.511635] RIP: 0010:handle_softirqs+0x71/0x290
  [   48.511635] Code: [...] 10 0a 00 00 00 31 c0 65 66 89 05 d5 f4 fa 7e fb bb ff ff ff ff <49> c7 c2 cb
  [   48.518527] RSP: 0018:ffffc90000310fa0 EFLAGS: 00000246
  [   48.519579] RAX: 0000000000000000 RBX: 00000000ffffffff RCX: 00000000000006e0
  [   48.522526] RDX: 0000000000000006 RSI: ffff88810791ae80 RDI: 0000000000000000
  [   48.523587] RBP: ffffc90000fabc88 R08: 00000005a0af4f7f R09: 0000000000000000
  [   48.525525] R10: 0000000561d2f29c R11: 0000000000006534 R12: 0000000000000280
  [   48.528525] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
  [   48.528525] FS:  00007f2f2885cd00(0000) GS:ffff888237c40000(0000) knlGS:0000000000000000
  [   48.531600] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  [   48.535520] CR2: 00007f2f287059f0 CR3: 0000000106a28002 CR4: 00000000003706f0
  [   48.537538] Call Trace:
  [   48.537538]  <IRQ>
  [   48.537538]  ? watchdog_timer_fn+0x1cd/0x250
  [   48.539590]  ? lockup_detector_update_enable+0x50/0x50
  [   48.539590]  ? __hrtimer_run_queues+0xff/0x280
  [   48.542520]  ? hrtimer_interrupt+0x103/0x230
  [   48.544524]  ? __sysvec_apic_timer_interrupt+0x4f/0x140
  [   48.545522]  ? sysvec_apic_timer_interrupt+0x3a/0x90
  [   48.547612]  ? asm_sysvec_apic_timer_interrupt+0x1a/0x20
  [   48.547612]  ? handle_softirqs+0x71/0x290
  [   48.547612]  irq_exit_rcu+0x63/0x80
  [   48.551585]  sysvec_apic_timer_interrupt+0x75/0x90
  [   48.552521]  </IRQ>
  [   48.553529]  <TASK>
  [   48.553529]  asm_sysvec_apic_timer_interrupt+0x1a/0x20
  [   48.555609] RIP: 0010:finish_task_switch.isra.0+0x90/0x260
  [   48.556526] Code: [...] 9f 58 0a 00 00 48 85 db 0f 85 89 01 00 00 4c 89 ff e8 53 d9 bd 00 fb 66 90 <4d> 85 ed 74
  [   48.562524] RSP: 0018:ffffc90000fabd38 EFLAGS: 00000282
  [   48.563589] RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffffff83385620
  [   48.563589] RDX: ffff888237c73ae4 RSI: 0000000000000000 RDI: ffff888237c6fd00
  [   48.568521] RBP: ffffc90000fabd68 R08: 0000000000000000 R09: 0000000000000000
  [   48.569528] R10: 0000000000000001 R11: 0000000000000000 R12: ffff8881009d0000
  [   48.573525] R13: ffff8881024e5400 R14: ffff88810791ae80 R15: ffff888237c6fd00
  [   48.575614]  ? finish_task_switch.isra.0+0x8d/0x260
  [   48.576523]  __schedule+0x364/0xac0
  [   48.577535]  schedule+0x2e/0x110
  [   48.578555]  pipe_read+0x301/0x400
  [   48.579589]  ? destroy_sched_domains_rcu+0x30/0x30
  [   48.579589]  vfs_read+0x2b3/0x2f0
  [   48.579589]  ksys_read+0x8b/0xc0
  [   48.583590]  do_syscall_64+0x3d/0xc0
  [   48.583590]  entry_SYSCALL_64_after_hwframe+0x4b/0x53
  [   48.586525] RIP: 0033:0x7f2f28703fa1
  [   48.587592] Code: [...] 00 00 00 0f 1f 44 00 00 f3 0f 1e fa 80 3d c5 23 14 00 00 74 13 31 c0 0f 05 <48> 3d 00 f0
  [   48.593534] RSP: 002b:00007ffd90f8cf88 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
  [   48.595589] RAX: ffffffffffffffda RBX: 00007ffd90f8d5e8 RCX: 00007f2f28703fa1
  [   48.595589] RDX: 0000000000000001 RSI: 00007ffd90f8cfb0 RDI: 0000000000000006
  [   48.599592] RBP: 00007ffd90f8d2f0 R08: 0000000000000064 R09: 0000000000000000
  [   48.602527] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
  [   48.603589] R13: 00007ffd90f8d608 R14: 00007f2f288d8000 R15: 0000000000f6bdb0
  [   48.605527]  </TASK>

In the test, two processes are communicating through pipe. Further debugging
with strace found that the above splat is triggered as read() syscall could
not receive the data even if the corresponding write() syscall in another
process successfully wrote data into the pipe.

The failed subtest is "send_signal_perf". The corresponding perf event has
sample_period 1 and config PERF_COUNT_SW_CPU_CLOCK. sample_period 1 means every
overflow event will trigger a call to the BPF program. So I suspect this may
overwhelm the system. So I increased the sample_period to 100,000 and the test
passed. The sample_period 10,000 still has the test failed.

In other parts of selftest, e.g., [1], sample_freq is used instead. So I
decided to use sample_freq = 1,000 since the test can pass as well.

  [1] https://lore.kernel.org/bpf/20240604070700.3032142-1-song@kernel.org/

Reported-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240605201203.2603846-1-yonghong.song@linux.dev
---
 tools/testing/selftests/bpf/prog_tests/send_signal.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/send_signal.c b/tools/testing/selftests/bpf/prog_tests/send_signal.c
index 920aee41bd58..6cc69900b310 100644
--- a/tools/testing/selftests/bpf/prog_tests/send_signal.c
+++ b/tools/testing/selftests/bpf/prog_tests/send_signal.c
@@ -156,7 +156,8 @@ static void test_send_signal_tracepoint(bool signal_thread)
 static void test_send_signal_perf(bool signal_thread)
 {
 	struct perf_event_attr attr = {
-		.sample_period = 1,
+		.freq = 1,
+		.sample_freq = 1000,
 		.type = PERF_TYPE_SOFTWARE,
 		.config = PERF_COUNT_SW_CPU_CLOCK,
 	};
-- 
cgit v1.2.3-58-ga151


From b24862bac7b5db326716ad07bbff7b6ee3b09a59 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Wed, 5 Jun 2024 16:33:14 +0100
Subject: selftests/bpf: Add btf_field_iter selftests

The added selftests verify that for every BTF kind we iterate correctly
over consituent strings and ids.

Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240605153314.3727466-1-alan.maguire@oracle.com
---
 .../selftests/bpf/prog_tests/btf_field_iter.c      | 161 +++++++++++++++++++++
 1 file changed, 161 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/btf_field_iter.c

diff --git a/tools/testing/selftests/bpf/prog_tests/btf_field_iter.c b/tools/testing/selftests/bpf/prog_tests/btf_field_iter.c
new file mode 100644
index 000000000000..32159d3eb281
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/btf_field_iter.c
@@ -0,0 +1,161 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2024, Oracle and/or its affiliates. */
+
+#include <test_progs.h>
+#include <bpf/btf.h>
+#include "btf_helpers.h"
+#include "bpf/libbpf_internal.h"
+
+struct field_data {
+	__u32 ids[5];
+	const char *strs[5];
+} fields[] = {
+	{ .ids = {},		.strs = {} },
+	{ .ids = {},		.strs = { "int" } },
+	{ .ids = {},		.strs = { "int64" } },
+	{ .ids = { 1 },		.strs = { "" } },
+	{ .ids = { 2, 1 },	.strs = { "" } },
+	{ .ids = { 3, 1 },	.strs = { "s1", "f1", "f2" } },
+	{ .ids = { 1, 5 },	.strs = { "u1", "f1", "f2" } },
+	{ .ids = {},		.strs = { "e1", "v1", "v2" } },
+	{ .ids = {},		.strs = { "fw1" } },
+	{ .ids = { 1 },		.strs = { "t" } },
+	{ .ids = { 2 },		.strs = { "" } },
+	{ .ids = { 1 },		.strs = { "" } },
+	{ .ids = { 3 },		.strs = { "" } },
+	{ .ids = { 1, 1, 3 },	.strs = { "", "p1", "p2" } },
+	{ .ids = { 13 },	.strs = { "func" } },
+	{ .ids = { 1 },		.strs = { "var1" } },
+	{ .ids = { 3 },		.strs = { "var2" } },
+	{ .ids = {},		.strs = { "float" } },
+	{ .ids = { 11 },	.strs = { "decltag" } },
+	{ .ids = { 6 },		.strs = { "typetag" } },
+	{ .ids = {},		.strs = { "e64", "eval1", "eval2", "eval3" } },
+	{ .ids = { 15, 16 },	.strs = { "datasec1" } }
+
+};
+
+/* Fabricate BTF with various types and check BTF field iteration finds types,
+ * strings expected.
+ */
+void test_btf_field_iter(void)
+{
+	struct btf *btf = NULL;
+	int id;
+
+	btf = btf__new_empty();
+	if (!ASSERT_OK_PTR(btf, "empty_btf"))
+		return;
+
+	btf__add_int(btf, "int", 4, BTF_INT_SIGNED);	/* [1] int */
+	btf__add_int(btf, "int64", 8, BTF_INT_SIGNED);	/* [2] int64 */
+	btf__add_ptr(btf, 1);				/* [3] int * */
+	btf__add_array(btf, 1, 2, 3);			/* [4] int64[3] */
+	btf__add_struct(btf, "s1", 12);			/* [5] struct s1 { */
+	btf__add_field(btf, "f1", 3, 0, 0);		/*      int *f1; */
+	btf__add_field(btf, "f2", 1, 0, 0);		/*	int f2; */
+							/* } */
+	btf__add_union(btf, "u1", 12);			/* [6] union u1 { */
+	btf__add_field(btf, "f1", 1, 0, 0);		/*	int f1; */
+	btf__add_field(btf, "f2", 5, 0, 0);		/*	struct s1 f2; */
+							/* } */
+	btf__add_enum(btf, "e1", 4);			/* [7] enum e1 { */
+	btf__add_enum_value(btf, "v1", 1);		/*	v1 = 1; */
+	btf__add_enum_value(btf, "v2", 2);		/*	v2 = 2; */
+							/* } */
+
+	btf__add_fwd(btf, "fw1", BTF_FWD_STRUCT);	/* [8] struct fw1; */
+	btf__add_typedef(btf, "t", 1);			/* [9] typedef int t; */
+	btf__add_volatile(btf, 2);			/* [10] volatile int64; */
+	btf__add_const(btf, 1);				/* [11] const int; */
+	btf__add_restrict(btf, 3);			/* [12] restrict int *; */
+	btf__add_func_proto(btf, 1);			/* [13] int (*)(int p1, int *p2); */
+	btf__add_func_param(btf, "p1", 1);
+	btf__add_func_param(btf, "p2", 3);
+
+	btf__add_func(btf, "func", BTF_FUNC_GLOBAL, 13);/* [14] int func(int p1, int *p2); */
+	btf__add_var(btf, "var1", BTF_VAR_STATIC, 1);	/* [15] static int var1; */
+	btf__add_var(btf, "var2", BTF_VAR_STATIC, 3);	/* [16] static int *var2; */
+	btf__add_float(btf, "float", 4);		/* [17] float; */
+	btf__add_decl_tag(btf, "decltag", 11, -1);	/* [18] decltag const int; */
+	btf__add_type_tag(btf, "typetag", 6);		/* [19] typetag union u1; */
+	btf__add_enum64(btf, "e64", 8, true);		/* [20] enum { */
+	btf__add_enum64_value(btf, "eval1", 1000);	/*	 eval1 = 1000, */
+	btf__add_enum64_value(btf, "eval2", 2000);	/*	 eval2 = 2000, */
+	btf__add_enum64_value(btf, "eval3", 3000);	/*	 eval3 = 3000 */
+							/* } */
+	btf__add_datasec(btf, "datasec1", 12);		/* [21] datasec datasec1 */
+	btf__add_datasec_var_info(btf, 15, 0, 4);
+	btf__add_datasec_var_info(btf, 16, 4, 8);
+
+	VALIDATE_RAW_BTF(
+		btf,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int64' size=8 bits_offset=0 nr_bits=64 encoding=SIGNED",
+		"[3] PTR '(anon)' type_id=1",
+		"[4] ARRAY '(anon)' type_id=2 index_type_id=1 nr_elems=3",
+		"[5] STRUCT 's1' size=12 vlen=2\n"
+		"\t'f1' type_id=3 bits_offset=0\n"
+		"\t'f2' type_id=1 bits_offset=0",
+		"[6] UNION 'u1' size=12 vlen=2\n"
+		"\t'f1' type_id=1 bits_offset=0\n"
+		"\t'f2' type_id=5 bits_offset=0",
+		"[7] ENUM 'e1' encoding=UNSIGNED size=4 vlen=2\n"
+		"\t'v1' val=1\n"
+		"\t'v2' val=2",
+		"[8] FWD 'fw1' fwd_kind=struct",
+		"[9] TYPEDEF 't' type_id=1",
+		"[10] VOLATILE '(anon)' type_id=2",
+		"[11] CONST '(anon)' type_id=1",
+		"[12] RESTRICT '(anon)' type_id=3",
+		"[13] FUNC_PROTO '(anon)' ret_type_id=1 vlen=2\n"
+		"\t'p1' type_id=1\n"
+		"\t'p2' type_id=3",
+		"[14] FUNC 'func' type_id=13 linkage=global",
+		"[15] VAR 'var1' type_id=1, linkage=static",
+		"[16] VAR 'var2' type_id=3, linkage=static",
+		"[17] FLOAT 'float' size=4",
+		"[18] DECL_TAG 'decltag' type_id=11 component_idx=-1",
+		"[19] TYPE_TAG 'typetag' type_id=6",
+		"[20] ENUM64 'e64' encoding=SIGNED size=8 vlen=3\n"
+		"\t'eval1' val=1000\n"
+		"\t'eval2' val=2000\n"
+		"\t'eval3' val=3000",
+		"[21] DATASEC 'datasec1' size=12 vlen=2\n"
+		"\ttype_id=15 offset=0 size=4\n"
+		"\ttype_id=16 offset=4 size=8");
+
+	for (id = 1; id < btf__type_cnt(btf); id++) {
+		struct btf_type *t = btf_type_by_id(btf, id);
+		struct btf_field_iter it_strs, it_ids;
+		int str_idx = 0, id_idx = 0;
+		__u32 *next_str, *next_id;
+
+		if (!ASSERT_OK_PTR(t, "btf_type_by_id"))
+			break;
+		if (!ASSERT_OK(btf_field_iter_init(&it_strs, t, BTF_FIELD_ITER_STRS),
+			       "iter_init_strs"))
+			break;
+		if (!ASSERT_OK(btf_field_iter_init(&it_ids, t, BTF_FIELD_ITER_IDS),
+			       "iter_init_ids"))
+			break;
+		while ((next_str = btf_field_iter_next(&it_strs))) {
+			const char *str = btf__str_by_offset(btf, *next_str);
+
+			if (!ASSERT_OK(strcmp(fields[id].strs[str_idx], str), "field_str_match"))
+				break;
+			str_idx++;
+		}
+		/* ensure no more strings are expected */
+		ASSERT_EQ(fields[id].strs[str_idx], NULL, "field_str_cnt");
+
+		while ((next_id = btf_field_iter_next(&it_ids))) {
+			if (!ASSERT_EQ(*next_id, fields[id].ids[id_idx], "field_id_match"))
+				break;
+			id_idx++;
+		}
+		/* ensure no more ids are expected */
+		ASSERT_EQ(fields[id].ids[id_idx], 0, "field_id_cnt");
+	}
+	btf__free(btf);
+}
-- 
cgit v1.2.3-58-ga151


From 08ac454e258e38813afb906650f19acce3afd982 Mon Sep 17 00:00:00 2001
From: Mykyta Yatsenko <yatsenko@meta.com>
Date: Wed, 5 Jun 2024 18:51:35 +0100
Subject: libbpf: Auto-attach struct_ops BPF maps in BPF skeleton

Similarly to `bpf_program`, support `bpf_map` automatic attachment in
`bpf_object__attach_skeleton`. Currently only struct_ops maps could be
attached.

On bpftool side, code-generate links in skeleton struct for struct_ops maps.
Similarly to `bpf_program_skeleton`, set links in `bpf_map_skeleton`.

On libbpf side, extend `bpf_map` with new `autoattach` field to support
enabling or disabling autoattach functionality, introducing
getter/setter for this field.

`bpf_object__(attach|detach)_skeleton` is extended with
attaching/detaching struct_ops maps logic.

Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240605175135.117127-1-yatsenko@meta.com
---
 tools/bpf/bpftool/gen.c  | 36 ++++++++++++++++++++++++---
 tools/lib/bpf/libbpf.c   | 64 +++++++++++++++++++++++++++++++++++++++++++++---
 tools/lib/bpf/libbpf.h   | 18 ++++++++++++++
 tools/lib/bpf/libbpf.map |  2 ++
 4 files changed, 113 insertions(+), 7 deletions(-)

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index d244a7de387e..4a4eedfcd479 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -848,7 +848,7 @@ out:
 }
 
 static void
-codegen_maps_skeleton(struct bpf_object *obj, size_t map_cnt, bool mmaped)
+codegen_maps_skeleton(struct bpf_object *obj, size_t map_cnt, bool mmaped, bool populate_links)
 {
 	struct bpf_map *map;
 	char ident[256];
@@ -888,6 +888,14 @@ codegen_maps_skeleton(struct bpf_object *obj, size_t map_cnt, bool mmaped)
 			printf("\ts->maps[%zu].mmaped = (void **)&obj->%s;\n",
 				i, ident);
 		}
+
+		if (populate_links && bpf_map__type(map) == BPF_MAP_TYPE_STRUCT_OPS) {
+			codegen("\
+				\n\
+					s->maps[%zu].link = &obj->links.%s;\n\
+				",
+				i, ident);
+		}
 		i++;
 	}
 }
@@ -1141,7 +1149,7 @@ static void gen_st_ops_shadow_init(struct btf *btf, struct bpf_object *obj)
 static int do_skeleton(int argc, char **argv)
 {
 	char header_guard[MAX_OBJ_NAME_LEN + sizeof("__SKEL_H__")];
-	size_t map_cnt = 0, prog_cnt = 0, file_sz, mmap_sz;
+	size_t map_cnt = 0, prog_cnt = 0, attach_map_cnt = 0, file_sz, mmap_sz;
 	DECLARE_LIBBPF_OPTS(bpf_object_open_opts, opts);
 	char obj_name[MAX_OBJ_NAME_LEN] = "", *obj_data;
 	struct bpf_object *obj = NULL;
@@ -1225,6 +1233,10 @@ static int do_skeleton(int argc, char **argv)
 			      bpf_map__name(map));
 			continue;
 		}
+
+		if (bpf_map__type(map) == BPF_MAP_TYPE_STRUCT_OPS)
+			attach_map_cnt++;
+
 		map_cnt++;
 	}
 	bpf_object__for_each_program(prog, obj) {
@@ -1297,6 +1309,9 @@ static int do_skeleton(int argc, char **argv)
 				       bpf_program__name(prog));
 		}
 		printf("\t} progs;\n");
+	}
+
+	if (prog_cnt + attach_map_cnt) {
 		printf("\tstruct {\n");
 		bpf_object__for_each_program(prog, obj) {
 			if (use_loader)
@@ -1306,6 +1321,19 @@ static int do_skeleton(int argc, char **argv)
 				printf("\t\tstruct bpf_link *%s;\n",
 				       bpf_program__name(prog));
 		}
+
+		bpf_object__for_each_map(map, obj) {
+			if (!get_map_ident(map, ident, sizeof(ident)))
+				continue;
+			if (bpf_map__type(map) != BPF_MAP_TYPE_STRUCT_OPS)
+				continue;
+
+			if (use_loader)
+				printf("t\tint %s_fd;\n", ident);
+			else
+				printf("\t\tstruct bpf_link *%s;\n", ident);
+		}
+
 		printf("\t} links;\n");
 	}
 
@@ -1448,7 +1476,7 @@ static int do_skeleton(int argc, char **argv)
 		obj_name
 	);
 
-	codegen_maps_skeleton(obj, map_cnt, true /*mmaped*/);
+	codegen_maps_skeleton(obj, map_cnt, true /*mmaped*/, true /*links*/);
 	codegen_progs_skeleton(obj, prog_cnt, true /*populate_links*/);
 
 	codegen("\
@@ -1786,7 +1814,7 @@ static int do_subskeleton(int argc, char **argv)
 		}
 	}
 
-	codegen_maps_skeleton(obj, map_cnt, false /*mmaped*/);
+	codegen_maps_skeleton(obj, map_cnt, false /*mmaped*/, false /*links*/);
 	codegen_progs_skeleton(obj, prog_cnt, false /*links*/);
 
 	codegen("\
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index d1627a2ca30b..4a28fac4908a 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -572,6 +572,7 @@ struct bpf_map {
 	bool pinned;
 	bool reused;
 	bool autocreate;
+	bool autoattach;
 	__u64 map_extra;
 };
 
@@ -1400,6 +1401,7 @@ static int init_struct_ops_maps(struct bpf_object *obj, const char *sec_name,
 		map->def.value_size = type->size;
 		map->def.max_entries = 1;
 		map->def.map_flags = strcmp(sec_name, STRUCT_OPS_LINK_SEC) == 0 ? BPF_F_LINK : 0;
+		map->autoattach = true;
 
 		map->st_ops = calloc(1, sizeof(*map->st_ops));
 		if (!map->st_ops)
@@ -4819,6 +4821,20 @@ int bpf_map__set_autocreate(struct bpf_map *map, bool autocreate)
 	return 0;
 }
 
+int bpf_map__set_autoattach(struct bpf_map *map, bool autoattach)
+{
+	if (!bpf_map__is_struct_ops(map))
+		return libbpf_err(-EINVAL);
+
+	map->autoattach = autoattach;
+	return 0;
+}
+
+bool bpf_map__autoattach(const struct bpf_map *map)
+{
+	return map->autoattach;
+}
+
 int bpf_map__reuse_fd(struct bpf_map *map, int fd)
 {
 	struct bpf_map_info info;
@@ -12900,8 +12916,10 @@ struct bpf_link *bpf_map__attach_struct_ops(const struct bpf_map *map)
 	__u32 zero = 0;
 	int err, fd;
 
-	if (!bpf_map__is_struct_ops(map))
+	if (!bpf_map__is_struct_ops(map)) {
+		pr_warn("map '%s': can't attach non-struct_ops map\n", map->name);
 		return libbpf_err_ptr(-EINVAL);
+	}
 
 	if (map->fd < 0) {
 		pr_warn("map '%s': can't attach BPF map without FD (was it created?)\n", map->name);
@@ -13945,6 +13963,35 @@ int bpf_object__attach_skeleton(struct bpf_object_skeleton *s)
 		 */
 	}
 
+	/* Skeleton is created with earlier version of bpftool
+	 * which does not support auto-attachment
+	 */
+	if (s->map_skel_sz < sizeof(struct bpf_map_skeleton))
+		return 0;
+
+	for (i = 0; i < s->map_cnt; i++) {
+		struct bpf_map *map = *s->maps[i].map;
+		struct bpf_link **link = s->maps[i].link;
+
+		if (!map->autocreate || !map->autoattach)
+			continue;
+
+		if (*link)
+			continue;
+
+		/* only struct_ops maps can be attached */
+		if (!bpf_map__is_struct_ops(map))
+			continue;
+		*link = bpf_map__attach_struct_ops(map);
+
+		if (!*link) {
+			err = -errno;
+			pr_warn("map '%s': failed to auto-attach: %d\n",
+				bpf_map__name(map), err);
+			return libbpf_err(err);
+		}
+	}
+
 	return 0;
 }
 
@@ -13958,6 +14005,18 @@ void bpf_object__detach_skeleton(struct bpf_object_skeleton *s)
 		bpf_link__destroy(*link);
 		*link = NULL;
 	}
+
+	if (s->map_skel_sz < sizeof(struct bpf_map_skeleton))
+		return;
+
+	for (i = 0; i < s->map_cnt; i++) {
+		struct bpf_link **link = s->maps[i].link;
+
+		if (link) {
+			bpf_link__destroy(*link);
+			*link = NULL;
+		}
+	}
 }
 
 void bpf_object__destroy_skeleton(struct bpf_object_skeleton *s)
@@ -13965,8 +14024,7 @@ void bpf_object__destroy_skeleton(struct bpf_object_skeleton *s)
 	if (!s)
 		return;
 
-	if (s->progs)
-		bpf_object__detach_skeleton(s);
+	bpf_object__detach_skeleton(s);
 	if (s->obj)
 		bpf_object__close(*s->obj);
 	free(s->maps);
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 26e4e35528c5..64a6a3d323e3 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -978,6 +978,23 @@ bpf_object__prev_map(const struct bpf_object *obj, const struct bpf_map *map);
 LIBBPF_API int bpf_map__set_autocreate(struct bpf_map *map, bool autocreate);
 LIBBPF_API bool bpf_map__autocreate(const struct bpf_map *map);
 
+/**
+ * @brief **bpf_map__set_autoattach()** sets whether libbpf has to auto-attach
+ * map during BPF skeleton attach phase.
+ * @param map the BPF map instance
+ * @param autoattach whether to attach map during BPF skeleton attach phase
+ * @return 0 on success; negative error code, otherwise
+ */
+LIBBPF_API int bpf_map__set_autoattach(struct bpf_map *map, bool autoattach);
+
+/**
+ * @brief **bpf_map__autoattach()** returns whether BPF map is configured to
+ * auto-attach during BPF skeleton attach phase.
+ * @param map the BPF map instance
+ * @return true if map is set to auto-attach during skeleton attach phase; false, otherwise
+ */
+LIBBPF_API bool bpf_map__autoattach(const struct bpf_map *map);
+
 /**
  * @brief **bpf_map__fd()** gets the file descriptor of the passed
  * BPF map
@@ -1672,6 +1689,7 @@ struct bpf_map_skeleton {
 	const char *name;
 	struct bpf_map **map;
 	void **mmaped;
+	struct bpf_link **link;
 };
 
 struct bpf_prog_skeleton {
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index c1ce8aa3520b..40595233dc7f 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -419,6 +419,8 @@ LIBBPF_1.4.0 {
 
 LIBBPF_1.5.0 {
 	global:
+		bpf_map__autoattach;
+		bpf_map__set_autoattach;
 		bpf_program__attach_sockmap;
 		ring__consume_n;
 		ring_buffer__consume_n;
-- 
cgit v1.2.3-58-ga151


From 9abdfd8a212332c64f6d0a27fc2ad69e9e0335d1 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 30 May 2024 15:41:08 +0800
Subject: selftests/bpf: Use connect_to_fd_opts in do_test in bpf_tcp_ca

This patch uses connect_to_fd_opts() instead of using connect_fd_to_fd()
and settcpca() in do_test() in prog_tests/bpf_tcp_ca.c to accept a struct
network_helper_opts argument.

Then define a dctcp dedicated post_socket_cb callback stg_post_socket_cb(),
invoking both settcpca() and bpf_map_update_elem() in it, and set it in
test_dctcp(). For passing map_fd into stg_post_socket_cb() callback, a new
member map_fd is added in struct cb_opts.

Add another "const struct network_helper_opts *cli_opts" to do_test() to
separate it from the server "opts".

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/876ec90430865bc468e3b7f6fb2648420b075548.1717054461.git.tanggeliang@kylinos.cn
---
 .../testing/selftests/bpf/prog_tests/bpf_tcp_ca.c  | 61 ++++++++++++----------
 1 file changed, 34 insertions(+), 27 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index ebc7d4616880..2f9d373feb0a 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -25,6 +25,7 @@ static int expected_stg = 0xeB9F;
 
 struct cb_opts {
 	const char *cc;
+	int map_fd;
 };
 
 static int settcpca(int fd, const char *tcp_ca)
@@ -39,9 +40,9 @@ static int settcpca(int fd, const char *tcp_ca)
 }
 
 static void do_test(const struct network_helper_opts *opts,
+		    const struct network_helper_opts *cli_opts,
 		    const struct bpf_map *sk_stg_map)
 {
-	struct cb_opts *cb_opts = (struct cb_opts *)opts->cb_opts;
 	int lfd = -1, fd = -1;
 	int err;
 
@@ -49,25 +50,9 @@ static void do_test(const struct network_helper_opts *opts,
 	if (!ASSERT_NEQ(lfd, -1, "socket"))
 		return;
 
-	fd = socket(AF_INET6, SOCK_STREAM, 0);
-	if (!ASSERT_NEQ(fd, -1, "socket")) {
-		close(lfd);
-		return;
-	}
-
-	if (settcpca(fd, cb_opts->cc))
-		goto done;
-
-	if (sk_stg_map) {
-		err = bpf_map_update_elem(bpf_map__fd(sk_stg_map), &fd,
-					  &expected_stg, BPF_NOEXIST);
-		if (!ASSERT_OK(err, "bpf_map_update_elem(sk_stg_map)"))
-			goto done;
-	}
-
 	/* connect to server */
-	err = connect_fd_to_fd(fd, lfd, 0);
-	if (!ASSERT_NEQ(err, -1, "connect"))
+	fd = connect_to_fd_opts(lfd, cli_opts);
+	if (!ASSERT_NEQ(fd, -1, "connect_to_fd_opts"))
 		goto done;
 
 	if (sk_stg_map) {
@@ -116,7 +101,7 @@ static void test_cubic(void)
 		return;
 	}
 
-	do_test(&opts, NULL);
+	do_test(&opts, &opts, NULL);
 
 	ASSERT_EQ(cubic_skel->bss->bpf_cubic_acked_called, 1, "pkts_acked called");
 
@@ -124,6 +109,23 @@ static void test_cubic(void)
 	bpf_cubic__destroy(cubic_skel);
 }
 
+static int stg_post_socket_cb(int fd, void *opts)
+{
+	struct cb_opts *cb_opts = (struct cb_opts *)opts;
+	int err;
+
+	err = settcpca(fd, cb_opts->cc);
+	if (err)
+		return err;
+
+	err = bpf_map_update_elem(cb_opts->map_fd, &fd,
+				  &expected_stg, BPF_NOEXIST);
+	if (!ASSERT_OK(err, "bpf_map_update_elem(sk_stg_map)"))
+		return err;
+
+	return 0;
+}
+
 static void test_dctcp(void)
 {
 	struct cb_opts cb_opts = {
@@ -133,6 +135,10 @@ static void test_dctcp(void)
 		.post_socket_cb	= cc_cb,
 		.cb_opts	= &cb_opts,
 	};
+	struct network_helper_opts cli_opts = {
+		.post_socket_cb	= stg_post_socket_cb,
+		.cb_opts	= &cb_opts,
+	};
 	struct bpf_dctcp *dctcp_skel;
 	struct bpf_link *link;
 
@@ -146,7 +152,8 @@ static void test_dctcp(void)
 		return;
 	}
 
-	do_test(&opts, dctcp_skel->maps.sk_stg_map);
+	cb_opts.map_fd = bpf_map__fd(dctcp_skel->maps.sk_stg_map);
+	do_test(&opts, &cli_opts, dctcp_skel->maps.sk_stg_map);
 	ASSERT_EQ(dctcp_skel->bss->stg_result, expected_stg, "stg_result");
 
 	bpf_link__destroy(link);
@@ -350,14 +357,14 @@ static void test_update_ca(void)
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
 	ASSERT_OK_PTR(link, "attach_struct_ops");
 
-	do_test(&opts, NULL);
+	do_test(&opts, &opts, NULL);
 	saved_ca1_cnt = skel->bss->ca1_cnt;
 	ASSERT_GT(saved_ca1_cnt, 0, "ca1_ca1_cnt");
 
 	err = bpf_link__update_map(link, skel->maps.ca_update_2);
 	ASSERT_OK(err, "update_map");
 
-	do_test(&opts, NULL);
+	do_test(&opts, &opts, NULL);
 	ASSERT_EQ(skel->bss->ca1_cnt, saved_ca1_cnt, "ca2_ca1_cnt");
 	ASSERT_GT(skel->bss->ca2_cnt, 0, "ca2_ca2_cnt");
 
@@ -386,14 +393,14 @@ static void test_update_wrong(void)
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
 	ASSERT_OK_PTR(link, "attach_struct_ops");
 
-	do_test(&opts, NULL);
+	do_test(&opts, &opts, NULL);
 	saved_ca1_cnt = skel->bss->ca1_cnt;
 	ASSERT_GT(saved_ca1_cnt, 0, "ca1_ca1_cnt");
 
 	err = bpf_link__update_map(link, skel->maps.ca_wrong);
 	ASSERT_ERR(err, "update_map");
 
-	do_test(&opts, NULL);
+	do_test(&opts, &opts, NULL);
 	ASSERT_GT(skel->bss->ca1_cnt, saved_ca1_cnt, "ca2_ca1_cnt");
 
 	bpf_link__destroy(link);
@@ -423,7 +430,7 @@ static void test_mixed_links(void)
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
 	ASSERT_OK_PTR(link, "attach_struct_ops");
 
-	do_test(&opts, NULL);
+	do_test(&opts, &opts, NULL);
 	ASSERT_GT(skel->bss->ca1_cnt, 0, "ca1_ca1_cnt");
 
 	err = bpf_link__update_map(link, skel->maps.ca_no_link);
@@ -530,7 +537,7 @@ static void test_cc_cubic(void)
 		return;
 	}
 
-	do_test(&opts, NULL);
+	do_test(&opts, &opts, NULL);
 
 	bpf_link__destroy(link);
 	bpf_cc_cubic__destroy(cc_cubic_skel);
-- 
cgit v1.2.3-58-ga151


From fee97d0c9a14b5dd5cce0ec1df3a54a6b963f40c Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 30 May 2024 15:41:09 +0800
Subject: selftests/bpf: Add start_test helper in bpf_tcp_ca

For moving the "if (sk_stg_map)" block out of do_test(), extract the
code before this block as a new function start_test(). It creates
server-side and client-side sockets and returns them to the caller.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/48f2921ff9be958f5d3d28fe6bb7269a61cafa9f.1717054461.git.tanggeliang@kylinos.cn
---
 .../testing/selftests/bpf/prog_tests/bpf_tcp_ca.c  | 42 +++++++++++++++++-----
 1 file changed, 33 insertions(+), 9 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index 2f9d373feb0a..794651ce0629 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -39,6 +39,34 @@ static int settcpca(int fd, const char *tcp_ca)
 	return 0;
 }
 
+static bool start_test(char *addr_str,
+		       const struct network_helper_opts *srv_opts,
+		       const struct network_helper_opts *cli_opts,
+		       int *srv_fd, int *cli_fd)
+{
+	*srv_fd = start_server_str(AF_INET6, SOCK_STREAM, addr_str, 0, srv_opts);
+	if (!ASSERT_NEQ(*srv_fd, -1, "start_server_str"))
+		goto err;
+
+	/* connect to server */
+	*cli_fd = connect_to_fd_opts(*srv_fd, cli_opts);
+	if (!ASSERT_NEQ(*cli_fd, -1, "connect_to_fd_opts"))
+		goto err;
+
+	return true;
+
+err:
+	if (*srv_fd != -1) {
+		close(*srv_fd);
+		*srv_fd = -1;
+	}
+	if (*cli_fd != -1) {
+		close(*cli_fd);
+		*cli_fd = -1;
+	}
+	return false;
+}
+
 static void do_test(const struct network_helper_opts *opts,
 		    const struct network_helper_opts *cli_opts,
 		    const struct bpf_map *sk_stg_map)
@@ -46,13 +74,7 @@ static void do_test(const struct network_helper_opts *opts,
 	int lfd = -1, fd = -1;
 	int err;
 
-	lfd = start_server_str(AF_INET6, SOCK_STREAM, NULL, 0, opts);
-	if (!ASSERT_NEQ(lfd, -1, "socket"))
-		return;
-
-	/* connect to server */
-	fd = connect_to_fd_opts(lfd, cli_opts);
-	if (!ASSERT_NEQ(fd, -1, "connect_to_fd_opts"))
+	if (!start_test(NULL, opts, cli_opts, &lfd, &fd))
 		goto done;
 
 	if (sk_stg_map) {
@@ -68,8 +90,10 @@ static void do_test(const struct network_helper_opts *opts,
 	ASSERT_OK(send_recv_data(lfd, fd, total_bytes), "send_recv_data");
 
 done:
-	close(lfd);
-	close(fd);
+	if (lfd != -1)
+		close(lfd);
+	if (fd != -1)
+		close(fd);
 }
 
 static int cc_cb(int fd, void *opts)
-- 
cgit v1.2.3-58-ga151


From 224eeb5598c30ee835dc9fea4c7ad85a8fb7eda4 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 30 May 2024 15:41:10 +0800
Subject: selftests/bpf: Use start_test in test_dctcp_fallback in bpf_tcp_ca

The newly added helper start_test() can be used in test_dctcp_fallback()
too, to replace start_server_str() and connect_to_fd_opts(). In that
way, two network_helper_opts srv_opts and cli_opts are used instead of
the previously shared opts.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/792ca3bb013fa06e618176da02d75e4f79a76733.1717054461.git.tanggeliang@kylinos.cn
---
 tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c | 20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index 794651ce0629..d10217169ff8 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -229,17 +229,22 @@ static void test_invalid_license(void)
 static void test_dctcp_fallback(void)
 {
 	int err, lfd = -1, cli_fd = -1, srv_fd = -1;
-	struct network_helper_opts opts = {
-		.post_socket_cb	= cc_cb,
-	};
 	struct bpf_dctcp *dctcp_skel;
 	struct bpf_link *link = NULL;
 	struct cb_opts dctcp = {
 		.cc = "bpf_dctcp",
 	};
+	struct network_helper_opts srv_opts = {
+		.post_socket_cb = cc_cb,
+		.cb_opts = &dctcp,
+	};
 	struct cb_opts cubic = {
 		.cc = "cubic",
 	};
+	struct network_helper_opts cli_opts = {
+		.post_socket_cb = cc_cb,
+		.cb_opts = &cubic,
+	};
 	char srv_cc[16];
 	socklen_t cc_len = sizeof(srv_cc);
 
@@ -254,14 +259,7 @@ static void test_dctcp_fallback(void)
 	if (!ASSERT_OK_PTR(link, "dctcp link"))
 		goto done;
 
-	opts.cb_opts = &dctcp;
-	lfd = start_server_str(AF_INET6, SOCK_STREAM, "::1", 0, &opts);
-	if (!ASSERT_GE(lfd, 0, "lfd"))
-		goto done;
-
-	opts.cb_opts = &cubic;
-	cli_fd = connect_to_fd_opts(lfd, &opts);
-	if (!ASSERT_GE(cli_fd, 0, "cli_fd"))
+	if (!start_test("::1", &srv_opts, &cli_opts, &lfd, &cli_fd))
 		goto done;
 
 	srv_fd = accept(lfd, NULL, 0);
-- 
cgit v1.2.3-58-ga151


From cd984b2ed62423eb3daceacb21d651115a612af6 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 30 May 2024 15:41:11 +0800
Subject: selftests/bpf: Use start_test in test_dctcp in bpf_tcp_ca

The "if (sk_stg_map)" block in do_test() is only used by test_dctcp(),
it makes sense to move it from do_test() into test_dctcp(). Then
do_test() can be used by other tests except test_dctcp().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/9938916627b9810c877e5c03a621bc0ba5acf5c5.1717054461.git.tanggeliang@kylinos.cn
---
 .../testing/selftests/bpf/prog_tests/bpf_tcp_ca.c  | 27 ++++++++++++----------
 1 file changed, 15 insertions(+), 12 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index d10217169ff8..1b27d0232cbd 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -72,21 +72,10 @@ static void do_test(const struct network_helper_opts *opts,
 		    const struct bpf_map *sk_stg_map)
 {
 	int lfd = -1, fd = -1;
-	int err;
 
 	if (!start_test(NULL, opts, cli_opts, &lfd, &fd))
 		goto done;
 
-	if (sk_stg_map) {
-		int tmp_stg;
-
-		err = bpf_map_lookup_elem(bpf_map__fd(sk_stg_map), &fd,
-					  &tmp_stg);
-		if (!ASSERT_ERR(err, "bpf_map_lookup_elem(sk_stg_map)") ||
-				!ASSERT_EQ(errno, ENOENT, "bpf_map_lookup_elem(sk_stg_map)"))
-			goto done;
-	}
-
 	ASSERT_OK(send_recv_data(lfd, fd, total_bytes), "send_recv_data");
 
 done:
@@ -163,6 +152,7 @@ static void test_dctcp(void)
 		.post_socket_cb	= stg_post_socket_cb,
 		.cb_opts	= &cb_opts,
 	};
+	int lfd = -1, fd = -1, tmp_stg, err;
 	struct bpf_dctcp *dctcp_skel;
 	struct bpf_link *link;
 
@@ -177,11 +167,24 @@ static void test_dctcp(void)
 	}
 
 	cb_opts.map_fd = bpf_map__fd(dctcp_skel->maps.sk_stg_map);
-	do_test(&opts, &cli_opts, dctcp_skel->maps.sk_stg_map);
+	if (!start_test(NULL, &opts, &cli_opts, &lfd, &fd))
+		goto done;
+
+	err = bpf_map_lookup_elem(cb_opts.map_fd, &fd, &tmp_stg);
+	if (!ASSERT_ERR(err, "bpf_map_lookup_elem(sk_stg_map)") ||
+			!ASSERT_EQ(errno, ENOENT, "bpf_map_lookup_elem(sk_stg_map)"))
+		goto done;
+
+	ASSERT_OK(send_recv_data(lfd, fd, total_bytes), "send_recv_data");
 	ASSERT_EQ(dctcp_skel->bss->stg_result, expected_stg, "stg_result");
 
+done:
 	bpf_link__destroy(link);
 	bpf_dctcp__destroy(dctcp_skel);
+	if (lfd != -1)
+		close(lfd);
+	if (fd != -1)
+		close(fd);
 }
 
 static char *err_str;
-- 
cgit v1.2.3-58-ga151


From f85af9d955ac9601174e1c64f4b3308c1cae4a7e Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Thu, 30 May 2024 15:41:12 +0800
Subject: selftests/bpf: Drop useless arguments of do_test in bpf_tcp_ca

bpf_map_lookup_elem() has been removed from do_test(), it makes the
sk_stg_map argument of do_test() useless. In addition, two exactly the
same opts are passed in all the places where do_test() is invoked, so
cli_opts argument can be dropped too.

This patch drops these two useless arguments of do_test() in bpf_tcp_ca.c.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/7056eab111d78a05bce29d2821228dc93f240de4.1717054461.git.tanggeliang@kylinos.cn
---
 tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c | 20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index 1b27d0232cbd..67358adf5db3 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -67,13 +67,11 @@ err:
 	return false;
 }
 
-static void do_test(const struct network_helper_opts *opts,
-		    const struct network_helper_opts *cli_opts,
-		    const struct bpf_map *sk_stg_map)
+static void do_test(const struct network_helper_opts *opts)
 {
 	int lfd = -1, fd = -1;
 
-	if (!start_test(NULL, opts, cli_opts, &lfd, &fd))
+	if (!start_test(NULL, opts, opts, &lfd, &fd))
 		goto done;
 
 	ASSERT_OK(send_recv_data(lfd, fd, total_bytes), "send_recv_data");
@@ -114,7 +112,7 @@ static void test_cubic(void)
 		return;
 	}
 
-	do_test(&opts, &opts, NULL);
+	do_test(&opts);
 
 	ASSERT_EQ(cubic_skel->bss->bpf_cubic_acked_called, 1, "pkts_acked called");
 
@@ -382,14 +380,14 @@ static void test_update_ca(void)
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
 	ASSERT_OK_PTR(link, "attach_struct_ops");
 
-	do_test(&opts, &opts, NULL);
+	do_test(&opts);
 	saved_ca1_cnt = skel->bss->ca1_cnt;
 	ASSERT_GT(saved_ca1_cnt, 0, "ca1_ca1_cnt");
 
 	err = bpf_link__update_map(link, skel->maps.ca_update_2);
 	ASSERT_OK(err, "update_map");
 
-	do_test(&opts, &opts, NULL);
+	do_test(&opts);
 	ASSERT_EQ(skel->bss->ca1_cnt, saved_ca1_cnt, "ca2_ca1_cnt");
 	ASSERT_GT(skel->bss->ca2_cnt, 0, "ca2_ca2_cnt");
 
@@ -418,14 +416,14 @@ static void test_update_wrong(void)
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
 	ASSERT_OK_PTR(link, "attach_struct_ops");
 
-	do_test(&opts, &opts, NULL);
+	do_test(&opts);
 	saved_ca1_cnt = skel->bss->ca1_cnt;
 	ASSERT_GT(saved_ca1_cnt, 0, "ca1_ca1_cnt");
 
 	err = bpf_link__update_map(link, skel->maps.ca_wrong);
 	ASSERT_ERR(err, "update_map");
 
-	do_test(&opts, &opts, NULL);
+	do_test(&opts);
 	ASSERT_GT(skel->bss->ca1_cnt, saved_ca1_cnt, "ca2_ca1_cnt");
 
 	bpf_link__destroy(link);
@@ -455,7 +453,7 @@ static void test_mixed_links(void)
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
 	ASSERT_OK_PTR(link, "attach_struct_ops");
 
-	do_test(&opts, &opts, NULL);
+	do_test(&opts);
 	ASSERT_GT(skel->bss->ca1_cnt, 0, "ca1_ca1_cnt");
 
 	err = bpf_link__update_map(link, skel->maps.ca_no_link);
@@ -562,7 +560,7 @@ static void test_cc_cubic(void)
 		return;
 	}
 
-	do_test(&opts, &opts, NULL);
+	do_test(&opts);
 
 	bpf_link__destroy(link);
 	bpf_cc_cubic__destroy(cc_cubic_skel);
-- 
cgit v1.2.3-58-ga151


From cb6cf0820f22ca36dc8f95cf1bd196e5ec24e69d Mon Sep 17 00:00:00 2001
From: Milan Broz <gmazyland@gmail.com>
Date: Wed, 5 Jun 2024 17:33:40 +0200
Subject: r8152: Set NET_ADDR_STOLEN if using passthru MAC

Some docks support MAC pass-through - MAC address
is taken from another device.

Driver should indicate that with NET_ADDR_STOLEN flag.

This should help to avoid collisions if network interface
names are generated with MAC policy.

Reported and discussed here
https://github.com/systemd/systemd/issues/33104

Signed-off-by: Milan Broz <gmazyland@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20240605153340.25694-1-gmazyland@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/usb/r8152.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 09fe70bc45d4..15e12f46d0ea 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -1774,6 +1774,7 @@ static int vendor_mac_passthru_addr_read(struct r8152 *tp, struct sockaddr *sa)
 		goto amacout;
 	}
 	memcpy(sa->sa_data, buf, 6);
+	tp->netdev->addr_assign_type = NET_ADDR_STOLEN;
 	netif_info(tp, probe, tp->netdev,
 		   "Using pass-thru MAC addr %pM\n", sa->sa_data);
 
-- 
cgit v1.2.3-58-ga151


From a999973236543f0b8f6daeaa7ecba7488c3a593b Mon Sep 17 00:00:00 2001
From: MD Danish Anwar <danishanwar@ti.com>
Date: Wed, 5 Jun 2024 15:22:23 +0530
Subject: net: ti: icssg-prueth: Add multicast filtering support

Add multicast filtering support for ICSSG Driver. Multicast addresses will
be updated by __dev_mc_sync() API. icssg_prueth_add_macst () and
icssg_prueth_del_mcast() will be sync and unsync APIs for the driver
respectively.

To add a mac_address for a port, driver needs to call icssg_fdb_add_del()
and pass the mac_address and BIT(port_id) to the API. The ICSSG firmware
will then configure the rules and allow filtering.

If a mac_address is added to port0 and the same mac_address needs to be
added for port1, driver needs to pass BIT(port0) | BIT(port1) to the
icssg_fdb_add_del() API. If driver just pass BIT(port1) then the entry for
port0 will be overwritten / lost. This is a design constraint on the
firmware side.

To overcome this in the driver, to add any mac_address for let's say portX
driver first checks if the same mac_address is already added for any other
port. If yes driver calls icssg_fdb_add_del() with BIT(portX) |
BIT(other_existing_port). If not, driver calls icssg_fdb_add_del() with
BIT(portX).

The same thing is applicable for deleting mac_addresses as well. This
logic is in icssg_prueth_add_mcast / icssg_prueth_del_mcast APIs.

Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/ti/icssg/icssg_prueth.c | 38 +++++++++++++++++++++++++---
 1 file changed, 34 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
index 6e65aa0977d4..e13835100754 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
@@ -439,6 +439,37 @@ const struct icss_iep_clockops prueth_iep_clockops = {
 	.perout_enable = prueth_perout_enable,
 };
 
+static int icssg_prueth_add_mcast(struct net_device *ndev, const u8 *addr)
+{
+	struct prueth_emac *emac = netdev_priv(ndev);
+	int port_mask = BIT(emac->port_id);
+
+	port_mask |= icssg_fdb_lookup(emac, addr, 0);
+	icssg_fdb_add_del(emac, addr, 0, port_mask, true);
+	icssg_vtbl_modify(emac, 0, port_mask, port_mask, true);
+
+	return 0;
+}
+
+static int icssg_prueth_del_mcast(struct net_device *ndev, const u8 *addr)
+{
+	struct prueth_emac *emac = netdev_priv(ndev);
+	int port_mask = BIT(emac->port_id);
+	int other_port_mask;
+
+	other_port_mask = port_mask ^ icssg_fdb_lookup(emac, addr, 0);
+
+	icssg_fdb_add_del(emac, addr, 0, port_mask, false);
+	icssg_vtbl_modify(emac, 0, port_mask, port_mask, false);
+
+	if (other_port_mask) {
+		icssg_fdb_add_del(emac, addr, 0, other_port_mask, true);
+		icssg_vtbl_modify(emac, 0, other_port_mask, other_port_mask, true);
+	}
+
+	return 0;
+}
+
 /**
  * emac_ndo_open - EMAC device open
  * @ndev: network adapter device
@@ -599,6 +630,8 @@ static int emac_ndo_stop(struct net_device *ndev)
 
 	icssg_class_disable(prueth->miig_rt, prueth_emac_slice(emac));
 
+	__dev_mc_unsync(ndev, icssg_prueth_del_mcast);
+
 	atomic_set(&emac->tdown_cnt, emac->tx_ch_num);
 	/* ensure new tdown_cnt value is visible */
 	smp_mb__after_atomic();
@@ -675,10 +708,7 @@ static void emac_ndo_set_rx_mode_work(struct work_struct *work)
 		return;
 	}
 
-	if (!netdev_mc_empty(ndev)) {
-		emac_set_port_state(emac, ICSSG_EMAC_PORT_MC_FLOODING_ENABLE);
-		return;
-	}
+	__dev_mc_sync(ndev, icssg_prueth_add_mcast, icssg_prueth_del_mcast);
 }
 
 /**
-- 
cgit v1.2.3-58-ga151


From 3e61103b2f7887af0be402a79b9c70425ceba3e3 Mon Sep 17 00:00:00 2001
From: David Wei <dw@davidwei.uk>
Date: Wed, 5 Jun 2024 09:19:24 -0700
Subject: page_pool: remove WARN_ON() with OR

Having an OR in WARN_ON() makes me sad because it's impossible to tell
which condition is true when triggered.

Split a WARN_ON() with an OR in page_pool_disable_direct_recycling().

Signed-off-by: David Wei <dw@davidwei.uk>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/page_pool.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index f4444b4e39e6..3927a0a7fa9a 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -1027,8 +1027,8 @@ static void page_pool_disable_direct_recycling(struct page_pool *pool)
 	/* To avoid races with recycling and additional barriers make sure
 	 * pool and NAPI are unlinked when NAPI is disabled.
 	 */
-	WARN_ON(!test_bit(NAPI_STATE_SCHED, &pool->p.napi->state) ||
-		READ_ONCE(pool->p.napi->list_owner) != -1);
+	WARN_ON(!test_bit(NAPI_STATE_SCHED, &pool->p.napi->state));
+	WARN_ON(READ_ONCE(pool->p.napi->list_owner) != -1);
 
 	WRITE_ONCE(pool->p.napi, NULL);
 }
-- 
cgit v1.2.3-58-ga151


From 924ee531754827bf23eacf35e6408a7f44fb9021 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 5 Jun 2024 10:16:44 -0700
Subject: tools: ynl: make user space policies const

Dan, who's working on C++ YNL, pointed out that the C code
does not make policies const. Sprinkle some 'const's around.

Reported-by: Dan Melnic <dmm@meta.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/net/ynl/lib/ynl-priv.h |  8 ++++----
 tools/net/ynl/lib/ynl.c      | 10 +++++-----
 tools/net/ynl/lib/ynl.h      |  2 +-
 tools/net/ynl/ynl-gen-c.py   |  6 +++---
 4 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/tools/net/ynl/lib/ynl-priv.h b/tools/net/ynl/lib/ynl-priv.h
index 80791c34730c..3c09a7bbfba5 100644
--- a/tools/net/ynl/lib/ynl-priv.h
+++ b/tools/net/ynl/lib/ynl-priv.h
@@ -45,17 +45,17 @@ struct ynl_policy_attr {
 	enum ynl_policy_type type;
 	unsigned int len;
 	const char *name;
-	struct ynl_policy_nest *nest;
+	const struct ynl_policy_nest *nest;
 };
 
 struct ynl_policy_nest {
 	unsigned int max_attr;
-	struct ynl_policy_attr *table;
+	const struct ynl_policy_attr *table;
 };
 
 struct ynl_parse_arg {
 	struct ynl_sock *ys;
-	struct ynl_policy_nest *rsp_policy;
+	const struct ynl_policy_nest *rsp_policy;
 	void *data;
 };
 
@@ -119,7 +119,7 @@ struct ynl_dump_state {
 };
 
 struct ynl_ntf_info {
-	struct ynl_policy_nest *policy;
+	const struct ynl_policy_nest *policy;
 	ynl_parse_cb_t cb;
 	size_t alloc_sz;
 	void (*free)(struct ynl_ntf_base_type *ntf);
diff --git a/tools/net/ynl/lib/ynl.c b/tools/net/ynl/lib/ynl.c
index 4b9c091fc86b..fcb18a5a6d70 100644
--- a/tools/net/ynl/lib/ynl.c
+++ b/tools/net/ynl/lib/ynl.c
@@ -46,7 +46,7 @@
 
 /* -- Netlink boiler plate */
 static int
-ynl_err_walk_report_one(struct ynl_policy_nest *policy, unsigned int type,
+ynl_err_walk_report_one(const struct ynl_policy_nest *policy, unsigned int type,
 			char *str, int str_sz, int *n)
 {
 	if (!policy) {
@@ -75,8 +75,8 @@ ynl_err_walk_report_one(struct ynl_policy_nest *policy, unsigned int type,
 
 static int
 ynl_err_walk(struct ynl_sock *ys, void *start, void *end, unsigned int off,
-	     struct ynl_policy_nest *policy, char *str, int str_sz,
-	     struct ynl_policy_nest **nest_pol)
+	     const struct ynl_policy_nest *policy, char *str, int str_sz,
+	     const struct ynl_policy_nest **nest_pol)
 {
 	unsigned int astart_off, aend_off;
 	const struct nlattr *attr;
@@ -206,7 +206,7 @@ ynl_ext_ack_check(struct ynl_sock *ys, const struct nlmsghdr *nlh,
 		bad_attr[n] = '\0';
 	}
 	if (tb[NLMSGERR_ATTR_MISS_TYPE]) {
-		struct ynl_policy_nest *nest_pol = NULL;
+		const struct ynl_policy_nest *nest_pol = NULL;
 		unsigned int n, off, type;
 		void *start, *end;
 		int n2;
@@ -296,7 +296,7 @@ static int ynl_cb_done(const struct nlmsghdr *nlh, struct ynl_parse_arg *yarg)
 
 int ynl_attr_validate(struct ynl_parse_arg *yarg, const struct nlattr *attr)
 {
-	struct ynl_policy_attr *policy;
+	const struct ynl_policy_attr *policy;
 	unsigned int type, len;
 	unsigned char *data;
 
diff --git a/tools/net/ynl/lib/ynl.h b/tools/net/ynl/lib/ynl.h
index eef7c6324ed4..6cd570b283ea 100644
--- a/tools/net/ynl/lib/ynl.h
+++ b/tools/net/ynl/lib/ynl.h
@@ -76,7 +76,7 @@ struct ynl_sock {
 	struct ynl_ntf_base_type **ntf_last_next;
 
 	struct nlmsghdr *nlh;
-	struct ynl_policy_nest *req_policy;
+	const struct ynl_policy_nest *req_policy;
 	unsigned char *tx_buf;
 	unsigned char *rx_buf;
 	unsigned char raw_buf[];
diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py
index a42d62b23ee0..374ca5e86e24 100755
--- a/tools/net/ynl/ynl-gen-c.py
+++ b/tools/net/ynl/ynl-gen-c.py
@@ -1507,12 +1507,12 @@ def print_dump_prototype(ri):
 
 
 def put_typol_fwd(cw, struct):
-    cw.p(f'extern struct ynl_policy_nest {struct.render_name}_nest;')
+    cw.p(f'extern const struct ynl_policy_nest {struct.render_name}_nest;')
 
 
 def put_typol(cw, struct):
     type_max = struct.attr_set.max_name
-    cw.block_start(line=f'struct ynl_policy_attr {struct.render_name}_policy[{type_max} + 1] =')
+    cw.block_start(line=f'const struct ynl_policy_attr {struct.render_name}_policy[{type_max} + 1] =')
 
     for _, arg in struct.member_list():
         arg.attr_typol(cw)
@@ -1520,7 +1520,7 @@ def put_typol(cw, struct):
     cw.block_end(line=';')
     cw.nl()
 
-    cw.block_start(line=f'struct ynl_policy_nest {struct.render_name}_nest =')
+    cw.block_start(line=f'const struct ynl_policy_nest {struct.render_name}_nest =')
     cw.p(f'.max_attr = {type_max},')
     cw.p(f'.table = {struct.render_name}_policy,')
     cw.block_end(line=';')
-- 
cgit v1.2.3-58-ga151


From 28f961f9d5b7c3d9b9f93cc59e54477ba1278cf9 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Thu, 6 Jun 2024 17:23:44 +0300
Subject: dmaengine: ti: k3-udma-glue: clean up return in
 k3_udma_glue_rx_get_irq()

Currently the k3_udma_glue_rx_get_irq() function returns either negative
error codes or zero on error.  Generally, in the kernel, zero means
success so this be confusing and has caused bugs in the past.  Also the
"tx" version of this function only returns negative error codes.  Let's
clean this "rx" function so both functions match.

This patch has no effect on runtime.

Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Acked-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/dma/ti/k3-udma-glue.c                | 3 +++
 drivers/net/ethernet/ti/am65-cpsw-nuss.c     | 4 ++--
 drivers/net/ethernet/ti/icssg/icssg_common.c | 4 +---
 3 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/dma/ti/k3-udma-glue.c b/drivers/dma/ti/k3-udma-glue.c
index c9b93055dc9d..b96b448a0e69 100644
--- a/drivers/dma/ti/k3-udma-glue.c
+++ b/drivers/dma/ti/k3-udma-glue.c
@@ -1531,6 +1531,9 @@ int k3_udma_glue_rx_get_irq(struct k3_udma_glue_rx_channel *rx_chn,
 		flow->virq = k3_ringacc_get_ring_irq_num(flow->ringrx);
 	}
 
+	if (!flow->virq)
+		return -ENXIO;
+
 	return flow->virq;
 }
 EXPORT_SYMBOL_GPL(k3_udma_glue_rx_get_irq);
diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
index e6f87ac394fe..81d9f21086ec 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c
@@ -2424,10 +2424,10 @@ static int am65_cpsw_nuss_init_rx_chns(struct am65_cpsw_common *common)
 
 		rx_chn->irq = k3_udma_glue_rx_get_irq(rx_chn->rx_chn, i);
 
-		if (rx_chn->irq <= 0) {
+		if (rx_chn->irq < 0) {
 			dev_err(dev, "Failed to get rx dma irq %d\n",
 				rx_chn->irq);
-			ret = -ENXIO;
+			ret = rx_chn->irq;
 			goto err;
 		}
 	}
diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c
index 873126dfc173..2f716c0d7060 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_common.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_common.c
@@ -440,9 +440,7 @@ int prueth_init_rx_chns(struct prueth_emac *emac,
 			fdqring_id = k3_udma_glue_rx_flow_get_fdq_id(rx_chn->rx_chn,
 								     i);
 		ret = k3_udma_glue_rx_get_irq(rx_chn->rx_chn, i);
-		if (ret <= 0) {
-			if (!ret)
-				ret = -ENXIO;
+		if (ret < 0) {
 			netdev_err(ndev, "Failed to get rx dma irq");
 			goto fail;
 		}
-- 
cgit v1.2.3-58-ga151


From c1e156ae50ee520baf4e560b1284677ddfcea974 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Thu, 6 Jun 2024 16:49:38 +0200
Subject: lib: objagg: Fix spelling

Fixes: 0a020d416d0a ("lib: introduce initial implementation of object aggregation manager")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Amit Cohen <amcohen@nvidia.com>
Tested-by: Alexander Zubkov <green@qrator.net>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 lib/objagg.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/objagg.c b/lib/objagg.c
index 1e248629ed64..955538c90223 100644
--- a/lib/objagg.c
+++ b/lib/objagg.c
@@ -421,7 +421,7 @@ static struct objagg_obj *__objagg_obj_get(struct objagg *objagg, void *obj)
  *
  * There are 3 main options this function wraps:
  * 1) The object according to "obj" already exist. In that case
- *    the reference counter is incrementes and the object is returned.
+ *    the reference counter is incremented and the object is returned.
  * 2) The object does not exist, but it can be aggregated within
  *    another object. In that case, user ops->delta_create() is called
  *    to obtain delta data and a new object is created with returned
-- 
cgit v1.2.3-58-ga151


From 2aad28ec4543f4c3154f904bcd817e649ffb0ba6 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Thu, 6 Jun 2024 16:49:39 +0200
Subject: lib: test_objagg: Fix spelling

Fixes: 0a020d416d0a ("lib: introduce initial implementation of object aggregation manager")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Amit Cohen <amcohen@nvidia.com>
Tested-by: Alexander Zubkov <green@qrator.net>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 lib/test_objagg.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/test_objagg.c b/lib/test_objagg.c
index c0c957c50635..d34df4306b87 100644
--- a/lib/test_objagg.c
+++ b/lib/test_objagg.c
@@ -60,7 +60,7 @@ static struct objagg_obj *world_obj_get(struct world *world,
 	if (!world->key_refs[key_id_index(key_id)]) {
 		world->objagg_objs[key_id_index(key_id)] = objagg_obj;
 	} else if (world->objagg_objs[key_id_index(key_id)] != objagg_obj) {
-		pr_err("Key %u: God another object for the same key.\n",
+		pr_err("Key %u: Got another object for the same key.\n",
 		       key_id);
 		err = -EINVAL;
 		goto err_key_id_check;
-- 
cgit v1.2.3-58-ga151


From 06fcdf2494068ceb29686657f8e5d15084e6e0c3 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Thu, 6 Jun 2024 16:49:40 +0200
Subject: mlxsw: spectrum_acl_atcam: Fix wrong comment

The key is encoded, not encrypted.

Fixes: c22291f7cf45 ("mlxsw: spectrum: acl: Implement delta for ERP")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Amit Cohen <amcohen@nvidia.com>
Tested-by: Alexander Zubkov <green@qrator.net>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_atcam.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_atcam.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_atcam.c
index 4b713832fdd5..a7473e782b56 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_atcam.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_atcam.c
@@ -491,7 +491,7 @@ __mlxsw_sp_acl_atcam_entry_add(struct mlxsw_sp *mlxsw_sp,
 	       sizeof(aentry->enc_key));
 
 	/* Compute all needed delta information and clear the delta bits
-	 * from the encrypted key.
+	 * from the encoded key.
 	 */
 	delta = mlxsw_sp_acl_erp_delta(aentry->erp_mask);
 	aentry->delta_info.start = mlxsw_sp_acl_erp_delta_start(delta);
-- 
cgit v1.2.3-58-ga151


From b4a3a89fffcdf09702b1f161b914e52abca1894d Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Thu, 6 Jun 2024 16:49:41 +0200
Subject: lib: objagg: Fix general protection fault

The library supports aggregation of objects into other objects only if
the parent object does not have a parent itself. That is, nesting is not
supported.

Aggregation happens in two cases: Without and with hints, where hints
are a pre-computed recommendation on how to aggregate the provided
objects.

Nesting is not possible in the first case due to a check that prevents
it, but in the second case there is no check because the assumption is
that nesting cannot happen when creating objects based on hints. The
violation of this assumption leads to various warnings and eventually to
a general protection fault [1].

Before fixing the root cause, error out when nesting happens and warn.

[1]
general protection fault, probably for non-canonical address 0xdead000000000d90: 0000 [#1] PREEMPT SMP PTI
CPU: 1 PID: 1083 Comm: kworker/1:9 Tainted: G        W          6.9.0-rc6-custom-gd9b4f1cca7fb #7
Hardware name: Mellanox Technologies Ltd. MSN3700/VMOD0005, BIOS 5.11 01/06/2019
Workqueue: mlxsw_core mlxsw_sp_acl_tcam_vregion_rehash_work
RIP: 0010:mlxsw_sp_acl_erp_bf_insert+0x25/0x80
[...]
Call Trace:
 <TASK>
 mlxsw_sp_acl_atcam_entry_add+0x256/0x3c0
 mlxsw_sp_acl_tcam_entry_create+0x5e/0xa0
 mlxsw_sp_acl_tcam_vchunk_migrate_one+0x16b/0x270
 mlxsw_sp_acl_tcam_vregion_rehash_work+0xbe/0x510
 process_one_work+0x151/0x370
 worker_thread+0x2cb/0x3e0
 kthread+0xd0/0x100
 ret_from_fork+0x34/0x50
 ret_from_fork_asm+0x1a/0x30
 </TASK>

Fixes: 9069a3817d82 ("lib: objagg: implement optimization hints assembly and use hints for object creation")
Reported-by: Alexander Zubkov <green@qrator.net>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Amit Cohen <amcohen@nvidia.com>
Tested-by: Alexander Zubkov <green@qrator.net>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 lib/objagg.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/lib/objagg.c b/lib/objagg.c
index 955538c90223..0f99ea5f5371 100644
--- a/lib/objagg.c
+++ b/lib/objagg.c
@@ -167,6 +167,9 @@ static int objagg_obj_parent_assign(struct objagg *objagg,
 {
 	void *delta_priv;
 
+	if (WARN_ON(!objagg_obj_is_root(parent)))
+		return -EINVAL;
+
 	delta_priv = objagg->ops->delta_create(objagg->priv, parent->obj,
 					       objagg_obj->obj);
 	if (IS_ERR(delta_priv))
-- 
cgit v1.2.3-58-ga151


From 97d833ceb27dc19f8777d63f90be4a27b5daeedf Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Thu, 6 Jun 2024 16:49:42 +0200
Subject: mlxsw: spectrum_acl_erp: Fix object nesting warning

ACLs in Spectrum-2 and newer ASICs can reside in the algorithmic TCAM
(A-TCAM) or in the ordinary circuit TCAM (C-TCAM). The former can
contain more ACLs (i.e., tc filters), but the number of masks in each
region (i.e., tc chain) is limited.

In order to mitigate the effects of the above limitation, the device
allows filters to share a single mask if their masks only differ in up
to 8 consecutive bits. For example, dst_ip/25 can be represented using
dst_ip/24 with a delta of 1 bit. The C-TCAM does not have a limit on the
number of masks being used (and therefore does not support mask
aggregation), but can contain a limited number of filters.

The driver uses the "objagg" library to perform the mask aggregation by
passing it objects that consist of the filter's mask and whether the
filter is to be inserted into the A-TCAM or the C-TCAM since filters in
different TCAMs cannot share a mask.

The set of created objects is dependent on the insertion order of the
filters and is not necessarily optimal. Therefore, the driver will
periodically ask the library to compute a more optimal set ("hints") by
looking at all the existing objects.

When the library asks the driver whether two objects can be aggregated
the driver only compares the provided masks and ignores the A-TCAM /
C-TCAM indication. This is the right thing to do since the goal is to
move as many filters as possible to the A-TCAM. The driver also forbids
two identical masks from being aggregated since this can only happen if
one was intentionally put in the C-TCAM to avoid a conflict in the
A-TCAM.

The above can result in the following set of hints:

H1: {mask X, A-TCAM} -> H2: {mask Y, A-TCAM} // X is Y + delta
H3: {mask Y, C-TCAM} -> H4: {mask Z, A-TCAM} // Y is Z + delta

After getting the hints from the library the driver will start migrating
filters from one region to another while consulting the computed hints
and instructing the device to perform a lookup in both regions during
the transition.

Assuming a filter with mask X is being migrated into the A-TCAM in the
new region, the hints lookup will return H1. Since H2 is the parent of
H1, the library will try to find the object associated with it and
create it if necessary in which case another hints lookup (recursive)
will be performed. This hints lookup for {mask Y, A-TCAM} will either
return H2 or H3 since the driver passes the library an object comparison
function that ignores the A-TCAM / C-TCAM indication.

This can eventually lead to nested objects which are not supported by
the library [1].

Fix by removing the object comparison function from both the driver and
the library as the driver was the only user. That way the lookup will
only return exact matches.

I do not have a reliable reproducer that can reproduce the issue in a
timely manner, but before the fix the issue would reproduce in several
minutes and with the fix it does not reproduce in over an hour.

Note that the current usefulness of the hints is limited because they
include the C-TCAM indication and represent aggregation that cannot
actually happen. This will be addressed in net-next.

[1]
WARNING: CPU: 0 PID: 153 at lib/objagg.c:170 objagg_obj_parent_assign+0xb5/0xd0
Modules linked in:
CPU: 0 PID: 153 Comm: kworker/0:18 Not tainted 6.9.0-rc6-custom-g70fbc2c1c38b #42
Hardware name: Mellanox Technologies Ltd. MSN3700C/VMOD0008, BIOS 5.11 10/10/2018
Workqueue: mlxsw_core mlxsw_sp_acl_tcam_vregion_rehash_work
RIP: 0010:objagg_obj_parent_assign+0xb5/0xd0
[...]
Call Trace:
 <TASK>
 __objagg_obj_get+0x2bb/0x580
 objagg_obj_get+0xe/0x80
 mlxsw_sp_acl_erp_mask_get+0xb5/0xf0
 mlxsw_sp_acl_atcam_entry_add+0xe8/0x3c0
 mlxsw_sp_acl_tcam_entry_create+0x5e/0xa0
 mlxsw_sp_acl_tcam_vchunk_migrate_one+0x16b/0x270
 mlxsw_sp_acl_tcam_vregion_rehash_work+0xbe/0x510
 process_one_work+0x151/0x370

Fixes: 9069a3817d82 ("lib: objagg: implement optimization hints assembly and use hints for object creation")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Amit Cohen <amcohen@nvidia.com>
Tested-by: Alexander Zubkov <green@qrator.net>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_erp.c | 13 -------------
 include/linux/objagg.h                                 |  1 -
 lib/objagg.c                                           | 15 ---------------
 3 files changed, 29 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_erp.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_erp.c
index d231f4d2888b..9eee229303cc 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_erp.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_erp.c
@@ -1217,18 +1217,6 @@ static bool mlxsw_sp_acl_erp_delta_check(void *priv, const void *parent_obj,
 	return err ? false : true;
 }
 
-static int mlxsw_sp_acl_erp_hints_obj_cmp(const void *obj1, const void *obj2)
-{
-	const struct mlxsw_sp_acl_erp_key *key1 = obj1;
-	const struct mlxsw_sp_acl_erp_key *key2 = obj2;
-
-	/* For hints purposes, two objects are considered equal
-	 * in case the masks are the same. Does not matter what
-	 * the "ctcam" value is.
-	 */
-	return memcmp(key1->mask, key2->mask, sizeof(key1->mask));
-}
-
 static void *mlxsw_sp_acl_erp_delta_create(void *priv, void *parent_obj,
 					   void *obj)
 {
@@ -1308,7 +1296,6 @@ static void mlxsw_sp_acl_erp_root_destroy(void *priv, void *root_priv)
 static const struct objagg_ops mlxsw_sp_acl_erp_objagg_ops = {
 	.obj_size = sizeof(struct mlxsw_sp_acl_erp_key),
 	.delta_check = mlxsw_sp_acl_erp_delta_check,
-	.hints_obj_cmp = mlxsw_sp_acl_erp_hints_obj_cmp,
 	.delta_create = mlxsw_sp_acl_erp_delta_create,
 	.delta_destroy = mlxsw_sp_acl_erp_delta_destroy,
 	.root_create = mlxsw_sp_acl_erp_root_create,
diff --git a/include/linux/objagg.h b/include/linux/objagg.h
index 78021777df46..6df5b887dc54 100644
--- a/include/linux/objagg.h
+++ b/include/linux/objagg.h
@@ -8,7 +8,6 @@ struct objagg_ops {
 	size_t obj_size;
 	bool (*delta_check)(void *priv, const void *parent_obj,
 			    const void *obj);
-	int (*hints_obj_cmp)(const void *obj1, const void *obj2);
 	void * (*delta_create)(void *priv, void *parent_obj, void *obj);
 	void (*delta_destroy)(void *priv, void *delta_priv);
 	void * (*root_create)(void *priv, void *obj, unsigned int root_id);
diff --git a/lib/objagg.c b/lib/objagg.c
index 0f99ea5f5371..363e43e849ac 100644
--- a/lib/objagg.c
+++ b/lib/objagg.c
@@ -906,20 +906,6 @@ static const struct objagg_opt_algo *objagg_opt_algos[] = {
 	[OBJAGG_OPT_ALGO_SIMPLE_GREEDY] = &objagg_opt_simple_greedy,
 };
 
-static int objagg_hints_obj_cmp(struct rhashtable_compare_arg *arg,
-				const void *obj)
-{
-	struct rhashtable *ht = arg->ht;
-	struct objagg_hints *objagg_hints =
-			container_of(ht, struct objagg_hints, node_ht);
-	const struct objagg_ops *ops = objagg_hints->ops;
-	const char *ptr = obj;
-
-	ptr += ht->p.key_offset;
-	return ops->hints_obj_cmp ? ops->hints_obj_cmp(ptr, arg->key) :
-				    memcmp(ptr, arg->key, ht->p.key_len);
-}
-
 /**
  * objagg_hints_get - obtains hints instance
  * @objagg:		objagg instance
@@ -958,7 +944,6 @@ struct objagg_hints *objagg_hints_get(struct objagg *objagg,
 				offsetof(struct objagg_hints_node, obj);
 	objagg_hints->ht_params.head_offset =
 				offsetof(struct objagg_hints_node, ht_node);
-	objagg_hints->ht_params.obj_cmpfn = objagg_hints_obj_cmp;
 
 	err = rhashtable_init(&objagg_hints->node_ht, &objagg_hints->ht_params);
 	if (err)
-- 
cgit v1.2.3-58-ga151


From 75d8d7a63065b18df9555dbaab0b42d4c6f20943 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Thu, 6 Jun 2024 16:49:43 +0200
Subject: mlxsw: spectrum_acl: Fix ACL scale regression and firmware errors

ACLs that reside in the algorithmic TCAM (A-TCAM) in Spectrum-2 and
newer ASICs can share the same mask if their masks only differ in up to
8 consecutive bits. For example, consider the following filters:

 # tc filter add dev swp1 ingress pref 1 proto ip flower dst_ip 192.0.2.0/24 action drop
 # tc filter add dev swp1 ingress pref 1 proto ip flower dst_ip 198.51.100.128/25 action drop

The second filter can use the same mask as the first (dst_ip/24) with a
delta of 1 bit.

However, the above only works because the two filters have different
values in the common unmasked part (dst_ip/24). When entries have the
same value in the common unmasked part they create undesired collisions
in the device since many entries now have the same key. This leads to
firmware errors such as [1] and to a reduced scale.

Fix by adjusting the hash table key to only include the value in the
common unmasked part. That is, without including the delta bits. That
way the driver will detect the collision during filter insertion and
spill the filter into the circuit TCAM (C-TCAM).

Add a test case that fails without the fix and adjust existing cases
that check C-TCAM spillage according to the above limitation.

[1]
mlxsw_spectrum2 0000:06:00.0: EMAD reg access failed (tid=3379b18a00003394,reg_id=3027(ptce3),type=write,status=8(resource not available))

Fixes: c22291f7cf45 ("mlxsw: spectrum: acl: Implement delta for ERP")
Reported-by: Alexander Zubkov <green@qrator.net>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Amit Cohen <amcohen@nvidia.com>
Tested-by: Alexander Zubkov <green@qrator.net>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../ethernet/mellanox/mlxsw/spectrum_acl_atcam.c   | 18 ++++---
 .../mellanox/mlxsw/spectrum_acl_bloom_filter.c     |  2 +-
 .../ethernet/mellanox/mlxsw/spectrum_acl_tcam.h    |  9 ++--
 .../drivers/net/mlxsw/spectrum-2/tc_flower.sh      | 55 ++++++++++++++++++++--
 4 files changed, 63 insertions(+), 21 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_atcam.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_atcam.c
index a7473e782b56..07cb1e26ca3e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_atcam.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_atcam.c
@@ -391,7 +391,8 @@ mlxsw_sp_acl_atcam_region_entry_insert(struct mlxsw_sp *mlxsw_sp,
 	if (err)
 		return err;
 
-	lkey_id = aregion->ops->lkey_id_get(aregion, aentry->enc_key, erp_id);
+	lkey_id = aregion->ops->lkey_id_get(aregion, aentry->ht_key.enc_key,
+					    erp_id);
 	if (IS_ERR(lkey_id))
 		return PTR_ERR(lkey_id);
 	aentry->lkey_id = lkey_id;
@@ -399,7 +400,7 @@ mlxsw_sp_acl_atcam_region_entry_insert(struct mlxsw_sp *mlxsw_sp,
 	kvdl_index = mlxsw_afa_block_first_kvdl_index(rulei->act_block);
 	mlxsw_reg_ptce3_pack(ptce3_pl, true, MLXSW_REG_PTCE3_OP_WRITE_WRITE,
 			     priority, region->tcam_region_info,
-			     aentry->enc_key, erp_id,
+			     aentry->ht_key.enc_key, erp_id,
 			     aentry->delta_info.start,
 			     aentry->delta_info.mask,
 			     aentry->delta_info.value,
@@ -428,7 +429,7 @@ mlxsw_sp_acl_atcam_region_entry_remove(struct mlxsw_sp *mlxsw_sp,
 
 	mlxsw_reg_ptce3_pack(ptce3_pl, false, MLXSW_REG_PTCE3_OP_WRITE_WRITE, 0,
 			     region->tcam_region_info,
-			     aentry->enc_key, erp_id,
+			     aentry->ht_key.enc_key, erp_id,
 			     aentry->delta_info.start,
 			     aentry->delta_info.mask,
 			     aentry->delta_info.value,
@@ -457,7 +458,7 @@ mlxsw_sp_acl_atcam_region_entry_action_replace(struct mlxsw_sp *mlxsw_sp,
 	kvdl_index = mlxsw_afa_block_first_kvdl_index(rulei->act_block);
 	mlxsw_reg_ptce3_pack(ptce3_pl, true, MLXSW_REG_PTCE3_OP_WRITE_UPDATE,
 			     priority, region->tcam_region_info,
-			     aentry->enc_key, erp_id,
+			     aentry->ht_key.enc_key, erp_id,
 			     aentry->delta_info.start,
 			     aentry->delta_info.mask,
 			     aentry->delta_info.value,
@@ -480,15 +481,13 @@ __mlxsw_sp_acl_atcam_entry_add(struct mlxsw_sp *mlxsw_sp,
 	int err;
 
 	mlxsw_afk_encode(afk, region->key_info, &rulei->values,
-			 aentry->ht_key.full_enc_key, mask);
+			 aentry->ht_key.enc_key, mask);
 
 	erp_mask = mlxsw_sp_acl_erp_mask_get(aregion, mask, false);
 	if (IS_ERR(erp_mask))
 		return PTR_ERR(erp_mask);
 	aentry->erp_mask = erp_mask;
 	aentry->ht_key.erp_id = mlxsw_sp_acl_erp_mask_erp_id(erp_mask);
-	memcpy(aentry->enc_key, aentry->ht_key.full_enc_key,
-	       sizeof(aentry->enc_key));
 
 	/* Compute all needed delta information and clear the delta bits
 	 * from the encoded key.
@@ -497,9 +496,8 @@ __mlxsw_sp_acl_atcam_entry_add(struct mlxsw_sp *mlxsw_sp,
 	aentry->delta_info.start = mlxsw_sp_acl_erp_delta_start(delta);
 	aentry->delta_info.mask = mlxsw_sp_acl_erp_delta_mask(delta);
 	aentry->delta_info.value =
-		mlxsw_sp_acl_erp_delta_value(delta,
-					     aentry->ht_key.full_enc_key);
-	mlxsw_sp_acl_erp_delta_clear(delta, aentry->enc_key);
+		mlxsw_sp_acl_erp_delta_value(delta, aentry->ht_key.enc_key);
+	mlxsw_sp_acl_erp_delta_clear(delta, aentry->ht_key.enc_key);
 
 	/* Add rule to the list of A-TCAM rules, assuming this
 	 * rule is intended to A-TCAM. In case this rule does
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_bloom_filter.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_bloom_filter.c
index 95f63fcf4ba1..a54eedb69a3f 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_bloom_filter.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_bloom_filter.c
@@ -249,7 +249,7 @@ __mlxsw_sp_acl_bf_key_encode(struct mlxsw_sp_acl_atcam_region *aregion,
 		memcpy(chunk + pad_bytes, &erp_region_id,
 		       sizeof(erp_region_id));
 		memcpy(chunk + key_offset,
-		       &aentry->enc_key[chunk_key_offsets[chunk_index]],
+		       &aentry->ht_key.enc_key[chunk_key_offsets[chunk_index]],
 		       chunk_key_len);
 		chunk += chunk_len;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h
index 79a1d8606512..010204f73ea4 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h
@@ -167,9 +167,9 @@ struct mlxsw_sp_acl_atcam_region {
 };
 
 struct mlxsw_sp_acl_atcam_entry_ht_key {
-	char full_enc_key[MLXSW_REG_PTCEX_FLEX_KEY_BLOCKS_LEN]; /* Encoded
-								 * key.
-								 */
+	char enc_key[MLXSW_REG_PTCEX_FLEX_KEY_BLOCKS_LEN]; /* Encoded key, minus
+							    * delta bits.
+							    */
 	u8 erp_id;
 };
 
@@ -181,9 +181,6 @@ struct mlxsw_sp_acl_atcam_entry {
 	struct rhash_head ht_node;
 	struct list_head list; /* Member in entries_list */
 	struct mlxsw_sp_acl_atcam_entry_ht_key ht_key;
-	char enc_key[MLXSW_REG_PTCEX_FLEX_KEY_BLOCKS_LEN]; /* Encoded key,
-							    * minus delta bits.
-							    */
 	struct {
 		u16 start;
 		u8 mask;
diff --git a/tools/testing/selftests/drivers/net/mlxsw/spectrum-2/tc_flower.sh b/tools/testing/selftests/drivers/net/mlxsw/spectrum-2/tc_flower.sh
index 31252bc8775e..4994bea5daf8 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/spectrum-2/tc_flower.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/spectrum-2/tc_flower.sh
@@ -11,7 +11,7 @@ ALL_TESTS="single_mask_test identical_filters_test two_masks_test \
 	multiple_masks_test ctcam_edge_cases_test delta_simple_test \
 	delta_two_masks_one_key_test delta_simple_rehash_test \
 	bloom_simple_test bloom_complex_test bloom_delta_test \
-	max_erp_entries_test max_group_size_test"
+	max_erp_entries_test max_group_size_test collision_test"
 NUM_NETIFS=2
 source $lib_dir/lib.sh
 source $lib_dir/tc_common.sh
@@ -457,7 +457,7 @@ delta_two_masks_one_key_test()
 {
 	# If 2 keys are the same and only differ in mask in a way that
 	# they belong under the same ERP (second is delta of the first),
-	# there should be no C-TCAM spill.
+	# there should be C-TCAM spill.
 
 	RET=0
 
@@ -474,8 +474,8 @@ delta_two_masks_one_key_test()
 	tp_record "mlxsw:*" "tc filter add dev $h2 ingress protocol ip \
 		   pref 2 handle 102 flower $tcflags dst_ip 192.0.2.2 \
 		   action drop"
-	tp_check_hits "mlxsw:mlxsw_sp_acl_atcam_entry_add_ctcam_spill" 0
-	check_err $? "incorrect C-TCAM spill while inserting the second rule"
+	tp_check_hits "mlxsw:mlxsw_sp_acl_atcam_entry_add_ctcam_spill" 1
+	check_err $? "C-TCAM spill did not happen while inserting the second rule"
 
 	$MZ $h1 -c 1 -p 64 -a $h1mac -b $h2mac -A 192.0.2.1 -B 192.0.2.2 \
 		-t ip -q
@@ -1087,6 +1087,53 @@ max_group_size_test()
 	log_test "max ACL group size test ($tcflags). max size $max_size"
 }
 
+collision_test()
+{
+	# Filters cannot share an eRP if in the common unmasked part (i.e.,
+	# without the delta bits) they have the same values. If the driver does
+	# not prevent such configuration (by spilling into the C-TCAM), then
+	# multiple entries will be present in the device with the same key,
+	# leading to collisions and a reduced scale.
+	#
+	# Create such a scenario and make sure all the filters are successfully
+	# added.
+
+	RET=0
+
+	local ret
+
+	if [[ "$tcflags" != "skip_sw" ]]; then
+		return 0;
+	fi
+
+	# Add a single dst_ip/24 filter and multiple dst_ip/32 filters that all
+	# have the same values in the common unmasked part (dst_ip/24).
+
+	tc filter add dev $h2 ingress pref 1 proto ipv4 handle 101 \
+		flower $tcflags dst_ip 198.51.100.0/24 \
+		action drop
+
+	for i in {0..255}; do
+		tc filter add dev $h2 ingress pref 2 proto ipv4 \
+			handle $((102 + i)) \
+			flower $tcflags dst_ip 198.51.100.${i}/32 \
+			action drop
+		ret=$?
+		[[ $ret -ne 0 ]] && break
+	done
+
+	check_err $ret "failed to add all the filters"
+
+	for i in {255..0}; do
+		tc filter del dev $h2 ingress pref 2 proto ipv4 \
+			handle $((102 + i)) flower
+	done
+
+	tc filter del dev $h2 ingress pref 1 proto ipv4 handle 101 flower
+
+	log_test "collision test ($tcflags)"
+}
+
 setup_prepare()
 {
 	h1=${NETIFS[p1]}
-- 
cgit v1.2.3-58-ga151


From b334b924c9b709bc969644fb5c406f5c9d01dceb Mon Sep 17 00:00:00 2001
From: Valentin Schneider <vschneid@redhat.com>
Date: Thu, 6 Jun 2024 17:11:37 +0200
Subject: net: tcp/dccp: prepare for tw_timer un-pinning

The TCP timewait timer is proving to be problematic for setups where
scheduler CPU isolation is achieved at runtime via cpusets (as opposed to
statically via isolcpus=domains).

What happens there is a CPU goes through tcp_time_wait(), arming the
time_wait timer, then gets isolated. TCP_TIMEWAIT_LEN later, the timer
fires, causing interference for the now-isolated CPU. This is conceptually
similar to the issue described in commit e02b93124855 ("workqueue: Unbind
kworkers before sending them to exit()")

Move inet_twsk_schedule() to within inet_twsk_hashdance(), with the ehash
lock held. Expand the lock's critical section from inet_twsk_kill() to
inet_twsk_deschedule_put(), serializing the scheduling vs descheduling of
the timer. IOW, this prevents the following race:

			     tcp_time_wait()
			       inet_twsk_hashdance()
  inet_twsk_deschedule_put()
    del_timer_sync()
			       inet_twsk_schedule()

Thanks to Paolo Abeni for suggesting to leverage the ehash lock.

This also restores a comment from commit ec94c2696f0b ("tcp/dccp: avoid
one atomic operation for timewait hashdance") as inet_twsk_hashdance() had
a "Step 1" and "Step 3" comment, but the "Step 2" had gone missing.

inet_twsk_deschedule_put() now acquires the ehash spinlock to synchronize
with inet_twsk_hashdance_schedule().

To ease possible regression search, actual un-pin is done in next patch.

Link: https://lore.kernel.org/all/ZPhpfMjSiHVjQkTk@localhost.localdomain/
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Valentin Schneider <vschneid@redhat.com>
Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_timewait_sock.h |  6 +++--
 net/dccp/minisocks.c             |  3 +--
 net/ipv4/inet_timewait_sock.c    | 52 ++++++++++++++++++++++++++++++++++------
 net/ipv4/tcp_ipv4.c              |  2 +-
 net/ipv4/tcp_minisocks.c         |  3 +--
 5 files changed, 52 insertions(+), 14 deletions(-)

diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h
index 2a536eea9424..5b43d220243d 100644
--- a/include/net/inet_timewait_sock.h
+++ b/include/net/inet_timewait_sock.h
@@ -93,8 +93,10 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk,
 					   struct inet_timewait_death_row *dr,
 					   const int state);
 
-void inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
-			 struct inet_hashinfo *hashinfo);
+void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw,
+				  struct sock *sk,
+				  struct inet_hashinfo *hashinfo,
+				  int timeo);
 
 void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo,
 			  bool rearm);
diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c
index 251a57cf5822..deb52d7d31b4 100644
--- a/net/dccp/minisocks.c
+++ b/net/dccp/minisocks.c
@@ -59,11 +59,10 @@ void dccp_time_wait(struct sock *sk, int state, int timeo)
 		 * we complete the initialization.
 		 */
 		local_bh_disable();
-		inet_twsk_schedule(tw, timeo);
 		/* Linkage updates.
 		 * Note that access to tw after this point is illegal.
 		 */
-		inet_twsk_hashdance(tw, sk, &dccp_hashinfo);
+		inet_twsk_hashdance_schedule(tw, sk, &dccp_hashinfo, timeo);
 		local_bh_enable();
 	} else {
 		/* Sorry, if we're out of memory, just CLOSE this
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index e28075f0006e..628d33a41ce5 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -96,9 +96,13 @@ static void inet_twsk_add_node_rcu(struct inet_timewait_sock *tw,
  * Enter the time wait state. This is called with locally disabled BH.
  * Essentially we whip up a timewait bucket, copy the relevant info into it
  * from the SK, and mess with hash chains and list linkage.
+ *
+ * The caller must not access @tw anymore after this function returns.
  */
-void inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
-			   struct inet_hashinfo *hashinfo)
+void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw,
+				  struct sock *sk,
+				  struct inet_hashinfo *hashinfo,
+				  int timeo)
 {
 	const struct inet_sock *inet = inet_sk(sk);
 	const struct inet_connection_sock *icsk = inet_csk(sk);
@@ -129,26 +133,33 @@ void inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
 
 	spin_lock(lock);
 
+	/* Step 2: Hash TW into tcp ehash chain */
 	inet_twsk_add_node_rcu(tw, &ehead->chain);
 
 	/* Step 3: Remove SK from hash chain */
 	if (__sk_nulls_del_node_init_rcu(sk))
 		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
 
-	spin_unlock(lock);
 
+	/* Ensure above writes are committed into memory before updating the
+	 * refcount.
+	 * Provides ordering vs later refcount_inc().
+	 */
+	smp_wmb();
 	/* tw_refcnt is set to 3 because we have :
 	 * - one reference for bhash chain.
 	 * - one reference for ehash chain.
 	 * - one reference for timer.
-	 * We can use atomic_set() because prior spin_lock()/spin_unlock()
-	 * committed into memory all tw fields.
 	 * Also note that after this point, we lost our implicit reference
 	 * so we are not allowed to use tw anymore.
 	 */
 	refcount_set(&tw->tw_refcnt, 3);
+
+	inet_twsk_schedule(tw, timeo);
+
+	spin_unlock(lock);
 }
-EXPORT_SYMBOL_GPL(inet_twsk_hashdance);
+EXPORT_SYMBOL_GPL(inet_twsk_hashdance_schedule);
 
 static void tw_timer_handler(struct timer_list *t)
 {
@@ -217,7 +228,34 @@ EXPORT_SYMBOL_GPL(inet_twsk_alloc);
  */
 void inet_twsk_deschedule_put(struct inet_timewait_sock *tw)
 {
-	if (del_timer_sync(&tw->tw_timer))
+	struct inet_hashinfo *hashinfo = tw->tw_dr->hashinfo;
+	spinlock_t *lock = inet_ehash_lockp(hashinfo, tw->tw_hash);
+
+	/* inet_twsk_purge() walks over all sockets, including tw ones,
+	 * and removes them via inet_twsk_deschedule_put() after a
+	 * refcount_inc_not_zero().
+	 *
+	 * inet_twsk_hashdance_schedule() must (re)init the refcount before
+	 * arming the timer, i.e. inet_twsk_purge can obtain a reference to
+	 * a twsk that did not yet schedule the timer.
+	 *
+	 * The ehash lock synchronizes these two:
+	 * After acquiring the lock, the timer is always scheduled (else
+	 * timer_shutdown returns false), because hashdance_schedule releases
+	 * the ehash lock only after completing the timer initialization.
+	 *
+	 * Without grabbing the ehash lock, we get:
+	 * 1) cpu x sets twsk refcount to 3
+	 * 2) cpu y bumps refcount to 4
+	 * 3) cpu y calls inet_twsk_deschedule_put() and shuts timer down
+	 * 4) cpu x tries to start timer, but mod_timer is a noop post-shutdown
+	 * -> timer refcount is never decremented.
+	 */
+	spin_lock(lock);
+	/*  Makes sure hashdance_schedule() has completed */
+	spin_unlock(lock);
+
+	if (timer_shutdown_sync(&tw->tw_timer))
 		inet_twsk_kill(tw);
 	inet_twsk_put(tw);
 }
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 3613e08ca794..e61c7c974745 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -157,7 +157,7 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp)
 	if (ts_recent_stamp &&
 	    (!twp || (reuse && time_after32(ktime_get_seconds(),
 					    ts_recent_stamp)))) {
-		/* inet_twsk_hashdance() sets sk_refcnt after putting twsk
+		/* inet_twsk_hashdance_schedule() sets sk_refcnt after putting twsk
 		 * and releasing the bucket lock.
 		 */
 		if (unlikely(!refcount_inc_not_zero(&sktw->sk_refcnt)))
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 4c894e540730..fc9a850ed9bd 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -350,11 +350,10 @@ void tcp_time_wait(struct sock *sk, int state, int timeo)
 		 * we complete the initialization.
 		 */
 		local_bh_disable();
-		inet_twsk_schedule(tw, timeo);
 		/* Linkage updates.
 		 * Note that access to tw after this point is illegal.
 		 */
-		inet_twsk_hashdance(tw, sk, net->ipv4.tcp_death_row.hashinfo);
+		inet_twsk_hashdance_schedule(tw, sk, net->ipv4.tcp_death_row.hashinfo, timeo);
 		local_bh_enable();
 	} else {
 		/* Sorry, if we're out of memory, just CLOSE this
-- 
cgit v1.2.3-58-ga151


From c75ad7c759ccead1be462d4cec42bd4ad3578989 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 6 Jun 2024 17:11:38 +0200
Subject: net: tcp: un-pin the tw_timer

After previous patch, even if timer fires immediately on another CPU,
context that schedules the timer now holds the ehash spinlock, so timer
cannot reap tw socket until ehash lock is released.

BH disable is moved into hashdance_schedule.

Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dccp/minisocks.c          | 6 ------
 net/ipv4/inet_timewait_sock.c | 6 ++++--
 net/ipv4/tcp_minisocks.c      | 6 ------
 3 files changed, 4 insertions(+), 14 deletions(-)

diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c
index deb52d7d31b4..fecc8190064f 100644
--- a/net/dccp/minisocks.c
+++ b/net/dccp/minisocks.c
@@ -54,16 +54,10 @@ void dccp_time_wait(struct sock *sk, int state, int timeo)
 		if (state == DCCP_TIME_WAIT)
 			timeo = DCCP_TIMEWAIT_LEN;
 
-		/* tw_timer is pinned, so we need to make sure BH are disabled
-		 * in following section, otherwise timer handler could run before
-		 * we complete the initialization.
-		 */
-		local_bh_disable();
 		/* Linkage updates.
 		 * Note that access to tw after this point is illegal.
 		 */
 		inet_twsk_hashdance_schedule(tw, sk, &dccp_hashinfo, timeo);
-		local_bh_enable();
 	} else {
 		/* Sorry, if we're out of memory, just CLOSE this
 		 * socket up.  We've got bigger problems than
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index 628d33a41ce5..b2d97c816c99 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -93,7 +93,7 @@ static void inet_twsk_add_node_rcu(struct inet_timewait_sock *tw,
 }
 
 /*
- * Enter the time wait state. This is called with locally disabled BH.
+ * Enter the time wait state.
  * Essentially we whip up a timewait bucket, copy the relevant info into it
  * from the SK, and mess with hash chains and list linkage.
  *
@@ -118,6 +118,7 @@ void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw,
 			hashinfo->bhash_size)];
 	bhead2 = inet_bhashfn_portaddr(hashinfo, sk, twsk_net(tw), inet->inet_num);
 
+	local_bh_disable();
 	spin_lock(&bhead->lock);
 	spin_lock(&bhead2->lock);
 
@@ -158,6 +159,7 @@ void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw,
 	inet_twsk_schedule(tw, timeo);
 
 	spin_unlock(lock);
+	local_bh_enable();
 }
 EXPORT_SYMBOL_GPL(inet_twsk_hashdance_schedule);
 
@@ -203,7 +205,7 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk,
 		tw->tw_prot	    = sk->sk_prot_creator;
 		atomic64_set(&tw->tw_cookie, atomic64_read(&sk->sk_cookie));
 		twsk_net_set(tw, sock_net(sk));
-		timer_setup(&tw->tw_timer, tw_timer_handler, TIMER_PINNED);
+		timer_setup(&tw->tw_timer, tw_timer_handler, 0);
 		/*
 		 * Because we use RCU lookups, we should not set tw_refcnt
 		 * to a non null value before everything is setup for this
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index fc9a850ed9bd..bc67f6b9efae 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -345,16 +345,10 @@ void tcp_time_wait(struct sock *sk, int state, int timeo)
 		if (state == TCP_TIME_WAIT)
 			timeo = TCP_TIMEWAIT_LEN;
 
-		/* tw_timer is pinned, so we need to make sure BH are disabled
-		 * in following section, otherwise timer handler could run before
-		 * we complete the initialization.
-		 */
-		local_bh_disable();
 		/* Linkage updates.
 		 * Note that access to tw after this point is illegal.
 		 */
 		inet_twsk_hashdance_schedule(tw, sk, net->ipv4.tcp_death_row.hashinfo, timeo);
-		local_bh_enable();
 	} else {
 		/* Sorry, if we're out of memory, just CLOSE this
 		 * socket up.  We've got bigger problems than
-- 
cgit v1.2.3-58-ga151


From f81d0dd2fde35fd1acc30b3f4de6aaf57d514551 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 6 Jun 2024 17:11:39 +0200
Subject: tcp: move inet_twsk_schedule helper out of header

Its no longer used outside inet_timewait_sock.c, so move it there.

Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/inet_timewait_sock.h | 5 -----
 net/ipv4/inet_timewait_sock.c    | 5 +++++
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h
index 5b43d220243d..f88b68269012 100644
--- a/include/net/inet_timewait_sock.h
+++ b/include/net/inet_timewait_sock.h
@@ -101,11 +101,6 @@ void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw,
 void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo,
 			  bool rearm);
 
-static inline void inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo)
-{
-	__inet_twsk_schedule(tw, timeo, false);
-}
-
 static inline void inet_twsk_reschedule(struct inet_timewait_sock *tw, int timeo)
 {
 	__inet_twsk_schedule(tw, timeo, true);
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index b2d97c816c99..337390ba85b4 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -92,6 +92,11 @@ static void inet_twsk_add_node_rcu(struct inet_timewait_sock *tw,
 	hlist_nulls_add_head_rcu(&tw->tw_node, list);
 }
 
+static void inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo)
+{
+	__inet_twsk_schedule(tw, timeo, false);
+}
+
 /*
  * Enter the time wait state.
  * Essentially we whip up a timewait bucket, copy the relevant info into it
-- 
cgit v1.2.3-58-ga151


From c917b26e1686b99f18a3c9fbce508a7c264c6706 Mon Sep 17 00:00:00 2001
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Thu, 6 Jun 2024 19:15:49 +0300
Subject: net: dsa: hellcreek: Replace kernel.h with what is used

kernel.h is included solely for some other existing headers.
Include them directly and get rid of kernel.h.

While at it, sort headers alphabetically for easier maintenance.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Kurt Kanzenbach <kurt@linutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/hirschmann/hellcreek.h | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/net/dsa/hirschmann/hellcreek.h b/drivers/net/dsa/hirschmann/hellcreek.h
index 6874cb9dc361..9c2ed2ba79da 100644
--- a/drivers/net/dsa/hirschmann/hellcreek.h
+++ b/drivers/net/dsa/hirschmann/hellcreek.h
@@ -12,14 +12,16 @@
 
 #include <linux/bitmap.h>
 #include <linux/bitops.h>
+#include <linux/container_of.h>
 #include <linux/device.h>
-#include <linux/kernel.h>
-#include <linux/mutex.h>
-#include <linux/workqueue.h>
 #include <linux/leds.h>
+#include <linux/mutex.h>
 #include <linux/platform_data/hirschmann-hellcreek.h>
 #include <linux/ptp_clock_kernel.h>
 #include <linux/timecounter.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
+
 #include <net/dsa.h>
 #include <net/pkt_sched.h>
 
-- 
cgit v1.2.3-58-ga151


From 5380d64f8d766576ac5c0f627418b2d0e1d2641f Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 6 Jun 2024 12:29:05 -0700
Subject: rtnetlink: move rtnl_lock handling out of af_netlink

Now that we have an intermediate layer of code for handling
rtnl-level netlink dump quirks, we can move the rtnl_lock
taking there.

For dump handlers with RTNL_FLAG_DUMP_SPLIT_NLM_DONE we can
avoid taking rtnl_lock just to generate NLM_DONE, once again.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/rtnetlink.c     | 9 +++++++--
 net/netlink/af_netlink.c | 2 --
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 4668d6718040..eabfc8290f5e 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -6486,6 +6486,7 @@ static int rtnl_mdb_del(struct sk_buff *skb, struct nlmsghdr *nlh,
 
 static int rtnl_dumpit(struct sk_buff *skb, struct netlink_callback *cb)
 {
+	const bool needs_lock = !(cb->flags & RTNL_FLAG_DUMP_UNLOCKED);
 	rtnl_dumpit_func dumpit = cb->data;
 	int err;
 
@@ -6495,7 +6496,11 @@ static int rtnl_dumpit(struct sk_buff *skb, struct netlink_callback *cb)
 	if (!dumpit)
 		return 0;
 
+	if (needs_lock)
+		rtnl_lock();
 	err = dumpit(skb, cb);
+	if (needs_lock)
+		rtnl_unlock();
 
 	/* Old dump handlers used to send NLM_DONE as in a separate recvmsg().
 	 * Some applications which parse netlink manually depend on this.
@@ -6515,7 +6520,8 @@ static int rtnetlink_dump_start(struct sock *ssk, struct sk_buff *skb,
 				const struct nlmsghdr *nlh,
 				struct netlink_dump_control *control)
 {
-	if (control->flags & RTNL_FLAG_DUMP_SPLIT_NLM_DONE) {
+	if (control->flags & RTNL_FLAG_DUMP_SPLIT_NLM_DONE ||
+	    !(control->flags & RTNL_FLAG_DUMP_UNLOCKED)) {
 		WARN_ON(control->data);
 		control->data = control->dump;
 		control->dump = rtnl_dumpit;
@@ -6703,7 +6709,6 @@ static int __net_init rtnetlink_net_init(struct net *net)
 	struct netlink_kernel_cfg cfg = {
 		.groups		= RTNLGRP_MAX,
 		.input		= rtnetlink_rcv,
-		.cb_mutex	= &rtnl_mutex,
 		.flags		= NL_CFG_F_NONROOT_RECV,
 		.bind		= rtnetlink_bind,
 	};
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index fa9c090cf629..8bbbe75e75db 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -2330,8 +2330,6 @@ static int netlink_dump(struct sock *sk, bool lock_taken)
 
 		cb->extack = &extack;
 
-		if (cb->flags & RTNL_FLAG_DUMP_UNLOCKED)
-			extra_mutex = NULL;
 		if (extra_mutex)
 			mutex_lock(extra_mutex);
 		nlk->dump_done_errno = cb->dump(skb, cb);
-- 
cgit v1.2.3-58-ga151


From 5fbf57a937f418fe204f9dbb7735e91984f4ee6a Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 6 Jun 2024 12:29:06 -0700
Subject: net: netlink: remove the cb_mutex "injection" from netlink core

Back in 2007, in commit af65bdfce98d ("[NETLINK]: Switch cb_lock spinlock
to mutex and allow to override it") netlink core was extended to allow
subsystems to replace the dump mutex lock with its own lock.

The mechanism was used by rtnetlink to take rtnl_lock but it isn't
sufficiently flexible for other users. Over the 17 years since
it was added no other user appeared. Since rtnetlink needs conditional
locking now, and doesn't use it either, axe this feature complete.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netlink.h  |  1 -
 net/netlink/af_netlink.c | 18 +++---------------
 2 files changed, 3 insertions(+), 16 deletions(-)

diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index 5df7340d4dab..b332c2048c75 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -47,7 +47,6 @@ struct netlink_kernel_cfg {
 	unsigned int	groups;
 	unsigned int	flags;
 	void		(*input)(struct sk_buff *skb);
-	struct mutex	*cb_mutex;
 	int		(*bind)(struct net *net, int group);
 	void		(*unbind)(struct net *net, int group);
 	void            (*release) (struct sock *sk, unsigned long *groups);
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 8bbbe75e75db..0b7a89db3ab7 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -636,8 +636,7 @@ static struct proto netlink_proto = {
 };
 
 static int __netlink_create(struct net *net, struct socket *sock,
-			    struct mutex *dump_cb_mutex, int protocol,
-			    int kern)
+			    int protocol, int kern)
 {
 	struct sock *sk;
 	struct netlink_sock *nlk;
@@ -655,7 +654,6 @@ static int __netlink_create(struct net *net, struct socket *sock,
 	lockdep_set_class_and_name(&nlk->nl_cb_mutex,
 					   nlk_cb_mutex_keys + protocol,
 					   nlk_cb_mutex_key_strings[protocol]);
-	nlk->dump_cb_mutex = dump_cb_mutex;
 	init_waitqueue_head(&nlk->wait);
 
 	sk->sk_destruct = netlink_sock_destruct;
@@ -667,7 +665,6 @@ static int netlink_create(struct net *net, struct socket *sock, int protocol,
 			  int kern)
 {
 	struct module *module = NULL;
-	struct mutex *cb_mutex;
 	struct netlink_sock *nlk;
 	int (*bind)(struct net *net, int group);
 	void (*unbind)(struct net *net, int group);
@@ -696,7 +693,6 @@ static int netlink_create(struct net *net, struct socket *sock, int protocol,
 		module = nl_table[protocol].module;
 	else
 		err = -EPROTONOSUPPORT;
-	cb_mutex = nl_table[protocol].cb_mutex;
 	bind = nl_table[protocol].bind;
 	unbind = nl_table[protocol].unbind;
 	release = nl_table[protocol].release;
@@ -705,7 +701,7 @@ static int netlink_create(struct net *net, struct socket *sock, int protocol,
 	if (err < 0)
 		goto out;
 
-	err = __netlink_create(net, sock, cb_mutex, protocol, kern);
+	err = __netlink_create(net, sock, protocol, kern);
 	if (err < 0)
 		goto out_module;
 
@@ -2016,7 +2012,6 @@ __netlink_kernel_create(struct net *net, int unit, struct module *module,
 	struct sock *sk;
 	struct netlink_sock *nlk;
 	struct listeners *listeners = NULL;
-	struct mutex *cb_mutex = cfg ? cfg->cb_mutex : NULL;
 	unsigned int groups;
 
 	BUG_ON(!nl_table);
@@ -2027,7 +2022,7 @@ __netlink_kernel_create(struct net *net, int unit, struct module *module,
 	if (sock_create_lite(PF_NETLINK, SOCK_DGRAM, unit, &sock))
 		return NULL;
 
-	if (__netlink_create(net, sock, cb_mutex, unit, 1) < 0)
+	if (__netlink_create(net, sock, unit, 1) < 0)
 		goto out_sock_release_nosk;
 
 	sk = sock->sk;
@@ -2055,7 +2050,6 @@ __netlink_kernel_create(struct net *net, int unit, struct module *module,
 	if (!nl_table[unit].registered) {
 		nl_table[unit].groups = groups;
 		rcu_assign_pointer(nl_table[unit].listeners, listeners);
-		nl_table[unit].cb_mutex = cb_mutex;
 		nl_table[unit].module = module;
 		if (cfg) {
 			nl_table[unit].bind = cfg->bind;
@@ -2326,15 +2320,9 @@ static int netlink_dump(struct sock *sk, bool lock_taken)
 	netlink_skb_set_owner_r(skb, sk);
 
 	if (nlk->dump_done_errno > 0) {
-		struct mutex *extra_mutex = nlk->dump_cb_mutex;
-
 		cb->extack = &extack;
 
-		if (extra_mutex)
-			mutex_lock(extra_mutex);
 		nlk->dump_done_errno = cb->dump(skb, cb);
-		if (extra_mutex)
-			mutex_unlock(extra_mutex);
 
 		/* EMSGSIZE plus something already in the skb means
 		 * that there's more to dump but current skb has filled up.
-- 
cgit v1.2.3-58-ga151


From 77f7541248fcfcc79afff68ff9acf1f09ab8e1e4 Mon Sep 17 00:00:00 2001
From: Marek Behún <kabel@kernel.org>
Date: Wed, 5 Jun 2024 15:33:28 +0200
Subject: net: dsa: deduplicate code adding / deleting the port address to fdb
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The sequence
  if (dsa_switch_supports_uc_filtering(ds))
    dsa_port_standalone_host_fdb_add(dp, addr, 0);
  if (!ether_addr_equal(addr, conduit->dev_addr))
    dev_uc_add(conduit, addr);
is executed both in dsa_user_open() and dsa_user_set_mac_addr().

Its reverse is executed both in dsa_user_close() and
dsa_user_set_mac_addr().

Refactor these sequences into new functions dsa_user_host_uc_install()
and dsa_user_host_uc_uninstall().

Signed-off-by: Marek Behún <kabel@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/user.c | 91 ++++++++++++++++++++++++++++++----------------------------
 1 file changed, 47 insertions(+), 44 deletions(-)

diff --git a/net/dsa/user.c b/net/dsa/user.c
index 867c5fe9a4da..efbb34c711cb 100644
--- a/net/dsa/user.c
+++ b/net/dsa/user.c
@@ -355,60 +355,82 @@ static int dsa_user_get_iflink(const struct net_device *dev)
 	return READ_ONCE(dsa_user_to_conduit(dev)->ifindex);
 }
 
-static int dsa_user_open(struct net_device *dev)
+static int dsa_user_host_uc_install(struct net_device *dev, const u8 *addr)
 {
 	struct net_device *conduit = dsa_user_to_conduit(dev);
 	struct dsa_port *dp = dsa_user_to_port(dev);
 	struct dsa_switch *ds = dp->ds;
 	int err;
 
-	err = dev_open(conduit, NULL);
-	if (err < 0) {
-		netdev_err(dev, "failed to open conduit %s\n", conduit->name);
-		goto out;
-	}
-
 	if (dsa_switch_supports_uc_filtering(ds)) {
-		err = dsa_port_standalone_host_fdb_add(dp, dev->dev_addr, 0);
+		err = dsa_port_standalone_host_fdb_add(dp, addr, 0);
 		if (err)
 			goto out;
 	}
 
-	if (!ether_addr_equal(dev->dev_addr, conduit->dev_addr)) {
-		err = dev_uc_add(conduit, dev->dev_addr);
+	if (!ether_addr_equal(addr, conduit->dev_addr)) {
+		err = dev_uc_add(conduit, addr);
 		if (err < 0)
 			goto del_host_addr;
 	}
 
-	err = dsa_port_enable_rt(dp, dev->phydev);
-	if (err)
-		goto del_unicast;
-
 	return 0;
 
-del_unicast:
-	if (!ether_addr_equal(dev->dev_addr, conduit->dev_addr))
-		dev_uc_del(conduit, dev->dev_addr);
 del_host_addr:
 	if (dsa_switch_supports_uc_filtering(ds))
-		dsa_port_standalone_host_fdb_del(dp, dev->dev_addr, 0);
+		dsa_port_standalone_host_fdb_del(dp, addr, 0);
 out:
 	return err;
 }
 
-static int dsa_user_close(struct net_device *dev)
+static void dsa_user_host_uc_uninstall(struct net_device *dev)
 {
 	struct net_device *conduit = dsa_user_to_conduit(dev);
 	struct dsa_port *dp = dsa_user_to_port(dev);
 	struct dsa_switch *ds = dp->ds;
 
-	dsa_port_disable_rt(dp);
-
 	if (!ether_addr_equal(dev->dev_addr, conduit->dev_addr))
 		dev_uc_del(conduit, dev->dev_addr);
 
 	if (dsa_switch_supports_uc_filtering(ds))
 		dsa_port_standalone_host_fdb_del(dp, dev->dev_addr, 0);
+}
+
+static int dsa_user_open(struct net_device *dev)
+{
+	struct net_device *conduit = dsa_user_to_conduit(dev);
+	struct dsa_port *dp = dsa_user_to_port(dev);
+	int err;
+
+	err = dev_open(conduit, NULL);
+	if (err < 0) {
+		netdev_err(dev, "failed to open conduit %s\n", conduit->name);
+		goto out;
+	}
+
+	err = dsa_user_host_uc_install(dev, dev->dev_addr);
+	if (err)
+		goto out;
+
+	err = dsa_port_enable_rt(dp, dev->phydev);
+	if (err)
+		goto out_del_host_uc;
+
+	return 0;
+
+out_del_host_uc:
+	dsa_user_host_uc_uninstall(dev);
+out:
+	return err;
+}
+
+static int dsa_user_close(struct net_device *dev)
+{
+	struct dsa_port *dp = dsa_user_to_port(dev);
+
+	dsa_port_disable_rt(dp);
+
+	dsa_user_host_uc_uninstall(dev);
 
 	return 0;
 }
@@ -448,7 +470,6 @@ static void dsa_user_set_rx_mode(struct net_device *dev)
 
 static int dsa_user_set_mac_address(struct net_device *dev, void *a)
 {
-	struct net_device *conduit = dsa_user_to_conduit(dev);
 	struct dsa_port *dp = dsa_user_to_port(dev);
 	struct dsa_switch *ds = dp->ds;
 	struct sockaddr *addr = a;
@@ -470,34 +491,16 @@ static int dsa_user_set_mac_address(struct net_device *dev, void *a)
 	if (!(dev->flags & IFF_UP))
 		goto out_change_dev_addr;
 
-	if (dsa_switch_supports_uc_filtering(ds)) {
-		err = dsa_port_standalone_host_fdb_add(dp, addr->sa_data, 0);
-		if (err)
-			return err;
-	}
-
-	if (!ether_addr_equal(addr->sa_data, conduit->dev_addr)) {
-		err = dev_uc_add(conduit, addr->sa_data);
-		if (err < 0)
-			goto del_unicast;
-	}
-
-	if (!ether_addr_equal(dev->dev_addr, conduit->dev_addr))
-		dev_uc_del(conduit, dev->dev_addr);
+	err = dsa_user_host_uc_install(dev, addr->sa_data);
+	if (err)
+		return err;
 
-	if (dsa_switch_supports_uc_filtering(ds))
-		dsa_port_standalone_host_fdb_del(dp, dev->dev_addr, 0);
+	dsa_user_host_uc_uninstall(dev);
 
 out_change_dev_addr:
 	eth_hw_addr_set(dev, addr->sa_data);
 
 	return 0;
-
-del_unicast:
-	if (dsa_switch_supports_uc_filtering(ds))
-		dsa_port_standalone_host_fdb_del(dp, addr->sa_data, 0);
-
-	return err;
 }
 
 struct dsa_user_dump_ctx {
-- 
cgit v1.2.3-58-ga151


From eef8e906aea270f4b8912a0b51403b80aec54d30 Mon Sep 17 00:00:00 2001
From: Marek Behún <kabel@kernel.org>
Date: Wed, 5 Jun 2024 15:33:29 +0200
Subject: net: dsa: update the unicast MAC address when changing conduit
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When changing DSA user interface conduit while the user interface is up,
DSA exhibits different behavior in comparison to when the interface is
down. This different behavior concerns the primary unicast MAC address
stored in the port standalone FDB and in the conduit device UC database.

If we put a switch port down while changing the conduit with
  ip link set sw0p0 down
  ip link set sw0p0 type dsa conduit conduit1
  ip link set sw0p0 up
we delete the address in dsa_user_close() and install the (possibly
different) address in dsa_user_open().

But when changing the conduit on the fly, the old address is not
deleted and the new one is not installed.

Since we explicitly want to support live-changing the conduit, uninstall
the old address before calling dsa_port_assign_conduit() and install the
(possibly different) new address after the call.

Because conduit change might also trigger address change (the user
interface is supposed to inherit the conduit interface MAC address if no
address is defined in hardware (dp->mac is a zero address)), move the
eth_hw_addr_inherit() call from dsa_user_change_conduit() to
dsa_port_change_conduit(), just before installing the new address.

Although this is in theory a flaw in DSA core, it needs not be
backported, since there is currently no DSA driver that can be affected
by this. The only DSA driver that supports changing conduit is felix,
and, as explained by Vladimir Oltean [1]:

  There are 2 reasons why with felix the bug does not manifest itself.

  First is because both the 'ocelot' and the alternate 'ocelot-8021q'
  tagging protocols have the 'promisc_on_conduit = true' flag. So the
  unicast address doesn't have to be in the conduit's RX filter -
  neither the old or the new conduit.

  Second, dsa_user_host_uc_install() theoretically leaves behind host
  FDB entries installed towards the wrong (old) CPU port. But in
  felix_fdb_add(), we treat any FDB entry requested towards any CPU port
  as if it was a multicast FDB entry programmed towards _all_ CPU ports.
  For that reason, it is installed towards the port mask of the PGID_CPU
  port group ID:

	if (dsa_port_is_cpu(dp))
		port = PGID_CPU;

Therefore no Fixes tag for this change.

[1] https://lore.kernel.org/netdev/20240507201827.47suw4fwcjrbungy@skbuf/
Signed-off-by: Marek Behún <kabel@kernel.org>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Tested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/dsa/port.c | 40 ++++++++++++++++++++++++++++++++++++++++
 net/dsa/user.c | 10 ++--------
 net/dsa/user.h |  2 ++
 3 files changed, 44 insertions(+), 8 deletions(-)

diff --git a/net/dsa/port.c b/net/dsa/port.c
index e23db9507546..25258b33e59e 100644
--- a/net/dsa/port.c
+++ b/net/dsa/port.c
@@ -1467,10 +1467,34 @@ int dsa_port_change_conduit(struct dsa_port *dp, struct net_device *conduit,
 	 */
 	dsa_user_unsync_ha(dev);
 
+	/* If live-changing, we also need to uninstall the user device address
+	 * from the port FDB and the conduit interface.
+	 */
+	if (dev->flags & IFF_UP)
+		dsa_user_host_uc_uninstall(dev);
+
 	err = dsa_port_assign_conduit(dp, conduit, extack, true);
 	if (err)
 		goto rewind_old_addrs;
 
+	/* If the port doesn't have its own MAC address and relies on the DSA
+	 * conduit's one, inherit it again from the new DSA conduit.
+	 */
+	if (is_zero_ether_addr(dp->mac))
+		eth_hw_addr_inherit(dev, conduit);
+
+	/* If live-changing, we need to install the user device address to the
+	 * port FDB and the conduit interface.
+	 */
+	if (dev->flags & IFF_UP) {
+		err = dsa_user_host_uc_install(dev, dev->dev_addr);
+		if (err) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Failed to install host UC address");
+			goto rewind_addr_inherit;
+		}
+	}
+
 	dsa_user_sync_ha(dev);
 
 	if (vlan_filtering) {
@@ -1500,10 +1524,26 @@ rewind_new_vlan:
 rewind_new_addrs:
 	dsa_user_unsync_ha(dev);
 
+	if (dev->flags & IFF_UP)
+		dsa_user_host_uc_uninstall(dev);
+
+rewind_addr_inherit:
+	if (is_zero_ether_addr(dp->mac))
+		eth_hw_addr_inherit(dev, old_conduit);
+
 	dsa_port_assign_conduit(dp, old_conduit, NULL, false);
 
 /* Restore the objects on the old CPU port */
 rewind_old_addrs:
+	if (dev->flags & IFF_UP) {
+		tmp = dsa_user_host_uc_install(dev, dev->dev_addr);
+		if (tmp) {
+			dev_err(ds->dev,
+				"port %d failed to restore host UC address: %pe\n",
+				dp->index, ERR_PTR(tmp));
+		}
+	}
+
 	dsa_user_sync_ha(dev);
 
 	if (vlan_filtering) {
diff --git a/net/dsa/user.c b/net/dsa/user.c
index efbb34c711cb..e8f56a40b614 100644
--- a/net/dsa/user.c
+++ b/net/dsa/user.c
@@ -355,7 +355,7 @@ static int dsa_user_get_iflink(const struct net_device *dev)
 	return READ_ONCE(dsa_user_to_conduit(dev)->ifindex);
 }
 
-static int dsa_user_host_uc_install(struct net_device *dev, const u8 *addr)
+int dsa_user_host_uc_install(struct net_device *dev, const u8 *addr)
 {
 	struct net_device *conduit = dsa_user_to_conduit(dev);
 	struct dsa_port *dp = dsa_user_to_port(dev);
@@ -383,7 +383,7 @@ out:
 	return err;
 }
 
-static void dsa_user_host_uc_uninstall(struct net_device *dev)
+void dsa_user_host_uc_uninstall(struct net_device *dev)
 {
 	struct net_device *conduit = dsa_user_to_conduit(dev);
 	struct dsa_port *dp = dsa_user_to_port(dev);
@@ -2882,12 +2882,6 @@ int dsa_user_change_conduit(struct net_device *dev, struct net_device *conduit,
 			    ERR_PTR(err));
 	}
 
-	/* If the port doesn't have its own MAC address and relies on the DSA
-	 * conduit's one, inherit it again from the new DSA conduit.
-	 */
-	if (is_zero_ether_addr(dp->mac))
-		eth_hw_addr_inherit(dev, conduit);
-
 	return 0;
 
 out_revert_conduit_link:
diff --git a/net/dsa/user.h b/net/dsa/user.h
index 996069130bea..016884bead3c 100644
--- a/net/dsa/user.h
+++ b/net/dsa/user.h
@@ -42,6 +42,8 @@ int dsa_user_suspend(struct net_device *user_dev);
 int dsa_user_resume(struct net_device *user_dev);
 int dsa_user_register_notifier(void);
 void dsa_user_unregister_notifier(void);
+int dsa_user_host_uc_install(struct net_device *dev, const u8 *addr);
+void dsa_user_host_uc_uninstall(struct net_device *dev);
 void dsa_user_sync_ha(struct net_device *dev);
 void dsa_user_unsync_ha(struct net_device *dev);
 void dsa_user_setup_tagger(struct net_device *user);
-- 
cgit v1.2.3-58-ga151


From 2ebb87f45b3c6adc97b29291102ecb97274f913f Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Fri, 7 Jun 2024 11:56:56 -0700
Subject: isdn: add missing MODULE_DESCRIPTION() macros

make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/hardware/mISDN/hfcpci.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/hardware/mISDN/hfcmulti.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/hardware/mISDN/hfcsusb.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/hardware/mISDN/avmfritz.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/hardware/mISDN/speedfax.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/hardware/mISDN/mISDNinfineon.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/hardware/mISDN/w6692.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/hardware/mISDN/netjet.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/hardware/mISDN/mISDNipac.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/hardware/mISDN/mISDNisar.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/mISDN/mISDN_core.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/mISDN/mISDN_dsp.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/isdn/mISDN/l1oip.o

Add the missing invocations of the MODULE_DESCRIPTION() macro.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://lore.kernel.org/r/20240607-md-drivers-isdn-v1-1-81fb7001bc3a@quicinc.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/isdn/hardware/mISDN/avmfritz.c      | 1 +
 drivers/isdn/hardware/mISDN/hfcmulti.c      | 1 +
 drivers/isdn/hardware/mISDN/hfcpci.c        | 1 +
 drivers/isdn/hardware/mISDN/hfcsusb.c       | 1 +
 drivers/isdn/hardware/mISDN/mISDNinfineon.c | 1 +
 drivers/isdn/hardware/mISDN/mISDNipac.c     | 1 +
 drivers/isdn/hardware/mISDN/mISDNisar.c     | 1 +
 drivers/isdn/hardware/mISDN/netjet.c        | 1 +
 drivers/isdn/hardware/mISDN/speedfax.c      | 1 +
 drivers/isdn/hardware/mISDN/w6692.c         | 1 +
 drivers/isdn/mISDN/core.c                   | 1 +
 drivers/isdn/mISDN/dsp_core.c               | 1 +
 drivers/isdn/mISDN/l1oip_core.c             | 1 +
 13 files changed, 13 insertions(+)

diff --git a/drivers/isdn/hardware/mISDN/avmfritz.c b/drivers/isdn/hardware/mISDN/avmfritz.c
index f68569bfef7a..509b362d6465 100644
--- a/drivers/isdn/hardware/mISDN/avmfritz.c
+++ b/drivers/isdn/hardware/mISDN/avmfritz.c
@@ -159,6 +159,7 @@ set_debug(const char *val, const struct kernel_param *kp)
 }
 
 MODULE_AUTHOR("Karsten Keil");
+MODULE_DESCRIPTION("mISDN driver for AVM FRITZ!CARD PCI ISDN cards");
 MODULE_LICENSE("GPL v2");
 MODULE_VERSION(AVMFRITZ_REV);
 module_param_call(debug, set_debug, param_get_uint, &debug, S_IRUGO | S_IWUSR);
diff --git a/drivers/isdn/hardware/mISDN/hfcmulti.c b/drivers/isdn/hardware/mISDN/hfcmulti.c
index 2e5cb9dde3ec..0d2928d8aeae 100644
--- a/drivers/isdn/hardware/mISDN/hfcmulti.c
+++ b/drivers/isdn/hardware/mISDN/hfcmulti.c
@@ -221,6 +221,7 @@ static uint	hwid = HWID_NONE;
 static int	HFC_cnt, E1_cnt, bmask_cnt, Port_cnt, PCM_cnt = 99;
 
 MODULE_AUTHOR("Andreas Eversberg");
+MODULE_DESCRIPTION("mISDN driver for hfc-4s/hfc-8s/hfc-e1 based cards");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(HFC_MULTI_VERSION);
 module_param(debug, uint, S_IRUGO | S_IWUSR);
diff --git a/drivers/isdn/hardware/mISDN/hfcpci.c b/drivers/isdn/hardware/mISDN/hfcpci.c
index fe391de1aba3..ce7bccc9faa3 100644
--- a/drivers/isdn/hardware/mISDN/hfcpci.c
+++ b/drivers/isdn/hardware/mISDN/hfcpci.c
@@ -48,6 +48,7 @@ static struct timer_list hfc_tl;
 static unsigned long hfc_jiffies;
 
 MODULE_AUTHOR("Karsten Keil");
+MODULE_DESCRIPTION("mISDN driver for CCD's hfc-pci based cards");
 MODULE_LICENSE("GPL");
 module_param(debug, uint, S_IRUGO | S_IWUSR);
 module_param(poll, uint, S_IRUGO | S_IWUSR);
diff --git a/drivers/isdn/hardware/mISDN/hfcsusb.c b/drivers/isdn/hardware/mISDN/hfcsusb.c
index b82b89888a5e..e54419a4e731 100644
--- a/drivers/isdn/hardware/mISDN/hfcsusb.c
+++ b/drivers/isdn/hardware/mISDN/hfcsusb.c
@@ -31,6 +31,7 @@ static DEFINE_RWLOCK(HFClock);
 
 
 MODULE_AUTHOR("Martin Bachem");
+MODULE_DESCRIPTION("mISDN driver for Colognechip HFC-S USB chip");
 MODULE_LICENSE("GPL");
 module_param(debug, uint, S_IRUGO | S_IWUSR);
 module_param(poll, int, 0);
diff --git a/drivers/isdn/hardware/mISDN/mISDNinfineon.c b/drivers/isdn/hardware/mISDN/mISDNinfineon.c
index 88d592bafdb0..30876a012711 100644
--- a/drivers/isdn/hardware/mISDN/mISDNinfineon.c
+++ b/drivers/isdn/hardware/mISDN/mISDNinfineon.c
@@ -245,6 +245,7 @@ set_debug(const char *val, const struct kernel_param *kp)
 }
 
 MODULE_AUTHOR("Karsten Keil");
+MODULE_DESCRIPTION("mISDN driver for cards based on Infineon ISDN chipsets");
 MODULE_LICENSE("GPL v2");
 MODULE_VERSION(INFINEON_REV);
 module_param_call(debug, set_debug, param_get_uint, &debug, S_IRUGO | S_IWUSR);
diff --git a/drivers/isdn/hardware/mISDN/mISDNipac.c b/drivers/isdn/hardware/mISDN/mISDNipac.c
index 4f8d85bb3ce1..d0b7271fbda1 100644
--- a/drivers/isdn/hardware/mISDN/mISDNipac.c
+++ b/drivers/isdn/hardware/mISDN/mISDNipac.c
@@ -21,6 +21,7 @@
 
 MODULE_AUTHOR("Karsten Keil");
 MODULE_VERSION(ISAC_REV);
+MODULE_DESCRIPTION("mISDN driver for ISAC specific functions");
 MODULE_LICENSE("GPL v2");
 
 #define ReadISAC(is, o)		(is->read_reg(is->dch.hw, o + is->off))
diff --git a/drivers/isdn/hardware/mISDN/mISDNisar.c b/drivers/isdn/hardware/mISDN/mISDNisar.c
index 48b3d43e2502..b3e03c410544 100644
--- a/drivers/isdn/hardware/mISDN/mISDNisar.c
+++ b/drivers/isdn/hardware/mISDN/mISDNisar.c
@@ -22,6 +22,7 @@
 #define ISAR_REV	"2.1"
 
 MODULE_AUTHOR("Karsten Keil");
+MODULE_DESCRIPTION("mISDN driver for ISAR (Siemens PSB 7110) specific functions");
 MODULE_LICENSE("GPL v2");
 MODULE_VERSION(ISAR_REV);
 
diff --git a/drivers/isdn/hardware/mISDN/netjet.c b/drivers/isdn/hardware/mISDN/netjet.c
index 566c790a9481..d163850c295e 100644
--- a/drivers/isdn/hardware/mISDN/netjet.c
+++ b/drivers/isdn/hardware/mISDN/netjet.c
@@ -114,6 +114,7 @@ set_debug(const char *val, const struct kernel_param *kp)
 }
 
 MODULE_AUTHOR("Karsten Keil");
+MODULE_DESCRIPTION("mISDN driver for NETJet cards");
 MODULE_LICENSE("GPL v2");
 MODULE_VERSION(NETJET_REV);
 module_param_call(debug, set_debug, param_get_uint, &debug, S_IRUGO | S_IWUSR);
diff --git a/drivers/isdn/hardware/mISDN/speedfax.c b/drivers/isdn/hardware/mISDN/speedfax.c
index b530c78eca8e..0c405261d940 100644
--- a/drivers/isdn/hardware/mISDN/speedfax.c
+++ b/drivers/isdn/hardware/mISDN/speedfax.c
@@ -97,6 +97,7 @@ set_debug(const char *val, const struct kernel_param *kp)
 }
 
 MODULE_AUTHOR("Karsten Keil");
+MODULE_DESCRIPTION("mISDN driver for Sedlbauer Speedfax+ cards");
 MODULE_LICENSE("GPL v2");
 MODULE_VERSION(SPEEDFAX_REV);
 MODULE_FIRMWARE("isdn/ISAR.BIN");
diff --git a/drivers/isdn/hardware/mISDN/w6692.c b/drivers/isdn/hardware/mISDN/w6692.c
index f3b8db7b48fe..ee69212ac351 100644
--- a/drivers/isdn/hardware/mISDN/w6692.c
+++ b/drivers/isdn/hardware/mISDN/w6692.c
@@ -101,6 +101,7 @@ set_debug(const char *val, const struct kernel_param *kp)
 }
 
 MODULE_AUTHOR("Karsten Keil");
+MODULE_DESCRIPTION("mISDN driver for Winbond w6692 based cards");
 MODULE_LICENSE("GPL v2");
 MODULE_VERSION(W6692_REV);
 module_param_call(debug, set_debug, param_get_uint, &debug, S_IRUGO | S_IWUSR);
diff --git a/drivers/isdn/mISDN/core.c b/drivers/isdn/mISDN/core.c
index ab8513a7acd5..e34a7a46754e 100644
--- a/drivers/isdn/mISDN/core.c
+++ b/drivers/isdn/mISDN/core.c
@@ -14,6 +14,7 @@
 static u_int debug;
 
 MODULE_AUTHOR("Karsten Keil");
+MODULE_DESCRIPTION("Modular ISDN core driver");
 MODULE_LICENSE("GPL");
 module_param(debug, uint, S_IRUGO | S_IWUSR);
 
diff --git a/drivers/isdn/mISDN/dsp_core.c b/drivers/isdn/mISDN/dsp_core.c
index fae95f166688..753232e9fc36 100644
--- a/drivers/isdn/mISDN/dsp_core.c
+++ b/drivers/isdn/mISDN/dsp_core.c
@@ -172,6 +172,7 @@ module_param(debug, uint, S_IRUGO | S_IWUSR);
 module_param(options, uint, S_IRUGO | S_IWUSR);
 module_param(poll, uint, S_IRUGO | S_IWUSR);
 module_param(dtmfthreshold, uint, S_IRUGO | S_IWUSR);
+MODULE_DESCRIPTION("mISDN driver for Digital Audio Processing of transparent data");
 MODULE_LICENSE("GPL");
 
 /*int spinnest = 0;*/
diff --git a/drivers/isdn/mISDN/l1oip_core.c b/drivers/isdn/mISDN/l1oip_core.c
index f010b35a0531..a5ad88a960d0 100644
--- a/drivers/isdn/mISDN/l1oip_core.c
+++ b/drivers/isdn/mISDN/l1oip_core.c
@@ -245,6 +245,7 @@ static int debug;
 static int ulaw;
 
 MODULE_AUTHOR("Andreas Eversberg");
+MODULE_DESCRIPTION("mISDN driver for tunneling layer 1 over IP");
 MODULE_LICENSE("GPL");
 module_param_array(type, uint, NULL, S_IRUGO | S_IWUSR);
 module_param_array(codec, uint, NULL, S_IRUGO | S_IWUSR);
-- 
cgit v1.2.3-58-ga151


From a2fe35df41c4cfce44f9f87d417cbd44a27b7362 Mon Sep 17 00:00:00 2001
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Fri, 7 Jun 2024 14:22:32 -0700
Subject: net: intel: Use *-y instead of *-objs in Makefile

*-objs suffix is reserved rather for (user-space) host programs while
usually *-y suffix is used for kernel drivers (although *-objs works
for that purpose for now).

Let's correct the old usages of *-objs in Makefiles.

Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Reviewed-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240607-next-2024-06-03-intel-next-batch-v3-1-d1470cee3347@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/e1000/Makefile   | 2 +-
 drivers/net/ethernet/intel/e1000e/Makefile  | 7 +++----
 drivers/net/ethernet/intel/i40e/Makefile    | 2 +-
 drivers/net/ethernet/intel/iavf/Makefile    | 5 ++---
 drivers/net/ethernet/intel/igb/Makefile     | 6 +++---
 drivers/net/ethernet/intel/igbvf/Makefile   | 6 +-----
 drivers/net/ethernet/intel/igc/Makefile     | 6 +++---
 drivers/net/ethernet/intel/ixgbe/Makefile   | 8 ++++----
 drivers/net/ethernet/intel/ixgbevf/Makefile | 6 +-----
 drivers/net/ethernet/intel/libeth/Makefile  | 2 +-
 drivers/net/ethernet/intel/libie/Makefile   | 2 +-
 11 files changed, 21 insertions(+), 31 deletions(-)

diff --git a/drivers/net/ethernet/intel/e1000/Makefile b/drivers/net/ethernet/intel/e1000/Makefile
index 314c52d44b7c..79491dec47e1 100644
--- a/drivers/net/ethernet/intel/e1000/Makefile
+++ b/drivers/net/ethernet/intel/e1000/Makefile
@@ -7,4 +7,4 @@
 
 obj-$(CONFIG_E1000) += e1000.o
 
-e1000-objs := e1000_main.o e1000_hw.o e1000_ethtool.o e1000_param.o
+e1000-y := e1000_main.o e1000_hw.o e1000_ethtool.o e1000_param.o
diff --git a/drivers/net/ethernet/intel/e1000e/Makefile b/drivers/net/ethernet/intel/e1000e/Makefile
index 0baa15503c38..18f22b6374d5 100644
--- a/drivers/net/ethernet/intel/e1000e/Makefile
+++ b/drivers/net/ethernet/intel/e1000e/Makefile
@@ -10,7 +10,6 @@ subdir-ccflags-y += -I$(src)
 
 obj-$(CONFIG_E1000E) += e1000e.o
 
-e1000e-objs := 82571.o ich8lan.o 80003es2lan.o \
-	       mac.o manage.o nvm.o phy.o \
-	       param.o ethtool.o netdev.o ptp.o
-
+e1000e-y := 82571.o ich8lan.o 80003es2lan.o \
+	    mac.o manage.o nvm.o phy.o \
+	    param.o ethtool.o netdev.o ptp.o
diff --git a/drivers/net/ethernet/intel/i40e/Makefile b/drivers/net/ethernet/intel/i40e/Makefile
index cad93f323bd5..9faa4339a76c 100644
--- a/drivers/net/ethernet/intel/i40e/Makefile
+++ b/drivers/net/ethernet/intel/i40e/Makefile
@@ -10,7 +10,7 @@ subdir-ccflags-y += -I$(src)
 
 obj-$(CONFIG_I40E) += i40e.o
 
-i40e-objs := i40e_main.o \
+i40e-y := i40e_main.o \
 	i40e_ethtool.o	\
 	i40e_adminq.o	\
 	i40e_common.o	\
diff --git a/drivers/net/ethernet/intel/iavf/Makefile b/drivers/net/ethernet/intel/iavf/Makefile
index 2d154a4e2fd7..356ac9faa5bf 100644
--- a/drivers/net/ethernet/intel/iavf/Makefile
+++ b/drivers/net/ethernet/intel/iavf/Makefile
@@ -11,6 +11,5 @@ subdir-ccflags-y += -I$(src)
 
 obj-$(CONFIG_IAVF) += iavf.o
 
-iavf-objs := iavf_main.o iavf_ethtool.o iavf_virtchnl.o iavf_fdir.o \
-	     iavf_adv_rss.o \
-	     iavf_txrx.o iavf_common.o iavf_adminq.o
+iavf-y := iavf_main.o iavf_ethtool.o iavf_virtchnl.o iavf_fdir.o \
+	  iavf_adv_rss.o iavf_txrx.o iavf_common.o iavf_adminq.o
diff --git a/drivers/net/ethernet/intel/igb/Makefile b/drivers/net/ethernet/intel/igb/Makefile
index 394c1e0656b9..463c0d26b9d4 100644
--- a/drivers/net/ethernet/intel/igb/Makefile
+++ b/drivers/net/ethernet/intel/igb/Makefile
@@ -6,6 +6,6 @@
 
 obj-$(CONFIG_IGB) += igb.o
 
-igb-objs := igb_main.o igb_ethtool.o e1000_82575.o \
-	    e1000_mac.o e1000_nvm.o e1000_phy.o e1000_mbx.o \
-	    e1000_i210.o igb_ptp.o igb_hwmon.o
+igb-y := igb_main.o igb_ethtool.o e1000_82575.o \
+	 e1000_mac.o e1000_nvm.o e1000_phy.o e1000_mbx.o \
+	 e1000_i210.o igb_ptp.o igb_hwmon.o
diff --git a/drivers/net/ethernet/intel/igbvf/Makefile b/drivers/net/ethernet/intel/igbvf/Makefile
index afd3e36eae75..902711d5e691 100644
--- a/drivers/net/ethernet/intel/igbvf/Makefile
+++ b/drivers/net/ethernet/intel/igbvf/Makefile
@@ -6,8 +6,4 @@
 
 obj-$(CONFIG_IGBVF) += igbvf.o
 
-igbvf-objs := vf.o \
-              mbx.o \
-              ethtool.o \
-              netdev.o
-
+igbvf-y := vf.o mbx.o ethtool.o netdev.o
diff --git a/drivers/net/ethernet/intel/igc/Makefile b/drivers/net/ethernet/intel/igc/Makefile
index ebffd3054285..efc5e7983dad 100644
--- a/drivers/net/ethernet/intel/igc/Makefile
+++ b/drivers/net/ethernet/intel/igc/Makefile
@@ -6,7 +6,7 @@
 #
 
 obj-$(CONFIG_IGC) += igc.o
-igc-$(CONFIG_IGC_LEDS) += igc_leds.o
 
-igc-objs := igc_main.o igc_mac.o igc_i225.o igc_base.o igc_nvm.o igc_phy.o \
-igc_diag.o igc_ethtool.o igc_ptp.o igc_dump.o igc_tsn.o igc_xdp.o
+igc-y := igc_main.o igc_mac.o igc_i225.o igc_base.o igc_nvm.o igc_phy.o \
+	 igc_diag.o igc_ethtool.o igc_ptp.o igc_dump.o igc_tsn.o igc_xdp.o
+igc-$(CONFIG_IGC_LEDS) += igc_leds.o
diff --git a/drivers/net/ethernet/intel/ixgbe/Makefile b/drivers/net/ethernet/intel/ixgbe/Makefile
index 4fb0d9e3f2da..965e5ce1b326 100644
--- a/drivers/net/ethernet/intel/ixgbe/Makefile
+++ b/drivers/net/ethernet/intel/ixgbe/Makefile
@@ -6,10 +6,10 @@
 
 obj-$(CONFIG_IXGBE) += ixgbe.o
 
-ixgbe-objs := ixgbe_main.o ixgbe_common.o ixgbe_ethtool.o \
-              ixgbe_82599.o ixgbe_82598.o ixgbe_phy.o ixgbe_sriov.o \
-              ixgbe_mbx.o ixgbe_x540.o ixgbe_x550.o ixgbe_lib.o ixgbe_ptp.o \
-              ixgbe_xsk.o
+ixgbe-y := ixgbe_main.o ixgbe_common.o ixgbe_ethtool.o \
+           ixgbe_82599.o ixgbe_82598.o ixgbe_phy.o ixgbe_sriov.o \
+           ixgbe_mbx.o ixgbe_x540.o ixgbe_x550.o ixgbe_lib.o ixgbe_ptp.o \
+           ixgbe_xsk.o
 
 ixgbe-$(CONFIG_IXGBE_DCB) +=  ixgbe_dcb.o ixgbe_dcb_82598.o \
                               ixgbe_dcb_82599.o ixgbe_dcb_nl.o
diff --git a/drivers/net/ethernet/intel/ixgbevf/Makefile b/drivers/net/ethernet/intel/ixgbevf/Makefile
index 186a4bb24fde..01d3e892f3fa 100644
--- a/drivers/net/ethernet/intel/ixgbevf/Makefile
+++ b/drivers/net/ethernet/intel/ixgbevf/Makefile
@@ -6,9 +6,5 @@
 
 obj-$(CONFIG_IXGBEVF) += ixgbevf.o
 
-ixgbevf-objs := vf.o \
-                mbx.o \
-                ethtool.o \
-                ixgbevf_main.o
+ixgbevf-y := vf.o mbx.o ethtool.o ixgbevf_main.o
 ixgbevf-$(CONFIG_IXGBEVF_IPSEC) += ipsec.o
-
diff --git a/drivers/net/ethernet/intel/libeth/Makefile b/drivers/net/ethernet/intel/libeth/Makefile
index cb99203d1dd2..52492b081132 100644
--- a/drivers/net/ethernet/intel/libeth/Makefile
+++ b/drivers/net/ethernet/intel/libeth/Makefile
@@ -3,4 +3,4 @@
 
 obj-$(CONFIG_LIBETH)		+= libeth.o
 
-libeth-objs			+= rx.o
+libeth-y			:= rx.o
diff --git a/drivers/net/ethernet/intel/libie/Makefile b/drivers/net/ethernet/intel/libie/Makefile
index bf42c5aeeedd..ffd27fab916a 100644
--- a/drivers/net/ethernet/intel/libie/Makefile
+++ b/drivers/net/ethernet/intel/libie/Makefile
@@ -3,4 +3,4 @@
 
 obj-$(CONFIG_LIBIE)	+= libie.o
 
-libie-objs		+= rx.o
+libie-y			:= rx.o
-- 
cgit v1.2.3-58-ga151


From 1d4ce389da2b84e0e24aad5e83fe9c742adc3f99 Mon Sep 17 00:00:00 2001
From: Jacob Keller <jacob.e.keller@intel.com>
Date: Fri, 7 Jun 2024 14:22:33 -0700
Subject: ice: add and use roundup_u64 instead of open coding equivalent

In ice_ptp_cfg_clkout(), the ice driver needs to calculate the nearest next
second of a current time value specified in nanoseconds. It implements this
using div64_u64, because the time value is a u64. It could use div_u64
since NSEC_PER_SEC is smaller than 32-bits.

Ideally this would be implemented directly with roundup(), but that can't
work on all platforms due to a division which requires using the specific
macros and functions due to platform restrictions, and to ensure that the
most appropriate and fast instructions are used.

The kernel doesn't currently provide any 64-bit equivalents for doing
roundup. Attempting to use roundup() on a 32-bit platform will result in a
link failure due to not having a direct 64-bit division.

The closest equivalent for this is DIV64_U64_ROUND_UP, which does a
division always rounding up. However, this only computes the division, and
forces use of the div64_u64 in cases where the divisor is a 32bit value and
could make use of div_u64.

Introduce DIV_U64_ROUND_UP based on div_u64, and then use it to implement
roundup_u64 which takes a u64 input value and a u32 rounding value.

The name roundup_u64 matches the naming scheme of div_u64, and future
patches could implement roundup64_u64 if they need to round by a multiple
that is greater than 32-bits.

Replace the logic in ice_ptp.c which does this equivalent with the newly
added roundup_u64.

Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240607-next-2024-06-03-intel-next-batch-v3-2-d1470cee3347@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_ptp.c |  3 +--
 include/linux/math64.h                   | 28 ++++++++++++++++++++++++++++
 2 files changed, 29 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_ptp.c b/drivers/net/ethernet/intel/ice/ice_ptp.c
index adbb9cffe20c..b7ab6fdf710d 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp.c
@@ -1773,8 +1773,7 @@ static int ice_ptp_cfg_clkout(struct ice_pf *pf, unsigned int chan,
 	 * maintaining phase
 	 */
 	if (start_time < current_time)
-		start_time = div64_u64(current_time + NSEC_PER_SEC - 1,
-				       NSEC_PER_SEC) * NSEC_PER_SEC + phase;
+		start_time = roundup_u64(current_time, NSEC_PER_SEC) + phase;
 
 	if (ice_is_e810(hw))
 		start_time -= E810_OUT_PROP_DELAY_NS;
diff --git a/include/linux/math64.h b/include/linux/math64.h
index d34def7f9a8c..6aaccc1626ab 100644
--- a/include/linux/math64.h
+++ b/include/linux/math64.h
@@ -297,6 +297,19 @@ u64 mul_u64_u64_div_u64(u64 a, u64 mul, u64 div);
 #define DIV64_U64_ROUND_UP(ll, d)	\
 	({ u64 _tmp = (d); div64_u64((ll) + _tmp - 1, _tmp); })
 
+/**
+ * DIV_U64_ROUND_UP - unsigned 64bit divide with 32bit divisor rounded up
+ * @ll: unsigned 64bit dividend
+ * @d: unsigned 32bit divisor
+ *
+ * Divide unsigned 64bit dividend by unsigned 32bit divisor
+ * and round up.
+ *
+ * Return: dividend / divisor rounded up
+ */
+#define DIV_U64_ROUND_UP(ll, d)		\
+	({ u32 _tmp = (d); div_u64((ll) + _tmp - 1, _tmp); })
+
 /**
  * DIV64_U64_ROUND_CLOSEST - unsigned 64bit divide with 64bit divisor rounded to nearest integer
  * @dividend: unsigned 64bit dividend
@@ -342,4 +355,19 @@ u64 mul_u64_u64_div_u64(u64 a, u64 mul, u64 div);
 		div_s64((__x - (__d / 2)), __d);	\
 }							\
 )
+
+/**
+ * roundup_u64 - Round up a 64bit value to the next specified 32bit multiple
+ * @x: the value to up
+ * @y: 32bit multiple to round up to
+ *
+ * Rounds @x to the next multiple of @y. For 32bit @x values, see roundup and
+ * the faster round_up() for powers of 2.
+ *
+ * Return: rounded up value.
+ */
+static inline u64 roundup_u64(u64 x, u32 y)
+{
+	return DIV_U64_ROUND_UP(x, y) * y;
+}
 #endif /* _LINUX_MATH64_H */
-- 
cgit v1.2.3-58-ga151


From dee55767dc8ca3fc6127a792e2a8b69c5e03fdd2 Mon Sep 17 00:00:00 2001
From: Michal Schmidt <mschmidt@redhat.com>
Date: Fri, 7 Jun 2024 14:22:34 -0700
Subject: ice: use irq_update_affinity_hint()

irq_set_affinity_hint() is deprecated. Use irq_update_affinity_hint()
instead. This removes the side-effect of actually applying the affinity.

The driver does not really need to worry about spreading its IRQs across
CPUs. The core code already takes care of that.
On the contrary, when the driver applies affinities by itself, it breaks
the users' expectations:
 1. The user configures irqbalance with IRQBALANCE_BANNED_CPULIST in
    order to prevent IRQs from being moved to certain CPUs that run a
    real-time workload.
 2. ice reconfigures VSIs at runtime due to a MIB change
    (ice_dcb_process_lldp_set_mib_change). Reopening a VSI resets the
    affinity in ice_vsi_req_irq_msix().
 3. ice has no idea about irqbalance's config, so it may move an IRQ to
    a banned CPU. The real-time workload suffers unacceptable latency.

I am not sure if updating the affinity hints is at all useful, because
irqbalance ignores them since 2016 ([1]), but at least it's harmless.

This ice change is similar to i40e commit d34c54d1739c ("i40e: Use
irq_update_affinity_hint()").

[1] https://github.com/Irqbalance/irqbalance/commit/dcc411e7bfdd

Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Sunil Goutham <sgoutham@marvell.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://lore.kernel.org/r/20240607-next-2024-06-03-intel-next-batch-v3-3-d1470cee3347@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_lib.c  | 4 ++--
 drivers/net/ethernet/intel/ice/ice_main.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index 7629b0190578..f559e60992fa 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -2580,8 +2580,8 @@ void ice_vsi_free_irq(struct ice_vsi *vsi)
 		if (!IS_ENABLED(CONFIG_RFS_ACCEL))
 			irq_set_affinity_notifier(irq_num, NULL);
 
-		/* clear the affinity_mask in the IRQ descriptor */
-		irq_set_affinity_hint(irq_num, NULL);
+		/* clear the affinity_hint in the IRQ descriptor */
+		irq_update_affinity_hint(irq_num, NULL);
 		synchronize_irq(irq_num);
 		devm_free_irq(ice_pf_to_dev(pf), irq_num, vsi->q_vectors[i]);
 	}
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 1b61ca3a6eb6..7d9a4e856f61 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -2607,7 +2607,7 @@ static int ice_vsi_req_irq_msix(struct ice_vsi *vsi, char *basename)
 		}
 
 		/* assign the mask for this irq */
-		irq_set_affinity_hint(irq_num, &q_vector->affinity_mask);
+		irq_update_affinity_hint(irq_num, &q_vector->affinity_mask);
 	}
 
 	err = ice_set_cpu_rx_rmap(vsi);
@@ -2625,7 +2625,7 @@ free_q_irqs:
 		irq_num = vsi->q_vectors[vector]->irq.virq;
 		if (!IS_ENABLED(CONFIG_RFS_ACCEL))
 			irq_set_affinity_notifier(irq_num, NULL);
-		irq_set_affinity_hint(irq_num, NULL);
+		irq_update_affinity_hint(irq_num, NULL);
 		devm_free_irq(dev, irq_num, &vsi->q_vectors[vector]);
 	}
 	return err;
-- 
cgit v1.2.3-58-ga151


From 4938ba733ee2147724c4b10025fe0d349d3b605f Mon Sep 17 00:00:00 2001
From: Rameshkumar Sundaram <quic_ramess@quicinc.com>
Date: Tue, 11 Jun 2024 09:42:31 +0300
Subject: wifi: ath12k: modify remain on channel for single wiphy

When multiple radios are advertised as a single wiphy which supports various
bands, vdev creation for the vif is deferred until channel is assigned to it.
If a remain on channel (RoC) request is received from mac80211, select the
corresponding radio (ar) based on channel and create a vdev on that radio to
initiate an RoC scan.

Note that on RoC completion this vdev is not deleted. If a new RoC/hw scan
request is seen on that same vif for a different band the vdev will be deleted
and created on the new radio supporting the request.

Also if the RoC scan is requested when the vdev is in started state, no
switching to new radio is allowed and RoC request can be accepted only on
channels within same radio.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Rameshkumar Sundaram <quic_ramess@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240528082739.1226758-1-quic_ramess@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 68 +++++++++++++++++++++++++++++++----
 1 file changed, 62 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 509c02bffdae..d1529612fddd 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -3372,7 +3372,7 @@ static void ath12k_mac_op_bss_info_changed(struct ieee80211_hw *hw,
 static struct ath12k*
 ath12k_mac_select_scan_device(struct ieee80211_hw *hw,
 			      struct ieee80211_vif *vif,
-			      struct ieee80211_scan_request *req)
+			      u32 center_freq)
 {
 	struct ath12k_hw *ah = hw->priv;
 	enum nl80211_band band;
@@ -3389,9 +3389,9 @@ ath12k_mac_select_scan_device(struct ieee80211_hw *hw,
 	 * split the hw request and perform multiple scans
 	 */
 
-	if (req->req.channels[0]->center_freq < ATH12K_MIN_5G_FREQ)
+	if (center_freq < ATH12K_MIN_5G_FREQ)
 		band = NL80211_BAND_2GHZ;
-	else if (req->req.channels[0]->center_freq < ATH12K_MIN_6G_FREQ)
+	else if (center_freq < ATH12K_MIN_6G_FREQ)
 		band = NL80211_BAND_5GHZ;
 	else
 		band = NL80211_BAND_6GHZ;
@@ -3591,7 +3591,7 @@ static int ath12k_mac_op_hw_scan(struct ieee80211_hw *hw,
 	/* Since the targeted scan device could depend on the frequency
 	 * requested in the hw_req, select the corresponding radio
 	 */
-	ar = ath12k_mac_select_scan_device(hw, vif, hw_req);
+	ar = ath12k_mac_select_scan_device(hw, vif, hw_req->req.channels[0]->center_freq);
 	if (!ar)
 		return -EINVAL;
 
@@ -8416,12 +8416,68 @@ static int ath12k_mac_op_remain_on_channel(struct ieee80211_hw *hw,
 	struct ath12k_vif *arvif = ath12k_vif_to_arvif(vif);
 	struct ath12k_hw *ah = ath12k_hw_to_ah(hw);
 	struct ath12k_wmi_scan_req_arg arg;
-	struct ath12k *ar;
+	struct ath12k *ar, *prev_ar;
 	u32 scan_time_msec;
+	bool create = true;
 	int ret;
 
-	ar = ath12k_ah_to_ar(ah, 0);
+	if (ah->num_radio == 1) {
+		WARN_ON(!arvif->is_created);
+		ar = ath12k_ah_to_ar(ah, 0);
+		goto scan;
+	}
+
+	ar = ath12k_mac_select_scan_device(hw, vif, chan->center_freq);
+	if (!ar)
+		return -EINVAL;
+
+	/* If the vif is already assigned to a specific vdev of an ar,
+	 * check whether its already started, vdev which is started
+	 * are not allowed to switch to a new radio.
+	 * If the vdev is not started, but was earlier created on a
+	 * different ar, delete that vdev and create a new one. We don't
+	 * delete at the scan stop as an optimization to avoid redundant
+	 * delete-create vdev's for the same ar, in case the request is
+	 * always on the same band for the vif
+	 */
+	if (arvif->is_created) {
+		if (WARN_ON(!arvif->ar))
+			return -EINVAL;
+
+		if (ar != arvif->ar && arvif->is_started)
+			return -EBUSY;
+
+		if (ar != arvif->ar) {
+			/* backup the previously used ar ptr, since the vdev delete
+			 * would assign the arvif->ar to NULL after the call
+			 */
+			prev_ar = arvif->ar;
+			mutex_lock(&prev_ar->conf_mutex);
+			ret = ath12k_mac_vdev_delete(prev_ar, vif);
+			mutex_unlock(&prev_ar->conf_mutex);
+			if (ret) {
+				ath12k_warn(prev_ar->ab,
+					    "unable to delete scan vdev for roc: %d\n",
+					    ret);
+				return ret;
+			}
+		} else {
+			create = false;
+		}
+	}
+
+	if (create) {
+		mutex_lock(&ar->conf_mutex);
+		ret = ath12k_mac_vdev_create(ar, vif);
+		mutex_unlock(&ar->conf_mutex);
+		if (ret) {
+			ath12k_warn(ar->ab, "unable to create scan vdev for roc: %d\n",
+				    ret);
+			return -EINVAL;
+		}
+	}
 
+scan:
 	mutex_lock(&ar->conf_mutex);
 	spin_lock_bh(&ar->data_lock);
 
-- 
cgit v1.2.3-58-ga151


From 8b2a12749b08726f006303814ecc1c37024d3617 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Tue, 11 Jun 2024 09:42:33 +0300
Subject: wifi: ath12k: fix ACPI warning when resume

Currently ACPI notification handler is installed when driver loads and only
gets removed when driver unloads. During resume after firmware is reloaded,
ath12k tries to install it by default. Since it is installed already, ACPI
subsystem rejects it and returns an error:

[   83.094206] ath12k_pci 0000:03:00.0: failed to install DSM notify callback: 7

Fix it by removing that handler when going to suspend. This also avoid any
possible ACPI call to firmware before firmware is reloaded/reinitialized.

Note ab->acpi also needs to be cleared in ath12k_acpi_stop() such that we
are in a clean state when ACPI structures are reinitialized in
ath12k_acpi_start().

Tested-on: WCN7850 HW2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Fixes: 576771c9fa21 ("wifi: ath12k: ACPI TAS support")
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240531024000.9291-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/acpi.c | 2 ++
 drivers/net/wireless/ath/ath12k/core.c | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/acpi.c b/drivers/net/wireless/ath/ath12k/acpi.c
index 443ba12e01f3..0555d35aab47 100644
--- a/drivers/net/wireless/ath/ath12k/acpi.c
+++ b/drivers/net/wireless/ath/ath12k/acpi.c
@@ -391,4 +391,6 @@ void ath12k_acpi_stop(struct ath12k_base *ab)
 	acpi_remove_notify_handler(ACPI_HANDLE(ab->dev),
 				   ACPI_DEVICE_NOTIFY,
 				   ath12k_acpi_dsm_notify);
+
+	memset(&ab->acpi, 0, sizeof(ab->acpi));
 }
diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c
index 4c3eab4686c2..da02d3528737 100644
--- a/drivers/net/wireless/ath/ath12k/core.c
+++ b/drivers/net/wireless/ath/ath12k/core.c
@@ -83,6 +83,8 @@ int ath12k_core_suspend_late(struct ath12k_base *ab)
 	if (!ab->hw_params->supports_suspend)
 		return -EOPNOTSUPP;
 
+	ath12k_acpi_stop(ab);
+
 	ath12k_hif_irq_disable(ab);
 	ath12k_hif_ce_irq_disable(ab);
 
-- 
cgit v1.2.3-58-ga151


From 53d7c99719e29819eaeb9f373d262a6b56ea7cce Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Tue, 11 Jun 2024 09:42:33 +0300
Subject: wifi: ath11k: fix RCU documentation in ath11k_mac_op_ipv6_changed()

Current documentation on RCU in ath11k_mac_op_ipv6_changed() says:

	/* Note: read_lock_bh() calls rcu_read_lock() */
	read_lock_bh(&idev->lock);

This is wrong because without enabling CONFIG_PREEMPT_RT
rcu_read_lock() is not called by read_lock_bh(). The reason
why current code works even in a CONFIG_PREEMPT_RT=n kernel
is because atomic_notifier_call_chain() already does that for
us, see:

	int atomic_notifier_call_chain()
	{
		...
		rcu_read_lock();
		ret = notifier_call_chain(&nh->head, val, v, -1, NULL);
		rcu_read_unlock();
		...
	}

and backtrace:

	ath11k_mac_op_ipv6_changed
	ieee80211_ifa6_changed
	notifier_call_chain
	atomic_notifier_call_chain

So update the comment to make it correct.

This is found during code review, compile tested only.

Fixes: feafe59c8975 ("wifi: ath11k: use RCU when accessing struct inet6_dev::ac_list")
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240531022411.6543-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath11k/mac.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index a1800c75d32b..29af015a16f9 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -9020,7 +9020,12 @@ static void ath11k_mac_op_ipv6_changed(struct ieee80211_hw *hw,
 	offload = &arvif->arp_ns_offload;
 	count = 0;
 
-	/* Note: read_lock_bh() calls rcu_read_lock() */
+	/* The _ipv6_changed() is called with RCU lock already held in
+	 * atomic_notifier_call_chain(), so we don't need to call
+	 * rcu_read_lock() again here. But note that with CONFIG_PREEMPT_RT
+	 * enabled, read_lock_bh() also calls rcu_read_lock(). This is OK
+	 * because RCU read critical section is allowed to get nested.
+	 */
 	read_lock_bh(&idev->lock);
 
 	memset(offload->ipv6_addr, 0, sizeof(offload->ipv6_addr));
-- 
cgit v1.2.3-58-ga151


From 98b303c9bf05dae932efbd71e18d81f6c64f20d8 Mon Sep 17 00:00:00 2001
From: Kenta Tada <tadakentaso@gmail.com>
Date: Fri, 7 Jun 2024 20:17:04 +0900
Subject: bpftool: Query only cgroup-related attach types

When CONFIG_NETKIT=y,
bpftool-cgroup shows error even if the cgroup's path is correct:

$ bpftool cgroup tree /sys/fs/cgroup
CgroupPath
ID       AttachType      AttachFlags     Name
Error: can't query bpf programs attached to /sys/fs/cgroup: No such device or address

>From strace and kernel tracing, I found netkit returned ENXIO and this command failed.
I think this AttachType(BPF_NETKIT_PRIMARY) is not relevant to cgroup.

bpftool-cgroup should query just only cgroup-related attach types.

v2->v3:
  - removed an unnecessary check

v1->v2:
  - used an array of cgroup attach types

Signed-off-by: Kenta Tada <tadakentaso@gmail.com>
Reviewed-by: Quentin Monnet <qmo@kernel.org>
Link: https://lore.kernel.org/r/20240607111704.6716-1-tadakentaso@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/cgroup.c | 40 ++++++++++++++++++++++++++++++++++++----
 1 file changed, 36 insertions(+), 4 deletions(-)

diff --git a/tools/bpf/bpftool/cgroup.c b/tools/bpf/bpftool/cgroup.c
index af6898c0f388..9af426d43299 100644
--- a/tools/bpf/bpftool/cgroup.c
+++ b/tools/bpf/bpftool/cgroup.c
@@ -19,6 +19,38 @@
 
 #include "main.h"
 
+static const int cgroup_attach_types[] = {
+	BPF_CGROUP_INET_INGRESS,
+	BPF_CGROUP_INET_EGRESS,
+	BPF_CGROUP_INET_SOCK_CREATE,
+	BPF_CGROUP_INET_SOCK_RELEASE,
+	BPF_CGROUP_INET4_BIND,
+	BPF_CGROUP_INET6_BIND,
+	BPF_CGROUP_INET4_POST_BIND,
+	BPF_CGROUP_INET6_POST_BIND,
+	BPF_CGROUP_INET4_CONNECT,
+	BPF_CGROUP_INET6_CONNECT,
+	BPF_CGROUP_UNIX_CONNECT,
+	BPF_CGROUP_INET4_GETPEERNAME,
+	BPF_CGROUP_INET6_GETPEERNAME,
+	BPF_CGROUP_UNIX_GETPEERNAME,
+	BPF_CGROUP_INET4_GETSOCKNAME,
+	BPF_CGROUP_INET6_GETSOCKNAME,
+	BPF_CGROUP_UNIX_GETSOCKNAME,
+	BPF_CGROUP_UDP4_SENDMSG,
+	BPF_CGROUP_UDP6_SENDMSG,
+	BPF_CGROUP_UNIX_SENDMSG,
+	BPF_CGROUP_UDP4_RECVMSG,
+	BPF_CGROUP_UDP6_RECVMSG,
+	BPF_CGROUP_UNIX_RECVMSG,
+	BPF_CGROUP_SOCK_OPS,
+	BPF_CGROUP_DEVICE,
+	BPF_CGROUP_SYSCTL,
+	BPF_CGROUP_GETSOCKOPT,
+	BPF_CGROUP_SETSOCKOPT,
+	BPF_LSM_CGROUP
+};
+
 #define HELP_SPEC_ATTACH_FLAGS						\
 	"ATTACH_FLAGS := { multi | override }"
 
@@ -183,13 +215,13 @@ static int count_attached_bpf_progs(int cgroup_fd, enum bpf_attach_type type)
 
 static int cgroup_has_attached_progs(int cgroup_fd)
 {
-	enum bpf_attach_type type;
+	unsigned int i = 0;
 	bool no_prog = true;
 
-	for (type = 0; type < __MAX_BPF_ATTACH_TYPE; type++) {
-		int count = count_attached_bpf_progs(cgroup_fd, type);
+	for (i = 0; i < ARRAY_SIZE(cgroup_attach_types); i++) {
+		int count = count_attached_bpf_progs(cgroup_fd, cgroup_attach_types[i]);
 
-		if (count < 0 && errno != EINVAL)
+		if (count < 0)
 			return -1;
 
 		if (count > 0) {
-- 
cgit v1.2.3-58-ga151


From 65a8368bf34e51b4499205658d7015d40e5a3f80 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Tue, 11 Jun 2024 09:42:33 +0300
Subject: wifi: ath11k: use 'time_left' variable with wait_event_timeout()

There is a confusing pattern in the kernel to use a variable named 'timeout' to
store the result of wait_event_timeout() causing patterns like:

	timeout = wait_event_timeout(...)
	if (!timeout) return -ETIMEDOUT;

with all kinds of permutations. Use 'time_left' as a variable to make the code
self explaining.

Fix to the proper variable type 'long' while here.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240603091541.8367-2-wsa+renesas@sang-engineering.com
---
 drivers/net/wireless/ath/ath11k/qmi.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/qmi.c b/drivers/net/wireless/ath/ath11k/qmi.c
index aa160e6fe24f..1bc648920ab6 100644
--- a/drivers/net/wireless/ath/ath11k/qmi.c
+++ b/drivers/net/wireless/ath/ath11k/qmi.c
@@ -2859,7 +2859,7 @@ int ath11k_qmi_firmware_start(struct ath11k_base *ab,
 
 int ath11k_qmi_fwreset_from_cold_boot(struct ath11k_base *ab)
 {
-	int timeout;
+	long time_left;
 
 	if (!ath11k_core_coldboot_cal_support(ab) ||
 	    ab->hw_params.cbcal_restart_fw == 0)
@@ -2867,11 +2867,11 @@ int ath11k_qmi_fwreset_from_cold_boot(struct ath11k_base *ab)
 
 	ath11k_dbg(ab, ATH11K_DBG_QMI, "wait for cold boot done\n");
 
-	timeout = wait_event_timeout(ab->qmi.cold_boot_waitq,
-				     (ab->qmi.cal_done == 1),
-				     ATH11K_COLD_BOOT_FW_RESET_DELAY);
+	time_left = wait_event_timeout(ab->qmi.cold_boot_waitq,
+				       (ab->qmi.cal_done == 1),
+				       ATH11K_COLD_BOOT_FW_RESET_DELAY);
 
-	if (timeout <= 0) {
+	if (time_left <= 0) {
 		ath11k_warn(ab, "Coldboot Calibration timed out\n");
 		return -ETIMEDOUT;
 	}
@@ -2886,7 +2886,7 @@ EXPORT_SYMBOL(ath11k_qmi_fwreset_from_cold_boot);
 
 static int ath11k_qmi_process_coldboot_calibration(struct ath11k_base *ab)
 {
-	int timeout;
+	long time_left;
 	int ret;
 
 	ret = ath11k_qmi_wlanfw_mode_send(ab, ATH11K_FIRMWARE_MODE_COLD_BOOT);
@@ -2897,10 +2897,10 @@ static int ath11k_qmi_process_coldboot_calibration(struct ath11k_base *ab)
 
 	ath11k_dbg(ab, ATH11K_DBG_QMI, "Coldboot calibration wait started\n");
 
-	timeout = wait_event_timeout(ab->qmi.cold_boot_waitq,
-				     (ab->qmi.cal_done  == 1),
-				     ATH11K_COLD_BOOT_FW_RESET_DELAY);
-	if (timeout <= 0) {
+	time_left = wait_event_timeout(ab->qmi.cold_boot_waitq,
+				       (ab->qmi.cal_done  == 1),
+				       ATH11K_COLD_BOOT_FW_RESET_DELAY);
+	if (time_left <= 0) {
 		ath11k_warn(ab, "coldboot calibration timed out\n");
 		return 0;
 	}
-- 
cgit v1.2.3-58-ga151


From 9e32058ae62f1d8fdf1ee7b177a1307f90335fbd Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Tue, 11 Jun 2024 09:42:33 +0300
Subject: wifi: ath12k: avoid unnecessary MSDU drop in the Rx error process

Currently, in the Rx error processing handler, once an MSDU drop is
detected, the subsequent MSDUs get unintentionally dropped due to the
previous drop flag being retained across all MSDU processing, leading
to the discarding of valid MSDUs. To resolve this issue, the drop flag
should be reset to false before processing each descriptor.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240604062641.2956288-1-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp_rx.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index cb1f308f096b..54aea3c22311 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -3402,7 +3402,7 @@ int ath12k_dp_rx_process_err(struct ath12k_base *ab, struct napi_struct *napi,
 	struct ath12k *ar;
 	dma_addr_t paddr;
 	bool is_frag;
-	bool drop = false;
+	bool drop;
 	int pdev_id;
 
 	tot_n_bufs_reaped = 0;
@@ -3420,7 +3420,9 @@ int ath12k_dp_rx_process_err(struct ath12k_base *ab, struct napi_struct *napi,
 
 	while (budget &&
 	       (reo_desc = ath12k_hal_srng_dst_get_next_entry(ab, srng))) {
+		drop = false;
 		ab->soc_stats.err_ring_pkts++;
+
 		ret = ath12k_hal_desc_reo_parse_err(ab, reo_desc, &paddr,
 						    &desc_bank);
 		if (ret) {
-- 
cgit v1.2.3-58-ga151


From d2b0ca38d362ebf16ca79cd7f309d5bb8b581deb Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Tue, 11 Jun 2024 09:42:34 +0300
Subject: wifi: ath11k: fix wrong handling of CCMP256 and GCMP ciphers

Currently for CCMP256, GCMP128 and GCMP256 ciphers, in ath11k_install_key()
IEEE80211_KEY_FLAG_GENERATE_IV_MGMT is not set. And in ath11k_mac_mgmt_tx_wmi()
a length of IEEE80211_CCMP_MIC_LEN is reserved for all ciphers.

This results in unexpected management frame drop in case either of above 3 ciphers
is used. The reason is, without IEEE80211_KEY_FLAG_GENERATE_IV_MGMT set, mac80211
will not generate CCMP/GCMP headers in frame for ath11k. Also MIC length reserved
is wrong. Such frame is dropped later by hardware:

ath11k_pci 0000:5a:00.0: mac tx mgmt frame, buf id 0
ath11k_pci 0000:5a:00.0: mgmt tx compl ev pdev_id 1, desc_id 0, status 1

From user point of view, we have observed very low throughput due to this issue:
action frames are all dropped so ADDBA response from DUT never reaches AP. AP
can not use aggregation thus throughput is low.

Fix this by setting IEEE80211_KEY_FLAG_GENERATE_IV_MGMT flag and by reserving proper
MIC length for those ciphers.

Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.30
Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Fixes: d5c65159f289 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Reported-by: Yaroslav Isakov <yaroslav.isakov@gmail.com>
Tested-by: Yaroslav Isakov <yaroslav.isakov@gmail.com>
Closes: https://lore.kernel.org/all/CADS+iDX5=JtJr0apAtAQ02WWBxgOFEv8G063vuGYwDTC8AVZaw@mail.gmail.com
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240605014826.22498-1-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dp_rx.c |  3 +--
 drivers/net/wireless/ath/ath11k/dp_rx.h |  3 +++
 drivers/net/wireless/ath/ath11k/mac.c   | 15 +++++++++++----
 3 files changed, 15 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.c b/drivers/net/wireless/ath/ath11k/dp_rx.c
index 198fb359a688..86485580dd89 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.c
@@ -1877,8 +1877,7 @@ static void ath11k_dp_rx_h_csum_offload(struct ath11k *ar, struct sk_buff *msdu)
 			  CHECKSUM_NONE : CHECKSUM_UNNECESSARY;
 }
 
-static int ath11k_dp_rx_crypto_mic_len(struct ath11k *ar,
-				       enum hal_encrypt_type enctype)
+int ath11k_dp_rx_crypto_mic_len(struct ath11k *ar, enum hal_encrypt_type enctype)
 {
 	switch (enctype) {
 	case HAL_ENCRYPT_TYPE_OPEN:
diff --git a/drivers/net/wireless/ath/ath11k/dp_rx.h b/drivers/net/wireless/ath/ath11k/dp_rx.h
index 623da3bf9dc8..c322e30caa96 100644
--- a/drivers/net/wireless/ath/ath11k/dp_rx.h
+++ b/drivers/net/wireless/ath/ath11k/dp_rx.h
@@ -1,6 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause-Clear */
 /*
  * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
+ * Copyright (c) 2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 #ifndef ATH11K_DP_RX_H
 #define ATH11K_DP_RX_H
@@ -95,4 +96,6 @@ int ath11k_peer_rx_frag_setup(struct ath11k *ar, const u8 *peer_mac, int vdev_id
 int ath11k_dp_rx_pktlog_start(struct ath11k_base *ab);
 int ath11k_dp_rx_pktlog_stop(struct ath11k_base *ab, bool stop_timer);
 
+int ath11k_dp_rx_crypto_mic_len(struct ath11k *ar, enum hal_encrypt_type enctype);
+
 #endif /* ATH11K_DP_RX_H */
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index 29af015a16f9..bb56160048ba 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -4229,6 +4229,7 @@ static int ath11k_install_key(struct ath11k_vif *arvif,
 
 	switch (key->cipher) {
 	case WLAN_CIPHER_SUITE_CCMP:
+	case WLAN_CIPHER_SUITE_CCMP_256:
 		arg.key_cipher = WMI_CIPHER_AES_CCM;
 		/* TODO: Re-check if flag is valid */
 		key->flags |= IEEE80211_KEY_FLAG_GENERATE_IV_MGMT;
@@ -4238,12 +4239,10 @@ static int ath11k_install_key(struct ath11k_vif *arvif,
 		arg.key_txmic_len = 8;
 		arg.key_rxmic_len = 8;
 		break;
-	case WLAN_CIPHER_SUITE_CCMP_256:
-		arg.key_cipher = WMI_CIPHER_AES_CCM;
-		break;
 	case WLAN_CIPHER_SUITE_GCMP:
 	case WLAN_CIPHER_SUITE_GCMP_256:
 		arg.key_cipher = WMI_CIPHER_AES_GCM;
+		key->flags |= IEEE80211_KEY_FLAG_GENERATE_IV_MGMT;
 		break;
 	default:
 		ath11k_warn(ar->ab, "cipher %d is not supported\n", key->cipher);
@@ -5903,7 +5902,10 @@ static int ath11k_mac_mgmt_tx_wmi(struct ath11k *ar, struct ath11k_vif *arvif,
 {
 	struct ath11k_base *ab = ar->ab;
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+	struct ath11k_skb_cb *skb_cb = ATH11K_SKB_CB(skb);
 	struct ieee80211_tx_info *info;
+	enum hal_encrypt_type enctype;
+	unsigned int mic_len;
 	dma_addr_t paddr;
 	int buf_id;
 	int ret;
@@ -5927,7 +5929,12 @@ static int ath11k_mac_mgmt_tx_wmi(struct ath11k *ar, struct ath11k_vif *arvif,
 		     ieee80211_is_deauth(hdr->frame_control) ||
 		     ieee80211_is_disassoc(hdr->frame_control)) &&
 		     ieee80211_has_protected(hdr->frame_control)) {
-			skb_put(skb, IEEE80211_CCMP_MIC_LEN);
+			if (!(skb_cb->flags & ATH11K_SKB_CIPHER_SET))
+				ath11k_warn(ab, "WMI management tx frame without ATH11K_SKB_CIPHER_SET");
+
+			enctype = ath11k_dp_tx_get_encrypt_type(skb_cb->cipher);
+			mic_len = ath11k_dp_rx_crypto_mic_len(ar, enctype);
+			skb_put(skb, mic_len);
 		}
 	}
 
-- 
cgit v1.2.3-58-ga151


From 8dd65ccbdf91e2fe3ab6e4da158b38f81746c3b4 Mon Sep 17 00:00:00 2001
From: Aditya Kumar Singh <quic_adisi@quicinc.com>
Date: Tue, 11 Jun 2024 09:42:32 +0300
Subject: wifi: ath12k: fix per pdev debugfs registration

Function ath12k_debugfs_register() is called once inside the function
ath12k_mac_hw_register(). However, with single wiphy model, there could
be multiple pdevs registered under single hardware. Hence, need to register
debugfs for each one of them.

Move the caller inside the loop which iterates over all underlying pdevs.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Fixes: f8bde02a26b9 ("wifi: ath12k: initial debugfs support")
Signed-off-by: Aditya Kumar Singh <quic_adisi@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240529043043.2488031-2-quic_adisi@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index d1529612fddd..4b71dfbfe728 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -9133,9 +9133,9 @@ static int ath12k_mac_hw_register(struct ath12k_hw *ah)
 			ath12k_err(ar->ab, "ath12k regd update failed: %d\n", ret);
 			goto err_unregister_hw;
 		}
-	}
 
-	ath12k_debugfs_register(ar);
+		ath12k_debugfs_register(ar);
+	}
 
 	return 0;
 
-- 
cgit v1.2.3-58-ga151


From 4e1eff38d37a27328251ea47602703b811a028e1 Mon Sep 17 00:00:00 2001
From: Aditya Kumar Singh <quic_adisi@quicinc.com>
Date: Tue, 11 Jun 2024 09:42:32 +0300
Subject: wifi: ath12k: unregister per pdev debugfs

During normal de-initialization path or if any error happens while
registering the hardware, there is no support to unregister the per pdev
debugfs. Add support for the same.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aditya Kumar Singh <quic_adisi@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240529043043.2488031-3-quic_adisi@quicinc.com
---
 drivers/net/wireless/ath/ath12k/debugfs.c | 10 ++++++++++
 drivers/net/wireless/ath/ath12k/debugfs.h |  6 +++++-
 drivers/net/wireless/ath/ath12k/mac.c     |  7 ++++++-
 3 files changed, 21 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/debugfs.c b/drivers/net/wireless/ath/ath12k/debugfs.c
index 8d8ba951093b..3b464cd40da6 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs.c
+++ b/drivers/net/wireless/ath/ath12k/debugfs.c
@@ -88,3 +88,13 @@ void ath12k_debugfs_register(struct ath12k *ar)
 				    &fops_simulate_radar);
 	}
 }
+
+void ath12k_debugfs_unregister(struct ath12k *ar)
+{
+	if (!ar->debug.debugfs_pdev)
+		return;
+
+	/* TODO: Remove symlink under ieee80211/phy* */
+	debugfs_remove_recursive(ar->debug.debugfs_pdev);
+	ar->debug.debugfs_pdev = NULL;
+}
diff --git a/drivers/net/wireless/ath/ath12k/debugfs.h b/drivers/net/wireless/ath/ath12k/debugfs.h
index a62f2a550b23..8d64ba03aa9a 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs.h
+++ b/drivers/net/wireless/ath/ath12k/debugfs.h
@@ -11,7 +11,7 @@
 void ath12k_debugfs_soc_create(struct ath12k_base *ab);
 void ath12k_debugfs_soc_destroy(struct ath12k_base *ab);
 void ath12k_debugfs_register(struct ath12k *ar);
-
+void ath12k_debugfs_unregister(struct ath12k *ar);
 #else
 static inline void ath12k_debugfs_soc_create(struct ath12k_base *ab)
 {
@@ -25,6 +25,10 @@ static inline void ath12k_debugfs_register(struct ath12k *ar)
 {
 }
 
+static inline void ath12k_debugfs_unregister(struct ath12k *ar)
+{
+}
+
 #endif /* CONFIG_ATH12K_DEBUGFS */
 
 #endif /* _ATH12K_DEBUGFS_H_ */
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 4b71dfbfe728..2259ce22cc8b 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -8892,8 +8892,10 @@ static void ath12k_mac_hw_unregister(struct ath12k_hw *ah)
 	struct ath12k *ar;
 	int i;
 
-	for_each_ar(ah, ar, i)
+	for_each_ar(ah, ar, i) {
 		cancel_work_sync(&ar->regd_update_work);
+		ath12k_debugfs_unregister(ar);
+	}
 
 	ieee80211_unregister_hw(hw);
 
@@ -9140,6 +9142,9 @@ static int ath12k_mac_hw_register(struct ath12k_hw *ah)
 	return 0;
 
 err_unregister_hw:
+	for_each_ar(ah, ar, i)
+		ath12k_debugfs_unregister(ar);
+
 	ieee80211_unregister_hw(hw);
 
 err_free_if_combs:
-- 
cgit v1.2.3-58-ga151


From 5a16da9cfb40d45e797df3ae1523ac64fe0b98ae Mon Sep 17 00:00:00 2001
From: Aditya Kumar Singh <quic_adisi@quicinc.com>
Date: Tue, 11 Jun 2024 09:42:33 +0300
Subject: wifi: ath12k: handle symlink cleanup for per pdev debugfs dentry

Whenever per pdev debugfs directory is created, a symlink to it is also
placed in ieee80211/phy* directory. During clean up of per pdev debugfs,
this symlink also needs to be cleaned up.

Add changes to clean up the symlink whenever the per pdev debugfs is
removed.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aditya Kumar Singh <quic_adisi@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240529043043.2488031-4-quic_adisi@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.h    | 1 +
 drivers/net/wireless/ath/ath12k/debugfs.c | 8 ++++++--
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index d03326a68a9f..b483899ca136 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -474,6 +474,7 @@ struct ath12k_fw_stats {
 
 struct ath12k_debug {
 	struct dentry *debugfs_pdev;
+	struct dentry *debugfs_pdev_symlink;
 };
 
 struct ath12k_per_peer_tx_stats {
diff --git a/drivers/net/wireless/ath/ath12k/debugfs.c b/drivers/net/wireless/ath/ath12k/debugfs.c
index 3b464cd40da6..751a9b04386a 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs.c
+++ b/drivers/net/wireless/ath/ath12k/debugfs.c
@@ -80,7 +80,9 @@ void ath12k_debugfs_register(struct ath12k *ar)
 
 	/* Create a symlink under ieee80211/phy* */
 	scnprintf(buf, sizeof(buf), "../../ath12k/%pd2", ar->debug.debugfs_pdev);
-	debugfs_create_symlink("ath12k", hw->wiphy->debugfsdir, buf);
+	ar->debug.debugfs_pdev_symlink = debugfs_create_symlink("ath12k",
+								hw->wiphy->debugfsdir,
+								buf);
 
 	if (ar->mac.sbands[NL80211_BAND_5GHZ].channels) {
 		debugfs_create_file("dfs_simulate_radar", 0200,
@@ -94,7 +96,9 @@ void ath12k_debugfs_unregister(struct ath12k *ar)
 	if (!ar->debug.debugfs_pdev)
 		return;
 
-	/* TODO: Remove symlink under ieee80211/phy* */
+	/* Remove symlink under ieee80211/phy* */
+	debugfs_remove(ar->debug.debugfs_pdev_symlink);
 	debugfs_remove_recursive(ar->debug.debugfs_pdev);
+	ar->debug.debugfs_pdev_symlink = NULL;
 	ar->debug.debugfs_pdev = NULL;
 }
-- 
cgit v1.2.3-58-ga151


From 983e44f0ee003bf6ca28519fbcabaa7adb77827b Mon Sep 17 00:00:00 2001
From: Chris Packham <chris.packham@alliedtelesis.co.nz>
Date: Fri, 7 Jun 2024 14:08:43 +1200
Subject: net: dsa: Fix typo in NET_DSA_TAG_RTL4_A Kconfig

Fix a minor typo in the help text for the NET_DSA_TAG_RTL4_A config
option.

Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Link: https://lore.kernel.org/r/20240607020843.1380735-1-chris.packham@alliedtelesis.co.nz
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/dsa/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig
index 8e698bea99a3..8d5bf869eb14 100644
--- a/net/dsa/Kconfig
+++ b/net/dsa/Kconfig
@@ -129,7 +129,7 @@ config NET_DSA_TAG_RTL4_A
 	tristate "Tag driver for Realtek 4 byte protocol A tags"
 	help
 	  Say Y or M if you want to enable support for tagging frames for the
-	  Realtek switches with 4 byte protocol A tags, sich as found in
+	  Realtek switches with 4 byte protocol A tags, such as found in
 	  the Realtek RTL8366RB.
 
 config NET_DSA_TAG_RTL8_4
-- 
cgit v1.2.3-58-ga151


From 45403b12c29c5d9510ace1ad767ea4b13a4caf38 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Fri, 7 Jun 2024 01:44:19 -0700
Subject: ip_tunnel: Move stats allocation to core

With commit 34d21de99cea9 ("net: Move {l,t,d}stats allocation to core and
convert veth & vrf"), stats allocation could be done on net core instead
of this driver.

With this new approach, the driver doesn't have to bother with error
handling (allocation failure checking, making sure free happens in the
right spot, etc). This is core responsibility now.

Move ip_tunnel driver to leverage the core allocation.

All the ip_tunnel_init() users call ip_tunnel_init() as part of their
.ndo_init callback. The .ndo_init callback is called before the stats
allocation in netdev_register(), thus, the allocation will happen before
the netdev is visible.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240607084420.3932875-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/ip_tunnel.c | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
index bccef2fcf620..5cffad42fe8c 100644
--- a/net/ipv4/ip_tunnel.c
+++ b/net/ipv4/ip_tunnel.c
@@ -1099,7 +1099,6 @@ static void ip_tunnel_dev_free(struct net_device *dev)
 
 	gro_cells_destroy(&tunnel->gro_cells);
 	dst_cache_destroy(&tunnel->dst_cache);
-	free_percpu(dev->tstats);
 }
 
 void ip_tunnel_dellink(struct net_device *dev, struct list_head *head)
@@ -1313,20 +1312,15 @@ int ip_tunnel_init(struct net_device *dev)
 
 	dev->needs_free_netdev = true;
 	dev->priv_destructor = ip_tunnel_dev_free;
-	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
-	if (!dev->tstats)
-		return -ENOMEM;
+	dev->pcpu_stat_type = NETDEV_PCPU_STAT_TSTATS;
 
 	err = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL);
-	if (err) {
-		free_percpu(dev->tstats);
+	if (err)
 		return err;
-	}
 
 	err = gro_cells_init(&tunnel->gro_cells, dev);
 	if (err) {
 		dst_cache_destroy(&tunnel->dst_cache);
-		free_percpu(dev->tstats);
 		return err;
 	}
 
-- 
cgit v1.2.3-58-ga151


From fa59dc2f6fc616fde3941f62ccd4adcaa21ccd47 Mon Sep 17 00:00:00 2001
From: Jeremy Kerr <jk@codeconstruct.com.au>
Date: Fri, 7 Jun 2024 18:25:24 +0800
Subject: net: core,vrf: Change pcpu_dstat fields to u64_stats_t

The pcpu_sw_netstats and pcpu_lstats structs both contain a set of
u64_stats_t fields for individual stats, but pcpu_dstats uses u64s
instead.

Make this consistent by using u64_stats_t across all stats types.

The per-cpu dstats are only used by the vrf driver at present, so update
that driver as part of this change.

Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240607-dstats-v3-1-cc781fe116f7@codeconstruct.com.au
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/vrf.c         | 38 ++++++++++++++++++++++----------------
 include/linux/netdevice.h | 12 ++++++------
 2 files changed, 28 insertions(+), 22 deletions(-)

diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c
index 3a252ac5dd28..5018831b2a79 100644
--- a/drivers/net/vrf.c
+++ b/drivers/net/vrf.c
@@ -126,8 +126,8 @@ static void vrf_rx_stats(struct net_device *dev, int len)
 	struct pcpu_dstats *dstats = this_cpu_ptr(dev->dstats);
 
 	u64_stats_update_begin(&dstats->syncp);
-	dstats->rx_packets++;
-	dstats->rx_bytes += len;
+	u64_stats_inc(&dstats->rx_packets);
+	u64_stats_add(&dstats->rx_bytes, len);
 	u64_stats_update_end(&dstats->syncp);
 }
 
@@ -150,11 +150,11 @@ static void vrf_get_stats64(struct net_device *dev,
 		dstats = per_cpu_ptr(dev->dstats, i);
 		do {
 			start = u64_stats_fetch_begin(&dstats->syncp);
-			tbytes = dstats->tx_bytes;
-			tpkts = dstats->tx_packets;
-			tdrops = dstats->tx_drops;
-			rbytes = dstats->rx_bytes;
-			rpkts = dstats->rx_packets;
+			tbytes = u64_stats_read(&dstats->tx_bytes);
+			tpkts = u64_stats_read(&dstats->tx_packets);
+			tdrops = u64_stats_read(&dstats->tx_drops);
+			rbytes = u64_stats_read(&dstats->rx_bytes);
+			rpkts = u64_stats_read(&dstats->rx_packets);
 		} while (u64_stats_fetch_retry(&dstats->syncp, start));
 		stats->tx_bytes += tbytes;
 		stats->tx_packets += tpkts;
@@ -408,10 +408,15 @@ static int vrf_local_xmit(struct sk_buff *skb, struct net_device *dev,
 
 	skb->protocol = eth_type_trans(skb, dev);
 
-	if (likely(__netif_rx(skb) == NET_RX_SUCCESS))
+	if (likely(__netif_rx(skb) == NET_RX_SUCCESS)) {
 		vrf_rx_stats(dev, len);
-	else
-		this_cpu_inc(dev->dstats->rx_drops);
+	} else {
+		struct pcpu_dstats *dstats = this_cpu_ptr(dev->dstats);
+
+		u64_stats_update_begin(&dstats->syncp);
+		u64_stats_inc(&dstats->rx_drops);
+		u64_stats_update_end(&dstats->syncp);
+	}
 
 	return NETDEV_TX_OK;
 }
@@ -599,19 +604,20 @@ static netdev_tx_t is_ip_tx_frame(struct sk_buff *skb, struct net_device *dev)
 
 static netdev_tx_t vrf_xmit(struct sk_buff *skb, struct net_device *dev)
 {
+	struct pcpu_dstats *dstats = this_cpu_ptr(dev->dstats);
+
 	int len = skb->len;
 	netdev_tx_t ret = is_ip_tx_frame(skb, dev);
 
+	u64_stats_update_begin(&dstats->syncp);
 	if (likely(ret == NET_XMIT_SUCCESS || ret == NET_XMIT_CN)) {
-		struct pcpu_dstats *dstats = this_cpu_ptr(dev->dstats);
 
-		u64_stats_update_begin(&dstats->syncp);
-		dstats->tx_packets++;
-		dstats->tx_bytes += len;
-		u64_stats_update_end(&dstats->syncp);
+		u64_stats_inc(&dstats->tx_packets);
+		u64_stats_add(&dstats->tx_bytes, len);
 	} else {
-		this_cpu_inc(dev->dstats->tx_drops);
+		u64_stats_inc(&dstats->tx_drops);
 	}
+	u64_stats_update_end(&dstats->syncp);
 
 	return ret;
 }
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index d20c6c99eb88..f148a01dd1d1 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2731,12 +2731,12 @@ struct pcpu_sw_netstats {
 } __aligned(4 * sizeof(u64));
 
 struct pcpu_dstats {
-	u64			rx_packets;
-	u64			rx_bytes;
-	u64			rx_drops;
-	u64			tx_packets;
-	u64			tx_bytes;
-	u64			tx_drops;
+	u64_stats_t		rx_packets;
+	u64_stats_t		rx_bytes;
+	u64_stats_t		rx_drops;
+	u64_stats_t		tx_packets;
+	u64_stats_t		tx_bytes;
+	u64_stats_t		tx_drops;
 	struct u64_stats_sync	syncp;
 } __aligned(8 * sizeof(u64));
 
-- 
cgit v1.2.3-58-ga151


From 94b601bc4f8528262b2b83194b7fd3fc2a6da75a Mon Sep 17 00:00:00 2001
From: Jeremy Kerr <jk@codeconstruct.com.au>
Date: Fri, 7 Jun 2024 18:25:25 +0800
Subject: net: core: Implement dstats-type stats collections

We currently have dev_get_tstats64() for collecting per-cpu stats of
type pcpu_sw_netstats ("tstats"). However, tstats doesn't allow for
accounting tx/rx drops. We do have a stats variant that does have stats
for dropped packets: struct pcpu_dstats, but there are no core helpers
for using those stats.

The VRF driver uses dstats, by providing its own collation/fetch
functions to do so.

This change adds a common implementation for dstats-type collection,
used when pcpu_stat_type == NETDEV_PCPU_STAT_DSTAT. This is based on the
VRF driver's existing stats collator (plus the unused tx_drops stat from
there). We will switch the VRF driver to use this in the next change.

Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240607-dstats-v3-2-cc781fe116f7@codeconstruct.com.au
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/dev.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 50 insertions(+)

diff --git a/net/core/dev.c b/net/core/dev.c
index e62698c7a0e6..c361a7b69da8 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -10703,6 +10703,54 @@ void netdev_run_todo(void)
 		wake_up(&netdev_unregistering_wq);
 }
 
+/* Collate per-cpu network dstats statistics
+ *
+ * Read per-cpu network statistics from dev->dstats and populate the related
+ * fields in @s.
+ */
+static void dev_fetch_dstats(struct rtnl_link_stats64 *s,
+			     const struct pcpu_dstats __percpu *dstats)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		u64 rx_packets, rx_bytes, rx_drops;
+		u64 tx_packets, tx_bytes, tx_drops;
+		const struct pcpu_dstats *stats;
+		unsigned int start;
+
+		stats = per_cpu_ptr(dstats, cpu);
+		do {
+			start = u64_stats_fetch_begin(&stats->syncp);
+			rx_packets = u64_stats_read(&stats->rx_packets);
+			rx_bytes   = u64_stats_read(&stats->rx_bytes);
+			rx_drops   = u64_stats_read(&stats->rx_drops);
+			tx_packets = u64_stats_read(&stats->tx_packets);
+			tx_bytes   = u64_stats_read(&stats->tx_bytes);
+			tx_drops   = u64_stats_read(&stats->tx_drops);
+		} while (u64_stats_fetch_retry(&stats->syncp, start));
+
+		s->rx_packets += rx_packets;
+		s->rx_bytes   += rx_bytes;
+		s->rx_dropped += rx_drops;
+		s->tx_packets += tx_packets;
+		s->tx_bytes   += tx_bytes;
+		s->tx_dropped += tx_drops;
+	}
+}
+
+/* ndo_get_stats64 implementation for dtstats-based accounting.
+ *
+ * Populate @s from dev->stats and dev->dstats. This is used internally by the
+ * core for NETDEV_PCPU_STAT_DSTAT-type stats collection.
+ */
+static void dev_get_dstats64(const struct net_device *dev,
+			     struct rtnl_link_stats64 *s)
+{
+	netdev_stats_to_stats64(s, &dev->stats);
+	dev_fetch_dstats(s, dev->dstats);
+}
+
 /* Convert net_device_stats to rtnl_link_stats64. rtnl_link_stats64 has
  * all the same fields in the same order as net_device_stats, with only
  * the type differing, but rtnl_link_stats64 may have additional fields
@@ -10779,6 +10827,8 @@ struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev,
 		netdev_stats_to_stats64(storage, ops->ndo_get_stats(dev));
 	} else if (dev->pcpu_stat_type == NETDEV_PCPU_STAT_TSTATS) {
 		dev_get_tstats64(dev, storage);
+	} else if (dev->pcpu_stat_type == NETDEV_PCPU_STAT_DSTATS) {
+		dev_get_dstats64(dev, storage);
 	} else {
 		netdev_stats_to_stats64(storage, &dev->stats);
 	}
-- 
cgit v1.2.3-58-ga151


From 2202576d4631b977a95162a387c0c9fb4ca6819c Mon Sep 17 00:00:00 2001
From: Jeremy Kerr <jk@codeconstruct.com.au>
Date: Fri, 7 Jun 2024 18:25:26 +0800
Subject: net: vrf: move to generic dstat helpers

The vrf driver has its own dstats-to-rtnl_link_stats64 collection, but
we now have a generic implementation for dstats collection, so switch to
this.

In doing so, we fix a minor issue where the (non-percpu)
dev->stats->tx_errors value was never collected into rtnl_link_stats64,
as the generic dev_get_dstats64() consumes the starting values from
dev->stats.

Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240607-dstats-v3-3-cc781fe116f7@codeconstruct.com.au
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/vrf.c | 28 ----------------------------
 1 file changed, 28 deletions(-)

diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c
index 5018831b2a79..9af316cdd8b3 100644
--- a/drivers/net/vrf.c
+++ b/drivers/net/vrf.c
@@ -137,33 +137,6 @@ static void vrf_tx_error(struct net_device *vrf_dev, struct sk_buff *skb)
 	kfree_skb(skb);
 }
 
-static void vrf_get_stats64(struct net_device *dev,
-			    struct rtnl_link_stats64 *stats)
-{
-	int i;
-
-	for_each_possible_cpu(i) {
-		const struct pcpu_dstats *dstats;
-		u64 tbytes, tpkts, tdrops, rbytes, rpkts;
-		unsigned int start;
-
-		dstats = per_cpu_ptr(dev->dstats, i);
-		do {
-			start = u64_stats_fetch_begin(&dstats->syncp);
-			tbytes = u64_stats_read(&dstats->tx_bytes);
-			tpkts = u64_stats_read(&dstats->tx_packets);
-			tdrops = u64_stats_read(&dstats->tx_drops);
-			rbytes = u64_stats_read(&dstats->rx_bytes);
-			rpkts = u64_stats_read(&dstats->rx_packets);
-		} while (u64_stats_fetch_retry(&dstats->syncp, start));
-		stats->tx_bytes += tbytes;
-		stats->tx_packets += tpkts;
-		stats->tx_dropped += tdrops;
-		stats->rx_bytes += rbytes;
-		stats->rx_packets += rpkts;
-	}
-}
-
 static struct vrf_map *netns_vrf_map(struct net *net)
 {
 	struct netns_vrf *nn_vrf = net_generic(net, vrf_net_id);
@@ -1201,7 +1174,6 @@ static const struct net_device_ops vrf_netdev_ops = {
 	.ndo_uninit		= vrf_dev_uninit,
 	.ndo_start_xmit		= vrf_xmit,
 	.ndo_set_mac_address	= eth_mac_addr,
-	.ndo_get_stats64	= vrf_get_stats64,
 	.ndo_add_slave		= vrf_add_slave,
 	.ndo_del_slave		= vrf_del_slave,
 };
-- 
cgit v1.2.3-58-ga151


From 7e0620bc6a5ec6b340a0be40054f294ca26c010f Mon Sep 17 00:00:00 2001
From: "Matthieu Baerts (NGI0)" <matttbe@kernel.org>
Date: Fri, 7 Jun 2024 18:31:02 +0200
Subject: selftests: net: lib: ignore possible errors

No need to disable errexit temporary, simply ignore the only possible
and not handled error.

Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://lore.kernel.org/r/20240607-upstream-net-next-20240607-selftests-mptcp-net-lib-v1-1-e36986faac94@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/lib.sh | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/tools/testing/selftests/net/lib.sh b/tools/testing/selftests/net/lib.sh
index 9155c914c064..b2572aff6286 100644
--- a/tools/testing/selftests/net/lib.sh
+++ b/tools/testing/selftests/net/lib.sh
@@ -128,25 +128,17 @@ slowwait_for_counter()
 cleanup_ns()
 {
 	local ns=""
-	local errexit=0
 	local ret=0
 
-	# disable errexit temporary
-	if [[ $- =~ "e" ]]; then
-		errexit=1
-		set +e
-	fi
-
 	for ns in "$@"; do
 		[ -z "${ns}" ] && continue
-		ip netns delete "${ns}" &> /dev/null
+		ip netns delete "${ns}" &> /dev/null || true
 		if ! busywait $BUSYWAIT_TIMEOUT ip netns list \| grep -vq "^$ns$" &> /dev/null; then
 			echo "Warn: Failed to remove namespace $ns"
 			ret=1
 		fi
 	done
 
-	[ $errexit -eq 1 ] && set -e
 	return $ret
 }
 
-- 
cgit v1.2.3-58-ga151


From 92fe5670271a58bb951646225ea367bb86f7feb3 Mon Sep 17 00:00:00 2001
From: "Matthieu Baerts (NGI0)" <matttbe@kernel.org>
Date: Fri, 7 Jun 2024 18:31:03 +0200
Subject: selftests: net: lib: remove ns from list after clean-up

Instead of only appending items to the list, removing them when the
netns has been deleted.

By doing that, we can make sure 'cleanup_all_ns()' is not trying to
remove already deleted netns.

Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://lore.kernel.org/r/20240607-upstream-net-next-20240607-selftests-mptcp-net-lib-v1-2-e36986faac94@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/lib.sh | 21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/net/lib.sh b/tools/testing/selftests/net/lib.sh
index b2572aff6286..c7a8cfb477cc 100644
--- a/tools/testing/selftests/net/lib.sh
+++ b/tools/testing/selftests/net/lib.sh
@@ -125,6 +125,20 @@ slowwait_for_counter()
 	slowwait "$timeout" until_counter_is ">= $((base + delta))" "$@"
 }
 
+remove_ns_list()
+{
+	local item=$1
+	local ns
+	local ns_list=("${NS_LIST[@]}")
+	NS_LIST=()
+
+	for ns in "${ns_list[@]}"; do
+		if [ "${ns}" != "${item}" ]; then
+			NS_LIST+=("${ns}")
+		fi
+	done
+}
+
 cleanup_ns()
 {
 	local ns=""
@@ -136,6 +150,8 @@ cleanup_ns()
 		if ! busywait $BUSYWAIT_TIMEOUT ip netns list \| grep -vq "^$ns$" &> /dev/null; then
 			echo "Warn: Failed to remove namespace $ns"
 			ret=1
+		else
+			remove_ns_list "${ns}"
 		fi
 	done
 
@@ -154,17 +170,14 @@ setup_ns()
 	local ns=""
 	local ns_name=""
 	local ns_list=()
-	local ns_exist=
 	for ns_name in "$@"; do
 		# Some test may setup/remove same netns multi times
 		if unset ${ns_name} 2> /dev/null; then
 			ns="${ns_name,,}-$(mktemp -u XXXXXX)"
 			eval readonly ${ns_name}="$ns"
-			ns_exist=false
 		else
 			eval ns='$'${ns_name}
 			cleanup_ns "$ns"
-			ns_exist=true
 		fi
 
 		if ! ip netns add "$ns"; then
@@ -173,7 +186,7 @@ setup_ns()
 			return $ksft_skip
 		fi
 		ip -n "$ns" link set lo up
-		! $ns_exist && ns_list+=("$ns")
+		ns_list+=("$ns")
 	done
 	NS_LIST+=("${ns_list[@]}")
 }
-- 
cgit v1.2.3-58-ga151


From 577db6bd5750b6de70f7017dfa2ad98d8934993b Mon Sep 17 00:00:00 2001
From: "Matthieu Baerts (NGI0)" <matttbe@kernel.org>
Date: Fri, 7 Jun 2024 18:31:04 +0200
Subject: selftests: net: lib: do not set ns var as readonly

It sounds good to mark the global netns variable as 'readonly', but Bash
doesn't allow the creation of local variables with the same name.

Because it looks like 'readonly' is mainly used here to check if a netns
with that name has already been set, it sounds fine to check if a
variable with this name has already been set instead. By doing that, we
avoid having to modify helpers from MPTCP selftests using the same
variable name as the one used to store the created netns name.

While at it, also avoid an unnecessary call to 'eval' to set a local
variable.

Reviewed-by: Geliang Tang <geliang@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://lore.kernel.org/r/20240607-upstream-net-next-20240607-selftests-mptcp-net-lib-v1-3-e36986faac94@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/lib.sh | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/net/lib.sh b/tools/testing/selftests/net/lib.sh
index c7a8cfb477cc..114b927fee25 100644
--- a/tools/testing/selftests/net/lib.sh
+++ b/tools/testing/selftests/net/lib.sh
@@ -172,11 +172,11 @@ setup_ns()
 	local ns_list=()
 	for ns_name in "$@"; do
 		# Some test may setup/remove same netns multi times
-		if unset ${ns_name} 2> /dev/null; then
+		if [ -z "${!ns_name}" ]; then
 			ns="${ns_name,,}-$(mktemp -u XXXXXX)"
-			eval readonly ${ns_name}="$ns"
+			eval "${ns_name}=${ns}"
 		else
-			eval ns='$'${ns_name}
+			ns="${!ns_name}"
 			cleanup_ns "$ns"
 		fi
 
-- 
cgit v1.2.3-58-ga151


From f8a2d2f874b710b31dcc482af7a59acedb2640ff Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Fri, 7 Jun 2024 18:31:05 +0200
Subject: selftests: net: lib: remove 'ns' var in setup_ns

The helper setup_ns() doesn't work when a net namespace named "ns" is
passed to it.

For example, in net/mptcp/diag.sh, the name of the namespace is "ns". If
"setup_ns ns" is used in it, diag.sh fails with errors:

  Invalid netns name "./mptcp_connect"
  Cannot open network namespace "10000": No such file or directory
  Cannot open network namespace "10000": No such file or directory

That is because "ns" is also a local variable in setup_ns, and it will
not set the value for the global variable that has been giving in
argument. To solve this, we could rename the variable, but it sounds
better to drop it, as we can resolve the name using the variable passed
in argument instead.

The other local variables -- "ns_list" and "ns_name" -- are more
unlikely to conflict with existing global variables. They don't seem to
be currently used in any other net selftests.

Co-developed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://lore.kernel.org/r/20240607-upstream-net-next-20240607-selftests-mptcp-net-lib-v1-4-e36986faac94@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/lib.sh | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/net/lib.sh b/tools/testing/selftests/net/lib.sh
index 114b927fee25..915f319bcc8b 100644
--- a/tools/testing/selftests/net/lib.sh
+++ b/tools/testing/selftests/net/lib.sh
@@ -167,26 +167,30 @@ cleanup_all_ns()
 # setup_ns local remote
 setup_ns()
 {
-	local ns=""
 	local ns_name=""
 	local ns_list=()
 	for ns_name in "$@"; do
+		# avoid conflicts with local var: internal error
+		if [ "${ns_name}" = "ns_name" ]; then
+			echo "Failed to setup namespace '${ns_name}': invalid name"
+			cleanup_ns "${ns_list[@]}"
+			exit $ksft_fail
+		fi
+
 		# Some test may setup/remove same netns multi times
 		if [ -z "${!ns_name}" ]; then
-			ns="${ns_name,,}-$(mktemp -u XXXXXX)"
-			eval "${ns_name}=${ns}"
+			eval "${ns_name}=${ns_name,,}-$(mktemp -u XXXXXX)"
 		else
-			ns="${!ns_name}"
-			cleanup_ns "$ns"
+			cleanup_ns "${!ns_name}"
 		fi
 
-		if ! ip netns add "$ns"; then
+		if ! ip netns add "${!ns_name}"; then
 			echo "Failed to create namespace $ns_name"
 			cleanup_ns "${ns_list[@]}"
 			return $ksft_skip
 		fi
-		ip -n "$ns" link set lo up
-		ns_list+=("$ns")
+		ip -n "${!ns_name}" link set lo up
+		ns_list+=("${!ns_name}")
 	done
 	NS_LIST+=("${ns_list[@]}")
 }
-- 
cgit v1.2.3-58-ga151


From f265d3119a29f1882cab156cb371a77865e17a13 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Fri, 7 Jun 2024 18:31:06 +0200
Subject: selftests: mptcp: lib: use setup/cleanup_ns helpers

This patch includes lib.sh into mptcp_lib.sh, uses setup_ns helper
defined in lib.sh to set up namespaces in mptcp_lib_ns_init(), and
uses cleanup_ns to delete namespaces in mptcp_lib_ns_exit().

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://lore.kernel.org/r/20240607-upstream-net-next-20240607-selftests-mptcp-net-lib-v1-5-e36986faac94@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/mptcp/mptcp_lib.sh | 17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_lib.sh b/tools/testing/selftests/net/mptcp/mptcp_lib.sh
index 6ffa9b7a3260..d9e30516dc72 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_lib.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_lib.sh
@@ -1,6 +1,8 @@
 #! /bin/bash
 # SPDX-License-Identifier: GPL-2.0
 
+. "$(dirname "${0}")/../lib.sh"
+
 readonly KSFT_PASS=0
 readonly KSFT_FAIL=1
 readonly KSFT_SKIP=4
@@ -438,17 +440,13 @@ mptcp_lib_check_tools() {
 }
 
 mptcp_lib_ns_init() {
-	local sec rndh
-
-	sec=$(date +%s)
-	rndh=$(printf %x "${sec}")-$(mktemp -u XXXXXX)
+	if ! setup_ns ${@}; then
+		mptcp_lib_pr_fail "Failed to setup namespace ${@}"
+		exit ${KSFT_FAIL}
+	fi
 
 	local netns
 	for netns in "${@}"; do
-		eval "${netns}=${netns}-${rndh}"
-
-		ip netns add "${!netns}" || exit ${KSFT_SKIP}
-		ip -net "${!netns}" link set lo up
 		ip netns exec "${!netns}" sysctl -q net.mptcp.enabled=1
 		ip netns exec "${!netns}" sysctl -q net.ipv4.conf.all.rp_filter=0
 		ip netns exec "${!netns}" sysctl -q net.ipv4.conf.default.rp_filter=0
@@ -456,9 +454,10 @@ mptcp_lib_ns_init() {
 }
 
 mptcp_lib_ns_exit() {
+	cleanup_ns "${@}"
+
 	local netns
 	for netns in "${@}"; do
-		ip netns del "${netns}"
 		rm -f /tmp/"${netns}".{nstat,out}
 	done
 }
-- 
cgit v1.2.3-58-ga151


From 1af3bc912eac91de6eb2a67a0cc7cfd5e1f11f2c Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Fri, 7 Jun 2024 18:31:07 +0200
Subject: selftests: mptcp: lib: use wait_local_port_listen helper

This patch includes net_helper.sh into mptcp_lib.sh, uses the helper
wait_local_port_listen() defined in it to implement the similar mptcp
helper. This can drop some duplicate code.

It looks like this helper from net_helper.sh was originally coming from
MPTCP, but MPTCP selftests have not been updated to use it from this
shared place.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://lore.kernel.org/r/20240607-upstream-net-next-20240607-selftests-mptcp-net-lib-v1-6-e36986faac94@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/mptcp/mptcp_lib.sh | 16 ++--------------
 1 file changed, 2 insertions(+), 14 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_lib.sh b/tools/testing/selftests/net/mptcp/mptcp_lib.sh
index d9e30516dc72..194c8fc2e55a 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_lib.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_lib.sh
@@ -2,6 +2,7 @@
 # SPDX-License-Identifier: GPL-2.0
 
 . "$(dirname "${0}")/../lib.sh"
+. "$(dirname "${0}")/../net_helper.sh"
 
 readonly KSFT_PASS=0
 readonly KSFT_FAIL=1
@@ -363,20 +364,7 @@ mptcp_lib_check_transfer() {
 
 # $1: ns, $2: port
 mptcp_lib_wait_local_port_listen() {
-	local listener_ns="${1}"
-	local port="${2}"
-
-	local port_hex
-	port_hex="$(printf "%04X" "${port}")"
-
-	local _
-	for _ in $(seq 10); do
-		ip netns exec "${listener_ns}" cat /proc/net/tcp* | \
-			awk "BEGIN {rc=1} {if (\$2 ~ /:${port_hex}\$/ && \$4 ~ /0A/) \
-			     {rc=0; exit}} END {exit rc}" &&
-			break
-		sleep 0.1
-	done
+	wait_local_port_listen "${@}" "tcp"
 }
 
 mptcp_lib_check_output() {
-- 
cgit v1.2.3-58-ga151


From 3966a668bfeef49e265e4975a2cdc55fb931036d Mon Sep 17 00:00:00 2001
From: Dmitry Safonov <0x7f454c46@gmail.com>
Date: Fri, 7 Jun 2024 00:25:55 +0100
Subject: net/tcp: Use static_branch_tcp_{md5,ao} to drop ifdefs

It's possible to clean-up some ifdefs by hiding that
tcp_{md5,ao}_needed static branch is defined and compiled only
under related configs, since commit 4c8530dc7d7d ("net/tcp: Only produce
AO/MD5 logs if there are any keys").

Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Dmitry Safonov <0x7f454c46@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h   | 14 ++++----------
 net/ipv4/tcp_ipv4.c |  8 ++------
 2 files changed, 6 insertions(+), 16 deletions(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index a70fc39090fe..e5427b05129b 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -2386,21 +2386,15 @@ static inline void tcp_get_current_key(const struct sock *sk,
 
 static inline bool tcp_key_is_md5(const struct tcp_key *key)
 {
-#ifdef CONFIG_TCP_MD5SIG
-	if (static_branch_unlikely(&tcp_md5_needed.key) &&
-	    key->type == TCP_KEY_MD5)
-		return true;
-#endif
+	if (static_branch_tcp_md5())
+		return key->type == TCP_KEY_MD5;
 	return false;
 }
 
 static inline bool tcp_key_is_ao(const struct tcp_key *key)
 {
-#ifdef CONFIG_TCP_AO
-	if (static_branch_unlikely(&tcp_ao_needed.key) &&
-	    key->type == TCP_KEY_AO)
-		return true;
-#endif
+	if (static_branch_tcp_ao())
+		return key->type == TCP_KEY_AO;
 	return false;
 }
 
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index e61c7c974745..de0c8f43448a 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1054,12 +1054,10 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb)
 #else
 	if (0) {
 #endif
-#ifdef CONFIG_TCP_MD5SIG
-	} else if (static_branch_unlikely(&tcp_md5_needed.key)) {
+	} else if (static_branch_tcp_md5()) {
 		key.md5_key = tcp_twsk_md5_key(tcptw);
 		if (key.md5_key)
 			key.type = TCP_KEY_MD5;
-#endif
 	}
 
 	tcp_v4_send_ack(sk, skb,
@@ -1128,8 +1126,7 @@ static void tcp_v4_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb,
 #else
 	if (0) {
 #endif
-#ifdef CONFIG_TCP_MD5SIG
-	} else if (static_branch_unlikely(&tcp_md5_needed.key)) {
+	} else if (static_branch_tcp_md5()) {
 		const union tcp_md5_addr *addr;
 		int l3index;
 
@@ -1138,7 +1135,6 @@ static void tcp_v4_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb,
 		key.md5_key = tcp_md5_do_lookup(sk, l3index, addr, AF_INET);
 		if (key.md5_key)
 			key.type = TCP_KEY_MD5;
-#endif
 	}
 
 	tcp_v4_send_ack(sk, skb, seq,
-- 
cgit v1.2.3-58-ga151


From 72863087f635367323693b9ab83c3107e0353c5f Mon Sep 17 00:00:00 2001
From: Dmitry Safonov <0x7f454c46@gmail.com>
Date: Fri, 7 Jun 2024 00:25:56 +0100
Subject: net/tcp: Add a helper tcp_ao_hdr_maclen()

It's going to be used more in TCP-AO tracepoints.

Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Dmitry Safonov <0x7f454c46@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp_ao.h | 5 +++++
 net/ipv4/tcp_ao.c    | 2 +-
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/include/net/tcp_ao.h b/include/net/tcp_ao.h
index 5d8e9ed2c005..198e02004ad2 100644
--- a/include/net/tcp_ao.h
+++ b/include/net/tcp_ao.h
@@ -19,6 +19,11 @@ struct tcp_ao_hdr {
 	u8	rnext_keyid;
 };
 
+static inline u8 tcp_ao_hdr_maclen(const struct tcp_ao_hdr *aoh)
+{
+	return aoh->length - sizeof(struct tcp_ao_hdr);
+}
+
 struct tcp_ao_counters {
 	atomic64_t	pkt_good;
 	atomic64_t	pkt_bad;
diff --git a/net/ipv4/tcp_ao.c b/net/ipv4/tcp_ao.c
index 37c42b63ff99..50ae43c92829 100644
--- a/net/ipv4/tcp_ao.c
+++ b/net/ipv4/tcp_ao.c
@@ -884,8 +884,8 @@ tcp_ao_verify_hash(const struct sock *sk, const struct sk_buff *skb,
 		   const struct tcp_ao_hdr *aoh, struct tcp_ao_key *key,
 		   u8 *traffic_key, u8 *phash, u32 sne, int l3index)
 {
-	u8 maclen = aoh->length - sizeof(struct tcp_ao_hdr);
 	const struct tcphdr *th = tcp_hdr(skb);
+	u8 maclen = tcp_ao_hdr_maclen(aoh);
 	void *hash_buf = NULL;
 
 	if (maclen != tcp_ao_maclen(key)) {
-- 
cgit v1.2.3-58-ga151


From 811efc06e5f30a57030451b2d1998aa81273baf8 Mon Sep 17 00:00:00 2001
From: Dmitry Safonov <0x7f454c46@gmail.com>
Date: Fri, 7 Jun 2024 00:25:57 +0100
Subject: net/tcp: Move tcp_inbound_hash() from headers

Two reasons:
1. It's grown up enough
2. In order to not do header spaghetti by including
   <trace/events/tcp.h>, which is necessary for TCP tracepoints.

While at it, unexport and make static tcp_inbound_ao_hash().

Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Dmitry Safonov <0x7f454c46@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp.h | 78 +++----------------------------------------------------
 net/ipv4/tcp.c    | 74 ++++++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 76 insertions(+), 76 deletions(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index e5427b05129b..2aac11e7e1cc 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1863,12 +1863,6 @@ tcp_md5_do_lookup_any_l3index(const struct sock *sk,
 	return __tcp_md5_do_lookup(sk, 0, addr, family, true);
 }
 
-enum skb_drop_reason
-tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb,
-		     const void *saddr, const void *daddr,
-		     int family, int l3index, const __u8 *hash_location);
-
-
 #define tcp_twsk_md5_key(twsk)	((twsk)->tw_md5_key)
 #else
 static inline struct tcp_md5sig_key *
@@ -1885,13 +1879,6 @@ tcp_md5_do_lookup_any_l3index(const struct sock *sk,
 	return NULL;
 }
 
-static inline enum skb_drop_reason
-tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb,
-		     const void *saddr, const void *daddr,
-		     int family, int l3index, const __u8 *hash_location)
-{
-	return SKB_NOT_DROPPED_YET;
-}
 #define tcp_twsk_md5_key(twsk)	NULL
 #endif
 
@@ -2806,66 +2793,9 @@ static inline bool tcp_ao_required(struct sock *sk, const void *saddr,
 	return false;
 }
 
-/* Called with rcu_read_lock() */
-static inline enum skb_drop_reason
-tcp_inbound_hash(struct sock *sk, const struct request_sock *req,
-		 const struct sk_buff *skb,
-		 const void *saddr, const void *daddr,
-		 int family, int dif, int sdif)
-{
-	const struct tcphdr *th = tcp_hdr(skb);
-	const struct tcp_ao_hdr *aoh;
-	const __u8 *md5_location;
-	int l3index;
-
-	/* Invalid option or two times meet any of auth options */
-	if (tcp_parse_auth_options(th, &md5_location, &aoh)) {
-		tcp_hash_fail("TCP segment has incorrect auth options set",
-			      family, skb, "");
-		return SKB_DROP_REASON_TCP_AUTH_HDR;
-	}
-
-	if (req) {
-		if (tcp_rsk_used_ao(req) != !!aoh) {
-			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOBAD);
-			tcp_hash_fail("TCP connection can't start/end using TCP-AO",
-				      family, skb, "%s",
-				      !aoh ? "missing AO" : "AO signed");
-			return SKB_DROP_REASON_TCP_AOFAILURE;
-		}
-	}
-
-	/* sdif set, means packet ingressed via a device
-	 * in an L3 domain and dif is set to the l3mdev
-	 */
-	l3index = sdif ? dif : 0;
-
-	/* Fast path: unsigned segments */
-	if (likely(!md5_location && !aoh)) {
-		/* Drop if there's TCP-MD5 or TCP-AO key with any rcvid/sndid
-		 * for the remote peer. On TCP-AO established connection
-		 * the last key is impossible to remove, so there's
-		 * always at least one current_key.
-		 */
-		if (tcp_ao_required(sk, saddr, family, l3index, true)) {
-			tcp_hash_fail("AO hash is required, but not found",
-					family, skb, "L3 index %d", l3index);
-			return SKB_DROP_REASON_TCP_AONOTFOUND;
-		}
-		if (unlikely(tcp_md5_do_lookup(sk, l3index, saddr, family))) {
-			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5NOTFOUND);
-			tcp_hash_fail("MD5 Hash not found",
-				      family, skb, "L3 index %d", l3index);
-			return SKB_DROP_REASON_TCP_MD5NOTFOUND;
-		}
-		return SKB_NOT_DROPPED_YET;
-	}
-
-	if (aoh)
-		return tcp_inbound_ao_hash(sk, skb, family, req, l3index, aoh);
-
-	return tcp_inbound_md5_hash(sk, skb, saddr, daddr, family,
-				    l3index, md5_location);
-}
+enum skb_drop_reason tcp_inbound_hash(struct sock *sk,
+		const struct request_sock *req, const struct sk_buff *skb,
+		const void *saddr, const void *daddr,
+		int family, int dif, int sdif);
 
 #endif	/* _TCP_H */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 6553221694ec..17a4a8e4855d 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -4463,7 +4463,7 @@ int tcp_md5_hash_key(struct tcp_sigpool *hp,
 EXPORT_SYMBOL(tcp_md5_hash_key);
 
 /* Called with rcu_read_lock() */
-enum skb_drop_reason
+static enum skb_drop_reason
 tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb,
 		     const void *saddr, const void *daddr,
 		     int family, int l3index, const __u8 *hash_location)
@@ -4517,10 +4517,80 @@ tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb,
 	}
 	return SKB_NOT_DROPPED_YET;
 }
-EXPORT_SYMBOL(tcp_inbound_md5_hash);
+#else
+static inline enum skb_drop_reason
+tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb,
+		     const void *saddr, const void *daddr,
+		     int family, int l3index, const __u8 *hash_location)
+{
+	return SKB_NOT_DROPPED_YET;
+}
 
 #endif
 
+/* Called with rcu_read_lock() */
+enum skb_drop_reason
+tcp_inbound_hash(struct sock *sk, const struct request_sock *req,
+		 const struct sk_buff *skb,
+		 const void *saddr, const void *daddr,
+		 int family, int dif, int sdif)
+{
+	const struct tcphdr *th = tcp_hdr(skb);
+	const struct tcp_ao_hdr *aoh;
+	const __u8 *md5_location;
+	int l3index;
+
+	/* Invalid option or two times meet any of auth options */
+	if (tcp_parse_auth_options(th, &md5_location, &aoh)) {
+		tcp_hash_fail("TCP segment has incorrect auth options set",
+			      family, skb, "");
+		return SKB_DROP_REASON_TCP_AUTH_HDR;
+	}
+
+	if (req) {
+		if (tcp_rsk_used_ao(req) != !!aoh) {
+			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOBAD);
+			tcp_hash_fail("TCP connection can't start/end using TCP-AO",
+				      family, skb, "%s",
+				      !aoh ? "missing AO" : "AO signed");
+			return SKB_DROP_REASON_TCP_AOFAILURE;
+		}
+	}
+
+	/* sdif set, means packet ingressed via a device
+	 * in an L3 domain and dif is set to the l3mdev
+	 */
+	l3index = sdif ? dif : 0;
+
+	/* Fast path: unsigned segments */
+	if (likely(!md5_location && !aoh)) {
+		/* Drop if there's TCP-MD5 or TCP-AO key with any rcvid/sndid
+		 * for the remote peer. On TCP-AO established connection
+		 * the last key is impossible to remove, so there's
+		 * always at least one current_key.
+		 */
+		if (tcp_ao_required(sk, saddr, family, l3index, true)) {
+			tcp_hash_fail("AO hash is required, but not found",
+				      family, skb, "L3 index %d", l3index);
+			return SKB_DROP_REASON_TCP_AONOTFOUND;
+		}
+		if (unlikely(tcp_md5_do_lookup(sk, l3index, saddr, family))) {
+			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5NOTFOUND);
+			tcp_hash_fail("MD5 Hash not found",
+				      family, skb, "L3 index %d", l3index);
+			return SKB_DROP_REASON_TCP_MD5NOTFOUND;
+		}
+		return SKB_NOT_DROPPED_YET;
+	}
+
+	if (aoh)
+		return tcp_inbound_ao_hash(sk, skb, family, req, l3index, aoh);
+
+	return tcp_inbound_md5_hash(sk, skb, saddr, daddr, family,
+				    l3index, md5_location);
+}
+EXPORT_SYMBOL_GPL(tcp_inbound_hash);
+
 void tcp_done(struct sock *sk)
 {
 	struct request_sock *req;
-- 
cgit v1.2.3-58-ga151


From 96be3dcd013df6aa79acf32e739c0a35b89a4f50 Mon Sep 17 00:00:00 2001
From: Dmitry Safonov <0x7f454c46@gmail.com>
Date: Fri, 7 Jun 2024 00:25:58 +0100
Subject: net/tcp: Add tcp-md5 and tcp-ao tracepoints

Instead of forcing userspace to parse dmesg (that's what currently is
happening, at least in codebase of my current company), provide a better
way, that can be enabled/disabled in runtime.

Currently, there are already tcp events, add hashing related ones there,
too. Rasdaemon currently exercises net_dev_xmit_timeout,
devlink_health_report, but it'll be trivial to teach it to deal with
failed hashes. Otherwise, BGP may trace/log them itself. Especially
exciting for possible investigations is key rotation (RNext_key
requests).

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Dmitry Safonov <0x7f454c46@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/tcp.h | 317 +++++++++++++++++++++++++++++++++++++++++++++
 net/ipv4/tcp.c             |  17 +++
 net/ipv4/tcp_ao.c          |  13 ++
 net/ipv4/tcp_input.c       |   8 +-
 net/ipv4/tcp_output.c      |   2 +
 5 files changed, 355 insertions(+), 2 deletions(-)

diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
index 49b5ee091cf6..1c8bd8e186b8 100644
--- a/include/trace/events/tcp.h
+++ b/include/trace/events/tcp.h
@@ -411,6 +411,323 @@ TRACE_EVENT(tcp_cong_state_set,
 		  __entry->cong_state)
 );
 
+DECLARE_EVENT_CLASS(tcp_hash_event,
+
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb),
+
+	TP_ARGS(sk, skb),
+
+	TP_STRUCT__entry(
+		__field(__u64, net_cookie)
+		__field(const void *, skbaddr)
+		__field(const void *, skaddr)
+		__field(int, state)
+
+		/* sockaddr_in6 is always bigger than sockaddr_in */
+		__array(__u8, saddr, sizeof(struct sockaddr_in6))
+		__array(__u8, daddr, sizeof(struct sockaddr_in6))
+		__field(int, l3index)
+
+		__field(__u16, sport)
+		__field(__u16, dport)
+		__field(__u16, family)
+
+		__field(bool, fin)
+		__field(bool, syn)
+		__field(bool, rst)
+		__field(bool, psh)
+		__field(bool, ack)
+	),
+
+	TP_fast_assign(
+		const struct tcphdr *th = (const struct tcphdr *)skb->data;
+
+		__entry->net_cookie = sock_net(sk)->net_cookie;
+		__entry->skbaddr = skb;
+		__entry->skaddr = sk;
+		__entry->state = sk->sk_state;
+
+		memset(__entry->saddr, 0, sizeof(struct sockaddr_in6));
+		memset(__entry->daddr, 0, sizeof(struct sockaddr_in6));
+		TP_STORE_ADDR_PORTS_SKB(skb, th, __entry->saddr, __entry->daddr);
+		__entry->l3index = inet_sdif(skb) ? inet_iif(skb) : 0;
+
+		/* For filtering use */
+		__entry->sport = ntohs(th->source);
+		__entry->dport = ntohs(th->dest);
+		__entry->family = sk->sk_family;
+
+		__entry->fin = th->fin;
+		__entry->syn = th->syn;
+		__entry->rst = th->rst;
+		__entry->psh = th->psh;
+		__entry->ack = th->ack;
+	),
+
+	TP_printk("net=%llu state=%s family=%s src=%pISpc dest=%pISpc L3index=%d [%c%c%c%c%c]",
+		  __entry->net_cookie,
+		  show_tcp_state_name(__entry->state),
+		  show_family_name(__entry->family),
+		  __entry->saddr, __entry->daddr,
+		  __entry->l3index,
+		  __entry->fin ? 'F' : ' ',
+		  __entry->syn ? 'S' : ' ',
+		  __entry->rst ? 'R' : ' ',
+		  __entry->psh ? 'P' : ' ',
+		  __entry->ack ? '.' : ' ')
+);
+
+DEFINE_EVENT(tcp_hash_event, tcp_hash_bad_header,
+
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb),
+	TP_ARGS(sk, skb)
+);
+
+DEFINE_EVENT(tcp_hash_event, tcp_hash_md5_required,
+
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb),
+	TP_ARGS(sk, skb)
+);
+
+DEFINE_EVENT(tcp_hash_event, tcp_hash_md5_unexpected,
+
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb),
+	TP_ARGS(sk, skb)
+);
+
+DEFINE_EVENT(tcp_hash_event, tcp_hash_md5_mismatch,
+
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb),
+	TP_ARGS(sk, skb)
+);
+
+DEFINE_EVENT(tcp_hash_event, tcp_hash_ao_required,
+
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb),
+	TP_ARGS(sk, skb)
+);
+
+DECLARE_EVENT_CLASS(tcp_ao_event,
+
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb,
+		 const __u8 keyid, const __u8 rnext, const __u8 maclen),
+
+	TP_ARGS(sk, skb, keyid, rnext, maclen),
+
+	TP_STRUCT__entry(
+		__field(__u64, net_cookie)
+		__field(const void *, skbaddr)
+		__field(const void *, skaddr)
+		__field(int, state)
+
+		/* sockaddr_in6 is always bigger than sockaddr_in */
+		__array(__u8, saddr, sizeof(struct sockaddr_in6))
+		__array(__u8, daddr, sizeof(struct sockaddr_in6))
+		__field(int, l3index)
+
+		__field(__u16, sport)
+		__field(__u16, dport)
+		__field(__u16, family)
+
+		__field(bool, fin)
+		__field(bool, syn)
+		__field(bool, rst)
+		__field(bool, psh)
+		__field(bool, ack)
+
+		__field(__u8, keyid)
+		__field(__u8, rnext)
+		__field(__u8, maclen)
+	),
+
+	TP_fast_assign(
+		const struct tcphdr *th = (const struct tcphdr *)skb->data;
+
+		__entry->net_cookie = sock_net(sk)->net_cookie;
+		__entry->skbaddr = skb;
+		__entry->skaddr = sk;
+		__entry->state = sk->sk_state;
+
+		memset(__entry->saddr, 0, sizeof(struct sockaddr_in6));
+		memset(__entry->daddr, 0, sizeof(struct sockaddr_in6));
+		TP_STORE_ADDR_PORTS_SKB(skb, th, __entry->saddr, __entry->daddr);
+		__entry->l3index = inet_sdif(skb) ? inet_iif(skb) : 0;
+
+		/* For filtering use */
+		__entry->sport = ntohs(th->source);
+		__entry->dport = ntohs(th->dest);
+		__entry->family = sk->sk_family;
+
+		__entry->fin = th->fin;
+		__entry->syn = th->syn;
+		__entry->rst = th->rst;
+		__entry->psh = th->psh;
+		__entry->ack = th->ack;
+
+		__entry->keyid = keyid;
+		__entry->rnext = rnext;
+		__entry->maclen = maclen;
+	),
+
+	TP_printk("net=%llu state=%s family=%s src=%pISpc dest=%pISpc L3index=%d [%c%c%c%c%c] keyid=%u rnext=%u maclen=%u",
+		  __entry->net_cookie,
+		  show_tcp_state_name(__entry->state),
+		  show_family_name(__entry->family),
+		  __entry->saddr, __entry->daddr,
+		  __entry->l3index,
+		  __entry->fin ? 'F' : ' ',
+		  __entry->syn ? 'S' : ' ',
+		  __entry->rst ? 'R' : ' ',
+		  __entry->psh ? 'P' : ' ',
+		  __entry->ack ? '.' : ' ',
+		  __entry->keyid, __entry->rnext, __entry->maclen)
+);
+
+DEFINE_EVENT(tcp_ao_event, tcp_ao_handshake_failure,
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb,
+		 const __u8 keyid, const __u8 rnext, const __u8 maclen),
+	TP_ARGS(sk, skb, keyid, rnext, maclen)
+);
+
+DEFINE_EVENT(tcp_ao_event, tcp_ao_wrong_maclen,
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb,
+		 const __u8 keyid, const __u8 rnext, const __u8 maclen),
+	TP_ARGS(sk, skb, keyid, rnext, maclen)
+);
+
+DEFINE_EVENT(tcp_ao_event, tcp_ao_mismatch,
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb,
+		 const __u8 keyid, const __u8 rnext, const __u8 maclen),
+	TP_ARGS(sk, skb, keyid, rnext, maclen)
+);
+
+DEFINE_EVENT(tcp_ao_event, tcp_ao_key_not_found,
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb,
+		 const __u8 keyid, const __u8 rnext, const __u8 maclen),
+	TP_ARGS(sk, skb, keyid, rnext, maclen)
+);
+
+DEFINE_EVENT(tcp_ao_event, tcp_ao_rnext_request,
+	TP_PROTO(const struct sock *sk, const struct sk_buff *skb,
+		 const __u8 keyid, const __u8 rnext, const __u8 maclen),
+	TP_ARGS(sk, skb, keyid, rnext, maclen)
+);
+
+DECLARE_EVENT_CLASS(tcp_ao_event_sk,
+
+	TP_PROTO(const struct sock *sk, const __u8 keyid, const __u8 rnext),
+
+	TP_ARGS(sk, keyid, rnext),
+
+	TP_STRUCT__entry(
+		__field(__u64, net_cookie)
+		__field(const void *, skaddr)
+		__field(int, state)
+
+		/* sockaddr_in6 is always bigger than sockaddr_in */
+		__array(__u8, saddr, sizeof(struct sockaddr_in6))
+		__array(__u8, daddr, sizeof(struct sockaddr_in6))
+
+		__field(__u16, sport)
+		__field(__u16, dport)
+		__field(__u16, family)
+
+		__field(__u8, keyid)
+		__field(__u8, rnext)
+	),
+
+	TP_fast_assign(
+		const struct inet_sock *inet = inet_sk(sk);
+
+		__entry->net_cookie = sock_net(sk)->net_cookie;
+		__entry->skaddr = sk;
+		__entry->state = sk->sk_state;
+
+		memset(__entry->saddr, 0, sizeof(struct sockaddr_in6));
+		memset(__entry->daddr, 0, sizeof(struct sockaddr_in6));
+		TP_STORE_ADDR_PORTS(__entry, inet, sk);
+
+		/* For filtering use */
+		__entry->sport = ntohs(inet->inet_sport);
+		__entry->dport = ntohs(inet->inet_dport);
+		__entry->family = sk->sk_family;
+
+		__entry->keyid = keyid;
+		__entry->rnext = rnext;
+	),
+
+	TP_printk("net=%llu state=%s family=%s src=%pISpc dest=%pISpc keyid=%u rnext=%u",
+		  __entry->net_cookie,
+		  show_tcp_state_name(__entry->state),
+		  show_family_name(__entry->family),
+		  __entry->saddr, __entry->daddr,
+		  __entry->keyid, __entry->rnext)
+);
+
+DEFINE_EVENT(tcp_ao_event_sk, tcp_ao_synack_no_key,
+	TP_PROTO(const struct sock *sk, const __u8 keyid, const __u8 rnext),
+	TP_ARGS(sk, keyid, rnext)
+);
+
+DECLARE_EVENT_CLASS(tcp_ao_event_sne,
+
+	TP_PROTO(const struct sock *sk, __u32 new_sne),
+
+	TP_ARGS(sk, new_sne),
+
+	TP_STRUCT__entry(
+		__field(__u64, net_cookie)
+		__field(const void *, skaddr)
+		__field(int, state)
+
+		/* sockaddr_in6 is always bigger than sockaddr_in */
+		__array(__u8, saddr, sizeof(struct sockaddr_in6))
+		__array(__u8, daddr, sizeof(struct sockaddr_in6))
+
+		__field(__u16, sport)
+		__field(__u16, dport)
+		__field(__u16, family)
+
+		__field(__u32, new_sne)
+	),
+
+	TP_fast_assign(
+		const struct inet_sock *inet = inet_sk(sk);
+
+		__entry->net_cookie = sock_net(sk)->net_cookie;
+		__entry->skaddr = sk;
+		__entry->state = sk->sk_state;
+
+		memset(__entry->saddr, 0, sizeof(struct sockaddr_in6));
+		memset(__entry->daddr, 0, sizeof(struct sockaddr_in6));
+		TP_STORE_ADDR_PORTS(__entry, inet, sk);
+
+		/* For filtering use */
+		__entry->sport = ntohs(inet->inet_sport);
+		__entry->dport = ntohs(inet->inet_dport);
+		__entry->family = sk->sk_family;
+
+		__entry->new_sne = new_sne;
+	),
+
+	TP_printk("net=%llu state=%s family=%s src=%pISpc dest=%pISpc sne=%u",
+		  __entry->net_cookie,
+		  show_tcp_state_name(__entry->state),
+		  show_family_name(__entry->family),
+		  __entry->saddr, __entry->daddr,
+		  __entry->new_sne)
+);
+
+DEFINE_EVENT(tcp_ao_event_sne, tcp_ao_snd_sne_update,
+	TP_PROTO(const struct sock *sk, __u32 new_sne),
+	TP_ARGS(sk, new_sne)
+);
+
+DEFINE_EVENT(tcp_ao_event_sne, tcp_ao_rcv_sne_update,
+	TP_PROTO(const struct sock *sk, __u32 new_sne),
+	TP_ARGS(sk, new_sne)
+);
+
 #endif /* _TRACE_TCP_H */
 
 /* This part must be outside protection */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 17a4a8e4855d..73152ce1367e 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -282,6 +282,7 @@
 #include <asm/ioctls.h>
 #include <net/busy_poll.h>
 #include <net/hotdata.h>
+#include <trace/events/tcp.h>
 #include <net/rps.h>
 
 /* Track pending CMSGs. */
@@ -4484,6 +4485,7 @@ tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb,
 	if (!key && hash_location) {
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5UNEXPECTED);
 		tcp_hash_fail("Unexpected MD5 Hash found", family, skb, "");
+		trace_tcp_hash_md5_unexpected(sk, skb);
 		return SKB_DROP_REASON_TCP_MD5UNEXPECTED;
 	}
 
@@ -4513,6 +4515,7 @@ tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb,
 					      l3index);
 			}
 		}
+		trace_tcp_hash_md5_mismatch(sk, skb);
 		return SKB_DROP_REASON_TCP_MD5FAILURE;
 	}
 	return SKB_NOT_DROPPED_YET;
@@ -4544,15 +4547,27 @@ tcp_inbound_hash(struct sock *sk, const struct request_sock *req,
 	if (tcp_parse_auth_options(th, &md5_location, &aoh)) {
 		tcp_hash_fail("TCP segment has incorrect auth options set",
 			      family, skb, "");
+		trace_tcp_hash_bad_header(sk, skb);
 		return SKB_DROP_REASON_TCP_AUTH_HDR;
 	}
 
 	if (req) {
 		if (tcp_rsk_used_ao(req) != !!aoh) {
+			u8 keyid, rnext, maclen;
+
+			if (aoh) {
+				keyid = aoh->keyid;
+				rnext = aoh->rnext_keyid;
+				maclen = tcp_ao_hdr_maclen(aoh);
+			} else {
+				keyid = rnext = maclen = 0;
+			}
+
 			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOBAD);
 			tcp_hash_fail("TCP connection can't start/end using TCP-AO",
 				      family, skb, "%s",
 				      !aoh ? "missing AO" : "AO signed");
+			trace_tcp_ao_handshake_failure(sk, skb, keyid, rnext, maclen);
 			return SKB_DROP_REASON_TCP_AOFAILURE;
 		}
 	}
@@ -4572,12 +4587,14 @@ tcp_inbound_hash(struct sock *sk, const struct request_sock *req,
 		if (tcp_ao_required(sk, saddr, family, l3index, true)) {
 			tcp_hash_fail("AO hash is required, but not found",
 				      family, skb, "L3 index %d", l3index);
+			trace_tcp_hash_ao_required(sk, skb);
 			return SKB_DROP_REASON_TCP_AONOTFOUND;
 		}
 		if (unlikely(tcp_md5_do_lookup(sk, l3index, saddr, family))) {
 			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5NOTFOUND);
 			tcp_hash_fail("MD5 Hash not found",
 				      family, skb, "L3 index %d", l3index);
+			trace_tcp_hash_md5_required(sk, skb);
 			return SKB_DROP_REASON_TCP_MD5NOTFOUND;
 		}
 		return SKB_NOT_DROPPED_YET;
diff --git a/net/ipv4/tcp_ao.c b/net/ipv4/tcp_ao.c
index 50ae43c92829..1e5087c6cd7d 100644
--- a/net/ipv4/tcp_ao.c
+++ b/net/ipv4/tcp_ao.c
@@ -16,6 +16,7 @@
 #include <net/tcp.h>
 #include <net/ipv6.h>
 #include <net/icmp.h>
+#include <trace/events/tcp.h>
 
 DEFINE_STATIC_KEY_DEFERRED_FALSE(tcp_ao_needed, HZ);
 
@@ -895,6 +896,8 @@ tcp_ao_verify_hash(const struct sock *sk, const struct sk_buff *skb,
 		tcp_hash_fail("AO hash wrong length", family, skb,
 			      "%u != %d L3index: %d", maclen,
 			      tcp_ao_maclen(key), l3index);
+		trace_tcp_ao_wrong_maclen(sk, skb, aoh->keyid,
+					  aoh->rnext_keyid, maclen);
 		return SKB_DROP_REASON_TCP_AOFAILURE;
 	}
 
@@ -911,6 +914,8 @@ tcp_ao_verify_hash(const struct sock *sk, const struct sk_buff *skb,
 		atomic64_inc(&key->pkt_bad);
 		tcp_hash_fail("AO hash mismatch", family, skb,
 			      "L3index: %d", l3index);
+		trace_tcp_ao_mismatch(sk, skb, aoh->keyid,
+				      aoh->rnext_keyid, maclen);
 		kfree(hash_buf);
 		return SKB_DROP_REASON_TCP_AOFAILURE;
 	}
@@ -927,6 +932,7 @@ tcp_inbound_ao_hash(struct sock *sk, const struct sk_buff *skb,
 		    int l3index, const struct tcp_ao_hdr *aoh)
 {
 	const struct tcphdr *th = tcp_hdr(skb);
+	u8 maclen = tcp_ao_hdr_maclen(aoh);
 	u8 *phash = (u8 *)(aoh + 1); /* hash goes just after the header */
 	struct tcp_ao_info *info;
 	enum skb_drop_reason ret;
@@ -941,6 +947,8 @@ tcp_inbound_ao_hash(struct sock *sk, const struct sk_buff *skb,
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOKEYNOTFOUND);
 		tcp_hash_fail("AO key not found", family, skb,
 			      "keyid: %u L3index: %d", aoh->keyid, l3index);
+		trace_tcp_ao_key_not_found(sk, skb, aoh->keyid,
+					   aoh->rnext_keyid, maclen);
 		return SKB_DROP_REASON_TCP_AOUNEXPECTED;
 	}
 
@@ -981,6 +989,9 @@ tcp_inbound_ao_hash(struct sock *sk, const struct sk_buff *skb,
 		current_key = READ_ONCE(info->current_key);
 		/* Key rotation: the peer asks us to use new key (RNext) */
 		if (unlikely(aoh->rnext_keyid != current_key->sndid)) {
+			trace_tcp_ao_rnext_request(sk, skb, current_key->sndid,
+						   aoh->rnext_keyid,
+						   tcp_ao_hdr_maclen(aoh));
 			/* If the key is not found we do nothing. */
 			key = tcp_ao_established_key(info, aoh->rnext_keyid, -1);
 			if (key)
@@ -1048,6 +1059,8 @@ key_not_found:
 	atomic64_inc(&info->counters.key_not_found);
 	tcp_hash_fail("Requested by the peer AO key id not found",
 		      family, skb, "L3index: %d", l3index);
+	trace_tcp_ao_key_not_found(sk, skb, aoh->keyid,
+				   aoh->rnext_keyid, maclen);
 	return SKB_DROP_REASON_TCP_AOKEYNOTFOUND;
 }
 
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index eb187450e4d7..d0a1e34d69f6 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3578,8 +3578,10 @@ static void tcp_snd_sne_update(struct tcp_sock *tp, u32 ack)
 
 	ao = rcu_dereference_protected(tp->ao_info,
 				       lockdep_sock_is_held((struct sock *)tp));
-	if (ao && ack < tp->snd_una)
+	if (ao && ack < tp->snd_una) {
 		ao->snd_sne++;
+		trace_tcp_ao_snd_sne_update((struct sock *)tp, ao->snd_sne);
+	}
 #endif
 }
 
@@ -3604,8 +3606,10 @@ static void tcp_rcv_sne_update(struct tcp_sock *tp, u32 seq)
 
 	ao = rcu_dereference_protected(tp->ao_info,
 				       lockdep_sock_is_held((struct sock *)tp));
-	if (ao && seq < tp->rcv_nxt)
+	if (ao && seq < tp->rcv_nxt) {
 		ao->rcv_sne++;
+		trace_tcp_ao_rcv_sne_update((struct sock *)tp, ao->rcv_sne);
+	}
 #endif
 }
 
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 090fb0c24599..16c48df8df4c 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -3768,6 +3768,7 @@ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst,
 #ifdef CONFIG_TCP_AO
 		struct tcp_ao_key *ao_key = NULL;
 		u8 keyid = tcp_rsk(req)->ao_keyid;
+		u8 rnext = tcp_rsk(req)->ao_rcv_next;
 
 		ao_key = tcp_sk(sk)->af_specific->ao_lookup(sk, req_to_sk(req),
 							    keyid, -1);
@@ -3777,6 +3778,7 @@ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst,
 		 * ao_keyid (RFC5925 RNextKeyID), so let's keep it simple here.
 		 */
 		if (unlikely(!ao_key)) {
+			trace_tcp_ao_synack_no_key(sk, keyid, rnext);
 			rcu_read_unlock();
 			kfree_skb(skb);
 			net_warn_ratelimited("TCP-AO: the keyid %u from SYN packet is not present - not sending SYNACK\n",
-- 
cgit v1.2.3-58-ga151


From 78b1b27db91c7a94297a8b6a665fe7e86dfc5750 Mon Sep 17 00:00:00 2001
From: Dmitry Safonov <0x7f454c46@gmail.com>
Date: Fri, 7 Jun 2024 00:25:59 +0100
Subject: net/tcp: Remove tcp_hash_fail()

Now there are tracepoints, that cover all functionality of
tcp_hash_fail(), but also wire up missing places
They are also faster, can be disabled and provide filtering.

This potentially may create a regression if a userspace depends on dmesg
logs. Fingers crossed, let's see if anyone complains in reality.

Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Dmitry Safonov <0x7f454c46@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/tcp_ao.h | 37 -------------------------------------
 net/ipv4/tcp.c       | 25 -------------------------
 net/ipv4/tcp_ao.c    |  9 ---------
 3 files changed, 71 deletions(-)

diff --git a/include/net/tcp_ao.h b/include/net/tcp_ao.h
index 198e02004ad2..1d46460d0fef 100644
--- a/include/net/tcp_ao.h
+++ b/include/net/tcp_ao.h
@@ -149,43 +149,6 @@ extern struct static_key_false_deferred tcp_ao_needed;
 #define static_branch_tcp_ao()	false
 #endif
 
-static inline bool tcp_hash_should_produce_warnings(void)
-{
-	return static_branch_tcp_md5() || static_branch_tcp_ao();
-}
-
-#define tcp_hash_fail(msg, family, skb, fmt, ...)			\
-do {									\
-	const struct tcphdr *th = tcp_hdr(skb);				\
-	char hdr_flags[6];						\
-	char *f = hdr_flags;						\
-									\
-	if (!tcp_hash_should_produce_warnings())			\
-		break;							\
-	if (th->fin)							\
-		*f++ = 'F';						\
-	if (th->syn)							\
-		*f++ = 'S';						\
-	if (th->rst)							\
-		*f++ = 'R';						\
-	if (th->psh)							\
-		*f++ = 'P';						\
-	if (th->ack)							\
-		*f++ = '.';						\
-	*f = 0;								\
-	if ((family) == AF_INET) {					\
-		net_info_ratelimited("%s for %pI4.%d->%pI4.%d [%s] " fmt "\n", \
-				msg, &ip_hdr(skb)->saddr, ntohs(th->source), \
-				&ip_hdr(skb)->daddr, ntohs(th->dest),	\
-				hdr_flags, ##__VA_ARGS__);		\
-	} else {							\
-		net_info_ratelimited("%s for [%pI6c].%d->[%pI6c].%d [%s]" fmt "\n", \
-				msg, &ipv6_hdr(skb)->saddr, ntohs(th->source), \
-				&ipv6_hdr(skb)->daddr, ntohs(th->dest),	\
-				hdr_flags, ##__VA_ARGS__);		\
-	}								\
-} while (0)
-
 #ifdef CONFIG_TCP_AO
 /* TCP-AO structures and functions */
 struct tcp4_ao_context {
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 73152ce1367e..e03a342c9162 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -4484,7 +4484,6 @@ tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb,
 
 	if (!key && hash_location) {
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5UNEXPECTED);
-		tcp_hash_fail("Unexpected MD5 Hash found", family, skb, "");
 		trace_tcp_hash_md5_unexpected(sk, skb);
 		return SKB_DROP_REASON_TCP_MD5UNEXPECTED;
 	}
@@ -4500,21 +4499,6 @@ tcp_inbound_md5_hash(const struct sock *sk, const struct sk_buff *skb,
 							 NULL, skb);
 	if (genhash || memcmp(hash_location, newhash, 16) != 0) {
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5FAILURE);
-		if (family == AF_INET) {
-			tcp_hash_fail("MD5 Hash failed", AF_INET, skb, "%s L3 index %d",
-				      genhash ? "tcp_v4_calc_md5_hash failed"
-				      : "", l3index);
-		} else {
-			if (genhash) {
-				tcp_hash_fail("MD5 Hash failed",
-					      AF_INET6, skb, "L3 index %d",
-					      l3index);
-			} else {
-				tcp_hash_fail("MD5 Hash mismatch",
-					      AF_INET6, skb, "L3 index %d",
-					      l3index);
-			}
-		}
 		trace_tcp_hash_md5_mismatch(sk, skb);
 		return SKB_DROP_REASON_TCP_MD5FAILURE;
 	}
@@ -4545,8 +4529,6 @@ tcp_inbound_hash(struct sock *sk, const struct request_sock *req,
 
 	/* Invalid option or two times meet any of auth options */
 	if (tcp_parse_auth_options(th, &md5_location, &aoh)) {
-		tcp_hash_fail("TCP segment has incorrect auth options set",
-			      family, skb, "");
 		trace_tcp_hash_bad_header(sk, skb);
 		return SKB_DROP_REASON_TCP_AUTH_HDR;
 	}
@@ -4564,9 +4546,6 @@ tcp_inbound_hash(struct sock *sk, const struct request_sock *req,
 			}
 
 			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOBAD);
-			tcp_hash_fail("TCP connection can't start/end using TCP-AO",
-				      family, skb, "%s",
-				      !aoh ? "missing AO" : "AO signed");
 			trace_tcp_ao_handshake_failure(sk, skb, keyid, rnext, maclen);
 			return SKB_DROP_REASON_TCP_AOFAILURE;
 		}
@@ -4585,15 +4564,11 @@ tcp_inbound_hash(struct sock *sk, const struct request_sock *req,
 		 * always at least one current_key.
 		 */
 		if (tcp_ao_required(sk, saddr, family, l3index, true)) {
-			tcp_hash_fail("AO hash is required, but not found",
-				      family, skb, "L3 index %d", l3index);
 			trace_tcp_hash_ao_required(sk, skb);
 			return SKB_DROP_REASON_TCP_AONOTFOUND;
 		}
 		if (unlikely(tcp_md5_do_lookup(sk, l3index, saddr, family))) {
 			NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5NOTFOUND);
-			tcp_hash_fail("MD5 Hash not found",
-				      family, skb, "L3 index %d", l3index);
 			trace_tcp_hash_md5_required(sk, skb);
 			return SKB_DROP_REASON_TCP_MD5NOTFOUND;
 		}
diff --git a/net/ipv4/tcp_ao.c b/net/ipv4/tcp_ao.c
index 1e5087c6cd7d..0de863aa5f66 100644
--- a/net/ipv4/tcp_ao.c
+++ b/net/ipv4/tcp_ao.c
@@ -893,9 +893,6 @@ tcp_ao_verify_hash(const struct sock *sk, const struct sk_buff *skb,
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOBAD);
 		atomic64_inc(&info->counters.pkt_bad);
 		atomic64_inc(&key->pkt_bad);
-		tcp_hash_fail("AO hash wrong length", family, skb,
-			      "%u != %d L3index: %d", maclen,
-			      tcp_ao_maclen(key), l3index);
 		trace_tcp_ao_wrong_maclen(sk, skb, aoh->keyid,
 					  aoh->rnext_keyid, maclen);
 		return SKB_DROP_REASON_TCP_AOFAILURE;
@@ -912,8 +909,6 @@ tcp_ao_verify_hash(const struct sock *sk, const struct sk_buff *skb,
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOBAD);
 		atomic64_inc(&info->counters.pkt_bad);
 		atomic64_inc(&key->pkt_bad);
-		tcp_hash_fail("AO hash mismatch", family, skb,
-			      "L3index: %d", l3index);
 		trace_tcp_ao_mismatch(sk, skb, aoh->keyid,
 				      aoh->rnext_keyid, maclen);
 		kfree(hash_buf);
@@ -945,8 +940,6 @@ tcp_inbound_ao_hash(struct sock *sk, const struct sk_buff *skb,
 	info = rcu_dereference(tcp_sk(sk)->ao_info);
 	if (!info) {
 		NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOKEYNOTFOUND);
-		tcp_hash_fail("AO key not found", family, skb,
-			      "keyid: %u L3index: %d", aoh->keyid, l3index);
 		trace_tcp_ao_key_not_found(sk, skb, aoh->keyid,
 					   aoh->rnext_keyid, maclen);
 		return SKB_DROP_REASON_TCP_AOUNEXPECTED;
@@ -1057,8 +1050,6 @@ verify_hash:
 key_not_found:
 	NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOKEYNOTFOUND);
 	atomic64_inc(&info->counters.key_not_found);
-	tcp_hash_fail("Requested by the peer AO key id not found",
-		      family, skb, "L3index: %d", l3index);
 	trace_tcp_ao_key_not_found(sk, skb, aoh->keyid,
 				   aoh->rnext_keyid, maclen);
 	return SKB_DROP_REASON_TCP_AOKEYNOTFOUND;
-- 
cgit v1.2.3-58-ga151


From efe46fb18e7891a44b05ce62f68eea3d87b9342c Mon Sep 17 00:00:00 2001
From: Dmitry Safonov <0x7f454c46@gmail.com>
Date: Fri, 7 Jun 2024 00:26:00 +0100
Subject: Documentation/tcp-ao: Add a few lines on tracepoints

Signed-off-by: Dmitry Safonov <0x7f454c46@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/tcp_ao.rst | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/Documentation/networking/tcp_ao.rst b/Documentation/networking/tcp_ao.rst
index 8a58321acce7..e96e62d1dab3 100644
--- a/Documentation/networking/tcp_ao.rst
+++ b/Documentation/networking/tcp_ao.rst
@@ -337,6 +337,15 @@ TCP-AO per-socket counters are also duplicated with per-netns counters,
 exposed with SNMP. Those are ``TCPAOGood``, ``TCPAOBad``, ``TCPAOKeyNotFound``,
 ``TCPAORequired`` and ``TCPAODroppedIcmps``.
 
+For monitoring purposes, there are following TCP-AO trace events:
+``tcp_hash_bad_header``, ``tcp_hash_ao_required``, ``tcp_ao_handshake_failure``,
+``tcp_ao_wrong_maclen``, ``tcp_ao_wrong_maclen``, ``tcp_ao_key_not_found``,
+``tcp_ao_rnext_request``, ``tcp_ao_synack_no_key``, ``tcp_ao_snd_sne_update``,
+``tcp_ao_rcv_sne_update``. It's possible to separately enable any of them and
+one can filter them by net-namespace, 4-tuple, family, L3 index, and TCP header
+flags. If a segment has a TCP-AO header, the filters may also include
+keyid, rnext, and maclen. SNE updates include the rolled-over numbers.
+
 RFC 5925 very permissively specifies how TCP port matching can be done for
 MKTs::
 
-- 
cgit v1.2.3-58-ga151


From 5f12dd57a071daf896c00cb4572f2baebbddbb46 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 12 Jun 2024 10:05:33 +0200
Subject: wifi: mac80211: correct EHT EIRP TPE parsing

For the EHT EIRP transmit power envelope, the 320 MHz is in
the last octet, but if we've copied 4 octets (count == 3),
the next one is at index 4 not 5 (count + 2). Fix this, and
just hardcode the offset since count is always 3 here.

Fixes: 39dc8b8ea387 ("wifi: mac80211: pass parsed TPE data to drivers")
Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240612100533.f96c1e0fb758.I2f301c4341abb44dafd29128e7e32c66dc0e296d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/parse.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mac80211/parse.c b/net/mac80211/parse.c
index 28aae14db8a9..279c5143b335 100644
--- a/net/mac80211/parse.c
+++ b/net/mac80211/parse.c
@@ -235,7 +235,7 @@ static void ieee80211_parse_tpe(struct ieee80211_parsed_tpe *tpe,
 		*cnt_out = count + 1;
 		/* separately take 320 MHz if present */
 		if (count == 3 && len > sizeof(*env) + count + 1) {
-			out[4] = env->variable[count + 2];
+			out[4] = env->variable[4];
 			*cnt_out = 5;
 		}
 		break;
-- 
cgit v1.2.3-58-ga151


From 7f12e26a194d0043441f870708093d9c2c3bad7d Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 7 Jun 2024 20:17:17 +0200
Subject: wifi: cfg80211: make hash table duplicates more survivable

Jiazi Li reported that they occasionally see hash table duplicates
as evidenced by the WARN_ON() in rb_insert_bss() in this code.  It
isn't clear how that happens, nor have I been able to reproduce it,
but if it does happen, the kernel crashes later, when it tries to
unhash the entry that's now not hashed.

Try to make this situation more survivable by removing the BSS from
the list(s) as well, that way it's fully leaked here (as had been
the intent in the hash insert error path), and no longer reachable
through the list(s) so it shouldn't be unhashed again later.

Link: https://lore.kernel.org/r/20231026013528.GA24122@Jiazi.Li
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240607181726.36835-2-johannes@sipsolutions.net
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/scan.c | 46 ++++++++++++++++++++++++++++++++++------------
 1 file changed, 34 insertions(+), 12 deletions(-)

diff --git a/net/wireless/scan.c b/net/wireless/scan.c
index 2f2a3163968a..9b31274a1376 100644
--- a/net/wireless/scan.c
+++ b/net/wireless/scan.c
@@ -1604,7 +1604,7 @@ struct cfg80211_bss *__cfg80211_get_bss(struct wiphy *wiphy,
 }
 EXPORT_SYMBOL(__cfg80211_get_bss);
 
-static void rb_insert_bss(struct cfg80211_registered_device *rdev,
+static bool rb_insert_bss(struct cfg80211_registered_device *rdev,
 			  struct cfg80211_internal_bss *bss)
 {
 	struct rb_node **p = &rdev->bss_tree.rb_node;
@@ -1620,7 +1620,7 @@ static void rb_insert_bss(struct cfg80211_registered_device *rdev,
 
 		if (WARN_ON(!cmp)) {
 			/* will sort of leak this BSS */
-			return;
+			return false;
 		}
 
 		if (cmp < 0)
@@ -1631,6 +1631,7 @@ static void rb_insert_bss(struct cfg80211_registered_device *rdev,
 
 	rb_link_node(&bss->rbn, parent, p);
 	rb_insert_color(&bss->rbn, &rdev->bss_tree);
+	return true;
 }
 
 static struct cfg80211_internal_bss *
@@ -1657,6 +1658,34 @@ rb_find_bss(struct cfg80211_registered_device *rdev,
 	return NULL;
 }
 
+static void cfg80211_insert_bss(struct cfg80211_registered_device *rdev,
+				struct cfg80211_internal_bss *bss)
+{
+	lockdep_assert_held(&rdev->bss_lock);
+
+	if (!rb_insert_bss(rdev, bss))
+		return;
+	list_add_tail(&bss->list, &rdev->bss_list);
+	rdev->bss_entries++;
+}
+
+static void cfg80211_rehash_bss(struct cfg80211_registered_device *rdev,
+                                struct cfg80211_internal_bss *bss)
+{
+	lockdep_assert_held(&rdev->bss_lock);
+
+	rb_erase(&bss->rbn, &rdev->bss_tree);
+	if (!rb_insert_bss(rdev, bss)) {
+		list_del(&bss->list);
+		if (!list_empty(&bss->hidden_list))
+			list_del_init(&bss->hidden_list);
+		if (!list_empty(&bss->pub.nontrans_list))
+			list_del_init(&bss->pub.nontrans_list);
+		rdev->bss_entries--;
+	}
+	rdev->bss_generation++;
+}
+
 static bool cfg80211_combine_bsses(struct cfg80211_registered_device *rdev,
 				   struct cfg80211_internal_bss *new)
 {
@@ -1969,9 +1998,7 @@ __cfg80211_bss_update(struct cfg80211_registered_device *rdev,
 			bss_ref_get(rdev, bss_from_pub(tmp->pub.transmitted_bss));
 		}
 
-		list_add_tail(&new->list, &rdev->bss_list);
-		rdev->bss_entries++;
-		rb_insert_bss(rdev, new);
+		cfg80211_insert_bss(rdev, new);
 		found = new;
 	}
 
@@ -3349,19 +3376,14 @@ void cfg80211_update_assoc_bss_entry(struct wireless_dev *wdev,
 		if (!WARN_ON(!__cfg80211_unlink_bss(rdev, new)))
 			rdev->bss_generation++;
 	}
-
-	rb_erase(&cbss->rbn, &rdev->bss_tree);
-	rb_insert_bss(rdev, cbss);
-	rdev->bss_generation++;
+	cfg80211_rehash_bss(rdev, cbss);
 
 	list_for_each_entry_safe(nontrans_bss, tmp,
 				 &cbss->pub.nontrans_list,
 				 nontrans_list) {
 		bss = bss_from_pub(nontrans_bss);
 		bss->pub.channel = chan;
-		rb_erase(&bss->rbn, &rdev->bss_tree);
-		rb_insert_bss(rdev, bss);
-		rdev->bss_generation++;
+		cfg80211_rehash_bss(rdev, bss);
 	}
 
 done:
-- 
cgit v1.2.3-58-ga151


From 7b7890f3861634d1ec762a2697072dff42f65319 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Fri, 7 Jun 2024 03:20:43 -0700
Subject: wifi: mac80211: Move stats allocation to core

With commit 34d21de99cea9 ("net: Move {l,t,d}stats allocation to core and
convert veth & vrf"), stats allocation could be done on net core instead
of this driver.

With this new approach, the driver doesn't have to bother with error
handling (allocation failure checking, making sure free happens in the
right spot, etc). This is core responsibility now.

Move mac80211 driver to leverage the core allocation.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://msgid.link/20240607102045.235071-1-leitao@debian.org
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/iface.c | 13 +------------
 1 file changed, 1 insertion(+), 12 deletions(-)

diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index d1a49ee4a194..ac4d4df65298 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -1458,11 +1458,6 @@ int ieee80211_do_open(struct wireless_dev *wdev, bool coming_up)
 	return res;
 }
 
-static void ieee80211_if_free(struct net_device *dev)
-{
-	free_percpu(dev->tstats);
-}
-
 static void ieee80211_if_setup(struct net_device *dev)
 {
 	ether_setup(dev);
@@ -1470,7 +1465,6 @@ static void ieee80211_if_setup(struct net_device *dev)
 	dev->priv_flags |= IFF_NO_QUEUE;
 	dev->netdev_ops = &ieee80211_dataif_ops;
 	dev->needs_free_netdev = true;
-	dev->priv_destructor = ieee80211_if_free;
 }
 
 static void ieee80211_iface_process_skb(struct ieee80211_local *local,
@@ -2103,11 +2097,7 @@ int ieee80211_if_add(struct ieee80211_local *local, const char *name,
 
 		dev_net_set(ndev, wiphy_net(local->hw.wiphy));
 
-		ndev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
-		if (!ndev->tstats) {
-			free_netdev(ndev);
-			return -ENOMEM;
-		}
+		ndev->pcpu_stat_type = NETDEV_PCPU_STAT_TSTATS;
 
 		ndev->needed_headroom = local->tx_headroom +
 					4*6 /* four MAC addresses */
@@ -2120,7 +2110,6 @@ int ieee80211_if_add(struct ieee80211_local *local, const char *name,
 
 		ret = dev_alloc_name(ndev, ndev->name);
 		if (ret < 0) {
-			ieee80211_if_free(ndev);
 			free_netdev(ndev);
 			return ret;
 		}
-- 
cgit v1.2.3-58-ga151


From c018411d355518a0b2a304d7543564cdd1b808b6 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Fri, 7 Jun 2024 03:20:44 -0700
Subject: wifi: mac80211: Remove generic .ndo_get_stats64

Commit 3e2f544dd8a33 ("net: get stats64 if device if driver is
configured") moved the callback to dev_get_tstats64() to net core, so,
unless the driver is doing some custom stats collection, it does not
need to set .ndo_get_stats64.

Since this driver is now relying in NETDEV_PCPU_STAT_TSTATS, then, it
doesn't need to set the dev_get_tstats64() generic .ndo_get_stats64
function pointer.

In this driver specifically, .ndo_get_stats64 basically points to
dev_fetch_sw_netstats(). Now it will point to dev_get_tstats64(), which
calls netdev_stats_to_stats64() and dev_fetch_sw_netstats().
netdev_stats_to_stats64() seems irrelevant for this driver.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://msgid.link/20240607102045.235071-2-leitao@debian.org
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/iface.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index ac4d4df65298..6b7580c61e0c 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -815,12 +815,6 @@ static void ieee80211_uninit(struct net_device *dev)
 	ieee80211_teardown_sdata(IEEE80211_DEV_TO_SUB_IF(dev));
 }
 
-static void
-ieee80211_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
-{
-	dev_fetch_sw_netstats(stats, dev->tstats);
-}
-
 static int ieee80211_netdev_setup_tc(struct net_device *dev,
 				     enum tc_setup_type type, void *type_data)
 {
@@ -837,7 +831,6 @@ static const struct net_device_ops ieee80211_dataif_ops = {
 	.ndo_start_xmit		= ieee80211_subif_start_xmit,
 	.ndo_set_rx_mode	= ieee80211_set_multicast_list,
 	.ndo_set_mac_address 	= ieee80211_change_mac,
-	.ndo_get_stats64	= ieee80211_get_stats64,
 	.ndo_setup_tc		= ieee80211_netdev_setup_tc,
 };
 
@@ -877,7 +870,6 @@ static const struct net_device_ops ieee80211_monitorif_ops = {
 	.ndo_set_rx_mode	= ieee80211_set_multicast_list,
 	.ndo_set_mac_address 	= ieee80211_change_mac,
 	.ndo_select_queue	= ieee80211_monitor_select_queue,
-	.ndo_get_stats64	= ieee80211_get_stats64,
 };
 
 static int ieee80211_netdev_fill_forward_path(struct net_device_path_ctx *ctx,
@@ -945,7 +937,6 @@ static const struct net_device_ops ieee80211_dataif_8023_ops = {
 	.ndo_start_xmit		= ieee80211_subif_start_xmit_8023,
 	.ndo_set_rx_mode	= ieee80211_set_multicast_list,
 	.ndo_set_mac_address	= ieee80211_change_mac,
-	.ndo_get_stats64	= ieee80211_get_stats64,
 	.ndo_fill_forward_path	= ieee80211_netdev_fill_forward_path,
 	.ndo_setup_tc		= ieee80211_netdev_setup_tc,
 };
-- 
cgit v1.2.3-58-ga151


From 9ee0d44f055276fe2802b2f65058e920853f4f99 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Thu, 6 Jun 2024 10:06:52 +0800
Subject: wifi: cfg80211: fix typo in cfg80211_calculate_bitrate_he()

rates_996 is mistakenly written as rates_969, fix it.

Fixes: c4cbaf7973a7 ("cfg80211: Add support for HE")
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Link: https://msgid.link/20240606020653.33205-2-quic_bqiang@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/util.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/wireless/util.c b/net/wireless/util.c
index 082c6f9c5416..d262d37c1519 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -1504,7 +1504,7 @@ static u32 cfg80211_calculate_bitrate_he(struct rate_info *rate)
 		  5120, /*  0.833333... */
 	};
 	u32 rates_160M[3] = { 960777777, 907400000, 816666666 };
-	u32 rates_969[3] =  { 480388888, 453700000, 408333333 };
+	u32 rates_996[3] =  { 480388888, 453700000, 408333333 };
 	u32 rates_484[3] =  { 229411111, 216666666, 195000000 };
 	u32 rates_242[3] =  { 114711111, 108333333,  97500000 };
 	u32 rates_106[3] =  {  40000000,  37777777,  34000000 };
@@ -1529,7 +1529,7 @@ static u32 cfg80211_calculate_bitrate_he(struct rate_info *rate)
 	else if (rate->bw == RATE_INFO_BW_80 ||
 		 (rate->bw == RATE_INFO_BW_HE_RU &&
 		  rate->he_ru_alloc == NL80211_RATE_INFO_HE_RU_ALLOC_996))
-		result = rates_969[rate->he_gi];
+		result = rates_996[rate->he_gi];
 	else if (rate->bw == RATE_INFO_BW_40 ||
 		 (rate->bw == RATE_INFO_BW_HE_RU &&
 		  rate->he_ru_alloc == NL80211_RATE_INFO_HE_RU_ALLOC_484))
-- 
cgit v1.2.3-58-ga151


From bcbd771cd5d68c0c52567556097d75f9fc4e7cd6 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Thu, 6 Jun 2024 10:06:53 +0800
Subject: wifi: cfg80211: handle 2x996 RU allocation in
 cfg80211_calculate_bitrate_he()

Currently NL80211_RATE_INFO_HE_RU_ALLOC_2x996 is not handled in
cfg80211_calculate_bitrate_he(), leading to below warning:

kernel: invalid HE MCS: bw:6, ru:6
kernel: WARNING: CPU: 0 PID: 2312 at net/wireless/util.c:1501 cfg80211_calculate_bitrate_he+0x22b/0x270 [cfg80211]

Fix it by handling 2x996 RU allocation in the same way as 160 MHz bandwidth.

Fixes: c4cbaf7973a7 ("cfg80211: Add support for HE")
Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Link: https://msgid.link/20240606020653.33205-3-quic_bqiang@quicinc.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/util.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/net/wireless/util.c b/net/wireless/util.c
index d262d37c1519..af6ec719567f 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -1524,7 +1524,9 @@ static u32 cfg80211_calculate_bitrate_he(struct rate_info *rate)
 	if (WARN_ON_ONCE(rate->nss < 1 || rate->nss > 8))
 		return 0;
 
-	if (rate->bw == RATE_INFO_BW_160)
+	if (rate->bw == RATE_INFO_BW_160 ||
+	    (rate->bw == RATE_INFO_BW_HE_RU &&
+	     rate->he_ru_alloc == NL80211_RATE_INFO_HE_RU_ALLOC_2x996))
 		result = rates_160M[rate->he_gi];
 	else if (rate->bw == RATE_INFO_BW_80 ||
 		 (rate->bw == RATE_INFO_BW_HE_RU &&
-- 
cgit v1.2.3-58-ga151


From f3269b7912f75b6e34926334377ede45656a8950 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 23 May 2024 12:09:42 +0200
Subject: wifi: nl80211: expose can-monitor channel property

It may be possible to monitor on disabled channels per the
can-monitor flag, but evidently I forgot to expose that out
to userspace. Fix that.

Fixes: a110a3b79177 ("wifi: cfg80211: optionally support monitor on disabled channels")
Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240523120945.9a2c19a51e53.I50fa1b1a18b70f63a5095131ac23dc2e71f3d426@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/nl80211.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 8ff5f79d446a..fd6e33a1eb2a 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -1204,6 +1204,9 @@ static int nl80211_msg_put_channel(struct sk_buff *msg, struct wiphy *wiphy,
 		if ((chan->flags & IEEE80211_CHAN_NO_6GHZ_AFC_CLIENT) &&
 		    nla_put_flag(msg, NL80211_FREQUENCY_ATTR_NO_6GHZ_AFC_CLIENT))
 			goto nla_put_failure;
+		if ((chan->flags & IEEE80211_CHAN_CAN_MONITOR) &&
+		    nla_put_flag(msg, NL80211_FREQUENCY_ATTR_CAN_MONITOR))
+			goto nla_put_failure;
 	}
 
 	if (nla_put_u32(msg, NL80211_FREQUENCY_ATTR_MAX_TX_POWER,
-- 
cgit v1.2.3-58-ga151


From 8682ad36870790ad6a9a03ac237c8d87d34b26d5 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 23 May 2024 12:09:43 +0200
Subject: wifi: cfg80211: use BIT() for flag enums

Use BIT(x) instead of 1<<x, in part because it's mostly
missing spaces anyway, in part because it reads nicer.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240523120945.c21598fbf49c.Ib8f26c5e9f508aee19fdfa1fd4b5995f084c46d4@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h | 102 ++++++++++++++++++++++++-------------------------
 1 file changed, 51 insertions(+), 51 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 5da9bb0ac6a4..051d4eb227bf 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -127,31 +127,31 @@ struct wiphy;
  *	even if it is otherwise disabled.
  */
 enum ieee80211_channel_flags {
-	IEEE80211_CHAN_DISABLED		= 1<<0,
-	IEEE80211_CHAN_NO_IR		= 1<<1,
-	IEEE80211_CHAN_PSD		= 1<<2,
-	IEEE80211_CHAN_RADAR		= 1<<3,
-	IEEE80211_CHAN_NO_HT40PLUS	= 1<<4,
-	IEEE80211_CHAN_NO_HT40MINUS	= 1<<5,
-	IEEE80211_CHAN_NO_OFDM		= 1<<6,
-	IEEE80211_CHAN_NO_80MHZ		= 1<<7,
-	IEEE80211_CHAN_NO_160MHZ	= 1<<8,
-	IEEE80211_CHAN_INDOOR_ONLY	= 1<<9,
-	IEEE80211_CHAN_IR_CONCURRENT	= 1<<10,
-	IEEE80211_CHAN_NO_20MHZ		= 1<<11,
-	IEEE80211_CHAN_NO_10MHZ		= 1<<12,
-	IEEE80211_CHAN_NO_HE		= 1<<13,
-	IEEE80211_CHAN_1MHZ		= 1<<14,
-	IEEE80211_CHAN_2MHZ		= 1<<15,
-	IEEE80211_CHAN_4MHZ		= 1<<16,
-	IEEE80211_CHAN_8MHZ		= 1<<17,
-	IEEE80211_CHAN_16MHZ		= 1<<18,
-	IEEE80211_CHAN_NO_320MHZ	= 1<<19,
-	IEEE80211_CHAN_NO_EHT		= 1<<20,
-	IEEE80211_CHAN_DFS_CONCURRENT	= 1<<21,
-	IEEE80211_CHAN_NO_6GHZ_VLP_CLIENT = 1<<22,
-	IEEE80211_CHAN_NO_6GHZ_AFC_CLIENT = 1<<23,
-	IEEE80211_CHAN_CAN_MONITOR	= 1<<24,
+	IEEE80211_CHAN_DISABLED			= BIT(0),
+	IEEE80211_CHAN_NO_IR			= BIT(1),
+	IEEE80211_CHAN_PSD			= BIT(2),
+	IEEE80211_CHAN_RADAR			= BIT(3),
+	IEEE80211_CHAN_NO_HT40PLUS		= BIT(4),
+	IEEE80211_CHAN_NO_HT40MINUS		= BIT(5),
+	IEEE80211_CHAN_NO_OFDM			= BIT(6),
+	IEEE80211_CHAN_NO_80MHZ			= BIT(7),
+	IEEE80211_CHAN_NO_160MHZ		= BIT(8),
+	IEEE80211_CHAN_INDOOR_ONLY		= BIT(9),
+	IEEE80211_CHAN_IR_CONCURRENT		= BIT(10),
+	IEEE80211_CHAN_NO_20MHZ			= BIT(11),
+	IEEE80211_CHAN_NO_10MHZ			= BIT(12),
+	IEEE80211_CHAN_NO_HE			= BIT(13),
+	IEEE80211_CHAN_1MHZ			= BIT(14),
+	IEEE80211_CHAN_2MHZ			= BIT(15),
+	IEEE80211_CHAN_4MHZ			= BIT(16),
+	IEEE80211_CHAN_8MHZ			= BIT(17),
+	IEEE80211_CHAN_16MHZ			= BIT(18),
+	IEEE80211_CHAN_NO_320MHZ		= BIT(19),
+	IEEE80211_CHAN_NO_EHT			= BIT(20),
+	IEEE80211_CHAN_DFS_CONCURRENT		= BIT(21),
+	IEEE80211_CHAN_NO_6GHZ_VLP_CLIENT	= BIT(22),
+	IEEE80211_CHAN_NO_6GHZ_AFC_CLIENT	= BIT(23),
+	IEEE80211_CHAN_CAN_MONITOR		= BIT(24),
 };
 
 #define IEEE80211_CHAN_NO_HT40 \
@@ -229,13 +229,13 @@ struct ieee80211_channel {
  * @IEEE80211_RATE_SUPPORTS_10MHZ: Rate can be used in 10 MHz mode
  */
 enum ieee80211_rate_flags {
-	IEEE80211_RATE_SHORT_PREAMBLE	= 1<<0,
-	IEEE80211_RATE_MANDATORY_A	= 1<<1,
-	IEEE80211_RATE_MANDATORY_B	= 1<<2,
-	IEEE80211_RATE_MANDATORY_G	= 1<<3,
-	IEEE80211_RATE_ERP_G		= 1<<4,
-	IEEE80211_RATE_SUPPORTS_5MHZ	= 1<<5,
-	IEEE80211_RATE_SUPPORTS_10MHZ	= 1<<6,
+	IEEE80211_RATE_SHORT_PREAMBLE	= BIT(0),
+	IEEE80211_RATE_MANDATORY_A	= BIT(1),
+	IEEE80211_RATE_MANDATORY_B	= BIT(2),
+	IEEE80211_RATE_MANDATORY_G	= BIT(3),
+	IEEE80211_RATE_ERP_G		= BIT(4),
+	IEEE80211_RATE_SUPPORTS_5MHZ	= BIT(5),
+	IEEE80211_RATE_SUPPORTS_10MHZ	= BIT(6),
 };
 
 /**
@@ -1957,9 +1957,9 @@ struct rate_info {
  * @BSS_PARAM_FLAGS_SHORT_SLOT_TIME: whether short slot time is enabled
  */
 enum bss_param_flags {
-	BSS_PARAM_FLAGS_CTS_PROT	= 1<<0,
-	BSS_PARAM_FLAGS_SHORT_PREAMBLE	= 1<<1,
-	BSS_PARAM_FLAGS_SHORT_SLOT_TIME	= 1<<2,
+	BSS_PARAM_FLAGS_CTS_PROT	= BIT(0),
+	BSS_PARAM_FLAGS_SHORT_PREAMBLE	= BIT(1),
+	BSS_PARAM_FLAGS_SHORT_SLOT_TIME	= BIT(2),
 };
 
 /**
@@ -2266,13 +2266,13 @@ static inline int cfg80211_get_station(struct net_device *dev,
  * @MONITOR_FLAG_ACTIVE: active monitor, ACKs frames on its MAC address
  */
 enum monitor_flags {
-	MONITOR_FLAG_CHANGED		= 1<<__NL80211_MNTR_FLAG_INVALID,
-	MONITOR_FLAG_FCSFAIL		= 1<<NL80211_MNTR_FLAG_FCSFAIL,
-	MONITOR_FLAG_PLCPFAIL		= 1<<NL80211_MNTR_FLAG_PLCPFAIL,
-	MONITOR_FLAG_CONTROL		= 1<<NL80211_MNTR_FLAG_CONTROL,
-	MONITOR_FLAG_OTHER_BSS		= 1<<NL80211_MNTR_FLAG_OTHER_BSS,
-	MONITOR_FLAG_COOK_FRAMES	= 1<<NL80211_MNTR_FLAG_COOK_FRAMES,
-	MONITOR_FLAG_ACTIVE		= 1<<NL80211_MNTR_FLAG_ACTIVE,
+	MONITOR_FLAG_CHANGED		= BIT(__NL80211_MNTR_FLAG_INVALID),
+	MONITOR_FLAG_FCSFAIL		= BIT(NL80211_MNTR_FLAG_FCSFAIL),
+	MONITOR_FLAG_PLCPFAIL		= BIT(NL80211_MNTR_FLAG_PLCPFAIL),
+	MONITOR_FLAG_CONTROL		= BIT(NL80211_MNTR_FLAG_CONTROL),
+	MONITOR_FLAG_OTHER_BSS		= BIT(NL80211_MNTR_FLAG_OTHER_BSS),
+	MONITOR_FLAG_COOK_FRAMES	= BIT(NL80211_MNTR_FLAG_COOK_FRAMES),
+	MONITOR_FLAG_ACTIVE		= BIT(NL80211_MNTR_FLAG_ACTIVE),
 };
 
 /**
@@ -3399,15 +3399,15 @@ enum cfg80211_connect_params_changed {
  * @WIPHY_PARAM_TXQ_QUANTUM: TXQ scheduler quantum
  */
 enum wiphy_params_flags {
-	WIPHY_PARAM_RETRY_SHORT		= 1 << 0,
-	WIPHY_PARAM_RETRY_LONG		= 1 << 1,
-	WIPHY_PARAM_FRAG_THRESHOLD	= 1 << 2,
-	WIPHY_PARAM_RTS_THRESHOLD	= 1 << 3,
-	WIPHY_PARAM_COVERAGE_CLASS	= 1 << 4,
-	WIPHY_PARAM_DYN_ACK		= 1 << 5,
-	WIPHY_PARAM_TXQ_LIMIT		= 1 << 6,
-	WIPHY_PARAM_TXQ_MEMORY_LIMIT	= 1 << 7,
-	WIPHY_PARAM_TXQ_QUANTUM		= 1 << 8,
+	WIPHY_PARAM_RETRY_SHORT		= BIT(0),
+	WIPHY_PARAM_RETRY_LONG		= BIT(1),
+	WIPHY_PARAM_FRAG_THRESHOLD	= BIT(2),
+	WIPHY_PARAM_RTS_THRESHOLD	= BIT(3),
+	WIPHY_PARAM_COVERAGE_CLASS	= BIT(4),
+	WIPHY_PARAM_DYN_ACK		= BIT(5),
+	WIPHY_PARAM_TXQ_LIMIT		= BIT(6),
+	WIPHY_PARAM_TXQ_MEMORY_LIMIT	= BIT(7),
+	WIPHY_PARAM_TXQ_QUANTUM		= BIT(8),
 };
 
 #define IEEE80211_DEFAULT_AIRTIME_WEIGHT	256
-- 
cgit v1.2.3-58-ga151


From 92a5df6687da755edd4bfe1acb714f00f56d2e06 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 23 May 2024 12:09:44 +0200
Subject: wifi: ieee80211: remove unused enum ieee80211_client_reg_power

This has never been used, and it's really not directly
representing the spec, so shouldn't have been here in
the first place. Remove it.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240523120945.32ed8fc1522d.Id4480d162e1921478e33d145890dc16c263b57bf@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/linux/ieee80211.h | 18 ------------------
 1 file changed, 18 deletions(-)

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 769008a51809..456a55bd6c6b 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -2426,24 +2426,6 @@ enum ieee80211_ap_reg_power {
 		IEEE80211_REG_AP_POWER_AFTER_LAST - 1,
 };
 
-/**
- * enum ieee80211_client_reg_power - regulatory power for a client
- *
- * @IEEE80211_REG_UNSET_CLIENT: Client has no regulatory power mode
- * @IEEE80211_REG_DEFAULT_CLIENT: Default Client
- * @IEEE80211_REG_SUBORDINATE_CLIENT: Subordinate Client
- * @IEEE80211_REG_CLIENT_POWER_AFTER_LAST: internal
- * @IEEE80211_REG_CLIENT_POWER_MAX: maximum value
- */
-enum ieee80211_client_reg_power {
-	IEEE80211_REG_UNSET_CLIENT,
-	IEEE80211_REG_DEFAULT_CLIENT,
-	IEEE80211_REG_SUBORDINATE_CLIENT,
-	IEEE80211_REG_CLIENT_POWER_AFTER_LAST,
-	IEEE80211_REG_CLIENT_POWER_MAX =
-		IEEE80211_REG_CLIENT_POWER_AFTER_LAST - 1,
-};
-
 /* 802.11ax HE MAC capabilities */
 #define IEEE80211_HE_MAC_CAP0_HTC_HE				0x01
 #define IEEE80211_HE_MAC_CAP0_TWT_REQ				0x02
-- 
cgit v1.2.3-58-ga151


From 0a9314ad5f45aeb10326dce33c5d70b85f74ef2d Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 23 May 2024 12:09:45 +0200
Subject: wifi: cfg80211: move enum ieee80211_ap_reg_power to cfg80211

This really shouldn't have been in ieee80211.h, since it
doesn't directly represent the spec. Move it to cfg80211
rather than mac80211 since upcoming changes will use it
there.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240523120945.962b16c831cd.I5745962525b1b176c5b90d37b3720fc100eee406@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/linux/ieee80211.h | 20 --------------------
 include/net/cfg80211.h    | 15 +++++++++++++++
 2 files changed, 15 insertions(+), 20 deletions(-)

diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h
index 456a55bd6c6b..30cef3b940eb 100644
--- a/include/linux/ieee80211.h
+++ b/include/linux/ieee80211.h
@@ -2406,26 +2406,6 @@ int ieee80211_get_vht_max_nss(struct ieee80211_vht_cap *cap,
 			      int mcs, bool ext_nss_bw_capable,
 			      unsigned int max_vht_nss);
 
-/**
- * enum ieee80211_ap_reg_power - regulatory power for a Access Point
- *
- * @IEEE80211_REG_UNSET_AP: Access Point has no regulatory power mode
- * @IEEE80211_REG_LPI_AP: Indoor Access Point
- * @IEEE80211_REG_SP_AP: Standard power Access Point
- * @IEEE80211_REG_VLP_AP: Very low power Access Point
- * @IEEE80211_REG_AP_POWER_AFTER_LAST: internal
- * @IEEE80211_REG_AP_POWER_MAX: maximum value
- */
-enum ieee80211_ap_reg_power {
-	IEEE80211_REG_UNSET_AP,
-	IEEE80211_REG_LPI_AP,
-	IEEE80211_REG_SP_AP,
-	IEEE80211_REG_VLP_AP,
-	IEEE80211_REG_AP_POWER_AFTER_LAST,
-	IEEE80211_REG_AP_POWER_MAX =
-		IEEE80211_REG_AP_POWER_AFTER_LAST - 1,
-};
-
 /* 802.11ax HE MAC capabilities */
 #define IEEE80211_HE_MAC_CAP0_HTC_HE				0x01
 #define IEEE80211_HE_MAC_CAP0_TWT_REQ				0x02
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 051d4eb227bf..fb7d76513e1c 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -6081,6 +6081,21 @@ void wiphy_delayed_work_cancel(struct wiphy *wiphy,
 void wiphy_delayed_work_flush(struct wiphy *wiphy,
 			      struct wiphy_delayed_work *dwork);
 
+/**
+ * enum ieee80211_ap_reg_power - regulatory power for an Access Point
+ *
+ * @IEEE80211_REG_UNSET_AP: Access Point has no regulatory power mode
+ * @IEEE80211_REG_LPI_AP: Indoor Access Point
+ * @IEEE80211_REG_SP_AP: Standard power Access Point
+ * @IEEE80211_REG_VLP_AP: Very low power Access Point
+ */
+enum ieee80211_ap_reg_power {
+	IEEE80211_REG_UNSET_AP,
+	IEEE80211_REG_LPI_AP,
+	IEEE80211_REG_SP_AP,
+	IEEE80211_REG_VLP_AP,
+};
+
 /**
  * struct wireless_dev - wireless device state
  *
-- 
cgit v1.2.3-58-ga151


From ca41bfe367d54a7fa966c11852d50d4191187725 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 23 May 2024 12:09:46 +0200
Subject: wifi: mac80211: refactor channel checks

To later introduce an override for VLP APs being allowed despite
NO-IR flags, which is somewhat similar in construction to being
allowed to monitor on disabled channels, refactor the code that
checks channel flags to have not a 'monitor' argument but a set
of 'permitting' flags that permit the operation without checking
for 'prohibited' flags.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240523120945.3da28ded4a50.I90cffc633d0510293d511f60097dc75e719b55f0@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/chan.c    | 19 +++++++++++--------
 net/wireless/core.h    |  3 ++-
 net/wireless/nl80211.c |  2 +-
 3 files changed, 14 insertions(+), 10 deletions(-)

diff --git a/net/wireless/chan.c b/net/wireless/chan.c
index 3414b2c3abcc..360480604515 100644
--- a/net/wireless/chan.c
+++ b/net/wireless/chan.c
@@ -1145,7 +1145,8 @@ EXPORT_SYMBOL(cfg80211_chandef_dfs_cac_time);
 
 static bool cfg80211_secondary_chans_ok(struct wiphy *wiphy,
 					u32 center_freq, u32 bandwidth,
-					u32 prohibited_flags, bool monitor)
+					u32 prohibited_flags,
+					u32 permitting_flags)
 {
 	struct ieee80211_channel *c;
 	u32 freq, start_freq, end_freq;
@@ -1157,7 +1158,7 @@ static bool cfg80211_secondary_chans_ok(struct wiphy *wiphy,
 		c = ieee80211_get_channel_khz(wiphy, freq);
 		if (!c)
 			return false;
-		if (monitor && c->flags & IEEE80211_CHAN_CAN_MONITOR)
+		if (c->flags & permitting_flags)
 			continue;
 		if (c->flags & prohibited_flags)
 			return false;
@@ -1221,7 +1222,8 @@ static bool cfg80211_edmg_usable(struct wiphy *wiphy, u8 edmg_channels,
 
 bool _cfg80211_chandef_usable(struct wiphy *wiphy,
 			      const struct cfg80211_chan_def *chandef,
-			      u32 prohibited_flags, bool monitor)
+			      u32 prohibited_flags,
+			      u32 permitting_flags)
 {
 	struct ieee80211_sta_ht_cap *ht_cap;
 	struct ieee80211_sta_vht_cap *vht_cap;
@@ -1383,22 +1385,23 @@ bool _cfg80211_chandef_usable(struct wiphy *wiphy,
 
 	if (!cfg80211_secondary_chans_ok(wiphy,
 					 ieee80211_chandef_to_khz(chandef),
-					 width, prohibited_flags, monitor))
+					 width, prohibited_flags,
+					 permitting_flags))
 		return false;
 
 	if (!chandef->center_freq2)
 		return true;
 	return cfg80211_secondary_chans_ok(wiphy,
 					   MHZ_TO_KHZ(chandef->center_freq2),
-					   width, prohibited_flags, monitor);
+					   width, prohibited_flags,
+					   permitting_flags);
 }
 
 bool cfg80211_chandef_usable(struct wiphy *wiphy,
 			     const struct cfg80211_chan_def *chandef,
 			     u32 prohibited_flags)
 {
-	return _cfg80211_chandef_usable(wiphy, chandef, prohibited_flags,
-					false);
+	return _cfg80211_chandef_usable(wiphy, chandef, prohibited_flags, 0);
 }
 EXPORT_SYMBOL(cfg80211_chandef_usable);
 
@@ -1541,7 +1544,7 @@ static bool _cfg80211_reg_can_beacon(struct wiphy *wiphy,
 		prohibited_flags = IEEE80211_CHAN_DISABLED;
 	}
 
-	res = cfg80211_chandef_usable(wiphy, chandef, prohibited_flags);
+	res = _cfg80211_chandef_usable(wiphy, chandef, prohibited_flags, 0);
 
 	trace_cfg80211_return_bool(res);
 	return res;
diff --git a/net/wireless/core.h b/net/wireless/core.h
index 118f2f619828..470a18dc4cbf 100644
--- a/net/wireless/core.h
+++ b/net/wireless/core.h
@@ -494,7 +494,8 @@ bool cfg80211_wdev_on_sub_chan(struct wireless_dev *wdev,
 			       bool primary_only);
 bool _cfg80211_chandef_usable(struct wiphy *wiphy,
 			      const struct cfg80211_chan_def *chandef,
-			      u32 prohibited_flags, bool monitor);
+			      u32 prohibited_flags,
+			      u32 permitting_flags);
 
 static inline unsigned int elapsed_jiffies_msecs(unsigned long start)
 {
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index fd6e33a1eb2a..4628280abf1d 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -3347,7 +3347,7 @@ static int _nl80211_parse_chandef(struct cfg80211_registered_device *rdev,
 
 	if (!_cfg80211_chandef_usable(&rdev->wiphy, chandef,
 				      IEEE80211_CHAN_DISABLED,
-				      monitor)) {
+				      monitor ? IEEE80211_CHAN_CAN_MONITOR : 0)) {
 		NL_SET_ERR_MSG(extack, "(extension) channel is disabled");
 		return -EINVAL;
 	}
-- 
cgit v1.2.3-58-ga151


From 459662e83db6eb33f648213bba2c509f4d6dead9 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 23 May 2024 12:09:47 +0200
Subject: wifi: cfg80211: refactor 6 GHz AP power type parsing

Add cfg80211_get_6ghz_power_type() to parse the 6 GHz
power type from a given set of elements, which is now
only inside cfg80211_6ghz_power_type_valid().

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240523120945.84cdffd94085.I76f434ee12552e8be91273f3b2d776179eaa62f1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/core.h |  4 ++++
 net/wireless/scan.c | 69 ++++++++++++++++++++++++++++++++---------------------
 2 files changed, 46 insertions(+), 27 deletions(-)

diff --git a/net/wireless/core.h b/net/wireless/core.h
index 470a18dc4cbf..41c8c0e3ba2e 100644
--- a/net/wireless/core.h
+++ b/net/wireless/core.h
@@ -533,6 +533,10 @@ struct cfg80211_internal_bss *
 cfg80211_bss_update(struct cfg80211_registered_device *rdev,
 		    struct cfg80211_internal_bss *tmp,
 		    bool signal_valid, unsigned long ts);
+
+enum ieee80211_ap_reg_power
+cfg80211_get_6ghz_power_type(const u8 *elems, size_t elems_len);
+
 #ifdef CONFIG_CFG80211_DEVELOPER_WARNINGS
 #define CFG80211_DEV_WARN_ON(cond)	WARN_ON(cond)
 #else
diff --git a/net/wireless/scan.c b/net/wireless/scan.c
index 9b31274a1376..88ce365764f5 100644
--- a/net/wireless/scan.c
+++ b/net/wireless/scan.c
@@ -2163,38 +2163,53 @@ struct cfg80211_inform_single_bss_data {
 	u64 cannot_use_reasons;
 };
 
-static bool cfg80211_6ghz_power_type_valid(const u8 *ie, size_t ielen,
-					   const u32 flags)
+enum ieee80211_ap_reg_power
+cfg80211_get_6ghz_power_type(const u8 *elems, size_t elems_len)
 {
-	const struct element *tmp;
+	const struct ieee80211_he_6ghz_oper *he_6ghz_oper;
 	struct ieee80211_he_operation *he_oper;
+	const struct element *tmp;
 
-	tmp = cfg80211_find_ext_elem(WLAN_EID_EXT_HE_OPERATION, ie, ielen);
-	if (tmp && tmp->datalen >= sizeof(*he_oper) + 1 &&
-	    tmp->datalen >= ieee80211_he_oper_size(tmp->data + 1)) {
-		const struct ieee80211_he_6ghz_oper *he_6ghz_oper;
-
-		he_oper = (void *)&tmp->data[1];
-		he_6ghz_oper = ieee80211_he_6ghz_oper(he_oper);
-
-		if (!he_6ghz_oper)
-			return false;
+	tmp = cfg80211_find_ext_elem(WLAN_EID_EXT_HE_OPERATION,
+				     elems, elems_len);
+	if (!tmp || tmp->datalen < sizeof(*he_oper) + 1 ||
+	    tmp->datalen < ieee80211_he_oper_size(tmp->data + 1))
+		return IEEE80211_REG_UNSET_AP;
+
+	he_oper = (void *)&tmp->data[1];
+	he_6ghz_oper = ieee80211_he_6ghz_oper(he_oper);
+
+	if (!he_6ghz_oper)
+		return IEEE80211_REG_UNSET_AP;
+
+	switch (u8_get_bits(he_6ghz_oper->control,
+			    IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) {
+	case IEEE80211_6GHZ_CTRL_REG_LPI_AP:
+	case IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP:
+		return IEEE80211_REG_LPI_AP;
+	case IEEE80211_6GHZ_CTRL_REG_SP_AP:
+	case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP:
+		return IEEE80211_REG_SP_AP;
+	case IEEE80211_6GHZ_CTRL_REG_VLP_AP:
+		return IEEE80211_REG_VLP_AP;
+	default:
+		return IEEE80211_REG_UNSET_AP;
+	}
+}
 
-		switch (u8_get_bits(he_6ghz_oper->control,
-				    IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) {
-		case IEEE80211_6GHZ_CTRL_REG_LPI_AP:
-		case IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP:
-			return true;
-		case IEEE80211_6GHZ_CTRL_REG_SP_AP:
-		case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP:
-			return !(flags & IEEE80211_CHAN_NO_6GHZ_AFC_CLIENT);
-		case IEEE80211_6GHZ_CTRL_REG_VLP_AP:
-			return !(flags & IEEE80211_CHAN_NO_6GHZ_VLP_CLIENT);
-		default:
-			return false;
-		}
+static bool cfg80211_6ghz_power_type_valid(const u8 *elems, size_t elems_len,
+					   const u32 flags)
+{
+	switch (cfg80211_get_6ghz_power_type(elems, elems_len)) {
+	case IEEE80211_REG_LPI_AP:
+		return true;
+	case IEEE80211_REG_SP_AP:
+		return !(flags & IEEE80211_CHAN_NO_6GHZ_AFC_CLIENT);
+	case IEEE80211_REG_VLP_AP:
+		return !(flags & IEEE80211_CHAN_NO_6GHZ_VLP_CLIENT);
+	default:
+		return false;
 	}
-	return false;
 }
 
 /* Returned bss is reference counted and must be cleaned up appropriately. */
-- 
cgit v1.2.3-58-ga151


From 9fd171a71b9d4cd8855891c0ba7e2a152139b41a Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 23 May 2024 12:09:48 +0200
Subject: wifi: cfg80211: refactor regulatory beaconing checking

There are two functions exported now, with different settings,
refactor to just export a single function that take a struct
with different settings. This will make it easier to add more
parameters.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240523120945.d44c34dadfc2.I59b4403108e0dbf7fc6ae8f7522e1af520cffb1c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h | 54 ++++++++++++++++++++++++++++++++++++++++++++------
 net/wireless/chan.c    | 30 +++++++++++-----------------
 2 files changed, 60 insertions(+), 24 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index fb7d76513e1c..45ffeb110d36 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -8800,6 +8800,31 @@ static inline void cfg80211_report_obss_beacon(struct wiphy *wiphy,
 					sig_dbm);
 }
 
+/**
+ * struct cfg80211_beaconing_check_config - beacon check configuration
+ * @iftype: the interface type to check for
+ * @relax: allow IR-relaxation conditions to apply (e.g. another
+ *	interface connected already on the same channel)
+ *	NOTE: If this is set, wiphy mutex must be held.
+ */
+struct cfg80211_beaconing_check_config {
+	enum nl80211_iftype iftype;
+	bool relax;
+};
+
+/**
+ * cfg80211_reg_check_beaconing - check if beaconing is allowed
+ * @wiphy: the wiphy
+ * @chandef: the channel definition
+ * @cfg: additional parameters for the checking
+ *
+ * Return: %true if there is no secondary channel or the secondary channel(s)
+ * can be used for beaconing (i.e. is not a radar channel etc.)
+ */
+bool cfg80211_reg_check_beaconing(struct wiphy *wiphy,
+				  struct cfg80211_chan_def *chandef,
+				  struct cfg80211_beaconing_check_config *cfg);
+
 /**
  * cfg80211_reg_can_beacon - check if beaconing is allowed
  * @wiphy: the wiphy
@@ -8809,9 +8834,17 @@ static inline void cfg80211_report_obss_beacon(struct wiphy *wiphy,
  * Return: %true if there is no secondary channel or the secondary channel(s)
  * can be used for beaconing (i.e. is not a radar channel etc.)
  */
-bool cfg80211_reg_can_beacon(struct wiphy *wiphy,
-			     struct cfg80211_chan_def *chandef,
-			     enum nl80211_iftype iftype);
+static inline bool
+cfg80211_reg_can_beacon(struct wiphy *wiphy,
+			struct cfg80211_chan_def *chandef,
+			enum nl80211_iftype iftype)
+{
+	struct cfg80211_beaconing_check_config config = {
+		.iftype = iftype,
+	};
+
+	return cfg80211_reg_check_beaconing(wiphy, chandef, &config);
+}
 
 /**
  * cfg80211_reg_can_beacon_relax - check if beaconing is allowed with relaxation
@@ -8826,9 +8859,18 @@ bool cfg80211_reg_can_beacon(struct wiphy *wiphy,
  *
  * Context: Requires the wiphy mutex to be held.
  */
-bool cfg80211_reg_can_beacon_relax(struct wiphy *wiphy,
-				   struct cfg80211_chan_def *chandef,
-				   enum nl80211_iftype iftype);
+static inline bool
+cfg80211_reg_can_beacon_relax(struct wiphy *wiphy,
+			      struct cfg80211_chan_def *chandef,
+			      enum nl80211_iftype iftype)
+{
+	struct cfg80211_beaconing_check_config config = {
+		.iftype = iftype,
+		.relax = true,
+	};
+
+	return cfg80211_reg_check_beaconing(wiphy, chandef, &config);
+}
 
 /**
  * cfg80211_ch_switch_notify - update wdev channel and notify userspace
diff --git a/net/wireless/chan.c b/net/wireless/chan.c
index 360480604515..8b1796130b28 100644
--- a/net/wireless/chan.c
+++ b/net/wireless/chan.c
@@ -1550,22 +1550,12 @@ static bool _cfg80211_reg_can_beacon(struct wiphy *wiphy,
 	return res;
 }
 
-bool cfg80211_reg_can_beacon(struct wiphy *wiphy,
-			     struct cfg80211_chan_def *chandef,
-			     enum nl80211_iftype iftype)
-{
-	return _cfg80211_reg_can_beacon(wiphy, chandef, iftype, true);
-}
-EXPORT_SYMBOL(cfg80211_reg_can_beacon);
-
-bool cfg80211_reg_can_beacon_relax(struct wiphy *wiphy,
-				   struct cfg80211_chan_def *chandef,
-				   enum nl80211_iftype iftype)
+bool cfg80211_reg_check_beaconing(struct wiphy *wiphy,
+				  struct cfg80211_chan_def *chandef,
+				  struct cfg80211_beaconing_check_config *cfg)
 {
 	struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
-	bool check_no_ir;
-
-	lockdep_assert_held(&rdev->wiphy.mtx);
+	bool check_no_ir = true;
 
 	/*
 	 * Under certain conditions suggested by some regulatory bodies a
@@ -1573,12 +1563,16 @@ bool cfg80211_reg_can_beacon_relax(struct wiphy *wiphy,
 	 * only if such relaxations are not enabled and the conditions are not
 	 * met.
 	 */
-	check_no_ir = !cfg80211_ir_permissive_chan(wiphy, iftype,
-						   chandef->chan);
+	if (cfg->relax) {
+		lockdep_assert_held(&rdev->wiphy.mtx);
+		check_no_ir = !cfg80211_ir_permissive_chan(wiphy, cfg->iftype,
+							   chandef->chan);
+	}
 
-	return _cfg80211_reg_can_beacon(wiphy, chandef, iftype, check_no_ir);
+	return _cfg80211_reg_can_beacon(wiphy, chandef, cfg->iftype,
+					check_no_ir);
 }
-EXPORT_SYMBOL(cfg80211_reg_can_beacon_relax);
+EXPORT_SYMBOL(cfg80211_reg_check_beaconing);
 
 int cfg80211_set_monitor_channel(struct cfg80211_registered_device *rdev,
 				 struct cfg80211_chan_def *chandef)
-- 
cgit v1.2.3-58-ga151


From c1d8bd8d777d55f6708ca6e47c54dbe9f66f9bbb Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 23 May 2024 12:09:49 +0200
Subject: wifi: cfg80211: add regulatory flag to allow VLP AP operation

Add a regulatory flag to allow VLP AP operation even on
channels otherwise marked NO_IR, which may be possible
in some regulatory domains/countries.

Note that this requires checking also when the beacon is
changed, since that may change the regulatory power type.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240523120945.63792ce19790.Ie2a02750d283b78fbf3c686b10565fb0388889e2@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h       |  6 ++++++
 include/uapi/linux/nl80211.h |  6 ++++++
 net/wireless/chan.c          | 37 ++++++++++++++++++++++++++-----------
 net/wireless/nl80211.c       | 27 +++++++++++++++++++++++++--
 net/wireless/reg.c           |  2 ++
 net/wireless/trace.h         | 15 +++++++++------
 6 files changed, 74 insertions(+), 19 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 45ffeb110d36..6f992aff74ae 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -125,6 +125,8 @@ struct wiphy;
  * @IEEE80211_CHAN_CAN_MONITOR: This channel can be used for monitor
  *	mode even in the presence of other (regulatory) restrictions,
  *	even if it is otherwise disabled.
+ * @IEEE80211_CHAN_ALLOW_6GHZ_VLP_AP: Allow using this channel for AP operation
+ *	with very low power (VLP), even if otherwise set to NO_IR.
  */
 enum ieee80211_channel_flags {
 	IEEE80211_CHAN_DISABLED			= BIT(0),
@@ -152,6 +154,7 @@ enum ieee80211_channel_flags {
 	IEEE80211_CHAN_NO_6GHZ_VLP_CLIENT	= BIT(22),
 	IEEE80211_CHAN_NO_6GHZ_AFC_CLIENT	= BIT(23),
 	IEEE80211_CHAN_CAN_MONITOR		= BIT(24),
+	IEEE80211_CHAN_ALLOW_6GHZ_VLP_AP	= BIT(25),
 };
 
 #define IEEE80211_CHAN_NO_HT40 \
@@ -8806,9 +8809,12 @@ static inline void cfg80211_report_obss_beacon(struct wiphy *wiphy,
  * @relax: allow IR-relaxation conditions to apply (e.g. another
  *	interface connected already on the same channel)
  *	NOTE: If this is set, wiphy mutex must be held.
+ * @reg_power: &enum ieee80211_ap_reg_power value indicating the
+ *	advertised/used 6 GHz regulatory power setting
  */
 struct cfg80211_beaconing_check_config {
 	enum nl80211_iftype iftype;
+	enum ieee80211_ap_reg_power reg_power;
 	bool relax;
 };
 
diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index f917bc6c9b6f..6ae3997061b6 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -4277,6 +4277,8 @@ enum nl80211_wmm_rule {
  * @NL80211_FREQUENCY_ATTR_CAN_MONITOR: This channel can be used in monitor
  *	mode despite other (regulatory) restrictions, even if the channel is
  *	otherwise completely disabled.
+ * @NL80211_FREQUENCY_ATTR_ALLOW_6GHZ_VLP_AP: This channel can be used for a
+ *	very low power (VLP) AP, despite being NO_IR.
  * @NL80211_FREQUENCY_ATTR_MAX: highest frequency attribute number
  *	currently defined
  * @__NL80211_FREQUENCY_ATTR_AFTER_LAST: internal use
@@ -4320,6 +4322,7 @@ enum nl80211_frequency_attr {
 	NL80211_FREQUENCY_ATTR_NO_6GHZ_VLP_CLIENT,
 	NL80211_FREQUENCY_ATTR_NO_6GHZ_AFC_CLIENT,
 	NL80211_FREQUENCY_ATTR_CAN_MONITOR,
+	NL80211_FREQUENCY_ATTR_ALLOW_6GHZ_VLP_AP,
 
 	/* keep last */
 	__NL80211_FREQUENCY_ATTR_AFTER_LAST,
@@ -4529,6 +4532,8 @@ enum nl80211_sched_scan_match_attr {
  *	Should be used together with %NL80211_RRF_DFS only.
  * @NL80211_RRF_NO_6GHZ_VLP_CLIENT: Client connection to VLP AP not allowed
  * @NL80211_RRF_NO_6GHZ_AFC_CLIENT: Client connection to AFC AP not allowed
+ * @NL80211_RRF_ALLOW_6GHZ_VLP_AP: Very low power (VLP) AP can be permitted
+ *	despite NO_IR configuration.
  */
 enum nl80211_reg_rule_flags {
 	NL80211_RRF_NO_OFDM		= 1<<0,
@@ -4553,6 +4558,7 @@ enum nl80211_reg_rule_flags {
 	NL80211_RRF_DFS_CONCURRENT	= 1<<21,
 	NL80211_RRF_NO_6GHZ_VLP_CLIENT	= 1<<22,
 	NL80211_RRF_NO_6GHZ_AFC_CLIENT	= 1<<23,
+	NL80211_RRF_ALLOW_6GHZ_VLP_AP	= 1<<24,
 };
 
 #define NL80211_RRF_PASSIVE_SCAN	NL80211_RRF_NO_IR
diff --git a/net/wireless/chan.c b/net/wireless/chan.c
index 8b1796130b28..bf2fdcd42019 100644
--- a/net/wireless/chan.c
+++ b/net/wireless/chan.c
@@ -1523,28 +1523,38 @@ static bool cfg80211_ir_permissive_chan(struct wiphy *wiphy,
 static bool _cfg80211_reg_can_beacon(struct wiphy *wiphy,
 				     struct cfg80211_chan_def *chandef,
 				     enum nl80211_iftype iftype,
-				     bool check_no_ir)
+				     u32 prohibited_flags,
+				     u32 permitting_flags)
 {
-	bool res;
-	u32 prohibited_flags = IEEE80211_CHAN_DISABLED;
+	bool res, check_radar;
 	int dfs_required;
 
-	trace_cfg80211_reg_can_beacon(wiphy, chandef, iftype, check_no_ir);
+	trace_cfg80211_reg_can_beacon(wiphy, chandef, iftype,
+				      prohibited_flags,
+				      permitting_flags);
 
-	if (check_no_ir)
-		prohibited_flags |= IEEE80211_CHAN_NO_IR;
+	if (!_cfg80211_chandef_usable(wiphy, chandef,
+				      IEEE80211_CHAN_DISABLED, 0))
+		return false;
 
 	dfs_required = cfg80211_chandef_dfs_required(wiphy, chandef, iftype);
-	if (dfs_required != 0)
-		prohibited_flags |= IEEE80211_CHAN_RADAR;
+	check_radar = dfs_required != 0;
 
 	if (dfs_required > 0 &&
 	    cfg80211_chandef_dfs_available(wiphy, chandef)) {
 		/* We can skip IEEE80211_CHAN_NO_IR if chandef dfs available */
-		prohibited_flags = IEEE80211_CHAN_DISABLED;
+		prohibited_flags &= ~IEEE80211_CHAN_NO_IR;
+		check_radar = false;
 	}
 
-	res = _cfg80211_chandef_usable(wiphy, chandef, prohibited_flags, 0);
+	if (check_radar &&
+	    !_cfg80211_chandef_usable(wiphy, chandef,
+				      IEEE80211_CHAN_RADAR, 0))
+		return false;
+
+	res = _cfg80211_chandef_usable(wiphy, chandef,
+				       prohibited_flags,
+				       permitting_flags);
 
 	trace_cfg80211_return_bool(res);
 	return res;
@@ -1555,6 +1565,7 @@ bool cfg80211_reg_check_beaconing(struct wiphy *wiphy,
 				  struct cfg80211_beaconing_check_config *cfg)
 {
 	struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
+	u32 permitting_flags = 0;
 	bool check_no_ir = true;
 
 	/*
@@ -1569,8 +1580,12 @@ bool cfg80211_reg_check_beaconing(struct wiphy *wiphy,
 							   chandef->chan);
 	}
 
+	if (cfg->reg_power == IEEE80211_REG_VLP_AP)
+		permitting_flags |= IEEE80211_CHAN_ALLOW_6GHZ_VLP_AP;
+
 	return _cfg80211_reg_can_beacon(wiphy, chandef, cfg->iftype,
-					check_no_ir);
+					check_no_ir ? IEEE80211_CHAN_NO_IR : 0,
+					permitting_flags);
 }
 EXPORT_SYMBOL(cfg80211_reg_check_beaconing);
 
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 4628280abf1d..a94e73c133f7 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -1207,6 +1207,9 @@ static int nl80211_msg_put_channel(struct sk_buff *msg, struct wiphy *wiphy,
 		if ((chan->flags & IEEE80211_CHAN_CAN_MONITOR) &&
 		    nla_put_flag(msg, NL80211_FREQUENCY_ATTR_CAN_MONITOR))
 			goto nla_put_failure;
+		if ((chan->flags & IEEE80211_CHAN_ALLOW_6GHZ_VLP_AP) &&
+		    nla_put_flag(msg, NL80211_FREQUENCY_ATTR_ALLOW_6GHZ_VLP_AP))
+			goto nla_put_failure;
 	}
 
 	if (nla_put_u32(msg, NL80211_FREQUENCY_ATTR_MAX_TX_POWER,
@@ -5954,6 +5957,7 @@ static int nl80211_validate_ap_phy_operation(struct cfg80211_ap_settings *params
 static int nl80211_start_ap(struct sk_buff *skb, struct genl_info *info)
 {
 	struct cfg80211_registered_device *rdev = info->user_ptr[0];
+	struct cfg80211_beaconing_check_config beacon_check = {};
 	unsigned int link_id = nl80211_link_id(info->attrs);
 	struct net_device *dev = info->user_ptr[1];
 	struct wireless_dev *wdev = dev->ieee80211_ptr;
@@ -6103,8 +6107,13 @@ static int nl80211_start_ap(struct sk_buff *skb, struct genl_info *info)
 		goto out;
 	}
 
-	if (!cfg80211_reg_can_beacon_relax(&rdev->wiphy, &params->chandef,
-					   wdev->iftype)) {
+	beacon_check.iftype = wdev->iftype;
+	beacon_check.relax = true;
+	beacon_check.reg_power =
+		cfg80211_get_6ghz_power_type(params->beacon.tail,
+					     params->beacon.tail_len);
+	if (!cfg80211_reg_check_beaconing(&rdev->wiphy, &params->chandef,
+					  &beacon_check)) {
 		err = -EINVAL;
 		goto out;
 	}
@@ -6261,6 +6270,7 @@ out:
 static int nl80211_set_beacon(struct sk_buff *skb, struct genl_info *info)
 {
 	struct cfg80211_registered_device *rdev = info->user_ptr[0];
+	struct cfg80211_beaconing_check_config beacon_check = {};
 	unsigned int link_id = nl80211_link_id(info->attrs);
 	struct net_device *dev = info->user_ptr[1];
 	struct wireless_dev *wdev = dev->ieee80211_ptr;
@@ -6287,6 +6297,19 @@ static int nl80211_set_beacon(struct sk_buff *skb, struct genl_info *info)
 	if (err)
 		goto out;
 
+	/* recheck beaconing is permitted with possibly changed power type */
+	beacon_check.iftype = wdev->iftype;
+	beacon_check.relax = true;
+	beacon_check.reg_power =
+		cfg80211_get_6ghz_power_type(params->beacon.tail,
+					     params->beacon.tail_len);
+	if (!cfg80211_reg_check_beaconing(&rdev->wiphy,
+					  &wdev->links[link_id].ap.chandef,
+					  &beacon_check)) {
+		err = -EINVAL;
+		goto out;
+	}
+
 	attr = info->attrs[NL80211_ATTR_FILS_DISCOVERY];
 	if (attr) {
 		err = nl80211_parse_fils_discovery(rdev, attr,
diff --git a/net/wireless/reg.c b/net/wireless/reg.c
index 3cef0021a3db..4a27f3823e25 100644
--- a/net/wireless/reg.c
+++ b/net/wireless/reg.c
@@ -1600,6 +1600,8 @@ static u32 map_regdom_flags(u32 rd_flags)
 		channel_flags |= IEEE80211_CHAN_NO_6GHZ_AFC_CLIENT;
 	if (rd_flags & NL80211_RRF_PSD)
 		channel_flags |= IEEE80211_CHAN_PSD;
+	if (rd_flags & NL80211_RRF_ALLOW_6GHZ_VLP_AP)
+		channel_flags |= IEEE80211_CHAN_ALLOW_6GHZ_VLP_AP;
 	return channel_flags;
 }
 
diff --git a/net/wireless/trace.h b/net/wireless/trace.h
index 6ef9294747e3..5c26f065bd68 100644
--- a/net/wireless/trace.h
+++ b/net/wireless/trace.h
@@ -3389,23 +3389,26 @@ TRACE_EVENT(cfg80211_cqm_rssi_notify,
 
 TRACE_EVENT(cfg80211_reg_can_beacon,
 	TP_PROTO(struct wiphy *wiphy, struct cfg80211_chan_def *chandef,
-		 enum nl80211_iftype iftype, bool check_no_ir),
-	TP_ARGS(wiphy, chandef, iftype, check_no_ir),
+		 enum nl80211_iftype iftype, u32 prohibited_flags,
+		 u32 permitting_flags),
+	TP_ARGS(wiphy, chandef, iftype, prohibited_flags, permitting_flags),
 	TP_STRUCT__entry(
 		WIPHY_ENTRY
 		CHAN_DEF_ENTRY
 		__field(enum nl80211_iftype, iftype)
-		__field(bool, check_no_ir)
+		__field(u32, prohibited_flags)
+		__field(u32, permitting_flags)
 	),
 	TP_fast_assign(
 		WIPHY_ASSIGN;
 		CHAN_DEF_ASSIGN(chandef);
 		__entry->iftype = iftype;
-		__entry->check_no_ir = check_no_ir;
+		__entry->prohibited_flags = prohibited_flags;
+		__entry->permitting_flags = permitting_flags;
 	),
-	TP_printk(WIPHY_PR_FMT ", " CHAN_DEF_PR_FMT ", iftype=%d check_no_ir=%s",
+	TP_printk(WIPHY_PR_FMT ", " CHAN_DEF_PR_FMT ", iftype=%d prohibited_flags=0x%x permitting_flags=0x%x",
 		  WIPHY_PR_ARG, CHAN_DEF_PR_ARG, __entry->iftype,
-		  BOOL_TO_STR(__entry->check_no_ir))
+		  __entry->prohibited_flags, __entry->permitting_flags)
 );
 
 TRACE_EVENT(cfg80211_chandef_dfs_required,
-- 
cgit v1.2.3-58-ga151


From 91b193d546683558a8799ffb2e2f935d3800633e Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Sun, 2 Jun 2024 10:22:00 +0300
Subject: wifi: cfg80211: reject non-conformant 6 GHz center frequencies

On 6 GHz (and also 5 GHz to some degree), only a specific set of center
frequencies should be used depending on the channel bandwidth. Verify
this is the case on 6 GHz. For 5 GHz, we are more accepting as there are
APs that got it wrong historically.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240602102200.876b10a2beda.I0d3d0daea4014e99654437ff6691378dbe452652@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/chan.c       | 38 ++++++++++++++++++++++++++++++++++++++
 net/wireless/tests/chan.c | 22 +++++++++++-----------
 2 files changed, 49 insertions(+), 11 deletions(-)

diff --git a/net/wireless/chan.c b/net/wireless/chan.c
index bf2fdcd42019..e579d7e1425f 100644
--- a/net/wireless/chan.c
+++ b/net/wireless/chan.c
@@ -263,6 +263,37 @@ static int cfg80211_chandef_get_width(const struct cfg80211_chan_def *c)
 	return nl80211_chan_width_to_mhz(c->width);
 }
 
+static bool cfg80211_valid_center_freq(u32 center,
+				       enum nl80211_chan_width width)
+{
+	int bw;
+	int step;
+
+	/* We only do strict verification on 6 GHz */
+	if (center < 5955 || center > 7115)
+		return true;
+
+	bw = nl80211_chan_width_to_mhz(width);
+	if (bw < 0)
+		return false;
+
+	/* Validate that the channels bw is entirely within the 6 GHz band */
+	if (center - bw / 2 < 5945 || center + bw / 2 > 7125)
+		return false;
+
+	/* With 320 MHz the permitted channels overlap */
+	if (bw == 320)
+		step = 160;
+	else
+		step = bw;
+
+	/*
+	 * Valid channels are packed from lowest frequency towards higher ones.
+	 * So test that the lower frequency alignes with one of these steps.
+	 */
+	return (center - bw / 2 - 5945) % step == 0;
+}
+
 bool cfg80211_chandef_valid(const struct cfg80211_chan_def *chandef)
 {
 	u32 control_freq, oper_freq;
@@ -374,6 +405,13 @@ bool cfg80211_chandef_valid(const struct cfg80211_chan_def *chandef)
 		return false;
 	}
 
+	if (!cfg80211_valid_center_freq(chandef->center_freq1, chandef->width))
+		return false;
+
+	if (chandef->width == NL80211_CHAN_WIDTH_80P80 &&
+	    !cfg80211_valid_center_freq(chandef->center_freq2, chandef->width))
+		return false;
+
 	/* channel 14 is only for IEEE 802.11b */
 	if (chandef->center_freq1 == 2484 &&
 	    chandef->width != NL80211_CHAN_WIDTH_20_NOHT)
diff --git a/net/wireless/tests/chan.c b/net/wireless/tests/chan.c
index d02258ac2dab..74bbee25085f 100644
--- a/net/wireless/tests/chan.c
+++ b/net/wireless/tests/chan.c
@@ -113,16 +113,16 @@ static const struct chandef_compat_case {
 		},
 	},
 	{
-		.desc = "different primary 160 MHz",
+		.desc = "different primary 320 MHz",
 		.c1 = {
 			.width = NL80211_CHAN_WIDTH_320,
 			.chan = &chan_6ghz_105,
-			.center_freq1 = 6475 + 150,
+			.center_freq1 = 6475 + 110,
 		},
 		.c2 = {
 			.width = NL80211_CHAN_WIDTH_320,
 			.chan = &chan_6ghz_105,
-			.center_freq1 = 6475 - 10,
+			.center_freq1 = 6475 - 50,
 		},
 	},
 	{
@@ -131,12 +131,12 @@ static const struct chandef_compat_case {
 		.c1 = {
 			.width = NL80211_CHAN_WIDTH_160,
 			.chan = &chan_6ghz_105,
-			.center_freq1 = 6475 + 70,
+			.center_freq1 = 6475 + 30,
 		},
 		.c2 = {
 			.width = NL80211_CHAN_WIDTH_320,
 			.chan = &chan_6ghz_105,
-			.center_freq1 = 6475 - 10,
+			.center_freq1 = 6475 - 50,
 		},
 		.compat = true,
 	},
@@ -145,12 +145,12 @@ static const struct chandef_compat_case {
 		.c1 = {
 			.width = NL80211_CHAN_WIDTH_160,
 			.chan = &chan_6ghz_105,
-			.center_freq1 = 6475 + 70,
+			.center_freq1 = 6475 + 30,
 		},
 		.c2 = {
 			.width = NL80211_CHAN_WIDTH_320,
 			.chan = &chan_6ghz_105,
-			.center_freq1 = 6475 - 10,
+			.center_freq1 = 6475 - 50,
 			.punctured = 0xf,
 		},
 		.compat = true,
@@ -160,13 +160,13 @@ static const struct chandef_compat_case {
 		.c1 = {
 			.width = NL80211_CHAN_WIDTH_160,
 			.chan = &chan_6ghz_105,
-			.center_freq1 = 6475 + 70,
+			.center_freq1 = 6475 + 30,
 			.punctured = 0xc0,
 		},
 		.c2 = {
 			.width = NL80211_CHAN_WIDTH_320,
 			.chan = &chan_6ghz_105,
-			.center_freq1 = 6475 - 10,
+			.center_freq1 = 6475 - 50,
 			.punctured = 0xc000,
 		},
 		.compat = true,
@@ -176,13 +176,13 @@ static const struct chandef_compat_case {
 		.c1 = {
 			.width = NL80211_CHAN_WIDTH_160,
 			.chan = &chan_6ghz_105,
-			.center_freq1 = 6475 + 70,
+			.center_freq1 = 6475 + 30,
 			.punctured = 0x80,
 		},
 		.c2 = {
 			.width = NL80211_CHAN_WIDTH_320,
 			.chan = &chan_6ghz_105,
-			.center_freq1 = 6475 - 10,
+			.center_freq1 = 6475 - 50,
 			.punctured = 0xc000,
 		},
 	},
-- 
cgit v1.2.3-58-ga151


From f737b70434f332042fcbe37f85ee3ee424959403 Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Wed, 5 Jun 2024 13:57:16 +0300
Subject: wifi: nl80211: remove the FTMs per burst limit for NDP ranging

In NDP ranging, the number of NDP exchanges is not negotiated
and thus is not limited by the protocol. Remove the limit on
FTMs per burst for trigger based and non trigger based ranging.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605135233.916e228537d9.I5fe4c1cefa1c1328726e7615dd5a0d861c694381@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/nl80211.c |  3 +--
 net/wireless/pmsr.c    | 10 +++++++++-
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index a94e73c133f7..fcac7dedcd61 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -315,8 +315,7 @@ nl80211_pmsr_ftm_req_attr_policy[NL80211_PMSR_FTM_REQ_ATTR_MAX + 1] = {
 	[NL80211_PMSR_FTM_REQ_ATTR_BURST_PERIOD] = { .type = NLA_U16 },
 	[NL80211_PMSR_FTM_REQ_ATTR_BURST_DURATION] =
 		NLA_POLICY_MAX(NLA_U8, 15),
-	[NL80211_PMSR_FTM_REQ_ATTR_FTMS_PER_BURST] =
-		NLA_POLICY_MAX(NLA_U8, 31),
+	[NL80211_PMSR_FTM_REQ_ATTR_FTMS_PER_BURST] = { .type = NLA_U8 },
 	[NL80211_PMSR_FTM_REQ_ATTR_NUM_FTMR_RETRIES] = { .type = NLA_U8 },
 	[NL80211_PMSR_FTM_REQ_ATTR_REQUEST_LCI] = { .type = NLA_FLAG },
 	[NL80211_PMSR_FTM_REQ_ATTR_REQUEST_CIVICLOC] = { .type = NLA_FLAG },
diff --git a/net/wireless/pmsr.c b/net/wireless/pmsr.c
index c569c37da317..0396fa19bdf1 100644
--- a/net/wireless/pmsr.c
+++ b/net/wireless/pmsr.c
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 */
 /*
- * Copyright (C) 2018 - 2021, 2023 Intel Corporation
+ * Copyright (C) 2018 - 2021, 2023 - 2024 Intel Corporation
  */
 #include <net/cfg80211.h>
 #include "core.h"
@@ -148,6 +148,14 @@ static int pmsr_parse_ftm(struct cfg80211_registered_device *rdev,
 		return -EINVAL;
 	}
 
+	if (out->ftm.ftms_per_burst > 31 && !out->ftm.non_trigger_based &&
+	    !out->ftm.trigger_based) {
+		NL_SET_ERR_MSG_ATTR(info->extack,
+				    tb[NL80211_PMSR_FTM_REQ_ATTR_FTMS_PER_BURST],
+				    "FTM: FTMs per burst must be set lower than 31");
+		return -ERANGE;
+	}
+
 	if ((out->ftm.trigger_based || out->ftm.non_trigger_based) &&
 	    out->ftm.preamble != NL80211_PREAMBLE_HE) {
 		NL_SET_ERR_MSG_ATTR(info->extack,
-- 
cgit v1.2.3-58-ga151


From a7bb6b5d36766e749d5d4cf5391f2ebce608968a Mon Sep 17 00:00:00 2001
From: Avraham Stern <avraham.stern@intel.com>
Date: Wed, 5 Jun 2024 13:57:17 +0300
Subject: wifi: mac80211_hwsim: add 320 MHz to hwsim channel widths

Setting a channel with 320 MHz channel width over hwsim results in an
array-index-out-of-bounds error. Fix it by adding 320 MHz to hwsim
supported channel widths.

Signed-off-by: Avraham Stern <avraham.stern@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605135233.a766c1465566.Ib859c7233511b61b8a34022cfceeb4971c739d80@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index c5d896994e70..fbf24870209d 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -2361,6 +2361,7 @@ static const char * const hwsim_chanwidths[] = {
 	[NL80211_CHAN_WIDTH_4] = "4MHz",
 	[NL80211_CHAN_WIDTH_8] = "8MHz",
 	[NL80211_CHAN_WIDTH_16] = "16MHz",
+	[NL80211_CHAN_WIDTH_320] = "eht320",
 };
 
 static int mac80211_hwsim_config(struct ieee80211_hw *hw, u32 changed)
-- 
cgit v1.2.3-58-ga151


From 03ecd745dde181f537bf84374caafb121463136b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 5 Jun 2024 13:57:18 +0300
Subject: wifi: mac80211: fix erroneous errors for STA changes

When e.g. wpa_supplicant sets only the MLD "sta" authorized
state, the code actually applies that change, but then returns
an error to userspace anyway because there were no changes to
the link station, and no link ID was given. However, it's not
incorrect to not have a link ID when wanting to change only
the MLD peer ("sta") state, so the code shouldn't require it.

To fix this, separate the "new_link" argument out into a new
three-state enum, because if modify is called on a link STA
only, it should return an error if no link is given or if it
doesn't exist. For modify on the MLD "sta", not having a link
ID is OK, but if there is one it should be validated.

This seems to not have mattered much as wpa_supplicant just
prints a message and continues, and the authorized state was
already set before this error return. However, in the later
code powersave recalculation etc. will be skipped, so that it
may result in never allowing powersave on MLO connections.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605135233.48e2b8af07e3.Ib9793c383fcba118c05100e024f4a11a1c3d0e85@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c | 55 +++++++++++++++++++++++++++++++++++-------------------
 1 file changed, 36 insertions(+), 19 deletions(-)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index afb361a043d9..65e7cf635d76 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1814,8 +1814,15 @@ static void sta_apply_mesh_params(struct ieee80211_local *local,
 #endif
 }
 
+enum sta_link_apply_mode {
+	STA_LINK_MODE_NEW,
+	STA_LINK_MODE_STA_MODIFY,
+	STA_LINK_MODE_LINK_MODIFY,
+};
+
 static int sta_link_apply_parameters(struct ieee80211_local *local,
-				     struct sta_info *sta, bool new_link,
+				     struct sta_info *sta,
+				     enum sta_link_apply_mode mode,
 				     struct link_station_parameters *params)
 {
 	int ret = 0;
@@ -1827,18 +1834,29 @@ static int sta_link_apply_parameters(struct ieee80211_local *local,
 	struct link_sta_info *link_sta =
 		rcu_dereference_protected(sta->link[link_id],
 					  lockdep_is_held(&local->hw.wiphy->mtx));
-
-	/*
-	 * If there are no changes, then accept a link that exist,
-	 * unless it's a new link.
-	 */
-	if (params->link_id >= 0 && !new_link &&
-	    !params->link_mac && !params->txpwr_set &&
-	    !params->supported_rates_len &&
-	    !params->ht_capa && !params->vht_capa &&
-	    !params->he_capa && !params->eht_capa &&
-	    !params->opmode_notif_used)
-		return 0;
+	bool changes = params->link_mac ||
+		       params->txpwr_set ||
+		       params->supported_rates_len ||
+		       params->ht_capa ||
+		       params->vht_capa ||
+		       params->he_capa ||
+		       params->eht_capa ||
+		       params->opmode_notif_used;
+
+	switch (mode) {
+	case STA_LINK_MODE_NEW:
+		if (!params->link_mac)
+			return -EINVAL;
+		break;
+	case STA_LINK_MODE_LINK_MODIFY:
+		break;
+	case STA_LINK_MODE_STA_MODIFY:
+		if (params->link_id >= 0)
+			break;
+		if (!changes)
+			return 0;
+		break;
+	}
 
 	if (!link || !link_sta)
 		return -EINVAL;
@@ -1848,15 +1866,13 @@ static int sta_link_apply_parameters(struct ieee80211_local *local,
 		return -EINVAL;
 
 	if (params->link_mac) {
-		if (new_link) {
+		if (mode == STA_LINK_MODE_NEW) {
 			memcpy(link_sta->addr, params->link_mac, ETH_ALEN);
 			memcpy(link_sta->pub->addr, params->link_mac, ETH_ALEN);
 		} else if (!ether_addr_equal(link_sta->addr,
 					     params->link_mac)) {
 			return -EINVAL;
 		}
-	} else if (new_link) {
-		return -EINVAL;
 	}
 
 	if (params->txpwr_set) {
@@ -2028,7 +2044,7 @@ static int sta_apply_parameters(struct ieee80211_local *local,
 	if (params->listen_interval >= 0)
 		sta->listen_interval = params->listen_interval;
 
-	ret = sta_link_apply_parameters(local, sta, false,
+	ret = sta_link_apply_parameters(local, sta, STA_LINK_MODE_STA_MODIFY,
 					&params->link_sta_params);
 	if (ret)
 		return ret;
@@ -5005,7 +5021,7 @@ ieee80211_add_link_station(struct wiphy *wiphy, struct net_device *dev,
 	if (ret)
 		return ret;
 
-	ret = sta_link_apply_parameters(local, sta, true, params);
+	ret = sta_link_apply_parameters(local, sta, STA_LINK_MODE_NEW, params);
 	if (ret) {
 		ieee80211_sta_free_link(sta, params->link_id);
 		return ret;
@@ -5032,7 +5048,8 @@ ieee80211_mod_link_station(struct wiphy *wiphy, struct net_device *dev,
 	if (!(sta->sta.valid_links & BIT(params->link_id)))
 		return -EINVAL;
 
-	return sta_link_apply_parameters(local, sta, false, params);
+	return sta_link_apply_parameters(local, sta, STA_LINK_MODE_LINK_MODIFY,
+					 params);
 }
 
 static int
-- 
cgit v1.2.3-58-ga151


From 642508a42f74d7467aae7c56dff3016db64a25bd Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 5 Jun 2024 13:57:19 +0300
Subject: wifi: mac80211: clean up 'ret' in sta_link_apply_parameters()

There's no need to have the always-zero ret variable in
the function scope, move it into the inner scope only.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605135233.eb7a24632d98.I72d7fe1da89d4b89bcfd0f5fb9057e3e69355cfe@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 65e7cf635d76..3236477424b8 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1825,7 +1825,6 @@ static int sta_link_apply_parameters(struct ieee80211_local *local,
 				     enum sta_link_apply_mode mode,
 				     struct link_station_parameters *params)
 {
-	int ret = 0;
 	struct ieee80211_supported_band *sband;
 	struct ieee80211_sub_if_data *sdata = sta->sdata;
 	u32 link_id = params->link_id < 0 ? 0 : params->link_id;
@@ -1876,6 +1875,8 @@ static int sta_link_apply_parameters(struct ieee80211_local *local,
 	}
 
 	if (params->txpwr_set) {
+		int ret;
+
 		link_sta->pub->txpwr.type = params->txpwr.type;
 		if (params->txpwr.type == NL80211_TX_POWER_LIMITED)
 			link_sta->pub->txpwr.power = params->txpwr.power;
@@ -1928,7 +1929,7 @@ static int sta_link_apply_parameters(struct ieee80211_local *local,
 
 	ieee80211_sta_init_nss(link_sta);
 
-	return ret;
+	return 0;
 }
 
 static int sta_apply_parameters(struct ieee80211_local *local,
-- 
cgit v1.2.3-58-ga151


From 671b6b1b1ea5c260549c8f7401828147c40fa9cf Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Wed, 5 Jun 2024 13:57:20 +0300
Subject: wifi: cfg80211: honor WIPHY_FLAG_SPLIT_SCAN_6GHZ in
 cfg80211_conn_scan

If a user uses iw to connect to a network and we don't have any
information about the existing networks, cfg80211 will trigger a scan
internally even if the user didn't ask for a scan. This scan is
implemented by cfg80211_conn_scan(). This function called rdev_scan()
directly without honoring the WIPHY_FLAG_SPLIT_SCAN_6GHZ flag.
Use cfg80211_scan instead, this will split the scan if the low level
driver asked to.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605135233.33f03661476a.I7b5be20a55aafe012cd9ddb3b4ba2d46b256ace4@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/sme.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/wireless/sme.c b/net/wireless/sme.c
index a8ad55f11133..e419aa8c4a5a 100644
--- a/net/wireless/sme.c
+++ b/net/wireless/sme.c
@@ -5,7 +5,7 @@
  * (for nl80211's connect() and wext)
  *
  * Copyright 2009	Johannes Berg <johannes@sipsolutions.net>
- * Copyright (C) 2009, 2020, 2022-2023 Intel Corporation. All rights reserved.
+ * Copyright (C) 2009, 2020, 2022-2024 Intel Corporation. All rights reserved.
  * Copyright 2017	Intel Deutschland GmbH
  */
 
@@ -130,7 +130,7 @@ static int cfg80211_conn_scan(struct wireless_dev *wdev)
 
 	rdev->scan_req = request;
 
-	err = rdev_scan(rdev, request);
+	err = cfg80211_scan(rdev);
 	if (!err) {
 		wdev->conn->state = CFG80211_CONN_SCANNING;
 		nl80211_send_scan_start(rdev, wdev);
-- 
cgit v1.2.3-58-ga151


From 10776262264042199bc3137f277ec2564f37907b Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Wed, 5 Jun 2024 14:04:57 +0300
Subject: wifi: iwlwifi: mvm: remove unneeded debugfs entries

These entries are not used, remove them.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240605140327.0c7c520814d5.I19cefb3d81b03a5be94c029cfffd1c8b8c437182@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/coex.h |  69 +------
 drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c | 221 -----------------------
 2 files changed, 1 insertion(+), 289 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/coex.h b/drivers/net/wireless/intel/iwlwifi/fw/api/coex.h
index bc27e15488f5..b97a43353779 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/coex.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/coex.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2023 Intel Corporation
+ * Copyright (C) 2023-2024 Intel Corporation
  * Copyright (C) 2013-2014, 2018-2019 Intel Corporation
  * Copyright (C) 2013-2014 Intel Mobile Communications GmbH
  * Copyright (C) 2017 Intel Deutschland GmbH
@@ -77,73 +77,6 @@ struct iwl_bt_coex_ci_cmd {
 	__le32 secondary_ch_phy_id;
 } __packed; /* BT_CI_MSG_API_S_VER_2 */
 
-#define BT_MBOX(n_dw, _msg, _pos, _nbits)	\
-	BT_MBOX##n_dw##_##_msg##_POS = (_pos),	\
-	BT_MBOX##n_dw##_##_msg = BITS(_nbits) << BT_MBOX##n_dw##_##_msg##_POS
-
-enum iwl_bt_mxbox_dw0 {
-	BT_MBOX(0, LE_SLAVE_LAT, 0, 3),
-	BT_MBOX(0, LE_PROF1, 3, 1),
-	BT_MBOX(0, LE_PROF2, 4, 1),
-	BT_MBOX(0, LE_PROF_OTHER, 5, 1),
-	BT_MBOX(0, CHL_SEQ_N, 8, 4),
-	BT_MBOX(0, INBAND_S, 13, 1),
-	BT_MBOX(0, LE_MIN_RSSI, 16, 4),
-	BT_MBOX(0, LE_SCAN, 20, 1),
-	BT_MBOX(0, LE_ADV, 21, 1),
-	BT_MBOX(0, LE_MAX_TX_POWER, 24, 4),
-	BT_MBOX(0, OPEN_CON_1, 28, 2),
-};
-
-enum iwl_bt_mxbox_dw1 {
-	BT_MBOX(1, BR_MAX_TX_POWER, 0, 4),
-	BT_MBOX(1, IP_SR, 4, 1),
-	BT_MBOX(1, LE_MSTR, 5, 1),
-	BT_MBOX(1, AGGR_TRFC_LD, 8, 6),
-	BT_MBOX(1, MSG_TYPE, 16, 3),
-	BT_MBOX(1, SSN, 19, 2),
-};
-
-enum iwl_bt_mxbox_dw2 {
-	BT_MBOX(2, SNIFF_ACT, 0, 3),
-	BT_MBOX(2, PAG, 3, 1),
-	BT_MBOX(2, INQUIRY, 4, 1),
-	BT_MBOX(2, CONN, 5, 1),
-	BT_MBOX(2, SNIFF_INTERVAL, 8, 5),
-	BT_MBOX(2, DISC, 13, 1),
-	BT_MBOX(2, SCO_TX_ACT, 16, 2),
-	BT_MBOX(2, SCO_RX_ACT, 18, 2),
-	BT_MBOX(2, ESCO_RE_TX, 20, 2),
-	BT_MBOX(2, SCO_DURATION, 24, 6),
-};
-
-enum iwl_bt_mxbox_dw3 {
-	BT_MBOX(3, SCO_STATE, 0, 1),
-	BT_MBOX(3, SNIFF_STATE, 1, 1),
-	BT_MBOX(3, A2DP_STATE, 2, 1),
-	BT_MBOX(3, ACL_STATE, 3, 1),
-	BT_MBOX(3, MSTR_STATE, 4, 1),
-	BT_MBOX(3, OBX_STATE, 5, 1),
-	BT_MBOX(3, A2DP_SRC, 6, 1),
-	BT_MBOX(3, OPEN_CON_2, 8, 2),
-	BT_MBOX(3, TRAFFIC_LOAD, 10, 2),
-	BT_MBOX(3, CHL_SEQN_LSB, 12, 1),
-	BT_MBOX(3, INBAND_P, 13, 1),
-	BT_MBOX(3, MSG_TYPE_2, 16, 3),
-	BT_MBOX(3, SSN_2, 19, 2),
-	BT_MBOX(3, UPDATE_REQUEST, 21, 1),
-};
-
-#define BT_MBOX_MSG(_notif, _num, _field)				     \
-	((le32_to_cpu((_notif)->mbox_msg[(_num)]) & BT_MBOX##_num##_##_field)\
-	>> BT_MBOX##_num##_##_field##_POS)
-
-#define BT_MBOX_PRINT(_num, _field, _end)				    \
-			pos += scnprintf(buf + pos, bufsz - pos,	    \
-					 "\t%s: %d%s",			    \
-					 #_field,			    \
-					 BT_MBOX_MSG(notif, _num, _field),  \
-					 true ? "\n" : ", ")
 enum iwl_bt_activity_grading {
 	BT_OFF			= 0,
 	BT_ON_NO_CONNECTION	= 1,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
index cb164b733cb3..91ca830a7b60 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c
@@ -151,37 +151,6 @@ static ssize_t iwl_dbgfs_tx_flush_write(struct iwl_mvm *mvm, char *buf,
 	return ret;
 }
 
-static ssize_t iwl_dbgfs_sta_drain_write(struct iwl_mvm *mvm, char *buf,
-					 size_t count, loff_t *ppos)
-{
-	struct iwl_mvm_sta *mvmsta;
-	int sta_id, drain, ret;
-
-	if (!iwl_mvm_firmware_running(mvm) ||
-	    mvm->fwrt.cur_fw_img != IWL_UCODE_REGULAR)
-		return -EIO;
-
-	if (sscanf(buf, "%d %d", &sta_id, &drain) != 2)
-		return -EINVAL;
-	if (sta_id < 0 || sta_id >= mvm->fw->ucode_capa.num_stations)
-		return -EINVAL;
-	if (drain < 0 || drain > 1)
-		return -EINVAL;
-
-	mutex_lock(&mvm->mutex);
-
-	mvmsta = iwl_mvm_sta_from_staid_protected(mvm, sta_id);
-
-	if (!mvmsta)
-		ret = -ENOENT;
-	else
-		ret = iwl_mvm_drain_sta(mvm, mvmsta, drain) ? : count;
-
-	mutex_unlock(&mvm->mutex);
-
-	return ret;
-}
-
 static ssize_t iwl_dbgfs_sram_read(struct file *file, char __user *user_buf,
 				   size_t count, loff_t *ppos)
 {
@@ -568,186 +537,6 @@ static ssize_t iwl_dbgfs_disable_power_off_write(struct iwl_mvm *mvm, char *buf,
 	return ret ?: count;
 }
 
-static
-int iwl_mvm_coex_dump_mbox(struct iwl_bt_coex_profile_notif *notif, char *buf,
-			   int pos, int bufsz)
-{
-	pos += scnprintf(buf+pos, bufsz-pos, "MBOX dw0:\n");
-
-	BT_MBOX_PRINT(0, LE_SLAVE_LAT, false);
-	BT_MBOX_PRINT(0, LE_PROF1, false);
-	BT_MBOX_PRINT(0, LE_PROF2, false);
-	BT_MBOX_PRINT(0, LE_PROF_OTHER, false);
-	BT_MBOX_PRINT(0, CHL_SEQ_N, false);
-	BT_MBOX_PRINT(0, INBAND_S, false);
-	BT_MBOX_PRINT(0, LE_MIN_RSSI, false);
-	BT_MBOX_PRINT(0, LE_SCAN, false);
-	BT_MBOX_PRINT(0, LE_ADV, false);
-	BT_MBOX_PRINT(0, LE_MAX_TX_POWER, false);
-	BT_MBOX_PRINT(0, OPEN_CON_1, true);
-
-	pos += scnprintf(buf+pos, bufsz-pos, "MBOX dw1:\n");
-
-	BT_MBOX_PRINT(1, BR_MAX_TX_POWER, false);
-	BT_MBOX_PRINT(1, IP_SR, false);
-	BT_MBOX_PRINT(1, LE_MSTR, false);
-	BT_MBOX_PRINT(1, AGGR_TRFC_LD, false);
-	BT_MBOX_PRINT(1, MSG_TYPE, false);
-	BT_MBOX_PRINT(1, SSN, true);
-
-	pos += scnprintf(buf+pos, bufsz-pos, "MBOX dw2:\n");
-
-	BT_MBOX_PRINT(2, SNIFF_ACT, false);
-	BT_MBOX_PRINT(2, PAG, false);
-	BT_MBOX_PRINT(2, INQUIRY, false);
-	BT_MBOX_PRINT(2, CONN, false);
-	BT_MBOX_PRINT(2, SNIFF_INTERVAL, false);
-	BT_MBOX_PRINT(2, DISC, false);
-	BT_MBOX_PRINT(2, SCO_TX_ACT, false);
-	BT_MBOX_PRINT(2, SCO_RX_ACT, false);
-	BT_MBOX_PRINT(2, ESCO_RE_TX, false);
-	BT_MBOX_PRINT(2, SCO_DURATION, true);
-
-	pos += scnprintf(buf+pos, bufsz-pos, "MBOX dw3:\n");
-
-	BT_MBOX_PRINT(3, SCO_STATE, false);
-	BT_MBOX_PRINT(3, SNIFF_STATE, false);
-	BT_MBOX_PRINT(3, A2DP_STATE, false);
-	BT_MBOX_PRINT(3, A2DP_SRC, false);
-	BT_MBOX_PRINT(3, ACL_STATE, false);
-	BT_MBOX_PRINT(3, MSTR_STATE, false);
-	BT_MBOX_PRINT(3, OBX_STATE, false);
-	BT_MBOX_PRINT(3, OPEN_CON_2, false);
-	BT_MBOX_PRINT(3, TRAFFIC_LOAD, false);
-	BT_MBOX_PRINT(3, CHL_SEQN_LSB, false);
-	BT_MBOX_PRINT(3, INBAND_P, false);
-	BT_MBOX_PRINT(3, MSG_TYPE_2, false);
-	BT_MBOX_PRINT(3, SSN_2, false);
-	BT_MBOX_PRINT(3, UPDATE_REQUEST, true);
-
-	return pos;
-}
-
-static ssize_t iwl_dbgfs_bt_notif_read(struct file *file, char __user *user_buf,
-				       size_t count, loff_t *ppos)
-{
-	struct iwl_mvm *mvm = file->private_data;
-	struct iwl_bt_coex_profile_notif *notif = &mvm->last_bt_notif;
-	char *buf;
-	int ret, pos = 0, bufsz = sizeof(char) * 1024;
-
-	buf = kmalloc(bufsz, GFP_KERNEL);
-	if (!buf)
-		return -ENOMEM;
-
-	mutex_lock(&mvm->mutex);
-
-	pos += iwl_mvm_coex_dump_mbox(notif, buf, pos, bufsz);
-
-	pos += scnprintf(buf + pos, bufsz - pos, "bt_ci_compliance = %d\n",
-			 notif->bt_ci_compliance);
-	pos += scnprintf(buf + pos, bufsz - pos, "primary_ch_lut = %d\n",
-			 le32_to_cpu(notif->primary_ch_lut));
-	pos += scnprintf(buf + pos, bufsz - pos, "secondary_ch_lut = %d\n",
-			 le32_to_cpu(notif->secondary_ch_lut));
-	pos += scnprintf(buf + pos,
-			 bufsz - pos, "bt_activity_grading = %d\n",
-			 le32_to_cpu(notif->bt_activity_grading));
-	pos += scnprintf(buf + pos, bufsz - pos, "bt_rrc = %d\n",
-			 notif->rrc_status & 0xF);
-	pos += scnprintf(buf + pos, bufsz - pos, "bt_ttc = %d\n",
-			 notif->ttc_status & 0xF);
-
-	pos += scnprintf(buf + pos, bufsz - pos, "sync_sco = %d\n",
-			 IWL_MVM_BT_COEX_SYNC2SCO);
-	pos += scnprintf(buf + pos, bufsz - pos, "mplut = %d\n",
-			 IWL_MVM_BT_COEX_MPLUT);
-
-	mutex_unlock(&mvm->mutex);
-
-	ret = simple_read_from_buffer(user_buf, count, ppos, buf, pos);
-	kfree(buf);
-
-	return ret;
-}
-#undef BT_MBOX_PRINT
-
-static ssize_t iwl_dbgfs_bt_cmd_read(struct file *file, char __user *user_buf,
-				     size_t count, loff_t *ppos)
-{
-	struct iwl_mvm *mvm = file->private_data;
-	struct iwl_bt_coex_ci_cmd *cmd = &mvm->last_bt_ci_cmd;
-	char buf[256];
-	int bufsz = sizeof(buf);
-	int pos = 0;
-
-	mutex_lock(&mvm->mutex);
-
-	pos += scnprintf(buf + pos, bufsz - pos, "Channel inhibition CMD\n");
-	pos += scnprintf(buf + pos, bufsz - pos,
-			 "\tPrimary Channel Bitmap 0x%016llx\n",
-			 le64_to_cpu(cmd->bt_primary_ci));
-	pos += scnprintf(buf + pos, bufsz - pos,
-			 "\tSecondary Channel Bitmap 0x%016llx\n",
-			 le64_to_cpu(cmd->bt_secondary_ci));
-
-	mutex_unlock(&mvm->mutex);
-
-	return simple_read_from_buffer(user_buf, count, ppos, buf, pos);
-}
-
-static ssize_t
-iwl_dbgfs_bt_tx_prio_write(struct iwl_mvm *mvm, char *buf,
-			   size_t count, loff_t *ppos)
-{
-	u32 bt_tx_prio;
-
-	if (sscanf(buf, "%u", &bt_tx_prio) != 1)
-		return -EINVAL;
-	if (bt_tx_prio > 4)
-		return -EINVAL;
-
-	mvm->bt_tx_prio = bt_tx_prio;
-
-	return count;
-}
-
-static ssize_t
-iwl_dbgfs_bt_force_ant_write(struct iwl_mvm *mvm, char *buf,
-			     size_t count, loff_t *ppos)
-{
-	static const char * const modes_str[BT_FORCE_ANT_MAX] = {
-		[BT_FORCE_ANT_DIS] = "dis",
-		[BT_FORCE_ANT_AUTO] = "auto",
-		[BT_FORCE_ANT_BT] = "bt",
-		[BT_FORCE_ANT_WIFI] = "wifi",
-	};
-	int ret, bt_force_ant_mode;
-
-	ret = match_string(modes_str, ARRAY_SIZE(modes_str), buf);
-	if (ret < 0)
-		return ret;
-
-	bt_force_ant_mode = ret;
-	ret = 0;
-	mutex_lock(&mvm->mutex);
-	if (mvm->bt_force_ant_mode == bt_force_ant_mode)
-		goto out;
-
-	mvm->bt_force_ant_mode = bt_force_ant_mode;
-	IWL_DEBUG_COEX(mvm, "Force mode: %s\n",
-		       modes_str[mvm->bt_force_ant_mode]);
-
-	if (iwl_mvm_firmware_running(mvm))
-		ret = iwl_mvm_send_bt_init_conf(mvm);
-	else
-		ret = 0;
-
-out:
-	mutex_unlock(&mvm->mutex);
-	return ret ?: count;
-}
-
 static ssize_t iwl_dbgfs_fw_ver_read(struct file *file, char __user *user_buf,
 				     size_t count, loff_t *ppos)
 {
@@ -2165,15 +1954,12 @@ MVM_DEBUGFS_WRITE_FILE_OPS(stop_ctdp, 8);
 MVM_DEBUGFS_WRITE_FILE_OPS(start_ctdp, 8);
 MVM_DEBUGFS_WRITE_FILE_OPS(force_ctkill, 8);
 MVM_DEBUGFS_WRITE_FILE_OPS(tx_flush, 16);
-MVM_DEBUGFS_WRITE_FILE_OPS(sta_drain, 8);
 MVM_DEBUGFS_WRITE_FILE_OPS(send_echo_cmd, 8);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(sram, 64);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(set_nic_temperature, 64);
 MVM_DEBUGFS_READ_FILE_OPS(nic_temp);
 MVM_DEBUGFS_READ_FILE_OPS(stations);
 MVM_DEBUGFS_READ_LINK_STA_FILE_OPS(rs_data);
-MVM_DEBUGFS_READ_FILE_OPS(bt_notif);
-MVM_DEBUGFS_READ_FILE_OPS(bt_cmd);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(disable_power_off, 64);
 MVM_DEBUGFS_READ_FILE_OPS(fw_rx_stats);
 MVM_DEBUGFS_READ_FILE_OPS(drv_rx_stats);
@@ -2183,8 +1969,6 @@ MVM_DEBUGFS_READ_FILE_OPS(phy_integration_ver);
 MVM_DEBUGFS_READ_FILE_OPS(tas_get_status);
 MVM_DEBUGFS_WRITE_FILE_OPS(fw_restart, 10);
 MVM_DEBUGFS_WRITE_FILE_OPS(fw_nmi, 10);
-MVM_DEBUGFS_WRITE_FILE_OPS(bt_tx_prio, 10);
-MVM_DEBUGFS_WRITE_FILE_OPS(bt_force_ant, 10);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(scan_ant_rxchain, 8);
 MVM_DEBUGFS_READ_WRITE_FILE_OPS(fw_dbg_conf, 8);
 MVM_DEBUGFS_WRITE_FILE_OPS(fw_dbg_collect, 64);
@@ -2371,7 +2155,6 @@ void iwl_mvm_dbgfs_register(struct iwl_mvm *mvm)
 	spin_lock_init(&mvm->drv_stats_lock);
 
 	MVM_DEBUGFS_ADD_FILE(tx_flush, mvm->debugfs_dir, 0200);
-	MVM_DEBUGFS_ADD_FILE(sta_drain, mvm->debugfs_dir, 0200);
 	MVM_DEBUGFS_ADD_FILE(sram, mvm->debugfs_dir, 0600);
 	MVM_DEBUGFS_ADD_FILE(set_nic_temperature, mvm->debugfs_dir, 0600);
 	MVM_DEBUGFS_ADD_FILE(nic_temp, mvm->debugfs_dir, 0400);
@@ -2380,8 +2163,6 @@ void iwl_mvm_dbgfs_register(struct iwl_mvm *mvm)
 	MVM_DEBUGFS_ADD_FILE(start_ctdp, mvm->debugfs_dir, 0200);
 	MVM_DEBUGFS_ADD_FILE(force_ctkill, mvm->debugfs_dir, 0200);
 	MVM_DEBUGFS_ADD_FILE(stations, mvm->debugfs_dir, 0400);
-	MVM_DEBUGFS_ADD_FILE(bt_notif, mvm->debugfs_dir, 0400);
-	MVM_DEBUGFS_ADD_FILE(bt_cmd, mvm->debugfs_dir, 0400);
 	MVM_DEBUGFS_ADD_FILE(disable_power_off, mvm->debugfs_dir, 0600);
 	MVM_DEBUGFS_ADD_FILE(fw_ver, mvm->debugfs_dir, 0400);
 	MVM_DEBUGFS_ADD_FILE(fw_rx_stats, mvm->debugfs_dir, 0400);
@@ -2389,8 +2170,6 @@ void iwl_mvm_dbgfs_register(struct iwl_mvm *mvm)
 	MVM_DEBUGFS_ADD_FILE(fw_system_stats, mvm->debugfs_dir, 0400);
 	MVM_DEBUGFS_ADD_FILE(fw_restart, mvm->debugfs_dir, 0200);
 	MVM_DEBUGFS_ADD_FILE(fw_nmi, mvm->debugfs_dir, 0200);
-	MVM_DEBUGFS_ADD_FILE(bt_tx_prio, mvm->debugfs_dir, 0200);
-	MVM_DEBUGFS_ADD_FILE(bt_force_ant, mvm->debugfs_dir, 0200);
 	MVM_DEBUGFS_ADD_FILE(scan_ant_rxchain, mvm->debugfs_dir, 0600);
 	MVM_DEBUGFS_ADD_FILE(prph_reg, mvm->debugfs_dir, 0600);
 	MVM_DEBUGFS_ADD_FILE(fw_dbg_conf, mvm->debugfs_dir, 0600);
-- 
cgit v1.2.3-58-ga151


From 2cbeb1a38758bfd165a7f284933e24b4552da65d Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Wed, 5 Jun 2024 14:04:58 +0300
Subject: wifi: iwlwifi: bump minimum API version in BZ/SC to 90

Stop supporting older FWs.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240605140327.ff8477233010.Ic8c73bd6749cc5f8ab5297807bb0be9bd96a59fa@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/bz.c | 2 +-
 drivers/net/wireless/intel/iwlwifi/cfg/sc.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c
index 45905e57e084..e3d3abfc5fe4 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c
@@ -13,7 +13,7 @@
 #define IWL_BZ_UCODE_API_MAX	91
 
 /* Lowest firmware API version supported */
-#define IWL_BZ_UCODE_API_MIN	80
+#define IWL_BZ_UCODE_API_MIN	90
 
 /* NVM versions */
 #define IWL_BZ_NVM_VERSION		0x0a1d
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
index 653279e08927..7ad3322a3305 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
@@ -13,7 +13,7 @@
 #define IWL_SC_UCODE_API_MAX	91
 
 /* Lowest firmware API version supported */
-#define IWL_SC_UCODE_API_MIN	82
+#define IWL_SC_UCODE_API_MIN	90
 
 /* NVM versions */
 #define IWL_SC_NVM_VERSION		0x0a1d
-- 
cgit v1.2.3-58-ga151


From 62a5c4029552a5645d0600fcb311a14131ccced1 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 5 Jun 2024 14:04:59 +0300
Subject: wifi: iwlwifi: mvm: fix DTIM skip powersave config

When entering D3 we want to configure skip over DTIM, but
it can't use the deflink configuration, that will not even
exist. Adjust the code to handle multiple links by taking
the min skip, even if we should only have a single active
link at this point.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605140327.bccf980fadb4.Idc98b9f3634f39d2fae9bd9916f5d050ccd48f95@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/power.c | 64 +++++++++++++-------------
 1 file changed, 33 insertions(+), 31 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
index cc7176b0be4f..bc363e8427e4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c
@@ -282,62 +282,64 @@ static bool iwl_mvm_power_allow_uapsd(struct iwl_mvm *mvm,
 	return data.allow_uapsd;
 }
 
-static bool iwl_mvm_power_is_radar(struct ieee80211_vif *vif)
+static bool iwl_mvm_power_is_radar(struct ieee80211_bss_conf *link_conf)
 {
 	struct ieee80211_chanctx_conf *chanctx_conf;
-	struct ieee80211_bss_conf *link_conf;
-	bool radar_detect = false;
-	unsigned int link_id;
 
-	rcu_read_lock();
-	for_each_vif_active_link(vif, link_conf, link_id) {
-		chanctx_conf = rcu_dereference(link_conf->chanctx_conf);
-		/* this happens on link switching, just ignore inactive ones */
-		if (!chanctx_conf)
-			continue;
+	chanctx_conf = rcu_dereference(link_conf->chanctx_conf);
 
-		radar_detect = !!(chanctx_conf->def.chan->flags &
-				  IEEE80211_CHAN_RADAR);
-		if (radar_detect)
-			goto out;
-	}
+	/* this happens on link switching, just ignore inactive ones */
+	if (!chanctx_conf)
+		return false;
 
-out:
-	rcu_read_unlock();
-	return radar_detect;
+	return chanctx_conf->def.chan->flags & IEEE80211_CHAN_RADAR;
 }
 
 static void iwl_mvm_power_config_skip_dtim(struct iwl_mvm *mvm,
 					   struct ieee80211_vif *vif,
 					   struct iwl_mac_power_cmd *cmd)
 {
-	int dtimper = vif->bss_conf.dtim_period ?: 1;
-	int skip;
+	struct ieee80211_bss_conf *link_conf;
+	unsigned int min_link_skip = ~0;
+	unsigned int link_id;
 
 	/* disable, in case we're supposed to override */
 	cmd->skip_dtim_periods = 0;
 	cmd->flags &= ~cpu_to_le16(POWER_FLAGS_SKIP_OVER_DTIM_MSK);
 
-	if (iwl_mvm_power_is_radar(vif))
+	if (!test_bit(IWL_MVM_STATUS_IN_D3, &mvm->status)) {
+		if (iwlmvm_mod_params.power_scheme != IWL_POWER_SCHEME_LP)
+			return;
+		cmd->skip_dtim_periods = 2;
+		cmd->flags |= cpu_to_le16(POWER_FLAGS_SKIP_OVER_DTIM_MSK);
 		return;
+	}
 
-	if (dtimper >= 10)
-		return;
+	rcu_read_lock();
+	for_each_vif_active_link(vif, link_conf, link_id) {
+		unsigned int dtimper = link_conf->dtim_period ?: 1;
+		unsigned int dtimper_tu = dtimper * link_conf->beacon_int;
+		unsigned int skip;
 
-	if (!test_bit(IWL_MVM_STATUS_IN_D3, &mvm->status)) {
-		if (iwlmvm_mod_params.power_scheme != IWL_POWER_SCHEME_LP)
+		if (dtimper >= 10 || iwl_mvm_power_is_radar(link_conf)) {
+			rcu_read_unlock();
 			return;
-		skip = 2;
-	} else {
-		int dtimper_tu = dtimper * vif->bss_conf.beacon_int;
+		}
 
 		if (WARN_ON(!dtimper_tu))
-			return;
+			continue;
+
 		/* configure skip over dtim up to 900 TU DTIM interval */
-		skip = max_t(u8, 1, 900 / dtimper_tu);
+		skip = max_t(int, 1, 900 / dtimper_tu);
+		min_link_skip = min(min_link_skip, skip);
 	}
+	rcu_read_unlock();
+
+	/* no WARN_ON, can only happen with WARN_ON above */
+	if (min_link_skip == ~0)
+		return;
 
-	cmd->skip_dtim_periods = skip;
+	cmd->skip_dtim_periods = min_link_skip;
 	cmd->flags |= cpu_to_le16(POWER_FLAGS_SKIP_OVER_DTIM_MSK);
 }
 
-- 
cgit v1.2.3-58-ga151


From 57bb72fad7bcb9ed0b2d7ec1b73a116ac3c6a2ba Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 5 Jun 2024 14:05:00 +0300
Subject: wifi: iwlwifi: move TXQ bytecount limit to queue code

This really isn't correct to be in the opmode, do the clamping
(and power-of-2 fixup that may be necessary due to this, or even
otherwise) in the queue code. Also move down the retrying of the
allocation, it should be after all the size fixups, but also it
just makes sense, and avoids retrying same-size allocations in
the case of the BZ-family A-step workaround.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605140327.000a0a1e807d.Ib822590d5aca76ff3168418ae2c139b3d43d81ed@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c  | 27 ++++++---------------------
 drivers/net/wireless/intel/iwlwifi/queue/tx.c | 22 ++++++++++++++++++++--
 2 files changed, 26 insertions(+), 23 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index 2a8ad718ce78..c57755e5ff5e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -857,12 +857,6 @@ int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
 		size = iwl_mvm_get_queue_size(sta);
 	}
 
-	/* take the min with bc tbl entries allowed */
-	size = min_t(u32, size, mvm->trans->txqs.bc_tbl_size / sizeof(u16));
-
-	/* size needs to be power of 2 values for calculating read/write pointers */
-	size = rounddown_pow_of_two(size);
-
 	if (sta) {
 		struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta);
 		struct ieee80211_link_sta *link_sta;
@@ -887,22 +881,13 @@ int iwl_mvm_tvqm_enable_txq(struct iwl_mvm *mvm,
 	if (!sta_mask)
 		return -EINVAL;
 
-	do {
-		queue = iwl_trans_txq_alloc(mvm->trans, 0, sta_mask,
-					    tid, size, timeout);
-
-		if (queue < 0)
-			IWL_DEBUG_TX_QUEUES(mvm,
-					    "Failed allocating TXQ of size %d for sta mask %x tid %d, ret: %d\n",
-					    size, sta_mask, tid, queue);
-		size /= 2;
-	} while (queue < 0 && size >= 16);
-
-	if (queue < 0)
-		return queue;
+	queue = iwl_trans_txq_alloc(mvm->trans, 0, sta_mask,
+				    tid, size, timeout);
 
-	IWL_DEBUG_TX_QUEUES(mvm, "Enabling TXQ #%d for sta mask 0x%x tid %d\n",
-			    queue, sta_mask, tid);
+	if (queue >= 0)
+		IWL_DEBUG_TX_QUEUES(mvm,
+				    "Enabling TXQ #%d for sta mask 0x%x tid %d\n",
+				    queue, sta_mask, tid);
 
 	return queue;
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/queue/tx.c b/drivers/net/wireless/intel/iwlwifi/queue/tx.c
index 6229c785c845..726035dfa5cd 100644
--- a/drivers/net/wireless/intel/iwlwifi/queue/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/queue/tx.c
@@ -1237,11 +1237,29 @@ int iwl_txq_dyn_alloc(struct iwl_trans *trans, u32 flags, u32 sta_mask,
 	};
 	int ret;
 
+	/* take the min with bytecount table entries allowed */
+	size = min_t(u32, size, trans->txqs.bc_tbl_size / sizeof(u16));
+	/* but must be power of 2 values for calculating read/write pointers */
+	size = rounddown_pow_of_two(size);
+
 	if (trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_BZ &&
-	    trans->hw_rev_step == SILICON_A_STEP)
+	    trans->hw_rev_step == SILICON_A_STEP) {
 		size = 4096;
+		txq = iwl_txq_dyn_alloc_dma(trans, size, timeout);
+	} else {
+		do {
+			txq = iwl_txq_dyn_alloc_dma(trans, size, timeout);
+			if (!IS_ERR(txq))
+				break;
+
+			IWL_DEBUG_TX_QUEUES(trans,
+					    "Failed allocating TXQ of size %d for sta mask %x tid %d, ret: %ld\n",
+					    size, sta_mask, tid,
+					    PTR_ERR(txq));
+			size /= 2;
+		} while (size >= 16);
+	}
 
-	txq = iwl_txq_dyn_alloc_dma(trans, size, timeout);
 	if (IS_ERR(txq))
 		return PTR_ERR(txq);
 
-- 
cgit v1.2.3-58-ga151


From dc8f854fec78c35316e9616c993a7b3586cc30ed Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 5 Jun 2024 14:05:01 +0300
Subject: wifi: iwlwifi: api: fix includes in debug.h

This needs to include dbg-tlv.h since it uses the value of
IWL_FW_INI_ALLOCATION_NUM from that file.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605140327.2d25691283eb.I0909621a0e293a8a21d4f1de6e5fd59c22e4b212@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/debug.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h b/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h
index b31ae6889bd0..bea0f4668cc8 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/debug.h
@@ -1,11 +1,12 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2005-2014, 2018-2023 Intel Corporation
+ * Copyright (C) 2005-2014, 2018-2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
 #ifndef __iwl_fw_api_debug_h__
 #define __iwl_fw_api_debug_h__
+#include "dbg-tlv.h"
 
 /**
  * enum iwl_debug_cmds - debug commands
-- 
cgit v1.2.3-58-ga151


From 49101078be769e9c50a8bfbc8caaab56db9d12d9 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 5 Jun 2024 14:05:02 +0300
Subject: wifi: iwlwifi: pcie: integrate TX queue code

The TX queue code was mostly moved out to support an internal
transport that we were never going to publish, but we're no
longer using that. Since we're also going to be dissolving
the virtual transport layer entirely, integrate the TX queue
code into the PCIe layer.

This also has a small kernel of already removing the virtual
transport function layer, since iwl_trans_send_cmd() calls
iwl_trans_pcie_send_hcmd() directly now, even if that still
calls the transport send_cmd method for now, we'll clean it
up later.

Also, not everything is renamed yet.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605140327.936b13f45071.Ib219ce01a1e67bcad79d5131626db950252aaa46@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/Makefile        |    1 -
 drivers/net/wireless/intel/iwlwifi/iwl-trans.c     |   61 +-
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h     |   71 +-
 .../wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c   |    2 +-
 .../net/wireless/intel/iwlwifi/pcie/ctxt-info.c    |    4 +-
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h |  211 ++-
 drivers/net/wireless/intel/iwlwifi/pcie/rx.c       |    9 +-
 .../net/wireless/intel/iwlwifi/pcie/trans-gen2.c   |   11 +-
 drivers/net/wireless/intel/iwlwifi/pcie/trans.c    |  114 +-
 drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c  | 1130 +++++++++++-
 drivers/net/wireless/intel/iwlwifi/pcie/tx.c       |  991 +++++++++-
 drivers/net/wireless/intel/iwlwifi/queue/tx.c      | 1918 --------------------
 drivers/net/wireless/intel/iwlwifi/queue/tx.h      |  191 --
 13 files changed, 2366 insertions(+), 2348 deletions(-)
 delete mode 100644 drivers/net/wireless/intel/iwlwifi/queue/tx.c
 delete mode 100644 drivers/net/wireless/intel/iwlwifi/queue/tx.h

diff --git a/drivers/net/wireless/intel/iwlwifi/Makefile b/drivers/net/wireless/intel/iwlwifi/Makefile
index 93cce88f2e27..64c123314245 100644
--- a/drivers/net/wireless/intel/iwlwifi/Makefile
+++ b/drivers/net/wireless/intel/iwlwifi/Makefile
@@ -14,7 +14,6 @@ iwlwifi-$(CONFIG_IWLMVM) += cfg/7000.o cfg/8000.o cfg/9000.o cfg/22000.o
 iwlwifi-$(CONFIG_IWLMVM) += cfg/ax210.o cfg/bz.o cfg/sc.o
 iwlwifi-objs		+= iwl-dbg-tlv.o
 iwlwifi-objs		+= iwl-trans.o
-iwlwifi-objs		+= queue/tx.o
 
 iwlwifi-objs		+= fw/img.o fw/notif-wait.o fw/rs.o
 iwlwifi-objs		+= fw/dbg.o fw/pnvm.o fw/dump.o
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.c b/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
index f95098c21c7d..327db93e8d84 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
@@ -2,7 +2,7 @@
 /*
  * Copyright (C) 2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
- * Copyright (C) 2019-2021, 2023 Intel Corporation
+ * Copyright (C) 2019-2021, 2023-2024 Intel Corporation
  */
 #include <linux/kernel.h>
 #include <linux/bsearch.h>
@@ -11,7 +11,6 @@
 #include "iwl-trans.h"
 #include "iwl-drv.h"
 #include "iwl-fh.h"
-#include "queue/tx.h"
 #include <linux/dmapool.h>
 #include "fw/api/commands.h"
 
@@ -42,17 +41,6 @@ struct iwl_trans *iwl_trans_alloc(unsigned int priv_size,
 
 	WARN_ON(!ops->wait_txq_empty && !ops->wait_tx_queues_empty);
 
-	if (trans->trans_cfg->gen2) {
-		trans->txqs.tfd.addr_size = 64;
-		trans->txqs.tfd.max_tbs = IWL_TFH_NUM_TBS;
-		trans->txqs.tfd.size = sizeof(struct iwl_tfh_tfd);
-	} else {
-		trans->txqs.tfd.addr_size = 36;
-		trans->txqs.tfd.max_tbs = IWL_NUM_OF_TBS;
-		trans->txqs.tfd.size = sizeof(struct iwl_tfd);
-	}
-	trans->max_skb_frags = IWL_TRANS_MAX_FRAGS(trans);
-
 	return trans;
 }
 
@@ -78,31 +66,6 @@ int iwl_trans_init(struct iwl_trans *trans)
 	if (WARN_ON(trans->trans_cfg->gen2 && txcmd_size >= txcmd_align))
 		return -EINVAL;
 
-	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ)
-		trans->txqs.bc_tbl_size =
-			sizeof(struct iwl_gen3_bc_tbl_entry) * TFD_QUEUE_BC_SIZE_GEN3_BZ;
-	else if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_AX210)
-		trans->txqs.bc_tbl_size =
-			sizeof(struct iwl_gen3_bc_tbl_entry) * TFD_QUEUE_BC_SIZE_GEN3_AX210;
-	else
-		trans->txqs.bc_tbl_size = sizeof(struct iwlagn_scd_bc_tbl);
-	/*
-	 * For gen2 devices, we use a single allocation for each byte-count
-	 * table, but they're pretty small (1k) so use a DMA pool that we
-	 * allocate here.
-	 */
-	if (trans->trans_cfg->gen2) {
-		trans->txqs.bc_pool = dmam_pool_create("iwlwifi:bc", trans->dev,
-						       trans->txqs.bc_tbl_size,
-						       256, 0);
-		if (!trans->txqs.bc_pool)
-			return -ENOMEM;
-	}
-
-	/* Some things must not change even if the config does */
-	WARN_ON(trans->txqs.tfd.addr_size !=
-		(trans->trans_cfg->gen2 ? 64 : 36));
-
 	snprintf(trans->dev_cmd_pool_name, sizeof(trans->dev_cmd_pool_name),
 		 "iwl_cmd_pool:%s", dev_name(trans->dev));
 	trans->dev_cmd_pool =
@@ -112,12 +75,6 @@ int iwl_trans_init(struct iwl_trans *trans)
 	if (!trans->dev_cmd_pool)
 		return -ENOMEM;
 
-	trans->txqs.tso_hdr_page = alloc_percpu(struct iwl_tso_hdr_page);
-	if (!trans->txqs.tso_hdr_page) {
-		kmem_cache_destroy(trans->dev_cmd_pool);
-		return -ENOMEM;
-	}
-
 	/* Initialize the wait queue for commands */
 	init_waitqueue_head(&trans->wait_command_queue);
 
@@ -126,20 +83,6 @@ int iwl_trans_init(struct iwl_trans *trans)
 
 void iwl_trans_free(struct iwl_trans *trans)
 {
-	int i;
-
-	if (trans->txqs.tso_hdr_page) {
-		for_each_possible_cpu(i) {
-			struct iwl_tso_hdr_page *p =
-				per_cpu_ptr(trans->txqs.tso_hdr_page, i);
-
-			if (p && p->page)
-				__free_page(p->page);
-		}
-
-		free_percpu(trans->txqs.tso_hdr_page);
-	}
-
 	kmem_cache_destroy(trans->dev_cmd_pool);
 }
 
@@ -180,7 +123,7 @@ int iwl_trans_send_cmd(struct iwl_trans *trans, struct iwl_host_cmd *cmd)
 			cmd->id = DEF_ID(cmd->id);
 	}
 
-	ret = iwl_trans_txq_send_hcmd(trans, cmd);
+	ret = iwl_trans_pcie_send_hcmd(trans, cmd);
 
 	if (!(cmd->flags & CMD_ASYNC))
 		lock_map_release(&trans->sync_cmd_lockdep_map);
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index b93cef7b2330..27f92b393f8e 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -26,11 +26,9 @@
  * DOC: Transport layer - what is it ?
  *
  * The transport layer is the layer that deals with the HW directly. It provides
- * an abstraction of the underlying HW to the upper layer. The transport layer
- * doesn't provide any policy, algorithm or anything of this kind, but only
- * mechanisms to make the HW do something. It is not completely stateless but
- * close to it.
- * We will have an implementation for each different supported bus.
+ * the PCIe access to the underlying hardwarwe. The transport layer doesn't
+ * provide any policy, algorithm or anything of this kind, but only mechanisms
+ * to make the HW do something. It is not completely stateless but close to it.
  */
 
 /**
@@ -167,12 +165,6 @@ struct iwl_device_tx_cmd {
  */
 #define IWL_MAX_CMD_TBS_PER_TFD	2
 
-/* We need 2 entries for the TX command and header, and another one might
- * be needed for potential data in the SKB's head. The remaining ones can
- * be used for frags.
- */
-#define IWL_TRANS_MAX_FRAGS(trans) ((trans)->txqs.tfd.max_tbs - 3)
-
 /**
  * enum iwl_hcmd_dataflag - flag for each one of the chunks of the command
  *
@@ -998,54 +990,6 @@ struct iwl_txq {
 	bool overflow_tx;
 };
 
-/**
- * struct iwl_trans_txqs - transport tx queues data
- *
- * @bc_table_dword: true if the BC table expects DWORD (as opposed to bytes)
- * @page_offs: offset from skb->cb to mac header page pointer
- * @dev_cmd_offs: offset from skb->cb to iwl_device_tx_cmd pointer
- * @queue_used: bit mask of used queues
- * @queue_stopped: bit mask of stopped queues
- * @txq: array of TXQ data structures representing the TXQs
- * @scd_bc_tbls: gen1 pointer to the byte count table of the scheduler
- * @queue_alloc_cmd_ver: queue allocation command version
- * @bc_pool: bytecount DMA allocations pool
- * @bc_tbl_size: bytecount table size
- * @tso_hdr_page: page allocated (per CPU) for A-MSDU headers when doing TSO
- *	(and similar usage)
- * @tfd: TFD data
- * @tfd.max_tbs: max number of buffers per TFD
- * @tfd.size: TFD size
- * @tfd.addr_size: TFD/TB address size
- */
-struct iwl_trans_txqs {
-	unsigned long queue_used[BITS_TO_LONGS(IWL_MAX_TVQM_QUEUES)];
-	unsigned long queue_stopped[BITS_TO_LONGS(IWL_MAX_TVQM_QUEUES)];
-	struct iwl_txq *txq[IWL_MAX_TVQM_QUEUES];
-	struct dma_pool *bc_pool;
-	size_t bc_tbl_size;
-	bool bc_table_dword;
-	u8 page_offs;
-	u8 dev_cmd_offs;
-	struct iwl_tso_hdr_page __percpu *tso_hdr_page;
-
-	struct {
-		u8 fifo;
-		u8 q_id;
-		unsigned int wdg_timeout;
-	} cmd;
-
-	struct {
-		u8 max_tbs;
-		u16 size;
-		u8 addr_size;
-	} tfd;
-
-	struct iwl_dma_ptr scd_bc_tbls;
-
-	u8 queue_alloc_cmd_ver;
-};
-
 /**
  * struct iwl_trans - transport common data
  *
@@ -1099,7 +1043,6 @@ struct iwl_trans_txqs {
  *	This mode is set dynamically, depending on the WoWLAN values
  *	configured from the userspace at runtime.
  * @name: the device name
- * @txqs: transport tx queues data.
  * @mbx_addr_0_step: step address data 0
  * @mbx_addr_1_step: step address data 1
  * @pcie_link_speed: current PCIe link speed (%PCI_EXP_LNKSTA_CLS_*),
@@ -1169,7 +1112,6 @@ struct iwl_trans {
 	enum iwl_plat_pm_mode system_pm_mode;
 
 	const char *name;
-	struct iwl_trans_txqs txqs;
 	u32 mbx_addr_0_step;
 	u32 mbx_addr_1_step;
 
@@ -1656,10 +1598,13 @@ static inline bool iwl_trans_is_hw_error_value(u32 val)
 }
 
 /*****************************************************
-* driver (transport) register/unregister functions
-******************************************************/
+ * PCIe handling
+ *****************************************************/
 int __must_check iwl_pci_register_driver(void);
 void iwl_pci_unregister_driver(void);
 void iwl_trans_pcie_remove(struct iwl_trans *trans, bool rescan);
 
+int iwl_trans_pcie_send_hcmd(struct iwl_trans *trans,
+			     struct iwl_host_cmd *cmd);
+
 #endif /* __iwl_trans_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c b/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c
index ebf11f276b20..e63efbf809f0 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c
@@ -216,7 +216,7 @@ int iwl_pcie_ctxt_info_gen3_init(struct iwl_trans *trans,
 	ctxt_info_gen3->cr_tail_idx_arr_base_addr =
 		cpu_to_le64(trans_pcie->prph_info_dma_addr + 3 * PAGE_SIZE / 4);
 	ctxt_info_gen3->mtr_base_addr =
-		cpu_to_le64(trans->txqs.txq[trans->txqs.cmd.q_id]->dma_addr);
+		cpu_to_le64(trans_pcie->txqs.txq[trans_pcie->txqs.cmd.q_id]->dma_addr);
 	ctxt_info_gen3->mcr_base_addr =
 		cpu_to_le64(trans_pcie->rxq->used_bd_dma);
 	ctxt_info_gen3->mtr_size =
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info.c b/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info.c
index 0fa92704cd14..344e4d5a1c6e 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
  * Copyright (C) 2017 Intel Deutschland GmbH
- * Copyright (C) 2018-2023 Intel Corporation
+ * Copyright (C) 2018-2024 Intel Corporation
  */
 #include "iwl-trans.h"
 #include "iwl-fh.h"
@@ -218,7 +218,7 @@ int iwl_pcie_ctxt_info_init(struct iwl_trans *trans,
 
 	/* initialize TX command queue */
 	ctxt_info->hcmd_cfg.cmd_queue_addr =
-		cpu_to_le64(trans->txqs.txq[trans->txqs.cmd.q_id]->dma_addr);
+		cpu_to_le64(trans_pcie->txqs.txq[trans_pcie->txqs.cmd.q_id]->dma_addr);
 	ctxt_info->hcmd_cfg.cmd_queue_size =
 		TFD_QUEUE_CB_SIZE(IWL_CMD_QUEUE_SIZE);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index a7eebe400b5b..3bdbb907d91a 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2003-2015, 2018-2023 Intel Corporation
+ * Copyright (C) 2003-2015, 2018-2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -22,7 +22,6 @@
 #include "iwl-io.h"
 #include "iwl-op-mode.h"
 #include "iwl-drv.h"
-#include "queue/tx.h"
 #include "iwl-context-info.h"
 
 /*
@@ -286,6 +285,54 @@ enum iwl_pcie_imr_status {
 	IMR_D2S_ERROR,
 };
 
+/**
+ * struct iwl_pcie_txqs - TX queues data
+ *
+ * @bc_table_dword: true if the BC table expects DWORD (as opposed to bytes)
+ * @page_offs: offset from skb->cb to mac header page pointer
+ * @dev_cmd_offs: offset from skb->cb to iwl_device_tx_cmd pointer
+ * @queue_used: bit mask of used queues
+ * @queue_stopped: bit mask of stopped queues
+ * @txq: array of TXQ data structures representing the TXQs
+ * @scd_bc_tbls: gen1 pointer to the byte count table of the scheduler
+ * @queue_alloc_cmd_ver: queue allocation command version
+ * @bc_pool: bytecount DMA allocations pool
+ * @bc_tbl_size: bytecount table size
+ * @tso_hdr_page: page allocated (per CPU) for A-MSDU headers when doing TSO
+ *	(and similar usage)
+ * @tfd: TFD data
+ * @tfd.max_tbs: max number of buffers per TFD
+ * @tfd.size: TFD size
+ * @tfd.addr_size: TFD/TB address size
+ */
+struct iwl_pcie_txqs {
+	unsigned long queue_used[BITS_TO_LONGS(IWL_MAX_TVQM_QUEUES)];
+	unsigned long queue_stopped[BITS_TO_LONGS(IWL_MAX_TVQM_QUEUES)];
+	struct iwl_txq *txq[IWL_MAX_TVQM_QUEUES];
+	struct dma_pool *bc_pool;
+	size_t bc_tbl_size;
+	bool bc_table_dword;
+	u8 page_offs;
+	u8 dev_cmd_offs;
+	struct iwl_tso_hdr_page __percpu *tso_hdr_page;
+
+	struct {
+		u8 fifo;
+		u8 q_id;
+		unsigned int wdg_timeout;
+	} cmd;
+
+	struct {
+		u8 max_tbs;
+		u16 size;
+		u8 addr_size;
+	} tfd;
+
+	struct iwl_dma_ptr scd_bc_tbls;
+
+	u8 queue_alloc_cmd_ver;
+};
+
 /**
  * struct iwl_trans_pcie - PCIe transport specific data
  * @rxq: all the RX queue data
@@ -367,6 +414,7 @@ enum iwl_pcie_imr_status {
  * @is_down: indicates the NIC is down
  * @isr_stats: interrupt statistics
  * @napi_dev: (fake) netdev for NAPI registration
+ * @txqs: transport tx queues data.
  */
 struct iwl_trans_pcie {
 	struct iwl_rxq *rxq;
@@ -464,6 +512,8 @@ struct iwl_trans_pcie {
 	enum iwl_pcie_imr_status imr_status;
 	wait_queue_head_t imr_waitq;
 	char rf_name[32];
+
+	struct iwl_pcie_txqs txqs;
 };
 
 static inline struct iwl_trans_pcie *
@@ -538,6 +588,17 @@ void iwl_pcie_disable_ict(struct iwl_trans *trans);
 /*****************************************************
 * TX / HCMD
 ******************************************************/
+/* We need 2 entries for the TX command and header, and another one might
+ * be needed for potential data in the SKB's head. The remaining ones can
+ * be used for frags.
+ */
+#define IWL_TRANS_PCIE_MAX_FRAGS(trans_pcie) ((trans_pcie)->txqs.tfd.max_tbs - 3)
+
+struct iwl_tso_hdr_page {
+	struct page *page;
+	u8 *pos;
+};
+
 int iwl_pcie_tx_init(struct iwl_trans *trans);
 void iwl_pcie_tx_start(struct iwl_trans *trans, u32 scd_base_addr);
 int iwl_pcie_tx_stop(struct iwl_trans *trans);
@@ -552,10 +613,154 @@ void iwl_trans_pcie_txq_set_shared_mode(struct iwl_trans *trans, u32 txq_id,
 int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 		      struct iwl_device_tx_cmd *dev_cmd, int txq_id);
 void iwl_pcie_txq_check_wrptrs(struct iwl_trans *trans);
-int iwl_trans_pcie_send_hcmd(struct iwl_trans *trans, struct iwl_host_cmd *cmd);
 void iwl_pcie_hcmd_complete(struct iwl_trans *trans,
 			    struct iwl_rx_cmd_buffer *rxb);
 void iwl_trans_pcie_tx_reset(struct iwl_trans *trans);
+int iwl_pcie_txq_alloc(struct iwl_trans *trans, struct iwl_txq *txq,
+		       int slots_num, bool cmd_queue);
+
+struct iwl_tso_hdr_page *iwl_pcie_get_page_hdr(struct iwl_trans *trans,
+					       size_t len, struct sk_buff *skb);
+void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb);
+
+static inline dma_addr_t
+iwl_txq_get_first_tb_dma(struct iwl_txq *txq, int idx)
+{
+	return txq->first_tb_dma +
+	       sizeof(struct iwl_pcie_first_tb_buf) * idx;
+}
+
+static inline u16 iwl_txq_get_cmd_index(const struct iwl_txq *q, u32 index)
+{
+	return index & (q->n_window - 1);
+}
+
+static inline void *iwl_txq_get_tfd(struct iwl_trans *trans,
+				    struct iwl_txq *txq, int idx)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+
+	if (trans->trans_cfg->gen2)
+		idx = iwl_txq_get_cmd_index(txq, idx);
+
+	return (u8 *)txq->tfds + trans_pcie->txqs.tfd.size * idx;
+}
+
+/*
+ * We need this inline in case dma_addr_t is only 32-bits - since the
+ * hardware is always 64-bit, the issue can still occur in that case,
+ * so use u64 for 'phys' here to force the addition in 64-bit.
+ */
+static inline bool iwl_txq_crosses_4g_boundary(u64 phys, u16 len)
+{
+	return upper_32_bits(phys) != upper_32_bits(phys + len);
+}
+
+int iwl_txq_space(struct iwl_trans *trans, const struct iwl_txq *q);
+
+static inline void iwl_txq_stop(struct iwl_trans *trans, struct iwl_txq *txq)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+
+	if (!test_and_set_bit(txq->id, trans_pcie->txqs.queue_stopped)) {
+		iwl_op_mode_queue_full(trans->op_mode, txq->id);
+		IWL_DEBUG_TX_QUEUES(trans, "Stop hwq %d\n", txq->id);
+	} else {
+		IWL_DEBUG_TX_QUEUES(trans, "hwq %d already stopped\n",
+				    txq->id);
+	}
+}
+
+/**
+ * iwl_txq_inc_wrap - increment queue index, wrap back to beginning
+ * @trans: the transport (for configuration data)
+ * @index: current index
+ */
+static inline int iwl_txq_inc_wrap(struct iwl_trans *trans, int index)
+{
+	return ++index &
+		(trans->trans_cfg->base_params->max_tfd_queue_size - 1);
+}
+
+/**
+ * iwl_txq_dec_wrap - decrement queue index, wrap back to end
+ * @trans: the transport (for configuration data)
+ * @index: current index
+ */
+static inline int iwl_txq_dec_wrap(struct iwl_trans *trans, int index)
+{
+	return --index &
+		(trans->trans_cfg->base_params->max_tfd_queue_size - 1);
+}
+
+void iwl_txq_log_scd_error(struct iwl_trans *trans, struct iwl_txq *txq);
+
+static inline void
+iwl_trans_pcie_wake_queue(struct iwl_trans *trans, struct iwl_txq *txq)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+
+	if (test_and_clear_bit(txq->id, trans_pcie->txqs.queue_stopped)) {
+		IWL_DEBUG_TX_QUEUES(trans, "Wake hwq %d\n", txq->id);
+		iwl_op_mode_queue_not_full(trans->op_mode, txq->id);
+	}
+}
+
+int iwl_txq_gen2_set_tb(struct iwl_trans *trans,
+			struct iwl_tfh_tfd *tfd, dma_addr_t addr,
+			u16 len);
+
+static inline void iwl_txq_set_tfd_invalid_gen2(struct iwl_trans *trans,
+						struct iwl_tfh_tfd *tfd)
+{
+	tfd->num_tbs = 0;
+
+	iwl_txq_gen2_set_tb(trans, tfd, trans->invalid_tx_cmd.dma,
+			    trans->invalid_tx_cmd.size);
+}
+
+void iwl_txq_gen2_tfd_unmap(struct iwl_trans *trans,
+			    struct iwl_cmd_meta *meta,
+			    struct iwl_tfh_tfd *tfd);
+
+int iwl_txq_dyn_alloc(struct iwl_trans *trans, u32 flags,
+		      u32 sta_mask, u8 tid,
+		      int size, unsigned int timeout);
+
+int iwl_txq_gen2_tx(struct iwl_trans *trans, struct sk_buff *skb,
+		    struct iwl_device_tx_cmd *dev_cmd, int txq_id);
+
+void iwl_txq_dyn_free(struct iwl_trans *trans, int queue);
+void iwl_txq_gen2_tx_free(struct iwl_trans *trans);
+int iwl_txq_init(struct iwl_trans *trans, struct iwl_txq *txq,
+		 int slots_num, bool cmd_queue);
+int iwl_txq_gen2_init(struct iwl_trans *trans, int txq_id,
+		      int queue_size);
+
+static inline u16 iwl_txq_gen1_tfd_tb_get_len(struct iwl_trans *trans,
+					      void *_tfd, u8 idx)
+{
+	struct iwl_tfd *tfd;
+	struct iwl_tfd_tb *tb;
+
+	if (trans->trans_cfg->gen2) {
+		struct iwl_tfh_tfd *tfh_tfd = _tfd;
+		struct iwl_tfh_tb *tfh_tb = &tfh_tfd->tbs[idx];
+
+		return le16_to_cpu(tfh_tb->tb_len);
+	}
+
+	tfd = (struct iwl_tfd *)_tfd;
+	tb = &tfd->tbs[idx];
+
+	return le16_to_cpu(tb->hi_n_len) >> 4;
+}
+
+void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
+		      struct sk_buff_head *skbs, bool is_flush);
+void iwl_pcie_set_q_ptrs(struct iwl_trans *trans, int txq_id, int ptr);
+void iwl_pcie_freeze_txq_timer(struct iwl_trans *trans,
+			       unsigned long txqs, bool freeze);
 
 /*****************************************************
 * Error handling
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index 984d7bcd381f..afb88eab8174 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2003-2014, 2018-2023 Intel Corporation
+ * Copyright (C) 2003-2014, 2018-2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -1301,7 +1301,7 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans,
 				int i)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_txq *txq = trans->txqs.txq[trans->txqs.cmd.q_id];
+	struct iwl_txq *txq = trans_pcie->txqs.txq[trans_pcie->txqs.cmd.q_id];
 	bool page_stolen = false;
 	int max_len = trans_pcie->rx_buf_bytes;
 	u32 offset = 0;
@@ -1678,6 +1678,7 @@ irqreturn_t iwl_pcie_irq_rx_msix_handler(int irq, void *dev_id)
  */
 static void iwl_pcie_irq_handle_error(struct iwl_trans *trans)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int i;
 
 	/* W/A for WiFi/WiMAX coex and WiMAX own the RF */
@@ -1694,9 +1695,9 @@ static void iwl_pcie_irq_handle_error(struct iwl_trans *trans)
 	}
 
 	for (i = 0; i < trans->trans_cfg->base_params->num_of_queues; i++) {
-		if (!trans->txqs.txq[i])
+		if (!trans_pcie->txqs.txq[i])
 			continue;
-		del_timer(&trans->txqs.txq[i]->stuck_timer);
+		del_timer(&trans_pcie->txqs.txq[i]->stuck_timer);
 	}
 
 	/* The STATUS_FW_ERROR bit is set in this function. This must happen
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
index a4a4772330cf..c1446d550b4f 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
  * Copyright (C) 2017 Intel Deutschland GmbH
- * Copyright (C) 2018-2023 Intel Corporation
+ * Copyright (C) 2018-2024 Intel Corporation
  */
 #include "iwl-trans.h"
 #include "iwl-prph.h"
@@ -247,7 +247,7 @@ static int iwl_pcie_gen2_nic_init(struct iwl_trans *trans)
 		return -ENOMEM;
 
 	/* Allocate or reset and init all Tx and Command queues */
-	if (iwl_txq_gen2_init(trans, trans->txqs.cmd.q_id, queue_size))
+	if (iwl_txq_gen2_init(trans, trans_pcie->txqs.cmd.q_id, queue_size))
 		return -ENOMEM;
 
 	/* enable shadow regs in HW */
@@ -346,9 +346,10 @@ void iwl_trans_pcie_gen2_fw_alive(struct iwl_trans *trans, u32 scd_addr)
 	iwl_pcie_reset_ict(trans);
 
 	/* make sure all queue are not stopped/used */
-	memset(trans->txqs.queue_stopped, 0,
-	       sizeof(trans->txqs.queue_stopped));
-	memset(trans->txqs.queue_used, 0, sizeof(trans->txqs.queue_used));
+	memset(trans_pcie->txqs.queue_stopped, 0,
+	       sizeof(trans_pcie->txqs.queue_stopped));
+	memset(trans_pcie->txqs.queue_used, 0,
+	       sizeof(trans_pcie->txqs.queue_used));
 
 	/* now that we got alive we can free the fw image & the context info.
 	 * paging memory cannot be freed included since FW will still use it
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index d5a887b3a4bb..48ae38f3120d 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2007-2015, 2018-2023 Intel Corporation
+ * Copyright (C) 2007-2015, 2018-2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -1956,12 +1956,12 @@ static void iwl_trans_pcie_configure(struct iwl_trans *trans,
 	/* free all first - we might be reconfigured for a different size */
 	iwl_pcie_free_rbs_pool(trans);
 
-	trans->txqs.cmd.q_id = trans_cfg->cmd_queue;
-	trans->txqs.cmd.fifo = trans_cfg->cmd_fifo;
-	trans->txqs.cmd.wdg_timeout = trans_cfg->cmd_q_wdg_timeout;
-	trans->txqs.page_offs = trans_cfg->cb_data_offs;
-	trans->txqs.dev_cmd_offs = trans_cfg->cb_data_offs + sizeof(void *);
-	trans->txqs.queue_alloc_cmd_ver = trans_cfg->queue_alloc_cmd_ver;
+	trans_pcie->txqs.cmd.q_id = trans_cfg->cmd_queue;
+	trans_pcie->txqs.cmd.fifo = trans_cfg->cmd_fifo;
+	trans_pcie->txqs.cmd.wdg_timeout = trans_cfg->cmd_q_wdg_timeout;
+	trans_pcie->txqs.page_offs = trans_cfg->cb_data_offs;
+	trans_pcie->txqs.dev_cmd_offs = trans_cfg->cb_data_offs + sizeof(void *);
+	trans_pcie->txqs.queue_alloc_cmd_ver = trans_cfg->queue_alloc_cmd_ver;
 
 	if (WARN_ON(trans_cfg->n_no_reclaim_cmds > MAX_NO_RECLAIM_CMDS))
 		trans_pcie->n_no_reclaim_cmds = 0;
@@ -1980,7 +1980,7 @@ static void iwl_trans_pcie_configure(struct iwl_trans *trans,
 	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_AX210)
 		trans_pcie->supported_dma_mask = DMA_BIT_MASK(11);
 
-	trans->txqs.bc_table_dword = trans_cfg->bc_table_dword;
+	trans_pcie->txqs.bc_table_dword = trans_cfg->bc_table_dword;
 	trans_pcie->scd_set_active = trans_cfg->scd_set_active;
 
 	trans->command_groups = trans_cfg->command_groups;
@@ -2079,6 +2079,19 @@ void iwl_trans_pcie_free(struct iwl_trans *trans)
 					      trans->dev);
 
 	mutex_destroy(&trans_pcie->mutex);
+
+	if (trans_pcie->txqs.tso_hdr_page) {
+		for_each_possible_cpu(i) {
+			struct iwl_tso_hdr_page *p =
+				per_cpu_ptr(trans_pcie->txqs.tso_hdr_page, i);
+
+			if (p && p->page)
+				__free_page(p->page);
+		}
+
+		free_percpu(trans_pcie->txqs.tso_hdr_page);
+	}
+
 	iwl_trans_free(trans);
 }
 
@@ -2392,6 +2405,7 @@ static int iwl_trans_pcie_rxq_dma_data(struct iwl_trans *trans, int queue,
 
 static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans, int txq_idx)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_txq *txq;
 	unsigned long now = jiffies;
 	bool overflow_tx;
@@ -2401,11 +2415,11 @@ static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans, int txq_idx)
 	if (test_bit(STATUS_TRANS_DEAD, &trans->status))
 		return -ENODEV;
 
-	if (!test_bit(txq_idx, trans->txqs.queue_used))
+	if (!test_bit(txq_idx, trans_pcie->txqs.queue_used))
 		return -EINVAL;
 
 	IWL_DEBUG_TX_QUEUES(trans, "Emptying queue %d...\n", txq_idx);
-	txq = trans->txqs.txq[txq_idx];
+	txq = trans_pcie->txqs.txq[txq_idx];
 
 	spin_lock_bh(&txq->lock);
 	overflow_tx = txq->overflow_tx ||
@@ -2453,6 +2467,7 @@ static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans, int txq_idx)
 
 static int iwl_trans_pcie_wait_txqs_empty(struct iwl_trans *trans, u32 txq_bm)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int cnt;
 	int ret = 0;
 
@@ -2461,9 +2476,9 @@ static int iwl_trans_pcie_wait_txqs_empty(struct iwl_trans *trans, u32 txq_bm)
 	     cnt < trans->trans_cfg->base_params->num_of_queues;
 	     cnt++) {
 
-		if (cnt == trans->txqs.cmd.q_id)
+		if (cnt == trans_pcie->txqs.cmd.q_id)
 			continue;
-		if (!test_bit(cnt, trans->txqs.queue_used))
+		if (!test_bit(cnt, trans_pcie->txqs.queue_used))
 			continue;
 		if (!(BIT(cnt) & txq_bm))
 			continue;
@@ -2636,12 +2651,13 @@ static int iwl_dbgfs_tx_queue_seq_show(struct seq_file *seq, void *v)
 	struct iwl_dbgfs_tx_queue_priv *priv = seq->private;
 	struct iwl_dbgfs_tx_queue_state *state = v;
 	struct iwl_trans *trans = priv->trans;
-	struct iwl_txq *txq = trans->txqs.txq[state->pos];
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *txq = trans_pcie->txqs.txq[state->pos];
 
 	seq_printf(seq, "hwq %.3u: used=%d stopped=%d ",
 		   (unsigned int)state->pos,
-		   !!test_bit(state->pos, trans->txqs.queue_used),
-		   !!test_bit(state->pos, trans->txqs.queue_stopped));
+		   !!test_bit(state->pos, trans_pcie->txqs.queue_used),
+		   !!test_bit(state->pos, trans_pcie->txqs.queue_stopped));
 	if (txq)
 		seq_printf(seq,
 			   "read=%u write=%u need_update=%d frozen=%d n_window=%d ampdu=%d",
@@ -2651,7 +2667,7 @@ static int iwl_dbgfs_tx_queue_seq_show(struct seq_file *seq, void *v)
 	else
 		seq_puts(seq, "(unallocated)");
 
-	if (state->pos == trans->txqs.cmd.q_id)
+	if (state->pos == trans_pcie->txqs.cmd.q_id)
 		seq_puts(seq, " (HCMD)");
 	seq_puts(seq, "\n");
 
@@ -3068,10 +3084,11 @@ static void iwl_trans_pcie_debugfs_cleanup(struct iwl_trans *trans)
 
 static u32 iwl_trans_pcie_get_cmdlen(struct iwl_trans *trans, void *tfd)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	u32 cmdlen = 0;
 	int i;
 
-	for (i = 0; i < trans->txqs.tfd.max_tbs; i++)
+	for (i = 0; i < trans_pcie->txqs.tfd.max_tbs; i++)
 		cmdlen += iwl_txq_gen1_tfd_tb_get_len(trans, tfd, i);
 
 	return cmdlen;
@@ -3340,7 +3357,7 @@ iwl_trans_pcie_dump_data(struct iwl_trans *trans,
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_fw_error_dump_data *data;
-	struct iwl_txq *cmdq = trans->txqs.txq[trans->txqs.cmd.q_id];
+	struct iwl_txq *cmdq = trans_pcie->txqs.txq[trans_pcie->txqs.cmd.q_id];
 	struct iwl_fw_error_dump_txcmd *txcmd;
 	struct iwl_trans_dump_data *dump_data;
 	u32 len, num_rbs = 0, monitor_len = 0;
@@ -3407,7 +3424,7 @@ iwl_trans_pcie_dump_data(struct iwl_trans *trans,
 	data = (void *)dump_data->data;
 
 	if (dump_mask & BIT(IWL_FW_ERROR_DUMP_TXCMD) && cmdq) {
-		u16 tfd_size = trans->txqs.tfd.size;
+		u16 tfd_size = trans_pcie->txqs.tfd.size;
 
 		data->type = cpu_to_le32(IWL_FW_ERROR_DUMP_TXCMD);
 		txcmd = (void *)data->data;
@@ -3543,7 +3560,7 @@ static const struct iwl_trans_ops trans_ops_pcie = {
 	.send_cmd = iwl_pcie_enqueue_hcmd,
 
 	.tx = iwl_trans_pcie_tx,
-	.reclaim = iwl_txq_reclaim,
+	.reclaim = iwl_pcie_reclaim,
 
 	.txq_disable = iwl_trans_pcie_txq_disable,
 	.txq_enable = iwl_trans_pcie_txq_enable,
@@ -3552,7 +3569,7 @@ static const struct iwl_trans_ops trans_ops_pcie = {
 
 	.wait_tx_queues_empty = iwl_trans_pcie_wait_txqs_empty,
 
-	.freeze_txq_timer = iwl_trans_txq_freeze_timer,
+	.freeze_txq_timer = iwl_pcie_freeze_txq_timer,
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	.debugfs_cleanup = iwl_trans_pcie_debugfs_cleanup,
 #endif
@@ -3568,9 +3585,9 @@ static const struct iwl_trans_ops trans_ops_pcie_gen2 = {
 	.send_cmd = iwl_pcie_gen2_enqueue_hcmd,
 
 	.tx = iwl_txq_gen2_tx,
-	.reclaim = iwl_txq_reclaim,
+	.reclaim = iwl_pcie_reclaim,
 
-	.set_q_ptrs = iwl_txq_set_q_ptrs,
+	.set_q_ptrs = iwl_pcie_set_q_ptrs,
 
 	.txq_alloc = iwl_txq_dyn_alloc,
 	.txq_free = iwl_txq_dyn_free,
@@ -3618,13 +3635,58 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 
 	trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
+	if (trans->trans_cfg->gen2) {
+		trans_pcie->txqs.tfd.addr_size = 64;
+		trans_pcie->txqs.tfd.max_tbs = IWL_TFH_NUM_TBS;
+		trans_pcie->txqs.tfd.size = sizeof(struct iwl_tfh_tfd);
+	} else {
+		trans_pcie->txqs.tfd.addr_size = 36;
+		trans_pcie->txqs.tfd.max_tbs = IWL_NUM_OF_TBS;
+		trans_pcie->txqs.tfd.size = sizeof(struct iwl_tfd);
+	}
+	trans->max_skb_frags = IWL_TRANS_PCIE_MAX_FRAGS(trans_pcie);
+
+	trans_pcie->txqs.tso_hdr_page = alloc_percpu(struct iwl_tso_hdr_page);
+	if (!trans_pcie->txqs.tso_hdr_page) {
+		ret = -ENOMEM;
+		goto out_free_trans;
+	}
+
+	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ)
+		trans_pcie->txqs.bc_tbl_size =
+			sizeof(struct iwl_gen3_bc_tbl_entry) * TFD_QUEUE_BC_SIZE_GEN3_BZ;
+	else if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_AX210)
+		trans_pcie->txqs.bc_tbl_size =
+			sizeof(struct iwl_gen3_bc_tbl_entry) * TFD_QUEUE_BC_SIZE_GEN3_AX210;
+	else
+		trans_pcie->txqs.bc_tbl_size = sizeof(struct iwlagn_scd_bc_tbl);
+	/*
+	 * For gen2 devices, we use a single allocation for each byte-count
+	 * table, but they're pretty small (1k) so use a DMA pool that we
+	 * allocate here.
+	 */
+	if (trans->trans_cfg->gen2) {
+		trans_pcie->txqs.bc_pool =
+			dmam_pool_create("iwlwifi:bc", trans->dev,
+					 trans_pcie->txqs.bc_tbl_size,
+					 256, 0);
+		if (!trans_pcie->txqs.bc_pool) {
+			ret = -ENOMEM;
+			goto out_free_tso;
+		}
+	}
+
+	/* Some things must not change even if the config does */
+	WARN_ON(trans_pcie->txqs.tfd.addr_size !=
+		(trans->trans_cfg->gen2 ? 64 : 36));
+
 	/* Initialize NAPI here - it should be before registering to mac80211
 	 * in the opmode but after the HW struct is allocated.
 	 */
 	trans_pcie->napi_dev = alloc_netdev_dummy(sizeof(struct iwl_trans_pcie *));
 	if (!trans_pcie->napi_dev) {
 		ret = -ENOMEM;
-		goto out_free_trans;
+		goto out_free_tso;
 	}
 	/* The private struct in netdev is a pointer to struct iwl_trans_pcie */
 	priv = netdev_priv(trans_pcie->napi_dev);
@@ -3663,7 +3725,7 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 
 	pci_set_master(pdev);
 
-	addr_size = trans->txqs.tfd.addr_size;
+	addr_size = trans_pcie->txqs.tfd.addr_size;
 	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(addr_size));
 	if (ret) {
 		ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
@@ -3766,6 +3828,8 @@ out_no_pci:
 	destroy_workqueue(trans_pcie->rba.alloc_wq);
 out_free_ndev:
 	free_netdev(trans_pcie->napi_dev);
+out_free_tso:
+	free_percpu(trans_pcie->txqs.tso_hdr_page);
 out_free_trans:
 	iwl_trans_free(trans);
 	return ERR_PTR(ret);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
index aabbef114bc2..b897de1b9226 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
  * Copyright (C) 2017 Intel Deutschland GmbH
- * Copyright (C) 2018-2020, 2023 Intel Corporation
+ * Copyright (C) 2018-2020, 2023-2024 Intel Corporation
  */
 #include <net/tso.h>
 #include <linux/tcp.h>
@@ -11,7 +11,1127 @@
 #include "iwl-io.h"
 #include "internal.h"
 #include "fw/api/tx.h"
-#include "queue/tx.h"
+#include "fw/api/commands.h"
+#include "fw/api/datapath.h"
+#include "iwl-scd.h"
+
+static struct page *get_workaround_page(struct iwl_trans *trans,
+					struct sk_buff *skb)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct page **page_ptr;
+	struct page *ret;
+
+	page_ptr = (void *)((u8 *)skb->cb + trans_pcie->txqs.page_offs);
+
+	ret = alloc_page(GFP_ATOMIC);
+	if (!ret)
+		return NULL;
+
+	/* set the chaining pointer to the previous page if there */
+	*(void **)((u8 *)page_address(ret) + PAGE_SIZE - sizeof(void *)) = *page_ptr;
+	*page_ptr = ret;
+
+	return ret;
+}
+
+/*
+ * Add a TB and if needed apply the FH HW bug workaround;
+ * meta != NULL indicates that it's a page mapping and we
+ * need to dma_unmap_page() and set the meta->tbs bit in
+ * this case.
+ */
+static int iwl_txq_gen2_set_tb_with_wa(struct iwl_trans *trans,
+				       struct sk_buff *skb,
+				       struct iwl_tfh_tfd *tfd,
+				       dma_addr_t phys, void *virt,
+				       u16 len, struct iwl_cmd_meta *meta)
+{
+	dma_addr_t oldphys = phys;
+	struct page *page;
+	int ret;
+
+	if (unlikely(dma_mapping_error(trans->dev, phys)))
+		return -ENOMEM;
+
+	if (likely(!iwl_txq_crosses_4g_boundary(phys, len))) {
+		ret = iwl_txq_gen2_set_tb(trans, tfd, phys, len);
+
+		if (ret < 0)
+			goto unmap;
+
+		if (meta)
+			meta->tbs |= BIT(ret);
+
+		ret = 0;
+		goto trace;
+	}
+
+	/*
+	 * Work around a hardware bug. If (as expressed in the
+	 * condition above) the TB ends on a 32-bit boundary,
+	 * then the next TB may be accessed with the wrong
+	 * address.
+	 * To work around it, copy the data elsewhere and make
+	 * a new mapping for it so the device will not fail.
+	 */
+
+	if (WARN_ON(len > PAGE_SIZE - sizeof(void *))) {
+		ret = -ENOBUFS;
+		goto unmap;
+	}
+
+	page = get_workaround_page(trans, skb);
+	if (!page) {
+		ret = -ENOMEM;
+		goto unmap;
+	}
+
+	memcpy(page_address(page), virt, len);
+
+	phys = dma_map_single(trans->dev, page_address(page), len,
+			      DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(trans->dev, phys)))
+		return -ENOMEM;
+	ret = iwl_txq_gen2_set_tb(trans, tfd, phys, len);
+	if (ret < 0) {
+		/* unmap the new allocation as single */
+		oldphys = phys;
+		meta = NULL;
+		goto unmap;
+	}
+	IWL_DEBUG_TX(trans,
+		     "TB bug workaround: copied %d bytes from 0x%llx to 0x%llx\n",
+		     len, (unsigned long long)oldphys,
+		     (unsigned long long)phys);
+
+	ret = 0;
+unmap:
+	if (meta)
+		dma_unmap_page(trans->dev, oldphys, len, DMA_TO_DEVICE);
+	else
+		dma_unmap_single(trans->dev, oldphys, len, DMA_TO_DEVICE);
+trace:
+	trace_iwlwifi_dev_tx_tb(trans->dev, skb, virt, phys, len);
+
+	return ret;
+}
+
+static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
+				    struct sk_buff *skb,
+				    struct iwl_tfh_tfd *tfd, int start_len,
+				    u8 hdr_len,
+				    struct iwl_device_tx_cmd *dev_cmd)
+{
+#ifdef CONFIG_INET
+	struct iwl_tx_cmd_gen2 *tx_cmd = (void *)dev_cmd->payload;
+	struct ieee80211_hdr *hdr = (void *)skb->data;
+	unsigned int snap_ip_tcp_hdrlen, ip_hdrlen, total_len, hdr_room;
+	unsigned int mss = skb_shinfo(skb)->gso_size;
+	u16 length, amsdu_pad;
+	u8 *start_hdr;
+	struct iwl_tso_hdr_page *hdr_page;
+	struct tso_t tso;
+
+	trace_iwlwifi_dev_tx(trans->dev, skb, tfd, sizeof(*tfd),
+			     &dev_cmd->hdr, start_len, 0);
+
+	ip_hdrlen = skb_network_header_len(skb);
+	snap_ip_tcp_hdrlen = 8 + ip_hdrlen + tcp_hdrlen(skb);
+	total_len = skb->len - snap_ip_tcp_hdrlen - hdr_len;
+	amsdu_pad = 0;
+
+	/* total amount of header we may need for this A-MSDU */
+	hdr_room = DIV_ROUND_UP(total_len, mss) *
+		(3 + snap_ip_tcp_hdrlen + sizeof(struct ethhdr));
+
+	/* Our device supports 9 segments at most, it will fit in 1 page */
+	hdr_page = iwl_pcie_get_page_hdr(trans, hdr_room, skb);
+	if (!hdr_page)
+		return -ENOMEM;
+
+	start_hdr = hdr_page->pos;
+
+	/*
+	 * Pull the ieee80211 header to be able to use TSO core,
+	 * we will restore it for the tx_status flow.
+	 */
+	skb_pull(skb, hdr_len);
+
+	/*
+	 * Remove the length of all the headers that we don't actually
+	 * have in the MPDU by themselves, but that we duplicate into
+	 * all the different MSDUs inside the A-MSDU.
+	 */
+	le16_add_cpu(&tx_cmd->len, -snap_ip_tcp_hdrlen);
+
+	tso_start(skb, &tso);
+
+	while (total_len) {
+		/* this is the data left for this subframe */
+		unsigned int data_left = min_t(unsigned int, mss, total_len);
+		unsigned int tb_len;
+		dma_addr_t tb_phys;
+		u8 *subf_hdrs_start = hdr_page->pos;
+
+		total_len -= data_left;
+
+		memset(hdr_page->pos, 0, amsdu_pad);
+		hdr_page->pos += amsdu_pad;
+		amsdu_pad = (4 - (sizeof(struct ethhdr) + snap_ip_tcp_hdrlen +
+				  data_left)) & 0x3;
+		ether_addr_copy(hdr_page->pos, ieee80211_get_DA(hdr));
+		hdr_page->pos += ETH_ALEN;
+		ether_addr_copy(hdr_page->pos, ieee80211_get_SA(hdr));
+		hdr_page->pos += ETH_ALEN;
+
+		length = snap_ip_tcp_hdrlen + data_left;
+		*((__be16 *)hdr_page->pos) = cpu_to_be16(length);
+		hdr_page->pos += sizeof(length);
+
+		/*
+		 * This will copy the SNAP as well which will be considered
+		 * as MAC header.
+		 */
+		tso_build_hdr(skb, hdr_page->pos, &tso, data_left, !total_len);
+
+		hdr_page->pos += snap_ip_tcp_hdrlen;
+
+		tb_len = hdr_page->pos - start_hdr;
+		tb_phys = dma_map_single(trans->dev, start_hdr,
+					 tb_len, DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(trans->dev, tb_phys)))
+			goto out_err;
+		/*
+		 * No need for _with_wa, this is from the TSO page and
+		 * we leave some space at the end of it so can't hit
+		 * the buggy scenario.
+		 */
+		iwl_txq_gen2_set_tb(trans, tfd, tb_phys, tb_len);
+		trace_iwlwifi_dev_tx_tb(trans->dev, skb, start_hdr,
+					tb_phys, tb_len);
+		/* add this subframe's headers' length to the tx_cmd */
+		le16_add_cpu(&tx_cmd->len, hdr_page->pos - subf_hdrs_start);
+
+		/* prepare the start_hdr for the next subframe */
+		start_hdr = hdr_page->pos;
+
+		/* put the payload */
+		while (data_left) {
+			int ret;
+
+			tb_len = min_t(unsigned int, tso.size, data_left);
+			tb_phys = dma_map_single(trans->dev, tso.data,
+						 tb_len, DMA_TO_DEVICE);
+			ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd,
+							  tb_phys, tso.data,
+							  tb_len, NULL);
+			if (ret)
+				goto out_err;
+
+			data_left -= tb_len;
+			tso_build_data(skb, &tso, tb_len);
+		}
+	}
+
+	/* re -add the WiFi header */
+	skb_push(skb, hdr_len);
+
+	return 0;
+
+out_err:
+#endif
+	return -EINVAL;
+}
+
+static struct
+iwl_tfh_tfd *iwl_txq_gen2_build_tx_amsdu(struct iwl_trans *trans,
+					 struct iwl_txq *txq,
+					 struct iwl_device_tx_cmd *dev_cmd,
+					 struct sk_buff *skb,
+					 struct iwl_cmd_meta *out_meta,
+					 int hdr_len,
+					 int tx_cmd_len)
+{
+	int idx = iwl_txq_get_cmd_index(txq, txq->write_ptr);
+	struct iwl_tfh_tfd *tfd = iwl_txq_get_tfd(trans, txq, idx);
+	dma_addr_t tb_phys;
+	int len;
+	void *tb1_addr;
+
+	tb_phys = iwl_txq_get_first_tb_dma(txq, idx);
+
+	/*
+	 * No need for _with_wa, the first TB allocation is aligned up
+	 * to a 64-byte boundary and thus can't be at the end or cross
+	 * a page boundary (much less a 2^32 boundary).
+	 */
+	iwl_txq_gen2_set_tb(trans, tfd, tb_phys, IWL_FIRST_TB_SIZE);
+
+	/*
+	 * The second TB (tb1) points to the remainder of the TX command
+	 * and the 802.11 header - dword aligned size
+	 * (This calculation modifies the TX command, so do it before the
+	 * setup of the first TB)
+	 */
+	len = tx_cmd_len + sizeof(struct iwl_cmd_header) + hdr_len -
+	      IWL_FIRST_TB_SIZE;
+
+	/* do not align A-MSDU to dword as the subframe header aligns it */
+
+	/* map the data for TB1 */
+	tb1_addr = ((u8 *)&dev_cmd->hdr) + IWL_FIRST_TB_SIZE;
+	tb_phys = dma_map_single(trans->dev, tb1_addr, len, DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(trans->dev, tb_phys)))
+		goto out_err;
+	/*
+	 * No need for _with_wa(), we ensure (via alignment) that the data
+	 * here can never cross or end at a page boundary.
+	 */
+	iwl_txq_gen2_set_tb(trans, tfd, tb_phys, len);
+
+	if (iwl_txq_gen2_build_amsdu(trans, skb, tfd, len + IWL_FIRST_TB_SIZE,
+				     hdr_len, dev_cmd))
+		goto out_err;
+
+	/* building the A-MSDU might have changed this data, memcpy it now */
+	memcpy(&txq->first_tb_bufs[idx], dev_cmd, IWL_FIRST_TB_SIZE);
+	return tfd;
+
+out_err:
+	iwl_txq_gen2_tfd_unmap(trans, out_meta, tfd);
+	return NULL;
+}
+
+static int iwl_txq_gen2_tx_add_frags(struct iwl_trans *trans,
+				     struct sk_buff *skb,
+				     struct iwl_tfh_tfd *tfd,
+				     struct iwl_cmd_meta *out_meta)
+{
+	int i;
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+		dma_addr_t tb_phys;
+		unsigned int fragsz = skb_frag_size(frag);
+		int ret;
+
+		if (!fragsz)
+			continue;
+
+		tb_phys = skb_frag_dma_map(trans->dev, frag, 0,
+					   fragsz, DMA_TO_DEVICE);
+		ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd, tb_phys,
+						  skb_frag_address(frag),
+						  fragsz, out_meta);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static struct
+iwl_tfh_tfd *iwl_txq_gen2_build_tx(struct iwl_trans *trans,
+				   struct iwl_txq *txq,
+				   struct iwl_device_tx_cmd *dev_cmd,
+				   struct sk_buff *skb,
+				   struct iwl_cmd_meta *out_meta,
+				   int hdr_len,
+				   int tx_cmd_len,
+				   bool pad)
+{
+	int idx = iwl_txq_get_cmd_index(txq, txq->write_ptr);
+	struct iwl_tfh_tfd *tfd = iwl_txq_get_tfd(trans, txq, idx);
+	dma_addr_t tb_phys;
+	int len, tb1_len, tb2_len;
+	void *tb1_addr;
+	struct sk_buff *frag;
+
+	tb_phys = iwl_txq_get_first_tb_dma(txq, idx);
+
+	/* The first TB points to bi-directional DMA data */
+	memcpy(&txq->first_tb_bufs[idx], dev_cmd, IWL_FIRST_TB_SIZE);
+
+	/*
+	 * No need for _with_wa, the first TB allocation is aligned up
+	 * to a 64-byte boundary and thus can't be at the end or cross
+	 * a page boundary (much less a 2^32 boundary).
+	 */
+	iwl_txq_gen2_set_tb(trans, tfd, tb_phys, IWL_FIRST_TB_SIZE);
+
+	/*
+	 * The second TB (tb1) points to the remainder of the TX command
+	 * and the 802.11 header - dword aligned size
+	 * (This calculation modifies the TX command, so do it before the
+	 * setup of the first TB)
+	 */
+	len = tx_cmd_len + sizeof(struct iwl_cmd_header) + hdr_len -
+	      IWL_FIRST_TB_SIZE;
+
+	if (pad)
+		tb1_len = ALIGN(len, 4);
+	else
+		tb1_len = len;
+
+	/* map the data for TB1 */
+	tb1_addr = ((u8 *)&dev_cmd->hdr) + IWL_FIRST_TB_SIZE;
+	tb_phys = dma_map_single(trans->dev, tb1_addr, tb1_len, DMA_TO_DEVICE);
+	if (unlikely(dma_mapping_error(trans->dev, tb_phys)))
+		goto out_err;
+	/*
+	 * No need for _with_wa(), we ensure (via alignment) that the data
+	 * here can never cross or end at a page boundary.
+	 */
+	iwl_txq_gen2_set_tb(trans, tfd, tb_phys, tb1_len);
+	trace_iwlwifi_dev_tx(trans->dev, skb, tfd, sizeof(*tfd), &dev_cmd->hdr,
+			     IWL_FIRST_TB_SIZE + tb1_len, hdr_len);
+
+	/* set up TFD's third entry to point to remainder of skb's head */
+	tb2_len = skb_headlen(skb) - hdr_len;
+
+	if (tb2_len > 0) {
+		int ret;
+
+		tb_phys = dma_map_single(trans->dev, skb->data + hdr_len,
+					 tb2_len, DMA_TO_DEVICE);
+		ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd, tb_phys,
+						  skb->data + hdr_len, tb2_len,
+						  NULL);
+		if (ret)
+			goto out_err;
+	}
+
+	if (iwl_txq_gen2_tx_add_frags(trans, skb, tfd, out_meta))
+		goto out_err;
+
+	skb_walk_frags(skb, frag) {
+		int ret;
+
+		tb_phys = dma_map_single(trans->dev, frag->data,
+					 skb_headlen(frag), DMA_TO_DEVICE);
+		ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd, tb_phys,
+						  frag->data,
+						  skb_headlen(frag), NULL);
+		if (ret)
+			goto out_err;
+		if (iwl_txq_gen2_tx_add_frags(trans, frag, tfd, out_meta))
+			goto out_err;
+	}
+
+	return tfd;
+
+out_err:
+	iwl_txq_gen2_tfd_unmap(trans, out_meta, tfd);
+	return NULL;
+}
+
+static
+struct iwl_tfh_tfd *iwl_txq_gen2_build_tfd(struct iwl_trans *trans,
+					   struct iwl_txq *txq,
+					   struct iwl_device_tx_cmd *dev_cmd,
+					   struct sk_buff *skb,
+					   struct iwl_cmd_meta *out_meta)
+{
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+	int idx = iwl_txq_get_cmd_index(txq, txq->write_ptr);
+	struct iwl_tfh_tfd *tfd = iwl_txq_get_tfd(trans, txq, idx);
+	int len, hdr_len;
+	bool amsdu;
+
+	/* There must be data left over for TB1 or this code must be changed */
+	BUILD_BUG_ON(sizeof(struct iwl_tx_cmd_gen2) < IWL_FIRST_TB_SIZE);
+	BUILD_BUG_ON(sizeof(struct iwl_cmd_header) +
+		     offsetofend(struct iwl_tx_cmd_gen2, dram_info) >
+		     IWL_FIRST_TB_SIZE);
+	BUILD_BUG_ON(sizeof(struct iwl_tx_cmd_gen3) < IWL_FIRST_TB_SIZE);
+	BUILD_BUG_ON(sizeof(struct iwl_cmd_header) +
+		     offsetofend(struct iwl_tx_cmd_gen3, dram_info) >
+		     IWL_FIRST_TB_SIZE);
+
+	memset(tfd, 0, sizeof(*tfd));
+
+	if (trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_AX210)
+		len = sizeof(struct iwl_tx_cmd_gen2);
+	else
+		len = sizeof(struct iwl_tx_cmd_gen3);
+
+	amsdu = ieee80211_is_data_qos(hdr->frame_control) &&
+			(*ieee80211_get_qos_ctl(hdr) &
+			 IEEE80211_QOS_CTL_A_MSDU_PRESENT);
+
+	hdr_len = ieee80211_hdrlen(hdr->frame_control);
+
+	/*
+	 * Only build A-MSDUs here if doing so by GSO, otherwise it may be
+	 * an A-MSDU for other reasons, e.g. NAN or an A-MSDU having been
+	 * built in the higher layers already.
+	 */
+	if (amsdu && skb_shinfo(skb)->gso_size)
+		return iwl_txq_gen2_build_tx_amsdu(trans, txq, dev_cmd, skb,
+						    out_meta, hdr_len, len);
+	return iwl_txq_gen2_build_tx(trans, txq, dev_cmd, skb, out_meta,
+				      hdr_len, len, !amsdu);
+}
+
+int iwl_txq_space(struct iwl_trans *trans, const struct iwl_txq *q)
+{
+	unsigned int max;
+	unsigned int used;
+
+	/*
+	 * To avoid ambiguity between empty and completely full queues, there
+	 * should always be less than max_tfd_queue_size elements in the queue.
+	 * If q->n_window is smaller than max_tfd_queue_size, there is no need
+	 * to reserve any queue entries for this purpose.
+	 */
+	if (q->n_window < trans->trans_cfg->base_params->max_tfd_queue_size)
+		max = q->n_window;
+	else
+		max = trans->trans_cfg->base_params->max_tfd_queue_size - 1;
+
+	/*
+	 * max_tfd_queue_size is a power of 2, so the following is equivalent to
+	 * modulo by max_tfd_queue_size and is well defined.
+	 */
+	used = (q->write_ptr - q->read_ptr) &
+		(trans->trans_cfg->base_params->max_tfd_queue_size - 1);
+
+	if (WARN_ON(used > max))
+		return 0;
+
+	return max - used;
+}
+
+/*
+ * iwl_pcie_gen2_update_byte_tbl - Set up entry in Tx byte-count array
+ */
+static void iwl_pcie_gen2_update_byte_tbl(struct iwl_trans *trans,
+					  struct iwl_txq *txq, u16 byte_cnt,
+					  int num_tbs)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	int idx = iwl_txq_get_cmd_index(txq, txq->write_ptr);
+	u8 filled_tfd_size, num_fetch_chunks;
+	u16 len = byte_cnt;
+	__le16 bc_ent;
+
+	if (WARN(idx >= txq->n_window, "%d >= %d\n", idx, txq->n_window))
+		return;
+
+	filled_tfd_size = offsetof(struct iwl_tfh_tfd, tbs) +
+			  num_tbs * sizeof(struct iwl_tfh_tb);
+	/*
+	 * filled_tfd_size contains the number of filled bytes in the TFD.
+	 * Dividing it by 64 will give the number of chunks to fetch
+	 * to SRAM- 0 for one chunk, 1 for 2 and so on.
+	 * If, for example, TFD contains only 3 TBs then 32 bytes
+	 * of the TFD are used, and only one chunk of 64 bytes should
+	 * be fetched
+	 */
+	num_fetch_chunks = DIV_ROUND_UP(filled_tfd_size, 64) - 1;
+
+	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_AX210) {
+		struct iwl_gen3_bc_tbl_entry *scd_bc_tbl_gen3 = txq->bc_tbl.addr;
+
+		/* Starting from AX210, the HW expects bytes */
+		WARN_ON(trans_pcie->txqs.bc_table_dword);
+		WARN_ON(len > 0x3FFF);
+		bc_ent = cpu_to_le16(len | (num_fetch_chunks << 14));
+		scd_bc_tbl_gen3[idx].tfd_offset = bc_ent;
+	} else {
+		struct iwlagn_scd_bc_tbl *scd_bc_tbl = txq->bc_tbl.addr;
+
+		/* Before AX210, the HW expects DW */
+		WARN_ON(!trans_pcie->txqs.bc_table_dword);
+		len = DIV_ROUND_UP(len, 4);
+		WARN_ON(len > 0xFFF);
+		bc_ent = cpu_to_le16(len | (num_fetch_chunks << 12));
+		scd_bc_tbl->tfd_offset[idx] = bc_ent;
+	}
+}
+
+static u8 iwl_txq_gen2_get_num_tbs(struct iwl_tfh_tfd *tfd)
+{
+	return le16_to_cpu(tfd->num_tbs) & 0x1f;
+}
+
+int iwl_txq_gen2_set_tb(struct iwl_trans *trans, struct iwl_tfh_tfd *tfd,
+			dma_addr_t addr, u16 len)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	int idx = iwl_txq_gen2_get_num_tbs(tfd);
+	struct iwl_tfh_tb *tb;
+
+	/* Only WARN here so we know about the issue, but we mess up our
+	 * unmap path because not every place currently checks for errors
+	 * returned from this function - it can only return an error if
+	 * there's no more space, and so when we know there is enough we
+	 * don't always check ...
+	 */
+	WARN(iwl_txq_crosses_4g_boundary(addr, len),
+	     "possible DMA problem with iova:0x%llx, len:%d\n",
+	     (unsigned long long)addr, len);
+
+	if (WARN_ON(idx >= IWL_TFH_NUM_TBS))
+		return -EINVAL;
+	tb = &tfd->tbs[idx];
+
+	/* Each TFD can point to a maximum max_tbs Tx buffers */
+	if (le16_to_cpu(tfd->num_tbs) >= trans_pcie->txqs.tfd.max_tbs) {
+		IWL_ERR(trans, "Error can not send more than %d chunks\n",
+			trans_pcie->txqs.tfd.max_tbs);
+		return -EINVAL;
+	}
+
+	put_unaligned_le64(addr, &tb->addr);
+	tb->tb_len = cpu_to_le16(len);
+
+	tfd->num_tbs = cpu_to_le16(idx + 1);
+
+	return idx;
+}
+
+void iwl_txq_gen2_tfd_unmap(struct iwl_trans *trans,
+			    struct iwl_cmd_meta *meta,
+			    struct iwl_tfh_tfd *tfd)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	int i, num_tbs;
+
+	/* Sanity check on number of chunks */
+	num_tbs = iwl_txq_gen2_get_num_tbs(tfd);
+
+	if (num_tbs > trans_pcie->txqs.tfd.max_tbs) {
+		IWL_ERR(trans, "Too many chunks: %i\n", num_tbs);
+		return;
+	}
+
+	/* first TB is never freed - it's the bidirectional DMA data */
+	for (i = 1; i < num_tbs; i++) {
+		if (meta->tbs & BIT(i))
+			dma_unmap_page(trans->dev,
+				       le64_to_cpu(tfd->tbs[i].addr),
+				       le16_to_cpu(tfd->tbs[i].tb_len),
+				       DMA_TO_DEVICE);
+		else
+			dma_unmap_single(trans->dev,
+					 le64_to_cpu(tfd->tbs[i].addr),
+					 le16_to_cpu(tfd->tbs[i].tb_len),
+					 DMA_TO_DEVICE);
+	}
+
+	iwl_txq_set_tfd_invalid_gen2(trans, tfd);
+}
+
+static void iwl_txq_gen2_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq)
+{
+	/* rd_ptr is bounded by TFD_QUEUE_SIZE_MAX and
+	 * idx is bounded by n_window
+	 */
+	int idx = iwl_txq_get_cmd_index(txq, txq->read_ptr);
+	struct sk_buff *skb;
+
+	lockdep_assert_held(&txq->lock);
+
+	if (!txq->entries)
+		return;
+
+	iwl_txq_gen2_tfd_unmap(trans, &txq->entries[idx].meta,
+			       iwl_txq_get_tfd(trans, txq, idx));
+
+	skb = txq->entries[idx].skb;
+
+	/* Can be called from irqs-disabled context
+	 * If skb is not NULL, it means that the whole queue is being
+	 * freed and that the queue is not empty - free the skb
+	 */
+	if (skb) {
+		iwl_op_mode_free_skb(trans->op_mode, skb);
+		txq->entries[idx].skb = NULL;
+	}
+}
+
+/*
+ * iwl_txq_inc_wr_ptr - Send new write index to hardware
+ */
+static void iwl_txq_inc_wr_ptr(struct iwl_trans *trans, struct iwl_txq *txq)
+{
+	lockdep_assert_held(&txq->lock);
+
+	IWL_DEBUG_TX(trans, "Q:%d WR: 0x%x\n", txq->id, txq->write_ptr);
+
+	/*
+	 * if not in power-save mode, uCode will never sleep when we're
+	 * trying to tx (during RFKILL, we're not trying to tx).
+	 */
+	iwl_write32(trans, HBUS_TARG_WRPTR, txq->write_ptr | (txq->id << 16));
+}
+
+int iwl_txq_gen2_tx(struct iwl_trans *trans, struct sk_buff *skb,
+		    struct iwl_device_tx_cmd *dev_cmd, int txq_id)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_cmd_meta *out_meta;
+	struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
+	u16 cmd_len;
+	int idx;
+	void *tfd;
+
+	if (WARN_ONCE(txq_id >= IWL_MAX_TVQM_QUEUES,
+		      "queue %d out of range", txq_id))
+		return -EINVAL;
+
+	if (WARN_ONCE(!test_bit(txq_id, trans_pcie->txqs.queue_used),
+		      "TX on unused queue %d\n", txq_id))
+		return -EINVAL;
+
+	if (skb_is_nonlinear(skb) &&
+	    skb_shinfo(skb)->nr_frags > IWL_TRANS_PCIE_MAX_FRAGS(trans_pcie) &&
+	    __skb_linearize(skb))
+		return -ENOMEM;
+
+	spin_lock(&txq->lock);
+
+	if (iwl_txq_space(trans, txq) < txq->high_mark) {
+		iwl_txq_stop(trans, txq);
+
+		/* don't put the packet on the ring, if there is no room */
+		if (unlikely(iwl_txq_space(trans, txq) < 3)) {
+			struct iwl_device_tx_cmd **dev_cmd_ptr;
+
+			dev_cmd_ptr = (void *)((u8 *)skb->cb +
+					       trans_pcie->txqs.dev_cmd_offs);
+
+			*dev_cmd_ptr = dev_cmd;
+			__skb_queue_tail(&txq->overflow_q, skb);
+			spin_unlock(&txq->lock);
+			return 0;
+		}
+	}
+
+	idx = iwl_txq_get_cmd_index(txq, txq->write_ptr);
+
+	/* Set up driver data for this TFD */
+	txq->entries[idx].skb = skb;
+	txq->entries[idx].cmd = dev_cmd;
+
+	dev_cmd->hdr.sequence =
+		cpu_to_le16((u16)(QUEUE_TO_SEQ(txq_id) |
+			    INDEX_TO_SEQ(idx)));
+
+	/* Set up first empty entry in queue's array of Tx/cmd buffers */
+	out_meta = &txq->entries[idx].meta;
+	out_meta->flags = 0;
+
+	tfd = iwl_txq_gen2_build_tfd(trans, txq, dev_cmd, skb, out_meta);
+	if (!tfd) {
+		spin_unlock(&txq->lock);
+		return -1;
+	}
+
+	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_AX210) {
+		struct iwl_tx_cmd_gen3 *tx_cmd_gen3 =
+			(void *)dev_cmd->payload;
+
+		cmd_len = le16_to_cpu(tx_cmd_gen3->len);
+	} else {
+		struct iwl_tx_cmd_gen2 *tx_cmd_gen2 =
+			(void *)dev_cmd->payload;
+
+		cmd_len = le16_to_cpu(tx_cmd_gen2->len);
+	}
+
+	/* Set up entry for this TFD in Tx byte-count array */
+	iwl_pcie_gen2_update_byte_tbl(trans, txq, cmd_len,
+				      iwl_txq_gen2_get_num_tbs(tfd));
+
+	/* start timer if queue currently empty */
+	if (txq->read_ptr == txq->write_ptr && txq->wd_timeout)
+		mod_timer(&txq->stuck_timer, jiffies + txq->wd_timeout);
+
+	/* Tell device the write index *just past* this latest filled TFD */
+	txq->write_ptr = iwl_txq_inc_wrap(trans, txq->write_ptr);
+	iwl_txq_inc_wr_ptr(trans, txq);
+	/*
+	 * At this point the frame is "transmitted" successfully
+	 * and we will get a TX status notification eventually.
+	 */
+	spin_unlock(&txq->lock);
+	return 0;
+}
+
+/*************** HOST COMMAND QUEUE FUNCTIONS   *****/
+
+/*
+ * iwl_txq_gen2_unmap -  Unmap any remaining DMA mappings and free skb's
+ */
+static void iwl_txq_gen2_unmap(struct iwl_trans *trans, int txq_id)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
+
+	spin_lock_bh(&txq->lock);
+	while (txq->write_ptr != txq->read_ptr) {
+		IWL_DEBUG_TX_REPLY(trans, "Q %d Free %d\n",
+				   txq_id, txq->read_ptr);
+
+		if (txq_id != trans_pcie->txqs.cmd.q_id) {
+			int idx = iwl_txq_get_cmd_index(txq, txq->read_ptr);
+			struct sk_buff *skb = txq->entries[idx].skb;
+
+			if (!WARN_ON_ONCE(!skb))
+				iwl_pcie_free_tso_page(trans, skb);
+		}
+		iwl_txq_gen2_free_tfd(trans, txq);
+		txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr);
+	}
+
+	while (!skb_queue_empty(&txq->overflow_q)) {
+		struct sk_buff *skb = __skb_dequeue(&txq->overflow_q);
+
+		iwl_op_mode_free_skb(trans->op_mode, skb);
+	}
+
+	spin_unlock_bh(&txq->lock);
+
+	/* just in case - this queue may have been stopped */
+	iwl_trans_pcie_wake_queue(trans, txq);
+}
+
+static void iwl_txq_gen2_free_memory(struct iwl_trans *trans,
+				     struct iwl_txq *txq)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct device *dev = trans->dev;
+
+	/* De-alloc circular buffer of TFDs */
+	if (txq->tfds) {
+		dma_free_coherent(dev,
+				  trans_pcie->txqs.tfd.size * txq->n_window,
+				  txq->tfds, txq->dma_addr);
+		dma_free_coherent(dev,
+				  sizeof(*txq->first_tb_bufs) * txq->n_window,
+				  txq->first_tb_bufs, txq->first_tb_dma);
+	}
+
+	kfree(txq->entries);
+	if (txq->bc_tbl.addr)
+		dma_pool_free(trans_pcie->txqs.bc_pool,
+			      txq->bc_tbl.addr, txq->bc_tbl.dma);
+	kfree(txq);
+}
+
+/*
+ * iwl_pcie_txq_free - Deallocate DMA queue.
+ * @txq: Transmit queue to deallocate.
+ *
+ * Empty queue by removing and destroying all BD's.
+ * Free all buffers.
+ * 0-fill, but do not free "txq" descriptor structure.
+ */
+static void iwl_txq_gen2_free(struct iwl_trans *trans, int txq_id)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *txq;
+	int i;
+
+	if (WARN_ONCE(txq_id >= IWL_MAX_TVQM_QUEUES,
+		      "queue %d out of range", txq_id))
+		return;
+
+	txq = trans_pcie->txqs.txq[txq_id];
+
+	if (WARN_ON(!txq))
+		return;
+
+	iwl_txq_gen2_unmap(trans, txq_id);
+
+	/* De-alloc array of command/tx buffers */
+	if (txq_id == trans_pcie->txqs.cmd.q_id)
+		for (i = 0; i < txq->n_window; i++) {
+			kfree_sensitive(txq->entries[i].cmd);
+			kfree_sensitive(txq->entries[i].free_buf);
+		}
+	del_timer_sync(&txq->stuck_timer);
+
+	iwl_txq_gen2_free_memory(trans, txq);
+
+	trans_pcie->txqs.txq[txq_id] = NULL;
+
+	clear_bit(txq_id, trans_pcie->txqs.queue_used);
+}
+
+static struct iwl_txq *
+iwl_txq_dyn_alloc_dma(struct iwl_trans *trans, int size, unsigned int timeout)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	size_t bc_tbl_size, bc_tbl_entries;
+	struct iwl_txq *txq;
+	int ret;
+
+	WARN_ON(!trans_pcie->txqs.bc_tbl_size);
+
+	bc_tbl_size = trans_pcie->txqs.bc_tbl_size;
+	bc_tbl_entries = bc_tbl_size / sizeof(u16);
+
+	if (WARN_ON(size > bc_tbl_entries))
+		return ERR_PTR(-EINVAL);
+
+	txq = kzalloc(sizeof(*txq), GFP_KERNEL);
+	if (!txq)
+		return ERR_PTR(-ENOMEM);
+
+	txq->bc_tbl.addr = dma_pool_alloc(trans_pcie->txqs.bc_pool, GFP_KERNEL,
+					  &txq->bc_tbl.dma);
+	if (!txq->bc_tbl.addr) {
+		IWL_ERR(trans, "Scheduler BC Table allocation failed\n");
+		kfree(txq);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	ret = iwl_pcie_txq_alloc(trans, txq, size, false);
+	if (ret) {
+		IWL_ERR(trans, "Tx queue alloc failed\n");
+		goto error;
+	}
+	ret = iwl_txq_init(trans, txq, size, false);
+	if (ret) {
+		IWL_ERR(trans, "Tx queue init failed\n");
+		goto error;
+	}
+
+	txq->wd_timeout = msecs_to_jiffies(timeout);
+
+	return txq;
+
+error:
+	iwl_txq_gen2_free_memory(trans, txq);
+	return ERR_PTR(ret);
+}
+
+static int iwl_pcie_txq_alloc_response(struct iwl_trans *trans,
+				       struct iwl_txq *txq,
+				       struct iwl_host_cmd *hcmd)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_tx_queue_cfg_rsp *rsp;
+	int ret, qid;
+	u32 wr_ptr;
+
+	if (WARN_ON(iwl_rx_packet_payload_len(hcmd->resp_pkt) !=
+		    sizeof(*rsp))) {
+		ret = -EINVAL;
+		goto error_free_resp;
+	}
+
+	rsp = (void *)hcmd->resp_pkt->data;
+	qid = le16_to_cpu(rsp->queue_number);
+	wr_ptr = le16_to_cpu(rsp->write_pointer);
+
+	if (qid >= ARRAY_SIZE(trans_pcie->txqs.txq)) {
+		WARN_ONCE(1, "queue index %d unsupported", qid);
+		ret = -EIO;
+		goto error_free_resp;
+	}
+
+	if (test_and_set_bit(qid, trans_pcie->txqs.queue_used)) {
+		WARN_ONCE(1, "queue %d already used", qid);
+		ret = -EIO;
+		goto error_free_resp;
+	}
+
+	if (WARN_ONCE(trans_pcie->txqs.txq[qid],
+		      "queue %d already allocated\n", qid)) {
+		ret = -EIO;
+		goto error_free_resp;
+	}
+
+	txq->id = qid;
+	trans_pcie->txqs.txq[qid] = txq;
+	wr_ptr &= (trans->trans_cfg->base_params->max_tfd_queue_size - 1);
+
+	/* Place first TFD at index corresponding to start sequence number */
+	txq->read_ptr = wr_ptr;
+	txq->write_ptr = wr_ptr;
+
+	IWL_DEBUG_TX_QUEUES(trans, "Activate queue %d\n", qid);
+
+	iwl_free_resp(hcmd);
+	return qid;
+
+error_free_resp:
+	iwl_free_resp(hcmd);
+	iwl_txq_gen2_free_memory(trans, txq);
+	return ret;
+}
+
+int iwl_txq_dyn_alloc(struct iwl_trans *trans, u32 flags, u32 sta_mask,
+		      u8 tid, int size, unsigned int timeout)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *txq;
+	union {
+		struct iwl_tx_queue_cfg_cmd old;
+		struct iwl_scd_queue_cfg_cmd new;
+	} cmd;
+	struct iwl_host_cmd hcmd = {
+		.flags = CMD_WANT_SKB,
+	};
+	int ret;
+
+	/* take the min with bytecount table entries allowed */
+	size = min_t(u32, size, trans_pcie->txqs.bc_tbl_size / sizeof(u16));
+	/* but must be power of 2 values for calculating read/write pointers */
+	size = rounddown_pow_of_two(size);
+
+	if (trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_BZ &&
+	    trans->hw_rev_step == SILICON_A_STEP) {
+		size = 4096;
+		txq = iwl_txq_dyn_alloc_dma(trans, size, timeout);
+	} else {
+		do {
+			txq = iwl_txq_dyn_alloc_dma(trans, size, timeout);
+			if (!IS_ERR(txq))
+				break;
+
+			IWL_DEBUG_TX_QUEUES(trans,
+					    "Failed allocating TXQ of size %d for sta mask %x tid %d, ret: %ld\n",
+					    size, sta_mask, tid,
+					    PTR_ERR(txq));
+			size /= 2;
+		} while (size >= 16);
+	}
+
+	if (IS_ERR(txq))
+		return PTR_ERR(txq);
+
+	if (trans_pcie->txqs.queue_alloc_cmd_ver == 0) {
+		memset(&cmd.old, 0, sizeof(cmd.old));
+		cmd.old.tfdq_addr = cpu_to_le64(txq->dma_addr);
+		cmd.old.byte_cnt_addr = cpu_to_le64(txq->bc_tbl.dma);
+		cmd.old.cb_size = cpu_to_le32(TFD_QUEUE_CB_SIZE(size));
+		cmd.old.flags = cpu_to_le16(flags | TX_QUEUE_CFG_ENABLE_QUEUE);
+		cmd.old.tid = tid;
+
+		if (hweight32(sta_mask) != 1) {
+			ret = -EINVAL;
+			goto error;
+		}
+		cmd.old.sta_id = ffs(sta_mask) - 1;
+
+		hcmd.id = SCD_QUEUE_CFG;
+		hcmd.len[0] = sizeof(cmd.old);
+		hcmd.data[0] = &cmd.old;
+	} else if (trans_pcie->txqs.queue_alloc_cmd_ver == 3) {
+		memset(&cmd.new, 0, sizeof(cmd.new));
+		cmd.new.operation = cpu_to_le32(IWL_SCD_QUEUE_ADD);
+		cmd.new.u.add.tfdq_dram_addr = cpu_to_le64(txq->dma_addr);
+		cmd.new.u.add.bc_dram_addr = cpu_to_le64(txq->bc_tbl.dma);
+		cmd.new.u.add.cb_size = cpu_to_le32(TFD_QUEUE_CB_SIZE(size));
+		cmd.new.u.add.flags = cpu_to_le32(flags);
+		cmd.new.u.add.sta_mask = cpu_to_le32(sta_mask);
+		cmd.new.u.add.tid = tid;
+
+		hcmd.id = WIDE_ID(DATA_PATH_GROUP, SCD_QUEUE_CONFIG_CMD);
+		hcmd.len[0] = sizeof(cmd.new);
+		hcmd.data[0] = &cmd.new;
+	} else {
+		ret = -EOPNOTSUPP;
+		goto error;
+	}
+
+	ret = iwl_trans_send_cmd(trans, &hcmd);
+	if (ret)
+		goto error;
+
+	return iwl_pcie_txq_alloc_response(trans, txq, &hcmd);
+
+error:
+	iwl_txq_gen2_free_memory(trans, txq);
+	return ret;
+}
+
+void iwl_txq_dyn_free(struct iwl_trans *trans, int queue)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+
+	if (WARN(queue >= IWL_MAX_TVQM_QUEUES,
+		 "queue %d out of range", queue))
+		return;
+
+	/*
+	 * Upon HW Rfkill - we stop the device, and then stop the queues
+	 * in the op_mode. Just for the sake of the simplicity of the op_mode,
+	 * allow the op_mode to call txq_disable after it already called
+	 * stop_device.
+	 */
+	if (!test_and_clear_bit(queue, trans_pcie->txqs.queue_used)) {
+		WARN_ONCE(test_bit(STATUS_DEVICE_ENABLED, &trans->status),
+			  "queue %d not used", queue);
+		return;
+	}
+
+	iwl_txq_gen2_free(trans, queue);
+
+	IWL_DEBUG_TX_QUEUES(trans, "Deactivate queue %d\n", queue);
+}
+
+void iwl_txq_gen2_tx_free(struct iwl_trans *trans)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	int i;
+
+	memset(trans_pcie->txqs.queue_used, 0,
+	       sizeof(trans_pcie->txqs.queue_used));
+
+	/* Free all TX queues */
+	for (i = 0; i < ARRAY_SIZE(trans_pcie->txqs.txq); i++) {
+		if (!trans_pcie->txqs.txq[i])
+			continue;
+
+		iwl_txq_gen2_free(trans, i);
+	}
+}
+
+int iwl_txq_gen2_init(struct iwl_trans *trans, int txq_id, int queue_size)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *queue;
+	int ret;
+
+	/* alloc and init the tx queue */
+	if (!trans_pcie->txqs.txq[txq_id]) {
+		queue = kzalloc(sizeof(*queue), GFP_KERNEL);
+		if (!queue) {
+			IWL_ERR(trans, "Not enough memory for tx queue\n");
+			return -ENOMEM;
+		}
+		trans_pcie->txqs.txq[txq_id] = queue;
+		ret = iwl_pcie_txq_alloc(trans, queue, queue_size, true);
+		if (ret) {
+			IWL_ERR(trans, "Tx %d queue init failed\n", txq_id);
+			goto error;
+		}
+	} else {
+		queue = trans_pcie->txqs.txq[txq_id];
+	}
+
+	ret = iwl_txq_init(trans, queue, queue_size,
+			   (txq_id == trans_pcie->txqs.cmd.q_id));
+	if (ret) {
+		IWL_ERR(trans, "Tx %d queue alloc failed\n", txq_id);
+		goto error;
+	}
+	trans_pcie->txqs.txq[txq_id]->id = txq_id;
+	set_bit(txq_id, trans_pcie->txqs.queue_used);
+
+	return 0;
+
+error:
+	iwl_txq_gen2_tx_free(trans);
+	return ret;
+}
 
 /*************** HOST COMMAND QUEUE FUNCTIONS   *****/
 
@@ -28,7 +1148,7 @@ int iwl_pcie_gen2_enqueue_hcmd(struct iwl_trans *trans,
 			       struct iwl_host_cmd *cmd)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_txq *txq = trans->txqs.txq[trans->txqs.cmd.q_id];
+	struct iwl_txq *txq = trans_pcie->txqs.txq[trans_pcie->txqs.cmd.q_id];
 	struct iwl_device_cmd *out_cmd;
 	struct iwl_cmd_meta *out_meta;
 	void *dup_buf = NULL;
@@ -143,7 +1263,7 @@ int iwl_pcie_gen2_enqueue_hcmd(struct iwl_trans *trans,
 		cpu_to_le16(cmd_size - sizeof(struct iwl_cmd_header_wide));
 	out_cmd->hdr_wide.reserved = 0;
 	out_cmd->hdr_wide.sequence =
-		cpu_to_le16(QUEUE_TO_SEQ(trans->txqs.cmd.q_id) |
+		cpu_to_le16(QUEUE_TO_SEQ(trans_pcie->txqs.cmd.q_id) |
 					 INDEX_TO_SEQ(txq->write_ptr));
 
 	cmd_pos = sizeof(struct iwl_cmd_header_wide);
@@ -191,7 +1311,7 @@ int iwl_pcie_gen2_enqueue_hcmd(struct iwl_trans *trans,
 		     "Sending command %s (%.2x.%.2x), seq: 0x%04X, %d bytes at %d[%d]:%d\n",
 		     iwl_get_cmd_string(trans, cmd->id), group_id,
 		     out_cmd->hdr.cmd, le16_to_cpu(out_cmd->hdr.sequence),
-		     cmd_size, txq->write_ptr, idx, trans->txqs.cmd.q_id);
+		     cmd_size, txq->write_ptr, idx, trans_pcie->txqs.cmd.q_id);
 
 	/* start the TFD with the minimum copy bytes */
 	tb0_size = min_t(int, copy_size, IWL_FIRST_TB_SIZE);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index fa8eba47dc4c..fd47c4400b4e 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -1,16 +1,22 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2003-2014, 2018-2021, 2023 Intel Corporation
+ * Copyright (C) 2003-2014, 2018-2021, 2023-2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
 #include <linux/etherdevice.h>
 #include <linux/ieee80211.h>
+#include <linux/dmapool.h>
 #include <linux/slab.h>
 #include <linux/sched.h>
+#include <linux/tcp.h>
 #include <net/ip6_checksum.h>
 #include <net/tso.h>
 
+#include "fw/api/commands.h"
+#include "fw/api/datapath.h"
+#include "fw/api/debug.h"
+#include "iwl-fh.h"
 #include "iwl-debug.h"
 #include "iwl-csr.h"
 #include "iwl-prph.h"
@@ -72,6 +78,7 @@ void iwl_pcie_free_dma_ptr(struct iwl_trans *trans, struct iwl_dma_ptr *ptr)
 static void iwl_pcie_txq_inc_wr_ptr(struct iwl_trans *trans,
 				    struct iwl_txq *txq)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	u32 reg = 0;
 	int txq_id = txq->id;
 
@@ -84,7 +91,7 @@ static void iwl_pcie_txq_inc_wr_ptr(struct iwl_trans *trans,
 	 * 3. there is a chance that the NIC is asleep
 	 */
 	if (!trans->trans_cfg->base_params->shadow_reg_enable &&
-	    txq_id != trans->txqs.cmd.q_id &&
+	    txq_id != trans_pcie->txqs.cmd.q_id &&
 	    test_bit(STATUS_TPOWER_PMI, &trans->status)) {
 		/*
 		 * wake up nic if it's powered down ...
@@ -115,12 +122,13 @@ static void iwl_pcie_txq_inc_wr_ptr(struct iwl_trans *trans,
 
 void iwl_pcie_txq_check_wrptrs(struct iwl_trans *trans)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int i;
 
 	for (i = 0; i < trans->trans_cfg->base_params->num_of_queues; i++) {
-		struct iwl_txq *txq = trans->txqs.txq[i];
+		struct iwl_txq *txq = trans_pcie->txqs.txq[i];
 
-		if (!test_bit(i, trans->txqs.queue_used))
+		if (!test_bit(i, trans_pcie->txqs.queue_used))
 			continue;
 
 		spin_lock_bh(&txq->lock);
@@ -132,23 +140,43 @@ void iwl_pcie_txq_check_wrptrs(struct iwl_trans *trans)
 	}
 }
 
+static inline void iwl_pcie_gen1_tfd_set_tb(struct iwl_tfd *tfd,
+					    u8 idx, dma_addr_t addr, u16 len)
+{
+	struct iwl_tfd_tb *tb = &tfd->tbs[idx];
+	u16 hi_n_len = len << 4;
+
+	put_unaligned_le32(addr, &tb->lo);
+	hi_n_len |= iwl_get_dma_hi_addr(addr);
+
+	tb->hi_n_len = cpu_to_le16(hi_n_len);
+
+	tfd->num_tbs = idx + 1;
+}
+
+static inline u8 iwl_txq_gen1_tfd_get_num_tbs(struct iwl_tfd *tfd)
+{
+	return tfd->num_tbs & 0x1f;
+}
+
 static int iwl_pcie_txq_build_tfd(struct iwl_trans *trans, struct iwl_txq *txq,
 				  dma_addr_t addr, u16 len, bool reset)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	void *tfd;
 	u32 num_tbs;
 
-	tfd = (u8 *)txq->tfds + trans->txqs.tfd.size * txq->write_ptr;
+	tfd = (u8 *)txq->tfds + trans_pcie->txqs.tfd.size * txq->write_ptr;
 
 	if (reset)
-		memset(tfd, 0, trans->txqs.tfd.size);
+		memset(tfd, 0, trans_pcie->txqs.tfd.size);
 
-	num_tbs = iwl_txq_gen1_tfd_get_num_tbs(trans, tfd);
+	num_tbs = iwl_txq_gen1_tfd_get_num_tbs(tfd);
 
 	/* Each TFD can point to a maximum max_tbs Tx buffers */
-	if (num_tbs >= trans->txqs.tfd.max_tbs) {
+	if (num_tbs >= trans_pcie->txqs.tfd.max_tbs) {
 		IWL_ERR(trans, "Error can not send more than %d chunks\n",
-			trans->txqs.tfd.max_tbs);
+			trans_pcie->txqs.tfd.max_tbs);
 		return -EINVAL;
 	}
 
@@ -156,7 +184,7 @@ static int iwl_pcie_txq_build_tfd(struct iwl_trans *trans, struct iwl_txq *txq,
 		 "Unaligned address = %llx\n", (unsigned long long)addr))
 		return -EINVAL;
 
-	iwl_pcie_gen1_tfd_set_tb(trans, tfd, num_tbs, addr, len);
+	iwl_pcie_gen1_tfd_set_tb(tfd, num_tbs, addr, len);
 
 	return num_tbs;
 }
@@ -181,12 +209,147 @@ static void iwl_pcie_clear_cmd_in_flight(struct iwl_trans *trans)
 	spin_unlock(&trans_pcie->reg_lock);
 }
 
+void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct page **page_ptr;
+	struct page *next;
+
+	page_ptr = (void *)((u8 *)skb->cb + trans_pcie->txqs.page_offs);
+	next = *page_ptr;
+	*page_ptr = NULL;
+
+	while (next) {
+		struct page *tmp = next;
+
+		next = *(void **)((u8 *)page_address(next) + PAGE_SIZE -
+				  sizeof(void *));
+		__free_page(tmp);
+	}
+}
+
+static inline dma_addr_t
+iwl_txq_gen1_tfd_tb_get_addr(struct iwl_tfd *tfd, u8 idx)
+{
+	struct iwl_tfd_tb *tb = &tfd->tbs[idx];
+	dma_addr_t addr;
+	dma_addr_t hi_len;
+
+	addr = get_unaligned_le32(&tb->lo);
+
+	if (sizeof(dma_addr_t) <= sizeof(u32))
+		return addr;
+
+	hi_len = le16_to_cpu(tb->hi_n_len) & 0xF;
+
+	/*
+	 * shift by 16 twice to avoid warnings on 32-bit
+	 * (where this code never runs anyway due to the
+	 * if statement above)
+	 */
+	return addr | ((hi_len << 16) << 16);
+}
+
+static void iwl_txq_set_tfd_invalid_gen1(struct iwl_trans *trans,
+					 struct iwl_tfd *tfd)
+{
+	tfd->num_tbs = 0;
+
+	iwl_pcie_gen1_tfd_set_tb(tfd, 0, trans->invalid_tx_cmd.dma,
+				 trans->invalid_tx_cmd.size);
+}
+
+static void iwl_txq_gen1_tfd_unmap(struct iwl_trans *trans,
+				   struct iwl_cmd_meta *meta,
+				   struct iwl_txq *txq, int index)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	int i, num_tbs;
+	struct iwl_tfd *tfd = iwl_txq_get_tfd(trans, txq, index);
+
+	/* Sanity check on number of chunks */
+	num_tbs = iwl_txq_gen1_tfd_get_num_tbs(tfd);
+
+	if (num_tbs > trans_pcie->txqs.tfd.max_tbs) {
+		IWL_ERR(trans, "Too many chunks: %i\n", num_tbs);
+		/* @todo issue fatal error, it is quite serious situation */
+		return;
+	}
+
+	/* first TB is never freed - it's the bidirectional DMA data */
+
+	for (i = 1; i < num_tbs; i++) {
+		if (meta->tbs & BIT(i))
+			dma_unmap_page(trans->dev,
+				       iwl_txq_gen1_tfd_tb_get_addr(tfd, i),
+				       iwl_txq_gen1_tfd_tb_get_len(trans,
+								   tfd, i),
+				       DMA_TO_DEVICE);
+		else
+			dma_unmap_single(trans->dev,
+					 iwl_txq_gen1_tfd_tb_get_addr(tfd, i),
+					 iwl_txq_gen1_tfd_tb_get_len(trans,
+								     tfd, i),
+					 DMA_TO_DEVICE);
+	}
+
+	meta->tbs = 0;
+
+	iwl_txq_set_tfd_invalid_gen1(trans, tfd);
+}
+
+/**
+ * iwl_txq_free_tfd - Free all chunks referenced by TFD [txq->q.read_ptr]
+ * @trans: transport private data
+ * @txq: tx queue
+ *
+ * Does NOT advance any TFD circular buffer read/write indexes
+ * Does NOT free the TFD itself (which is within circular buffer)
+ */
+static void iwl_txq_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq)
+{
+	/* rd_ptr is bounded by TFD_QUEUE_SIZE_MAX and
+	 * idx is bounded by n_window
+	 */
+	int rd_ptr = txq->read_ptr;
+	int idx = iwl_txq_get_cmd_index(txq, rd_ptr);
+	struct sk_buff *skb;
+
+	lockdep_assert_held(&txq->lock);
+
+	if (!txq->entries)
+		return;
+
+	/* We have only q->n_window txq->entries, but we use
+	 * TFD_QUEUE_SIZE_MAX tfds
+	 */
+	if (trans->trans_cfg->gen2)
+		iwl_txq_gen2_tfd_unmap(trans, &txq->entries[idx].meta,
+				       iwl_txq_get_tfd(trans, txq, rd_ptr));
+	else
+		iwl_txq_gen1_tfd_unmap(trans, &txq->entries[idx].meta,
+				       txq, rd_ptr);
+
+	/* free SKB */
+	skb = txq->entries[idx].skb;
+
+	/* Can be called from irqs-disabled context
+	 * If skb is not NULL, it means that the whole queue is being
+	 * freed and that the queue is not empty - free the skb
+	 */
+	if (skb) {
+		iwl_op_mode_free_skb(trans->op_mode, skb);
+		txq->entries[idx].skb = NULL;
+	}
+}
+
 /*
  * iwl_pcie_txq_unmap -  Unmap any remaining DMA mappings and free skb's
  */
 static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id)
 {
-	struct iwl_txq *txq = trans->txqs.txq[txq_id];
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
 
 	if (!txq) {
 		IWL_ERR(trans, "Trying to free a queue that wasn't allocated?\n");
@@ -198,19 +361,19 @@ static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id)
 		IWL_DEBUG_TX_REPLY(trans, "Q %d Free %d\n",
 				   txq_id, txq->read_ptr);
 
-		if (txq_id != trans->txqs.cmd.q_id) {
+		if (txq_id != trans_pcie->txqs.cmd.q_id) {
 			struct sk_buff *skb = txq->entries[txq->read_ptr].skb;
 
 			if (WARN_ON_ONCE(!skb))
 				continue;
 
-			iwl_txq_free_tso_page(trans, skb);
+			iwl_pcie_free_tso_page(trans, skb);
 		}
 		iwl_txq_free_tfd(trans, txq);
 		txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr);
 
 		if (txq->read_ptr == txq->write_ptr &&
-		    txq_id == trans->txqs.cmd.q_id)
+		    txq_id == trans_pcie->txqs.cmd.q_id)
 			iwl_pcie_clear_cmd_in_flight(trans);
 	}
 
@@ -223,7 +386,7 @@ static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id)
 	spin_unlock_bh(&txq->lock);
 
 	/* just in case - this queue may have been stopped */
-	iwl_wake_queue(trans, txq);
+	iwl_trans_pcie_wake_queue(trans, txq);
 }
 
 /*
@@ -236,7 +399,8 @@ static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id)
  */
 static void iwl_pcie_txq_free(struct iwl_trans *trans, int txq_id)
 {
-	struct iwl_txq *txq = trans->txqs.txq[txq_id];
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
 	struct device *dev = trans->dev;
 	int i;
 
@@ -246,7 +410,7 @@ static void iwl_pcie_txq_free(struct iwl_trans *trans, int txq_id)
 	iwl_pcie_txq_unmap(trans, txq_id);
 
 	/* De-alloc array of command/tx buffers */
-	if (txq_id == trans->txqs.cmd.q_id)
+	if (txq_id == trans_pcie->txqs.cmd.q_id)
 		for (i = 0; i < txq->n_window; i++) {
 			kfree_sensitive(txq->entries[i].cmd);
 			kfree_sensitive(txq->entries[i].free_buf);
@@ -255,7 +419,7 @@ static void iwl_pcie_txq_free(struct iwl_trans *trans, int txq_id)
 	/* De-alloc circular buffer of TFDs */
 	if (txq->tfds) {
 		dma_free_coherent(dev,
-				  trans->txqs.tfd.size *
+				  trans_pcie->txqs.tfd.size *
 				  trans->trans_cfg->base_params->max_tfd_queue_size,
 				  txq->tfds, txq->dma_addr);
 		txq->dma_addr = 0;
@@ -285,9 +449,10 @@ void iwl_pcie_tx_start(struct iwl_trans *trans, u32 scd_base_addr)
 				SCD_CONTEXT_MEM_LOWER_BOUND) / sizeof(u32);
 
 	/* make sure all queue are not stopped/used */
-	memset(trans->txqs.queue_stopped, 0,
-	       sizeof(trans->txqs.queue_stopped));
-	memset(trans->txqs.queue_used, 0, sizeof(trans->txqs.queue_used));
+	memset(trans_pcie->txqs.queue_stopped, 0,
+	       sizeof(trans_pcie->txqs.queue_stopped));
+	memset(trans_pcie->txqs.queue_used, 0,
+	       sizeof(trans_pcie->txqs.queue_used));
 
 	trans_pcie->scd_base_addr =
 		iwl_read_prph(trans, SCD_SRAM_BASE_ADDR);
@@ -301,7 +466,7 @@ void iwl_pcie_tx_start(struct iwl_trans *trans, u32 scd_base_addr)
 			    NULL, clear_dwords);
 
 	iwl_write_prph(trans, SCD_DRAM_BASE_ADDR,
-		       trans->txqs.scd_bc_tbls.dma >> 10);
+		       trans_pcie->txqs.scd_bc_tbls.dma >> 10);
 
 	/* The chain extension of the SCD doesn't work well. This feature is
 	 * enabled by default by the HW, so we need to disable it manually.
@@ -309,9 +474,9 @@ void iwl_pcie_tx_start(struct iwl_trans *trans, u32 scd_base_addr)
 	if (trans->trans_cfg->base_params->scd_chain_ext_wa)
 		iwl_write_prph(trans, SCD_CHAINEXT_EN, 0);
 
-	iwl_trans_ac_txq_enable(trans, trans->txqs.cmd.q_id,
-				trans->txqs.cmd.fifo,
-				trans->txqs.cmd.wdg_timeout);
+	iwl_trans_ac_txq_enable(trans, trans_pcie->txqs.cmd.q_id,
+				trans_pcie->txqs.cmd.fifo,
+				trans_pcie->txqs.cmd.wdg_timeout);
 
 	/* Activate all Tx DMA/FIFO channels */
 	iwl_scd_activate_fifos(trans);
@@ -347,7 +512,7 @@ void iwl_trans_pcie_tx_reset(struct iwl_trans *trans)
 
 	for (txq_id = 0; txq_id < trans->trans_cfg->base_params->num_of_queues;
 	     txq_id++) {
-		struct iwl_txq *txq = trans->txqs.txq[txq_id];
+		struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
 		if (trans->trans_cfg->gen2)
 			iwl_write_direct64(trans,
 					   FH_MEM_CBBC_QUEUE(trans, txq_id),
@@ -422,9 +587,10 @@ int iwl_pcie_tx_stop(struct iwl_trans *trans)
 	 * queues. This happens when we have an rfkill interrupt.
 	 * Since we stop Tx altogether - mark the queues as stopped.
 	 */
-	memset(trans->txqs.queue_stopped, 0,
-	       sizeof(trans->txqs.queue_stopped));
-	memset(trans->txqs.queue_used, 0, sizeof(trans->txqs.queue_used));
+	memset(trans_pcie->txqs.queue_stopped, 0,
+	       sizeof(trans_pcie->txqs.queue_stopped));
+	memset(trans_pcie->txqs.queue_used, 0,
+	       sizeof(trans_pcie->txqs.queue_used));
 
 	/* This can happen: start_hw, stop_device */
 	if (!trans_pcie->txq_memory)
@@ -448,7 +614,8 @@ void iwl_pcie_tx_free(struct iwl_trans *trans)
 	int txq_id;
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
-	memset(trans->txqs.queue_used, 0, sizeof(trans->txqs.queue_used));
+	memset(trans_pcie->txqs.queue_used, 0,
+	       sizeof(trans_pcie->txqs.queue_used));
 
 	/* Tx queues */
 	if (trans_pcie->txq_memory) {
@@ -456,7 +623,7 @@ void iwl_pcie_tx_free(struct iwl_trans *trans)
 		     txq_id < trans->trans_cfg->base_params->num_of_queues;
 		     txq_id++) {
 			iwl_pcie_txq_free(trans, txq_id);
-			trans->txqs.txq[txq_id] = NULL;
+			trans_pcie->txqs.txq[txq_id] = NULL;
 		}
 	}
 
@@ -465,7 +632,135 @@ void iwl_pcie_tx_free(struct iwl_trans *trans)
 
 	iwl_pcie_free_dma_ptr(trans, &trans_pcie->kw);
 
-	iwl_pcie_free_dma_ptr(trans, &trans->txqs.scd_bc_tbls);
+	iwl_pcie_free_dma_ptr(trans, &trans_pcie->txqs.scd_bc_tbls);
+}
+
+void iwl_txq_log_scd_error(struct iwl_trans *trans, struct iwl_txq *txq)
+{
+	u32 txq_id = txq->id;
+	u32 status;
+	bool active;
+	u8 fifo;
+
+	if (trans->trans_cfg->gen2) {
+		IWL_ERR(trans, "Queue %d is stuck %d %d\n", txq_id,
+			txq->read_ptr, txq->write_ptr);
+		/* TODO: access new SCD registers and dump them */
+		return;
+	}
+
+	status = iwl_read_prph(trans, SCD_QUEUE_STATUS_BITS(txq_id));
+	fifo = (status >> SCD_QUEUE_STTS_REG_POS_TXF) & 0x7;
+	active = !!(status & BIT(SCD_QUEUE_STTS_REG_POS_ACTIVE));
+
+	IWL_ERR(trans,
+		"Queue %d is %sactive on fifo %d and stuck for %u ms. SW [%d, %d] HW [%d, %d] FH TRB=0x0%x\n",
+		txq_id, active ? "" : "in", fifo,
+		jiffies_to_msecs(txq->wd_timeout),
+		txq->read_ptr, txq->write_ptr,
+		iwl_read_prph(trans, SCD_QUEUE_RDPTR(txq_id)) &
+			(trans->trans_cfg->base_params->max_tfd_queue_size - 1),
+			iwl_read_prph(trans, SCD_QUEUE_WRPTR(txq_id)) &
+			(trans->trans_cfg->base_params->max_tfd_queue_size - 1),
+			iwl_read_direct32(trans, FH_TX_TRB_REG(fifo)));
+}
+
+static void iwl_txq_stuck_timer(struct timer_list *t)
+{
+	struct iwl_txq *txq = from_timer(txq, t, stuck_timer);
+	struct iwl_trans *trans = txq->trans;
+
+	spin_lock(&txq->lock);
+	/* check if triggered erroneously */
+	if (txq->read_ptr == txq->write_ptr) {
+		spin_unlock(&txq->lock);
+		return;
+	}
+	spin_unlock(&txq->lock);
+
+	iwl_txq_log_scd_error(trans, txq);
+
+	iwl_force_nmi(trans);
+}
+
+int iwl_pcie_txq_alloc(struct iwl_trans *trans, struct iwl_txq *txq,
+		       int slots_num, bool cmd_queue)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	size_t num_entries = trans->trans_cfg->gen2 ?
+		slots_num : trans->trans_cfg->base_params->max_tfd_queue_size;
+	size_t tfd_sz;
+	size_t tb0_buf_sz;
+	int i;
+
+	if (WARN_ONCE(slots_num <= 0, "Invalid slots num:%d\n", slots_num))
+		return -EINVAL;
+
+	if (WARN_ON(txq->entries || txq->tfds))
+		return -EINVAL;
+
+	tfd_sz = trans_pcie->txqs.tfd.size * num_entries;
+
+	timer_setup(&txq->stuck_timer, iwl_txq_stuck_timer, 0);
+	txq->trans = trans;
+
+	txq->n_window = slots_num;
+
+	txq->entries = kcalloc(slots_num,
+			       sizeof(struct iwl_pcie_txq_entry),
+			       GFP_KERNEL);
+
+	if (!txq->entries)
+		goto error;
+
+	if (cmd_queue)
+		for (i = 0; i < slots_num; i++) {
+			txq->entries[i].cmd =
+				kmalloc(sizeof(struct iwl_device_cmd),
+					GFP_KERNEL);
+			if (!txq->entries[i].cmd)
+				goto error;
+		}
+
+	/* Circular buffer of transmit frame descriptors (TFDs),
+	 * shared with device
+	 */
+	txq->tfds = dma_alloc_coherent(trans->dev, tfd_sz,
+				       &txq->dma_addr, GFP_KERNEL);
+	if (!txq->tfds)
+		goto error;
+
+	BUILD_BUG_ON(sizeof(*txq->first_tb_bufs) != IWL_FIRST_TB_SIZE_ALIGN);
+
+	tb0_buf_sz = sizeof(*txq->first_tb_bufs) * slots_num;
+
+	txq->first_tb_bufs = dma_alloc_coherent(trans->dev, tb0_buf_sz,
+						&txq->first_tb_dma,
+						GFP_KERNEL);
+	if (!txq->first_tb_bufs)
+		goto err_free_tfds;
+
+	for (i = 0; i < num_entries; i++) {
+		void *tfd = iwl_txq_get_tfd(trans, txq, i);
+
+		if (trans->trans_cfg->gen2)
+			iwl_txq_set_tfd_invalid_gen2(trans, tfd);
+		else
+			iwl_txq_set_tfd_invalid_gen1(trans, tfd);
+	}
+
+	return 0;
+err_free_tfds:
+	dma_free_coherent(trans->dev, tfd_sz, txq->tfds, txq->dma_addr);
+	txq->tfds = NULL;
+error:
+	if (txq->entries && cmd_queue)
+		for (i = 0; i < slots_num; i++)
+			kfree(txq->entries[i].cmd);
+	kfree(txq->entries);
+	txq->entries = NULL;
+
+	return -ENOMEM;
 }
 
 /*
@@ -491,7 +786,7 @@ static int iwl_pcie_tx_alloc(struct iwl_trans *trans)
 		goto error;
 	}
 
-	ret = iwl_pcie_alloc_dma_ptr(trans, &trans->txqs.scd_bc_tbls,
+	ret = iwl_pcie_alloc_dma_ptr(trans, &trans_pcie->txqs.scd_bc_tbls,
 				     bc_tbls_size);
 	if (ret) {
 		IWL_ERR(trans, "Scheduler BC Table allocation failed\n");
@@ -517,7 +812,7 @@ static int iwl_pcie_tx_alloc(struct iwl_trans *trans)
 	/* Alloc and init all Tx queues, including the command queue (#4/#9) */
 	for (txq_id = 0; txq_id < trans->trans_cfg->base_params->num_of_queues;
 	     txq_id++) {
-		bool cmd_queue = (txq_id == trans->txqs.cmd.q_id);
+		bool cmd_queue = (txq_id == trans_pcie->txqs.cmd.q_id);
 
 		if (cmd_queue)
 			slots_num = max_t(u32, IWL_CMD_QUEUE_SIZE,
@@ -525,14 +820,14 @@ static int iwl_pcie_tx_alloc(struct iwl_trans *trans)
 		else
 			slots_num = max_t(u32, IWL_DEFAULT_QUEUE_SIZE,
 					  trans->cfg->min_ba_txq_size);
-		trans->txqs.txq[txq_id] = &trans_pcie->txq_memory[txq_id];
-		ret = iwl_txq_alloc(trans, trans->txqs.txq[txq_id], slots_num,
-				    cmd_queue);
+		trans_pcie->txqs.txq[txq_id] = &trans_pcie->txq_memory[txq_id];
+		ret = iwl_pcie_txq_alloc(trans, trans_pcie->txqs.txq[txq_id],
+					 slots_num, cmd_queue);
 		if (ret) {
 			IWL_ERR(trans, "Tx %d queue alloc failed\n", txq_id);
 			goto error;
 		}
-		trans->txqs.txq[txq_id]->id = txq_id;
+		trans_pcie->txqs.txq[txq_id]->id = txq_id;
 	}
 
 	return 0;
@@ -543,6 +838,68 @@ error:
 	return ret;
 }
 
+/*
+ * iwl_queue_init - Initialize queue's high/low-water and read/write indexes
+ */
+static int iwl_queue_init(struct iwl_txq *q, int slots_num)
+{
+	q->n_window = slots_num;
+
+	/* slots_num must be power-of-two size, otherwise
+	 * iwl_txq_get_cmd_index is broken.
+	 */
+	if (WARN_ON(!is_power_of_2(slots_num)))
+		return -EINVAL;
+
+	q->low_mark = q->n_window / 4;
+	if (q->low_mark < 4)
+		q->low_mark = 4;
+
+	q->high_mark = q->n_window / 8;
+	if (q->high_mark < 2)
+		q->high_mark = 2;
+
+	q->write_ptr = 0;
+	q->read_ptr = 0;
+
+	return 0;
+}
+
+int iwl_txq_init(struct iwl_trans *trans, struct iwl_txq *txq,
+		 int slots_num, bool cmd_queue)
+{
+	u32 tfd_queue_max_size =
+		trans->trans_cfg->base_params->max_tfd_queue_size;
+	int ret;
+
+	txq->need_update = false;
+
+	/* max_tfd_queue_size must be power-of-two size, otherwise
+	 * iwl_txq_inc_wrap and iwl_txq_dec_wrap are broken.
+	 */
+	if (WARN_ONCE(tfd_queue_max_size & (tfd_queue_max_size - 1),
+		      "Max tfd queue size must be a power of two, but is %d",
+		      tfd_queue_max_size))
+		return -EINVAL;
+
+	/* Initialize queue's high/low-water marks, and head/tail indexes */
+	ret = iwl_queue_init(txq, slots_num);
+	if (ret)
+		return ret;
+
+	spin_lock_init(&txq->lock);
+
+	if (cmd_queue) {
+		static struct lock_class_key iwl_txq_cmd_queue_lock_class;
+
+		lockdep_set_class(&txq->lock, &iwl_txq_cmd_queue_lock_class);
+	}
+
+	__skb_queue_head_init(&txq->overflow_q);
+
+	return 0;
+}
+
 int iwl_pcie_tx_init(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
@@ -571,7 +928,7 @@ int iwl_pcie_tx_init(struct iwl_trans *trans)
 	/* Alloc and init all Tx queues, including the command queue (#4/#9) */
 	for (txq_id = 0; txq_id < trans->trans_cfg->base_params->num_of_queues;
 	     txq_id++) {
-		bool cmd_queue = (txq_id == trans->txqs.cmd.q_id);
+		bool cmd_queue = (txq_id == trans_pcie->txqs.cmd.q_id);
 
 		if (cmd_queue)
 			slots_num = max_t(u32, IWL_CMD_QUEUE_SIZE,
@@ -579,7 +936,7 @@ int iwl_pcie_tx_init(struct iwl_trans *trans)
 		else
 			slots_num = max_t(u32, IWL_DEFAULT_QUEUE_SIZE,
 					  trans->cfg->min_ba_txq_size);
-		ret = iwl_txq_init(trans, trans->txqs.txq[txq_id], slots_num,
+		ret = iwl_txq_init(trans, trans_pcie->txqs.txq[txq_id], slots_num,
 				   cmd_queue);
 		if (ret) {
 			IWL_ERR(trans, "Tx %d queue init failed\n", txq_id);
@@ -593,7 +950,7 @@ int iwl_pcie_tx_init(struct iwl_trans *trans)
 		 * Circular buffer (TFD queue in DRAM) physical base address
 		 */
 		iwl_write_direct32(trans, FH_MEM_CBBC_QUEUE(trans, txq_id),
-				   trans->txqs.txq[txq_id]->dma_addr >> 8);
+				   trans_pcie->txqs.txq[txq_id]->dma_addr >> 8);
 	}
 
 	iwl_set_bits_prph(trans, SCD_GP_CTRL, SCD_GP_CTRL_AUTO_ACTIVE_MODE);
@@ -641,6 +998,41 @@ static int iwl_pcie_set_cmd_in_flight(struct iwl_trans *trans,
 	return 0;
 }
 
+static void iwl_txq_progress(struct iwl_txq *txq)
+{
+	lockdep_assert_held(&txq->lock);
+
+	if (!txq->wd_timeout)
+		return;
+
+	/*
+	 * station is asleep and we send data - that must
+	 * be uAPSD or PS-Poll. Don't rearm the timer.
+	 */
+	if (txq->frozen)
+		return;
+
+	/*
+	 * if empty delete timer, otherwise move timer forward
+	 * since we're making progress on this queue
+	 */
+	if (txq->read_ptr == txq->write_ptr)
+		del_timer(&txq->stuck_timer);
+	else
+		mod_timer(&txq->stuck_timer, jiffies + txq->wd_timeout);
+}
+
+static inline bool iwl_txq_used(const struct iwl_txq *q, int i)
+{
+	int index = iwl_txq_get_cmd_index(q, i);
+	int r = iwl_txq_get_cmd_index(q, q->read_ptr);
+	int w = iwl_txq_get_cmd_index(q, q->write_ptr);
+
+	return w >= r ?
+		(index >= r && index < w) :
+		!(index < r && index >= w);
+}
+
 /*
  * iwl_pcie_cmdq_reclaim - Reclaim TX command queue entries already Tx'd
  *
@@ -650,7 +1042,8 @@ static int iwl_pcie_set_cmd_in_flight(struct iwl_trans *trans,
  */
 static void iwl_pcie_cmdq_reclaim(struct iwl_trans *trans, int txq_id, int idx)
 {
-	struct iwl_txq *txq = trans->txqs.txq[txq_id];
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
 	int nfreed = 0;
 	u16 r;
 
@@ -661,7 +1054,7 @@ static void iwl_pcie_cmdq_reclaim(struct iwl_trans *trans, int txq_id, int idx)
 
 	if (idx >= trans->trans_cfg->base_params->max_tfd_queue_size ||
 	    (!iwl_txq_used(txq, idx))) {
-		WARN_ONCE(test_bit(txq_id, trans->txqs.queue_used),
+		WARN_ONCE(test_bit(txq_id, trans_pcie->txqs.queue_used),
 			  "%s: Read index for DMA queue txq id (%d), index %d is out of range [0-%d] %d %d.\n",
 			  __func__, txq_id, idx,
 			  trans->trans_cfg->base_params->max_tfd_queue_size,
@@ -720,11 +1113,11 @@ bool iwl_trans_pcie_txq_enable(struct iwl_trans *trans, int txq_id, u16 ssn,
 			       unsigned int wdg_timeout)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_txq *txq = trans->txqs.txq[txq_id];
+	struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
 	int fifo = -1;
 	bool scd_bug = false;
 
-	if (test_and_set_bit(txq_id, trans->txqs.queue_used))
+	if (test_and_set_bit(txq_id, trans_pcie->txqs.queue_used))
 		WARN_ONCE(1, "queue %d already used - expect issues", txq_id);
 
 	txq->wd_timeout = msecs_to_jiffies(wdg_timeout);
@@ -733,7 +1126,7 @@ bool iwl_trans_pcie_txq_enable(struct iwl_trans *trans, int txq_id, u16 ssn,
 		fifo = cfg->fifo;
 
 		/* Disable the scheduler prior configuring the cmd queue */
-		if (txq_id == trans->txqs.cmd.q_id &&
+		if (txq_id == trans_pcie->txqs.cmd.q_id &&
 		    trans_pcie->scd_set_active)
 			iwl_scd_enable_set_active(trans, 0);
 
@@ -741,7 +1134,7 @@ bool iwl_trans_pcie_txq_enable(struct iwl_trans *trans, int txq_id, u16 ssn,
 		iwl_scd_txq_set_inactive(trans, txq_id);
 
 		/* Set this queue as a chain-building queue unless it is CMD */
-		if (txq_id != trans->txqs.cmd.q_id)
+		if (txq_id != trans_pcie->txqs.cmd.q_id)
 			iwl_scd_txq_set_chain(trans, txq_id);
 
 		if (cfg->aggregate) {
@@ -811,7 +1204,7 @@ bool iwl_trans_pcie_txq_enable(struct iwl_trans *trans, int txq_id, u16 ssn,
 			       SCD_QUEUE_STTS_REG_MSK);
 
 		/* enable the scheduler for this queue (only) */
-		if (txq_id == trans->txqs.cmd.q_id &&
+		if (txq_id == trans_pcie->txqs.cmd.q_id &&
 		    trans_pcie->scd_set_active)
 			iwl_scd_enable_set_active(trans, BIT(txq_id));
 
@@ -830,7 +1223,8 @@ bool iwl_trans_pcie_txq_enable(struct iwl_trans *trans, int txq_id, u16 ssn,
 void iwl_trans_pcie_txq_set_shared_mode(struct iwl_trans *trans, u32 txq_id,
 					bool shared_mode)
 {
-	struct iwl_txq *txq = trans->txqs.txq[txq_id];
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
 
 	txq->ampdu = !shared_mode;
 }
@@ -843,8 +1237,8 @@ void iwl_trans_pcie_txq_disable(struct iwl_trans *trans, int txq_id,
 			SCD_TX_STTS_QUEUE_OFFSET(txq_id);
 	static const u32 zero_val[4] = {};
 
-	trans->txqs.txq[txq_id]->frozen_expiry_remainder = 0;
-	trans->txqs.txq[txq_id]->frozen = false;
+	trans_pcie->txqs.txq[txq_id]->frozen_expiry_remainder = 0;
+	trans_pcie->txqs.txq[txq_id]->frozen = false;
 
 	/*
 	 * Upon HW Rfkill - we stop the device, and then stop the queues
@@ -852,7 +1246,7 @@ void iwl_trans_pcie_txq_disable(struct iwl_trans *trans, int txq_id,
 	 * allow the op_mode to call txq_disable after it already called
 	 * stop_device.
 	 */
-	if (!test_and_clear_bit(txq_id, trans->txqs.queue_used)) {
+	if (!test_and_clear_bit(txq_id, trans_pcie->txqs.queue_used)) {
 		WARN_ONCE(test_bit(STATUS_DEVICE_ENABLED, &trans->status),
 			  "queue %d not used", txq_id);
 		return;
@@ -866,7 +1260,7 @@ void iwl_trans_pcie_txq_disable(struct iwl_trans *trans, int txq_id,
 	}
 
 	iwl_pcie_txq_unmap(trans, txq_id);
-	trans->txqs.txq[txq_id]->ampdu = false;
+	trans_pcie->txqs.txq[txq_id]->ampdu = false;
 
 	IWL_DEBUG_TX_QUEUES(trans, "Deactivate queue %d\n", txq_id);
 }
@@ -875,12 +1269,13 @@ void iwl_trans_pcie_txq_disable(struct iwl_trans *trans, int txq_id,
 
 static void iwl_trans_pcie_block_txq_ptrs(struct iwl_trans *trans, bool block)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int i;
 
 	for (i = 0; i < trans->trans_cfg->base_params->num_of_queues; i++) {
-		struct iwl_txq *txq = trans->txqs.txq[i];
+		struct iwl_txq *txq = trans_pcie->txqs.txq[i];
 
-		if (i == trans->txqs.cmd.q_id)
+		if (i == trans_pcie->txqs.cmd.q_id)
 			continue;
 
 		/* we skip the command queue (obviously) so it's OK to nest */
@@ -912,7 +1307,8 @@ static void iwl_trans_pcie_block_txq_ptrs(struct iwl_trans *trans, bool block)
 int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 			  struct iwl_host_cmd *cmd)
 {
-	struct iwl_txq *txq = trans->txqs.txq[trans->txqs.cmd.q_id];
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *txq = trans_pcie->txqs.txq[trans_pcie->txqs.cmd.q_id];
 	struct iwl_device_cmd *out_cmd;
 	struct iwl_cmd_meta *out_meta;
 	void *dup_buf = NULL;
@@ -1038,7 +1434,7 @@ int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 				    sizeof(struct iwl_cmd_header_wide));
 		out_cmd->hdr_wide.reserved = 0;
 		out_cmd->hdr_wide.sequence =
-			cpu_to_le16(QUEUE_TO_SEQ(trans->txqs.cmd.q_id) |
+			cpu_to_le16(QUEUE_TO_SEQ(trans_pcie->txqs.cmd.q_id) |
 						 INDEX_TO_SEQ(txq->write_ptr));
 
 		cmd_pos = sizeof(struct iwl_cmd_header_wide);
@@ -1046,7 +1442,7 @@ int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 	} else {
 		out_cmd->hdr.cmd = iwl_cmd_opcode(cmd->id);
 		out_cmd->hdr.sequence =
-			cpu_to_le16(QUEUE_TO_SEQ(trans->txqs.cmd.q_id) |
+			cpu_to_le16(QUEUE_TO_SEQ(trans_pcie->txqs.cmd.q_id) |
 						 INDEX_TO_SEQ(txq->write_ptr));
 		out_cmd->hdr.group_id = 0;
 
@@ -1097,7 +1493,7 @@ int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 		     iwl_get_cmd_string(trans, cmd->id),
 		     group_id, out_cmd->hdr.cmd,
 		     le16_to_cpu(out_cmd->hdr.sequence),
-		     cmd_size, txq->write_ptr, idx, trans->txqs.cmd.q_id);
+		     cmd_size, txq->write_ptr, idx, trans_pcie->txqs.cmd.q_id);
 
 	/* start the TFD with the minimum copy bytes */
 	tb0_size = min_t(int, copy_size, IWL_FIRST_TB_SIZE);
@@ -1196,14 +1592,14 @@ void iwl_pcie_hcmd_complete(struct iwl_trans *trans,
 	struct iwl_device_cmd *cmd;
 	struct iwl_cmd_meta *meta;
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	struct iwl_txq *txq = trans->txqs.txq[trans->txqs.cmd.q_id];
+	struct iwl_txq *txq = trans_pcie->txqs.txq[trans_pcie->txqs.cmd.q_id];
 
 	/* If a Tx command is being handled and it isn't in the actual
 	 * command queue then there a command routing bug has been introduced
 	 * in the queue management code. */
-	if (WARN(txq_id != trans->txqs.cmd.q_id,
+	if (WARN(txq_id != trans_pcie->txqs.cmd.q_id,
 		 "wrong command queue %d (should be %d), sequence 0x%X readp=%d writep=%d\n",
-		 txq_id, trans->txqs.cmd.q_id, sequence, txq->read_ptr,
+		 txq_id, trans_pcie->txqs.cmd.q_id, sequence, txq->read_ptr,
 		 txq->write_ptr)) {
 		iwl_print_hex_error(trans, pkt, 32);
 		return;
@@ -1306,12 +1702,58 @@ static int iwl_fill_data_tbs(struct iwl_trans *trans, struct sk_buff *skb,
 }
 
 #ifdef CONFIG_INET
+struct iwl_tso_hdr_page *iwl_pcie_get_page_hdr(struct iwl_trans *trans,
+					       size_t len, struct sk_buff *skb)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_tso_hdr_page *p = this_cpu_ptr(trans_pcie->txqs.tso_hdr_page);
+	struct page **page_ptr;
+
+	page_ptr = (void *)((u8 *)skb->cb + trans_pcie->txqs.page_offs);
+
+	if (WARN_ON(*page_ptr))
+		return NULL;
+
+	if (!p->page)
+		goto alloc;
+
+	/*
+	 * Check if there's enough room on this page
+	 *
+	 * Note that we put a page chaining pointer *last* in the
+	 * page - we need it somewhere, and if it's there then we
+	 * avoid DMA mapping the last bits of the page which may
+	 * trigger the 32-bit boundary hardware bug.
+	 *
+	 * (see also get_workaround_page() in tx-gen2.c)
+	 */
+	if (p->pos + len < (u8 *)page_address(p->page) + PAGE_SIZE -
+			   sizeof(void *))
+		goto out;
+
+	/* We don't have enough room on this page, get a new one. */
+	__free_page(p->page);
+
+alloc:
+	p->page = alloc_page(GFP_ATOMIC);
+	if (!p->page)
+		return NULL;
+	p->pos = page_address(p->page);
+	/* set the chaining pointer to NULL */
+	*(void **)((u8 *)page_address(p->page) + PAGE_SIZE - sizeof(void *)) = NULL;
+out:
+	*page_ptr = p->page;
+	get_page(p->page);
+	return p;
+}
+
 static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 				   struct iwl_txq *txq, u8 hdr_len,
 				   struct iwl_cmd_meta *out_meta,
 				   struct iwl_device_tx_cmd *dev_cmd,
 				   u16 tb1_len)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_tx_cmd *tx_cmd = (void *)dev_cmd->payload;
 	struct ieee80211_hdr *hdr = (void *)skb->data;
 	unsigned int snap_ip_tcp_hdrlen, ip_hdrlen, total_len, hdr_room;
@@ -1328,7 +1770,7 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 
 	trace_iwlwifi_dev_tx(trans->dev, skb,
 			     iwl_txq_get_tfd(trans, txq, txq->write_ptr),
-			     trans->txqs.tfd.size,
+			     trans_pcie->txqs.tfd.size,
 			     &dev_cmd->hdr, IWL_FIRST_TB_SIZE + tb1_len, 0);
 
 	ip_hdrlen = skb_network_header_len(skb);
@@ -1341,7 +1783,7 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 		(3 + snap_ip_tcp_hdrlen + sizeof(struct ethhdr)) + iv_len;
 
 	/* Our device supports 9 segments at most, it will fit in 1 page */
-	hdr_page = get_page_hdr(trans, hdr_room, skb);
+	hdr_page = iwl_pcie_get_page_hdr(trans, hdr_room, skb);
 	if (!hdr_page)
 		return -ENOMEM;
 
@@ -1450,9 +1892,61 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 }
 #endif /* CONFIG_INET */
 
+#define IWL_TX_CRC_SIZE 4
+#define IWL_TX_DELIMITER_SIZE 4
+
+/*
+ * iwl_txq_gen1_update_byte_cnt_tbl - Set up entry in Tx byte-count array
+ */
+static void iwl_txq_gen1_update_byte_cnt_tbl(struct iwl_trans *trans,
+					     struct iwl_txq *txq, u16 byte_cnt,
+					     int num_tbs)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwlagn_scd_bc_tbl *scd_bc_tbl;
+	int write_ptr = txq->write_ptr;
+	int txq_id = txq->id;
+	u8 sec_ctl = 0;
+	u16 len = byte_cnt + IWL_TX_CRC_SIZE + IWL_TX_DELIMITER_SIZE;
+	__le16 bc_ent;
+	struct iwl_device_tx_cmd *dev_cmd = txq->entries[txq->write_ptr].cmd;
+	struct iwl_tx_cmd *tx_cmd = (void *)dev_cmd->payload;
+	u8 sta_id = tx_cmd->sta_id;
+
+	scd_bc_tbl = trans_pcie->txqs.scd_bc_tbls.addr;
+
+	sec_ctl = tx_cmd->sec_ctl;
+
+	switch (sec_ctl & TX_CMD_SEC_MSK) {
+	case TX_CMD_SEC_CCM:
+		len += IEEE80211_CCMP_MIC_LEN;
+		break;
+	case TX_CMD_SEC_TKIP:
+		len += IEEE80211_TKIP_ICV_LEN;
+		break;
+	case TX_CMD_SEC_WEP:
+		len += IEEE80211_WEP_IV_LEN + IEEE80211_WEP_ICV_LEN;
+		break;
+	}
+	if (trans_pcie->txqs.bc_table_dword)
+		len = DIV_ROUND_UP(len, 4);
+
+	if (WARN_ON(len > 0xFFF || write_ptr >= TFD_QUEUE_SIZE_MAX))
+		return;
+
+	bc_ent = cpu_to_le16(len | (sta_id << 12));
+
+	scd_bc_tbl[txq_id].tfd_offset[write_ptr] = bc_ent;
+
+	if (write_ptr < TFD_QUEUE_SIZE_BC_DUP)
+		scd_bc_tbl[txq_id].tfd_offset[TFD_QUEUE_SIZE_MAX + write_ptr] =
+			bc_ent;
+}
+
 int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 		      struct iwl_device_tx_cmd *dev_cmd, int txq_id)
 {
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct ieee80211_hdr *hdr;
 	struct iwl_tx_cmd *tx_cmd = (struct iwl_tx_cmd *)dev_cmd->payload;
 	struct iwl_cmd_meta *out_meta;
@@ -1467,14 +1961,14 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 	u16 wifi_seq;
 	bool amsdu;
 
-	txq = trans->txqs.txq[txq_id];
+	txq = trans_pcie->txqs.txq[txq_id];
 
-	if (WARN_ONCE(!test_bit(txq_id, trans->txqs.queue_used),
+	if (WARN_ONCE(!test_bit(txq_id, trans_pcie->txqs.queue_used),
 		      "TX on unused queue %d\n", txq_id))
 		return -EINVAL;
 
 	if (skb_is_nonlinear(skb) &&
-	    skb_shinfo(skb)->nr_frags > IWL_TRANS_MAX_FRAGS(trans) &&
+	    skb_shinfo(skb)->nr_frags > IWL_TRANS_PCIE_MAX_FRAGS(trans_pcie) &&
 	    __skb_linearize(skb))
 		return -ENOMEM;
 
@@ -1495,7 +1989,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 			struct iwl_device_tx_cmd **dev_cmd_ptr;
 
 			dev_cmd_ptr = (void *)((u8 *)skb->cb +
-					       trans->txqs.dev_cmd_offs);
+					       trans_pcie->txqs.dev_cmd_offs);
 
 			*dev_cmd_ptr = dev_cmd;
 			__skb_queue_tail(&txq->overflow_q, skb);
@@ -1578,7 +2072,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 
 	trace_iwlwifi_dev_tx(trans->dev, skb,
 			     iwl_txq_get_tfd(trans, txq, txq->write_ptr),
-			     trans->txqs.tfd.size,
+			     trans_pcie->txqs.tfd.size,
 			     &dev_cmd->hdr, IWL_FIRST_TB_SIZE + tb1_len,
 			     hdr_len);
 
@@ -1613,8 +2107,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 	tfd = iwl_txq_get_tfd(trans, txq, txq->write_ptr);
 	/* Set up entry for this TFD in Tx byte-count array */
 	iwl_txq_gen1_update_byte_cnt_tbl(trans, txq, le16_to_cpu(tx_cmd->len),
-					 iwl_txq_gen1_tfd_get_num_tbs(trans,
-								      tfd));
+					 iwl_txq_gen1_tfd_get_num_tbs(tfd));
 
 	wait_write_ptr = ieee80211_has_morefrags(fc);
 
@@ -1649,3 +2142,359 @@ out_err:
 	spin_unlock(&txq->lock);
 	return -1;
 }
+
+static void iwl_txq_gen1_inval_byte_cnt_tbl(struct iwl_trans *trans,
+					    struct iwl_txq *txq)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwlagn_scd_bc_tbl *scd_bc_tbl = trans_pcie->txqs.scd_bc_tbls.addr;
+	int txq_id = txq->id;
+	int read_ptr = txq->read_ptr;
+	u8 sta_id = 0;
+	__le16 bc_ent;
+	struct iwl_device_tx_cmd *dev_cmd = txq->entries[read_ptr].cmd;
+	struct iwl_tx_cmd *tx_cmd = (void *)dev_cmd->payload;
+
+	WARN_ON(read_ptr >= TFD_QUEUE_SIZE_MAX);
+
+	if (txq_id != trans_pcie->txqs.cmd.q_id)
+		sta_id = tx_cmd->sta_id;
+
+	bc_ent = cpu_to_le16(1 | (sta_id << 12));
+
+	scd_bc_tbl[txq_id].tfd_offset[read_ptr] = bc_ent;
+
+	if (read_ptr < TFD_QUEUE_SIZE_BC_DUP)
+		scd_bc_tbl[txq_id].tfd_offset[TFD_QUEUE_SIZE_MAX + read_ptr] =
+			bc_ent;
+}
+
+/* Frees buffers until index _not_ inclusive */
+void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
+		      struct sk_buff_head *skbs, bool is_flush)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
+	int tfd_num, read_ptr, last_to_free;
+
+	/* This function is not meant to release cmd queue*/
+	if (WARN_ON(txq_id == trans_pcie->txqs.cmd.q_id))
+		return;
+
+	if (WARN_ON(!txq))
+		return;
+
+	tfd_num = iwl_txq_get_cmd_index(txq, ssn);
+
+	spin_lock_bh(&txq->lock);
+	read_ptr = iwl_txq_get_cmd_index(txq, txq->read_ptr);
+
+	if (!test_bit(txq_id, trans_pcie->txqs.queue_used)) {
+		IWL_DEBUG_TX_QUEUES(trans, "Q %d inactive - ignoring idx %d\n",
+				    txq_id, ssn);
+		goto out;
+	}
+
+	if (read_ptr == tfd_num)
+		goto out;
+
+	IWL_DEBUG_TX_REPLY(trans, "[Q %d] %d (%d) -> %d (%d)\n",
+			   txq_id, read_ptr, txq->read_ptr, tfd_num, ssn);
+
+	/* Since we free until index _not_ inclusive, the one before index is
+	 * the last we will free. This one must be used
+	 */
+	last_to_free = iwl_txq_dec_wrap(trans, tfd_num);
+
+	if (!iwl_txq_used(txq, last_to_free)) {
+		IWL_ERR(trans,
+			"%s: Read index for txq id (%d), last_to_free %d is out of range [0-%d] %d %d.\n",
+			__func__, txq_id, last_to_free,
+			trans->trans_cfg->base_params->max_tfd_queue_size,
+			txq->write_ptr, txq->read_ptr);
+
+		iwl_op_mode_time_point(trans->op_mode,
+				       IWL_FW_INI_TIME_POINT_FAKE_TX,
+				       NULL);
+		goto out;
+	}
+
+	if (WARN_ON(!skb_queue_empty(skbs)))
+		goto out;
+
+	for (;
+	     read_ptr != tfd_num;
+	     txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr),
+	     read_ptr = iwl_txq_get_cmd_index(txq, txq->read_ptr)) {
+		struct sk_buff *skb = txq->entries[read_ptr].skb;
+
+		if (WARN_ONCE(!skb, "no SKB at %d (%d) on queue %d\n",
+			      read_ptr, txq->read_ptr, txq_id))
+			continue;
+
+		iwl_pcie_free_tso_page(trans, skb);
+
+		__skb_queue_tail(skbs, skb);
+
+		txq->entries[read_ptr].skb = NULL;
+
+		if (!trans->trans_cfg->gen2)
+			iwl_txq_gen1_inval_byte_cnt_tbl(trans, txq);
+
+		iwl_txq_free_tfd(trans, txq);
+	}
+
+	iwl_txq_progress(txq);
+
+	if (iwl_txq_space(trans, txq) > txq->low_mark &&
+	    test_bit(txq_id, trans_pcie->txqs.queue_stopped)) {
+		struct sk_buff_head overflow_skbs;
+		struct sk_buff *skb;
+
+		__skb_queue_head_init(&overflow_skbs);
+		skb_queue_splice_init(&txq->overflow_q,
+				      is_flush ? skbs : &overflow_skbs);
+
+		/*
+		 * We are going to transmit from the overflow queue.
+		 * Remember this state so that wait_for_txq_empty will know we
+		 * are adding more packets to the TFD queue. It cannot rely on
+		 * the state of &txq->overflow_q, as we just emptied it, but
+		 * haven't TXed the content yet.
+		 */
+		txq->overflow_tx = true;
+
+		/*
+		 * This is tricky: we are in reclaim path which is non
+		 * re-entrant, so noone will try to take the access the
+		 * txq data from that path. We stopped tx, so we can't
+		 * have tx as well. Bottom line, we can unlock and re-lock
+		 * later.
+		 */
+		spin_unlock_bh(&txq->lock);
+
+		while ((skb = __skb_dequeue(&overflow_skbs))) {
+			struct iwl_device_tx_cmd *dev_cmd_ptr;
+
+			dev_cmd_ptr = *(void **)((u8 *)skb->cb +
+						 trans_pcie->txqs.dev_cmd_offs);
+
+			/*
+			 * Note that we can very well be overflowing again.
+			 * In that case, iwl_txq_space will be small again
+			 * and we won't wake mac80211's queue.
+			 */
+			iwl_trans_tx(trans, skb, dev_cmd_ptr, txq_id);
+		}
+
+		if (iwl_txq_space(trans, txq) > txq->low_mark)
+			iwl_trans_pcie_wake_queue(trans, txq);
+
+		spin_lock_bh(&txq->lock);
+		txq->overflow_tx = false;
+	}
+
+out:
+	spin_unlock_bh(&txq->lock);
+}
+
+/* Set wr_ptr of specific device and txq  */
+void iwl_pcie_set_q_ptrs(struct iwl_trans *trans, int txq_id, int ptr)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
+
+	spin_lock_bh(&txq->lock);
+
+	txq->write_ptr = ptr;
+	txq->read_ptr = txq->write_ptr;
+
+	spin_unlock_bh(&txq->lock);
+}
+
+void iwl_pcie_freeze_txq_timer(struct iwl_trans *trans,
+			       unsigned long txqs, bool freeze)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	int queue;
+
+	for_each_set_bit(queue, &txqs, BITS_PER_LONG) {
+		struct iwl_txq *txq = trans_pcie->txqs.txq[queue];
+		unsigned long now;
+
+		spin_lock_bh(&txq->lock);
+
+		now = jiffies;
+
+		if (txq->frozen == freeze)
+			goto next_queue;
+
+		IWL_DEBUG_TX_QUEUES(trans, "%s TXQ %d\n",
+				    freeze ? "Freezing" : "Waking", queue);
+
+		txq->frozen = freeze;
+
+		if (txq->read_ptr == txq->write_ptr)
+			goto next_queue;
+
+		if (freeze) {
+			if (unlikely(time_after(now,
+						txq->stuck_timer.expires))) {
+				/*
+				 * The timer should have fired, maybe it is
+				 * spinning right now on the lock.
+				 */
+				goto next_queue;
+			}
+			/* remember how long until the timer fires */
+			txq->frozen_expiry_remainder =
+				txq->stuck_timer.expires - now;
+			del_timer(&txq->stuck_timer);
+			goto next_queue;
+		}
+
+		/*
+		 * Wake a non-empty queue -> arm timer with the
+		 * remainder before it froze
+		 */
+		mod_timer(&txq->stuck_timer,
+			  now + txq->frozen_expiry_remainder);
+
+next_queue:
+		spin_unlock_bh(&txq->lock);
+	}
+}
+
+#define HOST_COMPLETE_TIMEOUT	(2 * HZ)
+
+static int iwl_trans_pcie_send_hcmd_sync(struct iwl_trans *trans,
+					 struct iwl_host_cmd *cmd)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	const char *cmd_str = iwl_get_cmd_string(trans, cmd->id);
+	struct iwl_txq *txq = trans_pcie->txqs.txq[trans_pcie->txqs.cmd.q_id];
+	int cmd_idx;
+	int ret;
+
+	IWL_DEBUG_INFO(trans, "Attempting to send sync command %s\n", cmd_str);
+
+	if (WARN(test_and_set_bit(STATUS_SYNC_HCMD_ACTIVE,
+				  &trans->status),
+		 "Command %s: a command is already active!\n", cmd_str))
+		return -EIO;
+
+	IWL_DEBUG_INFO(trans, "Setting HCMD_ACTIVE for command %s\n", cmd_str);
+
+	cmd_idx = trans->ops->send_cmd(trans, cmd);
+	if (cmd_idx < 0) {
+		ret = cmd_idx;
+		clear_bit(STATUS_SYNC_HCMD_ACTIVE, &trans->status);
+		IWL_ERR(trans, "Error sending %s: enqueue_hcmd failed: %d\n",
+			cmd_str, ret);
+		return ret;
+	}
+
+	ret = wait_event_timeout(trans->wait_command_queue,
+				 !test_bit(STATUS_SYNC_HCMD_ACTIVE,
+					   &trans->status),
+				 HOST_COMPLETE_TIMEOUT);
+	if (!ret) {
+		IWL_ERR(trans, "Error sending %s: time out after %dms.\n",
+			cmd_str, jiffies_to_msecs(HOST_COMPLETE_TIMEOUT));
+
+		IWL_ERR(trans, "Current CMD queue read_ptr %d write_ptr %d\n",
+			txq->read_ptr, txq->write_ptr);
+
+		clear_bit(STATUS_SYNC_HCMD_ACTIVE, &trans->status);
+		IWL_DEBUG_INFO(trans, "Clearing HCMD_ACTIVE for command %s\n",
+			       cmd_str);
+		ret = -ETIMEDOUT;
+
+		iwl_trans_sync_nmi(trans);
+		goto cancel;
+	}
+
+	if (test_bit(STATUS_FW_ERROR, &trans->status)) {
+		if (!test_and_clear_bit(STATUS_SUPPRESS_CMD_ERROR_ONCE,
+					&trans->status)) {
+			IWL_ERR(trans, "FW error in SYNC CMD %s\n", cmd_str);
+			dump_stack();
+		}
+		ret = -EIO;
+		goto cancel;
+	}
+
+	if (!(cmd->flags & CMD_SEND_IN_RFKILL) &&
+	    test_bit(STATUS_RFKILL_OPMODE, &trans->status)) {
+		IWL_DEBUG_RF_KILL(trans, "RFKILL in SYNC CMD... no rsp\n");
+		ret = -ERFKILL;
+		goto cancel;
+	}
+
+	if ((cmd->flags & CMD_WANT_SKB) && !cmd->resp_pkt) {
+		IWL_ERR(trans, "Error: Response NULL in '%s'\n", cmd_str);
+		ret = -EIO;
+		goto cancel;
+	}
+
+	return 0;
+
+cancel:
+	if (cmd->flags & CMD_WANT_SKB) {
+		/*
+		 * Cancel the CMD_WANT_SKB flag for the cmd in the
+		 * TX cmd queue. Otherwise in case the cmd comes
+		 * in later, it will possibly set an invalid
+		 * address (cmd->meta.source).
+		 */
+		txq->entries[cmd_idx].meta.flags &= ~CMD_WANT_SKB;
+	}
+
+	if (cmd->resp_pkt) {
+		iwl_free_resp(cmd);
+		cmd->resp_pkt = NULL;
+	}
+
+	return ret;
+}
+
+int iwl_trans_pcie_send_hcmd(struct iwl_trans *trans,
+			     struct iwl_host_cmd *cmd)
+{
+	/* Make sure the NIC is still alive in the bus */
+	if (test_bit(STATUS_TRANS_DEAD, &trans->status))
+		return -ENODEV;
+
+	if (!(cmd->flags & CMD_SEND_IN_RFKILL) &&
+	    test_bit(STATUS_RFKILL_OPMODE, &trans->status)) {
+		IWL_DEBUG_RF_KILL(trans, "Dropping CMD 0x%x: RF KILL\n",
+				  cmd->id);
+		return -ERFKILL;
+	}
+
+	if (unlikely(trans->system_pm_mode == IWL_PLAT_PM_MODE_D3 &&
+		     !(cmd->flags & CMD_SEND_IN_D3))) {
+		IWL_DEBUG_WOWLAN(trans, "Dropping CMD 0x%x: D3\n", cmd->id);
+		return -EHOSTDOWN;
+	}
+
+	if (cmd->flags & CMD_ASYNC) {
+		int ret;
+
+		/* An asynchronous command can not expect an SKB to be set. */
+		if (WARN_ON(cmd->flags & CMD_WANT_SKB))
+			return -EINVAL;
+
+		ret = trans->ops->send_cmd(trans, cmd);
+		if (ret < 0) {
+			IWL_ERR(trans,
+				"Error sending %s: enqueue_hcmd failed: %d\n",
+				iwl_get_cmd_string(trans, cmd->id), ret);
+			return ret;
+		}
+		return 0;
+	}
+
+	return iwl_trans_pcie_send_hcmd_sync(trans, cmd);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_pcie_send_hcmd);
diff --git a/drivers/net/wireless/intel/iwlwifi/queue/tx.c b/drivers/net/wireless/intel/iwlwifi/queue/tx.c
deleted file mode 100644
index 726035dfa5cd..000000000000
--- a/drivers/net/wireless/intel/iwlwifi/queue/tx.c
+++ /dev/null
@@ -1,1918 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
-/*
- * Copyright (C) 2020-2024 Intel Corporation
- */
-#include <net/tso.h>
-#include <linux/tcp.h>
-
-#include "iwl-debug.h"
-#include "iwl-io.h"
-#include "fw/api/commands.h"
-#include "fw/api/tx.h"
-#include "fw/api/datapath.h"
-#include "fw/api/debug.h"
-#include "queue/tx.h"
-#include "iwl-fh.h"
-#include "iwl-scd.h"
-#include <linux/dmapool.h>
-
-/*
- * iwl_txq_update_byte_tbl - Set up entry in Tx byte-count array
- */
-static void iwl_pcie_gen2_update_byte_tbl(struct iwl_trans *trans,
-					  struct iwl_txq *txq, u16 byte_cnt,
-					  int num_tbs)
-{
-	int idx = iwl_txq_get_cmd_index(txq, txq->write_ptr);
-	u8 filled_tfd_size, num_fetch_chunks;
-	u16 len = byte_cnt;
-	__le16 bc_ent;
-
-	if (WARN(idx >= txq->n_window, "%d >= %d\n", idx, txq->n_window))
-		return;
-
-	filled_tfd_size = offsetof(struct iwl_tfh_tfd, tbs) +
-			  num_tbs * sizeof(struct iwl_tfh_tb);
-	/*
-	 * filled_tfd_size contains the number of filled bytes in the TFD.
-	 * Dividing it by 64 will give the number of chunks to fetch
-	 * to SRAM- 0 for one chunk, 1 for 2 and so on.
-	 * If, for example, TFD contains only 3 TBs then 32 bytes
-	 * of the TFD are used, and only one chunk of 64 bytes should
-	 * be fetched
-	 */
-	num_fetch_chunks = DIV_ROUND_UP(filled_tfd_size, 64) - 1;
-
-	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_AX210) {
-		struct iwl_gen3_bc_tbl_entry *scd_bc_tbl_gen3 = txq->bc_tbl.addr;
-
-		/* Starting from AX210, the HW expects bytes */
-		WARN_ON(trans->txqs.bc_table_dword);
-		WARN_ON(len > 0x3FFF);
-		bc_ent = cpu_to_le16(len | (num_fetch_chunks << 14));
-		scd_bc_tbl_gen3[idx].tfd_offset = bc_ent;
-	} else {
-		struct iwlagn_scd_bc_tbl *scd_bc_tbl = txq->bc_tbl.addr;
-
-		/* Before AX210, the HW expects DW */
-		WARN_ON(!trans->txqs.bc_table_dword);
-		len = DIV_ROUND_UP(len, 4);
-		WARN_ON(len > 0xFFF);
-		bc_ent = cpu_to_le16(len | (num_fetch_chunks << 12));
-		scd_bc_tbl->tfd_offset[idx] = bc_ent;
-	}
-}
-
-/*
- * iwl_txq_inc_wr_ptr - Send new write index to hardware
- */
-void iwl_txq_inc_wr_ptr(struct iwl_trans *trans, struct iwl_txq *txq)
-{
-	lockdep_assert_held(&txq->lock);
-
-	IWL_DEBUG_TX(trans, "Q:%d WR: 0x%x\n", txq->id, txq->write_ptr);
-
-	/*
-	 * if not in power-save mode, uCode will never sleep when we're
-	 * trying to tx (during RFKILL, we're not trying to tx).
-	 */
-	iwl_write32(trans, HBUS_TARG_WRPTR, txq->write_ptr | (txq->id << 16));
-}
-
-static u8 iwl_txq_gen2_get_num_tbs(struct iwl_trans *trans,
-				   struct iwl_tfh_tfd *tfd)
-{
-	return le16_to_cpu(tfd->num_tbs) & 0x1f;
-}
-
-int iwl_txq_gen2_set_tb(struct iwl_trans *trans, struct iwl_tfh_tfd *tfd,
-			dma_addr_t addr, u16 len)
-{
-	int idx = iwl_txq_gen2_get_num_tbs(trans, tfd);
-	struct iwl_tfh_tb *tb;
-
-	/* Only WARN here so we know about the issue, but we mess up our
-	 * unmap path because not every place currently checks for errors
-	 * returned from this function - it can only return an error if
-	 * there's no more space, and so when we know there is enough we
-	 * don't always check ...
-	 */
-	WARN(iwl_txq_crosses_4g_boundary(addr, len),
-	     "possible DMA problem with iova:0x%llx, len:%d\n",
-	     (unsigned long long)addr, len);
-
-	if (WARN_ON(idx >= IWL_TFH_NUM_TBS))
-		return -EINVAL;
-	tb = &tfd->tbs[idx];
-
-	/* Each TFD can point to a maximum max_tbs Tx buffers */
-	if (le16_to_cpu(tfd->num_tbs) >= trans->txqs.tfd.max_tbs) {
-		IWL_ERR(trans, "Error can not send more than %d chunks\n",
-			trans->txqs.tfd.max_tbs);
-		return -EINVAL;
-	}
-
-	put_unaligned_le64(addr, &tb->addr);
-	tb->tb_len = cpu_to_le16(len);
-
-	tfd->num_tbs = cpu_to_le16(idx + 1);
-
-	return idx;
-}
-
-static void iwl_txq_set_tfd_invalid_gen2(struct iwl_trans *trans,
-					 struct iwl_tfh_tfd *tfd)
-{
-	tfd->num_tbs = 0;
-
-	iwl_txq_gen2_set_tb(trans, tfd, trans->invalid_tx_cmd.dma,
-			    trans->invalid_tx_cmd.size);
-}
-
-void iwl_txq_gen2_tfd_unmap(struct iwl_trans *trans, struct iwl_cmd_meta *meta,
-			    struct iwl_tfh_tfd *tfd)
-{
-	int i, num_tbs;
-
-	/* Sanity check on number of chunks */
-	num_tbs = iwl_txq_gen2_get_num_tbs(trans, tfd);
-
-	if (num_tbs > trans->txqs.tfd.max_tbs) {
-		IWL_ERR(trans, "Too many chunks: %i\n", num_tbs);
-		return;
-	}
-
-	/* first TB is never freed - it's the bidirectional DMA data */
-	for (i = 1; i < num_tbs; i++) {
-		if (meta->tbs & BIT(i))
-			dma_unmap_page(trans->dev,
-				       le64_to_cpu(tfd->tbs[i].addr),
-				       le16_to_cpu(tfd->tbs[i].tb_len),
-				       DMA_TO_DEVICE);
-		else
-			dma_unmap_single(trans->dev,
-					 le64_to_cpu(tfd->tbs[i].addr),
-					 le16_to_cpu(tfd->tbs[i].tb_len),
-					 DMA_TO_DEVICE);
-	}
-
-	iwl_txq_set_tfd_invalid_gen2(trans, tfd);
-}
-
-void iwl_txq_gen2_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq)
-{
-	/* rd_ptr is bounded by TFD_QUEUE_SIZE_MAX and
-	 * idx is bounded by n_window
-	 */
-	int idx = iwl_txq_get_cmd_index(txq, txq->read_ptr);
-	struct sk_buff *skb;
-
-	lockdep_assert_held(&txq->lock);
-
-	if (!txq->entries)
-		return;
-
-	iwl_txq_gen2_tfd_unmap(trans, &txq->entries[idx].meta,
-			       iwl_txq_get_tfd(trans, txq, idx));
-
-	skb = txq->entries[idx].skb;
-
-	/* Can be called from irqs-disabled context
-	 * If skb is not NULL, it means that the whole queue is being
-	 * freed and that the queue is not empty - free the skb
-	 */
-	if (skb) {
-		iwl_op_mode_free_skb(trans->op_mode, skb);
-		txq->entries[idx].skb = NULL;
-	}
-}
-
-static struct page *get_workaround_page(struct iwl_trans *trans,
-					struct sk_buff *skb)
-{
-	struct page **page_ptr;
-	struct page *ret;
-
-	page_ptr = (void *)((u8 *)skb->cb + trans->txqs.page_offs);
-
-	ret = alloc_page(GFP_ATOMIC);
-	if (!ret)
-		return NULL;
-
-	/* set the chaining pointer to the previous page if there */
-	*(void **)((u8 *)page_address(ret) + PAGE_SIZE - sizeof(void *)) = *page_ptr;
-	*page_ptr = ret;
-
-	return ret;
-}
-
-/*
- * Add a TB and if needed apply the FH HW bug workaround;
- * meta != NULL indicates that it's a page mapping and we
- * need to dma_unmap_page() and set the meta->tbs bit in
- * this case.
- */
-static int iwl_txq_gen2_set_tb_with_wa(struct iwl_trans *trans,
-				       struct sk_buff *skb,
-				       struct iwl_tfh_tfd *tfd,
-				       dma_addr_t phys, void *virt,
-				       u16 len, struct iwl_cmd_meta *meta)
-{
-	dma_addr_t oldphys = phys;
-	struct page *page;
-	int ret;
-
-	if (unlikely(dma_mapping_error(trans->dev, phys)))
-		return -ENOMEM;
-
-	if (likely(!iwl_txq_crosses_4g_boundary(phys, len))) {
-		ret = iwl_txq_gen2_set_tb(trans, tfd, phys, len);
-
-		if (ret < 0)
-			goto unmap;
-
-		if (meta)
-			meta->tbs |= BIT(ret);
-
-		ret = 0;
-		goto trace;
-	}
-
-	/*
-	 * Work around a hardware bug. If (as expressed in the
-	 * condition above) the TB ends on a 32-bit boundary,
-	 * then the next TB may be accessed with the wrong
-	 * address.
-	 * To work around it, copy the data elsewhere and make
-	 * a new mapping for it so the device will not fail.
-	 */
-
-	if (WARN_ON(len > PAGE_SIZE - sizeof(void *))) {
-		ret = -ENOBUFS;
-		goto unmap;
-	}
-
-	page = get_workaround_page(trans, skb);
-	if (!page) {
-		ret = -ENOMEM;
-		goto unmap;
-	}
-
-	memcpy(page_address(page), virt, len);
-
-	phys = dma_map_single(trans->dev, page_address(page), len,
-			      DMA_TO_DEVICE);
-	if (unlikely(dma_mapping_error(trans->dev, phys)))
-		return -ENOMEM;
-	ret = iwl_txq_gen2_set_tb(trans, tfd, phys, len);
-	if (ret < 0) {
-		/* unmap the new allocation as single */
-		oldphys = phys;
-		meta = NULL;
-		goto unmap;
-	}
-	IWL_DEBUG_TX(trans,
-		     "TB bug workaround: copied %d bytes from 0x%llx to 0x%llx\n",
-		     len, (unsigned long long)oldphys,
-		     (unsigned long long)phys);
-
-	ret = 0;
-unmap:
-	if (meta)
-		dma_unmap_page(trans->dev, oldphys, len, DMA_TO_DEVICE);
-	else
-		dma_unmap_single(trans->dev, oldphys, len, DMA_TO_DEVICE);
-trace:
-	trace_iwlwifi_dev_tx_tb(trans->dev, skb, virt, phys, len);
-
-	return ret;
-}
-
-#ifdef CONFIG_INET
-struct iwl_tso_hdr_page *get_page_hdr(struct iwl_trans *trans, size_t len,
-				      struct sk_buff *skb)
-{
-	struct iwl_tso_hdr_page *p = this_cpu_ptr(trans->txqs.tso_hdr_page);
-	struct page **page_ptr;
-
-	page_ptr = (void *)((u8 *)skb->cb + trans->txqs.page_offs);
-
-	if (WARN_ON(*page_ptr))
-		return NULL;
-
-	if (!p->page)
-		goto alloc;
-
-	/*
-	 * Check if there's enough room on this page
-	 *
-	 * Note that we put a page chaining pointer *last* in the
-	 * page - we need it somewhere, and if it's there then we
-	 * avoid DMA mapping the last bits of the page which may
-	 * trigger the 32-bit boundary hardware bug.
-	 *
-	 * (see also get_workaround_page() in tx-gen2.c)
-	 */
-	if (p->pos + len < (u8 *)page_address(p->page) + PAGE_SIZE -
-			   sizeof(void *))
-		goto out;
-
-	/* We don't have enough room on this page, get a new one. */
-	__free_page(p->page);
-
-alloc:
-	p->page = alloc_page(GFP_ATOMIC);
-	if (!p->page)
-		return NULL;
-	p->pos = page_address(p->page);
-	/* set the chaining pointer to NULL */
-	*(void **)((u8 *)page_address(p->page) + PAGE_SIZE - sizeof(void *)) = NULL;
-out:
-	*page_ptr = p->page;
-	get_page(p->page);
-	return p;
-}
-#endif
-
-static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
-				    struct sk_buff *skb,
-				    struct iwl_tfh_tfd *tfd, int start_len,
-				    u8 hdr_len,
-				    struct iwl_device_tx_cmd *dev_cmd)
-{
-#ifdef CONFIG_INET
-	struct iwl_tx_cmd_gen2 *tx_cmd = (void *)dev_cmd->payload;
-	struct ieee80211_hdr *hdr = (void *)skb->data;
-	unsigned int snap_ip_tcp_hdrlen, ip_hdrlen, total_len, hdr_room;
-	unsigned int mss = skb_shinfo(skb)->gso_size;
-	u16 length, amsdu_pad;
-	u8 *start_hdr;
-	struct iwl_tso_hdr_page *hdr_page;
-	struct tso_t tso;
-
-	trace_iwlwifi_dev_tx(trans->dev, skb, tfd, sizeof(*tfd),
-			     &dev_cmd->hdr, start_len, 0);
-
-	ip_hdrlen = skb_network_header_len(skb);
-	snap_ip_tcp_hdrlen = 8 + ip_hdrlen + tcp_hdrlen(skb);
-	total_len = skb->len - snap_ip_tcp_hdrlen - hdr_len;
-	amsdu_pad = 0;
-
-	/* total amount of header we may need for this A-MSDU */
-	hdr_room = DIV_ROUND_UP(total_len, mss) *
-		(3 + snap_ip_tcp_hdrlen + sizeof(struct ethhdr));
-
-	/* Our device supports 9 segments at most, it will fit in 1 page */
-	hdr_page = get_page_hdr(trans, hdr_room, skb);
-	if (!hdr_page)
-		return -ENOMEM;
-
-	start_hdr = hdr_page->pos;
-
-	/*
-	 * Pull the ieee80211 header to be able to use TSO core,
-	 * we will restore it for the tx_status flow.
-	 */
-	skb_pull(skb, hdr_len);
-
-	/*
-	 * Remove the length of all the headers that we don't actually
-	 * have in the MPDU by themselves, but that we duplicate into
-	 * all the different MSDUs inside the A-MSDU.
-	 */
-	le16_add_cpu(&tx_cmd->len, -snap_ip_tcp_hdrlen);
-
-	tso_start(skb, &tso);
-
-	while (total_len) {
-		/* this is the data left for this subframe */
-		unsigned int data_left = min_t(unsigned int, mss, total_len);
-		unsigned int tb_len;
-		dma_addr_t tb_phys;
-		u8 *subf_hdrs_start = hdr_page->pos;
-
-		total_len -= data_left;
-
-		memset(hdr_page->pos, 0, amsdu_pad);
-		hdr_page->pos += amsdu_pad;
-		amsdu_pad = (4 - (sizeof(struct ethhdr) + snap_ip_tcp_hdrlen +
-				  data_left)) & 0x3;
-		ether_addr_copy(hdr_page->pos, ieee80211_get_DA(hdr));
-		hdr_page->pos += ETH_ALEN;
-		ether_addr_copy(hdr_page->pos, ieee80211_get_SA(hdr));
-		hdr_page->pos += ETH_ALEN;
-
-		length = snap_ip_tcp_hdrlen + data_left;
-		*((__be16 *)hdr_page->pos) = cpu_to_be16(length);
-		hdr_page->pos += sizeof(length);
-
-		/*
-		 * This will copy the SNAP as well which will be considered
-		 * as MAC header.
-		 */
-		tso_build_hdr(skb, hdr_page->pos, &tso, data_left, !total_len);
-
-		hdr_page->pos += snap_ip_tcp_hdrlen;
-
-		tb_len = hdr_page->pos - start_hdr;
-		tb_phys = dma_map_single(trans->dev, start_hdr,
-					 tb_len, DMA_TO_DEVICE);
-		if (unlikely(dma_mapping_error(trans->dev, tb_phys)))
-			goto out_err;
-		/*
-		 * No need for _with_wa, this is from the TSO page and
-		 * we leave some space at the end of it so can't hit
-		 * the buggy scenario.
-		 */
-		iwl_txq_gen2_set_tb(trans, tfd, tb_phys, tb_len);
-		trace_iwlwifi_dev_tx_tb(trans->dev, skb, start_hdr,
-					tb_phys, tb_len);
-		/* add this subframe's headers' length to the tx_cmd */
-		le16_add_cpu(&tx_cmd->len, hdr_page->pos - subf_hdrs_start);
-
-		/* prepare the start_hdr for the next subframe */
-		start_hdr = hdr_page->pos;
-
-		/* put the payload */
-		while (data_left) {
-			int ret;
-
-			tb_len = min_t(unsigned int, tso.size, data_left);
-			tb_phys = dma_map_single(trans->dev, tso.data,
-						 tb_len, DMA_TO_DEVICE);
-			ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd,
-							  tb_phys, tso.data,
-							  tb_len, NULL);
-			if (ret)
-				goto out_err;
-
-			data_left -= tb_len;
-			tso_build_data(skb, &tso, tb_len);
-		}
-	}
-
-	/* re -add the WiFi header */
-	skb_push(skb, hdr_len);
-
-	return 0;
-
-out_err:
-#endif
-	return -EINVAL;
-}
-
-static struct
-iwl_tfh_tfd *iwl_txq_gen2_build_tx_amsdu(struct iwl_trans *trans,
-					 struct iwl_txq *txq,
-					 struct iwl_device_tx_cmd *dev_cmd,
-					 struct sk_buff *skb,
-					 struct iwl_cmd_meta *out_meta,
-					 int hdr_len,
-					 int tx_cmd_len)
-{
-	int idx = iwl_txq_get_cmd_index(txq, txq->write_ptr);
-	struct iwl_tfh_tfd *tfd = iwl_txq_get_tfd(trans, txq, idx);
-	dma_addr_t tb_phys;
-	int len;
-	void *tb1_addr;
-
-	tb_phys = iwl_txq_get_first_tb_dma(txq, idx);
-
-	/*
-	 * No need for _with_wa, the first TB allocation is aligned up
-	 * to a 64-byte boundary and thus can't be at the end or cross
-	 * a page boundary (much less a 2^32 boundary).
-	 */
-	iwl_txq_gen2_set_tb(trans, tfd, tb_phys, IWL_FIRST_TB_SIZE);
-
-	/*
-	 * The second TB (tb1) points to the remainder of the TX command
-	 * and the 802.11 header - dword aligned size
-	 * (This calculation modifies the TX command, so do it before the
-	 * setup of the first TB)
-	 */
-	len = tx_cmd_len + sizeof(struct iwl_cmd_header) + hdr_len -
-	      IWL_FIRST_TB_SIZE;
-
-	/* do not align A-MSDU to dword as the subframe header aligns it */
-
-	/* map the data for TB1 */
-	tb1_addr = ((u8 *)&dev_cmd->hdr) + IWL_FIRST_TB_SIZE;
-	tb_phys = dma_map_single(trans->dev, tb1_addr, len, DMA_TO_DEVICE);
-	if (unlikely(dma_mapping_error(trans->dev, tb_phys)))
-		goto out_err;
-	/*
-	 * No need for _with_wa(), we ensure (via alignment) that the data
-	 * here can never cross or end at a page boundary.
-	 */
-	iwl_txq_gen2_set_tb(trans, tfd, tb_phys, len);
-
-	if (iwl_txq_gen2_build_amsdu(trans, skb, tfd, len + IWL_FIRST_TB_SIZE,
-				     hdr_len, dev_cmd))
-		goto out_err;
-
-	/* building the A-MSDU might have changed this data, memcpy it now */
-	memcpy(&txq->first_tb_bufs[idx], dev_cmd, IWL_FIRST_TB_SIZE);
-	return tfd;
-
-out_err:
-	iwl_txq_gen2_tfd_unmap(trans, out_meta, tfd);
-	return NULL;
-}
-
-static int iwl_txq_gen2_tx_add_frags(struct iwl_trans *trans,
-				     struct sk_buff *skb,
-				     struct iwl_tfh_tfd *tfd,
-				     struct iwl_cmd_meta *out_meta)
-{
-	int i;
-
-	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
-		const skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
-		dma_addr_t tb_phys;
-		unsigned int fragsz = skb_frag_size(frag);
-		int ret;
-
-		if (!fragsz)
-			continue;
-
-		tb_phys = skb_frag_dma_map(trans->dev, frag, 0,
-					   fragsz, DMA_TO_DEVICE);
-		ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd, tb_phys,
-						  skb_frag_address(frag),
-						  fragsz, out_meta);
-		if (ret)
-			return ret;
-	}
-
-	return 0;
-}
-
-static struct
-iwl_tfh_tfd *iwl_txq_gen2_build_tx(struct iwl_trans *trans,
-				   struct iwl_txq *txq,
-				   struct iwl_device_tx_cmd *dev_cmd,
-				   struct sk_buff *skb,
-				   struct iwl_cmd_meta *out_meta,
-				   int hdr_len,
-				   int tx_cmd_len,
-				   bool pad)
-{
-	int idx = iwl_txq_get_cmd_index(txq, txq->write_ptr);
-	struct iwl_tfh_tfd *tfd = iwl_txq_get_tfd(trans, txq, idx);
-	dma_addr_t tb_phys;
-	int len, tb1_len, tb2_len;
-	void *tb1_addr;
-	struct sk_buff *frag;
-
-	tb_phys = iwl_txq_get_first_tb_dma(txq, idx);
-
-	/* The first TB points to bi-directional DMA data */
-	memcpy(&txq->first_tb_bufs[idx], dev_cmd, IWL_FIRST_TB_SIZE);
-
-	/*
-	 * No need for _with_wa, the first TB allocation is aligned up
-	 * to a 64-byte boundary and thus can't be at the end or cross
-	 * a page boundary (much less a 2^32 boundary).
-	 */
-	iwl_txq_gen2_set_tb(trans, tfd, tb_phys, IWL_FIRST_TB_SIZE);
-
-	/*
-	 * The second TB (tb1) points to the remainder of the TX command
-	 * and the 802.11 header - dword aligned size
-	 * (This calculation modifies the TX command, so do it before the
-	 * setup of the first TB)
-	 */
-	len = tx_cmd_len + sizeof(struct iwl_cmd_header) + hdr_len -
-	      IWL_FIRST_TB_SIZE;
-
-	if (pad)
-		tb1_len = ALIGN(len, 4);
-	else
-		tb1_len = len;
-
-	/* map the data for TB1 */
-	tb1_addr = ((u8 *)&dev_cmd->hdr) + IWL_FIRST_TB_SIZE;
-	tb_phys = dma_map_single(trans->dev, tb1_addr, tb1_len, DMA_TO_DEVICE);
-	if (unlikely(dma_mapping_error(trans->dev, tb_phys)))
-		goto out_err;
-	/*
-	 * No need for _with_wa(), we ensure (via alignment) that the data
-	 * here can never cross or end at a page boundary.
-	 */
-	iwl_txq_gen2_set_tb(trans, tfd, tb_phys, tb1_len);
-	trace_iwlwifi_dev_tx(trans->dev, skb, tfd, sizeof(*tfd), &dev_cmd->hdr,
-			     IWL_FIRST_TB_SIZE + tb1_len, hdr_len);
-
-	/* set up TFD's third entry to point to remainder of skb's head */
-	tb2_len = skb_headlen(skb) - hdr_len;
-
-	if (tb2_len > 0) {
-		int ret;
-
-		tb_phys = dma_map_single(trans->dev, skb->data + hdr_len,
-					 tb2_len, DMA_TO_DEVICE);
-		ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd, tb_phys,
-						  skb->data + hdr_len, tb2_len,
-						  NULL);
-		if (ret)
-			goto out_err;
-	}
-
-	if (iwl_txq_gen2_tx_add_frags(trans, skb, tfd, out_meta))
-		goto out_err;
-
-	skb_walk_frags(skb, frag) {
-		int ret;
-
-		tb_phys = dma_map_single(trans->dev, frag->data,
-					 skb_headlen(frag), DMA_TO_DEVICE);
-		ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd, tb_phys,
-						  frag->data,
-						  skb_headlen(frag), NULL);
-		if (ret)
-			goto out_err;
-		if (iwl_txq_gen2_tx_add_frags(trans, frag, tfd, out_meta))
-			goto out_err;
-	}
-
-	return tfd;
-
-out_err:
-	iwl_txq_gen2_tfd_unmap(trans, out_meta, tfd);
-	return NULL;
-}
-
-static
-struct iwl_tfh_tfd *iwl_txq_gen2_build_tfd(struct iwl_trans *trans,
-					   struct iwl_txq *txq,
-					   struct iwl_device_tx_cmd *dev_cmd,
-					   struct sk_buff *skb,
-					   struct iwl_cmd_meta *out_meta)
-{
-	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
-	int idx = iwl_txq_get_cmd_index(txq, txq->write_ptr);
-	struct iwl_tfh_tfd *tfd = iwl_txq_get_tfd(trans, txq, idx);
-	int len, hdr_len;
-	bool amsdu;
-
-	/* There must be data left over for TB1 or this code must be changed */
-	BUILD_BUG_ON(sizeof(struct iwl_tx_cmd_gen2) < IWL_FIRST_TB_SIZE);
-	BUILD_BUG_ON(sizeof(struct iwl_cmd_header) +
-		     offsetofend(struct iwl_tx_cmd_gen2, dram_info) >
-		     IWL_FIRST_TB_SIZE);
-	BUILD_BUG_ON(sizeof(struct iwl_tx_cmd_gen3) < IWL_FIRST_TB_SIZE);
-	BUILD_BUG_ON(sizeof(struct iwl_cmd_header) +
-		     offsetofend(struct iwl_tx_cmd_gen3, dram_info) >
-		     IWL_FIRST_TB_SIZE);
-
-	memset(tfd, 0, sizeof(*tfd));
-
-	if (trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_AX210)
-		len = sizeof(struct iwl_tx_cmd_gen2);
-	else
-		len = sizeof(struct iwl_tx_cmd_gen3);
-
-	amsdu = ieee80211_is_data_qos(hdr->frame_control) &&
-			(*ieee80211_get_qos_ctl(hdr) &
-			 IEEE80211_QOS_CTL_A_MSDU_PRESENT);
-
-	hdr_len = ieee80211_hdrlen(hdr->frame_control);
-
-	/*
-	 * Only build A-MSDUs here if doing so by GSO, otherwise it may be
-	 * an A-MSDU for other reasons, e.g. NAN or an A-MSDU having been
-	 * built in the higher layers already.
-	 */
-	if (amsdu && skb_shinfo(skb)->gso_size)
-		return iwl_txq_gen2_build_tx_amsdu(trans, txq, dev_cmd, skb,
-						    out_meta, hdr_len, len);
-	return iwl_txq_gen2_build_tx(trans, txq, dev_cmd, skb, out_meta,
-				      hdr_len, len, !amsdu);
-}
-
-int iwl_txq_space(struct iwl_trans *trans, const struct iwl_txq *q)
-{
-	unsigned int max;
-	unsigned int used;
-
-	/*
-	 * To avoid ambiguity between empty and completely full queues, there
-	 * should always be less than max_tfd_queue_size elements in the queue.
-	 * If q->n_window is smaller than max_tfd_queue_size, there is no need
-	 * to reserve any queue entries for this purpose.
-	 */
-	if (q->n_window < trans->trans_cfg->base_params->max_tfd_queue_size)
-		max = q->n_window;
-	else
-		max = trans->trans_cfg->base_params->max_tfd_queue_size - 1;
-
-	/*
-	 * max_tfd_queue_size is a power of 2, so the following is equivalent to
-	 * modulo by max_tfd_queue_size and is well defined.
-	 */
-	used = (q->write_ptr - q->read_ptr) &
-		(trans->trans_cfg->base_params->max_tfd_queue_size - 1);
-
-	if (WARN_ON(used > max))
-		return 0;
-
-	return max - used;
-}
-
-int iwl_txq_gen2_tx(struct iwl_trans *trans, struct sk_buff *skb,
-		    struct iwl_device_tx_cmd *dev_cmd, int txq_id)
-{
-	struct iwl_cmd_meta *out_meta;
-	struct iwl_txq *txq = trans->txqs.txq[txq_id];
-	u16 cmd_len;
-	int idx;
-	void *tfd;
-
-	if (WARN_ONCE(txq_id >= IWL_MAX_TVQM_QUEUES,
-		      "queue %d out of range", txq_id))
-		return -EINVAL;
-
-	if (WARN_ONCE(!test_bit(txq_id, trans->txqs.queue_used),
-		      "TX on unused queue %d\n", txq_id))
-		return -EINVAL;
-
-	if (skb_is_nonlinear(skb) &&
-	    skb_shinfo(skb)->nr_frags > IWL_TRANS_MAX_FRAGS(trans) &&
-	    __skb_linearize(skb))
-		return -ENOMEM;
-
-	spin_lock(&txq->lock);
-
-	if (iwl_txq_space(trans, txq) < txq->high_mark) {
-		iwl_txq_stop(trans, txq);
-
-		/* don't put the packet on the ring, if there is no room */
-		if (unlikely(iwl_txq_space(trans, txq) < 3)) {
-			struct iwl_device_tx_cmd **dev_cmd_ptr;
-
-			dev_cmd_ptr = (void *)((u8 *)skb->cb +
-					       trans->txqs.dev_cmd_offs);
-
-			*dev_cmd_ptr = dev_cmd;
-			__skb_queue_tail(&txq->overflow_q, skb);
-			spin_unlock(&txq->lock);
-			return 0;
-		}
-	}
-
-	idx = iwl_txq_get_cmd_index(txq, txq->write_ptr);
-
-	/* Set up driver data for this TFD */
-	txq->entries[idx].skb = skb;
-	txq->entries[idx].cmd = dev_cmd;
-
-	dev_cmd->hdr.sequence =
-		cpu_to_le16((u16)(QUEUE_TO_SEQ(txq_id) |
-			    INDEX_TO_SEQ(idx)));
-
-	/* Set up first empty entry in queue's array of Tx/cmd buffers */
-	out_meta = &txq->entries[idx].meta;
-	out_meta->flags = 0;
-
-	tfd = iwl_txq_gen2_build_tfd(trans, txq, dev_cmd, skb, out_meta);
-	if (!tfd) {
-		spin_unlock(&txq->lock);
-		return -1;
-	}
-
-	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_AX210) {
-		struct iwl_tx_cmd_gen3 *tx_cmd_gen3 =
-			(void *)dev_cmd->payload;
-
-		cmd_len = le16_to_cpu(tx_cmd_gen3->len);
-	} else {
-		struct iwl_tx_cmd_gen2 *tx_cmd_gen2 =
-			(void *)dev_cmd->payload;
-
-		cmd_len = le16_to_cpu(tx_cmd_gen2->len);
-	}
-
-	/* Set up entry for this TFD in Tx byte-count array */
-	iwl_pcie_gen2_update_byte_tbl(trans, txq, cmd_len,
-				      iwl_txq_gen2_get_num_tbs(trans, tfd));
-
-	/* start timer if queue currently empty */
-	if (txq->read_ptr == txq->write_ptr && txq->wd_timeout)
-		mod_timer(&txq->stuck_timer, jiffies + txq->wd_timeout);
-
-	/* Tell device the write index *just past* this latest filled TFD */
-	txq->write_ptr = iwl_txq_inc_wrap(trans, txq->write_ptr);
-	iwl_txq_inc_wr_ptr(trans, txq);
-	/*
-	 * At this point the frame is "transmitted" successfully
-	 * and we will get a TX status notification eventually.
-	 */
-	spin_unlock(&txq->lock);
-	return 0;
-}
-
-/*************** HOST COMMAND QUEUE FUNCTIONS   *****/
-
-/*
- * iwl_txq_gen2_unmap -  Unmap any remaining DMA mappings and free skb's
- */
-void iwl_txq_gen2_unmap(struct iwl_trans *trans, int txq_id)
-{
-	struct iwl_txq *txq = trans->txqs.txq[txq_id];
-
-	spin_lock_bh(&txq->lock);
-	while (txq->write_ptr != txq->read_ptr) {
-		IWL_DEBUG_TX_REPLY(trans, "Q %d Free %d\n",
-				   txq_id, txq->read_ptr);
-
-		if (txq_id != trans->txqs.cmd.q_id) {
-			int idx = iwl_txq_get_cmd_index(txq, txq->read_ptr);
-			struct sk_buff *skb = txq->entries[idx].skb;
-
-			if (!WARN_ON_ONCE(!skb))
-				iwl_txq_free_tso_page(trans, skb);
-		}
-		iwl_txq_gen2_free_tfd(trans, txq);
-		txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr);
-	}
-
-	while (!skb_queue_empty(&txq->overflow_q)) {
-		struct sk_buff *skb = __skb_dequeue(&txq->overflow_q);
-
-		iwl_op_mode_free_skb(trans->op_mode, skb);
-	}
-
-	spin_unlock_bh(&txq->lock);
-
-	/* just in case - this queue may have been stopped */
-	iwl_wake_queue(trans, txq);
-}
-
-static void iwl_txq_gen2_free_memory(struct iwl_trans *trans,
-				     struct iwl_txq *txq)
-{
-	struct device *dev = trans->dev;
-
-	/* De-alloc circular buffer of TFDs */
-	if (txq->tfds) {
-		dma_free_coherent(dev,
-				  trans->txqs.tfd.size * txq->n_window,
-				  txq->tfds, txq->dma_addr);
-		dma_free_coherent(dev,
-				  sizeof(*txq->first_tb_bufs) * txq->n_window,
-				  txq->first_tb_bufs, txq->first_tb_dma);
-	}
-
-	kfree(txq->entries);
-	if (txq->bc_tbl.addr)
-		dma_pool_free(trans->txqs.bc_pool,
-			      txq->bc_tbl.addr, txq->bc_tbl.dma);
-	kfree(txq);
-}
-
-/*
- * iwl_pcie_txq_free - Deallocate DMA queue.
- * @txq: Transmit queue to deallocate.
- *
- * Empty queue by removing and destroying all BD's.
- * Free all buffers.
- * 0-fill, but do not free "txq" descriptor structure.
- */
-static void iwl_txq_gen2_free(struct iwl_trans *trans, int txq_id)
-{
-	struct iwl_txq *txq;
-	int i;
-
-	if (WARN_ONCE(txq_id >= IWL_MAX_TVQM_QUEUES,
-		      "queue %d out of range", txq_id))
-		return;
-
-	txq = trans->txqs.txq[txq_id];
-
-	if (WARN_ON(!txq))
-		return;
-
-	iwl_txq_gen2_unmap(trans, txq_id);
-
-	/* De-alloc array of command/tx buffers */
-	if (txq_id == trans->txqs.cmd.q_id)
-		for (i = 0; i < txq->n_window; i++) {
-			kfree_sensitive(txq->entries[i].cmd);
-			kfree_sensitive(txq->entries[i].free_buf);
-		}
-	del_timer_sync(&txq->stuck_timer);
-
-	iwl_txq_gen2_free_memory(trans, txq);
-
-	trans->txqs.txq[txq_id] = NULL;
-
-	clear_bit(txq_id, trans->txqs.queue_used);
-}
-
-/*
- * iwl_queue_init - Initialize queue's high/low-water and read/write indexes
- */
-static int iwl_queue_init(struct iwl_txq *q, int slots_num)
-{
-	q->n_window = slots_num;
-
-	/* slots_num must be power-of-two size, otherwise
-	 * iwl_txq_get_cmd_index is broken. */
-	if (WARN_ON(!is_power_of_2(slots_num)))
-		return -EINVAL;
-
-	q->low_mark = q->n_window / 4;
-	if (q->low_mark < 4)
-		q->low_mark = 4;
-
-	q->high_mark = q->n_window / 8;
-	if (q->high_mark < 2)
-		q->high_mark = 2;
-
-	q->write_ptr = 0;
-	q->read_ptr = 0;
-
-	return 0;
-}
-
-int iwl_txq_init(struct iwl_trans *trans, struct iwl_txq *txq, int slots_num,
-		 bool cmd_queue)
-{
-	int ret;
-	u32 tfd_queue_max_size =
-		trans->trans_cfg->base_params->max_tfd_queue_size;
-
-	txq->need_update = false;
-
-	/* max_tfd_queue_size must be power-of-two size, otherwise
-	 * iwl_txq_inc_wrap and iwl_txq_dec_wrap are broken. */
-	if (WARN_ONCE(tfd_queue_max_size & (tfd_queue_max_size - 1),
-		      "Max tfd queue size must be a power of two, but is %d",
-		      tfd_queue_max_size))
-		return -EINVAL;
-
-	/* Initialize queue's high/low-water marks, and head/tail indexes */
-	ret = iwl_queue_init(txq, slots_num);
-	if (ret)
-		return ret;
-
-	spin_lock_init(&txq->lock);
-
-	if (cmd_queue) {
-		static struct lock_class_key iwl_txq_cmd_queue_lock_class;
-
-		lockdep_set_class(&txq->lock, &iwl_txq_cmd_queue_lock_class);
-	}
-
-	__skb_queue_head_init(&txq->overflow_q);
-
-	return 0;
-}
-
-void iwl_txq_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb)
-{
-	struct page **page_ptr;
-	struct page *next;
-
-	page_ptr = (void *)((u8 *)skb->cb + trans->txqs.page_offs);
-	next = *page_ptr;
-	*page_ptr = NULL;
-
-	while (next) {
-		struct page *tmp = next;
-
-		next = *(void **)((u8 *)page_address(next) + PAGE_SIZE -
-				  sizeof(void *));
-		__free_page(tmp);
-	}
-}
-
-void iwl_txq_log_scd_error(struct iwl_trans *trans, struct iwl_txq *txq)
-{
-	u32 txq_id = txq->id;
-	u32 status;
-	bool active;
-	u8 fifo;
-
-	if (trans->trans_cfg->gen2) {
-		IWL_ERR(trans, "Queue %d is stuck %d %d\n", txq_id,
-			txq->read_ptr, txq->write_ptr);
-		/* TODO: access new SCD registers and dump them */
-		return;
-	}
-
-	status = iwl_read_prph(trans, SCD_QUEUE_STATUS_BITS(txq_id));
-	fifo = (status >> SCD_QUEUE_STTS_REG_POS_TXF) & 0x7;
-	active = !!(status & BIT(SCD_QUEUE_STTS_REG_POS_ACTIVE));
-
-	IWL_ERR(trans,
-		"Queue %d is %sactive on fifo %d and stuck for %u ms. SW [%d, %d] HW [%d, %d] FH TRB=0x0%x\n",
-		txq_id, active ? "" : "in", fifo,
-		jiffies_to_msecs(txq->wd_timeout),
-		txq->read_ptr, txq->write_ptr,
-		iwl_read_prph(trans, SCD_QUEUE_RDPTR(txq_id)) &
-			(trans->trans_cfg->base_params->max_tfd_queue_size - 1),
-			iwl_read_prph(trans, SCD_QUEUE_WRPTR(txq_id)) &
-			(trans->trans_cfg->base_params->max_tfd_queue_size - 1),
-			iwl_read_direct32(trans, FH_TX_TRB_REG(fifo)));
-}
-
-static void iwl_txq_stuck_timer(struct timer_list *t)
-{
-	struct iwl_txq *txq = from_timer(txq, t, stuck_timer);
-	struct iwl_trans *trans = txq->trans;
-
-	spin_lock(&txq->lock);
-	/* check if triggered erroneously */
-	if (txq->read_ptr == txq->write_ptr) {
-		spin_unlock(&txq->lock);
-		return;
-	}
-	spin_unlock(&txq->lock);
-
-	iwl_txq_log_scd_error(trans, txq);
-
-	iwl_force_nmi(trans);
-}
-
-static void iwl_txq_set_tfd_invalid_gen1(struct iwl_trans *trans,
-					 struct iwl_tfd *tfd)
-{
-	tfd->num_tbs = 0;
-
-	iwl_pcie_gen1_tfd_set_tb(trans, tfd, 0, trans->invalid_tx_cmd.dma,
-				 trans->invalid_tx_cmd.size);
-}
-
-int iwl_txq_alloc(struct iwl_trans *trans, struct iwl_txq *txq, int slots_num,
-		  bool cmd_queue)
-{
-	size_t num_entries = trans->trans_cfg->gen2 ?
-		slots_num : trans->trans_cfg->base_params->max_tfd_queue_size;
-	size_t tfd_sz;
-	size_t tb0_buf_sz;
-	int i;
-
-	if (WARN_ONCE(slots_num <= 0, "Invalid slots num:%d\n", slots_num))
-		return -EINVAL;
-
-	if (WARN_ON(txq->entries || txq->tfds))
-		return -EINVAL;
-
-	tfd_sz = trans->txqs.tfd.size * num_entries;
-
-	timer_setup(&txq->stuck_timer, iwl_txq_stuck_timer, 0);
-	txq->trans = trans;
-
-	txq->n_window = slots_num;
-
-	txq->entries = kcalloc(slots_num,
-			       sizeof(struct iwl_pcie_txq_entry),
-			       GFP_KERNEL);
-
-	if (!txq->entries)
-		goto error;
-
-	if (cmd_queue)
-		for (i = 0; i < slots_num; i++) {
-			txq->entries[i].cmd =
-				kmalloc(sizeof(struct iwl_device_cmd),
-					GFP_KERNEL);
-			if (!txq->entries[i].cmd)
-				goto error;
-		}
-
-	/* Circular buffer of transmit frame descriptors (TFDs),
-	 * shared with device */
-	txq->tfds = dma_alloc_coherent(trans->dev, tfd_sz,
-				       &txq->dma_addr, GFP_KERNEL);
-	if (!txq->tfds)
-		goto error;
-
-	BUILD_BUG_ON(sizeof(*txq->first_tb_bufs) != IWL_FIRST_TB_SIZE_ALIGN);
-
-	tb0_buf_sz = sizeof(*txq->first_tb_bufs) * slots_num;
-
-	txq->first_tb_bufs = dma_alloc_coherent(trans->dev, tb0_buf_sz,
-						&txq->first_tb_dma,
-						GFP_KERNEL);
-	if (!txq->first_tb_bufs)
-		goto err_free_tfds;
-
-	for (i = 0; i < num_entries; i++) {
-		void *tfd = iwl_txq_get_tfd(trans, txq, i);
-
-		if (trans->trans_cfg->gen2)
-			iwl_txq_set_tfd_invalid_gen2(trans, tfd);
-		else
-			iwl_txq_set_tfd_invalid_gen1(trans, tfd);
-	}
-
-	return 0;
-err_free_tfds:
-	dma_free_coherent(trans->dev, tfd_sz, txq->tfds, txq->dma_addr);
-	txq->tfds = NULL;
-error:
-	if (txq->entries && cmd_queue)
-		for (i = 0; i < slots_num; i++)
-			kfree(txq->entries[i].cmd);
-	kfree(txq->entries);
-	txq->entries = NULL;
-
-	return -ENOMEM;
-}
-
-static struct iwl_txq *
-iwl_txq_dyn_alloc_dma(struct iwl_trans *trans, int size, unsigned int timeout)
-{
-	size_t bc_tbl_size, bc_tbl_entries;
-	struct iwl_txq *txq;
-	int ret;
-
-	WARN_ON(!trans->txqs.bc_tbl_size);
-
-	bc_tbl_size = trans->txqs.bc_tbl_size;
-	bc_tbl_entries = bc_tbl_size / sizeof(u16);
-
-	if (WARN_ON(size > bc_tbl_entries))
-		return ERR_PTR(-EINVAL);
-
-	txq = kzalloc(sizeof(*txq), GFP_KERNEL);
-	if (!txq)
-		return ERR_PTR(-ENOMEM);
-
-	txq->bc_tbl.addr = dma_pool_alloc(trans->txqs.bc_pool, GFP_KERNEL,
-					  &txq->bc_tbl.dma);
-	if (!txq->bc_tbl.addr) {
-		IWL_ERR(trans, "Scheduler BC Table allocation failed\n");
-		kfree(txq);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	ret = iwl_txq_alloc(trans, txq, size, false);
-	if (ret) {
-		IWL_ERR(trans, "Tx queue alloc failed\n");
-		goto error;
-	}
-	ret = iwl_txq_init(trans, txq, size, false);
-	if (ret) {
-		IWL_ERR(trans, "Tx queue init failed\n");
-		goto error;
-	}
-
-	txq->wd_timeout = msecs_to_jiffies(timeout);
-
-	return txq;
-
-error:
-	iwl_txq_gen2_free_memory(trans, txq);
-	return ERR_PTR(ret);
-}
-
-static int iwl_txq_alloc_response(struct iwl_trans *trans, struct iwl_txq *txq,
-				  struct iwl_host_cmd *hcmd)
-{
-	struct iwl_tx_queue_cfg_rsp *rsp;
-	int ret, qid;
-	u32 wr_ptr;
-
-	if (WARN_ON(iwl_rx_packet_payload_len(hcmd->resp_pkt) !=
-		    sizeof(*rsp))) {
-		ret = -EINVAL;
-		goto error_free_resp;
-	}
-
-	rsp = (void *)hcmd->resp_pkt->data;
-	qid = le16_to_cpu(rsp->queue_number);
-	wr_ptr = le16_to_cpu(rsp->write_pointer);
-
-	if (qid >= ARRAY_SIZE(trans->txqs.txq)) {
-		WARN_ONCE(1, "queue index %d unsupported", qid);
-		ret = -EIO;
-		goto error_free_resp;
-	}
-
-	if (test_and_set_bit(qid, trans->txqs.queue_used)) {
-		WARN_ONCE(1, "queue %d already used", qid);
-		ret = -EIO;
-		goto error_free_resp;
-	}
-
-	if (WARN_ONCE(trans->txqs.txq[qid],
-		      "queue %d already allocated\n", qid)) {
-		ret = -EIO;
-		goto error_free_resp;
-	}
-
-	txq->id = qid;
-	trans->txqs.txq[qid] = txq;
-	wr_ptr &= (trans->trans_cfg->base_params->max_tfd_queue_size - 1);
-
-	/* Place first TFD at index corresponding to start sequence number */
-	txq->read_ptr = wr_ptr;
-	txq->write_ptr = wr_ptr;
-
-	IWL_DEBUG_TX_QUEUES(trans, "Activate queue %d\n", qid);
-
-	iwl_free_resp(hcmd);
-	return qid;
-
-error_free_resp:
-	iwl_free_resp(hcmd);
-	iwl_txq_gen2_free_memory(trans, txq);
-	return ret;
-}
-
-int iwl_txq_dyn_alloc(struct iwl_trans *trans, u32 flags, u32 sta_mask,
-		      u8 tid, int size, unsigned int timeout)
-{
-	struct iwl_txq *txq;
-	union {
-		struct iwl_tx_queue_cfg_cmd old;
-		struct iwl_scd_queue_cfg_cmd new;
-	} cmd;
-	struct iwl_host_cmd hcmd = {
-		.flags = CMD_WANT_SKB,
-	};
-	int ret;
-
-	/* take the min with bytecount table entries allowed */
-	size = min_t(u32, size, trans->txqs.bc_tbl_size / sizeof(u16));
-	/* but must be power of 2 values for calculating read/write pointers */
-	size = rounddown_pow_of_two(size);
-
-	if (trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_BZ &&
-	    trans->hw_rev_step == SILICON_A_STEP) {
-		size = 4096;
-		txq = iwl_txq_dyn_alloc_dma(trans, size, timeout);
-	} else {
-		do {
-			txq = iwl_txq_dyn_alloc_dma(trans, size, timeout);
-			if (!IS_ERR(txq))
-				break;
-
-			IWL_DEBUG_TX_QUEUES(trans,
-					    "Failed allocating TXQ of size %d for sta mask %x tid %d, ret: %ld\n",
-					    size, sta_mask, tid,
-					    PTR_ERR(txq));
-			size /= 2;
-		} while (size >= 16);
-	}
-
-	if (IS_ERR(txq))
-		return PTR_ERR(txq);
-
-	if (trans->txqs.queue_alloc_cmd_ver == 0) {
-		memset(&cmd.old, 0, sizeof(cmd.old));
-		cmd.old.tfdq_addr = cpu_to_le64(txq->dma_addr);
-		cmd.old.byte_cnt_addr = cpu_to_le64(txq->bc_tbl.dma);
-		cmd.old.cb_size = cpu_to_le32(TFD_QUEUE_CB_SIZE(size));
-		cmd.old.flags = cpu_to_le16(flags | TX_QUEUE_CFG_ENABLE_QUEUE);
-		cmd.old.tid = tid;
-
-		if (hweight32(sta_mask) != 1) {
-			ret = -EINVAL;
-			goto error;
-		}
-		cmd.old.sta_id = ffs(sta_mask) - 1;
-
-		hcmd.id = SCD_QUEUE_CFG;
-		hcmd.len[0] = sizeof(cmd.old);
-		hcmd.data[0] = &cmd.old;
-	} else if (trans->txqs.queue_alloc_cmd_ver == 3) {
-		memset(&cmd.new, 0, sizeof(cmd.new));
-		cmd.new.operation = cpu_to_le32(IWL_SCD_QUEUE_ADD);
-		cmd.new.u.add.tfdq_dram_addr = cpu_to_le64(txq->dma_addr);
-		cmd.new.u.add.bc_dram_addr = cpu_to_le64(txq->bc_tbl.dma);
-		cmd.new.u.add.cb_size = cpu_to_le32(TFD_QUEUE_CB_SIZE(size));
-		cmd.new.u.add.flags = cpu_to_le32(flags);
-		cmd.new.u.add.sta_mask = cpu_to_le32(sta_mask);
-		cmd.new.u.add.tid = tid;
-
-		hcmd.id = WIDE_ID(DATA_PATH_GROUP, SCD_QUEUE_CONFIG_CMD);
-		hcmd.len[0] = sizeof(cmd.new);
-		hcmd.data[0] = &cmd.new;
-	} else {
-		ret = -EOPNOTSUPP;
-		goto error;
-	}
-
-	ret = iwl_trans_send_cmd(trans, &hcmd);
-	if (ret)
-		goto error;
-
-	return iwl_txq_alloc_response(trans, txq, &hcmd);
-
-error:
-	iwl_txq_gen2_free_memory(trans, txq);
-	return ret;
-}
-
-void iwl_txq_dyn_free(struct iwl_trans *trans, int queue)
-{
-	if (WARN(queue >= IWL_MAX_TVQM_QUEUES,
-		 "queue %d out of range", queue))
-		return;
-
-	/*
-	 * Upon HW Rfkill - we stop the device, and then stop the queues
-	 * in the op_mode. Just for the sake of the simplicity of the op_mode,
-	 * allow the op_mode to call txq_disable after it already called
-	 * stop_device.
-	 */
-	if (!test_and_clear_bit(queue, trans->txqs.queue_used)) {
-		WARN_ONCE(test_bit(STATUS_DEVICE_ENABLED, &trans->status),
-			  "queue %d not used", queue);
-		return;
-	}
-
-	iwl_txq_gen2_free(trans, queue);
-
-	IWL_DEBUG_TX_QUEUES(trans, "Deactivate queue %d\n", queue);
-}
-
-void iwl_txq_gen2_tx_free(struct iwl_trans *trans)
-{
-	int i;
-
-	memset(trans->txqs.queue_used, 0, sizeof(trans->txqs.queue_used));
-
-	/* Free all TX queues */
-	for (i = 0; i < ARRAY_SIZE(trans->txqs.txq); i++) {
-		if (!trans->txqs.txq[i])
-			continue;
-
-		iwl_txq_gen2_free(trans, i);
-	}
-}
-
-int iwl_txq_gen2_init(struct iwl_trans *trans, int txq_id, int queue_size)
-{
-	struct iwl_txq *queue;
-	int ret;
-
-	/* alloc and init the tx queue */
-	if (!trans->txqs.txq[txq_id]) {
-		queue = kzalloc(sizeof(*queue), GFP_KERNEL);
-		if (!queue) {
-			IWL_ERR(trans, "Not enough memory for tx queue\n");
-			return -ENOMEM;
-		}
-		trans->txqs.txq[txq_id] = queue;
-		ret = iwl_txq_alloc(trans, queue, queue_size, true);
-		if (ret) {
-			IWL_ERR(trans, "Tx %d queue init failed\n", txq_id);
-			goto error;
-		}
-	} else {
-		queue = trans->txqs.txq[txq_id];
-	}
-
-	ret = iwl_txq_init(trans, queue, queue_size,
-			   (txq_id == trans->txqs.cmd.q_id));
-	if (ret) {
-		IWL_ERR(trans, "Tx %d queue alloc failed\n", txq_id);
-		goto error;
-	}
-	trans->txqs.txq[txq_id]->id = txq_id;
-	set_bit(txq_id, trans->txqs.queue_used);
-
-	return 0;
-
-error:
-	iwl_txq_gen2_tx_free(trans);
-	return ret;
-}
-
-static inline dma_addr_t iwl_txq_gen1_tfd_tb_get_addr(struct iwl_trans *trans,
-						      struct iwl_tfd *tfd, u8 idx)
-{
-	struct iwl_tfd_tb *tb = &tfd->tbs[idx];
-	dma_addr_t addr;
-	dma_addr_t hi_len;
-
-	addr = get_unaligned_le32(&tb->lo);
-
-	if (sizeof(dma_addr_t) <= sizeof(u32))
-		return addr;
-
-	hi_len = le16_to_cpu(tb->hi_n_len) & 0xF;
-
-	/*
-	 * shift by 16 twice to avoid warnings on 32-bit
-	 * (where this code never runs anyway due to the
-	 * if statement above)
-	 */
-	return addr | ((hi_len << 16) << 16);
-}
-
-void iwl_txq_gen1_tfd_unmap(struct iwl_trans *trans,
-			    struct iwl_cmd_meta *meta,
-			    struct iwl_txq *txq, int index)
-{
-	int i, num_tbs;
-	struct iwl_tfd *tfd = iwl_txq_get_tfd(trans, txq, index);
-
-	/* Sanity check on number of chunks */
-	num_tbs = iwl_txq_gen1_tfd_get_num_tbs(trans, tfd);
-
-	if (num_tbs > trans->txqs.tfd.max_tbs) {
-		IWL_ERR(trans, "Too many chunks: %i\n", num_tbs);
-		/* @todo issue fatal error, it is quite serious situation */
-		return;
-	}
-
-	/* first TB is never freed - it's the bidirectional DMA data */
-
-	for (i = 1; i < num_tbs; i++) {
-		if (meta->tbs & BIT(i))
-			dma_unmap_page(trans->dev,
-				       iwl_txq_gen1_tfd_tb_get_addr(trans,
-								    tfd, i),
-				       iwl_txq_gen1_tfd_tb_get_len(trans,
-								   tfd, i),
-				       DMA_TO_DEVICE);
-		else
-			dma_unmap_single(trans->dev,
-					 iwl_txq_gen1_tfd_tb_get_addr(trans,
-								      tfd, i),
-					 iwl_txq_gen1_tfd_tb_get_len(trans,
-								     tfd, i),
-					 DMA_TO_DEVICE);
-	}
-
-	meta->tbs = 0;
-
-	iwl_txq_set_tfd_invalid_gen1(trans, tfd);
-}
-
-#define IWL_TX_CRC_SIZE 4
-#define IWL_TX_DELIMITER_SIZE 4
-
-/*
- * iwl_txq_gen1_update_byte_cnt_tbl - Set up entry in Tx byte-count array
- */
-void iwl_txq_gen1_update_byte_cnt_tbl(struct iwl_trans *trans,
-				      struct iwl_txq *txq, u16 byte_cnt,
-				      int num_tbs)
-{
-	struct iwlagn_scd_bc_tbl *scd_bc_tbl;
-	int write_ptr = txq->write_ptr;
-	int txq_id = txq->id;
-	u8 sec_ctl = 0;
-	u16 len = byte_cnt + IWL_TX_CRC_SIZE + IWL_TX_DELIMITER_SIZE;
-	__le16 bc_ent;
-	struct iwl_device_tx_cmd *dev_cmd = txq->entries[txq->write_ptr].cmd;
-	struct iwl_tx_cmd *tx_cmd = (void *)dev_cmd->payload;
-	u8 sta_id = tx_cmd->sta_id;
-
-	scd_bc_tbl = trans->txqs.scd_bc_tbls.addr;
-
-	sec_ctl = tx_cmd->sec_ctl;
-
-	switch (sec_ctl & TX_CMD_SEC_MSK) {
-	case TX_CMD_SEC_CCM:
-		len += IEEE80211_CCMP_MIC_LEN;
-		break;
-	case TX_CMD_SEC_TKIP:
-		len += IEEE80211_TKIP_ICV_LEN;
-		break;
-	case TX_CMD_SEC_WEP:
-		len += IEEE80211_WEP_IV_LEN + IEEE80211_WEP_ICV_LEN;
-		break;
-	}
-	if (trans->txqs.bc_table_dword)
-		len = DIV_ROUND_UP(len, 4);
-
-	if (WARN_ON(len > 0xFFF || write_ptr >= TFD_QUEUE_SIZE_MAX))
-		return;
-
-	bc_ent = cpu_to_le16(len | (sta_id << 12));
-
-	scd_bc_tbl[txq_id].tfd_offset[write_ptr] = bc_ent;
-
-	if (write_ptr < TFD_QUEUE_SIZE_BC_DUP)
-		scd_bc_tbl[txq_id].tfd_offset[TFD_QUEUE_SIZE_MAX + write_ptr] =
-			bc_ent;
-}
-
-void iwl_txq_gen1_inval_byte_cnt_tbl(struct iwl_trans *trans,
-				     struct iwl_txq *txq)
-{
-	struct iwlagn_scd_bc_tbl *scd_bc_tbl = trans->txqs.scd_bc_tbls.addr;
-	int txq_id = txq->id;
-	int read_ptr = txq->read_ptr;
-	u8 sta_id = 0;
-	__le16 bc_ent;
-	struct iwl_device_tx_cmd *dev_cmd = txq->entries[read_ptr].cmd;
-	struct iwl_tx_cmd *tx_cmd = (void *)dev_cmd->payload;
-
-	WARN_ON(read_ptr >= TFD_QUEUE_SIZE_MAX);
-
-	if (txq_id != trans->txqs.cmd.q_id)
-		sta_id = tx_cmd->sta_id;
-
-	bc_ent = cpu_to_le16(1 | (sta_id << 12));
-
-	scd_bc_tbl[txq_id].tfd_offset[read_ptr] = bc_ent;
-
-	if (read_ptr < TFD_QUEUE_SIZE_BC_DUP)
-		scd_bc_tbl[txq_id].tfd_offset[TFD_QUEUE_SIZE_MAX + read_ptr] =
-			bc_ent;
-}
-
-/*
- * iwl_txq_free_tfd - Free all chunks referenced by TFD [txq->q.read_ptr]
- * @trans - transport private data
- * @txq - tx queue
- * @dma_dir - the direction of the DMA mapping
- *
- * Does NOT advance any TFD circular buffer read/write indexes
- * Does NOT free the TFD itself (which is within circular buffer)
- */
-void iwl_txq_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq)
-{
-	/* rd_ptr is bounded by TFD_QUEUE_SIZE_MAX and
-	 * idx is bounded by n_window
-	 */
-	int rd_ptr = txq->read_ptr;
-	int idx = iwl_txq_get_cmd_index(txq, rd_ptr);
-	struct sk_buff *skb;
-
-	lockdep_assert_held(&txq->lock);
-
-	if (!txq->entries)
-		return;
-
-	/* We have only q->n_window txq->entries, but we use
-	 * TFD_QUEUE_SIZE_MAX tfds
-	 */
-	if (trans->trans_cfg->gen2)
-		iwl_txq_gen2_tfd_unmap(trans, &txq->entries[idx].meta,
-				       iwl_txq_get_tfd(trans, txq, rd_ptr));
-	else
-		iwl_txq_gen1_tfd_unmap(trans, &txq->entries[idx].meta,
-				       txq, rd_ptr);
-
-	/* free SKB */
-	skb = txq->entries[idx].skb;
-
-	/* Can be called from irqs-disabled context
-	 * If skb is not NULL, it means that the whole queue is being
-	 * freed and that the queue is not empty - free the skb
-	 */
-	if (skb) {
-		iwl_op_mode_free_skb(trans->op_mode, skb);
-		txq->entries[idx].skb = NULL;
-	}
-}
-
-void iwl_txq_progress(struct iwl_txq *txq)
-{
-	lockdep_assert_held(&txq->lock);
-
-	if (!txq->wd_timeout)
-		return;
-
-	/*
-	 * station is asleep and we send data - that must
-	 * be uAPSD or PS-Poll. Don't rearm the timer.
-	 */
-	if (txq->frozen)
-		return;
-
-	/*
-	 * if empty delete timer, otherwise move timer forward
-	 * since we're making progress on this queue
-	 */
-	if (txq->read_ptr == txq->write_ptr)
-		del_timer(&txq->stuck_timer);
-	else
-		mod_timer(&txq->stuck_timer, jiffies + txq->wd_timeout);
-}
-
-/* Frees buffers until index _not_ inclusive */
-void iwl_txq_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
-		     struct sk_buff_head *skbs, bool is_flush)
-{
-	struct iwl_txq *txq = trans->txqs.txq[txq_id];
-	int tfd_num, read_ptr, last_to_free;
-
-	/* This function is not meant to release cmd queue*/
-	if (WARN_ON(txq_id == trans->txqs.cmd.q_id))
-		return;
-
-	if (WARN_ON(!txq))
-		return;
-
-	tfd_num = iwl_txq_get_cmd_index(txq, ssn);
-
-	spin_lock_bh(&txq->lock);
-	read_ptr = iwl_txq_get_cmd_index(txq, txq->read_ptr);
-
-	if (!test_bit(txq_id, trans->txqs.queue_used)) {
-		IWL_DEBUG_TX_QUEUES(trans, "Q %d inactive - ignoring idx %d\n",
-				    txq_id, ssn);
-		goto out;
-	}
-
-	if (read_ptr == tfd_num)
-		goto out;
-
-	IWL_DEBUG_TX_REPLY(trans, "[Q %d] %d (%d) -> %d (%d)\n",
-			   txq_id, read_ptr, txq->read_ptr, tfd_num, ssn);
-
-	/*Since we free until index _not_ inclusive, the one before index is
-	 * the last we will free. This one must be used */
-	last_to_free = iwl_txq_dec_wrap(trans, tfd_num);
-
-	if (!iwl_txq_used(txq, last_to_free)) {
-		IWL_ERR(trans,
-			"%s: Read index for txq id (%d), last_to_free %d is out of range [0-%d] %d %d.\n",
-			__func__, txq_id, last_to_free,
-			trans->trans_cfg->base_params->max_tfd_queue_size,
-			txq->write_ptr, txq->read_ptr);
-
-		iwl_op_mode_time_point(trans->op_mode,
-				       IWL_FW_INI_TIME_POINT_FAKE_TX,
-				       NULL);
-		goto out;
-	}
-
-	if (WARN_ON(!skb_queue_empty(skbs)))
-		goto out;
-
-	for (;
-	     read_ptr != tfd_num;
-	     txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr),
-	     read_ptr = iwl_txq_get_cmd_index(txq, txq->read_ptr)) {
-		struct sk_buff *skb = txq->entries[read_ptr].skb;
-
-		if (WARN_ONCE(!skb, "no SKB at %d (%d) on queue %d\n",
-			      read_ptr, txq->read_ptr, txq_id))
-			continue;
-
-		iwl_txq_free_tso_page(trans, skb);
-
-		__skb_queue_tail(skbs, skb);
-
-		txq->entries[read_ptr].skb = NULL;
-
-		if (!trans->trans_cfg->gen2)
-			iwl_txq_gen1_inval_byte_cnt_tbl(trans, txq);
-
-		iwl_txq_free_tfd(trans, txq);
-	}
-
-	iwl_txq_progress(txq);
-
-	if (iwl_txq_space(trans, txq) > txq->low_mark &&
-	    test_bit(txq_id, trans->txqs.queue_stopped)) {
-		struct sk_buff_head overflow_skbs;
-		struct sk_buff *skb;
-
-		__skb_queue_head_init(&overflow_skbs);
-		skb_queue_splice_init(&txq->overflow_q,
-				      is_flush ? skbs : &overflow_skbs);
-
-		/*
-		 * We are going to transmit from the overflow queue.
-		 * Remember this state so that wait_for_txq_empty will know we
-		 * are adding more packets to the TFD queue. It cannot rely on
-		 * the state of &txq->overflow_q, as we just emptied it, but
-		 * haven't TXed the content yet.
-		 */
-		txq->overflow_tx = true;
-
-		/*
-		 * This is tricky: we are in reclaim path which is non
-		 * re-entrant, so noone will try to take the access the
-		 * txq data from that path. We stopped tx, so we can't
-		 * have tx as well. Bottom line, we can unlock and re-lock
-		 * later.
-		 */
-		spin_unlock_bh(&txq->lock);
-
-		while ((skb = __skb_dequeue(&overflow_skbs))) {
-			struct iwl_device_tx_cmd *dev_cmd_ptr;
-
-			dev_cmd_ptr = *(void **)((u8 *)skb->cb +
-						 trans->txqs.dev_cmd_offs);
-
-			/*
-			 * Note that we can very well be overflowing again.
-			 * In that case, iwl_txq_space will be small again
-			 * and we won't wake mac80211's queue.
-			 */
-			iwl_trans_tx(trans, skb, dev_cmd_ptr, txq_id);
-		}
-
-		if (iwl_txq_space(trans, txq) > txq->low_mark)
-			iwl_wake_queue(trans, txq);
-
-		spin_lock_bh(&txq->lock);
-		txq->overflow_tx = false;
-	}
-
-out:
-	spin_unlock_bh(&txq->lock);
-}
-
-/* Set wr_ptr of specific device and txq  */
-void iwl_txq_set_q_ptrs(struct iwl_trans *trans, int txq_id, int ptr)
-{
-	struct iwl_txq *txq = trans->txqs.txq[txq_id];
-
-	spin_lock_bh(&txq->lock);
-
-	txq->write_ptr = ptr;
-	txq->read_ptr = txq->write_ptr;
-
-	spin_unlock_bh(&txq->lock);
-}
-
-void iwl_trans_txq_freeze_timer(struct iwl_trans *trans, unsigned long txqs,
-				bool freeze)
-{
-	int queue;
-
-	for_each_set_bit(queue, &txqs, BITS_PER_LONG) {
-		struct iwl_txq *txq = trans->txqs.txq[queue];
-		unsigned long now;
-
-		spin_lock_bh(&txq->lock);
-
-		now = jiffies;
-
-		if (txq->frozen == freeze)
-			goto next_queue;
-
-		IWL_DEBUG_TX_QUEUES(trans, "%s TXQ %d\n",
-				    freeze ? "Freezing" : "Waking", queue);
-
-		txq->frozen = freeze;
-
-		if (txq->read_ptr == txq->write_ptr)
-			goto next_queue;
-
-		if (freeze) {
-			if (unlikely(time_after(now,
-						txq->stuck_timer.expires))) {
-				/*
-				 * The timer should have fired, maybe it is
-				 * spinning right now on the lock.
-				 */
-				goto next_queue;
-			}
-			/* remember how long until the timer fires */
-			txq->frozen_expiry_remainder =
-				txq->stuck_timer.expires - now;
-			del_timer(&txq->stuck_timer);
-			goto next_queue;
-		}
-
-		/*
-		 * Wake a non-empty queue -> arm timer with the
-		 * remainder before it froze
-		 */
-		mod_timer(&txq->stuck_timer,
-			  now + txq->frozen_expiry_remainder);
-
-next_queue:
-		spin_unlock_bh(&txq->lock);
-	}
-}
-
-#define HOST_COMPLETE_TIMEOUT	(2 * HZ)
-
-static int iwl_trans_txq_send_hcmd_sync(struct iwl_trans *trans,
-					struct iwl_host_cmd *cmd)
-{
-	const char *cmd_str = iwl_get_cmd_string(trans, cmd->id);
-	struct iwl_txq *txq = trans->txqs.txq[trans->txqs.cmd.q_id];
-	int cmd_idx;
-	int ret;
-
-	IWL_DEBUG_INFO(trans, "Attempting to send sync command %s\n", cmd_str);
-
-	if (WARN(test_and_set_bit(STATUS_SYNC_HCMD_ACTIVE,
-				  &trans->status),
-		 "Command %s: a command is already active!\n", cmd_str))
-		return -EIO;
-
-	IWL_DEBUG_INFO(trans, "Setting HCMD_ACTIVE for command %s\n", cmd_str);
-
-	cmd_idx = trans->ops->send_cmd(trans, cmd);
-	if (cmd_idx < 0) {
-		ret = cmd_idx;
-		clear_bit(STATUS_SYNC_HCMD_ACTIVE, &trans->status);
-		IWL_ERR(trans, "Error sending %s: enqueue_hcmd failed: %d\n",
-			cmd_str, ret);
-		return ret;
-	}
-
-	ret = wait_event_timeout(trans->wait_command_queue,
-				 !test_bit(STATUS_SYNC_HCMD_ACTIVE,
-					   &trans->status),
-				 HOST_COMPLETE_TIMEOUT);
-	if (!ret) {
-		IWL_ERR(trans, "Error sending %s: time out after %dms.\n",
-			cmd_str, jiffies_to_msecs(HOST_COMPLETE_TIMEOUT));
-
-		IWL_ERR(trans, "Current CMD queue read_ptr %d write_ptr %d\n",
-			txq->read_ptr, txq->write_ptr);
-
-		clear_bit(STATUS_SYNC_HCMD_ACTIVE, &trans->status);
-		IWL_DEBUG_INFO(trans, "Clearing HCMD_ACTIVE for command %s\n",
-			       cmd_str);
-		ret = -ETIMEDOUT;
-
-		iwl_trans_sync_nmi(trans);
-		goto cancel;
-	}
-
-	if (test_bit(STATUS_FW_ERROR, &trans->status)) {
-		if (!test_and_clear_bit(STATUS_SUPPRESS_CMD_ERROR_ONCE,
-					&trans->status)) {
-			IWL_ERR(trans, "FW error in SYNC CMD %s\n", cmd_str);
-			dump_stack();
-		}
-		ret = -EIO;
-		goto cancel;
-	}
-
-	if (!(cmd->flags & CMD_SEND_IN_RFKILL) &&
-	    test_bit(STATUS_RFKILL_OPMODE, &trans->status)) {
-		IWL_DEBUG_RF_KILL(trans, "RFKILL in SYNC CMD... no rsp\n");
-		ret = -ERFKILL;
-		goto cancel;
-	}
-
-	if ((cmd->flags & CMD_WANT_SKB) && !cmd->resp_pkt) {
-		IWL_ERR(trans, "Error: Response NULL in '%s'\n", cmd_str);
-		ret = -EIO;
-		goto cancel;
-	}
-
-	return 0;
-
-cancel:
-	if (cmd->flags & CMD_WANT_SKB) {
-		/*
-		 * Cancel the CMD_WANT_SKB flag for the cmd in the
-		 * TX cmd queue. Otherwise in case the cmd comes
-		 * in later, it will possibly set an invalid
-		 * address (cmd->meta.source).
-		 */
-		txq->entries[cmd_idx].meta.flags &= ~CMD_WANT_SKB;
-	}
-
-	if (cmd->resp_pkt) {
-		iwl_free_resp(cmd);
-		cmd->resp_pkt = NULL;
-	}
-
-	return ret;
-}
-
-int iwl_trans_txq_send_hcmd(struct iwl_trans *trans,
-			    struct iwl_host_cmd *cmd)
-{
-	/* Make sure the NIC is still alive in the bus */
-	if (test_bit(STATUS_TRANS_DEAD, &trans->status))
-		return -ENODEV;
-
-	if (!(cmd->flags & CMD_SEND_IN_RFKILL) &&
-	    test_bit(STATUS_RFKILL_OPMODE, &trans->status)) {
-		IWL_DEBUG_RF_KILL(trans, "Dropping CMD 0x%x: RF KILL\n",
-				  cmd->id);
-		return -ERFKILL;
-	}
-
-	if (unlikely(trans->system_pm_mode == IWL_PLAT_PM_MODE_D3 &&
-		     !(cmd->flags & CMD_SEND_IN_D3))) {
-		IWL_DEBUG_WOWLAN(trans, "Dropping CMD 0x%x: D3\n", cmd->id);
-		return -EHOSTDOWN;
-	}
-
-	if (cmd->flags & CMD_ASYNC) {
-		int ret;
-
-		/* An asynchronous command can not expect an SKB to be set. */
-		if (WARN_ON(cmd->flags & CMD_WANT_SKB))
-			return -EINVAL;
-
-		ret = trans->ops->send_cmd(trans, cmd);
-		if (ret < 0) {
-			IWL_ERR(trans,
-				"Error sending %s: enqueue_hcmd failed: %d\n",
-				iwl_get_cmd_string(trans, cmd->id), ret);
-			return ret;
-		}
-		return 0;
-	}
-
-	return iwl_trans_txq_send_hcmd_sync(trans, cmd);
-}
-
diff --git a/drivers/net/wireless/intel/iwlwifi/queue/tx.h b/drivers/net/wireless/intel/iwlwifi/queue/tx.h
deleted file mode 100644
index 124b29aac4a1..000000000000
--- a/drivers/net/wireless/intel/iwlwifi/queue/tx.h
+++ /dev/null
@@ -1,191 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
-/*
- * Copyright (C) 2020-2023 Intel Corporation
- */
-#ifndef __iwl_trans_queue_tx_h__
-#define __iwl_trans_queue_tx_h__
-#include "iwl-fh.h"
-#include "fw/api/tx.h"
-
-struct iwl_tso_hdr_page {
-	struct page *page;
-	u8 *pos;
-};
-
-static inline dma_addr_t
-iwl_txq_get_first_tb_dma(struct iwl_txq *txq, int idx)
-{
-	return txq->first_tb_dma +
-	       sizeof(struct iwl_pcie_first_tb_buf) * idx;
-}
-
-static inline u16 iwl_txq_get_cmd_index(const struct iwl_txq *q, u32 index)
-{
-	return index & (q->n_window - 1);
-}
-
-void iwl_txq_gen2_unmap(struct iwl_trans *trans, int txq_id);
-
-static inline void iwl_wake_queue(struct iwl_trans *trans,
-				  struct iwl_txq *txq)
-{
-	if (test_and_clear_bit(txq->id, trans->txqs.queue_stopped)) {
-		IWL_DEBUG_TX_QUEUES(trans, "Wake hwq %d\n", txq->id);
-		iwl_op_mode_queue_not_full(trans->op_mode, txq->id);
-	}
-}
-
-static inline void *iwl_txq_get_tfd(struct iwl_trans *trans,
-				    struct iwl_txq *txq, int idx)
-{
-	if (trans->trans_cfg->gen2)
-		idx = iwl_txq_get_cmd_index(txq, idx);
-
-	return (u8 *)txq->tfds + trans->txqs.tfd.size * idx;
-}
-
-int iwl_txq_alloc(struct iwl_trans *trans, struct iwl_txq *txq, int slots_num,
-		  bool cmd_queue);
-/*
- * We need this inline in case dma_addr_t is only 32-bits - since the
- * hardware is always 64-bit, the issue can still occur in that case,
- * so use u64 for 'phys' here to force the addition in 64-bit.
- */
-static inline bool iwl_txq_crosses_4g_boundary(u64 phys, u16 len)
-{
-	return upper_32_bits(phys) != upper_32_bits(phys + len);
-}
-
-int iwl_txq_space(struct iwl_trans *trans, const struct iwl_txq *q);
-
-static inline void iwl_txq_stop(struct iwl_trans *trans, struct iwl_txq *txq)
-{
-	if (!test_and_set_bit(txq->id, trans->txqs.queue_stopped)) {
-		iwl_op_mode_queue_full(trans->op_mode, txq->id);
-		IWL_DEBUG_TX_QUEUES(trans, "Stop hwq %d\n", txq->id);
-	} else {
-		IWL_DEBUG_TX_QUEUES(trans, "hwq %d already stopped\n",
-				    txq->id);
-	}
-}
-
-/**
- * iwl_txq_inc_wrap - increment queue index, wrap back to beginning
- * @trans: the transport (for configuration data)
- * @index: current index
- */
-static inline int iwl_txq_inc_wrap(struct iwl_trans *trans, int index)
-{
-	return ++index &
-		(trans->trans_cfg->base_params->max_tfd_queue_size - 1);
-}
-
-/**
- * iwl_txq_dec_wrap - decrement queue index, wrap back to end
- * @trans: the transport (for configuration data)
- * @index: current index
- */
-static inline int iwl_txq_dec_wrap(struct iwl_trans *trans, int index)
-{
-	return --index &
-		(trans->trans_cfg->base_params->max_tfd_queue_size - 1);
-}
-
-static inline bool iwl_txq_used(const struct iwl_txq *q, int i)
-{
-	int index = iwl_txq_get_cmd_index(q, i);
-	int r = iwl_txq_get_cmd_index(q, q->read_ptr);
-	int w = iwl_txq_get_cmd_index(q, q->write_ptr);
-
-	return w >= r ?
-		(index >= r && index < w) :
-		!(index < r && index >= w);
-}
-
-void iwl_txq_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb);
-
-void iwl_txq_log_scd_error(struct iwl_trans *trans, struct iwl_txq *txq);
-
-int iwl_txq_gen2_set_tb(struct iwl_trans *trans,
-			struct iwl_tfh_tfd *tfd, dma_addr_t addr,
-			u16 len);
-
-void iwl_txq_gen2_tfd_unmap(struct iwl_trans *trans,
-			    struct iwl_cmd_meta *meta,
-			    struct iwl_tfh_tfd *tfd);
-
-int iwl_txq_dyn_alloc(struct iwl_trans *trans, u32 flags,
-		      u32 sta_mask, u8 tid,
-		      int size, unsigned int timeout);
-
-int iwl_txq_gen2_tx(struct iwl_trans *trans, struct sk_buff *skb,
-		    struct iwl_device_tx_cmd *dev_cmd, int txq_id);
-
-void iwl_txq_dyn_free(struct iwl_trans *trans, int queue);
-void iwl_txq_gen2_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq);
-void iwl_txq_inc_wr_ptr(struct iwl_trans *trans, struct iwl_txq *txq);
-void iwl_txq_gen2_tx_free(struct iwl_trans *trans);
-int iwl_txq_init(struct iwl_trans *trans, struct iwl_txq *txq, int slots_num,
-		 bool cmd_queue);
-int iwl_txq_gen2_init(struct iwl_trans *trans, int txq_id, int queue_size);
-#ifdef CONFIG_INET
-struct iwl_tso_hdr_page *get_page_hdr(struct iwl_trans *trans, size_t len,
-				      struct sk_buff *skb);
-#endif
-static inline u8 iwl_txq_gen1_tfd_get_num_tbs(struct iwl_trans *trans,
-					      struct iwl_tfd *tfd)
-{
-	return tfd->num_tbs & 0x1f;
-}
-
-static inline u16 iwl_txq_gen1_tfd_tb_get_len(struct iwl_trans *trans,
-					      void *_tfd, u8 idx)
-{
-	struct iwl_tfd *tfd;
-	struct iwl_tfd_tb *tb;
-
-	if (trans->trans_cfg->gen2) {
-		struct iwl_tfh_tfd *tfh_tfd = _tfd;
-		struct iwl_tfh_tb *tfh_tb = &tfh_tfd->tbs[idx];
-
-		return le16_to_cpu(tfh_tb->tb_len);
-	}
-
-	tfd = (struct iwl_tfd *)_tfd;
-	tb = &tfd->tbs[idx];
-
-	return le16_to_cpu(tb->hi_n_len) >> 4;
-}
-
-static inline void iwl_pcie_gen1_tfd_set_tb(struct iwl_trans *trans,
-					    struct iwl_tfd *tfd,
-					    u8 idx, dma_addr_t addr, u16 len)
-{
-	struct iwl_tfd_tb *tb = &tfd->tbs[idx];
-	u16 hi_n_len = len << 4;
-
-	put_unaligned_le32(addr, &tb->lo);
-	hi_n_len |= iwl_get_dma_hi_addr(addr);
-
-	tb->hi_n_len = cpu_to_le16(hi_n_len);
-
-	tfd->num_tbs = idx + 1;
-}
-
-void iwl_txq_gen1_tfd_unmap(struct iwl_trans *trans,
-			    struct iwl_cmd_meta *meta,
-			    struct iwl_txq *txq, int index);
-void iwl_txq_gen1_inval_byte_cnt_tbl(struct iwl_trans *trans,
-				     struct iwl_txq *txq);
-void iwl_txq_gen1_update_byte_cnt_tbl(struct iwl_trans *trans,
-				      struct iwl_txq *txq, u16 byte_cnt,
-				      int num_tbs);
-void iwl_txq_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
-		     struct sk_buff_head *skbs, bool is_flush);
-void iwl_txq_set_q_ptrs(struct iwl_trans *trans, int txq_id, int ptr);
-void iwl_trans_txq_freeze_timer(struct iwl_trans *trans, unsigned long txqs,
-				bool freeze);
-void iwl_txq_progress(struct iwl_txq *txq);
-void iwl_txq_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq);
-int iwl_trans_txq_send_hcmd(struct iwl_trans *trans, struct iwl_host_cmd *cmd);
-#endif /* __iwl_trans_queue_tx_h__ */
-- 
cgit v1.2.3-58-ga151


From 0d91a2bfc3dcae0a586e67c904abef002c848e6c Mon Sep 17 00:00:00 2001
From: Yedidya Benshimol <yedidya.ben.shimol@intel.com>
Date: Wed, 5 Jun 2024 14:05:03 +0300
Subject: wifi: iwlwifi: remove struct iwl_trans_ops

This was needed when we had multiple types of transports. Now we only
have pcie, so there is no need for this ops.
Cleanup the code such as the different trans APIs will call the pcie
function directly, instead of calling the callback,
and remove struct iwl_trans_ops.

Signed-off-by: Yedidya Benshimol <yedidya.ben.shimol@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605140327.8315ff64f9f3.Ifdbc1f26d49766f7de553dcb5f613885f4ee65cc@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c  |   6 +-
 drivers/net/wireless/intel/iwlwifi/fw/dbg.c        |   6 +-
 drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c   |   6 -
 drivers/net/wireless/intel/iwlwifi/iwl-drv.c       |   2 +-
 drivers/net/wireless/intel/iwlwifi/iwl-trans.c     | 391 +++++++++++++-
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h     | 566 +++------------------
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |   2 -
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h |  43 ++
 drivers/net/wireless/intel/iwlwifi/pcie/trans.c    | 167 ++----
 drivers/net/wireless/intel/iwlwifi/pcie/tx.c       |  12 +-
 10 files changed, 552 insertions(+), 649 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
index 52b008ce53bd..826f129ecefa 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
@@ -2,7 +2,7 @@
 /******************************************************************************
  *
  * Copyright(c) 2003 - 2014 Intel Corporation. All rights reserved.
- * Copyright(C) 2018 - 2019, 2022 - 2023 Intel Corporation
+ * Copyright(C) 2018 - 2019, 2022 - 2024 Intel Corporation
  *
  * Portions of this file are derived from the ipw3945 project, as well
  * as portions of the ieee80211 subsystem header files.
@@ -145,8 +145,6 @@ int iwlagn_mac_setup_register(struct iwl_priv *priv,
 
 #ifdef CONFIG_PM_SLEEP
 	if (priv->fw->img[IWL_UCODE_WOWLAN].num_sec &&
-	    priv->trans->ops->d3_suspend &&
-	    priv->trans->ops->d3_resume &&
 	    device_can_wakeup(priv->trans->dev)) {
 		priv->wowlan_support.flags = WIPHY_WOWLAN_MAGIC_PKT |
 					     WIPHY_WOWLAN_DISCONNECT |
@@ -730,8 +728,6 @@ static int iwlagn_mac_ampdu_action(struct ieee80211_hw *hw,
 		ret = iwl_sta_rx_agg_stop(priv, sta, tid);
 		break;
 	case IEEE80211_AMPDU_TX_START:
-		if (!priv->trans->ops->txq_enable)
-			break;
 		if (!iwl_enable_tx_ampdu())
 			break;
 		IWL_DEBUG_HT(priv, "start Tx\n");
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
index 945ffc083d25..fa57df336785 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c
@@ -1168,17 +1168,13 @@ static int iwl_dump_ini_config_iter(struct iwl_fw_runtime *fwrt,
 		   le32_to_cpu(reg->dev_addr.offset);
 	int i;
 
-	/* we shouldn't get here if the trans doesn't have read_config32 */
-	if (WARN_ON_ONCE(!trans->ops->read_config32))
-		return -EOPNOTSUPP;
-
 	range->internal_base_addr = cpu_to_le32(addr);
 	range->range_data_size = reg->dev_addr.size;
 	for (i = 0; i < le32_to_cpu(reg->dev_addr.size); i += 4) {
 		int ret;
 		u32 tmp;
 
-		ret = trans->ops->read_config32(trans, addr + i, &tmp);
+		ret = iwl_trans_read_config32(trans, addr + i, &tmp);
 		if (ret < 0)
 			return ret;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
index 561d0c261123..9810a7f4a591 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
@@ -223,12 +223,6 @@ static int iwl_dbg_tlv_alloc_region(struct iwl_trans *trans,
 		return -EINVAL;
 	}
 
-	if (type == IWL_FW_INI_REGION_PCI_IOSF_CONFIG &&
-	    !trans->ops->read_config32) {
-		IWL_ERR(trans, "WRT: Unsupported region type %u\n", type);
-		return -EOPNOTSUPP;
-	}
-
 	if (type == IWL_FW_INI_REGION_INTERNAL_BUFFER) {
 		trans->dbg.imr_data.sram_addr =
 			le32_to_cpu(reg->internal_buffer.base_addr);
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
index 249dfd207fcf..aaaabd67f959 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
@@ -1836,7 +1836,7 @@ void iwl_drv_stop(struct iwl_drv *drv)
 	mutex_unlock(&iwlwifi_opmode_table_mtx);
 
 #ifdef CONFIG_IWLWIFI_DEBUGFS
-	drv->trans->ops->debugfs_cleanup(drv->trans);
+	iwl_trans_debugfs_cleanup(drv->trans);
 
 	debugfs_remove_recursive(drv->dbgfs_drv);
 #endif
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.c b/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
index 327db93e8d84..eeee5c35668e 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
@@ -13,10 +13,11 @@
 #include "iwl-fh.h"
 #include <linux/dmapool.h>
 #include "fw/api/commands.h"
+#include "pcie/internal.h"
+#include "iwl-context-info-gen3.h"
 
 struct iwl_trans *iwl_trans_alloc(unsigned int priv_size,
 				  struct device *dev,
-				  const struct iwl_trans_ops *ops,
 				  const struct iwl_cfg_trans_params *cfg_trans)
 {
 	struct iwl_trans *trans;
@@ -36,11 +37,8 @@ struct iwl_trans *iwl_trans_alloc(unsigned int priv_size,
 #endif
 
 	trans->dev = dev;
-	trans->ops = ops;
 	trans->num_rx_queues = 1;
 
-	WARN_ON(!ops->wait_txq_empty && !ops->wait_tx_queues_empty);
-
 	return trans;
 }
 
@@ -190,3 +188,388 @@ int iwl_cmd_groups_verify_sorted(const struct iwl_trans_config *trans)
 	return 0;
 }
 IWL_EXPORT_SYMBOL(iwl_cmd_groups_verify_sorted);
+
+void iwl_trans_configure(struct iwl_trans *trans,
+			 const struct iwl_trans_config *trans_cfg)
+{
+	trans->op_mode = trans_cfg->op_mode;
+
+	iwl_trans_pcie_configure(trans, trans_cfg);
+	WARN_ON(iwl_cmd_groups_verify_sorted(trans_cfg));
+}
+IWL_EXPORT_SYMBOL(iwl_trans_configure);
+
+int iwl_trans_start_hw(struct iwl_trans *trans)
+{
+	might_sleep();
+
+	return iwl_trans_pcie_start_hw(trans);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_start_hw);
+
+void iwl_trans_op_mode_leave(struct iwl_trans *trans)
+{
+	might_sleep();
+
+	iwl_trans_pcie_op_mode_leave(trans);
+
+	trans->op_mode = NULL;
+
+	trans->state = IWL_TRANS_NO_FW;
+}
+IWL_EXPORT_SYMBOL(iwl_trans_op_mode_leave);
+
+void iwl_trans_write8(struct iwl_trans *trans, u32 ofs, u8 val)
+{
+	iwl_trans_pcie_write8(trans, ofs, val);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_write8);
+
+void iwl_trans_write32(struct iwl_trans *trans, u32 ofs, u32 val)
+{
+	iwl_trans_pcie_write32(trans, ofs, val);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_write32);
+
+u32 iwl_trans_read32(struct iwl_trans *trans, u32 ofs)
+{
+	return iwl_trans_pcie_read32(trans, ofs);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_read32);
+
+u32 iwl_trans_read_prph(struct iwl_trans *trans, u32 ofs)
+{
+	return iwl_trans_pcie_read_prph(trans, ofs);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_read_prph);
+
+void iwl_trans_write_prph(struct iwl_trans *trans, u32 ofs, u32 val)
+{
+	return iwl_trans_pcie_write_prph(trans, ofs, val);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_write_prph);
+
+int iwl_trans_read_mem(struct iwl_trans *trans, u32 addr,
+		       void *buf, int dwords)
+{
+	return iwl_trans_pcie_read_mem(trans, addr, buf, dwords);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_read_mem);
+
+int iwl_trans_write_mem(struct iwl_trans *trans, u32 addr,
+			const void *buf, int dwords)
+{
+	return iwl_trans_pcie_write_mem(trans, addr, buf, dwords);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_write_mem);
+
+void iwl_trans_set_pmi(struct iwl_trans *trans, bool state)
+{
+	if (state)
+		set_bit(STATUS_TPOWER_PMI, &trans->status);
+	else
+		clear_bit(STATUS_TPOWER_PMI, &trans->status);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_set_pmi);
+
+int iwl_trans_sw_reset(struct iwl_trans *trans, bool retake_ownership)
+{
+	return iwl_trans_pcie_sw_reset(trans, retake_ownership);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_sw_reset);
+
+struct iwl_trans_dump_data *
+iwl_trans_dump_data(struct iwl_trans *trans, u32 dump_mask,
+		    const struct iwl_dump_sanitize_ops *sanitize_ops,
+		    void *sanitize_ctx)
+{
+	return iwl_trans_pcie_dump_data(trans, dump_mask,
+					sanitize_ops, sanitize_ctx);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_dump_data);
+
+int iwl_trans_d3_suspend(struct iwl_trans *trans, bool test, bool reset)
+{
+	might_sleep();
+
+	return iwl_trans_pcie_d3_suspend(trans, test, reset);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_d3_suspend);
+
+int iwl_trans_d3_resume(struct iwl_trans *trans, enum iwl_d3_status *status,
+			bool test, bool reset)
+{
+	might_sleep();
+
+	return iwl_trans_pcie_d3_resume(trans, status, test, reset);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_d3_resume);
+
+void iwl_trans_interrupts(struct iwl_trans *trans, bool enable)
+{
+	iwl_trans_pci_interrupts(trans, enable);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_interrupts);
+
+void iwl_trans_sync_nmi(struct iwl_trans *trans)
+{
+	iwl_trans_pcie_sync_nmi(trans);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_sync_nmi);
+
+int iwl_trans_write_imr_mem(struct iwl_trans *trans, u32 dst_addr,
+			    u64 src_addr, u32 byte_cnt)
+{
+	return iwl_trans_pcie_copy_imr(trans, dst_addr, src_addr, byte_cnt);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_write_imr_mem);
+
+void iwl_trans_set_bits_mask(struct iwl_trans *trans, u32 reg,
+			     u32 mask, u32 value)
+{
+	iwl_trans_pcie_set_bits_mask(trans, reg, mask, value);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_set_bits_mask);
+
+int iwl_trans_read_config32(struct iwl_trans *trans, u32 ofs,
+			    u32 *val)
+{
+	return iwl_trans_pcie_read_config32(trans, ofs, val);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_read_config32);
+
+bool _iwl_trans_grab_nic_access(struct iwl_trans *trans)
+{
+	return iwl_trans_pcie_grab_nic_access(trans);
+}
+IWL_EXPORT_SYMBOL(_iwl_trans_grab_nic_access);
+
+void __releases(nic_access)
+iwl_trans_release_nic_access(struct iwl_trans *trans)
+{
+	iwl_trans_pcie_release_nic_access(trans);
+	__release(nic_access);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_release_nic_access);
+
+void iwl_trans_fw_alive(struct iwl_trans *trans, u32 scd_addr)
+{
+	might_sleep();
+
+	trans->state = IWL_TRANS_FW_ALIVE;
+
+	if (trans->trans_cfg->gen2)
+		iwl_trans_pcie_gen2_fw_alive(trans, scd_addr);
+	else
+		iwl_trans_pcie_fw_alive(trans, scd_addr);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_fw_alive);
+
+int iwl_trans_start_fw(struct iwl_trans *trans, const struct fw_img *fw,
+		       bool run_in_rfkill)
+{
+	int ret;
+
+	might_sleep();
+
+	WARN_ON_ONCE(!trans->rx_mpdu_cmd);
+
+	clear_bit(STATUS_FW_ERROR, &trans->status);
+
+	if (trans->trans_cfg->gen2)
+		ret = iwl_trans_pcie_gen2_start_fw(trans, fw, run_in_rfkill);
+	else
+		ret = iwl_trans_pcie_start_fw(trans, fw, run_in_rfkill);
+
+	if (ret == 0)
+		trans->state = IWL_TRANS_FW_STARTED;
+
+	return ret;
+}
+IWL_EXPORT_SYMBOL(iwl_trans_start_fw);
+
+void iwl_trans_stop_device(struct iwl_trans *trans)
+{
+	might_sleep();
+
+	if (trans->trans_cfg->gen2)
+		iwl_trans_pcie_gen2_stop_device(trans);
+	else
+		iwl_trans_pcie_stop_device(trans);
+
+	trans->state = IWL_TRANS_NO_FW;
+}
+IWL_EXPORT_SYMBOL(iwl_trans_stop_device);
+
+int iwl_trans_tx(struct iwl_trans *trans, struct sk_buff *skb,
+		 struct iwl_device_tx_cmd *dev_cmd, int queue)
+{
+	if (unlikely(test_bit(STATUS_FW_ERROR, &trans->status)))
+		return -EIO;
+
+	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+		return -EIO;
+	}
+
+	if (trans->trans_cfg->gen2)
+		return iwl_txq_gen2_tx(trans, skb, dev_cmd, queue);
+
+	return iwl_trans_pcie_tx(trans, skb, dev_cmd, queue);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_tx);
+
+void iwl_trans_reclaim(struct iwl_trans *trans, int queue, int ssn,
+		       struct sk_buff_head *skbs, bool is_flush)
+{
+	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+		return;
+	}
+
+	iwl_pcie_reclaim(trans, queue, ssn, skbs, is_flush);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_reclaim);
+
+void iwl_trans_txq_disable(struct iwl_trans *trans, int queue,
+			   bool configure_scd)
+{
+	iwl_trans_pcie_txq_disable(trans, queue, configure_scd);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_txq_disable);
+
+bool iwl_trans_txq_enable_cfg(struct iwl_trans *trans, int queue, u16 ssn,
+			      const struct iwl_trans_txq_scd_cfg *cfg,
+			      unsigned int queue_wdg_timeout)
+{
+	might_sleep();
+
+	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+		return false;
+	}
+
+	return iwl_trans_pcie_txq_enable(trans, queue, ssn,
+					 cfg, queue_wdg_timeout);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_txq_enable_cfg);
+
+int iwl_trans_wait_txq_empty(struct iwl_trans *trans, int queue)
+{
+	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+		return -EIO;
+	}
+
+	return iwl_trans_pcie_wait_txq_empty(trans, queue);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_wait_txq_empty);
+
+int iwl_trans_wait_tx_queues_empty(struct iwl_trans *trans, u32 txqs)
+{
+	/* No need to wait if the firmware is not alive */
+	if (trans->state != IWL_TRANS_FW_ALIVE) {
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+		return -EIO;
+	}
+
+	return iwl_trans_pcie_wait_txqs_empty(trans, txqs);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_wait_tx_queues_empty);
+
+void iwl_trans_freeze_txq_timer(struct iwl_trans *trans,
+				unsigned long txqs, bool freeze)
+{
+	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+		return;
+	}
+
+	iwl_pcie_freeze_txq_timer(trans, txqs, freeze);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_freeze_txq_timer);
+
+void iwl_trans_txq_set_shared_mode(struct iwl_trans *trans,
+				   int txq_id, bool shared_mode)
+{
+	iwl_trans_pcie_txq_set_shared_mode(trans, txq_id, shared_mode);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_txq_set_shared_mode);
+
+#ifdef CONFIG_IWLWIFI_DEBUGFS
+void iwl_trans_debugfs_cleanup(struct iwl_trans *trans)
+{
+	iwl_trans_pcie_debugfs_cleanup(trans);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_debugfs_cleanup);
+#endif
+
+void iwl_trans_set_q_ptrs(struct iwl_trans *trans, int queue, int ptr)
+{
+	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+		return;
+	}
+
+	iwl_pcie_set_q_ptrs(trans, queue, ptr);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_set_q_ptrs);
+
+int iwl_trans_txq_alloc(struct iwl_trans *trans, u32 flags, u32 sta_mask,
+			u8 tid, int size, unsigned int wdg_timeout)
+{
+	might_sleep();
+
+	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
+		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+		return -EIO;
+	}
+
+	return iwl_txq_dyn_alloc(trans, flags, sta_mask, tid,
+				 size, wdg_timeout);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_txq_alloc);
+
+void iwl_trans_txq_free(struct iwl_trans *trans, int queue)
+{
+	iwl_txq_dyn_free(trans, queue);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_txq_free);
+
+int iwl_trans_get_rxq_dma_data(struct iwl_trans *trans, int queue,
+			       struct iwl_trans_rxq_dma_data *data)
+{
+	return iwl_trans_pcie_rxq_dma_data(trans, queue, data);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_get_rxq_dma_data);
+
+int iwl_trans_load_pnvm(struct iwl_trans *trans,
+			const struct iwl_pnvm_image *pnvm_data,
+			const struct iwl_ucode_capabilities *capa)
+{
+	return iwl_trans_pcie_ctx_info_gen3_load_pnvm(trans, pnvm_data, capa);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_load_pnvm);
+
+void iwl_trans_set_pnvm(struct iwl_trans *trans,
+			const struct iwl_ucode_capabilities *capa)
+{
+	iwl_trans_pcie_ctx_info_gen3_set_pnvm(trans, capa);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_set_pnvm);
+
+int iwl_trans_load_reduce_power(struct iwl_trans *trans,
+				const struct iwl_pnvm_image *payloads,
+				const struct iwl_ucode_capabilities *capa)
+{
+	return iwl_trans_pcie_ctx_info_gen3_load_reduce_power(trans, payloads,
+							      capa);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_load_reduce_power);
+
+void iwl_trans_set_reduce_power(struct iwl_trans *trans,
+				const struct iwl_ucode_capabilities *capa)
+{
+	iwl_trans_pcie_ctx_info_gen3_set_reduce_power(trans, capa);
+}
+IWL_EXPORT_SYMBOL(iwl_trans_set_reduce_power);
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index 27f92b393f8e..006eb51a705f 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -473,183 +473,6 @@ struct iwl_pnvm_image {
 	u32 version;
 };
 
-/**
- * struct iwl_trans_ops - transport specific operations
- *
- * All the handlers MUST be implemented
- *
- * @start_hw: starts the HW. From that point on, the HW can send interrupts.
- *	May sleep.
- * @op_mode_leave: Turn off the HW RF kill indication if on
- *	May sleep
- * @start_fw: allocates and inits all the resources for the transport
- *	layer. Also kick a fw image.
- *	May sleep
- * @fw_alive: called when the fw sends alive notification. If the fw provides
- *	the SCD base address in SRAM, then provide it here, or 0 otherwise.
- *	May sleep
- * @stop_device: stops the whole device (embedded CPU put to reset) and stops
- *	the HW. From that point on, the HW will be stopped but will still issue
- *	an interrupt if the HW RF kill switch is triggered.
- *	This callback must do the right thing and not crash even if %start_hw()
- *	was called but not &start_fw(). May sleep.
- * @d3_suspend: put the device into the correct mode for WoWLAN during
- *	suspend. This is optional, if not implemented WoWLAN will not be
- *	supported. This callback may sleep.
- * @d3_resume: resume the device after WoWLAN, enabling the opmode to
- *	talk to the WoWLAN image to get its status. This is optional, if not
- *	implemented WoWLAN will not be supported. This callback may sleep.
- * @send_cmd:send a host command. Must return -ERFKILL if RFkill is asserted.
- *	If RFkill is asserted in the middle of a SYNC host command, it must
- *	return -ERFKILL straight away.
- *	May sleep only if CMD_ASYNC is not set
- * @tx: send an skb. The transport relies on the op_mode to zero the
- *	the ieee80211_tx_info->driver_data. If the MPDU is an A-MSDU, all
- *	the CSUM will be taken care of (TCP CSUM and IP header in case of
- *	IPv4). If the MPDU is a single MSDU, the op_mode must compute the IP
- *	header if it is IPv4.
- *	Must be atomic
- * @reclaim: free packet until ssn. Returns a list of freed packets.
- *	Must be atomic
- * @set_q_ptrs: set queue pointers internally, after D3 when HW state changed
- * @txq_enable: setup a queue. To setup an AC queue, use the
- *	iwl_trans_ac_txq_enable wrapper. fw_alive must have been called before
- *	this one. The op_mode must not configure the HCMD queue. The scheduler
- *	configuration may be %NULL, in which case the hardware will not be
- *	configured. If true is returned, the operation mode needs to increment
- *	the sequence number of the packets routed to this queue because of a
- *	hardware scheduler bug. May sleep.
- * @txq_disable: de-configure a Tx queue to send AMPDUs
- *	Must be atomic
- * @txq_alloc: Allocate a new TX queue, may sleep.
- * @txq_free: Free a previously allocated TX queue.
- * @txq_set_shared_mode: change Tx queue shared/unshared marking
- * @wait_tx_queues_empty: wait until tx queues are empty. May sleep.
- * @wait_txq_empty: wait until specific tx queue is empty. May sleep.
- * @freeze_txq_timer: prevents the timer of the queue from firing until the
- *	queue is set to awake. Must be atomic.
- * @write8: write a u8 to a register at offset ofs from the BAR
- * @write32: write a u32 to a register at offset ofs from the BAR
- * @read32: read a u32 register at offset ofs from the BAR
- * @read_prph: read a DWORD from a periphery register
- * @write_prph: write a DWORD to a periphery register
- * @read_mem: read device's SRAM in DWORD
- * @write_mem: write device's SRAM in DWORD. If %buf is %NULL, then the memory
- *	will be zeroed.
- * @read_config32: read a u32 value from the device's config space at
- *	the given offset.
- * @configure: configure parameters required by the transport layer from
- *	the op_mode. May be called several times before start_fw, can't be
- *	called after that.
- * @set_pmi: set the power pmi state
- * @sw_reset: trigger software reset of the NIC
- * @grab_nic_access: wake the NIC to be able to access non-HBUS regs.
- *	Sleeping is not allowed between grab_nic_access and
- *	release_nic_access.
- * @release_nic_access: let the NIC go to sleep. The "flags" parameter
- *	must be the same one that was sent before to the grab_nic_access.
- * @set_bits_mask: set SRAM register according to value and mask.
- * @dump_data: return a vmalloc'ed buffer with debug data, maybe containing last
- *	TX'ed commands and similar. The buffer will be vfree'd by the caller.
- *	Note that the transport must fill in the proper file headers.
- * @debugfs_cleanup: used in the driver unload flow to make a proper cleanup
- *	of the trans debugfs
- * @sync_nmi: trigger a firmware NMI and wait for it to complete
- * @load_pnvm: save the pnvm data in DRAM
- * @set_pnvm: set the pnvm data in the prph scratch buffer, inside the
- *	context info.
- * @load_reduce_power: copy reduce power table to the corresponding DRAM memory
- * @set_reduce_power: set reduce power table addresses in the sratch buffer
- * @interrupts: disable/enable interrupts to transport
- * @imr_dma_data: set up IMR DMA
- * @rxq_dma_data: retrieve RX queue DMA data, see @struct iwl_trans_rxq_dma_data
- */
-struct iwl_trans_ops {
-
-	int (*start_hw)(struct iwl_trans *iwl_trans);
-	void (*op_mode_leave)(struct iwl_trans *iwl_trans);
-	int (*start_fw)(struct iwl_trans *trans, const struct fw_img *fw,
-			bool run_in_rfkill);
-	void (*fw_alive)(struct iwl_trans *trans, u32 scd_addr);
-	void (*stop_device)(struct iwl_trans *trans);
-
-	int (*d3_suspend)(struct iwl_trans *trans, bool test, bool reset);
-	int (*d3_resume)(struct iwl_trans *trans, enum iwl_d3_status *status,
-			 bool test, bool reset);
-
-	int (*send_cmd)(struct iwl_trans *trans, struct iwl_host_cmd *cmd);
-
-	int (*tx)(struct iwl_trans *trans, struct sk_buff *skb,
-		  struct iwl_device_tx_cmd *dev_cmd, int queue);
-	void (*reclaim)(struct iwl_trans *trans, int queue, int ssn,
-			struct sk_buff_head *skbs, bool is_flush);
-
-	void (*set_q_ptrs)(struct iwl_trans *trans, int queue, int ptr);
-
-	bool (*txq_enable)(struct iwl_trans *trans, int queue, u16 ssn,
-			   const struct iwl_trans_txq_scd_cfg *cfg,
-			   unsigned int queue_wdg_timeout);
-	void (*txq_disable)(struct iwl_trans *trans, int queue,
-			    bool configure_scd);
-	/* 22000 functions */
-	int (*txq_alloc)(struct iwl_trans *trans, u32 flags,
-			 u32 sta_mask, u8 tid,
-			 int size, unsigned int queue_wdg_timeout);
-	void (*txq_free)(struct iwl_trans *trans, int queue);
-	int (*rxq_dma_data)(struct iwl_trans *trans, int queue,
-			    struct iwl_trans_rxq_dma_data *data);
-
-	void (*txq_set_shared_mode)(struct iwl_trans *trans, u32 txq_id,
-				    bool shared);
-
-	int (*wait_tx_queues_empty)(struct iwl_trans *trans, u32 txq_bm);
-	int (*wait_txq_empty)(struct iwl_trans *trans, int queue);
-	void (*freeze_txq_timer)(struct iwl_trans *trans, unsigned long txqs,
-				 bool freeze);
-
-	void (*write8)(struct iwl_trans *trans, u32 ofs, u8 val);
-	void (*write32)(struct iwl_trans *trans, u32 ofs, u32 val);
-	u32 (*read32)(struct iwl_trans *trans, u32 ofs);
-	u32 (*read_prph)(struct iwl_trans *trans, u32 ofs);
-	void (*write_prph)(struct iwl_trans *trans, u32 ofs, u32 val);
-	int (*read_mem)(struct iwl_trans *trans, u32 addr,
-			void *buf, int dwords);
-	int (*write_mem)(struct iwl_trans *trans, u32 addr,
-			 const void *buf, int dwords);
-	int (*read_config32)(struct iwl_trans *trans, u32 ofs, u32 *val);
-	void (*configure)(struct iwl_trans *trans,
-			  const struct iwl_trans_config *trans_cfg);
-	void (*set_pmi)(struct iwl_trans *trans, bool state);
-	int (*sw_reset)(struct iwl_trans *trans, bool retake_ownership);
-	bool (*grab_nic_access)(struct iwl_trans *trans);
-	void (*release_nic_access)(struct iwl_trans *trans);
-	void (*set_bits_mask)(struct iwl_trans *trans, u32 reg, u32 mask,
-			      u32 value);
-
-	struct iwl_trans_dump_data *(*dump_data)(struct iwl_trans *trans,
-						 u32 dump_mask,
-						 const struct iwl_dump_sanitize_ops *sanitize_ops,
-						 void *sanitize_ctx);
-	void (*debugfs_cleanup)(struct iwl_trans *trans);
-	void (*sync_nmi)(struct iwl_trans *trans);
-	int (*load_pnvm)(struct iwl_trans *trans,
-			 const struct iwl_pnvm_image *pnvm_payloads,
-			 const struct iwl_ucode_capabilities *capa);
-	void (*set_pnvm)(struct iwl_trans *trans,
-			 const struct iwl_ucode_capabilities *capa);
-	int (*load_reduce_power)(struct iwl_trans *trans,
-				 const struct iwl_pnvm_image *payloads,
-				 const struct iwl_ucode_capabilities *capa);
-	void (*set_reduce_power)(struct iwl_trans *trans,
-				 const struct iwl_ucode_capabilities *capa);
-
-	void (*interrupts)(struct iwl_trans *trans, bool enable);
-	int (*imr_dma_data)(struct iwl_trans *trans,
-			    u32 dst_addr, u64 src_addr,
-			    u32 byte_cnt);
-
-};
-
 /**
  * enum iwl_trans_state - state of the transport layer
  *
@@ -994,7 +817,6 @@ struct iwl_txq {
  * struct iwl_trans - transport common data
  *
  * @csme_own: true if we couldn't get ownership on the device
- * @ops: pointer to iwl_trans_ops
  * @op_mode: pointer to the op_mode
  * @trans_cfg: the trans-specific configuration part
  * @cfg: pointer to the configuration
@@ -1055,7 +877,6 @@ struct iwl_txq {
  */
 struct iwl_trans {
 	bool csme_own;
-	const struct iwl_trans_ops *ops;
 	struct iwl_op_mode *op_mode;
 	const struct iwl_cfg_trans_params *trans_cfg;
 	const struct iwl_cfg *cfg;
@@ -1127,101 +948,29 @@ struct iwl_trans {
 const char *iwl_get_cmd_string(struct iwl_trans *trans, u32 id);
 int iwl_cmd_groups_verify_sorted(const struct iwl_trans_config *trans);
 
-static inline void iwl_trans_configure(struct iwl_trans *trans,
-				       const struct iwl_trans_config *trans_cfg)
-{
-	trans->op_mode = trans_cfg->op_mode;
-
-	trans->ops->configure(trans, trans_cfg);
-	WARN_ON(iwl_cmd_groups_verify_sorted(trans_cfg));
-}
-
-static inline int iwl_trans_start_hw(struct iwl_trans *trans)
-{
-	might_sleep();
-
-	return trans->ops->start_hw(trans);
-}
-
-static inline void iwl_trans_op_mode_leave(struct iwl_trans *trans)
-{
-	might_sleep();
-
-	if (trans->ops->op_mode_leave)
-		trans->ops->op_mode_leave(trans);
+void iwl_trans_configure(struct iwl_trans *trans,
+			 const struct iwl_trans_config *trans_cfg);
 
-	trans->op_mode = NULL;
-
-	trans->state = IWL_TRANS_NO_FW;
-}
-
-static inline void iwl_trans_fw_alive(struct iwl_trans *trans, u32 scd_addr)
-{
-	might_sleep();
-
-	trans->state = IWL_TRANS_FW_ALIVE;
-
-	trans->ops->fw_alive(trans, scd_addr);
-}
-
-static inline int iwl_trans_start_fw(struct iwl_trans *trans,
-				     const struct fw_img *fw,
-				     bool run_in_rfkill)
-{
-	int ret;
-
-	might_sleep();
-
-	WARN_ON_ONCE(!trans->rx_mpdu_cmd);
-
-	clear_bit(STATUS_FW_ERROR, &trans->status);
-	ret = trans->ops->start_fw(trans, fw, run_in_rfkill);
-	if (ret == 0)
-		trans->state = IWL_TRANS_FW_STARTED;
-
-	return ret;
-}
-
-static inline void iwl_trans_stop_device(struct iwl_trans *trans)
-{
-	might_sleep();
+int iwl_trans_start_hw(struct iwl_trans *trans);
 
-	trans->ops->stop_device(trans);
+void iwl_trans_op_mode_leave(struct iwl_trans *trans);
 
-	trans->state = IWL_TRANS_NO_FW;
-}
+void iwl_trans_fw_alive(struct iwl_trans *trans, u32 scd_addr);
 
-static inline int iwl_trans_d3_suspend(struct iwl_trans *trans, bool test,
-				       bool reset)
-{
-	might_sleep();
-	if (!trans->ops->d3_suspend)
-		return -EOPNOTSUPP;
+int iwl_trans_start_fw(struct iwl_trans *trans, const struct fw_img *fw,
+		       bool run_in_rfkill);
 
-	return trans->ops->d3_suspend(trans, test, reset);
-}
+void iwl_trans_stop_device(struct iwl_trans *trans);
 
-static inline int iwl_trans_d3_resume(struct iwl_trans *trans,
-				      enum iwl_d3_status *status,
-				      bool test, bool reset)
-{
-	might_sleep();
-	if (!trans->ops->d3_resume)
-		return -EOPNOTSUPP;
+int iwl_trans_d3_suspend(struct iwl_trans *trans, bool test, bool reset);
 
-	return trans->ops->d3_resume(trans, status, test, reset);
-}
+int iwl_trans_d3_resume(struct iwl_trans *trans, enum iwl_d3_status *status,
+			bool test, bool reset);
 
-static inline struct iwl_trans_dump_data *
+struct iwl_trans_dump_data *
 iwl_trans_dump_data(struct iwl_trans *trans, u32 dump_mask,
 		    const struct iwl_dump_sanitize_ops *sanitize_ops,
-		    void *sanitize_ctx)
-{
-	if (!trans->ops->dump_data)
-		return NULL;
-	return trans->ops->dump_data(trans, dump_mask,
-				     sanitize_ops, sanitize_ctx);
-}
+		    void *sanitize_ctx);
 
 static inline struct iwl_device_tx_cmd *
 iwl_trans_alloc_tx_cmd(struct iwl_trans *trans)
@@ -1237,109 +986,31 @@ static inline void iwl_trans_free_tx_cmd(struct iwl_trans *trans,
 	kmem_cache_free(trans->dev_cmd_pool, dev_cmd);
 }
 
-static inline int iwl_trans_tx(struct iwl_trans *trans, struct sk_buff *skb,
-			       struct iwl_device_tx_cmd *dev_cmd, int queue)
-{
-	if (unlikely(test_bit(STATUS_FW_ERROR, &trans->status)))
-		return -EIO;
+int iwl_trans_tx(struct iwl_trans *trans, struct sk_buff *skb,
+		 struct iwl_device_tx_cmd *dev_cmd, int queue);
 
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
-		return -EIO;
-	}
+void iwl_trans_reclaim(struct iwl_trans *trans, int queue, int ssn,
+		       struct sk_buff_head *skbs, bool is_flush);
 
-	return trans->ops->tx(trans, skb, dev_cmd, queue);
-}
+void iwl_trans_set_q_ptrs(struct iwl_trans *trans, int queue, int ptr);
 
-static inline void iwl_trans_reclaim(struct iwl_trans *trans, int queue,
-				     int ssn, struct sk_buff_head *skbs,
-				     bool is_flush)
-{
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
-		return;
-	}
+void iwl_trans_txq_disable(struct iwl_trans *trans, int queue,
+			   bool configure_scd);
 
-	trans->ops->reclaim(trans, queue, ssn, skbs, is_flush);
-}
+bool iwl_trans_txq_enable_cfg(struct iwl_trans *trans, int queue, u16 ssn,
+			      const struct iwl_trans_txq_scd_cfg *cfg,
+			      unsigned int queue_wdg_timeout);
 
-static inline void iwl_trans_set_q_ptrs(struct iwl_trans *trans, int queue,
-					int ptr)
-{
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
-		return;
-	}
+int iwl_trans_get_rxq_dma_data(struct iwl_trans *trans, int queue,
+			       struct iwl_trans_rxq_dma_data *data);
 
-	trans->ops->set_q_ptrs(trans, queue, ptr);
-}
-
-static inline void iwl_trans_txq_disable(struct iwl_trans *trans, int queue,
-					 bool configure_scd)
-{
-	trans->ops->txq_disable(trans, queue, configure_scd);
-}
-
-static inline bool
-iwl_trans_txq_enable_cfg(struct iwl_trans *trans, int queue, u16 ssn,
-			 const struct iwl_trans_txq_scd_cfg *cfg,
-			 unsigned int queue_wdg_timeout)
-{
-	might_sleep();
-
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
-		return false;
-	}
+void iwl_trans_txq_free(struct iwl_trans *trans, int queue);
 
-	return trans->ops->txq_enable(trans, queue, ssn,
-				      cfg, queue_wdg_timeout);
-}
-
-static inline int
-iwl_trans_get_rxq_dma_data(struct iwl_trans *trans, int queue,
-			   struct iwl_trans_rxq_dma_data *data)
-{
-	if (WARN_ON_ONCE(!trans->ops->rxq_dma_data))
-		return -EOPNOTSUPP;
+int iwl_trans_txq_alloc(struct iwl_trans *trans, u32 flags, u32 sta_mask,
+			u8 tid, int size, unsigned int wdg_timeout);
 
-	return trans->ops->rxq_dma_data(trans, queue, data);
-}
-
-static inline void
-iwl_trans_txq_free(struct iwl_trans *trans, int queue)
-{
-	if (WARN_ON_ONCE(!trans->ops->txq_free))
-		return;
-
-	trans->ops->txq_free(trans, queue);
-}
-
-static inline int
-iwl_trans_txq_alloc(struct iwl_trans *trans,
-		    u32 flags, u32 sta_mask, u8 tid,
-		    int size, unsigned int wdg_timeout)
-{
-	might_sleep();
-
-	if (WARN_ON_ONCE(!trans->ops->txq_alloc))
-		return -EOPNOTSUPP;
-
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
-		return -EIO;
-	}
-
-	return trans->ops->txq_alloc(trans, flags, sta_mask, tid,
-				     size, wdg_timeout);
-}
-
-static inline void iwl_trans_txq_set_shared_mode(struct iwl_trans *trans,
-						 int queue, bool shared_mode)
-{
-	if (trans->ops->txq_set_shared_mode)
-		trans->ops->txq_set_shared_mode(trans, queue, shared_mode);
-}
+void iwl_trans_txq_set_shared_mode(struct iwl_trans *trans,
+				   int txq_id, bool shared_mode);
 
 static inline void iwl_trans_txq_enable(struct iwl_trans *trans, int queue,
 					int fifo, int sta_id, int tid,
@@ -1372,78 +1043,32 @@ void iwl_trans_ac_txq_enable(struct iwl_trans *trans, int queue, int fifo,
 	iwl_trans_txq_enable_cfg(trans, queue, 0, &cfg, queue_wdg_timeout);
 }
 
-static inline void iwl_trans_freeze_txq_timer(struct iwl_trans *trans,
-					      unsigned long txqs,
-					      bool freeze)
-{
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
-		return;
-	}
+void iwl_trans_freeze_txq_timer(struct iwl_trans *trans,
+				unsigned long txqs, bool freeze);
 
-	if (trans->ops->freeze_txq_timer)
-		trans->ops->freeze_txq_timer(trans, txqs, freeze);
-}
+int iwl_trans_wait_tx_queues_empty(struct iwl_trans *trans, u32 txqs);
 
-static inline int iwl_trans_wait_tx_queues_empty(struct iwl_trans *trans,
-						 u32 txqs)
-{
-	if (WARN_ON_ONCE(!trans->ops->wait_tx_queues_empty))
-		return -EOPNOTSUPP;
-
-	/* No need to wait if the firmware is not alive */
-	if (trans->state != IWL_TRANS_FW_ALIVE) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
-		return -EIO;
-	}
-
-	return trans->ops->wait_tx_queues_empty(trans, txqs);
-}
+int iwl_trans_wait_txq_empty(struct iwl_trans *trans, int queue);
 
-static inline int iwl_trans_wait_txq_empty(struct iwl_trans *trans, int queue)
-{
-	if (WARN_ON_ONCE(!trans->ops->wait_txq_empty))
-		return -EOPNOTSUPP;
+void iwl_trans_write8(struct iwl_trans *trans, u32 ofs, u8 val);
 
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
-		return -EIO;
-	}
+void iwl_trans_write32(struct iwl_trans *trans, u32 ofs, u32 val);
 
-	return trans->ops->wait_txq_empty(trans, queue);
-}
+u32 iwl_trans_read32(struct iwl_trans *trans, u32 ofs);
 
-static inline void iwl_trans_write8(struct iwl_trans *trans, u32 ofs, u8 val)
-{
-	trans->ops->write8(trans, ofs, val);
-}
+u32 iwl_trans_read_prph(struct iwl_trans *trans, u32 ofs);
 
-static inline void iwl_trans_write32(struct iwl_trans *trans, u32 ofs, u32 val)
-{
-	trans->ops->write32(trans, ofs, val);
-}
+void iwl_trans_write_prph(struct iwl_trans *trans, u32 ofs, u32 val);
 
-static inline u32 iwl_trans_read32(struct iwl_trans *trans, u32 ofs)
-{
-	return trans->ops->read32(trans, ofs);
-}
+int iwl_trans_read_mem(struct iwl_trans *trans, u32 addr,
+		       void *buf, int dwords);
 
-static inline u32 iwl_trans_read_prph(struct iwl_trans *trans, u32 ofs)
-{
-	return trans->ops->read_prph(trans, ofs);
-}
+int iwl_trans_read_config32(struct iwl_trans *trans, u32 ofs,
+			    u32 *val);
 
-static inline void iwl_trans_write_prph(struct iwl_trans *trans, u32 ofs,
-					u32 val)
-{
-	return trans->ops->write_prph(trans, ofs, val);
-}
-
-static inline int iwl_trans_read_mem(struct iwl_trans *trans, u32 addr,
-				     void *buf, int dwords)
-{
-	return trans->ops->read_mem(trans, addr, buf, dwords);
-}
+#ifdef CONFIG_IWLWIFI_DEBUGFS
+void iwl_trans_debugfs_cleanup(struct iwl_trans *trans);
+#endif
 
 #define iwl_trans_read_mem_bytes(trans, addr, buf, bufsize)		      \
 	do {								      \
@@ -1452,14 +1077,8 @@ static inline int iwl_trans_read_mem(struct iwl_trans *trans, u32 addr,
 		iwl_trans_read_mem(trans, addr, buf, (bufsize) / sizeof(u32));\
 	} while (0)
 
-static inline int iwl_trans_write_imr_mem(struct iwl_trans *trans,
-					  u32 dst_addr, u64 src_addr,
-					  u32 byte_cnt)
-{
-	if (trans->ops->imr_dma_data)
-		return trans->ops->imr_dma_data(trans, dst_addr, src_addr, byte_cnt);
-	return 0;
-}
+int iwl_trans_write_imr_mem(struct iwl_trans *trans, u32 dst_addr,
+			    u64 src_addr, u32 byte_cnt);
 
 static inline u32 iwl_trans_read_mem32(struct iwl_trans *trans, u32 addr)
 {
@@ -1471,11 +1090,8 @@ static inline u32 iwl_trans_read_mem32(struct iwl_trans *trans, u32 addr)
 	return value;
 }
 
-static inline int iwl_trans_write_mem(struct iwl_trans *trans, u32 addr,
-				      const void *buf, int dwords)
-{
-	return trans->ops->write_mem(trans, addr, buf, dwords);
-}
+int iwl_trans_write_mem(struct iwl_trans *trans, u32 addr,
+			const void *buf, int dwords);
 
 static inline u32 iwl_trans_write_mem32(struct iwl_trans *trans, u32 addr,
 					u32 val)
@@ -1483,36 +1099,21 @@ static inline u32 iwl_trans_write_mem32(struct iwl_trans *trans, u32 addr,
 	return iwl_trans_write_mem(trans, addr, &val, 1);
 }
 
-static inline void iwl_trans_set_pmi(struct iwl_trans *trans, bool state)
-{
-	if (trans->ops->set_pmi)
-		trans->ops->set_pmi(trans, state);
-}
+void iwl_trans_set_pmi(struct iwl_trans *trans, bool state);
 
-static inline int iwl_trans_sw_reset(struct iwl_trans *trans,
-				     bool retake_ownership)
-{
-	if (trans->ops->sw_reset)
-		return trans->ops->sw_reset(trans, retake_ownership);
-	return 0;
-}
+int iwl_trans_sw_reset(struct iwl_trans *trans, bool retake_ownership);
 
-static inline void
-iwl_trans_set_bits_mask(struct iwl_trans *trans, u32 reg, u32 mask, u32 value)
-{
-	trans->ops->set_bits_mask(trans, reg, mask, value);
-}
+void iwl_trans_set_bits_mask(struct iwl_trans *trans, u32 reg,
+			     u32 mask, u32 value);
+
+bool _iwl_trans_grab_nic_access(struct iwl_trans *trans);
 
 #define iwl_trans_grab_nic_access(trans)		\
 	__cond_lock(nic_access,				\
-		    likely((trans)->ops->grab_nic_access(trans)))
+		    likely(_iwl_trans_grab_nic_access(trans)))
 
-static inline void __releases(nic_access)
-iwl_trans_release_nic_access(struct iwl_trans *trans)
-{
-	trans->ops->release_nic_access(trans);
-	__release(nic_access);
-}
+void __releases(nic_access)
+iwl_trans_release_nic_access(struct iwl_trans *trans);
 
 static inline void iwl_trans_fw_error(struct iwl_trans *trans, bool sync)
 {
@@ -1531,44 +1132,24 @@ static inline bool iwl_trans_fw_running(struct iwl_trans *trans)
 	return trans->state == IWL_TRANS_FW_ALIVE;
 }
 
-static inline void iwl_trans_sync_nmi(struct iwl_trans *trans)
-{
-	if (trans->ops->sync_nmi)
-		trans->ops->sync_nmi(trans);
-}
+void iwl_trans_sync_nmi(struct iwl_trans *trans);
 
 void iwl_trans_sync_nmi_with_addr(struct iwl_trans *trans, u32 inta_addr,
 				  u32 sw_err_bit);
 
-static inline int iwl_trans_load_pnvm(struct iwl_trans *trans,
-				      const struct iwl_pnvm_image *pnvm_data,
-				      const struct iwl_ucode_capabilities *capa)
-{
-	return trans->ops->load_pnvm(trans, pnvm_data, capa);
-}
+int iwl_trans_load_pnvm(struct iwl_trans *trans,
+			const struct iwl_pnvm_image *pnvm_data,
+			const struct iwl_ucode_capabilities *capa);
 
-static inline void iwl_trans_set_pnvm(struct iwl_trans *trans,
-				      const struct iwl_ucode_capabilities *capa)
-{
-	if (trans->ops->set_pnvm)
-		trans->ops->set_pnvm(trans, capa);
-}
+void iwl_trans_set_pnvm(struct iwl_trans *trans,
+			const struct iwl_ucode_capabilities *capa);
 
-static inline int iwl_trans_load_reduce_power
-				(struct iwl_trans *trans,
-				 const struct iwl_pnvm_image *payloads,
-				 const struct iwl_ucode_capabilities *capa)
-{
-	return trans->ops->load_reduce_power(trans, payloads, capa);
-}
+int iwl_trans_load_reduce_power(struct iwl_trans *trans,
+				const struct iwl_pnvm_image *payloads,
+				const struct iwl_ucode_capabilities *capa);
 
-static inline void
-iwl_trans_set_reduce_power(struct iwl_trans *trans,
-			   const struct iwl_ucode_capabilities *capa)
-{
-	if (trans->ops->set_reduce_power)
-		trans->ops->set_reduce_power(trans, capa);
-}
+void iwl_trans_set_reduce_power(struct iwl_trans *trans,
+				const struct iwl_ucode_capabilities *capa);
 
 static inline bool iwl_trans_dbg_ini_valid(struct iwl_trans *trans)
 {
@@ -1576,18 +1157,13 @@ static inline bool iwl_trans_dbg_ini_valid(struct iwl_trans *trans)
 		trans->dbg.external_ini_cfg != IWL_INI_CFG_STATE_NOT_LOADED;
 }
 
-static inline void iwl_trans_interrupts(struct iwl_trans *trans, bool enable)
-{
-	if (trans->ops->interrupts)
-		trans->ops->interrupts(trans, enable);
-}
+void iwl_trans_interrupts(struct iwl_trans *trans, bool enable);
 
 /*****************************************************
  * transport helper functions
  *****************************************************/
 struct iwl_trans *iwl_trans_alloc(unsigned int priv_size,
 			  struct device *dev,
-			  const struct iwl_trans_ops *ops,
 			  const struct iwl_cfg_trans_params *cfg_trans);
 int iwl_trans_init(struct iwl_trans *trans);
 void iwl_trans_free(struct iwl_trans *trans);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 88bc0baabf7e..b004de6caec3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -727,8 +727,6 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 
 #ifdef CONFIG_PM_SLEEP
 	if ((unified || mvm->fw->img[IWL_UCODE_WOWLAN].num_sec) &&
-	    mvm->trans->ops->d3_suspend &&
-	    mvm->trans->ops->d3_resume &&
 	    device_can_wakeup(mvm->trans->dev)) {
 		mvm->wowlan.flags |= WIPHY_WOWLAN_MAGIC_PKT |
 				     WIPHY_WOWLAN_DISCONNECT |
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index 3bdbb907d91a..3ce626c93651 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -761,6 +761,8 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 void iwl_pcie_set_q_ptrs(struct iwl_trans *trans, int txq_id, int ptr);
 void iwl_pcie_freeze_txq_timer(struct iwl_trans *trans,
 			       unsigned long txqs, bool freeze);
+int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans, int txq_idx);
+int iwl_trans_pcie_wait_txqs_empty(struct iwl_trans *trans, u32 txq_bm);
 
 /*****************************************************
 * Error handling
@@ -1027,12 +1029,51 @@ void iwl_trans_pcie_dump_regs(struct iwl_trans *trans);
 
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 void iwl_trans_pcie_dbgfs_register(struct iwl_trans *trans);
+void iwl_trans_pcie_debugfs_cleanup(struct iwl_trans *trans);
 #else
 static inline void iwl_trans_pcie_dbgfs_register(struct iwl_trans *trans) { }
 #endif
 
 void iwl_pcie_rx_allocator_work(struct work_struct *data);
 
+/* common trans ops for all generations transports */
+void iwl_trans_pcie_configure(struct iwl_trans *trans,
+			      const struct iwl_trans_config *trans_cfg);
+int iwl_trans_pcie_start_hw(struct iwl_trans *trans);
+void iwl_trans_pcie_op_mode_leave(struct iwl_trans *trans);
+void iwl_trans_pcie_write8(struct iwl_trans *trans, u32 ofs, u8 val);
+void iwl_trans_pcie_write32(struct iwl_trans *trans, u32 ofs, u32 val);
+u32 iwl_trans_pcie_read32(struct iwl_trans *trans, u32 ofs);
+u32 iwl_trans_pcie_read_prph(struct iwl_trans *trans, u32 reg);
+void iwl_trans_pcie_write_prph(struct iwl_trans *trans, u32 addr, u32 val);
+int iwl_trans_pcie_read_mem(struct iwl_trans *trans, u32 addr,
+			    void *buf, int dwords);
+int iwl_trans_pcie_write_mem(struct iwl_trans *trans, u32 addr,
+			     const void *buf, int dwords);
+int iwl_trans_pcie_sw_reset(struct iwl_trans *trans, bool retake_ownership);
+struct iwl_trans_dump_data *
+iwl_trans_pcie_dump_data(struct iwl_trans *trans, u32 dump_mask,
+			 const struct iwl_dump_sanitize_ops *sanitize_ops,
+			 void *sanitize_ctx);
+int iwl_trans_pcie_d3_resume(struct iwl_trans *trans,
+			     enum iwl_d3_status *status,
+			     bool test,  bool reset);
+int iwl_trans_pcie_d3_suspend(struct iwl_trans *trans, bool test, bool reset);
+void iwl_trans_pci_interrupts(struct iwl_trans *trans, bool enable);
+void iwl_trans_pcie_sync_nmi(struct iwl_trans *trans);
+void iwl_trans_pcie_set_bits_mask(struct iwl_trans *trans, u32 reg,
+				  u32 mask, u32 value);
+int iwl_trans_pcie_read_config32(struct iwl_trans *trans, u32 ofs,
+				 u32 *val);
+bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans);
+void iwl_trans_pcie_release_nic_access(struct iwl_trans *trans);
+
+/* transport gen 1 exported functions */
+void iwl_trans_pcie_fw_alive(struct iwl_trans *trans, u32 scd_addr);
+int iwl_trans_pcie_start_fw(struct iwl_trans *trans,
+			    const struct fw_img *fw, bool run_in_rfkill);
+void iwl_trans_pcie_stop_device(struct iwl_trans *trans);
+
 /* common functions that are used by gen2 transport */
 int iwl_pcie_gen2_apm_init(struct iwl_trans *trans);
 void iwl_pcie_apm_config(struct iwl_trans *trans);
@@ -1069,5 +1110,7 @@ void iwl_trans_pcie_copy_imr_fh(struct iwl_trans *trans,
 				u32 dst_addr, u64 src_addr, u32 byte_cnt);
 int iwl_trans_pcie_copy_imr(struct iwl_trans *trans,
 			    u32 dst_addr, u64 src_addr, u32 byte_cnt);
+int iwl_trans_pcie_rxq_dma_data(struct iwl_trans *trans, int queue,
+				struct iwl_trans_rxq_dma_data *data);
 
 #endif /* __iwl_trans_int_pcie_h__ */
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index 48ae38f3120d..94502e004719 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -127,8 +127,7 @@ out:
 	kfree(buf);
 }
 
-static int iwl_trans_pcie_sw_reset(struct iwl_trans *trans,
-				   bool retake_ownership)
+int iwl_trans_pcie_sw_reset(struct iwl_trans *trans, bool retake_ownership)
 {
 	/* Reset entire device - do controller reset (results in SHRD_HW_RST) */
 	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ) {
@@ -1336,8 +1335,8 @@ void iwl_pcie_synchronize_irqs(struct iwl_trans *trans)
 	}
 }
 
-static int iwl_trans_pcie_start_fw(struct iwl_trans *trans,
-				   const struct fw_img *fw, bool run_in_rfkill)
+int iwl_trans_pcie_start_fw(struct iwl_trans *trans,
+			    const struct fw_img *fw, bool run_in_rfkill)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	bool hw_rfkill;
@@ -1423,7 +1422,7 @@ out:
 	return ret;
 }
 
-static void iwl_trans_pcie_fw_alive(struct iwl_trans *trans, u32 scd_addr)
+void iwl_trans_pcie_fw_alive(struct iwl_trans *trans, u32 scd_addr)
 {
 	iwl_pcie_reset_ict(trans);
 	iwl_pcie_tx_start(trans, scd_addr);
@@ -1458,7 +1457,7 @@ void iwl_trans_pcie_handle_stop_rfkill(struct iwl_trans *trans,
 		iwl_trans_pcie_rf_kill(trans, hw_rfkill, false);
 }
 
-static void iwl_trans_pcie_stop_device(struct iwl_trans *trans)
+void iwl_trans_pcie_stop_device(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	bool was_in_rfkill;
@@ -1552,8 +1551,7 @@ static int iwl_pcie_d3_handshake(struct iwl_trans *trans, bool suspend)
 	return 0;
 }
 
-static int iwl_trans_pcie_d3_suspend(struct iwl_trans *trans, bool test,
-				     bool reset)
+int iwl_trans_pcie_d3_suspend(struct iwl_trans *trans, bool test, bool reset)
 {
 	int ret;
 
@@ -1571,9 +1569,9 @@ static int iwl_trans_pcie_d3_suspend(struct iwl_trans *trans, bool test,
 	return 0;
 }
 
-static int iwl_trans_pcie_d3_resume(struct iwl_trans *trans,
-				    enum iwl_d3_status *status,
-				    bool test,  bool reset)
+int iwl_trans_pcie_d3_resume(struct iwl_trans *trans,
+			     enum iwl_d3_status *status,
+			     bool test,  bool reset)
 {
 	struct iwl_trans_pcie *trans_pcie =  IWL_TRANS_GET_PCIE_TRANS(trans);
 	u32 val;
@@ -1874,7 +1872,7 @@ static int _iwl_trans_pcie_start_hw(struct iwl_trans *trans)
 	return 0;
 }
 
-static int iwl_trans_pcie_start_hw(struct iwl_trans *trans)
+int iwl_trans_pcie_start_hw(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int ret;
@@ -1886,7 +1884,7 @@ static int iwl_trans_pcie_start_hw(struct iwl_trans *trans)
 	return ret;
 }
 
-static void iwl_trans_pcie_op_mode_leave(struct iwl_trans *trans)
+void iwl_trans_pcie_op_mode_leave(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
@@ -1906,17 +1904,17 @@ static void iwl_trans_pcie_op_mode_leave(struct iwl_trans *trans)
 	iwl_pcie_synchronize_irqs(trans);
 }
 
-static void iwl_trans_pcie_write8(struct iwl_trans *trans, u32 ofs, u8 val)
+void iwl_trans_pcie_write8(struct iwl_trans *trans, u32 ofs, u8 val)
 {
 	writeb(val, IWL_TRANS_GET_PCIE_TRANS(trans)->hw_base + ofs);
 }
 
-static void iwl_trans_pcie_write32(struct iwl_trans *trans, u32 ofs, u32 val)
+void iwl_trans_pcie_write32(struct iwl_trans *trans, u32 ofs, u32 val)
 {
 	writel(val, IWL_TRANS_GET_PCIE_TRANS(trans)->hw_base + ofs);
 }
 
-static u32 iwl_trans_pcie_read32(struct iwl_trans *trans, u32 ofs)
+u32 iwl_trans_pcie_read32(struct iwl_trans *trans, u32 ofs)
 {
 	return readl(IWL_TRANS_GET_PCIE_TRANS(trans)->hw_base + ofs);
 }
@@ -1929,7 +1927,7 @@ static u32 iwl_trans_pcie_prph_msk(struct iwl_trans *trans)
 		return 0x000FFFFF;
 }
 
-static u32 iwl_trans_pcie_read_prph(struct iwl_trans *trans, u32 reg)
+u32 iwl_trans_pcie_read_prph(struct iwl_trans *trans, u32 reg)
 {
 	u32 mask = iwl_trans_pcie_prph_msk(trans);
 
@@ -1938,8 +1936,7 @@ static u32 iwl_trans_pcie_read_prph(struct iwl_trans *trans, u32 reg)
 	return iwl_trans_pcie_read32(trans, HBUS_TARG_PRPH_RDAT);
 }
 
-static void iwl_trans_pcie_write_prph(struct iwl_trans *trans, u32 addr,
-				      u32 val)
+void iwl_trans_pcie_write_prph(struct iwl_trans *trans, u32 addr, u32 val)
 {
 	u32 mask = iwl_trans_pcie_prph_msk(trans);
 
@@ -1948,8 +1945,8 @@ static void iwl_trans_pcie_write_prph(struct iwl_trans *trans, u32 addr,
 	iwl_trans_pcie_write32(trans, HBUS_TARG_PRPH_WDAT, val);
 }
 
-static void iwl_trans_pcie_configure(struct iwl_trans *trans,
-				     const struct iwl_trans_config *trans_cfg)
+void iwl_trans_pcie_configure(struct iwl_trans *trans,
+			      const struct iwl_trans_config *trans_cfg)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
@@ -2095,14 +2092,6 @@ void iwl_trans_pcie_free(struct iwl_trans *trans)
 	iwl_trans_free(trans);
 }
 
-static void iwl_trans_pcie_set_pmi(struct iwl_trans *trans, bool state)
-{
-	if (state)
-		set_bit(STATUS_TPOWER_PMI, &trans->status);
-	else
-		clear_bit(STATUS_TPOWER_PMI, &trans->status);
-}
-
 struct iwl_trans_pcie_removal {
 	struct pci_dev *pdev;
 	struct work_struct work;
@@ -2266,7 +2255,7 @@ out:
 	return true;
 }
 
-static bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
+bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
 {
 	bool ret;
 
@@ -2280,7 +2269,7 @@ static bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
 	return false;
 }
 
-static void iwl_trans_pcie_release_nic_access(struct iwl_trans *trans)
+void iwl_trans_pcie_release_nic_access(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
@@ -2310,8 +2299,8 @@ out:
 	spin_unlock_bh(&trans_pcie->reg_lock);
 }
 
-static int iwl_trans_pcie_read_mem(struct iwl_trans *trans, u32 addr,
-				   void *buf, int dwords)
+int iwl_trans_pcie_read_mem(struct iwl_trans *trans, u32 addr,
+			    void *buf, int dwords)
 {
 #define IWL_MAX_HW_ERRS 5
 	unsigned int num_consec_hw_errors = 0;
@@ -2360,8 +2349,8 @@ static int iwl_trans_pcie_read_mem(struct iwl_trans *trans, u32 addr,
 	return 0;
 }
 
-static int iwl_trans_pcie_write_mem(struct iwl_trans *trans, u32 addr,
-				    const void *buf, int dwords)
+int iwl_trans_pcie_write_mem(struct iwl_trans *trans, u32 addr,
+			     const void *buf, int dwords)
 {
 	int offs, ret = 0;
 	const u32 *vals = buf;
@@ -2378,8 +2367,8 @@ static int iwl_trans_pcie_write_mem(struct iwl_trans *trans, u32 addr,
 	return ret;
 }
 
-static int iwl_trans_pcie_read_config32(struct iwl_trans *trans, u32 ofs,
-					u32 *val)
+int iwl_trans_pcie_read_config32(struct iwl_trans *trans, u32 ofs,
+				 u32 *val)
 {
 	return pci_read_config_dword(IWL_TRANS_GET_PCIE_TRANS(trans)->pci_dev,
 				     ofs, val);
@@ -2387,8 +2376,8 @@ static int iwl_trans_pcie_read_config32(struct iwl_trans *trans, u32 ofs,
 
 #define IWL_FLUSH_WAIT_MS	2000
 
-static int iwl_trans_pcie_rxq_dma_data(struct iwl_trans *trans, int queue,
-				       struct iwl_trans_rxq_dma_data *data)
+int iwl_trans_pcie_rxq_dma_data(struct iwl_trans *trans, int queue,
+				struct iwl_trans_rxq_dma_data *data)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
@@ -2403,7 +2392,7 @@ static int iwl_trans_pcie_rxq_dma_data(struct iwl_trans *trans, int queue,
 	return 0;
 }
 
-static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans, int txq_idx)
+int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans, int txq_idx)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_txq *txq;
@@ -2465,7 +2454,7 @@ static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans, int txq_idx)
 	return 0;
 }
 
-static int iwl_trans_pcie_wait_txqs_empty(struct iwl_trans *trans, u32 txq_bm)
+int iwl_trans_pcie_wait_txqs_empty(struct iwl_trans *trans, u32 txq_bm)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	int cnt;
@@ -2491,8 +2480,8 @@ static int iwl_trans_pcie_wait_txqs_empty(struct iwl_trans *trans, u32 txq_bm)
 	return ret;
 }
 
-static void iwl_trans_pcie_set_bits_mask(struct iwl_trans *trans, u32 reg,
-					 u32 mask, u32 value)
+void iwl_trans_pcie_set_bits_mask(struct iwl_trans *trans, u32 reg,
+				  u32 mask, u32 value)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
@@ -3071,7 +3060,7 @@ void iwl_trans_pcie_dbgfs_register(struct iwl_trans *trans)
 	DEBUGFS_ADD_FILE(rf, dir, 0400);
 }
 
-static void iwl_trans_pcie_debugfs_cleanup(struct iwl_trans *trans)
+void iwl_trans_pcie_debugfs_cleanup(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct cont_rec *data = &trans_pcie->fw_mon_data;
@@ -3349,9 +3338,8 @@ static int iwl_trans_get_fw_monitor_len(struct iwl_trans *trans, u32 *len)
 	return 0;
 }
 
-static struct iwl_trans_dump_data *
-iwl_trans_pcie_dump_data(struct iwl_trans *trans,
-			 u32 dump_mask,
+struct iwl_trans_dump_data *
+iwl_trans_pcie_dump_data(struct iwl_trans *trans, u32 dump_mask,
 			 const struct iwl_dump_sanitize_ops *sanitize_ops,
 			 void *sanitize_ctx)
 {
@@ -3500,7 +3488,7 @@ iwl_trans_pcie_dump_data(struct iwl_trans *trans,
 	return dump_data;
 }
 
-static void iwl_trans_pci_interrupts(struct iwl_trans *trans, bool enable)
+void iwl_trans_pci_interrupts(struct iwl_trans *trans, bool enable)
 {
 	if (enable)
 		iwl_enable_interrupts(trans);
@@ -3508,7 +3496,7 @@ static void iwl_trans_pci_interrupts(struct iwl_trans *trans, bool enable)
 		iwl_disable_interrupts(trans);
 }
 
-static void iwl_trans_pcie_sync_nmi(struct iwl_trans *trans)
+void iwl_trans_pcie_sync_nmi(struct iwl_trans *trans)
 {
 	u32 inta_addr, sw_err_bit;
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
@@ -3527,81 +3515,6 @@ static void iwl_trans_pcie_sync_nmi(struct iwl_trans *trans)
 	iwl_trans_sync_nmi_with_addr(trans, inta_addr, sw_err_bit);
 }
 
-#define IWL_TRANS_COMMON_OPS						\
-	.op_mode_leave = iwl_trans_pcie_op_mode_leave,			\
-	.write8 = iwl_trans_pcie_write8,				\
-	.write32 = iwl_trans_pcie_write32,				\
-	.read32 = iwl_trans_pcie_read32,				\
-	.read_prph = iwl_trans_pcie_read_prph,				\
-	.write_prph = iwl_trans_pcie_write_prph,			\
-	.read_mem = iwl_trans_pcie_read_mem,				\
-	.write_mem = iwl_trans_pcie_write_mem,				\
-	.read_config32 = iwl_trans_pcie_read_config32,			\
-	.configure = iwl_trans_pcie_configure,				\
-	.set_pmi = iwl_trans_pcie_set_pmi,				\
-	.sw_reset = iwl_trans_pcie_sw_reset,				\
-	.grab_nic_access = iwl_trans_pcie_grab_nic_access,		\
-	.release_nic_access = iwl_trans_pcie_release_nic_access,	\
-	.set_bits_mask = iwl_trans_pcie_set_bits_mask,			\
-	.dump_data = iwl_trans_pcie_dump_data,				\
-	.d3_suspend = iwl_trans_pcie_d3_suspend,			\
-	.d3_resume = iwl_trans_pcie_d3_resume,				\
-	.interrupts = iwl_trans_pci_interrupts,				\
-	.sync_nmi = iwl_trans_pcie_sync_nmi,				\
-	.imr_dma_data = iwl_trans_pcie_copy_imr				\
-
-static const struct iwl_trans_ops trans_ops_pcie = {
-	IWL_TRANS_COMMON_OPS,
-	.start_hw = iwl_trans_pcie_start_hw,
-	.fw_alive = iwl_trans_pcie_fw_alive,
-	.start_fw = iwl_trans_pcie_start_fw,
-	.stop_device = iwl_trans_pcie_stop_device,
-
-	.send_cmd = iwl_pcie_enqueue_hcmd,
-
-	.tx = iwl_trans_pcie_tx,
-	.reclaim = iwl_pcie_reclaim,
-
-	.txq_disable = iwl_trans_pcie_txq_disable,
-	.txq_enable = iwl_trans_pcie_txq_enable,
-
-	.txq_set_shared_mode = iwl_trans_pcie_txq_set_shared_mode,
-
-	.wait_tx_queues_empty = iwl_trans_pcie_wait_txqs_empty,
-
-	.freeze_txq_timer = iwl_pcie_freeze_txq_timer,
-#ifdef CONFIG_IWLWIFI_DEBUGFS
-	.debugfs_cleanup = iwl_trans_pcie_debugfs_cleanup,
-#endif
-};
-
-static const struct iwl_trans_ops trans_ops_pcie_gen2 = {
-	IWL_TRANS_COMMON_OPS,
-	.start_hw = iwl_trans_pcie_start_hw,
-	.fw_alive = iwl_trans_pcie_gen2_fw_alive,
-	.start_fw = iwl_trans_pcie_gen2_start_fw,
-	.stop_device = iwl_trans_pcie_gen2_stop_device,
-
-	.send_cmd = iwl_pcie_gen2_enqueue_hcmd,
-
-	.tx = iwl_txq_gen2_tx,
-	.reclaim = iwl_pcie_reclaim,
-
-	.set_q_ptrs = iwl_pcie_set_q_ptrs,
-
-	.txq_alloc = iwl_txq_dyn_alloc,
-	.txq_free = iwl_txq_dyn_free,
-	.wait_txq_empty = iwl_trans_pcie_wait_txq_empty,
-	.rxq_dma_data = iwl_trans_pcie_rxq_dma_data,
-	.load_pnvm = iwl_trans_pcie_ctx_info_gen3_load_pnvm,
-	.set_pnvm = iwl_trans_pcie_ctx_info_gen3_set_pnvm,
-	.load_reduce_power = iwl_trans_pcie_ctx_info_gen3_load_reduce_power,
-	.set_reduce_power = iwl_trans_pcie_ctx_info_gen3_set_reduce_power,
-#ifdef CONFIG_IWLWIFI_DEBUGFS
-	.debugfs_cleanup = iwl_trans_pcie_debugfs_cleanup,
-#endif
-};
-
 struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 			       const struct pci_device_id *ent,
 			       const struct iwl_cfg_trans_params *cfg_trans)
@@ -3609,13 +3522,9 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 	struct iwl_trans_pcie *trans_pcie, **priv;
 	struct iwl_trans *trans;
 	int ret, addr_size;
-	const struct iwl_trans_ops *ops = &trans_ops_pcie_gen2;
 	void __iomem * const *table;
 	u32 bar0;
 
-	if (!cfg_trans->gen2)
-		ops = &trans_ops_pcie;
-
 	/* reassign our BAR 0 if invalid due to possible runtime PM races */
 	pci_read_config_dword(pdev, PCI_BASE_ADDRESS_0, &bar0);
 	if (bar0 == PCI_BASE_ADDRESS_MEM_TYPE_64) {
@@ -3628,7 +3537,7 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev,
 	if (ret)
 		return ERR_PTR(ret);
 
-	trans = iwl_trans_alloc(sizeof(struct iwl_trans_pcie), &pdev->dev, ops,
+	trans = iwl_trans_alloc(sizeof(struct iwl_trans_pcie), &pdev->dev,
 				cfg_trans);
 	if (!trans)
 		return ERR_PTR(-ENOMEM);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index fd47c4400b4e..f4452732417d 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -2385,7 +2385,11 @@ static int iwl_trans_pcie_send_hcmd_sync(struct iwl_trans *trans,
 
 	IWL_DEBUG_INFO(trans, "Setting HCMD_ACTIVE for command %s\n", cmd_str);
 
-	cmd_idx = trans->ops->send_cmd(trans, cmd);
+	if (trans->trans_cfg->gen2)
+		cmd_idx = iwl_pcie_gen2_enqueue_hcmd(trans, cmd);
+	else
+		cmd_idx = iwl_pcie_enqueue_hcmd(trans, cmd);
+
 	if (cmd_idx < 0) {
 		ret = cmd_idx;
 		clear_bit(STATUS_SYNC_HCMD_ACTIVE, &trans->status);
@@ -2485,7 +2489,11 @@ int iwl_trans_pcie_send_hcmd(struct iwl_trans *trans,
 		if (WARN_ON(cmd->flags & CMD_WANT_SKB))
 			return -EINVAL;
 
-		ret = trans->ops->send_cmd(trans, cmd);
+		if (trans->trans_cfg->gen2)
+			ret = iwl_pcie_gen2_enqueue_hcmd(trans, cmd);
+		else
+			ret = iwl_pcie_enqueue_hcmd(trans, cmd);
+
 		if (ret < 0) {
 			IWL_ERR(trans,
 				"Error sending %s: enqueue_hcmd failed: %d\n",
-- 
cgit v1.2.3-58-ga151


From bd40215b19d255b433a91a8bb8088937e5db4284 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Wed, 5 Jun 2024 14:05:06 +0300
Subject: wifi: iwlwifi: mvm: fix re-enabling EMLSR

When EMLSR gets unblocked, the current code checks if the last exit was
due to an EXIT reason (as opposed to a BLOCKING one), and if so, it
does nothing, as in this case a MLO scan was scheduled to run in 30
seconds.

But the code doesn't consider the time that passed from the last exit,
so if immediately after the exit a blocker occurred (e.g. non-BSS
interface), and lasts for more than 30 seconds, then the MLO scan and the
following link selection will decide not to enter EMLSR, and when the
unblocking event finally happens, the reason is still set to the EXIT one,
so it will do nothing, and we will not have the chance to re-enable EMLSR.

Fix this by checking also the time that has passed since the last exit,
only if it is less than 30 seconds, we can count on the scheduled MLO
scan.

Note that clearing the reason itself can't be done since it is needed
for the EMLSR prevention mechanism.

Fixes: 2f33561ea8f9 ("wifi: iwlwifi: mvm: trigger link selection after exiting EMLSR")
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240605140327.58556fc4cfa9.I4c55b3cd9f20b21b37f28258d0fb6842ba413966@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/link.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 02a475a1f1b8..73527781f89a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -1086,8 +1086,10 @@ static void iwl_mvm_esr_unblocked(struct iwl_mvm *mvm,
 
 	IWL_DEBUG_INFO(mvm, "EMLSR is unblocked\n");
 
-	/* We exited due to an EXIT reason, so MLO scan was scheduled already */
-	if (mvmvif->last_esr_exit.reason &&
+	/* If we exited due to an EXIT reason, and the exit was in less than
+	 * 30 seconds, then a MLO scan was scheduled already.
+	 */
+	if (!need_new_sel &&
 	    !(mvmvif->last_esr_exit.reason & IWL_MVM_BLOCK_ESR_REASONS)) {
 		IWL_DEBUG_INFO(mvm, "Wait for MLO scan\n");
 		return;
-- 
cgit v1.2.3-58-ga151


From bec2cdf548f92641f68677fcc43dc19aa4d0eceb Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Wed, 5 Jun 2024 14:05:07 +0300
Subject: wifi: iwlwifi: bump min API version for Qu/So devices

Stop supporting all FWs older than the max API version.
These FW versions were supported since v6.5.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://msgid.link/20240605140327.ad6d43fe9893.I96f769e7d5be3e6499d260451df781bd694a5142@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/22000.c | 4 ++--
 drivers/net/wireless/intel/iwlwifi/cfg/ax210.c | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
index d594694206b3..2e2fcb3807ef 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/22000.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
  * Copyright (C) 2015-2017 Intel Deutschland GmbH
- * Copyright (C) 2018-2023 Intel Corporation
+ * Copyright (C) 2018-2024 Intel Corporation
  */
 #include <linux/module.h>
 #include <linux/stringify.h>
@@ -13,7 +13,7 @@
 #define IWL_22000_UCODE_API_MAX	77
 
 /* Lowest firmware API version supported */
-#define IWL_22000_UCODE_API_MIN	50
+#define IWL_22000_UCODE_API_MIN	77
 
 /* NVM versions */
 #define IWL_22000_NVM_VERSION		0x0a1d
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/ax210.c b/drivers/net/wireless/intel/iwlwifi/cfg/ax210.c
index 25952d0bea99..975e8aed1526 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/ax210.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/ax210.c
@@ -13,7 +13,7 @@
 #define IWL_AX210_UCODE_API_MAX	89
 
 /* Lowest firmware API version supported */
-#define IWL_AX210_UCODE_API_MIN	59
+#define IWL_AX210_UCODE_API_MIN	77
 
 /* NVM versions */
 #define IWL_AX210_NVM_VERSION		0x0a1d
-- 
cgit v1.2.3-58-ga151


From 5f1fee9644759d07f3d3a468389c7c18027083d7 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 5 Jun 2024 14:07:36 +0300
Subject: wifi: iwlwifi: mvm: separate non-BSS/ROC EMLSR blocking

If non-BSS and remain-on-channel (ROC) blocking were to occur
simultaneously, they'd step on each other's toes, unblocking
when not yet supported. Disentangle these bits, and ROC doesn't
need to use the non_bss_link() function then.

Fixes: a1efeb823084 ("wifi: iwlwifi: mvm: Block EMLSR when a p2p/softAP vif is active")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605140556.461fcf7b95bb.Id0d21dcb739d426ff15ec068b5df8abaab58884d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c   | 10 +++++++---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h        |  5 ++++-
 drivers/net/wireless/intel/iwlwifi/mvm/time-event.c |  5 +++--
 3 files changed, 14 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index b004de6caec3..43d11db2444a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -4824,6 +4824,7 @@ int iwl_mvm_roc_common(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		       const struct iwl_mvm_roc_ops *ops)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
+	struct ieee80211_vif *bss_vif = iwl_mvm_get_bss_vif(mvm);
 	u32 lmac_id;
 	int ret;
 
@@ -4836,9 +4837,12 @@ int iwl_mvm_roc_common(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	 */
 	flush_work(&mvm->roc_done_wk);
 
-	ret = iwl_mvm_esr_non_bss_link(mvm, vif, 0, true);
-	if (ret)
-		return ret;
+	if (!IS_ERR_OR_NULL(bss_vif)) {
+		ret = iwl_mvm_block_esr_sync(mvm, bss_vif,
+					     IWL_MVM_ESR_BLOCKED_ROC);
+		if (ret)
+			return ret;
+	}
 
 	guard(mvm)(mvm);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 9b939225990a..8038764a4573 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -361,7 +361,9 @@ struct iwl_mvm_vif_link_info {
  * @IWL_MVM_ESR_BLOCKED_WOWLAN: WOWLAN is preventing the enablement of EMLSR
  * @IWL_MVM_ESR_BLOCKED_TPT: block EMLSR when there is not enough traffic
  * @IWL_MVM_ESR_BLOCKED_FW: FW didn't recommended/forced exit from EMLSR
- * @IWL_MVM_ESR_BLOCKED_NON_BSS: An active non-bssid link's preventing EMLSR
+ * @IWL_MVM_ESR_BLOCKED_NON_BSS: An active non-BSS interface's link is
+ *	preventing EMLSR
+ * @IWL_MVM_ESR_BLOCKED_ROC: remain-on-channel is preventing EMLSR
  * @IWL_MVM_ESR_EXIT_MISSED_BEACON: exited EMLSR due to missed beacons
  * @IWL_MVM_ESR_EXIT_LOW_RSSI: link is deactivated/not allowed for EMLSR
  *	due to low RSSI.
@@ -378,6 +380,7 @@ enum iwl_mvm_esr_state {
 	IWL_MVM_ESR_BLOCKED_TPT		= 0x4,
 	IWL_MVM_ESR_BLOCKED_FW		= 0x8,
 	IWL_MVM_ESR_BLOCKED_NON_BSS	= 0x10,
+	IWL_MVM_ESR_BLOCKED_ROC		= 0x20,
 	IWL_MVM_ESR_EXIT_MISSED_BEACON	= 0x10000,
 	IWL_MVM_ESR_EXIT_LOW_RSSI	= 0x20000,
 	IWL_MVM_ESR_EXIT_COEX		= 0x40000,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
index 8ee4498f4245..89f9af3393a1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
@@ -47,6 +47,7 @@ void iwl_mvm_te_clear_data(struct iwl_mvm *mvm,
 
 static void iwl_mvm_cleanup_roc(struct iwl_mvm *mvm)
 {
+	struct ieee80211_vif *bss_vif = iwl_mvm_get_bss_vif(mvm);
 	struct ieee80211_vif *vif = mvm->p2p_device_vif;
 
 	lockdep_assert_held(&mvm->mutex);
@@ -119,9 +120,9 @@ static void iwl_mvm_cleanup_roc(struct iwl_mvm *mvm)
 			iwl_mvm_rm_aux_sta(mvm);
 	}
 
+	if (vif && !IS_ERR_OR_NULL(bss_vif))
+		iwl_mvm_unblock_esr(mvm, bss_vif, IWL_MVM_ESR_BLOCKED_ROC);
 	mutex_unlock(&mvm->mutex);
-	if (vif)
-		iwl_mvm_esr_non_bss_link(mvm, vif, 0, false);
 }
 
 void iwl_mvm_roc_done_wk(struct work_struct *wk)
-- 
cgit v1.2.3-58-ga151


From 54d96912caf6bb1e6682cfdb225a67cb80545557 Mon Sep 17 00:00:00 2001
From: Shaul Triebitz <shaul.triebitz@intel.com>
Date: Wed, 5 Jun 2024 14:07:39 +0300
Subject: wifi: iwlwifi: mvm: use ROC for P2P device activities

This is the first step towards removing the P2P Device MAC.
Use ROC (which uses the AUX MAC) for P2P Device
discoveribility and action frames.

Signed-off-by: Shaul Triebitz <shaul.triebitz@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://msgid.link/20240605140556.8c90e457abbd.I8e340759ecb299e05b1809f3d8060429c6cbbd01@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/time-event.h |   4 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c  |   1 +
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |  39 ++++-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  17 +-
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c       |   2 +-
 .../net/wireless/intel/iwlwifi/mvm/time-event.c    | 177 ++++++++++++++-------
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c        |  38 +++--
 7 files changed, 196 insertions(+), 82 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/time-event.h b/drivers/net/wireless/intel/iwlwifi/fw/api/time-event.h
index 2e15be71c957..f4b827b58bd3 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/time-event.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/time-event.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2012-2014, 2018-2020, 2022-2023 Intel Corporation
+ * Copyright (C) 2012-2014, 2018-2020, 2022-2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -340,11 +340,13 @@ struct iwl_hs20_roc_res {
  * @ROC_ACTIVITY_HOTSPOT: ROC for hs20 activity
  * @ROC_ACTIVITY_P2P_DISC: ROC for p2p discoverability activity
  * @ROC_ACTIVITY_P2P_TXRX: ROC for p2p action frames activity
+ * @ROC_ACTIVITY_P2P_NEG: ROC for p2p negotiation (used also for TX)
  */
 enum iwl_roc_activity {
 	ROC_ACTIVITY_HOTSPOT,
 	ROC_ACTIVITY_P2P_DISC,
 	ROC_ACTIVITY_P2P_TXRX,
+	ROC_ACTIVITY_P2P_NEG,
 	ROC_NUM_ACTIVITIES
 }; /* ROC_ACTIVITY_API_E_VER_1 */
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
index ef7ffeaf6ff2..dfcc96f18b4f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c
@@ -296,6 +296,7 @@ int iwl_mvm_mac_ctxt_init(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 
 	INIT_LIST_HEAD(&mvmvif->time_event_data.list);
 	mvmvif->time_event_data.id = TE_MAX;
+	mvmvif->roc_activity = ROC_NUM_ACTIVITIES;
 
 	mvmvif->deflink.bcast_sta.sta_id = IWL_MVM_INVALID_STA;
 	mvmvif->deflink.mcast_sta.sta_id = IWL_MVM_INVALID_STA;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 43d11db2444a..81103a974b20 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -821,7 +821,7 @@ void iwl_mvm_mac_tx(struct ieee80211_hw *hw,
 	}
 
 	if (offchannel &&
-	    !test_bit(IWL_MVM_STATUS_ROC_RUNNING, &mvm->status) &&
+	    !test_bit(IWL_MVM_STATUS_ROC_P2P_RUNNING, &mvm->status) &&
 	    !test_bit(IWL_MVM_STATUS_ROC_AUX_RUNNING, &mvm->status))
 		goto drop;
 
@@ -1102,6 +1102,8 @@ static void iwl_mvm_cleanup_iterator(void *data, u8 *mac,
 	iwl_mvm_te_clear_data(mvm, &mvmvif->time_event_data);
 	spin_unlock_bh(&mvm->time_event_lock);
 
+	mvmvif->roc_activity = ROC_NUM_ACTIVITIES;
+
 	mvmvif->bf_enabled = false;
 	mvmvif->ba_enabled = false;
 	mvmvif->ap_sta = NULL;
@@ -4771,6 +4773,37 @@ static int iwl_mvm_roc_station(struct iwl_mvm *mvm,
 	return ret;
 }
 
+static int iwl_mvm_roc_p2p(struct iwl_mvm *mvm,
+			   struct ieee80211_channel *channel,
+			   struct ieee80211_vif *vif,
+			   int duration,
+			   enum ieee80211_roc_type type)
+{
+	enum iwl_roc_activity activity;
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	switch (type) {
+	case IEEE80211_ROC_TYPE_NORMAL:
+		activity = ROC_ACTIVITY_P2P_DISC;
+		break;
+	case IEEE80211_ROC_TYPE_MGMT_TX:
+		activity = ROC_ACTIVITY_P2P_NEG;
+		break;
+	default:
+		WARN_ONCE(1, "Got an invalid P2P ROC type\n");
+		return -EINVAL;
+	}
+
+	ret = iwl_mvm_mld_add_aux_sta(mvm,
+				      iwl_mvm_get_lmac_id(mvm, channel->band));
+	if (ret)
+		return ret;
+
+	return iwl_mvm_roc_add_cmd(mvm, channel, vif, duration, activity);
+}
+
 static int iwl_mvm_p2p_find_phy_ctxt(struct iwl_mvm *mvm,
 				     struct ieee80211_vif *vif,
 				     struct ieee80211_channel *channel)
@@ -4863,6 +4896,10 @@ int iwl_mvm_roc_common(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		return -EINVAL;
 	}
 
+	if (iwl_mvm_has_p2p_over_aux(mvm)) {
+		ret = iwl_mvm_roc_p2p(mvm, channel, vif, duration, type);
+		return ret;
+	}
 
 	ret = iwl_mvm_p2p_find_phy_ctxt(mvm, vif, channel);
 	if (ret)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 8038764a4573..bce517effe55 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -453,6 +453,8 @@ struct iwl_mvm_esr_exit {
  * @prevent_esr_done_wk: work that should be done when esr prevention ends.
  * @mlo_int_scan_wk: work for the internal MLO scan.
  * @unblock_esr_tpt_wk: work for unblocking EMLSR when tpt is high enough.
+ * @roc_activity: currently running ROC activity for this vif (or
+ *	ROC_NUM_ACTIVITIES if no activity is running).
  */
 struct iwl_mvm_vif {
 	struct iwl_mvm *mvm;
@@ -530,6 +532,7 @@ struct iwl_mvm_vif {
 
 	struct iwl_mvm_time_event_data time_event_data;
 	struct iwl_mvm_time_event_data hs_time_event_data;
+	enum iwl_roc_activity roc_activity;
 
 	/* TCP Checksum Offload */
 	netdev_features_t features;
@@ -1345,7 +1348,8 @@ DEFINE_GUARD(mvm, struct iwl_mvm *, mutex_lock(&_T->mutex), mutex_unlock(&_T->mu
  * enum iwl_mvm_status - MVM status bits
  * @IWL_MVM_STATUS_HW_RFKILL: HW RF-kill is asserted
  * @IWL_MVM_STATUS_HW_CTKILL: CT-kill is active
- * @IWL_MVM_STATUS_ROC_RUNNING: remain-on-channel is running
+ * @IWL_MVM_STATUS_ROC_P2P_RUNNING: remain-on-channel on P2P is running (when
+ *	P2P is not over AUX)
  * @IWL_MVM_STATUS_HW_RESTART_REQUESTED: HW restart was requested
  * @IWL_MVM_STATUS_IN_HW_RESTART: HW restart is active
  * @IWL_MVM_STATUS_ROC_AUX_RUNNING: AUX remain-on-channel is running
@@ -1359,7 +1363,7 @@ DEFINE_GUARD(mvm, struct iwl_mvm *, mutex_lock(&_T->mutex), mutex_unlock(&_T->mu
 enum iwl_mvm_status {
 	IWL_MVM_STATUS_HW_RFKILL,
 	IWL_MVM_STATUS_HW_CTKILL,
-	IWL_MVM_STATUS_ROC_RUNNING,
+	IWL_MVM_STATUS_ROC_P2P_RUNNING,
 	IWL_MVM_STATUS_HW_RESTART_REQUESTED,
 	IWL_MVM_STATUS_IN_HW_RESTART,
 	IWL_MVM_STATUS_ROC_AUX_RUNNING,
@@ -2773,6 +2777,13 @@ static inline void iwl_mvm_mei_set_sw_rfkill_state(struct iwl_mvm *mvm)
 					 sw_rfkill);
 }
 
+static inline bool iwl_mvm_has_p2p_over_aux(struct iwl_mvm *mvm)
+{
+	u32 cmd_id = WIDE_ID(MAC_CONF_GROUP, ROC_CMD);
+
+	return iwl_fw_lookup_cmd_ver(mvm->fw, cmd_id, 0) >= 4;
+}
+
 static inline bool iwl_mvm_mei_filter_scan(struct iwl_mvm *mvm,
 					   struct sk_buff *skb)
 {
@@ -2938,7 +2949,7 @@ void iwl_mvm_roc_duration_and_delay(struct ieee80211_vif *vif,
 int iwl_mvm_roc_add_cmd(struct iwl_mvm *mvm,
 			struct ieee80211_channel *channel,
 			struct ieee80211_vif *vif,
-			int duration, u32 activity);
+			int duration, enum iwl_roc_activity activity);
 
 /* EMLSR */
 bool iwl_mvm_vif_has_esr_cap(struct iwl_mvm *mvm, struct ieee80211_vif *vif);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 85178fe1b7ea..f8a4fefede8d 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -471,7 +471,7 @@ static const struct iwl_rx_handlers iwl_mvm_rx_handlers[] = {
 		       iwl_mvm_time_sync_msmt_confirm_event, RX_HANDLER_SYNC,
 		       struct iwl_time_msmt_cfm_notify),
 	RX_HANDLER_GRP(MAC_CONF_GROUP, ROC_NOTIF,
-		       iwl_mvm_rx_roc_notif, RX_HANDLER_SYNC,
+		       iwl_mvm_rx_roc_notif, RX_HANDLER_ASYNC_LOCKED,
 		       struct iwl_roc_notif),
 	RX_HANDLER_GRP(SCAN_GROUP, CHANNEL_SURVEY_NOTIF,
 		       iwl_mvm_rx_channel_survey_notif, RX_HANDLER_ASYNC_LOCKED,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
index 89f9af3393a1..d73e19a8fdda 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
@@ -53,7 +53,7 @@ static void iwl_mvm_cleanup_roc(struct iwl_mvm *mvm)
 	lockdep_assert_held(&mvm->mutex);
 
 	/*
-	 * Clear the ROC_RUNNING status bit.
+	 * Clear the ROC_P2P_RUNNING status bit.
 	 * This will cause the TX path to drop offchannel transmissions.
 	 * That would also be done by mac80211, but it is racy, in particular
 	 * in the case that the time event actually completed in the firmware.
@@ -63,7 +63,7 @@ static void iwl_mvm_cleanup_roc(struct iwl_mvm *mvm)
 	 * won't get stuck on the queue and be transmitted in the next
 	 * time event.
 	 */
-	if (test_and_clear_bit(IWL_MVM_STATUS_ROC_RUNNING, &mvm->status)) {
+	if (test_and_clear_bit(IWL_MVM_STATUS_ROC_P2P_RUNNING, &mvm->status)) {
 		struct iwl_mvm_vif *mvmvif;
 
 		synchronize_net();
@@ -100,7 +100,14 @@ static void iwl_mvm_cleanup_roc(struct iwl_mvm *mvm)
 		}
 	}
 
-	/* Do the same for AUX ROC */
+	/*
+	 * P2P AUX ROC and HS2.0 ROC do not run simultaneously.
+	 * Clear the ROC_AUX_RUNNING status bit.
+	 * This will cause the TX path to drop offchannel transmissions.
+	 * That would also be done by mac80211, but it is racy, in particular
+	 * in the case that the time event actually completed in the firmware
+	 * (which is handled in iwl_mvm_te_handle_notif).
+	 */
 	if (test_and_clear_bit(IWL_MVM_STATUS_ROC_AUX_RUNNING, &mvm->status)) {
 		synchronize_net();
 
@@ -379,7 +386,7 @@ static void iwl_mvm_te_handle_notif(struct iwl_mvm *mvm,
 		te_data->end_jiffies = TU_TO_EXP_TIME(te_data->duration);
 
 		if (te_data->vif->type == NL80211_IFTYPE_P2P_DEVICE) {
-			set_bit(IWL_MVM_STATUS_ROC_RUNNING, &mvm->status);
+			set_bit(IWL_MVM_STATUS_ROC_P2P_RUNNING, &mvm->status);
 			ieee80211_ready_on_channel(mvm->hw);
 		} else if (te_data->id == TE_CHANNEL_SWITCH_PERIOD) {
 			iwl_mvm_te_handle_notify_csa(mvm, te_data, notif);
@@ -389,14 +396,51 @@ static void iwl_mvm_te_handle_notif(struct iwl_mvm *mvm,
 	}
 }
 
+struct iwl_mvm_rx_roc_iterator_data {
+	u32 activity;
+	bool end_activity;
+	bool found;
+};
+
+static void iwl_mvm_rx_roc_iterator(void *_data, u8 *mac,
+				    struct ieee80211_vif *vif)
+{
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	struct iwl_mvm_rx_roc_iterator_data *data = _data;
+
+	if (mvmvif->roc_activity == data->activity) {
+		data->found = true;
+		if (data->end_activity)
+			mvmvif->roc_activity = ROC_NUM_ACTIVITIES;
+	}
+}
+
 void iwl_mvm_rx_roc_notif(struct iwl_mvm *mvm,
 			  struct iwl_rx_cmd_buffer *rxb)
 {
 	struct iwl_rx_packet *pkt = rxb_addr(rxb);
 	struct iwl_roc_notif *notif = (void *)pkt->data;
+	u32 activity = le32_to_cpu(notif->activity);
+	bool started = le32_to_cpu(notif->success) &&
+		le32_to_cpu(notif->started);
+	struct iwl_mvm_rx_roc_iterator_data data = {
+		.activity = activity,
+		.end_activity = !started,
+	};
+
+	/* Clear vif roc_activity if done (set to ROC_NUM_ACTIVITIES) */
+	ieee80211_iterate_active_interfaces_atomic(mvm->hw,
+						   IEEE80211_IFACE_ITER_NORMAL,
+						   iwl_mvm_rx_roc_iterator,
+						   &data);
+	/*
+	 * It is possible that the ROC was canceled
+	 * but the notification was already fired.
+	 */
+	if (!data.found)
+		return;
 
-	if (le32_to_cpu(notif->success) && le32_to_cpu(notif->started) &&
-	    le32_to_cpu(notif->activity) == ROC_ACTIVITY_HOTSPOT) {
+	if (started) {
 		set_bit(IWL_MVM_STATUS_ROC_AUX_RUNNING, &mvm->status);
 		ieee80211_ready_on_channel(mvm->hw);
 	} else {
@@ -725,6 +769,21 @@ static void iwl_mvm_cancel_session_protection(struct iwl_mvm *mvm,
 			"Couldn't send the SESSION_PROTECTION_CMD: %d\n", ret);
 }
 
+static void iwl_mvm_roc_rm_cmd(struct iwl_mvm *mvm, u32 activity)
+{
+	struct iwl_roc_req roc_cmd = {
+		.action = cpu_to_le32(FW_CTXT_ACTION_REMOVE),
+		.activity = cpu_to_le32(activity),
+	};
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+	ret = iwl_mvm_send_cmd_pdu(mvm, WIDE_ID(MAC_CONF_GROUP, ROC_CMD), 0,
+				   sizeof(roc_cmd), &roc_cmd);
+	if (ret)
+		IWL_ERR(mvm, "Couldn't send the ROC_CMD: %d\n", ret);
+}
+
 static bool __iwl_mvm_remove_time_event(struct iwl_mvm *mvm,
 					struct iwl_mvm_time_event_data *te_data,
 					u32 *uid)
@@ -734,6 +793,9 @@ static bool __iwl_mvm_remove_time_event(struct iwl_mvm *mvm,
 	struct iwl_mvm_vif *mvmvif;
 	enum nl80211_iftype iftype;
 	s8 link_id;
+	bool p2p_aux = iwl_mvm_has_p2p_over_aux(mvm);
+	u8 roc_ver = iwl_fw_lookup_cmd_ver(mvm->fw,
+					   WIDE_ID(MAC_CONF_GROUP, ROC_CMD), 0);
 
 	if (!vif)
 		return false;
@@ -758,14 +820,22 @@ static bool __iwl_mvm_remove_time_event(struct iwl_mvm *mvm,
 	iwl_mvm_te_clear_data(mvm, te_data);
 	spin_unlock_bh(&mvm->time_event_lock);
 
-	/* When session protection is used, the te_data->id field
-	 * is reused to save session protection's configuration.
-	 * For AUX ROC, HOT_SPOT_CMD is used and the te_data->id field is set
-	 * to HOT_SPOT_CMD.
-	 */
-	if (fw_has_capa(&mvm->fw->ucode_capa,
-			IWL_UCODE_TLV_CAPA_SESSION_PROT_CMD) &&
-	    id != HOT_SPOT_CMD) {
+	if ((p2p_aux && iftype == NL80211_IFTYPE_P2P_DEVICE) ||
+	    (roc_ver >= 3 && mvmvif->roc_activity == ROC_ACTIVITY_HOTSPOT)) {
+		if (mvmvif->roc_activity < ROC_NUM_ACTIVITIES) {
+			iwl_mvm_roc_rm_cmd(mvm, mvmvif->roc_activity);
+			mvmvif->roc_activity = ROC_NUM_ACTIVITIES;
+			iwl_mvm_roc_finished(mvm);
+		}
+		return false;
+	} else if (fw_has_capa(&mvm->fw->ucode_capa,
+			       IWL_UCODE_TLV_CAPA_SESSION_PROT_CMD) &&
+		   id != HOT_SPOT_CMD) {
+		/* When session protection is used, the te_data->id field
+		 * is reused to save session protection's configuration.
+		 * For AUX ROC, HOT_SPOT_CMD is used and the te_data->id
+		 * field is set to HOT_SPOT_CMD.
+		 */
 		if (mvmvif && id < SESSION_PROTECT_CONF_MAX_ID) {
 			/* Session protection is still ongoing. Cancel it */
 			iwl_mvm_cancel_session_protection(mvm, vif, id,
@@ -966,7 +1036,7 @@ void iwl_mvm_rx_session_protect_notif(struct iwl_mvm *mvm,
 		if (WARN_ON(mvmvif->time_event_data.id !=
 				le32_to_cpu(notif->conf_id)))
 			goto out_unlock;
-		set_bit(IWL_MVM_STATUS_ROC_RUNNING, &mvm->status);
+		set_bit(IWL_MVM_STATUS_ROC_P2P_RUNNING, &mvm->status);
 		ieee80211_ready_on_channel(mvm->hw); /* Start TE */
 	}
 
@@ -1015,7 +1085,7 @@ void iwl_mvm_roc_duration_and_delay(struct ieee80211_vif *vif,
 int iwl_mvm_roc_add_cmd(struct iwl_mvm *mvm,
 			struct ieee80211_channel *channel,
 			struct ieee80211_vif *vif,
-			int duration, u32 activity)
+			int duration, enum iwl_roc_activity activity)
 {
 	int res;
 	u32 duration_tu, delay;
@@ -1024,9 +1094,13 @@ int iwl_mvm_roc_add_cmd(struct iwl_mvm *mvm,
 		.activity = cpu_to_le32(activity),
 		.sta_id = cpu_to_le32(mvm->aux_sta.sta_id),
 	};
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 
 	lockdep_assert_held(&mvm->mutex);
 
+	if (WARN_ON(mvmvif->roc_activity != ROC_NUM_ACTIVITIES))
+		return -EBUSY;
+
 	/* Set the channel info data */
 	iwl_mvm_set_chan_info(mvm, &roc_req.channel_info,
 			      channel->hw_value,
@@ -1042,14 +1116,16 @@ int iwl_mvm_roc_add_cmd(struct iwl_mvm *mvm,
 		     "\t(requested = %ums, max_delay = %ums)\n",
 		     duration, delay);
 	IWL_DEBUG_TE(mvm,
-		     "Requesting to remain on channel %u for %utu\n",
-		     channel->hw_value, duration_tu);
+		     "Requesting to remain on channel %u for %utu. activity %u\n",
+		     channel->hw_value, duration_tu, activity);
 
 	/* Set the node address */
 	memcpy(roc_req.node_addr, vif->addr, ETH_ALEN);
 
 	res = iwl_mvm_send_cmd_pdu(mvm, WIDE_ID(MAC_CONF_GROUP, ROC_CMD),
 				   0, sizeof(roc_req), &roc_req);
+	if (!res)
+		mvmvif->roc_activity = activity;
 
 	return res;
 }
@@ -1192,61 +1268,39 @@ void iwl_mvm_cleanup_roc_te(struct iwl_mvm *mvm)
 		__iwl_mvm_remove_time_event(mvm, te_data, &uid);
 }
 
-static void iwl_mvm_roc_rm_cmd(struct iwl_mvm *mvm, u32 activity)
-{
-	int ret;
-	struct iwl_roc_req roc_cmd = {
-		.action = cpu_to_le32(FW_CTXT_ACTION_REMOVE),
-		.activity = cpu_to_le32(activity),
-	};
-
-	lockdep_assert_held(&mvm->mutex);
-	ret = iwl_mvm_send_cmd_pdu(mvm,
-				   WIDE_ID(MAC_CONF_GROUP, ROC_CMD),
-				   0, sizeof(roc_cmd), &roc_cmd);
-	WARN_ON(ret);
-}
-
-static void iwl_mvm_roc_station_remove(struct iwl_mvm *mvm,
-				       struct iwl_mvm_vif *mvmvif)
-{
-	u32 cmd_id = WIDE_ID(MAC_CONF_GROUP, ROC_CMD);
-	u8 fw_ver = iwl_fw_lookup_cmd_ver(mvm->fw, cmd_id,
-					  IWL_FW_CMD_VER_UNKNOWN);
-
-	if (fw_ver == IWL_FW_CMD_VER_UNKNOWN)
-		iwl_mvm_remove_aux_roc_te(mvm, mvmvif,
-					  &mvmvif->hs_time_event_data);
-	else if (fw_ver == 3)
-		iwl_mvm_roc_rm_cmd(mvm, ROC_ACTIVITY_HOTSPOT);
-	else
-		IWL_ERR(mvm, "ROC command version %d mismatch!\n", fw_ver);
-}
-
 void iwl_mvm_stop_roc(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
-	struct iwl_mvm_vif *mvmvif;
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm_time_event_data *te_data;
+	bool p2p_aux = iwl_mvm_has_p2p_over_aux(mvm);
+	u8 roc_ver = iwl_fw_lookup_cmd_ver(mvm->fw,
+					   WIDE_ID(MAC_CONF_GROUP, ROC_CMD), 0);
+	int iftype = vif->type;
 
 	mutex_lock(&mvm->mutex);
 
-	if (fw_has_capa(&mvm->fw->ucode_capa,
-			IWL_UCODE_TLV_CAPA_SESSION_PROT_CMD)) {
-		mvmvif = iwl_mvm_vif_from_mac80211(vif);
+	if (p2p_aux || (roc_ver >= 3 && iftype != NL80211_IFTYPE_P2P_DEVICE)) {
+		if (mvmvif->roc_activity < ROC_NUM_ACTIVITIES) {
+			iwl_mvm_roc_rm_cmd(mvm, mvmvif->roc_activity);
+			mvmvif->roc_activity = ROC_NUM_ACTIVITIES;
+		}
+		goto cleanup_roc;
+	} else if (fw_has_capa(&mvm->fw->ucode_capa,
+			       IWL_UCODE_TLV_CAPA_SESSION_PROT_CMD)) {
 		te_data = &mvmvif->time_event_data;
 
-		if (vif->type == NL80211_IFTYPE_P2P_DEVICE) {
+		if (iftype == NL80211_IFTYPE_P2P_DEVICE) {
 			if (te_data->id >= SESSION_PROTECT_CONF_MAX_ID) {
 				IWL_DEBUG_TE(mvm,
 					     "No remain on channel event\n");
 				return;
 			}
-
 			iwl_mvm_cancel_session_protection(mvm, vif,
 							  te_data->id,
 							  te_data->link_id);
 		} else {
-			iwl_mvm_roc_station_remove(mvm, mvmvif);
+			iwl_mvm_remove_aux_roc_te(mvm, mvmvif,
+						  &mvmvif->hs_time_event_data);
 		}
 		goto cleanup_roc;
 	}
@@ -1258,8 +1312,8 @@ void iwl_mvm_stop_roc(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 	}
 
 	mvmvif = iwl_mvm_vif_from_mac80211(te_data->vif);
-
-	if (te_data->vif->type == NL80211_IFTYPE_P2P_DEVICE)
+	iftype = te_data->vif->type;
+	if (iftype == NL80211_IFTYPE_P2P_DEVICE)
 		iwl_mvm_remove_time_event(mvm, mvmvif, te_data);
 	else
 		iwl_mvm_remove_aux_roc_te(mvm, mvmvif, te_data);
@@ -1270,9 +1324,10 @@ cleanup_roc:
 	 * (so the status bit isn't set) set it here so iwl_mvm_cleanup_roc will
 	 * cleanup things properly
 	 */
-	set_bit(vif->type == NL80211_IFTYPE_P2P_DEVICE ?
-		IWL_MVM_STATUS_ROC_RUNNING : IWL_MVM_STATUS_ROC_AUX_RUNNING,
-		&mvm->status);
+	if (p2p_aux || iftype != NL80211_IFTYPE_P2P_DEVICE)
+		set_bit(IWL_MVM_STATUS_ROC_AUX_RUNNING, &mvm->status);
+	else
+		set_bit(IWL_MVM_STATUS_ROC_P2P_RUNNING, &mvm->status);
 
 	/* Mutex is released inside this function */
 	iwl_mvm_cleanup_roc(mvm);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 599550f311de..7ff5ea5e7aca 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -802,10 +802,30 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 	if (info.control.vif) {
 		struct iwl_mvm_vif *mvmvif =
 			iwl_mvm_vif_from_mac80211(info.control.vif);
+		bool p2p_aux = iwl_mvm_has_p2p_over_aux(mvm);
 
-		if (info.control.vif->type == NL80211_IFTYPE_P2P_DEVICE ||
-		    info.control.vif->type == NL80211_IFTYPE_AP ||
-		    info.control.vif->type == NL80211_IFTYPE_ADHOC) {
+		if ((info.control.vif->type == NL80211_IFTYPE_P2P_DEVICE &&
+		     p2p_aux) ||
+		    (info.control.vif->type == NL80211_IFTYPE_STATION &&
+		     offchannel)) {
+			/*
+			 * IWL_MVM_OFFCHANNEL_QUEUE is used for ROC packets
+			 * that can be used in 2 different types of vifs, P2P
+			 * Device and STATION.
+			 * P2P Device uses the offchannel queue.
+			 * STATION (HS2.0) uses the auxiliary context of the FW,
+			 * and hence needs to be sent on the aux queue.
+			 * If P2P_DEV_OVER_AUX is supported (p2p_aux = true)
+			 * also P2P Device uses the aux queue.
+			 */
+			sta_id = mvm->aux_sta.sta_id;
+			queue = mvm->aux_queue;
+			if (WARN_ON(queue == IWL_MVM_INVALID_QUEUE))
+				return -1;
+		} else if (info.control.vif->type ==
+			   NL80211_IFTYPE_P2P_DEVICE ||
+			   info.control.vif->type == NL80211_IFTYPE_AP ||
+			   info.control.vif->type == NL80211_IFTYPE_ADHOC) {
 			u32 link_id = u32_get_bits(info.control.flags,
 						   IEEE80211_TX_CTRL_MLO_LINK);
 			struct iwl_mvm_vif_link_info *link;
@@ -831,18 +851,6 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 		} else if (info.control.vif->type == NL80211_IFTYPE_MONITOR) {
 			queue = mvm->snif_queue;
 			sta_id = mvm->snif_sta.sta_id;
-		} else if (info.control.vif->type == NL80211_IFTYPE_STATION &&
-			   offchannel) {
-			/*
-			 * IWL_MVM_OFFCHANNEL_QUEUE is used for ROC packets
-			 * that can be used in 2 different types of vifs, P2P &
-			 * STATION.
-			 * P2P uses the offchannel queue.
-			 * STATION (HS2.0) uses the auxiliary context of the FW,
-			 * and hence needs to be sent on the aux queue.
-			 */
-			sta_id = mvm->aux_sta.sta_id;
-			queue = mvm->aux_queue;
 		}
 	}
 
-- 
cgit v1.2.3-58-ga151


From aea9165ccfd1577be801fd75515fc9d4a9595b90 Mon Sep 17 00:00:00 2001
From: Erick Archer <erick.archer@outlook.com>
Date: Sun, 2 Jun 2024 18:13:09 +0200
Subject: wifi: at76c50x: use sizeof(*pointer) instead of sizeof(type)

It is preferred to use sizeof(*pointer) instead of sizeof(type)
due to the type of the variable can change and one needs not
change the former (unlike the latter). This patch has no effect
on runtime behavior.

At the same time remove some redundant NULL initializations.

Signed-off-by: Erick Archer <erick.archer@outlook.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/AS8PR02MB7237C784C14DBC943CB719F88BFE2@AS8PR02MB7237.eurprd02.prod.outlook.com
---
 drivers/net/wireless/atmel/at76c50x-usb.c | 44 +++++++++++++------------------
 1 file changed, 19 insertions(+), 25 deletions(-)

diff --git a/drivers/net/wireless/atmel/at76c50x-usb.c b/drivers/net/wireless/atmel/at76c50x-usb.c
index 0b55a272bfd6..0ca2f629683d 100644
--- a/drivers/net/wireless/atmel/at76c50x-usb.c
+++ b/drivers/net/wireless/atmel/at76c50x-usb.c
@@ -332,7 +332,7 @@ static int at76_dfu_get_status(struct usb_device *udev,
 
 	ret = usb_control_msg(udev, usb_rcvctrlpipe(udev, 0), DFU_GETSTATUS,
 			      USB_TYPE_CLASS | USB_DIR_IN | USB_RECIP_INTERFACE,
-			      0, 0, status, sizeof(struct dfu_status),
+			      0, 0, status, sizeof(*status),
 			      USB_CTRL_GET_TIMEOUT);
 	return ret;
 }
@@ -366,7 +366,7 @@ static int at76_usbdfu_download(struct usb_device *udev, u8 *buf, u32 size,
 	u32 dfu_timeout = 0;
 	int bsize = 0;
 	int blockno = 0;
-	struct dfu_status *dfu_stat_buf = NULL;
+	struct dfu_status *dfu_stat_buf;
 	u8 *dfu_state = NULL;
 	u8 *block = NULL;
 
@@ -378,7 +378,7 @@ static int at76_usbdfu_download(struct usb_device *udev, u8 *buf, u32 size,
 		return -EINVAL;
 	}
 
-	dfu_stat_buf = kmalloc(sizeof(struct dfu_status), GFP_KERNEL);
+	dfu_stat_buf = kmalloc(sizeof(*dfu_stat_buf), GFP_KERNEL);
 	if (!dfu_stat_buf) {
 		ret = -ENOMEM;
 		goto exit;
@@ -931,14 +931,12 @@ static void at76_dump_mib_mac_addr(struct at76_priv *priv)
 {
 	int i;
 	int ret;
-	struct mib_mac_addr *m = kmalloc(sizeof(struct mib_mac_addr),
-					 GFP_KERNEL);
+	struct mib_mac_addr *m = kmalloc(sizeof(*m), GFP_KERNEL);
 
 	if (!m)
 		return;
 
-	ret = at76_get_mib(priv->udev, MIB_MAC_ADDR, m,
-			   sizeof(struct mib_mac_addr));
+	ret = at76_get_mib(priv->udev, MIB_MAC_ADDR, m, sizeof(*m));
 	if (ret < 0) {
 		wiphy_err(priv->hw->wiphy,
 			  "at76_get_mib (MAC_ADDR) failed: %d\n", ret);
@@ -961,13 +959,12 @@ static void at76_dump_mib_mac_wep(struct at76_priv *priv)
 	int i;
 	int ret;
 	int key_len;
-	struct mib_mac_wep *m = kmalloc(sizeof(struct mib_mac_wep), GFP_KERNEL);
+	struct mib_mac_wep *m = kmalloc(sizeof(*m), GFP_KERNEL);
 
 	if (!m)
 		return;
 
-	ret = at76_get_mib(priv->udev, MIB_MAC_WEP, m,
-			   sizeof(struct mib_mac_wep));
+	ret = at76_get_mib(priv->udev, MIB_MAC_WEP, m, sizeof(*m));
 	if (ret < 0) {
 		wiphy_err(priv->hw->wiphy,
 			  "at76_get_mib (MAC_WEP) failed: %d\n", ret);
@@ -997,14 +994,12 @@ exit:
 static void at76_dump_mib_mac_mgmt(struct at76_priv *priv)
 {
 	int ret;
-	struct mib_mac_mgmt *m = kmalloc(sizeof(struct mib_mac_mgmt),
-					 GFP_KERNEL);
+	struct mib_mac_mgmt *m = kmalloc(sizeof(*m), GFP_KERNEL);
 
 	if (!m)
 		return;
 
-	ret = at76_get_mib(priv->udev, MIB_MAC_MGMT, m,
-			   sizeof(struct mib_mac_mgmt));
+	ret = at76_get_mib(priv->udev, MIB_MAC_MGMT, m, sizeof(*m));
 	if (ret < 0) {
 		wiphy_err(priv->hw->wiphy,
 			  "at76_get_mib (MAC_MGMT) failed: %d\n", ret);
@@ -1035,12 +1030,12 @@ exit:
 static void at76_dump_mib_mac(struct at76_priv *priv)
 {
 	int ret;
-	struct mib_mac *m = kmalloc(sizeof(struct mib_mac), GFP_KERNEL);
+	struct mib_mac *m = kmalloc(sizeof(*m), GFP_KERNEL);
 
 	if (!m)
 		return;
 
-	ret = at76_get_mib(priv->udev, MIB_MAC, m, sizeof(struct mib_mac));
+	ret = at76_get_mib(priv->udev, MIB_MAC, m, sizeof(*m));
 	if (ret < 0) {
 		wiphy_err(priv->hw->wiphy,
 			  "at76_get_mib (MAC) failed: %d\n", ret);
@@ -1072,12 +1067,12 @@ exit:
 static void at76_dump_mib_phy(struct at76_priv *priv)
 {
 	int ret;
-	struct mib_phy *m = kmalloc(sizeof(struct mib_phy), GFP_KERNEL);
+	struct mib_phy *m = kmalloc(sizeof(*m), GFP_KERNEL);
 
 	if (!m)
 		return;
 
-	ret = at76_get_mib(priv->udev, MIB_PHY, m, sizeof(struct mib_phy));
+	ret = at76_get_mib(priv->udev, MIB_PHY, m, sizeof(*m));
 	if (ret < 0) {
 		wiphy_err(priv->hw->wiphy,
 			  "at76_get_mib (PHY) failed: %d\n", ret);
@@ -1130,13 +1125,12 @@ exit:
 static void at76_dump_mib_mdomain(struct at76_priv *priv)
 {
 	int ret;
-	struct mib_mdomain *m = kmalloc(sizeof(struct mib_mdomain), GFP_KERNEL);
+	struct mib_mdomain *m = kmalloc(sizeof(*m), GFP_KERNEL);
 
 	if (!m)
 		return;
 
-	ret = at76_get_mib(priv->udev, MIB_MDOMAIN, m,
-			   sizeof(struct mib_mdomain));
+	ret = at76_get_mib(priv->udev, MIB_MDOMAIN, m, sizeof(*m));
 	if (ret < 0) {
 		wiphy_err(priv->hw->wiphy,
 			  "at76_get_mib (MDOMAIN) failed: %d\n", ret);
@@ -1375,7 +1369,7 @@ static int at76_startup_device(struct at76_priv *priv)
 		 priv->scan_min_time, priv->scan_max_time,
 		 priv->scan_mode == SCAN_TYPE_ACTIVE ? "active" : "passive");
 
-	memset(ccfg, 0, sizeof(struct at76_card_config));
+	memset(ccfg, 0, sizeof(*ccfg));
 	ccfg->promiscuous_mode = 0;
 	ccfg->short_retry_limit = priv->short_retry_limit;
 
@@ -1411,7 +1405,7 @@ static int at76_startup_device(struct at76_priv *priv)
 	ccfg->beacon_period = cpu_to_le16(priv->beacon_period);
 
 	ret = at76_set_card_command(priv->udev, CMD_STARTUP, &priv->card_config,
-				    sizeof(struct at76_card_config));
+				    sizeof(*ccfg));
 	if (ret < 0) {
 		wiphy_err(priv->hw->wiphy, "at76_set_card_command failed: %d\n",
 			  ret);
@@ -2443,7 +2437,7 @@ static int at76_probe(struct usb_interface *interface,
 	struct usb_device *udev;
 	int op_mode;
 	int need_ext_fw = 0;
-	struct mib_fw_version *fwv = NULL;
+	struct mib_fw_version *fwv;
 	int board_type = (int)id->driver_info;
 
 	udev = usb_get_dev(interface_to_usbdev(interface));
@@ -2531,7 +2525,7 @@ static int at76_probe(struct usb_interface *interface,
 
 	usb_set_intfdata(interface, priv);
 
-	memcpy(&priv->fw_version, fwv, sizeof(struct mib_fw_version));
+	memcpy(&priv->fw_version, fwv, sizeof(*fwv));
 	priv->board_type = board_type;
 
 	ret = at76_init_new_device(priv, interface);
-- 
cgit v1.2.3-58-ga151


From bbef1d006cb1c2774c4b44eb70441d784d3aef44 Mon Sep 17 00:00:00 2001
From: Erick Archer <erick.archer@outlook.com>
Date: Sun, 2 Jun 2024 18:13:10 +0200
Subject: wifi: at76c50x: prefer struct_size over open coded arithmetic

This is an effort to get rid of all multiplications from allocation
functions in order to prevent integer overflows [1][2].

As the "cmd_buf" variable is a pointer to "struct at76_command" and
this structure ends in a flexible array:

struct at76_command {
	[...]
	u8 data[];
} __packed;

the preferred way in the kernel is to use the struct_size() helper to
do the arithmetic instead of the calculation "size + count" in the
kmalloc() function.

Also, declare a new variable (total_size) since the return value of the
struct_size() helper is used several times.

At the same time, prepare for the coming implementation by GCC and Clang
of the __counted_by attribute. Flexible array members annotated with
__counted_by can have their accesses bounds-checked at run-time via
CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for
strcpy/memcpy-family functions). In this case, it is important to note
that the attribute used is "__counted_by_le" since the counter type is
"__le16".

This way, the code is more readable and safer.

Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#open-coded-arithmetic-in-allocator-arguments [1]
Link: https://github.com/KSPP/linux/issues/160 [2]
Signed-off-by: Erick Archer <erick.archer@outlook.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/AS8PR02MB7237578654CEDDFE5F8C17BA8BFE2@AS8PR02MB7237.eurprd02.prod.outlook.com
---
 drivers/net/wireless/atmel/at76c50x-usb.c | 12 ++++++------
 drivers/net/wireless/atmel/at76c50x-usb.h |  2 +-
 2 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/atmel/at76c50x-usb.c b/drivers/net/wireless/atmel/at76c50x-usb.c
index 0ca2f629683d..baa53cfefe48 100644
--- a/drivers/net/wireless/atmel/at76c50x-usb.c
+++ b/drivers/net/wireless/atmel/at76c50x-usb.c
@@ -721,9 +721,11 @@ static int at76_set_card_command(struct usb_device *udev, u8 cmd, void *buf,
 				 int buf_size)
 {
 	int ret;
-	struct at76_command *cmd_buf = kmalloc(sizeof(struct at76_command) +
-					       buf_size, GFP_KERNEL);
+	size_t total_size;
+	struct at76_command *cmd_buf;
 
+	total_size = struct_size(cmd_buf, data, buf_size);
+	cmd_buf = kmalloc(total_size, GFP_KERNEL);
 	if (!cmd_buf)
 		return -ENOMEM;
 
@@ -732,15 +734,13 @@ static int at76_set_card_command(struct usb_device *udev, u8 cmd, void *buf,
 	cmd_buf->size = cpu_to_le16(buf_size);
 	memcpy(cmd_buf->data, buf, buf_size);
 
-	at76_dbg_dump(DBG_CMD, cmd_buf, sizeof(struct at76_command) + buf_size,
+	at76_dbg_dump(DBG_CMD, cmd_buf, total_size,
 		      "issuing command %s (0x%02x)",
 		      at76_get_cmd_string(cmd), cmd);
 
 	ret = usb_control_msg(udev, usb_sndctrlpipe(udev, 0), 0x0e,
 			      USB_TYPE_VENDOR | USB_DIR_OUT | USB_RECIP_DEVICE,
-			      0, 0, cmd_buf,
-			      sizeof(struct at76_command) + buf_size,
-			      USB_CTRL_GET_TIMEOUT);
+			      0, 0, cmd_buf, total_size, USB_CTRL_GET_TIMEOUT);
 	kfree(cmd_buf);
 	return ret;
 }
diff --git a/drivers/net/wireless/atmel/at76c50x-usb.h b/drivers/net/wireless/atmel/at76c50x-usb.h
index 746e64dfd8aa..843146a0de64 100644
--- a/drivers/net/wireless/atmel/at76c50x-usb.h
+++ b/drivers/net/wireless/atmel/at76c50x-usb.h
@@ -151,7 +151,7 @@ struct at76_command {
 	u8 cmd;
 	u8 reserved;
 	__le16 size;
-	u8 data[];
+	u8 data[] __counted_by_le(size);
 } __packed;
 
 /* Length of Atmel-specific Rx header before 802.11 frame */
-- 
cgit v1.2.3-58-ga151


From 158d5a1b3caaaeafc838a28a78dc17ceba468fb6 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Mon, 3 Jun 2024 11:15:35 +0200
Subject: wifi: brcmfmac: use 'time_left' variable with wait_event_timeout()

There is a confusing pattern in the kernel to use a variable named 'timeout' to
store the result of wait_event_timeout() causing patterns like:

	timeout = wait_event_timeout(...)
	if (!timeout) return -ETIMEDOUT;

with all kinds of permutations. Use 'time_left' as a variable to make the code
self explaining.

Fix to the proper variable type 'long' while here.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240603091541.8367-3-wsa+renesas@sang-engineering.com
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
index 5fe0e671ecb3..1585a5653ee4 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c
@@ -4071,7 +4071,7 @@ static void brcmf_report_wowl_wakeind(struct wiphy *wiphy, struct brcmf_if *ifp)
 	struct cfg80211_wowlan_wakeup *wakeup;
 	u32 wakeind;
 	s32 err;
-	int timeout;
+	long time_left;
 
 	err = brcmf_fil_iovar_data_get(ifp, "wowl_wakeind", &wake_ind_le,
 				       sizeof(wake_ind_le));
@@ -4113,10 +4113,10 @@ static void brcmf_report_wowl_wakeind(struct wiphy *wiphy, struct brcmf_if *ifp)
 		}
 		if (wakeind & BRCMF_WOWL_PFN_FOUND) {
 			brcmf_dbg(INFO, "WOWL Wake indicator: BRCMF_WOWL_PFN_FOUND\n");
-			timeout = wait_event_timeout(cfg->wowl.nd_data_wait,
-				cfg->wowl.nd_data_completed,
-				BRCMF_ND_INFO_TIMEOUT);
-			if (!timeout)
+			time_left = wait_event_timeout(cfg->wowl.nd_data_wait,
+						       cfg->wowl.nd_data_completed,
+						       BRCMF_ND_INFO_TIMEOUT);
+			if (!time_left)
 				bphy_err(drvr, "No result for wowl net detect\n");
 			else
 				wakeup_data.net_detect = cfg->wowl.nd_info;
-- 
cgit v1.2.3-58-ga151


From 0c0668813cc0bb919b2b693c411e1d7d0dfdf39b Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Mon, 3 Jun 2024 11:15:36 +0200
Subject: wifi: mwl8k: use 'time_left' variable with
 wait_for_completion_timeout()

There is a confusing pattern in the kernel to use a variable named 'timeout' to
store the result of wait_for_completion_timeout() causing patterns like:

	timeout = wait_for_completion_timeout(...)
	if (!timeout) return -ETIMEDOUT;

with all kinds of permutations. Use 'time_left' as a variable to make the code
self explaining.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240603091541.8367-4-wsa+renesas@sang-engineering.com
---
 drivers/net/wireless/marvell/mwl8k.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/marvell/mwl8k.c b/drivers/net/wireless/marvell/mwl8k.c
index d3d07bb26335..241a02a0accd 100644
--- a/drivers/net/wireless/marvell/mwl8k.c
+++ b/drivers/net/wireless/marvell/mwl8k.c
@@ -2211,7 +2211,7 @@ static int mwl8k_post_cmd(struct ieee80211_hw *hw, struct mwl8k_cmd_pkt_hdr *cmd
 	dma_addr_t dma_addr;
 	unsigned int dma_size;
 	int rc;
-	unsigned long timeout = 0;
+	unsigned long time_left = 0;
 	u8 buf[32];
 	u32 bitmap = 0;
 
@@ -2258,8 +2258,8 @@ static int mwl8k_post_cmd(struct ieee80211_hw *hw, struct mwl8k_cmd_pkt_hdr *cmd
 	iowrite32(MWL8K_H2A_INT_DUMMY,
 		regs + MWL8K_HIU_H2A_INTERRUPT_EVENTS);
 
-	timeout = wait_for_completion_timeout(&cmd_wait,
-				msecs_to_jiffies(MWL8K_CMD_TIMEOUT_MS));
+	time_left = wait_for_completion_timeout(&cmd_wait,
+						msecs_to_jiffies(MWL8K_CMD_TIMEOUT_MS));
 
 	priv->hostcmd_wait = NULL;
 
@@ -2267,7 +2267,7 @@ static int mwl8k_post_cmd(struct ieee80211_hw *hw, struct mwl8k_cmd_pkt_hdr *cmd
 	dma_unmap_single(&priv->pdev->dev, dma_addr, dma_size,
 			 DMA_BIDIRECTIONAL);
 
-	if (!timeout) {
+	if (!time_left) {
 		wiphy_err(hw->wiphy, "Command %s timeout after %u ms\n",
 			  mwl8k_cmd_name(cmd->code, buf, sizeof(buf)),
 			  MWL8K_CMD_TIMEOUT_MS);
@@ -2275,7 +2275,7 @@ static int mwl8k_post_cmd(struct ieee80211_hw *hw, struct mwl8k_cmd_pkt_hdr *cmd
 	} else {
 		int ms;
 
-		ms = MWL8K_CMD_TIMEOUT_MS - jiffies_to_msecs(timeout);
+		ms = MWL8K_CMD_TIMEOUT_MS - jiffies_to_msecs(time_left);
 
 		rc = cmd->result ? -EINVAL : 0;
 		if (rc)
-- 
cgit v1.2.3-58-ga151


From a37f6947ff07896d5132d6fbfec538de01c74287 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Mon, 3 Jun 2024 11:15:37 +0200
Subject: wifi: p54: use 'time_left' variable with
 wait_for_completion_interruptible_timeout()

There is a confusing pattern in the kernel to use a variable named 'timeout' to
store the result of wait_for_completion_interruptible_timeout() causing patterns like:

	timeout = wait_for_completion_interruptible_timeout(...)
	if (!timeout) return -ETIMEDOUT;

with all kinds of permutations. Use 'time_left' as a variable to make the code
self explaining.

Fix to the proper variable type 'long' while here.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240603091541.8367-5-wsa+renesas@sang-engineering.com
---
 drivers/net/wireless/intersil/p54/fwio.c   |  6 +++---
 drivers/net/wireless/intersil/p54/p54pci.c |  8 ++++----
 drivers/net/wireless/intersil/p54/p54spi.c | 10 +++++-----
 3 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/intersil/p54/fwio.c b/drivers/net/wireless/intersil/p54/fwio.c
index c4fe70e05b9b..772084a9bd8d 100644
--- a/drivers/net/wireless/intersil/p54/fwio.c
+++ b/drivers/net/wireless/intersil/p54/fwio.c
@@ -216,7 +216,7 @@ int p54_download_eeprom(struct p54_common *priv, void *buf,
 	struct sk_buff *skb;
 	size_t eeprom_hdr_size;
 	int ret = 0;
-	long timeout;
+	long time_left;
 
 	if (priv->fw_var >= 0x509)
 		eeprom_hdr_size = sizeof(*eeprom_hdr);
@@ -245,9 +245,9 @@ int p54_download_eeprom(struct p54_common *priv, void *buf,
 
 	p54_tx(priv, skb);
 
-	timeout = wait_for_completion_interruptible_timeout(
+	time_left = wait_for_completion_interruptible_timeout(
 			&priv->eeprom_comp, HZ);
-	if (timeout <= 0) {
+	if (time_left <= 0) {
 		wiphy_err(priv->hw->wiphy,
 			"device does not respond or signal received!\n");
 		ret = -EBUSY;
diff --git a/drivers/net/wireless/intersil/p54/p54pci.c b/drivers/net/wireless/intersil/p54/p54pci.c
index e97ee547b9f3..6588f5477762 100644
--- a/drivers/net/wireless/intersil/p54/p54pci.c
+++ b/drivers/net/wireless/intersil/p54/p54pci.c
@@ -434,7 +434,7 @@ static int p54p_open(struct ieee80211_hw *dev)
 {
 	struct p54p_priv *priv = dev->priv;
 	int err;
-	long timeout;
+	long time_left;
 
 	init_completion(&priv->boot_comp);
 	err = request_irq(priv->pdev->irq, p54p_interrupt,
@@ -472,12 +472,12 @@ static int p54p_open(struct ieee80211_hw *dev)
 	P54P_WRITE(dev_int, cpu_to_le32(ISL38XX_DEV_INT_RESET));
 	P54P_READ(dev_int);
 
-	timeout = wait_for_completion_interruptible_timeout(
+	time_left = wait_for_completion_interruptible_timeout(
 			&priv->boot_comp, HZ);
-	if (timeout <= 0) {
+	if (time_left <= 0) {
 		wiphy_err(dev->wiphy, "Cannot boot firmware!\n");
 		p54p_stop(dev);
-		return timeout ? -ERESTARTSYS : -ETIMEDOUT;
+		return time_left ? -ERESTARTSYS : -ETIMEDOUT;
 	}
 
 	P54P_WRITE(int_enable, cpu_to_le32(ISL38XX_INT_IDENT_UPDATE));
diff --git a/drivers/net/wireless/intersil/p54/p54spi.c b/drivers/net/wireless/intersil/p54/p54spi.c
index 0073b5e0f9c9..d33a994906a7 100644
--- a/drivers/net/wireless/intersil/p54/p54spi.c
+++ b/drivers/net/wireless/intersil/p54/p54spi.c
@@ -518,7 +518,7 @@ out:
 static int p54spi_op_start(struct ieee80211_hw *dev)
 {
 	struct p54s_priv *priv = dev->priv;
-	unsigned long timeout;
+	long time_left;
 	int ret = 0;
 
 	if (mutex_lock_interruptible(&priv->mutex)) {
@@ -538,10 +538,10 @@ static int p54spi_op_start(struct ieee80211_hw *dev)
 
 	mutex_unlock(&priv->mutex);
 
-	timeout = msecs_to_jiffies(2000);
-	timeout = wait_for_completion_interruptible_timeout(&priv->fw_comp,
-							    timeout);
-	if (!timeout) {
+	time_left = msecs_to_jiffies(2000);
+	time_left = wait_for_completion_interruptible_timeout(&priv->fw_comp,
+							      time_left);
+	if (!time_left) {
 		dev_err(&priv->spi->dev, "firmware boot failed");
 		p54spi_power_off(priv);
 		ret = -1;
-- 
cgit v1.2.3-58-ga151


From a2ead3445a6317c5d2b53711cf9a98e54c395223 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Mon, 3 Jun 2024 11:15:39 +0200
Subject: wifi: zd1211rw: use 'time_left' variable with
 wait_for_completion_timeout()

There is a confusing pattern in the kernel to use a variable named 'timeout' to
store the result of wait_for_completion_timeout() causing patterns like:

	timeout = wait_for_completion_timeout(...)
	if (!timeout) return -ETIMEDOUT;

with all kinds of permutations. Use 'time_left' as a variable to make the code
self explaining.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240603091541.8367-7-wsa+renesas@sang-engineering.com
---
 drivers/net/wireless/zydas/zd1211rw/zd_usb.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/zydas/zd1211rw/zd_usb.c b/drivers/net/wireless/zydas/zd1211rw/zd_usb.c
index f3b567a13ded..2ee4218da1c5 100644
--- a/drivers/net/wireless/zydas/zd1211rw/zd_usb.c
+++ b/drivers/net/wireless/zydas/zd1211rw/zd_usb.c
@@ -1698,7 +1698,7 @@ int zd_usb_ioread16v(struct zd_usb *usb, u16 *values,
 	int r, i, req_len, actual_req_len, try_count = 0;
 	struct usb_device *udev;
 	struct usb_req_read_regs *req = NULL;
-	unsigned long timeout;
+	unsigned long time_left;
 	bool retry = false;
 
 	if (count < 1) {
@@ -1748,9 +1748,9 @@ retry_read:
 		goto error;
 	}
 
-	timeout = wait_for_completion_timeout(&usb->intr.read_regs.completion,
-					      msecs_to_jiffies(50));
-	if (!timeout) {
+	time_left = wait_for_completion_timeout(&usb->intr.read_regs.completion,
+						msecs_to_jiffies(50));
+	if (!time_left) {
 		disable_read_regs_int(usb);
 		dev_dbg_f(zd_usb_dev(usb), "read timed out\n");
 		r = -ETIMEDOUT;
-- 
cgit v1.2.3-58-ga151


From ebb79e96f1ea454fbcc8fe27dfe44e751bd74b4b Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:25 -0600
Subject: kbuild: bpf: Tell pahole to DECL_TAG kfuncs

With [0], pahole can now discover kfuncs and inject DECL_TAG
into BTF. With this commit, we will start shipping said DECL_TAGs
to downstream consumers if pahole supports it.

This is useful for feature probing kfuncs as well as generating
compilable prototypes. This is particularly important as kfuncs
do not have stable ABI.

[0]: https://git.kernel.org/pub/scm/devel/pahole/pahole.git/commit/?id=72e88f29c6f7e14201756e65bd66157427a61aaf

Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/324aac5c627bddb80d9968c30df6382846994cc8.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 scripts/Makefile.btf | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/Makefile.btf b/scripts/Makefile.btf
index bca8a8f26ea4..2597e3d4d6e0 100644
--- a/scripts/Makefile.btf
+++ b/scripts/Makefile.btf
@@ -19,7 +19,7 @@ pahole-flags-$(call test-ge, $(pahole-ver), 125)	+= --skip_encoding_btf_inconsis
 else
 
 # Switch to using --btf_features for v1.26 and later.
-pahole-flags-$(call test-ge, $(pahole-ver), 126)  = -j --btf_features=encode_force,var,float,enum64,decl_tag,type_tag,optimized_func,consistent_func
+pahole-flags-$(call test-ge, $(pahole-ver), 126)  = -j --btf_features=encode_force,var,float,enum64,decl_tag,type_tag,optimized_func,consistent_func,decl_tag_kfuncs
 
 endif
 
-- 
cgit v1.2.3-58-ga151


From 718135f5bd24ec10ff38aa0294a7da0a7b99fa89 Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:26 -0600
Subject: bpf: selftests: Fix bpf_iter_task_vma_new() prototype

bpf_iter_task_vma_new() is defined as taking a u64 as its 3rd argument.
u64 is a unsigned long long. bpf_experimental.h was defining the
prototype as unsigned long.

Fix by using __u64.

Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/fab4509bfee914f539166a91c3ff41e949f3df30.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/bpf_experimental.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/bpf_experimental.h b/tools/testing/selftests/bpf/bpf_experimental.h
index 3d9e4b8c6b81..8ee7a00b7c82 100644
--- a/tools/testing/selftests/bpf/bpf_experimental.h
+++ b/tools/testing/selftests/bpf/bpf_experimental.h
@@ -163,7 +163,7 @@ struct bpf_iter_task_vma;
 
 extern int bpf_iter_task_vma_new(struct bpf_iter_task_vma *it,
 				 struct task_struct *task,
-				 unsigned long addr) __ksym;
+				 __u64 addr) __ksym;
 extern struct vm_area_struct *bpf_iter_task_vma_next(struct bpf_iter_task_vma *it) __ksym;
 extern void bpf_iter_task_vma_destroy(struct bpf_iter_task_vma *it) __ksym;
 
-- 
cgit v1.2.3-58-ga151


From dff96e4f5078c6c61fc6c36dddf27b124c4318fc Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:27 -0600
Subject: bpf: selftests: Fix fentry test kfunc prototypes

Some prototypes in progs/get_func_ip_test.c were not in line with how the
actual kfuncs are defined in net/bpf/test_run.c. This causes compilation
errors when kfunc prototypes are generated from BTF.

Fix by aligning with actual kfunc definitions.

Also remove two unused prototypes.

Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/1e68870e7626b7b9c6420e65076b307fc404a2f0.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/get_func_ip_test.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/get_func_ip_test.c b/tools/testing/selftests/bpf/progs/get_func_ip_test.c
index 8956eb78a226..2011cacdeb18 100644
--- a/tools/testing/selftests/bpf/progs/get_func_ip_test.c
+++ b/tools/testing/selftests/bpf/progs/get_func_ip_test.c
@@ -5,13 +5,12 @@
 
 char _license[] SEC("license") = "GPL";
 
-extern const void bpf_fentry_test1 __ksym;
+extern int bpf_fentry_test1(int a) __ksym;
+extern int bpf_modify_return_test(int a, int *b) __ksym;
+
 extern const void bpf_fentry_test2 __ksym;
 extern const void bpf_fentry_test3 __ksym;
 extern const void bpf_fentry_test4 __ksym;
-extern const void bpf_modify_return_test __ksym;
-extern const void bpf_fentry_test6 __ksym;
-extern const void bpf_fentry_test7 __ksym;
 
 extern bool CONFIG_X86_KERNEL_IBT __kconfig __weak;
 
-- 
cgit v1.2.3-58-ga151


From 89f0b1abac497c47d0851b780abecc756c1e8734 Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:28 -0600
Subject: bpf: selftests: Fix bpf_cpumask_first_zero() kfunc prototype

The prototype in progs/nested_trust_common.h is not in line with how the
actual kfuncs are defined in kernel/bpf/cpumask.c. This causes compilation
errors when kfunc prototypes are generated from BTF.

Fix by aligning with actual kfunc definitions.

Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/437936a4e554b02e04566dd6e3f0a5d08370cc8c.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/nested_trust_common.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/nested_trust_common.h b/tools/testing/selftests/bpf/progs/nested_trust_common.h
index 83d33931136e..1784b496be2e 100644
--- a/tools/testing/selftests/bpf/progs/nested_trust_common.h
+++ b/tools/testing/selftests/bpf/progs/nested_trust_common.h
@@ -7,6 +7,6 @@
 #include <stdbool.h>
 
 bool bpf_cpumask_test_cpu(unsigned int cpu, const struct cpumask *cpumask) __ksym;
-bool bpf_cpumask_first_zero(const struct cpumask *cpumask) __ksym;
+__u32 bpf_cpumask_first_zero(const struct cpumask *cpumask) __ksym;
 
 #endif /* _NESTED_TRUST_COMMON_H */
-- 
cgit v1.2.3-58-ga151


From ac42f636dc11b2e8d6dea9dd5bb10a39c7bec342 Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:29 -0600
Subject: bpf: selftests: Fix bpf_map_sum_elem_count() kfunc prototype

The prototype in progs/map_percpu_stats.c is not in line with how the
actual kfuncs are defined in kernel/bpf/map_iter.c. This causes
compilation errors when kfunc prototypes are generated from BTF.

Fix by aligning with actual kfunc definitions.

Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/0497e11a71472dcb71ada7c90ad691523ae87c3b.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/map_percpu_stats.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/map_percpu_stats.c b/tools/testing/selftests/bpf/progs/map_percpu_stats.c
index 10b2325c1720..63245785eb69 100644
--- a/tools/testing/selftests/bpf/progs/map_percpu_stats.c
+++ b/tools/testing/selftests/bpf/progs/map_percpu_stats.c
@@ -7,7 +7,7 @@
 
 __u32 target_id;
 
-__s64 bpf_map_sum_elem_count(struct bpf_map *map) __ksym;
+__s64 bpf_map_sum_elem_count(const struct bpf_map *map) __ksym;
 
 SEC("iter/bpf_map")
 int dump_bpf_map(struct bpf_iter__bpf_map *ctx)
-- 
cgit v1.2.3-58-ga151


From 2b8dd87332cd2782b5b3f0c423bd6693e487ed30 Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:30 -0600
Subject: bpf: Make bpf_session_cookie() kfunc return long *

We will soon be generating kfunc prototypes from BTF. As part of that,
we need to align the manual signatures in bpf_kfuncs.h with the actual
kfunc definitions. There is currently a conflicting signature for
bpf_session_cookie() w.r.t. return type.

The original intent was to return long * and not __u64 *. You can see
evidence of that intent in a3a5113393cc ("selftests/bpf: Add kprobe
session cookie test").

Fix conflict by changing kfunc definition.

Fixes: 5c919acef851 ("bpf: Add support for kprobe session cookie")
Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/7043e1c251ab33151d6e3830f8ea1902ed2604ac.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/trace/bpf_trace.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index d1daeab1bbc1..bc16e21a2a44 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -3527,7 +3527,7 @@ __bpf_kfunc bool bpf_session_is_return(void)
 	return session_ctx->is_return;
 }
 
-__bpf_kfunc __u64 *bpf_session_cookie(void)
+__bpf_kfunc long *bpf_session_cookie(void)
 {
 	struct bpf_session_run_ctx *session_ctx;
 
-- 
cgit v1.2.3-58-ga151


From 0ce089cbdc6a393bf9ad04964427852800503a58 Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:31 -0600
Subject: bpf: selftests: Namespace struct_opt callbacks in bpf_dctcp

With generated kfunc prototypes, the existing callback names will
conflict. Fix by namespacing with a bpf_ prefix.

Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/efe7aadad8a054e5aeeba94b1d2e4502eee09d7a.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/bpf_dctcp.c | 36 +++++++++++++--------------
 1 file changed, 18 insertions(+), 18 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/bpf_dctcp.c b/tools/testing/selftests/bpf/progs/bpf_dctcp.c
index 3c9ffe340312..02f552e7fd4d 100644
--- a/tools/testing/selftests/bpf/progs/bpf_dctcp.c
+++ b/tools/testing/selftests/bpf/progs/bpf_dctcp.c
@@ -65,7 +65,7 @@ static void dctcp_reset(const struct tcp_sock *tp, struct bpf_dctcp *ca)
 }
 
 SEC("struct_ops")
-void BPF_PROG(dctcp_init, struct sock *sk)
+void BPF_PROG(bpf_dctcp_init, struct sock *sk)
 {
 	const struct tcp_sock *tp = tcp_sk(sk);
 	struct bpf_dctcp *ca = inet_csk_ca(sk);
@@ -77,7 +77,7 @@ void BPF_PROG(dctcp_init, struct sock *sk)
 				   (void *)fallback, sizeof(fallback)) == -EBUSY)
 			ebusy_cnt++;
 
-		/* Switch back to myself and the recurred dctcp_init()
+		/* Switch back to myself and the recurred bpf_dctcp_init()
 		 * will get -EBUSY for all bpf_setsockopt(TCP_CONGESTION),
 		 * except the last "cdg" one.
 		 */
@@ -112,7 +112,7 @@ void BPF_PROG(dctcp_init, struct sock *sk)
 }
 
 SEC("struct_ops")
-__u32 BPF_PROG(dctcp_ssthresh, struct sock *sk)
+__u32 BPF_PROG(bpf_dctcp_ssthresh, struct sock *sk)
 {
 	struct bpf_dctcp *ca = inet_csk_ca(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
@@ -122,7 +122,7 @@ __u32 BPF_PROG(dctcp_ssthresh, struct sock *sk)
 }
 
 SEC("struct_ops")
-void BPF_PROG(dctcp_update_alpha, struct sock *sk, __u32 flags)
+void BPF_PROG(bpf_dctcp_update_alpha, struct sock *sk, __u32 flags)
 {
 	const struct tcp_sock *tp = tcp_sk(sk);
 	struct bpf_dctcp *ca = inet_csk_ca(sk);
@@ -161,12 +161,12 @@ static void dctcp_react_to_loss(struct sock *sk)
 }
 
 SEC("struct_ops")
-void BPF_PROG(dctcp_state, struct sock *sk, __u8 new_state)
+void BPF_PROG(bpf_dctcp_state, struct sock *sk, __u8 new_state)
 {
 	if (new_state == TCP_CA_Recovery &&
 	    new_state != BPF_CORE_READ_BITFIELD(inet_csk(sk), icsk_ca_state))
 		dctcp_react_to_loss(sk);
-	/* We handle RTO in dctcp_cwnd_event to ensure that we perform only
+	/* We handle RTO in bpf_dctcp_cwnd_event to ensure that we perform only
 	 * one loss-adjustment per RTT.
 	 */
 }
@@ -208,7 +208,7 @@ static void dctcp_ece_ack_update(struct sock *sk, enum tcp_ca_event evt,
 }
 
 SEC("struct_ops")
-void BPF_PROG(dctcp_cwnd_event, struct sock *sk, enum tcp_ca_event ev)
+void BPF_PROG(bpf_dctcp_cwnd_event, struct sock *sk, enum tcp_ca_event ev)
 {
 	struct bpf_dctcp *ca = inet_csk_ca(sk);
 
@@ -227,7 +227,7 @@ void BPF_PROG(dctcp_cwnd_event, struct sock *sk, enum tcp_ca_event ev)
 }
 
 SEC("struct_ops")
-__u32 BPF_PROG(dctcp_cwnd_undo, struct sock *sk)
+__u32 BPF_PROG(bpf_dctcp_cwnd_undo, struct sock *sk)
 {
 	const struct bpf_dctcp *ca = inet_csk_ca(sk);
 
@@ -237,28 +237,28 @@ __u32 BPF_PROG(dctcp_cwnd_undo, struct sock *sk)
 extern void tcp_reno_cong_avoid(struct sock *sk, __u32 ack, __u32 acked) __ksym;
 
 SEC("struct_ops")
-void BPF_PROG(dctcp_cong_avoid, struct sock *sk, __u32 ack, __u32 acked)
+void BPF_PROG(bpf_dctcp_cong_avoid, struct sock *sk, __u32 ack, __u32 acked)
 {
 	tcp_reno_cong_avoid(sk, ack, acked);
 }
 
 SEC(".struct_ops")
 struct tcp_congestion_ops dctcp_nouse = {
-	.init		= (void *)dctcp_init,
-	.set_state	= (void *)dctcp_state,
+	.init		= (void *)bpf_dctcp_init,
+	.set_state	= (void *)bpf_dctcp_state,
 	.flags		= TCP_CONG_NEEDS_ECN,
 	.name		= "bpf_dctcp_nouse",
 };
 
 SEC(".struct_ops")
 struct tcp_congestion_ops dctcp = {
-	.init		= (void *)dctcp_init,
-	.in_ack_event   = (void *)dctcp_update_alpha,
-	.cwnd_event	= (void *)dctcp_cwnd_event,
-	.ssthresh	= (void *)dctcp_ssthresh,
-	.cong_avoid	= (void *)dctcp_cong_avoid,
-	.undo_cwnd	= (void *)dctcp_cwnd_undo,
-	.set_state	= (void *)dctcp_state,
+	.init		= (void *)bpf_dctcp_init,
+	.in_ack_event   = (void *)bpf_dctcp_update_alpha,
+	.cwnd_event	= (void *)bpf_dctcp_cwnd_event,
+	.ssthresh	= (void *)bpf_dctcp_ssthresh,
+	.cong_avoid	= (void *)bpf_dctcp_cong_avoid,
+	.undo_cwnd	= (void *)bpf_dctcp_cwnd_undo,
+	.set_state	= (void *)bpf_dctcp_state,
 	.flags		= TCP_CONG_NEEDS_ECN,
 	.name		= "bpf_dctcp",
 };
-- 
cgit v1.2.3-58-ga151


From ec209ad86324de84ef66990f0e9df0851e45e054 Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:32 -0600
Subject: bpf: verifier: Relax caller requirements for kfunc projection type
 args

Currently, if a kfunc accepts a projection type as an argument (eg
struct __sk_buff *), the caller must exactly provide exactly the same
type with provable provenance.

However in practice, kfuncs that accept projection types _must_ cast to
the underlying type before use b/c projection type layouts are
completely made up. Thus, it is ok to relax the verifier rules around
implicit conversions.

We will use this functionality in the next commit when we align kfuncs
to user-facing types.

Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/e2c025cb09ccfd4af1ec9e18284dc3cecff7514d.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/btf.h   |  1 +
 kernel/bpf/btf.c      | 13 ++++++++++---
 kernel/bpf/verifier.c | 10 +++++++++-
 3 files changed, 20 insertions(+), 4 deletions(-)

diff --git a/include/linux/btf.h b/include/linux/btf.h
index f9e56fd12a9f..56d91daacdba 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -531,6 +531,7 @@ s32 btf_find_dtor_kfunc(struct btf *btf, u32 btf_id);
 int register_btf_id_dtor_kfuncs(const struct btf_id_dtor_kfunc *dtors, u32 add_cnt,
 				struct module *owner);
 struct btf_struct_meta *btf_find_struct_meta(const struct btf *btf, u32 btf_id);
+bool btf_is_projection_of(const char *pname, const char *tname);
 bool btf_is_prog_ctx_type(struct bpf_verifier_log *log, const struct btf *btf,
 			   const struct btf_type *t, enum bpf_prog_type prog_type,
 			   int arg);
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 7928d920056f..ce4707968217 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -5820,6 +5820,15 @@ static int find_kern_ctx_type_id(enum bpf_prog_type prog_type)
 	return ctx_type->type;
 }
 
+bool btf_is_projection_of(const char *pname, const char *tname)
+{
+	if (strcmp(pname, "__sk_buff") == 0 && strcmp(tname, "sk_buff") == 0)
+		return true;
+	if (strcmp(pname, "xdp_md") == 0 && strcmp(tname, "xdp_buff") == 0)
+		return true;
+	return false;
+}
+
 bool btf_is_prog_ctx_type(struct bpf_verifier_log *log, const struct btf *btf,
 			  const struct btf_type *t, enum bpf_prog_type prog_type,
 			  int arg)
@@ -5882,9 +5891,7 @@ again:
 	 * int socket_filter_bpf_prog(struct __sk_buff *skb)
 	 * { // no fields of skb are ever used }
 	 */
-	if (strcmp(ctx_tname, "__sk_buff") == 0 && strcmp(tname, "sk_buff") == 0)
-		return true;
-	if (strcmp(ctx_tname, "xdp_md") == 0 && strcmp(tname, "xdp_buff") == 0)
+	if (btf_is_projection_of(ctx_tname, tname))
 		return true;
 	if (strcmp(ctx_tname, tname)) {
 		/* bpf_user_pt_regs_t is a typedef, so resolve it to
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 20ac9cfd54dd..dcac6119d810 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -11265,6 +11265,8 @@ static int process_kf_arg_ptr_to_btf_id(struct bpf_verifier_env *env,
 	bool strict_type_match = false;
 	const struct btf *reg_btf;
 	const char *reg_ref_tname;
+	bool taking_projection;
+	bool struct_same;
 	u32 reg_ref_id;
 
 	if (base_type(reg->type) == PTR_TO_BTF_ID) {
@@ -11308,7 +11310,13 @@ static int process_kf_arg_ptr_to_btf_id(struct bpf_verifier_env *env,
 
 	reg_ref_t = btf_type_skip_modifiers(reg_btf, reg_ref_id, &reg_ref_id);
 	reg_ref_tname = btf_name_by_offset(reg_btf, reg_ref_t->name_off);
-	if (!btf_struct_ids_match(&env->log, reg_btf, reg_ref_id, reg->off, meta->btf, ref_id, strict_type_match)) {
+	struct_same = btf_struct_ids_match(&env->log, reg_btf, reg_ref_id, reg->off, meta->btf, ref_id, strict_type_match);
+	/* If kfunc is accepting a projection type (ie. __sk_buff), it cannot
+	 * actually use it -- it must cast to the underlying type. So we allow
+	 * caller to pass in the underlying type.
+	 */
+	taking_projection = btf_is_projection_of(ref_tname, reg_ref_tname);
+	if (!taking_projection && !struct_same) {
 		verbose(env, "kernel function %s args#%d expected pointer to %s %s but R%d has a pointer to %s %s\n",
 			meta->func_name, argno, btf_type_str(ref_t), ref_tname, argno + 1,
 			btf_type_str(reg_ref_t), reg_ref_tname);
-- 
cgit v1.2.3-58-ga151


From cce4c40b960673f9e020835def310f1e89d3a940 Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:33 -0600
Subject: bpf: treewide: Align kfunc signatures to prog point-of-view

Previously, kfunc declarations in bpf_kfuncs.h (and others) used "user
facing" types for kfuncs prototypes while the actual kfunc definitions
used "kernel facing" types. More specifically: bpf_dynptr vs
bpf_dynptr_kern, __sk_buff vs sk_buff, and xdp_md vs xdp_buff.

It wasn't an issue before, as the verifier allows aliased types.
However, since we are now generating kfunc prototypes in vmlinux.h (in
addition to keeping bpf_kfuncs.h around), this conflict creates
compilation errors.

Fix this conflict by using "user facing" types in kfunc definitions.
This results in more casts, but otherwise has no additional runtime
cost.

Note, similar to 5b268d1ebcdc ("bpf: Have bpf_rdonly_cast() take a const
pointer"), we also make kfuncs take const arguments where appropriate in
order to make the kfunc more permissive.

Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/b58346a63a0e66bc9b7504da751b526b0b189a67.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 fs/verity/measure.c                                |  5 +--
 include/linux/bpf.h                                |  8 ++---
 kernel/bpf/crypto.c                                | 24 ++++++++-----
 kernel/bpf/helpers.c                               | 39 +++++++++++++++-------
 kernel/bpf/verifier.c                              |  2 +-
 kernel/trace/bpf_trace.c                           | 15 +++++----
 net/core/filter.c                                  | 32 +++++++++++-------
 .../testing/selftests/bpf/progs/ip_check_defrag.c  | 10 +++---
 .../selftests/bpf/progs/verifier_netfilter_ctx.c   |  6 ++--
 9 files changed, 88 insertions(+), 53 deletions(-)

diff --git a/fs/verity/measure.c b/fs/verity/measure.c
index 3969d54158d1..175d2f1bc089 100644
--- a/fs/verity/measure.c
+++ b/fs/verity/measure.c
@@ -111,14 +111,15 @@ __bpf_kfunc_start_defs();
 /**
  * bpf_get_fsverity_digest: read fsverity digest of file
  * @file: file to get digest from
- * @digest_ptr: (out) dynptr for struct fsverity_digest
+ * @digest_p: (out) dynptr for struct fsverity_digest
  *
  * Read fsverity_digest of *file* into *digest_ptr*.
  *
  * Return: 0 on success, a negative value on error.
  */
-__bpf_kfunc int bpf_get_fsverity_digest(struct file *file, struct bpf_dynptr_kern *digest_ptr)
+__bpf_kfunc int bpf_get_fsverity_digest(struct file *file, struct bpf_dynptr *digest_p)
 {
+	struct bpf_dynptr_kern *digest_ptr = (struct bpf_dynptr_kern *)digest_p;
 	const struct inode *inode = file_inode(file);
 	u32 dynptr_sz = __bpf_dynptr_size(digest_ptr);
 	struct fsverity_digest *arg;
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index a834f4b761bc..f636b4998bf7 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -3265,8 +3265,8 @@ u32 bpf_sock_convert_ctx_access(enum bpf_access_type type,
 				struct bpf_insn *insn_buf,
 				struct bpf_prog *prog,
 				u32 *target_size);
-int bpf_dynptr_from_skb_rdonly(struct sk_buff *skb, u64 flags,
-			       struct bpf_dynptr_kern *ptr);
+int bpf_dynptr_from_skb_rdonly(struct __sk_buff *skb, u64 flags,
+			       struct bpf_dynptr *ptr);
 #else
 static inline bool bpf_sock_common_is_valid_access(int off, int size,
 						   enum bpf_access_type type,
@@ -3288,8 +3288,8 @@ static inline u32 bpf_sock_convert_ctx_access(enum bpf_access_type type,
 {
 	return 0;
 }
-static inline int bpf_dynptr_from_skb_rdonly(struct sk_buff *skb, u64 flags,
-					     struct bpf_dynptr_kern *ptr)
+static inline int bpf_dynptr_from_skb_rdonly(struct __sk_buff *skb, u64 flags,
+					     struct bpf_dynptr *ptr)
 {
 	return -EOPNOTSUPP;
 }
diff --git a/kernel/bpf/crypto.c b/kernel/bpf/crypto.c
index 2bee4af91e38..3c1de0e5c0bd 100644
--- a/kernel/bpf/crypto.c
+++ b/kernel/bpf/crypto.c
@@ -311,11 +311,15 @@ static int bpf_crypto_crypt(const struct bpf_crypto_ctx *ctx,
  * Decrypts provided buffer using IV data and the crypto context. Crypto context must be configured.
  */
 __bpf_kfunc int bpf_crypto_decrypt(struct bpf_crypto_ctx *ctx,
-				   const struct bpf_dynptr_kern *src,
-				   const struct bpf_dynptr_kern *dst,
-				   const struct bpf_dynptr_kern *siv)
+				   const struct bpf_dynptr *src,
+				   const struct bpf_dynptr *dst,
+				   const struct bpf_dynptr *siv)
 {
-	return bpf_crypto_crypt(ctx, src, dst, siv, true);
+	const struct bpf_dynptr_kern *src_kern = (struct bpf_dynptr_kern *)src;
+	const struct bpf_dynptr_kern *dst_kern = (struct bpf_dynptr_kern *)dst;
+	const struct bpf_dynptr_kern *siv_kern = (struct bpf_dynptr_kern *)siv;
+
+	return bpf_crypto_crypt(ctx, src_kern, dst_kern, siv_kern, true);
 }
 
 /**
@@ -328,11 +332,15 @@ __bpf_kfunc int bpf_crypto_decrypt(struct bpf_crypto_ctx *ctx,
  * Encrypts provided buffer using IV data and the crypto context. Crypto context must be configured.
  */
 __bpf_kfunc int bpf_crypto_encrypt(struct bpf_crypto_ctx *ctx,
-				   const struct bpf_dynptr_kern *src,
-				   const struct bpf_dynptr_kern *dst,
-				   const struct bpf_dynptr_kern *siv)
+				   const struct bpf_dynptr *src,
+				   const struct bpf_dynptr *dst,
+				   const struct bpf_dynptr *siv)
 {
-	return bpf_crypto_crypt(ctx, src, dst, siv, false);
+	const struct bpf_dynptr_kern *src_kern = (struct bpf_dynptr_kern *)src;
+	const struct bpf_dynptr_kern *dst_kern = (struct bpf_dynptr_kern *)dst;
+	const struct bpf_dynptr_kern *siv_kern = (struct bpf_dynptr_kern *)siv;
+
+	return bpf_crypto_crypt(ctx, src_kern, dst_kern, siv_kern, false);
 }
 
 __bpf_kfunc_end_defs();
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 6f1abcb4b084..3ac521c48bba 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2459,9 +2459,10 @@ __bpf_kfunc struct task_struct *bpf_task_from_pid(s32 pid)
  * provided buffer, with its contents containing the data, if unable to obtain
  * direct pointer)
  */
-__bpf_kfunc void *bpf_dynptr_slice(const struct bpf_dynptr_kern *ptr, u32 offset,
+__bpf_kfunc void *bpf_dynptr_slice(const struct bpf_dynptr *p, u32 offset,
 				   void *buffer__opt, u32 buffer__szk)
 {
+	const struct bpf_dynptr_kern *ptr = (struct bpf_dynptr_kern *)p;
 	enum bpf_dynptr_type type;
 	u32 len = buffer__szk;
 	int err;
@@ -2543,9 +2544,11 @@ __bpf_kfunc void *bpf_dynptr_slice(const struct bpf_dynptr_kern *ptr, u32 offset
  * provided buffer, with its contents containing the data, if unable to obtain
  * direct pointer)
  */
-__bpf_kfunc void *bpf_dynptr_slice_rdwr(const struct bpf_dynptr_kern *ptr, u32 offset,
+__bpf_kfunc void *bpf_dynptr_slice_rdwr(const struct bpf_dynptr *p, u32 offset,
 					void *buffer__opt, u32 buffer__szk)
 {
+	const struct bpf_dynptr_kern *ptr = (struct bpf_dynptr_kern *)p;
+
 	if (!ptr->data || __bpf_dynptr_is_rdonly(ptr))
 		return NULL;
 
@@ -2571,11 +2574,12 @@ __bpf_kfunc void *bpf_dynptr_slice_rdwr(const struct bpf_dynptr_kern *ptr, u32 o
 	 * will be copied out into the buffer and the user will need to call
 	 * bpf_dynptr_write() to commit changes.
 	 */
-	return bpf_dynptr_slice(ptr, offset, buffer__opt, buffer__szk);
+	return bpf_dynptr_slice(p, offset, buffer__opt, buffer__szk);
 }
 
-__bpf_kfunc int bpf_dynptr_adjust(struct bpf_dynptr_kern *ptr, u32 start, u32 end)
+__bpf_kfunc int bpf_dynptr_adjust(const struct bpf_dynptr *p, u32 start, u32 end)
 {
+	struct bpf_dynptr_kern *ptr = (struct bpf_dynptr_kern *)p;
 	u32 size;
 
 	if (!ptr->data || start > end)
@@ -2592,36 +2596,45 @@ __bpf_kfunc int bpf_dynptr_adjust(struct bpf_dynptr_kern *ptr, u32 start, u32 en
 	return 0;
 }
 
-__bpf_kfunc bool bpf_dynptr_is_null(struct bpf_dynptr_kern *ptr)
+__bpf_kfunc bool bpf_dynptr_is_null(const struct bpf_dynptr *p)
 {
+	struct bpf_dynptr_kern *ptr = (struct bpf_dynptr_kern *)p;
+
 	return !ptr->data;
 }
 
-__bpf_kfunc bool bpf_dynptr_is_rdonly(struct bpf_dynptr_kern *ptr)
+__bpf_kfunc bool bpf_dynptr_is_rdonly(const struct bpf_dynptr *p)
 {
+	struct bpf_dynptr_kern *ptr = (struct bpf_dynptr_kern *)p;
+
 	if (!ptr->data)
 		return false;
 
 	return __bpf_dynptr_is_rdonly(ptr);
 }
 
-__bpf_kfunc __u32 bpf_dynptr_size(const struct bpf_dynptr_kern *ptr)
+__bpf_kfunc __u32 bpf_dynptr_size(const struct bpf_dynptr *p)
 {
+	struct bpf_dynptr_kern *ptr = (struct bpf_dynptr_kern *)p;
+
 	if (!ptr->data)
 		return -EINVAL;
 
 	return __bpf_dynptr_size(ptr);
 }
 
-__bpf_kfunc int bpf_dynptr_clone(struct bpf_dynptr_kern *ptr,
-				 struct bpf_dynptr_kern *clone__uninit)
+__bpf_kfunc int bpf_dynptr_clone(const struct bpf_dynptr *p,
+				 struct bpf_dynptr *clone__uninit)
 {
+	struct bpf_dynptr_kern *clone = (struct bpf_dynptr_kern *)clone__uninit;
+	struct bpf_dynptr_kern *ptr = (struct bpf_dynptr_kern *)p;
+
 	if (!ptr->data) {
-		bpf_dynptr_set_null(clone__uninit);
+		bpf_dynptr_set_null(clone);
 		return -EINVAL;
 	}
 
-	*clone__uninit = *ptr;
+	*clone = *ptr;
 
 	return 0;
 }
@@ -2986,7 +2999,9 @@ late_initcall(kfunc_init);
  */
 const void *__bpf_dynptr_data(const struct bpf_dynptr_kern *ptr, u32 len)
 {
-	return bpf_dynptr_slice(ptr, 0, NULL, len);
+	const struct bpf_dynptr *p = (struct bpf_dynptr *)ptr;
+
+	return bpf_dynptr_slice(p, 0, NULL, len);
 }
 
 /* Get a pointer to dynptr data up to len bytes for read write access. If
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index dcac6119d810..acc9dd830807 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -10914,7 +10914,7 @@ enum {
 };
 
 BTF_ID_LIST(kf_arg_btf_ids)
-BTF_ID(struct, bpf_dynptr_kern)
+BTF_ID(struct, bpf_dynptr)
 BTF_ID(struct, bpf_list_head)
 BTF_ID(struct, bpf_list_node)
 BTF_ID(struct, bpf_rb_root)
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index bc16e21a2a44..4b3fda456299 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -1369,8 +1369,8 @@ __bpf_kfunc void bpf_key_put(struct bpf_key *bkey)
 #ifdef CONFIG_SYSTEM_DATA_VERIFICATION
 /**
  * bpf_verify_pkcs7_signature - verify a PKCS#7 signature
- * @data_ptr: data to verify
- * @sig_ptr: signature of the data
+ * @data_p: data to verify
+ * @sig_p: signature of the data
  * @trusted_keyring: keyring with keys trusted for signature verification
  *
  * Verify the PKCS#7 signature *sig_ptr* against the supplied *data_ptr*
@@ -1378,10 +1378,12 @@ __bpf_kfunc void bpf_key_put(struct bpf_key *bkey)
  *
  * Return: 0 on success, a negative value on error.
  */
-__bpf_kfunc int bpf_verify_pkcs7_signature(struct bpf_dynptr_kern *data_ptr,
-			       struct bpf_dynptr_kern *sig_ptr,
+__bpf_kfunc int bpf_verify_pkcs7_signature(struct bpf_dynptr *data_p,
+			       struct bpf_dynptr *sig_p,
 			       struct bpf_key *trusted_keyring)
 {
+	struct bpf_dynptr_kern *data_ptr = (struct bpf_dynptr_kern *)data_p;
+	struct bpf_dynptr_kern *sig_ptr = (struct bpf_dynptr_kern *)sig_p;
 	const void *data, *sig;
 	u32 data_len, sig_len;
 	int ret;
@@ -1444,7 +1446,7 @@ __bpf_kfunc_start_defs();
  * bpf_get_file_xattr - get xattr of a file
  * @file: file to get xattr from
  * @name__str: name of the xattr
- * @value_ptr: output buffer of the xattr value
+ * @value_p: output buffer of the xattr value
  *
  * Get xattr *name__str* of *file* and store the output in *value_ptr*.
  *
@@ -1453,8 +1455,9 @@ __bpf_kfunc_start_defs();
  * Return: 0 on success, a negative value on error.
  */
 __bpf_kfunc int bpf_get_file_xattr(struct file *file, const char *name__str,
-				   struct bpf_dynptr_kern *value_ptr)
+				   struct bpf_dynptr *value_p)
 {
+	struct bpf_dynptr_kern *value_ptr = (struct bpf_dynptr_kern *)value_p;
 	struct dentry *dentry;
 	u32 value_len;
 	void *value;
diff --git a/net/core/filter.c b/net/core/filter.c
index 7c46ecba3b01..73722790cee3 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -11859,28 +11859,34 @@ bpf_sk_base_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 }
 
 __bpf_kfunc_start_defs();
-__bpf_kfunc int bpf_dynptr_from_skb(struct sk_buff *skb, u64 flags,
-				    struct bpf_dynptr_kern *ptr__uninit)
+__bpf_kfunc int bpf_dynptr_from_skb(struct __sk_buff *s, u64 flags,
+				    struct bpf_dynptr *ptr__uninit)
 {
+	struct bpf_dynptr_kern *ptr = (struct bpf_dynptr_kern *)ptr__uninit;
+	struct sk_buff *skb = (struct sk_buff *)s;
+
 	if (flags) {
-		bpf_dynptr_set_null(ptr__uninit);
+		bpf_dynptr_set_null(ptr);
 		return -EINVAL;
 	}
 
-	bpf_dynptr_init(ptr__uninit, skb, BPF_DYNPTR_TYPE_SKB, 0, skb->len);
+	bpf_dynptr_init(ptr, skb, BPF_DYNPTR_TYPE_SKB, 0, skb->len);
 
 	return 0;
 }
 
-__bpf_kfunc int bpf_dynptr_from_xdp(struct xdp_buff *xdp, u64 flags,
-				    struct bpf_dynptr_kern *ptr__uninit)
+__bpf_kfunc int bpf_dynptr_from_xdp(struct xdp_md *x, u64 flags,
+				    struct bpf_dynptr *ptr__uninit)
 {
+	struct bpf_dynptr_kern *ptr = (struct bpf_dynptr_kern *)ptr__uninit;
+	struct xdp_buff *xdp = (struct xdp_buff *)x;
+
 	if (flags) {
-		bpf_dynptr_set_null(ptr__uninit);
+		bpf_dynptr_set_null(ptr);
 		return -EINVAL;
 	}
 
-	bpf_dynptr_init(ptr__uninit, xdp, BPF_DYNPTR_TYPE_XDP, 0, xdp_get_buff_len(xdp));
+	bpf_dynptr_init(ptr, xdp, BPF_DYNPTR_TYPE_XDP, 0, xdp_get_buff_len(xdp));
 
 	return 0;
 }
@@ -11906,10 +11912,11 @@ __bpf_kfunc int bpf_sock_addr_set_sun_path(struct bpf_sock_addr_kern *sa_kern,
 	return 0;
 }
 
-__bpf_kfunc int bpf_sk_assign_tcp_reqsk(struct sk_buff *skb, struct sock *sk,
+__bpf_kfunc int bpf_sk_assign_tcp_reqsk(struct __sk_buff *s, struct sock *sk,
 					struct bpf_tcp_req_attrs *attrs, int attrs__sz)
 {
 #if IS_ENABLED(CONFIG_SYN_COOKIES)
+	struct sk_buff *skb = (struct sk_buff *)s;
 	const struct request_sock_ops *ops;
 	struct inet_request_sock *ireq;
 	struct tcp_request_sock *treq;
@@ -12004,16 +12011,17 @@ __bpf_kfunc int bpf_sk_assign_tcp_reqsk(struct sk_buff *skb, struct sock *sk,
 
 __bpf_kfunc_end_defs();
 
-int bpf_dynptr_from_skb_rdonly(struct sk_buff *skb, u64 flags,
-			       struct bpf_dynptr_kern *ptr__uninit)
+int bpf_dynptr_from_skb_rdonly(struct __sk_buff *skb, u64 flags,
+			       struct bpf_dynptr *ptr__uninit)
 {
+	struct bpf_dynptr_kern *ptr = (struct bpf_dynptr_kern *)ptr__uninit;
 	int err;
 
 	err = bpf_dynptr_from_skb(skb, flags, ptr__uninit);
 	if (err)
 		return err;
 
-	bpf_dynptr_set_rdonly(ptr__uninit);
+	bpf_dynptr_set_rdonly(ptr);
 
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/ip_check_defrag.c b/tools/testing/selftests/bpf/progs/ip_check_defrag.c
index 1c2b6c1616b0..645b2c9f7867 100644
--- a/tools/testing/selftests/bpf/progs/ip_check_defrag.c
+++ b/tools/testing/selftests/bpf/progs/ip_check_defrag.c
@@ -12,7 +12,7 @@
 #define IP_OFFSET		0x1FFF
 #define NEXTHDR_FRAGMENT	44
 
-extern int bpf_dynptr_from_skb(struct sk_buff *skb, __u64 flags,
+extern int bpf_dynptr_from_skb(struct __sk_buff *skb, __u64 flags,
 			      struct bpf_dynptr *ptr__uninit) __ksym;
 extern void *bpf_dynptr_slice(const struct bpf_dynptr *ptr, uint32_t offset,
 			      void *buffer, uint32_t buffer__sz) __ksym;
@@ -42,7 +42,7 @@ static bool is_frag_v6(struct ipv6hdr *ip6h)
 	return ip6h->nexthdr == NEXTHDR_FRAGMENT;
 }
 
-static int handle_v4(struct sk_buff *skb)
+static int handle_v4(struct __sk_buff *skb)
 {
 	struct bpf_dynptr ptr;
 	u8 iph_buf[20] = {};
@@ -64,7 +64,7 @@ static int handle_v4(struct sk_buff *skb)
 	return NF_ACCEPT;
 }
 
-static int handle_v6(struct sk_buff *skb)
+static int handle_v6(struct __sk_buff *skb)
 {
 	struct bpf_dynptr ptr;
 	struct ipv6hdr *ip6h;
@@ -89,9 +89,9 @@ static int handle_v6(struct sk_buff *skb)
 SEC("netfilter")
 int defrag(struct bpf_nf_ctx *ctx)
 {
-	struct sk_buff *skb = ctx->skb;
+	struct __sk_buff *skb = (struct __sk_buff *)ctx->skb;
 
-	switch (bpf_ntohs(skb->protocol)) {
+	switch (bpf_ntohs(ctx->skb->protocol)) {
 	case ETH_P_IP:
 		return handle_v4(skb);
 	case ETH_P_IPV6:
diff --git a/tools/testing/selftests/bpf/progs/verifier_netfilter_ctx.c b/tools/testing/selftests/bpf/progs/verifier_netfilter_ctx.c
index 65bba330e7e5..ab9f9f2620ed 100644
--- a/tools/testing/selftests/bpf/progs/verifier_netfilter_ctx.c
+++ b/tools/testing/selftests/bpf/progs/verifier_netfilter_ctx.c
@@ -79,7 +79,7 @@ int with_invalid_ctx_access_test5(struct bpf_nf_ctx *ctx)
 	return NF_ACCEPT;
 }
 
-extern int bpf_dynptr_from_skb(struct sk_buff *skb, __u64 flags,
+extern int bpf_dynptr_from_skb(struct __sk_buff *skb, __u64 flags,
                                struct bpf_dynptr *ptr__uninit) __ksym;
 extern void *bpf_dynptr_slice(const struct bpf_dynptr *ptr, uint32_t offset,
                                    void *buffer, uint32_t buffer__sz) __ksym;
@@ -90,8 +90,8 @@ __success __failure_unpriv
 __retval(0)
 int with_valid_ctx_access_test6(struct bpf_nf_ctx *ctx)
 {
+	struct __sk_buff *skb = (struct __sk_buff *)ctx->skb;
 	const struct nf_hook_state *state = ctx->state;
-	struct sk_buff *skb = ctx->skb;
 	const struct iphdr *iph;
 	const struct tcphdr *th;
 	u8 buffer_iph[20] = {};
@@ -99,7 +99,7 @@ int with_valid_ctx_access_test6(struct bpf_nf_ctx *ctx)
 	struct bpf_dynptr ptr;
 	uint8_t ihl;
 
-	if (skb->len <= 20 || bpf_dynptr_from_skb(skb, 0, &ptr))
+	if (ctx->skb->len <= 20 || bpf_dynptr_from_skb(skb, 0, &ptr))
 		return NF_ACCEPT;
 
 	iph = bpf_dynptr_slice(&ptr, 0, buffer_iph, sizeof(buffer_iph));
-- 
cgit v1.2.3-58-ga151


From f709124dd72fe7a3f6ba7764b2ed145c55c33e47 Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:34 -0600
Subject: bpf: selftests: nf: Opt out of using generated kfunc prototypes

The bpf-nf selftests play various games with aliased types such that
folks with CONFIG_NF_CONNTRACK=m/n configs can still build the
selftests. See commits:

1058b6a78db2 ("selftests/bpf: Do not fail build if CONFIG_NF_CONNTRACK=m/n")
92afc5329a5b ("selftests/bpf: Fix build errors if CONFIG_NF_CONNTRACK=m")

Thus, it is simpler if these selftests opt out of using generated kfunc
prototypes. The preprocessor macro this commit uses will be introduced
in the final commit.

Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/044a5b10cb3abd0d71cb1c818ee0bfc4a2239332.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/test_bpf_nf.c       | 1 +
 tools/testing/selftests/bpf/progs/test_bpf_nf_fail.c  | 1 +
 tools/testing/selftests/bpf/progs/xdp_synproxy_kern.c | 1 +
 3 files changed, 3 insertions(+)

diff --git a/tools/testing/selftests/bpf/progs/test_bpf_nf.c b/tools/testing/selftests/bpf/progs/test_bpf_nf.c
index 0289d8ce2b80..f7b330ddd007 100644
--- a/tools/testing/selftests/bpf/progs/test_bpf_nf.c
+++ b/tools/testing/selftests/bpf/progs/test_bpf_nf.c
@@ -1,4 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
+#define BPF_NO_KFUNC_PROTOTYPES
 #include <vmlinux.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_endian.h>
diff --git a/tools/testing/selftests/bpf/progs/test_bpf_nf_fail.c b/tools/testing/selftests/bpf/progs/test_bpf_nf_fail.c
index 0e4759ab38ff..a586f087ffeb 100644
--- a/tools/testing/selftests/bpf/progs/test_bpf_nf_fail.c
+++ b/tools/testing/selftests/bpf/progs/test_bpf_nf_fail.c
@@ -1,4 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
+#define BPF_NO_KFUNC_PROTOTYPES
 #include <vmlinux.h>
 #include <bpf/bpf_tracing.h>
 #include <bpf/bpf_helpers.h>
diff --git a/tools/testing/selftests/bpf/progs/xdp_synproxy_kern.c b/tools/testing/selftests/bpf/progs/xdp_synproxy_kern.c
index 7ea9785738b5..f8f5dc9f72b8 100644
--- a/tools/testing/selftests/bpf/progs/xdp_synproxy_kern.c
+++ b/tools/testing/selftests/bpf/progs/xdp_synproxy_kern.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: LGPL-2.1 OR BSD-2-Clause
 /* Copyright (c) 2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved. */
 
+#define BPF_NO_KFUNC_PROTOTYPES
 #include "vmlinux.h"
 
 #include <bpf/bpf_helpers.h>
-- 
cgit v1.2.3-58-ga151


From c567cba34585514f82600a10587c8813c50e3a7c Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:35 -0600
Subject: bpf: selftests: xfrm: Opt out of using generated kfunc prototypes

The xfrm_info selftest locally defines an aliased type such that folks
with CONFIG_XFRM_INTERFACE=m/n configs can still build the selftests.
See commit aa67961f3243 ("selftests/bpf: Allow building bpf tests with CONFIG_XFRM_INTERFACE=[m|n]").

Thus, it is simpler if this selftest opts out of using enerated kfunc
prototypes. The preprocessor macro this commit uses will be introduced
in the final commit.

Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/afe0bb1c50487f52542cdd5230c4aef9e36ce250.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/xfrm_info.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/testing/selftests/bpf/progs/xfrm_info.c b/tools/testing/selftests/bpf/progs/xfrm_info.c
index f6a501fbba2b..a1d9f106c3f0 100644
--- a/tools/testing/selftests/bpf/progs/xfrm_info.c
+++ b/tools/testing/selftests/bpf/progs/xfrm_info.c
@@ -1,4 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
+#define BPF_NO_KFUNC_PROTOTYPES
 #include "vmlinux.h"
 #include "bpf_tracing_net.h"
 #include <bpf/bpf_helpers.h>
-- 
cgit v1.2.3-58-ga151


From 770abbb5a25a5b767f1c60ba366aea503728e957 Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Wed, 12 Jun 2024 09:58:36 -0600
Subject: bpftool: Support dumping kfunc prototypes from BTF

This patch enables dumping kfunc prototypes from bpftool. This is useful
b/c with this patch, end users will no longer have to manually define
kfunc prototypes. For the kernel tree, this also means we can optionally
drop kfunc prototypes from:

        tools/testing/selftests/bpf/bpf_kfuncs.h
        tools/testing/selftests/bpf/bpf_experimental.h

Example usage:

        $ make PAHOLE=/home/dxu/dev/pahole/build/pahole -j30 vmlinux

        $ ./tools/bpf/bpftool/bpftool btf dump file ./vmlinux format c | rg "__ksym;" | head -3
        extern void cgroup_rstat_updated(struct cgroup *cgrp, int cpu) __weak __ksym;
        extern void cgroup_rstat_flush(struct cgroup *cgrp) __weak __ksym;
        extern struct bpf_key *bpf_lookup_user_key(u32 serial, u64 flags) __weak __ksym;

Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/bf6c08f9263c4bd9d10a717de95199d766a13f61.1718207789.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/btf.c | 55 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 55 insertions(+)

diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c
index af047dedde38..6789c7a4d5ca 100644
--- a/tools/bpf/bpftool/btf.c
+++ b/tools/bpf/bpftool/btf.c
@@ -20,6 +20,8 @@
 #include "json_writer.h"
 #include "main.h"
 
+#define KFUNC_DECL_TAG		"bpf_kfunc"
+
 static const char * const btf_kind_str[NR_BTF_KINDS] = {
 	[BTF_KIND_UNKN]		= "UNKNOWN",
 	[BTF_KIND_INT]		= "INT",
@@ -461,6 +463,49 @@ static int dump_btf_raw(const struct btf *btf,
 	return 0;
 }
 
+static int dump_btf_kfuncs(struct btf_dump *d, const struct btf *btf)
+{
+	LIBBPF_OPTS(btf_dump_emit_type_decl_opts, opts);
+	int cnt = btf__type_cnt(btf);
+	int i;
+
+	printf("\n/* BPF kfuncs */\n");
+	printf("#ifndef BPF_NO_KFUNC_PROTOTYPES\n");
+
+	for (i = 1; i < cnt; i++) {
+		const struct btf_type *t = btf__type_by_id(btf, i);
+		const char *name;
+		int err;
+
+		if (!btf_is_decl_tag(t))
+			continue;
+
+		if (btf_decl_tag(t)->component_idx != -1)
+			continue;
+
+		name = btf__name_by_offset(btf, t->name_off);
+		if (strncmp(name, KFUNC_DECL_TAG, sizeof(KFUNC_DECL_TAG)))
+			continue;
+
+		t = btf__type_by_id(btf, t->type);
+		if (!btf_is_func(t))
+			continue;
+
+		printf("extern ");
+
+		opts.field_name = btf__name_by_offset(btf, t->name_off);
+		err = btf_dump__emit_type_decl(d, t->type, &opts);
+		if (err)
+			return err;
+
+		printf(" __weak __ksym;\n");
+	}
+
+	printf("#endif\n\n");
+
+	return 0;
+}
+
 static void __printf(2, 0) btf_dump_printf(void *ctx,
 					   const char *fmt, va_list args)
 {
@@ -596,6 +641,12 @@ static int dump_btf_c(const struct btf *btf,
 	printf("#ifndef BPF_NO_PRESERVE_ACCESS_INDEX\n");
 	printf("#pragma clang attribute push (__attribute__((preserve_access_index)), apply_to = record)\n");
 	printf("#endif\n\n");
+	printf("#ifndef __ksym\n");
+	printf("#define __ksym __attribute__((section(\".ksyms\")))\n");
+	printf("#endif\n\n");
+	printf("#ifndef __weak\n");
+	printf("#define __weak __attribute__((weak))\n");
+	printf("#endif\n\n");
 
 	if (root_type_cnt) {
 		for (i = 0; i < root_type_cnt; i++) {
@@ -615,6 +666,10 @@ static int dump_btf_c(const struct btf *btf,
 			if (err)
 				goto done;
 		}
+
+		err = dump_btf_kfuncs(d, btf);
+		if (err)
+			goto done;
 	}
 
 	printf("#ifndef BPF_NO_PRESERVE_ACCESS_INDEX\n");
-- 
cgit v1.2.3-58-ga151


From 7fc45cb68696c7213c484ec81892bc8a986fde52 Mon Sep 17 00:00:00 2001
From: Shradha Gupta <shradhagupta@linux.microsoft.com>
Date: Mon, 10 Jun 2024 03:28:39 -0700
Subject: net: mana: Allow variable size indirection table

Allow variable size indirection table allocation in MANA instead
of using a constant value MANA_INDIRECT_TABLE_SIZE.
The size is now derived from the MANA_QUERY_VPORT_CONFIG and the
indirection table is allocated dynamically.

Signed-off-by: Shradha Gupta <shradhagupta@linux.microsoft.com>
Link: https://lore.kernel.org/r/1718015319-9609-1-git-send-email-shradhagupta@linux.microsoft.com
Reviewed-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 drivers/infiniband/hw/mana/qp.c                    | 10 +--
 drivers/net/ethernet/microsoft/mana/mana_en.c      | 85 ++++++++++++++++++----
 drivers/net/ethernet/microsoft/mana/mana_ethtool.c | 27 +++++--
 include/net/mana/gdma.h                            |  4 +-
 include/net/mana/mana.h                            |  9 ++-
 5 files changed, 104 insertions(+), 31 deletions(-)

diff --git a/drivers/infiniband/hw/mana/qp.c b/drivers/infiniband/hw/mana/qp.c
index ba13c5abf8ef..2d411a16a127 100644
--- a/drivers/infiniband/hw/mana/qp.c
+++ b/drivers/infiniband/hw/mana/qp.c
@@ -21,7 +21,7 @@ static int mana_ib_cfg_vport_steering(struct mana_ib_dev *dev,
 
 	gc = mdev_to_gc(dev);
 
-	req_buf_size = struct_size(req, indir_tab, MANA_INDIRECT_TABLE_SIZE);
+	req_buf_size = struct_size(req, indir_tab, MANA_INDIRECT_TABLE_DEF_SIZE);
 	req = kzalloc(req_buf_size, GFP_KERNEL);
 	if (!req)
 		return -ENOMEM;
@@ -41,18 +41,18 @@ static int mana_ib_cfg_vport_steering(struct mana_ib_dev *dev,
 	if (log_ind_tbl_size)
 		req->rss_enable = true;
 
-	req->num_indir_entries = MANA_INDIRECT_TABLE_SIZE;
+	req->num_indir_entries = MANA_INDIRECT_TABLE_DEF_SIZE;
 	req->indir_tab_offset = offsetof(struct mana_cfg_rx_steer_req_v2,
 					 indir_tab);
 	req->update_indir_tab = true;
 	req->cqe_coalescing_enable = 1;
 
 	/* The ind table passed to the hardware must have
-	 * MANA_INDIRECT_TABLE_SIZE entries. Adjust the verb
+	 * MANA_INDIRECT_TABLE_DEF_SIZE entries. Adjust the verb
 	 * ind_table to MANA_INDIRECT_TABLE_SIZE if required
 	 */
 	ibdev_dbg(&dev->ib_dev, "ind table size %u\n", 1 << log_ind_tbl_size);
-	for (i = 0; i < MANA_INDIRECT_TABLE_SIZE; i++) {
+	for (i = 0; i < MANA_INDIRECT_TABLE_DEF_SIZE; i++) {
 		req->indir_tab[i] = ind_table[i % (1 << log_ind_tbl_size)];
 		ibdev_dbg(&dev->ib_dev, "index %u handle 0x%llx\n", i,
 			  req->indir_tab[i]);
@@ -137,7 +137,7 @@ static int mana_ib_create_qp_rss(struct ib_qp *ibqp, struct ib_pd *pd,
 	}
 
 	ind_tbl_size = 1 << ind_tbl->log_ind_tbl_size;
-	if (ind_tbl_size > MANA_INDIRECT_TABLE_SIZE) {
+	if (ind_tbl_size > MANA_INDIRECT_TABLE_DEF_SIZE) {
 		ibdev_dbg(&mdev->ib_dev,
 			  "Indirect table size %d exceeding limit\n",
 			  ind_tbl_size);
diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index d087cf954f75..b89ad4afd66e 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -481,7 +481,7 @@ static int mana_get_tx_queue(struct net_device *ndev, struct sk_buff *skb,
 	struct sock *sk = skb->sk;
 	int txq;
 
-	txq = apc->indir_table[hash & MANA_INDIRECT_TABLE_MASK];
+	txq = apc->indir_table[hash & (apc->indir_table_sz - 1)];
 
 	if (txq != old_q && sk && sk_fullsock(sk) &&
 	    rcu_access_pointer(sk->sk_dst_cache))
@@ -721,6 +721,13 @@ static void mana_cleanup_port_context(struct mana_port_context *apc)
 	apc->rxqs = NULL;
 }
 
+static void mana_cleanup_indir_table(struct mana_port_context *apc)
+{
+	apc->indir_table_sz = 0;
+	kfree(apc->indir_table);
+	kfree(apc->rxobj_table);
+}
+
 static int mana_init_port_context(struct mana_port_context *apc)
 {
 	apc->rxqs = kcalloc(apc->num_queues, sizeof(struct mana_rxq *),
@@ -962,7 +969,16 @@ static int mana_query_vport_cfg(struct mana_port_context *apc, u32 vport_index,
 
 	*max_sq = resp.max_num_sq;
 	*max_rq = resp.max_num_rq;
-	*num_indir_entry = resp.num_indirection_ent;
+	if (resp.num_indirection_ent > 0 &&
+	    resp.num_indirection_ent <= MANA_INDIRECT_TABLE_MAX_SIZE &&
+	    is_power_of_2(resp.num_indirection_ent)) {
+		*num_indir_entry = resp.num_indirection_ent;
+	} else {
+		netdev_warn(apc->ndev,
+			    "Setting indirection table size to default %d for vPort %d\n",
+			    MANA_INDIRECT_TABLE_DEF_SIZE, apc->port_idx);
+		*num_indir_entry = MANA_INDIRECT_TABLE_DEF_SIZE;
+	}
 
 	apc->port_handle = resp.vport;
 	ether_addr_copy(apc->mac_addr, resp.mac_addr);
@@ -1054,14 +1070,13 @@ static int mana_cfg_vport_steering(struct mana_port_context *apc,
 				   bool update_default_rxobj, bool update_key,
 				   bool update_tab)
 {
-	u16 num_entries = MANA_INDIRECT_TABLE_SIZE;
 	struct mana_cfg_rx_steer_req_v2 *req;
 	struct mana_cfg_rx_steer_resp resp = {};
 	struct net_device *ndev = apc->ndev;
 	u32 req_buf_size;
 	int err;
 
-	req_buf_size = struct_size(req, indir_tab, num_entries);
+	req_buf_size = struct_size(req, indir_tab, apc->indir_table_sz);
 	req = kzalloc(req_buf_size, GFP_KERNEL);
 	if (!req)
 		return -ENOMEM;
@@ -1072,7 +1087,7 @@ static int mana_cfg_vport_steering(struct mana_port_context *apc,
 	req->hdr.req.msg_version = GDMA_MESSAGE_V2;
 
 	req->vport = apc->port_handle;
-	req->num_indir_entries = num_entries;
+	req->num_indir_entries = apc->indir_table_sz;
 	req->indir_tab_offset = offsetof(struct mana_cfg_rx_steer_req_v2,
 					 indir_tab);
 	req->rx_enable = rx;
@@ -1111,7 +1126,7 @@ static int mana_cfg_vport_steering(struct mana_port_context *apc,
 	}
 
 	netdev_info(ndev, "Configured steering vPort %llu entries %u\n",
-		    apc->port_handle, num_entries);
+		    apc->port_handle, apc->indir_table_sz);
 out:
 	kfree(req);
 	return err;
@@ -2344,11 +2359,33 @@ static int mana_create_vport(struct mana_port_context *apc,
 	return mana_create_txq(apc, net);
 }
 
+static int mana_rss_table_alloc(struct mana_port_context *apc)
+{
+	if (!apc->indir_table_sz) {
+		netdev_err(apc->ndev,
+			   "Indirection table size not set for vPort %d\n",
+			   apc->port_idx);
+		return -EINVAL;
+	}
+
+	apc->indir_table = kcalloc(apc->indir_table_sz, sizeof(u32), GFP_KERNEL);
+	if (!apc->indir_table)
+		return -ENOMEM;
+
+	apc->rxobj_table = kcalloc(apc->indir_table_sz, sizeof(mana_handle_t), GFP_KERNEL);
+	if (!apc->rxobj_table) {
+		kfree(apc->indir_table);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
 static void mana_rss_table_init(struct mana_port_context *apc)
 {
 	int i;
 
-	for (i = 0; i < MANA_INDIRECT_TABLE_SIZE; i++)
+	for (i = 0; i < apc->indir_table_sz; i++)
 		apc->indir_table[i] =
 			ethtool_rxfh_indir_default(i, apc->num_queues);
 }
@@ -2361,7 +2398,7 @@ int mana_config_rss(struct mana_port_context *apc, enum TRI_STATE rx,
 	int i;
 
 	if (update_tab) {
-		for (i = 0; i < MANA_INDIRECT_TABLE_SIZE; i++) {
+		for (i = 0; i < apc->indir_table_sz; i++) {
 			queue_idx = apc->indir_table[i];
 			apc->rxobj_table[i] = apc->rxqs[queue_idx]->rxobj;
 		}
@@ -2466,7 +2503,6 @@ static int mana_init_port(struct net_device *ndev)
 	struct mana_port_context *apc = netdev_priv(ndev);
 	u32 max_txq, max_rxq, max_queues;
 	int port_idx = apc->port_idx;
-	u32 num_indirect_entries;
 	int err;
 
 	err = mana_init_port_context(apc);
@@ -2474,7 +2510,7 @@ static int mana_init_port(struct net_device *ndev)
 		return err;
 
 	err = mana_query_vport_cfg(apc, port_idx, &max_txq, &max_rxq,
-				   &num_indirect_entries);
+				   &apc->indir_table_sz);
 	if (err) {
 		netdev_err(ndev, "Failed to query info for vPort %d\n",
 			   port_idx);
@@ -2723,6 +2759,10 @@ static int mana_probe_port(struct mana_context *ac, int port_idx,
 	if (err)
 		goto free_net;
 
+	err = mana_rss_table_alloc(apc);
+	if (err)
+		goto reset_apc;
+
 	netdev_lockdep_set_classes(ndev);
 
 	ndev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
@@ -2739,11 +2779,13 @@ static int mana_probe_port(struct mana_context *ac, int port_idx,
 	err = register_netdev(ndev);
 	if (err) {
 		netdev_err(ndev, "Unable to register netdev.\n");
-		goto reset_apc;
+		goto free_indir;
 	}
 
 	return 0;
 
+free_indir:
+	mana_cleanup_indir_table(apc);
 reset_apc:
 	kfree(apc->rxqs);
 	apc->rxqs = NULL;
@@ -2872,16 +2914,30 @@ int mana_probe(struct gdma_dev *gd, bool resuming)
 	if (!resuming) {
 		for (i = 0; i < ac->num_ports; i++) {
 			err = mana_probe_port(ac, i, &ac->ports[i]);
-			if (err)
+			/* we log the port for which the probe failed and stop
+			 * probes for subsequent ports.
+			 * Note that we keep running ports, for which the probes
+			 * were successful, unless add_adev fails too
+			 */
+			if (err) {
+				dev_err(dev, "Probe Failed for port %d\n", i);
 				break;
+			}
 		}
 	} else {
 		for (i = 0; i < ac->num_ports; i++) {
 			rtnl_lock();
 			err = mana_attach(ac->ports[i]);
 			rtnl_unlock();
-			if (err)
+			/* we log the port for which the attach failed and stop
+			 * attach for subsequent ports
+			 * Note that we keep running ports, for which the attach
+			 * were successful, unless add_adev fails too
+			 */
+			if (err) {
+				dev_err(dev, "Attach Failed for port %d\n", i);
 				break;
+			}
 		}
 	}
 
@@ -2897,6 +2953,7 @@ void mana_remove(struct gdma_dev *gd, bool suspending)
 {
 	struct gdma_context *gc = gd->gdma_context;
 	struct mana_context *ac = gd->driver_data;
+	struct mana_port_context *apc;
 	struct device *dev = gc->dev;
 	struct net_device *ndev;
 	int err;
@@ -2908,6 +2965,7 @@ void mana_remove(struct gdma_dev *gd, bool suspending)
 
 	for (i = 0; i < ac->num_ports; i++) {
 		ndev = ac->ports[i];
+		apc = netdev_priv(ndev);
 		if (!ndev) {
 			if (i == 0)
 				dev_err(dev, "No net device to remove\n");
@@ -2931,6 +2989,7 @@ void mana_remove(struct gdma_dev *gd, bool suspending)
 		}
 
 		unregister_netdevice(ndev);
+		mana_cleanup_indir_table(apc);
 
 		rtnl_unlock();
 
diff --git a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c
index ab2413d71f6c..146d5db1792f 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_ethtool.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_ethtool.c
@@ -245,7 +245,9 @@ static u32 mana_get_rxfh_key_size(struct net_device *ndev)
 
 static u32 mana_rss_indir_size(struct net_device *ndev)
 {
-	return MANA_INDIRECT_TABLE_SIZE;
+	struct mana_port_context *apc = netdev_priv(ndev);
+
+	return apc->indir_table_sz;
 }
 
 static int mana_get_rxfh(struct net_device *ndev,
@@ -257,7 +259,7 @@ static int mana_get_rxfh(struct net_device *ndev,
 	rxfh->hfunc = ETH_RSS_HASH_TOP; /* Toeplitz */
 
 	if (rxfh->indir) {
-		for (i = 0; i < MANA_INDIRECT_TABLE_SIZE; i++)
+		for (i = 0; i < apc->indir_table_sz; i++)
 			rxfh->indir[i] = apc->indir_table[i];
 	}
 
@@ -273,8 +275,8 @@ static int mana_set_rxfh(struct net_device *ndev,
 {
 	struct mana_port_context *apc = netdev_priv(ndev);
 	bool update_hash = false, update_table = false;
-	u32 save_table[MANA_INDIRECT_TABLE_SIZE];
 	u8 save_key[MANA_HASH_KEY_SIZE];
+	u32 *save_table;
 	int i, err;
 
 	if (!apc->port_is_up)
@@ -284,13 +286,19 @@ static int mana_set_rxfh(struct net_device *ndev,
 	    rxfh->hfunc != ETH_RSS_HASH_TOP)
 		return -EOPNOTSUPP;
 
+	save_table = kcalloc(apc->indir_table_sz, sizeof(u32), GFP_KERNEL);
+	if (!save_table)
+		return -ENOMEM;
+
 	if (rxfh->indir) {
-		for (i = 0; i < MANA_INDIRECT_TABLE_SIZE; i++)
-			if (rxfh->indir[i] >= apc->num_queues)
-				return -EINVAL;
+		for (i = 0; i < apc->indir_table_sz; i++)
+			if (rxfh->indir[i] >= apc->num_queues) {
+				err = -EINVAL;
+				goto cleanup;
+			}
 
 		update_table = true;
-		for (i = 0; i < MANA_INDIRECT_TABLE_SIZE; i++) {
+		for (i = 0; i < apc->indir_table_sz; i++) {
 			save_table[i] = apc->indir_table[i];
 			apc->indir_table[i] = rxfh->indir[i];
 		}
@@ -306,7 +314,7 @@ static int mana_set_rxfh(struct net_device *ndev,
 
 	if (err) { /* recover to original values */
 		if (update_table) {
-			for (i = 0; i < MANA_INDIRECT_TABLE_SIZE; i++)
+			for (i = 0; i < apc->indir_table_sz; i++)
 				apc->indir_table[i] = save_table[i];
 		}
 
@@ -316,6 +324,9 @@ static int mana_set_rxfh(struct net_device *ndev,
 		mana_config_rss(apc, TRI_STATE_TRUE, update_hash, update_table);
 	}
 
+cleanup:
+	kfree(save_table);
+
 	return err;
 }
 
diff --git a/include/net/mana/gdma.h b/include/net/mana/gdma.h
index 27684135bb4d..c547756c4284 100644
--- a/include/net/mana/gdma.h
+++ b/include/net/mana/gdma.h
@@ -543,11 +543,13 @@ enum {
  */
 #define GDMA_DRV_CAP_FLAG_1_NAPI_WKDONE_FIX BIT(2)
 #define GDMA_DRV_CAP_FLAG_1_HWC_TIMEOUT_RECONFIG BIT(3)
+#define GDMA_DRV_CAP_FLAG_1_VARIABLE_INDIRECTION_TABLE_SUPPORT BIT(5)
 
 #define GDMA_DRV_CAP_FLAGS1 \
 	(GDMA_DRV_CAP_FLAG_1_EQ_SHARING_MULTI_VPORT | \
 	 GDMA_DRV_CAP_FLAG_1_NAPI_WKDONE_FIX | \
-	 GDMA_DRV_CAP_FLAG_1_HWC_TIMEOUT_RECONFIG)
+	 GDMA_DRV_CAP_FLAG_1_HWC_TIMEOUT_RECONFIG | \
+	 GDMA_DRV_CAP_FLAG_1_VARIABLE_INDIRECTION_TABLE_SUPPORT)
 
 #define GDMA_DRV_CAP_FLAGS2 0
 
diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
index 561f6719fb4e..59823901b74f 100644
--- a/include/net/mana/mana.h
+++ b/include/net/mana/mana.h
@@ -30,8 +30,8 @@ enum TRI_STATE {
 };
 
 /* Number of entries for hardware indirection table must be in power of 2 */
-#define MANA_INDIRECT_TABLE_SIZE 64
-#define MANA_INDIRECT_TABLE_MASK (MANA_INDIRECT_TABLE_SIZE - 1)
+#define MANA_INDIRECT_TABLE_MAX_SIZE 512
+#define MANA_INDIRECT_TABLE_DEF_SIZE 64
 
 /* The Toeplitz hash key's length in bytes: should be multiple of 8 */
 #define MANA_HASH_KEY_SIZE 40
@@ -410,10 +410,11 @@ struct mana_port_context {
 	struct mana_tx_qp *tx_qp;
 
 	/* Indirection Table for RX & TX. The values are queue indexes */
-	u32 indir_table[MANA_INDIRECT_TABLE_SIZE];
+	u32 *indir_table;
+	u32 indir_table_sz;
 
 	/* Indirection table containing RxObject Handles */
-	mana_handle_t rxobj_table[MANA_INDIRECT_TABLE_SIZE];
+	mana_handle_t *rxobj_table;
 
 	/*  Hash key used by the NIC */
 	u8 hashkey[MANA_HASH_KEY_SIZE];
-- 
cgit v1.2.3-58-ga151


From b975d3ee5962237c1e2f5d5aeeaaf0dc2173486c Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Sun, 9 Jun 2024 00:10:39 +0200
Subject: net: add and use skb_get_hash_net

Years ago flow dissector gained ability to delegate flow dissection
to a bpf program, scoped per netns.

Unfortunately, skb_get_hash() only gets an sk_buff argument instead
of both net+skb.  This means the flow dissector needs to obtain the
netns pointer from somewhere else.

The netns is derived from skb->dev, and if that is not available, from
skb->sk.  If neither is set, we hit a (benign) WARN_ON_ONCE().

Trying both dev and sk covers most cases, but not all, as recently
reported by Christoph Paasch.

In case of nf-generated tcp reset, both sk and dev are NULL:

WARNING: .. net/core/flow_dissector.c:1104
 skb_flow_dissect_flow_keys include/linux/skbuff.h:1536 [inline]
 skb_get_hash include/linux/skbuff.h:1578 [inline]
 nft_trace_init+0x7d/0x120 net/netfilter/nf_tables_trace.c:320
 nft_do_chain+0xb26/0xb90 net/netfilter/nf_tables_core.c:268
 nft_do_chain_ipv4+0x7a/0xa0 net/netfilter/nft_chain_filter.c:23
 nf_hook_slow+0x57/0x160 net/netfilter/core.c:626
 __ip_local_out+0x21d/0x260 net/ipv4/ip_output.c:118
 ip_local_out+0x26/0x1e0 net/ipv4/ip_output.c:127
 nf_send_reset+0x58c/0x700 net/ipv4/netfilter/nf_reject_ipv4.c:308
 nft_reject_ipv4_eval+0x53/0x90 net/ipv4/netfilter/nft_reject_ipv4.c:30
 [..]

syzkaller did something like this:
table inet filter {
  chain input {
    type filter hook input priority filter; policy accept;
    meta nftrace set 1
    tcp dport 42 reject with tcp reset
   }
   chain output {
    type filter hook output priority filter; policy accept;
    # empty chain is enough
   }
}

... then sends a tcp packet to port 42.

Initial attempt to simply set skb->dev from nf_reject_ipv4 doesn't cover
all cases: skbs generated via ipv4 igmp_send_report trigger similar splat.

Moreover, Pablo Neira found that nft_hash.c uses __skb_get_hash_symmetric()
which would trigger same warn splat for such skbs.

Lets allow callers to pass the current netns explicitly.
The nf_trace infrastructure is adjusted to use the new helper.

__skb_get_hash_symmetric is handled in the next patch.

Reported-by: Christoph Paasch <cpaasch@apple.com>
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/494
Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240608221057.16070-2-fw@strlen.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/skbuff.h          | 12 ++++++++++--
 net/core/flow_dissector.c       | 15 +++++++++++----
 net/netfilter/nf_tables_trace.c |  2 +-
 3 files changed, 22 insertions(+), 7 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index fe7d8dbef77e..6e78019f899a 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1498,7 +1498,7 @@ __skb_set_sw_hash(struct sk_buff *skb, __u32 hash, bool is_l4)
 	__skb_set_hash(skb, hash, true, is_l4);
 }
 
-void __skb_get_hash(struct sk_buff *skb);
+void __skb_get_hash_net(const struct net *net, struct sk_buff *skb);
 u32 __skb_get_hash_symmetric(const struct sk_buff *skb);
 u32 skb_get_poff(const struct sk_buff *skb);
 u32 __skb_get_poff(const struct sk_buff *skb, const void *data,
@@ -1578,10 +1578,18 @@ void skb_flow_dissect_hash(const struct sk_buff *skb,
 			   struct flow_dissector *flow_dissector,
 			   void *target_container);
 
+static inline __u32 skb_get_hash_net(const struct net *net, struct sk_buff *skb)
+{
+	if (!skb->l4_hash && !skb->sw_hash)
+		__skb_get_hash_net(net, skb);
+
+	return skb->hash;
+}
+
 static inline __u32 skb_get_hash(struct sk_buff *skb)
 {
 	if (!skb->l4_hash && !skb->sw_hash)
-		__skb_get_hash(skb);
+		__skb_get_hash_net(NULL, skb);
 
 	return skb->hash;
 }
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 59fe46077b3c..702b4f0a70b6 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -1860,7 +1860,8 @@ u32 __skb_get_hash_symmetric(const struct sk_buff *skb)
 EXPORT_SYMBOL_GPL(__skb_get_hash_symmetric);
 
 /**
- * __skb_get_hash: calculate a flow hash
+ * __skb_get_hash_net: calculate a flow hash
+ * @net: associated network namespace, derived from @skb if NULL
  * @skb: sk_buff to calculate flow hash from
  *
  * This function calculates a flow hash based on src/dst addresses
@@ -1868,18 +1869,24 @@ EXPORT_SYMBOL_GPL(__skb_get_hash_symmetric);
  * on success, zero indicates no valid hash.  Also, sets l4_hash in skb
  * if hash is a canonical 4-tuple hash over transport ports.
  */
-void __skb_get_hash(struct sk_buff *skb)
+void __skb_get_hash_net(const struct net *net, struct sk_buff *skb)
 {
 	struct flow_keys keys;
 	u32 hash;
 
+	memset(&keys, 0, sizeof(keys));
+
+	__skb_flow_dissect(net, skb, &flow_keys_dissector,
+			   &keys, NULL, 0, 0, 0,
+			   FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL);
+
 	__flow_hash_secret_init();
 
-	hash = ___skb_get_hash(skb, &keys, &hashrnd);
+	hash = __flow_hash_from_keys(&keys, &hashrnd);
 
 	__skb_set_sw_hash(skb, hash, flow_keys_have_l4(&keys));
 }
-EXPORT_SYMBOL(__skb_get_hash);
+EXPORT_SYMBOL(__skb_get_hash_net);
 
 __u32 skb_get_hash_perturb(const struct sk_buff *skb,
 			   const siphash_key_t *perturb)
diff --git a/net/netfilter/nf_tables_trace.c b/net/netfilter/nf_tables_trace.c
index a83637e3f455..580c55268f65 100644
--- a/net/netfilter/nf_tables_trace.c
+++ b/net/netfilter/nf_tables_trace.c
@@ -317,7 +317,7 @@ void nft_trace_init(struct nft_traceinfo *info, const struct nft_pktinfo *pkt,
 	net_get_random_once(&trace_key, sizeof(trace_key));
 
 	info->skbid = (u32)siphash_3u32(hash32_ptr(skb),
-					skb_get_hash(skb),
+					skb_get_hash_net(nft_net(pkt), skb),
 					skb->skb_iif,
 					&trace_key);
 }
-- 
cgit v1.2.3-58-ga151


From d1dab4f71d372e00e2d34a9c32bf261623e3a95c Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Sun, 9 Jun 2024 00:10:40 +0200
Subject: net: add and use __skb_get_hash_symmetric_net

Similar to previous patch: apply same logic for
__skb_get_hash_symmetric and let callers pass the netns to the dissector
core.

Existing function is turned into a wrapper to avoid adjusting all
callers, nft_hash.c uses new function.

Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240608221057.16070-3-fw@strlen.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/skbuff.h    | 8 +++++++-
 net/core/flow_dissector.c | 6 +++---
 net/netfilter/nft_hash.c  | 3 ++-
 3 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 6e78019f899a..813406a9bd6c 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1498,8 +1498,14 @@ __skb_set_sw_hash(struct sk_buff *skb, __u32 hash, bool is_l4)
 	__skb_set_hash(skb, hash, true, is_l4);
 }
 
+u32 __skb_get_hash_symmetric_net(const struct net *net, const struct sk_buff *skb);
+
+static inline u32 __skb_get_hash_symmetric(const struct sk_buff *skb)
+{
+	return __skb_get_hash_symmetric_net(NULL, skb);
+}
+
 void __skb_get_hash_net(const struct net *net, struct sk_buff *skb);
-u32 __skb_get_hash_symmetric(const struct sk_buff *skb);
 u32 skb_get_poff(const struct sk_buff *skb);
 u32 __skb_get_poff(const struct sk_buff *skb, const void *data,
 		   const struct flow_keys_basic *keys, int hlen);
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 702b4f0a70b6..e479790db0f7 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -1845,19 +1845,19 @@ EXPORT_SYMBOL(make_flow_keys_digest);
 
 static struct flow_dissector flow_keys_dissector_symmetric __read_mostly;
 
-u32 __skb_get_hash_symmetric(const struct sk_buff *skb)
+u32 __skb_get_hash_symmetric_net(const struct net *net, const struct sk_buff *skb)
 {
 	struct flow_keys keys;
 
 	__flow_hash_secret_init();
 
 	memset(&keys, 0, sizeof(keys));
-	__skb_flow_dissect(NULL, skb, &flow_keys_dissector_symmetric,
+	__skb_flow_dissect(net, skb, &flow_keys_dissector_symmetric,
 			   &keys, NULL, 0, 0, 0, 0);
 
 	return __flow_hash_from_keys(&keys, &hashrnd);
 }
-EXPORT_SYMBOL_GPL(__skb_get_hash_symmetric);
+EXPORT_SYMBOL_GPL(__skb_get_hash_symmetric_net);
 
 /**
  * __skb_get_hash_net: calculate a flow hash
diff --git a/net/netfilter/nft_hash.c b/net/netfilter/nft_hash.c
index 92d47e469204..868d68302d22 100644
--- a/net/netfilter/nft_hash.c
+++ b/net/netfilter/nft_hash.c
@@ -51,7 +51,8 @@ static void nft_symhash_eval(const struct nft_expr *expr,
 	struct sk_buff *skb = pkt->skb;
 	u32 h;
 
-	h = reciprocal_scale(__skb_get_hash_symmetric(skb), priv->modulus);
+	h = reciprocal_scale(__skb_get_hash_symmetric_net(nft_net(pkt), skb),
+			     priv->modulus);
 
 	regs->data[priv->dreg] = h + priv->offset;
 }
-- 
cgit v1.2.3-58-ga151


From 721478fe6a5cd243e289105e3fdd0ecdd9d39ac3 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Tue, 11 Jun 2024 15:48:48 +0200
Subject: ravb: RAVB should select PAGE_POOL

If CONFIG_PAGE_POOL is not enabled:

    aarch64-linux-gnu-ld: Unexpected GOT/PLT entries detected!
    aarch64-linux-gnu-ld: Unexpected run-time procedure linkages detected!
    aarch64-linux-gnu-ld: drivers/net/ethernet/renesas/ravb_main.o: in function `ravb_rx_ring_refill':
    ravb_main.c:(.text+0x8d8): undefined reference to `page_pool_alloc_pages'
    aarch64-linux-gnu-ld: ravb_main.c:(.text+0x944): undefined reference to `page_pool_alloc_frag'
    aarch64-linux-gnu-ld: drivers/net/ethernet/renesas/ravb_main.o: in function `ravb_ring_init':
    ravb_main.c:(.text+0x1d4c): undefined reference to `page_pool_create'

Fixes: 966726324b7b1400 ("net: ravb: Allocate RX buffers via page pool")
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Paul Barker <paul.barker.ct@bp.renesas.com>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Link: https://lore.kernel.org/r/fa61b464ae1aa7630e9024f091991937941d49f1.1718113630.git.geert+renesas@glider.be
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/renesas/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/renesas/Kconfig b/drivers/net/ethernet/renesas/Kconfig
index b03fae7a0f72..83de7e56da61 100644
--- a/drivers/net/ethernet/renesas/Kconfig
+++ b/drivers/net/ethernet/renesas/Kconfig
@@ -33,6 +33,7 @@ config RAVB
 	select CRC32
 	select MII
 	select MDIO_BITBANG
+	select PAGE_POOL
 	select PHYLIB
 	select RESET_CONTROLLER
 	help
-- 
cgit v1.2.3-58-ga151


From 32b06603f8790b34835af43d259152eb57827413 Mon Sep 17 00:00:00 2001
From: Sean Anderson <sean.anderson@linux.dev>
Date: Tue, 11 Jun 2024 11:41:16 -0400
Subject: net: xilinx: axienet: Use NL_SET_ERR_MSG instead of netdev_err

This error message can be triggered by userspace. Use NL_SET_ERR_MSG so
the message is returned to the user and to avoid polluting the kernel
logs. Additionally, change the return value from EFAULT to EBUSY to
better reflect the error (which has nothing to do with addressing).

Signed-off-by: Sean Anderson <sean.anderson@linux.dev>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Radhey Shyam Pandey <radhey.shyam.pandey@amd.com>
Link: https://lore.kernel.org/r/20240611154116.2643662-1-sean.anderson@linux.dev
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
index c29809cd9201..5f98daa5b341 100644
--- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
+++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
@@ -1945,9 +1945,9 @@ axienet_ethtools_set_coalesce(struct net_device *ndev,
 	struct axienet_local *lp = netdev_priv(ndev);
 
 	if (netif_running(ndev)) {
-		netdev_err(ndev,
-			   "Please stop netif before applying configuration\n");
-		return -EFAULT;
+		NL_SET_ERR_MSG(extack,
+			       "Please stop netif before applying configuration");
+		return -EBUSY;
 	}
 
 	if (ecoalesce->rx_max_coalesced_frames)
-- 
cgit v1.2.3-58-ga151


From 3e453ca122d483eb519f934b6624215f0536301c Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Fri, 7 Jun 2024 17:13:53 +0200
Subject: net: ipv4,ipv6: Pass multipath hash computation through a helper

The following patches will add a sysctl to control multipath hash
seed. In order to centralize the hash computation, add a helper,
fib_multipath_hash_from_keys(), and have all IPv4 and IPv6 route.c
invocations of flow_hash_from_keys() go through this helper instead.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240607151357.421181-2-petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/ip_fib.h |  7 +++++++
 net/ipv4/route.c     | 12 ++++++------
 net/ipv6/route.c     | 12 ++++++------
 3 files changed, 19 insertions(+), 12 deletions(-)

diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 9b2f69ba5e49..b8b3c07e8f7b 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -521,6 +521,13 @@ void fib_nhc_update_mtu(struct fib_nh_common *nhc, u32 new, u32 orig);
 int fib_multipath_hash(const struct net *net, const struct flowi4 *fl4,
 		       const struct sk_buff *skb, struct flow_keys *flkeys);
 #endif
+
+static inline u32 fib_multipath_hash_from_keys(const struct net *net,
+					       struct flow_keys *keys)
+{
+	return flow_hash_from_keys(keys);
+}
+
 int fib_check_nh(struct net *net, struct fib_nh *nh, u32 table, u8 scope,
 		 struct netlink_ext_ack *extack);
 void fib_select_multipath(struct fib_result *res, int hash);
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index cb0bdf34ed50..54512acbead7 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1923,7 +1923,7 @@ static u32 fib_multipath_custom_hash_outer(const struct net *net,
 		hash_keys.ports.dst = keys.ports.dst;
 
 	*p_has_inner = !!(keys.control.flags & FLOW_DIS_ENCAPSULATION);
-	return flow_hash_from_keys(&hash_keys);
+	return fib_multipath_hash_from_keys(net, &hash_keys);
 }
 
 static u32 fib_multipath_custom_hash_inner(const struct net *net,
@@ -1972,7 +1972,7 @@ static u32 fib_multipath_custom_hash_inner(const struct net *net,
 	if (hash_fields & FIB_MULTIPATH_HASH_FIELD_INNER_DST_PORT)
 		hash_keys.ports.dst = keys.ports.dst;
 
-	return flow_hash_from_keys(&hash_keys);
+	return fib_multipath_hash_from_keys(net, &hash_keys);
 }
 
 static u32 fib_multipath_custom_hash_skb(const struct net *net,
@@ -2009,7 +2009,7 @@ static u32 fib_multipath_custom_hash_fl4(const struct net *net,
 	if (hash_fields & FIB_MULTIPATH_HASH_FIELD_DST_PORT)
 		hash_keys.ports.dst = fl4->fl4_dport;
 
-	return flow_hash_from_keys(&hash_keys);
+	return fib_multipath_hash_from_keys(net, &hash_keys);
 }
 
 /* if skb is set it will be used and fl4 can be NULL */
@@ -2030,7 +2030,7 @@ int fib_multipath_hash(const struct net *net, const struct flowi4 *fl4,
 			hash_keys.addrs.v4addrs.src = fl4->saddr;
 			hash_keys.addrs.v4addrs.dst = fl4->daddr;
 		}
-		mhash = flow_hash_from_keys(&hash_keys);
+		mhash = fib_multipath_hash_from_keys(net, &hash_keys);
 		break;
 	case 1:
 		/* skb is currently provided only when forwarding */
@@ -2064,7 +2064,7 @@ int fib_multipath_hash(const struct net *net, const struct flowi4 *fl4,
 			hash_keys.ports.dst = fl4->fl4_dport;
 			hash_keys.basic.ip_proto = fl4->flowi4_proto;
 		}
-		mhash = flow_hash_from_keys(&hash_keys);
+		mhash = fib_multipath_hash_from_keys(net, &hash_keys);
 		break;
 	case 2:
 		memset(&hash_keys, 0, sizeof(hash_keys));
@@ -2095,7 +2095,7 @@ int fib_multipath_hash(const struct net *net, const struct flowi4 *fl4,
 			hash_keys.addrs.v4addrs.src = fl4->saddr;
 			hash_keys.addrs.v4addrs.dst = fl4->daddr;
 		}
-		mhash = flow_hash_from_keys(&hash_keys);
+		mhash = fib_multipath_hash_from_keys(net, &hash_keys);
 		break;
 	case 3:
 		if (skb)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index ad5fff5a210c..1916de615398 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2372,7 +2372,7 @@ static u32 rt6_multipath_custom_hash_outer(const struct net *net,
 		hash_keys.ports.dst = keys.ports.dst;
 
 	*p_has_inner = !!(keys.control.flags & FLOW_DIS_ENCAPSULATION);
-	return flow_hash_from_keys(&hash_keys);
+	return fib_multipath_hash_from_keys(net, &hash_keys);
 }
 
 static u32 rt6_multipath_custom_hash_inner(const struct net *net,
@@ -2421,7 +2421,7 @@ static u32 rt6_multipath_custom_hash_inner(const struct net *net,
 	if (hash_fields & FIB_MULTIPATH_HASH_FIELD_INNER_DST_PORT)
 		hash_keys.ports.dst = keys.ports.dst;
 
-	return flow_hash_from_keys(&hash_keys);
+	return fib_multipath_hash_from_keys(net, &hash_keys);
 }
 
 static u32 rt6_multipath_custom_hash_skb(const struct net *net,
@@ -2460,7 +2460,7 @@ static u32 rt6_multipath_custom_hash_fl6(const struct net *net,
 	if (hash_fields & FIB_MULTIPATH_HASH_FIELD_DST_PORT)
 		hash_keys.ports.dst = fl6->fl6_dport;
 
-	return flow_hash_from_keys(&hash_keys);
+	return fib_multipath_hash_from_keys(net, &hash_keys);
 }
 
 /* if skb is set it will be used and fl6 can be NULL */
@@ -2482,7 +2482,7 @@ u32 rt6_multipath_hash(const struct net *net, const struct flowi6 *fl6,
 			hash_keys.tags.flow_label = (__force u32)flowi6_get_flowlabel(fl6);
 			hash_keys.basic.ip_proto = fl6->flowi6_proto;
 		}
-		mhash = flow_hash_from_keys(&hash_keys);
+		mhash = fib_multipath_hash_from_keys(net, &hash_keys);
 		break;
 	case 1:
 		if (skb) {
@@ -2514,7 +2514,7 @@ u32 rt6_multipath_hash(const struct net *net, const struct flowi6 *fl6,
 			hash_keys.ports.dst = fl6->fl6_dport;
 			hash_keys.basic.ip_proto = fl6->flowi6_proto;
 		}
-		mhash = flow_hash_from_keys(&hash_keys);
+		mhash = fib_multipath_hash_from_keys(net, &hash_keys);
 		break;
 	case 2:
 		memset(&hash_keys, 0, sizeof(hash_keys));
@@ -2551,7 +2551,7 @@ u32 rt6_multipath_hash(const struct net *net, const struct flowi6 *fl6,
 			hash_keys.tags.flow_label = (__force u32)flowi6_get_flowlabel(fl6);
 			hash_keys.basic.ip_proto = fl6->flowi6_proto;
 		}
-		mhash = flow_hash_from_keys(&hash_keys);
+		mhash = fib_multipath_hash_from_keys(net, &hash_keys);
 		break;
 	case 3:
 		if (skb)
-- 
cgit v1.2.3-58-ga151


From 4ee2a8cace3fb9a34aea6a56426f89d26dd514f3 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Fri, 7 Jun 2024 17:13:54 +0200
Subject: net: ipv4: Add a sysctl to set multipath hash seed

When calculating hashes for the purpose of multipath forwarding, both IPv4
and IPv6 code currently fall back on flow_hash_from_keys(). That uses a
randomly-generated seed. That's a fine choice by default, but unfortunately
some deployments may need a tighter control over the seed used.

In this patch, make the seed configurable by adding a new sysctl key,
net.ipv4.fib_multipath_hash_seed to control the seed. This seed is used
specifically for multipath forwarding and not for the other concerns that
flow_hash_from_keys() is used for, such as queue selection. Expose the knob
as sysctl because other such settings, such as headers to hash, are also
handled that way. Like those, the multipath hash seed is a per-netns
variable.

Despite being placed in the net.ipv4 namespace, the multipath seed sysctl
is used for both IPv4 and IPv6, similarly to e.g. a number of TCP
variables.

The seed used by flow_hash_from_keys() is a 128-bit quantity. However it
seems that usually the seed is a much more modest value. 32 bits seem
typical (Cisco, Cumulus), some systems go even lower. For that reason, and
to decouple the user interface from implementation details, go with a
32-bit quantity, which is then quadruplicated to form the siphash key.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240607151357.421181-3-petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/ip-sysctl.rst | 14 ++++++++
 include/net/flow_dissector.h           |  2 ++
 include/net/ip_fib.h                   | 23 +++++++++++-
 include/net/netns/ipv4.h               |  8 +++++
 net/core/flow_dissector.c              |  7 ++++
 net/ipv4/sysctl_net_ipv4.c             | 66 ++++++++++++++++++++++++++++++++++
 6 files changed, 119 insertions(+), 1 deletion(-)

diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst
index 6e99eccdb837..3616389c8c2d 100644
--- a/Documentation/networking/ip-sysctl.rst
+++ b/Documentation/networking/ip-sysctl.rst
@@ -131,6 +131,20 @@ fib_multipath_hash_fields - UNSIGNED INTEGER
 
 	Default: 0x0007 (source IP, destination IP and IP protocol)
 
+fib_multipath_hash_seed - UNSIGNED INTEGER
+	The seed value used when calculating hash for multipath routes. Applies
+	to both IPv4 and IPv6 datapath. Only present for kernels built with
+	CONFIG_IP_ROUTE_MULTIPATH enabled.
+
+	When set to 0, the seed value used for multipath routing defaults to an
+	internal random-generated one.
+
+	The actual hashing algorithm is not specified -- there is no guarantee
+	that a next hop distribution effected by a given seed will keep stable
+	across kernel versions.
+
+	Default: 0 (random)
+
 fib_sync_mem - UNSIGNED INTEGER
 	Amount of dirty memory from fib entries that can be backlogged before
 	synchronize_rcu is forced.
diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h
index 99626475c3f4..3e47e123934d 100644
--- a/include/net/flow_dissector.h
+++ b/include/net/flow_dissector.h
@@ -442,6 +442,8 @@ static inline bool flow_keys_have_l4(const struct flow_keys *keys)
 }
 
 u32 flow_hash_from_keys(struct flow_keys *keys);
+u32 flow_hash_from_keys_seed(struct flow_keys *keys,
+			     const siphash_key_t *keyval);
 void skb_flow_get_icmp_tci(const struct sk_buff *skb,
 			   struct flow_dissector_key_icmp *key_icmp,
 			   const void *data, int thoff, int hlen);
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index b8b3c07e8f7b..6e7984bfb986 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -520,13 +520,34 @@ void fib_nhc_update_mtu(struct fib_nh_common *nhc, u32 new, u32 orig);
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
 int fib_multipath_hash(const struct net *net, const struct flowi4 *fl4,
 		       const struct sk_buff *skb, struct flow_keys *flkeys);
-#endif
+
+static void
+fib_multipath_hash_construct_key(siphash_key_t *key, u32 mp_seed)
+{
+	u64 mp_seed_64 = mp_seed;
+
+	key->key[0] = (mp_seed_64 << 32) | mp_seed_64;
+	key->key[1] = key->key[0];
+}
 
+static inline u32 fib_multipath_hash_from_keys(const struct net *net,
+					       struct flow_keys *keys)
+{
+	siphash_aligned_key_t hash_key;
+	u32 mp_seed;
+
+	mp_seed = READ_ONCE(net->ipv4.sysctl_fib_multipath_hash_seed).mp_seed;
+	fib_multipath_hash_construct_key(&hash_key, mp_seed);
+
+	return flow_hash_from_keys_seed(keys, &hash_key);
+}
+#else
 static inline u32 fib_multipath_hash_from_keys(const struct net *net,
 					       struct flow_keys *keys)
 {
 	return flow_hash_from_keys(keys);
 }
+#endif
 
 int fib_check_nh(struct net *net, struct fib_nh *nh, u32 table, u8 scope,
 		 struct netlink_ext_ack *extack);
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index a91bb971f901..5fcd61ada622 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -40,6 +40,13 @@ struct inet_timewait_death_row {
 
 struct tcp_fastopen_context;
 
+#ifdef CONFIG_IP_ROUTE_MULTIPATH
+struct sysctl_fib_multipath_hash_seed {
+	u32 user_seed;
+	u32 mp_seed;
+};
+#endif
+
 struct netns_ipv4 {
 	/* Cacheline organization can be found documented in
 	 * Documentation/networking/net_cachelines/netns_ipv4_sysctl.rst.
@@ -246,6 +253,7 @@ struct netns_ipv4 {
 #endif
 #endif
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
+	struct sysctl_fib_multipath_hash_seed sysctl_fib_multipath_hash_seed;
 	u32 sysctl_fib_multipath_hash_fields;
 	u8 sysctl_fib_multipath_use_neigh;
 	u8 sysctl_fib_multipath_hash_policy;
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index e479790db0f7..e64a26379807 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -1806,6 +1806,13 @@ u32 flow_hash_from_keys(struct flow_keys *keys)
 }
 EXPORT_SYMBOL(flow_hash_from_keys);
 
+u32 flow_hash_from_keys_seed(struct flow_keys *keys,
+			     const siphash_key_t *keyval)
+{
+	return __flow_hash_from_keys(keys, keyval);
+}
+EXPORT_SYMBOL(flow_hash_from_keys_seed);
+
 static inline u32 ___skb_get_hash(const struct sk_buff *skb,
 				  struct flow_keys *keys,
 				  const siphash_key_t *keyval)
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index bb64c0ef092d..9140d20eb2d4 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -464,6 +464,61 @@ static int proc_fib_multipath_hash_fields(struct ctl_table *table, int write,
 
 	return ret;
 }
+
+static u32 proc_fib_multipath_hash_rand_seed __ro_after_init;
+
+static void proc_fib_multipath_hash_init_rand_seed(void)
+{
+	get_random_bytes(&proc_fib_multipath_hash_rand_seed,
+			 sizeof(proc_fib_multipath_hash_rand_seed));
+}
+
+static void proc_fib_multipath_hash_set_seed(struct net *net, u32 user_seed)
+{
+	struct sysctl_fib_multipath_hash_seed new = {
+		.user_seed = user_seed,
+		.mp_seed = (user_seed ? user_seed :
+			    proc_fib_multipath_hash_rand_seed),
+	};
+
+	WRITE_ONCE(net->ipv4.sysctl_fib_multipath_hash_seed, new);
+}
+
+static int proc_fib_multipath_hash_seed(struct ctl_table *table, int write,
+					void *buffer, size_t *lenp,
+					loff_t *ppos)
+{
+	struct sysctl_fib_multipath_hash_seed *mphs;
+	struct net *net = table->data;
+	struct ctl_table tmp;
+	u32 user_seed;
+	int ret;
+
+	mphs = &net->ipv4.sysctl_fib_multipath_hash_seed;
+	user_seed = mphs->user_seed;
+
+	tmp = *table;
+	tmp.data = &user_seed;
+
+	ret = proc_douintvec_minmax(&tmp, write, buffer, lenp, ppos);
+
+	if (write && ret == 0) {
+		proc_fib_multipath_hash_set_seed(net, user_seed);
+		call_netevent_notifiers(NETEVENT_IPV4_MPATH_HASH_UPDATE, net);
+	}
+
+	return ret;
+}
+#else
+
+static void proc_fib_multipath_hash_init_rand_seed(void)
+{
+}
+
+static void proc_fib_multipath_hash_set_seed(struct net *net, u32 user_seed)
+{
+}
+
 #endif
 
 static struct ctl_table ipv4_table[] = {
@@ -1072,6 +1127,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.extra1		= SYSCTL_ONE,
 		.extra2		= &fib_multipath_hash_fields_all_mask,
 	},
+	{
+		.procname	= "fib_multipath_hash_seed",
+		.data		= &init_net,
+		.maxlen		= sizeof(u32),
+		.mode		= 0644,
+		.proc_handler	= proc_fib_multipath_hash_seed,
+	},
 #endif
 	{
 		.procname	= "ip_unprivileged_port_start",
@@ -1550,6 +1612,8 @@ static __net_init int ipv4_sysctl_init_net(struct net *net)
 	if (!net->ipv4.sysctl_local_reserved_ports)
 		goto err_ports;
 
+	proc_fib_multipath_hash_set_seed(net, 0);
+
 	return 0;
 
 err_ports:
@@ -1584,6 +1648,8 @@ static __init int sysctl_ipv4_init(void)
 	if (!hdr)
 		return -ENOMEM;
 
+	proc_fib_multipath_hash_init_rand_seed();
+
 	if (register_pernet_subsys(&ipv4_sysctl_ops)) {
 		unregister_net_sysctl_table(hdr);
 		return -ENOMEM;
-- 
cgit v1.2.3-58-ga151


From 60bcfede3f9f54d4f1614906dd73e6acc23e9518 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Fri, 7 Jun 2024 17:13:55 +0200
Subject: mlxsw: spectrum_router: Apply user-defined multipath hash seed

When Spectrum machines compute hash for the purposes of ECMP routing, they
use a seed specified through RECR_v2 (Router ECMP Configuration Register).
Up until now mlxsw computed the seed by hashing the machine's base MAC.
Now that we can optionally have a user-provided seed, use that if possible.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240607151357.421181-4-petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 40ba314fbc72..800dfb64ec83 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -11450,12 +11450,16 @@ static int mlxsw_sp_mp_hash_init(struct mlxsw_sp *mlxsw_sp)
 {
 	bool old_inc_parsing_depth, new_inc_parsing_depth;
 	struct mlxsw_sp_mp_hash_config config = {};
+	struct net *net = mlxsw_sp_net(mlxsw_sp);
 	char recr2_pl[MLXSW_REG_RECR2_LEN];
 	unsigned long bit;
 	u32 seed;
 	int err;
 
-	seed = jhash(mlxsw_sp->base_mac, sizeof(mlxsw_sp->base_mac), 0);
+	seed = READ_ONCE(net->ipv4.sysctl_fib_multipath_hash_seed).user_seed;
+	if (!seed)
+		seed = jhash(mlxsw_sp->base_mac, sizeof(mlxsw_sp->base_mac), 0);
+
 	mlxsw_reg_recr2_pack(recr2_pl, seed);
 	mlxsw_sp_mp4_hash_init(mlxsw_sp, &config);
 	mlxsw_sp_mp6_hash_init(mlxsw_sp, &config);
-- 
cgit v1.2.3-58-ga151


From 6f51aed38a4f4923a699d741553c612dec85fd14 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Fri, 7 Jun 2024 17:13:56 +0200
Subject: selftests: forwarding: lib: Split sysctl_save() out of sysctl_set()

In order to be able to save the current value of a sysctl without changing
it, split the relevant bit out of sysctl_set() into a new helper.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240607151357.421181-5-petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/forwarding/lib.sh | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh
index eabbdf00d8ca..9086d2015296 100644
--- a/tools/testing/selftests/net/forwarding/lib.sh
+++ b/tools/testing/selftests/net/forwarding/lib.sh
@@ -1134,12 +1134,19 @@ bridge_ageing_time_get()
 }
 
 declare -A SYSCTL_ORIG
+sysctl_save()
+{
+	local key=$1; shift
+
+	SYSCTL_ORIG[$key]=$(sysctl -n $key)
+}
+
 sysctl_set()
 {
 	local key=$1; shift
 	local value=$1; shift
 
-	SYSCTL_ORIG[$key]=$(sysctl -n $key)
+	sysctl_save "$key"
 	sysctl -qw $key="$value"
 }
 
-- 
cgit v1.2.3-58-ga151


From 5f90d93b6108098f389f992e267588924e446049 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Fri, 7 Jun 2024 17:13:57 +0200
Subject: selftests: forwarding: router_mpath_hash: Add a new selftest

Add a selftest that exercises the sysctl added in the previous patches.

Test that set/get works as expected; that across seeds we eventually hit
all NHs (test_mpath_seed_*); and that a given seed keeps hitting the same
NHs even across seed changes (test_mpath_seed_stability_*).

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20240607151357.421181-6-petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/forwarding/Makefile    |   1 +
 .../selftests/net/forwarding/router_mpath_seed.sh  | 333 +++++++++++++++++++++
 2 files changed, 334 insertions(+)
 create mode 100755 tools/testing/selftests/net/forwarding/router_mpath_seed.sh

diff --git a/tools/testing/selftests/net/forwarding/Makefile b/tools/testing/selftests/net/forwarding/Makefile
index fa7b59ff4029..99576d7ecbf6 100644
--- a/tools/testing/selftests/net/forwarding/Makefile
+++ b/tools/testing/selftests/net/forwarding/Makefile
@@ -70,6 +70,7 @@ TEST_PROGS = bridge_fdb_learning_limit.sh \
 	router_broadcast.sh \
 	router_mpath_nh_res.sh \
 	router_mpath_nh.sh \
+	router_mpath_seed.sh \
 	router_multicast.sh \
 	router_multipath.sh \
 	router_nh.sh \
diff --git a/tools/testing/selftests/net/forwarding/router_mpath_seed.sh b/tools/testing/selftests/net/forwarding/router_mpath_seed.sh
new file mode 100755
index 000000000000..314cb906c1eb
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/router_mpath_seed.sh
@@ -0,0 +1,333 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+# +-------------------------+  +-------------------------+
+# |  H1                     |  |                      H2 |
+# |               $h1 +     |  | + $h2                   |
+# |      192.0.2.1/28 |     |  | | 192.0.2.34/28         |
+# |  2001:db8:1::1/64 |     |  | | 2001:db8:3::2/64      |
+# +-------------------|-----+  +-|-----------------------+
+#                     |          |
+# +-------------------|-----+  +-|-----------------------+
+# |  R1               |     |  | |                    R2 |
+# |             $rp11 +     |  | + $rp21                 |
+# |      192.0.2.2/28       |  |   192.0.2.33/28         |
+# |  2001:db8:1::2/64       |  |   2001:db8:3::1/64      |
+# |                         |  |                         |
+# |             $rp12 +     |  | + $rp22                 |
+# |     192.0.2.17/28 |     |  | | 192.0.2.18..27/28     |
+# | 2001:db8:2::17/64 |     |  | | 2001:db8:2::18..27/64 |
+# +-------------------|-----+  +-|-----------------------+
+#                     |          |
+#                     `----------'
+
+ALL_TESTS="
+	ping_ipv4
+	ping_ipv6
+	test_mpath_seed_stability_ipv4
+	test_mpath_seed_stability_ipv6
+	test_mpath_seed_get
+	test_mpath_seed_ipv4
+	test_mpath_seed_ipv6
+"
+NUM_NETIFS=6
+source lib.sh
+
+h1_create()
+{
+	simple_if_init $h1 192.0.2.1/28 2001:db8:1::1/64
+	ip -4 route add 192.0.2.32/28 vrf v$h1 nexthop via 192.0.2.2
+	ip -6 route add 2001:db8:3::/64 vrf v$h1 nexthop via 2001:db8:1::2
+}
+
+h1_destroy()
+{
+	ip -6 route del 2001:db8:3::/64 vrf v$h1 nexthop via 2001:db8:1::2
+	ip -4 route del 192.0.2.32/28 vrf v$h1 nexthop via 192.0.2.2
+	simple_if_fini $h1 192.0.2.1/28 2001:db8:1::1/64
+}
+
+h2_create()
+{
+	simple_if_init $h2 192.0.2.34/28 2001:db8:3::2/64
+	ip -4 route add 192.0.2.0/28 vrf v$h2 nexthop via 192.0.2.33
+	ip -6 route add 2001:db8:1::/64 vrf v$h2 nexthop via 2001:db8:3::1
+}
+
+h2_destroy()
+{
+	ip -6 route del 2001:db8:1::/64 vrf v$h2 nexthop via 2001:db8:3::1
+	ip -4 route del 192.0.2.0/28 vrf v$h2 nexthop via 192.0.2.33
+	simple_if_fini $h2 192.0.2.34/28 2001:db8:3::2/64
+}
+
+router1_create()
+{
+	simple_if_init $rp11 192.0.2.2/28 2001:db8:1::2/64
+	__simple_if_init $rp12 v$rp11 192.0.2.17/28 2001:db8:2::17/64
+}
+
+router1_destroy()
+{
+	__simple_if_fini $rp12 192.0.2.17/28 2001:db8:2::17/64
+	simple_if_fini $rp11 192.0.2.2/28 2001:db8:1::2/64
+}
+
+router2_create()
+{
+	simple_if_init $rp21 192.0.2.33/28 2001:db8:3::1/64
+	__simple_if_init $rp22 v$rp21 192.0.2.18/28 2001:db8:2::18/64
+	ip -4 route add 192.0.2.0/28 vrf v$rp21 nexthop via 192.0.2.17
+	ip -6 route add 2001:db8:1::/64 vrf v$rp21 nexthop via 2001:db8:2::17
+}
+
+router2_destroy()
+{
+	ip -6 route del 2001:db8:1::/64 vrf v$rp21 nexthop via 2001:db8:2::17
+	ip -4 route del 192.0.2.0/28 vrf v$rp21 nexthop via 192.0.2.17
+	__simple_if_fini $rp22 192.0.2.18/28 2001:db8:2::18/64
+	simple_if_fini $rp21 192.0.2.33/28 2001:db8:3::1/64
+}
+
+nexthops_create()
+{
+	local i
+	for i in $(seq 10); do
+		ip nexthop add id $((1000 + i)) via 192.0.2.18 dev $rp12
+		ip nexthop add id $((2000 + i)) via 2001:db8:2::18 dev $rp12
+	done
+
+	ip nexthop add id 1000 group $(seq -s / 1001 1010) hw_stats on
+	ip nexthop add id 2000 group $(seq -s / 2001 2010) hw_stats on
+	ip -4 route add 192.0.2.32/28 vrf v$rp11 nhid 1000
+	ip -6 route add 2001:db8:3::/64 vrf v$rp11 nhid 2000
+}
+
+nexthops_destroy()
+{
+	local i
+
+	ip -6 route del 2001:db8:3::/64 vrf v$rp11 nhid 2000
+	ip -4 route del 192.0.2.32/28 vrf v$rp11 nhid 1000
+	ip nexthop del id 2000
+	ip nexthop del id 1000
+
+	for i in $(seq 10 -1 1); do
+		ip nexthop del id $((2000 + i))
+		ip nexthop del id $((1000 + i))
+	done
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	rp11=${NETIFS[p2]}
+
+	rp12=${NETIFS[p3]}
+	rp22=${NETIFS[p4]}
+
+	rp21=${NETIFS[p5]}
+	h2=${NETIFS[p6]}
+
+	sysctl_save net.ipv4.fib_multipath_hash_seed
+
+	vrf_prepare
+
+	h1_create
+	h2_create
+	router1_create
+	router2_create
+
+	forwarding_enable
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	forwarding_restore
+
+	nexthops_destroy
+	router2_destroy
+	router1_destroy
+	h2_destroy
+	h1_destroy
+
+	vrf_cleanup
+
+	sysctl_restore net.ipv4.fib_multipath_hash_seed
+}
+
+ping_ipv4()
+{
+	ping_test $h1 192.0.2.34
+}
+
+ping_ipv6()
+{
+	ping6_test $h1 2001:db8:3::2
+}
+
+test_mpath_seed_get()
+{
+	RET=0
+
+	local i
+	for ((i = 0; i < 100; i++)); do
+		local seed_w=$((999331 * i))
+		sysctl -qw net.ipv4.fib_multipath_hash_seed=$seed_w
+		local seed_r=$(sysctl -n net.ipv4.fib_multipath_hash_seed)
+		((seed_r == seed_w))
+		check_err $? "mpath seed written as $seed_w, but read as $seed_r"
+	done
+
+	log_test "mpath seed set/get"
+}
+
+nh_stats_snapshot()
+{
+	local group_id=$1; shift
+
+	ip -j -s -s nexthop show id $group_id |
+	    jq -c '[.[].group_stats | sort_by(.id) | .[].packets]'
+}
+
+get_active_nh()
+{
+	local s0=$1; shift
+	local s1=$1; shift
+
+	jq -n --argjson s0 "$s0" --argjson s1 "$s1" -f /dev/stdin <<-"EOF"
+		[range($s0 | length)] |
+		map($s1[.] - $s0[.]) |
+		map(if . > 8 then 1 else 0 end) |
+		index(1)
+	EOF
+}
+
+probe_nh()
+{
+	local group_id=$1; shift
+	local -a mz=("$@")
+
+	local s0=$(nh_stats_snapshot $group_id)
+	"${mz[@]}"
+	local s1=$(nh_stats_snapshot $group_id)
+
+	get_active_nh "$s0" "$s1"
+}
+
+probe_seed()
+{
+	local group_id=$1; shift
+	local seed=$1; shift
+	local -a mz=("$@")
+
+	sysctl -qw net.ipv4.fib_multipath_hash_seed=$seed
+	probe_nh "$group_id" "${mz[@]}"
+}
+
+test_mpath_seed()
+{
+	local group_id=$1; shift
+	local what=$1; shift
+	local -a mz=("$@")
+	local ii
+
+	RET=0
+
+	local -a tally=(0 0 0 0 0 0 0 0 0 0)
+	for ((ii = 0; ii < 100; ii++)); do
+		local act=$(probe_seed $group_id $((999331 * ii)) "${mz[@]}")
+		((tally[act]++))
+	done
+
+	local tally_str="${tally[@]}"
+	for ((ii = 0; ii < ${#tally[@]}; ii++)); do
+		((tally[ii] > 0))
+		check_err $? "NH #$ii not hit, tally='$tally_str'"
+	done
+
+	log_test "mpath seed $what"
+	sysctl -qw net.ipv4.fib_multipath_hash_seed=0
+}
+
+test_mpath_seed_ipv4()
+{
+	test_mpath_seed 1000 IPv4 \
+		$MZ $h1 -A 192.0.2.1 -B 192.0.2.34 -q \
+			-p 64 -d 0 -c 10 -t udp
+}
+
+test_mpath_seed_ipv6()
+{
+	test_mpath_seed 2000 IPv6 \
+		$MZ -6 $h1 -A 2001:db8:1::1 -B 2001:db8:3::2 -q \
+			-p 64 -d 0 -c 10 -t udp
+}
+
+check_mpath_seed_stability()
+{
+	local seed=$1; shift
+	local act_0=$1; shift
+	local act_1=$1; shift
+
+	((act_0 == act_1))
+	check_err $? "seed $seed: active NH moved from $act_0 to $act_1 after seed change"
+}
+
+test_mpath_seed_stability()
+{
+	local group_id=$1; shift
+	local what=$1; shift
+	local -a mz=("$@")
+
+	RET=0
+
+	local seed_0=0
+	local seed_1=3221338814
+	local seed_2=3735928559
+
+	# Initial active NH before touching the seed at all.
+	local act_ini=$(probe_nh $group_id "${mz[@]}")
+
+	local act_0_0=$(probe_seed $group_id $seed_0 "${mz[@]}")
+	local act_1_0=$(probe_seed $group_id $seed_1 "${mz[@]}")
+	local act_2_0=$(probe_seed $group_id $seed_2 "${mz[@]}")
+
+	local act_0_1=$(probe_seed $group_id $seed_0 "${mz[@]}")
+	local act_1_1=$(probe_seed $group_id $seed_1 "${mz[@]}")
+	local act_2_1=$(probe_seed $group_id $seed_2 "${mz[@]}")
+
+	check_mpath_seed_stability initial $act_ini $act_0_0
+	check_mpath_seed_stability $seed_0 $act_0_0 $act_0_1
+	check_mpath_seed_stability $seed_1 $act_1_0 $act_1_1
+	check_mpath_seed_stability $seed_2 $act_2_0 $act_2_1
+
+	log_test "mpath seed stability $what"
+	sysctl -qw net.ipv4.fib_multipath_hash_seed=0
+}
+
+test_mpath_seed_stability_ipv4()
+{
+	test_mpath_seed_stability 1000 IPv4 \
+		$MZ $h1 -A 192.0.2.1 -B 192.0.2.34 -q \
+			-p 64 -d 0 -c 10 -t udp
+}
+
+test_mpath_seed_stability_ipv6()
+{
+	test_mpath_seed_stability 2000 IPv6 \
+		$MZ -6 $h1 -A 2001:db8:1::1 -B 2001:db8:3::2 -q \
+			-p 64 -d 0 -c 10 -t udp
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+nexthops_create
+
+tests_run
+
+exit $EXIT_STATUS
-- 
cgit v1.2.3-58-ga151


From dee4dd10c79aaca192b73520d8fb64628468ae0f Mon Sep 17 00:00:00 2001
From: Elad Yifee <eladwf@gmail.com>
Date: Fri, 7 Jun 2024 11:21:50 +0300
Subject: net: ethernet: mtk_eth_soc: ppe: add support for multiple PPEs

Add the missing pieces to allow multiple PPEs units, one for each GMAC.
mtk_gdm_config has been modified to work on targted mac ID,
the inner loop moved outside of the function to allow unrelated
operations like setting the MAC's PPE index.
Introduce a sanity check in flow_offload_replace to account for
non-MTK ingress devices.
Additional field 'ppe_idx' was added to struct mtk_mac in order
to keep track on the assigned PPE unit.

Signed-off-by: Elad Yifee <eladwf@gmail.com>
Link: https://lore.kernel.org/r/20240607082155.20021-1-eladwf@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c     | 112 +++++++++++++++---------
 drivers/net/ethernet/mediatek/mtk_eth_soc.h     |   8 +-
 drivers/net/ethernet/mediatek/mtk_ppe_offload.c |  17 +++-
 3 files changed, 92 insertions(+), 45 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index c84ce54a84a0..13d78d9b3197 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -80,7 +80,9 @@ static const struct mtk_reg_map mtk_reg_map = {
 		.fq_blen	= 0x1b2c,
 	},
 	.gdm1_cnt		= 0x2400,
-	.gdma_to_ppe		= 0x4444,
+	.gdma_to_ppe	= {
+		[0]		= 0x4444,
+	},
 	.ppe_base		= 0x0c00,
 	.wdma_base = {
 		[0]		= 0x2800,
@@ -144,7 +146,10 @@ static const struct mtk_reg_map mt7986_reg_map = {
 		.tx_sch_rate	= 0x4798,
 	},
 	.gdm1_cnt		= 0x1c00,
-	.gdma_to_ppe		= 0x3333,
+	.gdma_to_ppe	= {
+		[0]		= 0x3333,
+		[1]		= 0x4444,
+	},
 	.ppe_base		= 0x2000,
 	.wdma_base = {
 		[0]		= 0x4800,
@@ -192,7 +197,11 @@ static const struct mtk_reg_map mt7988_reg_map = {
 		.tx_sch_rate	= 0x4798,
 	},
 	.gdm1_cnt		= 0x1c00,
-	.gdma_to_ppe		= 0x3333,
+	.gdma_to_ppe	= {
+		[0]		= 0x3333,
+		[1]		= 0x4444,
+		[2]		= 0xcccc,
+	},
 	.ppe_base		= 0x2000,
 	.wdma_base = {
 		[0]		= 0x4800,
@@ -2015,6 +2024,7 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget,
 	struct mtk_rx_dma_v2 *rxd, trxd;
 	int done = 0, bytes = 0;
 	dma_addr_t dma_addr = DMA_MAPPING_ERROR;
+	int ppe_idx = 0;
 
 	while (done < budget) {
 		unsigned int pktlen, *rxdcsum;
@@ -2058,6 +2068,7 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget,
 			goto release_desc;
 
 		netdev = eth->netdev[mac];
+		ppe_idx = eth->mac[mac]->ppe_idx;
 
 		if (unlikely(test_bit(MTK_RESETTING, &eth->state)))
 			goto release_desc;
@@ -2181,7 +2192,7 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget,
 		}
 
 		if (reason == MTK_PPE_CPU_REASON_HIT_UNBIND_RATE_REACHED)
-			mtk_ppe_check_skb(eth->ppe[0], skb, hash);
+			mtk_ppe_check_skb(eth->ppe[ppe_idx], skb, hash);
 
 		skb_record_rx_queue(skb, 0);
 		napi_gro_receive(napi, skb);
@@ -3276,37 +3287,27 @@ static int mtk_start_dma(struct mtk_eth *eth)
 	return 0;
 }
 
-static void mtk_gdm_config(struct mtk_eth *eth, u32 config)
+static void mtk_gdm_config(struct mtk_eth *eth, u32 id, u32 config)
 {
-	int i;
+	u32 val;
 
 	if (MTK_HAS_CAPS(eth->soc->caps, MTK_SOC_MT7628))
 		return;
 
-	for (i = 0; i < MTK_MAX_DEVS; i++) {
-		u32 val;
-
-		if (!eth->netdev[i])
-			continue;
+	val = mtk_r32(eth, MTK_GDMA_FWD_CFG(id));
 
-		val = mtk_r32(eth, MTK_GDMA_FWD_CFG(i));
+	/* default setup the forward port to send frame to PDMA */
+	val &= ~0xffff;
 
-		/* default setup the forward port to send frame to PDMA */
-		val &= ~0xffff;
+	/* Enable RX checksum */
+	val |= MTK_GDMA_ICS_EN | MTK_GDMA_TCS_EN | MTK_GDMA_UCS_EN;
 
-		/* Enable RX checksum */
-		val |= MTK_GDMA_ICS_EN | MTK_GDMA_TCS_EN | MTK_GDMA_UCS_EN;
+	val |= config;
 
-		val |= config;
+	if (eth->netdev[id] && netdev_uses_dsa(eth->netdev[id]))
+		val |= MTK_GDMA_SPECIAL_TAG;
 
-		if (netdev_uses_dsa(eth->netdev[i]))
-			val |= MTK_GDMA_SPECIAL_TAG;
-
-		mtk_w32(eth, val, MTK_GDMA_FWD_CFG(i));
-	}
-	/* Reset and enable PSE */
-	mtk_w32(eth, RST_GL_PSE, MTK_RST_GL);
-	mtk_w32(eth, 0, MTK_RST_GL);
+	mtk_w32(eth, val, MTK_GDMA_FWD_CFG(id));
 }
 
 
@@ -3366,7 +3367,10 @@ static int mtk_open(struct net_device *dev)
 {
 	struct mtk_mac *mac = netdev_priv(dev);
 	struct mtk_eth *eth = mac->hw;
-	int i, err;
+	struct mtk_mac *target_mac;
+	int i, err, ppe_num;
+
+	ppe_num = eth->soc->ppe_num;
 
 	err = phylink_of_phy_connect(mac->phylink, mac->of_node, 0);
 	if (err) {
@@ -3390,18 +3394,38 @@ static int mtk_open(struct net_device *dev)
 		for (i = 0; i < ARRAY_SIZE(eth->ppe); i++)
 			mtk_ppe_start(eth->ppe[i]);
 
-		gdm_config = soc->offload_version ? soc->reg_map->gdma_to_ppe
-						  : MTK_GDMA_TO_PDMA;
-		mtk_gdm_config(eth, gdm_config);
+		for (i = 0; i < MTK_MAX_DEVS; i++) {
+			if (!eth->netdev[i])
+				break;
+
+			target_mac = netdev_priv(eth->netdev[i]);
+			if (!soc->offload_version) {
+				target_mac->ppe_idx = 0;
+				gdm_config = MTK_GDMA_TO_PDMA;
+			} else if (ppe_num >= 3 && target_mac->id == 2) {
+				target_mac->ppe_idx = 2;
+				gdm_config = soc->reg_map->gdma_to_ppe[2];
+			} else if (ppe_num >= 2 && target_mac->id == 1) {
+				target_mac->ppe_idx = 1;
+				gdm_config = soc->reg_map->gdma_to_ppe[1];
+			} else {
+				target_mac->ppe_idx = 0;
+				gdm_config = soc->reg_map->gdma_to_ppe[0];
+			}
+			mtk_gdm_config(eth, target_mac->id, gdm_config);
+		}
+		/* Reset and enable PSE */
+		mtk_w32(eth, RST_GL_PSE, MTK_RST_GL);
+		mtk_w32(eth, 0, MTK_RST_GL);
 
 		napi_enable(&eth->tx_napi);
 		napi_enable(&eth->rx_napi);
 		mtk_tx_irq_enable(eth, MTK_TX_DONE_INT);
 		mtk_rx_irq_enable(eth, soc->rx.irq_done_mask);
 		refcount_set(&eth->dma_refcnt, 1);
-	}
-	else
+	} else {
 		refcount_inc(&eth->dma_refcnt);
+	}
 
 	phylink_start(mac->phylink);
 	netif_tx_start_all_queues(dev);
@@ -3478,7 +3502,8 @@ static int mtk_stop(struct net_device *dev)
 	if (!refcount_dec_and_test(&eth->dma_refcnt))
 		return 0;
 
-	mtk_gdm_config(eth, MTK_GDMA_DROP_ALL);
+	for (i = 0; i < MTK_MAX_DEVS; i++)
+		mtk_gdm_config(eth, i, MTK_GDMA_DROP_ALL);
 
 	mtk_tx_irq_disable(eth, MTK_TX_DONE_INT);
 	mtk_rx_irq_disable(eth, eth->soc->rx.irq_done_mask);
@@ -4959,23 +4984,24 @@ static int mtk_probe(struct platform_device *pdev)
 	}
 
 	if (eth->soc->offload_version) {
-		u32 num_ppe = mtk_is_netsys_v2_or_greater(eth) ? 2 : 1;
+		u8 ppe_num = eth->soc->ppe_num;
 
-		num_ppe = min_t(u32, ARRAY_SIZE(eth->ppe), num_ppe);
-		for (i = 0; i < num_ppe; i++) {
-			u32 ppe_addr = eth->soc->reg_map->ppe_base + i * 0x400;
+		ppe_num = min_t(u8, ARRAY_SIZE(eth->ppe), ppe_num);
+		for (i = 0; i < ppe_num; i++) {
+			u32 ppe_addr = eth->soc->reg_map->ppe_base;
 
+			ppe_addr += (i == 2 ? 0xc00 : i * 0x400);
 			eth->ppe[i] = mtk_ppe_init(eth, eth->base + ppe_addr, i);
 
 			if (!eth->ppe[i]) {
 				err = -ENOMEM;
 				goto err_deinit_ppe;
 			}
-		}
+			err = mtk_eth_offload_init(eth, i);
 
-		err = mtk_eth_offload_init(eth);
-		if (err)
-			goto err_deinit_ppe;
+			if (err)
+				goto err_deinit_ppe;
+		}
 	}
 
 	for (i = 0; i < MTK_MAX_DEVS; i++) {
@@ -5083,6 +5109,7 @@ static const struct mtk_soc_data mt7621_data = {
 	.required_pctl = false,
 	.version = 1,
 	.offload_version = 1,
+	.ppe_num = 1,
 	.hash_offset = 2,
 	.foe_entry_size = MTK_FOE_ENTRY_V1_SIZE,
 	.tx = {
@@ -5111,6 +5138,7 @@ static const struct mtk_soc_data mt7622_data = {
 	.required_pctl = false,
 	.version = 1,
 	.offload_version = 2,
+	.ppe_num = 1,
 	.hash_offset = 2,
 	.has_accounting = true,
 	.foe_entry_size = MTK_FOE_ENTRY_V1_SIZE,
@@ -5139,6 +5167,7 @@ static const struct mtk_soc_data mt7623_data = {
 	.required_pctl = true,
 	.version = 1,
 	.offload_version = 1,
+	.ppe_num = 1,
 	.hash_offset = 2,
 	.foe_entry_size = MTK_FOE_ENTRY_V1_SIZE,
 	.disable_pll_modes = true,
@@ -5194,6 +5223,7 @@ static const struct mtk_soc_data mt7981_data = {
 	.required_pctl = false,
 	.version = 2,
 	.offload_version = 2,
+	.ppe_num = 2,
 	.hash_offset = 4,
 	.has_accounting = true,
 	.foe_entry_size = MTK_FOE_ENTRY_V2_SIZE,
@@ -5223,6 +5253,7 @@ static const struct mtk_soc_data mt7986_data = {
 	.required_pctl = false,
 	.version = 2,
 	.offload_version = 2,
+	.ppe_num = 2,
 	.hash_offset = 4,
 	.has_accounting = true,
 	.foe_entry_size = MTK_FOE_ENTRY_V2_SIZE,
@@ -5252,6 +5283,7 @@ static const struct mtk_soc_data mt7988_data = {
 	.required_pctl = false,
 	.version = 3,
 	.offload_version = 2,
+	.ppe_num = 3,
 	.hash_offset = 4,
 	.has_accounting = true,
 	.foe_entry_size = MTK_FOE_ENTRY_V3_SIZE,
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
index f5174f6cb1bb..eb1708b43aa3 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
@@ -1132,7 +1132,7 @@ struct mtk_reg_map {
 		u32	tx_sch_rate;	/* tx scheduler rate control registers */
 	} qdma;
 	u32	gdm1_cnt;
-	u32	gdma_to_ppe;
+	u32	gdma_to_ppe[3];
 	u32	ppe_base;
 	u32	wdma_base[3];
 	u32	pse_iq_sta;
@@ -1170,6 +1170,7 @@ struct mtk_soc_data {
 	u8		offload_version;
 	u8		hash_offset;
 	u8		version;
+	u8		ppe_num;
 	u16		foe_entry_size;
 	netdev_features_t hw_features;
 	bool		has_accounting;
@@ -1294,7 +1295,7 @@ struct mtk_eth {
 
 	struct metadata_dst		*dsa_meta[MTK_MAX_DSA_PORTS];
 
-	struct mtk_ppe			*ppe[2];
+	struct mtk_ppe			*ppe[3];
 	struct rhashtable		flow_table;
 
 	struct bpf_prog			__rcu *prog;
@@ -1319,6 +1320,7 @@ struct mtk_eth {
 struct mtk_mac {
 	int				id;
 	phy_interface_t			interface;
+	u8				ppe_idx;
 	int				speed;
 	struct device_node		*of_node;
 	struct phylink			*phylink;
@@ -1440,7 +1442,7 @@ int mtk_gmac_sgmii_path_setup(struct mtk_eth *eth, int mac_id);
 int mtk_gmac_gephy_path_setup(struct mtk_eth *eth, int mac_id);
 int mtk_gmac_rgmii_path_setup(struct mtk_eth *eth, int mac_id);
 
-int mtk_eth_offload_init(struct mtk_eth *eth);
+int mtk_eth_offload_init(struct mtk_eth *eth, u8 id);
 int mtk_eth_setup_tc(struct net_device *dev, enum tc_setup_type type,
 		     void *type_data);
 int mtk_flow_offload_cmd(struct mtk_eth *eth, struct flow_cls_offload *cls,
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
index aa262e6f4b85..f80af73d0a1b 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
@@ -245,10 +245,10 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f,
 			 int ppe_index)
 {
 	struct flow_rule *rule = flow_cls_offload_flow_rule(f);
+	struct net_device *idev = NULL, *odev = NULL;
 	struct flow_action_entry *act;
 	struct mtk_flow_data data = {};
 	struct mtk_foe_entry foe;
-	struct net_device *odev = NULL;
 	struct mtk_flow_entry *entry;
 	int offload_type = 0;
 	int wed_index = -1;
@@ -264,6 +264,17 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f,
 		struct flow_match_meta match;
 
 		flow_rule_match_meta(rule, &match);
+		if (mtk_is_netsys_v2_or_greater(eth)) {
+			idev = __dev_get_by_index(&init_net, match.key->ingress_ifindex);
+			if (idev) {
+				struct mtk_mac *mac = netdev_priv(idev);
+
+				if (WARN_ON(mac->ppe_idx >= eth->soc->ppe_num))
+					return -EINVAL;
+
+				ppe_index = mac->ppe_idx;
+			}
+		}
 	} else {
 		return -EOPNOTSUPP;
 	}
@@ -637,7 +648,9 @@ int mtk_eth_setup_tc(struct net_device *dev, enum tc_setup_type type,
 	}
 }
 
-int mtk_eth_offload_init(struct mtk_eth *eth)
+int mtk_eth_offload_init(struct mtk_eth *eth, u8 id)
 {
+	if (!eth->ppe[id] || !eth->ppe[id]->foe_table)
+		return 0;
 	return rhashtable_init(&eth->flow_table, &mtk_flow_ht_params);
 }
-- 
cgit v1.2.3-58-ga151


From b48a1540b73a3c3a9ef59ce34176cc8180c6ce57 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sun, 9 Jun 2024 17:33:51 +0000
Subject: flow_offload: add encapsulation control flag helpers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch adds two new helper functions:
  flow_rule_is_supp_enc_control_flags()
  flow_rule_has_enc_control_flags()

They are intended to be used for validating encapsulation control
flags, and compliment the similar helpers without "enc_" in the name.

The only difference is that they have their own error message,
to make it obvious if an unsupported flag error is related to
FLOW_DISSECTOR_KEY_CONTROL or FLOW_DISSECTOR_KEY_ENC_CONTROL.

flow_rule_has_enc_control_flags() is for drivers supporting
FLOW_DISSECTOR_KEY_ENC_CONTROL, but not supporting any
encapsulation control flags.
(Currently all 4 drivers fits this category)

flow_rule_is_supp_enc_control_flags() is currently only used
for the above helper, but should also be used by drivers once
they implement at least one encapsulation control flag.

There is AFAICT currently no need for an "enc_" variant of
flow_rule_match_has_control_flags(), as all drivers currently
supporting FLOW_DISSECTOR_KEY_ENC_CONTROL, are already calling
flow_rule_match_enc_control() directly.

Only compile tested.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://lore.kernel.org/r/20240609173358.193178-2-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/flow_offload.h | 35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/include/net/flow_offload.h b/include/net/flow_offload.h
index ec9f80509f60..292cd8f4b762 100644
--- a/include/net/flow_offload.h
+++ b/include/net/flow_offload.h
@@ -471,6 +471,28 @@ static inline bool flow_rule_is_supp_control_flags(const u32 supp_flags,
 	return false;
 }
 
+/**
+ * flow_rule_is_supp_enc_control_flags() - check for supported control flags
+ * @supp_enc_flags: encapsulation control flags supported by driver
+ * @enc_ctrl_flags: encapsulation control flags present in rule
+ * @extack: The netlink extended ACK for reporting errors.
+ *
+ * Return: true if only supported control flags are set, false otherwise.
+ */
+static inline bool flow_rule_is_supp_enc_control_flags(const u32 supp_enc_flags,
+						       const u32 enc_ctrl_flags,
+						       struct netlink_ext_ack *extack)
+{
+	if (likely((enc_ctrl_flags & ~supp_enc_flags) == 0))
+		return true;
+
+	NL_SET_ERR_MSG_FMT_MOD(extack,
+			       "Unsupported match on enc_control.flags %#x",
+			       enc_ctrl_flags);
+
+	return false;
+}
+
 /**
  * flow_rule_has_control_flags() - check for presence of any control flags
  * @ctrl_flags: control flags present in rule
@@ -484,6 +506,19 @@ static inline bool flow_rule_has_control_flags(const u32 ctrl_flags,
 	return !flow_rule_is_supp_control_flags(0, ctrl_flags, extack);
 }
 
+/**
+ * flow_rule_has_enc_control_flags() - check for presence of any control flags
+ * @enc_ctrl_flags: encapsulation control flags present in rule
+ * @extack: The netlink extended ACK for reporting errors.
+ *
+ * Return: true if control flags are set, false otherwise.
+ */
+static inline bool flow_rule_has_enc_control_flags(const u32 enc_ctrl_flags,
+						   struct netlink_ext_ack *extack)
+{
+	return !flow_rule_is_supp_enc_control_flags(0, enc_ctrl_flags, extack);
+}
+
 /**
  * flow_rule_match_has_control_flags() - match and check for any control flags
  * @rule: The flow_rule under evaluation.
-- 
cgit v1.2.3-58-ga151


From 2ede54f8786f6abae25eec41dd821259484be586 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sun, 9 Jun 2024 17:33:52 +0000
Subject: sfc: use flow_rule_is_supp_enc_control_flags()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Change the existing check for unsupported encapsulation control flags,
to use the new helper flow_rule_is_supp_enc_control_flags().

No functional change, only compile tested.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Acked-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://lore.kernel.org/r/20240609173358.193178-3-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/tc.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/sfc/tc.c b/drivers/net/ethernet/sfc/tc.c
index 9d140203e273..0d93164988fc 100644
--- a/drivers/net/ethernet/sfc/tc.c
+++ b/drivers/net/ethernet/sfc/tc.c
@@ -387,11 +387,8 @@ static int efx_tc_flower_parse_match(struct efx_nic *efx,
 		struct flow_match_control fm;
 
 		flow_rule_match_enc_control(rule, &fm);
-		if (fm.mask->flags) {
-			NL_SET_ERR_MSG_FMT_MOD(extack, "Unsupported match on enc_control.flags %#x",
-					       fm.mask->flags);
+		if (flow_rule_has_enc_control_flags(fm.mask->flags, extack))
 			return -EOPNOTSUPP;
-		}
 		if (!IS_ALL_ONES(fm.mask->addr_type)) {
 			NL_SET_ERR_MSG_FMT_MOD(extack, "Unsupported enc addr_type mask %u (key %u)",
 					       fm.mask->addr_type,
-- 
cgit v1.2.3-58-ga151


From 28d19ec9175534a6f4b38a042c3b83baf8563a8c Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sun, 9 Jun 2024 17:33:53 +0000
Subject: net/mlx5e: flower: validate encapsulation control flags
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Encapsulation control flags are currently not used anywhere,
so all flags are currently unsupported by all drivers.

This patch adds validation of this assumption, so that
encapsulation flags may be used in the future.

In case any encapsulation control flags are masked,
flow_rule_match_has_enc_control_flags() sets a NL extended
error message, and we return -EOPNOTSUPP.

Only compile tested.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://lore.kernel.org/r/20240609173358.193178-4-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
index 8dfb57f712b0..721f35e59757 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
@@ -850,6 +850,12 @@ int mlx5e_tc_tun_parse(struct net_device *filter_dev,
 		flow_rule_match_enc_control(rule, &match);
 		addr_type = match.key->addr_type;
 
+		if (flow_rule_has_enc_control_flags(match.mask->flags,
+						    extack)) {
+			err = -EOPNOTSUPP;
+			goto out;
+		}
+
 		/* For tunnel addr_type used same key id`s as for non-tunnel */
 		if (addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
 			struct flow_match_ipv4_addrs match;
-- 
cgit v1.2.3-58-ga151


From 34cdd984782039c924b0476893202b37f6e74742 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sun, 9 Jun 2024 17:33:54 +0000
Subject: nfp: flower: validate encapsulation control flags
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Encapsulation control flags are currently not used anywhere,
so all flags are currently unsupported by all drivers.

This patch adds validation of this assumption, so that
encapsulation flags may be used in the future.

In case any encapsulation control flags are masked,
flow_rule_match_has_enc_control_flags() sets a NL extended
error message, and we return -EOPNOTSUPP.

Only compile tested.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://lore.kernel.org/r/20240609173358.193178-5-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/netronome/nfp/flower/offload.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c
index 8e0a890381b6..46ffc2c20893 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/offload.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c
@@ -321,6 +321,10 @@ nfp_flower_calculate_key_layers(struct nfp_app *app,
 
 		flow_rule_match_enc_control(rule, &enc_ctl);
 
+		if (flow_rule_has_enc_control_flags(enc_ctl.mask->flags,
+						    extack))
+			return -EOPNOTSUPP;
+
 		if (enc_ctl.mask->addr_type != 0xffff) {
 			NL_SET_ERR_MSG_MOD(extack, "unsupported offload: wildcarded protocols on tunnels are not supported");
 			return -EOPNOTSUPP;
-- 
cgit v1.2.3-58-ga151


From 5a1b015d521d7e8d7e343c04a292151ad5de3611 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sun, 9 Jun 2024 17:33:55 +0000
Subject: ice: flower: validate encapsulation control flags
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Encapsulation control flags are currently not used anywhere,
so all flags are currently unsupported by all drivers.

This patch adds validation of this assumption, so that
encapsulation flags may be used in the future.

In case any encapsulation control flags are masked,
flow_rule_match_has_enc_control_flags() sets a NL extended
error message, and we return -EOPNOTSUPP.

Only compile tested.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Reviewed-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://lore.kernel.org/r/20240609173358.193178-6-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_tc_lib.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/intel/ice/ice_tc_lib.c b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
index 8bd24b33f3a6..e6923f8121a9 100644
--- a/drivers/net/ethernet/intel/ice/ice_tc_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
@@ -1353,6 +1353,7 @@ ice_parse_tunnel_attr(struct net_device *dev, struct flow_rule *rule,
 		      struct ice_tc_flower_fltr *fltr)
 {
 	struct ice_tc_flower_lyr_2_4_hdrs *headers = &fltr->outer_headers;
+	struct netlink_ext_ack *extack = fltr->extack;
 	struct flow_match_control enc_control;
 
 	fltr->tunnel_type = ice_tc_tun_get_type(dev);
@@ -1373,6 +1374,9 @@ ice_parse_tunnel_attr(struct net_device *dev, struct flow_rule *rule,
 
 	flow_rule_match_enc_control(rule, &enc_control);
 
+	if (flow_rule_has_enc_control_flags(enc_control.mask->flags, extack))
+		return -EOPNOTSUPP;
+
 	if (enc_control.key->addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
 		struct flow_match_ipv4_addrs match;
 
-- 
cgit v1.2.3-58-ga151


From 3ec8d7572a69d142d49f52b28ce8d84e5fef9131 Mon Sep 17 00:00:00 2001
From: Ole André Vadla Ravnås <oleavr@frida.re>
Date: Fri, 7 Jun 2024 09:40:17 +0200
Subject: CDC-NCM: add support for Apple's private interface
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Available on iOS/iPadOS >= 17, where this new interface is used by
developer tools using the new RemoteXPC protocol.

This private interface lacks a status endpoint, presumably because there
isn't a physical cable that can be unplugged, nor any speed changes to
be notified about.

Note that NCM interfaces are not exposed until a mode switch is
requested, which macOS does automatically.

The mode switch can be performed like this:

        uint8_t status;
        libusb_control_transfer(device_handle,
                LIBUSB_RECIPIENT_DEVICE | LIBUSB_REQUEST_TYPE_VENDOR |
                LIBUSB_ENDPOINT_IN,
                82, /* bRequest */
                0,  /* wValue   */
                3,  /* wIndex   */
                &status,
                sizeof(status),
                0);

Newer versions of usbmuxd do this automatically.

Co-developed-by: Håvard Sørbø <havard@hsorbo.no>
Signed-off-by: Håvard Sørbø <havard@hsorbo.no>
Signed-off-by: Ole André Vadla Ravnås <oleavr@frida.re>
Link: https://lore.kernel.org/r/20240607074117.31322-1-oleavr@frida.re
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/usb/cdc_ncm.c | 47 ++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 46 insertions(+), 1 deletion(-)

diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c
index bf76ecccc2e6..d5c47a2a62dc 100644
--- a/drivers/net/usb/cdc_ncm.c
+++ b/drivers/net/usb/cdc_ncm.c
@@ -933,7 +933,8 @@ int cdc_ncm_bind_common(struct usbnet *dev, struct usb_interface *intf, u8 data_
 
 	cdc_ncm_find_endpoints(dev, ctx->data);
 	cdc_ncm_find_endpoints(dev, ctx->control);
-	if (!dev->in || !dev->out || !dev->status) {
+	if (!dev->in || !dev->out ||
+	    (!dev->status && dev->driver_info->flags & FLAG_LINK_INTR)) {
 		dev_dbg(&intf->dev, "failed to collect endpoints\n");
 		goto error2;
 	}
@@ -1925,6 +1926,34 @@ static const struct driver_info cdc_ncm_zlp_info = {
 	.set_rx_mode = usbnet_cdc_update_filter,
 };
 
+/* Same as cdc_ncm_info, but with FLAG_SEND_ZLP */
+static const struct driver_info apple_tethering_interface_info = {
+	.description = "CDC NCM (Apple Tethering)",
+	.flags = FLAG_POINTTOPOINT | FLAG_NO_SETINT | FLAG_MULTI_PACKET
+			| FLAG_LINK_INTR | FLAG_ETHER | FLAG_SEND_ZLP,
+	.bind = cdc_ncm_bind,
+	.unbind = cdc_ncm_unbind,
+	.manage_power = usbnet_manage_power,
+	.status = cdc_ncm_status,
+	.rx_fixup = cdc_ncm_rx_fixup,
+	.tx_fixup = cdc_ncm_tx_fixup,
+	.set_rx_mode = usbnet_cdc_update_filter,
+};
+
+/* Same as apple_tethering_interface_info, but without FLAG_LINK_INTR */
+static const struct driver_info apple_private_interface_info = {
+	.description = "CDC NCM (Apple Private)",
+	.flags = FLAG_POINTTOPOINT | FLAG_NO_SETINT | FLAG_MULTI_PACKET
+			| FLAG_ETHER | FLAG_SEND_ZLP,
+	.bind = cdc_ncm_bind,
+	.unbind = cdc_ncm_unbind,
+	.manage_power = usbnet_manage_power,
+	.status = cdc_ncm_status,
+	.rx_fixup = cdc_ncm_rx_fixup,
+	.tx_fixup = cdc_ncm_tx_fixup,
+	.set_rx_mode = usbnet_cdc_update_filter,
+};
+
 /* Same as cdc_ncm_info, but with FLAG_WWAN */
 static const struct driver_info wwan_info = {
 	.description = "Mobile Broadband Network Device",
@@ -1954,6 +1983,22 @@ static const struct driver_info wwan_noarp_info = {
 };
 
 static const struct usb_device_id cdc_devs[] = {
+	/* iPhone */
+	{ USB_DEVICE_INTERFACE_NUMBER(0x05ac, 0x12a8, 2),
+		.driver_info = (unsigned long)&apple_tethering_interface_info,
+	},
+	{ USB_DEVICE_INTERFACE_NUMBER(0x05ac, 0x12a8, 4),
+		.driver_info = (unsigned long)&apple_private_interface_info,
+	},
+
+	/* iPad */
+	{ USB_DEVICE_INTERFACE_NUMBER(0x05ac, 0x12ab, 2),
+		.driver_info = (unsigned long)&apple_tethering_interface_info,
+	},
+	{ USB_DEVICE_INTERFACE_NUMBER(0x05ac, 0x12ab, 4),
+		.driver_info = (unsigned long)&apple_private_interface_info,
+	},
+
 	/* Ericsson MBM devices like F5521gw */
 	{ .match_flags = USB_DEVICE_ID_MATCH_INT_INFO
 		| USB_DEVICE_ID_MATCH_VENDOR,
-- 
cgit v1.2.3-58-ga151


From a3cfe84cca28f205761a0450016593b0d728165e Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadfed@meta.com>
Date: Thu, 6 Jun 2024 07:58:50 -0700
Subject: bpf: Add CHECKSUM_COMPLETE to bpf test progs

Add special flag to validate that TC BPF program properly updates
checksum information in skb.

Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240606145851.229116-1-vadfed@meta.com
---
 include/uapi/linux/bpf.h       |  2 ++
 net/bpf/test_run.c             | 28 +++++++++++++++++++++++++++-
 tools/include/uapi/linux/bpf.h |  2 ++
 3 files changed, 31 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 25ea393cf084..35bcf52dbc65 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -1425,6 +1425,8 @@ enum {
 #define BPF_F_TEST_RUN_ON_CPU	(1U << 0)
 /* If set, XDP frames will be transmitted after processing */
 #define BPF_F_TEST_XDP_LIVE_FRAMES	(1U << 1)
+/* If set, apply CHECKSUM_COMPLETE to skb and validate the checksum */
+#define BPF_F_TEST_SKB_CHECKSUM_COMPLETE	(1U << 2)
 
 /* type for BPF_ENABLE_STATS */
 enum bpf_stats_type {
diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 36ae54f57bf5..3c965e32fc33 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -983,7 +983,8 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr,
 	void *data;
 	int ret;
 
-	if (kattr->test.flags || kattr->test.cpu || kattr->test.batch_size)
+	if ((kattr->test.flags & ~BPF_F_TEST_SKB_CHECKSUM_COMPLETE) ||
+	    kattr->test.cpu || kattr->test.batch_size)
 		return -EINVAL;
 
 	data = bpf_test_init(kattr, kattr->test.data_size_in,
@@ -1031,6 +1032,7 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr,
 
 	skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN);
 	__skb_put(skb, size);
+
 	if (ctx && ctx->ifindex > 1) {
 		dev = dev_get_by_index(net, ctx->ifindex);
 		if (!dev) {
@@ -1066,9 +1068,19 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr,
 		__skb_push(skb, hh_len);
 	if (is_direct_pkt_access)
 		bpf_compute_data_pointers(skb);
+
 	ret = convert___skb_to_skb(skb, ctx);
 	if (ret)
 		goto out;
+
+	if (kattr->test.flags & BPF_F_TEST_SKB_CHECKSUM_COMPLETE) {
+		const int off = skb_network_offset(skb);
+		int len = skb->len - off;
+
+		skb->csum = skb_checksum(skb, off, len, 0);
+		skb->ip_summed = CHECKSUM_COMPLETE;
+	}
+
 	ret = bpf_test_run(prog, skb, repeat, &retval, &duration, false);
 	if (ret)
 		goto out;
@@ -1083,6 +1095,20 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr,
 		}
 		memset(__skb_push(skb, hh_len), 0, hh_len);
 	}
+
+	if (kattr->test.flags & BPF_F_TEST_SKB_CHECKSUM_COMPLETE) {
+		const int off = skb_network_offset(skb);
+		int len = skb->len - off;
+		__wsum csum;
+
+		csum = skb_checksum(skb, off, len, 0);
+
+		if (csum_fold(skb->csum) != csum_fold(csum)) {
+			ret = -EBADMSG;
+			goto out;
+		}
+	}
+
 	convert_skb_to___skb(skb, ctx);
 
 	size = skb->len;
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 25ea393cf084..35bcf52dbc65 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -1425,6 +1425,8 @@ enum {
 #define BPF_F_TEST_RUN_ON_CPU	(1U << 0)
 /* If set, XDP frames will be transmitted after processing */
 #define BPF_F_TEST_XDP_LIVE_FRAMES	(1U << 1)
+/* If set, apply CHECKSUM_COMPLETE to skb and validate the checksum */
+#define BPF_F_TEST_SKB_CHECKSUM_COMPLETE	(1U << 2)
 
 /* type for BPF_ENABLE_STATS */
 enum bpf_stats_type {
-- 
cgit v1.2.3-58-ga151


From 041c1dc988fdffd5eb0c13f1ce5d1b3b0125f208 Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadfed@meta.com>
Date: Thu, 6 Jun 2024 07:58:51 -0700
Subject: selftests/bpf: Validate CHECKSUM_COMPLETE option

Adjust skb program test to run with checksum validation.

Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240606145851.229116-2-vadfed@meta.com
---
 tools/testing/selftests/bpf/prog_tests/test_skb_pkt_end.c |  1 +
 tools/testing/selftests/bpf/progs/skb_pkt_end.c           | 11 ++++++++++-
 2 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/test_skb_pkt_end.c b/tools/testing/selftests/bpf/prog_tests/test_skb_pkt_end.c
index ae93411fd582..09ca13bdf6ca 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_skb_pkt_end.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_skb_pkt_end.c
@@ -11,6 +11,7 @@ static int sanity_run(struct bpf_program *prog)
 		.data_in = &pkt_v4,
 		.data_size_in = sizeof(pkt_v4),
 		.repeat = 1,
+		.flags = BPF_F_TEST_SKB_CHECKSUM_COMPLETE,
 	);
 
 	prog_fd = bpf_program__fd(prog);
diff --git a/tools/testing/selftests/bpf/progs/skb_pkt_end.c b/tools/testing/selftests/bpf/progs/skb_pkt_end.c
index db4abd2682fc..3bb4451524a1 100644
--- a/tools/testing/selftests/bpf/progs/skb_pkt_end.c
+++ b/tools/testing/selftests/bpf/progs/skb_pkt_end.c
@@ -33,6 +33,8 @@ int main_prog(struct __sk_buff *skb)
 	struct iphdr *ip = NULL;
 	struct tcphdr *tcp;
 	__u8 proto = 0;
+	int urg_ptr;
+	u32 offset;
 
 	if (!(ip = get_iphdr(skb)))
 		goto out;
@@ -48,7 +50,14 @@ int main_prog(struct __sk_buff *skb)
 	if (!tcp)
 		goto out;
 
-	return tcp->urg_ptr;
+	urg_ptr = tcp->urg_ptr;
+
+	/* Checksum validation part */
+	proto++;
+	offset = sizeof(struct ethhdr) + offsetof(struct iphdr, protocol);
+	bpf_skb_store_bytes(skb, offset, &proto, sizeof(proto), BPF_F_RECOMPUTE_CSUM);
+
+	return urg_ptr;
 out:
 	return -1;
 }
-- 
cgit v1.2.3-58-ga151


From 78746f93e903d022c692b9bb3a3e2570167b2dc2 Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Thu, 13 Jun 2024 10:19:25 -0600
Subject: bpf: Fix bpf_dynptr documentation comments

The function argument names were changed but the doc comment was not.
Fix htmldocs build warning by updating doc comments.

Fixes: cce4c40b9606 ("bpf: treewide: Align kfunc signatures to prog point-of-view")
Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/d0b0eb05f91e12e5795966153b11998d3fc1d433.1718295425.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/helpers.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 3ac521c48bba..229396172026 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2433,7 +2433,7 @@ __bpf_kfunc struct task_struct *bpf_task_from_pid(s32 pid)
 
 /**
  * bpf_dynptr_slice() - Obtain a read-only pointer to the dynptr data.
- * @ptr: The dynptr whose data slice to retrieve
+ * @p: The dynptr whose data slice to retrieve
  * @offset: Offset into the dynptr
  * @buffer__opt: User-provided buffer to copy contents into.  May be NULL
  * @buffer__szk: Size (in bytes) of the buffer if present. This is the
@@ -2504,7 +2504,7 @@ __bpf_kfunc void *bpf_dynptr_slice(const struct bpf_dynptr *p, u32 offset,
 
 /**
  * bpf_dynptr_slice_rdwr() - Obtain a writable pointer to the dynptr data.
- * @ptr: The dynptr whose data slice to retrieve
+ * @p: The dynptr whose data slice to retrieve
  * @offset: Offset into the dynptr
  * @buffer__opt: User-provided buffer to copy contents into. May be NULL
  * @buffer__szk: Size (in bytes) of the buffer if present. This is the
-- 
cgit v1.2.3-58-ga151


From 6a8260147745fe493d733d4e5f9b327da3720905 Mon Sep 17 00:00:00 2001
From: Daniel Xu <dxu@dxuuu.xyz>
Date: Thu, 13 Jun 2024 10:19:26 -0600
Subject: bpf: selftests: Do not use generated kfunc prototypes for arena progs

When selftests are built with a new enough clang, the arena selftests
opt-in to use LLVM address_space attribute annotations for arena
pointers.

These annotations are not emitted by kfunc prototype generation. This
causes compilation errors when clang sees conflicting prototypes.

Fix by opting arena selftests out of using generated kfunc prototypes.

Fixes: 770abbb5a25a ("bpftool: Support dumping kfunc prototypes from BTF")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/r/202406131810.c1B8hTm8-lkp@intel.com/
Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/fc59a617439ceea9ad8dfbb4786843c2169496ae.1718295425.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/arena_htab.c           | 1 +
 tools/testing/selftests/bpf/progs/arena_list.c           | 1 +
 tools/testing/selftests/bpf/progs/verifier_arena.c       | 1 +
 tools/testing/selftests/bpf/progs/verifier_arena_large.c | 1 +
 4 files changed, 4 insertions(+)

diff --git a/tools/testing/selftests/bpf/progs/arena_htab.c b/tools/testing/selftests/bpf/progs/arena_htab.c
index 1e6ac187a6a0..cf938cf8c11e 100644
--- a/tools/testing/selftests/bpf/progs/arena_htab.c
+++ b/tools/testing/selftests/bpf/progs/arena_htab.c
@@ -1,5 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2024 Meta Platforms, Inc. and affiliates. */
+#define BPF_NO_KFUNC_PROTOTYPES
 #include <vmlinux.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
diff --git a/tools/testing/selftests/bpf/progs/arena_list.c b/tools/testing/selftests/bpf/progs/arena_list.c
index 93bd0600eba0..3a2ddcacbea6 100644
--- a/tools/testing/selftests/bpf/progs/arena_list.c
+++ b/tools/testing/selftests/bpf/progs/arena_list.c
@@ -1,5 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2024 Meta Platforms, Inc. and affiliates. */
+#define BPF_NO_KFUNC_PROTOTYPES
 #include <vmlinux.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
diff --git a/tools/testing/selftests/bpf/progs/verifier_arena.c b/tools/testing/selftests/bpf/progs/verifier_arena.c
index 93144ae6df74..67509c5d3982 100644
--- a/tools/testing/selftests/bpf/progs/verifier_arena.c
+++ b/tools/testing/selftests/bpf/progs/verifier_arena.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2024 Meta Platforms, Inc. and affiliates. */
 
+#define BPF_NO_KFUNC_PROTOTYPES
 #include <vmlinux.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
diff --git a/tools/testing/selftests/bpf/progs/verifier_arena_large.c b/tools/testing/selftests/bpf/progs/verifier_arena_large.c
index ef66ea460264..6065f862d964 100644
--- a/tools/testing/selftests/bpf/progs/verifier_arena_large.c
+++ b/tools/testing/selftests/bpf/progs/verifier_arena_large.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2024 Meta Platforms, Inc. and affiliates. */
 
+#define BPF_NO_KFUNC_PROTOTYPES
 #include <vmlinux.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
-- 
cgit v1.2.3-58-ga151


From a90797993afcb0eaf6bf47a062ff47eb3810a6d5 Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadfed@meta.com>
Date: Thu, 13 Jun 2024 14:18:13 -0700
Subject: bpf: verifier: make kfuncs args nullalble

Some arguments to kfuncs might be NULL in some cases. But currently it's
not possible to pass NULL to any BTF structures because the check for
the suffix is located after all type checks. Move it to earlier place
to allow nullable args.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Link: https://lore.kernel.org/r/20240613211817.1551967-2-vadfed@meta.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index acc9dd830807..e857b08e1f2d 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -11187,6 +11187,9 @@ get_kfunc_ptr_arg_type(struct bpf_verifier_env *env,
 	if (btf_is_prog_ctx_type(&env->log, meta->btf, t, resolve_prog_type(env->prog), argno))
 		return KF_ARG_PTR_TO_CTX;
 
+	if (is_kfunc_arg_nullable(meta->btf, &args[argno]) && register_is_null(reg))
+		return KF_ARG_PTR_TO_NULL;
+
 	if (is_kfunc_arg_alloc_obj(meta->btf, &args[argno]))
 		return KF_ARG_PTR_TO_ALLOC_BTF_ID;
 
@@ -11232,9 +11235,6 @@ get_kfunc_ptr_arg_type(struct bpf_verifier_env *env,
 	if (is_kfunc_arg_callback(env, meta->btf, &args[argno]))
 		return KF_ARG_PTR_TO_CALLBACK;
 
-	if (is_kfunc_arg_nullable(meta->btf, &args[argno]) && register_is_null(reg))
-		return KF_ARG_PTR_TO_NULL;
-
 	if (argno + 1 < nargs &&
 	    (is_kfunc_arg_mem_size(meta->btf, &args[argno + 1], &regs[regno + 1]) ||
 	     is_kfunc_arg_const_mem_size(meta->btf, &args[argno + 1], &regs[regno + 1])))
-- 
cgit v1.2.3-58-ga151


From 65d6d61d25968d1f13a478a6f303ed8d6b978a77 Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadfed@meta.com>
Date: Thu, 13 Jun 2024 14:18:14 -0700
Subject: bpf: crypto: make state and IV dynptr nullable

Some ciphers do not require state and IV buffer, but with current
implementation 0-sized dynptr is always needed. With adjustment to
verifier we can provide NULL instead of 0-sized dynptr. Make crypto
kfuncs ready for this.

Reviewed-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Link: https://lore.kernel.org/r/20240613211817.1551967-3-vadfed@meta.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/crypto.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/kernel/bpf/crypto.c b/kernel/bpf/crypto.c
index 3c1de0e5c0bd..94854cd9c4cc 100644
--- a/kernel/bpf/crypto.c
+++ b/kernel/bpf/crypto.c
@@ -275,7 +275,7 @@ static int bpf_crypto_crypt(const struct bpf_crypto_ctx *ctx,
 	if (__bpf_dynptr_is_rdonly(dst))
 		return -EINVAL;
 
-	siv_len = __bpf_dynptr_size(siv);
+	siv_len = siv ? __bpf_dynptr_size(siv) : 0;
 	src_len = __bpf_dynptr_size(src);
 	dst_len = __bpf_dynptr_size(dst);
 	if (!src_len || !dst_len)
@@ -303,42 +303,42 @@ static int bpf_crypto_crypt(const struct bpf_crypto_ctx *ctx,
 
 /**
  * bpf_crypto_decrypt() - Decrypt buffer using configured context and IV provided.
- * @ctx:	The crypto context being used. The ctx must be a trusted pointer.
- * @src:	bpf_dynptr to the encrypted data. Must be a trusted pointer.
- * @dst:	bpf_dynptr to the buffer where to store the result. Must be a trusted pointer.
- * @siv:	bpf_dynptr to IV data and state data to be used by decryptor.
+ * @ctx:		The crypto context being used. The ctx must be a trusted pointer.
+ * @src:		bpf_dynptr to the encrypted data. Must be a trusted pointer.
+ * @dst:		bpf_dynptr to the buffer where to store the result. Must be a trusted pointer.
+ * @siv__nullable:	bpf_dynptr to IV data and state data to be used by decryptor. May be NULL.
  *
  * Decrypts provided buffer using IV data and the crypto context. Crypto context must be configured.
  */
 __bpf_kfunc int bpf_crypto_decrypt(struct bpf_crypto_ctx *ctx,
 				   const struct bpf_dynptr *src,
 				   const struct bpf_dynptr *dst,
-				   const struct bpf_dynptr *siv)
+				   const struct bpf_dynptr *siv__nullable)
 {
 	const struct bpf_dynptr_kern *src_kern = (struct bpf_dynptr_kern *)src;
 	const struct bpf_dynptr_kern *dst_kern = (struct bpf_dynptr_kern *)dst;
-	const struct bpf_dynptr_kern *siv_kern = (struct bpf_dynptr_kern *)siv;
+	const struct bpf_dynptr_kern *siv_kern = (struct bpf_dynptr_kern *)siv__nullable;
 
 	return bpf_crypto_crypt(ctx, src_kern, dst_kern, siv_kern, true);
 }
 
 /**
  * bpf_crypto_encrypt() - Encrypt buffer using configured context and IV provided.
- * @ctx:	The crypto context being used. The ctx must be a trusted pointer.
- * @src:	bpf_dynptr to the plain data. Must be a trusted pointer.
- * @dst:	bpf_dynptr to buffer where to store the result. Must be a trusted pointer.
- * @siv:	bpf_dynptr to IV data and state data to be used by decryptor.
+ * @ctx:		The crypto context being used. The ctx must be a trusted pointer.
+ * @src:		bpf_dynptr to the plain data. Must be a trusted pointer.
+ * @dst:		bpf_dynptr to the buffer where to store the result. Must be a trusted pointer.
+ * @siv__nullable:	bpf_dynptr to IV data and state data to be used by decryptor. May be NULL.
  *
  * Encrypts provided buffer using IV data and the crypto context. Crypto context must be configured.
  */
 __bpf_kfunc int bpf_crypto_encrypt(struct bpf_crypto_ctx *ctx,
 				   const struct bpf_dynptr *src,
 				   const struct bpf_dynptr *dst,
-				   const struct bpf_dynptr *siv)
+				   const struct bpf_dynptr *siv__nullable)
 {
 	const struct bpf_dynptr_kern *src_kern = (struct bpf_dynptr_kern *)src;
 	const struct bpf_dynptr_kern *dst_kern = (struct bpf_dynptr_kern *)dst;
-	const struct bpf_dynptr_kern *siv_kern = (struct bpf_dynptr_kern *)siv;
+	const struct bpf_dynptr_kern *siv_kern = (struct bpf_dynptr_kern *)siv__nullable;
 
 	return bpf_crypto_crypt(ctx, src_kern, dst_kern, siv_kern, false);
 }
-- 
cgit v1.2.3-58-ga151


From 9363dc8ddc4e222c4259013ae5428070712910b9 Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadfed@meta.com>
Date: Thu, 13 Jun 2024 14:18:15 -0700
Subject: selftests: bpf: crypto: use NULL instead of 0-sized dynptr

Adjust selftests to use nullable option for state and IV arg.

Reviewed-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Link: https://lore.kernel.org/r/20240613211817.1551967-4-vadfed@meta.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/crypto_sanity.c | 16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/crypto_sanity.c b/tools/testing/selftests/bpf/progs/crypto_sanity.c
index 1be0a3fa5efd..645be6cddf36 100644
--- a/tools/testing/selftests/bpf/progs/crypto_sanity.c
+++ b/tools/testing/selftests/bpf/progs/crypto_sanity.c
@@ -89,7 +89,7 @@ int decrypt_sanity(struct __sk_buff *skb)
 {
 	struct __crypto_ctx_value *v;
 	struct bpf_crypto_ctx *ctx;
-	struct bpf_dynptr psrc, pdst, iv;
+	struct bpf_dynptr psrc, pdst;
 	int err;
 
 	err = skb_dynptr_validate(skb, &psrc);
@@ -114,12 +114,8 @@ int decrypt_sanity(struct __sk_buff *skb)
 	 * production code, a percpu map should be used to store the result.
 	 */
 	bpf_dynptr_from_mem(dst, sizeof(dst), 0, &pdst);
-	/* iv dynptr has to be initialized with 0 size, but proper memory region
-	 * has to be provided anyway
-	 */
-	bpf_dynptr_from_mem(dst, 0, 0, &iv);
 
-	status = bpf_crypto_decrypt(ctx, &psrc, &pdst, &iv);
+	status = bpf_crypto_decrypt(ctx, &psrc, &pdst, NULL);
 
 	return TC_ACT_SHOT;
 }
@@ -129,7 +125,7 @@ int encrypt_sanity(struct __sk_buff *skb)
 {
 	struct __crypto_ctx_value *v;
 	struct bpf_crypto_ctx *ctx;
-	struct bpf_dynptr psrc, pdst, iv;
+	struct bpf_dynptr psrc, pdst;
 	int err;
 
 	status = 0;
@@ -156,12 +152,8 @@ int encrypt_sanity(struct __sk_buff *skb)
 	 * production code, a percpu map should be used to store the result.
 	 */
 	bpf_dynptr_from_mem(dst, sizeof(dst), 0, &pdst);
-	/* iv dynptr has to be initialized with 0 size, but proper memory region
-	 * has to be provided anyway
-	 */
-	bpf_dynptr_from_mem(dst, 0, 0, &iv);
 
-	status = bpf_crypto_encrypt(ctx, &psrc, &pdst, &iv);
+	status = bpf_crypto_encrypt(ctx, &psrc, &pdst, NULL);
 
 	return TC_ACT_SHOT;
 }
-- 
cgit v1.2.3-58-ga151


From 9b560751f75f7b2484fa22c781be68f4f9fec2b0 Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadfed@meta.com>
Date: Thu, 13 Jun 2024 14:18:16 -0700
Subject: selftests: bpf: crypto: adjust bench to use nullable IV
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The bench shows some improvements, around 4% faster on decrypt.

Before:

Benchmark 'crypto-decrypt' started.
Iter   0 (325.719us): hits    5.105M/s (  5.105M/prod), drops 0.000M/s, total operations    5.105M/s
Iter   1 (-17.295us): hits    5.224M/s (  5.224M/prod), drops 0.000M/s, total operations    5.224M/s
Iter   2 (  5.504us): hits    4.630M/s (  4.630M/prod), drops 0.000M/s, total operations    4.630M/s
Iter   3 (  9.239us): hits    5.148M/s (  5.148M/prod), drops 0.000M/s, total operations    5.148M/s
Iter   4 ( 37.885us): hits    5.198M/s (  5.198M/prod), drops 0.000M/s, total operations    5.198M/s
Iter   5 (-53.282us): hits    5.167M/s (  5.167M/prod), drops 0.000M/s, total operations    5.167M/s
Iter   6 (-17.809us): hits    5.186M/s (  5.186M/prod), drops 0.000M/s, total operations    5.186M/s
Summary: hits    5.092 ± 0.228M/s (  5.092M/prod), drops    0.000 ±0.000M/s, total operations    5.092 ± 0.228M/s

After:

Benchmark 'crypto-decrypt' started.
Iter   0 (268.912us): hits    5.312M/s (  5.312M/prod), drops 0.000M/s, total operations    5.312M/s
Iter   1 (124.869us): hits    5.354M/s (  5.354M/prod), drops 0.000M/s, total operations    5.354M/s
Iter   2 (-36.801us): hits    5.334M/s (  5.334M/prod), drops 0.000M/s, total operations    5.334M/s
Iter   3 (254.628us): hits    5.334M/s (  5.334M/prod), drops 0.000M/s, total operations    5.334M/s
Iter   4 (-77.691us): hits    5.275M/s (  5.275M/prod), drops 0.000M/s, total operations    5.275M/s
Iter   5 (-164.510us): hits    5.313M/s (  5.313M/prod), drops 0.000M/s, total operations    5.313M/s
Iter   6 (-81.376us): hits    5.346M/s (  5.346M/prod), drops 0.000M/s, total operations    5.346M/s
Summary: hits    5.326 ± 0.029M/s (  5.326M/prod), drops    0.000 ±0.000M/s, total operations    5.326 ± 0.029M/s

Reviewed-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Link: https://lore.kernel.org/r/20240613211817.1551967-5-vadfed@meta.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/crypto_bench.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/crypto_bench.c b/tools/testing/selftests/bpf/progs/crypto_bench.c
index e61fe0882293..4ac956b26240 100644
--- a/tools/testing/selftests/bpf/progs/crypto_bench.c
+++ b/tools/testing/selftests/bpf/progs/crypto_bench.c
@@ -57,7 +57,7 @@ int crypto_encrypt(struct __sk_buff *skb)
 {
 	struct __crypto_ctx_value *v;
 	struct bpf_crypto_ctx *ctx;
-	struct bpf_dynptr psrc, pdst, iv;
+	struct bpf_dynptr psrc, pdst;
 
 	v = crypto_ctx_value_lookup();
 	if (!v) {
@@ -73,9 +73,8 @@ int crypto_encrypt(struct __sk_buff *skb)
 
 	bpf_dynptr_from_skb(skb, 0, &psrc);
 	bpf_dynptr_from_mem(dst, len, 0, &pdst);
-	bpf_dynptr_from_mem(dst, 0, 0, &iv);
 
-	status = bpf_crypto_encrypt(ctx, &psrc, &pdst, &iv);
+	status = bpf_crypto_encrypt(ctx, &psrc, &pdst, NULL);
 	__sync_add_and_fetch(&hits, 1);
 
 	return 0;
@@ -84,7 +83,7 @@ int crypto_encrypt(struct __sk_buff *skb)
 SEC("tc")
 int crypto_decrypt(struct __sk_buff *skb)
 {
-	struct bpf_dynptr psrc, pdst, iv;
+	struct bpf_dynptr psrc, pdst;
 	struct __crypto_ctx_value *v;
 	struct bpf_crypto_ctx *ctx;
 
@@ -98,9 +97,8 @@ int crypto_decrypt(struct __sk_buff *skb)
 
 	bpf_dynptr_from_skb(skb, 0, &psrc);
 	bpf_dynptr_from_mem(dst, len, 0, &pdst);
-	bpf_dynptr_from_mem(dst, 0, 0, &iv);
 
-	status = bpf_crypto_decrypt(ctx, &psrc, &pdst, &iv);
+	status = bpf_crypto_decrypt(ctx, &psrc, &pdst, NULL);
 	__sync_add_and_fetch(&hits, 1);
 
 	return 0;
-- 
cgit v1.2.3-58-ga151


From 2d45ab1eda469c802728d0a74e1601de5e71c098 Mon Sep 17 00:00:00 2001
From: Vadim Fedorenko <vadfed@meta.com>
Date: Thu, 13 Jun 2024 14:18:17 -0700
Subject: selftests: bpf: add testmod kfunc for nullable params

Add special test to be sure that only __nullable BTF params can be
replaced by NULL. This patch adds fake kfuncs in bpf_testmod to
properly test different params.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Vadim Fedorenko <vadfed@meta.com>
Link: https://lore.kernel.org/r/20240613211817.1551967-6-vadfed@meta.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../selftests/bpf/bpf_testmod/bpf_testmod.c        |  6 +++
 .../selftests/bpf/bpf_testmod/bpf_testmod_kfunc.h  |  1 +
 .../bpf/prog_tests/kfunc_param_nullable.c          | 11 ++++++
 .../bpf/progs/test_kfunc_param_nullable.c          | 43 ++++++++++++++++++++++
 4 files changed, 61 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/kfunc_param_nullable.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_kfunc_param_nullable.c

diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index 0a09732cde4b..49f9a311e49b 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -154,6 +154,11 @@ __bpf_kfunc void bpf_kfunc_common_test(void)
 {
 }
 
+__bpf_kfunc void bpf_kfunc_dynptr_test(struct bpf_dynptr *ptr,
+				       struct bpf_dynptr *ptr__nullable)
+{
+}
+
 struct bpf_testmod_btf_type_tag_1 {
 	int a;
 };
@@ -363,6 +368,7 @@ BTF_ID_FLAGS(func, bpf_iter_testmod_seq_new, KF_ITER_NEW)
 BTF_ID_FLAGS(func, bpf_iter_testmod_seq_next, KF_ITER_NEXT | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_iter_testmod_seq_destroy, KF_ITER_DESTROY)
 BTF_ID_FLAGS(func, bpf_kfunc_common_test)
+BTF_ID_FLAGS(func, bpf_kfunc_dynptr_test)
 BTF_KFUNCS_END(bpf_testmod_common_kfunc_ids)
 
 static const struct btf_kfunc_id_set bpf_testmod_common_kfunc_set = {
diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod_kfunc.h b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod_kfunc.h
index b0d586a6751f..f9809517e7fa 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod_kfunc.h
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod_kfunc.h
@@ -134,4 +134,5 @@ int bpf_kfunc_call_sock_sendmsg(struct sendmsg_args *args) __ksym;
 int bpf_kfunc_call_kernel_getsockname(struct addr_args *args) __ksym;
 int bpf_kfunc_call_kernel_getpeername(struct addr_args *args) __ksym;
 
+void bpf_kfunc_dynptr_test(struct bpf_dynptr *ptr, struct bpf_dynptr *ptr__nullable) __ksym;
 #endif /* _BPF_TESTMOD_KFUNC_H */
diff --git a/tools/testing/selftests/bpf/prog_tests/kfunc_param_nullable.c b/tools/testing/selftests/bpf/prog_tests/kfunc_param_nullable.c
new file mode 100644
index 000000000000..c8f4dcaac7c7
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/kfunc_param_nullable.c
@@ -0,0 +1,11 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/* Copyright (c) 2024 Meta Platforms, Inc */
+
+#include <test_progs.h>
+#include "test_kfunc_param_nullable.skel.h"
+
+void test_kfunc_param_nullable(void)
+{
+	RUN_TESTS(test_kfunc_param_nullable);
+}
diff --git a/tools/testing/selftests/bpf/progs/test_kfunc_param_nullable.c b/tools/testing/selftests/bpf/progs/test_kfunc_param_nullable.c
new file mode 100644
index 000000000000..7c75e9b8f455
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_kfunc_param_nullable.c
@@ -0,0 +1,43 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2024 Meta Platforms, Inc */
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+#include "bpf_misc.h"
+#include "bpf_kfuncs.h"
+#include "../bpf_testmod/bpf_testmod_kfunc.h"
+
+SEC("tc")
+int kfunc_dynptr_nullable_test1(struct __sk_buff *skb)
+{
+	struct bpf_dynptr data;
+
+	bpf_dynptr_from_skb(skb, 0, &data);
+	bpf_kfunc_dynptr_test(&data, NULL);
+
+	return 0;
+}
+
+SEC("tc")
+int kfunc_dynptr_nullable_test2(struct __sk_buff *skb)
+{
+	struct bpf_dynptr data;
+
+	bpf_dynptr_from_skb(skb, 0, &data);
+	bpf_kfunc_dynptr_test(&data, &data);
+
+	return 0;
+}
+
+SEC("tc")
+__failure __msg("expected pointer to stack or dynptr_ptr")
+int kfunc_dynptr_nullable_test3(struct __sk_buff *skb)
+{
+	struct bpf_dynptr data;
+
+	bpf_dynptr_from_skb(skb, 0, &data);
+	bpf_kfunc_dynptr_test(NULL, &data);
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-58-ga151


From c7f75954212b5e64f6b1f2375215b02fd79758ce Mon Sep 17 00:00:00 2001
From: Martin Schiller <ms@dev.tdt.de>
Date: Tue, 11 Jun 2024 15:54:23 +0200
Subject: dt-bindings: net: dsa: lantiq,gswip: convert to YAML schema

Convert the lantiq,gswip bindings to YAML format.

Also add this new file to the MAINTAINERS file.

Furthermore, the CPU port has to specify a phy-mode and either a phy or
a fixed-link. Since GSWIP is connected using a SoC internal protocol
there's no PHY involved. Add phy-mode = "internal" and a fixed-link to
the example code to describe the communication between the PMAC
(Ethernet controller) and GSWIP switch.

Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://lore.kernel.org/r/20240611135434.3180973-2-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../devicetree/bindings/net/dsa/lantiq,gswip.yaml  | 202 +++++++++++++++++++++
 .../devicetree/bindings/net/dsa/lantiq-gswip.txt   | 146 ---------------
 MAINTAINERS                                        |   1 +
 3 files changed, 203 insertions(+), 146 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml
 delete mode 100644 Documentation/devicetree/bindings/net/dsa/lantiq-gswip.txt

diff --git a/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml b/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml
new file mode 100644
index 000000000000..f3154b19af78
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml
@@ -0,0 +1,202 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/dsa/lantiq,gswip.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Lantiq GSWIP Ethernet switches
+
+allOf:
+  - $ref: dsa.yaml#/$defs/ethernet-ports
+
+maintainers:
+  - Hauke Mehrtens <hauke@hauke-m.de>
+
+properties:
+  compatible:
+    enum:
+      - lantiq,xrx200-gswip
+      - lantiq,xrx300-gswip
+      - lantiq,xrx330-gswip
+
+  reg:
+    minItems: 3
+    maxItems: 3
+
+  reg-names:
+    items:
+      - const: switch
+      - const: mdio
+      - const: mii
+
+  mdio:
+    $ref: /schemas/net/mdio.yaml#
+    unevaluatedProperties: false
+
+    properties:
+      compatible:
+        const: lantiq,xrx200-mdio
+
+    required:
+      - compatible
+
+  gphy-fw:
+    type: object
+    properties:
+      '#address-cells':
+        const: 1
+
+      '#size-cells':
+        const: 0
+
+      compatible:
+        items:
+          - enum:
+              - lantiq,xrx200-gphy-fw
+              - lantiq,xrx300-gphy-fw
+              - lantiq,xrx330-gphy-fw
+          - const: lantiq,gphy-fw
+
+      lantiq,rcu:
+        $ref: /schemas/types.yaml#/definitions/phandle
+        description: phandle to the RCU syscon
+
+    patternProperties:
+      "^gphy@[0-9a-f]{1,2}$":
+        type: object
+
+        additionalProperties: false
+
+        properties:
+          reg:
+            minimum: 0
+            maximum: 255
+            description:
+              Offset of the GPHY firmware register in the RCU register range
+
+          resets:
+            items:
+              - description: GPHY reset line
+
+          reset-names:
+            items:
+              - const: gphy
+
+        required:
+          - reg
+
+    required:
+      - compatible
+      - lantiq,rcu
+
+    additionalProperties: false
+
+required:
+  - compatible
+  - reg
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    switch@e108000 {
+            compatible = "lantiq,xrx200-gswip";
+            reg = <0xe108000 0x3100>,  /* switch */
+                  <0xe10b100 0xd8>,    /* mdio */
+                  <0xe10b1d8 0x130>;   /* mii */
+            dsa,member = <0 0>;
+
+            ports {
+                    #address-cells = <1>;
+                    #size-cells = <0>;
+
+                    port@0 {
+                            reg = <0>;
+                            label = "lan3";
+                            phy-mode = "rgmii";
+                            phy-handle = <&phy0>;
+                    };
+
+                    port@1 {
+                            reg = <1>;
+                            label = "lan4";
+                            phy-mode = "rgmii";
+                            phy-handle = <&phy1>;
+                    };
+
+                    port@2 {
+                            reg = <2>;
+                            label = "lan2";
+                            phy-mode = "internal";
+                            phy-handle = <&phy11>;
+                    };
+
+                    port@4 {
+                            reg = <4>;
+                            label = "lan1";
+                            phy-mode = "internal";
+                            phy-handle = <&phy13>;
+                    };
+
+                    port@5 {
+                            reg = <5>;
+                            label = "wan";
+                            phy-mode = "rgmii";
+                            phy-handle = <&phy5>;
+                    };
+
+                    port@6 {
+                            reg = <0x6>;
+                            phy-mode = "internal";
+                            ethernet = <&eth0>;
+
+                            fixed-link {
+                                    speed = <1000>;
+                                    full-duplex;
+                            };
+                    };
+            };
+
+            mdio {
+                    #address-cells = <1>;
+                    #size-cells = <0>;
+                    compatible = "lantiq,xrx200-mdio";
+
+                    phy0: ethernet-phy@0 {
+                            reg = <0x0>;
+                    };
+                    phy1: ethernet-phy@1 {
+                            reg = <0x1>;
+                    };
+                    phy5: ethernet-phy@5 {
+                            reg = <0x5>;
+                    };
+                    phy11: ethernet-phy@11 {
+                            reg = <0x11>;
+                    };
+                    phy13: ethernet-phy@13 {
+                            reg = <0x13>;
+                    };
+            };
+
+            gphy-fw {
+                    #address-cells = <1>;
+                    #size-cells = <0>;
+                    compatible = "lantiq,xrx200-gphy-fw", "lantiq,gphy-fw";
+                    lantiq,rcu = <&rcu0>;
+
+                    gphy@20 {
+                            reg = <0x20>;
+
+                            resets = <&reset0 31 30>;
+                            reset-names = "gphy";
+                    };
+
+                    gphy@68 {
+                            reg = <0x68>;
+
+                            resets = <&reset0 29 28>;
+                            reset-names = "gphy";
+                    };
+            };
+    };
diff --git a/Documentation/devicetree/bindings/net/dsa/lantiq-gswip.txt b/Documentation/devicetree/bindings/net/dsa/lantiq-gswip.txt
deleted file mode 100644
index 8bb1eff21cb1..000000000000
--- a/Documentation/devicetree/bindings/net/dsa/lantiq-gswip.txt
+++ /dev/null
@@ -1,146 +0,0 @@
-Lantiq GSWIP Ethernet switches
-==================================
-
-Required properties for GSWIP core:
-
-- compatible	: "lantiq,xrx200-gswip" for the embedded GSWIP in the
-		  xRX200 SoC
-		  "lantiq,xrx300-gswip" for the embedded GSWIP in the
-		  xRX300 SoC
-		  "lantiq,xrx330-gswip" for the embedded GSWIP in the
-		  xRX330 SoC
-- reg		: memory range of the GSWIP core registers
-		: memory range of the GSWIP MDIO registers
-		: memory range of the GSWIP MII registers
-
-See Documentation/devicetree/bindings/net/dsa/dsa.txt for a list of
-additional required and optional properties.
-
-
-Required properties for MDIO bus:
-- compatible	: "lantiq,xrx200-mdio" for the MDIO bus inside the GSWIP
-		  core of the xRX200 SoC and the PHYs connected to it.
-
-See Documentation/devicetree/bindings/net/mdio.txt for a list of additional
-required and optional properties.
-
-
-Required properties for GPHY firmware loading:
-- compatible	: "lantiq,xrx200-gphy-fw", "lantiq,gphy-fw"
-		  "lantiq,xrx300-gphy-fw", "lantiq,gphy-fw"
-		  "lantiq,xrx330-gphy-fw", "lantiq,gphy-fw"
-		  for the loading of the firmware into the embedded
-		  GPHY core of the SoC.
-- lantiq,rcu	: reference to the rcu syscon
-
-The GPHY firmware loader has a list of GPHY entries, one for each
-embedded GPHY
-
-- reg		: Offset of the GPHY firmware register in the RCU
-		  register range
-- resets	: list of resets of the embedded GPHY
-- reset-names	: list of names of the resets
-
-Example:
-
-Ethernet switch on the VRX200 SoC:
-
-switch@e108000 {
-	#address-cells = <1>;
-	#size-cells = <0>;
-	compatible = "lantiq,xrx200-gswip";
-	reg = <	0xe108000 0x3100	/* switch */
-		0xe10b100 0xd8		/* mdio */
-		0xe10b1d8 0x130		/* mii */
-		>;
-	dsa,member = <0 0>;
-
-	ports {
-		#address-cells = <1>;
-		#size-cells = <0>;
-
-		port@0 {
-			reg = <0>;
-			label = "lan3";
-			phy-mode = "rgmii";
-			phy-handle = <&phy0>;
-		};
-
-		port@1 {
-			reg = <1>;
-			label = "lan4";
-			phy-mode = "rgmii";
-			phy-handle = <&phy1>;
-		};
-
-		port@2 {
-			reg = <2>;
-			label = "lan2";
-			phy-mode = "internal";
-			phy-handle = <&phy11>;
-		};
-
-		port@4 {
-			reg = <4>;
-			label = "lan1";
-			phy-mode = "internal";
-			phy-handle = <&phy13>;
-		};
-
-		port@5 {
-			reg = <5>;
-			label = "wan";
-			phy-mode = "rgmii";
-			phy-handle = <&phy5>;
-		};
-
-		port@6 {
-			reg = <0x6>;
-			ethernet = <&eth0>;
-		};
-	};
-
-	mdio {
-		#address-cells = <1>;
-		#size-cells = <0>;
-		compatible = "lantiq,xrx200-mdio";
-		reg = <0>;
-
-		phy0: ethernet-phy@0 {
-			reg = <0x0>;
-		};
-		phy1: ethernet-phy@1 {
-			reg = <0x1>;
-		};
-		phy5: ethernet-phy@5 {
-			reg = <0x5>;
-		};
-		phy11: ethernet-phy@11 {
-			reg = <0x11>;
-		};
-		phy13: ethernet-phy@13 {
-			reg = <0x13>;
-		};
-	};
-
-	gphy-fw {
-		compatible = "lantiq,xrx200-gphy-fw", "lantiq,gphy-fw";
-		lantiq,rcu = <&rcu0>;
-		#address-cells = <1>;
-		#size-cells = <0>;
-
-		gphy@20 {
-			reg = <0x20>;
-
-			resets = <&reset0 31 30>;
-			reset-names = "gphy";
-		};
-
-		gphy@68 {
-			reg = <0x68>;
-
-			resets = <&reset0 29 28>;
-			reset-names = "gphy";
-		};
-	};
-};
diff --git a/MAINTAINERS b/MAINTAINERS
index 5de76400d7fd..e66b7d4324ae 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -12449,6 +12449,7 @@ LANTIQ / INTEL Ethernet drivers
 M:	Hauke Mehrtens <hauke@hauke-m.de>
 L:	netdev@vger.kernel.org
 S:	Maintained
+F:	Documentation/devicetree/bindings/net/dsa/lantiq,gswip.yaml
 F:	drivers/net/dsa/lantiq_gswip.c
 F:	drivers/net/dsa/lantiq_pce.h
 F:	drivers/net/ethernet/lantiq_xrx200.c
-- 
cgit v1.2.3-58-ga151


From b98f122ebdac28b0c932f3f4474eb0927c39297b Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Tue, 11 Jun 2024 15:54:24 +0200
Subject: net: dsa: lantiq_gswip: Only allow phy-mode = "internal" on the CPU
 port

Add the CPU port to gswip_xrx200_phylink_get_caps() and
gswip_xrx300_phylink_get_caps(). It connects through a SoC-internal bus,
so the only allowed phy-mode is PHY_INTERFACE_MODE_INTERNAL.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Link: https://lore.kernel.org/r/20240611135434.3180973-3-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lantiq_gswip.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index a557049e34f5..b9c7076ce32f 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -1516,6 +1516,7 @@ static void gswip_xrx200_phylink_get_caps(struct dsa_switch *ds, int port,
 	case 2:
 	case 3:
 	case 4:
+	case 6:
 		__set_bit(PHY_INTERFACE_MODE_INTERNAL,
 			  config->supported_interfaces);
 		break;
@@ -1547,6 +1548,7 @@ static void gswip_xrx300_phylink_get_caps(struct dsa_switch *ds, int port,
 	case 2:
 	case 3:
 	case 4:
+	case 6:
 		__set_bit(PHY_INTERFACE_MODE_INTERNAL,
 			  config->supported_interfaces);
 		break;
-- 
cgit v1.2.3-58-ga151


From dd6d364e1895211df8a8fe02c2a5a0b2a7049957 Mon Sep 17 00:00:00 2001
From: Martin Schiller <ms@dev.tdt.de>
Date: Tue, 11 Jun 2024 15:54:25 +0200
Subject: net: dsa: lantiq_gswip: add terminating \n where missing

Some dev_err are missing the terminating \n. Let's add that.

Suggested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://lore.kernel.org/r/20240611135434.3180973-4-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lantiq_gswip.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index b9c7076ce32f..fe64baf0d7f1 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -836,7 +836,7 @@ static int gswip_setup(struct dsa_switch *ds)
 
 	err = gswip_pce_load_microcode(priv);
 	if (err) {
-		dev_err(priv->dev, "writing PCE microcode failed, %i", err);
+		dev_err(priv->dev, "writing PCE microcode failed, %i\n", err);
 		return err;
 	}
 
@@ -1792,7 +1792,7 @@ static u32 gswip_bcm_ram_entry_read(struct gswip_priv *priv, u32 table,
 	err = gswip_switch_r_timeout(priv, GSWIP_BM_RAM_CTRL,
 				     GSWIP_BM_RAM_CTRL_BAS);
 	if (err) {
-		dev_err(priv->dev, "timeout while reading table: %u, index: %u",
+		dev_err(priv->dev, "timeout while reading table: %u, index: %u\n",
 			table, index);
 		return 0;
 	}
@@ -2021,7 +2021,7 @@ static void gswip_gphy_fw_remove(struct gswip_priv *priv,
 
 	ret = regmap_write(priv->rcu_regmap, gphy_fw->fw_addr_offset, 0);
 	if (ret)
-		dev_err(priv->dev, "can not reset GPHY FW pointer");
+		dev_err(priv->dev, "can not reset GPHY FW pointer\n");
 
 	clk_disable_unprepare(gphy_fw->clk_gate);
 
@@ -2050,7 +2050,7 @@ static int gswip_gphy_fw_list(struct gswip_priv *priv,
 			priv->gphy_fw_name_cfg = &xrx200a2x_gphy_data;
 			break;
 		default:
-			dev_err(dev, "unknown GSWIP version: 0x%x", version);
+			dev_err(dev, "unknown GSWIP version: 0x%x\n", version);
 			return -ENOENT;
 		}
 	}
@@ -2060,7 +2060,7 @@ static int gswip_gphy_fw_list(struct gswip_priv *priv,
 		priv->gphy_fw_name_cfg = match->data;
 
 	if (!priv->gphy_fw_name_cfg) {
-		dev_err(dev, "GPHY compatible type not supported");
+		dev_err(dev, "GPHY compatible type not supported\n");
 		return -ENOENT;
 	}
 
@@ -2163,7 +2163,7 @@ static int gswip_probe(struct platform_device *pdev)
 			return -EINVAL;
 		break;
 	default:
-		dev_err(dev, "unknown GSWIP version: 0x%x", version);
+		dev_err(dev, "unknown GSWIP version: 0x%x\n", version);
 		return -ENOENT;
 	}
 
@@ -2191,7 +2191,7 @@ static int gswip_probe(struct platform_device *pdev)
 		goto gphy_fw_remove;
 	}
 	if (!dsa_is_cpu_port(priv->ds, priv->hw_info->cpu_port)) {
-		dev_err(dev, "wrong CPU port defined, HW only supports port: %i",
+		dev_err(dev, "wrong CPU port defined, HW only supports port: %i\n",
 			priv->hw_info->cpu_port);
 		err = -EINVAL;
 		goto disable_switch;
-- 
cgit v1.2.3-58-ga151


From 1763b155da022ac0f984463e68cb0cda8ffc1fe8 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Tue, 11 Jun 2024 15:54:26 +0200
Subject: net: dsa: lantiq_gswip: Use dev_err_probe where appropriate

dev_err_probe() can be used to simplify the existing code. Also it means
we get rid of the following warning which is seen whenever the PMAC
(Ethernet controller which connects to GSWIP's CPU port) has not been
probed yet:
  gswip 1e108000.switch: dsa switch register failed: -517

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Link: https://lore.kernel.org/r/20240611135434.3180973-5-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lantiq_gswip.c | 53 ++++++++++++++++++++----------------------
 1 file changed, 25 insertions(+), 28 deletions(-)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index fe64baf0d7f1..37cc0247dc78 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -1931,11 +1931,9 @@ static int gswip_gphy_fw_load(struct gswip_priv *priv, struct gswip_gphy_fw *gph
 	msleep(200);
 
 	ret = request_firmware(&fw, gphy_fw->fw_name, dev);
-	if (ret) {
-		dev_err(dev, "failed to load firmware: %s, error: %i\n",
-			gphy_fw->fw_name, ret);
-		return ret;
-	}
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to load firmware: %s\n",
+				     gphy_fw->fw_name);
 
 	/* GPHY cores need the firmware code in a persistent and contiguous
 	 * memory area with a 16 kB boundary aligned start address.
@@ -1948,9 +1946,9 @@ static int gswip_gphy_fw_load(struct gswip_priv *priv, struct gswip_gphy_fw *gph
 		dev_addr = ALIGN(dma_addr, XRX200_GPHY_FW_ALIGN);
 		memcpy(fw_addr, fw->data, fw->size);
 	} else {
-		dev_err(dev, "failed to alloc firmware memory\n");
 		release_firmware(fw);
-		return -ENOMEM;
+		return dev_err_probe(dev, -ENOMEM,
+				     "failed to alloc firmware memory\n");
 	}
 
 	release_firmware(fw);
@@ -1977,8 +1975,8 @@ static int gswip_gphy_fw_probe(struct gswip_priv *priv,
 
 	gphy_fw->clk_gate = devm_clk_get(dev, gphyname);
 	if (IS_ERR(gphy_fw->clk_gate)) {
-		dev_err(dev, "Failed to lookup gate clock\n");
-		return PTR_ERR(gphy_fw->clk_gate);
+		return dev_err_probe(dev, PTR_ERR(gphy_fw->clk_gate),
+				     "Failed to lookup gate clock\n");
 	}
 
 	ret = of_property_read_u32(gphy_fw_np, "reg", &gphy_fw->fw_addr_offset);
@@ -1998,8 +1996,8 @@ static int gswip_gphy_fw_probe(struct gswip_priv *priv,
 		gphy_fw->fw_name = priv->gphy_fw_name_cfg->ge_firmware_name;
 		break;
 	default:
-		dev_err(dev, "Unknown GPHY mode %d\n", gphy_mode);
-		return -EINVAL;
+		return dev_err_probe(dev, -EINVAL, "Unknown GPHY mode %d\n",
+				     gphy_mode);
 	}
 
 	gphy_fw->reset = of_reset_control_array_get_exclusive(gphy_fw_np);
@@ -2050,8 +2048,9 @@ static int gswip_gphy_fw_list(struct gswip_priv *priv,
 			priv->gphy_fw_name_cfg = &xrx200a2x_gphy_data;
 			break;
 		default:
-			dev_err(dev, "unknown GSWIP version: 0x%x\n", version);
-			return -ENOENT;
+			return dev_err_probe(dev, -ENOENT,
+					     "unknown GSWIP version: 0x%x\n",
+					     version);
 		}
 	}
 
@@ -2059,10 +2058,9 @@ static int gswip_gphy_fw_list(struct gswip_priv *priv,
 	if (match && match->data)
 		priv->gphy_fw_name_cfg = match->data;
 
-	if (!priv->gphy_fw_name_cfg) {
-		dev_err(dev, "GPHY compatible type not supported\n");
-		return -ENOENT;
-	}
+	if (!priv->gphy_fw_name_cfg)
+		return dev_err_probe(dev, -ENOENT,
+				     "GPHY compatible type not supported\n");
 
 	priv->num_gphy_fw = of_get_available_child_count(gphy_fw_list_np);
 	if (!priv->num_gphy_fw)
@@ -2163,8 +2161,8 @@ static int gswip_probe(struct platform_device *pdev)
 			return -EINVAL;
 		break;
 	default:
-		dev_err(dev, "unknown GSWIP version: 0x%x\n", version);
-		return -ENOENT;
+		return dev_err_probe(dev, -ENOENT,
+				     "unknown GSWIP version: 0x%x\n", version);
 	}
 
 	/* bring up the mdio bus */
@@ -2172,28 +2170,27 @@ static int gswip_probe(struct platform_device *pdev)
 	if (gphy_fw_np) {
 		err = gswip_gphy_fw_list(priv, gphy_fw_np, version);
 		of_node_put(gphy_fw_np);
-		if (err) {
-			dev_err(dev, "gphy fw probe failed\n");
-			return err;
-		}
+		if (err)
+			return dev_err_probe(dev, err,
+					     "gphy fw probe failed\n");
 	}
 
 	/* bring up the mdio bus */
 	err = gswip_mdio(priv);
 	if (err) {
-		dev_err(dev, "mdio probe failed\n");
+		dev_err_probe(dev, err, "mdio probe failed\n");
 		goto gphy_fw_remove;
 	}
 
 	err = dsa_register_switch(priv->ds);
 	if (err) {
-		dev_err(dev, "dsa switch register failed: %i\n", err);
+		dev_err_probe(dev, err, "dsa switch registration failed\n");
 		goto gphy_fw_remove;
 	}
 	if (!dsa_is_cpu_port(priv->ds, priv->hw_info->cpu_port)) {
-		dev_err(dev, "wrong CPU port defined, HW only supports port: %i\n",
-			priv->hw_info->cpu_port);
-		err = -EINVAL;
+		err = dev_err_probe(dev, -EINVAL,
+				    "wrong CPU port defined, HW only supports port: %i\n",
+				    priv->hw_info->cpu_port);
 		goto disable_switch;
 	}
 
-- 
cgit v1.2.3-58-ga151


From f5ebf9ab60940b00c36dfe64add41c80f3daff6a Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Tue, 11 Jun 2024 15:54:27 +0200
Subject: net: dsa: lantiq_gswip: Don't manually call gswip_port_enable()

We don't need to manually call gswip_port_enable() from within
gswip_setup() for the CPU port. DSA does this automatically for us.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Link: https://lore.kernel.org/r/20240611135434.3180973-6-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lantiq_gswip.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index 37cc0247dc78..c1f9419af35f 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -898,8 +898,6 @@ static int gswip_setup(struct dsa_switch *ds)
 
 	ds->mtu_enforcement_ingress = true;
 
-	gswip_port_enable(ds, cpu_port, NULL);
-
 	ds->configure_vlan_while_not_filtering = false;
 
 	return 0;
-- 
cgit v1.2.3-58-ga151


From 86b9ea6412af41914ef6549f85a849c3b987f4f3 Mon Sep 17 00:00:00 2001
From: Martin Schiller <ms@dev.tdt.de>
Date: Tue, 11 Jun 2024 15:54:28 +0200
Subject: net: dsa: lantiq_gswip: do also enable or disable cpu port

Before commit 74be4babe72f ("net: dsa: do not enable or disable non user
ports"), gswip_port_enable/disable() were also executed for the cpu port
in gswip_setup() which disabled the cpu port during initialization.

Let's restore this by removing the dsa_is_user_port checks. Also, let's
clean up the gswip_port_enable() function so that we only have to check
for the cpu port once. The operation reordering done here is safe.

Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://lore.kernel.org/r/20240611135434.3180973-7-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lantiq_gswip.c | 24 ++++++++----------------
 1 file changed, 8 insertions(+), 16 deletions(-)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index c1f9419af35f..8ec329d0c136 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -695,13 +695,18 @@ static int gswip_port_enable(struct dsa_switch *ds, int port,
 	struct gswip_priv *priv = ds->priv;
 	int err;
 
-	if (!dsa_is_user_port(ds, port))
-		return 0;
-
 	if (!dsa_is_cpu_port(ds, port)) {
+		u32 mdio_phy = 0;
+
 		err = gswip_add_single_port_br(priv, port, true);
 		if (err)
 			return err;
+
+		if (phydev)
+			mdio_phy = phydev->mdio.addr & GSWIP_MDIO_PHY_ADDR_MASK;
+
+		gswip_mdio_mask(priv, GSWIP_MDIO_PHY_ADDR_MASK, mdio_phy,
+				GSWIP_MDIO_PHYp(port));
 	}
 
 	/* RMON Counter Enable for port */
@@ -714,16 +719,6 @@ static int gswip_port_enable(struct dsa_switch *ds, int port,
 	gswip_switch_mask(priv, 0, GSWIP_SDMA_PCTRL_EN,
 			  GSWIP_SDMA_PCTRLp(port));
 
-	if (!dsa_is_cpu_port(ds, port)) {
-		u32 mdio_phy = 0;
-
-		if (phydev)
-			mdio_phy = phydev->mdio.addr & GSWIP_MDIO_PHY_ADDR_MASK;
-
-		gswip_mdio_mask(priv, GSWIP_MDIO_PHY_ADDR_MASK, mdio_phy,
-				GSWIP_MDIO_PHYp(port));
-	}
-
 	return 0;
 }
 
@@ -731,9 +726,6 @@ static void gswip_port_disable(struct dsa_switch *ds, int port)
 {
 	struct gswip_priv *priv = ds->priv;
 
-	if (!dsa_is_user_port(ds, port))
-		return;
-
 	gswip_switch_mask(priv, GSWIP_FDMA_PCTRL_EN, 0,
 			  GSWIP_FDMA_PCTRLp(port));
 	gswip_switch_mask(priv, GSWIP_SDMA_PCTRL_EN, 0,
-- 
cgit v1.2.3-58-ga151


From 7168ec1b06691295db6b335e5f5f6c86c7061213 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Tue, 11 Jun 2024 15:54:29 +0200
Subject: net: dsa: lantiq_gswip: Use dsa_is_cpu_port() in
 gswip_port_change_mtu()

Make the check for the CPU port in gswip_port_change_mtu() consistent
with other areas of the driver by using dsa_is_cpu_port().

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Link: https://lore.kernel.org/r/20240611135434.3180973-8-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lantiq_gswip.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index 8ec329d0c136..58c069f964dd 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -1464,12 +1464,11 @@ static int gswip_port_max_mtu(struct dsa_switch *ds, int port)
 static int gswip_port_change_mtu(struct dsa_switch *ds, int port, int new_mtu)
 {
 	struct gswip_priv *priv = ds->priv;
-	int cpu_port = priv->hw_info->cpu_port;
 
 	/* CPU port always has maximum mtu of user ports, so use it to set
 	 * switch frame size, including 8 byte special header.
 	 */
-	if (port == cpu_port) {
+	if (dsa_is_cpu_port(ds, port)) {
 		new_mtu += 8;
 		gswip_switch_w(priv, VLAN_ETH_HLEN + new_mtu + ETH_FCS_LEN,
 			       GSWIP_MAC_FLEN);
-- 
cgit v1.2.3-58-ga151


From c927b6e47b5cc7324217bf5fe7e6ccd0633971a0 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Tue, 11 Jun 2024 15:54:30 +0200
Subject: net: dsa: lantiq_gswip: Change literal 6 to ETH_ALEN

The addr variable in gswip_port_fdb_dump() stores a mac address. Use
ETH_ALEN to make this consistent across other drivers.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Link: https://lore.kernel.org/r/20240611135434.3180973-9-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lantiq_gswip.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index 58c069f964dd..525a62a21601 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -1413,7 +1413,7 @@ static int gswip_port_fdb_dump(struct dsa_switch *ds, int port,
 {
 	struct gswip_priv *priv = ds->priv;
 	struct gswip_pce_table_entry mac_bridge = {0,};
-	unsigned char addr[6];
+	unsigned char addr[ETH_ALEN];
 	int i;
 	int err;
 
-- 
cgit v1.2.3-58-ga151


From e6c34597f89ac98c06176eed57f125252015a330 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Tue, 11 Jun 2024 15:54:31 +0200
Subject: net: dsa: lantiq_gswip: Consistently use macros for the mac bridge
 table

Only bits [5:0] in mac_bridge.key[3] are reserved for the FID.
Also, for dynamic (learned) entries, bits [7:4] in mac_bridge.val[0]
represents the port.

Introduce new macros GSWIP_TABLE_MAC_BRIDGE_KEY3_FID and
GSWIP_TABLE_MAC_BRIDGE_VAL0_PORT macro and use it throughout the driver.
Also rename and update GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC to use the
BIT() macro. This makes the driver code easier to understand.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://lore.kernel.org/r/20240611135434.3180973-10-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lantiq_gswip.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index 525a62a21601..cd88b00cfdc1 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -236,7 +236,9 @@
 #define GSWIP_TABLE_ACTIVE_VLAN		0x01
 #define GSWIP_TABLE_VLAN_MAPPING	0x02
 #define GSWIP_TABLE_MAC_BRIDGE		0x0b
-#define  GSWIP_TABLE_MAC_BRIDGE_STATIC	0x01	/* Static not, aging entry */
+#define  GSWIP_TABLE_MAC_BRIDGE_KEY3_FID	GENMASK(5, 0)	/* Filtering identifier */
+#define  GSWIP_TABLE_MAC_BRIDGE_VAL0_PORT	GENMASK(7, 4)	/* Port on learned entries */
+#define  GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC	BIT(0)		/* Static, non-aging entry */
 
 #define XRX200_GPHY_FW_ALIGN	(16 * 1024)
 
@@ -1304,10 +1306,11 @@ static void gswip_port_fast_age(struct dsa_switch *ds, int port)
 		if (!mac_bridge.valid)
 			continue;
 
-		if (mac_bridge.val[1] & GSWIP_TABLE_MAC_BRIDGE_STATIC)
+		if (mac_bridge.val[1] & GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC)
 			continue;
 
-		if (((mac_bridge.val[0] & GENMASK(7, 4)) >> 4) != port)
+		if (port != FIELD_GET(GSWIP_TABLE_MAC_BRIDGE_VAL0_PORT,
+				      mac_bridge.val[0]))
 			continue;
 
 		mac_bridge.valid = false;
@@ -1382,9 +1385,9 @@ static int gswip_port_fdb(struct dsa_switch *ds, int port,
 	mac_bridge.key[0] = addr[5] | (addr[4] << 8);
 	mac_bridge.key[1] = addr[3] | (addr[2] << 8);
 	mac_bridge.key[2] = addr[1] | (addr[0] << 8);
-	mac_bridge.key[3] = fid;
+	mac_bridge.key[3] = FIELD_PREP(GSWIP_TABLE_MAC_BRIDGE_KEY3_FID, fid);
 	mac_bridge.val[0] = add ? BIT(port) : 0; /* port map */
-	mac_bridge.val[1] = GSWIP_TABLE_MAC_BRIDGE_STATIC;
+	mac_bridge.val[1] = GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC;
 	mac_bridge.valid = add;
 
 	err = gswip_pce_table_entry_write(priv, &mac_bridge);
@@ -1438,14 +1441,15 @@ static int gswip_port_fdb_dump(struct dsa_switch *ds, int port,
 		addr[2] = (mac_bridge.key[1] >> 8) & 0xff;
 		addr[1] = mac_bridge.key[2] & 0xff;
 		addr[0] = (mac_bridge.key[2] >> 8) & 0xff;
-		if (mac_bridge.val[1] & GSWIP_TABLE_MAC_BRIDGE_STATIC) {
+		if (mac_bridge.val[1] & GSWIP_TABLE_MAC_BRIDGE_VAL1_STATIC) {
 			if (mac_bridge.val[0] & BIT(port)) {
 				err = cb(addr, 0, true, data);
 				if (err)
 					return err;
 			}
 		} else {
-			if (((mac_bridge.val[0] & GENMASK(7, 4)) >> 4) == port) {
+			if (port == FIELD_GET(GSWIP_TABLE_MAC_BRIDGE_VAL0_PORT,
+					      mac_bridge.val[0])) {
 				err = cb(addr, 0, false, data);
 				if (err)
 					return err;
-- 
cgit v1.2.3-58-ga151


From b068706b7831ccf7c7f1a56a65862fbcc28d061f Mon Sep 17 00:00:00 2001
From: Martin Schiller <ms@dev.tdt.de>
Date: Tue, 11 Jun 2024 15:54:32 +0200
Subject: net: dsa: lantiq_gswip: Remove dead code from
 gswip_add_single_port_br()

The port validation in gswip_add_single_port_br() is superfluous and
can be omitted.

Suggested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://lore.kernel.org/r/20240611135434.3180973-11-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lantiq_gswip.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index cd88b00cfdc1..2bbc7dd45418 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -655,14 +655,8 @@ static int gswip_add_single_port_br(struct gswip_priv *priv, int port, bool add)
 	struct gswip_pce_table_entry vlan_active = {0,};
 	struct gswip_pce_table_entry vlan_mapping = {0,};
 	unsigned int cpu_port = priv->hw_info->cpu_port;
-	unsigned int max_ports = priv->hw_info->max_ports;
 	int err;
 
-	if (port >= max_ports) {
-		dev_err(priv->dev, "single port for %i supported\n", port);
-		return -EIO;
-	}
-
 	vlan_active.index = port + 1;
 	vlan_active.table = GSWIP_TABLE_ACTIVE_VLAN;
 	vlan_active.key[0] = 0; /* vid */
-- 
cgit v1.2.3-58-ga151


From e19fbe3996aae35a467ebad35ff2b8d84975a65c Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Tue, 11 Jun 2024 15:54:33 +0200
Subject: net: dsa: lantiq_gswip: Update comments in
 gswip_port_vlan_filtering()

Update the comments in gswip_port_vlan_filtering() so it's clear that
there are two separate cases, one for "tag based VLAN" and another one
for "port based VLAN".

Suggested-by: Martin Schiller <ms@dev.tdt.de>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Link: https://lore.kernel.org/r/20240611135434.3180973-12-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lantiq_gswip.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index 2bbc7dd45418..ec52c62eadce 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -780,7 +780,7 @@ static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port,
 	}
 
 	if (vlan_filtering) {
-		/* Use port based VLAN tag */
+		/* Use tag based VLAN */
 		gswip_switch_mask(priv,
 				  GSWIP_PCE_VCTRL_VSR,
 				  GSWIP_PCE_VCTRL_UVR | GSWIP_PCE_VCTRL_VIMR |
@@ -789,7 +789,7 @@ static int gswip_port_vlan_filtering(struct dsa_switch *ds, int port,
 		gswip_switch_mask(priv, GSWIP_PCE_PCTRL_0_TVM, 0,
 				  GSWIP_PCE_PCTRL_0p(port));
 	} else {
-		/* Use port based VLAN tag */
+		/* Use port based VLAN */
 		gswip_switch_mask(priv,
 				  GSWIP_PCE_VCTRL_UVR | GSWIP_PCE_VCTRL_VIMR |
 				  GSWIP_PCE_VCTRL_VEMR,
-- 
cgit v1.2.3-58-ga151


From 3b0a95ed7782dce88a5ef4860dcaab962cec9527 Mon Sep 17 00:00:00 2001
From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Tue, 11 Jun 2024 15:54:34 +0200
Subject: net: dsa: lantiq_gswip: Improve error message in gswip_port_fdb()

Print that no FID is found for bridge %s instead of the incorrect
message that the port is not part of a bridge.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Link: https://lore.kernel.org/r/20240611135434.3180973-13-ms@dev.tdt.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lantiq_gswip.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/dsa/lantiq_gswip.c b/drivers/net/dsa/lantiq_gswip.c
index ec52c62eadce..fcd4505f4925 100644
--- a/drivers/net/dsa/lantiq_gswip.c
+++ b/drivers/net/dsa/lantiq_gswip.c
@@ -1370,7 +1370,8 @@ static int gswip_port_fdb(struct dsa_switch *ds, int port,
 	}
 
 	if (fid == -1) {
-		dev_err(priv->dev, "Port not part of a bridge\n");
+		dev_err(priv->dev, "no FID found for bridge %s\n",
+			bridge->name);
 		return -EINVAL;
 	}
 
-- 
cgit v1.2.3-58-ga151


From 5f703ce5c981ee02c00e210d5b155bbbfbf11263 Mon Sep 17 00:00:00 2001
From: Lukasz Majewski <lukma@denx.de>
Date: Mon, 10 Jun 2024 15:39:14 +0200
Subject: net: hsr: Send supervisory frames to HSR network with ProxyNodeTable
 data

This patch provides support for sending supervision HSR frames with
MAC addresses stored in ProxyNodeTable when RedBox (i.e. HSR-SAN) is
enabled.

Supervision frames with RedBox MAC address (appended as second TLV)
are only send for ProxyNodeTable nodes.

This patch series shall be tested with hsr_redbox.sh script.

Signed-off-by: Lukasz Majewski <lukma@denx.de>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/hsr/hsr_device.c   | 63 +++++++++++++++++++++++++++++++++++++++++---------
 net/hsr/hsr_forward.c  | 37 +++++++++++++++++++++++++++--
 net/hsr/hsr_framereg.c | 12 ++++++++++
 net/hsr/hsr_framereg.h |  2 ++
 net/hsr/hsr_main.h     |  4 +++-
 net/hsr/hsr_netlink.c  |  1 +
 6 files changed, 105 insertions(+), 14 deletions(-)

diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c
index e6904288d40d..e4cc6b78dcfc 100644
--- a/net/hsr/hsr_device.c
+++ b/net/hsr/hsr_device.c
@@ -73,9 +73,15 @@ static void hsr_check_announce(struct net_device *hsr_dev)
 			mod_timer(&hsr->announce_timer, jiffies +
 				  msecs_to_jiffies(HSR_ANNOUNCE_INTERVAL));
 		}
+
+		if (hsr->redbox && !timer_pending(&hsr->announce_proxy_timer))
+			mod_timer(&hsr->announce_proxy_timer, jiffies +
+				  msecs_to_jiffies(HSR_ANNOUNCE_INTERVAL) / 2);
 	} else {
 		/* Deactivate the announce timer  */
 		timer_delete(&hsr->announce_timer);
+		if (hsr->redbox)
+			timer_delete(&hsr->announce_proxy_timer);
 	}
 }
 
@@ -279,10 +285,11 @@ out:
 	return NULL;
 }
 
-static void send_hsr_supervision_frame(struct hsr_port *master,
-				       unsigned long *interval)
+static void send_hsr_supervision_frame(struct hsr_port *port,
+				       unsigned long *interval,
+				       const unsigned char *addr)
 {
-	struct hsr_priv *hsr = master->hsr;
+	struct hsr_priv *hsr = port->hsr;
 	__u8 type = HSR_TLV_LIFE_CHECK;
 	struct hsr_sup_payload *hsr_sp;
 	struct hsr_sup_tlv *hsr_stlv;
@@ -296,9 +303,9 @@ static void send_hsr_supervision_frame(struct hsr_port *master,
 		hsr->announce_count++;
 	}
 
-	skb = hsr_init_skb(master);
+	skb = hsr_init_skb(port);
 	if (!skb) {
-		netdev_warn_once(master->dev, "HSR: Could not send supervision frame\n");
+		netdev_warn_once(port->dev, "HSR: Could not send supervision frame\n");
 		return;
 	}
 
@@ -321,11 +328,12 @@ static void send_hsr_supervision_frame(struct hsr_port *master,
 	hsr_stag->tlv.HSR_TLV_length = hsr->prot_version ?
 				sizeof(struct hsr_sup_payload) : 12;
 
-	/* Payload: MacAddressA */
+	/* Payload: MacAddressA / SAN MAC from ProxyNodeTable */
 	hsr_sp = skb_put(skb, sizeof(struct hsr_sup_payload));
-	ether_addr_copy(hsr_sp->macaddress_A, master->dev->dev_addr);
+	ether_addr_copy(hsr_sp->macaddress_A, addr);
 
-	if (hsr->redbox) {
+	if (hsr->redbox &&
+	    hsr_is_node_in_db(&hsr->proxy_node_db, addr)) {
 		hsr_stlv = skb_put(skb, sizeof(struct hsr_sup_tlv));
 		hsr_stlv->HSR_TLV_type = PRP_TLV_REDBOX_MAC;
 		hsr_stlv->HSR_TLV_length = sizeof(struct hsr_sup_payload);
@@ -340,13 +348,14 @@ static void send_hsr_supervision_frame(struct hsr_port *master,
 		return;
 	}
 
-	hsr_forward_skb(skb, master);
+	hsr_forward_skb(skb, port);
 	spin_unlock_bh(&hsr->seqnr_lock);
 	return;
 }
 
 static void send_prp_supervision_frame(struct hsr_port *master,
-				       unsigned long *interval)
+				       unsigned long *interval,
+				       const unsigned char *addr)
 {
 	struct hsr_priv *hsr = master->hsr;
 	struct hsr_sup_payload *hsr_sp;
@@ -396,7 +405,7 @@ static void hsr_announce(struct timer_list *t)
 
 	rcu_read_lock();
 	master = hsr_port_get_hsr(hsr, HSR_PT_MASTER);
-	hsr->proto_ops->send_sv_frame(master, &interval);
+	hsr->proto_ops->send_sv_frame(master, &interval, master->dev->dev_addr);
 
 	if (is_admin_up(master->dev))
 		mod_timer(&hsr->announce_timer, jiffies + interval);
@@ -404,6 +413,37 @@ static void hsr_announce(struct timer_list *t)
 	rcu_read_unlock();
 }
 
+/* Announce (supervision frame) timer function for RedBox
+ */
+static void hsr_proxy_announce(struct timer_list *t)
+{
+	struct hsr_priv *hsr = from_timer(hsr, t, announce_proxy_timer);
+	struct hsr_port *interlink;
+	unsigned long interval = 0;
+	struct hsr_node *node;
+
+	rcu_read_lock();
+	/* RedBOX sends supervisory frames to HSR network with MAC addresses
+	 * of SAN nodes stored in ProxyNodeTable.
+	 */
+	interlink = hsr_port_get_hsr(hsr, HSR_PT_INTERLINK);
+	list_for_each_entry_rcu(node, &hsr->proxy_node_db, mac_list) {
+		if (hsr_addr_is_redbox(hsr, node->macaddress_A))
+			continue;
+		hsr->proto_ops->send_sv_frame(interlink, &interval,
+					      node->macaddress_A);
+	}
+
+	if (is_admin_up(interlink->dev)) {
+		if (!interval)
+			interval = msecs_to_jiffies(HSR_ANNOUNCE_INTERVAL);
+
+		mod_timer(&hsr->announce_proxy_timer, jiffies + interval);
+	}
+
+	rcu_read_unlock();
+}
+
 void hsr_del_ports(struct hsr_priv *hsr)
 {
 	struct hsr_port *port;
@@ -590,6 +630,7 @@ int hsr_dev_finalize(struct net_device *hsr_dev, struct net_device *slave[2],
 	timer_setup(&hsr->announce_timer, hsr_announce, 0);
 	timer_setup(&hsr->prune_timer, hsr_prune_nodes, 0);
 	timer_setup(&hsr->prune_proxy_timer, hsr_prune_proxy_nodes, 0);
+	timer_setup(&hsr->announce_proxy_timer, hsr_proxy_announce, 0);
 
 	ether_addr_copy(hsr->sup_multicast_addr, def_multicast_addr);
 	hsr->sup_multicast_addr[ETH_ALEN - 1] = multicast_spec;
diff --git a/net/hsr/hsr_forward.c b/net/hsr/hsr_forward.c
index 05a61b8286ec..960ef386bc3a 100644
--- a/net/hsr/hsr_forward.c
+++ b/net/hsr/hsr_forward.c
@@ -117,6 +117,35 @@ static bool is_supervision_frame(struct hsr_priv *hsr, struct sk_buff *skb)
 	return true;
 }
 
+static bool is_proxy_supervision_frame(struct hsr_priv *hsr,
+				       struct sk_buff *skb)
+{
+	struct hsr_sup_payload *payload;
+	struct ethhdr *eth_hdr;
+	u16 total_length = 0;
+
+	eth_hdr = (struct ethhdr *)skb_mac_header(skb);
+
+	/* Get the HSR protocol revision. */
+	if (eth_hdr->h_proto == htons(ETH_P_HSR))
+		total_length = sizeof(struct hsrv1_ethhdr_sp);
+	else
+		total_length = sizeof(struct hsrv0_ethhdr_sp);
+
+	if (!pskb_may_pull(skb, total_length + sizeof(struct hsr_sup_payload)))
+		return false;
+
+	skb_pull(skb, total_length);
+	payload = (struct hsr_sup_payload *)skb->data;
+	skb_push(skb, total_length);
+
+	/* For RedBox (HSR-SAN) check if we have received the supervision
+	 * frame with MAC addresses from own ProxyNodeTable.
+	 */
+	return hsr_is_node_in_db(&hsr->proxy_node_db,
+				 payload->macaddress_A);
+}
+
 static struct sk_buff *create_stripped_skb_hsr(struct sk_buff *skb_in,
 					       struct hsr_frame_info *frame)
 {
@@ -499,7 +528,8 @@ static void hsr_forward_do(struct hsr_frame_info *frame)
 					   frame->sequence_nr))
 			continue;
 
-		if (frame->is_supervision && port->type == HSR_PT_MASTER) {
+		if (frame->is_supervision && port->type == HSR_PT_MASTER &&
+		    !frame->is_proxy_supervision) {
 			hsr_handle_sup_frame(frame);
 			continue;
 		}
@@ -637,6 +667,9 @@ static int fill_frame_info(struct hsr_frame_info *frame,
 
 	memset(frame, 0, sizeof(*frame));
 	frame->is_supervision = is_supervision_frame(port->hsr, skb);
+	if (frame->is_supervision && hsr->redbox)
+		frame->is_proxy_supervision =
+			is_proxy_supervision_frame(port->hsr, skb);
 
 	n_db = &hsr->node_db;
 	if (port->type == HSR_PT_INTERLINK)
@@ -688,7 +721,7 @@ void hsr_forward_skb(struct sk_buff *skb, struct hsr_port *port)
 	/* Gets called for ingress frames as well as egress from master port.
 	 * So check and increment stats for master port only here.
 	 */
-	if (port->type == HSR_PT_MASTER) {
+	if (port->type == HSR_PT_MASTER || port->type == HSR_PT_INTERLINK) {
 		port->dev->stats.tx_packets++;
 		port->dev->stats.tx_bytes += skb->len;
 	}
diff --git a/net/hsr/hsr_framereg.c b/net/hsr/hsr_framereg.c
index 614df9649794..73bc6f659812 100644
--- a/net/hsr/hsr_framereg.c
+++ b/net/hsr/hsr_framereg.c
@@ -36,6 +36,14 @@ static bool seq_nr_after(u16 a, u16 b)
 #define seq_nr_before(a, b)		seq_nr_after((b), (a))
 #define seq_nr_before_or_eq(a, b)	(!seq_nr_after((a), (b)))
 
+bool hsr_addr_is_redbox(struct hsr_priv *hsr, unsigned char *addr)
+{
+	if (!hsr->redbox || !is_valid_ether_addr(hsr->macaddress_redbox))
+		return false;
+
+	return ether_addr_equal(addr, hsr->macaddress_redbox);
+}
+
 bool hsr_addr_is_self(struct hsr_priv *hsr, unsigned char *addr)
 {
 	struct hsr_self_node *sn;
@@ -591,6 +599,10 @@ void hsr_prune_proxy_nodes(struct timer_list *t)
 
 	spin_lock_bh(&hsr->list_lock);
 	list_for_each_entry_safe(node, tmp, &hsr->proxy_node_db, mac_list) {
+		/* Don't prune RedBox node. */
+		if (hsr_addr_is_redbox(hsr, node->macaddress_A))
+			continue;
+
 		timestamp = node->time_in[HSR_PT_INTERLINK];
 
 		/* Prune old entries */
diff --git a/net/hsr/hsr_framereg.h b/net/hsr/hsr_framereg.h
index 7619e31c1d2d..993fa950d814 100644
--- a/net/hsr/hsr_framereg.h
+++ b/net/hsr/hsr_framereg.h
@@ -22,6 +22,7 @@ struct hsr_frame_info {
 	struct hsr_node *node_src;
 	u16 sequence_nr;
 	bool is_supervision;
+	bool is_proxy_supervision;
 	bool is_vlan;
 	bool is_local_dest;
 	bool is_local_exclusive;
@@ -35,6 +36,7 @@ struct hsr_node *hsr_get_node(struct hsr_port *port, struct list_head *node_db,
 			      enum hsr_port_type rx_port);
 void hsr_handle_sup_frame(struct hsr_frame_info *frame);
 bool hsr_addr_is_self(struct hsr_priv *hsr, unsigned char *addr);
+bool hsr_addr_is_redbox(struct hsr_priv *hsr, unsigned char *addr);
 
 void hsr_addr_subst_source(struct hsr_node *node, struct sk_buff *skb);
 void hsr_addr_subst_dest(struct hsr_node *node_src, struct sk_buff *skb,
diff --git a/net/hsr/hsr_main.h b/net/hsr/hsr_main.h
index 23850b16d1ea..ab1f8d35d9dc 100644
--- a/net/hsr/hsr_main.h
+++ b/net/hsr/hsr_main.h
@@ -170,7 +170,8 @@ struct hsr_node;
 
 struct hsr_proto_ops {
 	/* format and send supervision frame */
-	void (*send_sv_frame)(struct hsr_port *port, unsigned long *interval);
+	void (*send_sv_frame)(struct hsr_port *port, unsigned long *interval,
+			      const unsigned char addr[ETH_ALEN]);
 	void (*handle_san_frame)(bool san, enum hsr_port_type port,
 				 struct hsr_node *node);
 	bool (*drop_frame)(struct hsr_frame_info *frame, struct hsr_port *port);
@@ -197,6 +198,7 @@ struct hsr_priv {
 	struct list_head	proxy_node_db;	/* RedBox HSR proxy nodes */
 	struct hsr_self_node	__rcu *self_node;	/* MACs of slaves */
 	struct timer_list	announce_timer;	/* Supervision frame dispatch */
+	struct timer_list	announce_proxy_timer;
 	struct timer_list	prune_timer;
 	struct timer_list	prune_proxy_timer;
 	int announce_count;
diff --git a/net/hsr/hsr_netlink.c b/net/hsr/hsr_netlink.c
index 898f18c6da53..f6ff0b61e08a 100644
--- a/net/hsr/hsr_netlink.c
+++ b/net/hsr/hsr_netlink.c
@@ -131,6 +131,7 @@ static void hsr_dellink(struct net_device *dev, struct list_head *head)
 	del_timer_sync(&hsr->prune_timer);
 	del_timer_sync(&hsr->prune_proxy_timer);
 	del_timer_sync(&hsr->announce_timer);
+	timer_delete_sync(&hsr->announce_proxy_timer);
 
 	hsr_debugfs_term(hsr);
 	hsr_del_ports(hsr);
-- 
cgit v1.2.3-58-ga151


From 8a9044e5169bab7a8edadb4ceb748391657f0d7f Mon Sep 17 00:00:00 2001
From: Christophe Roullier <christophe.roullier@foss.st.com>
Date: Tue, 11 Jun 2024 10:35:59 +0200
Subject: dt-bindings: net: add STM32MP13 compatible in documentation for stm32

New STM32 SOC have 2 GMACs instances.
GMAC IP version is SNPS 4.20.

Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 .../devicetree/bindings/net/stm32-dwmac.yaml       | 43 ++++++++++++++++++----
 1 file changed, 36 insertions(+), 7 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/stm32-dwmac.yaml b/Documentation/devicetree/bindings/net/stm32-dwmac.yaml
index 7ccf75676b6d..f6e5e0626a3f 100644
--- a/Documentation/devicetree/bindings/net/stm32-dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/stm32-dwmac.yaml
@@ -22,18 +22,17 @@ select:
         enum:
           - st,stm32-dwmac
           - st,stm32mp1-dwmac
+          - st,stm32mp13-dwmac
   required:
     - compatible
 
-allOf:
-  - $ref: snps,dwmac.yaml#
-
 properties:
   compatible:
     oneOf:
       - items:
           - enum:
               - st,stm32mp1-dwmac
+              - st,stm32mp13-dwmac
           - const: snps,dwmac-4.20a
       - items:
           - enum:
@@ -75,12 +74,15 @@ properties:
   st,syscon:
     $ref: /schemas/types.yaml#/definitions/phandle-array
     items:
-      - items:
+      - minItems: 2
+        items:
           - description: phandle to the syscon node which encompases the glue register
           - description: offset of the control register
+          - description: field to set mask in register
     description:
       Should be phandle/offset pair. The phandle to the syscon node which
-      encompases the glue register, and the offset of the control register
+      encompases the glue register, the offset of the control register and
+      the mask to set bitfield in control register
 
   st,ext-phyclk:
     description:
@@ -112,12 +114,39 @@ required:
 
 unevaluatedProperties: false
 
+allOf:
+  - $ref: snps,dwmac.yaml#
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - st,stm32mp1-dwmac
+              - st,stm32-dwmac
+    then:
+      properties:
+        st,syscon:
+          items:
+            minItems: 2
+            maxItems: 2
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - st,stm32mp13-dwmac
+    then:
+      properties:
+        st,syscon:
+          items:
+            minItems: 3
+            maxItems: 3
+
 examples:
   - |
     #include <dt-bindings/interrupt-controller/arm-gic.h>
     #include <dt-bindings/clock/stm32mp1-clks.h>
-    #include <dt-bindings/reset/stm32mp1-resets.h>
-    #include <dt-bindings/mfd/stm32h7-rcc.h>
     //Example 1
      ethernet0: ethernet@5800a000 {
            compatible = "st,stm32mp1-dwmac", "snps,dwmac-4.20a";
-- 
cgit v1.2.3-58-ga151


From 582ac134963e2d5cf6c45db027e156fcfb7f7678 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Tue, 11 Jun 2024 10:36:00 +0200
Subject: net: stmmac: dwmac-stm32: Separate out external clock rate validation

Pull the external clock frequency validation into a separate function,
to avoid conflating it with external clock DT property decoding and
clock mux register configuration. This should make the code easier to
read and understand.

This does change the code behavior slightly. The clock mux PMCR register
setting now depends solely on the DT properties which configure the clock
mux between external clock and internal RCC generated clock. The mux PMCR
register settings no longer depend on the supplied clock frequency, that
supplied clock frequency is now only validated, and if the clock frequency
is invalid for a mode, it is rejected.

Previously, the code would switch the PMCR register clock mux to internal
RCC generated clock if external clock couldn't provide suitable frequency,
without checking whether the RCC generated clock frequency is correct. Such
behavior is risky at best, user should have configured their clock correctly
in the first place, so this behavior is removed here.

Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 51 ++++++++++++++++++-----
 1 file changed, 41 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
index c92dfc4ecf57..2fd2620ebed6 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
@@ -157,25 +157,54 @@ static int stm32_dwmac_init(struct plat_stmmacenet_data *plat_dat, bool resume)
 	return stm32_dwmac_clk_enable(dwmac, resume);
 }
 
+static int stm32mp1_validate_ethck_rate(struct plat_stmmacenet_data *plat_dat)
+{
+	struct stm32_dwmac *dwmac = plat_dat->bsp_priv;
+	const u32 clk_rate = clk_get_rate(dwmac->clk_eth_ck);
+
+	switch (plat_dat->mac_interface) {
+	case PHY_INTERFACE_MODE_MII:
+	case PHY_INTERFACE_MODE_GMII:
+		if (clk_rate == ETH_CK_F_25M)
+			return 0;
+		break;
+	case PHY_INTERFACE_MODE_RMII:
+		if (clk_rate == ETH_CK_F_25M || clk_rate == ETH_CK_F_50M)
+			return 0;
+		break;
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		if (clk_rate == ETH_CK_F_25M || clk_rate == ETH_CK_F_125M)
+			return 0;
+		break;
+	default:
+		break;
+	}
+
+	dev_err(dwmac->dev, "Mode %s does not match eth-ck frequency %d Hz",
+		phy_modes(plat_dat->mac_interface), clk_rate);
+	return -EINVAL;
+}
+
 static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
 {
 	struct stm32_dwmac *dwmac = plat_dat->bsp_priv;
-	u32 reg = dwmac->mode_reg, clk_rate;
-	int val;
+	u32 reg = dwmac->mode_reg;
+	int val, ret;
 
-	clk_rate = clk_get_rate(dwmac->clk_eth_ck);
 	dwmac->enable_eth_ck = false;
 	switch (plat_dat->mac_interface) {
 	case PHY_INTERFACE_MODE_MII:
-		if (clk_rate == ETH_CK_F_25M && dwmac->ext_phyclk)
+		if (dwmac->ext_phyclk)
 			dwmac->enable_eth_ck = true;
 		val = SYSCFG_PMCR_ETH_SEL_MII;
 		pr_debug("SYSCFG init : PHY_INTERFACE_MODE_MII\n");
 		break;
 	case PHY_INTERFACE_MODE_GMII:
 		val = SYSCFG_PMCR_ETH_SEL_GMII;
-		if (clk_rate == ETH_CK_F_25M &&
-		    (dwmac->eth_clk_sel_reg || dwmac->ext_phyclk)) {
+		if (dwmac->eth_clk_sel_reg || dwmac->ext_phyclk) {
 			dwmac->enable_eth_ck = true;
 			val |= SYSCFG_PMCR_ETH_CLK_SEL;
 		}
@@ -183,8 +212,7 @@ static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
 		break;
 	case PHY_INTERFACE_MODE_RMII:
 		val = SYSCFG_PMCR_ETH_SEL_RMII;
-		if ((clk_rate == ETH_CK_F_25M || clk_rate == ETH_CK_F_50M) &&
-		    (dwmac->eth_ref_clk_sel_reg || dwmac->ext_phyclk)) {
+		if (dwmac->eth_ref_clk_sel_reg || dwmac->ext_phyclk) {
 			dwmac->enable_eth_ck = true;
 			val |= SYSCFG_PMCR_ETH_REF_CLK_SEL;
 		}
@@ -195,8 +223,7 @@ static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
 	case PHY_INTERFACE_MODE_RGMII_RXID:
 	case PHY_INTERFACE_MODE_RGMII_TXID:
 		val = SYSCFG_PMCR_ETH_SEL_RGMII;
-		if ((clk_rate == ETH_CK_F_25M || clk_rate == ETH_CK_F_125M) &&
-		    (dwmac->eth_clk_sel_reg || dwmac->ext_phyclk)) {
+		if (dwmac->eth_clk_sel_reg || dwmac->ext_phyclk) {
 			dwmac->enable_eth_ck = true;
 			val |= SYSCFG_PMCR_ETH_CLK_SEL;
 		}
@@ -209,6 +236,10 @@ static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
 		return -EINVAL;
 	}
 
+	ret = stm32mp1_validate_ethck_rate(plat_dat);
+	if (ret)
+		return ret;
+
 	/* Need to update PMCCLRR (clear register) */
 	regmap_write(dwmac->regmap, reg + SYSCFG_PMCCLRR_OFFSET,
 		     dwmac->ops->syscfg_eth_mask);
-- 
cgit v1.2.3-58-ga151


From 63b0aa8ea73f7a15912d5e6e44714f98bda8a03c Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Tue, 11 Jun 2024 10:36:01 +0200
Subject: net: stmmac: dwmac-stm32: Separate out external clock selector

Pull the external clock selector into a separate function, to avoid
conflating it with external clock rate validation and clock mux
register configuration. This should make the code easier to read and
understand.

The dwmac->enable_eth_ck variable in the end indicates whether the MAC
clock are supplied by external oscillator (true) or internal RCC clock
IP (false). The dwmac->enable_eth_ck value is set based on multiple DT
properties, some of them deprecated, some of them specific to bus mode.

The following DT properties and variables are taken into account. In
each case, if the property is present or true, MAC clock is supplied
by external oscillator.
- "st,ext-phyclk", assigned to variable dwmac->ext_phyclk
  - Used in any mode (MII/RMII/GMII/RGMII)
  - The only non-deprecated DT property of the three
- "st,eth-clk-sel", assigned to variable dwmac->eth_clk_sel_reg
  - Valid only in GMII/RGMII mode
  - Deprecated property, backward compatibility only
- "st,eth-ref-clk-sel", assigned to variable dwmac->eth_ref_clk_sel_reg
  - Valid only in RMII mode
  - Deprecated property, backward compatibility only

The stm32mp1_select_ethck_external() function handles the aforementioned
DT properties and sets dwmac->enable_eth_ck accordingly.

The stm32mp1_set_mode() is adjusted to call stm32mp1_select_ethck_external()
first and then only use dwmac->enable_eth_ck to determine hardware clock mux
settings.

No functional change intended.

Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 50 +++++++++++++++++------
 1 file changed, 38 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
index 2fd2620ebed6..767994061ea8 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
@@ -157,6 +157,37 @@ static int stm32_dwmac_init(struct plat_stmmacenet_data *plat_dat, bool resume)
 	return stm32_dwmac_clk_enable(dwmac, resume);
 }
 
+static int stm32mp1_select_ethck_external(struct plat_stmmacenet_data *plat_dat)
+{
+	struct stm32_dwmac *dwmac = plat_dat->bsp_priv;
+
+	switch (plat_dat->mac_interface) {
+	case PHY_INTERFACE_MODE_MII:
+		dwmac->enable_eth_ck = dwmac->ext_phyclk;
+		return 0;
+	case PHY_INTERFACE_MODE_GMII:
+		dwmac->enable_eth_ck = dwmac->eth_clk_sel_reg ||
+				       dwmac->ext_phyclk;
+		return 0;
+	case PHY_INTERFACE_MODE_RMII:
+		dwmac->enable_eth_ck = dwmac->eth_ref_clk_sel_reg ||
+				       dwmac->ext_phyclk;
+		return 0;
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		dwmac->enable_eth_ck = dwmac->eth_clk_sel_reg ||
+				       dwmac->ext_phyclk;
+		return 0;
+	default:
+		dwmac->enable_eth_ck = false;
+		dev_err(dwmac->dev, "Mode %s not supported",
+			phy_modes(plat_dat->mac_interface));
+		return -EINVAL;
+	}
+}
+
 static int stm32mp1_validate_ethck_rate(struct plat_stmmacenet_data *plat_dat)
 {
 	struct stm32_dwmac *dwmac = plat_dat->bsp_priv;
@@ -194,28 +225,25 @@ static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
 	u32 reg = dwmac->mode_reg;
 	int val, ret;
 
-	dwmac->enable_eth_ck = false;
+	ret = stm32mp1_select_ethck_external(plat_dat);
+	if (ret)
+		return ret;
+
 	switch (plat_dat->mac_interface) {
 	case PHY_INTERFACE_MODE_MII:
-		if (dwmac->ext_phyclk)
-			dwmac->enable_eth_ck = true;
 		val = SYSCFG_PMCR_ETH_SEL_MII;
 		pr_debug("SYSCFG init : PHY_INTERFACE_MODE_MII\n");
 		break;
 	case PHY_INTERFACE_MODE_GMII:
 		val = SYSCFG_PMCR_ETH_SEL_GMII;
-		if (dwmac->eth_clk_sel_reg || dwmac->ext_phyclk) {
-			dwmac->enable_eth_ck = true;
+		if (dwmac->enable_eth_ck)
 			val |= SYSCFG_PMCR_ETH_CLK_SEL;
-		}
 		pr_debug("SYSCFG init : PHY_INTERFACE_MODE_GMII\n");
 		break;
 	case PHY_INTERFACE_MODE_RMII:
 		val = SYSCFG_PMCR_ETH_SEL_RMII;
-		if (dwmac->eth_ref_clk_sel_reg || dwmac->ext_phyclk) {
-			dwmac->enable_eth_ck = true;
+		if (dwmac->enable_eth_ck)
 			val |= SYSCFG_PMCR_ETH_REF_CLK_SEL;
-		}
 		pr_debug("SYSCFG init : PHY_INTERFACE_MODE_RMII\n");
 		break;
 	case PHY_INTERFACE_MODE_RGMII:
@@ -223,10 +251,8 @@ static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
 	case PHY_INTERFACE_MODE_RGMII_RXID:
 	case PHY_INTERFACE_MODE_RGMII_TXID:
 		val = SYSCFG_PMCR_ETH_SEL_RGMII;
-		if (dwmac->eth_clk_sel_reg || dwmac->ext_phyclk) {
-			dwmac->enable_eth_ck = true;
+		if (dwmac->enable_eth_ck)
 			val |= SYSCFG_PMCR_ETH_CLK_SEL;
-		}
 		pr_debug("SYSCFG init : PHY_INTERFACE_MODE_RGMII\n");
 		break;
 	default:
-- 
cgit v1.2.3-58-ga151


From 92704f8c0e7a3868bf22a7d36ae205b880195e66 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Tue, 11 Jun 2024 10:36:02 +0200
Subject: net: stmmac: dwmac-stm32: Extract PMCR configuration

Pull the PMCR clock mux configuration into a separate function. This is
the final change of three, which moves external clock rate validation,
external clock selector decoding, and clock mux configuration into
separate functions. This should make the code easier to understand.
No functional change intended.

Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 27 ++++++++++++++---------
 1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
index 767994061ea8..aa413edd1ef7 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
@@ -219,15 +219,11 @@ static int stm32mp1_validate_ethck_rate(struct plat_stmmacenet_data *plat_dat)
 	return -EINVAL;
 }
 
-static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
+static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat)
 {
 	struct stm32_dwmac *dwmac = plat_dat->bsp_priv;
 	u32 reg = dwmac->mode_reg;
-	int val, ret;
-
-	ret = stm32mp1_select_ethck_external(plat_dat);
-	if (ret)
-		return ret;
+	int val;
 
 	switch (plat_dat->mac_interface) {
 	case PHY_INTERFACE_MODE_MII:
@@ -262,10 +258,6 @@ static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
 		return -EINVAL;
 	}
 
-	ret = stm32mp1_validate_ethck_rate(plat_dat);
-	if (ret)
-		return ret;
-
 	/* Need to update PMCCLRR (clear register) */
 	regmap_write(dwmac->regmap, reg + SYSCFG_PMCCLRR_OFFSET,
 		     dwmac->ops->syscfg_eth_mask);
@@ -275,6 +267,21 @@ static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
 				 dwmac->ops->syscfg_eth_mask, val);
 }
 
+static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
+{
+	int ret;
+
+	ret = stm32mp1_select_ethck_external(plat_dat);
+	if (ret)
+		return ret;
+
+	ret = stm32mp1_validate_ethck_rate(plat_dat);
+	if (ret)
+		return ret;
+
+	return stm32mp1_configure_pmcr(plat_dat);
+}
+
 static int stm32mcu_set_mode(struct plat_stmmacenet_data *plat_dat)
 {
 	struct stm32_dwmac *dwmac = plat_dat->bsp_priv;
-- 
cgit v1.2.3-58-ga151


From c60a54b52026bd2c9a88ae00f2aac7a67fed8e38 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Tue, 11 Jun 2024 10:36:03 +0200
Subject: net: stmmac: dwmac-stm32: Clean up the debug prints

Use dev_err()/dev_dbg() and phy_modes() to print PHY mode instead of
pr_debug() and hand-written PHY mode decoding. This way, each debug
print has associated device with it and duplicated mode decoding is
removed.

Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
index aa413edd1ef7..75981ac2cbb5 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
@@ -228,19 +228,16 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat)
 	switch (plat_dat->mac_interface) {
 	case PHY_INTERFACE_MODE_MII:
 		val = SYSCFG_PMCR_ETH_SEL_MII;
-		pr_debug("SYSCFG init : PHY_INTERFACE_MODE_MII\n");
 		break;
 	case PHY_INTERFACE_MODE_GMII:
 		val = SYSCFG_PMCR_ETH_SEL_GMII;
 		if (dwmac->enable_eth_ck)
 			val |= SYSCFG_PMCR_ETH_CLK_SEL;
-		pr_debug("SYSCFG init : PHY_INTERFACE_MODE_GMII\n");
 		break;
 	case PHY_INTERFACE_MODE_RMII:
 		val = SYSCFG_PMCR_ETH_SEL_RMII;
 		if (dwmac->enable_eth_ck)
 			val |= SYSCFG_PMCR_ETH_REF_CLK_SEL;
-		pr_debug("SYSCFG init : PHY_INTERFACE_MODE_RMII\n");
 		break;
 	case PHY_INTERFACE_MODE_RGMII:
 	case PHY_INTERFACE_MODE_RGMII_ID:
@@ -249,15 +246,16 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat)
 		val = SYSCFG_PMCR_ETH_SEL_RGMII;
 		if (dwmac->enable_eth_ck)
 			val |= SYSCFG_PMCR_ETH_CLK_SEL;
-		pr_debug("SYSCFG init : PHY_INTERFACE_MODE_RGMII\n");
 		break;
 	default:
-		pr_debug("SYSCFG init :  Do not manage %d interface\n",
-			 plat_dat->mac_interface);
+		dev_err(dwmac->dev, "Mode %s not supported",
+			phy_modes(plat_dat->mac_interface));
 		/* Do not manage others interfaces */
 		return -EINVAL;
 	}
 
+	dev_dbg(dwmac->dev, "Mode %s", phy_modes(plat_dat->mac_interface));
+
 	/* Need to update PMCCLRR (clear register) */
 	regmap_write(dwmac->regmap, reg + SYSCFG_PMCCLRR_OFFSET,
 		     dwmac->ops->syscfg_eth_mask);
@@ -291,19 +289,19 @@ static int stm32mcu_set_mode(struct plat_stmmacenet_data *plat_dat)
 	switch (plat_dat->mac_interface) {
 	case PHY_INTERFACE_MODE_MII:
 		val = SYSCFG_MCU_ETH_SEL_MII;
-		pr_debug("SYSCFG init : PHY_INTERFACE_MODE_MII\n");
 		break;
 	case PHY_INTERFACE_MODE_RMII:
 		val = SYSCFG_MCU_ETH_SEL_RMII;
-		pr_debug("SYSCFG init : PHY_INTERFACE_MODE_RMII\n");
 		break;
 	default:
-		pr_debug("SYSCFG init :  Do not manage %d interface\n",
-			 plat_dat->mac_interface);
+		dev_err(dwmac->dev, "Mode %s not supported",
+			phy_modes(plat_dat->mac_interface));
 		/* Do not manage others interfaces */
 		return -EINVAL;
 	}
 
+	dev_dbg(dwmac->dev, "Mode %s", phy_modes(plat_dat->mac_interface));
+
 	return regmap_update_bits(dwmac->regmap, reg,
 				 dwmac->ops->syscfg_eth_mask, val << 23);
 }
-- 
cgit v1.2.3-58-ga151


From cbfad55322ceee283ff1dd053725b93c86018b88 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Tue, 11 Jun 2024 10:36:04 +0200
Subject: net: stmmac: dwmac-stm32: Fix Mhz to MHz

Trivial, fix up the comments using 'Mhz' to 'MHz'.
No functional change.

Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
index 75981ac2cbb5..bed2be129b2d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
@@ -58,7 +58,7 @@
  * Below table summarizes the clock requirement and clock sources for
  * supported phy interface modes.
  * __________________________________________________________________________
- *|PHY_MODE | Normal | PHY wo crystal|   PHY wo crystal   |No 125Mhz from PHY|
+ *|PHY_MODE | Normal | PHY wo crystal|   PHY wo crystal   |No 125MHz from PHY|
  *|         |        |      25MHz    |        50MHz       |                  |
  * ---------------------------------------------------------------------------
  *|  MII    |	 -   |     eth-ck    |	      n/a	  |	  n/a        |
@@ -367,7 +367,7 @@ static int stm32mp1_parse_data(struct stm32_dwmac *dwmac,
 	/* Gigabit Ethernet 125MHz clock selection. */
 	dwmac->eth_clk_sel_reg = of_property_read_bool(np, "st,eth-clk-sel");
 
-	/* Ethernet 50Mhz RMII clock selection */
+	/* Ethernet 50MHz RMII clock selection */
 	dwmac->eth_ref_clk_sel_reg =
 		of_property_read_bool(np, "st,eth-ref-clk-sel");
 
-- 
cgit v1.2.3-58-ga151


From 4f37dc467ffe5b56ab6b27cca207246172c0d861 Mon Sep 17 00:00:00 2001
From: Christophe Roullier <christophe.roullier@foss.st.com>
Date: Tue, 11 Jun 2024 10:36:05 +0200
Subject: net: stmmac: dwmac-stm32: Mask support for PMCR configuration

Add possibility to have second argument in syscon property to manage
mask. This mask will be used to address right BITFIELDS of PMCR register.

Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 28 +++++++++++++++--------
 1 file changed, 19 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
index bed2be129b2d..09ff0be0bdcd 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
@@ -90,6 +90,7 @@ struct stm32_dwmac {
 	int eth_ref_clk_sel_reg;
 	int irq_pwr_wakeup;
 	u32 mode_reg;		 /* MAC glue-logic mode register */
+	u32 mode_mask;
 	struct regmap *regmap;
 	u32 speed;
 	const struct stm32_ops *ops;
@@ -102,8 +103,8 @@ struct stm32_ops {
 	void (*resume)(struct stm32_dwmac *dwmac);
 	int (*parse_data)(struct stm32_dwmac *dwmac,
 			  struct device *dev);
-	u32 syscfg_eth_mask;
 	bool clk_rx_enable_in_suspend;
+	u32 syscfg_clr_off;
 };
 
 static int stm32_dwmac_clk_enable(struct stm32_dwmac *dwmac, bool resume)
@@ -256,13 +257,16 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat)
 
 	dev_dbg(dwmac->dev, "Mode %s", phy_modes(plat_dat->mac_interface));
 
+	/* Shift value at correct ethernet MAC offset in SYSCFG_PMCSETR */
+	val <<= ffs(dwmac->mode_mask) - ffs(SYSCFG_MP1_ETH_MASK);
+
 	/* Need to update PMCCLRR (clear register) */
-	regmap_write(dwmac->regmap, reg + SYSCFG_PMCCLRR_OFFSET,
-		     dwmac->ops->syscfg_eth_mask);
+	regmap_write(dwmac->regmap, dwmac->ops->syscfg_clr_off,
+		     dwmac->mode_mask);
 
 	/* Update PMCSETR (set register) */
 	return regmap_update_bits(dwmac->regmap, reg,
-				 dwmac->ops->syscfg_eth_mask, val);
+				 dwmac->mode_mask, val);
 }
 
 static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
@@ -303,7 +307,7 @@ static int stm32mcu_set_mode(struct plat_stmmacenet_data *plat_dat)
 	dev_dbg(dwmac->dev, "Mode %s", phy_modes(plat_dat->mac_interface));
 
 	return regmap_update_bits(dwmac->regmap, reg,
-				 dwmac->ops->syscfg_eth_mask, val << 23);
+				 SYSCFG_MCU_ETH_MASK, val << 23);
 }
 
 static void stm32_dwmac_clk_disable(struct stm32_dwmac *dwmac, bool suspend)
@@ -348,8 +352,15 @@ static int stm32_dwmac_parse_data(struct stm32_dwmac *dwmac,
 		return PTR_ERR(dwmac->regmap);
 
 	err = of_property_read_u32_index(np, "st,syscon", 1, &dwmac->mode_reg);
-	if (err)
+	if (err) {
 		dev_err(dev, "Can't get sysconfig mode offset (%d)\n", err);
+		return err;
+	}
+
+	dwmac->mode_mask = SYSCFG_MP1_ETH_MASK;
+	err = of_property_read_u32_index(np, "st,syscon", 2, &dwmac->mode_mask);
+	if (err)
+		dev_dbg(dev, "Warning sysconfig register mask not set\n");
 
 	return err;
 }
@@ -540,8 +551,7 @@ static SIMPLE_DEV_PM_OPS(stm32_dwmac_pm_ops,
 	stm32_dwmac_suspend, stm32_dwmac_resume);
 
 static struct stm32_ops stm32mcu_dwmac_data = {
-	.set_mode = stm32mcu_set_mode,
-	.syscfg_eth_mask = SYSCFG_MCU_ETH_MASK
+	.set_mode = stm32mcu_set_mode
 };
 
 static struct stm32_ops stm32mp1_dwmac_data = {
@@ -549,7 +559,7 @@ static struct stm32_ops stm32mp1_dwmac_data = {
 	.suspend = stm32mp1_suspend,
 	.resume = stm32mp1_resume,
 	.parse_data = stm32mp1_parse_data,
-	.syscfg_eth_mask = SYSCFG_MP1_ETH_MASK,
+	.syscfg_clr_off = 0x44,
 	.clk_rx_enable_in_suspend = true
 };
 
-- 
cgit v1.2.3-58-ga151


From 50bbc039311464af9a4c90be828750450e8cbed3 Mon Sep 17 00:00:00 2001
From: Christophe Roullier <christophe.roullier@foss.st.com>
Date: Tue, 11 Jun 2024 10:36:06 +0200
Subject: net: stmmac: dwmac-stm32: add management of stm32mp13 for stm32

Add Ethernet support for STM32MP13.
STM32MP13 is STM32 SOC with 2 GMACs instances.
GMAC IP version is SNPS 4.20.
GMAC IP configure with 1 RX and 1 TX queue.
DMA HW capability register supported
RX Checksum Offload Engine supported
TX Checksum insertion supported
Wake-Up On Lan supported
TSO supported

Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 32 ++++++++++++++++++++---
 1 file changed, 28 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
index 09ff0be0bdcd..b2db0e26c4e4 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
@@ -104,6 +104,7 @@ struct stm32_ops {
 	int (*parse_data)(struct stm32_dwmac *dwmac,
 			  struct device *dev);
 	bool clk_rx_enable_in_suspend;
+	bool is_mp13;
 	u32 syscfg_clr_off;
 };
 
@@ -224,11 +225,18 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat)
 {
 	struct stm32_dwmac *dwmac = plat_dat->bsp_priv;
 	u32 reg = dwmac->mode_reg;
-	int val;
+	int val = 0;
 
 	switch (plat_dat->mac_interface) {
 	case PHY_INTERFACE_MODE_MII:
-		val = SYSCFG_PMCR_ETH_SEL_MII;
+		/*
+		 * STM32MP15xx supports both MII and GMII, STM32MP13xx MII only.
+		 * SYSCFG_PMCSETR ETH_SELMII is present only on STM32MP15xx and
+		 * acts as a selector between 0:GMII and 1:MII. As STM32MP13xx
+		 * supports only MII, ETH_SELMII is not present.
+		 */
+		if (!dwmac->ops->is_mp13)  /* Select MII mode on STM32MP15xx */
+			val |= SYSCFG_PMCR_ETH_SEL_MII;
 		break;
 	case PHY_INTERFACE_MODE_GMII:
 		val = SYSCFG_PMCR_ETH_SEL_GMII;
@@ -359,8 +367,12 @@ static int stm32_dwmac_parse_data(struct stm32_dwmac *dwmac,
 
 	dwmac->mode_mask = SYSCFG_MP1_ETH_MASK;
 	err = of_property_read_u32_index(np, "st,syscon", 2, &dwmac->mode_mask);
-	if (err)
-		dev_dbg(dev, "Warning sysconfig register mask not set\n");
+	if (err) {
+		if (dwmac->ops->is_mp13)
+			dev_err(dev, "Sysconfig register mask must be set (%d)\n", err);
+		else
+			dev_dbg(dev, "Warning sysconfig register mask not set\n");
+	}
 
 	return err;
 }
@@ -560,12 +572,24 @@ static struct stm32_ops stm32mp1_dwmac_data = {
 	.resume = stm32mp1_resume,
 	.parse_data = stm32mp1_parse_data,
 	.syscfg_clr_off = 0x44,
+	.is_mp13 = false,
+	.clk_rx_enable_in_suspend = true
+};
+
+static struct stm32_ops stm32mp13_dwmac_data = {
+	.set_mode = stm32mp1_set_mode,
+	.suspend = stm32mp1_suspend,
+	.resume = stm32mp1_resume,
+	.parse_data = stm32mp1_parse_data,
+	.syscfg_clr_off = 0x08,
+	.is_mp13 = true,
 	.clk_rx_enable_in_suspend = true
 };
 
 static const struct of_device_id stm32_dwmac_match[] = {
 	{ .compatible = "st,stm32-dwmac", .data = &stm32mcu_dwmac_data},
 	{ .compatible = "st,stm32mp1-dwmac", .data = &stm32mp1_dwmac_data},
+	{ .compatible = "st,stm32mp13-dwmac", .data = &stm32mp13_dwmac_data},
 	{ }
 };
 MODULE_DEVICE_TABLE(of, stm32_dwmac_match);
-- 
cgit v1.2.3-58-ga151


From 404dbd26322f50c8123bf5bff9a409356889035f Mon Sep 17 00:00:00 2001
From: Chris Lew <quic_clew@quicinc.com>
Date: Wed, 12 Jun 2024 12:01:56 +0530
Subject: net: qrtr: ns: Ignore ENODEV failures in ns

Ignore the ENODEV failures returned by kernel_sendmsg(). These errors
indicate that either the local port has been closed or the remote has
gone down. Neither of these scenarios are fatal and will eventually be
handled through packets that are later queued on the control port.

Signed-off-by: Chris Lew <quic_clew@quicinc.com>
Signed-off-by: Sarannya Sasikumar <quic_sarannya@quicinc.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240612063156.1377210-1-quic_sarannya@quicinc.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/qrtr/ns.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/net/qrtr/ns.c b/net/qrtr/ns.c
index 654a3cc0d347..3de9350cbf30 100644
--- a/net/qrtr/ns.c
+++ b/net/qrtr/ns.c
@@ -132,8 +132,8 @@ static int service_announce_new(struct sockaddr_qrtr *dest,
 	return kernel_sendmsg(qrtr_ns.sock, &msg, &iv, 1, sizeof(pkt));
 }
 
-static int service_announce_del(struct sockaddr_qrtr *dest,
-				struct qrtr_server *srv)
+static void service_announce_del(struct sockaddr_qrtr *dest,
+				 struct qrtr_server *srv)
 {
 	struct qrtr_ctrl_pkt pkt;
 	struct msghdr msg = { };
@@ -157,10 +157,10 @@ static int service_announce_del(struct sockaddr_qrtr *dest,
 	msg.msg_namelen = sizeof(*dest);
 
 	ret = kernel_sendmsg(qrtr_ns.sock, &msg, &iv, 1, sizeof(pkt));
-	if (ret < 0)
+	if (ret < 0 && ret != -ENODEV)
 		pr_err("failed to announce del service\n");
 
-	return ret;
+	return;
 }
 
 static void lookup_notify(struct sockaddr_qrtr *to, struct qrtr_server *srv,
@@ -188,7 +188,7 @@ static void lookup_notify(struct sockaddr_qrtr *to, struct qrtr_server *srv,
 	msg.msg_namelen = sizeof(*to);
 
 	ret = kernel_sendmsg(qrtr_ns.sock, &msg, &iv, 1, sizeof(pkt));
-	if (ret < 0)
+	if (ret < 0 && ret != -ENODEV)
 		pr_err("failed to send lookup notification\n");
 }
 
@@ -207,6 +207,9 @@ static int announce_servers(struct sockaddr_qrtr *sq)
 	xa_for_each(&node->servers, index, srv) {
 		ret = service_announce_new(sq, srv);
 		if (ret < 0) {
+			if (ret == -ENODEV)
+				continue;
+
 			pr_err("failed to announce new service\n");
 			return ret;
 		}
@@ -369,7 +372,7 @@ static int ctrl_cmd_bye(struct sockaddr_qrtr *from)
 		msg.msg_namelen = sizeof(sq);
 
 		ret = kernel_sendmsg(qrtr_ns.sock, &msg, &iv, 1, sizeof(pkt));
-		if (ret < 0) {
+		if (ret < 0 && ret != -ENODEV) {
 			pr_err("failed to send bye cmd\n");
 			return ret;
 		}
@@ -443,7 +446,7 @@ static int ctrl_cmd_del_client(struct sockaddr_qrtr *from,
 		msg.msg_namelen = sizeof(sq);
 
 		ret = kernel_sendmsg(qrtr_ns.sock, &msg, &iv, 1, sizeof(pkt));
-		if (ret < 0) {
+		if (ret < 0 && ret != -ENODEV) {
 			pr_err("failed to send del client cmd\n");
 			return ret;
 		}
-- 
cgit v1.2.3-58-ga151


From 124e8c2b1b5d08a10d3a44ed082eaaf98a78c91f Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Wed, 12 Jun 2024 18:38:12 -0700
Subject: bpf: Relax tuple len requirement for sk helpers.

__bpf_skc_lookup() safely handles incorrect values of tuple len,
hence we can allow zero to be passed as tuple len.
This patch alone doesn't make an observable verifier difference.
It's a trivial improvement that might simplify bpf programs.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240613013815.953-2-alexei.starovoitov@gmail.com
---
 net/core/filter.c | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/net/core/filter.c b/net/core/filter.c
index 73722790cee3..f1c37c85b858 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -6815,7 +6815,7 @@ static const struct bpf_func_proto bpf_skc_lookup_tcp_proto = {
 	.ret_type	= RET_PTR_TO_SOCK_COMMON_OR_NULL,
 	.arg1_type	= ARG_PTR_TO_CTX,
 	.arg2_type	= ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type	= ARG_CONST_SIZE,
+	.arg3_type	= ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type	= ARG_ANYTHING,
 	.arg5_type	= ARG_ANYTHING,
 };
@@ -6834,7 +6834,7 @@ static const struct bpf_func_proto bpf_sk_lookup_tcp_proto = {
 	.ret_type	= RET_PTR_TO_SOCKET_OR_NULL,
 	.arg1_type	= ARG_PTR_TO_CTX,
 	.arg2_type	= ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type	= ARG_CONST_SIZE,
+	.arg3_type	= ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type	= ARG_ANYTHING,
 	.arg5_type	= ARG_ANYTHING,
 };
@@ -6853,7 +6853,7 @@ static const struct bpf_func_proto bpf_sk_lookup_udp_proto = {
 	.ret_type	= RET_PTR_TO_SOCKET_OR_NULL,
 	.arg1_type	= ARG_PTR_TO_CTX,
 	.arg2_type	= ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type	= ARG_CONST_SIZE,
+	.arg3_type	= ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type	= ARG_ANYTHING,
 	.arg5_type	= ARG_ANYTHING,
 };
@@ -6877,7 +6877,7 @@ static const struct bpf_func_proto bpf_tc_skc_lookup_tcp_proto = {
 	.ret_type	= RET_PTR_TO_SOCK_COMMON_OR_NULL,
 	.arg1_type	= ARG_PTR_TO_CTX,
 	.arg2_type	= ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type	= ARG_CONST_SIZE,
+	.arg3_type	= ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type	= ARG_ANYTHING,
 	.arg5_type	= ARG_ANYTHING,
 };
@@ -6901,7 +6901,7 @@ static const struct bpf_func_proto bpf_tc_sk_lookup_tcp_proto = {
 	.ret_type	= RET_PTR_TO_SOCKET_OR_NULL,
 	.arg1_type	= ARG_PTR_TO_CTX,
 	.arg2_type	= ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type	= ARG_CONST_SIZE,
+	.arg3_type	= ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type	= ARG_ANYTHING,
 	.arg5_type	= ARG_ANYTHING,
 };
@@ -6925,7 +6925,7 @@ static const struct bpf_func_proto bpf_tc_sk_lookup_udp_proto = {
 	.ret_type	= RET_PTR_TO_SOCKET_OR_NULL,
 	.arg1_type	= ARG_PTR_TO_CTX,
 	.arg2_type	= ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type	= ARG_CONST_SIZE,
+	.arg3_type	= ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type	= ARG_ANYTHING,
 	.arg5_type	= ARG_ANYTHING,
 };
@@ -6963,7 +6963,7 @@ static const struct bpf_func_proto bpf_xdp_sk_lookup_udp_proto = {
 	.ret_type       = RET_PTR_TO_SOCKET_OR_NULL,
 	.arg1_type      = ARG_PTR_TO_CTX,
 	.arg2_type      = ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type      = ARG_CONST_SIZE,
+	.arg3_type      = ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type      = ARG_ANYTHING,
 	.arg5_type      = ARG_ANYTHING,
 };
@@ -6987,7 +6987,7 @@ static const struct bpf_func_proto bpf_xdp_skc_lookup_tcp_proto = {
 	.ret_type       = RET_PTR_TO_SOCK_COMMON_OR_NULL,
 	.arg1_type      = ARG_PTR_TO_CTX,
 	.arg2_type      = ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type      = ARG_CONST_SIZE,
+	.arg3_type      = ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type      = ARG_ANYTHING,
 	.arg5_type      = ARG_ANYTHING,
 };
@@ -7011,7 +7011,7 @@ static const struct bpf_func_proto bpf_xdp_sk_lookup_tcp_proto = {
 	.ret_type       = RET_PTR_TO_SOCKET_OR_NULL,
 	.arg1_type      = ARG_PTR_TO_CTX,
 	.arg2_type      = ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type      = ARG_CONST_SIZE,
+	.arg3_type      = ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type      = ARG_ANYTHING,
 	.arg5_type      = ARG_ANYTHING,
 };
@@ -7031,7 +7031,7 @@ static const struct bpf_func_proto bpf_sock_addr_skc_lookup_tcp_proto = {
 	.ret_type	= RET_PTR_TO_SOCK_COMMON_OR_NULL,
 	.arg1_type	= ARG_PTR_TO_CTX,
 	.arg2_type	= ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type	= ARG_CONST_SIZE,
+	.arg3_type	= ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type	= ARG_ANYTHING,
 	.arg5_type	= ARG_ANYTHING,
 };
@@ -7050,7 +7050,7 @@ static const struct bpf_func_proto bpf_sock_addr_sk_lookup_tcp_proto = {
 	.ret_type	= RET_PTR_TO_SOCKET_OR_NULL,
 	.arg1_type	= ARG_PTR_TO_CTX,
 	.arg2_type	= ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type	= ARG_CONST_SIZE,
+	.arg3_type	= ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type	= ARG_ANYTHING,
 	.arg5_type	= ARG_ANYTHING,
 };
@@ -7069,7 +7069,7 @@ static const struct bpf_func_proto bpf_sock_addr_sk_lookup_udp_proto = {
 	.ret_type	= RET_PTR_TO_SOCKET_OR_NULL,
 	.arg1_type	= ARG_PTR_TO_CTX,
 	.arg2_type	= ARG_PTR_TO_MEM | MEM_RDONLY,
-	.arg3_type	= ARG_CONST_SIZE,
+	.arg3_type	= ARG_CONST_SIZE_OR_ZERO,
 	.arg4_type	= ARG_ANYTHING,
 	.arg5_type	= ARG_ANYTHING,
 };
-- 
cgit v1.2.3-58-ga151


From 98d7ca374ba4b39e7535613d40e159f09ca14da2 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Wed, 12 Jun 2024 18:38:13 -0700
Subject: bpf: Track delta between "linked" registers.

Compilers can generate the code
  r1 = r2
  r1 += 0x1
  if r2 < 1000 goto ...
  use knowledge of r2 range in subsequent r1 operations

So remember constant delta between r2 and r1 and update r1 after 'if' condition.

Unfortunately LLVM still uses this pattern for loops with 'can_loop' construct:
for (i = 0; i < 1000 && can_loop; i++)

The "undo" pass was introduced in LLVM
https://reviews.llvm.org/D121937
to prevent this optimization, but it cannot cover all cases.
Instead of fighting middle end optimizer in BPF backend teach the verifier
about this pattern.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240613013815.953-3-alexei.starovoitov@gmail.com
---
 include/linux/bpf_verifier.h                   | 12 +++-
 kernel/bpf/log.c                               |  4 +-
 kernel/bpf/verifier.c                          | 95 +++++++++++++++++++++++---
 tools/testing/selftests/bpf/verifier/precise.c | 22 +++---
 4 files changed, 109 insertions(+), 24 deletions(-)

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index 50aa87f8d77f..2b54e25d2364 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -73,7 +73,10 @@ enum bpf_iter_state {
 struct bpf_reg_state {
 	/* Ordering of fields matters.  See states_equal() */
 	enum bpf_reg_type type;
-	/* Fixed part of pointer offset, pointer types only */
+	/*
+	 * Fixed part of pointer offset, pointer types only.
+	 * Or constant delta between "linked" scalars with the same ID.
+	 */
 	s32 off;
 	union {
 		/* valid when type == PTR_TO_PACKET */
@@ -167,6 +170,13 @@ struct bpf_reg_state {
 	 * Similarly to dynptrs, we use ID to track "belonging" of a reference
 	 * to a specific instance of bpf_iter.
 	 */
+	/*
+	 * Upper bit of ID is used to remember relationship between "linked"
+	 * registers. Example:
+	 * r1 = r2;    both will have r1->id == r2->id == N
+	 * r1 += 10;   r1->id == N | BPF_ADD_CONST and r1->off == 10
+	 */
+#define BPF_ADD_CONST (1U << 31)
 	u32 id;
 	/* PTR_TO_SOCKET and PTR_TO_TCP_SOCK could be a ptr returned
 	 * from a pointer-cast helper, bpf_sk_fullsock() and
diff --git a/kernel/bpf/log.c b/kernel/bpf/log.c
index 4bd8f17a9f24..3f4ae92e549f 100644
--- a/kernel/bpf/log.c
+++ b/kernel/bpf/log.c
@@ -708,7 +708,9 @@ static void print_reg_state(struct bpf_verifier_env *env,
 		verbose(env, "%s", btf_type_name(reg->btf, reg->btf_id));
 	verbose(env, "(");
 	if (reg->id)
-		verbose_a("id=%d", reg->id);
+		verbose_a("id=%d", reg->id & ~BPF_ADD_CONST);
+	if (reg->id & BPF_ADD_CONST)
+		verbose(env, "%+d", reg->off);
 	if (reg->ref_obj_id)
 		verbose_a("ref_obj_id=%d", reg->ref_obj_id);
 	if (type_is_non_owning_ref(reg->type))
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index e857b08e1f2d..dcbbf5f64c5d 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -3991,7 +3991,7 @@ static bool idset_contains(struct bpf_idset *s, u32 id)
 	u32 i;
 
 	for (i = 0; i < s->count; ++i)
-		if (s->ids[i] == id)
+		if (s->ids[i] == (id & ~BPF_ADD_CONST))
 			return true;
 
 	return false;
@@ -4001,7 +4001,7 @@ static int idset_push(struct bpf_idset *s, u32 id)
 {
 	if (WARN_ON_ONCE(s->count >= ARRAY_SIZE(s->ids)))
 		return -EFAULT;
-	s->ids[s->count++] = id;
+	s->ids[s->count++] = id & ~BPF_ADD_CONST;
 	return 0;
 }
 
@@ -4438,8 +4438,20 @@ static bool __is_pointer_value(bool allow_ptr_leaks,
 static void assign_scalar_id_before_mov(struct bpf_verifier_env *env,
 					struct bpf_reg_state *src_reg)
 {
-	if (src_reg->type == SCALAR_VALUE && !src_reg->id &&
-	    !tnum_is_const(src_reg->var_off))
+	if (src_reg->type != SCALAR_VALUE)
+		return;
+
+	if (src_reg->id & BPF_ADD_CONST) {
+		/*
+		 * The verifier is processing rX = rY insn and
+		 * rY->id has special linked register already.
+		 * Cleared it, since multiple rX += const are not supported.
+		 */
+		src_reg->id = 0;
+		src_reg->off = 0;
+	}
+
+	if (!src_reg->id && !tnum_is_const(src_reg->var_off))
 		/* Ensure that src_reg has a valid ID that will be copied to
 		 * dst_reg and then will be used by find_equal_scalars() to
 		 * propagate min/max range.
@@ -14042,6 +14054,7 @@ static int adjust_reg_min_max_vals(struct bpf_verifier_env *env,
 	struct bpf_func_state *state = vstate->frame[vstate->curframe];
 	struct bpf_reg_state *regs = state->regs, *dst_reg, *src_reg;
 	struct bpf_reg_state *ptr_reg = NULL, off_reg = {0};
+	bool alu32 = (BPF_CLASS(insn->code) != BPF_ALU64);
 	u8 opcode = BPF_OP(insn->code);
 	int err;
 
@@ -14064,11 +14077,7 @@ static int adjust_reg_min_max_vals(struct bpf_verifier_env *env,
 
 	if (dst_reg->type != SCALAR_VALUE)
 		ptr_reg = dst_reg;
-	else
-		/* Make sure ID is cleared otherwise dst_reg min/max could be
-		 * incorrectly propagated into other registers by find_equal_scalars()
-		 */
-		dst_reg->id = 0;
+
 	if (BPF_SRC(insn->code) == BPF_X) {
 		src_reg = &regs[insn->src_reg];
 		if (src_reg->type != SCALAR_VALUE) {
@@ -14132,7 +14141,43 @@ static int adjust_reg_min_max_vals(struct bpf_verifier_env *env,
 		verbose(env, "verifier internal error: no src_reg\n");
 		return -EINVAL;
 	}
-	return adjust_scalar_min_max_vals(env, insn, dst_reg, *src_reg);
+	err = adjust_scalar_min_max_vals(env, insn, dst_reg, *src_reg);
+	if (err)
+		return err;
+	/*
+	 * Compilers can generate the code
+	 * r1 = r2
+	 * r1 += 0x1
+	 * if r2 < 1000 goto ...
+	 * use r1 in memory access
+	 * So remember constant delta between r2 and r1 and update r1 after
+	 * 'if' condition.
+	 */
+	if (env->bpf_capable && BPF_OP(insn->code) == BPF_ADD &&
+	    dst_reg->id && is_reg_const(src_reg, alu32)) {
+		u64 val = reg_const_value(src_reg, alu32);
+
+		if ((dst_reg->id & BPF_ADD_CONST) ||
+		    /* prevent overflow in find_equal_scalars() later */
+		    val > (u32)S32_MAX) {
+			/*
+			 * If the register already went through rX += val
+			 * we cannot accumulate another val into rx->off.
+			 */
+			dst_reg->off = 0;
+			dst_reg->id = 0;
+		} else {
+			dst_reg->id |= BPF_ADD_CONST;
+			dst_reg->off = val;
+		}
+	} else {
+		/*
+		 * Make sure ID is cleared otherwise dst_reg min/max could be
+		 * incorrectly propagated into other registers by find_equal_scalars()
+		 */
+		dst_reg->id = 0;
+	}
+	return 0;
 }
 
 /* check validity of 32-bit and 64-bit arithmetic operations */
@@ -15104,12 +15149,36 @@ static bool try_match_pkt_pointers(const struct bpf_insn *insn,
 static void find_equal_scalars(struct bpf_verifier_state *vstate,
 			       struct bpf_reg_state *known_reg)
 {
+	struct bpf_reg_state fake_reg;
 	struct bpf_func_state *state;
 	struct bpf_reg_state *reg;
 
 	bpf_for_each_reg_in_vstate(vstate, state, reg, ({
-		if (reg->type == SCALAR_VALUE && reg->id == known_reg->id)
+		if (reg->type != SCALAR_VALUE || reg == known_reg)
+			continue;
+		if ((reg->id & ~BPF_ADD_CONST) != (known_reg->id & ~BPF_ADD_CONST))
+			continue;
+		if ((!(reg->id & BPF_ADD_CONST) && !(known_reg->id & BPF_ADD_CONST)) ||
+		    reg->off == known_reg->off) {
 			copy_register_state(reg, known_reg);
+		} else {
+			s32 saved_off = reg->off;
+
+			fake_reg.type = SCALAR_VALUE;
+			__mark_reg_known(&fake_reg, (s32)reg->off - (s32)known_reg->off);
+
+			/* reg = known_reg; reg += delta */
+			copy_register_state(reg, known_reg);
+			/*
+			 * Must preserve off, id and add_const flag,
+			 * otherwise another find_equal_scalars() will be incorrect.
+			 */
+			reg->off = saved_off;
+
+			scalar32_min_max_add(reg, &fake_reg);
+			scalar_min_max_add(reg, &fake_reg);
+			reg->var_off = tnum_add(reg->var_off, fake_reg.var_off);
+		}
 	}));
 }
 
@@ -16738,6 +16807,10 @@ static bool regsafe(struct bpf_verifier_env *env, struct bpf_reg_state *rold,
 		}
 		if (!rold->precise && exact == NOT_EXACT)
 			return true;
+		if ((rold->id & BPF_ADD_CONST) != (rcur->id & BPF_ADD_CONST))
+			return false;
+		if ((rold->id & BPF_ADD_CONST) && (rold->off != rcur->off))
+			return false;
 		/* Why check_ids() for scalar registers?
 		 *
 		 * Consider the following BPF code:
diff --git a/tools/testing/selftests/bpf/verifier/precise.c b/tools/testing/selftests/bpf/verifier/precise.c
index 0a9293a57211..90643ccc221d 100644
--- a/tools/testing/selftests/bpf/verifier/precise.c
+++ b/tools/testing/selftests/bpf/verifier/precise.c
@@ -39,12 +39,12 @@
 	.result = VERBOSE_ACCEPT,
 	.errstr =
 	"mark_precise: frame0: last_idx 26 first_idx 20\
-	mark_precise: frame0: regs=r2 stack= before 25\
-	mark_precise: frame0: regs=r2 stack= before 24\
-	mark_precise: frame0: regs=r2 stack= before 23\
-	mark_precise: frame0: regs=r2 stack= before 22\
-	mark_precise: frame0: regs=r2 stack= before 20\
-	mark_precise: frame0: parent state regs=r2 stack=:\
+	mark_precise: frame0: regs=r2,r9 stack= before 25\
+	mark_precise: frame0: regs=r2,r9 stack= before 24\
+	mark_precise: frame0: regs=r2,r9 stack= before 23\
+	mark_precise: frame0: regs=r2,r9 stack= before 22\
+	mark_precise: frame0: regs=r2,r9 stack= before 20\
+	mark_precise: frame0: parent state regs=r2,r9 stack=:\
 	mark_precise: frame0: last_idx 19 first_idx 10\
 	mark_precise: frame0: regs=r2,r9 stack= before 19\
 	mark_precise: frame0: regs=r9 stack= before 18\
@@ -100,11 +100,11 @@
 	.errstr =
 	"26: (85) call bpf_probe_read_kernel#113\
 	mark_precise: frame0: last_idx 26 first_idx 22\
-	mark_precise: frame0: regs=r2 stack= before 25\
-	mark_precise: frame0: regs=r2 stack= before 24\
-	mark_precise: frame0: regs=r2 stack= before 23\
-	mark_precise: frame0: regs=r2 stack= before 22\
-	mark_precise: frame0: parent state regs=r2 stack=:\
+	mark_precise: frame0: regs=r2,r9 stack= before 25\
+	mark_precise: frame0: regs=r2,r9 stack= before 24\
+	mark_precise: frame0: regs=r2,r9 stack= before 23\
+	mark_precise: frame0: regs=r2,r9 stack= before 22\
+	mark_precise: frame0: parent state regs=r2,r9 stack=:\
 	mark_precise: frame0: last_idx 20 first_idx 20\
 	mark_precise: frame0: regs=r2,r9 stack= before 20\
 	mark_precise: frame0: parent state regs=r2,r9 stack=:\
-- 
cgit v1.2.3-58-ga151


From 6870bdb3f4f2991193449f9de57109b3e263f55c Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Wed, 12 Jun 2024 18:38:14 -0700
Subject: bpf: Support can_loop/cond_break on big endian

Add big endian support for can_loop/cond_break macros.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/bpf/20240613013815.953-4-alexei.starovoitov@gmail.com
---
 tools/testing/selftests/bpf/bpf_experimental.h | 28 ++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/tools/testing/selftests/bpf/bpf_experimental.h b/tools/testing/selftests/bpf/bpf_experimental.h
index 8ee7a00b7c82..eede6fc2ccb4 100644
--- a/tools/testing/selftests/bpf/bpf_experimental.h
+++ b/tools/testing/selftests/bpf/bpf_experimental.h
@@ -351,6 +351,7 @@ l_true:												\
 	l_continue:;					\
 	})
 #else
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
 #define can_loop					\
 	({ __label__ l_break, l_continue;		\
 	bool ret = true;				\
@@ -376,6 +377,33 @@ l_true:												\
 	l_break: break;					\
 	l_continue:;					\
 	})
+#else
+#define can_loop					\
+	({ __label__ l_break, l_continue;		\
+	bool ret = true;				\
+	asm volatile goto("1:.byte 0xe5;		\
+		      .byte 0;				\
+		      .long (((%l[l_break] - 1b - 8) / 8) & 0xffff) << 16;	\
+		      .short 0"				\
+		      :::: l_break);			\
+	goto l_continue;				\
+	l_break: ret = false;				\
+	l_continue:;					\
+	ret;						\
+	})
+
+#define cond_break					\
+	({ __label__ l_break, l_continue;		\
+	asm volatile goto("1:.byte 0xe5;		\
+		      .byte 0;				\
+		      .long (((%l[l_break] - 1b - 8) / 8) & 0xffff) << 16;	\
+		      .short 0"				\
+		      :::: l_break);			\
+	goto l_continue;				\
+	l_break: break;					\
+	l_continue:;					\
+	})
+#endif
 #endif
 
 #ifndef bpf_nop_mov
-- 
cgit v1.2.3-58-ga151


From dedf56d775c0bebbc3003bfb988dddaf0a583c28 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Wed, 12 Jun 2024 18:38:15 -0700
Subject: selftests/bpf: Add tests for add_const

Improve arena based tests and add several C and asm tests
with specific pattern.
These tests would have failed without add_const verifier support.

Also add several loop_inside_iter*() tests that are not related to add_const,
but nice to have.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240613013815.953-5-alexei.starovoitov@gmail.com
---
 tools/testing/selftests/bpf/progs/arena_htab.c     |  16 +-
 .../bpf/progs/verifier_iterating_callbacks.c       | 236 +++++++++++++++++++++
 2 files changed, 249 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/arena_htab.c b/tools/testing/selftests/bpf/progs/arena_htab.c
index cf938cf8c11e..81eaa94afeb0 100644
--- a/tools/testing/selftests/bpf/progs/arena_htab.c
+++ b/tools/testing/selftests/bpf/progs/arena_htab.c
@@ -19,25 +19,35 @@ void __arena *htab_for_user;
 bool skip = false;
 
 int zero = 0;
+char __arena arr1[100000];
+char arr2[1000];
 
 SEC("syscall")
 int arena_htab_llvm(void *ctx)
 {
 #if defined(__BPF_FEATURE_ADDR_SPACE_CAST) || defined(BPF_ARENA_FORCE_ASM)
 	struct htab __arena *htab;
+	char __arena *arr = arr1;
 	__u64 i;
 
 	htab = bpf_alloc(sizeof(*htab));
 	cast_kern(htab);
 	htab_init(htab);
 
+	cast_kern(arr);
+
 	/* first run. No old elems in the table */
-	for (i = zero; i < 1000; i++)
+	for (i = zero; i < 100000 && can_loop; i++) {
 		htab_update_elem(htab, i, i);
+		arr[i] = i;
+	}
 
-	/* should replace all elems with new ones */
-	for (i = zero; i < 1000; i++)
+	/* should replace some elems with new ones */
+	for (i = zero; i < 1000 && can_loop; i++) {
 		htab_update_elem(htab, i, i);
+		/* Access mem to make the verifier use bounded loop logic */
+		arr2[i] = i;
+	}
 	cast_user(htab);
 	htab_for_user = htab;
 #else
diff --git a/tools/testing/selftests/bpf/progs/verifier_iterating_callbacks.c b/tools/testing/selftests/bpf/progs/verifier_iterating_callbacks.c
index bd676d7e615f..53679252e8a1 100644
--- a/tools/testing/selftests/bpf/progs/verifier_iterating_callbacks.c
+++ b/tools/testing/selftests/bpf/progs/verifier_iterating_callbacks.c
@@ -405,4 +405,240 @@ int cond_break5(const void *ctx)
 	return cnt1 > 1 && cnt2 > 1 ? 1 : 0;
 }
 
+#define ARR2_SZ 1000
+SEC(".data.arr2")
+char arr2[ARR2_SZ];
+
+SEC("socket")
+__success __flag(BPF_F_TEST_STATE_FREQ)
+int loop_inside_iter(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, sum = 0;
+	__u64 i = 0;
+
+	bpf_iter_num_new(&it, 0, ARR2_SZ);
+	while ((v = bpf_iter_num_next(&it))) {
+		if (i < ARR2_SZ)
+			sum += arr2[i++];
+	}
+	bpf_iter_num_destroy(&it);
+	return sum;
+}
+
+SEC("socket")
+__success __flag(BPF_F_TEST_STATE_FREQ)
+int loop_inside_iter_signed(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, sum = 0;
+	long i = 0;
+
+	bpf_iter_num_new(&it, 0, ARR2_SZ);
+	while ((v = bpf_iter_num_next(&it))) {
+		if (i < ARR2_SZ && i >= 0)
+			sum += arr2[i++];
+	}
+	bpf_iter_num_destroy(&it);
+	return sum;
+}
+
+volatile const int limit = ARR2_SZ;
+
+SEC("socket")
+__success __flag(BPF_F_TEST_STATE_FREQ)
+int loop_inside_iter_volatile_limit(const void *ctx)
+{
+	struct bpf_iter_num it;
+	int *v, sum = 0;
+	__u64 i = 0;
+
+	bpf_iter_num_new(&it, 0, ARR2_SZ);
+	while ((v = bpf_iter_num_next(&it))) {
+		if (i < limit)
+			sum += arr2[i++];
+	}
+	bpf_iter_num_destroy(&it);
+	return sum;
+}
+
+#define ARR_LONG_SZ 1000
+
+SEC(".data.arr_long")
+long arr_long[ARR_LONG_SZ];
+
+SEC("socket")
+__success
+int test1(const void *ctx)
+{
+	long i;
+
+	for (i = 0; i < ARR_LONG_SZ && can_loop; i++)
+		arr_long[i] = i;
+	return 0;
+}
+
+SEC("socket")
+__success
+int test2(const void *ctx)
+{
+	__u64 i;
+
+	for (i = zero; i < ARR_LONG_SZ && can_loop; i++) {
+		barrier_var(i);
+		arr_long[i] = i;
+	}
+	return 0;
+}
+
+SEC(".data.arr_foo")
+struct {
+	int a;
+	int b;
+} arr_foo[ARR_LONG_SZ];
+
+SEC("socket")
+__success
+int test3(const void *ctx)
+{
+	__u64 i;
+
+	for (i = zero; i < ARR_LONG_SZ && can_loop; i++) {
+		barrier_var(i);
+		arr_foo[i].a = i;
+		arr_foo[i].b = i;
+	}
+	return 0;
+}
+
+SEC("socket")
+__success
+int test4(const void *ctx)
+{
+	long i;
+
+	for (i = zero + ARR_LONG_SZ - 1; i < ARR_LONG_SZ && i >= 0 && can_loop; i--) {
+		barrier_var(i);
+		arr_foo[i].a = i;
+		arr_foo[i].b = i;
+	}
+	return 0;
+}
+
+char buf[10] SEC(".data.buf");
+
+SEC("socket")
+__description("check add const")
+__success
+__naked void check_add_const(void)
+{
+	/* typical LLVM generated loop with may_goto */
+	asm volatile ("			\
+	call %[bpf_ktime_get_ns];	\
+	if r0 > 9 goto l1_%=;		\
+l0_%=:	r1 = %[buf];			\
+	r2 = r0;			\
+	r1 += r2;			\
+	r3 = *(u8 *)(r1 +0);		\
+	.byte 0xe5; /* may_goto */	\
+	.byte 0; /* regs */		\
+	.short 4; /* off of l1_%=: */	\
+	.long 0; /* imm */		\
+	r0 = r2;			\
+	r0 += 1;			\
+	if r2 < 9 goto l0_%=;		\
+	exit;				\
+l1_%=:	r0 = 0;				\
+	exit;				\
+"	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm_ptr(buf)
+	: __clobber_common);
+}
+
+SEC("socket")
+__failure
+__msg("*(u8 *)(r7 +0) = r0")
+__msg("invalid access to map value, value_size=10 off=10 size=1")
+__naked void check_add_const_3regs(void)
+{
+	asm volatile (
+	"r6 = %[buf];"
+	"r7 = %[buf];"
+	"call %[bpf_ktime_get_ns];"
+	"r1 = r0;"              /* link r0.id == r1.id == r2.id */
+	"r2 = r0;"
+	"r1 += 1;"              /* r1 == r0+1 */
+	"r2 += 2;"              /* r2 == r0+2 */
+	"if r0 > 8 goto 1f;"    /* r0 range [0, 8]  */
+	"r6 += r1;"             /* r1 range [1, 9]  */
+	"r7 += r2;"             /* r2 range [2, 10] */
+	"*(u8 *)(r6 +0) = r0;"  /* safe, within bounds   */
+	"*(u8 *)(r7 +0) = r0;"  /* unsafe, out of bounds */
+	"1: exit;"
+	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm_ptr(buf)
+	: __clobber_common);
+}
+
+SEC("socket")
+__failure
+__msg("*(u8 *)(r8 -1) = r0")
+__msg("invalid access to map value, value_size=10 off=10 size=1")
+__naked void check_add_const_3regs_2if(void)
+{
+	asm volatile (
+	"r6 = %[buf];"
+	"r7 = %[buf];"
+	"r8 = %[buf];"
+	"call %[bpf_ktime_get_ns];"
+	"if r0 < 2 goto 1f;"
+	"r1 = r0;"              /* link r0.id == r1.id == r2.id */
+	"r2 = r0;"
+	"r1 += 1;"              /* r1 == r0+1 */
+	"r2 += 2;"              /* r2 == r0+2 */
+	"if r2 > 11 goto 1f;"   /* r2 range [0, 11] -> r0 range [-2, 9]; r1 range [-1, 10] */
+	"if r0 s< 0 goto 1f;"   /* r0 range [0, 9] -> r1 range [1, 10]; r2 range [2, 11]; */
+	"r6 += r0;"             /* r0 range [0, 9]  */
+	"r7 += r1;"             /* r1 range [1, 10] */
+	"r8 += r2;"             /* r2 range [2, 11] */
+	"*(u8 *)(r6 +0) = r0;"  /* safe, within bounds   */
+	"*(u8 *)(r7 -1) = r0;"  /* safe */
+	"*(u8 *)(r8 -1) = r0;"  /* unsafe */
+	"1: exit;"
+	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm_ptr(buf)
+	: __clobber_common);
+}
+
+SEC("socket")
+__failure
+__flag(BPF_F_TEST_STATE_FREQ)
+__naked void check_add_const_regsafe_off(void)
+{
+	asm volatile (
+	"r8 = %[buf];"
+	"call %[bpf_ktime_get_ns];"
+	"r6 = r0;"
+	"call %[bpf_ktime_get_ns];"
+	"r7 = r0;"
+	"call %[bpf_ktime_get_ns];"
+	"r1 = r0;"              /* same ids for r1 and r0 */
+	"if r6 > r7 goto 1f;"   /* this jump can't be predicted */
+	"r1 += 1;"              /* r1.off == +1 */
+	"goto 2f;"
+	"1: r1 += 100;"         /* r1.off == +100 */
+	"goto +0;"              /* verify r1.off in regsafe() after this insn */
+	"2: if r0 > 8 goto 3f;" /* r0 range [0,8], r1 range either [1,9] or [100,108]*/
+	"r8 += r1;"
+	"*(u8 *)(r8 +0) = r0;"  /* potentially unsafe, buf size is 10 */
+	"3: exit;"
+	:
+	: __imm(bpf_ktime_get_ns),
+	  __imm_ptr(buf)
+	: __clobber_common);
+}
+
 char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-58-ga151


From 72421f35540c3e8830be8897ef1b99d2b7aa0981 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 12 Jun 2024 17:32:22 -0700
Subject: eth: lan966x: don't clear unsupported stats

Commit 12c2d0a5b8e2 ("net: lan966x: add ethtool configuration and statistics")
added support for various standard stats. We should not clear the stats
which are not collected by the device. Core code uses a special
initializer to detect when device does not report given stat.

Acked-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Link: https://lore.kernel.org/r/20240613003222.3327368-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c b/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c
index 06811c60d598..c0fc85ac5db3 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c
@@ -376,7 +376,6 @@ static void lan966x_get_eth_mac_stats(struct net_device *dev,
 		lan966x->stats[idx + SYS_COUNT_TX_PMAC_BC];
 	mac_stats->SingleCollisionFrames =
 		lan966x->stats[idx + SYS_COUNT_TX_COL];
-	mac_stats->MultipleCollisionFrames = 0;
 	mac_stats->FramesReceivedOK =
 		lan966x->stats[idx + SYS_COUNT_RX_UC] +
 		lan966x->stats[idx + SYS_COUNT_RX_MC] +
@@ -384,26 +383,19 @@ static void lan966x_get_eth_mac_stats(struct net_device *dev,
 	mac_stats->FrameCheckSequenceErrors =
 		lan966x->stats[idx + SYS_COUNT_RX_CRC] +
 		lan966x->stats[idx + SYS_COUNT_RX_CRC];
-	mac_stats->AlignmentErrors = 0;
 	mac_stats->OctetsTransmittedOK =
 		lan966x->stats[idx + SYS_COUNT_TX_OCT] +
 		lan966x->stats[idx + SYS_COUNT_TX_PMAC_OCT];
 	mac_stats->FramesWithDeferredXmissions =
 		lan966x->stats[idx + SYS_COUNT_TX_MM_HOLD];
-	mac_stats->LateCollisions = 0;
-	mac_stats->FramesAbortedDueToXSColls = 0;
-	mac_stats->FramesLostDueToIntMACXmitError = 0;
-	mac_stats->CarrierSenseErrors = 0;
 	mac_stats->OctetsReceivedOK =
 		lan966x->stats[idx + SYS_COUNT_RX_OCT];
-	mac_stats->FramesLostDueToIntMACRcvError = 0;
 	mac_stats->MulticastFramesXmittedOK =
 		lan966x->stats[idx + SYS_COUNT_TX_MC] +
 		lan966x->stats[idx + SYS_COUNT_TX_PMAC_MC];
 	mac_stats->BroadcastFramesXmittedOK =
 		lan966x->stats[idx + SYS_COUNT_TX_BC] +
 		lan966x->stats[idx + SYS_COUNT_TX_PMAC_BC];
-	mac_stats->FramesWithExcessiveDeferral = 0;
 	mac_stats->MulticastFramesReceivedOK =
 		lan966x->stats[idx + SYS_COUNT_RX_MC];
 	mac_stats->BroadcastFramesReceivedOK =
-- 
cgit v1.2.3-58-ga151


From 9e42a2ea7f6703e2092c39171c2bf1fd7eec0bd3 Mon Sep 17 00:00:00 2001
From: Chris Packham <chris.packham@alliedtelesis.co.nz>
Date: Tue, 11 Jun 2024 17:34:14 +1200
Subject: net: phy: realtek: add support for rtl8224 2.5Gbps PHY

The Realtek RTL8224 PHY is a 2.5Gbps capable PHY. It only uses the
clause 45 MDIO interface and can leverage the support that has already
been added for the other 822x PHYs.

Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Link: https://lore.kernel.org/r/20240611053415.2111723-1-chris.packham@alliedtelesis.co.nz
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/realtek.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c
index 7ab41f95dae5..2174893c974f 100644
--- a/drivers/net/phy/realtek.c
+++ b/drivers/net/phy/realtek.c
@@ -1317,6 +1317,14 @@ static struct phy_driver realtek_drvs[] = {
 		.resume         = rtlgen_resume,
 		.read_page      = rtl821x_read_page,
 		.write_page     = rtl821x_write_page,
+	}, {
+		PHY_ID_MATCH_EXACT(0x001ccad0),
+		.name		= "RTL8224 2.5Gbps PHY",
+		.get_features   = rtl822x_c45_get_features,
+		.config_aneg    = rtl822x_c45_config_aneg,
+		.read_status    = rtl822x_c45_read_status,
+		.suspend        = genphy_c45_pma_suspend,
+		.resume         = rtlgen_c45_resume,
 	}, {
 		PHY_ID_MATCH_EXACT(0x001cc961),
 		.name		= "RTL8366RB Gigabit Ethernet",
-- 
cgit v1.2.3-58-ga151


From e575d3a6dd22123888defb622b1742aa2d45b942 Mon Sep 17 00:00:00 2001
From: Cosmin Ratiu <cratiu@nvidia.com>
Date: Fri, 14 Jun 2024 00:00:31 +0300
Subject: net/mlx5: Correct TASR typo into TSAR

TSAR is the correct spelling (Transmit Scheduling ARbiter).

Signed-off-by: Cosmin Ratiu <cratiu@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240613210036.1125203-2-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c | 2 +-
 include/linux/mlx5/mlx5_ifc.h                     | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
index d2ebe56c3977..20146a2dc7f4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
@@ -531,7 +531,7 @@ static bool esw_qos_element_type_supported(struct mlx5_core_dev *dev, int type)
 	switch (type) {
 	case SCHEDULING_CONTEXT_ELEMENT_TYPE_TSAR:
 		return MLX5_CAP_QOS(dev, esw_element_type) &
-		       ELEMENT_TYPE_CAP_MASK_TASR;
+		       ELEMENT_TYPE_CAP_MASK_TSAR;
 	case SCHEDULING_CONTEXT_ELEMENT_TYPE_VPORT:
 		return MLX5_CAP_QOS(dev, esw_element_type) &
 		       ELEMENT_TYPE_CAP_MASK_VPORT;
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 17acd0f3ca8e..466dcda40bb5 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -3914,7 +3914,7 @@ enum {
 };
 
 enum {
-	ELEMENT_TYPE_CAP_MASK_TASR		= 1 << 0,
+	ELEMENT_TYPE_CAP_MASK_TSAR		= 1 << 0,
 	ELEMENT_TYPE_CAP_MASK_VPORT		= 1 << 1,
 	ELEMENT_TYPE_CAP_MASK_VPORT_TC		= 1 << 2,
 	ELEMENT_TYPE_CAP_MASK_PARA_VPORT_TC	= 1 << 3,
-- 
cgit v1.2.3-58-ga151


From 49d37d05f216ce026ebd297b76a2db5bff7a4b4d Mon Sep 17 00:00:00 2001
From: Chris Mi <cmi@nvidia.com>
Date: Fri, 14 Jun 2024 00:00:32 +0300
Subject: net/mlx5: CT: Separate CT and CT-NAT tuple entries

Currently a ct entry is stored in both ct and ct-nat tables. ct
action is directed to the ct table, while ct nat action is directed
to the nat table. ct-nat entries perform the nat header rewrites,
if required. The current design assures that a ct action will match
in hardware even if the tuple has nat configured, it will just not
execute it. However, storing each connection in two tables increases
the system's memory consumption while reducing its insertion rate.

Offload a connection to either ct or the ct-nat table. Add a miss
fall-through rule from ct-nat table to the ct table allowing ct(nat)
action on non-natted connections.

ct action on natted connections, by default, will be handled by the
software miss path.

Signed-off-by: Paul Blakey <paulb@nvidia.com>
Signed-off-by: Chris Mi <cmi@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240613210036.1125203-3-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c | 187 ++++++++++++++++-----
 1 file changed, 143 insertions(+), 44 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
index fadfa8b50beb..b49d87a51f21 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
@@ -69,6 +69,8 @@ struct mlx5_tc_ct_priv {
 	struct rhashtable ct_tuples_nat_ht;
 	struct mlx5_flow_table *ct;
 	struct mlx5_flow_table *ct_nat;
+	struct mlx5_flow_group *ct_nat_miss_group;
+	struct mlx5_flow_handle *ct_nat_miss_rule;
 	struct mlx5e_post_act *post_act;
 	struct mutex control_lock; /* guards parallel adds/dels */
 	struct mapping_ctx *zone_mapping;
@@ -141,6 +143,8 @@ struct mlx5_ct_counter {
 
 enum {
 	MLX5_CT_ENTRY_FLAG_VALID,
+	MLX5_CT_ENTRY_IN_CT_TABLE,
+	MLX5_CT_ENTRY_IN_CT_NAT_TABLE,
 };
 
 struct mlx5_ct_entry {
@@ -198,9 +202,15 @@ static const struct rhashtable_params tuples_nat_ht_params = {
 };
 
 static bool
-mlx5_tc_ct_entry_has_nat(struct mlx5_ct_entry *entry)
+mlx5_tc_ct_entry_in_ct_table(struct mlx5_ct_entry *entry)
 {
-	return !!(entry->tuple_nat_node.next);
+	return test_bit(MLX5_CT_ENTRY_IN_CT_TABLE, &entry->flags);
+}
+
+static bool
+mlx5_tc_ct_entry_in_ct_nat_table(struct mlx5_ct_entry *entry)
+{
+	return test_bit(MLX5_CT_ENTRY_IN_CT_NAT_TABLE, &entry->flags);
 }
 
 static int
@@ -526,8 +536,10 @@ static void
 mlx5_tc_ct_entry_del_rules(struct mlx5_tc_ct_priv *ct_priv,
 			   struct mlx5_ct_entry *entry)
 {
-	mlx5_tc_ct_entry_del_rule(ct_priv, entry, true);
-	mlx5_tc_ct_entry_del_rule(ct_priv, entry, false);
+	if (mlx5_tc_ct_entry_in_ct_nat_table(entry))
+		mlx5_tc_ct_entry_del_rule(ct_priv, entry, true);
+	if (mlx5_tc_ct_entry_in_ct_table(entry))
+		mlx5_tc_ct_entry_del_rule(ct_priv, entry, false);
 
 	atomic_dec(&ct_priv->debugfs.stats.offloaded);
 }
@@ -814,7 +826,7 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv,
 					      &zone_rule->mh,
 					      zone_restore_id,
 					      nat,
-					      mlx5_tc_ct_entry_has_nat(entry));
+					      mlx5_tc_ct_entry_in_ct_nat_table(entry));
 	if (err) {
 		ct_dbg("Failed to create ct entry mod hdr");
 		goto err_mod_hdr;
@@ -888,7 +900,7 @@ mlx5_tc_ct_entry_replace_rule(struct mlx5_tc_ct_priv *ct_priv,
 	*old_attr = *attr;
 
 	err = mlx5_tc_ct_entry_create_mod_hdr(ct_priv, attr, flow_rule, &mh, zone_restore_id,
-					      nat, mlx5_tc_ct_entry_has_nat(entry));
+					      nat, mlx5_tc_ct_entry_in_ct_nat_table(entry));
 	if (err) {
 		ct_dbg("Failed to create ct entry mod hdr");
 		goto err_mod_hdr;
@@ -957,11 +969,13 @@ static void mlx5_tc_ct_entry_remove_from_tuples(struct mlx5_ct_entry *entry)
 {
 	struct mlx5_tc_ct_priv *ct_priv = entry->ct_priv;
 
-	rhashtable_remove_fast(&ct_priv->ct_tuples_nat_ht,
-			       &entry->tuple_nat_node,
-			       tuples_nat_ht_params);
-	rhashtable_remove_fast(&ct_priv->ct_tuples_ht, &entry->tuple_node,
-			       tuples_ht_params);
+	if (mlx5_tc_ct_entry_in_ct_nat_table(entry))
+		rhashtable_remove_fast(&ct_priv->ct_tuples_nat_ht,
+				       &entry->tuple_nat_node,
+				       tuples_nat_ht_params);
+	if (mlx5_tc_ct_entry_in_ct_table(entry))
+		rhashtable_remove_fast(&ct_priv->ct_tuples_ht, &entry->tuple_node,
+				       tuples_ht_params);
 }
 
 static void mlx5_tc_ct_entry_del(struct mlx5_ct_entry *entry)
@@ -1100,21 +1114,26 @@ mlx5_tc_ct_entry_add_rules(struct mlx5_tc_ct_priv *ct_priv,
 		return err;
 	}
 
-	err = mlx5_tc_ct_entry_add_rule(ct_priv, flow_rule, entry, false,
-					zone_restore_id);
-	if (err)
-		goto err_orig;
+	if (mlx5_tc_ct_entry_in_ct_table(entry)) {
+		err = mlx5_tc_ct_entry_add_rule(ct_priv, flow_rule, entry, false,
+						zone_restore_id);
+		if (err)
+			goto err_orig;
+	}
 
-	err = mlx5_tc_ct_entry_add_rule(ct_priv, flow_rule, entry, true,
-					zone_restore_id);
-	if (err)
-		goto err_nat;
+	if (mlx5_tc_ct_entry_in_ct_nat_table(entry)) {
+		err = mlx5_tc_ct_entry_add_rule(ct_priv, flow_rule, entry, true,
+						zone_restore_id);
+		if (err)
+			goto err_nat;
+	}
 
 	atomic_inc(&ct_priv->debugfs.stats.offloaded);
 	return 0;
 
 err_nat:
-	mlx5_tc_ct_entry_del_rule(ct_priv, entry, false);
+	if (mlx5_tc_ct_entry_in_ct_table(entry))
+		mlx5_tc_ct_entry_del_rule(ct_priv, entry, false);
 err_orig:
 	mlx5_tc_ct_counter_put(ct_priv, entry);
 	return err;
@@ -1128,15 +1147,19 @@ mlx5_tc_ct_entry_replace_rules(struct mlx5_tc_ct_priv *ct_priv,
 {
 	int err;
 
-	err = mlx5_tc_ct_entry_replace_rule(ct_priv, flow_rule, entry, false,
-					    zone_restore_id);
-	if (err)
-		return err;
+	if (mlx5_tc_ct_entry_in_ct_table(entry)) {
+		err = mlx5_tc_ct_entry_replace_rule(ct_priv, flow_rule, entry, false,
+						    zone_restore_id);
+		if (err)
+			return err;
+	}
 
-	err = mlx5_tc_ct_entry_replace_rule(ct_priv, flow_rule, entry, true,
-					    zone_restore_id);
-	if (err)
-		mlx5_tc_ct_entry_del_rule(ct_priv, entry, false);
+	if (mlx5_tc_ct_entry_in_ct_nat_table(entry)) {
+		err = mlx5_tc_ct_entry_replace_rule(ct_priv, flow_rule, entry, true,
+						    zone_restore_id);
+		if (err && mlx5_tc_ct_entry_in_ct_table(entry))
+			mlx5_tc_ct_entry_del_rule(ct_priv, entry, false);
+	}
 	return err;
 }
 
@@ -1224,18 +1247,24 @@ mlx5_tc_ct_block_flow_offload_add(struct mlx5_ct_ft *ft,
 	if (err)
 		goto err_entries;
 
-	err = rhashtable_lookup_insert_fast(&ct_priv->ct_tuples_ht,
-					    &entry->tuple_node,
-					    tuples_ht_params);
-	if (err)
-		goto err_tuple;
-
 	if (memcmp(&entry->tuple, &entry->tuple_nat, sizeof(entry->tuple))) {
 		err = rhashtable_lookup_insert_fast(&ct_priv->ct_tuples_nat_ht,
 						    &entry->tuple_nat_node,
 						    tuples_nat_ht_params);
 		if (err)
 			goto err_tuple_nat;
+
+		set_bit(MLX5_CT_ENTRY_IN_CT_NAT_TABLE, &entry->flags);
+	}
+
+	if (!mlx5_tc_ct_entry_in_ct_nat_table(entry)) {
+		err = rhashtable_lookup_insert_fast(&ct_priv->ct_tuples_ht,
+						    &entry->tuple_node,
+						    tuples_ht_params);
+		if (err)
+			goto err_tuple;
+
+		set_bit(MLX5_CT_ENTRY_IN_CT_TABLE, &entry->flags);
 	}
 	spin_unlock_bh(&ct_priv->ht_lock);
 
@@ -1251,17 +1280,10 @@ mlx5_tc_ct_block_flow_offload_add(struct mlx5_ct_ft *ft,
 
 err_rules:
 	spin_lock_bh(&ct_priv->ht_lock);
-	if (mlx5_tc_ct_entry_has_nat(entry))
-		rhashtable_remove_fast(&ct_priv->ct_tuples_nat_ht,
-				       &entry->tuple_nat_node, tuples_nat_ht_params);
-err_tuple_nat:
-	rhashtable_remove_fast(&ct_priv->ct_tuples_ht,
-			       &entry->tuple_node,
-			       tuples_ht_params);
 err_tuple:
-	rhashtable_remove_fast(&ft->ct_entries_ht,
-			       &entry->node,
-			       cts_ht_params);
+	mlx5_tc_ct_entry_remove_from_tuples(entry);
+err_tuple_nat:
+	rhashtable_remove_fast(&ft->ct_entries_ht, &entry->node, cts_ht_params);
 err_entries:
 	spin_unlock_bh(&ct_priv->ht_lock);
 err_set:
@@ -2149,6 +2171,76 @@ mlx5_ct_tc_remove_dbgfs(struct mlx5_tc_ct_priv *ct_priv)
 	debugfs_remove_recursive(ct_priv->debugfs.root);
 }
 
+static struct mlx5_flow_handle *
+tc_ct_add_miss_rule(struct mlx5_flow_table *ft,
+		    struct mlx5_flow_table *next_ft)
+{
+	struct mlx5_flow_destination dest = {};
+	struct mlx5_flow_act act = {};
+
+	act.flags  = FLOW_ACT_IGNORE_FLOW_LEVEL | FLOW_ACT_NO_APPEND;
+	act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
+	dest.type  = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
+	dest.ft = next_ft;
+
+	return mlx5_add_flow_rules(ft, NULL, &act, &dest, 1);
+}
+
+static int
+tc_ct_add_ct_table_miss_rule(struct mlx5_flow_table *from,
+			     struct mlx5_flow_table *to,
+			     struct mlx5_flow_group **miss_group,
+			     struct mlx5_flow_handle **miss_rule)
+{
+	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
+	struct mlx5_flow_group *group;
+	struct mlx5_flow_handle *rule;
+	unsigned int max_fte = from->max_fte;
+	u32 *flow_group_in;
+	int err = 0;
+
+	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
+	if (!flow_group_in)
+		return -ENOMEM;
+
+	/* create miss group */
+	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index,
+		 max_fte - 2);
+	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index,
+		 max_fte - 1);
+	group = mlx5_create_flow_group(from, flow_group_in);
+	if (IS_ERR(group)) {
+		err = PTR_ERR(group);
+		goto err_miss_grp;
+	}
+
+	/* add miss rule to next fdb */
+	rule = tc_ct_add_miss_rule(from, to);
+	if (IS_ERR(rule)) {
+		err = PTR_ERR(rule);
+		goto err_miss_rule;
+	}
+
+	*miss_group = group;
+	*miss_rule = rule;
+	kvfree(flow_group_in);
+	return 0;
+
+err_miss_rule:
+	mlx5_destroy_flow_group(group);
+err_miss_grp:
+	kvfree(flow_group_in);
+	return err;
+}
+
+static void
+tc_ct_del_ct_table_miss_rule(struct mlx5_flow_group *miss_group,
+			     struct mlx5_flow_handle *miss_rule)
+{
+	mlx5_del_flow_rules(miss_rule);
+	mlx5_destroy_flow_group(miss_group);
+}
+
 #define INIT_ERR_PREFIX "tc ct offload init failed"
 
 struct mlx5_tc_ct_priv *
@@ -2212,6 +2304,12 @@ mlx5_tc_ct_init(struct mlx5e_priv *priv, struct mlx5_fs_chains *chains,
 		goto err_ct_nat_tbl;
 	}
 
+	err = tc_ct_add_ct_table_miss_rule(ct_priv->ct_nat, ct_priv->ct,
+					   &ct_priv->ct_nat_miss_group,
+					   &ct_priv->ct_nat_miss_rule);
+	if (err)
+		goto err_ct_zone_ht;
+
 	ct_priv->post_act = post_act;
 	mutex_init(&ct_priv->control_lock);
 	if (rhashtable_init(&ct_priv->zone_ht, &zone_params))
@@ -2273,6 +2371,7 @@ mlx5_tc_ct_clean(struct mlx5_tc_ct_priv *ct_priv)
 	ct_priv->fs_ops->destroy(ct_priv->fs);
 	kfree(ct_priv->fs);
 
+	tc_ct_del_ct_table_miss_rule(ct_priv->ct_nat_miss_group, ct_priv->ct_nat_miss_rule);
 	mlx5_chains_destroy_global_table(chains, ct_priv->ct_nat);
 	mlx5_chains_destroy_global_table(chains, ct_priv->ct);
 	mapping_destroy(ct_priv->zone_mapping);
-- 
cgit v1.2.3-58-ga151


From f070d422bab90b4554992e9d935764ce165dc101 Mon Sep 17 00:00:00 2001
From: Moshe Shemesh <moshe@nvidia.com>
Date: Fri, 14 Jun 2024 00:00:33 +0300
Subject: net/mlx5: Replace strcpy with strscpy

Replace deprecated strcpy with strscpy.

Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240613210036.1125203-4-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 32cdacc34a0d..a47d6419160d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -3353,9 +3353,9 @@ static int mlx5_fs_mode_get(struct devlink *devlink, u32 id,
 	struct mlx5_core_dev *dev = devlink_priv(devlink);
 
 	if (dev->priv.steering->mode == MLX5_FLOW_STEERING_MODE_SMFS)
-		strcpy(ctx->val.vstr, "smfs");
+		strscpy(ctx->val.vstr, "smfs", sizeof(ctx->val.vstr));
 	else
-		strcpy(ctx->val.vstr, "dmfs");
+		strscpy(ctx->val.vstr, "dmfs", sizeof(ctx->val.vstr));
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From a9dbb4ac58c02cdc2e2c7aa505699721fc426947 Mon Sep 17 00:00:00 2001
From: Gal Pressman <gal@nvidia.com>
Date: Fri, 14 Jun 2024 00:00:34 +0300
Subject: net/mlx5e: Fix outdated comment in features check

The code no longer treats only UDP tunnels, adjust the outdated comment.

Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Cosmin Ratiu <cratiu@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240613210036.1125203-5-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 10492349ff26..db4819f11541 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4910,7 +4910,7 @@ static netdev_features_t mlx5e_tunnel_features_check(struct mlx5e_priv *priv,
 	}
 
 out:
-	/* Disable CSUM and GSO if the udp dport is not offloaded by HW */
+	/* Disable CSUM and GSO if skb cannot be offloaded by HW */
 	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
-- 
cgit v1.2.3-58-ga151


From fac15a72b8e5c2baea5346fe56403f3dcf12755e Mon Sep 17 00:00:00 2001
From: Gal Pressman <gal@nvidia.com>
Date: Fri, 14 Jun 2024 00:00:35 +0300
Subject: net/mlx5e: Use tcp_v[46]_check checksum helpers

Use the tcp specific helpers to calculate the tcp pseudo header checksum
instead of the csum_*_magic ones.

Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240613210036.1125203-6-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index 43f018567faf..225da8d691fc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -1182,9 +1182,8 @@ static void mlx5e_lro_update_hdr(struct sk_buff *skb, struct mlx5_cqe64 *cqe,
 		check = csum_partial(tcp, tcp->doff * 4,
 				     csum_unfold((__force __sum16)cqe->check_sum));
 		/* Almost done, don't forget the pseudo header */
-		tcp->check = csum_tcpudp_magic(ipv4->saddr, ipv4->daddr,
-					       tot_len - sizeof(struct iphdr),
-					       IPPROTO_TCP, check);
+		tcp->check = tcp_v4_check(tot_len - sizeof(struct iphdr),
+					  ipv4->saddr, ipv4->daddr, check);
 	} else {
 		u16 payload_len = tot_len - sizeof(struct ipv6hdr);
 		struct ipv6hdr *ipv6 = ip_p;
@@ -1199,8 +1198,8 @@ static void mlx5e_lro_update_hdr(struct sk_buff *skb, struct mlx5_cqe64 *cqe,
 		check = csum_partial(tcp, tcp->doff * 4,
 				     csum_unfold((__force __sum16)cqe->check_sum));
 		/* Almost done, don't forget the pseudo header */
-		tcp->check = csum_ipv6_magic(&ipv6->saddr, &ipv6->daddr, payload_len,
-					     IPPROTO_TCP, check);
+		tcp->check = tcp_v6_check(payload_len, &ipv6->saddr,
+					  &ipv6->daddr, check);
 	}
 }
 
-- 
cgit v1.2.3-58-ga151


From 296eaab825060d0f25e89b2f85ab9fc26128cea8 Mon Sep 17 00:00:00 2001
From: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Date: Fri, 14 Jun 2024 00:00:36 +0300
Subject: net/mlx5e: Support SWP-mode offload L4 csum calculation

Calculate the pseudo-header checksum for both IPSec transport mode and
IPSec tunnel mode for mlx5 devices that do not implement a pure hardware
checksum offload for L4 checksum calculation. Introduce a capability bit
that identifies such mlx5 devices.

Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Cosmin Ratiu <cratiu@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/20240613210036.1125203-7-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h  | 37 ++++++++++++++++++++++
 .../mellanox/mlx5/core/en_accel/ipsec_rxtx.h       |  6 +++-
 include/linux/mlx5/mlx5_ifc.h                      |  3 +-
 3 files changed, 44 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index d1f0f868d494..5ec468268d1a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -6,6 +6,8 @@
 
 #include "en.h"
 #include <linux/indirect_call_wrapper.h>
+#include <net/ip6_checksum.h>
+#include <net/tcp.h>
 
 #define MLX5E_TX_WQE_EMPTY_DS_COUNT (sizeof(struct mlx5e_tx_wqe) / MLX5_SEND_WQE_DS)
 
@@ -479,6 +481,41 @@ mlx5e_set_eseg_swp(struct sk_buff *skb, struct mlx5_wqe_eth_seg *eseg,
 	}
 }
 
+static inline void
+mlx5e_swp_encap_csum_partial(struct mlx5_core_dev *mdev, struct sk_buff *skb, bool tunnel)
+{
+	const struct iphdr *ip = tunnel ? inner_ip_hdr(skb) : ip_hdr(skb);
+	const struct ipv6hdr *ip6;
+	struct tcphdr *th;
+	struct udphdr *uh;
+	int len;
+
+	if (!MLX5_CAP_ETH(mdev, swp_csum_l4_partial) || !skb_is_gso(skb))
+		return;
+
+	if (skb_is_gso_tcp(skb)) {
+		th = inner_tcp_hdr(skb);
+		len = skb_shinfo(skb)->gso_size + inner_tcp_hdrlen(skb);
+
+		if (ip->version == 4) {
+			th->check = ~tcp_v4_check(len, ip->saddr, ip->daddr, 0);
+		} else {
+			ip6 = tunnel ? inner_ipv6_hdr(skb) : ipv6_hdr(skb);
+			th->check = ~tcp_v6_check(len, &ip6->saddr, &ip6->daddr, 0);
+		}
+	} else if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) {
+		uh = (struct udphdr *)skb_inner_transport_header(skb);
+		len = skb_shinfo(skb)->gso_size + sizeof(struct udphdr);
+
+		if (ip->version == 4) {
+			uh->check = ~udp_v4_check(len, ip->saddr, ip->daddr, 0);
+		} else {
+			ip6 = tunnel ? inner_ipv6_hdr(skb) : ipv6_hdr(skb);
+			uh->check = ~udp_v6_check(len, &ip6->saddr, &ip6->daddr, 0);
+		}
+	}
+}
+
 #define MLX5E_STOP_ROOM(wqebbs) ((wqebbs) * 2 - 1)
 
 static inline u16 mlx5e_stop_room_for_wqe(struct mlx5_core_dev *mdev, u16 wqe_size)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
index 359050f0b54d..3cc640669247 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
@@ -116,6 +116,7 @@ static inline bool
 mlx5e_ipsec_txwqe_build_eseg_csum(struct mlx5e_txqsq *sq, struct sk_buff *skb,
 				  struct mlx5_wqe_eth_seg *eseg)
 {
+	struct mlx5_core_dev *mdev = sq->mdev;
 	u8 inner_ipproto;
 
 	if (!mlx5e_ipsec_eseg_meta(eseg))
@@ -125,9 +126,12 @@ mlx5e_ipsec_txwqe_build_eseg_csum(struct mlx5e_txqsq *sq, struct sk_buff *skb,
 	inner_ipproto = xfrm_offload(skb)->inner_ipproto;
 	if (inner_ipproto) {
 		eseg->cs_flags |= MLX5_ETH_WQE_L3_INNER_CSUM;
-		if (inner_ipproto == IPPROTO_TCP || inner_ipproto == IPPROTO_UDP)
+		if (inner_ipproto == IPPROTO_TCP || inner_ipproto == IPPROTO_UDP) {
+			mlx5e_swp_encap_csum_partial(mdev, skb, true);
 			eseg->cs_flags |= MLX5_ETH_WQE_L4_INNER_CSUM;
+		}
 	} else if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) {
+		mlx5e_swp_encap_csum_partial(mdev, skb, false);
 		eseg->cs_flags |= MLX5_ETH_WQE_L4_CSUM;
 		sq->stats->csum_partial_inner++;
 	}
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 466dcda40bb5..66b921c81c0f 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -1093,7 +1093,8 @@ struct mlx5_ifc_per_protocol_networking_offload_caps_bits {
 	u8         tunnel_stateless_ip_over_ip_tx[0x1];
 	u8         reserved_at_2e[0x2];
 	u8         max_vxlan_udp_ports[0x8];
-	u8         reserved_at_38[0x6];
+	u8         swp_csum_l4_partial[0x1];
+	u8         reserved_at_39[0x5];
 	u8         max_geneve_opt_len[0x1];
 	u8         tunnel_stateless_geneve_rx[0x1];
 
-- 
cgit v1.2.3-58-ga151


From 6c3282a6b296385bee2c383442c39f507b0d51dd Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Thu, 13 Jun 2024 11:36:06 +0100
Subject: net: stmmac: add select_pcs() platform method

Allow platform drivers to provide their logic to select an appropriate
PCS.

Tested-by: Romain Gantois <romain.gantois@bootlin.com>
Reviewed-by: Romain Gantois <romain.gantois@bootlin.com>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://lore.kernel.org/r/E1sHhoM-00Fesu-8E@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 7 +++++++
 include/linux/stmmac.h                            | 4 +++-
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index bbedf2a8c60f..302aa4080de3 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -949,6 +949,13 @@ static struct phylink_pcs *stmmac_mac_select_pcs(struct phylink_config *config,
 						 phy_interface_t interface)
 {
 	struct stmmac_priv *priv = netdev_priv(to_net_dev(config->dev));
+	struct phylink_pcs *pcs;
+
+	if (priv->plat->select_pcs) {
+		pcs = priv->plat->select_pcs(priv, interface);
+		if (!IS_ERR(pcs))
+			return pcs;
+	}
 
 	if (priv->hw->xpcs)
 		return &priv->hw->xpcs->pcs;
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index 8f0f156d50d3..9c54f82901a1 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -13,7 +13,7 @@
 #define __STMMAC_PLATFORM_DATA
 
 #include <linux/platform_device.h>
-#include <linux/phy.h>
+#include <linux/phylink.h>
 
 #define MTL_MAX_RX_QUEUES	8
 #define MTL_MAX_TX_QUEUES	8
@@ -271,6 +271,8 @@ struct plat_stmmacenet_data {
 	void (*dump_debug_regs)(void *priv);
 	int (*pcs_init)(struct stmmac_priv *priv);
 	void (*pcs_exit)(struct stmmac_priv *priv);
+	struct phylink_pcs *(*select_pcs)(struct stmmac_priv *priv,
+					  phy_interface_t interface);
 	void *bsp_priv;
 	struct clk *stmmac_clk;
 	struct clk *pclk;
-- 
cgit v1.2.3-58-ga151


From 135553da844c427639785af605517375c7bbc0c4 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Thu, 13 Jun 2024 11:36:11 +0100
Subject: net: stmmac: dwmac-intel: provide a select_pcs() implementation

Move the code returning the XPCS into dwmac-intel, which is the only
user of XPCS. Fill in the select_pcs() implementation only when we are
going to setup the XPCS, thus when it should be present.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://lore.kernel.org/r/E1sHhoR-00FetB-CP@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 11 +++++++++++
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c |  3 ---
 2 files changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
index 56649edb18cd..094d34c4193c 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
@@ -443,6 +443,16 @@ static void common_default_data(struct plat_stmmacenet_data *plat)
 	plat->rx_queues_cfg[0].pkt_route = 0x0;
 }
 
+static struct phylink_pcs *intel_mgbe_select_pcs(struct stmmac_priv *priv,
+						 phy_interface_t interface)
+{
+	/* plat->mdio_bus_data->has_xpcs has been set true, so there
+	 * should always be an XPCS. The original code would always
+	 * return this if present.
+	 */
+	return &priv->hw->xpcs->pcs;
+}
+
 static int intel_mgbe_common_data(struct pci_dev *pdev,
 				  struct plat_stmmacenet_data *plat)
 {
@@ -587,6 +597,7 @@ static int intel_mgbe_common_data(struct pci_dev *pdev,
 	    plat->phy_interface == PHY_INTERFACE_MODE_1000BASEX) {
 		plat->mdio_bus_data->has_xpcs = true;
 		plat->mdio_bus_data->default_an_inband = true;
+		plat->select_pcs = intel_mgbe_select_pcs;
 	}
 
 	/* Ensure mdio bus scan skips intel serdes and pcs-xpcs */
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 302aa4080de3..e9e2a95c91a3 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -957,9 +957,6 @@ static struct phylink_pcs *stmmac_mac_select_pcs(struct phylink_config *config,
 			return pcs;
 	}
 
-	if (priv->hw->xpcs)
-		return &priv->hw->xpcs->pcs;
-
 	return priv->hw->phylink_pcs;
 }
 
-- 
cgit v1.2.3-58-ga151


From 804c9866e0783bbdd43f35e3bda03eea212a7321 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Thu, 13 Jun 2024 11:36:16 +0100
Subject: net: stmmac: dwmac-rzn1: provide select_pcs() implementation

Provide a .select_pcs() implementation which returns the phylink PCS
that was created in the .pcs_init() method.

Tested-by: Romain Gantois <romain.gantois@bootlin.com>
Reviewed-by: Romain Gantois <romain.gantois@bootlin.com>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://lore.kernel.org/r/E1sHhoW-00FetH-GD@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-rzn1.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rzn1.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rzn1.c
index 848cf3c01f4a..59a7bd560f96 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rzn1.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rzn1.c
@@ -39,6 +39,12 @@ static void rzn1_dwmac_pcs_exit(struct stmmac_priv *priv)
 		miic_destroy(priv->hw->phylink_pcs);
 }
 
+static struct phylink_pcs *rzn1_dwmac_select_pcs(struct stmmac_priv *priv,
+						 phy_interface_t interface)
+{
+	return priv->hw->phylink_pcs;
+}
+
 static int rzn1_dwmac_probe(struct platform_device *pdev)
 {
 	struct plat_stmmacenet_data *plat_dat;
@@ -57,6 +63,7 @@ static int rzn1_dwmac_probe(struct platform_device *pdev)
 	plat_dat->bsp_priv = plat_dat;
 	plat_dat->pcs_init = rzn1_dwmac_pcs_init;
 	plat_dat->pcs_exit = rzn1_dwmac_pcs_exit;
+	plat_dat->select_pcs = rzn1_dwmac_select_pcs;
 
 	ret = stmmac_dvr_probe(dev, plat_dat, &stmmac_res);
 	if (ret)
-- 
cgit v1.2.3-58-ga151


From 98a6d9f192d32148a8b3ae4c8734fffb10bf0fb0 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Thu, 13 Jun 2024 11:36:21 +0100
Subject: net: stmmac: dwmac-socfpga: provide select_pcs() implementation

Provide a .select_pcs() implementation which returns the phylink PCS
that was created in the .pcs_init() method.

Reviewed-by: Romain Gantois <romain.gantois@bootlin.com>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://lore.kernel.org/r/E1sHhob-00FetN-Vp@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
index b3d45f9dfb55..fdb4c773ec98 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
@@ -429,6 +429,12 @@ static void socfpga_dwmac_pcs_exit(struct stmmac_priv *priv)
 		lynx_pcs_destroy(priv->hw->phylink_pcs);
 }
 
+static struct phylink_pcs *socfpga_dwmac_select_pcs(struct stmmac_priv *priv,
+						    phy_interface_t interface)
+{
+	return priv->hw->phylink_pcs;
+}
+
 static int socfpga_dwmac_probe(struct platform_device *pdev)
 {
 	struct plat_stmmacenet_data *plat_dat;
@@ -478,6 +484,7 @@ static int socfpga_dwmac_probe(struct platform_device *pdev)
 	plat_dat->fix_mac_speed = socfpga_dwmac_fix_mac_speed;
 	plat_dat->pcs_init = socfpga_dwmac_pcs_init;
 	plat_dat->pcs_exit = socfpga_dwmac_pcs_exit;
+	plat_dat->select_pcs = socfpga_dwmac_select_pcs;
 
 	ret = stmmac_dvr_probe(&pdev->dev, plat_dat, &stmmac_res);
 	if (ret)
-- 
cgit v1.2.3-58-ga151


From 93f84152e4aed84a246767f13da06f4a907328ef Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Thu, 13 Jun 2024 11:36:27 +0100
Subject: net: stmmac: clean up stmmac_mac_select_pcs()

Since all platform providers of PCS now populate the select_pcs()
method, there is no need for the common code to look at
priv->hw->phylink_pcs, so remove it.

Reviewed-by: Romain Gantois <romain.gantois@bootlin.com>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://lore.kernel.org/r/E1sHhoh-00FetT-3S@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index e9e2a95c91a3..5ddbb0d44373 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -957,7 +957,7 @@ static struct phylink_pcs *stmmac_mac_select_pcs(struct phylink_config *config,
 			return pcs;
 	}
 
-	return priv->hw->phylink_pcs;
+	return NULL;
 }
 
 static void stmmac_mac_config(struct phylink_config *config, unsigned int mode,
-- 
cgit v1.2.3-58-ga151


From afc5625e20971e8294a35b2f284a91b0601f9fce Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Thu, 13 Jun 2024 21:21:42 +0300
Subject: atm: clean up a put_user() calls

Unlike copy_from_user(), put_user() and get_user() return -EFAULT on
error.  Use the error code directly instead of setting it.

Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/04a018e8-7433-4f67-8ddd-9357a0114f87@moroto.mountain
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/atm/ioctl.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/atm/ioctl.c b/net/atm/ioctl.c
index f81f8d56f5c0..0f7a39aeccc8 100644
--- a/net/atm/ioctl.c
+++ b/net/atm/ioctl.c
@@ -68,7 +68,7 @@ static int do_vcc_ioctl(struct socket *sock, unsigned int cmd,
 			goto done;
 		}
 		error = put_user(sk->sk_sndbuf - sk_wmem_alloc_get(sk),
-				 (int __user *)argp) ? -EFAULT : 0;
+				 (int __user *)argp);
 		goto done;
 	case SIOCINQ:
 	{
@@ -83,7 +83,7 @@ static int do_vcc_ioctl(struct socket *sock, unsigned int cmd,
 		skb = skb_peek(&sk->sk_receive_queue);
 		amount = skb ? skb->len : 0;
 		spin_unlock_irq(&sk->sk_receive_queue.lock);
-		error = put_user(amount, (int __user *)argp) ? -EFAULT : 0;
+		error = put_user(amount, (int __user *)argp);
 		goto done;
 	}
 	case ATM_SETSC:
-- 
cgit v1.2.3-58-ga151


From cae7fd47dfe15e54710e3e550340876382a6db19 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Thu, 13 Jun 2024 16:07:54 +0200
Subject: mlxsw: port: Edit maximum MTU value

Currently mlxsw driver supports up to 10000 bytes for maximum MTU, this
value is not accurate, we can support up to 10K bytes. Change the value to
the maximum supported MTU by firmware.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://lore.kernel.org/r/666f51681234aeef09d771833ccb6e94bd323c88.1718275854.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/port.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/port.h b/drivers/net/ethernet/mellanox/mlxsw/port.h
index ac4d4ea51597..aa309615eff3 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/port.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/port.h
@@ -6,7 +6,7 @@
 
 #include <linux/types.h>
 
-#define MLXSW_PORT_MAX_MTU		10000
+#define MLXSW_PORT_MAX_MTU		(10 * 1024)
 
 #define MLXSW_PORT_DEFAULT_VID		1
 
-- 
cgit v1.2.3-58-ga151


From d361536fc2dfd22e65a84f4f2a13b2b8f4f0739b Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Thu, 13 Jun 2024 16:07:55 +0200
Subject: mlxsw: Adjust MTU value to hardware check

Ethernet frame consists of - Ethernet header, payload, FCS. The MTU value
which is used by user is the size of the payload, which means that when
user sets MTU to X, the total frame size will be larger due to the addition
of the Ethernet header and FCS.

Spectrum ASICs take into account Ethernet header and FCS as part of packet
size for MTU check. Adjust MTU value when user sets MTU, to configure the
MTU size which is required by hardware. The Tx header length which was used
by the driver is not relevant for such calculation, take into account
Ethernet header (with VLAN extension) and FCS.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://lore.kernel.org/r/f3203c2477bb8ed18b1e79642fa3e3713e1e55bb.1718275854.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/port.h     | 1 +
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/port.h b/drivers/net/ethernet/mellanox/mlxsw/port.h
index aa309615eff3..0a73b1a4526e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/port.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/port.h
@@ -7,6 +7,7 @@
 #include <linux/types.h>
 
 #define MLXSW_PORT_MAX_MTU		(10 * 1024)
+#define MLXSW_PORT_ETH_FRAME_HDR	(ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN)
 
 #define MLXSW_PORT_DEFAULT_VID		1
 
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 030ed71f945d..879daa18ccca 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -425,7 +425,7 @@ static int mlxsw_sp_port_mtu_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu)
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
 	char pmtu_pl[MLXSW_REG_PMTU_LEN];
 
-	mtu += MLXSW_TXHDR_LEN + ETH_HLEN;
+	mtu += MLXSW_PORT_ETH_FRAME_HDR;
 	if (mtu > mlxsw_sp_port->max_mtu)
 		return -EINVAL;
 
-- 
cgit v1.2.3-58-ga151


From 753aacfc032d3c20083cf7944d393aca08606feb Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Thu, 13 Jun 2024 16:07:56 +0200
Subject: mlxsw: spectrum: Set more accurate values for netdevice min/max MTU

Currently, the driver uses ETH_MAX_MTU as maximum MTU of netdevices,
instead, use the accurate value which is supported by the driver.
Subtract Ethernet headers which are taken into account by hardware for
MTU checking, as described in the previous patch.

Set minimum MTU to ETH_MIN_MTU, as zero MTU is not really supported.

With this change:
a. The stack will do the MTU checking, so we can remove it from the driver.
b. User space will be able to query the actual MTU limits.
   Before this patch:
   $ ip -j -d link show dev swp1 | jq | grep mtu
    "mtu": 1500,
    "min_mtu": 0,
    "max_mtu": 65535,

   With this patch:
   $ ip -j -d link show dev swp1 | jq | grep mtu
    "mtu": 1500,
    "min_mtu": 68,
    "max_mtu": 10218,

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://lore.kernel.org/r/be8232e38c196ecb607f82c5e000ea427ce22abb.1718275854.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 879daa18ccca..c5856f4d6b8b 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -426,8 +426,6 @@ static int mlxsw_sp_port_mtu_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu)
 	char pmtu_pl[MLXSW_REG_PMTU_LEN];
 
 	mtu += MLXSW_PORT_ETH_FRAME_HDR;
-	if (mtu > mlxsw_sp_port->max_mtu)
-		return -EINVAL;
 
 	mlxsw_reg_pmtu_pack(pmtu_pl, mlxsw_sp_port->local_port, mtu);
 	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pmtu), pmtu_pl);
@@ -1697,8 +1695,8 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u16 local_port,
 			 NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_HW_TC;
 	dev->hw_features |= NETIF_F_HW_TC | NETIF_F_LOOPBACK;
 
-	dev->min_mtu = 0;
-	dev->max_mtu = ETH_MAX_MTU;
+	dev->min_mtu = ETH_MIN_MTU;
+	dev->max_mtu = MLXSW_PORT_MAX_MTU - MLXSW_PORT_ETH_FRAME_HDR;
 
 	/* Each packet needs to have a Tx header (metadata) on top all other
 	 * headers.
-- 
cgit v1.2.3-58-ga151


From 3e7856545d369a6dcb3f93b21a9672b69f918650 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Thu, 13 Jun 2024 16:07:57 +0200
Subject: mlxsw: Use the same maximum MTU value throughout the driver

Currently, the driver uses two different values for maximum MTU, one is
stored in mlxsw_port->dev->max_mtu and the second is stored in
mlxsw_port->max_mtu. The second one is set to value which is queried from
firmware. This value was never tested, and unfortunately is not really
supported. That means that with the existing code, user can set MTU to
X, which is not really supported by firmware and which is bigger than
buffer size which is allocated in pci.

To make the driver consistent, use only mlxsw_port->dev->max_mtu for
maximum MTU value, for buffers headroom add Ethernet frame headers, which
are not included in mlxsw_port->dev->max_mtu. Remove mlxsw_port->max_mtu.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://lore.kernel.org/r/89fa6f804386b918d337e736e14ac291bb947483.1718275854.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/spectrum.c     | 23 ----------------------
 drivers/net/ethernet/mellanox/mlxsw/spectrum.h     |  1 -
 .../net/ethernet/mellanox/mlxsw/spectrum_buffers.c |  8 ++++++--
 3 files changed, 6 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index c5856f4d6b8b..f064789f3240 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -405,21 +405,6 @@ static int mlxsw_sp_port_dev_addr_init(struct mlxsw_sp_port *mlxsw_sp_port)
 					  mlxsw_sp_port->dev->dev_addr);
 }
 
-static int mlxsw_sp_port_max_mtu_get(struct mlxsw_sp_port *mlxsw_sp_port, int *p_max_mtu)
-{
-	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
-	char pmtu_pl[MLXSW_REG_PMTU_LEN];
-	int err;
-
-	mlxsw_reg_pmtu_pack(pmtu_pl, mlxsw_sp_port->local_port, 0);
-	err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(pmtu), pmtu_pl);
-	if (err)
-		return err;
-
-	*p_max_mtu = mlxsw_reg_pmtu_max_mtu_get(pmtu_pl);
-	return 0;
-}
-
 static int mlxsw_sp_port_mtu_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu)
 {
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
@@ -1725,13 +1710,6 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u16 local_port,
 		goto err_max_speed_get;
 	}
 
-	err = mlxsw_sp_port_max_mtu_get(mlxsw_sp_port, &mlxsw_sp_port->max_mtu);
-	if (err) {
-		dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to get maximum MTU\n",
-			mlxsw_sp_port->local_port);
-		goto err_port_max_mtu_get;
-	}
-
 	err = mlxsw_sp_port_mtu_set(mlxsw_sp_port, ETH_DATA_LEN);
 	if (err) {
 		dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to set MTU\n",
@@ -1875,7 +1853,6 @@ err_port_ets_init:
 err_port_buffers_init:
 err_port_admin_status_set:
 err_port_mtu_set:
-err_port_max_mtu_get:
 err_max_speed_get:
 err_port_speed_by_width_set:
 err_port_system_port_mapping_set:
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 3beb5d0847ab..bb0586b45c8d 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -359,7 +359,6 @@ struct mlxsw_sp_port {
 		u16 egr_types;
 		struct mlxsw_sp_ptp_port_stats stats;
 	} ptp;
-	int max_mtu;
 	u32 max_speed;
 	struct mlxsw_sp_hdroom *hdroom;
 	u64 module_overheat_initial_val;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c
index c9f1c79f3f9d..1b9ed393fbd4 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c
@@ -399,11 +399,13 @@ void mlxsw_sp_hdroom_bufs_reset_sizes(struct mlxsw_sp_port *mlxsw_sp_port,
 				      struct mlxsw_sp_hdroom *hdroom)
 {
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	unsigned int max_mtu = mlxsw_sp_port->dev->max_mtu;
 	u16 reserve_cells;
 	int i;
 
+	max_mtu += MLXSW_PORT_ETH_FRAME_HDR;
 	/* Internal buffer. */
-	reserve_cells = mlxsw_sp_hdroom_int_buf_size_get(mlxsw_sp, mlxsw_sp_port->max_mtu,
+	reserve_cells = mlxsw_sp_hdroom_int_buf_size_get(mlxsw_sp, max_mtu,
 							 mlxsw_sp_port->max_speed);
 	reserve_cells = mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port, reserve_cells);
 	hdroom->int_buf.reserve_cells = reserve_cells;
@@ -613,7 +615,9 @@ static int mlxsw_sp_port_headroom_init(struct mlxsw_sp_port *mlxsw_sp_port)
 	mlxsw_sp_hdroom_bufs_reset_sizes(mlxsw_sp_port, &hdroom);
 
 	/* Buffer 9 is used for control traffic. */
-	size9 = mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port, mlxsw_sp_port->max_mtu);
+	size9 = mlxsw_sp_port_headroom_8x_adjust(mlxsw_sp_port,
+						 mlxsw_sp_port->dev->max_mtu +
+						 MLXSW_PORT_ETH_FRAME_HDR);
 	hdroom.bufs.buf[9].size_cells = mlxsw_sp_bytes_cells(mlxsw_sp, size9);
 
 	return __mlxsw_sp_hdroom_configure(mlxsw_sp_port, &hdroom, true);
-- 
cgit v1.2.3-58-ga151


From 4be3dcc9bf04d0957235b45ba18983f1f751a3a1 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Thu, 13 Jun 2024 16:07:58 +0200
Subject: selftests: forwarding: Add test for minimum and maximum MTU

Add cases to check minimum and maximum MTU which are exposed via
"ip -d link show". Test configuration and traffic. Use VLAN devices as
usually VLAN header (4 bytes) is not included in the MTU, and drivers
should configure hardware correctly to send maximum MTU payload size
in VLAN tagged packets.

$ ./min_max_mtu.sh
TEST: ping						[ OK ]
TEST: ping6						[ OK ]
TEST: Test maximum MTU configuration			[ OK ]
TEST: Test traffic, packet size is maximum MTU		[ OK ]
TEST: Test minimum MTU configuration			[ OK ]
TEST: Test traffic, packet size is minimum MTU		[ OK ]

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://lore.kernel.org/r/89de8be8989db7a97f3b39e3c9da695673e78d2e.1718275854.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/forwarding/Makefile    |   1 +
 .../selftests/net/forwarding/min_max_mtu.sh        | 283 +++++++++++++++++++++
 2 files changed, 284 insertions(+)
 create mode 100755 tools/testing/selftests/net/forwarding/min_max_mtu.sh

diff --git a/tools/testing/selftests/net/forwarding/Makefile b/tools/testing/selftests/net/forwarding/Makefile
index 99576d7ecbf6..224346426ef2 100644
--- a/tools/testing/selftests/net/forwarding/Makefile
+++ b/tools/testing/selftests/net/forwarding/Makefile
@@ -39,6 +39,7 @@ TEST_PROGS = bridge_fdb_learning_limit.sh \
 	ipip_hier_gre.sh \
 	lib_sh_test.sh \
 	local_termination.sh \
+	min_max_mtu.sh \
 	mirror_gre_bound.sh \
 	mirror_gre_bridge_1d.sh \
 	mirror_gre_bridge_1d_vlan.sh \
diff --git a/tools/testing/selftests/net/forwarding/min_max_mtu.sh b/tools/testing/selftests/net/forwarding/min_max_mtu.sh
new file mode 100755
index 000000000000..97bb8b221bed
--- /dev/null
+++ b/tools/testing/selftests/net/forwarding/min_max_mtu.sh
@@ -0,0 +1,283 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+# +--------------------+
+# | H1                 |
+# |                    |
+# |           $h1.10 + |
+# |     192.0.2.2/24 | |
+# | 2001:db8:1::2/64 | |
+# |                  | |
+# |              $h1 + |
+# |                  | |
+# +------------------|-+
+#                    |
+# +------------------|-+
+# | SW               | |
+# |            $swp1 + |
+# |                  | |
+# |         $swp1.10 + |
+# |     192.0.2.1/24   |
+# | 2001:db8:1::1/64   |
+# |                    |
+# +--------------------+
+
+ALL_TESTS="
+	ping_ipv4
+	ping_ipv6
+	max_mtu_config_test
+	max_mtu_traffic_test
+	min_mtu_config_test
+	min_mtu_traffic_test
+"
+
+NUM_NETIFS=2
+source lib.sh
+
+h1_create()
+{
+	simple_if_init $h1
+	vlan_create $h1 10 v$h1 192.0.2.2/24 2001:db8:1::2/64
+}
+
+h1_destroy()
+{
+	vlan_destroy $h1 10 192.0.2.2/24 2001:db8:1::2/64
+	simple_if_fini $h1
+}
+
+switch_create()
+{
+	ip li set dev $swp1 up
+	vlan_create $swp1 10 "" 192.0.2.1/24 2001:db8:1::1/64
+}
+
+switch_destroy()
+{
+	ip li set dev $swp1 down
+	vlan_destroy $swp1 10
+}
+
+setup_prepare()
+{
+	h1=${NETIFS[p1]}
+	swp1=${NETIFS[p2]}
+
+	vrf_prepare
+
+	h1_create
+
+	switch_create
+
+	forwarding_enable
+}
+
+cleanup()
+{
+	pre_cleanup
+
+	forwarding_restore
+
+	switch_destroy
+
+	h1_destroy
+
+	vrf_cleanup
+}
+
+ping_ipv4()
+{
+	ping_test $h1.10 192.0.2.1
+}
+
+ping_ipv6()
+{
+	ping6_test $h1.10 2001:db8:1::1
+}
+
+min_max_mtu_get_if()
+{
+	local dev=$1; shift
+	local min_max=$1; shift
+
+	ip -d -j link show $dev | jq ".[].$min_max"
+}
+
+ensure_compatible_min_max_mtu()
+{
+	local min_max=$1; shift
+
+	local mtu=$(min_max_mtu_get_if ${NETIFS[p1]} $min_max)
+	local i
+
+	for ((i = 2; i <= NUM_NETIFS; ++i)); do
+		local current_mtu=$(min_max_mtu_get_if ${NETIFS[p$i]} $min_max)
+
+		if [ $current_mtu -ne $mtu ]; then
+			return 1
+		fi
+	done
+}
+
+mtu_set_if()
+{
+	local dev=$1; shift
+	local mtu=$1; shift
+	local should_fail=${1:-0}; shift
+
+	mtu_set $dev $mtu 2>/dev/null
+	check_err_fail $should_fail $? "Set MTU $mtu for $dev"
+}
+
+mtu_set_all_if()
+{
+	local mtu=$1; shift
+	local i
+
+	for ((i = 1; i <= NUM_NETIFS; ++i)); do
+		mtu_set_if ${NETIFS[p$i]} $mtu
+		mtu_set_if ${NETIFS[p$i]}.10 $mtu
+	done
+}
+
+mtu_restore_all_if()
+{
+	local i
+
+	for ((i = 1; i <= NUM_NETIFS; ++i)); do
+		mtu_restore ${NETIFS[p$i]}.10
+		mtu_restore ${NETIFS[p$i]}
+	done
+}
+
+mtu_test_ping4()
+{
+	local mtu=$1; shift
+	local should_fail=$1; shift
+
+	# Ping adds 8 bytes for ICMP header and 20 bytes for IP header
+	local ping_headers_len=$((20 + 8))
+	local pkt_size=$((mtu - ping_headers_len))
+
+	ping_do $h1.10 192.0.2.1 "-s $pkt_size -M do"
+	check_err_fail $should_fail $? "Ping, packet size: $pkt_size"
+}
+
+mtu_test_ping6()
+{
+	local mtu=$1; shift
+	local should_fail=$1; shift
+
+	# Ping adds 8 bytes for ICMP header and 40 bytes for IPv6 header
+	local ping6_headers_len=$((40 + 8))
+	local pkt_size=$((mtu - ping6_headers_len))
+
+	ping6_do $h1.10 2001:db8:1::1 "-s $pkt_size -M do"
+	check_err_fail $should_fail $? "Ping6, packet size: $pkt_size"
+}
+
+max_mtu_config_test()
+{
+	local i
+
+	RET=0
+
+	for ((i = 1; i <= NUM_NETIFS; ++i)); do
+		local dev=${NETIFS[p$i]}
+		local max_mtu=$(min_max_mtu_get_if $dev "max_mtu")
+		local should_fail
+
+		should_fail=0
+		mtu_set_if $dev $max_mtu $should_fail
+		mtu_restore $dev
+
+		should_fail=1
+		mtu_set_if $dev $((max_mtu + 1)) $should_fail
+		mtu_restore $dev
+	done
+
+	log_test "Test maximum MTU configuration"
+}
+
+max_mtu_traffic_test()
+{
+	local should_fail
+	local max_mtu
+
+	RET=0
+
+	if ! ensure_compatible_min_max_mtu "max_mtu"; then
+		log_test_xfail "Topology has incompatible maximum MTU values"
+		return
+	fi
+
+	max_mtu=$(min_max_mtu_get_if ${NETIFS[p1]} "max_mtu")
+
+	should_fail=0
+	mtu_set_all_if $max_mtu
+	mtu_test_ping4 $max_mtu $should_fail
+	mtu_test_ping6 $max_mtu $should_fail
+	mtu_restore_all_if
+
+	should_fail=1
+	mtu_set_all_if $((max_mtu - 1))
+	mtu_test_ping4 $max_mtu $should_fail
+	mtu_test_ping6 $max_mtu $should_fail
+	mtu_restore_all_if
+
+	log_test "Test traffic, packet size is maximum MTU"
+}
+
+min_mtu_config_test()
+{
+	local i
+
+	RET=0
+
+	for ((i = 1; i <= NUM_NETIFS; ++i)); do
+		local dev=${NETIFS[p$i]}
+		local min_mtu=$(min_max_mtu_get_if $dev "min_mtu")
+		local should_fail
+
+		should_fail=0
+		mtu_set_if $dev $min_mtu $should_fail
+		mtu_restore $dev
+
+		should_fail=1
+		mtu_set_if $dev $((min_mtu - 1)) $should_fail
+		mtu_restore $dev
+	done
+
+	log_test "Test minimum MTU configuration"
+}
+
+min_mtu_traffic_test()
+{
+	local should_fail=0
+	local min_mtu
+
+	RET=0
+
+	if ! ensure_compatible_min_max_mtu "min_mtu"; then
+		log_test_xfail "Topology has incompatible minimum MTU values"
+		return
+	fi
+
+	min_mtu=$(min_max_mtu_get_if ${NETIFS[p1]} "min_mtu")
+	mtu_set_all_if $min_mtu
+	mtu_test_ping4 $min_mtu $should_fail
+	# Do not test minimum MTU with IPv6, as IPv6 requires higher MTU.
+
+	mtu_restore_all_if
+
+	log_test "Test traffic, packet size is minimum MTU"
+}
+
+trap cleanup EXIT
+
+setup_prepare
+setup_wait
+
+tests_run
+
+exit $EXIT_STATUS
-- 
cgit v1.2.3-58-ga151


From 934c29999b57b835d65442da6f741d5e27f3b584 Mon Sep 17 00:00:00 2001
From: Sagi Grimberg <sagi@grimberg.me>
Date: Thu, 13 Jun 2024 14:35:04 +0300
Subject: net: micro-optimize skb_datagram_iter

We only use the mapping in a single context in a short and contained scope,
so kmap_local_page is sufficient and cheaper. This will also allow
skb_datagram_iter to be called from softirq context.

Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Link: https://lore.kernel.org/r/20240613113504.1079860-1-sagi@grimberg.me
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/datagram.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/core/datagram.c b/net/core/datagram.c
index e614cfd8e14a..95f242591fd2 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -417,14 +417,14 @@ static int __skb_datagram_iter(const struct sk_buff *skb, int offset,
 		end = start + skb_frag_size(frag);
 		if ((copy = end - offset) > 0) {
 			struct page *page = skb_frag_page(frag);
-			u8 *vaddr = kmap(page);
+			u8 *vaddr = kmap_local_page(page);
 
 			if (copy > len)
 				copy = len;
 			n = INDIRECT_CALL_1(cb, simple_copy_to_iter,
 					vaddr + skb_frag_off(frag) + offset - start,
 					copy, data, to);
-			kunmap(page);
+			kunmap_local(vaddr);
 			offset += n;
 			if (n != copy)
 				goto short_copy;
-- 
cgit v1.2.3-58-ga151


From d98995b4bf981519dde4af0a081c393d62474039 Mon Sep 17 00:00:00 2001
From: Jianbo Liu <jianbol@nvidia.com>
Date: Mon, 3 Jun 2024 13:26:37 +0300
Subject: net/mlx5: Reimplement write combining test

The test of write combining was added before in mlx5_ib driver. It
opens UD QP and posts NOP WQEs, and uses BlueFlame doorbell. When
BlueFlame is used, WQEs get written directly to a PCI BAR of the
device (in addition to memory) so that the device handles them without
having to access memory.

In this test, the WQEs written in memory are different from the ones
written to the BlueFlame which request CQE update. By checking the
completion reports posted on CQ, we can know if BlueFlame succeeds or
not. The write combining must be supported if BlueFlame succeeds as
its register is written using write combining.

This patch reimplements the test in the same way, but using a pair of
SQ and CQ only. It is moved to mlx5_core as a general feature used by
both mlx5_core and mlx5_ib.

Besides, save write combine test result of the PCI function, so that
its thousands of child functions such as SF can query without paying
the time and resource penalty by itself. The test function is called
only after failing to get the cached result. With this enhancement,
all thousands of SFs of the PF attached to same driver no longer need
to perform WC check explicitly, which is already done in the system.
This saves several commands per SF, thereby speeds up SF creation and
also saves completion EQ creation.

Signed-off-by: Jianbo Liu <jianbol@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://lore.kernel.org/r/4ff5a8cc4c5b5b0d98397baa45a5019bcdbf096e.1717409369.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 drivers/infiniband/hw/mlx5/main.c                |  19 +-
 drivers/infiniband/hw/mlx5/mem.c                 | 198 -----------
 drivers/infiniband/hw/mlx5/mlx5_ib.h             |   3 -
 drivers/infiniband/hw/mlx5/qp.c                  |  16 -
 drivers/net/ethernet/mellanox/mlx5/core/Makefile |   2 +-
 drivers/net/ethernet/mellanox/mlx5/core/main.c   |   2 +
 drivers/net/ethernet/mellanox/mlx5/core/wc.c     | 434 +++++++++++++++++++++++
 include/linux/mlx5/driver.h                      |  11 +
 8 files changed, 451 insertions(+), 234 deletions(-)
 create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/wc.c

diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 2366c46eebc8..00e292422801 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -1810,7 +1810,7 @@ static int set_ucontext_resp(struct ib_ucontext *uctx,
 	}
 
 	resp->qp_tab_size = 1 << MLX5_CAP_GEN(dev->mdev, log_max_qp);
-	if (dev->wc_support)
+	if (mlx5_wc_support_get(dev->mdev))
 		resp->bf_reg_size = 1 << MLX5_CAP_GEN(dev->mdev,
 						      log_bf_reg_size);
 	resp->cache_line_size = cache_line_size();
@@ -2337,7 +2337,7 @@ static int mlx5_ib_mmap(struct ib_ucontext *ibcontext, struct vm_area_struct *vm
 	switch (command) {
 	case MLX5_IB_MMAP_WC_PAGE:
 	case MLX5_IB_MMAP_ALLOC_WC:
-		if (!dev->wc_support)
+		if (!mlx5_wc_support_get(dev->mdev))
 			return -EPERM;
 		fallthrough;
 	case MLX5_IB_MMAP_NC_PAGE:
@@ -3612,7 +3612,7 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_UAR_OBJ_ALLOC)(
 	    alloc_type != MLX5_IB_UAPI_UAR_ALLOC_TYPE_NC)
 		return -EOPNOTSUPP;
 
-	if (!to_mdev(c->ibucontext.device)->wc_support &&
+	if (!mlx5_wc_support_get(to_mdev(c->ibucontext.device)->mdev) &&
 	    alloc_type == MLX5_IB_UAPI_UAR_ALLOC_TYPE_BF)
 		return -EOPNOTSUPP;
 
@@ -3766,18 +3766,6 @@ err_mp:
 	return err;
 }
 
-static int mlx5_ib_enable_driver(struct ib_device *dev)
-{
-	struct mlx5_ib_dev *mdev = to_mdev(dev);
-	int ret;
-
-	ret = mlx5_ib_test_wc(mdev);
-	mlx5_ib_dbg(mdev, "Write-Combining %s",
-		    mdev->wc_support ? "supported" : "not supported");
-
-	return ret;
-}
-
 static const struct ib_device_ops mlx5_ib_dev_ops = {
 	.owner = THIS_MODULE,
 	.driver_id = RDMA_DRIVER_MLX5,
@@ -3808,7 +3796,6 @@ static const struct ib_device_ops mlx5_ib_dev_ops = {
 	.drain_rq = mlx5_ib_drain_rq,
 	.drain_sq = mlx5_ib_drain_sq,
 	.device_group = &mlx5_attr_group,
-	.enable_driver = mlx5_ib_enable_driver,
 	.get_dev_fw_str = get_dev_fw_str,
 	.get_dma_mr = mlx5_ib_get_dma_mr,
 	.get_link_layer = mlx5_ib_port_link_layer,
diff --git a/drivers/infiniband/hw/mlx5/mem.c b/drivers/infiniband/hw/mlx5/mem.c
index 5a22be14d958..af321f6ef7f5 100644
--- a/drivers/infiniband/hw/mlx5/mem.c
+++ b/drivers/infiniband/hw/mlx5/mem.c
@@ -30,10 +30,8 @@
  * SOFTWARE.
  */
 
-#include <linux/io.h>
 #include <rdma/ib_umem_odp.h>
 #include "mlx5_ib.h"
-#include <linux/jiffies.h>
 
 /*
  * Fill in a physical address list. ib_umem_num_dma_blocks() entries will be
@@ -95,199 +93,3 @@ unsigned long __mlx5_umem_find_best_quantized_pgoff(
 		return 0;
 	return page_size;
 }
-
-#define WR_ID_BF 0xBF
-#define WR_ID_END 0xBAD
-#define TEST_WC_NUM_WQES 255
-#define TEST_WC_POLLING_MAX_TIME_JIFFIES msecs_to_jiffies(100)
-static int post_send_nop(struct mlx5_ib_dev *dev, struct ib_qp *ibqp, u64 wr_id,
-			 bool signaled)
-{
-	struct mlx5_ib_qp *qp = to_mqp(ibqp);
-	struct mlx5_wqe_ctrl_seg *ctrl;
-	struct mlx5_bf *bf = &qp->bf;
-	__be32 mmio_wqe[16] = {};
-	unsigned long flags;
-	unsigned int idx;
-
-	if (unlikely(dev->mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR))
-		return -EIO;
-
-	spin_lock_irqsave(&qp->sq.lock, flags);
-
-	idx = qp->sq.cur_post & (qp->sq.wqe_cnt - 1);
-	ctrl = mlx5_frag_buf_get_wqe(&qp->sq.fbc, idx);
-
-	memset(ctrl, 0, sizeof(struct mlx5_wqe_ctrl_seg));
-	ctrl->fm_ce_se = signaled ? MLX5_WQE_CTRL_CQ_UPDATE : 0;
-	ctrl->opmod_idx_opcode =
-		cpu_to_be32(((u32)(qp->sq.cur_post) << 8) | MLX5_OPCODE_NOP);
-	ctrl->qpn_ds = cpu_to_be32((sizeof(struct mlx5_wqe_ctrl_seg) / 16) |
-				   (qp->trans_qp.base.mqp.qpn << 8));
-
-	qp->sq.wrid[idx] = wr_id;
-	qp->sq.w_list[idx].opcode = MLX5_OPCODE_NOP;
-	qp->sq.wqe_head[idx] = qp->sq.head + 1;
-	qp->sq.cur_post += DIV_ROUND_UP(sizeof(struct mlx5_wqe_ctrl_seg),
-					MLX5_SEND_WQE_BB);
-	qp->sq.w_list[idx].next = qp->sq.cur_post;
-	qp->sq.head++;
-
-	memcpy(mmio_wqe, ctrl, sizeof(*ctrl));
-	((struct mlx5_wqe_ctrl_seg *)&mmio_wqe)->fm_ce_se |=
-		MLX5_WQE_CTRL_CQ_UPDATE;
-
-	/* Make sure that descriptors are written before
-	 * updating doorbell record and ringing the doorbell
-	 */
-	wmb();
-
-	qp->db.db[MLX5_SND_DBR] = cpu_to_be32(qp->sq.cur_post);
-
-	/* Make sure doorbell record is visible to the HCA before
-	 * we hit doorbell
-	 */
-	wmb();
-	__iowrite64_copy(bf->bfreg->map + bf->offset, mmio_wqe,
-			 sizeof(mmio_wqe) / 8);
-
-	bf->offset ^= bf->buf_size;
-
-	spin_unlock_irqrestore(&qp->sq.lock, flags);
-
-	return 0;
-}
-
-static int test_wc_poll_cq_result(struct mlx5_ib_dev *dev, struct ib_cq *cq)
-{
-	int ret;
-	struct ib_wc wc = {};
-	unsigned long end = jiffies + TEST_WC_POLLING_MAX_TIME_JIFFIES;
-
-	do {
-		ret = ib_poll_cq(cq, 1, &wc);
-		if (ret < 0 || wc.status)
-			return ret < 0 ? ret : -EINVAL;
-		if (ret)
-			break;
-	} while (!time_after(jiffies, end));
-
-	if (!ret)
-		return -ETIMEDOUT;
-
-	if (wc.wr_id != WR_ID_BF)
-		ret = 0;
-
-	return ret;
-}
-
-static int test_wc_do_send(struct mlx5_ib_dev *dev, struct ib_qp *qp)
-{
-	int err, i;
-
-	for (i = 0; i < TEST_WC_NUM_WQES; i++) {
-		err = post_send_nop(dev, qp, WR_ID_BF, false);
-		if (err)
-			return err;
-	}
-
-	return post_send_nop(dev, qp, WR_ID_END, true);
-}
-
-int mlx5_ib_test_wc(struct mlx5_ib_dev *dev)
-{
-	struct ib_cq_init_attr cq_attr = { .cqe = TEST_WC_NUM_WQES + 1 };
-	int port_type_cap = MLX5_CAP_GEN(dev->mdev, port_type);
-	struct ib_qp_init_attr qp_init_attr = {
-		.cap = { .max_send_wr = TEST_WC_NUM_WQES },
-		.qp_type = IB_QPT_UD,
-		.sq_sig_type = IB_SIGNAL_REQ_WR,
-		.create_flags = MLX5_IB_QP_CREATE_WC_TEST,
-	};
-	struct ib_qp_attr qp_attr = { .port_num = 1 };
-	struct ib_device *ibdev = &dev->ib_dev;
-	struct ib_qp *qp;
-	struct ib_cq *cq;
-	struct ib_pd *pd;
-	int ret;
-
-	if (!MLX5_CAP_GEN(dev->mdev, bf))
-		return 0;
-
-	if (!dev->mdev->roce.roce_en &&
-	    port_type_cap == MLX5_CAP_PORT_TYPE_ETH) {
-		if (mlx5_core_is_pf(dev->mdev))
-			dev->wc_support = arch_can_pci_mmap_wc();
-		return 0;
-	}
-
-	ret = mlx5_alloc_bfreg(dev->mdev, &dev->wc_bfreg, true, false);
-	if (ret)
-		goto print_err;
-
-	if (!dev->wc_bfreg.wc)
-		goto out1;
-
-	pd = ib_alloc_pd(ibdev, 0);
-	if (IS_ERR(pd)) {
-		ret = PTR_ERR(pd);
-		goto out1;
-	}
-
-	cq = ib_create_cq(ibdev, NULL, NULL, NULL, &cq_attr);
-	if (IS_ERR(cq)) {
-		ret = PTR_ERR(cq);
-		goto out2;
-	}
-
-	qp_init_attr.recv_cq = cq;
-	qp_init_attr.send_cq = cq;
-	qp = ib_create_qp(pd, &qp_init_attr);
-	if (IS_ERR(qp)) {
-		ret = PTR_ERR(qp);
-		goto out3;
-	}
-
-	qp_attr.qp_state = IB_QPS_INIT;
-	ret = ib_modify_qp(qp, &qp_attr,
-			   IB_QP_STATE | IB_QP_PORT | IB_QP_PKEY_INDEX |
-				   IB_QP_QKEY);
-	if (ret)
-		goto out4;
-
-	qp_attr.qp_state = IB_QPS_RTR;
-	ret = ib_modify_qp(qp, &qp_attr, IB_QP_STATE);
-	if (ret)
-		goto out4;
-
-	qp_attr.qp_state = IB_QPS_RTS;
-	ret = ib_modify_qp(qp, &qp_attr, IB_QP_STATE | IB_QP_SQ_PSN);
-	if (ret)
-		goto out4;
-
-	ret = test_wc_do_send(dev, qp);
-	if (ret < 0)
-		goto out4;
-
-	ret = test_wc_poll_cq_result(dev, cq);
-	if (ret > 0) {
-		dev->wc_support = true;
-		ret = 0;
-	}
-
-out4:
-	ib_destroy_qp(qp);
-out3:
-	ib_destroy_cq(cq);
-out2:
-	ib_dealloc_pd(pd);
-out1:
-	mlx5_free_bfreg(dev->mdev, &dev->wc_bfreg);
-print_err:
-	if (ret)
-		mlx5_ib_err(
-			dev,
-			"Error %d while trying to test write-combining support\n",
-			ret);
-	return ret;
-}
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index f255a12e26a0..b68779e9d86c 100644
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -341,7 +341,6 @@ struct mlx5_ib_flow_db {
  * rely on the range reserved for that use in the ib_qp_create_flags enum.
  */
 #define MLX5_IB_QP_CREATE_SQPN_QP1	IB_QP_CREATE_RESERVED_START
-#define MLX5_IB_QP_CREATE_WC_TEST	(IB_QP_CREATE_RESERVED_START << 1)
 
 struct wr_list {
 	u16	opcode;
@@ -1123,7 +1122,6 @@ struct mlx5_ib_dev {
 	u8				ib_active:1;
 	u8				is_rep:1;
 	u8				lag_active:1;
-	u8				wc_support:1;
 	u8				fill_delay;
 	struct umr_common		umrc;
 	/* sync used page count stats
@@ -1149,7 +1147,6 @@ struct mlx5_ib_dev {
 	/* Array with num_ports elements */
 	struct mlx5_ib_port	*port;
 	struct mlx5_sq_bfreg	bfreg;
-	struct mlx5_sq_bfreg	wc_bfreg;
 	struct mlx5_sq_bfreg	fp_bfreg;
 	struct mlx5_ib_delay_drop	delay_drop;
 	const struct mlx5_ib_profile	*profile;
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index e2164f813607..e8c0fead4062 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -1107,8 +1107,6 @@ static int _create_kernel_qp(struct mlx5_ib_dev *dev,
 
 	if (init_attr->qp_type == MLX5_IB_QPT_REG_UMR)
 		qp->bf.bfreg = &dev->fp_bfreg;
-	else if (qp->flags & MLX5_IB_QP_CREATE_WC_TEST)
-		qp->bf.bfreg = &dev->wc_bfreg;
 	else
 		qp->bf.bfreg = &dev->bfreg;
 
@@ -2959,14 +2957,6 @@ static void process_create_flag(struct mlx5_ib_dev *dev, int *flags, int flag,
 		return;
 	}
 
-	if (flag == MLX5_IB_QP_CREATE_WC_TEST) {
-		/*
-		 * Special case, if condition didn't meet, it won't be error,
-		 * just different in-kernel flow.
-		 */
-		*flags &= ~MLX5_IB_QP_CREATE_WC_TEST;
-		return;
-	}
 	mlx5_ib_dbg(dev, "Verbs create QP flag 0x%X is not supported\n", flag);
 }
 
@@ -3027,8 +3017,6 @@ static int process_create_flags(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
 			    IB_QP_CREATE_PCI_WRITE_END_PADDING,
 			    MLX5_CAP_GEN(mdev, end_pad), qp);
 
-	process_create_flag(dev, &create_flags, MLX5_IB_QP_CREATE_WC_TEST,
-			    qp_type != MLX5_IB_QPT_REG_UMR, qp);
 	process_create_flag(dev, &create_flags, MLX5_IB_QP_CREATE_SQPN_QP1,
 			    true, qp);
 
@@ -4609,10 +4597,6 @@ static bool mlx5_ib_modify_qp_allowed(struct mlx5_ib_dev *dev,
 	if (qp->type == IB_QPT_RAW_PACKET || qp->type == MLX5_IB_QPT_REG_UMR)
 		return true;
 
-	/* Internal QP used for wc testing, with NOPs in wq */
-	if (qp->flags & MLX5_IB_QP_CREATE_WC_TEST)
-		return true;
-
 	return false;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 76dc5a9b9648..1289475e7be7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -17,7 +17,7 @@ mlx5_core-y :=	main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
 		fs_counters.o fs_ft_pool.o rl.o lag/debugfs.o lag/lag.o dev.o events.o wq.o lib/gid.o \
 		lib/devcom.o lib/pci_vsc.o lib/dm.o lib/fs_ttc.o diag/fs_tracepoint.o \
 		diag/fw_tracer.o diag/crdump.o devlink.o diag/rsc_dump.o diag/reporter_vnic.o \
-		fw_reset.o qos.o lib/tout.o lib/aso.o
+		fw_reset.o qos.o lib/tout.o lib/aso.o wc.o
 
 #
 # Netdev basic
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 459a836a5d9c..527da58c7953 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -1819,6 +1819,7 @@ int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx)
 	mutex_init(&dev->intf_state_mutex);
 	lockdep_set_class(&dev->intf_state_mutex, &dev->lock_key);
 	mutex_init(&dev->mlx5e_res.uplink_netdev_lock);
+	mutex_init(&dev->wc_state_lock);
 
 	mutex_init(&priv->bfregs.reg_head.lock);
 	mutex_init(&priv->bfregs.wc_head.lock);
@@ -1916,6 +1917,7 @@ void mlx5_mdev_uninit(struct mlx5_core_dev *dev)
 	mutex_destroy(&priv->alloc_mutex);
 	mutex_destroy(&priv->bfregs.wc_head.lock);
 	mutex_destroy(&priv->bfregs.reg_head.lock);
+	mutex_destroy(&dev->wc_state_lock);
 	mutex_destroy(&dev->mlx5e_res.uplink_netdev_lock);
 	mutex_destroy(&dev->intf_state_mutex);
 	lockdep_unregister_key(&dev->lock_key);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wc.c b/drivers/net/ethernet/mellanox/mlx5/core/wc.c
new file mode 100644
index 000000000000..1bed75eca97d
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/wc.c
@@ -0,0 +1,434 @@
+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+// Copyright (c) 2024, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+
+#include <linux/io.h>
+#include <linux/mlx5/transobj.h>
+#include "lib/clock.h"
+#include "mlx5_core.h"
+#include "wq.h"
+
+#define TEST_WC_NUM_WQES 255
+#define TEST_WC_LOG_CQ_SZ (order_base_2(TEST_WC_NUM_WQES))
+#define TEST_WC_SQ_LOG_WQ_SZ TEST_WC_LOG_CQ_SZ
+#define TEST_WC_POLLING_MAX_TIME_JIFFIES msecs_to_jiffies(100)
+
+struct mlx5_wc_cq {
+	/* data path - accessed per cqe */
+	struct mlx5_cqwq wq;
+
+	/* data path - accessed per napi poll */
+	struct mlx5_core_cq mcq;
+
+	/* control */
+	struct mlx5_core_dev *mdev;
+	struct mlx5_wq_ctrl wq_ctrl;
+};
+
+struct mlx5_wc_sq {
+	/* data path */
+	u16 cc;
+	u16 pc;
+
+	/* read only */
+	struct mlx5_wq_cyc wq;
+	u32 sqn;
+
+	/* control path */
+	struct mlx5_wq_ctrl wq_ctrl;
+
+	struct mlx5_wc_cq cq;
+	struct mlx5_sq_bfreg bfreg;
+};
+
+static int mlx5_wc_create_cqwq(struct mlx5_core_dev *mdev, void *cqc,
+			       struct mlx5_wc_cq *cq)
+{
+	struct mlx5_core_cq *mcq = &cq->mcq;
+	struct mlx5_wq_param param = {};
+	int err;
+	u32 i;
+
+	err = mlx5_cqwq_create(mdev, &param, cqc, &cq->wq, &cq->wq_ctrl);
+	if (err)
+		return err;
+
+	mcq->cqe_sz     = 64;
+	mcq->set_ci_db  = cq->wq_ctrl.db.db;
+	mcq->arm_db     = cq->wq_ctrl.db.db + 1;
+
+	for (i = 0; i < mlx5_cqwq_get_size(&cq->wq); i++) {
+		struct mlx5_cqe64 *cqe = mlx5_cqwq_get_wqe(&cq->wq, i);
+
+		cqe->op_own = 0xf1;
+	}
+
+	cq->mdev = mdev;
+
+	return 0;
+}
+
+static int create_wc_cq(struct mlx5_wc_cq *cq, void *cqc_data)
+{
+	u32 out[MLX5_ST_SZ_DW(create_cq_out)];
+	struct mlx5_core_dev *mdev = cq->mdev;
+	struct mlx5_core_cq *mcq = &cq->mcq;
+	int err, inlen, eqn;
+	void *in, *cqc;
+
+	err = mlx5_comp_eqn_get(mdev, 0, &eqn);
+	if (err)
+		return err;
+
+	inlen = MLX5_ST_SZ_BYTES(create_cq_in) +
+		sizeof(u64) * cq->wq_ctrl.buf.npages;
+	in = kvzalloc(inlen, GFP_KERNEL);
+	if (!in)
+		return -ENOMEM;
+
+	cqc = MLX5_ADDR_OF(create_cq_in, in, cq_context);
+
+	memcpy(cqc, cqc_data, MLX5_ST_SZ_BYTES(cqc));
+
+	mlx5_fill_page_frag_array(&cq->wq_ctrl.buf,
+				  (__be64 *)MLX5_ADDR_OF(create_cq_in, in, pas));
+
+	MLX5_SET(cqc,   cqc, cq_period_mode, MLX5_CQ_PERIOD_MODE_START_FROM_EQE);
+	MLX5_SET(cqc,   cqc, c_eqn_or_apu_element, eqn);
+	MLX5_SET(cqc,   cqc, uar_page,      mdev->priv.uar->index);
+	MLX5_SET(cqc,   cqc, log_page_size, cq->wq_ctrl.buf.page_shift -
+					    MLX5_ADAPTER_PAGE_SHIFT);
+	MLX5_SET64(cqc, cqc, dbr_addr,      cq->wq_ctrl.db.dma);
+
+	err = mlx5_core_create_cq(mdev, mcq, in, inlen, out, sizeof(out));
+
+	kvfree(in);
+
+	return err;
+}
+
+static int mlx5_wc_create_cq(struct mlx5_core_dev *mdev, struct mlx5_wc_cq *cq)
+{
+	void *cqc;
+	int err;
+
+	cqc = kvzalloc(MLX5_ST_SZ_BYTES(cqc), GFP_KERNEL);
+	if (!cqc)
+		return -ENOMEM;
+
+	MLX5_SET(cqc, cqc, log_cq_size, TEST_WC_LOG_CQ_SZ);
+	MLX5_SET(cqc, cqc, uar_page, mdev->priv.uar->index);
+	if (MLX5_CAP_GEN(mdev, cqe_128_always) && cache_line_size() >= 128)
+		MLX5_SET(cqc, cqc, cqe_sz, CQE_STRIDE_128_PAD);
+
+	err = mlx5_wc_create_cqwq(mdev, cqc, cq);
+	if (err) {
+		mlx5_core_err(mdev, "Failed to create wc cq wq, err=%d\n", err);
+		goto err_create_cqwq;
+	}
+
+	err = create_wc_cq(cq, cqc);
+	if (err) {
+		mlx5_core_err(mdev, "Failed to create wc cq, err=%d\n", err);
+		goto err_create_cq;
+	}
+
+	kvfree(cqc);
+	return 0;
+
+err_create_cq:
+	mlx5_wq_destroy(&cq->wq_ctrl);
+err_create_cqwq:
+	kvfree(cqc);
+	return err;
+}
+
+static void mlx5_wc_destroy_cq(struct mlx5_wc_cq *cq)
+{
+	mlx5_core_destroy_cq(cq->mdev, &cq->mcq);
+	mlx5_wq_destroy(&cq->wq_ctrl);
+}
+
+static int create_wc_sq(struct mlx5_core_dev *mdev, void *sqc_data,
+			struct mlx5_wc_sq *sq)
+{
+	void *in, *sqc, *wq;
+	int inlen, err;
+	u8 ts_format;
+
+	inlen = MLX5_ST_SZ_BYTES(create_sq_in) +
+		sizeof(u64) * sq->wq_ctrl.buf.npages;
+	in = kvzalloc(inlen, GFP_KERNEL);
+	if (!in)
+		return -ENOMEM;
+
+	sqc = MLX5_ADDR_OF(create_sq_in, in, ctx);
+	wq = MLX5_ADDR_OF(sqc, sqc, wq);
+
+	memcpy(sqc, sqc_data, MLX5_ST_SZ_BYTES(sqc));
+	MLX5_SET(sqc,  sqc, cqn, sq->cq.mcq.cqn);
+
+	MLX5_SET(sqc,  sqc, state, MLX5_SQC_STATE_RST);
+	MLX5_SET(sqc,  sqc, flush_in_error_en, 1);
+
+	ts_format = mlx5_is_real_time_sq(mdev) ?
+			MLX5_TIMESTAMP_FORMAT_REAL_TIME :
+			MLX5_TIMESTAMP_FORMAT_FREE_RUNNING;
+	MLX5_SET(sqc, sqc, ts_format, ts_format);
+
+	MLX5_SET(wq,   wq, wq_type,       MLX5_WQ_TYPE_CYCLIC);
+	MLX5_SET(wq,   wq, uar_page,      sq->bfreg.index);
+	MLX5_SET(wq,   wq, log_wq_pg_sz,  sq->wq_ctrl.buf.page_shift -
+					  MLX5_ADAPTER_PAGE_SHIFT);
+	MLX5_SET64(wq, wq, dbr_addr,      sq->wq_ctrl.db.dma);
+
+	mlx5_fill_page_frag_array(&sq->wq_ctrl.buf,
+				  (__be64 *)MLX5_ADDR_OF(wq, wq, pas));
+
+	err = mlx5_core_create_sq(mdev, in, inlen, &sq->sqn);
+	if (err) {
+		mlx5_core_err(mdev, "Failed to create wc sq, err=%d\n", err);
+		goto err_create_sq;
+	}
+
+	memset(in, 0,  MLX5_ST_SZ_BYTES(modify_sq_in));
+	MLX5_SET(modify_sq_in, in, sq_state, MLX5_SQC_STATE_RST);
+	sqc = MLX5_ADDR_OF(modify_sq_in, in, ctx);
+	MLX5_SET(sqc, sqc, state, MLX5_SQC_STATE_RDY);
+
+	err = mlx5_core_modify_sq(mdev, sq->sqn, in);
+	if (err) {
+		mlx5_core_err(mdev, "Failed to set wc sq(sqn=0x%x) ready, err=%d\n",
+			      sq->sqn, err);
+		goto err_modify_sq;
+	}
+
+	kvfree(in);
+	return 0;
+
+err_modify_sq:
+	mlx5_core_destroy_sq(mdev, sq->sqn);
+err_create_sq:
+	kvfree(in);
+	return err;
+}
+
+static int mlx5_wc_create_sq(struct mlx5_core_dev *mdev, struct mlx5_wc_sq *sq)
+{
+	struct mlx5_wq_param param = {};
+	void *sqc_data, *wq;
+	int err;
+
+	sqc_data = kvzalloc(MLX5_ST_SZ_BYTES(sqc), GFP_KERNEL);
+	if (!sqc_data)
+		return -ENOMEM;
+
+	wq = MLX5_ADDR_OF(sqc, sqc_data, wq);
+	MLX5_SET(wq, wq, log_wq_stride, ilog2(MLX5_SEND_WQE_BB));
+	MLX5_SET(wq, wq, pd, mdev->mlx5e_res.hw_objs.pdn);
+	MLX5_SET(wq, wq, log_wq_sz, TEST_WC_SQ_LOG_WQ_SZ);
+
+	err = mlx5_wq_cyc_create(mdev, &param, wq, &sq->wq, &sq->wq_ctrl);
+	if (err) {
+		mlx5_core_err(mdev, "Failed to create wc sq wq, err=%d\n", err);
+		goto err_create_wq_cyc;
+	}
+
+	err = create_wc_sq(mdev, sqc_data, sq);
+	if (err)
+		goto err_create_sq;
+
+	mlx5_core_dbg(mdev, "wc sq->sqn = 0x%x created\n", sq->sqn);
+
+	kvfree(sqc_data);
+	return 0;
+
+err_create_sq:
+	mlx5_wq_destroy(&sq->wq_ctrl);
+err_create_wq_cyc:
+	kvfree(sqc_data);
+	return err;
+}
+
+static void mlx5_wc_destroy_sq(struct mlx5_wc_sq *sq)
+{
+	mlx5_core_destroy_sq(sq->cq.mdev, sq->sqn);
+	mlx5_wq_destroy(&sq->wq_ctrl);
+}
+
+static void mlx5_wc_post_nop(struct mlx5_wc_sq *sq, bool signaled)
+{
+	int buf_size = (1 << MLX5_CAP_GEN(sq->cq.mdev, log_bf_reg_size)) / 2;
+	struct mlx5_wqe_ctrl_seg *ctrl;
+	__be32 mmio_wqe[16] = {};
+	u16 pi;
+
+	pi = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->pc);
+	ctrl = mlx5_wq_cyc_get_wqe(&sq->wq, pi);
+	memset(ctrl, 0, sizeof(*ctrl));
+	ctrl->opmod_idx_opcode =
+		cpu_to_be32((sq->pc << MLX5_WQE_CTRL_WQE_INDEX_SHIFT) | MLX5_OPCODE_NOP);
+	ctrl->qpn_ds =
+		cpu_to_be32((sq->sqn << MLX5_WQE_CTRL_QPN_SHIFT) |
+			    DIV_ROUND_UP(sizeof(struct mlx5_wqe_ctrl_seg), MLX5_SEND_WQE_DS));
+	if (signaled)
+		ctrl->fm_ce_se |= MLX5_WQE_CTRL_CQ_UPDATE;
+
+	memcpy(mmio_wqe, ctrl, sizeof(*ctrl));
+	((struct mlx5_wqe_ctrl_seg *)&mmio_wqe)->fm_ce_se |=
+		MLX5_WQE_CTRL_CQ_UPDATE;
+
+	/* ensure wqe is visible to device before updating doorbell record */
+	dma_wmb();
+
+	sq->pc++;
+	sq->wq.db[MLX5_SND_DBR] = cpu_to_be32(sq->pc);
+
+	/* ensure doorbell record is visible to device before ringing the
+	 * doorbell
+	 */
+	wmb();
+
+	__iowrite64_copy(sq->bfreg.map + sq->bfreg.offset, mmio_wqe,
+			 sizeof(mmio_wqe) / 8);
+
+	sq->bfreg.offset ^= buf_size;
+}
+
+static int mlx5_wc_poll_cq(struct mlx5_wc_sq *sq)
+{
+	struct mlx5_wc_cq *cq = &sq->cq;
+	struct mlx5_cqe64 *cqe;
+
+	cqe = mlx5_cqwq_get_cqe(&cq->wq);
+	if (!cqe)
+		return -ETIMEDOUT;
+
+	/* sq->cc must be updated only after mlx5_cqwq_update_db_record(),
+	 * otherwise a cq overrun may occur
+	 */
+	mlx5_cqwq_pop(&cq->wq);
+
+	if (get_cqe_opcode(cqe) == MLX5_CQE_REQ) {
+		int wqe_counter = be16_to_cpu(cqe->wqe_counter);
+		struct mlx5_core_dev *mdev = cq->mdev;
+
+		if (wqe_counter == TEST_WC_NUM_WQES - 1)
+			mdev->wc_state = MLX5_WC_STATE_UNSUPPORTED;
+		else
+			mdev->wc_state = MLX5_WC_STATE_SUPPORTED;
+
+		mlx5_core_dbg(mdev, "wc wqe_counter = 0x%x\n", wqe_counter);
+	}
+
+	mlx5_cqwq_update_db_record(&cq->wq);
+
+	/* ensure cq space is freed before enabling more cqes */
+	wmb();
+
+	sq->cc++;
+
+	return 0;
+}
+
+static void mlx5_core_test_wc(struct mlx5_core_dev *mdev)
+{
+	unsigned long expires;
+	struct mlx5_wc_sq *sq;
+	int i, err;
+
+	if (mdev->wc_state != MLX5_WC_STATE_UNINITIALIZED)
+		return;
+
+	sq = kzalloc(sizeof(*sq), GFP_KERNEL);
+	if (!sq)
+		return;
+
+	err = mlx5_alloc_bfreg(mdev, &sq->bfreg, true, false);
+	if (err) {
+		mlx5_core_err(mdev, "Failed to alloc bfreg for wc, err=%d\n", err);
+		goto err_alloc_bfreg;
+	}
+
+	err = mlx5_wc_create_cq(mdev, &sq->cq);
+	if (err)
+		goto err_create_cq;
+
+	err = mlx5_wc_create_sq(mdev, sq);
+	if (err)
+		goto err_create_sq;
+
+	for (i = 0; i < TEST_WC_NUM_WQES - 1; i++)
+		mlx5_wc_post_nop(sq, false);
+
+	mlx5_wc_post_nop(sq, true);
+
+	expires = jiffies + TEST_WC_POLLING_MAX_TIME_JIFFIES;
+	do {
+		err = mlx5_wc_poll_cq(sq);
+		if (err)
+			usleep_range(2, 10);
+	} while (mdev->wc_state == MLX5_WC_STATE_UNINITIALIZED &&
+		 time_is_after_jiffies(expires));
+
+	mlx5_wc_destroy_sq(sq);
+
+err_create_sq:
+	mlx5_wc_destroy_cq(&sq->cq);
+err_create_cq:
+	mlx5_free_bfreg(mdev, &sq->bfreg);
+err_alloc_bfreg:
+	kfree(sq);
+}
+
+bool mlx5_wc_support_get(struct mlx5_core_dev *mdev)
+{
+	struct mlx5_core_dev *parent = NULL;
+
+	if (!MLX5_CAP_GEN(mdev, bf)) {
+		mlx5_core_dbg(mdev, "BlueFlame not supported\n");
+		goto out;
+	}
+
+	if (!MLX5_CAP_GEN(mdev, log_max_sq)) {
+		mlx5_core_dbg(mdev, "SQ not supported\n");
+		goto out;
+	}
+
+	if (mdev->wc_state != MLX5_WC_STATE_UNINITIALIZED)
+		/* No need to lock anything as we perform WC test only
+		 * once for whole device and was already done.
+		 */
+		goto out;
+
+	mutex_lock(&mdev->wc_state_lock);
+
+	if (mdev->wc_state != MLX5_WC_STATE_UNINITIALIZED)
+		goto unlock;
+
+#ifdef CONFIG_MLX5_SF
+	if (mlx5_core_is_sf(mdev))
+		parent = mdev->priv.parent_mdev;
+#endif
+
+	if (parent) {
+		mutex_lock(&parent->wc_state_lock);
+
+		mlx5_core_test_wc(parent);
+
+		mlx5_core_dbg(mdev, "parent set wc_state=%d\n",
+			      parent->wc_state);
+		mdev->wc_state = parent->wc_state;
+
+		mutex_unlock(&parent->wc_state_lock);
+	}
+
+	mlx5_core_test_wc(mdev);
+
+unlock:
+	mutex_unlock(&mdev->wc_state_lock);
+out:
+	mlx5_core_dbg(mdev, "wc_state=%d\n", mdev->wc_state);
+
+	return mdev->wc_state == MLX5_WC_STATE_SUPPORTED;
+}
+EXPORT_SYMBOL(mlx5_wc_support_get);
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 779cfdf2e9d6..0d31f77396fc 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -766,6 +766,12 @@ struct mlx5_hca_cap {
 	u32 max[MLX5_UN_SZ_DW(hca_cap_union)];
 };
 
+enum mlx5_wc_state {
+	MLX5_WC_STATE_UNINITIALIZED,
+	MLX5_WC_STATE_UNSUPPORTED,
+	MLX5_WC_STATE_SUPPORTED,
+};
+
 struct mlx5_core_dev {
 	struct device *device;
 	enum mlx5_coredev_type coredev_type;
@@ -824,6 +830,9 @@ struct mlx5_core_dev {
 #endif
 	u64 num_ipsec_offloads;
 	struct mlx5_sd          *sd;
+	enum mlx5_wc_state wc_state;
+	/* sync write combining state */
+	struct mutex wc_state_lock;
 };
 
 struct mlx5_db {
@@ -1375,4 +1384,6 @@ static inline bool mlx5_is_macsec_roce_supported(struct mlx5_core_dev *mdev)
 enum {
 	MLX5_OCTWORD = 16,
 };
+
+bool mlx5_wc_support_get(struct mlx5_core_dev *mdev);
 #endif /* MLX5_DRIVER_H */
-- 
cgit v1.2.3-58-ga151


From b339e0a39dc37726712b9f0485d78fe4306d1667 Mon Sep 17 00:00:00 2001
From: Patrisious Haddad <phaddad@nvidia.com>
Date: Thu, 13 Jun 2024 21:00:04 +0300
Subject: RDMA/mlx5: Add Qcounters
 req_transport_retries_exceeded/req_rnr_retries_exceeded

The req_transport_retries_exceeded counter shows the number of times
requester detected transport retries exceed error.

The req_rnr_retries_exceeded counter show the number of times the
requester detected RNR NAKs retries exceed error.

Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Link: https://lore.kernel.org/r/250466af94f4989d638fab168e246035530e912f.1718301543.git.leon@kernel.org
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
---
 drivers/infiniband/hw/mlx5/counters.c | 4 ++++
 include/linux/mlx5/mlx5_ifc.h         | 6 +++++-
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/mlx5/counters.c b/drivers/infiniband/hw/mlx5/counters.c
index 8300ce622835..4f6c1968a2ee 100644
--- a/drivers/infiniband/hw/mlx5/counters.c
+++ b/drivers/infiniband/hw/mlx5/counters.c
@@ -83,6 +83,8 @@ static const struct mlx5_ib_counter extended_err_cnts[] = {
 	INIT_Q_COUNTER(resp_remote_access_errors),
 	INIT_Q_COUNTER(resp_cqe_flush_error),
 	INIT_Q_COUNTER(req_cqe_flush_error),
+	INIT_Q_COUNTER(req_transport_retries_exceeded),
+	INIT_Q_COUNTER(req_rnr_retries_exceeded),
 };
 
 static const struct mlx5_ib_counter roce_accl_cnts[] = {
@@ -102,6 +104,8 @@ static const struct mlx5_ib_counter vport_extended_err_cnts[] = {
 	INIT_VPORT_Q_COUNTER(resp_remote_access_errors),
 	INIT_VPORT_Q_COUNTER(resp_cqe_flush_error),
 	INIT_VPORT_Q_COUNTER(req_cqe_flush_error),
+	INIT_VPORT_Q_COUNTER(req_transport_retries_exceeded),
+	INIT_VPORT_Q_COUNTER(req_rnr_retries_exceeded),
 };
 
 static const struct mlx5_ib_counter vport_roce_accl_cnts[] = {
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 5df52e15f7d6..09d9d87d62c6 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -5629,7 +5629,11 @@ struct mlx5_ifc_query_q_counter_out_bits {
 
 	u8         local_ack_timeout_err[0x20];
 
-	u8         reserved_at_320[0xa0];
+	u8         reserved_at_320[0x60];
+
+	u8         req_rnr_retries_exceeded[0x20];
+
+	u8         reserved_at_3a0[0x20];
 
 	u8         resp_local_length_error[0x20];
 
-- 
cgit v1.2.3-58-ga151


From 3d530eeaf8e79770048e35b1b93f8ffb36a8a133 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <wsa+renesas@sang-engineering.com>
Date: Mon, 3 Jun 2024 11:15:38 +0200
Subject: wifi: rtw89: use 'time_left' variable with
 wait_for_completion_timeout()

There is a confusing pattern in the kernel to use a variable named 'timeout' to
store the result of wait_for_completion_timeout() causing patterns like:

	timeout = wait_for_completion_timeout(...)
	if (!timeout) return -ETIMEDOUT;

with all kinds of permutations. Use 'time_left' as a variable to make the code
self explaining.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240603091541.8367-6-wsa+renesas@sang-engineering.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 3e99b63a7995..fe403fb7e8a2 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -4054,15 +4054,15 @@ void rtw89_core_update_beacon_work(struct work_struct *work)
 int rtw89_wait_for_cond(struct rtw89_wait_info *wait, unsigned int cond)
 {
 	struct completion *cmpl = &wait->completion;
-	unsigned long timeout;
+	unsigned long time_left;
 	unsigned int cur;
 
 	cur = atomic_cmpxchg(&wait->cond, RTW89_WAIT_COND_IDLE, cond);
 	if (cur != RTW89_WAIT_COND_IDLE)
 		return -EBUSY;
 
-	timeout = wait_for_completion_timeout(cmpl, RTW89_WAIT_FOR_COND_TIMEOUT);
-	if (timeout == 0) {
+	time_left = wait_for_completion_timeout(cmpl, RTW89_WAIT_FOR_COND_TIMEOUT);
+	if (time_left == 0) {
 		atomic_set(&wait->cond, RTW89_WAIT_COND_IDLE);
 		return -ETIMEDOUT;
 	}
-- 
cgit v1.2.3-58-ga151


From 5a72e198a6e965eb5c658e63711c889430075839 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 7 Jun 2024 15:06:57 +0800
Subject: wifi: rtw89: 8852bx: move common code from 8852b to 8852b_common

New upcoming chip is RTL8852BE-VT (or RTL8852BTE; 8852BT PCIE interface),
which is a variant of 8852B, and many codes excepting to RF calibration
can be shared, so move common code to an new kernel module named
rtw89_8852b_common.ko.

No logic change.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240607070659.80263-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/Kconfig         |    4 +
 drivers/net/wireless/realtek/rtw89/Makefile        |    3 +
 drivers/net/wireless/realtek/rtw89/rtw8852b.c      | 1691 +------------------
 drivers/net/wireless/realtek/rtw89/rtw8852b.h      |  122 --
 .../net/wireless/realtek/rtw89/rtw8852b_common.c   | 1701 ++++++++++++++++++++
 .../net/wireless/realtek/rtw89/rtw8852b_common.h   |  339 ++++
 drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c  |   19 +-
 7 files changed, 2079 insertions(+), 1800 deletions(-)
 create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
 create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8852b_common.h

diff --git a/drivers/net/wireless/realtek/rtw89/Kconfig b/drivers/net/wireless/realtek/rtw89/Kconfig
index eaea4eaeb361..3c9f864805b1 100644
--- a/drivers/net/wireless/realtek/rtw89/Kconfig
+++ b/drivers/net/wireless/realtek/rtw89/Kconfig
@@ -22,6 +22,9 @@ config RTW89_8851B
 config RTW89_8852A
 	tristate
 
+config RTW89_8852B_COMMON
+	tristate
+
 config RTW89_8852B
 	tristate
 
@@ -59,6 +62,7 @@ config RTW89_8852BE
 	select RTW89_CORE
 	select RTW89_PCI
 	select RTW89_8852B
+	select RTW89_8852B_COMMON
 	help
 	  Select this option will enable support for 8852BE chipset
 
diff --git a/drivers/net/wireless/realtek/rtw89/Makefile b/drivers/net/wireless/realtek/rtw89/Makefile
index 86a553fb0136..688847ce7084 100644
--- a/drivers/net/wireless/realtek/rtw89/Makefile
+++ b/drivers/net/wireless/realtek/rtw89/Makefile
@@ -39,6 +39,9 @@ rtw89_8852a-objs := rtw8852a.o \
 obj-$(CONFIG_RTW89_8852AE) += rtw89_8852ae.o
 rtw89_8852ae-objs := rtw8852ae.o
 
+obj-$(CONFIG_RTW89_8852B_COMMON) += rtw89_8852b_common.o
+rtw89_8852b_common-objs := rtw8852b_common.o
+
 obj-$(CONFIG_RTW89_8852B) += rtw89_8852b.o
 rtw89_8852b-objs := rtw8852b.o \
 		    rtw8852b_table.o \
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index 517ad1a763fa..381c881fee69 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -8,6 +8,7 @@
 #include "phy.h"
 #include "reg.h"
 #include "rtw8852b.h"
+#include "rtw8852b_common.h"
 #include "rtw8852b_rfk.h"
 #include "rtw8852b_table.h"
 #include "txrx.h"
@@ -65,167 +66,6 @@ static const struct rtw89_dle_mem rtw8852b_dle_mem_pcie[] = {
 			       NULL},
 };
 
-static const struct rtw89_reg3_def rtw8852b_pmac_ht20_mcs7_tbl[] = {
-	{0x4580, 0x0000ffff, 0x0},
-	{0x4580, 0xffff0000, 0x0},
-	{0x4584, 0x0000ffff, 0x0},
-	{0x4584, 0xffff0000, 0x0},
-	{0x4580, 0x0000ffff, 0x1},
-	{0x4578, 0x00ffffff, 0x2018b},
-	{0x4570, 0x03ffffff, 0x7},
-	{0x4574, 0x03ffffff, 0x32407},
-	{0x45b8, 0x00000010, 0x0},
-	{0x45b8, 0x00000100, 0x0},
-	{0x45b8, 0x00000080, 0x0},
-	{0x45b8, 0x00000008, 0x0},
-	{0x45a0, 0x0000ff00, 0x0},
-	{0x45a0, 0xff000000, 0x1},
-	{0x45a4, 0x0000ff00, 0x2},
-	{0x45a4, 0xff000000, 0x3},
-	{0x45b8, 0x00000020, 0x0},
-	{0x4568, 0xe0000000, 0x0},
-	{0x45b8, 0x00000002, 0x1},
-	{0x456c, 0xe0000000, 0x0},
-	{0x45b4, 0x00006000, 0x0},
-	{0x45b4, 0x00001800, 0x1},
-	{0x45b8, 0x00000040, 0x0},
-	{0x45b8, 0x00000004, 0x0},
-	{0x45b8, 0x00000200, 0x0},
-	{0x4598, 0xf8000000, 0x0},
-	{0x45b8, 0x00100000, 0x0},
-	{0x45a8, 0x00000fc0, 0x0},
-	{0x45b8, 0x00200000, 0x0},
-	{0x45b0, 0x00000038, 0x0},
-	{0x45b0, 0x000001c0, 0x0},
-	{0x45a0, 0x000000ff, 0x0},
-	{0x45b8, 0x00400000, 0x0},
-	{0x4590, 0x000007ff, 0x0},
-	{0x45b0, 0x00000e00, 0x0},
-	{0x45ac, 0x0000001f, 0x0},
-	{0x45b8, 0x00800000, 0x0},
-	{0x45a8, 0x0003f000, 0x0},
-	{0x45b8, 0x01000000, 0x0},
-	{0x45b0, 0x00007000, 0x0},
-	{0x45b0, 0x00038000, 0x0},
-	{0x45a0, 0x00ff0000, 0x0},
-	{0x45b8, 0x02000000, 0x0},
-	{0x4590, 0x003ff800, 0x0},
-	{0x45b0, 0x001c0000, 0x0},
-	{0x45ac, 0x000003e0, 0x0},
-	{0x45b8, 0x04000000, 0x0},
-	{0x45a8, 0x00fc0000, 0x0},
-	{0x45b8, 0x08000000, 0x0},
-	{0x45b0, 0x00e00000, 0x0},
-	{0x45b0, 0x07000000, 0x0},
-	{0x45a4, 0x000000ff, 0x0},
-	{0x45b8, 0x10000000, 0x0},
-	{0x4594, 0x000007ff, 0x0},
-	{0x45b0, 0x38000000, 0x0},
-	{0x45ac, 0x00007c00, 0x0},
-	{0x45b8, 0x20000000, 0x0},
-	{0x45a8, 0x3f000000, 0x0},
-	{0x45b8, 0x40000000, 0x0},
-	{0x45b4, 0x00000007, 0x0},
-	{0x45b4, 0x00000038, 0x0},
-	{0x45a4, 0x00ff0000, 0x0},
-	{0x45b8, 0x80000000, 0x0},
-	{0x4594, 0x003ff800, 0x0},
-	{0x45b4, 0x000001c0, 0x0},
-	{0x4598, 0xf8000000, 0x0},
-	{0x45b8, 0x00100000, 0x0},
-	{0x45a8, 0x00000fc0, 0x7},
-	{0x45b8, 0x00200000, 0x0},
-	{0x45b0, 0x00000038, 0x0},
-	{0x45b0, 0x000001c0, 0x0},
-	{0x45a0, 0x000000ff, 0x0},
-	{0x45b4, 0x06000000, 0x0},
-	{0x45b0, 0x00000007, 0x0},
-	{0x45b8, 0x00080000, 0x0},
-	{0x45a8, 0x0000003f, 0x0},
-	{0x457c, 0xffe00000, 0x1},
-	{0x4530, 0xffffffff, 0x0},
-	{0x4588, 0x00003fff, 0x0},
-	{0x4598, 0x000001ff, 0x0},
-	{0x4534, 0xffffffff, 0x0},
-	{0x4538, 0xffffffff, 0x0},
-	{0x453c, 0xffffffff, 0x0},
-	{0x4588, 0x0fffc000, 0x0},
-	{0x4598, 0x0003fe00, 0x0},
-	{0x4540, 0xffffffff, 0x0},
-	{0x4544, 0xffffffff, 0x0},
-	{0x4548, 0xffffffff, 0x0},
-	{0x458c, 0x00003fff, 0x0},
-	{0x4598, 0x07fc0000, 0x0},
-	{0x454c, 0xffffffff, 0x0},
-	{0x4550, 0xffffffff, 0x0},
-	{0x4554, 0xffffffff, 0x0},
-	{0x458c, 0x0fffc000, 0x0},
-	{0x459c, 0x000001ff, 0x0},
-	{0x4558, 0xffffffff, 0x0},
-	{0x455c, 0xffffffff, 0x0},
-	{0x4530, 0xffffffff, 0x4e790001},
-	{0x4588, 0x00003fff, 0x0},
-	{0x4598, 0x000001ff, 0x1},
-	{0x4534, 0xffffffff, 0x0},
-	{0x4538, 0xffffffff, 0x4b},
-	{0x45ac, 0x38000000, 0x7},
-	{0x4588, 0xf0000000, 0x0},
-	{0x459c, 0x7e000000, 0x0},
-	{0x45b8, 0x00040000, 0x0},
-	{0x45b8, 0x00020000, 0x0},
-	{0x4590, 0xffc00000, 0x0},
-	{0x45b8, 0x00004000, 0x0},
-	{0x4578, 0xff000000, 0x0},
-	{0x45b8, 0x00000400, 0x0},
-	{0x45b8, 0x00000800, 0x0},
-	{0x45b8, 0x00001000, 0x0},
-	{0x45b8, 0x00002000, 0x0},
-	{0x45b4, 0x00018000, 0x0},
-	{0x45ac, 0x07800000, 0x0},
-	{0x45b4, 0x00000600, 0x2},
-	{0x459c, 0x0001fe00, 0x80},
-	{0x45ac, 0x00078000, 0x3},
-	{0x459c, 0x01fe0000, 0x1},
-};
-
-static const struct rtw89_reg3_def rtw8852b_btc_preagc_en_defs[] = {
-	{0x46D0, GENMASK(1, 0), 0x3},
-	{0x4790, GENMASK(1, 0), 0x3},
-	{0x4AD4, GENMASK(31, 0), 0xf},
-	{0x4AE0, GENMASK(31, 0), 0xf},
-	{0x4688, GENMASK(31, 24), 0x80},
-	{0x476C, GENMASK(31, 24), 0x80},
-	{0x4694, GENMASK(7, 0), 0x80},
-	{0x4694, GENMASK(15, 8), 0x80},
-	{0x4778, GENMASK(7, 0), 0x80},
-	{0x4778, GENMASK(15, 8), 0x80},
-	{0x4AE4, GENMASK(23, 0), 0x780D1E},
-	{0x4AEC, GENMASK(23, 0), 0x780D1E},
-	{0x469C, GENMASK(31, 26), 0x34},
-	{0x49F0, GENMASK(31, 26), 0x34},
-};
-
-static DECLARE_PHY_REG3_TBL(rtw8852b_btc_preagc_en_defs);
-
-static const struct rtw89_reg3_def rtw8852b_btc_preagc_dis_defs[] = {
-	{0x46D0, GENMASK(1, 0), 0x0},
-	{0x4790, GENMASK(1, 0), 0x0},
-	{0x4AD4, GENMASK(31, 0), 0x60},
-	{0x4AE0, GENMASK(31, 0), 0x60},
-	{0x4688, GENMASK(31, 24), 0x1a},
-	{0x476C, GENMASK(31, 24), 0x1a},
-	{0x4694, GENMASK(7, 0), 0x2a},
-	{0x4694, GENMASK(15, 8), 0x2a},
-	{0x4778, GENMASK(7, 0), 0x2a},
-	{0x4778, GENMASK(15, 8), 0x2a},
-	{0x4AE4, GENMASK(23, 0), 0x79E99E},
-	{0x4AEC, GENMASK(23, 0), 0x79E99E},
-	{0x469C, GENMASK(31, 26), 0x26},
-	{0x49F0, GENMASK(31, 26), 0x26},
-};
-
-static DECLARE_PHY_REG3_TBL(rtw8852b_btc_preagc_dis_defs);
-
 static const u32 rtw8852b_h2c_regs[RTW89_H2CREG_MAX] = {
 	R_AX_H2CREG_DATA0, R_AX_H2CREG_DATA1,  R_AX_H2CREG_DATA2,
 	R_AX_H2CREG_DATA3
@@ -591,806 +431,6 @@ static int rtw8852b_pwr_off_func(struct rtw89_dev *rtwdev)
 	return 0;
 }
 
-static void rtw8852be_efuse_parsing(struct rtw89_efuse *efuse,
-				    struct rtw8852b_efuse *map)
-{
-	ether_addr_copy(efuse->addr, map->e.mac_addr);
-	efuse->rfe_type = map->rfe_type;
-	efuse->xtal_cap = map->xtal_k;
-}
-
-static void rtw8852b_efuse_parsing_tssi(struct rtw89_dev *rtwdev,
-					struct rtw8852b_efuse *map)
-{
-	struct rtw89_tssi_info *tssi = &rtwdev->tssi;
-	struct rtw8852b_tssi_offset *ofst[] = {&map->path_a_tssi, &map->path_b_tssi};
-	u8 i, j;
-
-	tssi->thermal[RF_PATH_A] = map->path_a_therm;
-	tssi->thermal[RF_PATH_B] = map->path_b_therm;
-
-	for (i = 0; i < RF_PATH_NUM_8852B; i++) {
-		memcpy(tssi->tssi_cck[i], ofst[i]->cck_tssi,
-		       sizeof(ofst[i]->cck_tssi));
-
-		for (j = 0; j < TSSI_CCK_CH_GROUP_NUM; j++)
-			rtw89_debug(rtwdev, RTW89_DBG_TSSI,
-				    "[TSSI][EFUSE] path=%d cck[%d]=0x%x\n",
-				    i, j, tssi->tssi_cck[i][j]);
-
-		memcpy(tssi->tssi_mcs[i], ofst[i]->bw40_tssi,
-		       sizeof(ofst[i]->bw40_tssi));
-		memcpy(tssi->tssi_mcs[i] + TSSI_MCS_2G_CH_GROUP_NUM,
-		       ofst[i]->bw40_1s_tssi_5g, sizeof(ofst[i]->bw40_1s_tssi_5g));
-
-		for (j = 0; j < TSSI_MCS_CH_GROUP_NUM; j++)
-			rtw89_debug(rtwdev, RTW89_DBG_TSSI,
-				    "[TSSI][EFUSE] path=%d mcs[%d]=0x%x\n",
-				    i, j, tssi->tssi_mcs[i][j]);
-	}
-}
-
-static bool _decode_efuse_gain(u8 data, s8 *high, s8 *low)
-{
-	if (high)
-		*high = sign_extend32(FIELD_GET(GENMASK(7,  4), data), 3);
-	if (low)
-		*low = sign_extend32(FIELD_GET(GENMASK(3,  0), data), 3);
-
-	return data != 0xff;
-}
-
-static void rtw8852b_efuse_parsing_gain_offset(struct rtw89_dev *rtwdev,
-					       struct rtw8852b_efuse *map)
-{
-	struct rtw89_phy_efuse_gain *gain = &rtwdev->efuse_gain;
-	bool valid = false;
-
-	valid |= _decode_efuse_gain(map->rx_gain_2g_cck,
-				    &gain->offset[RF_PATH_A][RTW89_GAIN_OFFSET_2G_CCK],
-				    &gain->offset[RF_PATH_B][RTW89_GAIN_OFFSET_2G_CCK]);
-	valid |= _decode_efuse_gain(map->rx_gain_2g_ofdm,
-				    &gain->offset[RF_PATH_A][RTW89_GAIN_OFFSET_2G_OFDM],
-				    &gain->offset[RF_PATH_B][RTW89_GAIN_OFFSET_2G_OFDM]);
-	valid |= _decode_efuse_gain(map->rx_gain_5g_low,
-				    &gain->offset[RF_PATH_A][RTW89_GAIN_OFFSET_5G_LOW],
-				    &gain->offset[RF_PATH_B][RTW89_GAIN_OFFSET_5G_LOW]);
-	valid |= _decode_efuse_gain(map->rx_gain_5g_mid,
-				    &gain->offset[RF_PATH_A][RTW89_GAIN_OFFSET_5G_MID],
-				    &gain->offset[RF_PATH_B][RTW89_GAIN_OFFSET_5G_MID]);
-	valid |= _decode_efuse_gain(map->rx_gain_5g_high,
-				    &gain->offset[RF_PATH_A][RTW89_GAIN_OFFSET_5G_HIGH],
-				    &gain->offset[RF_PATH_B][RTW89_GAIN_OFFSET_5G_HIGH]);
-
-	gain->offset_valid = valid;
-}
-
-static int rtw8852b_read_efuse(struct rtw89_dev *rtwdev, u8 *log_map,
-			       enum rtw89_efuse_block block)
-{
-	struct rtw89_efuse *efuse = &rtwdev->efuse;
-	struct rtw8852b_efuse *map;
-
-	map = (struct rtw8852b_efuse *)log_map;
-
-	efuse->country_code[0] = map->country_code[0];
-	efuse->country_code[1] = map->country_code[1];
-	rtw8852b_efuse_parsing_tssi(rtwdev, map);
-	rtw8852b_efuse_parsing_gain_offset(rtwdev, map);
-
-	switch (rtwdev->hci.type) {
-	case RTW89_HCI_TYPE_PCIE:
-		rtw8852be_efuse_parsing(efuse, map);
-		break;
-	default:
-		return -EOPNOTSUPP;
-	}
-
-	rtw89_info(rtwdev, "chip rfe_type is %d\n", efuse->rfe_type);
-
-	return 0;
-}
-
-static void rtw8852b_phycap_parsing_power_cal(struct rtw89_dev *rtwdev, u8 *phycap_map)
-{
-#define PWR_K_CHK_OFFSET 0x5E9
-#define PWR_K_CHK_VALUE 0xAA
-	u32 offset = PWR_K_CHK_OFFSET - rtwdev->chip->phycap_addr;
-
-	if (phycap_map[offset] == PWR_K_CHK_VALUE)
-		rtwdev->efuse.power_k_valid = true;
-}
-
-static void rtw8852b_phycap_parsing_tssi(struct rtw89_dev *rtwdev, u8 *phycap_map)
-{
-	struct rtw89_tssi_info *tssi = &rtwdev->tssi;
-	static const u32 tssi_trim_addr[RF_PATH_NUM_8852B] = {0x5D6, 0x5AB};
-	u32 addr = rtwdev->chip->phycap_addr;
-	bool pg = false;
-	u32 ofst;
-	u8 i, j;
-
-	for (i = 0; i < RF_PATH_NUM_8852B; i++) {
-		for (j = 0; j < TSSI_TRIM_CH_GROUP_NUM; j++) {
-			/* addrs are in decreasing order */
-			ofst = tssi_trim_addr[i] - addr - j;
-			tssi->tssi_trim[i][j] = phycap_map[ofst];
-
-			if (phycap_map[ofst] != 0xff)
-				pg = true;
-		}
-	}
-
-	if (!pg) {
-		memset(tssi->tssi_trim, 0, sizeof(tssi->tssi_trim));
-		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
-			    "[TSSI][TRIM] no PG, set all trim info to 0\n");
-	}
-
-	for (i = 0; i < RF_PATH_NUM_8852B; i++)
-		for (j = 0; j < TSSI_TRIM_CH_GROUP_NUM; j++)
-			rtw89_debug(rtwdev, RTW89_DBG_TSSI,
-				    "[TSSI] path=%d idx=%d trim=0x%x addr=0x%x\n",
-				    i, j, tssi->tssi_trim[i][j],
-				    tssi_trim_addr[i] - j);
-}
-
-static void rtw8852b_phycap_parsing_thermal_trim(struct rtw89_dev *rtwdev,
-						 u8 *phycap_map)
-{
-	struct rtw89_power_trim_info *info = &rtwdev->pwr_trim;
-	static const u32 thm_trim_addr[RF_PATH_NUM_8852B] = {0x5DF, 0x5DC};
-	u32 addr = rtwdev->chip->phycap_addr;
-	u8 i;
-
-	for (i = 0; i < RF_PATH_NUM_8852B; i++) {
-		info->thermal_trim[i] = phycap_map[thm_trim_addr[i] - addr];
-
-		rtw89_debug(rtwdev, RTW89_DBG_RFK,
-			    "[THERMAL][TRIM] path=%d thermal_trim=0x%x\n",
-			    i, info->thermal_trim[i]);
-
-		if (info->thermal_trim[i] != 0xff)
-			info->pg_thermal_trim = true;
-	}
-}
-
-static void rtw8852b_thermal_trim(struct rtw89_dev *rtwdev)
-{
-#define __thm_setting(raw)				\
-({							\
-	u8 __v = (raw);					\
-	((__v & 0x1) << 3) | ((__v & 0x1f) >> 1);	\
-})
-	struct rtw89_power_trim_info *info = &rtwdev->pwr_trim;
-	u8 i, val;
-
-	if (!info->pg_thermal_trim) {
-		rtw89_debug(rtwdev, RTW89_DBG_RFK,
-			    "[THERMAL][TRIM] no PG, do nothing\n");
-
-		return;
-	}
-
-	for (i = 0; i < RF_PATH_NUM_8852B; i++) {
-		val = __thm_setting(info->thermal_trim[i]);
-		rtw89_write_rf(rtwdev, i, RR_TM2, RR_TM2_OFF, val);
-
-		rtw89_debug(rtwdev, RTW89_DBG_RFK,
-			    "[THERMAL][TRIM] path=%d thermal_setting=0x%x\n",
-			    i, val);
-	}
-#undef __thm_setting
-}
-
-static void rtw8852b_phycap_parsing_pa_bias_trim(struct rtw89_dev *rtwdev,
-						 u8 *phycap_map)
-{
-	struct rtw89_power_trim_info *info = &rtwdev->pwr_trim;
-	static const u32 pabias_trim_addr[RF_PATH_NUM_8852B] = {0x5DE, 0x5DB};
-	u32 addr = rtwdev->chip->phycap_addr;
-	u8 i;
-
-	for (i = 0; i < RF_PATH_NUM_8852B; i++) {
-		info->pa_bias_trim[i] = phycap_map[pabias_trim_addr[i] - addr];
-
-		rtw89_debug(rtwdev, RTW89_DBG_RFK,
-			    "[PA_BIAS][TRIM] path=%d pa_bias_trim=0x%x\n",
-			    i, info->pa_bias_trim[i]);
-
-		if (info->pa_bias_trim[i] != 0xff)
-			info->pg_pa_bias_trim = true;
-	}
-}
-
-static void rtw8852b_pa_bias_trim(struct rtw89_dev *rtwdev)
-{
-	struct rtw89_power_trim_info *info = &rtwdev->pwr_trim;
-	u8 pabias_2g, pabias_5g;
-	u8 i;
-
-	if (!info->pg_pa_bias_trim) {
-		rtw89_debug(rtwdev, RTW89_DBG_RFK,
-			    "[PA_BIAS][TRIM] no PG, do nothing\n");
-
-		return;
-	}
-
-	for (i = 0; i < RF_PATH_NUM_8852B; i++) {
-		pabias_2g = FIELD_GET(GENMASK(3, 0), info->pa_bias_trim[i]);
-		pabias_5g = FIELD_GET(GENMASK(7, 4), info->pa_bias_trim[i]);
-
-		rtw89_debug(rtwdev, RTW89_DBG_RFK,
-			    "[PA_BIAS][TRIM] path=%d 2G=0x%x 5G=0x%x\n",
-			    i, pabias_2g, pabias_5g);
-
-		rtw89_write_rf(rtwdev, i, RR_BIASA, RR_BIASA_TXG, pabias_2g);
-		rtw89_write_rf(rtwdev, i, RR_BIASA, RR_BIASA_TXA, pabias_5g);
-	}
-}
-
-static void rtw8852b_phycap_parsing_gain_comp(struct rtw89_dev *rtwdev, u8 *phycap_map)
-{
-	static const u32 comp_addrs[][RTW89_SUBBAND_2GHZ_5GHZ_NR] = {
-		{0x5BB, 0x5BA, 0, 0x5B9, 0x5B8},
-		{0x590, 0x58F, 0, 0x58E, 0x58D},
-	};
-	struct rtw89_phy_efuse_gain *gain = &rtwdev->efuse_gain;
-	u32 phycap_addr = rtwdev->chip->phycap_addr;
-	bool valid = false;
-	int path, i;
-	u8 data;
-
-	for (path = 0; path < 2; path++)
-		for (i = 0; i < RTW89_SUBBAND_2GHZ_5GHZ_NR; i++) {
-			if (comp_addrs[path][i] == 0)
-				continue;
-
-			data = phycap_map[comp_addrs[path][i] - phycap_addr];
-			valid |= _decode_efuse_gain(data, NULL,
-						    &gain->comp[path][i]);
-		}
-
-	gain->comp_valid = valid;
-}
-
-static int rtw8852b_read_phycap(struct rtw89_dev *rtwdev, u8 *phycap_map)
-{
-	rtw8852b_phycap_parsing_power_cal(rtwdev, phycap_map);
-	rtw8852b_phycap_parsing_tssi(rtwdev, phycap_map);
-	rtw8852b_phycap_parsing_thermal_trim(rtwdev, phycap_map);
-	rtw8852b_phycap_parsing_pa_bias_trim(rtwdev, phycap_map);
-	rtw8852b_phycap_parsing_gain_comp(rtwdev, phycap_map);
-
-	return 0;
-}
-
-static void rtw8852b_power_trim(struct rtw89_dev *rtwdev)
-{
-	rtw8852b_thermal_trim(rtwdev);
-	rtw8852b_pa_bias_trim(rtwdev);
-}
-
-static void rtw8852b_set_channel_mac(struct rtw89_dev *rtwdev,
-				     const struct rtw89_chan *chan,
-				     u8 mac_idx)
-{
-	u32 rf_mod = rtw89_mac_reg_by_idx(rtwdev, R_AX_WMAC_RFMOD, mac_idx);
-	u32 sub_carr = rtw89_mac_reg_by_idx(rtwdev, R_AX_TX_SUB_CARRIER_VALUE, mac_idx);
-	u32 chk_rate = rtw89_mac_reg_by_idx(rtwdev, R_AX_TXRATE_CHK, mac_idx);
-	u8 txsc20 = 0, txsc40 = 0;
-
-	switch (chan->band_width) {
-	case RTW89_CHANNEL_WIDTH_80:
-		txsc40 = rtw89_phy_get_txsc(rtwdev, chan, RTW89_CHANNEL_WIDTH_40);
-		fallthrough;
-	case RTW89_CHANNEL_WIDTH_40:
-		txsc20 = rtw89_phy_get_txsc(rtwdev, chan, RTW89_CHANNEL_WIDTH_20);
-		break;
-	default:
-		break;
-	}
-
-	switch (chan->band_width) {
-	case RTW89_CHANNEL_WIDTH_80:
-		rtw89_write8_mask(rtwdev, rf_mod, B_AX_WMAC_RFMOD_MASK, BIT(1));
-		rtw89_write32(rtwdev, sub_carr, txsc20 | (txsc40 << 4));
-		break;
-	case RTW89_CHANNEL_WIDTH_40:
-		rtw89_write8_mask(rtwdev, rf_mod, B_AX_WMAC_RFMOD_MASK, BIT(0));
-		rtw89_write32(rtwdev, sub_carr, txsc20);
-		break;
-	case RTW89_CHANNEL_WIDTH_20:
-		rtw89_write8_clr(rtwdev, rf_mod, B_AX_WMAC_RFMOD_MASK);
-		rtw89_write32(rtwdev, sub_carr, 0);
-		break;
-	default:
-		break;
-	}
-
-	if (chan->channel > 14) {
-		rtw89_write8_clr(rtwdev, chk_rate, B_AX_BAND_MODE);
-		rtw89_write8_set(rtwdev, chk_rate,
-				 B_AX_CHECK_CCK_EN | B_AX_RTS_LIMIT_IN_OFDM6);
-	} else {
-		rtw89_write8_set(rtwdev, chk_rate, B_AX_BAND_MODE);
-		rtw89_write8_clr(rtwdev, chk_rate,
-				 B_AX_CHECK_CCK_EN | B_AX_RTS_LIMIT_IN_OFDM6);
-	}
-}
-
-static const u32 rtw8852b_sco_barker_threshold[14] = {
-	0x1cfea, 0x1d0e1, 0x1d1d7, 0x1d2cd, 0x1d3c3, 0x1d4b9, 0x1d5b0, 0x1d6a6,
-	0x1d79c, 0x1d892, 0x1d988, 0x1da7f, 0x1db75, 0x1ddc4
-};
-
-static const u32 rtw8852b_sco_cck_threshold[14] = {
-	0x27de3, 0x27f35, 0x28088, 0x281da, 0x2832d, 0x2847f, 0x285d2, 0x28724,
-	0x28877, 0x289c9, 0x28b1c, 0x28c6e, 0x28dc1, 0x290ed
-};
-
-static void rtw8852b_ctrl_sco_cck(struct rtw89_dev *rtwdev, u8 primary_ch)
-{
-	u8 ch_element = primary_ch - 1;
-
-	rtw89_phy_write32_mask(rtwdev, R_RXSCOBC, B_RXSCOBC_TH,
-			       rtw8852b_sco_barker_threshold[ch_element]);
-	rtw89_phy_write32_mask(rtwdev, R_RXSCOCCK, B_RXSCOCCK_TH,
-			       rtw8852b_sco_cck_threshold[ch_element]);
-}
-
-static u8 rtw8852b_sco_mapping(u8 central_ch)
-{
-	if (central_ch == 1)
-		return 109;
-	else if (central_ch >= 2 && central_ch <= 6)
-		return 108;
-	else if (central_ch >= 7 && central_ch <= 10)
-		return 107;
-	else if (central_ch >= 11 && central_ch <= 14)
-		return 106;
-	else if (central_ch == 36 || central_ch == 38)
-		return 51;
-	else if (central_ch >= 40 && central_ch <= 58)
-		return 50;
-	else if (central_ch >= 60 && central_ch <= 64)
-		return 49;
-	else if (central_ch == 100 || central_ch == 102)
-		return 48;
-	else if (central_ch >= 104 && central_ch <= 126)
-		return 47;
-	else if (central_ch >= 128 && central_ch <= 151)
-		return 46;
-	else if (central_ch >= 153 && central_ch <= 177)
-		return 45;
-	else
-		return 0;
-}
-
-struct rtw8852b_bb_gain {
-	u32 gain_g[BB_PATH_NUM_8852B];
-	u32 gain_a[BB_PATH_NUM_8852B];
-	u32 gain_mask;
-};
-
-static const struct rtw8852b_bb_gain bb_gain_lna[LNA_GAIN_NUM] = {
-	{ .gain_g = {0x4678, 0x475C}, .gain_a = {0x45DC, 0x4740},
-	  .gain_mask = 0x00ff0000 },
-	{ .gain_g = {0x4678, 0x475C}, .gain_a = {0x45DC, 0x4740},
-	  .gain_mask = 0xff000000 },
-	{ .gain_g = {0x467C, 0x4760}, .gain_a = {0x4660, 0x4744},
-	  .gain_mask = 0x000000ff },
-	{ .gain_g = {0x467C, 0x4760}, .gain_a = {0x4660, 0x4744},
-	  .gain_mask = 0x0000ff00 },
-	{ .gain_g = {0x467C, 0x4760}, .gain_a = {0x4660, 0x4744},
-	  .gain_mask = 0x00ff0000 },
-	{ .gain_g = {0x467C, 0x4760}, .gain_a = {0x4660, 0x4744},
-	  .gain_mask = 0xff000000 },
-	{ .gain_g = {0x4680, 0x4764}, .gain_a = {0x4664, 0x4748},
-	  .gain_mask = 0x000000ff },
-};
-
-static const struct rtw8852b_bb_gain bb_gain_tia[TIA_GAIN_NUM] = {
-	{ .gain_g = {0x4680, 0x4764}, .gain_a = {0x4664, 0x4748},
-	  .gain_mask = 0x00ff0000 },
-	{ .gain_g = {0x4680, 0x4764}, .gain_a = {0x4664, 0x4748},
-	  .gain_mask = 0xff000000 },
-};
-
-static void rtw8852b_set_gain_error(struct rtw89_dev *rtwdev,
-				    enum rtw89_subband subband,
-				    enum rtw89_rf_path path)
-{
-	const struct rtw89_phy_bb_gain_info *gain = &rtwdev->bb_gain.ax;
-	u8 gain_band = rtw89_subband_to_bb_gain_band(subband);
-	s32 val;
-	u32 reg;
-	u32 mask;
-	int i;
-
-	for (i = 0; i < LNA_GAIN_NUM; i++) {
-		if (subband == RTW89_CH_2G)
-			reg = bb_gain_lna[i].gain_g[path];
-		else
-			reg = bb_gain_lna[i].gain_a[path];
-
-		mask = bb_gain_lna[i].gain_mask;
-		val = gain->lna_gain[gain_band][path][i];
-		rtw89_phy_write32_mask(rtwdev, reg, mask, val);
-	}
-
-	for (i = 0; i < TIA_GAIN_NUM; i++) {
-		if (subband == RTW89_CH_2G)
-			reg = bb_gain_tia[i].gain_g[path];
-		else
-			reg = bb_gain_tia[i].gain_a[path];
-
-		mask = bb_gain_tia[i].gain_mask;
-		val = gain->tia_gain[gain_band][path][i];
-		rtw89_phy_write32_mask(rtwdev, reg, mask, val);
-	}
-}
-
-static void rtw8852b_set_gain_offset(struct rtw89_dev *rtwdev,
-				     enum rtw89_subband subband,
-				     enum rtw89_phy_idx phy_idx)
-{
-	static const u32 gain_err_addr[2] = {R_P0_AGC_RSVD, R_P1_AGC_RSVD};
-	static const u32 rssi_ofst_addr[2] = {R_PATH0_G_TIA1_LNA6_OP1DB_V1,
-					      R_PATH1_G_TIA1_LNA6_OP1DB_V1};
-	struct rtw89_hal *hal = &rtwdev->hal;
-	struct rtw89_phy_efuse_gain *efuse_gain = &rtwdev->efuse_gain;
-	enum rtw89_gain_offset gain_ofdm_band;
-	s32 offset_a, offset_b;
-	s32 offset_ofdm, offset_cck;
-	s32 tmp;
-	u8 path;
-
-	if (!efuse_gain->comp_valid)
-		goto next;
-
-	for (path = RF_PATH_A; path < BB_PATH_NUM_8852B; path++) {
-		tmp = efuse_gain->comp[path][subband];
-		tmp = clamp_t(s32, tmp << 2, S8_MIN, S8_MAX);
-		rtw89_phy_write32_mask(rtwdev, gain_err_addr[path], MASKBYTE0, tmp);
-	}
-
-next:
-	if (!efuse_gain->offset_valid)
-		return;
-
-	gain_ofdm_band = rtw89_subband_to_gain_offset_band_of_ofdm(subband);
-
-	offset_a = -efuse_gain->offset[RF_PATH_A][gain_ofdm_band];
-	offset_b = -efuse_gain->offset[RF_PATH_B][gain_ofdm_band];
-
-	tmp = -((offset_a << 2) + (efuse_gain->offset_base[RTW89_PHY_0] >> 2));
-	tmp = clamp_t(s32, tmp, S8_MIN, S8_MAX);
-	rtw89_phy_write32_mask(rtwdev, rssi_ofst_addr[RF_PATH_A], B_PATH0_R_G_OFST_MASK, tmp);
-
-	tmp = -((offset_b << 2) + (efuse_gain->offset_base[RTW89_PHY_0] >> 2));
-	tmp = clamp_t(s32, tmp, S8_MIN, S8_MAX);
-	rtw89_phy_write32_mask(rtwdev, rssi_ofst_addr[RF_PATH_B], B_PATH0_R_G_OFST_MASK, tmp);
-
-	if (hal->antenna_rx == RF_B) {
-		offset_ofdm = -efuse_gain->offset[RF_PATH_B][gain_ofdm_band];
-		offset_cck = -efuse_gain->offset[RF_PATH_B][0];
-	} else {
-		offset_ofdm = -efuse_gain->offset[RF_PATH_A][gain_ofdm_band];
-		offset_cck = -efuse_gain->offset[RF_PATH_A][0];
-	}
-
-	tmp = (offset_ofdm << 4) + efuse_gain->offset_base[RTW89_PHY_0];
-	tmp = clamp_t(s32, tmp, S8_MIN, S8_MAX);
-	rtw89_phy_write32_idx(rtwdev, R_P0_RPL1, B_P0_RPL1_BIAS_MASK, tmp, phy_idx);
-
-	tmp = (offset_ofdm << 4) + efuse_gain->rssi_base[RTW89_PHY_0];
-	tmp = clamp_t(s32, tmp, S8_MIN, S8_MAX);
-	rtw89_phy_write32_idx(rtwdev, R_P1_RPL1, B_P0_RPL1_BIAS_MASK, tmp, phy_idx);
-
-	if (subband == RTW89_CH_2G) {
-		tmp = (offset_cck << 3) + (efuse_gain->offset_base[RTW89_PHY_0] >> 1);
-		tmp = clamp_t(s32, tmp, S8_MIN >> 1, S8_MAX >> 1);
-		rtw89_phy_write32_mask(rtwdev, R_RX_RPL_OFST,
-				       B_RX_RPL_OFST_CCK_MASK, tmp);
-	}
-}
-
-static
-void rtw8852b_set_rxsc_rpl_comp(struct rtw89_dev *rtwdev, enum rtw89_subband subband)
-{
-	const struct rtw89_phy_bb_gain_info *gain = &rtwdev->bb_gain.ax;
-	u8 band = rtw89_subband_to_bb_gain_band(subband);
-	u32 val;
-
-	val = FIELD_PREP(B_P0_RPL1_20_MASK, (gain->rpl_ofst_20[band][RF_PATH_A] +
-					     gain->rpl_ofst_20[band][RF_PATH_B]) / 2) |
-	      FIELD_PREP(B_P0_RPL1_40_MASK, (gain->rpl_ofst_40[band][RF_PATH_A][0] +
-					     gain->rpl_ofst_40[band][RF_PATH_B][0]) / 2) |
-	      FIELD_PREP(B_P0_RPL1_41_MASK, (gain->rpl_ofst_40[band][RF_PATH_A][1] +
-					     gain->rpl_ofst_40[band][RF_PATH_B][1]) / 2);
-	val >>= B_P0_RPL1_SHIFT;
-	rtw89_phy_write32_mask(rtwdev, R_P0_RPL1, B_P0_RPL1_MASK, val);
-	rtw89_phy_write32_mask(rtwdev, R_P1_RPL1, B_P0_RPL1_MASK, val);
-
-	val = FIELD_PREP(B_P0_RTL2_42_MASK, (gain->rpl_ofst_40[band][RF_PATH_A][2] +
-					     gain->rpl_ofst_40[band][RF_PATH_B][2]) / 2) |
-	      FIELD_PREP(B_P0_RTL2_80_MASK, (gain->rpl_ofst_80[band][RF_PATH_A][0] +
-					     gain->rpl_ofst_80[band][RF_PATH_B][0]) / 2) |
-	      FIELD_PREP(B_P0_RTL2_81_MASK, (gain->rpl_ofst_80[band][RF_PATH_A][1] +
-					     gain->rpl_ofst_80[band][RF_PATH_B][1]) / 2) |
-	      FIELD_PREP(B_P0_RTL2_8A_MASK, (gain->rpl_ofst_80[band][RF_PATH_A][10] +
-					     gain->rpl_ofst_80[band][RF_PATH_B][10]) / 2);
-	rtw89_phy_write32(rtwdev, R_P0_RPL2, val);
-	rtw89_phy_write32(rtwdev, R_P1_RPL2, val);
-
-	val = FIELD_PREP(B_P0_RTL3_82_MASK, (gain->rpl_ofst_80[band][RF_PATH_A][2] +
-					     gain->rpl_ofst_80[band][RF_PATH_B][2]) / 2) |
-	      FIELD_PREP(B_P0_RTL3_83_MASK, (gain->rpl_ofst_80[band][RF_PATH_A][3] +
-					     gain->rpl_ofst_80[band][RF_PATH_B][3]) / 2) |
-	      FIELD_PREP(B_P0_RTL3_84_MASK, (gain->rpl_ofst_80[band][RF_PATH_A][4] +
-					     gain->rpl_ofst_80[band][RF_PATH_B][4]) / 2) |
-	      FIELD_PREP(B_P0_RTL3_89_MASK, (gain->rpl_ofst_80[band][RF_PATH_A][9] +
-					     gain->rpl_ofst_80[band][RF_PATH_B][9]) / 2);
-	rtw89_phy_write32(rtwdev, R_P0_RPL3, val);
-	rtw89_phy_write32(rtwdev, R_P1_RPL3, val);
-}
-
-static void rtw8852b_ctrl_ch(struct rtw89_dev *rtwdev,
-			     const struct rtw89_chan *chan,
-			     enum rtw89_phy_idx phy_idx)
-{
-	u8 central_ch = chan->channel;
-	u8 subband = chan->subband_type;
-	u8 sco_comp;
-	bool is_2g = central_ch <= 14;
-
-	/* Path A */
-	if (is_2g)
-		rtw89_phy_write32_idx(rtwdev, R_PATH0_BAND_SEL_V1,
-				      B_PATH0_BAND_SEL_MSK_V1, 1, phy_idx);
-	else
-		rtw89_phy_write32_idx(rtwdev, R_PATH0_BAND_SEL_V1,
-				      B_PATH0_BAND_SEL_MSK_V1, 0, phy_idx);
-
-	/* Path B */
-	if (is_2g)
-		rtw89_phy_write32_idx(rtwdev, R_PATH1_BAND_SEL_V1,
-				      B_PATH1_BAND_SEL_MSK_V1, 1, phy_idx);
-	else
-		rtw89_phy_write32_idx(rtwdev, R_PATH1_BAND_SEL_V1,
-				      B_PATH1_BAND_SEL_MSK_V1, 0, phy_idx);
-
-	/* SCO compensate FC setting */
-	sco_comp = rtw8852b_sco_mapping(central_ch);
-	rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_INV, sco_comp, phy_idx);
-
-	if (chan->band_type == RTW89_BAND_6G)
-		return;
-
-	/* CCK parameters */
-	if (central_ch == 14) {
-		rtw89_phy_write32_mask(rtwdev, R_TXFIR0, B_TXFIR_C01, 0x3b13ff);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIR2, B_TXFIR_C23, 0x1c42de);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIR4, B_TXFIR_C45, 0xfdb0ad);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIR6, B_TXFIR_C67, 0xf60f6e);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIR8, B_TXFIR_C89, 0xfd8f92);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIRA, B_TXFIR_CAB, 0x2d011);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIRC, B_TXFIR_CCD, 0x1c02c);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIRE, B_TXFIR_CEF, 0xfff00a);
-	} else {
-		rtw89_phy_write32_mask(rtwdev, R_TXFIR0, B_TXFIR_C01, 0x3d23ff);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIR2, B_TXFIR_C23, 0x29b354);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIR4, B_TXFIR_C45, 0xfc1c8);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIR6, B_TXFIR_C67, 0xfdb053);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIR8, B_TXFIR_C89, 0xf86f9a);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIRA, B_TXFIR_CAB, 0xfaef92);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIRC, B_TXFIR_CCD, 0xfe5fcc);
-		rtw89_phy_write32_mask(rtwdev, R_TXFIRE, B_TXFIR_CEF, 0xffdff5);
-	}
-
-	rtw8852b_set_gain_error(rtwdev, subband, RF_PATH_A);
-	rtw8852b_set_gain_error(rtwdev, subband, RF_PATH_B);
-	rtw8852b_set_gain_offset(rtwdev, subband, phy_idx);
-	rtw8852b_set_rxsc_rpl_comp(rtwdev, subband);
-}
-
-static void rtw8852b_bw_setting(struct rtw89_dev *rtwdev, u8 bw, u8 path)
-{
-	static const u32 adc_sel[2] = {0xC0EC, 0xC1EC};
-	static const u32 wbadc_sel[2] = {0xC0E4, 0xC1E4};
-
-	switch (bw) {
-	case RTW89_CHANNEL_WIDTH_5:
-		rtw89_phy_write32_mask(rtwdev, adc_sel[path], 0x6000, 0x1);
-		rtw89_phy_write32_mask(rtwdev, wbadc_sel[path], 0x30, 0x0);
-		break;
-	case RTW89_CHANNEL_WIDTH_10:
-		rtw89_phy_write32_mask(rtwdev, adc_sel[path], 0x6000, 0x2);
-		rtw89_phy_write32_mask(rtwdev, wbadc_sel[path], 0x30, 0x1);
-		break;
-	case RTW89_CHANNEL_WIDTH_20:
-		rtw89_phy_write32_mask(rtwdev, adc_sel[path], 0x6000, 0x0);
-		rtw89_phy_write32_mask(rtwdev, wbadc_sel[path], 0x30, 0x2);
-		break;
-	case RTW89_CHANNEL_WIDTH_40:
-		rtw89_phy_write32_mask(rtwdev, adc_sel[path], 0x6000, 0x0);
-		rtw89_phy_write32_mask(rtwdev, wbadc_sel[path], 0x30, 0x2);
-		break;
-	case RTW89_CHANNEL_WIDTH_80:
-		rtw89_phy_write32_mask(rtwdev, adc_sel[path], 0x6000, 0x0);
-		rtw89_phy_write32_mask(rtwdev, wbadc_sel[path], 0x30, 0x2);
-		break;
-	default:
-		rtw89_warn(rtwdev, "Fail to set ADC\n");
-	}
-}
-
-static void rtw8852b_ctrl_bw(struct rtw89_dev *rtwdev, u8 pri_ch, u8 bw,
-			     enum rtw89_phy_idx phy_idx)
-{
-	u32 rx_path_0;
-
-	switch (bw) {
-	case RTW89_CHANNEL_WIDTH_5:
-		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x0, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_SBW, 0x1, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_PRICH, 0x0, phy_idx);
-
-		/*Set RF mode at 3 */
-		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
-				      B_P0_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
-				      B_P1_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
-		break;
-	case RTW89_CHANNEL_WIDTH_10:
-		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x0, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_SBW, 0x2, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_PRICH, 0x0, phy_idx);
-
-		/*Set RF mode at 3 */
-		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
-				      B_P0_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
-				      B_P1_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
-		break;
-	case RTW89_CHANNEL_WIDTH_20:
-		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x0, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_SBW, 0x0, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_PRICH, 0x0, phy_idx);
-
-		/*Set RF mode at 3 */
-		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
-				      B_P0_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
-				      B_P1_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
-		break;
-	case RTW89_CHANNEL_WIDTH_40:
-		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x1, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_SBW, 0x0, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_PRICH,
-				      pri_ch, phy_idx);
-
-		/*Set RF mode at 3 */
-		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
-				      B_P0_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
-				      B_P1_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
-		/*CCK primary channel */
-		if (pri_ch == RTW89_SC_20_UPPER)
-			rtw89_phy_write32_mask(rtwdev, R_RXSC, B_RXSC_EN, 1);
-		else
-			rtw89_phy_write32_mask(rtwdev, R_RXSC, B_RXSC_EN, 0);
-
-		break;
-	case RTW89_CHANNEL_WIDTH_80:
-		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x2, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_SBW, 0x0, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_PRICH,
-				      pri_ch, phy_idx);
-
-		/*Set RF mode at A */
-		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
-				      B_P0_RFMODE_ORI_RX_ALL, 0xaaa, phy_idx);
-		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
-				      B_P1_RFMODE_ORI_RX_ALL, 0xaaa, phy_idx);
-		break;
-	default:
-		rtw89_warn(rtwdev, "Fail to switch bw (bw:%d, pri ch:%d)\n", bw,
-			   pri_ch);
-	}
-
-	rtw8852b_bw_setting(rtwdev, bw, RF_PATH_A);
-	rtw8852b_bw_setting(rtwdev, bw, RF_PATH_B);
-
-	rx_path_0 = rtw89_phy_read32_idx(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0,
-					 phy_idx);
-	if (rx_path_0 == 0x1)
-		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
-				      B_P1_RFMODE_ORI_RX_ALL, 0x111, phy_idx);
-	else if (rx_path_0 == 0x2)
-		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
-				      B_P0_RFMODE_ORI_RX_ALL, 0x111, phy_idx);
-}
-
-static void rtw8852b_ctrl_cck_en(struct rtw89_dev *rtwdev, bool cck_en)
-{
-	if (cck_en) {
-		rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC, B_ENABLE_CCK, 1);
-		rtw89_phy_write32_mask(rtwdev, R_RXCCA, B_RXCCA_DIS, 0);
-	} else {
-		rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC, B_ENABLE_CCK, 0);
-		rtw89_phy_write32_mask(rtwdev, R_RXCCA, B_RXCCA_DIS, 1);
-	}
-}
-
-static void rtw8852b_5m_mask(struct rtw89_dev *rtwdev, const struct rtw89_chan *chan,
-			     enum rtw89_phy_idx phy_idx)
-{
-	u8 pri_ch = chan->pri_ch_idx;
-	bool mask_5m_low;
-	bool mask_5m_en;
-
-	switch (chan->band_width) {
-	case RTW89_CHANNEL_WIDTH_40:
-		/* Prich=1: Mask 5M High, Prich=2: Mask 5M Low */
-		mask_5m_en = true;
-		mask_5m_low = pri_ch == RTW89_SC_20_LOWER;
-		break;
-	case RTW89_CHANNEL_WIDTH_80:
-		/* Prich=3: Mask 5M High, Prich=4: Mask 5M Low, Else: Disable */
-		mask_5m_en = pri_ch == RTW89_SC_20_UPMOST ||
-			     pri_ch == RTW89_SC_20_LOWEST;
-		mask_5m_low = pri_ch == RTW89_SC_20_LOWEST;
-		break;
-	default:
-		mask_5m_en = false;
-		break;
-	}
-
-	if (!mask_5m_en) {
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_EN, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_EN, 0x0);
-		rtw89_phy_write32_idx(rtwdev, R_ASSIGN_SBD_OPT_V1,
-				      B_ASSIGN_SBD_OPT_EN_V1, 0x0, phy_idx);
-		return;
-	}
-
-	if (mask_5m_low) {
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_TH, 0x4);
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_EN, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_SB2, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_SB0, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_TH, 0x4);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_EN, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_SB2, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_SB0, 0x1);
-	} else {
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_TH, 0x4);
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_EN, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_SB2, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_SB0, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_TH, 0x4);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_EN, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_SB2, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_SB0, 0x0);
-	}
-	rtw89_phy_write32_idx(rtwdev, R_ASSIGN_SBD_OPT_V1,
-			      B_ASSIGN_SBD_OPT_EN_V1, 0x1, phy_idx);
-}
-
-static void rtw8852b_bb_reset_all(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
-{
-	rtw89_phy_write32_idx(rtwdev, R_S0_HW_SI_DIS, B_S0_HW_SI_DIS_W_R_TRIG, 0x7, phy_idx);
-	rtw89_phy_write32_idx(rtwdev, R_S1_HW_SI_DIS, B_S1_HW_SI_DIS_W_R_TRIG, 0x7, phy_idx);
-	fsleep(1);
-	rtw89_phy_write32_idx(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_ALL, 1, phy_idx);
-	rtw89_phy_write32_idx(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_ALL, 0, phy_idx);
-	rtw89_phy_write32_idx(rtwdev, R_S0_HW_SI_DIS, B_S0_HW_SI_DIS_W_R_TRIG, 0x0, phy_idx);
-	rtw89_phy_write32_idx(rtwdev, R_S1_HW_SI_DIS, B_S1_HW_SI_DIS_W_R_TRIG, 0x0, phy_idx);
-	rtw89_phy_write32_idx(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_ALL, 1, phy_idx);
-}
-
 static void rtw8852b_bb_reset_en(struct rtw89_dev *rtwdev, enum rtw89_band band,
 				 enum rtw89_phy_idx phy_idx, bool en)
 {
@@ -1422,87 +462,20 @@ static void rtw8852b_bb_reset(struct rtw89_dev *rtwdev,
 	rtw89_phy_write32_set(rtwdev, R_P0_TSSI_TRK, B_P0_TSSI_TRK_EN);
 	rtw89_phy_write32_set(rtwdev, R_P1_TXPW_RSTB, B_P1_TXPW_RSTB_MANON);
 	rtw89_phy_write32_set(rtwdev, R_P1_TSSI_TRK, B_P1_TSSI_TRK_EN);
-	rtw8852b_bb_reset_all(rtwdev, phy_idx);
+	rtw8852bx_bb_reset_all(rtwdev, phy_idx);
 	rtw89_phy_write32_clr(rtwdev, R_P0_TXPW_RSTB, B_P0_TXPW_RSTB_MANON);
 	rtw89_phy_write32_clr(rtwdev, R_P0_TSSI_TRK, B_P0_TSSI_TRK_EN);
 	rtw89_phy_write32_clr(rtwdev, R_P1_TXPW_RSTB, B_P1_TXPW_RSTB_MANON);
 	rtw89_phy_write32_clr(rtwdev, R_P1_TSSI_TRK, B_P1_TSSI_TRK_EN);
 }
 
-static void rtw8852b_bb_macid_ctrl_init(struct rtw89_dev *rtwdev,
-					enum rtw89_phy_idx phy_idx)
-{
-	u32 addr;
-
-	for (addr = R_AX_PWR_MACID_LMT_TABLE0;
-	     addr <= R_AX_PWR_MACID_LMT_TABLE127; addr += 4)
-		rtw89_mac_txpwr_write32(rtwdev, phy_idx, addr, 0);
-}
-
-static void rtw8852b_bb_sethw(struct rtw89_dev *rtwdev)
-{
-	struct rtw89_phy_efuse_gain *gain = &rtwdev->efuse_gain;
-
-	rtw89_phy_write32_clr(rtwdev, R_P0_EN_SOUND_WO_NDP, B_P0_EN_SOUND_WO_NDP);
-	rtw89_phy_write32_clr(rtwdev, R_P1_EN_SOUND_WO_NDP, B_P1_EN_SOUND_WO_NDP);
-
-	rtw8852b_bb_macid_ctrl_init(rtwdev, RTW89_PHY_0);
-
-	/* read these registers after loading BB parameters */
-	gain->offset_base[RTW89_PHY_0] =
-		rtw89_phy_read32_mask(rtwdev, R_P0_RPL1, B_P0_RPL1_BIAS_MASK);
-	gain->rssi_base[RTW89_PHY_0] =
-		rtw89_phy_read32_mask(rtwdev, R_P1_RPL1, B_P0_RPL1_BIAS_MASK);
-}
-
-static void rtw8852b_bb_set_pop(struct rtw89_dev *rtwdev)
-{
-	if (rtwdev->hw->conf.flags & IEEE80211_CONF_MONITOR)
-		rtw89_phy_write32_clr(rtwdev, R_PKT_CTRL, B_PKT_POP_EN);
-}
-
-static void rtw8852b_set_channel_bb(struct rtw89_dev *rtwdev, const struct rtw89_chan *chan,
-				    enum rtw89_phy_idx phy_idx)
-{
-	bool cck_en = chan->channel <= 14;
-	u8 pri_ch_idx = chan->pri_ch_idx;
-	u8 band = chan->band_type, chan_idx;
-
-	if (cck_en)
-		rtw8852b_ctrl_sco_cck(rtwdev,  chan->primary_channel);
-
-	rtw8852b_ctrl_ch(rtwdev, chan, phy_idx);
-	rtw8852b_ctrl_bw(rtwdev, pri_ch_idx, chan->band_width, phy_idx);
-	rtw8852b_ctrl_cck_en(rtwdev, cck_en);
-	if (chan->band_type == RTW89_BAND_5G) {
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_BT_SHARE_V1,
-				       B_PATH0_BT_SHARE_V1, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_BTG_PATH_V1,
-				       B_PATH0_BTG_PATH_V1, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_BT_SHARE_V1,
-				       B_PATH1_BT_SHARE_V1, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_BTG_PATH_V1,
-				       B_PATH1_BTG_PATH_V1, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_BT_SHARE, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_BT_SEG0, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_BT_DYN_DC_EST_EN_V1,
-				       B_BT_DYN_DC_EST_EN_MSK, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_GNT_BT_WGT_EN, B_GNT_BT_WGT_EN, 0x0);
-	}
-	chan_idx = rtw89_encode_chan_idx(rtwdev, chan->primary_channel, band);
-	rtw89_phy_write32_mask(rtwdev, R_MAC_PIN_SEL, B_CH_IDX_SEG0, chan_idx);
-	rtw8852b_5m_mask(rtwdev, chan, phy_idx);
-	rtw8852b_bb_set_pop(rtwdev);
-	rtw8852b_bb_reset_all(rtwdev, phy_idx);
-}
-
 static void rtw8852b_set_channel(struct rtw89_dev *rtwdev,
 				 const struct rtw89_chan *chan,
 				 enum rtw89_mac_idx mac_idx,
 				 enum rtw89_phy_idx phy_idx)
 {
-	rtw8852b_set_channel_mac(rtwdev, chan, mac_idx);
-	rtw8852b_set_channel_bb(rtwdev, chan, phy_idx);
+	rtw8852bx_set_channel_mac(rtwdev, chan, mac_idx);
+	rtw8852bx_set_channel_bb(rtwdev, chan, phy_idx);
 	rtw8852b_set_channel_rf(rtwdev, chan, phy_idx);
 }
 
@@ -1602,540 +575,6 @@ static void rtw8852b_rfk_track(struct rtw89_dev *rtwdev)
 	rtw8852b_dpk_track(rtwdev);
 }
 
-static u32 rtw8852b_bb_cal_txpwr_ref(struct rtw89_dev *rtwdev,
-				     enum rtw89_phy_idx phy_idx, s16 ref)
-{
-	const u16 tssi_16dbm_cw = 0x12c;
-	const u8 base_cw_0db = 0x27;
-	const s8 ofst_int = 0;
-	s16 pwr_s10_3;
-	s16 rf_pwr_cw;
-	u16 bb_pwr_cw;
-	u32 pwr_cw;
-	u32 tssi_ofst_cw;
-
-	pwr_s10_3 = (ref << 1) + (s16)(ofst_int) + (s16)(base_cw_0db << 3);
-	bb_pwr_cw = FIELD_GET(GENMASK(2, 0), pwr_s10_3);
-	rf_pwr_cw = FIELD_GET(GENMASK(8, 3), pwr_s10_3);
-	rf_pwr_cw = clamp_t(s16, rf_pwr_cw, 15, 63);
-	pwr_cw = (rf_pwr_cw << 3) | bb_pwr_cw;
-
-	tssi_ofst_cw = (u32)((s16)tssi_16dbm_cw + (ref << 1) - (16 << 3));
-	rtw89_debug(rtwdev, RTW89_DBG_TXPWR,
-		    "[TXPWR] tssi_ofst_cw=%d rf_cw=0x%x bb_cw=0x%x\n",
-		    tssi_ofst_cw, rf_pwr_cw, bb_pwr_cw);
-
-	return FIELD_PREP(B_DPD_TSSI_CW, tssi_ofst_cw) |
-	       FIELD_PREP(B_DPD_PWR_CW, pwr_cw) |
-	       FIELD_PREP(B_DPD_REF, ref);
-}
-
-static void rtw8852b_set_txpwr_ref(struct rtw89_dev *rtwdev,
-				   enum rtw89_phy_idx phy_idx)
-{
-	static const u32 addr[RF_PATH_NUM_8852B] = {0x5800, 0x7800};
-	const u32 mask = B_DPD_TSSI_CW | B_DPD_PWR_CW | B_DPD_REF;
-	const u8 ofst_ofdm = 0x4;
-	const u8 ofst_cck = 0x8;
-	const s16 ref_ofdm = 0;
-	const s16 ref_cck = 0;
-	u32 val;
-	u8 i;
-
-	rtw89_debug(rtwdev, RTW89_DBG_TXPWR, "[TXPWR] set txpwr reference\n");
-
-	rtw89_mac_txpwr_write32_mask(rtwdev, phy_idx, R_AX_PWR_RATE_CTRL,
-				     B_AX_PWR_REF, 0x0);
-
-	rtw89_debug(rtwdev, RTW89_DBG_TXPWR, "[TXPWR] set bb ofdm txpwr ref\n");
-	val = rtw8852b_bb_cal_txpwr_ref(rtwdev, phy_idx, ref_ofdm);
-
-	for (i = 0; i < RF_PATH_NUM_8852B; i++)
-		rtw89_phy_write32_idx(rtwdev, addr[i] + ofst_ofdm, mask, val,
-				      phy_idx);
-
-	rtw89_debug(rtwdev, RTW89_DBG_TXPWR, "[TXPWR] set bb cck txpwr ref\n");
-	val = rtw8852b_bb_cal_txpwr_ref(rtwdev, phy_idx, ref_cck);
-
-	for (i = 0; i < RF_PATH_NUM_8852B; i++)
-		rtw89_phy_write32_idx(rtwdev, addr[i] + ofst_cck, mask, val,
-				      phy_idx);
-}
-
-static void rtw8852b_bb_set_tx_shape_dfir(struct rtw89_dev *rtwdev,
-					  const struct rtw89_chan *chan,
-					  u8 tx_shape_idx,
-					  enum rtw89_phy_idx phy_idx)
-{
-#define __DFIR_CFG_ADDR(i) (R_TXFIR0 + ((i) << 2))
-#define __DFIR_CFG_MASK 0xffffffff
-#define __DFIR_CFG_NR 8
-#define __DECL_DFIR_PARAM(_name, _val...) \
-	static const u32 param_ ## _name[] = {_val}; \
-	static_assert(ARRAY_SIZE(param_ ## _name) == __DFIR_CFG_NR)
-
-	__DECL_DFIR_PARAM(flat,
-			  0x023D23FF, 0x0029B354, 0x000FC1C8, 0x00FDB053,
-			  0x00F86F9A, 0x06FAEF92, 0x00FE5FCC, 0x00FFDFF5);
-	__DECL_DFIR_PARAM(sharp,
-			  0x023D83FF, 0x002C636A, 0x0013F204, 0x00008090,
-			  0x00F87FB0, 0x06F99F83, 0x00FDBFBA, 0x00003FF5);
-	__DECL_DFIR_PARAM(sharp_14,
-			  0x023B13FF, 0x001C42DE, 0x00FDB0AD, 0x00F60F6E,
-			  0x00FD8F92, 0x0602D011, 0x0001C02C, 0x00FFF00A);
-	u8 ch = chan->channel;
-	const u32 *param;
-	u32 addr;
-	int i;
-
-	if (ch > 14) {
-		rtw89_warn(rtwdev,
-			   "set tx shape dfir by unknown ch: %d on 2G\n", ch);
-		return;
-	}
-
-	if (ch == 14)
-		param = param_sharp_14;
-	else
-		param = tx_shape_idx == 0 ? param_flat : param_sharp;
-
-	for (i = 0; i < __DFIR_CFG_NR; i++) {
-		addr = __DFIR_CFG_ADDR(i);
-		rtw89_debug(rtwdev, RTW89_DBG_TXPWR,
-			    "set tx shape dfir: 0x%x: 0x%x\n", addr, param[i]);
-		rtw89_phy_write32_idx(rtwdev, addr, __DFIR_CFG_MASK, param[i],
-				      phy_idx);
-	}
-
-#undef __DECL_DFIR_PARAM
-#undef __DFIR_CFG_NR
-#undef __DFIR_CFG_MASK
-#undef __DECL_CFG_ADDR
-}
-
-static void rtw8852b_set_tx_shape(struct rtw89_dev *rtwdev,
-				  const struct rtw89_chan *chan,
-				  enum rtw89_phy_idx phy_idx)
-{
-	const struct rtw89_rfe_parms *rfe_parms = rtwdev->rfe_parms;
-	u8 band = chan->band_type;
-	u8 regd = rtw89_regd_get(rtwdev, band);
-	u8 tx_shape_cck = (*rfe_parms->tx_shape.lmt)[band][RTW89_RS_CCK][regd];
-	u8 tx_shape_ofdm = (*rfe_parms->tx_shape.lmt)[band][RTW89_RS_OFDM][regd];
-
-	if (band == RTW89_BAND_2G)
-		rtw8852b_bb_set_tx_shape_dfir(rtwdev, chan, tx_shape_cck, phy_idx);
-
-	rtw89_phy_write32_mask(rtwdev, R_DCFO_OPT, B_TXSHAPE_TRIANGULAR_CFG,
-			       tx_shape_ofdm);
-}
-
-static void rtw8852b_set_txpwr(struct rtw89_dev *rtwdev,
-			       const struct rtw89_chan *chan,
-			       enum rtw89_phy_idx phy_idx)
-{
-	rtw89_phy_set_txpwr_byrate(rtwdev, chan, phy_idx);
-	rtw89_phy_set_txpwr_offset(rtwdev, chan, phy_idx);
-	rtw8852b_set_tx_shape(rtwdev, chan, phy_idx);
-	rtw89_phy_set_txpwr_limit(rtwdev, chan, phy_idx);
-	rtw89_phy_set_txpwr_limit_ru(rtwdev, chan, phy_idx);
-}
-
-static void rtw8852b_set_txpwr_ctrl(struct rtw89_dev *rtwdev,
-				    enum rtw89_phy_idx phy_idx)
-{
-	rtw8852b_set_txpwr_ref(rtwdev, phy_idx);
-}
-
-static
-void rtw8852b_set_txpwr_ul_tb_offset(struct rtw89_dev *rtwdev,
-				     s8 pw_ofst, enum rtw89_mac_idx mac_idx)
-{
-	u32 reg;
-
-	if (pw_ofst < -16 || pw_ofst > 15) {
-		rtw89_warn(rtwdev, "[ULTB] Err pwr_offset=%d\n", pw_ofst);
-		return;
-	}
-
-	reg = rtw89_mac_reg_by_idx(rtwdev, R_AX_PWR_UL_TB_CTRL, mac_idx);
-	rtw89_write32_set(rtwdev, reg, B_AX_PWR_UL_TB_CTRL_EN);
-
-	reg = rtw89_mac_reg_by_idx(rtwdev, R_AX_PWR_UL_TB_1T, mac_idx);
-	rtw89_write32_mask(rtwdev, reg, B_AX_PWR_UL_TB_1T_MASK, pw_ofst);
-
-	pw_ofst = max_t(s8, pw_ofst - 3, -16);
-	reg = rtw89_mac_reg_by_idx(rtwdev, R_AX_PWR_UL_TB_2T, mac_idx);
-	rtw89_write32_mask(rtwdev, reg, B_AX_PWR_UL_TB_2T_MASK, pw_ofst);
-}
-
-static int
-rtw8852b_init_txpwr_unit(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
-{
-	int ret;
-
-	ret = rtw89_mac_txpwr_write32(rtwdev, phy_idx, R_AX_PWR_UL_CTRL2, 0x07763333);
-	if (ret)
-		return ret;
-
-	ret = rtw89_mac_txpwr_write32(rtwdev, phy_idx, R_AX_PWR_COEXT_CTRL, 0x01ebf000);
-	if (ret)
-		return ret;
-
-	ret = rtw89_mac_txpwr_write32(rtwdev, phy_idx, R_AX_PWR_UL_CTRL0, 0x0002f8ff);
-	if (ret)
-		return ret;
-
-	rtw8852b_set_txpwr_ul_tb_offset(rtwdev, 0, phy_idx == RTW89_PHY_1 ?
-						   RTW89_MAC_1 : RTW89_MAC_0);
-
-	return 0;
-}
-
-void rtw8852b_bb_set_plcp_tx(struct rtw89_dev *rtwdev)
-{
-	const struct rtw89_reg3_def *def = rtw8852b_pmac_ht20_mcs7_tbl;
-	u8 i;
-
-	for (i = 0; i < ARRAY_SIZE(rtw8852b_pmac_ht20_mcs7_tbl); i++, def++)
-		rtw89_phy_write32_mask(rtwdev, def->addr, def->mask, def->data);
-}
-
-static void rtw8852b_stop_pmac_tx(struct rtw89_dev *rtwdev,
-				  struct rtw8852b_bb_pmac_info *tx_info,
-				  enum rtw89_phy_idx idx)
-{
-	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC Stop Tx");
-	if (tx_info->mode == CONT_TX)
-		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_PRD, B_PMAC_CTX_EN, 0, idx);
-	else if (tx_info->mode == PKTS_TX)
-		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_PRD, B_PMAC_PTX_EN, 0, idx);
-}
-
-static void rtw8852b_start_pmac_tx(struct rtw89_dev *rtwdev,
-				   struct rtw8852b_bb_pmac_info *tx_info,
-				   enum rtw89_phy_idx idx)
-{
-	enum rtw8852b_pmac_mode mode = tx_info->mode;
-	u32 pkt_cnt = tx_info->tx_cnt;
-	u16 period = tx_info->period;
-
-	if (mode == CONT_TX && !tx_info->is_cck) {
-		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_PRD, B_PMAC_CTX_EN, 1, idx);
-		rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC CTx Start");
-	} else if (mode == PKTS_TX) {
-		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_PRD, B_PMAC_PTX_EN, 1, idx);
-		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_PRD,
-				      B_PMAC_TX_PRD_MSK, period, idx);
-		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_CNT, B_PMAC_TX_CNT_MSK,
-				      pkt_cnt, idx);
-		rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC PTx Start");
-	}
-
-	rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_CTRL, B_PMAC_TXEN_DIS, 1, idx);
-	rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_CTRL, B_PMAC_TXEN_DIS, 0, idx);
-}
-
-void rtw8852b_bb_set_pmac_tx(struct rtw89_dev *rtwdev,
-			     struct rtw8852b_bb_pmac_info *tx_info,
-			     enum rtw89_phy_idx idx)
-{
-	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
-
-	if (!tx_info->en_pmac_tx) {
-		rtw8852b_stop_pmac_tx(rtwdev, tx_info, idx);
-		rtw89_phy_write32_idx(rtwdev, R_PD_CTRL, B_PD_HIT_DIS, 0, idx);
-		if (chan->band_type == RTW89_BAND_2G)
-			rtw89_phy_write32_clr(rtwdev, R_RXCCA, B_RXCCA_DIS);
-		return;
-	}
-
-	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC Tx Enable");
-
-	rtw89_phy_write32_idx(rtwdev, R_PMAC_GNT, B_PMAC_GNT_TXEN, 1, idx);
-	rtw89_phy_write32_idx(rtwdev, R_PMAC_GNT, B_PMAC_GNT_RXEN, 1, idx);
-	rtw89_phy_write32_idx(rtwdev, R_PMAC_RX_CFG1, B_PMAC_OPT1_MSK, 0x3f, idx);
-	rtw89_phy_write32_idx(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_ALL, 0, idx);
-	rtw89_phy_write32_idx(rtwdev, R_PD_CTRL, B_PD_HIT_DIS, 1, idx);
-	rtw89_phy_write32_set(rtwdev, R_RXCCA, B_RXCCA_DIS);
-	rtw89_phy_write32_idx(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_ALL, 1, idx);
-
-	rtw8852b_start_pmac_tx(rtwdev, tx_info, idx);
-}
-
-void rtw8852b_bb_set_pmac_pkt_tx(struct rtw89_dev *rtwdev, u8 enable,
-				 u16 tx_cnt, u16 period, u16 tx_time,
-				 enum rtw89_phy_idx idx)
-{
-	struct rtw8852b_bb_pmac_info tx_info = {0};
-
-	tx_info.en_pmac_tx = enable;
-	tx_info.is_cck = 0;
-	tx_info.mode = PKTS_TX;
-	tx_info.tx_cnt = tx_cnt;
-	tx_info.period = period;
-	tx_info.tx_time = tx_time;
-
-	rtw8852b_bb_set_pmac_tx(rtwdev, &tx_info, idx);
-}
-
-void rtw8852b_bb_set_power(struct rtw89_dev *rtwdev, s16 pwr_dbm,
-			   enum rtw89_phy_idx idx)
-{
-	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC CFG Tx PWR = %d", pwr_dbm);
-
-	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_PWR_EN, 1, idx);
-	rtw89_phy_write32_idx(rtwdev, R_TXPWR, B_TXPWR_MSK, pwr_dbm, idx);
-}
-
-void rtw8852b_bb_cfg_tx_path(struct rtw89_dev *rtwdev, u8 tx_path)
-{
-	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_MOD, 7, RTW89_PHY_0);
-
-	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC CFG Tx Path = %d", tx_path);
-
-	if (tx_path == RF_PATH_A) {
-		rtw89_phy_write32_mask(rtwdev, R_TXPATH_SEL, B_TXPATH_SEL_MSK, 1);
-		rtw89_phy_write32_mask(rtwdev, R_TXNSS_MAP, B_TXNSS_MAP_MSK, 0);
-	} else if (tx_path == RF_PATH_B) {
-		rtw89_phy_write32_mask(rtwdev, R_TXPATH_SEL, B_TXPATH_SEL_MSK, 2);
-		rtw89_phy_write32_mask(rtwdev, R_TXNSS_MAP, B_TXNSS_MAP_MSK, 0);
-	} else if (tx_path == RF_PATH_AB) {
-		rtw89_phy_write32_mask(rtwdev, R_TXPATH_SEL, B_TXPATH_SEL_MSK, 3);
-		rtw89_phy_write32_mask(rtwdev, R_TXNSS_MAP, B_TXNSS_MAP_MSK, 4);
-	} else {
-		rtw89_debug(rtwdev, RTW89_DBG_TSSI, "Error Tx Path");
-	}
-}
-
-void rtw8852b_bb_tx_mode_switch(struct rtw89_dev *rtwdev,
-				enum rtw89_phy_idx idx, u8 mode)
-{
-	if (mode != 0)
-		return;
-
-	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "Tx mode switch");
-
-	rtw89_phy_write32_idx(rtwdev, R_PMAC_GNT, B_PMAC_GNT_TXEN, 0, idx);
-	rtw89_phy_write32_idx(rtwdev, R_PMAC_GNT, B_PMAC_GNT_RXEN, 0, idx);
-	rtw89_phy_write32_idx(rtwdev, R_PMAC_RX_CFG1, B_PMAC_OPT1_MSK, 0, idx);
-	rtw89_phy_write32_idx(rtwdev, R_PMAC_RXMOD, B_PMAC_RXMOD_MSK, 0, idx);
-	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_DPD_EN, 0, idx);
-	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_MOD, 0, idx);
-	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_PWR_EN, 0, idx);
-}
-
-void rtw8852b_bb_backup_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx idx,
-			     struct rtw8852b_bb_tssi_bak *bak)
-{
-	s32 tmp;
-
-	bak->tx_path = rtw89_phy_read32_idx(rtwdev, R_TXPATH_SEL, B_TXPATH_SEL_MSK, idx);
-	bak->rx_path = rtw89_phy_read32_idx(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0, idx);
-	bak->p0_rfmode = rtw89_phy_read32_idx(rtwdev, R_P0_RFMODE, MASKDWORD, idx);
-	bak->p0_rfmode_ftm = rtw89_phy_read32_idx(rtwdev, R_P0_RFMODE_FTM_RX, MASKDWORD, idx);
-	bak->p1_rfmode = rtw89_phy_read32_idx(rtwdev, R_P1_RFMODE, MASKDWORD, idx);
-	bak->p1_rfmode_ftm = rtw89_phy_read32_idx(rtwdev, R_P1_RFMODE_FTM_RX, MASKDWORD, idx);
-	tmp = rtw89_phy_read32_idx(rtwdev, R_TXPWR, B_TXPWR_MSK, idx);
-	bak->tx_pwr = sign_extend32(tmp, 8);
-}
-
-void rtw8852b_bb_restore_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx idx,
-			      const struct rtw8852b_bb_tssi_bak *bak)
-{
-	rtw89_phy_write32_idx(rtwdev, R_TXPATH_SEL, B_TXPATH_SEL_MSK, bak->tx_path, idx);
-	if (bak->tx_path == RF_AB)
-		rtw89_phy_write32_mask(rtwdev, R_TXNSS_MAP, B_TXNSS_MAP_MSK, 0x4);
-	else
-		rtw89_phy_write32_mask(rtwdev, R_TXNSS_MAP, B_TXNSS_MAP_MSK, 0x0);
-	rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0, bak->rx_path, idx);
-	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_PWR_EN, 1, idx);
-	rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE, MASKDWORD, bak->p0_rfmode, idx);
-	rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_FTM_RX, MASKDWORD, bak->p0_rfmode_ftm, idx);
-	rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE, MASKDWORD, bak->p1_rfmode, idx);
-	rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_FTM_RX, MASKDWORD, bak->p1_rfmode_ftm, idx);
-	rtw89_phy_write32_idx(rtwdev, R_TXPWR, B_TXPWR_MSK, bak->tx_pwr, idx);
-}
-
-static void rtw8852b_ctrl_nbtg_bt_tx(struct rtw89_dev *rtwdev, bool en,
-				     enum rtw89_phy_idx phy_idx)
-{
-	rtw89_phy_write_reg3_tbl(rtwdev, en ? &rtw8852b_btc_preagc_en_defs_tbl :
-						 &rtw8852b_btc_preagc_dis_defs_tbl);
-}
-
-static void rtw8852b_ctrl_btg_bt_rx(struct rtw89_dev *rtwdev, bool en,
-				    enum rtw89_phy_idx phy_idx)
-{
-	if (en) {
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_BT_SHARE_V1,
-				       B_PATH0_BT_SHARE_V1, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_BTG_PATH_V1,
-				       B_PATH0_BTG_PATH_V1, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_G_LNA6_OP1DB_V1,
-				       B_PATH1_G_LNA6_OP1DB_V1, 0x20);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_G_TIA0_LNA6_OP1DB_V1,
-				       B_PATH1_G_TIA0_LNA6_OP1DB_V1, 0x30);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_BT_SHARE_V1,
-				       B_PATH1_BT_SHARE_V1, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_BTG_PATH_V1,
-				       B_PATH1_BTG_PATH_V1, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_PMAC_GNT, B_PMAC_GNT_P1, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_BT_SHARE, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_BT_SEG0, 0x2);
-		rtw89_phy_write32_mask(rtwdev, R_BT_DYN_DC_EST_EN_V1,
-				       B_BT_DYN_DC_EST_EN_MSK, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_GNT_BT_WGT_EN, B_GNT_BT_WGT_EN, 0x1);
-	} else {
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_BT_SHARE_V1,
-				       B_PATH0_BT_SHARE_V1, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PATH0_BTG_PATH_V1,
-				       B_PATH0_BTG_PATH_V1, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_G_LNA6_OP1DB_V1,
-				       B_PATH1_G_LNA6_OP1DB_V1, 0x1a);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_G_TIA0_LNA6_OP1DB_V1,
-				       B_PATH1_G_TIA0_LNA6_OP1DB_V1, 0x2a);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_BT_SHARE_V1,
-				       B_PATH1_BT_SHARE_V1, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PATH1_BTG_PATH_V1,
-				       B_PATH1_BTG_PATH_V1, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_PMAC_GNT, B_PMAC_GNT_P1, 0xc);
-		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_BT_SHARE, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_BT_SEG0, 0x0);
-		rtw89_phy_write32_mask(rtwdev, R_BT_DYN_DC_EST_EN_V1,
-				       B_BT_DYN_DC_EST_EN_MSK, 0x1);
-		rtw89_phy_write32_mask(rtwdev, R_GNT_BT_WGT_EN, B_GNT_BT_WGT_EN, 0x0);
-	}
-}
-
-void rtw8852b_bb_ctrl_rx_path(struct rtw89_dev *rtwdev,
-			      enum rtw89_rf_path_bit rx_path)
-{
-	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
-	u32 rst_mask0;
-	u32 rst_mask1;
-
-	if (rx_path == RF_A) {
-		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0, 1);
-		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG0, 1);
-		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG1, 1);
-		rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 0);
-		rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 0);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_USER_MAX, 4);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 0);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHETB_MAX_NSS, 0);
-	} else if (rx_path == RF_B) {
-		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0, 2);
-		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG0, 2);
-		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG1, 2);
-		rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 0);
-		rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 0);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_USER_MAX, 4);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 0);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHETB_MAX_NSS, 0);
-	} else if (rx_path == RF_AB) {
-		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0, 3);
-		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG0, 3);
-		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG1, 3);
-		rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 1);
-		rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 1);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_USER_MAX, 4);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 1);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHETB_MAX_NSS, 1);
-	}
-
-	rtw8852b_set_gain_offset(rtwdev, chan->subband_type, RTW89_PHY_0);
-
-	if (chan->band_type == RTW89_BAND_2G &&
-	    (rx_path == RF_B || rx_path == RF_AB))
-		rtw8852b_ctrl_btg_bt_rx(rtwdev, true, RTW89_PHY_0);
-	else
-		rtw8852b_ctrl_btg_bt_rx(rtwdev, false, RTW89_PHY_0);
-
-	rst_mask0 = B_P0_TXPW_RSTB_MANON | B_P0_TXPW_RSTB_TSSI;
-	rst_mask1 = B_P1_TXPW_RSTB_MANON | B_P1_TXPW_RSTB_TSSI;
-	if (rx_path == RF_A) {
-		rtw89_phy_write32_mask(rtwdev, R_P0_TXPW_RSTB, rst_mask0, 1);
-		rtw89_phy_write32_mask(rtwdev, R_P0_TXPW_RSTB, rst_mask0, 3);
-	} else {
-		rtw89_phy_write32_mask(rtwdev, R_P1_TXPW_RSTB, rst_mask1, 1);
-		rtw89_phy_write32_mask(rtwdev, R_P1_TXPW_RSTB, rst_mask1, 3);
-	}
-}
-
-static void rtw8852b_bb_ctrl_rf_mode_rx_path(struct rtw89_dev *rtwdev,
-					     enum rtw89_rf_path_bit rx_path)
-{
-	if (rx_path == RF_A) {
-		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE,
-				       B_P0_RFMODE_ORI_TXRX_FTM_TX, 0x1233312);
-		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE_FTM_RX,
-				       B_P0_RFMODE_FTM_RX, 0x333);
-		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE,
-				       B_P1_RFMODE_ORI_TXRX_FTM_TX, 0x1111111);
-		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE_FTM_RX,
-				       B_P1_RFMODE_FTM_RX, 0x111);
-	} else if (rx_path == RF_B) {
-		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE,
-				       B_P0_RFMODE_ORI_TXRX_FTM_TX, 0x1111111);
-		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE_FTM_RX,
-				       B_P0_RFMODE_FTM_RX, 0x111);
-		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE,
-				       B_P1_RFMODE_ORI_TXRX_FTM_TX, 0x1233312);
-		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE_FTM_RX,
-				       B_P1_RFMODE_FTM_RX, 0x333);
-	} else if (rx_path == RF_AB) {
-		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE,
-				       B_P0_RFMODE_ORI_TXRX_FTM_TX, 0x1233312);
-		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE_FTM_RX,
-				       B_P0_RFMODE_FTM_RX, 0x333);
-		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE,
-				       B_P1_RFMODE_ORI_TXRX_FTM_TX, 0x1233312);
-		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE_FTM_RX,
-				       B_P1_RFMODE_FTM_RX, 0x333);
-	}
-}
-
-static void rtw8852b_bb_cfg_txrx_path(struct rtw89_dev *rtwdev)
-{
-	struct rtw89_hal *hal = &rtwdev->hal;
-	enum rtw89_rf_path_bit rx_path = hal->antenna_rx ? hal->antenna_rx : RF_AB;
-
-	rtw8852b_bb_ctrl_rx_path(rtwdev, rx_path);
-	rtw8852b_bb_ctrl_rf_mode_rx_path(rtwdev, rx_path);
-
-	if (rtwdev->hal.rx_nss == 1) {
-		rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 0);
-		rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 0);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 0);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHETB_MAX_NSS, 0);
-	} else {
-		rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 1);
-		rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 1);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 1);
-		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHETB_MAX_NSS, 1);
-	}
-
-	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_MOD, 0x0, RTW89_PHY_0);
-}
-
-static u8 rtw8852b_get_thermal(struct rtw89_dev *rtwdev, enum rtw89_rf_path rf_path)
-{
-	if (rtwdev->is_tssi_mode[rf_path]) {
-		u32 addr = 0x1c10 + (rf_path << 13);
-
-		return rtw89_phy_read32_mask(rtwdev, addr, 0x3F000000);
-	}
-
-	rtw89_write_rf(rtwdev, rf_path, RR_TM, RR_TM_TRI, 0x1);
-	rtw89_write_rf(rtwdev, rf_path, RR_TM, RR_TM_TRI, 0x0);
-	rtw89_write_rf(rtwdev, rf_path, RR_TM, RR_TM_TRI, 0x1);
-
-	fsleep(200);
-
-	return rtw89_read_rf(rtwdev, rf_path, RR_TM, RR_TM_VAL);
-}
-
 static void rtw8852b_btc_set_rfe(struct rtw89_dev *rtwdev)
 {
 	const struct rtw89_btc_ver *ver = rtwdev->btc.ver;
@@ -2402,15 +841,15 @@ static void rtw8852b_btc_set_wl_rx_gain(struct rtw89_dev *rtwdev, u32 level)
 	switch (level) {
 	case 0: /* original */
 	default:
-		rtw8852b_ctrl_nbtg_bt_tx(rtwdev, false, RTW89_PHY_0);
+		rtw8852bx_ctrl_nbtg_bt_tx(rtwdev, false, RTW89_PHY_0);
 		btc->dm.wl_lna2 = 0;
 		break;
 	case 1: /* for FDD free-run */
-		rtw8852b_ctrl_nbtg_bt_tx(rtwdev, true, RTW89_PHY_0);
+		rtw8852bx_ctrl_nbtg_bt_tx(rtwdev, true, RTW89_PHY_0);
 		btc->dm.wl_lna2 = 0;
 		break;
 	case 2: /* for BTG Co-Rx*/
-		rtw8852b_ctrl_nbtg_bt_tx(rtwdev, false, RTW89_PHY_0);
+		rtw8852bx_ctrl_nbtg_bt_tx(rtwdev, false, RTW89_PHY_0);
 		btc->dm.wl_lna2 = 1;
 		break;
 	}
@@ -2418,105 +857,19 @@ static void rtw8852b_btc_set_wl_rx_gain(struct rtw89_dev *rtwdev, u32 level)
 	rtw8852b_btc_set_wl_lna2(rtwdev, btc->dm.wl_lna2);
 }
 
-static void rtw8852b_fill_freq_with_ppdu(struct rtw89_dev *rtwdev,
-					 struct rtw89_rx_phy_ppdu *phy_ppdu,
-					 struct ieee80211_rx_status *status)
-{
-	u16 chan = phy_ppdu->chan_idx;
-	enum nl80211_band band;
-	u8 ch;
-
-	if (chan == 0)
-		return;
-
-	rtw89_decode_chan_idx(rtwdev, chan, &ch, &band);
-	status->freq = ieee80211_channel_to_frequency(ch, band);
-	status->band = band;
-}
-
-static void rtw8852b_query_ppdu(struct rtw89_dev *rtwdev,
-				struct rtw89_rx_phy_ppdu *phy_ppdu,
-				struct ieee80211_rx_status *status)
-{
-	u8 path;
-	u8 *rx_power = phy_ppdu->rssi;
-
-	status->signal = RTW89_RSSI_RAW_TO_DBM(max(rx_power[RF_PATH_A], rx_power[RF_PATH_B]));
-	for (path = 0; path < rtwdev->chip->rf_path_num; path++) {
-		status->chains |= BIT(path);
-		status->chain_signal[path] = RTW89_RSSI_RAW_TO_DBM(rx_power[path]);
-	}
-	if (phy_ppdu->valid)
-		rtw8852b_fill_freq_with_ppdu(rtwdev, phy_ppdu, status);
-}
-
-static int rtw8852b_mac_enable_bb_rf(struct rtw89_dev *rtwdev)
-{
-	int ret;
-
-	rtw89_write8_set(rtwdev, R_AX_SYS_FUNC_EN,
-			 B_AX_FEN_BBRSTB | B_AX_FEN_BB_GLB_RSTN);
-	rtw89_write32_mask(rtwdev, R_AX_SPS_DIG_ON_CTRL0, B_AX_REG_ZCDC_H_MASK, 0x1);
-	rtw89_write32_set(rtwdev, R_AX_WLRF_CTRL, B_AX_AFC_AFEDIG);
-	rtw89_write32_clr(rtwdev, R_AX_WLRF_CTRL, B_AX_AFC_AFEDIG);
-	rtw89_write32_set(rtwdev, R_AX_WLRF_CTRL, B_AX_AFC_AFEDIG);
-
-	ret = rtw89_mac_write_xtal_si(rtwdev, XTAL_SI_WL_RFC_S0, 0xC7,
-				      FULL_BIT_MASK);
-	if (ret)
-		return ret;
-
-	ret = rtw89_mac_write_xtal_si(rtwdev, XTAL_SI_WL_RFC_S1, 0xC7,
-				      FULL_BIT_MASK);
-	if (ret)
-		return ret;
-
-	rtw89_write8(rtwdev, R_AX_PHYREG_SET, PHYREG_SET_XYN_CYCLE);
-
-	return 0;
-}
-
-static int rtw8852b_mac_disable_bb_rf(struct rtw89_dev *rtwdev)
-{
-	u8 wl_rfc_s0;
-	u8 wl_rfc_s1;
-	int ret;
-
-	rtw89_write32_clr(rtwdev, R_AX_WLRF_CTRL, B_AX_AFC_AFEDIG);
-	rtw89_write8_clr(rtwdev, R_AX_SYS_FUNC_EN,
-			 B_AX_FEN_BBRSTB | B_AX_FEN_BB_GLB_RSTN);
-
-	ret = rtw89_mac_read_xtal_si(rtwdev, XTAL_SI_WL_RFC_S0, &wl_rfc_s0);
-	if (ret)
-		return ret;
-	wl_rfc_s0 &= ~XTAL_SI_RF00S_EN;
-	ret = rtw89_mac_write_xtal_si(rtwdev, XTAL_SI_WL_RFC_S0, wl_rfc_s0,
-				      FULL_BIT_MASK);
-	if (ret)
-		return ret;
-
-	ret = rtw89_mac_read_xtal_si(rtwdev, XTAL_SI_WL_RFC_S1, &wl_rfc_s1);
-	if (ret)
-		return ret;
-	wl_rfc_s1 &= ~XTAL_SI_RF10S_EN;
-	ret = rtw89_mac_write_xtal_si(rtwdev, XTAL_SI_WL_RFC_S1, wl_rfc_s1,
-				      FULL_BIT_MASK);
-	return ret;
-}
-
 static const struct rtw89_chip_ops rtw8852b_chip_ops = {
-	.enable_bb_rf		= rtw8852b_mac_enable_bb_rf,
-	.disable_bb_rf		= rtw8852b_mac_disable_bb_rf,
+	.enable_bb_rf		= rtw8852bx_mac_enable_bb_rf,
+	.disable_bb_rf		= rtw8852bx_mac_disable_bb_rf,
 	.bb_preinit		= NULL,
 	.bb_postinit		= NULL,
 	.bb_reset		= rtw8852b_bb_reset,
-	.bb_sethw		= rtw8852b_bb_sethw,
+	.bb_sethw		= rtw8852bx_bb_sethw,
 	.read_rf		= rtw89_phy_read_rf_v1,
 	.write_rf		= rtw89_phy_write_rf_v1,
 	.set_channel		= rtw8852b_set_channel,
 	.set_channel_help	= rtw8852b_set_channel_help,
-	.read_efuse		= rtw8852b_read_efuse,
-	.read_phycap		= rtw8852b_read_phycap,
+	.read_efuse		= rtw8852bx_read_efuse,
+	.read_phycap		= rtw8852bx_read_phycap,
 	.fem_setup		= NULL,
 	.rfe_gpio		= NULL,
 	.rfk_hw_init		= NULL,
@@ -2526,16 +879,16 @@ static const struct rtw89_chip_ops rtw8852b_chip_ops = {
 	.rfk_band_changed	= rtw8852b_rfk_band_changed,
 	.rfk_scan		= rtw8852b_rfk_scan,
 	.rfk_track		= rtw8852b_rfk_track,
-	.power_trim		= rtw8852b_power_trim,
-	.set_txpwr		= rtw8852b_set_txpwr,
-	.set_txpwr_ctrl		= rtw8852b_set_txpwr_ctrl,
-	.init_txpwr_unit	= rtw8852b_init_txpwr_unit,
-	.get_thermal		= rtw8852b_get_thermal,
-	.ctrl_btg_bt_rx		= rtw8852b_ctrl_btg_bt_rx,
-	.query_ppdu		= rtw8852b_query_ppdu,
-	.ctrl_nbtg_bt_tx	= rtw8852b_ctrl_nbtg_bt_tx,
-	.cfg_txrx_path		= rtw8852b_bb_cfg_txrx_path,
-	.set_txpwr_ul_tb_offset	= rtw8852b_set_txpwr_ul_tb_offset,
+	.power_trim		= rtw8852bx_power_trim,
+	.set_txpwr		= rtw8852bx_set_txpwr,
+	.set_txpwr_ctrl		= rtw8852bx_set_txpwr_ctrl,
+	.init_txpwr_unit	= rtw8852bx_init_txpwr_unit,
+	.get_thermal		= rtw8852bx_get_thermal,
+	.ctrl_btg_bt_rx		= rtw8852bx_ctrl_btg_bt_rx,
+	.query_ppdu		= rtw8852bx_query_ppdu,
+	.ctrl_nbtg_bt_tx	= rtw8852bx_ctrl_nbtg_bt_tx,
+	.cfg_txrx_path		= rtw8852bx_bb_cfg_txrx_path,
+	.set_txpwr_ul_tb_offset	= rtw8852bx_set_txpwr_ul_tb_offset,
 	.pwr_on_func		= rtw8852b_pwr_on_func,
 	.pwr_off_func		= rtw8852b_pwr_off_func,
 	.query_rxdesc		= rtw89_core_query_rxdesc,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.h b/drivers/net/wireless/realtek/rtw89/rtw8852b.h
index 4f9b3d476879..5ec7180fd355 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.h
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.h
@@ -10,128 +10,6 @@
 #define RF_PATH_NUM_8852B 2
 #define BB_PATH_NUM_8852B 2
 
-enum rtw8852b_pmac_mode {
-	NONE_TEST,
-	PKTS_TX,
-	PKTS_RX,
-	CONT_TX
-};
-
-struct rtw8852b_u_efuse {
-	u8 rsvd[0x88];
-	u8 mac_addr[ETH_ALEN];
-};
-
-struct rtw8852b_e_efuse {
-	u8 mac_addr[ETH_ALEN];
-};
-
-struct rtw8852b_tssi_offset {
-	u8 cck_tssi[TSSI_CCK_CH_GROUP_NUM];
-	u8 bw40_tssi[TSSI_MCS_2G_CH_GROUP_NUM];
-	u8 rsvd[7];
-	u8 bw40_1s_tssi_5g[TSSI_MCS_5G_CH_GROUP_NUM];
-} __packed;
-
-struct rtw8852b_efuse {
-	u8 rsvd[0x210];
-	struct rtw8852b_tssi_offset path_a_tssi;
-	u8 rsvd1[10];
-	struct rtw8852b_tssi_offset path_b_tssi;
-	u8 rsvd2[94];
-	u8 channel_plan;
-	u8 xtal_k;
-	u8 rsvd3;
-	u8 iqk_lck;
-	u8 rsvd4[5];
-	u8 reg_setting:2;
-	u8 tx_diversity:1;
-	u8 rx_diversity:2;
-	u8 ac_mode:1;
-	u8 module_type:2;
-	u8 rsvd5;
-	u8 shared_ant:1;
-	u8 coex_type:3;
-	u8 ant_iso:1;
-	u8 radio_on_off:1;
-	u8 rsvd6:2;
-	u8 eeprom_version;
-	u8 customer_id;
-	u8 tx_bb_swing_2g;
-	u8 tx_bb_swing_5g;
-	u8 tx_cali_pwr_trk_mode;
-	u8 trx_path_selection;
-	u8 rfe_type;
-	u8 country_code[2];
-	u8 rsvd7[3];
-	u8 path_a_therm;
-	u8 path_b_therm;
-	u8 rsvd8[2];
-	u8 rx_gain_2g_ofdm;
-	u8 rsvd9;
-	u8 rx_gain_2g_cck;
-	u8 rsvd10;
-	u8 rx_gain_5g_low;
-	u8 rsvd11;
-	u8 rx_gain_5g_mid;
-	u8 rsvd12;
-	u8 rx_gain_5g_high;
-	u8 rsvd13[35];
-	u8 path_a_cck_pwr_idx[6];
-	u8 path_a_bw40_1tx_pwr_idx[5];
-	u8 path_a_ofdm_1tx_pwr_idx_diff:4;
-	u8 path_a_bw20_1tx_pwr_idx_diff:4;
-	u8 path_a_bw20_2tx_pwr_idx_diff:4;
-	u8 path_a_bw40_2tx_pwr_idx_diff:4;
-	u8 path_a_cck_2tx_pwr_idx_diff:4;
-	u8 path_a_ofdm_2tx_pwr_idx_diff:4;
-	u8 rsvd14[0xf2];
-	union {
-		struct rtw8852b_u_efuse u;
-		struct rtw8852b_e_efuse e;
-	};
-} __packed;
-
-struct rtw8852b_bb_pmac_info {
-	u8 en_pmac_tx:1;
-	u8 is_cck:1;
-	u8 mode:3;
-	u8 rsvd:3;
-	u16 tx_cnt;
-	u16 period;
-	u16 tx_time;
-	u8 duty_cycle;
-};
-
-struct rtw8852b_bb_tssi_bak {
-	u8 tx_path;
-	u8 rx_path;
-	u32 p0_rfmode;
-	u32 p0_rfmode_ftm;
-	u32 p1_rfmode;
-	u32 p1_rfmode_ftm;
-	s16 tx_pwr; /* S9 */
-};
-
 extern const struct rtw89_chip_info rtw8852b_chip_info;
 
-void rtw8852b_bb_set_plcp_tx(struct rtw89_dev *rtwdev);
-void rtw8852b_bb_set_pmac_tx(struct rtw89_dev *rtwdev,
-			     struct rtw8852b_bb_pmac_info *tx_info,
-			     enum rtw89_phy_idx idx);
-void rtw8852b_bb_set_pmac_pkt_tx(struct rtw89_dev *rtwdev, u8 enable,
-				 u16 tx_cnt, u16 period, u16 tx_time,
-				 enum rtw89_phy_idx idx);
-void rtw8852b_bb_set_power(struct rtw89_dev *rtwdev, s16 pwr_dbm,
-			   enum rtw89_phy_idx idx);
-void rtw8852b_bb_cfg_tx_path(struct rtw89_dev *rtwdev, u8 tx_path);
-void rtw8852b_bb_ctrl_rx_path(struct rtw89_dev *rtwdev,
-			      enum rtw89_rf_path_bit rx_path);
-void rtw8852b_bb_tx_mode_switch(struct rtw89_dev *rtwdev,
-				enum rtw89_phy_idx idx, u8 mode);
-void rtw8852b_bb_backup_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx idx,
-			     struct rtw8852b_bb_tssi_bak *bak);
-void rtw8852b_bb_restore_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx idx,
-			      const struct rtw8852b_bb_tssi_bak *bak);
-
 #endif
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
new file mode 100644
index 000000000000..0826d257ddba
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
@@ -0,0 +1,1701 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/* Copyright(c) 2024  Realtek Corporation
+ */
+
+#include "debug.h"
+#include "mac.h"
+#include "phy.h"
+#include "reg.h"
+#include "rtw8852b_common.h"
+
+static const struct rtw89_reg3_def rtw8852bx_pmac_ht20_mcs7_tbl[] = {
+	{0x4580, 0x0000ffff, 0x0},
+	{0x4580, 0xffff0000, 0x0},
+	{0x4584, 0x0000ffff, 0x0},
+	{0x4584, 0xffff0000, 0x0},
+	{0x4580, 0x0000ffff, 0x1},
+	{0x4578, 0x00ffffff, 0x2018b},
+	{0x4570, 0x03ffffff, 0x7},
+	{0x4574, 0x03ffffff, 0x32407},
+	{0x45b8, 0x00000010, 0x0},
+	{0x45b8, 0x00000100, 0x0},
+	{0x45b8, 0x00000080, 0x0},
+	{0x45b8, 0x00000008, 0x0},
+	{0x45a0, 0x0000ff00, 0x0},
+	{0x45a0, 0xff000000, 0x1},
+	{0x45a4, 0x0000ff00, 0x2},
+	{0x45a4, 0xff000000, 0x3},
+	{0x45b8, 0x00000020, 0x0},
+	{0x4568, 0xe0000000, 0x0},
+	{0x45b8, 0x00000002, 0x1},
+	{0x456c, 0xe0000000, 0x0},
+	{0x45b4, 0x00006000, 0x0},
+	{0x45b4, 0x00001800, 0x1},
+	{0x45b8, 0x00000040, 0x0},
+	{0x45b8, 0x00000004, 0x0},
+	{0x45b8, 0x00000200, 0x0},
+	{0x4598, 0xf8000000, 0x0},
+	{0x45b8, 0x00100000, 0x0},
+	{0x45a8, 0x00000fc0, 0x0},
+	{0x45b8, 0x00200000, 0x0},
+	{0x45b0, 0x00000038, 0x0},
+	{0x45b0, 0x000001c0, 0x0},
+	{0x45a0, 0x000000ff, 0x0},
+	{0x45b8, 0x00400000, 0x0},
+	{0x4590, 0x000007ff, 0x0},
+	{0x45b0, 0x00000e00, 0x0},
+	{0x45ac, 0x0000001f, 0x0},
+	{0x45b8, 0x00800000, 0x0},
+	{0x45a8, 0x0003f000, 0x0},
+	{0x45b8, 0x01000000, 0x0},
+	{0x45b0, 0x00007000, 0x0},
+	{0x45b0, 0x00038000, 0x0},
+	{0x45a0, 0x00ff0000, 0x0},
+	{0x45b8, 0x02000000, 0x0},
+	{0x4590, 0x003ff800, 0x0},
+	{0x45b0, 0x001c0000, 0x0},
+	{0x45ac, 0x000003e0, 0x0},
+	{0x45b8, 0x04000000, 0x0},
+	{0x45a8, 0x00fc0000, 0x0},
+	{0x45b8, 0x08000000, 0x0},
+	{0x45b0, 0x00e00000, 0x0},
+	{0x45b0, 0x07000000, 0x0},
+	{0x45a4, 0x000000ff, 0x0},
+	{0x45b8, 0x10000000, 0x0},
+	{0x4594, 0x000007ff, 0x0},
+	{0x45b0, 0x38000000, 0x0},
+	{0x45ac, 0x00007c00, 0x0},
+	{0x45b8, 0x20000000, 0x0},
+	{0x45a8, 0x3f000000, 0x0},
+	{0x45b8, 0x40000000, 0x0},
+	{0x45b4, 0x00000007, 0x0},
+	{0x45b4, 0x00000038, 0x0},
+	{0x45a4, 0x00ff0000, 0x0},
+	{0x45b8, 0x80000000, 0x0},
+	{0x4594, 0x003ff800, 0x0},
+	{0x45b4, 0x000001c0, 0x0},
+	{0x4598, 0xf8000000, 0x0},
+	{0x45b8, 0x00100000, 0x0},
+	{0x45a8, 0x00000fc0, 0x7},
+	{0x45b8, 0x00200000, 0x0},
+	{0x45b0, 0x00000038, 0x0},
+	{0x45b0, 0x000001c0, 0x0},
+	{0x45a0, 0x000000ff, 0x0},
+	{0x45b4, 0x06000000, 0x0},
+	{0x45b0, 0x00000007, 0x0},
+	{0x45b8, 0x00080000, 0x0},
+	{0x45a8, 0x0000003f, 0x0},
+	{0x457c, 0xffe00000, 0x1},
+	{0x4530, 0xffffffff, 0x0},
+	{0x4588, 0x00003fff, 0x0},
+	{0x4598, 0x000001ff, 0x0},
+	{0x4534, 0xffffffff, 0x0},
+	{0x4538, 0xffffffff, 0x0},
+	{0x453c, 0xffffffff, 0x0},
+	{0x4588, 0x0fffc000, 0x0},
+	{0x4598, 0x0003fe00, 0x0},
+	{0x4540, 0xffffffff, 0x0},
+	{0x4544, 0xffffffff, 0x0},
+	{0x4548, 0xffffffff, 0x0},
+	{0x458c, 0x00003fff, 0x0},
+	{0x4598, 0x07fc0000, 0x0},
+	{0x454c, 0xffffffff, 0x0},
+	{0x4550, 0xffffffff, 0x0},
+	{0x4554, 0xffffffff, 0x0},
+	{0x458c, 0x0fffc000, 0x0},
+	{0x459c, 0x000001ff, 0x0},
+	{0x4558, 0xffffffff, 0x0},
+	{0x455c, 0xffffffff, 0x0},
+	{0x4530, 0xffffffff, 0x4e790001},
+	{0x4588, 0x00003fff, 0x0},
+	{0x4598, 0x000001ff, 0x1},
+	{0x4534, 0xffffffff, 0x0},
+	{0x4538, 0xffffffff, 0x4b},
+	{0x45ac, 0x38000000, 0x7},
+	{0x4588, 0xf0000000, 0x0},
+	{0x459c, 0x7e000000, 0x0},
+	{0x45b8, 0x00040000, 0x0},
+	{0x45b8, 0x00020000, 0x0},
+	{0x4590, 0xffc00000, 0x0},
+	{0x45b8, 0x00004000, 0x0},
+	{0x4578, 0xff000000, 0x0},
+	{0x45b8, 0x00000400, 0x0},
+	{0x45b8, 0x00000800, 0x0},
+	{0x45b8, 0x00001000, 0x0},
+	{0x45b8, 0x00002000, 0x0},
+	{0x45b4, 0x00018000, 0x0},
+	{0x45ac, 0x07800000, 0x0},
+	{0x45b4, 0x00000600, 0x2},
+	{0x459c, 0x0001fe00, 0x80},
+	{0x45ac, 0x00078000, 0x3},
+	{0x459c, 0x01fe0000, 0x1},
+};
+
+static const struct rtw89_reg3_def rtw8852bx_btc_preagc_en_defs[] = {
+	{0x46D0, GENMASK(1, 0), 0x3},
+	{0x4790, GENMASK(1, 0), 0x3},
+	{0x4AD4, GENMASK(31, 0), 0xf},
+	{0x4AE0, GENMASK(31, 0), 0xf},
+	{0x4688, GENMASK(31, 24), 0x80},
+	{0x476C, GENMASK(31, 24), 0x80},
+	{0x4694, GENMASK(7, 0), 0x80},
+	{0x4694, GENMASK(15, 8), 0x80},
+	{0x4778, GENMASK(7, 0), 0x80},
+	{0x4778, GENMASK(15, 8), 0x80},
+	{0x4AE4, GENMASK(23, 0), 0x780D1E},
+	{0x4AEC, GENMASK(23, 0), 0x780D1E},
+	{0x469C, GENMASK(31, 26), 0x34},
+	{0x49F0, GENMASK(31, 26), 0x34},
+};
+
+static DECLARE_PHY_REG3_TBL(rtw8852bx_btc_preagc_en_defs);
+
+static const struct rtw89_reg3_def rtw8852bx_btc_preagc_dis_defs[] = {
+	{0x46D0, GENMASK(1, 0), 0x0},
+	{0x4790, GENMASK(1, 0), 0x0},
+	{0x4AD4, GENMASK(31, 0), 0x60},
+	{0x4AE0, GENMASK(31, 0), 0x60},
+	{0x4688, GENMASK(31, 24), 0x1a},
+	{0x476C, GENMASK(31, 24), 0x1a},
+	{0x4694, GENMASK(7, 0), 0x2a},
+	{0x4694, GENMASK(15, 8), 0x2a},
+	{0x4778, GENMASK(7, 0), 0x2a},
+	{0x4778, GENMASK(15, 8), 0x2a},
+	{0x4AE4, GENMASK(23, 0), 0x79E99E},
+	{0x4AEC, GENMASK(23, 0), 0x79E99E},
+	{0x469C, GENMASK(31, 26), 0x26},
+	{0x49F0, GENMASK(31, 26), 0x26},
+};
+
+static DECLARE_PHY_REG3_TBL(rtw8852bx_btc_preagc_dis_defs);
+
+static void rtw8852be_efuse_parsing(struct rtw89_efuse *efuse,
+				    struct rtw8852bx_efuse *map)
+{
+	ether_addr_copy(efuse->addr, map->e.mac_addr);
+	efuse->rfe_type = map->rfe_type;
+	efuse->xtal_cap = map->xtal_k;
+}
+
+static void rtw8852bx_efuse_parsing_tssi(struct rtw89_dev *rtwdev,
+					 struct rtw8852bx_efuse *map)
+{
+	struct rtw89_tssi_info *tssi = &rtwdev->tssi;
+	struct rtw8852bx_tssi_offset *ofst[] = {&map->path_a_tssi, &map->path_b_tssi};
+	u8 i, j;
+
+	tssi->thermal[RF_PATH_A] = map->path_a_therm;
+	tssi->thermal[RF_PATH_B] = map->path_b_therm;
+
+	for (i = 0; i < RF_PATH_NUM_8852BX; i++) {
+		memcpy(tssi->tssi_cck[i], ofst[i]->cck_tssi,
+		       sizeof(ofst[i]->cck_tssi));
+
+		for (j = 0; j < TSSI_CCK_CH_GROUP_NUM; j++)
+			rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+				    "[TSSI][EFUSE] path=%d cck[%d]=0x%x\n",
+				    i, j, tssi->tssi_cck[i][j]);
+
+		memcpy(tssi->tssi_mcs[i], ofst[i]->bw40_tssi,
+		       sizeof(ofst[i]->bw40_tssi));
+		memcpy(tssi->tssi_mcs[i] + TSSI_MCS_2G_CH_GROUP_NUM,
+		       ofst[i]->bw40_1s_tssi_5g, sizeof(ofst[i]->bw40_1s_tssi_5g));
+
+		for (j = 0; j < TSSI_MCS_CH_GROUP_NUM; j++)
+			rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+				    "[TSSI][EFUSE] path=%d mcs[%d]=0x%x\n",
+				    i, j, tssi->tssi_mcs[i][j]);
+	}
+}
+
+static bool _decode_efuse_gain(u8 data, s8 *high, s8 *low)
+{
+	if (high)
+		*high = sign_extend32(FIELD_GET(GENMASK(7,  4), data), 3);
+	if (low)
+		*low = sign_extend32(FIELD_GET(GENMASK(3,  0), data), 3);
+
+	return data != 0xff;
+}
+
+static void rtw8852bx_efuse_parsing_gain_offset(struct rtw89_dev *rtwdev,
+						struct rtw8852bx_efuse *map)
+{
+	struct rtw89_phy_efuse_gain *gain = &rtwdev->efuse_gain;
+	bool valid = false;
+
+	valid |= _decode_efuse_gain(map->rx_gain_2g_cck,
+				    &gain->offset[RF_PATH_A][RTW89_GAIN_OFFSET_2G_CCK],
+				    &gain->offset[RF_PATH_B][RTW89_GAIN_OFFSET_2G_CCK]);
+	valid |= _decode_efuse_gain(map->rx_gain_2g_ofdm,
+				    &gain->offset[RF_PATH_A][RTW89_GAIN_OFFSET_2G_OFDM],
+				    &gain->offset[RF_PATH_B][RTW89_GAIN_OFFSET_2G_OFDM]);
+	valid |= _decode_efuse_gain(map->rx_gain_5g_low,
+				    &gain->offset[RF_PATH_A][RTW89_GAIN_OFFSET_5G_LOW],
+				    &gain->offset[RF_PATH_B][RTW89_GAIN_OFFSET_5G_LOW]);
+	valid |= _decode_efuse_gain(map->rx_gain_5g_mid,
+				    &gain->offset[RF_PATH_A][RTW89_GAIN_OFFSET_5G_MID],
+				    &gain->offset[RF_PATH_B][RTW89_GAIN_OFFSET_5G_MID]);
+	valid |= _decode_efuse_gain(map->rx_gain_5g_high,
+				    &gain->offset[RF_PATH_A][RTW89_GAIN_OFFSET_5G_HIGH],
+				    &gain->offset[RF_PATH_B][RTW89_GAIN_OFFSET_5G_HIGH]);
+
+	gain->offset_valid = valid;
+}
+
+static int __rtw8852bx_read_efuse(struct rtw89_dev *rtwdev, u8 *log_map,
+				  enum rtw89_efuse_block block)
+{
+	struct rtw89_efuse *efuse = &rtwdev->efuse;
+	struct rtw8852bx_efuse *map;
+
+	map = (struct rtw8852bx_efuse *)log_map;
+
+	efuse->country_code[0] = map->country_code[0];
+	efuse->country_code[1] = map->country_code[1];
+	rtw8852bx_efuse_parsing_tssi(rtwdev, map);
+	rtw8852bx_efuse_parsing_gain_offset(rtwdev, map);
+
+	switch (rtwdev->hci.type) {
+	case RTW89_HCI_TYPE_PCIE:
+		rtw8852be_efuse_parsing(efuse, map);
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	rtw89_info(rtwdev, "chip rfe_type is %d\n", efuse->rfe_type);
+
+	return 0;
+}
+
+static void rtw8852bx_phycap_parsing_power_cal(struct rtw89_dev *rtwdev, u8 *phycap_map)
+{
+#define PWR_K_CHK_OFFSET 0x5E9
+#define PWR_K_CHK_VALUE 0xAA
+	u32 offset = PWR_K_CHK_OFFSET - rtwdev->chip->phycap_addr;
+
+	if (phycap_map[offset] == PWR_K_CHK_VALUE)
+		rtwdev->efuse.power_k_valid = true;
+}
+
+static void rtw8852bx_phycap_parsing_tssi(struct rtw89_dev *rtwdev, u8 *phycap_map)
+{
+	struct rtw89_tssi_info *tssi = &rtwdev->tssi;
+	static const u32 tssi_trim_addr[RF_PATH_NUM_8852BX] = {0x5D6, 0x5AB};
+	u32 addr = rtwdev->chip->phycap_addr;
+	bool pg = false;
+	u32 ofst;
+	u8 i, j;
+
+	for (i = 0; i < RF_PATH_NUM_8852BX; i++) {
+		for (j = 0; j < TSSI_TRIM_CH_GROUP_NUM; j++) {
+			/* addrs are in decreasing order */
+			ofst = tssi_trim_addr[i] - addr - j;
+			tssi->tssi_trim[i][j] = phycap_map[ofst];
+
+			if (phycap_map[ofst] != 0xff)
+				pg = true;
+		}
+	}
+
+	if (!pg) {
+		memset(tssi->tssi_trim, 0, sizeof(tssi->tssi_trim));
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+			    "[TSSI][TRIM] no PG, set all trim info to 0\n");
+	}
+
+	for (i = 0; i < RF_PATH_NUM_8852BX; i++)
+		for (j = 0; j < TSSI_TRIM_CH_GROUP_NUM; j++)
+			rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+				    "[TSSI] path=%d idx=%d trim=0x%x addr=0x%x\n",
+				    i, j, tssi->tssi_trim[i][j],
+				    tssi_trim_addr[i] - j);
+}
+
+static void rtw8852bx_phycap_parsing_thermal_trim(struct rtw89_dev *rtwdev,
+						  u8 *phycap_map)
+{
+	struct rtw89_power_trim_info *info = &rtwdev->pwr_trim;
+	static const u32 thm_trim_addr[RF_PATH_NUM_8852BX] = {0x5DF, 0x5DC};
+	u32 addr = rtwdev->chip->phycap_addr;
+	u8 i;
+
+	for (i = 0; i < RF_PATH_NUM_8852BX; i++) {
+		info->thermal_trim[i] = phycap_map[thm_trim_addr[i] - addr];
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[THERMAL][TRIM] path=%d thermal_trim=0x%x\n",
+			    i, info->thermal_trim[i]);
+
+		if (info->thermal_trim[i] != 0xff)
+			info->pg_thermal_trim = true;
+	}
+}
+
+static void rtw8852bx_thermal_trim(struct rtw89_dev *rtwdev)
+{
+#define __thm_setting(raw)				\
+({							\
+	u8 __v = (raw);					\
+	((__v & 0x1) << 3) | ((__v & 0x1f) >> 1);	\
+})
+	struct rtw89_power_trim_info *info = &rtwdev->pwr_trim;
+	u8 i, val;
+
+	if (!info->pg_thermal_trim) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[THERMAL][TRIM] no PG, do nothing\n");
+
+		return;
+	}
+
+	for (i = 0; i < RF_PATH_NUM_8852BX; i++) {
+		val = __thm_setting(info->thermal_trim[i]);
+		rtw89_write_rf(rtwdev, i, RR_TM2, RR_TM2_OFF, val);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[THERMAL][TRIM] path=%d thermal_setting=0x%x\n",
+			    i, val);
+	}
+#undef __thm_setting
+}
+
+static void rtw8852bx_phycap_parsing_pa_bias_trim(struct rtw89_dev *rtwdev,
+						  u8 *phycap_map)
+{
+	struct rtw89_power_trim_info *info = &rtwdev->pwr_trim;
+	static const u32 pabias_trim_addr[RF_PATH_NUM_8852BX] = {0x5DE, 0x5DB};
+	u32 addr = rtwdev->chip->phycap_addr;
+	u8 i;
+
+	for (i = 0; i < RF_PATH_NUM_8852BX; i++) {
+		info->pa_bias_trim[i] = phycap_map[pabias_trim_addr[i] - addr];
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[PA_BIAS][TRIM] path=%d pa_bias_trim=0x%x\n",
+			    i, info->pa_bias_trim[i]);
+
+		if (info->pa_bias_trim[i] != 0xff)
+			info->pg_pa_bias_trim = true;
+	}
+}
+
+static void rtw8852bx_pa_bias_trim(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_power_trim_info *info = &rtwdev->pwr_trim;
+	u8 pabias_2g, pabias_5g;
+	u8 i;
+
+	if (!info->pg_pa_bias_trim) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[PA_BIAS][TRIM] no PG, do nothing\n");
+
+		return;
+	}
+
+	for (i = 0; i < RF_PATH_NUM_8852BX; i++) {
+		pabias_2g = FIELD_GET(GENMASK(3, 0), info->pa_bias_trim[i]);
+		pabias_5g = FIELD_GET(GENMASK(7, 4), info->pa_bias_trim[i]);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[PA_BIAS][TRIM] path=%d 2G=0x%x 5G=0x%x\n",
+			    i, pabias_2g, pabias_5g);
+
+		rtw89_write_rf(rtwdev, i, RR_BIASA, RR_BIASA_TXG, pabias_2g);
+		rtw89_write_rf(rtwdev, i, RR_BIASA, RR_BIASA_TXA, pabias_5g);
+	}
+}
+
+static void rtw8852bx_phycap_parsing_gain_comp(struct rtw89_dev *rtwdev, u8 *phycap_map)
+{
+	static const u32 comp_addrs[][RTW89_SUBBAND_2GHZ_5GHZ_NR] = {
+		{0x5BB, 0x5BA, 0, 0x5B9, 0x5B8},
+		{0x590, 0x58F, 0, 0x58E, 0x58D},
+	};
+	struct rtw89_phy_efuse_gain *gain = &rtwdev->efuse_gain;
+	u32 phycap_addr = rtwdev->chip->phycap_addr;
+	bool valid = false;
+	int path, i;
+	u8 data;
+
+	for (path = 0; path < 2; path++)
+		for (i = 0; i < RTW89_SUBBAND_2GHZ_5GHZ_NR; i++) {
+			if (comp_addrs[path][i] == 0)
+				continue;
+
+			data = phycap_map[comp_addrs[path][i] - phycap_addr];
+			valid |= _decode_efuse_gain(data, NULL,
+						    &gain->comp[path][i]);
+		}
+
+	gain->comp_valid = valid;
+}
+
+static int __rtw8852bx_read_phycap(struct rtw89_dev *rtwdev, u8 *phycap_map)
+{
+	rtw8852bx_phycap_parsing_power_cal(rtwdev, phycap_map);
+	rtw8852bx_phycap_parsing_tssi(rtwdev, phycap_map);
+	rtw8852bx_phycap_parsing_thermal_trim(rtwdev, phycap_map);
+	rtw8852bx_phycap_parsing_pa_bias_trim(rtwdev, phycap_map);
+	rtw8852bx_phycap_parsing_gain_comp(rtwdev, phycap_map);
+
+	return 0;
+}
+
+static void __rtw8852bx_power_trim(struct rtw89_dev *rtwdev)
+{
+	rtw8852bx_thermal_trim(rtwdev);
+	rtw8852bx_pa_bias_trim(rtwdev);
+}
+
+static void __rtw8852bx_set_channel_mac(struct rtw89_dev *rtwdev,
+					const struct rtw89_chan *chan,
+					u8 mac_idx)
+{
+	u32 rf_mod = rtw89_mac_reg_by_idx(rtwdev, R_AX_WMAC_RFMOD, mac_idx);
+	u32 sub_carr = rtw89_mac_reg_by_idx(rtwdev, R_AX_TX_SUB_CARRIER_VALUE, mac_idx);
+	u32 chk_rate = rtw89_mac_reg_by_idx(rtwdev, R_AX_TXRATE_CHK, mac_idx);
+	u8 txsc20 = 0, txsc40 = 0;
+
+	switch (chan->band_width) {
+	case RTW89_CHANNEL_WIDTH_80:
+		txsc40 = rtw89_phy_get_txsc(rtwdev, chan, RTW89_CHANNEL_WIDTH_40);
+		fallthrough;
+	case RTW89_CHANNEL_WIDTH_40:
+		txsc20 = rtw89_phy_get_txsc(rtwdev, chan, RTW89_CHANNEL_WIDTH_20);
+		break;
+	default:
+		break;
+	}
+
+	switch (chan->band_width) {
+	case RTW89_CHANNEL_WIDTH_80:
+		rtw89_write8_mask(rtwdev, rf_mod, B_AX_WMAC_RFMOD_MASK, BIT(1));
+		rtw89_write32(rtwdev, sub_carr, txsc20 | (txsc40 << 4));
+		break;
+	case RTW89_CHANNEL_WIDTH_40:
+		rtw89_write8_mask(rtwdev, rf_mod, B_AX_WMAC_RFMOD_MASK, BIT(0));
+		rtw89_write32(rtwdev, sub_carr, txsc20);
+		break;
+	case RTW89_CHANNEL_WIDTH_20:
+		rtw89_write8_clr(rtwdev, rf_mod, B_AX_WMAC_RFMOD_MASK);
+		rtw89_write32(rtwdev, sub_carr, 0);
+		break;
+	default:
+		break;
+	}
+
+	if (chan->channel > 14) {
+		rtw89_write8_clr(rtwdev, chk_rate, B_AX_BAND_MODE);
+		rtw89_write8_set(rtwdev, chk_rate,
+				 B_AX_CHECK_CCK_EN | B_AX_RTS_LIMIT_IN_OFDM6);
+	} else {
+		rtw89_write8_set(rtwdev, chk_rate, B_AX_BAND_MODE);
+		rtw89_write8_clr(rtwdev, chk_rate,
+				 B_AX_CHECK_CCK_EN | B_AX_RTS_LIMIT_IN_OFDM6);
+	}
+}
+
+static const u32 rtw8852bx_sco_barker_threshold[14] = {
+	0x1cfea, 0x1d0e1, 0x1d1d7, 0x1d2cd, 0x1d3c3, 0x1d4b9, 0x1d5b0, 0x1d6a6,
+	0x1d79c, 0x1d892, 0x1d988, 0x1da7f, 0x1db75, 0x1ddc4
+};
+
+static const u32 rtw8852bx_sco_cck_threshold[14] = {
+	0x27de3, 0x27f35, 0x28088, 0x281da, 0x2832d, 0x2847f, 0x285d2, 0x28724,
+	0x28877, 0x289c9, 0x28b1c, 0x28c6e, 0x28dc1, 0x290ed
+};
+
+static void rtw8852bx_ctrl_sco_cck(struct rtw89_dev *rtwdev, u8 primary_ch)
+{
+	u8 ch_element = primary_ch - 1;
+
+	rtw89_phy_write32_mask(rtwdev, R_RXSCOBC, B_RXSCOBC_TH,
+			       rtw8852bx_sco_barker_threshold[ch_element]);
+	rtw89_phy_write32_mask(rtwdev, R_RXSCOCCK, B_RXSCOCCK_TH,
+			       rtw8852bx_sco_cck_threshold[ch_element]);
+}
+
+static u8 rtw8852bx_sco_mapping(u8 central_ch)
+{
+	if (central_ch == 1)
+		return 109;
+	else if (central_ch >= 2 && central_ch <= 6)
+		return 108;
+	else if (central_ch >= 7 && central_ch <= 10)
+		return 107;
+	else if (central_ch >= 11 && central_ch <= 14)
+		return 106;
+	else if (central_ch == 36 || central_ch == 38)
+		return 51;
+	else if (central_ch >= 40 && central_ch <= 58)
+		return 50;
+	else if (central_ch >= 60 && central_ch <= 64)
+		return 49;
+	else if (central_ch == 100 || central_ch == 102)
+		return 48;
+	else if (central_ch >= 104 && central_ch <= 126)
+		return 47;
+	else if (central_ch >= 128 && central_ch <= 151)
+		return 46;
+	else if (central_ch >= 153 && central_ch <= 177)
+		return 45;
+	else
+		return 0;
+}
+
+struct rtw8852bx_bb_gain {
+	u32 gain_g[BB_PATH_NUM_8852BX];
+	u32 gain_a[BB_PATH_NUM_8852BX];
+	u32 gain_mask;
+};
+
+static const struct rtw8852bx_bb_gain bb_gain_lna[LNA_GAIN_NUM] = {
+	{ .gain_g = {0x4678, 0x475C}, .gain_a = {0x45DC, 0x4740},
+	  .gain_mask = 0x00ff0000 },
+	{ .gain_g = {0x4678, 0x475C}, .gain_a = {0x45DC, 0x4740},
+	  .gain_mask = 0xff000000 },
+	{ .gain_g = {0x467C, 0x4760}, .gain_a = {0x4660, 0x4744},
+	  .gain_mask = 0x000000ff },
+	{ .gain_g = {0x467C, 0x4760}, .gain_a = {0x4660, 0x4744},
+	  .gain_mask = 0x0000ff00 },
+	{ .gain_g = {0x467C, 0x4760}, .gain_a = {0x4660, 0x4744},
+	  .gain_mask = 0x00ff0000 },
+	{ .gain_g = {0x467C, 0x4760}, .gain_a = {0x4660, 0x4744},
+	  .gain_mask = 0xff000000 },
+	{ .gain_g = {0x4680, 0x4764}, .gain_a = {0x4664, 0x4748},
+	  .gain_mask = 0x000000ff },
+};
+
+static const struct rtw8852bx_bb_gain bb_gain_tia[TIA_GAIN_NUM] = {
+	{ .gain_g = {0x4680, 0x4764}, .gain_a = {0x4664, 0x4748},
+	  .gain_mask = 0x00ff0000 },
+	{ .gain_g = {0x4680, 0x4764}, .gain_a = {0x4664, 0x4748},
+	  .gain_mask = 0xff000000 },
+};
+
+static void rtw8852bx_set_gain_error(struct rtw89_dev *rtwdev,
+				     enum rtw89_subband subband,
+				     enum rtw89_rf_path path)
+{
+	const struct rtw89_phy_bb_gain_info *gain = &rtwdev->bb_gain.ax;
+	u8 gain_band = rtw89_subband_to_bb_gain_band(subband);
+	s32 val;
+	u32 reg;
+	u32 mask;
+	int i;
+
+	for (i = 0; i < LNA_GAIN_NUM; i++) {
+		if (subband == RTW89_CH_2G)
+			reg = bb_gain_lna[i].gain_g[path];
+		else
+			reg = bb_gain_lna[i].gain_a[path];
+
+		mask = bb_gain_lna[i].gain_mask;
+		val = gain->lna_gain[gain_band][path][i];
+		rtw89_phy_write32_mask(rtwdev, reg, mask, val);
+	}
+
+	for (i = 0; i < TIA_GAIN_NUM; i++) {
+		if (subband == RTW89_CH_2G)
+			reg = bb_gain_tia[i].gain_g[path];
+		else
+			reg = bb_gain_tia[i].gain_a[path];
+
+		mask = bb_gain_tia[i].gain_mask;
+		val = gain->tia_gain[gain_band][path][i];
+		rtw89_phy_write32_mask(rtwdev, reg, mask, val);
+	}
+}
+
+static void rtw8852bx_set_gain_offset(struct rtw89_dev *rtwdev,
+				      enum rtw89_subband subband,
+				      enum rtw89_phy_idx phy_idx)
+{
+	static const u32 gain_err_addr[2] = {R_P0_AGC_RSVD, R_P1_AGC_RSVD};
+	static const u32 rssi_ofst_addr[2] = {R_PATH0_G_TIA1_LNA6_OP1DB_V1,
+					      R_PATH1_G_TIA1_LNA6_OP1DB_V1};
+	struct rtw89_hal *hal = &rtwdev->hal;
+	struct rtw89_phy_efuse_gain *efuse_gain = &rtwdev->efuse_gain;
+	enum rtw89_gain_offset gain_ofdm_band;
+	s32 offset_a, offset_b;
+	s32 offset_ofdm, offset_cck;
+	s32 tmp;
+	u8 path;
+
+	if (!efuse_gain->comp_valid)
+		goto next;
+
+	for (path = RF_PATH_A; path < BB_PATH_NUM_8852BX; path++) {
+		tmp = efuse_gain->comp[path][subband];
+		tmp = clamp_t(s32, tmp << 2, S8_MIN, S8_MAX);
+		rtw89_phy_write32_mask(rtwdev, gain_err_addr[path], MASKBYTE0, tmp);
+	}
+
+next:
+	if (!efuse_gain->offset_valid)
+		return;
+
+	gain_ofdm_band = rtw89_subband_to_gain_offset_band_of_ofdm(subband);
+
+	offset_a = -efuse_gain->offset[RF_PATH_A][gain_ofdm_band];
+	offset_b = -efuse_gain->offset[RF_PATH_B][gain_ofdm_band];
+
+	tmp = -((offset_a << 2) + (efuse_gain->offset_base[RTW89_PHY_0] >> 2));
+	tmp = clamp_t(s32, tmp, S8_MIN, S8_MAX);
+	rtw89_phy_write32_mask(rtwdev, rssi_ofst_addr[RF_PATH_A], B_PATH0_R_G_OFST_MASK, tmp);
+
+	tmp = -((offset_b << 2) + (efuse_gain->offset_base[RTW89_PHY_0] >> 2));
+	tmp = clamp_t(s32, tmp, S8_MIN, S8_MAX);
+	rtw89_phy_write32_mask(rtwdev, rssi_ofst_addr[RF_PATH_B], B_PATH0_R_G_OFST_MASK, tmp);
+
+	if (hal->antenna_rx == RF_B) {
+		offset_ofdm = -efuse_gain->offset[RF_PATH_B][gain_ofdm_band];
+		offset_cck = -efuse_gain->offset[RF_PATH_B][0];
+	} else {
+		offset_ofdm = -efuse_gain->offset[RF_PATH_A][gain_ofdm_band];
+		offset_cck = -efuse_gain->offset[RF_PATH_A][0];
+	}
+
+	tmp = (offset_ofdm << 4) + efuse_gain->offset_base[RTW89_PHY_0];
+	tmp = clamp_t(s32, tmp, S8_MIN, S8_MAX);
+	rtw89_phy_write32_idx(rtwdev, R_P0_RPL1, B_P0_RPL1_BIAS_MASK, tmp, phy_idx);
+
+	tmp = (offset_ofdm << 4) + efuse_gain->rssi_base[RTW89_PHY_0];
+	tmp = clamp_t(s32, tmp, S8_MIN, S8_MAX);
+	rtw89_phy_write32_idx(rtwdev, R_P1_RPL1, B_P0_RPL1_BIAS_MASK, tmp, phy_idx);
+
+	if (subband == RTW89_CH_2G) {
+		tmp = (offset_cck << 3) + (efuse_gain->offset_base[RTW89_PHY_0] >> 1);
+		tmp = clamp_t(s32, tmp, S8_MIN >> 1, S8_MAX >> 1);
+		rtw89_phy_write32_mask(rtwdev, R_RX_RPL_OFST,
+				       B_RX_RPL_OFST_CCK_MASK, tmp);
+	}
+}
+
+static
+void rtw8852bx_set_rxsc_rpl_comp(struct rtw89_dev *rtwdev, enum rtw89_subband subband)
+{
+	const struct rtw89_phy_bb_gain_info *gain = &rtwdev->bb_gain.ax;
+	u8 band = rtw89_subband_to_bb_gain_band(subband);
+	u32 val;
+
+	val = u32_encode_bits((gain->rpl_ofst_20[band][RF_PATH_A] +
+			       gain->rpl_ofst_20[band][RF_PATH_B]) >> 1, B_P0_RPL1_20_MASK) |
+	      u32_encode_bits((gain->rpl_ofst_40[band][RF_PATH_A][0] +
+			       gain->rpl_ofst_40[band][RF_PATH_B][0]) >> 1, B_P0_RPL1_40_MASK) |
+	      u32_encode_bits((gain->rpl_ofst_40[band][RF_PATH_A][1] +
+			       gain->rpl_ofst_40[band][RF_PATH_B][1]) >> 1, B_P0_RPL1_41_MASK);
+	val >>= B_P0_RPL1_SHIFT;
+	rtw89_phy_write32_mask(rtwdev, R_P0_RPL1, B_P0_RPL1_MASK, val);
+	rtw89_phy_write32_mask(rtwdev, R_P1_RPL1, B_P0_RPL1_MASK, val);
+
+	val = u32_encode_bits((gain->rpl_ofst_40[band][RF_PATH_A][2] +
+			       gain->rpl_ofst_40[band][RF_PATH_B][2]) >> 1, B_P0_RTL2_42_MASK) |
+	      u32_encode_bits((gain->rpl_ofst_80[band][RF_PATH_A][0] +
+			       gain->rpl_ofst_80[band][RF_PATH_B][0]) >> 1, B_P0_RTL2_80_MASK) |
+	      u32_encode_bits((gain->rpl_ofst_80[band][RF_PATH_A][1] +
+			       gain->rpl_ofst_80[band][RF_PATH_B][1]) >> 1, B_P0_RTL2_81_MASK) |
+	      u32_encode_bits((gain->rpl_ofst_80[band][RF_PATH_A][10] +
+			       gain->rpl_ofst_80[band][RF_PATH_B][10]) >> 1, B_P0_RTL2_8A_MASK);
+	rtw89_phy_write32(rtwdev, R_P0_RPL2, val);
+	rtw89_phy_write32(rtwdev, R_P1_RPL2, val);
+
+	val = u32_encode_bits((gain->rpl_ofst_80[band][RF_PATH_A][2] +
+			       gain->rpl_ofst_80[band][RF_PATH_B][2]) >> 1, B_P0_RTL3_82_MASK) |
+	      u32_encode_bits((gain->rpl_ofst_80[band][RF_PATH_A][3] +
+			       gain->rpl_ofst_80[band][RF_PATH_B][3]) >> 1, B_P0_RTL3_83_MASK) |
+	      u32_encode_bits((gain->rpl_ofst_80[band][RF_PATH_A][4] +
+			       gain->rpl_ofst_80[band][RF_PATH_B][4]) >> 1, B_P0_RTL3_84_MASK) |
+	      u32_encode_bits((gain->rpl_ofst_80[band][RF_PATH_A][9] +
+			       gain->rpl_ofst_80[band][RF_PATH_B][9]) >> 1, B_P0_RTL3_89_MASK);
+	rtw89_phy_write32(rtwdev, R_P0_RPL3, val);
+	rtw89_phy_write32(rtwdev, R_P1_RPL3, val);
+}
+
+static void rtw8852bx_ctrl_ch(struct rtw89_dev *rtwdev,
+			      const struct rtw89_chan *chan,
+			      enum rtw89_phy_idx phy_idx)
+{
+	u8 central_ch = chan->channel;
+	u8 subband = chan->subband_type;
+	u8 sco_comp;
+	bool is_2g = central_ch <= 14;
+
+	/* Path A */
+	if (is_2g)
+		rtw89_phy_write32_idx(rtwdev, R_PATH0_BAND_SEL_V1,
+				      B_PATH0_BAND_SEL_MSK_V1, 1, phy_idx);
+	else
+		rtw89_phy_write32_idx(rtwdev, R_PATH0_BAND_SEL_V1,
+				      B_PATH0_BAND_SEL_MSK_V1, 0, phy_idx);
+
+	/* Path B */
+	if (is_2g)
+		rtw89_phy_write32_idx(rtwdev, R_PATH1_BAND_SEL_V1,
+				      B_PATH1_BAND_SEL_MSK_V1, 1, phy_idx);
+	else
+		rtw89_phy_write32_idx(rtwdev, R_PATH1_BAND_SEL_V1,
+				      B_PATH1_BAND_SEL_MSK_V1, 0, phy_idx);
+
+	/* SCO compensate FC setting */
+	sco_comp = rtw8852bx_sco_mapping(central_ch);
+	rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_INV, sco_comp, phy_idx);
+
+	if (chan->band_type == RTW89_BAND_6G)
+		return;
+
+	/* CCK parameters */
+	if (central_ch == 14) {
+		rtw89_phy_write32_mask(rtwdev, R_TXFIR0, B_TXFIR_C01, 0x3b13ff);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIR2, B_TXFIR_C23, 0x1c42de);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIR4, B_TXFIR_C45, 0xfdb0ad);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIR6, B_TXFIR_C67, 0xf60f6e);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIR8, B_TXFIR_C89, 0xfd8f92);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIRA, B_TXFIR_CAB, 0x2d011);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIRC, B_TXFIR_CCD, 0x1c02c);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIRE, B_TXFIR_CEF, 0xfff00a);
+	} else {
+		rtw89_phy_write32_mask(rtwdev, R_TXFIR0, B_TXFIR_C01, 0x3d23ff);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIR2, B_TXFIR_C23, 0x29b354);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIR4, B_TXFIR_C45, 0xfc1c8);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIR6, B_TXFIR_C67, 0xfdb053);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIR8, B_TXFIR_C89, 0xf86f9a);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIRA, B_TXFIR_CAB, 0xfaef92);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIRC, B_TXFIR_CCD, 0xfe5fcc);
+		rtw89_phy_write32_mask(rtwdev, R_TXFIRE, B_TXFIR_CEF, 0xffdff5);
+	}
+
+	rtw8852bx_set_gain_error(rtwdev, subband, RF_PATH_A);
+	rtw8852bx_set_gain_error(rtwdev, subband, RF_PATH_B);
+	rtw8852bx_set_gain_offset(rtwdev, subband, phy_idx);
+	rtw8852bx_set_rxsc_rpl_comp(rtwdev, subband);
+}
+
+static void rtw8852b_bw_setting(struct rtw89_dev *rtwdev, u8 bw, u8 path)
+{
+	static const u32 adc_sel[2] = {0xC0EC, 0xC1EC};
+	static const u32 wbadc_sel[2] = {0xC0E4, 0xC1E4};
+
+	switch (bw) {
+	case RTW89_CHANNEL_WIDTH_5:
+		rtw89_phy_write32_mask(rtwdev, adc_sel[path], 0x6000, 0x1);
+		rtw89_phy_write32_mask(rtwdev, wbadc_sel[path], 0x30, 0x0);
+		break;
+	case RTW89_CHANNEL_WIDTH_10:
+		rtw89_phy_write32_mask(rtwdev, adc_sel[path], 0x6000, 0x2);
+		rtw89_phy_write32_mask(rtwdev, wbadc_sel[path], 0x30, 0x1);
+		break;
+	case RTW89_CHANNEL_WIDTH_20:
+		rtw89_phy_write32_mask(rtwdev, adc_sel[path], 0x6000, 0x0);
+		rtw89_phy_write32_mask(rtwdev, wbadc_sel[path], 0x30, 0x2);
+		break;
+	case RTW89_CHANNEL_WIDTH_40:
+		rtw89_phy_write32_mask(rtwdev, adc_sel[path], 0x6000, 0x0);
+		rtw89_phy_write32_mask(rtwdev, wbadc_sel[path], 0x30, 0x2);
+		break;
+	case RTW89_CHANNEL_WIDTH_80:
+		rtw89_phy_write32_mask(rtwdev, adc_sel[path], 0x6000, 0x0);
+		rtw89_phy_write32_mask(rtwdev, wbadc_sel[path], 0x30, 0x2);
+		break;
+	default:
+		rtw89_warn(rtwdev, "Fail to set ADC\n");
+	}
+}
+
+static void rtw8852bx_ctrl_bw(struct rtw89_dev *rtwdev, u8 pri_ch, u8 bw,
+			      enum rtw89_phy_idx phy_idx)
+{
+	u32 rx_path_0;
+
+	switch (bw) {
+	case RTW89_CHANNEL_WIDTH_5:
+		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x0, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_SBW, 0x1, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_PRICH, 0x0, phy_idx);
+
+		/*Set RF mode at 3 */
+		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
+				      B_P0_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
+				      B_P1_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
+		break;
+	case RTW89_CHANNEL_WIDTH_10:
+		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x0, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_SBW, 0x2, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_PRICH, 0x0, phy_idx);
+
+		/*Set RF mode at 3 */
+		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
+				      B_P0_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
+				      B_P1_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
+		break;
+	case RTW89_CHANNEL_WIDTH_20:
+		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x0, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_SBW, 0x0, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_PRICH, 0x0, phy_idx);
+
+		/*Set RF mode at 3 */
+		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
+				      B_P0_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
+				      B_P1_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
+		break;
+	case RTW89_CHANNEL_WIDTH_40:
+		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x1, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_SBW, 0x0, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_PRICH,
+				      pri_ch, phy_idx);
+
+		/*Set RF mode at 3 */
+		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
+				      B_P0_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
+				      B_P1_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
+		/*CCK primary channel */
+		if (pri_ch == RTW89_SC_20_UPPER)
+			rtw89_phy_write32_mask(rtwdev, R_RXSC, B_RXSC_EN, 1);
+		else
+			rtw89_phy_write32_mask(rtwdev, R_RXSC, B_RXSC_EN, 0);
+
+		break;
+	case RTW89_CHANNEL_WIDTH_80:
+		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x2, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_SBW, 0x0, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_CHBW_MOD_PRICH,
+				      pri_ch, phy_idx);
+
+		/*Set RF mode at A */
+		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
+				      B_P0_RFMODE_ORI_RX_ALL, 0xaaa, phy_idx);
+		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
+				      B_P1_RFMODE_ORI_RX_ALL, 0xaaa, phy_idx);
+		break;
+	default:
+		rtw89_warn(rtwdev, "Fail to switch bw (bw:%d, pri ch:%d)\n", bw,
+			   pri_ch);
+	}
+
+	rtw8852b_bw_setting(rtwdev, bw, RF_PATH_A);
+	rtw8852b_bw_setting(rtwdev, bw, RF_PATH_B);
+
+	rx_path_0 = rtw89_phy_read32_idx(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0,
+					 phy_idx);
+	if (rx_path_0 == 0x1)
+		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
+				      B_P1_RFMODE_ORI_RX_ALL, 0x111, phy_idx);
+	else if (rx_path_0 == 0x2)
+		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
+				      B_P0_RFMODE_ORI_RX_ALL, 0x111, phy_idx);
+}
+
+static void rtw8852bx_ctrl_cck_en(struct rtw89_dev *rtwdev, bool cck_en)
+{
+	if (cck_en) {
+		rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC, B_ENABLE_CCK, 1);
+		rtw89_phy_write32_mask(rtwdev, R_RXCCA, B_RXCCA_DIS, 0);
+	} else {
+		rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC, B_ENABLE_CCK, 0);
+		rtw89_phy_write32_mask(rtwdev, R_RXCCA, B_RXCCA_DIS, 1);
+	}
+}
+
+static void rtw8852bx_5m_mask(struct rtw89_dev *rtwdev, const struct rtw89_chan *chan,
+			      enum rtw89_phy_idx phy_idx)
+{
+	u8 pri_ch = chan->pri_ch_idx;
+	bool mask_5m_low;
+	bool mask_5m_en;
+
+	switch (chan->band_width) {
+	case RTW89_CHANNEL_WIDTH_40:
+		/* Prich=1: Mask 5M High, Prich=2: Mask 5M Low */
+		mask_5m_en = true;
+		mask_5m_low = pri_ch == RTW89_SC_20_LOWER;
+		break;
+	case RTW89_CHANNEL_WIDTH_80:
+		/* Prich=3: Mask 5M High, Prich=4: Mask 5M Low, Else: Disable */
+		mask_5m_en = pri_ch == RTW89_SC_20_UPMOST ||
+			     pri_ch == RTW89_SC_20_LOWEST;
+		mask_5m_low = pri_ch == RTW89_SC_20_LOWEST;
+		break;
+	default:
+		mask_5m_en = false;
+		break;
+	}
+
+	if (!mask_5m_en) {
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_EN, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_EN, 0x0);
+		rtw89_phy_write32_idx(rtwdev, R_ASSIGN_SBD_OPT_V1,
+				      B_ASSIGN_SBD_OPT_EN_V1, 0x0, phy_idx);
+		return;
+	}
+
+	if (mask_5m_low) {
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_TH, 0x4);
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_EN, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_SB2, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_SB0, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_TH, 0x4);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_EN, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_SB2, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_SB0, 0x1);
+	} else {
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_TH, 0x4);
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_EN, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_SB2, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_5MDET_V1, B_PATH0_5MDET_SB0, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_TH, 0x4);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_EN, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_SB2, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_5MDET_V1, B_PATH1_5MDET_SB0, 0x0);
+	}
+	rtw89_phy_write32_idx(rtwdev, R_ASSIGN_SBD_OPT_V1,
+			      B_ASSIGN_SBD_OPT_EN_V1, 0x1, phy_idx);
+}
+
+static void __rtw8852bx_bb_reset_all(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
+{
+	rtw89_phy_write32_idx(rtwdev, R_S0_HW_SI_DIS, B_S0_HW_SI_DIS_W_R_TRIG, 0x7, phy_idx);
+	rtw89_phy_write32_idx(rtwdev, R_S1_HW_SI_DIS, B_S1_HW_SI_DIS_W_R_TRIG, 0x7, phy_idx);
+	fsleep(1);
+	rtw89_phy_write32_idx(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_ALL, 1, phy_idx);
+	rtw89_phy_write32_idx(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_ALL, 0, phy_idx);
+	rtw89_phy_write32_idx(rtwdev, R_S0_HW_SI_DIS, B_S0_HW_SI_DIS_W_R_TRIG, 0x0, phy_idx);
+	rtw89_phy_write32_idx(rtwdev, R_S1_HW_SI_DIS, B_S1_HW_SI_DIS_W_R_TRIG, 0x0, phy_idx);
+	rtw89_phy_write32_idx(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_ALL, 1, phy_idx);
+}
+
+static void rtw8852bx_bb_macid_ctrl_init(struct rtw89_dev *rtwdev,
+					 enum rtw89_phy_idx phy_idx)
+{
+	u32 addr;
+
+	for (addr = R_AX_PWR_MACID_LMT_TABLE0;
+	     addr <= R_AX_PWR_MACID_LMT_TABLE127; addr += 4)
+		rtw89_mac_txpwr_write32(rtwdev, phy_idx, addr, 0);
+}
+
+static void __rtw8852bx_bb_sethw(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_phy_efuse_gain *gain = &rtwdev->efuse_gain;
+
+	rtw89_phy_write32_clr(rtwdev, R_P0_EN_SOUND_WO_NDP, B_P0_EN_SOUND_WO_NDP);
+	rtw89_phy_write32_clr(rtwdev, R_P1_EN_SOUND_WO_NDP, B_P1_EN_SOUND_WO_NDP);
+
+	rtw8852bx_bb_macid_ctrl_init(rtwdev, RTW89_PHY_0);
+
+	/* read these registers after loading BB parameters */
+	gain->offset_base[RTW89_PHY_0] =
+		rtw89_phy_read32_mask(rtwdev, R_P0_RPL1, B_P0_RPL1_BIAS_MASK);
+	gain->rssi_base[RTW89_PHY_0] =
+		rtw89_phy_read32_mask(rtwdev, R_P1_RPL1, B_P0_RPL1_BIAS_MASK);
+}
+
+static void rtw8852bx_bb_set_pop(struct rtw89_dev *rtwdev)
+{
+	if (rtwdev->hw->conf.flags & IEEE80211_CONF_MONITOR)
+		rtw89_phy_write32_clr(rtwdev, R_PKT_CTRL, B_PKT_POP_EN);
+}
+
+static
+void __rtw8852bx_set_channel_bb(struct rtw89_dev *rtwdev, const struct rtw89_chan *chan,
+				enum rtw89_phy_idx phy_idx)
+{
+	bool cck_en = chan->channel <= 14;
+	u8 pri_ch_idx = chan->pri_ch_idx;
+	u8 band = chan->band_type, chan_idx;
+
+	if (cck_en)
+		rtw8852bx_ctrl_sco_cck(rtwdev,  chan->primary_channel);
+
+	rtw8852bx_ctrl_ch(rtwdev, chan, phy_idx);
+	rtw8852bx_ctrl_bw(rtwdev, pri_ch_idx, chan->band_width, phy_idx);
+	rtw8852bx_ctrl_cck_en(rtwdev, cck_en);
+	if (chan->band_type == RTW89_BAND_5G) {
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_BT_SHARE_V1,
+				       B_PATH0_BT_SHARE_V1, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_BTG_PATH_V1,
+				       B_PATH0_BTG_PATH_V1, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_BT_SHARE_V1,
+				       B_PATH1_BT_SHARE_V1, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_BTG_PATH_V1,
+				       B_PATH1_BTG_PATH_V1, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_BT_SHARE, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_BT_SEG0, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_BT_DYN_DC_EST_EN_V1,
+				       B_BT_DYN_DC_EST_EN_MSK, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_GNT_BT_WGT_EN, B_GNT_BT_WGT_EN, 0x0);
+	}
+	chan_idx = rtw89_encode_chan_idx(rtwdev, chan->primary_channel, band);
+	rtw89_phy_write32_mask(rtwdev, R_MAC_PIN_SEL, B_CH_IDX_SEG0, chan_idx);
+	rtw8852bx_5m_mask(rtwdev, chan, phy_idx);
+	rtw8852bx_bb_set_pop(rtwdev);
+	__rtw8852bx_bb_reset_all(rtwdev, phy_idx);
+}
+
+static u32 rtw8852bx_bb_cal_txpwr_ref(struct rtw89_dev *rtwdev,
+				      enum rtw89_phy_idx phy_idx, s16 ref)
+{
+	const u16 tssi_16dbm_cw = 0x12c;
+	const u8 base_cw_0db = 0x27;
+	const s8 ofst_int = 0;
+	s16 pwr_s10_3;
+	s16 rf_pwr_cw;
+	u16 bb_pwr_cw;
+	u32 pwr_cw;
+	u32 tssi_ofst_cw;
+
+	pwr_s10_3 = (ref << 1) + (s16)(ofst_int) + (s16)(base_cw_0db << 3);
+	bb_pwr_cw = u16_get_bits(pwr_s10_3, GENMASK(2, 0));
+	rf_pwr_cw = u16_get_bits(pwr_s10_3, GENMASK(8, 3));
+	rf_pwr_cw = clamp_t(s16, rf_pwr_cw, 15, 63);
+	pwr_cw = (rf_pwr_cw << 3) | bb_pwr_cw;
+
+	tssi_ofst_cw = (u32)((s16)tssi_16dbm_cw + (ref << 1) - (16 << 3));
+	rtw89_debug(rtwdev, RTW89_DBG_TXPWR,
+		    "[TXPWR] tssi_ofst_cw=%d rf_cw=0x%x bb_cw=0x%x\n",
+		    tssi_ofst_cw, rf_pwr_cw, bb_pwr_cw);
+
+	return u32_encode_bits(tssi_ofst_cw, B_DPD_TSSI_CW) |
+	       u32_encode_bits(pwr_cw, B_DPD_PWR_CW) |
+	       u32_encode_bits(ref, B_DPD_REF);
+}
+
+static void rtw8852bx_set_txpwr_ref(struct rtw89_dev *rtwdev,
+				    enum rtw89_phy_idx phy_idx)
+{
+	static const u32 addr[RF_PATH_NUM_8852BX] = {0x5800, 0x7800};
+	const u32 mask = B_DPD_TSSI_CW | B_DPD_PWR_CW | B_DPD_REF;
+	const u8 ofst_ofdm = 0x4;
+	const u8 ofst_cck = 0x8;
+	const s16 ref_ofdm = 0;
+	const s16 ref_cck = 0;
+	u32 val;
+	u8 i;
+
+	rtw89_debug(rtwdev, RTW89_DBG_TXPWR, "[TXPWR] set txpwr reference\n");
+
+	rtw89_mac_txpwr_write32_mask(rtwdev, phy_idx, R_AX_PWR_RATE_CTRL,
+				     B_AX_PWR_REF, 0x0);
+
+	rtw89_debug(rtwdev, RTW89_DBG_TXPWR, "[TXPWR] set bb ofdm txpwr ref\n");
+	val = rtw8852bx_bb_cal_txpwr_ref(rtwdev, phy_idx, ref_ofdm);
+
+	for (i = 0; i < RF_PATH_NUM_8852BX; i++)
+		rtw89_phy_write32_idx(rtwdev, addr[i] + ofst_ofdm, mask, val,
+				      phy_idx);
+
+	rtw89_debug(rtwdev, RTW89_DBG_TXPWR, "[TXPWR] set bb cck txpwr ref\n");
+	val = rtw8852bx_bb_cal_txpwr_ref(rtwdev, phy_idx, ref_cck);
+
+	for (i = 0; i < RF_PATH_NUM_8852BX; i++)
+		rtw89_phy_write32_idx(rtwdev, addr[i] + ofst_cck, mask, val,
+				      phy_idx);
+}
+
+static void rtw8852bx_bb_set_tx_shape_dfir(struct rtw89_dev *rtwdev,
+					   const struct rtw89_chan *chan,
+					   u8 tx_shape_idx,
+					   enum rtw89_phy_idx phy_idx)
+{
+#define __DFIR_CFG_ADDR(i) (R_TXFIR0 + ((i) << 2))
+#define __DFIR_CFG_MASK 0xffffffff
+#define __DFIR_CFG_NR 8
+#define __DECL_DFIR_PARAM(_name, _val...) \
+	static const u32 param_ ## _name[] = {_val}; \
+	static_assert(ARRAY_SIZE(param_ ## _name) == __DFIR_CFG_NR)
+
+	__DECL_DFIR_PARAM(flat,
+			  0x023D23FF, 0x0029B354, 0x000FC1C8, 0x00FDB053,
+			  0x00F86F9A, 0x06FAEF92, 0x00FE5FCC, 0x00FFDFF5);
+	__DECL_DFIR_PARAM(sharp,
+			  0x023D83FF, 0x002C636A, 0x0013F204, 0x00008090,
+			  0x00F87FB0, 0x06F99F83, 0x00FDBFBA, 0x00003FF5);
+	__DECL_DFIR_PARAM(sharp_14,
+			  0x023B13FF, 0x001C42DE, 0x00FDB0AD, 0x00F60F6E,
+			  0x00FD8F92, 0x0602D011, 0x0001C02C, 0x00FFF00A);
+	u8 ch = chan->channel;
+	const u32 *param;
+	u32 addr;
+	int i;
+
+	if (ch > 14) {
+		rtw89_warn(rtwdev,
+			   "set tx shape dfir by unknown ch: %d on 2G\n", ch);
+		return;
+	}
+
+	if (ch == 14)
+		param = param_sharp_14;
+	else
+		param = tx_shape_idx == 0 ? param_flat : param_sharp;
+
+	for (i = 0; i < __DFIR_CFG_NR; i++) {
+		addr = __DFIR_CFG_ADDR(i);
+		rtw89_debug(rtwdev, RTW89_DBG_TXPWR,
+			    "set tx shape dfir: 0x%x: 0x%x\n", addr, param[i]);
+		rtw89_phy_write32_idx(rtwdev, addr, __DFIR_CFG_MASK, param[i],
+				      phy_idx);
+	}
+
+#undef __DECL_DFIR_PARAM
+#undef __DFIR_CFG_NR
+#undef __DFIR_CFG_MASK
+#undef __DECL_CFG_ADDR
+}
+
+static void rtw8852bx_set_tx_shape(struct rtw89_dev *rtwdev,
+				   const struct rtw89_chan *chan,
+				   enum rtw89_phy_idx phy_idx)
+{
+	const struct rtw89_rfe_parms *rfe_parms = rtwdev->rfe_parms;
+	u8 band = chan->band_type;
+	u8 regd = rtw89_regd_get(rtwdev, band);
+	u8 tx_shape_cck = (*rfe_parms->tx_shape.lmt)[band][RTW89_RS_CCK][regd];
+	u8 tx_shape_ofdm = (*rfe_parms->tx_shape.lmt)[band][RTW89_RS_OFDM][regd];
+
+	if (band == RTW89_BAND_2G)
+		rtw8852bx_bb_set_tx_shape_dfir(rtwdev, chan, tx_shape_cck, phy_idx);
+
+	rtw89_phy_write32_mask(rtwdev, R_DCFO_OPT, B_TXSHAPE_TRIANGULAR_CFG,
+			       tx_shape_ofdm);
+}
+
+static void __rtw8852bx_set_txpwr(struct rtw89_dev *rtwdev,
+				  const struct rtw89_chan *chan,
+				  enum rtw89_phy_idx phy_idx)
+{
+	rtw89_phy_set_txpwr_byrate(rtwdev, chan, phy_idx);
+	rtw89_phy_set_txpwr_offset(rtwdev, chan, phy_idx);
+	rtw8852bx_set_tx_shape(rtwdev, chan, phy_idx);
+	rtw89_phy_set_txpwr_limit(rtwdev, chan, phy_idx);
+	rtw89_phy_set_txpwr_limit_ru(rtwdev, chan, phy_idx);
+}
+
+static void __rtw8852bx_set_txpwr_ctrl(struct rtw89_dev *rtwdev,
+				       enum rtw89_phy_idx phy_idx)
+{
+	rtw8852bx_set_txpwr_ref(rtwdev, phy_idx);
+}
+
+static
+void __rtw8852bx_set_txpwr_ul_tb_offset(struct rtw89_dev *rtwdev,
+					s8 pw_ofst, enum rtw89_mac_idx mac_idx)
+{
+	u32 reg;
+
+	if (pw_ofst < -16 || pw_ofst > 15) {
+		rtw89_warn(rtwdev, "[ULTB] Err pwr_offset=%d\n", pw_ofst);
+		return;
+	}
+
+	reg = rtw89_mac_reg_by_idx(rtwdev, R_AX_PWR_UL_TB_CTRL, mac_idx);
+	rtw89_write32_set(rtwdev, reg, B_AX_PWR_UL_TB_CTRL_EN);
+
+	reg = rtw89_mac_reg_by_idx(rtwdev, R_AX_PWR_UL_TB_1T, mac_idx);
+	rtw89_write32_mask(rtwdev, reg, B_AX_PWR_UL_TB_1T_MASK, pw_ofst);
+
+	pw_ofst = max_t(s8, pw_ofst - 3, -16);
+	reg = rtw89_mac_reg_by_idx(rtwdev, R_AX_PWR_UL_TB_2T, mac_idx);
+	rtw89_write32_mask(rtwdev, reg, B_AX_PWR_UL_TB_2T_MASK, pw_ofst);
+}
+
+static int
+__rtw8852bx_init_txpwr_unit(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
+{
+	int ret;
+
+	ret = rtw89_mac_txpwr_write32(rtwdev, phy_idx, R_AX_PWR_UL_CTRL2, 0x07763333);
+	if (ret)
+		return ret;
+
+	ret = rtw89_mac_txpwr_write32(rtwdev, phy_idx, R_AX_PWR_COEXT_CTRL, 0x01ebf000);
+	if (ret)
+		return ret;
+
+	ret = rtw89_mac_txpwr_write32(rtwdev, phy_idx, R_AX_PWR_UL_CTRL0, 0x0002f8ff);
+	if (ret)
+		return ret;
+
+	rtw8852bx_set_txpwr_ul_tb_offset(rtwdev, 0, phy_idx == RTW89_PHY_1 ?
+						   RTW89_MAC_1 : RTW89_MAC_0);
+
+	return 0;
+}
+
+static
+void __rtw8852bx_bb_set_plcp_tx(struct rtw89_dev *rtwdev)
+{
+	const struct rtw89_reg3_def *def = rtw8852bx_pmac_ht20_mcs7_tbl;
+	u8 i;
+
+	for (i = 0; i < ARRAY_SIZE(rtw8852bx_pmac_ht20_mcs7_tbl); i++, def++)
+		rtw89_phy_write32_mask(rtwdev, def->addr, def->mask, def->data);
+}
+
+static void rtw8852bx_stop_pmac_tx(struct rtw89_dev *rtwdev,
+				   struct rtw8852bx_bb_pmac_info *tx_info,
+				   enum rtw89_phy_idx idx)
+{
+	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC Stop Tx");
+	if (tx_info->mode == CONT_TX)
+		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_PRD, B_PMAC_CTX_EN, 0, idx);
+	else if (tx_info->mode == PKTS_TX)
+		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_PRD, B_PMAC_PTX_EN, 0, idx);
+}
+
+static void rtw8852bx_start_pmac_tx(struct rtw89_dev *rtwdev,
+				    struct rtw8852bx_bb_pmac_info *tx_info,
+				    enum rtw89_phy_idx idx)
+{
+	enum rtw8852bx_pmac_mode mode = tx_info->mode;
+	u32 pkt_cnt = tx_info->tx_cnt;
+	u16 period = tx_info->period;
+
+	if (mode == CONT_TX && !tx_info->is_cck) {
+		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_PRD, B_PMAC_CTX_EN, 1, idx);
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC CTx Start");
+	} else if (mode == PKTS_TX) {
+		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_PRD, B_PMAC_PTX_EN, 1, idx);
+		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_PRD,
+				      B_PMAC_TX_PRD_MSK, period, idx);
+		rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_CNT, B_PMAC_TX_CNT_MSK,
+				      pkt_cnt, idx);
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC PTx Start");
+	}
+
+	rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_CTRL, B_PMAC_TXEN_DIS, 1, idx);
+	rtw89_phy_write32_idx(rtwdev, R_PMAC_TX_CTRL, B_PMAC_TXEN_DIS, 0, idx);
+}
+
+static
+void rtw8852bx_bb_set_pmac_tx(struct rtw89_dev *rtwdev,
+			      struct rtw8852bx_bb_pmac_info *tx_info,
+			      enum rtw89_phy_idx idx)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+
+	if (!tx_info->en_pmac_tx) {
+		rtw8852bx_stop_pmac_tx(rtwdev, tx_info, idx);
+		rtw89_phy_write32_idx(rtwdev, R_PD_CTRL, B_PD_HIT_DIS, 0, idx);
+		if (chan->band_type == RTW89_BAND_2G)
+			rtw89_phy_write32_clr(rtwdev, R_RXCCA, B_RXCCA_DIS);
+		return;
+	}
+
+	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC Tx Enable");
+
+	rtw89_phy_write32_idx(rtwdev, R_PMAC_GNT, B_PMAC_GNT_TXEN, 1, idx);
+	rtw89_phy_write32_idx(rtwdev, R_PMAC_GNT, B_PMAC_GNT_RXEN, 1, idx);
+	rtw89_phy_write32_idx(rtwdev, R_PMAC_RX_CFG1, B_PMAC_OPT1_MSK, 0x3f, idx);
+	rtw89_phy_write32_idx(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_ALL, 0, idx);
+	rtw89_phy_write32_idx(rtwdev, R_PD_CTRL, B_PD_HIT_DIS, 1, idx);
+	rtw89_phy_write32_set(rtwdev, R_RXCCA, B_RXCCA_DIS);
+	rtw89_phy_write32_idx(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_ALL, 1, idx);
+
+	rtw8852bx_start_pmac_tx(rtwdev, tx_info, idx);
+}
+
+static
+void __rtw8852bx_bb_set_pmac_pkt_tx(struct rtw89_dev *rtwdev, u8 enable,
+				    u16 tx_cnt, u16 period, u16 tx_time,
+				    enum rtw89_phy_idx idx)
+{
+	struct rtw8852bx_bb_pmac_info tx_info = {0};
+
+	tx_info.en_pmac_tx = enable;
+	tx_info.is_cck = 0;
+	tx_info.mode = PKTS_TX;
+	tx_info.tx_cnt = tx_cnt;
+	tx_info.period = period;
+	tx_info.tx_time = tx_time;
+
+	rtw8852bx_bb_set_pmac_tx(rtwdev, &tx_info, idx);
+}
+
+static
+void __rtw8852bx_bb_set_power(struct rtw89_dev *rtwdev, s16 pwr_dbm,
+			      enum rtw89_phy_idx idx)
+{
+	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC CFG Tx PWR = %d", pwr_dbm);
+
+	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_PWR_EN, 1, idx);
+	rtw89_phy_write32_idx(rtwdev, R_TXPWR, B_TXPWR_MSK, pwr_dbm, idx);
+}
+
+static
+void __rtw8852bx_bb_cfg_tx_path(struct rtw89_dev *rtwdev, u8 tx_path)
+{
+	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_MOD, 7, RTW89_PHY_0);
+
+	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "PMAC CFG Tx Path = %d", tx_path);
+
+	if (tx_path == RF_PATH_A) {
+		rtw89_phy_write32_mask(rtwdev, R_TXPATH_SEL, B_TXPATH_SEL_MSK, 1);
+		rtw89_phy_write32_mask(rtwdev, R_TXNSS_MAP, B_TXNSS_MAP_MSK, 0);
+	} else if (tx_path == RF_PATH_B) {
+		rtw89_phy_write32_mask(rtwdev, R_TXPATH_SEL, B_TXPATH_SEL_MSK, 2);
+		rtw89_phy_write32_mask(rtwdev, R_TXNSS_MAP, B_TXNSS_MAP_MSK, 0);
+	} else if (tx_path == RF_PATH_AB) {
+		rtw89_phy_write32_mask(rtwdev, R_TXPATH_SEL, B_TXPATH_SEL_MSK, 3);
+		rtw89_phy_write32_mask(rtwdev, R_TXNSS_MAP, B_TXNSS_MAP_MSK, 4);
+	} else {
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI, "Error Tx Path");
+	}
+}
+
+static
+void __rtw8852bx_bb_tx_mode_switch(struct rtw89_dev *rtwdev,
+				   enum rtw89_phy_idx idx, u8 mode)
+{
+	if (mode != 0)
+		return;
+
+	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "Tx mode switch");
+
+	rtw89_phy_write32_idx(rtwdev, R_PMAC_GNT, B_PMAC_GNT_TXEN, 0, idx);
+	rtw89_phy_write32_idx(rtwdev, R_PMAC_GNT, B_PMAC_GNT_RXEN, 0, idx);
+	rtw89_phy_write32_idx(rtwdev, R_PMAC_RX_CFG1, B_PMAC_OPT1_MSK, 0, idx);
+	rtw89_phy_write32_idx(rtwdev, R_PMAC_RXMOD, B_PMAC_RXMOD_MSK, 0, idx);
+	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_DPD_EN, 0, idx);
+	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_MOD, 0, idx);
+	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_PWR_EN, 0, idx);
+}
+
+static
+void __rtw8852bx_bb_backup_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx idx,
+				struct rtw8852bx_bb_tssi_bak *bak)
+{
+	s32 tmp;
+
+	bak->tx_path = rtw89_phy_read32_idx(rtwdev, R_TXPATH_SEL, B_TXPATH_SEL_MSK, idx);
+	bak->rx_path = rtw89_phy_read32_idx(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0, idx);
+	bak->p0_rfmode = rtw89_phy_read32_idx(rtwdev, R_P0_RFMODE, MASKDWORD, idx);
+	bak->p0_rfmode_ftm = rtw89_phy_read32_idx(rtwdev, R_P0_RFMODE_FTM_RX, MASKDWORD, idx);
+	bak->p1_rfmode = rtw89_phy_read32_idx(rtwdev, R_P1_RFMODE, MASKDWORD, idx);
+	bak->p1_rfmode_ftm = rtw89_phy_read32_idx(rtwdev, R_P1_RFMODE_FTM_RX, MASKDWORD, idx);
+	tmp = rtw89_phy_read32_idx(rtwdev, R_TXPWR, B_TXPWR_MSK, idx);
+	bak->tx_pwr = sign_extend32(tmp, 8);
+}
+
+static
+void __rtw8852bx_bb_restore_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx idx,
+				 const struct rtw8852bx_bb_tssi_bak *bak)
+{
+	rtw89_phy_write32_idx(rtwdev, R_TXPATH_SEL, B_TXPATH_SEL_MSK, bak->tx_path, idx);
+	if (bak->tx_path == RF_AB)
+		rtw89_phy_write32_mask(rtwdev, R_TXNSS_MAP, B_TXNSS_MAP_MSK, 0x4);
+	else
+		rtw89_phy_write32_mask(rtwdev, R_TXNSS_MAP, B_TXNSS_MAP_MSK, 0x0);
+	rtw89_phy_write32_idx(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0, bak->rx_path, idx);
+	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_PWR_EN, 1, idx);
+	rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE, MASKDWORD, bak->p0_rfmode, idx);
+	rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_FTM_RX, MASKDWORD, bak->p0_rfmode_ftm, idx);
+	rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE, MASKDWORD, bak->p1_rfmode, idx);
+	rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_FTM_RX, MASKDWORD, bak->p1_rfmode_ftm, idx);
+	rtw89_phy_write32_idx(rtwdev, R_TXPWR, B_TXPWR_MSK, bak->tx_pwr, idx);
+}
+
+static void __rtw8852bx_ctrl_nbtg_bt_tx(struct rtw89_dev *rtwdev, bool en,
+					enum rtw89_phy_idx phy_idx)
+{
+	rtw89_phy_write_reg3_tbl(rtwdev, en ? &rtw8852bx_btc_preagc_en_defs_tbl :
+						 &rtw8852bx_btc_preagc_dis_defs_tbl);
+}
+
+static void __rtw8852bx_ctrl_btg_bt_rx(struct rtw89_dev *rtwdev, bool en,
+				       enum rtw89_phy_idx phy_idx)
+{
+	if (en) {
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_BT_SHARE_V1,
+				       B_PATH0_BT_SHARE_V1, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_BTG_PATH_V1,
+				       B_PATH0_BTG_PATH_V1, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_G_LNA6_OP1DB_V1,
+				       B_PATH1_G_LNA6_OP1DB_V1, 0x20);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_G_TIA0_LNA6_OP1DB_V1,
+				       B_PATH1_G_TIA0_LNA6_OP1DB_V1, 0x30);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_BT_SHARE_V1,
+				       B_PATH1_BT_SHARE_V1, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_BTG_PATH_V1,
+				       B_PATH1_BTG_PATH_V1, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_PMAC_GNT, B_PMAC_GNT_P1, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_BT_SHARE, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_BT_SEG0, 0x2);
+		rtw89_phy_write32_mask(rtwdev, R_BT_DYN_DC_EST_EN_V1,
+				       B_BT_DYN_DC_EST_EN_MSK, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_GNT_BT_WGT_EN, B_GNT_BT_WGT_EN, 0x1);
+	} else {
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_BT_SHARE_V1,
+				       B_PATH0_BT_SHARE_V1, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_BTG_PATH_V1,
+				       B_PATH0_BTG_PATH_V1, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_G_LNA6_OP1DB_V1,
+				       B_PATH1_G_LNA6_OP1DB_V1, 0x1a);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_G_TIA0_LNA6_OP1DB_V1,
+				       B_PATH1_G_TIA0_LNA6_OP1DB_V1, 0x2a);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_BT_SHARE_V1,
+				       B_PATH1_BT_SHARE_V1, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_BTG_PATH_V1,
+				       B_PATH1_BTG_PATH_V1, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PMAC_GNT, B_PMAC_GNT_P1, 0xc);
+		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_BT_SHARE, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_BT_SEG0, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_BT_DYN_DC_EST_EN_V1,
+				       B_BT_DYN_DC_EST_EN_MSK, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_GNT_BT_WGT_EN, B_GNT_BT_WGT_EN, 0x0);
+	}
+}
+
+static
+void __rtw8852bx_bb_ctrl_rx_path(struct rtw89_dev *rtwdev,
+				 enum rtw89_rf_path_bit rx_path)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	u32 rst_mask0;
+	u32 rst_mask1;
+
+	if (rx_path == RF_A) {
+		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0, 1);
+		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG0, 1);
+		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG1, 1);
+		rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 0);
+		rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 0);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_USER_MAX, 4);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 0);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHETB_MAX_NSS, 0);
+	} else if (rx_path == RF_B) {
+		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0, 2);
+		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG0, 2);
+		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG1, 2);
+		rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 0);
+		rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 0);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_USER_MAX, 4);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 0);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHETB_MAX_NSS, 0);
+	} else if (rx_path == RF_AB) {
+		rtw89_phy_write32_mask(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0, 3);
+		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG0, 3);
+		rtw89_phy_write32_mask(rtwdev, R_FC0_BW_V1, B_ANT_RX_1RCCA_SEG1, 3);
+		rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 1);
+		rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 1);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_USER_MAX, 4);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 1);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHETB_MAX_NSS, 1);
+	}
+
+	rtw8852bx_set_gain_offset(rtwdev, chan->subband_type, RTW89_PHY_0);
+
+	if (chan->band_type == RTW89_BAND_2G &&
+	    (rx_path == RF_B || rx_path == RF_AB))
+		rtw8852bx_ctrl_btg_bt_rx(rtwdev, true, RTW89_PHY_0);
+	else
+		rtw8852bx_ctrl_btg_bt_rx(rtwdev, false, RTW89_PHY_0);
+
+	rst_mask0 = B_P0_TXPW_RSTB_MANON | B_P0_TXPW_RSTB_TSSI;
+	rst_mask1 = B_P1_TXPW_RSTB_MANON | B_P1_TXPW_RSTB_TSSI;
+	if (rx_path == RF_A) {
+		rtw89_phy_write32_mask(rtwdev, R_P0_TXPW_RSTB, rst_mask0, 1);
+		rtw89_phy_write32_mask(rtwdev, R_P0_TXPW_RSTB, rst_mask0, 3);
+	} else {
+		rtw89_phy_write32_mask(rtwdev, R_P1_TXPW_RSTB, rst_mask1, 1);
+		rtw89_phy_write32_mask(rtwdev, R_P1_TXPW_RSTB, rst_mask1, 3);
+	}
+}
+
+static void rtw8852bx_bb_ctrl_rf_mode_rx_path(struct rtw89_dev *rtwdev,
+					      enum rtw89_rf_path_bit rx_path)
+{
+	if (rx_path == RF_A) {
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE,
+				       B_P0_RFMODE_ORI_TXRX_FTM_TX, 0x1233312);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE_FTM_RX,
+				       B_P0_RFMODE_FTM_RX, 0x333);
+		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE,
+				       B_P1_RFMODE_ORI_TXRX_FTM_TX, 0x1111111);
+		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE_FTM_RX,
+				       B_P1_RFMODE_FTM_RX, 0x111);
+	} else if (rx_path == RF_B) {
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE,
+				       B_P0_RFMODE_ORI_TXRX_FTM_TX, 0x1111111);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE_FTM_RX,
+				       B_P0_RFMODE_FTM_RX, 0x111);
+		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE,
+				       B_P1_RFMODE_ORI_TXRX_FTM_TX, 0x1233312);
+		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE_FTM_RX,
+				       B_P1_RFMODE_FTM_RX, 0x333);
+	} else if (rx_path == RF_AB) {
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE,
+				       B_P0_RFMODE_ORI_TXRX_FTM_TX, 0x1233312);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFMODE_FTM_RX,
+				       B_P0_RFMODE_FTM_RX, 0x333);
+		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE,
+				       B_P1_RFMODE_ORI_TXRX_FTM_TX, 0x1233312);
+		rtw89_phy_write32_mask(rtwdev, R_P1_RFMODE_FTM_RX,
+				       B_P1_RFMODE_FTM_RX, 0x333);
+	}
+}
+
+static void __rtw8852bx_bb_cfg_txrx_path(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_hal *hal = &rtwdev->hal;
+	enum rtw89_rf_path_bit rx_path = hal->antenna_rx ? hal->antenna_rx : RF_AB;
+
+	rtw8852bx_bb_ctrl_rx_path(rtwdev, rx_path);
+	rtw8852bx_bb_ctrl_rf_mode_rx_path(rtwdev, rx_path);
+
+	if (rtwdev->hal.rx_nss == 1) {
+		rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 0);
+		rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 0);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 0);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHETB_MAX_NSS, 0);
+	} else {
+		rtw89_phy_write32_mask(rtwdev, R_RXHT_MCS_LIMIT, B_RXHT_MCS_LIMIT, 1);
+		rtw89_phy_write32_mask(rtwdev, R_RXVHT_MCS_LIMIT, B_RXVHT_MCS_LIMIT, 1);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHE_MAX_NSS, 1);
+		rtw89_phy_write32_mask(rtwdev, R_RXHE, B_RXHETB_MAX_NSS, 1);
+	}
+
+	rtw89_phy_write32_idx(rtwdev, R_MAC_SEL, B_MAC_SEL_MOD, 0x0, RTW89_PHY_0);
+}
+
+static u8 __rtw8852bx_get_thermal(struct rtw89_dev *rtwdev, enum rtw89_rf_path rf_path)
+{
+	if (rtwdev->is_tssi_mode[rf_path]) {
+		u32 addr = 0x1c10 + (rf_path << 13);
+
+		return rtw89_phy_read32_mask(rtwdev, addr, 0x3F000000);
+	}
+
+	rtw89_write_rf(rtwdev, rf_path, RR_TM, RR_TM_TRI, 0x1);
+	rtw89_write_rf(rtwdev, rf_path, RR_TM, RR_TM_TRI, 0x0);
+	rtw89_write_rf(rtwdev, rf_path, RR_TM, RR_TM_TRI, 0x1);
+
+	fsleep(200);
+
+	return rtw89_read_rf(rtwdev, rf_path, RR_TM, RR_TM_VAL);
+}
+
+static void rtw8852bx_fill_freq_with_ppdu(struct rtw89_dev *rtwdev,
+					  struct rtw89_rx_phy_ppdu *phy_ppdu,
+					  struct ieee80211_rx_status *status)
+{
+	u16 chan = phy_ppdu->chan_idx;
+	enum nl80211_band band;
+	u8 ch;
+
+	if (chan == 0)
+		return;
+
+	rtw89_decode_chan_idx(rtwdev, chan, &ch, &band);
+	status->freq = ieee80211_channel_to_frequency(ch, band);
+	status->band = band;
+}
+
+static void __rtw8852bx_query_ppdu(struct rtw89_dev *rtwdev,
+				   struct rtw89_rx_phy_ppdu *phy_ppdu,
+				   struct ieee80211_rx_status *status)
+{
+	u8 path;
+	u8 *rx_power = phy_ppdu->rssi;
+
+	status->signal = RTW89_RSSI_RAW_TO_DBM(max(rx_power[RF_PATH_A], rx_power[RF_PATH_B]));
+	for (path = 0; path < rtwdev->chip->rf_path_num; path++) {
+		status->chains |= BIT(path);
+		status->chain_signal[path] = RTW89_RSSI_RAW_TO_DBM(rx_power[path]);
+	}
+	if (phy_ppdu->valid)
+		rtw8852bx_fill_freq_with_ppdu(rtwdev, phy_ppdu, status);
+}
+
+static int __rtw8852bx_mac_enable_bb_rf(struct rtw89_dev *rtwdev)
+{
+	int ret;
+
+	rtw89_write8_set(rtwdev, R_AX_SYS_FUNC_EN,
+			 B_AX_FEN_BBRSTB | B_AX_FEN_BB_GLB_RSTN);
+	rtw89_write32_mask(rtwdev, R_AX_SPS_DIG_ON_CTRL0, B_AX_REG_ZCDC_H_MASK, 0x1);
+	rtw89_write32_set(rtwdev, R_AX_WLRF_CTRL, B_AX_AFC_AFEDIG);
+	rtw89_write32_clr(rtwdev, R_AX_WLRF_CTRL, B_AX_AFC_AFEDIG);
+	rtw89_write32_set(rtwdev, R_AX_WLRF_CTRL, B_AX_AFC_AFEDIG);
+
+	ret = rtw89_mac_write_xtal_si(rtwdev, XTAL_SI_WL_RFC_S0, 0xC7,
+				      FULL_BIT_MASK);
+	if (ret)
+		return ret;
+
+	ret = rtw89_mac_write_xtal_si(rtwdev, XTAL_SI_WL_RFC_S1, 0xC7,
+				      FULL_BIT_MASK);
+	if (ret)
+		return ret;
+
+	rtw89_write8(rtwdev, R_AX_PHYREG_SET, PHYREG_SET_XYN_CYCLE);
+
+	return 0;
+}
+
+static int __rtw8852bx_mac_disable_bb_rf(struct rtw89_dev *rtwdev)
+{
+	u8 wl_rfc_s0;
+	u8 wl_rfc_s1;
+	int ret;
+
+	rtw89_write32_clr(rtwdev, R_AX_WLRF_CTRL, B_AX_AFC_AFEDIG);
+	rtw89_write8_clr(rtwdev, R_AX_SYS_FUNC_EN,
+			 B_AX_FEN_BBRSTB | B_AX_FEN_BB_GLB_RSTN);
+
+	ret = rtw89_mac_read_xtal_si(rtwdev, XTAL_SI_WL_RFC_S0, &wl_rfc_s0);
+	if (ret)
+		return ret;
+	wl_rfc_s0 &= ~XTAL_SI_RF00S_EN;
+	ret = rtw89_mac_write_xtal_si(rtwdev, XTAL_SI_WL_RFC_S0, wl_rfc_s0,
+				      FULL_BIT_MASK);
+	if (ret)
+		return ret;
+
+	ret = rtw89_mac_read_xtal_si(rtwdev, XTAL_SI_WL_RFC_S1, &wl_rfc_s1);
+	if (ret)
+		return ret;
+	wl_rfc_s1 &= ~XTAL_SI_RF10S_EN;
+	ret = rtw89_mac_write_xtal_si(rtwdev, XTAL_SI_WL_RFC_S1, wl_rfc_s1,
+				      FULL_BIT_MASK);
+	return ret;
+}
+
+const struct rtw8852bx_info rtw8852bx_info = {
+	.mac_enable_bb_rf = __rtw8852bx_mac_enable_bb_rf,
+	.mac_disable_bb_rf = __rtw8852bx_mac_disable_bb_rf,
+	.bb_sethw = __rtw8852bx_bb_sethw,
+	.bb_reset_all = __rtw8852bx_bb_reset_all,
+	.bb_cfg_txrx_path = __rtw8852bx_bb_cfg_txrx_path,
+	.bb_cfg_tx_path = __rtw8852bx_bb_cfg_tx_path,
+	.bb_ctrl_rx_path = __rtw8852bx_bb_ctrl_rx_path,
+	.bb_set_plcp_tx = __rtw8852bx_bb_set_plcp_tx,
+	.bb_set_power = __rtw8852bx_bb_set_power,
+	.bb_set_pmac_pkt_tx = __rtw8852bx_bb_set_pmac_pkt_tx,
+	.bb_backup_tssi = __rtw8852bx_bb_backup_tssi,
+	.bb_restore_tssi = __rtw8852bx_bb_restore_tssi,
+	.bb_tx_mode_switch = __rtw8852bx_bb_tx_mode_switch,
+	.set_channel_mac = __rtw8852bx_set_channel_mac,
+	.set_channel_bb = __rtw8852bx_set_channel_bb,
+	.ctrl_nbtg_bt_tx = __rtw8852bx_ctrl_nbtg_bt_tx,
+	.ctrl_btg_bt_rx = __rtw8852bx_ctrl_btg_bt_rx,
+	.query_ppdu = __rtw8852bx_query_ppdu,
+	.read_efuse = __rtw8852bx_read_efuse,
+	.read_phycap = __rtw8852bx_read_phycap,
+	.power_trim = __rtw8852bx_power_trim,
+	.set_txpwr = __rtw8852bx_set_txpwr,
+	.set_txpwr_ctrl = __rtw8852bx_set_txpwr_ctrl,
+	.init_txpwr_unit = __rtw8852bx_init_txpwr_unit,
+	.set_txpwr_ul_tb_offset = __rtw8852bx_set_txpwr_ul_tb_offset,
+	.get_thermal = __rtw8852bx_get_thermal,
+};
+EXPORT_SYMBOL(rtw8852bx_info);
+
+MODULE_AUTHOR("Realtek Corporation");
+MODULE_DESCRIPTION("Realtek 802.11ax wireless 8852B common routines");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.h b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.h
new file mode 100644
index 000000000000..2348a13545ac
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.h
@@ -0,0 +1,339 @@
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+/* Copyright(c) 2024  Realtek Corporation
+ */
+
+#ifndef __RTW89_8852BX_H__
+#define __RTW89_8852BX_H__
+
+#include "core.h"
+
+#define RF_PATH_NUM_8852BX 2
+#define BB_PATH_NUM_8852BX 2
+
+enum rtw8852bx_pmac_mode {
+	NONE_TEST,
+	PKTS_TX,
+	PKTS_RX,
+	CONT_TX
+};
+
+struct rtw8852bx_u_efuse {
+	u8 rsvd[0x88];
+	u8 mac_addr[ETH_ALEN];
+};
+
+struct rtw8852bx_e_efuse {
+	u8 mac_addr[ETH_ALEN];
+};
+
+struct rtw8852bx_tssi_offset {
+	u8 cck_tssi[TSSI_CCK_CH_GROUP_NUM];
+	u8 bw40_tssi[TSSI_MCS_2G_CH_GROUP_NUM];
+	u8 rsvd[7];
+	u8 bw40_1s_tssi_5g[TSSI_MCS_5G_CH_GROUP_NUM];
+} __packed;
+
+struct rtw8852bx_efuse {
+	u8 rsvd[0x210];
+	struct rtw8852bx_tssi_offset path_a_tssi;
+	u8 rsvd1[10];
+	struct rtw8852bx_tssi_offset path_b_tssi;
+	u8 rsvd2[94];
+	u8 channel_plan;
+	u8 xtal_k;
+	u8 rsvd3;
+	u8 iqk_lck;
+	u8 rsvd4[5];
+	u8 reg_setting:2;
+	u8 tx_diversity:1;
+	u8 rx_diversity:2;
+	u8 ac_mode:1;
+	u8 module_type:2;
+	u8 rsvd5;
+	u8 shared_ant:1;
+	u8 coex_type:3;
+	u8 ant_iso:1;
+	u8 radio_on_off:1;
+	u8 rsvd6:2;
+	u8 eeprom_version;
+	u8 customer_id;
+	u8 tx_bb_swing_2g;
+	u8 tx_bb_swing_5g;
+	u8 tx_cali_pwr_trk_mode;
+	u8 trx_path_selection;
+	u8 rfe_type;
+	u8 country_code[2];
+	u8 rsvd7[3];
+	u8 path_a_therm;
+	u8 path_b_therm;
+	u8 rsvd8[2];
+	u8 rx_gain_2g_ofdm;
+	u8 rsvd9;
+	u8 rx_gain_2g_cck;
+	u8 rsvd10;
+	u8 rx_gain_5g_low;
+	u8 rsvd11;
+	u8 rx_gain_5g_mid;
+	u8 rsvd12;
+	u8 rx_gain_5g_high;
+	u8 rsvd13[35];
+	u8 path_a_cck_pwr_idx[6];
+	u8 path_a_bw40_1tx_pwr_idx[5];
+	u8 path_a_ofdm_1tx_pwr_idx_diff:4;
+	u8 path_a_bw20_1tx_pwr_idx_diff:4;
+	u8 path_a_bw20_2tx_pwr_idx_diff:4;
+	u8 path_a_bw40_2tx_pwr_idx_diff:4;
+	u8 path_a_cck_2tx_pwr_idx_diff:4;
+	u8 path_a_ofdm_2tx_pwr_idx_diff:4;
+	u8 rsvd14[0xf2];
+	union {
+		struct rtw8852bx_u_efuse u;
+		struct rtw8852bx_e_efuse e;
+	};
+} __packed;
+
+struct rtw8852bx_bb_pmac_info {
+	u8 en_pmac_tx:1;
+	u8 is_cck:1;
+	u8 mode:3;
+	u8 rsvd:3;
+	u16 tx_cnt;
+	u16 period;
+	u16 tx_time;
+	u8 duty_cycle;
+};
+
+struct rtw8852bx_bb_tssi_bak {
+	u8 tx_path;
+	u8 rx_path;
+	u32 p0_rfmode;
+	u32 p0_rfmode_ftm;
+	u32 p1_rfmode;
+	u32 p1_rfmode_ftm;
+	s16 tx_pwr; /* S9 */
+};
+
+struct rtw8852bx_info {
+	int (*mac_enable_bb_rf)(struct rtw89_dev *rtwdev);
+	int (*mac_disable_bb_rf)(struct rtw89_dev *rtwdev);
+	void (*bb_sethw)(struct rtw89_dev *rtwdev);
+	void (*bb_reset_all)(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx);
+	void (*bb_cfg_txrx_path)(struct rtw89_dev *rtwdev);
+	void (*bb_cfg_tx_path)(struct rtw89_dev *rtwdev, u8 tx_path);
+	void (*bb_ctrl_rx_path)(struct rtw89_dev *rtwdev,
+				enum rtw89_rf_path_bit rx_path);
+	void (*bb_set_plcp_tx)(struct rtw89_dev *rtwdev);
+	void (*bb_set_power)(struct rtw89_dev *rtwdev, s16 pwr_dbm,
+			     enum rtw89_phy_idx idx);
+	void (*bb_set_pmac_pkt_tx)(struct rtw89_dev *rtwdev, u8 enable,
+				   u16 tx_cnt, u16 period, u16 tx_time,
+				   enum rtw89_phy_idx idx);
+	void (*bb_backup_tssi)(struct rtw89_dev *rtwdev, enum rtw89_phy_idx idx,
+			       struct rtw8852bx_bb_tssi_bak *bak);
+	void (*bb_restore_tssi)(struct rtw89_dev *rtwdev, enum rtw89_phy_idx idx,
+				const struct rtw8852bx_bb_tssi_bak *bak);
+	void (*bb_tx_mode_switch)(struct rtw89_dev *rtwdev,
+				  enum rtw89_phy_idx idx, u8 mode);
+	void (*set_channel_mac)(struct rtw89_dev *rtwdev,
+				const struct rtw89_chan *chan, u8 mac_idx);
+	void (*set_channel_bb)(struct rtw89_dev *rtwdev, const struct rtw89_chan *chan,
+			       enum rtw89_phy_idx phy_idx);
+	void (*ctrl_nbtg_bt_tx)(struct rtw89_dev *rtwdev, bool en,
+				enum rtw89_phy_idx phy_idx);
+	void (*ctrl_btg_bt_rx)(struct rtw89_dev *rtwdev, bool en,
+			       enum rtw89_phy_idx phy_idx);
+	void (*query_ppdu)(struct rtw89_dev *rtwdev,
+			   struct rtw89_rx_phy_ppdu *phy_ppdu,
+			   struct ieee80211_rx_status *status);
+	int (*read_efuse)(struct rtw89_dev *rtwdev, u8 *log_map,
+			  enum rtw89_efuse_block block);
+	int (*read_phycap)(struct rtw89_dev *rtwdev, u8 *phycap_map);
+	void (*power_trim)(struct rtw89_dev *rtwdev);
+	void (*set_txpwr)(struct rtw89_dev *rtwdev,
+			  const struct rtw89_chan *chan,
+			  enum rtw89_phy_idx phy_idx);
+	void (*set_txpwr_ctrl)(struct rtw89_dev *rtwdev,
+			       enum rtw89_phy_idx phy_idx);
+	int (*init_txpwr_unit)(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx);
+	void (*set_txpwr_ul_tb_offset)(struct rtw89_dev *rtwdev,
+				       s8 pw_ofst, enum rtw89_mac_idx mac_idx);
+	u8 (*get_thermal)(struct rtw89_dev *rtwdev, enum rtw89_rf_path rf_path);
+};
+
+extern const struct rtw8852bx_info rtw8852bx_info;
+
+static inline
+int rtw8852bx_mac_enable_bb_rf(struct rtw89_dev *rtwdev)
+{
+	return rtw8852bx_info.mac_enable_bb_rf(rtwdev);
+}
+
+static inline
+int rtw8852bx_mac_disable_bb_rf(struct rtw89_dev *rtwdev)
+{
+	return rtw8852bx_info.mac_disable_bb_rf(rtwdev);
+}
+
+static inline
+void rtw8852bx_bb_sethw(struct rtw89_dev *rtwdev)
+{
+	rtw8852bx_info.bb_sethw(rtwdev);
+}
+
+static inline
+void rtw8852bx_bb_reset_all(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
+{
+	rtw8852bx_info.bb_reset_all(rtwdev, phy_idx);
+}
+
+static inline
+void rtw8852bx_bb_cfg_txrx_path(struct rtw89_dev *rtwdev)
+{
+	rtw8852bx_info.bb_cfg_txrx_path(rtwdev);
+}
+
+static inline
+void rtw8852bx_bb_cfg_tx_path(struct rtw89_dev *rtwdev, u8 tx_path)
+{
+	rtw8852bx_info.bb_cfg_tx_path(rtwdev, tx_path);
+}
+
+static inline
+void rtw8852bx_bb_ctrl_rx_path(struct rtw89_dev *rtwdev,
+			       enum rtw89_rf_path_bit rx_path)
+{
+	rtw8852bx_info.bb_ctrl_rx_path(rtwdev, rx_path);
+}
+
+static inline
+void rtw8852bx_bb_set_plcp_tx(struct rtw89_dev *rtwdev)
+{
+	rtw8852bx_info.bb_set_plcp_tx(rtwdev);
+}
+
+static inline
+void rtw8852bx_bb_set_power(struct rtw89_dev *rtwdev, s16 pwr_dbm,
+			    enum rtw89_phy_idx idx)
+{
+	rtw8852bx_info.bb_set_power(rtwdev, pwr_dbm, idx);
+}
+
+static inline
+void rtw8852bx_bb_set_pmac_pkt_tx(struct rtw89_dev *rtwdev, u8 enable,
+				  u16 tx_cnt, u16 period, u16 tx_time,
+				  enum rtw89_phy_idx idx)
+{
+	rtw8852bx_info.bb_set_pmac_pkt_tx(rtwdev, enable, tx_cnt, period, tx_time, idx);
+}
+
+static inline
+void rtw8852bx_bb_backup_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx idx,
+			      struct rtw8852bx_bb_tssi_bak *bak)
+{
+	rtw8852bx_info.bb_backup_tssi(rtwdev, idx, bak);
+}
+
+static inline
+void rtw8852bx_bb_restore_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx idx,
+			       const struct rtw8852bx_bb_tssi_bak *bak)
+{
+	rtw8852bx_info.bb_restore_tssi(rtwdev, idx, bak);
+}
+
+static inline
+void rtw8852bx_bb_tx_mode_switch(struct rtw89_dev *rtwdev,
+				 enum rtw89_phy_idx idx, u8 mode)
+{
+	rtw8852bx_info.bb_tx_mode_switch(rtwdev, idx, mode);
+}
+
+static inline
+void rtw8852bx_set_channel_mac(struct rtw89_dev *rtwdev,
+			       const struct rtw89_chan *chan, u8 mac_idx)
+{
+	rtw8852bx_info.set_channel_mac(rtwdev, chan, mac_idx);
+}
+
+static inline
+void rtw8852bx_set_channel_bb(struct rtw89_dev *rtwdev, const struct rtw89_chan *chan,
+			      enum rtw89_phy_idx phy_idx)
+{
+	rtw8852bx_info.set_channel_bb(rtwdev, chan, phy_idx);
+}
+
+static inline
+void rtw8852bx_ctrl_nbtg_bt_tx(struct rtw89_dev *rtwdev, bool en,
+			       enum rtw89_phy_idx phy_idx)
+{
+	rtw8852bx_info.ctrl_nbtg_bt_tx(rtwdev, en, phy_idx);
+}
+
+static inline
+void rtw8852bx_ctrl_btg_bt_rx(struct rtw89_dev *rtwdev, bool en,
+			      enum rtw89_phy_idx phy_idx)
+{
+	rtw8852bx_info.ctrl_btg_bt_rx(rtwdev, en, phy_idx);
+}
+
+static inline
+void rtw8852bx_query_ppdu(struct rtw89_dev *rtwdev,
+			  struct rtw89_rx_phy_ppdu *phy_ppdu,
+			  struct ieee80211_rx_status *status)
+{
+	rtw8852bx_info.query_ppdu(rtwdev, phy_ppdu, status);
+}
+
+static inline
+int rtw8852bx_read_efuse(struct rtw89_dev *rtwdev, u8 *log_map,
+			 enum rtw89_efuse_block block)
+{
+	return rtw8852bx_info.read_efuse(rtwdev, log_map, block);
+}
+
+static inline
+int rtw8852bx_read_phycap(struct rtw89_dev *rtwdev, u8 *phycap_map)
+{
+	return rtw8852bx_info.read_phycap(rtwdev, phycap_map);
+}
+
+static inline
+void rtw8852bx_power_trim(struct rtw89_dev *rtwdev)
+{
+	rtw8852bx_info.power_trim(rtwdev);
+}
+
+static inline
+void rtw8852bx_set_txpwr(struct rtw89_dev *rtwdev,
+			 const struct rtw89_chan *chan,
+			 enum rtw89_phy_idx phy_idx)
+{
+	rtw8852bx_info.set_txpwr(rtwdev, chan, phy_idx);
+}
+
+static inline
+void rtw8852bx_set_txpwr_ctrl(struct rtw89_dev *rtwdev,
+			      enum rtw89_phy_idx phy_idx)
+{
+	rtw8852bx_info.set_txpwr_ctrl(rtwdev, phy_idx);
+}
+
+static inline
+int rtw8852bx_init_txpwr_unit(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
+{
+	return rtw8852bx_info.init_txpwr_unit(rtwdev, phy_idx);
+}
+
+static inline
+void rtw8852bx_set_txpwr_ul_tb_offset(struct rtw89_dev *rtwdev,
+				      s8 pw_ofst, enum rtw89_mac_idx mac_idx)
+{
+	rtw8852bx_info.set_txpwr_ul_tb_offset(rtwdev, pw_ofst, mac_idx);
+}
+
+static inline
+u8 rtw8852bx_get_thermal(struct rtw89_dev *rtwdev, enum rtw89_rf_path rf_path)
+{
+	return rtw8852bx_info.get_thermal(rtwdev, rf_path);
+}
+
+#endif
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c
index 259df67836a0..7c90e98dc1d9 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c
@@ -8,6 +8,7 @@
 #include "phy.h"
 #include "reg.h"
 #include "rtw8852b.h"
+#include "rtw8852b_common.h"
 #include "rtw8852b_rfk.h"
 #include "rtw8852b_rfk_table.h"
 #include "rtw8852b_table.h"
@@ -3433,13 +3434,13 @@ static void _tssi_hw_tx(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
 		rx_path = RF_ABCD; /* don't change path, but still set others */
 
 	if (enable) {
-		rtw8852b_bb_set_plcp_tx(rtwdev);
-		rtw8852b_bb_cfg_tx_path(rtwdev, path);
-		rtw8852b_bb_ctrl_rx_path(rtwdev, rx_path);
-		rtw8852b_bb_set_power(rtwdev, pwr_dbm, phy);
+		rtw8852bx_bb_set_plcp_tx(rtwdev);
+		rtw8852bx_bb_cfg_tx_path(rtwdev, path);
+		rtw8852bx_bb_ctrl_rx_path(rtwdev, rx_path);
+		rtw8852bx_bb_set_power(rtwdev, pwr_dbm, phy);
 	}
 
-	rtw8852b_bb_set_pmac_pkt_tx(rtwdev, enable, cnt, period, 20, phy);
+	rtw8852bx_bb_set_pmac_pkt_tx(rtwdev, enable, cnt, period, 20, phy);
 }
 
 static void _tssi_backup_bb_registers(struct rtw89_dev *rtwdev,
@@ -3578,7 +3579,7 @@ static void _tssi_alimentk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
 	u32 tssi_cw_rpt[RTW8852B_TSSI_PATH_NR] = {0};
 	u8 channel = chan->channel;
 	u8 ch_idx = _tssi_ch_to_idx(rtwdev, channel);
-	struct rtw8852b_bb_tssi_bak tssi_bak;
+	struct rtw8852bx_bb_tssi_bak tssi_bak;
 	s32 aliment_diff, tssi_cw_default;
 	u32 start_time, finish_time;
 	u32 bb_reg_backup[8] = {0};
@@ -3626,7 +3627,7 @@ static void _tssi_alimentk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
 	else
 		band = TSSI_ALIMK_2G;
 
-	rtw8852b_bb_backup_tssi(rtwdev, phy, &tssi_bak);
+	rtw8852bx_bb_backup_tssi(rtwdev, phy, &tssi_bak);
 	_tssi_backup_bb_registers(rtwdev, phy, bb_reg, bb_reg_backup, ARRAY_SIZE(bb_reg_backup));
 
 	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_AVG, B_P0_TSSI_AVG, 0x8);
@@ -3730,8 +3731,8 @@ static void _tssi_alimentk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
 
 out:
 	_tssi_reload_bb_registers(rtwdev, phy, bb_reg, bb_reg_backup, ARRAY_SIZE(bb_reg_backup));
-	rtw8852b_bb_restore_tssi(rtwdev, phy, &tssi_bak);
-	rtw8852b_bb_tx_mode_switch(rtwdev, phy, 0);
+	rtw8852bx_bb_restore_tssi(rtwdev, phy, &tssi_bak);
+	rtw8852bx_bb_tx_mode_switch(rtwdev, phy, 0);
 
 	finish_time = ktime_get_ns();
 	tssi_info->tssi_alimk_time += finish_time - start_time;
-- 
cgit v1.2.3-58-ga151


From 2ab298cf638c105e248156d7a7b2301f071f6543 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 7 Jun 2024 15:06:58 +0800
Subject: wifi: rtw89: 8852bx: add extra handles for 8852BT in 8852b_common

The channel configuration of 8852BT is very similar but a little different
to 8852B, so use chip ID as condition to add extra handles including
external loss compensation, ADC configurations, spur settings and so on.

Don't affect existing 8852BE.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240607070659.80263-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/Makefile        |   3 +-
 drivers/net/wireless/realtek/rtw89/core.h          |   1 +
 drivers/net/wireless/realtek/rtw89/phy.h           |   1 +
 drivers/net/wireless/realtek/rtw89/reg.h           |   6 +
 .../net/wireless/realtek/rtw89/rtw8852b_common.c   | 195 ++++++++++++++++++++-
 .../net/wireless/realtek/rtw89/rtw8852b_common.h   |   7 +
 drivers/net/wireless/realtek/rtw89/util.c          | 106 +++++++++++
 drivers/net/wireless/realtek/rtw89/util.h          |   5 +
 8 files changed, 315 insertions(+), 9 deletions(-)
 create mode 100644 drivers/net/wireless/realtek/rtw89/util.c

diff --git a/drivers/net/wireless/realtek/rtw89/Makefile b/drivers/net/wireless/realtek/rtw89/Makefile
index 688847ce7084..1f1050a7a89d 100644
--- a/drivers/net/wireless/realtek/rtw89/Makefile
+++ b/drivers/net/wireless/realtek/rtw89/Makefile
@@ -17,7 +17,8 @@ rtw89_core-y += core.o \
 		ps.o \
 		chan.o \
 		ser.o \
-		acpi.o
+		acpi.o \
+		util.o
 
 rtw89_core-$(CONFIG_PM) += wow.o
 
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 68221f0b156e..874119b5007d 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -132,6 +132,7 @@ enum rtw89_hci_type {
 enum rtw89_core_chip_id {
 	RTL8852A,
 	RTL8852B,
+	RTL8852BT,
 	RTL8852C,
 	RTL8851B,
 	RTL8922A,
diff --git a/drivers/net/wireless/realtek/rtw89/phy.h b/drivers/net/wireless/realtek/rtw89/phy.h
index 082231ebbee5..6e53ac9140fb 100644
--- a/drivers/net/wireless/realtek/rtw89/phy.h
+++ b/drivers/net/wireless/realtek/rtw89/phy.h
@@ -129,6 +129,7 @@
 #define EDCCA_HL_DIFF_NORMAL 8
 #define RSSI_UNIT_CONVER 110
 #define EDCCA_UNIT_CONVER 128
+#define EDCCA_PWROFST_DEFAULT 18
 
 enum rtw89_phy_c2h_ra_func {
 	RTW89_PHY_C2H_FUNC_STS_RPT,
diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index 01cbd0312102..dc78ffc39ae7 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -7872,6 +7872,7 @@
 #define B_UPD_CLK_ADC_ON BIT(24)
 #define B_ENABLE_CCK BIT(5)
 #define R_RSTB_ASYNC 0x0704
+#define B_RSTB_ASYNC_BW80 GENMASK(9, 8)
 #define B_RSTB_ASYNC_ALL BIT(1)
 #define R_P0_ANT_SW 0x0728
 #define B_P0_HW_ANTSW_DIS_BY_GNT_BT BIT(12)
@@ -8379,6 +8380,7 @@
 #define B_CDD_EVM_CHK_EN BIT(0)
 #define R_PATH0_BAND_SEL_V1 0x4738
 #define B_PATH0_BAND_SEL_MSK_V1 BIT(17)
+#define B_PATH0_BAND_NRBW_EN_V1 BIT(16)
 #define R_PATH0_BT_SHARE_V1 0x4738
 #define B_PATH0_BT_SHARE_V1 BIT(19)
 #define R_PATH0_BTG_PATH_V1 0x4738
@@ -8422,6 +8424,7 @@
 #define B_PATH1_G_TIA1_LNA6_OP1DB_V1 GENMASK(15, 8)
 #define R_PATH1_BAND_SEL_V1 0x4AA4
 #define B_PATH1_BAND_SEL_MSK_V1 BIT(17)
+#define B_PATH1_BAND_NRBW_EN_V1 BIT(16)
 #define R_PATH1_BT_SHARE_V1 0x4AA4
 #define B_PATH1_BT_SHARE_V1 BIT(19)
 #define R_PATH1_BTG_PATH_V1 0x4AA4
@@ -8442,6 +8445,8 @@
 #define B_SEG0R_PD_SPATIAL_REUSE_EN_MSK_V1 BIT(30)
 #define B_SEG0R_PD_SPATIAL_REUSE_EN_MSK BIT(29)
 #define B_SEG0R_PD_LOWER_BOUND_MSK GENMASK(10, 6)
+#define R_PWOFST 0x488C
+#define B_PWOFST GENMASK(21, 17)
 #define R_2P4G_BAND 0x4970
 #define B_2P4G_BAND_SEL BIT(1)
 #define R_FC0_BW 0x4974
@@ -9095,6 +9100,7 @@
 #define R_ADCMOD 0xC0E8
 #define B_ADCMOD_LP GENMASK(31, 16)
 #define R_DCIM 0xC0EC
+#define B_DCIM_RC GENMASK(23, 16)
 #define B_DCIM_FR GENMASK(14, 13)
 #define R_ADDCK0D 0xC0F0
 #define B_ADDCK0D_VAL2 GENMASK(31, 26)
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
index 0826d257ddba..60e9bd5b1376 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
@@ -7,6 +7,7 @@
 #include "phy.h"
 #include "reg.h"
 #include "rtw8852b_common.h"
+#include "util.h"
 
 static const struct rtw89_reg3_def rtw8852bx_pmac_ht20_mcs7_tbl[] = {
 	{0x4580, 0x0000ffff, 0x0},
@@ -609,6 +610,27 @@ static void rtw8852bx_set_gain_error(struct rtw89_dev *rtwdev,
 	}
 }
 
+static void rtw8852bt_ext_loss_avg_update(struct rtw89_dev *rtwdev,
+					  s8 ext_loss_a, s8 ext_loss_b)
+{
+	s8 ext_loss_avg;
+	u64 linear;
+	u8 pwrofst;
+
+	if (ext_loss_a == ext_loss_b) {
+		ext_loss_avg = ext_loss_a;
+	} else {
+		linear = rtw89_db_2_linear(abs(ext_loss_a - ext_loss_b)) + 1;
+		linear = DIV_ROUND_CLOSEST_ULL(linear / 2, 1 << RTW89_LINEAR_FRAC_BITS);
+		ext_loss_avg = rtw89_linear_2_db(linear);
+		ext_loss_avg += min(ext_loss_a, ext_loss_b);
+	}
+
+	pwrofst = max(DIV_ROUND_CLOSEST(ext_loss_avg, 4) + 16, EDCCA_PWROFST_DEFAULT);
+
+	rtw89_phy_write32_mask(rtwdev, R_PWOFST, B_PWOFST, pwrofst);
+}
+
 static void rtw8852bx_set_gain_offset(struct rtw89_dev *rtwdev,
 				      enum rtw89_subband subband,
 				      enum rtw89_phy_idx phy_idx)
@@ -619,6 +641,7 @@ static void rtw8852bx_set_gain_offset(struct rtw89_dev *rtwdev,
 	struct rtw89_hal *hal = &rtwdev->hal;
 	struct rtw89_phy_efuse_gain *efuse_gain = &rtwdev->efuse_gain;
 	enum rtw89_gain_offset gain_ofdm_band;
+	s8 ext_loss_a = 0, ext_loss_b = 0;
 	s32 offset_a, offset_b;
 	s32 offset_ofdm, offset_cck;
 	s32 tmp;
@@ -635,7 +658,7 @@ static void rtw8852bx_set_gain_offset(struct rtw89_dev *rtwdev,
 
 next:
 	if (!efuse_gain->offset_valid)
-		return;
+		goto ext_loss;
 
 	gain_ofdm_band = rtw89_subband_to_gain_offset_band_of_ofdm(subband);
 
@@ -672,6 +695,13 @@ next:
 		rtw89_phy_write32_mask(rtwdev, R_RX_RPL_OFST,
 				       B_RX_RPL_OFST_CCK_MASK, tmp);
 	}
+
+	ext_loss_a = (offset_a << 2) + (efuse_gain->offset_base[RTW89_PHY_0] >> 2);
+	ext_loss_b = (offset_b << 2) + (efuse_gain->offset_base[RTW89_PHY_0] >> 2);
+
+ext_loss:
+	if (rtwdev->chip->chip_id == RTL8852BT)
+		rtw8852bt_ext_loss_avg_update(rtwdev, ext_loss_a, ext_loss_b);
 }
 
 static
@@ -804,10 +834,79 @@ static void rtw8852b_bw_setting(struct rtw89_dev *rtwdev, u8 bw, u8 path)
 	}
 }
 
+static
+void rtw8852bt_adc_cfg(struct rtw89_dev *rtwdev, u8 bw, u8 path)
+{
+	static const u32 rck_reset_count[2] = {0xC0E8, 0xC1E8};
+	static const u32 adc_op5_bw_sel[2] = {0xC0D8, 0xC1D8};
+	static const u32 adc_sample_td[2] = {0xC0D4, 0xC1D4};
+	static const u32 adc_rst_cycle[2] = {0xC0EC, 0xC1EC};
+	static const u32 decim_filter[2] = {0xC0EC, 0xC1EC};
+	static const u32 rck_offset[2] = {0xC0C4, 0xC1C4};
+	static const u32 rx_adc_clk[2] = {0x12A0, 0x32A0};
+	static const u32 wbadc_sel[2] = {0xC0E4, 0xC1E4};
+	static const u32 idac2_1[2] = {0xC0D4, 0xC1D4};
+	static const u32 idac2[2] = {0xC0D4, 0xC1D4};
+	static const u32 upd_clk_adc = {0x704};
+
+	if (rtwdev->chip->chip_id != RTL8852BT)
+		return;
+
+	rtw89_phy_write32_mask(rtwdev, idac2[path], B_P0_CFCH_CTL, 0x8);
+	rtw89_phy_write32_mask(rtwdev, rck_reset_count[path], B_ADCMOD_LP, 0x9);
+	rtw89_phy_write32_mask(rtwdev, wbadc_sel[path], B_WDADC_SEL, 0x2);
+	rtw89_phy_write32_mask(rtwdev, rx_adc_clk[path], B_P0_RXCK_ADJ, 0x49);
+	rtw89_phy_write32_mask(rtwdev, decim_filter[path], B_DCIM_FR, 0x0);
+
+	switch (bw) {
+	case RTW89_CHANNEL_WIDTH_5:
+	case RTW89_CHANNEL_WIDTH_10:
+	case RTW89_CHANNEL_WIDTH_20:
+	case RTW89_CHANNEL_WIDTH_40:
+		rtw89_phy_write32_mask(rtwdev, idac2_1[path], B_P0_CFCH_EN, 0x2);
+		rtw89_phy_write32_mask(rtwdev, adc_sample_td[path], B_P0_CFCH_BW0, 0x3);
+		rtw89_phy_write32_mask(rtwdev, adc_op5_bw_sel[path], B_P0_CFCH_BW1, 0xf);
+		rtw89_phy_write32_mask(rtwdev, rck_offset[path], B_DRCK_MUL, 0x0);
+		/* Tx TSSI ADC update */
+		rtw89_phy_write32_mask(rtwdev, upd_clk_adc, B_RSTB_ASYNC_BW80, 0);
+
+		if (rtwdev->efuse.rfe_type >= 51)
+			rtw89_phy_write32_mask(rtwdev, adc_rst_cycle[path], B_DCIM_RC, 0x2);
+		else
+			rtw89_phy_write32_mask(rtwdev, adc_rst_cycle[path], B_DCIM_RC, 0x3);
+		break;
+	case RTW89_CHANNEL_WIDTH_80:
+		rtw89_phy_write32_mask(rtwdev, idac2_1[path], B_P0_CFCH_EN, 0x2);
+		rtw89_phy_write32_mask(rtwdev, adc_sample_td[path], B_P0_CFCH_BW0, 0x2);
+		rtw89_phy_write32_mask(rtwdev, adc_op5_bw_sel[path], B_P0_CFCH_BW1, 0x8);
+		rtw89_phy_write32_mask(rtwdev, rck_offset[path], B_DRCK_MUL, 0x0);
+		rtw89_phy_write32_mask(rtwdev, adc_rst_cycle[path], B_DCIM_RC, 0x3);
+		/* Tx TSSI ADC update */
+		rtw89_phy_write32_mask(rtwdev, upd_clk_adc, B_RSTB_ASYNC_BW80, 1);
+		break;
+	case RTW89_CHANNEL_WIDTH_160:
+		rtw89_phy_write32_mask(rtwdev, idac2_1[path], B_P0_CFCH_EN, 0x0);
+		rtw89_phy_write32_mask(rtwdev, adc_sample_td[path], B_P0_CFCH_BW0, 0x2);
+		rtw89_phy_write32_mask(rtwdev, adc_op5_bw_sel[path], B_P0_CFCH_BW1, 0x4);
+		rtw89_phy_write32_mask(rtwdev, rck_offset[path], B_DRCK_MUL, 0x6);
+		rtw89_phy_write32_mask(rtwdev, adc_rst_cycle[path], B_DCIM_RC, 0x3);
+		/* Tx TSSI ADC update */
+		rtw89_phy_write32_mask(rtwdev, upd_clk_adc, B_RSTB_ASYNC_BW80, 2);
+		break;
+	default:
+		rtw89_warn(rtwdev, "Fail to set ADC\n");
+		break;
+	}
+}
+
 static void rtw8852bx_ctrl_bw(struct rtw89_dev *rtwdev, u8 pri_ch, u8 bw,
 			      enum rtw89_phy_idx phy_idx)
 {
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	u32 rx_path_0;
+	u32 val;
+
+	rx_path_0 = rtw89_phy_read32_idx(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0, phy_idx);
 
 	switch (bw) {
 	case RTW89_CHANNEL_WIDTH_5:
@@ -820,6 +919,12 @@ static void rtw8852bx_ctrl_bw(struct rtw89_dev *rtwdev, u8 pri_ch, u8 bw,
 				      B_P0_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
 		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
 				      B_P1_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
+		if (chip_id == RTL8852BT) {
+			rtw89_phy_write32_idx(rtwdev, R_PATH0_BAND_SEL_V1,
+					      B_PATH0_BAND_NRBW_EN_V1, 0x0, phy_idx);
+			rtw89_phy_write32_idx(rtwdev, R_PATH1_BAND_SEL_V1,
+					      B_PATH1_BAND_NRBW_EN_V1, 0x0, phy_idx);
+		}
 		break;
 	case RTW89_CHANNEL_WIDTH_10:
 		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x0, phy_idx);
@@ -831,6 +936,12 @@ static void rtw8852bx_ctrl_bw(struct rtw89_dev *rtwdev, u8 pri_ch, u8 bw,
 				      B_P0_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
 		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
 				      B_P1_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
+		if (chip_id == RTL8852BT) {
+			rtw89_phy_write32_idx(rtwdev, R_PATH0_BAND_SEL_V1,
+					      B_PATH0_BAND_NRBW_EN_V1, 0x0, phy_idx);
+			rtw89_phy_write32_idx(rtwdev, R_PATH1_BAND_SEL_V1,
+					      B_PATH1_BAND_NRBW_EN_V1, 0x0, phy_idx);
+		}
 		break;
 	case RTW89_CHANNEL_WIDTH_20:
 		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x0, phy_idx);
@@ -842,6 +953,12 @@ static void rtw8852bx_ctrl_bw(struct rtw89_dev *rtwdev, u8 pri_ch, u8 bw,
 				      B_P0_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
 		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
 				      B_P1_RFMODE_ORI_RX_ALL, 0x333, phy_idx);
+		if (chip_id == RTL8852BT) {
+			rtw89_phy_write32_idx(rtwdev, R_PATH0_BAND_SEL_V1,
+					      B_PATH0_BAND_NRBW_EN_V1, 0x1, phy_idx);
+			rtw89_phy_write32_idx(rtwdev, R_PATH1_BAND_SEL_V1,
+					      B_PATH1_BAND_NRBW_EN_V1, 0x1, phy_idx);
+		}
 		break;
 	case RTW89_CHANNEL_WIDTH_40:
 		rtw89_phy_write32_idx(rtwdev, R_FC0_BW_V1, B_FC0_BW_SET, 0x1, phy_idx);
@@ -868,21 +985,25 @@ static void rtw8852bx_ctrl_bw(struct rtw89_dev *rtwdev, u8 pri_ch, u8 bw,
 				      pri_ch, phy_idx);
 
 		/*Set RF mode at A */
+		val = chip_id == RTL8852BT ? 0x333 : 0xaaa;
 		rtw89_phy_write32_idx(rtwdev, R_P0_RFMODE_ORI_RX,
-				      B_P0_RFMODE_ORI_RX_ALL, 0xaaa, phy_idx);
+				      B_P0_RFMODE_ORI_RX_ALL, val, phy_idx);
 		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
-				      B_P1_RFMODE_ORI_RX_ALL, 0xaaa, phy_idx);
+				      B_P1_RFMODE_ORI_RX_ALL, val, phy_idx);
 		break;
 	default:
 		rtw89_warn(rtwdev, "Fail to switch bw (bw:%d, pri ch:%d)\n", bw,
 			   pri_ch);
 	}
 
-	rtw8852b_bw_setting(rtwdev, bw, RF_PATH_A);
-	rtw8852b_bw_setting(rtwdev, bw, RF_PATH_B);
+	if (chip_id == RTL8852B) {
+		rtw8852b_bw_setting(rtwdev, bw, RF_PATH_A);
+		rtw8852b_bw_setting(rtwdev, bw, RF_PATH_B);
+	} else if (chip_id == RTL8852BT) {
+		rtw8852bt_adc_cfg(rtwdev, bw, RF_PATH_A);
+		rtw8852bt_adc_cfg(rtwdev, bw, RF_PATH_B);
+	}
 
-	rx_path_0 = rtw89_phy_read32_idx(rtwdev, R_CHBW_MOD_V1, B_ANT_RX_SEG0,
-					 phy_idx);
 	if (rx_path_0 == 0x1)
 		rtw89_phy_write32_idx(rtwdev, R_P1_RFMODE_ORI_RX,
 				      B_P1_RFMODE_ORI_RX_ALL, 0x111, phy_idx);
@@ -1001,10 +1122,56 @@ static void rtw8852bx_bb_set_pop(struct rtw89_dev *rtwdev)
 		rtw89_phy_write32_clr(rtwdev, R_PKT_CTRL, B_PKT_POP_EN);
 }
 
+static u32 rtw8852bt_spur_freq(struct rtw89_dev *rtwdev,
+			       const struct rtw89_chan *chan)
+{
+	u8 center_chan = chan->channel;
+
+	switch (chan->band_type) {
+	case RTW89_BAND_5G:
+		if (center_chan == 151 || center_chan == 153 ||
+		    center_chan == 155 || center_chan == 163)
+			return 5760;
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+#define CARRIER_SPACING_312_5 312500 /* 312.5 kHz */
+#define CARRIER_SPACING_78_125 78125 /* 78.125 kHz */
+#define MAX_TONE_NUM 2048
+
+static void rtw8852bt_set_csi_tone_idx(struct rtw89_dev *rtwdev,
+				       const struct rtw89_chan *chan,
+				       enum rtw89_phy_idx phy_idx)
+{
+	s32 freq_diff, csi_idx, csi_tone_idx;
+	u32 spur_freq;
+
+	spur_freq = rtw8852bt_spur_freq(rtwdev, chan);
+	if (spur_freq == 0) {
+		rtw89_phy_write32_idx(rtwdev, R_SEG0CSI_EN_V1, B_SEG0CSI_EN,
+				      0, phy_idx);
+		return;
+	}
+
+	freq_diff = (spur_freq - chan->freq) * 1000000;
+	csi_idx = s32_div_u32_round_closest(freq_diff, CARRIER_SPACING_78_125);
+	s32_div_u32_round_down(csi_idx, MAX_TONE_NUM, &csi_tone_idx);
+
+	rtw89_phy_write32_idx(rtwdev, R_SEG0CSI_V1, B_SEG0CSI_IDX,
+			      csi_tone_idx, phy_idx);
+	rtw89_phy_write32_idx(rtwdev, R_SEG0CSI_EN_V1, B_SEG0CSI_EN, 1, phy_idx);
+}
+
 static
 void __rtw8852bx_set_channel_bb(struct rtw89_dev *rtwdev, const struct rtw89_chan *chan,
 				enum rtw89_phy_idx phy_idx)
 {
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	bool cck_en = chan->channel <= 14;
 	u8 pri_ch_idx = chan->pri_ch_idx;
 	u8 band = chan->band_type, chan_idx;
@@ -1015,7 +1182,9 @@ void __rtw8852bx_set_channel_bb(struct rtw89_dev *rtwdev, const struct rtw89_cha
 	rtw8852bx_ctrl_ch(rtwdev, chan, phy_idx);
 	rtw8852bx_ctrl_bw(rtwdev, pri_ch_idx, chan->band_width, phy_idx);
 	rtw8852bx_ctrl_cck_en(rtwdev, cck_en);
-	if (chan->band_type == RTW89_BAND_5G) {
+	if (chip_id == RTL8852BT)
+		rtw8852bt_set_csi_tone_idx(rtwdev, chan, phy_idx);
+	if (chip_id == RTL8852B && chan->band_type == RTW89_BAND_5G) {
 		rtw89_phy_write32_mask(rtwdev, R_PATH0_BT_SHARE_V1,
 				       B_PATH0_BT_SHARE_V1, 0x0);
 		rtw89_phy_write32_mask(rtwdev, R_PATH0_BTG_PATH_V1,
@@ -1614,6 +1783,8 @@ static void __rtw8852bx_query_ppdu(struct rtw89_dev *rtwdev,
 
 static int __rtw8852bx_mac_enable_bb_rf(struct rtw89_dev *rtwdev)
 {
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
+	u32 val32;
 	int ret;
 
 	rtw89_write8_set(rtwdev, R_AX_SYS_FUNC_EN,
@@ -1623,6 +1794,13 @@ static int __rtw8852bx_mac_enable_bb_rf(struct rtw89_dev *rtwdev)
 	rtw89_write32_clr(rtwdev, R_AX_WLRF_CTRL, B_AX_AFC_AFEDIG);
 	rtw89_write32_set(rtwdev, R_AX_WLRF_CTRL, B_AX_AFC_AFEDIG);
 
+	if (chip_id == RTL8852BT) {
+		val32 = rtw89_read32(rtwdev, R_AX_AFE_OFF_CTRL1);
+		val32 = u32_replace_bits(val32, 0x1, B_AX_S0_LDO_VSEL_F_MASK);
+		val32 = u32_replace_bits(val32, 0x1, B_AX_S1_LDO_VSEL_F_MASK);
+		rtw89_write32(rtwdev, R_AX_AFE_OFF_CTRL1, val32);
+	}
+
 	ret = rtw89_mac_write_xtal_si(rtwdev, XTAL_SI_WL_RFC_S0, 0xC7,
 				      FULL_BIT_MASK);
 	if (ret)
@@ -1693,6 +1871,7 @@ const struct rtw8852bx_info rtw8852bx_info = {
 	.init_txpwr_unit = __rtw8852bx_init_txpwr_unit,
 	.set_txpwr_ul_tb_offset = __rtw8852bx_set_txpwr_ul_tb_offset,
 	.get_thermal = __rtw8852bx_get_thermal,
+	.adc_cfg = rtw8852bt_adc_cfg,
 };
 EXPORT_SYMBOL(rtw8852bx_info);
 
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.h b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.h
index 2348a13545ac..b8e3f20ba189 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.h
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.h
@@ -158,6 +158,7 @@ struct rtw8852bx_info {
 	void (*set_txpwr_ul_tb_offset)(struct rtw89_dev *rtwdev,
 				       s8 pw_ofst, enum rtw89_mac_idx mac_idx);
 	u8 (*get_thermal)(struct rtw89_dev *rtwdev, enum rtw89_rf_path rf_path);
+	void (*adc_cfg)(struct rtw89_dev *rtwdev, u8 bw, u8 path);
 };
 
 extern const struct rtw8852bx_info rtw8852bx_info;
@@ -336,4 +337,10 @@ u8 rtw8852bx_get_thermal(struct rtw89_dev *rtwdev, enum rtw89_rf_path rf_path)
 	return rtw8852bx_info.get_thermal(rtwdev, rf_path);
 }
 
+static inline
+void rtw8852bx_adc_cfg(struct rtw89_dev *rtwdev, u8 bw, u8 path)
+{
+	rtw8852bx_info.adc_cfg(rtwdev, bw, path);
+}
+
 #endif
diff --git a/drivers/net/wireless/realtek/rtw89/util.c b/drivers/net/wireless/realtek/rtw89/util.c
new file mode 100644
index 000000000000..e71956ce9853
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/util.c
@@ -0,0 +1,106 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/* Copyright(c) 2024  Realtek Corporation
+ */
+
+#include "util.h"
+
+#define FRAC_ROWS 3
+#define FRAC_ROW_MAX (FRAC_ROWS - 1)
+#define NORM_ROW_MIN FRAC_ROWS
+
+static const u32 db_invert_table[12][8] = {
+	/* rows 0~2 in unit of U(32,3) */
+	{10, 13, 16, 20, 25, 32, 40, 50},
+	{64, 80, 101, 128, 160, 201, 256, 318},
+	{401, 505, 635, 800, 1007, 1268, 1596, 2010},
+	/* rows 3~11 in unit of U(32,0) */
+	{316, 398, 501, 631, 794, 1000, 1259, 1585},
+	{1995, 2512, 3162, 3981, 5012, 6310, 7943, 10000},
+	{12589, 15849, 19953, 25119, 31623, 39811, 50119, 63098},
+	{79433, 100000, 125893, 158489, 199526, 251189, 316228, 398107},
+	{501187, 630957, 794328, 1000000, 1258925, 1584893, 1995262, 2511886},
+	{3162278, 3981072, 5011872, 6309573, 7943282, 1000000, 12589254,
+	 15848932},
+	{19952623, 25118864, 31622777, 39810717, 50118723, 63095734, 79432823,
+	 100000000},
+	{125892541, 158489319, 199526232, 251188643, 316227766, 398107171,
+	 501187234, 630957345},
+	{794328235, 1000000000, 1258925412, 1584893192, 1995262315, 2511886432U,
+	 3162277660U, 3981071706U},
+};
+
+u32 rtw89_linear_2_db(u64 val)
+{
+	u8 i, j;
+	u32 dB;
+
+	for (i = 0; i < 12; i++) {
+		for (j = 0; j < 8; j++) {
+			if (i <= FRAC_ROW_MAX &&
+			    (val << RTW89_LINEAR_FRAC_BITS) <= db_invert_table[i][j])
+				goto cnt;
+			else if (i > FRAC_ROW_MAX && val <= db_invert_table[i][j])
+				goto cnt;
+		}
+	}
+
+	return 96; /* maximum 96 dB */
+
+cnt:
+	/* special cases */
+	if (j == 0 && i == 0)
+		goto end;
+
+	if (i == NORM_ROW_MIN && j == 0) {
+		if (db_invert_table[NORM_ROW_MIN][0] - val >
+		    val - (db_invert_table[FRAC_ROW_MAX][7] >> RTW89_LINEAR_FRAC_BITS)) {
+			i = FRAC_ROW_MAX;
+			j = 7;
+		}
+		goto end;
+	}
+
+	if (i <= FRAC_ROW_MAX)
+		val <<= RTW89_LINEAR_FRAC_BITS;
+
+	/* compare difference to get precise dB */
+	if (j == 0) {
+		if (db_invert_table[i][j] - val >
+		    val - db_invert_table[i - 1][7]) {
+			i--;
+			j = 7;
+		}
+	} else {
+		if (db_invert_table[i][j] - val >
+		    val - db_invert_table[i][j - 1]) {
+			j--;
+		}
+	}
+end:
+	dB = (i << 3) + j + 1;
+
+	return dB;
+}
+EXPORT_SYMBOL(rtw89_linear_2_db);
+
+u64 rtw89_db_2_linear(u32 db)
+{
+	u64 linear;
+	u8 i, j;
+
+	if (db > 96)
+		db = 96;
+	else if (db < 1)
+		return 1;
+
+	i = (db - 1) >> 3;
+	j = (db - 1) & 0x7;
+
+	linear = db_invert_table[i][j];
+
+	if (i >= NORM_ROW_MIN)
+		linear = linear << RTW89_LINEAR_FRAC_BITS;
+
+	return linear;
+}
+EXPORT_SYMBOL(rtw89_db_2_linear);
diff --git a/drivers/net/wireless/realtek/rtw89/util.h b/drivers/net/wireless/realtek/rtw89/util.h
index e2ed4565025d..e82e7df052d8 100644
--- a/drivers/net/wireless/realtek/rtw89/util.h
+++ b/drivers/net/wireless/realtek/rtw89/util.h
@@ -6,6 +6,8 @@
 
 #include "core.h"
 
+#define RTW89_LINEAR_FRAC_BITS 3
+
 #define rtw89_iterate_vifs_bh(rtwdev, iterator, data)                          \
 	ieee80211_iterate_active_interfaces_atomic((rtwdev)->hw,               \
 			IEEE80211_IFACE_ITER_NORMAL, iterator, data)
@@ -55,4 +57,7 @@ static inline void ether_addr_copy_mask(u8 *dst, const u8 *src, u8 mask)
 	}
 }
 
+u32 rtw89_linear_2_db(u64 linear);
+u64 rtw89_db_2_linear(u32 db);
+
 #endif
-- 
cgit v1.2.3-58-ga151


From 190dc12d0649b7d8709c98c10926ea0c001f955b Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 7 Jun 2024 15:06:59 +0800
Subject: wifi: rtw89: 885xbx: apply common settings to 8851B, 8852B and 8852BT

Many common settings can share to 8851B, 8852B and 8852BT, so add an inline
function rtw89_is_rtl885xb() to be concise. Meanwhile review and align
settings for existing chips.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240607070659.80263-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/cam.c   |  1 +
 drivers/net/wireless/realtek/rtw89/coex.c  |  3 ++-
 drivers/net/wireless/realtek/rtw89/core.c  |  2 +-
 drivers/net/wireless/realtek/rtw89/core.h  | 10 +++++++
 drivers/net/wireless/realtek/rtw89/debug.c |  2 +-
 drivers/net/wireless/realtek/rtw89/mac.c   | 43 +++++++++++++++---------------
 drivers/net/wireless/realtek/rtw89/pci.c   | 38 ++++++++++++--------------
 drivers/net/wireless/realtek/rtw89/phy.c   |  2 +-
 drivers/net/wireless/realtek/rtw89/reg.h   |  2 ++
 9 files changed, 57 insertions(+), 46 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/cam.c b/drivers/net/wireless/realtek/rtw89/cam.c
index 35291efbbae9..4557c6e035a9 100644
--- a/drivers/net/wireless/realtek/rtw89/cam.c
+++ b/drivers/net/wireless/realtek/rtw89/cam.c
@@ -523,6 +523,7 @@ static u8 rtw89_get_addr_cam_entry_size(struct rtw89_dev *rtwdev)
 	case RTL8852A:
 	case RTL8852B:
 	case RTL8851B:
+	case RTL8852BT:
 		return ADDR_CAM_ENT_SIZE;
 	default:
 		return ADDR_CAM_ENT_SHORT_SIZE;
diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index c443b39ab3c6..d8fc93318a51 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -228,6 +228,7 @@ static u32 chip_id_to_bt_rom_code_id(u32 id)
 	case RTL8852A:
 	case RTL8852B:
 	case RTL8852C:
+	case RTL8852BT:
 		return 0x8852;
 	case RTL8851B:
 		return 0x8851;
@@ -9514,7 +9515,7 @@ static void _get_gnt(struct rtw89_dev *rtwdev, struct rtw89_mac_ax_coex_gnt *gnt
 	u32 val, status;
 
 	if (chip->chip_id == RTL8852A || chip->chip_id == RTL8852B ||
-	    chip->chip_id == RTL8851B) {
+	    chip->chip_id == RTL8851B || chip->chip_id == RTL8852BT) {
 		rtw89_mac_read_lte(rtwdev, R_AX_LTE_SW_CFG_1, &val);
 		rtw89_mac_read_lte(rtwdev, R_AX_GNT_VAL, &status);
 
diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index fe403fb7e8a2..87c38cdc691b 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -4382,7 +4382,7 @@ static void rtw89_read_chip_ver(struct rtw89_dev *rtwdev)
 
 	rtwdev->hal.cv = cv;
 
-	if (chip->chip_id == RTL8852B || chip->chip_id == RTL8851B) {
+	if (rtw89_is_rtl885xb(rtwdev)) {
 		ret = rtw89_mac_read_xtal_si(rtwdev, XTAL_SI_CV, &val);
 		if (ret)
 			return;
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 874119b5007d..e8ec0246bccd 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -6421,6 +6421,16 @@ static inline bool rtw89_is_mlo_1_1(struct rtw89_dev *rtwdev)
 	}
 }
 
+static inline bool rtw89_is_rtl885xb(struct rtw89_dev *rtwdev)
+{
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
+
+	if (chip_id == RTL8852B || chip_id == RTL8851B || chip_id == RTL8852BT)
+		return true;
+
+	return false;
+}
+
 int rtw89_core_tx_write(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif,
 			struct ieee80211_sta *sta, struct sk_buff *skb, int *qsel);
 int rtw89_h2c_tx(struct rtw89_dev *rtwdev,
diff --git a/drivers/net/wireless/realtek/rtw89/debug.c b/drivers/net/wireless/realtek/rtw89/debug.c
index 210f192ac9ec..49bbbd049316 100644
--- a/drivers/net/wireless/realtek/rtw89/debug.c
+++ b/drivers/net/wireless/realtek/rtw89/debug.c
@@ -2996,7 +2996,7 @@ static bool is_dbg_port_valid(struct rtw89_dev *rtwdev, u32 sel)
 	    sel >= RTW89_DBG_PORT_SEL_PCIE_TXDMA &&
 	    sel <= RTW89_DBG_PORT_SEL_PCIE_MISC2)
 		return false;
-	if (rtwdev->chip->chip_id == RTL8852B &&
+	if (rtw89_is_rtl885xb(rtwdev) &&
 	    sel >= RTW89_DBG_PORT_SEL_PTCL_C1 &&
 	    sel <= RTW89_DBG_PORT_SEL_TXTF_INFOH_C1)
 		return false;
diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index 824ece03d92d..b759fa8fbca9 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -1568,6 +1568,8 @@ static int dmac_func_en_ax(struct rtw89_dev *rtwdev)
 		 B_AX_DLE_CPUIO_CLK_EN | B_AX_PKT_IN_CLK_EN |
 		 B_AX_STA_SCH_CLK_EN | B_AX_TXPKT_CTRL_CLK_EN |
 		 B_AX_WD_RLS_CLK_EN | B_AX_BBRPT_CLK_EN);
+	if (chip_id == RTL8852BT)
+		val32 |= B_AX_AXIDMA_CLK_EN;
 	rtw89_write32(rtwdev, R_AX_DMAC_CLK_EN, val32);
 
 	return 0;
@@ -1577,7 +1579,7 @@ static int chip_func_en_ax(struct rtw89_dev *rtwdev)
 {
 	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B)
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev))
 		rtw89_write32_set(rtwdev, R_AX_SPS_DIG_ON_CTRL0,
 				  B_AX_OCP_L1_MASK);
 
@@ -2146,8 +2148,8 @@ int rtw89_mac_preload_init(struct rtw89_dev *rtwdev, enum rtw89_mac_idx mac_idx,
 {
 	const struct rtw89_chip_info *chip = rtwdev->chip;
 
-	if (chip->chip_id == RTL8852A || chip->chip_id == RTL8852B ||
-	    chip->chip_id == RTL8851B || !is_qta_poh(rtwdev))
+	if (chip->chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev) ||
+	    !is_qta_poh(rtwdev))
 		return 0;
 
 	return preload_init_set(rtwdev, mac_idx, mode);
@@ -2183,8 +2185,7 @@ static void _patch_ss2f_path(struct rtw89_dev *rtwdev)
 {
 	const struct rtw89_chip_info *chip = rtwdev->chip;
 
-	if (chip->chip_id == RTL8852A || chip->chip_id == RTL8852B ||
-	    chip->chip_id == RTL8851B)
+	if (chip->chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev))
 		return;
 
 	rtw89_write32_mask(rtwdev, R_AX_SS2FINFO_PATH, B_AX_SS_DEST_QUEUE_MASK,
@@ -2360,7 +2361,7 @@ static int scheduler_init_ax(struct rtw89_dev *rtwdev, u8 mac_idx)
 		rtw89_write32_mask(rtwdev, reg, B_AX_SIFS_MACTXEN_T1_MASK,
 				   SIFS_MACTXEN_T1);
 
-	if (rtwdev->chip->chip_id == RTL8852B || rtwdev->chip->chip_id == RTL8851B) {
+	if (rtw89_is_rtl885xb(rtwdev)) {
 		reg = rtw89_mac_reg_by_idx(rtwdev, R_AX_SCH_EXT_CTRL, mac_idx);
 		rtw89_write32_set(rtwdev, reg, B_AX_PORT_RST_TSF_ADV);
 	}
@@ -2588,7 +2589,9 @@ static int trxptcl_init_ax(struct rtw89_dev *rtwdev, u8 mac_idx)
 	case RTL8852A:
 		sifs = WMAC_SPEC_SIFS_OFDM_52A;
 		break;
+	case RTL8851B:
 	case RTL8852B:
+	case RTL8852BT:
 		sifs = WMAC_SPEC_SIFS_OFDM_52B;
 		break;
 	default:
@@ -2632,6 +2635,7 @@ static int rmac_init_ax(struct rtw89_dev *rtwdev, u8 mac_idx)
 #define RX_MAX_LEN_UNIT 512
 #define PLD_RLS_MAX_PG 127
 #define RX_SPEC_MAX_LEN (11454 + RX_MAX_LEN_UNIT)
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	int ret;
 	u32 reg, rx_max_len, rx_qta;
 	u16 val;
@@ -2652,6 +2656,8 @@ static int rmac_init_ax(struct rtw89_dev *rtwdev, u8 mac_idx)
 			       B_AX_RX_DLK_DATA_TIME_MASK);
 	val = u16_replace_bits(val, TRXCFG_RMAC_CCA_TO,
 			       B_AX_RX_DLK_CCA_TIME_MASK);
+	if (chip_id == RTL8852BT)
+		val |= B_AX_RX_DLK_RST_EN;
 	rtw89_write16(rtwdev, reg, val);
 
 	reg = rtw89_mac_reg_by_idx(rtwdev, R_AX_RCR, mac_idx);
@@ -2668,8 +2674,7 @@ static int rmac_init_ax(struct rtw89_dev *rtwdev, u8 mac_idx)
 	rx_max_len /= RX_MAX_LEN_UNIT;
 	rtw89_write32_mask(rtwdev, reg, B_AX_RX_MPDU_MAX_LEN_MASK, rx_max_len);
 
-	if (rtwdev->chip->chip_id == RTL8852A &&
-	    rtwdev->hal.cv == CHIP_CBV) {
+	if (chip_id == RTL8852A && rtwdev->hal.cv == CHIP_CBV) {
 		rtw89_write16_mask(rtwdev,
 				   rtw89_mac_reg_by_idx(rtwdev, R_AX_DLK_PROTECT_CTL, mac_idx),
 				   B_AX_RX_DLK_CCA_TIME_MASK, 0);
@@ -2700,7 +2705,7 @@ static int cmac_com_init_ax(struct rtw89_dev *rtwdev, u8 mac_idx)
 	val = u32_replace_bits(val, 0, B_AX_TXSC_80M_MASK);
 	rtw89_write32(rtwdev, reg, val);
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B) {
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)) {
 		reg = rtw89_mac_reg_by_idx(rtwdev, R_AX_PTCL_RRSR1, mac_idx);
 		rtw89_write32_mask(rtwdev, reg, B_AX_RRSR_RATE_EN_MASK, RRSR_OFDM_CCK_EN);
 	}
@@ -2766,11 +2771,10 @@ static int ptcl_init_ax(struct rtw89_dev *rtwdev, u8 mac_idx)
 
 static int cmac_dma_init_ax(struct rtw89_dev *rtwdev, u8 mac_idx)
 {
-	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	u32 reg;
 	int ret;
 
-	if (chip_id != RTL8852B)
+	if (!rtw89_is_rtl885xb(rtwdev))
 		return 0;
 
 	ret = rtw89_mac_check_mac_en(rtwdev, mac_idx, RTW89_CMAC_SEL);
@@ -3587,13 +3591,11 @@ static int enable_imr_ax(struct rtw89_dev *rtwdev, u8 mac_idx,
 
 static void err_imr_ctrl_ax(struct rtw89_dev *rtwdev, bool en)
 {
-	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
-
 	rtw89_write32(rtwdev, R_AX_DMAC_ERR_IMR,
 		      en ? DMAC_ERR_IMR_EN : DMAC_ERR_IMR_DIS);
 	rtw89_write32(rtwdev, R_AX_CMAC_ERR_IMR,
 		      en ? CMAC0_ERR_IMR_EN : CMAC0_ERR_IMR_DIS);
-	if (chip_id != RTL8852B && rtwdev->mac.dle_info.c1_rx_qta)
+	if (!rtw89_is_rtl885xb(rtwdev) && rtwdev->mac.dle_info.c1_rx_qta)
 		rtw89_write32(rtwdev, R_AX_CMAC_ERR_IMR_C1,
 			      en ? CMAC1_ERR_IMR_EN : CMAC1_ERR_IMR_DIS);
 }
@@ -3719,10 +3721,9 @@ static int rtw89_mac_feat_init(struct rtw89_dev *rtwdev)
 
 static void rtw89_disable_fw_watchdog(struct rtw89_dev *rtwdev)
 {
-	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	u32 val32;
 
-	if (chip_id == RTL8852B || chip_id == RTL8851B) {
+	if (rtw89_is_rtl885xb(rtwdev)) {
 		rtw89_write32_clr(rtwdev, R_AX_PLATFORM_ENABLE, B_AX_APB_WRAP_EN);
 		rtw89_write32_set(rtwdev, R_AX_PLATFORM_ENABLE, B_AX_APB_WRAP_EN);
 		return;
@@ -3818,7 +3819,7 @@ static void rtw89_mac_dmac_func_pre_en_ax(struct rtw89_dev *rtwdev)
 	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	u32 val;
 
-	if (chip_id == RTL8851B)
+	if (chip_id == RTL8851B || chip_id == RTL8852BT)
 		val = B_AX_DISPATCHER_CLK_EN | B_AX_AXIDMA_CLK_EN;
 	else
 		val = B_AX_DISPATCHER_CLK_EN;
@@ -5463,18 +5464,19 @@ void rtw89_mac_flush_txq(struct rtw89_dev *rtwdev, u32 queues, bool drop)
 
 int rtw89_mac_coex_init(struct rtw89_dev *rtwdev, const struct rtw89_mac_ax_coex *coex)
 {
+	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	u8 val;
 	u16 val16;
 	u32 val32;
 	int ret;
 
 	rtw89_write8_set(rtwdev, R_AX_GPIO_MUXCFG, B_AX_ENBT);
-	if (rtwdev->chip->chip_id != RTL8851B)
+	if (chip_id != RTL8851B && chip_id != RTL8852BT)
 		rtw89_write8_set(rtwdev, R_AX_BTC_FUNC_EN, B_AX_PTA_WL_TX_EN);
 	rtw89_write8_set(rtwdev, R_AX_BT_COEX_CFG_2 + 1, B_AX_GNT_BT_POLARITY >> 8);
 	rtw89_write8_set(rtwdev, R_AX_CSR_MODE, B_AX_STATIS_BT_EN | B_AX_WL_ACT_MSK);
 	rtw89_write8_set(rtwdev, R_AX_CSR_MODE + 2, B_AX_BT_CNT_RST >> 16);
-	if (rtwdev->chip->chip_id != RTL8851B)
+	if (chip_id != RTL8851B && chip_id != RTL8852BT)
 		rtw89_write8_clr(rtwdev, R_AX_TRXPTCL_RESP_0 + 3, B_AX_RSP_CHK_BTCCA >> 24);
 
 	val16 = rtw89_read16(rtwdev, R_AX_CCA_CFG_0);
@@ -5757,8 +5759,7 @@ bool rtw89_mac_get_ctrl_path(struct rtw89_dev *rtwdev)
 
 	if (chip->chip_id == RTL8852C || chip->chip_id == RTL8922A)
 		return false;
-	else if (chip->chip_id == RTL8852A || chip->chip_id == RTL8852B ||
-		 chip->chip_id == RTL8851B)
+	else if (chip->chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev))
 		val = rtw89_read8_mask(rtwdev, R_AX_SYS_SDIO_CTRL + 3,
 				       B_AX_LTE_MUX_CTRL_PATH >> 24);
 
diff --git a/drivers/net/wireless/realtek/rtw89/pci.c b/drivers/net/wireless/realtek/rtw89/pci.c
index 36cb0e351eea..b189e1ed98a8 100644
--- a/drivers/net/wireless/realtek/rtw89/pci.c
+++ b/drivers/net/wireless/realtek/rtw89/pci.c
@@ -2039,7 +2039,7 @@ static int rtw89_pci_write_config_byte(struct rtw89_dev *rtwdev, u16 addr,
 	if (!ret)
 		return 0;
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B)
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev))
 		ret = rtw89_dbi_write8(rtwdev, addr, data);
 
 	return ret;
@@ -2057,7 +2057,7 @@ static int rtw89_pci_read_config_byte(struct rtw89_dev *rtwdev, u16 addr,
 	if (!ret)
 		return 0;
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B)
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev))
 		ret = rtw89_dbi_read8(rtwdev, addr, value);
 
 	return ret;
@@ -2137,10 +2137,9 @@ __get_target(struct rtw89_dev *rtwdev, u16 *target, enum rtw89_pcie_phy phy_rate
 
 static int rtw89_pci_autok_x(struct rtw89_dev *rtwdev)
 {
-	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	int ret;
 
-	if (chip_id != RTL8852B && chip_id != RTL8851B)
+	if (!rtw89_is_rtl885xb(rtwdev))
 		return 0;
 
 	ret = rtw89_write16_mdio_mask(rtwdev, RAC_REG_FLD_0, BAC_AUTOK_N_MASK,
@@ -2150,14 +2149,13 @@ static int rtw89_pci_autok_x(struct rtw89_dev *rtwdev)
 
 static int rtw89_pci_auto_refclk_cal(struct rtw89_dev *rtwdev, bool autook_en)
 {
-	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	enum rtw89_pcie_phy phy_rate;
 	u16 val16, mgn_set, div_set, tar;
 	u8 val8, bdr_ori;
 	bool l1_flag = false;
 	int ret = 0;
 
-	if (chip_id != RTL8852B && chip_id != RTL8851B)
+	if (!rtw89_is_rtl885xb(rtwdev))
 		return 0;
 
 	ret = rtw89_pci_read_config_byte(rtwdev, RTW89_PCIE_PHY_RATE, &val8);
@@ -2398,7 +2396,7 @@ static void rtw89_pci_l1off_pwroff(struct rtw89_dev *rtwdev)
 {
 	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 
-	if (chip_id != RTL8852A && chip_id != RTL8852B && chip_id != RTL8851B)
+	if (chip_id != RTL8852A && !rtw89_is_rtl885xb(rtwdev))
 		return;
 
 	rtw89_write32_clr(rtwdev, R_AX_PCIE_PS_CTRL, B_AX_L1OFF_PWR_OFF_EN);
@@ -2428,7 +2426,7 @@ static void rtw89_pci_aphy_pwrcut(struct rtw89_dev *rtwdev)
 {
 	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 
-	if (chip_id != RTL8852A && chip_id != RTL8852B && chip_id != RTL8851B)
+	if (chip_id != RTL8852A && !rtw89_is_rtl885xb(rtwdev))
 		return;
 
 	rtw89_write32_clr(rtwdev, R_AX_SYS_PW_CTRL, B_AX_PSUS_OFF_CAPC_EN);
@@ -2438,7 +2436,7 @@ static void rtw89_pci_hci_ldo(struct rtw89_dev *rtwdev)
 {
 	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)) {
 		rtw89_write32_set(rtwdev, R_AX_SYS_SDIO_CTRL,
 				  B_AX_PCIE_DIS_L2_CTRL_LDO_HCI);
 		rtw89_write32_clr(rtwdev, R_AX_SYS_SDIO_CTRL,
@@ -2451,9 +2449,7 @@ static void rtw89_pci_hci_ldo(struct rtw89_dev *rtwdev)
 
 static int rtw89_pci_dphy_delay(struct rtw89_dev *rtwdev)
 {
-	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
-
-	if (chip_id != RTL8852B && chip_id != RTL8851B)
+	if (!rtw89_is_rtl885xb(rtwdev))
 		return 0;
 
 	return rtw89_write16_mdio_mask(rtwdev, RAC_REG_REV2, BAC_CMU_EN_DLY_MASK,
@@ -2715,7 +2711,7 @@ static int rtw89_pci_mode_op(struct rtw89_dev *rtwdev)
 					   B_AX_PCIE_RX_APPLEN_MASK, 0);
 	}
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B) {
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)) {
 		rtw89_write32_mask(rtwdev, R_AX_PCIE_INIT_CFG1, B_AX_PCIE_MAX_TXDMA_MASK, tx_burst);
 		rtw89_write32_mask(rtwdev, R_AX_PCIE_INIT_CFG1, B_AX_PCIE_MAX_RXDMA_MASK, rx_burst);
 	} else if (chip_id == RTL8852C) {
@@ -2723,7 +2719,7 @@ static int rtw89_pci_mode_op(struct rtw89_dev *rtwdev)
 		rtw89_write32_mask(rtwdev, R_AX_HAXI_INIT_CFG1, B_AX_HAXI_MAX_RXDMA_MASK, rx_burst);
 	}
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B) {
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)) {
 		if (tag_mode == MAC_AX_TAG_SGL) {
 			val32 = rtw89_read32(rtwdev, R_AX_PCIE_INIT_CFG1) &
 					    ~B_AX_LATENCY_CONTROL;
@@ -2738,7 +2734,7 @@ static int rtw89_pci_mode_op(struct rtw89_dev *rtwdev)
 	rtw89_write32_mask(rtwdev, info->exp_ctrl_reg, info->max_tag_num_mask,
 			   info->multi_tag_num);
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B) {
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)) {
 		rtw89_write32_mask(rtwdev, R_AX_PCIE_INIT_CFG2, B_AX_WD_ITVL_IDLE,
 				   wd_dma_idle_intvl);
 		rtw89_write32_mask(rtwdev, R_AX_PCIE_INIT_CFG2, B_AX_WD_ITVL_ACT,
@@ -2951,7 +2947,7 @@ static int rtw89_pci_ops_mac_post_init_ax(struct rtw89_dev *rtwdev)
 		/* ltr sw trigger */
 		rtw89_write32_set(rtwdev, R_AX_LTR_CTRL_0, B_AX_APP_LTR_ACT);
 	}
-	if (chip_id == RTL8852A || chip_id == RTL8852B) {
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)) {
 		/* ADDR info 8-byte mode */
 		rtw89_write32_set(rtwdev, R_AX_TX_ADDRESS_INFO_MODE_SETTING,
 				  B_AX_HOST_ADDR_INFO_8B_SEL);
@@ -3760,7 +3756,7 @@ static void rtw89_pci_clkreq_set_ax(struct rtw89_dev *rtwdev, bool enable)
 	if (ret)
 		rtw89_err(rtwdev, "failed to set CLKREQ Delay\n");
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)) {
 		if (enable)
 			ret = rtw89_pci_config_byte_set(rtwdev,
 							RTW89_PCIE_L1_CTRL,
@@ -3812,7 +3808,7 @@ static void rtw89_pci_aspm_set_ax(struct rtw89_dev *rtwdev, bool enable)
 	if (ret)
 		rtw89_warn(rtwdev, "failed to read ASPM Delay\n");
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)) {
 		if (enable)
 			ret = rtw89_pci_config_byte_set(rtwdev,
 							RTW89_PCIE_L1_CTRL,
@@ -3911,7 +3907,7 @@ static void rtw89_pci_l1ss_set_ax(struct rtw89_dev *rtwdev, bool enable)
 	enum rtw89_core_chip_id chip_id = rtwdev->chip->chip_id;
 	int ret;
 
-	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)) {
 		if (enable)
 			ret = rtw89_pci_config_byte_set(rtwdev,
 							RTW89_PCIE_TIMER_CTRL,
@@ -4108,7 +4104,7 @@ static int __maybe_unused rtw89_pci_suspend(struct device *dev)
 	rtw89_write32_set(rtwdev, R_AX_RSV_CTRL, B_AX_WLOCK_1C_BIT6);
 	rtw89_write32_set(rtwdev, R_AX_RSV_CTRL, B_AX_R_DIS_PRST);
 	rtw89_write32_clr(rtwdev, R_AX_RSV_CTRL, B_AX_WLOCK_1C_BIT6);
-	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)) {
 		rtw89_write32_clr(rtwdev, R_AX_SYS_SDIO_CTRL,
 				  B_AX_PCIE_DIS_L2_CTRL_LDO_HCI);
 		rtw89_write32_set(rtwdev, R_AX_PCIE_INIT_CFG1,
@@ -4142,7 +4138,7 @@ static int __maybe_unused rtw89_pci_resume(struct device *dev)
 	rtw89_write32_set(rtwdev, R_AX_RSV_CTRL, B_AX_WLOCK_1C_BIT6);
 	rtw89_write32_clr(rtwdev, R_AX_RSV_CTRL, B_AX_R_DIS_PRST);
 	rtw89_write32_clr(rtwdev, R_AX_RSV_CTRL, B_AX_WLOCK_1C_BIT6);
-	if (chip_id == RTL8852A || chip_id == RTL8852B || chip_id == RTL8851B) {
+	if (chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)) {
 		rtw89_write32_set(rtwdev, R_AX_SYS_SDIO_CTRL,
 				  B_AX_PCIE_DIS_L2_CTRL_LDO_HCI);
 		rtw89_write32_clr(rtwdev, R_AX_PCIE_INIT_CFG1,
diff --git a/drivers/net/wireless/realtek/rtw89/phy.c b/drivers/net/wireless/realtek/rtw89/phy.c
index a82b4c56a6f4..60f20e596321 100644
--- a/drivers/net/wireless/realtek/rtw89/phy.c
+++ b/drivers/net/wireless/realtek/rtw89/phy.c
@@ -1676,7 +1676,7 @@ static void rtw89_phy_preinit_rf_nctl_ax(struct rtw89_dev *rtwdev)
 	rtw89_phy_write32_set(rtwdev, R_P0_PATH_RST, 0x8000000);
 	if (chip->chip_id != RTL8851B)
 		rtw89_phy_write32_set(rtwdev, R_P1_PATH_RST, 0x8000000);
-	if (chip->chip_id == RTL8852B)
+	if (chip->chip_id == RTL8852B || chip->chip_id == RTL8852BT)
 		rtw89_phy_write32_set(rtwdev, R_IOQ_IQK_DPK, 0x2);
 
 	/* check 0x8080 */
diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index dc78ffc39ae7..5a59234301ca 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -3169,6 +3169,8 @@
 #define R_AX_DLK_PROTECT_CTL_C1 0xEE02
 #define B_AX_RX_DLK_CCA_TIME_MASK GENMASK(15, 8)
 #define B_AX_RX_DLK_DATA_TIME_MASK GENMASK(7, 4)
+#define B_AX_RX_DLK_RST_EN BIT(1)
+#define B_AX_RX_DLK_INT_EN BIT(0)
 
 #define R_AX_PLCP_HDR_FLTR 0xCE04
 #define R_AX_PLCP_HDR_FLTR_C1 0xEE04
-- 
cgit v1.2.3-58-ga151


From 8a00f7dfed62f1f74388bbd659e8d69d66886786 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 7 Jun 2024 22:02:51 +0800
Subject: wifi: rtw89: adopt firmware whose version is equal or less but
 closest

Version C of 8922AE hardware will use the same firmware of version B, so
extend rule of firmware recognition to allow less but closest firmware
version. Originally only accept firmware with matched version.

Tested on version A/B/C of 8922AE.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240607140251.8295-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c | 34 +++++++++++++++++++++++++--------
 1 file changed, 26 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index be39a8468d32..23204b2706c9 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -462,7 +462,7 @@ int rtw89_mfw_recognize(struct rtw89_dev *rtwdev, enum rtw89_fw_type type,
 	const u8 *mfw = firmware->data;
 	u32 mfw_len = firmware->size;
 	const struct rtw89_mfw_hdr *mfw_hdr = (const struct rtw89_mfw_hdr *)mfw;
-	const struct rtw89_mfw_info *mfw_info;
+	const struct rtw89_mfw_info *mfw_info = NULL, *tmp;
 	int i;
 
 	if (mfw_hdr->sig != RTW89_MFW_SIG) {
@@ -476,15 +476,27 @@ int rtw89_mfw_recognize(struct rtw89_dev *rtwdev, enum rtw89_fw_type type,
 	}
 
 	for (i = 0; i < mfw_hdr->fw_nr; i++) {
-		mfw_info = &mfw_hdr->info[i];
-		if (mfw_info->type == type) {
-			if (mfw_info->cv == rtwdev->hal.cv && !mfw_info->mp)
-				goto found;
-			if (type == RTW89_FW_LOGFMT)
-				goto found;
+		tmp = &mfw_hdr->info[i];
+		if (tmp->type != type)
+			continue;
+
+		if (type == RTW89_FW_LOGFMT) {
+			mfw_info = tmp;
+			goto found;
+		}
+
+		/* Version order of WiFi firmware in firmware file are not in order,
+		 * pass all firmware to find the equal or less but closest version.
+		 */
+		if (tmp->cv <= rtwdev->hal.cv && !tmp->mp) {
+			if (!mfw_info || mfw_info->cv < tmp->cv)
+				mfw_info = tmp;
 		}
 	}
 
+	if (mfw_info)
+		goto found;
+
 	if (!nowarn)
 		rtw89_err(rtwdev, "no suitable firmware found\n");
 	return -ENOENT;
@@ -606,10 +618,16 @@ int __rtw89_fw_recognize_from_elm(struct rtw89_dev *rtwdev,
 	struct rtw89_hal *hal = &rtwdev->hal;
 	struct rtw89_fw_suit *fw_suit;
 
-	if (hal->cv != elm->u.bbmcu.cv)
+	/* Version of BB MCU is in decreasing order in firmware file, so take
+	 * first equal or less version, which is equal or less but closest version.
+	 */
+	if (hal->cv < elm->u.bbmcu.cv)
 		return 1; /* ignore this element */
 
 	fw_suit = rtw89_fw_suit_get(rtwdev, type);
+	if (fw_suit->data)
+		return 1; /* ignore this element (a firmware is taken already) */
+
 	fw_suit->data = elm->u.bbmcu.contents;
 	fw_suit->size = le32_to_cpu(elm->size);
 
-- 
cgit v1.2.3-58-ga151


From 1fd4b3fe52efd5ad1647966f619c10988e7a4457 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Tue, 11 Jun 2024 10:19:00 +0800
Subject: wifi: rtw89: pci: support 36-bit PCI DMA address

Modern platforms can install more than 4GB memory, so DMA address can
larger than 32 bits. If a platform doesn't enable IOMMU, kernel needs extra
works of swiotlb to help DMA that packets reside on memory over 4GB.

The DMA addressing capability of Realtek WiFi chips is 36 bits, so set
LSB 4 bits of high 32-bit address to register and TX/RX descriptor, which
below figure shows 3-level pointers in TX direction, and RX direction is
similar but 2-level pointers only.

  +--------+
  |        | register to head of TX BD
  +---|----+
      |       +---------+
      +-----> |  TX BD  | (in memory)
              +----|----+
                   |        +---------+
                   +------> |  TX WD  | (in memory)
                            +----|----+
                                 |        +--------+
                                 +------> |   skb  |
                                          +--------+

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240611021901.26394-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/pci.c       | 85 +++++++++++++++++++++-----
 drivers/net/wireless/realtek/rtw89/pci.h       | 24 ++++++--
 drivers/net/wireless/realtek/rtw89/rtw8851be.c |  1 +
 drivers/net/wireless/realtek/rtw89/rtw8852ae.c |  1 +
 drivers/net/wireless/realtek/rtw89/rtw8852be.c |  1 +
 drivers/net/wireless/realtek/rtw89/rtw8852ce.c |  1 +
 drivers/net/wireless/realtek/rtw89/rtw8922ae.c |  1 +
 7 files changed, 95 insertions(+), 19 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/pci.c b/drivers/net/wireless/realtek/rtw89/pci.c
index b189e1ed98a8..ccb4bfe1a909 100644
--- a/drivers/net/wireless/realtek/rtw89/pci.c
+++ b/drivers/net/wireless/realtek/rtw89/pci.c
@@ -1298,10 +1298,12 @@ u32 rtw89_pci_fill_txaddr_info(struct rtw89_dev *rtwdev,
 			       dma_addr_t dma, u8 *add_info_nr)
 {
 	struct rtw89_pci_tx_addr_info_32 *txaddr_info = txaddr_info_addr;
+	__le16 option;
 
 	txaddr_info->length = cpu_to_le16(total_len);
-	txaddr_info->option = cpu_to_le16(RTW89_PCI_ADDR_MSDU_LS |
-					  RTW89_PCI_ADDR_NUM(1));
+	option = cpu_to_le16(RTW89_PCI_ADDR_MSDU_LS | RTW89_PCI_ADDR_NUM(1));
+	option |= le16_encode_bits(upper_32_bits(dma), RTW89_PCI_ADDR_HIGH_MASK);
+	txaddr_info->option = option;
 	txaddr_info->dma = cpu_to_le32(dma);
 
 	*add_info_nr = 1;
@@ -1328,6 +1330,8 @@ u32 rtw89_pci_fill_txaddr_info_v1(struct rtw89_dev *rtwdev,
 		length_option = FIELD_PREP(B_PCIADDR_LEN_V1_MASK, len) |
 				FIELD_PREP(B_PCIADDR_HIGH_SEL_V1_MASK, 0) |
 				FIELD_PREP(B_PCIADDR_LS_V1_MASK, remain == 0);
+		length_option |= u16_encode_bits(upper_32_bits(dma),
+						 B_PCIADDR_HIGH_SEL_V1_MASK);
 		txaddr_info->length_opt = cpu_to_le16(length_option);
 		txaddr_info->dma_low_lsb = cpu_to_le16(FIELD_GET(GENMASK(15, 0), dma));
 		txaddr_info->dma_low_msb = cpu_to_le16(FIELD_GET(GENMASK(31, 16), dma));
@@ -1418,6 +1422,7 @@ static int rtw89_pci_fwcmd_submit(struct rtw89_dev *rtwdev,
 	struct sk_buff *skb = tx_req->skb;
 	struct rtw89_pci_tx_data *tx_data = RTW89_PCI_TX_SKB_CB(skb);
 	dma_addr_t dma;
+	__le16 opt;
 
 	txdesc = skb_push(skb, txdesc_size);
 	memset(txdesc, 0, txdesc_size);
@@ -1430,7 +1435,9 @@ static int rtw89_pci_fwcmd_submit(struct rtw89_dev *rtwdev,
 	}
 
 	tx_data->dma = dma;
-	txbd->option = cpu_to_le16(RTW89_PCI_TXBD_OPTION_LS);
+	opt = cpu_to_le16(RTW89_PCI_TXBD_OPT_LS);
+	opt |= le16_encode_bits(upper_32_bits(dma), RTW89_PCI_TXBD_OPT_DMA_HI);
+	txbd->opt = opt;
 	txbd->length = cpu_to_le16(skb->len);
 	txbd->dma = cpu_to_le32(tx_data->dma);
 	skb_queue_tail(&rtwpci->h2c_queue, skb);
@@ -1446,6 +1453,7 @@ static int rtw89_pci_txbd_submit(struct rtw89_dev *rtwdev,
 				 struct rtw89_core_tx_request *tx_req)
 {
 	struct rtw89_pci_tx_wd *txwd;
+	__le16 opt;
 	int ret;
 
 	/* FWCMD queue doesn't have wd pages. Instead, it submits the CMD
@@ -1470,7 +1478,9 @@ static int rtw89_pci_txbd_submit(struct rtw89_dev *rtwdev,
 
 	list_add_tail(&txwd->list, &tx_ring->busy_pages);
 
-	txbd->option = cpu_to_le16(RTW89_PCI_TXBD_OPTION_LS);
+	opt = cpu_to_le16(RTW89_PCI_TXBD_OPT_LS);
+	opt |= le16_encode_bits(upper_32_bits(txwd->paddr), RTW89_PCI_TXBD_OPT_DMA_HI);
+	txbd->opt = opt;
 	txbd->length = cpu_to_le16(txwd->len);
 	txbd->dma = cpu_to_le32(txwd->paddr);
 
@@ -1569,6 +1579,25 @@ const struct rtw89_pci_bd_ram rtw89_bd_ram_table_single[RTW89_TXCH_NUM] = {
 };
 EXPORT_SYMBOL(rtw89_bd_ram_table_single);
 
+static void rtw89_pci_init_wp_16sel(struct rtw89_dev *rtwdev)
+{
+	const struct rtw89_pci_info *info = rtwdev->pci_info;
+	u32 addr = info->wp_sel_addr;
+	u32 val;
+	int i;
+
+	if (!info->wp_sel_addr)
+		return;
+
+	for (i = 0; i < 16; i += 4) {
+		val = u32_encode_bits(i + 0, MASKBYTE0) |
+		      u32_encode_bits(i + 1, MASKBYTE1) |
+		      u32_encode_bits(i + 2, MASKBYTE2) |
+		      u32_encode_bits(i + 3, MASKBYTE3);
+		rtw89_write32(rtwdev, addr + i, val);
+	}
+}
+
 static void rtw89_pci_reset_trx_rings(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_pci *rtwpci = (struct rtw89_pci *)rtwdev->priv;
@@ -1607,6 +1636,7 @@ static void rtw89_pci_reset_trx_rings(struct rtw89_dev *rtwdev)
 			rtw89_write32(rtwdev, addr_bdram, val32);
 		}
 		rtw89_write32(rtwdev, addr_desa_l, bd_ring->dma);
+		rtw89_write32(rtwdev, addr_desa_l + 4, upper_32_bits(bd_ring->dma));
 	}
 
 	for (i = 0; i < RTW89_RXCH_NUM; i++) {
@@ -1626,10 +1656,13 @@ static void rtw89_pci_reset_trx_rings(struct rtw89_dev *rtwdev)
 
 		rtw89_write16(rtwdev, addr_num, bd_ring->len);
 		rtw89_write32(rtwdev, addr_desa_l, bd_ring->dma);
+		rtw89_write32(rtwdev, addr_desa_l + 4, upper_32_bits(bd_ring->dma));
 
 		if (info->rx_ring_eq_is_full)
 			rtw89_write16(rtwdev, addr_idx, bd_ring->wp);
 	}
+
+	rtw89_pci_init_wp_16sel(rtwdev);
 }
 
 static void rtw89_pci_release_tx_ring(struct rtw89_dev *rtwdev,
@@ -2990,6 +3023,27 @@ static void rtw89_pci_declaim_device(struct rtw89_dev *rtwdev,
 	pci_disable_device(pdev);
 }
 
+static void rtw89_pci_cfg_dac(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_pci *rtwpci = (struct rtw89_pci *)rtwdev->priv;
+	const struct rtw89_chip_info *chip = rtwdev->chip;
+
+	if (!rtwpci->enable_dac)
+		return;
+
+	switch (chip->chip_id) {
+	case RTL8852A:
+	case RTL8852B:
+	case RTL8851B:
+	case RTL8852BT:
+		break;
+	default:
+		return;
+	}
+
+	rtw89_pci_config_byte_set(rtwdev, RTW89_PCIE_L1_CTRL, RTW89_PCIE_BIT_EN_64BITS);
+}
+
 static int rtw89_pci_setup_mapping(struct rtw89_dev *rtwdev,
 				   struct pci_dev *pdev)
 {
@@ -3004,16 +3058,17 @@ static int rtw89_pci_setup_mapping(struct rtw89_dev *rtwdev,
 		goto err;
 	}
 
-	ret = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32));
-	if (ret) {
-		rtw89_err(rtwdev, "failed to set dma mask to 32-bit\n");
-		goto err_release_regions;
-	}
-
-	ret = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
-	if (ret) {
-		rtw89_err(rtwdev, "failed to set consistent dma mask to 32-bit\n");
-		goto err_release_regions;
+	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(36));
+	if (!ret) {
+		rtwpci->enable_dac = true;
+		rtw89_pci_cfg_dac(rtwdev);
+	} else {
+		ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+		if (ret) {
+			rtw89_err(rtwdev,
+				  "failed to set dma and consistent mask to 32/36-bit\n");
+			goto err_release_regions;
+		}
 	}
 
 	resource_len = pci_resource_len(pdev, bar_id);
@@ -3164,6 +3219,7 @@ static int rtw89_pci_init_rx_bd(struct rtw89_dev *rtwdev, struct pci_dev *pdev,
 	memset(rx_bd, 0, sizeof(*rx_bd));
 	rx_bd->buf_size = cpu_to_le16(buf_sz);
 	rx_bd->dma = cpu_to_le32(dma);
+	rx_bd->opt = le16_encode_bits(upper_32_bits(dma), RTW89_PCI_RXBD_OPT_DMA_HI);
 	rx_info->dma = dma;
 
 	return 0;
@@ -4151,6 +4207,7 @@ static int __maybe_unused rtw89_pci_resume(struct device *dev)
 	}
 	rtw89_pci_l2_hci_ldo(rtwdev);
 	rtw89_pci_disable_eq(rtwdev);
+	rtw89_pci_cfg_dac(rtwdev);
 	rtw89_pci_filter_out(rtwdev);
 	rtw89_pci_link_cfg(rtwdev);
 	rtw89_pci_l1ss_cfg(rtwdev);
diff --git a/drivers/net/wireless/realtek/rtw89/pci.h b/drivers/net/wireless/realtek/rtw89/pci.h
index 7666753ae983..48c3ab735db2 100644
--- a/drivers/net/wireless/realtek/rtw89/pci.h
+++ b/drivers/net/wireless/realtek/rtw89/pci.h
@@ -724,6 +724,11 @@
 #define B_AX_CH11_BUSY			BIT(1)
 #define B_AX_CH10_BUSY			BIT(0)
 
+#define R_AX_WP_ADDR_H_SEL0_3 0x1334
+#define R_AX_WP_ADDR_H_SEL4_7 0x1338
+#define R_AX_WP_ADDR_H_SEL8_11 0x133C
+#define R_AX_WP_ADDR_H_SEL12_15 0x1340
+
 #define R_BE_HAXI_DMA_STOP1 0xB010
 #define B_BE_STOP_WPDMA BIT(31)
 #define B_BE_STOP_CH14 BIT(14)
@@ -823,6 +828,11 @@
 #define R_BE_RPQ0_RXBD_DESA_L_V1 0xB308
 #define R_BE_RPQ0_RXBD_DESA_H_V1 0xB30C
 
+#define R_BE_WP_ADDR_H_SEL0_3_V1 0xB420
+#define R_BE_WP_ADDR_H_SEL4_7_V1 0xB424
+#define R_BE_WP_ADDR_H_SEL8_11_V1 0xB428
+#define R_BE_WP_ADDR_H_SEL12_15_V1 0xB42C
+
 /* Configure */
 #define R_AX_PCIE_INIT_CFG2		0x1004
 #define B_AX_WD_ITVL_IDLE		GENMASK(27, 24)
@@ -1055,6 +1065,7 @@
 #define RTW89_PCIE_TIMER_CTRL		0x0718
 #define RTW89_PCIE_BIT_L1SUB		BIT(5)
 #define RTW89_PCIE_L1_CTRL		0x0719
+#define RTW89_PCIE_BIT_EN_64BITS	BIT(5)
 #define RTW89_PCIE_BIT_CLK		BIT(4)
 #define RTW89_PCIE_BIT_L1		BIT(3)
 #define RTW89_PCIE_CLK_CTRL		0x0725
@@ -1304,6 +1315,7 @@ struct rtw89_pci_info {
 	u32 rpwm_addr;
 	u32 cpwm_addr;
 	u32 mit_addr;
+	u32 wp_sel_addr;
 	u32 tx_dma_ch_mask;
 	const struct rtw89_pci_bd_idx_addr *bd_idx_addr_low_power;
 	const struct rtw89_pci_ch_dma_addr_set *dma_addr_set;
@@ -1330,11 +1342,11 @@ struct rtw89_pci_rx_info {
 	u32 fs:1, ls:1, tag:13, len:14;
 };
 
-#define RTW89_PCI_TXBD_OPTION_LS	BIT(14)
-
 struct rtw89_pci_tx_bd_32 {
 	__le16 length;
-	__le16 option;
+	__le16 opt;
+#define RTW89_PCI_TXBD_OPT_LS		BIT(14)
+#define RTW89_PCI_TXBD_OPT_DMA_HI	GENMASK(13, 6)
 	__le32 dma;
 } __packed;
 
@@ -1349,7 +1361,7 @@ struct rtw89_pci_tx_wp_info {
 
 #define RTW89_PCI_ADDR_MSDU_LS		BIT(15)
 #define RTW89_PCI_ADDR_LS		BIT(14)
-#define RTW89_PCI_ADDR_HIGH(a)		(((a) << 6) & GENMASK(13, 6))
+#define RTW89_PCI_ADDR_HIGH_MASK	GENMASK(13, 6)
 #define RTW89_PCI_ADDR_NUM(x)		((x) & GENMASK(5, 0))
 
 struct rtw89_pci_tx_addr_info_32 {
@@ -1386,7 +1398,8 @@ struct rtw89_pci_rpp_fmt {
 
 struct rtw89_pci_rx_bd_32 {
 	__le16 buf_size;
-	__le16 rsvd;
+	__le16 opt;
+#define RTW89_PCI_RXBD_OPT_DMA_HI	GENMASK(13, 6)
 	__le32 dma;
 } __packed;
 
@@ -1475,6 +1488,7 @@ struct rtw89_pci {
 	bool running;
 	bool low_power;
 	bool under_recovery;
+	bool enable_dac;
 	struct rtw89_pci_tx_ring tx_rings[RTW89_TXCH_NUM];
 	struct rtw89_pci_rx_ring rx_rings[RTW89_RXCH_NUM];
 	struct sk_buff_head h2c_queue;
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851be.c b/drivers/net/wireless/realtek/rtw89/rtw8851be.c
index ec3629d95fda..d334924faec8 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8851be.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8851be.c
@@ -46,6 +46,7 @@ static const struct rtw89_pci_info rtw8851b_pci_info = {
 	.rpwm_addr		= R_AX_PCIE_HRPWM,
 	.cpwm_addr		= R_AX_CPWM,
 	.mit_addr		= R_AX_INT_MIT_RX,
+	.wp_sel_addr		= 0,
 	.tx_dma_ch_mask		= BIT(RTW89_TXCH_ACH4) | BIT(RTW89_TXCH_ACH5) |
 				  BIT(RTW89_TXCH_ACH6) | BIT(RTW89_TXCH_ACH7) |
 				  BIT(RTW89_TXCH_CH10) | BIT(RTW89_TXCH_CH11),
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852ae.c b/drivers/net/wireless/realtek/rtw89/rtw8852ae.c
index fdee5dd4ba14..9a675e2193bc 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852ae.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852ae.c
@@ -46,6 +46,7 @@ static const struct rtw89_pci_info rtw8852a_pci_info = {
 	.rpwm_addr		= R_AX_PCIE_HRPWM,
 	.cpwm_addr		= R_AX_CPWM,
 	.mit_addr		= R_AX_INT_MIT_RX,
+	.wp_sel_addr		= 0,
 	.tx_dma_ch_mask		= 0,
 	.bd_idx_addr_low_power	= NULL,
 	.dma_addr_set		= &rtw89_pci_ch_dma_addr_set,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852be.c b/drivers/net/wireless/realtek/rtw89/rtw8852be.c
index 5f941122655c..d8f9d92ca0fb 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852be.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852be.c
@@ -46,6 +46,7 @@ static const struct rtw89_pci_info rtw8852b_pci_info = {
 	.rpwm_addr		= R_AX_PCIE_HRPWM,
 	.cpwm_addr		= R_AX_CPWM,
 	.mit_addr		= R_AX_INT_MIT_RX,
+	.wp_sel_addr		= 0,
 	.tx_dma_ch_mask		= BIT(RTW89_TXCH_ACH4) | BIT(RTW89_TXCH_ACH5) |
 				  BIT(RTW89_TXCH_ACH6) | BIT(RTW89_TXCH_ACH7) |
 				  BIT(RTW89_TXCH_CH10) | BIT(RTW89_TXCH_CH11),
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852ce.c b/drivers/net/wireless/realtek/rtw89/rtw8852ce.c
index e07c7f3ade41..8aaad7d58c0d 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852ce.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852ce.c
@@ -55,6 +55,7 @@ static const struct rtw89_pci_info rtw8852c_pci_info = {
 	.rpwm_addr		= R_AX_PCIE_HRPWM_V1,
 	.cpwm_addr		= R_AX_PCIE_CRPWM,
 	.mit_addr		= R_AX_INT_MIT_RX_V1,
+	.wp_sel_addr		= R_AX_WP_ADDR_H_SEL0_3,
 	.tx_dma_ch_mask		= 0,
 	.bd_idx_addr_low_power	= &rtw8852c_bd_idx_addr_low_power,
 	.dma_addr_set		= &rtw89_pci_ch_dma_addr_set_v1,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8922ae.c b/drivers/net/wireless/realtek/rtw89/rtw8922ae.c
index ce8aaa9501e1..47f855a7a268 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8922ae.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8922ae.c
@@ -46,6 +46,7 @@ static const struct rtw89_pci_info rtw8922a_pci_info = {
 	.rpwm_addr		= R_BE_PCIE_HRPWM,
 	.cpwm_addr		= R_BE_PCIE_CRPWM,
 	.mit_addr		= R_BE_PCIE_MIT_CH_EN,
+	.wp_sel_addr		= R_BE_WP_ADDR_H_SEL0_3_V1,
 	.tx_dma_ch_mask		= 0,
 	.bd_idx_addr_low_power	= NULL,
 	.dma_addr_set		= &rtw89_pci_ch_dma_addr_set_be,
-- 
cgit v1.2.3-58-ga151


From 94298477f81a1701fc4e1b5a0ce9672acab5dcb2 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Tue, 11 Jun 2024 10:19:01 +0800
Subject: wifi: rtw89: pci: fix RX tag race condition resulting in wrong RX
 length

Read 32 bits RX info to a local variable to fix race condition between
reading RX length and RX tag.

Another solution is to get RX tag at first statement, but adopted solution
can save some memory read, and also save 15 bytes binary code.

RX tag, a sequence number, is used to ensure that RX data has been DMA to
memory completely, so driver must check sequence number is expected before
reading other data.

This potential problem happens only after enabling 36-bit DMA.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://msgid.link/20240611021901.26394-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/pci.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/pci.c b/drivers/net/wireless/realtek/rtw89/pci.c
index ccb4bfe1a909..02afeb3acce4 100644
--- a/drivers/net/wireless/realtek/rtw89/pci.c
+++ b/drivers/net/wireless/realtek/rtw89/pci.c
@@ -183,14 +183,17 @@ static void rtw89_pci_sync_skb_for_device(struct rtw89_dev *rtwdev,
 static void rtw89_pci_rxbd_info_update(struct rtw89_dev *rtwdev,
 				       struct sk_buff *skb)
 {
-	struct rtw89_pci_rxbd_info *rxbd_info;
 	struct rtw89_pci_rx_info *rx_info = RTW89_PCI_RX_SKB_CB(skb);
+	struct rtw89_pci_rxbd_info *rxbd_info;
+	__le32 info;
 
 	rxbd_info = (struct rtw89_pci_rxbd_info *)skb->data;
-	rx_info->fs = le32_get_bits(rxbd_info->dword, RTW89_PCI_RXBD_FS);
-	rx_info->ls = le32_get_bits(rxbd_info->dword, RTW89_PCI_RXBD_LS);
-	rx_info->len = le32_get_bits(rxbd_info->dword, RTW89_PCI_RXBD_WRITE_SIZE);
-	rx_info->tag = le32_get_bits(rxbd_info->dword, RTW89_PCI_RXBD_TAG);
+	info = rxbd_info->dword;
+
+	rx_info->fs = le32_get_bits(info, RTW89_PCI_RXBD_FS);
+	rx_info->ls = le32_get_bits(info, RTW89_PCI_RXBD_LS);
+	rx_info->len = le32_get_bits(info, RTW89_PCI_RXBD_WRITE_SIZE);
+	rx_info->tag = le32_get_bits(info, RTW89_PCI_RXBD_TAG);
 }
 
 static int rtw89_pci_validate_rx_tag(struct rtw89_dev *rtwdev,
-- 
cgit v1.2.3-58-ga151


From 0a3e5c1b670f8c386bbd232ac30d74da3653a70a Mon Sep 17 00:00:00 2001
From: Joe Damato <jdamato@fastly.com>
Date: Wed, 12 Jun 2024 20:08:56 +0000
Subject: net/mlx5e: Add txq to sq stats mapping

mlx5 currently maps txqs to an sq via priv->txq2sq. It is useful to map
txqs to sq_stats, as well, for direct access to stats.

Add priv->txq2sq_stats and insert mappings. The mappings will be used
next to tabulate stats information.

Signed-off-by: Joe Damato <jdamato@fastly.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h      |  2 ++
 drivers/net/ethernet/mellanox/mlx5/core/en/qos.c  | 13 +++++++++++--
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 11 ++++++++++-
 3 files changed, 23 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index bec784d25d7b..6a343a8f162f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -867,6 +867,8 @@ struct mlx5e_priv {
 	/* priv data path fields - start */
 	struct mlx5e_selq selq;
 	struct mlx5e_txqsq **txq2sq;
+	struct mlx5e_sq_stats **txq2sq_stats;
+
 #ifdef CONFIG_MLX5_CORE_EN_DCB
 	struct mlx5e_dcbx_dp       dcbx_dp;
 #endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/qos.c b/drivers/net/ethernet/mellanox/mlx5/core/en/qos.c
index 6743806b8480..f0744a45db92 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/qos.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/qos.c
@@ -170,6 +170,7 @@ int mlx5e_activate_qos_sq(void *data, u16 node_qid, u32 hw_id)
 	mlx5e_tx_disable_queue(netdev_get_tx_queue(priv->netdev, qid));
 
 	priv->txq2sq[qid] = sq;
+	priv->txq2sq_stats[qid] = sq->stats;
 
 	/* Make the change to txq2sq visible before the queue is started.
 	 * As mlx5e_xmit runs under a spinlock, there is an implicit ACQUIRE,
@@ -186,6 +187,7 @@ int mlx5e_activate_qos_sq(void *data, u16 node_qid, u32 hw_id)
 void mlx5e_deactivate_qos_sq(struct mlx5e_priv *priv, u16 qid)
 {
 	struct mlx5e_txqsq *sq;
+	u16 txq_ix;
 
 	sq = mlx5e_get_qos_sq(priv, qid);
 	if (!sq) /* Handle the case when the SQ failed to open. */
@@ -194,7 +196,10 @@ void mlx5e_deactivate_qos_sq(struct mlx5e_priv *priv, u16 qid)
 	qos_dbg(sq->mdev, "Deactivate QoS SQ qid %u\n", qid);
 	mlx5e_deactivate_txqsq(sq);
 
-	priv->txq2sq[mlx5e_qid_from_qos(&priv->channels, qid)] = NULL;
+	txq_ix = mlx5e_qid_from_qos(&priv->channels, qid);
+
+	priv->txq2sq[txq_ix] = NULL;
+	priv->txq2sq_stats[txq_ix] = NULL;
 
 	/* Make the change to txq2sq visible before the queue is started again.
 	 * As mlx5e_xmit runs under a spinlock, there is an implicit ACQUIRE,
@@ -325,6 +330,7 @@ void mlx5e_qos_deactivate_queues(struct mlx5e_channel *c)
 {
 	struct mlx5e_params *params = &c->priv->channels.params;
 	struct mlx5e_txqsq __rcu **qos_sqs;
+	u16 txq_ix;
 	int i;
 
 	qos_sqs = mlx5e_state_dereference(c->priv, c->qos_sqs);
@@ -342,8 +348,11 @@ void mlx5e_qos_deactivate_queues(struct mlx5e_channel *c)
 		qos_dbg(c->mdev, "Deactivate QoS SQ qid %u\n", qid);
 		mlx5e_deactivate_txqsq(sq);
 
+		txq_ix = mlx5e_qid_from_qos(&c->priv->channels, qid);
+
 		/* The queue is disabled, no synchronization with datapath is needed. */
-		c->priv->txq2sq[mlx5e_qid_from_qos(&c->priv->channels, qid)] = NULL;
+		c->priv->txq2sq[txq_ix] = NULL;
+		c->priv->txq2sq_stats[txq_ix] = NULL;
 	}
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index db4819f11541..445bd0f8c6a2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3125,6 +3125,7 @@ static void mlx5e_build_txq_maps(struct mlx5e_priv *priv)
 			struct mlx5e_txqsq *sq = &c->sq[tc];
 
 			priv->txq2sq[sq->txq_ix] = sq;
+			priv->txq2sq_stats[sq->txq_ix] = sq->stats;
 		}
 	}
 
@@ -3139,6 +3140,7 @@ static void mlx5e_build_txq_maps(struct mlx5e_priv *priv)
 		struct mlx5e_txqsq *sq = &c->ptpsq[tc].txqsq;
 
 		priv->txq2sq[sq->txq_ix] = sq;
+		priv->txq2sq_stats[sq->txq_ix] = sq->stats;
 	}
 
 out:
@@ -5848,9 +5850,13 @@ int mlx5e_priv_init(struct mlx5e_priv *priv,
 	if (!priv->txq2sq)
 		goto err_destroy_workqueue;
 
+	priv->txq2sq_stats = kcalloc_node(num_txqs, sizeof(*priv->txq2sq_stats), GFP_KERNEL, node);
+	if (!priv->txq2sq_stats)
+		goto err_free_txq2sq;
+
 	priv->tx_rates = kcalloc_node(num_txqs, sizeof(*priv->tx_rates), GFP_KERNEL, node);
 	if (!priv->tx_rates)
-		goto err_free_txq2sq;
+		goto err_free_txq2sq_stats;
 
 	priv->channel_stats =
 		kcalloc_node(nch, sizeof(*priv->channel_stats), GFP_KERNEL, node);
@@ -5861,6 +5867,8 @@ int mlx5e_priv_init(struct mlx5e_priv *priv,
 
 err_free_tx_rates:
 	kfree(priv->tx_rates);
+err_free_txq2sq_stats:
+	kfree(priv->txq2sq_stats);
 err_free_txq2sq:
 	kfree(priv->txq2sq);
 err_destroy_workqueue:
@@ -5884,6 +5892,7 @@ void mlx5e_priv_cleanup(struct mlx5e_priv *priv)
 		kvfree(priv->channel_stats[i]);
 	kfree(priv->channel_stats);
 	kfree(priv->tx_rates);
+	kfree(priv->txq2sq_stats);
 	kfree(priv->txq2sq);
 	destroy_workqueue(priv->wq);
 	mlx5e_selq_cleanup(&priv->selq);
-- 
cgit v1.2.3-58-ga151


From 7b66ae536a78216f9c4892533db2d76a8b60a81d Mon Sep 17 00:00:00 2001
From: Joe Damato <jdamato@fastly.com>
Date: Wed, 12 Jun 2024 20:08:57 +0000
Subject: net/mlx5e: Add per queue netdev-genl stats

./cli.py --spec netlink/specs/netdev.yaml \
         --dump qstats-get --json '{"scope": "queue"}'

...snip

 {'ifindex': 7,
  'queue-id': 62,
  'queue-type': 'rx',
  'rx-alloc-fail': 0,
  'rx-bytes': 105965251,
  'rx-packets': 179790},
 {'ifindex': 7,
  'queue-id': 0,
  'queue-type': 'tx',
  'tx-bytes': 9402665,
  'tx-packets': 17551},

...snip

Also tested with the script tools/testing/selftests/drivers/net/stats.py
in several scenarios to ensure stats tallying was correct:

- on boot (default queue counts)
- adjusting queue count up or down (ethtool -L eth0 combined ...)

The tools/testing/selftests/drivers/net/stats.py brings the device up,
so to test with the device down, I did the following:

$ ip link show eth4
7: eth4: <BROADCAST,MULTICAST> mtu 9000 qdisc mq state DOWN [..snip..]
  [..snip..]

$ cat /proc/net/dev | grep eth4
eth4: 235710489  434811 [..snip rx..] 2878744 21227  [..snip tx..]

$ ./cli.py --spec ../../../Documentation/netlink/specs/netdev.yaml \
           --dump qstats-get --json '{"ifindex": 7}'
[{'ifindex': 7,
  'rx-alloc-fail': 0,
  'rx-bytes': 235710489,
  'rx-packets': 434811,
  'tx-bytes': 2878744,
  'tx-packets': 21227}]

Compare the values in /proc/net/dev match the output of cli for the same
device, even while the device is down.

Note that while the device is down, per queue stats output nothing
(because the device is down there are no queues):

$ ./cli.py --spec ../../../Documentation/netlink/specs/netdev.yaml \
           --dump qstats-get --json '{"scope": "queue", "ifindex": 7}'
[]

Signed-off-by: Joe Damato <jdamato@fastly.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 132 ++++++++++++++++++++++
 1 file changed, 132 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 445bd0f8c6a2..0eba4c5bb2ec 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -39,6 +39,7 @@
 #include <linux/debugfs.h>
 #include <linux/if_bridge.h>
 #include <linux/filter.h>
+#include <net/netdev_queues.h>
 #include <net/page_pool/types.h>
 #include <net/pkt_sched.h>
 #include <net/xdp_sock_drv.h>
@@ -5298,6 +5299,136 @@ static bool mlx5e_tunnel_any_tx_proto_supported(struct mlx5_core_dev *mdev)
 	return (mlx5_vxlan_allowed(mdev->vxlan) || mlx5_geneve_tx_allowed(mdev));
 }
 
+static void mlx5e_get_queue_stats_rx(struct net_device *dev, int i,
+				     struct netdev_queue_stats_rx *stats)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct mlx5e_channel_stats *channel_stats;
+	struct mlx5e_rq_stats *xskrq_stats;
+	struct mlx5e_rq_stats *rq_stats;
+
+	ASSERT_RTNL();
+	if (mlx5e_is_uplink_rep(priv))
+		return;
+
+	channel_stats = priv->channel_stats[i];
+	xskrq_stats = &channel_stats->xskrq;
+	rq_stats = &channel_stats->rq;
+
+	stats->packets = rq_stats->packets + xskrq_stats->packets;
+	stats->bytes = rq_stats->bytes + xskrq_stats->bytes;
+	stats->alloc_fail = rq_stats->buff_alloc_err +
+			    xskrq_stats->buff_alloc_err;
+}
+
+static void mlx5e_get_queue_stats_tx(struct net_device *dev, int i,
+				     struct netdev_queue_stats_tx *stats)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct mlx5e_sq_stats *sq_stats;
+
+	ASSERT_RTNL();
+	/* no special case needed for ptp htb etc since txq2sq_stats is kept up
+	 * to date for active sq_stats, otherwise get_base_stats takes care of
+	 * inactive sqs.
+	 */
+	sq_stats = priv->txq2sq_stats[i];
+	stats->packets = sq_stats->packets;
+	stats->bytes = sq_stats->bytes;
+}
+
+static void mlx5e_get_base_stats(struct net_device *dev,
+				 struct netdev_queue_stats_rx *rx,
+				 struct netdev_queue_stats_tx *tx)
+{
+	struct mlx5e_priv *priv = netdev_priv(dev);
+	struct mlx5e_ptp *ptp_channel;
+	int i, tc;
+
+	ASSERT_RTNL();
+	if (!mlx5e_is_uplink_rep(priv)) {
+		rx->packets = 0;
+		rx->bytes = 0;
+		rx->alloc_fail = 0;
+
+		for (i = priv->channels.params.num_channels; i < priv->stats_nch; i++) {
+			struct netdev_queue_stats_rx rx_i = {0};
+
+			mlx5e_get_queue_stats_rx(dev, i, &rx_i);
+
+			rx->packets += rx_i.packets;
+			rx->bytes += rx_i.bytes;
+			rx->alloc_fail += rx_i.alloc_fail;
+		}
+
+		/* always report PTP RX stats from base as there is no
+		 * corresponding channel to report them under in
+		 * mlx5e_get_queue_stats_rx.
+		 */
+		if (priv->rx_ptp_opened) {
+			struct mlx5e_rq_stats *rq_stats = &priv->ptp_stats.rq;
+
+			rx->packets += rq_stats->packets;
+			rx->bytes += rq_stats->bytes;
+		}
+	}
+
+	tx->packets = 0;
+	tx->bytes = 0;
+
+	for (i = 0; i < priv->stats_nch; i++) {
+		struct mlx5e_channel_stats *channel_stats = priv->channel_stats[i];
+
+		/* handle two cases:
+		 *
+		 *  1. channels which are active. In this case,
+		 *     report only deactivated TCs on these channels.
+		 *
+		 *  2. channels which were deactivated
+		 *     (i > priv->channels.params.num_channels)
+		 *     must have all of their TCs [0 .. priv->max_opened_tc)
+		 *     examined because deactivated channels will not be in the
+		 *     range of [0..real_num_tx_queues) and will not have their
+		 *     stats reported by mlx5e_get_queue_stats_tx.
+		 */
+		if (i < priv->channels.params.num_channels)
+			tc = mlx5e_get_dcb_num_tc(&priv->channels.params);
+		else
+			tc = 0;
+
+		for (; tc < priv->max_opened_tc; tc++) {
+			struct mlx5e_sq_stats *sq_stats = &channel_stats->sq[tc];
+
+			tx->packets += sq_stats->packets;
+			tx->bytes += sq_stats->bytes;
+		}
+	}
+
+	/* if PTP TX was opened at some point and has since either:
+	 *    -  been shutdown and set to NULL, or
+	 *    -  simply disabled (bit unset)
+	 *
+	 * report stats directly from the ptp_stats structures as these queues
+	 * are now unavailable and there is no txq index to retrieve these
+	 * stats via calls to mlx5e_get_queue_stats_tx.
+	 */
+	ptp_channel = priv->channels.ptp;
+	if (priv->tx_ptp_opened && (!ptp_channel || !test_bit(MLX5E_PTP_STATE_TX, ptp_channel->state))) {
+		for (tc = 0; tc < priv->max_opened_tc; tc++) {
+			struct mlx5e_sq_stats *sq_stats = &priv->ptp_stats.sq[tc];
+
+			tx->packets += sq_stats->packets;
+			tx->bytes   += sq_stats->bytes;
+		}
+	}
+}
+
+static const struct netdev_stat_ops mlx5e_stat_ops = {
+	.get_queue_stats_rx  = mlx5e_get_queue_stats_rx,
+	.get_queue_stats_tx  = mlx5e_get_queue_stats_tx,
+	.get_base_stats      = mlx5e_get_base_stats,
+};
+
 static void mlx5e_build_nic_netdev(struct net_device *netdev)
 {
 	struct mlx5e_priv *priv = netdev_priv(netdev);
@@ -5315,6 +5446,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
 
 	netdev->watchdog_timeo    = 15 * HZ;
 
+	netdev->stat_ops	  = &mlx5e_stat_ops;
 	netdev->ethtool_ops	  = &mlx5e_ethtool_ops;
 
 	netdev->vlan_features    |= NETIF_F_SG;
-- 
cgit v1.2.3-58-ga151


From 71839a929d9ee97daab319d370fc0f6720daef43 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Wed, 5 Jun 2024 14:21:04 +0200
Subject: dt-bindings: net: wireless: qcom,ath11k: describe the ath11k on
 QCA6390

Add a PCI compatible for the ATH11K module on QCA6390 and describe the
power inputs from the PMU that it consumes.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240605122106.23818-2-brgl@bgdev.pl
---
 .../bindings/net/wireless/qcom,ath11k-pci.yaml     | 46 ++++++++++++++++++++++
 1 file changed, 46 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml
index 41d023797d7d..8675d7d0215c 100644
--- a/Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml
+++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath11k-pci.yaml
@@ -17,6 +17,7 @@ description: |
 properties:
   compatible:
     enum:
+      - pci17cb,1101  # QCA6390
       - pci17cb,1103  # WCN6855
 
   reg:
@@ -28,10 +29,55 @@ properties:
       string to uniquely identify variant of the calibration data for designs
       with colliding bus and device ids
 
+  vddrfacmn-supply:
+    description: VDD_RFA_CMN supply regulator handle
+
+  vddaon-supply:
+    description: VDD_AON supply regulator handle
+
+  vddwlcx-supply:
+    description: VDD_WL_CX supply regulator handle
+
+  vddwlmx-supply:
+    description: VDD_WL_MX supply regulator handle
+
+  vddrfa0p8-supply:
+    description: VDD_RFA_0P8 supply regulator handle
+
+  vddrfa1p2-supply:
+    description: VDD_RFA_1P2 supply regulator handle
+
+  vddrfa1p7-supply:
+    description: VDD_RFA_1P7 supply regulator handle
+
+  vddpcie0p9-supply:
+    description: VDD_PCIE_0P9 supply regulator handle
+
+  vddpcie1p8-supply:
+    description: VDD_PCIE_1P8 supply regulator handle
+
 required:
   - compatible
   - reg
 
+allOf:
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: pci17cb,1101
+    then:
+      required:
+        - vddrfacmn-supply
+        - vddaon-supply
+        - vddwlcx-supply
+        - vddwlmx-supply
+        - vddrfa0p8-supply
+        - vddrfa1p2-supply
+        - vddrfa1p7-supply
+        - vddpcie0p9-supply
+        - vddpcie1p8-supply
+
 additionalProperties: false
 
 examples:
-- 
cgit v1.2.3-58-ga151


From aa17d384971b62a8a7dc901c96f615efc7328a62 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Wed, 5 Jun 2024 14:21:05 +0200
Subject: dt-bindings: net: wireless: describe the ath12k PCI module

Add device-tree bindings for the ATH12K module found in the WCN7850
package.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240605122106.23818-3-brgl@bgdev.pl
---
 .../bindings/net/wireless/qcom,ath12k.yaml         | 99 ++++++++++++++++++++++
 1 file changed, 99 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/wireless/qcom,ath12k.yaml

diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath12k.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath12k.yaml
new file mode 100644
index 000000000000..1b5884015b15
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath12k.yaml
@@ -0,0 +1,99 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+# Copyright (c) 2024 Linaro Limited
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/wireless/qcom,ath12k.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm Technologies ath12k wireless devices (PCIe)
+
+maintainers:
+  - Jeff Johnson <quic_jjohnson@quicinc.com>
+  - Kalle Valo <kvalo@kernel.org>
+
+description:
+  Qualcomm Technologies IEEE 802.11be PCIe devices.
+
+properties:
+  compatible:
+    enum:
+      - pci17cb,1107  # WCN7850
+
+  reg:
+    maxItems: 1
+
+  vddaon-supply:
+    description: VDD_AON supply regulator handle
+
+  vddwlcx-supply:
+    description: VDD_WLCX supply regulator handle
+
+  vddwlmx-supply:
+    description: VDD_WLMX supply regulator handle
+
+  vddrfacmn-supply:
+    description: VDD_RFA_CMN supply regulator handle
+
+  vddrfa0p8-supply:
+    description: VDD_RFA_0P8 supply regulator handle
+
+  vddrfa1p2-supply:
+    description: VDD_RFA_1P2 supply regulator handle
+
+  vddrfa1p8-supply:
+    description: VDD_RFA_1P8 supply regulator handle
+
+  vddpcie0p9-supply:
+    description: VDD_PCIE_0P9 supply regulator handle
+
+  vddpcie1p8-supply:
+    description: VDD_PCIE_1P8 supply regulator handle
+
+required:
+  - compatible
+  - reg
+  - vddaon-supply
+  - vddwlcx-supply
+  - vddwlmx-supply
+  - vddrfacmn-supply
+  - vddrfa0p8-supply
+  - vddrfa1p2-supply
+  - vddrfa1p8-supply
+  - vddpcie0p9-supply
+  - vddpcie1p8-supply
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/qcom,rpmh.h>
+    #include <dt-bindings/gpio/gpio.h>
+    pcie {
+        #address-cells = <3>;
+        #size-cells = <2>;
+
+        pcie@0 {
+            device_type = "pci";
+            reg = <0x0 0x0 0x0 0x0 0x0>;
+            #address-cells = <3>;
+            #size-cells = <2>;
+            ranges;
+
+            bus-range = <0x01 0xff>;
+
+            wifi@0 {
+                compatible = "pci17cb,1107";
+                reg = <0x10000 0x0 0x0 0x0 0x0>;
+
+                vddaon-supply = <&vreg_pmu_aon_0p59>;
+                vddwlcx-supply = <&vreg_pmu_wlcx_0p8>;
+                vddwlmx-supply = <&vreg_pmu_wlmx_0p85>;
+                vddrfacmn-supply = <&vreg_pmu_rfa_cmn>;
+                vddrfa0p8-supply = <&vreg_pmu_rfa_0p8>;
+                vddrfa1p2-supply = <&vreg_pmu_rfa_1p2>;
+                vddrfa1p8-supply = <&vreg_pmu_rfa_1p8>;
+                vddpcie0p9-supply = <&vreg_pmu_pcie_0p9>;
+                vddpcie1p8-supply = <&vreg_pmu_pcie_1p8>;
+            };
+        };
+    };
-- 
cgit v1.2.3-58-ga151


From 54fcc6189dfb822eea984fa2b3e477a02447279d Mon Sep 17 00:00:00 2001
From: Antony Antony <antony.antony@secunet.com>
Date: Tue, 11 Jun 2024 08:31:29 +0200
Subject: xfrm: Fix input error path memory access

When there is a misconfiguration of input state slow path
KASAN report error. Fix this error.
west login:
[   52.987278] eth1: renamed from veth11
[   53.078814] eth1: renamed from veth21
[   53.181355] eth1: renamed from veth31
[   54.921702] ==================================================================
[   54.922602] BUG: KASAN: wild-memory-access in xfrmi_rcv_cb+0x2d/0x295
[   54.923393] Read of size 8 at addr 6b6b6b6b00000000 by task ping/512
[   54.924169]
[   54.924386] CPU: 0 PID: 512 Comm: ping Not tainted 6.9.0-08574-gcd29a4313a1b #25
[   54.925290] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
[   54.926401] Call Trace:
[   54.926731]  <IRQ>
[   54.927009]  dump_stack_lvl+0x2a/0x3b
[   54.927478]  kasan_report+0x84/0xa6
[   54.927930]  ? xfrmi_rcv_cb+0x2d/0x295
[   54.928410]  xfrmi_rcv_cb+0x2d/0x295
[   54.928872]  ? xfrm4_rcv_cb+0x3d/0x5e
[   54.929354]  xfrm4_rcv_cb+0x46/0x5e
[   54.929804]  xfrm_rcv_cb+0x7e/0xa1
[   54.930240]  xfrm_input+0x1b3a/0x1b96
[   54.930715]  ? xfrm_offload+0x41/0x41
[   54.931182]  ? raw_rcv+0x292/0x292
[   54.931617]  ? nf_conntrack_confirm+0xa2/0xa2
[   54.932158]  ? skb_sec_path+0xd/0x3f
[   54.932610]  ? xfrmi_input+0x90/0xce
[   54.933066]  xfrm4_esp_rcv+0x33/0x54
[   54.933521]  ip_protocol_deliver_rcu+0xd7/0x1b2
[   54.934089]  ip_local_deliver_finish+0x110/0x120
[   54.934659]  ? ip_protocol_deliver_rcu+0x1b2/0x1b2
[   54.935248]  NF_HOOK.constprop.0+0xf8/0x138
[   54.935767]  ? ip_sublist_rcv_finish+0x68/0x68
[   54.936317]  ? secure_tcpv6_ts_off+0x23/0x168
[   54.936859]  ? ip_protocol_deliver_rcu+0x1b2/0x1b2
[   54.937454]  ? __xfrm_policy_check2.constprop.0+0x18d/0x18d
[   54.938135]  NF_HOOK.constprop.0+0xf8/0x138
[   54.938663]  ? ip_sublist_rcv_finish+0x68/0x68
[   54.939220]  ? __xfrm_policy_check2.constprop.0+0x18d/0x18d
[   54.939904]  ? ip_local_deliver_finish+0x120/0x120
[   54.940497]  __netif_receive_skb_one_core+0xc9/0x107
[   54.941121]  ? __netif_receive_skb_list_core+0x1c2/0x1c2
[   54.941771]  ? blk_mq_start_stopped_hw_queues+0xc7/0xf9
[   54.942413]  ? blk_mq_start_stopped_hw_queue+0x38/0x38
[   54.943044]  ? virtqueue_get_buf_ctx+0x295/0x46b
[   54.943618]  process_backlog+0xb3/0x187
[   54.944102]  __napi_poll.constprop.0+0x57/0x1a7
[   54.944669]  net_rx_action+0x1cb/0x380
[   54.945150]  ? __napi_poll.constprop.0+0x1a7/0x1a7
[   54.945744]  ? vring_new_virtqueue+0x17a/0x17a
[   54.946300]  ? note_interrupt+0x2cd/0x367
[   54.946805]  handle_softirqs+0x13c/0x2c9
[   54.947300]  do_softirq+0x5f/0x7d
[   54.947727]  </IRQ>
[   54.948014]  <TASK>
[   54.948300]  __local_bh_enable_ip+0x48/0x62
[   54.948832]  __neigh_event_send+0x3fd/0x4ca
[   54.949361]  neigh_resolve_output+0x1e/0x210
[   54.949896]  ip_finish_output2+0x4bf/0x4f0
[   54.950410]  ? __ip_finish_output+0x171/0x1b8
[   54.950956]  ip_send_skb+0x25/0x57
[   54.951390]  raw_sendmsg+0xf95/0x10c0
[   54.951850]  ? check_new_pages+0x45/0x71
[   54.952343]  ? raw_hash_sk+0x21b/0x21b
[   54.952815]  ? kernel_init_pages+0x42/0x51
[   54.953337]  ? prep_new_page+0x44/0x51
[   54.953811]  ? get_page_from_freelist+0x72b/0x915
[   54.954390]  ? signal_pending_state+0x77/0x77
[   54.954936]  ? preempt_count_sub+0x14/0xb3
[   54.955450]  ? __might_resched+0x8a/0x240
[   54.955951]  ? __might_sleep+0x25/0xa0
[   54.956424]  ? first_zones_zonelist+0x2c/0x43
[   54.956977]  ? __rcu_read_lock+0x2d/0x3a
[   54.957476]  ? __pte_offset_map+0x32/0xa4
[   54.957980]  ? __might_resched+0x8a/0x240
[   54.958483]  ? __might_sleep+0x25/0xa0
[   54.958963]  ? inet_send_prepare+0x54/0x54
[   54.959478]  ? sock_sendmsg_nosec+0x42/0x6c
[   54.960000]  sock_sendmsg_nosec+0x42/0x6c
[   54.960502]  __sys_sendto+0x15d/0x1cc
[   54.960966]  ? __x64_sys_getpeername+0x44/0x44
[   54.961522]  ? __handle_mm_fault+0x679/0xae4
[   54.962068]  ? find_vma+0x6b/0x8b
[   54.962497]  ? find_vma_intersection+0x8a/0x8a
[   54.963052]  ? handle_mm_fault+0x38/0x154
[   54.963556]  ? handle_mm_fault+0xeb/0x154
[   54.964059]  ? preempt_latency_start+0x29/0x34
[   54.964613]  ? preempt_count_sub+0x14/0xb3
[   54.965141]  ? up_read+0x4b/0x5c
[   54.965557]  __x64_sys_sendto+0x76/0x82
[   54.966041]  do_syscall_64+0x69/0xd5
[   54.966497]  entry_SYSCALL_64_after_hwframe+0x4b/0x53
[   54.967119] RIP: 0033:0x7f2d2fec9a73
[   54.967572] Code: 8b 15 a9 83 0c 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 80 3d 71 0b 0d 00 00 41 89 ca 74 14 b8 2c 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 75 c3 0f 1f 40 00 55 48 83 ec 30 44 89 4c 24
[   54.969747] RSP: 002b:00007ffe85756418 EFLAGS: 00000202 ORIG_RAX: 000000000000002c
[   54.970655] RAX: ffffffffffffffda RBX: 0000558bebad1340 RCX: 00007f2d2fec9a73
[   54.971511] RDX: 0000000000000040 RSI: 0000558bebad73c0 RDI: 0000000000000003
[   54.972366] RBP: 0000558bebad73c0 R08: 0000558bebad35c0 R09: 0000000000000010
[   54.973234] R10: 0000000000000000 R11: 0000000000000202 R12: 0000000000000040
[   54.974091] R13: 00007ffe85757b00 R14: 0000001d00000001 R15: 0000558bebad4680
[   54.974951]  </TASK>
[   54.975244] ==================================================================
[   54.976133] Disabling lock debugging due to kernel taint
[   54.976784] Oops: stack segment: 0000 [#1] PREEMPT DEBUG_PAGEALLOC KASAN
[   54.977603] CPU: 0 PID: 512 Comm: ping Tainted: G    B              6.9.0-08574-gcd29a4313a1b #25
[   54.978654] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
[   54.979750] RIP: 0010:xfrmi_rcv_cb+0x2d/0x295
[   54.980293] Code: 00 00 41 57 41 56 41 89 f6 41 55 41 54 55 53 48 89 fb 51 85 f6 75 31 48 89 df e8 d7 e8 ff ff 48 89 c5 48 89 c7 e8 8b a4 4f ff <48> 8b 7d 00 48 89 ee e8 eb f3 ff ff 49 89 c5 b8 01 00 00 00 4d 85
[   54.982462] RSP: 0018:ffffc90000007990 EFLAGS: 00010282
[   54.983099] RAX: 0000000000000001 RBX: ffff8881126e9900 RCX: fffffbfff07b77cd
[   54.983948] RDX: fffffbfff07b77cd RSI: fffffbfff07b77cd RDI: ffffffff83dbbe60
[   54.984794] RBP: 6b6b6b6b00000000 R08: 0000000000000008 R09: 0000000000000001
[   54.985647] R10: ffffffff83dbbe67 R11: fffffbfff07b77cc R12: 00000000ffffffff
[   54.986512] R13: 00000000ffffffff R14: 00000000ffffffff R15: 0000000000000002
[   54.987365] FS:  00007f2d2fc0dc40(0000) GS:ffffffff82eb2000(0000) knlGS:0000000000000000
[   54.988329] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   54.989026] CR2: 00007ffe85755ff8 CR3: 0000000109941000 CR4: 0000000000350ef0
[   54.989897] Call Trace:
[   54.990223]  <IRQ>
[   54.990500]  ? __die_body+0x1a/0x56
[   54.990950]  ? die+0x30/0x49
[   54.991326]  ? do_trap+0x9b/0x132
[   54.991751]  ? do_error_trap+0x7d/0xaf
[   54.992223]  ? exc_stack_segment+0x35/0x45
[   54.992734]  ? asm_exc_stack_segment+0x22/0x30
[   54.993294]  ? xfrmi_rcv_cb+0x2d/0x295
[   54.993764]  ? xfrm4_rcv_cb+0x3d/0x5e
[   54.994228]  xfrm4_rcv_cb+0x46/0x5e
[   54.994670]  xfrm_rcv_cb+0x7e/0xa1
[   54.995106]  xfrm_input+0x1b3a/0x1b96
[   54.995572]  ? xfrm_offload+0x41/0x41
[   54.996038]  ? raw_rcv+0x292/0x292
[   54.996472]  ? nf_conntrack_confirm+0xa2/0xa2
[   54.997011]  ? skb_sec_path+0xd/0x3f
[   54.997466]  ? xfrmi_input+0x90/0xce
[   54.997925]  xfrm4_esp_rcv+0x33/0x54
[   54.998378]  ip_protocol_deliver_rcu+0xd7/0x1b2
[   54.998944]  ip_local_deliver_finish+0x110/0x120
[   54.999520]  ? ip_protocol_deliver_rcu+0x1b2/0x1b2
[   55.000111]  NF_HOOK.constprop.0+0xf8/0x138
[   55.000630]  ? ip_sublist_rcv_finish+0x68/0x68
[   55.001195]  ? secure_tcpv6_ts_off+0x23/0x168
[   55.001743]  ? ip_protocol_deliver_rcu+0x1b2/0x1b2
[   55.002331]  ? __xfrm_policy_check2.constprop.0+0x18d/0x18d
[   55.003008]  NF_HOOK.constprop.0+0xf8/0x138
[   55.003527]  ? ip_sublist_rcv_finish+0x68/0x68
[   55.004078]  ? __xfrm_policy_check2.constprop.0+0x18d/0x18d
[   55.004755]  ? ip_local_deliver_finish+0x120/0x120
[   55.005351]  __netif_receive_skb_one_core+0xc9/0x107
[   55.005972]  ? __netif_receive_skb_list_core+0x1c2/0x1c2
[   55.006626]  ? blk_mq_start_stopped_hw_queues+0xc7/0xf9
[   55.007266]  ? blk_mq_start_stopped_hw_queue+0x38/0x38
[   55.007899]  ? virtqueue_get_buf_ctx+0x295/0x46b
[   55.008476]  process_backlog+0xb3/0x187
[   55.008961]  __napi_poll.constprop.0+0x57/0x1a7
[   55.009540]  net_rx_action+0x1cb/0x380
[   55.010020]  ? __napi_poll.constprop.0+0x1a7/0x1a7
[   55.010610]  ? vring_new_virtqueue+0x17a/0x17a
[   55.011173]  ? note_interrupt+0x2cd/0x367
[   55.011675]  handle_softirqs+0x13c/0x2c9
[   55.012169]  do_softirq+0x5f/0x7d
[   55.012597]  </IRQ>
[   55.012882]  <TASK>
[   55.013179]  __local_bh_enable_ip+0x48/0x62
[   55.013704]  __neigh_event_send+0x3fd/0x4ca
[   55.014227]  neigh_resolve_output+0x1e/0x210
[   55.014761]  ip_finish_output2+0x4bf/0x4f0
[   55.015278]  ? __ip_finish_output+0x171/0x1b8
[   55.015823]  ip_send_skb+0x25/0x57
[   55.016261]  raw_sendmsg+0xf95/0x10c0
[   55.016729]  ? check_new_pages+0x45/0x71
[   55.017229]  ? raw_hash_sk+0x21b/0x21b
[   55.017708]  ? kernel_init_pages+0x42/0x51
[   55.018225]  ? prep_new_page+0x44/0x51
[   55.018704]  ? get_page_from_freelist+0x72b/0x915
[   55.019292]  ? signal_pending_state+0x77/0x77
[   55.019840]  ? preempt_count_sub+0x14/0xb3
[   55.020357]  ? __might_resched+0x8a/0x240
[   55.020860]  ? __might_sleep+0x25/0xa0
[   55.021345]  ? first_zones_zonelist+0x2c/0x43
[   55.021896]  ? __rcu_read_lock+0x2d/0x3a
[   55.022396]  ? __pte_offset_map+0x32/0xa4
[   55.022901]  ? __might_resched+0x8a/0x240
[   55.023404]  ? __might_sleep+0x25/0xa0
[   55.023879]  ? inet_send_prepare+0x54/0x54
[   55.024391]  ? sock_sendmsg_nosec+0x42/0x6c
[   55.024918]  sock_sendmsg_nosec+0x42/0x6c
[   55.025428]  __sys_sendto+0x15d/0x1cc
[   55.025892]  ? __x64_sys_getpeername+0x44/0x44
[   55.026441]  ? __handle_mm_fault+0x679/0xae4
[   55.026988]  ? find_vma+0x6b/0x8b
[   55.027414]  ? find_vma_intersection+0x8a/0x8a
[   55.027966]  ? handle_mm_fault+0x38/0x154
[   55.028470]  ? handle_mm_fault+0xeb/0x154
[   55.028972]  ? preempt_latency_start+0x29/0x34
[   55.029532]  ? preempt_count_sub+0x14/0xb3
[   55.030047]  ? up_read+0x4b/0x5c
[   55.030463]  __x64_sys_sendto+0x76/0x82
[   55.030949]  do_syscall_64+0x69/0xd5
[   55.031406]  entry_SYSCALL_64_after_hwframe+0x4b/0x53
[   55.032028] RIP: 0033:0x7f2d2fec9a73
[   55.032481] Code: 8b 15 a9 83 0c 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 80 3d 71 0b 0d 00 00 41 89 ca 74 14 b8 2c 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 75 c3 0f 1f 40 00 55 48 83 ec 30 44 89 4c 24
[   55.034660] RSP: 002b:00007ffe85756418 EFLAGS: 00000202 ORIG_RAX: 000000000000002c
[   55.035567] RAX: ffffffffffffffda RBX: 0000558bebad1340 RCX: 00007f2d2fec9a73
[   55.036424] RDX: 0000000000000040 RSI: 0000558bebad73c0 RDI: 0000000000000003
[   55.037293] RBP: 0000558bebad73c0 R08: 0000558bebad35c0 R09: 0000000000000010
[   55.038153] R10: 0000000000000000 R11: 0000000000000202 R12: 0000000000000040
[   55.039012] R13: 00007ffe85757b00 R14: 0000001d00000001 R15: 0000558bebad4680
[   55.039871]  </TASK>
[   55.040167] Modules linked in:
[   55.040585] ---[ end trace 0000000000000000 ]---
[   55.041164] RIP: 0010:xfrmi_rcv_cb+0x2d/0x295
[   55.041714] Code: 00 00 41 57 41 56 41 89 f6 41 55 41 54 55 53 48 89 fb 51 85 f6 75 31 48 89 df e8 d7 e8 ff ff 48 89 c5 48 89 c7 e8 8b a4 4f ff <48> 8b 7d 00 48 89 ee e8 eb f3 ff ff 49 89 c5 b8 01 00 00 00 4d 85
[   55.043889] RSP: 0018:ffffc90000007990 EFLAGS: 00010282
[   55.044528] RAX: 0000000000000001 RBX: ffff8881126e9900 RCX: fffffbfff07b77cd
[   55.045386] RDX: fffffbfff07b77cd RSI: fffffbfff07b77cd RDI: ffffffff83dbbe60
[   55.046250] RBP: 6b6b6b6b00000000 R08: 0000000000000008 R09: 0000000000000001
[   55.047104] R10: ffffffff83dbbe67 R11: fffffbfff07b77cc R12: 00000000ffffffff
[   55.047960] R13: 00000000ffffffff R14: 00000000ffffffff R15: 0000000000000002
[   55.048820] FS:  00007f2d2fc0dc40(0000) GS:ffffffff82eb2000(0000) knlGS:0000000000000000
[   55.049805] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   55.050507] CR2: 00007ffe85755ff8 CR3: 0000000109941000 CR4: 0000000000350ef0
[   55.051366] Kernel panic - not syncing: Fatal exception in interrupt
[   55.052136] Kernel Offset: disabled
[   55.052577] ---[ end Kernel panic - not syncing: Fatal exception in interrupt ]---

Fixes: 304b44f0d5a4 ("xfrm: Add dir validation to "in" data path lookup")
Signed-off-by: Antony Antony <antony.antony@secunet.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/xfrm/xfrm_input.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index d2ea18dcb0cb..63c004103912 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -585,8 +585,11 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 		}
 
 		if (unlikely(x->dir && x->dir != XFRM_SA_DIR_IN)) {
+			secpath_reset(skb);
 			XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEDIRERROR);
+			xfrm_audit_state_notfound(skb, family, spi, seq);
 			xfrm_state_put(x);
+			x = NULL;
 			goto drop;
 		}
 
-- 
cgit v1.2.3-58-ga151


From 15f5fe9e84839dcc9eaa69b08ced9d24cb464369 Mon Sep 17 00:00:00 2001
From: Antony Antony <antony.antony@secunet.com>
Date: Tue, 11 Jun 2024 08:32:15 +0200
Subject: xfrm: Log input direction mismatch error in one place

Previously, the offload data path decrypted the packet before checking
the direction, leading to error logging and packet dropping. However,
dropped packets wouldn't be visible in tcpdump or audit log.

With this fix, the offload path, upon noticing SA direction mismatch,
will pass the packet to the stack without decrypting it. The L3 layer
will then log the error, audit, and drop ESP without decrypting or
decapsulating it.

This also ensures that the slow path records the error and audit log,
making dropped packets visible in tcpdump.

Fixes: 304b44f0d5a4 ("xfrm: Add dir validation to "in" data path lookup")
Signed-off-by: Antony Antony <antony.antony@secunet.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/ipv4/esp4_offload.c | 7 +++++++
 net/ipv6/esp6_offload.c | 7 +++++++
 net/xfrm/xfrm_input.c   | 5 -----
 3 files changed, 14 insertions(+), 5 deletions(-)

diff --git a/net/ipv4/esp4_offload.c b/net/ipv4/esp4_offload.c
index b3271957ad9a..3f28ecbdcaef 100644
--- a/net/ipv4/esp4_offload.c
+++ b/net/ipv4/esp4_offload.c
@@ -56,6 +56,13 @@ static struct sk_buff *esp4_gro_receive(struct list_head *head,
 		x = xfrm_state_lookup(dev_net(skb->dev), skb->mark,
 				      (xfrm_address_t *)&ip_hdr(skb)->daddr,
 				      spi, IPPROTO_ESP, AF_INET);
+
+		if (unlikely(x && x->dir && x->dir != XFRM_SA_DIR_IN)) {
+			/* non-offload path will record the error and audit log */
+			xfrm_state_put(x);
+			x = NULL;
+		}
+
 		if (!x)
 			goto out_reset;
 
diff --git a/net/ipv6/esp6_offload.c b/net/ipv6/esp6_offload.c
index 527b7caddbc6..919ebfabbe4e 100644
--- a/net/ipv6/esp6_offload.c
+++ b/net/ipv6/esp6_offload.c
@@ -83,6 +83,13 @@ static struct sk_buff *esp6_gro_receive(struct list_head *head,
 		x = xfrm_state_lookup(dev_net(skb->dev), skb->mark,
 				      (xfrm_address_t *)&ipv6_hdr(skb)->daddr,
 				      spi, IPPROTO_ESP, AF_INET6);
+
+		if (unlikely(x && x->dir && x->dir != XFRM_SA_DIR_IN)) {
+			/* non-offload path will record the error and audit log */
+			xfrm_state_put(x);
+			x = NULL;
+		}
+
 		if (!x)
 			goto out_reset;
 
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index 63c004103912..e95462b982b0 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -474,11 +474,6 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 	if (encap_type < 0 || (xo && xo->flags & XFRM_GRO)) {
 		x = xfrm_input_state(skb);
 
-		if (unlikely(x->dir && x->dir != XFRM_SA_DIR_IN)) {
-			XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEDIRERROR);
-			goto drop;
-		}
-
 		if (unlikely(x->km.state != XFRM_STATE_VALID)) {
 			if (x->km.state == XFRM_STATE_ACQ)
 				XFRM_INC_STATS(net, LINUX_MIB_XFRMACQUIREERROR);
-- 
cgit v1.2.3-58-ga151


From f22b4b55edb507a2b30981e133b66b642be4d13f Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 13 Jun 2024 14:33:16 -0700
Subject: net: make for_each_netdev_dump() a little more bug-proof

I find the behavior of xa_for_each_start() slightly counter-intuitive.
It doesn't end the iteration by making the index point after the last
element. IOW calling xa_for_each_start() again after it "finished"
will run the body of the loop for the last valid element, instead
of doing nothing.

This works fine for netlink dumps if they terminate correctly
(i.e. coalesce or carefully handle NLM_DONE), but as we keep getting
reminded legacy dumps are unlikely to go away.

Fixing this generically at the xa_for_each_start() level seems hard -
there is no index reserved for "end of iteration".
ifindexes are 31b wide, tho, and iterator is ulong so for
for_each_netdev_dump() it's safe to go to the next element.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f148a01dd1d1..85111502cf8f 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3021,7 +3021,8 @@ int call_netdevice_notifiers_info(unsigned long val,
 #define net_device_entry(lh)	list_entry(lh, struct net_device, dev_list)
 
 #define for_each_netdev_dump(net, d, ifindex)				\
-	xa_for_each_start(&(net)->dev_by_index, (ifindex), (d), (ifindex))
+	for (; (d = xa_find(&(net)->dev_by_index, &ifindex,		\
+			    ULONG_MAX, XA_PRESENT)); ifindex++)
 
 static inline struct net_device *next_net_device(struct net_device *dev)
 {
-- 
cgit v1.2.3-58-ga151


From d0e35656d83458d668593930f1568d464dde429c Mon Sep 17 00:00:00 2001
From: "D. Wythe" <alibuda@linux.alibaba.com>
Date: Fri, 14 Jun 2024 02:00:28 +0800
Subject: net/smc: refactoring initialization of smc sock

This patch aims to isolate the shared components of SMC socket
allocation by introducing smc_sk_init() for sock initialization
and __smc_create_clcsk() for the initialization of clcsock.

This is in preparation for the subsequent implementation of the
AF_INET version of SMC.

Signed-off-by: D. Wythe <alibuda@linux.alibaba.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Reviewed-by: Dust Li <dust.li@linux.alibaba.com>
Tested-by: Niklas Schnelle <schnelle@linux.ibm.com>
Tested-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/smc/af_smc.c | 86 +++++++++++++++++++++++++++++++-------------------------
 net/smc/smc.h    |  5 ++++
 2 files changed, 53 insertions(+), 38 deletions(-)

diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index c5f98c6b2561..0508ea20b3c3 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -361,25 +361,15 @@ static void smc_destruct(struct sock *sk)
 		return;
 }
 
-static struct sock *smc_sock_alloc(struct net *net, struct socket *sock,
-				   int protocol)
+void smc_sk_init(struct net *net, struct sock *sk, int protocol)
 {
-	struct smc_sock *smc;
-	struct proto *prot;
-	struct sock *sk;
-
-	prot = (protocol == SMCPROTO_SMC6) ? &smc_proto6 : &smc_proto;
-	sk = sk_alloc(net, PF_SMC, GFP_KERNEL, prot, 0);
-	if (!sk)
-		return NULL;
+	struct smc_sock *smc = smc_sk(sk);
 
-	sock_init_data(sock, sk); /* sets sk_refcnt to 1 */
 	sk->sk_state = SMC_INIT;
 	sk->sk_destruct = smc_destruct;
 	sk->sk_protocol = protocol;
 	WRITE_ONCE(sk->sk_sndbuf, 2 * READ_ONCE(net->smc.sysctl_wmem));
 	WRITE_ONCE(sk->sk_rcvbuf, 2 * READ_ONCE(net->smc.sysctl_rmem));
-	smc = smc_sk(sk);
 	INIT_WORK(&smc->tcp_listen_work, smc_tcp_listen_work);
 	INIT_WORK(&smc->connect_work, smc_connect_work);
 	INIT_DELAYED_WORK(&smc->conn.tx_work, smc_tx_work);
@@ -389,6 +379,24 @@ static struct sock *smc_sock_alloc(struct net *net, struct socket *sock,
 	sk->sk_prot->hash(sk);
 	mutex_init(&smc->clcsock_release_lock);
 	smc_init_saved_callbacks(smc);
+	smc->limit_smc_hs = net->smc.limit_smc_hs;
+	smc->use_fallback = false; /* assume rdma capability first */
+	smc->fallback_rsn = 0;
+}
+
+static struct sock *smc_sock_alloc(struct net *net, struct socket *sock,
+				   int protocol)
+{
+	struct proto *prot;
+	struct sock *sk;
+
+	prot = (protocol == SMCPROTO_SMC6) ? &smc_proto6 : &smc_proto;
+	sk = sk_alloc(net, PF_SMC, GFP_KERNEL, prot, 0);
+	if (!sk)
+		return NULL;
+
+	sock_init_data(sock, sk); /* sets sk_refcnt to 1 */
+	smc_sk_init(net, sk, protocol);
 
 	return sk;
 }
@@ -3303,6 +3311,31 @@ static const struct proto_ops smc_sock_ops = {
 	.splice_read	= smc_splice_read,
 };
 
+int smc_create_clcsk(struct net *net, struct sock *sk, int family)
+{
+	struct smc_sock *smc = smc_sk(sk);
+	int rc;
+
+	rc = sock_create_kern(net, family, SOCK_STREAM, IPPROTO_TCP,
+			      &smc->clcsock);
+	if (rc) {
+		sk_common_release(sk);
+		return rc;
+	}
+
+	/* smc_clcsock_release() does not wait smc->clcsock->sk's
+	 * destruction;  its sk_state might not be TCP_CLOSE after
+	 * smc->sk is close()d, and TCP timers can be fired later,
+	 * which need net ref.
+	 */
+	sk = smc->clcsock->sk;
+	__netns_tracker_free(net, &sk->ns_tracker, false);
+	sk->sk_net_refcnt = 1;
+	get_net_track(net, &sk->ns_tracker, GFP_KERNEL);
+	sock_inuse_add(net, 1);
+	return 0;
+}
+
 static int __smc_create(struct net *net, struct socket *sock, int protocol,
 			int kern, struct socket *clcsock)
 {
@@ -3328,35 +3361,12 @@ static int __smc_create(struct net *net, struct socket *sock, int protocol,
 
 	/* create internal TCP socket for CLC handshake and fallback */
 	smc = smc_sk(sk);
-	smc->use_fallback = false; /* assume rdma capability first */
-	smc->fallback_rsn = 0;
-
-	/* default behavior from limit_smc_hs in every net namespace */
-	smc->limit_smc_hs = net->smc.limit_smc_hs;
 
 	rc = 0;
-	if (!clcsock) {
-		rc = sock_create_kern(net, family, SOCK_STREAM, IPPROTO_TCP,
-				      &smc->clcsock);
-		if (rc) {
-			sk_common_release(sk);
-			goto out;
-		}
-
-		/* smc_clcsock_release() does not wait smc->clcsock->sk's
-		 * destruction;  its sk_state might not be TCP_CLOSE after
-		 * smc->sk is close()d, and TCP timers can be fired later,
-		 * which need net ref.
-		 */
-		sk = smc->clcsock->sk;
-		__netns_tracker_free(net, &sk->ns_tracker, false);
-		sk->sk_net_refcnt = 1;
-		get_net_track(net, &sk->ns_tracker, GFP_KERNEL);
-		sock_inuse_add(net, 1);
-	} else {
+	if (clcsock)
 		smc->clcsock = clcsock;
-	}
-
+	else
+		rc = smc_create_clcsk(net, sk, family);
 out:
 	return rc;
 }
diff --git a/net/smc/smc.h b/net/smc/smc.h
index 18c8b7870198..3edec1e133d8 100644
--- a/net/smc/smc.h
+++ b/net/smc/smc.h
@@ -34,6 +34,11 @@
 extern struct proto smc_proto;
 extern struct proto smc_proto6;
 
+/* smc sock initialization */
+void smc_sk_init(struct net *net, struct sock *sk, int protocol);
+/* clcsock initialization */
+int smc_create_clcsk(struct net *net, struct sock *sk, int family);
+
 #ifdef ATOMIC64_INIT
 #define KERNEL_HAS_ATOMIC64
 #endif
-- 
cgit v1.2.3-58-ga151


From 13543d02c90d6195b31bef8fb51dfeff77c0b368 Mon Sep 17 00:00:00 2001
From: "D. Wythe" <alibuda@linux.alibaba.com>
Date: Fri, 14 Jun 2024 02:00:29 +0800
Subject: net/smc: expose smc proto operations

Externalize smc proto operations (smc_xxx) to allow
access from files other than af_smc.c

This is in preparation for the subsequent implementation
of the AF_INET version of SMC.

Signed-off-by: D. Wythe <alibuda@linux.alibaba.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Reviewed-by: Dust Li <dust.li@linux.alibaba.com>
Tested-by: Niklas Schnelle <schnelle@linux.ibm.com>
Tested-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/smc/af_smc.c | 60 ++++++++++++++++++++++++++++----------------------------
 net/smc/smc.h    | 33 +++++++++++++++++++++++++++++++
 2 files changed, 63 insertions(+), 30 deletions(-)

diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index 0508ea20b3c3..7ee6f37813d6 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -170,15 +170,15 @@ static bool smc_hs_congested(const struct sock *sk)
 	return false;
 }
 
-static struct smc_hashinfo smc_v4_hashinfo = {
+struct smc_hashinfo smc_v4_hashinfo = {
 	.lock = __RW_LOCK_UNLOCKED(smc_v4_hashinfo.lock),
 };
 
-static struct smc_hashinfo smc_v6_hashinfo = {
+struct smc_hashinfo smc_v6_hashinfo = {
 	.lock = __RW_LOCK_UNLOCKED(smc_v6_hashinfo.lock),
 };
 
-static int smc_hash_sk(struct sock *sk)
+int smc_hash_sk(struct sock *sk)
 {
 	struct smc_hashinfo *h = sk->sk_prot->h.smc_hash;
 	struct hlist_head *head;
@@ -193,7 +193,7 @@ static int smc_hash_sk(struct sock *sk)
 	return 0;
 }
 
-static void smc_unhash_sk(struct sock *sk)
+void smc_unhash_sk(struct sock *sk)
 {
 	struct smc_hashinfo *h = sk->sk_prot->h.smc_hash;
 
@@ -207,7 +207,7 @@ static void smc_unhash_sk(struct sock *sk)
  * work which we didn't do because of user hold the sock_lock in the
  * BH context
  */
-static void smc_release_cb(struct sock *sk)
+void smc_release_cb(struct sock *sk)
 {
 	struct smc_sock *smc = smc_sk(sk);
 
@@ -307,7 +307,7 @@ static int __smc_release(struct smc_sock *smc)
 	return rc;
 }
 
-static int smc_release(struct socket *sock)
+int smc_release(struct socket *sock)
 {
 	struct sock *sk = sock->sk;
 	struct smc_sock *smc;
@@ -401,8 +401,8 @@ static struct sock *smc_sock_alloc(struct net *net, struct socket *sock,
 	return sk;
 }
 
-static int smc_bind(struct socket *sock, struct sockaddr *uaddr,
-		    int addr_len)
+int smc_bind(struct socket *sock, struct sockaddr *uaddr,
+	     int addr_len)
 {
 	struct sockaddr_in *addr = (struct sockaddr_in *)uaddr;
 	struct sock *sk = sock->sk;
@@ -1631,8 +1631,8 @@ out:
 	release_sock(&smc->sk);
 }
 
-static int smc_connect(struct socket *sock, struct sockaddr *addr,
-		       int alen, int flags)
+int smc_connect(struct socket *sock, struct sockaddr *addr,
+		int alen, int flags)
 {
 	struct sock *sk = sock->sk;
 	struct smc_sock *smc;
@@ -2613,7 +2613,7 @@ out:
 	read_unlock_bh(&listen_clcsock->sk_callback_lock);
 }
 
-static int smc_listen(struct socket *sock, int backlog)
+int smc_listen(struct socket *sock, int backlog)
 {
 	struct sock *sk = sock->sk;
 	struct smc_sock *smc;
@@ -2678,8 +2678,8 @@ out:
 	return rc;
 }
 
-static int smc_accept(struct socket *sock, struct socket *new_sock,
-		      struct proto_accept_arg *arg)
+int smc_accept(struct socket *sock, struct socket *new_sock,
+	       struct proto_accept_arg *arg)
 {
 	struct sock *sk = sock->sk, *nsk;
 	DECLARE_WAITQUEUE(wait, current);
@@ -2748,8 +2748,8 @@ out:
 	return rc;
 }
 
-static int smc_getname(struct socket *sock, struct sockaddr *addr,
-		       int peer)
+int smc_getname(struct socket *sock, struct sockaddr *addr,
+		int peer)
 {
 	struct smc_sock *smc;
 
@@ -2762,7 +2762,7 @@ static int smc_getname(struct socket *sock, struct sockaddr *addr,
 	return smc->clcsock->ops->getname(smc->clcsock, addr, peer);
 }
 
-static int smc_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
+int smc_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 {
 	struct sock *sk = sock->sk;
 	struct smc_sock *smc;
@@ -2800,8 +2800,8 @@ out:
 	return rc;
 }
 
-static int smc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
-		       int flags)
+int smc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
+		int flags)
 {
 	struct sock *sk = sock->sk;
 	struct smc_sock *smc;
@@ -2850,8 +2850,8 @@ static __poll_t smc_accept_poll(struct sock *parent)
 	return mask;
 }
 
-static __poll_t smc_poll(struct file *file, struct socket *sock,
-			     poll_table *wait)
+__poll_t smc_poll(struct file *file, struct socket *sock,
+		  poll_table *wait)
 {
 	struct sock *sk = sock->sk;
 	struct smc_sock *smc;
@@ -2903,7 +2903,7 @@ static __poll_t smc_poll(struct file *file, struct socket *sock,
 	return mask;
 }
 
-static int smc_shutdown(struct socket *sock, int how)
+int smc_shutdown(struct socket *sock, int how)
 {
 	struct sock *sk = sock->sk;
 	bool do_shutdown = true;
@@ -3043,8 +3043,8 @@ static int __smc_setsockopt(struct socket *sock, int level, int optname,
 	return rc;
 }
 
-static int smc_setsockopt(struct socket *sock, int level, int optname,
-			  sockptr_t optval, unsigned int optlen)
+int smc_setsockopt(struct socket *sock, int level, int optname,
+		   sockptr_t optval, unsigned int optlen)
 {
 	struct sock *sk = sock->sk;
 	struct smc_sock *smc;
@@ -3130,8 +3130,8 @@ out:
 	return rc;
 }
 
-static int smc_getsockopt(struct socket *sock, int level, int optname,
-			  char __user *optval, int __user *optlen)
+int smc_getsockopt(struct socket *sock, int level, int optname,
+		   char __user *optval, int __user *optlen)
 {
 	struct smc_sock *smc;
 	int rc;
@@ -3156,8 +3156,8 @@ static int smc_getsockopt(struct socket *sock, int level, int optname,
 	return rc;
 }
 
-static int smc_ioctl(struct socket *sock, unsigned int cmd,
-		     unsigned long arg)
+int smc_ioctl(struct socket *sock, unsigned int cmd,
+	      unsigned long arg)
 {
 	union smc_host_cursor cons, urg;
 	struct smc_connection *conn;
@@ -3243,9 +3243,9 @@ static int smc_ioctl(struct socket *sock, unsigned int cmd,
  * Note that subsequent recv() calls have to wait till all splice() processing
  * completed.
  */
-static ssize_t smc_splice_read(struct socket *sock, loff_t *ppos,
-			       struct pipe_inode_info *pipe, size_t len,
-			       unsigned int flags)
+ssize_t smc_splice_read(struct socket *sock, loff_t *ppos,
+			struct pipe_inode_info *pipe, size_t len,
+			unsigned int flags)
 {
 	struct sock *sk = sock->sk;
 	struct smc_sock *smc;
diff --git a/net/smc/smc.h b/net/smc/smc.h
index 3edec1e133d8..34b781e463c4 100644
--- a/net/smc/smc.h
+++ b/net/smc/smc.h
@@ -34,6 +34,39 @@
 extern struct proto smc_proto;
 extern struct proto smc_proto6;
 
+extern struct smc_hashinfo smc_v4_hashinfo;
+extern struct smc_hashinfo smc_v6_hashinfo;
+
+int smc_hash_sk(struct sock *sk);
+void smc_unhash_sk(struct sock *sk);
+void smc_release_cb(struct sock *sk);
+
+int smc_release(struct socket *sock);
+int smc_bind(struct socket *sock, struct sockaddr *uaddr,
+	     int addr_len);
+int smc_connect(struct socket *sock, struct sockaddr *addr,
+		int alen, int flags);
+int smc_accept(struct socket *sock, struct socket *new_sock,
+	       struct proto_accept_arg *arg);
+int smc_getname(struct socket *sock, struct sockaddr *addr,
+		int peer);
+__poll_t smc_poll(struct file *file, struct socket *sock,
+		  poll_table *wait);
+int smc_ioctl(struct socket *sock, unsigned int cmd,
+	      unsigned long arg);
+int smc_listen(struct socket *sock, int backlog);
+int smc_shutdown(struct socket *sock, int how);
+int smc_setsockopt(struct socket *sock, int level, int optname,
+		   sockptr_t optval, unsigned int optlen);
+int smc_getsockopt(struct socket *sock, int level, int optname,
+		   char __user *optval, int __user *optlen);
+int smc_sendmsg(struct socket *sock, struct msghdr *msg, size_t len);
+int smc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
+		int flags);
+ssize_t smc_splice_read(struct socket *sock, loff_t *ppos,
+			struct pipe_inode_info *pipe, size_t len,
+			unsigned int flags);
+
 /* smc sock initialization */
 void smc_sk_init(struct net *net, struct sock *sk, int protocol);
 /* clcsock initialization */
-- 
cgit v1.2.3-58-ga151


From d25a92ccae6bed02327b63d138e12e7806830f78 Mon Sep 17 00:00:00 2001
From: "D. Wythe" <alibuda@linux.alibaba.com>
Date: Fri, 14 Jun 2024 02:00:30 +0800
Subject: net/smc: Introduce IPPROTO_SMC

This patch allows to create smc socket via AF_INET,
similar to the following code,

/* create v4 smc sock */
v4 = socket(AF_INET, SOCK_STREAM, IPPROTO_SMC);

/* create v6 smc sock */
v6 = socket(AF_INET6, SOCK_STREAM, IPPROTO_SMC);

There are several reasons why we believe it is appropriate here:

1. For smc sockets, it actually use IPv4 (AF-INET) or IPv6 (AF-INET6)
address. There is no AF_SMC address at all.

2. Create smc socket in the AF_INET(6) path, which allows us to reuse
the infrastructure of AF_INET(6) path, such as common ebpf hooks.
Otherwise, smc have to implement it again in AF_SMC path.

Signed-off-by: D. Wythe <alibuda@linux.alibaba.com>
Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com>
Reviewed-by: Dust Li <dust.li@linux.alibaba.com>
Tested-by: Niklas Schnelle <schnelle@linux.ibm.com>
Tested-by: Wenjia Zhang <wenjia@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/in.h |   2 +
 net/smc/Makefile        |   2 +-
 net/smc/af_smc.c        |  16 ++++-
 net/smc/smc_inet.c      | 159 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/smc/smc_inet.h      |  22 +++++++
 5 files changed, 198 insertions(+), 3 deletions(-)
 create mode 100644 net/smc/smc_inet.c
 create mode 100644 net/smc/smc_inet.h

diff --git a/include/uapi/linux/in.h b/include/uapi/linux/in.h
index e682ab628dfa..d358add1611c 100644
--- a/include/uapi/linux/in.h
+++ b/include/uapi/linux/in.h
@@ -81,6 +81,8 @@ enum {
 #define IPPROTO_ETHERNET	IPPROTO_ETHERNET
   IPPROTO_RAW = 255,		/* Raw IP packets			*/
 #define IPPROTO_RAW		IPPROTO_RAW
+  IPPROTO_SMC = 256,		/* Shared Memory Communications		*/
+#define IPPROTO_SMC		IPPROTO_SMC
   IPPROTO_MPTCP = 262,		/* Multipath TCP connection		*/
 #define IPPROTO_MPTCP		IPPROTO_MPTCP
   IPPROTO_MAX
diff --git a/net/smc/Makefile b/net/smc/Makefile
index 2c510d543058..60f1c87d5212 100644
--- a/net/smc/Makefile
+++ b/net/smc/Makefile
@@ -4,6 +4,6 @@ obj-$(CONFIG_SMC)	+= smc.o
 obj-$(CONFIG_SMC_DIAG)	+= smc_diag.o
 smc-y := af_smc.o smc_pnet.o smc_ib.o smc_clc.o smc_core.o smc_wr.o smc_llc.o
 smc-y += smc_cdc.o smc_tx.o smc_rx.o smc_close.o smc_ism.o smc_netlink.o smc_stats.o
-smc-y += smc_tracepoint.o
+smc-y += smc_tracepoint.o smc_inet.o
 smc-$(CONFIG_SYSCTL) += smc_sysctl.o
 smc-$(CONFIG_SMC_LO) += smc_loopback.o
diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index 7ee6f37813d6..73a875573e7a 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -54,6 +54,7 @@
 #include "smc_tracepoint.h"
 #include "smc_sysctl.h"
 #include "smc_loopback.h"
+#include "smc_inet.h"
 
 static DEFINE_MUTEX(smc_server_lgr_pending);	/* serialize link group
 						 * creation on server
@@ -3575,10 +3576,15 @@ static int __init smc_init(void)
 		pr_err("%s: tcp_ulp_register fails with %d\n", __func__, rc);
 		goto out_lo;
 	}
-
+	rc = smc_inet_init();
+	if (rc) {
+		pr_err("%s: smc_inet_init fails with %d\n", __func__, rc);
+		goto out_ulp;
+	}
 	static_branch_enable(&tcp_have_smc);
 	return 0;
-
+out_ulp:
+	tcp_unregister_ulp(&smc_ulp_ops);
 out_lo:
 	smc_loopback_exit();
 out_ib:
@@ -3615,6 +3621,7 @@ out_pernet_subsys:
 static void __exit smc_exit(void)
 {
 	static_branch_disable(&tcp_have_smc);
+	smc_inet_exit();
 	tcp_unregister_ulp(&smc_ulp_ops);
 	sock_unregister(PF_SMC);
 	smc_core_exit();
@@ -3642,4 +3649,9 @@ MODULE_DESCRIPTION("smc socket address family");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS_NETPROTO(PF_SMC);
 MODULE_ALIAS_TCP_ULP("smc");
+/* 256 for IPPROTO_SMC and 1 for SOCK_STREAM */
+MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_INET, 256, 1);
+#if IS_ENABLED(CONFIG_IPV6)
+MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_INET6, 256, 1);
+#endif /* CONFIG_IPV6 */
 MODULE_ALIAS_GENL_FAMILY(SMC_GENL_FAMILY_NAME);
diff --git a/net/smc/smc_inet.c b/net/smc/smc_inet.c
new file mode 100644
index 000000000000..bece346dd8e9
--- /dev/null
+++ b/net/smc/smc_inet.c
@@ -0,0 +1,159 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Definitions for the IPPROTO_SMC (socket related)
+ *
+ *  Copyright IBM Corp. 2016, 2018
+ *  Copyright (c) 2024, Alibaba Inc.
+ *
+ *  Author: D. Wythe <alibuda@linux.alibaba.com>
+ */
+
+#include <net/protocol.h>
+#include <net/sock.h>
+
+#include "smc_inet.h"
+#include "smc.h"
+
+static int smc_inet_init_sock(struct sock *sk);
+
+static struct proto smc_inet_prot = {
+	.name		= "INET_SMC",
+	.owner		= THIS_MODULE,
+	.init		= smc_inet_init_sock,
+	.hash		= smc_hash_sk,
+	.unhash		= smc_unhash_sk,
+	.release_cb	= smc_release_cb,
+	.obj_size	= sizeof(struct smc_sock),
+	.h.smc_hash	= &smc_v4_hashinfo,
+	.slab_flags	= SLAB_TYPESAFE_BY_RCU,
+};
+
+static const struct proto_ops smc_inet_stream_ops = {
+	.family		= PF_INET,
+	.owner		= THIS_MODULE,
+	.release	= smc_release,
+	.bind		= smc_bind,
+	.connect	= smc_connect,
+	.socketpair	= sock_no_socketpair,
+	.accept		= smc_accept,
+	.getname	= smc_getname,
+	.poll		= smc_poll,
+	.ioctl		= smc_ioctl,
+	.listen		= smc_listen,
+	.shutdown	= smc_shutdown,
+	.setsockopt	= smc_setsockopt,
+	.getsockopt	= smc_getsockopt,
+	.sendmsg	= smc_sendmsg,
+	.recvmsg	= smc_recvmsg,
+	.mmap		= sock_no_mmap,
+	.splice_read	= smc_splice_read,
+};
+
+static struct inet_protosw smc_inet_protosw = {
+	.type		= SOCK_STREAM,
+	.protocol	= IPPROTO_SMC,
+	.prot		= &smc_inet_prot,
+	.ops		= &smc_inet_stream_ops,
+	.flags		= INET_PROTOSW_ICSK,
+};
+
+#if IS_ENABLED(CONFIG_IPV6)
+static struct proto smc_inet6_prot = {
+	.name		= "INET6_SMC",
+	.owner		= THIS_MODULE,
+	.init		= smc_inet_init_sock,
+	.hash		= smc_hash_sk,
+	.unhash		= smc_unhash_sk,
+	.release_cb	= smc_release_cb,
+	.obj_size	= sizeof(struct smc_sock),
+	.h.smc_hash	= &smc_v6_hashinfo,
+	.slab_flags	= SLAB_TYPESAFE_BY_RCU,
+};
+
+static const struct proto_ops smc_inet6_stream_ops = {
+	.family		= PF_INET6,
+	.owner		= THIS_MODULE,
+	.release	= smc_release,
+	.bind		= smc_bind,
+	.connect	= smc_connect,
+	.socketpair	= sock_no_socketpair,
+	.accept		= smc_accept,
+	.getname	= smc_getname,
+	.poll		= smc_poll,
+	.ioctl		= smc_ioctl,
+	.listen		= smc_listen,
+	.shutdown	= smc_shutdown,
+	.setsockopt	= smc_setsockopt,
+	.getsockopt	= smc_getsockopt,
+	.sendmsg	= smc_sendmsg,
+	.recvmsg	= smc_recvmsg,
+	.mmap		= sock_no_mmap,
+	.splice_read	= smc_splice_read,
+};
+
+static struct inet_protosw smc_inet6_protosw = {
+	.type		= SOCK_STREAM,
+	.protocol	= IPPROTO_SMC,
+	.prot		= &smc_inet6_prot,
+	.ops		= &smc_inet6_stream_ops,
+	.flags		= INET_PROTOSW_ICSK,
+};
+#endif /* CONFIG_IPV6 */
+
+static int smc_inet_init_sock(struct sock *sk)
+{
+	struct net *net = sock_net(sk);
+
+	/* init common smc sock */
+	smc_sk_init(net, sk, IPPROTO_SMC);
+	/* create clcsock */
+	return smc_create_clcsk(net, sk, sk->sk_family);
+}
+
+int __init smc_inet_init(void)
+{
+	int rc;
+
+	rc = proto_register(&smc_inet_prot, 1);
+	if (rc) {
+		pr_err("%s: proto_register smc_inet_prot fails with %d\n",
+		       __func__, rc);
+		return rc;
+	}
+	/* no return value */
+	inet_register_protosw(&smc_inet_protosw);
+
+#if IS_ENABLED(CONFIG_IPV6)
+	rc = proto_register(&smc_inet6_prot, 1);
+	if (rc) {
+		pr_err("%s: proto_register smc_inet6_prot fails with %d\n",
+		       __func__, rc);
+		goto out_inet6_prot;
+	}
+	rc = inet6_register_protosw(&smc_inet6_protosw);
+	if (rc) {
+		pr_err("%s: inet6_register_protosw smc_inet6_protosw fails with %d\n",
+		       __func__, rc);
+		goto out_inet6_protosw;
+	}
+	return rc;
+out_inet6_protosw:
+	proto_unregister(&smc_inet6_prot);
+out_inet6_prot:
+	inet_unregister_protosw(&smc_inet_protosw);
+	proto_unregister(&smc_inet_prot);
+#endif /* CONFIG_IPV6 */
+	return rc;
+}
+
+void smc_inet_exit(void)
+{
+#if IS_ENABLED(CONFIG_IPV6)
+	inet6_unregister_protosw(&smc_inet6_protosw);
+	proto_unregister(&smc_inet6_prot);
+#endif /* CONFIG_IPV6 */
+	inet_unregister_protosw(&smc_inet_protosw);
+	proto_unregister(&smc_inet_prot);
+}
diff --git a/net/smc/smc_inet.h b/net/smc/smc_inet.h
new file mode 100644
index 000000000000..a489c8a2b8ef
--- /dev/null
+++ b/net/smc/smc_inet.h
@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ *  Shared Memory Communications over RDMA (SMC-R) and RoCE
+ *
+ *  Definitions for the IPPROTO_SMC (socket related)
+
+ *  Copyright IBM Corp. 2016
+ *  Copyright (c) 2024, Alibaba Inc.
+ *
+ *  Author: D. Wythe <alibuda@linux.alibaba.com>
+ */
+#ifndef __INET_SMC
+#define __INET_SMC
+
+/* Initialize protocol registration on IPPROTO_SMC,
+ * @return 0 on success
+ */
+int smc_inet_init(void);
+
+void smc_inet_exit(void);
+
+#endif /* __INET_SMC */
-- 
cgit v1.2.3-58-ga151


From 670d4949bc8e7178420761130664f563453085bd Mon Sep 17 00:00:00 2001
From: Aaradhana Sahu <quic_aarasahu@quicinc.com>
Date: Mon, 17 Jun 2024 14:12:47 +0300
Subject: wifi: ath12k: Fix WARN_ON during firmware crash in split-phy

Whenever firmware is crashed in split-phy below WARN_ON() triggered:

WARNING: CPU: 3 PID: 82 at net/mac80211/driver-ops.c:41 drv_stop+0xac/0xbc
Modules linked in: ath12k qmi_helpers
CPU: 3 PID: 82 Comm: kworker/3:2 Tainted: G      D W          6.9.0-next-20240520-00113-gd981a3784e15 #39
Hardware name: Qualcomm Technologies, Inc. IPQ9574/AP-AL02-C9 (DT)
Workqueue: events_freezable ieee80211_restart_work
pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
pc : drv_stop+0xac/0xbc
lr : ieee80211_stop_device+0x54/0x64
sp : ffff8000848dbb20
x29: ffff8000848dbb20 x28: 0000000000000790 x27: ffff000014d78900
x26: ffff000014d791f8 x25: ffff000007f0d9b0 x24: 0000000000000018
x23: 0000000000000001 x22: 0000000000000000 x21: ffff000014d78e10
x20: ffff800081dc0000 x19: ffff000014d78900 x18: ffffffffffffffff
x17: ffff7fffbca84000 x16: ffff800083fe0000 x15: ffff800081dc0b48
x14: 0000000000000076 x13: 0000000000000076 x12: 0000000000000001
x11: 0000000000000000 x10: 0000000000000a60 x9 : ffff8000848db980
x8 : ffff000000dddfc0 x7 : 0000000000000400 x6 : ffff800083b012d8
x5 : ffff800083b012d8 x4 : 0000000000000000 x3 : ffff000014d78398
x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff000014d78900
Call trace:
 drv_stop+0xac/0xbc
 ieee80211_stop_device+0x54/0x64
 ieee80211_do_stop+0x5a0/0x790
 ieee80211_stop+0x4c/0x178
 __dev_close_many+0xb0/0x150
 dev_close_many+0x88/0x130
 dev_close.part.171+0x44/0x74
 dev_close+0x1c/0x28
 cfg80211_shutdown_all_interfaces+0x44/0xfc
 ieee80211_restart_work+0xfc/0x14c
 process_scheduled_works+0x18c/0x2dc
 worker_thread+0x13c/0x314
 kthread+0x118/0x124
 ret_from_fork+0x10/0x20
---[ end trace 0000000000000000 ]---

The warning in question is from drv_stop():

	if (WARN_ON(!local->started))
		return;

The sequence of WARN_ON() is:
Thread 1:
-Firmware crash calls ath12k_core_reset().
-Call ieee80211_restart_hw() inside
 ath12k_core_post_reconfigure_recovery() which schedules worker
 for both hardware.
-Wait for completion of ab->recovery_start.

Thread 2 (worker thread):
-One hardware acquires rtnl_lock() inside ieee80211_restart_hw() and
 calls ath12k_mac_wait_reconfigure() into ath12k_mac_op_start().
-Hardware is waiting for ab->reconfigure_complete but at this time
 recovery_start_count value is 1 because another worker thread
 (local->restart_work) is still waiting for rtnl_lock().
 recovery_start_count is not equal to number of radios
 (2 in split-phy). So ab->recovery_start complete does not set
 due to this, thread 1 is still waiting and not able to perform
 hif power down up and firmware reload.
-Wait timeout happens for ab->reconfigure_complete and comeback
 to caller (ath12k_mac_op_start()) and sends WMI command to
 crashed firmware and gets error.
-This returns error to drv_start() and local->started is set to false.
-Hardware calls cfg80211_shutdown_all_interfaces() after receiving error
 inside ieee80211_restart_work() and goes to drv_stop(), here we trigger
 WARN_ON as local->started is false.

To fix this issue call ieee80211_restart_hw() after firmware has been
reloaded. Now, each hardware can send WMI command to firmware
successfully. With this fix we don't need to wait for
ab->recovery_start completion so remove
ath12k_mac_wait_reconfigure().

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00209-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 HW2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Aaradhana Sahu <quic_aarasahu@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240529034405.2863150-1-quic_aarasahu@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.c | 21 ++++++++-------------
 drivers/net/wireless/ath/ath12k/core.h |  3 ---
 drivers/net/wireless/ath/ath12k/mac.c  | 23 -----------------------
 3 files changed, 8 insertions(+), 39 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c
index da02d3528737..197099f15e3f 100644
--- a/drivers/net/wireless/ath/ath12k/core.c
+++ b/drivers/net/wireless/ath/ath12k/core.c
@@ -1061,8 +1061,6 @@ static void ath12k_core_post_reconfigure_recovery(struct ath12k_base *ab)
 				mutex_unlock(&ar->conf_mutex);
 			}
 
-			/* Restart after all the link/radio halt */
-			ieee80211_restart_hw(ah->hw);
 			break;
 		case ATH12K_HW_STATE_OFF:
 			ath12k_warn(ab,
@@ -1089,7 +1087,8 @@ static void ath12k_core_post_reconfigure_recovery(struct ath12k_base *ab)
 static void ath12k_core_restart(struct work_struct *work)
 {
 	struct ath12k_base *ab = container_of(work, struct ath12k_base, restart_work);
-	int ret;
+	struct ath12k_hw *ah;
+	int ret, i;
 
 	ret = ath12k_core_reconfigure_on_crash(ab);
 	if (ret) {
@@ -1097,8 +1096,12 @@ static void ath12k_core_restart(struct work_struct *work)
 		return;
 	}
 
-	if (ab->is_reset)
-		complete_all(&ab->reconfigure_complete);
+	if (ab->is_reset) {
+		for (i = 0; i < ab->num_hw; i++) {
+			ah = ab->ah[i];
+			ieee80211_restart_hw(ah->hw);
+		}
+	}
 
 	complete(&ab->restart_completed);
 }
@@ -1152,20 +1155,14 @@ static void ath12k_core_reset(struct work_struct *work)
 	ath12k_dbg(ab, ATH12K_DBG_BOOT, "reset starting\n");
 
 	ab->is_reset = true;
-	atomic_set(&ab->recovery_start_count, 0);
-	reinit_completion(&ab->recovery_start);
 	atomic_set(&ab->recovery_count, 0);
 
 	ath12k_core_pre_reconfigure_recovery(ab);
 
-	reinit_completion(&ab->reconfigure_complete);
 	ath12k_core_post_reconfigure_recovery(ab);
 
 	ath12k_dbg(ab, ATH12K_DBG_BOOT, "waiting recovery start...\n");
 
-	time_left = wait_for_completion_timeout(&ab->recovery_start,
-						ATH12K_RECOVER_START_TIMEOUT_HZ);
-
 	ath12k_hif_irq_disable(ab);
 	ath12k_hif_ce_irq_disable(ab);
 
@@ -1277,8 +1274,6 @@ struct ath12k_base *ath12k_core_alloc(struct device *dev, size_t priv_size,
 	mutex_init(&ab->core_lock);
 	spin_lock_init(&ab->base_lock);
 	init_completion(&ab->reset_complete);
-	init_completion(&ab->reconfigure_complete);
-	init_completion(&ab->recovery_start);
 
 	INIT_LIST_HEAD(&ab->peers);
 	init_waitqueue_head(&ab->peer_mapping_wq);
diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index b483899ca136..82ec23a7329d 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -849,11 +849,8 @@ struct ath12k_base {
 	struct work_struct reset_work;
 	atomic_t reset_count;
 	atomic_t recovery_count;
-	atomic_t recovery_start_count;
 	bool is_reset;
 	struct completion reset_complete;
-	struct completion reconfigure_complete;
-	struct completion recovery_start;
 	/* continuous recovery fail count */
 	atomic_t fail_cont_count;
 	unsigned long reset_fail_timeout;
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 2259ce22cc8b..1f00a2362ae6 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -5834,28 +5834,6 @@ static int ath12k_mac_config_mon_status_default(struct ath12k *ar, bool enable)
 	/* TODO: Need to support new monitor mode */
 }
 
-static void ath12k_mac_wait_reconfigure(struct ath12k_base *ab)
-{
-	int recovery_start_count;
-
-	if (!ab->is_reset)
-		return;
-
-	recovery_start_count = atomic_inc_return(&ab->recovery_start_count);
-
-	ath12k_dbg(ab, ATH12K_DBG_MAC, "recovery start count %d\n", recovery_start_count);
-
-	if (recovery_start_count == ab->num_radios) {
-		complete(&ab->recovery_start);
-		ath12k_dbg(ab, ATH12K_DBG_MAC, "recovery started success\n");
-	}
-
-	ath12k_dbg(ab, ATH12K_DBG_MAC, "waiting reconfigure...\n");
-
-	wait_for_completion_timeout(&ab->reconfigure_complete,
-				    ATH12K_RECONFIGURE_TIMEOUT_HZ);
-}
-
 static int ath12k_mac_start(struct ath12k *ar)
 {
 	struct ath12k_hw *ah = ar->ah;
@@ -5987,7 +5965,6 @@ static int ath12k_mac_op_start(struct ieee80211_hw *hw)
 		break;
 	case ATH12K_HW_STATE_RESTARTING:
 		ah->state = ATH12K_HW_STATE_RESTARTED;
-		ath12k_mac_wait_reconfigure(ah->ab);
 		break;
 	case ATH12K_HW_STATE_RESTARTED:
 	case ATH12K_HW_STATE_WEDGED:
-- 
cgit v1.2.3-58-ga151


From 4f15b06e5782e9a44acc652034090b5fc40b49e5 Mon Sep 17 00:00:00 2001
From: Harshitha Prem <quic_hprem@quicinc.com>
Date: Mon, 17 Jun 2024 14:12:47 +0300
Subject: wifi: ath12k: Remove unused ath12k_base from ath12k_hw

Currently, device (ab) reference in hardware abstraction (ah)
is not used anywhere. Also, with multiple device group abstraction,
hardware abstraction would be coupled with device group abstraction
rather than single device.

Hence, remove the ab reference from hardware abstraction.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Harshitha Prem <quic_hprem@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20240529060939.4156281-1-quic_hprem@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.h | 2 --
 drivers/net/wireless/ath/ath12k/mac.c  | 1 -
 2 files changed, 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index 82ec23a7329d..d7a3e8fddb20 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -630,8 +630,6 @@ struct ath12k {
 
 struct ath12k_hw {
 	struct ieee80211_hw *hw;
-	struct ath12k_base *ab;
-
 	/* Protect the write operation of the hardware state ath12k_hw::state
 	 * between hardware start<=>reconfigure<=>stop transitions.
 	 */
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 1f00a2362ae6..bc2da8530651 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -9254,7 +9254,6 @@ static struct ath12k_hw *ath12k_mac_hw_allocate(struct ath12k_base *ab,
 
 	ah = ath12k_hw_to_ah(hw);
 	ah->hw = hw;
-	ah->ab = ab;
 	ah->num_radio = num_pdev_map;
 
 	mutex_init(&ah->hw_mutex);
-- 
cgit v1.2.3-58-ga151


From 58e185a0dc359a6c1c9eff348d7badfc9f722159 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 13 Jun 2024 10:50:06 +0100
Subject: libbpf: Add btf__distill_base() creating split BTF with distilled
 base BTF

To support more robust split BTF, adding supplemental context for the
base BTF type ids that split BTF refers to is required.  Without such
references, a simple shuffling of base BTF type ids (without any other
significant change) invalidates the split BTF.  Here the attempt is made
to store additional context to make split BTF more robust.

This context comes in the form of distilled base BTF providing minimal
information (name and - in some cases - size) for base INTs, FLOATs,
STRUCTs, UNIONs, ENUMs and ENUM64s along with modified split BTF that
points at that base and contains any additional types needed (such as
TYPEDEF, PTR and anonymous STRUCT/UNION declarations).  This
information constitutes the minimal BTF representation needed to
disambiguate or remove split BTF references to base BTF.  The rules
are as follows:

- INT, FLOAT, FWD are recorded in full.
- if a named base BTF STRUCT or UNION is referred to from split BTF, it
  will be encoded as a zero-member sized STRUCT/UNION (preserving
  size for later relocation checks).  Only base BTF STRUCT/UNIONs
  that are either embedded in split BTF STRUCT/UNIONs or that have
  multiple STRUCT/UNION instances of the same name will _need_ size
  checks at relocation time, but as it is possible a different set of
  types will be duplicates in the later to-be-resolved base BTF,
  we preserve size information for all named STRUCT/UNIONs.
- if an ENUM[64] is named, a ENUM forward representation (an ENUM
  with no values) of the same size is used.
- in all other cases, the type is added to the new split BTF.

Avoiding struct/union/enum/enum64 expansion is important to keep the
distilled base BTF representation to a minimum size.

When successful, new representations of the distilled base BTF and new
split BTF that refers to it are returned.  Both need to be freed by the
caller.

So to take a simple example, with split BTF with a type referring
to "struct sk_buff", we will generate distilled base BTF with a
0-member STRUCT sk_buff of the appropriate size, and the split BTF
will refer to it instead.

Tools like pahole can utilize such split BTF to populate the .BTF
section (split BTF) and an additional .BTF.base section.  Then
when the split BTF is loaded, the distilled base BTF can be used
to relocate split BTF to reference the current (and possibly changed)
base BTF.

So for example if "struct sk_buff" was id 502 when the split BTF was
originally generated,  we can use the distilled base BTF to see that
id 502 refers to a "struct sk_buff" and replace instances of id 502
with the current (relocated) base BTF sk_buff type id.

Distilled base BTF is small; when building a kernel with all modules
using distilled base BTF as a test, overall module size grew by only
5.3Mb total across ~2700 modules.

Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240613095014.357981-2-alan.maguire@oracle.com
---
 tools/lib/bpf/btf.c      | 319 ++++++++++++++++++++++++++++++++++++++++++++++-
 tools/lib/bpf/btf.h      |  21 ++++
 tools/lib/bpf/libbpf.map |   1 +
 3 files changed, 335 insertions(+), 6 deletions(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 775ca55a541c..407ed92b4134 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -1770,9 +1770,8 @@ static int btf_rewrite_str(struct btf_pipe *p, __u32 *str_off)
 	return 0;
 }
 
-int btf__add_type(struct btf *btf, const struct btf *src_btf, const struct btf_type *src_type)
+static int btf_add_type(struct btf_pipe *p, const struct btf_type *src_type)
 {
-	struct btf_pipe p = { .src = src_btf, .dst = btf };
 	struct btf_field_iter it;
 	struct btf_type *t;
 	__u32 *str_off;
@@ -1783,10 +1782,10 @@ int btf__add_type(struct btf *btf, const struct btf *src_btf, const struct btf_t
 		return libbpf_err(sz);
 
 	/* deconstruct BTF, if necessary, and invalidate raw_data */
-	if (btf_ensure_modifiable(btf))
+	if (btf_ensure_modifiable(p->dst))
 		return libbpf_err(-ENOMEM);
 
-	t = btf_add_type_mem(btf, sz);
+	t = btf_add_type_mem(p->dst, sz);
 	if (!t)
 		return libbpf_err(-ENOMEM);
 
@@ -1797,12 +1796,19 @@ int btf__add_type(struct btf *btf, const struct btf *src_btf, const struct btf_t
 		return libbpf_err(err);
 
 	while ((str_off = btf_field_iter_next(&it))) {
-		err = btf_rewrite_str(&p, str_off);
+		err = btf_rewrite_str(p, str_off);
 		if (err)
 			return libbpf_err(err);
 	}
 
-	return btf_commit_type(btf, sz);
+	return btf_commit_type(p->dst, sz);
+}
+
+int btf__add_type(struct btf *btf, const struct btf *src_btf, const struct btf_type *src_type)
+{
+	struct btf_pipe p = { .src = src_btf, .dst = btf };
+
+	return btf_add_type(&p, src_type);
 }
 
 static size_t btf_dedup_identity_hash_fn(long key, void *ctx);
@@ -5276,3 +5282,304 @@ int btf_ext_visit_str_offs(struct btf_ext *btf_ext, str_off_visit_fn visit, void
 
 	return 0;
 }
+
+struct btf_distill {
+	struct btf_pipe pipe;
+	int *id_map;
+	unsigned int split_start_id;
+	unsigned int split_start_str;
+	int diff_id;
+};
+
+static int btf_add_distilled_type_ids(struct btf_distill *dist, __u32 i)
+{
+	struct btf_type *split_t = btf_type_by_id(dist->pipe.src, i);
+	struct btf_field_iter it;
+	__u32 *id;
+	int err;
+
+	err = btf_field_iter_init(&it, split_t, BTF_FIELD_ITER_IDS);
+	if (err)
+		return err;
+	while ((id = btf_field_iter_next(&it))) {
+		struct btf_type *base_t;
+
+		if (!*id)
+			continue;
+		/* split BTF id, not needed */
+		if (*id >= dist->split_start_id)
+			continue;
+		/* already added ? */
+		if (dist->id_map[*id] > 0)
+			continue;
+
+		/* only a subset of base BTF types should be referenced from
+		 * split BTF; ensure nothing unexpected is referenced.
+		 */
+		base_t = btf_type_by_id(dist->pipe.src, *id);
+		switch (btf_kind(base_t)) {
+		case BTF_KIND_INT:
+		case BTF_KIND_FLOAT:
+		case BTF_KIND_FWD:
+		case BTF_KIND_ARRAY:
+		case BTF_KIND_STRUCT:
+		case BTF_KIND_UNION:
+		case BTF_KIND_TYPEDEF:
+		case BTF_KIND_ENUM:
+		case BTF_KIND_ENUM64:
+		case BTF_KIND_PTR:
+		case BTF_KIND_CONST:
+		case BTF_KIND_RESTRICT:
+		case BTF_KIND_VOLATILE:
+		case BTF_KIND_FUNC_PROTO:
+		case BTF_KIND_TYPE_TAG:
+			dist->id_map[*id] = *id;
+			break;
+		default:
+			pr_warn("unexpected reference to base type[%u] of kind [%u] when creating distilled base BTF.\n",
+				*id, btf_kind(base_t));
+			return -EINVAL;
+		}
+		/* If a base type is used, ensure types it refers to are
+		 * marked as used also; so for example if we find a PTR to INT
+		 * we need both the PTR and INT.
+		 *
+		 * The only exception is named struct/unions, since distilled
+		 * base BTF composite types have no members.
+		 */
+		if (btf_is_composite(base_t) && base_t->name_off)
+			continue;
+		err = btf_add_distilled_type_ids(dist, *id);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+static int btf_add_distilled_types(struct btf_distill *dist)
+{
+	bool adding_to_base = dist->pipe.dst->start_id == 1;
+	int id = btf__type_cnt(dist->pipe.dst);
+	struct btf_type *t;
+	int i, err = 0;
+
+
+	/* Add types for each of the required references to either distilled
+	 * base or split BTF, depending on type characteristics.
+	 */
+	for (i = 1; i < dist->split_start_id; i++) {
+		const char *name;
+		int kind;
+
+		if (!dist->id_map[i])
+			continue;
+		t = btf_type_by_id(dist->pipe.src, i);
+		kind = btf_kind(t);
+		name = btf__name_by_offset(dist->pipe.src, t->name_off);
+
+		switch (kind) {
+		case BTF_KIND_INT:
+		case BTF_KIND_FLOAT:
+		case BTF_KIND_FWD:
+			/* Named int, float, fwd are added to base. */
+			if (!adding_to_base)
+				continue;
+			err = btf_add_type(&dist->pipe, t);
+			break;
+		case BTF_KIND_STRUCT:
+		case BTF_KIND_UNION:
+			/* Named struct/union are added to base as 0-vlen
+			 * struct/union of same size.  Anonymous struct/unions
+			 * are added to split BTF as-is.
+			 */
+			if (adding_to_base) {
+				if (!t->name_off)
+					continue;
+				err = btf_add_composite(dist->pipe.dst, kind, name, t->size);
+			} else {
+				if (t->name_off)
+					continue;
+				err = btf_add_type(&dist->pipe, t);
+			}
+			break;
+		case BTF_KIND_ENUM:
+		case BTF_KIND_ENUM64:
+			/* Named enum[64]s are added to base as a sized
+			 * enum; relocation will match with appropriately-named
+			 * and sized enum or enum64.
+			 *
+			 * Anonymous enums are added to split BTF as-is.
+			 */
+			if (adding_to_base) {
+				if (!t->name_off)
+					continue;
+				err = btf__add_enum(dist->pipe.dst, name, t->size);
+			} else {
+				if (t->name_off)
+					continue;
+				err = btf_add_type(&dist->pipe, t);
+			}
+			break;
+		case BTF_KIND_ARRAY:
+		case BTF_KIND_TYPEDEF:
+		case BTF_KIND_PTR:
+		case BTF_KIND_CONST:
+		case BTF_KIND_RESTRICT:
+		case BTF_KIND_VOLATILE:
+		case BTF_KIND_FUNC_PROTO:
+		case BTF_KIND_TYPE_TAG:
+			/* All other types are added to split BTF. */
+			if (adding_to_base)
+				continue;
+			err = btf_add_type(&dist->pipe, t);
+			break;
+		default:
+			pr_warn("unexpected kind when adding base type '%s'[%u] of kind [%u] to distilled base BTF.\n",
+				name, i, kind);
+			return -EINVAL;
+
+		}
+		if (err < 0)
+			break;
+		dist->id_map[i] = id++;
+	}
+	return err;
+}
+
+/* Split BTF ids without a mapping will be shifted downwards since distilled
+ * base BTF is smaller than the original base BTF.  For those that have a
+ * mapping (either to base or updated split BTF), update the id based on
+ * that mapping.
+ */
+static int btf_update_distilled_type_ids(struct btf_distill *dist, __u32 i)
+{
+	struct btf_type *t = btf_type_by_id(dist->pipe.dst, i);
+	struct btf_field_iter it;
+	__u32 *id;
+	int err;
+
+	err = btf_field_iter_init(&it, t, BTF_FIELD_ITER_IDS);
+	if (err)
+		return err;
+	while ((id = btf_field_iter_next(&it))) {
+		if (dist->id_map[*id])
+			*id = dist->id_map[*id];
+		else if (*id >= dist->split_start_id)
+			*id -= dist->diff_id;
+	}
+	return 0;
+}
+
+/* Create updated split BTF with distilled base BTF; distilled base BTF
+ * consists of BTF information required to clarify the types that split
+ * BTF refers to, omitting unneeded details.  Specifically it will contain
+ * base types and memberless definitions of named structs, unions and enumerated
+ * types. Associated reference types like pointers, arrays and anonymous
+ * structs, unions and enumerated types will be added to split BTF.
+ * Size is recorded for named struct/unions to help guide matching to the
+ * target base BTF during later relocation.
+ *
+ * The only case where structs, unions or enumerated types are fully represented
+ * is when they are anonymous; in such cases, the anonymous type is added to
+ * split BTF in full.
+ *
+ * We return newly-created split BTF where the split BTF refers to a newly-created
+ * distilled base BTF. Both must be freed separately by the caller.
+ */
+int btf__distill_base(const struct btf *src_btf, struct btf **new_base_btf,
+		      struct btf **new_split_btf)
+{
+	struct btf *new_base = NULL, *new_split = NULL;
+	const struct btf *old_base;
+	unsigned int n = btf__type_cnt(src_btf);
+	struct btf_distill dist = {};
+	struct btf_type *t;
+	int i, err = 0;
+
+	/* src BTF must be split BTF. */
+	old_base = btf__base_btf(src_btf);
+	if (!new_base_btf || !new_split_btf || !old_base)
+		return libbpf_err(-EINVAL);
+
+	new_base = btf__new_empty();
+	if (!new_base)
+		return libbpf_err(-ENOMEM);
+	dist.id_map = calloc(n, sizeof(*dist.id_map));
+	if (!dist.id_map) {
+		err = -ENOMEM;
+		goto done;
+	}
+	dist.pipe.src = src_btf;
+	dist.pipe.dst = new_base;
+	dist.pipe.str_off_map = hashmap__new(btf_dedup_identity_hash_fn, btf_dedup_equal_fn, NULL);
+	if (IS_ERR(dist.pipe.str_off_map)) {
+		err = -ENOMEM;
+		goto done;
+	}
+	dist.split_start_id = btf__type_cnt(old_base);
+	dist.split_start_str = old_base->hdr->str_len;
+
+	/* Pass over src split BTF; generate the list of base BTF type ids it
+	 * references; these will constitute our distilled BTF set to be
+	 * distributed over base and split BTF as appropriate.
+	 */
+	for (i = src_btf->start_id; i < n; i++) {
+		err = btf_add_distilled_type_ids(&dist, i);
+		if (err < 0)
+			goto done;
+	}
+	/* Next add types for each of the required references to base BTF and split BTF
+	 * in turn.
+	 */
+	err = btf_add_distilled_types(&dist);
+	if (err < 0)
+		goto done;
+
+	/* Create new split BTF with distilled base BTF as its base; the final
+	 * state is split BTF with distilled base BTF that represents enough
+	 * about its base references to allow it to be relocated with the base
+	 * BTF available.
+	 */
+	new_split = btf__new_empty_split(new_base);
+	if (!new_split_btf) {
+		err = -errno;
+		goto done;
+	}
+	dist.pipe.dst = new_split;
+	/* First add all split types */
+	for (i = src_btf->start_id; i < n; i++) {
+		t = btf_type_by_id(src_btf, i);
+		err = btf_add_type(&dist.pipe, t);
+		if (err < 0)
+			goto done;
+	}
+	/* Now add distilled types to split BTF that are not added to base. */
+	err = btf_add_distilled_types(&dist);
+	if (err < 0)
+		goto done;
+
+	/* All split BTF ids will be shifted downwards since there are less base
+	 * BTF ids in distilled base BTF.
+	 */
+	dist.diff_id = dist.split_start_id - btf__type_cnt(new_base);
+
+	n = btf__type_cnt(new_split);
+	/* Now update base/split BTF ids. */
+	for (i = 1; i < n; i++) {
+		err = btf_update_distilled_type_ids(&dist, i);
+		if (err < 0)
+			break;
+	}
+done:
+	free(dist.id_map);
+	hashmap__free(dist.pipe.str_off_map);
+	if (err) {
+		btf__free(new_split);
+		btf__free(new_base);
+		return libbpf_err(err);
+	}
+	*new_base_btf = new_base;
+	*new_split_btf = new_split;
+
+	return 0;
+}
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index 8e6880d91c84..cb08ee9a5a10 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -107,6 +107,27 @@ LIBBPF_API struct btf *btf__new_empty(void);
  */
 LIBBPF_API struct btf *btf__new_empty_split(struct btf *base_btf);
 
+/**
+ * @brief **btf__distill_base()** creates new versions of the split BTF
+ * *src_btf* and its base BTF. The new base BTF will only contain the types
+ * needed to improve robustness of the split BTF to small changes in base BTF.
+ * When that split BTF is loaded against a (possibly changed) base, this
+ * distilled base BTF will help update references to that (possibly changed)
+ * base BTF.
+ *
+ * Both the new split and its associated new base BTF must be freed by
+ * the caller.
+ *
+ * If successful, 0 is returned and **new_base_btf** and **new_split_btf**
+ * will point at new base/split BTF. Both the new split and its associated
+ * new base BTF must be freed by the caller.
+ *
+ * A negative value is returned on error and the thread-local `errno` variable
+ * is set to the error code as well.
+ */
+LIBBPF_API int btf__distill_base(const struct btf *src_btf, struct btf **new_base_btf,
+				 struct btf **new_split_btf);
+
 LIBBPF_API struct btf *btf__parse(const char *path, struct btf_ext **btf_ext);
 LIBBPF_API struct btf *btf__parse_split(const char *path, struct btf *base_btf);
 LIBBPF_API struct btf *btf__parse_elf(const char *path, struct btf_ext **btf_ext);
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 40595233dc7f..175d1536e070 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -419,6 +419,7 @@ LIBBPF_1.4.0 {
 
 LIBBPF_1.5.0 {
 	global:
+		btf__distill_base;
 		bpf_map__autoattach;
 		bpf_map__set_autoattach;
 		bpf_program__attach_sockmap;
-- 
cgit v1.2.3-58-ga151


From eb20e727c4343ad591cff2bef243590c77f62cf1 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 13 Jun 2024 10:50:07 +0100
Subject: selftests/bpf: Test distilled base, split BTF generation

Test generation of split+distilled base BTF, ensuring that

- named base BTF STRUCTs and UNIONs are represented as 0-vlen sized
  STRUCT/UNIONs
- named ENUM[64]s are represented as 0-vlen named ENUM[64]s
- anonymous struct/unions are represented in full in split BTF
- anonymous enums are represented in full in split BTF
- types unreferenced from split BTF are not present in distilled
  base BTF

Also test that with vmlinux BTF and split BTF based upon it,
we only represent needed base types referenced from split BTF
in distilled base.

Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240613095014.357981-3-alan.maguire@oracle.com
---
 .../testing/selftests/bpf/prog_tests/btf_distill.c | 274 +++++++++++++++++++++
 1 file changed, 274 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/btf_distill.c

diff --git a/tools/testing/selftests/bpf/prog_tests/btf_distill.c b/tools/testing/selftests/bpf/prog_tests/btf_distill.c
new file mode 100644
index 000000000000..5c3a38747962
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/btf_distill.c
@@ -0,0 +1,274 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2024, Oracle and/or its affiliates. */
+
+#include <test_progs.h>
+#include <bpf/btf.h>
+#include "btf_helpers.h"
+
+/* Fabricate base, split BTF with references to base types needed; then create
+ * split BTF with distilled base BTF and ensure expectations are met:
+ *  - only referenced base types from split BTF are present
+ *  - struct/union/enum are represented as empty unless anonymous, when they
+ *    are represented in full in split BTF
+ */
+static void test_distilled_base(void)
+{
+	struct btf *btf1 = NULL, *btf2 = NULL, *btf3 = NULL, *btf4 = NULL;
+
+	btf1 = btf__new_empty();
+	if (!ASSERT_OK_PTR(btf1, "empty_main_btf"))
+		return;
+
+	btf__add_int(btf1, "int", 4, BTF_INT_SIGNED);	/* [1] int */
+	btf__add_ptr(btf1, 1);				/* [2] ptr to int */
+	btf__add_struct(btf1, "s1", 8);			/* [3] struct s1 { */
+	btf__add_field(btf1, "f1", 2, 0, 0);		/*      int *f1; */
+							/* } */
+	btf__add_struct(btf1, "", 12);			/* [4] struct { */
+	btf__add_field(btf1, "f1", 1, 0, 0);		/*	int f1; */
+	btf__add_field(btf1, "f2", 3, 32, 0);		/*	struct s1 f2; */
+							/* } */
+	btf__add_int(btf1, "unsigned int", 4, 0);	/* [5] unsigned int */
+	btf__add_union(btf1, "u1", 12);			/* [6] union u1 { */
+	btf__add_field(btf1, "f1", 1, 0, 0);		/*	int f1; */
+	btf__add_field(btf1, "f2", 2, 0, 0);		/*	int *f2; */
+							/* } */
+	btf__add_union(btf1, "", 4);			/* [7] union { */
+	btf__add_field(btf1, "f1", 1, 0, 0);		/*	int f1; */
+							/* } */
+	btf__add_enum(btf1, "e1", 4);			/* [8] enum e1 { */
+	btf__add_enum_value(btf1, "v1", 1);		/*	v1 = 1; */
+							/* } */
+	btf__add_enum(btf1, "", 4);			/* [9] enum { */
+	btf__add_enum_value(btf1, "av1", 2);		/*	av1 = 2; */
+							/* } */
+	btf__add_enum64(btf1, "e641", 8, true);		/* [10] enum64 { */
+	btf__add_enum64_value(btf1, "v1", 1024);	/*	v1 = 1024; */
+							/* } */
+	btf__add_enum64(btf1, "", 8, true);		/* [11] enum64 { */
+	btf__add_enum64_value(btf1, "v1", 1025);	/*	v1 = 1025; */
+							/* } */
+	btf__add_struct(btf1, "unneeded", 4);		/* [12] struct unneeded { */
+	btf__add_field(btf1, "f1", 1, 0, 0);		/*	int f1; */
+							/* } */
+	btf__add_struct(btf1, "embedded", 4);		/* [13] struct embedded { */
+	btf__add_field(btf1, "f1", 1, 0, 0);		/*	int f1; */
+							/* } */
+	btf__add_func_proto(btf1, 1);			/* [14] int (*)(int *p1); */
+	btf__add_func_param(btf1, "p1", 1);
+
+	btf__add_array(btf1, 1, 1, 3);			/* [15] int [3]; */
+
+	btf__add_struct(btf1, "from_proto", 4);		/* [16] struct from_proto { */
+	btf__add_field(btf1, "f1", 1, 0, 0);		/*	int f1; */
+							/* } */
+	btf__add_union(btf1, "u1", 4);			/* [17] union u1 { */
+	btf__add_field(btf1, "f1", 1, 0, 0);		/*	 int f1; */
+							/* } */
+	VALIDATE_RAW_BTF(
+		btf1,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] PTR '(anon)' type_id=1",
+		"[3] STRUCT 's1' size=8 vlen=1\n"
+		"\t'f1' type_id=2 bits_offset=0",
+		"[4] STRUCT '(anon)' size=12 vlen=2\n"
+		"\t'f1' type_id=1 bits_offset=0\n"
+		"\t'f2' type_id=3 bits_offset=32",
+		"[5] INT 'unsigned int' size=4 bits_offset=0 nr_bits=32 encoding=(none)",
+		"[6] UNION 'u1' size=12 vlen=2\n"
+		"\t'f1' type_id=1 bits_offset=0\n"
+		"\t'f2' type_id=2 bits_offset=0",
+		"[7] UNION '(anon)' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[8] ENUM 'e1' encoding=UNSIGNED size=4 vlen=1\n"
+		"\t'v1' val=1",
+		"[9] ENUM '(anon)' encoding=UNSIGNED size=4 vlen=1\n"
+		"\t'av1' val=2",
+		"[10] ENUM64 'e641' encoding=SIGNED size=8 vlen=1\n"
+		"\t'v1' val=1024",
+		"[11] ENUM64 '(anon)' encoding=SIGNED size=8 vlen=1\n"
+		"\t'v1' val=1025",
+		"[12] STRUCT 'unneeded' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[13] STRUCT 'embedded' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[14] FUNC_PROTO '(anon)' ret_type_id=1 vlen=1\n"
+		"\t'p1' type_id=1",
+		"[15] ARRAY '(anon)' type_id=1 index_type_id=1 nr_elems=3",
+		"[16] STRUCT 'from_proto' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[17] UNION 'u1' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0");
+
+	btf2 = btf__new_empty_split(btf1);
+	if (!ASSERT_OK_PTR(btf2, "empty_split_btf"))
+		goto cleanup;
+
+	btf__add_ptr(btf2, 3);				/* [18] ptr to struct s1 */
+	/* add ptr to struct anon */
+	btf__add_ptr(btf2, 4);				/* [19] ptr to struct (anon) */
+	btf__add_const(btf2, 6);			/* [20] const union u1 */
+	btf__add_restrict(btf2, 7);			/* [21] restrict union (anon) */
+	btf__add_volatile(btf2, 8);			/* [22] volatile enum e1 */
+	btf__add_typedef(btf2, "et", 9);		/* [23] typedef enum (anon) */
+	btf__add_const(btf2, 10);			/* [24] const enum64 e641 */
+	btf__add_ptr(btf2, 11);				/* [25] restrict enum64 (anon) */
+	btf__add_struct(btf2, "with_embedded", 4);	/* [26] struct with_embedded { */
+	btf__add_field(btf2, "f1", 13, 0, 0);		/*	struct embedded f1; */
+							/* } */
+	btf__add_func(btf2, "fn", BTF_FUNC_STATIC, 14);	/* [27] int fn(int p1); */
+	btf__add_typedef(btf2, "arraytype", 15);	/* [28] typedef int[3] foo; */
+	btf__add_func_proto(btf2, 1);			/* [29] int (*)(struct from proto p1); */
+	btf__add_func_param(btf2, "p1", 16);
+
+	VALIDATE_RAW_BTF(
+		btf2,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] PTR '(anon)' type_id=1",
+		"[3] STRUCT 's1' size=8 vlen=1\n"
+		"\t'f1' type_id=2 bits_offset=0",
+		"[4] STRUCT '(anon)' size=12 vlen=2\n"
+		"\t'f1' type_id=1 bits_offset=0\n"
+		"\t'f2' type_id=3 bits_offset=32",
+		"[5] INT 'unsigned int' size=4 bits_offset=0 nr_bits=32 encoding=(none)",
+		"[6] UNION 'u1' size=12 vlen=2\n"
+		"\t'f1' type_id=1 bits_offset=0\n"
+		"\t'f2' type_id=2 bits_offset=0",
+		"[7] UNION '(anon)' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[8] ENUM 'e1' encoding=UNSIGNED size=4 vlen=1\n"
+		"\t'v1' val=1",
+		"[9] ENUM '(anon)' encoding=UNSIGNED size=4 vlen=1\n"
+		"\t'av1' val=2",
+		"[10] ENUM64 'e641' encoding=SIGNED size=8 vlen=1\n"
+		"\t'v1' val=1024",
+		"[11] ENUM64 '(anon)' encoding=SIGNED size=8 vlen=1\n"
+		"\t'v1' val=1025",
+		"[12] STRUCT 'unneeded' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[13] STRUCT 'embedded' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[14] FUNC_PROTO '(anon)' ret_type_id=1 vlen=1\n"
+		"\t'p1' type_id=1",
+		"[15] ARRAY '(anon)' type_id=1 index_type_id=1 nr_elems=3",
+		"[16] STRUCT 'from_proto' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[17] UNION 'u1' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[18] PTR '(anon)' type_id=3",
+		"[19] PTR '(anon)' type_id=4",
+		"[20] CONST '(anon)' type_id=6",
+		"[21] RESTRICT '(anon)' type_id=7",
+		"[22] VOLATILE '(anon)' type_id=8",
+		"[23] TYPEDEF 'et' type_id=9",
+		"[24] CONST '(anon)' type_id=10",
+		"[25] PTR '(anon)' type_id=11",
+		"[26] STRUCT 'with_embedded' size=4 vlen=1\n"
+		"\t'f1' type_id=13 bits_offset=0",
+		"[27] FUNC 'fn' type_id=14 linkage=static",
+		"[28] TYPEDEF 'arraytype' type_id=15",
+		"[29] FUNC_PROTO '(anon)' ret_type_id=1 vlen=1\n"
+		"\t'p1' type_id=16");
+
+	if (!ASSERT_EQ(0, btf__distill_base(btf2, &btf3, &btf4),
+		       "distilled_base") ||
+	    !ASSERT_OK_PTR(btf3, "distilled_base") ||
+	    !ASSERT_OK_PTR(btf4, "distilled_split") ||
+	    !ASSERT_EQ(8, btf__type_cnt(btf3), "distilled_base_type_cnt"))
+		goto cleanup;
+
+	VALIDATE_RAW_BTF(
+		btf4,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] STRUCT 's1' size=8 vlen=0",
+		"[3] UNION 'u1' size=12 vlen=0",
+		"[4] ENUM 'e1' encoding=UNSIGNED size=4 vlen=0",
+		"[5] ENUM 'e641' encoding=UNSIGNED size=8 vlen=0",
+		"[6] STRUCT 'embedded' size=4 vlen=0",
+		"[7] STRUCT 'from_proto' size=4 vlen=0",
+		/* split BTF; these types should match split BTF above from 17-28, with
+		 * updated type id references
+		 */
+		"[8] PTR '(anon)' type_id=2",
+		"[9] PTR '(anon)' type_id=20",
+		"[10] CONST '(anon)' type_id=3",
+		"[11] RESTRICT '(anon)' type_id=21",
+		"[12] VOLATILE '(anon)' type_id=4",
+		"[13] TYPEDEF 'et' type_id=22",
+		"[14] CONST '(anon)' type_id=5",
+		"[15] PTR '(anon)' type_id=23",
+		"[16] STRUCT 'with_embedded' size=4 vlen=1\n"
+		"\t'f1' type_id=6 bits_offset=0",
+		"[17] FUNC 'fn' type_id=24 linkage=static",
+		"[18] TYPEDEF 'arraytype' type_id=25",
+		"[19] FUNC_PROTO '(anon)' ret_type_id=1 vlen=1\n"
+		"\t'p1' type_id=7",
+		/* split BTF types added from original base BTF below */
+		"[20] STRUCT '(anon)' size=12 vlen=2\n"
+		"\t'f1' type_id=1 bits_offset=0\n"
+		"\t'f2' type_id=2 bits_offset=32",
+		"[21] UNION '(anon)' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[22] ENUM '(anon)' encoding=UNSIGNED size=4 vlen=1\n"
+		"\t'av1' val=2",
+		"[23] ENUM64 '(anon)' encoding=SIGNED size=8 vlen=1\n"
+		"\t'v1' val=1025",
+		"[24] FUNC_PROTO '(anon)' ret_type_id=1 vlen=1\n"
+		"\t'p1' type_id=1",
+		"[25] ARRAY '(anon)' type_id=1 index_type_id=1 nr_elems=3");
+
+cleanup:
+	btf__free(btf4);
+	btf__free(btf3);
+	btf__free(btf2);
+	btf__free(btf1);
+}
+
+/* create split reference BTF from vmlinux + split BTF with a few type references;
+ * ensure the resultant split reference BTF is as expected, containing only types
+ * needed to disambiguate references from split BTF.
+ */
+static void test_distilled_base_vmlinux(void)
+{
+	struct btf *split_btf = NULL, *vmlinux_btf = btf__load_vmlinux_btf();
+	struct btf *split_dist = NULL, *base_dist = NULL;
+	__s32 int_id, myint_id;
+
+	if (!ASSERT_OK_PTR(vmlinux_btf, "load_vmlinux"))
+		return;
+	int_id = btf__find_by_name_kind(vmlinux_btf, "int", BTF_KIND_INT);
+	if (!ASSERT_GT(int_id, 0, "find_int"))
+		goto cleanup;
+	split_btf = btf__new_empty_split(vmlinux_btf);
+	if (!ASSERT_OK_PTR(split_btf, "new_split"))
+		goto cleanup;
+	myint_id = btf__add_typedef(split_btf, "myint", int_id);
+	btf__add_ptr(split_btf, myint_id);
+
+	if (!ASSERT_EQ(btf__distill_base(split_btf, &base_dist, &split_dist), 0,
+		       "distill_vmlinux_base"))
+		goto cleanup;
+
+	if (!ASSERT_OK_PTR(split_dist, "split_distilled") ||
+	    !ASSERT_OK_PTR(base_dist, "base_dist"))
+		goto cleanup;
+	VALIDATE_RAW_BTF(
+		split_dist,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] TYPEDEF 'myint' type_id=1",
+		"[3] PTR '(anon)' type_id=2");
+
+cleanup:
+	btf__free(split_dist);
+	btf__free(base_dist);
+	btf__free(split_btf);
+	btf__free(vmlinux_btf);
+}
+
+void test_btf_distill(void)
+{
+	if (test__start_subtest("distilled_base"))
+		test_distilled_base();
+	if (test__start_subtest("distilled_base_vmlinux"))
+		test_distilled_base_vmlinux();
+}
-- 
cgit v1.2.3-58-ga151


From 19e00c897d5031bed969dd79af28e899e038009f Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 13 Jun 2024 10:50:08 +0100
Subject: libbpf: Split BTF relocation

Map distilled base BTF type ids referenced in split BTF and their
references to the base BTF passed in, and if the mapping succeeds,
reparent the split BTF to the base BTF.

Relocation is done by first verifying that distilled base BTF
only consists of named INT, FLOAT, ENUM, FWD, STRUCT and
UNION kinds; then we sort these to speed lookups.  Once sorted,
the base BTF is iterated, and for each relevant kind we check
for an equivalent in distilled base BTF.  When found, the
mapping from distilled -> base BTF id and string offset is recorded.
In establishing mappings, we need to ensure we check STRUCT/UNION
size when the STRUCT/UNION is embedded in a split BTF STRUCT/UNION,
and when duplicate names exist for the same STRUCT/UNION.  Otherwise
size is ignored in matching STRUCT/UNIONs.

Once all mappings are established, we can update type ids
and string offsets in split BTF and reparent it to the new base.

Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240613095014.357981-4-alan.maguire@oracle.com
---
 tools/lib/bpf/Build             |   2 +-
 tools/lib/bpf/btf.c             |  17 ++
 tools/lib/bpf/btf.h             |  14 ++
 tools/lib/bpf/btf_relocate.c    | 506 ++++++++++++++++++++++++++++++++++++++++
 tools/lib/bpf/libbpf.map        |   1 +
 tools/lib/bpf/libbpf_internal.h |   3 +
 6 files changed, 542 insertions(+), 1 deletion(-)
 create mode 100644 tools/lib/bpf/btf_relocate.c

diff --git a/tools/lib/bpf/Build b/tools/lib/bpf/Build
index b6619199a706..336da6844d42 100644
--- a/tools/lib/bpf/Build
+++ b/tools/lib/bpf/Build
@@ -1,4 +1,4 @@
 libbpf-y := libbpf.o bpf.o nlattr.o btf.o libbpf_errno.o str_error.o \
 	    netlink.o bpf_prog_linfo.o libbpf_probes.o hashmap.o \
 	    btf_dump.o ringbuf.o strset.o linker.o gen_loader.o relo_core.o \
-	    usdt.o zip.o elf.o features.o
+	    usdt.o zip.o elf.o features.o btf_relocate.o
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 407ed92b4134..5e20354fbcfa 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -5583,3 +5583,20 @@ done:
 
 	return 0;
 }
+
+const struct btf_header *btf_header(const struct btf *btf)
+{
+	return btf->hdr;
+}
+
+void btf_set_base_btf(struct btf *btf, const struct btf *base_btf)
+{
+	btf->base_btf = (struct btf *)base_btf;
+	btf->start_id = btf__type_cnt(base_btf);
+	btf->start_str_off = base_btf->hdr->str_len;
+}
+
+int btf__relocate(struct btf *btf, const struct btf *base_btf)
+{
+	return libbpf_err(btf_relocate(btf, base_btf, NULL));
+}
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index cb08ee9a5a10..8a93120b7385 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -252,6 +252,20 @@ struct btf_dedup_opts {
 
 LIBBPF_API int btf__dedup(struct btf *btf, const struct btf_dedup_opts *opts);
 
+/**
+ * @brief **btf__relocate()** will check the split BTF *btf* for references
+ * to base BTF kinds, and verify those references are compatible with
+ * *base_btf*; if they are, *btf* is adjusted such that is re-parented to
+ * *base_btf* and type ids and strings are adjusted to accommodate this.
+ *
+ * If successful, 0 is returned and **btf** now has **base_btf** as its
+ * base.
+ *
+ * A negative value is returned on error and the thread-local `errno` variable
+ * is set to the error code as well.
+ */
+LIBBPF_API int btf__relocate(struct btf *btf, const struct btf *base_btf);
+
 struct btf_dump;
 
 struct btf_dump_opts {
diff --git a/tools/lib/bpf/btf_relocate.c b/tools/lib/bpf/btf_relocate.c
new file mode 100644
index 000000000000..eabb8755f662
--- /dev/null
+++ b/tools/lib/bpf/btf_relocate.c
@@ -0,0 +1,506 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2024, Oracle and/or its affiliates. */
+
+#ifndef _GNU_SOURCE
+#define _GNU_SOURCE
+#endif
+
+#include "btf.h"
+#include "bpf.h"
+#include "libbpf.h"
+#include "libbpf_internal.h"
+
+struct btf;
+
+struct btf_relocate {
+	struct btf *btf;
+	const struct btf *base_btf;
+	const struct btf *dist_base_btf;
+	unsigned int nr_base_types;
+	unsigned int nr_split_types;
+	unsigned int nr_dist_base_types;
+	int dist_str_len;
+	int base_str_len;
+	__u32 *id_map;
+	__u32 *str_map;
+};
+
+/* Set temporarily in relocation id_map if distilled base struct/union is
+ * embedded in a split BTF struct/union; in such a case, size information must
+ * match between distilled base BTF and base BTF representation of type.
+ */
+#define BTF_IS_EMBEDDED ((__u32)-1)
+
+/* <name, size, id> triple used in sorting/searching distilled base BTF. */
+struct btf_name_info {
+	const char *name;
+	/* set when search requires a size match */
+	int needs_size:1,
+	    size:31;
+	__u32 id;
+};
+
+static int btf_relocate_rewrite_type_id(struct btf_relocate *r, __u32 i)
+{
+	struct btf_type *t = btf_type_by_id(r->btf, i);
+	struct btf_field_iter it;
+	__u32 *id;
+	int err;
+
+	err = btf_field_iter_init(&it, t, BTF_FIELD_ITER_IDS);
+	if (err)
+		return err;
+
+	while ((id = btf_field_iter_next(&it)))
+		*id = r->id_map[*id];
+	return 0;
+}
+
+/* Simple string comparison used for sorting within BTF, since all distilled
+ * types are named.  If strings match, and size is non-zero for both elements
+ * fall back to using size for ordering.
+ */
+static int cmp_btf_name_size(const void *n1, const void *n2)
+{
+	const struct btf_name_info *ni1 = n1;
+	const struct btf_name_info *ni2 = n2;
+	int name_diff = strcmp(ni1->name, ni2->name);
+
+	if (!name_diff && ni1->needs_size && ni2->needs_size)
+		return ni2->size - ni1->size;
+	return name_diff;
+}
+
+/* Binary search with a small twist; find leftmost element that matches
+ * so that we can then iterate through all exact matches.  So for example
+ * searching { "a", "bb", "bb", "c" }  we would always match on the
+ * leftmost "bb".
+ */
+static struct btf_name_info *search_btf_name_size(struct btf_name_info *key,
+						  struct btf_name_info *vals,
+						  int nelems)
+{
+	struct btf_name_info *ret = NULL;
+	int high = nelems - 1;
+	int low = 0;
+
+	while (low <= high) {
+		int mid = (low + high)/2;
+		struct btf_name_info *val = &vals[mid];
+		int diff = cmp_btf_name_size(key, val);
+
+		if (diff == 0)
+			ret = val;
+		/* even if found, keep searching for leftmost match */
+		if (diff <= 0)
+			high = mid - 1;
+		else
+			low = mid + 1;
+	}
+	return ret;
+}
+
+/* If a member of a split BTF struct/union refers to a base BTF
+ * struct/union, mark that struct/union id temporarily in the id_map
+ * with BTF_IS_EMBEDDED.  Members can be const/restrict/volatile/typedef
+ * reference types, but if a pointer is encountered, the type is no longer
+ * considered embedded.
+ */
+static int btf_mark_embedded_composite_type_ids(struct btf_relocate *r, __u32 i)
+{
+	struct btf_type *t = btf_type_by_id(r->btf, i);
+	struct btf_field_iter it;
+	__u32 *id;
+	int err;
+
+	if (!btf_is_composite(t))
+		return 0;
+
+	err = btf_field_iter_init(&it, t, BTF_FIELD_ITER_IDS);
+	if (err)
+		return err;
+
+	while ((id = btf_field_iter_next(&it))) {
+		__u32 next_id = *id;
+
+		while (next_id) {
+			t = btf_type_by_id(r->btf, next_id);
+			switch (btf_kind(t)) {
+			case BTF_KIND_CONST:
+			case BTF_KIND_RESTRICT:
+			case BTF_KIND_VOLATILE:
+			case BTF_KIND_TYPEDEF:
+			case BTF_KIND_TYPE_TAG:
+				next_id = t->type;
+				break;
+			case BTF_KIND_ARRAY: {
+				struct btf_array *a = btf_array(t);
+
+				next_id = a->type;
+				break;
+			}
+			case BTF_KIND_STRUCT:
+			case BTF_KIND_UNION:
+				if (next_id < r->nr_dist_base_types)
+					r->id_map[next_id] = BTF_IS_EMBEDDED;
+				next_id = 0;
+				break;
+			default:
+				next_id = 0;
+				break;
+			}
+		}
+	}
+
+	return 0;
+}
+
+/* Build a map from distilled base BTF ids to base BTF ids. To do so, iterate
+ * through base BTF looking up distilled type (using binary search) equivalents.
+ */
+static int btf_relocate_map_distilled_base(struct btf_relocate *r)
+{
+	struct btf_name_info *dist_base_info_sorted, *dist_base_info_sorted_end;
+	struct btf_type *base_t, *dist_t;
+	__u8 *base_name_cnt = NULL;
+	int err = 0;
+	__u32 id;
+
+	/* generate a sort index array of name/type ids sorted by name for
+	 * distilled base BTF to speed name-based lookups.
+	 */
+	dist_base_info_sorted = calloc(r->nr_dist_base_types, sizeof(*dist_base_info_sorted));
+	if (!dist_base_info_sorted) {
+		err = -ENOMEM;
+		goto done;
+	}
+	dist_base_info_sorted_end = dist_base_info_sorted + r->nr_dist_base_types;
+	for (id = 0; id < r->nr_dist_base_types; id++) {
+		dist_t = btf_type_by_id(r->dist_base_btf, id);
+		dist_base_info_sorted[id].name = btf__name_by_offset(r->dist_base_btf,
+								     dist_t->name_off);
+		dist_base_info_sorted[id].id = id;
+		dist_base_info_sorted[id].size = dist_t->size;
+		dist_base_info_sorted[id].needs_size = true;
+	}
+	qsort(dist_base_info_sorted, r->nr_dist_base_types, sizeof(*dist_base_info_sorted),
+	      cmp_btf_name_size);
+
+	/* Mark distilled base struct/union members of split BTF structs/unions
+	 * in id_map with BTF_IS_EMBEDDED; this signals that these types
+	 * need to match both name and size, otherwise embeddding the base
+	 * struct/union in the split type is invalid.
+	 */
+	for (id = r->nr_dist_base_types; id < r->nr_split_types; id++) {
+		err = btf_mark_embedded_composite_type_ids(r, id);
+		if (err)
+			goto done;
+	}
+
+	/* Collect name counts for composite types in base BTF.  If multiple
+	 * instances of a struct/union of the same name exist, we need to use
+	 * size to determine which to map to since name alone is ambiguous.
+	 */
+	base_name_cnt = calloc(r->base_str_len, sizeof(*base_name_cnt));
+	if (!base_name_cnt) {
+		err = -ENOMEM;
+		goto done;
+	}
+	for (id = 1; id < r->nr_base_types; id++) {
+		base_t = btf_type_by_id(r->base_btf, id);
+		if (!btf_is_composite(base_t) || !base_t->name_off)
+			continue;
+		if (base_name_cnt[base_t->name_off] < 255)
+			base_name_cnt[base_t->name_off]++;
+	}
+
+	/* Now search base BTF for matching distilled base BTF types. */
+	for (id = 1; id < r->nr_base_types; id++) {
+		struct btf_name_info *dist_name_info, *dist_name_info_next = NULL;
+		struct btf_name_info base_name_info = {};
+		int dist_kind, base_kind;
+
+		base_t = btf_type_by_id(r->base_btf, id);
+		/* distilled base consists of named types only. */
+		if (!base_t->name_off)
+			continue;
+		base_kind = btf_kind(base_t);
+		base_name_info.id = id;
+		base_name_info.name = btf__name_by_offset(r->base_btf, base_t->name_off);
+		switch (base_kind) {
+		case BTF_KIND_INT:
+		case BTF_KIND_FLOAT:
+		case BTF_KIND_ENUM:
+		case BTF_KIND_ENUM64:
+			/* These types should match both name and size */
+			base_name_info.needs_size = true;
+			base_name_info.size = base_t->size;
+			break;
+		case BTF_KIND_FWD:
+			/* No size considerations for fwds. */
+			break;
+		case BTF_KIND_STRUCT:
+		case BTF_KIND_UNION:
+			/* Size only needs to be used for struct/union if there
+			 * are multiple types in base BTF with the same name.
+			 * If there are multiple _distilled_ types with the same
+			 * name (a very unlikely scenario), that doesn't matter
+			 * unless corresponding _base_ types to match them are
+			 * missing.
+			 */
+			base_name_info.needs_size = base_name_cnt[base_t->name_off] > 1;
+			base_name_info.size = base_t->size;
+			break;
+		default:
+			continue;
+		}
+		/* iterate over all matching distilled base types */
+		for (dist_name_info = search_btf_name_size(&base_name_info, dist_base_info_sorted,
+							   r->nr_dist_base_types);
+		     dist_name_info != NULL; dist_name_info = dist_name_info_next) {
+			/* Are there more distilled matches to process after
+			 * this one?
+			 */
+			dist_name_info_next = dist_name_info + 1;
+			if (dist_name_info_next >= dist_base_info_sorted_end ||
+			    cmp_btf_name_size(&base_name_info, dist_name_info_next))
+				dist_name_info_next = NULL;
+
+			if (!dist_name_info->id || dist_name_info->id > r->nr_dist_base_types) {
+				pr_warn("base BTF id [%d] maps to invalid distilled base BTF id [%d]\n",
+					id, dist_name_info->id);
+				err = -EINVAL;
+				goto done;
+			}
+			dist_t = btf_type_by_id(r->dist_base_btf, dist_name_info->id);
+			dist_kind = btf_kind(dist_t);
+
+			/* Validate that the found distilled type is compatible.
+			 * Do not error out on mismatch as another match may
+			 * occur for an identically-named type.
+			 */
+			switch (dist_kind) {
+			case BTF_KIND_FWD:
+				switch (base_kind) {
+				case BTF_KIND_FWD:
+					if (btf_kflag(dist_t) != btf_kflag(base_t))
+						continue;
+					break;
+				case BTF_KIND_STRUCT:
+					if (btf_kflag(base_t))
+						continue;
+					break;
+				case BTF_KIND_UNION:
+					if (!btf_kflag(base_t))
+						continue;
+					break;
+				default:
+					continue;
+				}
+				break;
+			case BTF_KIND_INT:
+				if (dist_kind != base_kind ||
+				    btf_int_encoding(base_t) != btf_int_encoding(dist_t))
+					continue;
+				break;
+			case BTF_KIND_FLOAT:
+				if (dist_kind != base_kind)
+					continue;
+				break;
+			case BTF_KIND_ENUM:
+				/* ENUM and ENUM64 are encoded as sized ENUM in
+				 * distilled base BTF.
+				 */
+				if (base_kind != dist_kind && base_kind != BTF_KIND_ENUM64)
+					continue;
+				break;
+			case BTF_KIND_STRUCT:
+			case BTF_KIND_UNION:
+				/* size verification is required for embedded
+				 * struct/unions.
+				 */
+				if (r->id_map[dist_name_info->id] == BTF_IS_EMBEDDED &&
+				    base_t->size != dist_t->size)
+					continue;
+				break;
+			default:
+				continue;
+			}
+			if (r->id_map[dist_name_info->id] &&
+			    r->id_map[dist_name_info->id] != BTF_IS_EMBEDDED) {
+				/* we already have a match; this tells us that
+				 * multiple base types of the same name
+				 * have the same size, since for cases where
+				 * multiple types have the same name we match
+				 * on name and size.  In this case, we have
+				 * no way of determining which to relocate
+				 * to in base BTF, so error out.
+				 */
+				pr_warn("distilled base BTF type '%s' [%u], size %u has multiple candidates of the same size (ids [%u, %u]) in base BTF\n",
+					base_name_info.name, dist_name_info->id,
+					base_t->size, id, r->id_map[dist_name_info->id]);
+				err = -EINVAL;
+				goto done;
+			}
+			/* map id and name */
+			r->id_map[dist_name_info->id] = id;
+			r->str_map[dist_t->name_off] = base_t->name_off;
+		}
+	}
+	/* ensure all distilled BTF ids now have a mapping... */
+	for (id = 1; id < r->nr_dist_base_types; id++) {
+		const char *name;
+
+		if (r->id_map[id] && r->id_map[id] != BTF_IS_EMBEDDED)
+			continue;
+		dist_t = btf_type_by_id(r->dist_base_btf, id);
+		name = btf__name_by_offset(r->dist_base_btf, dist_t->name_off);
+		pr_warn("distilled base BTF type '%s' [%d] is not mapped to base BTF id\n",
+			name, id);
+		err = -EINVAL;
+		break;
+	}
+done:
+	free(base_name_cnt);
+	free(dist_base_info_sorted);
+	return err;
+}
+
+/* distilled base should only have named int/float/enum/fwd/struct/union types. */
+static int btf_relocate_validate_distilled_base(struct btf_relocate *r)
+{
+	unsigned int i;
+
+	for (i = 1; i < r->nr_dist_base_types; i++) {
+		struct btf_type *t = btf_type_by_id(r->dist_base_btf, i);
+		int kind = btf_kind(t);
+
+		switch (kind) {
+		case BTF_KIND_INT:
+		case BTF_KIND_FLOAT:
+		case BTF_KIND_ENUM:
+		case BTF_KIND_STRUCT:
+		case BTF_KIND_UNION:
+		case BTF_KIND_FWD:
+			if (t->name_off)
+				break;
+			pr_warn("type [%d], kind [%d] is invalid for distilled base BTF; it is anonymous\n",
+				i, kind);
+			return -EINVAL;
+		default:
+			pr_warn("type [%d] in distilled based BTF has unexpected kind [%d]\n",
+				i, kind);
+			return -EINVAL;
+		}
+	}
+	return 0;
+}
+
+static int btf_relocate_rewrite_strs(struct btf_relocate *r, __u32 i)
+{
+	struct btf_type *t = btf_type_by_id(r->btf, i);
+	struct btf_field_iter it;
+	__u32 *str_off;
+	int off, err;
+
+	err = btf_field_iter_init(&it, t, BTF_FIELD_ITER_STRS);
+	if (err)
+		return err;
+
+	while ((str_off = btf_field_iter_next(&it))) {
+		if (!*str_off)
+			continue;
+		if (*str_off >= r->dist_str_len) {
+			*str_off += r->base_str_len - r->dist_str_len;
+		} else {
+			off = r->str_map[*str_off];
+			if (!off) {
+				pr_warn("string '%s' [offset %u] is not mapped to base BTF",
+					btf__str_by_offset(r->btf, off), *str_off);
+				return -ENOENT;
+			}
+			*str_off = off;
+		}
+	}
+	return 0;
+}
+
+/* If successful, output of relocation is updated BTF with base BTF pointing
+ * at base_btf, and type ids, strings adjusted accordingly.
+ */
+int btf_relocate(struct btf *btf, const struct btf *base_btf, __u32 **id_map)
+{
+	unsigned int nr_types = btf__type_cnt(btf);
+	const struct btf_header *dist_base_hdr;
+	const struct btf_header *base_hdr;
+	struct btf_relocate r = {};
+	int err = 0;
+	__u32 id, i;
+
+	r.dist_base_btf = btf__base_btf(btf);
+	if (!base_btf || r.dist_base_btf == base_btf)
+		return -EINVAL;
+
+	r.nr_dist_base_types = btf__type_cnt(r.dist_base_btf);
+	r.nr_base_types = btf__type_cnt(base_btf);
+	r.nr_split_types = nr_types - r.nr_dist_base_types;
+	r.btf = btf;
+	r.base_btf = base_btf;
+
+	r.id_map = calloc(nr_types, sizeof(*r.id_map));
+	r.str_map = calloc(btf_header(r.dist_base_btf)->str_len, sizeof(*r.str_map));
+	dist_base_hdr = btf_header(r.dist_base_btf);
+	base_hdr = btf_header(r.base_btf);
+	r.dist_str_len = dist_base_hdr->str_len;
+	r.base_str_len = base_hdr->str_len;
+	if (!r.id_map || !r.str_map) {
+		err = -ENOMEM;
+		goto err_out;
+	}
+
+	err = btf_relocate_validate_distilled_base(&r);
+	if (err)
+		goto err_out;
+
+	/* Split BTF ids need to be adjusted as base and distilled base
+	 * have different numbers of types, changing the start id of split
+	 * BTF.
+	 */
+	for (id = r.nr_dist_base_types; id < nr_types; id++)
+		r.id_map[id] = id + r.nr_base_types - r.nr_dist_base_types;
+
+	/* Build a map from distilled base ids to actual base BTF ids; it is used
+	 * to update split BTF id references.  Also build a str_map mapping from
+	 * distilled base BTF names to base BTF names.
+	 */
+	err = btf_relocate_map_distilled_base(&r);
+	if (err)
+		goto err_out;
+
+	/* Next, rewrite type ids in split BTF, replacing split ids with updated
+	 * ids based on number of types in base BTF, and base ids with
+	 * relocated ids from base_btf.
+	 */
+	for (i = 0, id = r.nr_dist_base_types; i < r.nr_split_types; i++, id++) {
+		err = btf_relocate_rewrite_type_id(&r, id);
+		if (err)
+			goto err_out;
+	}
+	/* String offsets now need to be updated using the str_map. */
+	for (i = 0; i < r.nr_split_types; i++) {
+		err = btf_relocate_rewrite_strs(&r, i + r.nr_dist_base_types);
+		if (err)
+			goto err_out;
+	}
+	/* Finally reset base BTF to be base_btf */
+	btf_set_base_btf(btf, base_btf);
+
+	if (id_map) {
+		*id_map = r.id_map;
+		r.id_map = NULL;
+	}
+err_out:
+	free(r.id_map);
+	free(r.str_map);
+	return err;
+}
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 175d1536e070..8f0d9ea3b1b4 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -420,6 +420,7 @@ LIBBPF_1.4.0 {
 LIBBPF_1.5.0 {
 	global:
 		btf__distill_base;
+		btf__relocate;
 		bpf_map__autoattach;
 		bpf_map__set_autoattach;
 		bpf_program__attach_sockmap;
diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index e2f06609c624..408df59e0771 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -234,6 +234,9 @@ struct btf_type;
 struct btf_type *btf_type_by_id(const struct btf *btf, __u32 type_id);
 const char *btf_kind_str(const struct btf_type *t);
 const struct btf_type *skip_mods_and_typedefs(const struct btf *btf, __u32 id, __u32 *res_id);
+const struct btf_header *btf_header(const struct btf *btf);
+void btf_set_base_btf(struct btf *btf, const struct btf *base_btf);
+int btf_relocate(struct btf *btf, const struct btf *base_btf, __u32 **id_map);
 
 static inline enum btf_func_linkage btf_func_linkage(const struct btf_type *t)
 {
-- 
cgit v1.2.3-58-ga151


From affdeb50616b190c3236cc2bf116e1b931a43be2 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 13 Jun 2024 10:50:09 +0100
Subject: selftests/bpf: Extend distilled BTF tests to cover BTF relocation

Ensure relocated BTF looks as expected; in this case identical to
original split BTF, with a few duplicate anonymous types added to
split BTF by the relocation process.  Also add relocation tests
for edge cases like missing type in base BTF and multiple types
of the same name.

Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240613095014.357981-5-alan.maguire@oracle.com
---
 .../testing/selftests/bpf/prog_tests/btf_distill.c | 278 +++++++++++++++++++++
 1 file changed, 278 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/btf_distill.c b/tools/testing/selftests/bpf/prog_tests/btf_distill.c
index 5c3a38747962..bfbe795823a2 100644
--- a/tools/testing/selftests/bpf/prog_tests/btf_distill.c
+++ b/tools/testing/selftests/bpf/prog_tests/btf_distill.c
@@ -217,7 +217,277 @@ static void test_distilled_base(void)
 		"\t'p1' type_id=1",
 		"[25] ARRAY '(anon)' type_id=1 index_type_id=1 nr_elems=3");
 
+	if (!ASSERT_EQ(btf__relocate(btf4, btf1), 0, "relocate_split"))
+		goto cleanup;
+
+	VALIDATE_RAW_BTF(
+		btf4,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] PTR '(anon)' type_id=1",
+		"[3] STRUCT 's1' size=8 vlen=1\n"
+		"\t'f1' type_id=2 bits_offset=0",
+		"[4] STRUCT '(anon)' size=12 vlen=2\n"
+		"\t'f1' type_id=1 bits_offset=0\n"
+		"\t'f2' type_id=3 bits_offset=32",
+		"[5] INT 'unsigned int' size=4 bits_offset=0 nr_bits=32 encoding=(none)",
+		"[6] UNION 'u1' size=12 vlen=2\n"
+		"\t'f1' type_id=1 bits_offset=0\n"
+		"\t'f2' type_id=2 bits_offset=0",
+		"[7] UNION '(anon)' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[8] ENUM 'e1' encoding=UNSIGNED size=4 vlen=1\n"
+		"\t'v1' val=1",
+		"[9] ENUM '(anon)' encoding=UNSIGNED size=4 vlen=1\n"
+		"\t'av1' val=2",
+		"[10] ENUM64 'e641' encoding=SIGNED size=8 vlen=1\n"
+		"\t'v1' val=1024",
+		"[11] ENUM64 '(anon)' encoding=SIGNED size=8 vlen=1\n"
+		"\t'v1' val=1025",
+		"[12] STRUCT 'unneeded' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[13] STRUCT 'embedded' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[14] FUNC_PROTO '(anon)' ret_type_id=1 vlen=1\n"
+		"\t'p1' type_id=1",
+		"[15] ARRAY '(anon)' type_id=1 index_type_id=1 nr_elems=3",
+		"[16] STRUCT 'from_proto' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[17] UNION 'u1' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[18] PTR '(anon)' type_id=3",
+		"[19] PTR '(anon)' type_id=30",
+		"[20] CONST '(anon)' type_id=6",
+		"[21] RESTRICT '(anon)' type_id=31",
+		"[22] VOLATILE '(anon)' type_id=8",
+		"[23] TYPEDEF 'et' type_id=32",
+		"[24] CONST '(anon)' type_id=10",
+		"[25] PTR '(anon)' type_id=33",
+		"[26] STRUCT 'with_embedded' size=4 vlen=1\n"
+		"\t'f1' type_id=13 bits_offset=0",
+		"[27] FUNC 'fn' type_id=34 linkage=static",
+		"[28] TYPEDEF 'arraytype' type_id=35",
+		"[29] FUNC_PROTO '(anon)' ret_type_id=1 vlen=1\n"
+		"\t'p1' type_id=16",
+		/* below here are (duplicate) anon base types added by distill
+		 * process to split BTF.
+		 */
+		"[30] STRUCT '(anon)' size=12 vlen=2\n"
+		"\t'f1' type_id=1 bits_offset=0\n"
+		"\t'f2' type_id=3 bits_offset=32",
+		"[31] UNION '(anon)' size=4 vlen=1\n"
+		"\t'f1' type_id=1 bits_offset=0",
+		"[32] ENUM '(anon)' encoding=UNSIGNED size=4 vlen=1\n"
+		"\t'av1' val=2",
+		"[33] ENUM64 '(anon)' encoding=SIGNED size=8 vlen=1\n"
+		"\t'v1' val=1025",
+		"[34] FUNC_PROTO '(anon)' ret_type_id=1 vlen=1\n"
+		"\t'p1' type_id=1",
+		"[35] ARRAY '(anon)' type_id=1 index_type_id=1 nr_elems=3");
+
+cleanup:
+	btf__free(btf4);
+	btf__free(btf3);
+	btf__free(btf2);
+	btf__free(btf1);
+}
+
+/* ensure we can cope with multiple types with the same name in
+ * distilled base BTF.  In this case because sizes are different,
+ * we can still disambiguate them.
+ */
+static void test_distilled_base_multi(void)
+{
+	struct btf *btf1 = NULL, *btf2 = NULL, *btf3 = NULL, *btf4 = NULL;
+
+	btf1 = btf__new_empty();
+	if (!ASSERT_OK_PTR(btf1, "empty_main_btf"))
+		return;
+	btf__add_int(btf1, "int", 4, BTF_INT_SIGNED);   /* [1] int */
+	btf__add_int(btf1, "int", 8, BTF_INT_SIGNED);	/* [2] int */
+	VALIDATE_RAW_BTF(
+		btf1,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int' size=8 bits_offset=0 nr_bits=64 encoding=SIGNED");
+	btf2 = btf__new_empty_split(btf1);
+	if (!ASSERT_OK_PTR(btf2, "empty_split_btf"))
+		goto cleanup;
+	btf__add_ptr(btf2, 1);
+	btf__add_const(btf2, 2);
+	VALIDATE_RAW_BTF(
+		btf2,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int' size=8 bits_offset=0 nr_bits=64 encoding=SIGNED",
+		"[3] PTR '(anon)' type_id=1",
+		"[4] CONST '(anon)' type_id=2");
+	if (!ASSERT_EQ(0, btf__distill_base(btf2, &btf3, &btf4),
+		       "distilled_base") ||
+	    !ASSERT_OK_PTR(btf3, "distilled_base") ||
+	    !ASSERT_OK_PTR(btf4, "distilled_split") ||
+	    !ASSERT_EQ(3, btf__type_cnt(btf3), "distilled_base_type_cnt"))
+		goto cleanup;
+	VALIDATE_RAW_BTF(
+		btf3,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int' size=8 bits_offset=0 nr_bits=64 encoding=SIGNED");
+	if (!ASSERT_EQ(btf__relocate(btf4, btf1), 0, "relocate_split"))
+		goto cleanup;
+
+	VALIDATE_RAW_BTF(
+		btf4,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int' size=8 bits_offset=0 nr_bits=64 encoding=SIGNED",
+		"[3] PTR '(anon)' type_id=1",
+		"[4] CONST '(anon)' type_id=2");
+
+cleanup:
+	btf__free(btf4);
+	btf__free(btf3);
+	btf__free(btf2);
+	btf__free(btf1);
+}
+
+/* If a needed type is not present in the base BTF we wish to relocate
+ * with, btf__relocate() should error our.
+ */
+static void test_distilled_base_missing_err(void)
+{
+	struct btf *btf1 = NULL, *btf2 = NULL, *btf3 = NULL, *btf4 = NULL, *btf5 = NULL;
+
+	btf1 = btf__new_empty();
+	if (!ASSERT_OK_PTR(btf1, "empty_main_btf"))
+		return;
+	btf__add_int(btf1, "int", 4, BTF_INT_SIGNED);   /* [1] int */
+	btf__add_int(btf1, "int", 8, BTF_INT_SIGNED);   /* [2] int */
+	VALIDATE_RAW_BTF(
+		btf1,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int' size=8 bits_offset=0 nr_bits=64 encoding=SIGNED");
+	btf2 = btf__new_empty_split(btf1);
+	if (!ASSERT_OK_PTR(btf2, "empty_split_btf"))
+		goto cleanup;
+	btf__add_ptr(btf2, 1);
+	btf__add_const(btf2, 2);
+	VALIDATE_RAW_BTF(
+		btf2,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int' size=8 bits_offset=0 nr_bits=64 encoding=SIGNED",
+		"[3] PTR '(anon)' type_id=1",
+		"[4] CONST '(anon)' type_id=2");
+	if (!ASSERT_EQ(0, btf__distill_base(btf2, &btf3, &btf4),
+		       "distilled_base") ||
+	    !ASSERT_OK_PTR(btf3, "distilled_base") ||
+	    !ASSERT_OK_PTR(btf4, "distilled_split") ||
+	    !ASSERT_EQ(3, btf__type_cnt(btf3), "distilled_base_type_cnt"))
+		goto cleanup;
+	VALIDATE_RAW_BTF(
+		btf3,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int' size=8 bits_offset=0 nr_bits=64 encoding=SIGNED");
+	btf5 = btf__new_empty();
+	if (!ASSERT_OK_PTR(btf5, "empty_reloc_btf"))
+		return;
+	btf__add_int(btf5, "int", 4, BTF_INT_SIGNED);   /* [1] int */
+	VALIDATE_RAW_BTF(
+		btf5,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED");
+	ASSERT_EQ(btf__relocate(btf4, btf5), -EINVAL, "relocate_split");
+
+cleanup:
+	btf__free(btf5);
+	btf__free(btf4);
+	btf__free(btf3);
+	btf__free(btf2);
+	btf__free(btf1);
+}
+
+/* With 2 types of same size in distilled base BTF, relocation should
+ * fail as we have no means to choose between them.
+ */
+static void test_distilled_base_multi_err(void)
+{
+	struct btf *btf1 = NULL, *btf2 = NULL, *btf3 = NULL, *btf4 = NULL;
+
+	btf1 = btf__new_empty();
+	if (!ASSERT_OK_PTR(btf1, "empty_main_btf"))
+		return;
+	btf__add_int(btf1, "int", 4, BTF_INT_SIGNED);   /* [1] int */
+	btf__add_int(btf1, "int", 4, BTF_INT_SIGNED);   /* [2] int */
+	VALIDATE_RAW_BTF(
+		btf1,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED");
+	btf2 = btf__new_empty_split(btf1);
+	if (!ASSERT_OK_PTR(btf2, "empty_split_btf"))
+		goto cleanup;
+	btf__add_ptr(btf2, 1);
+	btf__add_const(btf2, 2);
+	VALIDATE_RAW_BTF(
+		btf2,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[3] PTR '(anon)' type_id=1",
+		"[4] CONST '(anon)' type_id=2");
+	if (!ASSERT_EQ(0, btf__distill_base(btf2, &btf3, &btf4),
+		       "distilled_base") ||
+	    !ASSERT_OK_PTR(btf3, "distilled_base") ||
+	    !ASSERT_OK_PTR(btf4, "distilled_split") ||
+	    !ASSERT_EQ(3, btf__type_cnt(btf3), "distilled_base_type_cnt"))
+		goto cleanup;
+	VALIDATE_RAW_BTF(
+		btf3,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED");
+	ASSERT_EQ(btf__relocate(btf4, btf1), -EINVAL, "relocate_split");
+cleanup:
+	btf__free(btf4);
+	btf__free(btf3);
+	btf__free(btf2);
+	btf__free(btf1);
+}
+
+/* With 2 types of same size in base BTF, relocation should
+ * fail as we have no means to choose between them.
+ */
+static void test_distilled_base_multi_err2(void)
+{
+	struct btf *btf1 = NULL, *btf2 = NULL, *btf3 = NULL, *btf4 = NULL, *btf5 = NULL;
+
+	btf1 = btf__new_empty();
+	if (!ASSERT_OK_PTR(btf1, "empty_main_btf"))
+		return;
+	btf__add_int(btf1, "int", 4, BTF_INT_SIGNED);   /* [1] int */
+	VALIDATE_RAW_BTF(
+		btf1,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED");
+	btf2 = btf__new_empty_split(btf1);
+	if (!ASSERT_OK_PTR(btf2, "empty_split_btf"))
+		goto cleanup;
+	btf__add_ptr(btf2, 1);
+	VALIDATE_RAW_BTF(
+		btf2,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] PTR '(anon)' type_id=1");
+	if (!ASSERT_EQ(0, btf__distill_base(btf2, &btf3, &btf4),
+		       "distilled_base") ||
+	    !ASSERT_OK_PTR(btf3, "distilled_base") ||
+	    !ASSERT_OK_PTR(btf4, "distilled_split") ||
+	    !ASSERT_EQ(2, btf__type_cnt(btf3), "distilled_base_type_cnt"))
+		goto cleanup;
+	VALIDATE_RAW_BTF(
+		btf3,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED");
+	btf5 = btf__new_empty();
+	if (!ASSERT_OK_PTR(btf5, "empty_reloc_btf"))
+		return;
+	btf__add_int(btf5, "int", 4, BTF_INT_SIGNED);   /* [1] int */
+	btf__add_int(btf5, "int", 4, BTF_INT_SIGNED);   /* [2] int */
+	VALIDATE_RAW_BTF(
+		btf5,
+		"[1] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED",
+		"[2] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED");
+	ASSERT_EQ(btf__relocate(btf4, btf5), -EINVAL, "relocate_split");
 cleanup:
+	btf__free(btf5);
 	btf__free(btf4);
 	btf__free(btf3);
 	btf__free(btf2);
@@ -269,6 +539,14 @@ void test_btf_distill(void)
 {
 	if (test__start_subtest("distilled_base"))
 		test_distilled_base();
+	if (test__start_subtest("distilled_base_multi"))
+		test_distilled_base_multi();
+	if (test__start_subtest("distilled_base_missing_err"))
+		test_distilled_base_missing_err();
+	if (test__start_subtest("distilled_base_multi_err"))
+		test_distilled_base_multi_err();
+	if (test__start_subtest("distilled_base_multi_err2"))
+		test_distilled_base_multi_err2();
 	if (test__start_subtest("distilled_base_vmlinux"))
 		test_distilled_base_vmlinux();
 }
-- 
cgit v1.2.3-58-ga151


From c86f180ffc993975fed5907a869fc9b1555d0cfb Mon Sep 17 00:00:00 2001
From: Eduard Zingerman <eddyz87@gmail.com>
Date: Thu, 13 Jun 2024 10:50:10 +0100
Subject: libbpf: Make btf_parse_elf process .BTF.base transparently

Update btf_parse_elf() to check if .BTF.base section is present.
The logic is as follows:

  if .BTF.base section exists:
     distilled_base := btf_new(.BTF.base)
  if distilled_base:
     btf := btf_new(.BTF, .base_btf=distilled_base)
     if base_btf:
        btf_relocate(btf, base_btf)
  else:
     btf := btf_new(.BTF)
  return btf

In other words:
- if .BTF.base section exists, load BTF from it and use it as a base
  for .BTF load;
- if base_btf is specified and .BTF.base section exist, relocate newly
  loaded .BTF against base_btf.

Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240613095014.357981-6-alan.maguire@oracle.com
---
 tools/lib/bpf/btf.c | 164 +++++++++++++++++++++++++++++++++++-----------------
 tools/lib/bpf/btf.h |   1 +
 2 files changed, 111 insertions(+), 54 deletions(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 5e20354fbcfa..ef1b2f573c1b 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -116,6 +116,9 @@ struct btf {
 	/* whether strings are already deduplicated */
 	bool strs_deduped;
 
+	/* whether base_btf should be freed in btf_free for this instance */
+	bool owns_base;
+
 	/* BTF object FD, if loaded into kernel */
 	int fd;
 
@@ -969,6 +972,8 @@ void btf__free(struct btf *btf)
 	free(btf->raw_data);
 	free(btf->raw_data_swapped);
 	free(btf->type_offs);
+	if (btf->owns_base)
+		btf__free(btf->base_btf);
 	free(btf);
 }
 
@@ -1084,53 +1089,38 @@ struct btf *btf__new_split(const void *data, __u32 size, struct btf *base_btf)
 	return libbpf_ptr(btf_new(data, size, base_btf));
 }
 
-static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
-				 struct btf_ext **btf_ext)
+struct btf_elf_secs {
+	Elf_Data *btf_data;
+	Elf_Data *btf_ext_data;
+	Elf_Data *btf_base_data;
+};
+
+static int btf_find_elf_sections(Elf *elf, const char *path, struct btf_elf_secs *secs)
 {
-	Elf_Data *btf_data = NULL, *btf_ext_data = NULL;
-	int err = 0, fd = -1, idx = 0;
-	struct btf *btf = NULL;
 	Elf_Scn *scn = NULL;
-	Elf *elf = NULL;
+	Elf_Data *data;
 	GElf_Ehdr ehdr;
 	size_t shstrndx;
+	int idx = 0;
 
-	if (elf_version(EV_CURRENT) == EV_NONE) {
-		pr_warn("failed to init libelf for %s\n", path);
-		return ERR_PTR(-LIBBPF_ERRNO__LIBELF);
-	}
-
-	fd = open(path, O_RDONLY | O_CLOEXEC);
-	if (fd < 0) {
-		err = -errno;
-		pr_warn("failed to open %s: %s\n", path, strerror(errno));
-		return ERR_PTR(err);
-	}
-
-	err = -LIBBPF_ERRNO__FORMAT;
-
-	elf = elf_begin(fd, ELF_C_READ, NULL);
-	if (!elf) {
-		pr_warn("failed to open %s as ELF file\n", path);
-		goto done;
-	}
 	if (!gelf_getehdr(elf, &ehdr)) {
 		pr_warn("failed to get EHDR from %s\n", path);
-		goto done;
+		goto err;
 	}
 
 	if (elf_getshdrstrndx(elf, &shstrndx)) {
 		pr_warn("failed to get section names section index for %s\n",
 			path);
-		goto done;
+		goto err;
 	}
 
 	if (!elf_rawdata(elf_getscn(elf, shstrndx), NULL)) {
 		pr_warn("failed to get e_shstrndx from %s\n", path);
-		goto done;
+		goto err;
 	}
 
 	while ((scn = elf_nextscn(elf, scn)) != NULL) {
+		Elf_Data **field;
 		GElf_Shdr sh;
 		char *name;
 
@@ -1138,42 +1128,102 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
 		if (gelf_getshdr(scn, &sh) != &sh) {
 			pr_warn("failed to get section(%d) header from %s\n",
 				idx, path);
-			goto done;
+			goto err;
 		}
 		name = elf_strptr(elf, shstrndx, sh.sh_name);
 		if (!name) {
 			pr_warn("failed to get section(%d) name from %s\n",
 				idx, path);
-			goto done;
+			goto err;
 		}
-		if (strcmp(name, BTF_ELF_SEC) == 0) {
-			btf_data = elf_getdata(scn, 0);
-			if (!btf_data) {
-				pr_warn("failed to get section(%d, %s) data from %s\n",
-					idx, name, path);
-				goto done;
-			}
-			continue;
-		} else if (btf_ext && strcmp(name, BTF_EXT_ELF_SEC) == 0) {
-			btf_ext_data = elf_getdata(scn, 0);
-			if (!btf_ext_data) {
-				pr_warn("failed to get section(%d, %s) data from %s\n",
-					idx, name, path);
-				goto done;
-			}
+
+		if (strcmp(name, BTF_ELF_SEC) == 0)
+			field = &secs->btf_data;
+		else if (strcmp(name, BTF_EXT_ELF_SEC) == 0)
+			field = &secs->btf_ext_data;
+		else if (strcmp(name, BTF_BASE_ELF_SEC) == 0)
+			field = &secs->btf_base_data;
+		else
 			continue;
+
+		data = elf_getdata(scn, 0);
+		if (!data) {
+			pr_warn("failed to get section(%d, %s) data from %s\n",
+				idx, name, path);
+			goto err;
 		}
+		*field = data;
+	}
+
+	return 0;
+
+err:
+	return -LIBBPF_ERRNO__FORMAT;
+}
+
+static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
+				 struct btf_ext **btf_ext)
+{
+	struct btf_elf_secs secs = {};
+	struct btf *dist_base_btf = NULL;
+	struct btf *btf = NULL;
+	int err = 0, fd = -1;
+	Elf *elf = NULL;
+
+	if (elf_version(EV_CURRENT) == EV_NONE) {
+		pr_warn("failed to init libelf for %s\n", path);
+		return ERR_PTR(-LIBBPF_ERRNO__LIBELF);
+	}
+
+	fd = open(path, O_RDONLY | O_CLOEXEC);
+	if (fd < 0) {
+		err = -errno;
+		pr_warn("failed to open %s: %s\n", path, strerror(errno));
+		return ERR_PTR(err);
 	}
 
-	if (!btf_data) {
+	elf = elf_begin(fd, ELF_C_READ, NULL);
+	if (!elf) {
+		pr_warn("failed to open %s as ELF file\n", path);
+		goto done;
+	}
+
+	err = btf_find_elf_sections(elf, path, &secs);
+	if (err)
+		goto done;
+
+	if (!secs.btf_data) {
 		pr_warn("failed to find '%s' ELF section in %s\n", BTF_ELF_SEC, path);
 		err = -ENODATA;
 		goto done;
 	}
-	btf = btf_new(btf_data->d_buf, btf_data->d_size, base_btf);
-	err = libbpf_get_error(btf);
-	if (err)
+
+	if (secs.btf_base_data) {
+		dist_base_btf = btf_new(secs.btf_base_data->d_buf, secs.btf_base_data->d_size,
+					NULL);
+		if (IS_ERR(dist_base_btf)) {
+			err = PTR_ERR(dist_base_btf);
+			dist_base_btf = NULL;
+			goto done;
+		}
+	}
+
+	btf = btf_new(secs.btf_data->d_buf, secs.btf_data->d_size,
+		      dist_base_btf ?: base_btf);
+	if (IS_ERR(btf)) {
+		err = PTR_ERR(btf);
 		goto done;
+	}
+	if (dist_base_btf && base_btf) {
+		err = btf__relocate(btf, base_btf);
+		if (err)
+			goto done;
+		btf__free(dist_base_btf);
+		dist_base_btf = NULL;
+	}
+
+	if (dist_base_btf)
+		btf->owns_base = true;
 
 	switch (gelf_getclass(elf)) {
 	case ELFCLASS32:
@@ -1187,11 +1237,12 @@ static struct btf *btf_parse_elf(const char *path, struct btf *base_btf,
 		break;
 	}
 
-	if (btf_ext && btf_ext_data) {
-		*btf_ext = btf_ext__new(btf_ext_data->d_buf, btf_ext_data->d_size);
-		err = libbpf_get_error(*btf_ext);
-		if (err)
+	if (btf_ext && secs.btf_ext_data) {
+		*btf_ext = btf_ext__new(secs.btf_ext_data->d_buf, secs.btf_ext_data->d_size);
+		if (IS_ERR(*btf_ext)) {
+			err = PTR_ERR(*btf_ext);
 			goto done;
+		}
 	} else if (btf_ext) {
 		*btf_ext = NULL;
 	}
@@ -1205,6 +1256,7 @@ done:
 
 	if (btf_ext)
 		btf_ext__free(*btf_ext);
+	btf__free(dist_base_btf);
 	btf__free(btf);
 
 	return ERR_PTR(err);
@@ -5598,5 +5650,9 @@ void btf_set_base_btf(struct btf *btf, const struct btf *base_btf)
 
 int btf__relocate(struct btf *btf, const struct btf *base_btf)
 {
-	return libbpf_err(btf_relocate(btf, base_btf, NULL));
+	int err = btf_relocate(btf, base_btf, NULL);
+
+	if (!err)
+		btf->owns_base = false;
+	return libbpf_err(err);
 }
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index 8a93120b7385..b68d216837a9 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -18,6 +18,7 @@ extern "C" {
 
 #define BTF_ELF_SEC ".BTF"
 #define BTF_EXT_ELF_SEC ".BTF.ext"
+#define BTF_BASE_ELF_SEC ".BTF.base"
 #define MAPS_ELF_SEC ".maps"
 
 struct btf;
-- 
cgit v1.2.3-58-ga151


From 6ba77385f386053cea2a1cad33717de74a26db4e Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 13 Jun 2024 10:50:11 +0100
Subject: resolve_btfids: Handle presence of .BTF.base section

Now that btf_parse_elf() handles .BTF.base section presence,
we need to ensure that resolve_btfids uses .BTF.base when present
rather than the vmlinux base BTF passed in via the -B option.
Detect .BTF.base section presence and unset the base BTF path
to ensure that BTF ELF parsing will do the right thing.

Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240613095014.357981-7-alan.maguire@oracle.com
---
 tools/bpf/resolve_btfids/main.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/tools/bpf/resolve_btfids/main.c b/tools/bpf/resolve_btfids/main.c
index af393c7dee1f..936ef95c3d32 100644
--- a/tools/bpf/resolve_btfids/main.c
+++ b/tools/bpf/resolve_btfids/main.c
@@ -409,6 +409,14 @@ static int elf_collect(struct object *obj)
 			obj->efile.idlist       = data;
 			obj->efile.idlist_shndx = idx;
 			obj->efile.idlist_addr  = sh.sh_addr;
+		} else if (!strcmp(name, BTF_BASE_ELF_SEC)) {
+			/* If a .BTF.base section is found, do not resolve
+			 * BTF ids relative to vmlinux; resolve relative
+			 * to the .BTF.base section instead.  btf__parse_split()
+			 * will take care of this once the base BTF it is
+			 * passed is NULL.
+			 */
+			obj->base_btf_path = NULL;
 		}
 
 		if (compressed_section_fix(elf, scn, &sh))
-- 
cgit v1.2.3-58-ga151


From dd89a81d850fa9a65f67b4527c0e420d15bf836c Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Fri, 14 Jun 2024 08:25:18 -0400
Subject: fou: remove warn in gue_gro_receive on unsupported protocol

Drop the WARN_ON_ONCE inn gue_gro_receive if the encapsulated type is
not known or does not have a GRO handler.

Such a packet is easily constructed. Syzbot generates them and sets
off this warning.

Remove the warning as it is expected and not actionable.

The warning was previously reduced from WARN_ON to WARN_ON_ONCE in
commit 270136613bf7 ("fou: Do WARN_ON_ONCE in gue_gro_receive for bad
proto callbacks").

Signed-off-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240614122552.1649044-1-willemdebruijn.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/fou_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/fou_core.c b/net/ipv4/fou_core.c
index a8494f796dca..0abbc413e0fe 100644
--- a/net/ipv4/fou_core.c
+++ b/net/ipv4/fou_core.c
@@ -433,7 +433,7 @@ next_proto:
 
 	offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads;
 	ops = rcu_dereference(offloads[proto]);
-	if (WARN_ON_ONCE(!ops || !ops->callbacks.gro_receive))
+	if (!ops || !ops->callbacks.gro_receive)
 		goto out;
 
 	pp = call_gro_receive(ops->callbacks.gro_receive, head, skb);
-- 
cgit v1.2.3-58-ga151


From e275e19c918b502aa4534e56dc2e25b5095e9005 Mon Sep 17 00:00:00 2001
From: Shradha Gupta <shradhagupta@linux.microsoft.com>
Date: Fri, 14 Jun 2024 00:19:08 -0700
Subject: net: mana: Use mana_cleanup_port_context() for rxq cleanup

To cleanup rxqs in port context structures, instead of duplicating the
code, use existing function mana_cleanup_port_context() which does
the exact cleanup that's needed.

Signed-off-by: Shradha Gupta <shradhagupta@linux.microsoft.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Wei Liu <wei.liu@kernel.org>
Reviewed-by: Heng Qi <hengqi@linux.alibaba.com>
Link: https://lore.kernel.org/r/1718349548-28697-1-git-send-email-shradhagupta@linux.microsoft.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/microsoft/mana/mana_en.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index b89ad4afd66e..93e526e5dd16 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -2529,8 +2529,7 @@ static int mana_init_port(struct net_device *ndev)
 	return 0;
 
 reset_apc:
-	kfree(apc->rxqs);
-	apc->rxqs = NULL;
+	mana_cleanup_port_context(apc);
 	return err;
 }
 
@@ -2787,8 +2786,7 @@ static int mana_probe_port(struct mana_context *ac, int port_idx,
 free_indir:
 	mana_cleanup_indir_table(apc);
 reset_apc:
-	kfree(apc->rxqs);
-	apc->rxqs = NULL;
+	mana_cleanup_port_context(apc);
 free_net:
 	*ndev_storage = NULL;
 	netdev_err(ndev, "Failed to probe vPort %d: %d\n", port_idx, err);
-- 
cgit v1.2.3-58-ga151


From 0d9bb144276edfb53d61bd0b42788568b19bc718 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Sun, 16 Jun 2024 13:01:48 -0700
Subject: net: dwc-xlgmac: fix missing MODULE_DESCRIPTION() warning

With ARCH=hexagon, make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/ethernet/synopsys/dwc-xlgmac.o

With most other ARCH settings the MODULE_DESCRIPTION() is provided by
the macro invocation in dwc-xlgmac-pci.c. However, for hexagon, the
PCI bus is not enabled, and hence CONFIG_DWC_XLGMAC_PCI is not set.
As a result, dwc-xlgmac-pci.c is not compiled, and hence is not linked
into dwc-xlgmac.o.

To avoid this issue, relocate the MODULE_DESCRIPTION() and other
related macros from dwc-xlgmac-pci.c to dwc-xlgmac-common.c, since
that file already has an existing MODULE_LICENSE() and it is
unconditionally linked into dwc-xlgmac.o.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://lore.kernel.org/r/20240616-md-hexagon-drivers-net-ethernet-synopsys-v1-1-55852b60aef8@quicinc.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/synopsys/dwc-xlgmac-common.c | 7 +++++--
 drivers/net/ethernet/synopsys/dwc-xlgmac-pci.c    | 5 -----
 2 files changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/synopsys/dwc-xlgmac-common.c b/drivers/net/ethernet/synopsys/dwc-xlgmac-common.c
index f8e133604146..131786aa4d5b 100644
--- a/drivers/net/ethernet/synopsys/dwc-xlgmac-common.c
+++ b/drivers/net/ethernet/synopsys/dwc-xlgmac-common.c
@@ -21,8 +21,6 @@
 #include "dwc-xlgmac.h"
 #include "dwc-xlgmac-reg.h"
 
-MODULE_LICENSE("Dual BSD/GPL");
-
 static int debug = -1;
 module_param(debug, int, 0644);
 MODULE_PARM_DESC(debug, "DWC ethernet debug level (0=none,...,16=all)");
@@ -725,3 +723,8 @@ void xlgmac_print_all_hw_features(struct xlgmac_pdata *pdata)
 	XLGMAC_PR("=====================================================\n");
 	XLGMAC_PR("\n");
 }
+
+MODULE_DESCRIPTION(XLGMAC_DRV_DESC);
+MODULE_VERSION(XLGMAC_DRV_VERSION);
+MODULE_AUTHOR("Jie Deng <jiedeng@synopsys.com>");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/synopsys/dwc-xlgmac-pci.c b/drivers/net/ethernet/synopsys/dwc-xlgmac-pci.c
index fa8604d7b797..36fe538e3332 100644
--- a/drivers/net/ethernet/synopsys/dwc-xlgmac-pci.c
+++ b/drivers/net/ethernet/synopsys/dwc-xlgmac-pci.c
@@ -71,8 +71,3 @@ static struct pci_driver xlgmac_pci_driver = {
 };
 
 module_pci_driver(xlgmac_pci_driver);
-
-MODULE_DESCRIPTION(XLGMAC_DRV_DESC);
-MODULE_VERSION(XLGMAC_DRV_VERSION);
-MODULE_AUTHOR("Jie Deng <jiedeng@synopsys.com>");
-MODULE_LICENSE("Dual BSD/GPL");
-- 
cgit v1.2.3-58-ga151


From efb459303dd5dd6e198a0d58322dc04c3356dc23 Mon Sep 17 00:00:00 2001
From: Kory Maincent <kory.maincent@bootlin.com>
Date: Wed, 12 Jun 2024 17:04:02 +0200
Subject: net: Move dev_set_hwtstamp_phylib to net/core/dev.h

This declaration was added to the header to be called from ethtool.
ethtool is separated from core for code organization but it is not really
a separate entity, it controls very core things.
As ethtool is an internal stuff it is not wise to have it in netdevice.h.
Move the declaration to net/core/dev.h instead.

Remove the EXPORT_SYMBOL_GPL call as ethtool can not be built as a module.

Reviewed-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://lore.kernel.org/r/20240612-feature_ptp_netnext-v15-2-b2a086257b63@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/netdevice.h | 3 ---
 net/core/dev.h            | 4 ++++
 net/core/dev_ioctl.c      | 1 -
 3 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 85111502cf8f..c83b390191d4 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3904,9 +3904,6 @@ int generic_hwtstamp_get_lower(struct net_device *dev,
 int generic_hwtstamp_set_lower(struct net_device *dev,
 			       struct kernel_hwtstamp_config *kernel_cfg,
 			       struct netlink_ext_ack *extack);
-int dev_set_hwtstamp_phylib(struct net_device *dev,
-			    struct kernel_hwtstamp_config *cfg,
-			    struct netlink_ext_ack *extack);
 int dev_ethtool(struct net *net, struct ifreq *ifr, void __user *userdata);
 unsigned int dev_get_flags(const struct net_device *);
 int __dev_change_flags(struct net_device *dev, unsigned int flags,
diff --git a/net/core/dev.h b/net/core/dev.h
index b7b518bc2be5..58f88d28bc99 100644
--- a/net/core/dev.h
+++ b/net/core/dev.h
@@ -166,4 +166,8 @@ static inline void dev_xmit_recursion_dec(void)
 	__this_cpu_dec(softnet_data.xmit.recursion);
 }
 
+int dev_set_hwtstamp_phylib(struct net_device *dev,
+			    struct kernel_hwtstamp_config *cfg,
+			    struct netlink_ext_ack *extack);
+
 #endif
diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index 9a66cf5015f2..b9719ed3c3fd 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -363,7 +363,6 @@ int dev_set_hwtstamp_phylib(struct net_device *dev,
 
 	return 0;
 }
-EXPORT_SYMBOL_GPL(dev_set_hwtstamp_phylib);
 
 static int dev_set_hwtstamp(struct net_device *dev, struct ifreq *ifr)
 {
-- 
cgit v1.2.3-58-ga151


From 64ff013ce0986952fad1678d7faf8998584fc500 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 28 May 2024 10:17:42 +0100
Subject: wifi: wlcore: correctness fix fwlog reading

Fix the calculation of clear_offset, which may overflow the end of
the buffer. However, this is harmless if it does because in that case
it will be recalculated when we copy the chunk of messages at the
start of the buffer.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/E1sBsxi-00E8vQ-5r@rmk-PC.armlinux.org.uk
---
 drivers/net/wireless/ti/wlcore/event.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ti/wlcore/event.c b/drivers/net/wireless/ti/wlcore/event.c
index 2499dc908305..6c3a8ea9613e 100644
--- a/drivers/net/wireless/ti/wlcore/event.c
+++ b/drivers/net/wireless/ti/wlcore/event.c
@@ -83,7 +83,7 @@ int wlcore_event_fw_logger(struct wl1271 *wl)
 	/* Copy initial part up to the end of ring buffer */
 	len = min(actual_len, available_len);
 	wl12xx_copy_fwlog(wl, &buffer[start_loc], len);
-	clear_ptr = addr_ptr + start_loc + actual_len;
+	clear_ptr = addr_ptr + start_loc + len;
 	if (clear_ptr == buff_end_ptr)
 		clear_ptr = buff_start_ptr;
 
-- 
cgit v1.2.3-58-ga151


From b734d8830f70813e17f470daa60f535a6df4d570 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 28 May 2024 10:17:47 +0100
Subject: wifi: wl18xx: make wl18xx_tx_immediate_complete() more efficient

wl18xx_tx_immediate_complete() iterates through the completed transmit
descriptors in a circular fashion, and in doing so uses a modulus
operation that is not a power of two. This leads to inefficient code
generation, which can be easily solved by providing a helper to
increment to the next descriptor. Use this more efficient solution.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/E1sBsxn-00E8vW-9h@rmk-PC.armlinux.org.uk
---
 drivers/net/wireless/ti/wl18xx/tx.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ti/wl18xx/tx.c b/drivers/net/wireless/ti/wl18xx/tx.c
index 55d9b0861c53..beef393853ef 100644
--- a/drivers/net/wireless/ti/wl18xx/tx.c
+++ b/drivers/net/wireless/ti/wl18xx/tx.c
@@ -129,6 +129,14 @@ static void wl18xx_tx_complete_packet(struct wl1271 *wl, u8 tx_stat_byte)
 	wl1271_free_tx_id(wl, id);
 }
 
+static u8 wl18xx_next_tx_idx(u8 idx)
+{
+	if (++idx >= WL18XX_FW_MAX_TX_STATUS_DESC)
+		idx = 0;
+
+	return idx;
+}
+
 void wl18xx_tx_immediate_complete(struct wl1271 *wl)
 {
 	struct wl18xx_fw_status_priv *status_priv =
@@ -161,9 +169,8 @@ void wl18xx_tx_immediate_complete(struct wl1271 *wl)
 		return;
 	}
 
-	for (i = priv->last_fw_rls_idx;
-	     i != status_priv->fw_release_idx;
-	     i = (i + 1) % WL18XX_FW_MAX_TX_STATUS_DESC) {
+	for (i = priv->last_fw_rls_idx; i != status_priv->fw_release_idx;
+	     i = wl18xx_next_tx_idx(i)) {
 		wl18xx_tx_complete_packet(wl,
 			status_priv->released_tx_desc[i]);
 
-- 
cgit v1.2.3-58-ga151


From 97715e29cebc898a214558475331ea6a10c0f8d1 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 28 May 2024 10:17:52 +0100
Subject: wifi: wlcore: improve code in wlcore_fw_status()

Referring to status->counters.tx_lnk_free_pkts[i] multiple times leads
to less efficient code. Cache this value in a local variable. This
also makes the code clearer.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/E1sBsxs-00E8vc-DD@rmk-PC.armlinux.org.uk
---
 drivers/net/wireless/ti/wlcore/main.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c
index ef12169f8044..a98b26dc3cb8 100644
--- a/drivers/net/wireless/ti/wlcore/main.c
+++ b/drivers/net/wireless/ti/wlcore/main.c
@@ -412,18 +412,18 @@ static int wlcore_fw_status(struct wl1271 *wl, struct wl_fw_status *status)
 
 
 	for_each_set_bit(i, wl->links_map, wl->num_links) {
-		u8 diff;
+		u8 diff, tx_lnk_free_pkts;
 		lnk = &wl->links[i];
 
 		/* prevent wrap-around in freed-packets counter */
-		diff = (status->counters.tx_lnk_free_pkts[i] -
-		       lnk->prev_freed_pkts) & 0xff;
+		tx_lnk_free_pkts = status->counters.tx_lnk_free_pkts[i];
+		diff = (tx_lnk_free_pkts - lnk->prev_freed_pkts) & 0xff;
 
 		if (diff == 0)
 			continue;
 
 		lnk->allocated_pkts -= diff;
-		lnk->prev_freed_pkts = status->counters.tx_lnk_free_pkts[i];
+		lnk->prev_freed_pkts = tx_lnk_free_pkts;
 
 		/* accumulate the prev_freed_pkts counter */
 		lnk->total_freed_pkts += diff;
-- 
cgit v1.2.3-58-ga151


From dd265a7415f8b99fa9b4b055f1ad1015c1e6f250 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 28 May 2024 10:17:57 +0100
Subject: wifi: wlcore: pass "status" to wlcore_hw_convert_fw_status()

wlcore_fw_status() is passed a pointer to the struct wl_fw_status to
decode the status into, which is always wl->fw_status. Rather than
referencing wl->fw_status within wlcore_fw_status(), use the supplied
argument so that we access this member in a consistent manner.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/E1sBsxx-00E8vi-Gf@rmk-PC.armlinux.org.uk
---
 drivers/net/wireless/ti/wlcore/main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c
index a98b26dc3cb8..3defe49c5120 100644
--- a/drivers/net/wireless/ti/wlcore/main.c
+++ b/drivers/net/wireless/ti/wlcore/main.c
@@ -392,7 +392,7 @@ static int wlcore_fw_status(struct wl1271 *wl, struct wl_fw_status *status)
 	if (ret < 0)
 		return ret;
 
-	wlcore_hw_convert_fw_status(wl, wl->raw_fw_status, wl->fw_status);
+	wlcore_hw_convert_fw_status(wl, wl->raw_fw_status, status);
 
 	wl1271_debug(DEBUG_IRQ, "intr: 0x%x (fw_rx_counter = %d, "
 		     "drv_rx_counter = %d, tx_results_counter = %d)",
-- 
cgit v1.2.3-58-ga151


From 81271c2bc59ef9fd2b90c7f3e61115309bb2a44c Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 28 May 2024 10:18:02 +0100
Subject: wifi: wlcore: store AP encryption key type

Updates for WL18xx firmware 8.9.1.x.x need to know the AP encryption
key type. Store this when a new key is set. Patch extracted from:

https://git.ti.com/cgit/wilink8-wlan/build-utilites/tree/patches/kernel_patches/4.19.38/0023-wlcore-Fixing-PN-drift-on-encrypted-link-after-recov.patch?h=r8.9&id=a2ee50aa5190ed3b334373d6cd09b1bff56ffcf7

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/E1sBsy2-00E8vo-KK@rmk-PC.armlinux.org.uk
---
 drivers/net/wireless/ti/wlcore/main.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c
index 3defe49c5120..8f82666f379c 100644
--- a/drivers/net/wireless/ti/wlcore/main.c
+++ b/drivers/net/wireless/ti/wlcore/main.c
@@ -3537,6 +3537,10 @@ int wlcore_set_key(struct wl1271 *wl, enum set_key_cmd cmd,
 			return ret;
 		}
 
+		/* Store AP encryption key type */
+		if (wlvif->bss_type == BSS_TYPE_AP_BSS)
+			wlvif->encryption_type = key_type;
+
 		/*
 		 * reconfiguring arp response if the unicast (or common)
 		 * encryption key type was changed
-- 
cgit v1.2.3-58-ga151


From bb8edd900fd69e2d9865e29e4682e5289e8a2345 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 28 May 2024 10:18:07 +0100
Subject: wifi: wlcore: add pn16 support

TI Wl18xx firmware adds a "pn16" field for AES and TKIP keys as per
their patch:

https://git.ti.com/cgit/wilink8-wlan/build-utilites/tree/patches/kernel_patches/4.19.38/0023-wlcore-Fixing-PN-drift-on-encrypted-link-after-recov.patch?h=r8.9&id=a2ee50aa5190ed3b334373d6cd09b1bff56ffcf7

Add support for this, but rather than requiring the field to be
present (which would break existing firmwares), make it optional.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/E1sBsy7-00E8vu-Nc@rmk-PC.armlinux.org.uk
---
 drivers/net/wireless/ti/wlcore/cmd.c      |  9 ++++
 drivers/net/wireless/ti/wlcore/main.c     | 89 +++++++++++++++++++++++++++++--
 drivers/net/wireless/ti/wlcore/wlcore_i.h |  4 ++
 3 files changed, 97 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/ti/wlcore/cmd.c b/drivers/net/wireless/ti/wlcore/cmd.c
index a939fd89a7f5..0d1fcdca3869 100644
--- a/drivers/net/wireless/ti/wlcore/cmd.c
+++ b/drivers/net/wireless/ti/wlcore/cmd.c
@@ -332,6 +332,14 @@ int wl12xx_allocate_link(struct wl1271 *wl, struct wl12xx_vif *wlvif, u8 *hlid)
 			wl->fw_status->counters.tx_lnk_free_pkts[link];
 	wl->links[link].wlvif = wlvif;
 
+	/*
+	 * Take the last sec_pn16 value from the current FW status. On recovery,
+	 * we might not have fw_status yet, and tx_lnk_sec_pn16[] will be NULL.
+	 */
+	if (wl->fw_status->counters.tx_lnk_sec_pn16)
+		wl->links[link].prev_sec_pn16 =
+			le16_to_cpu(wl->fw_status->counters.tx_lnk_sec_pn16[link]);
+
 	/*
 	 * Take saved value for total freed packets from wlvif, in case this is
 	 * recovery/resume
@@ -360,6 +368,7 @@ void wl12xx_free_link(struct wl1271 *wl, struct wl12xx_vif *wlvif, u8 *hlid)
 
 	wl->links[*hlid].allocated_pkts = 0;
 	wl->links[*hlid].prev_freed_pkts = 0;
+	wl->links[*hlid].prev_sec_pn16 = 0;
 	wl->links[*hlid].ba_bitmap = 0;
 	eth_zero_addr(wl->links[*hlid].addr);
 
diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c
index 8f82666f379c..35d1114a28aa 100644
--- a/drivers/net/wireless/ti/wlcore/main.c
+++ b/drivers/net/wireless/ti/wlcore/main.c
@@ -379,6 +379,8 @@ static void wl12xx_irq_update_links_status(struct wl1271 *wl,
 
 static int wlcore_fw_status(struct wl1271 *wl, struct wl_fw_status *status)
 {
+	struct wl12xx_vif *wlvifsta;
+	struct wl12xx_vif *wlvifap;
 	struct wl12xx_vif *wlvif;
 	u32 old_tx_blk_count = wl->tx_blocks_available;
 	int avail, freed_blocks;
@@ -410,23 +412,100 @@ static int wlcore_fw_status(struct wl1271 *wl, struct wl_fw_status *status)
 		wl->tx_pkts_freed[i] = status->counters.tx_released_pkts[i];
 	}
 
+	/* Find an authorized STA vif */
+	wlvifsta = NULL;
+	wl12xx_for_each_wlvif_sta(wl, wlvif) {
+		if (wlvif->sta.hlid != WL12XX_INVALID_LINK_ID &&
+		    test_bit(WLVIF_FLAG_STA_AUTHORIZED, &wlvif->flags)) {
+			wlvifsta = wlvif;
+			break;
+		}
+	}
+
+	/* Find a started AP vif */
+	wlvifap = NULL;
+	wl12xx_for_each_wlvif(wl, wlvif) {
+		if (wlvif->bss_type == BSS_TYPE_AP_BSS &&
+		    wlvif->inconn_count == 0 &&
+		    test_bit(WLVIF_FLAG_AP_STARTED, &wlvif->flags)) {
+			wlvifap = wlvif;
+			break;
+		}
+	}
 
 	for_each_set_bit(i, wl->links_map, wl->num_links) {
+		u16 diff16, sec_pn16;
 		u8 diff, tx_lnk_free_pkts;
+
 		lnk = &wl->links[i];
 
 		/* prevent wrap-around in freed-packets counter */
 		tx_lnk_free_pkts = status->counters.tx_lnk_free_pkts[i];
 		diff = (tx_lnk_free_pkts - lnk->prev_freed_pkts) & 0xff;
 
-		if (diff == 0)
+		if (diff) {
+			lnk->allocated_pkts -= diff;
+			lnk->prev_freed_pkts = tx_lnk_free_pkts;
+		}
+
+		/* Get the current sec_pn16 value if present */
+		if (status->counters.tx_lnk_sec_pn16)
+			sec_pn16 = __le16_to_cpu(status->counters.tx_lnk_sec_pn16[i]);
+		else
+			sec_pn16 = 0;
+		/* prevent wrap-around in pn16 counter */
+		diff16 = (sec_pn16 - lnk->prev_sec_pn16) & 0xffff;
+
+		/* FIXME: since free_pkts is a 8-bit counter of packets that
+		 * rolls over, it can become zero. If it is zero, then we
+		 * omit processing below. Is that really correct?
+		 */
+		if (tx_lnk_free_pkts <= 0)
 			continue;
 
-		lnk->allocated_pkts -= diff;
-		lnk->prev_freed_pkts = tx_lnk_free_pkts;
+		/* For a station that has an authorized link: */
+		if (wlvifsta && wlvifsta->sta.hlid == i) {
+			if (wlvifsta->encryption_type == KEY_TKIP ||
+			    wlvifsta->encryption_type == KEY_AES) {
+				if (diff16) {
+					lnk->prev_sec_pn16 = sec_pn16;
+					/* accumulate the prev_freed_pkts
+					 * counter according to the PN from
+					 * firmware
+					 */
+					lnk->total_freed_pkts += diff16;
+				}
+			} else {
+				if (diff)
+					/* accumulate the prev_freed_pkts
+					 * counter according to the free packets
+					 * count from firmware
+					 */
+					lnk->total_freed_pkts += diff;
+			}
+		}
 
-		/* accumulate the prev_freed_pkts counter */
-		lnk->total_freed_pkts += diff;
+		/* For an AP that has been started */
+		if (wlvifap && test_bit(i, wlvifap->ap.sta_hlid_map)) {
+			if (wlvifap->encryption_type == KEY_TKIP ||
+			    wlvifap->encryption_type == KEY_AES) {
+				if (diff16) {
+					lnk->prev_sec_pn16 = sec_pn16;
+					/* accumulate the prev_freed_pkts
+					 * counter according to the PN from
+					 * firmware
+					 */
+					lnk->total_freed_pkts += diff16;
+				}
+			} else {
+				if (diff)
+					/* accumulate the prev_freed_pkts
+					 * counter according to the free packets
+					 * count from firmware
+					 */
+					lnk->total_freed_pkts += diff;
+			}
+		}
 	}
 
 	/* prevent wrap-around in total blocks counter */
diff --git a/drivers/net/wireless/ti/wlcore/wlcore_i.h b/drivers/net/wireless/ti/wlcore/wlcore_i.h
index eefae3f867b9..5fbed64302f1 100644
--- a/drivers/net/wireless/ti/wlcore/wlcore_i.h
+++ b/drivers/net/wireless/ti/wlcore/wlcore_i.h
@@ -151,6 +151,9 @@ struct wl_fw_status {
 		 */
 		u8 *tx_lnk_free_pkts;
 
+		/* PN16 of last TKIP/AES seq-num per HLID */
+		__le16 *tx_lnk_sec_pn16;
+
 		/* Cumulative counter of released Voice memory blocks */
 		u8 tx_voice_released_blks;
 
@@ -259,6 +262,7 @@ struct wl1271_link {
 	/* accounting for allocated / freed packets in FW */
 	u8 allocated_pkts;
 	u8 prev_freed_pkts;
+	u16 prev_sec_pn16;
 
 	u8 addr[ETH_ALEN];
 
-- 
cgit v1.2.3-58-ga151


From 9685262b5e5db162622648b5c6cba53e9de3b75f Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 28 May 2024 10:18:12 +0100
Subject: wifi: wl18xx: add support for reading 8.9.1 fw status

Add the necessary code to read the 8.9.1 firmware status into the
driver private status structure, augmenting the 8.9.0 firmware
status code. New structure layout taken from:

https://git.ti.com/cgit/wilink8-wlan/build-utilites/tree/patches/kernel_patches/4.19.38/0023-wlcore-Fixing-PN-drift-on-encrypted-link-after-recov.patch?h=r8.9&id=a2ee50aa5190ed3b334373d6cd09b1bff56ffcf7

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/E1sBsyC-00E8w0-Rz@rmk-PC.armlinux.org.uk
---
 drivers/net/wireless/ti/wl18xx/main.c   | 71 ++++++++++++++++++++++++++++++++-
 drivers/net/wireless/ti/wl18xx/wl18xx.h | 60 ++++++++++++++++++++++++++++
 2 files changed, 129 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ti/wl18xx/main.c b/drivers/net/wireless/ti/wl18xx/main.c
index 2ccac1cdec01..39d8eebb9b6e 100644
--- a/drivers/net/wireless/ti/wl18xx/main.c
+++ b/drivers/net/wireless/ti/wl18xx/main.c
@@ -1177,8 +1177,49 @@ static int wl18xx_hw_init(struct wl1271 *wl)
 	return ret;
 }
 
-static void wl18xx_convert_fw_status(struct wl1271 *wl, void *raw_fw_status,
-				     struct wl_fw_status *fw_status)
+static void wl18xx_convert_fw_status_8_9_1(struct wl1271 *wl,
+					   void *raw_fw_status,
+					   struct wl_fw_status *fw_status)
+{
+	struct wl18xx_fw_status_8_9_1 *int_fw_status = raw_fw_status;
+
+	fw_status->intr = le32_to_cpu(int_fw_status->intr);
+	fw_status->fw_rx_counter = int_fw_status->fw_rx_counter;
+	fw_status->drv_rx_counter = int_fw_status->drv_rx_counter;
+	fw_status->tx_results_counter = int_fw_status->tx_results_counter;
+	fw_status->rx_pkt_descs = int_fw_status->rx_pkt_descs;
+
+	fw_status->fw_localtime = le32_to_cpu(int_fw_status->fw_localtime);
+	fw_status->link_ps_bitmap = le32_to_cpu(int_fw_status->link_ps_bitmap);
+	fw_status->link_fast_bitmap =
+			le32_to_cpu(int_fw_status->link_fast_bitmap);
+	fw_status->total_released_blks =
+			le32_to_cpu(int_fw_status->total_released_blks);
+	fw_status->tx_total = le32_to_cpu(int_fw_status->tx_total);
+
+	fw_status->counters.tx_released_pkts =
+			int_fw_status->counters.tx_released_pkts;
+	fw_status->counters.tx_lnk_free_pkts =
+			int_fw_status->counters.tx_lnk_free_pkts;
+	fw_status->counters.tx_lnk_sec_pn16 =
+			int_fw_status->counters.tx_lnk_sec_pn16;
+	fw_status->counters.tx_voice_released_blks =
+			int_fw_status->counters.tx_voice_released_blks;
+	fw_status->counters.tx_last_rate =
+			int_fw_status->counters.tx_last_rate;
+	fw_status->counters.tx_last_rate_mbps =
+			int_fw_status->counters.tx_last_rate_mbps;
+	fw_status->counters.hlid =
+			int_fw_status->counters.hlid;
+
+	fw_status->log_start_addr = le32_to_cpu(int_fw_status->log_start_addr);
+
+	fw_status->priv = &int_fw_status->priv;
+}
+
+static void wl18xx_convert_fw_status_8_9_0(struct wl1271 *wl,
+					   void *raw_fw_status,
+					   struct wl_fw_status *fw_status)
 {
 	struct wl18xx_fw_status *int_fw_status = raw_fw_status;
 
@@ -1214,6 +1255,15 @@ static void wl18xx_convert_fw_status(struct wl1271 *wl, void *raw_fw_status,
 	fw_status->priv = &int_fw_status->priv;
 }
 
+static void wl18xx_convert_fw_status(struct wl1271 *wl, void *raw_fw_status,
+				     struct wl_fw_status *fw_status)
+{
+	if (wl->chip.fw_ver[FW_VER_MAJOR] == 0)
+		wl18xx_convert_fw_status_8_9_0(wl, raw_fw_status, fw_status);
+	else
+		wl18xx_convert_fw_status_8_9_1(wl, raw_fw_status, fw_status);
+}
+
 static void wl18xx_set_tx_desc_csum(struct wl1271 *wl,
 				    struct wl1271_tx_hw_descr *desc,
 				    struct sk_buff *skb)
@@ -1515,12 +1565,29 @@ static int wl18xx_handle_static_data(struct wl1271 *wl,
 {
 	struct wl18xx_static_data_priv *static_data_priv =
 		(struct wl18xx_static_data_priv *) static_data->priv;
+	size_t fw_status_len;
 
 	strscpy(wl->chip.phy_fw_ver_str, static_data_priv->phy_version,
 		sizeof(wl->chip.phy_fw_ver_str));
 
 	wl1271_info("PHY firmware version: %s", static_data_priv->phy_version);
 
+	/* Adjust the firmware status size according to the firmware version */
+	if (wl->chip.fw_ver[FW_VER_MAJOR] == 0)
+		fw_status_len = sizeof(struct wl18xx_fw_status);
+	else
+		fw_status_len = sizeof(struct wl18xx_fw_status_8_9_1);
+
+	if (wl->fw_status_len != fw_status_len) {
+		void *new_status = krealloc(wl->raw_fw_status, fw_status_len,
+					    GFP_KERNEL | __GFP_ZERO);
+		if (!new_status)
+			return -ENOMEM;
+
+		wl->raw_fw_status = new_status;
+		wl->fw_status_len = fw_status_len;
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/ti/wl18xx/wl18xx.h b/drivers/net/wireless/ti/wl18xx/wl18xx.h
index b642e0c437bb..7fed96d71b27 100644
--- a/drivers/net/wireless/ti/wl18xx/wl18xx.h
+++ b/drivers/net/wireless/ti/wl18xx/wl18xx.h
@@ -155,6 +155,66 @@ struct wl18xx_fw_status {
 	struct wl18xx_fw_status_priv priv;
 } __packed;
 
+struct wl18xx_fw_packet_counters_8_9_1 {
+	/* Cumulative counter of released packets per AC */
+	u8 tx_released_pkts[NUM_TX_QUEUES];
+
+	/* Cumulative counter of freed packets per HLID */
+	u8 tx_lnk_free_pkts[WL18XX_MAX_LINKS];
+
+	/* PN16 of last TKIP/AES seq-num per HLID */
+	__le16 tx_lnk_sec_pn16[WL18XX_MAX_LINKS];
+
+	/* Cumulative counter of released Voice memory blocks */
+	u8 tx_voice_released_blks;
+
+	/* Tx rate of the last transmitted packet */
+	u8 tx_last_rate;
+
+	/* Tx rate or Tx rate estimate pre-calculated by fw in mbps units */
+	u8 tx_last_rate_mbps;
+
+	/* hlid for which the rates were reported */
+	u8 hlid;
+} __packed;
+
+/* FW status registers */
+struct wl18xx_fw_status_8_9_1 {
+	__le32 intr;
+	u8  fw_rx_counter;
+	u8  drv_rx_counter;
+	u8  reserved;
+	u8  tx_results_counter;
+	__le32 rx_pkt_descs[WL18XX_NUM_RX_DESCRIPTORS];
+
+	__le32 fw_localtime;
+
+	/*
+	 * A bitmap (where each bit represents a single HLID)
+	 * to indicate if the station is in PS mode.
+	 */
+	__le32 link_ps_bitmap;
+
+	/*
+	 * A bitmap (where each bit represents a single HLID) to indicate
+	 * if the station is in Fast mode
+	 */
+	__le32 link_fast_bitmap;
+
+	/* Cumulative counter of total released mem blocks since FW-reset */
+	__le32 total_released_blks;
+
+	/* Size (in Memory Blocks) of TX pool */
+	__le32 tx_total;
+
+	struct wl18xx_fw_packet_counters_8_9_1 counters;
+
+	__le32 log_start_addr;
+
+	/* Private status to be used by the lower drivers */
+	struct wl18xx_fw_status_priv priv;
+} __packed;
+
 #define WL18XX_PHY_VERSION_MAX_LEN 20
 
 struct wl18xx_static_data_priv {
-- 
cgit v1.2.3-58-ga151


From 8c58f972219e132d1277caf9a76119b444a50505 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Tue, 28 May 2024 10:18:17 +0100
Subject: wifi: wl18xx: allow firmwares > 8.9.0.x.58

wlcore firmware versions are structured thusly:

	chip.if-type.major.sub-type.minor
e.g.	  8    9       0       0     58

With WL18xx ignoring the major firmware version, looking for a
firmware version that conforms to:

	chip >= 8
	if-type >= 9
	major (don't care)
	sub-type (don't care)
	minor >= 58

Each test is satisfied if the value read from the firmware is greater
than the minimum, but if it is equal (or we don't care about the
field), then the next field is checked.

Thus it doesn't recognise 8.9.1.x.0 as being newer than 8.9.0.x.58
since the major and sub-type numbers are "don't care" and the minor
needs to be greater or equal to 58.

We need to change the major version from "ignore" to "0" for this later
firmware to be correctly detected, and allow the dual-firmware version
support to work.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/E1sBsyH-00E8w6-Vu@rmk-PC.armlinux.org.uk
---
 drivers/net/wireless/ti/wl18xx/wl18xx.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ti/wl18xx/wl18xx.h b/drivers/net/wireless/ti/wl18xx/wl18xx.h
index 7fed96d71b27..de6c671c4be6 100644
--- a/drivers/net/wireless/ti/wl18xx/wl18xx.h
+++ b/drivers/net/wireless/ti/wl18xx/wl18xx.h
@@ -13,7 +13,7 @@
 /* minimum FW required for driver */
 #define WL18XX_CHIP_VER		8
 #define WL18XX_IFTYPE_VER	9
-#define WL18XX_MAJOR_VER	WLCORE_FW_VER_IGNORE
+#define WL18XX_MAJOR_VER	0
 #define WL18XX_SUBTYPE_VER	WLCORE_FW_VER_IGNORE
 #define WL18XX_MINOR_VER	58
 
-- 
cgit v1.2.3-58-ga151


From 70ed0bdab1ca17aef6731e2c146b69050ff447f6 Mon Sep 17 00:00:00 2001
From: Ajay Singh <ajay.kathat@microchip.com>
Date: Thu, 13 Jun 2024 16:06:40 +0200
Subject: wifi: wilc1000: disable power sequencer
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Driver systematically disables some power mechanism each time it starts the
chip firmware (so mostly when interface is brought up). This has a negative
impact on some specific scenarios when the chip is exposed as a
hotpluggable SDIO card (eg: WILC1000 SD):
- when executing suspend/resume sequence while interface has been brought
  up
- rebooting the platform while module is plugged and interface has been
  brought up
Those scenarios lead to mmc core trying to initialize again the chip which
is now unresponsive (because of the power sequencer setting), so it fails
in mmc_rescan->mmc_attach_sdio and enter a failure loop while trying to
send CMD5:
mmc0: error -110 whilst initialising SDIO card
mmc0: error -110 whilst initialising SDIO card
mmc0: error -110 whilst initialising SDIO card
[...]

Preventing the driver from disabling this "power sequencer" fixes those
enumeration issues without affecting nominal operations.

Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240613-wilc_suspend-v1-1-c2f766d0988c@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/sdio.c | 15 ---------------
 drivers/net/wireless/microchip/wilc1000/wlan.h |  1 -
 2 files changed, 16 deletions(-)

diff --git a/drivers/net/wireless/microchip/wilc1000/sdio.c b/drivers/net/wireless/microchip/wilc1000/sdio.c
index e6e20c86b791..6461a3ec0120 100644
--- a/drivers/net/wireless/microchip/wilc1000/sdio.c
+++ b/drivers/net/wireless/microchip/wilc1000/sdio.c
@@ -907,27 +907,12 @@ static int wilc_sdio_sync_ext(struct wilc *wilc, int nint)
 {
 	struct sdio_func *func = dev_to_sdio_func(wilc->dev);
 	struct wilc_sdio *sdio_priv = wilc->bus_data;
-	u32 reg;
 
 	if (nint > MAX_NUM_INT) {
 		dev_err(&func->dev, "Too many interrupts (%d)...\n", nint);
 		return -EINVAL;
 	}
 
-	/**
-	 *      Disable power sequencer
-	 **/
-	if (wilc_sdio_read_reg(wilc, WILC_MISC, &reg)) {
-		dev_err(&func->dev, "Failed read misc reg...\n");
-		return -EINVAL;
-	}
-
-	reg &= ~BIT(8);
-	if (wilc_sdio_write_reg(wilc, WILC_MISC, reg)) {
-		dev_err(&func->dev, "Failed write misc reg...\n");
-		return -EINVAL;
-	}
-
 	if (sdio_priv->irq_gpio) {
 		u32 reg;
 		int ret, i;
diff --git a/drivers/net/wireless/microchip/wilc1000/wlan.h b/drivers/net/wireless/microchip/wilc1000/wlan.h
index d72a0a81bbda..dd2fb3c2f06a 100644
--- a/drivers/net/wireless/microchip/wilc1000/wlan.h
+++ b/drivers/net/wireless/microchip/wilc1000/wlan.h
@@ -56,7 +56,6 @@
 #define WILC_HOST_RX_CTRL		(WILC_PERIPH_REG_BASE + 0x80)
 #define WILC_HOST_RX_EXTRA_SIZE		(WILC_PERIPH_REG_BASE + 0x84)
 #define WILC_HOST_TX_CTRL_1		(WILC_PERIPH_REG_BASE + 0x88)
-#define WILC_MISC			(WILC_PERIPH_REG_BASE + 0x428)
 #define WILC_INTR_REG_BASE		(WILC_PERIPH_REG_BASE + 0xa00)
 #define WILC_INTR_ENABLE		WILC_INTR_REG_BASE
 #define WILC_INTR2_ENABLE		(WILC_INTR_REG_BASE + 4)
-- 
cgit v1.2.3-58-ga151


From 301e44e973ebb28d31a97441846dfa9fef6aa62b Mon Sep 17 00:00:00 2001
From: Alexis Lothoré <alexis.lothore@bootlin.com>
Date: Thu, 13 Jun 2024 16:06:41 +0200
Subject: wifi: wilc1000: let host->chip suspend/resume notifications manage
 chip wake/sleep
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

host_wakeup_notify and host_sleep_notify are surrounded by chip_wakeup and
chip_allow_sleep calls, which theorically need to be protected with the
hif_cs lock. This lock protection is currently missing. Instead of adding
the lock where those two functions are called, move those in host->chip
suspend notifications to benefit from the lock already used there (in
bus_acquire/bus_release)

Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240613-wilc_suspend-v1-2-c2f766d0988c@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/sdio.c | 5 -----
 drivers/net/wireless/microchip/wilc1000/wlan.c | 8 ++++----
 2 files changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/microchip/wilc1000/sdio.c b/drivers/net/wireless/microchip/wilc1000/sdio.c
index 6461a3ec0120..4e478a2cd4cc 100644
--- a/drivers/net/wireless/microchip/wilc1000/sdio.c
+++ b/drivers/net/wireless/microchip/wilc1000/sdio.c
@@ -256,14 +256,12 @@ static int wilc_sdio_suspend(struct device *dev)
 	int ret;
 
 	dev_info(dev, "sdio suspend\n");
-	chip_wakeup(wilc);
 
 	if (!IS_ERR(wilc->rtc_clk))
 		clk_disable_unprepare(wilc->rtc_clk);
 
 	if (wilc->suspend_event) {
 		host_sleep_notify(wilc);
-		chip_allow_sleep(wilc);
 	}
 
 	ret = wilc_sdio_reset(wilc);
@@ -1003,14 +1001,11 @@ static int wilc_sdio_resume(struct device *dev)
 
 	dev_info(dev, "sdio resume\n");
 	sdio_release_host(func);
-	chip_wakeup(wilc);
 	wilc_sdio_init(wilc, true);
 
 	if (wilc->suspend_event)
 		host_wakeup_notify(wilc);
 
-	chip_allow_sleep(wilc);
-
 	return 0;
 }
 
diff --git a/drivers/net/wireless/microchip/wilc1000/wlan.c b/drivers/net/wireless/microchip/wilc1000/wlan.c
index 41a5c0ed7480..533939e71534 100644
--- a/drivers/net/wireless/microchip/wilc1000/wlan.c
+++ b/drivers/net/wireless/microchip/wilc1000/wlan.c
@@ -678,17 +678,17 @@ EXPORT_SYMBOL_GPL(chip_wakeup);
 
 void host_wakeup_notify(struct wilc *wilc)
 {
-	acquire_bus(wilc, WILC_BUS_ACQUIRE_ONLY);
+	acquire_bus(wilc, WILC_BUS_ACQUIRE_AND_WAKEUP);
 	wilc->hif_func->hif_write_reg(wilc, WILC_CORTUS_INTERRUPT_2, 1);
-	release_bus(wilc, WILC_BUS_RELEASE_ONLY);
+	release_bus(wilc, WILC_BUS_RELEASE_ALLOW_SLEEP);
 }
 EXPORT_SYMBOL_GPL(host_wakeup_notify);
 
 void host_sleep_notify(struct wilc *wilc)
 {
-	acquire_bus(wilc, WILC_BUS_ACQUIRE_ONLY);
+	acquire_bus(wilc, WILC_BUS_ACQUIRE_AND_WAKEUP);
 	wilc->hif_func->hif_write_reg(wilc, WILC_CORTUS_INTERRUPT_1, 1);
-	release_bus(wilc, WILC_BUS_RELEASE_ONLY);
+	release_bus(wilc, WILC_BUS_RELEASE_ALLOW_SLEEP);
 }
 EXPORT_SYMBOL_GPL(host_sleep_notify);
 
-- 
cgit v1.2.3-58-ga151


From 25d086ba196418a16b6328440226065aeefc6076 Mon Sep 17 00:00:00 2001
From: Alexis Lothoré <alexis.lothore@bootlin.com>
Date: Thu, 13 Jun 2024 16:06:42 +0200
Subject: wifi: wilc1000: do not keep sdio bus claimed during suspend/resume
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There is no reason to keep the MMC host claimed during suspend.

Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240613-wilc_suspend-v1-3-c2f766d0988c@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/sdio.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/wireless/microchip/wilc1000/sdio.c b/drivers/net/wireless/microchip/wilc1000/sdio.c
index 4e478a2cd4cc..fb4de3de1140 100644
--- a/drivers/net/wireless/microchip/wilc1000/sdio.c
+++ b/drivers/net/wireless/microchip/wilc1000/sdio.c
@@ -269,7 +269,6 @@ static int wilc_sdio_suspend(struct device *dev)
 		dev_err(&func->dev, "Fail reset sdio\n");
 		return ret;
 	}
-	sdio_claim_host(func);
 
 	return 0;
 }
@@ -1000,7 +999,6 @@ static int wilc_sdio_resume(struct device *dev)
 	struct wilc *wilc = sdio_get_drvdata(func);
 
 	dev_info(dev, "sdio resume\n");
-	sdio_release_host(func);
 	wilc_sdio_init(wilc, true);
 
 	if (wilc->suspend_event)
-- 
cgit v1.2.3-58-ga151


From a814112e9545b873bcd82b2aafb6e88a82e6949f Mon Sep 17 00:00:00 2001
From: Alexis Lothoré <alexis.lothore@bootlin.com>
Date: Thu, 13 Jun 2024 16:06:43 +0200
Subject: wifi: wilc1000: move sdio suspend method next to resume and pm ops
 definition
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Just move the suspend method next to the resume method in the sdio part

Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240613-wilc_suspend-v1-4-c2f766d0988c@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/sdio.c | 47 +++++++++++++-------------
 1 file changed, 23 insertions(+), 24 deletions(-)

diff --git a/drivers/net/wireless/microchip/wilc1000/sdio.c b/drivers/net/wireless/microchip/wilc1000/sdio.c
index fb4de3de1140..21398a3c600b 100644
--- a/drivers/net/wireless/microchip/wilc1000/sdio.c
+++ b/drivers/net/wireless/microchip/wilc1000/sdio.c
@@ -249,30 +249,6 @@ static bool wilc_sdio_is_init(struct wilc *wilc)
 	return sdio_priv->isinit;
 }
 
-static int wilc_sdio_suspend(struct device *dev)
-{
-	struct sdio_func *func = dev_to_sdio_func(dev);
-	struct wilc *wilc = sdio_get_drvdata(func);
-	int ret;
-
-	dev_info(dev, "sdio suspend\n");
-
-	if (!IS_ERR(wilc->rtc_clk))
-		clk_disable_unprepare(wilc->rtc_clk);
-
-	if (wilc->suspend_event) {
-		host_sleep_notify(wilc);
-	}
-
-	ret = wilc_sdio_reset(wilc);
-	if (ret) {
-		dev_err(&func->dev, "Fail reset sdio\n");
-		return ret;
-	}
-
-	return 0;
-}
-
 static int wilc_sdio_enable_interrupt(struct wilc *dev)
 {
 	struct sdio_func *func = container_of(dev->dev, struct sdio_func, dev);
@@ -993,6 +969,29 @@ static const struct wilc_hif_func wilc_hif_sdio = {
 	.hif_is_init = wilc_sdio_is_init,
 };
 
+static int wilc_sdio_suspend(struct device *dev)
+{
+	struct sdio_func *func = dev_to_sdio_func(dev);
+	struct wilc *wilc = sdio_get_drvdata(func);
+	int ret;
+
+	dev_info(dev, "sdio suspend\n");
+
+	if (!IS_ERR(wilc->rtc_clk))
+		clk_disable_unprepare(wilc->rtc_clk);
+
+	if (wilc->suspend_event)
+		host_sleep_notify(wilc);
+
+	ret = wilc_sdio_reset(wilc);
+	if (ret) {
+		dev_err(&func->dev, "Fail reset sdio\n");
+		return ret;
+	}
+
+	return 0;
+}
+
 static int wilc_sdio_resume(struct device *dev)
 {
 	struct sdio_func *func = dev_to_sdio_func(dev);
-- 
cgit v1.2.3-58-ga151


From 1e9e258d9fb5c46dbf78f1baf06ac3836a44f2fa Mon Sep 17 00:00:00 2001
From: Alexis Lothoré <alexis.lothore@bootlin.com>
Date: Thu, 13 Jun 2024 16:06:44 +0200
Subject: wifi: wilc1000: remove suspend/resume from cfg80211 part
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

WILC1000 suspend/resume implementation is currently composed of two parts:
suspend/resume ops implemented in cfg80211 ops, which merely sets a
flag, and suspend/resume ops in sdio/spi driver which, based on this flag,
execute or not the suspend/resume mechanism. This dual set of ops is not
really needed , so keep only the sdio part to implement suspend/resume.

While doing so, remove the now unused suspend_event flag.

Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240613-wilc_suspend-v1-5-c2f766d0988c@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/cfg80211.c | 19 -------------------
 drivers/net/wireless/microchip/wilc1000/netdev.h   |  1 -
 drivers/net/wireless/microchip/wilc1000/sdio.c     |  6 ++----
 3 files changed, 2 insertions(+), 24 deletions(-)

diff --git a/drivers/net/wireless/microchip/wilc1000/cfg80211.c b/drivers/net/wireless/microchip/wilc1000/cfg80211.c
index ba02e6cfd3ae..eb37b228d54e 100644
--- a/drivers/net/wireless/microchip/wilc1000/cfg80211.c
+++ b/drivers/net/wireless/microchip/wilc1000/cfg80211.c
@@ -1617,23 +1617,6 @@ static int del_virtual_intf(struct wiphy *wiphy, struct wireless_dev *wdev)
 	return 0;
 }
 
-static int wilc_suspend(struct wiphy *wiphy, struct cfg80211_wowlan *wow)
-{
-	struct wilc *wl = wiphy_priv(wiphy);
-
-	if (!wow && wilc_wlan_get_num_conn_ifcs(wl))
-		wl->suspend_event = true;
-	else
-		wl->suspend_event = false;
-
-	return 0;
-}
-
-static int wilc_resume(struct wiphy *wiphy)
-{
-	return 0;
-}
-
 static void wilc_set_wakeup(struct wiphy *wiphy, bool enabled)
 {
 	struct wilc *wl = wiphy_priv(wiphy);
@@ -1739,8 +1722,6 @@ static const struct cfg80211_ops wilc_cfg80211_ops = {
 	.set_power_mgmt = set_power_mgmt,
 	.set_cqm_rssi_config = set_cqm_rssi_config,
 
-	.suspend = wilc_suspend,
-	.resume = wilc_resume,
 	.set_wakeup = wilc_set_wakeup,
 	.set_tx_power = set_tx_power,
 	.get_tx_power = get_tx_power,
diff --git a/drivers/net/wireless/microchip/wilc1000/netdev.h b/drivers/net/wireless/microchip/wilc1000/netdev.h
index e59d1b6f5d7d..95bc8b8fe65a 100644
--- a/drivers/net/wireless/microchip/wilc1000/netdev.h
+++ b/drivers/net/wireless/microchip/wilc1000/netdev.h
@@ -272,7 +272,6 @@ struct wilc {
 	const struct firmware *firmware;
 
 	struct device *dev;
-	bool suspend_event;
 
 	struct workqueue_struct *hif_workqueue;
 	struct wilc_cfg cfg;
diff --git a/drivers/net/wireless/microchip/wilc1000/sdio.c b/drivers/net/wireless/microchip/wilc1000/sdio.c
index 21398a3c600b..4e7014670945 100644
--- a/drivers/net/wireless/microchip/wilc1000/sdio.c
+++ b/drivers/net/wireless/microchip/wilc1000/sdio.c
@@ -980,8 +980,7 @@ static int wilc_sdio_suspend(struct device *dev)
 	if (!IS_ERR(wilc->rtc_clk))
 		clk_disable_unprepare(wilc->rtc_clk);
 
-	if (wilc->suspend_event)
-		host_sleep_notify(wilc);
+	host_sleep_notify(wilc);
 
 	ret = wilc_sdio_reset(wilc);
 	if (ret) {
@@ -1000,8 +999,7 @@ static int wilc_sdio_resume(struct device *dev)
 	dev_info(dev, "sdio resume\n");
 	wilc_sdio_init(wilc, true);
 
-	if (wilc->suspend_event)
-		host_wakeup_notify(wilc);
+	host_wakeup_notify(wilc);
 
 	return 0;
 }
-- 
cgit v1.2.3-58-ga151


From 5ad8897fd15a3716601509ed4a7bc47bf0cab661 Mon Sep 17 00:00:00 2001
From: Alexis Lothoré <alexis.lothore@bootlin.com>
Date: Thu, 13 Jun 2024 16:06:45 +0200
Subject: wifi: wilc1000: disable SDIO func IRQ before suspend
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Issuing a system suspend command raises the following warning:
WARNING: CPU: 0 PID: 15 at drivers/mmc/core/sdio.c:1042 mmc_sdio_suspend+0xd4/0x19c
CPU: 0 PID: 15 Comm: kworker/u2:1 Not tainted 6.7.0-rc1-wt+ #710
Hardware name: Atmel SAMA5
Workqueue: events_unbound async_run_entry_fn
 unwind_backtrace from show_stack+0x18/0x1c
 show_stack from dump_stack_lvl+0x34/0x48
 dump_stack_lvl from __warn+0x98/0x160
 __warn from warn_slowpath_fmt+0xcc/0x140
 warn_slowpath_fmt from mmc_sdio_suspend+0xd4/0x19c
 mmc_sdio_suspend from mmc_bus_suspend+0x50/0x70
 mmc_bus_suspend from dpm_run_callback+0xe4/0x248
 dpm_run_callback from __device_suspend+0x234/0x91c
 __device_suspend from async_suspend+0x24/0x9c
 async_suspend from async_run_entry_fn+0x6c/0x210
 async_run_entry_fn from process_one_work+0x3a0/0x870
[...]

This warning is due to a check in SDIO core ensuring that interrupts do not
remain enabled for cards being powered down during suspend. WILC driver
currently does not set the MMC_PM_KEEP_POWER flag, so disable interrupt
when entering resume.

Signed-off-by: Alexis Lothoré <alexis.lothore@bootlin.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://msgid.link/20240613-wilc_suspend-v1-6-c2f766d0988c@bootlin.com
---
 drivers/net/wireless/microchip/wilc1000/sdio.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/microchip/wilc1000/sdio.c b/drivers/net/wireless/microchip/wilc1000/sdio.c
index 4e7014670945..0043f7a0fdf9 100644
--- a/drivers/net/wireless/microchip/wilc1000/sdio.c
+++ b/drivers/net/wireless/microchip/wilc1000/sdio.c
@@ -982,6 +982,8 @@ static int wilc_sdio_suspend(struct device *dev)
 
 	host_sleep_notify(wilc);
 
+	wilc_sdio_disable_interrupt(wilc);
+
 	ret = wilc_sdio_reset(wilc);
 	if (ret) {
 		dev_err(&func->dev, "Fail reset sdio\n");
@@ -998,6 +1000,7 @@ static int wilc_sdio_resume(struct device *dev)
 
 	dev_info(dev, "sdio resume\n");
 	wilc_sdio_init(wilc, true);
+	wilc_sdio_enable_interrupt(wilc);
 
 	host_wakeup_notify(wilc);
 
-- 
cgit v1.2.3-58-ga151


From 041cc86b3653cbcdf6ab96c2f2ae34f3d0a99b0a Mon Sep 17 00:00:00 2001
From: Furong Xu <0x1207@gmail.com>
Date: Sat, 15 Jun 2024 17:56:11 +0800
Subject: net: stmmac: Enable TSO on VLANs

The TSO engine works well when the frames are not VLAN Tagged.
But it will produce broken segments when frames are VLAN Tagged.

The first segment is all good, while the second segment to the
last segment are broken, they lack of required VLAN tag.

An example here:
========
// 1st segment of a VLAN Tagged TSO frame, nothing wrong.
MacSrc > MacDst, ethertype 802.1Q (0x8100), length 1518: vlan 100, p 1, ethertype IPv4 (0x0800), HostA:42643 > HostB:5201: Flags [.], seq 1:1449

// 2nd to last segments of a VLAN Tagged TSO frame, VLAN tag is missing.
MacSrc > MacDst, ethertype IPv4 (0x0800), length 1514: HostA:42643 > HostB:5201: Flags [.], seq 1449:2897
MacSrc > MacDst, ethertype IPv4 (0x0800), length 1514: HostA:42643 > HostB:5201: Flags [.], seq 2897:4345
MacSrc > MacDst, ethertype IPv4 (0x0800), length 1514: HostA:42643 > HostB:5201: Flags [.], seq 4345:5793
MacSrc > MacDst, ethertype IPv4 (0x0800), length 1514: HostA:42643 > HostB:5201: Flags [P.], seq 5793:7241

// normal VLAN Tagged non-TSO frame, nothing wrong.
MacSrc > MacDst, ethertype 802.1Q (0x8100), length 1022: vlan 100, p 1, ethertype IPv4 (0x0800), HostA:42643 > HostB:5201: Flags [P.], seq 7241:8193
MacSrc > MacDst, ethertype 802.1Q (0x8100), length 70: vlan 100, p 1, ethertype IPv4 (0x0800), HostA:42643 > HostB:5201: Flags [F.], seq 8193
========

When transmitting VLAN Tagged TSO frames, never insert VLAN tag by HW,
always insert VLAN tag to SKB payload, then TSO works well on VLANs for
all MAC cores.

Tested on DWMAC CORE 5.10a, DWMAC CORE 5.20a and DWXGMAC CORE 3.20a

Signed-off-by: Furong Xu <0x1207@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240615095611.517323-1-0x1207@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 32 ++++++++++++++---------
 1 file changed, 19 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 5ddbb0d44373..83b654b7a9fd 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -4237,18 +4237,32 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct dma_desc *desc, *first, *mss_desc = NULL;
 	struct stmmac_priv *priv = netdev_priv(dev);
-	int nfrags = skb_shinfo(skb)->nr_frags;
-	u32 queue = skb_get_queue_mapping(skb);
+	int tmp_pay_len = 0, first_tx, nfrags;
 	unsigned int first_entry, tx_packets;
 	struct stmmac_txq_stats *txq_stats;
-	int tmp_pay_len = 0, first_tx;
 	struct stmmac_tx_queue *tx_q;
-	bool has_vlan, set_ic;
+	u32 pay_len, mss, queue;
 	u8 proto_hdr_len, hdr;
-	u32 pay_len, mss;
 	dma_addr_t des;
+	bool set_ic;
 	int i;
 
+	/* Always insert VLAN tag to SKB payload for TSO frames.
+	 *
+	 * Never insert VLAN tag by HW, since segments splited by
+	 * TSO engine will be un-tagged by mistake.
+	 */
+	if (skb_vlan_tag_present(skb)) {
+		skb = __vlan_hwaccel_push_inside(skb);
+		if (unlikely(!skb)) {
+			priv->xstats.tx_dropped++;
+			return NETDEV_TX_OK;
+		}
+	}
+
+	nfrags = skb_shinfo(skb)->nr_frags;
+	queue = skb_get_queue_mapping(skb);
+
 	tx_q = &priv->dma_conf.tx_queue[queue];
 	txq_stats = &priv->xstats.txq_stats[queue];
 	first_tx = tx_q->cur_tx;
@@ -4301,9 +4315,6 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
 			skb->data_len);
 	}
 
-	/* Check if VLAN can be inserted by HW */
-	has_vlan = stmmac_vlan_insert(priv, skb, tx_q);
-
 	first_entry = tx_q->cur_tx;
 	WARN_ON(tx_q->tx_skbuff[first_entry]);
 
@@ -4313,9 +4324,6 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
 		desc = &tx_q->dma_tx[first_entry];
 	first = desc;
 
-	if (has_vlan)
-		stmmac_set_desc_vlan(priv, first, STMMAC_VLAN_INSERT);
-
 	/* first descriptor: fill Headers on Buf1 */
 	des = dma_map_single(priv->device, skb->data, skb_headlen(skb),
 			     DMA_TO_DEVICE);
@@ -7682,8 +7690,6 @@ int stmmac_dvr_probe(struct device *device,
 		ndev->features |= NETIF_F_RXHASH;
 
 	ndev->vlan_features |= ndev->features;
-	/* TSO doesn't work on VLANs yet */
-	ndev->vlan_features &= ~NETIF_F_TSO;
 
 	/* MTU range: 46 - hw-specific max */
 	ndev->min_mtu = ETH_ZLEN - ETH_HLEN;
-- 
cgit v1.2.3-58-ga151


From 777b8afb8179155353ec14b1d8153122410aba29 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 15 Jun 2024 20:00:27 +0800
Subject: net: phy: introduce core support for phy-mode = "10g-qxgmii"

10G-QXGMII is a MAC-to-PHY interface defined by the USXGMII multiport
specification. It uses the same signaling as USXGMII, but it multiplexes
4 ports over the link, resulting in a maximum speed of 2.5G per port.

Some in-tree SoCs like the NXP LS1028A use "usxgmii" when they mean
either the single-port USXGMII or the quad-port 10G-QXGMII variant, and
they could get away just fine with that thus far. But there is a need to
distinguish between the 2 as far as SerDes drivers are concerned.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Luo Jie <quic_luoj@quicinc.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 Documentation/networking/phy.rst | 6 ++++++
 drivers/net/phy/phy-core.c       | 1 +
 drivers/net/phy/phylink.c        | 9 ++++++++-
 include/linux/phy.h              | 4 ++++
 include/linux/phylink.h          | 1 +
 5 files changed, 20 insertions(+), 1 deletion(-)

diff --git a/Documentation/networking/phy.rst b/Documentation/networking/phy.rst
index 1283240d7620..f64641417c54 100644
--- a/Documentation/networking/phy.rst
+++ b/Documentation/networking/phy.rst
@@ -327,6 +327,12 @@ Some of the interface modes are described below:
     This is the Penta SGMII mode, it is similar to QSGMII but it combines 5
     SGMII lines into a single link compared to 4 on QSGMII.
 
+``PHY_INTERFACE_MODE_10G_QXGMII``
+    Represents the 10G-QXGMII PHY-MAC interface as defined by the Cisco USXGMII
+    Multiport Copper Interface document. It supports 4 ports over a 10.3125 GHz
+    SerDes lane, each port having speeds of 2.5G / 1G / 100M / 10M achieved
+    through symbol replication. The PCS expects the standard USXGMII code word.
+
 Pause frames / flow control
 ===========================
 
diff --git a/drivers/net/phy/phy-core.c b/drivers/net/phy/phy-core.c
index 15f349e5995a..a235ea2264a7 100644
--- a/drivers/net/phy/phy-core.c
+++ b/drivers/net/phy/phy-core.c
@@ -141,6 +141,7 @@ int phy_interface_num_ports(phy_interface_t interface)
 		return 1;
 	case PHY_INTERFACE_MODE_QSGMII:
 	case PHY_INTERFACE_MODE_QUSGMII:
+	case PHY_INTERFACE_MODE_10G_QXGMII:
 		return 4;
 	case PHY_INTERFACE_MODE_PSGMII:
 		return 5;
diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 02427378acfd..6c24c48dcf0f 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -231,6 +231,7 @@ static int phylink_interface_max_speed(phy_interface_t interface)
 		return SPEED_1000;
 
 	case PHY_INTERFACE_MODE_2500BASEX:
+	case PHY_INTERFACE_MODE_10G_QXGMII:
 		return SPEED_2500;
 
 	case PHY_INTERFACE_MODE_5GBASER:
@@ -500,7 +501,11 @@ static unsigned long phylink_get_capabilities(phy_interface_t interface,
 
 	switch (interface) {
 	case PHY_INTERFACE_MODE_USXGMII:
-		caps |= MAC_10000FD | MAC_5000FD | MAC_2500FD;
+		caps |= MAC_10000FD | MAC_5000FD;
+		fallthrough;
+
+	case PHY_INTERFACE_MODE_10G_QXGMII:
+		caps |= MAC_2500FD;
 		fallthrough;
 
 	case PHY_INTERFACE_MODE_RGMII_TXID:
@@ -926,6 +931,7 @@ static int phylink_parse_mode(struct phylink *pl,
 		case PHY_INTERFACE_MODE_5GBASER:
 		case PHY_INTERFACE_MODE_25GBASER:
 		case PHY_INTERFACE_MODE_USXGMII:
+		case PHY_INTERFACE_MODE_10G_QXGMII:
 		case PHY_INTERFACE_MODE_10GKR:
 		case PHY_INTERFACE_MODE_10GBASER:
 		case PHY_INTERFACE_MODE_XLGMII:
@@ -1124,6 +1130,7 @@ static unsigned int phylink_pcs_neg_mode(unsigned int mode,
 	case PHY_INTERFACE_MODE_QSGMII:
 	case PHY_INTERFACE_MODE_QUSGMII:
 	case PHY_INTERFACE_MODE_USXGMII:
+	case PHY_INTERFACE_MODE_10G_QXGMII:
 		/* These protocols are designed for use with a PHY which
 		 * communicates its negotiation result back to the MAC via
 		 * inband communication. Note: there exist PHYs that run
diff --git a/include/linux/phy.h b/include/linux/phy.h
index e6e83304558e..205fccfc0f60 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -128,6 +128,7 @@ extern const int phy_10gbit_features_array[1];
  * @PHY_INTERFACE_MODE_10GKR: 10GBASE-KR - with Clause 73 AN
  * @PHY_INTERFACE_MODE_QUSGMII: Quad Universal SGMII
  * @PHY_INTERFACE_MODE_1000BASEKX: 1000Base-KX - with Clause 73 AN
+ * @PHY_INTERFACE_MODE_10G_QXGMII: 10G-QXGMII - 4 ports over 10G USXGMII
  * @PHY_INTERFACE_MODE_MAX: Book keeping
  *
  * Describes the interface between the MAC and PHY.
@@ -168,6 +169,7 @@ typedef enum {
 	PHY_INTERFACE_MODE_10GKR,
 	PHY_INTERFACE_MODE_QUSGMII,
 	PHY_INTERFACE_MODE_1000BASEKX,
+	PHY_INTERFACE_MODE_10G_QXGMII,
 	PHY_INTERFACE_MODE_MAX,
 } phy_interface_t;
 
@@ -289,6 +291,8 @@ static inline const char *phy_modes(phy_interface_t interface)
 		return "100base-x";
 	case PHY_INTERFACE_MODE_QUSGMII:
 		return "qusgmii";
+	case PHY_INTERFACE_MODE_10G_QXGMII:
+		return "10g-qxgmii";
 	default:
 		return "unknown";
 	}
diff --git a/include/linux/phylink.h b/include/linux/phylink.h
index a30a692acc32..2381e07429a2 100644
--- a/include/linux/phylink.h
+++ b/include/linux/phylink.h
@@ -654,6 +654,7 @@ static inline int phylink_get_link_timer_ns(phy_interface_t interface)
 	case PHY_INTERFACE_MODE_SGMII:
 	case PHY_INTERFACE_MODE_QSGMII:
 	case PHY_INTERFACE_MODE_USXGMII:
+	case PHY_INTERFACE_MODE_10G_QXGMII:
 		return 1600000;
 
 	case PHY_INTERFACE_MODE_1000BASEX:
-- 
cgit v1.2.3-58-ga151


From 5dfabcdd76b1455c5378abbcb2e68285cf7eb554 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 15 Jun 2024 20:00:28 +0800
Subject: dt-bindings: net: ethernet-controller: add 10g-qxgmii mode

Add the new interface mode 10g-qxgmii, which is similar to
usxgmii but extend to 4 channels to support maximum of 4
ports with the link speed 10M/100M/1G/2.5G.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Luo Jie <quic_luoj@quicinc.com>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 Documentation/devicetree/bindings/net/ethernet-controller.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/net/ethernet-controller.yaml b/Documentation/devicetree/bindings/net/ethernet-controller.yaml
index b2785b03139f..45819b235800 100644
--- a/Documentation/devicetree/bindings/net/ethernet-controller.yaml
+++ b/Documentation/devicetree/bindings/net/ethernet-controller.yaml
@@ -103,6 +103,7 @@ properties:
       - usxgmii
       - 10gbase-r
       - 25gbase-r
+      - 10g-qxgmii
 
   phy-mode:
     $ref: "#/properties/phy-connection-type"
-- 
cgit v1.2.3-58-ga151


From 8c379e3ce404347acf35fbb4cf8694dd8d35f2a0 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sun, 16 Jun 2024 09:57:56 +0200
Subject: net: microchip: Constify struct vcap_operations

"struct vcap_operations" are not modified in these drivers.

Constifying this structure moves some data to a read-only section, so
increase overall security.

In order to do it, "struct vcap_control" also needs to be adjusted to this
new const qualifier.

As an example, on a x86_64, with allmodconfig:
Before:
======
   text	   data	    bss	    dec	    hex	filename
  15176	   1094	     16	  16286	   3f9e	drivers/net/ethernet/microchip/lan966x/lan966x_vcap_impl.o

After:
=====
   text	   data	    bss	    dec	    hex	filename
  15268	    998	     16	  16282	   3f9a	drivers/net/ethernet/microchip/lan966x/lan966x_vcap_impl.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Daniel Machon <daniel.machon@microchip.com>
Link: https://lore.kernel.org/r/d8e76094d2e98ebb5bfc8205799b3a9db0b46220.1718524644.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/microchip/lan966x/lan966x_vcap_impl.c   | 2 +-
 drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.c     | 2 +-
 drivers/net/ethernet/microchip/vcap/vcap_api.h               | 2 +-
 drivers/net/ethernet/microchip/vcap/vcap_api_debugfs_kunit.c | 2 +-
 drivers/net/ethernet/microchip/vcap/vcap_api_kunit.c         | 2 +-
 5 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_impl.c b/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_impl.c
index a4414f63c9b1..a1471e38d118 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_impl.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_vcap_impl.c
@@ -581,7 +581,7 @@ static void lan966x_vcap_move(struct net_device *dev,
 	lan966x_vcap_wait_update(lan966x, admin->tgt_inst);
 }
 
-static struct vcap_operations lan966x_vcap_ops = {
+static const struct vcap_operations lan966x_vcap_ops = {
 	.validate_keyset = lan966x_vcap_validate_keyset,
 	.add_default_fields = lan966x_vcap_add_default_fields,
 	.cache_erase = lan966x_vcap_cache_erase,
diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.c b/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.c
index 187efa1fc904..967c8621c250 100644
--- a/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.c
+++ b/drivers/net/ethernet/microchip/sparx5/sparx5_vcap_impl.c
@@ -1507,7 +1507,7 @@ static void sparx5_vcap_move(struct net_device *ndev, struct vcap_admin *admin,
 	}
 }
 
-static struct vcap_operations sparx5_vcap_ops = {
+static const struct vcap_operations sparx5_vcap_ops = {
 	.validate_keyset = sparx5_vcap_validate_keyset,
 	.add_default_fields = sparx5_vcap_add_default_fields,
 	.cache_erase = sparx5_vcap_cache_erase,
diff --git a/drivers/net/ethernet/microchip/vcap/vcap_api.h b/drivers/net/ethernet/microchip/vcap/vcap_api.h
index 9eccfa633c1a..6069ad95c27e 100644
--- a/drivers/net/ethernet/microchip/vcap/vcap_api.h
+++ b/drivers/net/ethernet/microchip/vcap/vcap_api.h
@@ -271,7 +271,7 @@ struct vcap_operations {
 
 /* VCAP API Client control interface */
 struct vcap_control {
-	struct vcap_operations *ops;  /* client supplied operations */
+	const struct vcap_operations *ops;  /* client supplied operations */
 	const struct vcap_info *vcaps; /* client supplied vcap models */
 	const struct vcap_statistics *stats; /* client supplied vcap stats */
 	struct list_head list; /* list of vcap instances */
diff --git a/drivers/net/ethernet/microchip/vcap/vcap_api_debugfs_kunit.c b/drivers/net/ethernet/microchip/vcap/vcap_api_debugfs_kunit.c
index b23c11b0647c..9c9d38042125 100644
--- a/drivers/net/ethernet/microchip/vcap/vcap_api_debugfs_kunit.c
+++ b/drivers/net/ethernet/microchip/vcap/vcap_api_debugfs_kunit.c
@@ -221,7 +221,7 @@ static int vcap_test_port_info(struct net_device *ndev,
 	return 0;
 }
 
-static struct vcap_operations test_callbacks = {
+static const struct vcap_operations test_callbacks = {
 	.validate_keyset = test_val_keyset,
 	.add_default_fields = test_add_def_fields,
 	.cache_erase = test_cache_erase,
diff --git a/drivers/net/ethernet/microchip/vcap/vcap_api_kunit.c b/drivers/net/ethernet/microchip/vcap/vcap_api_kunit.c
index fe4e166de8a0..51d9423b08a6 100644
--- a/drivers/net/ethernet/microchip/vcap/vcap_api_kunit.c
+++ b/drivers/net/ethernet/microchip/vcap/vcap_api_kunit.c
@@ -211,7 +211,7 @@ static int vcap_test_port_info(struct net_device *ndev,
 	return 0;
 }
 
-static struct vcap_operations test_callbacks = {
+static const struct vcap_operations test_callbacks = {
 	.validate_keyset = test_val_keyset,
 	.add_default_fields = test_add_def_fields,
 	.cache_erase = test_cache_erase,
-- 
cgit v1.2.3-58-ga151


From e52e010395dc5232f7dd5c271367f4bdd18ab74c Mon Sep 17 00:00:00 2001
From: Kamal Heib <kheib@redhat.com>
Date: Mon, 17 Jun 2024 13:23:27 -0400
Subject: net/mlx4_en: Use ethtool_puts to fill priv flags strings

Use the ethtool_puts helper to print the priv flags strings into the
ethtool strings interface.

Signed-off-by: Kamal Heib <kheib@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240617172329.239819-2-kheib@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 619e1c3ef7f9..50a4a017a3f4 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -536,8 +536,7 @@ static void mlx4_en_get_strings(struct net_device *dev,
 		break;
 	case ETH_SS_PRIV_FLAGS:
 		for (i = 0; i < ARRAY_SIZE(mlx4_en_priv_flags); i++)
-			strcpy(data + i * ETH_GSTRING_LEN,
-			       mlx4_en_priv_flags[i]);
+			ethtool_puts(&data, mlx4_en_priv_flags[i]);
 		break;
 
 	}
-- 
cgit v1.2.3-58-ga151


From 4454929c345dbd657dd5a6ae816050d140d228c0 Mon Sep 17 00:00:00 2001
From: Kamal Heib <kheib@redhat.com>
Date: Mon, 17 Jun 2024 13:23:28 -0400
Subject: net/mlx4_en: Use ethtool_puts to fill selftest strings

Use the ethtool_puts helper to print the selftest strings into the
ethtool strings interface.

Signed-off-by: Kamal Heib <kheib@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240617172329.239819-3-kheib@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 50a4a017a3f4..fee02a94ed2f 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -459,10 +459,10 @@ static void mlx4_en_get_strings(struct net_device *dev,
 	switch (stringset) {
 	case ETH_SS_TEST:
 		for (i = 0; i < MLX4_EN_NUM_SELF_TEST - 2; i++)
-			strcpy(data + i * ETH_GSTRING_LEN, mlx4_en_test_names[i]);
+			ethtool_puts(&data, mlx4_en_test_names[i]);
 		if (priv->mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_UC_LOOPBACK)
 			for (; i < MLX4_EN_NUM_SELF_TEST; i++)
-				strcpy(data + i * ETH_GSTRING_LEN, mlx4_en_test_names[i]);
+				ethtool_puts(&data, mlx4_en_test_names[i]);
 		break;
 
 	case ETH_SS_STATS:
-- 
cgit v1.2.3-58-ga151


From 6c7dd432dcbccdad9474aa9d45ec0d52c609d937 Mon Sep 17 00:00:00 2001
From: Kamal Heib <kheib@redhat.com>
Date: Mon, 17 Jun 2024 13:23:29 -0400
Subject: net/mlx4_en: Use ethtool_puts/sprintf to fill stats strings

Use the ethtool_puts/ethtool_sprintf helper to print the stats strings
into the ethtool strings interface.

Signed-off-by: Kamal Heib <kheib@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240617172329.239819-4-kheib@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 52 ++++++++-----------------
 1 file changed, 17 insertions(+), 35 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index fee02a94ed2f..0606f18e5bbe 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -450,7 +450,6 @@ static void mlx4_en_get_strings(struct net_device *dev,
 				uint32_t stringset, uint8_t *data)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
-	int index = 0;
 	int i, strings = 0;
 	struct bitmap_iterator it;
 
@@ -470,68 +469,51 @@ static void mlx4_en_get_strings(struct net_device *dev,
 		for (i = 0; i < NUM_MAIN_STATS; i++, strings++,
 		     bitmap_iterator_inc(&it))
 			if (bitmap_iterator_test(&it))
-				strcpy(data + (index++) * ETH_GSTRING_LEN,
-				       main_strings[strings]);
+				ethtool_puts(&data, main_strings[strings]);
 
 		for (i = 0; i < NUM_PORT_STATS; i++, strings++,
 		     bitmap_iterator_inc(&it))
 			if (bitmap_iterator_test(&it))
-				strcpy(data + (index++) * ETH_GSTRING_LEN,
-				       main_strings[strings]);
+				ethtool_puts(&data, main_strings[strings]);
 
 		for (i = 0; i < NUM_PF_STATS; i++, strings++,
 		     bitmap_iterator_inc(&it))
 			if (bitmap_iterator_test(&it))
-				strcpy(data + (index++) * ETH_GSTRING_LEN,
-				       main_strings[strings]);
+				ethtool_puts(&data, main_strings[strings]);
 
 		for (i = 0; i < NUM_FLOW_STATS; i++, strings++,
 		     bitmap_iterator_inc(&it))
 			if (bitmap_iterator_test(&it))
-				strcpy(data + (index++) * ETH_GSTRING_LEN,
-				       main_strings[strings]);
+				ethtool_puts(&data, main_strings[strings]);
 
 		for (i = 0; i < NUM_PKT_STATS; i++, strings++,
 		     bitmap_iterator_inc(&it))
 			if (bitmap_iterator_test(&it))
-				strcpy(data + (index++) * ETH_GSTRING_LEN,
-				       main_strings[strings]);
+				ethtool_puts(&data, main_strings[strings]);
 
 		for (i = 0; i < NUM_XDP_STATS; i++, strings++,
 		     bitmap_iterator_inc(&it))
 			if (bitmap_iterator_test(&it))
-				strcpy(data + (index++) * ETH_GSTRING_LEN,
-				       main_strings[strings]);
+				ethtool_puts(&data, main_strings[strings]);
 
 		for (i = 0; i < NUM_PHY_STATS; i++, strings++,
 		     bitmap_iterator_inc(&it))
 			if (bitmap_iterator_test(&it))
-				strcpy(data + (index++) * ETH_GSTRING_LEN,
-				       main_strings[strings]);
+				ethtool_puts(&data, main_strings[strings]);
 
 		for (i = 0; i < priv->tx_ring_num[TX]; i++) {
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"tx%d_packets", i);
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"tx%d_bytes", i);
+			ethtool_sprintf(&data, "tx%d_packets", i);
+			ethtool_sprintf(&data, "tx%d_bytes", i);
 		}
 		for (i = 0; i < priv->rx_ring_num; i++) {
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"rx%d_packets", i);
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"rx%d_bytes", i);
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"rx%d_dropped", i);
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"rx%d_xdp_drop", i);
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"rx%d_xdp_redirect", i);
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"rx%d_xdp_redirect_fail", i);
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"rx%d_xdp_tx", i);
-			sprintf(data + (index++) * ETH_GSTRING_LEN,
-				"rx%d_xdp_tx_full", i);
+			ethtool_sprintf(&data, "rx%d_packets", i);
+			ethtool_sprintf(&data, "rx%d_bytes", i);
+			ethtool_sprintf(&data, "rx%d_dropped", i);
+			ethtool_sprintf(&data, "rx%d_xdp_drop", i);
+			ethtool_sprintf(&data, "rx%d_xdp_redirect", i);
+			ethtool_sprintf(&data, "rx%d_xdp_redirect_fail", i);
+			ethtool_sprintf(&data, "rx%d_xdp_tx", i);
+			ethtool_sprintf(&data, "rx%d_xdp_tx_full", i);
 		}
 		break;
 	case ETH_SS_PRIV_FLAGS:
-- 
cgit v1.2.3-58-ga151


From 382d1741b5b2feffef7942dd074206372afe1a96 Mon Sep 17 00:00:00 2001
From: Haiyang Zhang <haiyangz@microsoft.com>
Date: Mon, 17 Jun 2024 13:17:26 -0700
Subject: net: mana: Add support for page sizes other than 4KB on ARM64

As defined by the MANA Hardware spec, the queue size for DMA is 4KB
minimal, and power of 2. And, the HWC queue size has to be exactly
4KB.

To support page sizes other than 4KB on ARM64, define the minimal
queue size as a macro separately from the PAGE_SIZE, which we always
assumed it to be 4KB before supporting ARM64.

Also, add MANA specific macros and update code related to size
alignment, DMA region calculations, etc.

Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Michael Kelley <mhklinux@outlook.com>
Link: https://lore.kernel.org/r/1718655446-6576-1-git-send-email-haiyangz@microsoft.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/microsoft/Kconfig            |  2 +-
 drivers/net/ethernet/microsoft/mana/gdma_main.c   | 10 +++++-----
 drivers/net/ethernet/microsoft/mana/hw_channel.c  | 14 +++++++-------
 drivers/net/ethernet/microsoft/mana/mana_en.c     |  8 ++++----
 drivers/net/ethernet/microsoft/mana/shm_channel.c | 13 +++++++------
 include/net/mana/gdma.h                           | 10 +++++++++-
 include/net/mana/mana.h                           |  3 ++-
 7 files changed, 35 insertions(+), 25 deletions(-)

diff --git a/drivers/net/ethernet/microsoft/Kconfig b/drivers/net/ethernet/microsoft/Kconfig
index 286f0d5697a1..901fbffbf718 100644
--- a/drivers/net/ethernet/microsoft/Kconfig
+++ b/drivers/net/ethernet/microsoft/Kconfig
@@ -18,7 +18,7 @@ if NET_VENDOR_MICROSOFT
 config MICROSOFT_MANA
 	tristate "Microsoft Azure Network Adapter (MANA) support"
 	depends on PCI_MSI
-	depends on X86_64 || (ARM64 && !CPU_BIG_ENDIAN && ARM64_4K_PAGES)
+	depends on X86_64 || (ARM64 && !CPU_BIG_ENDIAN)
 	depends on PCI_HYPERV
 	select AUXILIARY_BUS
 	select PAGE_POOL
diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c b/drivers/net/ethernet/microsoft/mana/gdma_main.c
index 1332db9a08eb..e1d70d21e207 100644
--- a/drivers/net/ethernet/microsoft/mana/gdma_main.c
+++ b/drivers/net/ethernet/microsoft/mana/gdma_main.c
@@ -182,7 +182,7 @@ int mana_gd_alloc_memory(struct gdma_context *gc, unsigned int length,
 	dma_addr_t dma_handle;
 	void *buf;
 
-	if (length < PAGE_SIZE || !is_power_of_2(length))
+	if (length < MANA_PAGE_SIZE || !is_power_of_2(length))
 		return -EINVAL;
 
 	gmi->dev = gc->dev;
@@ -717,7 +717,7 @@ EXPORT_SYMBOL_NS(mana_gd_destroy_dma_region, NET_MANA);
 static int mana_gd_create_dma_region(struct gdma_dev *gd,
 				     struct gdma_mem_info *gmi)
 {
-	unsigned int num_page = gmi->length / PAGE_SIZE;
+	unsigned int num_page = gmi->length / MANA_PAGE_SIZE;
 	struct gdma_create_dma_region_req *req = NULL;
 	struct gdma_create_dma_region_resp resp = {};
 	struct gdma_context *gc = gd->gdma_context;
@@ -727,10 +727,10 @@ static int mana_gd_create_dma_region(struct gdma_dev *gd,
 	int err;
 	int i;
 
-	if (length < PAGE_SIZE || !is_power_of_2(length))
+	if (length < MANA_PAGE_SIZE || !is_power_of_2(length))
 		return -EINVAL;
 
-	if (offset_in_page(gmi->virt_addr) != 0)
+	if (!MANA_PAGE_ALIGNED(gmi->virt_addr))
 		return -EINVAL;
 
 	hwc = gc->hwc.driver_data;
@@ -751,7 +751,7 @@ static int mana_gd_create_dma_region(struct gdma_dev *gd,
 	req->page_addr_list_len = num_page;
 
 	for (i = 0; i < num_page; i++)
-		req->page_addr_list[i] = gmi->dma_handle +  i * PAGE_SIZE;
+		req->page_addr_list[i] = gmi->dma_handle +  i * MANA_PAGE_SIZE;
 
 	err = mana_gd_send_request(gc, req_msg_size, req, sizeof(resp), &resp);
 	if (err)
diff --git a/drivers/net/ethernet/microsoft/mana/hw_channel.c b/drivers/net/ethernet/microsoft/mana/hw_channel.c
index bbc4f9e16c98..cafded2f9382 100644
--- a/drivers/net/ethernet/microsoft/mana/hw_channel.c
+++ b/drivers/net/ethernet/microsoft/mana/hw_channel.c
@@ -362,12 +362,12 @@ static int mana_hwc_create_cq(struct hw_channel_context *hwc, u16 q_depth,
 	int err;
 
 	eq_size = roundup_pow_of_two(GDMA_EQE_SIZE * q_depth);
-	if (eq_size < MINIMUM_SUPPORTED_PAGE_SIZE)
-		eq_size = MINIMUM_SUPPORTED_PAGE_SIZE;
+	if (eq_size < MANA_MIN_QSIZE)
+		eq_size = MANA_MIN_QSIZE;
 
 	cq_size = roundup_pow_of_two(GDMA_CQE_SIZE * q_depth);
-	if (cq_size < MINIMUM_SUPPORTED_PAGE_SIZE)
-		cq_size = MINIMUM_SUPPORTED_PAGE_SIZE;
+	if (cq_size < MANA_MIN_QSIZE)
+		cq_size = MANA_MIN_QSIZE;
 
 	hwc_cq = kzalloc(sizeof(*hwc_cq), GFP_KERNEL);
 	if (!hwc_cq)
@@ -429,7 +429,7 @@ static int mana_hwc_alloc_dma_buf(struct hw_channel_context *hwc, u16 q_depth,
 
 	dma_buf->num_reqs = q_depth;
 
-	buf_size = PAGE_ALIGN(q_depth * max_msg_size);
+	buf_size = MANA_PAGE_ALIGN(q_depth * max_msg_size);
 
 	gmi = &dma_buf->mem_info;
 	err = mana_gd_alloc_memory(gc, buf_size, gmi);
@@ -497,8 +497,8 @@ static int mana_hwc_create_wq(struct hw_channel_context *hwc,
 	else
 		queue_size = roundup_pow_of_two(GDMA_MAX_SQE_SIZE * q_depth);
 
-	if (queue_size < MINIMUM_SUPPORTED_PAGE_SIZE)
-		queue_size = MINIMUM_SUPPORTED_PAGE_SIZE;
+	if (queue_size < MANA_MIN_QSIZE)
+		queue_size = MANA_MIN_QSIZE;
 
 	hwc_wq = kzalloc(sizeof(*hwc_wq), GFP_KERNEL);
 	if (!hwc_wq)
diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index 93e526e5dd16..3f094a8a4e97 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -1904,10 +1904,10 @@ static int mana_create_txq(struct mana_port_context *apc,
 	 *  to prevent overflow.
 	 */
 	txq_size = MAX_SEND_BUFFERS_PER_QUEUE * 32;
-	BUILD_BUG_ON(!PAGE_ALIGNED(txq_size));
+	BUILD_BUG_ON(!MANA_PAGE_ALIGNED(txq_size));
 
 	cq_size = MAX_SEND_BUFFERS_PER_QUEUE * COMP_ENTRY_SIZE;
-	cq_size = PAGE_ALIGN(cq_size);
+	cq_size = MANA_PAGE_ALIGN(cq_size);
 
 	gc = gd->gdma_context;
 
@@ -2204,8 +2204,8 @@ static struct mana_rxq *mana_create_rxq(struct mana_port_context *apc,
 	if (err)
 		goto out;
 
-	rq_size = PAGE_ALIGN(rq_size);
-	cq_size = PAGE_ALIGN(cq_size);
+	rq_size = MANA_PAGE_ALIGN(rq_size);
+	cq_size = MANA_PAGE_ALIGN(cq_size);
 
 	/* Create RQ */
 	memset(&spec, 0, sizeof(spec));
diff --git a/drivers/net/ethernet/microsoft/mana/shm_channel.c b/drivers/net/ethernet/microsoft/mana/shm_channel.c
index 5553af9c8085..0f1679ebad96 100644
--- a/drivers/net/ethernet/microsoft/mana/shm_channel.c
+++ b/drivers/net/ethernet/microsoft/mana/shm_channel.c
@@ -6,6 +6,7 @@
 #include <linux/io.h>
 #include <linux/mm.h>
 
+#include <net/mana/gdma.h>
 #include <net/mana/shm_channel.h>
 
 #define PAGE_FRAME_L48_WIDTH_BYTES 6
@@ -155,8 +156,8 @@ int mana_smc_setup_hwc(struct shm_channel *sc, bool reset_vf, u64 eq_addr,
 		return err;
 	}
 
-	if (!PAGE_ALIGNED(eq_addr) || !PAGE_ALIGNED(cq_addr) ||
-	    !PAGE_ALIGNED(rq_addr) || !PAGE_ALIGNED(sq_addr))
+	if (!MANA_PAGE_ALIGNED(eq_addr) || !MANA_PAGE_ALIGNED(cq_addr) ||
+	    !MANA_PAGE_ALIGNED(rq_addr) || !MANA_PAGE_ALIGNED(sq_addr))
 		return -EINVAL;
 
 	if ((eq_msix_index & VECTOR_MASK) != eq_msix_index)
@@ -183,7 +184,7 @@ int mana_smc_setup_hwc(struct shm_channel *sc, bool reset_vf, u64 eq_addr,
 
 	/* EQ addr: low 48 bits of frame address */
 	shmem = (u64 *)ptr;
-	frame_addr = PHYS_PFN(eq_addr);
+	frame_addr = MANA_PFN(eq_addr);
 	*shmem = frame_addr & PAGE_FRAME_L48_MASK;
 	all_addr_h4bits |= (frame_addr >> PAGE_FRAME_L48_WIDTH_BITS) <<
 		(frame_addr_seq++ * PAGE_FRAME_H4_WIDTH_BITS);
@@ -191,7 +192,7 @@ int mana_smc_setup_hwc(struct shm_channel *sc, bool reset_vf, u64 eq_addr,
 
 	/* CQ addr: low 48 bits of frame address */
 	shmem = (u64 *)ptr;
-	frame_addr = PHYS_PFN(cq_addr);
+	frame_addr = MANA_PFN(cq_addr);
 	*shmem = frame_addr & PAGE_FRAME_L48_MASK;
 	all_addr_h4bits |= (frame_addr >> PAGE_FRAME_L48_WIDTH_BITS) <<
 		(frame_addr_seq++ * PAGE_FRAME_H4_WIDTH_BITS);
@@ -199,7 +200,7 @@ int mana_smc_setup_hwc(struct shm_channel *sc, bool reset_vf, u64 eq_addr,
 
 	/* RQ addr: low 48 bits of frame address */
 	shmem = (u64 *)ptr;
-	frame_addr = PHYS_PFN(rq_addr);
+	frame_addr = MANA_PFN(rq_addr);
 	*shmem = frame_addr & PAGE_FRAME_L48_MASK;
 	all_addr_h4bits |= (frame_addr >> PAGE_FRAME_L48_WIDTH_BITS) <<
 		(frame_addr_seq++ * PAGE_FRAME_H4_WIDTH_BITS);
@@ -207,7 +208,7 @@ int mana_smc_setup_hwc(struct shm_channel *sc, bool reset_vf, u64 eq_addr,
 
 	/* SQ addr: low 48 bits of frame address */
 	shmem = (u64 *)ptr;
-	frame_addr = PHYS_PFN(sq_addr);
+	frame_addr = MANA_PFN(sq_addr);
 	*shmem = frame_addr & PAGE_FRAME_L48_MASK;
 	all_addr_h4bits |= (frame_addr >> PAGE_FRAME_L48_WIDTH_BITS) <<
 		(frame_addr_seq++ * PAGE_FRAME_H4_WIDTH_BITS);
diff --git a/include/net/mana/gdma.h b/include/net/mana/gdma.h
index c547756c4284..83963d9e804d 100644
--- a/include/net/mana/gdma.h
+++ b/include/net/mana/gdma.h
@@ -224,7 +224,15 @@ struct gdma_dev {
 	struct auxiliary_device *adev;
 };
 
-#define MINIMUM_SUPPORTED_PAGE_SIZE PAGE_SIZE
+/* MANA_PAGE_SIZE is the DMA unit */
+#define MANA_PAGE_SHIFT 12
+#define MANA_PAGE_SIZE BIT(MANA_PAGE_SHIFT)
+#define MANA_PAGE_ALIGN(x) ALIGN((x), MANA_PAGE_SIZE)
+#define MANA_PAGE_ALIGNED(addr) IS_ALIGNED((unsigned long)(addr), MANA_PAGE_SIZE)
+#define MANA_PFN(a) ((a) >> MANA_PAGE_SHIFT)
+
+/* Required by HW */
+#define MANA_MIN_QSIZE MANA_PAGE_SIZE
 
 #define GDMA_CQE_SIZE 64
 #define GDMA_EQE_SIZE 16
diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
index 59823901b74f..e39b8676fe54 100644
--- a/include/net/mana/mana.h
+++ b/include/net/mana/mana.h
@@ -42,7 +42,8 @@ enum TRI_STATE {
 
 #define MAX_SEND_BUFFERS_PER_QUEUE 256
 
-#define EQ_SIZE (8 * PAGE_SIZE)
+#define EQ_SIZE (8 * MANA_PAGE_SIZE)
+
 #define LOG2_EQ_THROTTLE 3
 
 #define MAX_PORTS_IN_MANA_DEV 256
-- 
cgit v1.2.3-58-ga151


From 9f1f70dd850038b37e8b103ef0cc6f5eaeb6a8fc Mon Sep 17 00:00:00 2001
From: Hongfu Li <lihongfu@kylinos.cn>
Date: Mon, 17 Jun 2024 15:54:35 +0800
Subject: rds:Simplify the allocation of slab caches

Use the new KMEM_CACHE() macro instead of direct kmem_cache_create
to simplify the creation of SLAB caches.

Signed-off-by: Hongfu Li <lihongfu@kylinos.cn>
Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Reviewed-by: Allison Henderson <allison.henderson@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/rds/tcp.c      | 4 +---
 net/rds/tcp_recv.c | 4 +---
 2 files changed, 2 insertions(+), 6 deletions(-)

diff --git a/net/rds/tcp.c b/net/rds/tcp.c
index d8111ac83bb6..3dc6956f66f8 100644
--- a/net/rds/tcp.c
+++ b/net/rds/tcp.c
@@ -719,9 +719,7 @@ static int __init rds_tcp_init(void)
 {
 	int ret;
 
-	rds_tcp_conn_slab = kmem_cache_create("rds_tcp_connection",
-					      sizeof(struct rds_tcp_connection),
-					      0, 0, NULL);
+	rds_tcp_conn_slab = KMEM_CACHE(rds_tcp_connection, 0);
 	if (!rds_tcp_conn_slab) {
 		ret = -ENOMEM;
 		goto out;
diff --git a/net/rds/tcp_recv.c b/net/rds/tcp_recv.c
index c00f04a1a534..7997a19d1da3 100644
--- a/net/rds/tcp_recv.c
+++ b/net/rds/tcp_recv.c
@@ -337,9 +337,7 @@ out:
 
 int rds_tcp_recv_init(void)
 {
-	rds_tcp_incoming_slab = kmem_cache_create("rds_tcp_incoming",
-					sizeof(struct rds_tcp_incoming),
-					0, 0, NULL);
+	rds_tcp_incoming_slab = KMEM_CACHE(rds_tcp_incoming, 0);
 	if (!rds_tcp_incoming_slab)
 		return -ENOMEM;
 	return 0;
-- 
cgit v1.2.3-58-ga151


From 5e1e43893be21fb726b86e630e8e954289dcd8e4 Mon Sep 17 00:00:00 2001
From: Diogo Ivo <diogo.ivo@siemens.com>
Date: Mon, 17 Jun 2024 16:21:40 +0100
Subject: net: ti: icssg-prueth: Enable PTP timestamping support for SR1.0
 devices

Enable PTP support for AM65x SR1.0 devices by registering with the IEP
infrastructure in order to expose a PTP clock to userspace.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Signed-off-by: Diogo Ivo <diogo.ivo@siemens.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c | 51 +++++++++++++++++++++++-
 1 file changed, 50 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c b/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c
index 7b3304bbd7fc..fa98bdb11ece 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c
@@ -1011,16 +1011,44 @@ static int prueth_probe(struct platform_device *pdev)
 	dev_dbg(dev, "sram: pa %llx va %p size %zx\n", prueth->msmcram.pa,
 		prueth->msmcram.va, prueth->msmcram.size);
 
+	prueth->iep0 = icss_iep_get_idx(np, 0);
+	if (IS_ERR(prueth->iep0)) {
+		ret = dev_err_probe(dev, PTR_ERR(prueth->iep0),
+				    "iep0 get failed\n");
+		goto free_pool;
+	}
+
+	prueth->iep1 = icss_iep_get_idx(np, 1);
+	if (IS_ERR(prueth->iep1)) {
+		ret = dev_err_probe(dev, PTR_ERR(prueth->iep1),
+				    "iep1 get failed\n");
+		goto put_iep0;
+	}
+
+	ret = icss_iep_init(prueth->iep0, NULL, NULL, 0);
+	if (ret) {
+		dev_err_probe(dev, ret, "failed to init iep0\n");
+		goto put_iep;
+	}
+
+	ret = icss_iep_init(prueth->iep1, NULL, NULL, 0);
+	if (ret) {
+		dev_err_probe(dev, ret, "failed to init iep1\n");
+		goto exit_iep0;
+	}
+
 	if (eth0_node) {
 		ret = prueth_netdev_init(prueth, eth0_node);
 		if (ret) {
 			dev_err_probe(dev, ret, "netdev init %s failed\n",
 				      eth0_node->name);
-			goto free_pool;
+			goto exit_iep;
 		}
 
 		if (of_find_property(eth0_node, "ti,half-duplex-capable", NULL))
 			prueth->emac[PRUETH_MAC0]->half_duplex = 1;
+
+		prueth->emac[PRUETH_MAC0]->iep = prueth->iep0;
 	}
 
 	if (eth1_node) {
@@ -1033,6 +1061,8 @@ static int prueth_probe(struct platform_device *pdev)
 
 		if (of_find_property(eth1_node, "ti,half-duplex-capable", NULL))
 			prueth->emac[PRUETH_MAC1]->half_duplex = 1;
+
+		prueth->emac[PRUETH_MAC1]->iep = prueth->iep1;
 	}
 
 	/* register the network devices */
@@ -1091,6 +1121,19 @@ netdev_exit:
 		prueth_netdev_exit(prueth, eth_node);
 	}
 
+exit_iep:
+	icss_iep_exit(prueth->iep1);
+exit_iep0:
+	icss_iep_exit(prueth->iep0);
+
+put_iep:
+	icss_iep_put(prueth->iep1);
+
+put_iep0:
+	icss_iep_put(prueth->iep0);
+	prueth->iep0 = NULL;
+	prueth->iep1 = NULL;
+
 free_pool:
 	gen_pool_free(prueth->sram_pool,
 		      (unsigned long)prueth->msmcram.va, msmc_ram_size);
@@ -1138,6 +1181,12 @@ static void prueth_remove(struct platform_device *pdev)
 		prueth_netdev_exit(prueth, eth_node);
 	}
 
+	icss_iep_exit(prueth->iep1);
+	icss_iep_exit(prueth->iep0);
+
+	icss_iep_put(prueth->iep1);
+	icss_iep_put(prueth->iep0);
+
 	gen_pool_free(prueth->sram_pool,
 		      (unsigned long)prueth->msmcram.va,
 		      MSMC_RAM_SIZE_SR1);
-- 
cgit v1.2.3-58-ga151


From 5758e03cf604aa282b9afa61aec3188c4a9b3fe7 Mon Sep 17 00:00:00 2001
From: Diogo Ivo <diogo.ivo@siemens.com>
Date: Mon, 17 Jun 2024 16:21:41 +0100
Subject: net: ti: icss-iep: Remove spinlock-based synchronization

As all sources of concurrency in hardware register access occur in
non-interrupt context eliminate spinlock-based synchronization and
rely on the mutex-based synchronization that is already present.

Signed-off-by: Diogo Ivo <diogo.ivo@siemens.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/ti/icssg/icss_iep.c | 14 --------------
 1 file changed, 14 deletions(-)

diff --git a/drivers/net/ethernet/ti/icssg/icss_iep.c b/drivers/net/ethernet/ti/icssg/icss_iep.c
index 3025e9c18970..d52e42fa64f2 100644
--- a/drivers/net/ethernet/ti/icssg/icss_iep.c
+++ b/drivers/net/ethernet/ti/icssg/icss_iep.c
@@ -110,7 +110,6 @@ struct icss_iep {
 	struct ptp_clock_info ptp_info;
 	struct ptp_clock *ptp_clock;
 	struct mutex ptp_clk_mutex;	/* PHC access serializer */
-	spinlock_t irq_lock; /* CMP IRQ vs icss_iep_ptp_enable access */
 	u32 def_inc;
 	s16 slow_cmp_inc;
 	u32 slow_cmp_count;
@@ -192,14 +191,11 @@ static void icss_iep_update_to_next_boundary(struct icss_iep *iep, u64 start_ns)
  */
 static void icss_iep_settime(struct icss_iep *iep, u64 ns)
 {
-	unsigned long flags;
-
 	if (iep->ops && iep->ops->settime) {
 		iep->ops->settime(iep->clockops_data, ns);
 		return;
 	}
 
-	spin_lock_irqsave(&iep->irq_lock, flags);
 	if (iep->pps_enabled || iep->perout_enabled)
 		writel(0, iep->base + iep->plat_data->reg_offs[ICSS_IEP_SYNC_CTRL_REG]);
 
@@ -210,7 +206,6 @@ static void icss_iep_settime(struct icss_iep *iep, u64 ns)
 		writel(IEP_SYNC_CTRL_SYNC_N_EN(0) | IEP_SYNC_CTRL_SYNC_EN,
 		       iep->base + iep->plat_data->reg_offs[ICSS_IEP_SYNC_CTRL_REG]);
 	}
-	spin_unlock_irqrestore(&iep->irq_lock, flags);
 }
 
 /**
@@ -546,7 +541,6 @@ static int icss_iep_perout_enable_hw(struct icss_iep *iep,
 static int icss_iep_perout_enable(struct icss_iep *iep,
 				  struct ptp_perout_request *req, int on)
 {
-	unsigned long flags;
 	int ret = 0;
 
 	mutex_lock(&iep->ptp_clk_mutex);
@@ -559,11 +553,9 @@ static int icss_iep_perout_enable(struct icss_iep *iep,
 	if (iep->perout_enabled == !!on)
 		goto exit;
 
-	spin_lock_irqsave(&iep->irq_lock, flags);
 	ret = icss_iep_perout_enable_hw(iep, req, on);
 	if (!ret)
 		iep->perout_enabled = !!on;
-	spin_unlock_irqrestore(&iep->irq_lock, flags);
 
 exit:
 	mutex_unlock(&iep->ptp_clk_mutex);
@@ -575,7 +567,6 @@ static int icss_iep_pps_enable(struct icss_iep *iep, int on)
 {
 	struct ptp_clock_request rq;
 	struct timespec64 ts;
-	unsigned long flags;
 	int ret = 0;
 	u64 ns;
 
@@ -589,8 +580,6 @@ static int icss_iep_pps_enable(struct icss_iep *iep, int on)
 	if (iep->pps_enabled == !!on)
 		goto exit;
 
-	spin_lock_irqsave(&iep->irq_lock, flags);
-
 	rq.perout.index = 0;
 	if (on) {
 		ns = icss_iep_gettime(iep, NULL);
@@ -607,8 +596,6 @@ static int icss_iep_pps_enable(struct icss_iep *iep, int on)
 	if (!ret)
 		iep->pps_enabled = !!on;
 
-	spin_unlock_irqrestore(&iep->irq_lock, flags);
-
 exit:
 	mutex_unlock(&iep->ptp_clk_mutex);
 
@@ -853,7 +840,6 @@ static int icss_iep_probe(struct platform_device *pdev)
 
 	iep->ptp_info = icss_iep_ptp_info;
 	mutex_init(&iep->ptp_clk_mutex);
-	spin_lock_init(&iep->irq_lock);
 	dev_set_drvdata(dev, iep);
 	icss_iep_disable(iep);
 
-- 
cgit v1.2.3-58-ga151


From 5056860cf8eab681a08913ec03432c57c83e71be Mon Sep 17 00:00:00 2001
From: Diogo Ivo <diogo.ivo@siemens.com>
Date: Mon, 17 Jun 2024 16:21:42 +0100
Subject: dt-bindings: net: Add IEP interrupt

The IEP interrupt is used in order to support both capture events, where
an incoming external signal gets timestamped on arrival, and compare
events, where an interrupt is generated internally when the IEP counter
reaches a programmed value.

Signed-off-by: Diogo Ivo <diogo.ivo@siemens.com>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/devicetree/bindings/net/ti,icss-iep.yaml | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/ti,icss-iep.yaml b/Documentation/devicetree/bindings/net/ti,icss-iep.yaml
index f5c22d6dcaee..e36e3a622904 100644
--- a/Documentation/devicetree/bindings/net/ti,icss-iep.yaml
+++ b/Documentation/devicetree/bindings/net/ti,icss-iep.yaml
@@ -28,6 +28,15 @@ properties:
     maxItems: 1
     description: phandle to the IEP source clock
 
+  interrupts:
+    maxItems: 1
+    description:
+      Interrupt specifier for capture/compare IRQ.
+
+  interrupt-names:
+    items:
+      - const: iep_cap_cmp
+
 required:
   - compatible
   - reg
-- 
cgit v1.2.3-58-ga151


From f18ad402cd8bdf8e4a85dd5f209ceefceceffd9c Mon Sep 17 00:00:00 2001
From: Diogo Ivo <diogo.ivo@siemens.com>
Date: Mon, 17 Jun 2024 16:21:43 +0100
Subject: net: ti: icss-iep: Enable compare events

The IEP module supports compare events, in which a value is written to a
hardware register and when the IEP counter reaches the written value an
interrupt is generated. Add handling for this interrupt in order to
support PPS events.

Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Diogo Ivo <diogo.ivo@siemens.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/ti/icssg/icss_iep.c | 74 ++++++++++++++++++++++++++++++++
 1 file changed, 74 insertions(+)

diff --git a/drivers/net/ethernet/ti/icssg/icss_iep.c b/drivers/net/ethernet/ti/icssg/icss_iep.c
index d52e42fa64f2..003668dee738 100644
--- a/drivers/net/ethernet/ti/icssg/icss_iep.c
+++ b/drivers/net/ethernet/ti/icssg/icss_iep.c
@@ -17,6 +17,7 @@
 #include <linux/timekeeping.h>
 #include <linux/interrupt.h>
 #include <linux/of_irq.h>
+#include <linux/workqueue.h>
 
 #include "icss_iep.h"
 
@@ -121,6 +122,7 @@ struct icss_iep {
 	int cap_cmp_irq;
 	u64 period;
 	u32 latch_enable;
+	struct work_struct work;
 };
 
 /**
@@ -563,6 +565,57 @@ exit:
 	return ret;
 }
 
+static void icss_iep_cap_cmp_work(struct work_struct *work)
+{
+	struct icss_iep *iep = container_of(work, struct icss_iep, work);
+	const u32 *reg_offs = iep->plat_data->reg_offs;
+	struct ptp_clock_event pevent;
+	unsigned int val;
+	u64 ns, ns_next;
+
+	mutex_lock(&iep->ptp_clk_mutex);
+
+	ns = readl(iep->base + reg_offs[ICSS_IEP_CMP1_REG0]);
+	if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT) {
+		val = readl(iep->base + reg_offs[ICSS_IEP_CMP1_REG1]);
+		ns |= (u64)val << 32;
+	}
+	/* set next event */
+	ns_next = ns + iep->period;
+	writel(lower_32_bits(ns_next),
+	       iep->base + reg_offs[ICSS_IEP_CMP1_REG0]);
+	if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT)
+		writel(upper_32_bits(ns_next),
+		       iep->base + reg_offs[ICSS_IEP_CMP1_REG1]);
+
+	pevent.pps_times.ts_real = ns_to_timespec64(ns);
+	pevent.type = PTP_CLOCK_PPSUSR;
+	pevent.index = 0;
+	ptp_clock_event(iep->ptp_clock, &pevent);
+	dev_dbg(iep->dev, "IEP:pps ts: %llu next:%llu:\n", ns, ns_next);
+
+	mutex_unlock(&iep->ptp_clk_mutex);
+}
+
+static irqreturn_t icss_iep_cap_cmp_irq(int irq, void *dev_id)
+{
+	struct icss_iep *iep = (struct icss_iep *)dev_id;
+	const u32 *reg_offs = iep->plat_data->reg_offs;
+	unsigned int val;
+
+	val = readl(iep->base + reg_offs[ICSS_IEP_CMP_STAT_REG]);
+	/* The driver only enables CMP1 */
+	if (val & BIT(1)) {
+		/* Clear the event */
+		writel(BIT(1), iep->base + reg_offs[ICSS_IEP_CMP_STAT_REG]);
+		if (iep->pps_enabled || iep->perout_enabled)
+			schedule_work(&iep->work);
+		return IRQ_HANDLED;
+	}
+
+	return IRQ_NONE;
+}
+
 static int icss_iep_pps_enable(struct icss_iep *iep, int on)
 {
 	struct ptp_clock_request rq;
@@ -591,6 +644,8 @@ static int icss_iep_pps_enable(struct icss_iep *iep, int on)
 		ret = icss_iep_perout_enable_hw(iep, &rq.perout, on);
 	} else {
 		ret = icss_iep_perout_enable_hw(iep, &rq.perout, on);
+		if (iep->cap_cmp_irq)
+			cancel_work_sync(&iep->work);
 	}
 
 	if (!ret)
@@ -764,6 +819,8 @@ int icss_iep_init(struct icss_iep *iep, const struct icss_iep_clockops *clkops,
 	if (iep->ops && iep->ops->perout_enable) {
 		iep->ptp_info.n_per_out = 1;
 		iep->ptp_info.pps = 1;
+	} else if (iep->cap_cmp_irq) {
+		iep->ptp_info.pps = 1;
 	}
 
 	if (iep->ops && iep->ops->extts_enable)
@@ -804,6 +861,7 @@ static int icss_iep_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct icss_iep *iep;
 	struct clk *iep_clk;
+	int ret, irq;
 
 	iep = devm_kzalloc(dev, sizeof(*iep), GFP_KERNEL);
 	if (!iep)
@@ -814,6 +872,22 @@ static int icss_iep_probe(struct platform_device *pdev)
 	if (IS_ERR(iep->base))
 		return -ENODEV;
 
+	irq = platform_get_irq_byname_optional(pdev, "iep_cap_cmp");
+	if (irq == -EPROBE_DEFER)
+		return irq;
+
+	if (irq > 0) {
+		ret = devm_request_irq(dev, irq, icss_iep_cap_cmp_irq,
+				       IRQF_TRIGGER_HIGH, "iep_cap_cmp", iep);
+		if (ret) {
+			dev_info(iep->dev, "cap_cmp irq request failed: %x\n",
+				 ret);
+		} else {
+			iep->cap_cmp_irq = irq;
+			INIT_WORK(&iep->work, icss_iep_cap_cmp_work);
+		}
+	}
+
 	iep_clk = devm_clk_get(dev, NULL);
 	if (IS_ERR(iep_clk))
 		return PTR_ERR(iep_clk);
-- 
cgit v1.2.3-58-ga151


From 71be1189c92bc1e92f99f8f675e03b24ed00c6ff Mon Sep 17 00:00:00 2001
From: Diogo Ivo <diogo.ivo@siemens.com>
Date: Mon, 17 Jun 2024 16:21:44 +0100
Subject: arm64: dts: ti: iot2050: Add IEP interrupts for SR1.0 devices

Add the interrupts needed for PTP Hardware Clock support via IEP
in SR1.0 devices.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Diogo Ivo <diogo.ivo@siemens.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/arm64/boot/dts/ti/k3-am65-iot2050-common-pg1.dtsi | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/arch/arm64/boot/dts/ti/k3-am65-iot2050-common-pg1.dtsi b/arch/arm64/boot/dts/ti/k3-am65-iot2050-common-pg1.dtsi
index ef7897763ef8..0a29ed172215 100644
--- a/arch/arm64/boot/dts/ti/k3-am65-iot2050-common-pg1.dtsi
+++ b/arch/arm64/boot/dts/ti/k3-am65-iot2050-common-pg1.dtsi
@@ -73,3 +73,15 @@
 		    "rx0", "rx1",
 		    "rxmgm0", "rxmgm1";
 };
+
+&icssg0_iep0 {
+	interrupt-parent = <&icssg0_intc>;
+	interrupts = <7 7 7>;
+	interrupt-names = "iep_cap_cmp";
+};
+
+&icssg0_iep1 {
+	interrupt-parent = <&icssg0_intc>;
+	interrupts = <56 8 8>;
+	interrupt-names = "iep_cap_cmp";
+};
-- 
cgit v1.2.3-58-ga151


From c53795d48ee8f385c6a9e394651e7ee914baaeba Mon Sep 17 00:00:00 2001
From: Yan Zhai <yan@cloudflare.com>
Date: Mon, 17 Jun 2024 11:09:04 -0700
Subject: net: add rx_sk to trace_kfree_skb

skb does not include enough information to find out receiving
sockets/services and netns/containers on packet drops. In theory
skb->dev tells about netns, but it can get cleared/reused, e.g. by TCP
stack for OOO packet lookup. Similarly, skb->sk often identifies a local
sender, and tells nothing about a receiver.

Allow passing an extra receiving socket to the tracepoint to improve
the visibility on receiving drops.

Signed-off-by: Yan Zhai <yan@cloudflare.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/trace/events/skb.h | 11 +++++++----
 net/core/dev.c             |  2 +-
 net/core/drop_monitor.c    |  9 ++++++---
 net/core/skbuff.c          |  2 +-
 4 files changed, 15 insertions(+), 9 deletions(-)

diff --git a/include/trace/events/skb.h b/include/trace/events/skb.h
index 07e0715628ec..b877133cd93a 100644
--- a/include/trace/events/skb.h
+++ b/include/trace/events/skb.h
@@ -24,13 +24,14 @@ DEFINE_DROP_REASON(FN, FN)
 TRACE_EVENT(kfree_skb,
 
 	TP_PROTO(struct sk_buff *skb, void *location,
-		 enum skb_drop_reason reason),
+		 enum skb_drop_reason reason, struct sock *rx_sk),
 
-	TP_ARGS(skb, location, reason),
+	TP_ARGS(skb, location, reason, rx_sk),
 
 	TP_STRUCT__entry(
 		__field(void *,		skbaddr)
 		__field(void *,		location)
+		__field(void *,		rx_sk)
 		__field(unsigned short,	protocol)
 		__field(enum skb_drop_reason,	reason)
 	),
@@ -38,12 +39,14 @@ TRACE_EVENT(kfree_skb,
 	TP_fast_assign(
 		__entry->skbaddr = skb;
 		__entry->location = location;
+		__entry->rx_sk = rx_sk;
 		__entry->protocol = ntohs(skb->protocol);
 		__entry->reason = reason;
 	),
 
-	TP_printk("skbaddr=%p protocol=%u location=%pS reason: %s",
-		  __entry->skbaddr, __entry->protocol, __entry->location,
+	TP_printk("skbaddr=%p rx_sk=%p protocol=%u location=%pS reason: %s",
+		  __entry->skbaddr, __entry->rx_sk, __entry->protocol,
+		  __entry->location,
 		  __print_symbolic(__entry->reason,
 				   DEFINE_DROP_REASON(FN, FNe)))
 );
diff --git a/net/core/dev.c b/net/core/dev.c
index c361a7b69da8..093d82bf0e28 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5234,7 +5234,7 @@ static __latent_entropy void net_tx_action(struct softirq_action *h)
 				trace_consume_skb(skb, net_tx_action);
 			else
 				trace_kfree_skb(skb, net_tx_action,
-						get_kfree_skb_cb(skb)->reason);
+						get_kfree_skb_cb(skb)->reason, NULL);
 
 			if (skb->fclone != SKB_FCLONE_UNAVAILABLE)
 				__kfree_skb(skb);
diff --git a/net/core/drop_monitor.c b/net/core/drop_monitor.c
index 430ed18f8584..2e0ae3328232 100644
--- a/net/core/drop_monitor.c
+++ b/net/core/drop_monitor.c
@@ -109,7 +109,8 @@ static u32 net_dm_queue_len = 1000;
 struct net_dm_alert_ops {
 	void (*kfree_skb_probe)(void *ignore, struct sk_buff *skb,
 				void *location,
-				enum skb_drop_reason reason);
+				enum skb_drop_reason reason,
+				struct sock *rx_sk);
 	void (*napi_poll_probe)(void *ignore, struct napi_struct *napi,
 				int work, int budget);
 	void (*work_item_func)(struct work_struct *work);
@@ -264,7 +265,8 @@ out:
 
 static void trace_kfree_skb_hit(void *ignore, struct sk_buff *skb,
 				void *location,
-				enum skb_drop_reason reason)
+				enum skb_drop_reason reason,
+				struct sock *rx_sk)
 {
 	trace_drop_common(skb, location);
 }
@@ -491,7 +493,8 @@ static const struct net_dm_alert_ops net_dm_alert_summary_ops = {
 static void net_dm_packet_trace_kfree_skb_hit(void *ignore,
 					      struct sk_buff *skb,
 					      void *location,
-					      enum skb_drop_reason reason)
+					      enum skb_drop_reason reason,
+					      struct sock *rx_sk)
 {
 	ktime_t tstamp = ktime_get_real();
 	struct per_cpu_dm_data *data;
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index c8ac79851cd6..8973db4eabd4 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1203,7 +1203,7 @@ bool __kfree_skb_reason(struct sk_buff *skb, enum skb_drop_reason reason)
 	if (reason == SKB_CONSUMED)
 		trace_consume_skb(skb, __builtin_return_address(0));
 	else
-		trace_kfree_skb(skb, __builtin_return_address(0), reason);
+		trace_kfree_skb(skb, __builtin_return_address(0), reason, NULL);
 	return true;
 }
 
-- 
cgit v1.2.3-58-ga151


From ba8de796baf4bdc03530774fb284fe3c97875566 Mon Sep 17 00:00:00 2001
From: Yan Zhai <yan@cloudflare.com>
Date: Mon, 17 Jun 2024 11:09:09 -0700
Subject: net: introduce sk_skb_reason_drop function

Long used destructors kfree_skb and kfree_skb_reason do not pass
receiving socket to packet drop tracepoints trace_kfree_skb.
This makes it hard to track packet drops of a certain netns (container)
or a socket (user application).

The naming of these destructors are also not consistent with most sk/skb
operating functions, i.e. functions named "sk_xxx" or "skb_xxx".
Introduce a new functions sk_skb_reason_drop as drop-in replacement for
kfree_skb_reason on local receiving path. Callers can now pass receiving
sockets to the tracepoints.

kfree_skb and kfree_skb_reason are still usable but they are now just
inline helpers that call sk_skb_reason_drop.

Note it is not feasible to do the same to consume_skb. Packets not
dropped can flow through multiple receive handlers, and have multiple
receiving sockets. Leave it untouched for now.

Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/skbuff.h | 10 ++++++++--
 net/core/skbuff.c      | 22 ++++++++++++----------
 2 files changed, 20 insertions(+), 12 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 813406a9bd6c..f4cda3fbdb75 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1251,8 +1251,14 @@ static inline bool skb_data_unref(const struct sk_buff *skb,
 	return true;
 }
 
-void __fix_address
-kfree_skb_reason(struct sk_buff *skb, enum skb_drop_reason reason);
+void __fix_address sk_skb_reason_drop(struct sock *sk, struct sk_buff *skb,
+				      enum skb_drop_reason reason);
+
+static inline void
+kfree_skb_reason(struct sk_buff *skb, enum skb_drop_reason reason)
+{
+	sk_skb_reason_drop(NULL, skb, reason);
+}
 
 /**
  *	kfree_skb - free an sk_buff with 'NOT_SPECIFIED' reason
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 8973db4eabd4..2315c088e91d 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1190,7 +1190,8 @@ void __kfree_skb(struct sk_buff *skb)
 EXPORT_SYMBOL(__kfree_skb);
 
 static __always_inline
-bool __kfree_skb_reason(struct sk_buff *skb, enum skb_drop_reason reason)
+bool __sk_skb_reason_drop(struct sock *sk, struct sk_buff *skb,
+			  enum skb_drop_reason reason)
 {
 	if (unlikely(!skb_unref(skb)))
 		return false;
@@ -1203,26 +1204,27 @@ bool __kfree_skb_reason(struct sk_buff *skb, enum skb_drop_reason reason)
 	if (reason == SKB_CONSUMED)
 		trace_consume_skb(skb, __builtin_return_address(0));
 	else
-		trace_kfree_skb(skb, __builtin_return_address(0), reason, NULL);
+		trace_kfree_skb(skb, __builtin_return_address(0), reason, sk);
 	return true;
 }
 
 /**
- *	kfree_skb_reason - free an sk_buff with special reason
+ *	sk_skb_reason_drop - free an sk_buff with special reason
+ *	@sk: the socket to receive @skb, or NULL if not applicable
  *	@skb: buffer to free
  *	@reason: reason why this skb is dropped
  *
- *	Drop a reference to the buffer and free it if the usage count has
- *	hit zero. Meanwhile, pass the drop reason to 'kfree_skb'
- *	tracepoint.
+ *	Drop a reference to the buffer and free it if the usage count has hit
+ *	zero. Meanwhile, pass the receiving socket and drop reason to
+ *	'kfree_skb' tracepoint.
  */
 void __fix_address
-kfree_skb_reason(struct sk_buff *skb, enum skb_drop_reason reason)
+sk_skb_reason_drop(struct sock *sk, struct sk_buff *skb, enum skb_drop_reason reason)
 {
-	if (__kfree_skb_reason(skb, reason))
+	if (__sk_skb_reason_drop(sk, skb, reason))
 		__kfree_skb(skb);
 }
-EXPORT_SYMBOL(kfree_skb_reason);
+EXPORT_SYMBOL(sk_skb_reason_drop);
 
 #define KFREE_SKB_BULK_SIZE	16
 
@@ -1261,7 +1263,7 @@ kfree_skb_list_reason(struct sk_buff *segs, enum skb_drop_reason reason)
 	while (segs) {
 		struct sk_buff *next = segs->next;
 
-		if (__kfree_skb_reason(segs, reason)) {
+		if (__sk_skb_reason_drop(NULL, segs, reason)) {
 			skb_poison_list(segs);
 			kfree_skb_add_bulk(segs, &sa, reason);
 		}
-- 
cgit v1.2.3-58-ga151


From 7467de17635fb9d240987a0b9574fb2283a53b73 Mon Sep 17 00:00:00 2001
From: Yan Zhai <yan@cloudflare.com>
Date: Mon, 17 Jun 2024 11:09:13 -0700
Subject: ping: use sk_skb_reason_drop to free rx packets

Replace kfree_skb_reason with sk_skb_reason_drop and pass the receiving
socket to the tracepoint.

Signed-off-by: Yan Zhai <yan@cloudflare.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/ping.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index 823306487a82..619ddc087957 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -946,7 +946,7 @@ static enum skb_drop_reason __ping_queue_rcv_skb(struct sock *sk,
 	pr_debug("ping_queue_rcv_skb(sk=%p,sk->num=%d,skb=%p)\n",
 		 inet_sk(sk), inet_sk(sk)->inet_num, skb);
 	if (sock_queue_rcv_skb_reason(sk, skb, &reason) < 0) {
-		kfree_skb_reason(skb, reason);
+		sk_skb_reason_drop(sk, skb, reason);
 		pr_debug("ping_queue_rcv_skb -> failed\n");
 		return reason;
 	}
-- 
cgit v1.2.3-58-ga151


From ce9a2424e9da2cd4e790f2498621bc2aa5e5d298 Mon Sep 17 00:00:00 2001
From: Yan Zhai <yan@cloudflare.com>
Date: Mon, 17 Jun 2024 11:09:16 -0700
Subject: net: raw: use sk_skb_reason_drop to free rx packets

Replace kfree_skb_reason with sk_skb_reason_drop and pass the receiving
socket to the tracepoint.

Signed-off-by: Yan Zhai <yan@cloudflare.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/raw.c | 4 ++--
 net/ipv6/raw.c | 8 ++++----
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 1a0953650356..474dfd263c8b 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -301,7 +301,7 @@ static int raw_rcv_skb(struct sock *sk, struct sk_buff *skb)
 
 	ipv4_pktinfo_prepare(sk, skb, true);
 	if (sock_queue_rcv_skb_reason(sk, skb, &reason) < 0) {
-		kfree_skb_reason(skb, reason);
+		sk_skb_reason_drop(sk, skb, reason);
 		return NET_RX_DROP;
 	}
 
@@ -312,7 +312,7 @@ int raw_rcv(struct sock *sk, struct sk_buff *skb)
 {
 	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) {
 		atomic_inc(&sk->sk_drops);
-		kfree_skb_reason(skb, SKB_DROP_REASON_XFRM_POLICY);
+		sk_skb_reason_drop(sk, skb, SKB_DROP_REASON_XFRM_POLICY);
 		return NET_RX_DROP;
 	}
 	nf_reset_ct(skb);
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index f838366e8256..608fa9d05b55 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -362,14 +362,14 @@ static inline int rawv6_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	if ((raw6_sk(sk)->checksum || rcu_access_pointer(sk->sk_filter)) &&
 	    skb_checksum_complete(skb)) {
 		atomic_inc(&sk->sk_drops);
-		kfree_skb_reason(skb, SKB_DROP_REASON_SKB_CSUM);
+		sk_skb_reason_drop(sk, skb, SKB_DROP_REASON_SKB_CSUM);
 		return NET_RX_DROP;
 	}
 
 	/* Charge it to the socket. */
 	skb_dst_drop(skb);
 	if (sock_queue_rcv_skb_reason(sk, skb, &reason) < 0) {
-		kfree_skb_reason(skb, reason);
+		sk_skb_reason_drop(sk, skb, reason);
 		return NET_RX_DROP;
 	}
 
@@ -390,7 +390,7 @@ int rawv6_rcv(struct sock *sk, struct sk_buff *skb)
 
 	if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) {
 		atomic_inc(&sk->sk_drops);
-		kfree_skb_reason(skb, SKB_DROP_REASON_XFRM_POLICY);
+		sk_skb_reason_drop(sk, skb, SKB_DROP_REASON_XFRM_POLICY);
 		return NET_RX_DROP;
 	}
 	nf_reset_ct(skb);
@@ -415,7 +415,7 @@ int rawv6_rcv(struct sock *sk, struct sk_buff *skb)
 	if (inet_test_bit(HDRINCL, sk)) {
 		if (skb_checksum_complete(skb)) {
 			atomic_inc(&sk->sk_drops);
-			kfree_skb_reason(skb, SKB_DROP_REASON_SKB_CSUM);
+			sk_skb_reason_drop(sk, skb, SKB_DROP_REASON_SKB_CSUM);
 			return NET_RX_DROP;
 		}
 	}
-- 
cgit v1.2.3-58-ga151


From 46a02aa357529d7b038096955976b14f7c44aa23 Mon Sep 17 00:00:00 2001
From: Yan Zhai <yan@cloudflare.com>
Date: Mon, 17 Jun 2024 11:09:20 -0700
Subject: tcp: use sk_skb_reason_drop to free rx packets

Replace kfree_skb_reason with sk_skb_reason_drop and pass the receiving
socket to the tracepoint.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/r/202406011539.jhwBd7DX-lkp@intel.com/
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/syncookies.c | 2 +-
 net/ipv4/tcp_input.c  | 2 +-
 net/ipv4/tcp_ipv4.c   | 6 +++---
 net/ipv6/syncookies.c | 2 +-
 net/ipv6/tcp_ipv6.c   | 6 +++---
 5 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index b61d36810fe3..1948d15f1f28 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -496,6 +496,6 @@ out:
 out_free:
 	reqsk_free(req);
 out_drop:
-	kfree_skb_reason(skb, reason);
+	sk_skb_reason_drop(sk, skb, reason);
 	return NULL;
 }
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index d0a1e34d69f6..f513d1f927ad 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4860,7 +4860,7 @@ static void tcp_drop_reason(struct sock *sk, struct sk_buff *skb,
 			    enum skb_drop_reason reason)
 {
 	sk_drops_add(sk, skb);
-	kfree_skb_reason(skb, reason);
+	sk_skb_reason_drop(sk, skb, reason);
 }
 
 /* This one checks to see if we can put data from the
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index de0c8f43448a..8e49d69279d5 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1932,7 +1932,7 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)
 reset:
 	tcp_v4_send_reset(rsk, skb, sk_rst_convert_drop_reason(reason));
 discard:
-	kfree_skb_reason(skb, reason);
+	sk_skb_reason_drop(sk, skb, reason);
 	/* Be careful here. If this function gets more complicated and
 	 * gcc suffers from register pressure on the x86, sk (in %ebx)
 	 * might be destroyed here. This current version compiles correctly,
@@ -2168,8 +2168,8 @@ int tcp_v4_rcv(struct sk_buff *skb)
 	int dif = inet_iif(skb);
 	const struct iphdr *iph;
 	const struct tcphdr *th;
+	struct sock *sk = NULL;
 	bool refcounted;
-	struct sock *sk;
 	int ret;
 	u32 isn;
 
@@ -2368,7 +2368,7 @@ bad_packet:
 discard_it:
 	SKB_DR_OR(drop_reason, NOT_SPECIFIED);
 	/* Discard frame. */
-	kfree_skb_reason(skb, drop_reason);
+	sk_skb_reason_drop(sk, skb, drop_reason);
 	return 0;
 
 discard_and_relse:
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index bfad1e89b6a6..9d83eadd308b 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -275,6 +275,6 @@ out:
 out_free:
 	reqsk_free(req);
 out_drop:
-	kfree_skb_reason(skb, reason);
+	sk_skb_reason_drop(sk, skb, reason);
 	return NULL;
 }
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 223b71790e44..200fea92f12f 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1674,7 +1674,7 @@ reset:
 discard:
 	if (opt_skb)
 		__kfree_skb(opt_skb);
-	kfree_skb_reason(skb, reason);
+	sk_skb_reason_drop(sk, skb, reason);
 	return 0;
 csum_err:
 	reason = SKB_DROP_REASON_TCP_CSUM;
@@ -1747,8 +1747,8 @@ INDIRECT_CALLABLE_SCOPE int tcp_v6_rcv(struct sk_buff *skb)
 	int dif = inet6_iif(skb);
 	const struct tcphdr *th;
 	const struct ipv6hdr *hdr;
+	struct sock *sk = NULL;
 	bool refcounted;
-	struct sock *sk;
 	int ret;
 	u32 isn;
 	struct net *net = dev_net(skb->dev);
@@ -1940,7 +1940,7 @@ bad_packet:
 
 discard_it:
 	SKB_DR_OR(drop_reason, NOT_SPECIFIED);
-	kfree_skb_reason(skb, drop_reason);
+	sk_skb_reason_drop(sk, skb, drop_reason);
 	return 0;
 
 discard_and_relse:
-- 
cgit v1.2.3-58-ga151


From fc0cc9248843b37243fa5fd3287a121ec41d291f Mon Sep 17 00:00:00 2001
From: Yan Zhai <yan@cloudflare.com>
Date: Mon, 17 Jun 2024 11:09:24 -0700
Subject: udp: use sk_skb_reason_drop to free rx packets

Replace kfree_skb_reason with sk_skb_reason_drop and pass the receiving
socket to the tracepoint.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/r/202406011751.NpVN0sSk-lkp@intel.com/
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/udp.c | 10 +++++-----
 net/ipv6/udp.c | 10 +++++-----
 2 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index c9ca6d285347..d08bf16d476d 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -2074,7 +2074,7 @@ static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
 		}
 		UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
 		trace_udp_fail_queue_rcv_skb(rc, sk, skb);
-		kfree_skb_reason(skb, drop_reason);
+		sk_skb_reason_drop(sk, skb, drop_reason);
 		return -1;
 	}
 
@@ -2196,7 +2196,7 @@ csum_error:
 drop:
 	__UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
 	atomic_inc(&sk->sk_drops);
-	kfree_skb_reason(skb, drop_reason);
+	sk_skb_reason_drop(sk, skb, drop_reason);
 	return -1;
 }
 
@@ -2383,7 +2383,7 @@ static int udp_unicast_rcv_skb(struct sock *sk, struct sk_buff *skb,
 int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
 		   int proto)
 {
-	struct sock *sk;
+	struct sock *sk = NULL;
 	struct udphdr *uh;
 	unsigned short ulen;
 	struct rtable *rt = skb_rtable(skb);
@@ -2460,7 +2460,7 @@ no_sk:
 	 * Hmm.  We got an UDP packet to a port to which we
 	 * don't wanna listen.  Ignore it.
 	 */
-	kfree_skb_reason(skb, drop_reason);
+	sk_skb_reason_drop(sk, skb, drop_reason);
 	return 0;
 
 short_packet:
@@ -2485,7 +2485,7 @@ csum_error:
 	__UDP_INC_STATS(net, UDP_MIB_CSUMERRORS, proto == IPPROTO_UDPLITE);
 drop:
 	__UDP_INC_STATS(net, UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
-	kfree_skb_reason(skb, drop_reason);
+	sk_skb_reason_drop(sk, skb, drop_reason);
 	return 0;
 }
 
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index c81a07ac0463..b56f0b9f4307 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -673,7 +673,7 @@ static int __udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
 		}
 		UDP6_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
 		trace_udp_fail_queue_rcv_skb(rc, sk, skb);
-		kfree_skb_reason(skb, drop_reason);
+		sk_skb_reason_drop(sk, skb, drop_reason);
 		return -1;
 	}
 
@@ -776,7 +776,7 @@ csum_error:
 drop:
 	__UDP6_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
 	atomic_inc(&sk->sk_drops);
-	kfree_skb_reason(skb, drop_reason);
+	sk_skb_reason_drop(sk, skb, drop_reason);
 	return -1;
 }
 
@@ -940,8 +940,8 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
 	enum skb_drop_reason reason = SKB_DROP_REASON_NOT_SPECIFIED;
 	const struct in6_addr *saddr, *daddr;
 	struct net *net = dev_net(skb->dev);
+	struct sock *sk = NULL;
 	struct udphdr *uh;
-	struct sock *sk;
 	bool refcounted;
 	u32 ulen = 0;
 
@@ -1033,7 +1033,7 @@ no_sk:
 	__UDP6_INC_STATS(net, UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
 	icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0);
 
-	kfree_skb_reason(skb, reason);
+	sk_skb_reason_drop(sk, skb, reason);
 	return 0;
 
 short_packet:
@@ -1054,7 +1054,7 @@ csum_error:
 	__UDP6_INC_STATS(net, UDP_MIB_CSUMERRORS, proto == IPPROTO_UDPLITE);
 discard:
 	__UDP6_INC_STATS(net, UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
-	kfree_skb_reason(skb, reason);
+	sk_skb_reason_drop(sk, skb, reason);
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From e2e7d78d9a25c78dc829da400bcec857b8c41b78 Mon Sep 17 00:00:00 2001
From: Yan Zhai <yan@cloudflare.com>
Date: Mon, 17 Jun 2024 11:09:27 -0700
Subject: af_packet: use sk_skb_reason_drop to free rx packets

Replace kfree_skb_reason with sk_skb_reason_drop and pass the receiving
socket to the tracepoint.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/r/202406011859.Aacus8GV-lkp@intel.com/
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/packet/af_packet.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index fce390887591..42d29b8a84fc 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -2121,7 +2121,7 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev,
 		      struct packet_type *pt, struct net_device *orig_dev)
 {
 	enum skb_drop_reason drop_reason = SKB_CONSUMED;
-	struct sock *sk;
+	struct sock *sk = NULL;
 	struct sockaddr_ll *sll;
 	struct packet_sock *po;
 	u8 *skb_head = skb->data;
@@ -2226,7 +2226,7 @@ drop_n_restore:
 		skb->len = skb_len;
 	}
 drop:
-	kfree_skb_reason(skb, drop_reason);
+	sk_skb_reason_drop(sk, skb, drop_reason);
 	return 0;
 }
 
@@ -2234,7 +2234,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 		       struct packet_type *pt, struct net_device *orig_dev)
 {
 	enum skb_drop_reason drop_reason = SKB_CONSUMED;
-	struct sock *sk;
+	struct sock *sk = NULL;
 	struct packet_sock *po;
 	struct sockaddr_ll *sll;
 	union tpacket_uhdr h;
@@ -2494,7 +2494,7 @@ drop_n_restore:
 		skb->len = skb_len;
 	}
 drop:
-	kfree_skb_reason(skb, drop_reason);
+	sk_skb_reason_drop(sk, skb, drop_reason);
 	return 0;
 
 drop_n_account:
@@ -2503,7 +2503,7 @@ drop_n_account:
 	drop_reason = SKB_DROP_REASON_PACKET_SOCK_ERROR;
 
 	sk->sk_data_ready(sk);
-	kfree_skb_reason(copy_skb, drop_reason);
+	sk_skb_reason_drop(sk, copy_skb, drop_reason);
 	goto drop_n_restore;
 }
 
-- 
cgit v1.2.3-58-ga151


From c25c961fc7f36682f0a530150f1b7453ebc344cd Mon Sep 17 00:00:00 2001
From: Matthias Schiffer <mschiffer@universe-factory.net>
Date: Tue, 18 Jun 2024 09:17:12 +0200
Subject: net: dsa: mt7530: factor out bridge join/leave logic
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

As preparation for implementing bridge port isolation, move the logic to
add and remove bits in the port matrix into a new helper
mt7530_update_port_member(), which is called from
mt7530_port_bridge_join() and mt7530_port_bridge_leave().

Another part of the preparation is using dsa_port_offloads_bridge_dev()
instead of dsa_port_offloads_bridge() to check for bridge membership, as
we don't have a struct dsa_bridge in mt7530_port_bridge_flags().

The port matrix setting is slightly streamlined, now always first setting
the mt7530_port's pm field and then writing the port matrix from that
field into the hardware register, instead of duplicating the bit
manipulation for both the struct field and the register.

mt7530_port_bridge_join() was previously using |= to update the port
matrix with the port bitmap, which was unnecessary, as pm would only
have the CPU port set before joining a bridge; a simple assignment can
be used for both joining and leaving (and will also work when individual
bits are added/removed in port_bitmap with regard to the previous port
matrix, which is what happens with port isolation).

No functional change intended.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Reviewed-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 105 ++++++++++++++++++++++-------------------------
 1 file changed, 48 insertions(+), 57 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index 598434d8d6e4..9ce27ce07d77 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -1302,6 +1302,52 @@ mt7530_stp_state_set(struct dsa_switch *ds, int port, u8 state)
 		   FID_PST(FID_BRIDGED, stp_state));
 }
 
+static void mt7530_update_port_member(struct mt7530_priv *priv, int port,
+				      const struct net_device *bridge_dev,
+				      bool join) __must_hold(&priv->reg_mutex)
+{
+	struct dsa_port *dp = dsa_to_port(priv->ds, port), *other_dp;
+	struct mt7530_port *p = &priv->ports[port], *other_p;
+	struct dsa_port *cpu_dp = dp->cpu_dp;
+	u32 port_bitmap = BIT(cpu_dp->index);
+	int other_port;
+
+	dsa_switch_for_each_user_port(other_dp, priv->ds) {
+		other_port = other_dp->index;
+		other_p = &priv->ports[other_port];
+
+		if (dp == other_dp)
+			continue;
+
+		/* Add/remove this port to/from the port matrix of the other
+		 * ports in the same bridge. If the port is disabled, port
+		 * matrix is kept and not being setup until the port becomes
+		 * enabled.
+		 */
+		if (!dsa_port_offloads_bridge_dev(other_dp, bridge_dev))
+			continue;
+
+		if (join) {
+			other_p->pm |= PCR_MATRIX(BIT(port));
+			port_bitmap |= BIT(other_port);
+		} else {
+			other_p->pm &= ~PCR_MATRIX(BIT(port));
+		}
+
+		if (other_p->enable)
+			mt7530_rmw(priv, MT7530_PCR_P(other_port),
+				   PCR_MATRIX_MASK, other_p->pm);
+	}
+
+	/* Add/remove the all other ports to this port matrix. For !join
+	 * (leaving the bridge), only the CPU port will remain in the port matrix
+	 * of this port.
+	 */
+	p->pm = PCR_MATRIX(port_bitmap);
+	if (priv->ports[port].enable)
+		mt7530_rmw(priv, MT7530_PCR_P(port), PCR_MATRIX_MASK, p->pm);
+}
+
 static int
 mt7530_port_pre_bridge_flags(struct dsa_switch *ds, int port,
 			     struct switchdev_brport_flags flags,
@@ -1345,39 +1391,11 @@ mt7530_port_bridge_join(struct dsa_switch *ds, int port,
 			struct dsa_bridge bridge, bool *tx_fwd_offload,
 			struct netlink_ext_ack *extack)
 {
-	struct dsa_port *dp = dsa_to_port(ds, port), *other_dp;
-	struct dsa_port *cpu_dp = dp->cpu_dp;
-	u32 port_bitmap = BIT(cpu_dp->index);
 	struct mt7530_priv *priv = ds->priv;
 
 	mutex_lock(&priv->reg_mutex);
 
-	dsa_switch_for_each_user_port(other_dp, ds) {
-		int other_port = other_dp->index;
-
-		if (dp == other_dp)
-			continue;
-
-		/* Add this port to the port matrix of the other ports in the
-		 * same bridge. If the port is disabled, port matrix is kept
-		 * and not being setup until the port becomes enabled.
-		 */
-		if (!dsa_port_offloads_bridge(other_dp, &bridge))
-			continue;
-
-		if (priv->ports[other_port].enable)
-			mt7530_set(priv, MT7530_PCR_P(other_port),
-				   PCR_MATRIX(BIT(port)));
-		priv->ports[other_port].pm |= PCR_MATRIX(BIT(port));
-
-		port_bitmap |= BIT(other_port);
-	}
-
-	/* Add the all other ports to this port matrix. */
-	if (priv->ports[port].enable)
-		mt7530_rmw(priv, MT7530_PCR_P(port),
-			   PCR_MATRIX_MASK, PCR_MATRIX(port_bitmap));
-	priv->ports[port].pm |= PCR_MATRIX(port_bitmap);
+	mt7530_update_port_member(priv, port, bridge.dev, true);
 
 	/* Set to fallback mode for independent VLAN learning */
 	mt7530_rmw(priv, MT7530_PCR_P(port), PCR_PORT_VLAN_MASK,
@@ -1478,38 +1496,11 @@ static void
 mt7530_port_bridge_leave(struct dsa_switch *ds, int port,
 			 struct dsa_bridge bridge)
 {
-	struct dsa_port *dp = dsa_to_port(ds, port), *other_dp;
-	struct dsa_port *cpu_dp = dp->cpu_dp;
 	struct mt7530_priv *priv = ds->priv;
 
 	mutex_lock(&priv->reg_mutex);
 
-	dsa_switch_for_each_user_port(other_dp, ds) {
-		int other_port = other_dp->index;
-
-		if (dp == other_dp)
-			continue;
-
-		/* Remove this port from the port matrix of the other ports
-		 * in the same bridge. If the port is disabled, port matrix
-		 * is kept and not being setup until the port becomes enabled.
-		 */
-		if (!dsa_port_offloads_bridge(other_dp, &bridge))
-			continue;
-
-		if (priv->ports[other_port].enable)
-			mt7530_clear(priv, MT7530_PCR_P(other_port),
-				     PCR_MATRIX(BIT(port)));
-		priv->ports[other_port].pm &= ~PCR_MATRIX(BIT(port));
-	}
-
-	/* Set the cpu port to be the only one in the port matrix of
-	 * this port.
-	 */
-	if (priv->ports[port].enable)
-		mt7530_rmw(priv, MT7530_PCR_P(port), PCR_MATRIX_MASK,
-			   PCR_MATRIX(BIT(cpu_dp->index)));
-	priv->ports[port].pm = PCR_MATRIX(BIT(cpu_dp->index));
+	mt7530_update_port_member(priv, port, bridge.dev, false);
 
 	/* When a port is removed from the bridge, the port would be set up
 	 * back to the default as is at initial boot which is a VLAN-unaware
-- 
cgit v1.2.3-58-ga151


From 3d49ee2127c26fd2c77944fd2e3168c057f99439 Mon Sep 17 00:00:00 2001
From: Matthias Schiffer <mschiffer@universe-factory.net>
Date: Tue, 18 Jun 2024 09:17:13 +0200
Subject: net: dsa: mt7530: add support for bridge port isolation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Remove a pair of ports from the port matrix when both ports have the
isolated flag set.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Reviewed-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Tested-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/mt7530.c | 18 ++++++++++++++++--
 drivers/net/dsa/mt7530.h |  1 +
 2 files changed, 17 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index 9ce27ce07d77..ec18e68bf3a8 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -1311,6 +1311,7 @@ static void mt7530_update_port_member(struct mt7530_priv *priv, int port,
 	struct dsa_port *cpu_dp = dp->cpu_dp;
 	u32 port_bitmap = BIT(cpu_dp->index);
 	int other_port;
+	bool isolated;
 
 	dsa_switch_for_each_user_port(other_dp, priv->ds) {
 		other_port = other_dp->index;
@@ -1327,7 +1328,9 @@ static void mt7530_update_port_member(struct mt7530_priv *priv, int port,
 		if (!dsa_port_offloads_bridge_dev(other_dp, bridge_dev))
 			continue;
 
-		if (join) {
+		isolated = p->isolated && other_p->isolated;
+
+		if (join && !isolated) {
 			other_p->pm |= PCR_MATRIX(BIT(port));
 			port_bitmap |= BIT(other_port);
 		} else {
@@ -1354,7 +1357,7 @@ mt7530_port_pre_bridge_flags(struct dsa_switch *ds, int port,
 			     struct netlink_ext_ack *extack)
 {
 	if (flags.mask & ~(BR_LEARNING | BR_FLOOD | BR_MCAST_FLOOD |
-			   BR_BCAST_FLOOD))
+			   BR_BCAST_FLOOD | BR_ISOLATED))
 		return -EINVAL;
 
 	return 0;
@@ -1383,6 +1386,17 @@ mt7530_port_bridge_flags(struct dsa_switch *ds, int port,
 		mt7530_rmw(priv, MT753X_MFC, BC_FFP(BIT(port)),
 			   flags.val & BR_BCAST_FLOOD ? BC_FFP(BIT(port)) : 0);
 
+	if (flags.mask & BR_ISOLATED) {
+		struct dsa_port *dp = dsa_to_port(ds, port);
+		struct net_device *bridge_dev = dsa_port_bridge_dev_get(dp);
+
+		priv->ports[port].isolated = !!(flags.val & BR_ISOLATED);
+
+		mutex_lock(&priv->reg_mutex);
+		mt7530_update_port_member(priv, port, bridge_dev, true);
+		mutex_unlock(&priv->reg_mutex);
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/dsa/mt7530.h b/drivers/net/dsa/mt7530.h
index 2ea4e24628c6..28592123070b 100644
--- a/drivers/net/dsa/mt7530.h
+++ b/drivers/net/dsa/mt7530.h
@@ -721,6 +721,7 @@ struct mt7530_fdb {
  */
 struct mt7530_port {
 	bool enable;
+	bool isolated;
 	u32 pm;
 	u16 pvid;
 	struct phylink_pcs *sgmii_pcs;
-- 
cgit v1.2.3-58-ga151


From 94f228ac4f5b8c22f83865e51d3346c2d44d370d Mon Sep 17 00:00:00 2001
From: Lingbo Kong <quic_lingbok@quicinc.com>
Date: Tue, 11 Jun 2024 10:25:50 +0800
Subject: wifi: ath11k: fix ack signal strength calculation

Currently, the calculation of ack signal strength is incorrect.

This is because before calculating the ack signal strength, ath11k need
to determine whether the hardware and firmware support db2dbm. If the
hardware and firmware support db2dbm, do not need to add noise floor,
otherwise, need to add noise floor.

Besides, the value of ack_rssi passed by firmware to ath11k should be a
signed number, so change its type to s8.

After that, "iw wlan0 station dump" show the correct ack signal strength.

Such as:
root@CDCCSTEX0799733-LIN:~# iw wlp88s0 station dump
Station 00:03:7f:12:df:df (on wlp88s0)
        inactive time:  75 ms
        rx bytes:       11599
        rx packets:     99
        tx bytes:       9029
        tx packets:     81
        tx retries:     4
        tx failed:      0
        rx drop misc:   2
        signal:         -16 dBm
        signal avg:     -24 dBm
        tx bitrate:     1560.0 MBit/s VHT-MCS 9 80MHz VHT-NSS 4
        tx duration:    9230 us
        rx bitrate:     1560.0 MBit/s VHT-MCS 9 80MHz VHT-NSS 4
        rx duration:    7201 us
        last ack signal:-23 dBm
        avg ack signal: -22 dBm

Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.30
Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Lingbo Kong <quic_lingbok@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240611022550.59078-1-quic_lingbok@quicinc.com
---
 drivers/net/wireless/ath/ath11k/dp_tx.c  | 16 ++++++++++++----
 drivers/net/wireless/ath/ath11k/dp_tx.h  |  4 ++--
 drivers/net/wireless/ath/ath11k/hal_tx.h |  4 ++--
 3 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.c b/drivers/net/wireless/ath/ath11k/dp_tx.c
index 21819b741701..8522c67baabf 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.c
@@ -353,8 +353,12 @@ ath11k_dp_tx_htt_tx_complete_buf(struct ath11k_base *ab,
 	if (ts->acked) {
 		if (!(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 			info->flags |= IEEE80211_TX_STAT_ACK;
-			info->status.ack_signal = ATH11K_DEFAULT_NOISE_FLOOR +
-						  ts->ack_rssi;
+			info->status.ack_signal = ts->ack_rssi;
+
+			if (!test_bit(WMI_TLV_SERVICE_HW_DB2DBM_CONVERSION_SUPPORT,
+				      ab->wmi_ab.svc_map))
+				info->status.ack_signal += ATH11K_DEFAULT_NOISE_FLOOR;
+
 			info->status.flags |=
 				IEEE80211_TX_STATUS_ACK_SIGNAL_VALID;
 		} else {
@@ -584,8 +588,12 @@ static void ath11k_dp_tx_complete_msdu(struct ath11k *ar,
 	if (ts->status == HAL_WBM_TQM_REL_REASON_FRAME_ACKED &&
 	    !(info->flags & IEEE80211_TX_CTL_NO_ACK)) {
 		info->flags |= IEEE80211_TX_STAT_ACK;
-		info->status.ack_signal = ATH11K_DEFAULT_NOISE_FLOOR +
-					  ts->ack_rssi;
+		info->status.ack_signal = ts->ack_rssi;
+
+		if (!test_bit(WMI_TLV_SERVICE_HW_DB2DBM_CONVERSION_SUPPORT,
+			      ab->wmi_ab.svc_map))
+			info->status.ack_signal += ATH11K_DEFAULT_NOISE_FLOOR;
+
 		info->status.flags |= IEEE80211_TX_STATUS_ACK_SIGNAL_VALID;
 	}
 
diff --git a/drivers/net/wireless/ath/ath11k/dp_tx.h b/drivers/net/wireless/ath/ath11k/dp_tx.h
index 61be2265e09f..795fe3b8fa0d 100644
--- a/drivers/net/wireless/ath/ath11k/dp_tx.h
+++ b/drivers/net/wireless/ath/ath11k/dp_tx.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause-Clear */
 /*
  * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
- * Copyright (c) 2021, 2023 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2021, 2023-2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #ifndef ATH11K_DP_TX_H
@@ -13,7 +13,7 @@
 struct ath11k_dp_htt_wbm_tx_status {
 	u32 msdu_id;
 	bool acked;
-	int ack_rssi;
+	s8 ack_rssi;
 	u16 peer_id;
 };
 
diff --git a/drivers/net/wireless/ath/ath11k/hal_tx.h b/drivers/net/wireless/ath/ath11k/hal_tx.h
index c5e88364afe5..46d17abd808b 100644
--- a/drivers/net/wireless/ath/ath11k/hal_tx.h
+++ b/drivers/net/wireless/ath/ath11k/hal_tx.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause-Clear */
 /*
  * Copyright (c) 2018-2019 The Linux Foundation. All rights reserved.
- * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2022, 2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #ifndef ATH11K_HAL_TX_H
@@ -54,7 +54,7 @@ struct hal_tx_info {
 struct hal_tx_status {
 	enum hal_wbm_rel_src_module buf_rel_source;
 	enum hal_wbm_tqm_rel_reason status;
-	u8 ack_rssi;
+	s8 ack_rssi;
 	u32 flags; /* %HAL_TX_STATUS_FLAGS_ */
 	u32 ppdu_id;
 	u8 try_cnt;
-- 
cgit v1.2.3-58-ga151


From aadeee47e9d7ae98f7116f63fa63fcceffffc65c Mon Sep 17 00:00:00 2001
From: Lingbo Kong <quic_lingbok@quicinc.com>
Date: Sat, 9 Mar 2024 20:11:29 +0800
Subject: wifi: ath11k: modify the calculation of the average signal strength
 in station mode

Currently, the calculation of the average signal strength in station mode
is incorrect.

This is because before calculating the average signal strength, ath11k need
to determine whether the hardware and firmware support db2dbm, if the
hardware and firmware support db2dbm, do not need to add noise floor,
otherwise, need to add noise floor.

Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.23

Signed-off-by: Lingbo Kong <quic_lingbok@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240309121129.5379-1-quic_lingbok@quicinc.com
---
 drivers/net/wireless/ath/ath11k/mac.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index bb56160048ba..779009774567 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -8984,8 +8984,11 @@ static void ath11k_mac_op_sta_statistics(struct ieee80211_hw *hw,
 		sinfo->filled |= BIT_ULL(NL80211_STA_INFO_SIGNAL);
 	}
 
-	sinfo->signal_avg = ewma_avg_rssi_read(&arsta->avg_rssi) +
-		ATH11K_DEFAULT_NOISE_FLOOR;
+	sinfo->signal_avg = ewma_avg_rssi_read(&arsta->avg_rssi);
+
+	if (!db2dbm)
+		sinfo->signal_avg += ATH11K_DEFAULT_NOISE_FLOOR;
+
 	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_SIGNAL_AVG);
 }
 
-- 
cgit v1.2.3-58-ga151


From 59ca26c9a6e844204d25299774caa5441c0e68eb Mon Sep 17 00:00:00 2001
From: Aaradhana Sahu <quic_aarasahu@quicinc.com>
Date: Tue, 11 Jun 2024 08:40:16 +0530
Subject: wifi: ath12k: fix NULL pointer access in ath12k_mac_op_get_survey()

Smatch throws below error

drivers/net/wireless/ath/ath12k/mac.c:8318 ath12k_mac_op_get_survey() error: we previously assumed 'sband' could be null

Currently, we access sband inside the null check of the sband
in ath12k_mac_op_get_survey().

Fix this issue by removing the entire if block, because decrement
idx is unnecessary since there are no more band to test.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Fixes: 70e3be54bbdd ("wifi: ath12k: fix survey dump collection in 6 GHz")
Signed-off-by: Aaradhana Sahu <quic_aarasahu@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240611031017.297927-2-quic_aarasahu@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index bc2da8530651..00a158c22971 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -8291,10 +8291,6 @@ static int ath12k_mac_op_get_survey(struct ieee80211_hw *hw, int idx,
 
 	if (!sband)
 		sband = hw->wiphy->bands[NL80211_BAND_6GHZ];
-	if (!sband || idx >= sband->n_channels) {
-		idx -= sband->n_channels;
-		sband = NULL;
-	}
 
 	if (!sband || idx >= sband->n_channels)
 		return -ENOENT;
-- 
cgit v1.2.3-58-ga151


From 19b77e7c656a3e125319cc3ef347b397cf042bf6 Mon Sep 17 00:00:00 2001
From: Aaradhana Sahu <quic_aarasahu@quicinc.com>
Date: Tue, 11 Jun 2024 08:40:17 +0530
Subject: wifi: ath12k: fix uninitialize symbol error on
 ath12k_peer_assoc_h_he()

Smatch throws following errors

drivers/net/wireless/ath/ath12k/mac.c:1922 ath12k_peer_assoc_h_he() error: uninitialized symbol 'rx_mcs_80'.
drivers/net/wireless/ath/ath12k/mac.c:1922 ath12k_peer_assoc_h_he() error: uninitialized symbol 'rx_mcs_160'.
drivers/net/wireless/ath/ath12k/mac.c:1924 ath12k_peer_assoc_h_he() error: uninitialized symbol 'rx_mcs_80'.

In ath12k_peer_assoc_h_he() rx_mcs_80 and rx_mcs_160 variables
remain uninitialized in the following conditions:
1. Whenever the value of mcs_80 become equal to
   IEEE80211_HE_MCS_NOT_SUPPORTED then rx_mcs_80 remains uninitialized.
2. Whenever phy capability is not supported 160 channel width and
   value of mcs_160 become equal to IEEE80211_HE_MCS_NOT_SUPPORTED
   then rx_mcs_160 remains uninitialized.

Initialize these variables during declaration.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00188-QCAHKSWPL_SILICONZ-1

Signed-off-by: Aaradhana Sahu <quic_aarasahu@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240611031017.297927-3-quic_aarasahu@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 00a158c22971..2a4ca73dfc22 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -2050,7 +2050,9 @@ static void ath12k_peer_assoc_h_he(struct ath12k *ar,
 {
 	const struct ieee80211_sta_he_cap *he_cap = &sta->deflink.he_cap;
 	int i;
-	u8 ampdu_factor, rx_mcs_80, rx_mcs_160, max_nss;
+	u8 ampdu_factor, max_nss;
+	u8 rx_mcs_80 = IEEE80211_HE_MCS_NOT_SUPPORTED;
+	u8 rx_mcs_160 = IEEE80211_HE_MCS_NOT_SUPPORTED;
 	u16 mcs_160_map, mcs_80_map;
 	bool support_160;
 	u16 v;
-- 
cgit v1.2.3-58-ga151


From 3005c63a3673eaf3202b2e0c951e5092539bf758 Mon Sep 17 00:00:00 2001
From: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Date: Wed, 12 Jun 2024 15:53:36 -0700
Subject: wifi: ath12k: fix legacy peer association due to missing HT or 6 GHz
 capabilities

Currently SMPS configuration failed when the Information
Elements (IEs) did not contain HT or 6 GHz capabilities. This
caused legacy peer association to fail as legacy peers do not
have HT or 6 GHz capabilities. Fix this by not returning an
error when SMPS configuration fails due to the absence of HT
or 6 GHz capabilities.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Fixes: f0e61dc7ecf9 ("wifi: ath12k: refactor SMPS configuration")
Reported-by: Aditya Kumar Singh <quic_adisi@quicinc.com>
Reported-by: Zachary Smith <dr.z.smith@gmail.com>
Closes: https://lore.kernel.org/all/CAM=znoFPcXrn5GhDmDmo50Syic3-hXpWvD+vkv8KX5o_ZTo8kQ@mail.gmail.com/
Signed-off-by: Pradeep Kumar Chitrapu <quic_pradeepc@quicinc.com>
Reported-by: Aditya Kumar Singh <quic_adisi@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240612225336.2303119-1-quic_pradeepc@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 2a4ca73dfc22..af19870eb8c0 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -2257,9 +2257,6 @@ static int ath12k_get_smps_from_capa(const struct ieee80211_sta_ht_cap *ht_cap,
 				     const struct ieee80211_he_6ghz_capa *he_6ghz_capa,
 				     int *smps)
 {
-	if (!ht_cap->ht_supported && !he_6ghz_capa->capa)
-		return -EOPNOTSUPP;
-
 	if (ht_cap->ht_supported)
 		*smps = u16_get_bits(ht_cap->cap, IEEE80211_HT_CAP_SM_PS);
 	else
@@ -2279,6 +2276,9 @@ static void ath12k_peer_assoc_h_smps(struct ieee80211_sta *sta,
 	const struct ieee80211_sta_ht_cap *ht_cap = &sta->deflink.ht_cap;
 	int smps;
 
+	if (!ht_cap->ht_supported && !he_6ghz_capa->capa)
+		return;
+
 	if (ath12k_get_smps_from_capa(ht_cap, he_6ghz_capa, &smps))
 		return;
 
@@ -2758,6 +2758,9 @@ static int ath12k_setup_peer_smps(struct ath12k *ar, struct ath12k_vif *arvif,
 {
 	int smps, ret = 0;
 
+	if (!ht_cap->ht_supported && !he_6ghz_capa)
+		return 0;
+
 	ret = ath12k_get_smps_from_capa(ht_cap, he_6ghz_capa, &smps);
 	if (ret < 0)
 		return ret;
-- 
cgit v1.2.3-58-ga151


From db163a463bb93cd3e37e1e7b10b9726fb6f95857 Mon Sep 17 00:00:00 2001
From: Ajith C <quic_ajithc@quicinc.com>
Date: Thu, 13 Jun 2024 11:05:28 +0530
Subject: wifi: ath12k: fix firmware crash due to invalid peer nss

Currently, if the access point receives an association
request containing an Extended HE Capabilities Information
Element with an invalid MCS-NSS, it triggers a firmware
crash.

This issue arises when EHT-PHY capabilities shows support
for a bandwidth and MCS-NSS set for that particular
bandwidth is filled by zeros and due to this, driver obtains
peer_nss as 0 and sending this value to firmware causes
crash.

Address this issue by implementing a validation step for
the peer_nss value before passing it to the firmware. If
the value is greater than zero, proceed with forwarding
it to the firmware. However, if the value is invalid,
reject the association request to prevent potential
firmware crashes.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Signed-off-by: Ajith C <quic_ajithc@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240613053528.2541645-1-quic_ajithc@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index af19870eb8c0..f9caed8ffd63 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -4092,6 +4092,11 @@ static int ath12k_station_assoc(struct ath12k *ar,
 
 	ath12k_peer_assoc_prepare(ar, vif, sta, &peer_arg, reassoc);
 
+	if (peer_arg.peer_nss < 1) {
+		ath12k_warn(ar->ab,
+			    "invalid peer NSS %d\n", peer_arg.peer_nss);
+		return -EINVAL;
+	}
 	ret = ath12k_wmi_send_peer_assoc_cmd(ar, &peer_arg);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to run peer assoc for STA %pM vdev %i: %d\n",
-- 
cgit v1.2.3-58-ga151


From 253ec89c9013b0e61f5c54344df7c065d54934d1 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Thu, 13 Jun 2024 21:08:13 +0530
Subject: wifi: ath12k: fix mbssid max interface advertisement

The Current method for advertising the maximum MBSSID interface count
assumes single radio per wiphy (multi wiphy model). However, this
assumption is incorrect for multi radio per wiphy (single wiphy model).
Therefore, populate the parameter for each radio present in the MAC
abstraction layer (ah). This approach ensure scalability for both single
wiphy and multi wiphy models.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Fixes: 519a545cfee7 ("wifi: ath12k: advertise driver capabilities for MBSSID and EMA")
Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240613153813.3509837-1-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index f9caed8ffd63..17b7fab03d50 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -8942,6 +8942,7 @@ static int ath12k_mac_hw_register(struct ath12k_hw *ah)
 	u32 ht_cap = U32_MAX, antennas_rx = 0, antennas_tx = 0;
 	bool is_6ghz = false, is_raw_mode = false, is_monitor_disable = false;
 	u8 *mac_addr = NULL;
+	u8 mbssid_max_interfaces = 0;
 
 	wiphy->max_ap_assoc_sta = 0;
 
@@ -8985,6 +8986,8 @@ static int ath12k_mac_hw_register(struct ath12k_hw *ah)
 			mac_addr = ar->mac_addr;
 		else
 			mac_addr = ab->mac_addr;
+
+		mbssid_max_interfaces += TARGET_NUM_VDEVS;
 	}
 
 	wiphy->available_antennas_rx = antennas_rx;
@@ -9077,7 +9080,7 @@ static int ath12k_mac_hw_register(struct ath12k_hw *ah)
 	wiphy->iftype_ext_capab = ath12k_iftypes_ext_capa;
 	wiphy->num_iftype_ext_capab = ARRAY_SIZE(ath12k_iftypes_ext_capa);
 
-	wiphy->mbssid_max_interfaces = TARGET_NUM_VDEVS;
+	wiphy->mbssid_max_interfaces = mbssid_max_interfaces;
 	wiphy->ema_max_profile_periodicity = TARGET_EMA_MAX_PROFILE_PERIOD;
 
 	if (is_6ghz) {
-- 
cgit v1.2.3-58-ga151


From e6fed01554fabbe5d3c0172559badcb897eb3254 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Tue, 18 Jun 2024 09:53:44 -0700
Subject: net: arcnet: com20020-isa: add missing MODULE_DESCRIPTION() macro

With ARCH=m68k, make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/arcnet/com20020-isa.o

Add the missing invocation of the MODULE_DESCRIPTION() macro.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240618-md-m68k-drivers-net-arcnet-v1-1-90e42bc58102@quicinc.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/arcnet/com20020-isa.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/arcnet/com20020-isa.c b/drivers/net/arcnet/com20020-isa.c
index 293a621e654c..fef2ac2852a8 100644
--- a/drivers/net/arcnet/com20020-isa.c
+++ b/drivers/net/arcnet/com20020-isa.c
@@ -137,6 +137,7 @@ module_param(backplane, int, 0);
 module_param(clockp, int, 0);
 module_param(clockm, int, 0);
 
+MODULE_DESCRIPTION("ARCnet COM20020 chipset ISA driver");
 MODULE_LICENSE("GPL");
 
 static struct net_device *my_dev;
-- 
cgit v1.2.3-58-ga151


From deb9d5766206988e788037c43a12df892c366ad7 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Tue, 18 Jun 2024 10:26:20 -0700
Subject: net: amd: add missing MODULE_DESCRIPTION() macros

With ARCH=m68k, make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/ethernet/amd/a2065.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/ethernet/amd/ariadne.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/ethernet/amd/atarilance.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/ethernet/amd/hplance.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/ethernet/amd/7990.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/ethernet/amd/mvme147.o
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/ethernet/amd/sun3lance.o

Add the missing invocation of the MODULE_DESCRIPTION() macro to all
files which have a MODULE_LICENSE().

This includes drivers/net/ethernet/amd/lance.c which, although it did
not produce a warning with the m68k allmodconfig configuration, may
cause this warning with other configurations.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240618-md-m68k-drivers-net-ethernet-amd-v1-1-50ee7a9ad50e@quicinc.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/amd/7990.c       | 1 +
 drivers/net/ethernet/amd/a2065.c      | 1 +
 drivers/net/ethernet/amd/ariadne.c    | 1 +
 drivers/net/ethernet/amd/atarilance.c | 1 +
 drivers/net/ethernet/amd/hplance.c    | 1 +
 drivers/net/ethernet/amd/lance.c      | 1 +
 drivers/net/ethernet/amd/mvme147.c    | 1 +
 drivers/net/ethernet/amd/sun3lance.c  | 1 +
 8 files changed, 8 insertions(+)

diff --git a/drivers/net/ethernet/amd/7990.c b/drivers/net/ethernet/amd/7990.c
index ef512cf89abf..27792a52b6cf 100644
--- a/drivers/net/ethernet/amd/7990.c
+++ b/drivers/net/ethernet/amd/7990.c
@@ -667,4 +667,5 @@ void lance_poll(struct net_device *dev)
 EXPORT_SYMBOL_GPL(lance_poll);
 #endif
 
+MODULE_DESCRIPTION("LANCE Ethernet IC generic routines");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/amd/a2065.c b/drivers/net/ethernet/amd/a2065.c
index 68983b717145..1ca26a8c40eb 100644
--- a/drivers/net/ethernet/amd/a2065.c
+++ b/drivers/net/ethernet/amd/a2065.c
@@ -781,4 +781,5 @@ static void __exit a2065_cleanup_module(void)
 module_init(a2065_init_module);
 module_exit(a2065_cleanup_module);
 
+MODULE_DESCRIPTION("Commodore A2065 Ethernet driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/amd/ariadne.c b/drivers/net/ethernet/amd/ariadne.c
index 38153e633231..fa201da567ed 100644
--- a/drivers/net/ethernet/amd/ariadne.c
+++ b/drivers/net/ethernet/amd/ariadne.c
@@ -790,4 +790,5 @@ static void __exit ariadne_cleanup_module(void)
 module_init(ariadne_init_module);
 module_exit(ariadne_cleanup_module);
 
+MODULE_DESCRIPTION("Ariadne Ethernet Driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/amd/atarilance.c b/drivers/net/ethernet/amd/atarilance.c
index 751454d305c6..8c8cc7d0f42d 100644
--- a/drivers/net/ethernet/amd/atarilance.c
+++ b/drivers/net/ethernet/amd/atarilance.c
@@ -79,6 +79,7 @@ static int lance_debug = 1;
 #endif
 module_param(lance_debug, int, 0);
 MODULE_PARM_DESC(lance_debug, "atarilance debug level (0-3)");
+MODULE_DESCRIPTION("Atari LANCE Ethernet driver");
 MODULE_LICENSE("GPL");
 
 /* Print debug messages on probing? */
diff --git a/drivers/net/ethernet/amd/hplance.c b/drivers/net/ethernet/amd/hplance.c
index 055fda11c572..df42294530cb 100644
--- a/drivers/net/ethernet/amd/hplance.c
+++ b/drivers/net/ethernet/amd/hplance.c
@@ -234,4 +234,5 @@ static void __exit hplance_cleanup_module(void)
 module_init(hplance_init_module);
 module_exit(hplance_cleanup_module);
 
+MODULE_DESCRIPTION("HP300 on-board LANCE Ethernet driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/amd/lance.c b/drivers/net/ethernet/amd/lance.c
index 6cf38180cc01..b1e6620ad41d 100644
--- a/drivers/net/ethernet/amd/lance.c
+++ b/drivers/net/ethernet/amd/lance.c
@@ -385,6 +385,7 @@ static void __exit lance_cleanup_module(void)
 }
 module_exit(lance_cleanup_module);
 #endif /* MODULE */
+MODULE_DESCRIPTION("AMD LANCE/PCnet Ethernet driver");
 MODULE_LICENSE("GPL");
 
 
diff --git a/drivers/net/ethernet/amd/mvme147.c b/drivers/net/ethernet/amd/mvme147.c
index 410c7b67eba4..c156566c0906 100644
--- a/drivers/net/ethernet/amd/mvme147.c
+++ b/drivers/net/ethernet/amd/mvme147.c
@@ -178,6 +178,7 @@ static int m147lance_close(struct net_device *dev)
 	return 0;
 }
 
+MODULE_DESCRIPTION("MVME147 LANCE Ethernet driver");
 MODULE_LICENSE("GPL");
 
 static struct net_device *dev_mvme147_lance;
diff --git a/drivers/net/ethernet/amd/sun3lance.c b/drivers/net/ethernet/amd/sun3lance.c
index 246f34c43765..c60df4a21158 100644
--- a/drivers/net/ethernet/amd/sun3lance.c
+++ b/drivers/net/ethernet/amd/sun3lance.c
@@ -74,6 +74,7 @@ static int lance_debug = 1;
 #endif
 module_param(lance_debug, int, 0);
 MODULE_PARM_DESC(lance_debug, "SUN3 Lance debug level (0-3)");
+MODULE_DESCRIPTION("Sun3/Sun3x on-board LANCE Ethernet driver");
 MODULE_LICENSE("GPL");
 
 #define	DPRINTK(n,a) \
-- 
cgit v1.2.3-58-ga151


From 5e736135ad00106bc4b7eff3827162bc1e08b2e3 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Tue, 18 Jun 2024 10:41:54 -0700
Subject: net: ethernet: mac89x0: add missing MODULE_DESCRIPTION() macro

With ARCH=m68k, make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/ethernet/cirrus/mac89x0.o

Add the missing invocation of the MODULE_DESCRIPTION() macro.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240618-md-m68k-drivers-net-ethernet-cirrus-v1-1-07f5bd0b64cb@quicinc.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cirrus/mac89x0.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/cirrus/mac89x0.c b/drivers/net/ethernet/cirrus/mac89x0.c
index 887876f35f10..84b300fee2bb 100644
--- a/drivers/net/ethernet/cirrus/mac89x0.c
+++ b/drivers/net/ethernet/cirrus/mac89x0.c
@@ -554,6 +554,7 @@ static int set_mac_address(struct net_device *dev, void *addr)
 	return 0;
 }
 
+MODULE_DESCRIPTION("Macintosh CS89x0-based Ethernet driver");
 MODULE_LICENSE("GPL");
 
 static void mac89x0_device_remove(struct platform_device *pdev)
-- 
cgit v1.2.3-58-ga151


From 2b0cd6b7270e8fd8c50c3b0b2ebf1372c7b6d384 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Tue, 18 Jun 2024 10:56:28 -0700
Subject: net: smc9194: add missing MODULE_DESCRIPTION() macro

With ARCH=m68k, make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/net/ethernet/smsc/smc9194.o

Add the missing invocation of the MODULE_DESCRIPTION() macro.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240618-md-m68k-drivers-net-ethernet-smsc-v1-1-ad3d7200421e@quicinc.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/smsc/smc9194.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/smsc/smc9194.c b/drivers/net/ethernet/smsc/smc9194.c
index af661c65ffe2..e2e7b1c68563 100644
--- a/drivers/net/ethernet/smsc/smc9194.c
+++ b/drivers/net/ethernet/smsc/smc9194.c
@@ -1501,6 +1501,7 @@ static void smc_set_multicast_list(struct net_device *dev)
 #ifdef MODULE
 
 static struct net_device *devSMC9194;
+MODULE_DESCRIPTION("SMC 9194 Ethernet driver");
 MODULE_LICENSE("GPL");
 
 module_param_hw(io, int, ioport, 0);
-- 
cgit v1.2.3-58-ga151


From c8bd1f7f3e61fc6c562c806045f3ccd2cc819c01 Mon Sep 17 00:00:00 2001
From: Jiri Pirko <jiri@nvidia.com>
Date: Tue, 18 Jun 2024 16:44:56 +0200
Subject: virtio_net: add support for Byte Queue Limits

Add support for Byte Queue Limits (BQL).

Tested on qemu emulated virtio_net device with 1, 2 and 4 queues.
Tested with fq_codel and pfifo_fast. Super netperf with 50 threads is
running in background. Netperf TCP_RR results:

NOBQL FQC 1q:  159.56  159.33  158.50  154.31    agv: 157.925
NOBQL FQC 2q:  184.64  184.96  174.73  174.15    agv: 179.62
NOBQL FQC 4q:  994.46  441.96  416.50  499.56    agv: 588.12
NOBQL PFF 1q:  148.68  148.92  145.95  149.48    agv: 148.2575
NOBQL PFF 2q:  171.86  171.20  170.42  169.42    agv: 170.725
NOBQL PFF 4q: 1505.23 1137.23 2488.70 3507.99    agv: 2159.7875
  BQL FQC 1q: 1332.80 1297.97 1351.41 1147.57    agv: 1282.4375
  BQL FQC 2q:  768.30  817.72  864.43  974.40    agv: 856.2125
  BQL FQC 4q:  945.66  942.68  878.51  822.82    agv: 897.4175
  BQL PFF 1q:  149.69  151.49  149.40  147.47    agv: 149.5125
  BQL PFF 2q: 2059.32  798.74 1844.12  381.80    agv: 1270.995
  BQL PFF 4q: 1871.98 4420.02 4916.59 13268.16   agv: 6119.1875

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://lore.kernel.org/r/20240618144456.1688998-1-jiri@resnulli.us
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 81 ++++++++++++++++++++++++++++++++++--------------
 1 file changed, 57 insertions(+), 24 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 61a57d134544..9f9b86874173 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -47,7 +47,8 @@ module_param(napi_tx, bool, 0644);
 #define VIRTIO_XDP_TX		BIT(0)
 #define VIRTIO_XDP_REDIR	BIT(1)
 
-#define VIRTIO_XDP_FLAG	BIT(0)
+#define VIRTIO_XDP_FLAG		BIT(0)
+#define VIRTIO_ORPHAN_FLAG	BIT(1)
 
 /* RX packet size EWMA. The average packet size is used to determine the packet
  * buffer size when refilling RX rings. As the entire RX ring may be refilled
@@ -85,6 +86,8 @@ struct virtnet_stat_desc {
 struct virtnet_sq_free_stats {
 	u64 packets;
 	u64 bytes;
+	u64 napi_packets;
+	u64 napi_bytes;
 };
 
 struct virtnet_sq_stats {
@@ -506,29 +509,50 @@ static struct xdp_frame *ptr_to_xdp(void *ptr)
 	return (struct xdp_frame *)((unsigned long)ptr & ~VIRTIO_XDP_FLAG);
 }
 
-static void __free_old_xmit(struct send_queue *sq, bool in_napi,
-			    struct virtnet_sq_free_stats *stats)
+static bool is_orphan_skb(void *ptr)
+{
+	return (unsigned long)ptr & VIRTIO_ORPHAN_FLAG;
+}
+
+static void *skb_to_ptr(struct sk_buff *skb, bool orphan)
+{
+	return (void *)((unsigned long)skb | (orphan ? VIRTIO_ORPHAN_FLAG : 0));
+}
+
+static struct sk_buff *ptr_to_skb(void *ptr)
+{
+	return (struct sk_buff *)((unsigned long)ptr & ~VIRTIO_ORPHAN_FLAG);
+}
+
+static void __free_old_xmit(struct send_queue *sq, struct netdev_queue *txq,
+			    bool in_napi, struct virtnet_sq_free_stats *stats)
 {
 	unsigned int len;
 	void *ptr;
 
 	while ((ptr = virtqueue_get_buf(sq->vq, &len)) != NULL) {
-		++stats->packets;
-
 		if (!is_xdp_frame(ptr)) {
-			struct sk_buff *skb = ptr;
+			struct sk_buff *skb = ptr_to_skb(ptr);
 
 			pr_debug("Sent skb %p\n", skb);
 
-			stats->bytes += skb->len;
+			if (is_orphan_skb(ptr)) {
+				stats->packets++;
+				stats->bytes += skb->len;
+			} else {
+				stats->napi_packets++;
+				stats->napi_bytes += skb->len;
+			}
 			napi_consume_skb(skb, in_napi);
 		} else {
 			struct xdp_frame *frame = ptr_to_xdp(ptr);
 
+			stats->packets++;
 			stats->bytes += xdp_get_frame_len(frame);
 			xdp_return_frame(frame);
 		}
 	}
+	netdev_tx_completed_queue(txq, stats->napi_packets, stats->napi_bytes);
 }
 
 /* Converting between virtqueue no. and kernel tx/rx queue no.
@@ -955,21 +979,22 @@ static void virtnet_rq_unmap_free_buf(struct virtqueue *vq, void *buf)
 	virtnet_rq_free_buf(vi, rq, buf);
 }
 
-static void free_old_xmit(struct send_queue *sq, bool in_napi)
+static void free_old_xmit(struct send_queue *sq, struct netdev_queue *txq,
+			  bool in_napi)
 {
 	struct virtnet_sq_free_stats stats = {0};
 
-	__free_old_xmit(sq, in_napi, &stats);
+	__free_old_xmit(sq, txq, in_napi, &stats);
 
 	/* Avoid overhead when no packets have been processed
 	 * happens when called speculatively from start_xmit.
 	 */
-	if (!stats.packets)
+	if (!stats.packets && !stats.napi_packets)
 		return;
 
 	u64_stats_update_begin(&sq->stats.syncp);
-	u64_stats_add(&sq->stats.bytes, stats.bytes);
-	u64_stats_add(&sq->stats.packets, stats.packets);
+	u64_stats_add(&sq->stats.bytes, stats.bytes + stats.napi_bytes);
+	u64_stats_add(&sq->stats.packets, stats.packets + stats.napi_packets);
 	u64_stats_update_end(&sq->stats.syncp);
 }
 
@@ -1003,7 +1028,9 @@ static void check_sq_full_and_disable(struct virtnet_info *vi,
 	 * early means 16 slots are typically wasted.
 	 */
 	if (sq->vq->num_free < 2+MAX_SKB_FRAGS) {
-		netif_stop_subqueue(dev, qnum);
+		struct netdev_queue *txq = netdev_get_tx_queue(dev, qnum);
+
+		netif_tx_stop_queue(txq);
 		u64_stats_update_begin(&sq->stats.syncp);
 		u64_stats_inc(&sq->stats.stop);
 		u64_stats_update_end(&sq->stats.syncp);
@@ -1012,7 +1039,7 @@ static void check_sq_full_and_disable(struct virtnet_info *vi,
 				virtqueue_napi_schedule(&sq->napi, sq->vq);
 		} else if (unlikely(!virtqueue_enable_cb_delayed(sq->vq))) {
 			/* More just got used, free them then recheck. */
-			free_old_xmit(sq, false);
+			free_old_xmit(sq, txq, false);
 			if (sq->vq->num_free >= 2+MAX_SKB_FRAGS) {
 				netif_start_subqueue(dev, qnum);
 				u64_stats_update_begin(&sq->stats.syncp);
@@ -1138,7 +1165,8 @@ static int virtnet_xdp_xmit(struct net_device *dev,
 	}
 
 	/* Free up any pending old buffers before queueing new ones. */
-	__free_old_xmit(sq, false, &stats);
+	__free_old_xmit(sq, netdev_get_tx_queue(dev, sq - vi->sq),
+			false, &stats);
 
 	for (i = 0; i < n; i++) {
 		struct xdp_frame *xdpf = frames[i];
@@ -2313,7 +2341,7 @@ static void virtnet_poll_cleantx(struct receive_queue *rq)
 
 		do {
 			virtqueue_disable_cb(sq->vq);
-			free_old_xmit(sq, true);
+			free_old_xmit(sq, txq, true);
 		} while (unlikely(!virtqueue_enable_cb_delayed(sq->vq)));
 
 		if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS) {
@@ -2412,6 +2440,7 @@ static int virtnet_enable_queue_pair(struct virtnet_info *vi, int qp_index)
 		goto err_xdp_reg_mem_model;
 
 	virtnet_napi_enable(vi->rq[qp_index].vq, &vi->rq[qp_index].napi);
+	netdev_tx_reset_queue(netdev_get_tx_queue(vi->dev, qp_index));
 	virtnet_napi_tx_enable(vi, vi->sq[qp_index].vq, &vi->sq[qp_index].napi);
 
 	return 0;
@@ -2471,7 +2500,7 @@ static int virtnet_poll_tx(struct napi_struct *napi, int budget)
 	txq = netdev_get_tx_queue(vi->dev, index);
 	__netif_tx_lock(txq, raw_smp_processor_id());
 	virtqueue_disable_cb(sq->vq);
-	free_old_xmit(sq, true);
+	free_old_xmit(sq, txq, true);
 
 	if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS) {
 		if (netif_tx_queue_stopped(txq)) {
@@ -2505,7 +2534,7 @@ static int virtnet_poll_tx(struct napi_struct *napi, int budget)
 	return 0;
 }
 
-static int xmit_skb(struct send_queue *sq, struct sk_buff *skb)
+static int xmit_skb(struct send_queue *sq, struct sk_buff *skb, bool orphan)
 {
 	struct virtio_net_hdr_mrg_rxbuf *hdr;
 	const unsigned char *dest = ((struct ethhdr *)skb->data)->h_dest;
@@ -2549,7 +2578,8 @@ static int xmit_skb(struct send_queue *sq, struct sk_buff *skb)
 			return num_sg;
 		num_sg++;
 	}
-	return virtqueue_add_outbuf(sq->vq, sq->sg, num_sg, skb, GFP_ATOMIC);
+	return virtqueue_add_outbuf(sq->vq, sq->sg, num_sg,
+				    skb_to_ptr(skb, orphan), GFP_ATOMIC);
 }
 
 static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
@@ -2559,24 +2589,25 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct send_queue *sq = &vi->sq[qnum];
 	int err;
 	struct netdev_queue *txq = netdev_get_tx_queue(dev, qnum);
-	bool kick = !netdev_xmit_more();
+	bool xmit_more = netdev_xmit_more();
 	bool use_napi = sq->napi.weight;
+	bool kick;
 
 	/* Free up any pending old buffers before queueing new ones. */
 	do {
 		if (use_napi)
 			virtqueue_disable_cb(sq->vq);
 
-		free_old_xmit(sq, false);
+		free_old_xmit(sq, txq, false);
 
-	} while (use_napi && kick &&
+	} while (use_napi && !xmit_more &&
 	       unlikely(!virtqueue_enable_cb_delayed(sq->vq)));
 
 	/* timestamp packet in software */
 	skb_tx_timestamp(skb);
 
 	/* Try to transmit */
-	err = xmit_skb(sq, skb);
+	err = xmit_skb(sq, skb, !use_napi);
 
 	/* This should not happen! */
 	if (unlikely(err)) {
@@ -2598,7 +2629,9 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	check_sq_full_and_disable(vi, dev, sq);
 
-	if (kick || netif_xmit_stopped(txq)) {
+	kick = use_napi ? __netdev_tx_sent_queue(txq, skb->len, xmit_more) :
+			  !xmit_more || netif_xmit_stopped(txq);
+	if (kick) {
 		if (virtqueue_kick_prepare(sq->vq) && virtqueue_notify(sq->vq)) {
 			u64_stats_update_begin(&sq->stats.syncp);
 			u64_stats_inc(&sq->stats.kicks);
-- 
cgit v1.2.3-58-ga151


From 89f5e607772b4d6e467adf8bbe530a5036abb415 Mon Sep 17 00:00:00 2001
From: Lukasz Majewski <lukma@denx.de>
Date: Tue, 18 Jun 2024 14:58:17 +0200
Subject: net: hsr: cosmetic: Remove extra white space

This change just removes extra (i.e. not needed) white space in
prp_drop_frame() function.

No functional changes.

Signed-off-by: Lukasz Majewski <lukma@denx.de>
Reviewed-by: Hangbin Liu <liuhangbin@gmail.com>
Link: https://lore.kernel.org/r/20240618125817.1111070-1-lukma@denx.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/hsr/hsr_forward.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/hsr/hsr_forward.c b/net/hsr/hsr_forward.c
index 960ef386bc3a..b38060246e62 100644
--- a/net/hsr/hsr_forward.c
+++ b/net/hsr/hsr_forward.c
@@ -421,9 +421,9 @@ static int hsr_xmit(struct sk_buff *skb, struct hsr_port *port,
 bool prp_drop_frame(struct hsr_frame_info *frame, struct hsr_port *port)
 {
 	return ((frame->port_rcv->type == HSR_PT_SLAVE_A &&
-		 port->type ==  HSR_PT_SLAVE_B) ||
+		 port->type == HSR_PT_SLAVE_B) ||
 		(frame->port_rcv->type == HSR_PT_SLAVE_B &&
-		 port->type ==  HSR_PT_SLAVE_A));
+		 port->type == HSR_PT_SLAVE_A));
 }
 
 bool hsr_drop_frame(struct hsr_frame_info *frame, struct hsr_port *port)
-- 
cgit v1.2.3-58-ga151


From 39fa294f580a3be47fc2b502c143dec76737c8e6 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Tue, 18 Jun 2024 13:34:40 +0200
Subject: mlxsw: pci: Split NAPI setup/teardown into two steps

mlxsw_pci_cq_napi_setup() includes both NAPI initialization and
enablement, similar to teardown function. Next patches will add support
for page pool in mlxsw driver, then we use NAPI instance for page pool.

Page pool initialization should be done before NAPI enablement, same for
page pool destruction which should be done after NAPI disablement.

As preparation, split NAPI setup/teardown into two steps, then page pool
setup will be done between the phases.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/8dbf37e859f07247498fca17109b8858ff2b0498.1718709196.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/pci.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index bf66d996e32e..fe6df5a92616 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -832,13 +832,10 @@ static void mlxsw_pci_cq_napi_setup(struct mlxsw_pci_queue *q,
 			       mlxsw_pci_napi_poll_cq_rx);
 		break;
 	}
-
-	napi_enable(&q->u.cq.napi);
 }
 
 static void mlxsw_pci_cq_napi_teardown(struct mlxsw_pci_queue *q)
 {
-	napi_disable(&q->u.cq.napi);
 	netif_napi_del(&q->u.cq.napi);
 }
 
@@ -875,6 +872,7 @@ static int mlxsw_pci_cq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 	if (err)
 		return err;
 	mlxsw_pci_cq_napi_setup(q, mlxsw_pci_cq_type(mlxsw_pci, q));
+	napi_enable(&q->u.cq.napi);
 	mlxsw_pci_queue_doorbell_consumer_ring(mlxsw_pci, q);
 	mlxsw_pci_queue_doorbell_arm_consumer_ring(mlxsw_pci, q);
 	return 0;
@@ -883,6 +881,7 @@ static int mlxsw_pci_cq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 static void mlxsw_pci_cq_fini(struct mlxsw_pci *mlxsw_pci,
 			      struct mlxsw_pci_queue *q)
 {
+	napi_disable(&q->u.cq.napi);
 	mlxsw_pci_cq_napi_teardown(q);
 	mlxsw_cmd_hw2sw_cq(mlxsw_pci->core, q->num);
 }
-- 
cgit v1.2.3-58-ga151


From 7555b7f3385fb0cf5416cf38b4f7bd2bc1312f23 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Tue, 18 Jun 2024 13:34:41 +0200
Subject: mlxsw: pci: Store CQ pointer as part of RDQ structure

Next patches will add support for page pool in mlxsw driver. Page pool will
be used to allocate buffers for RDQ and will use NAPI instance of the
appropriate CQ (RDQ is mapped 1:1 to CQ).

To allow pool initialization as part of CQ init, when NAPI is initialized,
page_pool structure will be as part of CQ structure. Later, the allocations
for RDQ will be done from the pool in the appropriate CQ. To allow access
to the appropriate pool, set CQ pointer as part of RDQ initialization.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/d60918ca1e142a554af1df9c1152cdac83854a3b.1718709196.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/pci.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index fe6df5a92616..8751da09a6c1 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -92,6 +92,9 @@ struct mlxsw_pci_queue {
 		struct {
 			struct tasklet_struct tasklet;
 		} eq;
+		struct {
+			struct mlxsw_pci_queue *cq;
+		} rdq;
 	} u;
 };
 
@@ -434,6 +437,7 @@ static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 
 	cq = mlxsw_pci_cq_get(mlxsw_pci, cq_num);
 	cq->u.cq.dq = q;
+	q->u.rdq.cq = cq;
 
 	mlxsw_pci_queue_doorbell_producer_ring(mlxsw_pci, q);
 
@@ -455,6 +459,7 @@ rollback:
 		elem_info = mlxsw_pci_queue_elem_info_get(q, i);
 		mlxsw_pci_rdq_skb_free(mlxsw_pci, elem_info);
 	}
+	q->u.rdq.cq = NULL;
 	cq->u.cq.dq = NULL;
 	mlxsw_cmd_hw2sw_rdq(mlxsw_pci->core, q->num);
 
-- 
cgit v1.2.3-58-ga151


From 5642c6a086939bd1c0bba692f3a4afe4df668a50 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Tue, 18 Jun 2024 13:34:42 +0200
Subject: mlxsw: pci: Initialize page pool per CQ

Next patch will use page pool to allocate buffers for RDQ. Initialize
page pool for each CQ, which is mapped 1:1 to RDQ. Page pool for each Rx
queue enhances Rx side performance by reclaiming buffers back to each queue
specific pool.

When only one NAPI instance is the consumer of pages from page pool, it is
recommended to pass it as part of 'page_pool_params', then page pool APIs
will be done without special locks. mlxsw driver holds NAPI instance per
CQ, so add page pool per CQ and use the existing NAPI instance.

For now, pages are not allocated from the pool, next patch will use it.

Some notes regarding 'page_pool_params':
* Use PP_FLAG_DMA_MAP to allow page pool handles DMA mapping, for now
  do not use sync flag, as only the device writes to this memory and we
  read it only when it finishes writing there. This will probably be
  changed when we will support XDP.
* Define 'order' according to maximum MTU and take into account software
  overhead. Some round up are done, which means that we allocate more pages
  than we really need. This can be improved later by using fragmented
  buffers.
* Use pool_size = MLXSW_PCI_WQE_COUNT. This will be the size of 'ptr_ring',
  and should be the maximum amount of packets that page pool will allocate
  memory for. In our case, this is the queue size, defined as
  MLXSW_PCI_WQE_COUNT.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/02e5856ae7c572d4293ce6bb92c286ee6cfec800.1718709196.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/Kconfig |  1 +
 drivers/net/ethernet/mellanox/mlxsw/pci.c   | 60 ++++++++++++++++++++++++++++-
 2 files changed, 60 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/Kconfig b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
index a510bf2cff2f..74f7e27b490f 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
@@ -33,6 +33,7 @@ config MLXSW_CORE_THERMAL
 config MLXSW_PCI
 	tristate "PCI bus implementation for Mellanox Technologies Switch ASICs"
 	depends on PCI && HAS_IOMEM && MLXSW_CORE
+	select PAGE_POOL
 	default m
 	help
 	  This is PCI bus implementation for Mellanox Technologies Switch ASICs.
diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index 8751da09a6c1..2486f0fde5d9 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -13,6 +13,7 @@
 #include <linux/if_vlan.h>
 #include <linux/log2.h>
 #include <linux/string.h>
+#include <net/page_pool/helpers.h>
 
 #include "pci_hw.h"
 #include "pci.h"
@@ -88,6 +89,7 @@ struct mlxsw_pci_queue {
 			enum mlxsw_pci_cqe_v v;
 			struct mlxsw_pci_queue *dq;
 			struct napi_struct napi;
+			struct page_pool *page_pool;
 		} cq;
 		struct {
 			struct tasklet_struct tasklet;
@@ -338,6 +340,12 @@ static void mlxsw_pci_sdq_fini(struct mlxsw_pci *mlxsw_pci,
 	mlxsw_cmd_hw2sw_sdq(mlxsw_pci->core, q->num);
 }
 
+#define MLXSW_PCI_SKB_HEADROOM (NET_SKB_PAD + NET_IP_ALIGN)
+
+#define MLXSW_PCI_RX_BUF_SW_OVERHEAD		\
+		(MLXSW_PCI_SKB_HEADROOM +	\
+		SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
+
 static int mlxsw_pci_wqe_frag_map(struct mlxsw_pci *mlxsw_pci, char *wqe,
 				  int index, char *frag_data, size_t frag_len,
 				  int direction)
@@ -844,9 +852,47 @@ static void mlxsw_pci_cq_napi_teardown(struct mlxsw_pci_queue *q)
 	netif_napi_del(&q->u.cq.napi);
 }
 
+static int mlxsw_pci_cq_page_pool_init(struct mlxsw_pci_queue *q,
+				       enum mlxsw_pci_cq_type cq_type)
+{
+	struct page_pool_params pp_params = {};
+	struct mlxsw_pci *mlxsw_pci = q->pci;
+	struct page_pool *page_pool;
+	u32 max_pkt_size;
+
+	if (cq_type != MLXSW_PCI_CQ_RDQ)
+		return 0;
+
+	max_pkt_size = MLXSW_PORT_MAX_MTU + MLXSW_PCI_RX_BUF_SW_OVERHEAD;
+	pp_params.order = get_order(max_pkt_size);
+	pp_params.flags = PP_FLAG_DMA_MAP;
+	pp_params.pool_size = MLXSW_PCI_WQE_COUNT;
+	pp_params.nid = dev_to_node(&mlxsw_pci->pdev->dev);
+	pp_params.dev = &mlxsw_pci->pdev->dev;
+	pp_params.napi = &q->u.cq.napi;
+	pp_params.dma_dir = DMA_FROM_DEVICE;
+
+	page_pool = page_pool_create(&pp_params);
+	if (IS_ERR(page_pool))
+		return PTR_ERR(page_pool);
+
+	q->u.cq.page_pool = page_pool;
+	return 0;
+}
+
+static void mlxsw_pci_cq_page_pool_fini(struct mlxsw_pci_queue *q,
+					enum mlxsw_pci_cq_type cq_type)
+{
+	if (cq_type != MLXSW_PCI_CQ_RDQ)
+		return;
+
+	page_pool_destroy(q->u.cq.page_pool);
+}
+
 static int mlxsw_pci_cq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 			     struct mlxsw_pci_queue *q)
 {
+	enum mlxsw_pci_cq_type cq_type = mlxsw_pci_cq_type(mlxsw_pci, q);
 	int i;
 	int err;
 
@@ -876,17 +922,29 @@ static int mlxsw_pci_cq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 	err = mlxsw_cmd_sw2hw_cq(mlxsw_pci->core, mbox, q->num);
 	if (err)
 		return err;
-	mlxsw_pci_cq_napi_setup(q, mlxsw_pci_cq_type(mlxsw_pci, q));
+	mlxsw_pci_cq_napi_setup(q, cq_type);
+
+	err = mlxsw_pci_cq_page_pool_init(q, cq_type);
+	if (err)
+		goto err_page_pool_init;
+
 	napi_enable(&q->u.cq.napi);
 	mlxsw_pci_queue_doorbell_consumer_ring(mlxsw_pci, q);
 	mlxsw_pci_queue_doorbell_arm_consumer_ring(mlxsw_pci, q);
 	return 0;
+
+err_page_pool_init:
+	mlxsw_pci_cq_napi_teardown(q);
+	return err;
 }
 
 static void mlxsw_pci_cq_fini(struct mlxsw_pci *mlxsw_pci,
 			      struct mlxsw_pci_queue *q)
 {
+	enum mlxsw_pci_cq_type cq_type = mlxsw_pci_cq_type(mlxsw_pci, q);
+
 	napi_disable(&q->u.cq.napi);
+	mlxsw_pci_cq_page_pool_fini(q, cq_type);
 	mlxsw_pci_cq_napi_teardown(q);
 	mlxsw_cmd_hw2sw_cq(mlxsw_pci->core, q->num);
 }
-- 
cgit v1.2.3-58-ga151


From b5b60bb491b26b0e5961e021b2988562eaa6a8c7 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Tue, 18 Jun 2024 13:34:43 +0200
Subject: mlxsw: pci: Use page pool for Rx buffers allocation

As part of driver init, all Rx queues are filled with buffers for
hardware usage. Later, when a packet is received, a new buffer should be
allocated to be used by hardware instead of the received buffer.
Packet's processing time includes allocation time, which can be improved
using page pool.

Using page pool, DMA mapping is done only for first allocation of buffers.
As subsequent buffers allocation avoid DMA mapping, it results in
performance improvement. The purpose of page pool is to allocate pages fast
from cache without locking. This lockless guarantee naturally comes from
running under a NAPI.

Use page pool to allocate the data buffer only, so hardware will use it to
fill the packet. At completion time, attach the data buffer (now filled
with packet payload) to new SKB which is allocated around the received
buffer. SKB building at completion time prevents cache miss for each
packet, as now the SKB is allocated right before packets will be handled by
networking stack.

Page pool for each Rx queue enhances Rx side performance by reclaiming
buffers back to each queue specific pool. This change significantly
improves driver performance, CPU can handle about 345% of the packets per
second it previously handled.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/1cf788a8f43c70aae6d526018ef77becb27ad6d3.1718709196.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/pci.c | 103 +++++++++++++++++++-----------
 1 file changed, 64 insertions(+), 39 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index 2486f0fde5d9..4b34db2ae8f0 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -62,6 +62,7 @@ struct mlxsw_pci_mem_item {
 };
 
 struct mlxsw_pci_queue_elem_info {
+	struct page *page;
 	char *elem; /* pointer to actual dma mapped element mem chunk */
 	union {
 		struct {
@@ -346,6 +347,19 @@ static void mlxsw_pci_sdq_fini(struct mlxsw_pci *mlxsw_pci,
 		(MLXSW_PCI_SKB_HEADROOM +	\
 		SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
 
+static void
+mlxsw_pci_wqe_rx_frag_set(struct mlxsw_pci *mlxsw_pci, struct page *page,
+			  char *wqe, int index, size_t frag_len)
+{
+	dma_addr_t mapaddr;
+
+	mapaddr = page_pool_get_dma_addr(page);
+	mapaddr += MLXSW_PCI_SKB_HEADROOM;
+
+	mlxsw_pci_wqe_address_set(wqe, index, mapaddr);
+	mlxsw_pci_wqe_byte_count_set(wqe, index, frag_len);
+}
+
 static int mlxsw_pci_wqe_frag_map(struct mlxsw_pci *mlxsw_pci, char *wqe,
 				  int index, char *frag_data, size_t frag_len,
 				  int direction)
@@ -375,43 +389,46 @@ static void mlxsw_pci_wqe_frag_unmap(struct mlxsw_pci *mlxsw_pci, char *wqe,
 	dma_unmap_single(&pdev->dev, mapaddr, frag_len, direction);
 }
 
-static int mlxsw_pci_rdq_skb_alloc(struct mlxsw_pci *mlxsw_pci,
-				   struct mlxsw_pci_queue_elem_info *elem_info,
-				   gfp_t gfp)
+static struct sk_buff *mlxsw_pci_rdq_build_skb(struct page *page,
+					       u16 byte_count)
+{
+	void *data = page_address(page);
+	unsigned int allocated_size;
+	struct sk_buff *skb;
+
+	allocated_size = page_size(page);
+	skb = napi_build_skb(data, allocated_size);
+	if (unlikely(!skb))
+		return ERR_PTR(-ENOMEM);
+
+	skb_reserve(skb, MLXSW_PCI_SKB_HEADROOM);
+	skb_put(skb, byte_count);
+	return skb;
+}
+
+static int mlxsw_pci_rdq_page_alloc(struct mlxsw_pci_queue *q,
+				    struct mlxsw_pci_queue_elem_info *elem_info)
 {
+	struct mlxsw_pci_queue *cq = q->u.rdq.cq;
 	size_t buf_len = MLXSW_PORT_MAX_MTU;
 	char *wqe = elem_info->elem;
-	struct sk_buff *skb;
-	int err;
+	struct page *page;
 
-	skb = __netdev_alloc_skb_ip_align(NULL, buf_len, gfp);
-	if (!skb)
+	page = page_pool_dev_alloc_pages(cq->u.cq.page_pool);
+	if (unlikely(!page))
 		return -ENOMEM;
 
-	err = mlxsw_pci_wqe_frag_map(mlxsw_pci, wqe, 0, skb->data,
-				     buf_len, DMA_FROM_DEVICE);
-	if (err)
-		goto err_frag_map;
-
-	elem_info->u.rdq.skb = skb;
+	mlxsw_pci_wqe_rx_frag_set(q->pci, page, wqe, 0, buf_len);
+	elem_info->page = page;
 	return 0;
-
-err_frag_map:
-	dev_kfree_skb_any(skb);
-	return err;
 }
 
-static void mlxsw_pci_rdq_skb_free(struct mlxsw_pci *mlxsw_pci,
-				   struct mlxsw_pci_queue_elem_info *elem_info)
+static void mlxsw_pci_rdq_page_free(struct mlxsw_pci_queue *q,
+				    struct mlxsw_pci_queue_elem_info *elem_info)
 {
-	struct sk_buff *skb;
-	char *wqe;
+	struct mlxsw_pci_queue *cq = q->u.rdq.cq;
 
-	skb = elem_info->u.rdq.skb;
-	wqe = elem_info->elem;
-
-	mlxsw_pci_wqe_frag_unmap(mlxsw_pci, wqe, 0, DMA_FROM_DEVICE);
-	dev_kfree_skb_any(skb);
+	page_pool_put_page(cq->u.cq.page_pool, elem_info->page, -1, false);
 }
 
 static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
@@ -452,7 +469,7 @@ static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 	for (i = 0; i < q->count; i++) {
 		elem_info = mlxsw_pci_queue_elem_info_producer_get(q);
 		BUG_ON(!elem_info);
-		err = mlxsw_pci_rdq_skb_alloc(mlxsw_pci, elem_info, GFP_KERNEL);
+		err = mlxsw_pci_rdq_page_alloc(q, elem_info);
 		if (err)
 			goto rollback;
 		/* Everything is set up, ring doorbell to pass elem to HW */
@@ -465,7 +482,7 @@ static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 rollback:
 	for (i--; i >= 0; i--) {
 		elem_info = mlxsw_pci_queue_elem_info_get(q, i);
-		mlxsw_pci_rdq_skb_free(mlxsw_pci, elem_info);
+		mlxsw_pci_rdq_page_free(q, elem_info);
 	}
 	q->u.rdq.cq = NULL;
 	cq->u.cq.dq = NULL;
@@ -483,7 +500,7 @@ static void mlxsw_pci_rdq_fini(struct mlxsw_pci *mlxsw_pci,
 	mlxsw_cmd_hw2sw_rdq(mlxsw_pci->core, q->num);
 	for (i = 0; i < q->count; i++) {
 		elem_info = mlxsw_pci_queue_elem_info_get(q, i);
-		mlxsw_pci_rdq_skb_free(mlxsw_pci, elem_info);
+		mlxsw_pci_rdq_page_free(q, elem_info);
 	}
 }
 
@@ -618,26 +635,38 @@ static void mlxsw_pci_cqe_rdq_handle(struct mlxsw_pci *mlxsw_pci,
 {
 	struct pci_dev *pdev = mlxsw_pci->pdev;
 	struct mlxsw_pci_queue_elem_info *elem_info;
+	struct mlxsw_pci_queue *cq = q->u.rdq.cq;
 	struct mlxsw_rx_info rx_info = {};
-	char wqe[MLXSW_PCI_WQE_SIZE];
 	struct sk_buff *skb;
+	struct page *page;
 	u16 byte_count;
 	int err;
 
 	elem_info = mlxsw_pci_queue_elem_info_consumer_get(q);
-	skb = elem_info->u.rdq.skb;
-	memcpy(wqe, elem_info->elem, MLXSW_PCI_WQE_SIZE);
 
 	if (q->consumer_counter++ != consumer_counter_limit)
 		dev_dbg_ratelimited(&pdev->dev, "Consumer counter does not match limit in RDQ\n");
 
-	err = mlxsw_pci_rdq_skb_alloc(mlxsw_pci, elem_info, GFP_ATOMIC);
+	byte_count = mlxsw_pci_cqe_byte_count_get(cqe);
+	if (mlxsw_pci_cqe_crc_get(cqe_v, cqe))
+		byte_count -= ETH_FCS_LEN;
+
+	page = elem_info->page;
+
+	err = mlxsw_pci_rdq_page_alloc(q, elem_info);
 	if (err) {
-		dev_err_ratelimited(&pdev->dev, "Failed to alloc skb for RDQ\n");
+		dev_err_ratelimited(&pdev->dev, "Failed to alloc page\n");
+		goto out;
+	}
+
+	skb = mlxsw_pci_rdq_build_skb(page, byte_count);
+	if (IS_ERR(skb)) {
+		dev_err_ratelimited(&pdev->dev, "Failed to build skb for RDQ\n");
+		page_pool_recycle_direct(cq->u.cq.page_pool, page);
 		goto out;
 	}
 
-	mlxsw_pci_wqe_frag_unmap(mlxsw_pci, wqe, 0, DMA_FROM_DEVICE);
+	skb_mark_for_recycle(skb);
 
 	if (mlxsw_pci_cqe_lag_get(cqe_v, cqe)) {
 		rx_info.is_lag = true;
@@ -670,10 +699,6 @@ static void mlxsw_pci_cqe_rdq_handle(struct mlxsw_pci *mlxsw_pci,
 
 	mlxsw_pci_skb_cb_ts_set(mlxsw_pci, skb, cqe_v, cqe);
 
-	byte_count = mlxsw_pci_cqe_byte_count_get(cqe);
-	if (mlxsw_pci_cqe_crc_get(cqe_v, cqe))
-		byte_count -= ETH_FCS_LEN;
-	skb_put(skb, byte_count);
 	mlxsw_core_skb_receive(mlxsw_pci->core, skb, &rx_info);
 
 out:
-- 
cgit v1.2.3-58-ga151


From 0f3cd437a1d88b2a77dfb146c88a297bea974ceb Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Tue, 18 Jun 2024 13:34:44 +0200
Subject: mlxsw: pci: Optimize data buffer access

Before accessing data buffer, call net_prefetch() to load it into the
cache. This change improves driver performance, CPU can handle about
7.1% more packets per second.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/1fa07c510890866a6f201163ab7e78890ba28b3b.1718709196.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/pci.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index 4b34db2ae8f0..6f41747e8a76 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -396,6 +396,7 @@ static struct sk_buff *mlxsw_pci_rdq_build_skb(struct page *page,
 	unsigned int allocated_size;
 	struct sk_buff *skb;
 
+	net_prefetch(data);
 	allocated_size = page_size(page);
 	skb = napi_build_skb(data, allocated_size);
 	if (unlikely(!skb))
-- 
cgit v1.2.3-58-ga151


From e8441b1f6b64e7157b9e7777a2f7b79c2215cba1 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Tue, 18 Jun 2024 13:34:45 +0200
Subject: mlxsw: pci: Do not store SKB for RDQ elements

The previous patch used page pool to allocate buffers for RDQ. With this
change, 'elem_info->u.rdq.skb' is not used anymore, as we do not allocate
SKB before getting the packet, we hold page pointer and build the SKB
around it once packet is received.

Remove the union and store SKB pointer for SDQ only.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/23a531008936dc9a1a298643fb1e4f9a7b8e6eb3.1718709196.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/pci.c | 19 +++++++------------
 1 file changed, 7 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index 6f41747e8a76..b530b6a80ab7 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -64,14 +64,9 @@ struct mlxsw_pci_mem_item {
 struct mlxsw_pci_queue_elem_info {
 	struct page *page;
 	char *elem; /* pointer to actual dma mapped element mem chunk */
-	union {
-		struct {
-			struct sk_buff *skb;
-		} sdq;
-		struct {
-			struct sk_buff *skb;
-		} rdq;
-	} u;
+	struct {
+		struct sk_buff *skb;
+	} sdq;
 };
 
 struct mlxsw_pci_queue {
@@ -557,8 +552,8 @@ static void mlxsw_pci_cqe_sdq_handle(struct mlxsw_pci *mlxsw_pci,
 
 	spin_lock(&q->lock);
 	elem_info = mlxsw_pci_queue_elem_info_consumer_get(q);
-	tx_info = mlxsw_skb_cb(elem_info->u.sdq.skb)->tx_info;
-	skb = elem_info->u.sdq.skb;
+	tx_info = mlxsw_skb_cb(elem_info->sdq.skb)->tx_info;
+	skb = elem_info->sdq.skb;
 	wqe = elem_info->elem;
 	for (i = 0; i < MLXSW_PCI_WQE_SG_ENTRIES; i++)
 		mlxsw_pci_wqe_frag_unmap(mlxsw_pci, wqe, i, DMA_TO_DEVICE);
@@ -573,7 +568,7 @@ static void mlxsw_pci_cqe_sdq_handle(struct mlxsw_pci *mlxsw_pci,
 
 	if (skb)
 		dev_kfree_skb_any(skb);
-	elem_info->u.sdq.skb = NULL;
+	elem_info->sdq.skb = NULL;
 
 	if (q->consumer_counter++ != consumer_counter_limit)
 		dev_dbg_ratelimited(&pdev->dev, "Consumer counter does not match limit in SDQ\n");
@@ -2007,7 +2002,7 @@ static int mlxsw_pci_skb_transmit(void *bus_priv, struct sk_buff *skb,
 		goto unlock;
 	}
 	mlxsw_skb_cb(skb)->tx_info = *tx_info;
-	elem_info->u.sdq.skb = skb;
+	elem_info->sdq.skb = skb;
 
 	wqe = elem_info->elem;
 	mlxsw_pci_wqe_c_set(wqe, 1); /* always report completion */
-- 
cgit v1.2.3-58-ga151


From d94ae6415becdf437d58e60d1f153af39dddb247 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Tue, 18 Jun 2024 13:34:46 +0200
Subject: mlxsw: pci: Use napi_consume_skb() to free SKB as part of Tx
 completion

Currently, as part of Tx completion, the driver calls dev_kfree_skb_any()
to free the SKB. For this flow, the correct function is napi_consume_skb().
This function and dev_consume_skb_any() were added to be used for consumed
SKBs, which were not dropped, so the skb:kfree_skb tracepoint is not
triggered, and we can get better diagnostics about dropped packets.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Link: https://lore.kernel.org/r/a9f9f3dc884c0d1be4bd4c9d72030c88c7ac004f.1718709196.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/pci.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index b530b6a80ab7..cb043379c01c 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -541,7 +541,7 @@ static void mlxsw_pci_cqe_sdq_handle(struct mlxsw_pci *mlxsw_pci,
 				     struct mlxsw_pci_queue *q,
 				     u16 consumer_counter_limit,
 				     enum mlxsw_pci_cqe_v cqe_v,
-				     char *cqe)
+				     char *cqe, int budget)
 {
 	struct pci_dev *pdev = mlxsw_pci->pdev;
 	struct mlxsw_pci_queue_elem_info *elem_info;
@@ -567,7 +567,7 @@ static void mlxsw_pci_cqe_sdq_handle(struct mlxsw_pci *mlxsw_pci,
 	}
 
 	if (skb)
-		dev_kfree_skb_any(skb);
+		napi_consume_skb(skb, budget);
 	elem_info->sdq.skb = NULL;
 
 	if (q->consumer_counter++ != consumer_counter_limit)
@@ -819,7 +819,7 @@ static int mlxsw_pci_napi_poll_cq_tx(struct napi_struct *napi, int budget)
 		mlxsw_pci_queue_doorbell_consumer_ring(mlxsw_pci, q);
 
 		mlxsw_pci_cqe_sdq_handle(mlxsw_pci, sdq,
-					 wqe_counter, q->u.cq.v, ncqe);
+					 wqe_counter, q->u.cq.v, ncqe, budget);
 
 		work_done++;
 	}
-- 
cgit v1.2.3-58-ga151


From 4aaa49a282ad581e313afde8a423301cf4fe2ecc Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Tue, 18 Jun 2024 17:32:50 -0700
Subject: ionic: remove missed doorbell per-queue timer

Remove the timer-per-queue mechanics from the missed doorbell
check in preparation for the new missed doorbell fix.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240619003257.6138-2-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_dev.c  |  4 ---
 drivers/net/ethernet/pensando/ionic/ionic_lif.c  | 36 +++++-------------------
 drivers/net/ethernet/pensando/ionic/ionic_lif.h  |  2 --
 drivers/net/ethernet/pensando/ionic/ionic_txrx.c | 22 ++++++---------
 4 files changed, 15 insertions(+), 49 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
index 874499337132..89b4310f244c 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
@@ -703,10 +703,6 @@ void ionic_q_post(struct ionic_queue *q, bool ring_doorbell)
 				 q->dbval | q->head_idx);
 
 		q->dbell_jiffies = jiffies;
-
-		if (q_to_qcq(q)->napi_qcq)
-			mod_timer(&q_to_qcq(q)->napi_qcq->napi_deadline,
-				  jiffies + IONIC_NAPI_DEADLINE);
 	}
 }
 
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
index b8fdfb355386..1f02b32755fc 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
@@ -213,13 +213,6 @@ void ionic_link_status_check_request(struct ionic_lif *lif, bool can_sleep)
 	}
 }
 
-static void ionic_napi_deadline(struct timer_list *timer)
-{
-	struct ionic_qcq *qcq = container_of(timer, struct ionic_qcq, napi_deadline);
-
-	napi_schedule(&qcq->napi);
-}
-
 static irqreturn_t ionic_isr(int irq, void *data)
 {
 	struct napi_struct *napi = data;
@@ -343,7 +336,6 @@ static int ionic_qcq_disable(struct ionic_lif *lif, struct ionic_qcq *qcq, int f
 		synchronize_irq(qcq->intr.vector);
 		irq_set_affinity_hint(qcq->intr.vector, NULL);
 		napi_disable(&qcq->napi);
-		del_timer_sync(&qcq->napi_deadline);
 	}
 
 	/* If there was a previous fw communcation error, don't bother with
@@ -478,7 +470,6 @@ static void ionic_link_qcq_interrupts(struct ionic_qcq *src_qcq,
 {
 	n_qcq->intr.vector = src_qcq->intr.vector;
 	n_qcq->intr.index = src_qcq->intr.index;
-	n_qcq->napi_qcq = src_qcq->napi_qcq;
 }
 
 static int ionic_alloc_qcq_interrupt(struct ionic_lif *lif, struct ionic_qcq *qcq)
@@ -832,11 +823,8 @@ static int ionic_lif_txq_init(struct ionic_lif *lif, struct ionic_qcq *qcq)
 	q->dbell_deadline = IONIC_TX_DOORBELL_DEADLINE;
 	q->dbell_jiffies = jiffies;
 
-	if (test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state)) {
+	if (test_bit(IONIC_LIF_F_SPLIT_INTR, lif->state))
 		netif_napi_add(lif->netdev, &qcq->napi, ionic_tx_napi);
-		qcq->napi_qcq = qcq;
-		timer_setup(&qcq->napi_deadline, ionic_napi_deadline, 0);
-	}
 
 	qcq->flags |= IONIC_QCQ_F_INITED;
 
@@ -909,9 +897,6 @@ static int ionic_lif_rxq_init(struct ionic_lif *lif, struct ionic_qcq *qcq)
 	else
 		netif_napi_add(lif->netdev, &qcq->napi, ionic_txrx_napi);
 
-	qcq->napi_qcq = qcq;
-	timer_setup(&qcq->napi_deadline, ionic_napi_deadline, 0);
-
 	qcq->flags |= IONIC_QCQ_F_INITED;
 
 	return 0;
@@ -1166,7 +1151,6 @@ static int ionic_adminq_napi(struct napi_struct *napi, int budget)
 	struct ionic_dev *idev = &lif->ionic->idev;
 	unsigned long irqflags;
 	unsigned int flags = 0;
-	bool resched = false;
 	int rx_work = 0;
 	int tx_work = 0;
 	int n_work = 0;
@@ -1203,15 +1187,12 @@ static int ionic_adminq_napi(struct napi_struct *napi, int budget)
 		ionic_intr_credits(idev->intr_ctrl, intr->index, credits, flags);
 	}
 
-	if (!a_work && ionic_adminq_poke_doorbell(&lif->adminqcq->q))
-		resched = true;
-	if (lif->hwstamp_rxq && !rx_work && ionic_rxq_poke_doorbell(&lif->hwstamp_rxq->q))
-		resched = true;
-	if (lif->hwstamp_txq && !tx_work && ionic_txq_poke_doorbell(&lif->hwstamp_txq->q))
-		resched = true;
-	if (resched)
-		mod_timer(&lif->adminqcq->napi_deadline,
-			  jiffies + IONIC_NAPI_DEADLINE);
+	if (!a_work)
+		ionic_adminq_poke_doorbell(&lif->adminqcq->q);
+	if (lif->hwstamp_rxq && !rx_work)
+		ionic_rxq_poke_doorbell(&lif->hwstamp_rxq->q);
+	if (lif->hwstamp_txq && !tx_work)
+		ionic_txq_poke_doorbell(&lif->hwstamp_txq->q);
 
 	return work_done;
 }
@@ -3502,9 +3483,6 @@ static int ionic_lif_adminq_init(struct ionic_lif *lif)
 
 	netif_napi_add(lif->netdev, &qcq->napi, ionic_adminq_napi);
 
-	qcq->napi_qcq = qcq;
-	timer_setup(&qcq->napi_deadline, ionic_napi_deadline, 0);
-
 	napi_enable(&qcq->napi);
 
 	if (qcq->flags & IONIC_QCQ_F_INTR) {
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.h b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
index 08f4266fe2aa..a029206c0bc8 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
@@ -84,11 +84,9 @@ struct ionic_qcq {
 	u32 cmb_pgid;
 	u32 cmb_order;
 	struct dim dim;
-	struct timer_list napi_deadline;
 	struct ionic_queue q;
 	struct ionic_cq cq;
 	struct napi_struct napi;
-	struct ionic_qcq *napi_qcq;
 	struct ionic_intr_info intr;
 	struct dentry *dentry;
 };
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
index 9e6dee2fc1d4..a4e923376484 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
@@ -868,9 +868,6 @@ void ionic_rx_fill(struct ionic_queue *q)
 
 	q->dbell_deadline = IONIC_RX_MIN_DOORBELL_DEADLINE;
 	q->dbell_jiffies = jiffies;
-
-	mod_timer(&q_to_qcq(q)->napi_qcq->napi_deadline,
-		  jiffies + IONIC_NAPI_DEADLINE);
 }
 
 void ionic_rx_empty(struct ionic_queue *q)
@@ -953,8 +950,8 @@ int ionic_tx_napi(struct napi_struct *napi, int budget)
 				   work_done, flags);
 	}
 
-	if (!work_done && ionic_txq_poke_doorbell(&qcq->q))
-		mod_timer(&qcq->napi_deadline, jiffies + IONIC_NAPI_DEADLINE);
+	if (!work_done)
+		ionic_txq_poke_doorbell(&qcq->q);
 
 	return work_done;
 }
@@ -996,8 +993,8 @@ int ionic_rx_napi(struct napi_struct *napi, int budget)
 				   work_done, flags);
 	}
 
-	if (!work_done && ionic_rxq_poke_doorbell(&qcq->q))
-		mod_timer(&qcq->napi_deadline, jiffies + IONIC_NAPI_DEADLINE);
+	if (!work_done)
+		ionic_rxq_poke_doorbell(&qcq->q);
 
 	return work_done;
 }
@@ -1010,7 +1007,6 @@ int ionic_txrx_napi(struct napi_struct *napi, int budget)
 	struct ionic_qcq *txqcq;
 	struct ionic_lif *lif;
 	struct ionic_cq *txcq;
-	bool resched = false;
 	u32 rx_work_done = 0;
 	u32 tx_work_done = 0;
 	u32 flags = 0;
@@ -1042,12 +1038,10 @@ int ionic_txrx_napi(struct napi_struct *napi, int budget)
 				   tx_work_done + rx_work_done, flags);
 	}
 
-	if (!rx_work_done && ionic_rxq_poke_doorbell(&rxqcq->q))
-		resched = true;
-	if (!tx_work_done && ionic_txq_poke_doorbell(&txqcq->q))
-		resched = true;
-	if (resched)
-		mod_timer(&rxqcq->napi_deadline, jiffies + IONIC_NAPI_DEADLINE);
+	if (!rx_work_done)
+		ionic_rxq_poke_doorbell(&rxqcq->q);
+	if (!tx_work_done)
+		ionic_txq_poke_doorbell(&txqcq->q);
 
 	return rx_work_done;
 }
-- 
cgit v1.2.3-58-ga151


From d458d4b4fd4353e71a095f6c76e1271dc2fee33a Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@amd.com>
Date: Tue, 18 Jun 2024 17:32:51 -0700
Subject: ionic: Keep interrupt affinity up to date

Currently the driver either sets the initial interrupt affinity for its
adminq and tx/rx queues on probe or resets it on various
down/up/reconfigure flows. If any user and/or user process
(i.e. irqbalance) changes IRQ affinity for any of the driver's interrupts
that will be reset to driver defaults whenever any down/up/reconfigure
operation happens. This is incorrect and is fixed by making 2 changes:

1. Allocate an array of cpumasks that's only allocated on probe and
   destroyed on remove.
2. Update the cpumask(s) for interrupts that are in use by registering
   for affinity notifiers.

Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240619003257.6138-3-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic.h     |  1 +
 drivers/net/ethernet/pensando/ionic/ionic_dev.h |  4 +-
 drivers/net/ethernet/pensando/ionic/ionic_lif.c | 85 +++++++++++++++++++++++--
 3 files changed, 81 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic.h b/drivers/net/ethernet/pensando/ionic/ionic.h
index 2ccc2c2a06e3..438172cfb170 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic.h
@@ -54,6 +54,7 @@ struct ionic {
 	unsigned int nrxqs_per_lif;
 	unsigned int nintrs;
 	DECLARE_BITMAP(intrs, IONIC_INTR_CTRL_REGS_MAX);
+	cpumask_var_t *affinity_masks;
 	struct work_struct nb_work;
 	struct notifier_block nb;
 	struct rw_semaphore vf_op_lock;	/* lock for VF operations */
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.h b/drivers/net/ethernet/pensando/ionic/ionic_dev.h
index f30eee4a5a80..7dbd3b8b0e36 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_dev.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.h
@@ -280,9 +280,9 @@ struct ionic_intr_info {
 	u64 rearm_count;
 	unsigned int index;
 	unsigned int vector;
-	unsigned int cpu;
 	u32 dim_coal_hw;
-	cpumask_t affinity_mask;
+	cpumask_var_t *affinity_mask;
+	struct irq_affinity_notify aff_notify;
 };
 
 struct ionic_cq {
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
index 1f02b32755fc..46cb143b5941 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
@@ -265,6 +265,18 @@ static void ionic_intr_free(struct ionic *ionic, int index)
 		clear_bit(index, ionic->intrs);
 }
 
+static void ionic_irq_aff_notify(struct irq_affinity_notify *notify,
+				 const cpumask_t *mask)
+{
+	struct ionic_intr_info *intr = container_of(notify, struct ionic_intr_info, aff_notify);
+
+	cpumask_copy(*intr->affinity_mask, mask);
+}
+
+static void ionic_irq_aff_release(struct kref __always_unused *ref)
+{
+}
+
 static int ionic_qcq_enable(struct ionic_qcq *qcq)
 {
 	struct ionic_queue *q = &qcq->q;
@@ -299,8 +311,10 @@ static int ionic_qcq_enable(struct ionic_qcq *qcq)
 
 	if (qcq->flags & IONIC_QCQ_F_INTR) {
 		napi_enable(&qcq->napi);
+		irq_set_affinity_notifier(qcq->intr.vector,
+					  &qcq->intr.aff_notify);
 		irq_set_affinity_hint(qcq->intr.vector,
-				      &qcq->intr.affinity_mask);
+				      *qcq->intr.affinity_mask);
 		ionic_intr_mask(idev->intr_ctrl, qcq->intr.index,
 				IONIC_INTR_MASK_CLEAR);
 	}
@@ -334,6 +348,7 @@ static int ionic_qcq_disable(struct ionic_lif *lif, struct ionic_qcq *qcq, int f
 		ionic_intr_mask(idev->intr_ctrl, qcq->intr.index,
 				IONIC_INTR_MASK_SET);
 		synchronize_irq(qcq->intr.vector);
+		irq_set_affinity_notifier(qcq->intr.vector, NULL);
 		irq_set_affinity_hint(qcq->intr.vector, NULL);
 		napi_disable(&qcq->napi);
 	}
@@ -474,6 +489,7 @@ static void ionic_link_qcq_interrupts(struct ionic_qcq *src_qcq,
 
 static int ionic_alloc_qcq_interrupt(struct ionic_lif *lif, struct ionic_qcq *qcq)
 {
+	cpumask_var_t *affinity_mask;
 	int err;
 
 	if (!(qcq->flags & IONIC_QCQ_F_INTR)) {
@@ -505,10 +521,19 @@ static int ionic_alloc_qcq_interrupt(struct ionic_lif *lif, struct ionic_qcq *qc
 	}
 
 	/* try to get the irq on the local numa node first */
-	qcq->intr.cpu = cpumask_local_spread(qcq->intr.index,
-					     dev_to_node(lif->ionic->dev));
-	if (qcq->intr.cpu != -1)
-		cpumask_set_cpu(qcq->intr.cpu, &qcq->intr.affinity_mask);
+	affinity_mask = &lif->ionic->affinity_masks[qcq->intr.index];
+	if (cpumask_empty(*affinity_mask)) {
+		unsigned int cpu;
+
+		cpu = cpumask_local_spread(qcq->intr.index,
+					   dev_to_node(lif->ionic->dev));
+		if (cpu != -1)
+			cpumask_set_cpu(cpu, *affinity_mask);
+	}
+
+	qcq->intr.affinity_mask = affinity_mask;
+	qcq->intr.aff_notify.notify = ionic_irq_aff_notify;
+	qcq->intr.aff_notify.release = ionic_irq_aff_release;
 
 	netdev_dbg(lif->netdev, "%s: Interrupt index %d\n", qcq->q.name, qcq->intr.index);
 	return 0;
@@ -3120,6 +3145,44 @@ err_out:
 	return err;
 }
 
+static int ionic_affinity_masks_alloc(struct ionic *ionic)
+{
+	cpumask_var_t *affinity_masks;
+	int nintrs = ionic->nintrs;
+	int i;
+
+	affinity_masks = kcalloc(nintrs, sizeof(cpumask_var_t), GFP_KERNEL);
+	if (!affinity_masks)
+		return -ENOMEM;
+
+	for (i = 0; i < nintrs; i++) {
+		if (!zalloc_cpumask_var_node(&affinity_masks[i], GFP_KERNEL,
+					     dev_to_node(ionic->dev)))
+			goto err_out;
+	}
+
+	ionic->affinity_masks = affinity_masks;
+
+	return 0;
+
+err_out:
+	for (--i; i >= 0; i--)
+		free_cpumask_var(affinity_masks[i]);
+	kfree(affinity_masks);
+
+	return -ENOMEM;
+}
+
+static void ionic_affinity_masks_free(struct ionic *ionic)
+{
+	int i;
+
+	for (i = 0; i < ionic->nintrs; i++)
+		free_cpumask_var(ionic->affinity_masks[i]);
+	kfree(ionic->affinity_masks);
+	ionic->affinity_masks = NULL;
+}
+
 int ionic_lif_alloc(struct ionic *ionic)
 {
 	struct device *dev = ionic->dev;
@@ -3211,11 +3274,15 @@ int ionic_lif_alloc(struct ionic *ionic)
 
 	ionic_debugfs_add_lif(lif);
 
+	err = ionic_affinity_masks_alloc(ionic);
+	if (err)
+		goto err_out_free_lif_info;
+
 	/* allocate control queues and txrx queue arrays */
 	ionic_lif_queue_identify(lif);
 	err = ionic_qcqs_alloc(lif);
 	if (err)
-		goto err_out_free_lif_info;
+		goto err_out_free_affinity_masks;
 
 	/* allocate rss indirection table */
 	tbl_sz = le16_to_cpu(lif->ionic->ident.lif.eth.rss_ind_tbl_sz);
@@ -3237,6 +3304,8 @@ int ionic_lif_alloc(struct ionic *ionic)
 
 err_out_free_qcqs:
 	ionic_qcqs_free(lif);
+err_out_free_affinity_masks:
+	ionic_affinity_masks_free(lif->ionic);
 err_out_free_lif_info:
 	dma_free_coherent(dev, lif->info_sz, lif->info, lif->info_pa);
 	lif->info = NULL;
@@ -3410,6 +3479,8 @@ void ionic_lif_free(struct ionic_lif *lif)
 	if (!test_bit(IONIC_LIF_F_FW_RESET, lif->state))
 		ionic_lif_reset(lif);
 
+	ionic_affinity_masks_free(lif->ionic);
+
 	/* free lif info */
 	kfree(lif->identity);
 	dma_free_coherent(dev, lif->info_sz, lif->info, lif->info_pa);
@@ -3487,7 +3558,7 @@ static int ionic_lif_adminq_init(struct ionic_lif *lif)
 
 	if (qcq->flags & IONIC_QCQ_F_INTR) {
 		irq_set_affinity_hint(qcq->intr.vector,
-				      &qcq->intr.affinity_mask);
+				      *qcq->intr.affinity_mask);
 		ionic_intr_mask(idev->intr_ctrl, qcq->intr.index,
 				IONIC_INTR_MASK_CLEAR);
 	}
-- 
cgit v1.2.3-58-ga151


From 9e25450da7006cd6f425248a5b38dad4adb3c981 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Tue, 18 Jun 2024 17:32:52 -0700
Subject: ionic: add private workqueue per-device

Instead of using the system's default workqueue, add a private
workqueue for the device to use for its little jobs.  This is
to better support the new work items we will be adding in the
next patches for PF and VF specific jobs, without inundating
the system workqueue in a couple of customer cases where our
devices get scaled out to 100-200 VFs.

Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240619003257.6138-4-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic.h      |  1 +
 drivers/net/ethernet/pensando/ionic/ionic_dev.c  | 21 +++++++++++++++++----
 drivers/net/ethernet/pensando/ionic/ionic_lif.c  | 14 +++++++-------
 drivers/net/ethernet/pensando/ionic/ionic_lif.h  |  2 +-
 drivers/net/ethernet/pensando/ionic/ionic_main.c |  2 +-
 5 files changed, 27 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic.h b/drivers/net/ethernet/pensando/ionic/ionic.h
index 438172cfb170..df29c977a702 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic.h
@@ -47,6 +47,7 @@ struct ionic {
 	struct ionic_dev_bar bars[IONIC_BARS_MAX];
 	unsigned int num_bars;
 	struct ionic_identity ident;
+	struct workqueue_struct *wq;
 	struct ionic_lif *lif;
 	unsigned int nnqs_per_lif;
 	unsigned int neqs_per_lif;
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
index 89b4310f244c..342863fd0b16 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
@@ -43,11 +43,11 @@ static void ionic_watchdog_cb(struct timer_list *t)
 
 		work->type = IONIC_DW_TYPE_RX_MODE;
 		netdev_dbg(lif->netdev, "deferred: rx_mode\n");
-		ionic_lif_deferred_enqueue(&lif->deferred, work);
+		ionic_lif_deferred_enqueue(lif, work);
 	}
 }
 
-static void ionic_watchdog_init(struct ionic *ionic)
+static int ionic_watchdog_init(struct ionic *ionic)
 {
 	struct ionic_dev *idev = &ionic->idev;
 
@@ -63,6 +63,15 @@ static void ionic_watchdog_init(struct ionic *ionic)
 	idev->fw_status_ready = true;
 	idev->fw_generation = IONIC_FW_STS_F_GENERATION &
 			      ioread8(&idev->dev_info_regs->fw_status);
+
+	ionic->wq = alloc_workqueue("%s-wq", WQ_UNBOUND, 0,
+				    dev_name(ionic->dev));
+	if (!ionic->wq) {
+		dev_err(ionic->dev, "alloc_workqueue failed");
+		return -ENOMEM;
+	}
+
+	return 0;
 }
 
 void ionic_init_devinfo(struct ionic *ionic)
@@ -94,6 +103,7 @@ int ionic_dev_setup(struct ionic *ionic)
 	struct device *dev = ionic->dev;
 	int size;
 	u32 sig;
+	int err;
 
 	/* BAR0: dev_cmd and interrupts */
 	if (num_bars < 1) {
@@ -129,7 +139,9 @@ int ionic_dev_setup(struct ionic *ionic)
 		return -EFAULT;
 	}
 
-	ionic_watchdog_init(ionic);
+	err = ionic_watchdog_init(ionic);
+	if (err)
+		return err;
 
 	idev->db_pages = bar->vaddr;
 	idev->phy_db_pages = bar->bus_addr;
@@ -161,6 +173,7 @@ void ionic_dev_teardown(struct ionic *ionic)
 	idev->phy_cmb_pages = 0;
 	idev->cmb_npages = 0;
 
+	destroy_workqueue(ionic->wq);
 	mutex_destroy(&idev->cmb_inuse_lock);
 }
 
@@ -273,7 +286,7 @@ do_check_time:
 			if (work) {
 				work->type = IONIC_DW_TYPE_LIF_RESET;
 				work->fw_status = fw_status_ready;
-				ionic_lif_deferred_enqueue(&lif->deferred, work);
+				ionic_lif_deferred_enqueue(lif, work);
 			}
 		}
 	}
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
index 46cb143b5941..2882517f26fb 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
@@ -126,13 +126,13 @@ static void ionic_lif_deferred_work(struct work_struct *work)
 	} while (true);
 }
 
-void ionic_lif_deferred_enqueue(struct ionic_deferred *def,
+void ionic_lif_deferred_enqueue(struct ionic_lif *lif,
 				struct ionic_deferred_work *work)
 {
-	spin_lock_bh(&def->lock);
-	list_add_tail(&work->list, &def->list);
-	spin_unlock_bh(&def->lock);
-	schedule_work(&def->work);
+	spin_lock_bh(&lif->deferred.lock);
+	list_add_tail(&work->list, &lif->deferred.list);
+	spin_unlock_bh(&lif->deferred.lock);
+	queue_work(lif->ionic->wq, &lif->deferred.work);
 }
 
 static void ionic_link_status_check(struct ionic_lif *lif)
@@ -207,7 +207,7 @@ void ionic_link_status_check_request(struct ionic_lif *lif, bool can_sleep)
 		}
 
 		work->type = IONIC_DW_TYPE_LINK_STATUS;
-		ionic_lif_deferred_enqueue(&lif->deferred, work);
+		ionic_lif_deferred_enqueue(lif, work);
 	} else {
 		ionic_link_status_check(lif);
 	}
@@ -1389,7 +1389,7 @@ static void ionic_ndo_set_rx_mode(struct net_device *netdev)
 	}
 	work->type = IONIC_DW_TYPE_RX_MODE;
 	netdev_dbg(lif->netdev, "deferred: rx_mode\n");
-	ionic_lif_deferred_enqueue(&lif->deferred, work);
+	ionic_lif_deferred_enqueue(lif, work);
 }
 
 static __le64 ionic_netdev_features_to_nic(netdev_features_t features)
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.h b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
index a029206c0bc8..e4a5ae70793e 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
@@ -331,7 +331,7 @@ static inline bool ionic_txq_hwstamp_enabled(struct ionic_queue *q)
 void ionic_link_status_check_request(struct ionic_lif *lif, bool can_sleep);
 void ionic_get_stats64(struct net_device *netdev,
 		       struct rtnl_link_stats64 *ns);
-void ionic_lif_deferred_enqueue(struct ionic_deferred *def,
+void ionic_lif_deferred_enqueue(struct ionic_lif *lif,
 				struct ionic_deferred_work *work);
 int ionic_lif_alloc(struct ionic *ionic);
 int ionic_lif_init(struct ionic_lif *lif);
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_main.c b/drivers/net/ethernet/pensando/ionic/ionic_main.c
index c1259324b0be..0f817c3f92d8 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_main.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_main.c
@@ -287,7 +287,7 @@ bool ionic_notifyq_service(struct ionic_cq *cq)
 				clear_bit(IONIC_LIF_F_FW_STOPPING, lif->state);
 			} else {
 				work->type = IONIC_DW_TYPE_LIF_RESET;
-				ionic_lif_deferred_enqueue(&lif->deferred, work);
+				ionic_lif_deferred_enqueue(lif, work);
 			}
 		}
 		break;
-- 
cgit v1.2.3-58-ga151


From 4ded136c78f8ae17d140b8aae3cd4459a6827577 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Tue, 18 Jun 2024 17:32:53 -0700
Subject: ionic: add work item for missed-doorbell check

Add the first queued work for checking on the missed doorbell.
This is a delayed work item that reschedules itself every cycle
starting at probe.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240619003257.6138-5-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic.h        |  1 +
 .../net/ethernet/pensando/ionic/ionic_bus_pci.c    |  2 +
 drivers/net/ethernet/pensando/ionic/ionic_dev.c    | 66 ++++++++++++++++++++++
 drivers/net/ethernet/pensando/ionic/ionic_dev.h    |  3 +-
 drivers/net/ethernet/pensando/ionic/ionic_lif.c    |  2 +
 5 files changed, 73 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic.h b/drivers/net/ethernet/pensando/ionic/ionic.h
index df29c977a702..106ee5b2ceff 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic.h
@@ -56,6 +56,7 @@ struct ionic {
 	unsigned int nintrs;
 	DECLARE_BITMAP(intrs, IONIC_INTR_CTRL_REGS_MAX);
 	cpumask_var_t *affinity_masks;
+	struct delayed_work doorbell_check_dwork;
 	struct work_struct nb_work;
 	struct notifier_block nb;
 	struct rw_semaphore vf_op_lock;	/* lock for VF operations */
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
index a7146d50f814..f362e76756df 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
@@ -377,6 +377,7 @@ static int ionic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	mod_timer(&ionic->watchdog_timer,
 		  round_jiffies(jiffies + ionic->watchdog_period));
+	ionic_queue_doorbell_check(ionic, IONIC_NAPI_DEADLINE);
 
 	return 0;
 
@@ -411,6 +412,7 @@ static void ionic_remove(struct pci_dev *pdev)
 		if (test_and_clear_bit(IONIC_LIF_F_FW_RESET, ionic->lif->state))
 			set_bit(IONIC_LIF_F_FW_STOPPING, ionic->lif->state);
 
+		cancel_delayed_work_sync(&ionic->doorbell_check_dwork);
 		ionic_lif_unregister(ionic->lif);
 		ionic_devlink_unregister(ionic);
 		ionic_lif_deinit(ionic->lif);
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
index 342863fd0b16..5d5990e7376e 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
@@ -47,6 +47,61 @@ static void ionic_watchdog_cb(struct timer_list *t)
 	}
 }
 
+static void ionic_napi_schedule_do_softirq(struct napi_struct *napi)
+{
+	local_bh_disable();
+	napi_schedule(napi);
+	local_bh_enable();
+}
+
+static int ionic_get_preferred_cpu(struct ionic *ionic,
+				   struct ionic_intr_info *intr)
+{
+	int cpu;
+
+	cpu = cpumask_first_and(*intr->affinity_mask, cpu_online_mask);
+	if (cpu >= nr_cpu_ids)
+		cpu = cpumask_local_spread(0, dev_to_node(ionic->dev));
+
+	return cpu;
+}
+
+static void ionic_doorbell_check_dwork(struct work_struct *work)
+{
+	struct ionic *ionic = container_of(work, struct ionic,
+					   doorbell_check_dwork.work);
+	struct ionic_lif *lif = ionic->lif;
+
+	mutex_lock(&lif->queue_lock);
+
+	if (test_bit(IONIC_LIF_F_FW_STOPPING, lif->state) ||
+	    test_bit(IONIC_LIF_F_FW_RESET, lif->state)) {
+		mutex_unlock(&lif->queue_lock);
+		return;
+	}
+
+	ionic_napi_schedule_do_softirq(&lif->adminqcq->napi);
+
+	if (test_bit(IONIC_LIF_F_UP, lif->state)) {
+		int i;
+
+		for (i = 0; i < lif->nxqs; i++) {
+			ionic_napi_schedule_do_softirq(&lif->txqcqs[i]->napi);
+			ionic_napi_schedule_do_softirq(&lif->rxqcqs[i]->napi);
+		}
+
+		if (lif->hwstamp_txq &&
+		    lif->hwstamp_txq->flags & IONIC_QCQ_F_INTR)
+			ionic_napi_schedule_do_softirq(&lif->hwstamp_txq->napi);
+		if (lif->hwstamp_rxq &&
+		    lif->hwstamp_rxq->flags & IONIC_QCQ_F_INTR)
+			ionic_napi_schedule_do_softirq(&lif->hwstamp_rxq->napi);
+	}
+	mutex_unlock(&lif->queue_lock);
+
+	ionic_queue_doorbell_check(ionic, IONIC_NAPI_DEADLINE);
+}
+
 static int ionic_watchdog_init(struct ionic *ionic)
 {
 	struct ionic_dev *idev = &ionic->idev;
@@ -70,10 +125,21 @@ static int ionic_watchdog_init(struct ionic *ionic)
 		dev_err(ionic->dev, "alloc_workqueue failed");
 		return -ENOMEM;
 	}
+	INIT_DELAYED_WORK(&ionic->doorbell_check_dwork,
+			  ionic_doorbell_check_dwork);
 
 	return 0;
 }
 
+void ionic_queue_doorbell_check(struct ionic *ionic, int delay)
+{
+	int cpu;
+
+	cpu = ionic_get_preferred_cpu(ionic, &ionic->lif->adminqcq->intr);
+	queue_delayed_work_on(cpu, ionic->wq, &ionic->doorbell_check_dwork,
+			      delay);
+}
+
 void ionic_init_devinfo(struct ionic *ionic)
 {
 	struct ionic_dev *idev = &ionic->idev;
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.h b/drivers/net/ethernet/pensando/ionic/ionic_dev.h
index 7dbd3b8b0e36..d87e6020cfb1 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_dev.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.h
@@ -28,7 +28,7 @@
 #define IONIC_DEV_INFO_REG_COUNT	32
 #define IONIC_DEV_CMD_REG_COUNT		32
 
-#define IONIC_NAPI_DEADLINE		(HZ / 200)	/* 5ms */
+#define IONIC_NAPI_DEADLINE		(HZ)		/* 1 sec */
 #define IONIC_ADMIN_DOORBELL_DEADLINE	(HZ / 2)	/* 500ms */
 #define IONIC_TX_DOORBELL_DEADLINE	(HZ / 100)	/* 10ms */
 #define IONIC_RX_MIN_DOORBELL_DEADLINE	(HZ / 100)	/* 10ms */
@@ -386,6 +386,7 @@ bool ionic_q_is_posted(struct ionic_queue *q, unsigned int pos);
 
 int ionic_heartbeat_check(struct ionic *ionic);
 bool ionic_is_fw_running(struct ionic_dev *idev);
+void ionic_queue_doorbell_check(struct ionic *ionic, int delay);
 
 bool ionic_adminq_poke_doorbell(struct ionic_queue *q);
 bool ionic_txq_poke_doorbell(struct ionic_queue *q);
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
index 2882517f26fb..267c592d5864 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
@@ -1191,6 +1191,7 @@ static int ionic_adminq_napi(struct napi_struct *napi, int budget)
 	if (lif->adminqcq && lif->adminqcq->flags & IONIC_QCQ_F_INITED)
 		a_work = ionic_cq_service(&lif->adminqcq->cq, budget,
 					  ionic_adminq_service, NULL, NULL);
+
 	spin_unlock_irqrestore(&lif->adminq_lock, irqflags);
 
 	if (lif->hwstamp_rxq)
@@ -3406,6 +3407,7 @@ int ionic_restart_lif(struct ionic_lif *lif)
 	clear_bit(IONIC_LIF_F_FW_RESET, lif->state);
 	ionic_link_status_check_request(lif, CAN_SLEEP);
 	netif_device_attach(lif->netdev);
+	ionic_queue_doorbell_check(ionic, IONIC_NAPI_DEADLINE);
 
 	return 0;
 
-- 
cgit v1.2.3-58-ga151


From d7f9bc68591803a9bca34847b76b2fc8f12e5dfd Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Tue, 18 Jun 2024 17:32:54 -0700
Subject: ionic: add per-queue napi_schedule for doorbell check

Add a work item for each queue that will be run on the queue's
preferred cpu and will schedule another napi.  This napi is
run in case the device missed a doorbell and didn't process
a packet.  This is a problem for the Elba asic that happens
very rarely.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240619003257.6138-6-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_dev.c | 23 +++++++++++++++++++++--
 drivers/net/ethernet/pensando/ionic/ionic_dev.h |  1 +
 drivers/net/ethernet/pensando/ionic/ionic_lif.c |  2 ++
 drivers/net/ethernet/pensando/ionic/ionic_lif.h |  1 +
 4 files changed, 25 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
index 5d5990e7376e..56e7c120d492 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
@@ -54,6 +54,13 @@ static void ionic_napi_schedule_do_softirq(struct napi_struct *napi)
 	local_bh_enable();
 }
 
+void ionic_doorbell_napi_work(struct work_struct *work)
+{
+	struct ionic_qcq *qcq = container_of(work, struct ionic_qcq,
+					     doorbell_napi_work);
+	ionic_napi_schedule_do_softirq(&qcq->napi);
+}
+
 static int ionic_get_preferred_cpu(struct ionic *ionic,
 				   struct ionic_intr_info *intr)
 {
@@ -66,6 +73,18 @@ static int ionic_get_preferred_cpu(struct ionic *ionic,
 	return cpu;
 }
 
+static void ionic_queue_dbell_napi_work(struct ionic *ionic,
+					struct ionic_qcq *qcq)
+{
+	int cpu;
+
+	if (!(qcq->flags & IONIC_QCQ_F_INTR))
+		return;
+
+	cpu = ionic_get_preferred_cpu(ionic, &qcq->intr);
+	queue_work_on(cpu, ionic->wq, &qcq->doorbell_napi_work);
+}
+
 static void ionic_doorbell_check_dwork(struct work_struct *work)
 {
 	struct ionic *ionic = container_of(work, struct ionic,
@@ -86,8 +105,8 @@ static void ionic_doorbell_check_dwork(struct work_struct *work)
 		int i;
 
 		for (i = 0; i < lif->nxqs; i++) {
-			ionic_napi_schedule_do_softirq(&lif->txqcqs[i]->napi);
-			ionic_napi_schedule_do_softirq(&lif->rxqcqs[i]->napi);
+			ionic_queue_dbell_napi_work(ionic, lif->txqcqs[i]);
+			ionic_queue_dbell_napi_work(ionic, lif->rxqcqs[i]);
 		}
 
 		if (lif->hwstamp_txq &&
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.h b/drivers/net/ethernet/pensando/ionic/ionic_dev.h
index d87e6020cfb1..92f16b6c5662 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_dev.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.h
@@ -386,6 +386,7 @@ bool ionic_q_is_posted(struct ionic_queue *q, unsigned int pos);
 
 int ionic_heartbeat_check(struct ionic *ionic);
 bool ionic_is_fw_running(struct ionic_dev *idev);
+void ionic_doorbell_napi_work(struct work_struct *work);
 void ionic_queue_doorbell_check(struct ionic *ionic, int delay);
 
 bool ionic_adminq_poke_doorbell(struct ionic_queue *q);
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
index 267c592d5864..8a9719945240 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
@@ -344,6 +344,7 @@ static int ionic_qcq_disable(struct ionic_lif *lif, struct ionic_qcq *qcq, int f
 	if (qcq->flags & IONIC_QCQ_F_INTR) {
 		struct ionic_dev *idev = &lif->ionic->idev;
 
+		cancel_work_sync(&qcq->doorbell_napi_work);
 		cancel_work_sync(&qcq->dim.work);
 		ionic_intr_mask(idev->intr_ctrl, qcq->intr.index,
 				IONIC_INTR_MASK_SET);
@@ -690,6 +691,7 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type,
 
 	INIT_WORK(&new->dim.work, ionic_dim_work);
 	new->dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_CQE;
+	INIT_WORK(&new->doorbell_napi_work, ionic_doorbell_napi_work);
 
 	*qcq = new;
 
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.h b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
index e4a5ae70793e..40b28d0b858f 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
@@ -88,6 +88,7 @@ struct ionic_qcq {
 	struct ionic_cq cq;
 	struct napi_struct napi;
 	struct ionic_intr_info intr;
+	struct work_struct doorbell_napi_work;
 	struct dentry *dentry;
 };
 
-- 
cgit v1.2.3-58-ga151


From 55a3982ec721dabd5a4c2f16bfb03deb032e45c2 Mon Sep 17 00:00:00 2001
From: Shannon Nelson <shannon.nelson@amd.com>
Date: Tue, 18 Jun 2024 17:32:55 -0700
Subject: ionic: check for queue deadline in doorbell_napi_work

Check the deadline against the last time run and only
schedule a new napi if we haven't been run recently.

Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240619003257.6138-7-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_dev.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
index 56e7c120d492..ec36ace6d010 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
@@ -58,7 +58,14 @@ void ionic_doorbell_napi_work(struct work_struct *work)
 {
 	struct ionic_qcq *qcq = container_of(work, struct ionic_qcq,
 					     doorbell_napi_work);
-	ionic_napi_schedule_do_softirq(&qcq->napi);
+	unsigned long now, then, dif;
+
+	now = READ_ONCE(jiffies);
+	then = qcq->q.dbell_jiffies;
+	dif = now - then;
+
+	if (dif > qcq->q.dbell_deadline)
+		ionic_napi_schedule_do_softirq(&qcq->napi);
 }
 
 static int ionic_get_preferred_cpu(struct ionic *ionic,
-- 
cgit v1.2.3-58-ga151


From f703d56c03050b5d8f9bea6842569798b5407386 Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@amd.com>
Date: Tue, 18 Jun 2024 17:32:56 -0700
Subject: ionic: Use an u16 for rx_copybreak

We only support (u16)-1 size for rx_copybreak, so we can reduce the
field size and move a couple other fields around to save a little
space in the ionic_lif struct.

Before:
	/* size: 17440, cachelines: 273, members: 56 */
	/* sum members: 17403, holes: 9, sum holes: 37 */
	/* last cacheline: 32 bytes */

After:
	/* size: 17424, cachelines: 273, members: 56 */
	/* sum members: 17401, holes: 7, sum holes: 23 */
	/* last cacheline: 16 bytes */

Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240619003257.6138-8-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic_ethtool.c | 11 ++++++++++-
 drivers/net/ethernet/pensando/ionic/ionic_lif.h     |  6 +++---
 2 files changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c
index 91183965a6b7..185a03514ae3 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c
@@ -11,6 +11,8 @@
 #include "ionic_ethtool.h"
 #include "ionic_stats.h"
 
+#define IONIC_MAX_RX_COPYBREAK	min(U16_MAX, IONIC_MAX_BUF_LEN)
+
 static void ionic_get_stats_strings(struct ionic_lif *lif, u8 *buf)
 {
 	u32 i;
@@ -872,10 +874,17 @@ static int ionic_set_tunable(struct net_device *dev,
 			     const void *data)
 {
 	struct ionic_lif *lif = netdev_priv(dev);
+	u32 rx_copybreak;
 
 	switch (tuna->id) {
 	case ETHTOOL_RX_COPYBREAK:
-		lif->rx_copybreak = *(u32 *)data;
+		rx_copybreak = *(u32 *)data;
+		if (rx_copybreak > IONIC_MAX_RX_COPYBREAK) {
+			netdev_err(dev, "Max supported rx_copybreak size: %u\n",
+				   IONIC_MAX_RX_COPYBREAK);
+			return -EINVAL;
+		}
+		lif->rx_copybreak = (u16)rx_copybreak;
 		break;
 	default:
 		return -EOPNOTSUPP;
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.h b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
index 40b28d0b858f..5bd501355670 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
@@ -206,10 +206,10 @@ struct ionic_lif {
 	unsigned int nxqs;
 	unsigned int ntxq_descs;
 	unsigned int nrxq_descs;
-	u32 rx_copybreak;
 	u64 rxq_features;
-	u16 rx_mode;
 	u64 hw_features;
+	u16 rx_copybreak;
+	u16 rx_mode;
 	bool registered;
 	u16 lif_type;
 	unsigned int link_down_count;
@@ -225,11 +225,11 @@ struct ionic_lif {
 	u32 info_sz;
 	struct ionic_qtype_info qtype_info[IONIC_QTYPE_MAX];
 
-	u16 rss_types;
 	u8 rss_hash_key[IONIC_RSS_HASH_KEY_SIZE];
 	u8 *rss_ind_tbl;
 	dma_addr_t rss_ind_tbl_pa;
 	u32 rss_ind_tbl_sz;
+	u16 rss_types;
 
 	struct ionic_rx_filters rx_filters;
 	u32 rx_coalesce_usecs;		/* what the user asked for */
-- 
cgit v1.2.3-58-ga151


From da0262c2c931eff97714ec1e89827796b783d234 Mon Sep 17 00:00:00 2001
From: Brett Creeley <brett.creeley@amd.com>
Date: Tue, 18 Jun 2024 17:32:57 -0700
Subject: ionic: Only run the doorbell workaround for certain asic_type

If the doorbell workaround isn't required for a certain
asic_type then there is no need to run the associated
code. Since newer FW versions are finally reporting their
asic_type we can use a flag to determine whether or not to
do the workaround.

Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Link: https://lore.kernel.org/r/20240619003257.6138-9-shannon.nelson@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/pensando/ionic/ionic.h         |  4 ++++
 drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c |  3 ++-
 drivers/net/ethernet/pensando/ionic/ionic_dev.c     | 16 ++++++++++++++--
 drivers/net/ethernet/pensando/ionic/ionic_lif.c     | 21 +++++++++++++--------
 drivers/net/ethernet/pensando/ionic/ionic_lif.h     |  1 +
 drivers/net/ethernet/pensando/ionic/ionic_txrx.c    | 14 ++++++++------
 6 files changed, 42 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/pensando/ionic/ionic.h b/drivers/net/ethernet/pensando/ionic/ionic.h
index 106ee5b2ceff..1c61390677f7 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic.h
@@ -18,6 +18,8 @@ struct ionic_lif;
 #define PCI_DEVICE_ID_PENSANDO_IONIC_ETH_PF	0x1002
 #define PCI_DEVICE_ID_PENSANDO_IONIC_ETH_VF	0x1003
 
+#define IONIC_ASIC_TYPE_ELBA	2
+
 #define DEVCMD_TIMEOUT			5
 #define IONIC_ADMINQ_TIME_SLICE		msecs_to_jiffies(100)
 
@@ -96,4 +98,6 @@ int ionic_port_identify(struct ionic *ionic);
 int ionic_port_init(struct ionic *ionic);
 int ionic_port_reset(struct ionic *ionic);
 
+bool ionic_doorbell_wa(struct ionic *ionic);
+
 #endif /* _IONIC_H_ */
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
index f362e76756df..b93791d6b593 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_bus_pci.c
@@ -412,7 +412,8 @@ static void ionic_remove(struct pci_dev *pdev)
 		if (test_and_clear_bit(IONIC_LIF_F_FW_RESET, ionic->lif->state))
 			set_bit(IONIC_LIF_F_FW_STOPPING, ionic->lif->state);
 
-		cancel_delayed_work_sync(&ionic->doorbell_check_dwork);
+		if (ionic->lif->doorbell_wa)
+			cancel_delayed_work_sync(&ionic->doorbell_check_dwork);
 		ionic_lif_unregister(ionic->lif);
 		ionic_devlink_unregister(ionic);
 		ionic_lif_deinit(ionic->lif);
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
index ec36ace6d010..9e42d599840d 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c
@@ -128,6 +128,13 @@ static void ionic_doorbell_check_dwork(struct work_struct *work)
 	ionic_queue_doorbell_check(ionic, IONIC_NAPI_DEADLINE);
 }
 
+bool ionic_doorbell_wa(struct ionic *ionic)
+{
+	u8 asic_type = ionic->idev.dev_info.asic_type;
+
+	return !asic_type || asic_type == IONIC_ASIC_TYPE_ELBA;
+}
+
 static int ionic_watchdog_init(struct ionic *ionic)
 {
 	struct ionic_dev *idev = &ionic->idev;
@@ -151,8 +158,10 @@ static int ionic_watchdog_init(struct ionic *ionic)
 		dev_err(ionic->dev, "alloc_workqueue failed");
 		return -ENOMEM;
 	}
-	INIT_DELAYED_WORK(&ionic->doorbell_check_dwork,
-			  ionic_doorbell_check_dwork);
+
+	if (ionic_doorbell_wa(ionic))
+		INIT_DELAYED_WORK(&ionic->doorbell_check_dwork,
+				  ionic_doorbell_check_dwork);
 
 	return 0;
 }
@@ -161,6 +170,9 @@ void ionic_queue_doorbell_check(struct ionic *ionic, int delay)
 {
 	int cpu;
 
+	if (!ionic->lif->doorbell_wa)
+		return;
+
 	cpu = ionic_get_preferred_cpu(ionic, &ionic->lif->adminqcq->intr);
 	queue_delayed_work_on(cpu, ionic->wq, &ionic->doorbell_check_dwork,
 			      delay);
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
index 8a9719945240..38ce35462737 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c
@@ -344,7 +344,8 @@ static int ionic_qcq_disable(struct ionic_lif *lif, struct ionic_qcq *qcq, int f
 	if (qcq->flags & IONIC_QCQ_F_INTR) {
 		struct ionic_dev *idev = &lif->ionic->idev;
 
-		cancel_work_sync(&qcq->doorbell_napi_work);
+		if (lif->doorbell_wa)
+			cancel_work_sync(&qcq->doorbell_napi_work);
 		cancel_work_sync(&qcq->dim.work);
 		ionic_intr_mask(idev->intr_ctrl, qcq->intr.index,
 				IONIC_INTR_MASK_SET);
@@ -691,7 +692,8 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type,
 
 	INIT_WORK(&new->dim.work, ionic_dim_work);
 	new->dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_CQE;
-	INIT_WORK(&new->doorbell_napi_work, ionic_doorbell_napi_work);
+	if (lif->doorbell_wa)
+		INIT_WORK(&new->doorbell_napi_work, ionic_doorbell_napi_work);
 
 	*qcq = new;
 
@@ -1215,12 +1217,14 @@ static int ionic_adminq_napi(struct napi_struct *napi, int budget)
 		ionic_intr_credits(idev->intr_ctrl, intr->index, credits, flags);
 	}
 
-	if (!a_work)
-		ionic_adminq_poke_doorbell(&lif->adminqcq->q);
-	if (lif->hwstamp_rxq && !rx_work)
-		ionic_rxq_poke_doorbell(&lif->hwstamp_rxq->q);
-	if (lif->hwstamp_txq && !tx_work)
-		ionic_txq_poke_doorbell(&lif->hwstamp_txq->q);
+	if (lif->doorbell_wa) {
+		if (!a_work)
+			ionic_adminq_poke_doorbell(&lif->adminqcq->q);
+		if (lif->hwstamp_rxq && !rx_work)
+			ionic_rxq_poke_doorbell(&lif->hwstamp_rxq->q);
+		if (lif->hwstamp_txq && !tx_work)
+			ionic_txq_poke_doorbell(&lif->hwstamp_txq->q);
+	}
 
 	return work_done;
 }
@@ -3749,6 +3753,7 @@ int ionic_lif_init(struct ionic_lif *lif)
 		goto err_out_notifyq_deinit;
 
 	lif->rx_copybreak = IONIC_RX_COPYBREAK_DEFAULT;
+	lif->doorbell_wa = ionic_doorbell_wa(lif->ionic);
 
 	set_bit(IONIC_LIF_F_INITED, lif->state);
 
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.h b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
index 5bd501355670..3e1005293c4a 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_lif.h
+++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.h
@@ -211,6 +211,7 @@ struct ionic_lif {
 	u16 rx_copybreak;
 	u16 rx_mode;
 	bool registered;
+	bool doorbell_wa;
 	u16 lif_type;
 	unsigned int link_down_count;
 	unsigned int nmcast;
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
index a4e923376484..5bf13a5d411c 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c
@@ -950,7 +950,7 @@ int ionic_tx_napi(struct napi_struct *napi, int budget)
 				   work_done, flags);
 	}
 
-	if (!work_done)
+	if (!work_done && cq->bound_q->lif->doorbell_wa)
 		ionic_txq_poke_doorbell(&qcq->q);
 
 	return work_done;
@@ -993,7 +993,7 @@ int ionic_rx_napi(struct napi_struct *napi, int budget)
 				   work_done, flags);
 	}
 
-	if (!work_done)
+	if (!work_done && cq->bound_q->lif->doorbell_wa)
 		ionic_rxq_poke_doorbell(&qcq->q);
 
 	return work_done;
@@ -1038,10 +1038,12 @@ int ionic_txrx_napi(struct napi_struct *napi, int budget)
 				   tx_work_done + rx_work_done, flags);
 	}
 
-	if (!rx_work_done)
-		ionic_rxq_poke_doorbell(&rxqcq->q);
-	if (!tx_work_done)
-		ionic_txq_poke_doorbell(&txqcq->q);
+	if (lif->doorbell_wa) {
+		if (!rx_work_done)
+			ionic_rxq_poke_doorbell(&rxqcq->q);
+		if (!tx_work_done)
+			ionic_txq_poke_doorbell(&txqcq->q);
+	}
 
 	return rx_work_done;
 }
-- 
cgit v1.2.3-58-ga151


From b1dc3c68e977273397619d901b34f25634666c24 Mon Sep 17 00:00:00 2001
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Fri, 12 Apr 2024 20:33:32 +0300
Subject: can: mcp251x: Fix up includes

This driver is including the legacy GPIO header <linux/gpio.h>
but the only thing it is using from that header is the wrong
define for GPIOF_DIR_OUT.

Fix it up by using GPIO_LINE_DIRECTION_* macros respectively.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Link: https://lore.kernel.org/all/20240412173332.186685-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251x.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/can/spi/mcp251x.c b/drivers/net/can/spi/mcp251x.c
index 79c4bab5f724..643974b3f329 100644
--- a/drivers/net/can/spi/mcp251x.c
+++ b/drivers/net/can/spi/mcp251x.c
@@ -28,7 +28,6 @@
 #include <linux/device.h>
 #include <linux/ethtool.h>
 #include <linux/freezer.h>
-#include <linux/gpio.h>
 #include <linux/gpio/driver.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
@@ -482,9 +481,9 @@ static int mcp251x_gpio_get_direction(struct gpio_chip *chip,
 				      unsigned int offset)
 {
 	if (mcp251x_gpio_is_input(offset))
-		return GPIOF_DIR_IN;
+		return GPIO_LINE_DIRECTION_IN;
 
-	return GPIOF_DIR_OUT;
+	return GPIO_LINE_DIRECTION_OUT;
 }
 
 static int mcp251x_gpio_get(struct gpio_chip *chip, unsigned int offset)
-- 
cgit v1.2.3-58-ga151


From 5ca3801388f8ae69b145812410927d5e8156b67a Mon Sep 17 00:00:00 2001
From: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Thu, 2 May 2024 15:38:52 +0300
Subject: can: sja1000: plx_pci: Reuse predefined CTI subvendor ID

There is predefined PCI_SUBVENDOR_ID_CONNECT_TECH, use it in the driver.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/all/20240502123852.2631577-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/sja1000/plx_pci.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/can/sja1000/plx_pci.c b/drivers/net/can/sja1000/plx_pci.c
index 5de1ebb0c6f0..67e5316c6372 100644
--- a/drivers/net/can/sja1000/plx_pci.c
+++ b/drivers/net/can/sja1000/plx_pci.c
@@ -122,7 +122,6 @@ struct plx_pci_card {
 #define TEWS_PCI_VENDOR_ID		0x1498
 #define TEWS_PCI_DEVICE_ID_TMPC810	0x032A
 
-#define CTI_PCI_VENDOR_ID		0x12c4
 #define CTI_PCI_DEVICE_ID_CRG001	0x0900
 
 #define MOXA_PCI_VENDOR_ID		0x1393
@@ -358,7 +357,7 @@ static const struct pci_device_id plx_pci_tbl[] = {
 	{
 		/* Connect Tech Inc. CANpro/104-Plus Opto (CRG001) card */
 		PCI_VENDOR_ID_PLX, PCI_DEVICE_ID_PLX_9030,
-		CTI_PCI_VENDOR_ID, CTI_PCI_DEVICE_ID_CRG001,
+		PCI_SUBVENDOR_ID_CONNECT_TECH, CTI_PCI_DEVICE_ID_CRG001,
 		0, 0,
 		(kernel_ulong_t)&plx_pci_card_info_cti
 	},
-- 
cgit v1.2.3-58-ga151


From 58b34cd646b4af9c3176808c4f8e8724402ab805 Mon Sep 17 00:00:00 2001
From: Mans Rullgard <mans@mansr.com>
Date: Sat, 27 Apr 2024 16:26:48 +0100
Subject: can: Kconfig: remove obsolete help text for slcan

Commit cfcb4465e992 ("can: slcan: remove legacy infrastructure")
removed the 10-device limit.  Update the Kconfig help text accordingly.

Signed-off-by: Mans Rullgard <mans@mansr.com>
Acked-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Link: https://lore.kernel.org/all/20240427152648.25434-1-mans@mansr.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/Kconfig | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig
index 2e31db55d927..7f9b60a42d29 100644
--- a/drivers/net/can/Kconfig
+++ b/drivers/net/can/Kconfig
@@ -187,9 +187,8 @@ config CAN_SLCAN
 	  slcand) can be found in the can-utils at the linux-can project, see
 	  https://github.com/linux-can/can-utils for details.
 
-	  The slcan driver supports up to 10 CAN netdevices by default which
-	  can be changed by the 'maxdev=xx' module option. This driver can
-	  also be built as a module. If so, the module will be called slcan.
+	  This driver can also be built as a module. If so, the module
+	  will be called slcan.
 
 config CAN_SUN4I
 	tristate "Allwinner A10 CAN controller"
-- 
cgit v1.2.3-58-ga151


From 8416ac9c87bd77ad6cb3ee0505d0ca765949e0c9 Mon Sep 17 00:00:00 2001
From: Harini T <harini.t@amd.com>
Date: Fri, 3 May 2024 11:35:52 +0530
Subject: dt-bindings: can: xilinx_can: Modify the title to indicate CAN and
 CANFD controllers are supported

Xilinx CAN binding documentation supports CAN and CANFD controllers.
Modify the title to indicate that both controllers are supported.

Signed-off-by: Harini T <harini.t@amd.com>
Link: https://lore.kernel.org/all/20240503060553.8520-2-harini.t@amd.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 Documentation/devicetree/bindings/net/can/xilinx,can.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/net/can/xilinx,can.yaml b/Documentation/devicetree/bindings/net/can/xilinx,can.yaml
index 8d4e5af6fd6c..40835497050a 100644
--- a/Documentation/devicetree/bindings/net/can/xilinx,can.yaml
+++ b/Documentation/devicetree/bindings/net/can/xilinx,can.yaml
@@ -5,7 +5,7 @@ $id: http://devicetree.org/schemas/net/can/xilinx,can.yaml#
 $schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title:
-  Xilinx Axi CAN/Zynq CANPS controller
+  Xilinx CAN and CANFD controller
 
 maintainers:
   - Appana Durga Kedareswara rao <appana.durga.rao@xilinx.com>
-- 
cgit v1.2.3-58-ga151


From ba63a7e08523be3496cc1b7e5f77d306144a3a40 Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <socketcan@hartkopp.net>
Date: Sat, 20 Apr 2024 21:47:46 +0200
Subject: can: isotp: remove ISO 15675-2 specification version where possible

With the new ISO 15765-2:2024 release the former documentation and comments
have to be reworked. This patch removes the ISO specification version/date
where possible.

Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net>
Acked-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Acked-by: Francesco Valla <valla.francesco@gmail.com>
Link: https://lore.kernel.org/all/20240420194746.4885-1-socketcan@hartkopp.net
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 include/uapi/linux/can/isotp.h |  2 +-
 net/can/Kconfig                | 11 +++++------
 net/can/isotp.c                | 11 ++++++-----
 3 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/include/uapi/linux/can/isotp.h b/include/uapi/linux/can/isotp.h
index 6cde62371b6f..bd990917f7c4 100644
--- a/include/uapi/linux/can/isotp.h
+++ b/include/uapi/linux/can/isotp.h
@@ -2,7 +2,7 @@
 /*
  * linux/can/isotp.h
  *
- * Definitions for isotp CAN sockets (ISO 15765-2:2016)
+ * Definitions for ISO 15765-2 CAN transport protocol sockets
  *
  * Copyright (c) 2020 Volkswagen Group Electronic Research
  * All rights reserved.
diff --git a/net/can/Kconfig b/net/can/Kconfig
index cb56be8e3862..af64a6f76458 100644
--- a/net/can/Kconfig
+++ b/net/can/Kconfig
@@ -56,18 +56,17 @@ config CAN_GW
 source "net/can/j1939/Kconfig"
 
 config CAN_ISOTP
-	tristate "ISO 15765-2:2016 CAN transport protocol"
+	tristate "ISO 15765-2 CAN transport protocol"
 	help
 	  CAN Transport Protocols offer support for segmented Point-to-Point
 	  communication between CAN nodes via two defined CAN Identifiers.
+	  This protocol driver implements segmented data transfers for CAN CC
+	  (aka Classical CAN, CAN 2.0B) and CAN FD frame types which were
+	  introduced with ISO 15765-2:2016.
 	  As CAN frames can only transport a small amount of data bytes
-	  (max. 8 bytes for 'classic' CAN and max. 64 bytes for CAN FD) this
+	  (max. 8 bytes for CAN CC and max. 64 bytes for CAN FD) this
 	  segmentation is needed to transport longer Protocol Data Units (PDU)
 	  as needed e.g. for vehicle diagnosis (UDS, ISO 14229) or IP-over-CAN
 	  traffic.
-	  This protocol driver implements data transfers according to
-	  ISO 15765-2:2016 for 'classic' CAN and CAN FD frame types.
-	  If you want to perform automotive vehicle diagnostic services (UDS),
-	  say 'y'.
 
 endif
diff --git a/net/can/isotp.c b/net/can/isotp.c
index 25bac0fafc83..16046931542a 100644
--- a/net/can/isotp.c
+++ b/net/can/isotp.c
@@ -72,7 +72,7 @@
 #include <net/sock.h>
 #include <net/net_namespace.h>
 
-MODULE_DESCRIPTION("PF_CAN isotp 15765-2:2016 protocol");
+MODULE_DESCRIPTION("PF_CAN ISO 15765-2 transport protocol");
 MODULE_LICENSE("Dual BSD/GPL");
 MODULE_AUTHOR("Oliver Hartkopp <socketcan@hartkopp.net>");
 MODULE_ALIAS("can-proto-6");
@@ -83,10 +83,11 @@ MODULE_ALIAS("can-proto-6");
 			 (CAN_EFF_MASK | CAN_EFF_FLAG | CAN_RTR_FLAG) : \
 			 (CAN_SFF_MASK | CAN_EFF_FLAG | CAN_RTR_FLAG))
 
-/* ISO 15765-2:2016 supports more than 4095 byte per ISO PDU as the FF_DL can
- * take full 32 bit values (4 Gbyte). We would need some good concept to handle
- * this between user space and kernel space. For now set the static buffer to
- * something about 8 kbyte to be able to test this new functionality.
+/* Since ISO 15765-2:2016 the CAN isotp protocol supports more than 4095
+ * byte per ISO PDU as the FF_DL can take full 32 bit values (4 Gbyte).
+ * We would need some good concept to handle this between user space and
+ * kernel space. For now set the static buffer to something about 8 kbyte
+ * to be able to test this new functionality.
  */
 #define DEFAULT_MAX_PDU_SIZE 8300
 
-- 
cgit v1.2.3-58-ga151


From e562bad35fe38af22437caea9bdf911ff59cf56b Mon Sep 17 00:00:00 2001
From: Harini T <harini.t@amd.com>
Date: Fri, 3 May 2024 11:35:53 +0530
Subject: can: xilinx_can: Document driver description to list all supported
 IPs

Xilinx CAN driver supports AXI CAN, AXI CANFD, CANPS and CANFD PS IPs.
Document all supported IPs in comment description.

Signed-off-by: Harini T <harini.t@amd.com>
Link: https://lore.kernel.org/all/20240503060553.8520-3-harini.t@amd.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/xilinx_can.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/can/xilinx_can.c b/drivers/net/can/xilinx_can.c
index fae0120473f8..d944911d7f05 100644
--- a/drivers/net/can/xilinx_can.c
+++ b/drivers/net/can/xilinx_can.c
@@ -6,7 +6,7 @@
  * Copyright (C) 2017 - 2018 Sandvik Mining and Construction Oy
  *
  * Description:
- * This driver is developed for Axi CAN IP and for Zynq CANPS Controller.
+ * This driver is developed for AXI CAN IP, AXI CANFD IP, CANPS and CANFD PS Controller.
  */
 
 #include <linux/bitfield.h>
-- 
cgit v1.2.3-58-ga151


From 67711e04254c388aa2f3125f23fbb897bbc07d56 Mon Sep 17 00:00:00 2001
From: Francesco Valla <valla.francesco@gmail.com>
Date: Wed, 1 May 2024 11:24:13 +0200
Subject: Documentation: networking: document ISO 15765-2

Document basic concepts, APIs and behaviour of the ISO 15675-2 (ISO-TP)
CAN stack.

Signed-off-by: Francesco Valla <valla.francesco@gmail.com>
Reviewed-by: Bagas Sanjaya <bagasdotme@gmail.com>
Reviewed-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Link: https://lore.kernel.org/all/20240501092413.414700-2-valla.francesco@gmail.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 Documentation/networking/index.rst      |   1 +
 Documentation/networking/iso15765-2.rst | 386 ++++++++++++++++++++++++++++++++
 MAINTAINERS                             |   1 +
 3 files changed, 388 insertions(+)
 create mode 100644 Documentation/networking/iso15765-2.rst

diff --git a/Documentation/networking/index.rst b/Documentation/networking/index.rst
index a6443851a142..cd5574005dba 100644
--- a/Documentation/networking/index.rst
+++ b/Documentation/networking/index.rst
@@ -19,6 +19,7 @@ Contents:
    caif/index
    ethtool-netlink
    ieee802154
+   iso15765-2
    j1939
    kapi
    msg_zerocopy
diff --git a/Documentation/networking/iso15765-2.rst b/Documentation/networking/iso15765-2.rst
new file mode 100644
index 000000000000..0e9d96074178
--- /dev/null
+++ b/Documentation/networking/iso15765-2.rst
@@ -0,0 +1,386 @@
+.. SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause)
+
+====================
+ISO 15765-2 (ISO-TP)
+====================
+
+Overview
+========
+
+ISO 15765-2, also known as ISO-TP, is a transport protocol specifically defined
+for diagnostic communication on CAN. It is widely used in the automotive
+industry, for example as the transport protocol for UDSonCAN (ISO 14229-3) or
+emission-related diagnostic services (ISO 15031-5).
+
+ISO-TP can be used both on CAN CC (aka Classical CAN) and CAN FD (CAN with
+Flexible Datarate) based networks. It is also designed to be compatible with a
+CAN network using SAE J1939 as data link layer (however, this is not a
+requirement).
+
+Specifications used
+-------------------
+
+* ISO 15765-2:2024 : Road vehicles - Diagnostic communication over Controller
+  Area Network (DoCAN). Part 2: Transport protocol and network layer services.
+
+Addressing
+----------
+
+In its simplest form, ISO-TP is based on two kinds of addressing modes for the
+nodes connected to the same network:
+
+* physical addressing is implemented by two node-specific addresses and is used
+  in 1-to-1 communication.
+
+* functional addressing is implemented by one node-specific address and is used
+  in 1-to-N communication.
+
+Three different addressing formats can be employed:
+
+* "normal" : each address is represented simply by a CAN ID.
+
+* "extended": each address is represented by a CAN ID plus the first byte of
+  the CAN payload; both the CAN ID and the byte inside the payload shall be
+  different between two addresses.
+
+* "mixed": each address is represented by a CAN ID plus the first byte of
+  the CAN payload; the CAN ID is different between two addresses, but the
+  additional byte is the same.
+
+Transport protocol and associated frame types
+---------------------------------------------
+
+When transmitting data using the ISO-TP protocol, the payload can either fit
+inside one single CAN message or not, also considering the overhead the protocol
+is generating and the optional extended addressing. In the first case, the data
+is transmitted at once using a so-called Single Frame (SF). In the second case,
+ISO-TP defines a multi-frame protocol, in which the sender provides (through a
+First Frame - FF) the PDU length which is to be transmitted and also asks for a
+Flow Control (FC) frame, which provides the maximum supported size of a macro
+data block (``blocksize``) and the minimum time between the single CAN messages
+composing such block (``stmin``). Once this information has been received, the
+sender starts to send frames containing fragments of the data payload (called
+Consecutive Frames - CF), stopping after every ``blocksize``-sized block to wait
+confirmation from the receiver which should then send another Flow Control
+frame to inform the sender about its availability to receive more data.
+
+How to Use ISO-TP
+=================
+
+As with others CAN protocols, the ISO-TP stack support is built into the
+Linux network subsystem for the CAN bus, aka. Linux-CAN or SocketCAN, and
+thus follows the same socket API.
+
+Creation and basic usage of an ISO-TP socket
+--------------------------------------------
+
+To use the ISO-TP stack, ``#include <linux/can/isotp.h>`` shall be used. A
+socket can then be created using the ``PF_CAN`` protocol family, the
+``SOCK_DGRAM`` type (as the underlying protocol is datagram-based by design)
+and the ``CAN_ISOTP`` protocol:
+
+.. code-block:: C
+
+    s = socket(PF_CAN, SOCK_DGRAM, CAN_ISOTP);
+
+After the socket has been successfully created, ``bind(2)`` shall be called to
+bind the socket to the desired CAN interface; to do so:
+
+* a TX CAN ID shall be specified as part of the sockaddr supplied to the call
+  itself.
+
+* a RX CAN ID shall also be specified, unless broadcast flags have been set
+  through socket option (explained below).
+
+Once bound to an interface, the socket can be read from and written to using
+the usual ``read(2)`` and ``write(2)`` system calls, as well as ``send(2)``,
+``sendmsg(2)``, ``recv(2)`` and ``recvmsg(2)``.
+Unlike the CAN_RAW socket API, only the ISO-TP data field (the actual payload)
+is sent and received by the userspace application using these calls. The address
+information and the protocol information are automatically filled by the ISO-TP
+stack using the configuration supplied during socket creation. In the same way,
+the stack will use the transport mechanism when required (i.e., when the size
+of the data payload exceeds the MTU of the underlying CAN bus).
+
+The sockaddr structure used for SocketCAN has extensions for use with ISO-TP,
+as specified below:
+
+.. code-block:: C
+
+    struct sockaddr_can {
+        sa_family_t can_family;
+        int         can_ifindex;
+        union {
+            struct { canid_t rx_id, tx_id; } tp;
+        ...
+        } can_addr;
+    }
+
+* ``can_family`` and ``can_ifindex`` serve the same purpose as for other
+  SocketCAN sockets.
+
+* ``can_addr.tp.rx_id`` specifies the receive (RX) CAN ID and will be used as
+  a RX filter.
+
+* ``can_addr.tp.tx_id`` specifies the transmit (TX) CAN ID
+
+ISO-TP socket options
+---------------------
+
+When creating an ISO-TP socket, reasonable defaults are set. Some options can
+be modified with ``setsockopt(2)`` and/or read back with ``getsockopt(2)``.
+
+General options
+~~~~~~~~~~~~~~~
+
+General socket options can be passed using the ``CAN_ISOTP_OPTS`` optname:
+
+.. code-block:: C
+
+    struct can_isotp_options opts;
+    ret = setsockopt(s, SOL_CAN_ISOTP, CAN_ISOTP_OPTS, &opts, sizeof(opts))
+
+where the ``can_isotp_options`` structure has the following contents:
+
+.. code-block:: C
+
+    struct can_isotp_options {
+        u32 flags;
+        u32 frame_txtime;
+        u8  ext_address;
+        u8  txpad_content;
+        u8  rxpad_content;
+        u8  rx_ext_address;
+    };
+
+* ``flags``: modifiers to be applied to the default behaviour of the ISO-TP
+  stack. Following flags are available:
+
+  * ``CAN_ISOTP_LISTEN_MODE``: listen only (do not send FC frames); normally
+    used as a testing feature.
+
+  * ``CAN_ISOTP_EXTEND_ADDR``: use the byte specified in ``ext_address`` as an
+    additional address component. This enables the "mixed" addressing format if
+    used alone, or the "extended" addressing format if used in conjunction with
+    ``CAN_ISOTP_RX_EXT_ADDR``.
+
+  * ``CAN_ISOTP_TX_PADDING``: enable padding for transmitted frames, using
+    ``txpad_content`` as value for the padding bytes.
+
+  * ``CAN_ISOTP_RX_PADDING``: enable padding for the received frames, using
+    ``rxpad_content`` as value for the padding bytes.
+
+  * ``CAN_ISOTP_CHK_PAD_LEN``: check for correct padding length on the received
+    frames.
+
+  * ``CAN_ISOTP_CHK_PAD_DATA``: check padding bytes on the received frames
+    against ``rxpad_content``; if ``CAN_ISOTP_RX_PADDING`` is not specified,
+    this flag is ignored.
+
+  * ``CAN_ISOTP_HALF_DUPLEX``: force ISO-TP socket in half duplex mode
+    (that is, transport mechanism can only be incoming or outgoing at the same
+    time, not both).
+
+  * ``CAN_ISOTP_FORCE_TXSTMIN``: ignore stmin from received FC; normally
+    used as a testing feature.
+
+  * ``CAN_ISOTP_FORCE_RXSTMIN``: ignore CFs depending on rx stmin; normally
+    used as a testing feature.
+
+  * ``CAN_ISOTP_RX_EXT_ADDR``: use ``rx_ext_address`` instead of ``ext_address``
+    as extended addressing byte on the reception path. If used in conjunction
+    with ``CAN_ISOTP_EXTEND_ADDR``, this flag effectively enables the "extended"
+    addressing format.
+
+  * ``CAN_ISOTP_WAIT_TX_DONE``: wait until the frame is sent before returning
+    from ``write(2)`` and ``send(2)`` calls (i.e., blocking write operations).
+
+  * ``CAN_ISOTP_SF_BROADCAST``: use 1-to-N functional addressing (cannot be
+    specified alongside ``CAN_ISOTP_CF_BROADCAST``).
+
+  * ``CAN_ISOTP_CF_BROADCAST``: use 1-to-N transmission without flow control
+    (cannot be specified alongside ``CAN_ISOTP_SF_BROADCAST``).
+    NOTE: this is not covered by the ISO 15765-2 standard.
+
+  * ``CAN_ISOTP_DYN_FC_PARMS``: enable dynamic update of flow control
+    parameters.
+
+* ``frame_txtime``: frame transmission time (defined as N_As/N_Ar inside the
+  ISO standard); if ``0``, the default (or the last set value) is used.
+  To set the transmission time to ``0``, the ``CAN_ISOTP_FRAME_TXTIME_ZERO``
+  macro (equal to 0xFFFFFFFF) shall be used.
+
+* ``ext_address``: extended addressing byte, used if the
+  ``CAN_ISOTP_EXTEND_ADDR`` flag is specified.
+
+* ``txpad_content``: byte used as padding value for transmitted frames.
+
+* ``rxpad_content``: byte used as padding value for received frames.
+
+* ``rx_ext_address``: extended addressing byte for the reception path, used if
+  the ``CAN_ISOTP_RX_EXT_ADDR`` flag is specified.
+
+Flow Control options
+~~~~~~~~~~~~~~~~~~~~
+
+Flow Control (FC) options can be passed using the ``CAN_ISOTP_RECV_FC`` optname
+to provide the communication parameters for receiving ISO-TP PDUs.
+
+.. code-block:: C
+
+    struct can_isotp_fc_options fc_opts;
+    ret = setsockopt(s, SOL_CAN_ISOTP, CAN_ISOTP_RECV_FC, &fc_opts, sizeof(fc_opts));
+
+where the ``can_isotp_fc_options`` structure has the following contents:
+
+.. code-block:: C
+
+    struct can_isotp_options {
+        u8 bs;
+        u8 stmin;
+        u8 wftmax;
+    };
+
+* ``bs``: blocksize provided in flow control frames.
+
+* ``stmin``: minimum separation time provided in flow control frames; can
+  have the following values (others are reserved):
+
+  * 0x00 - 0x7F : 0 - 127 ms
+
+  * 0xF1 - 0xF9 : 100 us - 900 us
+
+* ``wftmax``: maximum number of wait frames provided in flow control frames.
+
+Link Layer options
+~~~~~~~~~~~~~~~~~~
+
+Link Layer (LL) options can be passed using the ``CAN_ISOTP_LL_OPTS`` optname:
+
+.. code-block:: C
+
+    struct can_isotp_ll_options ll_opts;
+    ret = setsockopt(s, SOL_CAN_ISOTP, CAN_ISOTP_LL_OPTS, &ll_opts, sizeof(ll_opts));
+
+where the ``can_isotp_ll_options`` structure has the following contents:
+
+.. code-block:: C
+
+    struct can_isotp_ll_options {
+        u8 mtu;
+        u8 tx_dl;
+        u8 tx_flags;
+    };
+
+* ``mtu``: generated and accepted CAN frame type, can be equal to ``CAN_MTU``
+  for classical CAN frames or ``CANFD_MTU`` for CAN FD frames.
+
+* ``tx_dl``: maximum payload length for transmitted frames, can have one value
+  among: 8, 12, 16, 20, 24, 32, 48, 64. Values above 8 only apply to CAN FD
+  traffic (i.e.: ``mtu = CANFD_MTU``).
+
+* ``tx_flags``: flags set into ``struct canfd_frame.flags`` at frame creation.
+  Only applies to CAN FD traffic (i.e.: ``mtu = CANFD_MTU``).
+
+Transmission stmin
+~~~~~~~~~~~~~~~~~~
+
+The transmission minimum separation time (stmin) can be forced using the
+``CAN_ISOTP_TX_STMIN`` optname and providing an stmin value in microseconds as
+a 32bit unsigned integer; this will overwrite the value sent by the receiver in
+flow control frames:
+
+.. code-block:: C
+
+    uint32_t stmin;
+    ret = setsockopt(s, SOL_CAN_ISOTP, CAN_ISOTP_TX_STMIN, &stmin, sizeof(stmin));
+
+Reception stmin
+~~~~~~~~~~~~~~~
+
+The reception minimum separation time (stmin) can be forced using the
+``CAN_ISOTP_RX_STMIN`` optname and providing an stmin value in microseconds as
+a 32bit unsigned integer; received Consecutive Frames (CF) which timestamps
+differ less than this value will be ignored:
+
+.. code-block:: C
+
+    uint32_t stmin;
+    ret = setsockopt(s, SOL_CAN_ISOTP, CAN_ISOTP_RX_STMIN, &stmin, sizeof(stmin));
+
+Multi-frame transport support
+-----------------------------
+
+The ISO-TP stack contained inside the Linux kernel supports the multi-frame
+transport mechanism defined by the standard, with the following constraints:
+
+* the maximum size of a PDU is defined by a module parameter, with an hard
+  limit imposed at build time.
+
+* when a transmission is in progress, subsequent calls to ``write(2)`` will
+  block, while calls to ``send(2)`` will either block or fail depending on the
+  presence of the ``MSG_DONTWAIT`` flag.
+
+* no support is present for sending "wait frames": whether a PDU can be fully
+  received or not is decided when the First Frame is received.
+
+Errors
+------
+
+Following errors are reported to userspace:
+
+RX path errors
+~~~~~~~~~~~~~~
+
+============ ===============================================================
+-ETIMEDOUT   timeout of data reception
+-EILSEQ      sequence number mismatch during a multi-frame reception
+-EBADMSG     data reception with wrong padding
+============ ===============================================================
+
+TX path errors
+~~~~~~~~~~~~~~
+
+========== =================================================================
+-ECOMM     flow control reception timeout
+-EMSGSIZE  flow control reception overflow
+-EBADMSG   flow control reception with wrong layout/padding
+========== =================================================================
+
+Examples
+========
+
+Basic node example
+------------------
+
+Following example implements a node using "normal" physical addressing, with
+RX ID equal to 0x18DAF142 and a TX ID equal to 0x18DA42F1. All options are left
+to their default.
+
+.. code-block:: C
+
+  int s;
+  struct sockaddr_can addr;
+  int ret;
+
+  s = socket(PF_CAN, SOCK_DGRAM, CAN_ISOTP);
+  if (s < 0)
+      exit(1);
+
+  addr.can_family = AF_CAN;
+  addr.can_ifindex = if_nametoindex("can0");
+  addr.tp.tx_id = 0x18DA42F1 | CAN_EFF_FLAG;
+  addr.tp.rx_id = 0x18DAF142 | CAN_EFF_FLAG;
+
+  ret = bind(s, (struct sockaddr *)&addr, sizeof(addr));
+  if (ret < 0)
+      exit(1);
+
+  /* Data can now be received using read(s, ...) and sent using write(s, ...) */
+
+Additional examples
+-------------------
+
+More complete (and complex) examples can be found inside the ``isotp*`` userland
+tools, distributed as part of the ``can-utils`` utilities at:
+https://github.com/linux-can/can-utils
diff --git a/MAINTAINERS b/MAINTAINERS
index e66b7d4324ae..17e3fa3eb6fd 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4842,6 +4842,7 @@ W:	https://github.com/linux-can
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can.git
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next.git
 F:	Documentation/networking/can.rst
+F:	Documentation/networking/iso15765-2.rst
 F:	include/linux/can/can-ml.h
 F:	include/linux/can/core.h
 F:	include/linux/can/skb.h
-- 
cgit v1.2.3-58-ga151


From f9f608e38b9cff2dfa6fc80fe9792a0441379d60 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Sun, 26 May 2024 00:25:09 +0100
Subject: can: mscan: remove unused struct 'mscan_state'

'mscan_state' is unused since the original
commit afa17a500a36 ("net/can: add driver for mscan family &
mpc52xx_mscan").

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/all/20240525232509.191735-1-linux@treblig.org
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/mscan/mscan.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/net/can/mscan/mscan.c b/drivers/net/can/mscan/mscan.c
index a6829cdc0e81..8c2a7bc64d3d 100644
--- a/drivers/net/can/mscan/mscan.c
+++ b/drivers/net/can/mscan/mscan.c
@@ -34,12 +34,6 @@ static const struct can_bittiming_const mscan_bittiming_const = {
 	.brp_inc = 1,
 };
 
-struct mscan_state {
-	u8 mode;
-	u8 canrier;
-	u8 cantier;
-};
-
 static enum can_state state_map[] = {
 	CAN_STATE_ERROR_ACTIVE,
 	CAN_STATE_ERROR_WARNING,
-- 
cgit v1.2.3-58-ga151


From b501d261a5b39fed653a90ab5db9f3c3d93c2486 Mon Sep 17 00:00:00 2001
From: Jiawen Wu <jiawenwu@trustnetic.com>
Date: Tue, 18 Jun 2024 18:16:07 +0800
Subject: net: txgbe: add FDIR ATR support

Add flow director ATR filter. ATR mode is enabled by default to filter
TCP packets.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/wangxun/libwx/wx_hw.c      |  27 ++-
 drivers/net/ethernet/wangxun/libwx/wx_hw.h      |   2 +
 drivers/net/ethernet/wangxun/libwx/wx_lib.c     |  31 ++-
 drivers/net/ethernet/wangxun/libwx/wx_lib.h     |   1 +
 drivers/net/ethernet/wangxun/libwx/wx_type.h    |  52 +++-
 drivers/net/ethernet/wangxun/txgbe/Makefile     |   1 +
 drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c | 302 ++++++++++++++++++++++++
 drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h |  10 +
 drivers/net/ethernet/wangxun/txgbe/txgbe_main.c |   9 +
 drivers/net/ethernet/wangxun/txgbe/txgbe_type.h | 121 ++++++++++
 10 files changed, 548 insertions(+), 8 deletions(-)
 create mode 100644 drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c
 create mode 100644 drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h

diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.c b/drivers/net/ethernet/wangxun/libwx/wx_hw.c
index 7c4b6881a93f..8fb38f83a615 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_hw.c
+++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.c
@@ -1147,8 +1147,15 @@ static void wx_enable_rx(struct wx *wx)
 static void wx_set_rxpba(struct wx *wx)
 {
 	u32 rxpktsize, txpktsize, txpbthresh;
+	u32 pbsize = wx->mac.rx_pb_size;
 
-	rxpktsize = wx->mac.rx_pb_size << WX_RDB_PB_SZ_SHIFT;
+	if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags)) {
+		if (test_bit(WX_FLAG_FDIR_HASH, wx->flags) ||
+		    test_bit(WX_FLAG_FDIR_PERFECT, wx->flags))
+			pbsize -= 64; /* Default 64KB */
+	}
+
+	rxpktsize = pbsize << WX_RDB_PB_SZ_SHIFT;
 	wr32(wx, WX_RDB_PB_SZ(0), rxpktsize);
 
 	/* Only support an equally distributed Tx packet buffer strategy. */
@@ -1261,7 +1268,7 @@ static void wx_configure_port(struct wx *wx)
  *  Stops the receive data path and waits for the HW to internally empty
  *  the Rx security block
  **/
-static int wx_disable_sec_rx_path(struct wx *wx)
+int wx_disable_sec_rx_path(struct wx *wx)
 {
 	u32 secrx;
 
@@ -1271,6 +1278,7 @@ static int wx_disable_sec_rx_path(struct wx *wx)
 	return read_poll_timeout(rd32, secrx, secrx & WX_RSC_ST_RSEC_RDY,
 				 1000, 40000, false, wx, WX_RSC_ST);
 }
+EXPORT_SYMBOL(wx_disable_sec_rx_path);
 
 /**
  *  wx_enable_sec_rx_path - Enables the receive data path
@@ -1278,11 +1286,12 @@ static int wx_disable_sec_rx_path(struct wx *wx)
  *
  *  Enables the receive data path.
  **/
-static void wx_enable_sec_rx_path(struct wx *wx)
+void wx_enable_sec_rx_path(struct wx *wx)
 {
 	wr32m(wx, WX_RSC_CTL, WX_RSC_CTL_RX_DIS, 0);
 	WX_WRITE_FLUSH(wx);
 }
+EXPORT_SYMBOL(wx_enable_sec_rx_path);
 
 static void wx_vlan_strip_control(struct wx *wx, bool enable)
 {
@@ -1499,6 +1508,13 @@ static void wx_configure_tx_ring(struct wx *wx,
 		txdctl |= ring->count / 128 << WX_PX_TR_CFG_TR_SIZE_SHIFT;
 	txdctl |= 0x20 << WX_PX_TR_CFG_WTHRESH_SHIFT;
 
+	ring->atr_count = 0;
+	if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags) &&
+	    test_bit(WX_FLAG_FDIR_HASH, wx->flags))
+		ring->atr_sample_rate = wx->atr_sample_rate;
+	else
+		ring->atr_sample_rate = 0;
+
 	/* reinitialize tx_buffer_info */
 	memset(ring->tx_buffer_info, 0,
 	       sizeof(struct wx_tx_buffer) * ring->count);
@@ -1732,7 +1748,9 @@ void wx_configure(struct wx *wx)
 
 	wx_set_rx_mode(wx->netdev);
 	wx_restore_vlan(wx);
-	wx_enable_sec_rx_path(wx);
+
+	if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags))
+		wx->configure_fdir(wx);
 
 	wx_configure_tx(wx);
 	wx_configure_rx(wx);
@@ -1959,6 +1977,7 @@ int wx_sw_init(struct wx *wx)
 	}
 
 	bitmap_zero(wx->state, WX_STATE_NBITS);
+	bitmap_zero(wx->flags, WX_PF_FLAGS_NBITS);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.h b/drivers/net/ethernet/wangxun/libwx/wx_hw.h
index 9e219fa717a2..11fb33349482 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_hw.h
+++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.h
@@ -28,6 +28,8 @@ void wx_mac_set_default_filter(struct wx *wx, u8 *addr);
 void wx_flush_sw_mac_table(struct wx *wx);
 int wx_set_mac(struct net_device *netdev, void *p);
 void wx_disable_rx(struct wx *wx);
+int wx_disable_sec_rx_path(struct wx *wx);
+void wx_enable_sec_rx_path(struct wx *wx);
 void wx_set_rx_mode(struct net_device *netdev);
 int wx_change_mtu(struct net_device *netdev, int new_mtu);
 void wx_disable_rx_queue(struct wx *wx, struct wx_ring *ring);
diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_lib.c
index 68bde91b67a0..6fbdbb6c9bb5 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_lib.c
+++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.c
@@ -148,10 +148,11 @@ static struct wx_dec_ptype wx_ptype_lookup[256] = {
 	[0xFD] = WX_PTT(IP, IPV6, IGMV, IPV6, SCTP, PAY4),
 };
 
-static struct wx_dec_ptype wx_decode_ptype(const u8 ptype)
+struct wx_dec_ptype wx_decode_ptype(const u8 ptype)
 {
 	return wx_ptype_lookup[ptype];
 }
+EXPORT_SYMBOL(wx_decode_ptype);
 
 /* wx_test_staterr - tests bits in Rx descriptor status and error fields */
 static __le32 wx_test_staterr(union wx_rx_desc *rx_desc,
@@ -1453,6 +1454,7 @@ static void wx_tx_csum(struct wx_ring *tx_ring, struct wx_tx_buffer *first,
 static netdev_tx_t wx_xmit_frame_ring(struct sk_buff *skb,
 				      struct wx_ring *tx_ring)
 {
+	struct wx *wx = netdev_priv(tx_ring->netdev);
 	u16 count = TXD_USE_COUNT(skb_headlen(skb));
 	struct wx_tx_buffer *first;
 	u8 hdr_len = 0, ptype;
@@ -1498,6 +1500,10 @@ static netdev_tx_t wx_xmit_frame_ring(struct sk_buff *skb,
 		goto out_drop;
 	else if (!tso)
 		wx_tx_csum(tx_ring, first, ptype);
+
+	if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags) && tx_ring->atr_sample_rate)
+		wx->atr(tx_ring, first, ptype);
+
 	wx_tx_map(tx_ring, first, hdr_len);
 
 	return NETDEV_TX_OK;
@@ -1574,8 +1580,27 @@ static void wx_set_rss_queues(struct wx *wx)
 	f = &wx->ring_feature[RING_F_RSS];
 	f->indices = f->limit;
 
-	wx->num_rx_queues = f->limit;
-	wx->num_tx_queues = f->limit;
+	if (!(test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags)))
+		goto out;
+
+	clear_bit(WX_FLAG_FDIR_HASH, wx->flags);
+
+	/* Use Flow Director in addition to RSS to ensure the best
+	 * distribution of flows across cores, even when an FDIR flow
+	 * isn't matched.
+	 */
+	if (f->indices > 1) {
+		f = &wx->ring_feature[RING_F_FDIR];
+
+		f->indices = f->limit;
+
+		if (!(test_bit(WX_FLAG_FDIR_PERFECT, wx->flags)))
+			set_bit(WX_FLAG_FDIR_HASH, wx->flags);
+	}
+
+out:
+	wx->num_rx_queues = f->indices;
+	wx->num_tx_queues = f->indices;
 }
 
 static void wx_set_num_queues(struct wx *wx)
diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.h b/drivers/net/ethernet/wangxun/libwx/wx_lib.h
index c41b29ea812f..fdeb0c315b75 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_lib.h
+++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.h
@@ -7,6 +7,7 @@
 #ifndef _WX_LIB_H_
 #define _WX_LIB_H_
 
+struct wx_dec_ptype wx_decode_ptype(const u8 ptype);
 void wx_alloc_rx_buffers(struct wx_ring *rx_ring, u16 cleaned_count);
 u16 wx_desc_unused(struct wx_ring *ring);
 netdev_tx_t wx_xmit_frame(struct sk_buff *skb,
diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h
index 5aaf7b1fa2db..b1f9bab06e90 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_type.h
+++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h
@@ -503,6 +503,34 @@ enum WX_MSCA_CMD_value {
 #define WX_PTYPE_TYP_TCP             0x04
 #define WX_PTYPE_TYP_SCTP            0x05
 
+/* Packet type non-ip values */
+enum wx_l2_ptypes {
+	WX_PTYPE_L2_ABORTED = (WX_PTYPE_PKT_MAC),
+	WX_PTYPE_L2_MAC = (WX_PTYPE_PKT_MAC | WX_PTYPE_TYP_MAC),
+
+	WX_PTYPE_L2_IPV4_FRAG = (WX_PTYPE_PKT_IP | WX_PTYPE_TYP_IPFRAG),
+	WX_PTYPE_L2_IPV4 = (WX_PTYPE_PKT_IP | WX_PTYPE_TYP_IP),
+	WX_PTYPE_L2_IPV4_UDP = (WX_PTYPE_PKT_IP | WX_PTYPE_TYP_UDP),
+	WX_PTYPE_L2_IPV4_TCP = (WX_PTYPE_PKT_IP | WX_PTYPE_TYP_TCP),
+	WX_PTYPE_L2_IPV4_SCTP = (WX_PTYPE_PKT_IP | WX_PTYPE_TYP_SCTP),
+	WX_PTYPE_L2_IPV6_FRAG = (WX_PTYPE_PKT_IP | WX_PTYPE_PKT_IPV6 |
+				 WX_PTYPE_TYP_IPFRAG),
+	WX_PTYPE_L2_IPV6 = (WX_PTYPE_PKT_IP | WX_PTYPE_PKT_IPV6 |
+			    WX_PTYPE_TYP_IP),
+	WX_PTYPE_L2_IPV6_UDP = (WX_PTYPE_PKT_IP | WX_PTYPE_PKT_IPV6 |
+				WX_PTYPE_TYP_UDP),
+	WX_PTYPE_L2_IPV6_TCP = (WX_PTYPE_PKT_IP | WX_PTYPE_PKT_IPV6 |
+				WX_PTYPE_TYP_TCP),
+	WX_PTYPE_L2_IPV6_SCTP = (WX_PTYPE_PKT_IP | WX_PTYPE_PKT_IPV6 |
+				 WX_PTYPE_TYP_SCTP),
+
+	WX_PTYPE_L2_TUN4_MAC = (WX_PTYPE_TUN_IPV4 | WX_PTYPE_PKT_IGM),
+	WX_PTYPE_L2_TUN6_MAC = (WX_PTYPE_TUN_IPV6 | WX_PTYPE_PKT_IGM),
+};
+
+#define WX_PTYPE_PKT(_pt)            ((_pt) & 0x30)
+#define WX_PTYPE_TYPL4(_pt)          ((_pt) & 0x07)
+
 #define WX_RXD_PKTTYPE(_rxd) \
 	((le32_to_cpu((_rxd)->wb.lower.lo_dword.data) >> 9) & 0xFF)
 #define WX_RXD_IPV6EX(_rxd) \
@@ -552,6 +580,9 @@ enum wx_tx_flags {
 	WX_TX_FLAGS_OUTER_IPV4	= 0x100,
 	WX_TX_FLAGS_LINKSEC	= 0x200,
 	WX_TX_FLAGS_IPSEC	= 0x400,
+
+	/* software defined flags */
+	WX_TX_FLAGS_SW_VLAN	= 0x40,
 };
 
 /* VLAN info */
@@ -900,7 +931,13 @@ struct wx_ring {
 					 */
 	u16 next_to_use;
 	u16 next_to_clean;
-	u16 next_to_alloc;
+	union {
+		u16 next_to_alloc;
+		struct {
+			u8 atr_sample_rate;
+			u8 atr_count;
+		};
+	};
 
 	struct wx_queue_stats stats;
 	struct u64_stats_sync syncp;
@@ -939,6 +976,7 @@ struct wx_ring_feature {
 enum wx_ring_f_enum {
 	RING_F_NONE = 0,
 	RING_F_RSS,
+	RING_F_FDIR,
 	RING_F_ARRAY_SIZE  /* must be last in enum set */
 };
 
@@ -986,9 +1024,18 @@ enum wx_state {
 	WX_STATE_RESETTING,
 	WX_STATE_NBITS,		/* must be last */
 };
+
+enum wx_pf_flags {
+	WX_FLAG_FDIR_CAPABLE,
+	WX_FLAG_FDIR_HASH,
+	WX_FLAG_FDIR_PERFECT,
+	WX_PF_FLAGS_NBITS               /* must be last */
+};
+
 struct wx {
 	unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
 	DECLARE_BITMAP(state, WX_STATE_NBITS);
+	DECLARE_BITMAP(flags, WX_PF_FLAGS_NBITS);
 
 	void *priv;
 	u8 __iomem *hw_addr;
@@ -1077,6 +1124,9 @@ struct wx {
 	u64 hw_csum_rx_error;
 	u64 alloc_rx_buff_failed;
 
+	u32 atr_sample_rate;
+	void (*atr)(struct wx_ring *ring, struct wx_tx_buffer *first, u8 ptype);
+	void (*configure_fdir)(struct wx *wx);
 	void (*do_reset)(struct net_device *netdev);
 };
 
diff --git a/drivers/net/ethernet/wangxun/txgbe/Makefile b/drivers/net/ethernet/wangxun/txgbe/Makefile
index 42718875277c..f74576fe7062 100644
--- a/drivers/net/ethernet/wangxun/txgbe/Makefile
+++ b/drivers/net/ethernet/wangxun/txgbe/Makefile
@@ -10,4 +10,5 @@ txgbe-objs := txgbe_main.o \
               txgbe_hw.o \
               txgbe_phy.o \
               txgbe_irq.o \
+              txgbe_fdir.o \
               txgbe_ethtool.o
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c
new file mode 100644
index 000000000000..6158209e84cb
--- /dev/null
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c
@@ -0,0 +1,302 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2015 - 2024 Beijing WangXun Technology Co., Ltd. */
+
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+
+#include "../libwx/wx_type.h"
+#include "../libwx/wx_lib.h"
+#include "../libwx/wx_hw.h"
+#include "txgbe_type.h"
+#include "txgbe_fdir.h"
+
+/* These defines allow us to quickly generate all of the necessary instructions
+ * in the function below by simply calling out TXGBE_COMPUTE_SIG_HASH_ITERATION
+ * for values 0 through 15
+ */
+#define TXGBE_ATR_COMMON_HASH_KEY \
+		(TXGBE_ATR_BUCKET_HASH_KEY & TXGBE_ATR_SIGNATURE_HASH_KEY)
+#define TXGBE_COMPUTE_SIG_HASH_ITERATION(_n) \
+do { \
+	u32 n = (_n); \
+	if (TXGBE_ATR_COMMON_HASH_KEY & (0x01 << n)) \
+		common_hash ^= lo_hash_dword >> n; \
+	else if (TXGBE_ATR_BUCKET_HASH_KEY & (0x01 << n)) \
+		bucket_hash ^= lo_hash_dword >> n; \
+	else if (TXGBE_ATR_SIGNATURE_HASH_KEY & (0x01 << n)) \
+		sig_hash ^= lo_hash_dword << (16 - n); \
+	if (TXGBE_ATR_COMMON_HASH_KEY & (0x01 << (n + 16))) \
+		common_hash ^= hi_hash_dword >> n; \
+	else if (TXGBE_ATR_BUCKET_HASH_KEY & (0x01 << (n + 16))) \
+		bucket_hash ^= hi_hash_dword >> n; \
+	else if (TXGBE_ATR_SIGNATURE_HASH_KEY & (0x01 << (n + 16))) \
+		sig_hash ^= hi_hash_dword << (16 - n); \
+} while (0)
+
+/**
+ *  txgbe_atr_compute_sig_hash - Compute the signature hash
+ *  @input: input bitstream to compute the hash on
+ *  @common: compressed common input dword
+ *  @hash: pointer to the computed hash
+ *
+ *  This function is almost identical to the function above but contains
+ *  several optimizations such as unwinding all of the loops, letting the
+ *  compiler work out all of the conditional ifs since the keys are static
+ *  defines, and computing two keys at once since the hashed dword stream
+ *  will be the same for both keys.
+ **/
+static void txgbe_atr_compute_sig_hash(union txgbe_atr_hash_dword input,
+				       union txgbe_atr_hash_dword common,
+				       u32 *hash)
+{
+	u32 sig_hash = 0, bucket_hash = 0, common_hash = 0;
+	u32 hi_hash_dword, lo_hash_dword, flow_vm_vlan;
+	u32 i;
+
+	/* record the flow_vm_vlan bits as they are a key part to the hash */
+	flow_vm_vlan = ntohl(input.dword);
+
+	/* generate common hash dword */
+	hi_hash_dword = ntohl(common.dword);
+
+	/* low dword is word swapped version of common */
+	lo_hash_dword = (hi_hash_dword >> 16) | (hi_hash_dword << 16);
+
+	/* apply flow ID/VM pool/VLAN ID bits to hash words */
+	hi_hash_dword ^= flow_vm_vlan ^ (flow_vm_vlan >> 16);
+
+	/* Process bits 0 and 16 */
+	TXGBE_COMPUTE_SIG_HASH_ITERATION(0);
+
+	/* apply flow ID/VM pool/VLAN ID bits to lo hash dword, we had to
+	 * delay this because bit 0 of the stream should not be processed
+	 * so we do not add the VLAN until after bit 0 was processed
+	 */
+	lo_hash_dword ^= flow_vm_vlan ^ (flow_vm_vlan << 16);
+
+	/* Process remaining 30 bit of the key */
+	for (i = 1; i <= 15; i++)
+		TXGBE_COMPUTE_SIG_HASH_ITERATION(i);
+
+	/* combine common_hash result with signature and bucket hashes */
+	bucket_hash ^= common_hash;
+	bucket_hash &= TXGBE_ATR_HASH_MASK;
+
+	sig_hash ^= common_hash << 16;
+	sig_hash &= TXGBE_ATR_HASH_MASK << 16;
+
+	/* return completed signature hash */
+	*hash = sig_hash ^ bucket_hash;
+}
+
+static int txgbe_fdir_check_cmd_complete(struct wx *wx)
+{
+	u32 val;
+
+	return read_poll_timeout_atomic(rd32, val,
+					!(val & TXGBE_RDB_FDIR_CMD_CMD_MASK),
+					10, 100, false,
+					wx, TXGBE_RDB_FDIR_CMD);
+}
+
+/**
+ *  txgbe_fdir_add_signature_filter - Adds a signature hash filter
+ *  @wx: pointer to hardware structure
+ *  @input: unique input dword
+ *  @common: compressed common input dword
+ *  @queue: queue index to direct traffic to
+ *
+ *  @return: 0 on success and negative on failure
+ **/
+static int txgbe_fdir_add_signature_filter(struct wx *wx,
+					   union txgbe_atr_hash_dword input,
+					   union txgbe_atr_hash_dword common,
+					   u8 queue)
+{
+	u32 fdirhashcmd, fdircmd;
+	u8 flow_type;
+	int err;
+
+	/* Get the flow_type in order to program FDIRCMD properly
+	 * lowest 2 bits are FDIRCMD.L4TYPE, third lowest bit is FDIRCMD.IPV6
+	 * fifth is FDIRCMD.TUNNEL_FILTER
+	 */
+	flow_type = input.formatted.flow_type;
+	switch (flow_type) {
+	case TXGBE_ATR_FLOW_TYPE_TCPV4:
+	case TXGBE_ATR_FLOW_TYPE_UDPV4:
+	case TXGBE_ATR_FLOW_TYPE_SCTPV4:
+	case TXGBE_ATR_FLOW_TYPE_TCPV6:
+	case TXGBE_ATR_FLOW_TYPE_UDPV6:
+	case TXGBE_ATR_FLOW_TYPE_SCTPV6:
+		break;
+	default:
+		wx_err(wx, "Error on flow type input\n");
+		return -EINVAL;
+	}
+
+	/* configure FDIRCMD register */
+	fdircmd = TXGBE_RDB_FDIR_CMD_CMD_ADD_FLOW |
+		  TXGBE_RDB_FDIR_CMD_FILTER_UPDATE |
+		  TXGBE_RDB_FDIR_CMD_LAST | TXGBE_RDB_FDIR_CMD_QUEUE_EN;
+	fdircmd |= TXGBE_RDB_FDIR_CMD_FLOW_TYPE(flow_type);
+	fdircmd |= TXGBE_RDB_FDIR_CMD_RX_QUEUE(queue);
+
+	txgbe_atr_compute_sig_hash(input, common, &fdirhashcmd);
+	fdirhashcmd |= TXGBE_RDB_FDIR_HASH_BUCKET_VALID;
+	wr32(wx, TXGBE_RDB_FDIR_HASH, fdirhashcmd);
+	wr32(wx, TXGBE_RDB_FDIR_CMD, fdircmd);
+
+	wx_dbg(wx, "Tx Queue=%x hash=%x\n", queue, (u32)fdirhashcmd);
+
+	err = txgbe_fdir_check_cmd_complete(wx);
+	if (err)
+		wx_err(wx, "Flow Director command did not complete!\n");
+
+	return err;
+}
+
+void txgbe_atr(struct wx_ring *ring, struct wx_tx_buffer *first, u8 ptype)
+{
+	union txgbe_atr_hash_dword common = { .dword = 0 };
+	union txgbe_atr_hash_dword input = { .dword = 0 };
+	struct wx_q_vector *q_vector = ring->q_vector;
+	struct wx_dec_ptype dptype;
+	union network_header {
+		struct ipv6hdr *ipv6;
+		struct iphdr *ipv4;
+		void *raw;
+	} hdr;
+	struct tcphdr *th;
+
+	/* if ring doesn't have a interrupt vector, cannot perform ATR */
+	if (!q_vector)
+		return;
+
+	ring->atr_count++;
+	dptype = wx_decode_ptype(ptype);
+	if (dptype.etype) {
+		if (WX_PTYPE_TYPL4(ptype) != WX_PTYPE_TYP_TCP)
+			return;
+		hdr.raw = (void *)skb_inner_network_header(first->skb);
+		th = inner_tcp_hdr(first->skb);
+	} else {
+		if (WX_PTYPE_PKT(ptype) != WX_PTYPE_PKT_IP ||
+		    WX_PTYPE_TYPL4(ptype) != WX_PTYPE_TYP_TCP)
+			return;
+		hdr.raw = (void *)skb_network_header(first->skb);
+		th = tcp_hdr(first->skb);
+	}
+
+	/* skip this packet since it is invalid or the socket is closing */
+	if (!th || th->fin)
+		return;
+
+	/* sample on all syn packets or once every atr sample count */
+	if (!th->syn && ring->atr_count < ring->atr_sample_rate)
+		return;
+
+	/* reset sample count */
+	ring->atr_count = 0;
+
+	/* src and dst are inverted, think how the receiver sees them
+	 *
+	 * The input is broken into two sections, a non-compressed section
+	 * containing vm_pool, vlan_id, and flow_type.  The rest of the data
+	 * is XORed together and stored in the compressed dword.
+	 */
+	input.formatted.vlan_id = htons((u16)ptype);
+
+	/* since src port and flex bytes occupy the same word XOR them together
+	 * and write the value to source port portion of compressed dword
+	 */
+	if (first->tx_flags & WX_TX_FLAGS_SW_VLAN)
+		common.port.src ^= th->dest ^ first->skb->protocol;
+	else if (first->tx_flags & WX_TX_FLAGS_HW_VLAN)
+		common.port.src ^= th->dest ^ first->skb->vlan_proto;
+	else
+		common.port.src ^= th->dest ^ first->protocol;
+	common.port.dst ^= th->source;
+
+	if (WX_PTYPE_PKT_IPV6 & WX_PTYPE_PKT(ptype)) {
+		input.formatted.flow_type = TXGBE_ATR_FLOW_TYPE_TCPV6;
+		common.ip ^= hdr.ipv6->saddr.s6_addr32[0] ^
+					 hdr.ipv6->saddr.s6_addr32[1] ^
+					 hdr.ipv6->saddr.s6_addr32[2] ^
+					 hdr.ipv6->saddr.s6_addr32[3] ^
+					 hdr.ipv6->daddr.s6_addr32[0] ^
+					 hdr.ipv6->daddr.s6_addr32[1] ^
+					 hdr.ipv6->daddr.s6_addr32[2] ^
+					 hdr.ipv6->daddr.s6_addr32[3];
+	} else {
+		input.formatted.flow_type = TXGBE_ATR_FLOW_TYPE_TCPV4;
+		common.ip ^= hdr.ipv4->saddr ^ hdr.ipv4->daddr;
+	}
+
+	/* This assumes the Rx queue and Tx queue are bound to the same CPU */
+	txgbe_fdir_add_signature_filter(q_vector->wx, input, common,
+					ring->queue_index);
+}
+
+/**
+ *  txgbe_fdir_enable - Initialize Flow Director control registers
+ *  @wx: pointer to hardware structure
+ *  @fdirctrl: value to write to flow director control register
+ **/
+static void txgbe_fdir_enable(struct wx *wx, u32 fdirctrl)
+{
+	u32 val;
+	int ret;
+
+	/* Prime the keys for hashing */
+	wr32(wx, TXGBE_RDB_FDIR_HKEY, TXGBE_ATR_BUCKET_HASH_KEY);
+	wr32(wx, TXGBE_RDB_FDIR_SKEY, TXGBE_ATR_SIGNATURE_HASH_KEY);
+
+	wr32(wx, TXGBE_RDB_FDIR_CTL, fdirctrl);
+	WX_WRITE_FLUSH(wx);
+	ret = read_poll_timeout(rd32, val, val & TXGBE_RDB_FDIR_CTL_INIT_DONE,
+				1000, 10000, false, wx, TXGBE_RDB_FDIR_CTL);
+
+	if (ret < 0)
+		wx_dbg(wx, "Flow Director poll time exceeded!\n");
+}
+
+/**
+ *  txgbe_init_fdir_signature -Initialize Flow Director sig filters
+ *  @wx: pointer to hardware structure
+ **/
+static void txgbe_init_fdir_signature(struct wx *wx)
+{
+	u32 fdirctrl = TXGBE_FDIR_PBALLOC_64K;
+	u32 flex = 0;
+
+	flex = rd32(wx, TXGBE_RDB_FDIR_FLEX_CFG(0));
+	flex &= ~TXGBE_RDB_FDIR_FLEX_CFG_FIELD0;
+
+	flex |= (TXGBE_RDB_FDIR_FLEX_CFG_BASE_MAC |
+		 TXGBE_RDB_FDIR_FLEX_CFG_OFST(0x6));
+	wr32(wx, TXGBE_RDB_FDIR_FLEX_CFG(0), flex);
+
+	/* Continue setup of fdirctrl register bits:
+	 *  Move the flexible bytes to use the ethertype - shift 6 words
+	 *  Set the maximum length per hash bucket to 0xA filters
+	 *  Send interrupt when 64 filters are left
+	 */
+	fdirctrl |= TXGBE_RDB_FDIR_CTL_HASH_BITS(0xF) |
+		    TXGBE_RDB_FDIR_CTL_MAX_LENGTH(0xA) |
+		    TXGBE_RDB_FDIR_CTL_FULL_THRESH(4);
+
+	/* write hashes and fdirctrl register, poll for completion */
+	txgbe_fdir_enable(wx, fdirctrl);
+}
+
+void txgbe_configure_fdir(struct wx *wx)
+{
+	wx_disable_sec_rx_path(wx);
+
+	if (test_bit(WX_FLAG_FDIR_HASH, wx->flags))
+		txgbe_init_fdir_signature(wx);
+
+	wx_enable_sec_rx_path(wx);
+}
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h
new file mode 100644
index 000000000000..ed245b66dc2a
--- /dev/null
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h
@@ -0,0 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2015 - 2024 Beijing WangXun Technology Co., Ltd. */
+
+#ifndef _TXGBE_FDIR_H_
+#define _TXGBE_FDIR_H_
+
+void txgbe_atr(struct wx_ring *ring, struct wx_tx_buffer *first, u8 ptype);
+void txgbe_configure_fdir(struct wx *wx);
+
+#endif /* _TXGBE_FDIR_H_ */
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
index 8c7a74981b90..ce49fb725541 100644
--- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
@@ -18,6 +18,7 @@
 #include "txgbe_hw.h"
 #include "txgbe_phy.h"
 #include "txgbe_irq.h"
+#include "txgbe_fdir.h"
 #include "txgbe_ethtool.h"
 
 char txgbe_driver_name[] = "txgbe";
@@ -257,6 +258,14 @@ static int txgbe_sw_init(struct wx *wx)
 						   num_online_cpus());
 	wx->rss_enabled = true;
 
+	wx->ring_feature[RING_F_FDIR].limit = min_t(int, TXGBE_MAX_FDIR_INDICES,
+						    num_online_cpus());
+	set_bit(WX_FLAG_FDIR_CAPABLE, wx->flags);
+	set_bit(WX_FLAG_FDIR_HASH, wx->flags);
+	wx->atr_sample_rate = TXGBE_DEFAULT_ATR_SAMPLE_RATE;
+	wx->atr = txgbe_atr;
+	wx->configure_fdir = txgbe_configure_fdir;
+
 	/* enable itr by default in dynamic mode */
 	wx->rx_itr_setting = 1;
 	wx->tx_itr_setting = 1;
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h
index f434a7865cb7..5b8c55df35fe 100644
--- a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h
@@ -89,6 +89,39 @@
 #define TXGBE_XPCS_IDA_ADDR                     0x13000
 #define TXGBE_XPCS_IDA_DATA                     0x13004
 
+/********************************* Flow Director *****************************/
+#define TXGBE_RDB_FDIR_CTL                      0x19500
+#define TXGBE_RDB_FDIR_CTL_INIT_DONE            BIT(3)
+#define TXGBE_RDB_FDIR_CTL_PERFECT_MATCH        BIT(4)
+#define TXGBE_RDB_FDIR_CTL_DROP_Q(v)            FIELD_PREP(GENMASK(14, 8), v)
+#define TXGBE_RDB_FDIR_CTL_HASH_BITS(v)         FIELD_PREP(GENMASK(23, 20), v)
+#define TXGBE_RDB_FDIR_CTL_MAX_LENGTH(v)        FIELD_PREP(GENMASK(27, 24), v)
+#define TXGBE_RDB_FDIR_CTL_FULL_THRESH(v)       FIELD_PREP(GENMASK(31, 28), v)
+#define TXGBE_RDB_FDIR_HASH                     0x19528
+#define TXGBE_RDB_FDIR_HASH_SIG_SW_INDEX(v)     FIELD_PREP(GENMASK(31, 16), v)
+#define TXGBE_RDB_FDIR_HASH_BUCKET_VALID        BIT(15)
+#define TXGBE_RDB_FDIR_CMD                      0x1952C
+#define TXGBE_RDB_FDIR_CMD_CMD_MASK             GENMASK(1, 0)
+#define TXGBE_RDB_FDIR_CMD_CMD(v)               FIELD_PREP(GENMASK(1, 0), v)
+#define TXGBE_RDB_FDIR_CMD_CMD_ADD_FLOW         TXGBE_RDB_FDIR_CMD_CMD(1)
+#define TXGBE_RDB_FDIR_CMD_CMD_REMOVE_FLOW      TXGBE_RDB_FDIR_CMD_CMD(2)
+#define TXGBE_RDB_FDIR_CMD_CMD_QUERY_REM_FILT   TXGBE_RDB_FDIR_CMD_CMD(3)
+#define TXGBE_RDB_FDIR_CMD_FILTER_VALID         BIT(2)
+#define TXGBE_RDB_FDIR_CMD_FILTER_UPDATE        BIT(3)
+#define TXGBE_RDB_FDIR_CMD_FLOW_TYPE(v)         FIELD_PREP(GENMASK(6, 5), v)
+#define TXGBE_RDB_FDIR_CMD_DROP                 BIT(9)
+#define TXGBE_RDB_FDIR_CMD_LAST                 BIT(11)
+#define TXGBE_RDB_FDIR_CMD_QUEUE_EN             BIT(15)
+#define TXGBE_RDB_FDIR_CMD_RX_QUEUE(v)          FIELD_PREP(GENMASK(22, 16), v)
+#define TXGBE_RDB_FDIR_CMD_VT_POOL(v)           FIELD_PREP(GENMASK(29, 24), v)
+#define TXGBE_RDB_FDIR_HKEY                     0x19568
+#define TXGBE_RDB_FDIR_SKEY                     0x1956C
+#define TXGBE_RDB_FDIR_FLEX_CFG(_i)             (0x19580 + ((_i) * 4))
+#define TXGBE_RDB_FDIR_FLEX_CFG_FIELD0          GENMASK(7, 0)
+#define TXGBE_RDB_FDIR_FLEX_CFG_BASE_MAC        FIELD_PREP(GENMASK(1, 0), 0)
+#define TXGBE_RDB_FDIR_FLEX_CFG_MSK             BIT(2)
+#define TXGBE_RDB_FDIR_FLEX_CFG_OFST(v)         FIELD_PREP(GENMASK(7, 3), v)
+
 /* Checksum and EEPROM pointers */
 #define TXGBE_EEPROM_LAST_WORD                  0x800
 #define TXGBE_EEPROM_CHECKSUM                   0x2F
@@ -112,6 +145,91 @@
 #define TXGBE_SP_RX_PB_SIZE     512
 #define TXGBE_SP_TDB_PB_SZ      (160 * 1024) /* 160KB Packet Buffer */
 
+#define TXGBE_DEFAULT_ATR_SAMPLE_RATE           20
+
+/* Software ATR hash keys */
+#define TXGBE_ATR_BUCKET_HASH_KEY               0x3DAD14E2
+#define TXGBE_ATR_SIGNATURE_HASH_KEY            0x174D3614
+
+/* Software ATR input stream values and masks */
+#define TXGBE_ATR_HASH_MASK                     0x7fff
+#define TXGBE_ATR_L4TYPE_MASK                   0x3
+#define TXGBE_ATR_L4TYPE_UDP                    0x1
+#define TXGBE_ATR_L4TYPE_TCP                    0x2
+#define TXGBE_ATR_L4TYPE_SCTP                   0x3
+#define TXGBE_ATR_L4TYPE_IPV6_MASK              0x4
+#define TXGBE_ATR_L4TYPE_TUNNEL_MASK            0x10
+
+enum txgbe_atr_flow_type {
+	TXGBE_ATR_FLOW_TYPE_IPV4                = 0x0,
+	TXGBE_ATR_FLOW_TYPE_UDPV4               = 0x1,
+	TXGBE_ATR_FLOW_TYPE_TCPV4               = 0x2,
+	TXGBE_ATR_FLOW_TYPE_SCTPV4              = 0x3,
+	TXGBE_ATR_FLOW_TYPE_IPV6                = 0x4,
+	TXGBE_ATR_FLOW_TYPE_UDPV6               = 0x5,
+	TXGBE_ATR_FLOW_TYPE_TCPV6               = 0x6,
+	TXGBE_ATR_FLOW_TYPE_SCTPV6              = 0x7,
+	TXGBE_ATR_FLOW_TYPE_TUNNELED_IPV4       = 0x10,
+	TXGBE_ATR_FLOW_TYPE_TUNNELED_UDPV4      = 0x11,
+	TXGBE_ATR_FLOW_TYPE_TUNNELED_TCPV4      = 0x12,
+	TXGBE_ATR_FLOW_TYPE_TUNNELED_SCTPV4     = 0x13,
+	TXGBE_ATR_FLOW_TYPE_TUNNELED_IPV6       = 0x14,
+	TXGBE_ATR_FLOW_TYPE_TUNNELED_UDPV6      = 0x15,
+	TXGBE_ATR_FLOW_TYPE_TUNNELED_TCPV6      = 0x16,
+	TXGBE_ATR_FLOW_TYPE_TUNNELED_SCTPV6     = 0x17,
+};
+
+/* Flow Director ATR input struct. */
+union txgbe_atr_input {
+	/* Byte layout in order, all values with MSB first:
+	 *
+	 * vm_pool    - 1 byte
+	 * flow_type  - 1 byte
+	 * vlan_id    - 2 bytes
+	 * dst_ip     - 16 bytes
+	 * src_ip     - 16 bytes
+	 * src_port   - 2 bytes
+	 * dst_port   - 2 bytes
+	 * flex_bytes - 2 bytes
+	 * bkt_hash   - 2 bytes
+	 */
+	struct {
+		u8 vm_pool;
+		u8 flow_type;
+		__be16 vlan_id;
+		__be32 dst_ip[4];
+		__be32 src_ip[4];
+		__be16 src_port;
+		__be16 dst_port;
+		__be16 flex_bytes;
+		__be16 bkt_hash;
+	} formatted;
+	__be32 dword_stream[11];
+};
+
+/* Flow Director compressed ATR hash input struct */
+union txgbe_atr_hash_dword {
+	struct {
+		u8 vm_pool;
+		u8 flow_type;
+		__be16 vlan_id;
+	} formatted;
+	__be32 ip;
+	struct {
+		__be16 src;
+		__be16 dst;
+	} port;
+	__be16 flex_bytes;
+	__be32 dword;
+};
+
+enum txgbe_fdir_pballoc_type {
+	TXGBE_FDIR_PBALLOC_NONE = 0,
+	TXGBE_FDIR_PBALLOC_64K  = 1,
+	TXGBE_FDIR_PBALLOC_128K = 2,
+	TXGBE_FDIR_PBALLOC_256K = 3,
+};
+
 /* TX/RX descriptor defines */
 #define TXGBE_DEFAULT_TXD               512
 #define TXGBE_DEFAULT_TX_WORK           256
@@ -196,6 +314,9 @@ struct txgbe {
 	struct gpio_chip *gpio;
 	unsigned int gpio_irq;
 	unsigned int link_irq;
+
+	/* flow director */
+	union txgbe_atr_input fdir_mask;
 };
 
 #endif /* _TXGBE_TYPE_H_ */
-- 
cgit v1.2.3-58-ga151


From 4bdb441105dcea40ac6e01756314c0fdf5173ce1 Mon Sep 17 00:00:00 2001
From: Jiawen Wu <jiawenwu@trustnetic.com>
Date: Tue, 18 Jun 2024 18:16:08 +0800
Subject: net: txgbe: support Flow Director perfect filters

Support the addition and deletion of Flow Director filters.

Supported fields: src-ip, dst-ip, src-port, dst-port
Supported flow-types: tcp4, udp4, sctp4, ipv4

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/wangxun/libwx/wx_lib.c        |  31 ++
 drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c | 427 +++++++++++++++++++++
 drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c    | 343 ++++++++++++++++-
 drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h    |  10 +
 drivers/net/ethernet/wangxun/txgbe/txgbe_main.c    |   9 +
 drivers/net/ethernet/wangxun/txgbe/txgbe_type.h    |  26 ++
 6 files changed, 845 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/wangxun/libwx/wx_lib.c b/drivers/net/ethernet/wangxun/libwx/wx_lib.c
index 6fbdbb6c9bb5..ac0e1d42fe55 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_lib.c
+++ b/drivers/net/ethernet/wangxun/libwx/wx_lib.c
@@ -2705,6 +2705,7 @@ int wx_set_features(struct net_device *netdev, netdev_features_t features)
 {
 	netdev_features_t changed = netdev->features ^ features;
 	struct wx *wx = netdev_priv(netdev);
+	bool need_reset = false;
 
 	if (features & NETIF_F_RXHASH) {
 		wr32m(wx, WX_RDB_RA_CTL, WX_RDB_RA_CTL_RSS_EN,
@@ -2722,6 +2723,36 @@ int wx_set_features(struct net_device *netdev, netdev_features_t features)
 	else if (changed & (NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER))
 		wx_set_rx_mode(netdev);
 
+	if (!(test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags)))
+		return 0;
+
+	/* Check if Flow Director n-tuple support was enabled or disabled.  If
+	 * the state changed, we need to reset.
+	 */
+	switch (features & NETIF_F_NTUPLE) {
+	case NETIF_F_NTUPLE:
+		/* turn off ATR, enable perfect filters and reset */
+		if (!(test_and_set_bit(WX_FLAG_FDIR_PERFECT, wx->flags)))
+			need_reset = true;
+
+		clear_bit(WX_FLAG_FDIR_HASH, wx->flags);
+		break;
+	default:
+		/* turn off perfect filters, enable ATR and reset */
+		if (test_and_clear_bit(WX_FLAG_FDIR_PERFECT, wx->flags))
+			need_reset = true;
+
+		/* We cannot enable ATR if RSS is disabled */
+		if (wx->ring_feature[RING_F_RSS].limit <= 1)
+			break;
+
+		set_bit(WX_FLAG_FDIR_HASH, wx->flags);
+		break;
+	}
+
+	if (need_reset)
+		wx->do_reset(netdev);
+
 	return 0;
 }
 EXPORT_SYMBOL(wx_set_features);
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c
index 31fde3fa7c6b..d98314b26c19 100644
--- a/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_ethtool.c
@@ -9,6 +9,7 @@
 #include "../libwx/wx_type.h"
 #include "../libwx/wx_lib.h"
 #include "txgbe_type.h"
+#include "txgbe_fdir.h"
 #include "txgbe_ethtool.h"
 
 static int txgbe_set_ringparam(struct net_device *netdev,
@@ -79,6 +80,430 @@ static int txgbe_set_channels(struct net_device *dev,
 	return txgbe_setup_tc(dev, netdev_get_num_tc(dev));
 }
 
+static int txgbe_get_ethtool_fdir_entry(struct txgbe *txgbe,
+					struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fsp =
+		(struct ethtool_rx_flow_spec *)&cmd->fs;
+	union txgbe_atr_input *mask = &txgbe->fdir_mask;
+	struct txgbe_fdir_filter *rule = NULL;
+	struct hlist_node *node;
+
+	/* report total rule count */
+	cmd->data = (1024 << TXGBE_FDIR_PBALLOC_64K) - 2;
+
+	hlist_for_each_entry_safe(rule, node, &txgbe->fdir_filter_list,
+				  fdir_node) {
+		if (fsp->location <= rule->sw_idx)
+			break;
+	}
+
+	if (!rule || fsp->location != rule->sw_idx)
+		return -EINVAL;
+
+	/* set flow type field */
+	switch (rule->filter.formatted.flow_type) {
+	case TXGBE_ATR_FLOW_TYPE_TCPV4:
+		fsp->flow_type = TCP_V4_FLOW;
+		break;
+	case TXGBE_ATR_FLOW_TYPE_UDPV4:
+		fsp->flow_type = UDP_V4_FLOW;
+		break;
+	case TXGBE_ATR_FLOW_TYPE_SCTPV4:
+		fsp->flow_type = SCTP_V4_FLOW;
+		break;
+	case TXGBE_ATR_FLOW_TYPE_IPV4:
+		fsp->flow_type = IP_USER_FLOW;
+		fsp->h_u.usr_ip4_spec.ip_ver = ETH_RX_NFC_IP4;
+		fsp->h_u.usr_ip4_spec.proto = 0;
+		fsp->m_u.usr_ip4_spec.proto = 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	fsp->h_u.tcp_ip4_spec.psrc = rule->filter.formatted.src_port;
+	fsp->m_u.tcp_ip4_spec.psrc = mask->formatted.src_port;
+	fsp->h_u.tcp_ip4_spec.pdst = rule->filter.formatted.dst_port;
+	fsp->m_u.tcp_ip4_spec.pdst = mask->formatted.dst_port;
+	fsp->h_u.tcp_ip4_spec.ip4src = rule->filter.formatted.src_ip[0];
+	fsp->m_u.tcp_ip4_spec.ip4src = mask->formatted.src_ip[0];
+	fsp->h_u.tcp_ip4_spec.ip4dst = rule->filter.formatted.dst_ip[0];
+	fsp->m_u.tcp_ip4_spec.ip4dst = mask->formatted.dst_ip[0];
+	fsp->h_ext.vlan_etype = rule->filter.formatted.flex_bytes;
+	fsp->m_ext.vlan_etype = mask->formatted.flex_bytes;
+	fsp->h_ext.data[1] = htonl(rule->filter.formatted.vm_pool);
+	fsp->m_ext.data[1] = htonl(mask->formatted.vm_pool);
+	fsp->flow_type |= FLOW_EXT;
+
+	/* record action */
+	if (rule->action == TXGBE_RDB_FDIR_DROP_QUEUE)
+		fsp->ring_cookie = RX_CLS_FLOW_DISC;
+	else
+		fsp->ring_cookie = rule->action;
+
+	return 0;
+}
+
+static int txgbe_get_ethtool_fdir_all(struct txgbe *txgbe,
+				      struct ethtool_rxnfc *cmd,
+				      u32 *rule_locs)
+{
+	struct txgbe_fdir_filter *rule;
+	struct hlist_node *node;
+	int cnt = 0;
+
+	/* report total rule count */
+	cmd->data = (1024 << TXGBE_FDIR_PBALLOC_64K) - 2;
+
+	hlist_for_each_entry_safe(rule, node, &txgbe->fdir_filter_list,
+				  fdir_node) {
+		if (cnt == cmd->rule_cnt)
+			return -EMSGSIZE;
+		rule_locs[cnt] = rule->sw_idx;
+		cnt++;
+	}
+
+	cmd->rule_cnt = cnt;
+
+	return 0;
+}
+
+static int txgbe_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
+			   u32 *rule_locs)
+{
+	struct wx *wx = netdev_priv(dev);
+	struct txgbe *txgbe = wx->priv;
+	int ret = -EOPNOTSUPP;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_GRXRINGS:
+		cmd->data = wx->num_rx_queues;
+		ret = 0;
+		break;
+	case ETHTOOL_GRXCLSRLCNT:
+		cmd->rule_cnt = txgbe->fdir_filter_count;
+		ret = 0;
+		break;
+	case ETHTOOL_GRXCLSRULE:
+		ret = txgbe_get_ethtool_fdir_entry(txgbe, cmd);
+		break;
+	case ETHTOOL_GRXCLSRLALL:
+		ret = txgbe_get_ethtool_fdir_all(txgbe, cmd, (u32 *)rule_locs);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+static int txgbe_flowspec_to_flow_type(struct ethtool_rx_flow_spec *fsp,
+				       u8 *flow_type)
+{
+	switch (fsp->flow_type & ~FLOW_EXT) {
+	case TCP_V4_FLOW:
+		*flow_type = TXGBE_ATR_FLOW_TYPE_TCPV4;
+		break;
+	case UDP_V4_FLOW:
+		*flow_type = TXGBE_ATR_FLOW_TYPE_UDPV4;
+		break;
+	case SCTP_V4_FLOW:
+		*flow_type = TXGBE_ATR_FLOW_TYPE_SCTPV4;
+		break;
+	case IP_USER_FLOW:
+		switch (fsp->h_u.usr_ip4_spec.proto) {
+		case IPPROTO_TCP:
+			*flow_type = TXGBE_ATR_FLOW_TYPE_TCPV4;
+			break;
+		case IPPROTO_UDP:
+			*flow_type = TXGBE_ATR_FLOW_TYPE_UDPV4;
+			break;
+		case IPPROTO_SCTP:
+			*flow_type = TXGBE_ATR_FLOW_TYPE_SCTPV4;
+			break;
+		case 0:
+			if (!fsp->m_u.usr_ip4_spec.proto) {
+				*flow_type = TXGBE_ATR_FLOW_TYPE_IPV4;
+				break;
+			}
+			fallthrough;
+		default:
+			return -EINVAL;
+		}
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static bool txgbe_match_ethtool_fdir_entry(struct txgbe *txgbe,
+					   struct txgbe_fdir_filter *input)
+{
+	struct txgbe_fdir_filter *rule = NULL;
+	struct hlist_node *node2;
+
+	hlist_for_each_entry_safe(rule, node2, &txgbe->fdir_filter_list,
+				  fdir_node) {
+		if (rule->filter.formatted.bkt_hash ==
+		    input->filter.formatted.bkt_hash &&
+		    rule->action == input->action) {
+			wx_dbg(txgbe->wx, "FDIR entry already exist\n");
+			return true;
+		}
+	}
+	return false;
+}
+
+static int txgbe_update_ethtool_fdir_entry(struct txgbe *txgbe,
+					   struct txgbe_fdir_filter *input,
+					   u16 sw_idx)
+{
+	struct hlist_node *node = NULL, *parent = NULL;
+	struct txgbe_fdir_filter *rule;
+	struct wx *wx = txgbe->wx;
+	bool deleted = false;
+	int err;
+
+	hlist_for_each_entry_safe(rule, node, &txgbe->fdir_filter_list,
+				  fdir_node) {
+		/* hash found, or no matching entry */
+		if (rule->sw_idx >= sw_idx)
+			break;
+		parent = node;
+	}
+
+	/* if there is an old rule occupying our place remove it */
+	if (rule && rule->sw_idx == sw_idx) {
+		/* hardware filters are only configured when interface is up,
+		 * and we should not issue filter commands while the interface
+		 * is down
+		 */
+		if (netif_running(wx->netdev) &&
+		    (!input || rule->filter.formatted.bkt_hash !=
+		     input->filter.formatted.bkt_hash)) {
+			err = txgbe_fdir_erase_perfect_filter(wx,
+							      &rule->filter,
+							      sw_idx);
+			if (err)
+				return -EINVAL;
+		}
+
+		hlist_del(&rule->fdir_node);
+		kfree(rule);
+		txgbe->fdir_filter_count--;
+		deleted = true;
+	}
+
+	/* If we weren't given an input, then this was a request to delete a
+	 * filter. We should return -EINVAL if the filter wasn't found, but
+	 * return 0 if the rule was successfully deleted.
+	 */
+	if (!input)
+		return deleted ? 0 : -EINVAL;
+
+	/* initialize node and set software index */
+	INIT_HLIST_NODE(&input->fdir_node);
+
+	/* add filter to the list */
+	if (parent)
+		hlist_add_behind(&input->fdir_node, parent);
+	else
+		hlist_add_head(&input->fdir_node,
+			       &txgbe->fdir_filter_list);
+
+	/* update counts */
+	txgbe->fdir_filter_count++;
+
+	return 0;
+}
+
+static int txgbe_add_ethtool_fdir_entry(struct txgbe *txgbe,
+					struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fsp =
+		(struct ethtool_rx_flow_spec *)&cmd->fs;
+	struct txgbe_fdir_filter *input;
+	union txgbe_atr_input mask;
+	struct wx *wx = txgbe->wx;
+	int err = -EINVAL;
+	u16 ptype = 0;
+	u8 queue;
+
+	if (!(test_bit(WX_FLAG_FDIR_PERFECT, wx->flags)))
+		return -EOPNOTSUPP;
+
+	/* ring_cookie is a masked into a set of queues and txgbe pools or
+	 * we use drop index
+	 */
+	if (fsp->ring_cookie == RX_CLS_FLOW_DISC) {
+		queue = TXGBE_RDB_FDIR_DROP_QUEUE;
+	} else {
+		u32 ring = ethtool_get_flow_spec_ring(fsp->ring_cookie);
+
+		if (ring >= wx->num_rx_queues)
+			return -EINVAL;
+
+		/* Map the ring onto the absolute queue index */
+		queue = wx->rx_ring[ring]->reg_idx;
+	}
+
+	/* Don't allow indexes to exist outside of available space */
+	if (fsp->location >= ((1024 << TXGBE_FDIR_PBALLOC_64K) - 2)) {
+		wx_err(wx, "Location out of range\n");
+		return -EINVAL;
+	}
+
+	input = kzalloc(sizeof(*input), GFP_ATOMIC);
+	if (!input)
+		return -ENOMEM;
+
+	memset(&mask, 0, sizeof(union txgbe_atr_input));
+
+	/* set SW index */
+	input->sw_idx = fsp->location;
+
+	/* record flow type */
+	if (txgbe_flowspec_to_flow_type(fsp,
+					&input->filter.formatted.flow_type)) {
+		wx_err(wx, "Unrecognized flow type\n");
+		goto err_out;
+	}
+
+	mask.formatted.flow_type = TXGBE_ATR_L4TYPE_IPV6_MASK |
+				   TXGBE_ATR_L4TYPE_MASK;
+
+	if (input->filter.formatted.flow_type == TXGBE_ATR_FLOW_TYPE_IPV4)
+		mask.formatted.flow_type &= TXGBE_ATR_L4TYPE_IPV6_MASK;
+
+	/* Copy input into formatted structures */
+	input->filter.formatted.src_ip[0] = fsp->h_u.tcp_ip4_spec.ip4src;
+	mask.formatted.src_ip[0] = fsp->m_u.tcp_ip4_spec.ip4src;
+	input->filter.formatted.dst_ip[0] = fsp->h_u.tcp_ip4_spec.ip4dst;
+	mask.formatted.dst_ip[0] = fsp->m_u.tcp_ip4_spec.ip4dst;
+	input->filter.formatted.src_port = fsp->h_u.tcp_ip4_spec.psrc;
+	mask.formatted.src_port = fsp->m_u.tcp_ip4_spec.psrc;
+	input->filter.formatted.dst_port = fsp->h_u.tcp_ip4_spec.pdst;
+	mask.formatted.dst_port = fsp->m_u.tcp_ip4_spec.pdst;
+
+	if (fsp->flow_type & FLOW_EXT) {
+		input->filter.formatted.vm_pool =
+				(unsigned char)ntohl(fsp->h_ext.data[1]);
+		mask.formatted.vm_pool =
+				(unsigned char)ntohl(fsp->m_ext.data[1]);
+		input->filter.formatted.flex_bytes =
+						fsp->h_ext.vlan_etype;
+		mask.formatted.flex_bytes = fsp->m_ext.vlan_etype;
+	}
+
+	switch (input->filter.formatted.flow_type) {
+	case TXGBE_ATR_FLOW_TYPE_TCPV4:
+		ptype = WX_PTYPE_L2_IPV4_TCP;
+		break;
+	case TXGBE_ATR_FLOW_TYPE_UDPV4:
+		ptype = WX_PTYPE_L2_IPV4_UDP;
+		break;
+	case TXGBE_ATR_FLOW_TYPE_SCTPV4:
+		ptype = WX_PTYPE_L2_IPV4_SCTP;
+		break;
+	case TXGBE_ATR_FLOW_TYPE_IPV4:
+		ptype = WX_PTYPE_L2_IPV4;
+		break;
+	default:
+		break;
+	}
+
+	input->filter.formatted.vlan_id = htons(ptype);
+	if (mask.formatted.flow_type & TXGBE_ATR_L4TYPE_MASK)
+		mask.formatted.vlan_id = htons(0xFFFF);
+	else
+		mask.formatted.vlan_id = htons(0xFFF8);
+
+	/* determine if we need to drop or route the packet */
+	if (fsp->ring_cookie == RX_CLS_FLOW_DISC)
+		input->action = TXGBE_RDB_FDIR_DROP_QUEUE;
+	else
+		input->action = fsp->ring_cookie;
+
+	spin_lock(&txgbe->fdir_perfect_lock);
+
+	if (hlist_empty(&txgbe->fdir_filter_list)) {
+		/* save mask and program input mask into HW */
+		memcpy(&txgbe->fdir_mask, &mask, sizeof(mask));
+		err = txgbe_fdir_set_input_mask(wx, &mask);
+		if (err)
+			goto err_unlock;
+	} else if (memcmp(&txgbe->fdir_mask, &mask, sizeof(mask))) {
+		wx_err(wx, "Hardware only supports one mask per port. To change the mask you must first delete all the rules.\n");
+		goto err_unlock;
+	}
+
+	/* apply mask and compute/store hash */
+	txgbe_atr_compute_perfect_hash(&input->filter, &mask);
+
+	/* check if new entry does not exist on filter list */
+	if (txgbe_match_ethtool_fdir_entry(txgbe, input))
+		goto err_unlock;
+
+	/* only program filters to hardware if the net device is running, as
+	 * we store the filters in the Rx buffer which is not allocated when
+	 * the device is down
+	 */
+	if (netif_running(wx->netdev)) {
+		err = txgbe_fdir_write_perfect_filter(wx, &input->filter,
+						      input->sw_idx, queue);
+		if (err)
+			goto err_unlock;
+	}
+
+	txgbe_update_ethtool_fdir_entry(txgbe, input, input->sw_idx);
+
+	spin_unlock(&txgbe->fdir_perfect_lock);
+
+	return 0;
+err_unlock:
+	spin_unlock(&txgbe->fdir_perfect_lock);
+err_out:
+	kfree(input);
+	return err;
+}
+
+static int txgbe_del_ethtool_fdir_entry(struct txgbe *txgbe,
+					struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fsp =
+		(struct ethtool_rx_flow_spec *)&cmd->fs;
+	int err = 0;
+
+	spin_lock(&txgbe->fdir_perfect_lock);
+	err = txgbe_update_ethtool_fdir_entry(txgbe, NULL, fsp->location);
+	spin_unlock(&txgbe->fdir_perfect_lock);
+
+	return err;
+}
+
+static int txgbe_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
+{
+	struct wx *wx = netdev_priv(dev);
+	struct txgbe *txgbe = wx->priv;
+	int ret = -EOPNOTSUPP;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		ret = txgbe_add_ethtool_fdir_entry(txgbe, cmd);
+		break;
+	case ETHTOOL_SRXCLSRLDEL:
+		ret = txgbe_del_ethtool_fdir_entry(txgbe, cmd);
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
 static const struct ethtool_ops txgbe_ethtool_ops = {
 	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
 				     ETHTOOL_COALESCE_TX_MAX_FRAMES_IRQ,
@@ -100,6 +525,8 @@ static const struct ethtool_ops txgbe_ethtool_ops = {
 	.set_coalesce		= wx_set_coalesce,
 	.get_channels		= wx_get_channels,
 	.set_channels		= txgbe_set_channels,
+	.get_rxnfc		= txgbe_get_rxnfc,
+	.set_rxnfc		= txgbe_set_rxnfc,
 	.get_msglevel		= wx_get_msglevel,
 	.set_msglevel		= wx_set_msglevel,
 };
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c
index 6158209e84cb..ef50efbaec0f 100644
--- a/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.c
@@ -90,6 +90,71 @@ static void txgbe_atr_compute_sig_hash(union txgbe_atr_hash_dword input,
 	*hash = sig_hash ^ bucket_hash;
 }
 
+#define TXGBE_COMPUTE_BKT_HASH_ITERATION(_n) \
+do { \
+	u32 n = (_n); \
+	if (TXGBE_ATR_BUCKET_HASH_KEY & (0x01 << n)) \
+		bucket_hash ^= lo_hash_dword >> n; \
+	if (TXGBE_ATR_BUCKET_HASH_KEY & (0x01 << (n + 16))) \
+		bucket_hash ^= hi_hash_dword >> n; \
+} while (0)
+
+/**
+ *  txgbe_atr_compute_perfect_hash - Compute the perfect filter hash
+ *  @input: input bitstream to compute the hash on
+ *  @input_mask: mask for the input bitstream
+ *
+ *  This function serves two main purposes.  First it applies the input_mask
+ *  to the atr_input resulting in a cleaned up atr_input data stream.
+ *  Secondly it computes the hash and stores it in the bkt_hash field at
+ *  the end of the input byte stream.  This way it will be available for
+ *  future use without needing to recompute the hash.
+ **/
+void txgbe_atr_compute_perfect_hash(union txgbe_atr_input *input,
+				    union txgbe_atr_input *input_mask)
+{
+	u32 hi_hash_dword, lo_hash_dword, flow_vm_vlan;
+	u32 bucket_hash = 0;
+	__be32 hi_dword = 0;
+	u32 i = 0;
+
+	/* Apply masks to input data */
+	for (i = 0; i < 11; i++)
+		input->dword_stream[i] &= input_mask->dword_stream[i];
+
+	/* record the flow_vm_vlan bits as they are a key part to the hash */
+	flow_vm_vlan = ntohl(input->dword_stream[0]);
+
+	/* generate common hash dword */
+	for (i = 1; i <= 10; i++)
+		hi_dword ^= input->dword_stream[i];
+	hi_hash_dword = ntohl(hi_dword);
+
+	/* low dword is word swapped version of common */
+	lo_hash_dword = (hi_hash_dword >> 16) | (hi_hash_dword << 16);
+
+	/* apply flow ID/VM pool/VLAN ID bits to hash words */
+	hi_hash_dword ^= flow_vm_vlan ^ (flow_vm_vlan >> 16);
+
+	/* Process bits 0 and 16 */
+	TXGBE_COMPUTE_BKT_HASH_ITERATION(0);
+
+	/* apply flow ID/VM pool/VLAN ID bits to lo hash dword, we had to
+	 * delay this because bit 0 of the stream should not be processed
+	 * so we do not add the VLAN until after bit 0 was processed
+	 */
+	lo_hash_dword ^= flow_vm_vlan ^ (flow_vm_vlan << 16);
+
+	/* Process remaining 30 bit of the key */
+	for (i = 1; i <= 15; i++)
+		TXGBE_COMPUTE_BKT_HASH_ITERATION(i);
+
+	/* Limit hash to 13 bits since max bucket count is 8K.
+	 * Store result at the end of the input stream.
+	 */
+	input->formatted.bkt_hash = (__force __be16)(bucket_hash & 0x1FFF);
+}
+
 static int txgbe_fdir_check_cmd_complete(struct wx *wx)
 {
 	u32 val;
@@ -239,6 +304,188 @@ void txgbe_atr(struct wx_ring *ring, struct wx_tx_buffer *first, u8 ptype)
 					ring->queue_index);
 }
 
+int txgbe_fdir_set_input_mask(struct wx *wx, union txgbe_atr_input *input_mask)
+{
+	u32 fdirm = 0, fdirtcpm = 0, flex = 0;
+
+	/* Program the relevant mask registers. If src/dst_port or src/dst_addr
+	 * are zero, then assume a full mask for that field.  Also assume that
+	 * a VLAN of 0 is unspecified, so mask that out as well.  L4type
+	 * cannot be masked out in this implementation.
+	 *
+	 * This also assumes IPv4 only.  IPv6 masking isn't supported at this
+	 * point in time.
+	 */
+
+	/* verify bucket hash is cleared on hash generation */
+	if (input_mask->formatted.bkt_hash)
+		wx_dbg(wx, "bucket hash should always be 0 in mask\n");
+
+	/* Program FDIRM and verify partial masks */
+	switch (input_mask->formatted.vm_pool & 0x7F) {
+	case 0x0:
+		fdirm |= TXGBE_RDB_FDIR_OTHER_MSK_POOL;
+		break;
+	case 0x7F:
+		break;
+	default:
+		wx_err(wx, "Error on vm pool mask\n");
+		return -EINVAL;
+	}
+
+	switch (input_mask->formatted.flow_type & TXGBE_ATR_L4TYPE_MASK) {
+	case 0x0:
+		fdirm |= TXGBE_RDB_FDIR_OTHER_MSK_L4P;
+		if (input_mask->formatted.dst_port ||
+		    input_mask->formatted.src_port) {
+			wx_err(wx, "Error on src/dst port mask\n");
+			return -EINVAL;
+		}
+		break;
+	case TXGBE_ATR_L4TYPE_MASK:
+		break;
+	default:
+		wx_err(wx, "Error on flow type mask\n");
+		return -EINVAL;
+	}
+
+	/* Now mask VM pool and destination IPv6 - bits 5 and 2 */
+	wr32(wx, TXGBE_RDB_FDIR_OTHER_MSK, fdirm);
+
+	flex = rd32(wx, TXGBE_RDB_FDIR_FLEX_CFG(0));
+	flex &= ~TXGBE_RDB_FDIR_FLEX_CFG_FIELD0;
+	flex |= (TXGBE_RDB_FDIR_FLEX_CFG_BASE_MAC |
+		 TXGBE_RDB_FDIR_FLEX_CFG_OFST(0x6));
+
+	switch ((__force u16)input_mask->formatted.flex_bytes & 0xFFFF) {
+	case 0x0000:
+		/* Mask Flex Bytes */
+		flex |= TXGBE_RDB_FDIR_FLEX_CFG_MSK;
+		break;
+	case 0xFFFF:
+		break;
+	default:
+		wx_err(wx, "Error on flexible byte mask\n");
+		return -EINVAL;
+	}
+	wr32(wx, TXGBE_RDB_FDIR_FLEX_CFG(0), flex);
+
+	/* store the TCP/UDP port masks, bit reversed from port layout */
+	fdirtcpm = ntohs(input_mask->formatted.dst_port);
+	fdirtcpm <<= TXGBE_RDB_FDIR_PORT_DESTINATION_SHIFT;
+	fdirtcpm |= ntohs(input_mask->formatted.src_port);
+
+	/* write both the same so that UDP and TCP use the same mask */
+	wr32(wx, TXGBE_RDB_FDIR_TCP_MSK, ~fdirtcpm);
+	wr32(wx, TXGBE_RDB_FDIR_UDP_MSK, ~fdirtcpm);
+	wr32(wx, TXGBE_RDB_FDIR_SCTP_MSK, ~fdirtcpm);
+
+	/* store source and destination IP masks (little-enian) */
+	wr32(wx, TXGBE_RDB_FDIR_SA4_MSK,
+	     ntohl(~input_mask->formatted.src_ip[0]));
+	wr32(wx, TXGBE_RDB_FDIR_DA4_MSK,
+	     ntohl(~input_mask->formatted.dst_ip[0]));
+
+	return 0;
+}
+
+int txgbe_fdir_write_perfect_filter(struct wx *wx,
+				    union txgbe_atr_input *input,
+				    u16 soft_id, u8 queue)
+{
+	u32 fdirport, fdirvlan, fdirhash, fdircmd;
+	int err = 0;
+
+	/* currently IPv6 is not supported, must be programmed with 0 */
+	wr32(wx, TXGBE_RDB_FDIR_IP6(2), ntohl(input->formatted.src_ip[0]));
+	wr32(wx, TXGBE_RDB_FDIR_IP6(1), ntohl(input->formatted.src_ip[1]));
+	wr32(wx, TXGBE_RDB_FDIR_IP6(0), ntohl(input->formatted.src_ip[2]));
+
+	/* record the source address (little-endian) */
+	wr32(wx, TXGBE_RDB_FDIR_SA, ntohl(input->formatted.src_ip[0]));
+
+	/* record the first 32 bits of the destination address
+	 * (little-endian)
+	 */
+	wr32(wx, TXGBE_RDB_FDIR_DA, ntohl(input->formatted.dst_ip[0]));
+
+	/* record source and destination port (little-endian)*/
+	fdirport = ntohs(input->formatted.dst_port);
+	fdirport <<= TXGBE_RDB_FDIR_PORT_DESTINATION_SHIFT;
+	fdirport |= ntohs(input->formatted.src_port);
+	wr32(wx, TXGBE_RDB_FDIR_PORT, fdirport);
+
+	/* record packet type and flex_bytes (little-endian) */
+	fdirvlan = ntohs(input->formatted.flex_bytes);
+	fdirvlan <<= TXGBE_RDB_FDIR_FLEX_FLEX_SHIFT;
+	fdirvlan |= ntohs(input->formatted.vlan_id);
+	wr32(wx, TXGBE_RDB_FDIR_FLEX, fdirvlan);
+
+	/* configure FDIRHASH register */
+	fdirhash = (__force u32)input->formatted.bkt_hash |
+		   TXGBE_RDB_FDIR_HASH_BUCKET_VALID |
+		   TXGBE_RDB_FDIR_HASH_SIG_SW_INDEX(soft_id);
+	wr32(wx, TXGBE_RDB_FDIR_HASH, fdirhash);
+
+	/* flush all previous writes to make certain registers are
+	 * programmed prior to issuing the command
+	 */
+	WX_WRITE_FLUSH(wx);
+
+	/* configure FDIRCMD register */
+	fdircmd = TXGBE_RDB_FDIR_CMD_CMD_ADD_FLOW |
+		  TXGBE_RDB_FDIR_CMD_FILTER_UPDATE |
+		  TXGBE_RDB_FDIR_CMD_LAST | TXGBE_RDB_FDIR_CMD_QUEUE_EN;
+	if (queue == TXGBE_RDB_FDIR_DROP_QUEUE)
+		fdircmd |= TXGBE_RDB_FDIR_CMD_DROP;
+	fdircmd |= TXGBE_RDB_FDIR_CMD_FLOW_TYPE(input->formatted.flow_type);
+	fdircmd |= TXGBE_RDB_FDIR_CMD_RX_QUEUE(queue);
+	fdircmd |= TXGBE_RDB_FDIR_CMD_VT_POOL(input->formatted.vm_pool);
+
+	wr32(wx, TXGBE_RDB_FDIR_CMD, fdircmd);
+	err = txgbe_fdir_check_cmd_complete(wx);
+	if (err)
+		wx_err(wx, "Flow Director command did not complete!\n");
+
+	return err;
+}
+
+int txgbe_fdir_erase_perfect_filter(struct wx *wx,
+				    union txgbe_atr_input *input,
+				    u16 soft_id)
+{
+	u32 fdirhash, fdircmd;
+	int err = 0;
+
+	/* configure FDIRHASH register */
+	fdirhash = (__force u32)input->formatted.bkt_hash;
+	fdirhash |= TXGBE_RDB_FDIR_HASH_SIG_SW_INDEX(soft_id);
+	wr32(wx, TXGBE_RDB_FDIR_HASH, fdirhash);
+
+	/* flush hash to HW */
+	WX_WRITE_FLUSH(wx);
+
+	/* Query if filter is present */
+	wr32(wx, TXGBE_RDB_FDIR_CMD, TXGBE_RDB_FDIR_CMD_CMD_QUERY_REM_FILT);
+
+	err = txgbe_fdir_check_cmd_complete(wx);
+	if (err) {
+		wx_err(wx, "Flow Director command did not complete!\n");
+		return err;
+	}
+
+	fdircmd = rd32(wx, TXGBE_RDB_FDIR_CMD);
+	/* if filter exists in hardware then remove it */
+	if (fdircmd & TXGBE_RDB_FDIR_CMD_FILTER_VALID) {
+		wr32(wx, TXGBE_RDB_FDIR_HASH, fdirhash);
+		WX_WRITE_FLUSH(wx);
+		wr32(wx, TXGBE_RDB_FDIR_CMD,
+		     TXGBE_RDB_FDIR_CMD_CMD_REMOVE_FLOW);
+	}
+
+	return 0;
+}
+
 /**
  *  txgbe_fdir_enable - Initialize Flow Director control registers
  *  @wx: pointer to hardware structure
@@ -291,12 +538,106 @@ static void txgbe_init_fdir_signature(struct wx *wx)
 	txgbe_fdir_enable(wx, fdirctrl);
 }
 
+/**
+ *  txgbe_init_fdir_perfect - Initialize Flow Director perfect filters
+ *  @wx: pointer to hardware structure
+ **/
+static void txgbe_init_fdir_perfect(struct wx *wx)
+{
+	u32 fdirctrl = TXGBE_FDIR_PBALLOC_64K;
+
+	/* Continue setup of fdirctrl register bits:
+	 *  Turn perfect match filtering on
+	 *  Report hash in RSS field of Rx wb descriptor
+	 *  Initialize the drop queue
+	 *  Move the flexible bytes to use the ethertype - shift 6 words
+	 *  Set the maximum length per hash bucket to 0xA filters
+	 *  Send interrupt when 64 (0x4 * 16) filters are left
+	 */
+	fdirctrl |= TXGBE_RDB_FDIR_CTL_PERFECT_MATCH |
+		    TXGBE_RDB_FDIR_CTL_DROP_Q(TXGBE_RDB_FDIR_DROP_QUEUE) |
+		    TXGBE_RDB_FDIR_CTL_HASH_BITS(0xF) |
+		    TXGBE_RDB_FDIR_CTL_MAX_LENGTH(0xA) |
+		    TXGBE_RDB_FDIR_CTL_FULL_THRESH(4);
+
+	/* write hashes and fdirctrl register, poll for completion */
+	txgbe_fdir_enable(wx, fdirctrl);
+}
+
+static void txgbe_fdir_filter_restore(struct wx *wx)
+{
+	struct txgbe_fdir_filter *filter;
+	struct txgbe *txgbe = wx->priv;
+	struct hlist_node *node;
+	u8 queue = 0;
+	int ret = 0;
+
+	spin_lock(&txgbe->fdir_perfect_lock);
+
+	if (!hlist_empty(&txgbe->fdir_filter_list))
+		ret = txgbe_fdir_set_input_mask(wx, &txgbe->fdir_mask);
+
+	if (ret)
+		goto unlock;
+
+	hlist_for_each_entry_safe(filter, node,
+				  &txgbe->fdir_filter_list, fdir_node) {
+		if (filter->action == TXGBE_RDB_FDIR_DROP_QUEUE) {
+			queue = TXGBE_RDB_FDIR_DROP_QUEUE;
+		} else {
+			u32 ring = ethtool_get_flow_spec_ring(filter->action);
+
+			if (ring >= wx->num_rx_queues) {
+				wx_err(wx, "FDIR restore failed, ring:%u\n",
+				       ring);
+				continue;
+			}
+
+			/* Map the ring onto the absolute queue index */
+			queue = wx->rx_ring[ring]->reg_idx;
+		}
+
+		ret = txgbe_fdir_write_perfect_filter(wx,
+						      &filter->filter,
+						      filter->sw_idx,
+						      queue);
+		if (ret)
+			wx_err(wx, "FDIR restore failed, index:%u\n",
+			       filter->sw_idx);
+	}
+
+unlock:
+	spin_unlock(&txgbe->fdir_perfect_lock);
+}
+
 void txgbe_configure_fdir(struct wx *wx)
 {
 	wx_disable_sec_rx_path(wx);
 
-	if (test_bit(WX_FLAG_FDIR_HASH, wx->flags))
+	if (test_bit(WX_FLAG_FDIR_HASH, wx->flags)) {
 		txgbe_init_fdir_signature(wx);
+	} else if (test_bit(WX_FLAG_FDIR_PERFECT, wx->flags)) {
+		txgbe_init_fdir_perfect(wx);
+		txgbe_fdir_filter_restore(wx);
+	}
 
 	wx_enable_sec_rx_path(wx);
 }
+
+void txgbe_fdir_filter_exit(struct wx *wx)
+{
+	struct txgbe_fdir_filter *filter;
+	struct txgbe *txgbe = wx->priv;
+	struct hlist_node *node;
+
+	spin_lock(&txgbe->fdir_perfect_lock);
+
+	hlist_for_each_entry_safe(filter, node,
+				  &txgbe->fdir_filter_list, fdir_node) {
+		hlist_del(&filter->fdir_node);
+		kfree(filter);
+	}
+	txgbe->fdir_filter_count = 0;
+
+	spin_unlock(&txgbe->fdir_perfect_lock);
+}
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h
index ed245b66dc2a..1f44ce60becb 100644
--- a/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_fdir.h
@@ -4,7 +4,17 @@
 #ifndef _TXGBE_FDIR_H_
 #define _TXGBE_FDIR_H_
 
+void txgbe_atr_compute_perfect_hash(union txgbe_atr_input *input,
+				    union txgbe_atr_input *input_mask);
 void txgbe_atr(struct wx_ring *ring, struct wx_tx_buffer *first, u8 ptype);
+int txgbe_fdir_set_input_mask(struct wx *wx, union txgbe_atr_input *input_mask);
+int txgbe_fdir_write_perfect_filter(struct wx *wx,
+				    union txgbe_atr_input *input,
+				    u16 soft_id, u8 queue);
+int txgbe_fdir_erase_perfect_filter(struct wx *wx,
+				    union txgbe_atr_input *input,
+				    u16 soft_id);
 void txgbe_configure_fdir(struct wx *wx);
+void txgbe_fdir_filter_exit(struct wx *wx);
 
 #endif /* _TXGBE_FDIR_H_ */
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
index ce49fb725541..41e9ebf11e41 100644
--- a/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_main.c
@@ -283,6 +283,12 @@ static int txgbe_sw_init(struct wx *wx)
 	return 0;
 }
 
+static void txgbe_init_fdir(struct txgbe *txgbe)
+{
+	txgbe->fdir_filter_count = 0;
+	spin_lock_init(&txgbe->fdir_perfect_lock);
+}
+
 /**
  * txgbe_open - Called when a network interface is made active
  * @netdev: network interface device structure
@@ -361,6 +367,7 @@ static int txgbe_close(struct net_device *netdev)
 	txgbe_down(wx);
 	wx_free_irq(wx);
 	wx_free_resources(wx);
+	txgbe_fdir_filter_exit(wx);
 	wx_control_hw(wx, false);
 
 	return 0;
@@ -669,6 +676,8 @@ static int txgbe_probe(struct pci_dev *pdev,
 	txgbe->wx = wx;
 	wx->priv = txgbe;
 
+	txgbe_init_fdir(txgbe);
+
 	err = txgbe_setup_misc_irq(txgbe);
 	if (err)
 		goto err_release_hw;
diff --git a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h
index 5b8c55df35fe..959102c4c379 100644
--- a/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h
+++ b/drivers/net/ethernet/wangxun/txgbe/txgbe_type.h
@@ -90,6 +90,7 @@
 #define TXGBE_XPCS_IDA_DATA                     0x13004
 
 /********************************* Flow Director *****************************/
+#define TXGBE_RDB_FDIR_DROP_QUEUE               127
 #define TXGBE_RDB_FDIR_CTL                      0x19500
 #define TXGBE_RDB_FDIR_CTL_INIT_DONE            BIT(3)
 #define TXGBE_RDB_FDIR_CTL_PERFECT_MATCH        BIT(4)
@@ -97,6 +98,13 @@
 #define TXGBE_RDB_FDIR_CTL_HASH_BITS(v)         FIELD_PREP(GENMASK(23, 20), v)
 #define TXGBE_RDB_FDIR_CTL_MAX_LENGTH(v)        FIELD_PREP(GENMASK(27, 24), v)
 #define TXGBE_RDB_FDIR_CTL_FULL_THRESH(v)       FIELD_PREP(GENMASK(31, 28), v)
+#define TXGBE_RDB_FDIR_IP6(_i)                  (0x1950C + ((_i) * 4)) /* 0-2 */
+#define TXGBE_RDB_FDIR_SA                       0x19518
+#define TXGBE_RDB_FDIR_DA                       0x1951C
+#define TXGBE_RDB_FDIR_PORT                     0x19520
+#define TXGBE_RDB_FDIR_PORT_DESTINATION_SHIFT   16
+#define TXGBE_RDB_FDIR_FLEX                     0x19524
+#define TXGBE_RDB_FDIR_FLEX_FLEX_SHIFT          16
 #define TXGBE_RDB_FDIR_HASH                     0x19528
 #define TXGBE_RDB_FDIR_HASH_SIG_SW_INDEX(v)     FIELD_PREP(GENMASK(31, 16), v)
 #define TXGBE_RDB_FDIR_HASH_BUCKET_VALID        BIT(15)
@@ -114,8 +122,16 @@
 #define TXGBE_RDB_FDIR_CMD_QUEUE_EN             BIT(15)
 #define TXGBE_RDB_FDIR_CMD_RX_QUEUE(v)          FIELD_PREP(GENMASK(22, 16), v)
 #define TXGBE_RDB_FDIR_CMD_VT_POOL(v)           FIELD_PREP(GENMASK(29, 24), v)
+#define TXGBE_RDB_FDIR_DA4_MSK                  0x1953C
+#define TXGBE_RDB_FDIR_SA4_MSK                  0x19540
+#define TXGBE_RDB_FDIR_TCP_MSK                  0x19544
+#define TXGBE_RDB_FDIR_UDP_MSK                  0x19548
+#define TXGBE_RDB_FDIR_SCTP_MSK                 0x19560
 #define TXGBE_RDB_FDIR_HKEY                     0x19568
 #define TXGBE_RDB_FDIR_SKEY                     0x1956C
+#define TXGBE_RDB_FDIR_OTHER_MSK                0x19570
+#define TXGBE_RDB_FDIR_OTHER_MSK_POOL           BIT(2)
+#define TXGBE_RDB_FDIR_OTHER_MSK_L4P            BIT(3)
 #define TXGBE_RDB_FDIR_FLEX_CFG(_i)             (0x19580 + ((_i) * 4))
 #define TXGBE_RDB_FDIR_FLEX_CFG_FIELD0          GENMASK(7, 0)
 #define TXGBE_RDB_FDIR_FLEX_CFG_BASE_MAC        FIELD_PREP(GENMASK(1, 0), 0)
@@ -230,6 +246,13 @@ enum txgbe_fdir_pballoc_type {
 	TXGBE_FDIR_PBALLOC_256K = 3,
 };
 
+struct txgbe_fdir_filter {
+	struct hlist_node fdir_node;
+	union txgbe_atr_input filter;
+	u16 sw_idx;
+	u16 action;
+};
+
 /* TX/RX descriptor defines */
 #define TXGBE_DEFAULT_TXD               512
 #define TXGBE_DEFAULT_TX_WORK           256
@@ -316,7 +339,10 @@ struct txgbe {
 	unsigned int link_irq;
 
 	/* flow director */
+	struct hlist_head fdir_filter_list;
 	union txgbe_atr_input fdir_mask;
+	int fdir_filter_count;
+	spinlock_t fdir_perfect_lock; /* spinlock for FDIR */
 };
 
 #endif /* _TXGBE_TYPE_H_ */
-- 
cgit v1.2.3-58-ga151


From 34744a7749b3f600dd54a2d39b8db99fecf54901 Mon Sep 17 00:00:00 2001
From: Jiawen Wu <jiawenwu@trustnetic.com>
Date: Tue, 18 Jun 2024 18:16:09 +0800
Subject: net: txgbe: add FDIR info to ethtool ops

Add flow director filter match and miss statistics to ethtool -S.
And change the number of queues when using flow director for ehtool -l.

Signed-off-by: Jiawen Wu <jiawenwu@trustnetic.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/wangxun/libwx/wx_ethtool.c | 39 ++++++++++++++++++++++---
 drivers/net/ethernet/wangxun/libwx/wx_hw.c      |  5 ++++
 drivers/net/ethernet/wangxun/libwx/wx_type.h    |  4 +++
 3 files changed, 44 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c b/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c
index cc3bec42ed8e..abe5921dde02 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c
+++ b/drivers/net/ethernet/wangxun/libwx/wx_ethtool.c
@@ -43,6 +43,11 @@ static const struct wx_stats wx_gstrings_stats[] = {
 	WX_STAT("alloc_rx_buff_failed", alloc_rx_buff_failed),
 };
 
+static const struct wx_stats wx_gstrings_fdir_stats[] = {
+	WX_STAT("fdir_match", stats.fdirmatch),
+	WX_STAT("fdir_miss", stats.fdirmiss),
+};
+
 /* drivers allocates num_tx_queues and num_rx_queues symmetrically so
  * we set the num_rx_queues to evaluate to num_tx_queues. This is
  * used because we do not have a good way to get the max number of
@@ -55,13 +60,17 @@ static const struct wx_stats wx_gstrings_stats[] = {
 		(WX_NUM_TX_QUEUES + WX_NUM_RX_QUEUES) * \
 		(sizeof(struct wx_queue_stats) / sizeof(u64)))
 #define WX_GLOBAL_STATS_LEN  ARRAY_SIZE(wx_gstrings_stats)
+#define WX_FDIR_STATS_LEN  ARRAY_SIZE(wx_gstrings_fdir_stats)
 #define WX_STATS_LEN (WX_GLOBAL_STATS_LEN + WX_QUEUE_STATS_LEN)
 
 int wx_get_sset_count(struct net_device *netdev, int sset)
 {
+	struct wx *wx = netdev_priv(netdev);
+
 	switch (sset) {
 	case ETH_SS_STATS:
-		return WX_STATS_LEN;
+		return (wx->mac.type == wx_mac_sp) ?
+			WX_STATS_LEN + WX_FDIR_STATS_LEN : WX_STATS_LEN;
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -70,6 +79,7 @@ EXPORT_SYMBOL(wx_get_sset_count);
 
 void wx_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
 {
+	struct wx *wx = netdev_priv(netdev);
 	u8 *p = data;
 	int i;
 
@@ -77,6 +87,10 @@ void wx_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
 	case ETH_SS_STATS:
 		for (i = 0; i < WX_GLOBAL_STATS_LEN; i++)
 			ethtool_puts(&p, wx_gstrings_stats[i].stat_string);
+		if (wx->mac.type == wx_mac_sp) {
+			for (i = 0; i < WX_FDIR_STATS_LEN; i++)
+				ethtool_puts(&p, wx_gstrings_fdir_stats[i].stat_string);
+		}
 		for (i = 0; i < netdev->num_tx_queues; i++) {
 			ethtool_sprintf(&p, "tx_queue_%u_packets", i);
 			ethtool_sprintf(&p, "tx_queue_%u_bytes", i);
@@ -96,7 +110,7 @@ void wx_get_ethtool_stats(struct net_device *netdev,
 	struct wx *wx = netdev_priv(netdev);
 	struct wx_ring *ring;
 	unsigned int start;
-	int i, j;
+	int i, j, k;
 	char *p;
 
 	wx_update_stats(wx);
@@ -107,6 +121,13 @@ void wx_get_ethtool_stats(struct net_device *netdev,
 			   sizeof(u64)) ? *(u64 *)p : *(u32 *)p;
 	}
 
+	if (wx->mac.type == wx_mac_sp) {
+		for (k = 0; k < WX_FDIR_STATS_LEN; k++) {
+			p = (char *)wx + wx_gstrings_fdir_stats[k].stat_offset;
+			data[i++] = *(u64 *)p;
+		}
+	}
+
 	for (j = 0; j < netdev->num_tx_queues; j++) {
 		ring = wx->tx_ring[j];
 		if (!ring) {
@@ -172,17 +193,21 @@ EXPORT_SYMBOL(wx_get_pause_stats);
 
 void wx_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *info)
 {
+	unsigned int stats_len = WX_STATS_LEN;
 	struct wx *wx = netdev_priv(netdev);
 
+	if (wx->mac.type == wx_mac_sp)
+		stats_len += WX_FDIR_STATS_LEN;
+
 	strscpy(info->driver, wx->driver_name, sizeof(info->driver));
 	strscpy(info->fw_version, wx->eeprom_id, sizeof(info->fw_version));
 	strscpy(info->bus_info, pci_name(wx->pdev), sizeof(info->bus_info));
 	if (wx->num_tx_queues <= WX_NUM_TX_QUEUES) {
-		info->n_stats = WX_STATS_LEN -
+		info->n_stats = stats_len -
 				   (WX_NUM_TX_QUEUES - wx->num_tx_queues) *
 				   (sizeof(struct wx_queue_stats) / sizeof(u64)) * 2;
 	} else {
-		info->n_stats = WX_STATS_LEN;
+		info->n_stats = stats_len;
 	}
 }
 EXPORT_SYMBOL(wx_get_drvinfo);
@@ -383,6 +408,9 @@ void wx_get_channels(struct net_device *dev,
 
 	/* record RSS queues */
 	ch->combined_count = wx->ring_feature[RING_F_RSS].indices;
+
+	if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags))
+		ch->combined_count = wx->ring_feature[RING_F_FDIR].indices;
 }
 EXPORT_SYMBOL(wx_get_channels);
 
@@ -400,6 +428,9 @@ int wx_set_channels(struct net_device *dev,
 	if (count > wx_max_channels(wx))
 		return -EINVAL;
 
+	if (test_bit(WX_FLAG_FDIR_CAPABLE, wx->flags))
+		wx->ring_feature[RING_F_FDIR].limit = count;
+
 	wx->ring_feature[RING_F_RSS].limit = count;
 
 	return 0;
diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.c b/drivers/net/ethernet/wangxun/libwx/wx_hw.c
index 8fb38f83a615..44cd7a5866c1 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_hw.c
+++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.c
@@ -2352,6 +2352,11 @@ void wx_update_stats(struct wx *wx)
 	hwstats->b2ogprc += rd32(wx, WX_RDM_BMC2OS_CNT);
 	hwstats->rdmdrop += rd32(wx, WX_RDM_DRP_PKT);
 
+	if (wx->mac.type == wx_mac_sp) {
+		hwstats->fdirmatch += rd32(wx, WX_RDB_FDIR_MATCH);
+		hwstats->fdirmiss += rd32(wx, WX_RDB_FDIR_MISS);
+	}
+
 	for (i = 0; i < wx->mac.max_rx_queues; i++)
 		hwstats->qmprc += rd32(wx, WX_PX_MPRC(i));
 }
diff --git a/drivers/net/ethernet/wangxun/libwx/wx_type.h b/drivers/net/ethernet/wangxun/libwx/wx_type.h
index b1f9bab06e90..e0b7866f96ec 100644
--- a/drivers/net/ethernet/wangxun/libwx/wx_type.h
+++ b/drivers/net/ethernet/wangxun/libwx/wx_type.h
@@ -157,6 +157,8 @@
 #define WX_RDB_RA_CTL_RSS_IPV6_TCP   BIT(21)
 #define WX_RDB_RA_CTL_RSS_IPV4_UDP   BIT(22)
 #define WX_RDB_RA_CTL_RSS_IPV6_UDP   BIT(23)
+#define WX_RDB_FDIR_MATCH            0x19558
+#define WX_RDB_FDIR_MISS             0x1955C
 
 /******************************* PSR Registers *******************************/
 /* psr control */
@@ -1018,6 +1020,8 @@ struct wx_hw_stats {
 	u64 crcerrs;
 	u64 rlec;
 	u64 qmprc;
+	u64 fdirmatch;
+	u64 fdirmiss;
 };
 
 enum wx_state {
-- 
cgit v1.2.3-58-ga151


From a012f9a752a5856b29610057c76079fbdbe02abf Mon Sep 17 00:00:00 2001
From: Jackie Jone <jackie.jone@alliedtelesis.co.nz>
Date: Tue, 18 Jun 2024 14:33:28 -0700
Subject: igb: Add MII write support

To facilitate running PHY parametric tests, add support for the SIOCSMIIREG
ioctl. This allows a userspace application to write to the PHY registers
to enable the test modes.

Signed-off-by: Jackie Jone <jackie.jone@alliedtelesis.co.nz>
Acked-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Link: https://lore.kernel.org/r/20240618213330.982046-1-anthony.l.nguyen@intel.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/intel/igb/igb_main.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index fce2930ae6af..3af03a211c3c 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -9139,6 +9139,10 @@ static int igb_mii_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
 			return -EIO;
 		break;
 	case SIOCSMIIREG:
+		if (igb_write_phy_reg(&adapter->hw, data->reg_num & 0x1F,
+				      data->val_in))
+			return -EIO;
+		break;
 	default:
 		return -EOPNOTSUPP;
 	}
-- 
cgit v1.2.3-58-ga151


From 0d6b1eb660be4c7817fc4686a85caf430d6111d0 Mon Sep 17 00:00:00 2001
From: Uwe Kleine-König <u.kleine-koenig@baylibre.com>
Date: Wed, 19 Jun 2024 21:56:31 +0200
Subject: nfc: Drop explicit initialization of struct
 i2c_device_id::driver_data to 0
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

These drivers don't use the driver_data member of struct i2c_device_id,
so don't explicitly initialize this member.

This prepares putting driver_data in an anonymous union which requires
either no initialization or named designators. But it's also a nice
cleanup on its own.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@baylibre.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patch.msgid.link/20240619195631.2545407-2-u.kleine-koenig@baylibre.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/nfc/microread/i2c.c | 2 +-
 drivers/nfc/nfcmrvl/i2c.c   | 2 +-
 drivers/nfc/nxp-nci/i2c.c   | 2 +-
 drivers/nfc/pn533/i2c.c     | 2 +-
 drivers/nfc/pn544/i2c.c     | 2 +-
 drivers/nfc/s3fwrn5/i2c.c   | 2 +-
 drivers/nfc/st-nci/i2c.c    | 2 +-
 drivers/nfc/st21nfca/i2c.c  | 2 +-
 8 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/nfc/microread/i2c.c b/drivers/nfc/microread/i2c.c
index 642df4e0ce24..113b2e306e35 100644
--- a/drivers/nfc/microread/i2c.c
+++ b/drivers/nfc/microread/i2c.c
@@ -277,7 +277,7 @@ static void microread_i2c_remove(struct i2c_client *client)
 }
 
 static const struct i2c_device_id microread_i2c_id[] = {
-	{ MICROREAD_I2C_DRIVER_NAME, 0},
+	{ MICROREAD_I2C_DRIVER_NAME },
 	{ }
 };
 MODULE_DEVICE_TABLE(i2c, microread_i2c_id);
diff --git a/drivers/nfc/nfcmrvl/i2c.c b/drivers/nfc/nfcmrvl/i2c.c
index 74553134c1b1..39ecf2aeda80 100644
--- a/drivers/nfc/nfcmrvl/i2c.c
+++ b/drivers/nfc/nfcmrvl/i2c.c
@@ -252,7 +252,7 @@ static const struct of_device_id of_nfcmrvl_i2c_match[] __maybe_unused = {
 MODULE_DEVICE_TABLE(of, of_nfcmrvl_i2c_match);
 
 static const struct i2c_device_id nfcmrvl_i2c_id_table[] = {
-	{ "nfcmrvl_i2c", 0 },
+	{ "nfcmrvl_i2c" },
 	{}
 };
 MODULE_DEVICE_TABLE(i2c, nfcmrvl_i2c_id_table);
diff --git a/drivers/nfc/nxp-nci/i2c.c b/drivers/nfc/nxp-nci/i2c.c
index 3ae4b41c59ac..a8aced0b8010 100644
--- a/drivers/nfc/nxp-nci/i2c.c
+++ b/drivers/nfc/nxp-nci/i2c.c
@@ -322,7 +322,7 @@ static void nxp_nci_i2c_remove(struct i2c_client *client)
 }
 
 static const struct i2c_device_id nxp_nci_i2c_id_table[] = {
-	{"nxp-nci_i2c", 0},
+	{ "nxp-nci_i2c" },
 	{}
 };
 MODULE_DEVICE_TABLE(i2c, nxp_nci_i2c_id_table);
diff --git a/drivers/nfc/pn533/i2c.c b/drivers/nfc/pn533/i2c.c
index 438ab9553f7a..132c050a365d 100644
--- a/drivers/nfc/pn533/i2c.c
+++ b/drivers/nfc/pn533/i2c.c
@@ -249,7 +249,7 @@ static const struct of_device_id of_pn533_i2c_match[] __maybe_unused = {
 MODULE_DEVICE_TABLE(of, of_pn533_i2c_match);
 
 static const struct i2c_device_id pn533_i2c_id_table[] = {
-	{ PN533_I2C_DRIVER_NAME, 0 },
+	{ PN533_I2C_DRIVER_NAME },
 	{}
 };
 MODULE_DEVICE_TABLE(i2c, pn533_i2c_id_table);
diff --git a/drivers/nfc/pn544/i2c.c b/drivers/nfc/pn544/i2c.c
index 3f6d74832bac..9fe664960b38 100644
--- a/drivers/nfc/pn544/i2c.c
+++ b/drivers/nfc/pn544/i2c.c
@@ -44,7 +44,7 @@
 					 PN544_HCI_I2C_LLC_MAX_PAYLOAD)
 
 static const struct i2c_device_id pn544_hci_i2c_id_table[] = {
-	{"pn544", 0},
+	{ "pn544" },
 	{}
 };
 
diff --git a/drivers/nfc/s3fwrn5/i2c.c b/drivers/nfc/s3fwrn5/i2c.c
index 720d4a72493c..536c566e3f59 100644
--- a/drivers/nfc/s3fwrn5/i2c.c
+++ b/drivers/nfc/s3fwrn5/i2c.c
@@ -245,7 +245,7 @@ static void s3fwrn5_i2c_remove(struct i2c_client *client)
 }
 
 static const struct i2c_device_id s3fwrn5_i2c_id_table[] = {
-	{S3FWRN5_I2C_DRIVER_NAME, 0},
+	{ S3FWRN5_I2C_DRIVER_NAME },
 	{}
 };
 MODULE_DEVICE_TABLE(i2c, s3fwrn5_i2c_id_table);
diff --git a/drivers/nfc/st-nci/i2c.c b/drivers/nfc/st-nci/i2c.c
index d20a337e90b4..416770adbeba 100644
--- a/drivers/nfc/st-nci/i2c.c
+++ b/drivers/nfc/st-nci/i2c.c
@@ -257,7 +257,7 @@ static void st_nci_i2c_remove(struct i2c_client *client)
 }
 
 static const struct i2c_device_id st_nci_i2c_id_table[] = {
-	{ST_NCI_DRIVER_NAME, 0},
+	{ ST_NCI_DRIVER_NAME },
 	{}
 };
 MODULE_DEVICE_TABLE(i2c, st_nci_i2c_id_table);
diff --git a/drivers/nfc/st21nfca/i2c.c b/drivers/nfc/st21nfca/i2c.c
index 064a63db288b..02c3d11a19c4 100644
--- a/drivers/nfc/st21nfca/i2c.c
+++ b/drivers/nfc/st21nfca/i2c.c
@@ -573,7 +573,7 @@ static void st21nfca_hci_i2c_remove(struct i2c_client *client)
 }
 
 static const struct i2c_device_id st21nfca_hci_i2c_id_table[] = {
-	{ST21NFCA_HCI_DRIVER_NAME, 0},
+	{ ST21NFCA_HCI_DRIVER_NAME },
 	{}
 };
 MODULE_DEVICE_TABLE(i2c, st21nfca_hci_i2c_id_table);
-- 
cgit v1.2.3-58-ga151


From d21a103b612c4bb149faf2b5eb25f4667378824b Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Wed, 19 Jun 2024 16:01:19 +0200
Subject: net: stmmac: unexport stmmac_pltfr_init/exit()

These functions are only used within the compilation unit they're defined
in so there's no reason to export them.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240619140119.26777-1-brgl@bgdev.pl
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 10 ++++------
 drivers/net/ethernet/stmicro/stmmac/stmmac_platform.h |  5 -----
 2 files changed, 4 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
index 54797edc9b38..ad868e8d195d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
@@ -764,8 +764,8 @@ EXPORT_SYMBOL_GPL(stmmac_get_platform_resources);
  * Description: Call the platform's init callback (if any) and propagate
  * the return value.
  */
-int stmmac_pltfr_init(struct platform_device *pdev,
-		      struct plat_stmmacenet_data *plat)
+static int stmmac_pltfr_init(struct platform_device *pdev,
+			     struct plat_stmmacenet_data *plat)
 {
 	int ret = 0;
 
@@ -774,7 +774,6 @@ int stmmac_pltfr_init(struct platform_device *pdev,
 
 	return ret;
 }
-EXPORT_SYMBOL_GPL(stmmac_pltfr_init);
 
 /**
  * stmmac_pltfr_exit
@@ -782,13 +781,12 @@ EXPORT_SYMBOL_GPL(stmmac_pltfr_init);
  * @plat: driver data platform structure
  * Description: Call the platform's exit callback (if any).
  */
-void stmmac_pltfr_exit(struct platform_device *pdev,
-		       struct plat_stmmacenet_data *plat)
+static void stmmac_pltfr_exit(struct platform_device *pdev,
+			      struct plat_stmmacenet_data *plat)
 {
 	if (plat->exit)
 		plat->exit(pdev, plat->bsp_priv);
 }
-EXPORT_SYMBOL_GPL(stmmac_pltfr_exit);
 
 /**
  * stmmac_pltfr_probe
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.h b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.h
index bb6fc7e59aed..72dc1a32e46d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.h
@@ -17,11 +17,6 @@ devm_stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac);
 int stmmac_get_platform_resources(struct platform_device *pdev,
 				  struct stmmac_resources *stmmac_res);
 
-int stmmac_pltfr_init(struct platform_device *pdev,
-		      struct plat_stmmacenet_data *plat);
-void stmmac_pltfr_exit(struct platform_device *pdev,
-		       struct plat_stmmacenet_data *plat);
-
 int stmmac_pltfr_probe(struct platform_device *pdev,
 		       struct plat_stmmacenet_data *plat,
 		       struct stmmac_resources *res);
-- 
cgit v1.2.3-58-ga151


From 08f116c9ea6dfddf2a9649a10032902a4d72f6da Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Wed, 19 Jun 2024 03:52:36 -0700
Subject: wifi: mt76: un-embedd netdev from mt76_dev

Embedding net_device into structures prohibits the usage of flexible
arrays in the net_device structure. For more details, see the discussion
at [1].

Un-embed the net_devices from struct mt76_dev by converting them
into pointers, and allocating them dynamically. Use the leverage
alloc_netdev_dummy() to allocate the net_device object at
mt76_dma_init().

The free of the device occurs at mt76_dma_cleanup().

Link: https://lore.kernel.org/all/20240229225910.79e224cf@kernel.org/ [1]
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Kalle Valo <kvalo@kernel.org>
Link: https://patch.msgid.link/20240619105311.3144908-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/wireless/mediatek/mt76/debugfs.c      |  6 ++---
 drivers/net/wireless/mediatek/mt76/dma.c          | 31 ++++++++++++++++++-----
 drivers/net/wireless/mediatek/mt76/dma.h          |  9 +++++++
 drivers/net/wireless/mediatek/mt76/mt76.h         |  4 +--
 drivers/net/wireless/mediatek/mt76/mt7603/dma.c   |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7615/dma.c   |  6 ++---
 drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7915/dma.c   |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7921/pci.c   |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7925/pci.c   |  2 +-
 drivers/net/wireless/mediatek/mt76/mt792x_dma.c   |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7996/dma.c   |  2 +-
 12 files changed, 49 insertions(+), 21 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/debugfs.c b/drivers/net/wireless/mediatek/mt76/debugfs.c
index ae83be572b94..b6a2746c187d 100644
--- a/drivers/net/wireless/mediatek/mt76/debugfs.c
+++ b/drivers/net/wireless/mediatek/mt76/debugfs.c
@@ -33,8 +33,8 @@ mt76_napi_threaded_set(void *data, u64 val)
 	if (!mt76_is_mmio(dev))
 		return -EOPNOTSUPP;
 
-	if (dev->napi_dev.threaded != val)
-		return dev_set_threaded(&dev->napi_dev, val);
+	if (dev->napi_dev->threaded != val)
+		return dev_set_threaded(dev->napi_dev, val);
 
 	return 0;
 }
@@ -44,7 +44,7 @@ mt76_napi_threaded_get(void *data, u64 *val)
 {
 	struct mt76_dev *dev = data;
 
-	*val = dev->napi_dev.threaded;
+	*val = dev->napi_dev->threaded;
 	return 0;
 }
 
diff --git a/drivers/net/wireless/mediatek/mt76/dma.c b/drivers/net/wireless/mediatek/mt76/dma.c
index f4f88c444e21..5f46d6daeaa7 100644
--- a/drivers/net/wireless/mediatek/mt76/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/dma.c
@@ -916,7 +916,7 @@ int mt76_dma_rx_poll(struct napi_struct *napi, int budget)
 	struct mt76_dev *dev;
 	int qid, done = 0, cur;
 
-	dev = container_of(napi->dev, struct mt76_dev, napi_dev);
+	dev = mt76_priv(napi->dev);
 	qid = napi - dev->napi;
 
 	rcu_read_lock();
@@ -940,18 +940,35 @@ static int
 mt76_dma_init(struct mt76_dev *dev,
 	      int (*poll)(struct napi_struct *napi, int budget))
 {
+	struct mt76_dev **priv;
 	int i;
 
-	init_dummy_netdev(&dev->napi_dev);
-	init_dummy_netdev(&dev->tx_napi_dev);
-	snprintf(dev->napi_dev.name, sizeof(dev->napi_dev.name), "%s",
+	dev->napi_dev = alloc_netdev_dummy(sizeof(struct mt76_dev *));
+	if (!dev->napi_dev)
+		return -ENOMEM;
+
+	/* napi_dev private data points to mt76_dev parent, so, mt76_dev
+	 * can be retrieved given napi_dev
+	 */
+	priv = netdev_priv(dev->napi_dev);
+	*priv = dev;
+
+	dev->tx_napi_dev = alloc_netdev_dummy(sizeof(struct mt76_dev *));
+	if (!dev->tx_napi_dev) {
+		free_netdev(dev->napi_dev);
+		return -ENOMEM;
+	}
+	priv = netdev_priv(dev->tx_napi_dev);
+	*priv = dev;
+
+	snprintf(dev->napi_dev->name, sizeof(dev->napi_dev->name), "%s",
 		 wiphy_name(dev->hw->wiphy));
-	dev->napi_dev.threaded = 1;
+	dev->napi_dev->threaded = 1;
 	init_completion(&dev->mmio.wed_reset);
 	init_completion(&dev->mmio.wed_reset_complete);
 
 	mt76_for_each_q_rx(dev, i) {
-		netif_napi_add(&dev->napi_dev, &dev->napi[i], poll);
+		netif_napi_add(dev->napi_dev, &dev->napi[i], poll);
 		mt76_dma_rx_fill(dev, &dev->q_rx[i], false);
 		napi_enable(&dev->napi[i]);
 	}
@@ -1019,5 +1036,7 @@ void mt76_dma_cleanup(struct mt76_dev *dev)
 
 	mt76_free_pending_txwi(dev);
 	mt76_free_pending_rxwi(dev);
+	free_netdev(dev->napi_dev);
+	free_netdev(dev->tx_napi_dev);
 }
 EXPORT_SYMBOL_GPL(mt76_dma_cleanup);
diff --git a/drivers/net/wireless/mediatek/mt76/dma.h b/drivers/net/wireless/mediatek/mt76/dma.h
index 1de5a2b20f74..e3ddc7a83757 100644
--- a/drivers/net/wireless/mediatek/mt76/dma.h
+++ b/drivers/net/wireless/mediatek/mt76/dma.h
@@ -116,4 +116,13 @@ mt76_dma_should_drop_buf(bool *drop, u32 ctrl, u32 buf1, u32 info)
 	}
 }
 
+static inline void *mt76_priv(struct net_device *dev)
+{
+	struct mt76_dev **priv;
+
+	priv = netdev_priv(dev);
+
+	return *priv;
+}
+
 #endif
diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h
index 11b9f22ca7f3..15f83b5adac7 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76.h
@@ -831,8 +831,8 @@ struct mt76_dev {
 
 	struct mt76_mcu mcu;
 
-	struct net_device napi_dev;
-	struct net_device tx_napi_dev;
+	struct net_device *napi_dev;
+	struct net_device *tx_napi_dev;
 	spinlock_t rx_lock;
 	struct napi_struct napi[__MT_RXQ_MAX];
 	struct sk_buff_head rx_skb[__MT_RXQ_MAX];
diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/dma.c b/drivers/net/wireless/mediatek/mt76/mt7603/dma.c
index 14304b063715..ea017f22fff2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7603/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7603/dma.c
@@ -242,7 +242,7 @@ int mt7603_dma_init(struct mt7603_dev *dev)
 	if (ret)
 		return ret;
 
-	netif_napi_add_tx(&dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
+	netif_napi_add_tx(dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
 			  mt7603_poll_tx);
 	napi_enable(&dev->mt76.tx_napi);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/dma.c b/drivers/net/wireless/mediatek/mt76/mt7615/dma.c
index e7135b2f1742..bcf7864312d7 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/dma.c
@@ -67,7 +67,7 @@ static int mt7615_poll_tx(struct napi_struct *napi, int budget)
 {
 	struct mt7615_dev *dev;
 
-	dev = container_of(napi, struct mt7615_dev, mt76.tx_napi);
+	dev = mt76_priv(napi->dev);
 	if (!mt76_connac_pm_ref(&dev->mphy, &dev->pm)) {
 		napi_complete(napi);
 		queue_work(dev->mt76.wq, &dev->pm.wake_work);
@@ -89,7 +89,7 @@ static int mt7615_poll_rx(struct napi_struct *napi, int budget)
 	struct mt7615_dev *dev;
 	int done;
 
-	dev = container_of(napi->dev, struct mt7615_dev, mt76.napi_dev);
+	dev = mt76_priv(napi->dev);
 
 	if (!mt76_connac_pm_ref(&dev->mphy, &dev->pm)) {
 		napi_complete(napi);
@@ -282,7 +282,7 @@ int mt7615_dma_init(struct mt7615_dev *dev)
 	if (ret < 0)
 		return ret;
 
-	netif_napi_add_tx(&dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
+	netif_napi_add_tx(dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
 			  mt7615_poll_tx);
 	napi_enable(&dev->mt76.tx_napi);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c b/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c
index e5ad635d3c56..35b7ebc2c9c6 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x02_mmio.c
@@ -239,7 +239,7 @@ int mt76x02_dma_init(struct mt76x02_dev *dev)
 	if (ret)
 		return ret;
 
-	netif_napi_add_tx(&dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
+	netif_napi_add_tx(dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
 			  mt76x02_poll_tx);
 	napi_enable(&dev->mt76.tx_napi);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/dma.c b/drivers/net/wireless/mediatek/mt76/mt7915/dma.c
index 0baa82c8df5a..0c62272fe7d0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/dma.c
@@ -578,7 +578,7 @@ int mt7915_dma_init(struct mt7915_dev *dev, struct mt7915_phy *phy2)
 	if (ret < 0)
 		return ret;
 
-	netif_napi_add_tx(&dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
+	netif_napi_add_tx(dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
 			  mt7915_poll_tx);
 	napi_enable(&dev->mt76.tx_napi);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index f768e9389ac6..e75e7b6d3aaf 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -219,7 +219,7 @@ static int mt7921_dma_init(struct mt792x_dev *dev)
 	if (ret < 0)
 		return ret;
 
-	netif_napi_add_tx(&dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
+	netif_napi_add_tx(dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
 			  mt792x_poll_tx);
 	napi_enable(&dev->mt76.tx_napi);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/pci.c b/drivers/net/wireless/mediatek/mt76/mt7925/pci.c
index 07b74d492ce1..577574fb7a1e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/pci.c
@@ -254,7 +254,7 @@ static int mt7925_dma_init(struct mt792x_dev *dev)
 	if (ret < 0)
 		return ret;
 
-	netif_napi_add_tx(&dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
+	netif_napi_add_tx(dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
 			  mt792x_poll_tx);
 	napi_enable(&dev->mt76.tx_napi);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_dma.c b/drivers/net/wireless/mediatek/mt76/mt792x_dma.c
index 5cc2d59b774a..6f9db782338e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_dma.c
@@ -340,7 +340,7 @@ int mt792x_poll_rx(struct napi_struct *napi, int budget)
 	struct mt792x_dev *dev;
 	int done;
 
-	dev = container_of(napi->dev, struct mt792x_dev, mt76.napi_dev);
+	dev = mt76_priv(napi->dev);
 
 	if (!mt76_connac_pm_ref(&dev->mphy, &dev->pm)) {
 		napi_complete(napi);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/dma.c b/drivers/net/wireless/mediatek/mt76/mt7996/dma.c
index 73e633d0d700..69a7d9b2e38b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/dma.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/dma.c
@@ -641,7 +641,7 @@ int mt7996_dma_init(struct mt7996_dev *dev)
 	if (ret < 0)
 		return ret;
 
-	netif_napi_add_tx(&dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
+	netif_napi_add_tx(dev->mt76.tx_napi_dev, &dev->mt76.tx_napi,
 			  mt7996_poll_tx);
 	napi_enable(&dev->mt76.tx_napi);
 
-- 
cgit v1.2.3-58-ga151


From 01479f1b912aa499057fd29deddb1ffb9c72fa0a Mon Sep 17 00:00:00 2001
From: Frank Li <Frank.Li@nxp.com>
Date: Tue, 18 Jun 2024 17:53:45 -0400
Subject: dt-bindings: ptp: Convert ptp-qoirq to yaml format

Convert ptp-qoirq from txt to yaml format.

Additional change:
- Fixed example interrupts proptery. Need only 1 irq by check MPC8313 spec.
- Move Reference clock context under clk,sel.
- Interrupts is not required property.
- Use low case for hex value.
- Check reference manual of MPC8313, p1010 and so on, which dts use more
than 1 irqs. Only 1 irq for each ptp device. Check driver code
(drivers/ptp/ptp_qoriq.c) and only 1 irq used. So original description is
wrong.
- Remove comments for compatible string.

Signed-off-by: Frank Li <Frank.Li@nxp.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patch.msgid.link/20240618-ls_fman-v2-1-f00a82623d8e@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/fsl-fman.txt |   2 +-
 .../devicetree/bindings/net/fsl-tsec-phy.txt       |   2 +-
 Documentation/devicetree/bindings/ptp/fsl,ptp.yaml | 144 +++++++++++++++++++++
 .../devicetree/bindings/ptp/ptp-qoriq.txt          |  87 -------------
 MAINTAINERS                                        |   2 +-
 5 files changed, 147 insertions(+), 90 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/ptp/fsl,ptp.yaml
 delete mode 100644 Documentation/devicetree/bindings/ptp/ptp-qoriq.txt

diff --git a/Documentation/devicetree/bindings/net/fsl-fman.txt b/Documentation/devicetree/bindings/net/fsl-fman.txt
index bda4b41af074..5e02b4b286f6 100644
--- a/Documentation/devicetree/bindings/net/fsl-fman.txt
+++ b/Documentation/devicetree/bindings/net/fsl-fman.txt
@@ -237,7 +237,7 @@ Refer to Documentation/devicetree/bindings/net/fsl,fman-dtsec.yaml
 ============================================================================
 FMan IEEE 1588 Node
 
-Refer to Documentation/devicetree/bindings/ptp/ptp-qoriq.txt
+Refer to Documentation/devicetree/bindings/ptp/fsl,ptp.yaml
 
 =============================================================================
 FMan MDIO Node
diff --git a/Documentation/devicetree/bindings/net/fsl-tsec-phy.txt b/Documentation/devicetree/bindings/net/fsl-tsec-phy.txt
index 047bdf7bdd2f..9c9668c1b6a2 100644
--- a/Documentation/devicetree/bindings/net/fsl-tsec-phy.txt
+++ b/Documentation/devicetree/bindings/net/fsl-tsec-phy.txt
@@ -86,4 +86,4 @@ Example:
 
 * Gianfar PTP clock nodes
 
-Refer to Documentation/devicetree/bindings/ptp/ptp-qoriq.txt
+Refer to Documentation/devicetree/bindings/ptp/fsl,ptp.yaml
diff --git a/Documentation/devicetree/bindings/ptp/fsl,ptp.yaml b/Documentation/devicetree/bindings/ptp/fsl,ptp.yaml
new file mode 100644
index 000000000000..3bb8615e3e91
--- /dev/null
+++ b/Documentation/devicetree/bindings/ptp/fsl,ptp.yaml
@@ -0,0 +1,144 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/ptp/fsl,ptp.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Freescale QorIQ 1588 timer based PTP clock
+
+maintainers:
+  - Frank Li <Frank.Li@nxp.com>
+
+properties:
+  compatible:
+    enum:
+      - fsl,etsec-ptp
+      - fsl,fman-ptp-timer
+      - fsl,dpaa2-ptp
+      - fsl,enetc-ptp
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  clocks:
+    maxItems: 1
+
+  fsl,cksel:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description: |
+      Timer reference clock source.
+
+      Reference clock source is determined by the value, which is holded
+      in CKSEL bits in TMR_CTRL register. "fsl,cksel" property keeps the
+      value, which will be directly written in those bits, that is why,
+      according to reference manual, the next clock sources can be used:
+
+      For eTSEC,
+      <0> - external high precision timer reference clock (TSEC_TMR_CLK
+            input is used for this purpose);
+      <1> - eTSEC system clock;
+      <2> - eTSEC1 transmit clock;
+      <3> - RTC clock input.
+
+      For DPAA FMan,
+      <0> - external high precision timer reference clock (TMR_1588_CLK)
+      <1> - MAC system clock (1/2 FMan clock)
+      <2> - reserved
+      <3> - RTC clock oscillator
+
+  fsl,tclk-period:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description: Timer reference clock period in nanoseconds.
+
+  fsl,tmr-prsc:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description: Prescaler, divides the output clock.
+
+  fsl,tmr-add:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description: Frequency compensation value.
+
+  fsl,tmr-fiper1:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description: Fixed interval period pulse generator.
+
+  fsl,tmr-fiper2:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description: Fixed interval period pulse generator.
+
+  fsl,tmr-fiper3:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description:
+      Fixed interval period pulse generator.
+      Supported only on DPAA2 and ENETC hardware.
+
+  fsl,max-adj:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description: |
+      Maximum frequency adjustment in parts per billion.
+
+      These properties set the operational parameters for the PTP
+      clock. You must choose these carefully for the clock to work right.
+      Here is how to figure good values:
+
+      TimerOsc     = selected reference clock   MHz
+      tclk_period  = desired clock period       nanoseconds
+      NominalFreq  = 1000 / tclk_period         MHz
+      FreqDivRatio = TimerOsc / NominalFreq     (must be greater that 1.0)
+      tmr_add      = ceil(2^32 / FreqDivRatio)
+      OutputClock  = NominalFreq / tmr_prsc     MHz
+      PulseWidth   = 1 / OutputClock            microseconds
+      FiperFreq1   = desired frequency in Hz
+      FiperDiv1    = 1000000 * OutputClock / FiperFreq1
+      tmr_fiper1   = tmr_prsc * tclk_period * FiperDiv1 - tclk_period
+      max_adj      = 1000000000 * (FreqDivRatio - 1.0) - 1
+
+      The calculation for tmr_fiper2 is the same as for tmr_fiper1. The
+      driver expects that tmr_fiper1 will be correctly set to produce a 1
+      Pulse Per Second (PPS) signal, since this will be offered to the PPS
+      subsystem to synchronize the Linux clock.
+
+      When this attribute is not used, the IEEE 1588 timer reference clock
+      will use the eTSEC system clock (for Gianfar) or the MAC system
+      clock (for DPAA).
+
+  fsl,extts-fifo:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description:
+      The presence of this property indicates hardware
+      support for the external trigger stamp FIFO
+
+  little-endian:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description:
+      The presence of this property indicates the 1588 timer
+      support for the external trigger stamp FIFO.
+      IP block is little-endian mode. The default endian mode
+      is big-endian.
+
+required:
+  - compatible
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/irq.h>
+
+    phc@24e00 {
+        compatible = "fsl,etsec-ptp";
+        reg = <0x24e00 0xb0>;
+        interrupts = <12 IRQ_TYPE_LEVEL_LOW>;
+        interrupt-parent = <&ipic>;
+        fsl,cksel       = <1>;
+        fsl,tclk-period = <10>;
+        fsl,tmr-prsc    = <100>;
+        fsl,tmr-add     = <0x999999a4>;
+        fsl,tmr-fiper1  = <0x3b9ac9f6>;
+        fsl,tmr-fiper2  = <0x00018696>;
+        fsl,max-adj     = <659999998>;
+    };
diff --git a/Documentation/devicetree/bindings/ptp/ptp-qoriq.txt b/Documentation/devicetree/bindings/ptp/ptp-qoriq.txt
deleted file mode 100644
index 743eda754e65..000000000000
--- a/Documentation/devicetree/bindings/ptp/ptp-qoriq.txt
+++ /dev/null
@@ -1,87 +0,0 @@
-* Freescale QorIQ 1588 timer based PTP clock
-
-General Properties:
-
-  - compatible   Should be "fsl,etsec-ptp" for eTSEC
-                 Should be "fsl,fman-ptp-timer" for DPAA FMan
-                 Should be "fsl,dpaa2-ptp" for DPAA2
-                 Should be "fsl,enetc-ptp" for ENETC
-  - reg          Offset and length of the register set for the device
-  - interrupts   There should be at least two interrupts. Some devices
-                 have as many as four PTP related interrupts.
-
-Clock Properties:
-
-  - fsl,cksel        Timer reference clock source.
-  - fsl,tclk-period  Timer reference clock period in nanoseconds.
-  - fsl,tmr-prsc     Prescaler, divides the output clock.
-  - fsl,tmr-add      Frequency compensation value.
-  - fsl,tmr-fiper1   Fixed interval period pulse generator.
-  - fsl,tmr-fiper2   Fixed interval period pulse generator.
-  - fsl,tmr-fiper3   Fixed interval period pulse generator.
-                     Supported only on DPAA2 and ENETC hardware.
-  - fsl,max-adj      Maximum frequency adjustment in parts per billion.
-  - fsl,extts-fifo   The presence of this property indicates hardware
-		     support for the external trigger stamp FIFO.
-  - little-endian    The presence of this property indicates the 1588 timer
-		     IP block is little-endian mode. The default endian mode
-		     is big-endian.
-
-  These properties set the operational parameters for the PTP
-  clock. You must choose these carefully for the clock to work right.
-  Here is how to figure good values:
-
-  TimerOsc     = selected reference clock   MHz
-  tclk_period  = desired clock period       nanoseconds
-  NominalFreq  = 1000 / tclk_period         MHz
-  FreqDivRatio = TimerOsc / NominalFreq     (must be greater that 1.0)
-  tmr_add      = ceil(2^32 / FreqDivRatio)
-  OutputClock  = NominalFreq / tmr_prsc     MHz
-  PulseWidth   = 1 / OutputClock            microseconds
-  FiperFreq1   = desired frequency in Hz
-  FiperDiv1    = 1000000 * OutputClock / FiperFreq1
-  tmr_fiper1   = tmr_prsc * tclk_period * FiperDiv1 - tclk_period
-  max_adj      = 1000000000 * (FreqDivRatio - 1.0) - 1
-
-  The calculation for tmr_fiper2 is the same as for tmr_fiper1. The
-  driver expects that tmr_fiper1 will be correctly set to produce a 1
-  Pulse Per Second (PPS) signal, since this will be offered to the PPS
-  subsystem to synchronize the Linux clock.
-
-  Reference clock source is determined by the value, which is holded
-  in CKSEL bits in TMR_CTRL register. "fsl,cksel" property keeps the
-  value, which will be directly written in those bits, that is why,
-  according to reference manual, the next clock sources can be used:
-
-  For eTSEC,
-  <0> - external high precision timer reference clock (TSEC_TMR_CLK
-        input is used for this purpose);
-  <1> - eTSEC system clock;
-  <2> - eTSEC1 transmit clock;
-  <3> - RTC clock input.
-
-  For DPAA FMan,
-  <0> - external high precision timer reference clock (TMR_1588_CLK)
-  <1> - MAC system clock (1/2 FMan clock)
-  <2> - reserved
-  <3> - RTC clock oscillator
-
-  When this attribute is not used, the IEEE 1588 timer reference clock
-  will use the eTSEC system clock (for Gianfar) or the MAC system
-  clock (for DPAA).
-
-Example:
-
-	ptp_clock@24e00 {
-		compatible = "fsl,etsec-ptp";
-		reg = <0x24E00 0xB0>;
-		interrupts = <12 0x8 13 0x8>;
-		interrupt-parent = < &ipic >;
-		fsl,cksel       = <1>;
-		fsl,tclk-period = <10>;
-		fsl,tmr-prsc    = <100>;
-		fsl,tmr-add     = <0x999999A4>;
-		fsl,tmr-fiper1  = <0x3B9AC9F6>;
-		fsl,tmr-fiper2  = <0x00018696>;
-		fsl,max-adj     = <659999998>;
-	};
diff --git a/MAINTAINERS b/MAINTAINERS
index 462ff5be6f00..9d2fbf651848 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -8816,7 +8816,7 @@ FREESCALE QORIQ PTP CLOCK DRIVER
 M:	Yangbo Lu <yangbo.lu@nxp.com>
 L:	netdev@vger.kernel.org
 S:	Maintained
-F:	Documentation/devicetree/bindings/ptp/ptp-qoriq.txt
+F:	Documentation/devicetree/bindings/ptp/fsl,ptp.yaml
 F:	drivers/net/ethernet/freescale/dpaa2/dpaa2-ptp*
 F:	drivers/net/ethernet/freescale/dpaa2/dprtc*
 F:	drivers/net/ethernet/freescale/enetc/enetc_ptp.c
-- 
cgit v1.2.3-58-ga151


From 243996d172a68fba79765cfe20ab910c29ec5f50 Mon Sep 17 00:00:00 2001
From: Frank Li <Frank.Li@nxp.com>
Date: Tue, 18 Jun 2024 17:53:46 -0400
Subject: dt-bindings: net: Convert fsl-fman to yaml

Convert fsl-fman from txt to yaml format and split it fsl,fman.yam,
fsl,fman-port.yaml, fsl-muram.yaml, fsl-mdio.yaml.

Addition changes:
fsl,fman.yaml:
  - Fixed interrupts in example.
  - Fixed ethernet@e8000 miss } in example.
  - ptp-timer add label in example.
  - Ref to new fsl,fman*.yaml.
  - Reorder property in example.
  - Keep only one example.
  - Add const for #address-cells and #size-cells.
  - Use defined interrupt type.
  - ptp example use node name phc.

fsl,fman-port:
  - Keep only one example.

fsl,fman-mdio:
  - Add little-endian property.
  - Add ref to mdio.yaml.
  - Remove suppress-preamble.
  - Add #address-cells and #size-cells in example.
  - Remove clock-frequency, which already describe in mmio.yaml.

fsl,muram.yaml:
  - Add reg property.
  - Remove range property.
  - Use reg instead of range in example.

Signed-off-by: Frank Li <Frank.Li@nxp.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patch.msgid.link/20240618-ls_fman-v2-2-f00a82623d8e@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../devicetree/bindings/net/fsl,fman-mdio.yaml     | 123 +++++
 .../devicetree/bindings/net/fsl,fman-muram.yaml    |  40 ++
 .../devicetree/bindings/net/fsl,fman-port.yaml     |  75 +++
 .../devicetree/bindings/net/fsl,fman.yaml          | 204 ++++++++
 Documentation/devicetree/bindings/net/fsl-fman.txt | 548 ---------------------
 MAINTAINERS                                        |   2 +-
 6 files changed, 443 insertions(+), 549 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/net/fsl,fman-mdio.yaml
 create mode 100644 Documentation/devicetree/bindings/net/fsl,fman-muram.yaml
 create mode 100644 Documentation/devicetree/bindings/net/fsl,fman-port.yaml
 create mode 100644 Documentation/devicetree/bindings/net/fsl,fman.yaml
 delete mode 100644 Documentation/devicetree/bindings/net/fsl-fman.txt

diff --git a/Documentation/devicetree/bindings/net/fsl,fman-mdio.yaml b/Documentation/devicetree/bindings/net/fsl,fman-mdio.yaml
new file mode 100644
index 000000000000..6b2c0aa407a2
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/fsl,fman-mdio.yaml
@@ -0,0 +1,123 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/fsl,fman-mdio.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Freescale Frame Manager MDIO Device
+
+maintainers:
+  - Frank Li <Frank.Li@nxp.com>
+
+description: FMan MDIO Node.
+  The MDIO is a bus to which the PHY devices are connected.
+
+properties:
+  compatible:
+    enum:
+      - fsl,fman-mdio
+      - fsl,fman-xmdio
+      - fsl,fman-memac-mdio
+    description:
+      Must include "fsl,fman-mdio" for 1 Gb/s MDIO from FMan v2.
+      Must include "fsl,fman-xmdio" for 10 Gb/s MDIO from FMan v2.
+      Must include "fsl,fman-memac-mdio" for 1/10 Gb/s MDIO from
+      FMan v3.
+
+  reg:
+    maxItems: 1
+
+  clocks:
+    items:
+      - description: A reference to the input clock of the controller
+          from which the MDC frequency is derived.
+
+  interrupts:
+    maxItems: 1
+
+  fsl,fman-internal-mdio:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description:
+      Fman has internal MDIO for internal PCS(Physical
+      Coding Sublayer) PHYs and external MDIO for external PHYs.
+      The settings and programming routines for internal/external
+      MDIO are different. Must be included for internal MDIO.
+
+  fsl,erratum-a009885:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description: Indicates the presence of the A009885
+      erratum describing that the contents of MDIO_DATA may
+      become corrupt unless it is read within 16 MDC cycles
+      of MDIO_CFG[BSY] being cleared, when performing an
+      MDIO read operation.
+
+  fsl,erratum-a011043:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description:
+      Indicates the presence of the A011043 erratum
+      describing that the MDIO_CFG[MDIO_RD_ER] bit may be falsely
+      set when reading internal PCS registers. MDIO reads to
+      internal PCS registers may result in having the
+      MDIO_CFG[MDIO_RD_ER] bit set, even when there is no error and
+      read data (MDIO_DATA[MDIO_DATA]) is correct.
+      Software may get false read error when reading internal
+      PCS registers through MDIO. As a workaround, all internal
+      MDIO accesses should ignore the MDIO_CFG[MDIO_RD_ER] bit.
+
+      For internal PHY device on internal mdio bus, a PHY node should be created.
+      See the definition of the PHY node in booting-without-of.txt for an
+      example of how to define a PHY (Internal PHY has no interrupt line).
+      - For "fsl,fman-mdio" compatible internal mdio bus, the PHY is TBI PHY.
+      - For "fsl,fman-memac-mdio" compatible internal mdio bus, the PHY is PCS PHY.
+        The PCS PHY address should correspond to the value of the appropriate
+        MDEV_PORT.
+
+  little-endian:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description:
+      IP block is little-endian mode. The default endian mode is big-endian.
+
+required:
+  - compatible
+  - reg
+
+allOf:
+  - $ref: mdio.yaml#
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    mdio@f1000 {
+        compatible = "fsl,fman-xmdio";
+        reg = <0xf1000 0x1000>;
+        interrupts = <101 2 0 0>;
+    };
+
+  - |
+    mdio@e3120 {
+        compatible = "fsl,fman-mdio";
+        reg = <0xe3120 0xee0>;
+        fsl,fman-internal-mdio;
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        tbi-phy@8 {
+            reg = <0x8>;
+            device_type = "tbi-phy";
+        };
+    };
+
+  - |
+    mdio@f1000 {
+        compatible = "fsl,fman-memac-mdio";
+        reg = <0xf1000 0x1000>;
+        fsl,fman-internal-mdio;
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        pcsphy6: ethernet-phy@0 {
+            reg = <0x0>;
+        };
+    };
+
diff --git a/Documentation/devicetree/bindings/net/fsl,fman-muram.yaml b/Documentation/devicetree/bindings/net/fsl,fman-muram.yaml
new file mode 100644
index 000000000000..aa71acc7fa5b
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/fsl,fman-muram.yaml
@@ -0,0 +1,40 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/fsl,fman-muram.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Freescale Frame Manager MURAM Device
+
+maintainers:
+  - Frank Li <Frank.Li@nxp.com>
+
+description: |
+  FMan Internal memory - shared between all the FMan modules.
+  It contains data structures that are common and written to or read by
+  the modules.
+
+  FMan internal memory is split into the following parts:
+    Packet buffering (Tx/Rx FIFOs)
+    Frames internal context
+
+properties:
+  compatible:
+    enum:
+      - fsl,fman-muram
+
+  reg:
+    maxItems: 1
+
+required:
+  - compatible
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    muram@0 {
+        compatible = "fsl,fman-muram";
+        reg = <0x0 0x28000>;
+    };
diff --git a/Documentation/devicetree/bindings/net/fsl,fman-port.yaml b/Documentation/devicetree/bindings/net/fsl,fman-port.yaml
new file mode 100644
index 000000000000..9de445307830
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/fsl,fman-port.yaml
@@ -0,0 +1,75 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/fsl,fman-port.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Freescale Frame Manager Port Device
+
+maintainers:
+  - Frank Li <Frank.Li@nxp.com>
+
+description: |
+  The Frame Manager (FMan) supports several types of hardware ports:
+    Ethernet receiver (RX)
+    Ethernet transmitter (TX)
+    Offline/Host command (O/H)
+
+properties:
+  compatible:
+    enum:
+      - fsl,fman-v2-port-oh
+      - fsl,fman-v2-port-rx
+      - fsl,fman-v2-port-tx
+      - fsl,fman-v3-port-oh
+      - fsl,fman-v3-port-rx
+      - fsl,fman-v3-port-tx
+
+  cell-index:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description:
+      Specifies the hardware port id.
+      Each hardware port on the FMan has its own hardware PortID.
+      Super set of all hardware Port IDs available at FMan Reference
+      Manual under "FMan Hardware Ports in Freescale Devices" table.
+
+      Each hardware port is assigned a 4KB, port-specific page in
+      the FMan hardware port memory region (which is part of the
+      FMan memory map). The first 4 KB in the FMan hardware ports
+      memory region is used for what are called common registers.
+      The subsequent 63 4KB pages are allocated to the hardware
+      ports.
+      The page of a specific port is determined by the cell-index.
+
+  reg:
+    items:
+      - description: There is one reg region describing the port
+          configuration registers.
+
+  fsl,fman-10g-port:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description: The default port rate is 1G.
+      If this property exists, the port is s 10G port.
+
+  fsl,fman-best-effort-port:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description: The default port rate is 1G.
+      Can be defined only if 10G-support is set.
+      This property marks a best-effort 10G port (10G port that
+      may not be capable of line rate).
+
+required:
+  - compatible
+  - reg
+  - cell-index
+
+additionalProperties: false
+
+examples:
+  - |
+    port@a8000 {
+        compatible = "fsl,fman-v2-port-tx";
+        reg = <0xa8000 0x1000>;
+        cell-index = <0x28>;
+    };
+
diff --git a/Documentation/devicetree/bindings/net/fsl,fman.yaml b/Documentation/devicetree/bindings/net/fsl,fman.yaml
new file mode 100644
index 000000000000..7908f67413de
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/fsl,fman.yaml
@@ -0,0 +1,204 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/fsl,fman.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Freescale Frame Manager Device
+
+maintainers:
+  - Frank Li <Frank.Li@nxp.com>
+
+description:
+  Due to the fact that the FMan is an aggregation of sub-engines (ports, MACs,
+  etc.) the FMan node will have child nodes for each of them.
+
+properties:
+  compatible:
+    enum:
+      - fsl,fman
+    description:
+      FMan version can be determined via FM_IP_REV_1 register in the
+      FMan block. The offset is 0xc4 from the beginning of the
+      Frame Processing Manager memory map (0xc3000 from the
+      beginning of the FMan node).
+
+  cell-index:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description: |
+      Specifies the index of the FMan unit.
+
+      The cell-index value may be used by the SoC, to identify the
+      FMan unit in the SoC memory map. In the table below,
+      there's a description of the cell-index use in each SoC:
+
+      - P1023:
+      register[bit]      FMan unit  cell-index
+      ============================================================
+      DEVDISR[1]      1    0
+
+      - P2041, P3041, P4080 P5020, P5040:
+      register[bit]      FMan unit  cell-index
+      ============================================================
+      DCFG_DEVDISR2[6]    1    0
+      DCFG_DEVDISR2[14]    2    1
+        (Second FM available only in P4080 and P5040)
+
+      - B4860, T1040, T2080, T4240:
+      register[bit]      FMan unit  cell-index
+      ============================================================
+      DCFG_CCSR_DEVDISR2[24]    1    0
+      DCFG_CCSR_DEVDISR2[25]    2    1
+        (Second FM available only in T4240)
+
+      DEVDISR, DCFG_DEVDISR2 and DCFG_CCSR_DEVDISR2 are located in
+      the specific SoC "Device Configuration/Pin Control" Memory
+      Map.
+
+  reg:
+    items:
+      - description: BMI configuration registers.
+      - description: QMI configuration registers.
+      - description: DMA configuration registers.
+      - description: FPM configuration registers.
+      - description: FMan controller configuration registers.
+    minItems: 1
+
+  ranges: true
+
+  clocks:
+    maxItems: 1
+
+  clock-names:
+    items:
+      - const: fmanclk
+
+  interrupts:
+    items:
+      - description: The first element is associated with the event interrupts.
+      - description: the second element is associated with the error interrupts.
+
+  fsl,qman-channel-range:
+    $ref: /schemas/types.yaml#/definitions/uint32-array
+    description:
+      Specifies the range of the available dedicated
+      channels in the FMan. The first cell specifies the beginning
+      of the range and the second cell specifies the number of
+      channels
+    items:
+      - description: The first cell specifies the beginning of the range.
+      - description: |
+          The second cell specifies the number of channels.
+          Further information available at:
+          "Work Queue (WQ) Channel Assignments in the QMan" section
+          in DPAA Reference Manual.
+
+  fsl,qman:
+    $ref: /schemas/types.yaml#/definitions/phandle
+    description: See soc/fsl/qman.txt
+
+  fsl,bman:
+    $ref: /schemas/types.yaml#/definitions/phandle
+    description: See soc/fsl/bman.txt
+
+  fsl,erratum-a050385:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description: A boolean property. Indicates the presence of the
+      erratum A050385 which indicates that DMA transactions that are
+      split can result in a FMan lock.
+
+  '#address-cells':
+    const: 1
+
+  '#size-cells':
+    const: 1
+
+patternProperties:
+  '^muram@[a-f0-9]+$':
+    $ref: fsl,fman-muram.yaml
+
+  '^port@[a-f0-9]+$':
+    $ref: fsl,fman-port.yaml
+
+  '^ethernet@[a-f0-9]+$':
+    $ref: fsl,fman-dtsec.yaml
+
+  '^mdio@[a-f0-9]+$':
+    $ref: fsl,fman-mdio.yaml
+
+  '^phc@[a-f0-9]+$':
+    $ref: /schemas/ptp/fsl,ptp.yaml
+
+required:
+  - compatible
+  - cell-index
+  - reg
+  - ranges
+  - clocks
+  - clock-names
+  - interrupts
+  - fsl,qman-channel-range
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/irq.h>
+
+    fman@400000 {
+        compatible = "fsl,fman";
+        reg = <0x400000 0x100000>;
+        ranges = <0 0x400000 0x100000>;
+        #address-cells = <1>;
+        #size-cells = <1>;
+        cell-index = <1>;
+        clocks = <&fman_clk>;
+        clock-names = "fmanclk";
+        interrupts = <96 IRQ_TYPE_EDGE_FALLING>,
+                     <16 IRQ_TYPE_EDGE_FALLING>;
+        fsl,qman-channel-range = <0x40 0xc>;
+
+        muram@0 {
+            compatible = "fsl,fman-muram";
+            reg = <0x0 0x28000>;
+        };
+
+        port@81000 {
+            cell-index = <1>;
+            compatible = "fsl,fman-v2-port-oh";
+            reg = <0x81000 0x1000>;
+        };
+
+        fman1_rx_0x8: port@88000 {
+            cell-index = <0x8>;
+            compatible = "fsl,fman-v2-port-rx";
+            reg = <0x88000 0x1000>;
+        };
+
+        fman1_tx_0x28: port@a8000 {
+            cell-index = <0x28>;
+            compatible = "fsl,fman-v2-port-tx";
+            reg = <0xa8000 0x1000>;
+        };
+
+        ethernet@e0000 {
+            compatible = "fsl,fman-dtsec";
+            cell-index = <0>;
+            reg = <0xe0000 0x1000>;
+            ptp-timer = <&ptp_timer>;
+            fsl,fman-ports = <&fman1_rx_0x8 &fman1_tx_0x28>;
+            tbi-handle = <&tbi5>;
+        };
+
+        ptp_timer: phc@fe000 {
+            compatible = "fsl,fman-ptp-timer";
+            reg = <0xfe000 0x1000>;
+            interrupts = <12 IRQ_TYPE_LEVEL_LOW>;
+        };
+
+        mdio@f1000 {
+            compatible = "fsl,fman-xmdio";
+            reg = <0xf1000 0x1000>;
+            interrupts = <101 IRQ_TYPE_EDGE_FALLING>;
+        };
+    };
diff --git a/Documentation/devicetree/bindings/net/fsl-fman.txt b/Documentation/devicetree/bindings/net/fsl-fman.txt
deleted file mode 100644
index 5e02b4b286f6..000000000000
--- a/Documentation/devicetree/bindings/net/fsl-fman.txt
+++ /dev/null
@@ -1,548 +0,0 @@
-=============================================================================
-Freescale Frame Manager Device Bindings
-
-CONTENTS
-  - FMan Node
-  - FMan Port Node
-  - FMan MURAM Node
-  - FMan dTSEC/XGEC/mEMAC Node
-  - FMan IEEE 1588 Node
-  - FMan MDIO Node
-  - Example
-
-=============================================================================
-FMan Node
-
-DESCRIPTION
-
-Due to the fact that the FMan is an aggregation of sub-engines (ports, MACs,
-etc.) the FMan node will have child nodes for each of them.
-
-PROPERTIES
-
-- compatible
-		Usage: required
-		Value type: <stringlist>
-		Definition: Must include "fsl,fman"
-		FMan version can be determined via FM_IP_REV_1 register in the
-		FMan block. The offset is 0xc4 from the beginning of the
-		Frame Processing Manager memory map (0xc3000 from the
-		beginning of the FMan node).
-
-- cell-index
-		Usage: required
-		Value type: <u32>
-		Definition: Specifies the index of the FMan unit.
-
-		The cell-index value may be used by the SoC, to identify the
-		FMan unit in the SoC memory map. In the table below,
-		there's a description of the cell-index use in each SoC:
-
-		- P1023:
-		register[bit]			FMan unit	cell-index
-		============================================================
-		DEVDISR[1]			1		0
-
-		- P2041, P3041, P4080 P5020, P5040:
-		register[bit]			FMan unit	cell-index
-		============================================================
-		DCFG_DEVDISR2[6]		1		0
-		DCFG_DEVDISR2[14]		2		1
-			(Second FM available only in P4080 and P5040)
-
-		- B4860, T1040, T2080, T4240:
-		register[bit]			FMan unit	cell-index
-		============================================================
-		DCFG_CCSR_DEVDISR2[24]		1		0
-		DCFG_CCSR_DEVDISR2[25]		2		1
-			(Second FM available only in T4240)
-
-		DEVDISR, DCFG_DEVDISR2 and DCFG_CCSR_DEVDISR2 are located in
-		the specific SoC "Device Configuration/Pin Control" Memory
-		Map.
-
-- reg
-		Usage: required
-		Value type: <prop-encoded-array>
-		Definition: A standard property. Specifies the offset of the
-		following configuration registers:
-		- BMI configuration registers.
-		- QMI configuration registers.
-		- DMA configuration registers.
-		- FPM configuration registers.
-		- FMan controller configuration registers.
-
-- ranges
-		Usage: required
-		Value type: <prop-encoded-array>
-		Definition: A standard property.
-
-- clocks
-		Usage: required
-		Value type: <prop-encoded-array>
-		Definition: phandle for the fman input clock.
-
-- clock-names
-		usage: required
-		Value type: <stringlist>
-		Definition: "fmanclk" for the fman input clock.
-
-- interrupts
-		Usage: required
-		Value type: <prop-encoded-array>
-		Definition: A pair of IRQs are specified in this property.
-		The first element is associated with the event interrupts and
-		the second element is associated with the error interrupts.
-
-- fsl,qman-channel-range
-		Usage: required
-		Value type: <prop-encoded-array>
-		Definition: Specifies the range of the available dedicated
-		channels in the FMan. The first cell specifies the beginning
-		of the range and the second cell specifies the number of
-		channels.
-		Further information available at:
-		"Work Queue (WQ) Channel Assignments in the QMan" section
-		in DPAA Reference Manual.
-
-- fsl,qman
-- fsl,bman
-		Usage: required
-		Definition: See soc/fsl/qman.txt and soc/fsl/bman.txt
-
-- fsl,erratum-a050385
-		Usage: optional
-		Value type: boolean
-		Definition: A boolean property. Indicates the presence of the
-		erratum A050385 which indicates that DMA transactions that are
-		split can result in a FMan lock.
-
-=============================================================================
-FMan MURAM Node
-
-DESCRIPTION
-
-FMan Internal memory - shared between all the FMan modules.
-It contains data structures that are common and written to or read by
-the modules.
-FMan internal memory is split into the following parts:
-	Packet buffering (Tx/Rx FIFOs)
-	Frames internal context
-
-PROPERTIES
-
-- compatible
-		Usage: required
-		Value type: <stringlist>
-		Definition: Must include "fsl,fman-muram"
-
-- ranges
-		Usage: required
-		Value type: <prop-encoded-array>
-		Definition: A standard property.
-		Specifies the multi-user memory offset and the size within
-		the FMan.
-
-EXAMPLE
-
-muram@0 {
-	compatible = "fsl,fman-muram";
-	ranges = <0 0x000000 0x28000>;
-};
-
-=============================================================================
-FMan Port Node
-
-DESCRIPTION
-
-The Frame Manager (FMan) supports several types of hardware ports:
-	Ethernet receiver (RX)
-	Ethernet transmitter (TX)
-	Offline/Host command (O/H)
-
-PROPERTIES
-
-- compatible
-		Usage: required
-		Value type: <stringlist>
-		Definition: A standard property.
-		Must include one of the following:
-			- "fsl,fman-v2-port-oh" for FManV2 OH ports
-			- "fsl,fman-v2-port-rx" for FManV2 RX ports
-			- "fsl,fman-v2-port-tx" for FManV2 TX ports
-			- "fsl,fman-v3-port-oh" for FManV3 OH ports
-			- "fsl,fman-v3-port-rx" for FManV3 RX ports
-			- "fsl,fman-v3-port-tx" for FManV3 TX ports
-
-- cell-index
-		Usage: required
-		Value type: <u32>
-		Definition: Specifies the hardware port id.
-		Each hardware port on the FMan has its own hardware PortID.
-		Super set of all hardware Port IDs available at FMan Reference
-		Manual under "FMan Hardware Ports in Freescale Devices" table.
-
-		Each hardware port is assigned a 4KB, port-specific page in
-		the FMan hardware port memory region (which is part of the
-		FMan memory map). The first 4 KB in the FMan hardware ports
-		memory region is used for what are called common registers.
-		The subsequent 63 4KB pages are allocated to the hardware
-		ports.
-		The page of a specific port is determined by the cell-index.
-
-- reg
-		Usage: required
-		Value type: <prop-encoded-array>
-		Definition: There is one reg region describing the port
-		configuration registers.
-
-- fsl,fman-10g-port
-		Usage: optional
-		Value type: boolean
-		Definition: The default port rate is 1G.
-		If this property exists, the port is s 10G port.
-
-- fsl,fman-best-effort-port
-		Usage: optional
-		Value type: boolean
-		Definition: Can be defined only if 10G-support is set.
-		This property marks a best-effort 10G port (10G port that
-		may not be capable of line rate).
-
-EXAMPLE
-
-port@a8000 {
-	cell-index = <0x28>;
-	compatible = "fsl,fman-v2-port-tx";
-	reg = <0xa8000 0x1000>;
-};
-
-port@88000 {
-	cell-index = <0x8>;
-	compatible = "fsl,fman-v2-port-rx";
-	reg = <0x88000 0x1000>;
-};
-
-port@81000 {
-	cell-index = <0x1>;
-	compatible = "fsl,fman-v2-port-oh";
-	reg = <0x81000 0x1000>;
-};
-
-=============================================================================
-FMan dTSEC/XGEC/mEMAC Node
-
-Refer to Documentation/devicetree/bindings/net/fsl,fman-dtsec.yaml
-
-============================================================================
-FMan IEEE 1588 Node
-
-Refer to Documentation/devicetree/bindings/ptp/fsl,ptp.yaml
-
-=============================================================================
-FMan MDIO Node
-
-DESCRIPTION
-
-The MDIO is a bus to which the PHY devices are connected.
-
-PROPERTIES
-
-- compatible
-		Usage: required
-		Value type: <stringlist>
-		Definition: A standard property.
-		Must include "fsl,fman-mdio" for 1 Gb/s MDIO from FMan v2.
-		Must include "fsl,fman-xmdio" for 10 Gb/s MDIO from FMan v2.
-		Must include "fsl,fman-memac-mdio" for 1/10 Gb/s MDIO from
-		FMan v3.
-
-- reg
-		Usage: required
-		Value type: <prop-encoded-array>
-		Definition: A standard property.
-
-- clocks
-		Usage: optional
-		Value type: <phandle>
-		Definition: A reference to the input clock of the controller
-		from which the MDC frequency is derived.
-
-- clock-frequency
-		Usage: optional
-		Value type: <u32>
-		Definition: Specifies the external MDC frequency, in Hertz, to
-		be used. Requires that the input clock is specified in the
-		"clocks" property. See also: mdio.yaml.
-
-- suppress-preamble
-		Usage: optional
-		Value type: <boolean>
-		Definition: Disable generation of preamble bits. See also:
-		mdio.yaml.
-
-- interrupts
-		Usage: required for external MDIO
-		Value type: <prop-encoded-array>
-		Definition: Event interrupt of external MDIO controller.
-
-- fsl,fman-internal-mdio
-		Usage: required for internal MDIO
-		Value type: boolean
-		Definition: Fman has internal MDIO for internal PCS(Physical
-		Coding Sublayer) PHYs and external MDIO for external PHYs.
-		The settings and programming routines for internal/external
-		MDIO are different. Must be included for internal MDIO.
-
-- fsl,erratum-a009885
-		Usage: optional
-		Value type: <boolean>
-		Definition: Indicates the presence of the A009885
-		erratum describing that the contents of MDIO_DATA may
-		become corrupt unless it is read within 16 MDC cycles
-		of MDIO_CFG[BSY] being cleared, when performing an
-		MDIO read operation.
-
-- fsl,erratum-a011043
-		Usage: optional
-		Value type: <boolean>
-		Definition: Indicates the presence of the A011043 erratum
-		describing that the MDIO_CFG[MDIO_RD_ER] bit may be falsely
-		set when reading internal PCS registers. MDIO reads to
-		internal PCS registers may result in having the
-		MDIO_CFG[MDIO_RD_ER] bit set, even when there is no error and
-		read data (MDIO_DATA[MDIO_DATA]) is correct.
-		Software may get false read error when reading internal
-		PCS registers through MDIO. As a workaround, all internal
-		MDIO accesses should ignore the MDIO_CFG[MDIO_RD_ER] bit.
-
-For internal PHY device on internal mdio bus, a PHY node should be created.
-See the definition of the PHY node in booting-without-of.txt for an
-example of how to define a PHY (Internal PHY has no interrupt line).
-- For "fsl,fman-mdio" compatible internal mdio bus, the PHY is TBI PHY.
-- For "fsl,fman-memac-mdio" compatible internal mdio bus, the PHY is PCS PHY.
-  The PCS PHY address should correspond to the value of the appropriate
-  MDEV_PORT.
-
-EXAMPLE
-
-Example for FMan v2 external MDIO:
-
-mdio@f1000 {
-	compatible = "fsl,fman-xmdio";
-	reg = <0xf1000 0x1000>;
-	interrupts = <101 2 0 0>;
-};
-
-Example for FMan v2 internal MDIO:
-
-mdio@e3120 {
-	compatible = "fsl,fman-mdio";
-	reg = <0xe3120 0xee0>;
-	fsl,fman-internal-mdio;
-
-	tbi1: tbi-phy@8 {
-		reg = <0x8>;
-		device_type = "tbi-phy";
-	};
-};
-
-Example for FMan v3 internal MDIO:
-
-mdio@f1000 {
-	compatible = "fsl,fman-memac-mdio";
-	reg = <0xf1000 0x1000>;
-	fsl,fman-internal-mdio;
-
-	pcsphy6: ethernet-phy@0 {
-		reg = <0x0>;
-	};
-};
-
-=============================================================================
-Example
-
-fman@400000 {
-	#address-cells = <1>;
-	#size-cells = <1>;
-	cell-index = <1>;
-	compatible = "fsl,fman"
-	ranges = <0 0x400000 0x100000>;
-	reg = <0x400000 0x100000>;
-	clocks = <&fman_clk>;
-	clock-names = "fmanclk";
-	interrupts = <
-		96 2 0 0
-		16 2 1 1>;
-	fsl,qman-channel-range = <0x40 0xc>;
-
-	muram@0 {
-		compatible = "fsl,fman-muram";
-		reg = <0x0 0x28000>;
-	};
-
-	port@81000 {
-		cell-index = <1>;
-		compatible = "fsl,fman-v2-port-oh";
-		reg = <0x81000 0x1000>;
-	};
-
-	port@82000 {
-		cell-index = <2>;
-		compatible = "fsl,fman-v2-port-oh";
-		reg = <0x82000 0x1000>;
-	};
-
-	port@83000 {
-		cell-index = <3>;
-		compatible = "fsl,fman-v2-port-oh";
-		reg = <0x83000 0x1000>;
-	};
-
-	port@84000 {
-		cell-index = <4>;
-		compatible = "fsl,fman-v2-port-oh";
-		reg = <0x84000 0x1000>;
-	};
-
-	port@85000 {
-		cell-index = <5>;
-		compatible = "fsl,fman-v2-port-oh";
-		reg = <0x85000 0x1000>;
-	};
-
-	port@86000 {
-		cell-index = <6>;
-		compatible = "fsl,fman-v2-port-oh";
-		reg = <0x86000 0x1000>;
-	};
-
-	fman1_rx_0x8: port@88000 {
-		cell-index = <0x8>;
-		compatible = "fsl,fman-v2-port-rx";
-		reg = <0x88000 0x1000>;
-	};
-
-	fman1_rx_0x9: port@89000 {
-		cell-index = <0x9>;
-		compatible = "fsl,fman-v2-port-rx";
-		reg = <0x89000 0x1000>;
-	};
-
-	fman1_rx_0xa: port@8a000 {
-		cell-index = <0xa>;
-		compatible = "fsl,fman-v2-port-rx";
-		reg = <0x8a000 0x1000>;
-	};
-
-	fman1_rx_0xb: port@8b000 {
-		cell-index = <0xb>;
-		compatible = "fsl,fman-v2-port-rx";
-		reg = <0x8b000 0x1000>;
-	};
-
-	fman1_rx_0xc: port@8c000 {
-		cell-index = <0xc>;
-		compatible = "fsl,fman-v2-port-rx";
-		reg = <0x8c000 0x1000>;
-	};
-
-	fman1_rx_0x10: port@90000 {
-		cell-index = <0x10>;
-		compatible = "fsl,fman-v2-port-rx";
-		reg = <0x90000 0x1000>;
-	};
-
-	fman1_tx_0x28: port@a8000 {
-		cell-index = <0x28>;
-		compatible = "fsl,fman-v2-port-tx";
-		reg = <0xa8000 0x1000>;
-	};
-
-	fman1_tx_0x29: port@a9000 {
-		cell-index = <0x29>;
-		compatible = "fsl,fman-v2-port-tx";
-		reg = <0xa9000 0x1000>;
-	};
-
-	fman1_tx_0x2a: port@aa000 {
-		cell-index = <0x2a>;
-		compatible = "fsl,fman-v2-port-tx";
-		reg = <0xaa000 0x1000>;
-	};
-
-	fman1_tx_0x2b: port@ab000 {
-		cell-index = <0x2b>;
-		compatible = "fsl,fman-v2-port-tx";
-		reg = <0xab000 0x1000>;
-	};
-
-	fman1_tx_0x2c: port@ac0000 {
-		cell-index = <0x2c>;
-		compatible = "fsl,fman-v2-port-tx";
-		reg = <0xac000 0x1000>;
-	};
-
-	fman1_tx_0x30: port@b0000 {
-		cell-index = <0x30>;
-		compatible = "fsl,fman-v2-port-tx";
-		reg = <0xb0000 0x1000>;
-	};
-
-	ethernet@e0000 {
-		compatible = "fsl,fman-dtsec";
-		cell-index = <0>;
-		reg = <0xe0000 0x1000>;
-		fsl,fman-ports = <&fman1_rx_0x8 &fman1_tx_0x28>;
-		tbi-handle = <&tbi5>;
-	};
-
-	ethernet@e2000 {
-		compatible = "fsl,fman-dtsec";
-		cell-index = <1>;
-		reg = <0xe2000 0x1000>;
-		fsl,fman-ports = <&fman1_rx_0x9 &fman1_tx_0x29>;
-		tbi-handle = <&tbi6>;
-	};
-
-	ethernet@e4000 {
-		compatible = "fsl,fman-dtsec";
-		cell-index = <2>;
-		reg = <0xe4000 0x1000>;
-		fsl,fman-ports = <&fman1_rx_0xa &fman1_tx_0x2a>;
-		tbi-handle = <&tbi7>;
-	};
-
-	ethernet@e6000 {
-		compatible = "fsl,fman-dtsec";
-		cell-index = <3>;
-		reg = <0xe6000 0x1000>;
-		fsl,fman-ports = <&fman1_rx_0xb &fman1_tx_0x2b>;
-		tbi-handle = <&tbi8>;
-	};
-
-	ethernet@e8000 {
-		compatible = "fsl,fman-dtsec";
-		cell-index = <4>;
-		reg = <0xf0000 0x1000>;
-		fsl,fman-ports = <&fman1_rx_0xc &fman1_tx_0x2c>;
-		tbi-handle = <&tbi9>;
-
-	ethernet@f0000 {
-		cell-index = <8>;
-		compatible = "fsl,fman-xgec";
-		reg = <0xf0000 0x1000>;
-		fsl,fman-ports = <&fman1_rx_0x10 &fman1_tx_0x30>;
-	};
-
-	ptp-timer@fe000 {
-		compatible = "fsl,fman-ptp-timer";
-		reg = <0xfe000 0x1000>;
-	};
-
-	mdio@f1000 {
-		compatible = "fsl,fman-xmdio";
-		reg = <0xf1000 0x1000>;
-		interrupts = <101 2 0 0>;
-	};
-};
diff --git a/MAINTAINERS b/MAINTAINERS
index 9d2fbf651848..1e8a95217b36 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -8809,7 +8809,7 @@ M:	Madalin Bucur <madalin.bucur@nxp.com>
 R:	Sean Anderson <sean.anderson@seco.com>
 L:	netdev@vger.kernel.org
 S:	Maintained
-F:	Documentation/devicetree/bindings/net/fsl-fman.txt
+F:	Documentation/devicetree/bindings/net/fsl-fman.yaml
 F:	drivers/net/ethernet/freescale/fman
 
 FREESCALE QORIQ PTP CLOCK DRIVER
-- 
cgit v1.2.3-58-ga151


From 01793ed86b5d7df1e956520b5474940743eb7ed8 Mon Sep 17 00:00:00 2001
From: Leon Hwang <hffilwlqm@gmail.com>
Date: Mon, 10 Jun 2024 20:42:23 +0800
Subject: bpf, verifier: Correct tail_call_reachable for bpf prog

It's confusing to inspect 'prog->aux->tail_call_reachable' with drgn[0],
when bpf prog has tail call but 'tail_call_reachable' is false.

This patch corrects 'tail_call_reachable' when bpf prog has tail call.

Signed-off-by: Leon Hwang <hffilwlqm@gmail.com>
Link: https://lore.kernel.org/r/20240610124224.34673-2-hffilwlqm@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index dcbbf5f64c5d..ffe98a788c33 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -2982,8 +2982,10 @@ static int check_subprogs(struct bpf_verifier_env *env)
 
 		if (code == (BPF_JMP | BPF_CALL) &&
 		    insn[i].src_reg == 0 &&
-		    insn[i].imm == BPF_FUNC_tail_call)
+		    insn[i].imm == BPF_FUNC_tail_call) {
 			subprog[cur_subprog].has_tail_call = true;
+			subprog[cur_subprog].tail_call_reachable = true;
+		}
 		if (BPF_CLASS(code) == BPF_LD &&
 		    (BPF_MODE(code) == BPF_ABS || BPF_MODE(code) == BPF_IND))
 			subprog[cur_subprog].has_ld_abs = true;
-- 
cgit v1.2.3-58-ga151


From f663a03c8e35c5156bad073a4a8f5e673d656e3f Mon Sep 17 00:00:00 2001
From: Leon Hwang <hffilwlqm@gmail.com>
Date: Mon, 10 Jun 2024 20:42:24 +0800
Subject: bpf, x64: Remove tail call detection

As 'prog->aux->tail_call_reachable' is correct for tail call present,
it's unnecessary to detect tail call in x86 jit.

Therefore, let's remove it.

Signed-off-by: Leon Hwang <hffilwlqm@gmail.com>
Link: https://lore.kernel.org/r/20240610124224.34673-3-hffilwlqm@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 arch/x86/net/bpf_jit_comp.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index 5159c7a22922..7c130001fbfe 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -1234,13 +1234,11 @@ bool ex_handler_bpf(const struct exception_table_entry *x, struct pt_regs *regs)
 }
 
 static void detect_reg_usage(struct bpf_insn *insn, int insn_cnt,
-			     bool *regs_used, bool *tail_call_seen)
+			     bool *regs_used)
 {
 	int i;
 
 	for (i = 1; i <= insn_cnt; i++, insn++) {
-		if (insn->code == (BPF_JMP | BPF_TAIL_CALL))
-			*tail_call_seen = true;
 		if (insn->dst_reg == BPF_REG_6 || insn->src_reg == BPF_REG_6)
 			regs_used[0] = true;
 		if (insn->dst_reg == BPF_REG_7 || insn->src_reg == BPF_REG_7)
@@ -1324,7 +1322,6 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image, u8 *rw_image
 	struct bpf_insn *insn = bpf_prog->insnsi;
 	bool callee_regs_used[4] = {};
 	int insn_cnt = bpf_prog->len;
-	bool tail_call_seen = false;
 	bool seen_exit = false;
 	u8 temp[BPF_MAX_INSN_SIZE + BPF_INSN_SAFETY];
 	u64 arena_vm_start, user_vm_start;
@@ -1336,11 +1333,7 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image, u8 *rw_image
 	arena_vm_start = bpf_arena_get_kern_vm_start(bpf_prog->aux->arena);
 	user_vm_start = bpf_arena_get_user_vm_start(bpf_prog->aux->arena);
 
-	detect_reg_usage(insn, insn_cnt, callee_regs_used,
-			 &tail_call_seen);
-
-	/* tail call's presence in current prog implies it is reachable */
-	tail_call_reachable |= tail_call_seen;
+	detect_reg_usage(insn, insn_cnt, callee_regs_used);
 
 	emit_prologue(&prog, bpf_prog->aux->stack_depth,
 		      bpf_prog_was_classic(bpf_prog), tail_call_reachable,
-- 
cgit v1.2.3-58-ga151


From 9919c5c98cb25dbf7e76aadb9beab55a2a25f830 Mon Sep 17 00:00:00 2001
From: Rafael Passos <rafael@rcpassos.me>
Date: Fri, 14 Jun 2024 23:24:08 -0300
Subject: bpf: remove unused parameter in bpf_jit_binary_pack_finalize

Fixes a compiler warning. the bpf_jit_binary_pack_finalize function
was taking an extra bpf_prog parameter that went unused.
This removves it and updates the callers accordingly.

Signed-off-by: Rafael Passos <rafael@rcpassos.me>
Link: https://lore.kernel.org/r/20240615022641.210320-2-rafael@rcpassos.me
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 arch/arm64/net/bpf_jit_comp.c   | 3 +--
 arch/powerpc/net/bpf_jit_comp.c | 4 ++--
 arch/riscv/net/bpf_jit_core.c   | 5 ++---
 arch/x86/net/bpf_jit_comp.c     | 4 ++--
 include/linux/filter.h          | 3 +--
 kernel/bpf/core.c               | 3 +--
 6 files changed, 9 insertions(+), 13 deletions(-)

diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c
index 720336d28856..6edaeafd1499 100644
--- a/arch/arm64/net/bpf_jit_comp.c
+++ b/arch/arm64/net/bpf_jit_comp.c
@@ -1829,8 +1829,7 @@ skip_init_ctx:
 			prog->jited_len = 0;
 			goto out_free_hdr;
 		}
-		if (WARN_ON(bpf_jit_binary_pack_finalize(prog, ro_header,
-							 header))) {
+		if (WARN_ON(bpf_jit_binary_pack_finalize(ro_header, header))) {
 			/* ro_header has been freed */
 			ro_header = NULL;
 			prog = orig_prog;
diff --git a/arch/powerpc/net/bpf_jit_comp.c b/arch/powerpc/net/bpf_jit_comp.c
index 984655419da5..2a36cc2e7e9e 100644
--- a/arch/powerpc/net/bpf_jit_comp.c
+++ b/arch/powerpc/net/bpf_jit_comp.c
@@ -225,7 +225,7 @@ skip_init_ctx:
 	fp->jited_len = proglen + FUNCTION_DESCR_SIZE;
 
 	if (!fp->is_func || extra_pass) {
-		if (bpf_jit_binary_pack_finalize(fp, fhdr, hdr)) {
+		if (bpf_jit_binary_pack_finalize(fhdr, hdr)) {
 			fp = org_fp;
 			goto out_addrs;
 		}
@@ -348,7 +348,7 @@ void bpf_jit_free(struct bpf_prog *fp)
 		 * before freeing it.
 		 */
 		if (jit_data) {
-			bpf_jit_binary_pack_finalize(fp, jit_data->fhdr, jit_data->hdr);
+			bpf_jit_binary_pack_finalize(jit_data->fhdr, jit_data->hdr);
 			kvfree(jit_data->addrs);
 			kfree(jit_data);
 		}
diff --git a/arch/riscv/net/bpf_jit_core.c b/arch/riscv/net/bpf_jit_core.c
index 0a96abdaca65..6de753c667f4 100644
--- a/arch/riscv/net/bpf_jit_core.c
+++ b/arch/riscv/net/bpf_jit_core.c
@@ -178,8 +178,7 @@ skip_init_ctx:
 	prog->jited_len = prog_size - cfi_get_offset();
 
 	if (!prog->is_func || extra_pass) {
-		if (WARN_ON(bpf_jit_binary_pack_finalize(prog, jit_data->ro_header,
-							 jit_data->header))) {
+		if (WARN_ON(bpf_jit_binary_pack_finalize(jit_data->ro_header, jit_data->header))) {
 			/* ro_header has been freed */
 			jit_data->ro_header = NULL;
 			prog = orig_prog;
@@ -258,7 +257,7 @@ void bpf_jit_free(struct bpf_prog *prog)
 		 * before freeing it.
 		 */
 		if (jit_data) {
-			bpf_jit_binary_pack_finalize(prog, jit_data->ro_header, jit_data->header);
+			bpf_jit_binary_pack_finalize(jit_data->ro_header, jit_data->header);
 			kfree(jit_data);
 		}
 		hdr = bpf_jit_binary_pack_hdr(prog);
diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index 7c130001fbfe..d25d81c8ecc0 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -3356,7 +3356,7 @@ out_image:
 			 *
 			 * Both cases are serious bugs and justify WARN_ON.
 			 */
-			if (WARN_ON(bpf_jit_binary_pack_finalize(prog, header, rw_header))) {
+			if (WARN_ON(bpf_jit_binary_pack_finalize(header, rw_header))) {
 				/* header has been freed */
 				header = NULL;
 				goto out_image;
@@ -3435,7 +3435,7 @@ void bpf_jit_free(struct bpf_prog *prog)
 		 * before freeing it.
 		 */
 		if (jit_data) {
-			bpf_jit_binary_pack_finalize(prog, jit_data->header,
+			bpf_jit_binary_pack_finalize(jit_data->header,
 						     jit_data->rw_header);
 			kvfree(jit_data->addrs);
 			kfree(jit_data);
diff --git a/include/linux/filter.h b/include/linux/filter.h
index b02aea291b7e..dd41a93f06b2 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -1129,8 +1129,7 @@ bpf_jit_binary_pack_alloc(unsigned int proglen, u8 **ro_image,
 			  struct bpf_binary_header **rw_hdr,
 			  u8 **rw_image,
 			  bpf_jit_fill_hole_t bpf_fill_ill_insns);
-int bpf_jit_binary_pack_finalize(struct bpf_prog *prog,
-				 struct bpf_binary_header *ro_header,
+int bpf_jit_binary_pack_finalize(struct bpf_binary_header *ro_header,
 				 struct bpf_binary_header *rw_header);
 void bpf_jit_binary_pack_free(struct bpf_binary_header *ro_header,
 			      struct bpf_binary_header *rw_header);
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 1a6c3faa6e4a..f6951c33790d 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -1174,8 +1174,7 @@ bpf_jit_binary_pack_alloc(unsigned int proglen, u8 **image_ptr,
 }
 
 /* Copy JITed text from rw_header to its final location, the ro_header. */
-int bpf_jit_binary_pack_finalize(struct bpf_prog *prog,
-				 struct bpf_binary_header *ro_header,
+int bpf_jit_binary_pack_finalize(struct bpf_binary_header *ro_header,
 				 struct bpf_binary_header *rw_header)
 {
 	void *ptr;
-- 
cgit v1.2.3-58-ga151


From ab224b9ef7c4eaa752752455ea79bd7022209d5d Mon Sep 17 00:00:00 2001
From: Rafael Passos <rafael@rcpassos.me>
Date: Fri, 14 Jun 2024 23:24:09 -0300
Subject: bpf: remove unused parameter in __bpf_free_used_btfs

Fixes a compiler warning. The __bpf_free_used_btfs function
was taking an extra unused struct bpf_prog_aux *aux param

Signed-off-by: Rafael Passos <rafael@rcpassos.me>
Link: https://lore.kernel.org/r/20240615022641.210320-3-rafael@rcpassos.me
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 include/linux/bpf.h   | 3 +--
 kernel/bpf/core.c     | 5 ++---
 kernel/bpf/verifier.c | 3 +--
 3 files changed, 4 insertions(+), 7 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index f636b4998bf7..960780ef04e1 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -2933,8 +2933,7 @@ bpf_probe_read_kernel_common(void *dst, u32 size, const void *unsafe_ptr)
 	return ret;
 }
 
-void __bpf_free_used_btfs(struct bpf_prog_aux *aux,
-			  struct btf_mod_pair *used_btfs, u32 len);
+void __bpf_free_used_btfs(struct btf_mod_pair *used_btfs, u32 len);
 
 static inline struct bpf_prog *bpf_prog_get_type(u32 ufd,
 						 enum bpf_prog_type type)
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index f6951c33790d..ae2e1eeda0d4 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -2742,8 +2742,7 @@ static void bpf_free_used_maps(struct bpf_prog_aux *aux)
 	kfree(aux->used_maps);
 }
 
-void __bpf_free_used_btfs(struct bpf_prog_aux *aux,
-			  struct btf_mod_pair *used_btfs, u32 len)
+void __bpf_free_used_btfs(struct btf_mod_pair *used_btfs, u32 len)
 {
 #ifdef CONFIG_BPF_SYSCALL
 	struct btf_mod_pair *btf_mod;
@@ -2760,7 +2759,7 @@ void __bpf_free_used_btfs(struct bpf_prog_aux *aux,
 
 static void bpf_free_used_btfs(struct bpf_prog_aux *aux)
 {
-	__bpf_free_used_btfs(aux, aux->used_btfs, aux->used_btf_cnt);
+	__bpf_free_used_btfs(aux->used_btfs, aux->used_btf_cnt);
 	kfree(aux->used_btfs);
 }
 
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index ffe98a788c33..3f6be4923655 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -18694,8 +18694,7 @@ static void release_maps(struct bpf_verifier_env *env)
 /* drop refcnt of maps used by the rejected program */
 static void release_btfs(struct bpf_verifier_env *env)
 {
-	__bpf_free_used_btfs(env->prog->aux, env->used_btfs,
-			     env->used_btf_cnt);
+	__bpf_free_used_btfs(env->used_btfs, env->used_btf_cnt);
 }
 
 /* convert pseudo BPF_LD_IMM64 into generic BPF_LD_IMM64 */
-- 
cgit v1.2.3-58-ga151


From 21ab4980e02d495174bc64c00ceb4d3cf87fadb1 Mon Sep 17 00:00:00 2001
From: Rafael Passos <rafael@rcpassos.me>
Date: Fri, 14 Jun 2024 23:24:10 -0300
Subject: bpf: remove redeclaration of new_n in bpf_verifier_vlog

This new_n is defined in the start of this function.
Its value is overwritten by `new_n = min(n, log->len_total);`
a couple lines before my change,
rendering the shadow declaration unnecessary.

Signed-off-by: Rafael Passos <rafael@rcpassos.me>
Link: https://lore.kernel.org/r/20240615022641.210320-4-rafael@rcpassos.me
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/log.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/log.c b/kernel/bpf/log.c
index 3f4ae92e549f..5aebfc3051e3 100644
--- a/kernel/bpf/log.c
+++ b/kernel/bpf/log.c
@@ -91,7 +91,7 @@ void bpf_verifier_vlog(struct bpf_verifier_log *log, const char *fmt,
 			goto fail;
 	} else {
 		u64 new_end, new_start;
-		u32 buf_start, buf_end, new_n;
+		u32 buf_start, buf_end;
 
 		new_end = log->end_pos + n;
 		if (new_end - log->start_pos >= log->len_total)
-- 
cgit v1.2.3-58-ga151


From 34ad6ec972525b903d4680202d7b8360f71d0d89 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Fri, 21 Jun 2024 10:15:58 +0800
Subject: selftests/bpf: Drop type from network_helper_opts

The opts.{type, noconnect} is at least a bit non intuitive or unnecessary.
The only use case now is in test_bpf_ip_check_defrag_ok which ends up
bypassing most (or at least some) of the connect_to_fd_opts() logic. It's
much better that test should have its own connect_to_fd_opts() instead.

This patch adds a new "type" parameter for connect_to_fd_opts(), then
opts->type and getsockopt(SO_TYPE) can be replaced by "type" parameter in
it.

In connect_to_fd(), use getsockopt(SO_TYPE) to get "type" value and pass
it to connect_to_fd_opts().

In bpf_tcp_ca.c and cgroup_v1v2.c, "SOCK_STREAM" types are passed to
connect_to_fd_opts(), and in ip_check_defrag.c, different types "SOCK_RAW"
and "SOCK_DGRAM" are passed to it.

With these changes, the strcut member "type" of network_helper_opts can be
dropped now.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/cfd20b5ad4085c1d1af5e79df3b09013a407199f.1718932493.git.tanggeliang@kylinos.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/network_helpers.c      | 45 ++++++++++------------
 tools/testing/selftests/bpf/network_helpers.h      |  3 +-
 .../testing/selftests/bpf/prog_tests/bpf_tcp_ca.c  |  2 +-
 .../testing/selftests/bpf/prog_tests/cgroup_v1v2.c |  4 +-
 .../selftests/bpf/prog_tests/ip_check_defrag.c     |  5 +--
 5 files changed, 26 insertions(+), 33 deletions(-)

diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index e20caef06aae..c0646d5a4283 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -303,36 +303,16 @@ error_close:
 	return -1;
 }
 
-int connect_to_fd_opts(int server_fd, const struct network_helper_opts *opts)
+int connect_to_fd_opts(int server_fd, int type, const struct network_helper_opts *opts)
 {
 	struct sockaddr_storage addr;
 	struct sockaddr_in *addr_in;
-	socklen_t addrlen, optlen;
-	int fd, type, protocol;
+	socklen_t addrlen;
+	int fd;
 
 	if (!opts)
 		opts = &default_opts;
 
-	optlen = sizeof(type);
-
-	if (opts->type) {
-		type = opts->type;
-	} else {
-		if (getsockopt(server_fd, SOL_SOCKET, SO_TYPE, &type, &optlen)) {
-			log_err("getsockopt(SOL_TYPE)");
-			return -1;
-		}
-	}
-
-	if (opts->proto) {
-		protocol = opts->proto;
-	} else {
-		if (getsockopt(server_fd, SOL_SOCKET, SO_PROTOCOL, &protocol, &optlen)) {
-			log_err("getsockopt(SOL_PROTOCOL)");
-			return -1;
-		}
-	}
-
 	addrlen = sizeof(addr);
 	if (getsockname(server_fd, (struct sockaddr *)&addr, &addrlen)) {
 		log_err("Failed to get server addr");
@@ -340,7 +320,7 @@ int connect_to_fd_opts(int server_fd, const struct network_helper_opts *opts)
 	}
 
 	addr_in = (struct sockaddr_in *)&addr;
-	fd = socket(addr_in->sin_family, type, protocol);
+	fd = socket(addr_in->sin_family, type, opts->proto);
 	if (fd < 0) {
 		log_err("Failed to create client socket");
 		return -1;
@@ -369,8 +349,23 @@ int connect_to_fd(int server_fd, int timeout_ms)
 	struct network_helper_opts opts = {
 		.timeout_ms = timeout_ms,
 	};
+	int type, protocol;
+	socklen_t optlen;
+
+	optlen = sizeof(type);
+	if (getsockopt(server_fd, SOL_SOCKET, SO_TYPE, &type, &optlen)) {
+		log_err("getsockopt(SOL_TYPE)");
+		return -1;
+	}
+
+	optlen = sizeof(protocol);
+	if (getsockopt(server_fd, SOL_SOCKET, SO_PROTOCOL, &protocol, &optlen)) {
+		log_err("getsockopt(SOL_PROTOCOL)");
+		return -1;
+	}
+	opts.proto = protocol;
 
-	return connect_to_fd_opts(server_fd, &opts);
+	return connect_to_fd_opts(server_fd, type, &opts);
 }
 
 int connect_fd_to_fd(int client_fd, int server_fd, int timeout_ms)
diff --git a/tools/testing/selftests/bpf/network_helpers.h b/tools/testing/selftests/bpf/network_helpers.h
index 11eea8e2e4f1..c92bed35dfe2 100644
--- a/tools/testing/selftests/bpf/network_helpers.h
+++ b/tools/testing/selftests/bpf/network_helpers.h
@@ -25,7 +25,6 @@ struct network_helper_opts {
 	int timeout_ms;
 	bool must_fail;
 	bool noconnect;
-	int type;
 	int proto;
 	int (*post_socket_cb)(int fd, void *opts);
 	void *cb_opts;
@@ -61,7 +60,7 @@ void free_fds(int *fds, unsigned int nr_close_fds);
 int connect_to_addr(int type, const struct sockaddr_storage *addr, socklen_t len,
 		    const struct network_helper_opts *opts);
 int connect_to_fd(int server_fd, int timeout_ms);
-int connect_to_fd_opts(int server_fd, const struct network_helper_opts *opts);
+int connect_to_fd_opts(int server_fd, int type, const struct network_helper_opts *opts);
 int connect_fd_to_fd(int client_fd, int server_fd, int timeout_ms);
 int fastopen_connect(int server_fd, const char *data, unsigned int data_len,
 		     int timeout_ms);
diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index 67358adf5db3..164f237b24dd 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -49,7 +49,7 @@ static bool start_test(char *addr_str,
 		goto err;
 
 	/* connect to server */
-	*cli_fd = connect_to_fd_opts(*srv_fd, cli_opts);
+	*cli_fd = connect_to_fd_opts(*srv_fd, SOCK_STREAM, cli_opts);
 	if (!ASSERT_NEQ(*cli_fd, -1, "connect_to_fd_opts"))
 		goto err;
 
diff --git a/tools/testing/selftests/bpf/prog_tests/cgroup_v1v2.c b/tools/testing/selftests/bpf/prog_tests/cgroup_v1v2.c
index addf720428f7..9709c8db7275 100644
--- a/tools/testing/selftests/bpf/prog_tests/cgroup_v1v2.c
+++ b/tools/testing/selftests/bpf/prog_tests/cgroup_v1v2.c
@@ -32,7 +32,7 @@ static int run_test(int cgroup_fd, int server_fd, bool classid)
 		goto out;
 	}
 
-	fd = connect_to_fd_opts(server_fd, &opts);
+	fd = connect_to_fd_opts(server_fd, SOCK_STREAM, &opts);
 	if (fd < 0)
 		err = -1;
 	else
@@ -52,7 +52,7 @@ void test_cgroup_v1v2(void)
 	server_fd = start_server(AF_INET, SOCK_STREAM, NULL, port, 0);
 	if (!ASSERT_GE(server_fd, 0, "server_fd"))
 		return;
-	client_fd = connect_to_fd_opts(server_fd, &opts);
+	client_fd = connect_to_fd_opts(server_fd, SOCK_STREAM, &opts);
 	if (!ASSERT_GE(client_fd, 0, "client_fd")) {
 		close(server_fd);
 		return;
diff --git a/tools/testing/selftests/bpf/prog_tests/ip_check_defrag.c b/tools/testing/selftests/bpf/prog_tests/ip_check_defrag.c
index 284764e7179f..1607a05bf2c2 100644
--- a/tools/testing/selftests/bpf/prog_tests/ip_check_defrag.c
+++ b/tools/testing/selftests/bpf/prog_tests/ip_check_defrag.c
@@ -164,7 +164,6 @@ void test_bpf_ip_check_defrag_ok(bool ipv6)
 	};
 	struct network_helper_opts tx_ops = {
 		.timeout_ms = 1000,
-		.type = SOCK_RAW,
 		.proto = IPPROTO_RAW,
 		.noconnect = true,
 	};
@@ -201,7 +200,7 @@ void test_bpf_ip_check_defrag_ok(bool ipv6)
 	nstoken = open_netns(NS0);
 	if (!ASSERT_OK_PTR(nstoken, "setns ns0"))
 		goto out;
-	client_tx_fd = connect_to_fd_opts(srv_fd, &tx_ops);
+	client_tx_fd = connect_to_fd_opts(srv_fd, SOCK_RAW, &tx_ops);
 	close_netns(nstoken);
 	if (!ASSERT_GE(client_tx_fd, 0, "connect_to_fd_opts"))
 		goto out;
@@ -210,7 +209,7 @@ void test_bpf_ip_check_defrag_ok(bool ipv6)
 	nstoken = open_netns(NS0);
 	if (!ASSERT_OK_PTR(nstoken, "setns ns0"))
 		goto out;
-	client_rx_fd = connect_to_fd_opts(srv_fd, &rx_opts);
+	client_rx_fd = connect_to_fd_opts(srv_fd, SOCK_DGRAM, &rx_opts);
 	close_netns(nstoken);
 	if (!ASSERT_GE(client_rx_fd, 0, "connect_to_fd_opts"))
 		goto out;
-- 
cgit v1.2.3-58-ga151


From 08a5206240d3763e0c6d91a9a4a9bfbb8fc9600c Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Fri, 21 Jun 2024 10:15:59 +0800
Subject: selftests/bpf: Use connect_to_addr in connect_to_fd_opt

This patch moves "post_socket_cb" and "noconnect" into connect_to_addr(),
then connect_to_fd_opts() can be implemented by getsockname() and
connect_to_addr(). This change makes connect_to_* interfaces more unified.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/4569c30533e14c22fae6c05070aad809720551c1.1718932493.git.tanggeliang@kylinos.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/network_helpers.c | 33 ++++++---------------------
 1 file changed, 7 insertions(+), 26 deletions(-)

diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index c0646d5a4283..5d1b4f165def 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -293,9 +293,14 @@ int connect_to_addr(int type, const struct sockaddr_storage *addr, socklen_t add
 	if (settimeo(fd, opts->timeout_ms))
 		goto error_close;
 
-	if (connect_fd_to_addr(fd, addr, addrlen, opts->must_fail))
+	if (opts->post_socket_cb &&
+	    opts->post_socket_cb(fd, opts->cb_opts))
 		goto error_close;
 
+	if (!opts->noconnect)
+		if (connect_fd_to_addr(fd, addr, addrlen, opts->must_fail))
+			goto error_close;
+
 	return fd;
 
 error_close:
@@ -306,9 +311,7 @@ error_close:
 int connect_to_fd_opts(int server_fd, int type, const struct network_helper_opts *opts)
 {
 	struct sockaddr_storage addr;
-	struct sockaddr_in *addr_in;
 	socklen_t addrlen;
-	int fd;
 
 	if (!opts)
 		opts = &default_opts;
@@ -319,29 +322,7 @@ int connect_to_fd_opts(int server_fd, int type, const struct network_helper_opts
 		return -1;
 	}
 
-	addr_in = (struct sockaddr_in *)&addr;
-	fd = socket(addr_in->sin_family, type, opts->proto);
-	if (fd < 0) {
-		log_err("Failed to create client socket");
-		return -1;
-	}
-
-	if (settimeo(fd, opts->timeout_ms))
-		goto error_close;
-
-	if (opts->post_socket_cb &&
-	    opts->post_socket_cb(fd, opts->cb_opts))
-		goto error_close;
-
-	if (!opts->noconnect)
-		if (connect_fd_to_addr(fd, &addr, addrlen, opts->must_fail))
-			goto error_close;
-
-	return fd;
-
-error_close:
-	save_errno_close(fd);
-	return -1;
+	return connect_to_addr(type, &addr, addrlen, opts);
 }
 
 int connect_to_fd(int server_fd, int timeout_ms)
-- 
cgit v1.2.3-58-ga151


From bbca57aa378b43d25af2ec360b3e8bc4185d65cf Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Fri, 21 Jun 2024 10:16:00 +0800
Subject: selftests/bpf: Add client_socket helper

This patch extracts a new helper client_socket() from connect_to_fd_opts()
to create the client socket, but don't connect to the server. Then
connect_to_fd_opts() can be implemented using client_socket() and
connect_fd_to_addr(). This helper can be used in connect_to_addr() too,
and make "noconnect" opts useless.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/4169c554e1cee79223feea49a1adc459d55e1ffe.1718932493.git.tanggeliang@kylinos.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/network_helpers.c | 37 +++++++++++++++++++++------
 tools/testing/selftests/bpf/network_helpers.h |  2 ++
 2 files changed, 31 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index 5d1b4f165def..5f8214e2880d 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -249,6 +249,34 @@ error_close:
 	return -1;
 }
 
+int client_socket(int family, int type,
+		  const struct network_helper_opts *opts)
+{
+	int fd;
+
+	if (!opts)
+		opts = &default_opts;
+
+	fd = socket(family, type, opts->proto);
+	if (fd < 0) {
+		log_err("Failed to create client socket");
+		return -1;
+	}
+
+	if (settimeo(fd, opts->timeout_ms))
+		goto error_close;
+
+	if (opts->post_socket_cb &&
+	    opts->post_socket_cb(fd, opts->cb_opts))
+		goto error_close;
+
+	return fd;
+
+error_close:
+	save_errno_close(fd);
+	return -1;
+}
+
 static int connect_fd_to_addr(int fd,
 			      const struct sockaddr_storage *addr,
 			      socklen_t addrlen, const bool must_fail)
@@ -284,19 +312,12 @@ int connect_to_addr(int type, const struct sockaddr_storage *addr, socklen_t add
 	if (!opts)
 		opts = &default_opts;
 
-	fd = socket(addr->ss_family, type, opts->proto);
+	fd = client_socket(addr->ss_family, type, opts);
 	if (fd < 0) {
 		log_err("Failed to create client socket");
 		return -1;
 	}
 
-	if (settimeo(fd, opts->timeout_ms))
-		goto error_close;
-
-	if (opts->post_socket_cb &&
-	    opts->post_socket_cb(fd, opts->cb_opts))
-		goto error_close;
-
 	if (!opts->noconnect)
 		if (connect_fd_to_addr(fd, addr, addrlen, opts->must_fail))
 			goto error_close;
diff --git a/tools/testing/selftests/bpf/network_helpers.h b/tools/testing/selftests/bpf/network_helpers.h
index c92bed35dfe2..e89eadfb02d6 100644
--- a/tools/testing/selftests/bpf/network_helpers.h
+++ b/tools/testing/selftests/bpf/network_helpers.h
@@ -57,6 +57,8 @@ int *start_reuseport_server(int family, int type, const char *addr_str,
 int start_server_addr(int type, const struct sockaddr_storage *addr, socklen_t len,
 		      const struct network_helper_opts *opts);
 void free_fds(int *fds, unsigned int nr_close_fds);
+int client_socket(int family, int type,
+		  const struct network_helper_opts *opts);
 int connect_to_addr(int type, const struct sockaddr_storage *addr, socklen_t len,
 		    const struct network_helper_opts *opts);
 int connect_to_fd(int server_fd, int timeout_ms);
-- 
cgit v1.2.3-58-ga151


From 7f0d5140a6d69d3e63467a220a2a1e0c9ec1463a Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Fri, 21 Jun 2024 10:16:01 +0800
Subject: selftests/bpf: Drop noconnect from network_helper_opts

In test_bpf_ip_check_defrag_ok(), the new helper client_socket() can be
used to replace connect_to_fd_opts() with "noconnect" opts, and the strcut
member "noconnect" of network_helper_opts can be dropped now, always
connect to server in connect_to_fd_opts().

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/f45760becce51986e4e08283c7df0f933eb0da14.1718932493.git.tanggeliang@kylinos.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/network_helpers.c            |  5 ++---
 tools/testing/selftests/bpf/network_helpers.h            |  1 -
 tools/testing/selftests/bpf/prog_tests/ip_check_defrag.c | 13 ++++++-------
 3 files changed, 8 insertions(+), 11 deletions(-)

diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index 5f8214e2880d..44c2c8fa542a 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -318,9 +318,8 @@ int connect_to_addr(int type, const struct sockaddr_storage *addr, socklen_t add
 		return -1;
 	}
 
-	if (!opts->noconnect)
-		if (connect_fd_to_addr(fd, addr, addrlen, opts->must_fail))
-			goto error_close;
+	if (connect_fd_to_addr(fd, addr, addrlen, opts->must_fail))
+		goto error_close;
 
 	return fd;
 
diff --git a/tools/testing/selftests/bpf/network_helpers.h b/tools/testing/selftests/bpf/network_helpers.h
index e89eadfb02d6..9ea36524b9db 100644
--- a/tools/testing/selftests/bpf/network_helpers.h
+++ b/tools/testing/selftests/bpf/network_helpers.h
@@ -24,7 +24,6 @@ typedef __u16 __sum16;
 struct network_helper_opts {
 	int timeout_ms;
 	bool must_fail;
-	bool noconnect;
 	int proto;
 	int (*post_socket_cb)(int fd, void *opts);
 	void *cb_opts;
diff --git a/tools/testing/selftests/bpf/prog_tests/ip_check_defrag.c b/tools/testing/selftests/bpf/prog_tests/ip_check_defrag.c
index 1607a05bf2c2..4ddb8a5fece8 100644
--- a/tools/testing/selftests/bpf/prog_tests/ip_check_defrag.c
+++ b/tools/testing/selftests/bpf/prog_tests/ip_check_defrag.c
@@ -158,14 +158,13 @@ static int send_frags6(int client)
 
 void test_bpf_ip_check_defrag_ok(bool ipv6)
 {
+	int family = ipv6 ? AF_INET6 : AF_INET;
 	struct network_helper_opts rx_opts = {
 		.timeout_ms = 1000,
-		.noconnect = true,
 	};
 	struct network_helper_opts tx_ops = {
 		.timeout_ms = 1000,
 		.proto = IPPROTO_RAW,
-		.noconnect = true,
 	};
 	struct sockaddr_storage caddr;
 	struct ip_check_defrag *skel;
@@ -191,7 +190,7 @@ void test_bpf_ip_check_defrag_ok(bool ipv6)
 	nstoken = open_netns(NS1);
 	if (!ASSERT_OK_PTR(nstoken, "setns ns1"))
 		goto out;
-	srv_fd = start_server(ipv6 ? AF_INET6 : AF_INET, SOCK_DGRAM, NULL, SERVER_PORT, 0);
+	srv_fd = start_server(family, SOCK_DGRAM, NULL, SERVER_PORT, 0);
 	close_netns(nstoken);
 	if (!ASSERT_GE(srv_fd, 0, "start_server"))
 		goto out;
@@ -200,18 +199,18 @@ void test_bpf_ip_check_defrag_ok(bool ipv6)
 	nstoken = open_netns(NS0);
 	if (!ASSERT_OK_PTR(nstoken, "setns ns0"))
 		goto out;
-	client_tx_fd = connect_to_fd_opts(srv_fd, SOCK_RAW, &tx_ops);
+	client_tx_fd = client_socket(family, SOCK_RAW, &tx_ops);
 	close_netns(nstoken);
-	if (!ASSERT_GE(client_tx_fd, 0, "connect_to_fd_opts"))
+	if (!ASSERT_GE(client_tx_fd, 0, "client_socket"))
 		goto out;
 
 	/* Open rx socket in ns0 */
 	nstoken = open_netns(NS0);
 	if (!ASSERT_OK_PTR(nstoken, "setns ns0"))
 		goto out;
-	client_rx_fd = connect_to_fd_opts(srv_fd, SOCK_DGRAM, &rx_opts);
+	client_rx_fd = client_socket(family, SOCK_DGRAM, &rx_opts);
 	close_netns(nstoken);
-	if (!ASSERT_GE(client_rx_fd, 0, "connect_to_fd_opts"))
+	if (!ASSERT_GE(client_rx_fd, 0, "client_socket"))
 		goto out;
 
 	/* Bind rx socket to a premeditated port */
-- 
cgit v1.2.3-58-ga151


From fb69f71cf585aabb2f59c6d7958bccfaebe64f5d Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Fri, 21 Jun 2024 10:16:02 +0800
Subject: selftests/bpf: Use start_server_str in mptcp

Since start_server_str() is added now, it can be used in mptcp.c in
start_mptcp_server() instead of using helpers make_sockaddr() and
start_server_addr() to simplify the code.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/16fb3e2cd60b64b5470b0e69f1aa233feaf2717c.1718932493.git.tanggeliang@kylinos.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/mptcp.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/mptcp.c b/tools/testing/selftests/bpf/prog_tests/mptcp.c
index 274d2e033e39..d2ca32fa3b21 100644
--- a/tools/testing/selftests/bpf/prog_tests/mptcp.c
+++ b/tools/testing/selftests/bpf/prog_tests/mptcp.c
@@ -89,13 +89,8 @@ static int start_mptcp_server(int family, const char *addr_str, __u16 port,
 		.timeout_ms	= timeout_ms,
 		.proto		= IPPROTO_MPTCP,
 	};
-	struct sockaddr_storage addr;
-	socklen_t addrlen;
 
-	if (make_sockaddr(family, addr_str, port, &addr, &addrlen))
-		return -1;
-
-	return start_server_addr(SOCK_STREAM, &addr, addrlen, &opts);
+	return start_server_str(family, SOCK_STREAM, addr_str, port, &opts);
 }
 
 static int verify_tsk(int map_fd, int client_fd)
-- 
cgit v1.2.3-58-ga151


From 8cab7cdcf5aebec354ede98bca28c08dd9df924c Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Fri, 21 Jun 2024 10:16:03 +0800
Subject: selftests/bpf: Use start_server_str in test_tcp_check_syncookie_user

Since start_server_str() is added now, it can be used in script
test_tcp_check_syncookie_user.c instead of start_server_addr() to
simplify the code.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/5d2f442261d37cff16c1f1b21a2b188508ab67fa.1718932493.git.tanggeliang@kylinos.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../selftests/bpf/test_tcp_check_syncookie_user.c  | 29 +++-------------------
 1 file changed, 3 insertions(+), 26 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_tcp_check_syncookie_user.c b/tools/testing/selftests/bpf/test_tcp_check_syncookie_user.c
index aebc58c24dc5..3844f9b8232a 100644
--- a/tools/testing/selftests/bpf/test_tcp_check_syncookie_user.c
+++ b/tools/testing/selftests/bpf/test_tcp_check_syncookie_user.c
@@ -156,10 +156,6 @@ static int v6only_false(int fd, void *opts)
 int main(int argc, char **argv)
 {
 	struct network_helper_opts opts = { 0 };
-	struct sockaddr_in addr4;
-	struct sockaddr_in6 addr6;
-	struct sockaddr_in addr4dual;
-	struct sockaddr_in6 addr6dual;
 	int server = -1;
 	int server_v6 = -1;
 	int server_dual = -1;
@@ -181,36 +177,17 @@ int main(int argc, char **argv)
 		goto err;
 	}
 
-	memset(&addr4, 0, sizeof(addr4));
-	addr4.sin_family = AF_INET;
-	addr4.sin_addr.s_addr = htonl(INADDR_LOOPBACK);
-	addr4.sin_port = 0;
-	memcpy(&addr4dual, &addr4, sizeof(addr4dual));
-
-	memset(&addr6, 0, sizeof(addr6));
-	addr6.sin6_family = AF_INET6;
-	addr6.sin6_addr = in6addr_loopback;
-	addr6.sin6_port = 0;
-
-	memset(&addr6dual, 0, sizeof(addr6dual));
-	addr6dual.sin6_family = AF_INET6;
-	addr6dual.sin6_addr = in6addr_any;
-	addr6dual.sin6_port = 0;
-
-	server = start_server_addr(SOCK_STREAM, (struct sockaddr_storage *)&addr4,
-				   sizeof(addr4), NULL);
+	server = start_server_str(AF_INET, SOCK_STREAM, "127.0.0.1", 0, NULL);
 	if (server == -1)
 		goto err;
 
 	opts.post_socket_cb = v6only_true;
-	server_v6 = start_server_addr(SOCK_STREAM, (struct sockaddr_storage *)&addr6,
-				      sizeof(addr6), &opts);
+	server_v6 = start_server_str(AF_INET6, SOCK_STREAM, "::1", 0, &opts);
 	if (server_v6 == -1)
 		goto err;
 
 	opts.post_socket_cb = v6only_false;
-	server_dual = start_server_addr(SOCK_STREAM, (struct sockaddr_storage *)&addr6dual,
-					sizeof(addr6dual), &opts);
+	server_dual = start_server_str(AF_INET6, SOCK_STREAM, "::0", 0, &opts);
 	if (server_dual == -1)
 		goto err;
 
-- 
cgit v1.2.3-58-ga151


From 2851d357a485a358119bc7c1f4ad8f5780debeca Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Wed, 12 Jun 2024 16:19:44 +0200
Subject: can: kvaser_usb: Add support for Vining 800

Add support for Vining 800, a branded device based on the hydra
platform.

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240612141946.3352364-2-martin.jocic@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/usb/Kconfig                      | 1 +
 drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/drivers/net/can/usb/Kconfig b/drivers/net/can/usb/Kconfig
index bd58c636d465..d01b46a8071a 100644
--- a/drivers/net/can/usb/Kconfig
+++ b/drivers/net/can/usb/Kconfig
@@ -117,6 +117,7 @@ config CAN_KVASER_USB
 	    - Kvaser U100S
 	    - ATI Memorator Pro 2xHS v2
 	    - ATI USBcan Pro 2xHS v2
+	    - Vining 800
 
 	  If unsure, say N.
 
diff --git a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
index 8faf8a462c05..b828fad123b2 100644
--- a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
+++ b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
@@ -89,6 +89,7 @@
 #define USB_HYBRID_CANLIN_PRODUCT_ID 0x0115
 #define USB_HYBRID_PRO_CANLIN_PRODUCT_ID 0x0116
 #define USB_LEAF_V3_PRODUCT_ID 0x0117
+#define USB_VINING_800_PRODUCT_ID 0x0119
 
 static const struct kvaser_usb_driver_info kvaser_usb_driver_info_hydra = {
 	.quirks = KVASER_USB_QUIRK_HAS_HARDWARE_TIMESTAMP,
@@ -238,6 +239,8 @@ static const struct usb_device_id kvaser_usb_table[] = {
 		.driver_info = (kernel_ulong_t)&kvaser_usb_driver_info_hydra },
 	{ USB_DEVICE(KVASER_VENDOR_ID, USB_LEAF_V3_PRODUCT_ID),
 		.driver_info = (kernel_ulong_t)&kvaser_usb_driver_info_hydra },
+	{ USB_DEVICE(KVASER_VENDOR_ID, USB_VINING_800_PRODUCT_ID),
+		.driver_info = (kernel_ulong_t)&kvaser_usb_driver_info_hydra },
 	{ }
 };
 MODULE_DEVICE_TABLE(usb, kvaser_usb_table);
-- 
cgit v1.2.3-58-ga151


From 96a669a1958ff9d93fc52931e3571c21b13333f2 Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Wed, 12 Jun 2024 16:19:45 +0200
Subject: can: kvaser_usb: Add support for Kvaser USBcan Pro 5xCAN

Add support for Kvaser USBcan Pro 5xCAN, based on the hydra platform.

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240612141946.3352364-3-martin.jocic@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/usb/Kconfig                      | 1 +
 drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/drivers/net/can/usb/Kconfig b/drivers/net/can/usb/Kconfig
index d01b46a8071a..641bbc24464a 100644
--- a/drivers/net/can/usb/Kconfig
+++ b/drivers/net/can/usb/Kconfig
@@ -111,6 +111,7 @@ config CAN_KVASER_USB
 	    - Kvaser USBcan Light 4xHS
 	    - Kvaser USBcan Pro 2xHS v2
 	    - Kvaser USBcan Pro 4xHS
+	    - Kvaser USBcan Pro 5xCAN
 	    - Kvaser USBcan Pro 5xHS
 	    - Kvaser U100
 	    - Kvaser U100P
diff --git a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
index b828fad123b2..1416619c94ef 100644
--- a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
+++ b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
@@ -90,6 +90,7 @@
 #define USB_HYBRID_PRO_CANLIN_PRODUCT_ID 0x0116
 #define USB_LEAF_V3_PRODUCT_ID 0x0117
 #define USB_VINING_800_PRODUCT_ID 0x0119
+#define USB_USBCAN_PRO_5XCAN_PRODUCT_ID 0x011A
 
 static const struct kvaser_usb_driver_info kvaser_usb_driver_info_hydra = {
 	.quirks = KVASER_USB_QUIRK_HAS_HARDWARE_TIMESTAMP,
@@ -241,6 +242,8 @@ static const struct usb_device_id kvaser_usb_table[] = {
 		.driver_info = (kernel_ulong_t)&kvaser_usb_driver_info_hydra },
 	{ USB_DEVICE(KVASER_VENDOR_ID, USB_VINING_800_PRODUCT_ID),
 		.driver_info = (kernel_ulong_t)&kvaser_usb_driver_info_hydra },
+	{ USB_DEVICE(KVASER_VENDOR_ID, USB_USBCAN_PRO_5XCAN_PRODUCT_ID),
+		.driver_info = (kernel_ulong_t)&kvaser_usb_driver_info_hydra },
 	{ }
 };
 MODULE_DEVICE_TABLE(usb, kvaser_usb_table);
-- 
cgit v1.2.3-58-ga151


From 0135c4c6b84c50bf7f13626c5367d24ce419e1ce Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Wed, 12 Jun 2024 16:19:46 +0200
Subject: can: kvaser_usb: Add support for Kvaser Mini PCIe 1xCAN

Add support for Kvaser Mini PCIe 1xCAN, based on the hydra platform.

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240612141946.3352364-4-martin.jocic@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/usb/Kconfig                      | 1 +
 drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/drivers/net/can/usb/Kconfig b/drivers/net/can/usb/Kconfig
index 641bbc24464a..3e1fba12c0c3 100644
--- a/drivers/net/can/usb/Kconfig
+++ b/drivers/net/can/usb/Kconfig
@@ -91,6 +91,7 @@ config CAN_KVASER_USB
 	    - Kvaser Leaf Light R v2
 	    - Kvaser Mini PCI Express HS
 	    - Kvaser Mini PCI Express 2xHS
+	    - Kvaser Mini PCIe 1xCAN
 	    - Kvaser USBcan Light 2xHS
 	    - Kvaser USBcan II HS/HS
 	    - Kvaser USBcan II HS/LS
diff --git a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
index 1416619c94ef..6c70eb1ceb8b 100644
--- a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
+++ b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c
@@ -91,6 +91,7 @@
 #define USB_LEAF_V3_PRODUCT_ID 0x0117
 #define USB_VINING_800_PRODUCT_ID 0x0119
 #define USB_USBCAN_PRO_5XCAN_PRODUCT_ID 0x011A
+#define USB_MINI_PCIE_1XCAN_PRODUCT_ID 0x011B
 
 static const struct kvaser_usb_driver_info kvaser_usb_driver_info_hydra = {
 	.quirks = KVASER_USB_QUIRK_HAS_HARDWARE_TIMESTAMP,
@@ -244,6 +245,8 @@ static const struct usb_device_id kvaser_usb_table[] = {
 		.driver_info = (kernel_ulong_t)&kvaser_usb_driver_info_hydra },
 	{ USB_DEVICE(KVASER_VENDOR_ID, USB_USBCAN_PRO_5XCAN_PRODUCT_ID),
 		.driver_info = (kernel_ulong_t)&kvaser_usb_driver_info_hydra },
+	{ USB_DEVICE(KVASER_VENDOR_ID, USB_MINI_PCIE_1XCAN_PRODUCT_ID),
+		.driver_info = (kernel_ulong_t)&kvaser_usb_driver_info_hydra },
 	{ }
 };
 MODULE_DEVICE_TABLE(usb, kvaser_usb_table);
-- 
cgit v1.2.3-58-ga151


From cdbc9d055fc7540e6a88ec2bdd7fee46e026156c Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Fri, 14 Jun 2024 17:15:18 +0200
Subject: can: kvaser_pciefd: Group #defines together

Increases readability

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240614151524.2718287-2-martin.jocic@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/kvaser_pciefd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c
index 7b5028b67cd5..fa205091aafe 100644
--- a/drivers/net/can/kvaser_pciefd.c
+++ b/drivers/net/can/kvaser_pciefd.c
@@ -29,10 +29,10 @@ MODULE_DESCRIPTION("CAN driver for Kvaser CAN/PCIe devices");
 #define KVASER_PCIEFD_CAN_TX_MAX_COUNT 17U
 #define KVASER_PCIEFD_MAX_CAN_CHANNELS 8UL
 #define KVASER_PCIEFD_DMA_COUNT 2U
-
 #define KVASER_PCIEFD_DMA_SIZE (4U * 1024U)
 
 #define KVASER_PCIEFD_VENDOR 0x1a07
+
 /* Altera based devices */
 #define KVASER_PCIEFD_4HS_DEVICE_ID 0x000d
 #define KVASER_PCIEFD_2HS_V2_DEVICE_ID 0x000e
-- 
cgit v1.2.3-58-ga151


From ac765219c2c4e44f29063724c8d36435a3e61985 Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Fri, 14 Jun 2024 17:15:19 +0200
Subject: can: kvaser_pciefd: Skip redundant NULL pointer check in ISR

This check is already done at the creation of the net devices in
kvaser_pciefd_setup_can_ctrls called from kvaser_pciefd_probe.

If it fails, the driver won't load, so there should be no need to
repeat the check inside the ISR. The number of channels is read
from the FPGA and should be trusted.

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240614151524.2718287-3-martin.jocic@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/kvaser_pciefd.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c
index fa205091aafe..4832a93d34de 100644
--- a/drivers/net/can/kvaser_pciefd.c
+++ b/drivers/net/can/kvaser_pciefd.c
@@ -1701,12 +1701,6 @@ static irqreturn_t kvaser_pciefd_irq_handler(int irq, void *dev)
 		kvaser_pciefd_receive_irq(pcie);
 
 	for (i = 0; i < pcie->nr_channels; i++) {
-		if (!pcie->can[i]) {
-			dev_err(&pcie->pci->dev,
-				"IRQ mask points to unallocated controller\n");
-			break;
-		}
-
 		/* Check that mask matches channel (i) IRQ mask */
 		if (board_irq & irq_mask->kcan_tx[i])
 			kvaser_pciefd_transmit_irq(pcie->can[i]);
-- 
cgit v1.2.3-58-ga151


From 11d186697ceb10b68c6a1fd505635346b1ccd055 Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Fri, 14 Jun 2024 17:15:20 +0200
Subject: can: kvaser_pciefd: Remove unnecessary comment

The code speaks for itself.

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240614151524.2718287-4-martin.jocic@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/kvaser_pciefd.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c
index 4832a93d34de..8c9abc702b24 100644
--- a/drivers/net/can/kvaser_pciefd.c
+++ b/drivers/net/can/kvaser_pciefd.c
@@ -1701,7 +1701,6 @@ static irqreturn_t kvaser_pciefd_irq_handler(int irq, void *dev)
 		kvaser_pciefd_receive_irq(pcie);
 
 	for (i = 0; i < pcie->nr_channels; i++) {
-		/* Check that mask matches channel (i) IRQ mask */
 		if (board_irq & irq_mask->kcan_tx[i])
 			kvaser_pciefd_transmit_irq(pcie->can[i]);
 	}
-- 
cgit v1.2.3-58-ga151


From 0132a05df1e0119286f51bf757759a7f8e6edaa5 Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Fri, 14 Jun 2024 17:15:21 +0200
Subject: can: kvaser_pciefd: Add inline

Make the short function kvaser_pciefd_set_tx_irq inline. This function
is effectively three lines long and therefore inlining it should be
OK according to rule #15 of the Linux kernel coding style.

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240614151524.2718287-5-martin.jocic@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/kvaser_pciefd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c
index 8c9abc702b24..3d40d7b3d64c 100644
--- a/drivers/net/can/kvaser_pciefd.c
+++ b/drivers/net/can/kvaser_pciefd.c
@@ -550,7 +550,7 @@ static void kvaser_pciefd_disable_err_gen(struct kvaser_pciefd_can *can)
 	spin_unlock_irqrestore(&can->lock, irq);
 }
 
-static void kvaser_pciefd_set_tx_irq(struct kvaser_pciefd_can *can)
+static inline void kvaser_pciefd_set_tx_irq(struct kvaser_pciefd_can *can)
 {
 	u32 msk;
 
-- 
cgit v1.2.3-58-ga151


From cebfebefaa012afbe49ad5e4d6bb97d573beb32d Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Fri, 14 Jun 2024 17:15:22 +0200
Subject: can: kvaser_pciefd: Add unlikely

Use unlikely for some unexpected errors.

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240614151524.2718287-6-martin.jocic@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/kvaser_pciefd.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c
index 3d40d7b3d64c..b08084b0a95b 100644
--- a/drivers/net/can/kvaser_pciefd.c
+++ b/drivers/net/can/kvaser_pciefd.c
@@ -1619,7 +1619,7 @@ static int kvaser_pciefd_read_packet(struct kvaser_pciefd *pcie, int *start_pos,
 	/* Position does not point to the end of the package,
 	 * corrupted packet size?
 	 */
-	if ((*start_pos + size) != pos)
+	if (unlikely((*start_pos + size) != pos))
 		return -EIO;
 
 	/* Point to the next packet header, if any */
@@ -1658,10 +1658,10 @@ static void kvaser_pciefd_receive_irq(struct kvaser_pciefd *pcie)
 			  KVASER_PCIEFD_SRB_ADDR(pcie) + KVASER_PCIEFD_SRB_CMD_REG);
 	}
 
-	if (irq & KVASER_PCIEFD_SRB_IRQ_DOF0 ||
-	    irq & KVASER_PCIEFD_SRB_IRQ_DOF1 ||
-	    irq & KVASER_PCIEFD_SRB_IRQ_DUF0 ||
-	    irq & KVASER_PCIEFD_SRB_IRQ_DUF1)
+	if (unlikely(irq & KVASER_PCIEFD_SRB_IRQ_DOF0 ||
+		     irq & KVASER_PCIEFD_SRB_IRQ_DOF1 ||
+		     irq & KVASER_PCIEFD_SRB_IRQ_DUF0 ||
+		     irq & KVASER_PCIEFD_SRB_IRQ_DUF1))
 		dev_err(&pcie->pci->dev, "DMA IRQ error 0x%08X\n", irq);
 
 	iowrite32(irq, KVASER_PCIEFD_SRB_ADDR(pcie) + KVASER_PCIEFD_SRB_IRQ_REG);
-- 
cgit v1.2.3-58-ga151


From cbf88a6ba7bb6ce0d3131b119298f73bd7b18459 Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Fri, 14 Jun 2024 17:15:23 +0200
Subject: can: kvaser_pciefd: Rename board_irq to pci_irq

Rename the variable name board_irq in the ISR to pci_irq to
be more specific and to match the macro by which it is read.

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240614151524.2718287-7-martin.jocic@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/kvaser_pciefd.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c
index b08084b0a95b..8b2c18f2f23b 100644
--- a/drivers/net/can/kvaser_pciefd.c
+++ b/drivers/net/can/kvaser_pciefd.c
@@ -1691,17 +1691,17 @@ static irqreturn_t kvaser_pciefd_irq_handler(int irq, void *dev)
 {
 	struct kvaser_pciefd *pcie = (struct kvaser_pciefd *)dev;
 	const struct kvaser_pciefd_irq_mask *irq_mask = pcie->driver_data->irq_mask;
-	u32 board_irq = ioread32(KVASER_PCIEFD_PCI_IRQ_ADDR(pcie));
+	u32 pci_irq = ioread32(KVASER_PCIEFD_PCI_IRQ_ADDR(pcie));
 	int i;
 
-	if (!(board_irq & irq_mask->all))
+	if (!(pci_irq & irq_mask->all))
 		return IRQ_NONE;
 
-	if (board_irq & irq_mask->kcan_rx0)
+	if (pci_irq & irq_mask->kcan_rx0)
 		kvaser_pciefd_receive_irq(pcie);
 
 	for (i = 0; i < pcie->nr_channels; i++) {
-		if (board_irq & irq_mask->kcan_tx[i])
+		if (pci_irq & irq_mask->kcan_tx[i])
 			kvaser_pciefd_transmit_irq(pcie->can[i]);
 	}
 
-- 
cgit v1.2.3-58-ga151


From 26a1b0fe3f62ba4a853f9d8ba3aa91f27e13f0ff Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Fri, 14 Jun 2024 17:15:24 +0200
Subject: can: kvaser_pciefd: Change name of return code variable

Replace the variable name err used for return codes with the more
generic name ret. An upcoming patch series for adding MSI interrupts
will introduce code which also returns values other than return codes.
Renaming the variable to ret enables using it for both purposes.
This is applied to the whole file to make it consistent.

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240614151524.2718287-8-martin.jocic@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/kvaser_pciefd.c | 56 ++++++++++++++++++++---------------------
 1 file changed, 28 insertions(+), 28 deletions(-)

diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c
index 8b2c18f2f23b..24871c276b31 100644
--- a/drivers/net/can/kvaser_pciefd.c
+++ b/drivers/net/can/kvaser_pciefd.c
@@ -711,17 +711,17 @@ static void kvaser_pciefd_pwm_start(struct kvaser_pciefd_can *can)
 
 static int kvaser_pciefd_open(struct net_device *netdev)
 {
-	int err;
+	int ret;
 	struct kvaser_pciefd_can *can = netdev_priv(netdev);
 
-	err = open_candev(netdev);
-	if (err)
-		return err;
+	ret = open_candev(netdev);
+	if (ret)
+		return ret;
 
-	err = kvaser_pciefd_bus_on(can);
-	if (err) {
+	ret = kvaser_pciefd_bus_on(can);
+	if (ret) {
 		close_candev(netdev);
-		return err;
+		return ret;
 	}
 
 	return 0;
@@ -1032,15 +1032,15 @@ static int kvaser_pciefd_reg_candev(struct kvaser_pciefd *pcie)
 	int i;
 
 	for (i = 0; i < pcie->nr_channels; i++) {
-		int err = register_candev(pcie->can[i]->can.dev);
+		int ret = register_candev(pcie->can[i]->can.dev);
 
-		if (err) {
+		if (ret) {
 			int j;
 
 			/* Unregister all successfully registered devices. */
 			for (j = 0; j < i; j++)
 				unregister_candev(pcie->can[j]->can.dev);
-			return err;
+			return ret;
 		}
 	}
 
@@ -1726,7 +1726,7 @@ static void kvaser_pciefd_teardown_can_ctrls(struct kvaser_pciefd *pcie)
 static int kvaser_pciefd_probe(struct pci_dev *pdev,
 			       const struct pci_device_id *id)
 {
-	int err;
+	int ret;
 	struct kvaser_pciefd *pcie;
 	const struct kvaser_pciefd_irq_mask *irq_mask;
 	void __iomem *irq_en_base;
@@ -1740,37 +1740,37 @@ static int kvaser_pciefd_probe(struct pci_dev *pdev,
 	pcie->driver_data = (const struct kvaser_pciefd_driver_data *)id->driver_data;
 	irq_mask = pcie->driver_data->irq_mask;
 
-	err = pci_enable_device(pdev);
-	if (err)
-		return err;
+	ret = pci_enable_device(pdev);
+	if (ret)
+		return ret;
 
-	err = pci_request_regions(pdev, KVASER_PCIEFD_DRV_NAME);
-	if (err)
+	ret = pci_request_regions(pdev, KVASER_PCIEFD_DRV_NAME);
+	if (ret)
 		goto err_disable_pci;
 
 	pcie->reg_base = pci_iomap(pdev, 0, 0);
 	if (!pcie->reg_base) {
-		err = -ENOMEM;
+		ret = -ENOMEM;
 		goto err_release_regions;
 	}
 
-	err = kvaser_pciefd_setup_board(pcie);
-	if (err)
+	ret = kvaser_pciefd_setup_board(pcie);
+	if (ret)
 		goto err_pci_iounmap;
 
-	err = kvaser_pciefd_setup_dma(pcie);
-	if (err)
+	ret = kvaser_pciefd_setup_dma(pcie);
+	if (ret)
 		goto err_pci_iounmap;
 
 	pci_set_master(pdev);
 
-	err = kvaser_pciefd_setup_can_ctrls(pcie);
-	if (err)
+	ret = kvaser_pciefd_setup_can_ctrls(pcie);
+	if (ret)
 		goto err_teardown_can_ctrls;
 
-	err = request_irq(pcie->pci->irq, kvaser_pciefd_irq_handler,
+	ret = request_irq(pcie->pci->irq, kvaser_pciefd_irq_handler,
 			  IRQF_SHARED, KVASER_PCIEFD_DRV_NAME, pcie);
-	if (err)
+	if (ret)
 		goto err_teardown_can_ctrls;
 
 	iowrite32(KVASER_PCIEFD_SRB_IRQ_DPD0 | KVASER_PCIEFD_SRB_IRQ_DPD1,
@@ -1790,8 +1790,8 @@ static int kvaser_pciefd_probe(struct pci_dev *pdev,
 	iowrite32(KVASER_PCIEFD_SRB_CMD_RDB1,
 		  KVASER_PCIEFD_SRB_ADDR(pcie) + KVASER_PCIEFD_SRB_CMD_REG);
 
-	err = kvaser_pciefd_reg_candev(pcie);
-	if (err)
+	ret = kvaser_pciefd_reg_candev(pcie);
+	if (ret)
 		goto err_free_irq;
 
 	return 0;
@@ -1815,7 +1815,7 @@ err_release_regions:
 err_disable_pci:
 	pci_disable_device(pdev);
 
-	return err;
+	return ret;
 }
 
 static void kvaser_pciefd_remove_all_ctrls(struct kvaser_pciefd *pcie)
-- 
cgit v1.2.3-58-ga151


From 48f827d4f48f5243e37b9240029ce3f456d1f490 Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Thu, 20 Jun 2024 20:13:19 +0200
Subject: can: kvaser_pciefd: Move reset of DMA RX buffers to the end of the
 ISR

A new interrupt is triggered by resetting the DMA RX buffers.
Since MSI interrupts are faster than legacy interrupts, the reset
of the DMA buffers must be moved to the very end of the ISR,
otherwise a new MSI interrupt will be masked by the current one.

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240620181320.235465-2-martin.jocic@kvaser.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/kvaser_pciefd.c | 30 ++++++++++++++++++------------
 1 file changed, 18 insertions(+), 12 deletions(-)

diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c
index 24871c276b31..b4ffd56fdeff 100644
--- a/drivers/net/can/kvaser_pciefd.c
+++ b/drivers/net/can/kvaser_pciefd.c
@@ -1640,23 +1640,15 @@ static int kvaser_pciefd_read_buffer(struct kvaser_pciefd *pcie, int dma_buf)
 	return res;
 }
 
-static void kvaser_pciefd_receive_irq(struct kvaser_pciefd *pcie)
+static u32 kvaser_pciefd_receive_irq(struct kvaser_pciefd *pcie)
 {
 	u32 irq = ioread32(KVASER_PCIEFD_SRB_ADDR(pcie) + KVASER_PCIEFD_SRB_IRQ_REG);
 
-	if (irq & KVASER_PCIEFD_SRB_IRQ_DPD0) {
+	if (irq & KVASER_PCIEFD_SRB_IRQ_DPD0)
 		kvaser_pciefd_read_buffer(pcie, 0);
-		/* Reset DMA buffer 0 */
-		iowrite32(KVASER_PCIEFD_SRB_CMD_RDB0,
-			  KVASER_PCIEFD_SRB_ADDR(pcie) + KVASER_PCIEFD_SRB_CMD_REG);
-	}
 
-	if (irq & KVASER_PCIEFD_SRB_IRQ_DPD1) {
+	if (irq & KVASER_PCIEFD_SRB_IRQ_DPD1)
 		kvaser_pciefd_read_buffer(pcie, 1);
-		/* Reset DMA buffer 1 */
-		iowrite32(KVASER_PCIEFD_SRB_CMD_RDB1,
-			  KVASER_PCIEFD_SRB_ADDR(pcie) + KVASER_PCIEFD_SRB_CMD_REG);
-	}
 
 	if (unlikely(irq & KVASER_PCIEFD_SRB_IRQ_DOF0 ||
 		     irq & KVASER_PCIEFD_SRB_IRQ_DOF1 ||
@@ -1665,6 +1657,7 @@ static void kvaser_pciefd_receive_irq(struct kvaser_pciefd *pcie)
 		dev_err(&pcie->pci->dev, "DMA IRQ error 0x%08X\n", irq);
 
 	iowrite32(irq, KVASER_PCIEFD_SRB_ADDR(pcie) + KVASER_PCIEFD_SRB_IRQ_REG);
+	return irq;
 }
 
 static void kvaser_pciefd_transmit_irq(struct kvaser_pciefd_can *can)
@@ -1692,19 +1685,32 @@ static irqreturn_t kvaser_pciefd_irq_handler(int irq, void *dev)
 	struct kvaser_pciefd *pcie = (struct kvaser_pciefd *)dev;
 	const struct kvaser_pciefd_irq_mask *irq_mask = pcie->driver_data->irq_mask;
 	u32 pci_irq = ioread32(KVASER_PCIEFD_PCI_IRQ_ADDR(pcie));
+	u32 srb_irq = 0;
 	int i;
 
 	if (!(pci_irq & irq_mask->all))
 		return IRQ_NONE;
 
 	if (pci_irq & irq_mask->kcan_rx0)
-		kvaser_pciefd_receive_irq(pcie);
+		srb_irq = kvaser_pciefd_receive_irq(pcie);
 
 	for (i = 0; i < pcie->nr_channels; i++) {
 		if (pci_irq & irq_mask->kcan_tx[i])
 			kvaser_pciefd_transmit_irq(pcie->can[i]);
 	}
 
+	if (srb_irq & KVASER_PCIEFD_SRB_IRQ_DPD0) {
+		/* Reset DMA buffer 0, may trigger new interrupt */
+		iowrite32(KVASER_PCIEFD_SRB_CMD_RDB0,
+			  KVASER_PCIEFD_SRB_ADDR(pcie) + KVASER_PCIEFD_SRB_CMD_REG);
+	}
+
+	if (srb_irq & KVASER_PCIEFD_SRB_IRQ_DPD1) {
+		/* Reset DMA buffer 1, may trigger new interrupt */
+		iowrite32(KVASER_PCIEFD_SRB_CMD_RDB1,
+			  KVASER_PCIEFD_SRB_ADDR(pcie) + KVASER_PCIEFD_SRB_CMD_REG);
+	}
+
 	return IRQ_HANDLED;
 }
 
-- 
cgit v1.2.3-58-ga151


From dd1f05ba2a996804567b33faa1110f7a839c71ee Mon Sep 17 00:00:00 2001
From: Martin Jocic <martin.jocic@kvaser.com>
Date: Thu, 20 Jun 2024 20:13:20 +0200
Subject: can: kvaser_pciefd: Add MSI interrupts

Use MSI interrupts with fallback to INTx interrupts.

Signed-off-by: Martin Jocic <martin.jocic@kvaser.com>
Link: https://lore.kernel.org/all/20240620181320.235465-3-martin.jocic@kvaser.com
[mkl: kvaser_pciefd_probe(): call pci_free_irq_vectors() unconditionally]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/kvaser_pciefd.c | 24 ++++++++++++++++++++----
 1 file changed, 20 insertions(+), 4 deletions(-)

diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c
index b4ffd56fdeff..a60d9efd5f8d 100644
--- a/drivers/net/can/kvaser_pciefd.c
+++ b/drivers/net/can/kvaser_pciefd.c
@@ -1774,11 +1774,24 @@ static int kvaser_pciefd_probe(struct pci_dev *pdev,
 	if (ret)
 		goto err_teardown_can_ctrls;
 
-	ret = request_irq(pcie->pci->irq, kvaser_pciefd_irq_handler,
-			  IRQF_SHARED, KVASER_PCIEFD_DRV_NAME, pcie);
-	if (ret)
+	ret = pci_alloc_irq_vectors(pcie->pci, 1, 1, PCI_IRQ_INTX | PCI_IRQ_MSI);
+	if (ret < 0) {
+		dev_err(&pcie->pci->dev, "Failed to allocate IRQ vectors.\n");
 		goto err_teardown_can_ctrls;
+	}
+
+	ret = pci_irq_vector(pcie->pci, 0);
+	if (ret < 0)
+		goto err_pci_free_irq_vectors;
+
+	pcie->pci->irq = ret;
 
+	ret = request_irq(pcie->pci->irq, kvaser_pciefd_irq_handler,
+			  IRQF_SHARED, KVASER_PCIEFD_DRV_NAME, pcie);
+	if (ret) {
+		dev_err(&pcie->pci->dev, "Failed to request IRQ %d\n", pcie->pci->irq);
+		goto err_pci_free_irq_vectors;
+	}
 	iowrite32(KVASER_PCIEFD_SRB_IRQ_DPD0 | KVASER_PCIEFD_SRB_IRQ_DPD1,
 		  KVASER_PCIEFD_SRB_ADDR(pcie) + KVASER_PCIEFD_SRB_IRQ_REG);
 
@@ -1807,6 +1820,9 @@ err_free_irq:
 	iowrite32(0, irq_en_base);
 	free_irq(pcie->pci->irq, pcie);
 
+err_pci_free_irq_vectors:
+	pci_free_irq_vectors(pcie->pci);
+
 err_teardown_can_ctrls:
 	kvaser_pciefd_teardown_can_ctrls(pcie);
 	iowrite32(0, KVASER_PCIEFD_SRB_ADDR(pcie) + KVASER_PCIEFD_SRB_CTRL_REG);
@@ -1852,7 +1868,7 @@ static void kvaser_pciefd_remove(struct pci_dev *pdev)
 	iowrite32(0, KVASER_PCIEFD_PCI_IEN_ADDR(pcie));
 
 	free_irq(pcie->pci->irq, pcie);
-
+	pci_free_irq_vectors(pcie->pci);
 	pci_iounmap(pdev, pcie->reg_base);
 	pci_release_regions(pdev);
 	pci_disable_device(pdev);
-- 
cgit v1.2.3-58-ga151


From 1562a49d000c37a3110593e072eb4faa2f83538a Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Thu, 6 Jun 2024 16:24:22 +0200
Subject: can: hi311x: simplify with spi_get_device_match_data()

Use spi_get_device_match_data() helper to simplify a bit the driver.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/all/20240606142424.129709-1-krzysztof.kozlowski@linaro.org
[mkl: add intermediate cast to uintptr_t]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/hi311x.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/net/can/spi/hi311x.c b/drivers/net/can/spi/hi311x.c
index e1b8533a602e..148d974ebb21 100644
--- a/drivers/net/can/spi/hi311x.c
+++ b/drivers/net/can/spi/hi311x.c
@@ -830,7 +830,6 @@ static int hi3110_can_probe(struct spi_device *spi)
 	struct device *dev = &spi->dev;
 	struct net_device *net;
 	struct hi3110_priv *priv;
-	const void *match;
 	struct clk *clk;
 	u32 freq;
 	int ret;
@@ -874,11 +873,7 @@ static int hi3110_can_probe(struct spi_device *spi)
 		CAN_CTRLMODE_LISTENONLY |
 		CAN_CTRLMODE_BERR_REPORTING;
 
-	match = device_get_match_data(dev);
-	if (match)
-		priv->model = (enum hi3110_model)(uintptr_t)match;
-	else
-		priv->model = spi_get_device_id(spi)->driver_data;
+	priv->model = (enum hi3110_model)(uintptr_t)spi_get_device_match_data(spi);
 	priv->net = net;
 	priv->clk = clk;
 
-- 
cgit v1.2.3-58-ga151


From d4383d67a25b1972d41423cbeec56f3eecfc45cb Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Thu, 6 Jun 2024 16:24:23 +0200
Subject: can: mcp251x: simplify with spi_get_device_match_data()

Use spi_get_device_match_data() helper to simplify a bit the driver.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/all/20240606142424.129709-2-krzysztof.kozlowski@linaro.org
[mkl: add intermediate cast to uintptr_t]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251x.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/net/can/spi/mcp251x.c b/drivers/net/can/spi/mcp251x.c
index 643974b3f329..3b8736ff0345 100644
--- a/drivers/net/can/spi/mcp251x.c
+++ b/drivers/net/can/spi/mcp251x.c
@@ -1300,7 +1300,6 @@ MODULE_DEVICE_TABLE(spi, mcp251x_id_table);
 
 static int mcp251x_can_probe(struct spi_device *spi)
 {
-	const void *match = device_get_match_data(&spi->dev);
 	struct net_device *net;
 	struct mcp251x_priv *priv;
 	struct clk *clk;
@@ -1338,10 +1337,7 @@ static int mcp251x_can_probe(struct spi_device *spi)
 	priv->can.clock.freq = freq / 2;
 	priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES |
 		CAN_CTRLMODE_LOOPBACK | CAN_CTRLMODE_LISTENONLY;
-	if (match)
-		priv->model = (enum mcp251x_model)(uintptr_t)match;
-	else
-		priv->model = spi_get_device_id(spi)->driver_data;
+	priv->model = (enum mcp251x_model)(uintptr_t)spi_get_device_match_data(spi);
 	priv->net = net;
 	priv->clk = clk;
 
-- 
cgit v1.2.3-58-ga151


From 9cdae370c4ec415c4057f76c612089068b1ff01a Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Thu, 6 Jun 2024 16:24:24 +0200
Subject: can: mcp251xfd: simplify with spi_get_device_match_data()

Use spi_get_device_match_data() helper to simplify a bit the driver.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/all/20240606142424.129709-3-krzysztof.kozlowski@linaro.org
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
index 1d9057dc44f2..4b7e08e8ed56 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
@@ -1989,7 +1989,6 @@ MODULE_DEVICE_TABLE(spi, mcp251xfd_id_table);
 
 static int mcp251xfd_probe(struct spi_device *spi)
 {
-	const void *match;
 	struct net_device *ndev;
 	struct mcp251xfd_priv *priv;
 	struct gpio_desc *rx_int;
@@ -2081,13 +2080,7 @@ static int mcp251xfd_probe(struct spi_device *spi)
 	priv->pll_enable = pll_enable;
 	priv->reg_vdd = reg_vdd;
 	priv->reg_xceiver = reg_xceiver;
-
-	match = device_get_match_data(&spi->dev);
-	if (match)
-		priv->devtype_data = *(struct mcp251xfd_devtype_data *)match;
-	else
-		priv->devtype_data = *(struct mcp251xfd_devtype_data *)
-			spi_get_device_id(spi)->driver_data;
+	priv->devtype_data = *(struct mcp251xfd_devtype_data *)spi_get_device_match_data(spi);
 
 	/* Errata Reference:
 	 * mcp2517fd: DS80000792C 5., mcp2518fd: DS80000789C 4.
-- 
cgit v1.2.3-58-ga151


From cd5a46ce6fa62abedd7740e4bd9f3d82041210ee Mon Sep 17 00:00:00 2001
From: Martin Hundebøll <martin@geanix.com>
Date: Fri, 7 Jun 2024 12:52:08 +0200
Subject: can: m_can: don't enable transceiver when probing
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The m_can driver sets and clears the CCCR.INIT bit during probe (both
when testing the NON-ISO bit, and when configuring the chip). After
clearing the CCCR.INIT bit, the transceiver enters normal mode, where it
affects the CAN bus (i.e. it ACKs frames). This can cause troubles when
the m_can node is only used for monitoring the bus, as one cannot setup
listen-only mode before the device is probed.

Rework the probe flow, so that the CCCR.INIT bit is only cleared when
upping the device. First, the tcan4x5x driver is changed to stay in
standby mode during/after probe. This in turn requires changes when
setting bits in the CCCR register, as its CSR and CSA bits are always
high in standby mode.

Signed-off-by: Martin Hundebøll <martin@geanix.com>
Reviewed-by: Markus Schneider-Pargmann <msp@baylibre.com>
Tested-by: Markus Schneider-Pargmann <msp@baylibre.com>
Link: https://lore.kernel.org/all/20240607105210.155435-1-martin@geanix.com
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/m_can/m_can.c         | 169 +++++++++++++++++++++-------------
 drivers/net/can/m_can/tcan4x5x-core.c |  13 ++-
 2 files changed, 116 insertions(+), 66 deletions(-)

diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c
index 14b231c4d7ec..7f63f866083e 100644
--- a/drivers/net/can/m_can/m_can.c
+++ b/drivers/net/can/m_can/m_can.c
@@ -379,38 +379,72 @@ m_can_txe_fifo_read(struct m_can_classdev *cdev, u32 fgi, u32 offset, u32 *val)
 	return cdev->ops->read_fifo(cdev, addr_offset, val, 1);
 }
 
-static void m_can_config_endisable(struct m_can_classdev *cdev, bool enable)
-{
-	u32 cccr = m_can_read(cdev, M_CAN_CCCR);
-	u32 timeout = 10;
-	u32 val = 0;
-
-	/* Clear the Clock stop request if it was set */
-	if (cccr & CCCR_CSR)
-		cccr &= ~CCCR_CSR;
-
-	if (enable) {
-		/* enable m_can configuration */
-		m_can_write(cdev, M_CAN_CCCR, cccr | CCCR_INIT);
-		udelay(5);
-		/* CCCR.CCE can only be set/reset while CCCR.INIT = '1' */
-		m_can_write(cdev, M_CAN_CCCR, cccr | CCCR_INIT | CCCR_CCE);
-	} else {
-		m_can_write(cdev, M_CAN_CCCR, cccr & ~(CCCR_INIT | CCCR_CCE));
+static int m_can_cccr_update_bits(struct m_can_classdev *cdev, u32 mask, u32 val)
+{
+	u32 val_before = m_can_read(cdev, M_CAN_CCCR);
+	u32 val_after = (val_before & ~mask) | val;
+	size_t tries = 10;
+
+	if (!(mask & CCCR_INIT) && !(val_before & CCCR_INIT)) {
+		dev_err(cdev->dev,
+			"refusing to configure device when in normal mode\n");
+		return -EBUSY;
 	}
 
-	/* there's a delay for module initialization */
-	if (enable)
-		val = CCCR_INIT | CCCR_CCE;
+	/* The chip should be in standby mode when changing the CCCR register,
+	 * and some chips set the CSR and CSA bits when in standby. Furthermore,
+	 * the CSR and CSA bits should be written as zeros, even when they read
+	 * ones.
+	 */
+	val_after &= ~(CCCR_CSR | CCCR_CSA);
 
-	while ((m_can_read(cdev, M_CAN_CCCR) & (CCCR_INIT | CCCR_CCE)) != val) {
-		if (timeout == 0) {
-			netdev_warn(cdev->net, "Failed to init module\n");
-			return;
-		}
-		timeout--;
-		udelay(1);
+	while (tries--) {
+		u32 val_read;
+
+		/* Write the desired value in each try, as setting some bits in
+		 * the CCCR register require other bits to be set first. E.g.
+		 * setting the NISO bit requires setting the CCE bit first.
+		 */
+		m_can_write(cdev, M_CAN_CCCR, val_after);
+
+		val_read = m_can_read(cdev, M_CAN_CCCR) & ~(CCCR_CSR | CCCR_CSA);
+
+		if (val_read == val_after)
+			return 0;
+
+		usleep_range(1, 5);
 	}
+
+	return -ETIMEDOUT;
+}
+
+static int m_can_config_enable(struct m_can_classdev *cdev)
+{
+	int err;
+
+	/* CCCR_INIT must be set in order to set CCCR_CCE, but access to
+	 * configuration registers should only be enabled when in standby mode,
+	 * where CCCR_INIT is always set.
+	 */
+	err = m_can_cccr_update_bits(cdev, CCCR_CCE, CCCR_CCE);
+	if (err)
+		netdev_err(cdev->net, "failed to enable configuration mode\n");
+
+	return err;
+}
+
+static int m_can_config_disable(struct m_can_classdev *cdev)
+{
+	int err;
+
+	/* Only clear CCCR_CCE, since CCCR_INIT cannot be cleared while in
+	 * standby mode
+	 */
+	err = m_can_cccr_update_bits(cdev, CCCR_CCE, 0);
+	if (err)
+		netdev_err(cdev->net, "failed to disable configuration registers\n");
+
+	return err;
 }
 
 static void m_can_interrupt_enable(struct m_can_classdev *cdev, u32 interrupts)
@@ -1403,7 +1437,9 @@ static int m_can_chip_config(struct net_device *dev)
 	interrupts &= ~(IR_ARA | IR_ELO | IR_DRX | IR_TEFF | IR_TFE | IR_TCF |
 			IR_HPM | IR_RF1F | IR_RF1W | IR_RF1N | IR_RF0F);
 
-	m_can_config_endisable(cdev, true);
+	err = m_can_config_enable(cdev);
+	if (err)
+		return err;
 
 	/* RX Buffer/FIFO Element Size 64 bytes data field */
 	m_can_write(cdev, M_CAN_RXESC,
@@ -1521,7 +1557,9 @@ static int m_can_chip_config(struct net_device *dev)
 		    FIELD_PREP(TSCC_TCP_MASK, 0xf) |
 		    FIELD_PREP(TSCC_TSS_MASK, TSCC_TSS_INTERNAL));
 
-	m_can_config_endisable(cdev, false);
+	err = m_can_config_disable(cdev);
+	if (err)
+		return err;
 
 	if (cdev->ops->init)
 		cdev->ops->init(cdev);
@@ -1550,7 +1588,11 @@ static int m_can_start(struct net_device *dev)
 		cdev->tx_fifo_putidx = FIELD_GET(TXFQS_TFQPI_MASK,
 						 m_can_read(cdev, M_CAN_TXFQS));
 
-	return 0;
+	ret = m_can_cccr_update_bits(cdev, CCCR_INIT, 0);
+	if (ret)
+		netdev_err(dev, "failed to enter normal mode\n");
+
+	return ret;
 }
 
 static int m_can_set_mode(struct net_device *dev, enum can_mode mode)
@@ -1599,43 +1641,37 @@ static int m_can_check_core_release(struct m_can_classdev *cdev)
 }
 
 /* Selectable Non ISO support only in version 3.2.x
- * This function checks if the bit is writable.
+ * Return 1 if the bit is writable, 0 if it is not, or negative on error.
  */
-static bool m_can_niso_supported(struct m_can_classdev *cdev)
+static int m_can_niso_supported(struct m_can_classdev *cdev)
 {
-	u32 cccr_reg, cccr_poll = 0;
-	int niso_timeout = -ETIMEDOUT;
-	int i;
+	int ret, niso;
 
-	m_can_config_endisable(cdev, true);
-	cccr_reg = m_can_read(cdev, M_CAN_CCCR);
-	cccr_reg |= CCCR_NISO;
-	m_can_write(cdev, M_CAN_CCCR, cccr_reg);
+	ret = m_can_config_enable(cdev);
+	if (ret)
+		return ret;
 
-	for (i = 0; i <= 10; i++) {
-		cccr_poll = m_can_read(cdev, M_CAN_CCCR);
-		if (cccr_poll == cccr_reg) {
-			niso_timeout = 0;
-			break;
-		}
+	/* First try to set the NISO bit. */
+	niso = m_can_cccr_update_bits(cdev, CCCR_NISO, CCCR_NISO);
 
-		usleep_range(1, 5);
+	/* Then clear the it again. */
+	ret = m_can_cccr_update_bits(cdev, CCCR_NISO, 0);
+	if (ret) {
+		dev_err(cdev->dev, "failed to revert the NON-ISO bit in CCCR\n");
+		return ret;
 	}
 
-	/* Clear NISO */
-	cccr_reg &= ~(CCCR_NISO);
-	m_can_write(cdev, M_CAN_CCCR, cccr_reg);
-
-	m_can_config_endisable(cdev, false);
+	ret = m_can_config_disable(cdev);
+	if (ret)
+		return ret;
 
-	/* return false if time out (-ETIMEDOUT), else return true */
-	return !niso_timeout;
+	return niso == 0;
 }
 
 static int m_can_dev_setup(struct m_can_classdev *cdev)
 {
 	struct net_device *dev = cdev->net;
-	int m_can_version, err;
+	int m_can_version, err, niso;
 
 	m_can_version = m_can_check_core_release(cdev);
 	/* return if unsupported version */
@@ -1684,9 +1720,11 @@ static int m_can_dev_setup(struct m_can_classdev *cdev)
 		cdev->can.bittiming_const = &m_can_bittiming_const_31X;
 		cdev->can.data_bittiming_const = &m_can_data_bittiming_const_31X;
 
-		cdev->can.ctrlmode_supported |=
-			(m_can_niso_supported(cdev) ?
-			 CAN_CTRLMODE_FD_NON_ISO : 0);
+		niso = m_can_niso_supported(cdev);
+		if (niso < 0)
+			return niso;
+		if (niso)
+			cdev->can.ctrlmode_supported |= CAN_CTRLMODE_FD_NON_ISO;
 		break;
 	default:
 		dev_err(cdev->dev, "Unsupported version number: %2d",
@@ -1694,21 +1732,26 @@ static int m_can_dev_setup(struct m_can_classdev *cdev)
 		return -EINVAL;
 	}
 
-	if (cdev->ops->init)
-		cdev->ops->init(cdev);
-
-	return 0;
+	/* Forcing standby mode should be redundant, as the chip should be in
+	 * standby after a reset. Write the INIT bit anyways, should the chip
+	 * be configured by previous stage.
+	 */
+	return m_can_cccr_update_bits(cdev, CCCR_INIT, CCCR_INIT);
 }
 
 static void m_can_stop(struct net_device *dev)
 {
 	struct m_can_classdev *cdev = netdev_priv(dev);
+	int ret;
 
 	/* disable all interrupts */
 	m_can_disable_all_interrupts(cdev);
 
 	/* Set init mode to disengage from the network */
-	m_can_config_endisable(cdev, true);
+	ret = m_can_cccr_update_bits(cdev, CCCR_INIT, CCCR_INIT);
+	if (ret)
+		netdev_err(dev, "failed to enter standby mode: %pe\n",
+			   ERR_PTR(ret));
 
 	/* set the state as STOPPED */
 	cdev->can.state = CAN_STATE_STOPPED;
diff --git a/drivers/net/can/m_can/tcan4x5x-core.c b/drivers/net/can/m_can/tcan4x5x-core.c
index a42600dac70d..d723206ac7c9 100644
--- a/drivers/net/can/m_can/tcan4x5x-core.c
+++ b/drivers/net/can/m_can/tcan4x5x-core.c
@@ -453,10 +453,17 @@ static int tcan4x5x_can_probe(struct spi_device *spi)
 		goto out_power;
 	}
 
-	ret = tcan4x5x_init(mcan_class);
+	tcan4x5x_check_wake(priv);
+
+	ret = tcan4x5x_write_tcan_reg(mcan_class, TCAN4X5X_INT_EN, 0);
 	if (ret) {
-		dev_err(&spi->dev, "tcan initialization failed %pe\n",
-			ERR_PTR(ret));
+		dev_err(&spi->dev, "Disabling interrupts failed %pe\n", ERR_PTR(ret));
+		goto out_power;
+	}
+
+	ret = tcan4x5x_clear_interrupts(mcan_class);
+	if (ret) {
+		dev_err(&spi->dev, "Clearing interrupts failed %pe\n", ERR_PTR(ret));
 		goto out_power;
 	}
 
-- 
cgit v1.2.3-58-ga151


From a8bd4d7af70d3e042bcffbdc5aa785b30b1a393e Mon Sep 17 00:00:00 2001
From: Johan Jonker <jbx6244@gmail.com>
Date: Tue, 18 Jun 2024 18:13:56 +0200
Subject: ARM: dts: rockchip: rk3xxx: fix emac node

In the combined DT of rk3066a/rk3188 the emac node uses as place holder
the compatible string "snps,arc-emac". The last real user nSIM_700
of the "snps,arc-emac" compatible string in a driver was removed in 2019.
Rockchip emac nodes don't make use of this common fall back string.
In order to removed unused driver code replace this string with
"rockchip,rk3066-emac".
As we are there remove the blank lines and sort.

Signed-off-by: Johan Jonker <jbx6244@gmail.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/arm/boot/dts/rockchip/rk3066a.dtsi | 4 ----
 arch/arm/boot/dts/rockchip/rk3xxx.dtsi  | 7 ++-----
 2 files changed, 2 insertions(+), 9 deletions(-)

diff --git a/arch/arm/boot/dts/rockchip/rk3066a.dtsi b/arch/arm/boot/dts/rockchip/rk3066a.dtsi
index 30139f21de64..51ae0418a7db 100644
--- a/arch/arm/boot/dts/rockchip/rk3066a.dtsi
+++ b/arch/arm/boot/dts/rockchip/rk3066a.dtsi
@@ -879,7 +879,3 @@
 &wdt {
 	compatible = "rockchip,rk3066-wdt", "snps,dw-wdt";
 };
-
-&emac {
-	compatible = "rockchip,rk3066-emac";
-};
diff --git a/arch/arm/boot/dts/rockchip/rk3xxx.dtsi b/arch/arm/boot/dts/rockchip/rk3xxx.dtsi
index f37137f298d5..e6a78bcf9163 100644
--- a/arch/arm/boot/dts/rockchip/rk3xxx.dtsi
+++ b/arch/arm/boot/dts/rockchip/rk3xxx.dtsi
@@ -194,17 +194,14 @@
 	};
 
 	emac: ethernet@10204000 {
-		compatible = "snps,arc-emac";
+		compatible = "rockchip,rk3066-emac";
 		reg = <0x10204000 0x3c>;
 		interrupts = <GIC_SPI 19 IRQ_TYPE_LEVEL_HIGH>;
-
-		rockchip,grf = <&grf>;
-
 		clocks = <&cru HCLK_EMAC>, <&cru SCLK_MAC>;
 		clock-names = "hclk", "macref";
 		max-speed = <100>;
 		phy-mode = "rmii";
-
+		rockchip,grf = <&grf>;
 		status = "disabled";
 	};
 
-- 
cgit v1.2.3-58-ga151


From a119aec5bf343be35adb5bee6c60c2d6bb5fdd26 Mon Sep 17 00:00:00 2001
From: Johan Jonker <jbx6244@gmail.com>
Date: Tue, 18 Jun 2024 18:14:14 +0200
Subject: net: ethernet: arc: remove emac_arc driver

The last real user nSIM_700 of the "snps,arc-emac" compatible string in
a driver was removed in 2019. The use of this string in the combined DT of
rk3066a/rk3188 as place holder has also been replaced, so
remove emac_arc.c to clean up some code.

Signed-off-by: Johan Jonker <jbx6244@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/arc/Kconfig    | 10 -----
 drivers/net/ethernet/arc/Makefile   |  1 -
 drivers/net/ethernet/arc/emac_arc.c | 88 -------------------------------------
 3 files changed, 99 deletions(-)
 delete mode 100644 drivers/net/ethernet/arc/emac_arc.c

diff --git a/drivers/net/ethernet/arc/Kconfig b/drivers/net/ethernet/arc/Kconfig
index 0a67612af228..0d400a7d8d91 100644
--- a/drivers/net/ethernet/arc/Kconfig
+++ b/drivers/net/ethernet/arc/Kconfig
@@ -23,16 +23,6 @@ config ARC_EMAC_CORE
 	select PHYLIB
 	select CRC32
 
-config ARC_EMAC
-	tristate "ARC EMAC support"
-	select ARC_EMAC_CORE
-	depends on OF_IRQ
-	depends on ARC || COMPILE_TEST
-	help
-	  On some legacy ARC (Synopsys) FPGA boards such as ARCAngel4/ML50x
-	  non-standard on-chip ethernet device ARC EMAC 10/100 is used.
-	  Say Y here if you have such a board.  If unsure, say N.
-
 config EMAC_ROCKCHIP
 	tristate "Rockchip EMAC support"
 	select ARC_EMAC_CORE
diff --git a/drivers/net/ethernet/arc/Makefile b/drivers/net/ethernet/arc/Makefile
index d63ada577c8e..23586eefec44 100644
--- a/drivers/net/ethernet/arc/Makefile
+++ b/drivers/net/ethernet/arc/Makefile
@@ -5,5 +5,4 @@
 
 arc_emac-objs := emac_main.o emac_mdio.o
 obj-$(CONFIG_ARC_EMAC_CORE) += arc_emac.o
-obj-$(CONFIG_ARC_EMAC) += emac_arc.o
 obj-$(CONFIG_EMAC_ROCKCHIP) += emac_rockchip.o
diff --git a/drivers/net/ethernet/arc/emac_arc.c b/drivers/net/ethernet/arc/emac_arc.c
deleted file mode 100644
index a3afddb23ee8..000000000000
--- a/drivers/net/ethernet/arc/emac_arc.c
+++ /dev/null
@@ -1,88 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-or-later
-/**
- * DOC: emac_arc.c - ARC EMAC specific glue layer
- *
- * Copyright (C) 2014 Romain Perier
- *
- * Romain Perier  <romain.perier@gmail.com>
- */
-
-#include <linux/etherdevice.h>
-#include <linux/module.h>
-#include <linux/of_net.h>
-#include <linux/platform_device.h>
-
-#include "emac.h"
-
-#define DRV_NAME    "emac_arc"
-
-static int emac_arc_probe(struct platform_device *pdev)
-{
-	struct device *dev = &pdev->dev;
-	struct arc_emac_priv *priv;
-	phy_interface_t interface;
-	struct net_device *ndev;
-	int err;
-
-	if (!dev->of_node)
-		return -ENODEV;
-
-	ndev = alloc_etherdev(sizeof(struct arc_emac_priv));
-	if (!ndev)
-		return -ENOMEM;
-	platform_set_drvdata(pdev, ndev);
-	SET_NETDEV_DEV(ndev, dev);
-
-	priv = netdev_priv(ndev);
-	priv->drv_name = DRV_NAME;
-
-	err = of_get_phy_mode(dev->of_node, &interface);
-	if (err) {
-		if (err == -ENODEV)
-			interface = PHY_INTERFACE_MODE_MII;
-		else
-			goto out_netdev;
-	}
-
-	priv->clk = devm_clk_get(dev, "hclk");
-	if (IS_ERR(priv->clk)) {
-		dev_err(dev, "failed to retrieve host clock from device tree\n");
-		err = -EINVAL;
-		goto out_netdev;
-	}
-
-	err = arc_emac_probe(ndev, interface);
-out_netdev:
-	if (err)
-		free_netdev(ndev);
-	return err;
-}
-
-static void emac_arc_remove(struct platform_device *pdev)
-{
-	struct net_device *ndev = platform_get_drvdata(pdev);
-
-	arc_emac_remove(ndev);
-	free_netdev(ndev);
-}
-
-static const struct of_device_id emac_arc_dt_ids[] = {
-	{ .compatible = "snps,arc-emac" },
-	{ /* Sentinel */ }
-};
-MODULE_DEVICE_TABLE(of, emac_arc_dt_ids);
-
-static struct platform_driver emac_arc_driver = {
-	.probe = emac_arc_probe,
-	.remove_new = emac_arc_remove,
-	.driver = {
-		.name = DRV_NAME,
-		.of_match_table  = emac_arc_dt_ids,
-	},
-};
-
-module_platform_driver(emac_arc_driver);
-
-MODULE_AUTHOR("Romain Perier <romain.perier@gmail.com>");
-MODULE_DESCRIPTION("ARC EMAC platform driver");
-MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-58-ga151


From 8a3913c8e05beb376572b074ad3c86a955b7dbf4 Mon Sep 17 00:00:00 2001
From: Johan Jonker <jbx6244@gmail.com>
Date: Tue, 18 Jun 2024 18:14:32 +0200
Subject: dt-bindings: net: remove arc_emac.txt

The last real user nSIM_700 of the "snps,arc-emac" compatible string in
a driver was removed in 2019. The use of this string in the combined DT of
rk3066a/rk3188 as place holder has also been replaced, so
remove arc_emac.txt

Signed-off-by: Johan Jonker <jbx6244@gmail.com>
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/devicetree/bindings/net/arc_emac.txt | 46 ----------------------
 1 file changed, 46 deletions(-)
 delete mode 100644 Documentation/devicetree/bindings/net/arc_emac.txt

diff --git a/Documentation/devicetree/bindings/net/arc_emac.txt b/Documentation/devicetree/bindings/net/arc_emac.txt
deleted file mode 100644
index c73a0e9c625e..000000000000
--- a/Documentation/devicetree/bindings/net/arc_emac.txt
+++ /dev/null
@@ -1,46 +0,0 @@
-* Synopsys ARC EMAC 10/100 Ethernet driver (EMAC)
-
-Required properties:
-- compatible: Should be "snps,arc-emac"
-- reg: Address and length of the register set for the device
-- interrupts: Should contain the EMAC interrupts
-- max-speed: see ethernet.txt file in the same directory.
-- phy: see ethernet.txt file in the same directory.
-
-Optional properties:
-- phy-reset-gpios : Should specify the gpio for phy reset
-- phy-reset-duration : Reset duration in milliseconds.  Should present
-  only if property "phy-reset-gpios" is available.  Missing the property
-  will have the duration be 1 millisecond.  Numbers greater than 1000 are
-  invalid and 1 millisecond will be used instead.
-
-Clock handling:
-The clock frequency is needed to calculate and set polling period of EMAC.
-It must be provided by one of:
-- clock-frequency: CPU frequency.
-- clocks: reference to the clock supplying the EMAC.
-
-Child nodes of the driver are the individual PHY devices connected to the
-MDIO bus. They must have a "reg" property given the PHY address on the MDIO bus.
-
-Examples:
-
-	ethernet@c0fc2000 {
-		compatible = "snps,arc-emac";
-		reg = <0xc0fc2000 0x3c>;
-		interrupts = <6>;
-		mac-address = [ 00 11 22 33 44 55 ];
-
-		clock-frequency = <80000000>;
-		/* or */
-		clocks = <&emac_clock>;
-
-		max-speed = <100>;
-		phy = <&phy0>;
-
-		#address-cells = <1>;
-		#size-cells = <0>;
-		phy0: ethernet-phy@0 {
-			reg = <1>;
-		};
-	};
-- 
cgit v1.2.3-58-ga151


From 88f56254a27522d3ecf95896ff2e7f57753fe0a5 Mon Sep 17 00:00:00 2001
From: David Wei <dw@davidwei.uk>
Date: Tue, 18 Jun 2024 23:29:30 -0700
Subject: bnxt_en: split rx ring helpers out from ring helpers

To prepare for queue API implementation, split rx ring functions out
from ring helpers. These new helpers will be called from queue API
implementation.

Signed-off-by: David Wei <dw@davidwei.uk>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 300 +++++++++++++++++++-----------
 1 file changed, 193 insertions(+), 107 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 1f72adaef709..4a24cbdaf6e0 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -3316,37 +3316,12 @@ static void bnxt_free_tx_skbs(struct bnxt *bp)
 	}
 }
 
-static void bnxt_free_one_rx_ring_skbs(struct bnxt *bp, int ring_nr)
+static void bnxt_free_one_rx_ring(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
 {
-	struct bnxt_rx_ring_info *rxr = &bp->rx_ring[ring_nr];
 	struct pci_dev *pdev = bp->pdev;
-	struct bnxt_tpa_idx_map *map;
-	int i, max_idx, max_agg_idx;
+	int i, max_idx;
 
 	max_idx = bp->rx_nr_pages * RX_DESC_CNT;
-	max_agg_idx = bp->rx_agg_nr_pages * RX_DESC_CNT;
-	if (!rxr->rx_tpa)
-		goto skip_rx_tpa_free;
-
-	for (i = 0; i < bp->max_tpa; i++) {
-		struct bnxt_tpa_info *tpa_info = &rxr->rx_tpa[i];
-		u8 *data = tpa_info->data;
-
-		if (!data)
-			continue;
-
-		dma_unmap_single_attrs(&pdev->dev, tpa_info->mapping,
-				       bp->rx_buf_use_size, bp->rx_dir,
-				       DMA_ATTR_WEAK_ORDERING);
-
-		tpa_info->data = NULL;
-
-		skb_free_frag(data);
-	}
-
-skip_rx_tpa_free:
-	if (!rxr->rx_buf_ring)
-		goto skip_rx_buf_free;
 
 	for (i = 0; i < max_idx; i++) {
 		struct bnxt_sw_rx_bd *rx_buf = &rxr->rx_buf_ring[i];
@@ -3366,12 +3341,15 @@ skip_rx_tpa_free:
 			skb_free_frag(data);
 		}
 	}
+}
 
-skip_rx_buf_free:
-	if (!rxr->rx_agg_ring)
-		goto skip_rx_agg_free;
+static void bnxt_free_one_rx_agg_ring(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
+{
+	int i, max_idx;
 
-	for (i = 0; i < max_agg_idx; i++) {
+	max_idx = bp->rx_agg_nr_pages * RX_DESC_CNT;
+
+	for (i = 0; i < max_idx; i++) {
 		struct bnxt_sw_rx_agg_bd *rx_agg_buf = &rxr->rx_agg_ring[i];
 		struct page *page = rx_agg_buf->page;
 
@@ -3383,6 +3361,45 @@ skip_rx_buf_free:
 
 		page_pool_recycle_direct(rxr->page_pool, page);
 	}
+}
+
+static void bnxt_free_one_rx_ring_skbs(struct bnxt *bp, int ring_nr)
+{
+	struct bnxt_rx_ring_info *rxr = &bp->rx_ring[ring_nr];
+	struct pci_dev *pdev = bp->pdev;
+	struct bnxt_tpa_idx_map *map;
+	int i;
+
+	if (!rxr->rx_tpa)
+		goto skip_rx_tpa_free;
+
+	for (i = 0; i < bp->max_tpa; i++) {
+		struct bnxt_tpa_info *tpa_info = &rxr->rx_tpa[i];
+		u8 *data = tpa_info->data;
+
+		if (!data)
+			continue;
+
+		dma_unmap_single_attrs(&pdev->dev, tpa_info->mapping,
+				       bp->rx_buf_use_size, bp->rx_dir,
+				       DMA_ATTR_WEAK_ORDERING);
+
+		tpa_info->data = NULL;
+
+		skb_free_frag(data);
+	}
+
+skip_rx_tpa_free:
+	if (!rxr->rx_buf_ring)
+		goto skip_rx_buf_free;
+
+	bnxt_free_one_rx_ring(bp, rxr);
+
+skip_rx_buf_free:
+	if (!rxr->rx_agg_ring)
+		goto skip_rx_agg_free;
+
+	bnxt_free_one_rx_agg_ring(bp, rxr);
 
 skip_rx_agg_free:
 	map = rxr->rx_tpa_idx_map;
@@ -4061,37 +4078,55 @@ static void bnxt_init_rxbd_pages(struct bnxt_ring_struct *ring, u32 type)
 	}
 }
 
-static int bnxt_alloc_one_rx_ring(struct bnxt *bp, int ring_nr)
+static void bnxt_alloc_one_rx_ring_skb(struct bnxt *bp,
+				       struct bnxt_rx_ring_info *rxr,
+				       int ring_nr)
 {
-	struct bnxt_rx_ring_info *rxr = &bp->rx_ring[ring_nr];
-	struct net_device *dev = bp->dev;
 	u32 prod;
 	int i;
 
 	prod = rxr->rx_prod;
 	for (i = 0; i < bp->rx_ring_size; i++) {
 		if (bnxt_alloc_rx_data(bp, rxr, prod, GFP_KERNEL)) {
-			netdev_warn(dev, "init'ed rx ring %d with %d/%d skbs only\n",
+			netdev_warn(bp->dev, "init'ed rx ring %d with %d/%d skbs only\n",
 				    ring_nr, i, bp->rx_ring_size);
 			break;
 		}
 		prod = NEXT_RX(prod);
 	}
 	rxr->rx_prod = prod;
+}
 
-	if (!(bp->flags & BNXT_FLAG_AGG_RINGS))
-		return 0;
+static void bnxt_alloc_one_rx_ring_page(struct bnxt *bp,
+					struct bnxt_rx_ring_info *rxr,
+					int ring_nr)
+{
+	u32 prod;
+	int i;
 
 	prod = rxr->rx_agg_prod;
 	for (i = 0; i < bp->rx_agg_ring_size; i++) {
 		if (bnxt_alloc_rx_page(bp, rxr, prod, GFP_KERNEL)) {
-			netdev_warn(dev, "init'ed rx ring %d with %d/%d pages only\n",
+			netdev_warn(bp->dev, "init'ed rx ring %d with %d/%d pages only\n",
 				    ring_nr, i, bp->rx_ring_size);
 			break;
 		}
 		prod = NEXT_RX_AGG(prod);
 	}
 	rxr->rx_agg_prod = prod;
+}
+
+static int bnxt_alloc_one_rx_ring(struct bnxt *bp, int ring_nr)
+{
+	struct bnxt_rx_ring_info *rxr = &bp->rx_ring[ring_nr];
+	int i;
+
+	bnxt_alloc_one_rx_ring_skb(bp, rxr, ring_nr);
+
+	if (!(bp->flags & BNXT_FLAG_AGG_RINGS))
+		return 0;
+
+	bnxt_alloc_one_rx_ring_page(bp, rxr, ring_nr);
 
 	if (rxr->rx_tpa) {
 		dma_addr_t mapping;
@@ -4110,9 +4145,9 @@ static int bnxt_alloc_one_rx_ring(struct bnxt *bp, int ring_nr)
 	return 0;
 }
 
-static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr)
+static void bnxt_init_one_rx_ring_rxbd(struct bnxt *bp,
+				       struct bnxt_rx_ring_info *rxr)
 {
-	struct bnxt_rx_ring_info *rxr;
 	struct bnxt_ring_struct *ring;
 	u32 type;
 
@@ -4122,28 +4157,43 @@ static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr)
 	if (NET_IP_ALIGN == 2)
 		type |= RX_BD_FLAGS_SOP;
 
-	rxr = &bp->rx_ring[ring_nr];
 	ring = &rxr->rx_ring_struct;
 	bnxt_init_rxbd_pages(ring, type);
-
-	netif_queue_set_napi(bp->dev, ring_nr, NETDEV_QUEUE_TYPE_RX,
-			     &rxr->bnapi->napi);
-
-	if (BNXT_RX_PAGE_MODE(bp) && bp->xdp_prog) {
-		bpf_prog_add(bp->xdp_prog, 1);
-		rxr->xdp_prog = bp->xdp_prog;
-	}
 	ring->fw_ring_id = INVALID_HW_RING_ID;
+}
+
+static void bnxt_init_one_rx_agg_ring_rxbd(struct bnxt *bp,
+					   struct bnxt_rx_ring_info *rxr)
+{
+	struct bnxt_ring_struct *ring;
+	u32 type;
 
 	ring = &rxr->rx_agg_ring_struct;
 	ring->fw_ring_id = INVALID_HW_RING_ID;
-
 	if ((bp->flags & BNXT_FLAG_AGG_RINGS)) {
 		type = ((u32)BNXT_RX_PAGE_SIZE << RX_BD_LEN_SHIFT) |
 			RX_BD_TYPE_RX_AGG_BD | RX_BD_FLAGS_SOP;
 
 		bnxt_init_rxbd_pages(ring, type);
 	}
+}
+
+static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr)
+{
+	struct bnxt_rx_ring_info *rxr;
+
+	rxr = &bp->rx_ring[ring_nr];
+	bnxt_init_one_rx_ring_rxbd(bp, rxr);
+
+	netif_queue_set_napi(bp->dev, ring_nr, NETDEV_QUEUE_TYPE_RX,
+			     &rxr->bnapi->napi);
+
+	if (BNXT_RX_PAGE_MODE(bp) && bp->xdp_prog) {
+		bpf_prog_add(bp->xdp_prog, 1);
+		rxr->xdp_prog = bp->xdp_prog;
+	}
+
+	bnxt_init_one_rx_agg_ring_rxbd(bp, rxr);
 
 	return bnxt_alloc_one_rx_ring(bp, ring_nr);
 }
@@ -6868,6 +6918,48 @@ static void bnxt_set_db(struct bnxt *bp, struct bnxt_db_info *db, u32 ring_type,
 	bnxt_set_db_mask(bp, db, ring_type);
 }
 
+static int bnxt_hwrm_rx_ring_alloc(struct bnxt *bp,
+				   struct bnxt_rx_ring_info *rxr)
+{
+	struct bnxt_ring_struct *ring = &rxr->rx_ring_struct;
+	struct bnxt_napi *bnapi = rxr->bnapi;
+	u32 type = HWRM_RING_ALLOC_RX;
+	u32 map_idx = bnapi->index;
+	int rc;
+
+	rc = hwrm_ring_alloc_send_msg(bp, ring, type, map_idx);
+	if (rc)
+		return rc;
+
+	bnxt_set_db(bp, &rxr->rx_db, type, map_idx, ring->fw_ring_id);
+	bp->grp_info[map_idx].rx_fw_ring_id = ring->fw_ring_id;
+
+	return 0;
+}
+
+static int bnxt_hwrm_rx_agg_ring_alloc(struct bnxt *bp,
+				       struct bnxt_rx_ring_info *rxr)
+{
+	struct bnxt_ring_struct *ring = &rxr->rx_agg_ring_struct;
+	u32 type = HWRM_RING_ALLOC_AGG;
+	u32 grp_idx = ring->grp_idx;
+	u32 map_idx;
+	int rc;
+
+	map_idx = grp_idx + bp->rx_nr_rings;
+	rc = hwrm_ring_alloc_send_msg(bp, ring, type, map_idx);
+	if (rc)
+		return rc;
+
+	bnxt_set_db(bp, &rxr->rx_agg_db, type, map_idx,
+		    ring->fw_ring_id);
+	bnxt_db_write(bp, &rxr->rx_agg_db, rxr->rx_agg_prod);
+	bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod);
+	bp->grp_info[grp_idx].agg_fw_ring_id = ring->fw_ring_id;
+
+	return 0;
+}
+
 static int bnxt_hwrm_ring_alloc(struct bnxt *bp)
 {
 	bool agg_rings = !!(bp->flags & BNXT_FLAG_AGG_RINGS);
@@ -6933,24 +7025,21 @@ static int bnxt_hwrm_ring_alloc(struct bnxt *bp)
 		bnxt_set_db(bp, &txr->tx_db, type, map_idx, ring->fw_ring_id);
 	}
 
-	type = HWRM_RING_ALLOC_RX;
 	for (i = 0; i < bp->rx_nr_rings; i++) {
 		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
-		struct bnxt_ring_struct *ring = &rxr->rx_ring_struct;
-		struct bnxt_napi *bnapi = rxr->bnapi;
-		u32 map_idx = bnapi->index;
 
-		rc = hwrm_ring_alloc_send_msg(bp, ring, type, map_idx);
+		rc = bnxt_hwrm_rx_ring_alloc(bp, rxr);
 		if (rc)
 			goto err_out;
-		bnxt_set_db(bp, &rxr->rx_db, type, map_idx, ring->fw_ring_id);
 		/* If we have agg rings, post agg buffers first. */
 		if (!agg_rings)
 			bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod);
-		bp->grp_info[map_idx].rx_fw_ring_id = ring->fw_ring_id;
 		if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
 			struct bnxt_cp_ring_info *cpr2 = rxr->rx_cpr;
+			struct bnxt_napi *bnapi = rxr->bnapi;
 			u32 type2 = HWRM_RING_ALLOC_CMPL;
+			struct bnxt_ring_struct *ring;
+			u32 map_idx = bnapi->index;
 
 			ring = &cpr2->cp_ring_struct;
 			ring->handle = BNXT_SET_NQ_HDL(cpr2);
@@ -6964,23 +7053,10 @@ static int bnxt_hwrm_ring_alloc(struct bnxt *bp)
 	}
 
 	if (agg_rings) {
-		type = HWRM_RING_ALLOC_AGG;
 		for (i = 0; i < bp->rx_nr_rings; i++) {
-			struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
-			struct bnxt_ring_struct *ring =
-						&rxr->rx_agg_ring_struct;
-			u32 grp_idx = ring->grp_idx;
-			u32 map_idx = grp_idx + bp->rx_nr_rings;
-
-			rc = hwrm_ring_alloc_send_msg(bp, ring, type, map_idx);
+			rc = bnxt_hwrm_rx_agg_ring_alloc(bp, &bp->rx_ring[i]);
 			if (rc)
 				goto err_out;
-
-			bnxt_set_db(bp, &rxr->rx_agg_db, type, map_idx,
-				    ring->fw_ring_id);
-			bnxt_db_write(bp, &rxr->rx_agg_db, rxr->rx_agg_prod);
-			bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod);
-			bp->grp_info[grp_idx].agg_fw_ring_id = ring->fw_ring_id;
 		}
 	}
 err_out:
@@ -7020,6 +7096,50 @@ exit:
 	return 0;
 }
 
+static void bnxt_hwrm_rx_ring_free(struct bnxt *bp,
+				   struct bnxt_rx_ring_info *rxr,
+				   bool close_path)
+{
+	struct bnxt_ring_struct *ring = &rxr->rx_ring_struct;
+	u32 grp_idx = rxr->bnapi->index;
+	u32 cmpl_ring_id;
+
+	if (ring->fw_ring_id == INVALID_HW_RING_ID)
+		return;
+
+	cmpl_ring_id = bnxt_cp_ring_for_rx(bp, rxr);
+	hwrm_ring_free_send_msg(bp, ring,
+				RING_FREE_REQ_RING_TYPE_RX,
+				close_path ? cmpl_ring_id :
+				INVALID_HW_RING_ID);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+	bp->grp_info[grp_idx].rx_fw_ring_id = INVALID_HW_RING_ID;
+}
+
+static void bnxt_hwrm_rx_agg_ring_free(struct bnxt *bp,
+				       struct bnxt_rx_ring_info *rxr,
+				       bool close_path)
+{
+	struct bnxt_ring_struct *ring = &rxr->rx_agg_ring_struct;
+	u32 grp_idx = rxr->bnapi->index;
+	u32 type, cmpl_ring_id;
+
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
+		type = RING_FREE_REQ_RING_TYPE_RX_AGG;
+	else
+		type = RING_FREE_REQ_RING_TYPE_RX;
+
+	if (ring->fw_ring_id == INVALID_HW_RING_ID)
+		return;
+
+	cmpl_ring_id = bnxt_cp_ring_for_rx(bp, rxr);
+	hwrm_ring_free_send_msg(bp, ring, type,
+				close_path ? cmpl_ring_id :
+				INVALID_HW_RING_ID);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
+	bp->grp_info[grp_idx].agg_fw_ring_id = INVALID_HW_RING_ID;
+}
+
 static void bnxt_hwrm_ring_free(struct bnxt *bp, bool close_path)
 {
 	u32 type;
@@ -7044,42 +7164,8 @@ static void bnxt_hwrm_ring_free(struct bnxt *bp, bool close_path)
 	}
 
 	for (i = 0; i < bp->rx_nr_rings; i++) {
-		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
-		struct bnxt_ring_struct *ring = &rxr->rx_ring_struct;
-		u32 grp_idx = rxr->bnapi->index;
-
-		if (ring->fw_ring_id != INVALID_HW_RING_ID) {
-			u32 cmpl_ring_id = bnxt_cp_ring_for_rx(bp, rxr);
-
-			hwrm_ring_free_send_msg(bp, ring,
-						RING_FREE_REQ_RING_TYPE_RX,
-						close_path ? cmpl_ring_id :
-						INVALID_HW_RING_ID);
-			ring->fw_ring_id = INVALID_HW_RING_ID;
-			bp->grp_info[grp_idx].rx_fw_ring_id =
-				INVALID_HW_RING_ID;
-		}
-	}
-
-	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
-		type = RING_FREE_REQ_RING_TYPE_RX_AGG;
-	else
-		type = RING_FREE_REQ_RING_TYPE_RX;
-	for (i = 0; i < bp->rx_nr_rings; i++) {
-		struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
-		struct bnxt_ring_struct *ring = &rxr->rx_agg_ring_struct;
-		u32 grp_idx = rxr->bnapi->index;
-
-		if (ring->fw_ring_id != INVALID_HW_RING_ID) {
-			u32 cmpl_ring_id = bnxt_cp_ring_for_rx(bp, rxr);
-
-			hwrm_ring_free_send_msg(bp, ring, type,
-						close_path ? cmpl_ring_id :
-						INVALID_HW_RING_ID);
-			ring->fw_ring_id = INVALID_HW_RING_ID;
-			bp->grp_info[grp_idx].agg_fw_ring_id =
-				INVALID_HW_RING_ID;
-		}
+		bnxt_hwrm_rx_ring_free(bp, &bp->rx_ring[i], close_path);
+		bnxt_hwrm_rx_agg_ring_free(bp, &bp->rx_ring[i], close_path);
 	}
 
 	/* The completion rings are about to be freed.  After that the
-- 
cgit v1.2.3-58-ga151


From 2d694c27d32efc9467a8a20e4ad641ab5adfd07d Mon Sep 17 00:00:00 2001
From: David Wei <dw@davidwei.uk>
Date: Tue, 18 Jun 2024 23:29:31 -0700
Subject: bnxt_en: implement netdev_queue_mgmt_ops

Implement netdev_queue_mgmt_ops for bnxt added in [1].

Two bnxt_rx_ring_info structs are allocated to hold the new/old queue
memory. Queue memory is copied from/to the main bp->rx_ring[idx]
bnxt_rx_ring_info.

Queue memory is pre-allocated in bnxt_queue_mem_alloc() into a clone,
and then copied into bp->rx_ring[idx] in bnxt_queue_mem_start().

Similarly, when bp->rx_ring[idx] is stopped its queue memory is copied
into a clone, and then freed later in bnxt_queue_mem_free().

I tested this patchset with netdev_rx_queue_restart(), including
inducing errors in all places that returns an error code. In all cases,
the queue is left in a good working state.

Rx queues are created/destroyed using bnxt_hwrm_rx_ring_alloc() and
bnxt_hwrm_rx_ring_free(), which issue HWRM_RING_ALLOC and HWRM_RING_FREE
commands respectively to the firmware. By the time a HWRM_RING_FREE
response is received, there won't be any more completions from that
queue.

Thanks to Somnath for helping me with this patch. With their permission
I've added them as Acked-by.

[1]: https://lore.kernel.org/netdev/20240501232549.1327174-2-shailend@google.com/

Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: David Wei <dw@davidwei.uk>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 275 ++++++++++++++++++++++++++++++
 1 file changed, 275 insertions(+)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 4a24cbdaf6e0..1bd0c5973252 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -3996,6 +3996,62 @@ static int bnxt_alloc_cp_rings(struct bnxt *bp)
 	return 0;
 }
 
+static void bnxt_init_rx_ring_struct(struct bnxt *bp,
+				     struct bnxt_rx_ring_info *rxr)
+{
+	struct bnxt_ring_mem_info *rmem;
+	struct bnxt_ring_struct *ring;
+
+	ring = &rxr->rx_ring_struct;
+	rmem = &ring->ring_mem;
+	rmem->nr_pages = bp->rx_nr_pages;
+	rmem->page_size = HW_RXBD_RING_SIZE;
+	rmem->pg_arr = (void **)rxr->rx_desc_ring;
+	rmem->dma_arr = rxr->rx_desc_mapping;
+	rmem->vmem_size = SW_RXBD_RING_SIZE * bp->rx_nr_pages;
+	rmem->vmem = (void **)&rxr->rx_buf_ring;
+
+	ring = &rxr->rx_agg_ring_struct;
+	rmem = &ring->ring_mem;
+	rmem->nr_pages = bp->rx_agg_nr_pages;
+	rmem->page_size = HW_RXBD_RING_SIZE;
+	rmem->pg_arr = (void **)rxr->rx_agg_desc_ring;
+	rmem->dma_arr = rxr->rx_agg_desc_mapping;
+	rmem->vmem_size = SW_RXBD_AGG_RING_SIZE * bp->rx_agg_nr_pages;
+	rmem->vmem = (void **)&rxr->rx_agg_ring;
+}
+
+static void bnxt_reset_rx_ring_struct(struct bnxt *bp,
+				      struct bnxt_rx_ring_info *rxr)
+{
+	struct bnxt_ring_mem_info *rmem;
+	struct bnxt_ring_struct *ring;
+	int i;
+
+	rxr->page_pool->p.napi = NULL;
+	rxr->page_pool = NULL;
+
+	ring = &rxr->rx_ring_struct;
+	rmem = &ring->ring_mem;
+	rmem->pg_tbl = NULL;
+	rmem->pg_tbl_map = 0;
+	for (i = 0; i < rmem->nr_pages; i++) {
+		rmem->pg_arr[i] = NULL;
+		rmem->dma_arr[i] = 0;
+	}
+	*rmem->vmem = NULL;
+
+	ring = &rxr->rx_agg_ring_struct;
+	rmem = &ring->ring_mem;
+	rmem->pg_tbl = NULL;
+	rmem->pg_tbl_map = 0;
+	for (i = 0; i < rmem->nr_pages; i++) {
+		rmem->pg_arr[i] = NULL;
+		rmem->dma_arr[i] = 0;
+	}
+	*rmem->vmem = NULL;
+}
+
 static void bnxt_init_ring_struct(struct bnxt *bp)
 {
 	int i, j;
@@ -14914,6 +14970,224 @@ static const struct netdev_stat_ops bnxt_stat_ops = {
 	.get_base_stats		= bnxt_get_base_stats,
 };
 
+static int bnxt_alloc_rx_agg_bmap(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
+{
+	u16 mem_size;
+
+	rxr->rx_agg_bmap_size = bp->rx_agg_ring_mask + 1;
+	mem_size = rxr->rx_agg_bmap_size / 8;
+	rxr->rx_agg_bmap = kzalloc(mem_size, GFP_KERNEL);
+	if (!rxr->rx_agg_bmap)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static int bnxt_queue_mem_alloc(struct net_device *dev, void *qmem, int idx)
+{
+	struct bnxt_rx_ring_info *rxr, *clone;
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_ring_struct *ring;
+	int rc;
+
+	rxr = &bp->rx_ring[idx];
+	clone = qmem;
+	memcpy(clone, rxr, sizeof(*rxr));
+	bnxt_init_rx_ring_struct(bp, clone);
+	bnxt_reset_rx_ring_struct(bp, clone);
+
+	clone->rx_prod = 0;
+	clone->rx_agg_prod = 0;
+	clone->rx_sw_agg_prod = 0;
+	clone->rx_next_cons = 0;
+
+	rc = bnxt_alloc_rx_page_pool(bp, clone, rxr->page_pool->p.nid);
+	if (rc)
+		return rc;
+
+	ring = &clone->rx_ring_struct;
+	rc = bnxt_alloc_ring(bp, &ring->ring_mem);
+	if (rc)
+		goto err_free_rx_ring;
+
+	if (bp->flags & BNXT_FLAG_AGG_RINGS) {
+		ring = &clone->rx_agg_ring_struct;
+		rc = bnxt_alloc_ring(bp, &ring->ring_mem);
+		if (rc)
+			goto err_free_rx_agg_ring;
+
+		rc = bnxt_alloc_rx_agg_bmap(bp, clone);
+		if (rc)
+			goto err_free_rx_agg_ring;
+	}
+
+	bnxt_init_one_rx_ring_rxbd(bp, clone);
+	bnxt_init_one_rx_agg_ring_rxbd(bp, clone);
+
+	bnxt_alloc_one_rx_ring_skb(bp, clone, idx);
+	if (bp->flags & BNXT_FLAG_AGG_RINGS)
+		bnxt_alloc_one_rx_ring_page(bp, clone, idx);
+
+	return 0;
+
+err_free_rx_agg_ring:
+	bnxt_free_ring(bp, &clone->rx_agg_ring_struct.ring_mem);
+err_free_rx_ring:
+	bnxt_free_ring(bp, &clone->rx_ring_struct.ring_mem);
+	clone->page_pool->p.napi = NULL;
+	page_pool_destroy(clone->page_pool);
+	clone->page_pool = NULL;
+	return rc;
+}
+
+static void bnxt_queue_mem_free(struct net_device *dev, void *qmem)
+{
+	struct bnxt_rx_ring_info *rxr = qmem;
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_ring_struct *ring;
+
+	bnxt_free_one_rx_ring(bp, rxr);
+	bnxt_free_one_rx_agg_ring(bp, rxr);
+
+	/* At this point, this NAPI instance has another page pool associated
+	 * with it. Disconnect here before freeing the old page pool to avoid
+	 * warnings.
+	 */
+	rxr->page_pool->p.napi = NULL;
+	page_pool_destroy(rxr->page_pool);
+	rxr->page_pool = NULL;
+
+	ring = &rxr->rx_ring_struct;
+	bnxt_free_ring(bp, &ring->ring_mem);
+
+	ring = &rxr->rx_agg_ring_struct;
+	bnxt_free_ring(bp, &ring->ring_mem);
+
+	kfree(rxr->rx_agg_bmap);
+	rxr->rx_agg_bmap = NULL;
+}
+
+static void bnxt_copy_rx_ring(struct bnxt *bp,
+			      struct bnxt_rx_ring_info *dst,
+			      struct bnxt_rx_ring_info *src)
+{
+	struct bnxt_ring_mem_info *dst_rmem, *src_rmem;
+	struct bnxt_ring_struct *dst_ring, *src_ring;
+	int i;
+
+	dst_ring = &dst->rx_ring_struct;
+	dst_rmem = &dst_ring->ring_mem;
+	src_ring = &src->rx_ring_struct;
+	src_rmem = &src_ring->ring_mem;
+
+	WARN_ON(dst_rmem->nr_pages != src_rmem->nr_pages);
+	WARN_ON(dst_rmem->page_size != src_rmem->page_size);
+	WARN_ON(dst_rmem->flags != src_rmem->flags);
+	WARN_ON(dst_rmem->depth != src_rmem->depth);
+	WARN_ON(dst_rmem->vmem_size != src_rmem->vmem_size);
+	WARN_ON(dst_rmem->ctx_mem != src_rmem->ctx_mem);
+
+	dst_rmem->pg_tbl = src_rmem->pg_tbl;
+	dst_rmem->pg_tbl_map = src_rmem->pg_tbl_map;
+	*dst_rmem->vmem = *src_rmem->vmem;
+	for (i = 0; i < dst_rmem->nr_pages; i++) {
+		dst_rmem->pg_arr[i] = src_rmem->pg_arr[i];
+		dst_rmem->dma_arr[i] = src_rmem->dma_arr[i];
+	}
+
+	if (!(bp->flags & BNXT_FLAG_AGG_RINGS))
+		return;
+
+	dst_ring = &dst->rx_agg_ring_struct;
+	dst_rmem = &dst_ring->ring_mem;
+	src_ring = &src->rx_agg_ring_struct;
+	src_rmem = &src_ring->ring_mem;
+
+	WARN_ON(dst_rmem->nr_pages != src_rmem->nr_pages);
+	WARN_ON(dst_rmem->page_size != src_rmem->page_size);
+	WARN_ON(dst_rmem->flags != src_rmem->flags);
+	WARN_ON(dst_rmem->depth != src_rmem->depth);
+	WARN_ON(dst_rmem->vmem_size != src_rmem->vmem_size);
+	WARN_ON(dst_rmem->ctx_mem != src_rmem->ctx_mem);
+	WARN_ON(dst->rx_agg_bmap_size != src->rx_agg_bmap_size);
+
+	dst_rmem->pg_tbl = src_rmem->pg_tbl;
+	dst_rmem->pg_tbl_map = src_rmem->pg_tbl_map;
+	*dst_rmem->vmem = *src_rmem->vmem;
+	for (i = 0; i < dst_rmem->nr_pages; i++) {
+		dst_rmem->pg_arr[i] = src_rmem->pg_arr[i];
+		dst_rmem->dma_arr[i] = src_rmem->dma_arr[i];
+	}
+
+	dst->rx_agg_bmap = src->rx_agg_bmap;
+}
+
+static int bnxt_queue_start(struct net_device *dev, void *qmem, int idx)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_rx_ring_info *rxr, *clone;
+	struct bnxt_cp_ring_info *cpr;
+	int rc;
+
+	rxr = &bp->rx_ring[idx];
+	clone = qmem;
+
+	rxr->rx_prod = clone->rx_prod;
+	rxr->rx_agg_prod = clone->rx_agg_prod;
+	rxr->rx_sw_agg_prod = clone->rx_sw_agg_prod;
+	rxr->rx_next_cons = clone->rx_next_cons;
+	rxr->page_pool = clone->page_pool;
+
+	bnxt_copy_rx_ring(bp, rxr, clone);
+
+	rc = bnxt_hwrm_rx_ring_alloc(bp, rxr);
+	if (rc)
+		return rc;
+	rc = bnxt_hwrm_rx_agg_ring_alloc(bp, rxr);
+	if (rc)
+		goto err_free_hwrm_rx_ring;
+
+	bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod);
+	if (bp->flags & BNXT_FLAG_AGG_RINGS)
+		bnxt_db_write(bp, &rxr->rx_agg_db, rxr->rx_agg_prod);
+
+	napi_enable(&rxr->bnapi->napi);
+
+	cpr = &rxr->bnapi->cp_ring;
+	cpr->sw_stats->rx.rx_resets++;
+
+	return 0;
+
+err_free_hwrm_rx_ring:
+	bnxt_hwrm_rx_ring_free(bp, rxr, false);
+	return rc;
+}
+
+static int bnxt_queue_stop(struct net_device *dev, void *qmem, int idx)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_rx_ring_info *rxr;
+
+	rxr = &bp->rx_ring[idx];
+	napi_disable(&rxr->bnapi->napi);
+	bnxt_hwrm_rx_ring_free(bp, rxr, false);
+	bnxt_hwrm_rx_agg_ring_free(bp, rxr, false);
+	rxr->rx_next_cons = 0;
+
+	memcpy(qmem, rxr, sizeof(*rxr));
+	bnxt_init_rx_ring_struct(bp, qmem);
+
+	return 0;
+}
+
+static const struct netdev_queue_mgmt_ops bnxt_queue_mgmt_ops = {
+	.ndo_queue_mem_size	= sizeof(struct bnxt_rx_ring_info),
+	.ndo_queue_mem_alloc	= bnxt_queue_mem_alloc,
+	.ndo_queue_mem_free	= bnxt_queue_mem_free,
+	.ndo_queue_start	= bnxt_queue_start,
+	.ndo_queue_stop		= bnxt_queue_stop,
+};
+
 static void bnxt_remove_one(struct pci_dev *pdev)
 {
 	struct net_device *dev = pci_get_drvdata(pdev);
@@ -15379,6 +15653,7 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	dev->stat_ops = &bnxt_stat_ops;
 	dev->watchdog_timeo = BNXT_TX_TIMEOUT;
 	dev->ethtool_ops = &bnxt_ethtool_ops;
+	dev->queue_mgmt_ops = &bnxt_queue_mgmt_ops;
 	pci_set_drvdata(pdev, dev);
 
 	rc = bnxt_alloc_hwrm_resources(bp);
-- 
cgit v1.2.3-58-ga151


From c32fe1986f27cac329767d3497986e306cad1d5e Mon Sep 17 00:00:00 2001
From: "Csókás, Bence" <csokas.bence@prolan.hu>
Date: Wed, 19 Jun 2024 14:31:11 +0200
Subject: net: fec: Fix FEC_ECR_EN1588 being cleared on link-down
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

FEC_ECR_EN1588 bit gets cleared after MAC reset in `fec_stop()`, which
makes all 1588 functionality shut down, and all the extended registers
disappear, on link-down, making the adapter fall back to compatibility
"dumb mode". However, some functionality needs to be retained (e.g. PPS)
even without link.

Fixes: 6605b730c061 ("FEC: Add time stamping code and a PTP hardware clock")
Cc: Richard Cochran <richardcochran@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/netdev/5fa9fadc-a89d-467a-aae9-c65469ff5fe1@lunn.ch/
Signed-off-by: Csókás, Bence <csokas.bence@prolan.hu>
Reviewed-by: Wei Fang <wei.fang@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/freescale/fec_main.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index 881ece735dcf..fb19295529a2 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -1361,6 +1361,12 @@ fec_stop(struct net_device *ndev)
 		writel(FEC_ECR_ETHEREN, fep->hwp + FEC_ECNTRL);
 		writel(rmii_mode, fep->hwp + FEC_R_CNTRL);
 	}
+
+	if (fep->bufdesc_ex) {
+		val = readl(fep->hwp + FEC_ECNTRL);
+		val |= FEC_ECR_EN1588;
+		writel(val, fep->hwp + FEC_ECNTRL);
+	}
 }
 
 static void
-- 
cgit v1.2.3-58-ga151


From dcec8d291da8813b5e1c7c0967ae63463a8521f6 Mon Sep 17 00:00:00 2001
From: Lukasz Majewski <lukma@denx.de>
Date: Wed, 19 Jun 2024 16:58:09 +0200
Subject: net: dsa: ksz_common: Allow only up to two HSR HW offloaded ports for
 KSZ9477

The KSZ9477 allows HSR in-HW offloading for any of two selected ports.
This patch adds check if one tries to use more than two ports with
HSR offloading enabled.

The problem is with RedBox configuration (HSR-SAN) - when configuring:
ip link add name hsr0 type hsr slave1 lan1 slave2 lan2 interlink lan3 \
	supervision 45 version 1

The lan1 (port0) and lan2 (port1) are correctly configured as ports, which
can use HSR offloading on ksz9477.

However, when we do already have two bits set in hsr_ports, we need to
return (-ENOTSUPP), so the interlink port (lan3) would be used with
SW based HSR RedBox support.

Otherwise, I do see some strange network behavior, as some HSR frames are
visible on non-HSR network and vice versa.

This causes the switch connected to interlink port (lan3) to drop frames
and no communication is possible.

Moreover, conceptually - the interlink (i.e. HSR-SAN port - lan3/port2)
shall be only supported in software as it is also possible to use ksz9477
with only SW based HSR (i.e. port0/1 -> hsr0 with offloading, port2 ->
HSR-SAN/interlink, port4/5 -> hsr1 with SW based HSR).

Fixes: 5055cccfc2d1 ("net: hsr: Provide RedBox support (HSR-SAN)")
Signed-off-by: Lukasz Majewski <lukma@denx.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/microchip/ksz_common.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c
index 2818e24e2a51..debb554b60ed 100644
--- a/drivers/net/dsa/microchip/ksz_common.c
+++ b/drivers/net/dsa/microchip/ksz_common.c
@@ -3906,6 +3906,13 @@ static int ksz_hsr_join(struct dsa_switch *ds, int port, struct net_device *hsr,
 		return -EOPNOTSUPP;
 	}
 
+	/* KSZ9477 can only perform HSR offloading for up to two ports */
+	if (hweight8(dev->hsr_ports) >= 2) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Cannot offload more than two ports - using software HSR");
+		return -EOPNOTSUPP;
+	}
+
 	/* Self MAC address filtering, to avoid frames traversing
 	 * the HSR ring more than once.
 	 */
-- 
cgit v1.2.3-58-ga151


From 4558645d139c01d5635cac7e3528a3d19bbb6526 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 19 Jun 2024 17:27:41 -0700
Subject: docs: net: document guidance of implementing the SR-IOV NDOs

New drivers were prevented from adding ndo_set_vf_* callbacks
over the last few years. This was expected to result in broader
switchdev adoption, but seems to have had little effect.

Based on recent netdev meeting there is broad support for allowing
adding those ops.

There is a problem with the current API supporting a limited number
of VFs (100+, which is less than some modern HW supports).
We can try to solve it by adding similar functionality on devlink
ports, but that'd be another API variation to maintain.
So a netlink attribute reshuffling is a more likely outcome.

Document the guidance, make it clear that the API is frozen.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/index.rst |  1 +
 Documentation/networking/sriov.rst | 25 +++++++++++++++++++++++++
 2 files changed, 26 insertions(+)
 create mode 100644 Documentation/networking/sriov.rst

diff --git a/Documentation/networking/index.rst b/Documentation/networking/index.rst
index a6443851a142..b4b2a002f183 100644
--- a/Documentation/networking/index.rst
+++ b/Documentation/networking/index.rst
@@ -105,6 +105,7 @@ Contents:
    seg6-sysctl
    skbuff
    smc-sysctl
+   sriov
    statistics
    strparser
    switchdev
diff --git a/Documentation/networking/sriov.rst b/Documentation/networking/sriov.rst
new file mode 100644
index 000000000000..5deb4ff3154f
--- /dev/null
+++ b/Documentation/networking/sriov.rst
@@ -0,0 +1,25 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+===============
+NIC SR-IOV APIs
+===============
+
+Modern NICs are strongly encouraged to focus on implementing the ``switchdev``
+model (see :ref:`switchdev`) to configure forwarding and security of SR-IOV
+functionality.
+
+Legacy API
+==========
+
+The old SR-IOV API is implemented in ``rtnetlink`` Netlink family as part of
+the ``RTM_GETLINK`` and ``RTM_SETLINK`` commands. On the driver side
+it consists of a number of ``ndo_set_vf_*`` and ``ndo_get_vf_*`` callbacks.
+
+Since the legacy APIs do not integrate well with the rest of the stack
+the API is considered frozen; no new functionality or extensions
+will be accepted. New drivers should not implement the uncommon callbacks;
+namely the following callbacks are off limits:
+
+ - ``ndo_get_vf_port``
+ - ``ndo_set_vf_port``
+ - ``ndo_set_vf_rss_query_en``
-- 
cgit v1.2.3-58-ga151


From 39c469188b6d20db107c454cac3ddea941718850 Mon Sep 17 00:00:00 2001
From: Sai Krishna <saikrishnag@marvell.com>
Date: Thu, 20 Jun 2024 14:29:49 +0530
Subject: octeontx2-pf: Add ucast filter count configurability via devlink.

The existing method of reserving unicast filter count leads to wasted
MCAM entries if the functionality is not used or fewer entries are used.
Furthermore, the amount of MCAM entries differs amongst Octeon SoCs.
We implemented a means to adjust the UC filter count via devlink,
allowing for better use of MCAM entries across Netdev apps.

commands:

To get the current unicast filter count
 # devlink dev param show pci/0002:02:00.0 name unicast_filter_count

To change/set the unicast filter count
 # devlink dev param  set  pci/0002:02:00.0  name unicast_filter_count
 value 5 cmode runtime

Signed-off-by: Sai Krishna <saikrishnag@marvell.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/networking/devlink/octeontx2.rst     | 16 ++++++
 .../ethernet/marvell/octeontx2/nic/otx2_common.h   |  7 +--
 .../ethernet/marvell/octeontx2/nic/otx2_devlink.c  | 64 ++++++++++++++++++++++
 .../ethernet/marvell/octeontx2/nic/otx2_flows.c    | 20 ++++---
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   |  2 +-
 5 files changed, 95 insertions(+), 14 deletions(-)

diff --git a/Documentation/networking/devlink/octeontx2.rst b/Documentation/networking/devlink/octeontx2.rst
index 610de99b728a..d33a90dd44bf 100644
--- a/Documentation/networking/devlink/octeontx2.rst
+++ b/Documentation/networking/devlink/octeontx2.rst
@@ -40,3 +40,19 @@ The ``octeontx2 AF`` driver implements the following driver-specific parameters.
      - runtime
      - Use to set the quantum which hardware uses for scheduling among transmit queues.
        Hardware uses weighted DWRR algorithm to schedule among all transmit queues.
+
+The ``octeontx2 PF`` driver implements the following driver-specific parameters.
+
+.. list-table:: Driver-specific parameters implemented
+   :widths: 5 5 5 85
+
+   * - Name
+     - Type
+     - Mode
+     - Description
+   * - ``unicast_filter_count``
+     - u8
+     - runtime
+     - Set the maximum number of unicast filters that can be programmed for
+       the device. This can be used to achieve better device resource
+       utilization, avoiding over consumption of unused MCAM table entries.
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 24fbbef265a6..f27a3456ae64 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -346,12 +346,9 @@ struct otx2_flow_config {
 	u16			*def_ent;
 	u16			nr_flows;
 #define OTX2_DEFAULT_FLOWCOUNT		16
-#define OTX2_MAX_UNICAST_FLOWS		8
+#define OTX2_DEFAULT_UNICAST_FLOWS	4
 #define OTX2_MAX_VLAN_FLOWS		1
 #define OTX2_MAX_TC_FLOWS	OTX2_DEFAULT_FLOWCOUNT
-#define OTX2_MCAM_COUNT		(OTX2_DEFAULT_FLOWCOUNT + \
-				 OTX2_MAX_UNICAST_FLOWS + \
-				 OTX2_MAX_VLAN_FLOWS)
 	u16			unicast_offset;
 	u16			rx_vlan_offset;
 	u16			vf_vlan_offset;
@@ -365,6 +362,7 @@ struct otx2_flow_config {
 	u16                     max_flows;
 	refcount_t		mark_flows;
 	struct list_head	flow_list_tc;
+	u8			ucast_flt_cnt;
 	bool			ntuple;
 };
 
@@ -1067,6 +1065,7 @@ int otx2_handle_ntuple_tc_features(struct net_device *netdev,
 int otx2_smq_flush(struct otx2_nic *pfvf, int smq);
 void otx2_free_bufs(struct otx2_nic *pfvf, struct otx2_pool *pool,
 		    u64 iova, int size);
+int otx2_mcam_entry_init(struct otx2_nic *pfvf);
 
 /* tc support */
 int otx2_init_tc(struct otx2_nic *nic);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c
index 458d34a62e18..53f14aa944bd 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c
@@ -64,9 +64,68 @@ static int otx2_dl_mcam_count_get(struct devlink *devlink, u32 id,
 	return 0;
 }
 
+static int otx2_dl_ucast_flt_cnt_set(struct devlink *devlink, u32 id,
+				     struct devlink_param_gset_ctx *ctx,
+				     struct netlink_ext_ack *extack)
+{
+	struct otx2_devlink *otx2_dl = devlink_priv(devlink);
+	struct otx2_nic *pfvf = otx2_dl->pfvf;
+	int err;
+
+	pfvf->flow_cfg->ucast_flt_cnt = ctx->val.vu8;
+
+	otx2_mcam_flow_del(pfvf);
+	err = otx2_mcam_entry_init(pfvf);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static int otx2_dl_ucast_flt_cnt_get(struct devlink *devlink, u32 id,
+				     struct devlink_param_gset_ctx *ctx)
+{
+	struct otx2_devlink *otx2_dl = devlink_priv(devlink);
+	struct otx2_nic *pfvf = otx2_dl->pfvf;
+
+	ctx->val.vu8 = pfvf->flow_cfg ? pfvf->flow_cfg->ucast_flt_cnt : 0;
+
+	return 0;
+}
+
+static int otx2_dl_ucast_flt_cnt_validate(struct devlink *devlink, u32 id,
+					  union devlink_param_value val,
+					  struct netlink_ext_ack *extack)
+{
+	struct otx2_devlink *otx2_dl = devlink_priv(devlink);
+	struct otx2_nic *pfvf = otx2_dl->pfvf;
+
+	/* Check for UNICAST filter support*/
+	if (!(pfvf->flags & OTX2_FLAG_UCAST_FLTR_SUPPORT)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Unicast filter not enabled");
+		return -EINVAL;
+	}
+
+	if (!pfvf->flow_cfg) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "pfvf->flow_cfg not initialized");
+		return -EINVAL;
+	}
+
+	if (pfvf->flow_cfg->nr_flows) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Cannot modify count when there are active rules");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 enum otx2_dl_param_id {
 	OTX2_DEVLINK_PARAM_ID_BASE = DEVLINK_PARAM_GENERIC_ID_MAX,
 	OTX2_DEVLINK_PARAM_ID_MCAM_COUNT,
+	OTX2_DEVLINK_PARAM_ID_UCAST_FLT_CNT,
 };
 
 static const struct devlink_param otx2_dl_params[] = {
@@ -75,6 +134,11 @@ static const struct devlink_param otx2_dl_params[] = {
 			     BIT(DEVLINK_PARAM_CMODE_RUNTIME),
 			     otx2_dl_mcam_count_get, otx2_dl_mcam_count_set,
 			     otx2_dl_mcam_count_validate),
+	DEVLINK_PARAM_DRIVER(OTX2_DEVLINK_PARAM_ID_UCAST_FLT_CNT,
+			     "unicast_filter_count", DEVLINK_PARAM_TYPE_U8,
+			     BIT(DEVLINK_PARAM_CMODE_RUNTIME),
+			     otx2_dl_ucast_flt_cnt_get, otx2_dl_ucast_flt_cnt_set,
+			     otx2_dl_ucast_flt_cnt_validate),
 };
 
 static const struct devlink_ops otx2_devlink_ops = {
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
index bc5819237ed7..98c31a16c70b 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
@@ -12,8 +12,6 @@
 
 #define OTX2_DEFAULT_ACTION	0x1
 
-static int otx2_mcam_entry_init(struct otx2_nic *pfvf);
-
 struct otx2_flow {
 	struct ethtool_rx_flow_spec flow_spec;
 	struct list_head list;
@@ -161,7 +159,7 @@ exit:
 }
 EXPORT_SYMBOL(otx2_alloc_mcam_entries);
 
-static int otx2_mcam_entry_init(struct otx2_nic *pfvf)
+int otx2_mcam_entry_init(struct otx2_nic *pfvf)
 {
 	struct otx2_flow_config *flow_cfg = pfvf->flow_cfg;
 	struct npc_get_field_status_req *freq;
@@ -172,7 +170,7 @@ static int otx2_mcam_entry_init(struct otx2_nic *pfvf)
 	int ent, count;
 
 	vf_vlan_max_flows = pfvf->total_vfs * OTX2_PER_VF_VLAN_FLOWS;
-	count = OTX2_MAX_UNICAST_FLOWS +
+	count = flow_cfg->ucast_flt_cnt +
 			OTX2_MAX_VLAN_FLOWS + vf_vlan_max_flows;
 
 	flow_cfg->def_ent = devm_kmalloc_array(pfvf->dev, count,
@@ -214,7 +212,7 @@ static int otx2_mcam_entry_init(struct otx2_nic *pfvf)
 	flow_cfg->vf_vlan_offset = 0;
 	flow_cfg->unicast_offset = vf_vlan_max_flows;
 	flow_cfg->rx_vlan_offset = flow_cfg->unicast_offset +
-					OTX2_MAX_UNICAST_FLOWS;
+					flow_cfg->ucast_flt_cnt;
 	pfvf->flags |= OTX2_FLAG_UCAST_FLTR_SUPPORT;
 
 	/* Check if NPC_DMAC field is supported
@@ -255,6 +253,7 @@ static int otx2_mcam_entry_init(struct otx2_nic *pfvf)
 	refcount_set(&flow_cfg->mark_flows, 1);
 	return 0;
 }
+EXPORT_SYMBOL(otx2_mcam_entry_init);
 
 /* TODO : revisit on size */
 #define OTX2_DMAC_FLTR_BITMAP_SZ (4 * 2048 + 32)
@@ -302,6 +301,8 @@ int otx2_mcam_flow_init(struct otx2_nic *pf)
 	INIT_LIST_HEAD(&pf->flow_cfg->flow_list);
 	INIT_LIST_HEAD(&pf->flow_cfg->flow_list_tc);
 
+	pf->flow_cfg->ucast_flt_cnt = OTX2_DEFAULT_UNICAST_FLOWS;
+
 	/* Allocate bare minimum number of MCAM entries needed for
 	 * unicast and ntuple filters.
 	 */
@@ -314,7 +315,7 @@ int otx2_mcam_flow_init(struct otx2_nic *pf)
 		return 0;
 
 	pf->mac_table = devm_kzalloc(pf->dev, sizeof(struct otx2_mac_table)
-					* OTX2_MAX_UNICAST_FLOWS, GFP_KERNEL);
+					* pf->flow_cfg->ucast_flt_cnt, GFP_KERNEL);
 	if (!pf->mac_table)
 		return -ENOMEM;
 
@@ -356,7 +357,7 @@ static int otx2_do_add_macfilter(struct otx2_nic *pf, const u8 *mac)
 		return -ENOMEM;
 
 	/* dont have free mcam entries or uc list is greater than alloted */
-	if (netdev_uc_count(pf->netdev) > OTX2_MAX_UNICAST_FLOWS)
+	if (netdev_uc_count(pf->netdev) > pf->flow_cfg->ucast_flt_cnt)
 		return -ENOMEM;
 
 	mutex_lock(&pf->mbox.lock);
@@ -367,7 +368,7 @@ static int otx2_do_add_macfilter(struct otx2_nic *pf, const u8 *mac)
 	}
 
 	/* unicast offset starts with 32 0..31 for ntuple */
-	for (i = 0; i <  OTX2_MAX_UNICAST_FLOWS; i++) {
+	for (i = 0; i <  pf->flow_cfg->ucast_flt_cnt; i++) {
 		if (pf->mac_table[i].inuse)
 			continue;
 		ether_addr_copy(pf->mac_table[i].addr, mac);
@@ -410,7 +411,7 @@ static bool otx2_get_mcamentry_for_mac(struct otx2_nic *pf, const u8 *mac,
 {
 	int i;
 
-	for (i = 0; i < OTX2_MAX_UNICAST_FLOWS; i++) {
+	for (i = 0; i < pf->flow_cfg->ucast_flt_cnt; i++) {
 		if (!pf->mac_table[i].inuse)
 			continue;
 
@@ -1394,6 +1395,7 @@ int otx2_destroy_mcam_flows(struct otx2_nic *pfvf)
 	}
 
 	pfvf->flags &= ~OTX2_FLAG_MCAM_ENTRIES_ALLOC;
+	flow_cfg->max_flows = 0;
 	mutex_unlock(&pfvf->mbox.lock);
 
 	return 0;
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index f5bce3e326cc..ff05ea20409a 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -1714,7 +1714,7 @@ static void otx2_do_set_rx_mode(struct otx2_nic *pf)
 		return;
 
 	if ((netdev->flags & IFF_PROMISC) ||
-	    (netdev_uc_count(netdev) > OTX2_MAX_UNICAST_FLOWS)) {
+	    (netdev_uc_count(netdev) > pf->flow_cfg->ucast_flt_cnt)) {
 		promisc = true;
 	}
 
-- 
cgit v1.2.3-58-ga151


From a744e2d03a91507646ffff8a03a19a2f34a6798a Mon Sep 17 00:00:00 2001
From: James Chapman <jchapman@katalix.com>
Date: Thu, 20 Jun 2024 12:22:37 +0100
Subject: l2tp: remove unused list_head member in l2tp_tunnel

Remove an unused variable in struct l2tp_tunnel which was left behind
by commit c4d48a58f32c5 ("l2tp: convert l2tp_tunnel_list to idr").

Signed-off-by: James Chapman <jchapman@katalix.com>
Reviewed-by: Tom Parkin <tparkin@katalix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/l2tp/l2tp_core.c | 2 --
 net/l2tp/l2tp_core.h | 1 -
 2 files changed, 3 deletions(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index 88a34db265d8..69f8c9f5cdc7 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -1462,8 +1462,6 @@ int l2tp_tunnel_create(int fd, int version, u32 tunnel_id, u32 peer_tunnel_id,
 	/* Init delete workqueue struct */
 	INIT_WORK(&tunnel->del_work, l2tp_tunnel_del_work);
 
-	INIT_LIST_HEAD(&tunnel->list);
-
 	err = 0;
 err:
 	if (tunnelp)
diff --git a/net/l2tp/l2tp_core.h b/net/l2tp/l2tp_core.h
index 91ebf0a3f499..54dfba1eb91c 100644
--- a/net/l2tp/l2tp_core.h
+++ b/net/l2tp/l2tp_core.h
@@ -174,7 +174,6 @@ struct l2tp_tunnel {
 	enum l2tp_encap_type	encap;
 	struct l2tp_stats	stats;
 
-	struct list_head	list;		/* list node on per-namespace list of tunnels */
 	struct net		*l2tp_net;	/* the net we belong to */
 
 	refcount_t		ref_count;
-- 
cgit v1.2.3-58-ga151


From aa5e17e1f5ecb68d3f67a069f7345dbf1a8f274f Mon Sep 17 00:00:00 2001
From: James Chapman <jchapman@katalix.com>
Date: Thu, 20 Jun 2024 12:22:38 +0100
Subject: l2tp: store l2tpv3 sessions in per-net IDR

L2TPv3 sessions are currently held in one of two fixed-size hash
lists: either a per-net hashlist (IP-encap), or a per-tunnel hashlist
(UDP-encap), keyed by the L2TPv3 32-bit session_id.

In order to lookup L2TPv3 sessions in UDP-encap tunnels efficiently
without finding the tunnel first via sk_user_data, UDP sessions are
now kept in a per-net session list, keyed by session ID. Convert the
existing per-net hashlist to use an IDR for better performance when
there are many sessions and have L2TPv3 UDP sessions use the same IDR.

Although the L2TPv3 RFC states that the session ID alone identifies
the session, our implementation has allowed the same session ID to be
used in different L2TP UDP tunnels. To retain support for this, a new
per-net session hashtable is used, keyed by the sock and session
ID. If on creating a new session, a session already exists with that
ID in the IDR, the colliding sessions are added to the new hashtable
and the existing IDR entry is flagged. When looking up sessions, the
approach is to first check the IDR and if no unflagged match is found,
check the new hashtable. The sock is made available to session getters
where session ID collisions are to be considered. In this way, the new
hashtable is used only for session ID collisions so can be kept small.

For managing session removal, we need a list of colliding sessions
matching a given ID in order to update or remove the IDR entry of the
ID. This is necessary to detect session ID collisions when future
sessions are created. The list head is allocated on first collision
of a given ID and refcounted.

Signed-off-by: James Chapman <jchapman@katalix.com>
Reviewed-by: Tom Parkin <tparkin@katalix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/l2tp/l2tp_core.c | 240 +++++++++++++++++++++++++++++++++++++--------------
 net/l2tp/l2tp_core.h |  18 ++--
 net/l2tp/l2tp_ip.c   |   2 +-
 net/l2tp/l2tp_ip6.c  |   2 +-
 4 files changed, 188 insertions(+), 74 deletions(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index 69f8c9f5cdc7..d6bffdb16466 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -107,11 +107,17 @@ struct l2tp_net {
 	/* Lock for write access to l2tp_tunnel_idr */
 	spinlock_t l2tp_tunnel_idr_lock;
 	struct idr l2tp_tunnel_idr;
-	struct hlist_head l2tp_session_hlist[L2TP_HASH_SIZE_2];
-	/* Lock for write access to l2tp_session_hlist */
-	spinlock_t l2tp_session_hlist_lock;
+	/* Lock for write access to l2tp_v3_session_idr/htable */
+	spinlock_t l2tp_session_idr_lock;
+	struct idr l2tp_v3_session_idr;
+	struct hlist_head l2tp_v3_session_htable[16];
 };
 
+static inline unsigned long l2tp_v3_session_hashkey(struct sock *sk, u32 session_id)
+{
+	return ((unsigned long)sk) + session_id;
+}
+
 #if IS_ENABLED(CONFIG_IPV6)
 static bool l2tp_sk_is_v6(struct sock *sk)
 {
@@ -125,17 +131,6 @@ static inline struct l2tp_net *l2tp_pernet(const struct net *net)
 	return net_generic(net, l2tp_net_id);
 }
 
-/* Session hash global list for L2TPv3.
- * The session_id SHOULD be random according to RFC3931, but several
- * L2TP implementations use incrementing session_ids.  So we do a real
- * hash on the session_id, rather than a simple bitmask.
- */
-static inline struct hlist_head *
-l2tp_session_id_hash_2(struct l2tp_net *pn, u32 session_id)
-{
-	return &pn->l2tp_session_hlist[hash_32(session_id, L2TP_HASH_BITS_2)];
-}
-
 /* Session hash list.
  * The session_id SHOULD be random according to RFC2661, but several
  * L2TP implementations (Cisco and Microsoft) use incrementing
@@ -262,26 +257,40 @@ struct l2tp_session *l2tp_tunnel_get_session(struct l2tp_tunnel *tunnel,
 }
 EXPORT_SYMBOL_GPL(l2tp_tunnel_get_session);
 
-struct l2tp_session *l2tp_session_get(const struct net *net, u32 session_id)
+struct l2tp_session *l2tp_v3_session_get(const struct net *net, struct sock *sk, u32 session_id)
 {
-	struct hlist_head *session_list;
+	const struct l2tp_net *pn = l2tp_pernet(net);
 	struct l2tp_session *session;
 
-	session_list = l2tp_session_id_hash_2(l2tp_pernet(net), session_id);
-
 	rcu_read_lock_bh();
-	hlist_for_each_entry_rcu(session, session_list, global_hlist)
-		if (session->session_id == session_id) {
-			l2tp_session_inc_refcount(session);
-			rcu_read_unlock_bh();
+	session = idr_find(&pn->l2tp_v3_session_idr, session_id);
+	if (session && !hash_hashed(&session->hlist) &&
+	    refcount_inc_not_zero(&session->ref_count)) {
+		rcu_read_unlock_bh();
+		return session;
+	}
 
-			return session;
+	/* If we get here and session is non-NULL, the session_id
+	 * collides with one in another tunnel. If sk is non-NULL,
+	 * find the session matching sk.
+	 */
+	if (session && sk) {
+		unsigned long key = l2tp_v3_session_hashkey(sk, session->session_id);
+
+		hash_for_each_possible_rcu(pn->l2tp_v3_session_htable, session,
+					   hlist, key) {
+			if (session->tunnel->sock == sk &&
+			    refcount_inc_not_zero(&session->ref_count)) {
+				rcu_read_unlock_bh();
+				return session;
+			}
 		}
+	}
 	rcu_read_unlock_bh();
 
 	return NULL;
 }
-EXPORT_SYMBOL_GPL(l2tp_session_get);
+EXPORT_SYMBOL_GPL(l2tp_v3_session_get);
 
 struct l2tp_session *l2tp_session_get_nth(struct l2tp_tunnel *tunnel, int nth)
 {
@@ -313,12 +322,12 @@ struct l2tp_session *l2tp_session_get_by_ifname(const struct net *net,
 						const char *ifname)
 {
 	struct l2tp_net *pn = l2tp_pernet(net);
-	int hash;
+	unsigned long session_id, tmp;
 	struct l2tp_session *session;
 
 	rcu_read_lock_bh();
-	for (hash = 0; hash < L2TP_HASH_SIZE_2; hash++) {
-		hlist_for_each_entry_rcu(session, &pn->l2tp_session_hlist[hash], global_hlist) {
+	idr_for_each_entry_ul(&pn->l2tp_v3_session_idr, session, tmp, session_id) {
+		if (session) {
 			if (!strcmp(session->ifname, ifname)) {
 				l2tp_session_inc_refcount(session);
 				rcu_read_unlock_bh();
@@ -334,13 +343,106 @@ struct l2tp_session *l2tp_session_get_by_ifname(const struct net *net,
 }
 EXPORT_SYMBOL_GPL(l2tp_session_get_by_ifname);
 
+static void l2tp_session_coll_list_add(struct l2tp_session_coll_list *clist,
+				       struct l2tp_session *session)
+{
+	l2tp_session_inc_refcount(session);
+	WARN_ON_ONCE(session->coll_list);
+	session->coll_list = clist;
+	spin_lock(&clist->lock);
+	list_add(&session->clist, &clist->list);
+	spin_unlock(&clist->lock);
+}
+
+static int l2tp_session_collision_add(struct l2tp_net *pn,
+				      struct l2tp_session *session1,
+				      struct l2tp_session *session2)
+{
+	struct l2tp_session_coll_list *clist;
+
+	lockdep_assert_held(&pn->l2tp_session_idr_lock);
+
+	if (!session2)
+		return -EEXIST;
+
+	/* If existing session is in IP-encap tunnel, refuse new session */
+	if (session2->tunnel->encap == L2TP_ENCAPTYPE_IP)
+		return -EEXIST;
+
+	clist = session2->coll_list;
+	if (!clist) {
+		/* First collision. Allocate list to manage the collided sessions
+		 * and add the existing session to the list.
+		 */
+		clist = kmalloc(sizeof(*clist), GFP_ATOMIC);
+		if (!clist)
+			return -ENOMEM;
+
+		spin_lock_init(&clist->lock);
+		INIT_LIST_HEAD(&clist->list);
+		refcount_set(&clist->ref_count, 1);
+		l2tp_session_coll_list_add(clist, session2);
+	}
+
+	/* If existing session isn't already in the session hlist, add it. */
+	if (!hash_hashed(&session2->hlist))
+		hash_add(pn->l2tp_v3_session_htable, &session2->hlist,
+			 session2->hlist_key);
+
+	/* Add new session to the hlist and collision list */
+	hash_add(pn->l2tp_v3_session_htable, &session1->hlist,
+		 session1->hlist_key);
+	refcount_inc(&clist->ref_count);
+	l2tp_session_coll_list_add(clist, session1);
+
+	return 0;
+}
+
+static void l2tp_session_collision_del(struct l2tp_net *pn,
+				       struct l2tp_session *session)
+{
+	struct l2tp_session_coll_list *clist = session->coll_list;
+	unsigned long session_key = session->session_id;
+	struct l2tp_session *session2;
+
+	lockdep_assert_held(&pn->l2tp_session_idr_lock);
+
+	hash_del(&session->hlist);
+
+	if (clist) {
+		/* Remove session from its collision list. If there
+		 * are other sessions with the same ID, replace this
+		 * session's IDR entry with that session, otherwise
+		 * remove the IDR entry. If this is the last session,
+		 * the collision list data is freed.
+		 */
+		spin_lock(&clist->lock);
+		list_del_init(&session->clist);
+		session2 = list_first_entry_or_null(&clist->list, struct l2tp_session, clist);
+		if (session2) {
+			void *old = idr_replace(&pn->l2tp_v3_session_idr, session2, session_key);
+
+			WARN_ON_ONCE(IS_ERR_VALUE(old));
+		} else {
+			void *removed = idr_remove(&pn->l2tp_v3_session_idr, session_key);
+
+			WARN_ON_ONCE(removed != session);
+		}
+		session->coll_list = NULL;
+		spin_unlock(&clist->lock);
+		if (refcount_dec_and_test(&clist->ref_count))
+			kfree(clist);
+		l2tp_session_dec_refcount(session);
+	}
+}
+
 int l2tp_session_register(struct l2tp_session *session,
 			  struct l2tp_tunnel *tunnel)
 {
+	struct l2tp_net *pn = l2tp_pernet(tunnel->l2tp_net);
 	struct l2tp_session *session_walk;
-	struct hlist_head *g_head;
 	struct hlist_head *head;
-	struct l2tp_net *pn;
+	u32 session_key;
 	int err;
 
 	head = l2tp_session_id_hash(tunnel, session->session_id);
@@ -358,39 +460,45 @@ int l2tp_session_register(struct l2tp_session *session,
 		}
 
 	if (tunnel->version == L2TP_HDR_VER_3) {
-		pn = l2tp_pernet(tunnel->l2tp_net);
-		g_head = l2tp_session_id_hash_2(pn, session->session_id);
-
-		spin_lock_bh(&pn->l2tp_session_hlist_lock);
-
+		session_key = session->session_id;
+		spin_lock_bh(&pn->l2tp_session_idr_lock);
+		err = idr_alloc_u32(&pn->l2tp_v3_session_idr, NULL,
+				    &session_key, session_key, GFP_ATOMIC);
 		/* IP encap expects session IDs to be globally unique, while
-		 * UDP encap doesn't.
+		 * UDP encap doesn't. This isn't per the RFC, which says that
+		 * sessions are identified only by the session ID, but is to
+		 * support existing userspace which depends on it.
 		 */
-		hlist_for_each_entry(session_walk, g_head, global_hlist)
-			if (session_walk->session_id == session->session_id &&
-			    (session_walk->tunnel->encap == L2TP_ENCAPTYPE_IP ||
-			     tunnel->encap == L2TP_ENCAPTYPE_IP)) {
-				err = -EEXIST;
-				goto err_tlock_pnlock;
-			}
+		if (err == -ENOSPC && tunnel->encap == L2TP_ENCAPTYPE_UDP) {
+			struct l2tp_session *session2;
 
-		l2tp_tunnel_inc_refcount(tunnel);
-		hlist_add_head_rcu(&session->global_hlist, g_head);
-
-		spin_unlock_bh(&pn->l2tp_session_hlist_lock);
-	} else {
-		l2tp_tunnel_inc_refcount(tunnel);
+			session2 = idr_find(&pn->l2tp_v3_session_idr,
+					    session_key);
+			err = l2tp_session_collision_add(pn, session, session2);
+		}
+		spin_unlock_bh(&pn->l2tp_session_idr_lock);
+		if (err == -ENOSPC)
+			err = -EEXIST;
 	}
 
+	if (err)
+		goto err_tlock;
+
+	l2tp_tunnel_inc_refcount(tunnel);
+
 	hlist_add_head_rcu(&session->hlist, head);
 	spin_unlock_bh(&tunnel->hlist_lock);
 
+	if (tunnel->version == L2TP_HDR_VER_3) {
+		spin_lock_bh(&pn->l2tp_session_idr_lock);
+		idr_replace(&pn->l2tp_v3_session_idr, session, session_key);
+		spin_unlock_bh(&pn->l2tp_session_idr_lock);
+	}
+
 	trace_register_session(session);
 
 	return 0;
 
-err_tlock_pnlock:
-	spin_unlock_bh(&pn->l2tp_session_hlist_lock);
 err_tlock:
 	spin_unlock_bh(&tunnel->hlist_lock);
 
@@ -1218,13 +1326,19 @@ static void l2tp_session_unhash(struct l2tp_session *session)
 		hlist_del_init_rcu(&session->hlist);
 		spin_unlock_bh(&tunnel->hlist_lock);
 
-		/* For L2TPv3 we have a per-net hash: remove from there, too */
-		if (tunnel->version != L2TP_HDR_VER_2) {
+		/* For L2TPv3 we have a per-net IDR: remove from there, too */
+		if (tunnel->version == L2TP_HDR_VER_3) {
 			struct l2tp_net *pn = l2tp_pernet(tunnel->l2tp_net);
-
-			spin_lock_bh(&pn->l2tp_session_hlist_lock);
-			hlist_del_init_rcu(&session->global_hlist);
-			spin_unlock_bh(&pn->l2tp_session_hlist_lock);
+			struct l2tp_session *removed = session;
+
+			spin_lock_bh(&pn->l2tp_session_idr_lock);
+			if (hash_hashed(&session->hlist))
+				l2tp_session_collision_del(pn, session);
+			else
+				removed = idr_remove(&pn->l2tp_v3_session_idr,
+						     session->session_id);
+			WARN_ON_ONCE(removed && removed != session);
+			spin_unlock_bh(&pn->l2tp_session_idr_lock);
 		}
 
 		synchronize_rcu();
@@ -1649,8 +1763,9 @@ struct l2tp_session *l2tp_session_create(int priv_size, struct l2tp_tunnel *tunn
 
 		skb_queue_head_init(&session->reorder_q);
 
+		session->hlist_key = l2tp_v3_session_hashkey(tunnel->sock, session->session_id);
 		INIT_HLIST_NODE(&session->hlist);
-		INIT_HLIST_NODE(&session->global_hlist);
+		INIT_LIST_HEAD(&session->clist);
 
 		if (cfg) {
 			session->pwtype = cfg->pw_type;
@@ -1683,15 +1798,12 @@ EXPORT_SYMBOL_GPL(l2tp_session_create);
 static __net_init int l2tp_init_net(struct net *net)
 {
 	struct l2tp_net *pn = net_generic(net, l2tp_net_id);
-	int hash;
 
 	idr_init(&pn->l2tp_tunnel_idr);
 	spin_lock_init(&pn->l2tp_tunnel_idr_lock);
 
-	for (hash = 0; hash < L2TP_HASH_SIZE_2; hash++)
-		INIT_HLIST_HEAD(&pn->l2tp_session_hlist[hash]);
-
-	spin_lock_init(&pn->l2tp_session_hlist_lock);
+	idr_init(&pn->l2tp_v3_session_idr);
+	spin_lock_init(&pn->l2tp_session_idr_lock);
 
 	return 0;
 }
@@ -1701,7 +1813,6 @@ static __net_exit void l2tp_exit_net(struct net *net)
 	struct l2tp_net *pn = l2tp_pernet(net);
 	struct l2tp_tunnel *tunnel = NULL;
 	unsigned long tunnel_id, tmp;
-	int hash;
 
 	rcu_read_lock_bh();
 	idr_for_each_entry_ul(&pn->l2tp_tunnel_idr, tunnel, tmp, tunnel_id) {
@@ -1714,8 +1825,7 @@ static __net_exit void l2tp_exit_net(struct net *net)
 		flush_workqueue(l2tp_wq);
 	rcu_barrier();
 
-	for (hash = 0; hash < L2TP_HASH_SIZE_2; hash++)
-		WARN_ON_ONCE(!hlist_empty(&pn->l2tp_session_hlist[hash]));
+	idr_destroy(&pn->l2tp_v3_session_idr);
 	idr_destroy(&pn->l2tp_tunnel_idr);
 }
 
diff --git a/net/l2tp/l2tp_core.h b/net/l2tp/l2tp_core.h
index 54dfba1eb91c..bfccc4ca2644 100644
--- a/net/l2tp/l2tp_core.h
+++ b/net/l2tp/l2tp_core.h
@@ -23,10 +23,6 @@
 #define L2TP_HASH_BITS	4
 #define L2TP_HASH_SIZE	BIT(L2TP_HASH_BITS)
 
-/* System-wide session hash table size */
-#define L2TP_HASH_BITS_2	8
-#define L2TP_HASH_SIZE_2	BIT(L2TP_HASH_BITS_2)
-
 struct sk_buff;
 
 struct l2tp_stats {
@@ -61,6 +57,12 @@ struct l2tp_session_cfg {
 	char			*ifname;
 };
 
+struct l2tp_session_coll_list {
+	spinlock_t lock;	/* for access to list */
+	struct list_head list;
+	refcount_t ref_count;
+};
+
 /* Represents a session (pseudowire) instance.
  * Tracks runtime state including cookies, dataplane packet sequencing, and IO statistics.
  * Is linked into a per-tunnel session hashlist; and in the case of an L2TPv3 session into
@@ -88,8 +90,11 @@ struct l2tp_session {
 	u32			nr_oos;		/* NR of last OOS packet */
 	int			nr_oos_count;	/* for OOS recovery */
 	int			nr_oos_count_max;
-	struct hlist_node	hlist;		/* hash list node */
 	refcount_t		ref_count;
+	struct hlist_node	hlist;		/* per-net session hlist */
+	unsigned long		hlist_key;	/* key for session hlist */
+	struct l2tp_session_coll_list *coll_list; /* session collision list */
+	struct list_head	clist;		/* for coll_list */
 
 	char			name[L2TP_SESSION_NAME_MAX]; /* for logging */
 	char			ifname[IFNAMSIZ];
@@ -102,7 +107,6 @@ struct l2tp_session {
 	int			reorder_skip;	/* set if skip to next nr */
 	enum l2tp_pwtype	pwtype;
 	struct l2tp_stats	stats;
-	struct hlist_node	global_hlist;	/* global hash list node */
 
 	/* Session receive handler for data packets.
 	 * Each pseudowire implementation should implement this callback in order to
@@ -226,7 +230,7 @@ struct l2tp_tunnel *l2tp_tunnel_get_nth(const struct net *net, int nth);
 struct l2tp_session *l2tp_tunnel_get_session(struct l2tp_tunnel *tunnel,
 					     u32 session_id);
 
-struct l2tp_session *l2tp_session_get(const struct net *net, u32 session_id);
+struct l2tp_session *l2tp_v3_session_get(const struct net *net, struct sock *sk, u32 session_id);
 struct l2tp_session *l2tp_session_get_nth(struct l2tp_tunnel *tunnel, int nth);
 struct l2tp_session *l2tp_session_get_by_ifname(const struct net *net,
 						const char *ifname);
diff --git a/net/l2tp/l2tp_ip.c b/net/l2tp/l2tp_ip.c
index 19c8cc5289d5..e48aa177d74c 100644
--- a/net/l2tp/l2tp_ip.c
+++ b/net/l2tp/l2tp_ip.c
@@ -140,7 +140,7 @@ static int l2tp_ip_recv(struct sk_buff *skb)
 	}
 
 	/* Ok, this is a data packet. Lookup the session. */
-	session = l2tp_session_get(net, session_id);
+	session = l2tp_v3_session_get(net, NULL, session_id);
 	if (!session)
 		goto discard;
 
diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c
index 8780ec64f376..d217ff1f229e 100644
--- a/net/l2tp/l2tp_ip6.c
+++ b/net/l2tp/l2tp_ip6.c
@@ -150,7 +150,7 @@ static int l2tp_ip6_recv(struct sk_buff *skb)
 	}
 
 	/* Ok, this is a data packet. Lookup the session. */
-	session = l2tp_session_get(net, session_id);
+	session = l2tp_v3_session_get(net, NULL, session_id);
 	if (!session)
 		goto discard;
 
-- 
cgit v1.2.3-58-ga151


From 2a3339f6c9636aa39f2493865e4664df1ef2baed Mon Sep 17 00:00:00 2001
From: James Chapman <jchapman@katalix.com>
Date: Thu, 20 Jun 2024 12:22:39 +0100
Subject: l2tp: store l2tpv2 sessions in per-net IDR

L2TPv2 sessions are currently kept in a per-tunnel hashlist, keyed by
16-bit session_id. When handling received L2TPv2 packets, we need to
first derive the tunnel using the 16-bit tunnel_id or sock, then
lookup the session in a per-tunnel hlist using the 16-bit session_id.

We want to avoid using sk_user_data in the datapath and double lookups
on every packet. So instead, use a per-net IDR to hold L2TPv2
sessions, keyed by a 32-bit value derived from the 16-bit tunnel_id
and session_id. This will allow the L2TPv2 UDP receive datapath to
lookup a session with a single lookup without deriving the tunnel
first.

L2TPv2 sessions are held in their own IDR to avoid potential
key collisions with L2TPv3 sessions.

Signed-off-by: James Chapman <jchapman@katalix.com>
Reviewed-by: Tom Parkin <tparkin@katalix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/l2tp/l2tp_core.c | 70 +++++++++++++++++++++++++++++++++++++++++-----------
 net/l2tp/l2tp_core.h |  1 +
 2 files changed, 56 insertions(+), 15 deletions(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index d6bffdb16466..6f30b347fd46 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -107,12 +107,18 @@ struct l2tp_net {
 	/* Lock for write access to l2tp_tunnel_idr */
 	spinlock_t l2tp_tunnel_idr_lock;
 	struct idr l2tp_tunnel_idr;
-	/* Lock for write access to l2tp_v3_session_idr/htable */
+	/* Lock for write access to l2tp_v[23]_session_idr/htable */
 	spinlock_t l2tp_session_idr_lock;
+	struct idr l2tp_v2_session_idr;
 	struct idr l2tp_v3_session_idr;
 	struct hlist_head l2tp_v3_session_htable[16];
 };
 
+static inline u32 l2tp_v2_session_key(u16 tunnel_id, u16 session_id)
+{
+	return ((u32)tunnel_id) << 16 | session_id;
+}
+
 static inline unsigned long l2tp_v3_session_hashkey(struct sock *sk, u32 session_id)
 {
 	return ((unsigned long)sk) + session_id;
@@ -292,6 +298,24 @@ struct l2tp_session *l2tp_v3_session_get(const struct net *net, struct sock *sk,
 }
 EXPORT_SYMBOL_GPL(l2tp_v3_session_get);
 
+struct l2tp_session *l2tp_v2_session_get(const struct net *net, u16 tunnel_id, u16 session_id)
+{
+	u32 session_key = l2tp_v2_session_key(tunnel_id, session_id);
+	const struct l2tp_net *pn = l2tp_pernet(net);
+	struct l2tp_session *session;
+
+	rcu_read_lock_bh();
+	session = idr_find(&pn->l2tp_v2_session_idr, session_key);
+	if (session && refcount_inc_not_zero(&session->ref_count)) {
+		rcu_read_unlock_bh();
+		return session;
+	}
+	rcu_read_unlock_bh();
+
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(l2tp_v2_session_get);
+
 struct l2tp_session *l2tp_session_get_nth(struct l2tp_tunnel *tunnel, int nth)
 {
 	int hash;
@@ -477,23 +501,32 @@ int l2tp_session_register(struct l2tp_session *session,
 			err = l2tp_session_collision_add(pn, session, session2);
 		}
 		spin_unlock_bh(&pn->l2tp_session_idr_lock);
-		if (err == -ENOSPC)
-			err = -EEXIST;
+	} else {
+		session_key = l2tp_v2_session_key(tunnel->tunnel_id,
+						  session->session_id);
+		spin_lock_bh(&pn->l2tp_session_idr_lock);
+		err = idr_alloc_u32(&pn->l2tp_v2_session_idr, NULL,
+				    &session_key, session_key, GFP_ATOMIC);
+		spin_unlock_bh(&pn->l2tp_session_idr_lock);
 	}
 
-	if (err)
+	if (err) {
+		if (err == -ENOSPC)
+			err = -EEXIST;
 		goto err_tlock;
+	}
 
 	l2tp_tunnel_inc_refcount(tunnel);
 
 	hlist_add_head_rcu(&session->hlist, head);
 	spin_unlock_bh(&tunnel->hlist_lock);
 
-	if (tunnel->version == L2TP_HDR_VER_3) {
-		spin_lock_bh(&pn->l2tp_session_idr_lock);
+	spin_lock_bh(&pn->l2tp_session_idr_lock);
+	if (tunnel->version == L2TP_HDR_VER_3)
 		idr_replace(&pn->l2tp_v3_session_idr, session, session_key);
-		spin_unlock_bh(&pn->l2tp_session_idr_lock);
-	}
+	else
+		idr_replace(&pn->l2tp_v2_session_idr, session, session_key);
+	spin_unlock_bh(&pn->l2tp_session_idr_lock);
 
 	trace_register_session(session);
 
@@ -1321,25 +1354,30 @@ static void l2tp_session_unhash(struct l2tp_session *session)
 
 	/* Remove the session from core hashes */
 	if (tunnel) {
+		struct l2tp_net *pn = l2tp_pernet(tunnel->l2tp_net);
+		struct l2tp_session *removed = session;
+
 		/* Remove from the per-tunnel hash */
 		spin_lock_bh(&tunnel->hlist_lock);
 		hlist_del_init_rcu(&session->hlist);
 		spin_unlock_bh(&tunnel->hlist_lock);
 
-		/* For L2TPv3 we have a per-net IDR: remove from there, too */
+		/* Remove from per-net IDR */
+		spin_lock_bh(&pn->l2tp_session_idr_lock);
 		if (tunnel->version == L2TP_HDR_VER_3) {
-			struct l2tp_net *pn = l2tp_pernet(tunnel->l2tp_net);
-			struct l2tp_session *removed = session;
-
-			spin_lock_bh(&pn->l2tp_session_idr_lock);
 			if (hash_hashed(&session->hlist))
 				l2tp_session_collision_del(pn, session);
 			else
 				removed = idr_remove(&pn->l2tp_v3_session_idr,
 						     session->session_id);
-			WARN_ON_ONCE(removed && removed != session);
-			spin_unlock_bh(&pn->l2tp_session_idr_lock);
+		} else {
+			u32 session_key = l2tp_v2_session_key(tunnel->tunnel_id,
+							      session->session_id);
+			removed = idr_remove(&pn->l2tp_v2_session_idr,
+					     session_key);
 		}
+		WARN_ON_ONCE(removed && removed != session);
+		spin_unlock_bh(&pn->l2tp_session_idr_lock);
 
 		synchronize_rcu();
 	}
@@ -1802,6 +1840,7 @@ static __net_init int l2tp_init_net(struct net *net)
 	idr_init(&pn->l2tp_tunnel_idr);
 	spin_lock_init(&pn->l2tp_tunnel_idr_lock);
 
+	idr_init(&pn->l2tp_v2_session_idr);
 	idr_init(&pn->l2tp_v3_session_idr);
 	spin_lock_init(&pn->l2tp_session_idr_lock);
 
@@ -1825,6 +1864,7 @@ static __net_exit void l2tp_exit_net(struct net *net)
 		flush_workqueue(l2tp_wq);
 	rcu_barrier();
 
+	idr_destroy(&pn->l2tp_v2_session_idr);
 	idr_destroy(&pn->l2tp_v3_session_idr);
 	idr_destroy(&pn->l2tp_tunnel_idr);
 }
diff --git a/net/l2tp/l2tp_core.h b/net/l2tp/l2tp_core.h
index bfccc4ca2644..d80f15f5b9fc 100644
--- a/net/l2tp/l2tp_core.h
+++ b/net/l2tp/l2tp_core.h
@@ -231,6 +231,7 @@ struct l2tp_session *l2tp_tunnel_get_session(struct l2tp_tunnel *tunnel,
 					     u32 session_id);
 
 struct l2tp_session *l2tp_v3_session_get(const struct net *net, struct sock *sk, u32 session_id);
+struct l2tp_session *l2tp_v2_session_get(const struct net *net, u16 tunnel_id, u16 session_id);
 struct l2tp_session *l2tp_session_get_nth(struct l2tp_tunnel *tunnel, int nth);
 struct l2tp_session *l2tp_session_get_by_ifname(const struct net *net,
 						const char *ifname);
-- 
cgit v1.2.3-58-ga151


From ff6a2ac23cb027ff9980d633412db17d5f7a1e7c Mon Sep 17 00:00:00 2001
From: James Chapman <jchapman@katalix.com>
Date: Thu, 20 Jun 2024 12:22:40 +0100
Subject: l2tp: refactor udp recv to lookup to not use sk_user_data

Modify UDP decap to not use the tunnel pointer which comes from the
sock's sk_user_data when parsing the L2TP header. By looking up the
destination session using only the packet contents we avoid potential
UDP 5-tuple aliasing issues which arise from depending on the socket
that received the packet.

Drop the useless error messages on short packet or on failing to find
a session since the tunnel pointer might point to a different tunnel
if multiple sockets use the same 5-tuple.

Short packets (those not big enough to contain an L2TP header) are no
longer counted in the tunnel's invalid counter because we can't derive
the tunnel until we parse the l2tp header to lookup the session.

l2tp_udp_encap_recv was a small wrapper around l2tp_udp_recv_core which
used sk_user_data to derive a tunnel pointer in an RCU-safe way. But
we no longer need the tunnel pointer, so remove that code and combine
the two functions.

Signed-off-by: James Chapman <jchapman@katalix.com>
Reviewed-by: Tom Parkin <tparkin@katalix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/l2tp/l2tp_core.c | 96 ++++++++++++----------------------------------------
 1 file changed, 21 insertions(+), 75 deletions(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index 6f30b347fd46..2c6378a9f384 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -926,19 +926,14 @@ static void l2tp_session_queue_purge(struct l2tp_session *session)
 	}
 }
 
-/* Internal UDP receive frame. Do the real work of receiving an L2TP data frame
- * here. The skb is not on a list when we get here.
- * Returns 0 if the packet was a data packet and was successfully passed on.
- * Returns 1 if the packet was not a good data packet and could not be
- * forwarded.  All such packets are passed up to userspace to deal with.
- */
-static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb)
+/* UDP encapsulation receive handler. See net/ipv4/udp.c for details. */
+int l2tp_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 {
 	struct l2tp_session *session = NULL;
-	struct l2tp_tunnel *orig_tunnel = tunnel;
+	struct l2tp_tunnel *tunnel = NULL;
+	struct net *net = sock_net(sk);
 	unsigned char *ptr, *optr;
 	u16 hdrflags;
-	u32 tunnel_id, session_id;
 	u16 version;
 	int length;
 
@@ -948,11 +943,8 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb)
 	__skb_pull(skb, sizeof(struct udphdr));
 
 	/* Short packet? */
-	if (!pskb_may_pull(skb, L2TP_HDR_SIZE_MAX)) {
-		pr_debug_ratelimited("%s: recv short packet (len=%d)\n",
-				     tunnel->name, skb->len);
-		goto invalid;
-	}
+	if (!pskb_may_pull(skb, L2TP_HDR_SIZE_MAX))
+		goto pass;
 
 	/* Point to L2TP header */
 	optr = skb->data;
@@ -975,6 +967,8 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb)
 	ptr += 2;
 
 	if (version == L2TP_HDR_VER_2) {
+		u16 tunnel_id, session_id;
+
 		/* If length is present, skip it */
 		if (hdrflags & L2TP_HDRFLAG_L)
 			ptr += 2;
@@ -982,49 +976,35 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb)
 		/* Extract tunnel and session ID */
 		tunnel_id = ntohs(*(__be16 *)ptr);
 		ptr += 2;
-
-		if (tunnel_id != tunnel->tunnel_id) {
-			/* We are receiving trafic for another tunnel, probably
-			 * because we have several tunnels between the same
-			 * IP/port quadruple, look it up.
-			 */
-			struct l2tp_tunnel *alt_tunnel;
-
-			alt_tunnel = l2tp_tunnel_get(tunnel->l2tp_net, tunnel_id);
-			if (!alt_tunnel)
-				goto pass;
-			tunnel = alt_tunnel;
-		}
-
 		session_id = ntohs(*(__be16 *)ptr);
 		ptr += 2;
+
+		session = l2tp_v2_session_get(net, tunnel_id, session_id);
 	} else {
+		u32 session_id;
+
 		ptr += 2;	/* skip reserved bits */
-		tunnel_id = tunnel->tunnel_id;
 		session_id = ntohl(*(__be32 *)ptr);
 		ptr += 4;
-	}
 
-	/* Check protocol version */
-	if (version != tunnel->version) {
-		pr_debug_ratelimited("%s: recv protocol version mismatch: got %d expected %d\n",
-				     tunnel->name, version, tunnel->version);
-		goto invalid;
+		session = l2tp_v3_session_get(net, sk, session_id);
 	}
 
-	/* Find the session context */
-	session = l2tp_tunnel_get_session(tunnel, session_id);
 	if (!session || !session->recv_skb) {
 		if (session)
 			l2tp_session_dec_refcount(session);
 
 		/* Not found? Pass to userspace to deal with */
-		pr_debug_ratelimited("%s: no session found (%u/%u). Passing up.\n",
-				     tunnel->name, tunnel_id, session_id);
 		goto pass;
 	}
 
-	if (tunnel->version == L2TP_HDR_VER_3 &&
+	tunnel = session->tunnel;
+
+	/* Check protocol version */
+	if (version != tunnel->version)
+		goto invalid;
+
+	if (version == L2TP_HDR_VER_3 &&
 	    l2tp_v3_ensure_opt_in_linear(session, skb, &ptr, &optr)) {
 		l2tp_session_dec_refcount(session);
 		goto invalid;
@@ -1033,9 +1013,6 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb)
 	l2tp_recv_common(session, skb, ptr, optr, hdrflags, length);
 	l2tp_session_dec_refcount(session);
 
-	if (tunnel != orig_tunnel)
-		l2tp_tunnel_dec_refcount(tunnel);
-
 	return 0;
 
 invalid:
@@ -1045,42 +1022,11 @@ pass:
 	/* Put UDP header back */
 	__skb_push(skb, sizeof(struct udphdr));
 
-	if (tunnel != orig_tunnel)
-		l2tp_tunnel_dec_refcount(tunnel);
-
-	return 1;
-}
-
-/* UDP encapsulation receive and error receive handlers.
- * See net/ipv4/udp.c for details.
- *
- * Note that these functions are called from inside an
- * RCU-protected region, but without the socket being locked.
- *
- * Hence we use rcu_dereference_sk_user_data to access the
- * tunnel data structure rather the usual l2tp_sk_to_tunnel
- * accessor function.
- */
-int l2tp_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
-{
-	struct l2tp_tunnel *tunnel;
-
-	tunnel = rcu_dereference_sk_user_data(sk);
-	if (!tunnel)
-		goto pass_up;
-	if (WARN_ON(tunnel->magic != L2TP_TUNNEL_MAGIC))
-		goto pass_up;
-
-	if (l2tp_udp_recv_core(tunnel, skb))
-		goto pass_up;
-
-	return 0;
-
-pass_up:
 	return 1;
 }
 EXPORT_SYMBOL_GPL(l2tp_udp_encap_recv);
 
+/* UDP encapsulation receive error handler. See net/ipv4/udp.c for details. */
 static void l2tp_udp_encap_err_recv(struct sock *sk, struct sk_buff *skb, int err,
 				    __be16 port, u32 info, u8 *payload)
 {
-- 
cgit v1.2.3-58-ga151


From c37e0138ca5f3be6b69c3020470aecb94eb5d773 Mon Sep 17 00:00:00 2001
From: James Chapman <jchapman@katalix.com>
Date: Thu, 20 Jun 2024 12:22:41 +0100
Subject: l2tp: don't use sk_user_data in l2tp_udp_encap_err_recv

If UDP sockets are aliased, sk might be the wrong socket. There's no
benefit to using sk_user_data to do some checks on the associated
tunnel context. Just report the error anyway, like udp core does.

Signed-off-by: James Chapman <jchapman@katalix.com>
Reviewed-by: Tom Parkin <tparkin@katalix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/l2tp/l2tp_core.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index 2c6378a9f384..cbc5de1373cd 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -1030,12 +1030,6 @@ EXPORT_SYMBOL_GPL(l2tp_udp_encap_recv);
 static void l2tp_udp_encap_err_recv(struct sock *sk, struct sk_buff *skb, int err,
 				    __be16 port, u32 info, u8 *payload)
 {
-	struct l2tp_tunnel *tunnel;
-
-	tunnel = rcu_dereference_sk_user_data(sk);
-	if (!tunnel || tunnel->fd < 0)
-		return;
-
 	sk->sk_err = err;
 	sk_error_report(sk);
 
-- 
cgit v1.2.3-58-ga151


From 5f77c18ea55601822f9c495135a5b5d4b499d647 Mon Sep 17 00:00:00 2001
From: James Chapman <jchapman@katalix.com>
Date: Thu, 20 Jun 2024 12:22:42 +0100
Subject: l2tp: use IDR for all session lookups

Add generic session getter which uses IDR. Replace all users of
l2tp_tunnel_get_session which uses the per-tunnel session list to use
the generic getter.

Signed-off-by: James Chapman <jchapman@katalix.com>
Reviewed-by: Tom Parkin <tparkin@katalix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/l2tp/l2tp_core.c    | 10 ++++++++++
 net/l2tp/l2tp_core.h    |  2 ++
 net/l2tp/l2tp_netlink.c |  6 ++++--
 net/l2tp/l2tp_ppp.c     |  6 ++++--
 4 files changed, 20 insertions(+), 4 deletions(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index cbc5de1373cd..0e826a0260fe 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -316,6 +316,16 @@ struct l2tp_session *l2tp_v2_session_get(const struct net *net, u16 tunnel_id, u
 }
 EXPORT_SYMBOL_GPL(l2tp_v2_session_get);
 
+struct l2tp_session *l2tp_session_get(const struct net *net, struct sock *sk, int pver,
+				      u32 tunnel_id, u32 session_id)
+{
+	if (pver == L2TP_HDR_VER_2)
+		return l2tp_v2_session_get(net, tunnel_id, session_id);
+	else
+		return l2tp_v3_session_get(net, sk, session_id);
+}
+EXPORT_SYMBOL_GPL(l2tp_session_get);
+
 struct l2tp_session *l2tp_session_get_nth(struct l2tp_tunnel *tunnel, int nth)
 {
 	int hash;
diff --git a/net/l2tp/l2tp_core.h b/net/l2tp/l2tp_core.h
index d80f15f5b9fc..0e7c9b0bcc1e 100644
--- a/net/l2tp/l2tp_core.h
+++ b/net/l2tp/l2tp_core.h
@@ -232,6 +232,8 @@ struct l2tp_session *l2tp_tunnel_get_session(struct l2tp_tunnel *tunnel,
 
 struct l2tp_session *l2tp_v3_session_get(const struct net *net, struct sock *sk, u32 session_id);
 struct l2tp_session *l2tp_v2_session_get(const struct net *net, u16 tunnel_id, u16 session_id);
+struct l2tp_session *l2tp_session_get(const struct net *net, struct sock *sk, int pver,
+				      u32 tunnel_id, u32 session_id);
 struct l2tp_session *l2tp_session_get_nth(struct l2tp_tunnel *tunnel, int nth);
 struct l2tp_session *l2tp_session_get_by_ifname(const struct net *net,
 						const char *ifname);
diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c
index a901fd14fe3b..d105030520f9 100644
--- a/net/l2tp/l2tp_netlink.c
+++ b/net/l2tp/l2tp_netlink.c
@@ -61,7 +61,8 @@ static struct l2tp_session *l2tp_nl_session_get(struct genl_info *info)
 		session_id = nla_get_u32(info->attrs[L2TP_ATTR_SESSION_ID]);
 		tunnel = l2tp_tunnel_get(net, tunnel_id);
 		if (tunnel) {
-			session = l2tp_tunnel_get_session(tunnel, session_id);
+			session = l2tp_session_get(net, tunnel->sock, tunnel->version,
+						   tunnel_id, session_id);
 			l2tp_tunnel_dec_refcount(tunnel);
 		}
 	}
@@ -635,7 +636,8 @@ static int l2tp_nl_cmd_session_create(struct sk_buff *skb, struct genl_info *inf
 							   &cfg);
 
 	if (ret >= 0) {
-		session = l2tp_tunnel_get_session(tunnel, session_id);
+		session = l2tp_session_get(net, tunnel->sock, tunnel->version,
+					   tunnel_id, session_id);
 		if (session) {
 			ret = l2tp_session_notify(&l2tp_nl_family, info, session,
 						  L2TP_CMD_SESSION_CREATE);
diff --git a/net/l2tp/l2tp_ppp.c b/net/l2tp/l2tp_ppp.c
index 6146e4e67bbb..3596290047b2 100644
--- a/net/l2tp/l2tp_ppp.c
+++ b/net/l2tp/l2tp_ppp.c
@@ -753,7 +753,8 @@ static int pppol2tp_connect(struct socket *sock, struct sockaddr *uservaddr,
 	if (tunnel->peer_tunnel_id == 0)
 		tunnel->peer_tunnel_id = info.peer_tunnel_id;
 
-	session = l2tp_tunnel_get_session(tunnel, info.session_id);
+	session = l2tp_session_get(sock_net(sk), tunnel->sock, tunnel->version,
+				   info.tunnel_id, info.session_id);
 	if (session) {
 		drop_refcnt = true;
 
@@ -1045,7 +1046,8 @@ static int pppol2tp_tunnel_copy_stats(struct pppol2tp_ioc_stats *stats,
 	/* If session_id is set, search the corresponding session in the
 	 * context of this tunnel and record the session's statistics.
 	 */
-	session = l2tp_tunnel_get_session(tunnel, stats->session_id);
+	session = l2tp_session_get(tunnel->l2tp_net, tunnel->sock, tunnel->version,
+				   tunnel->tunnel_id, stats->session_id);
 	if (!session)
 		return -EBADR;
 
-- 
cgit v1.2.3-58-ga151


From 8c6245af4fc5b6d244fb0f953d493e848d1e1387 Mon Sep 17 00:00:00 2001
From: James Chapman <jchapman@katalix.com>
Date: Thu, 20 Jun 2024 12:22:43 +0100
Subject: l2tp: drop the now unused l2tp_tunnel_get_session

All users of l2tp_tunnel_get_session are now gone so it can be
removed.

Signed-off-by: James Chapman <jchapman@katalix.com>
Reviewed-by: Tom Parkin <tparkin@katalix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/l2tp/l2tp_core.c | 22 ----------------------
 net/l2tp/l2tp_core.h |  2 --
 2 files changed, 24 deletions(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index 0e826a0260fe..3ce689331542 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -241,28 +241,6 @@ struct l2tp_tunnel *l2tp_tunnel_get_nth(const struct net *net, int nth)
 }
 EXPORT_SYMBOL_GPL(l2tp_tunnel_get_nth);
 
-struct l2tp_session *l2tp_tunnel_get_session(struct l2tp_tunnel *tunnel,
-					     u32 session_id)
-{
-	struct hlist_head *session_list;
-	struct l2tp_session *session;
-
-	session_list = l2tp_session_id_hash(tunnel, session_id);
-
-	rcu_read_lock_bh();
-	hlist_for_each_entry_rcu(session, session_list, hlist)
-		if (session->session_id == session_id) {
-			l2tp_session_inc_refcount(session);
-			rcu_read_unlock_bh();
-
-			return session;
-		}
-	rcu_read_unlock_bh();
-
-	return NULL;
-}
-EXPORT_SYMBOL_GPL(l2tp_tunnel_get_session);
-
 struct l2tp_session *l2tp_v3_session_get(const struct net *net, struct sock *sk, u32 session_id)
 {
 	const struct l2tp_net *pn = l2tp_pernet(net);
diff --git a/net/l2tp/l2tp_core.h b/net/l2tp/l2tp_core.h
index 0e7c9b0bcc1e..bfff69f2e0a2 100644
--- a/net/l2tp/l2tp_core.h
+++ b/net/l2tp/l2tp_core.h
@@ -227,8 +227,6 @@ void l2tp_session_dec_refcount(struct l2tp_session *session);
  */
 struct l2tp_tunnel *l2tp_tunnel_get(const struct net *net, u32 tunnel_id);
 struct l2tp_tunnel *l2tp_tunnel_get_nth(const struct net *net, int nth);
-struct l2tp_session *l2tp_tunnel_get_session(struct l2tp_tunnel *tunnel,
-					     u32 session_id);
 
 struct l2tp_session *l2tp_v3_session_get(const struct net *net, struct sock *sk, u32 session_id);
 struct l2tp_session *l2tp_v2_session_get(const struct net *net, u16 tunnel_id, u16 session_id);
-- 
cgit v1.2.3-58-ga151


From d18d3f0a24fc4a3513495892ab1a3753628b341b Mon Sep 17 00:00:00 2001
From: James Chapman <jchapman@katalix.com>
Date: Thu, 20 Jun 2024 12:22:44 +0100
Subject: l2tp: replace hlist with simple list for per-tunnel session list

The per-tunnel session list is no longer used by the
datapath. However, we still need a list of sessions in the tunnel for
l2tp_session_get_nth, which is used by management code. (An
alternative might be to walk each session IDR list, matching only
sessions of a given tunnel.)

Replace the per-tunnel hlist with a per-tunnel list. In functions
which walk a list of sessions of a tunnel, walk this list instead.

Signed-off-by: James Chapman <jchapman@katalix.com>
Reviewed-by: Tom Parkin <tparkin@katalix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/l2tp/l2tp_core.c    | 109 +++++++++++++++++-------------------------------
 net/l2tp/l2tp_core.h    |  19 ++++-----
 net/l2tp/l2tp_debugfs.c |  13 +++---
 3 files changed, 50 insertions(+), 91 deletions(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index 3ce689331542..be4bcbf291a1 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -39,7 +39,6 @@
 #include <linux/ip.h>
 #include <linux/udp.h>
 #include <linux/l2tp.h>
-#include <linux/hash.h>
 #include <linux/sort.h>
 #include <linux/file.h>
 #include <linux/nsproxy.h>
@@ -137,18 +136,6 @@ static inline struct l2tp_net *l2tp_pernet(const struct net *net)
 	return net_generic(net, l2tp_net_id);
 }
 
-/* Session hash list.
- * The session_id SHOULD be random according to RFC2661, but several
- * L2TP implementations (Cisco and Microsoft) use incrementing
- * session_ids.  So we do a real hash on the session_id, rather than a
- * simple bitmask.
- */
-static inline struct hlist_head *
-l2tp_session_id_hash(struct l2tp_tunnel *tunnel, u32 session_id)
-{
-	return &tunnel->session_hlist[hash_32(session_id, L2TP_HASH_BITS)];
-}
-
 static void l2tp_tunnel_free(struct l2tp_tunnel *tunnel)
 {
 	trace_free_tunnel(tunnel);
@@ -306,21 +293,17 @@ EXPORT_SYMBOL_GPL(l2tp_session_get);
 
 struct l2tp_session *l2tp_session_get_nth(struct l2tp_tunnel *tunnel, int nth)
 {
-	int hash;
 	struct l2tp_session *session;
 	int count = 0;
 
 	rcu_read_lock_bh();
-	for (hash = 0; hash < L2TP_HASH_SIZE; hash++) {
-		hlist_for_each_entry_rcu(session, &tunnel->session_hlist[hash], hlist) {
-			if (++count > nth) {
-				l2tp_session_inc_refcount(session);
-				rcu_read_unlock_bh();
-				return session;
-			}
+	list_for_each_entry_rcu(session, &tunnel->session_list, list) {
+		if (++count > nth) {
+			l2tp_session_inc_refcount(session);
+			rcu_read_unlock_bh();
+			return session;
 		}
 	}
-
 	rcu_read_unlock_bh();
 
 	return NULL;
@@ -334,21 +317,23 @@ struct l2tp_session *l2tp_session_get_by_ifname(const struct net *net,
 						const char *ifname)
 {
 	struct l2tp_net *pn = l2tp_pernet(net);
-	unsigned long session_id, tmp;
+	unsigned long tunnel_id, tmp;
 	struct l2tp_session *session;
+	struct l2tp_tunnel *tunnel;
 
 	rcu_read_lock_bh();
-	idr_for_each_entry_ul(&pn->l2tp_v3_session_idr, session, tmp, session_id) {
-		if (session) {
-			if (!strcmp(session->ifname, ifname)) {
-				l2tp_session_inc_refcount(session);
-				rcu_read_unlock_bh();
-
-				return session;
+	idr_for_each_entry_ul(&pn->l2tp_tunnel_idr, tunnel, tmp, tunnel_id) {
+		if (tunnel) {
+			list_for_each_entry_rcu(session, &tunnel->session_list, list) {
+				if (!strcmp(session->ifname, ifname)) {
+					l2tp_session_inc_refcount(session);
+					rcu_read_unlock_bh();
+
+					return session;
+				}
 			}
 		}
 	}
-
 	rcu_read_unlock_bh();
 
 	return NULL;
@@ -452,25 +437,15 @@ int l2tp_session_register(struct l2tp_session *session,
 			  struct l2tp_tunnel *tunnel)
 {
 	struct l2tp_net *pn = l2tp_pernet(tunnel->l2tp_net);
-	struct l2tp_session *session_walk;
-	struct hlist_head *head;
 	u32 session_key;
 	int err;
 
-	head = l2tp_session_id_hash(tunnel, session->session_id);
-
-	spin_lock_bh(&tunnel->hlist_lock);
+	spin_lock_bh(&tunnel->list_lock);
 	if (!tunnel->acpt_newsess) {
 		err = -ENODEV;
 		goto err_tlock;
 	}
 
-	hlist_for_each_entry(session_walk, head, hlist)
-		if (session_walk->session_id == session->session_id) {
-			err = -EEXIST;
-			goto err_tlock;
-		}
-
 	if (tunnel->version == L2TP_HDR_VER_3) {
 		session_key = session->session_id;
 		spin_lock_bh(&pn->l2tp_session_idr_lock);
@@ -506,8 +481,8 @@ int l2tp_session_register(struct l2tp_session *session,
 
 	l2tp_tunnel_inc_refcount(tunnel);
 
-	hlist_add_head_rcu(&session->hlist, head);
-	spin_unlock_bh(&tunnel->hlist_lock);
+	list_add(&session->list, &tunnel->session_list);
+	spin_unlock_bh(&tunnel->list_lock);
 
 	spin_lock_bh(&pn->l2tp_session_idr_lock);
 	if (tunnel->version == L2TP_HDR_VER_3)
@@ -521,7 +496,7 @@ int l2tp_session_register(struct l2tp_session *session,
 	return 0;
 
 err_tlock:
-	spin_unlock_bh(&tunnel->hlist_lock);
+	spin_unlock_bh(&tunnel->list_lock);
 
 	return err;
 }
@@ -1275,20 +1250,19 @@ end:
 	return;
 }
 
-/* Remove an l2tp session from l2tp_core's hash lists. */
+/* Remove an l2tp session from l2tp_core's lists. */
 static void l2tp_session_unhash(struct l2tp_session *session)
 {
 	struct l2tp_tunnel *tunnel = session->tunnel;
 
-	/* Remove the session from core hashes */
 	if (tunnel) {
 		struct l2tp_net *pn = l2tp_pernet(tunnel->l2tp_net);
 		struct l2tp_session *removed = session;
 
-		/* Remove from the per-tunnel hash */
-		spin_lock_bh(&tunnel->hlist_lock);
-		hlist_del_init_rcu(&session->hlist);
-		spin_unlock_bh(&tunnel->hlist_lock);
+		/* Remove from the per-tunnel list */
+		spin_lock_bh(&tunnel->list_lock);
+		list_del_init(&session->list);
+		spin_unlock_bh(&tunnel->list_lock);
 
 		/* Remove from per-net IDR */
 		spin_lock_bh(&pn->l2tp_session_idr_lock);
@@ -1316,28 +1290,19 @@ static void l2tp_session_unhash(struct l2tp_session *session)
 static void l2tp_tunnel_closeall(struct l2tp_tunnel *tunnel)
 {
 	struct l2tp_session *session;
-	int hash;
+	struct list_head __rcu *pos;
+	struct list_head *tmp;
 
-	spin_lock_bh(&tunnel->hlist_lock);
+	spin_lock_bh(&tunnel->list_lock);
 	tunnel->acpt_newsess = false;
-	for (hash = 0; hash < L2TP_HASH_SIZE; hash++) {
-again:
-		hlist_for_each_entry_rcu(session, &tunnel->session_hlist[hash], hlist) {
-			hlist_del_init_rcu(&session->hlist);
-
-			spin_unlock_bh(&tunnel->hlist_lock);
-			l2tp_session_delete(session);
-			spin_lock_bh(&tunnel->hlist_lock);
-
-			/* Now restart from the beginning of this hash
-			 * chain.  We always remove a session from the
-			 * list so we are guaranteed to make forward
-			 * progress.
-			 */
-			goto again;
-		}
+	list_for_each_safe(pos, tmp, &tunnel->session_list) {
+		session = list_entry(pos, struct l2tp_session, list);
+		list_del_init(&session->list);
+		spin_unlock_bh(&tunnel->list_lock);
+		l2tp_session_delete(session);
+		spin_lock_bh(&tunnel->list_lock);
 	}
-	spin_unlock_bh(&tunnel->hlist_lock);
+	spin_unlock_bh(&tunnel->list_lock);
 }
 
 /* Tunnel socket destroy hook for UDP encapsulation */
@@ -1531,8 +1496,9 @@ int l2tp_tunnel_create(int fd, int version, u32 tunnel_id, u32 peer_tunnel_id,
 
 	tunnel->magic = L2TP_TUNNEL_MAGIC;
 	sprintf(&tunnel->name[0], "tunl %u", tunnel_id);
-	spin_lock_init(&tunnel->hlist_lock);
+	spin_lock_init(&tunnel->list_lock);
 	tunnel->acpt_newsess = true;
+	INIT_LIST_HEAD(&tunnel->session_list);
 
 	tunnel->encap = encap;
 
@@ -1732,6 +1698,7 @@ struct l2tp_session *l2tp_session_create(int priv_size, struct l2tp_tunnel *tunn
 		session->hlist_key = l2tp_v3_session_hashkey(tunnel->sock, session->session_id);
 		INIT_HLIST_NODE(&session->hlist);
 		INIT_LIST_HEAD(&session->clist);
+		INIT_LIST_HEAD(&session->list);
 
 		if (cfg) {
 			session->pwtype = cfg->pw_type;
diff --git a/net/l2tp/l2tp_core.h b/net/l2tp/l2tp_core.h
index bfff69f2e0a2..8ac81bc1bc6f 100644
--- a/net/l2tp/l2tp_core.h
+++ b/net/l2tp/l2tp_core.h
@@ -19,10 +19,6 @@
 #define L2TP_TUNNEL_MAGIC	0x42114DDA
 #define L2TP_SESSION_MAGIC	0x0C04EB7D
 
-/* Per tunnel session hash table size */
-#define L2TP_HASH_BITS	4
-#define L2TP_HASH_SIZE	BIT(L2TP_HASH_BITS)
-
 struct sk_buff;
 
 struct l2tp_stats {
@@ -65,8 +61,7 @@ struct l2tp_session_coll_list {
 
 /* Represents a session (pseudowire) instance.
  * Tracks runtime state including cookies, dataplane packet sequencing, and IO statistics.
- * Is linked into a per-tunnel session hashlist; and in the case of an L2TPv3 session into
- * an additional per-net ("global") hashlist.
+ * Is linked into a per-tunnel session list and a per-net ("global") IDR tree.
  */
 #define L2TP_SESSION_NAME_MAX 32
 struct l2tp_session {
@@ -90,6 +85,7 @@ struct l2tp_session {
 	u32			nr_oos;		/* NR of last OOS packet */
 	int			nr_oos_count;	/* for OOS recovery */
 	int			nr_oos_count_max;
+	struct list_head	list;		/* per-tunnel list node */
 	refcount_t		ref_count;
 	struct hlist_node	hlist;		/* per-net session hlist */
 	unsigned long		hlist_key;	/* key for session hlist */
@@ -118,7 +114,7 @@ struct l2tp_session {
 	/* Session close handler.
 	 * Each pseudowire implementation may implement this callback in order to carry
 	 * out pseudowire-specific shutdown actions.
-	 * The callback is called by core after unhashing the session and purging its
+	 * The callback is called by core after unlisting the session and purging its
 	 * reorder queue.
 	 */
 	void (*session_close)(struct l2tp_session *session);
@@ -154,7 +150,7 @@ struct l2tp_tunnel_cfg {
 /* Represents a tunnel instance.
  * Tracks runtime state including IO statistics.
  * Holds the tunnel socket (either passed from userspace or directly created by the kernel).
- * Maintains a hashlist of sessions belonging to the tunnel instance.
+ * Maintains a list of sessions belonging to the tunnel instance.
  * Is linked into a per-net list of tunnels.
  */
 #define L2TP_TUNNEL_NAME_MAX 20
@@ -164,12 +160,11 @@ struct l2tp_tunnel {
 	unsigned long		dead;
 
 	struct rcu_head rcu;
-	spinlock_t		hlist_lock;	/* write-protection for session_hlist */
+	spinlock_t		list_lock;	/* write-protection for session_list */
 	bool			acpt_newsess;	/* indicates whether this tunnel accepts
-						 * new sessions. Protected by hlist_lock.
+						 * new sessions. Protected by list_lock.
 						 */
-	struct hlist_head	session_hlist[L2TP_HASH_SIZE];
-						/* hashed list of sessions, hashed by id */
+	struct list_head	session_list;	/* list of sessions */
 	u32			tunnel_id;
 	u32			peer_tunnel_id;
 	int			version;	/* 2=>L2TPv2, 3=>L2TPv3 */
diff --git a/net/l2tp/l2tp_debugfs.c b/net/l2tp/l2tp_debugfs.c
index 4595b56d175d..8755ae521154 100644
--- a/net/l2tp/l2tp_debugfs.c
+++ b/net/l2tp/l2tp_debugfs.c
@@ -123,17 +123,14 @@ static void l2tp_dfs_seq_tunnel_show(struct seq_file *m, void *v)
 	struct l2tp_tunnel *tunnel = v;
 	struct l2tp_session *session;
 	int session_count = 0;
-	int hash;
 
 	rcu_read_lock_bh();
-	for (hash = 0; hash < L2TP_HASH_SIZE; hash++) {
-		hlist_for_each_entry_rcu(session, &tunnel->session_hlist[hash], hlist) {
-			/* Session ID of zero is a dummy/reserved value used by pppol2tp */
-			if (session->session_id == 0)
-				continue;
+	list_for_each_entry_rcu(session, &tunnel->session_list, list) {
+		/* Session ID of zero is a dummy/reserved value used by pppol2tp */
+		if (session->session_id == 0)
+			continue;
 
-			session_count++;
-		}
+		session_count++;
 	}
 	rcu_read_unlock_bh();
 
-- 
cgit v1.2.3-58-ga151


From e5efa3ff412dd1d3ff266be1eea6ee47c9d951c4 Mon Sep 17 00:00:00 2001
From: Herve Codina <herve.codina@bootlin.com>
Date: Thu, 20 Jun 2024 14:01:24 +0200
Subject: dt-bindings: net: mscc-miim: Add resets property

Add the (optional) resets property.
The mscc-miim device is impacted by the switch reset especially when the
mscc-miim device is used as part of the LAN966x PCI device.

Signed-off-by: Herve Codina <herve.codina@bootlin.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/devicetree/bindings/net/mscc,miim.yaml | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/mscc,miim.yaml b/Documentation/devicetree/bindings/net/mscc,miim.yaml
index 5b292e7c9e46..792f26b06b06 100644
--- a/Documentation/devicetree/bindings/net/mscc,miim.yaml
+++ b/Documentation/devicetree/bindings/net/mscc,miim.yaml
@@ -38,6 +38,16 @@ properties:
 
   clock-frequency: true
 
+  resets:
+    items:
+      - description:
+          Reset shared with all blocks attached to the Switch Core Register
+          Bus (CSR) including VRAP slave.
+
+  reset-names:
+    items:
+      - const: switch
+
 required:
   - compatible
   - reg
-- 
cgit v1.2.3-58-ga151


From 9e6d33937b42ca4867af3b341e5d09abca4a2746 Mon Sep 17 00:00:00 2001
From: Herve Codina <herve.codina@bootlin.com>
Date: Thu, 20 Jun 2024 14:01:25 +0200
Subject: net: mdio: mscc-miim: Handle the switch reset

The mscc-miim device can be impacted by the switch reset, at least when
this device is part of the LAN966x PCI device.

Handle this newly added (optional) resets property.

Signed-off-by: Herve Codina <herve.codina@bootlin.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mdio/mdio-mscc-miim.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/net/mdio/mdio-mscc-miim.c b/drivers/net/mdio/mdio-mscc-miim.c
index c29377c85307..62c47e0dd142 100644
--- a/drivers/net/mdio/mdio-mscc-miim.c
+++ b/drivers/net/mdio/mdio-mscc-miim.c
@@ -19,6 +19,7 @@
 #include <linux/platform_device.h>
 #include <linux/property.h>
 #include <linux/regmap.h>
+#include <linux/reset.h>
 
 #define MSCC_MIIM_REG_STATUS		0x0
 #define		MSCC_MIIM_STATUS_STAT_PENDING	BIT(2)
@@ -271,10 +272,17 @@ static int mscc_miim_probe(struct platform_device *pdev)
 	struct device_node *np = pdev->dev.of_node;
 	struct regmap *mii_regmap, *phy_regmap;
 	struct device *dev = &pdev->dev;
+	struct reset_control *reset;
 	struct mscc_miim_dev *miim;
 	struct mii_bus *bus;
 	int ret;
 
+	reset = devm_reset_control_get_optional_shared(dev, "switch");
+	if (IS_ERR(reset))
+		return dev_err_probe(dev, PTR_ERR(reset), "Failed to get reset\n");
+
+	reset_control_reset(reset);
+
 	mii_regmap = ocelot_regmap_from_resource(pdev, 0,
 						 &mscc_miim_regmap_config);
 	if (IS_ERR(mii_regmap))
-- 
cgit v1.2.3-58-ga151


From e85d3e6fea05c8ae21a40809a3c6b7adc97411c7 Mon Sep 17 00:00:00 2001
From: Matthias Schiffer <mschiffer@universe-factory.net>
Date: Thu, 20 Jun 2024 19:25:48 +0200
Subject: net: dsa: qca8k: do not write port mask twice in bridge join/leave

qca8k_port_bridge_join() set QCA8K_PORT_LOOKUP_CTRL() for i == port twice,
once in the loop handling all other port's masks, and finally at the end
with the accumulated port_mask.

The first time it would incorrectly set the port's own bit in the mask,
only to correct the mistake a moment later. qca8k_port_bridge_leave() had
the same issue, but here the regmap_clear_bits() was a no-op rather than
setting an unintended value.

Remove the duplicate assignment by skipping the whole loop iteration for
i == port. The unintended bit setting doesn't seem to have any negative
effects (even when not reverted right away), so the change is submitted
as a simple cleanup rather than a fix.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/qca/qca8k-common.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/qca/qca8k-common.c b/drivers/net/dsa/qca/qca8k-common.c
index 7f80035c5441..b33df84070d3 100644
--- a/drivers/net/dsa/qca/qca8k-common.c
+++ b/drivers/net/dsa/qca/qca8k-common.c
@@ -653,6 +653,8 @@ int qca8k_port_bridge_join(struct dsa_switch *ds, int port,
 	port_mask = BIT(cpu_port);
 
 	for (i = 0; i < QCA8K_NUM_PORTS; i++) {
+		if (i == port)
+			continue;
 		if (dsa_is_cpu_port(ds, i))
 			continue;
 		if (!dsa_port_offloads_bridge(dsa_to_port(ds, i), &bridge))
@@ -665,8 +667,7 @@ int qca8k_port_bridge_join(struct dsa_switch *ds, int port,
 				      BIT(port));
 		if (ret)
 			return ret;
-		if (i != port)
-			port_mask |= BIT(i);
+		port_mask |= BIT(i);
 	}
 
 	/* Add all other ports to this ports portvlan mask */
@@ -685,6 +686,8 @@ void qca8k_port_bridge_leave(struct dsa_switch *ds, int port,
 	cpu_port = dsa_to_port(ds, port)->cpu_dp->index;
 
 	for (i = 0; i < QCA8K_NUM_PORTS; i++) {
+		if (i == port)
+			continue;
 		if (dsa_is_cpu_port(ds, i))
 			continue;
 		if (!dsa_port_offloads_bridge(dsa_to_port(ds, i), &bridge))
-- 
cgit v1.2.3-58-ga151


From 412e1775f413c944b8c51bdadb675be957d83dc8 Mon Sep 17 00:00:00 2001
From: Matthias Schiffer <mschiffer@universe-factory.net>
Date: Thu, 20 Jun 2024 19:25:49 +0200
Subject: net: dsa: qca8k: factor out bridge join/leave logic

Most of the logic in qca8k_port_bridge_join() and qca8k_port_bridge_leave()
is the same. Refactor to reduce duplication and prepare for reusing the
code for implementing bridge port isolation.

dsa_port_offloads_bridge_dev() is used instead of
dsa_port_offloads_bridge(), passing the bridge in as a struct netdevice *,
as we won't have a struct dsa_bridge in qca8k_port_bridge_flags().

The error handling is changed slightly in the bridge leave case,
returning early and emitting an error message when a regmap access fails.
This shouldn't matter in practice, as there isn't much we can do if
communication with the switch breaks down in the middle of reconfiguration.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/qca/qca8k-common.c | 101 ++++++++++++++++++-------------------
 1 file changed, 50 insertions(+), 51 deletions(-)

diff --git a/drivers/net/dsa/qca/qca8k-common.c b/drivers/net/dsa/qca/qca8k-common.c
index b33df84070d3..09108fa99dbe 100644
--- a/drivers/net/dsa/qca/qca8k-common.c
+++ b/drivers/net/dsa/qca/qca8k-common.c
@@ -614,6 +614,49 @@ void qca8k_port_stp_state_set(struct dsa_switch *ds, int port, u8 state)
 	qca8k_port_configure_learning(ds, port, learning);
 }
 
+static int qca8k_update_port_member(struct qca8k_priv *priv, int port,
+				    const struct net_device *bridge_dev,
+				    bool join)
+{
+	struct dsa_port *dp = dsa_to_port(priv->ds, port), *other_dp;
+	u32 port_mask = BIT(dp->cpu_dp->index);
+	int i, ret;
+
+	for (i = 0; i < QCA8K_NUM_PORTS; i++) {
+		if (i == port)
+			continue;
+		if (dsa_is_cpu_port(priv->ds, i))
+			continue;
+
+		other_dp = dsa_to_port(priv->ds, i);
+		if (!dsa_port_offloads_bridge_dev(other_dp, bridge_dev))
+			continue;
+
+		/* Add/remove this port to/from the portvlan mask of the other
+		 * ports in the bridge
+		 */
+		if (join) {
+			port_mask |= BIT(i);
+			ret = regmap_set_bits(priv->regmap,
+					      QCA8K_PORT_LOOKUP_CTRL(i),
+					      BIT(port));
+		} else {
+			ret = regmap_clear_bits(priv->regmap,
+						QCA8K_PORT_LOOKUP_CTRL(i),
+						BIT(port));
+		}
+
+		if (ret)
+			return ret;
+	}
+
+	/* Add/remove all other ports to/from this port's portvlan mask */
+	ret = qca8k_rmw(priv, QCA8K_PORT_LOOKUP_CTRL(port),
+			QCA8K_PORT_LOOKUP_MEMBER, port_mask);
+
+	return ret;
+}
+
 int qca8k_port_pre_bridge_flags(struct dsa_switch *ds, int port,
 				struct switchdev_brport_flags flags,
 				struct netlink_ext_ack *extack)
@@ -646,65 +689,21 @@ int qca8k_port_bridge_join(struct dsa_switch *ds, int port,
 			   struct netlink_ext_ack *extack)
 {
 	struct qca8k_priv *priv = ds->priv;
-	int port_mask, cpu_port;
-	int i, ret;
-
-	cpu_port = dsa_to_port(ds, port)->cpu_dp->index;
-	port_mask = BIT(cpu_port);
 
-	for (i = 0; i < QCA8K_NUM_PORTS; i++) {
-		if (i == port)
-			continue;
-		if (dsa_is_cpu_port(ds, i))
-			continue;
-		if (!dsa_port_offloads_bridge(dsa_to_port(ds, i), &bridge))
-			continue;
-		/* Add this port to the portvlan mask of the other ports
-		 * in the bridge
-		 */
-		ret = regmap_set_bits(priv->regmap,
-				      QCA8K_PORT_LOOKUP_CTRL(i),
-				      BIT(port));
-		if (ret)
-			return ret;
-		port_mask |= BIT(i);
-	}
-
-	/* Add all other ports to this ports portvlan mask */
-	ret = qca8k_rmw(priv, QCA8K_PORT_LOOKUP_CTRL(port),
-			QCA8K_PORT_LOOKUP_MEMBER, port_mask);
-
-	return ret;
+	return qca8k_update_port_member(priv, port, bridge.dev, true);
 }
 
 void qca8k_port_bridge_leave(struct dsa_switch *ds, int port,
 			     struct dsa_bridge bridge)
 {
 	struct qca8k_priv *priv = ds->priv;
-	int cpu_port, i;
-
-	cpu_port = dsa_to_port(ds, port)->cpu_dp->index;
+	int err;
 
-	for (i = 0; i < QCA8K_NUM_PORTS; i++) {
-		if (i == port)
-			continue;
-		if (dsa_is_cpu_port(ds, i))
-			continue;
-		if (!dsa_port_offloads_bridge(dsa_to_port(ds, i), &bridge))
-			continue;
-		/* Remove this port to the portvlan mask of the other ports
-		 * in the bridge
-		 */
-		regmap_clear_bits(priv->regmap,
-				  QCA8K_PORT_LOOKUP_CTRL(i),
-				  BIT(port));
-	}
-
-	/* Set the cpu port to be the only one in the portvlan mask of
-	 * this port
-	 */
-	qca8k_rmw(priv, QCA8K_PORT_LOOKUP_CTRL(port),
-		  QCA8K_PORT_LOOKUP_MEMBER, BIT(cpu_port));
+	err = qca8k_update_port_member(priv, port, bridge.dev, false);
+	if (err)
+		dev_err(priv->dev,
+			"Failed to update switch config for bridge leave: %d\n",
+			err);
 }
 
 void qca8k_port_fast_age(struct dsa_switch *ds, int port)
-- 
cgit v1.2.3-58-ga151


From 422b64025ec10981c48f9367311846bf4bd38042 Mon Sep 17 00:00:00 2001
From: Matthias Schiffer <mschiffer@universe-factory.net>
Date: Thu, 20 Jun 2024 19:25:50 +0200
Subject: net: dsa: qca8k: add support for bridge port isolation

Remove a pair of ports from the port matrix when both ports have the
isolated flag set.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/qca/qca8k-common.c | 22 ++++++++++++++++++++--
 drivers/net/dsa/qca/qca8k.h        |  1 +
 2 files changed, 21 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/qca/qca8k-common.c b/drivers/net/dsa/qca/qca8k-common.c
index 09108fa99dbe..560c74c4ac3d 100644
--- a/drivers/net/dsa/qca/qca8k-common.c
+++ b/drivers/net/dsa/qca/qca8k-common.c
@@ -618,6 +618,7 @@ static int qca8k_update_port_member(struct qca8k_priv *priv, int port,
 				    const struct net_device *bridge_dev,
 				    bool join)
 {
+	bool isolated = !!(priv->port_isolated_map & BIT(port)), other_isolated;
 	struct dsa_port *dp = dsa_to_port(priv->ds, port), *other_dp;
 	u32 port_mask = BIT(dp->cpu_dp->index);
 	int i, ret;
@@ -632,10 +633,12 @@ static int qca8k_update_port_member(struct qca8k_priv *priv, int port,
 		if (!dsa_port_offloads_bridge_dev(other_dp, bridge_dev))
 			continue;
 
+		other_isolated = !!(priv->port_isolated_map & BIT(i));
+
 		/* Add/remove this port to/from the portvlan mask of the other
 		 * ports in the bridge
 		 */
-		if (join) {
+		if (join && !(isolated && other_isolated)) {
 			port_mask |= BIT(i);
 			ret = regmap_set_bits(priv->regmap,
 					      QCA8K_PORT_LOOKUP_CTRL(i),
@@ -661,7 +664,7 @@ int qca8k_port_pre_bridge_flags(struct dsa_switch *ds, int port,
 				struct switchdev_brport_flags flags,
 				struct netlink_ext_ack *extack)
 {
-	if (flags.mask & ~BR_LEARNING)
+	if (flags.mask & ~(BR_LEARNING | BR_ISOLATED))
 		return -EINVAL;
 
 	return 0;
@@ -671,6 +674,7 @@ int qca8k_port_bridge_flags(struct dsa_switch *ds, int port,
 			    struct switchdev_brport_flags flags,
 			    struct netlink_ext_ack *extack)
 {
+	struct qca8k_priv *priv = ds->priv;
 	int ret;
 
 	if (flags.mask & BR_LEARNING) {
@@ -680,6 +684,20 @@ int qca8k_port_bridge_flags(struct dsa_switch *ds, int port,
 			return ret;
 	}
 
+	if (flags.mask & BR_ISOLATED) {
+		struct dsa_port *dp = dsa_to_port(ds, port);
+		struct net_device *bridge_dev = dsa_port_bridge_dev_get(dp);
+
+		if (flags.val & BR_ISOLATED)
+			priv->port_isolated_map |= BIT(port);
+		else
+			priv->port_isolated_map &= ~BIT(port);
+
+		ret = qca8k_update_port_member(priv, port, bridge_dev, true);
+		if (ret)
+			return ret;
+	}
+
 	return 0;
 }
 
diff --git a/drivers/net/dsa/qca/qca8k.h b/drivers/net/dsa/qca/qca8k.h
index 2184d8d2d5a9..3664a2e2f1f6 100644
--- a/drivers/net/dsa/qca/qca8k.h
+++ b/drivers/net/dsa/qca/qca8k.h
@@ -451,6 +451,7 @@ struct qca8k_priv {
 	 * Bit 1: port enabled. Bit 0: port disabled.
 	 */
 	u8 port_enabled_map;
+	u8 port_isolated_map;
 	struct qca8k_ports_config ports_config;
 	struct regmap *regmap;
 	struct mii_bus *bus;
-- 
cgit v1.2.3-58-ga151


From b0d3969d2b4db82602492cad576b8de494a12ddf Mon Sep 17 00:00:00 2001
From: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Date: Thu, 20 Jun 2024 20:22:19 +0200
Subject: net: ethernet: rtsn: Add support for Renesas Ethernet-TSN
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add initial support for Renesas Ethernet-TSN End-station device of R-Car
V4H. The Ethernet End-station can connect to an Ethernet network using a
10 Mbps, 100 Mbps, or 1 Gbps full-duplex link via MII/GMII/RMII/RGMII.
Depending on the connected PHY.

The driver supports Rx checksum and offload and hardware timestamps.

While full power management and suspend/resume is not yet supported the
driver enables runtime PM in order to enable the module clock. While
explicit clock management using clk_enable() would suffice for the
supported SoC, the module could be reused on SoCs where the module is
part of a power domain.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS                           |    8 +
 drivers/net/ethernet/renesas/Kconfig  |   10 +
 drivers/net/ethernet/renesas/Makefile |    2 +
 drivers/net/ethernet/renesas/rtsn.c   | 1391 +++++++++++++++++++++++++++++++++
 drivers/net/ethernet/renesas/rtsn.h   |  464 +++++++++++
 5 files changed, 1875 insertions(+)
 create mode 100644 drivers/net/ethernet/renesas/rtsn.c
 create mode 100644 drivers/net/ethernet/renesas/rtsn.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 1e8a95217b36..d648af07cbd6 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -19032,6 +19032,14 @@ F:	drivers/net/ethernet/renesas/Makefile
 F:	drivers/net/ethernet/renesas/rcar_gen4*
 F:	drivers/net/ethernet/renesas/rswitch*
 
+RENESAS ETHERNET TSN DRIVER
+M:	Niklas Söderlund <niklas.soderlund@ragnatech.se>
+L:	netdev@vger.kernel.org
+L:	linux-renesas-soc@vger.kernel.org
+S:	Supported
+F:	Documentation/devicetree/bindings/net/renesas,ethertsn.yaml
+F:	drivers/net/ethernet/renesas/rtsn.*
+
 RENESAS IDT821034 ASoC CODEC
 M:	Herve Codina <herve.codina@bootlin.com>
 L:	alsa-devel@alsa-project.org (moderated for non-subscribers)
diff --git a/drivers/net/ethernet/renesas/Kconfig b/drivers/net/ethernet/renesas/Kconfig
index 83de7e56da61..9b7559c88bee 100644
--- a/drivers/net/ethernet/renesas/Kconfig
+++ b/drivers/net/ethernet/renesas/Kconfig
@@ -59,4 +59,14 @@ config RENESAS_GEN4_PTP
 	help
 	  Renesas R-Car Gen4 gPTP device driver.
 
+config RTSN
+	tristate "Renesas Ethernet-TSN support"
+	depends on ARCH_RENESAS || COMPILE_TEST
+	depends on PTP_1588_CLOCK
+	select CRC32
+	select PHYLIB
+	select RENESAS_GEN4_PTP
+	help
+	  Renesas Ethernet-TSN device driver.
+
 endif # NET_VENDOR_RENESAS
diff --git a/drivers/net/ethernet/renesas/Makefile b/drivers/net/ethernet/renesas/Makefile
index 9070acfd6aaf..f65fc76f8b4d 100644
--- a/drivers/net/ethernet/renesas/Makefile
+++ b/drivers/net/ethernet/renesas/Makefile
@@ -11,3 +11,5 @@ obj-$(CONFIG_RAVB) += ravb.o
 obj-$(CONFIG_RENESAS_ETHER_SWITCH) += rswitch.o
 
 obj-$(CONFIG_RENESAS_GEN4_PTP) += rcar_gen4_ptp.o
+
+obj-$(CONFIG_RTSN) += rtsn.o
diff --git a/drivers/net/ethernet/renesas/rtsn.c b/drivers/net/ethernet/renesas/rtsn.c
new file mode 100644
index 000000000000..ad69d47463cb
--- /dev/null
+++ b/drivers/net/ethernet/renesas/rtsn.c
@@ -0,0 +1,1391 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/* Renesas Ethernet-TSN device driver
+ *
+ * Copyright (C) 2022 Renesas Electronics Corporation
+ * Copyright (C) 2023 Niklas Söderlund <niklas.soderlund@ragnatech.se>
+ */
+
+#include <linux/clk.h>
+#include <linux/dma-mapping.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/module.h>
+#include <linux/net_tstamp.h>
+#include <linux/of.h>
+#include <linux/of_mdio.h>
+#include <linux/of_net.h>
+#include <linux/phy.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/reset.h>
+#include <linux/spinlock.h>
+
+#include "rtsn.h"
+#include "rcar_gen4_ptp.h"
+
+struct rtsn_private {
+	struct net_device *ndev;
+	struct platform_device *pdev;
+	void __iomem *base;
+	struct rcar_gen4_ptp_private *ptp_priv;
+	struct clk *clk;
+	struct reset_control *reset;
+
+	u32 num_tx_ring;
+	u32 num_rx_ring;
+	u32 tx_desc_bat_size;
+	dma_addr_t tx_desc_bat_dma;
+	struct rtsn_desc *tx_desc_bat;
+	u32 rx_desc_bat_size;
+	dma_addr_t rx_desc_bat_dma;
+	struct rtsn_desc *rx_desc_bat;
+	dma_addr_t tx_desc_dma;
+	dma_addr_t rx_desc_dma;
+	struct rtsn_ext_desc *tx_ring;
+	struct rtsn_ext_ts_desc *rx_ring;
+	struct sk_buff **tx_skb;
+	struct sk_buff **rx_skb;
+	spinlock_t lock;	/* Register access lock */
+	u32 cur_tx;
+	u32 dirty_tx;
+	u32 cur_rx;
+	u32 dirty_rx;
+	u8 ts_tag;
+	struct napi_struct napi;
+	struct rtnl_link_stats64 stats;
+
+	struct mii_bus *mii;
+	phy_interface_t iface;
+	int link;
+	int speed;
+
+	int tx_data_irq;
+	int rx_data_irq;
+};
+
+static u32 rtsn_read(struct rtsn_private *priv, enum rtsn_reg reg)
+{
+	return ioread32(priv->base + reg);
+}
+
+static void rtsn_write(struct rtsn_private *priv, enum rtsn_reg reg, u32 data)
+{
+	iowrite32(data, priv->base + reg);
+}
+
+static void rtsn_modify(struct rtsn_private *priv, enum rtsn_reg reg,
+			u32 clear, u32 set)
+{
+	rtsn_write(priv, reg, (rtsn_read(priv, reg) & ~clear) | set);
+}
+
+static int rtsn_reg_wait(struct rtsn_private *priv, enum rtsn_reg reg,
+			 u32 mask, u32 expected)
+{
+	u32 val;
+
+	return readl_poll_timeout(priv->base + reg, val,
+				  (val & mask) == expected,
+				  RTSN_INTERVAL_US, RTSN_TIMEOUT_US);
+}
+
+static void rtsn_ctrl_data_irq(struct rtsn_private *priv, bool enable)
+{
+	if (enable) {
+		rtsn_write(priv, TDIE0, TDIE_TDID_TDX(TX_CHAIN_IDX));
+		rtsn_write(priv, RDIE0, RDIE_RDID_RDX(RX_CHAIN_IDX));
+	} else {
+		rtsn_write(priv, TDID0, TDIE_TDID_TDX(TX_CHAIN_IDX));
+		rtsn_write(priv, RDID0, RDIE_RDID_RDX(RX_CHAIN_IDX));
+	}
+}
+
+static void rtsn_get_timestamp(struct rtsn_private *priv, struct timespec64 *ts)
+{
+	struct rcar_gen4_ptp_private *ptp_priv = priv->ptp_priv;
+
+	ptp_priv->info.gettime64(&ptp_priv->info, ts);
+}
+
+static int rtsn_tx_free(struct net_device *ndev, bool free_txed_only)
+{
+	struct rtsn_private *priv = netdev_priv(ndev);
+	struct rtsn_ext_desc *desc;
+	struct sk_buff *skb;
+	int free_num = 0;
+	int entry, size;
+
+	for (; priv->cur_tx - priv->dirty_tx > 0; priv->dirty_tx++) {
+		entry = priv->dirty_tx % priv->num_tx_ring;
+		desc = &priv->tx_ring[entry];
+		if (free_txed_only && (desc->die_dt & DT_MASK) != DT_FEMPTY)
+			break;
+
+		dma_rmb();
+		size = le16_to_cpu(desc->info_ds) & TX_DS;
+		skb = priv->tx_skb[entry];
+		if (skb) {
+			if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) {
+				struct skb_shared_hwtstamps shhwtstamps;
+				struct timespec64 ts;
+
+				rtsn_get_timestamp(priv, &ts);
+				memset(&shhwtstamps, 0, sizeof(shhwtstamps));
+				shhwtstamps.hwtstamp = timespec64_to_ktime(ts);
+				skb_tstamp_tx(skb, &shhwtstamps);
+			}
+			dma_unmap_single(ndev->dev.parent,
+					 le32_to_cpu(desc->dptr),
+					 size, DMA_TO_DEVICE);
+			dev_kfree_skb_any(priv->tx_skb[entry]);
+			free_num++;
+
+			priv->stats.tx_packets++;
+			priv->stats.tx_bytes += size;
+		}
+
+		desc->die_dt = DT_EEMPTY;
+	}
+
+	desc = &priv->tx_ring[priv->num_tx_ring];
+	desc->die_dt = DT_LINK;
+
+	return free_num;
+}
+
+static int rtsn_rx(struct net_device *ndev, int budget)
+{
+	struct rtsn_private *priv = netdev_priv(ndev);
+	unsigned int ndescriptors;
+	unsigned int rx_packets;
+	unsigned int i;
+	bool get_ts;
+
+	get_ts = priv->ptp_priv->tstamp_rx_ctrl &
+		RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT;
+
+	ndescriptors = priv->dirty_rx + priv->num_rx_ring - priv->cur_rx;
+	rx_packets = 0;
+	for (i = 0; i < ndescriptors; i++) {
+		const unsigned int entry = priv->cur_rx % priv->num_rx_ring;
+		struct rtsn_ext_ts_desc *desc = &priv->rx_ring[entry];
+		struct sk_buff *skb;
+		dma_addr_t dma_addr;
+		u16 pkt_len;
+
+		/* Stop processing descriptors if budget is consumed. */
+		if (rx_packets >= budget)
+			break;
+
+		/* Stop processing descriptors on first empty. */
+		if ((desc->die_dt & DT_MASK) == DT_FEMPTY)
+			break;
+
+		dma_rmb();
+		pkt_len = le16_to_cpu(desc->info_ds) & RX_DS;
+
+		skb = priv->rx_skb[entry];
+		priv->rx_skb[entry] = NULL;
+		dma_addr = le32_to_cpu(desc->dptr);
+		dma_unmap_single(ndev->dev.parent, dma_addr, PKT_BUF_SZ,
+				 DMA_FROM_DEVICE);
+
+		/* Get timestamp if enabled. */
+		if (get_ts) {
+			struct skb_shared_hwtstamps *shhwtstamps;
+			struct timespec64 ts;
+
+			shhwtstamps = skb_hwtstamps(skb);
+			memset(shhwtstamps, 0, sizeof(*shhwtstamps));
+
+			ts.tv_sec = (u64)le32_to_cpu(desc->ts_sec);
+			ts.tv_nsec = le32_to_cpu(desc->ts_nsec & cpu_to_le32(0x3fffffff));
+
+			shhwtstamps->hwtstamp = timespec64_to_ktime(ts);
+		}
+
+		skb_put(skb, pkt_len);
+		skb->protocol = eth_type_trans(skb, ndev);
+		napi_gro_receive(&priv->napi, skb);
+
+		/* Update statistics. */
+		priv->stats.rx_packets++;
+		priv->stats.rx_bytes += pkt_len;
+
+		/* Update counters. */
+		priv->cur_rx++;
+		rx_packets++;
+	}
+
+	/* Refill the RX ring buffers */
+	for (; priv->cur_rx - priv->dirty_rx > 0; priv->dirty_rx++) {
+		const unsigned int entry = priv->dirty_rx % priv->num_rx_ring;
+		struct rtsn_ext_ts_desc *desc = &priv->rx_ring[entry];
+		struct sk_buff *skb;
+		dma_addr_t dma_addr;
+
+		desc->info_ds = cpu_to_le16(PKT_BUF_SZ);
+
+		if (!priv->rx_skb[entry]) {
+			skb = napi_alloc_skb(&priv->napi,
+					     PKT_BUF_SZ + RTSN_ALIGN - 1);
+			if (!skb)
+				break;
+			skb_reserve(skb, NET_IP_ALIGN);
+			dma_addr = dma_map_single(ndev->dev.parent, skb->data,
+						  le16_to_cpu(desc->info_ds),
+						  DMA_FROM_DEVICE);
+			if (dma_mapping_error(ndev->dev.parent, dma_addr))
+				desc->info_ds = cpu_to_le16(0);
+			desc->dptr = cpu_to_le32(dma_addr);
+			skb_checksum_none_assert(skb);
+			priv->rx_skb[entry] = skb;
+		}
+
+		dma_wmb();
+		desc->die_dt = DT_FEMPTY | D_DIE;
+	}
+
+	priv->rx_ring[priv->num_rx_ring].die_dt = DT_LINK;
+
+	return rx_packets;
+}
+
+static int rtsn_poll(struct napi_struct *napi, int budget)
+{
+	struct rtsn_private *priv;
+	struct net_device *ndev;
+	unsigned long flags;
+	int work_done;
+
+	ndev = napi->dev;
+	priv = netdev_priv(ndev);
+
+	/* Processing RX Descriptor Ring */
+	work_done = rtsn_rx(ndev, budget);
+
+	/* Processing TX Descriptor Ring */
+	spin_lock_irqsave(&priv->lock, flags);
+	rtsn_tx_free(ndev, true);
+	netif_wake_subqueue(ndev, 0);
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	/* Re-enable TX/RX interrupts */
+	if (work_done < budget && napi_complete_done(napi, work_done)) {
+		spin_lock_irqsave(&priv->lock, flags);
+		rtsn_ctrl_data_irq(priv, true);
+		spin_unlock_irqrestore(&priv->lock, flags);
+	}
+
+	return work_done;
+}
+
+static int rtsn_desc_alloc(struct rtsn_private *priv)
+{
+	struct device *dev = &priv->pdev->dev;
+	unsigned int i;
+
+	priv->tx_desc_bat_size = sizeof(struct rtsn_desc) * TX_NUM_CHAINS;
+	priv->tx_desc_bat = dma_alloc_coherent(dev, priv->tx_desc_bat_size,
+					       &priv->tx_desc_bat_dma,
+					       GFP_KERNEL);
+
+	if (!priv->tx_desc_bat)
+		return -ENOMEM;
+
+	for (i = 0; i < TX_NUM_CHAINS; i++)
+		priv->tx_desc_bat[i].die_dt = DT_EOS;
+
+	priv->rx_desc_bat_size = sizeof(struct rtsn_desc) * RX_NUM_CHAINS;
+	priv->rx_desc_bat = dma_alloc_coherent(dev, priv->rx_desc_bat_size,
+					       &priv->rx_desc_bat_dma,
+					       GFP_KERNEL);
+
+	if (!priv->rx_desc_bat)
+		return -ENOMEM;
+
+	for (i = 0; i < RX_NUM_CHAINS; i++)
+		priv->rx_desc_bat[i].die_dt = DT_EOS;
+
+	return 0;
+}
+
+static void rtsn_desc_free(struct rtsn_private *priv)
+{
+	if (priv->tx_desc_bat)
+		dma_free_coherent(&priv->pdev->dev, priv->tx_desc_bat_size,
+				  priv->tx_desc_bat, priv->tx_desc_bat_dma);
+	priv->tx_desc_bat = NULL;
+
+	if (priv->rx_desc_bat)
+		dma_free_coherent(&priv->pdev->dev, priv->rx_desc_bat_size,
+				  priv->rx_desc_bat, priv->rx_desc_bat_dma);
+	priv->rx_desc_bat = NULL;
+}
+
+static void rtsn_chain_free(struct rtsn_private *priv)
+{
+	struct device *dev = &priv->pdev->dev;
+
+	dma_free_coherent(dev,
+			  sizeof(struct rtsn_ext_desc) * (priv->num_tx_ring + 1),
+			  priv->tx_ring, priv->tx_desc_dma);
+	priv->tx_ring = NULL;
+
+	dma_free_coherent(dev,
+			  sizeof(struct rtsn_ext_ts_desc) * (priv->num_rx_ring + 1),
+			  priv->rx_ring, priv->rx_desc_dma);
+	priv->rx_ring = NULL;
+
+	kfree(priv->tx_skb);
+	priv->tx_skb = NULL;
+
+	kfree(priv->rx_skb);
+	priv->rx_skb = NULL;
+}
+
+static int rtsn_chain_init(struct rtsn_private *priv, int tx_size, int rx_size)
+{
+	struct net_device *ndev = priv->ndev;
+	struct sk_buff *skb;
+	int i;
+
+	priv->num_tx_ring = tx_size;
+	priv->num_rx_ring = rx_size;
+
+	priv->tx_skb = kcalloc(tx_size, sizeof(*priv->tx_skb), GFP_KERNEL);
+	priv->rx_skb = kcalloc(rx_size, sizeof(*priv->rx_skb), GFP_KERNEL);
+
+	if (!priv->rx_skb || !priv->tx_skb)
+		goto error;
+
+	for (i = 0; i < rx_size; i++) {
+		skb = netdev_alloc_skb(ndev, PKT_BUF_SZ + RTSN_ALIGN - 1);
+		if (!skb)
+			goto error;
+		skb_reserve(skb, NET_IP_ALIGN);
+		priv->rx_skb[i] = skb;
+	}
+
+	/* Allocate TX, RX descriptors */
+	priv->tx_ring = dma_alloc_coherent(ndev->dev.parent,
+					   sizeof(struct rtsn_ext_desc) * (tx_size + 1),
+					   &priv->tx_desc_dma, GFP_KERNEL);
+	priv->rx_ring = dma_alloc_coherent(ndev->dev.parent,
+					   sizeof(struct rtsn_ext_ts_desc) * (rx_size + 1),
+					   &priv->rx_desc_dma, GFP_KERNEL);
+
+	if (!priv->tx_ring || !priv->rx_ring)
+		goto error;
+
+	return 0;
+error:
+	rtsn_chain_free(priv);
+
+	return -ENOMEM;
+}
+
+static void rtsn_chain_format(struct rtsn_private *priv)
+{
+	struct net_device *ndev = priv->ndev;
+	struct rtsn_ext_ts_desc *rx_desc;
+	struct rtsn_ext_desc *tx_desc;
+	struct rtsn_desc *bat_desc;
+	dma_addr_t dma_addr;
+	unsigned int i;
+
+	priv->cur_tx = 0;
+	priv->cur_rx = 0;
+	priv->dirty_rx = 0;
+	priv->dirty_tx = 0;
+
+	/* TX */
+	memset(priv->tx_ring, 0, sizeof(*tx_desc) * priv->num_tx_ring);
+	for (i = 0, tx_desc = priv->tx_ring; i < priv->num_tx_ring; i++, tx_desc++)
+		tx_desc->die_dt = DT_EEMPTY | D_DIE;
+
+	tx_desc->dptr = cpu_to_le32((u32)priv->tx_desc_dma);
+	tx_desc->die_dt = DT_LINK;
+
+	bat_desc = &priv->tx_desc_bat[TX_CHAIN_IDX];
+	bat_desc->die_dt = DT_LINK;
+	bat_desc->dptr = cpu_to_le32((u32)priv->tx_desc_dma);
+
+	/* RX */
+	memset(priv->rx_ring, 0, sizeof(*rx_desc) * priv->num_rx_ring);
+	for (i = 0, rx_desc = priv->rx_ring; i < priv->num_rx_ring; i++, rx_desc++) {
+		dma_addr = dma_map_single(ndev->dev.parent,
+					  priv->rx_skb[i]->data, PKT_BUF_SZ,
+					  DMA_FROM_DEVICE);
+		if (!dma_mapping_error(ndev->dev.parent, dma_addr))
+			rx_desc->info_ds = cpu_to_le16(PKT_BUF_SZ);
+		rx_desc->dptr = cpu_to_le32((u32)dma_addr);
+		rx_desc->die_dt = DT_FEMPTY | D_DIE;
+	}
+	rx_desc->dptr = cpu_to_le32((u32)priv->rx_desc_dma);
+	rx_desc->die_dt = DT_LINK;
+
+	bat_desc = &priv->rx_desc_bat[RX_CHAIN_IDX];
+	bat_desc->die_dt = DT_LINK;
+	bat_desc->dptr = cpu_to_le32((u32)priv->rx_desc_dma);
+}
+
+static int rtsn_dmac_init(struct rtsn_private *priv)
+{
+	int ret;
+
+	ret = rtsn_chain_init(priv, TX_CHAIN_SIZE, RX_CHAIN_SIZE);
+	if (ret)
+		return ret;
+
+	rtsn_chain_format(priv);
+
+	return 0;
+}
+
+static enum rtsn_mode rtsn_read_mode(struct rtsn_private *priv)
+{
+	return (rtsn_read(priv, OSR) & OSR_OPS) >> 1;
+}
+
+static int rtsn_wait_mode(struct rtsn_private *priv, enum rtsn_mode mode)
+{
+	unsigned int i;
+
+	/* Need to busy loop as mode changes can happen in atomic context. */
+	for (i = 0; i < RTSN_TIMEOUT_US / RTSN_INTERVAL_US; i++) {
+		if (rtsn_read_mode(priv) == mode)
+			return 0;
+
+		udelay(RTSN_INTERVAL_US);
+	}
+
+	return -ETIMEDOUT;
+}
+
+static int rtsn_change_mode(struct rtsn_private *priv, enum rtsn_mode mode)
+{
+	int ret;
+
+	rtsn_write(priv, OCR, mode);
+	ret = rtsn_wait_mode(priv, mode);
+	if (ret)
+		netdev_err(priv->ndev, "Failed to switch operation mode\n");
+	return ret;
+}
+
+static int rtsn_get_data_irq_status(struct rtsn_private *priv)
+{
+	u32 val;
+
+	val = rtsn_read(priv, TDIS0) | TDIS_TDS(TX_CHAIN_IDX);
+	val |= rtsn_read(priv, RDIS0) | RDIS_RDS(RX_CHAIN_IDX);
+
+	return val;
+}
+
+static irqreturn_t rtsn_irq(int irq, void *dev_id)
+{
+	struct rtsn_private *priv = dev_id;
+	int ret = IRQ_NONE;
+
+	spin_lock(&priv->lock);
+
+	if (rtsn_get_data_irq_status(priv)) {
+		/* Clear TX/RX irq status */
+		rtsn_write(priv, TDIS0, TDIS_TDS(TX_CHAIN_IDX));
+		rtsn_write(priv, RDIS0, RDIS_RDS(RX_CHAIN_IDX));
+
+		if (napi_schedule_prep(&priv->napi)) {
+			/* Disable TX/RX interrupts */
+			rtsn_ctrl_data_irq(priv, false);
+
+			__napi_schedule(&priv->napi);
+		}
+
+		ret = IRQ_HANDLED;
+	}
+
+	spin_unlock(&priv->lock);
+
+	return ret;
+}
+
+static int rtsn_request_irq(unsigned int irq, irq_handler_t handler,
+			    unsigned long flags, struct rtsn_private *priv,
+			    const char *ch)
+{
+	char *name;
+	int ret;
+
+	name = devm_kasprintf(&priv->pdev->dev, GFP_KERNEL, "%s:%s",
+			      priv->ndev->name, ch);
+	if (!name)
+		return -ENOMEM;
+
+	ret = request_irq(irq, handler, flags, name, priv);
+	if (ret)
+		netdev_err(priv->ndev, "Cannot request IRQ %s\n", name);
+
+	return ret;
+}
+
+static void rtsn_free_irqs(struct rtsn_private *priv)
+{
+	free_irq(priv->tx_data_irq, priv);
+	free_irq(priv->rx_data_irq, priv);
+}
+
+static int rtsn_request_irqs(struct rtsn_private *priv)
+{
+	int ret;
+
+	priv->rx_data_irq = platform_get_irq_byname(priv->pdev, "rx");
+	if (priv->rx_data_irq < 0)
+		return priv->rx_data_irq;
+
+	priv->tx_data_irq = platform_get_irq_byname(priv->pdev, "tx");
+	if (priv->tx_data_irq < 0)
+		return priv->tx_data_irq;
+
+	ret = rtsn_request_irq(priv->tx_data_irq, rtsn_irq, 0, priv, "tx");
+	if (ret)
+		return ret;
+
+	ret = rtsn_request_irq(priv->rx_data_irq, rtsn_irq, 0, priv, "rx");
+	if (ret) {
+		free_irq(priv->tx_data_irq, priv);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int rtsn_reset(struct rtsn_private *priv)
+{
+	reset_control_reset(priv->reset);
+	mdelay(1);
+
+	return rtsn_wait_mode(priv, OCR_OPC_DISABLE);
+}
+
+static int rtsn_axibmi_init(struct rtsn_private *priv)
+{
+	int ret;
+
+	ret = rtsn_reg_wait(priv, RR, RR_RST, RR_RST_COMPLETE);
+	if (ret)
+		return ret;
+
+	/* Set AXIWC */
+	rtsn_write(priv, AXIWC, AXIWC_DEFAULT);
+
+	/* Set AXIRC */
+	rtsn_write(priv, AXIRC, AXIRC_DEFAULT);
+
+	/* TX Descriptor chain setting */
+	rtsn_write(priv, TATLS0, TATLS0_TEDE | TATLS0_TATEN(TX_CHAIN_IDX));
+	rtsn_write(priv, TATLS1, priv->tx_desc_bat_dma + TX_CHAIN_ADDR_OFFSET);
+	rtsn_write(priv, TATLR, TATLR_TATL);
+
+	ret = rtsn_reg_wait(priv, TATLR, TATLR_TATL, 0);
+	if (ret)
+		return ret;
+
+	/* RX Descriptor chain setting */
+	rtsn_write(priv, RATLS0,
+		   RATLS0_RETS | RATLS0_REDE | RATLS0_RATEN(RX_CHAIN_IDX));
+	rtsn_write(priv, RATLS1, priv->rx_desc_bat_dma + RX_CHAIN_ADDR_OFFSET);
+	rtsn_write(priv, RATLR, RATLR_RATL);
+
+	ret = rtsn_reg_wait(priv, RATLR, RATLR_RATL, 0);
+	if (ret)
+		return ret;
+
+	/* Enable TX/RX interrupts */
+	rtsn_ctrl_data_irq(priv, true);
+
+	return 0;
+}
+
+static void rtsn_mhd_init(struct rtsn_private *priv)
+{
+	/* TX General setting */
+	rtsn_write(priv, TGC1, TGC1_STTV_DEFAULT | TGC1_TQTM_SFM);
+	rtsn_write(priv, TMS0, TMS_MFS_MAX);
+
+	/* RX Filter IP */
+	rtsn_write(priv, CFCR0, CFCR_SDID(RX_CHAIN_IDX));
+	rtsn_write(priv, FMSCR, FMSCR_FMSIE(RX_CHAIN_IDX));
+}
+
+static int rtsn_get_phy_params(struct rtsn_private *priv)
+{
+	int ret;
+
+	ret = of_get_phy_mode(priv->pdev->dev.of_node, &priv->iface);
+	if (ret)
+		return ret;
+
+	switch (priv->iface) {
+	case PHY_INTERFACE_MODE_MII:
+		priv->speed = 100;
+		break;
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		priv->speed = 1000;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static void rtsn_set_phy_interface(struct rtsn_private *priv)
+{
+	u32 val;
+
+	switch (priv->iface) {
+	case PHY_INTERFACE_MODE_MII:
+		val = MPIC_PIS_MII;
+		break;
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		val = MPIC_PIS_GMII;
+		break;
+	default:
+		return;
+	}
+
+	rtsn_modify(priv, MPIC, MPIC_PIS_MASK, val);
+}
+
+static void rtsn_set_rate(struct rtsn_private *priv)
+{
+	u32 val;
+
+	switch (priv->speed) {
+	case 10:
+		val = MPIC_LSC_10M;
+		break;
+	case 100:
+		val = MPIC_LSC_100M;
+		break;
+	case 1000:
+		val = MPIC_LSC_1G;
+		break;
+	default:
+		return;
+	}
+
+	rtsn_modify(priv, MPIC, MPIC_LSC_MASK, val);
+}
+
+static int rtsn_rmac_init(struct rtsn_private *priv)
+{
+	const u8 *mac_addr = priv->ndev->dev_addr;
+	int ret;
+
+	/* Set MAC address */
+	rtsn_write(priv, MRMAC0, (mac_addr[0] << 8) | mac_addr[1]);
+	rtsn_write(priv, MRMAC1, (mac_addr[2] << 24) | (mac_addr[3] << 16) |
+		   (mac_addr[4] << 8) | mac_addr[5]);
+
+	/* Set xMII type */
+	rtsn_set_phy_interface(priv);
+	rtsn_set_rate(priv);
+
+	/* Enable MII */
+	rtsn_modify(priv, MPIC, MPIC_PSMCS_MASK | MPIC_PSMHT_MASK,
+		    MPIC_PSMCS_DEFAULT | MPIC_PSMHT_DEFAULT);
+
+	/* Link verification */
+	rtsn_modify(priv, MLVC, MLVC_PLV, MLVC_PLV);
+	ret = rtsn_reg_wait(priv, MLVC, MLVC_PLV, 0);
+	if (ret)
+		return ret;
+
+	return ret;
+}
+
+static int rtsn_hw_init(struct rtsn_private *priv)
+{
+	int ret;
+
+	ret = rtsn_reset(priv);
+	if (ret)
+		return ret;
+
+	/* Change to CONFIG mode */
+	ret = rtsn_change_mode(priv, OCR_OPC_CONFIG);
+	if (ret)
+		return ret;
+
+	ret = rtsn_axibmi_init(priv);
+	if (ret)
+		return ret;
+
+	rtsn_mhd_init(priv);
+
+	ret = rtsn_rmac_init(priv);
+	if (ret)
+		return ret;
+
+	ret = rtsn_change_mode(priv, OCR_OPC_DISABLE);
+	if (ret)
+		return ret;
+
+	/* Change to OPERATION mode */
+	ret = rtsn_change_mode(priv, OCR_OPC_OPERATION);
+
+	return ret;
+}
+
+static int rtsn_mii_access(struct mii_bus *bus, bool read, int phyad,
+			   int regad, u16 data)
+{
+	struct rtsn_private *priv = bus->priv;
+	u32 val;
+	int ret;
+
+	val = MPSM_PDA(phyad) | MPSM_PRA(regad) | MPSM_PSME;
+
+	if (!read)
+		val |= MPSM_PSMAD | MPSM_PRD_SET(data);
+
+	rtsn_write(priv, MPSM, val);
+
+	ret = rtsn_reg_wait(priv, MPSM, MPSM_PSME, 0);
+	if (ret)
+		return ret;
+
+	if (read)
+		ret = MPSM_PRD_GET(rtsn_read(priv, MPSM));
+
+	return ret;
+}
+
+static int rtsn_mii_read(struct mii_bus *bus, int addr, int regnum)
+{
+	return rtsn_mii_access(bus, true, addr, regnum, 0);
+}
+
+static int rtsn_mii_write(struct mii_bus *bus, int addr, int regnum, u16 val)
+{
+	return rtsn_mii_access(bus, false, addr, regnum, val);
+}
+
+static int rtsn_mdio_alloc(struct rtsn_private *priv)
+{
+	struct platform_device *pdev = priv->pdev;
+	struct device *dev = &pdev->dev;
+	struct device_node *mdio_node;
+	struct mii_bus *mii;
+	int ret;
+
+	mii = mdiobus_alloc();
+	if (!mii)
+		return -ENOMEM;
+
+	mdio_node = of_get_child_by_name(dev->of_node, "mdio");
+	if (!mdio_node) {
+		ret = -ENODEV;
+		goto out_free_bus;
+	}
+
+	/* Enter config mode before registering the MDIO bus */
+	ret = rtsn_reset(priv);
+	if (ret)
+		goto out_free_bus;
+
+	ret = rtsn_change_mode(priv, OCR_OPC_CONFIG);
+	if (ret)
+		goto out_free_bus;
+
+	rtsn_modify(priv, MPIC, MPIC_PSMCS_MASK | MPIC_PSMHT_MASK,
+		    MPIC_PSMCS_DEFAULT | MPIC_PSMHT_DEFAULT);
+
+	/* Register the MDIO bus */
+	mii->name = "rtsn_mii";
+	snprintf(mii->id, MII_BUS_ID_SIZE, "%s-%x",
+		 pdev->name, pdev->id);
+	mii->priv = priv;
+	mii->read = rtsn_mii_read;
+	mii->write = rtsn_mii_write;
+	mii->parent = dev;
+
+	ret = of_mdiobus_register(mii, mdio_node);
+	of_node_put(mdio_node);
+	if (ret)
+		goto out_free_bus;
+
+	priv->mii = mii;
+
+	return 0;
+
+out_free_bus:
+	mdiobus_free(mii);
+	return ret;
+}
+
+static void rtsn_mdio_free(struct rtsn_private *priv)
+{
+	mdiobus_unregister(priv->mii);
+	mdiobus_free(priv->mii);
+	priv->mii = NULL;
+}
+
+static void rtsn_adjust_link(struct net_device *ndev)
+{
+	struct rtsn_private *priv = netdev_priv(ndev);
+	struct phy_device *phydev = ndev->phydev;
+	bool new_state = false;
+	unsigned long flags;
+
+	spin_lock_irqsave(&priv->lock, flags);
+
+	if (phydev->link) {
+		if (phydev->speed != priv->speed) {
+			new_state = true;
+			priv->speed = phydev->speed;
+		}
+
+		if (!priv->link) {
+			new_state = true;
+			priv->link = phydev->link;
+		}
+	} else if (priv->link) {
+		new_state = true;
+		priv->link = 0;
+		priv->speed = 0;
+	}
+
+	if (new_state) {
+		/* Need to transition to CONFIG mode before reconfiguring and
+		 * then back to the original mode. Any state change to/from
+		 * CONFIG or OPERATION must go over DISABLED to stop Rx/Tx.
+		 */
+		enum rtsn_mode orgmode = rtsn_read_mode(priv);
+
+		/* Transit to CONFIG */
+		if (orgmode != OCR_OPC_CONFIG) {
+			if (orgmode != OCR_OPC_DISABLE &&
+			    rtsn_change_mode(priv, OCR_OPC_DISABLE))
+				goto out;
+			if (rtsn_change_mode(priv, OCR_OPC_CONFIG))
+				goto out;
+		}
+
+		rtsn_set_rate(priv);
+
+		/* Transition to original mode */
+		if (orgmode != OCR_OPC_CONFIG) {
+			if (rtsn_change_mode(priv, OCR_OPC_DISABLE))
+				goto out;
+			if (orgmode != OCR_OPC_DISABLE &&
+			    rtsn_change_mode(priv, orgmode))
+				goto out;
+		}
+	}
+out:
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	if (new_state)
+		phy_print_status(phydev);
+}
+
+static int rtsn_phy_init(struct rtsn_private *priv)
+{
+	struct device_node *np = priv->ndev->dev.parent->of_node;
+	struct phy_device *phydev;
+	struct device_node *phy;
+
+	priv->link = 0;
+
+	phy = of_parse_phandle(np, "phy-handle", 0);
+	if (!phy)
+		return -ENOENT;
+
+	phydev = of_phy_connect(priv->ndev, phy, rtsn_adjust_link, 0,
+				priv->iface);
+	of_node_put(phy);
+	if (!phydev)
+		return -ENOENT;
+
+	/* Only support full-duplex mode */
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_10baseT_Half_BIT);
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_100baseT_Half_BIT);
+	phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_1000baseT_Half_BIT);
+
+	phy_attached_info(phydev);
+
+	return 0;
+}
+
+static void rtsn_phy_deinit(struct rtsn_private *priv)
+{
+	phy_disconnect(priv->ndev->phydev);
+	priv->ndev->phydev = NULL;
+}
+
+static int rtsn_init(struct rtsn_private *priv)
+{
+	int ret;
+
+	ret = rtsn_desc_alloc(priv);
+	if (ret)
+		return ret;
+
+	ret = rtsn_dmac_init(priv);
+	if (ret)
+		goto error_free_desc;
+
+	ret = rtsn_hw_init(priv);
+	if (ret)
+		goto error_free_chain;
+
+	ret = rtsn_phy_init(priv);
+	if (ret)
+		goto error_free_chain;
+
+	ret = rtsn_request_irqs(priv);
+	if (ret)
+		goto error_free_phy;
+
+	return 0;
+error_free_phy:
+	rtsn_phy_deinit(priv);
+error_free_chain:
+	rtsn_chain_free(priv);
+error_free_desc:
+	rtsn_desc_free(priv);
+	return ret;
+}
+
+static void rtsn_deinit(struct rtsn_private *priv)
+{
+	rtsn_free_irqs(priv);
+	rtsn_phy_deinit(priv);
+	rtsn_chain_free(priv);
+	rtsn_desc_free(priv);
+}
+
+static void rtsn_parse_mac_address(struct device_node *np,
+				   struct net_device *ndev)
+{
+	struct rtsn_private *priv = netdev_priv(ndev);
+	u8 addr[ETH_ALEN];
+	u32 mrmac0;
+	u32 mrmac1;
+
+	/* Try to read address from Device Tree. */
+	if (!of_get_mac_address(np, addr)) {
+		eth_hw_addr_set(ndev, addr);
+		return;
+	}
+
+	/* Try to read address from device. */
+	mrmac0 = rtsn_read(priv, MRMAC0);
+	mrmac1 = rtsn_read(priv, MRMAC1);
+
+	addr[0] = (mrmac0 >>  8) & 0xff;
+	addr[1] = (mrmac0 >>  0) & 0xff;
+	addr[2] = (mrmac1 >> 24) & 0xff;
+	addr[3] = (mrmac1 >> 16) & 0xff;
+	addr[4] = (mrmac1 >>  8) & 0xff;
+	addr[5] = (mrmac1 >>  0) & 0xff;
+
+	if (is_valid_ether_addr(addr)) {
+		eth_hw_addr_set(ndev, addr);
+		return;
+	}
+
+	/* Fallback to a random address */
+	eth_hw_addr_random(ndev);
+}
+
+static int rtsn_open(struct net_device *ndev)
+{
+	struct rtsn_private *priv = netdev_priv(ndev);
+	int ret;
+
+	napi_enable(&priv->napi);
+
+	ret = rtsn_init(priv);
+	if (ret) {
+		napi_disable(&priv->napi);
+		return ret;
+	}
+
+	phy_start(ndev->phydev);
+
+	netif_start_queue(ndev);
+
+	return 0;
+}
+
+static int rtsn_stop(struct net_device *ndev)
+{
+	struct rtsn_private *priv = netdev_priv(ndev);
+
+	phy_stop(priv->ndev->phydev);
+	napi_disable(&priv->napi);
+	rtsn_change_mode(priv, OCR_OPC_DISABLE);
+	rtsn_deinit(priv);
+
+	return 0;
+}
+
+static netdev_tx_t rtsn_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+{
+	struct rtsn_private *priv = netdev_priv(ndev);
+	struct rtsn_ext_desc *desc;
+	int ret = NETDEV_TX_OK;
+	unsigned long flags;
+	dma_addr_t dma_addr;
+	int entry;
+
+	spin_lock_irqsave(&priv->lock, flags);
+
+	/* Drop packet if it won't fit in a single descriptor. */
+	if (skb->len >= TX_DS) {
+		priv->stats.tx_dropped++;
+		priv->stats.tx_errors++;
+		goto out;
+	}
+
+	if (priv->cur_tx - priv->dirty_tx > priv->num_tx_ring) {
+		netif_stop_subqueue(ndev, 0);
+		ret = NETDEV_TX_BUSY;
+		goto out;
+	}
+
+	if (skb_put_padto(skb, ETH_ZLEN))
+		goto out;
+
+	dma_addr = dma_map_single(ndev->dev.parent, skb->data, skb->len,
+				  DMA_TO_DEVICE);
+	if (dma_mapping_error(ndev->dev.parent, dma_addr)) {
+		dev_kfree_skb_any(skb);
+		goto out;
+	}
+
+	entry = priv->cur_tx % priv->num_tx_ring;
+	priv->tx_skb[entry] = skb;
+	desc = &priv->tx_ring[entry];
+	desc->dptr = cpu_to_le32(dma_addr);
+	desc->info_ds = cpu_to_le16(skb->len);
+	desc->info1 = cpu_to_le64(skb->len);
+
+	if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) {
+		skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+		priv->ts_tag++;
+		desc->info_ds |= cpu_to_le16(TXC);
+		desc->info = priv->ts_tag;
+	}
+
+	skb_tx_timestamp(skb);
+	dma_wmb();
+
+	desc->die_dt = DT_FSINGLE | D_DIE;
+	priv->cur_tx++;
+
+	/* Start xmit */
+	rtsn_write(priv, TRCR0, BIT(TX_CHAIN_IDX));
+out:
+	spin_unlock_irqrestore(&priv->lock, flags);
+	return ret;
+}
+
+static void rtsn_get_stats64(struct net_device *ndev,
+			     struct rtnl_link_stats64 *storage)
+{
+	struct rtsn_private *priv = netdev_priv(ndev);
+	*storage = priv->stats;
+}
+
+static int rtsn_do_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd)
+{
+	if (!netif_running(ndev))
+		return -ENODEV;
+
+	return phy_do_ioctl_running(ndev, ifr, cmd);
+}
+
+static int rtsn_hwtstamp_get(struct net_device *ndev,
+			     struct kernel_hwtstamp_config *config)
+{
+	struct rcar_gen4_ptp_private *ptp_priv;
+	struct rtsn_private *priv;
+
+	if (!netif_running(ndev))
+		return -ENODEV;
+
+	priv = netdev_priv(ndev);
+	ptp_priv = priv->ptp_priv;
+
+	config->flags = 0;
+
+	config->tx_type =
+		ptp_priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF;
+
+	switch (ptp_priv->tstamp_rx_ctrl & RCAR_GEN4_RXTSTAMP_TYPE) {
+	case RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT:
+		config->rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT;
+		break;
+	case RCAR_GEN4_RXTSTAMP_TYPE_ALL:
+		config->rx_filter = HWTSTAMP_FILTER_ALL;
+		break;
+	default:
+		config->rx_filter = HWTSTAMP_FILTER_NONE;
+		break;
+	}
+
+	return 0;
+}
+
+static int rtsn_hwtstamp_set(struct net_device *ndev,
+			     struct kernel_hwtstamp_config *config,
+			     struct netlink_ext_ack *extack)
+{
+	struct rcar_gen4_ptp_private *ptp_priv;
+	struct rtsn_private *priv;
+	u32 tstamp_rx_ctrl;
+	u32 tstamp_tx_ctrl;
+
+	if (!netif_running(ndev))
+		return -ENODEV;
+
+	priv = netdev_priv(ndev);
+	ptp_priv = priv->ptp_priv;
+
+	if (config->flags)
+		return -EINVAL;
+
+	switch (config->tx_type) {
+	case HWTSTAMP_TX_OFF:
+		tstamp_tx_ctrl = 0;
+		break;
+	case HWTSTAMP_TX_ON:
+		tstamp_tx_ctrl = RCAR_GEN4_TXTSTAMP_ENABLED;
+		break;
+	default:
+		return -ERANGE;
+	}
+
+	switch (config->rx_filter) {
+	case HWTSTAMP_FILTER_NONE:
+		tstamp_rx_ctrl = 0;
+		break;
+	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
+		tstamp_rx_ctrl = RCAR_GEN4_RXTSTAMP_ENABLED |
+			RCAR_GEN4_RXTSTAMP_TYPE_V2_L2_EVENT;
+		break;
+	default:
+		config->rx_filter = HWTSTAMP_FILTER_ALL;
+		tstamp_rx_ctrl = RCAR_GEN4_RXTSTAMP_ENABLED |
+			RCAR_GEN4_RXTSTAMP_TYPE_ALL;
+		break;
+	}
+
+	ptp_priv->tstamp_tx_ctrl = tstamp_tx_ctrl;
+	ptp_priv->tstamp_rx_ctrl = tstamp_rx_ctrl;
+
+	return 0;
+}
+
+static const struct net_device_ops rtsn_netdev_ops = {
+	.ndo_open		= rtsn_open,
+	.ndo_stop		= rtsn_stop,
+	.ndo_start_xmit		= rtsn_start_xmit,
+	.ndo_get_stats64	= rtsn_get_stats64,
+	.ndo_eth_ioctl		= rtsn_do_ioctl,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_set_mac_address	= eth_mac_addr,
+	.ndo_hwtstamp_set	= rtsn_hwtstamp_set,
+	.ndo_hwtstamp_get	= rtsn_hwtstamp_get,
+};
+
+static int rtsn_get_ts_info(struct net_device *ndev,
+			    struct ethtool_ts_info *info)
+{
+	struct rtsn_private *priv = netdev_priv(ndev);
+
+	info->phc_index = ptp_clock_index(priv->ptp_priv->clock);
+	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
+		SOF_TIMESTAMPING_RX_SOFTWARE |
+		SOF_TIMESTAMPING_SOFTWARE |
+		SOF_TIMESTAMPING_TX_HARDWARE |
+		SOF_TIMESTAMPING_RX_HARDWARE |
+		SOF_TIMESTAMPING_RAW_HARDWARE;
+	info->tx_types = BIT(HWTSTAMP_TX_OFF) | BIT(HWTSTAMP_TX_ON);
+	info->rx_filters = BIT(HWTSTAMP_FILTER_NONE) | BIT(HWTSTAMP_FILTER_ALL);
+
+	return 0;
+}
+
+static const struct ethtool_ops rtsn_ethtool_ops = {
+	.nway_reset		= phy_ethtool_nway_reset,
+	.get_link		= ethtool_op_get_link,
+	.get_ts_info		= rtsn_get_ts_info,
+	.get_link_ksettings	= phy_ethtool_get_link_ksettings,
+	.set_link_ksettings	= phy_ethtool_set_link_ksettings,
+};
+
+static const struct of_device_id rtsn_match_table[] = {
+	{ .compatible = "renesas,r8a779g0-ethertsn", },
+	{ /* Sentinel */ }
+};
+
+MODULE_DEVICE_TABLE(of, rtsn_match_table);
+
+static int rtsn_probe(struct platform_device *pdev)
+{
+	struct rtsn_private *priv;
+	struct net_device *ndev;
+	struct resource *res;
+	int ret;
+
+	ndev = alloc_etherdev_mqs(sizeof(struct rtsn_private), TX_NUM_CHAINS,
+				  RX_NUM_CHAINS);
+	if (!ndev)
+		return -ENOMEM;
+
+	priv = netdev_priv(ndev);
+	priv->pdev = pdev;
+	priv->ndev = ndev;
+	priv->ptp_priv = rcar_gen4_ptp_alloc(pdev);
+
+	spin_lock_init(&priv->lock);
+	platform_set_drvdata(pdev, priv);
+
+	priv->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(priv->clk)) {
+		ret = PTR_ERR(priv->clk);
+		goto error_free;
+	}
+
+	priv->reset = devm_reset_control_get(&pdev->dev, NULL);
+	if (IS_ERR(priv->reset)) {
+		ret = PTR_ERR(priv->reset);
+		goto error_free;
+	}
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "tsnes");
+	if (!res) {
+		dev_err(&pdev->dev, "Can't find tsnes resource\n");
+		ret = -EINVAL;
+		goto error_free;
+	}
+
+	priv->base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(priv->base)) {
+		ret = PTR_ERR(priv->base);
+		goto error_free;
+	}
+
+	SET_NETDEV_DEV(ndev, &pdev->dev);
+
+	ndev->features = NETIF_F_RXCSUM;
+	ndev->hw_features = NETIF_F_RXCSUM;
+	ndev->base_addr = res->start;
+	ndev->netdev_ops = &rtsn_netdev_ops;
+	ndev->ethtool_ops = &rtsn_ethtool_ops;
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "gptp");
+	if (!res) {
+		dev_err(&pdev->dev, "Can't find gptp resource\n");
+		ret = -EINVAL;
+		goto error_free;
+	}
+
+	priv->ptp_priv->addr = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(priv->ptp_priv->addr)) {
+		ret = PTR_ERR(priv->ptp_priv->addr);
+		goto error_free;
+	}
+
+	ret = rtsn_get_phy_params(priv);
+	if (ret)
+		goto error_free;
+
+	pm_runtime_enable(&pdev->dev);
+	pm_runtime_get_sync(&pdev->dev);
+
+	netif_napi_add(ndev, &priv->napi, rtsn_poll);
+
+	rtsn_parse_mac_address(pdev->dev.of_node, ndev);
+
+	dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+
+	device_set_wakeup_capable(&pdev->dev, 1);
+
+	ret = rcar_gen4_ptp_register(priv->ptp_priv, RCAR_GEN4_PTP_REG_LAYOUT,
+				     clk_get_rate(priv->clk));
+	if (ret)
+		goto error_pm;
+
+	ret = rtsn_mdio_alloc(priv);
+	if (ret)
+		goto error_ptp;
+
+	ret = register_netdev(ndev);
+	if (ret)
+		goto error_mdio;
+
+	netdev_info(ndev, "MAC address %pM\n", ndev->dev_addr);
+
+	return 0;
+
+error_mdio:
+	rtsn_mdio_free(priv);
+error_ptp:
+	rcar_gen4_ptp_unregister(priv->ptp_priv);
+error_pm:
+	netif_napi_del(&priv->napi);
+	rtsn_change_mode(priv, OCR_OPC_DISABLE);
+	pm_runtime_put_sync(&pdev->dev);
+	pm_runtime_disable(&pdev->dev);
+error_free:
+	free_netdev(ndev);
+
+	return ret;
+}
+
+static int rtsn_remove(struct platform_device *pdev)
+{
+	struct rtsn_private *priv = platform_get_drvdata(pdev);
+
+	unregister_netdev(priv->ndev);
+	rtsn_mdio_free(priv);
+	rcar_gen4_ptp_unregister(priv->ptp_priv);
+	rtsn_change_mode(priv, OCR_OPC_DISABLE);
+	netif_napi_del(&priv->napi);
+
+	pm_runtime_put_sync(&pdev->dev);
+	pm_runtime_disable(&pdev->dev);
+
+	free_netdev(priv->ndev);
+
+	return 0;
+}
+
+static struct platform_driver rtsn_driver = {
+	.probe		= rtsn_probe,
+	.remove		= rtsn_remove,
+	.driver	= {
+		.name	= "rtsn",
+		.of_match_table	= rtsn_match_table,
+	}
+};
+module_platform_driver(rtsn_driver);
+
+MODULE_AUTHOR("Phong Hoang, Niklas Söderlund");
+MODULE_DESCRIPTION("Renesas Ethernet-TSN device driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/renesas/rtsn.h b/drivers/net/ethernet/renesas/rtsn.h
new file mode 100644
index 000000000000..3183e80d7e6b
--- /dev/null
+++ b/drivers/net/ethernet/renesas/rtsn.h
@@ -0,0 +1,464 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+/* Renesas Ethernet-TSN device driver
+ *
+ * Copyright (C) 2022 Renesas Electronics Corporation
+ * Copyright (C) 2023 Niklas Söderlund <niklas.soderlund@ragnatech.se>
+ */
+
+#ifndef __RTSN_H__
+#define __RTSN_H__
+
+#include <linux/types.h>
+
+#define AXIBMI	0x0000
+#define TSNMHD	0x1000
+#define RMSO	0x2000
+#define RMRO	0x3800
+
+enum rtsn_reg {
+	AXIWC		= AXIBMI + 0x0000,
+	AXIRC		= AXIBMI + 0x0004,
+	TDPC0		= AXIBMI + 0x0010,
+	TFT		= AXIBMI + 0x0090,
+	TATLS0		= AXIBMI + 0x00a0,
+	TATLS1		= AXIBMI + 0x00a4,
+	TATLR		= AXIBMI + 0x00a8,
+	RATLS0		= AXIBMI + 0x00b0,
+	RATLS1		= AXIBMI + 0x00b4,
+	RATLR		= AXIBMI + 0x00b8,
+	TSA0		= AXIBMI + 0x00c0,
+	TSS0		= AXIBMI + 0x00c4,
+	TRCR0		= AXIBMI + 0x0140,
+	RIDAUAS0	= AXIBMI + 0x0180,
+	RR		= AXIBMI + 0x0200,
+	TATS		= AXIBMI + 0x0210,
+	TATSR0		= AXIBMI + 0x0214,
+	TATSR1		= AXIBMI + 0x0218,
+	TATSR2		= AXIBMI + 0x021c,
+	RATS		= AXIBMI + 0x0220,
+	RATSR0		= AXIBMI + 0x0224,
+	RATSR1		= AXIBMI + 0x0228,
+	RATSR2		= AXIBMI + 0x022c,
+	RIDASM0		= AXIBMI + 0x0240,
+	RIDASAM0	= AXIBMI + 0x0244,
+	RIDACAM0	= AXIBMI + 0x0248,
+	EIS0		= AXIBMI + 0x0300,
+	EIE0		= AXIBMI + 0x0304,
+	EID0		= AXIBMI + 0x0308,
+	EIS1		= AXIBMI + 0x0310,
+	EIE1		= AXIBMI + 0x0314,
+	EID1		= AXIBMI + 0x0318,
+	TCEIS0		= AXIBMI + 0x0340,
+	TCEIE0		= AXIBMI + 0x0344,
+	TCEID0		= AXIBMI + 0x0348,
+	RFSEIS0		= AXIBMI + 0x04c0,
+	RFSEIE0		= AXIBMI + 0x04c4,
+	RFSEID0		= AXIBMI + 0x04c8,
+	RFEIS0		= AXIBMI + 0x0540,
+	RFEIE0		= AXIBMI + 0x0544,
+	RFEID0		= AXIBMI + 0x0548,
+	RCEIS0		= AXIBMI + 0x05c0,
+	RCEIE0		= AXIBMI + 0x05c4,
+	RCEID0		= AXIBMI + 0x05c8,
+	RIDAOIS		= AXIBMI + 0x0640,
+	RIDAOIE		= AXIBMI + 0x0644,
+	RIDAOID		= AXIBMI + 0x0648,
+	TSFEIS		= AXIBMI + 0x06c0,
+	TSFEIE		= AXIBMI + 0x06c4,
+	TSFEID		= AXIBMI + 0x06c8,
+	TSCEIS		= AXIBMI + 0x06d0,
+	TSCEIE		= AXIBMI + 0x06d4,
+	TSCEID		= AXIBMI + 0x06d8,
+	DIS		= AXIBMI + 0x0b00,
+	DIE		= AXIBMI + 0x0b04,
+	DID		= AXIBMI + 0x0b08,
+	TDIS0		= AXIBMI + 0x0b10,
+	TDIE0		= AXIBMI + 0x0b14,
+	TDID0		= AXIBMI + 0x0b18,
+	RDIS0		= AXIBMI + 0x0b90,
+	RDIE0		= AXIBMI + 0x0b94,
+	RDID0		= AXIBMI + 0x0b98,
+	TSDIS		= AXIBMI + 0x0c10,
+	TSDIE		= AXIBMI + 0x0c14,
+	TSDID		= AXIBMI + 0x0c18,
+	GPOUT		= AXIBMI + 0x6000,
+
+	OCR		= TSNMHD + 0x0000,
+	OSR		= TSNMHD + 0x0004,
+	SWR		= TSNMHD + 0x0008,
+	SIS		= TSNMHD + 0x000c,
+	GIS		= TSNMHD + 0x0010,
+	GIE		= TSNMHD + 0x0014,
+	GID		= TSNMHD + 0x0018,
+	TIS1		= TSNMHD + 0x0020,
+	TIE1		= TSNMHD + 0x0024,
+	TID1		= TSNMHD + 0x0028,
+	TIS2		= TSNMHD + 0x0030,
+	TIE2		= TSNMHD + 0x0034,
+	TID2		= TSNMHD + 0x0038,
+	RIS		= TSNMHD + 0x0040,
+	RIE		= TSNMHD + 0x0044,
+	RID		= TSNMHD + 0x0048,
+	TGC1		= TSNMHD + 0x0050,
+	TGC2		= TSNMHD + 0x0054,
+	TFS0		= TSNMHD + 0x0060,
+	TCF0		= TSNMHD + 0x0070,
+	TCR1		= TSNMHD + 0x0080,
+	TCR2		= TSNMHD + 0x0084,
+	TCR3		= TSNMHD + 0x0088,
+	TCR4		= TSNMHD + 0x008c,
+	TMS0		= TSNMHD + 0x0090,
+	TSR1		= TSNMHD + 0x00b0,
+	TSR2		= TSNMHD + 0x00b4,
+	TSR3		= TSNMHD + 0x00b8,
+	TSR4		= TSNMHD + 0x00bc,
+	TSR5		= TSNMHD + 0x00c0,
+	RGC		= TSNMHD + 0x00d0,
+	RDFCR		= TSNMHD + 0x00d4,
+	RCFCR		= TSNMHD + 0x00d8,
+	REFCNCR		= TSNMHD + 0x00dc,
+	RSR1		= TSNMHD + 0x00e0,
+	RSR2		= TSNMHD + 0x00e4,
+	RSR3		= TSNMHD + 0x00e8,
+	TCIS		= TSNMHD + 0x01e0,
+	TCIE		= TSNMHD + 0x01e4,
+	TCID		= TSNMHD + 0x01e8,
+	TPTPC		= TSNMHD + 0x01f0,
+	TTML		= TSNMHD + 0x01f4,
+	TTJ		= TSNMHD + 0x01f8,
+	TCC		= TSNMHD + 0x0200,
+	TCS		= TSNMHD + 0x0204,
+	TGS		= TSNMHD + 0x020c,
+	TACST0		= TSNMHD + 0x0210,
+	TACST1		= TSNMHD + 0x0214,
+	TACST2		= TSNMHD + 0x0218,
+	TALIT0		= TSNMHD + 0x0220,
+	TALIT1		= TSNMHD + 0x0224,
+	TALIT2		= TSNMHD + 0x0228,
+	TAEN0		= TSNMHD + 0x0230,
+	TAEN1		= TSNMHD + 0x0234,
+	TASFE		= TSNMHD + 0x0240,
+	TACLL0		= TSNMHD + 0x0250,
+	TACLL1		= TSNMHD + 0x0254,
+	TACLL2		= TSNMHD + 0x0258,
+	CACC		= TSNMHD + 0x0260,
+	CCS		= TSNMHD + 0x0264,
+	CAIV0		= TSNMHD + 0x0270,
+	CAUL0		= TSNMHD + 0x0290,
+	TOCST0		= TSNMHD + 0x0300,
+	TOCST1		= TSNMHD + 0x0304,
+	TOCST2		= TSNMHD + 0x0308,
+	TOLIT0		= TSNMHD + 0x0310,
+	TOLIT1		= TSNMHD + 0x0314,
+	TOLIT2		= TSNMHD + 0x0318,
+	TOEN0		= TSNMHD + 0x0320,
+	TOEN1		= TSNMHD + 0x0324,
+	TOSFE		= TSNMHD + 0x0330,
+	TCLR0		= TSNMHD + 0x0340,
+	TCLR1		= TSNMHD + 0x0344,
+	TCLR2		= TSNMHD + 0x0348,
+	TSMS		= TSNMHD + 0x0350,
+	COCC		= TSNMHD + 0x0360,
+	COIV0		= TSNMHD + 0x03b0,
+	COUL0		= TSNMHD + 0x03d0,
+	QSTMACU0	= TSNMHD + 0x0400,
+	QSTMACD0	= TSNMHD + 0x0404,
+	QSTMAMU0	= TSNMHD + 0x0408,
+	QSTMAMD0	= TSNMHD + 0x040c,
+	QSFTVL0		= TSNMHD + 0x0410,
+	QSFTVLM0	= TSNMHD + 0x0414,
+	QSFTMSD0	= TSNMHD + 0x0418,
+	QSFTGMI0	= TSNMHD + 0x041c,
+	QSFTLS		= TSNMHD + 0x0600,
+	QSFTLIS		= TSNMHD + 0x0604,
+	QSFTLIE		= TSNMHD + 0x0608,
+	QSFTLID		= TSNMHD + 0x060c,
+	QSMSMC		= TSNMHD + 0x0610,
+	QSGTMC		= TSNMHD + 0x0614,
+	QSEIS		= TSNMHD + 0x0618,
+	QSEIE		= TSNMHD + 0x061c,
+	QSEID		= TSNMHD + 0x0620,
+	QGACST0		= TSNMHD + 0x0630,
+	QGACST1		= TSNMHD + 0x0634,
+	QGACST2		= TSNMHD + 0x0638,
+	QGALIT1		= TSNMHD + 0x0640,
+	QGALIT2		= TSNMHD + 0x0644,
+	QGAEN0		= TSNMHD + 0x0648,
+	QGAEN1		= TSNMHD + 0x074c,
+	QGIGS		= TSNMHD + 0x0650,
+	QGGC		= TSNMHD + 0x0654,
+	QGATL0		= TSNMHD + 0x0664,
+	QGATL1		= TSNMHD + 0x0668,
+	QGATL2		= TSNMHD + 0x066c,
+	QGOCST0		= TSNMHD + 0x0670,
+	QGOCST1		= TSNMHD + 0x0674,
+	QGOCST2		= TSNMHD + 0x0678,
+	QGOLIT0		= TSNMHD + 0x067c,
+	QGOLIT1		= TSNMHD + 0x0680,
+	QGOLIT2		= TSNMHD + 0x0684,
+	QGOEN0		= TSNMHD + 0x0688,
+	QGOEN1		= TSNMHD + 0x068c,
+	QGTRO		= TSNMHD + 0x0690,
+	QGTR1		= TSNMHD + 0x0694,
+	QGTR2		= TSNMHD + 0x0698,
+	QGFSMS		= TSNMHD + 0x069c,
+	QTMIS		= TSNMHD + 0x06e0,
+	QTMIE		= TSNMHD + 0x06e4,
+	QTMID		= TSNMHD + 0x06e8,
+	QMEC		= TSNMHD + 0x0700,
+	QMMC		= TSNMHD + 0x0704,
+	QRFDC		= TSNMHD + 0x0708,
+	QYFDC		= TSNMHD + 0x070c,
+	QVTCMC0		= TSNMHD + 0x0710,
+	QMCBSC0		= TSNMHD + 0x0750,
+	QMCIRC0		= TSNMHD + 0x0790,
+	QMEBSC0		= TSNMHD + 0x07d0,
+	QMEIRC0		= TSNMHD + 0x0710,
+	QMCFC		= TSNMHD + 0x0850,
+	QMEIS		= TSNMHD + 0x0860,
+	QMEIE		= TSNMHD + 0x0864,
+	QMEID		= TSNMHD + 0x086c,
+	QSMFC0		= TSNMHD + 0x0870,
+	QMSPPC0		= TSNMHD + 0x08b0,
+	QMSRPC0		= TSNMHD + 0x08f0,
+	QGPPC0		= TSNMHD + 0x0930,
+	QGRPC0		= TSNMHD + 0x0950,
+	QMDPC0		= TSNMHD + 0x0970,
+	QMGPC0		= TSNMHD + 0x09b0,
+	QMYPC0		= TSNMHD + 0x09f0,
+	QMRPC0		= TSNMHD + 0x0a30,
+	MQSTMACU	= TSNMHD + 0x0a70,
+	MQSTMACD	= TSNMHD + 0x0a74,
+	MQSTMAMU	= TSNMHD + 0x0a78,
+	MQSTMAMD	= TSNMHD + 0x0a7c,
+	MQSFTVL		= TSNMHD + 0x0a80,
+	MQSFTVLM	= TSNMHD + 0x0a84,
+	MQSFTMSD	= TSNMHD + 0x0a88,
+	MQSFTGMI	= TSNMHD + 0x0a8c,
+
+	CFCR0		= RMSO + 0x0800,
+	FMSCR		= RMSO + 0x0c10,
+
+	MMC		= RMRO + 0x0000,
+	MPSM		= RMRO + 0x0010,
+	MPIC		= RMRO + 0x0014,
+	MTFFC		= RMRO + 0x0020,
+	MTPFC		= RMRO + 0x0024,
+	MTATC0		= RMRO + 0x0040,
+	MRGC		= RMRO + 0x0080,
+	MRMAC0		= RMRO + 0x0084,
+	MRMAC1		= RMRO + 0x0088,
+	MRAFC		= RMRO + 0x008c,
+	MRSCE		= RMRO + 0x0090,
+	MRSCP		= RMRO + 0x0094,
+	MRSCC		= RMRO + 0x0098,
+	MRFSCE		= RMRO + 0x009c,
+	MRFSCP		= RMRO + 0x00a0,
+	MTRC		= RMRO + 0x00a4,
+	MPFC		= RMRO + 0x0100,
+	MLVC		= RMRO + 0x0340,
+	MEEEC		= RMRO + 0x0350,
+	MLBC		= RMRO + 0x0360,
+	MGMR		= RMRO + 0x0400,
+	MMPFTCT		= RMRO + 0x0410,
+	MAPFTCT		= RMRO + 0x0414,
+	MPFRCT		= RMRO + 0x0418,
+	MFCICT		= RMRO + 0x041c,
+	MEEECT		= RMRO + 0x0420,
+	MEIS		= RMRO + 0x0500,
+	MEIE		= RMRO + 0x0504,
+	MEID		= RMRO + 0x0508,
+	MMIS0		= RMRO + 0x0510,
+	MMIE0		= RMRO + 0x0514,
+	MMID0		= RMRO + 0x0518,
+	MMIS1		= RMRO + 0x0520,
+	MMIE1		= RMRO + 0x0524,
+	MMID1		= RMRO + 0x0528,
+	MMIS2		= RMRO + 0x0530,
+	MMIE2		= RMRO + 0x0534,
+	MMID2		= RMRO + 0x0538,
+	MXMS		= RMRO + 0x0600,
+
+};
+
+/* AXIBMI */
+#define RR_RATRR		BIT(0)
+#define RR_TATRR		BIT(1)
+#define RR_RST			(RR_RATRR | RR_TATRR)
+#define RR_RST_COMPLETE		0x03
+
+#define AXIWC_DEFAULT		0xffff
+#define AXIRC_DEFAULT		0xffff
+
+#define TATLS0_TEDE		BIT(1)
+#define TATLS0_TATEN_SHIFT	24
+#define TATLS0_TATEN(n)		((n) << TATLS0_TATEN_SHIFT)
+#define TATLR_TATL		BIT(31)
+
+#define RATLS0_RETS		BIT(2)
+#define RATLS0_REDE		BIT(3)
+#define RATLS0_RATEN_SHIFT	24
+#define RATLS0_RATEN(n)		((n) << RATLS0_RATEN_SHIFT)
+#define RATLR_RATL		BIT(31)
+
+#define DIE_DID_TDICX(n)	BIT((n))
+#define DIE_DID_RDICX(n)	BIT((n) + 8)
+#define TDIE_TDID_TDX(n)	BIT(n)
+#define RDIE_RDID_RDX(n)	BIT(n)
+#define TDIS_TDS(n)		BIT(n)
+#define RDIS_RDS(n)		BIT(n)
+
+/* MHD */
+#define OSR_OPS			0x07
+#define SWR_SWR			BIT(0)
+
+#define TGC1_TQTM_SFM		0xff00
+#define TGC1_STTV_DEFAULT	0x03
+
+#define TMS_MFS_MAX		0x2800
+
+/* RMAC System */
+#define CFCR_SDID(n)		((n) << 16)
+#define FMSCR_FMSIE(n)		((n) << 0)
+
+/* RMAC */
+#define MPIC_PIS_MASK		GENMASK(1, 0)
+#define MPIC_PIS_MII		0
+#define MPIC_PIS_RMII		0x01
+#define MPIC_PIS_GMII		0x02
+#define MPIC_PIS_RGMII		0x03
+#define MPIC_LSC_SHIFT		2
+#define MPIC_LSC_MASK		GENMASK(3, MPIC_LSC_SHIFT)
+#define MPIC_LSC_10M		(0 << MPIC_LSC_SHIFT)
+#define MPIC_LSC_100M		(0x01 << MPIC_LSC_SHIFT)
+#define MPIC_LSC_1G		(0x02 << MPIC_LSC_SHIFT)
+#define MPIC_PSMCS_SHIFT	16
+#define MPIC_PSMCS_MASK		GENMASK(21, MPIC_PSMCS_SHIFT)
+#define MPIC_PSMCS_DEFAULT	(0x0a << MPIC_PSMCS_SHIFT)
+#define MPIC_PSMHT_SHIFT	24
+#define MPIC_PSMHT_MASK		GENMASK(26, MPIC_PSMHT_SHIFT)
+#define MPIC_PSMHT_DEFAULT	(0x07 << MPIC_PSMHT_SHIFT)
+
+#define MLVC_PASE		BIT(8)
+#define MLVC_PSE		BIT(16)
+#define MLVC_PLV		BIT(17)
+
+#define MPSM_PSME		BIT(0)
+#define MPSM_PSMAD		BIT(1)
+#define MPSM_PDA_SHIFT		3
+#define MPSM_PDA_MASK		GENMASK(7, 3)
+#define MPSM_PDA(n)		(((n) << MPSM_PDA_SHIFT) & MPSM_PDA_MASK)
+#define MPSM_PRA_SHIFT		8
+#define MPSM_PRA_MASK		GENMASK(12, 8)
+#define MPSM_PRA(n)		(((n) << MPSM_PRA_SHIFT) & MPSM_PRA_MASK)
+#define MPSM_PRD_SHIFT		16
+#define MPSM_PRD_SET(n)		((n) << MPSM_PRD_SHIFT)
+#define MPSM_PRD_GET(n)		((n) >> MPSM_PRD_SHIFT)
+
+#define GPOUT_RDM		BIT(13)
+#define GPOUT_TDM		BIT(14)
+
+/* RTSN */
+#define RTSN_INTERVAL_US	1000
+#define RTSN_TIMEOUT_US		1000000
+
+#define TX_NUM_CHAINS		1
+#define RX_NUM_CHAINS		1
+
+#define TX_CHAIN_SIZE		1024
+#define RX_CHAIN_SIZE		1024
+
+#define TX_CHAIN_IDX		0
+#define RX_CHAIN_IDX		0
+
+#define TX_CHAIN_ADDR_OFFSET	(sizeof(struct rtsn_desc) * TX_CHAIN_IDX)
+#define RX_CHAIN_ADDR_OFFSET	(sizeof(struct rtsn_desc) * RX_CHAIN_IDX)
+
+#define PKT_BUF_SZ		1584
+#define RTSN_ALIGN		128
+
+enum rtsn_mode {
+	OCR_OPC_DISABLE,
+	OCR_OPC_CONFIG,
+	OCR_OPC_OPERATION,
+};
+
+/* Descriptors */
+enum RX_DS_CC_BIT {
+	RX_DS	= 0x0fff, /* Data size */
+	RX_TR	= 0x1000, /* Truncation indication */
+	RX_EI	= 0x2000, /* Error indication */
+	RX_PS	= 0xc000, /* Padding selection */
+};
+
+enum TX_FS_TAGL_BIT {
+	TX_DS	= 0x0fff, /* Data size */
+	TX_TAGL	= 0xf000, /* Frame tag LSBs */
+};
+
+enum DIE_DT {
+	/* HW/SW arbitration */
+	DT_FEMPTY_IS	= 0x10,
+	DT_FEMPTY_IC	= 0x20,
+	DT_FEMPTY_ND	= 0x30,
+	DT_FEMPTY	= 0x40,
+	DT_FEMPTY_START	= 0x50,
+	DT_FEMPTY_MID	= 0x60,
+	DT_FEMPTY_END	= 0x70,
+
+	/* Frame data */
+	DT_FSINGLE	= 0x80,
+	DT_FSTART	= 0x90,
+	DT_FMID		= 0xa0,
+	DT_FEND		= 0xb0,
+
+	/* Chain control */
+	DT_LEMPTY	= 0xc0,
+	DT_EEMPTY	= 0xd0,
+	DT_LINK		= 0xe0,
+	DT_EOS		= 0xf0,
+
+	DT_MASK		= 0xf0,
+	D_DIE		= 0x08,
+};
+
+struct rtsn_desc {
+	__le16 info_ds;
+	__u8 info;
+	u8 die_dt;
+	__le32 dptr;
+} __packed;
+
+struct rtsn_ts_desc {
+	__le16 info_ds;
+	__u8 info;
+	u8 die_dt;
+	__le32 dptr;
+	__le32 ts_nsec;
+	__le32 ts_sec;
+} __packed;
+
+struct rtsn_ext_desc {
+	__le16 info_ds;
+	__u8 info;
+	u8 die_dt;
+	__le32 dptr;
+	__le64 info1;
+} __packed;
+
+struct rtsn_ext_ts_desc {
+	__le16 info_ds;
+	__u8 info;
+	u8 die_dt;
+	__le32 dptr;
+	__le64 info1;
+	__le32 ts_nsec;
+	__le32 ts_sec;
+} __packed;
+
+enum EXT_INFO_DS_BIT {
+	TXC = 0x4000,
+};
+
+#endif
-- 
cgit v1.2.3-58-ga151


From d7dd13ea54af8496aca2762a758d817d6813e81c Mon Sep 17 00:00:00 2001
From: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Date: Sun, 16 Jun 2024 22:27:34 +0300
Subject: wifi: rtw88: usb: Further limit the TX aggregation

Currently the number of frames sent to the chip in a single USB Request
Block is limited only by the size of the TX buffer, which is 20 KiB.
Testing reveals that as many as 13 frames get aggregated. This is more
than what any of the chips would like to receive. RTL8822CU, RTL8822BU,
and RTL8821CU want at most 3 frames, and RTL8723DU wants only 1 frame
per URB.

RTL8723DU in particular reliably malfunctions during a speed test if it
receives more than 1 frame per URB. All traffic seems to stop. Pinging
the AP no longer works.

Fix this problem by limiting the number of frames sent to the chip in a
single URB according to what each chip likes.

Also configure RTL8822CU, RTL8822BU, and RTL8821CU to expect 3 frames
per URB.

RTL8703B may or may not be found in USB devices. Declare that it wants
only 1 frame per URB, just in case.

Tested with RTL8723DU and RTL8811CU.

Cc: stable@vger.kernel.org
Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/cb46ea35-7e59-4742-9c1f-01ceeaad36fb@gmail.com
---
 drivers/net/wireless/realtek/rtw88/mac.c      | 9 +++++++++
 drivers/net/wireless/realtek/rtw88/main.h     | 2 ++
 drivers/net/wireless/realtek/rtw88/reg.h      | 1 +
 drivers/net/wireless/realtek/rtw88/rtw8703b.c | 1 +
 drivers/net/wireless/realtek/rtw88/rtw8723d.c | 1 +
 drivers/net/wireless/realtek/rtw88/rtw8821c.c | 1 +
 drivers/net/wireless/realtek/rtw88/rtw8822b.c | 1 +
 drivers/net/wireless/realtek/rtw88/rtw8822c.c | 1 +
 drivers/net/wireless/realtek/rtw88/usb.c      | 4 +++-
 9 files changed, 20 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw88/mac.c b/drivers/net/wireless/realtek/rtw88/mac.c
index 0dba8aae7716..564f5988ee82 100644
--- a/drivers/net/wireless/realtek/rtw88/mac.c
+++ b/drivers/net/wireless/realtek/rtw88/mac.c
@@ -1201,6 +1201,15 @@ static int __priority_queue_cfg(struct rtw_dev *rtwdev,
 	rtw_write16(rtwdev, REG_FIFOPAGE_CTRL_2 + 2, fifo->rsvd_boundary);
 	rtw_write16(rtwdev, REG_BCNQ1_BDNY_V1, fifo->rsvd_boundary);
 	rtw_write32(rtwdev, REG_RXFF_BNDY, chip->rxff_size - C2H_PKT_BUF - 1);
+
+	if (rtwdev->hci.type == RTW_HCI_TYPE_USB) {
+		rtw_write8_mask(rtwdev, REG_AUTO_LLT_V1, BIT_MASK_BLK_DESC_NUM,
+				chip->usb_tx_agg_desc_num);
+
+		rtw_write8(rtwdev, REG_AUTO_LLT_V1 + 3, chip->usb_tx_agg_desc_num);
+		rtw_write8_set(rtwdev, REG_TXDMA_OFFSET_CHK + 1, BIT(1));
+	}
+
 	rtw_write8_set(rtwdev, REG_AUTO_LLT_V1, BIT_AUTO_INIT_LLT_V1);
 
 	if (!check_hw_ready(rtwdev, REG_AUTO_LLT_V1, BIT_AUTO_INIT_LLT_V1, 0))
diff --git a/drivers/net/wireless/realtek/rtw88/main.h b/drivers/net/wireless/realtek/rtw88/main.h
index 49894331f7b4..49a3fd4fb7dc 100644
--- a/drivers/net/wireless/realtek/rtw88/main.h
+++ b/drivers/net/wireless/realtek/rtw88/main.h
@@ -1197,6 +1197,8 @@ struct rtw_chip_info {
 	u16 fw_fifo_addr[RTW_FW_FIFO_MAX];
 	const struct rtw_fwcd_segs *fwcd_segs;
 
+	u8 usb_tx_agg_desc_num;
+
 	u8 default_1ss_tx_path;
 
 	bool path_div_supported;
diff --git a/drivers/net/wireless/realtek/rtw88/reg.h b/drivers/net/wireless/realtek/rtw88/reg.h
index b122f226924b..02ef9a77316b 100644
--- a/drivers/net/wireless/realtek/rtw88/reg.h
+++ b/drivers/net/wireless/realtek/rtw88/reg.h
@@ -270,6 +270,7 @@
 #define BIT_MASK_BCN_HEAD_1_V1	0xfff
 #define REG_AUTO_LLT_V1		0x0208
 #define BIT_AUTO_INIT_LLT_V1	BIT(0)
+#define BIT_MASK_BLK_DESC_NUM	GENMASK(7, 4)
 #define REG_DWBCN0_CTRL		0x0208
 #define BIT_BCN_VALID		BIT(16)
 #define REG_TXDMA_OFFSET_CHK	0x020C
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8703b.c b/drivers/net/wireless/realtek/rtw88/rtw8703b.c
index 8919f9e11f03..222608de33cd 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8703b.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8703b.c
@@ -2013,6 +2013,7 @@ const struct rtw_chip_info rtw8703b_hw_spec = {
 	.tx_stbc = false,
 	.max_power_index = 0x3f,
 	.ampdu_density = IEEE80211_HT_MPDU_DENSITY_16,
+	.usb_tx_agg_desc_num = 1, /* Not sure if this chip has USB interface */
 
 	.path_div_supported = false,
 	.ht_supported = true,
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8723d.c b/drivers/net/wireless/realtek/rtw88/rtw8723d.c
index f8df4c84d39f..3fba4054d45f 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8723d.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8723d.c
@@ -2171,6 +2171,7 @@ const struct rtw_chip_info rtw8723d_hw_spec = {
 	.band = RTW_BAND_2G,
 	.page_size = TX_PAGE_SIZE,
 	.dig_min = 0x20,
+	.usb_tx_agg_desc_num = 1,
 	.ht_supported = true,
 	.vht_supported = false,
 	.lps_deep_mode_supported = 0,
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8821c.c b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
index fe5d8e188350..526e8de77b3e 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8821c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8821c.c
@@ -2008,6 +2008,7 @@ const struct rtw_chip_info rtw8821c_hw_spec = {
 	.band = RTW_BAND_2G | RTW_BAND_5G,
 	.page_size = TX_PAGE_SIZE,
 	.dig_min = 0x1c,
+	.usb_tx_agg_desc_num = 3,
 	.ht_supported = true,
 	.vht_supported = true,
 	.lps_deep_mode_supported = BIT(LPS_DEEP_MODE_LCLK),
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822b.c b/drivers/net/wireless/realtek/rtw88/rtw8822b.c
index 3017a9760da8..2456ff242818 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822b.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822b.c
@@ -2548,6 +2548,7 @@ const struct rtw_chip_info rtw8822b_hw_spec = {
 	.band = RTW_BAND_2G | RTW_BAND_5G,
 	.page_size = TX_PAGE_SIZE,
 	.dig_min = 0x1c,
+	.usb_tx_agg_desc_num = 3,
 	.ht_supported = true,
 	.vht_supported = true,
 	.lps_deep_mode_supported = BIT(LPS_DEEP_MODE_LCLK),
diff --git a/drivers/net/wireless/realtek/rtw88/rtw8822c.c b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
index cd965edc29ce..62376d1cca22 100644
--- a/drivers/net/wireless/realtek/rtw88/rtw8822c.c
+++ b/drivers/net/wireless/realtek/rtw88/rtw8822c.c
@@ -5366,6 +5366,7 @@ const struct rtw_chip_info rtw8822c_hw_spec = {
 	.band = RTW_BAND_2G | RTW_BAND_5G,
 	.page_size = TX_PAGE_SIZE,
 	.dig_min = 0x20,
+	.usb_tx_agg_desc_num = 3,
 	.default_1ss_tx_path = BB_PATH_A,
 	.path_div_supported = true,
 	.ht_supported = true,
diff --git a/drivers/net/wireless/realtek/rtw88/usb.c b/drivers/net/wireless/realtek/rtw88/usb.c
index d204d138afe2..057c0ffbe944 100644
--- a/drivers/net/wireless/realtek/rtw88/usb.c
+++ b/drivers/net/wireless/realtek/rtw88/usb.c
@@ -379,7 +379,9 @@ static bool rtw_usb_tx_agg_skb(struct rtw_usb *rtwusb, struct sk_buff_head *list
 
 		skb_iter = skb_peek(list);
 
-		if (skb_iter && skb_iter->len + skb_head->len <= RTW_USB_MAX_XMITBUF_SZ)
+		if (skb_iter &&
+		    skb_iter->len + skb_head->len <= RTW_USB_MAX_XMITBUF_SZ &&
+		    agg_num < rtwdev->chip->usb_tx_agg_desc_num)
 			__skb_unlink(skb_iter, list);
 		else
 			skb_iter = NULL;
-- 
cgit v1.2.3-58-ga151


From 6c3b5970b0c408e39d776113cf4624c1f551d68b Mon Sep 17 00:00:00 2001
From: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Date: Wed, 19 Jun 2024 10:40:17 +0800
Subject: wifi: rtl8xxxu: use swap() in rtl8xxxu_switch_ports()

Use existing swap() function rather than duplicating its implementation.

./drivers/net/wireless/realtek/rtl8xxxu/core.c:6749:30-31: WARNING opportunity for swap().

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=9358
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240619024017.53246-1-jiapeng.chong@linux.alibaba.com
---
 drivers/net/wireless/realtek/rtl8xxxu/core.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/core.c b/drivers/net/wireless/realtek/rtl8xxxu/core.c
index 89a841b4e8d5..e19fd31edfda 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/core.c
@@ -6679,7 +6679,6 @@ static void rtl8xxxu_switch_ports(struct rtl8xxxu_priv *priv)
 	u8 macid[ETH_ALEN], bssid[ETH_ALEN], macid_1[ETH_ALEN], bssid_1[ETH_ALEN];
 	u8 msr, bcn_ctrl, bcn_ctrl_1, atimwnd[2], atimwnd_1[2];
 	struct rtl8xxxu_vif *rtlvif;
-	struct ieee80211_vif *vif;
 	u8 tsftr[8], tsftr_1[8];
 	int i;
 
@@ -6744,10 +6743,7 @@ static void rtl8xxxu_switch_ports(struct rtl8xxxu_priv *priv)
 	/* write bcn ctl */
 	rtl8xxxu_write8(priv, REG_BEACON_CTRL, bcn_ctrl_1);
 	rtl8xxxu_write8(priv, REG_BEACON_CTRL_1, bcn_ctrl);
-
-	vif = priv->vifs[0];
-	priv->vifs[0] = priv->vifs[1];
-	priv->vifs[1] = vif;
+	swap(priv->vifs[0], priv->vifs[1]);
 
 	/* priv->vifs[0] is NULL here, based on how this function is currently
 	 * called from rtl8xxxu_add_interface().
-- 
cgit v1.2.3-58-ga151


From 3226607302ca5a74dee6f1c817580c713ef9d0dd Mon Sep 17 00:00:00 2001
From: Taehee Yoo <ap420073@gmail.com>
Date: Fri, 21 Jun 2024 04:16:37 +0000
Subject: selftests: net: change shebang to bash in amt.sh

amt.sh is written in bash, not sh.
So, shebang should be bash.

Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Acked-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/amt.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/amt.sh b/tools/testing/selftests/net/amt.sh
index 7e7ed6c558da..d458b45c775b 100755
--- a/tools/testing/selftests/net/amt.sh
+++ b/tools/testing/selftests/net/amt.sh
@@ -1,4 +1,4 @@
-#!/bin/sh
+#!/bin/bash
 # SPDX-License-Identifier: GPL-2.0
 
 # Author: Taehee Yoo <ap420073@gmail.com>
-- 
cgit v1.2.3-58-ga151


From 639ac8ce8b65a44f176e3e831e39d5dde738ac71 Mon Sep 17 00:00:00 2001
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Date: Mon, 10 Jun 2024 09:44:31 +0200
Subject: ice: store representor ID in bridge port

It is used to get representor structure during cleaning.

Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_eswitch_br.c | 4 +++-
 drivers/net/ethernet/intel/ice/ice_eswitch_br.h | 1 +
 drivers/net/ethernet/intel/ice/ice_repr.c       | 7 ++-----
 drivers/net/ethernet/intel/ice/ice_repr.h       | 3 +--
 4 files changed, 7 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_eswitch_br.c b/drivers/net/ethernet/intel/ice/ice_eswitch_br.c
index ac5beecd028b..f5aceb32bf4d 100644
--- a/drivers/net/ethernet/intel/ice/ice_eswitch_br.c
+++ b/drivers/net/ethernet/intel/ice/ice_eswitch_br.c
@@ -896,7 +896,8 @@ ice_eswitch_br_port_deinit(struct ice_esw_br *bridge,
 	if (br_port->type == ICE_ESWITCH_BR_UPLINK_PORT && vsi->back) {
 		vsi->back->br_port = NULL;
 	} else {
-		struct ice_repr *repr = ice_repr_get_by_vsi(vsi);
+		struct ice_repr *repr =
+			ice_repr_get(vsi->back, br_port->repr_id);
 
 		if (repr)
 			repr->br_port = NULL;
@@ -937,6 +938,7 @@ ice_eswitch_br_vf_repr_port_init(struct ice_esw_br *bridge,
 	br_port->vsi = repr->src_vsi;
 	br_port->vsi_idx = br_port->vsi->idx;
 	br_port->type = ICE_ESWITCH_BR_VF_REPR_PORT;
+	br_port->repr_id = repr->id;
 	repr->br_port = br_port;
 
 	err = xa_insert(&bridge->ports, br_port->vsi_idx, br_port, GFP_KERNEL);
diff --git a/drivers/net/ethernet/intel/ice/ice_eswitch_br.h b/drivers/net/ethernet/intel/ice/ice_eswitch_br.h
index 85a8fadb2928..c15c7344d7f8 100644
--- a/drivers/net/ethernet/intel/ice/ice_eswitch_br.h
+++ b/drivers/net/ethernet/intel/ice/ice_eswitch_br.h
@@ -46,6 +46,7 @@ struct ice_esw_br_port {
 	enum ice_esw_br_port_type type;
 	u16 vsi_idx;
 	u16 pvid;
+	u32 repr_id;
 	struct xarray vlans;
 };
 
diff --git a/drivers/net/ethernet/intel/ice/ice_repr.c b/drivers/net/ethernet/intel/ice/ice_repr.c
index d367f4c66dcd..fe83f305cc7d 100644
--- a/drivers/net/ethernet/intel/ice/ice_repr.c
+++ b/drivers/net/ethernet/intel/ice/ice_repr.c
@@ -415,12 +415,9 @@ err_repr_add:
 	return ERR_PTR(err);
 }
 
-struct ice_repr *ice_repr_get_by_vsi(struct ice_vsi *vsi)
+struct ice_repr *ice_repr_get(struct ice_pf *pf, u32 id)
 {
-	if (!vsi->vf)
-		return NULL;
-
-	return xa_load(&vsi->back->eswitch.reprs, vsi->vf->repr_id);
+	return xa_load(&pf->eswitch.reprs, id);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_repr.h b/drivers/net/ethernet/intel/ice/ice_repr.h
index cff730b15ca0..488661b2900b 100644
--- a/drivers/net/ethernet/intel/ice/ice_repr.h
+++ b/drivers/net/ethernet/intel/ice/ice_repr.h
@@ -35,9 +35,8 @@ void ice_repr_stop_tx_queues(struct ice_repr *repr);
 struct ice_repr *ice_netdev_to_repr(const struct net_device *netdev);
 bool ice_is_port_repr_netdev(const struct net_device *netdev);
 
-struct ice_repr *ice_repr_get_by_vsi(struct ice_vsi *vsi);
-
 void ice_repr_inc_tx_stats(struct ice_repr *repr, unsigned int len,
 			   int xmit_status);
 void ice_repr_inc_rx_stats(struct net_device *netdev, unsigned int len);
+struct ice_repr *ice_repr_get(struct ice_pf *pf, u32 id);
 #endif
-- 
cgit v1.2.3-58-ga151


From 8d2f518c0c9dc86be677c037a6e2f5e7da268866 Mon Sep 17 00:00:00 2001
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Date: Mon, 10 Jun 2024 09:44:32 +0200
Subject: ice: move devlink locking outside the port creation

In case of subfunction lock will be taken for whole port creation and
removing. Do the same in VF case.

Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/devlink/devlink.c      | 2 --
 drivers/net/ethernet/intel/ice/devlink/devlink_port.c | 4 ++--
 drivers/net/ethernet/intel/ice/ice_eswitch.c          | 9 +++++++--
 drivers/net/ethernet/intel/ice/ice_repr.c             | 2 --
 4 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/devlink/devlink.c b/drivers/net/ethernet/intel/ice/devlink/devlink.c
index 704e9ad5144e..f774781ab514 100644
--- a/drivers/net/ethernet/intel/ice/devlink/devlink.c
+++ b/drivers/net/ethernet/intel/ice/devlink/devlink.c
@@ -794,10 +794,8 @@ int ice_devlink_rate_init_tx_topology(struct devlink *devlink, struct ice_vsi *v
 
 	tc_node = pi->root->children[0];
 	mutex_lock(&pi->sched_lock);
-	devl_lock(devlink);
 	for (i = 0; i < tc_node->num_children; i++)
 		ice_traverse_tx_tree(devlink, tc_node->children[i], tc_node, pf);
-	devl_unlock(devlink);
 	mutex_unlock(&pi->sched_lock);
 
 	return 0;
diff --git a/drivers/net/ethernet/intel/ice/devlink/devlink_port.c b/drivers/net/ethernet/intel/ice/devlink/devlink_port.c
index 13e6790d3cae..c9fbeebf7fb9 100644
--- a/drivers/net/ethernet/intel/ice/devlink/devlink_port.c
+++ b/drivers/net/ethernet/intel/ice/devlink/devlink_port.c
@@ -407,7 +407,7 @@ int ice_devlink_create_vf_port(struct ice_vf *vf)
 	devlink_port_attrs_set(devlink_port, &attrs);
 	devlink = priv_to_devlink(pf);
 
-	err = devlink_port_register(devlink, devlink_port, vsi->idx);
+	err = devl_port_register(devlink, devlink_port, vsi->idx);
 	if (err) {
 		dev_err(dev, "Failed to create devlink port for VF %d, error %d\n",
 			vf->vf_id, err);
@@ -426,5 +426,5 @@ int ice_devlink_create_vf_port(struct ice_vf *vf)
 void ice_devlink_destroy_vf_port(struct ice_vf *vf)
 {
 	devl_rate_leaf_destroy(&vf->devlink_port);
-	devlink_port_unregister(&vf->devlink_port);
+	devl_port_unregister(&vf->devlink_port);
 }
diff --git a/drivers/net/ethernet/intel/ice/ice_eswitch.c b/drivers/net/ethernet/intel/ice/ice_eswitch.c
index b102db8b829a..7b57a6561a5a 100644
--- a/drivers/net/ethernet/intel/ice/ice_eswitch.c
+++ b/drivers/net/ethernet/intel/ice/ice_eswitch.c
@@ -423,6 +423,7 @@ static void ice_eswitch_start_reprs(struct ice_pf *pf)
 int
 ice_eswitch_attach(struct ice_pf *pf, struct ice_vf *vf)
 {
+	struct devlink *devlink = priv_to_devlink(pf);
 	struct ice_repr *repr;
 	int err;
 
@@ -437,7 +438,9 @@ ice_eswitch_attach(struct ice_pf *pf, struct ice_vf *vf)
 
 	ice_eswitch_stop_reprs(pf);
 
+	devl_lock(devlink);
 	repr = ice_repr_add_vf(vf);
+	devl_unlock(devlink);
 	if (IS_ERR(repr)) {
 		err = PTR_ERR(repr);
 		goto err_create_repr;
@@ -460,7 +463,9 @@ ice_eswitch_attach(struct ice_pf *pf, struct ice_vf *vf)
 err_xa_alloc:
 	ice_eswitch_release_repr(pf, repr);
 err_setup_repr:
+	devl_lock(devlink);
 	ice_repr_rem_vf(repr);
+	devl_unlock(devlink);
 err_create_repr:
 	if (xa_empty(&pf->eswitch.reprs))
 		ice_eswitch_disable_switchdev(pf);
@@ -484,6 +489,7 @@ void ice_eswitch_detach(struct ice_pf *pf, struct ice_vf *vf)
 		ice_eswitch_disable_switchdev(pf);
 
 	ice_eswitch_release_repr(pf, repr);
+	devl_lock(devlink);
 	ice_repr_rem_vf(repr);
 
 	if (xa_empty(&pf->eswitch.reprs)) {
@@ -491,12 +497,11 @@ void ice_eswitch_detach(struct ice_pf *pf, struct ice_vf *vf)
 		 * no point in keeping the nodes
 		 */
 		ice_devlink_rate_clear_tx_topology(ice_get_main_vsi(pf));
-		devl_lock(devlink);
 		devl_rate_nodes_destroy(devlink);
-		devl_unlock(devlink);
 	} else {
 		ice_eswitch_start_reprs(pf);
 	}
+	devl_unlock(devlink);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_repr.c b/drivers/net/ethernet/intel/ice/ice_repr.c
index fe83f305cc7d..35a6ac8c0466 100644
--- a/drivers/net/ethernet/intel/ice/ice_repr.c
+++ b/drivers/net/ethernet/intel/ice/ice_repr.c
@@ -285,9 +285,7 @@ ice_repr_reg_netdev(struct net_device *netdev)
 
 static void ice_repr_remove_node(struct devlink_port *devlink_port)
 {
-	devl_lock(devlink_port->devlink);
 	devl_rate_leaf_destroy(devlink_port);
-	devl_unlock(devlink_port->devlink);
 }
 
 /**
-- 
cgit v1.2.3-58-ga151


From 4d364df2b5ed1718580f22df1458cfecf82787a2 Mon Sep 17 00:00:00 2001
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Date: Mon, 10 Jun 2024 09:44:33 +0200
Subject: ice: move VSI configuration outside repr setup

It is needed because subfunction port representor shouldn't configure
the source VSI during representor creation.

Move the code to separate function and call it only in case the VF port
representor is being created.

Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_eswitch.c | 55 ++++++++++++++++++++--------
 drivers/net/ethernet/intel/ice/ice_eswitch.h | 10 +++++
 drivers/net/ethernet/intel/ice/ice_repr.c    |  7 ++++
 3 files changed, 57 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_eswitch.c b/drivers/net/ethernet/intel/ice/ice_eswitch.c
index 7b57a6561a5a..3f73f46111fc 100644
--- a/drivers/net/ethernet/intel/ice/ice_eswitch.c
+++ b/drivers/net/ethernet/intel/ice/ice_eswitch.c
@@ -117,17 +117,10 @@ static int ice_eswitch_setup_repr(struct ice_pf *pf, struct ice_repr *repr)
 	struct ice_vsi *vsi = repr->src_vsi;
 	struct metadata_dst *dst;
 
-	ice_remove_vsi_fltr(&pf->hw, vsi->idx);
 	repr->dst = metadata_dst_alloc(0, METADATA_HW_PORT_MUX,
 				       GFP_KERNEL);
 	if (!repr->dst)
-		goto err_add_mac_fltr;
-
-	if (ice_vsi_update_security(vsi, ice_vsi_ctx_clear_antispoof))
-		goto err_dst_free;
-
-	if (ice_vsi_add_vlan_zero(vsi))
-		goto err_update_security;
+		return -ENOMEM;
 
 	netif_keep_dst(uplink_vsi->netdev);
 
@@ -136,16 +129,48 @@ static int ice_eswitch_setup_repr(struct ice_pf *pf, struct ice_repr *repr)
 	dst->u.port_info.lower_dev = uplink_vsi->netdev;
 
 	return 0;
+}
 
-err_update_security:
+/**
+ * ice_eswitch_cfg_vsi - configure VSI to work in slow-path
+ * @vsi: VSI structure of representee
+ * @mac: representee MAC
+ *
+ * Return: 0 on success, non-zero on error.
+ */
+int ice_eswitch_cfg_vsi(struct ice_vsi *vsi, const u8 *mac)
+{
+	int err;
+
+	ice_remove_vsi_fltr(&vsi->back->hw, vsi->idx);
+
+	err = ice_vsi_update_security(vsi, ice_vsi_ctx_clear_antispoof);
+	if (err)
+		goto err_update_security;
+
+	err = ice_vsi_add_vlan_zero(vsi);
+	if (err)
+		goto err_vlan_zero;
+
+	return 0;
+
+err_vlan_zero:
 	ice_vsi_update_security(vsi, ice_vsi_ctx_set_antispoof);
-err_dst_free:
-	metadata_dst_free(repr->dst);
-	repr->dst = NULL;
-err_add_mac_fltr:
-	ice_fltr_add_mac_and_broadcast(vsi, repr->parent_mac, ICE_FWD_TO_VSI);
+err_update_security:
+	ice_fltr_add_mac_and_broadcast(vsi, mac, ICE_FWD_TO_VSI);
 
-	return -ENODEV;
+	return err;
+}
+
+/**
+ * ice_eswitch_decfg_vsi - unroll changes done to VSI for switchdev
+ * @vsi: VSI structure of representee
+ * @mac: representee MAC
+ */
+void ice_eswitch_decfg_vsi(struct ice_vsi *vsi, const u8 *mac)
+{
+	ice_vsi_update_security(vsi, ice_vsi_ctx_set_antispoof);
+	ice_fltr_add_mac_and_broadcast(vsi, mac, ICE_FWD_TO_VSI);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_eswitch.h b/drivers/net/ethernet/intel/ice/ice_eswitch.h
index e2e5c0c75e7d..9a25606e9740 100644
--- a/drivers/net/ethernet/intel/ice/ice_eswitch.h
+++ b/drivers/net/ethernet/intel/ice/ice_eswitch.h
@@ -28,6 +28,9 @@ netdev_tx_t
 ice_eswitch_port_start_xmit(struct sk_buff *skb, struct net_device *netdev);
 struct net_device *ice_eswitch_get_target(struct ice_rx_ring *rx_ring,
 					  union ice_32b_rx_flex_desc *rx_desc);
+
+int ice_eswitch_cfg_vsi(struct ice_vsi *vsi, const u8 *mac);
+void ice_eswitch_decfg_vsi(struct ice_vsi *vsi, const u8 *mac);
 #else /* CONFIG_ICE_SWITCHDEV */
 static inline void ice_eswitch_detach(struct ice_pf *pf, struct ice_vf *vf) { }
 
@@ -85,5 +88,12 @@ ice_eswitch_get_target(struct ice_rx_ring *rx_ring,
 {
 	return rx_ring->netdev;
 }
+
+static inline int ice_eswitch_cfg_vsi(struct ice_vsi *vsi, const u8 *mac)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline void ice_eswitch_decfg_vsi(struct ice_vsi *vsi, const u8 *mac) { }
 #endif /* CONFIG_ICE_SWITCHDEV */
 #endif /* _ICE_ESWITCH_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_repr.c b/drivers/net/ethernet/intel/ice/ice_repr.c
index 35a6ac8c0466..bdda3401e343 100644
--- a/drivers/net/ethernet/intel/ice/ice_repr.c
+++ b/drivers/net/ethernet/intel/ice/ice_repr.c
@@ -306,6 +306,7 @@ static void ice_repr_rem(struct ice_repr *repr)
 void ice_repr_rem_vf(struct ice_repr *repr)
 {
 	ice_repr_remove_node(&repr->vf->devlink_port);
+	ice_eswitch_decfg_vsi(repr->src_vsi, repr->parent_mac);
 	unregister_netdev(repr->netdev);
 	ice_devlink_destroy_vf_port(repr->vf);
 	ice_virtchnl_set_dflt_ops(repr->vf);
@@ -401,11 +402,17 @@ struct ice_repr *ice_repr_add_vf(struct ice_vf *vf)
 	if (err)
 		goto err_netdev;
 
+	err = ice_eswitch_cfg_vsi(repr->src_vsi, repr->parent_mac);
+	if (err)
+		goto err_cfg_vsi;
+
 	ice_virtchnl_set_repr_ops(vf);
 	ice_repr_set_tx_topology(vf->pf);
 
 	return repr;
 
+err_cfg_vsi:
+	unregister_netdev(repr->netdev);
 err_netdev:
 	ice_repr_rem(repr);
 err_repr_add:
-- 
cgit v1.2.3-58-ga151


From fff5cca345a66a341870512e58ca469042249f25 Mon Sep 17 00:00:00 2001
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Date: Mon, 10 Jun 2024 09:44:34 +0200
Subject: ice: update representor when VSI is ready

In case of reset of VF VSI can be reallocated. To handle this case it
should be properly updated.

Reload representor as vsi->vsi_num can be different than the one stored
when representor was created.

Instead of only changing antispoof do whole VSI configuration for
eswitch.

Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_eswitch.c | 21 ++++++++++++++-------
 drivers/net/ethernet/intel/ice/ice_eswitch.h |  4 ++--
 drivers/net/ethernet/intel/ice/ice_vf_lib.c  |  2 +-
 3 files changed, 17 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_eswitch.c b/drivers/net/ethernet/intel/ice/ice_eswitch.c
index 3f73f46111fc..4f539b1c7781 100644
--- a/drivers/net/ethernet/intel/ice/ice_eswitch.c
+++ b/drivers/net/ethernet/intel/ice/ice_eswitch.c
@@ -178,16 +178,16 @@ void ice_eswitch_decfg_vsi(struct ice_vsi *vsi, const u8 *mac)
  * @repr_id: representor ID
  * @vsi: VSI for which port representor is configured
  */
-void ice_eswitch_update_repr(unsigned long repr_id, struct ice_vsi *vsi)
+void ice_eswitch_update_repr(unsigned long *repr_id, struct ice_vsi *vsi)
 {
 	struct ice_pf *pf = vsi->back;
 	struct ice_repr *repr;
-	int ret;
+	int err;
 
 	if (!ice_is_switchdev_running(pf))
 		return;
 
-	repr = xa_load(&pf->eswitch.reprs, repr_id);
+	repr = xa_load(&pf->eswitch.reprs, *repr_id);
 	if (!repr)
 		return;
 
@@ -197,12 +197,19 @@ void ice_eswitch_update_repr(unsigned long repr_id, struct ice_vsi *vsi)
 	if (repr->br_port)
 		repr->br_port->vsi = vsi;
 
-	ret = ice_vsi_update_security(vsi, ice_vsi_ctx_clear_antispoof);
-	if (ret) {
-		ice_fltr_add_mac_and_broadcast(vsi, repr->parent_mac,
-					       ICE_FWD_TO_VSI);
+	err = ice_eswitch_cfg_vsi(vsi, repr->parent_mac);
+	if (err)
 		dev_err(ice_pf_to_dev(pf), "Failed to update VSI of port representor %d",
 			repr->id);
+
+	/* The VSI number is different, reload the PR with new id */
+	if (repr->id != vsi->vsi_num) {
+		xa_erase(&pf->eswitch.reprs, repr->id);
+		repr->id = vsi->vsi_num;
+		if (xa_insert(&pf->eswitch.reprs, repr->id, repr, GFP_KERNEL))
+			dev_err(ice_pf_to_dev(pf), "Failed to reload port representor %d",
+				repr->id);
+		*repr_id = repr->id;
 	}
 }
 
diff --git a/drivers/net/ethernet/intel/ice/ice_eswitch.h b/drivers/net/ethernet/intel/ice/ice_eswitch.h
index 9a25606e9740..09194d514f9b 100644
--- a/drivers/net/ethernet/intel/ice/ice_eswitch.h
+++ b/drivers/net/ethernet/intel/ice/ice_eswitch.h
@@ -18,7 +18,7 @@ ice_eswitch_mode_set(struct devlink *devlink, u16 mode,
 		     struct netlink_ext_ack *extack);
 bool ice_is_eswitch_mode_switchdev(struct ice_pf *pf);
 
-void ice_eswitch_update_repr(unsigned long repr_id, struct ice_vsi *vsi);
+void ice_eswitch_update_repr(unsigned long *repr_id, struct ice_vsi *vsi);
 
 void ice_eswitch_stop_all_tx_queues(struct ice_pf *pf);
 
@@ -47,7 +47,7 @@ ice_eswitch_set_target_vsi(struct sk_buff *skb,
 			   struct ice_tx_offload_params *off) { }
 
 static inline void
-ice_eswitch_update_repr(unsigned long repr_id, struct ice_vsi *vsi) { }
+ice_eswitch_update_repr(unsigned long *repr_id, struct ice_vsi *vsi) { }
 
 static inline int ice_eswitch_configure(struct ice_pf *pf)
 {
diff --git a/drivers/net/ethernet/intel/ice/ice_vf_lib.c b/drivers/net/ethernet/intel/ice/ice_vf_lib.c
index 48a8d462d76a..5635e9da2212 100644
--- a/drivers/net/ethernet/intel/ice/ice_vf_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_vf_lib.c
@@ -948,7 +948,7 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
 		goto out_unlock;
 	}
 
-	ice_eswitch_update_repr(vf->repr_id, vsi);
+	ice_eswitch_update_repr(&vf->repr_id, vsi);
 
 	/* if the VF has been reset allow it to come up again */
 	ice_mbx_clear_malvf(&vf->mbx_info);
-- 
cgit v1.2.3-58-ga151


From 717d6313bba1b3179f0bf1026aaec6b7e26f484e Mon Sep 17 00:00:00 2001
From: Jiri Olsa <jolsa@kernel.org>
Date: Wed, 19 Jun 2024 10:16:24 +0200
Subject: bpf: Change bpf_session_cookie return value to __u64 *

This reverts [1] and changes return value for bpf_session_cookie
in bpf selftests. Having long * might lead to problems on 32-bit
architectures.

Fixes: 2b8dd87332cd ("bpf: Make bpf_session_cookie() kfunc return long *")
Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240619081624.1620152-1-jolsa@kernel.org
---
 kernel/trace/bpf_trace.c                                        | 2 +-
 tools/testing/selftests/bpf/bpf_kfuncs.h                        | 2 +-
 tools/testing/selftests/bpf/progs/kprobe_multi_session_cookie.c | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 4b3fda456299..cd098846e251 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -3530,7 +3530,7 @@ __bpf_kfunc bool bpf_session_is_return(void)
 	return session_ctx->is_return;
 }
 
-__bpf_kfunc long *bpf_session_cookie(void)
+__bpf_kfunc __u64 *bpf_session_cookie(void)
 {
 	struct bpf_session_run_ctx *session_ctx;
 
diff --git a/tools/testing/selftests/bpf/bpf_kfuncs.h b/tools/testing/selftests/bpf/bpf_kfuncs.h
index be91a6919315..3b6675ab4086 100644
--- a/tools/testing/selftests/bpf/bpf_kfuncs.h
+++ b/tools/testing/selftests/bpf/bpf_kfuncs.h
@@ -77,5 +77,5 @@ extern int bpf_verify_pkcs7_signature(struct bpf_dynptr *data_ptr,
 				      struct bpf_key *trusted_keyring) __ksym;
 
 extern bool bpf_session_is_return(void) __ksym __weak;
-extern long *bpf_session_cookie(void) __ksym __weak;
+extern __u64 *bpf_session_cookie(void) __ksym __weak;
 #endif
diff --git a/tools/testing/selftests/bpf/progs/kprobe_multi_session_cookie.c b/tools/testing/selftests/bpf/progs/kprobe_multi_session_cookie.c
index d49070803e22..0835b5edf685 100644
--- a/tools/testing/selftests/bpf/progs/kprobe_multi_session_cookie.c
+++ b/tools/testing/selftests/bpf/progs/kprobe_multi_session_cookie.c
@@ -25,7 +25,7 @@ int BPF_PROG(trigger)
 
 static int check_cookie(__u64 val, __u64 *result)
 {
-	long *cookie;
+	__u64 *cookie;
 
 	if (bpf_get_current_pid_tgid() >> 32 != pid)
 		return 1;
-- 
cgit v1.2.3-58-ga151


From 651337c7ca82c259bf5c8fe9beda9673531a0031 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Tue, 18 Jun 2024 11:38:32 -0700
Subject: bpftool: Allow compile-time checks of BPF map auto-attach support in
 skeleton

New versions of bpftool now emit additional link placeholders for BPF
maps (struct_ops maps are the only maps right now that support
attachment), and set up BPF skeleton in such a way that libbpf will
auto-attach BPF maps automatically, assumming libbpf is recent enough
(v1.5+). Old libbpf will do nothing with those links and won't attempt
to auto-attach maps. This allows user code to handle both pre-v1.5 and
v1.5+ versions of libbpf at runtime, if necessary.

But if users don't have (or don't want to) control bpftool version that
generates skeleton, then they can't just assume that skeleton will have
link placeholders. To make this detection possible and easy, let's add
the following to generated skeleton header file:

  #define BPF_SKEL_SUPPORTS_MAP_AUTO_ATTACH 1

This can be used during compilation time to guard code that accesses
skel->links.<map> slots.

Note, if auto-attachment is undesirable, libbpf allows to disable this
through bpf_map__set_autoattach(map, false). This is necessary only on
libbpf v1.5+, older libbpf doesn't support map auto-attach anyways.

Libbpf version can be detected at compilation time using
LIBBPF_MAJOR_VERSION and LIBBPF_MINOR_VERSION macros, or at runtime with
libbpf_major_version() and libbpf_minor_version() APIs.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Quentin Monnet <qmo@kernel.org>
Link: https://lore.kernel.org/bpf/20240618183832.2535876-1-andrii@kernel.org
---
 tools/bpf/bpftool/gen.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index 4a4eedfcd479..51eaed76db97 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -1272,6 +1272,8 @@ static int do_skeleton(int argc, char **argv)
 		#include <stdlib.h>					    \n\
 		#include <bpf/libbpf.h>					    \n\
 									    \n\
+		#define BPF_SKEL_SUPPORTS_MAP_AUTO_ATTACH 1		    \n\
+									    \n\
 		struct %1$s {						    \n\
 			struct bpf_object_skeleton *skeleton;		    \n\
 			struct bpf_object *obj;				    \n\
-- 
cgit v1.2.3-58-ga151


From 6ddf3a9abd9fdfdd63d8c906fc1393f7950c23f4 Mon Sep 17 00:00:00 2001
From: Matt Bobrowski <mattbobrowski@google.com>
Date: Tue, 18 Jun 2024 19:29:22 +0000
Subject: bpf: Add security_file_post_open() LSM hook to sleepable_lsm_hooks

The new generic LSM hook security_file_post_open() was recently added
to the LSM framework in commit 8f46ff5767b0b ("security: Introduce
file_post_open hook"). Let's proactively add this generic LSM hook to
the sleepable_lsm_hooks BTF ID set, because I can't see there being
any strong reasons not to, and it's only a matter of time before
someone else comes around and asks for it to be there.

security_file_post_open() is inherently sleepable as it's purposely
situated in the kernel that allows LSMs to directly read out the
contents of the backing file if need be. Additionally, it's called
directly after security_file_open(), and that LSM hook in itself
already exists in the sleepable_lsm_hooks BTF ID set.

Signed-off-by: Matt Bobrowski <mattbobrowski@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240618192923.379852-1-mattbobrowski@google.com
---
 kernel/bpf/bpf_lsm.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/kernel/bpf/bpf_lsm.c b/kernel/bpf/bpf_lsm.c
index 68240c3c6e7d..08a338e1f231 100644
--- a/kernel/bpf/bpf_lsm.c
+++ b/kernel/bpf/bpf_lsm.c
@@ -280,6 +280,7 @@ BTF_ID(func, bpf_lsm_cred_prepare)
 BTF_ID(func, bpf_lsm_file_ioctl)
 BTF_ID(func, bpf_lsm_file_lock)
 BTF_ID(func, bpf_lsm_file_open)
+BTF_ID(func, bpf_lsm_file_post_open)
 BTF_ID(func, bpf_lsm_file_receive)
 
 BTF_ID(func, bpf_lsm_inode_create)
-- 
cgit v1.2.3-58-ga151


From cc5083d1f3881624ad2de1f3cbb3a07e152cb254 Mon Sep 17 00:00:00 2001
From: Donglin Peng <dolinux.peng@gmail.com>
Date: Wed, 19 Jun 2024 05:23:55 -0700
Subject: libbpf: Checking the btf_type kind when fixing variable offsets

I encountered an issue when building the test_progs from the repository [1]:

  $ pwd
  /work/Qemu/x86_64/linux-6.10-rc2/tools/testing/selftests/bpf/

  $ make test_progs V=1
  [...]
  ./tools/sbin/bpftool gen object ./ip_check_defrag.bpf.linked2.o ./ip_check_defrag.bpf.linked1.o
  libbpf: failed to find symbol for variable 'bpf_dynptr_slice' in section '.ksyms'
  Error: failed to link './ip_check_defrag.bpf.linked1.o': No such file or directory (2)
  [...]

Upon investigation, I discovered that the btf_types referenced in the '.ksyms'
section had a kind of BTF_KIND_FUNC instead of BTF_KIND_VAR:

  $ bpftool btf dump file ./ip_check_defrag.bpf.linked1.o
  [...]
  [2] DATASEC '.ksyms' size=0 vlen=2
        type_id=16 offset=0 size=0 (FUNC 'bpf_dynptr_from_skb')
        type_id=17 offset=0 size=0 (FUNC 'bpf_dynptr_slice')
  [...]
  [16] FUNC 'bpf_dynptr_from_skb' type_id=82 linkage=extern
  [17] FUNC 'bpf_dynptr_slice' type_id=85 linkage=extern
  [...]

For a detailed analysis, please refer to [2]. We can add a kind checking to
fix the issue.

  [1] https://github.com/eddyz87/bpf/tree/binsort-btf-dedup
  [2] https://lore.kernel.org/all/0c0ef20c-c05e-4db9-bad7-2cbc0d6dfae7@oracle.com/

Fixes: 8fd27bf69b86 ("libbpf: Add BPF static linker BTF and BTF.ext support")
Signed-off-by: Donglin Peng <dolinux.peng@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240619122355.426405-1-dolinux.peng@gmail.com
---
 tools/lib/bpf/linker.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/tools/lib/bpf/linker.c b/tools/lib/bpf/linker.c
index fa11a671da3e..9cd3d4109788 100644
--- a/tools/lib/bpf/linker.c
+++ b/tools/lib/bpf/linker.c
@@ -2227,10 +2227,17 @@ static int linker_fixup_btf(struct src_obj *obj)
 		vi = btf_var_secinfos(t);
 		for (j = 0, m = btf_vlen(t); j < m; j++, vi++) {
 			const struct btf_type *vt = btf__type_by_id(obj->btf, vi->type);
-			const char *var_name = btf__str_by_offset(obj->btf, vt->name_off);
-			int var_linkage = btf_var(vt)->linkage;
+			const char *var_name;
+			int var_linkage;
 			Elf64_Sym *sym;
 
+			/* could be a variable or function */
+			if (!btf_is_var(vt))
+				continue;
+
+			var_name = btf__str_by_offset(obj->btf, vt->name_off);
+			var_linkage = btf_var(vt)->linkage;
+
 			/* no need to patch up static or extern vars */
 			if (var_linkage != BTF_VAR_GLOBAL_ALLOCATED)
 				continue;
-- 
cgit v1.2.3-58-ga151


From f06ae6194f278444201e0b041a00192d794f83b6 Mon Sep 17 00:00:00 2001
From: Cupertino Miranda <cupertino.miranda@oracle.com>
Date: Mon, 17 Jun 2024 15:14:57 +0100
Subject: selftests/bpf: Support checks against a regular expression

Add support for __regex and __regex_unpriv macros to check the test
execution output against a regular expression. This is similar to __msg
and __msg_unpriv, however those expect do substring matching.

Signed-off-by: Cupertino Miranda <cupertino.miranda@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240617141458.471620-2-cupertino.miranda@oracle.com
---
 tools/testing/selftests/bpf/progs/bpf_misc.h |  11 ++-
 tools/testing/selftests/bpf/test_loader.c    | 115 ++++++++++++++++++++-------
 2 files changed, 96 insertions(+), 30 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index fb2f5513e29e..c0280bd2f340 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -7,9 +7,9 @@
  *
  * The test_loader sequentially loads each program in a skeleton.
  * Programs could be loaded in privileged and unprivileged modes.
- * - __success, __failure, __msg imply privileged mode;
- * - __success_unpriv, __failure_unpriv, __msg_unpriv imply
- *   unprivileged mode.
+ * - __success, __failure, __msg, __regex imply privileged mode;
+ * - __success_unpriv, __failure_unpriv, __msg_unpriv, __regex_unpriv
+ *   imply unprivileged mode.
  * If combination of privileged and unprivileged attributes is present
  * both modes are used. If none are present privileged mode is implied.
  *
@@ -24,6 +24,9 @@
  *                   Multiple __msg attributes could be specified.
  * __msg_unpriv      Same as __msg but for unprivileged mode.
  *
+ * __regex           Same as __msg, but using a regular expression.
+ * __regex_unpriv    Same as __msg_unpriv but using a regular expression.
+ *
  * __success         Expect program load success in privileged mode.
  * __success_unpriv  Expect program load success in unprivileged mode.
  *
@@ -59,10 +62,12 @@
  * __auxiliary_unpriv  Same, but load program in unprivileged mode.
  */
 #define __msg(msg)		__attribute__((btf_decl_tag("comment:test_expect_msg=" msg)))
+#define __regex(regex)		__attribute__((btf_decl_tag("comment:test_expect_regex=" regex)))
 #define __failure		__attribute__((btf_decl_tag("comment:test_expect_failure")))
 #define __success		__attribute__((btf_decl_tag("comment:test_expect_success")))
 #define __description(desc)	__attribute__((btf_decl_tag("comment:test_description=" desc)))
 #define __msg_unpriv(msg)	__attribute__((btf_decl_tag("comment:test_expect_msg_unpriv=" msg)))
+#define __regex_unpriv(regex)	__attribute__((btf_decl_tag("comment:test_expect_regex_unpriv=" regex)))
 #define __failure_unpriv	__attribute__((btf_decl_tag("comment:test_expect_failure_unpriv")))
 #define __success_unpriv	__attribute__((btf_decl_tag("comment:test_expect_success_unpriv")))
 #define __log_level(lvl)	__attribute__((btf_decl_tag("comment:test_log_level="#lvl)))
diff --git a/tools/testing/selftests/bpf/test_loader.c b/tools/testing/selftests/bpf/test_loader.c
index 524c38e9cde4..f14e10b0de96 100644
--- a/tools/testing/selftests/bpf/test_loader.c
+++ b/tools/testing/selftests/bpf/test_loader.c
@@ -2,6 +2,7 @@
 /* Copyright (c) 2022 Meta Platforms, Inc. and affiliates. */
 #include <linux/capability.h>
 #include <stdlib.h>
+#include <regex.h>
 #include <test_progs.h>
 #include <bpf/btf.h>
 
@@ -17,9 +18,11 @@
 #define TEST_TAG_EXPECT_FAILURE "comment:test_expect_failure"
 #define TEST_TAG_EXPECT_SUCCESS "comment:test_expect_success"
 #define TEST_TAG_EXPECT_MSG_PFX "comment:test_expect_msg="
+#define TEST_TAG_EXPECT_REGEX_PFX "comment:test_expect_regex="
 #define TEST_TAG_EXPECT_FAILURE_UNPRIV "comment:test_expect_failure_unpriv"
 #define TEST_TAG_EXPECT_SUCCESS_UNPRIV "comment:test_expect_success_unpriv"
 #define TEST_TAG_EXPECT_MSG_PFX_UNPRIV "comment:test_expect_msg_unpriv="
+#define TEST_TAG_EXPECT_REGEX_PFX_UNPRIV "comment:test_expect_regex_unpriv="
 #define TEST_TAG_LOG_LEVEL_PFX "comment:test_log_level="
 #define TEST_TAG_PROG_FLAGS_PFX "comment:test_prog_flags="
 #define TEST_TAG_DESCRIPTION_PFX "comment:test_description="
@@ -46,10 +49,16 @@ enum mode {
 	UNPRIV = 2
 };
 
+struct expect_msg {
+	const char *substr; /* substring match */
+	const char *regex_str; /* regex-based match */
+	regex_t regex;
+};
+
 struct test_subspec {
 	char *name;
 	bool expect_failure;
-	const char **expect_msgs;
+	struct expect_msg *expect_msgs;
 	size_t expect_msg_cnt;
 	int retval;
 	bool execute;
@@ -89,6 +98,16 @@ void test_loader_fini(struct test_loader *tester)
 
 static void free_test_spec(struct test_spec *spec)
 {
+	int i;
+
+	/* Deallocate expect_msgs arrays. */
+	for (i = 0; i < spec->priv.expect_msg_cnt; i++)
+		if (spec->priv.expect_msgs[i].regex_str)
+			regfree(&spec->priv.expect_msgs[i].regex);
+	for (i = 0; i < spec->unpriv.expect_msg_cnt; i++)
+		if (spec->unpriv.expect_msgs[i].regex_str)
+			regfree(&spec->unpriv.expect_msgs[i].regex);
+
 	free(spec->priv.name);
 	free(spec->unpriv.name);
 	free(spec->priv.expect_msgs);
@@ -100,18 +119,38 @@ static void free_test_spec(struct test_spec *spec)
 	spec->unpriv.expect_msgs = NULL;
 }
 
-static int push_msg(const char *msg, struct test_subspec *subspec)
+static int push_msg(const char *substr, const char *regex_str, struct test_subspec *subspec)
 {
 	void *tmp;
+	int regcomp_res;
+	char error_msg[100];
+	struct expect_msg *msg;
 
-	tmp = realloc(subspec->expect_msgs, (1 + subspec->expect_msg_cnt) * sizeof(void *));
+	tmp = realloc(subspec->expect_msgs,
+		      (1 + subspec->expect_msg_cnt) * sizeof(struct expect_msg));
 	if (!tmp) {
 		ASSERT_FAIL("failed to realloc memory for messages\n");
 		return -ENOMEM;
 	}
 	subspec->expect_msgs = tmp;
-	subspec->expect_msgs[subspec->expect_msg_cnt++] = msg;
+	msg = &subspec->expect_msgs[subspec->expect_msg_cnt];
+
+	if (substr) {
+		msg->substr = substr;
+		msg->regex_str = NULL;
+	} else {
+		msg->regex_str = regex_str;
+		msg->substr = NULL;
+		regcomp_res = regcomp(&msg->regex, regex_str, REG_EXTENDED|REG_NEWLINE);
+		if (regcomp_res != 0) {
+			regerror(regcomp_res, &msg->regex, error_msg, sizeof(error_msg));
+			PRINT_FAIL("Regexp compilation error in '%s': '%s'\n",
+				   regex_str, error_msg);
+			return -EINVAL;
+		}
+	}
 
+	subspec->expect_msg_cnt += 1;
 	return 0;
 }
 
@@ -233,13 +272,25 @@ static int parse_test_spec(struct test_loader *tester,
 			spec->mode_mask |= UNPRIV;
 		} else if (str_has_pfx(s, TEST_TAG_EXPECT_MSG_PFX)) {
 			msg = s + sizeof(TEST_TAG_EXPECT_MSG_PFX) - 1;
-			err = push_msg(msg, &spec->priv);
+			err = push_msg(msg, NULL, &spec->priv);
 			if (err)
 				goto cleanup;
 			spec->mode_mask |= PRIV;
 		} else if (str_has_pfx(s, TEST_TAG_EXPECT_MSG_PFX_UNPRIV)) {
 			msg = s + sizeof(TEST_TAG_EXPECT_MSG_PFX_UNPRIV) - 1;
-			err = push_msg(msg, &spec->unpriv);
+			err = push_msg(msg, NULL, &spec->unpriv);
+			if (err)
+				goto cleanup;
+			spec->mode_mask |= UNPRIV;
+		} else if (str_has_pfx(s, TEST_TAG_EXPECT_REGEX_PFX)) {
+			msg = s + sizeof(TEST_TAG_EXPECT_REGEX_PFX) - 1;
+			err = push_msg(NULL, msg, &spec->priv);
+			if (err)
+				goto cleanup;
+			spec->mode_mask |= PRIV;
+		} else if (str_has_pfx(s, TEST_TAG_EXPECT_REGEX_PFX_UNPRIV)) {
+			msg = s + sizeof(TEST_TAG_EXPECT_REGEX_PFX_UNPRIV) - 1;
+			err = push_msg(NULL, msg, &spec->unpriv);
 			if (err)
 				goto cleanup;
 			spec->mode_mask |= UNPRIV;
@@ -337,16 +388,13 @@ static int parse_test_spec(struct test_loader *tester,
 		}
 
 		if (!spec->unpriv.expect_msgs) {
-			size_t sz = spec->priv.expect_msg_cnt * sizeof(void *);
+			for (i = 0; i < spec->priv.expect_msg_cnt; i++) {
+				struct expect_msg *msg = &spec->priv.expect_msgs[i];
 
-			spec->unpriv.expect_msgs = malloc(sz);
-			if (!spec->unpriv.expect_msgs) {
-				PRINT_FAIL("failed to allocate memory for unpriv.expect_msgs\n");
-				err = -ENOMEM;
-				goto cleanup;
+				err = push_msg(msg->substr, msg->regex_str, &spec->unpriv);
+				if (err)
+					goto cleanup;
 			}
-			memcpy(spec->unpriv.expect_msgs, spec->priv.expect_msgs, sz);
-			spec->unpriv.expect_msg_cnt = spec->priv.expect_msg_cnt;
 		}
 	}
 
@@ -402,27 +450,40 @@ static void validate_case(struct test_loader *tester,
 			  struct bpf_program *prog,
 			  int load_err)
 {
-	int i, j;
+	int i, j, err;
+	char *match;
+	regmatch_t reg_match[1];
 
 	for (i = 0; i < subspec->expect_msg_cnt; i++) {
-		char *match;
-		const char *expect_msg;
-
-		expect_msg = subspec->expect_msgs[i];
+		struct expect_msg *msg = &subspec->expect_msgs[i];
+
+		if (msg->substr) {
+			match = strstr(tester->log_buf + tester->next_match_pos, msg->substr);
+			if (match)
+				tester->next_match_pos = match - tester->log_buf + strlen(msg->substr);
+		} else {
+			err = regexec(&msg->regex,
+				      tester->log_buf + tester->next_match_pos, 1, reg_match, 0);
+			if (err == 0) {
+				match = tester->log_buf + tester->next_match_pos + reg_match[0].rm_so;
+				tester->next_match_pos += reg_match[0].rm_eo;
+			} else {
+				match = NULL;
+			}
+		}
 
-		match = strstr(tester->log_buf + tester->next_match_pos, expect_msg);
 		if (!ASSERT_OK_PTR(match, "expect_msg")) {
-			/* if we are in verbose mode, we've already emitted log */
 			if (env.verbosity == VERBOSE_NONE)
 				emit_verifier_log(tester->log_buf, true /*force*/);
-			for (j = 0; j < i; j++)
-				fprintf(stderr,
-					"MATCHED  MSG: '%s'\n", subspec->expect_msgs[j]);
-			fprintf(stderr, "EXPECTED MSG: '%s'\n", expect_msg);
+			for (j = 0; j <= i; j++) {
+				msg = &subspec->expect_msgs[j];
+				fprintf(stderr, "%s %s: '%s'\n",
+					j < i ? "MATCHED " : "EXPECTED",
+					msg->substr ? "SUBSTR" : " REGEX",
+					msg->substr ?: msg->regex_str);
+			}
 			return;
 		}
-
-		tester->next_match_pos = match - tester->log_buf + strlen(expect_msg);
 	}
 }
 
-- 
cgit v1.2.3-58-ga151


From 3e23c99764d465ae411f0729fd6d2e0e3edd0ade Mon Sep 17 00:00:00 2001
From: Cupertino Miranda <cupertino.miranda@oracle.com>
Date: Mon, 17 Jun 2024 15:14:58 +0100
Subject: selftests/bpf: Match tests against regular expression

This patch changes a few tests to make use of regular expressions.
Fixed tests otherwise fail when compiled with GCC.

Signed-off-by: Cupertino Miranda <cupertino.miranda@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240617141458.471620-3-cupertino.miranda@oracle.com
---
 tools/testing/selftests/bpf/progs/dynptr_fail.c          | 6 +++---
 tools/testing/selftests/bpf/progs/rbtree_fail.c          | 2 +-
 tools/testing/selftests/bpf/progs/refcounted_kptr_fail.c | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/dynptr_fail.c b/tools/testing/selftests/bpf/progs/dynptr_fail.c
index 66a60bfb5867..64cc9d936a13 100644
--- a/tools/testing/selftests/bpf/progs/dynptr_fail.c
+++ b/tools/testing/selftests/bpf/progs/dynptr_fail.c
@@ -964,7 +964,7 @@ int dynptr_invalidate_slice_reinit(void *ctx)
  * mem_or_null pointers.
  */
 SEC("?raw_tp")
-__failure __msg("R1 type=scalar expected=percpu_ptr_")
+__failure __regex("R[0-9]+ type=scalar expected=percpu_ptr_")
 int dynptr_invalidate_slice_or_null(void *ctx)
 {
 	struct bpf_dynptr ptr;
@@ -982,7 +982,7 @@ int dynptr_invalidate_slice_or_null(void *ctx)
 
 /* Destruction of dynptr should also any slices obtained from it */
 SEC("?raw_tp")
-__failure __msg("R7 invalid mem access 'scalar'")
+__failure __regex("R[0-9]+ invalid mem access 'scalar'")
 int dynptr_invalidate_slice_failure(void *ctx)
 {
 	struct bpf_dynptr ptr1;
@@ -1069,7 +1069,7 @@ int dynptr_read_into_slot(void *ctx)
 
 /* bpf_dynptr_slice()s are read-only and cannot be written to */
 SEC("?tc")
-__failure __msg("R0 cannot write into rdonly_mem")
+__failure __regex("R[0-9]+ cannot write into rdonly_mem")
 int skb_invalid_slice_write(struct __sk_buff *skb)
 {
 	struct bpf_dynptr ptr;
diff --git a/tools/testing/selftests/bpf/progs/rbtree_fail.c b/tools/testing/selftests/bpf/progs/rbtree_fail.c
index 3fecf1c6dfe5..b722a1e1ddef 100644
--- a/tools/testing/selftests/bpf/progs/rbtree_fail.c
+++ b/tools/testing/selftests/bpf/progs/rbtree_fail.c
@@ -105,7 +105,7 @@ long rbtree_api_remove_unadded_node(void *ctx)
 }
 
 SEC("?tc")
-__failure __msg("Unreleased reference id=3 alloc_insn=10")
+__failure __regex("Unreleased reference id=3 alloc_insn=[0-9]+")
 long rbtree_api_remove_no_drop(void *ctx)
 {
 	struct bpf_rb_node *res;
diff --git a/tools/testing/selftests/bpf/progs/refcounted_kptr_fail.c b/tools/testing/selftests/bpf/progs/refcounted_kptr_fail.c
index 1553b9c16aa7..f8d4b7cfcd68 100644
--- a/tools/testing/selftests/bpf/progs/refcounted_kptr_fail.c
+++ b/tools/testing/selftests/bpf/progs/refcounted_kptr_fail.c
@@ -32,7 +32,7 @@ static bool less(struct bpf_rb_node *a, const struct bpf_rb_node *b)
 }
 
 SEC("?tc")
-__failure __msg("Unreleased reference id=4 alloc_insn=21")
+__failure __regex("Unreleased reference id=4 alloc_insn=[0-9]+")
 long rbtree_refcounted_node_ref_escapes(void *ctx)
 {
 	struct node_acquire *n, *m;
@@ -73,7 +73,7 @@ long refcount_acquire_maybe_null(void *ctx)
 }
 
 SEC("?tc")
-__failure __msg("Unreleased reference id=3 alloc_insn=9")
+__failure __regex("Unreleased reference id=3 alloc_insn=[0-9]+")
 long rbtree_refcounted_node_ref_escapes_owning_input(void *ctx)
 {
 	struct node_acquire *n, *m;
-- 
cgit v1.2.3-58-ga151


From 2bb138cb20a6a347cfed84381430cd25e05f118e Mon Sep 17 00:00:00 2001
From: Puranjay Mohan <puranjay@kernel.org>
Date: Wed, 19 Jun 2024 13:13:34 +0000
Subject: bpf, arm64: Inline bpf_get_current_task/_btf() helpers

On ARM64, the pointer to task_struct is always available in the sp_el0
register and therefore the calls to bpf_get_current_task() and
bpf_get_current_task_btf() can be inlined into a single MRS instruction.

Here is the difference before and after this change:

Before:

; struct task_struct *task = bpf_get_current_task_btf();
  54:   mov     x10, #0xffffffffffff7978        // #-34440
  58:   movk    x10, #0x802b, lsl #16
  5c:   movk    x10, #0x8000, lsl #32
  60:   blr     x10          -------------->    0xffff8000802b7978 <+0>:     mrs     x0, sp_el0
  64:   add     x7, x0, #0x0 <--------------    0xffff8000802b797c <+4>:     ret

After:

; struct task_struct *task = bpf_get_current_task_btf();
  54:   mrs     x7, sp_el0

This shows around 1% performance improvement in artificial microbenchmark.

Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Xu Kuohai <xukuohai@huawei.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240619131334.4297-1-puranjay@kernel.org
---
 arch/arm64/net/bpf_jit_comp.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c
index 6edaeafd1499..751331f5ba90 100644
--- a/arch/arm64/net/bpf_jit_comp.c
+++ b/arch/arm64/net/bpf_jit_comp.c
@@ -1244,6 +1244,13 @@ emit_cond_jmp:
 			break;
 		}
 
+		/* Implement helper call to bpf_get_current_task/_btf() inline */
+		if (insn->src_reg == 0 && (insn->imm == BPF_FUNC_get_current_task ||
+					   insn->imm == BPF_FUNC_get_current_task_btf)) {
+			emit(A64_MRS_SP_EL0(r0), ctx);
+			break;
+		}
+
 		ret = bpf_jit_get_func_addr(ctx->prog, insn, extra_pass,
 					    &func_addr, &func_addr_fixed);
 		if (ret < 0)
@@ -2580,6 +2587,8 @@ bool bpf_jit_inlines_helper_call(s32 imm)
 {
 	switch (imm) {
 	case BPF_FUNC_get_smp_processor_id:
+	case BPF_FUNC_get_current_task:
+	case BPF_FUNC_get_current_task_btf:
 		return true;
 	default:
 		return false;
-- 
cgit v1.2.3-58-ga151


From cd387ce54834bc7808082c471fd745ce85a0e21f Mon Sep 17 00:00:00 2001
From: Mykyta Yatsenko <yatsenko@meta.com>
Date: Fri, 21 Jun 2024 19:03:24 +0100
Subject: selftests/bpf: Test struct_ops bpf map auto-attach

Adding selftest to verify that struct_ops maps are auto attached by
bpf skeleton's `*__attach` function.

Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240621180324.238379-1-yatsenko@meta.com
---
 .../testing/selftests/bpf/prog_tests/bpf_tcp_ca.c  | 35 ++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index 164f237b24dd..bceff5900016 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -185,6 +185,39 @@ done:
 		close(fd);
 }
 
+static void test_dctcp_autoattach_map(void)
+{
+	struct cb_opts cb_opts = {
+		.cc = "bpf_dctcp",
+	};
+	struct network_helper_opts opts = {
+		.post_socket_cb	= cc_cb,
+		.cb_opts	= &cb_opts,
+	};
+	struct bpf_dctcp *dctcp_skel;
+	struct bpf_link *link;
+
+	dctcp_skel = bpf_dctcp__open_and_load();
+	if (!ASSERT_OK_PTR(dctcp_skel, "bpf_dctcp__open_and_load"))
+		return;
+
+	bpf_map__set_autoattach(dctcp_skel->maps.dctcp, true);
+	bpf_map__set_autoattach(dctcp_skel->maps.dctcp_nouse, false);
+
+	if (!ASSERT_OK(bpf_dctcp__attach(dctcp_skel), "bpf_dctcp__attach"))
+		goto destroy;
+
+	/* struct_ops is auto-attached  */
+	link = dctcp_skel->links.dctcp;
+	if (!ASSERT_OK_PTR(link, "link"))
+		goto destroy;
+
+	do_test(&opts);
+
+destroy:
+	bpf_dctcp__destroy(dctcp_skel);
+}
+
 static char *err_str;
 static bool found;
 
@@ -598,4 +631,6 @@ void test_bpf_tcp_ca(void)
 		test_tcp_ca_kfunc();
 	if (test__start_subtest("cc_cubic"))
 		test_cc_cubic();
+	if (test__start_subtest("dctcp_autoattach_map"))
+		test_dctcp_autoattach_map();
 }
-- 
cgit v1.2.3-58-ga151


From d1cf840854bb603c0718a011bc993f69f2df014e Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 20 Jun 2024 10:17:28 +0100
Subject: libbpf: BTF relocation followup fixing naming, loop logic

Use less verbose names in BTF relocation code and fix off-by-one error
and typo in btf_relocate.c.  Simplify loop over matching distilled
types, moving from assigning a _next value in loop body to moving
match check conditions into the guard.

Suggested-by: Andrii Nakryiko <andrii.nakryiko@gmail.com>
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240620091733.1967885-2-alan.maguire@oracle.com
---
 tools/lib/bpf/btf_relocate.c | 72 +++++++++++++++++++-------------------------
 1 file changed, 31 insertions(+), 41 deletions(-)

diff --git a/tools/lib/bpf/btf_relocate.c b/tools/lib/bpf/btf_relocate.c
index eabb8755f662..23a41fb03e0d 100644
--- a/tools/lib/bpf/btf_relocate.c
+++ b/tools/lib/bpf/btf_relocate.c
@@ -160,7 +160,7 @@ static int btf_mark_embedded_composite_type_ids(struct btf_relocate *r, __u32 i)
  */
 static int btf_relocate_map_distilled_base(struct btf_relocate *r)
 {
-	struct btf_name_info *dist_base_info_sorted, *dist_base_info_sorted_end;
+	struct btf_name_info *info, *info_end;
 	struct btf_type *base_t, *dist_t;
 	__u8 *base_name_cnt = NULL;
 	int err = 0;
@@ -169,26 +169,24 @@ static int btf_relocate_map_distilled_base(struct btf_relocate *r)
 	/* generate a sort index array of name/type ids sorted by name for
 	 * distilled base BTF to speed name-based lookups.
 	 */
-	dist_base_info_sorted = calloc(r->nr_dist_base_types, sizeof(*dist_base_info_sorted));
-	if (!dist_base_info_sorted) {
+	info = calloc(r->nr_dist_base_types, sizeof(*info));
+	if (!info) {
 		err = -ENOMEM;
 		goto done;
 	}
-	dist_base_info_sorted_end = dist_base_info_sorted + r->nr_dist_base_types;
+	info_end = info + r->nr_dist_base_types;
 	for (id = 0; id < r->nr_dist_base_types; id++) {
 		dist_t = btf_type_by_id(r->dist_base_btf, id);
-		dist_base_info_sorted[id].name = btf__name_by_offset(r->dist_base_btf,
-								     dist_t->name_off);
-		dist_base_info_sorted[id].id = id;
-		dist_base_info_sorted[id].size = dist_t->size;
-		dist_base_info_sorted[id].needs_size = true;
+		info[id].name = btf__name_by_offset(r->dist_base_btf, dist_t->name_off);
+		info[id].id = id;
+		info[id].size = dist_t->size;
+		info[id].needs_size = true;
 	}
-	qsort(dist_base_info_sorted, r->nr_dist_base_types, sizeof(*dist_base_info_sorted),
-	      cmp_btf_name_size);
+	qsort(info, r->nr_dist_base_types, sizeof(*info), cmp_btf_name_size);
 
 	/* Mark distilled base struct/union members of split BTF structs/unions
 	 * in id_map with BTF_IS_EMBEDDED; this signals that these types
-	 * need to match both name and size, otherwise embeddding the base
+	 * need to match both name and size, otherwise embedding the base
 	 * struct/union in the split type is invalid.
 	 */
 	for (id = r->nr_dist_base_types; id < r->nr_split_types; id++) {
@@ -216,8 +214,7 @@ static int btf_relocate_map_distilled_base(struct btf_relocate *r)
 
 	/* Now search base BTF for matching distilled base BTF types. */
 	for (id = 1; id < r->nr_base_types; id++) {
-		struct btf_name_info *dist_name_info, *dist_name_info_next = NULL;
-		struct btf_name_info base_name_info = {};
+		struct btf_name_info *dist_info, base_info = {};
 		int dist_kind, base_kind;
 
 		base_t = btf_type_by_id(r->base_btf, id);
@@ -225,16 +222,16 @@ static int btf_relocate_map_distilled_base(struct btf_relocate *r)
 		if (!base_t->name_off)
 			continue;
 		base_kind = btf_kind(base_t);
-		base_name_info.id = id;
-		base_name_info.name = btf__name_by_offset(r->base_btf, base_t->name_off);
+		base_info.id = id;
+		base_info.name = btf__name_by_offset(r->base_btf, base_t->name_off);
 		switch (base_kind) {
 		case BTF_KIND_INT:
 		case BTF_KIND_FLOAT:
 		case BTF_KIND_ENUM:
 		case BTF_KIND_ENUM64:
 			/* These types should match both name and size */
-			base_name_info.needs_size = true;
-			base_name_info.size = base_t->size;
+			base_info.needs_size = true;
+			base_info.size = base_t->size;
 			break;
 		case BTF_KIND_FWD:
 			/* No size considerations for fwds. */
@@ -248,31 +245,24 @@ static int btf_relocate_map_distilled_base(struct btf_relocate *r)
 			 * unless corresponding _base_ types to match them are
 			 * missing.
 			 */
-			base_name_info.needs_size = base_name_cnt[base_t->name_off] > 1;
-			base_name_info.size = base_t->size;
+			base_info.needs_size = base_name_cnt[base_t->name_off] > 1;
+			base_info.size = base_t->size;
 			break;
 		default:
 			continue;
 		}
 		/* iterate over all matching distilled base types */
-		for (dist_name_info = search_btf_name_size(&base_name_info, dist_base_info_sorted,
-							   r->nr_dist_base_types);
-		     dist_name_info != NULL; dist_name_info = dist_name_info_next) {
-			/* Are there more distilled matches to process after
-			 * this one?
-			 */
-			dist_name_info_next = dist_name_info + 1;
-			if (dist_name_info_next >= dist_base_info_sorted_end ||
-			    cmp_btf_name_size(&base_name_info, dist_name_info_next))
-				dist_name_info_next = NULL;
-
-			if (!dist_name_info->id || dist_name_info->id > r->nr_dist_base_types) {
+		for (dist_info = search_btf_name_size(&base_info, info, r->nr_dist_base_types);
+		     dist_info != NULL && dist_info < info_end &&
+		     cmp_btf_name_size(&base_info, dist_info) == 0;
+		     dist_info++) {
+			if (!dist_info->id || dist_info->id >= r->nr_dist_base_types) {
 				pr_warn("base BTF id [%d] maps to invalid distilled base BTF id [%d]\n",
-					id, dist_name_info->id);
+					id, dist_info->id);
 				err = -EINVAL;
 				goto done;
 			}
-			dist_t = btf_type_by_id(r->dist_base_btf, dist_name_info->id);
+			dist_t = btf_type_by_id(r->dist_base_btf, dist_info->id);
 			dist_kind = btf_kind(dist_t);
 
 			/* Validate that the found distilled type is compatible.
@@ -319,15 +309,15 @@ static int btf_relocate_map_distilled_base(struct btf_relocate *r)
 				/* size verification is required for embedded
 				 * struct/unions.
 				 */
-				if (r->id_map[dist_name_info->id] == BTF_IS_EMBEDDED &&
+				if (r->id_map[dist_info->id] == BTF_IS_EMBEDDED &&
 				    base_t->size != dist_t->size)
 					continue;
 				break;
 			default:
 				continue;
 			}
-			if (r->id_map[dist_name_info->id] &&
-			    r->id_map[dist_name_info->id] != BTF_IS_EMBEDDED) {
+			if (r->id_map[dist_info->id] &&
+			    r->id_map[dist_info->id] != BTF_IS_EMBEDDED) {
 				/* we already have a match; this tells us that
 				 * multiple base types of the same name
 				 * have the same size, since for cases where
@@ -337,13 +327,13 @@ static int btf_relocate_map_distilled_base(struct btf_relocate *r)
 				 * to in base BTF, so error out.
 				 */
 				pr_warn("distilled base BTF type '%s' [%u], size %u has multiple candidates of the same size (ids [%u, %u]) in base BTF\n",
-					base_name_info.name, dist_name_info->id,
-					base_t->size, id, r->id_map[dist_name_info->id]);
+					base_info.name, dist_info->id,
+					base_t->size, id, r->id_map[dist_info->id]);
 				err = -EINVAL;
 				goto done;
 			}
 			/* map id and name */
-			r->id_map[dist_name_info->id] = id;
+			r->id_map[dist_info->id] = id;
 			r->str_map[dist_t->name_off] = base_t->name_off;
 		}
 	}
@@ -362,7 +352,7 @@ static int btf_relocate_map_distilled_base(struct btf_relocate *r)
 	}
 done:
 	free(base_name_cnt);
-	free(dist_base_info_sorted);
+	free(info);
 	return err;
 }
 
-- 
cgit v1.2.3-58-ga151


From d4e48e3dd45017abdd69a19285d197de897ef44f Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 20 Jun 2024 10:17:29 +0100
Subject: module, bpf: Store BTF base pointer in struct module

...as this will allow split BTF modules with a base BTF
representation (rather than the full vmlinux BTF at time of
BTF encoding) to resolve their references to kernel types in a
way that is more resilient to small changes in kernel types.

This will allow modules that are not built every time the kernel
is to provide more resilient BTF, rather than have it invalidated
every time BTF ids for core kernel types change.

Fields are ordered to avoid holes in struct module.

Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240620091733.1967885-3-alan.maguire@oracle.com
---
 include/linux/module.h | 2 ++
 kernel/module/main.c   | 5 ++++-
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/include/linux/module.h b/include/linux/module.h
index ffa1c603163c..b79d926cae8a 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -509,7 +509,9 @@ struct module {
 #endif
 #ifdef CONFIG_DEBUG_INFO_BTF_MODULES
 	unsigned int btf_data_size;
+	unsigned int btf_base_data_size;
 	void *btf_data;
+	void *btf_base_data;
 #endif
 #ifdef CONFIG_JUMP_LABEL
 	struct jump_entry *jump_entries;
diff --git a/kernel/module/main.c b/kernel/module/main.c
index d18a94b973e1..d9592195c5bb 100644
--- a/kernel/module/main.c
+++ b/kernel/module/main.c
@@ -2166,6 +2166,8 @@ static int find_module_sections(struct module *mod, struct load_info *info)
 #endif
 #ifdef CONFIG_DEBUG_INFO_BTF_MODULES
 	mod->btf_data = any_section_objs(info, ".BTF", 1, &mod->btf_data_size);
+	mod->btf_base_data = any_section_objs(info, ".BTF.base", 1,
+					      &mod->btf_base_data_size);
 #endif
 #ifdef CONFIG_JUMP_LABEL
 	mod->jump_entries = section_objs(info, "__jump_table",
@@ -2590,8 +2592,9 @@ static noinline int do_init_module(struct module *mod)
 	}
 
 #ifdef CONFIG_DEBUG_INFO_BTF_MODULES
-	/* .BTF is not SHF_ALLOC and will get removed, so sanitize pointer */
+	/* .BTF is not SHF_ALLOC and will get removed, so sanitize pointers */
 	mod->btf_data = NULL;
+	mod->btf_base_data = NULL;
 #endif
 	/*
 	 * We want to free module_init, but be aware that kallsyms may be
-- 
cgit v1.2.3-58-ga151


From e7ac331b30555cf1a0826784a346f36dbf800451 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 20 Jun 2024 10:17:30 +0100
Subject: libbpf: Split field iter code into its own file kernel

This will allow it to be shared with the kernel.  No functional change.

Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240620091733.1967885-4-alan.maguire@oracle.com
---
 tools/lib/bpf/Build      |   2 +-
 tools/lib/bpf/btf.c      | 162 ---------------------------------------------
 tools/lib/bpf/btf_iter.c | 169 +++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 170 insertions(+), 163 deletions(-)
 create mode 100644 tools/lib/bpf/btf_iter.c

diff --git a/tools/lib/bpf/Build b/tools/lib/bpf/Build
index 336da6844d42..e2cd558ca0b4 100644
--- a/tools/lib/bpf/Build
+++ b/tools/lib/bpf/Build
@@ -1,4 +1,4 @@
 libbpf-y := libbpf.o bpf.o nlattr.o btf.o libbpf_errno.o str_error.o \
 	    netlink.o bpf_prog_linfo.o libbpf_probes.o hashmap.o \
 	    btf_dump.o ringbuf.o strset.o linker.o gen_loader.o relo_core.o \
-	    usdt.o zip.o elf.o features.o btf_relocate.o
+	    usdt.o zip.o elf.o features.o btf_iter.o btf_relocate.o
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index ef1b2f573c1b..0c0f60cad769 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -5093,168 +5093,6 @@ struct btf *btf__load_module_btf(const char *module_name, struct btf *vmlinux_bt
 	return btf__parse_split(path, vmlinux_btf);
 }
 
-int btf_field_iter_init(struct btf_field_iter *it, struct btf_type *t, enum btf_field_iter_kind iter_kind)
-{
-	it->p = NULL;
-	it->m_idx = -1;
-	it->off_idx = 0;
-	it->vlen = 0;
-
-	switch (iter_kind) {
-	case BTF_FIELD_ITER_IDS:
-		switch (btf_kind(t)) {
-		case BTF_KIND_UNKN:
-		case BTF_KIND_INT:
-		case BTF_KIND_FLOAT:
-		case BTF_KIND_ENUM:
-		case BTF_KIND_ENUM64:
-			it->desc = (struct btf_field_desc) {};
-			break;
-		case BTF_KIND_FWD:
-		case BTF_KIND_CONST:
-		case BTF_KIND_VOLATILE:
-		case BTF_KIND_RESTRICT:
-		case BTF_KIND_PTR:
-		case BTF_KIND_TYPEDEF:
-		case BTF_KIND_FUNC:
-		case BTF_KIND_VAR:
-		case BTF_KIND_DECL_TAG:
-		case BTF_KIND_TYPE_TAG:
-			it->desc = (struct btf_field_desc) { 1, {offsetof(struct btf_type, type)} };
-			break;
-		case BTF_KIND_ARRAY:
-			it->desc = (struct btf_field_desc) {
-				2, {sizeof(struct btf_type) + offsetof(struct btf_array, type),
-				    sizeof(struct btf_type) + offsetof(struct btf_array, index_type)}
-			};
-			break;
-		case BTF_KIND_STRUCT:
-		case BTF_KIND_UNION:
-			it->desc = (struct btf_field_desc) {
-				0, {},
-				sizeof(struct btf_member),
-				1, {offsetof(struct btf_member, type)}
-			};
-			break;
-		case BTF_KIND_FUNC_PROTO:
-			it->desc = (struct btf_field_desc) {
-				1, {offsetof(struct btf_type, type)},
-				sizeof(struct btf_param),
-				1, {offsetof(struct btf_param, type)}
-			};
-			break;
-		case BTF_KIND_DATASEC:
-			it->desc = (struct btf_field_desc) {
-				0, {},
-				sizeof(struct btf_var_secinfo),
-				1, {offsetof(struct btf_var_secinfo, type)}
-			};
-			break;
-		default:
-			return -EINVAL;
-		}
-		break;
-	case BTF_FIELD_ITER_STRS:
-		switch (btf_kind(t)) {
-		case BTF_KIND_UNKN:
-			it->desc = (struct btf_field_desc) {};
-			break;
-		case BTF_KIND_INT:
-		case BTF_KIND_FLOAT:
-		case BTF_KIND_FWD:
-		case BTF_KIND_ARRAY:
-		case BTF_KIND_CONST:
-		case BTF_KIND_VOLATILE:
-		case BTF_KIND_RESTRICT:
-		case BTF_KIND_PTR:
-		case BTF_KIND_TYPEDEF:
-		case BTF_KIND_FUNC:
-		case BTF_KIND_VAR:
-		case BTF_KIND_DECL_TAG:
-		case BTF_KIND_TYPE_TAG:
-		case BTF_KIND_DATASEC:
-			it->desc = (struct btf_field_desc) {
-				1, {offsetof(struct btf_type, name_off)}
-			};
-			break;
-		case BTF_KIND_ENUM:
-			it->desc = (struct btf_field_desc) {
-				1, {offsetof(struct btf_type, name_off)},
-				sizeof(struct btf_enum),
-				1, {offsetof(struct btf_enum, name_off)}
-			};
-			break;
-		case BTF_KIND_ENUM64:
-			it->desc = (struct btf_field_desc) {
-				1, {offsetof(struct btf_type, name_off)},
-				sizeof(struct btf_enum64),
-				1, {offsetof(struct btf_enum64, name_off)}
-			};
-			break;
-		case BTF_KIND_STRUCT:
-		case BTF_KIND_UNION:
-			it->desc = (struct btf_field_desc) {
-				1, {offsetof(struct btf_type, name_off)},
-				sizeof(struct btf_member),
-				1, {offsetof(struct btf_member, name_off)}
-			};
-			break;
-		case BTF_KIND_FUNC_PROTO:
-			it->desc = (struct btf_field_desc) {
-				1, {offsetof(struct btf_type, name_off)},
-				sizeof(struct btf_param),
-				1, {offsetof(struct btf_param, name_off)}
-			};
-			break;
-		default:
-			return -EINVAL;
-		}
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	if (it->desc.m_sz)
-		it->vlen = btf_vlen(t);
-
-	it->p = t;
-	return 0;
-}
-
-__u32 *btf_field_iter_next(struct btf_field_iter *it)
-{
-	if (!it->p)
-		return NULL;
-
-	if (it->m_idx < 0) {
-		if (it->off_idx < it->desc.t_off_cnt)
-			return it->p + it->desc.t_offs[it->off_idx++];
-		/* move to per-member iteration */
-		it->m_idx = 0;
-		it->p += sizeof(struct btf_type);
-		it->off_idx = 0;
-	}
-
-	/* if type doesn't have members, stop */
-	if (it->desc.m_sz == 0) {
-		it->p = NULL;
-		return NULL;
-	}
-
-	if (it->off_idx >= it->desc.m_off_cnt) {
-		/* exhausted this member's fields, go to the next member */
-		it->m_idx++;
-		it->p += it->desc.m_sz;
-		it->off_idx = 0;
-	}
-
-	if (it->m_idx < it->vlen)
-		return it->p + it->desc.m_offs[it->off_idx++];
-
-	it->p = NULL;
-	return NULL;
-}
-
 int btf_ext_visit_type_ids(struct btf_ext *btf_ext, type_id_visit_fn visit, void *ctx)
 {
 	const struct btf_ext_info *seg;
diff --git a/tools/lib/bpf/btf_iter.c b/tools/lib/bpf/btf_iter.c
new file mode 100644
index 000000000000..c308aa60285d
--- /dev/null
+++ b/tools/lib/bpf/btf_iter.c
@@ -0,0 +1,169 @@
+// SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
+/* Copyright (c) 2021 Facebook */
+/* Copyright (c) 2024, Oracle and/or its affiliates. */
+
+#include "btf.h"
+#include "libbpf_internal.h"
+
+int btf_field_iter_init(struct btf_field_iter *it, struct btf_type *t,
+			enum btf_field_iter_kind iter_kind)
+{
+	it->p = NULL;
+	it->m_idx = -1;
+	it->off_idx = 0;
+	it->vlen = 0;
+
+	switch (iter_kind) {
+	case BTF_FIELD_ITER_IDS:
+		switch (btf_kind(t)) {
+		case BTF_KIND_UNKN:
+		case BTF_KIND_INT:
+		case BTF_KIND_FLOAT:
+		case BTF_KIND_ENUM:
+		case BTF_KIND_ENUM64:
+			it->desc = (struct btf_field_desc) {};
+			break;
+		case BTF_KIND_FWD:
+		case BTF_KIND_CONST:
+		case BTF_KIND_VOLATILE:
+		case BTF_KIND_RESTRICT:
+		case BTF_KIND_PTR:
+		case BTF_KIND_TYPEDEF:
+		case BTF_KIND_FUNC:
+		case BTF_KIND_VAR:
+		case BTF_KIND_DECL_TAG:
+		case BTF_KIND_TYPE_TAG:
+			it->desc = (struct btf_field_desc) { 1, {offsetof(struct btf_type, type)} };
+			break;
+		case BTF_KIND_ARRAY:
+			it->desc = (struct btf_field_desc) {
+				2, {sizeof(struct btf_type) + offsetof(struct btf_array, type),
+				sizeof(struct btf_type) + offsetof(struct btf_array, index_type)}
+			};
+			break;
+		case BTF_KIND_STRUCT:
+		case BTF_KIND_UNION:
+			it->desc = (struct btf_field_desc) {
+				0, {},
+				sizeof(struct btf_member),
+				1, {offsetof(struct btf_member, type)}
+			};
+			break;
+		case BTF_KIND_FUNC_PROTO:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, type)},
+				sizeof(struct btf_param),
+				1, {offsetof(struct btf_param, type)}
+			};
+			break;
+		case BTF_KIND_DATASEC:
+			it->desc = (struct btf_field_desc) {
+				0, {},
+				sizeof(struct btf_var_secinfo),
+				1, {offsetof(struct btf_var_secinfo, type)}
+			};
+			break;
+		default:
+			return -EINVAL;
+		}
+		break;
+	case BTF_FIELD_ITER_STRS:
+		switch (btf_kind(t)) {
+		case BTF_KIND_UNKN:
+			it->desc = (struct btf_field_desc) {};
+			break;
+		case BTF_KIND_INT:
+		case BTF_KIND_FLOAT:
+		case BTF_KIND_FWD:
+		case BTF_KIND_ARRAY:
+		case BTF_KIND_CONST:
+		case BTF_KIND_VOLATILE:
+		case BTF_KIND_RESTRICT:
+		case BTF_KIND_PTR:
+		case BTF_KIND_TYPEDEF:
+		case BTF_KIND_FUNC:
+		case BTF_KIND_VAR:
+		case BTF_KIND_DECL_TAG:
+		case BTF_KIND_TYPE_TAG:
+		case BTF_KIND_DATASEC:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, name_off)}
+			};
+			break;
+		case BTF_KIND_ENUM:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, name_off)},
+				sizeof(struct btf_enum),
+				1, {offsetof(struct btf_enum, name_off)}
+			};
+			break;
+		case BTF_KIND_ENUM64:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, name_off)},
+				sizeof(struct btf_enum64),
+				1, {offsetof(struct btf_enum64, name_off)}
+			};
+			break;
+		case BTF_KIND_STRUCT:
+		case BTF_KIND_UNION:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, name_off)},
+				sizeof(struct btf_member),
+				1, {offsetof(struct btf_member, name_off)}
+			};
+			break;
+		case BTF_KIND_FUNC_PROTO:
+			it->desc = (struct btf_field_desc) {
+				1, {offsetof(struct btf_type, name_off)},
+				sizeof(struct btf_param),
+				1, {offsetof(struct btf_param, name_off)}
+			};
+			break;
+		default:
+			return -EINVAL;
+		}
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (it->desc.m_sz)
+		it->vlen = btf_vlen(t);
+
+	it->p = t;
+	return 0;
+}
+
+__u32 *btf_field_iter_next(struct btf_field_iter *it)
+{
+	if (!it->p)
+		return NULL;
+
+	if (it->m_idx < 0) {
+		if (it->off_idx < it->desc.t_off_cnt)
+			return it->p + it->desc.t_offs[it->off_idx++];
+		/* move to per-member iteration */
+		it->m_idx = 0;
+		it->p += sizeof(struct btf_type);
+		it->off_idx = 0;
+	}
+
+	/* if type doesn't have members, stop */
+	if (it->desc.m_sz == 0) {
+		it->p = NULL;
+		return NULL;
+	}
+
+	if (it->off_idx >= it->desc.m_off_cnt) {
+		/* exhausted this member's fields, go to the next member */
+		it->m_idx++;
+		it->p += it->desc.m_sz;
+		it->off_idx = 0;
+	}
+
+	if (it->m_idx < it->vlen)
+		return it->p + it->desc.m_offs[it->off_idx++];
+
+	it->p = NULL;
+	return NULL;
+}
-- 
cgit v1.2.3-58-ga151


From 8646db238997df36c6ad71a9d7e0b52ceee221b2 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 20 Jun 2024 10:17:31 +0100
Subject: libbpf,bpf: Share BTF relocate-related code with kernel

Share relocation implementation with the kernel.  As part of this,
we also need the type/string iteration functions so also share
btf_iter.c file. Relocation code in kernel and userspace is identical
save for the impementation of the reparenting of split BTF to the
relocated base BTF and retrieval of the BTF header from "struct btf";
these small functions need separate user-space and kernel implementations
for the separate "struct btf"s they operate upon.

One other wrinkle on the kernel side is we have to map .BTF.ids in
modules as they were generated with the type ids used at BTF encoding
time. btf_relocate() optionally returns an array mapping from old BTF
ids to relocated ids, so we use that to fix up these references where
needed for kfuncs.

Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240620091733.1967885-5-alan.maguire@oracle.com
---
 include/linux/btf.h          |  64 ++++++++++++++++
 kernel/bpf/Makefile          |   8 +-
 kernel/bpf/btf.c             | 176 ++++++++++++++++++++++++++++++-------------
 tools/lib/bpf/btf_iter.c     |   8 ++
 tools/lib/bpf/btf_relocate.c |  23 ++++++
 5 files changed, 226 insertions(+), 53 deletions(-)

diff --git a/include/linux/btf.h b/include/linux/btf.h
index 56d91daacdba..d199fa17abb4 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -140,6 +140,7 @@ extern const struct file_operations btf_fops;
 const char *btf_get_name(const struct btf *btf);
 void btf_get(struct btf *btf);
 void btf_put(struct btf *btf);
+const struct btf_header *btf_header(const struct btf *btf);
 int btf_new_fd(const union bpf_attr *attr, bpfptr_t uattr, u32 uattr_sz);
 struct btf *btf_get_by_fd(int fd);
 int btf_get_info_by_fd(const struct btf *btf,
@@ -212,8 +213,10 @@ int btf_get_fd_by_id(u32 id);
 u32 btf_obj_id(const struct btf *btf);
 bool btf_is_kernel(const struct btf *btf);
 bool btf_is_module(const struct btf *btf);
+bool btf_is_vmlinux(const struct btf *btf);
 struct module *btf_try_get_module(const struct btf *btf);
 u32 btf_nr_types(const struct btf *btf);
+struct btf *btf_base_btf(const struct btf *btf);
 bool btf_member_is_reg_int(const struct btf *btf, const struct btf_type *s,
 			   const struct btf_member *m,
 			   u32 expected_offset, u32 expected_size);
@@ -339,6 +342,11 @@ static inline u8 btf_int_offset(const struct btf_type *t)
 	return BTF_INT_OFFSET(*(u32 *)(t + 1));
 }
 
+static inline __u8 btf_int_bits(const struct btf_type *t)
+{
+	return BTF_INT_BITS(*(__u32 *)(t + 1));
+}
+
 static inline bool btf_type_is_scalar(const struct btf_type *t)
 {
 	return btf_type_is_int(t) || btf_type_is_enum(t);
@@ -478,6 +486,11 @@ static inline struct btf_param *btf_params(const struct btf_type *t)
 	return (struct btf_param *)(t + 1);
 }
 
+static inline struct btf_decl_tag *btf_decl_tag(const struct btf_type *t)
+{
+	return (struct btf_decl_tag *)(t + 1);
+}
+
 static inline int btf_id_cmp_func(const void *a, const void *b)
 {
 	const int *pa = a, *pb = b;
@@ -515,9 +528,38 @@ static inline const struct bpf_struct_ops_desc *bpf_struct_ops_find(struct btf *
 }
 #endif
 
+enum btf_field_iter_kind {
+	BTF_FIELD_ITER_IDS,
+	BTF_FIELD_ITER_STRS,
+};
+
+struct btf_field_desc {
+	/* once-per-type offsets */
+	int t_off_cnt, t_offs[2];
+	/* member struct size, or zero, if no members */
+	int m_sz;
+	/* repeated per-member offsets */
+	int m_off_cnt, m_offs[1];
+};
+
+struct btf_field_iter {
+	struct btf_field_desc desc;
+	void *p;
+	int m_idx;
+	int off_idx;
+	int vlen;
+};
+
 #ifdef CONFIG_BPF_SYSCALL
 const struct btf_type *btf_type_by_id(const struct btf *btf, u32 type_id);
+void btf_set_base_btf(struct btf *btf, const struct btf *base_btf);
+int btf_relocate(struct btf *btf, const struct btf *base_btf, __u32 **map_ids);
+int btf_field_iter_init(struct btf_field_iter *it, struct btf_type *t,
+			enum btf_field_iter_kind iter_kind);
+__u32 *btf_field_iter_next(struct btf_field_iter *it);
+
 const char *btf_name_by_offset(const struct btf *btf, u32 offset);
+const char *btf_str_by_offset(const struct btf *btf, u32 offset);
 struct btf *btf_parse_vmlinux(void);
 struct btf *bpf_prog_get_target_btf(const struct bpf_prog *prog);
 u32 *btf_kfunc_id_set_contains(const struct btf *btf, u32 kfunc_btf_id,
@@ -544,6 +586,28 @@ static inline const struct btf_type *btf_type_by_id(const struct btf *btf,
 {
 	return NULL;
 }
+
+static inline void btf_set_base_btf(struct btf *btf, const struct btf *base_btf)
+{
+}
+
+static inline int btf_relocate(void *log, struct btf *btf, const struct btf *base_btf,
+			       __u32 **map_ids)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int btf_field_iter_init(struct btf_field_iter *it, struct btf_type *t,
+				      enum btf_field_iter_kind iter_kind)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline __u32 *btf_field_iter_next(struct btf_field_iter *it)
+{
+	return NULL;
+}
+
 static inline const char *btf_name_by_offset(const struct btf *btf,
 					     u32 offset)
 {
diff --git a/kernel/bpf/Makefile b/kernel/bpf/Makefile
index 7eb9ad3a3ae6..0291eef9ce92 100644
--- a/kernel/bpf/Makefile
+++ b/kernel/bpf/Makefile
@@ -50,5 +50,11 @@ endif
 obj-$(CONFIG_BPF_PRELOAD) += preload/
 
 obj-$(CONFIG_BPF_SYSCALL) += relo_core.o
-$(obj)/relo_core.o: $(srctree)/tools/lib/bpf/relo_core.c FORCE
+obj-$(CONFIG_BPF_SYSCALL) += btf_iter.o
+obj-$(CONFIG_BPF_SYSCALL) += btf_relocate.o
+
+# Some source files are common to libbpf.
+vpath %.c $(srctree)/kernel/bpf:$(srctree)/tools/lib/bpf
+
+$(obj)/%.o: %.c FORCE
 	$(call if_changed_rule,cc_o_c)
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index ce4707968217..8e12cb80ba73 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -274,6 +274,7 @@ struct btf {
 	u32 start_str_off; /* first string offset (0 for base BTF) */
 	char name[MODULE_NAME_LEN];
 	bool kernel_btf;
+	__u32 *base_id_map; /* map from distilled base BTF -> vmlinux BTF ids */
 };
 
 enum verifier_phase {
@@ -530,6 +531,11 @@ static bool btf_type_is_decl_tag_target(const struct btf_type *t)
 	       btf_type_is_var(t) || btf_type_is_typedef(t);
 }
 
+bool btf_is_vmlinux(const struct btf *btf)
+{
+	return btf->kernel_btf && !btf->base_btf;
+}
+
 u32 btf_nr_types(const struct btf *btf)
 {
 	u32 total = 0;
@@ -772,7 +778,7 @@ static bool __btf_name_char_ok(char c, bool first)
 	return true;
 }
 
-static const char *btf_str_by_offset(const struct btf *btf, u32 offset)
+const char *btf_str_by_offset(const struct btf *btf, u32 offset)
 {
 	while (offset < btf->start_str_off)
 		btf = btf->base_btf;
@@ -1670,14 +1676,8 @@ static void btf_free_kfunc_set_tab(struct btf *btf)
 
 	if (!tab)
 		return;
-	/* For module BTF, we directly assign the sets being registered, so
-	 * there is nothing to free except kfunc_set_tab.
-	 */
-	if (btf_is_module(btf))
-		goto free_tab;
 	for (hook = 0; hook < ARRAY_SIZE(tab->sets); hook++)
 		kfree(tab->sets[hook]);
-free_tab:
 	kfree(tab);
 	btf->kfunc_set_tab = NULL;
 }
@@ -1735,7 +1735,12 @@ static void btf_free(struct btf *btf)
 	kvfree(btf->types);
 	kvfree(btf->resolved_sizes);
 	kvfree(btf->resolved_ids);
-	kvfree(btf->data);
+	/* vmlinux does not allocate btf->data, it simply points it at
+	 * __start_BTF.
+	 */
+	if (!btf_is_vmlinux(btf))
+		kvfree(btf->data);
+	kvfree(btf->base_id_map);
 	kfree(btf);
 }
 
@@ -1764,6 +1769,23 @@ void btf_put(struct btf *btf)
 	}
 }
 
+struct btf *btf_base_btf(const struct btf *btf)
+{
+	return btf->base_btf;
+}
+
+const struct btf_header *btf_header(const struct btf *btf)
+{
+	return &btf->hdr;
+}
+
+void btf_set_base_btf(struct btf *btf, const struct btf *base_btf)
+{
+	btf->base_btf = (struct btf *)base_btf;
+	btf->start_id = btf_nr_types(base_btf);
+	btf->start_str_off = base_btf->hdr.str_len;
+}
+
 static int env_resolve_init(struct btf_verifier_env *env)
 {
 	struct btf *btf = env->btf;
@@ -6083,23 +6105,15 @@ int get_kern_ctx_btf_id(struct bpf_verifier_log *log, enum bpf_prog_type prog_ty
 BTF_ID_LIST(bpf_ctx_convert_btf_id)
 BTF_ID(struct, bpf_ctx_convert)
 
-struct btf *btf_parse_vmlinux(void)
+static struct btf *btf_parse_base(struct btf_verifier_env *env, const char *name,
+				  void *data, unsigned int data_size)
 {
-	struct btf_verifier_env *env = NULL;
-	struct bpf_verifier_log *log;
 	struct btf *btf = NULL;
 	int err;
 
 	if (!IS_ENABLED(CONFIG_DEBUG_INFO_BTF))
 		return ERR_PTR(-ENOENT);
 
-	env = kzalloc(sizeof(*env), GFP_KERNEL | __GFP_NOWARN);
-	if (!env)
-		return ERR_PTR(-ENOMEM);
-
-	log = &env->log;
-	log->level = BPF_LOG_KERNEL;
-
 	btf = kzalloc(sizeof(*btf), GFP_KERNEL | __GFP_NOWARN);
 	if (!btf) {
 		err = -ENOMEM;
@@ -6107,10 +6121,10 @@ struct btf *btf_parse_vmlinux(void)
 	}
 	env->btf = btf;
 
-	btf->data = __start_BTF;
-	btf->data_size = __stop_BTF - __start_BTF;
+	btf->data = data;
+	btf->data_size = data_size;
 	btf->kernel_btf = true;
-	snprintf(btf->name, sizeof(btf->name), "vmlinux");
+	snprintf(btf->name, sizeof(btf->name), "%s", name);
 
 	err = btf_parse_hdr(env);
 	if (err)
@@ -6130,20 +6144,11 @@ struct btf *btf_parse_vmlinux(void)
 	if (err)
 		goto errout;
 
-	/* btf_parse_vmlinux() runs under bpf_verifier_lock */
-	bpf_ctx_convert.t = btf_type_by_id(btf, bpf_ctx_convert_btf_id[0]);
-
 	refcount_set(&btf->refcnt, 1);
 
-	err = btf_alloc_id(btf);
-	if (err)
-		goto errout;
-
-	btf_verifier_env_free(env);
 	return btf;
 
 errout:
-	btf_verifier_env_free(env);
 	if (btf) {
 		kvfree(btf->types);
 		kfree(btf);
@@ -6151,19 +6156,61 @@ errout:
 	return ERR_PTR(err);
 }
 
+struct btf *btf_parse_vmlinux(void)
+{
+	struct btf_verifier_env *env = NULL;
+	struct bpf_verifier_log *log;
+	struct btf *btf;
+	int err;
+
+	env = kzalloc(sizeof(*env), GFP_KERNEL | __GFP_NOWARN);
+	if (!env)
+		return ERR_PTR(-ENOMEM);
+
+	log = &env->log;
+	log->level = BPF_LOG_KERNEL;
+	btf = btf_parse_base(env, "vmlinux", __start_BTF, __stop_BTF - __start_BTF);
+	if (IS_ERR(btf))
+		goto err_out;
+
+	/* btf_parse_vmlinux() runs under bpf_verifier_lock */
+	bpf_ctx_convert.t = btf_type_by_id(btf, bpf_ctx_convert_btf_id[0]);
+	err = btf_alloc_id(btf);
+	if (err) {
+		btf_free(btf);
+		btf = ERR_PTR(err);
+	}
+err_out:
+	btf_verifier_env_free(env);
+	return btf;
+}
+
 #ifdef CONFIG_DEBUG_INFO_BTF_MODULES
 
-static struct btf *btf_parse_module(const char *module_name, const void *data, unsigned int data_size)
+/* If .BTF_ids section was created with distilled base BTF, both base and
+ * split BTF ids will need to be mapped to actual base/split ids for
+ * BTF now that it has been relocated.
+ */
+static __u32 btf_relocate_id(const struct btf *btf, __u32 id)
+{
+	if (!btf->base_btf || !btf->base_id_map)
+		return id;
+	return btf->base_id_map[id];
+}
+
+static struct btf *btf_parse_module(const char *module_name, const void *data,
+				    unsigned int data_size, void *base_data,
+				    unsigned int base_data_size)
 {
+	struct btf *btf = NULL, *vmlinux_btf, *base_btf = NULL;
 	struct btf_verifier_env *env = NULL;
 	struct bpf_verifier_log *log;
-	struct btf *btf = NULL, *base_btf;
-	int err;
+	int err = 0;
 
-	base_btf = bpf_get_btf_vmlinux();
-	if (IS_ERR(base_btf))
-		return base_btf;
-	if (!base_btf)
+	vmlinux_btf = bpf_get_btf_vmlinux();
+	if (IS_ERR(vmlinux_btf))
+		return vmlinux_btf;
+	if (!vmlinux_btf)
 		return ERR_PTR(-EINVAL);
 
 	env = kzalloc(sizeof(*env), GFP_KERNEL | __GFP_NOWARN);
@@ -6173,6 +6220,16 @@ static struct btf *btf_parse_module(const char *module_name, const void *data, u
 	log = &env->log;
 	log->level = BPF_LOG_KERNEL;
 
+	if (base_data) {
+		base_btf = btf_parse_base(env, ".BTF.base", base_data, base_data_size);
+		if (IS_ERR(base_btf)) {
+			err = PTR_ERR(base_btf);
+			goto errout;
+		}
+	} else {
+		base_btf = vmlinux_btf;
+	}
+
 	btf = kzalloc(sizeof(*btf), GFP_KERNEL | __GFP_NOWARN);
 	if (!btf) {
 		err = -ENOMEM;
@@ -6212,12 +6269,22 @@ static struct btf *btf_parse_module(const char *module_name, const void *data, u
 	if (err)
 		goto errout;
 
+	if (base_btf != vmlinux_btf) {
+		err = btf_relocate(btf, vmlinux_btf, &btf->base_id_map);
+		if (err)
+			goto errout;
+		btf_free(base_btf);
+		base_btf = vmlinux_btf;
+	}
+
 	btf_verifier_env_free(env);
 	refcount_set(&btf->refcnt, 1);
 	return btf;
 
 errout:
 	btf_verifier_env_free(env);
+	if (base_btf != vmlinux_btf)
+		btf_free(base_btf);
 	if (btf) {
 		kvfree(btf->data);
 		kvfree(btf->types);
@@ -7770,7 +7837,8 @@ static int btf_module_notify(struct notifier_block *nb, unsigned long op,
 			err = -ENOMEM;
 			goto out;
 		}
-		btf = btf_parse_module(mod->name, mod->btf_data, mod->btf_data_size);
+		btf = btf_parse_module(mod->name, mod->btf_data, mod->btf_data_size,
+				       mod->btf_base_data, mod->btf_base_data_size);
 		if (IS_ERR(btf)) {
 			kfree(btf_mod);
 			if (!IS_ENABLED(CONFIG_MODULE_ALLOW_BTF_MISMATCH)) {
@@ -8094,7 +8162,7 @@ static int btf_populate_kfunc_set(struct btf *btf, enum btf_kfunc_hook hook,
 	bool add_filter = !!kset->filter;
 	struct btf_kfunc_set_tab *tab;
 	struct btf_id_set8 *set;
-	u32 set_cnt;
+	u32 set_cnt, i;
 	int ret;
 
 	if (hook >= BTF_KFUNC_HOOK_MAX) {
@@ -8140,21 +8208,15 @@ static int btf_populate_kfunc_set(struct btf *btf, enum btf_kfunc_hook hook,
 		goto end;
 	}
 
-	/* We don't need to allocate, concatenate, and sort module sets, because
-	 * only one is allowed per hook. Hence, we can directly assign the
-	 * pointer and return.
-	 */
-	if (!vmlinux_set) {
-		tab->sets[hook] = add_set;
-		goto do_add_filter;
-	}
-
 	/* In case of vmlinux sets, there may be more than one set being
 	 * registered per hook. To create a unified set, we allocate a new set
 	 * and concatenate all individual sets being registered. While each set
 	 * is individually sorted, they may become unsorted when concatenated,
 	 * hence re-sorting the final set again is required to make binary
 	 * searching the set using btf_id_set8_contains function work.
+	 *
+	 * For module sets, we need to allocate as we may need to relocate
+	 * BTF ids.
 	 */
 	set_cnt = set ? set->cnt : 0;
 
@@ -8184,11 +8246,14 @@ static int btf_populate_kfunc_set(struct btf *btf, enum btf_kfunc_hook hook,
 
 	/* Concatenate the two sets */
 	memcpy(set->pairs + set->cnt, add_set->pairs, add_set->cnt * sizeof(set->pairs[0]));
+	/* Now that the set is copied, update with relocated BTF ids */
+	for (i = set->cnt; i < set->cnt + add_set->cnt; i++)
+		set->pairs[i].id = btf_relocate_id(btf, set->pairs[i].id);
+
 	set->cnt += add_set->cnt;
 
 	sort(set->pairs, set->cnt, sizeof(set->pairs[0]), btf_id_cmp_func, NULL);
 
-do_add_filter:
 	if (add_filter) {
 		hook_filter = &tab->hook_filters[hook];
 		hook_filter->filters[hook_filter->nr_filters++] = kset->filter;
@@ -8308,7 +8373,7 @@ static int __register_btf_kfunc_id_set(enum btf_kfunc_hook hook,
 		return PTR_ERR(btf);
 
 	for (i = 0; i < kset->set->cnt; i++) {
-		ret = btf_check_kfunc_protos(btf, kset->set->pairs[i].id,
+		ret = btf_check_kfunc_protos(btf, btf_relocate_id(btf, kset->set->pairs[i].id),
 					     kset->set->pairs[i].flags);
 		if (ret)
 			goto err_out;
@@ -8372,7 +8437,7 @@ static int btf_check_dtor_kfuncs(struct btf *btf, const struct btf_id_dtor_kfunc
 	u32 nr_args, i;
 
 	for (i = 0; i < cnt; i++) {
-		dtor_btf_id = dtors[i].kfunc_btf_id;
+		dtor_btf_id = btf_relocate_id(btf, dtors[i].kfunc_btf_id);
 
 		dtor_func = btf_type_by_id(btf, dtor_btf_id);
 		if (!dtor_func || !btf_type_is_func(dtor_func))
@@ -8407,7 +8472,7 @@ int register_btf_id_dtor_kfuncs(const struct btf_id_dtor_kfunc *dtors, u32 add_c
 {
 	struct btf_id_dtor_kfunc_tab *tab;
 	struct btf *btf;
-	u32 tab_cnt;
+	u32 tab_cnt, i;
 	int ret;
 
 	btf = btf_get_module_btf(owner);
@@ -8458,6 +8523,13 @@ int register_btf_id_dtor_kfuncs(const struct btf_id_dtor_kfunc *dtors, u32 add_c
 	btf->dtor_kfunc_tab = tab;
 
 	memcpy(tab->dtors + tab->cnt, dtors, add_cnt * sizeof(tab->dtors[0]));
+
+	/* remap BTF ids based on BTF relocation (if any) */
+	for (i = tab_cnt; i < tab_cnt + add_cnt; i++) {
+		tab->dtors[i].btf_id = btf_relocate_id(btf, tab->dtors[i].btf_id);
+		tab->dtors[i].kfunc_btf_id = btf_relocate_id(btf, tab->dtors[i].kfunc_btf_id);
+	}
+
 	tab->cnt += add_cnt;
 
 	sort(tab->dtors, tab->cnt, sizeof(tab->dtors[0]), btf_id_cmp_func, NULL);
diff --git a/tools/lib/bpf/btf_iter.c b/tools/lib/bpf/btf_iter.c
index c308aa60285d..9a6c822c2294 100644
--- a/tools/lib/bpf/btf_iter.c
+++ b/tools/lib/bpf/btf_iter.c
@@ -2,8 +2,16 @@
 /* Copyright (c) 2021 Facebook */
 /* Copyright (c) 2024, Oracle and/or its affiliates. */
 
+#ifdef __KERNEL__
+#include <linux/bpf.h>
+#include <linux/btf.h>
+
+#define btf_var_secinfos(t)	(struct btf_var_secinfo *)btf_type_var_secinfo(t)
+
+#else
 #include "btf.h"
 #include "libbpf_internal.h"
+#endif
 
 int btf_field_iter_init(struct btf_field_iter *it, struct btf_type *t,
 			enum btf_field_iter_kind iter_kind)
diff --git a/tools/lib/bpf/btf_relocate.c b/tools/lib/bpf/btf_relocate.c
index 23a41fb03e0d..2281dbbafa11 100644
--- a/tools/lib/bpf/btf_relocate.c
+++ b/tools/lib/bpf/btf_relocate.c
@@ -5,11 +5,34 @@
 #define _GNU_SOURCE
 #endif
 
+#ifdef __KERNEL__
+#include <linux/bpf.h>
+#include <linux/bsearch.h>
+#include <linux/btf.h>
+#include <linux/sort.h>
+#include <linux/string.h>
+#include <linux/bpf_verifier.h>
+
+#define btf_type_by_id				(struct btf_type *)btf_type_by_id
+#define btf__type_cnt				btf_nr_types
+#define btf__base_btf				btf_base_btf
+#define btf__name_by_offset			btf_name_by_offset
+#define btf__str_by_offset			btf_str_by_offset
+#define btf_kflag				btf_type_kflag
+
+#define calloc(nmemb, sz)			kvcalloc(nmemb, sz, GFP_KERNEL | __GFP_NOWARN)
+#define free(ptr)				kvfree(ptr)
+#define qsort(base, num, sz, cmp)		sort(base, num, sz, cmp, NULL)
+
+#else
+
 #include "btf.h"
 #include "bpf.h"
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
+#endif /* __KERNEL__ */
+
 struct btf;
 
 struct btf_relocate {
-- 
cgit v1.2.3-58-ga151


From 46fb0b62ea29c0dbcb3e44f1d67aafe79bc6e045 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 20 Jun 2024 10:17:32 +0100
Subject: kbuild,bpf: Add module-specific pahole flags for distilled base BTF

Support creation of module BTF along with distilled base BTF;
the latter is stored in a .BTF.base ELF section and supplements
split BTF references to base BTF with information about base types,
allowing for later relocation of split BTF with a (possibly
changed) base.  resolve_btfids detects the presence of a .BTF.base
section and will use it instead of the base BTF it is passed in
BTF id resolution.

Modules will be built with a distilled .BTF.base section for external
module build, i.e.

make -C. -M=path2/module

...while in-tree module build as part of a normal kernel build will
not generate distilled base BTF; this is because in-tree modules
change with the kernel and do not require BTF relocation for the
running vmlinux.

Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20240620091733.1967885-6-alan.maguire@oracle.com
---
 scripts/Makefile.btf      | 5 +++++
 scripts/Makefile.modfinal | 2 +-
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/scripts/Makefile.btf b/scripts/Makefile.btf
index 2597e3d4d6e0..b75f09f3f424 100644
--- a/scripts/Makefile.btf
+++ b/scripts/Makefile.btf
@@ -21,8 +21,13 @@ else
 # Switch to using --btf_features for v1.26 and later.
 pahole-flags-$(call test-ge, $(pahole-ver), 126)  = -j --btf_features=encode_force,var,float,enum64,decl_tag,type_tag,optimized_func,consistent_func,decl_tag_kfuncs
 
+ifneq ($(KBUILD_EXTMOD),)
+module-pahole-flags-$(call test-ge, $(pahole-ver), 126) += --btf_features=distilled_base
+endif
+
 endif
 
 pahole-flags-$(CONFIG_PAHOLE_HAS_LANG_EXCLUDE)		+= --lang_exclude=rust
 
 export PAHOLE_FLAGS := $(pahole-flags-y)
+export MODULE_PAHOLE_FLAGS := $(module-pahole-flags-y)
diff --git a/scripts/Makefile.modfinal b/scripts/Makefile.modfinal
index 3bec9043e4f3..1fa98b5e952b 100644
--- a/scripts/Makefile.modfinal
+++ b/scripts/Makefile.modfinal
@@ -41,7 +41,7 @@ quiet_cmd_btf_ko = BTF [M] $@
 	if [ ! -f vmlinux ]; then					\
 		printf "Skipping BTF generation for %s due to unavailability of vmlinux\n" $@ 1>&2; \
 	else								\
-		LLVM_OBJCOPY="$(OBJCOPY)" $(PAHOLE) -J $(PAHOLE_FLAGS) --btf_base vmlinux $@; \
+		LLVM_OBJCOPY="$(OBJCOPY)" $(PAHOLE) -J $(PAHOLE_FLAGS) $(MODULE_PAHOLE_FLAGS) --btf_base vmlinux $@; \
 		$(RESOLVE_BTFIDS) -b vmlinux $@; 			\
 	fi;
 
-- 
cgit v1.2.3-58-ga151


From 47a8cf0c5b3f6769b9d558301735c75119a0a165 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 20 Jun 2024 10:17:33 +0100
Subject: selftests/bpf: Add kfunc_call test for simple dtor in bpf_testmod

add simple kfuncs to create/destroy a context type to bpf_testmod,
register them and add a kfunc_call test to use them.  This provides
test coverage for registration of dtor kfuncs from modules.

By transferring the context pointer to a map value as a __kptr
we also trigger the map-based dtor cleanup logic, improving test
coverage.

Suggested-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240620091733.1967885-7-alan.maguire@oracle.com
---
 .../selftests/bpf/bpf_testmod/bpf_testmod.c        | 46 ++++++++++++++++++++++
 .../selftests/bpf/bpf_testmod/bpf_testmod_kfunc.h  |  9 +++++
 .../testing/selftests/bpf/prog_tests/kfunc_call.c  |  1 +
 .../testing/selftests/bpf/progs/kfunc_call_test.c  | 37 +++++++++++++++++
 4 files changed, 93 insertions(+)

diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index 49f9a311e49b..d8bd01d8560b 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -159,6 +159,37 @@ __bpf_kfunc void bpf_kfunc_dynptr_test(struct bpf_dynptr *ptr,
 {
 }
 
+__bpf_kfunc struct bpf_testmod_ctx *
+bpf_testmod_ctx_create(int *err)
+{
+	struct bpf_testmod_ctx *ctx;
+
+	ctx = kzalloc(sizeof(*ctx), GFP_ATOMIC);
+	if (!ctx) {
+		*err = -ENOMEM;
+		return NULL;
+	}
+	refcount_set(&ctx->usage, 1);
+
+	return ctx;
+}
+
+static void testmod_free_cb(struct rcu_head *head)
+{
+	struct bpf_testmod_ctx *ctx;
+
+	ctx = container_of(head, struct bpf_testmod_ctx, rcu);
+	kfree(ctx);
+}
+
+__bpf_kfunc void bpf_testmod_ctx_release(struct bpf_testmod_ctx *ctx)
+{
+	if (!ctx)
+		return;
+	if (refcount_dec_and_test(&ctx->usage))
+		call_rcu(&ctx->rcu, testmod_free_cb);
+}
+
 struct bpf_testmod_btf_type_tag_1 {
 	int a;
 };
@@ -369,8 +400,14 @@ BTF_ID_FLAGS(func, bpf_iter_testmod_seq_next, KF_ITER_NEXT | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_iter_testmod_seq_destroy, KF_ITER_DESTROY)
 BTF_ID_FLAGS(func, bpf_kfunc_common_test)
 BTF_ID_FLAGS(func, bpf_kfunc_dynptr_test)
+BTF_ID_FLAGS(func, bpf_testmod_ctx_create, KF_ACQUIRE | KF_RET_NULL)
+BTF_ID_FLAGS(func, bpf_testmod_ctx_release, KF_RELEASE)
 BTF_KFUNCS_END(bpf_testmod_common_kfunc_ids)
 
+BTF_ID_LIST(bpf_testmod_dtor_ids)
+BTF_ID(struct, bpf_testmod_ctx)
+BTF_ID(func, bpf_testmod_ctx_release)
+
 static const struct btf_kfunc_id_set bpf_testmod_common_kfunc_set = {
 	.owner = THIS_MODULE,
 	.set   = &bpf_testmod_common_kfunc_ids,
@@ -904,6 +941,12 @@ extern int bpf_fentry_test1(int a);
 
 static int bpf_testmod_init(void)
 {
+	const struct btf_id_dtor_kfunc bpf_testmod_dtors[] = {
+		{
+			.btf_id		= bpf_testmod_dtor_ids[0],
+			.kfunc_btf_id	= bpf_testmod_dtor_ids[1]
+		},
+	};
 	int ret;
 
 	ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_UNSPEC, &bpf_testmod_common_kfunc_set);
@@ -912,6 +955,9 @@ static int bpf_testmod_init(void)
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_SYSCALL, &bpf_testmod_kfunc_set);
 	ret = ret ?: register_bpf_struct_ops(&bpf_bpf_testmod_ops, bpf_testmod_ops);
 	ret = ret ?: register_bpf_struct_ops(&bpf_testmod_ops2, bpf_testmod_ops2);
+	ret = ret ?: register_btf_id_dtor_kfuncs(bpf_testmod_dtors,
+						 ARRAY_SIZE(bpf_testmod_dtors),
+						 THIS_MODULE);
 	if (ret < 0)
 		return ret;
 	if (bpf_fentry_test1(0) < 0)
diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod_kfunc.h b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod_kfunc.h
index f9809517e7fa..e587a79f2239 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod_kfunc.h
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod_kfunc.h
@@ -80,6 +80,11 @@ struct sendmsg_args {
 	int msglen;
 };
 
+struct bpf_testmod_ctx {
+	struct callback_head	rcu;
+	refcount_t		usage;
+};
+
 struct prog_test_ref_kfunc *
 bpf_kfunc_call_test_acquire(unsigned long *scalar_ptr) __ksym;
 void bpf_kfunc_call_test_release(struct prog_test_ref_kfunc *p) __ksym;
@@ -135,4 +140,8 @@ int bpf_kfunc_call_kernel_getsockname(struct addr_args *args) __ksym;
 int bpf_kfunc_call_kernel_getpeername(struct addr_args *args) __ksym;
 
 void bpf_kfunc_dynptr_test(struct bpf_dynptr *ptr, struct bpf_dynptr *ptr__nullable) __ksym;
+
+struct bpf_testmod_ctx *bpf_testmod_ctx_create(int *err) __ksym;
+void bpf_testmod_ctx_release(struct bpf_testmod_ctx *ctx) __ksym;
+
 #endif /* _BPF_TESTMOD_KFUNC_H */
diff --git a/tools/testing/selftests/bpf/prog_tests/kfunc_call.c b/tools/testing/selftests/bpf/prog_tests/kfunc_call.c
index 2eb71559713c..5b743212292f 100644
--- a/tools/testing/selftests/bpf/prog_tests/kfunc_call.c
+++ b/tools/testing/selftests/bpf/prog_tests/kfunc_call.c
@@ -78,6 +78,7 @@ static struct kfunc_test_params kfunc_tests[] = {
 	SYSCALL_TEST(kfunc_syscall_test, 0),
 	SYSCALL_NULL_CTX_TEST(kfunc_syscall_test_null, 0),
 	TC_TEST(kfunc_call_test_static_unused_arg, 0),
+	TC_TEST(kfunc_call_ctx, 0),
 };
 
 struct syscall_test_args {
diff --git a/tools/testing/selftests/bpf/progs/kfunc_call_test.c b/tools/testing/selftests/bpf/progs/kfunc_call_test.c
index cf68d1e48a0f..f502f755f567 100644
--- a/tools/testing/selftests/bpf/progs/kfunc_call_test.c
+++ b/tools/testing/selftests/bpf/progs/kfunc_call_test.c
@@ -177,4 +177,41 @@ int kfunc_call_test_static_unused_arg(struct __sk_buff *skb)
 	return actual != expected ? -1 : 0;
 }
 
+struct ctx_val {
+	struct bpf_testmod_ctx __kptr *ctx;
+};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, int);
+	__type(value, struct ctx_val);
+} ctx_map SEC(".maps");
+
+SEC("tc")
+int kfunc_call_ctx(struct __sk_buff *skb)
+{
+	struct bpf_testmod_ctx *ctx;
+	int err = 0;
+
+	ctx = bpf_testmod_ctx_create(&err);
+	if (!ctx && !err)
+		err = -1;
+	if (ctx) {
+		int key = 0;
+		struct ctx_val *ctx_val = bpf_map_lookup_elem(&ctx_map, &key);
+
+		/* Transfer ctx to map to be freed via implicit dtor call
+		 * on cleanup.
+		 */
+		if (ctx_val)
+			ctx = bpf_kptr_xchg(&ctx_val->ctx, ctx);
+		if (ctx) {
+			bpf_testmod_ctx_release(ctx);
+			err = -1;
+		}
+	}
+	return err;
+}
+
 char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-58-ga151


From 185d72112b951404968cfaa9f77cd9ee19207205 Mon Sep 17 00:00:00 2001
From: Sean Anderson <sean.anderson@linux.dev>
Date: Thu, 20 Jun 2024 16:39:43 -0400
Subject: net: xilinx: axienet: Enable multicast by default

We support multicast addresses, so enable it by default.

Signed-off-by: Sean Anderson <sean.anderson@linux.dev>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
index 5f98daa5b341..e342f387c3dd 100644
--- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
+++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
@@ -2254,7 +2254,6 @@ static int axienet_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, ndev);
 
 	SET_NETDEV_DEV(ndev, &pdev->dev);
-	ndev->flags &= ~IFF_MULTICAST;  /* clear multicast */
 	ndev->features = NETIF_F_SG;
 	ndev->ethtool_ops = &axienet_ethtool_ops;
 
-- 
cgit v1.2.3-58-ga151


From 04efaebd72d1d3d9991841051fafc6b195f3676d Mon Sep 17 00:00:00 2001
From: Dave Thaler <dthaler1968@googlemail.com>
Date: Sun, 23 Jun 2024 08:04:53 -0700
Subject: bpf, docs: Address comments from IETF Area Directors

This patch does the following to address IETF feedback:

* Remove mention of "program type" and reference future
  docs (and mention platform-specific docs exist) for
  helper functions and BTF. Addresses Roman Danyliw's
  comments based on GENART review from Ines Robles [0].

* Add reference for endianness as requested by John
  Scudder [1].

* Added bit numbers to top of 32-bit wide format diagrams
  as requested by Paul Wouters [2].

* Added more text about why BPF doesn't stand for anything, based
  on text from ebpf.io [3], as requested by Eric Vyncke and
  Gunter Van de Velde [4].

* Replaced "htobe16" (and similar) and the direction-specific
  description with just "be16" (and similar) and a direction-agnostic
  description, to match the direction-agnostic description in
  the Byteswap Instructions section. Based on feedback from Eric
  Vyncke [5].

[0] https://mailarchive.ietf.org/arch/msg/bpf/DvDgDWOiwk05OyNlWlAmELZFPlM/

[1] https://mailarchive.ietf.org/arch/msg/bpf/eKNXpU4jCLjsbZDSw8LjI29M3tM/

[2] https://mailarchive.ietf.org/arch/msg/bpf/hGk8HkYxeZTpdu9qW_MvbGKj7WU/

[3] https://ebpf.io/what-is-ebpf/#what-do-ebpf-and-bpf-stand-for

[4] https://mailarchive.ietf.org/arch/msg/bpf/i93lzdN3ewnzzS_JMbinCIYxAIU/

[5] https://mailarchive.ietf.org/arch/msg/bpf/KBWXbMeDcSrq4vsKR_KkBbV6hI4/

Acked-by: David Vernet <void@manifault.com>
Signed-off-by: Dave Thaler <dthaler1968@googlemail.com>
Link: https://lore.kernel.org/r/20240623150453.10613-1-dthaler1968@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 .../bpf/standardization/instruction-set.rst        | 80 ++++++++++++----------
 1 file changed, 45 insertions(+), 35 deletions(-)

diff --git a/Documentation/bpf/standardization/instruction-set.rst b/Documentation/bpf/standardization/instruction-set.rst
index 8d19810504b8..ab820d565052 100644
--- a/Documentation/bpf/standardization/instruction-set.rst
+++ b/Documentation/bpf/standardization/instruction-set.rst
@@ -5,12 +5,19 @@
 BPF Instruction Set Architecture (ISA)
 ======================================
 
-eBPF (which is no longer an acronym for anything), also commonly
+eBPF, also commonly
 referred to as BPF, is a technology with origins in the Linux kernel
 that can run untrusted programs in a privileged context such as an
 operating system kernel. This document specifies the BPF instruction
 set architecture (ISA).
 
+As a historical note, BPF originally stood for Berkeley Packet Filter,
+but now that it can do so much more than packet filtering, the acronym
+no longer makes sense. BPF is now considered a standalone term that
+does not stand for anything.  The original BPF is sometimes referred to
+as cBPF (classic BPF) to distinguish it from the now widely deployed
+eBPF (extended BPF).
+
 Documentation conventions
 =========================
 
@@ -18,7 +25,7 @@ The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
 "OPTIONAL" in this document are to be interpreted as described in
 BCP 14 `<https://www.rfc-editor.org/info/rfc2119>`_
-`RFC8174 <https://www.rfc-editor.org/info/rfc8174>`_
+`<https://www.rfc-editor.org/info/rfc8174>`_
 when, and only when, they appear in all capitals, as shown here.
 
 For brevity and consistency, this document refers to families
@@ -59,24 +66,18 @@ numbers.
 
 Functions
 ---------
-* htobe16: Takes an unsigned 16-bit number in host-endian format and
-  returns the equivalent number as an unsigned 16-bit number in big-endian
-  format.
-* htobe32: Takes an unsigned 32-bit number in host-endian format and
-  returns the equivalent number as an unsigned 32-bit number in big-endian
-  format.
-* htobe64: Takes an unsigned 64-bit number in host-endian format and
-  returns the equivalent number as an unsigned 64-bit number in big-endian
-  format.
-* htole16: Takes an unsigned 16-bit number in host-endian format and
-  returns the equivalent number as an unsigned 16-bit number in little-endian
-  format.
-* htole32: Takes an unsigned 32-bit number in host-endian format and
-  returns the equivalent number as an unsigned 32-bit number in little-endian
-  format.
-* htole64: Takes an unsigned 64-bit number in host-endian format and
-  returns the equivalent number as an unsigned 64-bit number in little-endian
-  format.
+
+The following byteswap functions are direction-agnostic.  That is,
+the same function is used for conversion in either direction discussed
+below.
+
+* be16: Takes an unsigned 16-bit number and converts it between
+  host byte order and big-endian
+  (`IEN137 <https://www.rfc-editor.org/ien/ien137.txt>`_) byte order.
+* be32: Takes an unsigned 32-bit number and converts it between
+  host byte order and big-endian byte order.
+* be64: Takes an unsigned 64-bit number and converts it between
+  host byte order and big-endian byte order.
 * bswap16: Takes an unsigned 16-bit number in either big- or little-endian
   format and returns the equivalent number with the same bit width but
   opposite endianness.
@@ -86,7 +87,12 @@ Functions
 * bswap64: Takes an unsigned 64-bit number in either big- or little-endian
   format and returns the equivalent number with the same bit width but
   opposite endianness.
-
+* le16: Takes an unsigned 16-bit number and converts it between
+  host byte order and little-endian byte order.
+* le32: Takes an unsigned 32-bit number and converts it between
+  host byte order and little-endian byte order.
+* le64: Takes an unsigned 64-bit number and converts it between
+  host byte order and little-endian byte order.
 
 Definitions
 -----------
@@ -437,8 +443,8 @@ and MUST be set to 0.
   =====  ========  =====  =================================================
   class  source    value  description
   =====  ========  =====  =================================================
-  ALU    TO_LE     0      convert between host byte order and little endian
-  ALU    TO_BE     1      convert between host byte order and big endian
+  ALU    LE        0      convert between host byte order and little endian
+  ALU    BE        1      convert between host byte order and big endian
   ALU64  Reserved  0      do byte swap unconditionally
   =====  ========  =====  =================================================
 
@@ -449,19 +455,19 @@ conformance group.
 
 Examples:
 
-``{END, TO_LE, ALU}`` with 'imm' = 16/32/64 means::
+``{END, LE, ALU}`` with 'imm' = 16/32/64 means::
 
-  dst = htole16(dst)
-  dst = htole32(dst)
-  dst = htole64(dst)
+  dst = le16(dst)
+  dst = le32(dst)
+  dst = le64(dst)
 
-``{END, TO_BE, ALU}`` with 'imm' = 16/32/64 means::
+``{END, BE, ALU}`` with 'imm' = 16/32/64 means::
 
-  dst = htobe16(dst)
-  dst = htobe32(dst)
-  dst = htobe64(dst)
+  dst = be16(dst)
+  dst = be32(dst)
+  dst = be64(dst)
 
-``{END, TO_LE, ALU64}`` with 'imm' = 16/32/64 means::
+``{END, TO, ALU64}`` with 'imm' = 16/32/64 means::
 
   dst = bswap16(dst)
   dst = bswap32(dst)
@@ -541,13 +547,17 @@ Helper functions are a concept whereby BPF programs can call into a
 set of function calls exposed by the underlying platform.
 
 Historically, each helper function was identified by a static ID
-encoded in the 'imm' field.  The available helper functions may differ
-for each program type, but static IDs are unique across all program types.
+encoded in the 'imm' field.  Further documentation of helper functions
+is outside the scope of this document and standardization is left for
+future work, but use is widely deployed and more information can be
+found in platform-specific documentation (e.g., Linux kernel documentation).
 
 Platforms that support the BPF Type Format (BTF) support identifying
 a helper function by a BTF ID encoded in the 'imm' field, where the BTF ID
 identifies the helper name and type.  Further documentation of BTF
-is outside the scope of this document and is left for future work.
+is outside the scope of this document and standardization is left for
+future work, but use is widely deployed and more information can be
+found in platform-specific documentation (e.g., Linux kernel documentation).
 
 Program-local functions
 ~~~~~~~~~~~~~~~~~~~~~~~
-- 
cgit v1.2.3-58-ga151


From 5a532459aa919d055d822d8db4ea2c5c8d511568 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Sun, 23 Jun 2024 14:52:24 +0100
Subject: bpf: fix build when CONFIG_DEBUG_INFO_BTF[_MODULES] is undefined

Kernel test robot reports that kernel build fails with
resilient split BTF changes.

Examining the associated config and code we see that
btf_relocate_id() is defined under CONFIG_DEBUG_INFO_BTF_MODULES.
Moving it outside the #ifdef solves the issue.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202406221742.d2srFLVI-lkp@intel.com/
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Link: https://lore.kernel.org/r/20240623135224.27981-1-alan.maguire@oracle.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/btf.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 8e12cb80ba73..4ff11779699e 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -6185,8 +6185,6 @@ err_out:
 	return btf;
 }
 
-#ifdef CONFIG_DEBUG_INFO_BTF_MODULES
-
 /* If .BTF_ids section was created with distilled base BTF, both base and
  * split BTF ids will need to be mapped to actual base/split ids for
  * BTF now that it has been relocated.
@@ -6198,6 +6196,8 @@ static __u32 btf_relocate_id(const struct btf *btf, __u32 id)
 	return btf->base_id_map[id];
 }
 
+#ifdef CONFIG_DEBUG_INFO_BTF_MODULES
+
 static struct btf *btf_parse_module(const char *module_name, const void *data,
 				    unsigned int data_size, void *base_data,
 				    unsigned int base_data_size)
-- 
cgit v1.2.3-58-ga151


From 568ebdaba6370c03360860f1524f646ddd5ca523 Mon Sep 17 00:00:00 2001
From: Lukas Bulwahn <lukas.bulwahn@redhat.com>
Date: Mon, 24 Jun 2024 09:03:26 +0200
Subject: MAINTAINERS: adjust file entry in FREESCALE QORIQ DPAA FMAN DRIVER

Commit 243996d172a6 ("dt-bindings: net: Convert fsl-fman to yaml") splits
the previous dt text file into four yaml files. It adjusts a corresponding
file entry in MAINTAINERS from txt to yaml, but this adjustment misses
that the file was split and renamed.

Hence, ./scripts/get_maintainer.pl --self-test=patterns complains about a
broken reference.

Adjust the file entry to match the four yaml files resulting from this
commit above.

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 0e3cb040fc16..9283dd04cc6e 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -8810,7 +8810,7 @@ M:	Madalin Bucur <madalin.bucur@nxp.com>
 R:	Sean Anderson <sean.anderson@seco.com>
 L:	netdev@vger.kernel.org
 S:	Maintained
-F:	Documentation/devicetree/bindings/net/fsl-fman.yaml
+F:	Documentation/devicetree/bindings/net/fsl,fman*.yaml
 F:	drivers/net/ethernet/freescale/fman
 
 FREESCALE QORIQ PTP CLOCK DRIVER
-- 
cgit v1.2.3-58-ga151


From c73a9683cb21012b6c0f14217974837151c527a8 Mon Sep 17 00:00:00 2001
From: Antoine Tenart <atenart@kernel.org>
Date: Mon, 24 Jun 2024 11:09:07 +0200
Subject: libbpf: Skip base btf sanity checks

When upgrading to libbpf 1.3 we noticed a big performance hit while
loading programs using CORE on non base-BTF symbols. This was tracked
down to the new BTF sanity check logic. The issue is the base BTF
definitions are checked first for the base BTF and then again for every
module BTF.

Loading 5 dummy programs (using libbpf-rs) that are using CORE on a
non-base BTF symbol on my system:
- Before this fix: 3s.
- With this fix: 0.1s.

Fix this by only checking the types starting at the BTF start id. This
should ensure the base BTF is still checked as expected but only once
(btf->start_id == 1 when creating the base BTF), and then only
additional types are checked for each module BTF.

Fixes: 3903802bb99a ("libbpf: Add basic BTF sanity validation")
Signed-off-by: Antoine Tenart <atenart@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Alan Maguire <alan.maguire@oracle.com>
Link: https://lore.kernel.org/bpf/20240624090908.171231-1-atenart@kernel.org
---
 tools/lib/bpf/btf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 0c0f60cad769..cd5dd6619214 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -601,7 +601,7 @@ static int btf_sanity_check(const struct btf *btf)
 	__u32 i, n = btf__type_cnt(btf);
 	int err;
 
-	for (i = 1; i < n; i++) {
+	for (i = btf->start_id; i < n; i++) {
 		t = btf_type_by_id(btf, i);
 		err = btf_validate_type(btf, t, i);
 		if (err)
-- 
cgit v1.2.3-58-ga151


From 3216b7bcd7197ca5da204d3eea81d308cae2a849 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Wed, 19 Jun 2024 17:21:11 +0300
Subject: wifi: ath12k: add ATH12K_DBG_WOW log level

Currently there is no dedicated log level for WoW, so create it for future use.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240604055407.12506-2-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/debug.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/debug.h b/drivers/net/wireless/ath/ath12k/debug.h
index aa685295f8a4..f7005917362c 100644
--- a/drivers/net/wireless/ath/ath12k/debug.h
+++ b/drivers/net/wireless/ath/ath12k/debug.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: BSD-3-Clause-Clear */
 /*
  * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved.
- * Copyright (c) 2021-2022 Qualcomm Innovation Center, Inc. All rights reserved.
+ * Copyright (c) 2021-2022, 2024 Qualcomm Innovation Center, Inc. All rights reserved.
  */
 
 #ifndef _ATH12K_DEBUG_H_
@@ -25,6 +25,7 @@ enum ath12k_debug_mask {
 	ATH12K_DBG_PCI		= 0x00001000,
 	ATH12K_DBG_DP_TX	= 0x00002000,
 	ATH12K_DBG_DP_RX	= 0x00004000,
+	ATH12K_DBG_WOW		= 0x00008000,
 	ATH12K_DBG_ANY		= 0xffffffff,
 };
 
-- 
cgit v1.2.3-58-ga151


From 5931741709195da9abb0fb0b8c0fed5d0898b275 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Wed, 19 Jun 2024 17:21:11 +0300
Subject: wifi: ath12k: implement WoW enable and wakeup commands

Implement WoW enable and WoW wakeup commands which are needed
for suspend/resume.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240604055407.12506-3-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/Makefile |   1 +
 drivers/net/wireless/ath/ath12k/core.c   |   2 +
 drivers/net/wireless/ath/ath12k/core.h   |   4 +
 drivers/net/wireless/ath/ath12k/wmi.c    | 105 +++++++++++++++++++
 drivers/net/wireless/ath/ath12k/wmi.h    | 174 +++++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/wow.c    |  79 ++++++++++++++
 drivers/net/wireless/ath/ath12k/wow.h    |  30 ++++++
 7 files changed, 395 insertions(+)
 create mode 100644 drivers/net/wireless/ath/ath12k/wow.c
 create mode 100644 drivers/net/wireless/ath/ath12k/wow.h

diff --git a/drivers/net/wireless/ath/ath12k/Makefile b/drivers/net/wireless/ath/ath12k/Makefile
index d42480db7463..f0dcc933aa95 100644
--- a/drivers/net/wireless/ath/ath12k/Makefile
+++ b/drivers/net/wireless/ath/ath12k/Makefile
@@ -26,6 +26,7 @@ ath12k-y += core.o \
 ath12k-$(CONFIG_ATH12K_DEBUGFS) += debugfs.o
 ath12k-$(CONFIG_ACPI) += acpi.o
 ath12k-$(CONFIG_ATH12K_TRACING) += trace.o
+ath12k-$(CONFIG_PM) += wow.o
 
 # for tracing framework to find trace.h
 CFLAGS_trace.o := -I$(src)
diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c
index 197099f15e3f..7603516500ba 100644
--- a/drivers/net/wireless/ath/ath12k/core.c
+++ b/drivers/net/wireless/ath/ath12k/core.c
@@ -16,6 +16,7 @@
 #include "hif.h"
 #include "fw.h"
 #include "debugfs.h"
+#include "wow.h"
 
 unsigned int ath12k_debug_mask;
 module_param_named(debug_mask, ath12k_debug_mask, uint, 0644);
@@ -1285,6 +1286,7 @@ struct ath12k_base *ath12k_core_alloc(struct device *dev, size_t priv_size,
 	timer_setup(&ab->rx_replenish_retry, ath12k_ce_rx_replenish_retry, 0);
 	init_completion(&ab->htc_suspend);
 	init_completion(&ab->restart_completed);
+	init_completion(&ab->wow.wakeup_completed);
 
 	ab->dev = dev;
 	ab->hif.bus = bus;
diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index d7a3e8fddb20..10aa3aca7250 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -765,6 +765,10 @@ struct ath12k_base {
 		const struct ath12k_hif_ops *ops;
 	} hif;
 
+	struct {
+		struct completion wakeup_completed;
+	} wow;
+
 	struct ath12k_ce ce;
 	struct timer_list rx_replenish_retry;
 	struct ath12k_hal hal;
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index d6e1d1398cdb..3eafa46d737b 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -7030,6 +7030,66 @@ exit:
 	kfree(tb);
 }
 
+static int ath12k_wmi_wow_wakeup_host_parse(struct ath12k_base *ab,
+					    u16 tag, u16 len,
+					    const void *ptr, void *data)
+{
+	const struct wmi_wow_ev_pg_fault_param *pf_param;
+	const struct wmi_wow_ev_param *param;
+	struct wmi_wow_ev_arg *arg = data;
+	int pf_len;
+
+	switch (tag) {
+	case WMI_TAG_WOW_EVENT_INFO:
+		param = ptr;
+		arg->wake_reason = le32_to_cpu(param->wake_reason);
+		ath12k_dbg(ab, ATH12K_DBG_WMI, "wow wakeup host reason %d %s\n",
+			   arg->wake_reason, wow_reason(arg->wake_reason));
+		break;
+
+	case WMI_TAG_ARRAY_BYTE:
+		if (arg && arg->wake_reason == WOW_REASON_PAGE_FAULT) {
+			pf_param = ptr;
+			pf_len = le32_to_cpu(pf_param->len);
+			if (pf_len > len - sizeof(pf_len) ||
+			    pf_len < 0) {
+				ath12k_warn(ab, "invalid wo reason page fault buffer len %d\n",
+					    pf_len);
+				return -EINVAL;
+			}
+			ath12k_dbg(ab, ATH12K_DBG_WMI, "wow_reason_page_fault len %d\n",
+				   pf_len);
+			ath12k_dbg_dump(ab, ATH12K_DBG_WMI,
+					"wow_reason_page_fault packet present",
+					"wow_pg_fault ",
+					pf_param->data,
+					pf_len);
+		}
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+static void ath12k_wmi_event_wow_wakeup_host(struct ath12k_base *ab, struct sk_buff *skb)
+{
+	struct wmi_wow_ev_arg arg = { };
+	int ret;
+
+	ret = ath12k_wmi_tlv_iter(ab, skb->data, skb->len,
+				  ath12k_wmi_wow_wakeup_host_parse,
+				  &arg);
+	if (ret) {
+		ath12k_warn(ab, "failed to parse wmi wow wakeup host event tlv: %d\n",
+			    ret);
+		return;
+	}
+
+	complete(&ab->wow.wakeup_completed);
+}
+
 static void ath12k_wmi_op_rx(struct ath12k_base *ab, struct sk_buff *skb)
 {
 	struct wmi_cmd_hdr *cmd_hdr;
@@ -7150,6 +7210,9 @@ static void ath12k_wmi_op_rx(struct ath12k_base *ab, struct sk_buff *skb)
 	case WMI_DIAG_EVENTID:
 		ath12k_wmi_diag_event(ab, skb);
 		break;
+	case WMI_WOW_WAKEUP_HOST_EVENTID:
+		ath12k_wmi_event_wow_wakeup_host(ab, skb);
+		break;
 	/* TODO: Add remaining events */
 	default:
 		ath12k_dbg(ab, ATH12K_DBG_WMI, "Unknown eventid: 0x%x\n", id);
@@ -7359,3 +7422,45 @@ void ath12k_wmi_detach(struct ath12k_base *ab)
 
 	ath12k_wmi_free_dbring_caps(ab);
 }
+
+int ath12k_wmi_wow_host_wakeup_ind(struct ath12k *ar)
+{
+	struct wmi_wow_host_wakeup_cmd *cmd;
+	struct sk_buff *skb;
+	size_t len;
+
+	len = sizeof(*cmd);
+	skb = ath12k_wmi_alloc_skb(ar->wmi->wmi_ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_wow_host_wakeup_cmd *)skb->data;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_WOW_HOSTWAKEUP_FROM_SLEEP_CMD,
+						 sizeof(*cmd));
+
+	ath12k_dbg(ar->ab, ATH12K_DBG_WMI, "wmi tlv wow host wakeup ind\n");
+
+	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_WOW_HOSTWAKEUP_FROM_SLEEP_CMDID);
+}
+
+int ath12k_wmi_wow_enable(struct ath12k *ar)
+{
+	struct wmi_wow_enable_cmd *cmd;
+	struct sk_buff *skb;
+	int len;
+
+	len = sizeof(*cmd);
+	skb = ath12k_wmi_alloc_skb(ar->wmi->wmi_ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_wow_enable_cmd *)skb->data;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_WOW_ENABLE_CMD,
+						 sizeof(*cmd));
+
+	cmd->enable = cpu_to_le32(1);
+	cmd->pause_iface_config = cpu_to_le32(WOW_IFACE_PAUSE_ENABLED);
+	ath12k_dbg(ar->ab, ATH12K_DBG_WMI, "wmi tlv wow enable\n");
+
+	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_WOW_ENABLE_CMDID);
+}
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index c2b86e187a03..9cee3a114645 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -4902,6 +4902,178 @@ struct wmi_twt_disable_event {
 	__le32 status;
 } __packed;
 
+/* WOW structures */
+enum wmi_wow_wakeup_event {
+	WOW_BMISS_EVENT = 0,
+	WOW_BETTER_AP_EVENT,
+	WOW_DEAUTH_RECVD_EVENT,
+	WOW_MAGIC_PKT_RECVD_EVENT,
+	WOW_GTK_ERR_EVENT,
+	WOW_FOURWAY_HSHAKE_EVENT,
+	WOW_EAPOL_RECVD_EVENT,
+	WOW_NLO_DETECTED_EVENT,
+	WOW_DISASSOC_RECVD_EVENT,
+	WOW_PATTERN_MATCH_EVENT,
+	WOW_CSA_IE_EVENT,
+	WOW_PROBE_REQ_WPS_IE_EVENT,
+	WOW_AUTH_REQ_EVENT,
+	WOW_ASSOC_REQ_EVENT,
+	WOW_HTT_EVENT,
+	WOW_RA_MATCH_EVENT,
+	WOW_HOST_AUTO_SHUTDOWN_EVENT,
+	WOW_IOAC_MAGIC_EVENT,
+	WOW_IOAC_SHORT_EVENT,
+	WOW_IOAC_EXTEND_EVENT,
+	WOW_IOAC_TIMER_EVENT,
+	WOW_DFS_PHYERR_RADAR_EVENT,
+	WOW_BEACON_EVENT,
+	WOW_CLIENT_KICKOUT_EVENT,
+	WOW_EVENT_MAX,
+};
+
+enum wmi_wow_interface_cfg {
+	WOW_IFACE_PAUSE_ENABLED,
+	WOW_IFACE_PAUSE_DISABLED
+};
+
+#define C2S(x) case x: return #x
+
+static inline const char *wow_wakeup_event(enum wmi_wow_wakeup_event ev)
+{
+	switch (ev) {
+	C2S(WOW_BMISS_EVENT);
+	C2S(WOW_BETTER_AP_EVENT);
+	C2S(WOW_DEAUTH_RECVD_EVENT);
+	C2S(WOW_MAGIC_PKT_RECVD_EVENT);
+	C2S(WOW_GTK_ERR_EVENT);
+	C2S(WOW_FOURWAY_HSHAKE_EVENT);
+	C2S(WOW_EAPOL_RECVD_EVENT);
+	C2S(WOW_NLO_DETECTED_EVENT);
+	C2S(WOW_DISASSOC_RECVD_EVENT);
+	C2S(WOW_PATTERN_MATCH_EVENT);
+	C2S(WOW_CSA_IE_EVENT);
+	C2S(WOW_PROBE_REQ_WPS_IE_EVENT);
+	C2S(WOW_AUTH_REQ_EVENT);
+	C2S(WOW_ASSOC_REQ_EVENT);
+	C2S(WOW_HTT_EVENT);
+	C2S(WOW_RA_MATCH_EVENT);
+	C2S(WOW_HOST_AUTO_SHUTDOWN_EVENT);
+	C2S(WOW_IOAC_MAGIC_EVENT);
+	C2S(WOW_IOAC_SHORT_EVENT);
+	C2S(WOW_IOAC_EXTEND_EVENT);
+	C2S(WOW_IOAC_TIMER_EVENT);
+	C2S(WOW_DFS_PHYERR_RADAR_EVENT);
+	C2S(WOW_BEACON_EVENT);
+	C2S(WOW_CLIENT_KICKOUT_EVENT);
+	C2S(WOW_EVENT_MAX);
+	default:
+		return NULL;
+	}
+}
+
+enum wmi_wow_wake_reason {
+	WOW_REASON_UNSPECIFIED = -1,
+	WOW_REASON_NLOD = 0,
+	WOW_REASON_AP_ASSOC_LOST,
+	WOW_REASON_LOW_RSSI,
+	WOW_REASON_DEAUTH_RECVD,
+	WOW_REASON_DISASSOC_RECVD,
+	WOW_REASON_GTK_HS_ERR,
+	WOW_REASON_EAP_REQ,
+	WOW_REASON_FOURWAY_HS_RECV,
+	WOW_REASON_TIMER_INTR_RECV,
+	WOW_REASON_PATTERN_MATCH_FOUND,
+	WOW_REASON_RECV_MAGIC_PATTERN,
+	WOW_REASON_P2P_DISC,
+	WOW_REASON_WLAN_HB,
+	WOW_REASON_CSA_EVENT,
+	WOW_REASON_PROBE_REQ_WPS_IE_RECV,
+	WOW_REASON_AUTH_REQ_RECV,
+	WOW_REASON_ASSOC_REQ_RECV,
+	WOW_REASON_HTT_EVENT,
+	WOW_REASON_RA_MATCH,
+	WOW_REASON_HOST_AUTO_SHUTDOWN,
+	WOW_REASON_IOAC_MAGIC_EVENT,
+	WOW_REASON_IOAC_SHORT_EVENT,
+	WOW_REASON_IOAC_EXTEND_EVENT,
+	WOW_REASON_IOAC_TIMER_EVENT,
+	WOW_REASON_ROAM_HO,
+	WOW_REASON_DFS_PHYERR_RADADR_EVENT,
+	WOW_REASON_BEACON_RECV,
+	WOW_REASON_CLIENT_KICKOUT_EVENT,
+	WOW_REASON_PAGE_FAULT = 0x3a,
+	WOW_REASON_DEBUG_TEST = 0xFF,
+};
+
+static inline const char *wow_reason(enum wmi_wow_wake_reason reason)
+{
+	switch (reason) {
+	C2S(WOW_REASON_UNSPECIFIED);
+	C2S(WOW_REASON_NLOD);
+	C2S(WOW_REASON_AP_ASSOC_LOST);
+	C2S(WOW_REASON_LOW_RSSI);
+	C2S(WOW_REASON_DEAUTH_RECVD);
+	C2S(WOW_REASON_DISASSOC_RECVD);
+	C2S(WOW_REASON_GTK_HS_ERR);
+	C2S(WOW_REASON_EAP_REQ);
+	C2S(WOW_REASON_FOURWAY_HS_RECV);
+	C2S(WOW_REASON_TIMER_INTR_RECV);
+	C2S(WOW_REASON_PATTERN_MATCH_FOUND);
+	C2S(WOW_REASON_RECV_MAGIC_PATTERN);
+	C2S(WOW_REASON_P2P_DISC);
+	C2S(WOW_REASON_WLAN_HB);
+	C2S(WOW_REASON_CSA_EVENT);
+	C2S(WOW_REASON_PROBE_REQ_WPS_IE_RECV);
+	C2S(WOW_REASON_AUTH_REQ_RECV);
+	C2S(WOW_REASON_ASSOC_REQ_RECV);
+	C2S(WOW_REASON_HTT_EVENT);
+	C2S(WOW_REASON_RA_MATCH);
+	C2S(WOW_REASON_HOST_AUTO_SHUTDOWN);
+	C2S(WOW_REASON_IOAC_MAGIC_EVENT);
+	C2S(WOW_REASON_IOAC_SHORT_EVENT);
+	C2S(WOW_REASON_IOAC_EXTEND_EVENT);
+	C2S(WOW_REASON_IOAC_TIMER_EVENT);
+	C2S(WOW_REASON_ROAM_HO);
+	C2S(WOW_REASON_DFS_PHYERR_RADADR_EVENT);
+	C2S(WOW_REASON_BEACON_RECV);
+	C2S(WOW_REASON_CLIENT_KICKOUT_EVENT);
+	C2S(WOW_REASON_PAGE_FAULT);
+	C2S(WOW_REASON_DEBUG_TEST);
+	default:
+		return NULL;
+	}
+}
+
+#undef C2S
+
+struct wmi_wow_enable_cmd {
+	__le32 tlv_header;
+	__le32 enable;
+	__le32 pause_iface_config;
+	__le32 flags;
+}  __packed;
+
+struct wmi_wow_host_wakeup_cmd {
+	__le32 tlv_header;
+	__le32 reserved;
+} __packed;
+
+struct wmi_wow_ev_param {
+	__le32 vdev_id;
+	__le32 flag;
+	__le32 wake_reason;
+	__le32 data_len;
+} __packed;
+
+struct wmi_wow_ev_pg_fault_param {
+	__le32 len;
+	u8 data[];
+} __packed;
+
+struct wmi_wow_ev_arg {
+	enum wmi_wow_wake_reason wake_reason;
+};
+
 void ath12k_wmi_init_qcn9274(struct ath12k_base *ab,
 			     struct ath12k_wmi_resource_config_arg *config);
 void ath12k_wmi_init_wcn7850(struct ath12k_base *ab,
@@ -5054,4 +5226,6 @@ ath12k_wmi_mac_phy_get_hw_link_id(const struct ath12k_wmi_mac_phy_caps_params *p
 			     WMI_CAPS_PARAMS_HW_LINK_ID);
 }
 
+int ath12k_wmi_wow_host_wakeup_ind(struct ath12k *ar);
+int ath12k_wmi_wow_enable(struct ath12k *ar);
 #endif
diff --git a/drivers/net/wireless/ath/ath12k/wow.c b/drivers/net/wireless/ath/ath12k/wow.c
new file mode 100644
index 000000000000..d6d3ed7332ab
--- /dev/null
+++ b/drivers/net/wireless/ath/ath12k/wow.c
@@ -0,0 +1,79 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2020 The Linux Foundation. All rights reserved.
+ * Copyright (c) 2022-2024 Qualcomm Innovation Center, Inc. All rights reserved.
+ */
+
+#include <linux/delay.h>
+
+#include "mac.h"
+#include "core.h"
+#include "hif.h"
+#include "debug.h"
+#include "wmi.h"
+#include "wow.h"
+
+int ath12k_wow_enable(struct ath12k *ar)
+{
+	struct ath12k_base *ab = ar->ab;
+	int i, ret;
+
+	clear_bit(ATH12K_FLAG_HTC_SUSPEND_COMPLETE, &ab->dev_flags);
+
+	/* The firmware might be busy and it can not enter WoW immediately.
+	 * In that case firmware notifies host with
+	 * ATH12K_HTC_MSG_NACK_SUSPEND message, asking host to try again
+	 * later. Per the firmware team there could be up to 10 loops.
+	 */
+	for (i = 0; i < ATH12K_WOW_RETRY_NUM; i++) {
+		reinit_completion(&ab->htc_suspend);
+
+		ret = ath12k_wmi_wow_enable(ar);
+		if (ret) {
+			ath12k_warn(ab, "failed to issue wow enable: %d\n", ret);
+			return ret;
+		}
+
+		ret = wait_for_completion_timeout(&ab->htc_suspend, 3 * HZ);
+		if (ret == 0) {
+			ath12k_warn(ab,
+				    "timed out while waiting for htc suspend completion\n");
+			return -ETIMEDOUT;
+		}
+
+		if (test_bit(ATH12K_FLAG_HTC_SUSPEND_COMPLETE, &ab->dev_flags))
+			/* success, suspend complete received */
+			return 0;
+
+		ath12k_warn(ab, "htc suspend not complete, retrying (try %d)\n",
+			    i);
+		msleep(ATH12K_WOW_RETRY_WAIT_MS);
+	}
+
+	ath12k_warn(ab, "htc suspend not complete, failing after %d tries\n", i);
+
+	return -ETIMEDOUT;
+}
+
+int ath12k_wow_wakeup(struct ath12k *ar)
+{
+	struct ath12k_base *ab = ar->ab;
+	int ret;
+
+	reinit_completion(&ab->wow.wakeup_completed);
+
+	ret = ath12k_wmi_wow_host_wakeup_ind(ar);
+	if (ret) {
+		ath12k_warn(ab, "failed to send wow wakeup indication: %d\n",
+			    ret);
+		return ret;
+	}
+
+	ret = wait_for_completion_timeout(&ab->wow.wakeup_completed, 3 * HZ);
+	if (ret == 0) {
+		ath12k_warn(ab, "timed out while waiting for wow wakeup completion\n");
+		return -ETIMEDOUT;
+	}
+
+	return 0;
+}
diff --git a/drivers/net/wireless/ath/ath12k/wow.h b/drivers/net/wireless/ath/ath12k/wow.h
new file mode 100644
index 000000000000..6817ceaa4cfe
--- /dev/null
+++ b/drivers/net/wireless/ath/ath12k/wow.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2020 The Linux Foundation. All rights reserved.
+ * Copyright (c) 2022, 2024 Qualcomm Innovation Center, Inc. All rights reserved.
+ */
+
+#ifndef ATH12K_WOW_H
+#define ATH12K_WOW_H
+
+#define ATH12K_WOW_RETRY_NUM		10
+#define ATH12K_WOW_RETRY_WAIT_MS	200
+
+#ifdef CONFIG_PM
+
+int ath12k_wow_enable(struct ath12k *ar);
+int ath12k_wow_wakeup(struct ath12k *ar);
+
+#else
+
+static inline int ath12k_wow_enable(struct ath12k *ar)
+{
+	return 0;
+}
+
+static inline int ath12k_wow_wakeup(struct ath12k *ar)
+{
+	return 0;
+}
+#endif /* CONFIG_PM */
+#endif /* ATH12K_WOW_H */
-- 
cgit v1.2.3-58-ga151


From 4a3c212eee0e36c1cd55bfbfb94041a0d3ad265b Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Wed, 19 Jun 2024 17:21:11 +0300
Subject: wifi: ath12k: add basic WoW functionalities

Implement basic WoW functionalities such as magic-packet, disconnect
and pattern. The logic is very similar to ath11k.

When WoW is configured, ath12k_core_suspend and ath12k_core_resume
are skipped (by checking ar->state) as we are not allowed to power
cycle firmware.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240604055407.12506-4-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.c |  46 +++-
 drivers/net/wireless/ath/ath12k/core.h |   5 +
 drivers/net/wireless/ath/ath12k/htc.c  |   6 +
 drivers/net/wireless/ath/ath12k/mac.c  |  14 ++
 drivers/net/wireless/ath/ath12k/wmi.c  | 143 ++++++++++++
 drivers/net/wireless/ath/ath12k/wmi.h  |  63 +++++
 drivers/net/wireless/ath/ath12k/wow.c  | 416 +++++++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/wow.h  |  32 +++
 8 files changed, 717 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.c b/drivers/net/wireless/ath/ath12k/core.c
index 7603516500ba..51252e8bc1ae 100644
--- a/drivers/net/wireless/ath/ath12k/core.c
+++ b/drivers/net/wireless/ath/ath12k/core.c
@@ -43,14 +43,38 @@ static int ath12k_core_rfkill_config(struct ath12k_base *ab)
 	return ret;
 }
 
-int ath12k_core_suspend(struct ath12k_base *ab)
+/* Check if we need to continue with suspend/resume operation.
+ * Return:
+ *	a negative value: error happens and don't continue.
+ *	0:  no error but don't continue.
+ *	positive value: no error and do continue.
+ */
+static int ath12k_core_continue_suspend_resume(struct ath12k_base *ab)
 {
 	struct ath12k *ar;
-	int ret, i;
 
 	if (!ab->hw_params->supports_suspend)
 		return -EOPNOTSUPP;
 
+	/* so far single_pdev_only chips have supports_suspend as true
+	 * so pass 0 as a dummy pdev_id here.
+	 */
+	ar = ab->pdevs[0].ar;
+	if (!ar || !ar->ah || ar->ah->state != ATH12K_HW_STATE_OFF)
+		return 0;
+
+	return 1;
+}
+
+int ath12k_core_suspend(struct ath12k_base *ab)
+{
+	struct ath12k *ar;
+	int ret, i;
+
+	ret = ath12k_core_continue_suspend_resume(ab);
+	if (ret <= 0)
+		return ret;
+
 	for (i = 0; i < ab->num_radios; i++) {
 		ar = ab->pdevs[i].ar;
 		if (!ar)
@@ -81,8 +105,11 @@ EXPORT_SYMBOL(ath12k_core_suspend);
 
 int ath12k_core_suspend_late(struct ath12k_base *ab)
 {
-	if (!ab->hw_params->supports_suspend)
-		return -EOPNOTSUPP;
+	int ret;
+
+	ret = ath12k_core_continue_suspend_resume(ab);
+	if (ret <= 0)
+		return ret;
 
 	ath12k_acpi_stop(ab);
 
@@ -99,8 +126,9 @@ int ath12k_core_resume_early(struct ath12k_base *ab)
 {
 	int ret;
 
-	if (!ab->hw_params->supports_suspend)
-		return -EOPNOTSUPP;
+	ret = ath12k_core_continue_suspend_resume(ab);
+	if (ret <= 0)
+		return ret;
 
 	reinit_completion(&ab->restart_completed);
 	ret = ath12k_hif_power_up(ab);
@@ -114,9 +142,11 @@ EXPORT_SYMBOL(ath12k_core_resume_early);
 int ath12k_core_resume(struct ath12k_base *ab)
 {
 	long time_left;
+	int ret;
 
-	if (!ab->hw_params->supports_suspend)
-		return -EOPNOTSUPP;
+	ret = ath12k_core_continue_suspend_resume(ab);
+	if (ret <= 0)
+		return ret;
 
 	time_left = wait_for_completion_timeout(&ab->restart_completed,
 						ATH12K_RESET_TIMEOUT_HZ);
diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index 10aa3aca7250..2fd08648cef3 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -28,6 +28,7 @@
 #include "dbring.h"
 #include "fw.h"
 #include "acpi.h"
+#include "wow.h"
 
 #define SM(_v, _f) (((_v) << _f##_LSB) & _f##_MASK)
 
@@ -607,6 +608,9 @@ struct ath12k {
 	struct work_struct wmi_mgmt_tx_work;
 	struct sk_buff_head wmi_mgmt_tx_queue;
 
+	struct ath12k_wow wow;
+	struct completion target_suspend;
+	bool target_suspend_ack;
 	struct ath12k_per_peer_tx_stats peer_tx_stats;
 	struct list_head ppdu_stats_info;
 	u32 ppdu_stat_list_depth;
@@ -767,6 +771,7 @@ struct ath12k_base {
 
 	struct {
 		struct completion wakeup_completed;
+		u32 wmi_conf_rx_decap_mode;
 	} wow;
 
 	struct ath12k_ce ce;
diff --git a/drivers/net/wireless/ath/ath12k/htc.c b/drivers/net/wireless/ath/ath12k/htc.c
index 2f2230f565bb..d13616bf07f4 100644
--- a/drivers/net/wireless/ath/ath12k/htc.c
+++ b/drivers/net/wireless/ath/ath12k/htc.c
@@ -244,6 +244,11 @@ static void ath12k_htc_suspend_complete(struct ath12k_base *ab, bool ack)
 	complete(&ab->htc_suspend);
 }
 
+static void ath12k_htc_wakeup_from_suspend(struct ath12k_base *ab)
+{
+	ath12k_dbg(ab, ATH12K_DBG_BOOT, "boot wakeup from suspend is received\n");
+}
+
 void ath12k_htc_rx_completion_handler(struct ath12k_base *ab,
 				      struct sk_buff *skb)
 {
@@ -349,6 +354,7 @@ void ath12k_htc_rx_completion_handler(struct ath12k_base *ab,
 			ath12k_htc_suspend_complete(ab, false);
 			break;
 		case ATH12K_HTC_MSG_WAKEUP_FROM_SUSPEND_ID:
+			ath12k_htc_wakeup_from_suspend(ab);
 			break;
 		default:
 			ath12k_warn(ab, "ignoring unsolicited htc ep0 event %u\n",
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 17b7fab03d50..59662ca241ca 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -15,6 +15,8 @@
 #include "dp_rx.h"
 #include "peer.h"
 #include "debugfs.h"
+#include "hif.h"
+#include "wow.h"
 
 #define CHAN2G(_channel, _freq, _flags) { \
 	.band                   = NL80211_BAND_2GHZ, \
@@ -8578,6 +8580,12 @@ static const struct ieee80211_ops ath12k_ops = {
 	.sta_statistics			= ath12k_mac_op_sta_statistics,
 	.remain_on_channel              = ath12k_mac_op_remain_on_channel,
 	.cancel_remain_on_channel       = ath12k_mac_op_cancel_remain_on_channel,
+
+#ifdef CONFIG_PM
+	.suspend			= ath12k_wow_op_suspend,
+	.resume				= ath12k_wow_op_resume,
+	.set_wakeup			= ath12k_wow_op_set_wakeup,
+#endif
 };
 
 static void ath12k_mac_update_ch_list(struct ath12k *ar,
@@ -9100,6 +9108,12 @@ static int ath12k_mac_hw_register(struct ath12k_hw *ah)
 		ieee80211_hw_set(hw, SUPPORT_FAST_XMIT);
 	}
 
+	ret = ath12k_wow_init(ar);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to init wow: %d\n", ret);
+		goto err_free_if_combs;
+	}
+
 	ret = ieee80211_register_hw(hw);
 	if (ret) {
 		ath12k_err(ab, "ieee80211 registration failed: %d\n", ret);
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 3eafa46d737b..f72ec153723c 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -3717,6 +3717,7 @@ int ath12k_wmi_cmd_init(struct ath12k_base *ab)
 		arg.res_cfg.is_reg_cc_ext_event_supported = true;
 
 	ab->hw_params->wmi_init(ab, &arg.res_cfg);
+	ab->wow.wmi_conf_rx_decap_mode = arg.res_cfg.rx_decap_mode;
 
 	arg.num_mem_chunks = wmi_ab->num_mem_chunks;
 	arg.hw_mode_id = wmi_ab->preferred_hw_mode;
@@ -7464,3 +7465,145 @@ int ath12k_wmi_wow_enable(struct ath12k *ar)
 
 	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_WOW_ENABLE_CMDID);
 }
+
+int ath12k_wmi_wow_add_wakeup_event(struct ath12k *ar, u32 vdev_id,
+				    enum wmi_wow_wakeup_event event,
+				    u32 enable)
+{
+	struct wmi_wow_add_del_event_cmd *cmd;
+	struct sk_buff *skb;
+	size_t len;
+
+	len = sizeof(*cmd);
+	skb = ath12k_wmi_alloc_skb(ar->wmi->wmi_ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_wow_add_del_event_cmd *)skb->data;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_WOW_ADD_DEL_EVT_CMD,
+						 sizeof(*cmd));
+	cmd->vdev_id = cpu_to_le32(vdev_id);
+	cmd->is_add = cpu_to_le32(enable);
+	cmd->event_bitmap = cpu_to_le32((1 << event));
+
+	ath12k_dbg(ar->ab, ATH12K_DBG_WMI, "wmi tlv wow add wakeup event %s enable %d vdev_id %d\n",
+		   wow_wakeup_event(event), enable, vdev_id);
+
+	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_WOW_ENABLE_DISABLE_WAKE_EVENT_CMDID);
+}
+
+int ath12k_wmi_wow_add_pattern(struct ath12k *ar, u32 vdev_id, u32 pattern_id,
+			       const u8 *pattern, const u8 *mask,
+			       int pattern_len, int pattern_offset)
+{
+	struct wmi_wow_add_pattern_cmd *cmd;
+	struct wmi_wow_bitmap_pattern_params *bitmap;
+	struct wmi_tlv *tlv;
+	struct sk_buff *skb;
+	void *ptr;
+	size_t len;
+
+	len = sizeof(*cmd) +
+	      sizeof(*tlv) +			/* array struct */
+	      sizeof(*bitmap) +			/* bitmap */
+	      sizeof(*tlv) +			/* empty ipv4 sync */
+	      sizeof(*tlv) +			/* empty ipv6 sync */
+	      sizeof(*tlv) +			/* empty magic */
+	      sizeof(*tlv) +			/* empty info timeout */
+	      sizeof(*tlv) + sizeof(u32);	/* ratelimit interval */
+
+	skb = ath12k_wmi_alloc_skb(ar->wmi->wmi_ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	/* cmd */
+	ptr = skb->data;
+	cmd = ptr;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_WOW_ADD_PATTERN_CMD,
+						 sizeof(*cmd));
+	cmd->vdev_id = cpu_to_le32(vdev_id);
+	cmd->pattern_id = cpu_to_le32(pattern_id);
+	cmd->pattern_type = cpu_to_le32(WOW_BITMAP_PATTERN);
+
+	ptr += sizeof(*cmd);
+
+	/* bitmap */
+	tlv = ptr;
+	tlv->header = ath12k_wmi_tlv_hdr(WMI_TAG_ARRAY_STRUCT, sizeof(*bitmap));
+
+	ptr += sizeof(*tlv);
+
+	bitmap = ptr;
+	bitmap->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_WOW_BITMAP_PATTERN_T,
+						    sizeof(*bitmap));
+	memcpy(bitmap->patternbuf, pattern, pattern_len);
+	memcpy(bitmap->bitmaskbuf, mask, pattern_len);
+	bitmap->pattern_offset = cpu_to_le32(pattern_offset);
+	bitmap->pattern_len = cpu_to_le32(pattern_len);
+	bitmap->bitmask_len = cpu_to_le32(pattern_len);
+	bitmap->pattern_id = cpu_to_le32(pattern_id);
+
+	ptr += sizeof(*bitmap);
+
+	/* ipv4 sync */
+	tlv = ptr;
+	tlv->header = ath12k_wmi_tlv_hdr(WMI_TAG_ARRAY_STRUCT, 0);
+
+	ptr += sizeof(*tlv);
+
+	/* ipv6 sync */
+	tlv = ptr;
+	tlv->header = ath12k_wmi_tlv_hdr(WMI_TAG_ARRAY_STRUCT, 0);
+
+	ptr += sizeof(*tlv);
+
+	/* magic */
+	tlv = ptr;
+	tlv->header = ath12k_wmi_tlv_hdr(WMI_TAG_ARRAY_STRUCT, 0);
+
+	ptr += sizeof(*tlv);
+
+	/* pattern info timeout */
+	tlv = ptr;
+	tlv->header = ath12k_wmi_tlv_hdr(WMI_TAG_ARRAY_UINT32, 0);
+
+	ptr += sizeof(*tlv);
+
+	/* ratelimit interval */
+	tlv = ptr;
+	tlv->header = ath12k_wmi_tlv_hdr(WMI_TAG_ARRAY_UINT32, sizeof(u32));
+
+	ath12k_dbg(ar->ab, ATH12K_DBG_WMI, "wmi tlv wow add pattern vdev_id %d pattern_id %d pattern_offset %d pattern_len %d\n",
+		   vdev_id, pattern_id, pattern_offset, pattern_len);
+
+	ath12k_dbg_dump(ar->ab, ATH12K_DBG_WMI, NULL, "wow pattern: ",
+			bitmap->patternbuf, pattern_len);
+	ath12k_dbg_dump(ar->ab, ATH12K_DBG_WMI, NULL, "wow bitmask: ",
+			bitmap->bitmaskbuf, pattern_len);
+
+	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_WOW_ADD_WAKE_PATTERN_CMDID);
+}
+
+int ath12k_wmi_wow_del_pattern(struct ath12k *ar, u32 vdev_id, u32 pattern_id)
+{
+	struct wmi_wow_del_pattern_cmd *cmd;
+	struct sk_buff *skb;
+	size_t len;
+
+	len = sizeof(*cmd);
+	skb = ath12k_wmi_alloc_skb(ar->wmi->wmi_ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_wow_del_pattern_cmd *)skb->data;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_WOW_DEL_PATTERN_CMD,
+						 sizeof(*cmd));
+	cmd->vdev_id = cpu_to_le32(vdev_id);
+	cmd->pattern_id = cpu_to_le32(pattern_id);
+	cmd->pattern_type = cpu_to_le32(WOW_BITMAP_PATTERN);
+
+	ath12k_dbg(ar->ab, ATH12K_DBG_WMI, "wmi tlv wow del pattern vdev_id %d pattern_id %d\n",
+		   vdev_id, pattern_id);
+
+	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_WOW_DEL_WAKE_PATTERN_CMDID);
+}
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 9cee3a114645..6cb06f6f7df4 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -5046,6 +5046,62 @@ static inline const char *wow_reason(enum wmi_wow_wake_reason reason)
 
 #undef C2S
 
+#define WOW_DEFAULT_BITMAP_PATTERN_SIZE		148
+#define WOW_DEFAULT_BITMASK_SIZE		148
+
+#define WOW_MIN_PATTERN_SIZE	1
+#define WOW_MAX_PATTERN_SIZE	148
+#define WOW_MAX_PKT_OFFSET	128
+#define WOW_HDR_LEN	(sizeof(struct ieee80211_hdr_3addr) + \
+	sizeof(struct rfc1042_hdr))
+#define WOW_MAX_REDUCE	(WOW_HDR_LEN - sizeof(struct ethhdr) - \
+	offsetof(struct ieee80211_hdr_3addr, addr1))
+
+struct wmi_wow_bitmap_pattern_params {
+	__le32 tlv_header;
+	u8 patternbuf[WOW_DEFAULT_BITMAP_PATTERN_SIZE];
+	u8 bitmaskbuf[WOW_DEFAULT_BITMASK_SIZE];
+	__le32 pattern_offset;
+	__le32 pattern_len;
+	__le32 bitmask_len;
+	__le32 pattern_id;
+} __packed;
+
+struct wmi_wow_add_pattern_cmd {
+	__le32 tlv_header;
+	__le32 vdev_id;
+	__le32 pattern_id;
+	__le32 pattern_type;
+} __packed;
+
+struct wmi_wow_del_pattern_cmd {
+	__le32 tlv_header;
+	__le32 vdev_id;
+	__le32 pattern_id;
+	__le32 pattern_type;
+} __packed;
+
+enum wmi_tlv_pattern_type {
+	WOW_PATTERN_MIN = 0,
+	WOW_BITMAP_PATTERN = WOW_PATTERN_MIN,
+	WOW_IPV4_SYNC_PATTERN,
+	WOW_IPV6_SYNC_PATTERN,
+	WOW_WILD_CARD_PATTERN,
+	WOW_TIMER_PATTERN,
+	WOW_MAGIC_PATTERN,
+	WOW_IPV6_RA_PATTERN,
+	WOW_IOAC_PKT_PATTERN,
+	WOW_IOAC_TMR_PATTERN,
+	WOW_PATTERN_MAX
+};
+
+struct wmi_wow_add_del_event_cmd {
+	__le32 tlv_header;
+	__le32 vdev_id;
+	__le32 is_add;
+	__le32 event_bitmap;
+} __packed;
+
 struct wmi_wow_enable_cmd {
 	__le32 tlv_header;
 	__le32 enable;
@@ -5228,4 +5284,11 @@ ath12k_wmi_mac_phy_get_hw_link_id(const struct ath12k_wmi_mac_phy_caps_params *p
 
 int ath12k_wmi_wow_host_wakeup_ind(struct ath12k *ar);
 int ath12k_wmi_wow_enable(struct ath12k *ar);
+int ath12k_wmi_wow_del_pattern(struct ath12k *ar, u32 vdev_id, u32 pattern_id);
+int ath12k_wmi_wow_add_pattern(struct ath12k *ar, u32 vdev_id, u32 pattern_id,
+			       const u8 *pattern, const u8 *mask,
+			       int pattern_len, int pattern_offset);
+int ath12k_wmi_wow_add_wakeup_event(struct ath12k *ar, u32 vdev_id,
+				    enum wmi_wow_wakeup_event event,
+				    u32 enable);
 #endif
diff --git a/drivers/net/wireless/ath/ath12k/wow.c b/drivers/net/wireless/ath/ath12k/wow.c
index d6d3ed7332ab..648e7cb970ea 100644
--- a/drivers/net/wireless/ath/ath12k/wow.c
+++ b/drivers/net/wireless/ath/ath12k/wow.c
@@ -7,12 +7,22 @@
 #include <linux/delay.h>
 
 #include "mac.h"
+
+#include <net/mac80211.h>
 #include "core.h"
 #include "hif.h"
 #include "debug.h"
 #include "wmi.h"
 #include "wow.h"
 
+static const struct wiphy_wowlan_support ath12k_wowlan_support = {
+	.flags = WIPHY_WOWLAN_DISCONNECT |
+		 WIPHY_WOWLAN_MAGIC_PKT,
+	.pattern_min_len = WOW_MIN_PATTERN_SIZE,
+	.pattern_max_len = WOW_MAX_PATTERN_SIZE,
+	.max_pkt_offset = WOW_MAX_PKT_OFFSET,
+};
+
 int ath12k_wow_enable(struct ath12k *ar)
 {
 	struct ath12k_base *ab = ar->ab;
@@ -77,3 +87,409 @@ int ath12k_wow_wakeup(struct ath12k *ar)
 
 	return 0;
 }
+
+static int ath12k_wow_vif_cleanup(struct ath12k_vif *arvif)
+{
+	struct ath12k *ar = arvif->ar;
+	int i, ret;
+
+	for (i = 0; i < WOW_EVENT_MAX; i++) {
+		ret = ath12k_wmi_wow_add_wakeup_event(ar, arvif->vdev_id, i, 0);
+		if (ret) {
+			ath12k_warn(ar->ab, "failed to issue wow wakeup for event %s on vdev %i: %d\n",
+				    wow_wakeup_event(i), arvif->vdev_id, ret);
+			return ret;
+		}
+	}
+
+	for (i = 0; i < ar->wow.max_num_patterns; i++) {
+		ret = ath12k_wmi_wow_del_pattern(ar, arvif->vdev_id, i);
+		if (ret) {
+			ath12k_warn(ar->ab, "failed to delete wow pattern %d for vdev %i: %d\n",
+				    i, arvif->vdev_id, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static int ath12k_wow_cleanup(struct ath12k *ar)
+{
+	struct ath12k_vif *arvif;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		ret = ath12k_wow_vif_cleanup(arvif);
+		if (ret) {
+			ath12k_warn(ar->ab, "failed to clean wow wakeups on vdev %i: %d\n",
+				    arvif->vdev_id, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+/* Convert a 802.3 format to a 802.11 format.
+ *         +------------+-----------+--------+----------------+
+ * 802.3:  |dest mac(6B)|src mac(6B)|type(2B)|     body...    |
+ *         +------------+-----------+--------+----------------+
+ *                |__         |_______    |____________  |________
+ *                   |                |                |          |
+ *         +--+------------+----+-----------+---------------+-----------+
+ * 802.11: |4B|dest mac(6B)| 6B |src mac(6B)|  8B  |type(2B)|  body...  |
+ *         +--+------------+----+-----------+---------------+-----------+
+ */
+static void
+ath12k_wow_convert_8023_to_80211(struct ath12k *ar,
+				 const struct cfg80211_pkt_pattern *eth_pattern,
+				 struct ath12k_pkt_pattern *i80211_pattern)
+{
+	size_t r1042_eth_ofs = offsetof(struct rfc1042_hdr, eth_type);
+	size_t a1_ofs = offsetof(struct ieee80211_hdr_3addr, addr1);
+	size_t a3_ofs = offsetof(struct ieee80211_hdr_3addr, addr3);
+	size_t i80211_hdr_len = sizeof(struct ieee80211_hdr_3addr);
+	size_t prot_ofs = offsetof(struct ethhdr, h_proto);
+	size_t src_ofs = offsetof(struct ethhdr, h_source);
+	u8 eth_bytemask[WOW_MAX_PATTERN_SIZE] = {};
+	const u8 *eth_pat = eth_pattern->pattern;
+	size_t eth_pat_len = eth_pattern->pattern_len;
+	size_t eth_pkt_ofs = eth_pattern->pkt_offset;
+	u8 *bytemask = i80211_pattern->bytemask;
+	u8 *pat = i80211_pattern->pattern;
+	size_t pat_len = 0;
+	size_t pkt_ofs = 0;
+	size_t delta;
+	int i;
+
+	/* convert bitmask to bytemask */
+	for (i = 0; i < eth_pat_len; i++)
+		if (eth_pattern->mask[i / 8] & BIT(i % 8))
+			eth_bytemask[i] = 0xff;
+
+	if (eth_pkt_ofs < ETH_ALEN) {
+		pkt_ofs = eth_pkt_ofs + a1_ofs;
+
+		if (eth_pkt_ofs + eth_pat_len < ETH_ALEN) {
+			memcpy(pat, eth_pat, eth_pat_len);
+			memcpy(bytemask, eth_bytemask, eth_pat_len);
+
+			pat_len = eth_pat_len;
+		} else if (eth_pkt_ofs + eth_pat_len < prot_ofs) {
+			memcpy(pat, eth_pat, ETH_ALEN - eth_pkt_ofs);
+			memcpy(bytemask, eth_bytemask, ETH_ALEN - eth_pkt_ofs);
+
+			delta = eth_pkt_ofs + eth_pat_len - src_ofs;
+			memcpy(pat + a3_ofs - pkt_ofs,
+			       eth_pat + ETH_ALEN - eth_pkt_ofs,
+			       delta);
+			memcpy(bytemask + a3_ofs - pkt_ofs,
+			       eth_bytemask + ETH_ALEN - eth_pkt_ofs,
+			       delta);
+
+			pat_len = a3_ofs - pkt_ofs + delta;
+		} else {
+			memcpy(pat, eth_pat, ETH_ALEN - eth_pkt_ofs);
+			memcpy(bytemask, eth_bytemask, ETH_ALEN - eth_pkt_ofs);
+
+			memcpy(pat + a3_ofs - pkt_ofs,
+			       eth_pat + ETH_ALEN - eth_pkt_ofs,
+			       ETH_ALEN);
+			memcpy(bytemask + a3_ofs - pkt_ofs,
+			       eth_bytemask + ETH_ALEN - eth_pkt_ofs,
+			       ETH_ALEN);
+
+			delta = eth_pkt_ofs + eth_pat_len - prot_ofs;
+			memcpy(pat + i80211_hdr_len + r1042_eth_ofs - pkt_ofs,
+			       eth_pat + prot_ofs - eth_pkt_ofs,
+			       delta);
+			memcpy(bytemask + i80211_hdr_len + r1042_eth_ofs - pkt_ofs,
+			       eth_bytemask + prot_ofs - eth_pkt_ofs,
+			       delta);
+
+			pat_len = i80211_hdr_len + r1042_eth_ofs - pkt_ofs + delta;
+		}
+	} else if (eth_pkt_ofs < prot_ofs) {
+		pkt_ofs = eth_pkt_ofs - ETH_ALEN + a3_ofs;
+
+		if (eth_pkt_ofs + eth_pat_len < prot_ofs) {
+			memcpy(pat, eth_pat, eth_pat_len);
+			memcpy(bytemask, eth_bytemask, eth_pat_len);
+
+			pat_len = eth_pat_len;
+		} else {
+			memcpy(pat, eth_pat, prot_ofs - eth_pkt_ofs);
+			memcpy(bytemask, eth_bytemask, prot_ofs - eth_pkt_ofs);
+
+			delta = eth_pkt_ofs + eth_pat_len - prot_ofs;
+			memcpy(pat + i80211_hdr_len + r1042_eth_ofs - pkt_ofs,
+			       eth_pat +  prot_ofs - eth_pkt_ofs,
+			       delta);
+			memcpy(bytemask + i80211_hdr_len + r1042_eth_ofs - pkt_ofs,
+			       eth_bytemask + prot_ofs - eth_pkt_ofs,
+			       delta);
+
+			pat_len =  i80211_hdr_len + r1042_eth_ofs - pkt_ofs + delta;
+		}
+	} else {
+		pkt_ofs = eth_pkt_ofs - prot_ofs + i80211_hdr_len + r1042_eth_ofs;
+
+		memcpy(pat, eth_pat, eth_pat_len);
+		memcpy(bytemask, eth_bytemask, eth_pat_len);
+
+		pat_len = eth_pat_len;
+	}
+
+	i80211_pattern->pattern_len = pat_len;
+	i80211_pattern->pkt_offset = pkt_ofs;
+}
+
+static int ath12k_vif_wow_set_wakeups(struct ath12k_vif *arvif,
+				      struct cfg80211_wowlan *wowlan)
+{
+	const struct cfg80211_pkt_pattern *patterns = wowlan->patterns;
+	struct ath12k *ar = arvif->ar;
+	unsigned long wow_mask = 0;
+	int pattern_id = 0;
+	int ret, i;
+
+	/* Setup requested WOW features */
+	switch (arvif->vdev_type) {
+	case WMI_VDEV_TYPE_IBSS:
+		__set_bit(WOW_BEACON_EVENT, &wow_mask);
+		fallthrough;
+	case WMI_VDEV_TYPE_AP:
+		__set_bit(WOW_DEAUTH_RECVD_EVENT, &wow_mask);
+		__set_bit(WOW_DISASSOC_RECVD_EVENT, &wow_mask);
+		__set_bit(WOW_PROBE_REQ_WPS_IE_EVENT, &wow_mask);
+		__set_bit(WOW_AUTH_REQ_EVENT, &wow_mask);
+		__set_bit(WOW_ASSOC_REQ_EVENT, &wow_mask);
+		__set_bit(WOW_HTT_EVENT, &wow_mask);
+		__set_bit(WOW_RA_MATCH_EVENT, &wow_mask);
+		break;
+	case WMI_VDEV_TYPE_STA:
+		if (wowlan->disconnect) {
+			__set_bit(WOW_DEAUTH_RECVD_EVENT, &wow_mask);
+			__set_bit(WOW_DISASSOC_RECVD_EVENT, &wow_mask);
+			__set_bit(WOW_BMISS_EVENT, &wow_mask);
+			__set_bit(WOW_CSA_IE_EVENT, &wow_mask);
+		}
+
+		if (wowlan->magic_pkt)
+			__set_bit(WOW_MAGIC_PKT_RECVD_EVENT, &wow_mask);
+		break;
+	default:
+		break;
+	}
+
+	for (i = 0; i < wowlan->n_patterns; i++) {
+		const struct cfg80211_pkt_pattern *eth_pattern = &patterns[i];
+		struct ath12k_pkt_pattern new_pattern = {};
+
+		if (WARN_ON(eth_pattern->pattern_len > WOW_MAX_PATTERN_SIZE))
+			return -EINVAL;
+
+		if (ar->ab->wow.wmi_conf_rx_decap_mode ==
+		    ATH12K_HW_TXRX_NATIVE_WIFI) {
+			ath12k_wow_convert_8023_to_80211(ar, eth_pattern,
+							 &new_pattern);
+
+			if (WARN_ON(new_pattern.pattern_len > WOW_MAX_PATTERN_SIZE))
+				return -EINVAL;
+		} else {
+			memcpy(new_pattern.pattern, eth_pattern->pattern,
+			       eth_pattern->pattern_len);
+
+			/* convert bitmask to bytemask */
+			for (i = 0; i < eth_pattern->pattern_len; i++)
+				if (eth_pattern->mask[i / 8] & BIT(i % 8))
+					new_pattern.bytemask[i] = 0xff;
+
+			new_pattern.pattern_len = eth_pattern->pattern_len;
+			new_pattern.pkt_offset = eth_pattern->pkt_offset;
+		}
+
+		ret = ath12k_wmi_wow_add_pattern(ar, arvif->vdev_id,
+						 pattern_id,
+						 new_pattern.pattern,
+						 new_pattern.bytemask,
+						 new_pattern.pattern_len,
+						 new_pattern.pkt_offset);
+		if (ret) {
+			ath12k_warn(ar->ab, "failed to add pattern %i to vdev %i: %d\n",
+				    pattern_id,
+				    arvif->vdev_id, ret);
+			return ret;
+		}
+
+		pattern_id++;
+		__set_bit(WOW_PATTERN_MATCH_EVENT, &wow_mask);
+	}
+
+	for (i = 0; i < WOW_EVENT_MAX; i++) {
+		if (!test_bit(i, &wow_mask))
+			continue;
+		ret = ath12k_wmi_wow_add_wakeup_event(ar, arvif->vdev_id, i, 1);
+		if (ret) {
+			ath12k_warn(ar->ab, "failed to enable wakeup event %s on vdev %i: %d\n",
+				    wow_wakeup_event(i), arvif->vdev_id, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static int ath12k_wow_set_wakeups(struct ath12k *ar,
+				  struct cfg80211_wowlan *wowlan)
+{
+	struct ath12k_vif *arvif;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		ret = ath12k_vif_wow_set_wakeups(arvif, wowlan);
+		if (ret) {
+			ath12k_warn(ar->ab, "failed to set wow wakeups on vdev %i: %d\n",
+				    arvif->vdev_id, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+int ath12k_wow_op_suspend(struct ieee80211_hw *hw,
+			  struct cfg80211_wowlan *wowlan)
+{
+	struct ath12k_hw *ah = ath12k_hw_to_ah(hw);
+	struct ath12k *ar = ath12k_ah_to_ar(ah, 0);
+	int ret;
+
+	mutex_lock(&ar->conf_mutex);
+
+	ret =  ath12k_wow_cleanup(ar);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to clear wow wakeup events: %d\n",
+			    ret);
+		goto exit;
+	}
+
+	ret = ath12k_wow_set_wakeups(ar, wowlan);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to set wow wakeup events: %d\n",
+			    ret);
+		goto cleanup;
+	}
+
+	ret = ath12k_mac_wait_tx_complete(ar);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to wait tx complete: %d\n", ret);
+		goto cleanup;
+	}
+
+	ret = ath12k_wow_enable(ar);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to start wow: %d\n", ret);
+		goto cleanup;
+	}
+
+	ath12k_hif_irq_disable(ar->ab);
+	ath12k_hif_ce_irq_disable(ar->ab);
+
+	ret = ath12k_hif_suspend(ar->ab);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to suspend hif: %d\n", ret);
+		goto wakeup;
+	}
+
+	goto exit;
+
+wakeup:
+	ath12k_wow_wakeup(ar);
+
+cleanup:
+	ath12k_wow_cleanup(ar);
+
+exit:
+	mutex_unlock(&ar->conf_mutex);
+	return ret ? 1 : 0;
+}
+
+void ath12k_wow_op_set_wakeup(struct ieee80211_hw *hw, bool enabled)
+{
+	struct ath12k_hw *ah = ath12k_hw_to_ah(hw);
+	struct ath12k *ar = ath12k_ah_to_ar(ah, 0);
+
+	mutex_lock(&ar->conf_mutex);
+	device_set_wakeup_enable(ar->ab->dev, enabled);
+	mutex_unlock(&ar->conf_mutex);
+}
+
+int ath12k_wow_op_resume(struct ieee80211_hw *hw)
+{
+	struct ath12k_hw *ah = ath12k_hw_to_ah(hw);
+	struct ath12k *ar = ath12k_ah_to_ar(ah, 0);
+	int ret;
+
+	mutex_lock(&ar->conf_mutex);
+
+	ret = ath12k_hif_resume(ar->ab);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to resume hif: %d\n", ret);
+		goto exit;
+	}
+
+	ath12k_hif_ce_irq_enable(ar->ab);
+	ath12k_hif_irq_enable(ar->ab);
+
+	ret = ath12k_wow_wakeup(ar);
+	if (ret)
+		ath12k_warn(ar->ab, "failed to wakeup from wow: %d\n", ret);
+
+exit:
+	if (ret) {
+		switch (ah->state) {
+		case ATH12K_HW_STATE_ON:
+			ah->state = ATH12K_HW_STATE_RESTARTING;
+			ret = 1;
+			break;
+		case ATH12K_HW_STATE_OFF:
+		case ATH12K_HW_STATE_RESTARTING:
+		case ATH12K_HW_STATE_RESTARTED:
+		case ATH12K_HW_STATE_WEDGED:
+			ath12k_warn(ar->ab, "encountered unexpected device state %d on resume, cannot recover\n",
+				    ah->state);
+			ret = -EIO;
+			break;
+		}
+	}
+
+	mutex_unlock(&ar->conf_mutex);
+	return ret;
+}
+
+int ath12k_wow_init(struct ath12k *ar)
+{
+	if (WARN_ON(!test_bit(WMI_TLV_SERVICE_WOW, ar->wmi->wmi_ab->svc_map)))
+		return -EINVAL;
+
+	ar->wow.wowlan_support = ath12k_wowlan_support;
+
+	if (ar->ab->wow.wmi_conf_rx_decap_mode == ATH12K_HW_TXRX_NATIVE_WIFI) {
+		ar->wow.wowlan_support.pattern_max_len -= WOW_MAX_REDUCE;
+		ar->wow.wowlan_support.max_pkt_offset -= WOW_MAX_REDUCE;
+	}
+
+	ar->wow.max_num_patterns = ATH12K_WOW_PATTERNS;
+	ar->wow.wowlan_support.n_patterns = ar->wow.max_num_patterns;
+	ar->ah->hw->wiphy->wowlan = &ar->wow.wowlan_support;
+
+	device_set_wakeup_capable(ar->ab->dev, true);
+
+	return 0;
+}
diff --git a/drivers/net/wireless/ath/ath12k/wow.h b/drivers/net/wireless/ath/ath12k/wow.h
index 6817ceaa4cfe..af9be5fadcc3 100644
--- a/drivers/net/wireless/ath/ath12k/wow.h
+++ b/drivers/net/wireless/ath/ath12k/wow.h
@@ -9,14 +9,46 @@
 
 #define ATH12K_WOW_RETRY_NUM		10
 #define ATH12K_WOW_RETRY_WAIT_MS	200
+#define ATH12K_WOW_PATTERNS		22
+
+struct ath12k_wow {
+	u32 max_num_patterns;
+	struct completion wakeup_completed;
+	struct wiphy_wowlan_support wowlan_support;
+};
+
+struct ath12k_pkt_pattern {
+	u8 pattern[WOW_MAX_PATTERN_SIZE];
+	u8 bytemask[WOW_MAX_PATTERN_SIZE];
+	int pattern_len;
+	int pkt_offset;
+};
+
+struct rfc1042_hdr {
+	u8 llc_dsap;
+	u8 llc_ssap;
+	u8 llc_ctrl;
+	u8 snap_oui[3];
+	__be16 eth_type;
+} __packed;
 
 #ifdef CONFIG_PM
 
+int ath12k_wow_init(struct ath12k *ar);
+int ath12k_wow_op_suspend(struct ieee80211_hw *hw,
+			  struct cfg80211_wowlan *wowlan);
+int ath12k_wow_op_resume(struct ieee80211_hw *hw);
+void ath12k_wow_op_set_wakeup(struct ieee80211_hw *hw, bool enabled);
 int ath12k_wow_enable(struct ath12k *ar);
 int ath12k_wow_wakeup(struct ath12k *ar);
 
 #else
 
+static inline int ath12k_wow_init(struct ath12k *ar)
+{
+	return 0;
+}
+
 static inline int ath12k_wow_enable(struct ath12k *ar)
 {
 	return 0;
-- 
cgit v1.2.3-58-ga151


From 16f474d6d49db5d93e79c2d4bdcc3a730cf1d1f1 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Tue, 4 Jun 2024 13:54:03 +0800
Subject: wifi: ath12k: add WoW net-detect functionality

Implement net-detect feature by setting flag
WIPHY_WOWLAN_NET_DETECT if firmware supports this
feature. Driver sets the related PNO configuration
to firmware before entering WoW and firmware then
scans periodically and wakes up host if a specific
SSID is found.

Note that firmware crashes if we enable it for both
P2P vdev and station vdev simultaneously because
firmware can only support one vdev at a time. Since
there is rare scenario for a P2P vdev to do net-detect,
skip it for P2P vdevs.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240604055407.12506-5-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.h |   2 +
 drivers/net/wireless/ath/ath12k/mac.c  |  12 ++
 drivers/net/wireless/ath/ath12k/wmi.c  | 145 +++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/wmi.h  | 166 +++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/wow.c  | 202 ++++++++++++++++++++++++++++++++-
 5 files changed, 524 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index 2fd08648cef3..81c4bfe7792f 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -630,6 +630,8 @@ struct ath12k {
 
 	u32 freq_low;
 	u32 freq_high;
+
+	bool nlo_enabled;
 };
 
 struct ath12k_hw {
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 59662ca241ca..cd8fb3797eea 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -9108,6 +9108,18 @@ static int ath12k_mac_hw_register(struct ath12k_hw *ah)
 		ieee80211_hw_set(hw, SUPPORT_FAST_XMIT);
 	}
 
+	if (test_bit(WMI_TLV_SERVICE_NLO, ar->wmi->wmi_ab->svc_map)) {
+		wiphy->max_sched_scan_ssids = WMI_PNO_MAX_SUPP_NETWORKS;
+		wiphy->max_match_sets = WMI_PNO_MAX_SUPP_NETWORKS;
+		wiphy->max_sched_scan_ie_len = WMI_PNO_MAX_IE_LENGTH;
+		wiphy->max_sched_scan_plans = WMI_PNO_MAX_SCHED_SCAN_PLANS;
+		wiphy->max_sched_scan_plan_interval =
+					WMI_PNO_MAX_SCHED_SCAN_PLAN_INT;
+		wiphy->max_sched_scan_plan_iterations =
+					WMI_PNO_MAX_SCHED_SCAN_PLAN_ITRNS;
+		wiphy->features |= NL80211_FEATURE_ND_RANDOM_MAC_ADDR;
+	}
+
 	ret = ath12k_wow_init(ar);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to init wow: %d\n", ret);
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index f72ec153723c..b69f20495ab4 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -7607,3 +7607,148 @@ int ath12k_wmi_wow_del_pattern(struct ath12k *ar, u32 vdev_id, u32 pattern_id)
 
 	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_WOW_DEL_WAKE_PATTERN_CMDID);
 }
+
+static struct sk_buff *
+ath12k_wmi_op_gen_config_pno_start(struct ath12k *ar, u32 vdev_id,
+				   struct wmi_pno_scan_req_arg *pno)
+{
+	struct nlo_configured_params *nlo_list;
+	size_t len, nlo_list_len, channel_list_len;
+	struct wmi_wow_nlo_config_cmd *cmd;
+	__le32 *channel_list;
+	struct wmi_tlv *tlv;
+	struct sk_buff *skb;
+	void *ptr;
+	u32 i;
+
+	len = sizeof(*cmd) +
+	      sizeof(*tlv) +
+	      /* TLV place holder for array of structures
+	       * nlo_configured_params(nlo_list)
+	       */
+	      sizeof(*tlv);
+	      /* TLV place holder for array of uint32 channel_list */
+
+	channel_list_len = sizeof(u32) * pno->a_networks[0].channel_count;
+	len += channel_list_len;
+
+	nlo_list_len = sizeof(*nlo_list) * pno->uc_networks_count;
+	len += nlo_list_len;
+
+	skb = ath12k_wmi_alloc_skb(ar->wmi->wmi_ab, len);
+	if (!skb)
+		return ERR_PTR(-ENOMEM);
+
+	ptr = skb->data;
+	cmd = ptr;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_NLO_CONFIG_CMD, sizeof(*cmd));
+
+	cmd->vdev_id = cpu_to_le32(pno->vdev_id);
+	cmd->flags = cpu_to_le32(WMI_NLO_CONFIG_START | WMI_NLO_CONFIG_SSID_HIDE_EN);
+
+	/* current FW does not support min-max range for dwell time */
+	cmd->active_dwell_time = cpu_to_le32(pno->active_max_time);
+	cmd->passive_dwell_time = cpu_to_le32(pno->passive_max_time);
+
+	if (pno->do_passive_scan)
+		cmd->flags |= cpu_to_le32(WMI_NLO_CONFIG_SCAN_PASSIVE);
+
+	cmd->fast_scan_period = cpu_to_le32(pno->fast_scan_period);
+	cmd->slow_scan_period = cpu_to_le32(pno->slow_scan_period);
+	cmd->fast_scan_max_cycles = cpu_to_le32(pno->fast_scan_max_cycles);
+	cmd->delay_start_time = cpu_to_le32(pno->delay_start_time);
+
+	if (pno->enable_pno_scan_randomization) {
+		cmd->flags |= cpu_to_le32(WMI_NLO_CONFIG_SPOOFED_MAC_IN_PROBE_REQ |
+					  WMI_NLO_CONFIG_RANDOM_SEQ_NO_IN_PROBE_REQ);
+		ether_addr_copy(cmd->mac_addr.addr, pno->mac_addr);
+		ether_addr_copy(cmd->mac_mask.addr, pno->mac_addr_mask);
+	}
+
+	ptr += sizeof(*cmd);
+
+	/* nlo_configured_params(nlo_list) */
+	cmd->no_of_ssids = cpu_to_le32(pno->uc_networks_count);
+	tlv = ptr;
+	tlv->header = ath12k_wmi_tlv_hdr(WMI_TAG_ARRAY_STRUCT, nlo_list_len);
+
+	ptr += sizeof(*tlv);
+	nlo_list = ptr;
+	for (i = 0; i < pno->uc_networks_count; i++) {
+		tlv = (struct wmi_tlv *)(&nlo_list[i].tlv_header);
+		tlv->header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_ARRAY_BYTE,
+						     sizeof(*nlo_list));
+
+		nlo_list[i].ssid.valid = cpu_to_le32(1);
+		nlo_list[i].ssid.ssid.ssid_len =
+			cpu_to_le32(pno->a_networks[i].ssid.ssid_len);
+		memcpy(nlo_list[i].ssid.ssid.ssid,
+		       pno->a_networks[i].ssid.ssid,
+		       le32_to_cpu(nlo_list[i].ssid.ssid.ssid_len));
+
+		if (pno->a_networks[i].rssi_threshold &&
+		    pno->a_networks[i].rssi_threshold > -300) {
+			nlo_list[i].rssi_cond.valid = cpu_to_le32(1);
+			nlo_list[i].rssi_cond.rssi =
+					cpu_to_le32(pno->a_networks[i].rssi_threshold);
+		}
+
+		nlo_list[i].bcast_nw_type.valid = cpu_to_le32(1);
+		nlo_list[i].bcast_nw_type.bcast_nw_type =
+					cpu_to_le32(pno->a_networks[i].bcast_nw_type);
+	}
+
+	ptr += nlo_list_len;
+	cmd->num_of_channels = cpu_to_le32(pno->a_networks[0].channel_count);
+	tlv = ptr;
+	tlv->header =  ath12k_wmi_tlv_hdr(WMI_TAG_ARRAY_UINT32, channel_list_len);
+	ptr += sizeof(*tlv);
+	channel_list = ptr;
+
+	for (i = 0; i < pno->a_networks[0].channel_count; i++)
+		channel_list[i] = cpu_to_le32(pno->a_networks[0].channels[i]);
+
+	ath12k_dbg(ar->ab, ATH12K_DBG_WMI, "wmi tlv start pno config vdev_id %d\n",
+		   vdev_id);
+
+	return skb;
+}
+
+static struct sk_buff *ath12k_wmi_op_gen_config_pno_stop(struct ath12k *ar,
+							 u32 vdev_id)
+{
+	struct wmi_wow_nlo_config_cmd *cmd;
+	struct sk_buff *skb;
+	size_t len;
+
+	len = sizeof(*cmd);
+	skb = ath12k_wmi_alloc_skb(ar->wmi->wmi_ab, len);
+	if (!skb)
+		return ERR_PTR(-ENOMEM);
+
+	cmd = (struct wmi_wow_nlo_config_cmd *)skb->data;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_NLO_CONFIG_CMD, len);
+
+	cmd->vdev_id = cpu_to_le32(vdev_id);
+	cmd->flags = cpu_to_le32(WMI_NLO_CONFIG_STOP);
+
+	ath12k_dbg(ar->ab, ATH12K_DBG_WMI,
+		   "wmi tlv stop pno config vdev_id %d\n", vdev_id);
+	return skb;
+}
+
+int ath12k_wmi_wow_config_pno(struct ath12k *ar, u32 vdev_id,
+			      struct wmi_pno_scan_req_arg  *pno_scan)
+{
+	struct sk_buff *skb;
+
+	if (pno_scan->enable)
+		skb = ath12k_wmi_op_gen_config_pno_start(ar, vdev_id, pno_scan);
+	else
+		skb = ath12k_wmi_op_gen_config_pno_stop(ar, vdev_id);
+
+	if (IS_ERR_OR_NULL(skb))
+		return -ENOMEM;
+
+	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_NETWORK_LIST_OFFLOAD_CONFIG_CMDID);
+}
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 6cb06f6f7df4..0842ab760f2b 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -5130,6 +5130,170 @@ struct wmi_wow_ev_arg {
 	enum wmi_wow_wake_reason wake_reason;
 };
 
+#define WMI_PNO_MAX_SCHED_SCAN_PLANS      2
+#define WMI_PNO_MAX_SCHED_SCAN_PLAN_INT   7200
+#define WMI_PNO_MAX_SCHED_SCAN_PLAN_ITRNS 100
+#define WMI_PNO_MAX_NETW_CHANNELS         26
+#define WMI_PNO_MAX_NETW_CHANNELS_EX      60
+#define WMI_PNO_MAX_SUPP_NETWORKS         WLAN_SCAN_PARAMS_MAX_SSID
+#define WMI_PNO_MAX_IE_LENGTH             WLAN_SCAN_PARAMS_MAX_IE_LEN
+
+/* size based of dot11 declaration without extra IEs as we will not carry those for PNO */
+#define WMI_PNO_MAX_PB_REQ_SIZE    450
+
+#define WMI_PNO_24GHZ_DEFAULT_CH     1
+#define WMI_PNO_5GHZ_DEFAULT_CH      36
+
+#define WMI_ACTIVE_MAX_CHANNEL_TIME 40
+#define WMI_PASSIVE_MAX_CHANNEL_TIME   110
+
+/* SSID broadcast type */
+enum wmi_ssid_bcast_type {
+	BCAST_UNKNOWN      = 0,
+	BCAST_NORMAL       = 1,
+	BCAST_HIDDEN       = 2,
+};
+
+#define WMI_NLO_MAX_SSIDS    16
+#define WMI_NLO_MAX_CHAN     48
+
+#define WMI_NLO_CONFIG_STOP                             BIT(0)
+#define WMI_NLO_CONFIG_START                            BIT(1)
+#define WMI_NLO_CONFIG_RESET                            BIT(2)
+#define WMI_NLO_CONFIG_SLOW_SCAN                        BIT(4)
+#define WMI_NLO_CONFIG_FAST_SCAN                        BIT(5)
+#define WMI_NLO_CONFIG_SSID_HIDE_EN                     BIT(6)
+
+/* This bit is used to indicate if EPNO or supplicant PNO is enabled.
+ * Only one of them can be enabled at a given time
+ */
+#define WMI_NLO_CONFIG_ENLO                             BIT(7)
+#define WMI_NLO_CONFIG_SCAN_PASSIVE                     BIT(8)
+#define WMI_NLO_CONFIG_ENLO_RESET                       BIT(9)
+#define WMI_NLO_CONFIG_SPOOFED_MAC_IN_PROBE_REQ         BIT(10)
+#define WMI_NLO_CONFIG_RANDOM_SEQ_NO_IN_PROBE_REQ       BIT(11)
+#define WMI_NLO_CONFIG_ENABLE_IE_WHITELIST_IN_PROBE_REQ BIT(12)
+#define WMI_NLO_CONFIG_ENABLE_CNLO_RSSI_CONFIG          BIT(13)
+
+struct wmi_nlo_ssid_params {
+	__le32 valid;
+	struct ath12k_wmi_ssid_params ssid;
+} __packed;
+
+struct wmi_nlo_enc_params {
+	__le32 valid;
+	__le32 enc_type;
+} __packed;
+
+struct wmi_nlo_auth_params {
+	__le32 valid;
+	__le32 auth_type;
+} __packed;
+
+struct wmi_nlo_bcast_nw_params {
+	__le32 valid;
+	__le32 bcast_nw_type;
+} __packed;
+
+struct wmi_nlo_rssi_params {
+	__le32 valid;
+	__le32 rssi;
+} __packed;
+
+struct nlo_configured_params {
+	/* TLV tag and len;*/
+	__le32 tlv_header;
+	struct wmi_nlo_ssid_params ssid;
+	struct wmi_nlo_enc_params enc_type;
+	struct wmi_nlo_auth_params auth_type;
+	struct wmi_nlo_rssi_params rssi_cond;
+
+	/* indicates if the SSID is hidden or not */
+	struct wmi_nlo_bcast_nw_params bcast_nw_type;
+} __packed;
+
+struct wmi_network_type_arg {
+	struct cfg80211_ssid ssid;
+	u32 authentication;
+	u32 encryption;
+	u32 bcast_nw_type;
+	u8 channel_count;
+	u16 channels[WMI_PNO_MAX_NETW_CHANNELS_EX];
+	s32 rssi_threshold;
+};
+
+struct wmi_pno_scan_req_arg {
+	u8 enable;
+	u8 vdev_id;
+	u8 uc_networks_count;
+	struct wmi_network_type_arg a_networks[WMI_PNO_MAX_SUPP_NETWORKS];
+	u32 fast_scan_period;
+	u32 slow_scan_period;
+	u8 fast_scan_max_cycles;
+
+	bool do_passive_scan;
+
+	u32 delay_start_time;
+	u32 active_min_time;
+	u32 active_max_time;
+	u32 passive_min_time;
+	u32 passive_max_time;
+
+	/* mac address randomization attributes */
+	u32 enable_pno_scan_randomization;
+	u8 mac_addr[ETH_ALEN];
+	u8 mac_addr_mask[ETH_ALEN];
+};
+
+struct wmi_wow_nlo_config_cmd {
+	__le32 tlv_header;
+	__le32 flags;
+	__le32 vdev_id;
+	__le32 fast_scan_max_cycles;
+	__le32 active_dwell_time;
+	__le32 passive_dwell_time;
+	__le32 probe_bundle_size;
+
+	/* ART = IRT */
+	__le32 rest_time;
+
+	/* max value that can be reached after scan_backoff_multiplier */
+	__le32 max_rest_time;
+
+	__le32 scan_backoff_multiplier;
+	__le32 fast_scan_period;
+
+	/* specific to windows */
+	__le32 slow_scan_period;
+
+	__le32 no_of_ssids;
+
+	__le32 num_of_channels;
+
+	/* NLO scan start delay time in milliseconds */
+	__le32 delay_start_time;
+
+	/* MAC Address to use in Probe Req as SA */
+	struct ath12k_wmi_mac_addr_params mac_addr;
+
+	/* Mask on which MAC has to be randomized */
+	struct ath12k_wmi_mac_addr_params mac_mask;
+
+	/* IE bitmap to use in Probe Req */
+	__le32 ie_bitmap[8];
+
+	/* Number of vendor OUIs. In the TLV vendor_oui[] */
+	__le32 num_vendor_oui;
+
+	/* Number of connected NLO band preferences */
+	__le32 num_cnlo_band_pref;
+
+	/* The TLVs will follow.
+	 * nlo_configured_params nlo_list[];
+	 * u32 channel_list[num_of_channels];
+	 */
+} __packed;
+
 void ath12k_wmi_init_qcn9274(struct ath12k_base *ab,
 			     struct ath12k_wmi_resource_config_arg *config);
 void ath12k_wmi_init_wcn7850(struct ath12k_base *ab,
@@ -5291,4 +5455,6 @@ int ath12k_wmi_wow_add_pattern(struct ath12k *ar, u32 vdev_id, u32 pattern_id,
 int ath12k_wmi_wow_add_wakeup_event(struct ath12k *ar, u32 vdev_id,
 				    enum wmi_wow_wakeup_event event,
 				    u32 enable);
+int ath12k_wmi_wow_config_pno(struct ath12k *ar, u32 vdev_id,
+			      struct wmi_pno_scan_req_arg  *pno_scan);
 #endif
diff --git a/drivers/net/wireless/ath/ath12k/wow.c b/drivers/net/wireless/ath/ath12k/wow.c
index 648e7cb970ea..5d4f11028403 100644
--- a/drivers/net/wireless/ath/ath12k/wow.c
+++ b/drivers/net/wireless/ath/ath12k/wow.c
@@ -23,6 +23,13 @@ static const struct wiphy_wowlan_support ath12k_wowlan_support = {
 	.max_pkt_offset = WOW_MAX_PKT_OFFSET,
 };
 
+static inline bool ath12k_wow_is_p2p_vdev(struct ath12k_vif *arvif)
+{
+	return (arvif->vdev_subtype == WMI_VDEV_SUBTYPE_P2P_DEVICE ||
+		arvif->vdev_subtype == WMI_VDEV_SUBTYPE_P2P_CLIENT ||
+		arvif->vdev_subtype == WMI_VDEV_SUBTYPE_P2P_GO);
+}
+
 int ath12k_wow_enable(struct ath12k *ar)
 {
 	struct ath12k_base *ab = ar->ab;
@@ -247,7 +254,101 @@ ath12k_wow_convert_8023_to_80211(struct ath12k *ar,
 	i80211_pattern->pkt_offset = pkt_ofs;
 }
 
-static int ath12k_vif_wow_set_wakeups(struct ath12k_vif *arvif,
+static int
+ath12k_wow_pno_check_and_convert(struct ath12k *ar, u32 vdev_id,
+				 const struct cfg80211_sched_scan_request *nd_config,
+				 struct wmi_pno_scan_req_arg *pno)
+{
+	int i, j;
+	u8 ssid_len;
+
+	pno->enable = 1;
+	pno->vdev_id = vdev_id;
+	pno->uc_networks_count = nd_config->n_match_sets;
+
+	if (!pno->uc_networks_count ||
+	    pno->uc_networks_count > WMI_PNO_MAX_SUPP_NETWORKS)
+		return -EINVAL;
+
+	if (nd_config->n_channels > WMI_PNO_MAX_NETW_CHANNELS_EX)
+		return -EINVAL;
+
+	/* Filling per profile params */
+	for (i = 0; i < pno->uc_networks_count; i++) {
+		ssid_len = nd_config->match_sets[i].ssid.ssid_len;
+
+		if (ssid_len == 0 || ssid_len > 32)
+			return -EINVAL;
+
+		pno->a_networks[i].ssid.ssid_len = ssid_len;
+
+		memcpy(pno->a_networks[i].ssid.ssid,
+		       nd_config->match_sets[i].ssid.ssid,
+		       ssid_len);
+		pno->a_networks[i].authentication = 0;
+		pno->a_networks[i].encryption     = 0;
+		pno->a_networks[i].bcast_nw_type  = 0;
+
+		/* Copying list of valid channel into request */
+		pno->a_networks[i].channel_count = nd_config->n_channels;
+		pno->a_networks[i].rssi_threshold = nd_config->match_sets[i].rssi_thold;
+
+		for (j = 0; j < nd_config->n_channels; j++) {
+			pno->a_networks[i].channels[j] =
+					nd_config->channels[j]->center_freq;
+		}
+	}
+
+	/* set scan to passive if no SSIDs are specified in the request */
+	if (nd_config->n_ssids == 0)
+		pno->do_passive_scan = true;
+	else
+		pno->do_passive_scan = false;
+
+	for (i = 0; i < nd_config->n_ssids; i++) {
+		for (j = 0; j < pno->uc_networks_count; j++) {
+			if (pno->a_networks[j].ssid.ssid_len ==
+				nd_config->ssids[i].ssid_len &&
+			    !memcmp(pno->a_networks[j].ssid.ssid,
+				    nd_config->ssids[i].ssid,
+				    pno->a_networks[j].ssid.ssid_len)) {
+				pno->a_networks[j].bcast_nw_type = BCAST_HIDDEN;
+				break;
+			}
+		}
+	}
+
+	if (nd_config->n_scan_plans == 2) {
+		pno->fast_scan_period = nd_config->scan_plans[0].interval * MSEC_PER_SEC;
+		pno->fast_scan_max_cycles = nd_config->scan_plans[0].iterations;
+		pno->slow_scan_period =
+			nd_config->scan_plans[1].interval * MSEC_PER_SEC;
+	} else if (nd_config->n_scan_plans == 1) {
+		pno->fast_scan_period = nd_config->scan_plans[0].interval * MSEC_PER_SEC;
+		pno->fast_scan_max_cycles = 1;
+		pno->slow_scan_period = nd_config->scan_plans[0].interval * MSEC_PER_SEC;
+	} else {
+		ath12k_warn(ar->ab, "Invalid number of PNO scan plans: %d",
+			    nd_config->n_scan_plans);
+	}
+
+	if (nd_config->flags & NL80211_SCAN_FLAG_RANDOM_ADDR) {
+		/* enable mac randomization */
+		pno->enable_pno_scan_randomization = 1;
+		memcpy(pno->mac_addr, nd_config->mac_addr, ETH_ALEN);
+		memcpy(pno->mac_addr_mask, nd_config->mac_addr_mask, ETH_ALEN);
+	}
+
+	pno->delay_start_time = nd_config->delay;
+
+	/* Current FW does not support min-max range for dwell time */
+	pno->active_max_time = WMI_ACTIVE_MAX_CHANNEL_TIME;
+	pno->passive_max_time = WMI_PASSIVE_MAX_CHANNEL_TIME;
+
+	return 0;
+}
+
+static int ath12k_wow_vif_set_wakeups(struct ath12k_vif *arvif,
 				      struct cfg80211_wowlan *wowlan)
 {
 	const struct cfg80211_pkt_pattern *patterns = wowlan->patterns;
@@ -280,6 +381,26 @@ static int ath12k_vif_wow_set_wakeups(struct ath12k_vif *arvif,
 
 		if (wowlan->magic_pkt)
 			__set_bit(WOW_MAGIC_PKT_RECVD_EVENT, &wow_mask);
+
+		if (wowlan->nd_config) {
+			struct wmi_pno_scan_req_arg *pno;
+			int ret;
+
+			pno = kzalloc(sizeof(*pno), GFP_KERNEL);
+			if (!pno)
+				return -ENOMEM;
+
+			ar->nlo_enabled = true;
+
+			ret = ath12k_wow_pno_check_and_convert(ar, arvif->vdev_id,
+							       wowlan->nd_config, pno);
+			if (!ret) {
+				ath12k_wmi_wow_config_pno(ar, arvif->vdev_id, pno);
+				__set_bit(WOW_NLO_DETECTED_EVENT, &wow_mask);
+			}
+
+			kfree(pno);
+		}
 		break;
 	default:
 		break;
@@ -352,7 +473,9 @@ static int ath12k_wow_set_wakeups(struct ath12k *ar,
 	lockdep_assert_held(&ar->conf_mutex);
 
 	list_for_each_entry(arvif, &ar->arvifs, list) {
-		ret = ath12k_vif_wow_set_wakeups(arvif, wowlan);
+		if (ath12k_wow_is_p2p_vdev(arvif))
+			continue;
+		ret = ath12k_wow_vif_set_wakeups(arvif, wowlan);
 		if (ret) {
 			ath12k_warn(ar->ab, "failed to set wow wakeups on vdev %i: %d\n",
 				    arvif->vdev_id, ret);
@@ -363,6 +486,66 @@ static int ath12k_wow_set_wakeups(struct ath12k *ar,
 	return 0;
 }
 
+static int ath12k_wow_vdev_clean_nlo(struct ath12k *ar, u32 vdev_id)
+{
+	struct wmi_pno_scan_req_arg *pno;
+	int ret;
+
+	if (!ar->nlo_enabled)
+		return 0;
+
+	pno = kzalloc(sizeof(*pno), GFP_KERNEL);
+	if (!pno)
+		return -ENOMEM;
+
+	pno->enable = 0;
+	ret = ath12k_wmi_wow_config_pno(ar, vdev_id, pno);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to disable PNO: %d", ret);
+		goto out;
+	}
+
+	ar->nlo_enabled = false;
+
+out:
+	kfree(pno);
+	return ret;
+}
+
+static int ath12k_wow_vif_clean_nlo(struct ath12k_vif *arvif)
+{
+	struct ath12k *ar = arvif->ar;
+
+	switch (arvif->vdev_type) {
+	case WMI_VDEV_TYPE_STA:
+		return ath12k_wow_vdev_clean_nlo(ar, arvif->vdev_id);
+	default:
+		return 0;
+	}
+}
+
+static int ath12k_wow_nlo_cleanup(struct ath12k *ar)
+{
+	struct ath12k_vif *arvif;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		if (ath12k_wow_is_p2p_vdev(arvif))
+			continue;
+
+		ret = ath12k_wow_vif_clean_nlo(arvif);
+		if (ret) {
+			ath12k_warn(ar->ab, "failed to clean nlo settings on vdev %i: %d\n",
+				    arvif->vdev_id, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
 int ath12k_wow_op_suspend(struct ieee80211_hw *hw,
 			  struct cfg80211_wowlan *wowlan)
 {
@@ -448,8 +631,16 @@ int ath12k_wow_op_resume(struct ieee80211_hw *hw)
 	ath12k_hif_irq_enable(ar->ab);
 
 	ret = ath12k_wow_wakeup(ar);
-	if (ret)
+	if (ret) {
 		ath12k_warn(ar->ab, "failed to wakeup from wow: %d\n", ret);
+		goto exit;
+	}
+
+	ret = ath12k_wow_nlo_cleanup(ar);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to cleanup nlo: %d\n", ret);
+		goto exit;
+	}
 
 exit:
 	if (ret) {
@@ -485,6 +676,11 @@ int ath12k_wow_init(struct ath12k *ar)
 		ar->wow.wowlan_support.max_pkt_offset -= WOW_MAX_REDUCE;
 	}
 
+	if (test_bit(WMI_TLV_SERVICE_NLO, ar->wmi->wmi_ab->svc_map)) {
+		ar->wow.wowlan_support.flags |= WIPHY_WOWLAN_NET_DETECT;
+		ar->wow.wowlan_support.max_nd_match_sets = WMI_PNO_MAX_SUPP_NETWORKS;
+	}
+
 	ar->wow.max_num_patterns = ATH12K_WOW_PATTERNS;
 	ar->wow.wowlan_support.n_patterns = ar->wow.max_num_patterns;
 	ar->ah->hw->wiphy->wowlan = &ar->wow.wowlan_support;
-- 
cgit v1.2.3-58-ga151


From 66a9448b1b897793a3237f832b5ce3bb255e28e7 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Wed, 19 Jun 2024 17:21:12 +0300
Subject: wifi: ath12k: implement hardware data filter

Host needs to set hardware data filter before entering WoW to
let firmware drop needless broadcast/multicast frames to avoid
frequent wakeup. Host clears hardware data filter when leaving WoW.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240604055407.12506-6-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/wmi.c | 31 ++++++++++++++++
 drivers/net/wireless/ath/ath12k/wmi.h | 21 +++++++++++
 drivers/net/wireless/ath/ath12k/wow.c | 66 +++++++++++++++++++++++++++++++++++
 3 files changed, 118 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index b69f20495ab4..b1f343efa4b0 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -7424,6 +7424,37 @@ void ath12k_wmi_detach(struct ath12k_base *ab)
 	ath12k_wmi_free_dbring_caps(ab);
 }
 
+int ath12k_wmi_hw_data_filter_cmd(struct ath12k *ar, struct wmi_hw_data_filter_arg *arg)
+{
+	struct wmi_hw_data_filter_cmd *cmd;
+	struct sk_buff *skb;
+	int len;
+
+	len = sizeof(*cmd);
+	skb = ath12k_wmi_alloc_skb(ar->wmi->wmi_ab, len);
+
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_hw_data_filter_cmd *)skb->data;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_HW_DATA_FILTER_CMD,
+						 sizeof(*cmd));
+	cmd->vdev_id = cpu_to_le32(arg->vdev_id);
+	cmd->enable = cpu_to_le32(arg->enable ? 1 : 0);
+
+	/* Set all modes in case of disable */
+	if (arg->enable)
+		cmd->hw_filter_bitmap = cpu_to_le32(arg->hw_filter_bitmap);
+	else
+		cmd->hw_filter_bitmap = cpu_to_le32((u32)~0U);
+
+	ath12k_dbg(ar->ab, ATH12K_DBG_WMI,
+		   "wmi hw data filter enable %d filter_bitmap 0x%x\n",
+		   arg->enable, arg->hw_filter_bitmap);
+
+	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_HW_DATA_FILTER_CMDID);
+}
+
 int ath12k_wmi_wow_host_wakeup_ind(struct ath12k *ar)
 {
 	struct wmi_wow_host_wakeup_cmd *cmd;
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 0842ab760f2b..62d7bb9e791d 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -5294,6 +5294,25 @@ struct wmi_wow_nlo_config_cmd {
 	 */
 } __packed;
 
+/* Definition of HW data filtering */
+enum hw_data_filter_type {
+	WMI_HW_DATA_FILTER_DROP_NON_ARP_BC = BIT(0),
+	WMI_HW_DATA_FILTER_DROP_NON_ICMPV6_MC = BIT(1),
+};
+
+struct wmi_hw_data_filter_cmd {
+	__le32 tlv_header;
+	__le32 vdev_id;
+	__le32 enable;
+	__le32 hw_filter_bitmap;
+} __packed;
+
+struct wmi_hw_data_filter_arg {
+	u32 vdev_id;
+	bool enable;
+	u32 hw_filter_bitmap;
+};
+
 void ath12k_wmi_init_qcn9274(struct ath12k_base *ab,
 			     struct ath12k_wmi_resource_config_arg *config);
 void ath12k_wmi_init_wcn7850(struct ath12k_base *ab,
@@ -5457,4 +5476,6 @@ int ath12k_wmi_wow_add_wakeup_event(struct ath12k *ar, u32 vdev_id,
 				    u32 enable);
 int ath12k_wmi_wow_config_pno(struct ath12k *ar, u32 vdev_id,
 			      struct wmi_pno_scan_req_arg  *pno_scan);
+int ath12k_wmi_hw_data_filter_cmd(struct ath12k *ar,
+				  struct wmi_hw_data_filter_arg *arg);
 #endif
diff --git a/drivers/net/wireless/ath/ath12k/wow.c b/drivers/net/wireless/ath/ath12k/wow.c
index 5d4f11028403..146281244f70 100644
--- a/drivers/net/wireless/ath/ath12k/wow.c
+++ b/drivers/net/wireless/ath/ath12k/wow.c
@@ -546,6 +546,59 @@ static int ath12k_wow_nlo_cleanup(struct ath12k *ar)
 	return 0;
 }
 
+static int ath12k_wow_set_hw_filter(struct ath12k *ar)
+{
+	struct wmi_hw_data_filter_arg arg;
+	struct ath12k_vif *arvif;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		if (arvif->vdev_type != WMI_VDEV_TYPE_STA)
+			continue;
+
+		arg.vdev_id = arvif->vdev_id;
+		arg.enable = true;
+		arg.hw_filter_bitmap = WMI_HW_DATA_FILTER_DROP_NON_ICMPV6_MC;
+		ret = ath12k_wmi_hw_data_filter_cmd(ar, &arg);
+		if (ret) {
+			ath12k_warn(ar->ab, "failed to set hw data filter on vdev %i: %d\n",
+				    arvif->vdev_id, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static int ath12k_wow_clear_hw_filter(struct ath12k *ar)
+{
+	struct wmi_hw_data_filter_arg arg;
+	struct ath12k_vif *arvif;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		if (arvif->vdev_type != WMI_VDEV_TYPE_STA)
+			continue;
+
+		arg.vdev_id = arvif->vdev_id;
+		arg.enable = false;
+		arg.hw_filter_bitmap = 0;
+		ret = ath12k_wmi_hw_data_filter_cmd(ar, &arg);
+
+		if (ret) {
+			ath12k_warn(ar->ab, "failed to clear hw data filter on vdev %i: %d\n",
+				    arvif->vdev_id, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
 int ath12k_wow_op_suspend(struct ieee80211_hw *hw,
 			  struct cfg80211_wowlan *wowlan)
 {
@@ -575,6 +628,13 @@ int ath12k_wow_op_suspend(struct ieee80211_hw *hw,
 		goto cleanup;
 	}
 
+	ret = ath12k_wow_set_hw_filter(ar);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to set hw filter: %d\n",
+			    ret);
+		goto cleanup;
+	}
+
 	ret = ath12k_wow_enable(ar);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to start wow: %d\n", ret);
@@ -642,6 +702,12 @@ int ath12k_wow_op_resume(struct ieee80211_hw *hw)
 		goto exit;
 	}
 
+	ret = ath12k_wow_clear_hw_filter(ar);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to clear hw filter: %d\n", ret);
+		goto exit;
+	}
+
 exit:
 	if (ret) {
 		switch (ah->state) {
-- 
cgit v1.2.3-58-ga151


From 1666108c74c438180dd69973e57909420836fd49 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Wed, 19 Jun 2024 17:21:12 +0300
Subject: wifi: ath12k: support ARP and NS offload

Support ARP and NS offload in WoW state.

Tested this way: put machine A with QCA6390 to WoW state,
ping/ping6 machine A from another machine B, check sniffer to see
any ARP response and Neighbor Advertisement from machine A.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240604055407.12506-7-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c |   1 +
 drivers/net/wireless/ath/ath12k/wmi.c | 148 ++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/wmi.h |  66 ++++++++++++
 drivers/net/wireless/ath/ath12k/wow.c | 191 ++++++++++++++++++++++++++++++++++
 4 files changed, 406 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index cd8fb3797eea..13186496728b 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -6,6 +6,7 @@
 
 #include <net/mac80211.h>
 #include <linux/etherdevice.h>
+
 #include "mac.h"
 #include "core.h"
 #include "debug.h"
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index b1f343efa4b0..8b80d5b6bc01 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -7783,3 +7783,151 @@ int ath12k_wmi_wow_config_pno(struct ath12k *ar, u32 vdev_id,
 
 	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_NETWORK_LIST_OFFLOAD_CONFIG_CMDID);
 }
+
+static void ath12k_wmi_fill_ns_offload(struct ath12k *ar,
+				       struct wmi_arp_ns_offload_arg *offload,
+				       void **ptr,
+				       bool enable,
+				       bool ext)
+{
+	struct wmi_ns_offload_params *ns;
+	struct wmi_tlv *tlv;
+	void *buf_ptr = *ptr;
+	u32 ns_cnt, ns_ext_tuples;
+	int i, max_offloads;
+
+	ns_cnt = offload->ipv6_count;
+
+	tlv  = buf_ptr;
+
+	if (ext) {
+		ns_ext_tuples = offload->ipv6_count - WMI_MAX_NS_OFFLOADS;
+		tlv->header = ath12k_wmi_tlv_hdr(WMI_TAG_ARRAY_STRUCT,
+						 ns_ext_tuples * sizeof(*ns));
+		i = WMI_MAX_NS_OFFLOADS;
+		max_offloads = offload->ipv6_count;
+	} else {
+		tlv->header = ath12k_wmi_tlv_hdr(WMI_TAG_ARRAY_STRUCT,
+						 WMI_MAX_NS_OFFLOADS * sizeof(*ns));
+		i = 0;
+		max_offloads = WMI_MAX_NS_OFFLOADS;
+	}
+
+	buf_ptr += sizeof(*tlv);
+
+	for (; i < max_offloads; i++) {
+		ns = buf_ptr;
+		ns->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_NS_OFFLOAD_TUPLE,
+							sizeof(*ns));
+
+		if (enable) {
+			if (i < ns_cnt)
+				ns->flags |= cpu_to_le32(WMI_NSOL_FLAGS_VALID);
+
+			memcpy(ns->target_ipaddr[0], offload->ipv6_addr[i], 16);
+			memcpy(ns->solicitation_ipaddr, offload->self_ipv6_addr[i], 16);
+
+			if (offload->ipv6_type[i])
+				ns->flags |= cpu_to_le32(WMI_NSOL_FLAGS_IS_IPV6_ANYCAST);
+
+			memcpy(ns->target_mac.addr, offload->mac_addr, ETH_ALEN);
+
+			if (!is_zero_ether_addr(ns->target_mac.addr))
+				ns->flags |= cpu_to_le32(WMI_NSOL_FLAGS_MAC_VALID);
+
+			ath12k_dbg(ar->ab, ATH12K_DBG_WMI,
+				   "wmi index %d ns_solicited %pI6 target %pI6",
+				   i, ns->solicitation_ipaddr,
+				   ns->target_ipaddr[0]);
+		}
+
+		buf_ptr += sizeof(*ns);
+	}
+
+	*ptr = buf_ptr;
+}
+
+static void ath12k_wmi_fill_arp_offload(struct ath12k *ar,
+					struct wmi_arp_ns_offload_arg *offload,
+					void **ptr,
+					bool enable)
+{
+	struct wmi_arp_offload_params *arp;
+	struct wmi_tlv *tlv;
+	void *buf_ptr = *ptr;
+	int i;
+
+	/* fill arp tuple */
+	tlv = buf_ptr;
+	tlv->header = ath12k_wmi_tlv_hdr(WMI_TAG_ARRAY_STRUCT,
+					 WMI_MAX_ARP_OFFLOADS * sizeof(*arp));
+	buf_ptr += sizeof(*tlv);
+
+	for (i = 0; i < WMI_MAX_ARP_OFFLOADS; i++) {
+		arp = buf_ptr;
+		arp->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_ARP_OFFLOAD_TUPLE,
+							 sizeof(*arp));
+
+		if (enable && i < offload->ipv4_count) {
+			/* Copy the target ip addr and flags */
+			arp->flags = cpu_to_le32(WMI_ARPOL_FLAGS_VALID);
+			memcpy(arp->target_ipaddr, offload->ipv4_addr[i], 4);
+
+			ath12k_dbg(ar->ab, ATH12K_DBG_WMI, "wmi arp offload address %pI4",
+				   arp->target_ipaddr);
+		}
+
+		buf_ptr += sizeof(*arp);
+	}
+
+	*ptr = buf_ptr;
+}
+
+int ath12k_wmi_arp_ns_offload(struct ath12k *ar,
+			      struct ath12k_vif *arvif,
+			      struct wmi_arp_ns_offload_arg *offload,
+			      bool enable)
+{
+	struct wmi_set_arp_ns_offload_cmd *cmd;
+	struct wmi_tlv *tlv;
+	struct sk_buff *skb;
+	void *buf_ptr;
+	size_t len;
+	u8 ns_cnt, ns_ext_tuples = 0;
+
+	ns_cnt = offload->ipv6_count;
+
+	len = sizeof(*cmd) +
+	      sizeof(*tlv) +
+	      WMI_MAX_NS_OFFLOADS * sizeof(struct wmi_ns_offload_params) +
+	      sizeof(*tlv) +
+	      WMI_MAX_ARP_OFFLOADS * sizeof(struct wmi_arp_offload_params);
+
+	if (ns_cnt > WMI_MAX_NS_OFFLOADS) {
+		ns_ext_tuples = ns_cnt - WMI_MAX_NS_OFFLOADS;
+		len += sizeof(*tlv) +
+		       ns_ext_tuples * sizeof(struct wmi_ns_offload_params);
+	}
+
+	skb = ath12k_wmi_alloc_skb(ar->wmi->wmi_ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	buf_ptr = skb->data;
+	cmd = buf_ptr;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_SET_ARP_NS_OFFLOAD_CMD,
+						 sizeof(*cmd));
+	cmd->flags = cpu_to_le32(0);
+	cmd->vdev_id = cpu_to_le32(arvif->vdev_id);
+	cmd->num_ns_ext_tuples = cpu_to_le32(ns_ext_tuples);
+
+	buf_ptr += sizeof(*cmd);
+
+	ath12k_wmi_fill_ns_offload(ar, offload, &buf_ptr, enable, 0);
+	ath12k_wmi_fill_arp_offload(ar, offload, &buf_ptr, enable);
+
+	if (ns_ext_tuples)
+		ath12k_wmi_fill_ns_offload(ar, offload, &buf_ptr, enable, 1);
+
+	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_SET_ARP_NS_OFFLOAD_CMDID);
+}
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 62d7bb9e791d..ac0517f46b72 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -24,6 +24,7 @@
 
 struct ath12k_base;
 struct ath12k;
+struct ath12k_vif;
 
 /* There is no signed version of __le32, so for a temporary solution come
  * up with our own version. The idea is from fs/ntfs/endian.h.
@@ -5313,6 +5314,66 @@ struct wmi_hw_data_filter_arg {
 	u32 hw_filter_bitmap;
 };
 
+#define WMI_IPV6_UC_TYPE     0
+#define WMI_IPV6_AC_TYPE     1
+
+#define WMI_IPV6_MAX_COUNT   16
+#define WMI_IPV4_MAX_COUNT   2
+
+struct wmi_arp_ns_offload_arg {
+	u8  ipv4_addr[WMI_IPV4_MAX_COUNT][4];
+	u32 ipv4_count;
+	u32 ipv6_count;
+	u8  ipv6_addr[WMI_IPV6_MAX_COUNT][16];
+	u8  self_ipv6_addr[WMI_IPV6_MAX_COUNT][16];
+	u8  ipv6_type[WMI_IPV6_MAX_COUNT];
+	bool ipv6_valid[WMI_IPV6_MAX_COUNT];
+	u8  mac_addr[ETH_ALEN];
+};
+
+#define WMI_MAX_NS_OFFLOADS           2
+#define WMI_MAX_ARP_OFFLOADS          2
+
+#define WMI_ARPOL_FLAGS_VALID              BIT(0)
+#define WMI_ARPOL_FLAGS_MAC_VALID          BIT(1)
+#define WMI_ARPOL_FLAGS_REMOTE_IP_VALID    BIT(2)
+
+struct wmi_arp_offload_params {
+	__le32 tlv_header;
+	__le32 flags;
+	u8 target_ipaddr[4];
+	u8 remote_ipaddr[4];
+	struct ath12k_wmi_mac_addr_params target_mac;
+} __packed;
+
+#define WMI_NSOL_FLAGS_VALID               BIT(0)
+#define WMI_NSOL_FLAGS_MAC_VALID           BIT(1)
+#define WMI_NSOL_FLAGS_REMOTE_IP_VALID     BIT(2)
+#define WMI_NSOL_FLAGS_IS_IPV6_ANYCAST     BIT(3)
+
+#define WMI_NSOL_MAX_TARGET_IPS    2
+
+struct wmi_ns_offload_params {
+	__le32 tlv_header;
+	__le32 flags;
+	u8 target_ipaddr[WMI_NSOL_MAX_TARGET_IPS][16];
+	u8 solicitation_ipaddr[16];
+	u8 remote_ipaddr[16];
+	struct ath12k_wmi_mac_addr_params target_mac;
+} __packed;
+
+struct wmi_set_arp_ns_offload_cmd {
+	__le32 tlv_header;
+	__le32 flags;
+	__le32 vdev_id;
+	__le32 num_ns_ext_tuples;
+	/* The TLVs follow:
+	 * wmi_ns_offload_params  ns[WMI_MAX_NS_OFFLOADS];
+	 * wmi_arp_offload_params arp[WMI_MAX_ARP_OFFLOADS];
+	 * wmi_ns_offload_params  ns_ext[num_ns_ext_tuples];
+	 */
+} __packed;
+
 void ath12k_wmi_init_qcn9274(struct ath12k_base *ab,
 			     struct ath12k_wmi_resource_config_arg *config);
 void ath12k_wmi_init_wcn7850(struct ath12k_base *ab,
@@ -5478,4 +5539,9 @@ int ath12k_wmi_wow_config_pno(struct ath12k *ar, u32 vdev_id,
 			      struct wmi_pno_scan_req_arg  *pno_scan);
 int ath12k_wmi_hw_data_filter_cmd(struct ath12k *ar,
 				  struct wmi_hw_data_filter_arg *arg);
+int ath12k_wmi_arp_ns_offload(struct ath12k *ar,
+			      struct ath12k_vif *arvif,
+			      struct wmi_arp_ns_offload_arg *offload,
+			      bool enable);
+
 #endif
diff --git a/drivers/net/wireless/ath/ath12k/wow.c b/drivers/net/wireless/ath/ath12k/wow.c
index 146281244f70..8dfa7853e7a4 100644
--- a/drivers/net/wireless/ath/ath12k/wow.c
+++ b/drivers/net/wireless/ath/ath12k/wow.c
@@ -5,6 +5,10 @@
  */
 
 #include <linux/delay.h>
+#include <linux/inetdevice.h>
+#include <net/addrconf.h>
+#include <net/if_inet6.h>
+#include <net/ipv6.h>
 
 #include "mac.h"
 
@@ -599,6 +603,179 @@ static int ath12k_wow_clear_hw_filter(struct ath12k *ar)
 	return 0;
 }
 
+static void ath12k_wow_generate_ns_mc_addr(struct ath12k_base *ab,
+					   struct wmi_arp_ns_offload_arg *offload)
+{
+	int i;
+
+	for (i = 0; i < offload->ipv6_count; i++) {
+		offload->self_ipv6_addr[i][0] = 0xff;
+		offload->self_ipv6_addr[i][1] = 0x02;
+		offload->self_ipv6_addr[i][11] = 0x01;
+		offload->self_ipv6_addr[i][12] = 0xff;
+		offload->self_ipv6_addr[i][13] =
+					offload->ipv6_addr[i][13];
+		offload->self_ipv6_addr[i][14] =
+					offload->ipv6_addr[i][14];
+		offload->self_ipv6_addr[i][15] =
+					offload->ipv6_addr[i][15];
+		ath12k_dbg(ab, ATH12K_DBG_WOW, "NS solicited addr %pI6\n",
+			   offload->self_ipv6_addr[i]);
+	}
+}
+
+static void ath12k_wow_prepare_ns_offload(struct ath12k_vif *arvif,
+					  struct wmi_arp_ns_offload_arg *offload)
+{
+	struct net_device *ndev = ieee80211_vif_to_wdev(arvif->vif)->netdev;
+	struct ath12k_base *ab = arvif->ar->ab;
+	struct inet6_ifaddr *ifa6;
+	struct ifacaddr6 *ifaca6;
+	struct inet6_dev *idev;
+	u32 count = 0, scope;
+
+	if (!ndev)
+		return;
+
+	idev = in6_dev_get(ndev);
+	if (!idev)
+		return;
+
+	ath12k_dbg(ab, ATH12K_DBG_WOW, "wow prepare ns offload\n");
+
+	read_lock_bh(&idev->lock);
+
+	/* get unicast address */
+	list_for_each_entry(ifa6, &idev->addr_list, if_list) {
+		if (count >= WMI_IPV6_MAX_COUNT)
+			goto unlock;
+
+		if (ifa6->flags & IFA_F_DADFAILED)
+			continue;
+
+		scope = ipv6_addr_src_scope(&ifa6->addr);
+		if (scope != IPV6_ADDR_SCOPE_LINKLOCAL &&
+		    scope != IPV6_ADDR_SCOPE_GLOBAL) {
+			ath12k_dbg(ab, ATH12K_DBG_WOW,
+				   "Unsupported ipv6 scope: %d\n", scope);
+			continue;
+		}
+
+		memcpy(offload->ipv6_addr[count], &ifa6->addr.s6_addr,
+		       sizeof(ifa6->addr.s6_addr));
+		offload->ipv6_type[count] = WMI_IPV6_UC_TYPE;
+		ath12k_dbg(ab, ATH12K_DBG_WOW, "mac count %d ipv6 uc %pI6 scope %d\n",
+			   count, offload->ipv6_addr[count],
+			   scope);
+		count++;
+	}
+
+	/* get anycast address */
+	rcu_read_lock();
+
+	for (ifaca6 = rcu_dereference(idev->ac_list); ifaca6;
+	     ifaca6 = rcu_dereference(ifaca6->aca_next)) {
+		if (count >= WMI_IPV6_MAX_COUNT) {
+			rcu_read_unlock();
+			goto unlock;
+		}
+
+		scope = ipv6_addr_src_scope(&ifaca6->aca_addr);
+		if (scope != IPV6_ADDR_SCOPE_LINKLOCAL &&
+		    scope != IPV6_ADDR_SCOPE_GLOBAL) {
+			ath12k_dbg(ab, ATH12K_DBG_WOW,
+				   "Unsupported ipv scope: %d\n", scope);
+			continue;
+		}
+
+		memcpy(offload->ipv6_addr[count], &ifaca6->aca_addr,
+		       sizeof(ifaca6->aca_addr));
+		offload->ipv6_type[count] = WMI_IPV6_AC_TYPE;
+		ath12k_dbg(ab, ATH12K_DBG_WOW, "mac count %d ipv6 ac %pI6 scope %d\n",
+			   count, offload->ipv6_addr[count],
+			   scope);
+		count++;
+	}
+
+	rcu_read_unlock();
+
+unlock:
+	read_unlock_bh(&idev->lock);
+
+	in6_dev_put(idev);
+
+	offload->ipv6_count = count;
+	ath12k_wow_generate_ns_mc_addr(ab, offload);
+}
+
+static void ath12k_wow_prepare_arp_offload(struct ath12k_vif *arvif,
+					   struct wmi_arp_ns_offload_arg *offload)
+{
+	struct ieee80211_vif *vif = arvif->vif;
+	struct ieee80211_vif_cfg vif_cfg = vif->cfg;
+	struct ath12k_base *ab = arvif->ar->ab;
+	u32 ipv4_cnt;
+
+	ath12k_dbg(ab, ATH12K_DBG_WOW, "wow prepare arp offload\n");
+
+	ipv4_cnt = min(vif_cfg.arp_addr_cnt, WMI_IPV4_MAX_COUNT);
+	memcpy(offload->ipv4_addr, vif_cfg.arp_addr_list, ipv4_cnt * sizeof(u32));
+	offload->ipv4_count = ipv4_cnt;
+
+	ath12k_dbg(ab, ATH12K_DBG_WOW,
+		   "wow arp_addr_cnt %d vif->addr %pM, offload_addr %pI4\n",
+		   vif_cfg.arp_addr_cnt, vif->addr, offload->ipv4_addr);
+}
+
+static int ath12k_wow_arp_ns_offload(struct ath12k *ar, bool enable)
+{
+	struct wmi_arp_ns_offload_arg *offload;
+	struct ath12k_vif *arvif;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	offload = kmalloc(sizeof(*offload), GFP_KERNEL);
+	if (!offload)
+		return -ENOMEM;
+
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		if (arvif->vdev_type != WMI_VDEV_TYPE_STA)
+			continue;
+
+		memset(offload, 0, sizeof(*offload));
+
+		memcpy(offload->mac_addr, arvif->vif->addr, ETH_ALEN);
+		ath12k_wow_prepare_ns_offload(arvif, offload);
+		ath12k_wow_prepare_arp_offload(arvif, offload);
+
+		ret = ath12k_wmi_arp_ns_offload(ar, arvif, offload, enable);
+		if (ret) {
+			ath12k_warn(ar->ab, "failed to set arp ns offload vdev %i: enable %d, ret %d\n",
+				    arvif->vdev_id, enable, ret);
+			return ret;
+		}
+	}
+
+	kfree(offload);
+
+	return 0;
+}
+
+static int ath12k_wow_protocol_offload(struct ath12k *ar, bool enable)
+{
+	int ret;
+
+	ret = ath12k_wow_arp_ns_offload(ar, enable);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to offload ARP and NS %d %d\n",
+			    enable, ret);
+		return ret;
+	}
+
+	return 0;
+}
+
 int ath12k_wow_op_suspend(struct ieee80211_hw *hw,
 			  struct cfg80211_wowlan *wowlan)
 {
@@ -622,6 +799,13 @@ int ath12k_wow_op_suspend(struct ieee80211_hw *hw,
 		goto cleanup;
 	}
 
+	ret = ath12k_wow_protocol_offload(ar, true);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to set wow protocol offload events: %d\n",
+			    ret);
+		goto cleanup;
+	}
+
 	ret = ath12k_mac_wait_tx_complete(ar);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to wait tx complete: %d\n", ret);
@@ -708,6 +892,13 @@ int ath12k_wow_op_resume(struct ieee80211_hw *hw)
 		goto exit;
 	}
 
+	ret = ath12k_wow_protocol_offload(ar, false);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to clear wow protocol offload events: %d\n",
+			    ret);
+		goto exit;
+	}
+
 exit:
 	if (ret) {
 		switch (ah->state) {
-- 
cgit v1.2.3-58-ga151


From aab4ae566fa15c19efadad16a8b54306b9b37e10 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Wed, 19 Jun 2024 17:21:12 +0300
Subject: wifi: ath12k: support GTK rekey offload

Host sets GTK related info to firmware before WoW is enabled, and
gets rekey replay_count and then disables GTK rekey when WoW quits.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240604055407.12506-8-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.h |   8 +++
 drivers/net/wireless/ath/ath12k/mac.c  |  38 +++++++++++
 drivers/net/wireless/ath/ath12k/wmi.c  | 112 +++++++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/wmi.h  |  39 ++++++++++++
 drivers/net/wireless/ath/ath12k/wow.c  |  46 +++++++++++++-
 5 files changed, 242 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index 81c4bfe7792f..9de3df924a1c 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -230,6 +230,13 @@ struct ath12k_vif_cache {
 	u32 bss_conf_changed;
 };
 
+struct ath12k_rekey_data {
+	u8 kck[NL80211_KCK_LEN];
+	u8 kek[NL80211_KCK_LEN];
+	u64 replay_ctr;
+	bool enable_offload;
+};
+
 struct ath12k_vif {
 	u32 vdev_id;
 	enum wmi_vdev_type vdev_type;
@@ -286,6 +293,7 @@ struct ath12k_vif {
 	u32 punct_bitmap;
 	bool ps;
 	struct ath12k_vif_cache *cache;
+	struct ath12k_rekey_data rekey_data;
 };
 
 struct ath12k_vif_iter {
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 13186496728b..261859ad967c 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -2842,6 +2842,7 @@ static void ath12k_bss_assoc(struct ath12k *ar,
 	}
 
 	arvif->is_up = true;
+	arvif->rekey_data.enable_offload = false;
 
 	ath12k_dbg(ar->ab, ATH12K_DBG_MAC,
 		   "mac vdev %d up (associated) bssid %pM aid %d\n",
@@ -2889,6 +2890,8 @@ static void ath12k_bss_disassoc(struct ath12k *ar,
 
 	arvif->is_up = false;
 
+	memset(&arvif->rekey_data, 0, sizeof(arvif->rekey_data));
+
 	cancel_delayed_work(&arvif->connection_loss_work);
 }
 
@@ -8545,6 +8548,40 @@ exit:
 	return ret;
 }
 
+static void ath12k_mac_op_set_rekey_data(struct ieee80211_hw *hw,
+					 struct ieee80211_vif *vif,
+					 struct cfg80211_gtk_rekey_data *data)
+{
+	struct ath12k_vif *arvif = ath12k_vif_to_arvif(vif);
+	struct ath12k_rekey_data *rekey_data = &arvif->rekey_data;
+	struct ath12k_hw *ah = ath12k_hw_to_ah(hw);
+	struct ath12k *ar = ath12k_ah_to_ar(ah, 0);
+
+	ath12k_dbg(ar->ab, ATH12K_DBG_MAC, "mac set rekey data vdev %d\n",
+		   arvif->vdev_id);
+
+	mutex_lock(&ar->conf_mutex);
+
+	memcpy(rekey_data->kck, data->kck, NL80211_KCK_LEN);
+	memcpy(rekey_data->kek, data->kek, NL80211_KEK_LEN);
+
+	/* The supplicant works on big-endian, the firmware expects it on
+	 * little endian.
+	 */
+	rekey_data->replay_ctr = get_unaligned_be64(data->replay_ctr);
+
+	arvif->rekey_data.enable_offload = true;
+
+	ath12k_dbg_dump(ar->ab, ATH12K_DBG_MAC, "kck", NULL,
+			rekey_data->kck, NL80211_KCK_LEN);
+	ath12k_dbg_dump(ar->ab, ATH12K_DBG_MAC, "kek", NULL,
+			rekey_data->kck, NL80211_KEK_LEN);
+	ath12k_dbg_dump(ar->ab, ATH12K_DBG_MAC, "replay ctr", NULL,
+			&rekey_data->replay_ctr, sizeof(rekey_data->replay_ctr));
+
+	mutex_unlock(&ar->conf_mutex);
+}
+
 static const struct ieee80211_ops ath12k_ops = {
 	.tx				= ath12k_mac_op_tx,
 	.wake_tx_queue			= ieee80211_handle_wake_tx_queue,
@@ -8560,6 +8597,7 @@ static const struct ieee80211_ops ath12k_ops = {
 	.hw_scan                        = ath12k_mac_op_hw_scan,
 	.cancel_hw_scan                 = ath12k_mac_op_cancel_hw_scan,
 	.set_key                        = ath12k_mac_op_set_key,
+	.set_rekey_data	                = ath12k_mac_op_set_rekey_data,
 	.sta_state                      = ath12k_mac_op_sta_state,
 	.sta_set_txpwr			= ath12k_mac_op_sta_set_txpwr,
 	.sta_rc_update			= ath12k_mac_op_sta_rc_update,
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 8b80d5b6bc01..0e41386bca45 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -7091,6 +7091,56 @@ static void ath12k_wmi_event_wow_wakeup_host(struct ath12k_base *ab, struct sk_b
 	complete(&ab->wow.wakeup_completed);
 }
 
+static void ath12k_wmi_gtk_offload_status_event(struct ath12k_base *ab,
+						struct sk_buff *skb)
+{
+	const struct wmi_gtk_offload_status_event *ev;
+	struct ath12k_vif *arvif;
+	__be64 replay_ctr_be;
+	u64 replay_ctr;
+	const void **tb;
+	int ret;
+
+	tb = ath12k_wmi_tlv_parse_alloc(ab, skb, GFP_ATOMIC);
+	if (IS_ERR(tb)) {
+		ret = PTR_ERR(tb);
+		ath12k_warn(ab, "failed to parse tlv: %d\n", ret);
+		return;
+	}
+
+	ev = tb[WMI_TAG_GTK_OFFLOAD_STATUS_EVENT];
+	if (!ev) {
+		ath12k_warn(ab, "failed to fetch gtk offload status ev");
+		kfree(tb);
+		return;
+	}
+
+	rcu_read_lock();
+	arvif = ath12k_mac_get_arvif_by_vdev_id(ab, le32_to_cpu(ev->vdev_id));
+	if (!arvif) {
+		rcu_read_unlock();
+		ath12k_warn(ab, "failed to get arvif for vdev_id:%d\n",
+			    le32_to_cpu(ev->vdev_id));
+		kfree(tb);
+		return;
+	}
+
+	replay_ctr = le64_to_cpu(ev->replay_ctr);
+	arvif->rekey_data.replay_ctr = replay_ctr;
+	ath12k_dbg(ab, ATH12K_DBG_WMI, "wmi gtk offload event refresh_cnt %d replay_ctr %llu\n",
+		   le32_to_cpu(ev->refresh_cnt), replay_ctr);
+
+	/* supplicant expects big-endian replay counter */
+	replay_ctr_be = cpu_to_be64(replay_ctr);
+
+	ieee80211_gtk_rekey_notify(arvif->vif, arvif->bssid,
+				   (void *)&replay_ctr_be, GFP_ATOMIC);
+
+	rcu_read_unlock();
+
+	kfree(tb);
+}
+
 static void ath12k_wmi_op_rx(struct ath12k_base *ab, struct sk_buff *skb)
 {
 	struct wmi_cmd_hdr *cmd_hdr;
@@ -7214,6 +7264,9 @@ static void ath12k_wmi_op_rx(struct ath12k_base *ab, struct sk_buff *skb)
 	case WMI_WOW_WAKEUP_HOST_EVENTID:
 		ath12k_wmi_event_wow_wakeup_host(ab, skb);
 		break;
+	case WMI_GTK_OFFLOAD_STATUS_EVENTID:
+		ath12k_wmi_gtk_offload_status_event(ab, skb);
+		break;
 	/* TODO: Add remaining events */
 	default:
 		ath12k_dbg(ab, ATH12K_DBG_WMI, "Unknown eventid: 0x%x\n", id);
@@ -7931,3 +7984,62 @@ int ath12k_wmi_arp_ns_offload(struct ath12k *ar,
 
 	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_SET_ARP_NS_OFFLOAD_CMDID);
 }
+
+int ath12k_wmi_gtk_rekey_offload(struct ath12k *ar,
+				 struct ath12k_vif *arvif, bool enable)
+{
+	struct ath12k_rekey_data *rekey_data = &arvif->rekey_data;
+	struct wmi_gtk_rekey_offload_cmd *cmd;
+	struct sk_buff *skb;
+	__le64 replay_ctr;
+	int len;
+
+	len = sizeof(*cmd);
+	skb =  ath12k_wmi_alloc_skb(ar->wmi->wmi_ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_gtk_rekey_offload_cmd *)skb->data;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_GTK_OFFLOAD_CMD, sizeof(*cmd));
+	cmd->vdev_id = cpu_to_le32(arvif->vdev_id);
+
+	if (enable) {
+		cmd->flags = cpu_to_le32(GTK_OFFLOAD_ENABLE_OPCODE);
+
+		/* the length in rekey_data and cmd is equal */
+		memcpy(cmd->kck, rekey_data->kck, sizeof(cmd->kck));
+		memcpy(cmd->kek, rekey_data->kek, sizeof(cmd->kek));
+
+		replay_ctr = cpu_to_le64(rekey_data->replay_ctr);
+		memcpy(cmd->replay_ctr, &replay_ctr,
+		       sizeof(replay_ctr));
+	} else {
+		cmd->flags = cpu_to_le32(GTK_OFFLOAD_DISABLE_OPCODE);
+	}
+
+	ath12k_dbg(ar->ab, ATH12K_DBG_WMI, "offload gtk rekey vdev: %d %d\n",
+		   arvif->vdev_id, enable);
+	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_GTK_OFFLOAD_CMDID);
+}
+
+int ath12k_wmi_gtk_rekey_getinfo(struct ath12k *ar,
+				 struct ath12k_vif *arvif)
+{
+	struct wmi_gtk_rekey_offload_cmd *cmd;
+	struct sk_buff *skb;
+	int len;
+
+	len = sizeof(*cmd);
+	skb =  ath12k_wmi_alloc_skb(ar->wmi->wmi_ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_gtk_rekey_offload_cmd *)skb->data;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_GTK_OFFLOAD_CMD, sizeof(*cmd));
+	cmd->vdev_id = cpu_to_le32(arvif->vdev_id);
+	cmd->flags = cpu_to_le32(GTK_OFFLOAD_REQUEST_STATUS_OPCODE);
+
+	ath12k_dbg(ar->ab, ATH12K_DBG_WMI, "get gtk rekey vdev_id: %d\n",
+		   arvif->vdev_id);
+	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_GTK_OFFLOAD_CMDID);
+}
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index ac0517f46b72..32f74993b129 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -5374,6 +5374,41 @@ struct wmi_set_arp_ns_offload_cmd {
 	 */
 } __packed;
 
+#define GTK_OFFLOAD_OPCODE_MASK             0xFF000000
+#define GTK_OFFLOAD_ENABLE_OPCODE           0x01000000
+#define GTK_OFFLOAD_DISABLE_OPCODE          0x02000000
+#define GTK_OFFLOAD_REQUEST_STATUS_OPCODE   0x04000000
+
+#define GTK_OFFLOAD_KEK_BYTES       16
+#define GTK_OFFLOAD_KCK_BYTES       16
+#define GTK_REPLAY_COUNTER_BYTES    8
+#define WMI_MAX_KEY_LEN             32
+#define IGTK_PN_SIZE                6
+
+struct wmi_gtk_offload_status_event {
+	__le32 vdev_id;
+	__le32 flags;
+	__le32 refresh_cnt;
+	__le64 replay_ctr;
+	u8 igtk_key_index;
+	u8 igtk_key_length;
+	u8 igtk_key_rsc[IGTK_PN_SIZE];
+	u8 igtk_key[WMI_MAX_KEY_LEN];
+	u8 gtk_key_index;
+	u8 gtk_key_length;
+	u8 gtk_key_rsc[GTK_REPLAY_COUNTER_BYTES];
+	u8 gtk_key[WMI_MAX_KEY_LEN];
+} __packed;
+
+struct wmi_gtk_rekey_offload_cmd {
+	__le32 tlv_header;
+	__le32 vdev_id;
+	__le32 flags;
+	u8 kek[GTK_OFFLOAD_KEK_BYTES];
+	u8 kck[GTK_OFFLOAD_KCK_BYTES];
+	u8 replay_ctr[GTK_REPLAY_COUNTER_BYTES];
+} __packed;
+
 void ath12k_wmi_init_qcn9274(struct ath12k_base *ab,
 			     struct ath12k_wmi_resource_config_arg *config);
 void ath12k_wmi_init_wcn7850(struct ath12k_base *ab,
@@ -5543,5 +5578,9 @@ int ath12k_wmi_arp_ns_offload(struct ath12k *ar,
 			      struct ath12k_vif *arvif,
 			      struct wmi_arp_ns_offload_arg *offload,
 			      bool enable);
+int ath12k_wmi_gtk_rekey_offload(struct ath12k *ar,
+				 struct ath12k_vif *arvif, bool enable);
+int ath12k_wmi_gtk_rekey_getinfo(struct ath12k *ar,
+				 struct ath12k_vif *arvif);
 
 #endif
diff --git a/drivers/net/wireless/ath/ath12k/wow.c b/drivers/net/wireless/ath/ath12k/wow.c
index 8dfa7853e7a4..5a7001edafe0 100644
--- a/drivers/net/wireless/ath/ath12k/wow.c
+++ b/drivers/net/wireless/ath/ath12k/wow.c
@@ -21,7 +21,9 @@
 
 static const struct wiphy_wowlan_support ath12k_wowlan_support = {
 	.flags = WIPHY_WOWLAN_DISCONNECT |
-		 WIPHY_WOWLAN_MAGIC_PKT,
+		 WIPHY_WOWLAN_MAGIC_PKT |
+		 WIPHY_WOWLAN_SUPPORTS_GTK_REKEY |
+		 WIPHY_WOWLAN_GTK_REKEY_FAILURE,
 	.pattern_min_len = WOW_MIN_PATTERN_SIZE,
 	.pattern_max_len = WOW_MAX_PATTERN_SIZE,
 	.max_pkt_offset = WOW_MAX_PKT_OFFSET,
@@ -762,6 +764,41 @@ static int ath12k_wow_arp_ns_offload(struct ath12k *ar, bool enable)
 	return 0;
 }
 
+static int ath12k_gtk_rekey_offload(struct ath12k *ar, bool enable)
+{
+	struct ath12k_vif *arvif;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		if (arvif->vdev_type != WMI_VDEV_TYPE_STA ||
+		    !arvif->is_up ||
+		    !arvif->rekey_data.enable_offload)
+			continue;
+
+		/* get rekey info before disable rekey offload */
+		if (!enable) {
+			ret = ath12k_wmi_gtk_rekey_getinfo(ar, arvif);
+			if (ret) {
+				ath12k_warn(ar->ab, "failed to request rekey info vdev %i, ret %d\n",
+					    arvif->vdev_id, ret);
+				return ret;
+			}
+		}
+
+		ret = ath12k_wmi_gtk_rekey_offload(ar, arvif, enable);
+
+		if (ret) {
+			ath12k_warn(ar->ab, "failed to offload gtk reky vdev %i: enable %d, ret %d\n",
+				    arvif->vdev_id, enable, ret);
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
 static int ath12k_wow_protocol_offload(struct ath12k *ar, bool enable)
 {
 	int ret;
@@ -773,6 +810,13 @@ static int ath12k_wow_protocol_offload(struct ath12k *ar, bool enable)
 		return ret;
 	}
 
+	ret = ath12k_gtk_rekey_offload(ar, enable);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to offload gtk rekey %d %d\n",
+			    enable, ret);
+		return ret;
+	}
+
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From 7af01e5695295444d47f4ead3e27e91d8c9e60f4 Mon Sep 17 00:00:00 2001
From: Baochen Qiang <quic_bqiang@quicinc.com>
Date: Wed, 19 Jun 2024 17:21:12 +0300
Subject: wifi: ath12k: handle keepalive during WoWLAN suspend and resume

With WoWLAN enabled and after sleeping for a rather long time,
we are seeing that with some APs, it is not able to wake up
the STA though the correct wake up pattern has been configured.
This is because the host doesn't send keepalive command to
firmware, thus firmware will not send any packet to the AP and
after a specific time the AP kicks out the STA.

So enable keepalive before going to suspend and disable it after
resume back.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Baochen Qiang <quic_bqiang@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240604055407.12506-9-quic_bqiang@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 31 +++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/mac.h |  4 +++
 drivers/net/wireless/ath/ath12k/wmi.c | 38 +++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/wmi.h | 46 +++++++++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/wow.c | 34 ++++++++++++++++++++++++++
 5 files changed, 153 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 261859ad967c..43801ff5bfd8 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -9418,3 +9418,34 @@ err:
 
 	return ret;
 }
+
+int ath12k_mac_vif_set_keepalive(struct ath12k_vif *arvif,
+				 enum wmi_sta_keepalive_method method,
+				 u32 interval)
+{
+	struct wmi_sta_keepalive_arg arg = {};
+	struct ath12k *ar = arvif->ar;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	if (arvif->vdev_type != WMI_VDEV_TYPE_STA)
+		return 0;
+
+	if (!test_bit(WMI_TLV_SERVICE_STA_KEEP_ALIVE, ar->ab->wmi_ab.svc_map))
+		return 0;
+
+	arg.vdev_id = arvif->vdev_id;
+	arg.enabled = 1;
+	arg.method = method;
+	arg.interval = interval;
+
+	ret = ath12k_wmi_sta_keepalive(ar, &arg);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to set keepalive on vdev %i: %d\n",
+			    arvif->vdev_id, ret);
+		return ret;
+	}
+
+	return 0;
+}
diff --git a/drivers/net/wireless/ath/ath12k/mac.h b/drivers/net/wireless/ath/ath12k/mac.h
index 69fd282b9dd3..2c6cb144827a 100644
--- a/drivers/net/wireless/ath/ath12k/mac.h
+++ b/drivers/net/wireless/ath/ath12k/mac.h
@@ -9,6 +9,7 @@
 
 #include <net/mac80211.h>
 #include <net/cfg80211.h>
+#include "wmi.h"
 
 struct ath12k;
 struct ath12k_base;
@@ -81,5 +82,8 @@ int ath12k_mac_rfkill_config(struct ath12k *ar);
 int ath12k_mac_wait_tx_complete(struct ath12k *ar);
 void ath12k_mac_handle_beacon(struct ath12k *ar, struct sk_buff *skb);
 void ath12k_mac_handle_beacon_miss(struct ath12k *ar, u32 vdev_id);
+int ath12k_mac_vif_set_keepalive(struct ath12k_vif *arvif,
+				 enum wmi_sta_keepalive_method method,
+				 u32 interval);
 
 #endif
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 0e41386bca45..7c5976366615 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -8043,3 +8043,41 @@ int ath12k_wmi_gtk_rekey_getinfo(struct ath12k *ar,
 		   arvif->vdev_id);
 	return ath12k_wmi_cmd_send(ar->wmi, skb, WMI_GTK_OFFLOAD_CMDID);
 }
+
+int ath12k_wmi_sta_keepalive(struct ath12k *ar,
+			     const struct wmi_sta_keepalive_arg *arg)
+{
+	struct wmi_sta_keepalive_arp_resp_params *arp;
+	struct ath12k_wmi_pdev *wmi = ar->wmi;
+	struct wmi_sta_keepalive_cmd *cmd;
+	struct sk_buff *skb;
+	size_t len;
+
+	len = sizeof(*cmd) + sizeof(*arp);
+	skb = ath12k_wmi_alloc_skb(wmi->wmi_ab, len);
+	if (!skb)
+		return -ENOMEM;
+
+	cmd = (struct wmi_sta_keepalive_cmd *)skb->data;
+	cmd->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_STA_KEEPALIVE_CMD, sizeof(*cmd));
+	cmd->vdev_id = cpu_to_le32(arg->vdev_id);
+	cmd->enabled = cpu_to_le32(arg->enabled);
+	cmd->interval = cpu_to_le32(arg->interval);
+	cmd->method = cpu_to_le32(arg->method);
+
+	arp = (struct wmi_sta_keepalive_arp_resp_params *)(cmd + 1);
+	arp->tlv_header = ath12k_wmi_tlv_cmd_hdr(WMI_TAG_STA_KEEPALVE_ARP_RESPONSE,
+						 sizeof(*arp));
+	if (arg->method == WMI_STA_KEEPALIVE_METHOD_UNSOLICITED_ARP_RESPONSE ||
+	    arg->method == WMI_STA_KEEPALIVE_METHOD_GRATUITOUS_ARP_REQUEST) {
+		arp->src_ip4_addr = cpu_to_le32(arg->src_ip4_addr);
+		arp->dest_ip4_addr = cpu_to_le32(arg->dest_ip4_addr);
+		ether_addr_copy(arp->dest_mac_addr.addr, arg->dest_mac_addr);
+	}
+
+	ath12k_dbg(ar->ab, ATH12K_DBG_WMI,
+		   "wmi sta keepalive vdev %d enabled %d method %d interval %d\n",
+		   arg->vdev_id, arg->enabled, arg->method, arg->interval);
+
+	return ath12k_wmi_cmd_send(wmi, skb, WMI_STA_KEEPALIVE_CMDID);
+}
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 32f74993b129..0ac7accc45dc 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -5409,6 +5409,50 @@ struct wmi_gtk_rekey_offload_cmd {
 	u8 replay_ctr[GTK_REPLAY_COUNTER_BYTES];
 } __packed;
 
+struct wmi_sta_keepalive_cmd {
+	__le32 tlv_header;
+	__le32 vdev_id;
+	__le32 enabled;
+
+	/* WMI_STA_KEEPALIVE_METHOD_ */
+	__le32 method;
+
+	/* in seconds */
+	__le32 interval;
+
+	/* following this structure is the TLV for struct
+	 * wmi_sta_keepalive_arp_resp_params
+	 */
+} __packed;
+
+struct wmi_sta_keepalive_arp_resp_params {
+	__le32 tlv_header;
+	__le32 src_ip4_addr;
+	__le32 dest_ip4_addr;
+	struct ath12k_wmi_mac_addr_params dest_mac_addr;
+} __packed;
+
+struct wmi_sta_keepalive_arg {
+	u32 vdev_id;
+	u32 enabled;
+	u32 method;
+	u32 interval;
+	u32 src_ip4_addr;
+	u32 dest_ip4_addr;
+	const u8 dest_mac_addr[ETH_ALEN];
+};
+
+enum wmi_sta_keepalive_method {
+	WMI_STA_KEEPALIVE_METHOD_NULL_FRAME = 1,
+	WMI_STA_KEEPALIVE_METHOD_UNSOLICITED_ARP_RESPONSE = 2,
+	WMI_STA_KEEPALIVE_METHOD_ETHERNET_LOOPBACK = 3,
+	WMI_STA_KEEPALIVE_METHOD_GRATUITOUS_ARP_REQUEST = 4,
+	WMI_STA_KEEPALIVE_METHOD_MGMT_VENDOR_ACTION = 5,
+};
+
+#define WMI_STA_KEEPALIVE_INTERVAL_DEFAULT	30
+#define WMI_STA_KEEPALIVE_INTERVAL_DISABLE	0
+
 void ath12k_wmi_init_qcn9274(struct ath12k_base *ab,
 			     struct ath12k_wmi_resource_config_arg *config);
 void ath12k_wmi_init_wcn7850(struct ath12k_base *ab,
@@ -5582,5 +5626,7 @@ int ath12k_wmi_gtk_rekey_offload(struct ath12k *ar,
 				 struct ath12k_vif *arvif, bool enable);
 int ath12k_wmi_gtk_rekey_getinfo(struct ath12k *ar,
 				 struct ath12k_vif *arvif);
+int ath12k_wmi_sta_keepalive(struct ath12k *ar,
+			     const struct wmi_sta_keepalive_arg *arg);
 
 #endif
diff --git a/drivers/net/wireless/ath/ath12k/wow.c b/drivers/net/wireless/ath/ath12k/wow.c
index 5a7001edafe0..685e8e98d845 100644
--- a/drivers/net/wireless/ath/ath12k/wow.c
+++ b/drivers/net/wireless/ath/ath12k/wow.c
@@ -820,6 +820,24 @@ static int ath12k_wow_protocol_offload(struct ath12k *ar, bool enable)
 	return 0;
 }
 
+static int ath12k_wow_set_keepalive(struct ath12k *ar,
+				    enum wmi_sta_keepalive_method method,
+				    u32 interval)
+{
+	struct ath12k_vif *arvif;
+	int ret;
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		ret = ath12k_mac_vif_set_keepalive(arvif, method, interval);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 int ath12k_wow_op_suspend(struct ieee80211_hw *hw,
 			  struct cfg80211_wowlan *wowlan)
 {
@@ -863,6 +881,14 @@ int ath12k_wow_op_suspend(struct ieee80211_hw *hw,
 		goto cleanup;
 	}
 
+	ret = ath12k_wow_set_keepalive(ar,
+				       WMI_STA_KEEPALIVE_METHOD_NULL_FRAME,
+				       WMI_STA_KEEPALIVE_INTERVAL_DEFAULT);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to enable wow keepalive: %d\n", ret);
+		goto cleanup;
+	}
+
 	ret = ath12k_wow_enable(ar);
 	if (ret) {
 		ath12k_warn(ar->ab, "failed to start wow: %d\n", ret);
@@ -943,6 +969,14 @@ int ath12k_wow_op_resume(struct ieee80211_hw *hw)
 		goto exit;
 	}
 
+	ret = ath12k_wow_set_keepalive(ar,
+				       WMI_STA_KEEPALIVE_METHOD_NULL_FRAME,
+				       WMI_STA_KEEPALIVE_INTERVAL_DISABLE);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to disable wow keepalive: %d\n", ret);
+		goto exit;
+	}
+
 exit:
 	if (ret) {
 		switch (ah->state) {
-- 
cgit v1.2.3-58-ga151


From 07e4fd4c0592ad57fe4c5033393d30362dbfaa5d Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:21:51 +0200
Subject: locking/local_lock: Introduce guard definition for local_lock.

Introduce lock guard definition for local_lock_t. There are no users
yet.

Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-2-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/local_lock.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/include/linux/local_lock.h b/include/linux/local_lock.h
index e55010fa7329..82366a37f447 100644
--- a/include/linux/local_lock.h
+++ b/include/linux/local_lock.h
@@ -51,4 +51,15 @@
 #define local_unlock_irqrestore(lock, flags)			\
 	__local_unlock_irqrestore(lock, flags)
 
+DEFINE_GUARD(local_lock, local_lock_t __percpu*,
+	     local_lock(_T),
+	     local_unlock(_T))
+DEFINE_GUARD(local_lock_irq, local_lock_t __percpu*,
+	     local_lock_irq(_T),
+	     local_unlock_irq(_T))
+DEFINE_LOCK_GUARD_1(local_lock_irqsave, local_lock_t __percpu,
+		    local_lock_irqsave(_T->lock, _T->flags),
+		    local_unlock_irqrestore(_T->lock, _T->flags),
+		    unsigned long flags)
+
 #endif
-- 
cgit v1.2.3-58-ga151


From c5bcab7558220bbad8bd4863576afd1b340ce29f Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:21:52 +0200
Subject: locking/local_lock: Add local nested BH locking infrastructure.

Add local_lock_nested_bh() locking. It is based on local_lock_t and the
naming follows the preempt_disable_nested() example.

For !PREEMPT_RT + !LOCKDEP it is a per-CPU annotation for locking
assumptions based on local_bh_disable(). The macro is optimized away
during compilation.
For !PREEMPT_RT + LOCKDEP the local_lock_nested_bh() is reduced to
the usual lock-acquire plus lockdep_assert_in_softirq() - ensuring that
BH is disabled.

For PREEMPT_RT local_lock_nested_bh() acquires the specified per-CPU
lock. It does not disable CPU migration because it relies on
local_bh_disable() disabling CPU migration.
With LOCKDEP it performans the usual lockdep checks as with !PREEMPT_RT.
Due to include hell the softirq check has been moved spinlock.c.

The intention is to use this locking in places where locking of a per-CPU
variable relies on BH being disabled. Instead of treating disabled
bottom halves as a big per-CPU lock, PREEMPT_RT can use this to reduce
the locking scope to what actually needs protecting.
A side effect is that it also documents the protection scope of the
per-CPU variables.

Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-3-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/local_lock.h          | 10 ++++++++++
 include/linux/local_lock_internal.h | 31 +++++++++++++++++++++++++++++++
 include/linux/lockdep.h             |  3 +++
 kernel/locking/spinlock.c           |  8 ++++++++
 4 files changed, 52 insertions(+)

diff --git a/include/linux/local_lock.h b/include/linux/local_lock.h
index 82366a37f447..091dc0b6bdfb 100644
--- a/include/linux/local_lock.h
+++ b/include/linux/local_lock.h
@@ -62,4 +62,14 @@ DEFINE_LOCK_GUARD_1(local_lock_irqsave, local_lock_t __percpu,
 		    local_unlock_irqrestore(_T->lock, _T->flags),
 		    unsigned long flags)
 
+#define local_lock_nested_bh(_lock)				\
+	__local_lock_nested_bh(_lock)
+
+#define local_unlock_nested_bh(_lock)				\
+	__local_unlock_nested_bh(_lock)
+
+DEFINE_GUARD(local_lock_nested_bh, local_lock_t __percpu*,
+	     local_lock_nested_bh(_T),
+	     local_unlock_nested_bh(_T))
+
 #endif
diff --git a/include/linux/local_lock_internal.h b/include/linux/local_lock_internal.h
index 975e33b793a7..8dd71fbbb6d2 100644
--- a/include/linux/local_lock_internal.h
+++ b/include/linux/local_lock_internal.h
@@ -62,6 +62,17 @@ do {								\
 	local_lock_debug_init(lock);				\
 } while (0)
 
+#define __spinlock_nested_bh_init(lock)				\
+do {								\
+	static struct lock_class_key __key;			\
+								\
+	debug_check_no_locks_freed((void *)lock, sizeof(*lock));\
+	lockdep_init_map_type(&(lock)->dep_map, #lock, &__key,  \
+			      0, LD_WAIT_CONFIG, LD_WAIT_INV,	\
+			      LD_LOCK_NORMAL);			\
+	local_lock_debug_init(lock);				\
+} while (0)
+
 #define __local_lock(lock)					\
 	do {							\
 		preempt_disable();				\
@@ -98,6 +109,15 @@ do {								\
 		local_irq_restore(flags);			\
 	} while (0)
 
+#define __local_lock_nested_bh(lock)				\
+	do {							\
+		lockdep_assert_in_softirq();			\
+		local_lock_acquire(this_cpu_ptr(lock));	\
+	} while (0)
+
+#define __local_unlock_nested_bh(lock)				\
+	local_lock_release(this_cpu_ptr(lock))
+
 #else /* !CONFIG_PREEMPT_RT */
 
 /*
@@ -138,4 +158,15 @@ typedef spinlock_t local_lock_t;
 
 #define __local_unlock_irqrestore(lock, flags)	__local_unlock(lock)
 
+#define __local_lock_nested_bh(lock)				\
+do {								\
+	lockdep_assert_in_softirq_func();			\
+	spin_lock(this_cpu_ptr(lock));				\
+} while (0)
+
+#define __local_unlock_nested_bh(lock)				\
+do {								\
+	spin_unlock(this_cpu_ptr((lock)));			\
+} while (0)
+
 #endif /* CONFIG_PREEMPT_RT */
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 08b0d1d9d78b..3f5a551579cc 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -600,6 +600,8 @@ do {									\
 		     (!in_softirq() || in_irq() || in_nmi()));		\
 } while (0)
 
+extern void lockdep_assert_in_softirq_func(void);
+
 #else
 # define might_lock(lock) do { } while (0)
 # define might_lock_read(lock) do { } while (0)
@@ -613,6 +615,7 @@ do {									\
 # define lockdep_assert_preemption_enabled() do { } while (0)
 # define lockdep_assert_preemption_disabled() do { } while (0)
 # define lockdep_assert_in_softirq() do { } while (0)
+# define lockdep_assert_in_softirq_func() do { } while (0)
 #endif
 
 #ifdef CONFIG_PROVE_RAW_LOCK_NESTING
diff --git a/kernel/locking/spinlock.c b/kernel/locking/spinlock.c
index 8475a0794f8c..438c6086d540 100644
--- a/kernel/locking/spinlock.c
+++ b/kernel/locking/spinlock.c
@@ -413,3 +413,11 @@ notrace int in_lock_functions(unsigned long addr)
 	&& addr < (unsigned long)__lock_text_end;
 }
 EXPORT_SYMBOL(in_lock_functions);
+
+#if defined(CONFIG_PROVE_LOCKING) && defined(CONFIG_PREEMPT_RT)
+void notrace lockdep_assert_in_softirq_func(void)
+{
+	lockdep_assert_in_softirq();
+}
+EXPORT_SYMBOL(lockdep_assert_in_softirq_func);
+#endif
-- 
cgit v1.2.3-58-ga151


From 43d7ca2907cb0e8a23d73934a5cf000e24b0aa22 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:21:53 +0200
Subject: net: Use __napi_alloc_frag_align() instead of open coding it.

The else condition within __netdev_alloc_frag_align() is an open coded
__napi_alloc_frag_align().

Use __napi_alloc_frag_align() instead of open coding it.
Move fragsz assignment before page_frag_alloc_align() invocation because
__napi_alloc_frag_align() also contains this statement.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-4-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/skbuff.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 2315c088e91d..1b52f69ad05e 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -318,19 +318,15 @@ void *__netdev_alloc_frag_align(unsigned int fragsz, unsigned int align_mask)
 {
 	void *data;
 
-	fragsz = SKB_DATA_ALIGN(fragsz);
 	if (in_hardirq() || irqs_disabled()) {
 		struct page_frag_cache *nc = this_cpu_ptr(&netdev_alloc_cache);
 
+		fragsz = SKB_DATA_ALIGN(fragsz);
 		data = __page_frag_alloc_align(nc, fragsz, GFP_ATOMIC,
 					       align_mask);
 	} else {
-		struct napi_alloc_cache *nc;
-
 		local_bh_disable();
-		nc = this_cpu_ptr(&napi_alloc_cache);
-		data = __page_frag_alloc_align(&nc->page, fragsz, GFP_ATOMIC,
-					       align_mask);
+		data = __napi_alloc_frag_align(fragsz, align_mask);
 		local_bh_enable();
 	}
 	return data;
-- 
cgit v1.2.3-58-ga151


From bdacf3e34945232037979b977d7f409f734a7226 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:21:54 +0200
Subject: net: Use nested-BH locking for napi_alloc_cache.

napi_alloc_cache is a per-CPU variable and relies on disabled BH for its
locking. Without per-CPU locking in local_bh_disable() on PREEMPT_RT
this data structure requires explicit locking.

Add a local_lock_t to the data structure and use local_lock_nested_bh()
for locking. This change adds only lockdep coverage and does not alter
the functional behaviour for !PREEMPT_RT.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-5-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/skbuff.c | 29 ++++++++++++++++++++++++-----
 1 file changed, 24 insertions(+), 5 deletions(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 1b52f69ad05e..eb9a7e65b5c8 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -277,6 +277,7 @@ static void *page_frag_alloc_1k(struct page_frag_1k *nc, gfp_t gfp_mask)
 #endif
 
 struct napi_alloc_cache {
+	local_lock_t bh_lock;
 	struct page_frag_cache page;
 	struct page_frag_1k page_small;
 	unsigned int skb_count;
@@ -284,7 +285,9 @@ struct napi_alloc_cache {
 };
 
 static DEFINE_PER_CPU(struct page_frag_cache, netdev_alloc_cache);
-static DEFINE_PER_CPU(struct napi_alloc_cache, napi_alloc_cache);
+static DEFINE_PER_CPU(struct napi_alloc_cache, napi_alloc_cache) = {
+	.bh_lock = INIT_LOCAL_LOCK(bh_lock),
+};
 
 /* Double check that napi_get_frags() allocates skbs with
  * skb->head being backed by slab, not a page fragment.
@@ -306,11 +309,16 @@ void napi_get_frags_check(struct napi_struct *napi)
 void *__napi_alloc_frag_align(unsigned int fragsz, unsigned int align_mask)
 {
 	struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache);
+	void *data;
 
 	fragsz = SKB_DATA_ALIGN(fragsz);
 
-	return __page_frag_alloc_align(&nc->page, fragsz, GFP_ATOMIC,
+	local_lock_nested_bh(&napi_alloc_cache.bh_lock);
+	data = __page_frag_alloc_align(&nc->page, fragsz, GFP_ATOMIC,
 				       align_mask);
+	local_unlock_nested_bh(&napi_alloc_cache.bh_lock);
+	return data;
+
 }
 EXPORT_SYMBOL(__napi_alloc_frag_align);
 
@@ -338,16 +346,20 @@ static struct sk_buff *napi_skb_cache_get(void)
 	struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache);
 	struct sk_buff *skb;
 
+	local_lock_nested_bh(&napi_alloc_cache.bh_lock);
 	if (unlikely(!nc->skb_count)) {
 		nc->skb_count = kmem_cache_alloc_bulk(net_hotdata.skbuff_cache,
 						      GFP_ATOMIC,
 						      NAPI_SKB_CACHE_BULK,
 						      nc->skb_cache);
-		if (unlikely(!nc->skb_count))
+		if (unlikely(!nc->skb_count)) {
+			local_unlock_nested_bh(&napi_alloc_cache.bh_lock);
 			return NULL;
+		}
 	}
 
 	skb = nc->skb_cache[--nc->skb_count];
+	local_unlock_nested_bh(&napi_alloc_cache.bh_lock);
 	kasan_mempool_unpoison_object(skb, kmem_cache_size(net_hotdata.skbuff_cache));
 
 	return skb;
@@ -740,9 +752,13 @@ struct sk_buff *__netdev_alloc_skb(struct net_device *dev, unsigned int len,
 		pfmemalloc = nc->pfmemalloc;
 	} else {
 		local_bh_disable();
+		local_lock_nested_bh(&napi_alloc_cache.bh_lock);
+
 		nc = this_cpu_ptr(&napi_alloc_cache.page);
 		data = page_frag_alloc(nc, len, gfp_mask);
 		pfmemalloc = nc->pfmemalloc;
+
+		local_unlock_nested_bh(&napi_alloc_cache.bh_lock);
 		local_bh_enable();
 	}
 
@@ -806,11 +822,11 @@ struct sk_buff *napi_alloc_skb(struct napi_struct *napi, unsigned int len)
 		goto skb_success;
 	}
 
-	nc = this_cpu_ptr(&napi_alloc_cache);
-
 	if (sk_memalloc_socks())
 		gfp_mask |= __GFP_MEMALLOC;
 
+	local_lock_nested_bh(&napi_alloc_cache.bh_lock);
+	nc = this_cpu_ptr(&napi_alloc_cache);
 	if (NAPI_HAS_SMALL_PAGE_FRAG && len <= SKB_WITH_OVERHEAD(1024)) {
 		/* we are artificially inflating the allocation size, but
 		 * that is not as bad as it may look like, as:
@@ -832,6 +848,7 @@ struct sk_buff *napi_alloc_skb(struct napi_struct *napi, unsigned int len)
 		data = page_frag_alloc(&nc->page, len, gfp_mask);
 		pfmemalloc = nc->page.pfmemalloc;
 	}
+	local_unlock_nested_bh(&napi_alloc_cache.bh_lock);
 
 	if (unlikely(!data))
 		return NULL;
@@ -1431,6 +1448,7 @@ static void napi_skb_cache_put(struct sk_buff *skb)
 	if (!kasan_mempool_poison_object(skb))
 		return;
 
+	local_lock_nested_bh(&napi_alloc_cache.bh_lock);
 	nc->skb_cache[nc->skb_count++] = skb;
 
 	if (unlikely(nc->skb_count == NAPI_SKB_CACHE_SIZE)) {
@@ -1442,6 +1460,7 @@ static void napi_skb_cache_put(struct sk_buff *skb)
 				     nc->skb_cache + NAPI_SKB_CACHE_HALF);
 		nc->skb_count = NAPI_SKB_CACHE_HALF;
 	}
+	local_unlock_nested_bh(&napi_alloc_cache.bh_lock);
 }
 
 void __napi_kfree_skb(struct sk_buff *skb, enum skb_drop_reason reason)
-- 
cgit v1.2.3-58-ga151


From 585aa621af6cd11d3d6f830965af3e45ee79ea51 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:21:55 +0200
Subject: net/tcp_sigpool: Use nested-BH locking for sigpool_scratch.

sigpool_scratch is a per-CPU variable and relies on disabled BH for its
locking. Without per-CPU locking in local_bh_disable() on PREEMPT_RT
this data structure requires explicit locking.

Make a struct with a pad member (original sigpool_scratch) and a
local_lock_t and use local_lock_nested_bh() for locking. This change
adds only lockdep coverage and does not alter the functional behaviour
for !PREEMPT_RT.

Cc: David Ahern <dsahern@kernel.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-6-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/tcp_sigpool.c | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/net/ipv4/tcp_sigpool.c b/net/ipv4/tcp_sigpool.c
index 8512cb09ebc0..d8a4f192873a 100644
--- a/net/ipv4/tcp_sigpool.c
+++ b/net/ipv4/tcp_sigpool.c
@@ -10,7 +10,14 @@
 #include <net/tcp.h>
 
 static size_t __scratch_size;
-static DEFINE_PER_CPU(void __rcu *, sigpool_scratch);
+struct sigpool_scratch {
+	local_lock_t bh_lock;
+	void __rcu *pad;
+};
+
+static DEFINE_PER_CPU(struct sigpool_scratch, sigpool_scratch) = {
+	.bh_lock = INIT_LOCAL_LOCK(bh_lock),
+};
 
 struct sigpool_entry {
 	struct crypto_ahash	*hash;
@@ -72,7 +79,7 @@ static int sigpool_reserve_scratch(size_t size)
 			break;
 		}
 
-		old_scratch = rcu_replace_pointer(per_cpu(sigpool_scratch, cpu),
+		old_scratch = rcu_replace_pointer(per_cpu(sigpool_scratch.pad, cpu),
 					scratch, lockdep_is_held(&cpool_mutex));
 		if (!cpu_online(cpu) || !old_scratch) {
 			kfree(old_scratch);
@@ -93,7 +100,7 @@ static void sigpool_scratch_free(void)
 	int cpu;
 
 	for_each_possible_cpu(cpu)
-		kfree(rcu_replace_pointer(per_cpu(sigpool_scratch, cpu),
+		kfree(rcu_replace_pointer(per_cpu(sigpool_scratch.pad, cpu),
 					  NULL, lockdep_is_held(&cpool_mutex)));
 	__scratch_size = 0;
 }
@@ -277,7 +284,8 @@ int tcp_sigpool_start(unsigned int id, struct tcp_sigpool *c) __cond_acquires(RC
 	/* Pairs with tcp_sigpool_reserve_scratch(), scratch area is
 	 * valid (allocated) until tcp_sigpool_end().
 	 */
-	c->scratch = rcu_dereference_bh(*this_cpu_ptr(&sigpool_scratch));
+	local_lock_nested_bh(&sigpool_scratch.bh_lock);
+	c->scratch = rcu_dereference_bh(*this_cpu_ptr(&sigpool_scratch.pad));
 	return 0;
 }
 EXPORT_SYMBOL_GPL(tcp_sigpool_start);
@@ -286,6 +294,7 @@ void tcp_sigpool_end(struct tcp_sigpool *c) __releases(RCU_BH)
 {
 	struct crypto_ahash *hash = crypto_ahash_reqtfm(c->req);
 
+	local_unlock_nested_bh(&sigpool_scratch.bh_lock);
 	rcu_read_unlock_bh();
 	ahash_request_free(c->req);
 	crypto_free_ahash(hash);
-- 
cgit v1.2.3-58-ga151


From ebad6d0334793f16a16e5320182f665292a31e0c Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:21:56 +0200
Subject: net/ipv4: Use nested-BH locking for ipv4_tcp_sk.

ipv4_tcp_sk is a per-CPU variable and relies on disabled BH for its
locking. Without per-CPU locking in local_bh_disable() on PREEMPT_RT
this data structure requires explicit locking.

Make a struct with a sock member (original ipv4_tcp_sk) and a
local_lock_t and use local_lock_nested_bh() for locking. This change
adds only lockdep coverage and does not alter the functional behaviour
for !PREEMPT_RT.

Cc: David Ahern <dsahern@kernel.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-7-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/sock.h  |  5 +++++
 net/ipv4/tcp_ipv4.c | 15 +++++++++++----
 2 files changed, 16 insertions(+), 4 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index b30ea0c342a6..cce23ac4d514 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -544,6 +544,11 @@ struct sock {
 	netns_tracker		ns_tracker;
 };
 
+struct sock_bh_locked {
+	struct sock *sock;
+	local_lock_t bh_lock;
+};
+
 enum sk_pacing {
 	SK_PACING_NONE		= 0,
 	SK_PACING_NEEDED	= 1,
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 8e49d69279d5..fd17f25ff288 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -93,7 +93,9 @@ static int tcp_v4_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key,
 struct inet_hashinfo tcp_hashinfo;
 EXPORT_SYMBOL(tcp_hashinfo);
 
-static DEFINE_PER_CPU(struct sock *, ipv4_tcp_sk);
+static DEFINE_PER_CPU(struct sock_bh_locked, ipv4_tcp_sk) = {
+	.bh_lock = INIT_LOCAL_LOCK(bh_lock),
+};
 
 static u32 tcp_v4_init_seq(const struct sk_buff *skb)
 {
@@ -882,7 +884,9 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb,
 	arg.tos = ip_hdr(skb)->tos;
 	arg.uid = sock_net_uid(net, sk && sk_fullsock(sk) ? sk : NULL);
 	local_bh_disable();
-	ctl_sk = this_cpu_read(ipv4_tcp_sk);
+	local_lock_nested_bh(&ipv4_tcp_sk.bh_lock);
+	ctl_sk = this_cpu_read(ipv4_tcp_sk.sock);
+
 	sock_net_set(ctl_sk, net);
 	if (sk) {
 		ctl_sk->sk_mark = (sk->sk_state == TCP_TIME_WAIT) ?
@@ -907,6 +911,7 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb,
 	sock_net_set(ctl_sk, &init_net);
 	__TCP_INC_STATS(net, TCP_MIB_OUTSEGS);
 	__TCP_INC_STATS(net, TCP_MIB_OUTRSTS);
+	local_unlock_nested_bh(&ipv4_tcp_sk.bh_lock);
 	local_bh_enable();
 
 #ifdef CONFIG_TCP_MD5SIG
@@ -1002,7 +1007,8 @@ static void tcp_v4_send_ack(const struct sock *sk,
 	arg.tos = tos;
 	arg.uid = sock_net_uid(net, sk_fullsock(sk) ? sk : NULL);
 	local_bh_disable();
-	ctl_sk = this_cpu_read(ipv4_tcp_sk);
+	local_lock_nested_bh(&ipv4_tcp_sk.bh_lock);
+	ctl_sk = this_cpu_read(ipv4_tcp_sk.sock);
 	sock_net_set(ctl_sk, net);
 	ctl_sk->sk_mark = (sk->sk_state == TCP_TIME_WAIT) ?
 			   inet_twsk(sk)->tw_mark : READ_ONCE(sk->sk_mark);
@@ -1017,6 +1023,7 @@ static void tcp_v4_send_ack(const struct sock *sk,
 
 	sock_net_set(ctl_sk, &init_net);
 	__TCP_INC_STATS(net, TCP_MIB_OUTSEGS);
+	local_unlock_nested_bh(&ipv4_tcp_sk.bh_lock);
 	local_bh_enable();
 }
 
@@ -3615,7 +3622,7 @@ void __init tcp_v4_init(void)
 
 		sk->sk_clockid = CLOCK_MONOTONIC;
 
-		per_cpu(ipv4_tcp_sk, cpu) = sk;
+		per_cpu(ipv4_tcp_sk.sock, cpu) = sk;
 	}
 	if (register_pernet_subsys(&tcp_sk_ops))
 		panic("Failed to create the TCP control socket.\n");
-- 
cgit v1.2.3-58-ga151


From c67ef53a88db5bcc29a6da20dcbc09206d49d0a9 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:21:57 +0200
Subject: netfilter: br_netfilter: Use nested-BH locking for
 brnf_frag_data_storage.

brnf_frag_data_storage is a per-CPU variable and relies on disabled BH
for its locking. Without per-CPU locking in local_bh_disable() on
PREEMPT_RT this data structure requires explicit locking.

Add a local_lock_t to the data structure and use local_lock_nested_bh()
for locking. This change adds only lockdep coverage and does not alter
the functional behaviour for !PREEMPT_RT.

Cc: Florian Westphal <fw@strlen.de>
Cc: Jozsef Kadlecsik <kadlec@netfilter.org>
Cc: Nikolay Aleksandrov <razor@blackwall.org>
Cc: Pablo Neira Ayuso <pablo@netfilter.org>
Cc: Roopa Prabhu <roopa@nvidia.com>
Cc: bridge@lists.linux.dev
Cc: coreteam@netfilter.org
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-8-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/bridge/br_netfilter_hooks.c | 20 ++++++++++++++++----
 1 file changed, 16 insertions(+), 4 deletions(-)

diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c
index bf30c50b5689..3c9f6538990e 100644
--- a/net/bridge/br_netfilter_hooks.c
+++ b/net/bridge/br_netfilter_hooks.c
@@ -137,6 +137,7 @@ static inline bool is_pppoe_ipv6(const struct sk_buff *skb,
 #define NF_BRIDGE_MAX_MAC_HEADER_LENGTH (PPPOE_SES_HLEN + ETH_HLEN)
 
 struct brnf_frag_data {
+	local_lock_t bh_lock;
 	char mac[NF_BRIDGE_MAX_MAC_HEADER_LENGTH];
 	u8 encap_size;
 	u8 size;
@@ -144,7 +145,9 @@ struct brnf_frag_data {
 	__be16 vlan_proto;
 };
 
-static DEFINE_PER_CPU(struct brnf_frag_data, brnf_frag_data_storage);
+static DEFINE_PER_CPU(struct brnf_frag_data, brnf_frag_data_storage) = {
+	.bh_lock = INIT_LOCAL_LOCK(bh_lock),
+};
 
 static void nf_bridge_info_free(struct sk_buff *skb)
 {
@@ -850,6 +853,7 @@ static int br_nf_dev_queue_xmit(struct net *net, struct sock *sk, struct sk_buff
 {
 	struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb);
 	unsigned int mtu, mtu_reserved;
+	int ret;
 
 	mtu_reserved = nf_bridge_mtu_reduction(skb);
 	mtu = skb->dev->mtu;
@@ -882,6 +886,7 @@ static int br_nf_dev_queue_xmit(struct net *net, struct sock *sk, struct sk_buff
 
 		IPCB(skb)->frag_max_size = nf_bridge->frag_max_size;
 
+		local_lock_nested_bh(&brnf_frag_data_storage.bh_lock);
 		data = this_cpu_ptr(&brnf_frag_data_storage);
 
 		if (skb_vlan_tag_present(skb)) {
@@ -897,7 +902,9 @@ static int br_nf_dev_queue_xmit(struct net *net, struct sock *sk, struct sk_buff
 		skb_copy_from_linear_data_offset(skb, -data->size, data->mac,
 						 data->size);
 
-		return br_nf_ip_fragment(net, sk, skb, br_nf_push_frag_xmit);
+		ret = br_nf_ip_fragment(net, sk, skb, br_nf_push_frag_xmit);
+		local_unlock_nested_bh(&brnf_frag_data_storage.bh_lock);
+		return ret;
 	}
 	if (IS_ENABLED(CONFIG_NF_DEFRAG_IPV6) &&
 	    skb->protocol == htons(ETH_P_IPV6)) {
@@ -909,6 +916,7 @@ static int br_nf_dev_queue_xmit(struct net *net, struct sock *sk, struct sk_buff
 
 		IP6CB(skb)->frag_max_size = nf_bridge->frag_max_size;
 
+		local_lock_nested_bh(&brnf_frag_data_storage.bh_lock);
 		data = this_cpu_ptr(&brnf_frag_data_storage);
 		data->encap_size = nf_bridge_encap_header_len(skb);
 		data->size = ETH_HLEN + data->encap_size;
@@ -916,8 +924,12 @@ static int br_nf_dev_queue_xmit(struct net *net, struct sock *sk, struct sk_buff
 		skb_copy_from_linear_data_offset(skb, -data->size, data->mac,
 						 data->size);
 
-		if (v6ops)
-			return v6ops->fragment(net, sk, skb, br_nf_push_frag_xmit);
+		if (v6ops) {
+			ret = v6ops->fragment(net, sk, skb, br_nf_push_frag_xmit);
+			local_unlock_nested_bh(&brnf_frag_data_storage.bh_lock);
+			return ret;
+		}
+		local_unlock_nested_bh(&brnf_frag_data_storage.bh_lock);
 
 		kfree_skb(skb);
 		return -EMSGSIZE;
-- 
cgit v1.2.3-58-ga151


From ecefbc09e8ee768ae85b7bb7a1de8c8287397d68 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:21:58 +0200
Subject: net: softnet_data: Make xmit per task.

Softirq is preemptible on PREEMPT_RT. Without a per-CPU lock in
local_bh_disable() there is no guarantee that only one device is
transmitting at a time.
With preemption and multiple senders it is possible that the per-CPU
`recursion' counter gets incremented by different threads and exceeds
XMIT_RECURSION_LIMIT leading to a false positive recursion alert.
The `more' member is subject to similar problems if set by one thread
for one driver and wrongly used by another driver within another thread.

Instead of adding a lock to protect the per-CPU variable it is simpler
to make xmit per-task. Sending and receiving skbs happens always
in thread context anyway.

Having a lock to protected the per-CPU counter would block/ serialize two
sending threads needlessly. It would also require a recursive lock to
ensure that the owner can increment the counter further.

Make the softnet_data.xmit a task_struct member on PREEMPT_RT. Add
needed wrapper.

Cc: Ben Segall <bsegall@google.com>
Cc: Daniel Bristot de Oliveira <bristot@redhat.com>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Valentin Schneider <vschneid@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-9-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/netdevice.h      | 42 +++++++++++++++++++++++++++++++-----------
 include/linux/netdevice_xmit.h | 13 +++++++++++++
 include/linux/sched.h          |  5 ++++-
 net/core/dev.c                 | 14 ++++++++++++++
 net/core/dev.h                 | 18 ++++++++++++++++++
 5 files changed, 80 insertions(+), 12 deletions(-)
 create mode 100644 include/linux/netdevice_xmit.h

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c83b390191d4..f6fc9066147d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -43,6 +43,7 @@
 
 #include <linux/netdev_features.h>
 #include <linux/neighbour.h>
+#include <linux/netdevice_xmit.h>
 #include <uapi/linux/netdevice.h>
 #include <uapi/linux/if_bonding.h>
 #include <uapi/linux/pkt_cls.h>
@@ -3223,13 +3224,7 @@ struct softnet_data {
 	struct sk_buff_head	xfrm_backlog;
 #endif
 	/* written and read only by owning cpu: */
-	struct {
-		u16 recursion;
-		u8  more;
-#ifdef CONFIG_NET_EGRESS
-		u8  skip_txqueue;
-#endif
-	} xmit;
+	struct netdev_xmit xmit;
 #ifdef CONFIG_RPS
 	/* input_queue_head should be written by cpu owning this struct,
 	 * and only read by other cpus. Worth using a cache line.
@@ -3257,10 +3252,18 @@ struct softnet_data {
 
 DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data);
 
+#ifndef CONFIG_PREEMPT_RT
 static inline int dev_recursion_level(void)
 {
 	return this_cpu_read(softnet_data.xmit.recursion);
 }
+#else
+static inline int dev_recursion_level(void)
+{
+	return current->net_xmit.recursion;
+}
+
+#endif
 
 void __netif_schedule(struct Qdisc *q);
 void netif_schedule_queue(struct netdev_queue *txq);
@@ -4872,18 +4875,35 @@ static inline ktime_t netdev_get_tstamp(struct net_device *dev,
 	return hwtstamps->hwtstamp;
 }
 
-static inline netdev_tx_t __netdev_start_xmit(const struct net_device_ops *ops,
-					      struct sk_buff *skb, struct net_device *dev,
-					      bool more)
+#ifndef CONFIG_PREEMPT_RT
+static inline void netdev_xmit_set_more(bool more)
 {
 	__this_cpu_write(softnet_data.xmit.more, more);
-	return ops->ndo_start_xmit(skb, dev);
 }
 
 static inline bool netdev_xmit_more(void)
 {
 	return __this_cpu_read(softnet_data.xmit.more);
 }
+#else
+static inline void netdev_xmit_set_more(bool more)
+{
+	current->net_xmit.more = more;
+}
+
+static inline bool netdev_xmit_more(void)
+{
+	return current->net_xmit.more;
+}
+#endif
+
+static inline netdev_tx_t __netdev_start_xmit(const struct net_device_ops *ops,
+					      struct sk_buff *skb, struct net_device *dev,
+					      bool more)
+{
+	netdev_xmit_set_more(more);
+	return ops->ndo_start_xmit(skb, dev);
+}
 
 static inline netdev_tx_t netdev_start_xmit(struct sk_buff *skb, struct net_device *dev,
 					    struct netdev_queue *txq, bool more)
diff --git a/include/linux/netdevice_xmit.h b/include/linux/netdevice_xmit.h
new file mode 100644
index 000000000000..38325e070296
--- /dev/null
+++ b/include/linux/netdevice_xmit.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+#ifndef _LINUX_NETDEVICE_XMIT_H
+#define _LINUX_NETDEVICE_XMIT_H
+
+struct netdev_xmit {
+	u16 recursion;
+	u8  more;
+#ifdef CONFIG_NET_EGRESS
+	u8  skip_txqueue;
+#endif
+};
+
+#endif
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 61591ac6eab6..5187486c2522 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -36,6 +36,7 @@
 #include <linux/signal_types.h>
 #include <linux/syscall_user_dispatch_types.h>
 #include <linux/mm_types_task.h>
+#include <linux/netdevice_xmit.h>
 #include <linux/task_io_accounting.h>
 #include <linux/posix-timers_types.h>
 #include <linux/restart_block.h>
@@ -975,7 +976,9 @@ struct task_struct {
 	/* delay due to memory thrashing */
 	unsigned                        in_thrashing:1;
 #endif
-
+#ifdef CONFIG_PREEMPT_RT
+	struct netdev_xmit		net_xmit;
+#endif
 	unsigned long			atomic_flags; /* Flags requiring atomic access. */
 
 	struct restart_block		restart_block;
diff --git a/net/core/dev.c b/net/core/dev.c
index 093d82bf0e28..95b9e4cc1767 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3940,6 +3940,7 @@ netdev_tx_queue_mapping(struct net_device *dev, struct sk_buff *skb)
 	return netdev_get_tx_queue(dev, netdev_cap_txqueue(dev, qm));
 }
 
+#ifndef CONFIG_PREEMPT_RT
 static bool netdev_xmit_txqueue_skipped(void)
 {
 	return __this_cpu_read(softnet_data.xmit.skip_txqueue);
@@ -3950,6 +3951,19 @@ void netdev_xmit_skip_txqueue(bool skip)
 	__this_cpu_write(softnet_data.xmit.skip_txqueue, skip);
 }
 EXPORT_SYMBOL_GPL(netdev_xmit_skip_txqueue);
+
+#else
+static bool netdev_xmit_txqueue_skipped(void)
+{
+	return current->net_xmit.skip_txqueue;
+}
+
+void netdev_xmit_skip_txqueue(bool skip)
+{
+	current->net_xmit.skip_txqueue = skip;
+}
+EXPORT_SYMBOL_GPL(netdev_xmit_skip_txqueue);
+#endif
 #endif /* CONFIG_NET_EGRESS */
 
 #ifdef CONFIG_NET_XGRESS
diff --git a/net/core/dev.h b/net/core/dev.h
index 58f88d28bc99..5654325c5b71 100644
--- a/net/core/dev.h
+++ b/net/core/dev.h
@@ -150,6 +150,8 @@ struct napi_struct *napi_by_id(unsigned int napi_id);
 void kick_defer_list_purge(struct softnet_data *sd, unsigned int cpu);
 
 #define XMIT_RECURSION_LIMIT	8
+
+#ifndef CONFIG_PREEMPT_RT
 static inline bool dev_xmit_recursion(void)
 {
 	return unlikely(__this_cpu_read(softnet_data.xmit.recursion) >
@@ -165,6 +167,22 @@ static inline void dev_xmit_recursion_dec(void)
 {
 	__this_cpu_dec(softnet_data.xmit.recursion);
 }
+#else
+static inline bool dev_xmit_recursion(void)
+{
+	return unlikely(current->net_xmit.recursion > XMIT_RECURSION_LIMIT);
+}
+
+static inline void dev_xmit_recursion_inc(void)
+{
+	current->net_xmit.recursion++;
+}
+
+static inline void dev_xmit_recursion_dec(void)
+{
+	current->net_xmit.recursion--;
+}
+#endif
 
 int dev_set_hwtstamp_phylib(struct net_device *dev,
 			    struct kernel_hwtstamp_config *cfg,
-- 
cgit v1.2.3-58-ga151


From a8760d0d1497878d01f901ef2038c7389187a8af Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:21:59 +0200
Subject: dev: Remove PREEMPT_RT ifdefs from backlog_lock.*().

The backlog_napi locking (previously RPS) relies on explicit locking if
either RPS or backlog NAPI is enabled. If both are disabled then locking
was achieved by disabling interrupts except on PREEMPT_RT. PREEMPT_RT
was excluded because the needed synchronisation was already provided
local_bh_disable().

Since the introduction of backlog NAPI and making it mandatory for
PREEMPT_RT the ifdef within backlog_lock.*() is obsolete and can be
removed.

Remove the ifdefs in backlog_lock.*().

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-10-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/dev.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 95b9e4cc1767..73c4d14e4feb 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -229,7 +229,7 @@ static inline void backlog_lock_irq_save(struct softnet_data *sd,
 {
 	if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
 		spin_lock_irqsave(&sd->input_pkt_queue.lock, *flags);
-	else if (!IS_ENABLED(CONFIG_PREEMPT_RT))
+	else
 		local_irq_save(*flags);
 }
 
@@ -237,7 +237,7 @@ static inline void backlog_lock_irq_disable(struct softnet_data *sd)
 {
 	if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
 		spin_lock_irq(&sd->input_pkt_queue.lock);
-	else if (!IS_ENABLED(CONFIG_PREEMPT_RT))
+	else
 		local_irq_disable();
 }
 
@@ -246,7 +246,7 @@ static inline void backlog_unlock_irq_restore(struct softnet_data *sd,
 {
 	if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
 		spin_unlock_irqrestore(&sd->input_pkt_queue.lock, *flags);
-	else if (!IS_ENABLED(CONFIG_PREEMPT_RT))
+	else
 		local_irq_restore(*flags);
 }
 
@@ -254,7 +254,7 @@ static inline void backlog_unlock_irq_enable(struct softnet_data *sd)
 {
 	if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
 		spin_unlock_irq(&sd->input_pkt_queue.lock);
-	else if (!IS_ENABLED(CONFIG_PREEMPT_RT))
+	else
 		local_irq_enable();
 }
 
-- 
cgit v1.2.3-58-ga151


From b22800f9d3b142bf2550dd47ff738b9feedc1093 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:22:00 +0200
Subject: dev: Use nested-BH locking for softnet_data.process_queue.

softnet_data::process_queue is a per-CPU variable and relies on disabled
BH for its locking. Without per-CPU locking in local_bh_disable() on
PREEMPT_RT this data structure requires explicit locking.

softnet_data::input_queue_head can be updated lockless. This is fine
because this value is only update CPU local by the local backlog_napi
thread.

Add a local_lock_t to softnet_data and use local_lock_nested_bh() for locking
of process_queue. This change adds only lockdep coverage and does not
alter the functional behaviour for !PREEMPT_RT.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-11-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/netdevice.h |  1 +
 net/core/dev.c            | 12 +++++++++++-
 2 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f6fc9066147d..4e81660b4462 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3202,6 +3202,7 @@ static inline bool dev_has_header(const struct net_device *dev)
 struct softnet_data {
 	struct list_head	poll_list;
 	struct sk_buff_head	process_queue;
+	local_lock_t		process_queue_bh_lock;
 
 	/* stats */
 	unsigned int		processed;
diff --git a/net/core/dev.c b/net/core/dev.c
index 73c4d14e4feb..8ef727c2ae2b 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -449,7 +449,9 @@ static RAW_NOTIFIER_HEAD(netdev_chain);
  *	queue in the local softnet handler.
  */
 
-DEFINE_PER_CPU_ALIGNED(struct softnet_data, softnet_data);
+DEFINE_PER_CPU_ALIGNED(struct softnet_data, softnet_data) = {
+	.process_queue_bh_lock = INIT_LOCAL_LOCK(process_queue_bh_lock),
+};
 EXPORT_PER_CPU_SYMBOL(softnet_data);
 
 /* Page_pool has a lockless array/stack to alloc/recycle pages.
@@ -5949,6 +5951,7 @@ static void flush_backlog(struct work_struct *work)
 	}
 	backlog_unlock_irq_enable(sd);
 
+	local_lock_nested_bh(&softnet_data.process_queue_bh_lock);
 	skb_queue_walk_safe(&sd->process_queue, skb, tmp) {
 		if (skb->dev->reg_state == NETREG_UNREGISTERING) {
 			__skb_unlink(skb, &sd->process_queue);
@@ -5956,6 +5959,7 @@ static void flush_backlog(struct work_struct *work)
 			rps_input_queue_head_incr(sd);
 		}
 	}
+	local_unlock_nested_bh(&softnet_data.process_queue_bh_lock);
 	local_bh_enable();
 }
 
@@ -6077,7 +6081,9 @@ static int process_backlog(struct napi_struct *napi, int quota)
 	while (again) {
 		struct sk_buff *skb;
 
+		local_lock_nested_bh(&softnet_data.process_queue_bh_lock);
 		while ((skb = __skb_dequeue(&sd->process_queue))) {
+			local_unlock_nested_bh(&softnet_data.process_queue_bh_lock);
 			rcu_read_lock();
 			__netif_receive_skb(skb);
 			rcu_read_unlock();
@@ -6086,7 +6092,9 @@ static int process_backlog(struct napi_struct *napi, int quota)
 				return work;
 			}
 
+			local_lock_nested_bh(&softnet_data.process_queue_bh_lock);
 		}
+		local_unlock_nested_bh(&softnet_data.process_queue_bh_lock);
 
 		backlog_lock_irq_disable(sd);
 		if (skb_queue_empty(&sd->input_pkt_queue)) {
@@ -6101,8 +6109,10 @@ static int process_backlog(struct napi_struct *napi, int quota)
 			napi->state &= NAPIF_STATE_THREADED;
 			again = false;
 		} else {
+			local_lock_nested_bh(&softnet_data.process_queue_bh_lock);
 			skb_queue_splice_tail_init(&sd->input_pkt_queue,
 						   &sd->process_queue);
+			local_unlock_nested_bh(&softnet_data.process_queue_bh_lock);
 		}
 		backlog_unlock_irq_enable(sd);
 	}
-- 
cgit v1.2.3-58-ga151


From 3414adbd6a6ad3702d0bdc49081ee7c9e9e1c600 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:22:01 +0200
Subject: lwt: Don't disable migration prio invoking BPF.

There is no need to explicitly disable migration if bottom halves are
also disabled. Disabling BH implies disabling migration.

Remove migrate_disable() and rely solely on disabling BH to remain on
the same CPU.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-12-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/lwt_bpf.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/net/core/lwt_bpf.c b/net/core/lwt_bpf.c
index 4a0797f0a154..a94943681e5a 100644
--- a/net/core/lwt_bpf.c
+++ b/net/core/lwt_bpf.c
@@ -40,10 +40,9 @@ static int run_lwt_bpf(struct sk_buff *skb, struct bpf_lwt_prog *lwt,
 {
 	int ret;
 
-	/* Migration disable and BH disable are needed to protect per-cpu
-	 * redirect_info between BPF prog and skb_do_redirect().
+	/* Disabling BH is needed to protect per-CPU bpf_redirect_info between
+	 * BPF prog and skb_do_redirect().
 	 */
-	migrate_disable();
 	local_bh_disable();
 	bpf_compute_data_pointers(skb);
 	ret = bpf_prog_run_save_cb(lwt->prog, skb);
@@ -78,7 +77,6 @@ static int run_lwt_bpf(struct sk_buff *skb, struct bpf_lwt_prog *lwt,
 	}
 
 	local_bh_enable();
-	migrate_enable();
 
 	return ret;
 }
-- 
cgit v1.2.3-58-ga151


From d1542d4ae4dfdc47c9b3205ebe849ed23af213dd Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:22:02 +0200
Subject: seg6: Use nested-BH locking for seg6_bpf_srh_states.

The access to seg6_bpf_srh_states is protected by disabling preemption.
Based on the code, the entry point is input_action_end_bpf() and
every other function (the bpf helper functions bpf_lwt_seg6_*()), that
is accessing seg6_bpf_srh_states, should be called from within
input_action_end_bpf().

input_action_end_bpf() accesses seg6_bpf_srh_states first at the top of
the function and then disables preemption. This looks wrong because if
preemption needs to be disabled as part of the locking mechanism then
the variable shouldn't be accessed beforehand.

Looking at how it is used via test_lwt_seg6local.sh then
input_action_end_bpf() is always invoked from softirq context. If this
is always the case then the preempt_disable() statement is superfluous.
If this is not always invoked from softirq then disabling only
preemption is not sufficient.

Replace the preempt_disable() statement with nested-BH locking. This is
not an equivalent replacement as it assumes that the invocation of
input_action_end_bpf() always occurs in softirq context and thus the
preempt_disable() is superfluous.
Add a local_lock_t the data structure and use local_lock_nested_bh() for
locking. Add lockdep_assert_held() to ensure the lock is held while the
per-CPU variable is referenced in the helper functions.

Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: David Ahern <dsahern@kernel.org>
Cc: Hao Luo <haoluo@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: KP Singh <kpsingh@kernel.org>
Cc: Martin KaFai Lau <martin.lau@linux.dev>
Cc: Song Liu <song@kernel.org>
Cc: Stanislav Fomichev <sdf@google.com>
Cc: Yonghong Song <yonghong.song@linux.dev>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-13-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/seg6_local.h |  1 +
 net/core/filter.c        |  3 +++
 net/ipv6/seg6_local.c    | 22 ++++++++++++++--------
 3 files changed, 18 insertions(+), 8 deletions(-)

diff --git a/include/net/seg6_local.h b/include/net/seg6_local.h
index 3fab9dec2ec4..888c1ce6f527 100644
--- a/include/net/seg6_local.h
+++ b/include/net/seg6_local.h
@@ -19,6 +19,7 @@ extern int seg6_lookup_nexthop(struct sk_buff *skb, struct in6_addr *nhaddr,
 extern bool seg6_bpf_has_valid_srh(struct sk_buff *skb);
 
 struct seg6_bpf_srh_state {
+	local_lock_t bh_lock;
 	struct ipv6_sr_hdr *srh;
 	u16 hdrlen;
 	bool valid;
diff --git a/net/core/filter.c b/net/core/filter.c
index b077e7467946..c4f5db324f04 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -6455,6 +6455,7 @@ BPF_CALL_4(bpf_lwt_seg6_store_bytes, struct sk_buff *, skb, u32, offset,
 	void *srh_tlvs, *srh_end, *ptr;
 	int srhoff = 0;
 
+	lockdep_assert_held(&srh_state->bh_lock);
 	if (srh == NULL)
 		return -EINVAL;
 
@@ -6511,6 +6512,7 @@ BPF_CALL_4(bpf_lwt_seg6_action, struct sk_buff *, skb,
 	int hdroff = 0;
 	int err;
 
+	lockdep_assert_held(&srh_state->bh_lock);
 	switch (action) {
 	case SEG6_LOCAL_ACTION_END_X:
 		if (!seg6_bpf_has_valid_srh(skb))
@@ -6587,6 +6589,7 @@ BPF_CALL_3(bpf_lwt_seg6_adjust_srh, struct sk_buff *, skb, u32, offset,
 	int srhoff = 0;
 	int ret;
 
+	lockdep_assert_held(&srh_state->bh_lock);
 	if (unlikely(srh == NULL))
 		return -EINVAL;
 
diff --git a/net/ipv6/seg6_local.c b/net/ipv6/seg6_local.c
index c434940131b1..c74705ead984 100644
--- a/net/ipv6/seg6_local.c
+++ b/net/ipv6/seg6_local.c
@@ -1380,7 +1380,9 @@ drop:
 	return err;
 }
 
-DEFINE_PER_CPU(struct seg6_bpf_srh_state, seg6_bpf_srh_states);
+DEFINE_PER_CPU(struct seg6_bpf_srh_state, seg6_bpf_srh_states) = {
+	.bh_lock	= INIT_LOCAL_LOCK(bh_lock),
+};
 
 bool seg6_bpf_has_valid_srh(struct sk_buff *skb)
 {
@@ -1388,6 +1390,7 @@ bool seg6_bpf_has_valid_srh(struct sk_buff *skb)
 		this_cpu_ptr(&seg6_bpf_srh_states);
 	struct ipv6_sr_hdr *srh = srh_state->srh;
 
+	lockdep_assert_held(&srh_state->bh_lock);
 	if (unlikely(srh == NULL))
 		return false;
 
@@ -1408,8 +1411,7 @@ bool seg6_bpf_has_valid_srh(struct sk_buff *skb)
 static int input_action_end_bpf(struct sk_buff *skb,
 				struct seg6_local_lwt *slwt)
 {
-	struct seg6_bpf_srh_state *srh_state =
-		this_cpu_ptr(&seg6_bpf_srh_states);
+	struct seg6_bpf_srh_state *srh_state;
 	struct ipv6_sr_hdr *srh;
 	int ret;
 
@@ -1420,10 +1422,14 @@ static int input_action_end_bpf(struct sk_buff *skb,
 	}
 	advance_nextseg(srh, &ipv6_hdr(skb)->daddr);
 
-	/* preempt_disable is needed to protect the per-CPU buffer srh_state,
-	 * which is also accessed by the bpf_lwt_seg6_* helpers
+	/* The access to the per-CPU buffer srh_state is protected by running
+	 * always in softirq context (with disabled BH). On PREEMPT_RT the
+	 * required locking is provided by the following local_lock_nested_bh()
+	 * statement. It is also accessed by the bpf_lwt_seg6_* helpers via
+	 * bpf_prog_run_save_cb().
 	 */
-	preempt_disable();
+	local_lock_nested_bh(&seg6_bpf_srh_states.bh_lock);
+	srh_state = this_cpu_ptr(&seg6_bpf_srh_states);
 	srh_state->srh = srh;
 	srh_state->hdrlen = srh->hdrlen << 3;
 	srh_state->valid = true;
@@ -1446,15 +1452,15 @@ static int input_action_end_bpf(struct sk_buff *skb,
 
 	if (srh_state->srh && !seg6_bpf_has_valid_srh(skb))
 		goto drop;
+	local_unlock_nested_bh(&seg6_bpf_srh_states.bh_lock);
 
-	preempt_enable();
 	if (ret != BPF_REDIRECT)
 		seg6_lookup_nexthop(skb, NULL, 0);
 
 	return dst_input(skb);
 
 drop:
-	preempt_enable();
+	local_unlock_nested_bh(&seg6_bpf_srh_states.bh_lock);
 	kfree_skb(skb);
 	return -EINVAL;
 }
-- 
cgit v1.2.3-58-ga151


From 78f520b7bbe579438dfc202226b3dac5607d8c7f Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:22:03 +0200
Subject: net: Use nested-BH locking for bpf_scratchpad.

bpf_scratchpad is a per-CPU variable and relies on disabled BH for its
locking. Without per-CPU locking in local_bh_disable() on PREEMPT_RT
this data structure requires explicit locking.

Add a local_lock_t to the data structure and use local_lock_nested_bh()
for locking. This change adds only lockdep coverage and does not alter
the functional behaviour for !PREEMPT_RT.

Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Hao Luo <haoluo@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: KP Singh <kpsingh@kernel.org>
Cc: Martin KaFai Lau <martin.lau@linux.dev>
Cc: Song Liu <song@kernel.org>
Cc: Stanislav Fomichev <sdf@google.com>
Cc: Yonghong Song <yonghong.song@linux.dev>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-14-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/filter.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/net/core/filter.c b/net/core/filter.c
index c4f5db324f04..adb887195c3a 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1658,9 +1658,12 @@ struct bpf_scratchpad {
 		__be32 diff[MAX_BPF_STACK / sizeof(__be32)];
 		u8     buff[MAX_BPF_STACK];
 	};
+	local_lock_t	bh_lock;
 };
 
-static DEFINE_PER_CPU(struct bpf_scratchpad, bpf_sp);
+static DEFINE_PER_CPU(struct bpf_scratchpad, bpf_sp) = {
+	.bh_lock	= INIT_LOCAL_LOCK(bh_lock),
+};
 
 static inline int __bpf_try_make_writable(struct sk_buff *skb,
 					  unsigned int write_len)
@@ -2021,6 +2024,7 @@ BPF_CALL_5(bpf_csum_diff, __be32 *, from, u32, from_size,
 	struct bpf_scratchpad *sp = this_cpu_ptr(&bpf_sp);
 	u32 diff_size = from_size + to_size;
 	int i, j = 0;
+	__wsum ret;
 
 	/* This is quite flexible, some examples:
 	 *
@@ -2034,12 +2038,15 @@ BPF_CALL_5(bpf_csum_diff, __be32 *, from, u32, from_size,
 		     diff_size > sizeof(sp->diff)))
 		return -EINVAL;
 
+	local_lock_nested_bh(&bpf_sp.bh_lock);
 	for (i = 0; i < from_size / sizeof(__be32); i++, j++)
 		sp->diff[j] = ~from[i];
 	for (i = 0; i <   to_size / sizeof(__be32); i++, j++)
 		sp->diff[j] = to[i];
 
-	return csum_partial(sp->diff, diff_size, seed);
+	ret = csum_partial(sp->diff, diff_size, seed);
+	local_unlock_nested_bh(&bpf_sp.bh_lock);
+	return ret;
 }
 
 static const struct bpf_func_proto bpf_csum_diff_proto = {
-- 
cgit v1.2.3-58-ga151


From 401cb7dae8130fd34eb84648e02ab4c506df7d5e Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:22:04 +0200
Subject: net: Reference bpf_redirect_info via task_struct on PREEMPT_RT.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The XDP redirect process is two staged:
- bpf_prog_run_xdp() is invoked to run a eBPF program which inspects the
  packet and makes decisions. While doing that, the per-CPU variable
  bpf_redirect_info is used.

- Afterwards xdp_do_redirect() is invoked and accesses bpf_redirect_info
  and it may also access other per-CPU variables like xskmap_flush_list.

At the very end of the NAPI callback, xdp_do_flush() is invoked which
does not access bpf_redirect_info but will touch the individual per-CPU
lists.

The per-CPU variables are only used in the NAPI callback hence disabling
bottom halves is the only protection mechanism. Users from preemptible
context (like cpu_map_kthread_run()) explicitly disable bottom halves
for protections reasons.
Without locking in local_bh_disable() on PREEMPT_RT this data structure
requires explicit locking.

PREEMPT_RT has forced-threaded interrupts enabled and every
NAPI-callback runs in a thread. If each thread has its own data
structure then locking can be avoided.

Create a struct bpf_net_context which contains struct bpf_redirect_info.
Define the variable on stack, use bpf_net_ctx_set() to save a pointer to
it, bpf_net_ctx_clear() removes it again.
The bpf_net_ctx_set() may nest. For instance a function can be used from
within NET_RX_SOFTIRQ/ net_rx_action which uses bpf_net_ctx_set() and
NET_TX_SOFTIRQ which does not. Therefore only the first invocations
updates the pointer.
Use bpf_net_ctx_get_ri() as a wrapper to retrieve the current struct
bpf_redirect_info. The returned data structure is zero initialized to
ensure nothing is leaked from stack. This is done on first usage of the
struct. bpf_net_ctx_set() sets bpf_redirect_info::kern_flags to 0 to
note that initialisation is required. First invocation of
bpf_net_ctx_get_ri() will memset() the data structure and update
bpf_redirect_info::kern_flags.
bpf_redirect_info::nh is excluded from memset because it is only used
once BPF_F_NEIGH is set which also sets the nh member. The kern_flags is
moved past nh to exclude it from memset.

The pointer to bpf_net_context is saved task's task_struct. Using
always the bpf_net_context approach has the advantage that there is
almost zero differences between PREEMPT_RT and non-PREEMPT_RT builds.

Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Eduard Zingerman <eddyz87@gmail.com>
Cc: Hao Luo <haoluo@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: KP Singh <kpsingh@kernel.org>
Cc: Martin KaFai Lau <martin.lau@linux.dev>
Cc: Song Liu <song@kernel.org>
Cc: Stanislav Fomichev <sdf@google.com>
Cc: Yonghong Song <yonghong.song@linux.dev>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-15-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/filter.h | 56 +++++++++++++++++++++++++++++++++++++++++---------
 include/linux/sched.h  |  3 +++
 kernel/bpf/cpumap.c    |  3 +++
 kernel/bpf/devmap.c    |  9 +++++++-
 kernel/fork.c          |  1 +
 net/bpf/test_run.c     | 11 +++++++++-
 net/core/dev.c         | 29 +++++++++++++++++++++++++-
 net/core/filter.c      | 44 +++++++++++----------------------------
 net/core/lwt_bpf.c     |  3 +++
 9 files changed, 114 insertions(+), 45 deletions(-)

diff --git a/include/linux/filter.h b/include/linux/filter.h
index b02aea291b7e..0a7f6e4a00b6 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -733,21 +733,59 @@ struct bpf_nh_params {
 	};
 };
 
+/* flags for bpf_redirect_info kern_flags */
+#define BPF_RI_F_RF_NO_DIRECT	BIT(0)	/* no napi_direct on return_frame */
+#define BPF_RI_F_RI_INIT	BIT(1)
+
 struct bpf_redirect_info {
 	u64 tgt_index;
 	void *tgt_value;
 	struct bpf_map *map;
 	u32 flags;
-	u32 kern_flags;
 	u32 map_id;
 	enum bpf_map_type map_type;
 	struct bpf_nh_params nh;
+	u32 kern_flags;
 };
 
-DECLARE_PER_CPU(struct bpf_redirect_info, bpf_redirect_info);
+struct bpf_net_context {
+	struct bpf_redirect_info ri;
+};
 
-/* flags for bpf_redirect_info kern_flags */
-#define BPF_RI_F_RF_NO_DIRECT	BIT(0)	/* no napi_direct on return_frame */
+static inline struct bpf_net_context *bpf_net_ctx_set(struct bpf_net_context *bpf_net_ctx)
+{
+	struct task_struct *tsk = current;
+
+	if (tsk->bpf_net_context != NULL)
+		return NULL;
+	bpf_net_ctx->ri.kern_flags = 0;
+
+	tsk->bpf_net_context = bpf_net_ctx;
+	return bpf_net_ctx;
+}
+
+static inline void bpf_net_ctx_clear(struct bpf_net_context *bpf_net_ctx)
+{
+	if (bpf_net_ctx)
+		current->bpf_net_context = NULL;
+}
+
+static inline struct bpf_net_context *bpf_net_ctx_get(void)
+{
+	return current->bpf_net_context;
+}
+
+static inline struct bpf_redirect_info *bpf_net_ctx_get_ri(void)
+{
+	struct bpf_net_context *bpf_net_ctx = bpf_net_ctx_get();
+
+	if (!(bpf_net_ctx->ri.kern_flags & BPF_RI_F_RI_INIT)) {
+		memset(&bpf_net_ctx->ri, 0, offsetof(struct bpf_net_context, ri.nh));
+		bpf_net_ctx->ri.kern_flags |= BPF_RI_F_RI_INIT;
+	}
+
+	return &bpf_net_ctx->ri;
+}
 
 /* Compute the linear packet data range [data, data_end) which
  * will be accessed by various program types (cls_bpf, act_bpf,
@@ -1018,25 +1056,23 @@ struct bpf_prog *bpf_patch_insn_single(struct bpf_prog *prog, u32 off,
 				       const struct bpf_insn *patch, u32 len);
 int bpf_remove_insns(struct bpf_prog *prog, u32 off, u32 cnt);
 
-void bpf_clear_redirect_map(struct bpf_map *map);
-
 static inline bool xdp_return_frame_no_direct(void)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 
 	return ri->kern_flags & BPF_RI_F_RF_NO_DIRECT;
 }
 
 static inline void xdp_set_return_frame_no_direct(void)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 
 	ri->kern_flags |= BPF_RI_F_RF_NO_DIRECT;
 }
 
 static inline void xdp_clear_return_frame_no_direct(void)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 
 	ri->kern_flags &= ~BPF_RI_F_RF_NO_DIRECT;
 }
@@ -1592,7 +1628,7 @@ static __always_inline long __bpf_xdp_redirect_map(struct bpf_map *map, u64 inde
 						   u64 flags, const u64 flag_mask,
 						   void *lookup_elem(struct bpf_map *map, u32 key))
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 	const u64 action_mask = XDP_ABORTED | XDP_DROP | XDP_PASS | XDP_TX;
 
 	/* Lower bits of the flags are used as return code on lookup failure */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 5187486c2522..5ff5e65a4627 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -54,6 +54,7 @@ struct bio_list;
 struct blk_plug;
 struct bpf_local_storage;
 struct bpf_run_ctx;
+struct bpf_net_context;
 struct capture_control;
 struct cfs_rq;
 struct fs_struct;
@@ -1509,6 +1510,8 @@ struct task_struct {
 	/* Used for BPF run context */
 	struct bpf_run_ctx		*bpf_ctx;
 #endif
+	/* Used by BPF for per-TASK xdp storage */
+	struct bpf_net_context		*bpf_net_context;
 
 #ifdef CONFIG_GCC_PLUGIN_STACKLEAK
 	unsigned long			lowest_stack;
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index a8e34416e960..66974bd02710 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -240,12 +240,14 @@ static int cpu_map_bpf_prog_run(struct bpf_cpu_map_entry *rcpu, void **frames,
 				int xdp_n, struct xdp_cpumap_stats *stats,
 				struct list_head *list)
 {
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 	int nframes;
 
 	if (!rcpu->prog)
 		return xdp_n;
 
 	rcu_read_lock_bh();
+	bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
 
 	nframes = cpu_map_bpf_prog_run_xdp(rcpu, frames, xdp_n, stats);
 
@@ -255,6 +257,7 @@ static int cpu_map_bpf_prog_run(struct bpf_cpu_map_entry *rcpu, void **frames,
 	if (unlikely(!list_empty(list)))
 		cpu_map_bpf_prog_run_skb(rcpu, list, stats);
 
+	bpf_net_ctx_clear(bpf_net_ctx);
 	rcu_read_unlock_bh(); /* resched point, may call do_softirq() */
 
 	return nframes;
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 7f3b34452243..fbfdfb60db8d 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -196,7 +196,14 @@ static void dev_map_free(struct bpf_map *map)
 	list_del_rcu(&dtab->list);
 	spin_unlock(&dev_map_lock);
 
-	bpf_clear_redirect_map(map);
+	/* bpf_redirect_info->map is assigned in __bpf_xdp_redirect_map()
+	 * during NAPI callback and cleared after the XDP redirect. There is no
+	 * explicit RCU read section which protects bpf_redirect_info->map but
+	 * local_bh_disable() also marks the beginning an RCU section. This
+	 * makes the complete softirq callback RCU protected. Thus after
+	 * following synchronize_rcu() there no bpf_redirect_info->map == map
+	 * assignment.
+	 */
 	synchronize_rcu();
 
 	/* Make sure prior __dev_map_entry_free() have completed. */
diff --git a/kernel/fork.c b/kernel/fork.c
index 99076dbe27d8..f314bdd7e610 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -2355,6 +2355,7 @@ __latent_entropy struct task_struct *copy_process(
 	RCU_INIT_POINTER(p->bpf_storage, NULL);
 	p->bpf_ctx = NULL;
 #endif
+	p->bpf_net_context =  NULL;
 
 	/* Perform scheduler related setup. Assign this task to a CPU. */
 	retval = sched_fork(clone_flags, p);
diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index 36ae54f57bf5..a6d7f790cdda 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -283,9 +283,10 @@ static int xdp_recv_frames(struct xdp_frame **frames, int nframes,
 static int xdp_test_run_batch(struct xdp_test_data *xdp, struct bpf_prog *prog,
 			      u32 repeat)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 	int err = 0, act, ret, i, nframes = 0, batch_sz;
 	struct xdp_frame **frames = xdp->frames;
+	struct bpf_redirect_info *ri;
 	struct xdp_page_head *head;
 	struct xdp_frame *frm;
 	bool redirect = false;
@@ -295,6 +296,8 @@ static int xdp_test_run_batch(struct xdp_test_data *xdp, struct bpf_prog *prog,
 	batch_sz = min_t(u32, repeat, xdp->batch_size);
 
 	local_bh_disable();
+	bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
+	ri = bpf_net_ctx_get_ri();
 	xdp_set_return_frame_no_direct();
 
 	for (i = 0; i < batch_sz; i++) {
@@ -359,6 +362,7 @@ out:
 	}
 
 	xdp_clear_return_frame_no_direct();
+	bpf_net_ctx_clear(bpf_net_ctx);
 	local_bh_enable();
 	return err;
 }
@@ -394,6 +398,7 @@ static int bpf_test_run_xdp_live(struct bpf_prog *prog, struct xdp_buff *ctx,
 static int bpf_test_run(struct bpf_prog *prog, void *ctx, u32 repeat,
 			u32 *retval, u32 *time, bool xdp)
 {
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 	struct bpf_prog_array_item item = {.prog = prog};
 	struct bpf_run_ctx *old_ctx;
 	struct bpf_cg_run_ctx run_ctx;
@@ -419,10 +424,14 @@ static int bpf_test_run(struct bpf_prog *prog, void *ctx, u32 repeat,
 	do {
 		run_ctx.prog_item = &item;
 		local_bh_disable();
+		bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
+
 		if (xdp)
 			*retval = bpf_prog_run_xdp(prog, ctx);
 		else
 			*retval = bpf_prog_run(prog, ctx);
+
+		bpf_net_ctx_clear(bpf_net_ctx);
 		local_bh_enable();
 	} while (bpf_test_timer_continue(&t, 1, repeat, &ret, time));
 	bpf_reset_run_ctx(old_ctx);
diff --git a/net/core/dev.c b/net/core/dev.c
index 8ef727c2ae2b..b94fb4e63a28 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4045,10 +4045,13 @@ sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret,
 {
 	struct bpf_mprog_entry *entry = rcu_dereference_bh(skb->dev->tcx_ingress);
 	enum skb_drop_reason drop_reason = SKB_DROP_REASON_TC_INGRESS;
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 	int sch_ret;
 
 	if (!entry)
 		return skb;
+
+	bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
 	if (*pt_prev) {
 		*ret = deliver_skb(skb, *pt_prev, orig_dev);
 		*pt_prev = NULL;
@@ -4077,10 +4080,12 @@ ingress_verdict:
 			break;
 		}
 		*ret = NET_RX_SUCCESS;
+		bpf_net_ctx_clear(bpf_net_ctx);
 		return NULL;
 	case TC_ACT_SHOT:
 		kfree_skb_reason(skb, drop_reason);
 		*ret = NET_RX_DROP;
+		bpf_net_ctx_clear(bpf_net_ctx);
 		return NULL;
 	/* used by tc_run */
 	case TC_ACT_STOLEN:
@@ -4090,8 +4095,10 @@ ingress_verdict:
 		fallthrough;
 	case TC_ACT_CONSUMED:
 		*ret = NET_RX_SUCCESS;
+		bpf_net_ctx_clear(bpf_net_ctx);
 		return NULL;
 	}
+	bpf_net_ctx_clear(bpf_net_ctx);
 
 	return skb;
 }
@@ -4101,11 +4108,14 @@ sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev)
 {
 	struct bpf_mprog_entry *entry = rcu_dereference_bh(dev->tcx_egress);
 	enum skb_drop_reason drop_reason = SKB_DROP_REASON_TC_EGRESS;
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 	int sch_ret;
 
 	if (!entry)
 		return skb;
 
+	bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
+
 	/* qdisc_skb_cb(skb)->pkt_len & tcx_set_ingress() was
 	 * already set by the caller.
 	 */
@@ -4121,10 +4131,12 @@ egress_verdict:
 		/* No need to push/pop skb's mac_header here on egress! */
 		skb_do_redirect(skb);
 		*ret = NET_XMIT_SUCCESS;
+		bpf_net_ctx_clear(bpf_net_ctx);
 		return NULL;
 	case TC_ACT_SHOT:
 		kfree_skb_reason(skb, drop_reason);
 		*ret = NET_XMIT_DROP;
+		bpf_net_ctx_clear(bpf_net_ctx);
 		return NULL;
 	/* used by tc_run */
 	case TC_ACT_STOLEN:
@@ -4134,8 +4146,10 @@ egress_verdict:
 		fallthrough;
 	case TC_ACT_CONSUMED:
 		*ret = NET_XMIT_SUCCESS;
+		bpf_net_ctx_clear(bpf_net_ctx);
 		return NULL;
 	}
+	bpf_net_ctx_clear(bpf_net_ctx);
 
 	return skb;
 }
@@ -6325,6 +6339,7 @@ enum {
 static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock,
 			   unsigned flags, u16 budget)
 {
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 	bool skip_schedule = false;
 	unsigned long timeout;
 	int rc;
@@ -6342,6 +6357,7 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock,
 	clear_bit(NAPI_STATE_IN_BUSY_POLL, &napi->state);
 
 	local_bh_disable();
+	bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
 
 	if (flags & NAPI_F_PREFER_BUSY_POLL) {
 		napi->defer_hard_irqs_count = READ_ONCE(napi->dev->napi_defer_hard_irqs);
@@ -6364,6 +6380,7 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock,
 	netpoll_poll_unlock(have_poll_lock);
 	if (rc == budget)
 		__busy_poll_stop(napi, skip_schedule);
+	bpf_net_ctx_clear(bpf_net_ctx);
 	local_bh_enable();
 }
 
@@ -6373,6 +6390,7 @@ static void __napi_busy_loop(unsigned int napi_id,
 {
 	unsigned long start_time = loop_end ? busy_loop_current_time() : 0;
 	int (*napi_poll)(struct napi_struct *napi, int budget);
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 	void *have_poll_lock = NULL;
 	struct napi_struct *napi;
 
@@ -6391,6 +6409,7 @@ restart:
 		int work = 0;
 
 		local_bh_disable();
+		bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
 		if (!napi_poll) {
 			unsigned long val = READ_ONCE(napi->state);
 
@@ -6421,6 +6440,7 @@ count:
 			__NET_ADD_STATS(dev_net(napi->dev),
 					LINUX_MIB_BUSYPOLLRXPACKETS, work);
 		skb_defer_free_flush(this_cpu_ptr(&softnet_data));
+		bpf_net_ctx_clear(bpf_net_ctx);
 		local_bh_enable();
 
 		if (!loop_end || loop_end(loop_end_arg, start_time))
@@ -6848,6 +6868,7 @@ static int napi_thread_wait(struct napi_struct *napi)
 
 static void napi_threaded_poll_loop(struct napi_struct *napi)
 {
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 	struct softnet_data *sd;
 	unsigned long last_qs = jiffies;
 
@@ -6856,6 +6877,8 @@ static void napi_threaded_poll_loop(struct napi_struct *napi)
 		void *have;
 
 		local_bh_disable();
+		bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
+
 		sd = this_cpu_ptr(&softnet_data);
 		sd->in_napi_threaded_poll = true;
 
@@ -6871,6 +6894,7 @@ static void napi_threaded_poll_loop(struct napi_struct *napi)
 			net_rps_action_and_irq_enable(sd);
 		}
 		skb_defer_free_flush(sd);
+		bpf_net_ctx_clear(bpf_net_ctx);
 		local_bh_enable();
 
 		if (!repoll)
@@ -6896,10 +6920,12 @@ static __latent_entropy void net_rx_action(struct softirq_action *h)
 	struct softnet_data *sd = this_cpu_ptr(&softnet_data);
 	unsigned long time_limit = jiffies +
 		usecs_to_jiffies(READ_ONCE(net_hotdata.netdev_budget_usecs));
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 	int budget = READ_ONCE(net_hotdata.netdev_budget);
 	LIST_HEAD(list);
 	LIST_HEAD(repoll);
 
+	bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
 start:
 	sd->in_net_rx_action = true;
 	local_irq_disable();
@@ -6952,7 +6978,8 @@ start:
 		sd->in_net_rx_action = false;
 
 	net_rps_action_and_irq_enable(sd);
-end:;
+end:
+	bpf_net_ctx_clear(bpf_net_ctx);
 }
 
 struct netdev_adjacent {
diff --git a/net/core/filter.c b/net/core/filter.c
index adb887195c3a..eb1c4425c06f 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -2483,9 +2483,6 @@ static const struct bpf_func_proto bpf_clone_redirect_proto = {
 	.arg3_type      = ARG_ANYTHING,
 };
 
-DEFINE_PER_CPU(struct bpf_redirect_info, bpf_redirect_info);
-EXPORT_PER_CPU_SYMBOL_GPL(bpf_redirect_info);
-
 static struct net_device *skb_get_peer_dev(struct net_device *dev)
 {
 	const struct net_device_ops *ops = dev->netdev_ops;
@@ -2498,7 +2495,7 @@ static struct net_device *skb_get_peer_dev(struct net_device *dev)
 
 int skb_do_redirect(struct sk_buff *skb)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 	struct net *net = dev_net(skb->dev);
 	struct net_device *dev;
 	u32 flags = ri->flags;
@@ -2531,7 +2528,7 @@ out_drop:
 
 BPF_CALL_2(bpf_redirect, u32, ifindex, u64, flags)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 
 	if (unlikely(flags & (~(BPF_F_INGRESS) | BPF_F_REDIRECT_INTERNAL)))
 		return TC_ACT_SHOT;
@@ -2552,7 +2549,7 @@ static const struct bpf_func_proto bpf_redirect_proto = {
 
 BPF_CALL_2(bpf_redirect_peer, u32, ifindex, u64, flags)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 
 	if (unlikely(flags))
 		return TC_ACT_SHOT;
@@ -2574,7 +2571,7 @@ static const struct bpf_func_proto bpf_redirect_peer_proto = {
 BPF_CALL_4(bpf_redirect_neigh, u32, ifindex, struct bpf_redir_neigh *, params,
 	   int, plen, u64, flags)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 
 	if (unlikely((plen && plen < sizeof(*params)) || flags))
 		return TC_ACT_SHOT;
@@ -4300,30 +4297,13 @@ void xdp_do_check_flushed(struct napi_struct *napi)
 }
 #endif
 
-void bpf_clear_redirect_map(struct bpf_map *map)
-{
-	struct bpf_redirect_info *ri;
-	int cpu;
-
-	for_each_possible_cpu(cpu) {
-		ri = per_cpu_ptr(&bpf_redirect_info, cpu);
-		/* Avoid polluting remote cacheline due to writes if
-		 * not needed. Once we pass this test, we need the
-		 * cmpxchg() to make sure it hasn't been changed in
-		 * the meantime by remote CPU.
-		 */
-		if (unlikely(READ_ONCE(ri->map) == map))
-			cmpxchg(&ri->map, map, NULL);
-	}
-}
-
 DEFINE_STATIC_KEY_FALSE(bpf_master_redirect_enabled_key);
 EXPORT_SYMBOL_GPL(bpf_master_redirect_enabled_key);
 
 u32 xdp_master_redirect(struct xdp_buff *xdp)
 {
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 	struct net_device *master, *slave;
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
 
 	master = netdev_master_upper_dev_get_rcu(xdp->rxq->dev);
 	slave = master->netdev_ops->ndo_xdp_get_xmit_slave(master, xdp);
@@ -4395,7 +4375,7 @@ static __always_inline int __xdp_do_redirect_frame(struct bpf_redirect_info *ri,
 			map = READ_ONCE(ri->map);
 
 			/* The map pointer is cleared when the map is being torn
-			 * down by bpf_clear_redirect_map()
+			 * down by dev_map_free()
 			 */
 			if (unlikely(!map)) {
 				err = -ENOENT;
@@ -4440,7 +4420,7 @@ err:
 int xdp_do_redirect(struct net_device *dev, struct xdp_buff *xdp,
 		    struct bpf_prog *xdp_prog)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 	enum bpf_map_type map_type = ri->map_type;
 
 	if (map_type == BPF_MAP_TYPE_XSKMAP)
@@ -4454,7 +4434,7 @@ EXPORT_SYMBOL_GPL(xdp_do_redirect);
 int xdp_do_redirect_frame(struct net_device *dev, struct xdp_buff *xdp,
 			  struct xdp_frame *xdpf, struct bpf_prog *xdp_prog)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 	enum bpf_map_type map_type = ri->map_type;
 
 	if (map_type == BPF_MAP_TYPE_XSKMAP)
@@ -4471,7 +4451,7 @@ static int xdp_do_generic_redirect_map(struct net_device *dev,
 				       enum bpf_map_type map_type, u32 map_id,
 				       u32 flags)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 	struct bpf_map *map;
 	int err;
 
@@ -4483,7 +4463,7 @@ static int xdp_do_generic_redirect_map(struct net_device *dev,
 			map = READ_ONCE(ri->map);
 
 			/* The map pointer is cleared when the map is being torn
-			 * down by bpf_clear_redirect_map()
+			 * down by dev_map_free()
 			 */
 			if (unlikely(!map)) {
 				err = -ENOENT;
@@ -4525,7 +4505,7 @@ err:
 int xdp_do_generic_redirect(struct net_device *dev, struct sk_buff *skb,
 			    struct xdp_buff *xdp, struct bpf_prog *xdp_prog)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 	enum bpf_map_type map_type = ri->map_type;
 	void *fwd = ri->tgt_value;
 	u32 map_id = ri->map_id;
@@ -4561,7 +4541,7 @@ err:
 
 BPF_CALL_2(bpf_xdp_redirect, u32, ifindex, u64, flags)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
+	struct bpf_redirect_info *ri = bpf_net_ctx_get_ri();
 
 	if (unlikely(flags))
 		return XDP_ABORTED;
diff --git a/net/core/lwt_bpf.c b/net/core/lwt_bpf.c
index a94943681e5a..afb05f58b64c 100644
--- a/net/core/lwt_bpf.c
+++ b/net/core/lwt_bpf.c
@@ -38,12 +38,14 @@ static inline struct bpf_lwt *bpf_lwt_lwtunnel(struct lwtunnel_state *lwt)
 static int run_lwt_bpf(struct sk_buff *skb, struct bpf_lwt_prog *lwt,
 		       struct dst_entry *dst, bool can_redirect)
 {
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 	int ret;
 
 	/* Disabling BH is needed to protect per-CPU bpf_redirect_info between
 	 * BPF prog and skb_do_redirect().
 	 */
 	local_bh_disable();
+	bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
 	bpf_compute_data_pointers(skb);
 	ret = bpf_prog_run_save_cb(lwt->prog, skb);
 
@@ -76,6 +78,7 @@ static int run_lwt_bpf(struct sk_buff *skb, struct bpf_lwt_prog *lwt,
 		break;
 	}
 
+	bpf_net_ctx_clear(bpf_net_ctx);
 	local_bh_enable();
 
 	return ret;
-- 
cgit v1.2.3-58-ga151


From 3f9fe37d9e16a6cfd5f4d1f536686ea71db3196f Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 20 Jun 2024 15:22:05 +0200
Subject: net: Move per-CPU flush-lists to bpf_net_context on PREEMPT_RT.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The per-CPU flush lists, which are accessed from within the NAPI callback
(xdp_do_flush() for instance), are per-CPU. There are subject to the
same problem as struct bpf_redirect_info.

Add the per-CPU lists cpu_map_flush_list, dev_map_flush_list and
xskmap_map_flush_list to struct bpf_net_context. Add wrappers for the
access. The lists initialized on first usage (similar to
bpf_net_ctx_get_ri()).

Cc: "Björn Töpel" <bjorn@kernel.org>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Eduard Zingerman <eddyz87@gmail.com>
Cc: Hao Luo <haoluo@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Fastabend <john.fastabend@gmail.com>
Cc: Jonathan Lemon <jonathan.lemon@gmail.com>
Cc: KP Singh <kpsingh@kernel.org>
Cc: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Cc: Magnus Karlsson <magnus.karlsson@intel.com>
Cc: Martin KaFai Lau <martin.lau@linux.dev>
Cc: Song Liu <song@kernel.org>
Cc: Stanislav Fomichev <sdf@google.com>
Cc: Yonghong Song <yonghong.song@linux.dev>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240620132727.660738-16-bigeasy@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/filter.h | 42 ++++++++++++++++++++++++++++++++++++++++++
 kernel/bpf/cpumap.c    | 19 +++----------------
 kernel/bpf/devmap.c    | 11 +++--------
 net/xdp/xsk.c          | 12 ++++--------
 4 files changed, 52 insertions(+), 32 deletions(-)

diff --git a/include/linux/filter.h b/include/linux/filter.h
index 0a7f6e4a00b6..c0349522de8f 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -736,6 +736,9 @@ struct bpf_nh_params {
 /* flags for bpf_redirect_info kern_flags */
 #define BPF_RI_F_RF_NO_DIRECT	BIT(0)	/* no napi_direct on return_frame */
 #define BPF_RI_F_RI_INIT	BIT(1)
+#define BPF_RI_F_CPU_MAP_INIT	BIT(2)
+#define BPF_RI_F_DEV_MAP_INIT	BIT(3)
+#define BPF_RI_F_XSK_MAP_INIT	BIT(4)
 
 struct bpf_redirect_info {
 	u64 tgt_index;
@@ -750,6 +753,9 @@ struct bpf_redirect_info {
 
 struct bpf_net_context {
 	struct bpf_redirect_info ri;
+	struct list_head cpu_map_flush_list;
+	struct list_head dev_map_flush_list;
+	struct list_head xskmap_map_flush_list;
 };
 
 static inline struct bpf_net_context *bpf_net_ctx_set(struct bpf_net_context *bpf_net_ctx)
@@ -787,6 +793,42 @@ static inline struct bpf_redirect_info *bpf_net_ctx_get_ri(void)
 	return &bpf_net_ctx->ri;
 }
 
+static inline struct list_head *bpf_net_ctx_get_cpu_map_flush_list(void)
+{
+	struct bpf_net_context *bpf_net_ctx = bpf_net_ctx_get();
+
+	if (!(bpf_net_ctx->ri.kern_flags & BPF_RI_F_CPU_MAP_INIT)) {
+		INIT_LIST_HEAD(&bpf_net_ctx->cpu_map_flush_list);
+		bpf_net_ctx->ri.kern_flags |= BPF_RI_F_CPU_MAP_INIT;
+	}
+
+	return &bpf_net_ctx->cpu_map_flush_list;
+}
+
+static inline struct list_head *bpf_net_ctx_get_dev_flush_list(void)
+{
+	struct bpf_net_context *bpf_net_ctx = bpf_net_ctx_get();
+
+	if (!(bpf_net_ctx->ri.kern_flags & BPF_RI_F_DEV_MAP_INIT)) {
+		INIT_LIST_HEAD(&bpf_net_ctx->dev_map_flush_list);
+		bpf_net_ctx->ri.kern_flags |= BPF_RI_F_DEV_MAP_INIT;
+	}
+
+	return &bpf_net_ctx->dev_map_flush_list;
+}
+
+static inline struct list_head *bpf_net_ctx_get_xskmap_flush_list(void)
+{
+	struct bpf_net_context *bpf_net_ctx = bpf_net_ctx_get();
+
+	if (!(bpf_net_ctx->ri.kern_flags & BPF_RI_F_XSK_MAP_INIT)) {
+		INIT_LIST_HEAD(&bpf_net_ctx->xskmap_map_flush_list);
+		bpf_net_ctx->ri.kern_flags |= BPF_RI_F_XSK_MAP_INIT;
+	}
+
+	return &bpf_net_ctx->xskmap_map_flush_list;
+}
+
 /* Compute the linear packet data range [data, data_end) which
  * will be accessed by various program types (cls_bpf, act_bpf,
  * lwt, ...). Subsystems allowing direct data access must (!)
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index 66974bd02710..068e994ed781 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -79,8 +79,6 @@ struct bpf_cpu_map {
 	struct bpf_cpu_map_entry __rcu **cpu_map;
 };
 
-static DEFINE_PER_CPU(struct list_head, cpu_map_flush_list);
-
 static struct bpf_map *cpu_map_alloc(union bpf_attr *attr)
 {
 	u32 value_size = attr->value_size;
@@ -709,7 +707,7 @@ static void bq_flush_to_queue(struct xdp_bulk_queue *bq)
  */
 static void bq_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_frame *xdpf)
 {
-	struct list_head *flush_list = this_cpu_ptr(&cpu_map_flush_list);
+	struct list_head *flush_list = bpf_net_ctx_get_cpu_map_flush_list();
 	struct xdp_bulk_queue *bq = this_cpu_ptr(rcpu->bulkq);
 
 	if (unlikely(bq->count == CPU_MAP_BULK_SIZE))
@@ -761,7 +759,7 @@ trace:
 
 void __cpu_map_flush(void)
 {
-	struct list_head *flush_list = this_cpu_ptr(&cpu_map_flush_list);
+	struct list_head *flush_list = bpf_net_ctx_get_cpu_map_flush_list();
 	struct xdp_bulk_queue *bq, *tmp;
 
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node) {
@@ -775,20 +773,9 @@ void __cpu_map_flush(void)
 #ifdef CONFIG_DEBUG_NET
 bool cpu_map_check_flush(void)
 {
-	if (list_empty(this_cpu_ptr(&cpu_map_flush_list)))
+	if (list_empty(bpf_net_ctx_get_cpu_map_flush_list()))
 		return false;
 	__cpu_map_flush();
 	return true;
 }
 #endif
-
-static int __init cpu_map_init(void)
-{
-	int cpu;
-
-	for_each_possible_cpu(cpu)
-		INIT_LIST_HEAD(&per_cpu(cpu_map_flush_list, cpu));
-	return 0;
-}
-
-subsys_initcall(cpu_map_init);
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index fbfdfb60db8d..317ac2d66ebd 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -83,7 +83,6 @@ struct bpf_dtab {
 	u32 n_buckets;
 };
 
-static DEFINE_PER_CPU(struct list_head, dev_flush_list);
 static DEFINE_SPINLOCK(dev_map_lock);
 static LIST_HEAD(dev_map_list);
 
@@ -415,7 +414,7 @@ out:
  */
 void __dev_flush(void)
 {
-	struct list_head *flush_list = this_cpu_ptr(&dev_flush_list);
+	struct list_head *flush_list = bpf_net_ctx_get_dev_flush_list();
 	struct xdp_dev_bulk_queue *bq, *tmp;
 
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node) {
@@ -429,7 +428,7 @@ void __dev_flush(void)
 #ifdef CONFIG_DEBUG_NET
 bool dev_check_flush(void)
 {
-	if (list_empty(this_cpu_ptr(&dev_flush_list)))
+	if (list_empty(bpf_net_ctx_get_dev_flush_list()))
 		return false;
 	__dev_flush();
 	return true;
@@ -460,7 +459,7 @@ static void *__dev_map_lookup_elem(struct bpf_map *map, u32 key)
 static void bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
 		       struct net_device *dev_rx, struct bpf_prog *xdp_prog)
 {
-	struct list_head *flush_list = this_cpu_ptr(&dev_flush_list);
+	struct list_head *flush_list = bpf_net_ctx_get_dev_flush_list();
 	struct xdp_dev_bulk_queue *bq = this_cpu_ptr(dev->xdp_bulkq);
 
 	if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
@@ -1160,15 +1159,11 @@ static struct notifier_block dev_map_notifier = {
 
 static int __init dev_map_init(void)
 {
-	int cpu;
-
 	/* Assure tracepoint shadow struct _bpf_dtab_netdev is in sync */
 	BUILD_BUG_ON(offsetof(struct bpf_dtab_netdev, dev) !=
 		     offsetof(struct _bpf_dtab_netdev, dev));
 	register_netdevice_notifier(&dev_map_notifier);
 
-	for_each_possible_cpu(cpu)
-		INIT_LIST_HEAD(&per_cpu(dev_flush_list, cpu));
 	return 0;
 }
 
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 7d1c0986f9bb..ed062e038389 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -35,8 +35,6 @@
 #define TX_BATCH_SIZE 32
 #define MAX_PER_SOCKET_BUDGET (TX_BATCH_SIZE)
 
-static DEFINE_PER_CPU(struct list_head, xskmap_flush_list);
-
 void xsk_set_rx_need_wakeup(struct xsk_buff_pool *pool)
 {
 	if (pool->cached_need_wakeup & XDP_WAKEUP_RX)
@@ -372,7 +370,7 @@ static int xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 
 int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp)
 {
-	struct list_head *flush_list = this_cpu_ptr(&xskmap_flush_list);
+	struct list_head *flush_list = bpf_net_ctx_get_xskmap_flush_list();
 	int err;
 
 	err = xsk_rcv(xs, xdp);
@@ -387,7 +385,7 @@ int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp)
 
 void __xsk_map_flush(void)
 {
-	struct list_head *flush_list = this_cpu_ptr(&xskmap_flush_list);
+	struct list_head *flush_list = bpf_net_ctx_get_xskmap_flush_list();
 	struct xdp_sock *xs, *tmp;
 
 	list_for_each_entry_safe(xs, tmp, flush_list, flush_node) {
@@ -399,7 +397,7 @@ void __xsk_map_flush(void)
 #ifdef CONFIG_DEBUG_NET
 bool xsk_map_check_flush(void)
 {
-	if (list_empty(this_cpu_ptr(&xskmap_flush_list)))
+	if (list_empty(bpf_net_ctx_get_xskmap_flush_list()))
 		return false;
 	__xsk_map_flush();
 	return true;
@@ -1772,7 +1770,7 @@ static struct pernet_operations xsk_net_ops = {
 
 static int __init xsk_init(void)
 {
-	int err, cpu;
+	int err;
 
 	err = proto_register(&xsk_proto, 0 /* no slab */);
 	if (err)
@@ -1790,8 +1788,6 @@ static int __init xsk_init(void)
 	if (err)
 		goto out_pernet;
 
-	for_each_possible_cpu(cpu)
-		INIT_LIST_HEAD(&per_cpu(xskmap_flush_list, cpu));
 	return 0;
 
 out_pernet:
-- 
cgit v1.2.3-58-ga151


From 07b87f9eea0c30675084d50c82532d20168da009 Mon Sep 17 00:00:00 2001
From: Steffen Klassert <steffen.klassert@secunet.com>
Date: Thu, 20 Jun 2024 08:47:24 +0200
Subject: xfrm: Fix unregister netdevice hang on hardware offload.

When offloading xfrm states to hardware, the offloading
device is attached to the skbs secpath. If a skb is free
is deferred, an unregister netdevice hangs because the
netdevice is still refcounted.

Fix this by removing the netdevice from the xfrm states
when the netdevice is unregistered. To find all xfrm states
that need to be cleared we add another list where skbs
linked to that are unlinked from the lists (deleted)
but not yet freed.

Fixes: d77e38e612a0 ("xfrm: Add an IPsec hardware offloading API")
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 include/net/xfrm.h    | 36 +++++++++---------------------
 net/xfrm/xfrm_state.c | 61 +++++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 69 insertions(+), 28 deletions(-)

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 77ebf5bcf0b9..7d4c2235252c 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -178,7 +178,10 @@ struct xfrm_state {
 		struct hlist_node	gclist;
 		struct hlist_node	bydst;
 	};
-	struct hlist_node	bysrc;
+	union {
+		struct hlist_node	dev_gclist;
+		struct hlist_node	bysrc;
+	};
 	struct hlist_node	byspi;
 	struct hlist_node	byseq;
 
@@ -1588,7 +1591,7 @@ void xfrm_state_update_stats(struct net *net);
 static inline void xfrm_dev_state_update_stats(struct xfrm_state *x)
 {
 	struct xfrm_dev_offload *xdo = &x->xso;
-	struct net_device *dev = xdo->dev;
+	struct net_device *dev = READ_ONCE(xdo->dev);
 
 	if (dev && dev->xfrmdev_ops &&
 	    dev->xfrmdev_ops->xdo_dev_state_update_stats)
@@ -1946,13 +1949,16 @@ int xfrm_dev_policy_add(struct net *net, struct xfrm_policy *xp,
 			struct xfrm_user_offload *xuo, u8 dir,
 			struct netlink_ext_ack *extack);
 bool xfrm_dev_offload_ok(struct sk_buff *skb, struct xfrm_state *x);
+void xfrm_dev_state_delete(struct xfrm_state *x);
+void xfrm_dev_state_free(struct xfrm_state *x);
 
 static inline void xfrm_dev_state_advance_esn(struct xfrm_state *x)
 {
 	struct xfrm_dev_offload *xso = &x->xso;
+	struct net_device *dev = READ_ONCE(xso->dev);
 
-	if (xso->dev && xso->dev->xfrmdev_ops->xdo_dev_state_advance_esn)
-		xso->dev->xfrmdev_ops->xdo_dev_state_advance_esn(x);
+	if (dev && dev->xfrmdev_ops->xdo_dev_state_advance_esn)
+		dev->xfrmdev_ops->xdo_dev_state_advance_esn(x);
 }
 
 static inline bool xfrm_dst_offload_ok(struct dst_entry *dst)
@@ -1973,28 +1979,6 @@ static inline bool xfrm_dst_offload_ok(struct dst_entry *dst)
 	return false;
 }
 
-static inline void xfrm_dev_state_delete(struct xfrm_state *x)
-{
-	struct xfrm_dev_offload *xso = &x->xso;
-
-	if (xso->dev)
-		xso->dev->xfrmdev_ops->xdo_dev_state_delete(x);
-}
-
-static inline void xfrm_dev_state_free(struct xfrm_state *x)
-{
-	struct xfrm_dev_offload *xso = &x->xso;
-	struct net_device *dev = xso->dev;
-
-	if (dev && dev->xfrmdev_ops) {
-		if (dev->xfrmdev_ops->xdo_dev_state_free)
-			dev->xfrmdev_ops->xdo_dev_state_free(x);
-		xso->dev = NULL;
-		xso->type = XFRM_DEV_OFFLOAD_UNSPECIFIED;
-		netdev_put(dev, &xso->dev_tracker);
-	}
-}
-
 static inline void xfrm_dev_policy_delete(struct xfrm_policy *x)
 {
 	struct xfrm_dev_offload *xdo = &x->xdo;
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 649bb739df0d..d531d2a1fae2 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -49,6 +49,7 @@ static struct kmem_cache *xfrm_state_cache __ro_after_init;
 
 static DECLARE_WORK(xfrm_state_gc_work, xfrm_state_gc_task);
 static HLIST_HEAD(xfrm_state_gc_list);
+static HLIST_HEAD(xfrm_state_dev_gc_list);
 
 static inline bool xfrm_state_hold_rcu(struct xfrm_state __rcu *x)
 {
@@ -214,6 +215,7 @@ static DEFINE_SPINLOCK(xfrm_state_afinfo_lock);
 static struct xfrm_state_afinfo __rcu *xfrm_state_afinfo[NPROTO];
 
 static DEFINE_SPINLOCK(xfrm_state_gc_lock);
+static DEFINE_SPINLOCK(xfrm_state_dev_gc_lock);
 
 int __xfrm_state_delete(struct xfrm_state *x);
 
@@ -683,6 +685,40 @@ struct xfrm_state *xfrm_state_alloc(struct net *net)
 }
 EXPORT_SYMBOL(xfrm_state_alloc);
 
+#ifdef CONFIG_XFRM_OFFLOAD
+void xfrm_dev_state_delete(struct xfrm_state *x)
+{
+	struct xfrm_dev_offload *xso = &x->xso;
+	struct net_device *dev = READ_ONCE(xso->dev);
+
+	if (dev) {
+		dev->xfrmdev_ops->xdo_dev_state_delete(x);
+		spin_lock_bh(&xfrm_state_dev_gc_lock);
+		hlist_add_head(&x->dev_gclist, &xfrm_state_dev_gc_list);
+		spin_unlock_bh(&xfrm_state_dev_gc_lock);
+	}
+}
+
+void xfrm_dev_state_free(struct xfrm_state *x)
+{
+	struct xfrm_dev_offload *xso = &x->xso;
+	struct net_device *dev = READ_ONCE(xso->dev);
+
+	if (dev && dev->xfrmdev_ops) {
+		spin_lock_bh(&xfrm_state_dev_gc_lock);
+		if (!hlist_unhashed(&x->dev_gclist))
+			hlist_del(&x->dev_gclist);
+		spin_unlock_bh(&xfrm_state_dev_gc_lock);
+
+		if (dev->xfrmdev_ops->xdo_dev_state_free)
+			dev->xfrmdev_ops->xdo_dev_state_free(x);
+		WRITE_ONCE(xso->dev, NULL);
+		xso->type = XFRM_DEV_OFFLOAD_UNSPECIFIED;
+		netdev_put(dev, &xso->dev_tracker);
+	}
+}
+#endif
+
 void __xfrm_state_destroy(struct xfrm_state *x, bool sync)
 {
 	WARN_ON(x->km.state != XFRM_STATE_DEAD);
@@ -848,6 +884,9 @@ EXPORT_SYMBOL(xfrm_state_flush);
 
 int xfrm_dev_state_flush(struct net *net, struct net_device *dev, bool task_valid)
 {
+	struct xfrm_state *x;
+	struct hlist_node *tmp;
+	struct xfrm_dev_offload *xso;
 	int i, err = 0, cnt = 0;
 
 	spin_lock_bh(&net->xfrm.xfrm_state_lock);
@@ -857,8 +896,6 @@ int xfrm_dev_state_flush(struct net *net, struct net_device *dev, bool task_vali
 
 	err = -ESRCH;
 	for (i = 0; i <= net->xfrm.state_hmask; i++) {
-		struct xfrm_state *x;
-		struct xfrm_dev_offload *xso;
 restart:
 		hlist_for_each_entry(x, net->xfrm.state_bydst+i, bydst) {
 			xso = &x->xso;
@@ -868,6 +905,8 @@ restart:
 				spin_unlock_bh(&net->xfrm.xfrm_state_lock);
 
 				err = xfrm_state_delete(x);
+				xfrm_dev_state_free(x);
+
 				xfrm_audit_state_delete(x, err ? 0 : 1,
 							task_valid);
 				xfrm_state_put(x);
@@ -884,6 +923,24 @@ restart:
 
 out:
 	spin_unlock_bh(&net->xfrm.xfrm_state_lock);
+
+	spin_lock_bh(&xfrm_state_dev_gc_lock);
+restart_gc:
+	hlist_for_each_entry_safe(x, tmp, &xfrm_state_dev_gc_list, dev_gclist) {
+		xso = &x->xso;
+
+		if (xso->dev == dev) {
+			spin_unlock_bh(&xfrm_state_dev_gc_lock);
+			xfrm_dev_state_free(x);
+			spin_lock_bh(&xfrm_state_dev_gc_lock);
+			goto restart_gc;
+		}
+
+	}
+	spin_unlock_bh(&xfrm_state_dev_gc_lock);
+
+	xfrm_flush_gc();
+
 	return err;
 }
 EXPORT_SYMBOL(xfrm_dev_state_flush);
-- 
cgit v1.2.3-58-ga151


From 3955802f160b5c61ac00d7e54da8d746f2e4a2d5 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Thu, 20 Jun 2024 13:56:13 -0700
Subject: af_unix: Define locking order for unix_table_double_lock().

When created, AF_UNIX socket is put into net->unx.table.buckets[],
and the hash is stored in sk->sk_hash.

  * unbound socket  : 0 <= sk_hash <= UNIX_HASH_MOD

When bind() is called, the socket could be moved to another bucket.

  * pathname socket : 0 <= sk_hash <= UNIX_HASH_MOD
  * abstract socket : UNIX_HASH_MOD + 1 <= sk_hash <= UNIX_HASH_MOD * 2 + 1

Then, we call unix_table_double_lock() which locks a single bucket
or two.

Let's define the order as unix_table_lock_cmp_fn() instead of using
spin_lock_nested().

The locking is always done in ascending order of sk->sk_hash, which
is the index of buckets/locks array allocated by kvmalloc_array().

  sk_hash_A < sk_hash_B
  <=> &locks[sk_hash_A].dep_map < &locks[sk_hash_B].dep_map

So, the relation of two sk->sk_hash can be derived from the addresses
of dep_map in the array of locks.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Kent Overstreet <kent.overstreet@linux.dev>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/unix/af_unix.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index e9c941e6a464..7889d4723959 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -126,6 +126,15 @@ static spinlock_t bsd_socket_locks[UNIX_HASH_SIZE / 2];
  *    hash table is protected with spinlock.
  *    each socket state is protected by separate spinlock.
  */
+#ifdef CONFIG_PROVE_LOCKING
+#define cmp_ptr(l, r)	(((l) > (r)) - ((l) < (r)))
+
+static int unix_table_lock_cmp_fn(const struct lockdep_map *a,
+				  const struct lockdep_map *b)
+{
+	return cmp_ptr(a, b);
+}
+#endif
 
 static unsigned int unix_unbound_hash(struct sock *sk)
 {
@@ -168,7 +177,7 @@ static void unix_table_double_lock(struct net *net,
 		swap(hash1, hash2);
 
 	spin_lock(&net->unx.table.locks[hash1]);
-	spin_lock_nested(&net->unx.table.locks[hash2], SINGLE_DEPTH_NESTING);
+	spin_lock(&net->unx.table.locks[hash2]);
 }
 
 static void unix_table_double_unlock(struct net *net,
@@ -3578,6 +3587,7 @@ static int __net_init unix_net_init(struct net *net)
 
 	for (i = 0; i < UNIX_HASH_SIZE; i++) {
 		spin_lock_init(&net->unx.table.locks[i]);
+		lock_set_cmp_fn(&net->unx.table.locks[i], unix_table_lock_cmp_fn, NULL);
 		INIT_HLIST_HEAD(&net->unx.table.buckets[i]);
 	}
 
-- 
cgit v1.2.3-58-ga151


From ed99822817cb728eee8786c1c921c69c6be206fe Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Thu, 20 Jun 2024 13:56:14 -0700
Subject: af_unix: Define locking order for U_LOCK_SECOND in
 unix_state_double_lock().

unix_dgram_connect() and unix_dgram_{send,recv}msg() lock the socket
and peer in ascending order of the socket address.

Let's define the order as unix_state_lock_cmp_fn() instead of using
unix_state_lock_nested().

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Kent Overstreet <kent.overstreet@linux.dev>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/unix/af_unix.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 7889d4723959..0657f599bbef 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -134,6 +134,18 @@ static int unix_table_lock_cmp_fn(const struct lockdep_map *a,
 {
 	return cmp_ptr(a, b);
 }
+
+static int unix_state_lock_cmp_fn(const struct lockdep_map *_a,
+				  const struct lockdep_map *_b)
+{
+	const struct unix_sock *a, *b;
+
+	a = container_of(_a, struct unix_sock, lock.dep_map);
+	b = container_of(_b, struct unix_sock, lock.dep_map);
+
+	/* unix_state_double_lock(): ascending address order. */
+	return cmp_ptr(a, b);
+}
 #endif
 
 static unsigned int unix_unbound_hash(struct sock *sk)
@@ -987,6 +999,7 @@ static struct sock *unix_create1(struct net *net, struct socket *sock, int kern,
 	u->path.dentry = NULL;
 	u->path.mnt = NULL;
 	spin_lock_init(&u->lock);
+	lock_set_cmp_fn(&u->lock, unix_state_lock_cmp_fn, NULL);
 	mutex_init(&u->iolock); /* single task reading lock */
 	mutex_init(&u->bindlock); /* single task binding lock */
 	init_waitqueue_head(&u->peer_wait);
@@ -1335,11 +1348,12 @@ static void unix_state_double_lock(struct sock *sk1, struct sock *sk2)
 		unix_state_lock(sk1);
 		return;
 	}
+
 	if (sk1 > sk2)
 		swap(sk1, sk2);
 
 	unix_state_lock(sk1);
-	unix_state_lock_nested(sk2, U_LOCK_SECOND);
+	unix_state_lock(sk2);
 }
 
 static void unix_state_double_unlock(struct sock *sk1, struct sock *sk2)
-- 
cgit v1.2.3-58-ga151


From 1ca27e0c8c13ac50a4acf9cdf77069e2d94a547d Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Thu, 20 Jun 2024 13:56:15 -0700
Subject: af_unix: Don't retry after unix_state_lock_nested() in
 unix_stream_connect().

When a SOCK_(STREAM|SEQPACKET) socket connect()s to another one, we need
to lock the two sockets to check their states in unix_stream_connect().

We use unix_state_lock() for the server and unix_state_lock_nested() for
client with tricky sk->sk_state check to avoid deadlock.

The possible deadlock scenario are the following:

  1) Self connect()
  2) Simultaneous connect()

The former is simple, attempt to grab the same lock, and the latter is
AB-BA deadlock.

After the server's unix_state_lock(), we check the server socket's state,
and if it's not TCP_LISTEN, connect() fails with -EINVAL.

Then, we avoid the former deadlock by checking the client's state before
unix_state_lock_nested().  If its state is not TCP_LISTEN, we can make
sure that the client and the server are not identical based on the state.

Also, the latter deadlock can be avoided in the same way.  Due to the
server sk->sk_state requirement, AB-BA deadlock could happen only with
TCP_LISTEN sockets.  So, if the client's state is TCP_LISTEN, we can
give up the second lock to avoid the deadlock.

  CPU 1                 CPU 2                  CPU 3
  connect(A -> B)       connect(B -> A)        listen(A)
  ---                   ---                    ---
  unix_state_lock(B)
  B->sk_state == TCP_LISTEN
  READ_ONCE(A->sk_state) == TCP_CLOSE
                            ^^^^^^^^^
                            ok, will lock A    unix_state_lock(A)
             .--------------'                  WRITE_ONCE(A->sk_state, TCP_LISTEN)
             |                                 unix_state_unlock(A)
             |
             |          unix_state_lock(A)
             |          A->sk_sk_state == TCP_LISTEN
             |          READ_ONCE(B->sk_state) == TCP_LISTEN
             v                                    ^^^^^^^^^^
  unix_state_lock_nested(A)                       Don't lock B !!

Currently, while checking the client's state, we also check if it's
TCP_ESTABLISHED, but this is unlikely and can be checked after we know
the state is not TCP_CLOSE.

Moreover, if it happens after the second lock, we now jump to the restart
label, but it's unlikely that the server is not found during the retry,
so the jump is mostly to revist the client state check.

Let's remove the retry logic and check the state against TCP_CLOSE first.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/unix/af_unix.c | 34 +++++++++-------------------------
 1 file changed, 9 insertions(+), 25 deletions(-)

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 0657f599bbef..88f2c5d039c4 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -1496,6 +1496,7 @@ static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr,
 	struct unix_sock *u = unix_sk(sk), *newu, *otheru;
 	struct net *net = sock_net(sk);
 	struct sk_buff *skb = NULL;
+	unsigned char state;
 	long timeo;
 	int err;
 
@@ -1546,7 +1547,6 @@ restart:
 		goto out;
 	}
 
-	/* Latch state of peer */
 	unix_state_lock(other);
 
 	/* Apparently VFS overslept socket death. Retry. */
@@ -1576,37 +1576,21 @@ restart:
 		goto restart;
 	}
 
-	/* Latch our state.
-
-	   It is tricky place. We need to grab our state lock and cannot
-	   drop lock on peer. It is dangerous because deadlock is
-	   possible. Connect to self case and simultaneous
-	   attempt to connect are eliminated by checking socket
-	   state. other is TCP_LISTEN, if sk is TCP_LISTEN we
-	   check this before attempt to grab lock.
-
-	   Well, and we have to recheck the state after socket locked.
+	/* self connect and simultaneous connect are eliminated
+	 * by rejecting TCP_LISTEN socket to avoid deadlock.
 	 */
-	switch (READ_ONCE(sk->sk_state)) {
-	case TCP_CLOSE:
-		/* This is ok... continue with connect */
-		break;
-	case TCP_ESTABLISHED:
-		/* Socket is already connected */
-		err = -EISCONN;
-		goto out_unlock;
-	default:
-		err = -EINVAL;
+	state = READ_ONCE(sk->sk_state);
+	if (unlikely(state != TCP_CLOSE)) {
+		err = state == TCP_ESTABLISHED ? -EISCONN : -EINVAL;
 		goto out_unlock;
 	}
 
 	unix_state_lock_nested(sk, U_LOCK_SECOND);
 
-	if (sk->sk_state != TCP_CLOSE) {
+	if (unlikely(sk->sk_state != TCP_CLOSE)) {
+		err = sk->sk_state == TCP_ESTABLISHED ? -EISCONN : -EINVAL;
 		unix_state_unlock(sk);
-		unix_state_unlock(other);
-		sock_put(other);
-		goto restart;
+		goto out_unlock;
 	}
 
 	err = security_unix_stream_connect(sk, other, newsk);
-- 
cgit v1.2.3-58-ga151


From 98f706de445b464f25220360210a4bcb9cc6c41a Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Thu, 20 Jun 2024 13:56:16 -0700
Subject: af_unix: Define locking order for U_LOCK_SECOND in
 unix_stream_connect().

While a SOCK_(STREAM|SEQPACKET) socket connect()s to another, we hold
two locks of them by unix_state_lock() and unix_state_lock_nested() in
unix_stream_connect().

Before unix_state_lock_nested(), the following is guaranteed by checking
sk->sk_state:

  1. The first socket is TCP_LISTEN
  2. The second socket is not the first one
  3. Simultaneous connect() must fail

So, the client state can be TCP_CLOSE or TCP_LISTEN or TCP_ESTABLISHED.

Let's define the expected states as unix_state_lock_cmp_fn() instead of
using unix_state_lock_nested().

Note that 2. is detected by debug_spin_lock_before() and 3. cannot be
expressed as lock_cmp_fn.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/af_unix.h |  1 -
 net/unix/af_unix.c    | 37 ++++++++++++++++++++++++++++++++++++-
 2 files changed, 36 insertions(+), 2 deletions(-)

diff --git a/include/net/af_unix.h b/include/net/af_unix.h
index b6eedf7650da..fd813ad73ab8 100644
--- a/include/net/af_unix.h
+++ b/include/net/af_unix.h
@@ -98,7 +98,6 @@ struct unix_sock {
 #define unix_state_unlock(s)	spin_unlock(&unix_sk(s)->lock)
 enum unix_socket_lock_class {
 	U_LOCK_NORMAL,
-	U_LOCK_SECOND,	/* for double locking, see unix_state_double_lock(). */
 	U_LOCK_DIAG, /* used while dumping icons, see sk_diag_dump_icons(). */
 	U_LOCK_GC_LISTENER, /* used for listening socket while determining gc
 			     * candidates to close a small race window.
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 88f2c5d039c4..a092d6999ae0 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -143,6 +143,41 @@ static int unix_state_lock_cmp_fn(const struct lockdep_map *_a,
 	a = container_of(_a, struct unix_sock, lock.dep_map);
 	b = container_of(_b, struct unix_sock, lock.dep_map);
 
+	if (a->sk.sk_state == TCP_LISTEN) {
+		/* unix_stream_connect(): Before the 2nd unix_state_lock(),
+		 *
+		 *   1. a is TCP_LISTEN.
+		 *   2. b is not a.
+		 *   3. concurrent connect(b -> a) must fail.
+		 *
+		 * Except for 2. & 3., the b's state can be any possible
+		 * value due to concurrent connect() or listen().
+		 *
+		 * 2. is detected in debug_spin_lock_before(), and 3. cannot
+		 * be expressed as lock_cmp_fn.
+		 */
+		switch (b->sk.sk_state) {
+		case TCP_CLOSE:
+		case TCP_ESTABLISHED:
+		case TCP_LISTEN:
+			return -1;
+		default:
+			/* Invalid case. */
+			return 0;
+		}
+	}
+
+	/* Should never happen.  Just to be symmetric. */
+	if (b->sk.sk_state == TCP_LISTEN) {
+		switch (b->sk.sk_state) {
+		case TCP_CLOSE:
+		case TCP_ESTABLISHED:
+			return 1;
+		default:
+			return 0;
+		}
+	}
+
 	/* unix_state_double_lock(): ascending address order. */
 	return cmp_ptr(a, b);
 }
@@ -1585,7 +1620,7 @@ restart:
 		goto out_unlock;
 	}
 
-	unix_state_lock_nested(sk, U_LOCK_SECOND);
+	unix_state_lock(sk);
 
 	if (unlikely(sk->sk_state != TCP_CLOSE)) {
 		err = sk->sk_state == TCP_ESTABLISHED ? -EISCONN : -EINVAL;
-- 
cgit v1.2.3-58-ga151


From b380b18102a0b72a9726077474a2915dfe8bbc08 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Thu, 20 Jun 2024 13:56:17 -0700
Subject: af_unix: Don't acquire unix_state_lock() for sock_i_ino().

sk_diag_dump_peer() and sk_diag_dump() call unix_state_lock() for
sock_i_ino() which reads SOCK_INODE(sk->sk_socket)->i_ino, but it's
protected by sk->sk_callback_lock.

Let's remove unnecessary unix_state_lock().

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/unix/diag.c | 30 ++++++++++--------------------
 1 file changed, 10 insertions(+), 20 deletions(-)

diff --git a/net/unix/diag.c b/net/unix/diag.c
index 937edf4afed4..d2d66727b0da 100644
--- a/net/unix/diag.c
+++ b/net/unix/diag.c
@@ -47,9 +47,7 @@ static int sk_diag_dump_peer(struct sock *sk, struct sk_buff *nlskb)
 
 	peer = unix_peer_get(sk);
 	if (peer) {
-		unix_state_lock(peer);
 		ino = sock_i_ino(peer);
-		unix_state_unlock(peer);
 		sock_put(peer);
 
 		return nla_put_u32(nlskb, UNIX_DIAG_PEER, ino);
@@ -180,22 +178,6 @@ out_nlmsg_trim:
 	return -EMSGSIZE;
 }
 
-static int sk_diag_dump(struct sock *sk, struct sk_buff *skb, struct unix_diag_req *req,
-			struct user_namespace *user_ns,
-			u32 portid, u32 seq, u32 flags)
-{
-	int sk_ino;
-
-	unix_state_lock(sk);
-	sk_ino = sock_i_ino(sk);
-	unix_state_unlock(sk);
-
-	if (!sk_ino)
-		return 0;
-
-	return sk_diag_fill(sk, skb, req, user_ns, portid, seq, flags, sk_ino);
-}
-
 static int unix_diag_dump(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct net *net = sock_net(skb->sk);
@@ -213,14 +195,22 @@ static int unix_diag_dump(struct sk_buff *skb, struct netlink_callback *cb)
 		num = 0;
 		spin_lock(&net->unx.table.locks[slot]);
 		sk_for_each(sk, &net->unx.table.buckets[slot]) {
+			int sk_ino;
+
 			if (num < s_num)
 				goto next;
+
 			if (!(req->udiag_states & (1 << READ_ONCE(sk->sk_state))))
 				goto next;
-			if (sk_diag_dump(sk, skb, req, sk_user_ns(skb->sk),
+
+			sk_ino = sock_i_ino(sk);
+			if (!sk_ino)
+				goto next;
+
+			if (sk_diag_fill(sk, skb, req, sk_user_ns(skb->sk),
 					 NETLINK_CB(cb->skb).portid,
 					 cb->nlh->nlmsg_seq,
-					 NLM_F_MULTI) < 0) {
+					 NLM_F_MULTI, sk_ino) < 0) {
 				spin_unlock(&net->unx.table.locks[slot]);
 				goto done;
 			}
-- 
cgit v1.2.3-58-ga151


From c4da4661d985fd3cbaea3ea6101e2dd0d2ad4b74 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Thu, 20 Jun 2024 13:56:18 -0700
Subject: af_unix: Remove U_LOCK_DIAG.

sk_diag_dump_icons() acquires embryo's lock by unix_state_lock_nested()
to fetch its peer.

The embryo's ->peer is set to NULL only when its parent listener is
close()d.  Then, unix_release_sock() is called for each embryo after
unlinking skb by skb_dequeue().

In sk_diag_dump_icons(), we hold the parent's recvq lock, so we need
not acquire unix_state_lock_nested(), and peer is always non-NULL.

Let's remove unnecessary unix_state_lock_nested() and non-NULL test
for peer.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/af_unix.h |  1 -
 net/unix/diag.c       | 17 +++--------------
 2 files changed, 3 insertions(+), 15 deletions(-)

diff --git a/include/net/af_unix.h b/include/net/af_unix.h
index fd813ad73ab8..c42645199cee 100644
--- a/include/net/af_unix.h
+++ b/include/net/af_unix.h
@@ -98,7 +98,6 @@ struct unix_sock {
 #define unix_state_unlock(s)	spin_unlock(&unix_sk(s)->lock)
 enum unix_socket_lock_class {
 	U_LOCK_NORMAL,
-	U_LOCK_DIAG, /* used while dumping icons, see sk_diag_dump_icons(). */
 	U_LOCK_GC_LISTENER, /* used for listening socket while determining gc
 			     * candidates to close a small race window.
 			     */
diff --git a/net/unix/diag.c b/net/unix/diag.c
index d2d66727b0da..9138af8b465e 100644
--- a/net/unix/diag.c
+++ b/net/unix/diag.c
@@ -73,20 +73,9 @@ static int sk_diag_dump_icons(struct sock *sk, struct sk_buff *nlskb)
 
 		buf = nla_data(attr);
 		i = 0;
-		skb_queue_walk(&sk->sk_receive_queue, skb) {
-			struct sock *req, *peer;
-
-			req = skb->sk;
-			/*
-			 * The state lock is outer for the same sk's
-			 * queue lock. With the other's queue locked it's
-			 * OK to lock the state.
-			 */
-			unix_state_lock_nested(req, U_LOCK_DIAG);
-			peer = unix_sk(req)->peer;
-			buf[i++] = (peer ? sock_i_ino(peer) : 0);
-			unix_state_unlock(req);
-		}
+		skb_queue_walk(&sk->sk_receive_queue, skb)
+			buf[i++] = sock_i_ino(unix_peer(skb->sk));
+
 		spin_unlock(&sk->sk_receive_queue.lock);
 	}
 
-- 
cgit v1.2.3-58-ga151


From 7202cb591624b860bd7b688d4bd0f5bee79c7f44 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Thu, 20 Jun 2024 13:56:19 -0700
Subject: af_unix: Remove U_LOCK_GC_LISTENER.

Commit 1971d13ffa84 ("af_unix: Suppress false-positive lockdep splat for
spin_lock() in __unix_gc().") added U_LOCK_GC_LISTENER for the old GC,
but it's no longer needed for the new GC.

Let's remove U_LOCK_GC_LISTENER and unix_state_lock_nested() as there's
no user.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/af_unix.h | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/include/net/af_unix.h b/include/net/af_unix.h
index c42645199cee..63129c79b8cb 100644
--- a/include/net/af_unix.h
+++ b/include/net/af_unix.h
@@ -96,18 +96,6 @@ struct unix_sock {
 
 #define unix_state_lock(s)	spin_lock(&unix_sk(s)->lock)
 #define unix_state_unlock(s)	spin_unlock(&unix_sk(s)->lock)
-enum unix_socket_lock_class {
-	U_LOCK_NORMAL,
-	U_LOCK_GC_LISTENER, /* used for listening socket while determining gc
-			     * candidates to close a small race window.
-			     */
-};
-
-static inline void unix_state_lock_nested(struct sock *sk,
-				   enum unix_socket_lock_class subclass)
-{
-	spin_lock_nested(&unix_sk(sk)->lock, subclass);
-}
 
 #define peer_wait peer_wq.wait
 
-- 
cgit v1.2.3-58-ga151


From 8647ece4814f3bfdb5f7a8e19f882c9b89299a07 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Thu, 20 Jun 2024 13:56:20 -0700
Subject: af_unix: Define locking order for U_RECVQ_LOCK_EMBRYO in
 unix_collect_skb().

While GC is cleaning up cyclic references by SCM_RIGHTS,
unix_collect_skb() collects skb in the socket's recvq.

If the socket is TCP_LISTEN, we need to collect skb in the
embryo's queue.  Then, both the listener's recvq lock and
the embroy's one are held.

The locking is always done in the listener -> embryo order.

Let's define it as unix_recvq_lock_cmp_fn() instead of using
spin_lock_nested().

Note that the reverse order is defined for consistency.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/unix/af_unix.c | 21 +++++++++++++++++++++
 net/unix/garbage.c |  8 +-------
 2 files changed, 22 insertions(+), 7 deletions(-)

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index a092d6999ae0..89675879038d 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -181,6 +181,25 @@ static int unix_state_lock_cmp_fn(const struct lockdep_map *_a,
 	/* unix_state_double_lock(): ascending address order. */
 	return cmp_ptr(a, b);
 }
+
+static int unix_recvq_lock_cmp_fn(const struct lockdep_map *_a,
+				  const struct lockdep_map *_b)
+{
+	const struct sock *a, *b;
+
+	a = container_of(_a, struct sock, sk_receive_queue.lock.dep_map);
+	b = container_of(_b, struct sock, sk_receive_queue.lock.dep_map);
+
+	/* unix_collect_skb(): listener -> embryo order. */
+	if (a->sk_state == TCP_LISTEN && unix_sk(b)->listener == a)
+		return -1;
+
+	/* Should never happen.  Just to be symmetric. */
+	if (b->sk_state == TCP_LISTEN && unix_sk(a)->listener == b)
+		return 1;
+
+	return 0;
+}
 #endif
 
 static unsigned int unix_unbound_hash(struct sock *sk)
@@ -1028,6 +1047,8 @@ static struct sock *unix_create1(struct net *net, struct socket *sock, int kern,
 	sk->sk_write_space	= unix_write_space;
 	sk->sk_max_ack_backlog	= READ_ONCE(net->unx.sysctl_max_dgram_qlen);
 	sk->sk_destruct		= unix_sock_destructor;
+	lock_set_cmp_fn(&sk->sk_receive_queue.lock, unix_recvq_lock_cmp_fn, NULL);
+
 	u = unix_sk(sk);
 	u->listener = NULL;
 	u->vertex = NULL;
diff --git a/net/unix/garbage.c b/net/unix/garbage.c
index dfe94a90ece4..eb8aa5171a68 100644
--- a/net/unix/garbage.c
+++ b/net/unix/garbage.c
@@ -337,11 +337,6 @@ static bool unix_vertex_dead(struct unix_vertex *vertex)
 	return true;
 }
 
-enum unix_recv_queue_lock_class {
-	U_RECVQ_LOCK_NORMAL,
-	U_RECVQ_LOCK_EMBRYO,
-};
-
 static void unix_collect_queue(struct unix_sock *u, struct sk_buff_head *hitlist)
 {
 	skb_queue_splice_init(&u->sk.sk_receive_queue, hitlist);
@@ -375,8 +370,7 @@ static void unix_collect_skb(struct list_head *scc, struct sk_buff_head *hitlist
 			skb_queue_walk(queue, skb) {
 				struct sk_buff_head *embryo_queue = &skb->sk->sk_receive_queue;
 
-				/* listener -> embryo order, the inversion never happens. */
-				spin_lock_nested(&embryo_queue->lock, U_RECVQ_LOCK_EMBRYO);
+				spin_lock(&embryo_queue->lock);
 				unix_collect_queue(unix_sk(skb->sk), hitlist);
 				spin_unlock(&embryo_queue->lock);
 			}
-- 
cgit v1.2.3-58-ga151


From faf489e6896d645a679d3d90a2d1d5d12c6b3e13 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Thu, 20 Jun 2024 13:56:21 -0700
Subject: af_unix: Set sk_peer_pid/sk_peer_cred locklessly for new socket.

init_peercred() is called in 3 places:

  1. socketpair() : both sockets
  2. connect()    : child socket
  3. listen()     : listening socket

The first two need not hold sk_peer_lock because no one can
touch the socket.

Let's set cred/pid without holding lock for the two cases and
rename the old init_peercred() to update_peercred() to properly
reflect the use case.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/unix/af_unix.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 89675879038d..d11664c2faad 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -750,6 +750,12 @@ static void unix_release_sock(struct sock *sk, int embrion)
 }
 
 static void init_peercred(struct sock *sk)
+{
+	sk->sk_peer_pid = get_pid(task_tgid(current));
+	sk->sk_peer_cred = get_current_cred();
+}
+
+static void update_peercred(struct sock *sk)
 {
 	const struct cred *old_cred;
 	struct pid *old_pid;
@@ -757,8 +763,7 @@ static void init_peercred(struct sock *sk)
 	spin_lock(&sk->sk_peer_lock);
 	old_pid = sk->sk_peer_pid;
 	old_cred = sk->sk_peer_cred;
-	sk->sk_peer_pid  = get_pid(task_tgid(current));
-	sk->sk_peer_cred = get_current_cred();
+	init_peercred(sk);
 	spin_unlock(&sk->sk_peer_lock);
 
 	put_pid(old_pid);
@@ -810,7 +815,7 @@ static int unix_listen(struct socket *sock, int backlog)
 	WRITE_ONCE(sk->sk_state, TCP_LISTEN);
 
 	/* set credentials so connect can copy them */
-	init_peercred(sk);
+	update_peercred(sk);
 	err = 0;
 
 out_unlock:
-- 
cgit v1.2.3-58-ga151


From e4bd881d987121dbf1a288641491955a53d9f8f7 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Thu, 20 Jun 2024 13:56:22 -0700
Subject: af_unix: Remove put_pid()/put_cred() in copy_peercred().

When (AF_UNIX, SOCK_STREAM) socket connect()s to a listening socket,
the listener's sk_peer_pid/sk_peer_cred are copied to the client in
copy_peercred().

Then, the client's sk_peer_pid and sk_peer_cred are always NULL, so
we need not call put_pid() and put_cred() there.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/unix/af_unix.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index d11664c2faad..3d0ace7ca017 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -772,9 +772,6 @@ static void update_peercred(struct sock *sk)
 
 static void copy_peercred(struct sock *sk, struct sock *peersk)
 {
-	const struct cred *old_cred;
-	struct pid *old_pid;
-
 	if (sk < peersk) {
 		spin_lock(&sk->sk_peer_lock);
 		spin_lock_nested(&peersk->sk_peer_lock, SINGLE_DEPTH_NESTING);
@@ -782,16 +779,12 @@ static void copy_peercred(struct sock *sk, struct sock *peersk)
 		spin_lock(&peersk->sk_peer_lock);
 		spin_lock_nested(&sk->sk_peer_lock, SINGLE_DEPTH_NESTING);
 	}
-	old_pid = sk->sk_peer_pid;
-	old_cred = sk->sk_peer_cred;
+
 	sk->sk_peer_pid  = get_pid(peersk->sk_peer_pid);
 	sk->sk_peer_cred = get_cred(peersk->sk_peer_cred);
 
 	spin_unlock(&sk->sk_peer_lock);
 	spin_unlock(&peersk->sk_peer_lock);
-
-	put_pid(old_pid);
-	put_cred(old_cred);
 }
 
 static int unix_listen(struct socket *sock, int backlog)
-- 
cgit v1.2.3-58-ga151


From 22e5751b0524fedd4f345412d8d3394387471ab7 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Thu, 20 Jun 2024 13:56:23 -0700
Subject: af_unix: Don't use spin_lock_nested() in copy_peercred().

When (AF_UNIX, SOCK_STREAM) socket connect()s to a listening socket,
the listener's sk_peer_pid/sk_peer_cred are copied to the client in
copy_peercred().

Then, two sk_peer_locks are held there; one is client's and another
is listener's.

However, the latter is not needed because we hold the listner's
unix_state_lock() there and unix_listen() cannot update the cred
concurrently.

Let's drop the unnecessary spin_lock() and use the bare spin_lock()
for the client to protect concurrent read by getsockopt(SO_PEERCRED).

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/unix/af_unix.c | 13 +++----------
 1 file changed, 3 insertions(+), 10 deletions(-)

diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 3d0ace7ca017..103a7909cb1a 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -772,19 +772,12 @@ static void update_peercred(struct sock *sk)
 
 static void copy_peercred(struct sock *sk, struct sock *peersk)
 {
-	if (sk < peersk) {
-		spin_lock(&sk->sk_peer_lock);
-		spin_lock_nested(&peersk->sk_peer_lock, SINGLE_DEPTH_NESTING);
-	} else {
-		spin_lock(&peersk->sk_peer_lock);
-		spin_lock_nested(&sk->sk_peer_lock, SINGLE_DEPTH_NESTING);
-	}
+	lockdep_assert_held(&unix_sk(peersk)->lock);
 
-	sk->sk_peer_pid  = get_pid(peersk->sk_peer_pid);
+	spin_lock(&sk->sk_peer_lock);
+	sk->sk_peer_pid = get_pid(peersk->sk_peer_pid);
 	sk->sk_peer_cred = get_cred(peersk->sk_peer_cred);
-
 	spin_unlock(&sk->sk_peer_lock);
-	spin_unlock(&peersk->sk_peer_lock);
 }
 
 static int unix_listen(struct socket *sock, int backlog)
-- 
cgit v1.2.3-58-ga151


From 759cc793ebfc2d1a02f357ae97e5dcdcd63f758f Mon Sep 17 00:00:00 2001
From: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Date: Fri, 21 Jun 2024 10:27:32 +0530
Subject: net: macb: queue tie-off or disable during WOL suspend

When GEM is used as a wake device, it is not mandatory for the RX DMA
to be active. The RX engine in IP only needs to receive and identify
a wake packet through an interrupt. The wake packet is of no further
significance; hence, it is not required to be copied into memory.
By disabling RX DMA during suspend, we can avoid unnecessary DMA
processing of any incoming traffic.

During suspend, perform either of the below operations:

- tie-off/dummy descriptor: Disable unused queues by connecting
  them to a looped descriptor chain without free slots.

- queue disable: The newer IP version allows disabling individual queues.

Co-developed-by: Harini Katakam <harini.katakam@amd.com>
Signed-off-by: Harini Katakam <harini.katakam@amd.com>
Signed-off-by: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Claudiu Beznea <claudiu.beznea@tuxon.dev>
Tested-by: Claudiu Beznea <claudiu.beznea@tuxon.dev> # on SAMA7G5
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/cadence/macb.h      |  7 ++++
 drivers/net/ethernet/cadence/macb_main.c | 60 ++++++++++++++++++++++++++++++--
 2 files changed, 64 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index aa5700ac9c00..50cd35ef21ad 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -645,6 +645,10 @@
 #define GEM_T2OFST_OFFSET			0 /* offset value */
 #define GEM_T2OFST_SIZE				7
 
+/* Bitfields in queue pointer registers */
+#define MACB_QUEUE_DISABLE_OFFSET		0 /* disable queue */
+#define MACB_QUEUE_DISABLE_SIZE			1
+
 /* Offset for screener type 2 compare values (T2CMPOFST).
  * Note the offset is applied after the specified point,
  * e.g. GEM_T2COMPOFST_ETYPE denotes the EtherType field, so an offset
@@ -733,6 +737,7 @@
 #define MACB_CAPS_NEEDS_RSTONUBR		0x00000100
 #define MACB_CAPS_MIIONRGMII			0x00000200
 #define MACB_CAPS_NEED_TSUCLK			0x00000400
+#define MACB_CAPS_QUEUE_DISABLE			0x00000800
 #define MACB_CAPS_PCS				0x01000000
 #define MACB_CAPS_HIGH_SPEED			0x02000000
 #define MACB_CAPS_CLK_HW_CHG			0x04000000
@@ -1254,6 +1259,8 @@ struct macb {
 	u32	(*macb_reg_readl)(struct macb *bp, int offset);
 	void	(*macb_reg_writel)(struct macb *bp, int offset, u32 value);
 
+	struct macb_dma_desc	*rx_ring_tieoff;
+	dma_addr_t		rx_ring_tieoff_dma;
 	size_t			rx_buffer_size;
 
 	unsigned int		rx_ring_size;
diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index 241ce9a2fa99..9fc8c5a82bf8 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -2477,6 +2477,12 @@ static void macb_free_consistent(struct macb *bp)
 	unsigned int q;
 	int size;
 
+	if (bp->rx_ring_tieoff) {
+		dma_free_coherent(&bp->pdev->dev, macb_dma_desc_get_size(bp),
+				  bp->rx_ring_tieoff, bp->rx_ring_tieoff_dma);
+		bp->rx_ring_tieoff = NULL;
+	}
+
 	bp->macbgem_ops.mog_free_rx_buffers(bp);
 
 	for (q = 0, queue = bp->queues; q < bp->num_queues; ++q, ++queue) {
@@ -2568,6 +2574,16 @@ static int macb_alloc_consistent(struct macb *bp)
 	if (bp->macbgem_ops.mog_alloc_rx_buffers(bp))
 		goto out_err;
 
+	/* Required for tie off descriptor for PM cases */
+	if (!(bp->caps & MACB_CAPS_QUEUE_DISABLE)) {
+		bp->rx_ring_tieoff = dma_alloc_coherent(&bp->pdev->dev,
+							macb_dma_desc_get_size(bp),
+							&bp->rx_ring_tieoff_dma,
+							GFP_KERNEL);
+		if (!bp->rx_ring_tieoff)
+			goto out_err;
+	}
+
 	return 0;
 
 out_err:
@@ -2575,6 +2591,19 @@ out_err:
 	return -ENOMEM;
 }
 
+static void macb_init_tieoff(struct macb *bp)
+{
+	struct macb_dma_desc *desc = bp->rx_ring_tieoff;
+
+	if (bp->caps & MACB_CAPS_QUEUE_DISABLE)
+		return;
+	/* Setup a wrapping descriptor with no free slots
+	 * (WRAP and USED) to tie off/disable unused RX queues.
+	 */
+	macb_set_addr(bp, desc, MACB_BIT(RX_WRAP) | MACB_BIT(RX_USED));
+	desc->ctrl = 0;
+}
+
 static void gem_init_rings(struct macb *bp)
 {
 	struct macb_queue *queue;
@@ -2598,6 +2627,7 @@ static void gem_init_rings(struct macb *bp)
 		gem_rx_refill(queue);
 	}
 
+	macb_init_tieoff(bp);
 }
 
 static void macb_init_rings(struct macb *bp)
@@ -2615,6 +2645,8 @@ static void macb_init_rings(struct macb *bp)
 	bp->queues[0].tx_head = 0;
 	bp->queues[0].tx_tail = 0;
 	desc->ctrl |= MACB_BIT(TX_WRAP);
+
+	macb_init_tieoff(bp);
 }
 
 static void macb_reset_hw(struct macb *bp)
@@ -5215,6 +5247,7 @@ static int __maybe_unused macb_suspend(struct device *dev)
 	unsigned long flags;
 	unsigned int q;
 	int err;
+	u32 tmp;
 
 	if (!device_may_wakeup(&bp->dev->dev))
 		phy_exit(bp->sgmii_phy);
@@ -5224,17 +5257,38 @@ static int __maybe_unused macb_suspend(struct device *dev)
 
 	if (bp->wol & MACB_WOL_ENABLED) {
 		spin_lock_irqsave(&bp->lock, flags);
-		/* Flush all status bits */
-		macb_writel(bp, TSR, -1);
-		macb_writel(bp, RSR, -1);
+
+		/* Disable Tx and Rx engines before  disabling the queues,
+		 * this is mandatory as per the IP spec sheet
+		 */
+		tmp = macb_readl(bp, NCR);
+		macb_writel(bp, NCR, tmp & ~(MACB_BIT(TE) | MACB_BIT(RE)));
 		for (q = 0, queue = bp->queues; q < bp->num_queues;
 		     ++q, ++queue) {
+			/* Disable RX queues */
+			if (bp->caps & MACB_CAPS_QUEUE_DISABLE) {
+				queue_writel(queue, RBQP, MACB_BIT(QUEUE_DISABLE));
+			} else {
+				/* Tie off RX queues */
+				queue_writel(queue, RBQP,
+					     lower_32_bits(bp->rx_ring_tieoff_dma));
+#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
+				queue_writel(queue, RBQPH,
+					     upper_32_bits(bp->rx_ring_tieoff_dma));
+#endif
+			}
 			/* Disable all interrupts */
 			queue_writel(queue, IDR, -1);
 			queue_readl(queue, ISR);
 			if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE)
 				queue_writel(queue, ISR, -1);
 		}
+		/* Enable Receive engine */
+		macb_writel(bp, NCR, tmp | MACB_BIT(RE));
+		/* Flush all status bits */
+		macb_writel(bp, TSR, -1);
+		macb_writel(bp, RSR, -1);
+
 		/* Change interrupt handler and
 		 * Enable WoL IRQ on queue 0
 		 */
-- 
cgit v1.2.3-58-ga151


From 3650a8cc5b34b1032428724c1b74349a753237a4 Mon Sep 17 00:00:00 2001
From: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Date: Fri, 21 Jun 2024 10:27:33 +0530
Subject: net: macb: Enable queue disable

Enable queue disable for Versal devices.

Signed-off-by: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Claudiu Beznea <claudiu.beznea@tuxon.dev>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/cadence/macb_main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index 9fc8c5a82bf8..4007b291526f 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -4949,7 +4949,8 @@ static const struct macb_config sama7g5_emac_config = {
 
 static const struct macb_config versal_config = {
 	.caps = MACB_CAPS_GIGABIT_MODE_AVAILABLE | MACB_CAPS_JUMBO |
-		MACB_CAPS_GEM_HAS_PTP | MACB_CAPS_BD_RD_PREFETCH | MACB_CAPS_NEED_TSUCLK,
+		MACB_CAPS_GEM_HAS_PTP | MACB_CAPS_BD_RD_PREFETCH | MACB_CAPS_NEED_TSUCLK |
+		MACB_CAPS_QUEUE_DISABLE,
 	.dma_burst_length = 16,
 	.clk_init = macb_clk_init,
 	.init = init_reset_optional,
-- 
cgit v1.2.3-58-ga151


From 0cb8de39a776e093c759e21d2c95df049f6ea9e9 Mon Sep 17 00:00:00 2001
From: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Date: Fri, 21 Jun 2024 10:27:34 +0530
Subject: net: macb: Add ARP support to WOL

Extend wake-on LAN support with an ARP packet.

Currently, if PHY supports WOL, ethtool ignores the modes supported
by MACB. This change extends the WOL modes with MACB supported modes.

Advertise wake-on LAN supported modes by default without relying on
dt node. By default, wake-on LAN will be in disabled state.
Using ethtool, users can enable/disable or choose packet types.

For wake-on LAN via ARP, ensure the IP address is assigned and
report an error otherwise.

Co-developed-by: Harini Katakam <harini.katakam@amd.com>
Signed-off-by: Harini Katakam <harini.katakam@amd.com>
Signed-off-by: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Claudiu Beznea <claudiu.beznea@tuxon.dev>
Tested-by: Claudiu Beznea <claudiu.beznea@tuxon.dev> # on SAMA7G5
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/cadence/macb.h      |  1 +
 drivers/net/ethernet/cadence/macb_main.c | 58 ++++++++++++++++++--------------
 2 files changed, 33 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index 50cd35ef21ad..122663ff7834 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -1306,6 +1306,7 @@ struct macb {
 	unsigned int		jumbo_max_len;
 
 	u32			wol;
+	u32			wolopts;
 
 	/* holds value of rx watermark value for pbuf_rxcutthru register */
 	u32			rx_watermark;
diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index 4007b291526f..cecc3d6e630f 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -38,6 +38,7 @@
 #include <linux/ptp_classify.h>
 #include <linux/reset.h>
 #include <linux/firmware/xlnx-zynqmp.h>
+#include <linux/inetdevice.h>
 #include "macb.h"
 
 /* This structure is only used for MACB on SiFive FU540 devices */
@@ -84,8 +85,7 @@ struct sifive_fu540_macb_mgmt {
 #define GEM_MTU_MIN_SIZE	ETH_MIN_MTU
 #define MACB_NETIF_LSO		NETIF_F_TSO
 
-#define MACB_WOL_HAS_MAGIC_PACKET	(0x1 << 0)
-#define MACB_WOL_ENABLED		(0x1 << 1)
+#define MACB_WOL_ENABLED		BIT(0)
 
 #define HS_SPEED_10000M			4
 #define MACB_SERDES_RATE_10G		1
@@ -3278,13 +3278,11 @@ static void macb_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 {
 	struct macb *bp = netdev_priv(netdev);
 
-	if (bp->wol & MACB_WOL_HAS_MAGIC_PACKET) {
-		phylink_ethtool_get_wol(bp->phylink, wol);
-		wol->supported |= WAKE_MAGIC;
+	phylink_ethtool_get_wol(bp->phylink, wol);
+	wol->supported |= (WAKE_MAGIC | WAKE_ARP);
 
-		if (bp->wol & MACB_WOL_ENABLED)
-			wol->wolopts |= WAKE_MAGIC;
-	}
+	/* Add macb wolopts to phy wolopts */
+	wol->wolopts |= bp->wolopts;
 }
 
 static int macb_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
@@ -3294,22 +3292,15 @@ static int macb_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 
 	/* Pass the order to phylink layer */
 	ret = phylink_ethtool_set_wol(bp->phylink, wol);
-	/* Don't manage WoL on MAC if handled by the PHY
-	 * or if there's a failure in talking to the PHY
-	 */
-	if (!ret || ret != -EOPNOTSUPP)
+	/* Don't manage WoL on MAC, if PHY set_wol() fails */
+	if (ret && ret != -EOPNOTSUPP)
 		return ret;
 
-	if (!(bp->wol & MACB_WOL_HAS_MAGIC_PACKET) ||
-	    (wol->wolopts & ~WAKE_MAGIC))
-		return -EOPNOTSUPP;
-
-	if (wol->wolopts & WAKE_MAGIC)
-		bp->wol |= MACB_WOL_ENABLED;
-	else
-		bp->wol &= ~MACB_WOL_ENABLED;
+	bp->wolopts = (wol->wolopts & WAKE_MAGIC) ? WAKE_MAGIC : 0;
+	bp->wolopts |= (wol->wolopts & WAKE_ARP) ? WAKE_ARP : 0;
+	bp->wol = (wol->wolopts) ? MACB_WOL_ENABLED : 0;
 
-	device_set_wakeup_enable(&bp->pdev->dev, bp->wol & MACB_WOL_ENABLED);
+	device_set_wakeup_enable(&bp->pdev->dev, bp->wol);
 
 	return 0;
 }
@@ -5086,9 +5077,7 @@ static int macb_probe(struct platform_device *pdev)
 		bp->max_tx_length = GEM_MAX_TX_LEN;
 
 	bp->wol = 0;
-	if (of_property_read_bool(np, "magic-packet"))
-		bp->wol |= MACB_WOL_HAS_MAGIC_PACKET;
-	device_set_wakeup_capable(&pdev->dev, bp->wol & MACB_WOL_HAS_MAGIC_PACKET);
+	device_set_wakeup_capable(&pdev->dev, 1);
 
 	bp->usrio = macb_config->usrio;
 
@@ -5244,7 +5233,9 @@ static int __maybe_unused macb_suspend(struct device *dev)
 {
 	struct net_device *netdev = dev_get_drvdata(dev);
 	struct macb *bp = netdev_priv(netdev);
+	struct in_ifaddr *ifa = NULL;
 	struct macb_queue *queue;
+	struct in_device *idev;
 	unsigned long flags;
 	unsigned int q;
 	int err;
@@ -5257,6 +5248,14 @@ static int __maybe_unused macb_suspend(struct device *dev)
 		return 0;
 
 	if (bp->wol & MACB_WOL_ENABLED) {
+		/* Check for IP address in WOL ARP mode */
+		idev = __in_dev_get_rcu(bp->dev);
+		if (idev && idev->ifa_list)
+			ifa = rcu_access_pointer(idev->ifa_list);
+		if ((bp->wolopts & WAKE_ARP) && !ifa) {
+			netdev_err(netdev, "IP address not assigned as required by WoL walk ARP\n");
+			return -EOPNOTSUPP;
+		}
 		spin_lock_irqsave(&bp->lock, flags);
 
 		/* Disable Tx and Rx engines before  disabling the queues,
@@ -5290,6 +5289,13 @@ static int __maybe_unused macb_suspend(struct device *dev)
 		macb_writel(bp, TSR, -1);
 		macb_writel(bp, RSR, -1);
 
+		tmp = (bp->wolopts & WAKE_MAGIC) ? MACB_BIT(MAG) : 0;
+		if (bp->wolopts & WAKE_ARP) {
+			tmp |= MACB_BIT(ARP);
+			/* write IP address into register */
+			tmp |= MACB_BFEXT(IP, be32_to_cpu(ifa->ifa_local));
+		}
+
 		/* Change interrupt handler and
 		 * Enable WoL IRQ on queue 0
 		 */
@@ -5305,7 +5311,7 @@ static int __maybe_unused macb_suspend(struct device *dev)
 				return err;
 			}
 			queue_writel(bp->queues, IER, GEM_BIT(WOL));
-			gem_writel(bp, WOL, MACB_BIT(MAG));
+			gem_writel(bp, WOL, tmp);
 		} else {
 			err = devm_request_irq(dev, bp->queues[0].irq, macb_wol_interrupt,
 					       IRQF_SHARED, netdev->name, bp->queues);
@@ -5317,7 +5323,7 @@ static int __maybe_unused macb_suspend(struct device *dev)
 				return err;
 			}
 			queue_writel(bp->queues, IER, MACB_BIT(WOL));
-			macb_writel(bp, WOL, MACB_BIT(MAG));
+			macb_writel(bp, WOL, tmp);
 		}
 		spin_unlock_irqrestore(&bp->lock, flags);
 
-- 
cgit v1.2.3-58-ga151


From 783bfe279e5427fdcda54bc269005f39ac94d202 Mon Sep 17 00:00:00 2001
From: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Date: Fri, 21 Jun 2024 10:27:35 +0530
Subject: dt-bindings: net: cdns,macb: Deprecate magic-packet property

WOL modes such as magic-packet should be an OS policy.
By default, advertise supported modes and use ethtool to activate
the required mode.

Suggested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Vineeth Karumanchi <vineeth.karumanchi@amd.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 Documentation/devicetree/bindings/net/cdns,macb.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/net/cdns,macb.yaml b/Documentation/devicetree/bindings/net/cdns,macb.yaml
index 2c71e2cf3a2f..3c30dd23cd4e 100644
--- a/Documentation/devicetree/bindings/net/cdns,macb.yaml
+++ b/Documentation/devicetree/bindings/net/cdns,macb.yaml
@@ -146,6 +146,7 @@ patternProperties:
 
       magic-packet:
         type: boolean
+        deprecated: true
         description:
           Indicates that the hardware supports waking up via magic packet.
 
-- 
cgit v1.2.3-58-ga151


From 73cfd947dbdb25ef9863ac49c4596a7d53ad4025 Mon Sep 17 00:00:00 2001
From: Elad Yifee <eladwf@gmail.com>
Date: Sun, 23 Jun 2024 20:51:09 +0300
Subject: net: ethernet: mtk_eth_soc: ppe: prevent ppe update for non-mtk
 devices

Introduce an additional validation to ensure that the PPE index
is modified exclusively for mtk_eth ingress devices.
This primarily addresses the issue related
to WED operation with multiple PPEs.

Fixes: dee4dd10c79a ("net: ethernet: mtk_eth_soc: ppe: add support for multiple PPEs")
Signed-off-by: Elad Yifee <eladwf@gmail.com>
Link: https://lore.kernel.org/r/20240623175113.24437-1-eladwf@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/mediatek/mtk_ppe_offload.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
index f80af73d0a1b..f20bb390df3a 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
@@ -266,7 +266,7 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f,
 		flow_rule_match_meta(rule, &match);
 		if (mtk_is_netsys_v2_or_greater(eth)) {
 			idev = __dev_get_by_index(&init_net, match.key->ingress_ifindex);
-			if (idev) {
+			if (idev && idev->netdev_ops == eth->netdev[0]->netdev_ops) {
 				struct mtk_mac *mac = netdev_priv(idev);
 
 				if (WARN_ON(mac->ppe_idx >= eth->soc->ppe_num))
-- 
cgit v1.2.3-58-ga151


From a8a8d89dbd2bd2b762b6d0226a1201ec33f7aeac Mon Sep 17 00:00:00 2001
From: James Chapman <jchapman@katalix.com>
Date: Mon, 24 Jun 2024 09:29:45 +0100
Subject: l2tp: remove incorrect __rcu attribute

This fixes a sparse warning.

Fixes: d18d3f0a24fc ("l2tp: replace hlist with simple list for per-tunnel session list")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202406220754.evK8Hrjw-lkp@intel.com/
Signed-off-by: James Chapman <jchapman@katalix.com>
Link: https://patch.msgid.link/20240624082945.1925009-1-jchapman@katalix.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/l2tp/l2tp_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index be4bcbf291a1..64f446f0930b 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -1290,7 +1290,7 @@ static void l2tp_session_unhash(struct l2tp_session *session)
 static void l2tp_tunnel_closeall(struct l2tp_tunnel *tunnel)
 {
 	struct l2tp_session *session;
-	struct list_head __rcu *pos;
+	struct list_head *pos;
 	struct list_head *tmp;
 
 	spin_lock_bh(&tunnel->list_lock);
-- 
cgit v1.2.3-58-ga151


From c4532232fa2a4f8d9b9a88135a666545157f3d13 Mon Sep 17 00:00:00 2001
From: Yujie Liu <yujie.liu@intel.com>
Date: Mon, 24 Jun 2024 13:55:39 +0800
Subject: selftests: net: remove unneeded IP_GRE config

It seems that there is no definition for config IP_GRE, and it is not a
dependency of other configs, so remove it.

linux$ find -name Kconfig | xargs grep "IP_GRE"
<-- nothing

There is a IPV6_GRE config defined in net/ipv6/Kconfig. It only depends
on NET_IPGRE_DEMUX but not IP_GRE.

Signed-off-by: Yujie Liu <yujie.liu@intel.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Link: https://patch.msgid.link/20240624055539.2092322-1-yujie.liu@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/config | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tools/testing/selftests/net/config b/tools/testing/selftests/net/config
index d4891f7a2bfa..f46e27cd1e70 100644
--- a/tools/testing/selftests/net/config
+++ b/tools/testing/selftests/net/config
@@ -26,7 +26,6 @@ CONFIG_INET_ESP=y
 CONFIG_INET_ESP_OFFLOAD=y
 CONFIG_NET_FOU=y
 CONFIG_NET_FOU_IP_TUNNELS=y
-CONFIG_IP_GRE=m
 CONFIG_NETFILTER=y
 CONFIG_NETFILTER_ADVANCED=y
 CONFIG_NF_CONNTRACK=m
-- 
cgit v1.2.3-58-ga151


From 605efd54b50437ed9f3915690539d0afddca9d95 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Mon, 13 May 2024 15:00:41 +0200
Subject: netfilter: nf_tables: make struct nft_trans first member of derived
 subtypes

There is 'struct nft_trans', the basic structure for all transactional
objects, and the the various different transactional objects, such as
nft_trans_table, chain, set, set_elem and so on.

Right now 'struct nft_trans' uses a flexible member at the tail
(data[]), and casting is needed to access the actual type-specific
members.

Change this to make the hierarchy visible in source code, i.e. make
struct nft_trans the first member of all derived subtypes.

This has several advantages:
1. pahole output reflects the real size needed by the particular subtype
2. allows to use container_of() to convert the base type to the actual
   object type instead of casting ->data to the overlay structure.
3. It makes it easy to add intermediate types.

'struct nft_trans' contains a 'binding_list' that is only needed
by two subtypes, so it should be part of the two subtypes, not in
the base structure.

But that makes it hard to interate over the binding_list, because
there is no common base structure.

A follow patch moves the bind list to a new struct:

 struct nft_trans_binding {
   struct nft_trans nft_trans;
   struct list_head binding_list;
 };

... and makes that structure the new 'first member' for both
nft_trans_chain and nft_trans_set.

No functional change intended in this patch.

Some numbers:
 struct nft_trans { /* size: 88, cachelines: 2, members: 5 */
 struct nft_trans_chain { /* size: 152, cachelines: 3, members: 10 */
 struct nft_trans_elem { /* size: 112, cachelines: 2, members: 4 */
 struct nft_trans_flowtable { /* size: 128, cachelines: 2, members: 5 */
 struct nft_trans_obj { /* size: 112, cachelines: 2, members: 4 */
 struct nft_trans_rule { /* size: 112, cachelines: 2, members: 5 */
 struct nft_trans_set { /* size: 120, cachelines: 2, members: 8 */
 struct nft_trans_table { /* size: 96, cachelines: 2, members: 2 */

Of particular interest is nft_trans_elem, which needs to be allocated
once for each pending (to be added or removed) set element.

Add BUILD_BUG_ON to check struct nft_trans is placed at the top of
the container structure.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h | 162 ++++++++++++++++++++++----------------
 net/netfilter/nf_tables_api.c     |  18 +++--
 2 files changed, 105 insertions(+), 75 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 2796153b03da..b25df037fceb 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1608,14 +1608,16 @@ static inline int nft_set_elem_is_dead(const struct nft_set_ext *ext)
 }
 
 /**
- *	struct nft_trans - nf_tables object update in transaction
+ * struct nft_trans - nf_tables object update in transaction
  *
- *	@list: used internally
- *	@binding_list: list of objects with possible bindings
- *	@msg_type: message type
- *	@put_net: ctx->net needs to be put
- *	@ctx: transaction context
- *	@data: internal information related to the transaction
+ * @list: used internally
+ * @binding_list: list of objects with possible bindings
+ * @msg_type: message type
+ * @put_net: ctx->net needs to be put
+ * @ctx: transaction context
+ *
+ * This is the information common to all objects in the transaction,
+ * this must always be the first member of derived sub-types.
  */
 struct nft_trans {
 	struct list_head		list;
@@ -1623,26 +1625,29 @@ struct nft_trans {
 	int				msg_type;
 	bool				put_net;
 	struct nft_ctx			ctx;
-	char				data[];
 };
 
 struct nft_trans_rule {
+	struct nft_trans		nft_trans;
 	struct nft_rule			*rule;
 	struct nft_flow_rule		*flow;
 	u32				rule_id;
 	bool				bound;
 };
 
-#define nft_trans_rule(trans)	\
-	(((struct nft_trans_rule *)trans->data)->rule)
-#define nft_trans_flow_rule(trans)	\
-	(((struct nft_trans_rule *)trans->data)->flow)
-#define nft_trans_rule_id(trans)	\
-	(((struct nft_trans_rule *)trans->data)->rule_id)
-#define nft_trans_rule_bound(trans)	\
-	(((struct nft_trans_rule *)trans->data)->bound)
+#define nft_trans_container_rule(trans)			\
+	container_of(trans, struct nft_trans_rule, nft_trans)
+#define nft_trans_rule(trans)				\
+	nft_trans_container_rule(trans)->rule
+#define nft_trans_flow_rule(trans)			\
+	nft_trans_container_rule(trans)->flow
+#define nft_trans_rule_id(trans)			\
+	nft_trans_container_rule(trans)->rule_id
+#define nft_trans_rule_bound(trans)			\
+	nft_trans_container_rule(trans)->bound
 
 struct nft_trans_set {
+	struct nft_trans		nft_trans;
 	struct nft_set			*set;
 	u32				set_id;
 	u32				gc_int;
@@ -1652,22 +1657,25 @@ struct nft_trans_set {
 	u32				size;
 };
 
-#define nft_trans_set(trans)	\
-	(((struct nft_trans_set *)trans->data)->set)
-#define nft_trans_set_id(trans)	\
-	(((struct nft_trans_set *)trans->data)->set_id)
-#define nft_trans_set_bound(trans)	\
-	(((struct nft_trans_set *)trans->data)->bound)
-#define nft_trans_set_update(trans)	\
-	(((struct nft_trans_set *)trans->data)->update)
-#define nft_trans_set_timeout(trans)	\
-	(((struct nft_trans_set *)trans->data)->timeout)
-#define nft_trans_set_gc_int(trans)	\
-	(((struct nft_trans_set *)trans->data)->gc_int)
-#define nft_trans_set_size(trans)	\
-	(((struct nft_trans_set *)trans->data)->size)
+#define nft_trans_container_set(trans)			\
+	container_of(trans, struct nft_trans_set, nft_trans)
+#define nft_trans_set(trans)				\
+	nft_trans_container_set(trans)->set
+#define nft_trans_set_id(trans)				\
+	nft_trans_container_set(trans)->set_id
+#define nft_trans_set_bound(trans)			\
+	nft_trans_container_set(trans)->bound
+#define nft_trans_set_update(trans)			\
+	nft_trans_container_set(trans)->update
+#define nft_trans_set_timeout(trans)			\
+	nft_trans_container_set(trans)->timeout
+#define nft_trans_set_gc_int(trans)			\
+	nft_trans_container_set(trans)->gc_int
+#define nft_trans_set_size(trans)			\
+	nft_trans_container_set(trans)->size
 
 struct nft_trans_chain {
+	struct nft_trans		nft_trans;
 	struct nft_chain		*chain;
 	bool				update;
 	char				*name;
@@ -1679,73 +1687,87 @@ struct nft_trans_chain {
 	struct list_head		hook_list;
 };
 
-#define nft_trans_chain(trans)	\
-	(((struct nft_trans_chain *)trans->data)->chain)
-#define nft_trans_chain_update(trans)	\
-	(((struct nft_trans_chain *)trans->data)->update)
-#define nft_trans_chain_name(trans)	\
-	(((struct nft_trans_chain *)trans->data)->name)
-#define nft_trans_chain_stats(trans)	\
-	(((struct nft_trans_chain *)trans->data)->stats)
-#define nft_trans_chain_policy(trans)	\
-	(((struct nft_trans_chain *)trans->data)->policy)
-#define nft_trans_chain_bound(trans)	\
-	(((struct nft_trans_chain *)trans->data)->bound)
-#define nft_trans_chain_id(trans)	\
-	(((struct nft_trans_chain *)trans->data)->chain_id)
-#define nft_trans_basechain(trans)	\
-	(((struct nft_trans_chain *)trans->data)->basechain)
-#define nft_trans_chain_hooks(trans)	\
-	(((struct nft_trans_chain *)trans->data)->hook_list)
+#define nft_trans_container_chain(trans)		\
+	container_of(trans, struct nft_trans_chain, nft_trans)
+#define nft_trans_chain(trans)				\
+	nft_trans_container_chain(trans)->chain
+#define nft_trans_chain_update(trans)			\
+	nft_trans_container_chain(trans)->update
+#define nft_trans_chain_name(trans)			\
+	nft_trans_container_chain(trans)->name
+#define nft_trans_chain_stats(trans)			\
+	nft_trans_container_chain(trans)->stats
+#define nft_trans_chain_policy(trans)			\
+	nft_trans_container_chain(trans)->policy
+#define nft_trans_chain_bound(trans)			\
+	nft_trans_container_chain(trans)->bound
+#define nft_trans_chain_id(trans)			\
+	nft_trans_container_chain(trans)->chain_id
+#define nft_trans_basechain(trans)			\
+	nft_trans_container_chain(trans)->basechain
+#define nft_trans_chain_hooks(trans)			\
+	nft_trans_container_chain(trans)->hook_list
 
 struct nft_trans_table {
+	struct nft_trans		nft_trans;
 	bool				update;
 };
 
-#define nft_trans_table_update(trans)	\
-	(((struct nft_trans_table *)trans->data)->update)
+#define nft_trans_container_table(trans)		\
+	container_of(trans, struct nft_trans_table, nft_trans)
+#define nft_trans_table_update(trans)			\
+	nft_trans_container_table(trans)->update
 
 struct nft_trans_elem {
+	struct nft_trans		nft_trans;
 	struct nft_set			*set;
 	struct nft_elem_priv		*elem_priv;
 	bool				bound;
 };
 
-#define nft_trans_elem_set(trans)	\
-	(((struct nft_trans_elem *)trans->data)->set)
-#define nft_trans_elem_priv(trans)	\
-	(((struct nft_trans_elem *)trans->data)->elem_priv)
-#define nft_trans_elem_set_bound(trans)	\
-	(((struct nft_trans_elem *)trans->data)->bound)
+#define nft_trans_container_elem(t)			\
+	container_of(t, struct nft_trans_elem, nft_trans)
+#define nft_trans_elem_set(trans)			\
+	nft_trans_container_elem(trans)->set
+#define nft_trans_elem_priv(trans)			\
+	nft_trans_container_elem(trans)->elem_priv
+#define nft_trans_elem_set_bound(trans)			\
+	nft_trans_container_elem(trans)->bound
 
 struct nft_trans_obj {
+	struct nft_trans		nft_trans;
 	struct nft_object		*obj;
 	struct nft_object		*newobj;
 	bool				update;
 };
 
-#define nft_trans_obj(trans)	\
-	(((struct nft_trans_obj *)trans->data)->obj)
-#define nft_trans_obj_newobj(trans) \
-	(((struct nft_trans_obj *)trans->data)->newobj)
-#define nft_trans_obj_update(trans)	\
-	(((struct nft_trans_obj *)trans->data)->update)
+#define nft_trans_container_obj(t)			\
+	container_of(t, struct nft_trans_obj, nft_trans)
+#define nft_trans_obj(trans)				\
+	nft_trans_container_obj(trans)->obj
+#define nft_trans_obj_newobj(trans)			\
+	nft_trans_container_obj(trans)->newobj
+#define nft_trans_obj_update(trans)			\
+	nft_trans_container_obj(trans)->update
 
 struct nft_trans_flowtable {
+	struct nft_trans		nft_trans;
 	struct nft_flowtable		*flowtable;
 	bool				update;
 	struct list_head		hook_list;
 	u32				flags;
 };
 
-#define nft_trans_flowtable(trans)	\
-	(((struct nft_trans_flowtable *)trans->data)->flowtable)
-#define nft_trans_flowtable_update(trans)	\
-	(((struct nft_trans_flowtable *)trans->data)->update)
-#define nft_trans_flowtable_hooks(trans)	\
-	(((struct nft_trans_flowtable *)trans->data)->hook_list)
-#define nft_trans_flowtable_flags(trans)	\
-	(((struct nft_trans_flowtable *)trans->data)->flags)
+#define nft_trans_container_flowtable(t)		\
+	container_of(t, struct nft_trans_flowtable, nft_trans)
+#define nft_trans_flowtable(trans)			\
+	nft_trans_container_flowtable(trans)->flowtable
+#define nft_trans_flowtable_update(trans)		\
+	nft_trans_container_flowtable(trans)->update
+#define nft_trans_flowtable_hooks(trans)		\
+	nft_trans_container_flowtable(trans)->hook_list
+#define nft_trans_flowtable_flags(trans)		\
+	nft_trans_container_flowtable(trans)->flags
 
 #define NFT_TRANS_GC_BATCHCOUNT	256
 
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index be3b4c90d2ed..19edd1bcecef 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -153,7 +153,7 @@ static struct nft_trans *nft_trans_alloc_gfp(const struct nft_ctx *ctx,
 {
 	struct nft_trans *trans;
 
-	trans = kzalloc(sizeof(struct nft_trans) + size, gfp);
+	trans = kzalloc(size, gfp);
 	if (trans == NULL)
 		return NULL;
 
@@ -10348,7 +10348,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 					     trans->msg_type, GFP_KERNEL);
 			break;
 		case NFT_MSG_NEWSETELEM:
-			te = (struct nft_trans_elem *)trans->data;
+			te = nft_trans_container_elem(trans);
 
 			nft_setelem_activate(net, te->set, te->elem_priv);
 			nf_tables_setelem_notify(&trans->ctx, te->set,
@@ -10363,7 +10363,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 			break;
 		case NFT_MSG_DELSETELEM:
 		case NFT_MSG_DESTROYSETELEM:
-			te = (struct nft_trans_elem *)trans->data;
+			te = nft_trans_container_elem(trans);
 
 			nf_tables_setelem_notify(&trans->ctx, te->set,
 						 te->elem_priv,
@@ -10643,7 +10643,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 				nft_trans_destroy(trans);
 				break;
 			}
-			te = (struct nft_trans_elem *)trans->data;
+			te = nft_trans_container_elem(trans);
 			nft_setelem_remove(net, te->set, te->elem_priv);
 			if (!nft_setelem_is_catchall(te->set, te->elem_priv))
 				atomic_dec(&te->set->nelems);
@@ -10656,7 +10656,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 			break;
 		case NFT_MSG_DELSETELEM:
 		case NFT_MSG_DESTROYSETELEM:
-			te = (struct nft_trans_elem *)trans->data;
+			te = nft_trans_container_elem(trans);
 
 			if (!nft_setelem_active_next(net, te->set, te->elem_priv)) {
 				nft_setelem_data_activate(net, te->set, te->elem_priv);
@@ -11588,6 +11588,14 @@ static int __init nf_tables_module_init(void)
 {
 	int err;
 
+	BUILD_BUG_ON(offsetof(struct nft_trans_table, nft_trans) != 0);
+	BUILD_BUG_ON(offsetof(struct nft_trans_chain, nft_trans) != 0);
+	BUILD_BUG_ON(offsetof(struct nft_trans_rule, nft_trans) != 0);
+	BUILD_BUG_ON(offsetof(struct nft_trans_set, nft_trans) != 0);
+	BUILD_BUG_ON(offsetof(struct nft_trans_elem, nft_trans) != 0);
+	BUILD_BUG_ON(offsetof(struct nft_trans_obj, nft_trans) != 0);
+	BUILD_BUG_ON(offsetof(struct nft_trans_flowtable, nft_trans) != 0);
+
 	err = register_pernet_subsys(&nf_tables_net_ops);
 	if (err < 0)
 		return err;
-- 
cgit v1.2.3-58-ga151


From 17d8f3ad36a5fa5c93afab90ed03ba7ec748dd03 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Mon, 24 Jun 2024 20:53:16 +0200
Subject: netfilter: nf_tables: move bind list_head into relevant subtypes

Only nft_trans_chain and nft_trans_set subtypes use the
trans->binding_list member.

Add a new common binding subtype and move the member there.

This reduces size of all other subtypes by 16 bytes on 64bit platforms.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h | 26 +++++++++-----
 net/netfilter/nf_tables_api.c     | 71 +++++++++++++++++++++++++++++++--------
 2 files changed, 75 insertions(+), 22 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index b25df037fceb..f72448095833 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1611,7 +1611,6 @@ static inline int nft_set_elem_is_dead(const struct nft_set_ext *ext)
  * struct nft_trans - nf_tables object update in transaction
  *
  * @list: used internally
- * @binding_list: list of objects with possible bindings
  * @msg_type: message type
  * @put_net: ctx->net needs to be put
  * @ctx: transaction context
@@ -1621,12 +1620,23 @@ static inline int nft_set_elem_is_dead(const struct nft_set_ext *ext)
  */
 struct nft_trans {
 	struct list_head		list;
-	struct list_head		binding_list;
 	int				msg_type;
 	bool				put_net;
 	struct nft_ctx			ctx;
 };
 
+/**
+ * struct nft_trans_binding - nf_tables object with binding support in transaction
+ * @nft_trans:    base structure, MUST be first member
+ * @binding_list: list of objects with possible bindings
+ *
+ * This is the base type used by objects that can be bound to a chain.
+ */
+struct nft_trans_binding {
+	struct nft_trans nft_trans;
+	struct list_head binding_list;
+};
+
 struct nft_trans_rule {
 	struct nft_trans		nft_trans;
 	struct nft_rule			*rule;
@@ -1647,7 +1657,7 @@ struct nft_trans_rule {
 	nft_trans_container_rule(trans)->bound
 
 struct nft_trans_set {
-	struct nft_trans		nft_trans;
+	struct nft_trans_binding	nft_trans_binding;
 	struct nft_set			*set;
 	u32				set_id;
 	u32				gc_int;
@@ -1657,8 +1667,8 @@ struct nft_trans_set {
 	u32				size;
 };
 
-#define nft_trans_container_set(trans)			\
-	container_of(trans, struct nft_trans_set, nft_trans)
+#define nft_trans_container_set(t)	\
+	container_of(t, struct nft_trans_set, nft_trans_binding.nft_trans)
 #define nft_trans_set(trans)				\
 	nft_trans_container_set(trans)->set
 #define nft_trans_set_id(trans)				\
@@ -1675,7 +1685,7 @@ struct nft_trans_set {
 	nft_trans_container_set(trans)->size
 
 struct nft_trans_chain {
-	struct nft_trans		nft_trans;
+	struct nft_trans_binding	nft_trans_binding;
 	struct nft_chain		*chain;
 	bool				update;
 	char				*name;
@@ -1687,8 +1697,8 @@ struct nft_trans_chain {
 	struct list_head		hook_list;
 };
 
-#define nft_trans_container_chain(trans)		\
-	container_of(trans, struct nft_trans_chain, nft_trans)
+#define nft_trans_container_chain(t)	\
+	container_of(t, struct nft_trans_chain, nft_trans_binding.nft_trans)
 #define nft_trans_chain(trans)				\
 	nft_trans_container_chain(trans)->chain
 #define nft_trans_chain_update(trans)			\
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 19edd1bcecef..c950938ef612 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -158,7 +158,6 @@ static struct nft_trans *nft_trans_alloc_gfp(const struct nft_ctx *ctx,
 		return NULL;
 
 	INIT_LIST_HEAD(&trans->list);
-	INIT_LIST_HEAD(&trans->binding_list);
 	trans->msg_type = msg_type;
 	trans->ctx	= *ctx;
 
@@ -171,10 +170,26 @@ static struct nft_trans *nft_trans_alloc(const struct nft_ctx *ctx,
 	return nft_trans_alloc_gfp(ctx, msg_type, size, GFP_KERNEL);
 }
 
+static struct nft_trans_binding *nft_trans_get_binding(struct nft_trans *trans)
+{
+	switch (trans->msg_type) {
+	case NFT_MSG_NEWCHAIN:
+	case NFT_MSG_NEWSET:
+		return container_of(trans, struct nft_trans_binding, nft_trans);
+	}
+
+	return NULL;
+}
+
 static void nft_trans_list_del(struct nft_trans *trans)
 {
+	struct nft_trans_binding *trans_binding;
+
 	list_del(&trans->list);
-	list_del(&trans->binding_list);
+
+	trans_binding = nft_trans_get_binding(trans);
+	if (trans_binding)
+		list_del(&trans_binding->binding_list);
 }
 
 static void nft_trans_destroy(struct nft_trans *trans)
@@ -372,21 +387,26 @@ static void nf_tables_unregister_hook(struct net *net,
 static void nft_trans_commit_list_add_tail(struct net *net, struct nft_trans *trans)
 {
 	struct nftables_pernet *nft_net = nft_pernet(net);
+	struct nft_trans_binding *binding;
+
+	list_add_tail(&trans->list, &nft_net->commit_list);
+
+	binding = nft_trans_get_binding(trans);
+	if (!binding)
+		return;
 
 	switch (trans->msg_type) {
 	case NFT_MSG_NEWSET:
 		if (!nft_trans_set_update(trans) &&
 		    nft_set_is_anonymous(nft_trans_set(trans)))
-			list_add_tail(&trans->binding_list, &nft_net->binding_list);
+			list_add_tail(&binding->binding_list, &nft_net->binding_list);
 		break;
 	case NFT_MSG_NEWCHAIN:
 		if (!nft_trans_chain_update(trans) &&
 		    nft_chain_binding(nft_trans_chain(trans)))
-			list_add_tail(&trans->binding_list, &nft_net->binding_list);
+			list_add_tail(&binding->binding_list, &nft_net->binding_list);
 		break;
 	}
-
-	list_add_tail(&trans->list, &nft_net->commit_list);
 }
 
 static int nft_trans_table_add(struct nft_ctx *ctx, int msg_type)
@@ -416,11 +436,27 @@ static int nft_deltable(struct nft_ctx *ctx)
 	return err;
 }
 
-static struct nft_trans *nft_trans_chain_add(struct nft_ctx *ctx, int msg_type)
+static struct nft_trans *
+nft_trans_alloc_chain(const struct nft_ctx *ctx, int msg_type)
 {
+	struct nft_trans_chain *trans_chain;
 	struct nft_trans *trans;
 
 	trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_chain));
+	if (!trans)
+		return NULL;
+
+	trans_chain = nft_trans_container_chain(trans);
+	INIT_LIST_HEAD(&trans_chain->nft_trans_binding.binding_list);
+
+	return trans;
+}
+
+static struct nft_trans *nft_trans_chain_add(struct nft_ctx *ctx, int msg_type)
+{
+	struct nft_trans *trans;
+
+	trans = nft_trans_alloc_chain(ctx, msg_type);
 	if (trans == NULL)
 		return ERR_PTR(-ENOMEM);
 
@@ -560,12 +596,16 @@ static int __nft_trans_set_add(const struct nft_ctx *ctx, int msg_type,
 			       struct nft_set *set,
 			       const struct nft_set_desc *desc)
 {
+	struct nft_trans_set *trans_set;
 	struct nft_trans *trans;
 
 	trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_set));
 	if (trans == NULL)
 		return -ENOMEM;
 
+	trans_set = nft_trans_container_set(trans);
+	INIT_LIST_HEAD(&trans_set->nft_trans_binding.binding_list);
+
 	if (msg_type == NFT_MSG_NEWSET && ctx->nla[NFTA_SET_ID] && !desc) {
 		nft_trans_set_id(trans) =
 			ntohl(nla_get_be32(ctx->nla[NFTA_SET_ID]));
@@ -2698,8 +2738,7 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy,
 	}
 
 	err = -ENOMEM;
-	trans = nft_trans_alloc(ctx, NFT_MSG_NEWCHAIN,
-				sizeof(struct nft_trans_chain));
+	trans = nft_trans_alloc_chain(ctx, NFT_MSG_NEWCHAIN);
 	if (trans == NULL)
 		goto err_trans;
 
@@ -2915,8 +2954,7 @@ static int nft_delchain_hook(struct nft_ctx *ctx,
 		list_move(&hook->list, &chain_del_list);
 	}
 
-	trans = nft_trans_alloc(ctx, NFT_MSG_DELCHAIN,
-				sizeof(struct nft_trans_chain));
+	trans = nft_trans_alloc_chain(ctx, NFT_MSG_DELCHAIN);
 	if (!trans) {
 		err = -ENOMEM;
 		goto err_chain_del_hook;
@@ -10147,6 +10185,7 @@ static void nft_gc_seq_end(struct nftables_pernet *nft_net, unsigned int gc_seq)
 static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 {
 	struct nftables_pernet *nft_net = nft_pernet(net);
+	struct nft_trans_binding *trans_binding;
 	struct nft_trans *trans, *next;
 	unsigned int base_seq, gc_seq;
 	LIST_HEAD(set_update_list);
@@ -10161,7 +10200,8 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 		return 0;
 	}
 
-	list_for_each_entry(trans, &nft_net->binding_list, binding_list) {
+	list_for_each_entry(trans_binding, &nft_net->binding_list, binding_list) {
+		trans = &trans_binding->nft_trans;
 		switch (trans->msg_type) {
 		case NFT_MSG_NEWSET:
 			if (!nft_trans_set_update(trans) &&
@@ -10179,6 +10219,9 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 				return -EINVAL;
 			}
 			break;
+		default:
+			WARN_ONCE(1, "Unhandled bind type %d", trans->msg_type);
+			break;
 		}
 	}
 
@@ -11589,9 +11632,9 @@ static int __init nf_tables_module_init(void)
 	int err;
 
 	BUILD_BUG_ON(offsetof(struct nft_trans_table, nft_trans) != 0);
-	BUILD_BUG_ON(offsetof(struct nft_trans_chain, nft_trans) != 0);
+	BUILD_BUG_ON(offsetof(struct nft_trans_chain, nft_trans_binding.nft_trans) != 0);
 	BUILD_BUG_ON(offsetof(struct nft_trans_rule, nft_trans) != 0);
-	BUILD_BUG_ON(offsetof(struct nft_trans_set, nft_trans) != 0);
+	BUILD_BUG_ON(offsetof(struct nft_trans_set, nft_trans_binding.nft_trans) != 0);
 	BUILD_BUG_ON(offsetof(struct nft_trans_elem, nft_trans) != 0);
 	BUILD_BUG_ON(offsetof(struct nft_trans_obj, nft_trans) != 0);
 	BUILD_BUG_ON(offsetof(struct nft_trans_flowtable, nft_trans) != 0);
-- 
cgit v1.2.3-58-ga151


From b3f4c216f7af37fa60e50d2ebb3ec9dd0f93886c Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Mon, 13 May 2024 15:00:43 +0200
Subject: netfilter: nf_tables: compact chain+ft transaction objects

Cover holes to reduce both structures by 8 byte.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index f72448095833..1f0607b671ac 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1687,10 +1687,10 @@ struct nft_trans_set {
 struct nft_trans_chain {
 	struct nft_trans_binding	nft_trans_binding;
 	struct nft_chain		*chain;
-	bool				update;
 	char				*name;
 	struct nft_stats __percpu	*stats;
 	u8				policy;
+	bool				update;
 	bool				bound;
 	u32				chain_id;
 	struct nft_base_chain		*basechain;
@@ -1763,9 +1763,9 @@ struct nft_trans_obj {
 struct nft_trans_flowtable {
 	struct nft_trans		nft_trans;
 	struct nft_flowtable		*flowtable;
-	bool				update;
 	struct list_head		hook_list;
 	u32				flags;
+	bool				update;
 };
 
 #define nft_trans_container_flowtable(t)		\
-- 
cgit v1.2.3-58-ga151


From 06fcaca2ed1f7a1d02b18364fceac4525332f178 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Mon, 13 May 2024 15:00:44 +0200
Subject: netfilter: nf_tables: reduce trans->ctx.table references

nft_ctx is huge, it should not be stored in nft_trans at all,
most information is not needed.

Preparation patch to remove trans->ctx, no change in behaviour intended.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c | 79 ++++++++++++++++++++++---------------------
 1 file changed, 41 insertions(+), 38 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index c950938ef612..60c435774db8 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -9472,14 +9472,15 @@ static void nft_chain_commit_drop_policy(struct nft_trans *trans)
 
 static void nft_chain_commit_update(struct nft_trans *trans)
 {
+	struct nft_table *table = trans->ctx.table;
 	struct nft_base_chain *basechain;
 
 	if (nft_trans_chain_name(trans)) {
-		rhltable_remove(&trans->ctx.table->chains_ht,
+		rhltable_remove(&table->chains_ht,
 				&trans->ctx.chain->rhlhead,
 				nft_chain_ht_params);
 		swap(trans->ctx.chain->name, nft_trans_chain_name(trans));
-		rhltable_insert_key(&trans->ctx.table->chains_ht,
+		rhltable_insert_key(&table->chains_ht,
 				    trans->ctx.chain->name,
 				    &trans->ctx.chain->rhlhead,
 				    nft_chain_ht_params);
@@ -10237,9 +10238,10 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 
 	/* 1.  Allocate space for next generation rules_gen_X[] */
 	list_for_each_entry_safe(trans, next, &nft_net->commit_list, list) {
+		struct nft_table *table = trans->ctx.table;
 		int ret;
 
-		ret = nf_tables_commit_audit_alloc(&adl, trans->ctx.table);
+		ret = nf_tables_commit_audit_alloc(&adl, table);
 		if (ret) {
 			nf_tables_commit_chain_prepare_cancel(net);
 			nf_tables_commit_audit_free(&adl);
@@ -10280,28 +10282,29 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 	net->nft.gencursor = nft_gencursor_next(net);
 
 	list_for_each_entry_safe(trans, next, &nft_net->commit_list, list) {
-		nf_tables_commit_audit_collect(&adl, trans->ctx.table,
-					       trans->msg_type);
+		struct nft_table *table = trans->ctx.table;
+
+		nf_tables_commit_audit_collect(&adl, table, trans->msg_type);
 		switch (trans->msg_type) {
 		case NFT_MSG_NEWTABLE:
 			if (nft_trans_table_update(trans)) {
-				if (!(trans->ctx.table->flags & __NFT_TABLE_F_UPDATE)) {
+				if (!(table->flags & __NFT_TABLE_F_UPDATE)) {
 					nft_trans_destroy(trans);
 					break;
 				}
-				if (trans->ctx.table->flags & NFT_TABLE_F_DORMANT)
-					nf_tables_table_disable(net, trans->ctx.table);
+				if (table->flags & NFT_TABLE_F_DORMANT)
+					nf_tables_table_disable(net, table);
 
-				trans->ctx.table->flags &= ~__NFT_TABLE_F_UPDATE;
+				table->flags &= ~__NFT_TABLE_F_UPDATE;
 			} else {
-				nft_clear(net, trans->ctx.table);
+				nft_clear(net, table);
 			}
 			nf_tables_table_notify(&trans->ctx, NFT_MSG_NEWTABLE);
 			nft_trans_destroy(trans);
 			break;
 		case NFT_MSG_DELTABLE:
 		case NFT_MSG_DESTROYTABLE:
-			list_del_rcu(&trans->ctx.table->list);
+			list_del_rcu(&table->list);
 			nf_tables_table_notify(&trans->ctx, trans->msg_type);
 			break;
 		case NFT_MSG_NEWCHAIN:
@@ -10324,7 +10327,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 			if (nft_trans_chain_update(trans)) {
 				nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN,
 						       &nft_trans_chain_hooks(trans));
-				if (!(trans->ctx.table->flags & NFT_TABLE_F_DORMANT)) {
+				if (!(table->flags & NFT_TABLE_F_DORMANT)) {
 					nft_netdev_unregister_hooks(net,
 								    &nft_trans_chain_hooks(trans),
 								    true);
@@ -10333,8 +10336,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 				nft_chain_del(trans->ctx.chain);
 				nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN,
 						       NULL);
-				nf_tables_unregister_hook(trans->ctx.net,
-							  trans->ctx.table,
+				nf_tables_unregister_hook(trans->ctx.net, table,
 							  trans->ctx.chain);
 			}
 			break;
@@ -10377,7 +10379,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 				 */
 				if (nft_set_is_anonymous(nft_trans_set(trans)) &&
 				    !list_empty(&nft_trans_set(trans)->bindings))
-					nft_use_dec(&trans->ctx.table->use);
+					nft_use_dec(&table->use);
 			}
 			nf_tables_set_notify(&trans->ctx, nft_trans_set(trans),
 					     NFT_MSG_NEWSET, GFP_KERNEL);
@@ -10575,37 +10577,39 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 
 	list_for_each_entry_safe_reverse(trans, next, &nft_net->commit_list,
 					 list) {
+		struct nft_table *table = trans->ctx.table;
+
 		switch (trans->msg_type) {
 		case NFT_MSG_NEWTABLE:
 			if (nft_trans_table_update(trans)) {
-				if (!(trans->ctx.table->flags & __NFT_TABLE_F_UPDATE)) {
+				if (!(table->flags & __NFT_TABLE_F_UPDATE)) {
 					nft_trans_destroy(trans);
 					break;
 				}
-				if (trans->ctx.table->flags & __NFT_TABLE_F_WAS_DORMANT) {
-					nf_tables_table_disable(net, trans->ctx.table);
-					trans->ctx.table->flags |= NFT_TABLE_F_DORMANT;
-				} else if (trans->ctx.table->flags & __NFT_TABLE_F_WAS_AWAKEN) {
-					trans->ctx.table->flags &= ~NFT_TABLE_F_DORMANT;
+				if (table->flags & __NFT_TABLE_F_WAS_DORMANT) {
+					nf_tables_table_disable(net, table);
+					table->flags |= NFT_TABLE_F_DORMANT;
+				} else if (table->flags & __NFT_TABLE_F_WAS_AWAKEN) {
+					table->flags &= ~NFT_TABLE_F_DORMANT;
 				}
-				if (trans->ctx.table->flags & __NFT_TABLE_F_WAS_ORPHAN) {
-					trans->ctx.table->flags &= ~NFT_TABLE_F_OWNER;
-					trans->ctx.table->nlpid = 0;
+				if (table->flags & __NFT_TABLE_F_WAS_ORPHAN) {
+					table->flags &= ~NFT_TABLE_F_OWNER;
+					table->nlpid = 0;
 				}
-				trans->ctx.table->flags &= ~__NFT_TABLE_F_UPDATE;
+				table->flags &= ~__NFT_TABLE_F_UPDATE;
 				nft_trans_destroy(trans);
 			} else {
-				list_del_rcu(&trans->ctx.table->list);
+				list_del_rcu(&table->list);
 			}
 			break;
 		case NFT_MSG_DELTABLE:
 		case NFT_MSG_DESTROYTABLE:
-			nft_clear(trans->ctx.net, trans->ctx.table);
+			nft_clear(trans->ctx.net, table);
 			nft_trans_destroy(trans);
 			break;
 		case NFT_MSG_NEWCHAIN:
 			if (nft_trans_chain_update(trans)) {
-				if (!(trans->ctx.table->flags & NFT_TABLE_F_DORMANT)) {
+				if (!(table->flags & NFT_TABLE_F_DORMANT)) {
 					nft_netdev_unregister_hooks(net,
 								    &nft_trans_chain_hooks(trans),
 								    true);
@@ -10618,10 +10622,9 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 					nft_trans_destroy(trans);
 					break;
 				}
-				nft_use_dec_restore(&trans->ctx.table->use);
+				nft_use_dec_restore(&table->use);
 				nft_chain_del(trans->ctx.chain);
-				nf_tables_unregister_hook(trans->ctx.net,
-							  trans->ctx.table,
+				nf_tables_unregister_hook(trans->ctx.net, table,
 							  trans->ctx.chain);
 			}
 			break;
@@ -10631,7 +10634,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 				list_splice(&nft_trans_chain_hooks(trans),
 					    &nft_trans_basechain(trans)->hook_list);
 			} else {
-				nft_use_inc_restore(&trans->ctx.table->use);
+				nft_use_inc_restore(&table->use);
 				nft_clear(trans->ctx.net, trans->ctx.chain);
 			}
 			nft_trans_destroy(trans);
@@ -10664,7 +10667,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 				nft_trans_destroy(trans);
 				break;
 			}
-			nft_use_dec_restore(&trans->ctx.table->use);
+			nft_use_dec_restore(&table->use);
 			if (nft_trans_set_bound(trans)) {
 				nft_trans_destroy(trans);
 				break;
@@ -10674,7 +10677,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 			break;
 		case NFT_MSG_DELSET:
 		case NFT_MSG_DESTROYSET:
-			nft_use_inc_restore(&trans->ctx.table->use);
+			nft_use_inc_restore(&table->use);
 			nft_clear(trans->ctx.net, nft_trans_set(trans));
 			if (nft_trans_set(trans)->flags & (NFT_SET_MAP | NFT_SET_OBJECT))
 				nft_map_activate(&trans->ctx, nft_trans_set(trans));
@@ -10720,13 +10723,13 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 				nft_obj_destroy(&trans->ctx, nft_trans_obj_newobj(trans));
 				nft_trans_destroy(trans);
 			} else {
-				nft_use_dec_restore(&trans->ctx.table->use);
+				nft_use_dec_restore(&table->use);
 				nft_obj_del(nft_trans_obj(trans));
 			}
 			break;
 		case NFT_MSG_DELOBJ:
 		case NFT_MSG_DESTROYOBJ:
-			nft_use_inc_restore(&trans->ctx.table->use);
+			nft_use_inc_restore(&table->use);
 			nft_clear(trans->ctx.net, nft_trans_obj(trans));
 			nft_trans_destroy(trans);
 			break;
@@ -10735,7 +10738,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 				nft_unregister_flowtable_net_hooks(net,
 						&nft_trans_flowtable_hooks(trans));
 			} else {
-				nft_use_dec_restore(&trans->ctx.table->use);
+				nft_use_dec_restore(&table->use);
 				list_del_rcu(&nft_trans_flowtable(trans)->list);
 				nft_unregister_flowtable_net_hooks(net,
 						&nft_trans_flowtable(trans)->hook_list);
@@ -10747,7 +10750,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 				list_splice(&nft_trans_flowtable_hooks(trans),
 					    &nft_trans_flowtable(trans)->hook_list);
 			} else {
-				nft_use_inc_restore(&trans->ctx.table->use);
+				nft_use_inc_restore(&table->use);
 				nft_clear(trans->ctx.net, nft_trans_flowtable(trans));
 			}
 			nft_trans_destroy(trans);
-- 
cgit v1.2.3-58-ga151


From 8965d42bcf54d42cbc72fe34a9d0ec3f8527debd Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Mon, 13 May 2024 15:00:45 +0200
Subject: netfilter: nf_tables: pass nft_chain to destroy function, not nft_ctx

It would be better to not store nft_ctx inside nft_trans object,
the netlink ctx strucutre is huge and most of its information is
never needed in places that use trans->ctx.

Avoid/reduce its usage if possible, no runtime behaviour change
intended.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h |  2 +-
 net/netfilter/nf_tables_api.c     | 17 ++++++++---------
 net/netfilter/nft_immediate.c     |  2 +-
 3 files changed, 10 insertions(+), 11 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 1f0607b671ac..328fdc140551 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1171,7 +1171,7 @@ static inline bool nft_chain_is_bound(struct nft_chain *chain)
 
 int nft_chain_add(struct nft_table *table, struct nft_chain *chain);
 void nft_chain_del(struct nft_chain *chain);
-void nf_tables_chain_destroy(struct nft_ctx *ctx);
+void nf_tables_chain_destroy(struct nft_chain *chain);
 
 struct nft_stats {
 	u64			bytes;
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 60c435774db8..bdc2d7f781ca 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -2118,9 +2118,9 @@ static void nf_tables_chain_free_chain_rules(struct nft_chain *chain)
 	kvfree(chain->blob_next);
 }
 
-void nf_tables_chain_destroy(struct nft_ctx *ctx)
+void nf_tables_chain_destroy(struct nft_chain *chain)
 {
-	struct nft_chain *chain = ctx->chain;
+	const struct nft_table *table = chain->table;
 	struct nft_hook *hook, *next;
 
 	if (WARN_ON(chain->use > 0))
@@ -2132,7 +2132,7 @@ void nf_tables_chain_destroy(struct nft_ctx *ctx)
 	if (nft_is_base_chain(chain)) {
 		struct nft_base_chain *basechain = nft_base_chain(chain);
 
-		if (nft_base_chain_netdev(ctx->family, basechain->ops.hooknum)) {
+		if (nft_base_chain_netdev(table->family, basechain->ops.hooknum)) {
 			list_for_each_entry_safe(hook, next,
 						 &basechain->hook_list, list) {
 				list_del_rcu(&hook->list);
@@ -2621,7 +2621,7 @@ err_chain_add:
 err_trans:
 	nft_use_dec_restore(&table->use);
 err_destroy_chain:
-	nf_tables_chain_destroy(ctx);
+	nf_tables_chain_destroy(chain);
 
 	return err;
 }
@@ -9532,7 +9532,7 @@ static void nft_commit_release(struct nft_trans *trans)
 		if (nft_trans_chain_update(trans))
 			nft_hooks_destroy(&nft_trans_chain_hooks(trans));
 		else
-			nf_tables_chain_destroy(&trans->ctx);
+			nf_tables_chain_destroy(nft_trans_chain(trans));
 		break;
 	case NFT_MSG_DELRULE:
 	case NFT_MSG_DESTROYRULE:
@@ -10524,7 +10524,7 @@ static void nf_tables_abort_release(struct nft_trans *trans)
 		if (nft_trans_chain_update(trans))
 			nft_hooks_destroy(&nft_trans_chain_hooks(trans));
 		else
-			nf_tables_chain_destroy(&trans->ctx);
+			nf_tables_chain_destroy(nft_trans_chain(trans));
 		break;
 	case NFT_MSG_NEWRULE:
 		nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans));
@@ -11411,7 +11411,7 @@ int __nft_release_basechain(struct nft_ctx *ctx)
 	}
 	nft_chain_del(ctx->chain);
 	nft_use_dec(&ctx->table->use);
-	nf_tables_chain_destroy(ctx);
+	nf_tables_chain_destroy(ctx->chain);
 
 	return 0;
 }
@@ -11486,10 +11486,9 @@ static void __nft_release_table(struct net *net, struct nft_table *table)
 		nft_obj_destroy(&ctx, obj);
 	}
 	list_for_each_entry_safe(chain, nc, &table->chains, list) {
-		ctx.chain = chain;
 		nft_chain_del(chain);
 		nft_use_dec(&table->use);
-		nf_tables_chain_destroy(&ctx);
+		nf_tables_chain_destroy(chain);
 	}
 	nf_tables_table_destroy(&ctx);
 }
diff --git a/net/netfilter/nft_immediate.c b/net/netfilter/nft_immediate.c
index 6475c7abc1fe..ac2422c215e5 100644
--- a/net/netfilter/nft_immediate.c
+++ b/net/netfilter/nft_immediate.c
@@ -221,7 +221,7 @@ static void nft_immediate_destroy(const struct nft_ctx *ctx,
 			list_del(&rule->list);
 			nf_tables_rule_destroy(&chain_ctx, rule);
 		}
-		nf_tables_chain_destroy(&chain_ctx);
+		nf_tables_chain_destroy(chain);
 		break;
 	default:
 		break;
-- 
cgit v1.2.3-58-ga151


From 0c2e0ee861de071c2e02c44dda5c44329ea8394d Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Mon, 13 May 2024 15:00:46 +0200
Subject: netfilter: nf_tables: pass more specific nft_trans_chain where
 possible

These functions pass a pointer to the base object type, use the
more specific one.  No functional change intended.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c | 51 ++++++++++++++++++++++---------------------
 1 file changed, 26 insertions(+), 25 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index bdc2d7f781ca..62a4da955574 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -448,6 +448,7 @@ nft_trans_alloc_chain(const struct nft_ctx *ctx, int msg_type)
 
 	trans_chain = nft_trans_container_chain(trans);
 	INIT_LIST_HEAD(&trans_chain->nft_trans_binding.binding_list);
+	trans_chain->chain = ctx->chain;
 
 	return trans;
 }
@@ -468,7 +469,6 @@ static struct nft_trans *nft_trans_chain_add(struct nft_ctx *ctx, int msg_type)
 				ntohl(nla_get_be32(ctx->nla[NFTA_CHAIN_ID]));
 		}
 	}
-	nft_trans_chain(trans) = ctx->chain;
 	nft_trans_commit_list_add_tail(ctx->net, trans);
 
 	return trans;
@@ -2089,18 +2089,19 @@ static struct nft_stats __percpu *nft_stats_alloc(const struct nlattr *attr)
 	return newstats;
 }
 
-static void nft_chain_stats_replace(struct nft_trans *trans)
+static void nft_chain_stats_replace(struct nft_trans_chain *trans)
 {
-	struct nft_base_chain *chain = nft_base_chain(trans->ctx.chain);
+	const struct nft_trans *t = &trans->nft_trans_binding.nft_trans;
+	struct nft_base_chain *chain = nft_base_chain(trans->chain);
 
-	if (!nft_trans_chain_stats(trans))
+	if (!trans->stats)
 		return;
 
-	nft_trans_chain_stats(trans) =
-		rcu_replace_pointer(chain->stats, nft_trans_chain_stats(trans),
-				    lockdep_commit_lock_is_held(trans->ctx.net));
+	trans->stats =
+		rcu_replace_pointer(chain->stats, trans->stats,
+				    lockdep_commit_lock_is_held(t->ctx.net));
 
-	if (!nft_trans_chain_stats(trans))
+	if (!trans->stats)
 		static_branch_inc(&nft_counters_enabled);
 }
 
@@ -9456,47 +9457,47 @@ static int nf_tables_validate(struct net *net)
  *
  * We defer the drop policy until the transaction has been finalized.
  */
-static void nft_chain_commit_drop_policy(struct nft_trans *trans)
+static void nft_chain_commit_drop_policy(struct nft_trans_chain *trans)
 {
 	struct nft_base_chain *basechain;
 
-	if (nft_trans_chain_policy(trans) != NF_DROP)
+	if (trans->policy != NF_DROP)
 		return;
 
-	if (!nft_is_base_chain(trans->ctx.chain))
+	if (!nft_is_base_chain(trans->chain))
 		return;
 
-	basechain = nft_base_chain(trans->ctx.chain);
+	basechain = nft_base_chain(trans->chain);
 	basechain->policy = NF_DROP;
 }
 
-static void nft_chain_commit_update(struct nft_trans *trans)
+static void nft_chain_commit_update(struct nft_trans_chain *trans)
 {
-	struct nft_table *table = trans->ctx.table;
+	struct nft_table *table = trans->nft_trans_binding.nft_trans.ctx.table;
 	struct nft_base_chain *basechain;
 
-	if (nft_trans_chain_name(trans)) {
+	if (trans->name) {
 		rhltable_remove(&table->chains_ht,
-				&trans->ctx.chain->rhlhead,
+				&trans->chain->rhlhead,
 				nft_chain_ht_params);
-		swap(trans->ctx.chain->name, nft_trans_chain_name(trans));
+		swap(trans->chain->name, trans->name);
 		rhltable_insert_key(&table->chains_ht,
-				    trans->ctx.chain->name,
-				    &trans->ctx.chain->rhlhead,
+				    trans->chain->name,
+				    &trans->chain->rhlhead,
 				    nft_chain_ht_params);
 	}
 
-	if (!nft_is_base_chain(trans->ctx.chain))
+	if (!nft_is_base_chain(trans->chain))
 		return;
 
 	nft_chain_stats_replace(trans);
 
-	basechain = nft_base_chain(trans->ctx.chain);
+	basechain = nft_base_chain(trans->chain);
 
-	switch (nft_trans_chain_policy(trans)) {
+	switch (trans->policy) {
 	case NF_DROP:
 	case NF_ACCEPT:
-		basechain->policy = nft_trans_chain_policy(trans);
+		basechain->policy = trans->policy;
 		break;
 	}
 }
@@ -10309,14 +10310,14 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 			break;
 		case NFT_MSG_NEWCHAIN:
 			if (nft_trans_chain_update(trans)) {
-				nft_chain_commit_update(trans);
+				nft_chain_commit_update(nft_trans_container_chain(trans));
 				nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN,
 						       &nft_trans_chain_hooks(trans));
 				list_splice(&nft_trans_chain_hooks(trans),
 					    &nft_trans_basechain(trans)->hook_list);
 				/* trans destroyed after rcu grace period */
 			} else {
-				nft_chain_commit_drop_policy(trans);
+				nft_chain_commit_drop_policy(nft_trans_container_chain(trans));
 				nft_clear(net, trans->ctx.chain);
 				nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN, NULL);
 				nft_trans_destroy(trans);
-- 
cgit v1.2.3-58-ga151


From d4f6f3994e133ff49e89d7e108500f45224a79e1 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Mon, 13 May 2024 15:00:47 +0200
Subject: netfilter: nf_tables: avoid usage of embedded nft_ctx

nft_ctx is stored in nft_trans object, but nft_ctx is large
(48 bytes on 64-bit platforms), it should not be embedded in
the transaction structures.

Reduce its usage so we can remove it eventually.

This replaces trans->ctx.chain with the chain pointer
already available in nft_trans_chain structure.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c     | 10 +++++-----
 net/netfilter/nf_tables_offload.c | 16 ++++++++--------
 2 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 62a4da955574..f4e39816104f 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -9682,10 +9682,10 @@ static void nf_tables_commit_chain_prepare_cancel(struct net *net)
 	struct nft_trans *trans, *next;
 
 	list_for_each_entry_safe(trans, next, &nft_net->commit_list, list) {
-		struct nft_chain *chain = trans->ctx.chain;
-
 		if (trans->msg_type == NFT_MSG_NEWRULE ||
 		    trans->msg_type == NFT_MSG_DELRULE) {
+			struct nft_chain *chain = trans->ctx.chain;
+
 			kvfree(chain->blob_next);
 			chain->blob_next = NULL;
 		}
@@ -10318,7 +10318,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 				/* trans destroyed after rcu grace period */
 			} else {
 				nft_chain_commit_drop_policy(nft_trans_container_chain(trans));
-				nft_clear(net, trans->ctx.chain);
+				nft_clear(net, nft_trans_chain(trans));
 				nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN, NULL);
 				nft_trans_destroy(trans);
 			}
@@ -10334,11 +10334,11 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 								    true);
 				}
 			} else {
-				nft_chain_del(trans->ctx.chain);
+				nft_chain_del(nft_trans_chain(trans));
 				nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN,
 						       NULL);
 				nf_tables_unregister_hook(trans->ctx.net, table,
-							  trans->ctx.chain);
+							  nft_trans_chain(trans));
 			}
 			break;
 		case NFT_MSG_NEWRULE:
diff --git a/net/netfilter/nf_tables_offload.c b/net/netfilter/nf_tables_offload.c
index 12ab78fa5d84..8d892a0d2438 100644
--- a/net/netfilter/nf_tables_offload.c
+++ b/net/netfilter/nf_tables_offload.c
@@ -518,18 +518,18 @@ static void nft_flow_rule_offload_abort(struct net *net,
 
 		switch (trans->msg_type) {
 		case NFT_MSG_NEWCHAIN:
-			if (!(trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD) ||
+			if (!(nft_trans_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD) ||
 			    nft_trans_chain_update(trans))
 				continue;
 
-			err = nft_flow_offload_chain(trans->ctx.chain, NULL,
+			err = nft_flow_offload_chain(nft_trans_chain(trans), NULL,
 						     FLOW_BLOCK_UNBIND);
 			break;
 		case NFT_MSG_DELCHAIN:
-			if (!(trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD))
+			if (!(nft_trans_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD))
 				continue;
 
-			err = nft_flow_offload_chain(trans->ctx.chain, NULL,
+			err = nft_flow_offload_chain(nft_trans_chain(trans), NULL,
 						     FLOW_BLOCK_BIND);
 			break;
 		case NFT_MSG_NEWRULE:
@@ -569,20 +569,20 @@ int nft_flow_rule_offload_commit(struct net *net)
 
 		switch (trans->msg_type) {
 		case NFT_MSG_NEWCHAIN:
-			if (!(trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD) ||
+			if (!(nft_trans_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD) ||
 			    nft_trans_chain_update(trans))
 				continue;
 
 			policy = nft_trans_chain_policy(trans);
-			err = nft_flow_offload_chain(trans->ctx.chain, &policy,
+			err = nft_flow_offload_chain(nft_trans_chain(trans), &policy,
 						     FLOW_BLOCK_BIND);
 			break;
 		case NFT_MSG_DELCHAIN:
-			if (!(trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD))
+			if (!(nft_trans_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD))
 				continue;
 
 			policy = nft_trans_chain_policy(trans);
-			err = nft_flow_offload_chain(trans->ctx.chain, &policy,
+			err = nft_flow_offload_chain(nft_trans_chain(trans), &policy,
 						     FLOW_BLOCK_UNBIND);
 			break;
 		case NFT_MSG_NEWRULE:
-- 
cgit v1.2.3-58-ga151


From 13f20bc9ec4f9f25935bf52337d3d1708787bd55 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Mon, 24 Jun 2024 20:57:03 +0200
Subject: netfilter: nf_tables: store chain pointer in rule transaction

Currently the chain can be derived from trans->ctx.chain, but
the ctx will go away soon.

Thus add the chain pointer to nft_trans_rule structure itself.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h |  3 +++
 net/netfilter/nf_tables_api.c     | 21 +++++++++++----------
 net/netfilter/nf_tables_offload.c | 16 ++++++++--------
 3 files changed, 22 insertions(+), 18 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 328fdc140551..86e6bd63a205 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1640,6 +1640,7 @@ struct nft_trans_binding {
 struct nft_trans_rule {
 	struct nft_trans		nft_trans;
 	struct nft_rule			*rule;
+	struct nft_chain		*chain;
 	struct nft_flow_rule		*flow;
 	u32				rule_id;
 	bool				bound;
@@ -1655,6 +1656,8 @@ struct nft_trans_rule {
 	nft_trans_container_rule(trans)->rule_id
 #define nft_trans_rule_bound(trans)			\
 	nft_trans_container_rule(trans)->bound
+#define nft_trans_rule_chain(trans)	\
+	nft_trans_container_rule(trans)->chain
 
 struct nft_trans_set {
 	struct nft_trans_binding	nft_trans_binding;
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index f4e39816104f..3e5980f0bf71 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -251,7 +251,7 @@ static void __nft_chain_trans_bind(const struct nft_ctx *ctx,
 				nft_trans_chain_bound(trans) = bind;
 			break;
 		case NFT_MSG_NEWRULE:
-			if (trans->ctx.chain == chain)
+			if (nft_trans_rule_chain(trans) == chain)
 				nft_trans_rule_bound(trans) = bind;
 			break;
 		}
@@ -541,6 +541,7 @@ static struct nft_trans *nft_trans_rule_add(struct nft_ctx *ctx, int msg_type,
 			ntohl(nla_get_be32(ctx->nla[NFTA_RULE_ID]));
 	}
 	nft_trans_rule(trans) = rule;
+	nft_trans_rule_chain(trans) = ctx->chain;
 	nft_trans_commit_list_add_tail(ctx->net, trans);
 
 	return trans;
@@ -4227,7 +4228,7 @@ static struct nft_rule *nft_rule_lookup_byid(const struct net *net,
 
 	list_for_each_entry(trans, &nft_net->commit_list, list) {
 		if (trans->msg_type == NFT_MSG_NEWRULE &&
-		    trans->ctx.chain == chain &&
+		    nft_trans_rule_chain(trans) == chain &&
 		    id == nft_trans_rule_id(trans))
 			return nft_trans_rule(trans);
 	}
@@ -9684,7 +9685,7 @@ static void nf_tables_commit_chain_prepare_cancel(struct net *net)
 	list_for_each_entry_safe(trans, next, &nft_net->commit_list, list) {
 		if (trans->msg_type == NFT_MSG_NEWRULE ||
 		    trans->msg_type == NFT_MSG_DELRULE) {
-			struct nft_chain *chain = trans->ctx.chain;
+			struct nft_chain *chain = nft_trans_rule_chain(trans);
 
 			kvfree(chain->blob_next);
 			chain->blob_next = NULL;
@@ -10250,7 +10251,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 		}
 		if (trans->msg_type == NFT_MSG_NEWRULE ||
 		    trans->msg_type == NFT_MSG_DELRULE) {
-			chain = trans->ctx.chain;
+			chain = nft_trans_rule_chain(trans);
 
 			ret = nf_tables_commit_chain_prepare(net, chain);
 			if (ret < 0) {
@@ -10346,7 +10347,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 			nf_tables_rule_notify(&trans->ctx,
 					      nft_trans_rule(trans),
 					      NFT_MSG_NEWRULE);
-			if (trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD)
+			if (nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD)
 				nft_flow_rule_destroy(nft_trans_flow_rule(trans));
 
 			nft_trans_destroy(trans);
@@ -10361,7 +10362,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 						 nft_trans_rule(trans),
 						 NFT_TRANS_COMMIT);
 
-			if (trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD)
+			if (nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD)
 				nft_flow_rule_destroy(nft_trans_flow_rule(trans));
 			break;
 		case NFT_MSG_NEWSET:
@@ -10645,20 +10646,20 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 				nft_trans_destroy(trans);
 				break;
 			}
-			nft_use_dec_restore(&trans->ctx.chain->use);
+			nft_use_dec_restore(&nft_trans_rule_chain(trans)->use);
 			list_del_rcu(&nft_trans_rule(trans)->list);
 			nft_rule_expr_deactivate(&trans->ctx,
 						 nft_trans_rule(trans),
 						 NFT_TRANS_ABORT);
-			if (trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD)
+			if (nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD)
 				nft_flow_rule_destroy(nft_trans_flow_rule(trans));
 			break;
 		case NFT_MSG_DELRULE:
 		case NFT_MSG_DESTROYRULE:
-			nft_use_inc_restore(&trans->ctx.chain->use);
+			nft_use_inc_restore(&nft_trans_rule_chain(trans)->use);
 			nft_clear(trans->ctx.net, nft_trans_rule(trans));
 			nft_rule_expr_activate(&trans->ctx, nft_trans_rule(trans));
-			if (trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD)
+			if (nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD)
 				nft_flow_rule_destroy(nft_trans_flow_rule(trans));
 
 			nft_trans_destroy(trans);
diff --git a/net/netfilter/nf_tables_offload.c b/net/netfilter/nf_tables_offload.c
index 8d892a0d2438..0619feb10abb 100644
--- a/net/netfilter/nf_tables_offload.c
+++ b/net/netfilter/nf_tables_offload.c
@@ -533,18 +533,18 @@ static void nft_flow_rule_offload_abort(struct net *net,
 						     FLOW_BLOCK_BIND);
 			break;
 		case NFT_MSG_NEWRULE:
-			if (!(trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD))
+			if (!(nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD))
 				continue;
 
-			err = nft_flow_offload_rule(trans->ctx.chain,
+			err = nft_flow_offload_rule(nft_trans_rule_chain(trans),
 						    nft_trans_rule(trans),
 						    NULL, FLOW_CLS_DESTROY);
 			break;
 		case NFT_MSG_DELRULE:
-			if (!(trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD))
+			if (!(nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD))
 				continue;
 
-			err = nft_flow_offload_rule(trans->ctx.chain,
+			err = nft_flow_offload_rule(nft_trans_rule_chain(trans),
 						    nft_trans_rule(trans),
 						    nft_trans_flow_rule(trans),
 						    FLOW_CLS_REPLACE);
@@ -586,7 +586,7 @@ int nft_flow_rule_offload_commit(struct net *net)
 						     FLOW_BLOCK_UNBIND);
 			break;
 		case NFT_MSG_NEWRULE:
-			if (!(trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD))
+			if (!(nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD))
 				continue;
 
 			if (trans->ctx.flags & NLM_F_REPLACE ||
@@ -594,16 +594,16 @@ int nft_flow_rule_offload_commit(struct net *net)
 				err = -EOPNOTSUPP;
 				break;
 			}
-			err = nft_flow_offload_rule(trans->ctx.chain,
+			err = nft_flow_offload_rule(nft_trans_rule_chain(trans),
 						    nft_trans_rule(trans),
 						    nft_trans_flow_rule(trans),
 						    FLOW_CLS_REPLACE);
 			break;
 		case NFT_MSG_DELRULE:
-			if (!(trans->ctx.chain->flags & NFT_CHAIN_HW_OFFLOAD))
+			if (!(nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD))
 				continue;
 
-			err = nft_flow_offload_rule(trans->ctx.chain,
+			err = nft_flow_offload_rule(nft_trans_rule_chain(trans),
 						    nft_trans_rule(trans),
 						    NULL, FLOW_CLS_DESTROY);
 			break;
-- 
cgit v1.2.3-58-ga151


From 551b3886401c3b25fd7a3d5ae31fb8ae29423cc4 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Mon, 13 May 2024 15:00:49 +0200
Subject: netfilter: nf_tables: reduce trans->ctx.chain references

These objects are the trans_chain subtype, so use the helper instead
of referencing trans->ctx, which will be removed soon.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 3e5980f0bf71..bd311b37fc61 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -1262,7 +1262,7 @@ static bool nft_table_pending_update(const struct nft_ctx *ctx)
 		    ((trans->msg_type == NFT_MSG_NEWCHAIN &&
 		      nft_trans_chain_update(trans)) ||
 		     (trans->msg_type == NFT_MSG_DELCHAIN &&
-		      nft_is_base_chain(trans->ctx.chain))))
+		      nft_is_base_chain(nft_trans_chain(trans)))))
 			return true;
 	}
 
@@ -2815,13 +2815,11 @@ static struct nft_chain *nft_chain_lookup_byid(const struct net *net,
 	struct nft_trans *trans;
 
 	list_for_each_entry(trans, &nft_net->commit_list, list) {
-		struct nft_chain *chain = trans->ctx.chain;
-
 		if (trans->msg_type == NFT_MSG_NEWCHAIN &&
-		    chain->table == table &&
+		    nft_trans_chain(trans)->table == table &&
 		    id == nft_trans_chain_id(trans) &&
-		    nft_active_genmask(chain, genmask))
-			return chain;
+		    nft_active_genmask(nft_trans_chain(trans), genmask))
+			return nft_trans_chain(trans);
 	}
 	return ERR_PTR(-ENOENT);
 }
@@ -10625,9 +10623,9 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 					break;
 				}
 				nft_use_dec_restore(&table->use);
-				nft_chain_del(trans->ctx.chain);
+				nft_chain_del(nft_trans_chain(trans));
 				nf_tables_unregister_hook(trans->ctx.net, table,
-							  trans->ctx.chain);
+							  nft_trans_chain(trans));
 			}
 			break;
 		case NFT_MSG_DELCHAIN:
@@ -10637,7 +10635,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 					    &nft_trans_basechain(trans)->hook_list);
 			} else {
 				nft_use_inc_restore(&table->use);
-				nft_clear(trans->ctx.net, trans->ctx.chain);
+				nft_clear(trans->ctx.net, nft_trans_chain(trans));
 			}
 			nft_trans_destroy(trans);
 			break;
-- 
cgit v1.2.3-58-ga151


From 0be908750162ed08ef0dfdec68e1619c8f693f83 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Mon, 13 May 2024 15:00:50 +0200
Subject: netfilter: nf_tables: pass nft_table to destroy function

No functional change intended.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_tables_api.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index bd311b37fc61..6958f922f95a 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -1656,15 +1656,15 @@ static int nf_tables_deltable(struct sk_buff *skb, const struct nfnl_info *info,
 	return nft_flush_table(&ctx);
 }
 
-static void nf_tables_table_destroy(struct nft_ctx *ctx)
+static void nf_tables_table_destroy(struct nft_table *table)
 {
-	if (WARN_ON(ctx->table->use > 0))
+	if (WARN_ON(table->use > 0))
 		return;
 
-	rhltable_destroy(&ctx->table->chains_ht);
-	kfree(ctx->table->name);
-	kfree(ctx->table->udata);
-	kfree(ctx->table);
+	rhltable_destroy(&table->chains_ht);
+	kfree(table->name);
+	kfree(table->udata);
+	kfree(table);
 }
 
 void nft_register_chain_type(const struct nft_chain_type *ctype)
@@ -9521,7 +9521,7 @@ static void nft_commit_release(struct nft_trans *trans)
 	switch (trans->msg_type) {
 	case NFT_MSG_DELTABLE:
 	case NFT_MSG_DESTROYTABLE:
-		nf_tables_table_destroy(&trans->ctx);
+		nf_tables_table_destroy(trans->ctx.table);
 		break;
 	case NFT_MSG_NEWCHAIN:
 		free_percpu(nft_trans_chain_stats(trans));
@@ -10518,7 +10518,7 @@ static void nf_tables_abort_release(struct nft_trans *trans)
 {
 	switch (trans->msg_type) {
 	case NFT_MSG_NEWTABLE:
-		nf_tables_table_destroy(&trans->ctx);
+		nf_tables_table_destroy(trans->ctx.table);
 		break;
 	case NFT_MSG_NEWCHAIN:
 		if (nft_trans_chain_update(trans))
@@ -11490,7 +11490,7 @@ static void __nft_release_table(struct net *net, struct nft_table *table)
 		nft_use_dec(&table->use);
 		nf_tables_chain_destroy(chain);
 	}
-	nf_tables_table_destroy(&ctx);
+	nf_tables_table_destroy(table);
 }
 
 static void __nft_release_tables(struct net *net)
-- 
cgit v1.2.3-58-ga151


From e169285f8c56b8d5702475de0582dc83650c6cee Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Mon, 13 May 2024 15:00:51 +0200
Subject: netfilter: nf_tables: do not store nft_ctx in transaction objects

nft_ctx is huge and most of the information stored within isn't used
at all.

Remove nft_ctx member from the base transaction structure and store
only what is needed.

After this change, relevant struct sizes are:

struct nft_trans_chain { /* size: 120 (-32), cachelines: 2, members: 10 */
struct nft_trans_elem { /* size: 72 (-40), cachelines: 2, members: 4 */
struct nft_trans_flowtable { /* size: 80 (-48), cachelines: 2, members: 5 */
struct nft_trans_obj { /* size: 72 (-40), cachelines: 2, members: 4 */
struct nft_trans_rule { /* size: 80 (-32), cachelines: 2, members: 6 */
struct nft_trans_set { /* size: 96 (-24), cachelines: 2, members: 8 */
struct nft_trans_table { /* size: 56 (-40), cachelines: 1, members: 2 */

struct nft_trans_elem can now be allocated from kmalloc-96 instead of
kmalloc-128 slab.
A further reduction by 8 bytes would even allow for kmalloc-64.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h |  43 ++++++++++--
 net/netfilter/nf_tables_api.c     | 140 ++++++++++++++++++++++----------------
 net/netfilter/nf_tables_offload.c |   8 +--
 3 files changed, 125 insertions(+), 66 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 86e6bd63a205..1e8da1b882ac 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1611,18 +1611,26 @@ static inline int nft_set_elem_is_dead(const struct nft_set_ext *ext)
  * struct nft_trans - nf_tables object update in transaction
  *
  * @list: used internally
+ * @net: struct net
+ * @table: struct nft_table the object resides in
  * @msg_type: message type
- * @put_net: ctx->net needs to be put
- * @ctx: transaction context
+ * @seq: netlink sequence number
+ * @flags: modifiers to new request
+ * @report: notify via unicast netlink message
+ * @put_net: net needs to be put
  *
  * This is the information common to all objects in the transaction,
  * this must always be the first member of derived sub-types.
  */
 struct nft_trans {
 	struct list_head		list;
+	struct net			*net;
+	struct nft_table		*table;
 	int				msg_type;
-	bool				put_net;
-	struct nft_ctx			ctx;
+	u32				seq;
+	u16				flags;
+	u8				report:1;
+	u8				put_net:1;
 };
 
 /**
@@ -1794,6 +1802,33 @@ struct nft_trans_gc {
 	struct rcu_head		rcu;
 };
 
+static inline void nft_ctx_update(struct nft_ctx *ctx,
+				  const struct nft_trans *trans)
+{
+	switch (trans->msg_type) {
+	case NFT_MSG_NEWRULE:
+	case NFT_MSG_DELRULE:
+	case NFT_MSG_DESTROYRULE:
+		ctx->chain = nft_trans_rule_chain(trans);
+		break;
+	case NFT_MSG_NEWCHAIN:
+	case NFT_MSG_DELCHAIN:
+	case NFT_MSG_DESTROYCHAIN:
+		ctx->chain = nft_trans_chain(trans);
+		break;
+	default:
+		ctx->chain = NULL;
+		break;
+	}
+
+	ctx->net = trans->net;
+	ctx->table = trans->table;
+	ctx->family = trans->table->family;
+	ctx->report = trans->report;
+	ctx->flags = trans->flags;
+	ctx->seq = trans->seq;
+}
+
 struct nft_trans_gc *nft_trans_gc_alloc(struct nft_set *set,
 					unsigned int gc_seq, gfp_t gfp);
 void nft_trans_gc_destroy(struct nft_trans_gc *trans);
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 6958f922f95a..02d75aefaa8e 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -159,7 +159,12 @@ static struct nft_trans *nft_trans_alloc_gfp(const struct nft_ctx *ctx,
 
 	INIT_LIST_HEAD(&trans->list);
 	trans->msg_type = msg_type;
-	trans->ctx	= *ctx;
+
+	trans->net = ctx->net;
+	trans->table = ctx->table;
+	trans->seq = ctx->seq;
+	trans->flags = ctx->flags;
+	trans->report = ctx->report;
 
 	return trans;
 }
@@ -1258,7 +1263,7 @@ static bool nft_table_pending_update(const struct nft_ctx *ctx)
 		return true;
 
 	list_for_each_entry(trans, &nft_net->commit_list, list) {
-		if (trans->ctx.table == ctx->table &&
+		if (trans->table == ctx->table &&
 		    ((trans->msg_type == NFT_MSG_NEWCHAIN &&
 		      nft_trans_chain_update(trans)) ||
 		     (trans->msg_type == NFT_MSG_DELCHAIN &&
@@ -2100,7 +2105,7 @@ static void nft_chain_stats_replace(struct nft_trans_chain *trans)
 
 	trans->stats =
 		rcu_replace_pointer(chain->stats, trans->stats,
-				    lockdep_commit_lock_is_held(t->ctx.net));
+				    lockdep_commit_lock_is_held(t->net));
 
 	if (!trans->stats)
 		static_branch_inc(&nft_counters_enabled);
@@ -2766,7 +2771,7 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy,
 		err = -EEXIST;
 		list_for_each_entry(tmp, &nft_net->commit_list, list) {
 			if (tmp->msg_type == NFT_MSG_NEWCHAIN &&
-			    tmp->ctx.table == table &&
+			    tmp->table == table &&
 			    nft_trans_chain_update(tmp) &&
 			    nft_trans_chain_name(tmp) &&
 			    strcmp(name, nft_trans_chain_name(tmp)) == 0) {
@@ -9472,7 +9477,7 @@ static void nft_chain_commit_drop_policy(struct nft_trans_chain *trans)
 
 static void nft_chain_commit_update(struct nft_trans_chain *trans)
 {
-	struct nft_table *table = trans->nft_trans_binding.nft_trans.ctx.table;
+	struct nft_table *table = trans->nft_trans_binding.nft_trans.table;
 	struct nft_base_chain *basechain;
 
 	if (trans->name) {
@@ -9501,7 +9506,8 @@ static void nft_chain_commit_update(struct nft_trans_chain *trans)
 	}
 }
 
-static void nft_obj_commit_update(struct nft_trans *trans)
+static void nft_obj_commit_update(const struct nft_ctx *ctx,
+				  struct nft_trans *trans)
 {
 	struct nft_object *newobj;
 	struct nft_object *obj;
@@ -9513,15 +9519,21 @@ static void nft_obj_commit_update(struct nft_trans *trans)
 		return;
 
 	obj->ops->update(obj, newobj);
-	nft_obj_destroy(&trans->ctx, newobj);
+	nft_obj_destroy(ctx, newobj);
 }
 
 static void nft_commit_release(struct nft_trans *trans)
 {
+	struct nft_ctx ctx = {
+		.net = trans->net,
+	};
+
+	nft_ctx_update(&ctx, trans);
+
 	switch (trans->msg_type) {
 	case NFT_MSG_DELTABLE:
 	case NFT_MSG_DESTROYTABLE:
-		nf_tables_table_destroy(trans->ctx.table);
+		nf_tables_table_destroy(trans->table);
 		break;
 	case NFT_MSG_NEWCHAIN:
 		free_percpu(nft_trans_chain_stats(trans));
@@ -9536,21 +9548,21 @@ static void nft_commit_release(struct nft_trans *trans)
 		break;
 	case NFT_MSG_DELRULE:
 	case NFT_MSG_DESTROYRULE:
-		nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans));
+		nf_tables_rule_destroy(&ctx, nft_trans_rule(trans));
 		break;
 	case NFT_MSG_DELSET:
 	case NFT_MSG_DESTROYSET:
-		nft_set_destroy(&trans->ctx, nft_trans_set(trans));
+		nft_set_destroy(&ctx, nft_trans_set(trans));
 		break;
 	case NFT_MSG_DELSETELEM:
 	case NFT_MSG_DESTROYSETELEM:
-		nf_tables_set_elem_destroy(&trans->ctx,
+		nf_tables_set_elem_destroy(&ctx,
 					   nft_trans_elem_set(trans),
 					   nft_trans_elem_priv(trans));
 		break;
 	case NFT_MSG_DELOBJ:
 	case NFT_MSG_DESTROYOBJ:
-		nft_obj_destroy(&trans->ctx, nft_trans_obj(trans));
+		nft_obj_destroy(&ctx, nft_trans_obj(trans));
 		break;
 	case NFT_MSG_DELFLOWTABLE:
 	case NFT_MSG_DESTROYFLOWTABLE:
@@ -9562,7 +9574,7 @@ static void nft_commit_release(struct nft_trans *trans)
 	}
 
 	if (trans->put_net)
-		put_net(trans->ctx.net);
+		put_net(trans->net);
 
 	kfree(trans);
 }
@@ -10042,7 +10054,7 @@ static void nf_tables_commit_release(struct net *net)
 
 	trans = list_last_entry(&nft_net->commit_list,
 				struct nft_trans, list);
-	get_net(trans->ctx.net);
+	get_net(trans->net);
 	WARN_ON_ONCE(trans->put_net);
 
 	trans->put_net = true;
@@ -10186,6 +10198,7 @@ static void nft_gc_seq_end(struct nftables_pernet *nft_net, unsigned int gc_seq)
 static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 {
 	struct nftables_pernet *nft_net = nft_pernet(net);
+	const struct nlmsghdr *nlh = nlmsg_hdr(skb);
 	struct nft_trans_binding *trans_binding;
 	struct nft_trans *trans, *next;
 	unsigned int base_seq, gc_seq;
@@ -10193,6 +10206,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 	struct nft_trans_elem *te;
 	struct nft_chain *chain;
 	struct nft_table *table;
+	struct nft_ctx ctx;
 	LIST_HEAD(adl);
 	int err;
 
@@ -10201,6 +10215,8 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 		return 0;
 	}
 
+	nft_ctx_init(&ctx, net, skb, nlh, NFPROTO_UNSPEC, NULL, NULL, NULL);
+
 	list_for_each_entry(trans_binding, &nft_net->binding_list, binding_list) {
 		trans = &trans_binding->nft_trans;
 		switch (trans->msg_type) {
@@ -10238,7 +10254,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 
 	/* 1.  Allocate space for next generation rules_gen_X[] */
 	list_for_each_entry_safe(trans, next, &nft_net->commit_list, list) {
-		struct nft_table *table = trans->ctx.table;
+		struct nft_table *table = trans->table;
 		int ret;
 
 		ret = nf_tables_commit_audit_alloc(&adl, table);
@@ -10282,7 +10298,9 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 	net->nft.gencursor = nft_gencursor_next(net);
 
 	list_for_each_entry_safe(trans, next, &nft_net->commit_list, list) {
-		struct nft_table *table = trans->ctx.table;
+		struct nft_table *table = trans->table;
+
+		nft_ctx_update(&ctx, trans);
 
 		nf_tables_commit_audit_collect(&adl, table, trans->msg_type);
 		switch (trans->msg_type) {
@@ -10299,18 +10317,18 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 			} else {
 				nft_clear(net, table);
 			}
-			nf_tables_table_notify(&trans->ctx, NFT_MSG_NEWTABLE);
+			nf_tables_table_notify(&ctx, NFT_MSG_NEWTABLE);
 			nft_trans_destroy(trans);
 			break;
 		case NFT_MSG_DELTABLE:
 		case NFT_MSG_DESTROYTABLE:
 			list_del_rcu(&table->list);
-			nf_tables_table_notify(&trans->ctx, trans->msg_type);
+			nf_tables_table_notify(&ctx, trans->msg_type);
 			break;
 		case NFT_MSG_NEWCHAIN:
 			if (nft_trans_chain_update(trans)) {
 				nft_chain_commit_update(nft_trans_container_chain(trans));
-				nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN,
+				nf_tables_chain_notify(&ctx, NFT_MSG_NEWCHAIN,
 						       &nft_trans_chain_hooks(trans));
 				list_splice(&nft_trans_chain_hooks(trans),
 					    &nft_trans_basechain(trans)->hook_list);
@@ -10318,14 +10336,14 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 			} else {
 				nft_chain_commit_drop_policy(nft_trans_container_chain(trans));
 				nft_clear(net, nft_trans_chain(trans));
-				nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN, NULL);
+				nf_tables_chain_notify(&ctx, NFT_MSG_NEWCHAIN, NULL);
 				nft_trans_destroy(trans);
 			}
 			break;
 		case NFT_MSG_DELCHAIN:
 		case NFT_MSG_DESTROYCHAIN:
 			if (nft_trans_chain_update(trans)) {
-				nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN,
+				nf_tables_chain_notify(&ctx, NFT_MSG_DELCHAIN,
 						       &nft_trans_chain_hooks(trans));
 				if (!(table->flags & NFT_TABLE_F_DORMANT)) {
 					nft_netdev_unregister_hooks(net,
@@ -10334,16 +10352,15 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 				}
 			} else {
 				nft_chain_del(nft_trans_chain(trans));
-				nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN,
+				nf_tables_chain_notify(&ctx, NFT_MSG_DELCHAIN,
 						       NULL);
-				nf_tables_unregister_hook(trans->ctx.net, table,
+				nf_tables_unregister_hook(ctx.net, ctx.table,
 							  nft_trans_chain(trans));
 			}
 			break;
 		case NFT_MSG_NEWRULE:
-			nft_clear(trans->ctx.net, nft_trans_rule(trans));
-			nf_tables_rule_notify(&trans->ctx,
-					      nft_trans_rule(trans),
+			nft_clear(net, nft_trans_rule(trans));
+			nf_tables_rule_notify(&ctx, nft_trans_rule(trans),
 					      NFT_MSG_NEWRULE);
 			if (nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD)
 				nft_flow_rule_destroy(nft_trans_flow_rule(trans));
@@ -10353,11 +10370,9 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 		case NFT_MSG_DELRULE:
 		case NFT_MSG_DESTROYRULE:
 			list_del_rcu(&nft_trans_rule(trans)->list);
-			nf_tables_rule_notify(&trans->ctx,
-					      nft_trans_rule(trans),
+			nf_tables_rule_notify(&ctx, nft_trans_rule(trans),
 					      trans->msg_type);
-			nft_rule_expr_deactivate(&trans->ctx,
-						 nft_trans_rule(trans),
+			nft_rule_expr_deactivate(&ctx, nft_trans_rule(trans),
 						 NFT_TRANS_COMMIT);
 
 			if (nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD)
@@ -10381,7 +10396,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 				    !list_empty(&nft_trans_set(trans)->bindings))
 					nft_use_dec(&table->use);
 			}
-			nf_tables_set_notify(&trans->ctx, nft_trans_set(trans),
+			nf_tables_set_notify(&ctx, nft_trans_set(trans),
 					     NFT_MSG_NEWSET, GFP_KERNEL);
 			nft_trans_destroy(trans);
 			break;
@@ -10389,14 +10404,14 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 		case NFT_MSG_DESTROYSET:
 			nft_trans_set(trans)->dead = 1;
 			list_del_rcu(&nft_trans_set(trans)->list);
-			nf_tables_set_notify(&trans->ctx, nft_trans_set(trans),
+			nf_tables_set_notify(&ctx, nft_trans_set(trans),
 					     trans->msg_type, GFP_KERNEL);
 			break;
 		case NFT_MSG_NEWSETELEM:
 			te = nft_trans_container_elem(trans);
 
 			nft_setelem_activate(net, te->set, te->elem_priv);
-			nf_tables_setelem_notify(&trans->ctx, te->set,
+			nf_tables_setelem_notify(&ctx, te->set,
 						 te->elem_priv,
 						 NFT_MSG_NEWSETELEM);
 			if (te->set->ops->commit &&
@@ -10410,7 +10425,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 		case NFT_MSG_DESTROYSETELEM:
 			te = nft_trans_container_elem(trans);
 
-			nf_tables_setelem_notify(&trans->ctx, te->set,
+			nf_tables_setelem_notify(&ctx, te->set,
 						 te->elem_priv,
 						 trans->msg_type);
 			nft_setelem_remove(net, te->set, te->elem_priv);
@@ -10426,13 +10441,13 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 			break;
 		case NFT_MSG_NEWOBJ:
 			if (nft_trans_obj_update(trans)) {
-				nft_obj_commit_update(trans);
-				nf_tables_obj_notify(&trans->ctx,
+				nft_obj_commit_update(&ctx, trans);
+				nf_tables_obj_notify(&ctx,
 						     nft_trans_obj(trans),
 						     NFT_MSG_NEWOBJ);
 			} else {
 				nft_clear(net, nft_trans_obj(trans));
-				nf_tables_obj_notify(&trans->ctx,
+				nf_tables_obj_notify(&ctx,
 						     nft_trans_obj(trans),
 						     NFT_MSG_NEWOBJ);
 				nft_trans_destroy(trans);
@@ -10441,14 +10456,14 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 		case NFT_MSG_DELOBJ:
 		case NFT_MSG_DESTROYOBJ:
 			nft_obj_del(nft_trans_obj(trans));
-			nf_tables_obj_notify(&trans->ctx, nft_trans_obj(trans),
+			nf_tables_obj_notify(&ctx, nft_trans_obj(trans),
 					     trans->msg_type);
 			break;
 		case NFT_MSG_NEWFLOWTABLE:
 			if (nft_trans_flowtable_update(trans)) {
 				nft_trans_flowtable(trans)->data.flags =
 					nft_trans_flowtable_flags(trans);
-				nf_tables_flowtable_notify(&trans->ctx,
+				nf_tables_flowtable_notify(&ctx,
 							   nft_trans_flowtable(trans),
 							   &nft_trans_flowtable_hooks(trans),
 							   NFT_MSG_NEWFLOWTABLE);
@@ -10456,7 +10471,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 					    &nft_trans_flowtable(trans)->hook_list);
 			} else {
 				nft_clear(net, nft_trans_flowtable(trans));
-				nf_tables_flowtable_notify(&trans->ctx,
+				nf_tables_flowtable_notify(&ctx,
 							   nft_trans_flowtable(trans),
 							   NULL,
 							   NFT_MSG_NEWFLOWTABLE);
@@ -10466,7 +10481,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 		case NFT_MSG_DELFLOWTABLE:
 		case NFT_MSG_DESTROYFLOWTABLE:
 			if (nft_trans_flowtable_update(trans)) {
-				nf_tables_flowtable_notify(&trans->ctx,
+				nf_tables_flowtable_notify(&ctx,
 							   nft_trans_flowtable(trans),
 							   &nft_trans_flowtable_hooks(trans),
 							   trans->msg_type);
@@ -10474,7 +10489,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 								   &nft_trans_flowtable_hooks(trans));
 			} else {
 				list_del_rcu(&nft_trans_flowtable(trans)->list);
-				nf_tables_flowtable_notify(&trans->ctx,
+				nf_tables_flowtable_notify(&ctx,
 							   nft_trans_flowtable(trans),
 							   NULL,
 							   trans->msg_type);
@@ -10516,9 +10531,13 @@ static void nf_tables_module_autoload(struct net *net)
 
 static void nf_tables_abort_release(struct nft_trans *trans)
 {
+	struct nft_ctx ctx = { };
+
+	nft_ctx_update(&ctx, trans);
+
 	switch (trans->msg_type) {
 	case NFT_MSG_NEWTABLE:
-		nf_tables_table_destroy(trans->ctx.table);
+		nf_tables_table_destroy(trans->table);
 		break;
 	case NFT_MSG_NEWCHAIN:
 		if (nft_trans_chain_update(trans))
@@ -10527,17 +10546,17 @@ static void nf_tables_abort_release(struct nft_trans *trans)
 			nf_tables_chain_destroy(nft_trans_chain(trans));
 		break;
 	case NFT_MSG_NEWRULE:
-		nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans));
+		nf_tables_rule_destroy(&ctx, nft_trans_rule(trans));
 		break;
 	case NFT_MSG_NEWSET:
-		nft_set_destroy(&trans->ctx, nft_trans_set(trans));
+		nft_set_destroy(&ctx, nft_trans_set(trans));
 		break;
 	case NFT_MSG_NEWSETELEM:
 		nft_set_elem_destroy(nft_trans_elem_set(trans),
 				     nft_trans_elem_priv(trans), true);
 		break;
 	case NFT_MSG_NEWOBJ:
-		nft_obj_destroy(&trans->ctx, nft_trans_obj(trans));
+		nft_obj_destroy(&ctx, nft_trans_obj(trans));
 		break;
 	case NFT_MSG_NEWFLOWTABLE:
 		if (nft_trans_flowtable_update(trans))
@@ -10569,6 +10588,9 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 	struct nft_trans *trans, *next;
 	LIST_HEAD(set_update_list);
 	struct nft_trans_elem *te;
+	struct nft_ctx ctx = {
+		.net = net,
+	};
 	int err = 0;
 
 	if (action == NFNL_ABORT_VALIDATE &&
@@ -10577,7 +10599,9 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 
 	list_for_each_entry_safe_reverse(trans, next, &nft_net->commit_list,
 					 list) {
-		struct nft_table *table = trans->ctx.table;
+		struct nft_table *table = trans->table;
+
+		nft_ctx_update(&ctx, trans);
 
 		switch (trans->msg_type) {
 		case NFT_MSG_NEWTABLE:
@@ -10604,7 +10628,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 			break;
 		case NFT_MSG_DELTABLE:
 		case NFT_MSG_DESTROYTABLE:
-			nft_clear(trans->ctx.net, table);
+			nft_clear(trans->net, table);
 			nft_trans_destroy(trans);
 			break;
 		case NFT_MSG_NEWCHAIN:
@@ -10624,7 +10648,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 				}
 				nft_use_dec_restore(&table->use);
 				nft_chain_del(nft_trans_chain(trans));
-				nf_tables_unregister_hook(trans->ctx.net, table,
+				nf_tables_unregister_hook(trans->net, table,
 							  nft_trans_chain(trans));
 			}
 			break;
@@ -10635,7 +10659,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 					    &nft_trans_basechain(trans)->hook_list);
 			} else {
 				nft_use_inc_restore(&table->use);
-				nft_clear(trans->ctx.net, nft_trans_chain(trans));
+				nft_clear(trans->net, nft_trans_chain(trans));
 			}
 			nft_trans_destroy(trans);
 			break;
@@ -10646,7 +10670,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 			}
 			nft_use_dec_restore(&nft_trans_rule_chain(trans)->use);
 			list_del_rcu(&nft_trans_rule(trans)->list);
-			nft_rule_expr_deactivate(&trans->ctx,
+			nft_rule_expr_deactivate(&ctx,
 						 nft_trans_rule(trans),
 						 NFT_TRANS_ABORT);
 			if (nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD)
@@ -10655,8 +10679,8 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 		case NFT_MSG_DELRULE:
 		case NFT_MSG_DESTROYRULE:
 			nft_use_inc_restore(&nft_trans_rule_chain(trans)->use);
-			nft_clear(trans->ctx.net, nft_trans_rule(trans));
-			nft_rule_expr_activate(&trans->ctx, nft_trans_rule(trans));
+			nft_clear(trans->net, nft_trans_rule(trans));
+			nft_rule_expr_activate(&ctx, nft_trans_rule(trans));
 			if (nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD)
 				nft_flow_rule_destroy(nft_trans_flow_rule(trans));
 
@@ -10678,9 +10702,9 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 		case NFT_MSG_DELSET:
 		case NFT_MSG_DESTROYSET:
 			nft_use_inc_restore(&table->use);
-			nft_clear(trans->ctx.net, nft_trans_set(trans));
+			nft_clear(trans->net, nft_trans_set(trans));
 			if (nft_trans_set(trans)->flags & (NFT_SET_MAP | NFT_SET_OBJECT))
-				nft_map_activate(&trans->ctx, nft_trans_set(trans));
+				nft_map_activate(&ctx, nft_trans_set(trans));
 
 			nft_trans_destroy(trans);
 			break;
@@ -10720,7 +10744,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 			break;
 		case NFT_MSG_NEWOBJ:
 			if (nft_trans_obj_update(trans)) {
-				nft_obj_destroy(&trans->ctx, nft_trans_obj_newobj(trans));
+				nft_obj_destroy(&ctx, nft_trans_obj_newobj(trans));
 				nft_trans_destroy(trans);
 			} else {
 				nft_use_dec_restore(&table->use);
@@ -10730,7 +10754,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 		case NFT_MSG_DELOBJ:
 		case NFT_MSG_DESTROYOBJ:
 			nft_use_inc_restore(&table->use);
-			nft_clear(trans->ctx.net, nft_trans_obj(trans));
+			nft_clear(trans->net, nft_trans_obj(trans));
 			nft_trans_destroy(trans);
 			break;
 		case NFT_MSG_NEWFLOWTABLE:
@@ -10751,7 +10775,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 					    &nft_trans_flowtable(trans)->hook_list);
 			} else {
 				nft_use_inc_restore(&table->use);
-				nft_clear(trans->ctx.net, nft_trans_flowtable(trans));
+				nft_clear(trans->net, nft_trans_flowtable(trans));
 			}
 			nft_trans_destroy(trans);
 			break;
diff --git a/net/netfilter/nf_tables_offload.c b/net/netfilter/nf_tables_offload.c
index 0619feb10abb..64675f1c7f29 100644
--- a/net/netfilter/nf_tables_offload.c
+++ b/net/netfilter/nf_tables_offload.c
@@ -513,7 +513,7 @@ static void nft_flow_rule_offload_abort(struct net *net,
 	int err = 0;
 
 	list_for_each_entry_continue_reverse(trans, &nft_net->commit_list, list) {
-		if (trans->ctx.family != NFPROTO_NETDEV)
+		if (trans->table->family != NFPROTO_NETDEV)
 			continue;
 
 		switch (trans->msg_type) {
@@ -564,7 +564,7 @@ int nft_flow_rule_offload_commit(struct net *net)
 	u8 policy;
 
 	list_for_each_entry(trans, &nft_net->commit_list, list) {
-		if (trans->ctx.family != NFPROTO_NETDEV)
+		if (trans->table->family != NFPROTO_NETDEV)
 			continue;
 
 		switch (trans->msg_type) {
@@ -589,8 +589,8 @@ int nft_flow_rule_offload_commit(struct net *net)
 			if (!(nft_trans_rule_chain(trans)->flags & NFT_CHAIN_HW_OFFLOAD))
 				continue;
 
-			if (trans->ctx.flags & NLM_F_REPLACE ||
-			    !(trans->ctx.flags & NLM_F_APPEND)) {
+			if (trans->flags & NLM_F_REPLACE ||
+			    !(trans->flags & NLM_F_APPEND)) {
 				err = -EOPNOTSUPP;
 				break;
 			}
-- 
cgit v1.2.3-58-ga151


From 53796b03295cf7ab1fc8600016fa6dfbf4a494a0 Mon Sep 17 00:00:00 2001
From: Ismael Luceno <iluceno@suse.de>
Date: Thu, 23 May 2024 18:54:44 +0200
Subject: ipvs: Avoid unnecessary calls to skb_is_gso_sctp

In the context of the SCTP SNAT/DNAT handler, these calls can only
return true.

Fixes: e10d3ba4d434 ("ipvs: Fix checksumming on GSO of SCTP packets")
Signed-off-by: Ismael Luceno <iluceno@suse.de>
Acked-by: Julian Anastasov <ja@ssi.bg>
Acked-by: Simon Horman <horms@kernel.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/ipvs/ip_vs_proto_sctp.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c
index 1e689c714127..83e452916403 100644
--- a/net/netfilter/ipvs/ip_vs_proto_sctp.c
+++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c
@@ -126,7 +126,7 @@ sctp_snat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 	if (sctph->source != cp->vport || payload_csum ||
 	    skb->ip_summed == CHECKSUM_PARTIAL) {
 		sctph->source = cp->vport;
-		if (!skb_is_gso(skb) || !skb_is_gso_sctp(skb))
+		if (!skb_is_gso(skb))
 			sctp_nat_csum(skb, sctph, sctphoff);
 	} else {
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
@@ -175,7 +175,7 @@ sctp_dnat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 	    (skb->ip_summed == CHECKSUM_PARTIAL &&
 	     !(skb_dst(skb)->dev->features & NETIF_F_SCTP_CRC))) {
 		sctph->dest = cp->dport;
-		if (!skb_is_gso(skb) || !skb_is_gso_sctp(skb))
+		if (!skb_is_gso(skb))
 			sctp_nat_csum(skb, sctph, sctphoff);
 	} else if (skb->ip_summed != CHECKSUM_PARTIAL) {
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
-- 
cgit v1.2.3-58-ga151


From 0b88d1654d556264bcd24a9cb6383f0888e30131 Mon Sep 17 00:00:00 2001
From: Yunjian Wang <wangyunjian@huawei.com>
Date: Fri, 31 May 2024 11:48:47 +0800
Subject: netfilter: nf_conncount: fix wrong variable type

Now there is a issue is that code checks reports a warning: implicit
narrowing conversion from type 'unsigned int' to small type 'u8' (the
'keylen' variable). Fix it by removing the 'keylen' variable.

Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nf_conncount.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/net/netfilter/nf_conncount.c b/net/netfilter/nf_conncount.c
index 8715617b02fe..34ba14e59e95 100644
--- a/net/netfilter/nf_conncount.c
+++ b/net/netfilter/nf_conncount.c
@@ -321,7 +321,6 @@ insert_tree(struct net *net,
 	struct nf_conncount_rb *rbconn;
 	struct nf_conncount_tuple *conn;
 	unsigned int count = 0, gc_count = 0;
-	u8 keylen = data->keylen;
 	bool do_gc = true;
 
 	spin_lock_bh(&nf_conncount_locks[hash]);
@@ -333,7 +332,7 @@ restart:
 		rbconn = rb_entry(*rbnode, struct nf_conncount_rb, node);
 
 		parent = *rbnode;
-		diff = key_diff(key, rbconn->key, keylen);
+		diff = key_diff(key, rbconn->key, data->keylen);
 		if (diff < 0) {
 			rbnode = &((*rbnode)->rb_left);
 		} else if (diff > 0) {
@@ -378,7 +377,7 @@ restart:
 
 	conn->tuple = *tuple;
 	conn->zone = *zone;
-	memcpy(rbconn->key, key, sizeof(u32) * keylen);
+	memcpy(rbconn->key, key, sizeof(u32) * data->keylen);
 
 	nf_conncount_list_init(&rbconn->list);
 	list_add(&conn->node, &rbconn->list.head);
@@ -403,7 +402,6 @@ count_tree(struct net *net,
 	struct rb_node *parent;
 	struct nf_conncount_rb *rbconn;
 	unsigned int hash;
-	u8 keylen = data->keylen;
 
 	hash = jhash2(key, data->keylen, conncount_rnd) % CONNCOUNT_SLOTS;
 	root = &data->root[hash];
@@ -414,7 +412,7 @@ count_tree(struct net *net,
 
 		rbconn = rb_entry(parent, struct nf_conncount_rb, node);
 
-		diff = key_diff(key, rbconn->key, keylen);
+		diff = key_diff(key, rbconn->key, data->keylen);
 		if (diff < 0) {
 			parent = rcu_dereference_raw(parent->rb_left);
 		} else if (diff > 0) {
-- 
cgit v1.2.3-58-ga151


From fe87a8deaad46c9a45381624f7e5f2f4fd145721 Mon Sep 17 00:00:00 2001
From: Lin Ma <linma@zju.edu.cn>
Date: Fri, 31 May 2024 09:28:47 +0800
Subject: netfilter: cttimeout: remove 'l3num' attr check

After commit dd2934a95701 ("netfilter: conntrack: remove l3->l4 mapping
information"), the attribute of type `CTA_TIMEOUT_L3PROTO` is not used
any more in function cttimeout_default_set.

However, the previous commit ea9cf2a55a7b ("netfilter: cttimeout: remove
set but not used variable 'l3num'") forgot to remove the attribute
present check when removing the related variable.

This commit removes that check to ensure consistency.

Signed-off-by: Lin Ma <linma@zju.edu.cn>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/nfnetlink_cttimeout.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c
index f466af4f8531..eab4f476b47f 100644
--- a/net/netfilter/nfnetlink_cttimeout.c
+++ b/net/netfilter/nfnetlink_cttimeout.c
@@ -366,8 +366,7 @@ static int cttimeout_default_set(struct sk_buff *skb,
 	__u8 l4num;
 	int ret;
 
-	if (!cda[CTA_TIMEOUT_L3PROTO] ||
-	    !cda[CTA_TIMEOUT_L4PROTO] ||
+	if (!cda[CTA_TIMEOUT_L4PROTO] ||
 	    !cda[CTA_TIMEOUT_DATA])
 		return -EINVAL;
 
-- 
cgit v1.2.3-58-ga151


From e29630247be24c3987e2b048f8e152771b32d38b Mon Sep 17 00:00:00 2001
From: Pablo Neira Ayuso <pablo@netfilter.org>
Date: Mon, 3 Jun 2024 20:16:59 +0200
Subject: netfilter: nf_tables: rise cap on SELinux secmark context

secmark context is artificially limited 256 bytes, rise it to 4Kbytes.

Fixes: fb961945457f ("netfilter: nf_tables: add SECMARK support")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/uapi/linux/netfilter/nf_tables.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index aa4094ca2444..639894ed1b97 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -1376,7 +1376,7 @@ enum nft_secmark_attributes {
 #define NFTA_SECMARK_MAX	(__NFTA_SECMARK_MAX - 1)
 
 /* Max security context length */
-#define NFT_SECMARK_CTX_MAXLEN		256
+#define NFT_SECMARK_CTX_MAXLEN		4096
 
 /**
  * enum nft_reject_types - nf_tables reject expression reject types
-- 
cgit v1.2.3-58-ga151


From d891317fe4fbe5a85522dccb66d3a159915d7983 Mon Sep 17 00:00:00 2001
From: Li RongQing <lirongqing@baidu.com>
Date: Fri, 21 Jun 2024 17:45:52 +0800
Subject: virtio_net: Remove u64_stats_update_begin()/end() for stats fetch

This place is fetching the stats, u64_stats_update_begin()/end()
should not be used, and the fetcher of stats is in the same context
as the updater of the stats, so don't need any protection

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Li RongQing <lirongqing@baidu.com>
Link: https://lore.kernel.org/20240621094552.53469-1-lirongqing@baidu.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index b1f8b720733e..f21297751310 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -2382,12 +2382,13 @@ static void virtnet_rx_dim_update(struct virtnet_info *vi, struct receive_queue
 	if (!rq->packets_in_napi)
 		return;
 
-	u64_stats_update_begin(&rq->stats.syncp);
+	/* Don't need protection when fetching stats, since fetcher and
+	 * updater of the stats are in same context
+	 */
 	dim_update_sample(rq->calls,
 			  u64_stats_read(&rq->stats.packets),
 			  u64_stats_read(&rq->stats.bytes),
 			  &cur_sample);
-	u64_stats_update_end(&rq->stats.syncp);
 
 	net_dim(&rq->dim, cur_sample);
 	rq->packets_in_napi = 0;
-- 
cgit v1.2.3-58-ga151


From 8d653d26ffc7cbc80335ea74f498237c5709cdfa Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Mon, 24 Jun 2024 15:25:24 +0200
Subject: ravb: Improve ravb_hw_info instance order
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Move ravb_gen2_hw_info before ravb_gen3_hw_info to match
ravb_match_table[] order.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Link: https://patch.msgid.link/a76febe3737e26365a784e9193da9363f22aa550.1719234830.git.geert+renesas@glider.be
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/renesas/ravb_main.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index c1546b916e4e..974e0bb9da19 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -2652,7 +2652,7 @@ static int ravb_mdio_release(struct ravb_private *priv)
 	return 0;
 }
 
-static const struct ravb_hw_info ravb_gen3_hw_info = {
+static const struct ravb_hw_info ravb_gen2_hw_info = {
 	.receive = ravb_rx_rcar,
 	.set_rate = ravb_set_rate_rcar,
 	.set_feature = ravb_set_features_rcar,
@@ -2668,16 +2668,13 @@ static const struct ravb_hw_info ravb_gen3_hw_info = {
 	.rx_buffer_size = SZ_2K +
 			  SKB_DATA_ALIGN(sizeof(struct skb_shared_info)),
 	.rx_desc_size = sizeof(struct ravb_ex_rx_desc),
-	.internal_delay = 1,
-	.tx_counters = 1,
-	.multi_irqs = 1,
-	.irq_en_dis = 1,
-	.ccc_gac = 1,
+	.aligned_tx = 1,
+	.gptp = 1,
 	.nc_queues = 1,
 	.magic_pkt = 1,
 };
 
-static const struct ravb_hw_info ravb_gen2_hw_info = {
+static const struct ravb_hw_info ravb_gen3_hw_info = {
 	.receive = ravb_rx_rcar,
 	.set_rate = ravb_set_rate_rcar,
 	.set_feature = ravb_set_features_rcar,
@@ -2693,8 +2690,11 @@ static const struct ravb_hw_info ravb_gen2_hw_info = {
 	.rx_buffer_size = SZ_2K +
 			  SKB_DATA_ALIGN(sizeof(struct skb_shared_info)),
 	.rx_desc_size = sizeof(struct ravb_ex_rx_desc),
-	.aligned_tx = 1,
-	.gptp = 1,
+	.internal_delay = 1,
+	.tx_counters = 1,
+	.multi_irqs = 1,
+	.irq_en_dis = 1,
+	.ccc_gac = 1,
 	.nc_queues = 1,
 	.magic_pkt = 1,
 };
-- 
cgit v1.2.3-58-ga151


From 6e0713cc820688deab490edd242f0bfec489738a Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Mon, 24 Jun 2024 15:25:25 +0200
Subject: ravb: Add MII support for R-Car V4M
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

All EtherAVB instances on R-Car Gen3/Gen4 SoCs support the RGMII
interface.  In addition, the first two EtherAVB instances on R-Car V4M
also support the MII interface, but this is not yet supported by the
driver.

Add support for MII on R-Car Gen4 by adding an R-Car Gen4-specific EMAC
initialization function that selects the MII clock instead of the RGMII
clock when the PHY interface is MII.  Note that all implementations of
EtherAVB on R-Car Gen4 SoCs have the APSR register, but only MII-capable
instances are documented to have the MIISELECT bit, which has a
documented value of zero when reserved.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Link: https://patch.msgid.link/3a21d1d6680864aa85afff9260234c2b8054020a.1719234830.git.geert+renesas@glider.be
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/renesas/ravb.h      |  1 +
 drivers/net/ethernet/renesas/ravb_main.c | 37 +++++++++++++++++++++++++++++++-
 2 files changed, 37 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/renesas/ravb.h b/drivers/net/ethernet/renesas/ravb.h
index 6b2444d31fcc..9893c91af105 100644
--- a/drivers/net/ethernet/renesas/ravb.h
+++ b/drivers/net/ethernet/renesas/ravb.h
@@ -258,6 +258,7 @@ enum APSR_BIT {
 	APSR_CMSW	= 0x00000010,
 	APSR_RDM	= 0x00002000,
 	APSR_TDM	= 0x00004000,
+	APSR_MIISELECT	= 0x01000000,	/* R-Car V4M only */
 };
 
 /* RCR */
diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 974e0bb9da19..6605e4f4af53 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -579,6 +579,16 @@ static void ravb_emac_init_rcar(struct net_device *ndev)
 	ravb_write(ndev, ECSIPR_ICDIP | ECSIPR_MPDIP | ECSIPR_LCHNGIP, ECSIPR);
 }
 
+static void ravb_emac_init_rcar_gen4(struct net_device *ndev)
+{
+	struct ravb_private *priv = netdev_priv(ndev);
+	bool mii = priv->phy_interface == PHY_INTERFACE_MODE_MII;
+
+	ravb_modify(ndev, APSR, APSR_MIISELECT, mii ? APSR_MIISELECT : 0);
+
+	ravb_emac_init_rcar(ndev);
+}
+
 /* E-MAC init function */
 static void ravb_emac_init(struct net_device *ndev)
 {
@@ -2699,6 +2709,31 @@ static const struct ravb_hw_info ravb_gen3_hw_info = {
 	.magic_pkt = 1,
 };
 
+static const struct ravb_hw_info ravb_gen4_hw_info = {
+	.receive = ravb_rx_rcar,
+	.set_rate = ravb_set_rate_rcar,
+	.set_feature = ravb_set_features_rcar,
+	.dmac_init = ravb_dmac_init_rcar,
+	.emac_init = ravb_emac_init_rcar_gen4,
+	.gstrings_stats = ravb_gstrings_stats,
+	.gstrings_size = sizeof(ravb_gstrings_stats),
+	.net_hw_features = NETIF_F_RXCSUM,
+	.net_features = NETIF_F_RXCSUM,
+	.stats_len = ARRAY_SIZE(ravb_gstrings_stats),
+	.tccr_mask = TCCR_TSRQ0 | TCCR_TSRQ1 | TCCR_TSRQ2 | TCCR_TSRQ3,
+	.rx_max_frame_size = SZ_2K,
+	.rx_buffer_size = SZ_2K +
+			  SKB_DATA_ALIGN(sizeof(struct skb_shared_info)),
+	.rx_desc_size = sizeof(struct ravb_ex_rx_desc),
+	.internal_delay = 1,
+	.tx_counters = 1,
+	.multi_irqs = 1,
+	.irq_en_dis = 1,
+	.ccc_gac = 1,
+	.nc_queues = 1,
+	.magic_pkt = 1,
+};
+
 static const struct ravb_hw_info ravb_rzv2m_hw_info = {
 	.receive = ravb_rx_rcar,
 	.set_rate = ravb_set_rate_rcar,
@@ -2751,7 +2786,7 @@ static const struct of_device_id ravb_match_table[] = {
 	{ .compatible = "renesas,etheravb-rcar-gen2", .data = &ravb_gen2_hw_info },
 	{ .compatible = "renesas,etheravb-r8a7795", .data = &ravb_gen3_hw_info },
 	{ .compatible = "renesas,etheravb-rcar-gen3", .data = &ravb_gen3_hw_info },
-	{ .compatible = "renesas,etheravb-rcar-gen4", .data = &ravb_gen3_hw_info },
+	{ .compatible = "renesas,etheravb-rcar-gen4", .data = &ravb_gen4_hw_info },
 	{ .compatible = "renesas,etheravb-rzv2m", .data = &ravb_rzv2m_hw_info },
 	{ .compatible = "renesas,rzg2l-gbeth", .data = &gbeth_hw_info },
 	{ }
-- 
cgit v1.2.3-58-ga151


From 0e942053e4dc42a760f48c1981f3239825430f15 Mon Sep 17 00:00:00 2001
From: Heng Qi <hengqi@linux.alibaba.com>
Date: Fri, 21 Jun 2024 18:13:49 +0800
Subject: linux/dim: move useful macros to .h file

Useful macros will be used effectively elsewhere.
These will be utilized in subsequent patches.

Signed-off-by: Heng Qi <hengqi@linux.alibaba.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240621101353.107425-2-hengqi@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/dim.h | 7 +++++++
 lib/dim/net_dim.c   | 6 ------
 2 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/include/linux/dim.h b/include/linux/dim.h
index f343bc9aa2ec..43398f5eade2 100644
--- a/include/linux/dim.h
+++ b/include/linux/dim.h
@@ -10,6 +10,13 @@
 #include <linux/types.h>
 #include <linux/workqueue.h>
 
+/* Number of DIM profiles and period mode. */
+#define NET_DIM_PARAMS_NUM_PROFILES 5
+#define NET_DIM_DEFAULT_RX_CQ_PKTS_FROM_EQE 256
+#define NET_DIM_DEFAULT_TX_CQ_PKTS_FROM_EQE 128
+#define NET_DIM_DEF_PROFILE_CQE 1
+#define NET_DIM_DEF_PROFILE_EQE 1
+
 /*
  * Number of events between DIM iterations.
  * Causes a moderation of the algorithm run.
diff --git a/lib/dim/net_dim.c b/lib/dim/net_dim.c
index 4e32f7aaac86..67d5beb34dc3 100644
--- a/lib/dim/net_dim.c
+++ b/lib/dim/net_dim.c
@@ -11,12 +11,6 @@
  *        There are different set of profiles for RX/TX CQs.
  *        Each profile size must be of NET_DIM_PARAMS_NUM_PROFILES
  */
-#define NET_DIM_PARAMS_NUM_PROFILES 5
-#define NET_DIM_DEFAULT_RX_CQ_PKTS_FROM_EQE 256
-#define NET_DIM_DEFAULT_TX_CQ_PKTS_FROM_EQE 128
-#define NET_DIM_DEF_PROFILE_CQE 1
-#define NET_DIM_DEF_PROFILE_EQE 1
-
 #define NET_DIM_RX_EQE_PROFILES { \
 	{.usec = 1,   .pkts = NET_DIM_DEFAULT_RX_CQ_PKTS_FROM_EQE,}, \
 	{.usec = 8,   .pkts = NET_DIM_DEFAULT_RX_CQ_PKTS_FROM_EQE,}, \
-- 
cgit v1.2.3-58-ga151


From b65e697a7c9e0ae28a6255257d8b9b3271960426 Mon Sep 17 00:00:00 2001
From: Heng Qi <hengqi@linux.alibaba.com>
Date: Fri, 21 Jun 2024 18:13:50 +0800
Subject: dim: make DIMLIB dependent on NET

DIMLIB's capabilities are supplied by the dim, net_dim, and
rdma_dim objects, and dim's interfaces solely act as a base for
net_dim and rdma_dim and are not explicitly used anywhere else.
rdma_dim is utilized by the infiniband driver, while net_dim
is for network devices, excluding the soc/fsl driver.

In this patch, net_dim relies on some NET's interfaces, thus
DIMLIB needs to explicitly depend on the NET Kconfig.

The soc/fsl driver uses the functions provided by net_dim, so
it also needs to depend on NET.

Signed-off-by: Heng Qi <hengqi@linux.alibaba.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240621101353.107425-3-hengqi@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/soc/fsl/Kconfig | 2 +-
 lib/Kconfig             | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/soc/fsl/Kconfig b/drivers/soc/fsl/Kconfig
index fcec6ed83d5e..a1e0bc8c1757 100644
--- a/drivers/soc/fsl/Kconfig
+++ b/drivers/soc/fsl/Kconfig
@@ -22,7 +22,7 @@ config FSL_GUTS
 
 config FSL_MC_DPIO
         tristate "QorIQ DPAA2 DPIO driver"
-        depends on FSL_MC_BUS
+        depends on FSL_MC_BUS && NET
         select SOC_BUS
         select FSL_GUTS
         select DIMLIB
diff --git a/lib/Kconfig b/lib/Kconfig
index b0a76dff5c18..b38849af6f13 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -623,6 +623,7 @@ config SIGNATURE
 
 config DIMLIB
 	tristate
+	depends on NET
 	help
 	  Dynamic Interrupt Moderation library.
 	  Implements an algorithm for dynamically changing CQ moderation values
-- 
cgit v1.2.3-58-ga151


From f750dfe825b904164688adeb147950e0e0c4d262 Mon Sep 17 00:00:00 2001
From: Heng Qi <hengqi@linux.alibaba.com>
Date: Fri, 21 Jun 2024 18:13:51 +0800
Subject: ethtool: provide customized dim profile management

The NetDIM library, currently leveraged by an array of NICs, delivers
excellent acceleration benefits. Nevertheless, NICs vary significantly
in their dim profile list prerequisites.

Specifically, virtio-net backends may present diverse sw or hw device
implementation, making a one-size-fits-all parameter list impractical.
On Alibaba Cloud, the virtio DPU's performance under the default DIM
profile falls short of expectations, partly due to a mismatch in
parameter configuration.

I also noticed that ice/idpf/ena and other NICs have customized
profilelist or placed some restrictions on dim capabilities.

Motivated by this, I tried adding new params for "ethtool -C" that provides
a per-device control to modify and access a device's interrupt parameters.

Usage
========
The target NIC is named ethx.

Assume that ethx only declares support for rx profile setting
(with DIM_PROFILE_RX flag set in profile_flags) and supports modification
of usec and pkt fields.

1. Query the currently customized list of the device

$ ethtool -c ethx
...
rx-profile:
{.usec =   1, .pkts = 256, .comps = n/a,},
{.usec =   8, .pkts = 256, .comps = n/a,},
{.usec =  64, .pkts = 256, .comps = n/a,},
{.usec = 128, .pkts = 256, .comps = n/a,},
{.usec = 256, .pkts = 256, .comps = n/a,}
tx-profile:   n/a

2. Tune
$ ethtool -C ethx rx-profile 1,1,n_2,n,n_3,3,n_4,4,n_n,5,n
"n" means do not modify this field.
$ ethtool -c ethx
...
rx-profile:
{.usec =   1, .pkts =   1, .comps = n/a,},
{.usec =   2, .pkts = 256, .comps = n/a,},
{.usec =   3, .pkts =   3, .comps = n/a,},
{.usec =   4, .pkts =   4, .comps = n/a,},
{.usec = 256, .pkts =   5, .comps = n/a,}
tx-profile:   n/a

3. Hint
If the device does not support some type of customized dim profiles,
the corresponding "n/a" will display.

If the "n/a" field is being modified, -EOPNOTSUPP will be reported.

Signed-off-by: Heng Qi <hengqi@linux.alibaba.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240621101353.107425-4-hengqi@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/ethtool.yaml     |  31 +++
 Documentation/networking/ethtool-netlink.rst |   8 +
 Documentation/networking/net_dim.rst         |  42 +++++
 include/linux/dim.h                          |  58 ++++++
 include/linux/ethtool.h                      |   4 +-
 include/linux/netdevice.h                    |   3 +
 include/uapi/linux/ethtool_netlink.h         |  22 +++
 lib/dim/net_dim.c                            |  70 +++++++
 net/Kconfig                                  |   1 +
 net/ethtool/coalesce.c                       | 273 ++++++++++++++++++++++++++-
 10 files changed, 509 insertions(+), 3 deletions(-)

diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml
index 00dc61358be8..6c2ab3d1c22f 100644
--- a/Documentation/netlink/specs/ethtool.yaml
+++ b/Documentation/netlink/specs/ethtool.yaml
@@ -414,6 +414,26 @@ attribute-sets:
         name: combined-count
         type: u32
 
+  -
+    name: irq-moderation
+    attributes:
+      -
+        name: usec
+        type: u32
+      -
+        name: pkts
+        type: u32
+      -
+        name: comps
+        type: u32
+  -
+    name: profile
+    attributes:
+      -
+        name: irq-moderation
+        type: nest
+        multi-attr: true
+        nested-attributes: irq-moderation
   -
     name: coalesce
     attributes:
@@ -502,6 +522,15 @@ attribute-sets:
       -
         name: tx-aggr-time-usecs
         type: u32
+      -
+        name: rx-profile
+        type: nest
+        nested-attributes: profile
+      -
+        name: tx-profile
+        type: nest
+        nested-attributes: profile
+
   -
     name: pause-stat
     attributes:
@@ -1325,6 +1354,8 @@ operations:
             - tx-aggr-max-bytes
             - tx-aggr-max-frames
             - tx-aggr-time-usecs
+            - rx-profile
+            - tx-profile
       dump: *coalesce-get-op
     -
       name: coalesce-set
diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 160bfb0ae8ba..7ec08e903bab 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -1033,6 +1033,8 @@ Kernel response contents:
   ``ETHTOOL_A_COALESCE_TX_AGGR_MAX_BYTES``     u32     max aggr size, Tx
   ``ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES``    u32     max aggr packets, Tx
   ``ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS``    u32     time (us), aggr, Tx
+  ``ETHTOOL_A_COALESCE_RX_PROFILE``            nested  profile of DIM, Rx
+  ``ETHTOOL_A_COALESCE_TX_PROFILE``            nested  profile of DIM, Tx
   ===========================================  ======  =======================
 
 Attributes are only included in reply if their value is not zero or the
@@ -1062,6 +1064,10 @@ block should be sent.
 This feature is mainly of interest for specific USB devices which does not cope
 well with frequent small-sized URBs transmissions.
 
+``ETHTOOL_A_COALESCE_RX_PROFILE`` and ``ETHTOOL_A_COALESCE_TX_PROFILE`` refer
+to DIM parameters, see `Generic Network Dynamic Interrupt Moderation (Net DIM)
+<https://www.kernel.org/doc/Documentation/networking/net_dim.rst>`_.
+
 COALESCE_SET
 ============
 
@@ -1098,6 +1104,8 @@ Request contents:
   ``ETHTOOL_A_COALESCE_TX_AGGR_MAX_BYTES``     u32     max aggr size, Tx
   ``ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES``    u32     max aggr packets, Tx
   ``ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS``    u32     time (us), aggr, Tx
+  ``ETHTOOL_A_COALESCE_RX_PROFILE``            nested  profile of DIM, Rx
+  ``ETHTOOL_A_COALESCE_TX_PROFILE``            nested  profile of DIM, Tx
   ===========================================  ======  =======================
 
 Request is rejected if it attributes declared as unsupported by driver (i.e.
diff --git a/Documentation/networking/net_dim.rst b/Documentation/networking/net_dim.rst
index 3bed9fd95336..8908fd7b0a8d 100644
--- a/Documentation/networking/net_dim.rst
+++ b/Documentation/networking/net_dim.rst
@@ -169,6 +169,48 @@ usage is not complete but it should make the outline of the usage clear.
 	...
   }
 
+
+Tuning DIM
+==========
+
+Net DIM serves a range of network devices and delivers excellent acceleration
+benefits. Yet, it has been observed that some preset configurations of DIM may
+not align seamlessly with the varying specifications of network devices, and
+this discrepancy has been identified as a factor to the suboptimal performance
+outcomes of DIM-enabled network devices, related to a mismatch in profiles.
+
+To address this issue, Net DIM introduces a per-device control to modify and
+access a device's ``rx-profile`` and ``tx-profile`` parameters:
+Assume that the target network device is named ethx, and ethx only declares
+support for RX profile setting and supports modification of ``usec`` field
+and ``pkts`` field (See the data structure:
+:c:type:`struct dim_cq_moder <dim_cq_moder>`).
+
+You can use ethtool to modify the current RX DIM profile where all
+values are 64::
+
+    $ ethtool -C ethx rx-profile 1,1,n_2,2,n_3,n,n_n,4,n_n,n,n
+
+``n`` means do not modify this field, and ``_`` separates structure
+elements of the profile array.
+
+Querying the current profiles using::
+
+    $ ethtool -c ethx
+    ...
+    rx-profile:
+    {.usec =   1, .pkts =   1, .comps = n/a,},
+    {.usec =   2, .pkts =   2, .comps = n/a,},
+    {.usec =   3, .pkts =  64, .comps = n/a,},
+    {.usec =  64, .pkts =   4, .comps = n/a,},
+    {.usec =  64, .pkts =  64, .comps = n/a,}
+    tx-profile:   n/a
+
+If the network device does not support specific fields of DIM profiles,
+the corresponding ``n/a`` will display. If the ``n/a`` field is being
+modified, error messages will be reported.
+
+
 Dynamic Interrupt Moderation (DIM) library API
 ==============================================
 
diff --git a/include/linux/dim.h b/include/linux/dim.h
index 43398f5eade2..e0f39bd85432 100644
--- a/include/linux/dim.h
+++ b/include/linux/dim.h
@@ -10,6 +10,8 @@
 #include <linux/types.h>
 #include <linux/workqueue.h>
 
+struct net_device;
+
 /* Number of DIM profiles and period mode. */
 #define NET_DIM_PARAMS_NUM_PROFILES 5
 #define NET_DIM_DEFAULT_RX_CQ_PKTS_FROM_EQE 256
@@ -45,12 +47,45 @@
  * @pkts: CQ packet counter suggestion (by DIM)
  * @comps: Completion counter
  * @cq_period_mode: CQ period count mode (from CQE/EQE)
+ * @rcu: for asynchronous kfree_rcu
  */
 struct dim_cq_moder {
 	u16 usec;
 	u16 pkts;
 	u16 comps;
 	u8 cq_period_mode;
+	struct rcu_head rcu;
+};
+
+#define DIM_PROFILE_RX		BIT(0)	/* support rx profile modification */
+#define DIM_PROFILE_TX		BIT(1)	/* support tx profile modification */
+
+#define DIM_COALESCE_USEC	BIT(0)	/* support usec field modification */
+#define DIM_COALESCE_PKTS	BIT(1)	/* support pkts field modification */
+#define DIM_COALESCE_COMPS	BIT(2)	/* support comps field modification */
+
+/**
+ * struct dim_irq_moder - Structure for irq moderation information.
+ * Used to collect irq moderation related information.
+ *
+ * @profile_flags: DIM_PROFILE_*
+ * @coal_flags: DIM_COALESCE_* for Rx and Tx
+ * @dim_rx_mode: Rx DIM period count mode: CQE or EQE
+ * @dim_tx_mode: Tx DIM period count mode: CQE or EQE
+ * @rx_profile: DIM profile list for Rx
+ * @tx_profile: DIM profile list for Tx
+ * @rx_dim_work: Rx DIM worker scheduled by net_dim()
+ * @tx_dim_work: Tx DIM worker scheduled by net_dim()
+ */
+struct dim_irq_moder {
+	u8 profile_flags;
+	u8 coal_flags;
+	u8 dim_rx_mode;
+	u8 dim_tx_mode;
+	struct dim_cq_moder __rcu *rx_profile;
+	struct dim_cq_moder __rcu *tx_profile;
+	void (*rx_dim_work)(struct work_struct *work);
+	void (*tx_dim_work)(struct work_struct *work);
 };
 
 /**
@@ -198,6 +233,29 @@ enum dim_step_result {
 	DIM_ON_EDGE,
 };
 
+/**
+ * net_dim_init_irq_moder - collect information to initialize irq moderation
+ * @dev: target network device
+ * @profile_flags: Rx or Tx profile modification capability
+ * @coal_flags: irq moderation params flags
+ * @rx_mode: CQ period mode for Rx
+ * @tx_mode: CQ period mode for Tx
+ * @rx_dim_work: Rx worker called after dim decision
+ * @tx_dim_work: Tx worker called after dim decision
+ *
+ * Return: 0 on success or a negative error code.
+ */
+int net_dim_init_irq_moder(struct net_device *dev, u8 profile_flags,
+			   u8 coal_flags, u8 rx_mode, u8 tx_mode,
+			   void (*rx_dim_work)(struct work_struct *work),
+			   void (*tx_dim_work)(struct work_struct *work));
+
+/**
+ * net_dim_free_irq_moder - free fields for irq moderation
+ * @dev: target network device
+ */
+void net_dim_free_irq_moder(struct net_device *dev);
+
 /**
  *	dim_on_top - check if current state is a good place to stop (top location)
  *	@dim: DIM context
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 6fd9107d3cc0..959196af7f5a 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -284,7 +284,9 @@ bool ethtool_convert_link_mode_to_legacy_u32(u32 *legacy_u32,
 #define ETHTOOL_COALESCE_TX_AGGR_MAX_BYTES	BIT(24)
 #define ETHTOOL_COALESCE_TX_AGGR_MAX_FRAMES	BIT(25)
 #define ETHTOOL_COALESCE_TX_AGGR_TIME_USECS	BIT(26)
-#define ETHTOOL_COALESCE_ALL_PARAMS		GENMASK(26, 0)
+#define ETHTOOL_COALESCE_RX_PROFILE		BIT(27)
+#define ETHTOOL_COALESCE_TX_PROFILE		BIT(28)
+#define ETHTOOL_COALESCE_ALL_PARAMS		GENMASK(28, 0)
 
 #define ETHTOOL_COALESCE_USECS						\
 	(ETHTOOL_COALESCE_RX_USECS | ETHTOOL_COALESCE_TX_USECS)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 4e81660b4462..cc18acd3c58b 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2402,6 +2402,9 @@ struct net_device {
 	/** @page_pools: page pools created for this netdevice */
 	struct hlist_head	page_pools;
 #endif
+
+	/** @irq_moder: dim parameters used if IS_ENABLED(CONFIG_DIMLIB). */
+	struct dim_irq_moder	*irq_moder;
 };
 #define to_net_dev(d) container_of(d, struct net_device, dev)
 
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index b49b804b9495..d15856c7e001 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -415,12 +415,34 @@ enum {
 	ETHTOOL_A_COALESCE_TX_AGGR_MAX_BYTES,		/* u32 */
 	ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES,		/* u32 */
 	ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS,		/* u32 */
+	/* nest - _A_PROFILE_IRQ_MODERATION */
+	ETHTOOL_A_COALESCE_RX_PROFILE,
+	/* nest - _A_PROFILE_IRQ_MODERATION */
+	ETHTOOL_A_COALESCE_TX_PROFILE,
 
 	/* add new constants above here */
 	__ETHTOOL_A_COALESCE_CNT,
 	ETHTOOL_A_COALESCE_MAX = (__ETHTOOL_A_COALESCE_CNT - 1)
 };
 
+enum {
+	ETHTOOL_A_PROFILE_UNSPEC,
+	/* nest, _A_IRQ_MODERATION_* */
+	ETHTOOL_A_PROFILE_IRQ_MODERATION,
+	__ETHTOOL_A_PROFILE_CNT,
+	ETHTOOL_A_PROFILE_MAX = (__ETHTOOL_A_PROFILE_CNT - 1)
+};
+
+enum {
+	ETHTOOL_A_IRQ_MODERATION_UNSPEC,
+	ETHTOOL_A_IRQ_MODERATION_USEC,			/* u32 */
+	ETHTOOL_A_IRQ_MODERATION_PKTS,			/* u32 */
+	ETHTOOL_A_IRQ_MODERATION_COMPS,			/* u32 */
+
+	__ETHTOOL_A_IRQ_MODERATION_CNT,
+	ETHTOOL_A_IRQ_MODERATION_MAX = (__ETHTOOL_A_IRQ_MODERATION_CNT - 1)
+};
+
 /* PAUSE */
 
 enum {
diff --git a/lib/dim/net_dim.c b/lib/dim/net_dim.c
index 67d5beb34dc3..0cd41277c7a3 100644
--- a/lib/dim/net_dim.c
+++ b/lib/dim/net_dim.c
@@ -4,6 +4,7 @@
  */
 
 #include <linux/dim.h>
+#include <linux/rtnetlink.h>
 
 /*
  * Net DIM profiles:
@@ -95,6 +96,75 @@ net_dim_get_def_tx_moderation(u8 cq_period_mode)
 }
 EXPORT_SYMBOL(net_dim_get_def_tx_moderation);
 
+int net_dim_init_irq_moder(struct net_device *dev, u8 profile_flags,
+			   u8 coal_flags, u8 rx_mode, u8 tx_mode,
+			   void (*rx_dim_work)(struct work_struct *work),
+			   void (*tx_dim_work)(struct work_struct *work))
+{
+	struct dim_cq_moder *rxp = NULL, *txp;
+	struct dim_irq_moder *moder;
+	int len;
+
+	dev->irq_moder = kzalloc(sizeof(*dev->irq_moder), GFP_KERNEL);
+	if (!dev->irq_moder)
+		return -ENOMEM;
+
+	moder = dev->irq_moder;
+	len = NET_DIM_PARAMS_NUM_PROFILES * sizeof(*moder->rx_profile);
+
+	moder->coal_flags = coal_flags;
+	moder->profile_flags = profile_flags;
+
+	if (profile_flags & DIM_PROFILE_RX) {
+		moder->rx_dim_work = rx_dim_work;
+		moder->dim_rx_mode = rx_mode;
+		rxp = kmemdup(rx_profile[rx_mode], len, GFP_KERNEL);
+		if (!rxp)
+			goto free_moder;
+
+		rcu_assign_pointer(moder->rx_profile, rxp);
+	}
+
+	if (profile_flags & DIM_PROFILE_TX) {
+		moder->tx_dim_work = tx_dim_work;
+		moder->dim_tx_mode = tx_mode;
+		txp = kmemdup(tx_profile[tx_mode], len, GFP_KERNEL);
+		if (!txp)
+			goto free_rxp;
+
+		rcu_assign_pointer(moder->tx_profile, txp);
+	}
+
+	return 0;
+
+free_rxp:
+	kfree(rxp);
+free_moder:
+	kfree(moder);
+	return -ENOMEM;
+}
+EXPORT_SYMBOL(net_dim_init_irq_moder);
+
+/* RTNL lock is held. */
+void net_dim_free_irq_moder(struct net_device *dev)
+{
+	struct dim_cq_moder *rxp, *txp;
+
+	if (!dev->irq_moder)
+		return;
+
+	rxp = rtnl_dereference(dev->irq_moder->rx_profile);
+	txp = rtnl_dereference(dev->irq_moder->tx_profile);
+
+	rcu_assign_pointer(dev->irq_moder->rx_profile, NULL);
+	rcu_assign_pointer(dev->irq_moder->tx_profile, NULL);
+
+	kfree_rcu(rxp, rcu);
+	kfree_rcu(txp, rcu);
+	kfree(dev->irq_moder);
+}
+EXPORT_SYMBOL(net_dim_free_irq_moder);
+
 static int net_dim_step(struct dim *dim)
 {
 	if (dim->tired == (NET_DIM_PARAMS_NUM_PROFILES * 2))
diff --git a/net/Kconfig b/net/Kconfig
index 9fe65fa26e48..d27d0deac0bf 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -508,6 +508,7 @@ config FAILOVER
 
 config ETHTOOL_NETLINK
 	bool "Netlink interface for ethtool"
+	select DIMLIB
 	default y
 	help
 	  An alternative userspace interface for ethtool based on generic
diff --git a/net/ethtool/coalesce.c b/net/ethtool/coalesce.c
index 83112c1a71ae..759b16e3d134 100644
--- a/net/ethtool/coalesce.c
+++ b/net/ethtool/coalesce.c
@@ -1,5 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0-only
 
+#include <linux/dim.h>
 #include "netlink.h"
 #include "common.h"
 
@@ -82,6 +83,14 @@ static int coalesce_prepare_data(const struct ethnl_req_info *req_base,
 static int coalesce_reply_size(const struct ethnl_req_info *req_base,
 			       const struct ethnl_reply_data *reply_base)
 {
+	int modersz = nla_total_size(0) + /* _PROFILE_IRQ_MODERATION, nest */
+		      nla_total_size(sizeof(u32)) + /* _IRQ_MODERATION_USEC */
+		      nla_total_size(sizeof(u32)) + /* _IRQ_MODERATION_PKTS */
+		      nla_total_size(sizeof(u32));  /* _IRQ_MODERATION_COMPS */
+
+	int total_modersz = nla_total_size(0) +  /* _{R,T}X_PROFILE, nest */
+			modersz * NET_DIM_PARAMS_NUM_PROFILES;
+
 	return nla_total_size(sizeof(u32)) +	/* _RX_USECS */
 	       nla_total_size(sizeof(u32)) +	/* _RX_MAX_FRAMES */
 	       nla_total_size(sizeof(u32)) +	/* _RX_USECS_IRQ */
@@ -108,7 +117,8 @@ static int coalesce_reply_size(const struct ethnl_req_info *req_base,
 	       nla_total_size(sizeof(u8)) +	/* _USE_CQE_MODE_RX */
 	       nla_total_size(sizeof(u32)) +	/* _TX_AGGR_MAX_BYTES */
 	       nla_total_size(sizeof(u32)) +	/* _TX_AGGR_MAX_FRAMES */
-	       nla_total_size(sizeof(u32));	/* _TX_AGGR_TIME_USECS */
+	       nla_total_size(sizeof(u32)) +	/* _TX_AGGR_TIME_USECS */
+	       total_modersz * 2;		/* _{R,T}X_PROFILE */
 }
 
 static bool coalesce_put_u32(struct sk_buff *skb, u16 attr_type, u32 val,
@@ -127,14 +137,84 @@ static bool coalesce_put_bool(struct sk_buff *skb, u16 attr_type, u32 val,
 	return nla_put_u8(skb, attr_type, !!val);
 }
 
+/**
+ * coalesce_put_profile - fill reply with a nla nest with four child nla nests.
+ * @skb: socket buffer the message is stored in
+ * @attr_type: nest attr type ETHTOOL_A_COALESCE_*X_PROFILE
+ * @profile: data passed to userspace
+ * @coal_flags: modifiable parameters supported by the driver
+ *
+ * Put a dim profile nest attribute. Refer to ETHTOOL_A_PROFILE_IRQ_MODERATION.
+ *
+ * Return: 0 on success or a negative error code.
+ */
+static int coalesce_put_profile(struct sk_buff *skb, u16 attr_type,
+				const struct dim_cq_moder *profile,
+				u8 coal_flags)
+{
+	struct nlattr *profile_attr, *moder_attr;
+	int i, ret;
+
+	if (!profile || !coal_flags)
+		return 0;
+
+	profile_attr = nla_nest_start(skb, attr_type);
+	if (!profile_attr)
+		return -EMSGSIZE;
+
+	for (i = 0; i < NET_DIM_PARAMS_NUM_PROFILES; i++) {
+		moder_attr = nla_nest_start(skb,
+					    ETHTOOL_A_PROFILE_IRQ_MODERATION);
+		if (!moder_attr) {
+			ret = -EMSGSIZE;
+			goto cancel_profile;
+		}
+
+		if (coal_flags & DIM_COALESCE_USEC) {
+			ret = nla_put_u32(skb, ETHTOOL_A_IRQ_MODERATION_USEC,
+					  profile[i].usec);
+			if (ret)
+				goto cancel_moder;
+		}
+
+		if (coal_flags & DIM_COALESCE_PKTS) {
+			ret = nla_put_u32(skb, ETHTOOL_A_IRQ_MODERATION_PKTS,
+					  profile[i].pkts);
+			if (ret)
+				goto cancel_moder;
+		}
+
+		if (coal_flags & DIM_COALESCE_COMPS) {
+			ret = nla_put_u32(skb, ETHTOOL_A_IRQ_MODERATION_COMPS,
+					  profile[i].comps);
+			if (ret)
+				goto cancel_moder;
+		}
+
+		nla_nest_end(skb, moder_attr);
+	}
+
+	nla_nest_end(skb, profile_attr);
+
+	return 0;
+
+cancel_moder:
+	nla_nest_cancel(skb, moder_attr);
+cancel_profile:
+	nla_nest_cancel(skb, profile_attr);
+	return ret;
+}
+
 static int coalesce_fill_reply(struct sk_buff *skb,
 			       const struct ethnl_req_info *req_base,
 			       const struct ethnl_reply_data *reply_base)
 {
 	const struct coalesce_reply_data *data = COALESCE_REPDATA(reply_base);
 	const struct kernel_ethtool_coalesce *kcoal = &data->kernel_coalesce;
+	struct dim_irq_moder *moder = req_base->dev->irq_moder;
 	const struct ethtool_coalesce *coal = &data->coalesce;
 	u32 supported = data->supported_params;
+	int ret = 0;
 
 	if (coalesce_put_u32(skb, ETHTOOL_A_COALESCE_RX_USECS,
 			     coal->rx_coalesce_usecs, supported) ||
@@ -192,11 +272,41 @@ static int coalesce_fill_reply(struct sk_buff *skb,
 			     kcoal->tx_aggr_time_usecs, supported))
 		return -EMSGSIZE;
 
-	return 0;
+	if (!moder)
+		return 0;
+
+	rcu_read_lock();
+	if (moder->profile_flags & DIM_PROFILE_RX) {
+		ret = coalesce_put_profile(skb, ETHTOOL_A_COALESCE_RX_PROFILE,
+					   rcu_dereference(moder->rx_profile),
+					   moder->coal_flags);
+		if (ret)
+			goto out;
+	}
+
+	if (moder->profile_flags & DIM_PROFILE_TX)
+		ret = coalesce_put_profile(skb, ETHTOOL_A_COALESCE_TX_PROFILE,
+					   rcu_dereference(moder->tx_profile),
+					   moder->coal_flags);
+
+out:
+	rcu_read_unlock();
+	return ret;
 }
 
 /* COALESCE_SET */
 
+static const struct nla_policy coalesce_irq_moderation_policy[] = {
+	[ETHTOOL_A_IRQ_MODERATION_USEC]	= { .type = NLA_U32 },
+	[ETHTOOL_A_IRQ_MODERATION_PKTS]	= { .type = NLA_U32 },
+	[ETHTOOL_A_IRQ_MODERATION_COMPS] = { .type = NLA_U32 },
+};
+
+static const struct nla_policy coalesce_profile_policy[] = {
+	[ETHTOOL_A_PROFILE_IRQ_MODERATION] =
+		NLA_POLICY_NESTED(coalesce_irq_moderation_policy),
+};
+
 const struct nla_policy ethnl_coalesce_set_policy[] = {
 	[ETHTOOL_A_COALESCE_HEADER]		=
 		NLA_POLICY_NESTED(ethnl_header_policy),
@@ -227,6 +337,10 @@ const struct nla_policy ethnl_coalesce_set_policy[] = {
 	[ETHTOOL_A_COALESCE_TX_AGGR_MAX_BYTES] = { .type = NLA_U32 },
 	[ETHTOOL_A_COALESCE_TX_AGGR_MAX_FRAMES] = { .type = NLA_U32 },
 	[ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS] = { .type = NLA_U32 },
+	[ETHTOOL_A_COALESCE_RX_PROFILE] =
+		NLA_POLICY_NESTED(coalesce_profile_policy),
+	[ETHTOOL_A_COALESCE_TX_PROFILE] =
+		NLA_POLICY_NESTED(coalesce_profile_policy),
 };
 
 static int
@@ -234,6 +348,7 @@ ethnl_set_coalesce_validate(struct ethnl_req_info *req_info,
 			    struct genl_info *info)
 {
 	const struct ethtool_ops *ops = req_info->dev->ethtool_ops;
+	struct dim_irq_moder *irq_moder = req_info->dev->irq_moder;
 	struct nlattr **tb = info->attrs;
 	u32 supported_params;
 	u16 a;
@@ -243,6 +358,12 @@ ethnl_set_coalesce_validate(struct ethnl_req_info *req_info,
 
 	/* make sure that only supported parameters are present */
 	supported_params = ops->supported_coalesce_params;
+	if (irq_moder && irq_moder->profile_flags & DIM_PROFILE_RX)
+		supported_params |= ETHTOOL_COALESCE_RX_PROFILE;
+
+	if (irq_moder && irq_moder->profile_flags & DIM_PROFILE_TX)
+		supported_params |= ETHTOOL_COALESCE_TX_PROFILE;
+
 	for (a = ETHTOOL_A_COALESCE_RX_USECS; a < __ETHTOOL_A_COALESCE_CNT; a++)
 		if (tb[a] && !(supported_params & attr_to_mask(a))) {
 			NL_SET_ERR_MSG_ATTR(info->extack, tb[a],
@@ -253,6 +374,138 @@ ethnl_set_coalesce_validate(struct ethnl_req_info *req_info,
 	return 1;
 }
 
+/**
+ * ethnl_update_irq_moder - update a specific field in the given profile
+ * @irq_moder: place that collects dim related information
+ * @irq_field: field in profile to modify
+ * @attr_type: attr type ETHTOOL_A_IRQ_MODERATION_*
+ * @tb: netlink attribute with new values or null
+ * @coal_bit: DIM_COALESCE_* bit from coal_flags
+ * @mod: pointer to bool for modification tracking
+ * @extack: netlink extended ack
+ *
+ * Return: 0 on success or a negative error code.
+ */
+static int ethnl_update_irq_moder(struct dim_irq_moder *irq_moder,
+				  u16 *irq_field, u16 attr_type,
+				  struct nlattr **tb,
+				  u8 coal_bit, bool *mod,
+				  struct netlink_ext_ack *extack)
+{
+	int ret = 0;
+	u32 val;
+
+	if (!tb[attr_type])
+		return 0;
+
+	if (irq_moder->coal_flags & coal_bit) {
+		val = nla_get_u32(tb[attr_type]);
+		if (*irq_field == val)
+			return 0;
+
+		*irq_field = val;
+		*mod = true;
+	} else {
+		NL_SET_BAD_ATTR(extack, tb[attr_type]);
+		ret = -EOPNOTSUPP;
+	}
+
+	return ret;
+}
+
+/**
+ * ethnl_update_profile - get a profile nest with child nests from userspace.
+ * @dev: netdevice to update the profile
+ * @dst: profile get from the driver and modified by ethnl_update_profile.
+ * @nests: nest attr ETHTOOL_A_COALESCE_*X_PROFILE to set profile.
+ * @mod: pointer to bool for modification tracking
+ * @extack: Netlink extended ack
+ *
+ * Layout of nests:
+ *   Nested ETHTOOL_A_COALESCE_*X_PROFILE attr
+ *     Nested ETHTOOL_A_PROFILE_IRQ_MODERATION attr
+ *       ETHTOOL_A_IRQ_MODERATION_USEC attr
+ *       ETHTOOL_A_IRQ_MODERATION_PKTS attr
+ *       ETHTOOL_A_IRQ_MODERATION_COMPS attr
+ *     ...
+ *     Nested ETHTOOL_A_PROFILE_IRQ_MODERATION attr
+ *       ETHTOOL_A_IRQ_MODERATION_USEC attr
+ *       ETHTOOL_A_IRQ_MODERATION_PKTS attr
+ *       ETHTOOL_A_IRQ_MODERATION_COMPS attr
+ *
+ * Return: 0 on success or a negative error code.
+ */
+static int ethnl_update_profile(struct net_device *dev,
+				struct dim_cq_moder __rcu **dst,
+				const struct nlattr *nests,
+				bool *mod,
+				struct netlink_ext_ack *extack)
+{
+	int len_irq_moder = ARRAY_SIZE(coalesce_irq_moderation_policy);
+	struct nlattr *tb[ARRAY_SIZE(coalesce_irq_moderation_policy)];
+	struct dim_irq_moder *irq_moder = dev->irq_moder;
+	struct dim_cq_moder *new_profile, *old_profile;
+	int ret, rem, i = 0, len;
+	struct nlattr *nest;
+
+	if (!nests)
+		return 0;
+
+	if (!*dst)
+		return -EOPNOTSUPP;
+
+	old_profile = rtnl_dereference(*dst);
+	len = NET_DIM_PARAMS_NUM_PROFILES * sizeof(*old_profile);
+	new_profile = kmemdup(old_profile, len, GFP_KERNEL);
+	if (!new_profile)
+		return -ENOMEM;
+
+	nla_for_each_nested_type(nest, ETHTOOL_A_PROFILE_IRQ_MODERATION,
+				 nests, rem) {
+		ret = nla_parse_nested(tb, len_irq_moder - 1, nest,
+				       coalesce_irq_moderation_policy,
+				       extack);
+		if (ret)
+			goto err_out;
+
+		ret = ethnl_update_irq_moder(irq_moder, &new_profile[i].usec,
+					     ETHTOOL_A_IRQ_MODERATION_USEC,
+					     tb, DIM_COALESCE_USEC,
+					     mod, extack);
+		if (ret)
+			goto err_out;
+
+		ret = ethnl_update_irq_moder(irq_moder, &new_profile[i].pkts,
+					     ETHTOOL_A_IRQ_MODERATION_PKTS,
+					     tb, DIM_COALESCE_PKTS,
+					     mod, extack);
+		if (ret)
+			goto err_out;
+
+		ret = ethnl_update_irq_moder(irq_moder, &new_profile[i].comps,
+					     ETHTOOL_A_IRQ_MODERATION_COMPS,
+					     tb, DIM_COALESCE_COMPS,
+					     mod, extack);
+		if (ret)
+			goto err_out;
+
+		i++;
+	}
+
+	/* After the profile is modified, dim itself is a dynamic
+	 * mechanism and will quickly fit to the appropriate
+	 * coalescing parameters according to the new profile.
+	 */
+	rcu_assign_pointer(*dst, new_profile);
+	kfree_rcu(old_profile, rcu);
+
+	return 0;
+
+err_out:
+	kfree(new_profile);
+	return ret;
+}
+
 static int
 __ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info,
 		     bool *dual_change)
@@ -317,6 +570,22 @@ __ethnl_set_coalesce(struct ethnl_req_info *req_info, struct genl_info *info,
 	ethnl_update_u32(&kernel_coalesce.tx_aggr_time_usecs,
 			 tb[ETHTOOL_A_COALESCE_TX_AGGR_TIME_USECS], &mod);
 
+	if (dev->irq_moder && dev->irq_moder->profile_flags & DIM_PROFILE_RX) {
+		ret = ethnl_update_profile(dev, &dev->irq_moder->rx_profile,
+					   tb[ETHTOOL_A_COALESCE_RX_PROFILE],
+					   &mod, info->extack);
+		if (ret < 0)
+			return ret;
+	}
+
+	if (dev->irq_moder && dev->irq_moder->profile_flags & DIM_PROFILE_TX) {
+		ret = ethnl_update_profile(dev, &dev->irq_moder->tx_profile,
+					   tb[ETHTOOL_A_COALESCE_TX_PROFILE],
+					   &mod, info->extack);
+		if (ret < 0)
+			return ret;
+	}
+
 	/* Update operation modes */
 	ethnl_update_bool32(&coalesce.use_adaptive_rx_coalesce,
 			    tb[ETHTOOL_A_COALESCE_USE_ADAPTIVE_RX], &mod_mode);
-- 
cgit v1.2.3-58-ga151


From 13ba28c5cd047e272f9dbbeb5c62c403873d8987 Mon Sep 17 00:00:00 2001
From: Heng Qi <hengqi@linux.alibaba.com>
Date: Fri, 21 Jun 2024 18:13:52 +0800
Subject: dim: add new interfaces for initialization and getting results

DIM-related mode and work have been collected in one same place,
so new interfaces are added to provide convenience.

Signed-off-by: Heng Qi <hengqi@linux.alibaba.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240621101353.107425-5-hengqi@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/dim.h | 48 +++++++++++++++++++++++++++++++++++++
 lib/dim/net_dim.c   | 68 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 116 insertions(+)

diff --git a/include/linux/dim.h b/include/linux/dim.h
index e0f39bd85432..1b581ff25a15 100644
--- a/include/linux/dim.h
+++ b/include/linux/dim.h
@@ -256,6 +256,54 @@ int net_dim_init_irq_moder(struct net_device *dev, u8 profile_flags,
  */
 void net_dim_free_irq_moder(struct net_device *dev);
 
+/**
+ * net_dim_setting - initialize DIM's cq mode and schedule worker
+ * @dev: target network device
+ * @dim: DIM context
+ * @is_tx: true indicates the tx direction, false indicates the rx direction
+ */
+void net_dim_setting(struct net_device *dev, struct dim *dim, bool is_tx);
+
+/**
+ * net_dim_work_cancel - synchronously cancel dim's worker
+ * @dim: DIM context
+ */
+void net_dim_work_cancel(struct dim *dim);
+
+/**
+ * net_dim_get_rx_irq_moder - get DIM rx results based on profile_ix
+ * @dev: target network device
+ * @dim: DIM context
+ *
+ * Return: DIM irq moderation
+ */
+struct dim_cq_moder
+net_dim_get_rx_irq_moder(struct net_device *dev, struct dim *dim);
+
+/**
+ * net_dim_get_tx_irq_moder - get DIM tx results based on profile_ix
+ * @dev: target network device
+ * @dim: DIM context
+ *
+ * Return: DIM irq moderation
+ */
+struct dim_cq_moder
+net_dim_get_tx_irq_moder(struct net_device *dev, struct dim *dim);
+
+/**
+ * net_dim_set_rx_mode - set DIM rx cq mode
+ * @dev: target network device
+ * @rx_mode: target rx cq mode
+ */
+void net_dim_set_rx_mode(struct net_device *dev, u8 rx_mode);
+
+/**
+ * net_dim_set_tx_mode - set DIM tx cq mode
+ * @dev: target network device
+ * @tx_mode: target tx cq mode
+ */
+void net_dim_set_tx_mode(struct net_device *dev, u8 tx_mode);
+
 /**
  *	dim_on_top - check if current state is a good place to stop (top location)
  *	@dim: DIM context
diff --git a/lib/dim/net_dim.c b/lib/dim/net_dim.c
index 0cd41277c7a3..d7e7028e9b19 100644
--- a/lib/dim/net_dim.c
+++ b/lib/dim/net_dim.c
@@ -165,6 +165,74 @@ void net_dim_free_irq_moder(struct net_device *dev)
 }
 EXPORT_SYMBOL(net_dim_free_irq_moder);
 
+void net_dim_setting(struct net_device *dev, struct dim *dim, bool is_tx)
+{
+	struct dim_irq_moder *irq_moder = dev->irq_moder;
+
+	if (!irq_moder)
+		return;
+
+	if (is_tx) {
+		INIT_WORK(&dim->work, irq_moder->tx_dim_work);
+		dim->mode = READ_ONCE(irq_moder->dim_tx_mode);
+		return;
+	}
+
+	INIT_WORK(&dim->work, irq_moder->rx_dim_work);
+	dim->mode = READ_ONCE(irq_moder->dim_rx_mode);
+}
+EXPORT_SYMBOL(net_dim_setting);
+
+void net_dim_work_cancel(struct dim *dim)
+{
+	cancel_work_sync(&dim->work);
+}
+EXPORT_SYMBOL(net_dim_work_cancel);
+
+struct dim_cq_moder net_dim_get_rx_irq_moder(struct net_device *dev,
+					     struct dim *dim)
+{
+	struct dim_cq_moder res, *profile;
+
+	rcu_read_lock();
+	profile = rcu_dereference(dev->irq_moder->rx_profile);
+	res = profile[dim->profile_ix];
+	rcu_read_unlock();
+
+	res.cq_period_mode = dim->mode;
+
+	return res;
+}
+EXPORT_SYMBOL(net_dim_get_rx_irq_moder);
+
+struct dim_cq_moder net_dim_get_tx_irq_moder(struct net_device *dev,
+					     struct dim *dim)
+{
+	struct dim_cq_moder res, *profile;
+
+	rcu_read_lock();
+	profile = rcu_dereference(dev->irq_moder->tx_profile);
+	res = profile[dim->profile_ix];
+	rcu_read_unlock();
+
+	res.cq_period_mode = dim->mode;
+
+	return res;
+}
+EXPORT_SYMBOL(net_dim_get_tx_irq_moder);
+
+void net_dim_set_rx_mode(struct net_device *dev, u8 rx_mode)
+{
+	WRITE_ONCE(dev->irq_moder->dim_rx_mode, rx_mode);
+}
+EXPORT_SYMBOL(net_dim_set_rx_mode);
+
+void net_dim_set_tx_mode(struct net_device *dev, u8 tx_mode)
+{
+	WRITE_ONCE(dev->irq_moder->dim_tx_mode, tx_mode);
+}
+EXPORT_SYMBOL(net_dim_set_tx_mode);
+
 static int net_dim_step(struct dim *dim)
 {
 	if (dim->tired == (NET_DIM_PARAMS_NUM_PROFILES * 2))
-- 
cgit v1.2.3-58-ga151


From dcb67f6a9ead650360a60009ba8a68be385c2bde Mon Sep 17 00:00:00 2001
From: Heng Qi <hengqi@linux.alibaba.com>
Date: Fri, 21 Jun 2024 18:13:53 +0800
Subject: virtio-net: support dim profile fine-tuning

Virtio-net has different types of back-end device implementations.
In order to effectively optimize the dim library's gains for different
device implementations, let's use the new interface params to
initialize and query dim results from a customized profile list.

Signed-off-by: Heng Qi <hengqi@linux.alibaba.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240621101353.107425-6-hengqi@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 54 +++++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 47 insertions(+), 7 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index f21297751310..0b4747e81464 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -2469,6 +2469,13 @@ err_xdp_reg_mem_model:
 	return err;
 }
 
+static void virtnet_cancel_dim(struct virtnet_info *vi, struct dim *dim)
+{
+	if (!virtio_has_feature(vi->vdev, VIRTIO_NET_F_VQ_NOTF_COAL))
+		return;
+	net_dim_work_cancel(dim);
+}
+
 static int virtnet_open(struct net_device *dev)
 {
 	struct virtnet_info *vi = netdev_priv(dev);
@@ -2495,7 +2502,7 @@ err_enable_qp:
 
 	for (i--; i >= 0; i--) {
 		virtnet_disable_queue_pair(vi, i);
-		cancel_work_sync(&vi->rq[i].dim.work);
+		virtnet_cancel_dim(vi, &vi->rq[i].dim);
 	}
 
 	return err;
@@ -2671,7 +2678,7 @@ static int virtnet_rx_resize(struct virtnet_info *vi,
 
 	if (running) {
 		napi_disable(&rq->napi);
-		cancel_work_sync(&rq->dim.work);
+		virtnet_cancel_dim(vi, &rq->dim);
 	}
 
 	err = virtqueue_resize(rq->vq, ring_num, virtnet_rq_unmap_free_buf);
@@ -2932,7 +2939,7 @@ static int virtnet_close(struct net_device *dev)
 
 	for (i = 0; i < vi->max_queue_pairs; i++) {
 		virtnet_disable_queue_pair(vi, i);
-		cancel_work_sync(&vi->rq[i].dim.work);
+		virtnet_cancel_dim(vi, &vi->rq[i].dim);
 	}
 
 	return 0;
@@ -4458,7 +4465,7 @@ static void virtnet_rx_dim_work(struct work_struct *work)
 	if (!rq->dim_enabled)
 		goto out;
 
-	update_moder = net_dim_get_rx_moderation(dim->mode, dim->profile_ix);
+	update_moder = net_dim_get_rx_irq_moder(dev, dim);
 	if (update_moder.usec != rq->intr_coal.max_usecs ||
 	    update_moder.pkts != rq->intr_coal.max_packets) {
 		err = virtnet_send_rx_ctrl_coal_vq_cmd(vi, qnum,
@@ -5158,6 +5165,36 @@ static void virtnet_tx_timeout(struct net_device *dev, unsigned int txqueue)
 		   jiffies_to_usecs(jiffies - READ_ONCE(txq->trans_start)));
 }
 
+static int virtnet_init_irq_moder(struct virtnet_info *vi)
+{
+	u8 profile_flags = 0, coal_flags = 0;
+	int ret, i;
+
+	profile_flags |= DIM_PROFILE_RX;
+	coal_flags |= DIM_COALESCE_USEC | DIM_COALESCE_PKTS;
+	ret = net_dim_init_irq_moder(vi->dev, profile_flags, coal_flags,
+				     DIM_CQ_PERIOD_MODE_START_FROM_EQE,
+				     0, virtnet_rx_dim_work, NULL);
+
+	if (ret)
+		return ret;
+
+	for (i = 0; i < vi->max_queue_pairs; i++)
+		net_dim_setting(vi->dev, &vi->rq[i].dim, false);
+
+	return 0;
+}
+
+static void virtnet_free_irq_moder(struct virtnet_info *vi)
+{
+	if (!virtio_has_feature(vi->vdev, VIRTIO_NET_F_VQ_NOTF_COAL))
+		return;
+
+	rtnl_lock();
+	net_dim_free_irq_moder(vi->dev);
+	rtnl_unlock();
+}
+
 static const struct net_device_ops virtnet_netdev = {
 	.ndo_open            = virtnet_open,
 	.ndo_stop   	     = virtnet_close,
@@ -5437,9 +5474,6 @@ static int virtnet_alloc_queues(struct virtnet_info *vi)
 					 virtnet_poll_tx,
 					 napi_tx ? napi_weight : 0);
 
-		INIT_WORK(&vi->rq[i].dim.work, virtnet_rx_dim_work);
-		vi->rq[i].dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE;
-
 		sg_init_table(vi->rq[i].sg, ARRAY_SIZE(vi->rq[i].sg));
 		ewma_pkt_len_init(&vi->rq[i].mrg_avg_pkt_len);
 		sg_init_table(vi->sq[i].sg, ARRAY_SIZE(vi->sq[i].sg));
@@ -5868,6 +5902,10 @@ static int virtnet_probe(struct virtio_device *vdev)
 		for (i = 0; i < vi->max_queue_pairs; i++)
 			if (vi->sq[i].napi.weight)
 				vi->sq[i].intr_coal.max_packets = 1;
+
+		err = virtnet_init_irq_moder(vi);
+		if (err)
+			goto free;
 	}
 
 #ifdef CONFIG_SYSFS
@@ -6019,6 +6057,8 @@ static void virtnet_remove(struct virtio_device *vdev)
 	disable_rx_mode_work(vi);
 	flush_work(&vi->rx_mode_work);
 
+	virtnet_free_irq_moder(vi);
+
 	unregister_netdev(vi->dev);
 
 	net_failover_destroy(vi->failover);
-- 
cgit v1.2.3-58-ga151


From 1108566ca509e67aa8abfbf914b1cd31e9ff51f8 Mon Sep 17 00:00:00 2001
From: Ziwei Xiao <ziweixiao@google.com>
Date: Tue, 25 Jun 2024 00:12:27 +0000
Subject: gve: Add adminq mutex lock

We were depending on the rtnl_lock to make sure there is only one adminq
command running at a time. But some commands may take too long to hold
the rtnl_lock, such as the upcoming flow steering operations. For such
situations, it can temporarily drop the rtnl_lock, and replace it for
these operations with a new adminq lock, which can ensure the adminq
command execution to be thread-safe.

Signed-off-by: Ziwei Xiao <ziweixiao@google.com>
Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240625001232.1476315-2-ziweixiao@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/google/gve/gve.h        |  1 +
 drivers/net/ethernet/google/gve/gve_adminq.c | 22 ++++++++++++----------
 2 files changed, 13 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h
index ae1e21c9b0a5..ca7fce17f2c0 100644
--- a/drivers/net/ethernet/google/gve/gve.h
+++ b/drivers/net/ethernet/google/gve/gve.h
@@ -724,6 +724,7 @@ struct gve_priv {
 	union gve_adminq_command *adminq;
 	dma_addr_t adminq_bus_addr;
 	struct dma_pool *adminq_pool;
+	struct mutex adminq_lock; /* Protects adminq command execution */
 	u32 adminq_mask; /* masks prod_cnt to adminq size */
 	u32 adminq_prod_cnt; /* free-running count of AQ cmds executed */
 	u32 adminq_cmd_fail; /* free-running count of AQ cmds failed */
diff --git a/drivers/net/ethernet/google/gve/gve_adminq.c b/drivers/net/ethernet/google/gve/gve_adminq.c
index 8ca0def176ef..2e0c1eb87b11 100644
--- a/drivers/net/ethernet/google/gve/gve_adminq.c
+++ b/drivers/net/ethernet/google/gve/gve_adminq.c
@@ -284,6 +284,7 @@ int gve_adminq_alloc(struct device *dev, struct gve_priv *priv)
 			    &priv->reg_bar0->adminq_base_address_lo);
 		iowrite32be(GVE_DRIVER_STATUS_RUN_MASK, &priv->reg_bar0->driver_status);
 	}
+	mutex_init(&priv->adminq_lock);
 	gve_set_admin_queue_ok(priv);
 	return 0;
 }
@@ -511,28 +512,29 @@ static int gve_adminq_issue_cmd(struct gve_priv *priv,
 	return 0;
 }
 
-/* This function is not threadsafe - the caller is responsible for any
- * necessary locks.
- * The caller is also responsible for making sure there are no commands
- * waiting to be executed.
- */
 static int gve_adminq_execute_cmd(struct gve_priv *priv,
 				  union gve_adminq_command *cmd_orig)
 {
 	u32 tail, head;
 	int err;
 
+	mutex_lock(&priv->adminq_lock);
 	tail = ioread32be(&priv->reg_bar0->adminq_event_counter);
 	head = priv->adminq_prod_cnt;
-	if (tail != head)
-		// This is not a valid path
-		return -EINVAL;
+	if (tail != head) {
+		err = -EINVAL;
+		goto out;
+	}
 
 	err = gve_adminq_issue_cmd(priv, cmd_orig);
 	if (err)
-		return err;
+		goto out;
 
-	return gve_adminq_kick_and_wait(priv);
+	err = gve_adminq_kick_and_wait(priv);
+
+out:
+	mutex_unlock(&priv->adminq_lock);
+	return err;
 }
 
 /* The device specifies that the management vector can either be the first irq
-- 
cgit v1.2.3-58-ga151


From fcfe6318dbac45e6c3c5c7a40f5dd5c08600a5e9 Mon Sep 17 00:00:00 2001
From: Jeroen de Borst <jeroendb@google.com>
Date: Tue, 25 Jun 2024 00:12:28 +0000
Subject: gve: Add adminq extended command

The adminq command is limited to 64 bytes per entry and it's 56 bytes
for the command itself at maximum. To support larger commands, we need
to dma_alloc a separate memory to put the command in that memory and
send the dma memory address instead of the actual command.

Introduce an extended adminq command to wrap the real command with the
inner opcode and the allocated dma memory address specified. Once the
device receives it, it can get the real command from the given dma
memory address. As designed with the device, all the extended commands
will use inner opcode larger than 0xFF.

Signed-off-by: Jeroen de Borst <jeroendb@google.com>
Co-developed-by: Ziwei Xiao <ziweixiao@google.com>
Signed-off-by: Ziwei Xiao <ziweixiao@google.com>
Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240625001232.1476315-3-ziweixiao@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/google/gve/gve_adminq.c | 31 ++++++++++++++++++++++++++++
 drivers/net/ethernet/google/gve/gve_adminq.h | 12 +++++++++++
 2 files changed, 43 insertions(+)

diff --git a/drivers/net/ethernet/google/gve/gve_adminq.c b/drivers/net/ethernet/google/gve/gve_adminq.c
index 2e0c1eb87b11..5b54ce369eb2 100644
--- a/drivers/net/ethernet/google/gve/gve_adminq.c
+++ b/drivers/net/ethernet/google/gve/gve_adminq.c
@@ -461,6 +461,8 @@ static int gve_adminq_issue_cmd(struct gve_priv *priv,
 
 	memcpy(cmd, cmd_orig, sizeof(*cmd_orig));
 	opcode = be32_to_cpu(READ_ONCE(cmd->opcode));
+	if (opcode == GVE_ADMINQ_EXTENDED_COMMAND)
+		opcode = be32_to_cpu(cmd->extended_command.inner_opcode);
 
 	switch (opcode) {
 	case GVE_ADMINQ_DESCRIBE_DEVICE:
@@ -537,6 +539,35 @@ out:
 	return err;
 }
 
+static int __maybe_unused gve_adminq_execute_extended_cmd(struct gve_priv *priv, u32 opcode,
+							  size_t cmd_size, void *cmd_orig)
+{
+	union gve_adminq_command cmd;
+	dma_addr_t inner_cmd_bus;
+	void *inner_cmd;
+	int err;
+
+	inner_cmd = dma_alloc_coherent(&priv->pdev->dev, cmd_size,
+				       &inner_cmd_bus, GFP_KERNEL);
+	if (!inner_cmd)
+		return -ENOMEM;
+
+	memcpy(inner_cmd, cmd_orig, cmd_size);
+
+	memset(&cmd, 0, sizeof(cmd));
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_EXTENDED_COMMAND);
+	cmd.extended_command = (struct gve_adminq_extended_command) {
+		.inner_opcode = cpu_to_be32(opcode),
+		.inner_length = cpu_to_be32(cmd_size),
+		.inner_command_addr = cpu_to_be64(inner_cmd_bus),
+	};
+
+	err = gve_adminq_execute_cmd(priv, &cmd);
+
+	dma_free_coherent(&priv->pdev->dev, cmd_size, inner_cmd, inner_cmd_bus);
+	return err;
+}
+
 /* The device specifies that the management vector can either be the first irq
  * or the last irq. ntfy_blk_msix_base_idx indicates the first irq assigned to
  * the ntfy blks. It if is 0 then the management vector is last, if it is 1 then
diff --git a/drivers/net/ethernet/google/gve/gve_adminq.h b/drivers/net/ethernet/google/gve/gve_adminq.h
index e64f0dbe744d..e0370ace8397 100644
--- a/drivers/net/ethernet/google/gve/gve_adminq.h
+++ b/drivers/net/ethernet/google/gve/gve_adminq.h
@@ -25,6 +25,9 @@ enum gve_adminq_opcodes {
 	GVE_ADMINQ_REPORT_LINK_SPEED		= 0xD,
 	GVE_ADMINQ_GET_PTYPE_MAP		= 0xE,
 	GVE_ADMINQ_VERIFY_DRIVER_COMPATIBILITY	= 0xF,
+
+	/* For commands that are larger than 56 bytes */
+	GVE_ADMINQ_EXTENDED_COMMAND		= 0xFF,
 };
 
 /* Admin queue status codes */
@@ -208,6 +211,14 @@ enum gve_driver_capbility {
 #define GVE_DRIVER_CAPABILITY_FLAGS3 0x0
 #define GVE_DRIVER_CAPABILITY_FLAGS4 0x0
 
+struct gve_adminq_extended_command {
+	__be32 inner_opcode;
+	__be32 inner_length;
+	__be64 inner_command_addr;
+};
+
+static_assert(sizeof(struct gve_adminq_extended_command) == 16);
+
 struct gve_driver_info {
 	u8 os_type;	/* 0x01 = Linux */
 	u8 driver_major;
@@ -432,6 +443,7 @@ union gve_adminq_command {
 			struct gve_adminq_get_ptype_map get_ptype_map;
 			struct gve_adminq_verify_driver_compatibility
 						verify_driver_compatibility;
+			struct gve_adminq_extended_command extended_command;
 		};
 	};
 	u8 reserved[64];
-- 
cgit v1.2.3-58-ga151


From 3519c00557e0725906384c42e9f490d311c7b93f Mon Sep 17 00:00:00 2001
From: Jeroen de Borst <jeroendb@google.com>
Date: Tue, 25 Jun 2024 00:12:29 +0000
Subject: gve: Add flow steering device option

Add a new device option to signal to the driver that the device supports
flow steering. This device option also carries the maximum number of
flow steering rules that the device can store.

Signed-off-by: Jeroen de Borst <jeroendb@google.com>
Co-developed-by: Ziwei Xiao <ziweixiao@google.com>
Signed-off-by: Ziwei Xiao <ziweixiao@google.com>
Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240625001232.1476315-4-ziweixiao@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/google/gve/gve.h        |  2 ++
 drivers/net/ethernet/google/gve/gve_adminq.c | 40 ++++++++++++++++++++++++++--
 drivers/net/ethernet/google/gve/gve_adminq.h | 11 ++++++++
 3 files changed, 51 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h
index ca7fce17f2c0..58213c15e084 100644
--- a/drivers/net/ethernet/google/gve/gve.h
+++ b/drivers/net/ethernet/google/gve/gve.h
@@ -786,6 +786,8 @@ struct gve_priv {
 
 	u16 header_buf_size; /* device configured, header-split supported if non-zero */
 	bool header_split_enabled; /* True if the header split is enabled by the user */
+
+	u32 max_flow_rules;
 };
 
 enum gve_service_task_flags_bit {
diff --git a/drivers/net/ethernet/google/gve/gve_adminq.c b/drivers/net/ethernet/google/gve/gve_adminq.c
index 5b54ce369eb2..088f543f0ba8 100644
--- a/drivers/net/ethernet/google/gve/gve_adminq.c
+++ b/drivers/net/ethernet/google/gve/gve_adminq.c
@@ -44,6 +44,7 @@ void gve_parse_device_option(struct gve_priv *priv,
 			     struct gve_device_option_jumbo_frames **dev_op_jumbo_frames,
 			     struct gve_device_option_dqo_qpl **dev_op_dqo_qpl,
 			     struct gve_device_option_buffer_sizes **dev_op_buffer_sizes,
+			     struct gve_device_option_flow_steering **dev_op_flow_steering,
 			     struct gve_device_option_modify_ring **dev_op_modify_ring)
 {
 	u32 req_feat_mask = be32_to_cpu(option->required_features_mask);
@@ -189,6 +190,23 @@ void gve_parse_device_option(struct gve_priv *priv,
 		if (option_length == GVE_DEVICE_OPTION_NO_MIN_RING_SIZE)
 			priv->default_min_ring_size = true;
 		break;
+	case GVE_DEV_OPT_ID_FLOW_STEERING:
+		if (option_length < sizeof(**dev_op_flow_steering) ||
+		    req_feat_mask != GVE_DEV_OPT_REQ_FEAT_MASK_FLOW_STEERING) {
+			dev_warn(&priv->pdev->dev, GVE_DEVICE_OPTION_ERROR_FMT,
+				 "Flow Steering",
+				 (int)sizeof(**dev_op_flow_steering),
+				 GVE_DEV_OPT_REQ_FEAT_MASK_FLOW_STEERING,
+				 option_length, req_feat_mask);
+			break;
+		}
+
+		if (option_length > sizeof(**dev_op_flow_steering))
+			dev_warn(&priv->pdev->dev,
+				 GVE_DEVICE_OPTION_TOO_BIG_FMT,
+				 "Flow Steering");
+		*dev_op_flow_steering = (void *)(option + 1);
+		break;
 	default:
 		/* If we don't recognize the option just continue
 		 * without doing anything.
@@ -208,6 +226,7 @@ gve_process_device_options(struct gve_priv *priv,
 			   struct gve_device_option_jumbo_frames **dev_op_jumbo_frames,
 			   struct gve_device_option_dqo_qpl **dev_op_dqo_qpl,
 			   struct gve_device_option_buffer_sizes **dev_op_buffer_sizes,
+			   struct gve_device_option_flow_steering **dev_op_flow_steering,
 			   struct gve_device_option_modify_ring **dev_op_modify_ring)
 {
 	const int num_options = be16_to_cpu(descriptor->num_device_options);
@@ -230,7 +249,7 @@ gve_process_device_options(struct gve_priv *priv,
 					dev_op_gqi_rda, dev_op_gqi_qpl,
 					dev_op_dqo_rda, dev_op_jumbo_frames,
 					dev_op_dqo_qpl, dev_op_buffer_sizes,
-					dev_op_modify_ring);
+					dev_op_flow_steering, dev_op_modify_ring);
 		dev_opt = next_opt;
 	}
 
@@ -838,6 +857,8 @@ static void gve_enable_supported_features(struct gve_priv *priv,
 					  *dev_op_dqo_qpl,
 					  const struct gve_device_option_buffer_sizes
 					  *dev_op_buffer_sizes,
+					  const struct gve_device_option_flow_steering
+					  *dev_op_flow_steering,
 					  const struct gve_device_option_modify_ring
 					  *dev_op_modify_ring)
 {
@@ -890,10 +911,23 @@ static void gve_enable_supported_features(struct gve_priv *priv,
 			priv->min_tx_desc_cnt = be16_to_cpu(dev_op_modify_ring->min_tx_ring_size);
 		}
 	}
+
+	if (dev_op_flow_steering &&
+	    (supported_features_mask & GVE_SUP_FLOW_STEERING_MASK)) {
+		if (dev_op_flow_steering->max_flow_rules) {
+			priv->max_flow_rules =
+				be32_to_cpu(dev_op_flow_steering->max_flow_rules);
+			priv->dev->hw_features |= NETIF_F_NTUPLE;
+			dev_info(&priv->pdev->dev,
+				 "FLOW STEERING device option enabled with max rule limit of %u.\n",
+				 priv->max_flow_rules);
+		}
+	}
 }
 
 int gve_adminq_describe_device(struct gve_priv *priv)
 {
+	struct gve_device_option_flow_steering *dev_op_flow_steering = NULL;
 	struct gve_device_option_buffer_sizes *dev_op_buffer_sizes = NULL;
 	struct gve_device_option_jumbo_frames *dev_op_jumbo_frames = NULL;
 	struct gve_device_option_modify_ring *dev_op_modify_ring = NULL;
@@ -930,6 +964,7 @@ int gve_adminq_describe_device(struct gve_priv *priv)
 					 &dev_op_gqi_qpl, &dev_op_dqo_rda,
 					 &dev_op_jumbo_frames, &dev_op_dqo_qpl,
 					 &dev_op_buffer_sizes,
+					 &dev_op_flow_steering,
 					 &dev_op_modify_ring);
 	if (err)
 		goto free_device_descriptor;
@@ -991,7 +1026,8 @@ int gve_adminq_describe_device(struct gve_priv *priv)
 
 	gve_enable_supported_features(priv, supported_features_mask,
 				      dev_op_jumbo_frames, dev_op_dqo_qpl,
-				      dev_op_buffer_sizes, dev_op_modify_ring);
+				      dev_op_buffer_sizes, dev_op_flow_steering,
+				      dev_op_modify_ring);
 
 free_device_descriptor:
 	dma_pool_free(priv->adminq_pool, descriptor, descriptor_bus);
diff --git a/drivers/net/ethernet/google/gve/gve_adminq.h b/drivers/net/ethernet/google/gve/gve_adminq.h
index e0370ace8397..e64a0e72e781 100644
--- a/drivers/net/ethernet/google/gve/gve_adminq.h
+++ b/drivers/net/ethernet/google/gve/gve_adminq.h
@@ -146,6 +146,14 @@ struct gve_device_option_modify_ring {
 
 static_assert(sizeof(struct gve_device_option_modify_ring) == 12);
 
+struct gve_device_option_flow_steering {
+	__be32 supported_features_mask;
+	__be32 reserved;
+	__be32 max_flow_rules;
+};
+
+static_assert(sizeof(struct gve_device_option_flow_steering) == 12);
+
 /* Terminology:
  *
  * RDA - Raw DMA Addressing - Buffers associated with SKBs are directly DMA
@@ -163,6 +171,7 @@ enum gve_dev_opt_id {
 	GVE_DEV_OPT_ID_DQO_QPL			= 0x7,
 	GVE_DEV_OPT_ID_JUMBO_FRAMES		= 0x8,
 	GVE_DEV_OPT_ID_BUFFER_SIZES		= 0xa,
+	GVE_DEV_OPT_ID_FLOW_STEERING		= 0xb,
 };
 
 enum gve_dev_opt_req_feat_mask {
@@ -174,12 +183,14 @@ enum gve_dev_opt_req_feat_mask {
 	GVE_DEV_OPT_REQ_FEAT_MASK_DQO_QPL		= 0x0,
 	GVE_DEV_OPT_REQ_FEAT_MASK_BUFFER_SIZES		= 0x0,
 	GVE_DEV_OPT_REQ_FEAT_MASK_MODIFY_RING		= 0x0,
+	GVE_DEV_OPT_REQ_FEAT_MASK_FLOW_STEERING		= 0x0,
 };
 
 enum gve_sup_feature_mask {
 	GVE_SUP_MODIFY_RING_MASK	= 1 << 0,
 	GVE_SUP_JUMBO_FRAMES_MASK	= 1 << 2,
 	GVE_SUP_BUFFER_SIZES_MASK	= 1 << 4,
+	GVE_SUP_FLOW_STEERING_MASK	= 1 << 5,
 };
 
 #define GVE_DEV_OPT_LEN_GQI_RAW_ADDRESSING 0x0
-- 
cgit v1.2.3-58-ga151


From 57718b60df9b67366291d1e4ce3e5dde7a7cf5a5 Mon Sep 17 00:00:00 2001
From: Jeroen de Borst <jeroendb@google.com>
Date: Tue, 25 Jun 2024 00:12:30 +0000
Subject: gve: Add flow steering adminq commands

Add new adminq commands for the driver to configure and query flow rules
that are stored in the device. Flow steering rules are assigned with a
location that determines the relative order of the rules.

Flow rules can run up to an order of millions. In such cases, storing
a full copy of the rules in the driver to prepare for the ethtool query
is infeasible while querying them from the device is better. That needs
to be optimized too so that we don't send a lot of adminq commands. The
solution here is to store a limited number of rules/rule ids in the
driver in a cache. Use dma_pool to allocate 4k bytes which lets device
write at most 46 flow rules(4096/88) or 1024 rule ids(4096/4) at a time.

For configuring flow rules, there are 3 sub-commands:
- ADD which adds a rule at the location supplied
- DEL which deletes the rule at the location supplied
- RESET which clears all currently active rules in the device

For querying flow rules, there are also 3 sub-commands:
- QUERY_RULES corresponds to ETHTOOL_GRXCLSRULE. It fills the rules in
  the allocated cache after querying the device
- QUERY_RULES_IDS corresponds to ETHTOOL_GRXCLSRLALL. It fills the
  rule_ids in the allocated cache after querying the device
- QUERY_RULES_STATS corresponds to ETHTOOL_GRXCLSRLCNT. It queries the
  device's current flow rule number and the supported max flow rule
  limit

Signed-off-by: Jeroen de Borst <jeroendb@google.com>
Co-developed-by: Ziwei Xiao <ziweixiao@google.com>
Signed-off-by: Ziwei Xiao <ziweixiao@google.com>
Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240625001232.1476315-5-ziweixiao@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/google/gve/gve.h         |  43 ++++++++
 drivers/net/ethernet/google/gve/gve_adminq.c  | 139 +++++++++++++++++++++++++-
 drivers/net/ethernet/google/gve/gve_adminq.h  |  80 +++++++++++++++
 drivers/net/ethernet/google/gve/gve_ethtool.c |   5 +-
 drivers/net/ethernet/google/gve/gve_main.c    |  51 +++++++++-
 5 files changed, 314 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h
index 58213c15e084..b9e9dd958f3c 100644
--- a/drivers/net/ethernet/google/gve/gve.h
+++ b/drivers/net/ethernet/google/gve/gve.h
@@ -60,6 +60,11 @@
 
 #define GVE_DEFAULT_RX_BUFFER_OFFSET 2048
 
+#define GVE_FLOW_RULES_CACHE_SIZE \
+	(GVE_ADMINQ_BUFFER_SIZE / sizeof(struct gve_adminq_queried_flow_rule))
+#define GVE_FLOW_RULE_IDS_CACHE_SIZE \
+	(GVE_ADMINQ_BUFFER_SIZE / sizeof(((struct gve_adminq_queried_flow_rule *)0)->location))
+
 #define GVE_XDP_ACTIONS 5
 
 #define GVE_GQ_TX_MIN_PKT_DESC_BYTES 182
@@ -678,6 +683,39 @@ enum gve_queue_format {
 	GVE_DQO_QPL_FORMAT		= 0x4,
 };
 
+struct gve_flow_spec {
+	__be32 src_ip[4];
+	__be32 dst_ip[4];
+	union {
+		struct {
+			__be16 src_port;
+			__be16 dst_port;
+		};
+		__be32 spi;
+	};
+	union {
+		u8 tos;
+		u8 tclass;
+	};
+};
+
+struct gve_flow_rule {
+	u32 location;
+	u16 flow_type;
+	u16 action;
+	struct gve_flow_spec key;
+	struct gve_flow_spec mask;
+};
+
+struct gve_flow_rules_cache {
+	bool rules_cache_synced; /* False if the driver's rules_cache is outdated */
+	struct gve_adminq_queried_flow_rule *rules_cache;
+	__be32 *rule_ids_cache;
+	/* The total number of queried rules that stored in the caches */
+	u32 rules_cache_num;
+	u32 rule_ids_cache_num;
+};
+
 struct gve_priv {
 	struct net_device *dev;
 	struct gve_tx_ring *tx; /* array of tx_cfg.num_queues */
@@ -744,6 +782,8 @@ struct gve_priv {
 	u32 adminq_report_link_speed_cnt;
 	u32 adminq_get_ptype_map_cnt;
 	u32 adminq_verify_driver_compatibility_cnt;
+	u32 adminq_query_flow_rules_cnt;
+	u32 adminq_cfg_flow_rule_cnt;
 
 	/* Global stats */
 	u32 interface_up_cnt; /* count of times interface turned up since last reset */
@@ -788,6 +828,9 @@ struct gve_priv {
 	bool header_split_enabled; /* True if the header split is enabled by the user */
 
 	u32 max_flow_rules;
+	u32 num_flow_rules;
+
+	struct gve_flow_rules_cache flow_rules_cache;
 };
 
 enum gve_service_task_flags_bit {
diff --git a/drivers/net/ethernet/google/gve/gve_adminq.c b/drivers/net/ethernet/google/gve/gve_adminq.c
index 088f543f0ba8..c5bbc1b7524e 100644
--- a/drivers/net/ethernet/google/gve/gve_adminq.c
+++ b/drivers/net/ethernet/google/gve/gve_adminq.c
@@ -287,6 +287,8 @@ int gve_adminq_alloc(struct device *dev, struct gve_priv *priv)
 	priv->adminq_report_stats_cnt = 0;
 	priv->adminq_report_link_speed_cnt = 0;
 	priv->adminq_get_ptype_map_cnt = 0;
+	priv->adminq_query_flow_rules_cnt = 0;
+	priv->adminq_cfg_flow_rule_cnt = 0;
 
 	/* Setup Admin queue with the device */
 	if (priv->pdev->revision < 0x1) {
@@ -526,6 +528,12 @@ static int gve_adminq_issue_cmd(struct gve_priv *priv,
 	case GVE_ADMINQ_VERIFY_DRIVER_COMPATIBILITY:
 		priv->adminq_verify_driver_compatibility_cnt++;
 		break;
+	case GVE_ADMINQ_QUERY_FLOW_RULES:
+		priv->adminq_query_flow_rules_cnt++;
+		break;
+	case GVE_ADMINQ_CONFIGURE_FLOW_RULE:
+		priv->adminq_cfg_flow_rule_cnt++;
+		break;
 	default:
 		dev_err(&priv->pdev->dev, "unknown AQ command opcode %d\n", opcode);
 	}
@@ -558,8 +566,8 @@ out:
 	return err;
 }
 
-static int __maybe_unused gve_adminq_execute_extended_cmd(struct gve_priv *priv, u32 opcode,
-							  size_t cmd_size, void *cmd_orig)
+static int gve_adminq_execute_extended_cmd(struct gve_priv *priv, u32 opcode,
+					   size_t cmd_size, void *cmd_orig)
 {
 	union gve_adminq_command cmd;
 	dma_addr_t inner_cmd_bus;
@@ -1190,3 +1198,130 @@ err:
 			  ptype_map_bus);
 	return err;
 }
+
+static int
+gve_adminq_configure_flow_rule(struct gve_priv *priv,
+			       struct gve_adminq_configure_flow_rule *flow_rule_cmd)
+{
+	int err = gve_adminq_execute_extended_cmd(priv,
+			GVE_ADMINQ_CONFIGURE_FLOW_RULE,
+			sizeof(struct gve_adminq_configure_flow_rule),
+			flow_rule_cmd);
+
+	if (err) {
+		dev_err(&priv->pdev->dev, "Timeout to configure the flow rule, trigger reset");
+		gve_reset(priv, true);
+	} else {
+		priv->flow_rules_cache.rules_cache_synced = false;
+	}
+
+	return err;
+}
+
+int gve_adminq_add_flow_rule(struct gve_priv *priv, struct gve_adminq_flow_rule *rule, u32 loc)
+{
+	struct gve_adminq_configure_flow_rule flow_rule_cmd = {
+		.opcode = cpu_to_be16(GVE_FLOW_RULE_CFG_ADD),
+		.location = cpu_to_be32(loc),
+		.rule = *rule,
+	};
+
+	return gve_adminq_configure_flow_rule(priv, &flow_rule_cmd);
+}
+
+int gve_adminq_del_flow_rule(struct gve_priv *priv, u32 loc)
+{
+	struct gve_adminq_configure_flow_rule flow_rule_cmd = {
+		.opcode = cpu_to_be16(GVE_FLOW_RULE_CFG_DEL),
+		.location = cpu_to_be32(loc),
+	};
+
+	return gve_adminq_configure_flow_rule(priv, &flow_rule_cmd);
+}
+
+int gve_adminq_reset_flow_rules(struct gve_priv *priv)
+{
+	struct gve_adminq_configure_flow_rule flow_rule_cmd = {
+		.opcode = cpu_to_be16(GVE_FLOW_RULE_CFG_RESET),
+	};
+
+	return gve_adminq_configure_flow_rule(priv, &flow_rule_cmd);
+}
+
+/* In the dma memory that the driver allocated for the device to query the flow rules, the device
+ * will first write it with a struct of gve_query_flow_rules_descriptor. Next to it, the device
+ * will write an array of rules or rule ids with the count that specified in the descriptor.
+ * For GVE_FLOW_RULE_QUERY_STATS, the device will only write the descriptor.
+ */
+static int gve_adminq_process_flow_rules_query(struct gve_priv *priv, u16 query_opcode,
+					       struct gve_query_flow_rules_descriptor *descriptor)
+{
+	struct gve_flow_rules_cache *flow_rules_cache = &priv->flow_rules_cache;
+	u32 num_queried_rules, total_memory_len, rule_info_len;
+	void *rule_info;
+
+	total_memory_len = be32_to_cpu(descriptor->total_length);
+	num_queried_rules = be32_to_cpu(descriptor->num_queried_rules);
+	rule_info = (void *)(descriptor + 1);
+
+	switch (query_opcode) {
+	case GVE_FLOW_RULE_QUERY_RULES:
+		rule_info_len = num_queried_rules * sizeof(*flow_rules_cache->rules_cache);
+		if (sizeof(*descriptor) + rule_info_len != total_memory_len) {
+			dev_err(&priv->dev->dev, "flow rules query is out of memory.\n");
+			return -ENOMEM;
+		}
+
+		memcpy(flow_rules_cache->rules_cache, rule_info, rule_info_len);
+		flow_rules_cache->rules_cache_num = num_queried_rules;
+		break;
+	case GVE_FLOW_RULE_QUERY_IDS:
+		rule_info_len = num_queried_rules * sizeof(*flow_rules_cache->rule_ids_cache);
+		if (sizeof(*descriptor) + rule_info_len != total_memory_len) {
+			dev_err(&priv->dev->dev, "flow rule ids query is out of memory.\n");
+			return -ENOMEM;
+		}
+
+		memcpy(flow_rules_cache->rule_ids_cache, rule_info, rule_info_len);
+		flow_rules_cache->rule_ids_cache_num = num_queried_rules;
+		break;
+	case GVE_FLOW_RULE_QUERY_STATS:
+		priv->num_flow_rules = be32_to_cpu(descriptor->num_flow_rules);
+		priv->max_flow_rules = be32_to_cpu(descriptor->max_flow_rules);
+		return 0;
+	default:
+		return -EINVAL;
+	}
+
+	return  0;
+}
+
+int gve_adminq_query_flow_rules(struct gve_priv *priv, u16 query_opcode, u32 starting_loc)
+{
+	struct gve_query_flow_rules_descriptor *descriptor;
+	union gve_adminq_command cmd;
+	dma_addr_t descriptor_bus;
+	int err = 0;
+
+	memset(&cmd, 0, sizeof(cmd));
+	descriptor = dma_pool_alloc(priv->adminq_pool, GFP_KERNEL, &descriptor_bus);
+	if (!descriptor)
+		return -ENOMEM;
+
+	cmd.opcode = cpu_to_be32(GVE_ADMINQ_QUERY_FLOW_RULES);
+	cmd.query_flow_rules = (struct gve_adminq_query_flow_rules) {
+		.opcode = cpu_to_be16(query_opcode),
+		.starting_rule_id = cpu_to_be32(starting_loc),
+		.available_length = cpu_to_be64(GVE_ADMINQ_BUFFER_SIZE),
+		.rule_descriptor_addr = cpu_to_be64(descriptor_bus),
+	};
+	err = gve_adminq_execute_cmd(priv, &cmd);
+	if (err)
+		goto out;
+
+	err = gve_adminq_process_flow_rules_query(priv, query_opcode, descriptor);
+
+out:
+	dma_pool_free(priv->adminq_pool, descriptor, descriptor_bus);
+	return err;
+}
diff --git a/drivers/net/ethernet/google/gve/gve_adminq.h b/drivers/net/ethernet/google/gve/gve_adminq.h
index e64a0e72e781..ed1370c9b197 100644
--- a/drivers/net/ethernet/google/gve/gve_adminq.h
+++ b/drivers/net/ethernet/google/gve/gve_adminq.h
@@ -25,11 +25,21 @@ enum gve_adminq_opcodes {
 	GVE_ADMINQ_REPORT_LINK_SPEED		= 0xD,
 	GVE_ADMINQ_GET_PTYPE_MAP		= 0xE,
 	GVE_ADMINQ_VERIFY_DRIVER_COMPATIBILITY	= 0xF,
+	GVE_ADMINQ_QUERY_FLOW_RULES		= 0x10,
 
 	/* For commands that are larger than 56 bytes */
 	GVE_ADMINQ_EXTENDED_COMMAND		= 0xFF,
 };
 
+/* The normal adminq command is restricted to be 56 bytes at maximum. For the
+ * longer adminq command, it is wrapped by GVE_ADMINQ_EXTENDED_COMMAND with
+ * inner opcode of gve_adminq_extended_cmd_opcodes specified. The inner command
+ * is written in the dma memory allocated by GVE_ADMINQ_EXTENDED_COMMAND.
+ */
+enum gve_adminq_extended_cmd_opcodes {
+	GVE_ADMINQ_CONFIGURE_FLOW_RULE	= 0x101,
+};
+
 /* Admin queue status codes */
 enum gve_adminq_statuses {
 	GVE_ADMINQ_COMMAND_UNSET			= 0x0,
@@ -434,6 +444,71 @@ struct gve_adminq_get_ptype_map {
 	__be64 ptype_map_addr;
 };
 
+/* Flow-steering related definitions */
+enum gve_adminq_flow_rule_cfg_opcode {
+	GVE_FLOW_RULE_CFG_ADD	= 0,
+	GVE_FLOW_RULE_CFG_DEL	= 1,
+	GVE_FLOW_RULE_CFG_RESET	= 2,
+};
+
+enum gve_adminq_flow_rule_query_opcode {
+	GVE_FLOW_RULE_QUERY_RULES	= 0,
+	GVE_FLOW_RULE_QUERY_IDS		= 1,
+	GVE_FLOW_RULE_QUERY_STATS	= 2,
+};
+
+enum gve_adminq_flow_type {
+	GVE_FLOW_TYPE_TCPV4,
+	GVE_FLOW_TYPE_UDPV4,
+	GVE_FLOW_TYPE_SCTPV4,
+	GVE_FLOW_TYPE_AHV4,
+	GVE_FLOW_TYPE_ESPV4,
+	GVE_FLOW_TYPE_TCPV6,
+	GVE_FLOW_TYPE_UDPV6,
+	GVE_FLOW_TYPE_SCTPV6,
+	GVE_FLOW_TYPE_AHV6,
+	GVE_FLOW_TYPE_ESPV6,
+};
+
+/* Flow-steering command */
+struct gve_adminq_flow_rule {
+	__be16 flow_type;
+	__be16 action; /* RX queue id */
+	struct gve_flow_spec key;
+	struct gve_flow_spec mask;
+};
+
+struct gve_adminq_configure_flow_rule {
+	__be16 opcode;
+	u8 padding[2];
+	struct gve_adminq_flow_rule rule;
+	__be32 location;
+};
+
+static_assert(sizeof(struct gve_adminq_configure_flow_rule) == 92);
+
+struct gve_query_flow_rules_descriptor {
+	__be32 num_flow_rules;
+	__be32 max_flow_rules;
+	__be32 num_queried_rules;
+	__be32 total_length;
+};
+
+struct gve_adminq_queried_flow_rule {
+	__be32 location;
+	struct gve_adminq_flow_rule flow_rule;
+};
+
+struct gve_adminq_query_flow_rules {
+	__be16 opcode;
+	u8 padding[2];
+	__be32 starting_rule_id;
+	__be64 available_length; /* The dma memory length that the driver allocated */
+	__be64 rule_descriptor_addr; /* The dma memory address */
+};
+
+static_assert(sizeof(struct gve_adminq_query_flow_rules) == 24);
+
 union gve_adminq_command {
 	struct {
 		__be32 opcode;
@@ -454,6 +529,7 @@ union gve_adminq_command {
 			struct gve_adminq_get_ptype_map get_ptype_map;
 			struct gve_adminq_verify_driver_compatibility
 						verify_driver_compatibility;
+			struct gve_adminq_query_flow_rules query_flow_rules;
 			struct gve_adminq_extended_command extended_command;
 		};
 	};
@@ -488,6 +564,10 @@ int gve_adminq_verify_driver_compatibility(struct gve_priv *priv,
 					   u64 driver_info_len,
 					   dma_addr_t driver_info_addr);
 int gve_adminq_report_link_speed(struct gve_priv *priv);
+int gve_adminq_add_flow_rule(struct gve_priv *priv, struct gve_adminq_flow_rule *rule, u32 loc);
+int gve_adminq_del_flow_rule(struct gve_priv *priv, u32 loc);
+int gve_adminq_reset_flow_rules(struct gve_priv *priv);
+int gve_adminq_query_flow_rules(struct gve_priv *priv, u16 query_opcode, u32 starting_loc);
 
 struct gve_ptype_lut;
 int gve_adminq_get_ptype_map_dqo(struct gve_priv *priv,
diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c
index fe1741d482b4..ffaa878d67bc 100644
--- a/drivers/net/ethernet/google/gve/gve_ethtool.c
+++ b/drivers/net/ethernet/google/gve/gve_ethtool.c
@@ -74,7 +74,8 @@ static const char gve_gstrings_adminq_stats[][ETH_GSTRING_LEN] = {
 	"adminq_create_tx_queue_cnt", "adminq_create_rx_queue_cnt",
 	"adminq_destroy_tx_queue_cnt", "adminq_destroy_rx_queue_cnt",
 	"adminq_dcfg_device_resources_cnt", "adminq_set_driver_parameter_cnt",
-	"adminq_report_stats_cnt", "adminq_report_link_speed_cnt", "adminq_get_ptype_map_cnt"
+	"adminq_report_stats_cnt", "adminq_report_link_speed_cnt", "adminq_get_ptype_map_cnt",
+	"adminq_query_flow_rules", "adminq_cfg_flow_rule",
 };
 
 static const char gve_gstrings_priv_flags[][ETH_GSTRING_LEN] = {
@@ -450,6 +451,8 @@ gve_get_ethtool_stats(struct net_device *netdev,
 	data[i++] = priv->adminq_report_stats_cnt;
 	data[i++] = priv->adminq_report_link_speed_cnt;
 	data[i++] = priv->adminq_get_ptype_map_cnt;
+	data[i++] = priv->adminq_query_flow_rules_cnt;
+	data[i++] = priv->adminq_cfg_flow_rule_cnt;
 }
 
 static void gve_get_channels(struct net_device *netdev,
diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c
index cabf7d4bcecb..fc142856e189 100644
--- a/drivers/net/ethernet/google/gve/gve_main.c
+++ b/drivers/net/ethernet/google/gve/gve_main.c
@@ -141,6 +141,49 @@ static void gve_get_stats(struct net_device *dev, struct rtnl_link_stats64 *s)
 	}
 }
 
+static int gve_alloc_flow_rule_caches(struct gve_priv *priv)
+{
+	struct gve_flow_rules_cache *flow_rules_cache = &priv->flow_rules_cache;
+	int err = 0;
+
+	if (!priv->max_flow_rules)
+		return 0;
+
+	flow_rules_cache->rules_cache =
+		kvcalloc(GVE_FLOW_RULES_CACHE_SIZE, sizeof(*flow_rules_cache->rules_cache),
+			 GFP_KERNEL);
+	if (!flow_rules_cache->rules_cache) {
+		dev_err(&priv->pdev->dev, "Cannot alloc flow rules cache\n");
+		return -ENOMEM;
+	}
+
+	flow_rules_cache->rule_ids_cache =
+		kvcalloc(GVE_FLOW_RULE_IDS_CACHE_SIZE, sizeof(*flow_rules_cache->rule_ids_cache),
+			 GFP_KERNEL);
+	if (!flow_rules_cache->rule_ids_cache) {
+		dev_err(&priv->pdev->dev, "Cannot alloc flow rule ids cache\n");
+		err = -ENOMEM;
+		goto free_rules_cache;
+	}
+
+	return 0;
+
+free_rules_cache:
+	kvfree(flow_rules_cache->rules_cache);
+	flow_rules_cache->rules_cache = NULL;
+	return err;
+}
+
+static void gve_free_flow_rule_caches(struct gve_priv *priv)
+{
+	struct gve_flow_rules_cache *flow_rules_cache = &priv->flow_rules_cache;
+
+	kvfree(flow_rules_cache->rule_ids_cache);
+	flow_rules_cache->rule_ids_cache = NULL;
+	kvfree(flow_rules_cache->rules_cache);
+	flow_rules_cache->rules_cache = NULL;
+}
+
 static int gve_alloc_counter_array(struct gve_priv *priv)
 {
 	priv->counter_array =
@@ -521,9 +564,12 @@ static int gve_setup_device_resources(struct gve_priv *priv)
 {
 	int err;
 
-	err = gve_alloc_counter_array(priv);
+	err = gve_alloc_flow_rule_caches(priv);
 	if (err)
 		return err;
+	err = gve_alloc_counter_array(priv);
+	if (err)
+		goto abort_with_flow_rule_caches;
 	err = gve_alloc_notify_blocks(priv);
 	if (err)
 		goto abort_with_counter;
@@ -575,6 +621,8 @@ abort_with_ntfy_blocks:
 	gve_free_notify_blocks(priv);
 abort_with_counter:
 	gve_free_counter_array(priv);
+abort_with_flow_rule_caches:
+	gve_free_flow_rule_caches(priv);
 
 	return err;
 }
@@ -606,6 +654,7 @@ static void gve_teardown_device_resources(struct gve_priv *priv)
 	kvfree(priv->ptype_lut_dqo);
 	priv->ptype_lut_dqo = NULL;
 
+	gve_free_flow_rule_caches(priv);
 	gve_free_counter_array(priv);
 	gve_free_notify_blocks(priv);
 	gve_free_stats_report(priv);
-- 
cgit v1.2.3-58-ga151


From 6f3bc487565d30ae9535e338efad213dd74c04ec Mon Sep 17 00:00:00 2001
From: Jeroen de Borst <jeroendb@google.com>
Date: Tue, 25 Jun 2024 00:12:31 +0000
Subject: gve: Add flow steering ethtool support

Implement the ethtool commands that can be used to configure and query
flow-steering rules.

A large part of this change consists of translating the ethtool
representation of 'ntuples' to our internal gve_flow_rule and vice-versa
in the new created gve_flow_rule.c

Considering the possible large amount of flow rules, the driver doesn't
store all the rules locally. When the user runs 'ethtool -n <nic>' to
check the registered rules, the driver will send adminq command to
query a limited amount of rules/rule ids(that filled in a 4096 bytes dma
memory) at a time as a cache for the ethtool queries. The adminq query
commands will be repeated for several times until the ethtool has
queried all the needed rules.

Signed-off-by: Jeroen de Borst <jeroendb@google.com>
Co-developed-by: Ziwei Xiao <ziweixiao@google.com>
Signed-off-by: Ziwei Xiao <ziweixiao@google.com>
Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com>
Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240625001232.1476315-6-ziweixiao@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/google/gve/Makefile        |   2 +-
 drivers/net/ethernet/google/gve/gve.h           |   8 +-
 drivers/net/ethernet/google/gve/gve_ethtool.c   |  67 +++++-
 drivers/net/ethernet/google/gve/gve_flow_rule.c | 298 ++++++++++++++++++++++++
 drivers/net/ethernet/google/gve/gve_main.c      |  32 ++-
 5 files changed, 398 insertions(+), 9 deletions(-)
 create mode 100644 drivers/net/ethernet/google/gve/gve_flow_rule.c

diff --git a/drivers/net/ethernet/google/gve/Makefile b/drivers/net/ethernet/google/gve/Makefile
index b9a6be76531b..9ed07080b38a 100644
--- a/drivers/net/ethernet/google/gve/Makefile
+++ b/drivers/net/ethernet/google/gve/Makefile
@@ -1,4 +1,4 @@
 # Makefile for the Google virtual Ethernet (gve) driver
 
 obj-$(CONFIG_GVE) += gve.o
-gve-objs := gve_main.o gve_tx.o gve_tx_dqo.o gve_rx.o gve_rx_dqo.o gve_ethtool.o gve_adminq.o gve_utils.o
+gve-objs := gve_main.o gve_tx.o gve_tx_dqo.o gve_rx.o gve_rx_dqo.o gve_ethtool.o gve_adminq.o gve_utils.o gve_flow_rule.o
diff --git a/drivers/net/ethernet/google/gve/gve.h b/drivers/net/ethernet/google/gve/gve.h
index b9e9dd958f3c..84ac004d3953 100644
--- a/drivers/net/ethernet/google/gve/gve.h
+++ b/drivers/net/ethernet/google/gve/gve.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: (GPL-2.0 OR MIT)
  * Google virtual Ethernet (gve) driver
  *
- * Copyright (C) 2015-2021 Google, Inc.
+ * Copyright (C) 2015-2024 Google LLC
  */
 
 #ifndef _GVE_H_
@@ -1170,6 +1170,12 @@ int gve_adjust_config(struct gve_priv *priv,
 int gve_adjust_queues(struct gve_priv *priv,
 		      struct gve_queue_config new_rx_config,
 		      struct gve_queue_config new_tx_config);
+/* flow steering rule */
+int gve_get_flow_rule_entry(struct gve_priv *priv, struct ethtool_rxnfc *cmd);
+int gve_get_flow_rule_ids(struct gve_priv *priv, struct ethtool_rxnfc *cmd, u32 *rule_locs);
+int gve_add_flow_rule(struct gve_priv *priv, struct ethtool_rxnfc *cmd);
+int gve_del_flow_rule(struct gve_priv *priv, struct ethtool_rxnfc *cmd);
+int gve_flow_rules_reset(struct gve_priv *priv);
 /* report stats handling */
 void gve_handle_report_stats(struct gve_priv *priv);
 /* exported by ethtool.c */
diff --git a/drivers/net/ethernet/google/gve/gve_ethtool.c b/drivers/net/ethernet/google/gve/gve_ethtool.c
index ffaa878d67bc..3480ff5c7ed6 100644
--- a/drivers/net/ethernet/google/gve/gve_ethtool.c
+++ b/drivers/net/ethernet/google/gve/gve_ethtool.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: (GPL-2.0 OR MIT)
 /* Google virtual Ethernet (gve) driver
  *
- * Copyright (C) 2015-2021 Google, Inc.
+ * Copyright (C) 2015-2024 Google LLC
  */
 
 #include <linux/rtnetlink.h>
@@ -775,6 +775,69 @@ static int gve_set_coalesce(struct net_device *netdev,
 	return 0;
 }
 
+static int gve_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	int err = 0;
+
+	if (!(netdev->features & NETIF_F_NTUPLE))
+		return -EOPNOTSUPP;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_SRXCLSRLINS:
+		err = gve_add_flow_rule(priv, cmd);
+		break;
+	case ETHTOOL_SRXCLSRLDEL:
+		err = gve_del_flow_rule(priv, cmd);
+		break;
+	case ETHTOOL_SRXFH:
+		err = -EOPNOTSUPP;
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
+static int gve_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd, u32 *rule_locs)
+{
+	struct gve_priv *priv = netdev_priv(netdev);
+	int err = 0;
+
+	switch (cmd->cmd) {
+	case ETHTOOL_GRXRINGS:
+		cmd->data = priv->rx_cfg.num_queues;
+		break;
+	case ETHTOOL_GRXCLSRLCNT:
+		if (!priv->max_flow_rules)
+			return -EOPNOTSUPP;
+
+		err = gve_adminq_query_flow_rules(priv, GVE_FLOW_RULE_QUERY_STATS, 0);
+		if (err)
+			return err;
+
+		cmd->rule_cnt = priv->num_flow_rules;
+		cmd->data = priv->max_flow_rules;
+		break;
+	case ETHTOOL_GRXCLSRULE:
+		err = gve_get_flow_rule_entry(priv, cmd);
+		break;
+	case ETHTOOL_GRXCLSRLALL:
+		err = gve_get_flow_rule_ids(priv, cmd, (u32 *)rule_locs);
+		break;
+	case ETHTOOL_GRXFH:
+		err = -EOPNOTSUPP;
+		break;
+	default:
+		err = -EOPNOTSUPP;
+		break;
+	}
+
+	return err;
+}
+
 const struct ethtool_ops gve_ethtool_ops = {
 	.supported_coalesce_params = ETHTOOL_COALESCE_USECS,
 	.supported_ring_params = ETHTOOL_RING_USE_TCP_DATA_SPLIT,
@@ -786,6 +849,8 @@ const struct ethtool_ops gve_ethtool_ops = {
 	.get_msglevel = gve_get_msglevel,
 	.set_channels = gve_set_channels,
 	.get_channels = gve_get_channels,
+	.set_rxnfc = gve_set_rxnfc,
+	.get_rxnfc = gve_get_rxnfc,
 	.get_link = ethtool_op_get_link,
 	.get_coalesce = gve_get_coalesce,
 	.set_coalesce = gve_set_coalesce,
diff --git a/drivers/net/ethernet/google/gve/gve_flow_rule.c b/drivers/net/ethernet/google/gve/gve_flow_rule.c
new file mode 100644
index 000000000000..0bb8cd1876a3
--- /dev/null
+++ b/drivers/net/ethernet/google/gve/gve_flow_rule.c
@@ -0,0 +1,298 @@
+// SPDX-License-Identifier: (GPL-2.0 OR MIT)
+/* Google virtual Ethernet (gve) driver
+ *
+ * Copyright (C) 2015-2024 Google LLC
+ */
+
+#include "gve.h"
+#include "gve_adminq.h"
+
+static
+int gve_fill_ethtool_flow_spec(struct ethtool_rx_flow_spec *fsp,
+			       struct gve_adminq_queried_flow_rule *rule)
+{
+	struct gve_adminq_flow_rule *flow_rule = &rule->flow_rule;
+	static const u16 flow_type_lut[] = {
+		[GVE_FLOW_TYPE_TCPV4]	= TCP_V4_FLOW,
+		[GVE_FLOW_TYPE_UDPV4]	= UDP_V4_FLOW,
+		[GVE_FLOW_TYPE_SCTPV4]	= SCTP_V4_FLOW,
+		[GVE_FLOW_TYPE_AHV4]	= AH_V4_FLOW,
+		[GVE_FLOW_TYPE_ESPV4]	= ESP_V4_FLOW,
+		[GVE_FLOW_TYPE_TCPV6]	= TCP_V6_FLOW,
+		[GVE_FLOW_TYPE_UDPV6]	= UDP_V6_FLOW,
+		[GVE_FLOW_TYPE_SCTPV6]	= SCTP_V6_FLOW,
+		[GVE_FLOW_TYPE_AHV6]	= AH_V6_FLOW,
+		[GVE_FLOW_TYPE_ESPV6]	= ESP_V6_FLOW,
+	};
+
+	if (be16_to_cpu(flow_rule->flow_type) >= ARRAY_SIZE(flow_type_lut))
+		return -EINVAL;
+
+	fsp->flow_type = flow_type_lut[be16_to_cpu(flow_rule->flow_type)];
+
+	memset(&fsp->h_u, 0, sizeof(fsp->h_u));
+	memset(&fsp->h_ext, 0, sizeof(fsp->h_ext));
+	memset(&fsp->m_u, 0, sizeof(fsp->m_u));
+	memset(&fsp->m_ext, 0, sizeof(fsp->m_ext));
+
+	switch (fsp->flow_type) {
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW:
+	case SCTP_V4_FLOW:
+		fsp->h_u.tcp_ip4_spec.ip4src = flow_rule->key.src_ip[0];
+		fsp->h_u.tcp_ip4_spec.ip4dst = flow_rule->key.dst_ip[0];
+		fsp->h_u.tcp_ip4_spec.psrc = flow_rule->key.src_port;
+		fsp->h_u.tcp_ip4_spec.pdst = flow_rule->key.dst_port;
+		fsp->h_u.tcp_ip4_spec.tos = flow_rule->key.tos;
+		fsp->m_u.tcp_ip4_spec.ip4src = flow_rule->mask.src_ip[0];
+		fsp->m_u.tcp_ip4_spec.ip4dst = flow_rule->mask.dst_ip[0];
+		fsp->m_u.tcp_ip4_spec.psrc = flow_rule->mask.src_port;
+		fsp->m_u.tcp_ip4_spec.pdst = flow_rule->mask.dst_port;
+		fsp->m_u.tcp_ip4_spec.tos = flow_rule->mask.tos;
+		break;
+	case AH_V4_FLOW:
+	case ESP_V4_FLOW:
+		fsp->h_u.ah_ip4_spec.ip4src = flow_rule->key.src_ip[0];
+		fsp->h_u.ah_ip4_spec.ip4dst = flow_rule->key.dst_ip[0];
+		fsp->h_u.ah_ip4_spec.spi = flow_rule->key.spi;
+		fsp->h_u.ah_ip4_spec.tos = flow_rule->key.tos;
+		fsp->m_u.ah_ip4_spec.ip4src = flow_rule->mask.src_ip[0];
+		fsp->m_u.ah_ip4_spec.ip4dst = flow_rule->mask.dst_ip[0];
+		fsp->m_u.ah_ip4_spec.spi = flow_rule->mask.spi;
+		fsp->m_u.ah_ip4_spec.tos = flow_rule->mask.tos;
+		break;
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW:
+	case SCTP_V6_FLOW:
+		memcpy(fsp->h_u.tcp_ip6_spec.ip6src, &flow_rule->key.src_ip,
+		       sizeof(struct in6_addr));
+		memcpy(fsp->h_u.tcp_ip6_spec.ip6dst, &flow_rule->key.dst_ip,
+		       sizeof(struct in6_addr));
+		fsp->h_u.tcp_ip6_spec.psrc = flow_rule->key.src_port;
+		fsp->h_u.tcp_ip6_spec.pdst = flow_rule->key.dst_port;
+		fsp->h_u.tcp_ip6_spec.tclass = flow_rule->key.tclass;
+		memcpy(fsp->m_u.tcp_ip6_spec.ip6src, &flow_rule->mask.src_ip,
+		       sizeof(struct in6_addr));
+		memcpy(fsp->m_u.tcp_ip6_spec.ip6dst, &flow_rule->mask.dst_ip,
+		       sizeof(struct in6_addr));
+		fsp->m_u.tcp_ip6_spec.psrc = flow_rule->mask.src_port;
+		fsp->m_u.tcp_ip6_spec.pdst = flow_rule->mask.dst_port;
+		fsp->m_u.tcp_ip6_spec.tclass = flow_rule->mask.tclass;
+		break;
+	case AH_V6_FLOW:
+	case ESP_V6_FLOW:
+		memcpy(fsp->h_u.ah_ip6_spec.ip6src, &flow_rule->key.src_ip,
+		       sizeof(struct in6_addr));
+		memcpy(fsp->h_u.ah_ip6_spec.ip6dst, &flow_rule->key.dst_ip,
+		       sizeof(struct in6_addr));
+		fsp->h_u.ah_ip6_spec.spi = flow_rule->key.spi;
+		fsp->h_u.ah_ip6_spec.tclass = flow_rule->key.tclass;
+		memcpy(fsp->m_u.ah_ip6_spec.ip6src, &flow_rule->mask.src_ip,
+		       sizeof(struct in6_addr));
+		memcpy(fsp->m_u.ah_ip6_spec.ip6dst, &flow_rule->mask.dst_ip,
+		       sizeof(struct in6_addr));
+		fsp->m_u.ah_ip6_spec.spi = flow_rule->mask.spi;
+		fsp->m_u.ah_ip6_spec.tclass = flow_rule->mask.tclass;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	fsp->ring_cookie = be16_to_cpu(flow_rule->action);
+
+	return 0;
+}
+
+static int gve_generate_flow_rule(struct gve_priv *priv, struct ethtool_rx_flow_spec *fsp,
+				  struct gve_adminq_flow_rule *rule)
+{
+	static const u16 flow_type_lut[] = {
+		[TCP_V4_FLOW]	= GVE_FLOW_TYPE_TCPV4,
+		[UDP_V4_FLOW]	= GVE_FLOW_TYPE_UDPV4,
+		[SCTP_V4_FLOW]	= GVE_FLOW_TYPE_SCTPV4,
+		[AH_V4_FLOW]	= GVE_FLOW_TYPE_AHV4,
+		[ESP_V4_FLOW]	= GVE_FLOW_TYPE_ESPV4,
+		[TCP_V6_FLOW]	= GVE_FLOW_TYPE_TCPV6,
+		[UDP_V6_FLOW]	= GVE_FLOW_TYPE_UDPV6,
+		[SCTP_V6_FLOW]	= GVE_FLOW_TYPE_SCTPV6,
+		[AH_V6_FLOW]	= GVE_FLOW_TYPE_AHV6,
+		[ESP_V6_FLOW]	= GVE_FLOW_TYPE_ESPV6,
+	};
+	u32 flow_type;
+
+	if (fsp->ring_cookie == RX_CLS_FLOW_DISC)
+		return -EOPNOTSUPP;
+
+	if (fsp->ring_cookie >= priv->rx_cfg.num_queues)
+		return -EINVAL;
+
+	rule->action = cpu_to_be16(fsp->ring_cookie);
+
+	flow_type = fsp->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT | FLOW_RSS);
+	if (!flow_type || flow_type >= ARRAY_SIZE(flow_type_lut))
+		return -EINVAL;
+
+	rule->flow_type = cpu_to_be16(flow_type_lut[flow_type]);
+
+	switch (flow_type) {
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW:
+	case SCTP_V4_FLOW:
+		rule->key.src_ip[0] = fsp->h_u.tcp_ip4_spec.ip4src;
+		rule->key.dst_ip[0] = fsp->h_u.tcp_ip4_spec.ip4dst;
+		rule->key.src_port = fsp->h_u.tcp_ip4_spec.psrc;
+		rule->key.dst_port = fsp->h_u.tcp_ip4_spec.pdst;
+		rule->mask.src_ip[0] = fsp->m_u.tcp_ip4_spec.ip4src;
+		rule->mask.dst_ip[0] = fsp->m_u.tcp_ip4_spec.ip4dst;
+		rule->mask.src_port = fsp->m_u.tcp_ip4_spec.psrc;
+		rule->mask.dst_port = fsp->m_u.tcp_ip4_spec.pdst;
+		break;
+	case AH_V4_FLOW:
+	case ESP_V4_FLOW:
+		rule->key.src_ip[0] = fsp->h_u.tcp_ip4_spec.ip4src;
+		rule->key.dst_ip[0] = fsp->h_u.tcp_ip4_spec.ip4dst;
+		rule->key.spi = fsp->h_u.ah_ip4_spec.spi;
+		rule->mask.src_ip[0] = fsp->m_u.tcp_ip4_spec.ip4src;
+		rule->mask.dst_ip[0] = fsp->m_u.tcp_ip4_spec.ip4dst;
+		rule->mask.spi = fsp->m_u.ah_ip4_spec.spi;
+		break;
+	case TCP_V6_FLOW:
+	case UDP_V6_FLOW:
+	case SCTP_V6_FLOW:
+		memcpy(&rule->key.src_ip, fsp->h_u.tcp_ip6_spec.ip6src,
+		       sizeof(struct in6_addr));
+		memcpy(&rule->key.dst_ip, fsp->h_u.tcp_ip6_spec.ip6dst,
+		       sizeof(struct in6_addr));
+		rule->key.src_port = fsp->h_u.tcp_ip6_spec.psrc;
+		rule->key.dst_port = fsp->h_u.tcp_ip6_spec.pdst;
+		memcpy(&rule->mask.src_ip, fsp->m_u.tcp_ip6_spec.ip6src,
+		       sizeof(struct in6_addr));
+		memcpy(&rule->mask.dst_ip, fsp->m_u.tcp_ip6_spec.ip6dst,
+		       sizeof(struct in6_addr));
+		rule->mask.src_port = fsp->m_u.tcp_ip6_spec.psrc;
+		rule->mask.dst_port = fsp->m_u.tcp_ip6_spec.pdst;
+		break;
+	case AH_V6_FLOW:
+	case ESP_V6_FLOW:
+		memcpy(&rule->key.src_ip, fsp->h_u.usr_ip6_spec.ip6src,
+		       sizeof(struct in6_addr));
+		memcpy(&rule->key.dst_ip, fsp->h_u.usr_ip6_spec.ip6dst,
+		       sizeof(struct in6_addr));
+		rule->key.spi = fsp->h_u.ah_ip6_spec.spi;
+		memcpy(&rule->mask.src_ip, fsp->m_u.usr_ip6_spec.ip6src,
+		       sizeof(struct in6_addr));
+		memcpy(&rule->mask.dst_ip, fsp->m_u.usr_ip6_spec.ip6dst,
+		       sizeof(struct in6_addr));
+		rule->key.spi = fsp->h_u.ah_ip6_spec.spi;
+		break;
+	default:
+		/* not doing un-parsed flow types */
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int gve_get_flow_rule_entry(struct gve_priv *priv, struct ethtool_rxnfc *cmd)
+{
+	struct gve_adminq_queried_flow_rule *rules_cache = priv->flow_rules_cache.rules_cache;
+	struct ethtool_rx_flow_spec *fsp = (struct ethtool_rx_flow_spec *)&cmd->fs;
+	u32 *cache_num = &priv->flow_rules_cache.rules_cache_num;
+	struct gve_adminq_queried_flow_rule *rule = NULL;
+	int err = 0;
+	u32 i;
+
+	if (!priv->max_flow_rules)
+		return -EOPNOTSUPP;
+
+	if (!priv->flow_rules_cache.rules_cache_synced ||
+	    fsp->location < be32_to_cpu(rules_cache[0].location) ||
+	    fsp->location > be32_to_cpu(rules_cache[*cache_num - 1].location)) {
+		err = gve_adminq_query_flow_rules(priv, GVE_FLOW_RULE_QUERY_RULES, fsp->location);
+		if (err)
+			return err;
+
+		priv->flow_rules_cache.rules_cache_synced = true;
+	}
+
+	for (i = 0; i < *cache_num; i++) {
+		if (fsp->location == be32_to_cpu(rules_cache[i].location)) {
+			rule = &rules_cache[i];
+			break;
+		}
+	}
+
+	if (!rule)
+		return -EINVAL;
+
+	err = gve_fill_ethtool_flow_spec(fsp, rule);
+
+	return err;
+}
+
+int gve_get_flow_rule_ids(struct gve_priv *priv, struct ethtool_rxnfc *cmd, u32 *rule_locs)
+{
+	__be32 *rule_ids_cache = priv->flow_rules_cache.rule_ids_cache;
+	u32 *cache_num = &priv->flow_rules_cache.rule_ids_cache_num;
+	u32 starting_rule_id = 0;
+	u32 i = 0, j = 0;
+	int err = 0;
+
+	if (!priv->max_flow_rules)
+		return -EOPNOTSUPP;
+
+	do {
+		err = gve_adminq_query_flow_rules(priv, GVE_FLOW_RULE_QUERY_IDS,
+						  starting_rule_id);
+		if (err)
+			return err;
+
+		for (i = 0; i < *cache_num; i++) {
+			if (j >= cmd->rule_cnt)
+				return -EMSGSIZE;
+
+			rule_locs[j++] = be32_to_cpu(rule_ids_cache[i]);
+			starting_rule_id = be32_to_cpu(rule_ids_cache[i]) + 1;
+		}
+	} while (*cache_num != 0);
+	cmd->data = priv->max_flow_rules;
+
+	return err;
+}
+
+int gve_add_flow_rule(struct gve_priv *priv, struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fsp = &cmd->fs;
+	struct gve_adminq_flow_rule *rule = NULL;
+	int err;
+
+	if (!priv->max_flow_rules)
+		return -EOPNOTSUPP;
+
+	rule = kvzalloc(sizeof(*rule), GFP_KERNEL);
+	if (!rule)
+		return -ENOMEM;
+
+	err = gve_generate_flow_rule(priv, fsp, rule);
+	if (err)
+		goto out;
+
+	err = gve_adminq_add_flow_rule(priv, rule, fsp->location);
+
+out:
+	kvfree(rule);
+	if (err)
+		dev_err(&priv->pdev->dev, "Failed to add the flow rule: %u", fsp->location);
+
+	return err;
+}
+
+int gve_del_flow_rule(struct gve_priv *priv, struct ethtool_rxnfc *cmd)
+{
+	struct ethtool_rx_flow_spec *fsp = (struct ethtool_rx_flow_spec *)&cmd->fs;
+
+	if (!priv->max_flow_rules)
+		return -EOPNOTSUPP;
+
+	return gve_adminq_del_flow_rule(priv, fsp->location);
+}
diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c
index fc142856e189..9744b426940e 100644
--- a/drivers/net/ethernet/google/gve/gve_main.c
+++ b/drivers/net/ethernet/google/gve/gve_main.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: (GPL-2.0 OR MIT)
 /* Google virtual Ethernet (gve) driver
  *
- * Copyright (C) 2015-2021 Google, Inc.
+ * Copyright (C) 2015-2024 Google LLC
  */
 
 #include <linux/bpf.h>
@@ -635,6 +635,12 @@ static void gve_teardown_device_resources(struct gve_priv *priv)
 
 	/* Tell device its resources are being freed */
 	if (gve_get_device_resources_ok(priv)) {
+		err = gve_flow_rules_reset(priv);
+		if (err) {
+			dev_err(&priv->pdev->dev,
+				"Failed to reset flow rules: err=%d\n", err);
+			gve_trigger_reset(priv);
+		}
 		/* detach the stats report */
 		err = gve_adminq_report_stats(priv, 0, 0x0, GVE_STATS_REPORT_TIMER_PERIOD);
 		if (err) {
@@ -1779,6 +1785,14 @@ static int gve_xdp(struct net_device *dev, struct netdev_bpf *xdp)
 	}
 }
 
+int gve_flow_rules_reset(struct gve_priv *priv)
+{
+	if (!priv->max_flow_rules)
+		return 0;
+
+	return gve_adminq_reset_flow_rules(priv);
+}
+
 int gve_adjust_config(struct gve_priv *priv,
 		      struct gve_tx_alloc_rings_cfg *tx_alloc_cfg,
 		      struct gve_rx_alloc_rings_cfg *rx_alloc_cfg)
@@ -2052,15 +2066,21 @@ static int gve_set_features(struct net_device *netdev,
 		netdev->features ^= NETIF_F_LRO;
 		if (netif_carrier_ok(netdev)) {
 			err = gve_adjust_config(priv, &tx_alloc_cfg, &rx_alloc_cfg);
-			if (err) {
-				/* Revert the change on error. */
-				netdev->features = orig_features;
-				return err;
-			}
+			if (err)
+				goto revert_features;
 		}
 	}
+	if ((netdev->features & NETIF_F_NTUPLE) && !(features & NETIF_F_NTUPLE)) {
+		err = gve_flow_rules_reset(priv);
+		if (err)
+			goto revert_features;
+	}
 
 	return 0;
+
+revert_features:
+	netdev->features = orig_features;
+	return err;
 }
 
 static const struct net_device_ops gve_netdev_ops = {
-- 
cgit v1.2.3-58-ga151


From c0c68e4d52ed7fdecec7dab05cd5b8540ce4813e Mon Sep 17 00:00:00 2001
From: Chris Packham <chris.packham@alliedtelesis.co.nz>
Date: Tue, 25 Jun 2024 09:18:57 +1200
Subject: dt-bindings: net: dsa: mediatek,mt7530: Minor wording fixes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Update the mt7530 binding with some minor updates that make the document
easier to read.

Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Acked-by: Arınç ÜNAL <arinc.unal@arinc9.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240624211858.1990601-1-chris.packham@alliedtelesis.co.nz
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml b/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml
index 1c2444121e60..7e405ad96eb2 100644
--- a/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml
+++ b/Documentation/devicetree/bindings/net/dsa/mediatek,mt7530.yaml
@@ -22,16 +22,16 @@ description: |
 
   The MT7988 SoC comes with a built-in switch similar to MT7531 as well as four
   Gigabit Ethernet PHYs. The switch registers are directly mapped into the SoC's
-  memory map rather than using MDIO. The switch got an internally connected 10G
+  memory map rather than using MDIO. The switch has an internally connected 10G
   CPU port and 4 user ports connected to the built-in Gigabit Ethernet PHYs.
 
-  MT7530 in MT7620AN, MT7620DA, MT7620DAN and MT7620NN SoCs has got 10/100 PHYs
+  The MT7530 in MT7620AN, MT7620DA, MT7620DAN and MT7620NN SoCs has 10/100 PHYs
   and the switch registers are directly mapped into SoC's memory map rather than
   using MDIO. The DSA driver currently doesn't support MT7620 variants.
 
   There is only the standalone version of MT7531.
 
-  Port 5 on MT7530 has got various ways of configuration:
+  Port 5 on MT7530 supports various configurations:
 
     - Port 5 can be used as a CPU port.
 
-- 
cgit v1.2.3-58-ga151


From d65f3767de20782e75d8a665fdc54f822f344802 Mon Sep 17 00:00:00 2001
From: Leon Hwang <hffilwlqm@gmail.com>
Date: Tue, 25 Jun 2024 22:53:51 +0800
Subject: bpf: Fix tailcall cases in test_bpf

Since f663a03c8e35 ("bpf, x64: Remove tail call detection"),
tail_call_reachable won't be detected in x86 JIT. And, tail_call_reachable
is provided by verifier.

Therefore, in test_bpf, the tail_call_reachable must be provided in test
cases before running.

Fix and test:

[  174.828662] test_bpf: #0 Tail call leaf jited:1 170 PASS
[  174.829574] test_bpf: #1 Tail call 2 jited:1 244 PASS
[  174.830363] test_bpf: #2 Tail call 3 jited:1 296 PASS
[  174.830924] test_bpf: #3 Tail call 4 jited:1 719 PASS
[  174.831863] test_bpf: #4 Tail call load/store leaf jited:1 197 PASS
[  174.832240] test_bpf: #5 Tail call load/store jited:1 326 PASS
[  174.832240] test_bpf: #6 Tail call error path, max count reached jited:1 2214 PASS
[  174.835713] test_bpf: #7 Tail call count preserved across function calls jited:1 609751 PASS
[  175.446098] test_bpf: #8 Tail call error path, NULL target jited:1 472 PASS
[  175.447597] test_bpf: #9 Tail call error path, index out of range jited:1 206 PASS
[  175.448833] test_bpf: test_tail_calls: Summary: 10 PASSED, 0 FAILED, [10/10 JIT'ed]

Reported-by: kernel test robot <oliver.sang@intel.com>
Closes: https://lore.kernel.org/oe-lkp/202406251415.c51865bc-oliver.sang@intel.com
Fixes: f663a03c8e35 ("bpf, x64: Remove tail call detection")
Signed-off-by: Leon Hwang <hffilwlqm@gmail.com>
Link: https://lore.kernel.org/r/20240625145351.40072-1-hffilwlqm@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 lib/test_bpf.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/lib/test_bpf.c b/lib/test_bpf.c
index ce5716c3999a..b7acc29bcc3b 100644
--- a/lib/test_bpf.c
+++ b/lib/test_bpf.c
@@ -15198,6 +15198,7 @@ struct tail_call_test {
 	int flags;
 	int result;
 	int stack_depth;
+	bool has_tail_call;
 };
 
 /* Flags that can be passed to tail call test cases */
@@ -15273,6 +15274,7 @@ static struct tail_call_test tail_call_tests[] = {
 			BPF_EXIT_INSN(),
 		},
 		.result = 3,
+		.has_tail_call = true,
 	},
 	{
 		"Tail call 3",
@@ -15283,6 +15285,7 @@ static struct tail_call_test tail_call_tests[] = {
 			BPF_EXIT_INSN(),
 		},
 		.result = 6,
+		.has_tail_call = true,
 	},
 	{
 		"Tail call 4",
@@ -15293,6 +15296,7 @@ static struct tail_call_test tail_call_tests[] = {
 			BPF_EXIT_INSN(),
 		},
 		.result = 10,
+		.has_tail_call = true,
 	},
 	{
 		"Tail call load/store leaf",
@@ -15323,6 +15327,7 @@ static struct tail_call_test tail_call_tests[] = {
 		},
 		.result = 0,
 		.stack_depth = 16,
+		.has_tail_call = true,
 	},
 	{
 		"Tail call error path, max count reached",
@@ -15335,6 +15340,7 @@ static struct tail_call_test tail_call_tests[] = {
 		},
 		.flags = FLAG_NEED_STATE | FLAG_RESULT_IN_STATE,
 		.result = (MAX_TAIL_CALL_CNT + 1) * MAX_TESTRUNS,
+		.has_tail_call = true,
 	},
 	{
 		"Tail call count preserved across function calls",
@@ -15357,6 +15363,7 @@ static struct tail_call_test tail_call_tests[] = {
 		.stack_depth = 8,
 		.flags = FLAG_NEED_STATE | FLAG_RESULT_IN_STATE,
 		.result = (MAX_TAIL_CALL_CNT + 1) * MAX_TESTRUNS,
+		.has_tail_call = true,
 	},
 	{
 		"Tail call error path, NULL target",
@@ -15369,6 +15376,7 @@ static struct tail_call_test tail_call_tests[] = {
 		},
 		.flags = FLAG_NEED_STATE | FLAG_RESULT_IN_STATE,
 		.result = MAX_TESTRUNS,
+		.has_tail_call = true,
 	},
 	{
 		"Tail call error path, index out of range",
@@ -15381,6 +15389,7 @@ static struct tail_call_test tail_call_tests[] = {
 		},
 		.flags = FLAG_NEED_STATE | FLAG_RESULT_IN_STATE,
 		.result = MAX_TESTRUNS,
+		.has_tail_call = true,
 	},
 };
 
@@ -15430,6 +15439,7 @@ static __init int prepare_tail_call_tests(struct bpf_array **pprogs)
 		fp->len = len;
 		fp->type = BPF_PROG_TYPE_SOCKET_FILTER;
 		fp->aux->stack_depth = test->stack_depth;
+		fp->aux->tail_call_reachable = test->has_tail_call;
 		memcpy(fp->insnsi, test->insns, len * sizeof(struct bpf_insn));
 
 		/* Relocate runtime tail call offsets and addresses */
-- 
cgit v1.2.3-58-ga151


From eee5528890d54b22b46f833002355a5ee94c3bb4 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@gmail.com>
Date: Mon, 24 Jun 2024 08:55:01 +0900
Subject: PCI: Add Edimax Vendor ID to pci_ids.h

Add the Edimax Vendor ID (0x1432) for an ethernet driver for Tehuti
Networks TN40xx chips. This ID can be used for Realtek 8180 and Ralink
rt28xx wireless drivers.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@gmail.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://patch.msgid.link/20240623235507.108147-2-fujita.tomonori@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/pci_ids.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 942a587bb97e..677aea20d3e1 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2126,6 +2126,8 @@
 
 #define PCI_VENDOR_ID_CHELSIO		0x1425
 
+#define PCI_VENDOR_ID_EDIMAX		0x1432
+
 #define PCI_VENDOR_ID_ADLINK		0x144a
 
 #define PCI_VENDOR_ID_SAMSUNG		0x144d
-- 
cgit v1.2.3-58-ga151


From ab61adc60001c83c3130107ca25737addf42fd6f Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@gmail.com>
Date: Mon, 24 Jun 2024 08:55:02 +0900
Subject: net: tn40xx: add pci driver for Tehuti Networks TN40xx chips

This just adds the scaffolding for an ethernet driver for Tehuti
Networks TN40xx chips.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@gmail.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Link: https://patch.msgid.link/20240623235507.108147-3-fujita.tomonori@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 MAINTAINERS                          |  8 +++++-
 drivers/net/ethernet/tehuti/Kconfig  | 12 ++++++++
 drivers/net/ethernet/tehuti/Makefile |  3 ++
 drivers/net/ethernet/tehuti/tn40.c   | 54 ++++++++++++++++++++++++++++++++++++
 drivers/net/ethernet/tehuti/tn40.h   |  9 ++++++
 5 files changed, 85 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/ethernet/tehuti/tn40.c
 create mode 100644 drivers/net/ethernet/tehuti/tn40.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 9283dd04cc6e..571c52b43cfb 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -22148,7 +22148,13 @@ TEHUTI ETHERNET DRIVER
 M:	Andy Gospodarek <andy@greyhouse.net>
 L:	netdev@vger.kernel.org
 S:	Supported
-F:	drivers/net/ethernet/tehuti/*
+F:	drivers/net/ethernet/tehuti/tehuti.*
+
+TEHUTI TN40XX ETHERNET DRIVER
+M:	FUJITA Tomonori <fujita.tomonori@gmail.com>
+L:	netdev@vger.kernel.org
+S:	Maintained
+F:	drivers/net/ethernet/tehuti/tn40*
 
 TELECOM CLOCK DRIVER FOR MCPL0010
 M:	Mark Gross <markgross@kernel.org>
diff --git a/drivers/net/ethernet/tehuti/Kconfig b/drivers/net/ethernet/tehuti/Kconfig
index 8735633765a1..849e3b4a71c1 100644
--- a/drivers/net/ethernet/tehuti/Kconfig
+++ b/drivers/net/ethernet/tehuti/Kconfig
@@ -23,4 +23,16 @@ config TEHUTI
 	help
 	  Tehuti Networks 10G Ethernet NIC
 
+config TEHUTI_TN40
+	tristate "Tehuti Networks TN40xx 10G Ethernet adapters"
+	depends on PCI
+	help
+	  This driver supports 10G Ethernet adapters using Tehuti Networks
+	  TN40xx chips. Currently, adapters with Applied Micro Circuits
+	  Corporation QT2025 are supported; Tehuti Networks TN9310,
+	  DLink DXE-810S, ASUS XG-C100F, and Edimax EN-9320.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called tn40xx.
+
 endif # NET_VENDOR_TEHUTI
diff --git a/drivers/net/ethernet/tehuti/Makefile b/drivers/net/ethernet/tehuti/Makefile
index 13a0ddd62088..1c468d99e476 100644
--- a/drivers/net/ethernet/tehuti/Makefile
+++ b/drivers/net/ethernet/tehuti/Makefile
@@ -4,3 +4,6 @@
 #
 
 obj-$(CONFIG_TEHUTI) += tehuti.o
+
+tn40xx-y := tn40.o
+obj-$(CONFIG_TEHUTI_TN40) += tn40xx.o
diff --git a/drivers/net/ethernet/tehuti/tn40.c b/drivers/net/ethernet/tehuti/tn40.c
new file mode 100644
index 000000000000..248b17366680
--- /dev/null
+++ b/drivers/net/ethernet/tehuti/tn40.c
@@ -0,0 +1,54 @@
+// SPDX-License-Identifier: GPL-2.0+
+/* Copyright (c) Tehuti Networks Ltd. */
+
+#include <linux/pci.h>
+
+#include "tn40.h"
+
+static int tn40_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	int ret;
+
+	ret = pci_enable_device(pdev);
+	if (ret)
+		return ret;
+	ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+	if (ret) {
+		dev_err(&pdev->dev, "failed to set DMA mask.\n");
+		goto err_disable_device;
+	}
+	return 0;
+err_disable_device:
+	pci_disable_device(pdev);
+	return ret;
+}
+
+static void tn40_remove(struct pci_dev *pdev)
+{
+	pci_disable_device(pdev);
+}
+
+static const struct pci_device_id tn40_id_table[] = {
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_TEHUTI, 0x4022,
+			 PCI_VENDOR_ID_TEHUTI, 0x3015) },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_TEHUTI, 0x4022,
+			 PCI_VENDOR_ID_DLINK, 0x4d00) },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_TEHUTI, 0x4022,
+			 PCI_VENDOR_ID_ASUSTEK, 0x8709) },
+	{ PCI_DEVICE_SUB(PCI_VENDOR_ID_TEHUTI, 0x4022,
+			 PCI_VENDOR_ID_EDIMAX, 0x8103) },
+	{ }
+};
+
+static struct pci_driver tn40_driver = {
+	.name = TN40_DRV_NAME,
+	.id_table = tn40_id_table,
+	.probe = tn40_probe,
+	.remove = tn40_remove,
+};
+
+module_pci_driver(tn40_driver);
+
+MODULE_DEVICE_TABLE(pci, tn40_id_table);
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Tehuti Network TN40xx Driver");
diff --git a/drivers/net/ethernet/tehuti/tn40.h b/drivers/net/ethernet/tehuti/tn40.h
new file mode 100644
index 000000000000..b45a2eef2850
--- /dev/null
+++ b/drivers/net/ethernet/tehuti/tn40.h
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) Tehuti Networks Ltd. */
+
+#ifndef _TN40_H_
+#define _TN40_H_
+
+#define TN40_DRV_NAME "tn40xx"
+
+#endif /* _TN40XX_H */
-- 
cgit v1.2.3-58-ga151


From ffa28c748b388b8afaf658df15a6a1ed6d310663 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@gmail.com>
Date: Mon, 24 Jun 2024 08:55:03 +0900
Subject: net: tn40xx: add register defines

This adds several defines to handle registers in Tehuti Networks
TN40xx chips for later patches.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@gmail.com>
Reviewed-by: Hans-Frieder Vogt <hfdevel@gmx.net>
Link: https://patch.msgid.link/20240623235507.108147-4-fujita.tomonori@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/tehuti/tn40.h      |   2 +
 drivers/net/ethernet/tehuti/tn40_regs.h | 245 ++++++++++++++++++++++++++++++++
 2 files changed, 247 insertions(+)
 create mode 100644 drivers/net/ethernet/tehuti/tn40_regs.h

diff --git a/drivers/net/ethernet/tehuti/tn40.h b/drivers/net/ethernet/tehuti/tn40.h
index b45a2eef2850..aec23208e9b1 100644
--- a/drivers/net/ethernet/tehuti/tn40.h
+++ b/drivers/net/ethernet/tehuti/tn40.h
@@ -4,6 +4,8 @@
 #ifndef _TN40_H_
 #define _TN40_H_
 
+#include "tn40_regs.h"
+
 #define TN40_DRV_NAME "tn40xx"
 
 #endif /* _TN40XX_H */
diff --git a/drivers/net/ethernet/tehuti/tn40_regs.h b/drivers/net/ethernet/tehuti/tn40_regs.h
new file mode 100644
index 000000000000..95171aa57a9e
--- /dev/null
+++ b/drivers/net/ethernet/tehuti/tn40_regs.h
@@ -0,0 +1,245 @@
+/* SPDX-License-Identifier: GPL-2.0+ */
+/* Copyright (c) Tehuti Networks Ltd. */
+
+#ifndef _TN40_REGS_H_
+#define _TN40_REGS_H_
+
+/* Register region size */
+#define TN40_REGS_SIZE 0x10000
+
+/* Registers from 0x0000-0x00fc were remapped to 0x4000-0x40fc */
+#define TN40_REG_TXD_CFG1_0 0x4000
+#define TN40_REG_TXD_CFG1_1 0x4004
+#define TN40_REG_TXD_CFG1_2 0x4008
+#define TN40_REG_TXD_CFG1_3 0x400C
+
+#define TN40_REG_RXF_CFG1_0 0x4010
+#define TN40_REG_RXF_CFG1_1 0x4014
+#define TN40_REG_RXF_CFG1_2 0x4018
+#define TN40_REG_RXF_CFG1_3 0x401C
+
+#define TN40_REG_RXD_CFG1_0 0x4020
+#define TN40_REG_RXD_CFG1_1 0x4024
+#define TN40_REG_RXD_CFG1_2 0x4028
+#define TN40_REG_RXD_CFG1_3 0x402C
+
+#define TN40_REG_TXF_CFG1_0 0x4030
+#define TN40_REG_TXF_CFG1_1 0x4034
+#define TN40_REG_TXF_CFG1_2 0x4038
+#define TN40_REG_TXF_CFG1_3 0x403C
+
+#define TN40_REG_TXD_CFG0_0 0x4040
+#define TN40_REG_TXD_CFG0_1 0x4044
+#define TN40_REG_TXD_CFG0_2 0x4048
+#define TN40_REG_TXD_CFG0_3 0x404C
+
+#define TN40_REG_RXF_CFG0_0 0x4050
+#define TN40_REG_RXF_CFG0_1 0x4054
+#define TN40_REG_RXF_CFG0_2 0x4058
+#define TN40_REG_RXF_CFG0_3 0x405C
+
+#define TN40_REG_RXD_CFG0_0 0x4060
+#define TN40_REG_RXD_CFG0_1 0x4064
+#define TN40_REG_RXD_CFG0_2 0x4068
+#define TN40_REG_RXD_CFG0_3 0x406C
+
+#define TN40_REG_TXF_CFG0_0 0x4070
+#define TN40_REG_TXF_CFG0_1 0x4074
+#define TN40_REG_TXF_CFG0_2 0x4078
+#define TN40_REG_TXF_CFG0_3 0x407C
+
+#define TN40_REG_TXD_WPTR_0 0x4080
+#define TN40_REG_TXD_WPTR_1 0x4084
+#define TN40_REG_TXD_WPTR_2 0x4088
+#define TN40_REG_TXD_WPTR_3 0x408C
+
+#define TN40_REG_RXF_WPTR_0 0x4090
+#define TN40_REG_RXF_WPTR_1 0x4094
+#define TN40_REG_RXF_WPTR_2 0x4098
+#define TN40_REG_RXF_WPTR_3 0x409C
+
+#define TN40_REG_RXD_WPTR_0 0x40A0
+#define TN40_REG_RXD_WPTR_1 0x40A4
+#define TN40_REG_RXD_WPTR_2 0x40A8
+#define TN40_REG_RXD_WPTR_3 0x40AC
+
+#define TN40_REG_TXF_WPTR_0 0x40B0
+#define TN40_REG_TXF_WPTR_1 0x40B4
+#define TN40_REG_TXF_WPTR_2 0x40B8
+#define TN40_REG_TXF_WPTR_3 0x40BC
+
+#define TN40_REG_TXD_RPTR_0 0x40C0
+#define TN40_REG_TXD_RPTR_1 0x40C4
+#define TN40_REG_TXD_RPTR_2 0x40C8
+#define TN40_REG_TXD_RPTR_3 0x40CC
+
+#define TN40_REG_RXF_RPTR_0 0x40D0
+#define TN40_REG_RXF_RPTR_1 0x40D4
+#define TN40_REG_RXF_RPTR_2 0x40D8
+#define TN40_REG_RXF_RPTR_3 0x40DC
+
+#define TN40_REG_RXD_RPTR_0 0x40E0
+#define TN40_REG_RXD_RPTR_1 0x40E4
+#define TN40_REG_RXD_RPTR_2 0x40E8
+#define TN40_REG_RXD_RPTR_3 0x40EC
+
+#define TN40_REG_TXF_RPTR_0 0x40F0
+#define TN40_REG_TXF_RPTR_1 0x40F4
+#define TN40_REG_TXF_RPTR_2 0x40F8
+#define TN40_REG_TXF_RPTR_3 0x40FC
+
+/* Hardware versioning */
+#define TN40_FPGA_VER 0x5030
+
+/* Registers from 0x0100-0x0150 were remapped to 0x5100-0x5150 */
+#define TN40_REG_ISR TN40_REG_ISR0
+#define TN40_REG_ISR0 0x5100
+
+#define TN40_REG_IMR TN40_REG_IMR0
+#define TN40_REG_IMR0 0x5110
+
+#define TN40_REG_RDINTCM0 0x5120
+#define TN40_REG_RDINTCM2 0x5128
+
+#define TN40_REG_TDINTCM0 0x5130
+
+#define TN40_REG_ISR_MSK0 0x5140
+
+#define TN40_REG_INIT_SEMAPHORE 0x5170
+#define TN40_REG_INIT_STATUS 0x5180
+
+#define TN40_REG_MAC_LNK_STAT 0x0200
+#define TN40_MAC_LINK_STAT 0x0004 /* Link state */
+
+#define TN40_REG_BLNK_LED 0x0210
+
+#define TN40_REG_GMAC_RXF_A 0x1240
+
+#define TN40_REG_UNC_MAC0_A 0x1250
+#define TN40_REG_UNC_MAC1_A 0x1260
+#define TN40_REG_UNC_MAC2_A 0x1270
+
+#define TN40_REG_VLAN_0 0x1800
+
+#define TN40_REG_MAX_FRAME_A 0x12C0
+
+#define TN40_REG_RX_MAC_MCST0 0x1A80
+#define TN40_REG_RX_MAC_MCST1 0x1A84
+#define TN40_MAC_MCST_NUM 15
+#define TN40_REG_RX_MCST_HASH0 0x1A00
+#define TN40_MAC_MCST_HASH_NUM 8
+
+#define TN40_REG_VPC 0x2300
+#define TN40_REG_VIC 0x2320
+#define TN40_REG_VGLB 0x2340
+
+#define TN40_REG_CLKPLL 0x5000
+
+/* MDIO interface */
+
+#define TN40_REG_MDIO_CMD_STAT 0x6030
+#define TN40_REG_MDIO_CMD 0x6034
+#define TN40_REG_MDIO_DATA 0x6038
+#define TN40_REG_MDIO_ADDR 0x603C
+#define TN40_GET_MDIO_BUSY(x) FIELD_GET(GENMASK(0, 0), (x))
+#define TN40_GET_MDIO_RD_ERR(x) FIELD_GET(GENMASK(1, 1), (x))
+
+#define TN40_REG_REVISION 0x6000
+#define TN40_REG_SCRATCH 0x6004
+#define TN40_REG_CTRLST 0x6008
+#define TN40_REG_MAC_ADDR_0 0x600C
+#define TN40_REG_MAC_ADDR_1 0x6010
+#define TN40_REG_FRM_LENGTH 0x6014
+#define TN40_REG_PAUSE_QUANT 0x6054
+#define TN40_REG_RX_FIFO_SECTION 0x601C
+#define TN40_REG_TX_FIFO_SECTION 0x6020
+#define TN40_REG_RX_FULLNESS 0x6024
+#define TN40_REG_TX_FULLNESS 0x6028
+#define TN40_REG_HASHTABLE 0x602C
+
+#define TN40_REG_RST_PORT 0x7000
+#define TN40_REG_DIS_PORT 0x7010
+#define TN40_REG_RST_QU 0x7020
+#define TN40_REG_DIS_QU 0x7030
+
+#define TN40_REG_CTRLST_TX_ENA 0x0001
+#define TN40_REG_CTRLST_RX_ENA 0x0002
+#define TN40_REG_CTRLST_PRM_ENA 0x0010
+#define TN40_REG_CTRLST_PAD_ENA 0x0020
+
+#define TN40_REG_CTRLST_BASE (TN40_REG_CTRLST_PAD_ENA | REG_CTRLST_PRM_ENA)
+
+/* TXD TXF RXF RXD  CONFIG 0x0000 --- 0x007c */
+#define TN40_TX_RX_CFG1_BASE 0xffffffff /*0-31 */
+#define TN40_TX_RX_CFG0_BASE 0xfffff000 /*31:12 */
+#define TN40_TX_RX_CFG0_RSVD 0x00000ffc /*11:2 */
+#define TN40_TX_RX_CFG0_SIZE 0x00000003 /*1:0 */
+
+/* TXD TXF RXF RXD  WRITE 0x0080 --- 0x00BC */
+#define TN40_TXF_WPTR_WR_PTR 0x00007ff8 /*14:3 */
+
+/* TXD TXF RXF RXD  READ  0x00CO --- 0x00FC */
+#define TN40_TXF_RPTR_RD_PTR 0x00007ff8 /*14:3 */
+
+/* The last 4 bits are dropped size is rounded to 16 */
+#define TN40_TXF_WPTR_MASK 0x7ff0
+
+/* regISR 0x0100 */
+/* regIMR 0x0110 */
+#define TN40_IMR_INPROG 0x80000000 /*31 */
+#define TN40_IR_LNKCHG1 0x10000000 /*28 */
+#define TN40_IR_LNKCHG0 0x08000000 /*27 */
+#define TN40_IR_GPIO 0x04000000 /*26 */
+#define TN40_IR_RFRSH 0x02000000 /*25 */
+#define TN40_IR_RSVD 0x01000000 /*24 */
+#define TN40_IR_SWI 0x00800000 /*23 */
+#define TN40_IR_RX_FREE_3 0x00400000 /*22 */
+#define TN40_IR_RX_FREE_2 0x00200000 /*21 */
+#define TN40_IR_RX_FREE_1 0x00100000 /*20 */
+#define TN40_IR_RX_FREE_0 0x00080000 /*19 */
+#define TN40_IR_TX_FREE_3 0x00040000 /*18 */
+#define TN40_IR_TX_FREE_2 0x00020000 /*17 */
+#define TN40_IR_TX_FREE_1 0x00010000 /*16 */
+#define TN40_IR_TX_FREE_0 0x00008000 /*15 */
+#define TN40_IR_RX_DESC_3 0x00004000 /*14 */
+#define TN40_IR_RX_DESC_2 0x00002000 /*13 */
+#define TN40_IR_RX_DESC_1 0x00001000 /*12 */
+#define TN40_IR_RX_DESC_0 0x00000800 /*11 */
+#define TN40_IR_PSE 0x00000400 /*10 */
+#define TN40_IR_TMR3 0x00000200 /* 9 */
+#define TN40_IR_TMR2 0x00000100 /* 8 */
+#define TN40_IR_TMR1 0x00000080 /* 7 */
+#define TN40_IR_TMR0 0x00000040 /* 6 */
+#define TN40_IR_VNT 0x00000020 /* 5 */
+#define TN40_IR_RxFL 0x00000010 /* 4 */
+#define TN40_IR_SDPERR 0x00000008 /* 3 */
+#define TN40_IR_TR 0x00000004 /* 2 */
+#define TN40_IR_PCIE_LINK 0x00000002 /* 1 */
+#define TN40_IR_PCIE_TOUT 0x00000001 /* 0 */
+
+#define TN40_IR_EXTRA						\
+	(TN40_IR_RX_FREE_0 | TN40_IR_LNKCHG0 | TN40_IR_LNKCHG1 |\
+	TN40_IR_PSE | TN40_IR_TMR0 | TN40_IR_PCIE_LINK |	\
+	TN40_IR_PCIE_TOUT)
+
+#define TN40_GMAC_RX_FILTER_OSEN 0x1000 /* shared OS enable */
+#define TN40_GMAC_RX_FILTER_TXFC 0x0400 /* Tx flow control */
+#define TN40_GMAC_RX_FILTER_RSV0 0x0200 /* reserved */
+#define TN40_GMAC_RX_FILTER_FDA 0x0100 /* filter out direct address */
+#define TN40_GMAC_RX_FILTER_AOF 0x0080 /* accept over run */
+#define TN40_GMAC_RX_FILTER_ACF 0x0040 /* accept control frames */
+#define TN40_GMAC_RX_FILTER_ARUNT 0x0020 /* accept under run */
+#define TN40_GMAC_RX_FILTER_ACRC 0x0010 /* accept crc error */
+#define TN40_GMAC_RX_FILTER_AM 0x0008 /* accept multicast */
+#define TN40_GMAC_RX_FILTER_AB 0x0004 /* accept broadcast */
+#define TN40_GMAC_RX_FILTER_PRM 0x0001 /* [0:1] promiscuous mode */
+
+#define TN40_MAX_FRAME_AB_VAL 0x3fff /* 13:0 */
+
+#define TN40_CLKPLL_PLLLKD 0x0200 /* 9 */
+#define TN40_CLKPLL_RSTEND 0x0100 /* 8 */
+#define TN40_CLKPLL_SFTRST 0x0001 /* 0 */
+
+#define TN40_CLKPLL_LKD (TN40_CLKPLL_PLLLKD | TN40_CLKPLL_RSTEND)
+
+#endif
-- 
cgit v1.2.3-58-ga151


From dd2a0ff554082f6972b6cc6ea8f9d67e123e054e Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@gmail.com>
Date: Mon, 24 Jun 2024 08:55:04 +0900
Subject: net: tn40xx: add basic Tx handling

This patch adds device specific structures to initialize the hardware
with basic Tx handling. The original driver loads the embedded
firmware in the header file. This driver is implemented to use the
firmware APIs.

The Tx logic uses three major data structures; two ring buffers with
NIC and one database. One ring buffer is used to send information
about packets to be sent for NIC. The other is used to get information
from NIC about packet that are sent. The database is used to keep the
information about DMA mapping. After a packet is sent, the db is used
to free the resource used for the packet.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@gmail.com>
Link: https://patch.msgid.link/20240623235507.108147-5-fujita.tomonori@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/tehuti/Kconfig |    1 +
 drivers/net/ethernet/tehuti/tn40.c  | 1253 +++++++++++++++++++++++++++++++++++
 drivers/net/ethernet/tehuti/tn40.h  |  157 +++++
 3 files changed, 1411 insertions(+)

diff --git a/drivers/net/ethernet/tehuti/Kconfig b/drivers/net/ethernet/tehuti/Kconfig
index 849e3b4a71c1..4198fd59e42e 100644
--- a/drivers/net/ethernet/tehuti/Kconfig
+++ b/drivers/net/ethernet/tehuti/Kconfig
@@ -26,6 +26,7 @@ config TEHUTI
 config TEHUTI_TN40
 	tristate "Tehuti Networks TN40xx 10G Ethernet adapters"
 	depends on PCI
+	select FW_LOADER
 	help
 	  This driver supports 10G Ethernet adapters using Tehuti Networks
 	  TN40xx chips. Currently, adapters with Applied Micro Circuits
diff --git a/drivers/net/ethernet/tehuti/tn40.c b/drivers/net/ethernet/tehuti/tn40.c
index 248b17366680..ce3c323aa073 100644
--- a/drivers/net/ethernet/tehuti/tn40.c
+++ b/drivers/net/ethernet/tehuti/tn40.c
@@ -1,12 +1,1176 @@
 // SPDX-License-Identifier: GPL-2.0+
 /* Copyright (c) Tehuti Networks Ltd. */
 
+#include <linux/bitfield.h>
+#include <linux/ethtool.h>
+#include <linux/firmware.h>
+#include <linux/if_vlan.h>
+#include <linux/iopoll.h>
+#include <linux/netdevice.h>
 #include <linux/pci.h>
+#include <linux/vmalloc.h>
 
 #include "tn40.h"
 
+#define TN40_SHORT_PACKET_SIZE 60
+#define TN40_FIRMWARE_NAME "tehuti/bdx.bin"
+
+static void tn40_enable_interrupts(struct tn40_priv *priv)
+{
+	tn40_write_reg(priv, TN40_REG_IMR, priv->isr_mask);
+}
+
+static void tn40_disable_interrupts(struct tn40_priv *priv)
+{
+	tn40_write_reg(priv, TN40_REG_IMR, 0);
+}
+
+static int tn40_fifo_alloc(struct tn40_priv *priv, struct tn40_fifo *f,
+			   int fsz_type,
+			   u16 reg_cfg0, u16 reg_cfg1,
+			   u16 reg_rptr, u16 reg_wptr)
+{
+	u16 memsz = TN40_FIFO_SIZE * (1 << fsz_type);
+	u64 cfg_base;
+
+	memset(f, 0, sizeof(struct tn40_fifo));
+	/* 1K extra space is allocated at the end of the fifo to simplify
+	 * processing of descriptors that wraps around fifo's end.
+	 */
+	f->va = dma_alloc_coherent(&priv->pdev->dev,
+				   memsz + TN40_FIFO_EXTRA_SPACE, &f->da,
+				   GFP_KERNEL);
+	if (!f->va)
+		return -ENOMEM;
+
+	f->reg_cfg0 = reg_cfg0;
+	f->reg_cfg1 = reg_cfg1;
+	f->reg_rptr = reg_rptr;
+	f->reg_wptr = reg_wptr;
+	f->rptr = 0;
+	f->wptr = 0;
+	f->memsz = memsz;
+	f->size_mask = memsz - 1;
+	cfg_base = lower_32_bits((f->da & TN40_TX_RX_CFG0_BASE) | fsz_type);
+	tn40_write_reg(priv, reg_cfg0, cfg_base);
+	tn40_write_reg(priv, reg_cfg1, upper_32_bits(f->da));
+	return 0;
+}
+
+static void tn40_fifo_free(struct tn40_priv *priv, struct tn40_fifo *f)
+{
+	dma_free_coherent(&priv->pdev->dev,
+			  f->memsz + TN40_FIFO_EXTRA_SPACE, f->va, f->da);
+}
+
+/* TX HW/SW interaction overview
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ * There are 2 types of TX communication channels between driver and NIC.
+ * 1) TX Free Fifo - TXF - Holds ack descriptors for sent packets.
+ * 2) TX Data Fifo - TXD - Holds descriptors of full buffers.
+ *
+ * Currently the NIC supports TSO, checksumming and gather DMA
+ * UFO and IP fragmentation is on the way.
+ *
+ * RX SW Data Structures
+ * ~~~~~~~~~~~~~~~~~~~~~
+ * TXDB is used to keep track of all skbs owned by SW and their DMA addresses.
+ * For TX case, ownership lasts from getting the packet via hard_xmit and
+ * until the HW acknowledges sending the packet by TXF descriptors.
+ * TXDB is implemented as a cyclic buffer.
+ *
+ * FIFO objects keep info about the fifo's size and location, relevant HW
+ * registers, usage and skb db. Each RXD and RXF fifo has their own fifo
+ * structure. Implemented as simple struct.
+ *
+ * TX SW Execution Flow
+ * ~~~~~~~~~~~~~~~~~~~~
+ * OS calls the driver's hard_xmit method with a packet to send. The driver
+ * creates DMA mappings, builds TXD descriptors and kicks the HW by updating
+ * TXD WPTR.
+ *
+ * When a packet is sent, The HW write a TXF descriptor and the SW
+ * frees the original skb. To prevent TXD fifo overflow without
+ * reading HW registers every time, the SW deploys "tx level"
+ * technique. Upon startup, the tx level is initialized to TXD fifo
+ * length. For every sent packet, the SW gets its TXD descriptor size
+ * (from a pre-calculated array) and subtracts it from tx level.  The
+ * size is also stored in txdb. When a TXF ack arrives, the SW fetched
+ * the size of the original TXD descriptor from the txdb and adds it
+ * to the tx level. When the Tx level drops below some predefined
+ * threshold, the driver stops the TX queue. When the TX level rises
+ * above that level, the tx queue is enabled again.
+ *
+ * This technique avoids excessive reading of RPTR and WPTR registers.
+ * As our benchmarks shows, it adds 1.5 Gbit/sec to NIC's throughput.
+ */
+static void tn40_do_tx_db_ptr_next(struct tn40_txdb *db,
+				   struct tn40_tx_map **pptr)
+{
+	++*pptr;
+	if (unlikely(*pptr == db->end))
+		*pptr = db->start;
+}
+
+static void tn40_tx_db_inc_rptr(struct tn40_txdb *db)
+{
+	tn40_do_tx_db_ptr_next(db, &db->rptr);
+}
+
+static void tn40_tx_db_inc_wptr(struct tn40_txdb *db)
+{
+	tn40_do_tx_db_ptr_next(db, &db->wptr);
+}
+
+static int tn40_tx_db_init(struct tn40_txdb *d, int sz_type)
+{
+	int memsz = TN40_FIFO_SIZE * (1 << (sz_type + 1));
+
+	d->start = vzalloc(memsz);
+	if (!d->start)
+		return -ENOMEM;
+	/* In order to differentiate between an empty db state and a full db
+	 * state at least one element should always be empty in order to
+	 * avoid rptr == wptr, which means that the db is empty.
+	 */
+	d->size = memsz / sizeof(struct tn40_tx_map) - 1;
+	d->end = d->start + d->size + 1;	/* just after last element */
+
+	/* All dbs are created empty */
+	d->rptr = d->start;
+	d->wptr = d->start;
+	return 0;
+}
+
+static void tn40_tx_db_close(struct tn40_txdb *d)
+{
+	if (d->start) {
+		vfree(d->start);
+		d->start = NULL;
+	}
+}
+
+/* Sizes of tx desc (including padding if needed) as function of the SKB's
+ * frag number
+ * 7 - is number of lwords in txd with one phys buffer
+ * 3 - is number of lwords used for every additional phys buffer
+ * for (i = 0; i < TN40_MAX_PBL; i++) {
+ *	lwords = 7 + (i * 3);
+ *	if (lwords & 1)
+ *		lwords++;	pad it with 1 lword
+ *	tn40_txd_sizes[i].bytes = lwords << 2;
+ *	tn40_txd_sizes[i].qwords = lwords >> 1;
+ * }
+ */
+static struct {
+	u16 bytes;
+	u16 qwords;	/* qword = 64 bit */
+} tn40_txd_sizes[] = {
+	{0x20, 0x04},
+	{0x28, 0x05},
+	{0x38, 0x07},
+	{0x40, 0x08},
+	{0x50, 0x0a},
+	{0x58, 0x0b},
+	{0x68, 0x0d},
+	{0x70, 0x0e},
+	{0x80, 0x10},
+	{0x88, 0x11},
+	{0x98, 0x13},
+	{0xa0, 0x14},
+	{0xb0, 0x16},
+	{0xb8, 0x17},
+	{0xc8, 0x19},
+	{0xd0, 0x1a},
+	{0xe0, 0x1c},
+	{0xe8, 0x1d},
+	{0xf8, 0x1f},
+};
+
+static void tn40_pbl_set(struct tn40_pbl *pbl, dma_addr_t dma, int len)
+{
+	pbl->len = cpu_to_le32(len);
+	pbl->pa_lo = cpu_to_le32(lower_32_bits(dma));
+	pbl->pa_hi = cpu_to_le32(upper_32_bits(dma));
+}
+
+static void tn40_txdb_set(struct tn40_txdb *db, dma_addr_t dma, int len)
+{
+	db->wptr->len = len;
+	db->wptr->addr.dma = dma;
+}
+
+struct tn40_mapping_info {
+	dma_addr_t dma;
+	size_t size;
+};
+
+/**
+ * tn40_tx_map_skb - create and store DMA mappings for skb's data blocks
+ * @priv: NIC private structure
+ * @skb: socket buffer to map
+ * @txdd: pointer to tx descriptor to be updated
+ * @pkt_len: pointer to unsigned long value
+ *
+ * This function creates DMA mappings for skb's data blocks and writes them to
+ * PBL of a new tx descriptor. It also stores them in the tx db, so they could
+ * be unmapped after the data has been sent. It is the responsibility of the
+ * caller to make sure that there is enough space in the txdb. The last
+ * element holds a pointer to skb itself and is marked with a zero length.
+ *
+ * Return: 0 on success and negative value on error.
+ */
+static int tn40_tx_map_skb(struct tn40_priv *priv, struct sk_buff *skb,
+			   struct tn40_txd_desc *txdd, unsigned int *pkt_len)
+{
+	struct tn40_mapping_info info[TN40_MAX_PBL];
+	int nr_frags = skb_shinfo(skb)->nr_frags;
+	struct tn40_pbl *pbl = &txdd->pbl[0];
+	struct tn40_txdb *db = &priv->txdb;
+	unsigned int size;
+	int i, len, ret;
+	dma_addr_t dma;
+
+	netdev_dbg(priv->ndev, "TX skb %p skbLen %d dataLen %d frags %d\n", skb,
+		   skb->len, skb->data_len, nr_frags);
+	if (nr_frags > TN40_MAX_PBL - 1) {
+		ret = skb_linearize(skb);
+		if (ret)
+			return ret;
+		nr_frags = skb_shinfo(skb)->nr_frags;
+	}
+	/* initial skb */
+	len = skb->len - skb->data_len;
+	dma = dma_map_single(&priv->pdev->dev, skb->data, len,
+			     DMA_TO_DEVICE);
+	ret = dma_mapping_error(&priv->pdev->dev, dma);
+	if (ret)
+		return ret;
+
+	tn40_txdb_set(db, dma, len);
+	tn40_pbl_set(pbl++, db->wptr->addr.dma, db->wptr->len);
+	*pkt_len = db->wptr->len;
+
+	for (i = 0; i < nr_frags; i++) {
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+
+		size = skb_frag_size(frag);
+		dma = skb_frag_dma_map(&priv->pdev->dev, frag, 0,
+				       size, DMA_TO_DEVICE);
+
+		ret = dma_mapping_error(&priv->pdev->dev, dma);
+		if (ret)
+			goto mapping_error;
+		info[i].dma = dma;
+		info[i].size = size;
+	}
+
+	for (i = 0; i < nr_frags; i++) {
+		tn40_tx_db_inc_wptr(db);
+		tn40_txdb_set(db, info[i].dma, info[i].size);
+		tn40_pbl_set(pbl++, db->wptr->addr.dma, db->wptr->len);
+		*pkt_len += db->wptr->len;
+	}
+
+	/* SHORT_PKT_FIX */
+	if (skb->len < TN40_SHORT_PACKET_SIZE)
+		++nr_frags;
+
+	/* Add skb clean up info. */
+	tn40_tx_db_inc_wptr(db);
+	db->wptr->len = -tn40_txd_sizes[nr_frags].bytes;
+	db->wptr->addr.skb = skb;
+	tn40_tx_db_inc_wptr(db);
+
+	return 0;
+ mapping_error:
+	dma_unmap_page(&priv->pdev->dev, db->wptr->addr.dma, db->wptr->len,
+		       DMA_TO_DEVICE);
+	for (; i > 0; i--)
+		dma_unmap_page(&priv->pdev->dev, info[i - 1].dma,
+			       info[i - 1].size, DMA_TO_DEVICE);
+	return -ENOMEM;
+}
+
+static int tn40_create_tx_ring(struct tn40_priv *priv)
+{
+	int ret;
+
+	ret = tn40_fifo_alloc(priv, &priv->txd_fifo0.m, priv->txd_size,
+			      TN40_REG_TXD_CFG0_0, TN40_REG_TXD_CFG1_0,
+			      TN40_REG_TXD_RPTR_0, TN40_REG_TXD_WPTR_0);
+	if (ret)
+		return ret;
+
+	ret = tn40_fifo_alloc(priv, &priv->txf_fifo0.m, priv->txf_size,
+			      TN40_REG_TXF_CFG0_0, TN40_REG_TXF_CFG1_0,
+			      TN40_REG_TXF_RPTR_0, TN40_REG_TXF_WPTR_0);
+	if (ret)
+		goto err_free_txd;
+
+	/* The TX db has to keep mappings for all packets sent (on
+	 * TxD) and not yet reclaimed (on TxF).
+	 */
+	ret = tn40_tx_db_init(&priv->txdb, max(priv->txd_size, priv->txf_size));
+	if (ret)
+		goto err_free_txf;
+
+	/* SHORT_PKT_FIX */
+	priv->b0_len = 64;
+	priv->b0_va = dma_alloc_coherent(&priv->pdev->dev, priv->b0_len,
+					 &priv->b0_dma, GFP_KERNEL);
+	if (!priv->b0_va)
+		goto err_free_db;
+
+	priv->tx_level = TN40_MAX_TX_LEVEL;
+	priv->tx_update_mark = priv->tx_level - 1024;
+	return 0;
+err_free_db:
+	tn40_tx_db_close(&priv->txdb);
+err_free_txf:
+	tn40_fifo_free(priv, &priv->txf_fifo0.m);
+err_free_txd:
+	tn40_fifo_free(priv, &priv->txd_fifo0.m);
+	return -ENOMEM;
+}
+
+/**
+ * tn40_tx_space - Calculate the available space in the TX fifo.
+ * @priv: NIC private structure
+ *
+ * Return: available space in TX fifo in bytes
+ */
+static int tn40_tx_space(struct tn40_priv *priv)
+{
+	struct tn40_txd_fifo *f = &priv->txd_fifo0;
+	int fsize;
+
+	f->m.rptr = tn40_read_reg(priv, f->m.reg_rptr) & TN40_TXF_WPTR_WR_PTR;
+	fsize = f->m.rptr - f->m.wptr;
+	if (fsize <= 0)
+		fsize = f->m.memsz + fsize;
+	return fsize;
+}
+
+#define TN40_TXD_FULL_CHECKSUM 7
+
+static netdev_tx_t tn40_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+{
+	struct tn40_priv *priv = netdev_priv(ndev);
+	struct tn40_txd_fifo *f = &priv->txd_fifo0;
+	int txd_checksum = TN40_TXD_FULL_CHECKSUM;
+	struct tn40_txd_desc *txdd;
+	int nr_frags, len, err;
+	unsigned int pkt_len;
+	int txd_vlan_id = 0;
+	int txd_lgsnd = 0;
+	int txd_vtag = 0;
+	int txd_mss = 0;
+
+	/* Build tx descriptor */
+	txdd = (struct tn40_txd_desc *)(f->m.va + f->m.wptr);
+	err = tn40_tx_map_skb(priv, skb, txdd, &pkt_len);
+	if (err) {
+		u64_stats_update_begin(&priv->syncp);
+		priv->stats.tx_dropped++;
+		u64_stats_update_end(&priv->syncp);
+		dev_kfree_skb(skb);
+		return NETDEV_TX_OK;
+	}
+	nr_frags = skb_shinfo(skb)->nr_frags;
+	if (unlikely(skb->ip_summed != CHECKSUM_PARTIAL))
+		txd_checksum = 0;
+
+	if (skb_shinfo(skb)->gso_size) {
+		txd_mss = skb_shinfo(skb)->gso_size;
+		txd_lgsnd = 1;
+		netdev_dbg(priv->ndev, "skb %p pkt len %d gso size = %d\n", skb,
+			   pkt_len, txd_mss);
+	}
+	if (skb_vlan_tag_present(skb)) {
+		/* Don't cut VLAN ID to 12 bits */
+		txd_vlan_id = skb_vlan_tag_get(skb);
+		txd_vtag = 1;
+	}
+	txdd->va_hi = 0;
+	txdd->va_lo = 0;
+	txdd->length = cpu_to_le16(pkt_len);
+	txdd->mss = cpu_to_le16(txd_mss);
+	txdd->txd_val1 =
+		cpu_to_le32(TN40_TXD_W1_VAL
+			    (tn40_txd_sizes[nr_frags].qwords, txd_checksum,
+			     txd_vtag, txd_lgsnd, txd_vlan_id));
+	netdev_dbg(priv->ndev, "=== w1 qwords[%d] %d =====\n", nr_frags,
+		   tn40_txd_sizes[nr_frags].qwords);
+	netdev_dbg(priv->ndev, "=== TxD desc =====================\n");
+	netdev_dbg(priv->ndev, "=== w1: 0x%x ================\n",
+		   txdd->txd_val1);
+	netdev_dbg(priv->ndev, "=== w2: mss 0x%x len 0x%x\n", txdd->mss,
+		   txdd->length);
+	/* SHORT_PKT_FIX */
+	if (pkt_len < TN40_SHORT_PACKET_SIZE) {
+		struct tn40_pbl *pbl = &txdd->pbl[++nr_frags];
+
+		txdd->length = cpu_to_le16(TN40_SHORT_PACKET_SIZE);
+		txdd->txd_val1 =
+			cpu_to_le32(TN40_TXD_W1_VAL
+				    (tn40_txd_sizes[nr_frags].qwords,
+				     txd_checksum, txd_vtag, txd_lgsnd,
+				     txd_vlan_id));
+		pbl->len = cpu_to_le32(TN40_SHORT_PACKET_SIZE - pkt_len);
+		pbl->pa_lo = cpu_to_le32(lower_32_bits(priv->b0_dma));
+		pbl->pa_hi = cpu_to_le32(upper_32_bits(priv->b0_dma));
+		netdev_dbg(priv->ndev, "=== SHORT_PKT_FIX   ==============\n");
+		netdev_dbg(priv->ndev, "=== nr_frags : %d   ==============\n",
+			   nr_frags);
+	}
+
+	/* Increment TXD write pointer. In case of fifo wrapping copy
+	 * reminder of the descriptor to the beginning.
+	 */
+	f->m.wptr += tn40_txd_sizes[nr_frags].bytes;
+	len = f->m.wptr - f->m.memsz;
+	if (unlikely(len >= 0)) {
+		f->m.wptr = len;
+		if (len > 0)
+			memcpy(f->m.va, f->m.va + f->m.memsz, len);
+	}
+	/* Force memory writes to complete before letting the HW know
+	 * there are new descriptors to fetch.
+	 */
+	wmb();
+
+	priv->tx_level -= tn40_txd_sizes[nr_frags].bytes;
+	if (priv->tx_level > priv->tx_update_mark) {
+		tn40_write_reg(priv, f->m.reg_wptr,
+			       f->m.wptr & TN40_TXF_WPTR_WR_PTR);
+	} else {
+		if (priv->tx_noupd++ > TN40_NO_UPD_PACKETS) {
+			priv->tx_noupd = 0;
+			tn40_write_reg(priv, f->m.reg_wptr,
+				       f->m.wptr & TN40_TXF_WPTR_WR_PTR);
+		}
+	}
+
+	u64_stats_update_begin(&priv->syncp);
+	priv->stats.tx_packets++;
+	priv->stats.tx_bytes += pkt_len;
+	u64_stats_update_end(&priv->syncp);
+	if (priv->tx_level < TN40_MIN_TX_LEVEL) {
+		netdev_dbg(priv->ndev, "TX Q STOP level %d\n", priv->tx_level);
+		netif_stop_queue(ndev);
+	}
+
+	return NETDEV_TX_OK;
+}
+
+static void tn40_tx_free_skbs(struct tn40_priv *priv)
+{
+	struct tn40_txdb *db = &priv->txdb;
+
+	while (db->rptr != db->wptr) {
+		if (likely(db->rptr->len))
+			dma_unmap_page(&priv->pdev->dev, db->rptr->addr.dma,
+				       db->rptr->len, DMA_TO_DEVICE);
+		else
+			dev_kfree_skb(db->rptr->addr.skb);
+		tn40_tx_db_inc_rptr(db);
+	}
+}
+
+static void tn40_destroy_tx_ring(struct tn40_priv *priv)
+{
+	tn40_tx_free_skbs(priv);
+	tn40_fifo_free(priv, &priv->txd_fifo0.m);
+	tn40_fifo_free(priv, &priv->txf_fifo0.m);
+	tn40_tx_db_close(&priv->txdb);
+	/* SHORT_PKT_FIX */
+	if (priv->b0_len) {
+		dma_free_coherent(&priv->pdev->dev, priv->b0_len, priv->b0_va,
+				  priv->b0_dma);
+		priv->b0_len = 0;
+	}
+}
+
+/**
+ * tn40_tx_push_desc - Push a descriptor to TxD fifo.
+ *
+ * @priv: NIC private structure
+ * @data: desc's data
+ * @size: desc's size
+ *
+ * This function pushes desc to TxD fifo and overlaps it if needed.
+ *
+ * This function does not check for available space, nor does it check
+ * that the data size is smaller than the fifo size. Checking for
+ * space is the responsibility of the caller.
+ */
+static void tn40_tx_push_desc(struct tn40_priv *priv, void *data, int size)
+{
+	struct tn40_txd_fifo *f = &priv->txd_fifo0;
+	int i = f->m.memsz - f->m.wptr;
+
+	if (size == 0)
+		return;
+
+	if (i > size) {
+		memcpy(f->m.va + f->m.wptr, data, size);
+		f->m.wptr += size;
+	} else {
+		memcpy(f->m.va + f->m.wptr, data, i);
+		f->m.wptr = size - i;
+		memcpy(f->m.va, data + i, f->m.wptr);
+	}
+	tn40_write_reg(priv, f->m.reg_wptr, f->m.wptr & TN40_TXF_WPTR_WR_PTR);
+}
+
+/**
+ * tn40_tx_push_desc_safe - push descriptor to TxD fifo in a safe way.
+ *
+ * @priv: NIC private structure
+ * @data: descriptor data
+ * @size: descriptor size
+ *
+ * This function does check for available space and, if necessary,
+ * waits for the NIC to read existing data before writing new data.
+ */
+static void tn40_tx_push_desc_safe(struct tn40_priv *priv, void *data, int size)
+{
+	int timer = 0;
+
+	while (size > 0) {
+		/* We subtract 8 because when the fifo is full rptr ==
+		 * wptr, which also means that fifo is empty, we can
+		 * understand the difference, but could the HW do the
+		 * same ???
+		 */
+		int avail = tn40_tx_space(priv) - 8;
+
+		if (avail <= 0) {
+			if (timer++ > 300) /* Prevent endless loop */
+				break;
+			/* Give the HW a chance to clean the fifo */
+			usleep_range(50, 60);
+			continue;
+		}
+		avail = min(avail, size);
+		netdev_dbg(priv->ndev,
+			   "about to push  %d bytes starting %p size %d\n",
+			   avail, data, size);
+		tn40_tx_push_desc(priv, data, avail);
+		size -= avail;
+		data += avail;
+	}
+}
+
+static int tn40_set_link_speed(struct tn40_priv *priv, u32 speed)
+{
+	u32 val;
+	int i;
+
+	netdev_dbg(priv->ndev, "speed %d\n", speed);
+	switch (speed) {
+	case SPEED_10000:
+	case SPEED_5000:
+	case SPEED_2500:
+		netdev_dbg(priv->ndev, "link_speed %d\n", speed);
+
+		tn40_write_reg(priv, 0x1010, 0x217);	/*ETHSD.REFCLK_CONF  */
+		tn40_write_reg(priv, 0x104c, 0x4c);	/*ETHSD.L0_RX_PCNT  */
+		tn40_write_reg(priv, 0x1050, 0x4c);	/*ETHSD.L1_RX_PCNT  */
+		tn40_write_reg(priv, 0x1054, 0x4c);	/*ETHSD.L2_RX_PCNT  */
+		tn40_write_reg(priv, 0x1058, 0x4c);	/*ETHSD.L3_RX_PCNT  */
+		tn40_write_reg(priv, 0x102c, 0x434);	/*ETHSD.L0_TX_PCNT  */
+		tn40_write_reg(priv, 0x1030, 0x434);	/*ETHSD.L1_TX_PCNT  */
+		tn40_write_reg(priv, 0x1034, 0x434);	/*ETHSD.L2_TX_PCNT  */
+		tn40_write_reg(priv, 0x1038, 0x434);	/*ETHSD.L3_TX_PCNT  */
+		tn40_write_reg(priv, 0x6300, 0x0400);	/*MAC.PCS_CTRL */
+
+		tn40_write_reg(priv, 0x1018, 0x00);	/*Mike2 */
+		udelay(5);
+		tn40_write_reg(priv, 0x1018, 0x04);	/*Mike2 */
+		udelay(5);
+		tn40_write_reg(priv, 0x1018, 0x06);	/*Mike2 */
+		udelay(5);
+		/*MikeFix1 */
+		/*L0: 0x103c , L1: 0x1040 , L2: 0x1044 , L3: 0x1048 =0x81644 */
+		tn40_write_reg(priv, 0x103c, 0x81644);	/*ETHSD.L0_TX_DCNT  */
+		tn40_write_reg(priv, 0x1040, 0x81644);	/*ETHSD.L1_TX_DCNT  */
+		tn40_write_reg(priv, 0x1044, 0x81644);	/*ETHSD.L2_TX_DCNT  */
+		tn40_write_reg(priv, 0x1048, 0x81644);	/*ETHSD.L3_TX_DCNT  */
+		tn40_write_reg(priv, 0x1014, 0x043);	/*ETHSD.INIT_STAT */
+		for (i = 1000; i; i--) {
+			usleep_range(50, 60);
+			/*ETHSD.INIT_STAT */
+			val = tn40_read_reg(priv, 0x1014);
+			if (val & (1 << 9)) {
+				/*ETHSD.INIT_STAT */
+				tn40_write_reg(priv, 0x1014, 0x3);
+				/*ETHSD.INIT_STAT */
+				val = tn40_read_reg(priv, 0x1014);
+
+				break;
+			}
+		}
+		if (!i)
+			netdev_err(priv->ndev, "MAC init timeout!\n");
+
+		tn40_write_reg(priv, 0x6350, 0x0);	/*MAC.PCS_IF_MODE */
+		tn40_write_reg(priv, TN40_REG_CTRLST, 0xC13);	/*0x93//0x13 */
+		tn40_write_reg(priv, 0x111c, 0x7ff);	/*MAC.MAC_RST_CNT */
+		usleep_range(2000, 2100);
+
+		tn40_write_reg(priv, 0x111c, 0x0);	/*MAC.MAC_RST_CNT */
+		break;
+
+	case SPEED_1000:
+	case SPEED_100:
+		tn40_write_reg(priv, 0x1010, 0x613);	/*ETHSD.REFCLK_CONF */
+		tn40_write_reg(priv, 0x104c, 0x4d);	/*ETHSD.L0_RX_PCNT  */
+		tn40_write_reg(priv, 0x1050, 0x0);	/*ETHSD.L1_RX_PCNT  */
+		tn40_write_reg(priv, 0x1054, 0x0);	/*ETHSD.L2_RX_PCNT  */
+		tn40_write_reg(priv, 0x1058, 0x0);	/*ETHSD.L3_RX_PCNT  */
+		tn40_write_reg(priv, 0x102c, 0x35);	/*ETHSD.L0_TX_PCNT  */
+		tn40_write_reg(priv, 0x1030, 0x0);	/*ETHSD.L1_TX_PCNT  */
+		tn40_write_reg(priv, 0x1034, 0x0);	/*ETHSD.L2_TX_PCNT  */
+		tn40_write_reg(priv, 0x1038, 0x0);	/*ETHSD.L3_TX_PCNT  */
+		tn40_write_reg(priv, 0x6300, 0x01140);	/*MAC.PCS_CTRL */
+
+		tn40_write_reg(priv, 0x1014, 0x043);	/*ETHSD.INIT_STAT */
+		for (i = 1000; i; i--) {
+			usleep_range(50, 60);
+			val = tn40_read_reg(priv, 0x1014); /*ETHSD.INIT_STAT */
+			if (val & (1 << 9)) {
+				/*ETHSD.INIT_STAT */
+				tn40_write_reg(priv, 0x1014, 0x3);
+				/*ETHSD.INIT_STAT */
+				val = tn40_read_reg(priv, 0x1014);
+
+				break;
+			}
+		}
+		if (!i)
+			netdev_err(priv->ndev, "MAC init timeout!\n");
+
+		tn40_write_reg(priv, 0x6350, 0x2b);	/*MAC.PCS_IF_MODE 1g */
+		tn40_write_reg(priv, 0x6310, 0x9801);	/*MAC.PCS_DEV_AB */
+
+		tn40_write_reg(priv, 0x6314, 0x1);	/*MAC.PCS_PART_AB */
+		tn40_write_reg(priv, 0x6348, 0xc8);	/*MAC.PCS_LINK_LO */
+		tn40_write_reg(priv, 0x634c, 0xc8);	/*MAC.PCS_LINK_HI */
+		usleep_range(50, 60);
+		tn40_write_reg(priv, TN40_REG_CTRLST, 0xC13);	/*0x93//0x13 */
+		tn40_write_reg(priv, 0x111c, 0x7ff);	/*MAC.MAC_RST_CNT */
+		usleep_range(2000, 2100);
+
+		tn40_write_reg(priv, 0x111c, 0x0);	/*MAC.MAC_RST_CNT */
+		tn40_write_reg(priv, 0x6300, 0x1140);	/*MAC.PCS_CTRL */
+		break;
+
+	case 0:		/* Link down */
+		tn40_write_reg(priv, 0x104c, 0x0);	/*ETHSD.L0_RX_PCNT  */
+		tn40_write_reg(priv, 0x1050, 0x0);	/*ETHSD.L1_RX_PCNT  */
+		tn40_write_reg(priv, 0x1054, 0x0);	/*ETHSD.L2_RX_PCNT  */
+		tn40_write_reg(priv, 0x1058, 0x0);	/*ETHSD.L3_RX_PCNT  */
+		tn40_write_reg(priv, 0x102c, 0x0);	/*ETHSD.L0_TX_PCNT  */
+		tn40_write_reg(priv, 0x1030, 0x0);	/*ETHSD.L1_TX_PCNT  */
+		tn40_write_reg(priv, 0x1034, 0x0);	/*ETHSD.L2_TX_PCNT  */
+		tn40_write_reg(priv, 0x1038, 0x0);	/*ETHSD.L3_TX_PCNT  */
+
+		tn40_write_reg(priv, TN40_REG_CTRLST, 0x800);
+		tn40_write_reg(priv, 0x111c, 0x7ff);	/*MAC.MAC_RST_CNT */
+		usleep_range(2000, 2100);
+
+		tn40_write_reg(priv, 0x111c, 0x0);	/*MAC.MAC_RST_CNT */
+		break;
+
+	default:
+		netdev_err(priv->ndev,
+			   "Link speed was not identified yet (%d)\n", speed);
+		speed = 0;
+		break;
+	}
+	return speed;
+}
+
+static void tn40_link_changed(struct tn40_priv *priv)
+{
+	u32 link = tn40_read_reg(priv,
+				 TN40_REG_MAC_LNK_STAT) & TN40_MAC_LINK_STAT;
+
+	netdev_dbg(priv->ndev, "link changed %u\n", link);
+}
+
+static void tn40_isr_extra(struct tn40_priv *priv, u32 isr)
+{
+	if (isr & (TN40_IR_LNKCHG0 | TN40_IR_LNKCHG1 | TN40_IR_TMR0)) {
+		netdev_dbg(priv->ndev, "isr = 0x%x\n", isr);
+		tn40_link_changed(priv);
+	}
+}
+
+static irqreturn_t tn40_isr_napi(int irq, void *dev)
+{
+	struct tn40_priv *priv = netdev_priv((struct net_device *)dev);
+	u32 isr;
+
+	isr = tn40_read_reg(priv, TN40_REG_ISR_MSK0);
+
+	if (unlikely(!isr)) {
+		tn40_enable_interrupts(priv);
+		return IRQ_NONE;	/* Not our interrupt */
+	}
+
+	if (isr & TN40_IR_EXTRA)
+		tn40_isr_extra(priv, isr);
+
+	if (isr & (TN40_IR_RX_DESC_0 | TN40_IR_TX_FREE_0 | TN40_IR_TMR1)) {
+		/* We get here if an interrupt has slept into the
+		 * small time window between these lines in
+		 * tn40_poll: tn40_enable_interrupts(priv); return 0;
+		 *
+		 * Currently interrupts are disabled (since we read
+		 * the ISR register) and we have failed to register
+		 * the next poll. So we read the regs to trigger the
+		 * chip and allow further interrupts.
+		 */
+		tn40_read_reg(priv, TN40_REG_TXF_WPTR_0);
+		tn40_read_reg(priv, TN40_REG_RXD_WPTR_0);
+	}
+
+	tn40_enable_interrupts(priv);
+	return IRQ_HANDLED;
+}
+
+static int tn40_fw_load(struct tn40_priv *priv)
+{
+	const struct firmware *fw = NULL;
+	int master, ret;
+	u32 val;
+
+	ret = request_firmware(&fw, TN40_FIRMWARE_NAME, &priv->pdev->dev);
+	if (ret)
+		return ret;
+
+	master = tn40_read_reg(priv, TN40_REG_INIT_SEMAPHORE);
+	if (!tn40_read_reg(priv, TN40_REG_INIT_STATUS) && master) {
+		netdev_dbg(priv->ndev, "Loading FW...\n");
+		tn40_tx_push_desc_safe(priv, (void *)fw->data, fw->size);
+		msleep(100);
+	}
+	ret = read_poll_timeout(tn40_read_reg, val, val, 2000, 400000, false,
+				priv, TN40_REG_INIT_STATUS);
+	if (master)
+		tn40_write_reg(priv, TN40_REG_INIT_SEMAPHORE, 1);
+
+	if (ret) {
+		netdev_err(priv->ndev, "firmware loading failed\n");
+		netdev_dbg(priv->ndev, "VPC: 0x%x VIC: 0x%x STATUS: 0x%xd\n",
+			   tn40_read_reg(priv, TN40_REG_VPC),
+			   tn40_read_reg(priv, TN40_REG_VIC),
+			   tn40_read_reg(priv, TN40_REG_INIT_STATUS));
+		ret = -EIO;
+	} else {
+		netdev_dbg(priv->ndev, "firmware loading success\n");
+	}
+	release_firmware(fw);
+	return ret;
+}
+
+static void tn40_restore_mac(struct net_device *ndev, struct tn40_priv *priv)
+{
+	u32 val;
+
+	netdev_dbg(priv->ndev, "mac0 =%x mac1 =%x mac2 =%x\n",
+		   tn40_read_reg(priv, TN40_REG_UNC_MAC0_A),
+		   tn40_read_reg(priv, TN40_REG_UNC_MAC1_A),
+		   tn40_read_reg(priv, TN40_REG_UNC_MAC2_A));
+
+	val = (ndev->dev_addr[0] << 8) | (ndev->dev_addr[1]);
+	tn40_write_reg(priv, TN40_REG_UNC_MAC2_A, val);
+	val = (ndev->dev_addr[2] << 8) | (ndev->dev_addr[3]);
+	tn40_write_reg(priv, TN40_REG_UNC_MAC1_A, val);
+	val = (ndev->dev_addr[4] << 8) | (ndev->dev_addr[5]);
+	tn40_write_reg(priv, TN40_REG_UNC_MAC0_A, val);
+
+	/* More then IP MAC address */
+	tn40_write_reg(priv, TN40_REG_MAC_ADDR_0,
+		       (ndev->dev_addr[3] << 24) | (ndev->dev_addr[2] << 16) |
+		       (ndev->dev_addr[1] << 8) | (ndev->dev_addr[0]));
+	tn40_write_reg(priv, TN40_REG_MAC_ADDR_1,
+		       (ndev->dev_addr[5] << 8) | (ndev->dev_addr[4]));
+
+	netdev_dbg(priv->ndev, "mac0 =%x mac1 =%x mac2 =%x\n",
+		   tn40_read_reg(priv, TN40_REG_UNC_MAC0_A),
+		   tn40_read_reg(priv, TN40_REG_UNC_MAC1_A),
+		   tn40_read_reg(priv, TN40_REG_UNC_MAC2_A));
+}
+
+static void tn40_hw_start(struct tn40_priv *priv)
+{
+	tn40_write_reg(priv, TN40_REG_FRM_LENGTH, 0X3FE0);
+	tn40_write_reg(priv, TN40_REG_GMAC_RXF_A, 0X10fd);
+	/*MikeFix1 */
+	/*L0: 0x103c , L1: 0x1040 , L2: 0x1044 , L3: 0x1048 =0x81644 */
+	tn40_write_reg(priv, 0x103c, 0x81644);	/*ETHSD.L0_TX_DCNT  */
+	tn40_write_reg(priv, 0x1040, 0x81644);	/*ETHSD.L1_TX_DCNT  */
+	tn40_write_reg(priv, 0x1044, 0x81644);	/*ETHSD.L2_TX_DCNT  */
+	tn40_write_reg(priv, 0x1048, 0x81644);	/*ETHSD.L3_TX_DCNT  */
+	tn40_write_reg(priv, TN40_REG_RX_FIFO_SECTION, 0x10);
+	tn40_write_reg(priv, TN40_REG_TX_FIFO_SECTION, 0xE00010);
+	tn40_write_reg(priv, TN40_REG_RX_FULLNESS, 0);
+	tn40_write_reg(priv, TN40_REG_TX_FULLNESS, 0);
+
+	tn40_write_reg(priv, TN40_REG_VGLB, 0);
+	tn40_write_reg(priv, TN40_REG_RDINTCM0, priv->rdintcm);
+	tn40_write_reg(priv, TN40_REG_RDINTCM2, 0);
+
+	/* old val = 0x300064 */
+	tn40_write_reg(priv, TN40_REG_TDINTCM0, priv->tdintcm);
+
+	/* Enable timer interrupt once in 2 secs. */
+	tn40_restore_mac(priv->ndev, priv);
+
+	/* Pause frame */
+	tn40_write_reg(priv, 0x12E0, 0x28);
+	tn40_write_reg(priv, TN40_REG_PAUSE_QUANT, 0xFFFF);
+	tn40_write_reg(priv, 0x6064, 0xF);
+
+	tn40_write_reg(priv, TN40_REG_GMAC_RXF_A,
+		       TN40_GMAC_RX_FILTER_OSEN | TN40_GMAC_RX_FILTER_TXFC |
+		       TN40_GMAC_RX_FILTER_AM | TN40_GMAC_RX_FILTER_AB);
+
+	tn40_enable_interrupts(priv);
+}
+
+static int tn40_hw_reset(struct tn40_priv *priv)
+{
+	u32 val;
+
+	/* Reset sequences: read, write 1, read, write 0 */
+	val = tn40_read_reg(priv, TN40_REG_CLKPLL);
+	tn40_write_reg(priv, TN40_REG_CLKPLL, (val | TN40_CLKPLL_SFTRST) + 0x8);
+	usleep_range(50, 60);
+	val = tn40_read_reg(priv, TN40_REG_CLKPLL);
+	tn40_write_reg(priv, TN40_REG_CLKPLL, val & ~TN40_CLKPLL_SFTRST);
+
+	/* Check that the PLLs are locked and reset ended */
+	val = read_poll_timeout(tn40_read_reg, val,
+				(val & TN40_CLKPLL_LKD) == TN40_CLKPLL_LKD,
+				10000, 700000, false, priv, TN40_REG_CLKPLL);
+	if (val)
+		return -EIO;
+
+	usleep_range(50, 60);
+	/* Do any PCI-E read transaction */
+	tn40_read_reg(priv, TN40_REG_RXD_CFG0_0);
+	return 0;
+}
+
+static void tn40_sw_reset(struct tn40_priv *priv)
+{
+	int i, ret;
+	u32 val;
+
+	/* 1. load MAC (obsolete) */
+	/* 2. disable Rx (and Tx) */
+	tn40_write_reg(priv, TN40_REG_GMAC_RXF_A, 0);
+	msleep(100);
+	/* 3. Disable port */
+	tn40_write_reg(priv, TN40_REG_DIS_PORT, 1);
+	/* 4. Disable queue */
+	tn40_write_reg(priv, TN40_REG_DIS_QU, 1);
+	/* 5. Wait until hw is disabled */
+	ret = read_poll_timeout(tn40_read_reg, val, val & 1, 10000, 500000,
+				false, priv, TN40_REG_RST_PORT);
+	if (ret)
+		netdev_err(priv->ndev, "SW reset timeout. continuing anyway\n");
+
+	/* 6. Disable interrupts */
+	tn40_write_reg(priv, TN40_REG_RDINTCM0, 0);
+	tn40_write_reg(priv, TN40_REG_TDINTCM0, 0);
+	tn40_write_reg(priv, TN40_REG_IMR, 0);
+	tn40_read_reg(priv, TN40_REG_ISR);
+
+	/* 7. Reset queue */
+	tn40_write_reg(priv, TN40_REG_RST_QU, 1);
+	/* 8. Reset port */
+	tn40_write_reg(priv, TN40_REG_RST_PORT, 1);
+	/* 9. Zero all read and write pointers */
+	for (i = TN40_REG_TXD_WPTR_0; i <= TN40_REG_TXF_RPTR_3; i += 0x10)
+		tn40_write_reg(priv, i, 0);
+	/* 10. Unset port disable */
+	tn40_write_reg(priv, TN40_REG_DIS_PORT, 0);
+	/* 11. Unset queue disable */
+	tn40_write_reg(priv, TN40_REG_DIS_QU, 0);
+	/* 12. Unset queue reset */
+	tn40_write_reg(priv, TN40_REG_RST_QU, 0);
+	/* 13. Unset port reset */
+	tn40_write_reg(priv, TN40_REG_RST_PORT, 0);
+	/* 14. Enable Rx */
+	/* Skipped. will be done later */
+}
+
+static int tn40_start(struct tn40_priv *priv)
+{
+	int ret;
+
+	ret = tn40_create_tx_ring(priv);
+	if (ret) {
+		netdev_err(priv->ndev, "failed to tx init %d\n", ret);
+		return ret;
+	}
+
+	ret = request_irq(priv->pdev->irq, &tn40_isr_napi, IRQF_SHARED,
+			  priv->ndev->name, priv->ndev);
+	if (ret) {
+		netdev_err(priv->ndev, "failed to request irq %d\n", ret);
+		goto err_tx_ring;
+	}
+
+	tn40_hw_start(priv);
+	return 0;
+err_tx_ring:
+	tn40_destroy_tx_ring(priv);
+	return ret;
+}
+
+static void tn40_stop(struct tn40_priv *priv)
+{
+	tn40_disable_interrupts(priv);
+	free_irq(priv->pdev->irq, priv->ndev);
+	tn40_sw_reset(priv);
+	tn40_destroy_tx_ring(priv);
+}
+
+static int tn40_close(struct net_device *ndev)
+{
+	struct tn40_priv *priv = netdev_priv(ndev);
+
+	tn40_stop(priv);
+	return 0;
+}
+
+static int tn40_open(struct net_device *dev)
+{
+	struct tn40_priv *priv = netdev_priv(dev);
+	int ret;
+
+	tn40_sw_reset(priv);
+	ret = tn40_start(priv);
+	if (ret) {
+		netdev_err(dev, "failed to start %d\n", ret);
+		return ret;
+	}
+	return 0;
+}
+
+static void __tn40_vlan_rx_vid(struct net_device *ndev, uint16_t vid,
+			       int enable)
+{
+	struct tn40_priv *priv = netdev_priv(ndev);
+	u32 reg, bit, val;
+
+	netdev_dbg(priv->ndev, "vid =%d value =%d\n", (int)vid, enable);
+	reg = TN40_REG_VLAN_0 + (vid / 32) * 4;
+	bit = 1 << vid % 32;
+	val = tn40_read_reg(priv, reg);
+	netdev_dbg(priv->ndev, "reg =%x, val =%x, bit =%d\n", reg, val, bit);
+	if (enable)
+		val |= bit;
+	else
+		val &= ~bit;
+	netdev_dbg(priv->ndev, "new val %x\n", val);
+	tn40_write_reg(priv, reg, val);
+}
+
+static int tn40_vlan_rx_add_vid(struct net_device *ndev,
+				__always_unused __be16 proto, u16 vid)
+{
+	__tn40_vlan_rx_vid(ndev, vid, 1);
+	return 0;
+}
+
+static int tn40_vlan_rx_kill_vid(struct net_device *ndev,
+				 __always_unused __be16 proto, u16 vid)
+{
+	__tn40_vlan_rx_vid(ndev, vid, 0);
+	return 0;
+}
+
+static void tn40_setmulti(struct net_device *ndev)
+{
+	u32 rxf_val = TN40_GMAC_RX_FILTER_AM | TN40_GMAC_RX_FILTER_AB |
+		TN40_GMAC_RX_FILTER_OSEN | TN40_GMAC_RX_FILTER_TXFC;
+	struct tn40_priv *priv = netdev_priv(ndev);
+	int i;
+
+	/* IMF - imperfect (hash) rx multicast filter */
+	/* PMF - perfect rx multicast filter */
+
+	/* FIXME: RXE(OFF) */
+	if (ndev->flags & IFF_PROMISC) {
+		rxf_val |= TN40_GMAC_RX_FILTER_PRM;
+	} else if (ndev->flags & IFF_ALLMULTI) {
+		/* set IMF to accept all multicast frames */
+		for (i = 0; i < TN40_MAC_MCST_HASH_NUM; i++)
+			tn40_write_reg(priv,
+				       TN40_REG_RX_MCST_HASH0 + i * 4, ~0);
+	} else if (netdev_mc_count(ndev)) {
+		struct netdev_hw_addr *mclist;
+		u32 reg, val;
+		u8 hash;
+
+		/* Set IMF to deny all multicast frames */
+		for (i = 0; i < TN40_MAC_MCST_HASH_NUM; i++)
+			tn40_write_reg(priv,
+				       TN40_REG_RX_MCST_HASH0 + i * 4, 0);
+
+		/* Set PMF to deny all multicast frames */
+		for (i = 0; i < TN40_MAC_MCST_NUM; i++) {
+			tn40_write_reg(priv,
+				       TN40_REG_RX_MAC_MCST0 + i * 8, 0);
+			tn40_write_reg(priv,
+				       TN40_REG_RX_MAC_MCST1 + i * 8, 0);
+		}
+		/* Use PMF to accept first MAC_MCST_NUM (15) addresses */
+
+		/* TBD: Sort the addresses and write them in ascending
+		 * order into RX_MAC_MCST regs. we skip this phase now
+		 * and accept ALL multicast frames through IMF. Accept
+		 * the rest of addresses throw IMF.
+		 */
+		netdev_for_each_mc_addr(mclist, ndev) {
+			hash = 0;
+			for (i = 0; i < ETH_ALEN; i++)
+				hash ^= mclist->addr[i];
+
+			reg = TN40_REG_RX_MCST_HASH0 + ((hash >> 5) << 2);
+			val = tn40_read_reg(priv, reg);
+			val |= (1 << (hash % 32));
+			tn40_write_reg(priv, reg, val);
+		}
+	} else {
+		rxf_val |= TN40_GMAC_RX_FILTER_AB;
+	}
+	tn40_write_reg(priv, TN40_REG_GMAC_RXF_A, rxf_val);
+	/* Enable RX */
+	/* FIXME: RXE(ON) */
+}
+
+static int tn40_set_mac(struct net_device *ndev, void *p)
+{
+	struct tn40_priv *priv = netdev_priv(ndev);
+	struct sockaddr *addr = p;
+
+	eth_hw_addr_set(ndev, addr->sa_data);
+	tn40_restore_mac(ndev, priv);
+	return 0;
+}
+
+static void tn40_mac_init(struct tn40_priv *priv)
+{
+	u8 addr[ETH_ALEN];
+	u64 val;
+
+	val = (u64)tn40_read_reg(priv, TN40_REG_UNC_MAC0_A);
+	val |= (u64)tn40_read_reg(priv, TN40_REG_UNC_MAC1_A) << 16;
+	val |= (u64)tn40_read_reg(priv, TN40_REG_UNC_MAC2_A) << 32;
+
+	u64_to_ether_addr(val, addr);
+	eth_hw_addr_set(priv->ndev, addr);
+}
+
+static void tn40_get_stats(struct net_device *ndev,
+			   struct rtnl_link_stats64 *stats)
+{
+	struct tn40_priv *priv = netdev_priv(ndev);
+	unsigned int start;
+
+	do {
+		start = u64_stats_fetch_begin(&priv->syncp);
+		stats->tx_packets = priv->stats.tx_packets;
+		stats->tx_bytes = priv->stats.tx_bytes;
+		stats->tx_dropped = priv->stats.tx_dropped;
+	} while (u64_stats_fetch_retry(&priv->syncp, start));
+}
+
+static const struct net_device_ops tn40_netdev_ops = {
+	.ndo_open = tn40_open,
+	.ndo_stop = tn40_close,
+	.ndo_start_xmit = tn40_start_xmit,
+	.ndo_validate_addr = eth_validate_addr,
+	.ndo_set_rx_mode = tn40_setmulti,
+	.ndo_get_stats64 = tn40_get_stats,
+	.ndo_set_mac_address = tn40_set_mac,
+	.ndo_vlan_rx_add_vid = tn40_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid = tn40_vlan_rx_kill_vid,
+};
+
+static int tn40_priv_init(struct tn40_priv *priv)
+{
+	int ret;
+
+	tn40_set_link_speed(priv, 0);
+
+	/* Set GPIO[9:0] to output 0 */
+	tn40_write_reg(priv, 0x51E0, 0x30010006);	/* GPIO_OE_ WR CMD */
+	tn40_write_reg(priv, 0x51F0, 0x0);	/* GPIO_OE_ DATA */
+	tn40_write_reg(priv, TN40_REG_MDIO_CMD_STAT, 0x3ec8);
+
+	/* we use tx descriptors to load a firmware. */
+	ret = tn40_create_tx_ring(priv);
+	if (ret)
+		return ret;
+	ret = tn40_fw_load(priv);
+	tn40_destroy_tx_ring(priv);
+	return ret;
+}
+
+static struct net_device *tn40_netdev_alloc(struct pci_dev *pdev)
+{
+	struct net_device *ndev;
+
+	ndev = devm_alloc_etherdev(&pdev->dev, sizeof(struct tn40_priv));
+	if (!ndev)
+		return NULL;
+	ndev->netdev_ops = &tn40_netdev_ops;
+	ndev->tx_queue_len = TN40_NDEV_TXQ_LEN;
+	ndev->mem_start = pci_resource_start(pdev, 0);
+	ndev->mem_end = pci_resource_end(pdev, 0);
+	ndev->min_mtu = ETH_ZLEN;
+	ndev->max_mtu = TN40_MAX_MTU;
+
+	ndev->features = NETIF_F_IP_CSUM |
+		NETIF_F_SG |
+		NETIF_F_FRAGLIST |
+		NETIF_F_TSO | NETIF_F_GRO |
+		NETIF_F_RXCSUM |
+		NETIF_F_RXHASH |
+		NETIF_F_HW_VLAN_CTAG_TX |
+		NETIF_F_HW_VLAN_CTAG_RX |
+		NETIF_F_HW_VLAN_CTAG_FILTER;
+	ndev->vlan_features = NETIF_F_IP_CSUM |
+			       NETIF_F_SG |
+			       NETIF_F_TSO | NETIF_F_GRO | NETIF_F_RXHASH;
+
+	if (dma_get_mask(&pdev->dev) == DMA_BIT_MASK(64)) {
+		ndev->features |= NETIF_F_HIGHDMA;
+		ndev->vlan_features |= NETIF_F_HIGHDMA;
+	}
+	ndev->hw_features |= ndev->features;
+
+	SET_NETDEV_DEV(ndev, &pdev->dev);
+	netif_stop_queue(ndev);
+	return ndev;
+}
+
 static int tn40_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
+	struct net_device *ndev;
+	struct tn40_priv *priv;
+	unsigned int nvec = 1;
+	void __iomem *regs;
 	int ret;
 
 	ret = pci_enable_device(pdev);
@@ -17,7 +1181,86 @@ static int tn40_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		dev_err(&pdev->dev, "failed to set DMA mask.\n");
 		goto err_disable_device;
 	}
+
+	ret = pci_request_regions(pdev, TN40_DRV_NAME);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to request PCI regions.\n");
+		goto err_disable_device;
+	}
+
+	pci_set_master(pdev);
+
+	regs = pci_iomap(pdev, 0, TN40_REGS_SIZE);
+	if (!regs) {
+		ret = -EIO;
+		dev_err(&pdev->dev, "failed to map PCI bar.\n");
+		goto err_free_regions;
+	}
+
+	ndev = tn40_netdev_alloc(pdev);
+	if (!ndev) {
+		ret = -ENOMEM;
+		dev_err(&pdev->dev, "failed to allocate netdev.\n");
+		goto err_iounmap;
+	}
+
+	priv = netdev_priv(ndev);
+	pci_set_drvdata(pdev, priv);
+
+	priv->regs = regs;
+	priv->pdev = pdev;
+	priv->ndev = ndev;
+	/* Initialize fifo sizes. */
+	priv->txd_size = 3;
+	priv->txf_size = 3;
+	priv->rxd_size = 3;
+	priv->rxf_size = 3;
+	/* Initialize the initial coalescing registers. */
+	priv->rdintcm = TN40_INT_REG_VAL(0x20, 1, 4, 12);
+	priv->tdintcm = TN40_INT_REG_VAL(0x20, 1, 0, 12);
+
+	ret = tn40_hw_reset(priv);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to reset HW.\n");
+		goto err_unset_drvdata;
+	}
+
+	ret = pci_alloc_irq_vectors(pdev, 1, nvec, PCI_IRQ_MSI);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "failed to allocate irq.\n");
+		goto err_unset_drvdata;
+	}
+
+	priv->stats_flag =
+		((tn40_read_reg(priv, TN40_FPGA_VER) & 0xFFF) != 308);
+	u64_stats_init(&priv->syncp);
+
+	priv->isr_mask = TN40_IR_RX_FREE_0 | TN40_IR_LNKCHG0 | TN40_IR_PSE |
+		TN40_IR_TMR0 | TN40_IR_RX_DESC_0 | TN40_IR_TX_FREE_0 |
+		TN40_IR_TMR1;
+
+	tn40_mac_init(priv);
+
+	ret = tn40_priv_init(priv);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to initialize tn40_priv.\n");
+		goto err_free_irq;
+	}
+
+	ret = register_netdev(ndev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to register netdev.\n");
+		goto err_free_irq;
+	}
 	return 0;
+err_free_irq:
+	pci_free_irq_vectors(pdev);
+err_unset_drvdata:
+	pci_set_drvdata(pdev, NULL);
+err_iounmap:
+	iounmap(regs);
+err_free_regions:
+	pci_release_regions(pdev);
 err_disable_device:
 	pci_disable_device(pdev);
 	return ret;
@@ -25,6 +1268,15 @@ err_disable_device:
 
 static void tn40_remove(struct pci_dev *pdev)
 {
+	struct tn40_priv *priv = pci_get_drvdata(pdev);
+	struct net_device *ndev = priv->ndev;
+
+	unregister_netdev(ndev);
+
+	pci_free_irq_vectors(priv->pdev);
+	pci_set_drvdata(pdev, NULL);
+	iounmap(priv->regs);
+	pci_release_regions(pdev);
 	pci_disable_device(pdev);
 }
 
@@ -51,4 +1303,5 @@ module_pci_driver(tn40_driver);
 
 MODULE_DEVICE_TABLE(pci, tn40_id_table);
 MODULE_LICENSE("GPL");
+MODULE_FIRMWARE(TN40_FIRMWARE_NAME);
 MODULE_DESCRIPTION("Tehuti Network TN40xx Driver");
diff --git a/drivers/net/ethernet/tehuti/tn40.h b/drivers/net/ethernet/tehuti/tn40.h
index aec23208e9b1..381257416d76 100644
--- a/drivers/net/ethernet/tehuti/tn40.h
+++ b/drivers/net/ethernet/tehuti/tn40.h
@@ -8,4 +8,161 @@
 
 #define TN40_DRV_NAME "tn40xx"
 
+#define TN40_MDIO_SPEED_1MHZ (1)
+#define TN40_MDIO_SPEED_6MHZ (6)
+
+/* netdev tx queue len for Luxor. The default value is 1000.
+ * ifconfig eth1 txqueuelen 3000 - to change it at runtime.
+ */
+#define TN40_NDEV_TXQ_LEN 1000
+
+#define TN40_FIFO_SIZE 4096
+#define TN40_FIFO_EXTRA_SPACE 1024
+
+#define TN40_TXF_DESC_SZ 16
+#define TN40_MAX_TX_LEVEL (priv->txd_fifo0.m.memsz - 16)
+#define TN40_MIN_TX_LEVEL 256
+#define TN40_NO_UPD_PACKETS 40
+#define TN40_MAX_MTU BIT(14)
+
+#define TN40_PCK_TH_MULT 128
+#define TN40_INT_COAL_MULT 2
+
+#define TN40_INT_REG_VAL(coal, coal_rc, rxf_th, pck_th) (	\
+	FIELD_PREP(GENMASK(14, 0), (coal)) |		\
+	FIELD_PREP(BIT(15), (coal_rc)) |		\
+	FIELD_PREP(GENMASK(19, 16), (rxf_th)) |		\
+	FIELD_PREP(GENMASK(31, 20), (pck_th))		\
+	)
+
+struct tn40_fifo {
+	dma_addr_t da; /* Physical address of fifo (used by HW) */
+	char *va; /* Virtual address of fifo (used by SW) */
+	u32 rptr, wptr;
+	 /* Cached values of RPTR and WPTR registers,
+	  * they're 32 bits on both 32 and 64 archs.
+	  */
+	u16 reg_cfg0;
+	u16 reg_cfg1;
+	u16 reg_rptr;
+	u16 reg_wptr;
+	u16 memsz; /* Memory size allocated for fifo */
+	u16 size_mask;
+	u16 pktsz; /* Skb packet size to allocate */
+	u16 rcvno; /* Number of buffers that come from this RXF */
+};
+
+struct tn40_txf_fifo {
+	struct tn40_fifo m; /* The minimal set of variables used by all fifos */
+};
+
+struct tn40_txd_fifo {
+	struct tn40_fifo m; /* The minimal set of variables used by all fifos */
+};
+
+union tn40_tx_dma_addr {
+	dma_addr_t dma;
+	struct sk_buff *skb;
+};
+
+/* Entry in the db.
+ * if len == 0 addr is dma
+ * if len != 0 addr is skb
+ */
+struct tn40_tx_map {
+	union tn40_tx_dma_addr addr;
+	int len;
+};
+
+/* tx database - implemented as circular fifo buffer */
+struct tn40_txdb {
+	struct tn40_tx_map *start; /* Points to the first element */
+	struct tn40_tx_map *end; /* Points just AFTER the last element */
+	struct tn40_tx_map *rptr; /* Points to the next element to read */
+	struct tn40_tx_map *wptr; /* Points to the next element to write */
+	int size; /* Number of elements in the db */
+};
+
+struct tn40_priv {
+	struct net_device *ndev;
+	struct pci_dev *pdev;
+
+	/* Tx FIFOs: 1 for data desc, 1 for empty (acks) desc */
+	struct tn40_txd_fifo txd_fifo0;
+	struct tn40_txf_fifo txf_fifo0;
+	struct tn40_txdb txdb;
+	int tx_level;
+	int tx_update_mark;
+	int tx_noupd;
+
+	int stats_flag;
+	struct rtnl_link_stats64 stats;
+	struct u64_stats_sync syncp;
+
+	u8 txd_size;
+	u8 txf_size;
+	u8 rxd_size;
+	u8 rxf_size;
+	u32 rdintcm;
+	u32 tdintcm;
+
+	u32 isr_mask;
+
+	void __iomem *regs;
+
+	/* SHORT_PKT_FIX */
+	u32 b0_len;
+	dma_addr_t b0_dma; /* Physical address of buffer */
+	char *b0_va; /* Virtual address of buffer */
+};
+
+#define TN40_MAX_PBL (19)
+/* PBL describes each virtual buffer to be transmitted from the host. */
+struct tn40_pbl {
+	__le32 pa_lo;
+	__le32 pa_hi;
+	__le32 len;
+};
+
+/* First word for TXD descriptor. It means: type = 3 for regular Tx packet,
+ * hw_csum = 7 for IP+UDP+TCP HW checksums.
+ */
+#define TN40_TXD_W1_VAL(bc, checksum, vtag, lgsnd, vlan_id) (		\
+	GENMASK(17, 16) |						\
+	FIELD_PREP(GENMASK(4, 0), (bc)) |				\
+	FIELD_PREP(GENMASK(7, 5), (checksum)) |				\
+	FIELD_PREP(BIT(8), (vtag)) |					\
+	FIELD_PREP(GENMASK(12, 9), (lgsnd)) |				\
+	FIELD_PREP(GENMASK(15, 13),					\
+		   FIELD_GET(GENMASK(15, 13), (vlan_id))) |		\
+	FIELD_PREP(GENMASK(31, 20),					\
+		   FIELD_GET(GENMASK(11, 0), (vlan_id)))		\
+	)
+
+struct tn40_txd_desc {
+	__le32 txd_val1;
+	__le16 mss;
+	__le16 length;
+	__le32 va_lo;
+	__le32 va_hi;
+	struct tn40_pbl pbl[]; /* Fragments */
+};
+
+struct tn40_txf_desc {
+	u32 status;
+	u32 va_lo; /* VAdr[31:0] */
+	u32 va_hi; /* VAdr[63:32] */
+	u32 pad;
+};
+
+static inline u32 tn40_read_reg(struct tn40_priv *priv, u32 reg)
+{
+	return readl(priv->regs + reg);
+}
+
+static inline void tn40_write_reg(struct tn40_priv *priv, u32 reg, u32 val)
+{
+	writel(val, priv->regs + reg);
+}
+
 #endif /* _TN40XX_H */
-- 
cgit v1.2.3-58-ga151


From 37c4947af44d51358cb03262c8e6eeac4a1cd642 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@gmail.com>
Date: Mon, 24 Jun 2024 08:55:05 +0900
Subject: net: tn40xx: add basic Rx handling

This patch adds basic Rx handling. The Rx logic uses three major data
structures; two ring buffers with NIC and one database. One ring
buffer is used to send information to NIC about memory to be stored
packets to be received. The other is used to get information from NIC
about received packets. The database is used to keep the information
about DMA mapping. After a packet arrived, the db is used to pass the
packet to the network stack.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@gmail.com>
Reviewed-by: Hans-Frieder Vogt <hfdevel@gmx.net>
Link: https://patch.msgid.link/20240623235507.108147-6-fujita.tomonori@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/tehuti/Kconfig |   1 +
 drivers/net/ethernet/tehuti/tn40.c  | 456 +++++++++++++++++++++++++++++++++++-
 drivers/net/ethernet/tehuti/tn40.h  |  52 ++++
 3 files changed, 508 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/tehuti/Kconfig b/drivers/net/ethernet/tehuti/Kconfig
index 4198fd59e42e..2b3b5a8c7fbf 100644
--- a/drivers/net/ethernet/tehuti/Kconfig
+++ b/drivers/net/ethernet/tehuti/Kconfig
@@ -26,6 +26,7 @@ config TEHUTI
 config TEHUTI_TN40
 	tristate "Tehuti Networks TN40xx 10G Ethernet adapters"
 	depends on PCI
+	select PAGE_POOL
 	select FW_LOADER
 	help
 	  This driver supports 10G Ethernet adapters using Tehuti Networks
diff --git a/drivers/net/ethernet/tehuti/tn40.c b/drivers/net/ethernet/tehuti/tn40.c
index ce3c323aa073..0b51a1d63aad 100644
--- a/drivers/net/ethernet/tehuti/tn40.c
+++ b/drivers/net/ethernet/tehuti/tn40.c
@@ -9,6 +9,7 @@
 #include <linux/netdevice.h>
 #include <linux/pci.h>
 #include <linux/vmalloc.h>
+#include <net/page_pool/helpers.h>
 
 #include "tn40.h"
 
@@ -63,6 +64,353 @@ static void tn40_fifo_free(struct tn40_priv *priv, struct tn40_fifo *f)
 			  f->memsz + TN40_FIFO_EXTRA_SPACE, f->va, f->da);
 }
 
+static struct tn40_rxdb *tn40_rxdb_alloc(int nelem)
+{
+	size_t size = sizeof(struct tn40_rxdb) + (nelem * sizeof(int)) +
+	    (nelem * sizeof(struct tn40_rx_map));
+	struct tn40_rxdb *db;
+	int i;
+
+	db = vzalloc(size);
+	if (db) {
+		db->stack = (int *)(db + 1);
+		db->elems = (void *)(db->stack + nelem);
+		db->nelem = nelem;
+		db->top = nelem;
+		/* make the first alloc close to db struct */
+		for (i = 0; i < nelem; i++)
+			db->stack[i] = nelem - i - 1;
+	}
+	return db;
+}
+
+static void tn40_rxdb_free(struct tn40_rxdb *db)
+{
+	vfree(db);
+}
+
+static int tn40_rxdb_alloc_elem(struct tn40_rxdb *db)
+{
+	return db->stack[--db->top];
+}
+
+static void *tn40_rxdb_addr_elem(struct tn40_rxdb *db, unsigned int n)
+{
+	return db->elems + n;
+}
+
+static int tn40_rxdb_available(struct tn40_rxdb *db)
+{
+	return db->top;
+}
+
+static void tn40_rxdb_free_elem(struct tn40_rxdb *db, unsigned int n)
+{
+	db->stack[db->top++] = n;
+}
+
+/**
+ * tn40_create_rx_ring - Initialize RX all related HW and SW resources
+ * @priv: NIC private structure
+ *
+ * create_rx_ring creates rxf and rxd fifos, updates the relevant HW registers,
+ * preallocates skbs for rx. It assumes that Rx is disabled in HW funcs are
+ * grouped for better cache usage
+ *
+ * RxD fifo is smaller then RxF fifo by design. Upon high load, RxD will be
+ * filled and packets will be dropped by the NIC without getting into the host
+ * or generating interrupts. In this situation the host has no chance of
+ * processing all the packets. Dropping packets by the NIC is cheaper, since it
+ * takes 0 CPU cycles.
+ *
+ * Return: 0 on success and negative value on error.
+ */
+static int tn40_create_rx_ring(struct tn40_priv *priv)
+{
+	struct page_pool_params pp = {
+		.dev = &priv->pdev->dev,
+		.napi = &priv->napi,
+		.dma_dir = DMA_FROM_DEVICE,
+		.netdev = priv->ndev,
+		.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV,
+		.max_len = PAGE_SIZE,
+	};
+	int ret, pkt_size, nr;
+
+	priv->page_pool = page_pool_create(&pp);
+	if (IS_ERR(priv->page_pool))
+		return PTR_ERR(priv->page_pool);
+
+	ret = tn40_fifo_alloc(priv, &priv->rxd_fifo0.m, priv->rxd_size,
+			      TN40_REG_RXD_CFG0_0, TN40_REG_RXD_CFG1_0,
+			      TN40_REG_RXD_RPTR_0, TN40_REG_RXD_WPTR_0);
+	if (ret)
+		goto err_destroy_page_pool;
+
+	ret = tn40_fifo_alloc(priv, &priv->rxf_fifo0.m, priv->rxf_size,
+			      TN40_REG_RXF_CFG0_0, TN40_REG_RXF_CFG1_0,
+			      TN40_REG_RXF_RPTR_0, TN40_REG_RXF_WPTR_0);
+	if (ret)
+		goto err_free_rxd;
+
+	pkt_size = priv->ndev->mtu + VLAN_ETH_HLEN;
+	priv->rxf_fifo0.m.pktsz = pkt_size;
+	nr = priv->rxf_fifo0.m.memsz / sizeof(struct tn40_rxf_desc);
+	priv->rxdb0 = tn40_rxdb_alloc(nr);
+	if (!priv->rxdb0) {
+		ret = -ENOMEM;
+		goto err_free_rxf;
+	}
+	return 0;
+err_free_rxf:
+	tn40_fifo_free(priv, &priv->rxf_fifo0.m);
+err_free_rxd:
+	tn40_fifo_free(priv, &priv->rxd_fifo0.m);
+err_destroy_page_pool:
+	page_pool_destroy(priv->page_pool);
+	return ret;
+}
+
+static void tn40_rx_free_buffers(struct tn40_priv *priv)
+{
+	struct tn40_rxdb *db = priv->rxdb0;
+	struct tn40_rx_map *dm;
+	u16 i;
+
+	netdev_dbg(priv->ndev, "total =%d free =%d busy =%d\n", db->nelem,
+		   tn40_rxdb_available(db),
+		   db->nelem - tn40_rxdb_available(db));
+
+	for (i = 0; i < db->nelem; i++) {
+		dm = tn40_rxdb_addr_elem(db, i);
+		if (dm->page)
+			page_pool_put_full_page(priv->page_pool, dm->page,
+						false);
+	}
+}
+
+static void tn40_destroy_rx_ring(struct tn40_priv *priv)
+{
+	if (priv->rxdb0) {
+		tn40_rx_free_buffers(priv);
+		tn40_rxdb_free(priv->rxdb0);
+		priv->rxdb0 = NULL;
+	}
+	tn40_fifo_free(priv, &priv->rxf_fifo0.m);
+	tn40_fifo_free(priv, &priv->rxd_fifo0.m);
+	page_pool_destroy(priv->page_pool);
+}
+
+static void tn40_set_rx_desc(struct tn40_priv *priv, int idx, u64 dma)
+{
+	struct tn40_rxf_fifo *f = &priv->rxf_fifo0;
+	struct tn40_rxf_desc *rxfd;
+	int delta;
+
+	rxfd = (struct tn40_rxf_desc *)(f->m.va + f->m.wptr);
+	rxfd->info = cpu_to_le32(0x10003);	/* INFO =1 BC =3 */
+	rxfd->va_lo = cpu_to_le32(idx);
+	rxfd->pa_lo = cpu_to_le32(lower_32_bits(dma));
+	rxfd->pa_hi = cpu_to_le32(upper_32_bits(dma));
+	rxfd->len = cpu_to_le32(f->m.pktsz);
+	f->m.wptr += sizeof(struct tn40_rxf_desc);
+	delta = f->m.wptr - f->m.memsz;
+	if (unlikely(delta >= 0)) {
+		f->m.wptr = delta;
+		if (delta > 0) {
+			memcpy(f->m.va, f->m.va + f->m.memsz, delta);
+			netdev_dbg(priv->ndev,
+				   "wrapped rxd descriptor\n");
+		}
+	}
+}
+
+/**
+ * tn40_rx_alloc_buffers - Fill rxf fifo with buffers.
+ *
+ * @priv: NIC's private structure
+ *
+ * rx_alloc_buffers allocates buffers via the page pool API, builds rxf descs
+ * and pushes them (rxf descr) into the rxf fifo. The pages are stored in rxdb.
+ * To calculate the free space, we uses the cached values of RPTR and WPTR
+ * when needed. This function also updates RPTR and WPTR.
+ */
+static void tn40_rx_alloc_buffers(struct tn40_priv *priv)
+{
+	struct tn40_rxf_fifo *f = &priv->rxf_fifo0;
+	struct tn40_rxdb *db = priv->rxdb0;
+	struct tn40_rx_map *dm;
+	struct page *page;
+	int dno, i, idx;
+
+	dno = tn40_rxdb_available(db) - 1;
+	for (i = dno; i > 0; i--) {
+		page = page_pool_dev_alloc_pages(priv->page_pool);
+		if (!page)
+			break;
+
+		idx = tn40_rxdb_alloc_elem(db);
+		tn40_set_rx_desc(priv, idx, page_pool_get_dma_addr(page));
+		dm = tn40_rxdb_addr_elem(db, idx);
+		dm->page = page;
+	}
+	if (i != dno)
+		tn40_write_reg(priv, f->m.reg_wptr,
+			       f->m.wptr & TN40_TXF_WPTR_WR_PTR);
+	netdev_dbg(priv->ndev, "write_reg 0x%04x f->m.reg_wptr 0x%x\n",
+		   f->m.reg_wptr, f->m.wptr & TN40_TXF_WPTR_WR_PTR);
+	netdev_dbg(priv->ndev, "read_reg  0x%04x f->m.reg_rptr=0x%x\n",
+		   f->m.reg_rptr, tn40_read_reg(priv, f->m.reg_rptr));
+	netdev_dbg(priv->ndev, "write_reg 0x%04x f->m.reg_wptr=0x%x\n",
+		   f->m.reg_wptr, tn40_read_reg(priv, f->m.reg_wptr));
+}
+
+static void tn40_recycle_rx_buffer(struct tn40_priv *priv,
+				   struct tn40_rxd_desc *rxdd)
+{
+	struct tn40_rxf_fifo *f = &priv->rxf_fifo0;
+	struct tn40_rx_map *dm;
+	int idx;
+
+	idx = le32_to_cpu(rxdd->va_lo);
+	dm = tn40_rxdb_addr_elem(priv->rxdb0, idx);
+	tn40_set_rx_desc(priv, idx, page_pool_get_dma_addr(dm->page));
+
+	tn40_write_reg(priv, f->m.reg_wptr, f->m.wptr & TN40_TXF_WPTR_WR_PTR);
+}
+
+static int tn40_rx_receive(struct tn40_priv *priv, int budget)
+{
+	struct tn40_rxd_fifo *f = &priv->rxd_fifo0;
+	u32 rxd_val1, rxd_err, pkt_id;
+	int tmp_len, size, done = 0;
+	struct tn40_rxdb *db = NULL;
+	struct tn40_rxd_desc *rxdd;
+	struct tn40_rx_map *dm;
+	struct sk_buff *skb;
+	u16 len, rxd_vlan;
+	int idx;
+
+	f->m.wptr = tn40_read_reg(priv, f->m.reg_wptr) & TN40_TXF_WPTR_WR_PTR;
+	size = f->m.wptr - f->m.rptr;
+	if (size < 0)
+		size += f->m.memsz;	/* Size is negative :-) */
+
+	while (size > 0) {
+		rxdd = (struct tn40_rxd_desc *)(f->m.va + f->m.rptr);
+		db = priv->rxdb0;
+
+		/* We have a chicken and egg problem here. If the
+		 * descriptor is wrapped we first need to copy the tail
+		 * of the descriptor to the end of the buffer before
+		 * extracting values from the descriptor. However in
+		 * order to know if the descriptor is wrapped we need to
+		 * obtain the length of the descriptor from (the
+		 * wrapped) descriptor. Luckily the length is the first
+		 * word of the descriptor. Descriptor lengths are
+		 * multiples of 8 bytes so in case of a wrapped
+		 * descriptor the first 8 bytes guaranteed to appear
+		 * before the end of the buffer. We first obtain the
+		 * length, we then copy the rest of the descriptor if
+		 * needed and then extract the rest of the values from
+		 * the descriptor.
+		 *
+		 * Do not change the order of operations as it will
+		 * break the code!!!
+		 */
+		rxd_val1 = le32_to_cpu(rxdd->rxd_val1);
+		tmp_len = TN40_GET_RXD_BC(rxd_val1) << 3;
+		pkt_id = TN40_GET_RXD_PKT_ID(rxd_val1);
+		size -= tmp_len;
+		/* CHECK FOR A PARTIALLY ARRIVED DESCRIPTOR */
+		if (size < 0) {
+			netdev_dbg(priv->ndev,
+				   "%s partially arrived desc tmp_len %d\n",
+				   __func__, tmp_len);
+			break;
+		}
+		/* make sure that the descriptor fully is arrived
+		 * before reading the rest of the descriptor.
+		 */
+		rmb();
+
+		/* A special treatment is given to non-contiguous
+		 * descriptors that start near the end, wraps around
+		 * and continue at the beginning. The second part is
+		 * copied right after the first, and then descriptor
+		 * is interpreted as normal. The fifo has an extra
+		 * space to allow such operations.
+		 */
+
+		/* HAVE WE REACHED THE END OF THE QUEUE? */
+		f->m.rptr += tmp_len;
+		tmp_len = f->m.rptr - f->m.memsz;
+		if (unlikely(tmp_len >= 0)) {
+			f->m.rptr = tmp_len;
+			if (tmp_len > 0) {
+				/* COPY PARTIAL DESCRIPTOR
+				 * TO THE END OF THE QUEUE
+				 */
+				netdev_dbg(priv->ndev,
+					   "wrapped desc rptr=%d tmp_len=%d\n",
+					   f->m.rptr, tmp_len);
+				memcpy(f->m.va + f->m.memsz, f->m.va, tmp_len);
+			}
+		}
+		idx = le32_to_cpu(rxdd->va_lo);
+		dm = tn40_rxdb_addr_elem(db, idx);
+		prefetch(dm);
+
+		len = le16_to_cpu(rxdd->len);
+		rxd_vlan = le16_to_cpu(rxdd->rxd_vlan);
+		/* CHECK FOR ERRORS */
+		rxd_err = TN40_GET_RXD_ERR(rxd_val1);
+		if (unlikely(rxd_err)) {
+			u64_stats_update_begin(&priv->syncp);
+			priv->stats.rx_errors++;
+			u64_stats_update_end(&priv->syncp);
+			tn40_recycle_rx_buffer(priv, rxdd);
+			continue;
+		}
+
+		skb = napi_build_skb(page_address(dm->page), PAGE_SIZE);
+		if (!skb) {
+			u64_stats_update_begin(&priv->syncp);
+			priv->stats.rx_dropped++;
+			u64_stats_update_end(&priv->syncp);
+			tn40_recycle_rx_buffer(priv, rxdd);
+			break;
+		}
+		skb_mark_for_recycle(skb);
+		skb_put(skb, len);
+		skb->protocol = eth_type_trans(skb, priv->ndev);
+		skb->ip_summed =
+		    (pkt_id == 0) ? CHECKSUM_NONE : CHECKSUM_UNNECESSARY;
+		if (TN40_GET_RXD_VTAG(rxd_val1))
+			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
+					       TN40_GET_RXD_VLAN_TCI(rxd_vlan));
+
+		dm->page = NULL;
+		tn40_rxdb_free_elem(db, idx);
+
+		napi_gro_receive(&priv->napi, skb);
+
+		u64_stats_update_begin(&priv->syncp);
+		priv->stats.rx_bytes += len;
+		u64_stats_update_end(&priv->syncp);
+
+		if (unlikely(++done >= budget))
+			break;
+	}
+	u64_stats_update_begin(&priv->syncp);
+	priv->stats.rx_packets += done;
+	u64_stats_update_end(&priv->syncp);
+	/* FIXME: Do something to minimize pci accesses */
+	tn40_write_reg(priv, f->m.reg_rptr, f->m.rptr & TN40_TXF_WPTR_WR_PTR);
+	tn40_rx_alloc_buffers(priv);
+	return done;
+}
+
 /* TX HW/SW interaction overview
  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  * There are 2 types of TX communication channels between driver and NIC.
@@ -464,6 +812,61 @@ static netdev_tx_t tn40_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 	return NETDEV_TX_OK;
 }
 
+static void tn40_tx_cleanup(struct tn40_priv *priv)
+{
+	struct tn40_txf_fifo *f = &priv->txf_fifo0;
+	struct tn40_txdb *db = &priv->txdb;
+	int tx_level = 0;
+
+	f->m.wptr = tn40_read_reg(priv, f->m.reg_wptr) & TN40_TXF_WPTR_MASK;
+
+	netif_tx_lock(priv->ndev);
+	while (f->m.wptr != f->m.rptr) {
+		f->m.rptr += TN40_TXF_DESC_SZ;
+		f->m.rptr &= f->m.size_mask;
+		/* Unmap all fragments */
+		/* First has to come tx_maps containing DMA */
+		do {
+			dma_addr_t addr = db->rptr->addr.dma;
+			size_t size =  db->rptr->len;
+
+			netif_tx_unlock(priv->ndev);
+			dma_unmap_page(&priv->pdev->dev, addr,
+				       size, DMA_TO_DEVICE);
+			netif_tx_lock(priv->ndev);
+			tn40_tx_db_inc_rptr(db);
+		} while (db->rptr->len > 0);
+		tx_level -= db->rptr->len; /* '-' Because the len is negative */
+
+		/* Now should come skb pointer - free it */
+		dev_kfree_skb_any(db->rptr->addr.skb);
+		netdev_dbg(priv->ndev, "dev_kfree_skb_any %p %d\n",
+			   db->rptr->addr.skb, -db->rptr->len);
+		tn40_tx_db_inc_rptr(db);
+	}
+
+	/* Let the HW know which TXF descriptors were cleaned */
+	tn40_write_reg(priv, f->m.reg_rptr, f->m.rptr & TN40_TXF_WPTR_WR_PTR);
+
+	/* We reclaimed resources, so in case the Q is stopped by xmit
+	 * callback, we resume the transmission and use tx_lock to
+	 * synchronize with xmit.
+	 */
+	priv->tx_level += tx_level;
+	if (priv->tx_noupd) {
+		priv->tx_noupd = 0;
+		tn40_write_reg(priv, priv->txd_fifo0.m.reg_wptr,
+			       priv->txd_fifo0.m.wptr & TN40_TXF_WPTR_WR_PTR);
+	}
+	if (unlikely(netif_queue_stopped(priv->ndev) &&
+		     netif_carrier_ok(priv->ndev) &&
+		     (priv->tx_level >= TN40_MAX_TX_LEVEL / 2))) {
+		netdev_dbg(priv->ndev, "TX Q WAKE level %d\n", priv->tx_level);
+		netif_wake_queue(priv->ndev);
+	}
+	netif_tx_unlock(priv->ndev);
+}
+
 static void tn40_tx_free_skbs(struct tn40_priv *priv)
 {
 	struct tn40_txdb *db = &priv->txdb;
@@ -725,6 +1128,10 @@ static irqreturn_t tn40_isr_napi(int irq, void *dev)
 		tn40_isr_extra(priv, isr);
 
 	if (isr & (TN40_IR_RX_DESC_0 | TN40_IR_TX_FREE_0 | TN40_IR_TMR1)) {
+		if (likely(napi_schedule_prep(&priv->napi))) {
+			__napi_schedule(&priv->napi);
+			return IRQ_HANDLED;
+		}
 		/* We get here if an interrupt has slept into the
 		 * small time window between these lines in
 		 * tn40_poll: tn40_enable_interrupts(priv); return 0;
@@ -742,6 +1149,25 @@ static irqreturn_t tn40_isr_napi(int irq, void *dev)
 	return IRQ_HANDLED;
 }
 
+static int tn40_poll(struct napi_struct *napi, int budget)
+{
+	struct tn40_priv *priv = container_of(napi, struct tn40_priv, napi);
+	int work_done;
+
+	tn40_tx_cleanup(priv);
+
+	if (!budget)
+		return 0;
+
+	work_done = tn40_rx_receive(priv, budget);
+	if (work_done == budget)
+		return budget;
+
+	if (napi_complete_done(napi, work_done))
+		tn40_enable_interrupts(priv);
+	return work_done;
+}
+
 static int tn40_fw_load(struct tn40_priv *priv)
 {
 	const struct firmware *fw = NULL;
@@ -822,6 +1248,8 @@ static void tn40_hw_start(struct tn40_priv *priv)
 	tn40_write_reg(priv, TN40_REG_TX_FULLNESS, 0);
 
 	tn40_write_reg(priv, TN40_REG_VGLB, 0);
+	tn40_write_reg(priv, TN40_REG_MAX_FRAME_A,
+		       priv->rxf_fifo0.m.pktsz & TN40_MAX_FRAME_AB_VAL);
 	tn40_write_reg(priv, TN40_REG_RDINTCM0, priv->rdintcm);
 	tn40_write_reg(priv, TN40_REG_RDINTCM2, 0);
 
@@ -921,15 +1349,30 @@ static int tn40_start(struct tn40_priv *priv)
 		return ret;
 	}
 
+	ret = tn40_create_rx_ring(priv);
+	if (ret) {
+		netdev_err(priv->ndev, "failed to rx init %d\n", ret);
+		goto err_tx_ring;
+	}
+
+	tn40_rx_alloc_buffers(priv);
+	if (tn40_rxdb_available(priv->rxdb0) != 1) {
+		ret = -ENOMEM;
+		netdev_err(priv->ndev, "failed to allocate rx buffers\n");
+		goto err_rx_ring;
+	}
+
 	ret = request_irq(priv->pdev->irq, &tn40_isr_napi, IRQF_SHARED,
 			  priv->ndev->name, priv->ndev);
 	if (ret) {
 		netdev_err(priv->ndev, "failed to request irq %d\n", ret);
-		goto err_tx_ring;
+		goto err_rx_ring;
 	}
 
 	tn40_hw_start(priv);
 	return 0;
+err_rx_ring:
+	tn40_destroy_rx_ring(priv);
 err_tx_ring:
 	tn40_destroy_tx_ring(priv);
 	return ret;
@@ -941,12 +1384,15 @@ static void tn40_stop(struct tn40_priv *priv)
 	free_irq(priv->pdev->irq, priv->ndev);
 	tn40_sw_reset(priv);
 	tn40_destroy_tx_ring(priv);
+	tn40_destroy_rx_ring(priv);
 }
 
 static int tn40_close(struct net_device *ndev)
 {
 	struct tn40_priv *priv = netdev_priv(ndev);
 
+	napi_disable(&priv->napi);
+	netif_napi_del(&priv->napi);
 	tn40_stop(priv);
 	return 0;
 }
@@ -962,6 +1408,8 @@ static int tn40_open(struct net_device *dev)
 		netdev_err(dev, "failed to start %d\n", ret);
 		return ret;
 	}
+	napi_enable(&priv->napi);
+	netif_start_queue(priv->ndev);
 	return 0;
 }
 
@@ -1092,6 +1540,11 @@ static void tn40_get_stats(struct net_device *ndev,
 		stats->tx_packets = priv->stats.tx_packets;
 		stats->tx_bytes = priv->stats.tx_bytes;
 		stats->tx_dropped = priv->stats.tx_dropped;
+
+		stats->rx_packets = priv->stats.rx_packets;
+		stats->rx_bytes = priv->stats.rx_bytes;
+		stats->rx_dropped = priv->stats.rx_dropped;
+		stats->rx_errors = priv->stats.rx_errors;
 	} while (u64_stats_fetch_retry(&priv->syncp, start));
 }
 
@@ -1206,6 +1659,7 @@ static int tn40_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	priv = netdev_priv(ndev);
 	pci_set_drvdata(pdev, priv);
+	netif_napi_add(ndev, &priv->napi, tn40_poll);
 
 	priv->regs = regs;
 	priv->pdev = pdev;
diff --git a/drivers/net/ethernet/tehuti/tn40.h b/drivers/net/ethernet/tehuti/tn40.h
index 381257416d76..20b400e58ef9 100644
--- a/drivers/net/ethernet/tehuti/tn40.h
+++ b/drivers/net/ethernet/tehuti/tn40.h
@@ -60,6 +60,25 @@ struct tn40_txd_fifo {
 	struct tn40_fifo m; /* The minimal set of variables used by all fifos */
 };
 
+struct tn40_rxf_fifo {
+	struct tn40_fifo m; /* The minimal set of variables used by all fifos */
+};
+
+struct tn40_rxd_fifo {
+	struct tn40_fifo m; /* The minimal set of variables used by all fifos */
+};
+
+struct tn40_rx_map {
+	struct page *page;
+};
+
+struct tn40_rxdb {
+	unsigned int *stack;
+	struct tn40_rx_map *elems;
+	unsigned int nelem;
+	unsigned int top;
+};
+
 union tn40_tx_dma_addr {
 	dma_addr_t dma;
 	struct sk_buff *skb;
@@ -87,6 +106,13 @@ struct tn40_priv {
 	struct net_device *ndev;
 	struct pci_dev *pdev;
 
+	struct napi_struct napi;
+	/* RX FIFOs: 1 for data (full) descs, and 2 for free descs */
+	struct tn40_rxd_fifo rxd_fifo0;
+	struct tn40_rxf_fifo rxf_fifo0;
+	struct tn40_rxdb *rxdb0; /* Rx dbs to store skb pointers */
+	struct page_pool *page_pool;
+
 	/* Tx FIFOs: 1 for data desc, 1 for empty (acks) desc */
 	struct tn40_txd_fifo txd_fifo0;
 	struct tn40_txf_fifo txf_fifo0;
@@ -116,6 +142,32 @@ struct tn40_priv {
 	char *b0_va; /* Virtual address of buffer */
 };
 
+/* RX FREE descriptor - 64bit */
+struct tn40_rxf_desc {
+	__le32 info; /* Buffer Count + Info - described below */
+	__le32 va_lo; /* VAdr[31:0] */
+	__le32 va_hi; /* VAdr[63:32] */
+	__le32 pa_lo; /* PAdr[31:0] */
+	__le32 pa_hi; /* PAdr[63:32] */
+	__le32 len; /* Buffer Length */
+};
+
+#define TN40_GET_RXD_BC(x) FIELD_GET(GENMASK(4, 0), (x))
+#define TN40_GET_RXD_ERR(x) FIELD_GET(GENMASK(26, 21), (x))
+#define TN40_GET_RXD_PKT_ID(x) FIELD_GET(GENMASK(30, 28), (x))
+#define TN40_GET_RXD_VTAG(x) FIELD_GET(BIT(31), (x))
+#define TN40_GET_RXD_VLAN_TCI(x) FIELD_GET(GENMASK(15, 0), (x))
+
+struct tn40_rxd_desc {
+	__le32 rxd_val1;
+	__le16 len;
+	__le16 rxd_vlan;
+	__le32 va_lo;
+	__le32 va_hi;
+	__le32 rss_lo;
+	__le32 rss_hash;
+};
+
 #define TN40_MAX_PBL (19)
 /* PBL describes each virtual buffer to be transmitted from the host. */
 struct tn40_pbl {
-- 
cgit v1.2.3-58-ga151


From 7fdbd2f2bb5dc33c50604ea66eb8ae16247cc518 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@gmail.com>
Date: Mon, 24 Jun 2024 08:55:06 +0900
Subject: net: tn40xx: add mdio bus support

This patch adds supports for mdio bus. A later path adds PHYLIB
support on the top of this.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@gmail.com>
Link: https://patch.msgid.link/20240623235507.108147-7-fujita.tomonori@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/tehuti/Makefile    |   2 +-
 drivers/net/ethernet/tehuti/tn40.h      |   4 +
 drivers/net/ethernet/tehuti/tn40_mdio.c | 142 ++++++++++++++++++++++++++++++++
 3 files changed, 147 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/ethernet/tehuti/tn40_mdio.c

diff --git a/drivers/net/ethernet/tehuti/Makefile b/drivers/net/ethernet/tehuti/Makefile
index 1c468d99e476..7a0fe586a243 100644
--- a/drivers/net/ethernet/tehuti/Makefile
+++ b/drivers/net/ethernet/tehuti/Makefile
@@ -5,5 +5,5 @@
 
 obj-$(CONFIG_TEHUTI) += tehuti.o
 
-tn40xx-y := tn40.o
+tn40xx-y := tn40.o tn40_mdio.o
 obj-$(CONFIG_TEHUTI_TN40) += tn40xx.o
diff --git a/drivers/net/ethernet/tehuti/tn40.h b/drivers/net/ethernet/tehuti/tn40.h
index 20b400e58ef9..4cbe0c3883c7 100644
--- a/drivers/net/ethernet/tehuti/tn40.h
+++ b/drivers/net/ethernet/tehuti/tn40.h
@@ -140,6 +140,8 @@ struct tn40_priv {
 	u32 b0_len;
 	dma_addr_t b0_dma; /* Physical address of buffer */
 	char *b0_va; /* Virtual address of buffer */
+
+	struct mii_bus *mdio;
 };
 
 /* RX FREE descriptor - 64bit */
@@ -217,4 +219,6 @@ static inline void tn40_write_reg(struct tn40_priv *priv, u32 reg, u32 val)
 	writel(val, priv->regs + reg);
 }
 
+int tn40_mdiobus_init(struct tn40_priv *priv);
+
 #endif /* _TN40XX_H */
diff --git a/drivers/net/ethernet/tehuti/tn40_mdio.c b/drivers/net/ethernet/tehuti/tn40_mdio.c
new file mode 100644
index 000000000000..af18615d64a8
--- /dev/null
+++ b/drivers/net/ethernet/tehuti/tn40_mdio.c
@@ -0,0 +1,142 @@
+// SPDX-License-Identifier: GPL-2.0+
+/* Copyright (c) Tehuti Networks Ltd. */
+
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+#include <linux/phylink.h>
+
+#include "tn40.h"
+
+#define TN40_MDIO_DEVAD_MASK GENMASK(4, 0)
+#define TN40_MDIO_PRTAD_MASK GENMASK(9, 5)
+#define TN40_MDIO_CMD_VAL(device, port)			\
+	(FIELD_PREP(TN40_MDIO_DEVAD_MASK, (device)) |	\
+	 (FIELD_PREP(TN40_MDIO_PRTAD_MASK, (port))))
+#define TN40_MDIO_CMD_READ BIT(15)
+
+static void tn40_mdio_set_speed(struct tn40_priv *priv, u32 speed)
+{
+	void __iomem *regs = priv->regs;
+	int mdio_cfg;
+
+	if (speed == TN40_MDIO_SPEED_1MHZ)
+		mdio_cfg = (0x7d << 7) | 0x08;	/* 1MHz */
+	else
+		mdio_cfg = 0xA08;	/* 6MHz */
+	mdio_cfg |= (1 << 6);
+	writel(mdio_cfg, regs + TN40_REG_MDIO_CMD_STAT);
+	msleep(100);
+}
+
+static u32 tn40_mdio_stat(struct tn40_priv *priv)
+{
+	void __iomem *regs = priv->regs;
+
+	return readl(regs + TN40_REG_MDIO_CMD_STAT);
+}
+
+static int tn40_mdio_wait_nobusy(struct tn40_priv *priv, u32 *val)
+{
+	u32 stat;
+	int ret;
+
+	ret = readx_poll_timeout_atomic(tn40_mdio_stat, priv, stat,
+					TN40_GET_MDIO_BUSY(stat) == 0, 10,
+					10000);
+	if (val)
+		*val = stat;
+	return ret;
+}
+
+static int tn40_mdio_read(struct tn40_priv *priv, int port, int device,
+			  u16 regnum)
+{
+	void __iomem *regs = priv->regs;
+	u32 i;
+
+	/* wait until MDIO is not busy */
+	if (tn40_mdio_wait_nobusy(priv, NULL))
+		return -EIO;
+
+	i = TN40_MDIO_CMD_VAL(device, port);
+	writel(i, regs + TN40_REG_MDIO_CMD);
+	writel((u32)regnum, regs + TN40_REG_MDIO_ADDR);
+	if (tn40_mdio_wait_nobusy(priv, NULL))
+		return -EIO;
+
+	writel(TN40_MDIO_CMD_READ | i, regs + TN40_REG_MDIO_CMD);
+	/* read CMD_STAT until not busy */
+	if (tn40_mdio_wait_nobusy(priv, NULL))
+		return -EIO;
+
+	return lower_16_bits(readl(regs + TN40_REG_MDIO_DATA));
+}
+
+static int tn40_mdio_write(struct tn40_priv *priv, int port, int device,
+			   u16 regnum, u16 data)
+{
+	void __iomem *regs = priv->regs;
+	u32 tmp_reg = 0;
+	int ret;
+
+	/* wait until MDIO is not busy */
+	if (tn40_mdio_wait_nobusy(priv, NULL))
+		return -EIO;
+	writel(TN40_MDIO_CMD_VAL(device, port), regs + TN40_REG_MDIO_CMD);
+	writel((u32)regnum, regs + TN40_REG_MDIO_ADDR);
+	if (tn40_mdio_wait_nobusy(priv, NULL))
+		return -EIO;
+	writel((u32)data, regs + TN40_REG_MDIO_DATA);
+	/* read CMD_STAT until not busy */
+	ret = tn40_mdio_wait_nobusy(priv, &tmp_reg);
+	if (ret)
+		return -EIO;
+
+	if (TN40_GET_MDIO_RD_ERR(tmp_reg)) {
+		dev_err(&priv->pdev->dev, "MDIO error after write command\n");
+		return -EIO;
+	}
+	return 0;
+}
+
+static int tn40_mdio_read_c45(struct mii_bus *mii_bus, int addr, int devnum,
+			      int regnum)
+{
+	return tn40_mdio_read(mii_bus->priv, addr, devnum, regnum);
+}
+
+static int tn40_mdio_write_c45(struct mii_bus *mii_bus, int addr, int devnum,
+			       int regnum, u16 val)
+{
+	return  tn40_mdio_write(mii_bus->priv, addr, devnum, regnum, val);
+}
+
+int tn40_mdiobus_init(struct tn40_priv *priv)
+{
+	struct pci_dev *pdev = priv->pdev;
+	struct mii_bus *bus;
+	int ret;
+
+	bus = devm_mdiobus_alloc(&pdev->dev);
+	if (!bus)
+		return -ENOMEM;
+
+	bus->name = TN40_DRV_NAME;
+	bus->parent = &pdev->dev;
+	snprintf(bus->id, MII_BUS_ID_SIZE, "tn40xx-%x-%x",
+		 pci_domain_nr(pdev->bus), pci_dev_id(pdev));
+	bus->priv = priv;
+
+	bus->read_c45 = tn40_mdio_read_c45;
+	bus->write_c45 = tn40_mdio_write_c45;
+
+	ret = devm_mdiobus_register(&pdev->dev, bus);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to register mdiobus %d %u %u\n",
+			ret, bus->state, MDIOBUS_UNREGISTERED);
+		return ret;
+	}
+	tn40_mdio_set_speed(priv, TN40_MDIO_SPEED_6MHZ);
+	priv->mdio = bus;
+	return 0;
+}
-- 
cgit v1.2.3-58-ga151


From 30824122422476bede5d8a25ed25bf7b720dc193 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@gmail.com>
Date: Mon, 24 Jun 2024 08:55:07 +0900
Subject: net: tn40xx: add phylink support

This patch adds supports for multiple PHY hardware with phylink. The
adapters with TN40xx chips use multiple PHY hardware; AMCC QT2025, TI
TLK10232, Aqrate AQR105, and Marvell 88X3120, 88X3310, and MV88E2010.

For now, the PCI ID table of this driver enables adapters using only
QT2025 PHY. I've tested this driver and the QT2025 PHY driver (SFP+
10G SR) with Edimax EN-9320 10G adapter.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@gmail.com>
Reviewed-by: Hans-Frieder Vogt <hfdevel@gmx.net>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://patch.msgid.link/20240623235507.108147-8-fujita.tomonori@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/tehuti/Kconfig    |  1 +
 drivers/net/ethernet/tehuti/Makefile   |  2 +-
 drivers/net/ethernet/tehuti/tn40.c     | 31 ++++++++++++--
 drivers/net/ethernet/tehuti/tn40.h     |  8 ++++
 drivers/net/ethernet/tehuti/tn40_phy.c | 76 ++++++++++++++++++++++++++++++++++
 5 files changed, 114 insertions(+), 4 deletions(-)
 create mode 100644 drivers/net/ethernet/tehuti/tn40_phy.c

diff --git a/drivers/net/ethernet/tehuti/Kconfig b/drivers/net/ethernet/tehuti/Kconfig
index 2b3b5a8c7fbf..6db2c9817445 100644
--- a/drivers/net/ethernet/tehuti/Kconfig
+++ b/drivers/net/ethernet/tehuti/Kconfig
@@ -28,6 +28,7 @@ config TEHUTI_TN40
 	depends on PCI
 	select PAGE_POOL
 	select FW_LOADER
+	select PHYLINK
 	help
 	  This driver supports 10G Ethernet adapters using Tehuti Networks
 	  TN40xx chips. Currently, adapters with Applied Micro Circuits
diff --git a/drivers/net/ethernet/tehuti/Makefile b/drivers/net/ethernet/tehuti/Makefile
index 7a0fe586a243..0d4f4d63a65c 100644
--- a/drivers/net/ethernet/tehuti/Makefile
+++ b/drivers/net/ethernet/tehuti/Makefile
@@ -5,5 +5,5 @@
 
 obj-$(CONFIG_TEHUTI) += tehuti.o
 
-tn40xx-y := tn40.o tn40_mdio.o
+tn40xx-y := tn40.o tn40_mdio.o tn40_phy.o
 obj-$(CONFIG_TEHUTI_TN40) += tn40xx.o
diff --git a/drivers/net/ethernet/tehuti/tn40.c b/drivers/net/ethernet/tehuti/tn40.c
index 0b51a1d63aad..11db9fde11fe 100644
--- a/drivers/net/ethernet/tehuti/tn40.c
+++ b/drivers/net/ethernet/tehuti/tn40.c
@@ -8,6 +8,7 @@
 #include <linux/iopoll.h>
 #include <linux/netdevice.h>
 #include <linux/pci.h>
+#include <linux/phylink.h>
 #include <linux/vmalloc.h>
 #include <net/page_pool/helpers.h>
 
@@ -966,7 +967,7 @@ static void tn40_tx_push_desc_safe(struct tn40_priv *priv, void *data, int size)
 	}
 }
 
-static int tn40_set_link_speed(struct tn40_priv *priv, u32 speed)
+int tn40_set_link_speed(struct tn40_priv *priv, u32 speed)
 {
 	u32 val;
 	int i;
@@ -1391,6 +1392,9 @@ static int tn40_close(struct net_device *ndev)
 {
 	struct tn40_priv *priv = netdev_priv(ndev);
 
+	phylink_stop(priv->phylink);
+	phylink_disconnect_phy(priv->phylink);
+
 	napi_disable(&priv->napi);
 	netif_napi_del(&priv->napi);
 	tn40_stop(priv);
@@ -1402,13 +1406,20 @@ static int tn40_open(struct net_device *dev)
 	struct tn40_priv *priv = netdev_priv(dev);
 	int ret;
 
+	ret = phylink_connect_phy(priv->phylink, priv->phydev);
+	if (ret) {
+		netdev_err(dev, "failed to connect to phy %d\n", ret);
+		return ret;
+	}
 	tn40_sw_reset(priv);
 	ret = tn40_start(priv);
 	if (ret) {
+		phylink_disconnect_phy(priv->phylink);
 		netdev_err(dev, "failed to start %d\n", ret);
 		return ret;
 	}
 	napi_enable(&priv->napi);
+	phylink_start(priv->phylink);
 	netif_start_queue(priv->ndev);
 	return 0;
 }
@@ -1685,6 +1696,12 @@ static int tn40_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_unset_drvdata;
 	}
 
+	ret = tn40_mdiobus_init(priv);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to initialize mdio bus.\n");
+		goto err_free_irq;
+	}
+
 	priv->stats_flag =
 		((tn40_read_reg(priv, TN40_FPGA_VER) & 0xFFF) != 308);
 	u64_stats_init(&priv->syncp);
@@ -1694,19 +1711,26 @@ static int tn40_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		TN40_IR_TMR1;
 
 	tn40_mac_init(priv);
+	ret = tn40_phy_register(priv);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to set up PHY.\n");
+		goto err_free_irq;
+	}
 
 	ret = tn40_priv_init(priv);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to initialize tn40_priv.\n");
-		goto err_free_irq;
+		goto err_unregister_phydev;
 	}
 
 	ret = register_netdev(ndev);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to register netdev.\n");
-		goto err_free_irq;
+		goto err_unregister_phydev;
 	}
 	return 0;
+err_unregister_phydev:
+	tn40_phy_unregister(priv);
 err_free_irq:
 	pci_free_irq_vectors(pdev);
 err_unset_drvdata:
@@ -1727,6 +1751,7 @@ static void tn40_remove(struct pci_dev *pdev)
 
 	unregister_netdev(ndev);
 
+	tn40_phy_unregister(priv);
 	pci_free_irq_vectors(priv->pdev);
 	pci_set_drvdata(pdev, NULL);
 	iounmap(priv->regs);
diff --git a/drivers/net/ethernet/tehuti/tn40.h b/drivers/net/ethernet/tehuti/tn40.h
index 4cbe0c3883c7..10368264f7b7 100644
--- a/drivers/net/ethernet/tehuti/tn40.h
+++ b/drivers/net/ethernet/tehuti/tn40.h
@@ -142,6 +142,9 @@ struct tn40_priv {
 	char *b0_va; /* Virtual address of buffer */
 
 	struct mii_bus *mdio;
+	struct phy_device *phydev;
+	struct phylink *phylink;
+	struct phylink_config phylink_config;
 };
 
 /* RX FREE descriptor - 64bit */
@@ -219,6 +222,11 @@ static inline void tn40_write_reg(struct tn40_priv *priv, u32 reg, u32 val)
 	writel(val, priv->regs + reg);
 }
 
+int tn40_set_link_speed(struct tn40_priv *priv, u32 speed);
+
 int tn40_mdiobus_init(struct tn40_priv *priv);
 
+int tn40_phy_register(struct tn40_priv *priv);
+void tn40_phy_unregister(struct tn40_priv *priv);
+
 #endif /* _TN40XX_H */
diff --git a/drivers/net/ethernet/tehuti/tn40_phy.c b/drivers/net/ethernet/tehuti/tn40_phy.c
new file mode 100644
index 000000000000..39eef7ca7958
--- /dev/null
+++ b/drivers/net/ethernet/tehuti/tn40_phy.c
@@ -0,0 +1,76 @@
+// SPDX-License-Identifier: GPL-2.0+
+/* Copyright (c) Tehuti Networks Ltd. */
+
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+#include <linux/phylink.h>
+
+#include "tn40.h"
+
+static struct tn40_priv *tn40_config_to_priv(struct phylink_config *config)
+{
+	return container_of(config, struct tn40_priv, phylink_config);
+}
+
+static void tn40_link_up(struct phylink_config *config, struct phy_device *phy,
+			 unsigned int mode, phy_interface_t interface,
+			 int speed, int duplex, bool tx_pause, bool rx_pause)
+{
+	struct tn40_priv *priv = tn40_config_to_priv(config);
+
+	tn40_set_link_speed(priv, speed);
+	netif_wake_queue(priv->ndev);
+}
+
+static void tn40_link_down(struct phylink_config *config, unsigned int mode,
+			   phy_interface_t interface)
+{
+	struct tn40_priv *priv = tn40_config_to_priv(config);
+
+	netif_stop_queue(priv->ndev);
+	tn40_set_link_speed(priv, 0);
+}
+
+static void tn40_mac_config(struct phylink_config *config, unsigned int mode,
+			    const struct phylink_link_state *state)
+{
+}
+
+static const struct phylink_mac_ops tn40_mac_ops = {
+	.mac_config = tn40_mac_config,
+	.mac_link_up = tn40_link_up,
+	.mac_link_down = tn40_link_down,
+};
+
+int tn40_phy_register(struct tn40_priv *priv)
+{
+	struct phylink_config *config;
+	struct phy_device *phydev;
+	struct phylink *phylink;
+
+	phydev = phy_find_first(priv->mdio);
+	if (!phydev) {
+		dev_err(&priv->pdev->dev, "PHY isn't found\n");
+		return -ENODEV;
+	}
+
+	config = &priv->phylink_config;
+	config->dev = &priv->ndev->dev;
+	config->type = PHYLINK_NETDEV;
+	config->mac_capabilities = MAC_10000FD;
+	__set_bit(PHY_INTERFACE_MODE_XAUI, config->supported_interfaces);
+
+	phylink = phylink_create(config, NULL, PHY_INTERFACE_MODE_XAUI,
+				 &tn40_mac_ops);
+	if (IS_ERR(phylink))
+		return PTR_ERR(phylink);
+
+	priv->phydev = phydev;
+	priv->phylink = phylink;
+	return 0;
+}
+
+void tn40_phy_unregister(struct tn40_priv *priv)
+{
+	phylink_destroy(priv->phylink);
+}
-- 
cgit v1.2.3-58-ga151


From dc494fdc1f0657b4786d4f556b0462fe5198506b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 12 Jun 2024 14:28:35 +0200
Subject: wifi: mac80211: refactor CSA queue block/unblock

This code is duplicated many times, refactor it into
new separate functions.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240612143037.1ad22f10392d.If21490c2c67aae28f3c54038363181ee920ce3d1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c         | 20 ++++----------------
 net/mac80211/ieee80211_i.h |  3 +++
 net/mac80211/iface.c       | 29 ++++++++++++++++++++++++-----
 net/mac80211/mlme.c        | 33 ++++++---------------------------
 4 files changed, 37 insertions(+), 48 deletions(-)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 3236477424b8..7eb2e5bedb6f 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -1615,11 +1615,7 @@ static int ieee80211_stop_ap(struct wiphy *wiphy, struct net_device *dev,
 	/* abort any running channel switch or color change */
 	link_conf->csa_active = false;
 	link_conf->color_change_active = false;
-	if (sdata->csa_blocked_queues) {
-		ieee80211_wake_vif_queues(local, sdata,
-					  IEEE80211_QUEUE_STOP_REASON_CSA);
-		sdata->csa_blocked_queues = false;
-	}
+	ieee80211_vif_unblock_queues_csa(sdata);
 
 	ieee80211_free_next_beacon(link);
 
@@ -3757,11 +3753,7 @@ static int __ieee80211_csa_finalize(struct ieee80211_link_data *link_data)
 
 	ieee80211_link_info_change_notify(sdata, link_data, changed);
 
-	if (sdata->csa_blocked_queues) {
-		ieee80211_wake_vif_queues(local, sdata,
-					  IEEE80211_QUEUE_STOP_REASON_CSA);
-		sdata->csa_blocked_queues = false;
-	}
+	ieee80211_vif_unblock_queues_csa(sdata);
 
 	err = drv_post_channel_switch(link_data);
 	if (err)
@@ -4038,12 +4030,8 @@ __ieee80211_channel_switch(struct wiphy *wiphy, struct net_device *dev,
 	link_data->csa.chanreq = chanreq;
 	link_conf->csa_active = true;
 
-	if (params->block_tx &&
-	    !ieee80211_hw_check(&local->hw, HANDLES_QUIET_CSA)) {
-		ieee80211_stop_vif_queues(local, sdata,
-					  IEEE80211_QUEUE_STOP_REASON_CSA);
-		sdata->csa_blocked_queues = true;
-	}
+	if (params->block_tx)
+		ieee80211_vif_block_queues_csa(sdata);
 
 	cfg80211_ch_switch_started_notify(sdata->dev,
 					  &link_data->csa.chanreq.oper, link_id,
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 3e735c9436d3..b482763e9f99 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -1823,6 +1823,9 @@ ieee80211_have_rx_timestamp(struct ieee80211_rx_status *status)
 void ieee80211_vif_inc_num_mcast(struct ieee80211_sub_if_data *sdata);
 void ieee80211_vif_dec_num_mcast(struct ieee80211_sub_if_data *sdata);
 
+void ieee80211_vif_block_queues_csa(struct ieee80211_sub_if_data *sdata);
+void ieee80211_vif_unblock_queues_csa(struct ieee80211_sub_if_data *sdata);
+
 /* This function returns the number of multicast stations connected to this
  * interface. It returns -1 if that number is not tracked, that is for netdevs
  * not in AP or AP_VLAN mode or when using 4addr.
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 6b7580c61e0c..f06e165d6c7a 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -544,11 +544,7 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, bool going_do
 	sdata->vif.bss_conf.csa_active = false;
 	if (sdata->vif.type == NL80211_IFTYPE_STATION)
 		sdata->deflink.u.mgd.csa.waiting_bcn = false;
-	if (sdata->csa_blocked_queues) {
-		ieee80211_wake_vif_queues(local, sdata,
-					  IEEE80211_QUEUE_STOP_REASON_CSA);
-		sdata->csa_blocked_queues = false;
-	}
+	ieee80211_vif_unblock_queues_csa(sdata);
 
 	wiphy_work_cancel(local->hw.wiphy, &sdata->deflink.csa.finalize_work);
 	wiphy_work_cancel(local->hw.wiphy,
@@ -2345,3 +2341,26 @@ void ieee80211_vif_dec_num_mcast(struct ieee80211_sub_if_data *sdata)
 	else if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN)
 		atomic_dec(&sdata->u.vlan.num_mcast_sta);
 }
+
+void ieee80211_vif_block_queues_csa(struct ieee80211_sub_if_data *sdata)
+{
+	struct ieee80211_local *local = sdata->local;
+
+	if (ieee80211_hw_check(&local->hw, HANDLES_QUIET_CSA))
+		return;
+
+	ieee80211_stop_vif_queues(local, sdata,
+				  IEEE80211_QUEUE_STOP_REASON_CSA);
+	sdata->csa_blocked_queues = true;
+}
+
+void ieee80211_vif_unblock_queues_csa(struct ieee80211_sub_if_data *sdata)
+{
+	struct ieee80211_local *local = sdata->local;
+
+	if (sdata->csa_blocked_queues) {
+		ieee80211_wake_vif_queues(local, sdata,
+					  IEEE80211_QUEUE_STOP_REASON_CSA);
+		sdata->csa_blocked_queues = false;
+	}
+}
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index ef3280fafbe9..ac376ec47a59 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2166,7 +2166,6 @@ static void ieee80211_csa_switch_work(struct wiphy *wiphy,
 static void ieee80211_chswitch_post_beacon(struct ieee80211_link_data *link)
 {
 	struct ieee80211_sub_if_data *sdata = link->sdata;
-	struct ieee80211_local *local = sdata->local;
 	struct ieee80211_if_managed *ifmgd = &sdata->u.mgd;
 	int ret;
 
@@ -2174,11 +2173,7 @@ static void ieee80211_chswitch_post_beacon(struct ieee80211_link_data *link)
 
 	WARN_ON(!link->conf->csa_active);
 
-	if (sdata->csa_blocked_queues) {
-		ieee80211_wake_vif_queues(local, sdata,
-					  IEEE80211_QUEUE_STOP_REASON_CSA);
-		sdata->csa_blocked_queues = false;
-	}
+	ieee80211_vif_unblock_queues_csa(sdata);
 
 	link->conf->csa_active = false;
 	link->u.mgd.csa.blocked_tx = false;
@@ -2242,11 +2237,7 @@ ieee80211_sta_abort_chanswitch(struct ieee80211_link_data *link)
 
 	ieee80211_link_unreserve_chanctx(link);
 
-	if (sdata->csa_blocked_queues) {
-		ieee80211_wake_vif_queues(local, sdata,
-					  IEEE80211_QUEUE_STOP_REASON_CSA);
-		sdata->csa_blocked_queues = false;
-	}
+	ieee80211_vif_unblock_queues_csa(sdata);
 
 	link->conf->csa_active = false;
 	link->u.mgd.csa.blocked_tx = false;
@@ -2571,12 +2562,8 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 	link->u.mgd.beacon_crc_valid = false;
 	link->u.mgd.csa.blocked_tx = csa_ie.mode;
 
-	if (csa_ie.mode &&
-	    !ieee80211_hw_check(&local->hw, HANDLES_QUIET_CSA)) {
-		ieee80211_stop_vif_queues(local, sdata,
-					  IEEE80211_QUEUE_STOP_REASON_CSA);
-		sdata->csa_blocked_queues = true;
-	}
+	if (csa_ie.mode)
+		ieee80211_vif_block_queues_csa(sdata);
 
 	cfg80211_ch_switch_started_notify(sdata->dev, &csa_ie.chanreq.oper,
 					  link->link_id, csa_ie.count,
@@ -3670,11 +3657,7 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata,
 	sdata->deflink.u.mgd.csa.blocked_tx = false;
 	sdata->deflink.u.mgd.csa.waiting_bcn = false;
 	sdata->deflink.u.mgd.csa.ignored_same_chan = false;
-	if (sdata->csa_blocked_queues) {
-		ieee80211_wake_vif_queues(local, sdata,
-					  IEEE80211_QUEUE_STOP_REASON_CSA);
-		sdata->csa_blocked_queues = false;
-	}
+	ieee80211_vif_unblock_queues_csa(sdata);
 
 	/* existing TX TSPEC sessions no longer exist */
 	memset(ifmgd->tx_tspec, 0, sizeof(ifmgd->tx_tspec));
@@ -4045,11 +4028,7 @@ static void __ieee80211_disconnect(struct ieee80211_sub_if_data *sdata)
 	sdata->vif.bss_conf.csa_active = false;
 	sdata->deflink.u.mgd.csa.waiting_bcn = false;
 	sdata->deflink.u.mgd.csa.blocked_tx = false;
-	if (sdata->csa_blocked_queues) {
-		ieee80211_wake_vif_queues(local, sdata,
-					  IEEE80211_QUEUE_STOP_REASON_CSA);
-		sdata->csa_blocked_queues = false;
-	}
+	ieee80211_vif_unblock_queues_csa(sdata);
 
 	ieee80211_report_disconnect(sdata, frame_buf, sizeof(frame_buf), tx,
 				    WLAN_REASON_DISASSOC_DUE_TO_INACTIVITY,
-- 
cgit v1.2.3-58-ga151


From 414e090bc41d27af8196d2f36c2e883d7b4f1478 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 12 Jun 2024 14:28:36 +0200
Subject: wifi: mac80211: restrict public action ECSA frame handling

Public action extended channel switch announcement (ECSA)
frames cannot be protected well, the spec is unclear about
what should happen in the presence of stations that can
receive protected dual and stations that cannot.

Mitigate these issues by not treating public action frames
as the absolute truth, only treat them as a hint to stop
transmitting (quiet mode), and do the remainder of the CSA
handling only when receiving the next beacon (or protected
action frame) that contains the CSA; or, if it doesn't,
simply stop being quiet and continue operating normally.

This limits the exposure to malicious ECSA public action
frames, since they cannot cause a disconnect now, only a
short interruption in traffic.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240612143037.ec7ccc45903e.Ife17d55c7ecbf98060f9c52889f3c8ba48798970@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/ibss.c        |  3 ++-
 net/mac80211/ieee80211_i.h |  4 +++-
 net/mac80211/mesh.c        |  2 +-
 net/mac80211/mlme.c        | 59 +++++++++++++++++++++++++++++++++++++++-------
 net/mac80211/spectmgmt.c   | 15 +++++++-----
 5 files changed, 65 insertions(+), 18 deletions(-)

diff --git a/net/mac80211/ibss.c b/net/mac80211/ibss.c
index bf338f3d4dd3..7db4c3ee7e6d 100644
--- a/net/mac80211/ibss.c
+++ b/net/mac80211/ibss.c
@@ -785,7 +785,8 @@ ieee80211_ibss_process_chanswitch(struct ieee80211_sub_if_data *sdata,
 	err = ieee80211_parse_ch_switch_ie(sdata, elems,
 					   ifibss->chandef.chan->band,
 					   vht_cap_info, &conn,
-					   ifibss->bssid, &csa_ie);
+					   ifibss->bssid, false,
+					   &csa_ie);
 	/* can't switch to destination channel, fail */
 	if (err < 0)
 		goto disconnect;
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index b482763e9f99..cb3e28f88089 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -2218,6 +2218,8 @@ void ieee80211_process_measurement_req(struct ieee80211_sub_if_data *sdata,
  * @conn: contains information about own capabilities and restrictions
  *	to decide which channel switch announcements can be accepted
  * @bssid: the currently connected bssid (for reporting)
+ * @unprot_action: whether the frame was an unprotected frame or not,
+ *	used for reporting
  * @csa_ie: parsed 802.11 csa elements on count, mode, chandef and mesh ttl.
  *	All of them will be filled with if success only.
  * Return: 0 on success, <0 on error and >0 if there is nothing to parse.
@@ -2227,7 +2229,7 @@ int ieee80211_parse_ch_switch_ie(struct ieee80211_sub_if_data *sdata,
 				 enum nl80211_band current_band,
 				 u32 vht_cap_info,
 				 struct ieee80211_conn_settings *conn,
-				 u8 *bssid,
+				 u8 *bssid, bool unprot_action,
 				 struct ieee80211_csa_ie *csa_ie);
 
 /* Suspend/resume and hw reconfiguration */
diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c
index 6d4510221c98..f94e4be0be12 100644
--- a/net/mac80211/mesh.c
+++ b/net/mac80211/mesh.c
@@ -1312,7 +1312,7 @@ ieee80211_mesh_process_chnswitch(struct ieee80211_sub_if_data *sdata,
 	memset(&params, 0, sizeof(params));
 	err = ieee80211_parse_ch_switch_ie(sdata, elems, sband->band,
 					   vht_cap_info, &conn,
-					   sdata->vif.addr,
+					   sdata->vif.addr, false,
 					   &csa_ie);
 	if (err < 0)
 		return false;
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index ac376ec47a59..fff158b1615f 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -2352,7 +2352,8 @@ ieee80211_sta_other_link_csa_disappeared(struct ieee80211_link_data *link,
 enum ieee80211_csa_source {
 	IEEE80211_CSA_SOURCE_BEACON,
 	IEEE80211_CSA_SOURCE_OTHER_LINK,
-	IEEE80211_CSA_SOURCE_ACTION,
+	IEEE80211_CSA_SOURCE_PROT_ACTION,
+	IEEE80211_CSA_SOURCE_UNPROT_ACTION,
 };
 
 static void
@@ -2393,7 +2394,9 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 						   current_band,
 						   bss->vht_cap_info,
 						   &link->u.mgd.conn,
-						   link->u.mgd.bssid, &csa_ie);
+						   link->u.mgd.bssid,
+						   source == IEEE80211_CSA_SOURCE_UNPROT_ACTION,
+						   &csa_ie);
 		if (res == 0) {
 			ch_switch.block_tx = csa_ie.mode;
 			ch_switch.chandef = csa_ie.chanreq.oper;
@@ -2412,12 +2415,17 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 		res = 1;
 	}
 
-	if (res < 0)
+	if (res < 0) {
+		/* ignore this case, not a protected frame */
+		if (source == IEEE80211_CSA_SOURCE_UNPROT_ACTION)
+			return;
 		goto drop_connection;
+	}
 
 	if (link->conf->csa_active) {
 		switch (source) {
-		case IEEE80211_CSA_SOURCE_ACTION:
+		case IEEE80211_CSA_SOURCE_PROT_ACTION:
+		case IEEE80211_CSA_SOURCE_UNPROT_ACTION:
 			/* already processing - disregard action frames */
 			return;
 		case IEEE80211_CSA_SOURCE_BEACON:
@@ -2466,9 +2474,35 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link,
 		}
 	}
 
-	/* nothing to do at all - no active CSA nor a new one */
-	if (res)
+	/* no active CSA nor a new one */
+	if (res) {
+		/*
+		 * However, we may have stopped queues when receiving a public
+		 * action frame that couldn't be protected, if it had the quiet
+		 * bit set. This is a trade-off, we want to be quiet as soon as
+		 * possible, but also don't trust the public action frame much,
+		 * as it can't be protected.
+		 */
+		if (unlikely(link->u.mgd.csa.blocked_tx)) {
+			link->u.mgd.csa.blocked_tx = false;
+			ieee80211_vif_unblock_queues_csa(sdata);
+		}
+		return;
+	}
+
+	/*
+	 * We don't really trust public action frames, but block queues (go to
+	 * quiet mode) for them anyway, we should get a beacon soon to either
+	 * know what the CSA really is, or figure out the public action frame
+	 * was actually an attack.
+	 */
+	if (source == IEEE80211_CSA_SOURCE_UNPROT_ACTION) {
+		if (csa_ie.mode) {
+			link->u.mgd.csa.blocked_tx = true;
+			ieee80211_vif_block_queues_csa(sdata);
+		}
 		return;
+	}
 
 	if (link->conf->chanreq.oper.chan->band !=
 	    csa_ie.chanreq.oper.chan->band) {
@@ -7453,12 +7487,16 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata,
 					mgmt->u.action.u.chan_switch.variable,
 					ies_len, true, NULL);
 
-			if (elems && !elems->parse_error)
+			if (elems && !elems->parse_error) {
+				enum ieee80211_csa_source src =
+					IEEE80211_CSA_SOURCE_PROT_ACTION;
+
 				ieee80211_sta_process_chanswitch(link,
 								 rx_status->mactime,
 								 rx_status->device_timestamp,
 								 elems, elems,
-								 IEEE80211_CSA_SOURCE_ACTION);
+								 src);
+			}
 			kfree(elems);
 		} else if (mgmt->u.action.category == WLAN_CATEGORY_PUBLIC) {
 			struct ieee802_11_elems *elems;
@@ -7479,6 +7517,9 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata,
 					ies_len, true, NULL);
 
 			if (elems && !elems->parse_error) {
+				enum ieee80211_csa_source src =
+					IEEE80211_CSA_SOURCE_UNPROT_ACTION;
+
 				/* for the handling code pretend it was an IE */
 				elems->ext_chansw_ie =
 					&mgmt->u.action.u.ext_chan_switch.data;
@@ -7487,7 +7528,7 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata,
 								 rx_status->mactime,
 								 rx_status->device_timestamp,
 								 elems, elems,
-								 IEEE80211_CSA_SOURCE_ACTION);
+								 src);
 			}
 
 			kfree(elems);
diff --git a/net/mac80211/spectmgmt.c b/net/mac80211/spectmgmt.c
index df96d3db1c0e..e91ca4ccdd37 100644
--- a/net/mac80211/spectmgmt.c
+++ b/net/mac80211/spectmgmt.c
@@ -223,7 +223,7 @@ int ieee80211_parse_ch_switch_ie(struct ieee80211_sub_if_data *sdata,
 				 enum nl80211_band current_band,
 				 u32 vht_cap_info,
 				 struct ieee80211_conn_settings *conn,
-				 u8 *bssid,
+				 u8 *bssid, bool unprot_action,
 				 struct ieee80211_csa_ie *csa_ie)
 {
 	enum nl80211_band new_band = current_band;
@@ -258,8 +258,10 @@ int ieee80211_parse_ch_switch_ie(struct ieee80211_sub_if_data *sdata,
 
 		if (!ieee80211_operating_class_to_band(new_op_class, &new_band)) {
 			new_op_class = 0;
-			sdata_info(sdata, "cannot understand ECSA IE operating class, %d, ignoring\n",
-				   ext_chansw_elem->new_operating_class);
+			if (!unprot_action)
+				sdata_info(sdata,
+					   "cannot understand ECSA IE operating class, %d, ignoring\n",
+					   ext_chansw_elem->new_operating_class);
 		} else {
 			new_chan_no = ext_chansw_elem->new_ch_num;
 			csa_ie->count = ext_chansw_elem->count;
@@ -293,9 +295,10 @@ int ieee80211_parse_ch_switch_ie(struct ieee80211_sub_if_data *sdata,
 	new_freq = ieee80211_channel_to_frequency(new_chan_no, new_band);
 	new_chan = ieee80211_get_channel(sdata->local->hw.wiphy, new_freq);
 	if (!new_chan || new_chan->flags & IEEE80211_CHAN_DISABLED) {
-		sdata_info(sdata,
-			   "BSS %pM switches to unsupported channel (%d MHz), disconnecting\n",
-			   bssid, new_freq);
+		if (!unprot_action)
+			sdata_info(sdata,
+				   "BSS %pM switches to unsupported channel (%d MHz), disconnecting\n",
+				   bssid, new_freq);
 		return -EINVAL;
 	}
 
-- 
cgit v1.2.3-58-ga151


From b777bdfc9be8a61ecf3c07962add3b5fc1279e98 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 12 Jun 2024 14:28:37 +0200
Subject: wifi: mac80211: handle protected dual of public action

The code currently handles ECSA (extended channel switch
announcement) public action frames. Handle also their
protected dual, which actually is protected.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240612143037.db642feb8b2e.I184fa5c9bffb68099171701e403c2aa733f60fde@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 22 ++++++++++++++--------
 net/mac80211/rx.c   |  1 +
 2 files changed, 15 insertions(+), 8 deletions(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index fff158b1615f..e76b887a46eb 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -7429,6 +7429,7 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata,
 {
 	struct ieee80211_link_data *link = &sdata->deflink;
 	struct ieee80211_rx_status *rx_status;
+	struct ieee802_11_elems *elems;
 	struct ieee80211_mgmt *mgmt;
 	u16 fc;
 	int ies_len;
@@ -7472,9 +7473,8 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata,
 		    !ether_addr_equal(mgmt->bssid, sdata->vif.cfg.ap_addr))
 			break;
 
-		if (mgmt->u.action.category == WLAN_CATEGORY_SPECTRUM_MGMT) {
-			struct ieee802_11_elems *elems;
-
+		switch (mgmt->u.action.category) {
+		case WLAN_CATEGORY_SPECTRUM_MGMT:
 			ies_len = skb->len -
 				  offsetof(struct ieee80211_mgmt,
 					   u.action.u.chan_switch.variable);
@@ -7498,9 +7498,9 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata,
 								 src);
 			}
 			kfree(elems);
-		} else if (mgmt->u.action.category == WLAN_CATEGORY_PUBLIC) {
-			struct ieee802_11_elems *elems;
-
+			break;
+		case WLAN_CATEGORY_PUBLIC:
+		case WLAN_CATEGORY_PROTECTED_DUAL_OF_ACTION:
 			ies_len = skb->len -
 				  offsetof(struct ieee80211_mgmt,
 					   u.action.u.ext_chan_switch.variable);
@@ -7517,8 +7517,13 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata,
 					ies_len, true, NULL);
 
 			if (elems && !elems->parse_error) {
-				enum ieee80211_csa_source src =
-					IEEE80211_CSA_SOURCE_UNPROT_ACTION;
+				enum ieee80211_csa_source src;
+
+				if (mgmt->u.action.category ==
+						WLAN_CATEGORY_PROTECTED_DUAL_OF_ACTION)
+					src = IEEE80211_CSA_SOURCE_PROT_ACTION;
+				else
+					src = IEEE80211_CSA_SOURCE_UNPROT_ACTION;
 
 				/* for the handling code pretend it was an IE */
 				elems->ext_chansw_ie =
@@ -7532,6 +7537,7 @@ void ieee80211_sta_rx_queued_mgmt(struct ieee80211_sub_if_data *sdata,
 			}
 
 			kfree(elems);
+			break;
 		}
 		break;
 	}
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 4914692750e5..0ff9062a130c 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -3617,6 +3617,7 @@ ieee80211_rx_h_action(struct ieee80211_rx_data *rx)
 
 		break;
 	case WLAN_CATEGORY_PUBLIC:
+	case WLAN_CATEGORY_PROTECTED_DUAL_OF_ACTION:
 		if (len < IEEE80211_MIN_ACTION_SIZE + 1)
 			goto invalid;
 		if (sdata->vif.type != NL80211_IFTYPE_STATION)
-- 
cgit v1.2.3-58-ga151


From 25af8ff51d7e027fbdf8ed84e349427311a6da4d Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 12 Jun 2024 14:32:03 +0200
Subject: wifi: mac80211: optionally pass chandef to ieee80211_sta_cap_rx_bw()

We'll need this function to take a new chandef in
(some) channel switching cases, so prepare for that
by allowing that to be passed and using it if so.
Clean up the code a little bit while at it.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240612143418.772313f08b6a.If9708249e5870671e745d4c2b02e03b25092bea3@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/ieee80211_i.h |  8 +++++++-
 net/mac80211/vht.c         | 48 +++++++++++++++++++++++-----------------------
 2 files changed, 31 insertions(+), 25 deletions(-)

diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index cb3e28f88089..e96404f9dc70 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -2158,7 +2158,13 @@ ieee80211_vht_cap_ie_to_sta_vht_cap(struct ieee80211_sub_if_data *sdata,
 				    const struct ieee80211_vht_cap *vht_cap_ie2,
 				    struct link_sta_info *link_sta);
 enum ieee80211_sta_rx_bandwidth
-ieee80211_sta_cap_rx_bw(struct link_sta_info *link_sta);
+_ieee80211_sta_cap_rx_bw(struct link_sta_info *link_sta,
+			 struct cfg80211_chan_def *chandef);
+static inline enum ieee80211_sta_rx_bandwidth
+ieee80211_sta_cap_rx_bw(struct link_sta_info *link_sta)
+{
+	return _ieee80211_sta_cap_rx_bw(link_sta, NULL);
+}
 enum ieee80211_sta_rx_bandwidth
 ieee80211_sta_cur_vht_bw(struct link_sta_info *link_sta);
 void ieee80211_sta_init_nss(struct link_sta_info *link_sta);
diff --git a/net/mac80211/vht.c b/net/mac80211/vht.c
index 642891cafbaf..c280945fc9d6 100644
--- a/net/mac80211/vht.c
+++ b/net/mac80211/vht.c
@@ -351,7 +351,8 @@ ieee80211_vht_cap_ie_to_sta_vht_cap(struct ieee80211_sub_if_data *sdata,
 
 /* FIXME: move this to some better location - parses HE/EHT now */
 enum ieee80211_sta_rx_bandwidth
-ieee80211_sta_cap_rx_bw(struct link_sta_info *link_sta)
+_ieee80211_sta_cap_rx_bw(struct link_sta_info *link_sta,
+			 struct cfg80211_chan_def *chandef)
 {
 	unsigned int link_id = link_sta->link_id;
 	struct ieee80211_sub_if_data *sdata = link_sta->sta->sdata;
@@ -361,44 +362,43 @@ ieee80211_sta_cap_rx_bw(struct link_sta_info *link_sta)
 	u32 cap_width;
 
 	if (he_cap->has_he) {
-		struct ieee80211_bss_conf *link_conf;
-		enum ieee80211_sta_rx_bandwidth ret;
+		enum nl80211_band band;
 		u8 info;
 
-		rcu_read_lock();
-		link_conf = rcu_dereference(sdata->vif.link_conf[link_id]);
+		if (chandef) {
+			band = chandef->chan->band;
+		} else {
+			struct ieee80211_bss_conf *link_conf;
+
+			rcu_read_lock();
+			link_conf = rcu_dereference(sdata->vif.link_conf[link_id]);
+			band = link_conf->chanreq.oper.chan->band;
+			rcu_read_unlock();
+		}
 
-		if (eht_cap->has_eht &&
-		    link_conf->chanreq.oper.chan->band == NL80211_BAND_6GHZ) {
+		if (eht_cap->has_eht && band == NL80211_BAND_6GHZ) {
 			info = eht_cap->eht_cap_elem.phy_cap_info[0];
 
-			if (info & IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ) {
-				ret = IEEE80211_STA_RX_BW_320;
-				goto out;
-			}
+			if (info & IEEE80211_EHT_PHY_CAP0_320MHZ_IN_6GHZ)
+				return IEEE80211_STA_RX_BW_320;
 		}
 
 		info = he_cap->he_cap_elem.phy_cap_info[0];
 
-		if (link_conf->chanreq.oper.chan->band == NL80211_BAND_2GHZ) {
+		if (band == NL80211_BAND_2GHZ) {
 			if (info & IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_IN_2G)
-				ret = IEEE80211_STA_RX_BW_40;
-			else
-				ret = IEEE80211_STA_RX_BW_20;
-			goto out;
+				return IEEE80211_STA_RX_BW_40;
+			return IEEE80211_STA_RX_BW_20;
 		}
 
 		if (info & IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_160MHZ_IN_5G ||
 		    info & IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_80PLUS80_MHZ_IN_5G)
-			ret = IEEE80211_STA_RX_BW_160;
-		else if (info & IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G)
-			ret = IEEE80211_STA_RX_BW_80;
-		else
-			ret = IEEE80211_STA_RX_BW_20;
-out:
-		rcu_read_unlock();
+			return IEEE80211_STA_RX_BW_160;
+
+		if (info & IEEE80211_HE_PHY_CAP0_CHANNEL_WIDTH_SET_40MHZ_80MHZ_IN_5G)
+			return IEEE80211_STA_RX_BW_80;
 
-		return ret;
+		return IEEE80211_STA_RX_BW_20;
 	}
 
 	if (!vht_cap->vht_supported)
-- 
cgit v1.2.3-58-ga151


From 7d2bad829c449b8cb172d20cb51d62c586fca94d Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 12 Jun 2024 14:32:04 +0200
Subject: wifi: mac80211: optionally pass chandef to ieee80211_sta_cur_vht_bw()

We'll need this as well for channel switching cases, so
add the ability now to pass the chandef to calculate for.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240612143418.f70e05d9f306.Ifa0ce267de4f0ef3c21d063fb0cbf50e84d7d6ff@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/ieee80211_i.h |  8 +++++++-
 net/mac80211/vht.c         | 25 ++++++++++++++++---------
 2 files changed, 23 insertions(+), 10 deletions(-)

diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index e96404f9dc70..33b8efff92c1 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -2166,7 +2166,13 @@ ieee80211_sta_cap_rx_bw(struct link_sta_info *link_sta)
 	return _ieee80211_sta_cap_rx_bw(link_sta, NULL);
 }
 enum ieee80211_sta_rx_bandwidth
-ieee80211_sta_cur_vht_bw(struct link_sta_info *link_sta);
+_ieee80211_sta_cur_vht_bw(struct link_sta_info *link_sta,
+			  struct cfg80211_chan_def *chandef);
+static inline enum ieee80211_sta_rx_bandwidth
+ieee80211_sta_cur_vht_bw(struct link_sta_info *link_sta)
+{
+	return _ieee80211_sta_cur_vht_bw(link_sta, NULL);
+}
 void ieee80211_sta_init_nss(struct link_sta_info *link_sta);
 enum ieee80211_sta_rx_bandwidth
 ieee80211_chan_width_to_rx_bw(enum nl80211_chan_width width);
diff --git a/net/mac80211/vht.c b/net/mac80211/vht.c
index c280945fc9d6..bf6ef45af757 100644
--- a/net/mac80211/vht.c
+++ b/net/mac80211/vht.c
@@ -503,22 +503,29 @@ ieee80211_chan_width_to_rx_bw(enum nl80211_chan_width width)
 
 /* FIXME: rename/move - this deals with everything not just VHT */
 enum ieee80211_sta_rx_bandwidth
-ieee80211_sta_cur_vht_bw(struct link_sta_info *link_sta)
+_ieee80211_sta_cur_vht_bw(struct link_sta_info *link_sta,
+			  struct cfg80211_chan_def *chandef)
 {
 	struct sta_info *sta = link_sta->sta;
-	struct ieee80211_bss_conf *link_conf;
 	enum nl80211_chan_width bss_width;
 	enum ieee80211_sta_rx_bandwidth bw;
 
-	rcu_read_lock();
-	link_conf = rcu_dereference(sta->sdata->vif.link_conf[link_sta->link_id]);
-	if (WARN_ON(!link_conf))
-		bss_width = NL80211_CHAN_WIDTH_20_NOHT;
-	else
+	if (chandef) {
+		bss_width = chandef->width;
+	} else {
+		struct ieee80211_bss_conf *link_conf;
+
+		rcu_read_lock();
+		link_conf = rcu_dereference(sta->sdata->vif.link_conf[link_sta->link_id]);
+		if (WARN_ON_ONCE(!link_conf)) {
+			rcu_read_unlock();
+			return IEEE80211_STA_RX_BW_20;
+		}
 		bss_width = link_conf->chanreq.oper.width;
-	rcu_read_unlock();
+		rcu_read_unlock();
+	}
 
-	bw = ieee80211_sta_cap_rx_bw(link_sta);
+	bw = _ieee80211_sta_cap_rx_bw(link_sta, chandef);
 	bw = min(bw, link_sta->cur_max_bandwidth);
 
 	/* Don't consider AP's bandwidth for TDLS peers, section 11.23.1 of
-- 
cgit v1.2.3-58-ga151


From b27512368591fc959768df1f7dacf2a96b1bd036 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 12 Jun 2024 14:32:05 +0200
Subject: wifi: mac80211: make ieee80211_chan_bw_change() able to use reserved

Make ieee80211_chan_bw_change() able to use the reserved chanreq
(really the chandef part of it) for the calculations, so it can
be used _without_ applying the changes first. Remove the comment
that indicates this is required, since it no longer is. However,
this capability only gets used later.

Also, this is not ideal, we really should not different so much
between reserved and non-reserved usage, to simplify. That's a
further cleanup later though.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240612143418.1a08cf83b8cb.Ie567bb272eb25ce487651088f13ad041f549651c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/chan.c | 31 +++++++++++++++++++------------
 1 file changed, 19 insertions(+), 12 deletions(-)

diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c
index ec16d7676088..a42ab3179d99 100644
--- a/net/mac80211/chan.c
+++ b/net/mac80211/chan.c
@@ -396,12 +396,9 @@ _ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local,
 	return IEEE80211_CHANCTX_CHANGE_MIN_WIDTH;
 }
 
-/* calling this function is assuming that station vif is updated to
- * lates changes by calling ieee80211_link_update_chanreq
- */
 static void ieee80211_chan_bw_change(struct ieee80211_local *local,
 				     struct ieee80211_chanctx *ctx,
-				     bool narrowed)
+				     bool reserved, bool narrowed)
 {
 	struct sta_info *sta;
 	struct ieee80211_supported_band *sband =
@@ -418,13 +415,17 @@ static void ieee80211_chan_bw_change(struct ieee80211_local *local,
 			continue;
 
 		for (link_id = 0; link_id < ARRAY_SIZE(sta->sdata->link); link_id++) {
-			struct ieee80211_bss_conf *link_conf =
-				rcu_dereference(sdata->vif.link_conf[link_id]);
+			struct ieee80211_link_data *link =
+				rcu_dereference(sdata->link[link_id]);
+			struct ieee80211_bss_conf *link_conf;
+			struct cfg80211_chan_def *new_chandef;
 			struct link_sta_info *link_sta;
 
-			if (!link_conf)
+			if (!link)
 				continue;
 
+			link_conf = link->conf;
+
 			if (rcu_access_pointer(link_conf->chanctx_conf) != &ctx->conf)
 				continue;
 
@@ -432,7 +433,13 @@ static void ieee80211_chan_bw_change(struct ieee80211_local *local,
 			if (!link_sta)
 				continue;
 
-			new_sta_bw = ieee80211_sta_cur_vht_bw(link_sta);
+			if (reserved)
+				new_chandef = &link->reserved.oper;
+			else
+				new_chandef = &link_conf->chanreq.oper;
+
+			new_sta_bw = _ieee80211_sta_cur_vht_bw(link_sta,
+							       new_chandef);
 
 			/* nothing change */
 			if (new_sta_bw == link_sta->pub->bandwidth)
@@ -466,12 +473,12 @@ void ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local,
 		return;
 
 	/* check is BW narrowed */
-	ieee80211_chan_bw_change(local, ctx, true);
+	ieee80211_chan_bw_change(local, ctx, false, true);
 
 	drv_change_chanctx(local, ctx, changed);
 
 	/* check is BW wider */
-	ieee80211_chan_bw_change(local, ctx, false);
+	ieee80211_chan_bw_change(local, ctx, false, false);
 }
 
 static void _ieee80211_change_chanctx(struct ieee80211_local *local,
@@ -505,7 +512,7 @@ static void _ieee80211_change_chanctx(struct ieee80211_local *local,
 	 * due to maybe not returning from it, e.g in case new context was added
 	 * first time with all parameters up to date.
 	 */
-	ieee80211_chan_bw_change(local, old_ctx, true);
+	ieee80211_chan_bw_change(local, old_ctx, false, true);
 
 	if (ieee80211_chanreq_identical(&ctx_req, chanreq)) {
 		ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for);
@@ -536,7 +543,7 @@ static void _ieee80211_change_chanctx(struct ieee80211_local *local,
 	drv_change_chanctx(local, ctx, changed);
 
 	/* check if BW is wider */
-	ieee80211_chan_bw_change(local, old_ctx, false);
+	ieee80211_chan_bw_change(local, old_ctx, false, false);
 }
 
 static void ieee80211_change_chanctx(struct ieee80211_local *local,
-- 
cgit v1.2.3-58-ga151


From dd7b1bdb5614419305089de399b7e6aa2fde8301 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 12 Jun 2024 14:32:06 +0200
Subject: wifi: mac80211: update STA/chandef width during switch

In channel switch without an additional channel context,
where the reassign logic kicks in, we also need to update
the station bandwidth and chandef minimum width correctly
to avoid having station rate control configured to wider
bandwidth than the channel context. Do that now.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240612143418.0bc3d28231b3.I51e76df86212057ca0469e235ba9bf4461cbee75@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/chan.c        | 67 +++++++++++++++++++++++++++++++++++-----------
 net/mac80211/ieee80211_i.h |  3 ++-
 net/mac80211/util.c        |  2 +-
 3 files changed, 55 insertions(+), 17 deletions(-)

diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c
index a42ab3179d99..942c882f1a1d 100644
--- a/net/mac80211/chan.c
+++ b/net/mac80211/chan.c
@@ -295,17 +295,24 @@ ieee80211_get_max_required_bw(struct ieee80211_link_data *link)
 static enum nl80211_chan_width
 ieee80211_get_chanctx_max_required_bw(struct ieee80211_local *local,
 				      struct ieee80211_chanctx *ctx,
-				      struct ieee80211_link_data *rsvd_for)
+				      struct ieee80211_link_data *rsvd_for,
+				      bool check_reserved)
 {
 	struct ieee80211_sub_if_data *sdata;
 	struct ieee80211_link_data *link;
 	enum nl80211_chan_width max_bw = NL80211_CHAN_WIDTH_20_NOHT;
 
+	if (WARN_ON(check_reserved && rsvd_for))
+		return ctx->conf.def.width;
+
 	for_each_sdata_link(local, link) {
 		enum nl80211_chan_width width = NL80211_CHAN_WIDTH_20_NOHT;
 
-		if (link != rsvd_for &&
-		    rcu_access_pointer(link->conf->chanctx_conf) != &ctx->conf)
+		if (check_reserved) {
+			if (link->reserved_chanctx != ctx)
+				continue;
+		} else if (link != rsvd_for &&
+			   rcu_access_pointer(link->conf->chanctx_conf) != &ctx->conf)
 			continue;
 
 		switch (link->sdata->vif.type) {
@@ -359,7 +366,8 @@ ieee80211_get_chanctx_max_required_bw(struct ieee80211_local *local,
 static u32
 _ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local,
 				  struct ieee80211_chanctx *ctx,
-				  struct ieee80211_link_data *rsvd_for)
+				  struct ieee80211_link_data *rsvd_for,
+				  bool check_reserved)
 {
 	enum nl80211_chan_width max_bw;
 	struct cfg80211_chan_def min_def;
@@ -379,7 +387,8 @@ _ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local,
 		return 0;
 	}
 
-	max_bw = ieee80211_get_chanctx_max_required_bw(local, ctx, rsvd_for);
+	max_bw = ieee80211_get_chanctx_max_required_bw(local, ctx, rsvd_for,
+						       check_reserved);
 
 	/* downgrade chandef up to max_bw */
 	min_def = ctx->conf.def;
@@ -465,9 +474,11 @@ static void ieee80211_chan_bw_change(struct ieee80211_local *local,
  */
 void ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local,
 				      struct ieee80211_chanctx *ctx,
-				      struct ieee80211_link_data *rsvd_for)
+				      struct ieee80211_link_data *rsvd_for,
+				      bool check_reserved)
 {
-	u32 changed = _ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for);
+	u32 changed = _ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for,
+							check_reserved);
 
 	if (!changed)
 		return;
@@ -515,7 +526,7 @@ static void _ieee80211_change_chanctx(struct ieee80211_local *local,
 	ieee80211_chan_bw_change(local, old_ctx, false, true);
 
 	if (ieee80211_chanreq_identical(&ctx_req, chanreq)) {
-		ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for);
+		ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for, false);
 		return;
 	}
 
@@ -536,7 +547,7 @@ static void _ieee80211_change_chanctx(struct ieee80211_local *local,
 	ctx->conf.ap = chanreq->ap;
 
 	/* check if min chanctx also changed */
-	changed |= _ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for);
+	changed |= _ieee80211_recalc_chanctx_min_def(local, ctx, rsvd_for, false);
 
 	ieee80211_add_wbrf(local, &ctx->conf.def);
 
@@ -663,7 +674,7 @@ ieee80211_alloc_chanctx(struct ieee80211_local *local,
 	ctx->conf.rx_chains_dynamic = 1;
 	ctx->mode = mode;
 	ctx->conf.radar_enabled = false;
-	_ieee80211_recalc_chanctx_min_def(local, ctx, NULL);
+	_ieee80211_recalc_chanctx_min_def(local, ctx, NULL, false);
 
 	return ctx;
 }
@@ -845,7 +856,7 @@ static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link,
 
 	if (new_ctx) {
 		/* recalc considering the link we'll use it for now */
-		ieee80211_recalc_chanctx_min_def(local, new_ctx, link);
+		ieee80211_recalc_chanctx_min_def(local, new_ctx, link, false);
 
 		ret = drv_assign_vif_chanctx(local, sdata, link->conf, new_ctx);
 		if (assign_on_failure || !ret) {
@@ -868,12 +879,12 @@ static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link,
 		ieee80211_recalc_chanctx_chantype(local, curr_ctx);
 		ieee80211_recalc_smps_chanctx(local, curr_ctx);
 		ieee80211_recalc_radar_chanctx(local, curr_ctx);
-		ieee80211_recalc_chanctx_min_def(local, curr_ctx, NULL);
+		ieee80211_recalc_chanctx_min_def(local, curr_ctx, NULL, false);
 	}
 
 	if (new_ctx && ieee80211_chanctx_num_assigned(local, new_ctx) > 0) {
 		ieee80211_recalc_txpower(sdata, false);
-		ieee80211_recalc_chanctx_min_def(local, new_ctx, NULL);
+		ieee80211_recalc_chanctx_min_def(local, new_ctx, NULL, false);
 	}
 
 	if (conf) {
@@ -1286,7 +1297,7 @@ ieee80211_link_use_reserved_reassign(struct ieee80211_link_data *link)
 	if (ieee80211_chanctx_refcount(local, old_ctx) == 0)
 		ieee80211_free_chanctx(local, old_ctx, false);
 
-	ieee80211_recalc_chanctx_min_def(local, new_ctx, NULL);
+	ieee80211_recalc_chanctx_min_def(local, new_ctx, NULL, false);
 	ieee80211_recalc_smps_chanctx(local, new_ctx);
 	ieee80211_recalc_radar_chanctx(local, new_ctx);
 
@@ -1552,6 +1563,31 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local)
 		goto err;
 	}
 
+	/* update station rate control and min width before switch */
+	list_for_each_entry(ctx, &local->chanctx_list, list) {
+		struct ieee80211_link_data *link;
+
+		if (ctx->replace_state != IEEE80211_CHANCTX_REPLACES_OTHER)
+			continue;
+
+		if (WARN_ON(!ctx->replace_ctx)) {
+			err = -EINVAL;
+			goto err;
+		}
+
+		list_for_each_entry(link, &ctx->reserved_links,
+				    reserved_chanctx_list) {
+			if (!ieee80211_link_has_in_place_reservation(link))
+				continue;
+
+			ieee80211_chan_bw_change(local,
+						 ieee80211_link_get_chanctx(link),
+						 true, true);
+		}
+
+		ieee80211_recalc_chanctx_min_def(local, ctx, NULL, true);
+	}
+
 	/*
 	 * All necessary vifs are ready. Perform the switch now depending on
 	 * reservations and driver capabilities.
@@ -1619,7 +1655,7 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local)
 		ieee80211_recalc_chanctx_chantype(local, ctx);
 		ieee80211_recalc_smps_chanctx(local, ctx);
 		ieee80211_recalc_radar_chanctx(local, ctx);
-		ieee80211_recalc_chanctx_min_def(local, ctx, NULL);
+		ieee80211_recalc_chanctx_min_def(local, ctx, NULL, false);
 
 		list_for_each_entry_safe(link, link_tmp, &ctx->reserved_links,
 					 reserved_chanctx_list) {
@@ -1632,6 +1668,7 @@ static int ieee80211_vif_use_reserved_switch(struct ieee80211_local *local)
 			link->reserved_chanctx = NULL;
 
 			ieee80211_link_chanctx_reservation_complete(link);
+			ieee80211_chan_bw_change(local, ctx, false, false);
 		}
 
 		/*
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 33b8efff92c1..501f02809135 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -2624,7 +2624,8 @@ void ieee80211_recalc_smps_chanctx(struct ieee80211_local *local,
 				   struct ieee80211_chanctx *chanctx);
 void ieee80211_recalc_chanctx_min_def(struct ieee80211_local *local,
 				      struct ieee80211_chanctx *ctx,
-				      struct ieee80211_link_data *rsvd_for);
+				      struct ieee80211_link_data *rsvd_for,
+				      bool check_reserved);
 bool ieee80211_is_radar_required(struct ieee80211_local *local);
 
 void ieee80211_dfs_cac_timer_work(struct wiphy *wiphy, struct wiphy_work *work);
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index c73ff7dfbdba..b3b8873a107b 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -2334,7 +2334,7 @@ void ieee80211_recalc_min_chandef(struct ieee80211_sub_if_data *sdata,
 
 		chanctx = container_of(chanctx_conf, struct ieee80211_chanctx,
 				       conf);
-		ieee80211_recalc_chanctx_min_def(local, chanctx, NULL);
+		ieee80211_recalc_chanctx_min_def(local, chanctx, NULL, false);
 	}
 }
 
-- 
cgit v1.2.3-58-ga151


From d42fcaece03654a4b21d2da88d68ed913e0b6c46 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 12 Jun 2024 14:35:56 +0200
Subject: wifi: mac80211: add ieee80211_tdls_sta_link_id()

We've open-coded this twice and will need it again,
add ieee80211_tdls_sta_link_id() to get the one link
ID for a TDLS STA.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240612143707.9f8141ae1725.I343822bbba0ae08dedb2f54a0ce87f2ae5ebeb2b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/sta_info.h | 6 ++++++
 net/mac80211/tx.c       | 6 ++----
 2 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h
index bd5e2f7146f6..9195d5a2de0a 100644
--- a/net/mac80211/sta_info.h
+++ b/net/mac80211/sta_info.h
@@ -727,6 +727,12 @@ struct sta_info {
 	struct ieee80211_sta sta;
 };
 
+static inline int ieee80211_tdls_sta_link_id(struct sta_info *sta)
+{
+	/* TDLS STA can only have a single link */
+	return sta->sta.valid_links ? __ffs(sta->sta.valid_links) : 0;
+}
+
 static inline enum nl80211_plink_state sta_plink_state(struct sta_info *sta)
 {
 #ifdef CONFIG_MAC80211_MESH
diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index f861d99e5f05..72a9ba8bc5fd 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -2774,8 +2774,7 @@ static struct sk_buff *ieee80211_build_hdr(struct ieee80211_sub_if_data *sdata,
 
 		if (tdls_peer) {
 			/* For TDLS only one link can be valid with peer STA */
-			int tdls_link_id = sta->sta.valid_links ?
-					   __ffs(sta->sta.valid_links) : 0;
+			int tdls_link_id = ieee80211_tdls_sta_link_id(sta);
 			struct ieee80211_link_data *link;
 
 			/* DA SA BSSID */
@@ -3101,8 +3100,7 @@ void ieee80211_check_fast_xmit(struct sta_info *sta)
 	case NL80211_IFTYPE_STATION:
 		if (test_sta_flag(sta, WLAN_STA_TDLS_PEER)) {
 			/* For TDLS only one link can be valid with peer STA */
-			int tdls_link_id = sta->sta.valid_links ?
-					   __ffs(sta->sta.valid_links) : 0;
+			int tdls_link_id = ieee80211_tdls_sta_link_id(sta);
 			struct ieee80211_link_data *link;
 
 			/* DA SA BSSID */
-- 
cgit v1.2.3-58-ga151


From 0b2d9d9aec2be212a28b7d14b5462c56d9adc3a3 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 12 Jun 2024 14:35:57 +0200
Subject: wifi: mac80211: correcty limit wider BW TDLS STAs

When updating a channel context, the code can apply wider
bandwidth TDLS STA channel definitions to each and every
channel context used by the device, an approach that will
surely lead to problems if there is ever more than one.

Restrict the wider BW TDLS STA consideration to only TDLS
STAs that are actually related to links using the channel
context being updated.

Fixes: 0fabfaafec3a ("mac80211: upgrade BW of TDLS peers when possible")
Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240612143707.1ad989acecde.I5c75c94d95c3f4ea84f8ff4253189f4b13bad5c3@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/chan.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c
index 942c882f1a1d..06a65dc6f6c6 100644
--- a/net/mac80211/chan.c
+++ b/net/mac80211/chan.c
@@ -793,13 +793,24 @@ void ieee80211_recalc_chanctx_chantype(struct ieee80211_local *local,
 
 	/* TDLS peers can sometimes affect the chandef width */
 	list_for_each_entry(sta, &local->sta_list, list) {
+		struct ieee80211_sub_if_data *sdata = sta->sdata;
 		struct ieee80211_chan_req tdls_chanreq = {};
+		int tdls_link_id;
+
 		if (!sta->uploaded ||
 		    !test_sta_flag(sta, WLAN_STA_TDLS_WIDER_BW) ||
 		    !test_sta_flag(sta, WLAN_STA_AUTHORIZED) ||
 		    !sta->tdls_chandef.chan)
 			continue;
 
+		tdls_link_id = ieee80211_tdls_sta_link_id(sta);
+		link = sdata_dereference(sdata->link[tdls_link_id], sdata);
+		if (!link)
+			continue;
+
+		if (rcu_access_pointer(link->conf->chanctx_conf) != conf)
+			continue;
+
 		tdls_chanreq.oper = sta->tdls_chandef;
 
 		/* note this always fills and returns &tmp if compat */
-- 
cgit v1.2.3-58-ga151


From 9cc88678db5be1d67c865dbe60afb56867698301 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 12 Jun 2024 14:38:10 +0200
Subject: wifi: mac80211: check SSID in beacon

Check that the SSID in beacons is correct, if it's not hidden
and beacon protection is enabled (otherwise there's no value).
If it doesn't match, disconnect.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240612143809.8b24a3d26a3d.I3e3ef31dbd2ec606be74d502a9d00dd9514c6885@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 32 ++++++++++++++++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index e76b887a46eb..3d207d79d11f 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -6667,6 +6667,29 @@ handle:
 	}
 }
 
+static bool ieee80211_mgd_ssid_mismatch(struct ieee80211_sub_if_data *sdata,
+					const struct ieee802_11_elems *elems)
+{
+	struct ieee80211_vif_cfg *cfg = &sdata->vif.cfg;
+	static u8 zero_ssid[IEEE80211_MAX_SSID_LEN];
+
+	if (!elems->ssid)
+		return false;
+
+	/* hidden SSID: zero length */
+	if (elems->ssid_len == 0)
+		return false;
+
+	if (elems->ssid_len != cfg->ssid_len)
+		return true;
+
+	/* hidden SSID: zeroed out */
+	if (memcmp(elems->ssid, zero_ssid, elems->ssid_len))
+		return false;
+
+	return memcmp(elems->ssid, cfg->ssid, cfg->ssid_len);
+}
+
 static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 				     struct ieee80211_hdr *hdr, size_t len,
 				     struct ieee80211_rx_status *rx_status)
@@ -6808,6 +6831,15 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 	elems = ieee802_11_parse_elems_full(&parse_params);
 	if (!elems)
 		return;
+
+	if (rx_status->flag & RX_FLAG_DECRYPTED &&
+	    ieee80211_mgd_ssid_mismatch(sdata, elems)) {
+		sdata_info(sdata, "SSID mismatch for AP %pM, disconnect\n",
+			   sdata->vif.cfg.ap_addr);
+		__ieee80211_disconnect(sdata);
+		return;
+	}
+
 	ncrc = elems->crc;
 
 	if (ieee80211_hw_check(&local->hw, PS_NULLFUNC_STACK) &&
-- 
cgit v1.2.3-58-ga151


From 5036eaffed3e29fb35c796989efb3c88ac79d87b Mon Sep 17 00:00:00 2001
From: Ilan Peer <ilan.peer@intel.com>
Date: Fri, 14 Jun 2024 09:35:41 +0200
Subject: wifi: cfg80211: Always call tracing

Call the tracing function even if the cfg80211 callbacks
are not set. This would allow better understanding of
user space actions.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Link: https://patch.msgid.link/20240614093541.018cb816e176.I28f68740a6b42144346f5c175c7874b0a669a364@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/rdev-ops.h | 56 +++++++++++++++++++------------------------------
 1 file changed, 21 insertions(+), 35 deletions(-)

diff --git a/net/wireless/rdev-ops.h b/net/wireless/rdev-ops.h
index 755af47b88b9..8f15658002ee 100644
--- a/net/wireless/rdev-ops.h
+++ b/net/wireless/rdev-ops.h
@@ -578,13 +578,11 @@ static inline int rdev_leave_ibss(struct cfg80211_registered_device *rdev,
 static inline int
 rdev_set_wiphy_params(struct cfg80211_registered_device *rdev, u32 changed)
 {
-	int ret;
-
-	if (!rdev->ops->set_wiphy_params)
-		return -EOPNOTSUPP;
+	int ret = -EOPNOTSUPP;
 
 	trace_rdev_set_wiphy_params(&rdev->wiphy, changed);
-	ret = rdev->ops->set_wiphy_params(&rdev->wiphy, changed);
+	if (rdev->ops->set_wiphy_params)
+		ret = rdev->ops->set_wiphy_params(&rdev->wiphy, changed);
 	trace_rdev_return_int(&rdev->wiphy, ret);
 	return ret;
 }
@@ -1425,13 +1423,11 @@ rdev_set_radar_background(struct cfg80211_registered_device *rdev,
 			  struct cfg80211_chan_def *chandef)
 {
 	struct wiphy *wiphy = &rdev->wiphy;
-	int ret;
-
-	if (!rdev->ops->set_radar_background)
-		return -EOPNOTSUPP;
+	int ret = -EOPNOTSUPP;
 
 	trace_rdev_set_radar_background(wiphy, chandef);
-	ret = rdev->ops->set_radar_background(wiphy, chandef);
+	if (rdev->ops->set_radar_background)
+		ret = rdev->ops->set_radar_background(wiphy, chandef);
 	trace_rdev_return_int(wiphy, ret);
 
 	return ret;
@@ -1468,13 +1464,11 @@ rdev_add_link_station(struct cfg80211_registered_device *rdev,
 		      struct net_device *dev,
 		      struct link_station_parameters *params)
 {
-	int ret;
-
-	if (!rdev->ops->add_link_station)
-		return -EOPNOTSUPP;
+	int ret = -EOPNOTSUPP;
 
 	trace_rdev_add_link_station(&rdev->wiphy, dev, params);
-	ret = rdev->ops->add_link_station(&rdev->wiphy, dev, params);
+	if (rdev->ops->add_link_station)
+		ret = rdev->ops->add_link_station(&rdev->wiphy, dev, params);
 	trace_rdev_return_int(&rdev->wiphy, ret);
 	return ret;
 }
@@ -1484,13 +1478,11 @@ rdev_mod_link_station(struct cfg80211_registered_device *rdev,
 		      struct net_device *dev,
 		      struct link_station_parameters *params)
 {
-	int ret;
-
-	if (!rdev->ops->mod_link_station)
-		return -EOPNOTSUPP;
+	int ret = -EOPNOTSUPP;
 
 	trace_rdev_mod_link_station(&rdev->wiphy, dev, params);
-	ret = rdev->ops->mod_link_station(&rdev->wiphy, dev, params);
+	if (rdev->ops->mod_link_station)
+		ret = rdev->ops->mod_link_station(&rdev->wiphy, dev, params);
 	trace_rdev_return_int(&rdev->wiphy, ret);
 	return ret;
 }
@@ -1500,13 +1492,11 @@ rdev_del_link_station(struct cfg80211_registered_device *rdev,
 		      struct net_device *dev,
 		      struct link_station_del_parameters *params)
 {
-	int ret;
-
-	if (!rdev->ops->del_link_station)
-		return -EOPNOTSUPP;
+	int ret = -EOPNOTSUPP;
 
 	trace_rdev_del_link_station(&rdev->wiphy, dev, params);
-	ret = rdev->ops->del_link_station(&rdev->wiphy, dev, params);
+	if (rdev->ops->del_link_station)
+		ret = rdev->ops->del_link_station(&rdev->wiphy, dev, params);
 	trace_rdev_return_int(&rdev->wiphy, ret);
 	return ret;
 }
@@ -1517,13 +1507,11 @@ rdev_set_hw_timestamp(struct cfg80211_registered_device *rdev,
 		      struct cfg80211_set_hw_timestamp *hwts)
 {
 	struct wiphy *wiphy = &rdev->wiphy;
-	int ret;
-
-	if (!rdev->ops->set_hw_timestamp)
-		return -EOPNOTSUPP;
+	int ret = -EOPNOTSUPP;
 
 	trace_rdev_set_hw_timestamp(wiphy, dev, hwts);
-	ret = rdev->ops->set_hw_timestamp(wiphy, dev, hwts);
+	if (rdev->ops->set_hw_timestamp)
+		ret = rdev->ops->set_hw_timestamp(wiphy, dev, hwts);
 	trace_rdev_return_int(wiphy, ret);
 
 	return ret;
@@ -1535,13 +1523,11 @@ rdev_set_ttlm(struct cfg80211_registered_device *rdev,
 	      struct cfg80211_ttlm_params *params)
 {
 	struct wiphy *wiphy = &rdev->wiphy;
-	int ret;
-
-	if (!rdev->ops->set_ttlm)
-		return -EOPNOTSUPP;
+	int ret = -EOPNOTSUPP;
 
 	trace_rdev_set_ttlm(wiphy, dev, params);
-	ret = rdev->ops->set_ttlm(wiphy, dev, params);
+	if (rdev->ops->set_ttlm)
+		ret = rdev->ops->set_ttlm(wiphy, dev, params);
 	trace_rdev_return_int(wiphy, ret);
 
 	return ret;
-- 
cgit v1.2.3-58-ga151


From 1decf05d0f4de78ef67dc3f794709258c689e09e Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Tue, 18 Jun 2024 19:25:56 +0300
Subject: wifi: mac80211: inform the low level if drv_stop() is a suspend

This will allow the low level driver to take different actions for
different flows.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618192529.739036208b6e.Ie18a2fe8e02bf2717549d39420b350cfdaf3d317@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/admtek/adm8211.c                     |  2 +-
 drivers/net/wireless/ath/ar5523/ar5523.c                  |  2 +-
 drivers/net/wireless/ath/ath10k/mac.c                     |  2 +-
 drivers/net/wireless/ath/ath11k/mac.c                     |  2 +-
 drivers/net/wireless/ath/ath12k/mac.c                     |  2 +-
 drivers/net/wireless/ath/ath5k/base.c                     |  2 +-
 drivers/net/wireless/ath/ath5k/base.h                     |  2 +-
 drivers/net/wireless/ath/ath9k/htc_drv_main.c             |  2 +-
 drivers/net/wireless/ath/ath9k/main.c                     |  2 +-
 drivers/net/wireless/ath/carl9170/main.c                  |  2 +-
 drivers/net/wireless/ath/wcn36xx/main.c                   |  2 +-
 drivers/net/wireless/atmel/at76c50x-usb.c                 |  2 +-
 drivers/net/wireless/broadcom/b43/main.c                  |  2 +-
 drivers/net/wireless/broadcom/b43legacy/main.c            |  2 +-
 .../wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c    |  2 +-
 drivers/net/wireless/intel/iwlegacy/3945-mac.c            |  2 +-
 drivers/net/wireless/intel/iwlegacy/4965-mac.c            |  2 +-
 drivers/net/wireless/intel/iwlegacy/4965.h                |  2 +-
 drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c         |  2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c         |  2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h              |  2 +-
 drivers/net/wireless/intersil/p54/main.c                  |  2 +-
 drivers/net/wireless/marvell/libertas_tf/main.c           |  2 +-
 drivers/net/wireless/marvell/mwl8k.c                      |  4 ++--
 drivers/net/wireless/mediatek/mt76/mt7603/main.c          |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7615/main.c          |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7615/usb.c           |  2 +-
 drivers/net/wireless/mediatek/mt76/mt76x0/pci.c           |  2 +-
 drivers/net/wireless/mediatek/mt76/mt76x0/usb.c           |  2 +-
 drivers/net/wireless/mediatek/mt76/mt76x2/pci_main.c      |  2 +-
 drivers/net/wireless/mediatek/mt76/mt76x2/usb_main.c      |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7915/main.c          |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7921/main.c          |  4 ++--
 drivers/net/wireless/mediatek/mt76/mt792x.h               |  4 ++--
 drivers/net/wireless/mediatek/mt76/mt792x_core.c          |  2 +-
 drivers/net/wireless/mediatek/mt76/mt792x_usb.c           |  4 ++--
 drivers/net/wireless/mediatek/mt76/mt7996/main.c          |  2 +-
 drivers/net/wireless/mediatek/mt7601u/main.c              |  2 +-
 drivers/net/wireless/purelifi/plfxlc/mac.c                |  2 +-
 drivers/net/wireless/purelifi/plfxlc/mac.h                |  2 +-
 drivers/net/wireless/purelifi/plfxlc/usb.c                |  4 ++--
 drivers/net/wireless/ralink/rt2x00/rt2x00.h               |  2 +-
 drivers/net/wireless/ralink/rt2x00/rt2x00mac.c            |  2 +-
 drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c        |  2 +-
 drivers/net/wireless/realtek/rtl818x/rtl8187/dev.c        |  2 +-
 drivers/net/wireless/realtek/rtl8xxxu/core.c              |  2 +-
 drivers/net/wireless/realtek/rtlwifi/core.c               |  4 ++--
 drivers/net/wireless/realtek/rtw88/mac80211.c             |  2 +-
 drivers/net/wireless/realtek/rtw89/mac80211.c             |  2 +-
 drivers/net/wireless/rsi/rsi_91x_mac80211.c               |  3 ++-
 drivers/net/wireless/silabs/wfx/sta.c                     |  2 +-
 drivers/net/wireless/silabs/wfx/sta.h                     |  2 +-
 drivers/net/wireless/st/cw1200/sta.c                      |  2 +-
 drivers/net/wireless/st/cw1200/sta.h                      |  2 +-
 drivers/net/wireless/ti/wl1251/main.c                     |  2 +-
 drivers/net/wireless/ti/wlcore/main.c                     |  2 +-
 drivers/net/wireless/virtual/mac80211_hwsim.c             |  2 +-
 drivers/net/wireless/zydas/zd1211rw/zd_mac.c              |  2 +-
 drivers/net/wireless/zydas/zd1211rw/zd_mac.h              |  2 +-
 drivers/net/wireless/zydas/zd1211rw/zd_usb.c              |  2 +-
 drivers/staging/vt6655/device_main.c                      |  2 +-
 drivers/staging/vt6656/main_usb.c                         |  2 +-
 include/net/mac80211.h                                    |  2 +-
 net/mac80211/driver-ops.c                                 |  6 +++---
 net/mac80211/driver-ops.h                                 |  2 +-
 net/mac80211/ieee80211_i.h                                |  2 +-
 net/mac80211/iface.c                                      |  4 ++--
 net/mac80211/pm.c                                         |  4 ++--
 net/mac80211/trace.h                                      | 15 ++++++++++++---
 net/mac80211/util.c                                       |  4 ++--
 70 files changed, 93 insertions(+), 83 deletions(-)

diff --git a/drivers/net/wireless/admtek/adm8211.c b/drivers/net/wireless/admtek/adm8211.c
index e3fd48dd3909..a2d87c3ad196 100644
--- a/drivers/net/wireless/admtek/adm8211.c
+++ b/drivers/net/wireless/admtek/adm8211.c
@@ -1550,7 +1550,7 @@ fail:
 	return retval;
 }
 
-static void adm8211_stop(struct ieee80211_hw *dev)
+static void adm8211_stop(struct ieee80211_hw *dev, bool suspend)
 {
 	struct adm8211_priv *priv = dev->priv;
 
diff --git a/drivers/net/wireless/ath/ar5523/ar5523.c b/drivers/net/wireless/ath/ar5523/ar5523.c
index 5a55db349cb5..156f3650c006 100644
--- a/drivers/net/wireless/ath/ar5523/ar5523.c
+++ b/drivers/net/wireless/ath/ar5523/ar5523.c
@@ -1061,7 +1061,7 @@ err:
 	return error;
 }
 
-static void ar5523_stop(struct ieee80211_hw *hw)
+static void ar5523_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct ar5523 *ar = hw->priv;
 
diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c
index 3bf67b2ecd6d..a5da32e87106 100644
--- a/drivers/net/wireless/ath/ath10k/mac.c
+++ b/drivers/net/wireless/ath/ath10k/mac.c
@@ -5363,7 +5363,7 @@ err:
 	return ret;
 }
 
-static void ath10k_stop(struct ieee80211_hw *hw)
+static void ath10k_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct ath10k *ar = hw->priv;
 	u32 opt;
diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index a1800c75d32b..c9a13b88c804 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c
@@ -6278,7 +6278,7 @@ err:
 	return ret;
 }
 
-static void ath11k_mac_op_stop(struct ieee80211_hw *hw)
+static void ath11k_mac_op_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct ath11k *ar = hw->priv;
 	struct htt_ppdu_stats_info *ppdu_stats, *tmp;
diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 509c02bffdae..bd3e021e2a81 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -6112,7 +6112,7 @@ static void ath12k_mac_stop(struct ath12k *ar)
 	atomic_set(&ar->num_pending_mgmt_tx, 0);
 }
 
-static void ath12k_mac_op_stop(struct ieee80211_hw *hw)
+static void ath12k_mac_op_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct ath12k_hw *ah = ath12k_hw_to_ah(hw);
 	struct ath12k *ar;
diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c
index 9f534ed2fbb3..abe41330fb69 100644
--- a/drivers/net/wireless/ath/ath5k/base.c
+++ b/drivers/net/wireless/ath/ath5k/base.c
@@ -2847,7 +2847,7 @@ static void ath5k_stop_tasklets(struct ath5k_hw *ah)
  * if another thread does a system call and the thread doing the
  * stop is preempted).
  */
-void ath5k_stop(struct ieee80211_hw *hw)
+void ath5k_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct ath5k_hw *ah = hw->priv;
 	int ret;
diff --git a/drivers/net/wireless/ath/ath5k/base.h b/drivers/net/wireless/ath/ath5k/base.h
index 97469d0fbad7..594e5b945cb7 100644
--- a/drivers/net/wireless/ath/ath5k/base.h
+++ b/drivers/net/wireless/ath/ath5k/base.h
@@ -92,7 +92,7 @@ void ath5k_vif_iter(void *data, u8 *mac, struct ieee80211_vif *vif);
 bool ath5k_any_vif_assoc(struct ath5k_hw *ah);
 
 int ath5k_start(struct ieee80211_hw *hw);
-void ath5k_stop(struct ieee80211_hw *hw);
+void ath5k_stop(struct ieee80211_hw *hw, bool suspend);
 
 void ath5k_beacon_update_timers(struct ath5k_hw *ah, u64 bc_tsf);
 int ath5k_beacon_update(struct ieee80211_hw *hw, struct ieee80211_vif *vif);
diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_main.c b/drivers/net/wireless/ath/ath9k/htc_drv_main.c
index b389e19381c4..8a03bcc2789e 100644
--- a/drivers/net/wireless/ath/ath9k/htc_drv_main.c
+++ b/drivers/net/wireless/ath/ath9k/htc_drv_main.c
@@ -973,7 +973,7 @@ static int ath9k_htc_start(struct ieee80211_hw *hw)
 	return ret;
 }
 
-static void ath9k_htc_stop(struct ieee80211_hw *hw)
+static void ath9k_htc_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct ath9k_htc_priv *priv = hw->priv;
 	struct ath_hw *ah = priv->ah;
diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c
index 01173aac3045..b92c89dad8de 100644
--- a/drivers/net/wireless/ath/ath9k/main.c
+++ b/drivers/net/wireless/ath/ath9k/main.c
@@ -895,7 +895,7 @@ static void ath9k_pending_key_del(struct ath_softc *sc, u8 keyix)
 	ath_key_delete(common, keyix);
 }
 
-static void ath9k_stop(struct ieee80211_hw *hw)
+static void ath9k_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct ath_softc *sc = hw->priv;
 	struct ath_hw *ah = sc->sc_ah;
diff --git a/drivers/net/wireless/ath/carl9170/main.c b/drivers/net/wireless/ath/carl9170/main.c
index 7e7797bf44b7..755c068e4197 100644
--- a/drivers/net/wireless/ath/carl9170/main.c
+++ b/drivers/net/wireless/ath/carl9170/main.c
@@ -439,7 +439,7 @@ static void carl9170_cancel_worker(struct ar9170 *ar)
 	cancel_work_sync(&ar->ampdu_work);
 }
 
-static void carl9170_op_stop(struct ieee80211_hw *hw)
+static void carl9170_op_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct ar9170 *ar = hw->priv;
 
diff --git a/drivers/net/wireless/ath/wcn36xx/main.c b/drivers/net/wireless/ath/wcn36xx/main.c
index e760d8002e09..408776562a7e 100644
--- a/drivers/net/wireless/ath/wcn36xx/main.c
+++ b/drivers/net/wireless/ath/wcn36xx/main.c
@@ -278,7 +278,7 @@ out_err:
 	return ret;
 }
 
-static void wcn36xx_stop(struct ieee80211_hw *hw)
+static void wcn36xx_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct wcn36xx *wcn = hw->priv;
 
diff --git a/drivers/net/wireless/atmel/at76c50x-usb.c b/drivers/net/wireless/atmel/at76c50x-usb.c
index baa53cfefe48..504e05ea30f2 100644
--- a/drivers/net/wireless/atmel/at76c50x-usb.c
+++ b/drivers/net/wireless/atmel/at76c50x-usb.c
@@ -1850,7 +1850,7 @@ error:
 	return 0;
 }
 
-static void at76_mac80211_stop(struct ieee80211_hw *hw)
+static void at76_mac80211_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct at76_priv *priv = hw->priv;
 
diff --git a/drivers/net/wireless/broadcom/b43/main.c b/drivers/net/wireless/broadcom/b43/main.c
index badb2f494035..8e56dcf9309d 100644
--- a/drivers/net/wireless/broadcom/b43/main.c
+++ b/drivers/net/wireless/broadcom/b43/main.c
@@ -5078,7 +5078,7 @@ static int b43_op_start(struct ieee80211_hw *hw)
 	return err;
 }
 
-static void b43_op_stop(struct ieee80211_hw *hw)
+static void b43_op_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct b43_wl *wl = hw_to_b43_wl(hw);
 	struct b43_wldev *dev = wl->current_dev;
diff --git a/drivers/net/wireless/broadcom/b43legacy/main.c b/drivers/net/wireless/broadcom/b43legacy/main.c
index 18eb610f600a..441d6440671b 100644
--- a/drivers/net/wireless/broadcom/b43legacy/main.c
+++ b/drivers/net/wireless/broadcom/b43legacy/main.c
@@ -3485,7 +3485,7 @@ out_mutex_unlock:
 	return err;
 }
 
-static void b43legacy_op_stop(struct ieee80211_hw *hw)
+static void b43legacy_op_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct b43legacy_wl *wl = hw_to_b43legacy_wl(hw);
 	struct b43legacy_wldev *dev = wl->current_dev;
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
index 860ef9c11c46..9ac6f3c4bdc7 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
@@ -457,7 +457,7 @@ static int brcms_ops_start(struct ieee80211_hw *hw)
 	return err;
 }
 
-static void brcms_ops_stop(struct ieee80211_hw *hw)
+static void brcms_ops_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct brcms_info *wl = hw->priv;
 	int status;
diff --git a/drivers/net/wireless/intel/iwlegacy/3945-mac.c b/drivers/net/wireless/intel/iwlegacy/3945-mac.c
index 075b705a8d7b..74fc76c00ebc 100644
--- a/drivers/net/wireless/intel/iwlegacy/3945-mac.c
+++ b/drivers/net/wireless/intel/iwlegacy/3945-mac.c
@@ -2813,7 +2813,7 @@ out_release_irq:
 }
 
 static void
-il3945_mac_stop(struct ieee80211_hw *hw)
+il3945_mac_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct il_priv *il = hw->priv;
 
diff --git a/drivers/net/wireless/intel/iwlegacy/4965-mac.c b/drivers/net/wireless/intel/iwlegacy/4965-mac.c
index d018f56be966..1600c344edbb 100644
--- a/drivers/net/wireless/intel/iwlegacy/4965-mac.c
+++ b/drivers/net/wireless/intel/iwlegacy/4965-mac.c
@@ -5820,7 +5820,7 @@ out:
 }
 
 void
-il4965_mac_stop(struct ieee80211_hw *hw)
+il4965_mac_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct il_priv *il = hw->priv;
 
diff --git a/drivers/net/wireless/intel/iwlegacy/4965.h b/drivers/net/wireless/intel/iwlegacy/4965.h
index 863e3792d153..951f2245fefb 100644
--- a/drivers/net/wireless/intel/iwlegacy/4965.h
+++ b/drivers/net/wireless/intel/iwlegacy/4965.h
@@ -151,7 +151,7 @@ void il4965_mac_tx(struct ieee80211_hw *hw,
 		   struct ieee80211_tx_control *control,
 		   struct sk_buff *skb);
 int il4965_mac_start(struct ieee80211_hw *hw);
-void il4965_mac_stop(struct ieee80211_hw *hw);
+void il4965_mac_stop(struct ieee80211_hw *hw, bool suspend);
 void il4965_configure_filter(struct ieee80211_hw *hw,
 			     unsigned int changed_flags,
 			     unsigned int *total_flags, u64 multicast);
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
index 826f129ecefa..74d163e56511 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/mac80211.c
@@ -300,7 +300,7 @@ static int iwlagn_mac_start(struct ieee80211_hw *hw)
 	return ret;
 }
 
-static void iwlagn_mac_stop(struct ieee80211_hw *hw)
+static void iwlagn_mac_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct iwl_priv *priv = IWL_MAC80211_GET_DVM(hw);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 81103a974b20..896e560f5a82 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1376,7 +1376,7 @@ void __iwl_mvm_mac_stop(struct iwl_mvm *mvm)
 	}
 }
 
-void iwl_mvm_mac_stop(struct ieee80211_hw *hw)
+void iwl_mvm_mac_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index bce517effe55..54c3553db219 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -2818,7 +2818,7 @@ int iwl_mvm_op_set_antenna(struct ieee80211_hw *hw, u32 tx_ant, u32 rx_ant);
 int iwl_mvm_mac_start(struct ieee80211_hw *hw);
 void iwl_mvm_mac_reconfig_complete(struct ieee80211_hw *hw,
 				   enum ieee80211_reconfig_type reconfig_type);
-void iwl_mvm_mac_stop(struct ieee80211_hw *hw);
+void iwl_mvm_mac_stop(struct ieee80211_hw *hw, bool suspend);
 static inline int iwl_mvm_mac_config(struct ieee80211_hw *hw, u32 changed)
 {
 	return 0;
diff --git a/drivers/net/wireless/intersil/p54/main.c b/drivers/net/wireless/intersil/p54/main.c
index 687841b2fa2a..42111bb53f58 100644
--- a/drivers/net/wireless/intersil/p54/main.c
+++ b/drivers/net/wireless/intersil/p54/main.c
@@ -197,7 +197,7 @@ out:
 	return err;
 }
 
-static void p54_stop(struct ieee80211_hw *dev)
+static void p54_stop(struct ieee80211_hw *dev, bool suspend)
 {
 	struct p54_common *priv = dev->priv;
 	int i;
diff --git a/drivers/net/wireless/marvell/libertas_tf/main.c b/drivers/net/wireless/marvell/libertas_tf/main.c
index 9cca69fe04d7..b47a832b9ae2 100644
--- a/drivers/net/wireless/marvell/libertas_tf/main.c
+++ b/drivers/net/wireless/marvell/libertas_tf/main.c
@@ -267,7 +267,7 @@ static int lbtf_op_start(struct ieee80211_hw *hw)
 	return 0;
 }
 
-static void lbtf_op_stop(struct ieee80211_hw *hw)
+static void lbtf_op_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct lbtf_private *priv = hw->priv;
 	unsigned long flags;
diff --git a/drivers/net/wireless/marvell/mwl8k.c b/drivers/net/wireless/marvell/mwl8k.c
index 241a02a0accd..b130e057370f 100644
--- a/drivers/net/wireless/marvell/mwl8k.c
+++ b/drivers/net/wireless/marvell/mwl8k.c
@@ -4768,7 +4768,7 @@ static int mwl8k_start(struct ieee80211_hw *hw)
 	return rc;
 }
 
-static void mwl8k_stop(struct ieee80211_hw *hw)
+static void mwl8k_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mwl8k_priv *priv = hw->priv;
 	int i;
@@ -6023,7 +6023,7 @@ static int mwl8k_reload_firmware(struct ieee80211_hw *hw, char *fw_image)
 	struct mwl8k_priv *priv = hw->priv;
 	struct mwl8k_vif *vif, *tmp_vif;
 
-	mwl8k_stop(hw);
+	mwl8k_stop(hw, false);
 	mwl8k_rxq_deinit(hw, 0);
 
 	/*
diff --git a/drivers/net/wireless/mediatek/mt76/mt7603/main.c b/drivers/net/wireless/mediatek/mt76/mt7603/main.c
index 9b49267b1eab..f35fa643c0da 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7603/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7603/main.c
@@ -23,7 +23,7 @@ mt7603_start(struct ieee80211_hw *hw)
 }
 
 static void
-mt7603_stop(struct ieee80211_hw *hw)
+mt7603_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt7603_dev *dev = hw->priv;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/main.c b/drivers/net/wireless/mediatek/mt76/mt7615/main.c
index c27acaf0eb1c..50e262c1622f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/main.c
@@ -91,7 +91,7 @@ out:
 	return ret;
 }
 
-static void mt7615_stop(struct ieee80211_hw *hw)
+static void mt7615_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt7615_dev *dev = mt7615_hw_dev(hw);
 	struct mt7615_phy *phy = mt7615_hw_phy(hw);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/usb.c b/drivers/net/wireless/mediatek/mt76/mt7615/usb.c
index df737e1ff27b..9335ca0776fe 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/usb.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/usb.c
@@ -79,7 +79,7 @@ static void mt7663u_copy(struct mt76_dev *dev, u32 offset,
 	mutex_unlock(&usb->usb_ctrl_mtx);
 }
 
-static void mt7663u_stop(struct ieee80211_hw *hw)
+static void mt7663u_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt7615_phy *phy = mt7615_hw_phy(hw);
 	struct mt7615_dev *dev = hw->priv;
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x0/pci.c b/drivers/net/wireless/mediatek/mt76/mt76x0/pci.c
index 79b7996ad1a8..2ecee7c5c80d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x0/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x0/pci.c
@@ -44,7 +44,7 @@ static void mt76x0e_stop_hw(struct mt76x02_dev *dev)
 	mt76_clear(dev, MT_WPDMA_GLO_CFG, MT_WPDMA_GLO_CFG_RX_DMA_EN);
 }
 
-static void mt76x0e_stop(struct ieee80211_hw *hw)
+static void mt76x0e_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt76x02_dev *dev = hw->priv;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x0/usb.c b/drivers/net/wireless/mediatek/mt76/mt76x0/usb.c
index bba44f289b4e..390f502e97f0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x0/usb.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x0/usb.c
@@ -77,7 +77,7 @@ static void mt76x0u_cleanup(struct mt76x02_dev *dev)
 	mt76u_queues_deinit(&dev->mt76);
 }
 
-static void mt76x0u_stop(struct ieee80211_hw *hw)
+static void mt76x0u_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt76x02_dev *dev = hw->priv;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/pci_main.c b/drivers/net/wireless/mediatek/mt76/mt76x2/pci_main.c
index bfc8c69f43fa..6accea551319 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x2/pci_main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x2/pci_main.c
@@ -24,7 +24,7 @@ mt76x2_start(struct ieee80211_hw *hw)
 }
 
 static void
-mt76x2_stop(struct ieee80211_hw *hw)
+mt76x2_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt76x02_dev *dev = hw->priv;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_main.c b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_main.c
index 9fe390fdd730..ba0241c36672 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x2/usb_main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x2/usb_main.c
@@ -22,7 +22,7 @@ static int mt76x2u_start(struct ieee80211_hw *hw)
 	return 0;
 }
 
-static void mt76x2u_stop(struct ieee80211_hw *hw)
+static void mt76x2u_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt76x02_dev *dev = hw->priv;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/main.c b/drivers/net/wireless/mediatek/mt76/mt7915/main.c
index 2624edbb59a1..049223df9beb 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/main.c
@@ -108,7 +108,7 @@ static int mt7915_start(struct ieee80211_hw *hw)
 	return ret;
 }
 
-static void mt7915_stop(struct ieee80211_hw *hw)
+static void mt7915_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt7915_dev *dev = mt7915_hw_dev(hw);
 	struct mt7915_phy *phy = mt7915_hw_phy(hw);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/main.c b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
index 3e3ad3518d85..4f30426afbb7 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
@@ -268,7 +268,7 @@ static int mt7921_start(struct ieee80211_hw *hw)
 	return err;
 }
 
-static void mt7921_stop(struct ieee80211_hw *hw)
+static void mt7921_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
 	int err = 0;
@@ -281,7 +281,7 @@ static void mt7921_stop(struct ieee80211_hw *hw)
 			return;
 	}
 
-	mt792x_stop(hw);
+	mt792x_stop(hw, false);
 }
 
 static int
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index 20578497a405..cf14a38c5e72 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -251,7 +251,7 @@ static inline bool mt792x_dma_need_reinit(struct mt792x_dev *dev)
 #define mt792x_mutex_release(dev)	\
 	mt76_connac_mutex_release(&(dev)->mt76, &(dev)->pm)
 
-void mt792x_stop(struct ieee80211_hw *hw);
+void mt792x_stop(struct ieee80211_hw *hw, bool suspend);
 void mt792x_pm_wake_work(struct work_struct *work);
 void mt792x_pm_power_save_work(struct work_struct *work);
 void mt792x_reset(struct mt76_dev *mdev);
@@ -368,7 +368,7 @@ void mt792xu_wr(struct mt76_dev *dev, u32 addr, u32 val);
 u32 mt792xu_rmw(struct mt76_dev *dev, u32 addr, u32 mask, u32 val);
 void mt792xu_copy(struct mt76_dev *dev, u32 offset, const void *data, int len);
 void mt792xu_disconnect(struct usb_interface *usb_intf);
-void mt792xu_stop(struct ieee80211_hw *hw);
+void mt792xu_stop(struct ieee80211_hw *hw, bool suspend);
 
 static inline void
 mt792x_skb_add_usb_sdio_hdr(struct mt792x_dev *dev, struct sk_buff *skb,
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_core.c b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
index a405af8d9052..4adca99eb9b8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_core.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
@@ -91,7 +91,7 @@ void mt792x_tx(struct ieee80211_hw *hw, struct ieee80211_tx_control *control,
 }
 EXPORT_SYMBOL_GPL(mt792x_tx);
 
-void mt792x_stop(struct ieee80211_hw *hw)
+void mt792x_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
 	struct mt792x_phy *phy = mt792x_hw_phy(hw);
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_usb.c b/drivers/net/wireless/mediatek/mt76/mt792x_usb.c
index b49668a4b784..76272a03b22e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_usb.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_usb.c
@@ -285,12 +285,12 @@ int mt792xu_init_reset(struct mt792x_dev *dev)
 }
 EXPORT_SYMBOL_GPL(mt792xu_init_reset);
 
-void mt792xu_stop(struct ieee80211_hw *hw)
+void mt792xu_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
 
 	mt76u_stop_tx(&dev->mt76);
-	mt792x_stop(hw);
+	mt792x_stop(hw, false);
 }
 EXPORT_SYMBOL_GPL(mt792xu_stop);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/main.c b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
index 7c97140d8255..bce082038219 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/main.c
@@ -93,7 +93,7 @@ static int mt7996_start(struct ieee80211_hw *hw)
 	return ret;
 }
 
-static void mt7996_stop(struct ieee80211_hw *hw)
+static void mt7996_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt7996_dev *dev = mt7996_hw_dev(hw);
 	struct mt7996_phy *phy = mt7996_hw_phy(hw);
diff --git a/drivers/net/wireless/mediatek/mt7601u/main.c b/drivers/net/wireless/mediatek/mt7601u/main.c
index a7330576486b..7570c6ceecea 100644
--- a/drivers/net/wireless/mediatek/mt7601u/main.c
+++ b/drivers/net/wireless/mediatek/mt7601u/main.c
@@ -28,7 +28,7 @@ out:
 	return ret;
 }
 
-static void mt7601u_stop(struct ieee80211_hw *hw)
+static void mt7601u_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mt7601u_dev *dev = hw->priv;
 
diff --git a/drivers/net/wireless/purelifi/plfxlc/mac.c b/drivers/net/wireless/purelifi/plfxlc/mac.c
index 641f847d47ab..eae93efa6150 100644
--- a/drivers/net/wireless/purelifi/plfxlc/mac.c
+++ b/drivers/net/wireless/purelifi/plfxlc/mac.c
@@ -111,7 +111,7 @@ int plfxlc_op_start(struct ieee80211_hw *hw)
 	return 0;
 }
 
-void plfxlc_op_stop(struct ieee80211_hw *hw)
+void plfxlc_op_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct plfxlc_mac *mac = plfxlc_hw_mac(hw);
 
diff --git a/drivers/net/wireless/purelifi/plfxlc/mac.h b/drivers/net/wireless/purelifi/plfxlc/mac.h
index 49b92413729b..9384acddcf26 100644
--- a/drivers/net/wireless/purelifi/plfxlc/mac.h
+++ b/drivers/net/wireless/purelifi/plfxlc/mac.h
@@ -178,7 +178,7 @@ int plfxlc_mac_rx(struct ieee80211_hw *hw, const u8 *buffer,
 void plfxlc_mac_tx_failed(struct urb *urb);
 void plfxlc_mac_tx_to_dev(struct sk_buff *skb, int error);
 int plfxlc_op_start(struct ieee80211_hw *hw);
-void plfxlc_op_stop(struct ieee80211_hw *hw);
+void plfxlc_op_stop(struct ieee80211_hw *hw, bool suspend);
 int plfxlc_restore_settings(struct plfxlc_mac *mac);
 
 #endif /* PLFXLC_MAC_H */
diff --git a/drivers/net/wireless/purelifi/plfxlc/usb.c b/drivers/net/wireless/purelifi/plfxlc/usb.c
index 311676c1ece0..15334940287d 100644
--- a/drivers/net/wireless/purelifi/plfxlc/usb.c
+++ b/drivers/net/wireless/purelifi/plfxlc/usb.c
@@ -408,7 +408,7 @@ void plfxlc_usb_init(struct plfxlc_usb *usb, struct ieee80211_hw *hw,
 
 void plfxlc_usb_release(struct plfxlc_usb *usb)
 {
-	plfxlc_op_stop(plfxlc_usb_to_hw(usb));
+	plfxlc_op_stop(plfxlc_usb_to_hw(usb), false);
 	plfxlc_usb_disable_tx(usb);
 	plfxlc_usb_disable_rx(usb);
 	usb_set_intfdata(usb->intf, NULL);
@@ -761,7 +761,7 @@ static void plfxlc_usb_resume(struct plfxlc_usb *usb)
 
 static void plfxlc_usb_stop(struct plfxlc_usb *usb)
 {
-	plfxlc_op_stop(plfxlc_usb_to_hw(usb));
+	plfxlc_op_stop(plfxlc_usb_to_hw(usb), false);
 	plfxlc_usb_disable_tx(usb);
 	plfxlc_usb_disable_rx(usb);
 
diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00.h b/drivers/net/wireless/ralink/rt2x00/rt2x00.h
index bb648f95dfdd..dfb4bb370f01 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/ralink/rt2x00/rt2x00.h
@@ -1450,7 +1450,7 @@ void rt2x00mac_tx(struct ieee80211_hw *hw,
 		  struct ieee80211_tx_control *control,
 		  struct sk_buff *skb);
 int rt2x00mac_start(struct ieee80211_hw *hw);
-void rt2x00mac_stop(struct ieee80211_hw *hw);
+void rt2x00mac_stop(struct ieee80211_hw *hw, bool suspend);
 void rt2x00mac_reconfig_complete(struct ieee80211_hw *hw,
 				 enum ieee80211_reconfig_type reconfig_type);
 int rt2x00mac_add_interface(struct ieee80211_hw *hw,
diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c b/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c
index 75fda72c14ca..451632488805 100644
--- a/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c
+++ b/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c
@@ -178,7 +178,7 @@ int rt2x00mac_start(struct ieee80211_hw *hw)
 }
 EXPORT_SYMBOL_GPL(rt2x00mac_start);
 
-void rt2x00mac_stop(struct ieee80211_hw *hw)
+void rt2x00mac_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct rt2x00_dev *rt2x00dev = hw->priv;
 
diff --git a/drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c b/drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c
index 77b6cb7e1f6b..ded8d4d59289 100644
--- a/drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c
+++ b/drivers/net/wireless/realtek/rtl818x/rtl8180/dev.c
@@ -1249,7 +1249,7 @@ static int rtl8180_start(struct ieee80211_hw *dev)
 	return ret;
 }
 
-static void rtl8180_stop(struct ieee80211_hw *dev)
+static void rtl8180_stop(struct ieee80211_hw *dev, bool suspend)
 {
 	struct rtl8180_priv *priv = dev->priv;
 	u8 reg;
diff --git a/drivers/net/wireless/realtek/rtl818x/rtl8187/dev.c b/drivers/net/wireless/realtek/rtl818x/rtl8187/dev.c
index 78d99afa373d..220ac5bdf279 100644
--- a/drivers/net/wireless/realtek/rtl818x/rtl8187/dev.c
+++ b/drivers/net/wireless/realtek/rtl818x/rtl8187/dev.c
@@ -1019,7 +1019,7 @@ rtl8187_start_exit:
 	return ret;
 }
 
-static void rtl8187_stop(struct ieee80211_hw *dev)
+static void rtl8187_stop(struct ieee80211_hw *dev, bool suspend)
 {
 	struct rtl8187_priv *priv = dev->priv;
 	struct sk_buff *skb;
diff --git a/drivers/net/wireless/realtek/rtl8xxxu/core.c b/drivers/net/wireless/realtek/rtl8xxxu/core.c
index 89a841b4e8d5..3685dbefc9bd 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/core.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/core.c
@@ -7521,7 +7521,7 @@ error_out:
 	return ret;
 }
 
-static void rtl8xxxu_stop(struct ieee80211_hw *hw)
+static void rtl8xxxu_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct rtl8xxxu_priv *priv = hw->priv;
 	unsigned long flags;
diff --git a/drivers/net/wireless/realtek/rtlwifi/core.c b/drivers/net/wireless/realtek/rtlwifi/core.c
index 42b7db12b1bd..7537f04b1930 100644
--- a/drivers/net/wireless/realtek/rtlwifi/core.c
+++ b/drivers/net/wireless/realtek/rtlwifi/core.c
@@ -144,7 +144,7 @@ static int rtl_op_start(struct ieee80211_hw *hw)
 	return err;
 }
 
-static void rtl_op_stop(struct ieee80211_hw *hw)
+static void rtl_op_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct rtl_priv *rtlpriv = rtl_priv(hw);
 	struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
@@ -547,7 +547,7 @@ static int rtl_op_suspend(struct ieee80211_hw *hw,
 	rtlhal->enter_pnp_sleep = true;
 
 	rtl_lps_leave(hw, true);
-	rtl_op_stop(hw);
+	rtl_op_stop(hw, false);
 	device_set_wakeup_enable(wiphy_dev(hw->wiphy), true);
 	return 0;
 }
diff --git a/drivers/net/wireless/realtek/rtw88/mac80211.c b/drivers/net/wireless/realtek/rtw88/mac80211.c
index 0acebbfa13c4..63326b352738 100644
--- a/drivers/net/wireless/realtek/rtw88/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw88/mac80211.c
@@ -62,7 +62,7 @@ static int rtw_ops_start(struct ieee80211_hw *hw)
 	return ret;
 }
 
-static void rtw_ops_stop(struct ieee80211_hw *hw)
+static void rtw_ops_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct rtw_dev *rtwdev = hw->priv;
 
diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c
index 41b286da3d59..722d09e9fbb5 100644
--- a/drivers/net/wireless/realtek/rtw89/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw89/mac80211.c
@@ -66,7 +66,7 @@ static int rtw89_ops_start(struct ieee80211_hw *hw)
 	return ret;
 }
 
-static void rtw89_ops_stop(struct ieee80211_hw *hw)
+static void rtw89_ops_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct rtw89_dev *rtwdev = hw->priv;
 
diff --git a/drivers/net/wireless/rsi/rsi_91x_mac80211.c b/drivers/net/wireless/rsi/rsi_91x_mac80211.c
index 211fa25b9a78..3425a473b9a1 100644
--- a/drivers/net/wireless/rsi/rsi_91x_mac80211.c
+++ b/drivers/net/wireless/rsi/rsi_91x_mac80211.c
@@ -410,10 +410,11 @@ static int rsi_mac80211_start(struct ieee80211_hw *hw)
 /**
  * rsi_mac80211_stop() - This is the last handler that 802.11 module calls.
  * @hw: Pointer to the ieee80211_hw structure.
+ * @suspend: true if the this was called from suspend flow.
  *
  * Return: None.
  */
-static void rsi_mac80211_stop(struct ieee80211_hw *hw)
+static void rsi_mac80211_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct rsi_hw *adapter = hw->priv;
 	struct rsi_common *common = adapter->priv;
diff --git a/drivers/net/wireless/silabs/wfx/sta.c b/drivers/net/wireless/silabs/wfx/sta.c
index a904602f02ce..216d43c8bd6e 100644
--- a/drivers/net/wireless/silabs/wfx/sta.c
+++ b/drivers/net/wireless/silabs/wfx/sta.c
@@ -805,7 +805,7 @@ int wfx_start(struct ieee80211_hw *hw)
 	return 0;
 }
 
-void wfx_stop(struct ieee80211_hw *hw)
+void wfx_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct wfx_dev *wdev = hw->priv;
 
diff --git a/drivers/net/wireless/silabs/wfx/sta.h b/drivers/net/wireless/silabs/wfx/sta.h
index c478ddcb934b..7817c7c6f3dd 100644
--- a/drivers/net/wireless/silabs/wfx/sta.h
+++ b/drivers/net/wireless/silabs/wfx/sta.h
@@ -20,7 +20,7 @@ struct wfx_sta_priv {
 
 /* mac80211 interface */
 int wfx_start(struct ieee80211_hw *hw);
-void wfx_stop(struct ieee80211_hw *hw);
+void wfx_stop(struct ieee80211_hw *hw, bool suspend);
 int wfx_config(struct ieee80211_hw *hw, u32 changed);
 int wfx_set_rts_threshold(struct ieee80211_hw *hw, u32 value);
 void wfx_set_default_unicast_key(struct ieee80211_hw *hw, struct ieee80211_vif *vif, int idx);
diff --git a/drivers/net/wireless/st/cw1200/sta.c b/drivers/net/wireless/st/cw1200/sta.c
index 8ef1d06b9bbd..c259da8161e4 100644
--- a/drivers/net/wireless/st/cw1200/sta.c
+++ b/drivers/net/wireless/st/cw1200/sta.c
@@ -90,7 +90,7 @@ out:
 	return ret;
 }
 
-void cw1200_stop(struct ieee80211_hw *dev)
+void cw1200_stop(struct ieee80211_hw *dev, bool suspend)
 {
 	struct cw1200_common *priv = dev->priv;
 	LIST_HEAD(list);
diff --git a/drivers/net/wireless/st/cw1200/sta.h b/drivers/net/wireless/st/cw1200/sta.h
index a49f187c7049..b955b92cfd73 100644
--- a/drivers/net/wireless/st/cw1200/sta.h
+++ b/drivers/net/wireless/st/cw1200/sta.h
@@ -13,7 +13,7 @@
 /* mac80211 API								*/
 
 int cw1200_start(struct ieee80211_hw *dev);
-void cw1200_stop(struct ieee80211_hw *dev);
+void cw1200_stop(struct ieee80211_hw *dev, bool suspend);
 int cw1200_add_interface(struct ieee80211_hw *dev,
 			 struct ieee80211_vif *vif);
 void cw1200_remove_interface(struct ieee80211_hw *dev,
diff --git a/drivers/net/wireless/ti/wl1251/main.c b/drivers/net/wireless/ti/wl1251/main.c
index 0da2d29dd7bd..bb53d681c11b 100644
--- a/drivers/net/wireless/ti/wl1251/main.c
+++ b/drivers/net/wireless/ti/wl1251/main.c
@@ -415,7 +415,7 @@ out:
 	return ret;
 }
 
-static void wl1251_op_stop(struct ieee80211_hw *hw)
+static void wl1251_op_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct wl1251 *wl = hw->priv;
 
diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c
index 35d1114a28aa..7e1d30f89855 100644
--- a/drivers/net/wireless/ti/wlcore/main.c
+++ b/drivers/net/wireless/ti/wlcore/main.c
@@ -2085,7 +2085,7 @@ static void wlcore_op_stop_locked(struct wl1271 *wl)
 	memset(wl->reg_ch_conf_last, 0, sizeof(wl->reg_ch_conf_last));
 }
 
-static void wlcore_op_stop(struct ieee80211_hw *hw)
+static void wlcore_op_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct wl1271 *wl = hw->priv;
 
diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index fbf24870209d..8491eb32f760 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -2098,7 +2098,7 @@ static int mac80211_hwsim_start(struct ieee80211_hw *hw)
 }
 
 
-static void mac80211_hwsim_stop(struct ieee80211_hw *hw)
+static void mac80211_hwsim_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct mac80211_hwsim_data *data = hw->priv;
 	int i;
diff --git a/drivers/net/wireless/zydas/zd1211rw/zd_mac.c b/drivers/net/wireless/zydas/zd1211rw/zd_mac.c
index 900c063bd724..f90c33d19b39 100644
--- a/drivers/net/wireless/zydas/zd1211rw/zd_mac.c
+++ b/drivers/net/wireless/zydas/zd1211rw/zd_mac.c
@@ -326,7 +326,7 @@ out:
 	return r;
 }
 
-void zd_op_stop(struct ieee80211_hw *hw)
+void zd_op_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct zd_mac *mac = zd_hw_mac(hw);
 	struct zd_chip *chip = &mac->chip;
diff --git a/drivers/net/wireless/zydas/zd1211rw/zd_mac.h b/drivers/net/wireless/zydas/zd1211rw/zd_mac.h
index 5ff84bdc5a4c..053748a474ec 100644
--- a/drivers/net/wireless/zydas/zd1211rw/zd_mac.h
+++ b/drivers/net/wireless/zydas/zd1211rw/zd_mac.h
@@ -303,7 +303,7 @@ void zd_mac_tx_failed(struct urb *urb);
 void zd_mac_tx_to_dev(struct sk_buff *skb, int error);
 
 int zd_op_start(struct ieee80211_hw *hw);
-void zd_op_stop(struct ieee80211_hw *hw);
+void zd_op_stop(struct ieee80211_hw *hw, bool suspend);
 int zd_restore_settings(struct zd_mac *mac);
 
 #ifdef DEBUG
diff --git a/drivers/net/wireless/zydas/zd1211rw/zd_usb.c b/drivers/net/wireless/zydas/zd1211rw/zd_usb.c
index 2ee4218da1c5..a8a94edf2a70 100644
--- a/drivers/net/wireless/zydas/zd1211rw/zd_usb.c
+++ b/drivers/net/wireless/zydas/zd1211rw/zd_usb.c
@@ -1476,7 +1476,7 @@ static void zd_usb_stop(struct zd_usb *usb)
 {
 	dev_dbg_f(zd_usb_dev(usb), "\n");
 
-	zd_op_stop(zd_usb_to_hw(usb));
+	zd_op_stop(zd_usb_to_hw(usb), false);
 
 	zd_usb_disable_tx(usb);
 	zd_usb_disable_rx(usb);
diff --git a/drivers/staging/vt6655/device_main.c b/drivers/staging/vt6655/device_main.c
index 283804b49e91..3ff8103366c1 100644
--- a/drivers/staging/vt6655/device_main.c
+++ b/drivers/staging/vt6655/device_main.c
@@ -1339,7 +1339,7 @@ err_free_rings:
 	return ret;
 }
 
-static void vnt_stop(struct ieee80211_hw *hw)
+static void vnt_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct vnt_private *priv = hw->priv;
 
diff --git a/drivers/staging/vt6656/main_usb.c b/drivers/staging/vt6656/main_usb.c
index 7bbed462f062..4f09e733e7a8 100644
--- a/drivers/staging/vt6656/main_usb.c
+++ b/drivers/staging/vt6656/main_usb.c
@@ -613,7 +613,7 @@ err:
 	return ret;
 }
 
-static void vnt_stop(struct ieee80211_hw *hw)
+static void vnt_stop(struct ieee80211_hw *hw, bool suspend)
 {
 	struct vnt_private *priv = hw->priv;
 	int i;
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index ecfa65ade226..9c96e8ae9ef7 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -4444,7 +4444,7 @@ struct ieee80211_ops {
 		   struct ieee80211_tx_control *control,
 		   struct sk_buff *skb);
 	int (*start)(struct ieee80211_hw *hw);
-	void (*stop)(struct ieee80211_hw *hw);
+	void (*stop)(struct ieee80211_hw *hw, bool suspend);
 #ifdef CONFIG_PM
 	int (*suspend)(struct ieee80211_hw *hw, struct cfg80211_wowlan *wowlan);
 	int (*resume)(struct ieee80211_hw *hw);
diff --git a/net/mac80211/driver-ops.c b/net/mac80211/driver-ops.c
index dce37ba8ebe3..bef43990f07a 100644
--- a/net/mac80211/driver-ops.c
+++ b/net/mac80211/driver-ops.c
@@ -33,7 +33,7 @@ int drv_start(struct ieee80211_local *local)
 	return ret;
 }
 
-void drv_stop(struct ieee80211_local *local)
+void drv_stop(struct ieee80211_local *local, bool suspend)
 {
 	might_sleep();
 	lockdep_assert_wiphy(local->hw.wiphy);
@@ -41,8 +41,8 @@ void drv_stop(struct ieee80211_local *local)
 	if (WARN_ON(!local->started))
 		return;
 
-	trace_drv_stop(local);
-	local->ops->stop(&local->hw);
+	trace_drv_stop(local, suspend);
+	local->ops->stop(&local->hw, suspend);
 	trace_drv_return_void(local);
 
 	/* sync away all work on the tasklet before clearing started */
diff --git a/net/mac80211/driver-ops.h b/net/mac80211/driver-ops.h
index d4e73d3630e0..d382d9729e85 100644
--- a/net/mac80211/driver-ops.h
+++ b/net/mac80211/driver-ops.h
@@ -88,7 +88,7 @@ static inline int drv_get_et_sset_count(struct ieee80211_sub_if_data *sdata,
 }
 
 int drv_start(struct ieee80211_local *local);
-void drv_stop(struct ieee80211_local *local);
+void drv_stop(struct ieee80211_local *local, bool suspend);
 
 #ifdef CONFIG_PM
 static inline int drv_suspend(struct ieee80211_local *local,
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 501f02809135..6349552e62a8 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -2246,7 +2246,7 @@ int ieee80211_parse_ch_switch_ie(struct ieee80211_sub_if_data *sdata,
 
 /* Suspend/resume and hw reconfiguration */
 int ieee80211_reconfig(struct ieee80211_local *local);
-void ieee80211_stop_device(struct ieee80211_local *local);
+void ieee80211_stop_device(struct ieee80211_local *local, bool suspend);
 
 int __ieee80211_suspend(struct ieee80211_hw *hw,
 			struct cfg80211_wowlan *wowlan);
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index f06e165d6c7a..64106097949a 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -698,7 +698,7 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, bool going_do
 		wiphy_delayed_work_flush(local->hw.wiphy, &local->scan_work);
 
 	if (local->open_count == 0) {
-		ieee80211_stop_device(local);
+		ieee80211_stop_device(local, false);
 
 		/* no reconfiguring after stop! */
 		return;
@@ -1435,7 +1435,7 @@ int ieee80211_do_open(struct wireless_dev *wdev, bool coming_up)
 	drv_remove_interface(local, sdata);
  err_stop:
 	if (!local->open_count)
-		drv_stop(local);
+		drv_stop(local, false);
  err_del_bss:
 	sdata->bss = NULL;
 	if (sdata->vif.type == NL80211_IFTYPE_AP_VLAN)
diff --git a/net/mac80211/pm.c b/net/mac80211/pm.c
index c1fa26e09479..d823d58303e8 100644
--- a/net/mac80211/pm.c
+++ b/net/mac80211/pm.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /*
  * Portions
- * Copyright (C) 2020-2021, 2023 Intel Corporation
+ * Copyright (C) 2020-2021, 2023-2024 Intel Corporation
  */
 #include <net/mac80211.h>
 #include <net/rtnetlink.h>
@@ -171,7 +171,7 @@ int __ieee80211_suspend(struct ieee80211_hw *hw, struct cfg80211_wowlan *wowlan)
 	WARN_ON(!list_empty(&local->chanctx_list));
 
 	/* stop hardware - this must stop RX */
-	ieee80211_stop_device(local);
+	ieee80211_stop_device(local, true);
 
  suspend:
 	local->suspended = true;
diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h
index b26aacfbc622..dc498cd8cd91 100644
--- a/net/mac80211/trace.h
+++ b/net/mac80211/trace.h
@@ -328,9 +328,18 @@ TRACE_EVENT(drv_set_wakeup,
 	TP_printk(LOCAL_PR_FMT " enabled:%d", LOCAL_PR_ARG, __entry->enabled)
 );
 
-DEFINE_EVENT(local_only_evt, drv_stop,
-	TP_PROTO(struct ieee80211_local *local),
-	TP_ARGS(local)
+TRACE_EVENT(drv_stop,
+	TP_PROTO(struct ieee80211_local *local, bool suspend),
+	TP_ARGS(local, suspend),
+	TP_STRUCT__entry(
+		LOCAL_ENTRY
+		__field(bool, suspend)
+	),
+	TP_fast_assign(
+		LOCAL_ASSIGN;
+		__entry->suspend = suspend;
+	),
+	TP_printk(LOCAL_PR_FMT " suspend:%d", LOCAL_PR_ARG, __entry->suspend)
 );
 
 DEFINE_EVENT(local_sdata_addr_evt, drv_add_interface,
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index b3b8873a107b..610f0a94796c 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -1565,7 +1565,7 @@ u32 ieee80211_sta_get_rates(struct ieee80211_sub_if_data *sdata,
 	return supp_rates;
 }
 
-void ieee80211_stop_device(struct ieee80211_local *local)
+void ieee80211_stop_device(struct ieee80211_local *local, bool suspend)
 {
 	ieee80211_handle_queued_frames(local);
 
@@ -1576,7 +1576,7 @@ void ieee80211_stop_device(struct ieee80211_local *local)
 
 	flush_workqueue(local->workqueue);
 	wiphy_work_flush(local->hw.wiphy, NULL);
-	drv_stop(local);
+	drv_stop(local, suspend);
 }
 
 static void ieee80211_flush_completed_scan(struct ieee80211_local *local,
-- 
cgit v1.2.3-58-ga151


From d1621b008e3ad866426499fd351983a01784fa34 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Tue, 18 Jun 2024 20:03:01 +0300
Subject: wifi: iwlwifi: pcie: fix a few legacy register accesses for new
 devices

Do not access legacy bits for new devices, this has no effect.
Somehow, wowlan worked despite the usage of the wrong bits. Now
that we want to keep the firmware loaded during suspend even without
wowlan, this change is needed.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618200104.399d4d215210.Id12e7fdb7bab9f2c4c0d292519b5c1b4753a8c84@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-io.c     |  4 ++--
 drivers/net/wireless/intel/iwlwifi/pcie/trans.c | 22 +++++++++++++++++-----
 2 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-io.c b/drivers/net/wireless/intel/iwlwifi/iwl-io.c
index c60f9466c5fd..060becfd64f3 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-io.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-io.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /*
- * Copyright (C) 2003-2014, 2018-2022 Intel Corporation
+ * Copyright (C) 2003-2014, 2018-2022, 2024 Intel Corporation
  * Copyright (C) 2015-2016 Intel Deutschland GmbH
  */
 #include <linux/delay.h>
@@ -460,7 +460,7 @@ int iwl_finish_nic_init(struct iwl_trans *trans)
 	 */
 	if (cfg_trans->device_family >= IWL_DEVICE_FAMILY_BZ) {
 		iwl_set_bit(trans, CSR_GP_CNTRL,
-			    CSR_GP_CNTRL_REG_FLAG_MAC_CLOCK_READY |
+			    CSR_GP_CNTRL_REG_FLAG_BZ_MAC_ACCESS_REQ |
 			    CSR_GP_CNTRL_REG_FLAG_MAC_INIT);
 		poll_ready = CSR_GP_CNTRL_REG_FLAG_MAC_STATUS;
 	} else {
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index 94502e004719..719ddc4b72c5 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1504,9 +1504,17 @@ void iwl_pcie_d3_complete_suspend(struct iwl_trans *trans,
 
 	iwl_pcie_synchronize_irqs(trans);
 
-	iwl_clear_bit(trans, CSR_GP_CNTRL,
-		      CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
-	iwl_clear_bit(trans, CSR_GP_CNTRL, CSR_GP_CNTRL_REG_FLAG_INIT_DONE);
+	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ) {
+		iwl_clear_bit(trans, CSR_GP_CNTRL,
+			      CSR_GP_CNTRL_REG_FLAG_BZ_MAC_ACCESS_REQ);
+		iwl_clear_bit(trans, CSR_GP_CNTRL,
+			      CSR_GP_CNTRL_REG_FLAG_MAC_INIT);
+	} else {
+		iwl_clear_bit(trans, CSR_GP_CNTRL,
+			      CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
+		iwl_clear_bit(trans, CSR_GP_CNTRL,
+			      CSR_GP_CNTRL_REG_FLAG_INIT_DONE);
+	}
 
 	if (reset) {
 		/*
@@ -1584,8 +1592,12 @@ int iwl_trans_pcie_d3_resume(struct iwl_trans *trans,
 		goto out;
 	}
 
-	iwl_set_bit(trans, CSR_GP_CNTRL,
-		    CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
+	if (trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_BZ)
+		iwl_set_bit(trans, CSR_GP_CNTRL,
+			    CSR_GP_CNTRL_REG_FLAG_BZ_MAC_ACCESS_REQ);
+	else
+		iwl_set_bit(trans, CSR_GP_CNTRL,
+			    CSR_GP_CNTRL_REG_FLAG_MAC_ACCESS_REQ);
 
 	ret = iwl_finish_nic_init(trans);
 	if (ret)
-- 
cgit v1.2.3-58-ga151


From 5c38bedac16a946402b627621b3d4bfcc9952479 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 18 Jun 2024 20:03:02 +0300
Subject: wifi: iwlwifi: mvm: unify and fix interface combinations

AP interfaces fundamentally cannot leave the channel, so multi-
channel operation with them isn't really possible. We shouldn't
advertise support for such, at least not as long as we don't
have full multi-radio support. Thus, remove the AP bit from the
interface combinations for two channels and add another set for
just one channel that has it.

Also, to avoid duplicating everything even more, unify the NAN
and non-NAN cases.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Reviewed-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618200104.3213638262ef.I2a0031b37623d7763fd0c5405477ea7206a3e923@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 43 +++++++++++++++--------
 1 file changed, 28 insertions(+), 15 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 896e560f5a82..1c1f9816cf5e 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -30,21 +30,28 @@
 #include "iwl-nvm-parse.h"
 #include "time-sync.h"
 
+#define IWL_MVM_LIMITS(ap)					\
+	{							\
+		.max = 1,					\
+		.types = BIT(NL80211_IFTYPE_STATION),		\
+	},							\
+	{							\
+		.max = 1,					\
+		.types = ap |					\
+			 BIT(NL80211_IFTYPE_P2P_CLIENT) |	\
+			 BIT(NL80211_IFTYPE_P2P_GO),		\
+	},							\
+	{							\
+		.max = 1,					\
+		.types = BIT(NL80211_IFTYPE_P2P_DEVICE),	\
+	}
+
 static const struct ieee80211_iface_limit iwl_mvm_limits[] = {
-	{
-		.max = 1,
-		.types = BIT(NL80211_IFTYPE_STATION),
-	},
-	{
-		.max = 1,
-		.types = BIT(NL80211_IFTYPE_AP) |
-			BIT(NL80211_IFTYPE_P2P_CLIENT) |
-			BIT(NL80211_IFTYPE_P2P_GO),
-	},
-	{
-		.max = 1,
-		.types = BIT(NL80211_IFTYPE_P2P_DEVICE),
-	},
+	IWL_MVM_LIMITS(0)
+};
+
+static const struct ieee80211_iface_limit iwl_mvm_limits_ap[] = {
+	IWL_MVM_LIMITS(BIT(NL80211_IFTYPE_AP))
 };
 
 static const struct ieee80211_iface_combination iwl_mvm_iface_combinations[] = {
@@ -52,7 +59,13 @@ static const struct ieee80211_iface_combination iwl_mvm_iface_combinations[] = {
 		.num_different_channels = 2,
 		.max_interfaces = 3,
 		.limits = iwl_mvm_limits,
-		.n_limits = ARRAY_SIZE(iwl_mvm_limits),
+		.n_limits = ARRAY_SIZE(iwl_mvm_limits) - 1,
+	},
+	{
+		.num_different_channels = 1,
+		.max_interfaces = 3,
+		.limits = iwl_mvm_limits_ap,
+		.n_limits = ARRAY_SIZE(iwl_mvm_limits_ap) - 1,
 	},
 };
 
-- 
cgit v1.2.3-58-ga151


From 0091eda01412805e0d2c8025638ac0992102a7fc Mon Sep 17 00:00:00 2001
From: Daniel Gabay <daniel.gabay@intel.com>
Date: Tue, 18 Jun 2024 20:03:03 +0300
Subject: wifi: iwlwifi: fix iwl_mvm_get_valid_rx_ant()

Fix incorrect use of _tx_ valid ant data in the function.

Fixes: 4ea1ed1d14d8 ("wifi: iwlwifi: mvm: support set_antenna()")
Signed-off-by: Daniel Gabay <daniel.gabay@intel.com>
Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618200104.b7c6a320c7dc.I3092eb5275056f2162b9694e583c310c38568b2a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 54c3553db219..4ced1e51c908 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1875,10 +1875,10 @@ static inline u8 iwl_mvm_get_valid_tx_ant(struct iwl_mvm *mvm)
 
 static inline u8 iwl_mvm_get_valid_rx_ant(struct iwl_mvm *mvm)
 {
-	u8 rx_ant = mvm->fw->valid_tx_ant;
+	u8 rx_ant = mvm->fw->valid_rx_ant;
 
 	if (mvm->nvm_data && mvm->nvm_data->valid_rx_ant)
-		rx_ant &= mvm->nvm_data->valid_tx_ant;
+		rx_ant &= mvm->nvm_data->valid_rx_ant;
 
 	if (mvm->set_rx_ant)
 		rx_ant &= mvm->set_rx_ant;
-- 
cgit v1.2.3-58-ga151


From e8bb19c1d5906d2f31ae61853fea2a9abc71e951 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Tue, 18 Jun 2024 19:44:03 +0300
Subject: wifi: iwlwifi: support fast resume

This will allow to suspend / resume the system without resetting the
firmware. This will allow to reduce the resume time.
In case the fast_resume fails, stop the device and bring it up from
scratch.

Raise the timeout for the D3_END notification since in some iterations,
it took 240ms.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618194245.03b8d2801044.I613d17c712de7a0d611cde4e14f37ebbe0c3c964@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/d3.c       | 85 +++++++++++++++++++++--
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 40 +++++++++--
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      | 14 +++-
 drivers/net/wireless/intel/iwlwifi/mvm/tt.c       |  2 +-
 4 files changed, 130 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
index 54f4acbbd05b..b4d650583ac2 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/d3.c
@@ -2493,6 +2493,9 @@ static void iwl_mvm_parse_wowlan_info_notif(struct iwl_mvm *mvm,
 		return;
 	}
 
+	if (mvm->fast_resume)
+		return;
+
 	iwl_mvm_convert_key_counters_v5(status, &data->gtk[0].sc);
 	iwl_mvm_convert_gtk_v3(status, data->gtk);
 	iwl_mvm_convert_igtk(status, &data->igtk[0]);
@@ -3049,7 +3052,7 @@ static bool iwl_mvm_check_rt_status(struct iwl_mvm *mvm,
 	if (iwl_mvm_rt_status(mvm->trans,
 			      mvm->trans->dbg.lmac_error_event_table[0],
 			      &err_id)) {
-		if (err_id == RF_KILL_INDICATOR_FOR_WOWLAN) {
+		if (err_id == RF_KILL_INDICATOR_FOR_WOWLAN && vif) {
 			struct cfg80211_wowlan_wakeup wakeup = {
 				.rfkill_release = true,
 			};
@@ -3366,7 +3369,7 @@ static int iwl_mvm_resume_firmware(struct iwl_mvm *mvm, bool test)
 	return ret;
 }
 
-#define IWL_MVM_D3_NOTIF_TIMEOUT (HZ / 5)
+#define IWL_MVM_D3_NOTIF_TIMEOUT (HZ / 3)
 
 static int iwl_mvm_d3_notif_wait(struct iwl_mvm *mvm,
 				 struct iwl_d3_data *d3_data)
@@ -3377,12 +3380,22 @@ static int iwl_mvm_d3_notif_wait(struct iwl_mvm *mvm,
 		WIDE_ID(SCAN_GROUP, OFFLOAD_MATCH_INFO_NOTIF),
 		WIDE_ID(PROT_OFFLOAD_GROUP, D3_END_NOTIFICATION)
 	};
+	static const u16 d3_fast_resume_notif[] = {
+		WIDE_ID(PROT_OFFLOAD_GROUP, D3_END_NOTIFICATION)
+	};
 	struct iwl_notification_wait wait_d3_notif;
 	int ret;
 
-	iwl_init_notification_wait(&mvm->notif_wait, &wait_d3_notif,
-				   d3_resume_notif, ARRAY_SIZE(d3_resume_notif),
-				   iwl_mvm_wait_d3_notif, d3_data);
+	if (mvm->fast_resume)
+		iwl_init_notification_wait(&mvm->notif_wait, &wait_d3_notif,
+					   d3_fast_resume_notif,
+					   ARRAY_SIZE(d3_fast_resume_notif),
+					   iwl_mvm_wait_d3_notif, d3_data);
+	else
+		iwl_init_notification_wait(&mvm->notif_wait, &wait_d3_notif,
+					   d3_resume_notif,
+					   ARRAY_SIZE(d3_resume_notif),
+					   iwl_mvm_wait_d3_notif, d3_data);
 
 	ret = iwl_mvm_resume_firmware(mvm, d3_data->test);
 	if (ret) {
@@ -3567,6 +3580,68 @@ void iwl_mvm_set_wakeup(struct ieee80211_hw *hw, bool enabled)
 	device_set_wakeup_enable(mvm->trans->dev, enabled);
 }
 
+void iwl_mvm_fast_suspend(struct iwl_mvm *mvm)
+{
+	struct iwl_d3_manager_config d3_cfg_cmd_data = {};
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	IWL_DEBUG_WOWLAN(mvm, "Starting fast suspend flow\n");
+
+	mvm->fast_resume = true;
+	set_bit(IWL_MVM_STATUS_IN_D3, &mvm->status);
+
+	WARN_ON(iwl_mvm_power_update_device(mvm));
+	mvm->trans->system_pm_mode = IWL_PLAT_PM_MODE_D3;
+	ret = iwl_mvm_send_cmd_pdu(mvm, D3_CONFIG_CMD, CMD_SEND_IN_D3,
+				   sizeof(d3_cfg_cmd_data), &d3_cfg_cmd_data);
+	if (ret)
+		IWL_ERR(mvm,
+			"fast suspend: couldn't send D3_CONFIG_CMD %d\n", ret);
+
+	WARN_ON(iwl_mvm_power_update_mac(mvm));
+
+	ret = iwl_trans_d3_suspend(mvm->trans, false, false);
+	if (ret)
+		IWL_ERR(mvm, "fast suspend: trans_d3_suspend failed %d\n", ret);
+}
+
+int iwl_mvm_fast_resume(struct iwl_mvm *mvm)
+{
+	struct iwl_d3_data d3_data = {
+		.notif_expected =
+			IWL_D3_NOTIF_D3_END_NOTIF,
+	};
+	int ret;
+
+	lockdep_assert_held(&mvm->mutex);
+
+	IWL_DEBUG_WOWLAN(mvm, "Starting the fast resume flow\n");
+
+	mvm->last_reset_or_resume_time_jiffies = jiffies;
+	iwl_fw_dbg_read_d3_debug_data(&mvm->fwrt);
+
+	if (iwl_mvm_check_rt_status(mvm, NULL)) {
+		set_bit(STATUS_FW_ERROR, &mvm->trans->status);
+		iwl_mvm_dump_nic_error_log(mvm);
+		iwl_dbg_tlv_time_point(&mvm->fwrt,
+				       IWL_FW_INI_TIME_POINT_FW_ASSERT, NULL);
+		iwl_fw_dbg_collect_desc(&mvm->fwrt, &iwl_dump_desc_assert,
+					false, 0);
+		return -ENODEV;
+	}
+	ret = iwl_mvm_d3_notif_wait(mvm, &d3_data);
+	clear_bit(IWL_MVM_STATUS_IN_D3, &mvm->status);
+	mvm->trans->system_pm_mode = IWL_PLAT_PM_MODE_DISABLED;
+	mvm->fast_resume = false;
+
+	if (ret)
+		IWL_ERR(mvm, "Couldn't get the d3 notif %d\n", ret);
+
+	return ret;
+}
+
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 static int iwl_mvm_d3_test_open(struct inode *inode, struct file *file)
 {
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 1c1f9816cf5e..3cffd03c6013 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1222,6 +1222,7 @@ static void iwl_mvm_restart_cleanup(struct iwl_mvm *mvm)
 
 int __iwl_mvm_mac_start(struct iwl_mvm *mvm)
 {
+	bool fast_resume = false;
 	int ret;
 
 	lockdep_assert_held(&mvm->mutex);
@@ -1247,6 +1248,30 @@ int __iwl_mvm_mac_start(struct iwl_mvm *mvm)
 		mvm->nvm_data = NULL;
 	}
 
+#ifdef CONFIG_PM
+	/* fast_resume will be cleared by iwl_mvm_fast_resume */
+	fast_resume = mvm->fast_resume;
+
+	if (fast_resume) {
+		ret = iwl_mvm_fast_resume(mvm);
+		if (ret) {
+			iwl_mvm_stop_device(mvm);
+			/* iwl_mvm_up() will be called further down */
+		} else {
+			/*
+			 * We clear IWL_MVM_STATUS_FIRMWARE_RUNNING upon
+			 * mac_down() so that debugfs will stop honoring
+			 * requests after we flush all the workers.
+			 * Set the IWL_MVM_STATUS_FIRMWARE_RUNNING bit again
+			 * now that we are back. This is a bit abusing the
+			 * flag since the firmware wasn't really ever stopped,
+			 * but this still serves the purpose.
+			 */
+			set_bit(IWL_MVM_STATUS_FIRMWARE_RUNNING, &mvm->status);
+		}
+	}
+#endif /* CONFIG_PM */
+
 	if (test_bit(IWL_MVM_STATUS_HW_RESTART_REQUESTED, &mvm->status)) {
 		/*
 		 * Now convert the HW_RESTART_REQUESTED flag to IN_HW_RESTART
@@ -1257,7 +1282,10 @@ int __iwl_mvm_mac_start(struct iwl_mvm *mvm)
 		/* Clean up some internal and mac80211 state on restart */
 		iwl_mvm_restart_cleanup(mvm);
 	}
-	ret = iwl_mvm_up(mvm);
+
+	/* we also want to load the firmware if fast_resume failed */
+	if (!fast_resume || ret)
+		ret = iwl_mvm_up(mvm);
 
 	iwl_dbg_tlv_time_point(&mvm->fwrt, IWL_FW_INI_TIME_POINT_POST_INIT,
 			       NULL);
@@ -1340,7 +1368,7 @@ void iwl_mvm_mac_reconfig_complete(struct ieee80211_hw *hw,
 	}
 }
 
-void __iwl_mvm_mac_stop(struct iwl_mvm *mvm)
+void __iwl_mvm_mac_stop(struct iwl_mvm *mvm, bool suspend)
 {
 	lockdep_assert_held(&mvm->mutex);
 
@@ -1356,7 +1384,11 @@ void __iwl_mvm_mac_stop(struct iwl_mvm *mvm)
 	if (!iwl_mvm_has_new_station_api(mvm->fw))
 		iwl_mvm_rm_aux_sta(mvm);
 
-	iwl_mvm_stop_device(mvm);
+	if (suspend &&
+	    mvm->trans->trans_cfg->device_family >= IWL_DEVICE_FAMILY_AX210)
+		iwl_mvm_fast_suspend(mvm);
+	else
+		iwl_mvm_stop_device(mvm);
 
 	iwl_mvm_async_handlers_purge(mvm);
 	/* async_handlers_list is empty and will stay empty: HW is stopped */
@@ -1425,7 +1457,7 @@ void iwl_mvm_mac_stop(struct ieee80211_hw *hw, bool suspend)
 	iwl_mvm_mei_set_sw_rfkill_state(mvm);
 
 	mutex_lock(&mvm->mutex);
-	__iwl_mvm_mac_stop(mvm);
+	__iwl_mvm_mac_stop(mvm, suspend);
 	mutex_unlock(&mvm->mutex);
 
 	/*
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 4ced1e51c908..3692f62ee895 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1161,6 +1161,7 @@ struct iwl_mvm {
 	struct ieee80211_channel **nd_channels;
 	int n_nd_channels;
 	bool net_detect;
+	bool fast_resume;
 	u8 offload_tid;
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	bool d3_wake_sysassert;
@@ -1739,7 +1740,7 @@ struct iwl_rate_info {
 	u8 ieee;	/* MAC header:  IWL_RATE_6M_IEEE, etc. */
 };
 
-void __iwl_mvm_mac_stop(struct iwl_mvm *mvm);
+void __iwl_mvm_mac_stop(struct iwl_mvm *mvm, bool suspend);
 int __iwl_mvm_mac_start(struct iwl_mvm *mvm);
 
 /******************
@@ -2261,11 +2262,22 @@ extern const struct file_operations iwl_dbgfs_d3_test_ops;
 #ifdef CONFIG_PM
 void iwl_mvm_set_last_nonqos_seq(struct iwl_mvm *mvm,
 				 struct ieee80211_vif *vif);
+void iwl_mvm_fast_suspend(struct iwl_mvm *mvm);
+int iwl_mvm_fast_resume(struct iwl_mvm *mvm);
 #else
 static inline void
 iwl_mvm_set_last_nonqos_seq(struct iwl_mvm *mvm, struct ieee80211_vif *vif)
 {
 }
+
+static inline void iwl_mvm_fast_suspend(struct iwl_mvm *mvm)
+{
+}
+
+static inline int iwl_mvm_fast_resume(struct iwl_mvm *mvm)
+{
+	return 0;
+}
 #endif
 void iwl_mvm_set_wowlan_qos_seq(struct iwl_mvm_sta *mvm_ap_sta,
 				struct iwl_wowlan_config_cmd *cmd);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
index 45b431ffbc97..718184be19b3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tt.c
@@ -299,7 +299,7 @@ static void check_exit_ctkill(struct work_struct *work)
 
 	ret = iwl_mvm_get_temp(mvm, &temp);
 
-	__iwl_mvm_mac_stop(mvm);
+	__iwl_mvm_mac_stop(mvm, false);
 
 	if (ret)
 		goto reschedule;
-- 
cgit v1.2.3-58-ga151


From f77bd9491f7fad1f1e3d12ff25469f75d97cb1d4 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 18 Jun 2024 19:44:04 +0300
Subject: wifi: iwlwifi: mvm: remove IWL_MVM_PARSE_NVM

This debug is not in use anymore. Remove it.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://patch.msgid.link/20240618194245.bded22049fae.I980c4a941d769f93cf74bfc90a7d7d9fb384dea1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/constants.h |  1 -
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c        | 10 +---------
 2 files changed, 1 insertion(+), 10 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
index 1abe9e9fd46b..c4c1e67b9ac7 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/constants.h
@@ -56,7 +56,6 @@
 #define IWL_MVM_RS_80_20_FAR_RANGE_TWEAK	1
 #define IWL_MVM_TOF_IS_RESPONDER		0
 #define IWL_MVM_HW_CSUM_DISABLE			0
-#define IWL_MVM_PARSE_NVM			0
 #define IWL_MVM_ADWELL_ENABLE			1
 #define IWL_MVM_ADWELL_MAX_BUDGET		0
 #define IWL_MVM_TCM_LOAD_MEDIUM_THRESH		10 /* percentage */
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index b8a785032513..36ed08f20f80 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -686,14 +686,6 @@ static int iwl_run_unified_mvm_ucode(struct iwl_mvm *mvm)
 			goto error;
 	}
 
-	if (IWL_MVM_PARSE_NVM && !mvm->nvm_data) {
-		ret = iwl_nvm_init(mvm);
-		if (ret) {
-			IWL_ERR(mvm, "Failed to read NVM: %d\n", ret);
-			goto error;
-		}
-	}
-
 	ret = iwl_mvm_send_cmd_pdu(mvm, WIDE_ID(REGULATORY_AND_NVM_GROUP,
 						NVM_ACCESS_COMPLETE),
 				   CMD_SEND_IN_RFKILL,
@@ -718,7 +710,7 @@ static int iwl_run_unified_mvm_ucode(struct iwl_mvm *mvm)
 		return ret;
 
 	/* Read the NVM only at driver load time, no need to do this twice */
-	if (!IWL_MVM_PARSE_NVM && !mvm->nvm_data) {
+	if (!mvm->nvm_data) {
 		mvm->nvm_data = iwl_get_nvm(mvm->trans, mvm->fw,
 					    mvm->set_tx_ant, mvm->set_rx_ant);
 		if (IS_ERR(mvm->nvm_data)) {
-- 
cgit v1.2.3-58-ga151


From 2917d04bce3906750b9b746736c54cc68740723e Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 18 Jun 2024 19:44:05 +0300
Subject: wifi: iwlwifi: trans: remove unused function parameter

iwl_trans_pcie_gen2_fw_alive doesn't use the scd_addr parameter,
it was there only because we needed the functio to have a prototype same
as iwl_trans_ops::fw_alive callback.
But now the ops is removed so no reason to keep the parameter.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Link: https://patch.msgid.link/20240618194245.1aa8bf13aea9.I9662c10c1db545dd8849af4bb4ab47708d4548d8@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-trans.c       | 2 +-
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h   | 2 +-
 drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.c b/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
index eeee5c35668e..bd21bbe7e7e3 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
@@ -359,7 +359,7 @@ void iwl_trans_fw_alive(struct iwl_trans *trans, u32 scd_addr)
 	trans->state = IWL_TRANS_FW_ALIVE;
 
 	if (trans->trans_cfg->gen2)
-		iwl_trans_pcie_gen2_fw_alive(trans, scd_addr);
+		iwl_trans_pcie_gen2_fw_alive(trans);
 	else
 		iwl_trans_pcie_fw_alive(trans, scd_addr);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index 3ce626c93651..2040e00c744a 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -1095,7 +1095,7 @@ void iwl_pcie_alloc_fw_monitor(struct iwl_trans *trans, u8 max_power);
 /* transport gen 2 exported functions */
 int iwl_trans_pcie_gen2_start_fw(struct iwl_trans *trans,
 				 const struct fw_img *fw, bool run_in_rfkill);
-void iwl_trans_pcie_gen2_fw_alive(struct iwl_trans *trans, u32 scd_addr);
+void iwl_trans_pcie_gen2_fw_alive(struct iwl_trans *trans);
 int iwl_trans_pcie_gen2_send_hcmd(struct iwl_trans *trans,
 				  struct iwl_host_cmd *cmd);
 void iwl_trans_pcie_gen2_stop_device(struct iwl_trans *trans);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
index c1446d550b4f..18dda89b7985 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c
@@ -339,7 +339,7 @@ static void iwl_pcie_get_rf_name(struct iwl_trans *trans)
 	pos += scnprintf(buf + pos, buflen - pos, "\n");
 }
 
-void iwl_trans_pcie_gen2_fw_alive(struct iwl_trans *trans, u32 scd_addr)
+void iwl_trans_pcie_gen2_fw_alive(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
-- 
cgit v1.2.3-58-ga151


From 74943017afdddcf682b1a162781b872d2188c5e6 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 18 Jun 2024 19:44:06 +0300
Subject: wifi: iwlwifi: bump FW API to 92 for BZ/SC devices

Start supporting API version 92 for new devices.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618194245.433cfbb6f042.I914da5bd0e2ed25148726f9fb55e7a60b895edfd@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/cfg/bz.c | 2 +-
 drivers/net/wireless/intel/iwlwifi/cfg/sc.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c
index e3d3abfc5fe4..3b6b8b410be5 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/bz.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/bz.c
@@ -10,7 +10,7 @@
 #include "fw/api/txq.h"
 
 /* Highest firmware API version supported */
-#define IWL_BZ_UCODE_API_MAX	91
+#define IWL_BZ_UCODE_API_MAX	92
 
 /* Lowest firmware API version supported */
 #define IWL_BZ_UCODE_API_MIN	90
diff --git a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
index 7ad3322a3305..4ccb0b7bdc20 100644
--- a/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
+++ b/drivers/net/wireless/intel/iwlwifi/cfg/sc.c
@@ -10,7 +10,7 @@
 #include "fw/api/txq.h"
 
 /* Highest firmware API version supported */
-#define IWL_SC_UCODE_API_MAX	91
+#define IWL_SC_UCODE_API_MAX	92
 
 /* Lowest firmware API version supported */
 #define IWL_SC_UCODE_API_MIN	90
-- 
cgit v1.2.3-58-ga151


From 2b94a6f792b0844a98c187eb35e090ce16bf5810 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Tue, 18 Jun 2024 19:44:07 +0300
Subject: wifi: iwlwifi: don't assume op_mode_nic_config exists

The new op_mode won't have this callback.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://patch.msgid.link/20240618194245.718983e7b832.I02fb130aa3d2044958cba7dced357031b92e33bb@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-op-mode.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-op-mode.h b/drivers/net/wireless/intel/iwlwifi/iwl-op-mode.h
index 1ca82f3e4ebf..595fa6ddf084 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-op-mode.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-op-mode.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2005-2014, 2018-2021 Intel Corporation
+ * Copyright (C) 2005-2014, 2018-2021, 2024 Intel Corporation
  * Copyright (C) 2013-2014 Intel Mobile Communications GmbH
  * Copyright (C) 2015 Intel Deutschland GmbH
  */
@@ -185,7 +185,8 @@ static inline void iwl_op_mode_cmd_queue_full(struct iwl_op_mode *op_mode)
 static inline void iwl_op_mode_nic_config(struct iwl_op_mode *op_mode)
 {
 	might_sleep();
-	op_mode->ops->nic_config(op_mode);
+	if (op_mode->ops->nic_config)
+		op_mode->ops->nic_config(op_mode);
 }
 
 static inline void iwl_op_mode_wimax_active(struct iwl_op_mode *op_mode)
-- 
cgit v1.2.3-58-ga151


From bffa3d04e4ab90dcd3ae74642977132df135417b Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Tue, 18 Jun 2024 19:44:08 +0300
Subject: wifi: iwlwifi: trans: remove unused status bits

Those bits are never used.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618194245.cd26318d9513.I035ec7f3f022cdc19a650eca3b89455e3b5eca18@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index 006eb51a705f..ac8f8ff5a99d 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -291,9 +291,6 @@ enum iwl_d3_status {
  * @STATUS_RFKILL_HW: the actual HW state of the RF-kill switch
  * @STATUS_RFKILL_OPMODE: RF-kill state reported to opmode
  * @STATUS_FW_ERROR: the fw is in error state
- * @STATUS_TRANS_GOING_IDLE: shutting down the trans, only special commands
- *	are sent
- * @STATUS_TRANS_IDLE: the trans is idle - general commands are not to be sent
  * @STATUS_TRANS_DEAD: trans is dead - avoid any read/write operation
  * @STATUS_SUPPRESS_CMD_ERROR_ONCE: suppress "FW error in SYNC CMD" once,
  *	e.g. for testing
@@ -306,8 +303,6 @@ enum iwl_trans_status {
 	STATUS_RFKILL_HW,
 	STATUS_RFKILL_OPMODE,
 	STATUS_FW_ERROR,
-	STATUS_TRANS_GOING_IDLE,
-	STATUS_TRANS_IDLE,
 	STATUS_TRANS_DEAD,
 	STATUS_SUPPRESS_CMD_ERROR_ONCE,
 };
-- 
cgit v1.2.3-58-ga151


From 7714a40561c3905764140cc3e971469671b07187 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 18 Jun 2024 19:44:09 +0300
Subject: wifi: iwlwifi: mvm: dissolve iwl_mvm_mac_remove_interface_common()

Since the function isn't actually common (any more), just
dissolve it.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618194245.adf54ab5bc4a.I90339f152bba73b622e05063d16ae914ae20053f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 21 +++------------------
 1 file changed, 3 insertions(+), 18 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 3cffd03c6013..4f16ada07e2f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -1899,12 +1899,8 @@ void iwl_mvm_prepare_mac_removal(struct iwl_mvm *mvm,
 	cancel_delayed_work_sync(&mvmvif->csa_work);
 }
 
-/* This function is doing the common part of removing the interface for
- * both - MLD and non-MLD modes. Returns true if removing the interface
- * is done
- */
-static bool iwl_mvm_mac_remove_interface_common(struct ieee80211_hw *hw,
-						struct ieee80211_vif *vif)
+static void iwl_mvm_mac_remove_interface(struct ieee80211_hw *hw,
+					 struct ieee80211_vif *vif)
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
@@ -1952,21 +1948,10 @@ static bool iwl_mvm_mac_remove_interface_common(struct ieee80211_hw *hw,
 			mvm->noa_duration = 0;
 		}
 #endif
-		return true;
+		goto out;
 	}
 
 	iwl_mvm_power_update_mac(mvm);
-	return false;
-}
-
-static void iwl_mvm_mac_remove_interface(struct ieee80211_hw *hw,
-					 struct ieee80211_vif *vif)
-{
-	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
-	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-
-	if (iwl_mvm_mac_remove_interface_common(hw, vif))
-		goto out;
 
 	/* Before the interface removal, mac80211 would cancel the ROC, and the
 	 * ROC worker would be scheduled if needed. The worker would be flushed
-- 
cgit v1.2.3-58-ga151


From a02079af2491b076a4bb6b1a5d668621efbb4b4e Mon Sep 17 00:00:00 2001
From: Ilan Peer <ilan.peer@intel.com>
Date: Tue, 18 Jun 2024 19:44:10 +0300
Subject: wifi: iwlwifi: mvm: Fix associated initiator key setting

When setting the keys for secure measurement with an AP the station
is associated with, the TK should only be referenced and not copied.

In addition set the cipher only when the correct station is found.

Fixes: 626be4bf99f6 ("wifi: iwlwifi: mvm: modify iwl_mvm_ftm_set_secured_ranging() parameters")
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618194245.be2a5327554d.Ie53220b075dacb23a8d073f6008aafd339368592@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
index 00d447e40cc6..afd90a52d4ec 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c
@@ -732,10 +732,7 @@ static void iter(struct ieee80211_hw *hw,
 
 	WARN_ON(!sta->mfp);
 
-	if (WARN_ON(key->keylen > sizeof(target->tk)))
-		return;
-
-	memcpy(target->tk, key->key, key->keylen);
+	target->tk = key->key;
 	*target->cipher = iwl_mvm_cipher_to_location_cipher(key->cipher);
 	WARN_ON(*target->cipher == IWL_LOCATION_CIPHER_INVALID);
 }
@@ -774,9 +771,7 @@ iwl_mvm_ftm_set_secured_ranging(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 		    !memcmp(vif->bss_conf.bssid, bssid, ETH_ALEN)) {
 			struct iwl_mvm_ftm_iter_data target;
 
-			target.cipher = cipher;
 			target.bssid = bssid;
-			target.tk = tk;
 			ieee80211_iter_keys(mvm->hw, vif, iter, &target);
 		} else {
 			memcpy(tk, entry->tk, sizeof(entry->tk));
-- 
cgit v1.2.3-58-ga151


From 67c638675dcdad588aeb71e67dde77b48a53a320 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 18 Jun 2024 19:44:11 +0300
Subject: wifi: iwlwifi: mvm: rename 'pldr_sync'

PLDR (product level device reset) is a Windows term, and
is something the driver triggers there, AFAICT.

Really what 'pldr_sync' here wants to capture is whether
or not the firmware will/may do a product reset during
initialization, which makes the device drop off the bus,
requiring a rescan. If this is the case, obviously the
init will fail/time out, so we don't want to report all
kinds of errors etc., hence this tracking variable.

Rename it to 'fw_product_reset' to capture the meaning
better.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618194245.ccf849642af8.I01dded6b2393771b7baf8b4b17336784d987c7c2@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c  | 12 ++++++------
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h |  7 ++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c |  2 +-
 3 files changed, 13 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 36ed08f20f80..0a8bd957bd6c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -408,7 +408,7 @@ static int iwl_mvm_load_ucode_wait_alive(struct iwl_mvm *mvm,
 						UREG_LMAC2_CURRENT_PC));
 		}
 
-		if (ret == -ETIMEDOUT && !mvm->pldr_sync)
+		if (ret == -ETIMEDOUT && !mvm->fw_product_reset)
 			iwl_fw_dbg_error_collect(&mvm->fwrt,
 						 FW_DBG_TRIGGER_ALIVE_TIMEOUT);
 
@@ -627,8 +627,8 @@ static int iwl_run_unified_mvm_ucode(struct iwl_mvm *mvm)
 	if (mvm->trans->trans_cfg->device_family == IWL_DEVICE_FAMILY_AX210) {
 		sb_cfg = iwl_read_umac_prph(mvm->trans, SB_MODIFY_CFG_FLAG);
 		/* if needed, we'll reset this on our way out later */
-		mvm->pldr_sync = sb_cfg == SB_CFG_RESIDES_IN_ROM;
-		if (mvm->pldr_sync && iwl_mei_pldr_req())
+		mvm->fw_product_reset = sb_cfg == SB_CFG_RESIDES_IN_ROM;
+		if (mvm->fw_product_reset && iwl_mei_pldr_req())
 			return -EBUSY;
 	}
 
@@ -647,7 +647,7 @@ static int iwl_run_unified_mvm_ucode(struct iwl_mvm *mvm)
 		IWL_ERR(mvm, "Failed to start RT ucode: %d\n", ret);
 
 		/* if we needed reset then fail here, but notify and remove */
-		if (mvm->pldr_sync) {
+		if (mvm->fw_product_reset) {
 			iwl_mei_alive_notif(false);
 			iwl_trans_pcie_remove(mvm->trans, true);
 		}
@@ -1407,14 +1407,14 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 	ret = iwl_mvm_load_rt_fw(mvm);
 	if (ret) {
 		IWL_ERR(mvm, "Failed to start RT ucode: %d\n", ret);
-		if (ret != -ERFKILL && !mvm->pldr_sync)
+		if (ret != -ERFKILL && !mvm->fw_product_reset)
 			iwl_fw_dbg_error_collect(&mvm->fwrt,
 						 FW_DBG_TRIGGER_DRIVER);
 		goto error;
 	}
 
 	/* FW loaded successfully */
-	mvm->pldr_sync = false;
+	mvm->fw_product_reset = false;
 
 	iwl_fw_disable_dbg_asserts(&mvm->fwrt);
 	iwl_get_shared_mem_conf(&mvm->fwrt);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 3692f62ee895..5bd388881092 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1325,7 +1325,12 @@ struct iwl_mvm {
 	bool sta_remove_requires_queue_remove;
 	bool mld_api_is_used;
 
-	bool pldr_sync;
+	/*
+	 * Indicates that firmware will do a product reset (and then
+	 * therefore fail to load) when we start it (due to OTP burn),
+	 * if so don't dump errors etc. since this is expected.
+	 */
+	bool fw_product_reset;
 
 	struct iwl_time_sync_data time_sync;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index f8a4fefede8d..008fe5fece87 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -875,7 +875,7 @@ get_nvm_from_fw:
 		IWL_ERR(mvm, "Failed to run INIT ucode: %d\n", ret);
 
 	/* no longer need this regardless of failure or not */
-	mvm->pldr_sync = false;
+	mvm->fw_product_reset = false;
 
 	return ret;
 }
-- 
cgit v1.2.3-58-ga151


From 3591e3b9a7cae0f412665f3f5f5f53a29a950d87 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 18 Jun 2024 19:44:12 +0300
Subject: wifi: iwlwifi: mei: clarify iwl_mei_pldr_req() docs

This isn't related to whether or not "fw can be loaded",
but rather requesting that ME go into a state where doing
a product reset is safe. This is related to FW load only
in the specific case of where it's used today in iwlmvm,
notably when it's known that the firmware itself will (or
at least may) do a product reset during load.

Clarify the documentation.

I was tempted to rename things too, but on the ME side it
really is also called PLDR (which is a Windows term and
may not even match the complete behaviour since doing a
full product reset from the driver also requires calling
an ACPI method first.) So keep the name aligned with ME.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618194245.ca2c55121a04.I889cd47210367ca9110411472ee696b796a37ab5@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h b/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h
index ee48b86674a6..4900de3cc0d3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h
+++ b/drivers/net/wireless/intel/iwlwifi/mei/iwl-mei.h
@@ -456,8 +456,11 @@ void iwl_mei_device_state(bool up);
 /**
  * iwl_mei_pldr_req() - must be called before loading the fw
  *
- * Return: 0 if the PLDR flow was successful and the fw can be loaded, negative
- *	value otherwise.
+ * Requests from the ME that it releases its potential bus access to
+ * the WiFi NIC so that the device can safely undergo product reset.
+ *
+ * Return: 0 if the request was successful and the device can be
+ *	reset, a negative error value otherwise
  */
 int iwl_mei_pldr_req(void);
 
-- 
cgit v1.2.3-58-ga151


From a9056a3716c5f77be845475e184daabaa771054c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 18 Jun 2024 19:44:13 +0300
Subject: wifi: iwlwifi: mvm: enable VLP AP on VLP channels

If channels are marked VLP, then we're actually also
allowed to be VLP AP/GO. Enable this.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618194245.e15f24fc6bc8.I33ed7d141fec731e79370ba6c7cfbe28776944a0@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index 712dabb30acc..c2656dee5a86 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -370,7 +370,9 @@ static u32 iwl_get_channel_flags(u8 ch_num, int ch_idx, enum nl80211_band band,
 		flags |= IEEE80211_CHAN_IR_CONCURRENT;
 
 	/* Set the AP type for the UHB case. */
-	if (!(nvm_flags & NVM_CHANNEL_VLP))
+	if (nvm_flags & NVM_CHANNEL_VLP)
+		flags |= IEEE80211_CHAN_ALLOW_6GHZ_VLP_AP;
+	else
 		flags |= IEEE80211_CHAN_NO_6GHZ_VLP_CLIENT;
 	if (!(nvm_flags & NVM_CHANNEL_AFC))
 		flags |= IEEE80211_CHAN_NO_6GHZ_AFC_CLIENT;
@@ -1661,7 +1663,9 @@ static u32 iwl_nvm_get_regdom_bw_flags(const u16 *nvm_chan,
 
 	/* Set the AP type for the UHB case. */
 	if (uats_enabled) {
-		if (!(nvm_flags & NVM_CHANNEL_VLP))
+		if (nvm_flags & NVM_CHANNEL_VLP)
+			flags |= NL80211_RRF_ALLOW_6GHZ_VLP_AP;
+		else
 			flags |= NL80211_RRF_NO_6GHZ_VLP_CLIENT;
 
 		if (!(nvm_flags & NVM_CHANNEL_AFC))
-- 
cgit v1.2.3-58-ga151


From f9802d531835b7591a9ddf99996e0f3ae59093f0 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 18 Jun 2024 19:59:45 +0300
Subject: wifi: iwlwifi: mvm: don't limit VLP/AFC to UATS-enabled

When UATS isn't enabled (no VLP/AFC AP support), we need to still
set the right bits in the channel/regulatory flags, so remove the
uats_enabled argument to the parsing etc.

Also, firmware deals just fine with getting the UATS table if it
supports the command even if the bits aren't set, so always send
it, since it's also needed if BIT(31) is set, but the driver need
not have any knowledge of that. Remove 'uats_enabled' entirely.

Fixes: 0d2fc8821a7d ("wifi: iwlwifi: nvm: parse the VLP/AFC bit from regulatory")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240618195731.a81e7234c4f6.Ic0131180d38e0f1ead2f7fa0e7583407ceaa0bd1@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/runtime.h    |  2 --
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c | 21 +++++++++------------
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h |  4 ++--
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c        | 15 +--------------
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  |  3 +--
 5 files changed, 13 insertions(+), 32 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
index 3fcc08b3c0db..048877fa7c71 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/runtime.h
@@ -103,7 +103,6 @@ struct iwl_txf_iter_data {
  * @cur_fw_img: current firmware image, must be maintained by
  *	the driver by calling &iwl_fw_set_current_image()
  * @dump: debug dump data
- * @uats_enabled: VLP or AFC AP is enabled
  * @uats_table: AP type table
  * @uefi_tables_lock_status: The status of the WIFI GUID UEFI variables lock:
  *	0: Unlocked, 1 and 2: Locked.
@@ -183,7 +182,6 @@ struct iwl_fw_runtime {
 	bool sgom_enabled;
 	struct iwl_mcc_allowed_ap_type_cmd uats_table;
 	u8 uefi_tables_lock_status;
-	bool uats_enabled;
 };
 
 void iwl_fw_runtime_init(struct iwl_fw_runtime *fwrt, struct iwl_trans *trans,
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
index c2656dee5a86..d902121da009 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
@@ -1610,8 +1610,7 @@ IWL_EXPORT_SYMBOL(iwl_parse_nvm_data);
 static u32 iwl_nvm_get_regdom_bw_flags(const u16 *nvm_chan,
 				       int ch_idx, u16 nvm_flags,
 				       struct iwl_reg_capa reg_capa,
-				       const struct iwl_cfg *cfg,
-				       bool uats_enabled)
+				       const struct iwl_cfg *cfg)
 {
 	u32 flags = NL80211_RRF_NO_HT40;
 
@@ -1662,15 +1661,13 @@ static u32 iwl_nvm_get_regdom_bw_flags(const u16 *nvm_chan,
 	}
 
 	/* Set the AP type for the UHB case. */
-	if (uats_enabled) {
-		if (nvm_flags & NVM_CHANNEL_VLP)
-			flags |= NL80211_RRF_ALLOW_6GHZ_VLP_AP;
-		else
-			flags |= NL80211_RRF_NO_6GHZ_VLP_CLIENT;
+	if (nvm_flags & NVM_CHANNEL_VLP)
+		flags |= NL80211_RRF_ALLOW_6GHZ_VLP_AP;
+	else
+		flags |= NL80211_RRF_NO_6GHZ_VLP_CLIENT;
 
-		if (!(nvm_flags & NVM_CHANNEL_AFC))
-			flags |= NL80211_RRF_NO_6GHZ_AFC_CLIENT;
-	}
+	if (!(nvm_flags & NVM_CHANNEL_AFC))
+		flags |= NL80211_RRF_NO_6GHZ_AFC_CLIENT;
 
 	/*
 	 * reg_capa is per regulatory domain so apply it for every channel
@@ -1726,7 +1723,7 @@ static struct iwl_reg_capa iwl_get_reg_capa(u32 flags, u8 resp_ver)
 struct ieee80211_regdomain *
 iwl_parse_nvm_mcc_info(struct device *dev, const struct iwl_cfg *cfg,
 		       int num_of_ch, __le32 *channels, u16 fw_mcc,
-		       u16 geo_info, u32 cap, u8 resp_ver, bool uats_enabled)
+		       u16 geo_info, u32 cap, u8 resp_ver)
 {
 	int ch_idx;
 	u16 ch_flags;
@@ -1793,7 +1790,7 @@ iwl_parse_nvm_mcc_info(struct device *dev, const struct iwl_cfg *cfg,
 
 		reg_rule_flags = iwl_nvm_get_regdom_bw_flags(nvm_chan, ch_idx,
 							     ch_flags, reg_capa,
-							     cfg, uats_enabled);
+							     cfg);
 
 		/* we can't continue the same rule */
 		if (ch_idx == 0 || prev_reg_rule_flags != reg_rule_flags ||
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
index 5a1b5ab62de1..fd82feb24b2d 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2005-2015, 2018-2023 Intel Corporation
+ * Copyright (C) 2005-2015, 2018-2024 Intel Corporation
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
 #ifndef __iwl_nvm_parse_h__
@@ -50,7 +50,7 @@ iwl_parse_nvm_data(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 struct ieee80211_regdomain *
 iwl_parse_nvm_mcc_info(struct device *dev, const struct iwl_cfg *cfg,
 		       int num_of_ch, __le32 *channels, u16 fw_mcc,
-		       u16 geo_info, u32 cap, u8 resp_ver, bool uats_enabled);
+		       u16 geo_info, u32 cap, u8 resp_ver);
 
 /**
  * struct iwl_nvm_section - describes an NVM section in memory.
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 0a8bd957bd6c..ef5acecd11fc 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -28,9 +28,6 @@
 #define MVM_UCODE_ALIVE_TIMEOUT	(2 * HZ)
 #define MVM_UCODE_CALIB_TIMEOUT	(2 * HZ)
 
-#define IWL_UATS_VLP_AP_SUPPORTED BIT(29)
-#define IWL_UATS_AFC_AP_SUPPORTED BIT(30)
-
 struct iwl_mvm_alive_data {
 	bool valid;
 	u32 scd_base_addr;
@@ -491,17 +488,11 @@ static void iwl_mvm_uats_init(struct iwl_mvm *mvm)
 		.dataflags[0] = IWL_HCMD_DFL_NOCOPY,
 	};
 
-	if (!(mvm->trans->trans_cfg->device_family >=
-	      IWL_DEVICE_FAMILY_AX210)) {
+	if (mvm->trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_AX210) {
 		IWL_DEBUG_RADIO(mvm, "UATS feature is not supported\n");
 		return;
 	}
 
-	if (!mvm->fwrt.uats_enabled) {
-		IWL_DEBUG_RADIO(mvm, "UATS feature is disabled\n");
-		return;
-	}
-
 	cmd_ver = iwl_fw_lookup_cmd_ver(mvm->fw, cmd.id,
 					IWL_FW_CMD_VER_UNKNOWN);
 	if (cmd_ver != 1) {
@@ -1223,10 +1214,6 @@ static void iwl_mvm_lari_cfg(struct iwl_mvm *mvm)
 					"Failed to send LARI_CONFIG_CHANGE (%d)\n",
 					ret);
 	}
-
-	if (le32_to_cpu(cmd.oem_uhb_allow_bitmap) & IWL_UATS_VLP_AP_SUPPORTED ||
-	    le32_to_cpu(cmd.oem_uhb_allow_bitmap) & IWL_UATS_AFC_AP_SUPPORTED)
-		mvm->fwrt.uats_enabled = true;
 }
 
 void iwl_mvm_get_bios_tables(struct iwl_mvm *mvm)
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 4f16ada07e2f..8c732becd281 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -151,8 +151,7 @@ struct ieee80211_regdomain *iwl_mvm_get_regdomain(struct wiphy *wiphy,
 				      resp->channels,
 				      __le16_to_cpu(resp->mcc),
 				      __le16_to_cpu(resp->geo_info),
-				      le32_to_cpu(resp->cap), resp_ver,
-				      mvm->fwrt.uats_enabled);
+				      le32_to_cpu(resp->cap), resp_ver);
 	/* Store the return source id */
 	src_id = resp->source_id;
 	if (IS_ERR_OR_NULL(regd)) {
-- 
cgit v1.2.3-58-ga151


From 6adae0b081454393ca5f7363fd3c7379c8e2a7a1 Mon Sep 17 00:00:00 2001
From: Golan Ben Ami <golan.ben.ami@intel.com>
Date: Thu, 13 Jun 2024 17:11:23 +0300
Subject: wifi: iwlwifi: remove AX101, AX201 and AX203 support from LNL

LNL is the codename for the upcoming Series 2 Core Ultra
processors designed by Intel. AX101, AX201 and AX203 devices
are not shiped on LNL platforms, so don't support them.

Signed-off-by: Golan Ben Ami <golan.ben.ami@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240613171043.f24a228dfd96.I989a2d3f1513211bc49ac8143ee4e9e341e1ee67@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 32 ++++++++++++++++++++++++++-
 1 file changed, 31 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
index d6da25e24818..9ad43464b702 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c
@@ -503,7 +503,37 @@ VISIBLE_IF_IWLWIFI_KUNIT const struct pci_device_id iwl_hw_card_ids[] = {
 	{IWL_PCI_DEVICE(0x2727, PCI_ANY_ID, iwl_bz_trans_cfg)},
 	{IWL_PCI_DEVICE(0x272D, PCI_ANY_ID, iwl_bz_trans_cfg)},
 	{IWL_PCI_DEVICE(0x272b, PCI_ANY_ID, iwl_bz_trans_cfg)},
-	{IWL_PCI_DEVICE(0xA840, PCI_ANY_ID, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0000, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0090, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0094, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0098, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x009C, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x00C0, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x00C4, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x00E0, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x00E4, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x00E8, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x00EC, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0100, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0110, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0114, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0118, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x011C, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0310, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0314, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0510, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x0A10, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x1671, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x1672, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x1771, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x1772, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x1791, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x1792, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x4090, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x40C4, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x40E0, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x4110, iwl_bz_trans_cfg)},
+	{IWL_PCI_DEVICE(0xA840, 0x4314, iwl_bz_trans_cfg)},
 	{IWL_PCI_DEVICE(0x7740, PCI_ANY_ID, iwl_bz_trans_cfg)},
 	{IWL_PCI_DEVICE(0x4D40, PCI_ANY_ID, iwl_bz_trans_cfg)},
 
-- 
cgit v1.2.3-58-ga151


From 8a4475a15b373bb5fcaf5ccd24f0f08217fa6d34 Mon Sep 17 00:00:00 2001
From: Daniel Gabay <daniel.gabay@intel.com>
Date: Tue, 25 Jun 2024 19:51:07 +0300
Subject: wifi: iwlwifi: remove MVM prefix from scan API

These are not MVM specific.

Signed-off-by: Daniel Gabay <daniel.gabay@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.09f672318944.Idffeab6a4dfc12effebd1c50815ae5c540afca74@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/scan.h  | 6 +++---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 8 ++++----
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      | 2 +-
 3 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
index 6684506f4fc4..c00f9422306d 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
@@ -443,10 +443,10 @@ struct iwl_periodic_scan_complete {
 /* UMAC Scan API */
 
 /* The maximum of either of these cannot exceed 8, because we use an
- * 8-bit mask (see IWL_MVM_SCAN_MASK in mvm.h).
+ * 8-bit mask (see enum iwl_scan_status).
  */
-#define IWL_MVM_MAX_UMAC_SCANS 4
-#define IWL_MVM_MAX_LMAC_SCANS 1
+#define IWL_MAX_UMAC_SCANS 4
+#define IWL_MAX_LMAC_SCANS 1
 
 enum scan_config_flags {
 	SCAN_CONFIG_FLAG_ACTIVATE			= BIT(0),
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 8c732becd281..d29531d95353 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -591,13 +591,13 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 	hw->wiphy->max_scan_ssids = PROBE_OPTION_MAX;
 
 	BUILD_BUG_ON(IWL_MVM_SCAN_STOPPING_MASK & IWL_MVM_SCAN_MASK);
-	BUILD_BUG_ON(IWL_MVM_MAX_UMAC_SCANS > HWEIGHT32(IWL_MVM_SCAN_MASK) ||
-		     IWL_MVM_MAX_LMAC_SCANS > HWEIGHT32(IWL_MVM_SCAN_MASK));
+	BUILD_BUG_ON(IWL_MAX_UMAC_SCANS > HWEIGHT32(IWL_MVM_SCAN_MASK) ||
+		     IWL_MAX_LMAC_SCANS > HWEIGHT32(IWL_MVM_SCAN_MASK));
 
 	if (fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_UMAC_SCAN))
-		mvm->max_scans = IWL_MVM_MAX_UMAC_SCANS;
+		mvm->max_scans = IWL_MAX_UMAC_SCANS;
 	else
-		mvm->max_scans = IWL_MVM_MAX_LMAC_SCANS;
+		mvm->max_scans = IWL_MAX_LMAC_SCANS;
 
 	if (mvm->nvm_data->bands[NL80211_BAND_2GHZ].n_channels)
 		hw->wiphy->bands[NL80211_BAND_2GHZ] =
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 5bd388881092..e51e834902a9 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1071,7 +1071,7 @@ struct iwl_mvm {
 	unsigned int max_scans;
 
 	/* UMAC scan tracking */
-	u32 scan_uid_status[IWL_MVM_MAX_UMAC_SCANS];
+	u32 scan_uid_status[IWL_MAX_UMAC_SCANS];
 
 	/* start time of last scan in TSF of the mac that requested the scan */
 	u64 scan_start;
-- 
cgit v1.2.3-58-ga151


From f9414f8067015b1db14d374a33fbafc8395eb958 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:08 +0300
Subject: wifi: iwlwifi: mvm: don't flush BSSes on restart with MLD API

If the firmware has MLD APIs, it will handle all timing and we
don't need to give it timestamps. Therefore, we don't care about
the timestamps stored in the BSS table, so there's no need to
flush the BSS table.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.c6d86dc2377e.I246d0fae0d23ed34b7cd9c3400edb004eb5ac1d0@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index ef5acecd11fc..52102a24f972 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -457,12 +457,14 @@ static int iwl_mvm_load_ucode_wait_alive(struct iwl_mvm *mvm,
 #endif
 
 	/*
+	 * For pre-MLD API (MLD API doesn't use the timestamps):
 	 * All the BSSes in the BSS table include the GP2 in the system
 	 * at the beacon Rx time, this is of course no longer relevant
 	 * since we are resetting the firmware.
 	 * Purge all the BSS table.
 	 */
-	cfg80211_bss_flush(mvm->hw->wiphy);
+	if (!mvm->mld_api_is_used)
+		cfg80211_bss_flush(mvm->hw->wiphy);
 
 	return 0;
 }
-- 
cgit v1.2.3-58-ga151


From 9215152677d4b321801a92b06f6d5248b2b4465f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:09 +0300
Subject: wifi: iwlwifi: mvm: use IWL_FW_CHECK for link ID check

The lookup function iwl_mvm_rcu_fw_link_id_to_link_conf() is
normally called with input from the firmware, so it should use
IWL_FW_CHECK() instead of WARN_ON().

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.4ea8fb7c47d4.I1c22af213f97f69bfc14674502511c1bc504adfb@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index e51e834902a9..8eeb405e4326 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1460,7 +1460,8 @@ iwl_mvm_rcu_dereference_vif_id(struct iwl_mvm *mvm, u8 vif_id, bool rcu)
 static inline struct ieee80211_bss_conf *
 iwl_mvm_rcu_fw_link_id_to_link_conf(struct iwl_mvm *mvm, u8 link_id, bool rcu)
 {
-	if (WARN_ON(link_id >= ARRAY_SIZE(mvm->link_id_to_link_conf)))
+	if (IWL_FW_CHECK(mvm, link_id >= ARRAY_SIZE(mvm->link_id_to_link_conf),
+			 "erroneous FW link ID: %d\n", link_id))
 		return NULL;
 
 	if (rcu)
-- 
cgit v1.2.3-58-ga151


From f9068fe4fd49f9e4409c30546d7e16238942ce62 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:10 +0300
Subject: wifi: iwlwifi: mvm: always unblock EMLSR on ROC end

Since we always block EMLSR for ROC, we also need to always
unblock it, even if we don't have a P2P device interface.
Fix this.

Fixes: a1efeb823084 ("wifi: iwlwifi: mvm: Block EMLSR when a p2p/softAP vif is active")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.96bbf98b716d.Id5a36954f8ebaa95142fd3d3a7a52bab5363b0bd@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/time-event.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
index d73e19a8fdda..98490866a7f5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
@@ -127,7 +127,7 @@ static void iwl_mvm_cleanup_roc(struct iwl_mvm *mvm)
 			iwl_mvm_rm_aux_sta(mvm);
 	}
 
-	if (vif && !IS_ERR_OR_NULL(bss_vif))
+	if (!IS_ERR_OR_NULL(bss_vif))
 		iwl_mvm_unblock_esr(mvm, bss_vif, IWL_MVM_ESR_BLOCKED_ROC);
 	mutex_unlock(&mvm->mutex);
 }
-- 
cgit v1.2.3-58-ga151


From c4fe1c84b2b301ed91dcfdc15f719fb94fd83e70 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Tue, 25 Jun 2024 19:51:11 +0300
Subject: wifi: iwlwifi: update the BA notification API

We don't use the new field, but at least, document the change.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.3d7887e2e374.I37bf709969d069ff0392e0976e62e06fb7a87bc9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/tx.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
index bbd176d88820..71669f81d93b 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
@@ -698,6 +698,7 @@ enum iwl_mvm_ba_resp_flags {
  * @query_frame_cnt: SCD query frame count
  * @txed: number of frames sent in the aggregation (all-TIDs)
  * @done: number of frames that were Acked by the BA (all-TIDs)
+ * @rts_retry_cnt: RTS retry count
  * @reserved: reserved (for alignment)
  * @wireless_time: Wireless-media time
  * @tx_rate: the rate the aggregation was sent at
@@ -718,7 +719,8 @@ struct iwl_mvm_compressed_ba_notif {
 	__le16 query_frame_cnt;
 	__le16 txed;
 	__le16 done;
-	__le16 reserved;
+	u8 rts_retry_cnt;
+	u8 reserved;
 	__le32 wireless_time;
 	__le32 tx_rate;
 	__le16 tfd_cnt;
-- 
cgit v1.2.3-58-ga151


From 3778a17ab054313a5513e3341eaff49aea65a88a Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Tue, 25 Jun 2024 19:51:12 +0300
Subject: wifi: iwlwifi: mvm: remove init_dbg module parameter

This is no longer used

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Link: https://patch.msgid.link/20240625194805.17a1484f2485.I095c7482ac517111081f8ff40312b48ffdd7ff94@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c  | 8 ++------
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 5 -----
 drivers/net/wireless/intel/iwlwifi/mvm/ops.c | 9 +--------
 3 files changed, 3 insertions(+), 19 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 52102a24f972..1f27ed9c3886 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -828,7 +828,7 @@ remove_notif:
 	iwl_remove_notification(&mvm->notif_wait, &calib_wait);
 out:
 	mvm->rfkill_safe_init_done = false;
-	if (iwlmvm_mod_params.init_dbg && !mvm->nvm_data) {
+	if (!mvm->nvm_data) {
 		/* we want to debug INIT and we have no NVM - fake */
 		mvm->nvm_data = kzalloc(sizeof(struct iwl_nvm_data) +
 					sizeof(struct ieee80211_channel) +
@@ -1357,9 +1357,6 @@ static int iwl_mvm_load_rt_fw(struct iwl_mvm *mvm)
 
 	if (ret) {
 		IWL_ERR(mvm, "Failed to run INIT ucode: %d\n", ret);
-
-		if (iwlmvm_mod_params.init_dbg)
-			return 0;
 		return ret;
 	}
 
@@ -1598,8 +1595,7 @@ int iwl_mvm_up(struct iwl_mvm *mvm)
 	IWL_DEBUG_INFO(mvm, "RT uCode started.\n");
 	return 0;
  error:
-	if (!iwlmvm_mod_params.init_dbg || !ret)
-		iwl_mvm_stop_device(mvm);
+	iwl_mvm_stop_device(mvm);
 	return ret;
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 8eeb405e4326..f2d0c6fbcee5 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -83,14 +83,9 @@ extern const struct ieee80211_ops iwl_mvm_mld_hw_ops;
 
 /**
  * struct iwl_mvm_mod_params - module parameters for iwlmvm
- * @init_dbg: if true, then the NIC won't be stopped if the INIT fw asserted.
- *	We will register to mac80211 to have testmode working. The NIC must not
- *	be up'ed after the INIT fw asserted. This is useful to be able to use
- *	proprietary tools over testmode to debug the INIT fw.
  * @power_scheme: one of enum iwl_power_scheme
  */
 struct iwl_mvm_mod_params {
-	bool init_dbg;
 	int power_scheme;
 };
 extern struct iwl_mvm_mod_params iwlmvm_mod_params;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 008fe5fece87..df0de859fed4 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -41,12 +41,8 @@ static const struct iwl_op_mode_ops iwl_mvm_ops_mq;
 
 struct iwl_mvm_mod_params iwlmvm_mod_params = {
 	.power_scheme = IWL_POWER_SCHEME_BPS,
-	/* rest of fields are 0 by default */
 };
 
-module_param_named(init_dbg, iwlmvm_mod_params.init_dbg, bool, 0444);
-MODULE_PARM_DESC(init_dbg,
-		 "set to true to debug an ASSERT in INIT fw (default: false");
 module_param_named(power_scheme, iwlmvm_mod_params.power_scheme, int, 0444);
 MODULE_PARM_DESC(power_scheme,
 		 "power management scheme: 1-active, 2-balanced, 3-low power, default: 2");
@@ -864,8 +860,7 @@ get_nvm_from_fw:
 		ret = iwl_mvm_init_mcc(mvm);
 	}
 
-	if (!iwlmvm_mod_params.init_dbg || !ret)
-		iwl_mvm_stop_device(mvm);
+	iwl_mvm_stop_device(mvm);
 
 	mutex_unlock(&mvm->mutex);
 	wiphy_unlock(mvm->hw->wiphy);
@@ -1500,8 +1495,6 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 	iwl_fw_flush_dumps(&mvm->fwrt);
 	iwl_fw_runtime_free(&mvm->fwrt);
 
-	if (iwlmvm_mod_params.init_dbg)
-		return op_mode;
 	iwl_phy_db_free(mvm->phy_db);
 	kfree(mvm->scan_cmd);
 	iwl_trans_op_mode_leave(trans);
-- 
cgit v1.2.3-58-ga151


From 702935aeb1446b5e2ed1a41add14702ae2b09c77 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:13 +0300
Subject: wifi: iwlwifi: fw: api: fix some kernel-doc

There are naming issues of structs vs. kernel-doc,
fix some that I noticed now.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.4c530804b4ff.I68b894b9cdbd9560d86b92646e9b6b17a6d5117e@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/config.h   |  4 ++--
 drivers/net/wireless/intel/iwlwifi/fw/api/phy-ctxt.h |  4 ++--
 drivers/net/wireless/intel/iwlwifi/fw/api/power.h    | 10 +++++-----
 drivers/net/wireless/intel/iwlwifi/fw/api/scan.h     |  2 +-
 4 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/config.h b/drivers/net/wireless/intel/iwlwifi/fw/api/config.h
index 4419631604b4..1fc65469990e 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/config.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/config.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2012-2014, 2018-2019, 2023 Intel Corporation
+ * Copyright (C) 2012-2014, 2018-2019, 2023-2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -76,7 +76,7 @@ struct iwl_phy_specific_cfg {
 } __packed; /* PHY_SPECIFIC_CONFIGURATION_API_VER_1*/
 
 /**
- * struct iwl_phy_cfg_cmd - Phy configuration command
+ * struct iwl_phy_cfg_cmd_v1 - Phy configuration command
  *
  * @phy_cfg: PHY configuration value, uses &enum iwl_fw_phy_cfg
  * @calib_control: calibration control data
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/phy-ctxt.h b/drivers/net/wireless/intel/iwlwifi/fw/api/phy-ctxt.h
index 08a2c416ce60..f192e02e4ba8 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/phy-ctxt.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/phy-ctxt.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2012-2014, 2018, 2020-2023 Intel Corporation
+ * Copyright (C) 2012-2014, 2018, 2020-2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -113,7 +113,7 @@ struct iwl_phy_context_cmd_tail {
 } __packed;
 
 /**
- * struct iwl_phy_context_cmd - config of the PHY context
+ * struct iwl_phy_context_cmd_v1 - config of the PHY context
  * ( PHY_CONTEXT_CMD = 0x8 )
  * @id_and_color: ID and color of the relevant Binding
  * @action: action to perform, see &enum iwl_ctxt_action
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/power.h b/drivers/net/wireless/intel/iwlwifi/fw/api/power.h
index 532d5cfa9162..19eb6008fe4b 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/power.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/power.h
@@ -462,7 +462,7 @@ struct iwl_per_chain_offset {
 } __packed; /* PER_CHAIN_LIMIT_OFFSET_PER_CHAIN_S_VER_1 */
 
 /**
- * struct iwl_geo_tx_power_profile_cmd_v1 - struct for PER_CHAIN_LIMIT_OFFSET_CMD cmd.
+ * struct iwl_geo_tx_power_profiles_cmd_v1 - struct for PER_CHAIN_LIMIT_OFFSET_CMD cmd.
  * @ops: operations, value from &enum iwl_geo_per_chain_offset_operation
  * @table: offset profile per band.
  */
@@ -472,7 +472,7 @@ struct iwl_geo_tx_power_profiles_cmd_v1 {
 } __packed; /* PER_CHAIN_LIMIT_OFFSET_CMD_VER_1 */
 
 /**
- * struct iwl_geo_tx_power_profile_cmd_v2 - struct for PER_CHAIN_LIMIT_OFFSET_CMD cmd.
+ * struct iwl_geo_tx_power_profiles_cmd_v2 - struct for PER_CHAIN_LIMIT_OFFSET_CMD cmd.
  * @ops: operations, value from &enum iwl_geo_per_chain_offset_operation
  * @table: offset profile per band.
  * @table_revision: 0 for not-South Korea, 1 for South Korea (the name is misleading)
@@ -484,7 +484,7 @@ struct iwl_geo_tx_power_profiles_cmd_v2 {
 } __packed; /* PER_CHAIN_LIMIT_OFFSET_CMD_VER_2 */
 
 /**
- * struct iwl_geo_tx_power_profile_cmd_v3 - struct for PER_CHAIN_LIMIT_OFFSET_CMD cmd.
+ * struct iwl_geo_tx_power_profiles_cmd_v3 - struct for PER_CHAIN_LIMIT_OFFSET_CMD cmd.
  * @ops: operations, value from &enum iwl_geo_per_chain_offset_operation
  * @table: offset profile per band.
  * @table_revision: 0 for not-South Korea, 1 for South Korea (the name is misleading)
@@ -496,7 +496,7 @@ struct iwl_geo_tx_power_profiles_cmd_v3 {
 } __packed; /* PER_CHAIN_LIMIT_OFFSET_CMD_VER_3 */
 
 /**
- * struct iwl_geo_tx_power_profile_cmd_v4 - struct for PER_CHAIN_LIMIT_OFFSET_CMD cmd.
+ * struct iwl_geo_tx_power_profiles_cmd_v4 - struct for PER_CHAIN_LIMIT_OFFSET_CMD cmd.
  * @ops: operations, value from &enum iwl_geo_per_chain_offset_operation
  * @table: offset profile per band.
  * @table_revision: 0 for not-South Korea, 1 for South Korea (the name is misleading)
@@ -508,7 +508,7 @@ struct iwl_geo_tx_power_profiles_cmd_v4 {
 } __packed; /* PER_CHAIN_LIMIT_OFFSET_CMD_VER_4 */
 
 /**
- * struct iwl_geo_tx_power_profile_cmd_v5 - struct for PER_CHAIN_LIMIT_OFFSET_CMD cmd.
+ * struct iwl_geo_tx_power_profiles_cmd_v5 - struct for PER_CHAIN_LIMIT_OFFSET_CMD cmd.
  * @ops: operations, value from &enum iwl_geo_per_chain_offset_operation
  * @table: offset profile per band.
  * @table_revision: 0 for not-South Korea, 1 for South Korea (the name is misleading)
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
index c00f9422306d..69a204c7f14a 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
@@ -423,7 +423,7 @@ struct iwl_lmac_scan_complete_notif {
 } __packed;
 
 /**
- * struct iwl_scan_offload_complete - PERIODIC_SCAN_COMPLETE_NTF_API_S_VER_2
+ * struct iwl_periodic_scan_complete - PERIODIC_SCAN_COMPLETE_NTF_API_S_VER_2
  * @last_schedule_line: last schedule line executed (fast or regular)
  * @last_schedule_iteration: last scan iteration executed before scan abort
  * @status: &enum iwl_scan_offload_complete_status
-- 
cgit v1.2.3-58-ga151


From 924d1cd504f39689e88c22b8bb7df7f1405626d4 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:14 +0300
Subject: wifi: iwlwifi: trans: make bad state warnings

Kalle reported that this triggers very occasionally, but
we don't even know which place, except that it wasn't one
with a warning. Make all of them warnings since this is
really not meant to happen and indicates driver bugs.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.be7a3a95afae.Ie8606d36783818c043c971bf0bc6f4df6a6e8f5b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-trans.c | 46 ++++++++++----------------
 1 file changed, 18 insertions(+), 28 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.c b/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
index bd21bbe7e7e3..3c9d91496c82 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.c
@@ -108,10 +108,9 @@ int iwl_trans_send_cmd(struct iwl_trans *trans, struct iwl_host_cmd *cmd)
 	if (unlikely(test_bit(STATUS_FW_ERROR, &trans->status)))
 		return -EIO;
 
-	if (unlikely(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+	if (WARN_ONCE(trans->state != IWL_TRANS_FW_ALIVE,
+		      "bad state = %d\n", trans->state))
 		return -EIO;
-	}
 
 	if (!(cmd->flags & CMD_ASYNC))
 		lock_map_acquire_read(&trans->sync_cmd_lockdep_map);
@@ -407,10 +406,9 @@ int iwl_trans_tx(struct iwl_trans *trans, struct sk_buff *skb,
 	if (unlikely(test_bit(STATUS_FW_ERROR, &trans->status)))
 		return -EIO;
 
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+	if (WARN_ONCE(trans->state != IWL_TRANS_FW_ALIVE,
+		      "bad state = %d\n", trans->state))
 		return -EIO;
-	}
 
 	if (trans->trans_cfg->gen2)
 		return iwl_txq_gen2_tx(trans, skb, dev_cmd, queue);
@@ -422,10 +420,9 @@ IWL_EXPORT_SYMBOL(iwl_trans_tx);
 void iwl_trans_reclaim(struct iwl_trans *trans, int queue, int ssn,
 		       struct sk_buff_head *skbs, bool is_flush)
 {
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+	if (WARN_ONCE(trans->state != IWL_TRANS_FW_ALIVE,
+		      "bad state = %d\n", trans->state))
 		return;
-	}
 
 	iwl_pcie_reclaim(trans, queue, ssn, skbs, is_flush);
 }
@@ -444,10 +441,9 @@ bool iwl_trans_txq_enable_cfg(struct iwl_trans *trans, int queue, u16 ssn,
 {
 	might_sleep();
 
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+	if (WARN_ONCE(trans->state != IWL_TRANS_FW_ALIVE,
+		      "bad state = %d\n", trans->state))
 		return false;
-	}
 
 	return iwl_trans_pcie_txq_enable(trans, queue, ssn,
 					 cfg, queue_wdg_timeout);
@@ -456,10 +452,9 @@ IWL_EXPORT_SYMBOL(iwl_trans_txq_enable_cfg);
 
 int iwl_trans_wait_txq_empty(struct iwl_trans *trans, int queue)
 {
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+	if (WARN_ONCE(trans->state != IWL_TRANS_FW_ALIVE,
+		      "bad state = %d\n", trans->state))
 		return -EIO;
-	}
 
 	return iwl_trans_pcie_wait_txq_empty(trans, queue);
 }
@@ -467,11 +462,9 @@ IWL_EXPORT_SYMBOL(iwl_trans_wait_txq_empty);
 
 int iwl_trans_wait_tx_queues_empty(struct iwl_trans *trans, u32 txqs)
 {
-	/* No need to wait if the firmware is not alive */
-	if (trans->state != IWL_TRANS_FW_ALIVE) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+	if (WARN_ONCE(trans->state != IWL_TRANS_FW_ALIVE,
+		      "bad state = %d\n", trans->state))
 		return -EIO;
-	}
 
 	return iwl_trans_pcie_wait_txqs_empty(trans, txqs);
 }
@@ -480,10 +473,9 @@ IWL_EXPORT_SYMBOL(iwl_trans_wait_tx_queues_empty);
 void iwl_trans_freeze_txq_timer(struct iwl_trans *trans,
 				unsigned long txqs, bool freeze)
 {
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+	if (WARN_ONCE(trans->state != IWL_TRANS_FW_ALIVE,
+		      "bad state = %d\n", trans->state))
 		return;
-	}
 
 	iwl_pcie_freeze_txq_timer(trans, txqs, freeze);
 }
@@ -506,10 +498,9 @@ IWL_EXPORT_SYMBOL(iwl_trans_debugfs_cleanup);
 
 void iwl_trans_set_q_ptrs(struct iwl_trans *trans, int queue, int ptr)
 {
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+	if (WARN_ONCE(trans->state != IWL_TRANS_FW_ALIVE,
+		      "bad state = %d\n", trans->state))
 		return;
-	}
 
 	iwl_pcie_set_q_ptrs(trans, queue, ptr);
 }
@@ -520,10 +511,9 @@ int iwl_trans_txq_alloc(struct iwl_trans *trans, u32 flags, u32 sta_mask,
 {
 	might_sleep();
 
-	if (WARN_ON_ONCE(trans->state != IWL_TRANS_FW_ALIVE)) {
-		IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state);
+	if (WARN_ONCE(trans->state != IWL_TRANS_FW_ALIVE,
+		      "bad state = %d\n", trans->state))
 		return -EIO;
-	}
 
 	return iwl_txq_dyn_alloc(trans, flags, sta_mask, tid,
 				 size, wdg_timeout);
-- 
cgit v1.2.3-58-ga151


From c0b047fd243360bc10dfff100d23de3eb08ebc73 Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:15 +0300
Subject: wifi: iwlwifi: mvm: don't log error for failed UATS table read

This causes unnecessary error level kernel messages if the platform
does not have any UATS table.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Reviewed-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.306b7eed8671.I6e9294335378dab38ef957866a0d39ec1a2df7f8@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 1f27ed9c3886..08c4898c8f1a 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -506,7 +506,7 @@ static void iwl_mvm_uats_init(struct iwl_mvm *mvm)
 
 	ret = iwl_uefi_get_uats_table(mvm->trans, &mvm->fwrt);
 	if (ret < 0) {
-		IWL_ERR(mvm, "failed to read UATS table (%d)\n", ret);
+		IWL_DEBUG_FW(mvm, "failed to read UATS table (%d)\n", ret);
 		return;
 	}
 
-- 
cgit v1.2.3-58-ga151


From 7b24e0b8a1533ae3519922f56935313e0338282e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:16 +0300
Subject: wifi: iwlwifi: dvm: fix kernel-doc warnings

Mostly the docs just aren't kernel-doc anyway, and one is a typo.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.8b12f56bf8c0.I64fa9df72ca0e862b96647c062b8c9464318e649@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/dvm/commands.h | 8 ++++----
 drivers/net/wireless/intel/iwlwifi/dvm/tt.h       | 4 ++--
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/commands.h b/drivers/net/wireless/intel/iwlwifi/dvm/commands.h
index 04864d3fda63..3f49c0bccb28 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/commands.h
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/commands.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2005-2014, 2023 Intel Corporation
+ * Copyright (C) 2005-2014, 2023-2024 Intel Corporation
  */
 /*
  * Please use this file (commands.h) only for uCode API definitions.
@@ -177,7 +177,7 @@ enum {
  *
  *****************************************************************************/
 
-/**
+/*
  * iwlagn rate_n_flags bit fields
  *
  * rate_n_flags format is used in following iwlagn commands:
@@ -251,7 +251,7 @@ enum {
 #define RATE_MCS_SGI_POS 13
 #define RATE_MCS_SGI_MSK 0x2000
 
-/**
+/*
  * rate_n_flags Tx antenna masks
  * bit14:16
  */
@@ -2767,7 +2767,7 @@ struct iwl_missed_beacon_notif {
  *
  *****************************************************************************/
 
-/**
+/*
  * SENSITIVITY_CMD = 0xa8 (command, has simple generic response)
  *
  * This command sets up the Rx signal detector for a sensitivity level that
diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/tt.h b/drivers/net/wireless/intel/iwlwifi/dvm/tt.h
index 23dfcda0dd86..4af792d41dce 100644
--- a/drivers/net/wireless/intel/iwlwifi/dvm/tt.h
+++ b/drivers/net/wireless/intel/iwlwifi/dvm/tt.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 /******************************************************************************
  *
- * Copyright(c) 2007 - 2014, 2023 Intel Corporation. All rights reserved.
+ * Copyright(c) 2007 - 2014, 2023-2024 Intel Corporation. All rights reserved.
  *
  * Portions of this file are derived from the ipw3945 project, as well
  * as portions of the ieee80211 subsystem header files.
@@ -64,7 +64,7 @@ struct iwl_tt_trans {
 };
 
 /**
- * struct iwl_tt_mgnt - Thermal Throttling Management structure
+ * struct iwl_tt_mgmt - Thermal Throttling Management structure
  * @advanced_tt:    advanced thermal throttle required
  * @state:          current Thermal Throttling state
  * @tt_power_mode:  Thermal Throttling power mode index
-- 
cgit v1.2.3-58-ga151


From ab713d0c883dcb59400d8cbdaae1d6a970c957ec Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:17 +0300
Subject: wifi: iwlwifi: pcie: fix kernel-doc

One typo, and a few things were missing. Fix that.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.c667bc035757.Iae0e5903a35f8e42f86deb27429131f22329b8dd@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index 2040e00c744a..f2bb34270ccf 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -272,7 +272,7 @@ enum iwl_pcie_fw_reset_state {
 };
 
 /**
- * enum wl_pcie_imr_status - imr dma transfer state
+ * enum iwl_pcie_imr_status - imr dma transfer state
  * @IMR_D2S_IDLE: default value of the dma transfer
  * @IMR_D2S_REQUESTED: dma transfer requested
  * @IMR_D2S_COMPLETED: dma transfer completed
@@ -300,6 +300,10 @@ enum iwl_pcie_imr_status {
  * @bc_tbl_size: bytecount table size
  * @tso_hdr_page: page allocated (per CPU) for A-MSDU headers when doing TSO
  *	(and similar usage)
+ * @cmd: command queue data
+ * @cmd.fifo: FIFO number
+ * @cmd.q_id: queue ID
+ * @cmd.wdg_timeout: watchdog timeout
  * @tfd: TFD data
  * @tfd.max_tbs: max number of buffers per TFD
  * @tfd.size: TFD size
-- 
cgit v1.2.3-58-ga151


From 7fe5da1cd0aa77cfd6c3c70dd47af9854eb2d032 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:18 +0300
Subject: wifi: iwlwifi: fix kernel-doc in iwl-trans.h

One mismatched enum name, and some missing docs.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.4846bf27dec1.I31fdfad01abc82b1340c59e51ece3db2242c8816@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index ac8f8ff5a99d..ecf94ec2c2b0 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -129,6 +129,11 @@ enum CMD_MODE {
  * For allocation of the command and tx queues, this establishes the overall
  * size of the largest command we send to uCode, except for commands that
  * aren't fully copied and use other TFD space.
+ *
+ * @hdr: command header
+ * @payload: payload for the command
+ * @hdr_wide: wide command header
+ * @payload_wide: payload for the wide command
  */
 struct iwl_device_cmd {
 	union {
@@ -273,7 +278,7 @@ static inline void iwl_free_rxb(struct iwl_rx_cmd_buffer *r)
 #define IWL_9000_MAX_RX_HW_QUEUES	1
 
 /**
- * enum iwl_wowlan_status - WoWLAN image/device status
+ * enum iwl_d3_status - WoWLAN image/device status
  * @IWL_D3_STATUS_ALIVE: firmware is still running after resume
  * @IWL_D3_STATUS_RESET: device was reset while suspended
  */
-- 
cgit v1.2.3-58-ga151


From e3f637b35f5d8a8031bfe2f60ffae7acd7e7f96c Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:19 +0300
Subject: wifi: iwlwifi: fix kernel-doc in iwl-fh.h

Some blocks aren't really kernel-doc, and some are misformatted
or with mismatched names. Fix that.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.16865e5503ac.I5401edbf9ecbc25e07aad929bb56255410173711@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-fh.h | 34 ++++++++++++++---------------
 1 file changed, 17 insertions(+), 17 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
index 6ba374efaacb..5c8f1868db64 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h
@@ -15,7 +15,7 @@
 /* Flow Handler Definitions */
 /****************************/
 
-/**
+/*
  * This I/O area is directly read/writable by driver (e.g. Linux uses writel())
  * Addresses are offsets from device's PCI hardware base address.
  */
@@ -24,7 +24,7 @@
 #define FH_MEM_LOWER_BOUND_GEN2              (0xa06000)
 #define FH_MEM_UPPER_BOUND_GEN2              (0xa08000)
 
-/**
+/*
  * Keep-Warm (KW) buffer base address.
  *
  * Driver must allocate a 4KByte buffer that is for keeping the
@@ -44,7 +44,7 @@
 #define FH_KW_MEM_ADDR_REG		     (FH_MEM_LOWER_BOUND + 0x97C)
 
 
-/**
+/*
  * TFD Circular Buffers Base (CBBC) addresses
  *
  * Device has 16 base pointer registers, one for each of 16 host-DRAM-resident
@@ -143,7 +143,7 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(struct iwl_trans *trans,
  */
 #define TFH_SRV_DMA_CHNL0_BC	(0x1F70)
 
-/**
+/*
  * Rx SRAM Control and Status Registers (RSCSR)
  *
  * These registers provide handshake between driver and device for the Rx queue
@@ -216,21 +216,21 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(struct iwl_trans *trans,
 #define FH_MEM_RSCSR_UPPER_BOUND	(FH_MEM_LOWER_BOUND + 0xC00)
 #define FH_MEM_RSCSR_CHNL0		(FH_MEM_RSCSR_LOWER_BOUND)
 
-/**
+/*
  * Physical base address of 8-byte Rx Status buffer.
  * Bit fields:
  *  31-0: Rx status buffer physical base address [35:4], must 16-byte aligned.
  */
 #define FH_RSCSR_CHNL0_STTS_WPTR_REG	(FH_MEM_RSCSR_CHNL0)
 
-/**
+/*
  * Physical base address of Rx Buffer Descriptor Circular Buffer.
  * Bit fields:
  *  27-0:  RBD CD physical base address [35:8], must be 256-byte aligned.
  */
 #define FH_RSCSR_CHNL0_RBDCB_BASE_REG	(FH_MEM_RSCSR_CHNL0 + 0x004)
 
-/**
+/*
  * Rx write pointer (index, really!).
  * Bit fields:
  *  11-0:  Index of driver's most recent prepared-to-be-filled RBD, + 1.
@@ -242,7 +242,7 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(struct iwl_trans *trans,
 #define FW_RSCSR_CHNL0_RXDCB_RDPTR_REG	(FH_MEM_RSCSR_CHNL0 + 0x00c)
 #define FH_RSCSR_CHNL0_RDPTR		FW_RSCSR_CHNL0_RXDCB_RDPTR_REG
 
-/**
+/*
  * Rx Config/Status Registers (RCSR)
  * Rx Config Reg for channel 0 (only channel used)
  *
@@ -300,7 +300,7 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(struct iwl_trans *trans,
 #define FH_RCSR_CHNL0_RX_CONFIG_IRQ_DEST_NO_INT_VAL    (0x00000000)
 #define FH_RCSR_CHNL0_RX_CONFIG_IRQ_DEST_INT_HOST_VAL  (0x00001000)
 
-/**
+/*
  * Rx Shared Status Registers (RSSR)
  *
  * After stopping Rx DMA channel (writing 0 to
@@ -356,7 +356,7 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(struct iwl_trans *trans,
 #define RFH_RBDBUF_RBD0_LSB 0xA08300
 #define RFH_RBDBUF_RBD_LSB(q) (RFH_RBDBUF_RBD0_LSB + (q) * 8)
 
-/**
+/*
  * RFH Status Register
  *
  * Bit fields:
@@ -440,7 +440,7 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(struct iwl_trans *trans,
 #define FH_TFDIB_CTRL0_REG(_chnl)  (FH_TFDIB_LOWER_BOUND + 0x8 * (_chnl))
 #define FH_TFDIB_CTRL1_REG(_chnl)  (FH_TFDIB_LOWER_BOUND + 0x8 * (_chnl) + 0x4)
 
-/**
+/*
  * Transmit DMA Channel Control/Status Registers (TCSR)
  *
  * Device has one configuration register for each of 8 Tx DMA/FIFO channels
@@ -501,7 +501,7 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(struct iwl_trans *trans,
 #define FH_TCSR_CHNL_TX_BUF_STS_REG_POS_TB_NUM		(20)
 #define FH_TCSR_CHNL_TX_BUF_STS_REG_POS_TB_IDX		(12)
 
-/**
+/*
  * Tx Shared Status Registers (TSSR)
  *
  * After stopping Tx DMA channel (writing 0 to
@@ -518,7 +518,7 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(struct iwl_trans *trans,
 
 #define FH_TSSR_TX_STATUS_REG		(FH_TSSR_LOWER_BOUND + 0x010)
 
-/**
+/*
  * Bit fields for TSSR(Tx Shared Status & Control) error status register:
  * 31:  Indicates an address error when accessed to internal memory
  *	uCode/driver must write "1" in order to clear this flag
@@ -634,7 +634,7 @@ enum iwl_tfd_tb_hi_n_len {
 };
 
 /**
- * struct iwl_tfd_tb transmit buffer descriptor within transmit frame descriptor
+ * struct iwl_tfd_tb - transmit buffer descriptor within transmit frame descriptor
  *
  * This structure contains dma address and length of transmission address
  *
@@ -648,7 +648,7 @@ struct iwl_tfd_tb {
 } __packed;
 
 /**
- * struct iwl_tfh_tb transmit buffer descriptor within transmit frame descriptor
+ * struct iwl_tfh_tb - transmit buffer descriptor within transmit frame descriptor
  *
  * This structure contains dma address and length of transmission address
  *
@@ -717,7 +717,7 @@ struct iwl_tfh_tfd {
 /* Fixed (non-configurable) rx data from phy */
 
 /**
- * struct iwlagn_schedq_bc_tbl scheduler byte count table
+ * struct iwlagn_scd_bc_tbl - scheduler byte count table
  *	base physical address provided by SCD_DRAM_BASE_ADDR
  * For devices up to 22000:
  * @tfd_offset:
@@ -734,7 +734,7 @@ struct iwlagn_scd_bc_tbl {
 } __packed;
 
 /**
- * struct iwl_gen3_bc_tbl_entry scheduler byte count table entry gen3
+ * struct iwl_gen3_bc_tbl_entry - scheduler byte count table entry gen3
  * For AX210 and on:
  * @tfd_offset: 0-12 - tx command byte count
  *		12-13 - number of 64 byte chunks
-- 
cgit v1.2.3-58-ga151


From fa1a1eb21c9caf1ad633d4bd3cb3b9c299c6a3a3 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:20 +0300
Subject: wifi: iwlwifi: fix prototype mismatch kernel-doc warnings

Fix all the prototype mismatch and "wrong kernel-doc identifier"
warnings, due to typos in or misformatting of the kernel-doc.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.6ec65cf9b88c.I7804114d7369f352e80a0e8430f7119af8e210de@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/alive.h    | 6 +++---
 drivers/net/wireless/intel/iwlwifi/fw/api/binding.h  | 4 ++--
 drivers/net/wireless/intel/iwlwifi/fw/api/d3.h       | 4 ++--
 drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h  | 2 +-
 drivers/net/wireless/intel/iwlwifi/fw/api/location.h | 2 +-
 drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h  | 2 +-
 drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h  | 4 ++--
 drivers/net/wireless/intel/iwlwifi/fw/api/offload.h  | 2 +-
 drivers/net/wireless/intel/iwlwifi/fw/api/phy.h      | 2 +-
 drivers/net/wireless/intel/iwlwifi/fw/api/rs.h       | 4 ++--
 drivers/net/wireless/intel/iwlwifi/fw/api/rx.h       | 4 ++--
 drivers/net/wireless/intel/iwlwifi/fw/api/scan.h     | 8 ++++----
 drivers/net/wireless/intel/iwlwifi/fw/api/tx.h       | 2 +-
 drivers/net/wireless/intel/iwlwifi/fw/error-dump.h   | 4 ++--
 drivers/net/wireless/intel/iwlwifi/iwl-config.h      | 3 ++-
 drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h   | 2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h         | 2 +-
 drivers/net/wireless/intel/iwlwifi/mvm/rs.h          | 6 +++---
 drivers/net/wireless/intel/iwlwifi/mvm/sta.h         | 2 +-
 19 files changed, 33 insertions(+), 32 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/alive.h b/drivers/net/wireless/intel/iwlwifi/fw/api/alive.h
index e00ab21e7358..ebe85fdf08d3 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/alive.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/alive.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2012-2014, 2018, 2020-2021 Intel Corporation
+ * Copyright (C) 2012-2014, 2018, 2020-2021, 2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -113,7 +113,7 @@ struct iwl_alive_ntf_v6 {
 } __packed; /* UCODE_ALIVE_NTFY_API_S_VER_6 */
 
 /**
- * enum iwl_extended_cfg_flag - commands driver may send before
+ * enum iwl_extended_cfg_flags - commands driver may send before
  *	finishing init flow
  * @IWL_INIT_DEBUG_CFG: driver is going to send debug config command
  * @IWL_INIT_NVM: driver is going to send NVM_ACCESS commands
@@ -126,7 +126,7 @@ enum iwl_extended_cfg_flags {
 };
 
 /**
- * struct iwl_extended_cfg_cmd - mark what commands ucode should wait for
+ * struct iwl_init_extended_cfg_cmd - mark what commands ucode should wait for
  * before finishing init flows
  * @init_flags: values from iwl_extended_cfg_flags
  */
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/binding.h b/drivers/net/wireless/intel/iwlwifi/fw/api/binding.h
index d9044ada6a43..2397fdc37fc5 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/binding.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/binding.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2012-2014, 2020, 2022 Intel Corporation
+ * Copyright (C) 2012-2014, 2020, 2022, 2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2016-2017 Intel Deutschland GmbH
  */
@@ -77,7 +77,7 @@ struct iwl_time_quota_data_v1 {
 } __packed; /* TIME_QUOTA_DATA_API_S_VER_1 */
 
 /**
- * struct iwl_time_quota_cmd - configuration of time quota between bindings
+ * struct iwl_time_quota_cmd_v1 - configuration of time quota between bindings
  * ( TIME_QUOTA_CMD = 0x2c )
  * @quotas: allocations per binding
  * Note: on non-CDB the fourth one is the auxilary mac and is
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/d3.h b/drivers/net/wireless/intel/iwlwifi/fw/api/d3.h
index bbaaf3c73115..ffee7927cf26 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/d3.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/d3.h
@@ -42,7 +42,7 @@ struct iwl_d3_manager_config {
 /* TODO: OFFLOADS_QUERY_API_S_VER_1 */
 
 /**
- * enum iwl_d3_proto_offloads - enabled protocol offloads
+ * enum iwl_proto_offloads - enabled protocol offloads
  * @IWL_D3_PROTO_OFFLOAD_ARP: ARP data is enabled
  * @IWL_D3_PROTO_OFFLOAD_NS: NS (Neighbor Solicitation) is enabled
  * @IWL_D3_PROTO_IPV4_VALID: IPv4 data is valid
@@ -195,7 +195,7 @@ struct iwl_wowlan_pattern_v1 {
 #define IWL_WOWLAN_MAX_PATTERNS	20
 
 /**
- * struct iwl_wowlan_patterns_cmd - WoWLAN wakeup patterns
+ * struct iwl_wowlan_patterns_cmd_v1 - WoWLAN wakeup patterns
  */
 struct iwl_wowlan_patterns_cmd_v1 {
 	/**
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h b/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h
index 6cfe8a779cc7..e71b3bc20253 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h
@@ -291,7 +291,7 @@ struct iwl_fw_ini_addr_val {
 } __packed; /* FW_TLV_DEBUG_ADDR_VALUE_VER_1 */
 
 /**
- * struct iwl_fw_ini_conf_tlv - configuration TLV to set register/memory.
+ * struct iwl_fw_ini_conf_set_tlv - configuration TLV to set register/memory.
  *
  * @hdr: debug header
  * @time_point: time point to apply config. One of &enum iwl_fw_ini_time_point
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
index 7c158a8dfed0..30a54c7fa001 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/location.h
@@ -446,7 +446,7 @@ struct iwl_tof_responder_config_cmd {
 #define IWL_LCI_CIVIC_IE_MAX_SIZE	400
 
 /**
- * struct iwl_tof_responder_dyn_config_cmd - Dynamic responder settings
+ * struct iwl_tof_responder_dyn_config_cmd_v2 - Dynamic responder settings
  * @lci_len: The length of the 1st (LCI) part in the @lci_civic buffer
  * @civic_len: The length of the 2nd (CIVIC) part in the @lci_civic buffer
  * @lci_civic: The LCI/CIVIC buffer. LCI data (if exists) comes first, then, if
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
index 754c5d655ad0..ca6fa66d1917 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/mac-cfg.h
@@ -144,7 +144,7 @@ struct iwl_missed_vap_notif {
 } __packed; /* MISSED_VAP_NTFY_API_S_VER_1 */
 
 /**
- * struct iwl_channel_switch_start_notif - Channel switch start notification
+ * struct iwl_channel_switch_start_notif_v1 - Channel switch start notification
  *
  * @id_and_color: ID and color of the MAC
  */
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h
index 644c8df780bc..9ff5d7e538fd 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h
@@ -120,7 +120,7 @@ struct iwl_nvm_access_cmd {
 } __packed; /* NVM_ACCESS_CMD_API_S_VER_2 */
 
 /**
- * struct iwl_nvm_access_resp_ver2 - response to NVM_ACCESS_CMD
+ * struct iwl_nvm_access_resp - response to NVM_ACCESS_CMD
  * @offset: offset in bytes into the section
  * @length: in bytes, either how much was written or read
  * @type: NVM_SECTION_TYPE_*
@@ -212,7 +212,7 @@ struct iwl_nvm_get_info_phy {
 #define IWL_NUM_CHANNELS	110
 
 /**
- * struct iwl_nvm_get_info_regulatory - regulatory information
+ * struct iwl_nvm_get_info_regulatory_v1 - regulatory information
  * @lar_enabled: is LAR enabled
  * @channel_profile: regulatory data of this channel
  * @reserved: reserved
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/offload.h b/drivers/net/wireless/intel/iwlwifi/fw/api/offload.h
index 2ed7acc09e5a..6a7bbfd6b2b7 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/offload.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/offload.h
@@ -60,7 +60,7 @@ struct iwl_stored_beacon_notif_common {
 } __packed;
 
 /**
- * struct iwl_stored_beacon_notif - Stored beacon notification
+ * struct iwl_stored_beacon_notif_v2 - Stored beacon notification
  *
  * @common: fields common for all versions
  * @data: beacon data, length in @byte_count
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/phy.h b/drivers/net/wireless/intel/iwlwifi/fw/api/phy.h
index 92e4b62c119f..c73d4d597857 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/phy.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/phy.h
@@ -195,7 +195,7 @@ struct ct_kill_notif {
 } __packed; /* CT_KILL_NOTIFICATION_API_S_VER_1, CT_KILL_NOTIFICATION_API_S_VER_2 */
 
 /**
-* enum ctdp_cmd_operation - CTDP command operations
+* enum iwl_mvm_ctdp_cmd_operation - CTDP command operations
 * @CTDP_CMD_OPERATION_START: update the current budget
 * @CTDP_CMD_OPERATION_STOP: stop ctdp
 * @CTDP_CMD_OPERATION_REPORT: get the average budget
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h b/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h
index a1a272433b09..1a60f0cdf972 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/rs.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2012-2014, 2018-2022 Intel Corporation
+ * Copyright (C) 2012-2014, 2018-2022, 2024 Intel Corporation
  * Copyright (C) 2017 Intel Deutschland GmbH
  */
 #ifndef __iwl_fw_api_rs_h__
@@ -9,7 +9,7 @@
 #include "mac.h"
 
 /**
- * enum iwl_tlc_mng_cfg_flags_enum - options for TLC config flags
+ * enum iwl_tlc_mng_cfg_flags - options for TLC config flags
  * @IWL_TLC_MNG_CFG_FLAGS_STBC_MSK: enable STBC. For HE this enables STBC for
  *				    bandwidths <= 80MHz
  * @IWL_TLC_MNG_CFG_FLAGS_LDPC_MSK: enable LDPC
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
index e71f29d0c694..8bd85ef90052 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2012-2014, 2018-2023 Intel Corporation
+ * Copyright (C) 2012-2014, 2018-2024 Intel Corporation
  * Copyright (C) 2013-2015 Intel Mobile Communications GmbH
  * Copyright (C) 2015-2017 Intel Deutschland GmbH
  */
@@ -976,7 +976,7 @@ struct iwl_ba_window_status_notif {
 } __packed; /* BA_WINDOW_STATUS_NTFY_API_S_VER_1 */
 
 /**
- * struct iwl_rfh_queue_config - RX queue configuration
+ * struct iwl_rfh_queue_data - RX queue configuration
  * @q_num: Q num
  * @enable: enable queue
  * @reserved: alignment
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
index 69a204c7f14a..8598031567bb 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/scan.h
@@ -149,7 +149,7 @@ struct iwl_scan_offload_profile_cfg_data {
 } __packed;
 
 /**
- * struct iwl_scan_offload_profile_cfg
+ * struct iwl_scan_offload_profile_cfg_v1 - scan offload profile config
  * @profiles:	profiles to search for match
  * @data:	the rest of the data for profile_cfg
  */
@@ -789,7 +789,7 @@ struct iwl_scan_req_umac_tail_v1 {
 } __packed;
 
 /**
- * struct iwl_scan_req_umac_tail - the rest of the UMAC scan request command
+ * struct iwl_scan_req_umac_tail_v2 - the rest of the UMAC scan request command
  *      parameters following channels configuration array.
  * @schedule: two scheduling plans.
  * @delay: delay in TUs before starting the first scan iteration
@@ -1085,7 +1085,7 @@ struct iwl_scan_req_params_v12 {
 } __packed; /* SCAN_REQUEST_PARAMS_API_S_VER_12 */
 
 /**
- * struct iwl_scan_req_params_v16
+ * struct iwl_scan_req_params_v17 - scan request parameters (v17)
  * @general_params: &struct iwl_scan_general_params_v11
  * @channel_params: &struct iwl_scan_channel_params_v7
  * @periodic_params: &struct iwl_scan_periodic_parms_v1
@@ -1111,7 +1111,7 @@ struct iwl_scan_req_umac_v12 {
 } __packed; /* SCAN_REQUEST_CMD_UMAC_API_S_VER_12 */
 
 /**
- * struct iwl_scan_req_umac_v16
+ * struct iwl_scan_req_umac_v17 - scan request command (v17)
  * @uid: scan id, &enum iwl_umac_scan_uid_offsets
  * @ooc_priority: out of channel priority - &enum iwl_scan_priority
  * @scan_params: scan parameters
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
index 71669f81d93b..c5277e2f8cd4 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/tx.h
@@ -866,7 +866,7 @@ enum iwl_dump_control {
 };
 
 /**
- * struct iwl_tx_path_flush_cmd -- queue/FIFO flush command
+ * struct iwl_tx_path_flush_cmd_v1 -- queue/FIFO flush command
  * @queues_ctl: bitmap of queues to flush
  * @flush_ctl: control flags
  * @reserved: reserved
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h b/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h
index 5c76e3b94968..e63b08b7d336 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/error-dump.h
@@ -248,7 +248,7 @@ struct iwl_fw_error_dump_mem {
 #define IWL_INI_DUMP_NAME_TYPE (BIT(31) | BIT(24))
 
 /**
- * struct iwl_fw_error_dump_data - data for one type
+ * struct iwl_fw_ini_error_dump_data - data for one type
  * @type: &enum iwl_fw_ini_region_type
  * @sub_type: sub type id
  * @sub_type_ver: sub type version
@@ -278,7 +278,7 @@ struct iwl_fw_ini_dump_entry {
 } __packed;
 
 /**
- * struct iwl_fw_error_dump_file - header of dump file
+ * struct iwl_fw_ini_dump_file_hdr - header of dump file
  * @barker: must be %IWL_FW_INI_ERROR_DUMP_BARKER
  * @file_len: the length of all the file including the header
  */
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
index 732889f96ca2..b2abd4fd1944 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h
@@ -241,7 +241,7 @@ enum iwl_cfg_trans_ltr_delay {
 };
 
 /**
- * struct iwl_cfg_trans - information needed to start the trans
+ * struct iwl_cfg_trans_params - information needed to start the trans
  *
  * These values are specific to the device ID and do not change when
  * multiple configs are used for a single device ID.  They values are
@@ -258,6 +258,7 @@ enum iwl_cfg_trans_ltr_delay {
  * @mq_rx_supported: multi-queue rx support
  * @integrated: discrete or integrated
  * @low_latency_xtal: use the low latency xtal if supported
+ * @bisr_workaround: BISR hardware workaround (for 22260 series devices)
  * @ltr_delay: LTR delay parameter, &enum iwl_cfg_trans_ltr_delay.
  * @imr_enabled: use the IMR if supported.
  */
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
index fd82feb24b2d..0c6c3fb8c6dd 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.h
@@ -38,7 +38,7 @@ iwl_parse_nvm_data(struct iwl_trans *trans, const struct iwl_cfg *cfg,
 		   u8 tx_chains, u8 rx_chains);
 
 /**
- * iwl_parse_mcc_info - parse MCC (mobile country code) info coming from FW
+ * iwl_parse_nvm_mcc_info - parse MCC (mobile country code) info coming from FW
  *
  * This function parses the regulatory channel data received as a
  * MCC_UPDATE_CMD command. It returns a newly allocation regulatory domain,
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index f2d0c6fbcee5..362973fdeac0 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -612,7 +612,7 @@ enum iwl_mvm_sched_scan_pass_all_states {
 };
 
 /**
- * struct iwl_mvm_tt_mgnt - Thermal Throttling Management structure
+ * struct iwl_mvm_tt_mgmt - Thermal Throttling Management structure
  * @ct_kill_exit: worker to exit thermal kill
  * @dynamic_smps: Is thermal throttling enabled dynamic_smps?
  * @tx_backoff: The current thremal throttling tx backoff in uSec.
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
index 6cd4ec4d8f34..20b9fa7deeda 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
@@ -3,7 +3,7 @@
  *
  * Copyright(c) 2015 Intel Mobile Communications GmbH
  * Copyright(c) 2017 Intel Deutschland GmbH
- * Copyright (C) 2003 - 2014, 2018 - 2023 Intel Corporation
+ * Copyright (C) 2003 - 2014, 2018 - 2024 Intel Corporation
  *****************************************************************************/
 
 #ifndef __rs_h__
@@ -402,7 +402,7 @@ void iwl_mvm_rs_tx_status(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 			  int tid, struct ieee80211_tx_info *info, bool ndp);
 
 /**
- * iwl_rate_control_register - Register the rate control algorithm callbacks
+ * iwl_mvm_rate_control_register - Register the rate control algorithm callbacks
  *
  * Since the rate control algorithm is hardware specific, there is no need
  * or reason to place it as a stand alone module.  The driver can call
@@ -414,7 +414,7 @@ void iwl_mvm_rs_tx_status(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 int iwl_mvm_rate_control_register(void);
 
 /**
- * iwl_rate_control_unregister - Unregister the rate control callbacks
+ * iwl_mvm_rate_control_unregister - Unregister the rate control callbacks
  *
  * This should be called after calling ieee80211_unregister_hw, but before
  * the driver is unloaded.
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
index 754a05a8c189..0dc83d6afb3c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h
@@ -478,7 +478,7 @@ struct iwl_mvm_int_sta {
 };
 
 /**
- * Send the STA info to the FW.
+ * iwl_mvm_sta_send_to_fw - Send the STA info to the FW.
  *
  * @mvm: the iwl_mvm* to use
  * @sta: the STA
-- 
cgit v1.2.3-58-ga151


From 7c8afa63652ebc326fb659579db6d0e17a7fe217 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:21 +0300
Subject: wifi: iwlwifi: fix remaining mistagged kernel-doc comments

There are some comments left that aren't really kernel-doc,
remove the extra * that tags them.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.98119856de4b.I9ca0cee675b166c4a7d58d619ce3278494398ea2@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-csr.h  | 6 ++----
 drivers/net/wireless/intel/iwlwifi/iwl-prph.h | 4 ++--
 2 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
index 4511d7fb2279..98563757ce2c 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
 /*
- * Copyright (C) 2005-2014, 2018-2023 Intel Corporation
+ * Copyright (C) 2005-2014, 2018-2024 Intel Corporation
  * Copyright (C) 2013-2014 Intel Mobile Communications GmbH
  * Copyright (C) 2016 Intel Deutschland GmbH
  */
@@ -304,9 +304,7 @@
 #define CSR_HW_RFID_IS_CDB(_val)       (((_val) & 0x10000000) >> 28)
 #define CSR_HW_RFID_IS_JACKET(_val)    (((_val) & 0x20000000) >> 29)
 
-/**
- *  hw_rev values
- */
+/* hw_rev values */
 enum {
 	SILICON_A_STEP = 0,
 	SILICON_B_STEP,
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
index 898e22e0d1ab..dc171c29eb7b 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h
@@ -96,7 +96,7 @@
 #define DTSC_PTAT_AVG		(0x00a10650)
 
 
-/**
+/*
  * Tx Scheduler
  *
  * The Tx Scheduler selects the next frame to be transmitted, choosing TFDs
@@ -169,7 +169,7 @@
  */
 #define SCD_MEM_LOWER_BOUND		(0x0000)
 
-/**
+/*
  * Max Tx window size is the max number of contiguous TFDs that the scheduler
  * can keep track of at one time when creating block-ack chains of frames.
  * Note that "64" matches the number of ack bits in a block-ack packet.
-- 
cgit v1.2.3-58-ga151


From ba59cfb7417ca93e4240948a8767ffb1a2c366ec Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:22 +0300
Subject: wifi: iwlwifi: fw: api: datapath: fix kernel-doc

Fix kernel-doc warnings in datapath.h.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.1a644d4c38f4.I6060819da2bfc948bee089a91626ff474300a896@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/datapath.h   | 27 +++++++++++++++-------
 1 file changed, 19 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h b/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h
index f272b6a4e72e..2ab38eaeb290 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/datapath.h
@@ -231,28 +231,33 @@ struct iwl_synced_time_rsp {
 #define PTP_CTX_MAX_DATA_SIZE   128
 
 /**
- * struct iwl_time_msmt_ptp_ctx - Vendor specific information element
+ * struct iwl_time_msmt_ptp_ctx - Vendor specific element
  * to allow a space for flexibility for the userspace App
  *
- * @element_id: element id of vendor specific ie
- * @length: length of vendor specific ie
- * @reserved: for alignment
- * @data: vendor specific data blob
+ * @ftm: FTM specific vendor element
+ * @ftm.element_id: element id of vendor specific ie
+ * @ftm.length: length of vendor specific ie
+ * @ftm.reserved: for alignment
+ * @ftm.data: vendor specific data blob
+ * @tm: TM specific vendor element
+ * @tm.element_id: element id of vendor specific ie
+ * @tm.length: length of vendor specific ie
+ * @tm.data: vendor specific data blob
  */
 struct iwl_time_msmt_ptp_ctx {
-	/* Differentiate between FTM and TM specific Vendor IEs */
+	/* Differentiate between FTM and TM specific Vendor elements */
 	union {
 		struct {
 			u8 element_id;
 			u8 length;
 			__le16 reserved;
 			u8 data[PTP_CTX_MAX_DATA_SIZE];
-		} ftm; /* FTM specific vendor IE */
+		} ftm;
 		struct {
 			u8 element_id;
 			u8 length;
 			u8 data[PTP_CTX_MAX_DATA_SIZE];
-		} tm; /* TM specific vendor IE */
+		} tm;
 	};
 } __packed /* PTP_CTX_VER_1 */;
 
@@ -531,6 +536,10 @@ struct iwl_rx_baid_cfg_cmd_remove {
 /**
  * struct iwl_rx_baid_cfg_cmd - BAID allocation/config command
  * @action: the action, from &enum iwl_rx_baid_action
+ * @alloc: allocation data
+ * @modify: modify data
+ * @remove_v1: remove data (version 1)
+ * @remove: remove data
  */
 struct iwl_rx_baid_cfg_cmd {
 	__le32 action;
@@ -565,6 +574,7 @@ enum iwl_scd_queue_cfg_operation {
 /**
  * struct iwl_scd_queue_cfg_cmd - scheduler queue allocation command
  * @operation: the operation, see &enum iwl_scd_queue_cfg_operation
+ * @u: union depending on command usage
  * @u.add.sta_mask: station mask
  * @u.add.tid: TID
  * @u.add.reserved: reserved
@@ -634,6 +644,7 @@ enum iwl_sec_key_flags {
 /**
  * struct iwl_sec_key_cmd - security key command
  * @action: action from &enum iwl_ctxt_action
+ * @u: union depending on command type
  * @u.add.sta_mask: station mask for the new key
  * @u.add.key_id: key ID (0-7) for the new key
  * @u.add.key_flags: key flags per &enum iwl_sec_key_flags
-- 
cgit v1.2.3-58-ga151


From a6db5c70efd79b5ae96c1c161b0b776b7529ed68 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 25 Jun 2024 19:51:23 +0300
Subject: wifi: iwlwifi: mvm: fix rs.h kernel-doc

Some things are mislabeled here, fix that.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240625194805.baa3bd60c8f8.Ibc4886f7fe696d57991689cc2885cde5cecc8f90@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/rs.h | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
index 20b9fa7deeda..ea81cb236d5c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h
@@ -198,11 +198,12 @@ struct rs_rate {
 /**
  * struct iwl_lq_sta_rs_fw - rate and related statistics for RS in FW
  * @last_rate_n_flags: last rate reported by FW
+ * @pers: persistent fields
  * @pers.sta_id: the id of the station
- * @chains: bitmask of chains reported in %chain_signal
- * @chain_signal: per chain signal strength
- * @last_rssi: last rssi reported
- * @drv: pointer back to the driver data
+ * @pers.chains: bitmask of chains reported in %chain_signal
+ * @pers.chain_signal: per chain signal strength
+ * @pers.last_rssi: last rssi reported
+ * @pers.drv: pointer back to the driver data
  */
 struct iwl_lq_sta_rs_fw {
 	/* last tx rate_n_flags */
@@ -213,11 +214,11 @@ struct iwl_lq_sta_rs_fw {
 		u32 sta_id;
 #ifdef CONFIG_MAC80211_DEBUGFS
 		/**
-		 * @dbg_fixed_rate: for debug, use fixed rate if not 0
+		 * @pers.dbg_fixed_rate: for debug, use fixed rate if not 0
 		 */
 		u32 dbg_fixed_rate;
 		/**
-		 * @dbg_agg_frame_count_lim: for debug, max number of
+		 * @pers.dbg_agg_frame_count_lim: for debug, max number of
 		 *	frames in A-MPDU
 		 */
 		u16 dbg_agg_frame_count_lim;
-- 
cgit v1.2.3-58-ga151


From 021d53a3d87eeb9dbba524ac515651242a2a7e3b Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Mon, 17 Jun 2024 19:52:17 +0800
Subject: wifi: mac80211: fix NULL dereference at band check in starting tx ba
 session

In MLD connection, link_data/link_conf are dynamically allocated. They
don't point to vif->bss_conf. So, there will be no chanreq assigned to
vif->bss_conf and then the chan will be NULL. Tweak the code to check
ht_supported/vht_supported/has_he/has_eht on sta deflink.

Crash log (with rtw89 version under MLO development):
[ 9890.526087] BUG: kernel NULL pointer dereference, address: 0000000000000000
[ 9890.526102] #PF: supervisor read access in kernel mode
[ 9890.526105] #PF: error_code(0x0000) - not-present page
[ 9890.526109] PGD 0 P4D 0
[ 9890.526114] Oops: 0000 [#1] PREEMPT SMP PTI
[ 9890.526119] CPU: 2 PID: 6367 Comm: kworker/u16:2 Kdump: loaded Tainted: G           OE      6.9.0 #1
[ 9890.526123] Hardware name: LENOVO 2356AD1/2356AD1, BIOS G7ETB3WW (2.73 ) 11/28/2018
[ 9890.526126] Workqueue: phy2 rtw89_core_ba_work [rtw89_core]
[ 9890.526203] RIP: 0010:ieee80211_start_tx_ba_session (net/mac80211/agg-tx.c:618 (discriminator 1)) mac80211
[ 9890.526279] Code: f7 e8 d5 93 3e ea 48 83 c4 28 89 d8 5b 41 5c 41 5d 41 5e 41 5f 5d c3 cc cc cc cc 49 8b 84 24 e0 f1 ff ff 48 8b 80 90 1b 00 00 <83> 38 03 0f 84 37 fe ff ff bb ea ff ff ff eb cc 49 8b 84 24 10 f3
All code
========
   0:	f7 e8                	imul   %eax
   2:	d5                   	(bad)
   3:	93                   	xchg   %eax,%ebx
   4:	3e ea                	ds (bad)
   6:	48 83 c4 28          	add    $0x28,%rsp
   a:	89 d8                	mov    %ebx,%eax
   c:	5b                   	pop    %rbx
   d:	41 5c                	pop    %r12
   f:	41 5d                	pop    %r13
  11:	41 5e                	pop    %r14
  13:	41 5f                	pop    %r15
  15:	5d                   	pop    %rbp
  16:	c3                   	retq
  17:	cc                   	int3
  18:	cc                   	int3
  19:	cc                   	int3
  1a:	cc                   	int3
  1b:	49 8b 84 24 e0 f1 ff 	mov    -0xe20(%r12),%rax
  22:	ff
  23:	48 8b 80 90 1b 00 00 	mov    0x1b90(%rax),%rax
  2a:*	83 38 03             	cmpl   $0x3,(%rax)		<-- trapping instruction
  2d:	0f 84 37 fe ff ff    	je     0xfffffffffffffe6a
  33:	bb ea ff ff ff       	mov    $0xffffffea,%ebx
  38:	eb cc                	jmp    0x6
  3a:	49                   	rex.WB
  3b:	8b                   	.byte 0x8b
  3c:	84 24 10             	test   %ah,(%rax,%rdx,1)
  3f:	f3                   	repz

Code starting with the faulting instruction
===========================================
   0:	83 38 03             	cmpl   $0x3,(%rax)
   3:	0f 84 37 fe ff ff    	je     0xfffffffffffffe40
   9:	bb ea ff ff ff       	mov    $0xffffffea,%ebx
   e:	eb cc                	jmp    0xffffffffffffffdc
  10:	49                   	rex.WB
  11:	8b                   	.byte 0x8b
  12:	84 24 10             	test   %ah,(%rax,%rdx,1)
  15:	f3                   	repz
[ 9890.526285] RSP: 0018:ffffb8db09013d68 EFLAGS: 00010246
[ 9890.526291] RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff9308e0d656c8
[ 9890.526295] RDX: 0000000000000000 RSI: ffffffffab99460b RDI: ffffffffab9a7685
[ 9890.526300] RBP: ffffb8db09013db8 R08: 0000000000000000 R09: 0000000000000873
[ 9890.526304] R10: ffff9308e0d64800 R11: 0000000000000002 R12: ffff9308e5ff6e70
[ 9890.526308] R13: ffff930952500e20 R14: ffff9309192a8c00 R15: 0000000000000000
[ 9890.526313] FS:  0000000000000000(0000) GS:ffff930b4e700000(0000) knlGS:0000000000000000
[ 9890.526316] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 9890.526318] CR2: 0000000000000000 CR3: 0000000391c58005 CR4: 00000000001706f0
[ 9890.526321] Call Trace:
[ 9890.526324]  <TASK>
[ 9890.526327] ? show_regs (arch/x86/kernel/dumpstack.c:479)
[ 9890.526335] ? __die (arch/x86/kernel/dumpstack.c:421 arch/x86/kernel/dumpstack.c:434)
[ 9890.526340] ? page_fault_oops (arch/x86/mm/fault.c:713)
[ 9890.526347] ? search_module_extables (kernel/module/main.c:3256 (discriminator 3))
[ 9890.526353] ? ieee80211_start_tx_ba_session (net/mac80211/agg-tx.c:618 (discriminator 1)) mac80211

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Link: https://patch.msgid.link/20240617115217.22344-1-kevin_yang@realtek.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/agg-tx.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c
index 21d55dc539f6..677bbbac9f16 100644
--- a/net/mac80211/agg-tx.c
+++ b/net/mac80211/agg-tx.c
@@ -616,7 +616,9 @@ int ieee80211_start_tx_ba_session(struct ieee80211_sta *pubsta, u16 tid,
 		return -EINVAL;
 
 	if (!pubsta->deflink.ht_cap.ht_supported &&
-	    sta->sdata->vif.bss_conf.chanreq.oper.chan->band != NL80211_BAND_6GHZ)
+	    !pubsta->deflink.vht_cap.vht_supported &&
+	    !pubsta->deflink.he_cap.has_he &&
+	    !pubsta->deflink.eht_cap.has_eht)
 		return -EINVAL;
 
 	if (WARN_ON_ONCE(!local->ops->ampdu_action))
-- 
cgit v1.2.3-58-ga151


From 1bbdb7f7a4ebf2596b06a6bb84505a4a37a66d2e Mon Sep 17 00:00:00 2001
From: Zijun Hu <quic_zijuhu@quicinc.com>
Date: Thu, 13 Jun 2024 22:04:36 +0800
Subject: net: rfkill: Correct return value in invalid parameter case

rfkill_set_hw_state_reason() does not return current combined
block state when its parameter @reason is invalid, that is
wrong according to its comments, fix it by correcting the
value returned.

Also reformat the WARN while at it.

Signed-off-by: Zijun Hu <quic_zijuhu@quicinc.com>
Link: https://patch.msgid.link/1718287476-28227-1-git-send-email-quic_zijuhu@quicinc.com
[edit/reformat commit message, remove unneeded variable]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/rfkill/core.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index c3feb4f49d09..7a5367628c05 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -546,10 +546,10 @@ bool rfkill_set_hw_state_reason(struct rfkill *rfkill,
 
 	BUG_ON(!rfkill);
 
-	if (WARN(reason &
-	    ~(RFKILL_HARD_BLOCK_SIGNAL | RFKILL_HARD_BLOCK_NOT_OWNER),
-	    "hw_state reason not supported: 0x%lx", reason))
-		return blocked;
+	if (WARN(reason & ~(RFKILL_HARD_BLOCK_SIGNAL |
+			    RFKILL_HARD_BLOCK_NOT_OWNER),
+		 "hw_state reason not supported: 0x%lx", reason))
+		return rfkill_blocked(rfkill);
 
 	spin_lock_irqsave(&rfkill->lock, flags);
 	prev = !!(rfkill->hard_block_reasons & reason);
-- 
cgit v1.2.3-58-ga151


From f531d13bdfe3f4f084aaa8acae2cb0f02295f5ae Mon Sep 17 00:00:00 2001
From: Eyal Birger <eyal.birger@gmail.com>
Date: Mon, 27 May 2024 20:29:14 -0700
Subject: xfrm: support sending NAT keepalives in ESP in UDP states

Add the ability to send out RFC-3948 NAT keepalives from the xfrm stack.

To use, Userspace sets an XFRM_NAT_KEEPALIVE_INTERVAL integer property when
creating XFRM outbound states which denotes the number of seconds between
keepalive messages.

Keepalive messages are sent from a per net delayed work which iterates over
the xfrm states. The logic is guarded by the xfrm state spinlock due to the
xfrm state walk iterator.

Possible future enhancements:

- Adding counters to keep track of sent keepalives.
- deduplicate NAT keepalives between states sharing the same nat keepalive
  parameters.
- provisioning hardware offloads for devices capable of implementing this.
- revise xfrm state list to use an rcu list in order to avoid running this
  under spinlock.

Suggested-by: Paul Wouters <paul.wouters@aiven.io>
Tested-by: Paul Wouters <paul.wouters@aiven.io>
Tested-by: Antony Antony <antony.antony@secunet.com>
Signed-off-by: Eyal Birger <eyal.birger@gmail.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 include/net/ipv6_stubs.h      |   3 +
 include/net/netns/xfrm.h      |   1 +
 include/net/xfrm.h            |  10 ++
 include/uapi/linux/xfrm.h     |   1 +
 net/ipv6/af_inet6.c           |   1 +
 net/ipv6/xfrm6_policy.c       |   7 +
 net/xfrm/Makefile             |   3 +-
 net/xfrm/xfrm_compat.c        |   6 +-
 net/xfrm/xfrm_nat_keepalive.c | 292 ++++++++++++++++++++++++++++++++++++++++++
 net/xfrm/xfrm_policy.c        |   8 ++
 net/xfrm/xfrm_state.c         |  17 +++
 net/xfrm/xfrm_user.c          |  15 +++
 12 files changed, 361 insertions(+), 3 deletions(-)
 create mode 100644 net/xfrm/xfrm_nat_keepalive.c

diff --git a/include/net/ipv6_stubs.h b/include/net/ipv6_stubs.h
index 485c39a89866..11cefd50704d 100644
--- a/include/net/ipv6_stubs.h
+++ b/include/net/ipv6_stubs.h
@@ -9,6 +9,7 @@
 #include <net/flow.h>
 #include <net/neighbour.h>
 #include <net/sock.h>
+#include <net/ipv6.h>
 
 /* structs from net/ip6_fib.h */
 struct fib6_info;
@@ -72,6 +73,8 @@ struct ipv6_stub {
 			     int (*output)(struct net *, struct sock *, struct sk_buff *));
 	struct net_device *(*ipv6_dev_find)(struct net *net, const struct in6_addr *addr,
 					    struct net_device *dev);
+	int (*ip6_xmit)(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6,
+			__u32 mark, struct ipv6_txoptions *opt, int tclass, u32 priority);
 };
 extern const struct ipv6_stub *ipv6_stub __read_mostly;
 
diff --git a/include/net/netns/xfrm.h b/include/net/netns/xfrm.h
index 423b52eca908..d489d9250bff 100644
--- a/include/net/netns/xfrm.h
+++ b/include/net/netns/xfrm.h
@@ -83,6 +83,7 @@ struct netns_xfrm {
 
 	spinlock_t xfrm_policy_lock;
 	struct mutex xfrm_cfg_mutex;
+	struct delayed_work	nat_keepalive_work;
 };
 
 #endif
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 77ebf5bcf0b9..46a214a76081 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -229,6 +229,10 @@ struct xfrm_state {
 	struct xfrm_encap_tmpl	*encap;
 	struct sock __rcu	*encap_sk;
 
+	/* NAT keepalive */
+	u32			nat_keepalive_interval; /* seconds */
+	time64_t		nat_keepalive_expiration;
+
 	/* Data for care-of address */
 	xfrm_address_t	*coaddr;
 
@@ -2203,4 +2207,10 @@ static inline int register_xfrm_state_bpf(void)
 }
 #endif
 
+int xfrm_nat_keepalive_init(unsigned short family);
+void xfrm_nat_keepalive_fini(unsigned short family);
+int xfrm_nat_keepalive_net_init(struct net *net);
+int xfrm_nat_keepalive_net_fini(struct net *net);
+void xfrm_nat_keepalive_state_updated(struct xfrm_state *x);
+
 #endif	/* _NET_XFRM_H */
diff --git a/include/uapi/linux/xfrm.h b/include/uapi/linux/xfrm.h
index d950d02ab791..f28701500714 100644
--- a/include/uapi/linux/xfrm.h
+++ b/include/uapi/linux/xfrm.h
@@ -321,6 +321,7 @@ enum xfrm_attr_type_t {
 	XFRMA_IF_ID,		/* __u32 */
 	XFRMA_MTIMER_THRESH,	/* __u32 in seconds for input SA */
 	XFRMA_SA_DIR,		/* __u8 */
+	XFRMA_NAT_KEEPALIVE_INTERVAL,	/* __u32 in seconds for NAT keepalive */
 	__XFRMA_MAX
 
 #define XFRMA_OUTPUT_MARK XFRMA_SET_MARK	/* Compatibility */
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 8041dc181bd4..2b893858b9a9 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -1060,6 +1060,7 @@ static const struct ipv6_stub ipv6_stub_impl = {
 	.nd_tbl	= &nd_tbl,
 	.ipv6_fragment = ip6_fragment,
 	.ipv6_dev_find = ipv6_dev_find,
+	.ip6_xmit = ip6_xmit,
 };
 
 static const struct ipv6_bpf_stub ipv6_bpf_stub_impl = {
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index cc885d3aa9e5..6837ff05f11a 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -284,8 +284,14 @@ int __init xfrm6_init(void)
 	ret = register_pernet_subsys(&xfrm6_net_ops);
 	if (ret)
 		goto out_protocol;
+
+	ret = xfrm_nat_keepalive_init(AF_INET6);
+	if (ret)
+		goto out_nat_keepalive;
 out:
 	return ret;
+out_nat_keepalive:
+	unregister_pernet_subsys(&xfrm6_net_ops);
 out_protocol:
 	xfrm6_protocol_fini();
 out_state:
@@ -297,6 +303,7 @@ out_policy:
 
 void xfrm6_fini(void)
 {
+	xfrm_nat_keepalive_fini(AF_INET6);
 	unregister_pernet_subsys(&xfrm6_net_ops);
 	xfrm6_protocol_fini();
 	xfrm6_policy_fini();
diff --git a/net/xfrm/Makefile b/net/xfrm/Makefile
index 547cec77ba03..512e0b2f8514 100644
--- a/net/xfrm/Makefile
+++ b/net/xfrm/Makefile
@@ -13,7 +13,8 @@ endif
 
 obj-$(CONFIG_XFRM) := xfrm_policy.o xfrm_state.o xfrm_hash.o \
 		      xfrm_input.o xfrm_output.o \
-		      xfrm_sysctl.o xfrm_replay.o xfrm_device.o
+		      xfrm_sysctl.o xfrm_replay.o xfrm_device.o \
+		      xfrm_nat_keepalive.o
 obj-$(CONFIG_XFRM_STATISTICS) += xfrm_proc.o
 obj-$(CONFIG_XFRM_ALGO) += xfrm_algo.o
 obj-$(CONFIG_XFRM_USER) += xfrm_user.o
diff --git a/net/xfrm/xfrm_compat.c b/net/xfrm/xfrm_compat.c
index 703d4172c7d7..91357ccaf4af 100644
--- a/net/xfrm/xfrm_compat.c
+++ b/net/xfrm/xfrm_compat.c
@@ -131,6 +131,7 @@ static const struct nla_policy compat_policy[XFRMA_MAX+1] = {
 	[XFRMA_IF_ID]		= { .type = NLA_U32 },
 	[XFRMA_MTIMER_THRESH]	= { .type = NLA_U32 },
 	[XFRMA_SA_DIR]          = NLA_POLICY_RANGE(NLA_U8, XFRM_SA_DIR_IN, XFRM_SA_DIR_OUT),
+	[XFRMA_NAT_KEEPALIVE_INTERVAL]	= { .type = NLA_U32 },
 };
 
 static struct nlmsghdr *xfrm_nlmsg_put_compat(struct sk_buff *skb,
@@ -280,9 +281,10 @@ static int xfrm_xlate64_attr(struct sk_buff *dst, const struct nlattr *src)
 	case XFRMA_IF_ID:
 	case XFRMA_MTIMER_THRESH:
 	case XFRMA_SA_DIR:
+	case XFRMA_NAT_KEEPALIVE_INTERVAL:
 		return xfrm_nla_cpy(dst, src, nla_len(src));
 	default:
-		BUILD_BUG_ON(XFRMA_MAX != XFRMA_SA_DIR);
+		BUILD_BUG_ON(XFRMA_MAX != XFRMA_NAT_KEEPALIVE_INTERVAL);
 		pr_warn_once("unsupported nla_type %d\n", src->nla_type);
 		return -EOPNOTSUPP;
 	}
@@ -437,7 +439,7 @@ static int xfrm_xlate32_attr(void *dst, const struct nlattr *nla,
 	int err;
 
 	if (type > XFRMA_MAX) {
-		BUILD_BUG_ON(XFRMA_MAX != XFRMA_SA_DIR);
+		BUILD_BUG_ON(XFRMA_MAX != XFRMA_NAT_KEEPALIVE_INTERVAL);
 		NL_SET_ERR_MSG(extack, "Bad attribute");
 		return -EOPNOTSUPP;
 	}
diff --git a/net/xfrm/xfrm_nat_keepalive.c b/net/xfrm/xfrm_nat_keepalive.c
new file mode 100644
index 000000000000..82f0a301683f
--- /dev/null
+++ b/net/xfrm/xfrm_nat_keepalive.c
@@ -0,0 +1,292 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * xfrm_nat_keepalive.c
+ *
+ * (c) 2024 Eyal Birger <eyal.birger@gmail.com>
+ */
+
+#include <net/inet_common.h>
+#include <net/ip6_checksum.h>
+#include <net/xfrm.h>
+
+static DEFINE_PER_CPU(struct sock *, nat_keepalive_sk_ipv4);
+#if IS_ENABLED(CONFIG_IPV6)
+static DEFINE_PER_CPU(struct sock *, nat_keepalive_sk_ipv6);
+#endif
+
+struct nat_keepalive {
+	struct net *net;
+	u16 family;
+	xfrm_address_t saddr;
+	xfrm_address_t daddr;
+	__be16 encap_sport;
+	__be16 encap_dport;
+	__u32 smark;
+};
+
+static void nat_keepalive_init(struct nat_keepalive *ka, struct xfrm_state *x)
+{
+	ka->net = xs_net(x);
+	ka->family = x->props.family;
+	ka->saddr = x->props.saddr;
+	ka->daddr = x->id.daddr;
+	ka->encap_sport = x->encap->encap_sport;
+	ka->encap_dport = x->encap->encap_dport;
+	ka->smark = xfrm_smark_get(0, x);
+}
+
+static int nat_keepalive_send_ipv4(struct sk_buff *skb,
+				   struct nat_keepalive *ka)
+{
+	struct net *net = ka->net;
+	struct flowi4 fl4;
+	struct rtable *rt;
+	struct sock *sk;
+	__u8 tos = 0;
+	int err;
+
+	flowi4_init_output(&fl4, 0 /* oif */, skb->mark, tos,
+			   RT_SCOPE_UNIVERSE, IPPROTO_UDP, 0,
+			   ka->daddr.a4, ka->saddr.a4, ka->encap_dport,
+			   ka->encap_sport, sock_net_uid(net, NULL));
+
+	rt = ip_route_output_key(net, &fl4);
+	if (IS_ERR(rt))
+		return PTR_ERR(rt);
+
+	skb_dst_set(skb, &rt->dst);
+
+	sk = *this_cpu_ptr(&nat_keepalive_sk_ipv4);
+	sock_net_set(sk, net);
+	err = ip_build_and_send_pkt(skb, sk, fl4.saddr, fl4.daddr, NULL, tos);
+	sock_net_set(sk, &init_net);
+	return err;
+}
+
+#if IS_ENABLED(CONFIG_IPV6)
+static int nat_keepalive_send_ipv6(struct sk_buff *skb,
+				   struct nat_keepalive *ka,
+				   struct udphdr *uh)
+{
+	struct net *net = ka->net;
+	struct dst_entry *dst;
+	struct flowi6 fl6;
+	struct sock *sk;
+	__wsum csum;
+	int err;
+
+	csum = skb_checksum(skb, 0, skb->len, 0);
+	uh->check = csum_ipv6_magic(&ka->saddr.in6, &ka->daddr.in6,
+				    skb->len, IPPROTO_UDP, csum);
+	if (uh->check == 0)
+		uh->check = CSUM_MANGLED_0;
+
+	memset(&fl6, 0, sizeof(fl6));
+	fl6.flowi6_mark = skb->mark;
+	fl6.saddr = ka->saddr.in6;
+	fl6.daddr = ka->daddr.in6;
+	fl6.flowi6_proto = IPPROTO_UDP;
+	fl6.fl6_sport = ka->encap_sport;
+	fl6.fl6_dport = ka->encap_dport;
+
+	sk = *this_cpu_ptr(&nat_keepalive_sk_ipv6);
+	sock_net_set(sk, net);
+	dst = ipv6_stub->ipv6_dst_lookup_flow(net, sk, &fl6, NULL);
+	if (IS_ERR(dst))
+		return PTR_ERR(dst);
+
+	skb_dst_set(skb, dst);
+	err = ipv6_stub->ip6_xmit(sk, skb, &fl6, skb->mark, NULL, 0, 0);
+	sock_net_set(sk, &init_net);
+	return err;
+}
+#endif
+
+static void nat_keepalive_send(struct nat_keepalive *ka)
+{
+	const int nat_ka_hdrs_len = max(sizeof(struct iphdr),
+					sizeof(struct ipv6hdr)) +
+				    sizeof(struct udphdr);
+	const u8 nat_ka_payload = 0xFF;
+	int err = -EAFNOSUPPORT;
+	struct sk_buff *skb;
+	struct udphdr *uh;
+
+	skb = alloc_skb(nat_ka_hdrs_len + sizeof(nat_ka_payload), GFP_ATOMIC);
+	if (unlikely(!skb))
+		return;
+
+	skb_reserve(skb, nat_ka_hdrs_len);
+
+	skb_put_u8(skb, nat_ka_payload);
+
+	uh = skb_push(skb, sizeof(*uh));
+	uh->source = ka->encap_sport;
+	uh->dest = ka->encap_dport;
+	uh->len = htons(skb->len);
+	uh->check = 0;
+
+	skb->mark = ka->smark;
+
+	switch (ka->family) {
+	case AF_INET:
+		err = nat_keepalive_send_ipv4(skb, ka);
+		break;
+#if IS_ENABLED(CONFIG_IPV6)
+	case AF_INET6:
+		err = nat_keepalive_send_ipv6(skb, ka, uh);
+		break;
+#endif
+	}
+	if (err)
+		kfree_skb(skb);
+}
+
+struct nat_keepalive_work_ctx {
+	time64_t next_run;
+	time64_t now;
+};
+
+static int nat_keepalive_work_single(struct xfrm_state *x, int count, void *ptr)
+{
+	struct nat_keepalive_work_ctx *ctx = ptr;
+	bool send_keepalive = false;
+	struct nat_keepalive ka;
+	time64_t next_run;
+	u32 interval;
+	int delta;
+
+	interval = x->nat_keepalive_interval;
+	if (!interval)
+		return 0;
+
+	spin_lock(&x->lock);
+
+	delta = (int)(ctx->now - x->lastused);
+	if (delta < interval) {
+		x->nat_keepalive_expiration = ctx->now + interval - delta;
+		next_run = x->nat_keepalive_expiration;
+	} else if (x->nat_keepalive_expiration > ctx->now) {
+		next_run = x->nat_keepalive_expiration;
+	} else {
+		next_run = ctx->now + interval;
+		nat_keepalive_init(&ka, x);
+		send_keepalive = true;
+	}
+
+	spin_unlock(&x->lock);
+
+	if (send_keepalive)
+		nat_keepalive_send(&ka);
+
+	if (!ctx->next_run || next_run < ctx->next_run)
+		ctx->next_run = next_run;
+	return 0;
+}
+
+static void nat_keepalive_work(struct work_struct *work)
+{
+	struct nat_keepalive_work_ctx ctx;
+	struct xfrm_state_walk walk;
+	struct net *net;
+
+	ctx.next_run = 0;
+	ctx.now = ktime_get_real_seconds();
+
+	net = container_of(work, struct net, xfrm.nat_keepalive_work.work);
+	xfrm_state_walk_init(&walk, IPPROTO_ESP, NULL);
+	xfrm_state_walk(net, &walk, nat_keepalive_work_single, &ctx);
+	xfrm_state_walk_done(&walk, net);
+	if (ctx.next_run)
+		schedule_delayed_work(&net->xfrm.nat_keepalive_work,
+				      (ctx.next_run - ctx.now) * HZ);
+}
+
+static int nat_keepalive_sk_init(struct sock * __percpu *socks,
+				 unsigned short family)
+{
+	struct sock *sk;
+	int err, i;
+
+	for_each_possible_cpu(i) {
+		err = inet_ctl_sock_create(&sk, family, SOCK_RAW, IPPROTO_UDP,
+					   &init_net);
+		if (err < 0)
+			goto err;
+
+		*per_cpu_ptr(socks, i) = sk;
+	}
+
+	return 0;
+err:
+	for_each_possible_cpu(i)
+		inet_ctl_sock_destroy(*per_cpu_ptr(socks, i));
+	return err;
+}
+
+static void nat_keepalive_sk_fini(struct sock * __percpu *socks)
+{
+	int i;
+
+	for_each_possible_cpu(i)
+		inet_ctl_sock_destroy(*per_cpu_ptr(socks, i));
+}
+
+void xfrm_nat_keepalive_state_updated(struct xfrm_state *x)
+{
+	struct net *net;
+
+	if (!x->nat_keepalive_interval)
+		return;
+
+	net = xs_net(x);
+	schedule_delayed_work(&net->xfrm.nat_keepalive_work, 0);
+}
+
+int __net_init xfrm_nat_keepalive_net_init(struct net *net)
+{
+	INIT_DELAYED_WORK(&net->xfrm.nat_keepalive_work, nat_keepalive_work);
+	return 0;
+}
+
+int xfrm_nat_keepalive_net_fini(struct net *net)
+{
+	cancel_delayed_work_sync(&net->xfrm.nat_keepalive_work);
+	return 0;
+}
+
+int xfrm_nat_keepalive_init(unsigned short family)
+{
+	int err = -EAFNOSUPPORT;
+
+	switch (family) {
+	case AF_INET:
+		err = nat_keepalive_sk_init(&nat_keepalive_sk_ipv4, PF_INET);
+		break;
+#if IS_ENABLED(CONFIG_IPV6)
+	case AF_INET6:
+		err = nat_keepalive_sk_init(&nat_keepalive_sk_ipv6, PF_INET6);
+		break;
+#endif
+	}
+
+	if (err)
+		pr_err("xfrm nat keepalive init: failed to init err:%d\n", err);
+	return err;
+}
+EXPORT_SYMBOL_GPL(xfrm_nat_keepalive_init);
+
+void xfrm_nat_keepalive_fini(unsigned short family)
+{
+	switch (family) {
+	case AF_INET:
+		nat_keepalive_sk_fini(&nat_keepalive_sk_ipv4);
+		break;
+#if IS_ENABLED(CONFIG_IPV6)
+	case AF_INET6:
+		nat_keepalive_sk_fini(&nat_keepalive_sk_ipv6);
+		break;
+#endif
+	}
+}
+EXPORT_SYMBOL_GPL(xfrm_nat_keepalive_fini);
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 475b904fe68b..6603d3bd171f 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -4289,8 +4289,14 @@ static int __net_init xfrm_net_init(struct net *net)
 	if (rv < 0)
 		goto out_sysctl;
 
+	rv = xfrm_nat_keepalive_net_init(net);
+	if (rv < 0)
+		goto out_nat_keepalive;
+
 	return 0;
 
+out_nat_keepalive:
+	xfrm_sysctl_fini(net);
 out_sysctl:
 	xfrm_policy_fini(net);
 out_policy:
@@ -4303,6 +4309,7 @@ out_statistics:
 
 static void __net_exit xfrm_net_exit(struct net *net)
 {
+	xfrm_nat_keepalive_net_fini(net);
 	xfrm_sysctl_fini(net);
 	xfrm_policy_fini(net);
 	xfrm_state_fini(net);
@@ -4364,6 +4371,7 @@ void __init xfrm_init(void)
 #endif
 
 	register_xfrm_state_bpf();
+	xfrm_nat_keepalive_init(AF_INET);
 }
 
 #ifdef CONFIG_AUDITSYSCALL
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 649bb739df0d..abadc857cd45 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -715,6 +715,7 @@ int __xfrm_state_delete(struct xfrm_state *x)
 		if (x->id.spi)
 			hlist_del_rcu(&x->byspi);
 		net->xfrm.state_num--;
+		xfrm_nat_keepalive_state_updated(x);
 		spin_unlock(&net->xfrm.xfrm_state_lock);
 
 		if (x->encap_sk)
@@ -1453,6 +1454,7 @@ static void __xfrm_state_insert(struct xfrm_state *x)
 	net->xfrm.state_num++;
 
 	xfrm_hash_grow_check(net, x->bydst.next != NULL);
+	xfrm_nat_keepalive_state_updated(x);
 }
 
 /* net->xfrm.xfrm_state_lock is held */
@@ -2871,6 +2873,21 @@ int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload,
 			goto error;
 	}
 
+	if (x->nat_keepalive_interval) {
+		if (x->dir != XFRM_SA_DIR_OUT) {
+			NL_SET_ERR_MSG(extack, "NAT keepalive is only supported for outbound SAs");
+			err = -EINVAL;
+			goto error;
+		}
+
+		if (!x->encap || x->encap->encap_type != UDP_ENCAP_ESPINUDP) {
+			NL_SET_ERR_MSG(extack,
+				       "NAT keepalive is only supported for UDP encapsulation");
+			err = -EINVAL;
+			goto error;
+		}
+	}
+
 error:
 	return err;
 }
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index e83c687bd64e..a552cfa623ea 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -833,6 +833,10 @@ static struct xfrm_state *xfrm_state_construct(struct net *net,
 	if (attrs[XFRMA_SA_DIR])
 		x->dir = nla_get_u8(attrs[XFRMA_SA_DIR]);
 
+	if (attrs[XFRMA_NAT_KEEPALIVE_INTERVAL])
+		x->nat_keepalive_interval =
+			nla_get_u32(attrs[XFRMA_NAT_KEEPALIVE_INTERVAL]);
+
 	err = __xfrm_init_state(x, false, attrs[XFRMA_OFFLOAD_DEV], extack);
 	if (err)
 		goto error;
@@ -1288,6 +1292,13 @@ static int copy_to_user_state_extra(struct xfrm_state *x,
 	}
 	if (x->dir)
 		ret = nla_put_u8(skb, XFRMA_SA_DIR, x->dir);
+
+	if (x->nat_keepalive_interval) {
+		ret = nla_put_u32(skb, XFRMA_NAT_KEEPALIVE_INTERVAL,
+				  x->nat_keepalive_interval);
+		if (ret)
+			goto out;
+	}
 out:
 	return ret;
 }
@@ -3165,6 +3176,7 @@ const struct nla_policy xfrma_policy[XFRMA_MAX+1] = {
 	[XFRMA_IF_ID]		= { .type = NLA_U32 },
 	[XFRMA_MTIMER_THRESH]   = { .type = NLA_U32 },
 	[XFRMA_SA_DIR]          = NLA_POLICY_RANGE(NLA_U8, XFRM_SA_DIR_IN, XFRM_SA_DIR_OUT),
+	[XFRMA_NAT_KEEPALIVE_INTERVAL] = { .type = NLA_U32 },
 };
 EXPORT_SYMBOL_GPL(xfrma_policy);
 
@@ -3474,6 +3486,9 @@ static inline unsigned int xfrm_sa_len(struct xfrm_state *x)
 	if (x->dir)
 		l += nla_total_size(sizeof(x->dir));
 
+	if (x->nat_keepalive_interval)
+		l += nla_total_size(sizeof(x->nat_keepalive_interval));
+
 	return l;
 }
 
-- 
cgit v1.2.3-58-ga151


From a6a6a98094116b60e5523a571d9443c53325f5b1 Mon Sep 17 00:00:00 2001
From: Uwe Kleine-König <u.kleine-koenig@baylibre.com>
Date: Tue, 25 Jun 2024 10:38:53 +0200
Subject: net: Drop explicit initialization of struct
 i2c_device_id::driver_data to 0
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

These drivers don't use the driver_data member of struct i2c_device_id,
so don't explicitly initialize this member.

This prepares putting driver_data in an anonymous union which requires
either no initialization or named designators. But it's also a nice
cleanup on its own.

While add it, also remove commas after the sentinel entries.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@baylibre.com>
Reviewed-by: Petr Machata <petrm@nvidia.com> # For mlxsw
Reviewed-by: Kory Maincent <Kory.maincent@bootlin.com>
Reviewed-by: Jeremy Kerr <jk@codeconstruct.com.au> # for mctp-i2c
Reviewed-by: Oleksij Rempel <o.rempel@pengutronix.de>
Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
Link: https://patch.msgid.link/20240625083853.2205977-2-u.kleine-koenig@baylibre.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lan9303_i2c.c                 | 2 +-
 drivers/net/dsa/microchip/ksz9477_i2c.c       | 4 ++--
 drivers/net/dsa/xrs700x/xrs700x_i2c.c         | 4 ++--
 drivers/net/ethernet/mellanox/mlxsw/minimal.c | 4 ++--
 drivers/net/mctp/mctp-i2c.c                   | 4 ++--
 drivers/net/pse-pd/pd692x0.c                  | 4 ++--
 drivers/net/pse-pd/tps23881.c                 | 4 ++--
 7 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/drivers/net/dsa/lan9303_i2c.c b/drivers/net/dsa/lan9303_i2c.c
index bbbec322bc4f..c62d27cdc117 100644
--- a/drivers/net/dsa/lan9303_i2c.c
+++ b/drivers/net/dsa/lan9303_i2c.c
@@ -89,7 +89,7 @@ static void lan9303_i2c_shutdown(struct i2c_client *client)
 /*-------------------------------------------------------------------------*/
 
 static const struct i2c_device_id lan9303_i2c_id[] = {
-	{ "lan9303", 0 },
+	{ "lan9303" },
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(i2c, lan9303_i2c_id);
diff --git a/drivers/net/dsa/microchip/ksz9477_i2c.c b/drivers/net/dsa/microchip/ksz9477_i2c.c
index 82bebee4615c..7d7560f23a73 100644
--- a/drivers/net/dsa/microchip/ksz9477_i2c.c
+++ b/drivers/net/dsa/microchip/ksz9477_i2c.c
@@ -72,8 +72,8 @@ static void ksz9477_i2c_shutdown(struct i2c_client *i2c)
 }
 
 static const struct i2c_device_id ksz9477_i2c_id[] = {
-	{ "ksz9477-switch", 0 },
-	{},
+	{ "ksz9477-switch" },
+	{}
 };
 
 MODULE_DEVICE_TABLE(i2c, ksz9477_i2c_id);
diff --git a/drivers/net/dsa/xrs700x/xrs700x_i2c.c b/drivers/net/dsa/xrs700x/xrs700x_i2c.c
index c1179d7311f7..9b731dea78c1 100644
--- a/drivers/net/dsa/xrs700x/xrs700x_i2c.c
+++ b/drivers/net/dsa/xrs700x/xrs700x_i2c.c
@@ -127,8 +127,8 @@ static void xrs700x_i2c_shutdown(struct i2c_client *i2c)
 }
 
 static const struct i2c_device_id xrs700x_i2c_id[] = {
-	{ "xrs700x-switch", 0 },
-	{},
+	{ "xrs700x-switch" },
+	{}
 };
 
 MODULE_DEVICE_TABLE(i2c, xrs700x_i2c_id);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/minimal.c b/drivers/net/ethernet/mellanox/mlxsw/minimal.c
index f0ceb196a6ce..431accdc6213 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/minimal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/minimal.c
@@ -702,8 +702,8 @@ static struct mlxsw_driver mlxsw_m_driver = {
 };
 
 static const struct i2c_device_id mlxsw_m_i2c_id[] = {
-	{ "mlxsw_minimal", 0},
-	{ },
+	{ "mlxsw_minimal" },
+	{ }
 };
 
 static struct i2c_driver mlxsw_m_i2c_driver = {
diff --git a/drivers/net/mctp/mctp-i2c.c b/drivers/net/mctp/mctp-i2c.c
index b37a9e4bade4..f9afea25044f 100644
--- a/drivers/net/mctp/mctp-i2c.c
+++ b/drivers/net/mctp/mctp-i2c.c
@@ -1042,8 +1042,8 @@ static struct notifier_block mctp_i2c_notifier = {
 };
 
 static const struct i2c_device_id mctp_i2c_id[] = {
-	{ "mctp-i2c-interface", 0 },
-	{},
+	{ "mctp-i2c-interface" },
+	{}
 };
 MODULE_DEVICE_TABLE(i2c, mctp_i2c_id);
 
diff --git a/drivers/net/pse-pd/pd692x0.c b/drivers/net/pse-pd/pd692x0.c
index 6488b941703c..820358b71f0f 100644
--- a/drivers/net/pse-pd/pd692x0.c
+++ b/drivers/net/pse-pd/pd692x0.c
@@ -1194,8 +1194,8 @@ static void pd692x0_i2c_remove(struct i2c_client *client)
 }
 
 static const struct i2c_device_id pd692x0_id[] = {
-	{ PD692X0_PSE_NAME, 0 },
-	{ },
+	{ PD692X0_PSE_NAME },
+	{ }
 };
 MODULE_DEVICE_TABLE(i2c, pd692x0_id);
 
diff --git a/drivers/net/pse-pd/tps23881.c b/drivers/net/pse-pd/tps23881.c
index 98ffbb1bbf13..61f6ad9c1934 100644
--- a/drivers/net/pse-pd/tps23881.c
+++ b/drivers/net/pse-pd/tps23881.c
@@ -794,8 +794,8 @@ static int tps23881_i2c_probe(struct i2c_client *client)
 }
 
 static const struct i2c_device_id tps23881_id[] = {
-	{ "tps23881", 0 },
-	{ },
+	{ "tps23881" },
+	{ }
 };
 MODULE_DEVICE_TABLE(i2c, tps23881_id);
 
-- 
cgit v1.2.3-58-ga151


From 8f8cea8f3ddbd8f64d4f08c2d8525989c4fb44d4 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Tue, 25 Jun 2024 15:47:34 +0200
Subject: mlxsw: pci: Store number of scatter/gather entries for maximum packet
 size

A previous patch-set used page pool for Rx buffers allocations. To
simplify the change, we first used page pool for one allocation per
packet - one continuous buffer is allocated for each packet. This can be
improved by using fragmented buffers, then memory consumption will be
significantly reduced.

WQE (Work Queue Element) includes up to 3 scatter/gather entries for
data. As preparation for fragmented buffer usage, calculate number of
scatter/gather entries which are required for packet according to
maximum MTU and store it for future use. For now use PAGE_SIZE for each
entry, which means that maximum buffer size is 3 * PAGE_SIZE. This is
enough for the maximum MTU which is supported in the driver now (10K).
Warn in an unlikely case of maximum MTU which requires more than 3 pages,
for now this warn should not happen with standard page size (>=4K) and
maximum MTU (10K).

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/98c3e3adb7e727e571ac538faf67cef262cec4fc.1719321422.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/pci.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index cb043379c01c..b54d8cbaa7de 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -111,6 +111,7 @@ struct mlxsw_pci {
 	bool cff_support;
 	enum mlxsw_cmd_mbox_config_profile_lag_mode lag_mode;
 	enum mlxsw_cmd_mbox_config_profile_flood_mode flood_mode;
+	u8 num_sg_entries; /* Number of scatter/gather entries for packets. */
 	struct mlxsw_pci_queue_type_group queues[MLXSW_PCI_QUEUE_TYPE_COUNT];
 	u32 doorbell_offset;
 	struct mlxsw_core *core;
@@ -427,6 +428,12 @@ static void mlxsw_pci_rdq_page_free(struct mlxsw_pci_queue *q,
 	page_pool_put_page(cq->u.cq.page_pool, elem_info->page, -1, false);
 }
 
+static u8 mlxsw_pci_num_sg_entries_get(u16 byte_count)
+{
+	return DIV_ROUND_UP(byte_count + MLXSW_PCI_RX_BUF_SW_OVERHEAD,
+			    PAGE_SIZE);
+}
+
 static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 			      struct mlxsw_pci_queue *q)
 {
@@ -1774,6 +1781,17 @@ static void mlxsw_pci_free_irq_vectors(struct mlxsw_pci *mlxsw_pci)
 	pci_free_irq_vectors(mlxsw_pci->pdev);
 }
 
+static void mlxsw_pci_num_sg_entries_set(struct mlxsw_pci *mlxsw_pci)
+{
+	u8 num_sg_entries;
+
+	num_sg_entries = mlxsw_pci_num_sg_entries_get(MLXSW_PORT_MAX_MTU);
+	mlxsw_pci->num_sg_entries = min(num_sg_entries,
+					MLXSW_PCI_WQE_SG_ENTRIES);
+
+	WARN_ON(num_sg_entries > MLXSW_PCI_WQE_SG_ENTRIES);
+}
+
 static int mlxsw_pci_init(void *bus_priv, struct mlxsw_core *mlxsw_core,
 			  const struct mlxsw_config_profile *profile,
 			  struct mlxsw_res *res)
@@ -1896,6 +1914,8 @@ static int mlxsw_pci_init(void *bus_priv, struct mlxsw_core *mlxsw_core,
 	if (err)
 		goto err_requery_resources;
 
+	mlxsw_pci_num_sg_entries_set(mlxsw_pci);
+
 	err = mlxsw_pci_napi_devs_init(mlxsw_pci);
 	if (err)
 		goto err_napi_devs_init;
-- 
cgit v1.2.3-58-ga151


From 36437f469d7e92635c8e07b63bd490f0c14c3cba Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Tue, 25 Jun 2024 15:47:35 +0200
Subject: mlxsw: pci: Use fragmented buffers

WQE (Work Queue Element) includes 3 scatter/gather entries for buffers.
The buffer can be split into 3 parts, software should set address and byte
count of each part.

A previous patch-set used page pool to allocate buffers, to simplify the
change, we first used one continuous buffer, which was allocated with
order > 0. This patch improves page pool usage to allocate the exact
number of pages which are required for packet.

As part of init, fill WQE.address[x] and WQE.byte_count* with pages which
are allocated from the pool. Fill x entries according to number of
scatter/gather entries which are required for maximum packet size. When a
packet is received, check the actual size and replace only the used pages.
Save bytes for software overhead only as part of the first entry.

This change also requires using fragmented SKB, till now all the buffer
was in the linear part. Note that 'skb->truesize' is decreased for small
packets.

For now the maximum buffer size is 3 * PAGE_SIZE which is enough, in
case that the driver will support larger MTU, we can use 'order' to
allocate more than one page per scatter/gather entry.

This change significantly reduces memory consumption of mlxsw driver. The
footprint is reduced by 26%.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/ee38898c692e7f644a7f3ea4d33aeddb4dd917d2.1719321422.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/pci.c | 163 +++++++++++++++++++++++-------
 1 file changed, 129 insertions(+), 34 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index b54d8cbaa7de..e794162a91d0 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -62,7 +62,7 @@ struct mlxsw_pci_mem_item {
 };
 
 struct mlxsw_pci_queue_elem_info {
-	struct page *page;
+	struct page *pages[MLXSW_PCI_WQE_SG_ENTRIES];
 	char *elem; /* pointer to actual dma mapped element mem chunk */
 	struct {
 		struct sk_buff *skb;
@@ -350,7 +350,11 @@ mlxsw_pci_wqe_rx_frag_set(struct mlxsw_pci *mlxsw_pci, struct page *page,
 	dma_addr_t mapaddr;
 
 	mapaddr = page_pool_get_dma_addr(page);
-	mapaddr += MLXSW_PCI_SKB_HEADROOM;
+
+	if (index == 0) {
+		mapaddr += MLXSW_PCI_SKB_HEADROOM;
+		frag_len = frag_len - MLXSW_PCI_RX_BUF_SW_OVERHEAD;
+	}
 
 	mlxsw_pci_wqe_address_set(wqe, index, mapaddr);
 	mlxsw_pci_wqe_byte_count_set(wqe, index, frag_len);
@@ -385,29 +389,56 @@ static void mlxsw_pci_wqe_frag_unmap(struct mlxsw_pci *mlxsw_pci, char *wqe,
 	dma_unmap_single(&pdev->dev, mapaddr, frag_len, direction);
 }
 
-static struct sk_buff *mlxsw_pci_rdq_build_skb(struct page *page,
+static struct sk_buff *mlxsw_pci_rdq_build_skb(struct page *pages[],
 					       u16 byte_count)
 {
-	void *data = page_address(page);
-	unsigned int allocated_size;
+	unsigned int linear_data_size;
 	struct sk_buff *skb;
+	int page_index = 0;
+	bool linear_only;
+	void *data;
 
+	data = page_address(pages[page_index]);
 	net_prefetch(data);
-	allocated_size = page_size(page);
-	skb = napi_build_skb(data, allocated_size);
+
+	skb = napi_build_skb(data, PAGE_SIZE);
 	if (unlikely(!skb))
 		return ERR_PTR(-ENOMEM);
 
+	linear_only = byte_count + MLXSW_PCI_RX_BUF_SW_OVERHEAD <= PAGE_SIZE;
+	linear_data_size = linear_only ? byte_count :
+					 PAGE_SIZE -
+					 MLXSW_PCI_RX_BUF_SW_OVERHEAD;
+
 	skb_reserve(skb, MLXSW_PCI_SKB_HEADROOM);
-	skb_put(skb, byte_count);
+	skb_put(skb, linear_data_size);
+
+	if (linear_only)
+		return skb;
+
+	byte_count -= linear_data_size;
+	page_index++;
+
+	while (byte_count > 0) {
+		unsigned int frag_size;
+		struct page *page;
+
+		page = pages[page_index];
+		frag_size = min(byte_count, PAGE_SIZE);
+		skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags,
+				page, 0, frag_size, PAGE_SIZE);
+		byte_count -= frag_size;
+		page_index++;
+	}
+
 	return skb;
 }
 
 static int mlxsw_pci_rdq_page_alloc(struct mlxsw_pci_queue *q,
-				    struct mlxsw_pci_queue_elem_info *elem_info)
+				    struct mlxsw_pci_queue_elem_info *elem_info,
+				    int index)
 {
 	struct mlxsw_pci_queue *cq = q->u.rdq.cq;
-	size_t buf_len = MLXSW_PORT_MAX_MTU;
 	char *wqe = elem_info->elem;
 	struct page *page;
 
@@ -415,17 +446,19 @@ static int mlxsw_pci_rdq_page_alloc(struct mlxsw_pci_queue *q,
 	if (unlikely(!page))
 		return -ENOMEM;
 
-	mlxsw_pci_wqe_rx_frag_set(q->pci, page, wqe, 0, buf_len);
-	elem_info->page = page;
+	mlxsw_pci_wqe_rx_frag_set(q->pci, page, wqe, index, PAGE_SIZE);
+	elem_info->pages[index] = page;
 	return 0;
 }
 
 static void mlxsw_pci_rdq_page_free(struct mlxsw_pci_queue *q,
-				    struct mlxsw_pci_queue_elem_info *elem_info)
+				    struct mlxsw_pci_queue_elem_info *elem_info,
+				    int index)
 {
 	struct mlxsw_pci_queue *cq = q->u.rdq.cq;
 
-	page_pool_put_page(cq->u.cq.page_pool, elem_info->page, -1, false);
+	page_pool_put_page(cq->u.cq.page_pool, elem_info->pages[index], -1,
+			   false);
 }
 
 static u8 mlxsw_pci_num_sg_entries_get(u16 byte_count)
@@ -434,6 +467,64 @@ static u8 mlxsw_pci_num_sg_entries_get(u16 byte_count)
 			    PAGE_SIZE);
 }
 
+static int
+mlxsw_pci_elem_info_pages_ref_store(const struct mlxsw_pci_queue *q,
+				    const struct mlxsw_pci_queue_elem_info *el,
+				    u16 byte_count, struct page *pages[],
+				    u8 *p_num_sg_entries)
+{
+	u8 num_sg_entries;
+	int i;
+
+	num_sg_entries = mlxsw_pci_num_sg_entries_get(byte_count);
+	if (WARN_ON_ONCE(num_sg_entries > q->pci->num_sg_entries))
+		return -EINVAL;
+
+	for (i = 0; i < num_sg_entries; i++)
+		pages[i] = el->pages[i];
+
+	*p_num_sg_entries = num_sg_entries;
+	return 0;
+}
+
+static int
+mlxsw_pci_rdq_pages_alloc(struct mlxsw_pci_queue *q,
+			  struct mlxsw_pci_queue_elem_info *elem_info,
+			  u8 num_sg_entries)
+{
+	struct page *old_pages[MLXSW_PCI_WQE_SG_ENTRIES];
+	struct mlxsw_pci_queue *cq = q->u.rdq.cq;
+	int i, err;
+
+	for (i = 0; i < num_sg_entries; i++) {
+		old_pages[i] = elem_info->pages[i];
+		err = mlxsw_pci_rdq_page_alloc(q, elem_info, i);
+		if (err) {
+			dev_err_ratelimited(&q->pci->pdev->dev, "Failed to alloc page\n");
+			goto err_page_alloc;
+		}
+	}
+
+	return 0;
+
+err_page_alloc:
+	for (i--; i >= 0; i--)
+		page_pool_recycle_direct(cq->u.cq.page_pool, old_pages[i]);
+
+	return err;
+}
+
+static void
+mlxsw_pci_rdq_pages_recycle(struct mlxsw_pci_queue *q, struct page *pages[],
+			    u8 num_sg_entries)
+{
+	struct mlxsw_pci_queue *cq = q->u.rdq.cq;
+	int i;
+
+	for (i = 0; i < num_sg_entries; i++)
+		page_pool_recycle_direct(cq->u.cq.page_pool, pages[i]);
+}
+
 static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 			      struct mlxsw_pci_queue *q)
 {
@@ -441,7 +532,7 @@ static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 	u8 sdq_count = mlxsw_pci->num_sdqs;
 	struct mlxsw_pci_queue *cq;
 	u8 cq_num;
-	int i;
+	int i, j;
 	int err;
 
 	q->producer_counter = 0;
@@ -472,9 +563,12 @@ static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 	for (i = 0; i < q->count; i++) {
 		elem_info = mlxsw_pci_queue_elem_info_producer_get(q);
 		BUG_ON(!elem_info);
-		err = mlxsw_pci_rdq_page_alloc(q, elem_info);
-		if (err)
-			goto rollback;
+
+		for (j = 0; j < mlxsw_pci->num_sg_entries; j++) {
+			err = mlxsw_pci_rdq_page_alloc(q, elem_info, j);
+			if (err)
+				goto rollback;
+		}
 		/* Everything is set up, ring doorbell to pass elem to HW */
 		q->producer_counter++;
 		mlxsw_pci_queue_doorbell_producer_ring(mlxsw_pci, q);
@@ -485,7 +579,9 @@ static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox,
 rollback:
 	for (i--; i >= 0; i--) {
 		elem_info = mlxsw_pci_queue_elem_info_get(q, i);
-		mlxsw_pci_rdq_page_free(q, elem_info);
+		for (j--; j >= 0; j--)
+			mlxsw_pci_rdq_page_free(q, elem_info, j);
+		j = mlxsw_pci->num_sg_entries;
 	}
 	q->u.rdq.cq = NULL;
 	cq->u.cq.dq = NULL;
@@ -498,12 +594,13 @@ static void mlxsw_pci_rdq_fini(struct mlxsw_pci *mlxsw_pci,
 			       struct mlxsw_pci_queue *q)
 {
 	struct mlxsw_pci_queue_elem_info *elem_info;
-	int i;
+	int i, j;
 
 	mlxsw_cmd_hw2sw_rdq(mlxsw_pci->core, q->num);
 	for (i = 0; i < q->count; i++) {
 		elem_info = mlxsw_pci_queue_elem_info_get(q, i);
-		mlxsw_pci_rdq_page_free(q, elem_info);
+		for (j = 0; j < mlxsw_pci->num_sg_entries; j++)
+			mlxsw_pci_rdq_page_free(q, elem_info, j);
 	}
 }
 
@@ -637,11 +734,11 @@ static void mlxsw_pci_cqe_rdq_handle(struct mlxsw_pci *mlxsw_pci,
 				     enum mlxsw_pci_cqe_v cqe_v, char *cqe)
 {
 	struct pci_dev *pdev = mlxsw_pci->pdev;
+	struct page *pages[MLXSW_PCI_WQE_SG_ENTRIES];
 	struct mlxsw_pci_queue_elem_info *elem_info;
-	struct mlxsw_pci_queue *cq = q->u.rdq.cq;
 	struct mlxsw_rx_info rx_info = {};
 	struct sk_buff *skb;
-	struct page *page;
+	u8 num_sg_entries;
 	u16 byte_count;
 	int err;
 
@@ -654,18 +751,19 @@ static void mlxsw_pci_cqe_rdq_handle(struct mlxsw_pci *mlxsw_pci,
 	if (mlxsw_pci_cqe_crc_get(cqe_v, cqe))
 		byte_count -= ETH_FCS_LEN;
 
-	page = elem_info->page;
+	err = mlxsw_pci_elem_info_pages_ref_store(q, elem_info, byte_count,
+						  pages, &num_sg_entries);
+	if (err)
+		goto out;
 
-	err = mlxsw_pci_rdq_page_alloc(q, elem_info);
-	if (err) {
-		dev_err_ratelimited(&pdev->dev, "Failed to alloc page\n");
+	err = mlxsw_pci_rdq_pages_alloc(q, elem_info, num_sg_entries);
+	if (err)
 		goto out;
-	}
 
-	skb = mlxsw_pci_rdq_build_skb(page, byte_count);
+	skb = mlxsw_pci_rdq_build_skb(pages, byte_count);
 	if (IS_ERR(skb)) {
 		dev_err_ratelimited(&pdev->dev, "Failed to build skb for RDQ\n");
-		page_pool_recycle_direct(cq->u.cq.page_pool, page);
+		mlxsw_pci_rdq_pages_recycle(q, pages, num_sg_entries);
 		goto out;
 	}
 
@@ -886,15 +984,12 @@ static int mlxsw_pci_cq_page_pool_init(struct mlxsw_pci_queue *q,
 	struct page_pool_params pp_params = {};
 	struct mlxsw_pci *mlxsw_pci = q->pci;
 	struct page_pool *page_pool;
-	u32 max_pkt_size;
 
 	if (cq_type != MLXSW_PCI_CQ_RDQ)
 		return 0;
 
-	max_pkt_size = MLXSW_PORT_MAX_MTU + MLXSW_PCI_RX_BUF_SW_OVERHEAD;
-	pp_params.order = get_order(max_pkt_size);
 	pp_params.flags = PP_FLAG_DMA_MAP;
-	pp_params.pool_size = MLXSW_PCI_WQE_COUNT;
+	pp_params.pool_size = MLXSW_PCI_WQE_COUNT * mlxsw_pci->num_sg_entries;
 	pp_params.nid = dev_to_node(&mlxsw_pci->pdev->dev);
 	pp_params.dev = &mlxsw_pci->pdev->dev;
 	pp_params.napi = &q->u.cq.napi;
-- 
cgit v1.2.3-58-ga151


From 2c7857677d695166e913123b23ed7e55d80f72be Mon Sep 17 00:00:00 2001
From: Lingbo Kong <quic_lingbok@quicinc.com>
Date: Fri, 21 Jun 2024 15:58:09 +0530
Subject: wifi: ath12k: Fix pdev id sent to firmware for single phy devices

Pdev id from mac phy capabilities will be sent as a part of
HTT/WMI command to firmware. This causes issue with single pdev
devices where firmware does not respond to the WMI/HTT request
sent from host.

For single pdev devices firmware expects pdev id as 1 for 5 GHz/6 GHz
phy and 2 for 2 GHz band. Add wrapper ath12k_mac_get_target_pdev_id()
to help fetch right pdev for single pdev devices.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Lingbo Kong <quic_lingbok@quicinc.com>
Signed-off-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240621102809.3984004-1-quic_rgnanase@quicinc.com
---
 drivers/net/wireless/ath/ath12k/mac.c | 76 +++++++++++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath12k/mac.h |  1 +
 2 files changed, 77 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/mac.c b/drivers/net/wireless/ath/ath12k/mac.c
index 43801ff5bfd8..7475dddca8c5 100644
--- a/drivers/net/wireless/ath/ath12k/mac.c
+++ b/drivers/net/wireless/ath/ath12k/mac.c
@@ -673,6 +673,82 @@ static struct ath12k *ath12k_get_ar_by_vif(struct ieee80211_hw *hw,
 	return NULL;
 }
 
+static struct ath12k_vif *ath12k_mac_get_vif_up(struct ath12k *ar)
+{
+	struct ath12k_vif *arvif;
+
+	lockdep_assert_held(&ar->conf_mutex);
+	list_for_each_entry(arvif, &ar->arvifs, list) {
+		if (arvif->is_up)
+			return arvif;
+	}
+
+	return NULL;
+}
+
+static bool ath12k_mac_band_match(enum nl80211_band band1, enum WMI_HOST_WLAN_BAND band2)
+{
+	switch (band1) {
+	case NL80211_BAND_2GHZ:
+		if (band2 & WMI_HOST_WLAN_2G_CAP)
+			return true;
+		break;
+	case NL80211_BAND_5GHZ:
+	case NL80211_BAND_6GHZ:
+		if (band2 & WMI_HOST_WLAN_5G_CAP)
+			return true;
+		break;
+	default:
+		return false;
+	}
+
+	return false;
+}
+
+static u8 ath12k_mac_get_target_pdev_id_from_vif(struct ath12k_vif *arvif)
+{
+	struct ath12k *ar = arvif->ar;
+	struct ath12k_base *ab = ar->ab;
+	struct ieee80211_vif *vif = arvif->vif;
+	struct cfg80211_chan_def def;
+	enum nl80211_band band;
+	u8 pdev_id = ab->fw_pdev[0].pdev_id;
+	int i;
+
+	if (WARN_ON(ath12k_mac_vif_chan(vif, &def)))
+		return pdev_id;
+
+	band = def.chan->band;
+
+	for (i = 0; i < ab->fw_pdev_count; i++) {
+		if (ath12k_mac_band_match(band, ab->fw_pdev[i].supported_bands))
+			return ab->fw_pdev[i].pdev_id;
+	}
+
+	return pdev_id;
+}
+
+u8 ath12k_mac_get_target_pdev_id(struct ath12k *ar)
+{
+	struct ath12k_vif *arvif;
+	struct ath12k_base *ab = ar->ab;
+
+	if (!ab->hw_params->single_pdev_only)
+		return ar->pdev->pdev_id;
+
+	arvif = ath12k_mac_get_vif_up(ar);
+
+	/* fw_pdev array has pdev ids derived from phy capability
+	 * service ready event (pdev_and_hw_link_ids).
+	 * If no vif is active, return default first index.
+	 */
+	if (!arvif)
+		return ar->ab->fw_pdev[0].pdev_id;
+
+	/* If active vif is found, return the pdev id matching chandef band */
+	return ath12k_mac_get_target_pdev_id_from_vif(arvif);
+}
+
 static void ath12k_pdev_caps_update(struct ath12k *ar)
 {
 	struct ath12k_base *ab = ar->ab;
diff --git a/drivers/net/wireless/ath/ath12k/mac.h b/drivers/net/wireless/ath/ath12k/mac.h
index 2c6cb144827a..5efbb6822628 100644
--- a/drivers/net/wireless/ath/ath12k/mac.h
+++ b/drivers/net/wireless/ath/ath12k/mac.h
@@ -85,5 +85,6 @@ void ath12k_mac_handle_beacon_miss(struct ath12k *ar, u32 vdev_id);
 int ath12k_mac_vif_set_keepalive(struct ath12k_vif *arvif,
 				 enum wmi_sta_keepalive_method method,
 				 u32 interval);
+u8 ath12k_mac_get_target_pdev_id(struct ath12k *ar);
 
 #endif
-- 
cgit v1.2.3-58-ga151


From 1eeafd64c7b455381b77c546e41bc267e13e2809 Mon Sep 17 00:00:00 2001
From: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Date: Mon, 24 Jun 2024 20:24:18 +0530
Subject: wifi: ath12k: fix peer metadata parsing

Currently, the Rx data path only supports parsing peer metadata of version
zero. However, the QCN9274 platform configures the peer metadata version
as V1B. When V1B peer metadata is parsed using the version zero logic,
invalid data is populated, causing valid packets to be dropped. To address
this issue, refactor the peer metadata version and add the version based
parsing to populate the data from peer metadata correctly.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Fixes: 287033810990 ("wifi: ath12k: add support for peer meta data version")
Signed-off-by: Karthikeyan Periyasamy <quic_periyasa@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240624145418.2043461-1-quic_periyasa@quicinc.com
---
 drivers/net/wireless/ath/ath12k/dp.h       |  1 +
 drivers/net/wireless/ath/ath12k/dp_rx.c    | 40 +++++++++++++++++++++++++-----
 drivers/net/wireless/ath/ath12k/hal_desc.h | 26 +++++++++++++++++--
 drivers/net/wireless/ath/ath12k/hw.h       |  2 --
 drivers/net/wireless/ath/ath12k/wmi.c      |  6 +++--
 drivers/net/wireless/ath/ath12k/wmi.h      | 11 ++++++--
 6 files changed, 72 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/dp.h b/drivers/net/wireless/ath/ath12k/dp.h
index 742094545089..b77497c14ac4 100644
--- a/drivers/net/wireless/ath/ath12k/dp.h
+++ b/drivers/net/wireless/ath/ath12k/dp.h
@@ -333,6 +333,7 @@ struct ath12k_dp {
 	struct dp_srng reo_except_ring;
 	struct dp_srng reo_cmd_ring;
 	struct dp_srng reo_status_ring;
+	enum ath12k_peer_metadata_version peer_metadata_ver;
 	struct dp_srng reo_dst_ring[DP_REO_DST_RING_MAX];
 	struct dp_tx_ring tx_ring[DP_TCL_NUM_RING_MAX];
 	struct hal_wbm_idle_scatter_list scatter_list[DP_IDLE_SCATTER_BUFS_MAX];
diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index 54aea3c22311..d7bb412ad391 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -2583,6 +2583,29 @@ static void ath12k_dp_rx_process_received_packets(struct ath12k_base *ab,
 	rcu_read_unlock();
 }
 
+static u16 ath12k_dp_rx_get_peer_id(struct ath12k_base *ab,
+				    enum ath12k_peer_metadata_version ver,
+				    __le32 peer_metadata)
+{
+	switch (ver) {
+	default:
+		ath12k_warn(ab, "Unknown peer metadata version: %d", ver);
+		fallthrough;
+	case ATH12K_PEER_METADATA_V0:
+		return le32_get_bits(peer_metadata,
+				     RX_MPDU_DESC_META_DATA_V0_PEER_ID);
+	case ATH12K_PEER_METADATA_V1:
+		return le32_get_bits(peer_metadata,
+				     RX_MPDU_DESC_META_DATA_V1_PEER_ID);
+	case ATH12K_PEER_METADATA_V1A:
+		return le32_get_bits(peer_metadata,
+				     RX_MPDU_DESC_META_DATA_V1A_PEER_ID);
+	case ATH12K_PEER_METADATA_V1B:
+		return le32_get_bits(peer_metadata,
+				     RX_MPDU_DESC_META_DATA_V1B_PEER_ID);
+	}
+}
+
 int ath12k_dp_rx_process(struct ath12k_base *ab, int ring_id,
 			 struct napi_struct *napi, int budget)
 {
@@ -2611,6 +2634,8 @@ try_again:
 	ath12k_hal_srng_access_begin(ab, srng);
 
 	while ((desc = ath12k_hal_srng_dst_get_next_entry(ab, srng))) {
+		struct rx_mpdu_desc *mpdu_info;
+		struct rx_msdu_desc *msdu_info;
 		enum hal_reo_dest_ring_push_reason push_reason;
 		u32 cookie;
 
@@ -2658,16 +2683,19 @@ try_again:
 			continue;
 		}
 
-		rxcb->is_first_msdu = !!(le32_to_cpu(desc->rx_msdu_info.info0) &
+		msdu_info = &desc->rx_msdu_info;
+		mpdu_info = &desc->rx_mpdu_info;
+
+		rxcb->is_first_msdu = !!(le32_to_cpu(msdu_info->info0) &
 					 RX_MSDU_DESC_INFO0_FIRST_MSDU_IN_MPDU);
-		rxcb->is_last_msdu = !!(le32_to_cpu(desc->rx_msdu_info.info0) &
+		rxcb->is_last_msdu = !!(le32_to_cpu(msdu_info->info0) &
 					RX_MSDU_DESC_INFO0_LAST_MSDU_IN_MPDU);
-		rxcb->is_continuation = !!(le32_to_cpu(desc->rx_msdu_info.info0) &
+		rxcb->is_continuation = !!(le32_to_cpu(msdu_info->info0) &
 					   RX_MSDU_DESC_INFO0_MSDU_CONTINUATION);
 		rxcb->mac_id = mac_id;
-		rxcb->peer_id = le32_get_bits(desc->rx_mpdu_info.peer_meta_data,
-					      RX_MPDU_DESC_META_DATA_PEER_ID);
-		rxcb->tid = le32_get_bits(desc->rx_mpdu_info.info0,
+		rxcb->peer_id = ath12k_dp_rx_get_peer_id(ab, dp->peer_metadata_ver,
+							 mpdu_info->peer_meta_data);
+		rxcb->tid = le32_get_bits(mpdu_info->info0,
 					  RX_MPDU_DESC_INFO0_TID);
 
 		__skb_queue_tail(&msdu_list, msdu);
diff --git a/drivers/net/wireless/ath/ath12k/hal_desc.h b/drivers/net/wireless/ath/ath12k/hal_desc.h
index 02b7db06b24e..739f73370015 100644
--- a/drivers/net/wireless/ath/ath12k/hal_desc.h
+++ b/drivers/net/wireless/ath/ath12k/hal_desc.h
@@ -597,8 +597,30 @@ struct hal_tlv_64_hdr {
 #define RX_MPDU_DESC_INFO0_MPDU_QOS_CTRL_VALID	BIT(27)
 #define RX_MPDU_DESC_INFO0_TID			GENMASK(31, 28)
 
-/* TODO revisit after meta data is concluded */
-#define RX_MPDU_DESC_META_DATA_PEER_ID		GENMASK(15, 0)
+/* Peer Metadata classification */
+
+/* Version 0 */
+#define RX_MPDU_DESC_META_DATA_V0_PEER_ID	GENMASK(15, 0)
+#define RX_MPDU_DESC_META_DATA_V0_VDEV_ID	GENMASK(23, 16)
+
+/* Version 1 */
+#define RX_MPDU_DESC_META_DATA_V1_PEER_ID		GENMASK(13, 0)
+#define RX_MPDU_DESC_META_DATA_V1_LOGICAL_LINK_ID	GENMASK(15, 14)
+#define RX_MPDU_DESC_META_DATA_V1_VDEV_ID		GENMASK(23, 16)
+#define RX_MPDU_DESC_META_DATA_V1_LMAC_ID		GENMASK(25, 24)
+#define RX_MPDU_DESC_META_DATA_V1_DEVICE_ID		GENMASK(28, 26)
+
+/* Version 1A */
+#define RX_MPDU_DESC_META_DATA_V1A_PEER_ID		GENMASK(13, 0)
+#define RX_MPDU_DESC_META_DATA_V1A_VDEV_ID		GENMASK(21, 14)
+#define RX_MPDU_DESC_META_DATA_V1A_LOGICAL_LINK_ID	GENMASK(25, 22)
+#define RX_MPDU_DESC_META_DATA_V1A_DEVICE_ID		GENMASK(28, 26)
+
+/* Version 1B */
+#define RX_MPDU_DESC_META_DATA_V1B_PEER_ID	GENMASK(13, 0)
+#define RX_MPDU_DESC_META_DATA_V1B_VDEV_ID	GENMASK(21, 14)
+#define RX_MPDU_DESC_META_DATA_V1B_HW_LINK_ID	GENMASK(25, 22)
+#define RX_MPDU_DESC_META_DATA_V1B_DEVICE_ID	GENMASK(28, 26)
 
 struct rx_mpdu_desc {
 	__le32 info0; /* %RX_MPDU_DESC_INFO */
diff --git a/drivers/net/wireless/ath/ath12k/hw.h b/drivers/net/wireless/ath/ath12k/hw.h
index af33bf11416b..e792eb6b249b 100644
--- a/drivers/net/wireless/ath/ath12k/hw.h
+++ b/drivers/net/wireless/ath/ath12k/hw.h
@@ -78,8 +78,6 @@
 #define TARGET_NUM_WDS_ENTRIES		32
 #define TARGET_DMA_BURST_SIZE		1
 #define TARGET_RX_BATCHMODE		1
-#define TARGET_RX_PEER_METADATA_VER_V1A	2
-#define TARGET_RX_PEER_METADATA_VER_V1B	3
 #define TARGET_EMA_MAX_PROFILE_PERIOD	8
 
 #define ATH12K_HW_DEFAULT_QUEUE		0
diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 7c5976366615..9f6be557365e 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -233,7 +233,7 @@ void ath12k_wmi_init_qcn9274(struct ath12k_base *ab,
 	config->beacon_tx_offload_max_vdev += config->ema_max_vap_cnt;
 
 	if (test_bit(WMI_TLV_SERVICE_PEER_METADATA_V1A_V1B_SUPPORT, ab->wmi_ab.svc_map))
-		config->dp_peer_meta_data_ver = TARGET_RX_PEER_METADATA_VER_V1B;
+		config->peer_metadata_ver = ATH12K_PEER_METADATA_V1B;
 }
 
 void ath12k_wmi_init_wcn7850(struct ath12k_base *ab,
@@ -3498,7 +3498,7 @@ ath12k_wmi_copy_resource_config(struct ath12k_wmi_resource_config_params *wmi_cf
 	wmi_cfg->sched_params = cpu_to_le32(tg_cfg->sched_params);
 	wmi_cfg->twt_ap_pdev_count = cpu_to_le32(tg_cfg->twt_ap_pdev_count);
 	wmi_cfg->twt_ap_sta_count = cpu_to_le32(tg_cfg->twt_ap_sta_count);
-	wmi_cfg->flags2 = le32_encode_bits(tg_cfg->dp_peer_meta_data_ver,
+	wmi_cfg->flags2 = le32_encode_bits(tg_cfg->peer_metadata_ver,
 					   WMI_RSRC_CFG_FLAGS2_RX_PEER_METADATA_VERSION);
 	wmi_cfg->host_service_flags = cpu_to_le32(tg_cfg->is_reg_cc_ext_event_supported <<
 				WMI_RSRC_CFG_HOST_SVC_FLAG_REG_CC_EXT_SUPPORT_BIT);
@@ -3729,6 +3729,8 @@ int ath12k_wmi_cmd_init(struct ath12k_base *ab)
 	arg.num_band_to_mac = ab->num_radios;
 	ath12k_fill_band_to_mac_param(ab, arg.band_to_mac);
 
+	ab->dp.peer_metadata_ver = arg.res_cfg.peer_metadata_ver;
+
 	return ath12k_init_cmd_send(&wmi_ab->wmi[0], &arg);
 }
 
diff --git a/drivers/net/wireless/ath/ath12k/wmi.h b/drivers/net/wireless/ath/ath12k/wmi.h
index 0ac7accc45dc..f1f52175a52b 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.h
+++ b/drivers/net/wireless/ath/ath12k/wmi.h
@@ -2294,6 +2294,13 @@ struct ath12k_wmi_host_mem_chunk_arg {
 	u32 req_id;
 };
 
+enum ath12k_peer_metadata_version {
+	ATH12K_PEER_METADATA_V0,
+	ATH12K_PEER_METADATA_V1,
+	ATH12K_PEER_METADATA_V1A,
+	ATH12K_PEER_METADATA_V1B
+};
+
 struct ath12k_wmi_resource_config_arg {
 	u32 num_vdevs;
 	u32 num_peers;
@@ -2356,10 +2363,10 @@ struct ath12k_wmi_resource_config_arg {
 	u32 sched_params;
 	u32 twt_ap_pdev_count;
 	u32 twt_ap_sta_count;
-	bool is_reg_cc_ext_event_supported;
-	u8  dp_peer_meta_data_ver;
+	enum ath12k_peer_metadata_version peer_metadata_ver;
 	u32 ema_max_vap_cnt;
 	u32 ema_max_profile_period;
+	bool is_reg_cc_ext_event_supported;
 };
 
 struct ath12k_wmi_init_cmd_arg {
-- 
cgit v1.2.3-58-ga151


From dbb5265a5d7cca1cdba7736dba313ab7d07bc19d Mon Sep 17 00:00:00 2001
From: Arend van Spriel <arend.vanspriel@broadcom.com>
Date: Mon, 17 Jun 2024 14:26:09 +0200
Subject: wifi: brcmsmac: advertise MFP_CAPABLE to enable WPA3

After being asked about support for WPA3 for BCM43224 chipset it
was found that all it takes is setting the MFP_CAPABLE flag and
mac80211 will take care of all that is needed [1].

Link: https://lore.kernel.org/linux-wireless/20200526155909.5807-2-Larry.Finger@lwfinger.net/ [1]
Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Tested-by: Reijer Boekhoff <reijerboekhoff@protonmail.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://patch.msgid.link/20240617122609.349582-1-arend.vanspriel@broadcom.com
---
 drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
index 9ac6f3c4bdc7..d86f28b8bc60 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c
@@ -1090,6 +1090,7 @@ static int ieee_hw_init(struct ieee80211_hw *hw)
 	ieee80211_hw_set(hw, AMPDU_AGGREGATION);
 	ieee80211_hw_set(hw, SIGNAL_DBM);
 	ieee80211_hw_set(hw, REPORTS_TX_ACK_STATUS);
+	ieee80211_hw_set(hw, MFP_CAPABLE);
 
 	hw->extra_tx_headroom = brcms_c_get_header_len();
 	hw->queues = N_TX_QUEUES;
-- 
cgit v1.2.3-58-ga151


From f3f942d6e5889ea9d46cecf8fc0b7893b5ba9792 Mon Sep 17 00:00:00 2001
From: Alex Bee <knaerzche@gmail.com>
Date: Sat, 22 Jun 2024 23:54:16 +0200
Subject: wifi: brcmfmac: of: Support interrupts-extended

The currently existing of_property_present check for interrupts does not
cover all ways interrupts can be defined in a device tree, e.g.
"interrupts-extended".

In order to support all current and future ways that can be done, drop that
check and call of_irq_parse_one to figure out if an interrupt is defined
and irq_create_of_mapping for the actual mapping and let it be handled by
the interrupt subsystem.

Signed-off-by: Alex Bee <knaerzche@gmail.com>
Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://patch.msgid.link/20240622215416.659208-1-knaerzche@gmail.com
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
index e406e11481a6..fe4f65756105 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
@@ -70,6 +70,7 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
 {
 	struct brcmfmac_sdio_pd *sdio = &settings->bus.sdio;
 	struct device_node *root, *np = dev->of_node;
+	struct of_phandle_args oirq;
 	const char *prop;
 	int irq;
 	int err;
@@ -129,10 +130,10 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
 		sdio->drive_strength = val;
 
 	/* make sure there are interrupts defined in the node */
-	if (!of_property_present(np, "interrupts"))
+	if (of_irq_parse_one(np, 0, &oirq))
 		return;
 
-	irq = irq_of_parse_and_map(np, 0);
+	irq = irq_create_of_mapping(&oirq);
 	if (!irq) {
 		brcmf_err("interrupt could not be mapped\n");
 		return;
-- 
cgit v1.2.3-58-ga151


From ec2b9a5e11e51fea1bb04c1e7e471952e887e874 Mon Sep 17 00:00:00 2001
From: Matt Bobrowski <mattbobrowski@google.com>
Date: Tue, 25 Jun 2024 06:28:56 +0000
Subject: bpf: add missing check_func_arg_reg_off() to prevent out-of-bounds
 memory accesses

Currently, it's possible to pass in a modified CONST_PTR_TO_DYNPTR to
a global function as an argument. The adverse effects of this is that
BPF helpers can continue to make use of this modified
CONST_PTR_TO_DYNPTR from within the context of the global function,
which can unintentionally result in out-of-bounds memory accesses and
therefore compromise overall system stability i.e.

[  244.157771] BUG: KASAN: slab-out-of-bounds in bpf_dynptr_data+0x137/0x140
[  244.161345] Read of size 8 at addr ffff88810914be68 by task test_progs/302
[  244.167151] CPU: 0 PID: 302 Comm: test_progs Tainted: G O E 6.10.0-rc3-00131-g66b586715063 #533
[  244.174318] Call Trace:
[  244.175787]  <TASK>
[  244.177356]  dump_stack_lvl+0x66/0xa0
[  244.179531]  print_report+0xce/0x670
[  244.182314]  ? __virt_addr_valid+0x200/0x3e0
[  244.184908]  kasan_report+0xd7/0x110
[  244.187408]  ? bpf_dynptr_data+0x137/0x140
[  244.189714]  ? bpf_dynptr_data+0x137/0x140
[  244.192020]  bpf_dynptr_data+0x137/0x140
[  244.194264]  bpf_prog_b02a02fdd2bdc5fa_global_call_bpf_dynptr_data+0x22/0x26
[  244.198044]  bpf_prog_b0fe7b9d7dc3abde_callback_adjust_bpf_dynptr_reg_off+0x1f/0x23
[  244.202136]  bpf_user_ringbuf_drain+0x2c7/0x570
[  244.204744]  ? 0xffffffffc0009e58
[  244.206593]  ? __pfx_bpf_user_ringbuf_drain+0x10/0x10
[  244.209795]  bpf_prog_33ab33f6a804ba2d_user_ringbuf_callback_const_ptr_to_dynptr_reg_off+0x47/0x4b
[  244.215922]  bpf_trampoline_6442502480+0x43/0xe3
[  244.218691]  __x64_sys_prlimit64+0x9/0xf0
[  244.220912]  do_syscall_64+0xc1/0x1d0
[  244.223043]  entry_SYSCALL_64_after_hwframe+0x77/0x7f
[  244.226458] RIP: 0033:0x7ffa3eb8f059
[  244.228582] Code: 08 89 e8 5b 5d c3 66 2e 0f 1f 84 00 00 00 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 8f 1d 0d 00 f7 d8 64 89 01 48
[  244.241307] RSP: 002b:00007ffa3e9c6eb8 EFLAGS: 00000206 ORIG_RAX: 000000000000012e
[  244.246474] RAX: ffffffffffffffda RBX: 00007ffa3e9c7cdc RCX: 00007ffa3eb8f059
[  244.250478] RDX: 00007ffa3eb162b4 RSI: 0000000000000000 RDI: 00007ffa3e9c7fb0
[  244.255396] RBP: 00007ffa3e9c6ed0 R08: 00007ffa3e9c76c0 R09: 0000000000000000
[  244.260195] R10: 0000000000000000 R11: 0000000000000206 R12: ffffffffffffff80
[  244.264201] R13: 000000000000001c R14: 00007ffc5d6b4260 R15: 00007ffa3e1c7000
[  244.268303]  </TASK>

Add a check_func_arg_reg_off() to the path in which the BPF verifier
verifies the arguments of global function arguments, specifically
those which take an argument of type ARG_PTR_TO_DYNPTR |
MEM_RDONLY. Also, process_dynptr_func() doesn't appear to perform any
explicit and strict type matching on the supplied register type, so
let's also enforce that a register either type PTR_TO_STACK or
CONST_PTR_TO_DYNPTR is by the caller.

Reported-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Matt Bobrowski <mattbobrowski@google.com>
Link: https://lore.kernel.org/r/20240625062857.92760-1-mattbobrowski@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 3f6be4923655..d3927d819465 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -7726,6 +7726,13 @@ static int process_dynptr_func(struct bpf_verifier_env *env, int regno, int insn
 	struct bpf_reg_state *regs = cur_regs(env), *reg = &regs[regno];
 	int err;
 
+	if (reg->type != PTR_TO_STACK && reg->type != CONST_PTR_TO_DYNPTR) {
+		verbose(env,
+			"arg#%d expected pointer to stack or const struct bpf_dynptr\n",
+			regno);
+		return -EINVAL;
+	}
+
 	/* MEM_UNINIT and MEM_RDONLY are exclusive, when applied to an
 	 * ARG_PTR_TO_DYNPTR (or ARG_PTR_TO_DYNPTR | DYNPTR_TYPE_*):
 	 */
@@ -9475,6 +9482,10 @@ static int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog,
 				return -EINVAL;
 			}
 		} else if (arg->arg_type == (ARG_PTR_TO_DYNPTR | MEM_RDONLY)) {
+			ret = check_func_arg_reg_off(env, reg, regno, ARG_PTR_TO_DYNPTR);
+			if (ret)
+				return ret;
+
 			ret = process_dynptr_func(env, regno, -1, arg->arg_type, 0);
 			if (ret)
 				return ret;
@@ -11976,12 +11987,6 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 			enum bpf_arg_type dynptr_arg_type = ARG_PTR_TO_DYNPTR;
 			int clone_ref_obj_id = 0;
 
-			if (reg->type != PTR_TO_STACK &&
-			    reg->type != CONST_PTR_TO_DYNPTR) {
-				verbose(env, "arg#%d expected pointer to stack or dynptr_ptr\n", i);
-				return -EINVAL;
-			}
-
 			if (reg->type == CONST_PTR_TO_DYNPTR)
 				dynptr_arg_type |= MEM_RDONLY;
 
-- 
cgit v1.2.3-58-ga151


From aa293983d2020390e286544b120f3cd0a3d40749 Mon Sep 17 00:00:00 2001
From: Matt Bobrowski <mattbobrowski@google.com>
Date: Tue, 25 Jun 2024 06:28:57 +0000
Subject: bpf: add new negative selftests to cover missing
 check_func_arg_reg_off() and reg->type check

Add new negative selftests which are intended to cover the
out-of-bounds memory access that could be performed on a
CONST_PTR_TO_DYNPTR within functions taking a ARG_PTR_TO_DYNPTR |
MEM_RDONLY as an argument, and acceptance of invalid register types
i.e. PTR_TO_BTF_ID within functions taking a ARG_PTR_TO_DYNPTR |
MEM_RDONLY.

Reported-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Matt Bobrowski <mattbobrowski@google.com>
Link: https://lore.kernel.org/r/20240625062857.92760-2-mattbobrowski@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/dynptr_fail.c    | 24 ++++++++++++++++++++++
 .../selftests/bpf/progs/test_kfunc_dynptr_param.c  |  2 +-
 .../bpf/progs/test_kfunc_param_nullable.c          |  2 +-
 .../selftests/bpf/progs/user_ringbuf_fail.c        | 22 ++++++++++++++++++++
 4 files changed, 48 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/dynptr_fail.c b/tools/testing/selftests/bpf/progs/dynptr_fail.c
index 64cc9d936a13..e35bc1eac52a 100644
--- a/tools/testing/selftests/bpf/progs/dynptr_fail.c
+++ b/tools/testing/selftests/bpf/progs/dynptr_fail.c
@@ -1686,3 +1686,27 @@ int test_dynptr_skb_small_buff(struct __sk_buff *skb)
 
 	return !!data;
 }
+
+__noinline long global_call_bpf_dynptr(const struct bpf_dynptr *dynptr)
+{
+	long ret = 0;
+	/* Avoid leaving this global function empty to avoid having the compiler
+	 * optimize away the call to this global function.
+	 */
+	__sink(ret);
+	return ret;
+}
+
+SEC("?raw_tp")
+__failure __msg("arg#1 expected pointer to stack or const struct bpf_dynptr")
+int test_dynptr_reg_type(void *ctx)
+{
+	struct task_struct *current = NULL;
+	/* R1 should be holding a PTR_TO_BTF_ID, so this shouldn't be a
+	 * reg->type that can be passed to a function accepting a
+	 * ARG_PTR_TO_DYNPTR | MEM_RDONLY. process_dynptr_func() should catch
+	 * this.
+	 */
+	global_call_bpf_dynptr((const struct bpf_dynptr *)current);
+	return 0;
+}
diff --git a/tools/testing/selftests/bpf/progs/test_kfunc_dynptr_param.c b/tools/testing/selftests/bpf/progs/test_kfunc_dynptr_param.c
index 2dde8e3fe4c9..e68667aec6a6 100644
--- a/tools/testing/selftests/bpf/progs/test_kfunc_dynptr_param.c
+++ b/tools/testing/selftests/bpf/progs/test_kfunc_dynptr_param.c
@@ -45,7 +45,7 @@ int BPF_PROG(not_valid_dynptr, int cmd, union bpf_attr *attr, unsigned int size)
 }
 
 SEC("?lsm.s/bpf")
-__failure __msg("arg#0 expected pointer to stack or dynptr_ptr")
+__failure __msg("arg#1 expected pointer to stack or const struct bpf_dynptr")
 int BPF_PROG(not_ptr_to_stack, int cmd, union bpf_attr *attr, unsigned int size)
 {
 	unsigned long val = 0;
diff --git a/tools/testing/selftests/bpf/progs/test_kfunc_param_nullable.c b/tools/testing/selftests/bpf/progs/test_kfunc_param_nullable.c
index 7c75e9b8f455..7ac7e1de34d8 100644
--- a/tools/testing/selftests/bpf/progs/test_kfunc_param_nullable.c
+++ b/tools/testing/selftests/bpf/progs/test_kfunc_param_nullable.c
@@ -29,7 +29,7 @@ int kfunc_dynptr_nullable_test2(struct __sk_buff *skb)
 }
 
 SEC("tc")
-__failure __msg("expected pointer to stack or dynptr_ptr")
+__failure __msg("expected pointer to stack or const struct bpf_dynptr")
 int kfunc_dynptr_nullable_test3(struct __sk_buff *skb)
 {
 	struct bpf_dynptr data;
diff --git a/tools/testing/selftests/bpf/progs/user_ringbuf_fail.c b/tools/testing/selftests/bpf/progs/user_ringbuf_fail.c
index 11ab25c42c36..54de0389f878 100644
--- a/tools/testing/selftests/bpf/progs/user_ringbuf_fail.c
+++ b/tools/testing/selftests/bpf/progs/user_ringbuf_fail.c
@@ -221,3 +221,25 @@ int user_ringbuf_callback_reinit_dynptr_ringbuf(void *ctx)
 	bpf_user_ringbuf_drain(&user_ringbuf, try_reinit_dynptr_ringbuf, NULL, 0);
 	return 0;
 }
+
+__noinline long global_call_bpf_dynptr_data(struct bpf_dynptr *dynptr)
+{
+	bpf_dynptr_data(dynptr, 0xA, 0xA);
+	return 0;
+}
+
+static long callback_adjust_bpf_dynptr_reg_off(struct bpf_dynptr *dynptr,
+					       void *ctx)
+{
+	global_call_bpf_dynptr_data(dynptr += 1024);
+	return 0;
+}
+
+SEC("?raw_tp")
+__failure __msg("dereference of modified dynptr_ptr ptr R1 off=16384 disallowed")
+int user_ringbuf_callback_const_ptr_to_dynptr_reg_off(void *ctx)
+{
+	bpf_user_ringbuf_drain(&user_ringbuf,
+			       callback_adjust_bpf_dynptr_reg_off, NULL, 0);
+	return 0;
+}
-- 
cgit v1.2.3-58-ga151


From d07980f7373b4c57c85478f8e815cc7b9b394c05 Mon Sep 17 00:00:00 2001
From: Ma Ke <make24@iscas.ac.cn>
Date: Sun, 23 Jun 2024 21:17:53 +0800
Subject: selftests/bpf: Don't close(-1) in serial_test_fexit_stress()

Guard close() with extra link_fd[i] > 0 and fexit_fd[i] > 0
check to prevent close(-1).

Signed-off-by: Ma Ke <make24@iscas.ac.cn>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240623131753.2133829-1-make24@iscas.ac.cn
---
 tools/testing/selftests/bpf/prog_tests/fexit_stress.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/fexit_stress.c b/tools/testing/selftests/bpf/prog_tests/fexit_stress.c
index 596536def43d..49b1ffc9af1f 100644
--- a/tools/testing/selftests/bpf/prog_tests/fexit_stress.c
+++ b/tools/testing/selftests/bpf/prog_tests/fexit_stress.c
@@ -50,9 +50,9 @@ void serial_test_fexit_stress(void)
 
 out:
 	for (i = 0; i < bpf_max_tramp_links; i++) {
-		if (link_fd[i])
+		if (link_fd[i] > 0)
 			close(link_fd[i]);
-		if (fexit_fd[i])
+		if (fexit_fd[i] > 0)
 			close(fexit_fd[i]);
 	}
 	free(fd);
-- 
cgit v1.2.3-58-ga151


From 0f31c2c61f6923747628c65a0fe36b2d4d7e21b0 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Mon, 24 Jun 2024 20:29:03 +0100
Subject: libbpf: Fix clang compilation error in btf_relocate.c

When building with clang for ARCH=i386, the following errors are
observed:

  CC      kernel/bpf/btf_relocate.o
./tools/lib/bpf/btf_relocate.c:206:23: error: implicit truncation from 'int' to a one-bit wide bit-field changes value from 1 to -1 [-Werror,-Wsingle-bit-bitfield-constant-conversion]
  206 |                 info[id].needs_size = true;
      |                                     ^ ~
./tools/lib/bpf/btf_relocate.c:256:25: error: implicit truncation from 'int' to a one-bit wide bit-field changes value from 1 to -1 [-Werror,-Wsingle-bit-bitfield-constant-conversion]
  256 |                         base_info.needs_size = true;
      |                                              ^ ~
2 errors generated.

The problem is we use 1-bit, 31-bit bitfields in a signed int.
Changing to

	bool needs_size: 1;
	unsigned int size:31;

...resolves the error and pahole reports that 4 bytes are used
for the underlying representation:

$ pahole btf_name_info tools/lib/bpf/btf_relocate.o
struct btf_name_info {
	const char  *              name;                 /*     0     8 */
	unsigned int               needs_size:1;         /*     8: 0  4 */
	unsigned int               size:31;              /*     8: 1  4 */
	__u32                      id;                   /*    12     4 */

	/* size: 16, cachelines: 1, members: 4 */
	/* last cacheline: 16 bytes */
};

Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240624192903.854261-1-alan.maguire@oracle.com
---
 tools/lib/bpf/btf_relocate.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/lib/bpf/btf_relocate.c b/tools/lib/bpf/btf_relocate.c
index 2281dbbafa11..17f8b32f94a0 100644
--- a/tools/lib/bpf/btf_relocate.c
+++ b/tools/lib/bpf/btf_relocate.c
@@ -58,8 +58,8 @@ struct btf_relocate {
 struct btf_name_info {
 	const char *name;
 	/* set when search requires a size match */
-	int needs_size:1,
-	    size:31;
+	bool needs_size: 1;
+	unsigned int size: 31;
 	__u32 id;
 };
 
-- 
cgit v1.2.3-58-ga151


From a12978712d9001b060bcc10eaae42ad5102abe2b Mon Sep 17 00:00:00 2001
From: Jiri Olsa <jolsa@kernel.org>
Date: Wed, 26 Jun 2024 15:47:19 +0200
Subject: selftests/bpf: Move ARRAY_SIZE to bpf_misc.h

ARRAY_SIZE is used on multiple places, move its definition in
bpf_misc.h header.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Alan Maguire <alan.maguire@oracle.com>
Link: https://lore.kernel.org/bpf/20240626134719.3893748-1-jolsa@kernel.org
---
 tools/testing/selftests/bpf/progs/bpf_misc.h                   | 4 ++++
 tools/testing/selftests/bpf/progs/iters.c                      | 2 --
 tools/testing/selftests/bpf/progs/kprobe_multi_session.c       | 3 +--
 tools/testing/selftests/bpf/progs/linked_list.c                | 5 +----
 tools/testing/selftests/bpf/progs/netif_receive_skb.c          | 5 +----
 tools/testing/selftests/bpf/progs/profiler.inc.h               | 5 +----
 tools/testing/selftests/bpf/progs/setget_sockopt.c             | 5 +----
 tools/testing/selftests/bpf/progs/test_bpf_ma.c                | 4 ----
 tools/testing/selftests/bpf/progs/test_sysctl_loop1.c          | 5 +----
 tools/testing/selftests/bpf/progs/test_sysctl_loop2.c          | 5 +----
 tools/testing/selftests/bpf/progs/test_sysctl_prog.c           | 5 +----
 tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.c  | 1 +
 tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.h  | 2 --
 tools/testing/selftests/bpf/progs/verifier_subprog_precision.c | 2 --
 14 files changed, 13 insertions(+), 40 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/bpf_misc.h b/tools/testing/selftests/bpf/progs/bpf_misc.h
index c0280bd2f340..81097a3f15eb 100644
--- a/tools/testing/selftests/bpf/progs/bpf_misc.h
+++ b/tools/testing/selftests/bpf/progs/bpf_misc.h
@@ -140,4 +140,8 @@
 /* make it look to compiler like value is read and written */
 #define __sink(expr) asm volatile("" : "+g"(expr))
 
+#ifndef ARRAY_SIZE
+#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
+#endif
+
 #endif
diff --git a/tools/testing/selftests/bpf/progs/iters.c b/tools/testing/selftests/bpf/progs/iters.c
index fe65e0952a1e..16bdc3e25591 100644
--- a/tools/testing/selftests/bpf/progs/iters.c
+++ b/tools/testing/selftests/bpf/progs/iters.c
@@ -7,8 +7,6 @@
 #include "bpf_misc.h"
 #include "bpf_compiler.h"
 
-#define ARRAY_SIZE(x) (int)(sizeof(x) / sizeof((x)[0]))
-
 static volatile int zero = 0;
 
 int my_pid;
diff --git a/tools/testing/selftests/bpf/progs/kprobe_multi_session.c b/tools/testing/selftests/bpf/progs/kprobe_multi_session.c
index bbba9eb46551..bd8b7fb7061e 100644
--- a/tools/testing/selftests/bpf/progs/kprobe_multi_session.c
+++ b/tools/testing/selftests/bpf/progs/kprobe_multi_session.c
@@ -4,8 +4,7 @@
 #include <bpf/bpf_tracing.h>
 #include <stdbool.h>
 #include "bpf_kfuncs.h"
-
-#define ARRAY_SIZE(x) (int)(sizeof(x) / sizeof((x)[0]))
+#include "bpf_misc.h"
 
 char _license[] SEC("license") = "GPL";
 
diff --git a/tools/testing/selftests/bpf/progs/linked_list.c b/tools/testing/selftests/bpf/progs/linked_list.c
index f69bf3e30321..421f40835acd 100644
--- a/tools/testing/selftests/bpf/progs/linked_list.c
+++ b/tools/testing/selftests/bpf/progs/linked_list.c
@@ -4,10 +4,7 @@
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_core_read.h>
 #include "bpf_experimental.h"
-
-#ifndef ARRAY_SIZE
-#define ARRAY_SIZE(x) (int)(sizeof(x) / sizeof((x)[0]))
-#endif
+#include "bpf_misc.h"
 
 #include "linked_list.h"
 
diff --git a/tools/testing/selftests/bpf/progs/netif_receive_skb.c b/tools/testing/selftests/bpf/progs/netif_receive_skb.c
index c0062645fc68..9e067dcbf607 100644
--- a/tools/testing/selftests/bpf/progs/netif_receive_skb.c
+++ b/tools/testing/selftests/bpf/progs/netif_receive_skb.c
@@ -5,6 +5,7 @@
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
 #include <bpf/bpf_core_read.h>
+#include "bpf_misc.h"
 
 #include <errno.h>
 
@@ -23,10 +24,6 @@ bool skip = false;
 #define BADPTR			0
 #endif
 
-#ifndef ARRAY_SIZE
-#define ARRAY_SIZE(x)	(sizeof(x) / sizeof((x)[0]))
-#endif
-
 struct {
 	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
 	__uint(max_entries, 1);
diff --git a/tools/testing/selftests/bpf/progs/profiler.inc.h b/tools/testing/selftests/bpf/progs/profiler.inc.h
index 6957d9f2805e..8bd1ebd7d6af 100644
--- a/tools/testing/selftests/bpf/progs/profiler.inc.h
+++ b/tools/testing/selftests/bpf/progs/profiler.inc.h
@@ -9,6 +9,7 @@
 #include "err.h"
 #include "bpf_experimental.h"
 #include "bpf_compiler.h"
+#include "bpf_misc.h"
 
 #ifndef NULL
 #define NULL 0
@@ -133,10 +134,6 @@ struct {
 	__uint(max_entries, 16);
 } disallowed_exec_inodes SEC(".maps");
 
-#ifndef ARRAY_SIZE
-#define ARRAY_SIZE(arr) (int)(sizeof(arr) / sizeof(arr[0]))
-#endif
-
 static INLINE bool IS_ERR(const void* ptr)
 {
 	return IS_ERR_VALUE((unsigned long)ptr);
diff --git a/tools/testing/selftests/bpf/progs/setget_sockopt.c b/tools/testing/selftests/bpf/progs/setget_sockopt.c
index 7a438600ae98..60518aed1ffc 100644
--- a/tools/testing/selftests/bpf/progs/setget_sockopt.c
+++ b/tools/testing/selftests/bpf/progs/setget_sockopt.c
@@ -6,10 +6,7 @@
 #include <bpf/bpf_core_read.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
-
-#ifndef ARRAY_SIZE
-#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
-#endif
+#include "bpf_misc.h"
 
 extern unsigned long CONFIG_HZ __kconfig;
 
diff --git a/tools/testing/selftests/bpf/progs/test_bpf_ma.c b/tools/testing/selftests/bpf/progs/test_bpf_ma.c
index 3494ca30fa7f..4a4e0b8d9b72 100644
--- a/tools/testing/selftests/bpf/progs/test_bpf_ma.c
+++ b/tools/testing/selftests/bpf/progs/test_bpf_ma.c
@@ -7,10 +7,6 @@
 #include "bpf_experimental.h"
 #include "bpf_misc.h"
 
-#ifndef ARRAY_SIZE
-#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
-#endif
-
 struct generic_map_value {
 	void *data;
 };
diff --git a/tools/testing/selftests/bpf/progs/test_sysctl_loop1.c b/tools/testing/selftests/bpf/progs/test_sysctl_loop1.c
index 7f74077d6622..548660e299a5 100644
--- a/tools/testing/selftests/bpf/progs/test_sysctl_loop1.c
+++ b/tools/testing/selftests/bpf/progs/test_sysctl_loop1.c
@@ -10,10 +10,7 @@
 #include <bpf/bpf_helpers.h>
 
 #include "bpf_compiler.h"
-
-#ifndef ARRAY_SIZE
-#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
-#endif
+#include "bpf_misc.h"
 
 /* tcp_mem sysctl has only 3 ints, but this test is doing TCP_MEM_LOOPS */
 #define TCP_MEM_LOOPS 28  /* because 30 doesn't fit into 512 bytes of stack */
diff --git a/tools/testing/selftests/bpf/progs/test_sysctl_loop2.c b/tools/testing/selftests/bpf/progs/test_sysctl_loop2.c
index 68a75436e8af..81249d119a8b 100644
--- a/tools/testing/selftests/bpf/progs/test_sysctl_loop2.c
+++ b/tools/testing/selftests/bpf/progs/test_sysctl_loop2.c
@@ -10,10 +10,7 @@
 #include <bpf/bpf_helpers.h>
 
 #include "bpf_compiler.h"
-
-#ifndef ARRAY_SIZE
-#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
-#endif
+#include "bpf_misc.h"
 
 /* tcp_mem sysctl has only 3 ints, but this test is doing TCP_MEM_LOOPS */
 #define TCP_MEM_LOOPS 20  /* because 30 doesn't fit into 512 bytes of stack */
diff --git a/tools/testing/selftests/bpf/progs/test_sysctl_prog.c b/tools/testing/selftests/bpf/progs/test_sysctl_prog.c
index efc3c61f7852..bbdd08764789 100644
--- a/tools/testing/selftests/bpf/progs/test_sysctl_prog.c
+++ b/tools/testing/selftests/bpf/progs/test_sysctl_prog.c
@@ -10,6 +10,7 @@
 #include <bpf/bpf_helpers.h>
 
 #include "bpf_compiler.h"
+#include "bpf_misc.h"
 
 /* Max supported length of a string with unsigned long in base 10 (pow2 - 1). */
 #define MAX_ULONG_STR_LEN 0xF
@@ -17,10 +18,6 @@
 /* Max supported length of sysctl value string (pow2). */
 #define MAX_VALUE_STR_LEN 0x40
 
-#ifndef ARRAY_SIZE
-#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
-#endif
-
 const char tcp_mem_name[] = "net/ipv4/tcp_mem";
 static __always_inline int is_tcp_mem(struct bpf_sysctl *ctx)
 {
diff --git a/tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.c b/tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.c
index c8e4553648bf..44ee0d037f95 100644
--- a/tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.c
+++ b/tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.c
@@ -9,6 +9,7 @@
 #include "bpf_kfuncs.h"
 #include "test_siphash.h"
 #include "test_tcp_custom_syncookie.h"
+#include "bpf_misc.h"
 
 #define MAX_PACKET_OFF 0xffff
 
diff --git a/tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.h b/tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.h
index 29a6a53cf229..f8b1b7e68d2e 100644
--- a/tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.h
+++ b/tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.h
@@ -7,8 +7,6 @@
 #define __packed __attribute__((__packed__))
 #define __force
 
-#define ARRAY_SIZE(arr)	(sizeof(arr) / sizeof((arr)[0]))
-
 #define swap(a, b)				\
 	do {					\
 		typeof(a) __tmp = (a);		\
diff --git a/tools/testing/selftests/bpf/progs/verifier_subprog_precision.c b/tools/testing/selftests/bpf/progs/verifier_subprog_precision.c
index 4a58e0398e72..6a6fad625f7e 100644
--- a/tools/testing/selftests/bpf/progs/verifier_subprog_precision.c
+++ b/tools/testing/selftests/bpf/progs/verifier_subprog_precision.c
@@ -8,8 +8,6 @@
 #include "bpf_misc.h"
 #include <../../../tools/include/linux/filter.h>
 
-#define ARRAY_SIZE(x) (sizeof(x) / sizeof(x[0]))
-
 int vals[] SEC(".data.vals") = {1, 2, 3, 4};
 
 __naked __noinline __used
-- 
cgit v1.2.3-58-ga151


From 0065199f4303bc2b05467e64779d392f9a29d61d Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Thu, 20 Jun 2024 13:58:19 +0800
Subject: wifi: rtw89: wow: append security header offset for different cipher

When creating EAPOL_KEY, SA_QUERY and ARP_RSP packet offload, we need
to append security header offset for different cipher as required by
the firmware. Only 8852A, 8852B, 8852BT and 8851B need it.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240620055825.17592-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c  | 67 +++++++++++++++++++-------------
 drivers/net/wireless/realtek/rtw89/wow.c | 22 +++++++----
 drivers/net/wireless/realtek/rtw89/wow.h | 30 ++++++++++++++
 3 files changed, 85 insertions(+), 34 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 23204b2706c9..3c666fd0f8ed 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -13,22 +13,20 @@
 #include "ps.h"
 #include "reg.h"
 #include "util.h"
+#include "wow.h"
 
 struct rtw89_eapol_2_of_2 {
-	struct ieee80211_hdr_3addr hdr;
 	u8 gtkbody[14];
 	u8 key_des_ver;
 	u8 rsvd[92];
-} __packed __aligned(2);
+} __packed;
 
 struct rtw89_sa_query {
-	struct ieee80211_hdr_3addr hdr;
 	u8 category;
 	u8 action;
-} __packed __aligned(2);
+} __packed;
 
 struct rtw89_arp_rsp {
-	struct ieee80211_hdr_3addr addr;
 	u8 llc_hdr[sizeof(rfc1042_header)];
 	__be16 llc_type;
 	struct arphdr arp_hdr;
@@ -36,7 +34,7 @@ struct rtw89_arp_rsp {
 	__be32 sender_ip;
 	u8 target_hw[ETH_ALEN];
 	__be32 target_ip;
-} __packed __aligned(2);
+} __packed;
 
 static const u8 mss_signature[] = {0x4D, 0x53, 0x53, 0x4B, 0x50, 0x4F, 0x4F, 0x4C};
 
@@ -2197,8 +2195,10 @@ static struct sk_buff *rtw89_eapol_get(struct rtw89_dev *rtwdev,
 				     0x8E, 0x01, 0x03, 0x00, 0x5F, 0x02, 0x03};
 	struct ieee80211_vif *vif = rtwvif_to_vif(rtwvif);
 	struct ieee80211_bss_conf *bss_conf = &vif->bss_conf;
+	u8 sec_hdr_len = rtw89_wow_get_sec_hdr_len(rtwdev);
 	struct rtw89_wow_param *rtw_wow = &rtwdev->wow;
 	struct rtw89_eapol_2_of_2 *eapol_pkt;
+	struct ieee80211_hdr_3addr *hdr;
 	struct sk_buff *skb;
 	u8 key_des_ver;
 
@@ -2211,17 +2211,21 @@ static struct sk_buff *rtw89_eapol_get(struct rtw89_dev *rtwdev,
 	else
 		key_des_ver = 0;
 
-	skb = dev_alloc_skb(sizeof(*eapol_pkt));
+	skb = dev_alloc_skb(sizeof(*hdr) + sec_hdr_len + sizeof(*eapol_pkt));
 	if (!skb)
 		return NULL;
 
+	hdr = skb_put_zero(skb, sizeof(*hdr));
+	hdr->frame_control = cpu_to_le16(IEEE80211_FTYPE_DATA |
+					 IEEE80211_FCTL_TODS |
+					 IEEE80211_FCTL_PROTECTED);
+	ether_addr_copy(hdr->addr1, bss_conf->bssid);
+	ether_addr_copy(hdr->addr2, vif->addr);
+	ether_addr_copy(hdr->addr3, bss_conf->bssid);
+
+	skb_put_zero(skb, sec_hdr_len);
+
 	eapol_pkt = skb_put_zero(skb, sizeof(*eapol_pkt));
-	eapol_pkt->hdr.frame_control = cpu_to_le16(IEEE80211_FTYPE_DATA |
-						   IEEE80211_FCTL_TODS |
-						   IEEE80211_FCTL_PROTECTED);
-	ether_addr_copy(eapol_pkt->hdr.addr1, bss_conf->bssid);
-	ether_addr_copy(eapol_pkt->hdr.addr2, vif->addr);
-	ether_addr_copy(eapol_pkt->hdr.addr3, bss_conf->bssid);
 	memcpy(eapol_pkt->gtkbody, gtkbody, sizeof(gtkbody));
 	eapol_pkt->key_des_ver = key_des_ver;
 
@@ -2233,20 +2237,26 @@ static struct sk_buff *rtw89_sa_query_get(struct rtw89_dev *rtwdev,
 {
 	struct ieee80211_vif *vif = rtwvif_to_vif(rtwvif);
 	struct ieee80211_bss_conf *bss_conf = &vif->bss_conf;
+	u8 sec_hdr_len = rtw89_wow_get_sec_hdr_len(rtwdev);
+	struct ieee80211_hdr_3addr *hdr;
 	struct rtw89_sa_query *sa_query;
 	struct sk_buff *skb;
 
-	skb = dev_alloc_skb(sizeof(*sa_query));
+	skb = dev_alloc_skb(sizeof(*hdr) + sec_hdr_len + sizeof(*sa_query));
 	if (!skb)
 		return NULL;
 
+	hdr = skb_put_zero(skb, sizeof(*hdr));
+	hdr->frame_control = cpu_to_le16(IEEE80211_FTYPE_MGMT |
+					 IEEE80211_STYPE_ACTION |
+					 IEEE80211_FCTL_PROTECTED);
+	ether_addr_copy(hdr->addr1, bss_conf->bssid);
+	ether_addr_copy(hdr->addr2, vif->addr);
+	ether_addr_copy(hdr->addr3, bss_conf->bssid);
+
+	skb_put_zero(skb, sec_hdr_len);
+
 	sa_query = skb_put_zero(skb, sizeof(*sa_query));
-	sa_query->hdr.frame_control = cpu_to_le16(IEEE80211_FTYPE_MGMT |
-						  IEEE80211_STYPE_ACTION |
-						  IEEE80211_FCTL_PROTECTED);
-	ether_addr_copy(sa_query->hdr.addr1, bss_conf->bssid);
-	ether_addr_copy(sa_query->hdr.addr2, vif->addr);
-	ether_addr_copy(sa_query->hdr.addr3, bss_conf->bssid);
 	sa_query->category = WLAN_CATEGORY_SA_QUERY;
 	sa_query->action = WLAN_ACTION_SA_QUERY_RESPONSE;
 
@@ -2256,17 +2266,19 @@ static struct sk_buff *rtw89_sa_query_get(struct rtw89_dev *rtwdev,
 static struct sk_buff *rtw89_arp_response_get(struct rtw89_dev *rtwdev,
 					      struct rtw89_vif *rtwvif)
 {
+	u8 sec_hdr_len = rtw89_wow_get_sec_hdr_len(rtwdev);
 	struct rtw89_wow_param *rtw_wow = &rtwdev->wow;
+	struct ieee80211_hdr_3addr *hdr;
 	struct rtw89_arp_rsp *arp_skb;
 	struct arphdr *arp_hdr;
 	struct sk_buff *skb;
 	__le16 fc;
 
-	skb = dev_alloc_skb(sizeof(struct rtw89_arp_rsp));
+	skb = dev_alloc_skb(sizeof(*hdr) + sec_hdr_len + sizeof(*arp_skb));
 	if (!skb)
 		return NULL;
 
-	arp_skb = skb_put_zero(skb, sizeof(*arp_skb));
+	hdr = skb_put_zero(skb, sizeof(*hdr));
 
 	if (rtw_wow->ptk_alg)
 		fc = cpu_to_le16(IEEE80211_FTYPE_DATA | IEEE80211_FCTL_TODS |
@@ -2274,11 +2286,14 @@ static struct sk_buff *rtw89_arp_response_get(struct rtw89_dev *rtwdev,
 	else
 		fc = cpu_to_le16(IEEE80211_FTYPE_DATA | IEEE80211_FCTL_TODS);
 
-	arp_skb->addr.frame_control = fc;
-	ether_addr_copy(arp_skb->addr.addr1, rtwvif->bssid);
-	ether_addr_copy(arp_skb->addr.addr2, rtwvif->mac_addr);
-	ether_addr_copy(arp_skb->addr.addr3, rtwvif->bssid);
+	hdr->frame_control = fc;
+	ether_addr_copy(hdr->addr1, rtwvif->bssid);
+	ether_addr_copy(hdr->addr2, rtwvif->mac_addr);
+	ether_addr_copy(hdr->addr3, rtwvif->bssid);
+
+	skb_put_zero(skb, sec_hdr_len);
 
+	arp_skb = skb_put_zero(skb, sizeof(*arp_skb));
 	memcpy(arp_skb->llc_hdr, rfc1042_header, sizeof(rfc1042_header));
 	arp_skb->llc_type = htons(ETH_P_ARP);
 
diff --git a/drivers/net/wireless/realtek/rtw89/wow.c b/drivers/net/wireless/realtek/rtw89/wow.c
index fa61484c3839..df22a53b8f41 100644
--- a/drivers/net/wireless/realtek/rtw89/wow.c
+++ b/drivers/net/wireless/realtek/rtw89/wow.c
@@ -27,17 +27,23 @@ void rtw89_wow_parse_akm(struct rtw89_dev *rtwdev, struct sk_buff *skb)
 	rtw_wow->akm = rsn_ie->akm_cipher_suite.type;
 }
 
+#define RTW89_CIPHER_INFO_DEF(cipher) \
+	{WLAN_CIPHER_SUITE_ ## cipher, .fw_alg = RTW89_WOW_FW_ALG_ ## cipher, \
+	 .len = WLAN_KEY_LEN_ ## cipher}
+
 static const struct rtw89_cipher_info rtw89_cipher_info_defs[] = {
-	{WLAN_CIPHER_SUITE_WEP40,	.fw_alg = 1,	.len = WLAN_KEY_LEN_WEP40,},
-	{WLAN_CIPHER_SUITE_WEP104,	.fw_alg = 2,	.len = WLAN_KEY_LEN_WEP104,},
-	{WLAN_CIPHER_SUITE_TKIP,	.fw_alg = 3,	.len = WLAN_KEY_LEN_TKIP,},
-	{WLAN_CIPHER_SUITE_CCMP,	.fw_alg = 6,	.len = WLAN_KEY_LEN_CCMP,},
-	{WLAN_CIPHER_SUITE_GCMP,	.fw_alg = 8,	.len = WLAN_KEY_LEN_GCMP,},
-	{WLAN_CIPHER_SUITE_CCMP_256,	.fw_alg = 7,	.len = WLAN_KEY_LEN_CCMP_256,},
-	{WLAN_CIPHER_SUITE_GCMP_256,	.fw_alg = 23,	.len = WLAN_KEY_LEN_GCMP_256,},
-	{WLAN_CIPHER_SUITE_AES_CMAC,	.fw_alg = 32,	.len = WLAN_KEY_LEN_AES_CMAC,},
+	RTW89_CIPHER_INFO_DEF(WEP40),
+	RTW89_CIPHER_INFO_DEF(WEP104),
+	RTW89_CIPHER_INFO_DEF(TKIP),
+	RTW89_CIPHER_INFO_DEF(CCMP),
+	RTW89_CIPHER_INFO_DEF(GCMP),
+	RTW89_CIPHER_INFO_DEF(CCMP_256),
+	RTW89_CIPHER_INFO_DEF(GCMP_256),
+	RTW89_CIPHER_INFO_DEF(AES_CMAC),
 };
 
+#undef RTW89_CIPHER_INFO_DEF
+
 static const
 struct rtw89_cipher_info *rtw89_cipher_alg_recognize(u32 cipher)
 {
diff --git a/drivers/net/wireless/realtek/rtw89/wow.h b/drivers/net/wireless/realtek/rtw89/wow.h
index e595aee0196d..0d90add0e88d 100644
--- a/drivers/net/wireless/realtek/rtw89/wow.h
+++ b/drivers/net/wireless/realtek/rtw89/wow.h
@@ -35,6 +35,17 @@ enum rtw89_wake_reason {
 	RTW89_WOW_RSN_RX_NLO = 0x55,
 };
 
+enum rtw89_fw_alg {
+	RTW89_WOW_FW_ALG_WEP40 = 0x1,
+	RTW89_WOW_FW_ALG_WEP104 = 0x2,
+	RTW89_WOW_FW_ALG_TKIP = 0x3,
+	RTW89_WOW_FW_ALG_CCMP = 0x6,
+	RTW89_WOW_FW_ALG_CCMP_256 = 0x7,
+	RTW89_WOW_FW_ALG_GCMP = 0x8,
+	RTW89_WOW_FW_ALG_GCMP_256 = 0x9,
+	RTW89_WOW_FW_ALG_AES_CMAC = 0xa,
+};
+
 struct rtw89_cipher_suite {
 	u8 oui[3];
 	u8 type;
@@ -64,6 +75,25 @@ struct rtw89_set_key_info_iter_data {
 	bool error;
 };
 
+static inline int rtw89_wow_get_sec_hdr_len(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_wow_param *rtw_wow = &rtwdev->wow;
+
+	if (!(rtwdev->chip->chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev)))
+		return 0;
+
+	switch (rtw_wow->ptk_alg) {
+	case RTW89_WOW_FW_ALG_WEP40:
+		return 4;
+	case RTW89_WOW_FW_ALG_TKIP:
+	case RTW89_WOW_FW_ALG_CCMP:
+	case RTW89_WOW_FW_ALG_GCMP_256:
+		return 8;
+	default:
+		return 0;
+	}
+}
+
 #ifdef CONFIG_PM
 int rtw89_wow_suspend(struct rtw89_dev *rtwdev, struct cfg80211_wowlan *wowlan);
 int rtw89_wow_resume(struct rtw89_dev *rtwdev);
-- 
cgit v1.2.3-58-ga151


From 0e5210217768625b43f099bcaafe627b098655d5 Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Thu, 20 Jun 2024 13:58:20 +0800
Subject: wifi: rtw89: wow: update WoWLAN reason register for different FW

Need to update WoWLAN wakeup reason register after firmware version
0.35.22.0 for RTL8922A, and 0.27.80.0 for RTL8852CE.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240620055825.17592-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.h     | 3 ++-
 drivers/net/wireless/realtek/rtw89/fw.c       | 2 ++
 drivers/net/wireless/realtek/rtw89/fw.h       | 6 ++++++
 drivers/net/wireless/realtek/rtw89/reg.h      | 4 ++++
 drivers/net/wireless/realtek/rtw89/rtw8851b.c | 6 +++++-
 drivers/net/wireless/realtek/rtw89/rtw8852a.c | 6 +++++-
 drivers/net/wireless/realtek/rtw89/rtw8852b.c | 6 +++++-
 drivers/net/wireless/realtek/rtw89/rtw8852c.c | 6 +++++-
 drivers/net/wireless/realtek/rtw89/rtw8922a.c | 6 +++++-
 drivers/net/wireless/realtek/rtw89/wow.c      | 7 ++++++-
 10 files changed, 45 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index e8ec0246bccd..f1de975c5356 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -4236,7 +4236,7 @@ struct rtw89_chip_info {
 	const u32 *c2h_regs;
 	struct rtw89_reg_def c2h_counter_reg;
 	const struct rtw89_page_regs *page_regs;
-	u32 wow_reason_reg;
+	const u32 *wow_reason_reg;
 	bool cfo_src_fd;
 	bool cfo_hw_comp;
 	const struct rtw89_reg_def *dcfo_comp;
@@ -4346,6 +4346,7 @@ enum rtw89_fw_feature {
 	RTW89_FW_FEATURE_NO_LPS_PG,
 	RTW89_FW_FEATURE_BEACON_FILTER,
 	RTW89_FW_FEATURE_MACID_PAUSE_SLEEP,
+	RTW89_FW_FEATURE_WOW_REASON_V1,
 };
 
 struct rtw89_fw_suit {
diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index 3c666fd0f8ed..a907d381ac77 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -675,10 +675,12 @@ static const struct __fw_feat_cfg fw_feat_tbl[] = {
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 36, 0, SCAN_OFFLOAD),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 40, 0, CRASH_TRIGGER),
 	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 56, 10, BEACON_FILTER),
+	__CFG_FW_FEAT(RTL8852C, ge, 0, 27, 80, 0, WOW_REASON_V1),
 	__CFG_FW_FEAT(RTL8922A, ge, 0, 34, 30, 0, CRASH_TRIGGER),
 	__CFG_FW_FEAT(RTL8922A, ge, 0, 34, 11, 0, MACID_PAUSE_SLEEP),
 	__CFG_FW_FEAT(RTL8922A, ge, 0, 34, 35, 0, SCAN_OFFLOAD),
 	__CFG_FW_FEAT(RTL8922A, ge, 0, 35, 12, 0, BEACON_FILTER),
+	__CFG_FW_FEAT(RTL8922A, ge, 0, 35, 22, 0, WOW_REASON_V1),
 };
 
 static void rtw89_fw_iterate_feature_cfg(struct rtw89_fw_info *fw,
diff --git a/drivers/net/wireless/realtek/rtw89/fw.h b/drivers/net/wireless/realtek/rtw89/fw.h
index 01fea0b004d3..c3b4324c621c 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.h
+++ b/drivers/net/wireless/realtek/rtw89/fw.h
@@ -4659,4 +4659,10 @@ const struct rtw89_rfe_parms *
 rtw89_load_rfe_data_from_fw(struct rtw89_dev *rtwdev,
 			    const struct rtw89_rfe_parms *init);
 
+enum rtw89_wow_wakeup_ver {
+	RTW89_WOW_REASON_V0,
+	RTW89_WOW_REASON_V1,
+	RTW89_WOW_REASON_NUM,
+};
+
 #endif
diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index 5a59234301ca..fd81cf863bea 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -311,6 +311,8 @@
 #define B_AX_S1_LDO2PWRCUT_F BIT(23)
 #define B_AX_S0_LDO_VSEL_F_MASK GENMASK(22, 21)
 
+#define R_AX_DBG_WOW 0x0504
+
 #define R_AX_SEC_CTRL 0x0C00
 #define B_AX_SEC_IDMEM_SIZE_CONFIG_MASK GENMASK(17, 16)
 
@@ -4315,6 +4317,8 @@
 
 #define R_BE_WLCPU_PORT_PC 0x03FC
 
+#define R_BE_DBG_WOW 0x0504
+
 #define R_BE_DCPU_PLATFORM_ENABLE 0x0888
 #define B_BE_DCPU_SYM_DPLT_MEM_MUX_EN BIT(10)
 #define B_BE_DCPU_WARM_EN BIT(9)
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8851b.c b/drivers/net/wireless/realtek/rtw89/rtw8851b.c
index 78e276f321c2..40cf84a79c46 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8851b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8851b.c
@@ -105,6 +105,10 @@ static const u32 rtw8851b_c2h_regs[RTW89_C2HREG_MAX] = {
 	R_AX_C2HREG_DATA3
 };
 
+static const u32 rtw8851b_wow_wakeup_regs[RTW89_WOW_REASON_NUM] = {
+	R_AX_C2HREG_DATA3 + 3, R_AX_C2HREG_DATA3 + 3,
+};
+
 static const struct rtw89_page_regs rtw8851b_page_regs = {
 	.hci_fc_ctrl	= R_AX_HCI_FC_CTRL,
 	.ch_page_ctrl	= R_AX_CH_PAGE_CTRL,
@@ -2509,7 +2513,7 @@ const struct rtw89_chip_info rtw8851b_chip_info = {
 	.c2h_counter_reg	= {R_AX_UDM1 + 1, B_AX_UDM1_HALMAC_C2H_ENQ_CNT_MASK >> 8},
 	.c2h_regs		= rtw8851b_c2h_regs,
 	.page_regs		= &rtw8851b_page_regs,
-	.wow_reason_reg		= R_AX_C2HREG_DATA3 + 3,
+	.wow_reason_reg		= rtw8851b_wow_wakeup_regs,
 	.cfo_src_fd		= true,
 	.cfo_hw_comp		= true,
 	.dcfo_comp		= &rtw8851b_dcfo_comp,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852a.c b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
index aebbceea93f8..08e148328c62 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852a.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852a.c
@@ -398,6 +398,10 @@ static const u32 rtw8852a_c2h_regs[RTW89_C2HREG_MAX] = {
 	R_AX_C2HREG_DATA3
 };
 
+static const u32 rtw8852a_wow_wakeup_regs[RTW89_WOW_REASON_NUM] = {
+	R_AX_C2HREG_DATA3 + 3, R_AX_C2HREG_DATA3 + 3,
+};
+
 static const struct rtw89_page_regs rtw8852a_page_regs = {
 	.hci_fc_ctrl	= R_AX_HCI_FC_CTRL,
 	.ch_page_ctrl	= R_AX_CH_PAGE_CTRL,
@@ -2225,7 +2229,7 @@ const struct rtw89_chip_info rtw8852a_chip_info = {
 	.c2h_regs		= rtw8852a_c2h_regs,
 	.c2h_counter_reg	= {R_AX_UDM1 + 1, B_AX_UDM1_HALMAC_C2H_ENQ_CNT_MASK >> 8},
 	.page_regs		= &rtw8852a_page_regs,
-	.wow_reason_reg		= R_AX_C2HREG_DATA3 + 3,
+	.wow_reason_reg		= rtw8852a_wow_wakeup_regs,
 	.cfo_src_fd		= false,
 	.cfo_hw_comp            = false,
 	.dcfo_comp		= &rtw8852a_dcfo_comp,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index 381c881fee69..24f9fd444ca0 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -76,6 +76,10 @@ static const u32 rtw8852b_c2h_regs[RTW89_C2HREG_MAX] = {
 	R_AX_C2HREG_DATA3
 };
 
+static const u32 rtw8852b_wow_wakeup_regs[RTW89_WOW_REASON_NUM] = {
+	R_AX_C2HREG_DATA3 + 3, R_AX_C2HREG_DATA3 + 3,
+};
+
 static const struct rtw89_page_regs rtw8852b_page_regs = {
 	.hci_fc_ctrl	= R_AX_HCI_FC_CTRL,
 	.ch_page_ctrl	= R_AX_CH_PAGE_CTRL,
@@ -1026,7 +1030,7 @@ const struct rtw89_chip_info rtw8852b_chip_info = {
 	.c2h_counter_reg	= {R_AX_UDM1 + 1, B_AX_UDM1_HALMAC_C2H_ENQ_CNT_MASK >> 8},
 	.c2h_regs		= rtw8852b_c2h_regs,
 	.page_regs		= &rtw8852b_page_regs,
-	.wow_reason_reg		= R_AX_C2HREG_DATA3 + 3,
+	.wow_reason_reg		= rtw8852b_wow_wakeup_regs,
 	.cfo_src_fd		= true,
 	.cfo_hw_comp		= true,
 	.dcfo_comp		= &rtw8852b_dcfo_comp,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c.c b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
index 544ea1d3b577..193168dc7b6c 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c.c
@@ -73,6 +73,10 @@ static const u32 rtw8852c_c2h_regs[RTW89_H2CREG_MAX] = {
 	R_AX_C2HREG_DATA3_V1
 };
 
+static const u32 rtw8852c_wow_wakeup_regs[RTW89_WOW_REASON_NUM] = {
+	R_AX_C2HREG_DATA3_V1 + 3, R_AX_DBG_WOW,
+};
+
 static const struct rtw89_page_regs rtw8852c_page_regs = {
 	.hci_fc_ctrl	= R_AX_HCI_FC_CTRL_V1,
 	.ch_page_ctrl	= R_AX_CH_PAGE_CTRL_V1,
@@ -3007,7 +3011,7 @@ const struct rtw89_chip_info rtw8852c_chip_info = {
 	.c2h_counter_reg	= {R_AX_UDM1 + 1, B_AX_UDM1_HALMAC_C2H_ENQ_CNT_MASK >> 8},
 	.c2h_regs		= rtw8852c_c2h_regs,
 	.page_regs		= &rtw8852c_page_regs,
-	.wow_reason_reg		= R_AX_C2HREG_DATA3_V1 + 3,
+	.wow_reason_reg		= rtw8852c_wow_wakeup_regs,
 	.cfo_src_fd		= false,
 	.cfo_hw_comp            = false,
 	.dcfo_comp		= &rtw8852c_dcfo_comp,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8922a.c b/drivers/net/wireless/realtek/rtw89/rtw8922a.c
index cc4f251caadd..2af568a3264d 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8922a.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8922a.c
@@ -85,6 +85,10 @@ static const u32 rtw8922a_c2h_regs[RTW89_H2CREG_MAX] = {
 	R_BE_C2HREG_DATA3
 };
 
+static const u32 rtw8922a_wow_wakeup_regs[RTW89_WOW_REASON_NUM] = {
+	R_AX_C2HREG_DATA3_V1 + 3, R_BE_DBG_WOW,
+};
+
 static const struct rtw89_page_regs rtw8922a_page_regs = {
 	.hci_fc_ctrl	= R_BE_HCI_FC_CTRL,
 	.ch_page_ctrl	= R_BE_CH_PAGE_CTRL,
@@ -2609,7 +2613,7 @@ const struct rtw89_chip_info rtw8922a_chip_info = {
 	.c2h_counter_reg	= {R_BE_UDM1 + 1, B_BE_UDM1_HALMAC_C2H_ENQ_CNT_MASK >> 8},
 	.c2h_regs		= rtw8922a_c2h_regs,
 	.page_regs		= &rtw8922a_page_regs,
-	.wow_reason_reg		= R_AX_C2HREG_DATA3_V1 + 3,
+	.wow_reason_reg		= rtw8922a_wow_wakeup_regs,
 	.cfo_src_fd		= true,
 	.cfo_hw_comp            = true,
 	.dcfo_comp		= NULL,
diff --git a/drivers/net/wireless/realtek/rtw89/wow.c b/drivers/net/wireless/realtek/rtw89/wow.c
index df22a53b8f41..a49265cda98f 100644
--- a/drivers/net/wireless/realtek/rtw89/wow.c
+++ b/drivers/net/wireless/realtek/rtw89/wow.c
@@ -723,13 +723,18 @@ static void rtw89_wow_show_wakeup_reason(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_wow_param *rtw_wow = &rtwdev->wow;
 	struct rtw89_wow_aoac_report *aoac_rpt = &rtw_wow->aoac_rpt;
-	u32 wow_reason_reg = rtwdev->chip->wow_reason_reg;
 	struct cfg80211_wowlan_nd_info nd_info;
 	struct cfg80211_wowlan_wakeup wakeup = {
 		.pattern_idx = -1,
 	};
+	u32 wow_reason_reg;
 	u8 reason;
 
+	if (RTW89_CHK_FW_FEATURE(WOW_REASON_V1, &rtwdev->fw))
+		wow_reason_reg = rtwdev->chip->wow_reason_reg[RTW89_WOW_REASON_V1];
+	else
+		wow_reason_reg = rtwdev->chip->wow_reason_reg[RTW89_WOW_REASON_V0];
+
 	reason = rtw89_read8(rtwdev, wow_reason_reg);
 	switch (reason) {
 	case RTW89_WOW_RSN_RX_DEAUTH:
-- 
cgit v1.2.3-58-ga151


From 6a03a349be1b2b8de06e725329d11448e11766d5 Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Thu, 20 Jun 2024 13:58:21 +0800
Subject: wifi: rtw89: wow: update config mac for 802.11ax chip

The 802.11ax chip also needs rtw89_mac_cpu_io_rx(), which notifies Firmware
to start or stop WoWLAN and waits until Firmware process over. Then,
the driver can continue processing to avoid unexpected behavior.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240620055825.17592-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/mac.c    | 29 +++++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/mac.h    |  1 +
 drivers/net/wireless/realtek/rtw89/mac_be.c | 20 --------------------
 drivers/net/wireless/realtek/rtw89/reg.h    |  1 +
 4 files changed, 31 insertions(+), 20 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index b759fa8fbca9..73462f3343e3 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -6320,9 +6320,30 @@ int rtw89_mac_ptk_drop_by_band_and_wait(struct rtw89_dev *rtwdev,
 	return ret;
 }
 
+int rtw89_mac_cpu_io_rx(struct rtw89_dev *rtwdev, bool wow_enable)
+{
+	struct rtw89_mac_h2c_info h2c_info = {};
+	struct rtw89_mac_c2h_info c2h_info = {};
+	u32 ret;
+
+	h2c_info.id = RTW89_FWCMD_H2CREG_FUNC_WOW_CPUIO_RX_CTRL;
+	h2c_info.content_len = sizeof(h2c_info.u.hdr);
+	h2c_info.u.hdr.w0 = u32_encode_bits(wow_enable, RTW89_H2CREG_WOW_CPUIO_RX_CTRL_EN);
+
+	ret = rtw89_fw_msg_reg(rtwdev, &h2c_info, &c2h_info);
+	if (ret)
+		return ret;
+
+	if (c2h_info.id != RTW89_FWCMD_C2HREG_FUNC_WOW_CPUIO_RX_ACK)
+		ret = -EINVAL;
+
+	return ret;
+}
+
 static int rtw89_wow_config_mac_ax(struct rtw89_dev *rtwdev, bool enable_wow)
 {
 	const struct rtw89_mac_gen_def *mac = rtwdev->chip->mac_def;
+	const struct rtw89_chip_info *chip = rtwdev->chip;
 	int ret;
 
 	if (enable_wow) {
@@ -6333,12 +6354,19 @@ static int rtw89_wow_config_mac_ax(struct rtw89_dev *rtwdev, bool enable_wow)
 		}
 
 		rtw89_write32_set(rtwdev, R_AX_RX_FUNCTION_STOP, B_AX_HDR_RX_STOP);
+		rtw89_mac_cpu_io_rx(rtwdev, enable_wow);
 		rtw89_write32_clr(rtwdev, mac->rx_fltr, B_AX_SNIFFER_MODE);
 		rtw89_mac_cfg_ppdu_status(rtwdev, RTW89_MAC_0, false);
 		rtw89_write32(rtwdev, R_AX_ACTION_FWD0, 0);
 		rtw89_write32(rtwdev, R_AX_ACTION_FWD1, 0);
 		rtw89_write32(rtwdev, R_AX_TF_FWD, 0);
 		rtw89_write32(rtwdev, R_AX_HW_RPT_FWD, 0);
+
+		if (chip->chip_id == RTL8852A || rtw89_is_rtl885xb(rtwdev))
+			rtw89_write8(rtwdev, R_BE_DBG_WOW_READY, WOWLAN_NOT_READY);
+		else
+			rtw89_write32_set(rtwdev, R_AX_DBG_WOW,
+					  B_AX_DBG_WOW_CPU_IO_RX_EN);
 	} else {
 		ret = rtw89_mac_resize_ple_rx_quota(rtwdev, false);
 		if (ret) {
@@ -6346,6 +6374,7 @@ static int rtw89_wow_config_mac_ax(struct rtw89_dev *rtwdev, bool enable_wow)
 			return ret;
 		}
 
+		rtw89_mac_cpu_io_rx(rtwdev, enable_wow);
 		rtw89_write32_clr(rtwdev, R_AX_RX_FUNCTION_STOP, B_AX_HDR_RX_STOP);
 		rtw89_mac_cfg_ppdu_status(rtwdev, RTW89_MAC_0, true);
 		rtw89_write32(rtwdev, R_AX_ACTION_FWD0, TRXCFG_MPDU_PROC_ACT_FRWD);
diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h
index a580cb719233..cc9f914d440c 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.h
+++ b/drivers/net/wireless/realtek/rtw89/mac.h
@@ -1446,5 +1446,6 @@ int rtw89_mac_dle_quota_change(struct rtw89_dev *rtwdev, enum rtw89_qta_mode mod
 int rtw89_mac_get_dle_rsvd_qt_cfg(struct rtw89_dev *rtwdev,
 				  enum rtw89_mac_dle_rsvd_qt_type type,
 				  struct rtw89_mac_dle_rsvd_qt_cfg *cfg);
+int rtw89_mac_cpu_io_rx(struct rtw89_dev *rtwdev, bool wow_enable);
 
 #endif
diff --git a/drivers/net/wireless/realtek/rtw89/mac_be.c b/drivers/net/wireless/realtek/rtw89/mac_be.c
index 934bdf3b398f..f212b67771d5 100644
--- a/drivers/net/wireless/realtek/rtw89/mac_be.c
+++ b/drivers/net/wireless/realtek/rtw89/mac_be.c
@@ -2312,26 +2312,6 @@ static void rtw89_mac_dump_qta_lost_be(struct rtw89_dev *rtwdev)
 	dump_err_status_dispatcher_be(rtwdev);
 }
 
-static int rtw89_mac_cpu_io_rx(struct rtw89_dev *rtwdev, bool wow_enable)
-{
-	struct rtw89_mac_h2c_info h2c_info = {};
-	struct rtw89_mac_c2h_info c2h_info = {};
-	u32 ret;
-
-	h2c_info.id = RTW89_FWCMD_H2CREG_FUNC_WOW_CPUIO_RX_CTRL;
-	h2c_info.content_len = sizeof(h2c_info.u.hdr);
-	h2c_info.u.hdr.w0 = u32_encode_bits(wow_enable, RTW89_H2CREG_WOW_CPUIO_RX_CTRL_EN);
-
-	ret = rtw89_fw_msg_reg(rtwdev, &h2c_info, &c2h_info);
-	if (ret)
-		return ret;
-
-	if (c2h_info.id != RTW89_FWCMD_C2HREG_FUNC_WOW_CPUIO_RX_ACK)
-		ret = -EINVAL;
-
-	return ret;
-}
-
 static int rtw89_wow_config_mac_be(struct rtw89_dev *rtwdev, bool enable_wow)
 {
 	if (enable_wow) {
diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index fd81cf863bea..c9f6bdc84d03 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -312,6 +312,7 @@
 #define B_AX_S0_LDO_VSEL_F_MASK GENMASK(22, 21)
 
 #define R_AX_DBG_WOW 0x0504
+#define B_AX_DBG_WOW_CPU_IO_RX_EN BIT(8)
 
 #define R_AX_SEC_CTRL 0x0C00
 #define B_AX_SEC_IDMEM_SIZE_CONFIG_MASK GENMASK(17, 16)
-- 
cgit v1.2.3-58-ga151


From dda364c345913fe03ddbe4d5ae14a2754c100296 Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Thu, 20 Jun 2024 13:58:22 +0800
Subject: wifi: rtw89: wow: fix GTK offload H2C skbuff issue

We mistakenly put skb too large and that may exceed skb->end.
Therefore, we fix it.

skbuff: skb_over_panic: text:ffffffffc09e9a9d len:416 put:204 head:ffff8fba04eca780 data:ffff8fba04eca7e0 tail:0x200 end:0x140 dev:<NULL>
------------[ cut here ]------------
kernel BUG at net/core/skbuff.c:192!
invalid opcode: 0000 [#1] PREEMPT SMP PTI
CPU: 1 PID: 4747 Comm: kworker/u4:44 Tainted: G           O       6.6.30-02659-gc18865c4dfbd #1 86547039b47e46935493f615ee31d0b2d711d35e
Hardware name: HP Meep/Meep, BIOS Google_Meep.11297.262.0 03/18/2021
Workqueue: events_unbound async_run_entry_fn
RIP: 0010:skb_panic+0x5d/0x60
Code: c6 63 8b 8f bb 4c 0f 45 f6 48 c7 c7 4d 89 8b bb 48 89 ce 44 89 d1 41 56 53 41 53 ff b0 c8 00 00 00 e8 27 5f 23 00 48 83 c4 20 <0f> 0b 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 0f 1f 44
RSP: 0018:ffffaa700144bad0 EFLAGS: 00010282
RAX: 0000000000000089 RBX: 0000000000000140 RCX: 14432c5aad26c900
RDX: 0000000000000000 RSI: 00000000ffffdfff RDI: 0000000000000001
RBP: ffffaa700144bae0 R08: 0000000000000000 R09: ffffaa700144b920
R10: 00000000ffffdfff R11: ffffffffbc28fbc0 R12: ffff8fba4e57a010
R13: 0000000000000000 R14: ffffffffbb8f8b63 R15: 0000000000000000
FS:  0000000000000000(0000) GS:ffff8fba7bd00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007999c4ad1000 CR3: 000000015503a000 CR4: 0000000000350ee0
Call Trace:
 <TASK>
 ? __die_body+0x1f/0x70
 ? die+0x3d/0x60
 ? do_trap+0xa4/0x110
 ? skb_panic+0x5d/0x60
 ? do_error_trap+0x6d/0x90
 ? skb_panic+0x5d/0x60
 ? handle_invalid_op+0x30/0x40
 ? skb_panic+0x5d/0x60
 ? exc_invalid_op+0x3c/0x50
 ? asm_exc_invalid_op+0x16/0x20
 ? skb_panic+0x5d/0x60
 skb_put+0x49/0x50
 rtw89_fw_h2c_wow_gtk_ofld+0xbd/0x220 [rtw89_core 778b32de31cd1f14df2d6721ae99ba8a83636fa5]
 rtw89_wow_resume+0x31f/0x540 [rtw89_core 778b32de31cd1f14df2d6721ae99ba8a83636fa5]
 rtw89_ops_resume+0x2b/0xa0 [rtw89_core 778b32de31cd1f14df2d6721ae99ba8a83636fa5]
 ieee80211_reconfig+0x84/0x13e0 [mac80211 818a894e3b77da6298269c59ed7cdff065a4ed52]
 ? __pfx_wiphy_resume+0x10/0x10 [cfg80211 1a793119e2aeb157c4ca4091ff8e1d9ae233b59d]
 ? dev_printk_emit+0x51/0x70
 ? _dev_info+0x6e/0x90
 ? __pfx_wiphy_resume+0x10/0x10 [cfg80211 1a793119e2aeb157c4ca4091ff8e1d9ae233b59d]
 wiphy_resume+0x89/0x180 [cfg80211 1a793119e2aeb157c4ca4091ff8e1d9ae233b59d]
 ? __pfx_wiphy_resume+0x10/0x10 [cfg80211 1a793119e2aeb157c4ca4091ff8e1d9ae233b59d]
 dpm_run_callback+0x3c/0x140
 device_resume+0x1f9/0x3c0
 ? __pfx_dpm_watchdog_handler+0x10/0x10
 async_resume+0x1d/0x30
 async_run_entry_fn+0x29/0xd0
 process_scheduled_works+0x1d8/0x3d0
 worker_thread+0x1fc/0x2f0
 kthread+0xed/0x110
 ? __pfx_worker_thread+0x10/0x10
 ? __pfx_kthread+0x10/0x10
 ret_from_fork+0x38/0x50
 ? __pfx_kthread+0x10/0x10
 ret_from_fork_asm+0x1b/0x30
 </TASK>
Modules linked in: ccm 8021q r8153_ecm cdc_ether usbnet r8152 mii dm_integrity async_xor xor async_tx lz4 lz4_compress zstd zstd_compress zram zsmalloc uinput rfcomm cmac algif_hash rtw89_8922ae(O) algif_skcipher rtw89_8922a(O) af_alg rtw89_pci(O) rtw89_core(O) btusb(O) snd_soc_sst_bxt_da7219_max98357a btbcm(O) snd_soc_hdac_hdmi btintel(O) snd_soc_intel_hda_dsp_common snd_sof_probes btrtl(O) btmtk(O) snd_hda_codec_hdmi snd_soc_dmic uvcvideo videobuf2_vmalloc uvc videobuf2_memops videobuf2_v4l2 videobuf2_common snd_sof_pci_intel_apl snd_sof_intel_hda_common snd_soc_hdac_hda snd_sof_intel_hda soundwire_intel soundwire_generic_allocation snd_sof_intel_hda_mlink soundwire_cadence snd_sof_pci snd_sof_xtensa_dsp mac80211 snd_soc_acpi_intel_match snd_soc_acpi snd_sof snd_sof_utils soundwire_bus snd_soc_max98357a snd_soc_avs snd_soc_hda_codec snd_hda_ext_core snd_intel_dspcfg snd_intel_sdw_acpi snd_soc_da7219 snd_hda_codec snd_hwdep snd_hda_core veth ip6table_nat xt_MASQUERADE xt_cgroup fuse bluetooth ecdh_generic
 cfg80211 ecc
gsmi: Log Shutdown Reason 0x03
---[ end trace 0000000000000000 ]---

Fixes: ed9a3c0d4dd9 ("wifi: rtw89: wow: construct EAPoL packet for GTK rekey offload")
Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240620055825.17592-5-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index a907d381ac77..e3fc9f168ae5 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -6767,10 +6767,8 @@ int rtw89_fw_h2c_wow_gtk_ofld(struct rtw89_dev *rtwdev,
 	skb_put(skb, len);
 	h2c = (struct rtw89_h2c_wow_gtk_ofld *)skb->data;
 
-	if (!enable) {
-		skb_put_zero(skb, sizeof(*gtk_info));
+	if (!enable)
 		goto hdr;
-	}
 
 	ret = rtw89_fw_h2c_add_general_pkt(rtwdev, rtwvif,
 					   RTW89_PKT_OFLD_TYPE_EAPOL_KEY,
-- 
cgit v1.2.3-58-ga151


From 60757f28408bcc63c4c0676b2a69a38adce30fc7 Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Thu, 20 Jun 2024 13:58:23 +0800
Subject: wifi: rtw89: wow: prevent to send unexpected H2C during download
 Firmware

While downloading Firmware in the resume flow, it is possible to receive
beacon and send H2C to Firmware. However, if Firmware receives unexpected
H2C during the download process, it will fail. Therefore, we prevent to
send unexpected H2C during download Firmware in WoWLAN mode.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240620055825.17592-6-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 87c38cdc691b..4acd4301d3c2 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -1921,7 +1921,8 @@ static void rtw89_vif_rx_stats_iter(void *data, u8 *mac,
 		return;
 
 	if (ieee80211_is_beacon(hdr->frame_control)) {
-		if (vif->type == NL80211_IFTYPE_STATION) {
+		if (vif->type == NL80211_IFTYPE_STATION &&
+		    !test_bit(RTW89_FLAG_WOWLAN, rtwdev->flags)) {
 			rtw89_vif_sync_bcn_tsf(rtwvif, hdr, skb->len);
 			rtw89_fw_h2c_rssi_offload(rtwdev, phy_ppdu);
 		}
-- 
cgit v1.2.3-58-ga151


From 2e2564877f57368c69a1ff09504447db63fb0640 Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Thu, 20 Jun 2024 13:58:24 +0800
Subject: wifi: rtw89: wow: enable beacon filter after swapping firmware

To avoid wake up by AP disconnection, but no beacon filter setting and
driver can't disconnect successfully. We need to enable beacon filter
after swapping firmware in WoWLAN mode.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240620055825.17592-7-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/wow.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/wow.c b/drivers/net/wireless/realtek/rtw89/wow.c
index a49265cda98f..9882064ef68d 100644
--- a/drivers/net/wireless/realtek/rtw89/wow.c
+++ b/drivers/net/wireless/realtek/rtw89/wow.c
@@ -1295,12 +1295,16 @@ static int rtw89_wow_disable_trx_pre(struct rtw89_dev *rtwdev)
 
 static int rtw89_wow_disable_trx_post(struct rtw89_dev *rtwdev)
 {
+	struct rtw89_wow_param *rtw_wow = &rtwdev->wow;
+	struct ieee80211_vif *vif = rtw_wow->wow_vif;
 	int ret;
 
 	ret = rtw89_mac_cfg_ppdu_status(rtwdev, RTW89_MAC_0, true);
 	if (ret)
 		rtw89_err(rtwdev, "cfg ppdu status\n");
 
+	rtw89_fw_h2c_set_bcn_fltr_cfg(rtwdev, vif, true);
+
 	return ret;
 }
 
-- 
cgit v1.2.3-58-ga151


From 792586591f446c688821989a21ea3e65f192e6f1 Mon Sep 17 00:00:00 2001
From: Chih-Kang Chang <gary.chang@realtek.com>
Date: Thu, 20 Jun 2024 13:58:25 +0800
Subject: wifi: rtw89: add polling for LPS H2C to ensure FW received

We add polling after sending LPS H2C to ensure that the Firmware is
received and executes RPWM thereafter. Otherwise, if the Firmware
executes RPWM without receiving LPS H2C, it will cause beacon loss in
WoWLAN mode due to the inability to obtain channel and bandwidth
information from H2C.

Signed-off-by: Chih-Kang Chang <gary.chang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240620055825.17592-8-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/fw.c  | 7 +++++++
 drivers/net/wireless/realtek/rtw89/reg.h | 2 ++
 2 files changed, 9 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/fw.c b/drivers/net/wireless/realtek/rtw89/fw.c
index e3fc9f168ae5..fbe08c162b93 100644
--- a/drivers/net/wireless/realtek/rtw89/fw.c
+++ b/drivers/net/wireless/realtek/rtw89/fw.c
@@ -2496,6 +2496,7 @@ int rtw89_fw_h2c_lps_ch_info(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
 	struct rtw89_h2c_lps_ch_info *h2c;
 	u32 len = sizeof(*h2c);
 	struct sk_buff *skb;
+	u32 done;
 	int ret;
 
 	if (chip->chip_gen != RTW89_CHIP_BE)
@@ -2519,12 +2520,18 @@ int rtw89_fw_h2c_lps_ch_info(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif)
 			      H2C_CAT_OUTSRC, H2C_CL_OUTSRC_DM,
 			      H2C_FUNC_FW_LPS_CH_INFO, 0, 0, len);
 
+	rtw89_phy_write32_mask(rtwdev, R_CHK_LPS_STAT, B_CHK_LPS_STAT, 0);
 	ret = rtw89_h2c_tx(rtwdev, skb, false);
 	if (ret) {
 		rtw89_err(rtwdev, "failed to send h2c\n");
 		goto fail;
 	}
 
+	ret = read_poll_timeout(rtw89_phy_read32_mask, done, done, 50, 5000,
+				true, rtwdev, R_CHK_LPS_STAT, B_CHK_LPS_STAT);
+	if (ret)
+		rtw89_warn(rtwdev, "h2c_lps_ch_info done polling timeout\n");
+
 	return 0;
 fail:
 	dev_kfree_skb_any(skb);
diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index c9f6bdc84d03..897884e9667f 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -7818,6 +7818,8 @@
 #define B_UPD_P0_EN BIT(31)
 #define R_EMLSR 0x0044
 #define B_EMLSR_PARM GENMASK(27, 12)
+#define R_CHK_LPS_STAT 0x0058
+#define B_CHK_LPS_STAT BIT(0)
 #define R_SPOOF_CG 0x00B4
 #define B_SPOOF_CG_EN BIT(17)
 #define R_CHINFO_SEG 0x00B4
-- 
cgit v1.2.3-58-ga151


From 9f8852a92e8270c1684117753fcd114e7563a93d Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 21 Jun 2024 15:34:57 +0800
Subject: wifi: rtw89: 8852bt: rfk: add TSSI

TSSI is transmitter signal strength indication, which is a close-loop
hardware circuit to feedback actual transmitting power as a reference for
next transmission.

It does full calibration when we are going to connect an AP. When switching
bands or channels for scanning or some reasons, reset hardware status to
prevent use wrong power value feedback from previous transmission.

It also loads tables of compensation values reflecting current temperature
into registers according to channel and band group to transmit packets
with expected power.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240621073458.55187-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/reg.h           |    1 +
 drivers/net/wireless/realtek/rtw89/rtw8852bt.h     |   13 +
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c | 1207 ++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h |   15 +
 .../wireless/realtek/rtw89/rtw8852bt_rfk_table.c   |  490 ++++++++
 .../wireless/realtek/rtw89/rtw8852bt_rfk_table.h   |   38 +
 6 files changed, 1764 insertions(+)
 create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8852bt.h
 create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
 create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
 create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk_table.c
 create mode 100644 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk_table.h

diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index 897884e9667f..ed4964312063 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -9017,6 +9017,7 @@
 #define R_IQRSN 0x8220
 #define B_IQRSN_K1 BIT(28)
 #define B_IQRSN_K2 BIT(16)
+#define R_DPD_CH0B 0x82BC
 #define R_RXCFIR_P0C0 0x8D40
 #define R_RXCFIR_P0C1 0x8D84
 #define R_RXCFIR_P0C2 0x8DC8
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt.h b/drivers/net/wireless/realtek/rtw89/rtw8852bt.h
new file mode 100644
index 000000000000..6177f36ad667
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+/* Copyright(c) 2024 Realtek Corporation
+ */
+
+#ifndef __RTW89_8852BT_H__
+#define __RTW89_8852BT_H__
+
+#include "core.h"
+
+#define RF_PATH_NUM_8852BT 2
+#define BB_PATH_NUM_8852BT 2
+
+#endif
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
new file mode 100644
index 000000000000..dcc15797f5db
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
@@ -0,0 +1,1207 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/* Copyright(c) 2024 Realtek Corporation
+ */
+
+#include "coex.h"
+#include "debug.h"
+#include "fw.h"
+#include "mac.h"
+#include "phy.h"
+#include "reg.h"
+#include "rtw8852bt.h"
+#include "rtw8852bt_rfk.h"
+#include "rtw8852bt_rfk_table.h"
+#include "rtw8852b_common.h"
+
+#define RTW8852BT_TSSI_PATH_NR 2
+
+#define _TSSI_DE_MASK GENMASK(21, 12)
+
+static const u32 _tssi_trigger[RTW8852BT_TSSI_PATH_NR] = {0x5820, 0x7820};
+static const u32 _tssi_cw_rpt_addr[RTW8852BT_TSSI_PATH_NR] = {0x1c18, 0x3c18};
+static const u32 _tssi_cw_default_addr[RTW8852BT_TSSI_PATH_NR][4] = {
+	{0x5634, 0x5630, 0x5630, 0x5630},
+	{0x7634, 0x7630, 0x7630, 0x7630} };
+static const u32 _tssi_cw_default_mask[4] = {
+	0x000003ff, 0x3ff00000, 0x000ffc00, 0x000003ff};
+static const u32 _tssi_de_cck_long[RF_PATH_NUM_8852BT] = {0x5858, 0x7858};
+static const u32 _tssi_de_cck_short[RF_PATH_NUM_8852BT] = {0x5860, 0x7860};
+static const u32 _tssi_de_mcs_20m[RF_PATH_NUM_8852BT] = {0x5838, 0x7838};
+static const u32 _tssi_de_mcs_40m[RF_PATH_NUM_8852BT] = {0x5840, 0x7840};
+static const u32 _tssi_de_mcs_80m[RF_PATH_NUM_8852BT] = {0x5848, 0x7848};
+static const u32 _tssi_de_mcs_80m_80m[RF_PATH_NUM_8852BT] = {0x5850, 0x7850};
+static const u32 _tssi_de_mcs_5m[RF_PATH_NUM_8852BT] = {0x5828, 0x7828};
+static const u32 _tssi_de_mcs_10m[RF_PATH_NUM_8852BT] = {0x5830, 0x7830};
+
+static u8 _kpath(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
+{
+	u8 val;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[RFK]dbcc_en: %x,PHY%d\n",
+		    rtwdev->dbcc_en, phy_idx);
+
+	if (!rtwdev->dbcc_en) {
+		val = RF_AB;
+	} else {
+		if (phy_idx == RTW89_PHY_0)
+			val = RF_A;
+		else
+			val = RF_B;
+	}
+	return val;
+}
+
+static void _wait_rx_mode(struct rtw89_dev *rtwdev, u8 kpath)
+{
+	u32 rf_mode;
+	u8 path;
+	int ret;
+
+	for (path = 0; path < RF_PATH_MAX; path++) {
+		if (!(kpath & BIT(path)))
+			continue;
+
+		ret = read_poll_timeout_atomic(rtw89_read_rf, rf_mode,
+					       rf_mode != 2, 2, 5000, false,
+					       rtwdev, path, RR_MOD, RR_MOD_MASK);
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[RFK] Wait S%d to Rx mode!! (ret = %d)\n", path, ret);
+	}
+}
+
+static void _tmac_tx_pause(struct rtw89_dev *rtwdev, enum rtw89_phy_idx band_idx,
+			   bool is_pause)
+{
+	if (!is_pause)
+		return;
+
+	_wait_rx_mode(rtwdev, _kpath(rtwdev, band_idx));
+}
+
+static void _tssi_dpk_off(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
+{
+	rtw89_phy_write32_mask(rtwdev, R_DPD_CH0A, BIT(24), 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_DPD_CH0B, BIT(24), 0x0);
+}
+
+static void _tssi_rf_setting(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			     enum rtw89_rf_path path)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	enum rtw89_band band = chan->band_type;
+
+	if (band == RTW89_BAND_2G)
+		rtw89_write_rf(rtwdev, path, RR_TXPOW, RR_TXPOW_TXG, 0x1);
+	else
+		rtw89_write_rf(rtwdev, path, RR_TXPOW, RR_TXPOW_TXA, 0x1);
+}
+
+static void _tssi_set_sys(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			  enum rtw89_rf_path path)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	enum rtw89_band band = chan->band_type;
+
+	rtw89_rfk_parser(rtwdev, &rtw8852bt_tssi_sys_defs_tbl);
+
+	if (chan->band_width == RTW89_CHANNEL_WIDTH_80)
+		rtw89_phy_write32_mask(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_BW80, 0x1);
+	else
+		rtw89_phy_write32_mask(rtwdev, R_RSTB_ASYNC, B_RSTB_ASYNC_BW80, 0x0);
+
+	if (path == RF_PATH_A)
+		rtw89_rfk_parser_by_cond(rtwdev, band == RTW89_BAND_2G,
+					 &rtw8852bt_tssi_sys_a_defs_2g_tbl,
+					 &rtw8852bt_tssi_sys_a_defs_5g_tbl);
+	else
+		rtw89_rfk_parser_by_cond(rtwdev, band == RTW89_BAND_2G,
+					 &rtw8852bt_tssi_sys_b_defs_2g_tbl,
+					 &rtw8852bt_tssi_sys_b_defs_5g_tbl);
+}
+
+static void _tssi_ini_txpwr_ctrl_bb(struct rtw89_dev *rtwdev,
+				    enum rtw89_phy_idx phy,
+				    enum rtw89_rf_path path)
+{
+	rtw89_rfk_parser_by_cond(rtwdev, path == RF_PATH_A,
+				 &rtw8852bt_tssi_init_txpwr_defs_a_tbl,
+				 &rtw8852bt_tssi_init_txpwr_defs_b_tbl);
+}
+
+static void _tssi_ini_txpwr_ctrl_bb_he_tb(struct rtw89_dev *rtwdev,
+					  enum rtw89_phy_idx phy,
+					  enum rtw89_rf_path path)
+{
+	rtw89_rfk_parser_by_cond(rtwdev, path == RF_PATH_A,
+				 &rtw8852bt_tssi_init_txpwr_he_tb_defs_a_tbl,
+				 &rtw8852bt_tssi_init_txpwr_he_tb_defs_b_tbl);
+}
+
+static void _tssi_set_dck(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			  enum rtw89_rf_path path)
+{
+	rtw89_rfk_parser_by_cond(rtwdev, path == RF_PATH_A,
+				 &rtw8852bt_tssi_dck_defs_a_tbl,
+				 &rtw8852bt_tssi_dck_defs_b_tbl);
+}
+
+static void _tssi_set_tmeter_tbl(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+				 enum rtw89_rf_path path)
+{
+#define RTW8852BT_TSSI_GET_VAL(ptr, idx)			\
+({							\
+	s8 *__ptr = (ptr);				\
+	u8 __idx = (idx), __i, __v;			\
+	u32 __val = 0;					\
+	for (__i = 0; __i < 4; __i++) {			\
+		__v = (__ptr[__idx + __i]);		\
+		__val |= (__v << (8 * __i));		\
+	}						\
+	__val;						\
+})
+	struct rtw89_fw_txpwr_track_cfg *trk = rtwdev->fw.elm_info.txpwr_trk;
+	struct rtw89_tssi_info *tssi_info = &rtwdev->tssi;
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	u8 ch = chan->channel;
+	u8 subband = chan->subband_type;
+	const s8 *thm_up_a = NULL;
+	const s8 *thm_down_a = NULL;
+	const s8 *thm_up_b = NULL;
+	const s8 *thm_down_b = NULL;
+	u8 thermal = 0xff;
+	s8 thm_ofst[64] = {0};
+	u32 tmp = 0;
+	u8 i, j;
+
+	switch (subband) {
+	default:
+	case RTW89_CH_2G:
+		thm_up_a = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_2GA_P][0];
+		thm_down_a = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_2GA_N][0];
+		thm_up_b = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_2GB_P][0];
+		thm_down_b = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_2GB_N][0];
+		break;
+	case RTW89_CH_5G_BAND_1:
+		thm_up_a = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GA_P][0];
+		thm_down_a = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GA_N][0];
+		thm_up_b = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GB_P][0];
+		thm_down_b = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GB_N][0];
+		break;
+	case RTW89_CH_5G_BAND_3:
+		thm_up_a = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GA_P][1];
+		thm_down_a = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GA_N][1];
+		thm_up_b = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GB_P][1];
+		thm_down_b = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GB_N][1];
+		break;
+	case RTW89_CH_5G_BAND_4:
+		thm_up_a = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GA_P][2];
+		thm_down_a = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GA_N][2];
+		thm_up_b = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GB_P][2];
+		thm_down_b = trk->delta[RTW89_FW_TXPWR_TRK_TYPE_5GB_N][2];
+		break;
+	}
+
+	if (path == RF_PATH_A) {
+		thermal = tssi_info->thermal[RF_PATH_A];
+
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+			    "[TSSI] ch=%d thermal_pathA=0x%x\n", ch, thermal);
+
+		rtw89_phy_write32_mask(rtwdev, R_P0_TMETER, B_P0_TMETER_DIS, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_P0_TMETER, B_P0_TMETER_TRK, 0x1);
+
+		if (thermal == 0xff) {
+			rtw89_phy_write32_mask(rtwdev, R_P0_TMETER, B_P0_TMETER, 32);
+			rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_VAL, 32);
+
+			for (i = 0; i < 64; i += 4) {
+				rtw89_phy_write32(rtwdev, R_P0_TSSI_BASE + i, 0x0);
+
+				rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+					    "[TSSI] write 0x%x val=0x%08x\n",
+					    R_P0_TSSI_BASE + i, 0x0);
+			}
+
+		} else {
+			rtw89_phy_write32_mask(rtwdev, R_P0_TMETER, B_P0_TMETER,
+					       thermal);
+			rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_VAL,
+					       thermal);
+
+			i = 0;
+			for (j = 0; j < 32; j++)
+				thm_ofst[j] = i < DELTA_SWINGIDX_SIZE ?
+					      -thm_down_a[i++] :
+					      -thm_down_a[DELTA_SWINGIDX_SIZE - 1];
+
+			i = 1;
+			for (j = 63; j >= 32; j--)
+				thm_ofst[j] = i < DELTA_SWINGIDX_SIZE ?
+					      thm_up_a[i++] :
+					      thm_up_a[DELTA_SWINGIDX_SIZE - 1];
+
+			for (i = 0; i < 64; i += 4) {
+				tmp = RTW8852BT_TSSI_GET_VAL(thm_ofst, i);
+				rtw89_phy_write32(rtwdev, R_P0_TSSI_BASE + i, tmp);
+
+				rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+					    "[TSSI] write 0x%x val=0x%08x\n",
+					    0x5c00 + i, tmp);
+			}
+		}
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, R_P0_RFCTM_RDY, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, R_P0_RFCTM_RDY, 0x0);
+
+	} else {
+		thermal = tssi_info->thermal[RF_PATH_B];
+
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+			    "[TSSI] ch=%d thermal_pathB=0x%x\n", ch, thermal);
+
+		rtw89_phy_write32_mask(rtwdev, R_P1_TMETER, B_P1_TMETER_DIS, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_P1_TMETER, B_P1_TMETER_TRK, 0x1);
+
+		if (thermal == 0xff) {
+			rtw89_phy_write32_mask(rtwdev, R_P1_TMETER, B_P1_TMETER, 32);
+			rtw89_phy_write32_mask(rtwdev, R_P1_RFCTM, B_P1_RFCTM_VAL, 32);
+
+			for (i = 0; i < 64; i += 4) {
+				rtw89_phy_write32(rtwdev, R_TSSI_THOF + i, 0x0);
+
+				rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+					    "[TSSI] write 0x%x val=0x%08x\n",
+					    0x7c00 + i, 0x0);
+			}
+
+		} else {
+			rtw89_phy_write32_mask(rtwdev, R_P1_TMETER, B_P1_TMETER,
+					       thermal);
+			rtw89_phy_write32_mask(rtwdev, R_P1_RFCTM, B_P1_RFCTM_VAL,
+					       thermal);
+
+			i = 0;
+			for (j = 0; j < 32; j++)
+				thm_ofst[j] = i < DELTA_SWINGIDX_SIZE ?
+					      -thm_down_b[i++] :
+					      -thm_down_b[DELTA_SWINGIDX_SIZE - 1];
+
+			i = 1;
+			for (j = 63; j >= 32; j--)
+				thm_ofst[j] = i < DELTA_SWINGIDX_SIZE ?
+					      thm_up_b[i++] :
+					      thm_up_b[DELTA_SWINGIDX_SIZE - 1];
+
+			for (i = 0; i < 64; i += 4) {
+				tmp = RTW8852BT_TSSI_GET_VAL(thm_ofst, i);
+				rtw89_phy_write32(rtwdev, R_TSSI_THOF + i, tmp);
+
+				rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+					    "[TSSI] write 0x%x val=0x%08x\n",
+					    0x7c00 + i, tmp);
+			}
+		}
+		rtw89_phy_write32_mask(rtwdev, R_P1_RFCTM, R_P1_RFCTM_RDY, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_P1_RFCTM, R_P1_RFCTM_RDY, 0x0);
+	}
+#undef RTW8852BT_TSSI_GET_VAL
+}
+
+static void _tssi_set_dac_gain_tbl(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+				   enum rtw89_rf_path path)
+{
+	rtw89_rfk_parser_by_cond(rtwdev, path == RF_PATH_A,
+				 &rtw8852bt_tssi_dac_gain_defs_a_tbl,
+				 &rtw8852bt_tssi_dac_gain_defs_b_tbl);
+}
+
+static void _tssi_slope_cal_org(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+				enum rtw89_rf_path path)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	enum rtw89_band band = chan->band_type;
+
+	if (path == RF_PATH_A)
+		rtw89_rfk_parser_by_cond(rtwdev, band == RTW89_BAND_2G,
+					 &rtw8852bt_tssi_slope_a_defs_2g_tbl,
+					 &rtw8852bt_tssi_slope_a_defs_5g_tbl);
+	else
+		rtw89_rfk_parser_by_cond(rtwdev, band == RTW89_BAND_2G,
+					 &rtw8852bt_tssi_slope_b_defs_2g_tbl,
+					 &rtw8852bt_tssi_slope_b_defs_5g_tbl);
+}
+
+static void _tssi_alignment_default(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+				    enum rtw89_rf_path path, bool all)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	enum rtw89_band band = chan->band_type;
+	const struct rtw89_rfk_tbl *tbl = NULL;
+	u8 ch = chan->channel;
+
+	if (path == RF_PATH_A) {
+		if (band == RTW89_BAND_2G)
+			tbl = &rtw8852bt_tssi_align_a_2g_all_defs_tbl;
+		else if (ch >= 36 && ch <= 64)
+			tbl = &rtw8852bt_tssi_align_a_5g1_all_defs_tbl;
+		else if (ch >= 100 && ch <= 144)
+			tbl = &rtw8852bt_tssi_align_a_5g2_all_defs_tbl;
+		else if (ch >= 149 && ch <= 177)
+			tbl = &rtw8852bt_tssi_align_a_5g3_all_defs_tbl;
+	} else {
+		if (ch >= 1 && ch <= 14)
+			tbl = &rtw8852bt_tssi_align_b_2g_all_defs_tbl;
+		else if (ch >= 36 && ch <= 64)
+			tbl = &rtw8852bt_tssi_align_b_5g1_all_defs_tbl;
+		else if (ch >= 100 && ch <= 144)
+			tbl = &rtw8852bt_tssi_align_b_5g2_all_defs_tbl;
+		else if (ch >= 149 && ch <= 177)
+			tbl = &rtw8852bt_tssi_align_b_5g3_all_defs_tbl;
+	}
+
+	if (tbl)
+		rtw89_rfk_parser(rtwdev, tbl);
+}
+
+static void _tssi_set_tssi_slope(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+				 enum rtw89_rf_path path)
+{
+	rtw89_rfk_parser_by_cond(rtwdev, path == RF_PATH_A,
+				 &rtw8852bt_tssi_slope_defs_a_tbl,
+				 &rtw8852bt_tssi_slope_defs_b_tbl);
+}
+
+static void _tssi_set_tssi_track(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+				 enum rtw89_rf_path path)
+{
+	if (path == RF_PATH_A)
+		rtw89_phy_write32_mask(rtwdev, R_P0_TSSIC, B_P0_TSSIC_BYPASS, 0x0);
+	else
+		rtw89_phy_write32_mask(rtwdev, R_P1_TSSIC, B_P1_TSSIC_BYPASS, 0x0);
+}
+
+static void _tssi_set_txagc_offset_mv_avg(struct rtw89_dev *rtwdev,
+					  enum rtw89_phy_idx phy,
+					  enum rtw89_rf_path path)
+{
+	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "======>%s   path=%d\n", __func__,
+		    path);
+
+	if (path == RF_PATH_A)
+		rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_MV_AVG,
+				       B_P0_TSSI_MV_MIX, 0x010);
+	else
+		rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_MV_AVG,
+				       B_P1_RFCTM_DEL, 0x010);
+}
+
+static void _tssi_enable(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
+{
+	u8 i;
+
+	for (i = 0; i < RF_PATH_NUM_8852BT; i++) {
+		_tssi_set_tssi_track(rtwdev, phy, i);
+		_tssi_set_txagc_offset_mv_avg(rtwdev, phy, i);
+
+		if (i == RF_PATH_A) {
+			rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_MV_AVG,
+					       B_P0_TSSI_MV_CLR, 0x0);
+			rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_AVG,
+					       B_P0_TSSI_EN, 0x0);
+			rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_AVG,
+					       B_P0_TSSI_EN, 0x1);
+			rtw89_write_rf(rtwdev, i, RR_TXGA_V1,
+				       RR_TXGA_V1_TRK_EN, 0x1);
+			rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_TRK,
+					       B_P0_TSSI_RFC, 0x3);
+
+			rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_TRK,
+					       B_P0_TSSI_OFT, 0xc0);
+			rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_TRK,
+					       B_P0_TSSI_OFT_EN, 0x0);
+			rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_TRK,
+					       B_P0_TSSI_OFT_EN, 0x1);
+
+			rtwdev->is_tssi_mode[RF_PATH_A] = true;
+		} else {
+			rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_MV_AVG,
+					       B_P1_TSSI_MV_CLR, 0x0);
+			rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_AVG,
+					       B_P1_TSSI_EN, 0x0);
+			rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_AVG,
+					       B_P1_TSSI_EN, 0x1);
+			rtw89_write_rf(rtwdev, i, RR_TXGA_V1,
+				       RR_TXGA_V1_TRK_EN, 0x1);
+			rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_TRK,
+					       B_P1_TSSI_RFC, 0x3);
+
+			rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_TRK,
+					       B_P1_TSSI_OFT, 0xc0);
+			rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_TRK,
+					       B_P1_TSSI_OFT_EN, 0x0);
+			rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_TRK,
+					       B_P1_TSSI_OFT_EN, 0x1);
+
+			rtwdev->is_tssi_mode[RF_PATH_B] = true;
+		}
+	}
+}
+
+static void _tssi_disable(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
+{
+	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_AVG, B_P0_TSSI_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_TRK, B_P0_TSSI_RFC, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_MV_AVG, B_P0_TSSI_MV_CLR, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_AVG, B_P1_TSSI_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_TRK, B_P1_TSSI_RFC, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_MV_AVG, B_P1_TSSI_MV_CLR, 0x1);
+
+	rtwdev->is_tssi_mode[RF_PATH_A] = false;
+	rtwdev->is_tssi_mode[RF_PATH_B] = false;
+}
+
+static u32 _tssi_get_cck_group(struct rtw89_dev *rtwdev, u8 ch)
+{
+	switch (ch) {
+	case 1 ... 2:
+		return 0;
+	case 3 ... 5:
+		return 1;
+	case 6 ... 8:
+		return 2;
+	case 9 ... 11:
+		return 3;
+	case 12 ... 13:
+		return 4;
+	case 14:
+		return 5;
+	}
+
+	return 0;
+}
+
+#define TSSI_EXTRA_GROUP_BIT (BIT(31))
+#define TSSI_EXTRA_GROUP(idx) (TSSI_EXTRA_GROUP_BIT | (idx))
+#define IS_TSSI_EXTRA_GROUP(group) ((group) & TSSI_EXTRA_GROUP_BIT)
+#define TSSI_EXTRA_GET_GROUP_IDX1(group) ((group) & ~TSSI_EXTRA_GROUP_BIT)
+#define TSSI_EXTRA_GET_GROUP_IDX2(group) (TSSI_EXTRA_GET_GROUP_IDX1(group) + 1)
+
+static u32 _tssi_get_ofdm_group(struct rtw89_dev *rtwdev, u8 ch)
+{
+	switch (ch) {
+	case 1 ... 2:
+		return 0;
+	case 3 ... 5:
+		return 1;
+	case 6 ... 8:
+		return 2;
+	case 9 ... 11:
+		return 3;
+	case 12 ... 14:
+		return 4;
+	case 36 ... 40:
+		return 5;
+	case 41 ... 43:
+		return TSSI_EXTRA_GROUP(5);
+	case 44 ... 48:
+		return 6;
+	case 49 ... 51:
+		return TSSI_EXTRA_GROUP(6);
+	case 52 ... 56:
+		return 7;
+	case 57 ... 59:
+		return TSSI_EXTRA_GROUP(7);
+	case 60 ... 64:
+		return 8;
+	case 100 ... 104:
+		return 9;
+	case 105 ... 107:
+		return TSSI_EXTRA_GROUP(9);
+	case 108 ... 112:
+		return 10;
+	case 113 ... 115:
+		return TSSI_EXTRA_GROUP(10);
+	case 116 ... 120:
+		return 11;
+	case 121 ... 123:
+		return TSSI_EXTRA_GROUP(11);
+	case 124 ... 128:
+		return 12;
+	case 129 ... 131:
+		return TSSI_EXTRA_GROUP(12);
+	case 132 ... 136:
+		return 13;
+	case 137 ... 139:
+		return TSSI_EXTRA_GROUP(13);
+	case 140 ... 144:
+		return 14;
+	case 149 ... 153:
+		return 15;
+	case 154 ... 156:
+		return TSSI_EXTRA_GROUP(15);
+	case 157 ... 161:
+		return 16;
+	case 162 ... 164:
+		return TSSI_EXTRA_GROUP(16);
+	case 165 ... 169:
+		return 17;
+	case 170 ... 172:
+		return TSSI_EXTRA_GROUP(17);
+	case 173 ... 177:
+		return 18;
+	}
+
+	return 0;
+}
+
+static u32 _tssi_get_trim_group(struct rtw89_dev *rtwdev, u8 ch)
+{
+	switch (ch) {
+	case 1 ... 8:
+		return 0;
+	case 9 ... 14:
+		return 1;
+	case 36 ... 48:
+		return 2;
+	case 52 ... 64:
+		return 3;
+	case 100 ... 112:
+		return 4;
+	case 116 ... 128:
+		return 5;
+	case 132 ... 144:
+		return 6;
+	case 149 ... 177:
+		return 7;
+	}
+
+	return 0;
+}
+
+static s8 _tssi_get_ofdm_de(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			    enum rtw89_rf_path path)
+{
+	struct rtw89_tssi_info *tssi_info = &rtwdev->tssi;
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	u8 ch = chan->channel;
+	u32 gidx, gidx_1st, gidx_2nd;
+	s8 de_1st;
+	s8 de_2nd;
+	s8 val;
+
+	gidx = _tssi_get_ofdm_group(rtwdev, ch);
+
+	rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+		    "[TSSI][TRIM]: path=%d mcs group_idx=0x%x\n", path, gidx);
+
+	if (IS_TSSI_EXTRA_GROUP(gidx)) {
+		gidx_1st = TSSI_EXTRA_GET_GROUP_IDX1(gidx);
+		gidx_2nd = TSSI_EXTRA_GET_GROUP_IDX2(gidx);
+		de_1st = tssi_info->tssi_mcs[path][gidx_1st];
+		de_2nd = tssi_info->tssi_mcs[path][gidx_2nd];
+		val = (de_1st + de_2nd) / 2;
+
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+			    "[TSSI][TRIM]: path=%d mcs de=%d 1st=%d 2nd=%d\n",
+			    path, val, de_1st, de_2nd);
+	} else {
+		val = tssi_info->tssi_mcs[path][gidx];
+
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+			    "[TSSI][TRIM]: path=%d mcs de=%d\n", path, val);
+	}
+
+	return val;
+}
+
+static s8 _tssi_get_ofdm_trim_de(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+				 enum rtw89_rf_path path)
+{
+	struct rtw89_tssi_info *tssi_info = &rtwdev->tssi;
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	u8 ch = chan->channel;
+	u32 tgidx, tgidx_1st, tgidx_2nd;
+	s8 tde_1st;
+	s8 tde_2nd;
+	s8 val;
+
+	tgidx = _tssi_get_trim_group(rtwdev, ch);
+
+	rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+		    "[TSSI][TRIM]: path=%d mcs trim_group_idx=0x%x\n",
+		    path, tgidx);
+
+	if (IS_TSSI_EXTRA_GROUP(tgidx)) {
+		tgidx_1st = TSSI_EXTRA_GET_GROUP_IDX1(tgidx);
+		tgidx_2nd = TSSI_EXTRA_GET_GROUP_IDX2(tgidx);
+		tde_1st = tssi_info->tssi_trim[path][tgidx_1st];
+		tde_2nd = tssi_info->tssi_trim[path][tgidx_2nd];
+		val = (tde_1st + tde_2nd) / 2;
+
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+			    "[TSSI][TRIM]: path=%d mcs trim_de=%d 1st=%d 2nd=%d\n",
+			    path, val, tde_1st, tde_2nd);
+	} else {
+		val = tssi_info->tssi_trim[path][tgidx];
+
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+			    "[TSSI][TRIM]: path=%d mcs trim_de=%d\n",
+			    path, val);
+	}
+
+	return val;
+}
+
+static void _tssi_set_efuse_to_de(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
+{
+	struct rtw89_tssi_info *tssi_info = &rtwdev->tssi;
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	u8 ch = chan->channel;
+	u8 gidx;
+	s8 ofdm_de;
+	s8 trim_de;
+	s32 val;
+	u32 i;
+
+	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "[TSSI][TRIM]: phy=%d ch=%d\n",
+		    phy, ch);
+
+	for (i = RF_PATH_A; i < RF_PATH_NUM_8852BT; i++) {
+		gidx = _tssi_get_cck_group(rtwdev, ch);
+		trim_de = _tssi_get_ofdm_trim_de(rtwdev, phy, i);
+		val = tssi_info->tssi_cck[i][gidx] + trim_de;
+
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+			    "[TSSI][TRIM]: path=%d cck[%d]=0x%x trim=0x%x\n",
+			    i, gidx, tssi_info->tssi_cck[i][gidx], trim_de);
+
+		rtw89_phy_write32_mask(rtwdev, _tssi_de_cck_long[i], _TSSI_DE_MASK, val);
+		rtw89_phy_write32_mask(rtwdev, _tssi_de_cck_short[i], _TSSI_DE_MASK, val);
+
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+			    "[TSSI] Set TSSI CCK DE 0x%x[21:12]=0x%x\n",
+			    _tssi_de_cck_long[i],
+			    rtw89_phy_read32_mask(rtwdev, _tssi_de_cck_long[i],
+						  _TSSI_DE_MASK));
+
+		ofdm_de = _tssi_get_ofdm_de(rtwdev, phy, i);
+		trim_de = _tssi_get_ofdm_trim_de(rtwdev, phy, i);
+		val = ofdm_de + trim_de;
+
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+			    "[TSSI][TRIM]: path=%d mcs=0x%x trim=0x%x\n",
+			    i, ofdm_de, trim_de);
+
+		rtw89_phy_write32_mask(rtwdev, _tssi_de_mcs_20m[i], _TSSI_DE_MASK, val);
+		rtw89_phy_write32_mask(rtwdev, _tssi_de_mcs_40m[i], _TSSI_DE_MASK, val);
+		rtw89_phy_write32_mask(rtwdev, _tssi_de_mcs_80m[i], _TSSI_DE_MASK, val);
+		rtw89_phy_write32_mask(rtwdev, _tssi_de_mcs_80m_80m[i],
+				       _TSSI_DE_MASK, val);
+		rtw89_phy_write32_mask(rtwdev, _tssi_de_mcs_5m[i], _TSSI_DE_MASK, val);
+		rtw89_phy_write32_mask(rtwdev, _tssi_de_mcs_10m[i], _TSSI_DE_MASK, val);
+
+		rtw89_debug(rtwdev, RTW89_DBG_TSSI,
+			    "[TSSI] Set TSSI MCS DE 0x%x[21:12]=0x%x\n",
+			    _tssi_de_mcs_20m[i],
+			    rtw89_phy_read32_mask(rtwdev, _tssi_de_mcs_20m[i],
+						  _TSSI_DE_MASK));
+	}
+}
+
+static void _tssi_alimentk_dump_result(struct rtw89_dev *rtwdev, enum rtw89_rf_path path)
+{
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[TSSI PA K]\n0x%x = 0x%08x\n0x%x = 0x%08x\n0x%x = 0x%08x\n0x%x = 0x%08x\n"
+		    "0x%x = 0x%08x\n0x%x = 0x%08x\n0x%x = 0x%08x\n0x%x = 0x%08x\n",
+		    R_TSSI_PA_K1 + (path << 13),
+		    rtw89_phy_read32(rtwdev, R_TSSI_PA_K1 + (path << 13)),
+		    R_TSSI_PA_K2 + (path << 13),
+		    rtw89_phy_read32(rtwdev, R_TSSI_PA_K2 + (path << 13)),
+		    R_P0_TSSI_ALIM1 + (path << 13),
+		    rtw89_phy_read32(rtwdev, R_P0_TSSI_ALIM1 + (path << 13)),
+		    R_P0_TSSI_ALIM3 + (path << 13),
+		    rtw89_phy_read32(rtwdev, R_P0_TSSI_ALIM3 + (path << 13)),
+		    R_TSSI_PA_K5 + (path << 13),
+		    rtw89_phy_read32(rtwdev, R_TSSI_PA_K5 + (path << 13)),
+		    R_P0_TSSI_ALIM2 + (path << 13),
+		    rtw89_phy_read32(rtwdev, R_P0_TSSI_ALIM2 + (path << 13)),
+		    R_P0_TSSI_ALIM4 + (path << 13),
+		    rtw89_phy_read32(rtwdev, R_P0_TSSI_ALIM4 + (path << 13)),
+		    R_TSSI_PA_K8 + (path << 13),
+		    rtw89_phy_read32(rtwdev, R_TSSI_PA_K8 + (path << 13)));
+}
+
+static void _tssi_alimentk_done(struct rtw89_dev *rtwdev,
+				enum rtw89_phy_idx phy, enum rtw89_rf_path path)
+{
+	struct rtw89_tssi_info *tssi_info = &rtwdev->tssi;
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	u8 channel = chan->channel;
+	u8 band;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "======>%s   phy=%d   path=%d\n", __func__, phy, path);
+
+	if (channel >= 1 && channel <= 14)
+		band = TSSI_ALIMK_2G;
+	else if (channel >= 36 && channel <= 64)
+		band = TSSI_ALIMK_5GL;
+	else if (channel >= 100 && channel <= 144)
+		band = TSSI_ALIMK_5GM;
+	else if (channel >= 149 && channel <= 177)
+		band = TSSI_ALIMK_5GH;
+	else
+		band = TSSI_ALIMK_2G;
+
+	if (tssi_info->alignment_done[path][band]) {
+		rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_ALIM1 + (path << 13), MASKDWORD,
+				       tssi_info->alignment_value[path][band][0]);
+		rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_ALIM3 + (path << 13), MASKDWORD,
+				       tssi_info->alignment_value[path][band][1]);
+		rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_ALIM2 + (path << 13), MASKDWORD,
+				       tssi_info->alignment_value[path][band][2]);
+		rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_ALIM4 + (path << 13), MASKDWORD,
+				       tssi_info->alignment_value[path][band][3]);
+	}
+
+	_tssi_alimentk_dump_result(rtwdev, path);
+}
+
+static void _tssi_hw_tx(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			enum rtw89_rf_path path, u16 cnt, u16 period, s16 pwr_dbm,
+			u8 enable)
+{
+	enum rtw89_rf_path_bit rx_path;
+
+	if (path == RF_PATH_A)
+		rx_path = RF_A;
+	else if (path == RF_PATH_B)
+		rx_path = RF_B;
+	else if (path == RF_PATH_AB)
+		rx_path = RF_AB;
+	else
+		rx_path = RF_ABCD; /* don't change path, but still set others */
+
+	if (enable) {
+		rtw8852bx_bb_set_plcp_tx(rtwdev);
+		rtw8852bx_bb_cfg_tx_path(rtwdev, path);
+		rtw8852bx_bb_ctrl_rx_path(rtwdev, rx_path);
+		rtw8852bx_bb_set_power(rtwdev, pwr_dbm, phy);
+	}
+
+	rtw8852bx_bb_set_pmac_pkt_tx(rtwdev, enable, cnt, period, 20, phy);
+}
+
+static void _tssi_backup_bb_registers(struct rtw89_dev *rtwdev,
+				      enum rtw89_phy_idx phy, const u32 reg[],
+				      u32 reg_backup[], u32 reg_num)
+{
+	u32 i;
+
+	for (i = 0; i < reg_num; i++) {
+		reg_backup[i] = rtw89_phy_read32_mask(rtwdev, reg[i], MASKDWORD);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[TSSI] Backup BB 0x%x = 0x%x\n", reg[i],
+			    reg_backup[i]);
+	}
+}
+
+static void _tssi_reload_bb_registers(struct rtw89_dev *rtwdev,
+				      enum rtw89_phy_idx phy, const u32 reg[],
+				      u32 reg_backup[], u32 reg_num)
+
+{
+	u32 i;
+
+	for (i = 0; i < reg_num; i++) {
+		rtw89_phy_write32_mask(rtwdev, reg[i], MASKDWORD, reg_backup[i]);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[TSSI] Reload BB 0x%x = 0x%x\n", reg[i],
+			    reg_backup[i]);
+	}
+}
+
+static u8 _tssi_ch_to_idx(struct rtw89_dev *rtwdev, u8 channel)
+{
+	u8 channel_index;
+
+	if (channel >= 1 && channel <= 14)
+		channel_index = channel - 1;
+	else if (channel >= 36 && channel <= 64)
+		channel_index = (channel - 36) / 2 + 14;
+	else if (channel >= 100 && channel <= 144)
+		channel_index = ((channel - 100) / 2) + 15 + 14;
+	else if (channel >= 149 && channel <= 177)
+		channel_index = ((channel - 149) / 2) + 38 + 14;
+	else
+		channel_index = 0;
+
+	return channel_index;
+}
+
+static bool _tssi_get_cw_report(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+				enum rtw89_rf_path path, const s16 *power,
+				u32 *tssi_cw_rpt)
+{
+	u32 tx_counter, tx_counter_tmp;
+	const int retry = 100;
+	u32 tmp;
+	int j, k;
+
+	for (j = 0; j < RTW8852BT_TSSI_PATH_NR; j++) {
+		rtw89_phy_write32_mask(rtwdev, _tssi_trigger[path], B_P0_TSSI_EN, 0x0);
+		rtw89_phy_write32_mask(rtwdev, _tssi_trigger[path], B_P0_TSSI_EN, 0x1);
+
+		tx_counter = rtw89_phy_read32_mask(rtwdev, R_TX_COUNTER, MASKLWORD);
+
+		tmp = rtw89_phy_read32_mask(rtwdev, _tssi_trigger[path], MASKDWORD);
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[TSSI PA K] 0x%x = 0x%08x   path=%d\n",
+			    _tssi_trigger[path], tmp, path);
+
+		if (j == 0)
+			_tssi_hw_tx(rtwdev, phy, path, 100, 5000, power[j], true);
+		else
+			_tssi_hw_tx(rtwdev, phy, RF_PATH_ABCD, 100, 5000, power[j], true);
+
+		tx_counter_tmp = rtw89_phy_read32_mask(rtwdev, R_TX_COUNTER, MASKLWORD);
+		tx_counter_tmp -= tx_counter;
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[TSSI PA K] First HWTXcounter=%d path=%d\n",
+			    tx_counter_tmp, path);
+
+		for (k = 0; k < retry; k++) {
+			tmp = rtw89_phy_read32_mask(rtwdev, _tssi_cw_rpt_addr[path],
+						    B_TSSI_CWRPT_RDY);
+			if (tmp)
+				break;
+
+			udelay(30);
+
+			tx_counter_tmp =
+				rtw89_phy_read32_mask(rtwdev, R_TX_COUNTER, MASKLWORD);
+			tx_counter_tmp -= tx_counter;
+
+			rtw89_debug(rtwdev, RTW89_DBG_RFK,
+				    "[TSSI PA K] Flow k = %d HWTXcounter=%d path=%d\n",
+				    k, tx_counter_tmp, path);
+		}
+
+		if (k >= retry) {
+			rtw89_debug(rtwdev, RTW89_DBG_RFK,
+				    "[TSSI PA K] TSSI finish bit k > %d mp:100ms normal:30us path=%d\n",
+				    k, path);
+
+			_tssi_hw_tx(rtwdev, phy, path, 100, 5000, power[j], false);
+			return false;
+		}
+
+		tssi_cw_rpt[j] =
+			rtw89_phy_read32_mask(rtwdev, _tssi_cw_rpt_addr[path],
+					      B_TSSI_CWRPT);
+
+		_tssi_hw_tx(rtwdev, phy, path, 100, 5000, power[j], false);
+
+		tx_counter_tmp = rtw89_phy_read32_mask(rtwdev, R_TX_COUNTER, MASKLWORD);
+		tx_counter_tmp -= tx_counter;
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[TSSI PA K] Final HWTXcounter=%d path=%d\n",
+			    tx_counter_tmp, path);
+	}
+
+	return true;
+}
+
+static void _tssi_alimentk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			   enum rtw89_rf_path path)
+{
+	static const u32 bb_reg[8] = {0x5820, 0x7820, 0x4978, 0x58e4,
+				      0x78e4, 0x49c0, 0x0d18, 0x0d80};
+	static const s16 power_2g[4] = {48, 20, 4, -8};
+	static const s16 power_5g[4] = {48, 20, 4, 4};
+	struct rtw89_tssi_info *tssi_info = &rtwdev->tssi;
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	s32 tssi_alim_offset_1, tssi_alim_offset_2, tssi_alim_offset_3;
+	u32 tssi_cw_rpt[RTW8852BT_TSSI_PATH_NR] = {};
+	u8 channel = chan->channel;
+	u8 ch_idx = _tssi_ch_to_idx(rtwdev, channel);
+	struct rtw8852bx_bb_tssi_bak tssi_bak;
+	s32 aliment_diff, tssi_cw_default;
+	u32 start_time, finish_time;
+	u32 bb_reg_backup[8] = {};
+	const s16 *power;
+	u8 band;
+	bool ok;
+	u32 tmp;
+	u8 j;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "======> %s   channel=%d   path=%d\n", __func__, channel,
+		    path);
+
+	start_time = ktime_get_ns();
+
+	if (chan->band_type == RTW89_BAND_2G)
+		power = power_2g;
+	else
+		power = power_5g;
+
+	if (channel >= 1 && channel <= 14)
+		band = TSSI_ALIMK_2G;
+	else if (channel >= 36 && channel <= 64)
+		band = TSSI_ALIMK_5GL;
+	else if (channel >= 100 && channel <= 144)
+		band = TSSI_ALIMK_5GM;
+	else if (channel >= 149 && channel <= 177)
+		band = TSSI_ALIMK_5GH;
+	else
+		band = TSSI_ALIMK_2G;
+
+	rtw8852bx_bb_backup_tssi(rtwdev, phy, &tssi_bak);
+	_tssi_backup_bb_registers(rtwdev, phy, bb_reg, bb_reg_backup,
+				  ARRAY_SIZE(bb_reg_backup));
+
+	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_AVG, B_P0_TSSI_AVG, 0x8);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_AVG, B_P1_TSSI_AVG, 0x8);
+	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_MV_AVG, B_P0_TSSI_MV_AVG, 0x2);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_MV_AVG, B_P1_TSSI_MV_AVG, 0x2);
+
+	ok = _tssi_get_cw_report(rtwdev, phy, path, power, tssi_cw_rpt);
+	if (!ok)
+		goto out;
+
+	for (j = 0; j < RTW8852BT_TSSI_PATH_NR; j++) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[TSSI PA K] power[%d]=%d  tssi_cw_rpt[%d]=%d\n", j,
+			    power[j], j, tssi_cw_rpt[j]);
+	}
+
+	tmp = rtw89_phy_read32_mask(rtwdev, _tssi_cw_default_addr[path][1],
+				    _tssi_cw_default_mask[1]);
+	tssi_cw_default = sign_extend32(tmp, 8);
+	tssi_alim_offset_1 = tssi_cw_rpt[0] - ((power[0] - power[1]) * 2) -
+			     tssi_cw_rpt[1] + tssi_cw_default;
+	aliment_diff = tssi_alim_offset_1 - tssi_cw_default;
+
+	tmp = rtw89_phy_read32_mask(rtwdev, _tssi_cw_default_addr[path][2],
+				    _tssi_cw_default_mask[2]);
+	tssi_cw_default = sign_extend32(tmp, 8);
+	tssi_alim_offset_2 = tssi_cw_default + aliment_diff;
+
+	tmp = rtw89_phy_read32_mask(rtwdev, _tssi_cw_default_addr[path][3],
+				    _tssi_cw_default_mask[3]);
+	tssi_cw_default = sign_extend32(tmp, 8);
+	tssi_alim_offset_3 = tssi_cw_default + aliment_diff;
+
+	if (path == RF_PATH_A) {
+		tmp = FIELD_PREP(B_P1_TSSI_ALIM11, tssi_alim_offset_1) |
+		      FIELD_PREP(B_P1_TSSI_ALIM12, tssi_alim_offset_2) |
+		      FIELD_PREP(B_P1_TSSI_ALIM13, tssi_alim_offset_3);
+
+		rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_ALIM1, B_P0_TSSI_ALIM1, tmp);
+		rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_ALIM2, B_P0_TSSI_ALIM2, tmp);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[TSSI PA K] tssi_alim_offset = 0x%x   0x%x   0x%x   0x%x\n",
+			    rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM3, B_P0_TSSI_ALIM31),
+			    rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM1, B_P0_TSSI_ALIM11),
+			    rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM1, B_P0_TSSI_ALIM12),
+			    rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM1, B_P0_TSSI_ALIM13));
+	} else {
+		tmp = FIELD_PREP(B_P1_TSSI_ALIM11, tssi_alim_offset_1) |
+		      FIELD_PREP(B_P1_TSSI_ALIM12, tssi_alim_offset_2) |
+		      FIELD_PREP(B_P1_TSSI_ALIM13, tssi_alim_offset_3);
+
+		rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_ALIM1, B_P1_TSSI_ALIM1, tmp);
+		rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_ALIM2, B_P1_TSSI_ALIM2, tmp);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[TSSI PA K] tssi_alim_offset = 0x%x   0x%x   0x%x   0x%x\n",
+			    rtw89_phy_read32_mask(rtwdev, R_P1_TSSI_ALIM3, B_P1_TSSI_ALIM31),
+			    rtw89_phy_read32_mask(rtwdev, R_P1_TSSI_ALIM1, B_P1_TSSI_ALIM11),
+			    rtw89_phy_read32_mask(rtwdev, R_P1_TSSI_ALIM1, B_P1_TSSI_ALIM12),
+			    rtw89_phy_read32_mask(rtwdev, R_P1_TSSI_ALIM1, B_P1_TSSI_ALIM13));
+	}
+
+	tssi_info->alignment_done[path][band] = true;
+	tssi_info->alignment_value[path][band][0] =
+		rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM1 + (path << 13), MASKDWORD);
+	tssi_info->alignment_value[path][band][1] =
+		rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM3 + (path << 13), MASKDWORD);
+	tssi_info->alignment_value[path][band][2] =
+		rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM2 + (path << 13), MASKDWORD);
+	tssi_info->alignment_value[path][band][3] =
+		rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM4 + (path << 13), MASKDWORD);
+
+	tssi_info->check_backup_aligmk[path][ch_idx] = true;
+	tssi_info->alignment_backup_by_ch[path][ch_idx][0] =
+		rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM1 + (path << 13), MASKDWORD);
+	tssi_info->alignment_backup_by_ch[path][ch_idx][1] =
+		rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM3 + (path << 13), MASKDWORD);
+	tssi_info->alignment_backup_by_ch[path][ch_idx][2] =
+		rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM2 + (path << 13), MASKDWORD);
+	tssi_info->alignment_backup_by_ch[path][ch_idx][3] =
+		rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_ALIM4 + (path << 13), MASKDWORD);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[TSSI PA K] tssi_info->alignment_value[path=%d][band=%d][0], 0x%x = 0x%08x\n",
+		    path, band, R_P0_TSSI_ALIM1 + (path << 13),
+		    tssi_info->alignment_value[path][band][0]);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[TSSI PA K] tssi_info->alignment_value[path=%d][band=%d][1], 0x%x = 0x%08x\n",
+		    path, band, R_P0_TSSI_ALIM3 + (path << 13),
+		    tssi_info->alignment_value[path][band][1]);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[TSSI PA K] tssi_info->alignment_value[path=%d][band=%d][2], 0x%x = 0x%08x\n",
+		    path, band, R_P0_TSSI_ALIM2 + (path << 13),
+		    tssi_info->alignment_value[path][band][2]);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[TSSI PA K] tssi_info->alignment_value[path=%d][band=%d][3], 0x%x = 0x%08x\n",
+		    path, band, R_P0_TSSI_ALIM4 + (path << 13),
+		    tssi_info->alignment_value[path][band][3]);
+
+out:
+	_tssi_reload_bb_registers(rtwdev, phy, bb_reg, bb_reg_backup,
+				  ARRAY_SIZE(bb_reg_backup));
+	rtw8852bx_bb_restore_tssi(rtwdev, phy, &tssi_bak);
+	rtw8852bx_bb_tx_mode_switch(rtwdev, phy, 0);
+
+	finish_time = ktime_get_ns();
+	tssi_info->tssi_alimk_time += finish_time - start_time;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[TSSI PA K] %s processing time = %d ms\n", __func__,
+		    tssi_info->tssi_alimk_time);
+}
+
+void rtw8852bt_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, bool hwtx_en)
+{
+	static const u32 reg[2] = {R_DPD_CH0A, R_DPD_CH0B};
+	u8 phy_map = rtw89_btc_phymap(rtwdev, phy, RF_AB);
+	u32 reg_backup[2] = {};
+	u32 tx_en;
+	u8 i;
+
+	_tssi_backup_bb_registers(rtwdev, phy, reg, reg_backup, 2);
+	rtw89_debug(rtwdev, RTW89_DBG_TSSI, "[TSSI] %s: phy=%d\n", __func__, phy);
+	rtw89_btc_ntfy_wl_rfk(rtwdev, phy_map, BTC_WRFKT_IQK, BTC_WRFK_ONESHOT_START);
+
+	_tssi_dpk_off(rtwdev, phy);
+	_tssi_disable(rtwdev, phy);
+
+	for (i = RF_PATH_A; i < RF_PATH_NUM_8852BT; i++) {
+		_tssi_rf_setting(rtwdev, phy, i);
+		_tssi_set_sys(rtwdev, phy, i);
+		_tssi_ini_txpwr_ctrl_bb(rtwdev, phy, i);
+		_tssi_ini_txpwr_ctrl_bb_he_tb(rtwdev, phy, i);
+		_tssi_set_dck(rtwdev, phy, i);
+		_tssi_set_tmeter_tbl(rtwdev, phy, i);
+		_tssi_set_dac_gain_tbl(rtwdev, phy, i);
+		_tssi_slope_cal_org(rtwdev, phy, i);
+		_tssi_alignment_default(rtwdev, phy, i, true);
+		_tssi_set_tssi_slope(rtwdev, phy, i);
+
+		rtw89_chip_stop_sch_tx(rtwdev, phy, &tx_en, RTW89_SCH_TX_SEL_ALL);
+		_tmac_tx_pause(rtwdev, phy, true);
+		if (hwtx_en)
+			_tssi_alimentk(rtwdev, phy, i);
+		_tmac_tx_pause(rtwdev, phy, false);
+		rtw89_chip_resume_sch_tx(rtwdev, phy, tx_en);
+	}
+
+	_tssi_enable(rtwdev, phy);
+	_tssi_set_efuse_to_de(rtwdev, phy);
+
+	_tssi_reload_bb_registers(rtwdev, phy, reg, reg_backup, 2);
+
+	rtw89_btc_ntfy_wl_rfk(rtwdev, phy_map, BTC_WRFKT_IQK, BTC_WRFK_ONESHOT_STOP);
+}
+
+void rtw8852bt_tssi_scan(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	struct rtw89_tssi_info *tssi_info = &rtwdev->tssi;
+	u8 channel = chan->channel;
+	u8 band;
+	u32 i;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "======>%s   phy=%d  channel=%d\n", __func__, phy, channel);
+
+	if (channel >= 1 && channel <= 14)
+		band = TSSI_ALIMK_2G;
+	else if (channel >= 36 && channel <= 64)
+		band = TSSI_ALIMK_5GL;
+	else if (channel >= 100 && channel <= 144)
+		band = TSSI_ALIMK_5GM;
+	else if (channel >= 149 && channel <= 177)
+		band = TSSI_ALIMK_5GH;
+	else
+		band = TSSI_ALIMK_2G;
+
+	_tssi_disable(rtwdev, phy);
+
+	for (i = RF_PATH_A; i < RTW8852BT_TSSI_PATH_NR; i++) {
+		_tssi_rf_setting(rtwdev, phy, i);
+		_tssi_set_sys(rtwdev, phy, i);
+		_tssi_set_tmeter_tbl(rtwdev, phy, i);
+
+		if (tssi_info->alignment_done[i][band])
+			_tssi_alimentk_done(rtwdev, phy, i);
+		else
+			_tssi_alignment_default(rtwdev, phy, i, true);
+	}
+
+	_tssi_enable(rtwdev, phy);
+	_tssi_set_efuse_to_de(rtwdev, phy);
+}
+
+static void rtw8852bt_tssi_default_txagc(struct rtw89_dev *rtwdev,
+					 enum rtw89_phy_idx phy, bool enable)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	u8 channel = chan->channel;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "======> %s   ch=%d\n",
+		    __func__, channel);
+
+	if (enable)
+		return;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "======>%s 1 SCAN_END Set 0x5818[7:0]=0x%x 0x7818[7:0]=0x%x\n",
+		    __func__,
+		    rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_TRK, B_P0_TSSI_OFT),
+		    rtw89_phy_read32_mask(rtwdev, R_P1_TSSI_TRK, B_P1_TSSI_OFT));
+
+	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_TRK, B_P0_TSSI_OFT, 0xc0);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_TRK, B_P1_TSSI_OFT,  0xc0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_TRK, B_P0_TSSI_OFT_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_TRK, B_P0_TSSI_OFT_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_TRK, B_P1_TSSI_OFT_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_TRK, B_P1_TSSI_OFT_EN, 0x1);
+
+	_tssi_alimentk_done(rtwdev, phy, RF_PATH_A);
+	_tssi_alimentk_done(rtwdev, phy, RF_PATH_B);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "======>%s 2 SCAN_END Set 0x5818[7:0]=0x%x 0x7818[7:0]=0x%x\n",
+		    __func__,
+		    rtw89_phy_read32_mask(rtwdev, R_P0_TSSI_TRK, B_P0_TSSI_OFT),
+		    rtw89_phy_read32_mask(rtwdev, R_P1_TSSI_TRK, B_P1_TSSI_OFT));
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "======> %s   SCAN_END\n", __func__);
+}
+
+void rtw8852bt_wifi_scan_notify(struct rtw89_dev *rtwdev, bool scan_start,
+				enum rtw89_phy_idx phy_idx)
+{
+	if (scan_start)
+		rtw8852bt_tssi_default_txagc(rtwdev, phy_idx, true);
+	else
+		rtw8852bt_tssi_default_txagc(rtwdev, phy_idx, false);
+}
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
new file mode 100644
index 000000000000..03fad3abe554
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+/* Copyright(c) 2024 Realtek Corporation
+ */
+
+#ifndef __RTW89_8852BT_RFK_H__
+#define __RTW89_8852BT_RFK_H__
+
+#include "core.h"
+
+void rtw8852bt_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, bool hwtx_en);
+void rtw8852bt_tssi_scan(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy);
+void rtw8852bt_wifi_scan_notify(struct rtw89_dev *rtwdev, bool scan_start,
+				enum rtw89_phy_idx phy_idx);
+
+#endif
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk_table.c b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk_table.c
new file mode 100644
index 000000000000..782144bb7f49
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk_table.c
@@ -0,0 +1,490 @@
+// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
+/* Copyright(c) 2024 Realtek Corporation
+ */
+
+#include "rtw8852bt_rfk_table.h"
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_sys_defs[] = {
+	RTW89_DECL_RFK_WM(0x12a8, 0x0000000f, 0x4),
+	RTW89_DECL_RFK_WM(0x32a8, 0x0000000f, 0x4),
+	RTW89_DECL_RFK_WM(0x12bc, 0x000ffff0, 0x5555),
+	RTW89_DECL_RFK_WM(0x32bc, 0x000ffff0, 0x5555),
+	RTW89_DECL_RFK_WM(0x0300, 0xff000000, 0x16),
+	RTW89_DECL_RFK_WM(0x0304, 0x000000ff, 0x19),
+	RTW89_DECL_RFK_WM(0x0314, 0xffff0000, 0x2041),
+	RTW89_DECL_RFK_WM(0x0318, 0xffffffff, 0x2041),
+	RTW89_DECL_RFK_WM(0x0318, 0xffffffff, 0x20012041),
+	RTW89_DECL_RFK_WM(0x0020, 0x00006000, 0x3),
+	RTW89_DECL_RFK_WM(0x0024, 0x00006000, 0x3),
+	RTW89_DECL_RFK_WM(0x0704, 0xffff0000, 0x601e),
+	RTW89_DECL_RFK_WM(0x2704, 0xffff0000, 0x601e),
+	RTW89_DECL_RFK_WM(0x0700, 0xf0000000, 0x4),
+	RTW89_DECL_RFK_WM(0x2700, 0xf0000000, 0x4),
+	RTW89_DECL_RFK_WM(0x0650, 0x3c000000, 0x0),
+	RTW89_DECL_RFK_WM(0x2650, 0x3c000000, 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_sys_defs);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_sys_a_defs_2g[] = {
+	RTW89_DECL_RFK_WM(0x120c, 0x000000ff, 0x33),
+	RTW89_DECL_RFK_WM(0x12c0, 0x0ff00000, 0x33),
+	RTW89_DECL_RFK_WM(0x58f8, 0x40000000, 0x1),
+	RTW89_DECL_RFK_WM(0x0304, 0x0000ff00, 0x1e),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_sys_a_defs_2g);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_sys_a_defs_5g[] = {
+	RTW89_DECL_RFK_WM(0x120c, 0x000000ff, 0x44),
+	RTW89_DECL_RFK_WM(0x12c0, 0x0ff00000, 0x44),
+	RTW89_DECL_RFK_WM(0x58f8, 0x40000000, 0x0),
+	RTW89_DECL_RFK_WM(0x0304, 0x0000ff00, 0x1d),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_sys_a_defs_5g);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_sys_b_defs_2g[] = {
+	RTW89_DECL_RFK_WM(0x32c0, 0x0ff00000, 0x33),
+	RTW89_DECL_RFK_WM(0x320c, 0x000000ff, 0x33),
+	RTW89_DECL_RFK_WM(0x78f8, 0x40000000, 0x1),
+	RTW89_DECL_RFK_WM(0x0304, 0x0000ff00, 0x1e),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_sys_b_defs_2g);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_sys_b_defs_5g[] = {
+	RTW89_DECL_RFK_WM(0x32c0, 0x0ff00000, 0x44),
+	RTW89_DECL_RFK_WM(0x320c, 0x000000ff, 0x44),
+	RTW89_DECL_RFK_WM(0x78f8, 0x40000000, 0x0),
+	RTW89_DECL_RFK_WM(0x0304, 0x0000ff00, 0x1d),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_sys_b_defs_5g);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_init_txpwr_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x566c, 0x00001000, 0x0),
+	RTW89_DECL_RFK_WM(0x5800, 0xffffffff, 0x003f807f),
+	RTW89_DECL_RFK_WM(0x580c, 0x0000007f, 0x40),
+	RTW89_DECL_RFK_WM(0x580c, 0x0fffff00, 0x00040),
+	RTW89_DECL_RFK_WM(0x5810, 0xffffffff, 0x59010000),
+	RTW89_DECL_RFK_WM(0x5814, 0x01ffffff, 0x002d000),
+	RTW89_DECL_RFK_WM(0x5814, 0xf8000000, 0x00),
+	RTW89_DECL_RFK_WM(0x5818, 0xffffffff, 0x002c1800),
+	RTW89_DECL_RFK_WM(0x581c, 0x3fffffff, 0x1dc80280),
+	RTW89_DECL_RFK_WM(0x5820, 0xffffffff, 0x00002080),
+	RTW89_DECL_RFK_WM(0x580c, 0x10000000, 0x1),
+	RTW89_DECL_RFK_WM(0x580c, 0x40000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5834, 0x3fffffff, 0x000115f2),
+	RTW89_DECL_RFK_WM(0x5838, 0x7fffffff, 0x0000121),
+	RTW89_DECL_RFK_WM(0x5854, 0x3fffffff, 0x000115f2),
+	RTW89_DECL_RFK_WM(0x5858, 0x7fffffff, 0x0000121),
+	RTW89_DECL_RFK_WM(0x5860, 0x80000000, 0x0),
+	RTW89_DECL_RFK_WM(0x5864, 0x07ffffff, 0x00801ff),
+	RTW89_DECL_RFK_WM(0x5898, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x589c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x58a4, 0x000000ff, 0x16),
+	RTW89_DECL_RFK_WM(0x58b0, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x58b4, 0x7fffffff, 0x0a002000),
+	RTW89_DECL_RFK_WM(0x58b8, 0x7fffffff, 0x00007628),
+	RTW89_DECL_RFK_WM(0x58bc, 0x07ffffff, 0x7a7807f),
+	RTW89_DECL_RFK_WM(0x58c0, 0xfffe0000, 0x003f),
+	RTW89_DECL_RFK_WM(0x58c4, 0xffffffff, 0x0003ffff),
+	RTW89_DECL_RFK_WM(0x58c8, 0x00ffffff, 0x000000),
+	RTW89_DECL_RFK_WM(0x58c8, 0xf0000000, 0x0),
+	RTW89_DECL_RFK_WM(0x58cc, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x58d0, 0x07ffffff, 0x2008101),
+	RTW89_DECL_RFK_WM(0x58d4, 0x000000ff, 0x00),
+	RTW89_DECL_RFK_WM(0x58d4, 0x0003fe00, 0x0ff),
+	RTW89_DECL_RFK_WM(0x58d4, 0x07fc0000, 0x100),
+	RTW89_DECL_RFK_WM(0x58d8, 0xffffffff, 0x8008016c),
+	RTW89_DECL_RFK_WM(0x58dc, 0x0001ffff, 0x0807f),
+	RTW89_DECL_RFK_WM(0x58dc, 0xfff00000, 0xc00),
+	RTW89_DECL_RFK_WM(0x58f0, 0x0003ffff, 0x001ff),
+	RTW89_DECL_RFK_WM(0x58f4, 0x000fffff, 0x000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_init_txpwr_defs_a);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_init_txpwr_defs_b[] = {
+	RTW89_DECL_RFK_WM(0x566c, 0x00001000, 0x0),
+	RTW89_DECL_RFK_WM(0x7800, 0xffffffff, 0x003f807f),
+	RTW89_DECL_RFK_WM(0x780c, 0x0000007f, 0x40),
+	RTW89_DECL_RFK_WM(0x780c, 0x0fffff00, 0x00040),
+	RTW89_DECL_RFK_WM(0x7810, 0xffffffff, 0x59010000),
+	RTW89_DECL_RFK_WM(0x7814, 0x01ffffff, 0x002d000),
+	RTW89_DECL_RFK_WM(0x7814, 0xf8000000, 0x00),
+	RTW89_DECL_RFK_WM(0x7818, 0xffffffff, 0x002c1800),
+	RTW89_DECL_RFK_WM(0x781c, 0x3fffffff, 0x1dc80280),
+	RTW89_DECL_RFK_WM(0x7820, 0xffffffff, 0x00002080),
+	RTW89_DECL_RFK_WM(0x780c, 0x10000000, 0x1),
+	RTW89_DECL_RFK_WM(0x780c, 0x40000000, 0x1),
+	RTW89_DECL_RFK_WM(0x7834, 0x3fffffff, 0x000115f2),
+	RTW89_DECL_RFK_WM(0x7838, 0x7fffffff, 0x0000121),
+	RTW89_DECL_RFK_WM(0x7854, 0x3fffffff, 0x000115f2),
+	RTW89_DECL_RFK_WM(0x7858, 0x7fffffff, 0x0000121),
+	RTW89_DECL_RFK_WM(0x7860, 0x80000000, 0x0),
+	RTW89_DECL_RFK_WM(0x7864, 0x07ffffff, 0x00801ff),
+	RTW89_DECL_RFK_WM(0x7898, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x789c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x78a4, 0x000000ff, 0x16),
+	RTW89_DECL_RFK_WM(0x78b0, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x78b4, 0x7fffffff, 0x0a002000),
+	RTW89_DECL_RFK_WM(0x78b8, 0x7fffffff, 0x00007628),
+	RTW89_DECL_RFK_WM(0x78bc, 0x07ffffff, 0x7a7807f),
+	RTW89_DECL_RFK_WM(0x78c0, 0xfffe0000, 0x003f),
+	RTW89_DECL_RFK_WM(0x78c4, 0xffffffff, 0x0003ffff),
+	RTW89_DECL_RFK_WM(0x78c8, 0x00ffffff, 0x000000),
+	RTW89_DECL_RFK_WM(0x78c8, 0xf0000000, 0x0),
+	RTW89_DECL_RFK_WM(0x78cc, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x78d0, 0x07ffffff, 0x2008101),
+	RTW89_DECL_RFK_WM(0x78d4, 0x000000ff, 0x00),
+	RTW89_DECL_RFK_WM(0x78d4, 0x0003fe00, 0x0ff),
+	RTW89_DECL_RFK_WM(0x78d4, 0x07fc0000, 0x100),
+	RTW89_DECL_RFK_WM(0x78d8, 0xffffffff, 0x8008016c),
+	RTW89_DECL_RFK_WM(0x78dc, 0x0001ffff, 0x0807f),
+	RTW89_DECL_RFK_WM(0x78dc, 0xfff00000, 0xc00),
+	RTW89_DECL_RFK_WM(0x78f0, 0x0003ffff, 0x001ff),
+	RTW89_DECL_RFK_WM(0x78f4, 0x000fffff, 0x000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_init_txpwr_defs_b);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_init_txpwr_he_tb_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x58a0, 0xffffffff, 0x000000fe),
+	RTW89_DECL_RFK_WM(0x58e4, 0x0000007f, 0x1f),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_init_txpwr_he_tb_defs_a);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_init_txpwr_he_tb_defs_b[] = {
+	RTW89_DECL_RFK_WM(0x78a0, 0xffffffff, 0x000000fe),
+	RTW89_DECL_RFK_WM(0x78e4, 0x0000007f, 0x1f),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_init_txpwr_he_tb_defs_b);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_dck_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x580c, 0x0fff0000, 0x000),
+	RTW89_DECL_RFK_WM(0x5814, 0x003ff000, 0x0ef),
+	RTW89_DECL_RFK_WM(0x5814, 0x18000000, 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_dck_defs_a);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_dck_defs_b[] = {
+	RTW89_DECL_RFK_WM(0x780c, 0x0fff0000, 0x000),
+	RTW89_DECL_RFK_WM(0x7814, 0x003ff000, 0x0ef),
+	RTW89_DECL_RFK_WM(0x7814, 0x18000000, 0x0),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_dck_defs_b);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_dac_gain_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x58b0, 0x00000400, 0x1),
+	RTW89_DECL_RFK_WM(0x58b0, 0x00000fff, 0x000),
+	RTW89_DECL_RFK_WM(0x58b0, 0x00000800, 0x1),
+	RTW89_DECL_RFK_WM(0x5a00, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a04, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a08, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a0c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a10, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a14, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a18, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a1c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a20, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a24, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a28, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a2c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a30, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a34, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a38, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a3c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a40, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a44, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a48, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a4c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a50, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a54, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a58, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a5c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a60, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a64, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a68, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a6c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a70, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a74, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a78, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a7c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a80, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a84, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a88, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a8c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a90, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a94, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a98, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5a9c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5aa0, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5aa4, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5aa8, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5aac, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5ab0, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5ab4, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5ab8, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5abc, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5ac0, 0xffffffff, 0x00000000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_dac_gain_defs_a);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_dac_gain_defs_b[] = {
+	RTW89_DECL_RFK_WM(0x78b0, 0x00000fff, 0x000),
+	RTW89_DECL_RFK_WM(0x78b0, 0x00000800, 0x1),
+	RTW89_DECL_RFK_WM(0x7a00, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a04, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a08, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a0c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a10, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a14, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a18, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a1c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a20, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a24, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a28, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a2c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a30, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a34, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a38, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a3c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a40, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a44, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a48, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a4c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a50, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a54, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a58, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a5c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a60, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a64, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a68, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a6c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a70, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a74, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a78, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a7c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a80, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a84, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a88, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a8c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a90, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a94, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a98, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7a9c, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7aa0, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7aa4, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7aa8, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7aac, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7ab0, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7ab4, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7ab8, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7abc, 0xffffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7ac0, 0xffffffff, 0x00000000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_dac_gain_defs_b);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_slope_a_defs_2g[] = {
+	RTW89_DECL_RFK_WM(0x5608, 0x07ffffff, 0x0801008),
+	RTW89_DECL_RFK_WM(0x560c, 0x07ffffff, 0x0201020),
+	RTW89_DECL_RFK_WM(0x5610, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x5614, 0x07ffffff, 0x0804008),
+	RTW89_DECL_RFK_WM(0x5618, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x561c, 0x000001ff, 0x008),
+	RTW89_DECL_RFK_WM(0x561c, 0xffff0000, 0x0808),
+	RTW89_DECL_RFK_WM(0x5620, 0xffffffff, 0x08081e28),
+	RTW89_DECL_RFK_WM(0x5624, 0xffffffff, 0x08080808),
+	RTW89_DECL_RFK_WM(0x5628, 0xffffffff, 0x08081e28),
+	RTW89_DECL_RFK_WM(0x562c, 0x0000ffff, 0x0808),
+	RTW89_DECL_RFK_WM(0x581c, 0x00100000, 0x1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_slope_a_defs_2g);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_slope_a_defs_5g[] = {
+	RTW89_DECL_RFK_WM(0x5608, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x560c, 0x07ffffff, 0x0201019),
+	RTW89_DECL_RFK_WM(0x5610, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x5614, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x5618, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x561c, 0x000001ff, 0x008),
+	RTW89_DECL_RFK_WM(0x561c, 0xffff0000, 0x0808),
+	RTW89_DECL_RFK_WM(0x5620, 0xffffffff, 0x08081808),
+	RTW89_DECL_RFK_WM(0x5624, 0xffffffff, 0x08080808),
+	RTW89_DECL_RFK_WM(0x5628, 0xffffffff, 0x08080808),
+	RTW89_DECL_RFK_WM(0x562c, 0x0000ffff, 0x0808),
+	RTW89_DECL_RFK_WM(0x581c, 0x00100000, 0x1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_slope_a_defs_5g);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_slope_b_defs_2g[] = {
+	RTW89_DECL_RFK_WM(0x7608, 0x07ffffff, 0x0801008),
+	RTW89_DECL_RFK_WM(0x760c, 0x07ffffff, 0x0201020),
+	RTW89_DECL_RFK_WM(0x7610, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x7614, 0x07ffffff, 0x0804008),
+	RTW89_DECL_RFK_WM(0x7618, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x761c, 0x000001ff, 0x008),
+	RTW89_DECL_RFK_WM(0x761c, 0xffff0000, 0x0808),
+	RTW89_DECL_RFK_WM(0x7620, 0xffffffff, 0x08081e28),
+	RTW89_DECL_RFK_WM(0x7624, 0xffffffff, 0x08080808),
+	RTW89_DECL_RFK_WM(0x7628, 0xffffffff, 0x08081e28),
+	RTW89_DECL_RFK_WM(0x762c, 0x0000ffff, 0x0808),
+	RTW89_DECL_RFK_WM(0x781c, 0x00100000, 0x1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_slope_b_defs_2g);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_slope_b_defs_5g[] = {
+	RTW89_DECL_RFK_WM(0x7608, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x760c, 0x07ffffff, 0x0201019),
+	RTW89_DECL_RFK_WM(0x7610, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x7614, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x7618, 0x07ffffff, 0x0201008),
+	RTW89_DECL_RFK_WM(0x761c, 0x000001ff, 0x008),
+	RTW89_DECL_RFK_WM(0x761c, 0xffff0000, 0x0808),
+	RTW89_DECL_RFK_WM(0x7620, 0xffffffff, 0x08081808),
+	RTW89_DECL_RFK_WM(0x7624, 0xffffffff, 0x08080808),
+	RTW89_DECL_RFK_WM(0x7628, 0xffffffff, 0x08080808),
+	RTW89_DECL_RFK_WM(0x762c, 0x0000ffff, 0x0808),
+	RTW89_DECL_RFK_WM(0x781c, 0x00100000, 0x1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_slope_b_defs_5g);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_align_a_2g_all_defs[] = {
+	RTW89_DECL_RFK_WM(0x5604, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5600, 0x3fffffff, 0x3f2d2721),
+	RTW89_DECL_RFK_WM(0x5604, 0x003fffff, 0x010101),
+	RTW89_DECL_RFK_WM(0x5630, 0x3fffffff, 0x029f57c0),
+	RTW89_DECL_RFK_WM(0x5634, 0x3fffffff, 0x00000077),
+	RTW89_DECL_RFK_WM(0x5638, 0x000fffff, 0x00000),
+	RTW89_DECL_RFK_WM(0x563c, 0x3fffffff, 0x029f5bc0),
+	RTW89_DECL_RFK_WM(0x5640, 0x3fffffff, 0x00000076),
+	RTW89_DECL_RFK_WM(0x5644, 0x000fffff, 0x00000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_align_a_2g_all_defs);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_align_a_5g1_all_defs[] = {
+	RTW89_DECL_RFK_WM(0x5604, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5600, 0x3fffffff, 0x3f2d2721),
+	RTW89_DECL_RFK_WM(0x5604, 0x003fffff, 0x010101),
+	RTW89_DECL_RFK_WM(0x5630, 0x3fffffff, 0x007ff3d7),
+	RTW89_DECL_RFK_WM(0x5634, 0x3fffffff, 0x00000068),
+	RTW89_DECL_RFK_WM(0x5638, 0x000fffff, 0x00000),
+	RTW89_DECL_RFK_WM(0x563c, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5640, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5644, 0x000fffff, 0x00000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_align_a_5g1_all_defs);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_align_a_5g2_all_defs[] = {
+	RTW89_DECL_RFK_WM(0x5604, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5600, 0x3fffffff, 0x3f2d2721),
+	RTW89_DECL_RFK_WM(0x5604, 0x003fffff, 0x010101),
+	RTW89_DECL_RFK_WM(0x5630, 0x3fffffff, 0x00a003db),
+	RTW89_DECL_RFK_WM(0x5634, 0x3fffffff, 0x00000065),
+	RTW89_DECL_RFK_WM(0x5638, 0x000fffff, 0x00000),
+	RTW89_DECL_RFK_WM(0x563c, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5640, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5644, 0x000fffff, 0x00000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_align_a_5g2_all_defs);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_align_a_5g3_all_defs[] = {
+	RTW89_DECL_RFK_WM(0x5604, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5600, 0x3fffffff, 0x3f2d2721),
+	RTW89_DECL_RFK_WM(0x5604, 0x003fffff, 0x010101),
+	RTW89_DECL_RFK_WM(0x5630, 0x3fffffff, 0x01101be2),
+	RTW89_DECL_RFK_WM(0x5634, 0x3fffffff, 0x00000065),
+	RTW89_DECL_RFK_WM(0x5638, 0x000fffff, 0x00000),
+	RTW89_DECL_RFK_WM(0x563c, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5640, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x5644, 0x000fffff, 0x00000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_align_a_5g3_all_defs);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_align_b_2g_all_defs[] = {
+	RTW89_DECL_RFK_WM(0x7604, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x7600, 0x3fffffff, 0x3f2d2721),
+	RTW89_DECL_RFK_WM(0x7604, 0x003fffff, 0x010101),
+	RTW89_DECL_RFK_WM(0x7630, 0x3fffffff, 0x023f3fb9),
+	RTW89_DECL_RFK_WM(0x7634, 0x3fffffff, 0x00000075),
+	RTW89_DECL_RFK_WM(0x7638, 0x000fffff, 0x00000),
+	RTW89_DECL_RFK_WM(0x763c, 0x3fffffff, 0x01df3fb8),
+	RTW89_DECL_RFK_WM(0x7640, 0x3fffffff, 0x00000074),
+	RTW89_DECL_RFK_WM(0x7644, 0x000fffff, 0x00000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_align_b_2g_all_defs);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_align_b_5g1_all_defs[] = {
+	RTW89_DECL_RFK_WM(0x7604, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x7600, 0x3fffffff, 0x3f2d2721),
+	RTW89_DECL_RFK_WM(0x7604, 0x003fffff, 0x010101),
+	RTW89_DECL_RFK_WM(0x7630, 0x3fffffff, 0x010017e0),
+	RTW89_DECL_RFK_WM(0x7634, 0x3fffffff, 0x00000069),
+	RTW89_DECL_RFK_WM(0x7638, 0x000fffff, 0x00000),
+	RTW89_DECL_RFK_WM(0x763c, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7640, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7644, 0x000fffff, 0x00000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_align_b_5g1_all_defs);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_align_b_5g2_all_defs[] = {
+	RTW89_DECL_RFK_WM(0x7604, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x7600, 0x3fffffff, 0x3f2d2721),
+	RTW89_DECL_RFK_WM(0x7604, 0x003fffff, 0x010101),
+	RTW89_DECL_RFK_WM(0x7630, 0x3fffffff, 0x01201fe2),
+	RTW89_DECL_RFK_WM(0x7634, 0x3fffffff, 0x00000066),
+	RTW89_DECL_RFK_WM(0x7638, 0x000fffff, 0x00000),
+	RTW89_DECL_RFK_WM(0x763c, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7640, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7644, 0x000fffff, 0x00000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_align_b_5g2_all_defs);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_align_b_5g3_all_defs[] = {
+	RTW89_DECL_RFK_WM(0x7604, 0x80000000, 0x1),
+	RTW89_DECL_RFK_WM(0x7600, 0x3fffffff, 0x3f2d2721),
+	RTW89_DECL_RFK_WM(0x7604, 0x003fffff, 0x010101),
+	RTW89_DECL_RFK_WM(0x7630, 0x3fffffff, 0x01602fe5),
+	RTW89_DECL_RFK_WM(0x7634, 0x3fffffff, 0x00000068),
+	RTW89_DECL_RFK_WM(0x7638, 0x000fffff, 0x00000),
+	RTW89_DECL_RFK_WM(0x763c, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7640, 0x3fffffff, 0x00000000),
+	RTW89_DECL_RFK_WM(0x7644, 0x000fffff, 0x00000),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_align_b_5g3_all_defs);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_slope_defs_a[] = {
+	RTW89_DECL_RFK_WM(0x5814, 0x00000800, 0x1),
+	RTW89_DECL_RFK_WM(0x581c, 0x20000000, 0x1),
+	RTW89_DECL_RFK_WM(0x5814, 0x20000000, 0x1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_slope_defs_a);
+
+static const struct rtw89_reg5_def rtw8852bt_tssi_slope_defs_b[] = {
+	RTW89_DECL_RFK_WM(0x7814, 0x00000800, 0x1),
+	RTW89_DECL_RFK_WM(0x781c, 0x20000000, 0x1),
+	RTW89_DECL_RFK_WM(0x7814, 0x20000000, 0x1),
+};
+
+RTW89_DECLARE_RFK_TBL(rtw8852bt_tssi_slope_defs_b);
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk_table.h b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk_table.h
new file mode 100644
index 000000000000..beb246237d17
--- /dev/null
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk_table.h
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */
+/* Copyright(c) 2024 Realtek Corporation
+ */
+
+#ifndef __RTW89_8852BT_RFK_TABLE_H__
+#define __RTW89_8852BT_RFK_TABLE_H__
+
+#include "phy.h"
+
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_sys_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_sys_a_defs_2g_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_sys_a_defs_5g_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_sys_b_defs_2g_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_sys_b_defs_5g_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_init_txpwr_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_init_txpwr_defs_b_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_init_txpwr_he_tb_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_init_txpwr_he_tb_defs_b_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_dck_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_dck_defs_b_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_dac_gain_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_dac_gain_defs_b_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_slope_a_defs_2g_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_slope_a_defs_5g_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_slope_b_defs_2g_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_slope_b_defs_5g_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_align_a_2g_all_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_align_a_5g1_all_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_align_a_5g2_all_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_align_a_5g3_all_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_align_b_2g_all_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_align_b_5g1_all_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_align_b_5g2_all_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_align_b_5g3_all_defs_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_slope_defs_a_tbl;
+extern const struct rtw89_rfk_tbl rtw8852bt_tssi_slope_defs_b_tbl;
+
+#endif
-- 
cgit v1.2.3-58-ga151


From 5b8c61b82eec9ed267d02331f7ff6b126d9f1253 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 21 Jun 2024 15:34:58 +0800
Subject: wifi: rtw89: 8852bt: rfk: add DPK

DPK is short for digital pre-distortion calibration. It can adjusts digital
waveform according to PA linear characteristics dynamically to enhance
TX EVM for high power.

Do this calibration when we are going to run on AP channel. To prevent
power offset out of boundary, it monitors thermal and set proper boundary
to register.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240621073458.55187-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.h          |    2 +
 drivers/net/wireless/realtek/rtw89/reg.h           |   24 +
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c | 1333 ++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h |    3 +
 4 files changed, 1362 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index f1de975c5356..e1bcac7e1324 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -4764,6 +4764,8 @@ struct rtw89_dpk_info {
 	u8 cur_idx[RTW89_DPK_RF_PATH];
 	u8 cur_k_set;
 	struct rtw89_dpk_bkup_para bp[RTW89_DPK_RF_PATH][RTW89_DPK_BKUP_NUM];
+	u8 max_dpk_txagc[RTW89_DPK_RF_PATH];
+	u32 dpk_order[RTW89_DPK_RF_PATH];
 };
 
 struct rtw89_fem_info {
diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index ed4964312063..177981a108e0 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -7877,6 +7877,7 @@
 #define R_RXCCA_BE1 0x0520
 #define B_RXCCA_BE1_DIS BIT(0)
 #define R_UPD_CLK_ADC 0x0700
+#define B_UPD_GEN_ON BIT(27)
 #define B_UPD_CLK_ADC_VAL GENMASK(26, 25)
 #define B_UPD_CLK_ADC_ON BIT(24)
 #define B_ENABLE_CCK BIT(5)
@@ -7945,6 +7946,8 @@
 #define B_MEASUREMENT_TRIG_MSK BIT(2)
 #define B_CCX_TRIG_OPT_MSK BIT(1)
 #define B_CCX_EN_MSK BIT(0)
+#define R_FAHM 0x0C1C
+#define B_RXTD_CKEN BIT(2)
 #define R_IFS_COUNTER 0x0C28
 #define B_IFS_CLM_PERIOD_MSK GENMASK(31, 16)
 #define B_IFS_CLM_COUNTER_UNIT_MSK GENMASK(15, 14)
@@ -7978,6 +7981,7 @@
 #define B_IQK_DPK_RST BIT(0)
 #define R_TX_COLLISION_T2R_ST 0x0C70
 #define B_TX_COLLISION_T2R_ST_M GENMASK(25, 20)
+#define B_TXRX_FORCE_VAL GENMASK(9, 0)
 #define R_TXGATING 0x0C74
 #define B_TXGATING_EN BIT(4)
 #define R_TXRFC 0x0C7C
@@ -8038,6 +8042,7 @@
 #define B_P0_RFMODE_FTM_RX GENMASK(11, 0)
 #define R_P0_NRBW 0x12B8
 #define B_P0_NRBW_DBG BIT(30)
+#define B_P0_NRBW_RSTB BIT(28)
 #define R_S0_RXDC 0x12D4
 #define B_S0_RXDC_I GENMASK(25, 16)
 #define B_S0_RXDC_Q GENMASK(31, 26)
@@ -8119,6 +8124,8 @@
 #define R_S0_ADDCK 0x1E00
 #define B_S0_ADDCK_I GENMASK(9, 0)
 #define B_S0_ADDCK_Q GENMASK(19, 10)
+#define R_TXCKEN_FORCE 0x2008
+#define B_TXCKEN_FORCE_ALL GENMASK(24, 0)
 #define R_EDCCA_RPT_SEL 0x20CC
 #define B_EDCCA_RPT_SEL_MSK GENMASK(2, 0)
 #define R_ADC_FIFO 0x20fc
@@ -8274,6 +8281,7 @@
 #define R_DCFO_COMP_S0 0x448C
 #define B_DCFO_COMP_S0_MSK GENMASK(11, 0)
 #define R_DCFO_WEIGHT 0x4490
+#define B_DAC_CLK_IDX BIT(31)
 #define B_DCFO_WEIGHT_MSK GENMASK(27, 24)
 #define R_DCFO_OPT 0x4494
 #define B_DCFO_OPT_EN BIT(29)
@@ -8636,6 +8644,8 @@
 #define B_P0_TMETER GENMASK(15, 10)
 #define B_P0_TMETER_DIS BIT(16)
 #define B_P0_TMETER_TRK BIT(24)
+#define R_P0_ADCFF_EN 0x58C8
+#define B_P0_ADCFF_EN BIT(24)
 #define R_P1_TSSIC 0x7814
 #define B_P1_TSSIC_BYPASS BIT(11)
 #define R_P0_TSSI_TRK 0x5818
@@ -8647,6 +8657,7 @@
 #define B_P0_TSSI_EN BIT(31)
 #define B_P0_TSSI_AVG GENMASK(15, 12)
 #define R_P0_RFCTM 0x5864
+#define B_P0_CLKG_FORCE GENMASK(31, 30)
 #define B_P0_RFCTM_EN BIT(29)
 #define B_P0_RFCTM_VAL GENMASK(25, 20)
 #define R_P0_RFCTM_RDY BIT(26)
@@ -8680,12 +8691,14 @@
 #define B_P0_RFM_BT_EN BIT(5)
 #define B_P0_RFM_OUT GENMASK(4, 0)
 #define R_P0_PATH_RST 0x58AC
+#define B_P0_PATH_RST BIT(27)
 #define R_P0_TXDPD 0x58D4
 #define B_P0_TXDPD GENMASK(31, 28)
 #define R_P0_TXPW_RSTB 0x58DC
 #define B_P0_TXPW_RSTB_MANON BIT(30)
 #define B_P0_TXPW_RSTB_TSSI BIT(31)
 #define R_P0_TSSI_MV_AVG 0x58E4
+#define B_P0_TXPW_RSTB GENMASK(28, 27)
 #define B_P0_TSSI_MV_MIX GENMASK(19, 11)
 #define B_P0_TSSI_MV_AVG GENMASK(13, 11)
 #define B_P0_TSSI_MV_CLR BIT(14)
@@ -8810,6 +8823,10 @@
 #define B_P1_TSSI_ALIM2 GENMASK(29, 0)
 #define R_P1_TSSI_ADC_CLK 0x766c
 #define B_P1_TSSI_ADC_CLK GENMASK(17, 16)
+#define R_P1_TXAGC_TH 0x7800
+#define B_P1_TXAGC_MAXMIN GENMASK(15, 0)
+#define R_P1_TXPW_FORCE 0x780C
+#define B_P1_TXPW_RDY BIT(15)
 #define R_P1_TSSIC 0x7814
 #define B_P1_TSSIC_BYPASS BIT(11)
 #define R_P1_TMETER 0x7810
@@ -8825,14 +8842,20 @@
 #define B_P1_TSSI_EN BIT(31)
 #define B_P1_TSSI_AVG GENMASK(15, 12)
 #define R_P1_RFCTM 0x7864
+#define B_P1_CLKG_FORCE GENMASK(31, 30)
+#define B_P1_GOT_TXRX GENMASK(28, 27)
 #define R_P1_RFCTM_RDY BIT(26)
 #define B_P1_RFCTM_VAL GENMASK(25, 20)
 #define B_P1_RFCTM_DEL GENMASK(19, 11)
 #define R_P1_PATH_RST 0x78AC
+#define B_P1_PATH_RST BIT(27)
+#define R_P1_ADCFF_EN 0x78C8
+#define B_P1_ADCFF_EN BIT(24)
 #define R_P1_TXPW_RSTB 0x78DC
 #define B_P1_TXPW_RSTB_MANON BIT(30)
 #define B_P1_TXPW_RSTB_TSSI BIT(31)
 #define R_P1_TSSI_MV_AVG 0x78E4
+#define B_P1_TXPW_RSTB GENMASK(28, 27)
 #define B_P1_TSSI_MV_MIX GENMASK(19, 11)
 #define B_P1_TSSI_MV_AVG GENMASK(13, 11)
 #define B_P1_TSSI_MV_CLR BIT(14)
@@ -9051,6 +9074,7 @@
 #define B_IQKINF2_FCNT GENMASK(23, 16)
 #define B_IQKINF2_KCNT GENMASK(15, 8)
 #define B_IQKINF2_NCTLV GENMASK(7, 0)
+#define R_RFK_ST 0xBFF8
 #define R_DCOF0 0xC000
 #define B_DCOF0_RST BIT(17)
 #define B_DCOF0_V GENMASK(4, 1)
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
index dcc15797f5db..d0a40eabfe2d 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
@@ -13,9 +13,77 @@
 #include "rtw8852bt_rfk_table.h"
 #include "rtw8852b_common.h"
 
+#define RTW8852BT_SS 2
 #define RTW8852BT_TSSI_PATH_NR 2
+#define RTW8852BT_DPK_VER 0x06
+#define DPK_RF_PATH_MAX_8852BT 2
 
 #define _TSSI_DE_MASK GENMASK(21, 12)
+#define DPK_TXAGC_LOWER 0x2e
+#define DPK_TXAGC_UPPER 0x3f
+#define DPK_TXAGC_INVAL 0xff
+#define RFREG_MASKRXBB 0x003e0
+#define RFREG_MASKMODE 0xf0000
+
+enum rf_mode {
+	RF_SHUT_DOWN = 0x0,
+	RF_STANDBY = 0x1,
+	RF_TX = 0x2,
+	RF_RX = 0x3,
+	RF_TXIQK = 0x4,
+	RF_DPK = 0x5,
+	RF_RXK1 = 0x6,
+	RF_RXK2 = 0x7,
+};
+
+enum rtw8852bt_dpk_id {
+	LBK_RXIQK	= 0x06,
+	SYNC		= 0x10,
+	MDPK_IDL	= 0x11,
+	MDPK_MPA	= 0x12,
+	GAIN_LOSS	= 0x13,
+	GAIN_CAL	= 0x14,
+	DPK_RXAGC	= 0x15,
+	KIP_PRESET	= 0x16,
+	KIP_RESTORE	= 0x17,
+	DPK_TXAGC	= 0x19,
+	D_KIP_PRESET	= 0x28,
+	D_TXAGC		= 0x29,
+	D_RXAGC		= 0x2a,
+	D_SYNC		= 0x2b,
+	D_GAIN_LOSS	= 0x2c,
+	D_MDPK_IDL	= 0x2d,
+	D_GAIN_NORM	= 0x2f,
+	D_KIP_THERMAL	= 0x30,
+	D_KIP_RESTORE	= 0x31
+};
+
+enum dpk_agc_step {
+	DPK_AGC_STEP_SYNC_DGAIN,
+	DPK_AGC_STEP_GAIN_ADJ,
+	DPK_AGC_STEP_GAIN_LOSS_IDX,
+	DPK_AGC_STEP_GL_GT_CRITERION,
+	DPK_AGC_STEP_GL_LT_CRITERION,
+	DPK_AGC_STEP_SET_TX_GAIN,
+};
+
+enum adc_ck {
+	ADC_NA = 0,
+	ADC_480M = 1,
+	ADC_960M = 2,
+	ADC_1920M = 3,
+};
+
+enum dac_ck {
+	DAC_40M = 0,
+	DAC_80M = 1,
+	DAC_120M = 2,
+	DAC_160M = 3,
+	DAC_240M = 4,
+	DAC_320M = 5,
+	DAC_480M = 6,
+	DAC_960M = 7,
+};
 
 static const u32 _tssi_trigger[RTW8852BT_TSSI_PATH_NR] = {0x5820, 0x7820};
 static const u32 _tssi_cw_rpt_addr[RTW8852BT_TSSI_PATH_NR] = {0x1c18, 0x3c18};
@@ -33,6 +101,116 @@ static const u32 _tssi_de_mcs_80m_80m[RF_PATH_NUM_8852BT] = {0x5850, 0x7850};
 static const u32 _tssi_de_mcs_5m[RF_PATH_NUM_8852BT] = {0x5828, 0x7828};
 static const u32 _tssi_de_mcs_10m[RF_PATH_NUM_8852BT] = {0x5830, 0x7830};
 
+static const u32 rtw8852bt_backup_bb_regs[] = {0x2344, 0x5800, 0x7800, 0x0704};
+static const u32 rtw8852bt_backup_rf_regs[] = {
+	0xde, 0xdf, 0x8b, 0x90, 0x97, 0x85, 0x5, 0x10005};
+static const u32 rtw8852bt_backup_kip_regs[] = {
+	0x813c, 0x8124, 0x8120, 0xc0d4, 0xc0d8, 0xc0c4, 0xc0ec,
+	0x823c, 0x8224, 0x8220, 0xc1d4, 0xc1d8, 0xc1c4, 0xc1ec};
+
+#define BACKUP_BB_REGS_NR ARRAY_SIZE(rtw8852bt_backup_bb_regs)
+#define BACKUP_RF_REGS_NR ARRAY_SIZE(rtw8852bt_backup_rf_regs)
+#define BACKUP_KIP_REGS_NR ARRAY_SIZE(rtw8852bt_backup_kip_regs)
+
+static void _rfk_get_thermal(struct rtw89_dev *rtwdev, u8 kidx, enum rtw89_rf_path path)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+
+	rtw89_write_rf(rtwdev, path, RR_TM, RR_TM_TRI, 0x1);
+	rtw89_write_rf(rtwdev, path, RR_TM, RR_TM_TRI, 0x0);
+	rtw89_write_rf(rtwdev, path, RR_TM, RR_TM_TRI, 0x1);
+
+	udelay(200);
+
+	dpk->bp[path][kidx].ther_dpk = rtw89_read_rf(rtwdev, path, RR_TM, RR_TM_VAL);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] thermal@DPK = 0x%x\n",
+		    dpk->bp[path][kidx].ther_dpk);
+}
+
+static void _rfk_backup_bb_reg(struct rtw89_dev *rtwdev, u32 backup_bb_reg_val[])
+{
+	u32 i;
+
+	for (i = 0; i < BACKUP_BB_REGS_NR; i++) {
+		backup_bb_reg_val[i] =
+		rtw89_phy_read32_mask(rtwdev, rtw8852bt_backup_bb_regs[i], MASKDWORD);
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[RFK]backup bb reg : %x, value =%x\n",
+			    rtw8852bt_backup_bb_regs[i], backup_bb_reg_val[i]);
+	}
+}
+
+static void _rfk_backup_kip_reg(struct rtw89_dev *rtwdev, u32 backup_kip_reg_val[])
+{
+	u32 i;
+
+	for (i = 0; i < BACKUP_KIP_REGS_NR; i++) {
+		backup_kip_reg_val[i] =
+			rtw89_phy_read32_mask(rtwdev, rtw8852bt_backup_kip_regs[i],
+					      MASKDWORD);
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] Backup 0x%x = %x\n",
+			    rtw8852bt_backup_kip_regs[i], backup_kip_reg_val[i]);
+	}
+}
+
+static
+void _rfk_backup_rf_reg(struct rtw89_dev *rtwdev, u32 backup_rf_reg_val[], u8 rf_path)
+{
+	u32 i;
+
+	for (i = 0; i < BACKUP_RF_REGS_NR; i++) {
+		backup_rf_reg_val[i] =
+			rtw89_read_rf(rtwdev, rf_path, rtw8852bt_backup_rf_regs[i],
+				      RFREG_MASK);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] Backup RF S%d 0x%x = %x\n",
+			    rf_path, rtw8852bt_backup_rf_regs[i], backup_rf_reg_val[i]);
+	}
+}
+
+static void _rfk_reload_bb_reg(struct rtw89_dev *rtwdev, const u32 backup_bb_reg_val[])
+{
+	u32 i;
+
+	for (i = 0; i < BACKUP_BB_REGS_NR; i++) {
+		rtw89_phy_write32_mask(rtwdev, rtw8852bt_backup_bb_regs[i],
+				       MASKDWORD, backup_bb_reg_val[i]);
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[RFK]restore bb reg : %x, value =%x\n",
+			    rtw8852bt_backup_bb_regs[i], backup_bb_reg_val[i]);
+	}
+}
+
+static void _rfk_reload_kip_reg(struct rtw89_dev *rtwdev, u32 backup_kip_reg_val[])
+{
+	u32 i;
+
+	for (i = 0; i < BACKUP_KIP_REGS_NR; i++) {
+		rtw89_phy_write32_mask(rtwdev, rtw8852bt_backup_kip_regs[i],
+				       MASKDWORD, backup_kip_reg_val[i]);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[RFK]restore kip reg : %x, value =%x\n",
+			    rtw8852bt_backup_kip_regs[i], backup_kip_reg_val[i]);
+	}
+}
+
+static void _rfk_reload_rf_reg(struct rtw89_dev *rtwdev,
+			       const u32 backup_rf_reg_val[], u8 rf_path)
+{
+	u32 i;
+
+	for (i = 0; i < BACKUP_RF_REGS_NR; i++) {
+		rtw89_write_rf(rtwdev, rf_path, rtw8852bt_backup_rf_regs[i],
+			       RFREG_MASK, backup_rf_reg_val[i]);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[RFK]restore rf S%d reg: %x, value =%x\n", rf_path,
+			    rtw8852bt_backup_rf_regs[i], backup_rf_reg_val[i]);
+	}
+}
+
 static u8 _kpath(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
 {
 	u8 val;
@@ -51,6 +229,124 @@ static u8 _kpath(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
 	return val;
 }
 
+static
+void _txck_force(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, bool force,
+		 enum dac_ck ck)
+{
+	rtw89_phy_write32_mask(rtwdev, R_P0_RXCK | (path << 13), B_P0_TXCK_ON, 0x0);
+
+	if (!force)
+		return;
+
+	rtw89_phy_write32_mask(rtwdev, R_P0_RXCK | (path << 13), B_P0_TXCK_VAL, ck);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RXCK | (path << 13), B_P0_TXCK_ON, 0x1);
+}
+
+static
+void _rxck_force(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, bool force,
+		 enum adc_ck ck)
+{
+	u32 bw = 0;
+
+	rtw89_phy_write32_mask(rtwdev, R_P0_RXCK | (path << 13), B_P0_RXCK_ON, 0x0);
+
+	if (!force)
+		return;
+
+	rtw89_phy_write32_mask(rtwdev, R_P0_RXCK | (path << 13), B_P0_RXCK_VAL, ck);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RXCK | (path << 13), B_P0_RXCK_ON, 0x1);
+
+	switch (ck) {
+	case ADC_480M:
+		bw = RTW89_CHANNEL_WIDTH_40;
+		break;
+	case ADC_960M:
+		bw = RTW89_CHANNEL_WIDTH_80;
+		break;
+	case ADC_1920M:
+		bw = RTW89_CHANNEL_WIDTH_160;
+		break;
+	default:
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "%s==>Invalid ck", __func__);
+		break;
+	}
+
+	rtw8852bx_adc_cfg(rtwdev, bw, path);
+}
+
+static void _rfk_bb_afe_setting(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+				enum rtw89_rf_path path, u8 kpath)
+{
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, MASKHWORD, 0x0303);
+	rtw89_phy_write32_mask(rtwdev, R_P0_ADCFF_EN, B_P0_ADCFF_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_ADCFF_EN, B_P1_ADCFF_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_CLKG_FORCE, 0x3);
+	rtw89_phy_write32_mask(rtwdev, R_P1_RFCTM, B_P1_CLKG_FORCE, 0x3);
+	rtw89_phy_write32_mask(rtwdev, R_TXCKEN_FORCE, B_TXCKEN_FORCE_ALL, 0x1ffffff);
+	rtw89_phy_write32_mask(rtwdev, R_FAHM, B_RXTD_CKEN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC, B_UPD_GEN_ON, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_TX_COLLISION_T2R_ST, B_TXRX_FORCE_VAL, 0x3ff);
+	rtw89_phy_write32_mask(rtwdev, R_IOQ_IQK_DPK, B_IOQ_IQK_DPK_CLKEN, 0x3);
+	rtw89_phy_write32_mask(rtwdev, R_IQK_DPK_RST, B_IQK_DPK_RST, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P0_PATH_RST, B_P0_PATH_RST, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_PATH_RST, B_P1_PATH_RST, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_PD_CTRL, B_PD_HIT_DIS, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_RXCCA, B_RXCCA_DIS, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_DCFO_WEIGHT, B_DAC_CLK_IDX, 0x1);
+
+	_txck_force(rtwdev, RF_PATH_A, true, DAC_960M);
+	_txck_force(rtwdev, RF_PATH_B, true, DAC_960M);
+	_rxck_force(rtwdev, RF_PATH_A, true, ADC_1920M);
+	_rxck_force(rtwdev, RF_PATH_B, true, ADC_1920M);
+
+	rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC,
+			       B_UPD_CLK_ADC_VAL | B_UPD_CLK_ADC_ON, 0x5);
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_DBG, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_DBGMOD, B_P1_DBGMOD_ON, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, MASKBYTE3, 0x1f);
+	udelay(1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, MASKBYTE3, 0x13);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, MASKHWORD, 0x0001);
+	udelay(1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, MASKHWORD, 0x0041);
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_RSTB, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, MASKHWORD, 0x3333);
+
+	rtw89_phy_write32_mask(rtwdev, R_TXPWRB_H, B_TXPWRB_RDY, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_DPD_OFT_EN, MASKLWORD, 0x0000);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TXPW_FORCE, B_P1_TXPW_RDY, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TXAGC_TH, MASKLWORD, 0x0000);
+}
+
+static void _rfk_bb_afe_restore(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+				enum rtw89_rf_path path, u8 kpath)
+{
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, MASKHWORD, 0x0303);
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_DBG, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P1_DBGMOD, B_P1_DBGMOD_ON, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_CLKG_FORCE, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P1_RFCTM, B_P1_CLKG_FORCE, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_TXCKEN_FORCE, B_TXCKEN_FORCE_ALL, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_FAHM, B_RXTD_CKEN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC, B_UPD_GEN_ON, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_TX_COLLISION_T2R_ST, B_TXRX_FORCE_VAL, 0x63);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RXCK, B_P0_TXCK_ALL, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_P1_RXCK, B_P1_TXCK_ALL, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC,
+			       B_UPD_CLK_ADC_VAL | B_UPD_CLK_ADC_ON, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, MASKHWORD, 0x0000);
+	rtw89_phy_write32_mask(rtwdev, R_P0_ADCFF_EN, B_P0_ADCFF_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P1_ADCFF_EN, B_P1_ADCFF_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_PD_CTRL, B_PD_HIT_DIS, 0x0);
+
+	rtw89_phy_write32_mask(rtwdev, R_TXPWRB_H, B_TXPWRB_RDY, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_MV_AVG, B_P0_TXPW_RSTB, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_MV_AVG, B_P0_TXPW_RSTB, 0x2);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TXPW_FORCE, B_P1_TXPW_RDY, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_MV_AVG, B_P1_TXPW_RSTB, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_MV_AVG, B_P1_TXPW_RSTB, 0x2);
+}
+
 static void _wait_rx_mode(struct rtw89_dev *rtwdev, u8 kpath)
 {
 	u32 rf_mode;
@@ -78,6 +374,1011 @@ static void _tmac_tx_pause(struct rtw89_dev *rtwdev, enum rtw89_phy_idx band_idx
 	_wait_rx_mode(rtwdev, _kpath(rtwdev, band_idx));
 }
 
+static void _dpk_onoff(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, bool off)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+	u8 val, kidx = dpk->cur_idx[path];
+	bool off_reverse;
+
+	val = dpk->is_dpk_enable && !off && dpk->bp[path][kidx].path_ok;
+
+	if (off)
+		off_reverse = false;
+	else
+		off_reverse = true;
+
+	val = dpk->is_dpk_enable & off_reverse & dpk->bp[path][kidx].path_ok;
+
+	rtw89_phy_write32_mask(rtwdev, R_DPD_CH0A + (path << 8) + (kidx << 2),
+			       BIT(24), val);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d[%d] DPK %s !!!\n", path,
+		    kidx, dpk->is_dpk_enable & off_reverse ? "enable" : "disable");
+}
+
+static void _dpk_one_shot(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			  enum rtw89_rf_path path, enum rtw8852bt_dpk_id id)
+{
+	u16 dpk_cmd;
+	u32 val;
+	int ret;
+
+	dpk_cmd = (id << 8) | (0x19 + (path << 4));
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_CFG, MASKDWORD, dpk_cmd);
+
+	ret = read_poll_timeout_atomic(rtw89_phy_read32_mask, val, val == 0x55,
+				       1, 30000, false,
+				       rtwdev, R_RFK_ST, MASKBYTE0);
+	if (ret)
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] one-shot 1 over 30ms!!!!\n");
+
+	udelay(1);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_RPT1, MASKDWORD, 0x00030000);
+
+	ret = read_poll_timeout_atomic(rtw89_phy_read32_mask, val, val == 0x8000,
+				       1, 2000, false,
+				       rtwdev, R_RPT_COM, MASKLWORD);
+	if (ret)
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] one-shot 2 over 2ms!!!!\n");
+
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, MASKBYTE0, 0x0);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DPK] one-shot for %s = 0x%04x\n",
+		    id == 0x06 ? "LBK_RXIQK" :
+		    id == 0x10 ? "SYNC" :
+		    id == 0x11 ? "MDPK_IDL" :
+		    id == 0x12 ? "MDPK_MPA" :
+		    id == 0x13 ? "GAIN_LOSS" :
+		    id == 0x14 ? "PWR_CAL" :
+		    id == 0x15 ? "DPK_RXAGC" :
+		    id == 0x16 ? "KIP_PRESET" :
+		    id == 0x17 ? "KIP_RESOTRE" :
+		    "DPK_TXAGC", dpk_cmd);
+}
+
+static void _dpk_rx_dck(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			enum rtw89_rf_path path)
+{
+	rtw89_write_rf(rtwdev, path, RR_DCK, RR_DCK_LV, 0x0);
+	rtw89_write_rf(rtwdev, path, RR_DCK, RR_DCK_LV, 0x1);
+
+	udelay(600);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d RXDCK\n", path);
+}
+
+static void _dpk_information(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			     enum rtw89_rf_path path)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+
+	u8 kidx = dpk->cur_idx[path];
+
+	dpk->bp[path][kidx].band = chan->band_type;
+	dpk->bp[path][kidx].ch = chan->channel;
+	dpk->bp[path][kidx].bw = chan->band_width;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DPK] S%d[%d] (PHY%d): TSSI %s/ DBCC %s/ %s/ CH%d/ %s\n",
+		    path, dpk->cur_idx[path], phy,
+		    rtwdev->is_tssi_mode[path] ? "on" : "off",
+		    rtwdev->dbcc_en ? "on" : "off",
+		    dpk->bp[path][kidx].band == 0 ? "2G" :
+		    dpk->bp[path][kidx].band == 1 ? "5G" : "6G",
+		    dpk->bp[path][kidx].ch,
+		    dpk->bp[path][kidx].bw == 0 ? "20M" :
+		    dpk->bp[path][kidx].bw == 1 ? "40M" : "80M");
+}
+
+static void _dpk_tssi_pause(struct rtw89_dev *rtwdev,
+			    enum rtw89_rf_path path, bool is_pause)
+{
+	rtw89_phy_write32_mask(rtwdev, R_P0_TSSI_TRK + (path << 13),
+			       B_P0_TSSI_TRK_EN, is_pause);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d TSSI %s\n", path,
+		    is_pause ? "pause" : "resume");
+}
+
+static void _dpk_kip_restore(struct rtw89_dev *rtwdev,
+			     enum rtw89_rf_path path)
+{
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_RPT, MASKDWORD, 0x00000000);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_SYSCFG, MASKDWORD, 0x80000000);
+
+	if (rtwdev->hal.cv > CHIP_CAV)
+		rtw89_phy_write32_mask(rtwdev, R_DPD_COM + (path << 8),
+				       B_DPD_COM_OF, 0x1);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d restore KIP\n", path);
+}
+
+static void _dpk_lbk_rxiqk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			   enum rtw89_rf_path path, u8 cur_rxbb, u32 rf_18)
+{
+	rtw89_phy_write32_mask(rtwdev, R_MDPK_RX_DCK, B_MDPK_RX_DCK_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_IQK_RES + (path << 8), B_IQK_RES_RXCFIR, 0x0);
+
+	rtw89_write_rf(rtwdev, path, RR_RSV4, RFREG_MASK, rf_18);
+	rtw89_write_rf(rtwdev, path, RR_MOD, RFREG_MASKMODE, 0xd);
+	rtw89_write_rf(rtwdev, path, RR_RXK, RR_RXK_PLLEN, 0x1);
+
+	if (cur_rxbb >= 0x11)
+		rtw89_write_rf(rtwdev, path, RR_TXIQK, RR_TXIQK_ATT1, 0x13);
+	else if (cur_rxbb <= 0xa)
+		rtw89_write_rf(rtwdev, path, RR_TXIQK, RR_TXIQK_ATT1, 0x00);
+	else
+		rtw89_write_rf(rtwdev, path, RR_TXIQK, RR_TXIQK_ATT1, 0x05);
+
+	rtw89_write_rf(rtwdev, path, RR_XGLNA2, RR_XGLNA2_SW, 0x0);
+	rtw89_write_rf(rtwdev, path, RR_RXKPLL, RR_RXKPLL_POW, 0x0);
+	rtw89_write_rf(rtwdev, path, RR_RXKPLL, RFREG_MASK, 0x80014);
+
+	udelay(100);
+
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_IQK_DIF4, B_IQK_DIF4_RXT, 0x025);
+
+	_dpk_one_shot(rtwdev, phy, path, LBK_RXIQK);
+
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+	rtw89_write_rf(rtwdev, path, RR_RXK, RR_RXK_PLLEN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_MDPK_RX_DCK, B_MDPK_RX_DCK_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_KPATH_CFG, B_KPATH_CFG_ED, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_LOAD_COEF + (path << 8), B_LOAD_COEF_DI, 0x1);
+	rtw89_write_rf(rtwdev, path, RR_MOD, RFREG_MASKMODE, 0x5);
+}
+
+static void _dpk_rf_setting(struct rtw89_dev *rtwdev, u8 gain,
+			    enum rtw89_rf_path path, u8 kidx)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+
+	if (dpk->bp[path][kidx].band == RTW89_BAND_2G) {
+		rtw89_write_rf(rtwdev, path, RR_MOD, RFREG_MASK, 0x50220);
+		rtw89_write_rf(rtwdev, path, RR_RXBB, RR_RXBB_FATT, 0xf2);
+		rtw89_write_rf(rtwdev, path, RR_LUTDBG, RR_LUTDBG_TIA, 0x1);
+		rtw89_write_rf(rtwdev, path, RR_TIA, RR_TIA_N6, 0x1);
+	} else {
+		rtw89_write_rf(rtwdev, path, RR_MOD, RFREG_MASK, 0x50220);
+		rtw89_write_rf(rtwdev, path, RR_RXA2, RR_RAA2_SWATT, 0x5);
+		rtw89_write_rf(rtwdev, path, RR_LUTDBG, RR_LUTDBG_TIA, 0x1);
+		rtw89_write_rf(rtwdev, path, RR_TIA, RR_TIA_N6, 0x1);
+		rtw89_write_rf(rtwdev, path, RR_RXA_LNA, RFREG_MASK, 0x920FC);
+		rtw89_write_rf(rtwdev, path, RR_XALNA2, RFREG_MASK, 0x002C0);
+		rtw89_write_rf(rtwdev, path, RR_IQGEN, RFREG_MASK, 0x38800);
+	}
+
+	rtw89_write_rf(rtwdev, path, RR_RCKD, RR_RCKD_BW, 0x1);
+	rtw89_write_rf(rtwdev, path, RR_BTC, RR_BTC_TXBB, dpk->bp[path][kidx].bw + 1);
+	rtw89_write_rf(rtwdev, path, RR_BTC, RR_BTC_RXBB, 0x0);
+}
+
+static void _dpk_bypass_rxcfir(struct rtw89_dev *rtwdev,
+			       enum rtw89_rf_path path, bool is_bypass)
+{
+	if (is_bypass) {
+		rtw89_phy_write32_mask(rtwdev, R_RXIQC + (path << 8),
+				       B_RXIQC_BYPASS2, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_RXIQC + (path << 8),
+				       B_RXIQC_BYPASS, 0x1);
+	} else {
+		rtw89_phy_write32_mask(rtwdev, R_RXIQC + (path << 8),
+				       B_RXIQC_BYPASS2, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_RXIQC + (path << 8),
+				       B_RXIQC_BYPASS, 0x0);
+	}
+}
+
+static
+void _dpk_tpg_sel(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, u8 kidx)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+
+	if (dpk->bp[path][kidx].bw == RTW89_CHANNEL_WIDTH_80)
+		rtw89_phy_write32_mask(rtwdev, R_TPG_MOD, B_TPG_MOD_F, 0x0);
+	else if (dpk->bp[path][kidx].bw == RTW89_CHANNEL_WIDTH_40)
+		rtw89_phy_write32_mask(rtwdev, R_TPG_MOD, B_TPG_MOD_F, 0x2);
+	else
+		rtw89_phy_write32_mask(rtwdev, R_TPG_MOD, B_TPG_MOD_F, 0x1);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] TPG_Select for %s\n",
+		    dpk->bp[path][kidx].bw == RTW89_CHANNEL_WIDTH_80 ? "80M" :
+		    dpk->bp[path][kidx].bw == RTW89_CHANNEL_WIDTH_40 ? "40M" : "20M");
+}
+
+static void _dpk_table_select(struct rtw89_dev *rtwdev,
+			      enum rtw89_rf_path path, u8 kidx, u8 gain)
+{
+	u8 val;
+
+	val = 0x80 + kidx * 0x20 + gain * 0x10;
+	rtw89_phy_write32_mask(rtwdev, R_DPD_CH0 + (path << 8), MASKBYTE3, val);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DPK] table select for Kidx[%d], Gain[%d] (0x%x)\n", kidx,
+		    gain, val);
+}
+
+static bool _dpk_sync_check(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, u8 kidx)
+{
+#define DPK_SYNC_TH_DC_I 200
+#define DPK_SYNC_TH_DC_Q 200
+#define DPK_SYNC_TH_CORR 170
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+	u8 corr_val, corr_idx;
+	u16 dc_i, dc_q;
+	u32 corr, dc;
+
+	rtw89_phy_write32_mask(rtwdev, R_KIP_RPT1, B_KIP_RPT1_SEL, 0x0);
+
+	corr = rtw89_phy_read32_mask(rtwdev, R_RPT_COM, MASKDWORD);
+	corr_idx = u32_get_bits(corr, B_PRT_COM_CORI);
+	corr_val = u32_get_bits(corr, B_PRT_COM_CORV);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DPK] S%d Corr_idx / Corr_val = %d / %d\n",
+		    path, corr_idx, corr_val);
+
+	dpk->corr_idx[path][kidx] = corr_idx;
+	dpk->corr_val[path][kidx] = corr_val;
+
+	rtw89_phy_write32_mask(rtwdev, R_KIP_RPT1, B_KIP_RPT1_SEL, 0x9);
+
+	dc = rtw89_phy_read32_mask(rtwdev, R_RPT_COM, MASKDWORD);
+	dc_i = u32_get_bits(dc, B_PRT_COM_DCI);
+	dc_q = u32_get_bits(dc, B_PRT_COM_DCQ);
+
+	dc_i = abs(sign_extend32(dc_i, 11));
+	dc_q = abs(sign_extend32(dc_q, 11));
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d DC I/Q, = %d / %d\n",
+		    path, dc_i, dc_q);
+
+	dpk->dc_i[path][kidx] = dc_i;
+	dpk->dc_q[path][kidx] = dc_q;
+
+	if (dc_i > DPK_SYNC_TH_DC_I || dc_q > DPK_SYNC_TH_DC_Q ||
+	    corr_val < DPK_SYNC_TH_CORR)
+		return true;
+	else
+		return false;
+}
+
+static void _dpk_sync(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+		      enum rtw89_rf_path path, u8 kidx)
+{
+	_dpk_one_shot(rtwdev, phy, path, SYNC);
+}
+
+static u16 _dpk_dgain_read(struct rtw89_dev *rtwdev)
+{
+	u16 dgain;
+
+	rtw89_phy_write32_mask(rtwdev, R_KIP_RPT1, B_KIP_RPT1_SEL, 0x0);
+
+	dgain = rtw89_phy_read32_mask(rtwdev, R_RPT_COM, B_PRT_COM_DCI);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] DGain = 0x%x\n", dgain);
+
+	return dgain;
+}
+
+static s8 _dpk_dgain_mapping(struct rtw89_dev *rtwdev, u16 dgain)
+{
+	static const u16 bnd[15] = {
+		0xbf1, 0xaa5, 0x97d, 0x875, 0x789, 0x6b7, 0x5fc, 0x556,
+		0x4c1, 0x43d, 0x3c7, 0x35e, 0x2ac, 0x262, 0x220
+	};
+	s8 offset;
+
+	if (dgain >= bnd[0])
+		offset = 0x6;
+	else if (bnd[0] > dgain && dgain >= bnd[1])
+		offset = 0x6;
+	else if (bnd[1] > dgain && dgain >= bnd[2])
+		offset = 0x5;
+	else if (bnd[2] > dgain && dgain >= bnd[3])
+		offset = 0x4;
+	else if (bnd[3] > dgain && dgain >= bnd[4])
+		offset = 0x3;
+	else if (bnd[4] > dgain && dgain >= bnd[5])
+		offset = 0x2;
+	else if (bnd[5] > dgain && dgain >= bnd[6])
+		offset = 0x1;
+	else if (bnd[6] > dgain && dgain >= bnd[7])
+		offset = 0x0;
+	else if (bnd[7] > dgain && dgain >= bnd[8])
+		offset = 0xff;
+	else if (bnd[8] > dgain && dgain >= bnd[9])
+		offset = 0xfe;
+	else if (bnd[9] > dgain && dgain >= bnd[10])
+		offset = 0xfd;
+	else if (bnd[10] > dgain && dgain >= bnd[11])
+		offset = 0xfc;
+	else if (bnd[11] > dgain && dgain >= bnd[12])
+		offset = 0xfb;
+	else if (bnd[12] > dgain && dgain >= bnd[13])
+		offset = 0xfa;
+	else if (bnd[13] > dgain && dgain >= bnd[14])
+		offset = 0xf9;
+	else if (bnd[14] > dgain)
+		offset = 0xf8;
+	else
+		offset = 0x0;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] DGain offset = %d\n", offset);
+
+	return offset;
+}
+
+static u8 _dpk_gainloss_read(struct rtw89_dev *rtwdev)
+{
+	rtw89_phy_write32_mask(rtwdev, R_KIP_RPT1, B_KIP_RPT1_SEL, 0x6);
+	rtw89_phy_write32_mask(rtwdev, R_DPK_CFG2, B_DPK_CFG2_ST, 0x1);
+
+	return rtw89_phy_read32_mask(rtwdev, R_RPT_COM, B_PRT_COM_GL);
+}
+
+static void _dpk_gainloss(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			  enum rtw89_rf_path path, u8 kidx)
+{
+	_dpk_one_shot(rtwdev, phy, path, GAIN_LOSS);
+
+	rtw89_phy_write32_mask(rtwdev, R_KIP_RPT1, B_KIP_RPT1_SEL, 0x6);
+	rtw89_phy_write32_mask(rtwdev, R_DPK_CFG2, B_DPK_CFG2_ST, 0x1);
+}
+
+static void _dpk_kip_preset(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			    enum rtw89_rf_path path, u8 kidx)
+{
+	_dpk_tpg_sel(rtwdev, path, kidx);
+	_dpk_one_shot(rtwdev, phy, path, KIP_PRESET);
+}
+
+static void _dpk_kip_pwr_clk_on(struct rtw89_dev *rtwdev,
+				enum rtw89_rf_path path)
+{
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_RPT, MASKDWORD, 0x00000080);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_SYSCFG, MASKDWORD, 0x807f030a);
+	rtw89_phy_write32_mask(rtwdev, R_CFIR_SYS + (path << 8), MASKDWORD, 0xce000a08);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] KIP Power/CLK on\n");
+}
+
+static
+u8 _dpk_txagc_check_8852bt(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, u8 txagc)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+
+	if (txagc >= dpk->max_dpk_txagc[path])
+		txagc = dpk->max_dpk_txagc[path];
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] Set TxAGC = 0x%x\n", txagc);
+
+	return txagc;
+}
+
+static void _dpk_kip_set_txagc(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			       enum rtw89_rf_path path, u8 txagc)
+{
+	u8 val;
+
+	val = _dpk_txagc_check_8852bt(rtwdev, path, txagc);
+	rtw89_write_rf(rtwdev, path, RR_TXAGC, RFREG_MASK, val);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+	_dpk_one_shot(rtwdev, phy, path, DPK_TXAGC);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] set TXAGC = 0x%x\n", txagc);
+}
+
+static void _dpk_kip_set_rxagc(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			       enum rtw89_rf_path path)
+{
+	rtw89_phy_write32_mask(rtwdev, R_KIP_MOD, B_KIP_MOD, 0x50220);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+	_dpk_one_shot(rtwdev, phy, path, DPK_RXAGC);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+}
+
+static u8 _dpk_set_offset(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			  enum rtw89_rf_path path, u8 txagc, s8 gain_offset)
+{
+	txagc = rtw89_read_rf(rtwdev, path, RR_TXAGC, RFREG_MASK);
+
+	if ((txagc - gain_offset) < DPK_TXAGC_LOWER)
+		txagc = DPK_TXAGC_LOWER;
+	else if ((txagc - gain_offset) > DPK_TXAGC_UPPER)
+		txagc = DPK_TXAGC_UPPER;
+	else
+		txagc = txagc - gain_offset;
+
+	_dpk_kip_set_txagc(rtwdev, phy, path, txagc);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] tmp_txagc (GL=%d) = 0x%x\n",
+		    gain_offset, txagc);
+	return txagc;
+}
+
+static bool _dpk_pas_read(struct rtw89_dev *rtwdev, enum rtw89_rf_path path,
+			  u8 is_check)
+{
+	u32 val1_i = 0, val1_q = 0, val2_i = 0, val2_q = 0;
+	u8 i;
+
+	rtw89_phy_write32_mask(rtwdev, R_KIP_RPT1, MASKBYTE2, 0x06);
+	rtw89_phy_write32_mask(rtwdev, R_DPK_CFG2, B_DPK_CFG2_ST, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_DPK_CFG3, MASKBYTE2, 0x08);
+
+	if (is_check) {
+		rtw89_phy_write32_mask(rtwdev, R_DPK_CFG3, MASKBYTE3, 0x00);
+		val1_i = rtw89_phy_read32_mask(rtwdev, R_RPT_COM, MASKHWORD);
+		val1_i = abs(sign_extend32(val1_i, 11));
+		val1_q = rtw89_phy_read32_mask(rtwdev, R_RPT_COM, MASKLWORD);
+		val1_q = abs(sign_extend32(val1_q, 11));
+
+		rtw89_phy_write32_mask(rtwdev, R_DPK_CFG3, MASKBYTE3, 0x1f);
+		val2_i = rtw89_phy_read32_mask(rtwdev, R_RPT_COM, MASKHWORD);
+		val2_i = abs(sign_extend32(val2_i, 11));
+		val2_q = rtw89_phy_read32_mask(rtwdev, R_RPT_COM, MASKLWORD);
+		val2_q = abs(sign_extend32(val2_q, 11));
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] PAS_delta = 0x%x\n",
+			    phy_div(val1_i * val1_i + val1_q * val1_q,
+				    val2_i * val2_i + val2_q * val2_q));
+	} else {
+		for (i = 0; i < 32; i++) {
+			rtw89_phy_write32_mask(rtwdev, R_DPK_CFG3, MASKBYTE3, i);
+			rtw89_debug(rtwdev, RTW89_DBG_RFK,
+				    "[DPK] PAS_Read[%02d]= 0x%08x\n", i,
+				    rtw89_phy_read32_mask(rtwdev, R_RPT_COM, MASKDWORD));
+		}
+	}
+
+	if (val1_i * val1_i + val1_q * val1_q >=
+	    (val2_i * val2_i + val2_q * val2_q) * 8 / 5)
+		return true;
+
+	return false;
+}
+
+static u8 _dpk_agc(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+		   enum rtw89_rf_path path, u8 kidx, u8 init_txagc,
+		   bool loss_only)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+	u8 goout = 0, agc_cnt = 0, limited_rxbb = 0, gl_cnt = 0;
+	u8 tmp_txagc, tmp_rxbb, tmp_gl_idx = 0;
+	u8 step = DPK_AGC_STEP_SYNC_DGAIN;
+	int limit = 200;
+	s8 offset = 0;
+	u16 dgain = 0;
+	u32 rf_18;
+
+	tmp_txagc = init_txagc;
+
+	tmp_rxbb = rtw89_read_rf(rtwdev, path, RR_MOD, RFREG_MASKRXBB);
+	rf_18 = rtw89_read_rf(rtwdev, path, RR_CFGCH, RFREG_MASK);
+
+	do {
+		switch (step) {
+		case DPK_AGC_STEP_SYNC_DGAIN:
+			_dpk_sync(rtwdev, phy, path, kidx);
+			if (agc_cnt == 0) {
+				if (chan->band_width < 2)
+					_dpk_bypass_rxcfir(rtwdev, path, true);
+				else
+					_dpk_lbk_rxiqk(rtwdev, phy, path,
+						       tmp_rxbb, rf_18);
+			}
+
+			if (_dpk_sync_check(rtwdev, path, kidx) == true) {
+				tmp_txagc = 0xff;
+				goout = 1;
+				break;
+			}
+
+			dgain = _dpk_dgain_read(rtwdev);
+			offset = _dpk_dgain_mapping(rtwdev, dgain);
+
+			if (loss_only == 1 || limited_rxbb == 1 || offset == 0)
+				step = DPK_AGC_STEP_GAIN_LOSS_IDX;
+			else
+				step = DPK_AGC_STEP_GAIN_ADJ;
+			break;
+		case DPK_AGC_STEP_GAIN_ADJ:
+			tmp_rxbb = rtw89_read_rf(rtwdev, path, RR_MOD, RFREG_MASKRXBB);
+
+			if (tmp_rxbb + offset > 0x1f) {
+				tmp_rxbb = 0x1f;
+				limited_rxbb = 1;
+			} else if (tmp_rxbb + offset < 0) {
+				tmp_rxbb = 0;
+				limited_rxbb = 1;
+			} else {
+				tmp_rxbb = tmp_rxbb + offset;
+			}
+
+			rtw89_write_rf(rtwdev, path, RR_MOD, RFREG_MASKRXBB, tmp_rxbb);
+			rtw89_debug(rtwdev, RTW89_DBG_RFK,
+				    "[DPK] Adjust RXBB (%d) = 0x%x\n", offset, tmp_rxbb);
+
+			if (chan->band_width == RTW89_CHANNEL_WIDTH_80)
+				_dpk_lbk_rxiqk(rtwdev, phy, path, tmp_rxbb, rf_18);
+			if (dgain > 1922 || dgain < 342)
+				step = DPK_AGC_STEP_SYNC_DGAIN;
+			else
+				step = DPK_AGC_STEP_GAIN_LOSS_IDX;
+
+			agc_cnt++;
+			break;
+		case DPK_AGC_STEP_GAIN_LOSS_IDX:
+			_dpk_gainloss(rtwdev, phy, path, kidx);
+
+			tmp_gl_idx = _dpk_gainloss_read(rtwdev);
+
+			if ((tmp_gl_idx == 0 && _dpk_pas_read(rtwdev, path, true)) ||
+			    tmp_gl_idx >= 7)
+				step = DPK_AGC_STEP_GL_GT_CRITERION;
+			else if (tmp_gl_idx == 0)
+				step = DPK_AGC_STEP_GL_LT_CRITERION;
+			else
+				step = DPK_AGC_STEP_SET_TX_GAIN;
+
+			gl_cnt++;
+			break;
+		case DPK_AGC_STEP_GL_GT_CRITERION:
+			if (tmp_txagc == 0x2e ||
+			    tmp_txagc == dpk->max_dpk_txagc[path]) {
+				goout = 1;
+				rtw89_debug(rtwdev, RTW89_DBG_RFK,
+					    "[DPK] Txagc@lower bound!!\n");
+			} else {
+				tmp_txagc = _dpk_set_offset(rtwdev, phy, path,
+							    tmp_txagc, 0x3);
+			}
+			step = DPK_AGC_STEP_GAIN_LOSS_IDX;
+			agc_cnt++;
+			break;
+
+		case DPK_AGC_STEP_GL_LT_CRITERION:
+			if (tmp_txagc == 0x3f || tmp_txagc == dpk->max_dpk_txagc[path]) {
+				goout = 1;
+				rtw89_debug(rtwdev, RTW89_DBG_RFK,
+					    "[DPK] Txagc@upper bound!!\n");
+			} else {
+				tmp_txagc = _dpk_set_offset(rtwdev, phy, path,
+							    tmp_txagc, 0xfe);
+			}
+			step = DPK_AGC_STEP_GAIN_LOSS_IDX;
+			agc_cnt++;
+			break;
+
+		case DPK_AGC_STEP_SET_TX_GAIN:
+			tmp_txagc = _dpk_set_offset(rtwdev, phy, path, tmp_txagc,
+						    tmp_gl_idx);
+			goout = 1;
+			agc_cnt++;
+			break;
+
+		default:
+			goout = 1;
+			break;
+		}
+	} while (!goout && agc_cnt < 6 && limit-- > 0);
+
+	if (gl_cnt >= 6)
+		_dpk_pas_read(rtwdev, path, false);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DPK] Txagc / RXBB for DPK = 0x%x / 0x%x\n", tmp_txagc, tmp_rxbb);
+
+	return tmp_txagc;
+}
+
+static void _dpk_set_mdpd_para(struct rtw89_dev *rtwdev,
+			       enum rtw89_rf_path path, u8 order)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+
+	switch (order) {
+	case 0: /* (5,3,1) */
+		rtw89_phy_write32_mask(rtwdev, R_LDL_NORM, B_LDL_NORM_OP, order);
+		rtw89_phy_write32_mask(rtwdev, R_LDL_NORM, B_LDL_NORM_PN, 0x3);
+		rtw89_phy_write32_mask(rtwdev, R_MDPK_SYNC, B_MDPK_SYNC_MAN, 0x1);
+		dpk->dpk_order[path] = 0x3;
+		break;
+	case 1: /* (5,3,0) */
+		rtw89_phy_write32_mask(rtwdev, R_LDL_NORM, B_LDL_NORM_OP, order);
+		rtw89_phy_write32_mask(rtwdev, R_LDL_NORM, B_LDL_NORM_PN, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_MDPK_SYNC, B_MDPK_SYNC_MAN, 0x0);
+		dpk->dpk_order[path] = 0x1;
+		break;
+	case 2: /* (5,0,0) */
+		rtw89_phy_write32_mask(rtwdev, R_LDL_NORM, B_LDL_NORM_OP, order);
+		rtw89_phy_write32_mask(rtwdev, R_LDL_NORM, B_LDL_NORM_PN, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_MDPK_SYNC, B_MDPK_SYNC_MAN, 0x0);
+		dpk->dpk_order[path] = 0x0;
+		break;
+	default:
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[DPK] Wrong MDPD order!!(0x%x)\n", order);
+		break;
+	}
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] Set %s for IDL\n",
+		    order == 0x0 ? "(5,3,1)" :
+		    order == 0x1 ? "(5,3,0)" : "(5,0,0)");
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DPK] Set MDPD order to 0x%x for IDL\n", order);
+}
+
+static void _dpk_idl_mpa(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			 enum rtw89_rf_path path, u8 kidx, u8 gain)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+
+	if (dpk->bp[path][kidx].bw < RTW89_CHANNEL_WIDTH_80 &&
+	    dpk->bp[path][kidx].band == RTW89_BAND_5G)
+		_dpk_set_mdpd_para(rtwdev, path, 0x2);
+	else
+		_dpk_set_mdpd_para(rtwdev, path, 0x0);
+
+	_dpk_one_shot(rtwdev, phy, path, MDPK_IDL);
+}
+
+static void _dpk_fill_result(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			     enum rtw89_rf_path path, u8 kidx, u8 gain, u8 txagc)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+	u8 gs = dpk->dpk_gs[phy];
+	u16 pwsf = 0x78;
+
+	rtw89_phy_write32_mask(rtwdev, R_COEF_SEL + (path << 8), BIT(8), kidx);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DPK] Fill txagc/ pwsf/ gs = 0x%x/ 0x%x/ 0x%x\n",
+		    txagc, pwsf, gs);
+
+	dpk->bp[path][kidx].txagc_dpk = txagc;
+	rtw89_phy_write32_mask(rtwdev, R_TXAGC_RFK + (path << 8),
+			       0x3F << ((gain << 3) + (kidx << 4)), txagc);
+
+	dpk->bp[path][kidx].pwsf = pwsf;
+	rtw89_phy_write32_mask(rtwdev, R_DPD_BND + (path << 8) + (kidx << 2),
+			       0x1FF << (gain << 4), pwsf);
+
+	rtw89_phy_write32_mask(rtwdev, R_LOAD_COEF + (path << 8), B_LOAD_COEF_MDPD, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_LOAD_COEF + (path << 8), B_LOAD_COEF_MDPD, 0x0);
+
+	dpk->bp[path][kidx].gs = gs;
+	if (dpk->dpk_gs[phy] == 0x7f)
+		rtw89_phy_write32_mask(rtwdev,
+				       R_DPD_CH0A + (path << 8) + (kidx << 2),
+				       MASKDWORD, 0x007f7f7f);
+	else
+		rtw89_phy_write32_mask(rtwdev,
+				       R_DPD_CH0A + (path << 8) + (kidx << 2),
+				       MASKDWORD, 0x005b5b5b);
+
+	rtw89_phy_write32_mask(rtwdev, R_DPD_CH0A + (path << 8) + (kidx << 2),
+			       B_DPD_ORDER_V1, dpk->dpk_order[path]);
+
+	rtw89_phy_write32_mask(rtwdev, R_DPD_V1 + (path << 8), MASKDWORD, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_MDPK_SYNC, B_MDPK_SYNC_SEL, 0x0);
+}
+
+static bool _dpk_reload_check(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			      enum rtw89_rf_path path)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+	u8 idx, cur_band, cur_ch;
+	bool is_reload = false;
+
+	cur_band = chan->band_type;
+	cur_ch = chan->channel;
+
+	for (idx = 0; idx < RTW89_DPK_BKUP_NUM; idx++) {
+		if (cur_band != dpk->bp[path][idx].band ||
+		    cur_ch != dpk->bp[path][idx].ch)
+			continue;
+
+		rtw89_phy_write32_mask(rtwdev, R_COEF_SEL + (path << 8),
+				       B_COEF_SEL_MDPD, idx);
+		dpk->cur_idx[path] = idx;
+		is_reload = true;
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[DPK] reload S%d[%d] success\n", path, idx);
+	}
+
+	return is_reload;
+}
+
+static
+void _rf_direct_cntrl(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, bool is_bybb)
+{
+	if (is_bybb)
+		rtw89_write_rf(rtwdev, path, RR_RSV1, RR_RSV1_RST, 0x1);
+	else
+		rtw89_write_rf(rtwdev, path, RR_RSV1, RR_RSV1_RST, 0x0);
+}
+
+static
+void _drf_direct_cntrl(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, bool is_bybb)
+{
+	if (is_bybb)
+		rtw89_write_rf(rtwdev, path, RR_BBDC, RR_BBDC_SEL, 0x1);
+	else
+		rtw89_write_rf(rtwdev, path, RR_BBDC, RR_BBDC_SEL, 0x0);
+}
+
+static bool _dpk_main(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+		      enum rtw89_rf_path path, u8 gain)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+	u8 txagc = 0x38, kidx = dpk->cur_idx[path];
+	bool is_fail = false;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DPK] ========= S%d[%d] DPK Start =========\n", path, kidx);
+
+	_rf_direct_cntrl(rtwdev, path, false);
+	_drf_direct_cntrl(rtwdev, path, false);
+
+	_dpk_kip_pwr_clk_on(rtwdev, path);
+	_dpk_kip_set_txagc(rtwdev, phy, path, txagc);
+	_dpk_rf_setting(rtwdev, gain, path, kidx);
+	_dpk_rx_dck(rtwdev, phy, path);
+	_dpk_kip_preset(rtwdev, phy, path, kidx);
+	_dpk_kip_set_rxagc(rtwdev, phy, path);
+	_dpk_table_select(rtwdev, path, kidx, gain);
+
+	txagc = _dpk_agc(rtwdev, phy, path, kidx, txagc, false);
+
+	_rfk_get_thermal(rtwdev, kidx, path);
+
+	if (txagc == 0xff) {
+		is_fail = true;
+		goto _error;
+	}
+
+	_dpk_idl_mpa(rtwdev, phy, path, kidx, gain);
+
+	rtw89_write_rf(rtwdev, path, RR_MOD, RFREG_MASKMODE, RF_RX);
+	_dpk_fill_result(rtwdev, phy, path, kidx, gain, txagc);
+
+_error:
+	if (!is_fail)
+		dpk->bp[path][kidx].path_ok = 1;
+	else
+		dpk->bp[path][kidx].path_ok = 0;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d[%d] DPK %s\n", path, kidx,
+		    is_fail ? "Check" : "Success");
+
+	_dpk_onoff(rtwdev, path, is_fail);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DPK] S%d[%d] DPK %s\n", path, kidx,
+		    is_fail ? "Check" : "Success");
+
+	return is_fail;
+}
+
+static void _dpk_cal_select(struct rtw89_dev *rtwdev,
+			    enum rtw89_phy_idx phy, u8 kpath)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+	u32 backup_kip_val[BACKUP_KIP_REGS_NR];
+	u32 backup_bb_val[BACKUP_BB_REGS_NR];
+	u32 backup_rf_val[RTW8852BT_SS][BACKUP_RF_REGS_NR];
+	bool reloaded[2] = {false};
+	u8 path;
+
+	for (path = 0; path < DPK_RF_PATH_MAX_8852BT; path++) {
+		reloaded[path] = _dpk_reload_check(rtwdev, phy, path);
+		if (!reloaded[path] && dpk->bp[path][0].ch != 0)
+			dpk->cur_idx[path] = !dpk->cur_idx[path];
+		else
+			_dpk_onoff(rtwdev, path, false);
+	}
+
+	_rfk_backup_bb_reg(rtwdev, backup_bb_val);
+	_rfk_backup_kip_reg(rtwdev, backup_kip_val);
+
+	for (path = 0; path < DPK_RF_PATH_MAX_8852BT; path++) {
+		_rfk_backup_rf_reg(rtwdev, backup_rf_val[path], path);
+		_dpk_information(rtwdev, phy, path);
+		if (rtwdev->is_tssi_mode[path])
+			_dpk_tssi_pause(rtwdev, path, true);
+	}
+
+	_rfk_bb_afe_setting(rtwdev, phy, path, kpath);
+
+	for (path = 0; path < DPK_RF_PATH_MAX_8852BT; path++)
+		_dpk_main(rtwdev, phy, path, 1);
+
+	_rfk_bb_afe_restore(rtwdev, phy, path, kpath);
+
+	_dpk_kip_restore(rtwdev, path);
+	_rfk_reload_bb_reg(rtwdev, backup_bb_val);
+	_rfk_reload_kip_reg(rtwdev, backup_kip_val);
+
+	for (path = 0; path < DPK_RF_PATH_MAX_8852BT; path++) {
+		_rfk_reload_rf_reg(rtwdev, backup_rf_val[path], path);
+		if (rtwdev->is_tssi_mode[path])
+			_dpk_tssi_pause(rtwdev, path, false);
+	}
+}
+
+static bool _dpk_bypass_check(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	struct rtw89_fem_info *fem = &rtwdev->fem;
+
+	if (fem->epa_2g && chan->band_type == RTW89_BAND_2G) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[DPK] Skip DPK due to 2G_ext_PA exist!!\n");
+		return true;
+	} else if (fem->epa_5g && chan->band_type == RTW89_BAND_5G) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[DPK] Skip DPK due to 5G_ext_PA exist!!\n");
+		return true;
+	} else if (fem->epa_6g && chan->band_type == RTW89_BAND_6G) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[DPK] Skip DPK due to 6G_ext_PA exist!!\n");
+		return true;
+	}
+
+	return false;
+}
+
+static void _dpk_force_bypass(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
+{
+	u8 path, kpath;
+
+	kpath = _kpath(rtwdev, phy);
+
+	for (path = 0; path < RTW8852BT_SS; path++) {
+		if (kpath & BIT(path))
+			_dpk_onoff(rtwdev, path, true);
+	}
+}
+
+static void _dpk_track(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+	s8 txagc_bb, txagc_bb_tp, ini_diff = 0, txagc_ofst;
+	s8 delta_ther[2] = {};
+	u8 trk_idx, txagc_rf;
+	u8 path, kidx;
+	u16 pwsf[2];
+	u8 cur_ther;
+	u32 tmp;
+
+	for (path = 0; path < RF_PATH_NUM_8852BT; path++) {
+		kidx = dpk->cur_idx[path];
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK_TRACK,
+			    "[DPK_TRK] ================[S%d[%d] (CH %d)]================\n",
+			    path, kidx, dpk->bp[path][kidx].ch);
+
+		cur_ther = ewma_thermal_read(&rtwdev->phystat.avg_thermal[path]);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK_TRACK,
+			    "[DPK_TRK] thermal now = %d\n", cur_ther);
+
+		if (dpk->bp[path][kidx].ch && cur_ther)
+			delta_ther[path] = dpk->bp[path][kidx].ther_dpk - cur_ther;
+
+		if (dpk->bp[path][kidx].band == RTW89_BAND_2G)
+			delta_ther[path] = delta_ther[path] * 3 / 2;
+		else
+			delta_ther[path] = delta_ther[path] * 5 / 2;
+
+		txagc_rf = rtw89_phy_read32_mask(rtwdev, R_TXAGC_BB + (path << 13),
+						 B_TXAGC_RF);
+
+		if (rtwdev->is_tssi_mode[path]) {
+			trk_idx = rtw89_read_rf(rtwdev, path, RR_TXA, RR_TXA_TRK);
+
+			rtw89_debug(rtwdev, RTW89_DBG_RFK_TRACK,
+				    "[DPK_TRK] txagc_RF / track_idx = 0x%x / %d\n",
+				    txagc_rf, trk_idx);
+
+			txagc_bb =
+				rtw89_phy_read32_mask(rtwdev, R_TXAGC_BB + (path << 13),
+						      MASKBYTE2);
+			txagc_bb_tp =
+				rtw89_phy_read32_mask(rtwdev, R_TXAGC_TP + (path << 13),
+						      B_TXAGC_TP);
+
+			rtw89_debug(rtwdev, RTW89_DBG_RFK_TRACK,
+				    "[DPK_TRK] txagc_bb_tp / txagc_bb = 0x%x / 0x%x\n",
+				    txagc_bb_tp, txagc_bb);
+
+			txagc_ofst =
+				rtw89_phy_read32_mask(rtwdev, R_TXAGC_BB + (path << 13),
+						      MASKBYTE3);
+
+			rtw89_debug(rtwdev, RTW89_DBG_RFK_TRACK,
+				    "[DPK_TRK] txagc_offset / delta_ther = %d / %d\n",
+				    txagc_ofst, delta_ther[path]);
+			tmp = rtw89_phy_read32_mask(rtwdev, R_DPD_COM + (path << 8),
+						    B_DPD_COM_OF);
+			if (tmp == 0x1) {
+				txagc_ofst = 0;
+				rtw89_debug(rtwdev, RTW89_DBG_RFK_TRACK,
+					    "[DPK_TRK] HW txagc offset mode\n");
+			}
+
+			if (txagc_rf && cur_ther)
+				ini_diff = txagc_ofst + (delta_ther[path]);
+
+			tmp = rtw89_phy_read32_mask(rtwdev,
+						    R_P0_TXDPD + (path << 13),
+						    B_P0_TXDPD);
+			if (tmp == 0x0) {
+				pwsf[0] = dpk->bp[path][kidx].pwsf +
+					  txagc_bb_tp - txagc_bb + ini_diff;
+				pwsf[1] = dpk->bp[path][kidx].pwsf +
+					  txagc_bb_tp - txagc_bb + ini_diff;
+			} else {
+				pwsf[0] = dpk->bp[path][kidx].pwsf + ini_diff;
+				pwsf[1] = dpk->bp[path][kidx].pwsf + ini_diff;
+			}
+		} else {
+			pwsf[0] = (dpk->bp[path][kidx].pwsf + delta_ther[path]) & 0x1ff;
+			pwsf[1] = (dpk->bp[path][kidx].pwsf + delta_ther[path]) & 0x1ff;
+		}
+
+		tmp = rtw89_phy_read32_mask(rtwdev, R_DPK_TRK, B_DPK_TRK_DIS);
+		if (!tmp && txagc_rf) {
+			rtw89_debug(rtwdev, RTW89_DBG_RFK_TRACK,
+				    "[DPK_TRK] New pwsf[0] / pwsf[1] = 0x%x / 0x%x\n",
+				    pwsf[0], pwsf[1]);
+
+			rtw89_phy_write32_mask(rtwdev,
+					       R_DPD_BND + (path << 8) + (kidx << 2),
+					       B_DPD_BND_0, pwsf[0]);
+			rtw89_phy_write32_mask(rtwdev,
+					       R_DPD_BND + (path << 8) + (kidx << 2),
+					       B_DPD_BND_1, pwsf[1]);
+		}
+	}
+}
+
+static void _set_dpd_backoff(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+	u8 tx_scale, ofdm_bkof, path, kpath;
+
+	kpath = _kpath(rtwdev, phy);
+
+	ofdm_bkof = rtw89_phy_read32_mask(rtwdev, R_DPD_BF + (phy << 13), B_DPD_BF_OFDM);
+	tx_scale = rtw89_phy_read32_mask(rtwdev, R_DPD_BF + (phy << 13), B_DPD_BF_SCA);
+
+	if (ofdm_bkof + tx_scale >= 44) {
+		/* move dpd backoff to bb, and set dpd backoff to 0 */
+		dpk->dpk_gs[phy] = 0x7f;
+		for (path = 0; path < RF_PATH_NUM_8852BT; path++) {
+			if (!(kpath & BIT(path)))
+				continue;
+
+			rtw89_phy_write32_mask(rtwdev, R_DPD_CH0A + (path << 8),
+					       B_DPD_CFG, 0x7f7f7f);
+			rtw89_debug(rtwdev, RTW89_DBG_RFK,
+				    "[RFK] Set S%d DPD backoff to 0dB\n", path);
+		}
+	} else {
+		dpk->dpk_gs[phy] = 0x5b;
+	}
+}
+
 static void _tssi_dpk_off(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
 {
 	rtw89_phy_write32_mask(rtwdev, R_DPD_CH0A, BIT(24), 0x0);
@@ -1077,6 +2378,38 @@ out:
 		    tssi_info->tssi_alimk_time);
 }
 
+void rtw8852bt_dpk_init(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
+
+	u8 path;
+
+	for (path = 0; path < 2; path++) {
+		dpk->cur_idx[path] = 0;
+		dpk->max_dpk_txagc[path] = 0x3F;
+	}
+
+	dpk->is_dpk_enable = true;
+	dpk->is_dpk_reload_en = false;
+	_set_dpd_backoff(rtwdev, RTW89_PHY_0);
+}
+
+void rtw8852bt_dpk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
+{
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DPK] ****** DPK Start (Ver: 0x%x) ******\n", RTW8852BT_DPK_VER);
+
+	if (_dpk_bypass_check(rtwdev, phy_idx))
+		_dpk_force_bypass(rtwdev, phy_idx);
+	else
+		_dpk_cal_select(rtwdev, phy_idx, RF_AB);
+}
+
+void rtw8852bt_dpk_track(struct rtw89_dev *rtwdev)
+{
+	_dpk_track(rtwdev);
+}
+
 void rtw8852bt_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, bool hwtx_en)
 {
 	static const u32 reg[2] = {R_DPD_CH0A, R_DPD_CH0B};
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
index 03fad3abe554..89640b42d4d0 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
@@ -7,6 +7,9 @@
 
 #include "core.h"
 
+void rtw8852bt_dpk_init(struct rtw89_dev *rtwdev);
+void rtw8852bt_dpk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy);
+void rtw8852bt_dpk_track(struct rtw89_dev *rtwdev);
 void rtw8852bt_tssi(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, bool hwtx_en);
 void rtw8852bt_tssi_scan(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy);
 void rtw8852bt_wifi_scan_notify(struct rtw89_dev *rtwdev, bool scan_start,
-- 
cgit v1.2.3-58-ga151


From 5f7e92c59b8e03ac90130223c2ebb4449373b9cd Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Fri, 21 Jun 2024 20:36:16 +0800
Subject: wifi: rtw89: 8852b: set AMSDU limit to 5000

Enlarge AMSDU limit to improve MAC efficient to get better TX throughput
from 851 to 895 Mbps. No change for RX throughput 902 Mbps.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240621123617.6687-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852b.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index 24f9fd444ca0..1a9c4951060a 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -944,7 +944,7 @@ const struct rtw89_chip_info rtw8852b_chip_info = {
 	.fifo_size		= 196608,
 	.small_fifo_size	= true,
 	.dle_scc_rsvd_size	= 98304,
-	.max_amsdu_limit	= 3500,
+	.max_amsdu_limit	= 5000,
 	.dis_2g_40m_ul_ofdma	= true,
 	.rsvd_ple_ofst		= 0x2f800,
 	.hfc_param_ini		= rtw8852b_hfc_param_ini_pcie,
-- 
cgit v1.2.3-58-ga151


From 2f35712ab82683554c660bc2456f05785835efbe Mon Sep 17 00:00:00 2001
From: Kuan-Chung Chen <damon.chen@realtek.com>
Date: Fri, 21 Jun 2024 20:36:17 +0800
Subject: wifi: rtw89: 8852b: fix definition of KIP register number

An incorrect definition caused DPK to fail to backup and
restore a set of KIP registers. Fixing this will improve
RX throughput from 902 to 997 Mbps.

Fixes: 5b8471ace5b1 ("wifi: rtw89: 8852b: rfk: add DPK")
Signed-off-by: Kuan-Chung Chen <damon.chen@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240621123617.6687-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c
index 7c90e98dc1d9..12354612441c 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_rfk.c
@@ -21,7 +21,7 @@
 #define RTW8852B_RF_REL_VERSION 34
 #define RTW8852B_DPK_VER 0x0d
 #define RTW8852B_DPK_RF_PATH 2
-#define RTW8852B_DPK_KIP_REG_NUM 2
+#define RTW8852B_DPK_KIP_REG_NUM 3
 
 #define _TSSI_DE_MASK GENMASK(21, 12)
 #define ADDC_T_AVG 100
-- 
cgit v1.2.3-58-ga151


From d0b4b8ef083ca46d5d318e66a30fb80e0abbb90d Mon Sep 17 00:00:00 2001
From: Martin Kaistra <martin.kaistra@linutronix.de>
Date: Mon, 24 Jun 2024 16:00:37 +0200
Subject: wifi: rtl8xxxu: 8188f: Limit TX power index

TX power index is read from the efuse on init, the values get written to
the TX power registers when the channel gets switched.

When the chip has not yet been calibrated, the efuse values are 0xFF,
which on some boards leads to USB timeouts for reading/writing registers
after the first frames have been sent.

The vendor driver (v5.11.5-1) checks for these invalid values and sets
default values instead. Implement something similar in
rtl8188fu_parse_efuse().

Fixes: c888183b21f3 ("wifi: rtl8xxxu: Support new chip RTL8188FU")
Signed-off-by: Martin Kaistra <martin.kaistra@linutronix.de>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240624140037.231657-1-martin.kaistra@linutronix.de
---
 drivers/net/wireless/realtek/rtl8xxxu/8188f.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtl8xxxu/8188f.c b/drivers/net/wireless/realtek/rtl8xxxu/8188f.c
index bd5a0603b4a2..3abf14d7044f 100644
--- a/drivers/net/wireless/realtek/rtl8xxxu/8188f.c
+++ b/drivers/net/wireless/realtek/rtl8xxxu/8188f.c
@@ -697,9 +697,14 @@ static void rtl8188fu_init_statistics(struct rtl8xxxu_priv *priv)
 	rtl8xxxu_write32(priv, REG_OFDM0_FA_RSTC, val32);
 }
 
+#define TX_POWER_INDEX_MAX 0x3F
+#define TX_POWER_INDEX_DEFAULT_CCK 0x22
+#define TX_POWER_INDEX_DEFAULT_HT40 0x27
+
 static int rtl8188fu_parse_efuse(struct rtl8xxxu_priv *priv)
 {
 	struct rtl8188fu_efuse *efuse = &priv->efuse_wifi.efuse8188fu;
+	int i;
 
 	if (efuse->rtl_id != cpu_to_le16(0x8129))
 		return -EINVAL;
@@ -713,6 +718,16 @@ static int rtl8188fu_parse_efuse(struct rtl8xxxu_priv *priv)
 	       efuse->tx_power_index_A.ht40_base,
 	       sizeof(efuse->tx_power_index_A.ht40_base));
 
+	for (i = 0; i < ARRAY_SIZE(priv->cck_tx_power_index_A); i++) {
+		if (priv->cck_tx_power_index_A[i] > TX_POWER_INDEX_MAX)
+			priv->cck_tx_power_index_A[i] = TX_POWER_INDEX_DEFAULT_CCK;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(priv->ht40_1s_tx_power_index_A); i++) {
+		if (priv->ht40_1s_tx_power_index_A[i] > TX_POWER_INDEX_MAX)
+			priv->ht40_1s_tx_power_index_A[i] = TX_POWER_INDEX_DEFAULT_HT40;
+	}
+
 	priv->ofdm_tx_power_diff[0].a = efuse->tx_power_index_A.ht20_ofdm_1s_diff.a;
 	priv->ht20_tx_power_diff[0].a = efuse->tx_power_index_A.ht20_ofdm_1s_diff.b;
 
-- 
cgit v1.2.3-58-ga151


From 8b8fe280155d9fb6f3f6310f05f613cbdf30196b Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 25 Jun 2024 18:24:53 -0700
Subject: selftests: drv-net: try to check if port is in use

We use random ports for communication. As Willem predicted
this leads to occasional failures. Try to check if port is
already in use by opening a socket and binding to that port.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240626012456.2326192-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/lib/py/utils.py | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/net/lib/py/utils.py b/tools/testing/selftests/net/lib/py/utils.py
index 0540ea24921d..dd9d2b9f2b20 100644
--- a/tools/testing/selftests/net/lib/py/utils.py
+++ b/tools/testing/selftests/net/lib/py/utils.py
@@ -1,8 +1,10 @@
 # SPDX-License-Identifier: GPL-2.0
 
+import errno
 import json as _json
 import random
 import re
+import socket
 import subprocess
 import time
 
@@ -79,9 +81,18 @@ def ip(args, json=None, ns=None, host=None):
 
 def rand_port():
     """
-    Get unprivileged port, for now just random, one day we may decide to check if used.
+    Get a random unprivileged port, try to make sure it's not already used.
     """
-    return random.randint(10000, 65535)
+    for _ in range(1000):
+        port = random.randint(10000, 65535)
+        try:
+            with socket.socket(socket.AF_INET6, socket.SOCK_STREAM) as s:
+                s.bind(("", port))
+            return port
+        except OSError as e:
+            if e.errno != errno.EADDRINUSE:
+                raise
+    raise Exception("Can't find any free unprivileged port")
 
 
 def wait_port_listen(port, proto="tcp", ns=None, host=None, sleep=0.005, deadline=5):
-- 
cgit v1.2.3-58-ga151


From af8e51644a70f612974a6e767fa7d896d3c23f88 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 25 Jun 2024 18:24:54 -0700
Subject: selftests: drv-net: add helper to wait for HW stats to sync

Some devices DMA stats to the host periodically. Add a helper
which can wait for that to happen, based on frequency reported
by the driver in ethtool.

Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240626012456.2326192-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/lib/py/env.py | 19 ++++++++++++++++++-
 tools/testing/selftests/net/lib/py/utils.py       |  4 ++++
 2 files changed, 22 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/drivers/net/lib/py/env.py b/tools/testing/selftests/drivers/net/lib/py/env.py
index edcedd7bffab..a5e800b8f103 100644
--- a/tools/testing/selftests/drivers/net/lib/py/env.py
+++ b/tools/testing/selftests/drivers/net/lib/py/env.py
@@ -1,9 +1,10 @@
 # SPDX-License-Identifier: GPL-2.0
 
 import os
+import time
 from pathlib import Path
 from lib.py import KsftSkipEx, KsftXfailEx
-from lib.py import cmd, ip
+from lib.py import cmd, ethtool, ip
 from lib.py import NetNS, NetdevSimDev
 from .remote import Remote
 
@@ -82,6 +83,8 @@ class NetDrvEpEnv:
 
         self.env = _load_env_file(src_path)
 
+        self._stats_settle_time = None
+
         # Things we try to destroy
         self.remote = None
         # These are for local testing state
@@ -222,3 +225,17 @@ class NetDrvEpEnv:
         if remote:
             if not self._require_cmd(comm, "remote"):
                 raise KsftSkipEx("Test requires (remote) command: " + comm)
+
+    def wait_hw_stats_settle(self):
+        """
+        Wait for HW stats to become consistent, some devices DMA HW stats
+        periodically so events won't be reflected until next sync.
+        Good drivers will tell us via ethtool what their sync period is.
+        """
+        if self._stats_settle_time is None:
+            data = ethtool("-c " + self.ifname, json=True)[0]
+
+            self._stats_settle_time = 0.025 + \
+                data.get('stats-block-usecs', 0) / 1000 / 1000
+
+        time.sleep(self._stats_settle_time)
diff --git a/tools/testing/selftests/net/lib/py/utils.py b/tools/testing/selftests/net/lib/py/utils.py
index dd9d2b9f2b20..a6c87bfe532f 100644
--- a/tools/testing/selftests/net/lib/py/utils.py
+++ b/tools/testing/selftests/net/lib/py/utils.py
@@ -79,6 +79,10 @@ def ip(args, json=None, ns=None, host=None):
     return tool('ip', args, json=json, host=host)
 
 
+def ethtool(args, json=None, ns=None, host=None):
+    return tool('ethtool', args, json=json, ns=ns, host=host)
+
+
 def rand_port():
     """
     Get a random unprivileged port, try to make sure it's not already used.
-- 
cgit v1.2.3-58-ga151


From 94fecaa6dcd04889eff4291707b0433ddfc09039 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 25 Jun 2024 18:24:55 -0700
Subject: selftests: drv-net: add ability to wait for at least N packets to
 load gen

Teach the load generator how to wait for at least given number
of packets to be received. This will be useful for filtering
where we'll want to send a non-trivial number of packets and
make sure they landed in right queues.

Reviewed-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240626012456.2326192-4-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/lib/py/load.py | 30 +++++++++++++++++-----
 1 file changed, 23 insertions(+), 7 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/lib/py/load.py b/tools/testing/selftests/drivers/net/lib/py/load.py
index abdb677bdb1c..f6bc57eb304a 100644
--- a/tools/testing/selftests/drivers/net/lib/py/load.py
+++ b/tools/testing/selftests/drivers/net/lib/py/load.py
@@ -18,15 +18,31 @@ class GenerateTraffic:
                                  background=True, host=env.remote)
 
         # Wait for traffic to ramp up
-        pkt = ip("-s link show dev " + env.ifname, json=True)[0]["stats64"]["rx"]["packets"]
+        if not self._wait_pkts(pps=1000):
+            self.stop(verbose=True)
+            raise Exception("iperf3 traffic did not ramp up")
+
+    def _wait_pkts(self, pkt_cnt=None, pps=None):
+        """
+        Wait until we've seen pkt_cnt or until traffic ramps up to pps.
+        Only one of pkt_cnt or pss can be specified.
+        """
+        pkt_start = ip("-s link show dev " + self.env.ifname, json=True)[0]["stats64"]["rx"]["packets"]
         for _ in range(50):
             time.sleep(0.1)
-            now = ip("-s link show dev " + env.ifname, json=True)[0]["stats64"]["rx"]["packets"]
-            if now - pkt > 1000:
-                return
-            pkt = now
-        self.stop(verbose=True)
-        raise Exception("iperf3 traffic did not ramp up")
+            pkt_now = ip("-s link show dev " + self.env.ifname, json=True)[0]["stats64"]["rx"]["packets"]
+            if pps:
+                if pkt_now - pkt_start > pps / 10:
+                    return True
+                pkt_start = pkt_now
+            elif pkt_cnt:
+                if pkt_now - pkt_start > pkt_cnt:
+                    return True
+        return False
+
+    def wait_pkts_and_stop(self, pkt_cnt):
+        failed = not self._wait_pkts(pkt_cnt=pkt_cnt)
+        self.stop(verbose=failed)
 
     def stop(self, verbose=None):
         self._iperf_client.process(terminate=True)
-- 
cgit v1.2.3-58-ga151


From f898c16a0624e7f2dcb0b1cda6916c9be6489197 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 25 Jun 2024 18:24:56 -0700
Subject: selftests: drv-net: rss_ctx: add tests for RSS configuration and
 contexts

Add tests focusing on indirection table configuration and
creating extra RSS contexts in drivers which support it.

  $ export NETIF=eth0 REMOTE_...
  $ ./drivers/net/hw/rss_ctx.py
  KTAP version 1
  1..8
  ok 1 rss_ctx.test_rss_key_indir
  ok 2 rss_ctx.test_rss_context
  ok 3 rss_ctx.test_rss_context4
  # Increasing queue count 44 -> 66
  # Failed to create context 32, trying to test what we got
  ok 4 rss_ctx.test_rss_context32 # SKIP Tested only 31 contexts, wanted 32
  ok 5 rss_ctx.test_rss_context_overlap
  ok 6 rss_ctx.test_rss_context_overlap2
  # .. sprays traffic like a headless chicken ..
  not ok 7 rss_ctx.test_rss_context_out_of_order
  ok 8 rss_ctx.test_rss_context4_create_with_cfg
  # Totals: pass:6 fail:1 xfail:0 xpass:0 skip:1 error:0

Note that rss_ctx.test_rss_context_out_of_order fails with the device
I tested with, but it seems to be a device / driver bug.

Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240626012456.2326192-5-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/hw/Makefile    |   1 +
 tools/testing/selftests/drivers/net/hw/rss_ctx.py  | 383 +++++++++++++++++++++
 tools/testing/selftests/drivers/net/lib/py/load.py |   7 +-
 tools/testing/selftests/net/lib/py/ksft.py         |   5 +
 tools/testing/selftests/net/lib/py/utils.py        |   8 +-
 5 files changed, 399 insertions(+), 5 deletions(-)
 create mode 100755 tools/testing/selftests/drivers/net/hw/rss_ctx.py

diff --git a/tools/testing/selftests/drivers/net/hw/Makefile b/tools/testing/selftests/drivers/net/hw/Makefile
index 4933d045ab66..c9f2f48fc30f 100644
--- a/tools/testing/selftests/drivers/net/hw/Makefile
+++ b/tools/testing/selftests/drivers/net/hw/Makefile
@@ -11,6 +11,7 @@ TEST_PROGS = \
 	hw_stats_l3_gre.sh \
 	loopback.sh \
 	pp_alloc_fail.py \
+	rss_ctx.py \
 	#
 
 TEST_FILES := \
diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
new file mode 100755
index 000000000000..68c7d40214eb
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
@@ -0,0 +1,383 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+import datetime
+import random
+from lib.py import ksft_run, ksft_pr, ksft_exit, ksft_eq, ksft_ge, ksft_lt
+from lib.py import NetDrvEpEnv
+from lib.py import NetdevFamily
+from lib.py import KsftSkipEx
+from lib.py import rand_port
+from lib.py import ethtool, ip, GenerateTraffic, CmdExitFailure
+
+
+def _rss_key_str(key):
+    return ":".join(["{:02x}".format(x) for x in key])
+
+
+def _rss_key_rand(length):
+    return [random.randint(0, 255) for _ in range(length)]
+
+
+def get_rss(cfg, context=0):
+    return ethtool(f"-x {cfg.ifname} context {context}", json=True)[0]
+
+
+def get_drop_err_sum(cfg):
+    stats = ip("-s -s link show dev " + cfg.ifname, json=True)[0]
+    cnt = 0
+    for key in ['errors', 'dropped', 'over_errors', 'fifo_errors',
+                'length_errors', 'crc_errors', 'missed_errors',
+                'frame_errors']:
+        cnt += stats["stats64"]["rx"][key]
+    return cnt, stats["stats64"]["tx"]["carrier_changes"]
+
+
+def ethtool_create(cfg, act, opts):
+    output = ethtool(f"{act} {cfg.ifname} {opts}").stdout
+    # Output will be something like: "New RSS context is 1" or
+    # "Added rule with ID 7", we want the integer from the end
+    return int(output.split()[-1])
+
+
+def require_ntuple(cfg):
+    features = ethtool(f"-k {cfg.ifname}", json=True)[0]
+    if not features["ntuple-filters"]["active"]:
+        # ntuple is more of a capability than a config knob, don't bother
+        # trying to enable it (until some driver actually needs it).
+        raise KsftSkipEx("Ntuple filters not enabled on the device: " + str(features["ntuple-filters"]))
+
+
+# Get Rx packet counts for all queues, as a simple list of integers
+# if @prev is specified the prev counts will be subtracted
+def _get_rx_cnts(cfg, prev=None):
+    cfg.wait_hw_stats_settle()
+    data = cfg.netdevnl.qstats_get({"ifindex": cfg.ifindex, "scope": ["queue"]}, dump=True)
+    data = [x for x in data if x['queue-type'] == "rx"]
+    max_q = max([x["queue-id"] for x in data])
+    queue_stats = [0] * (max_q + 1)
+    for q in data:
+        queue_stats[q["queue-id"]] = q["rx-packets"]
+        if prev and q["queue-id"] < len(prev):
+            queue_stats[q["queue-id"]] -= prev[q["queue-id"]]
+    return queue_stats
+
+
+def test_rss_key_indir(cfg):
+    """
+    Test basics like updating the main RSS key and indirection table.
+    """
+    if len(_get_rx_cnts(cfg)) < 2:
+        KsftSkipEx("Device has only one queue (or doesn't support queue stats)")
+
+    data = get_rss(cfg)
+    want_keys = ['rss-hash-key', 'rss-hash-function', 'rss-indirection-table']
+    for k in want_keys:
+        if k not in data:
+            raise KsftFailEx("ethtool results missing key: " + k)
+        if not data[k]:
+            raise KsftFailEx(f"ethtool results empty for '{k}': {data[k]}")
+
+    key_len = len(data['rss-hash-key'])
+
+    # Set the key
+    key = _rss_key_rand(key_len)
+    ethtool(f"-X {cfg.ifname} hkey " + _rss_key_str(key))
+
+    data = get_rss(cfg)
+    ksft_eq(key, data['rss-hash-key'])
+
+    # Set the indirection table
+    ethtool(f"-X {cfg.ifname} equal 2")
+    data = get_rss(cfg)
+    ksft_eq(0, min(data['rss-indirection-table']))
+    ksft_eq(1, max(data['rss-indirection-table']))
+
+    # Check we only get traffic on the first 2 queues
+    cnts = _get_rx_cnts(cfg)
+    GenerateTraffic(cfg).wait_pkts_and_stop(20000)
+    cnts = _get_rx_cnts(cfg, prev=cnts)
+    # 2 queues, 20k packets, must be at least 5k per queue
+    ksft_ge(cnts[0], 5000, "traffic on main context (1/2): " + str(cnts))
+    ksft_ge(cnts[1], 5000, "traffic on main context (2/2): " + str(cnts))
+    # The other queues should be unused
+    ksft_eq(sum(cnts[2:]), 0, "traffic on unused queues: " + str(cnts))
+
+    # Restore, and check traffic gets spread again
+    ethtool(f"-X {cfg.ifname} default")
+
+    cnts = _get_rx_cnts(cfg)
+    GenerateTraffic(cfg).wait_pkts_and_stop(20000)
+    cnts = _get_rx_cnts(cfg, prev=cnts)
+    # First two queues get less traffic than all the rest
+    ksft_lt(sum(cnts[:2]), sum(cnts[2:]), "traffic distributed: " + str(cnts))
+
+
+def test_rss_context(cfg, ctx_cnt=1, create_with_cfg=None):
+    """
+    Test separating traffic into RSS contexts.
+    The queues will be allocated 2 for each context:
+     ctx0  ctx1  ctx2  ctx3
+    [0 1] [2 3] [4 5] [6 7] ...
+    """
+
+    require_ntuple(cfg)
+
+    requested_ctx_cnt = ctx_cnt
+
+    # Try to allocate more queues when necessary
+    qcnt = len(_get_rx_cnts(cfg))
+    if qcnt >= 2 + 2 * ctx_cnt:
+        qcnt = None
+    else:
+        try:
+            ksft_pr(f"Increasing queue count {qcnt} -> {2 + 2 * ctx_cnt}")
+            ethtool(f"-L {cfg.ifname} combined {2 + 2 * ctx_cnt}")
+        except:
+            raise KsftSkipEx("Not enough queues for the test")
+
+    ntuple = []
+    ctx_id = []
+    ports = []
+    try:
+        # Use queues 0 and 1 for normal traffic
+        ethtool(f"-X {cfg.ifname} equal 2")
+
+        for i in range(ctx_cnt):
+            want_cfg = f"start {2 + i * 2} equal 2"
+            create_cfg = want_cfg if create_with_cfg else ""
+
+            try:
+                ctx_id.append(ethtool_create(cfg, "-X", f"context new {create_cfg}"))
+            except CmdExitFailure:
+                # try to carry on and skip at the end
+                if i == 0:
+                    raise
+                ksft_pr(f"Failed to create context {i + 1}, trying to test what we got")
+                ctx_cnt = i
+                break
+
+            if not create_with_cfg:
+                ethtool(f"-X {cfg.ifname} context {ctx_id[i]} {want_cfg}")
+
+            # Sanity check the context we just created
+            data = get_rss(cfg, ctx_id[i])
+            ksft_eq(min(data['rss-indirection-table']), 2 + i * 2, "Unexpected context cfg: " + str(data))
+            ksft_eq(max(data['rss-indirection-table']), 2 + i * 2 + 1, "Unexpected context cfg: " + str(data))
+
+            ports.append(rand_port())
+            flow = f"flow-type tcp{cfg.addr_ipver} dst-port {ports[i]} context {ctx_id[i]}"
+            ntuple.append(ethtool_create(cfg, "-N", flow))
+
+        for i in range(ctx_cnt):
+            cnts = _get_rx_cnts(cfg)
+            GenerateTraffic(cfg, port=ports[i]).wait_pkts_and_stop(20000)
+            cnts = _get_rx_cnts(cfg, prev=cnts)
+
+            ksft_lt(sum(cnts[ :2]), 10000, "traffic on main context:" + str(cnts))
+            ksft_ge(sum(cnts[2+i*2:4+i*2]), 20000, f"traffic on context {i}: " + str(cnts))
+            ksft_eq(sum(cnts[2:2+i*2] + cnts[4+i*2:]), 0, "traffic on other contexts: " + str(cnts))
+    finally:
+        for nid in ntuple:
+            ethtool(f"-N {cfg.ifname} delete {nid}")
+        for cid in ctx_id:
+            ethtool(f"-X {cfg.ifname} context {cid} delete")
+        ethtool(f"-X {cfg.ifname} default")
+        if qcnt:
+            ethtool(f"-L {cfg.ifname} combined {qcnt}")
+
+    if requested_ctx_cnt != ctx_cnt:
+        raise KsftSkipEx(f"Tested only {ctx_cnt} contexts, wanted {requested_ctx_cnt}")
+
+
+def test_rss_context4(cfg):
+    test_rss_context(cfg, 4)
+
+
+def test_rss_context32(cfg):
+    test_rss_context(cfg, 32)
+
+
+def test_rss_context4_create_with_cfg(cfg):
+    test_rss_context(cfg, 4, create_with_cfg=True)
+
+
+def test_rss_context_out_of_order(cfg, ctx_cnt=4):
+    """
+    Test separating traffic into RSS contexts.
+    Contexts are removed in semi-random order, and steering re-tested
+    to make sure removal doesn't break steering to surviving contexts.
+    Test requires 3 contexts to work.
+    """
+
+    require_ntuple(cfg)
+
+    requested_ctx_cnt = ctx_cnt
+
+    # Try to allocate more queues when necessary
+    qcnt = len(_get_rx_cnts(cfg))
+    if qcnt >= 2 + 2 * ctx_cnt:
+        qcnt = None
+    else:
+        try:
+            ksft_pr(f"Increasing queue count {qcnt} -> {2 + 2 * ctx_cnt}")
+            ethtool(f"-L {cfg.ifname} combined {2 + 2 * ctx_cnt}")
+        except:
+            raise KsftSkipEx("Not enough queues for the test")
+
+    ntuple = []
+    ctx_id = []
+    ports = []
+
+    def remove_ctx(idx):
+        ethtool(f"-N {cfg.ifname} delete {ntuple[idx]}")
+        ntuple[idx] = None
+        ethtool(f"-X {cfg.ifname} context {ctx_id[idx]} delete")
+        ctx_id[idx] = None
+
+    def check_traffic():
+        for i in range(ctx_cnt):
+            cnts = _get_rx_cnts(cfg)
+            GenerateTraffic(cfg, port=ports[i]).wait_pkts_and_stop(20000)
+            cnts = _get_rx_cnts(cfg, prev=cnts)
+
+            if ctx_id[i] is None:
+                ksft_lt(sum(cnts[ :2]), 10000, "traffic on main context:" + str(cnts))
+                ksft_ge(sum(cnts[2+i*2:4+i*2]), 20000, f"traffic on context {i}: " + str(cnts))
+                ksft_eq(sum(cnts[2:2+i*2] + cnts[4+i*2:]), 0, "traffic on other contexts: " + str(cnts))
+            else:
+                ksft_ge(sum(cnts[ :2]), 20000, "traffic on main context:" + str(cnts))
+                ksft_eq(sum(cnts[2: ]),     0, "traffic on other contexts: " + str(cnts))
+
+    try:
+        # Use queues 0 and 1 for normal traffic
+        ethtool(f"-X {cfg.ifname} equal 2")
+
+        for i in range(ctx_cnt):
+            ctx_id.append(ethtool_create(cfg, "-X", f"context new start {2 + i * 2} equal 2"))
+
+            ports.append(rand_port())
+            flow = f"flow-type tcp{cfg.addr_ipver} dst-port {ports[i]} context {ctx_id[i]}"
+            ntuple.append(ethtool_create(cfg, "-N", flow))
+
+        check_traffic()
+
+        # Remove middle context
+        remove_ctx(ctx_cnt // 2)
+        check_traffic()
+
+        # Remove first context
+        remove_ctx(0)
+        check_traffic()
+
+        # Remove last context
+        remove_ctx(-1)
+        check_traffic()
+
+    finally:
+        for nid in ntuple:
+            if nid is not None:
+                ethtool(f"-N {cfg.ifname} delete {nid}")
+        for cid in ctx_id:
+            if cid is not None:
+                ethtool(f"-X {cfg.ifname} context {cid} delete")
+        ethtool(f"-X {cfg.ifname} default")
+        if qcnt:
+            ethtool(f"-L {cfg.ifname} combined {qcnt}")
+
+    if requested_ctx_cnt != ctx_cnt:
+        raise KsftSkipEx(f"Tested only {ctx_cnt} contexts, wanted {requested_ctx_cnt}")
+
+
+def test_rss_context_overlap(cfg, other_ctx=0):
+    """
+    Test contexts overlapping with each other.
+    Use 4 queues for the main context, but only queues 2 and 3 for context 1.
+    """
+
+    require_ntuple(cfg)
+
+    queue_cnt = len(_get_rx_cnts(cfg))
+    if queue_cnt >= 4:
+        queue_cnt = None
+    else:
+        try:
+            ksft_pr(f"Increasing queue count {queue_cnt} -> 4")
+            ethtool(f"-L {cfg.ifname} combined 4")
+        except:
+            raise KsftSkipEx("Not enough queues for the test")
+
+    ctx_id = None
+    ntuple = None
+    if other_ctx == 0:
+        ethtool(f"-X {cfg.ifname} equal 4")
+    else:
+        other_ctx = ethtool_create(cfg, "-X", "context new")
+        ethtool(f"-X {cfg.ifname} context {other_ctx} equal 4")
+
+    try:
+        ctx_id = ethtool_create(cfg, "-X", "context new")
+        ethtool(f"-X {cfg.ifname} context {ctx_id} start 2 equal 2")
+
+        port = rand_port()
+        if other_ctx:
+            flow = f"flow-type tcp{cfg.addr_ipver} dst-port {port} context {other_ctx}"
+            ntuple = ethtool_create(cfg, "-N", flow)
+
+        # Test the main context
+        cnts = _get_rx_cnts(cfg)
+        GenerateTraffic(cfg, port=port).wait_pkts_and_stop(20000)
+        cnts = _get_rx_cnts(cfg, prev=cnts)
+
+        ksft_ge(sum(cnts[ :4]), 20000, "traffic on main context: " + str(cnts))
+        ksft_ge(sum(cnts[ :2]),  7000, "traffic on main context (1/2): " + str(cnts))
+        ksft_ge(sum(cnts[2:4]),  7000, "traffic on main context (2/2): " + str(cnts))
+        if other_ctx == 0:
+            ksft_eq(sum(cnts[4: ]),     0, "traffic on other queues: " + str(cnts))
+
+        # Now create a rule for context 1 and make sure traffic goes to a subset
+        if other_ctx:
+            ethtool(f"-N {cfg.ifname} delete {ntuple}")
+            ntuple = None
+        flow = f"flow-type tcp{cfg.addr_ipver} dst-port {port} context {ctx_id}"
+        ntuple = ethtool_create(cfg, "-N", flow)
+
+        cnts = _get_rx_cnts(cfg)
+        GenerateTraffic(cfg, port=port).wait_pkts_and_stop(20000)
+        cnts = _get_rx_cnts(cfg, prev=cnts)
+
+        ksft_lt(sum(cnts[ :2]),  7000, "traffic on main context: " + str(cnts))
+        ksft_ge(sum(cnts[2:4]), 20000, "traffic on extra context: " + str(cnts))
+        if other_ctx == 0:
+            ksft_eq(sum(cnts[4: ]),     0, "traffic on other queues: " + str(cnts))
+    finally:
+        if ntuple is not None:
+            ethtool(f"-N {cfg.ifname} delete {ntuple}")
+        if ctx_id:
+            ethtool(f"-X {cfg.ifname} context {ctx_id} delete")
+        if other_ctx == 0:
+            ethtool(f"-X {cfg.ifname} default")
+        else:
+            ethtool(f"-X {cfg.ifname} context {other_ctx} delete")
+        if queue_cnt:
+            ethtool(f"-L {cfg.ifname} combined {queue_cnt}")
+
+
+def test_rss_context_overlap2(cfg):
+    test_rss_context_overlap(cfg, True)
+
+
+def main() -> None:
+    with NetDrvEpEnv(__file__, nsim_test=False) as cfg:
+        cfg.netdevnl = NetdevFamily()
+
+        ksft_run([test_rss_key_indir,
+                  test_rss_context, test_rss_context4, test_rss_context32,
+                  test_rss_context_overlap, test_rss_context_overlap2,
+                  test_rss_context_out_of_order, test_rss_context4_create_with_cfg],
+                 args=(cfg, ))
+    ksft_exit()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/tools/testing/selftests/drivers/net/lib/py/load.py b/tools/testing/selftests/drivers/net/lib/py/load.py
index f6bc57eb304a..d9c10613ae67 100644
--- a/tools/testing/selftests/drivers/net/lib/py/load.py
+++ b/tools/testing/selftests/drivers/net/lib/py/load.py
@@ -5,13 +5,14 @@ import time
 from lib.py import ksft_pr, cmd, ip, rand_port, wait_port_listen
 
 class GenerateTraffic:
-    def __init__(self, env):
+    def __init__(self, env, port=None):
         env.require_cmd("iperf3", remote=True)
 
         self.env = env
 
-        port = rand_port()
-        self._iperf_server = cmd(f"iperf3 -s -p {port}", background=True)
+        if port is None:
+            port = rand_port()
+        self._iperf_server = cmd(f"iperf3 -s -1 -p {port}", background=True)
         wait_port_listen(port)
         time.sleep(0.1)
         self._iperf_client = cmd(f"iperf3 -c {env.addr} -P 16 -p {port} -t 86400",
diff --git a/tools/testing/selftests/net/lib/py/ksft.py b/tools/testing/selftests/net/lib/py/ksft.py
index 4769b4eb1ea1..b6ce3f33d41e 100644
--- a/tools/testing/selftests/net/lib/py/ksft.py
+++ b/tools/testing/selftests/net/lib/py/ksft.py
@@ -57,6 +57,11 @@ def ksft_ge(a, b, comment=""):
         _fail("Check failed", a, "<", b, comment)
 
 
+def ksft_lt(a, b, comment=""):
+    if a >= b:
+        _fail("Check failed", a, ">=", b, comment)
+
+
 class ksft_raises:
     def __init__(self, expected_type):
         self.exception = None
diff --git a/tools/testing/selftests/net/lib/py/utils.py b/tools/testing/selftests/net/lib/py/utils.py
index a6c87bfe532f..405aa510aaf2 100644
--- a/tools/testing/selftests/net/lib/py/utils.py
+++ b/tools/testing/selftests/net/lib/py/utils.py
@@ -9,6 +9,10 @@ import subprocess
 import time
 
 
+class CmdExitFailure(Exception):
+    pass
+
+
 class cmd:
     def __init__(self, comm, shell=True, fail=True, ns=None, background=False, host=None, timeout=5):
         if ns:
@@ -43,8 +47,8 @@ class cmd:
         if self.proc.returncode != 0 and fail:
             if len(stderr) > 0 and stderr[-1] == "\n":
                 stderr = stderr[:-1]
-            raise Exception("Command failed: %s\nSTDOUT: %s\nSTDERR: %s" %
-                            (self.proc.args, stdout, stderr))
+            raise CmdExitFailure("Command failed: %s\nSTDOUT: %s\nSTDERR: %s" %
+                                 (self.proc.args, stdout, stderr))
 
 
 class bkg(cmd):
-- 
cgit v1.2.3-58-ga151


From 3d94d1ac3792abee651274e3cfbb9a12c9f95988 Mon Sep 17 00:00:00 2001
From: Christophe Roullier <christophe.roullier@foss.st.com>
Date: Mon, 24 Jun 2024 09:10:51 +0200
Subject: dt-bindings: net: add STM32MP25 compatible in documentation for stm32

New STM32 SOC have 2 GMACs instances.
GMAC IP version is SNPS 5.30

Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 Documentation/devicetree/bindings/net/stm32-dwmac.yaml | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/net/stm32-dwmac.yaml b/Documentation/devicetree/bindings/net/stm32-dwmac.yaml
index f6e5e0626a3f..bf23838fe6e8 100644
--- a/Documentation/devicetree/bindings/net/stm32-dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/stm32-dwmac.yaml
@@ -23,12 +23,17 @@ select:
           - st,stm32-dwmac
           - st,stm32mp1-dwmac
           - st,stm32mp13-dwmac
+          - st,stm32mp25-dwmac
   required:
     - compatible
 
 properties:
   compatible:
     oneOf:
+      - items:
+          - enum:
+              - st,stm32mp25-dwmac
+          - const: snps,dwmac-5.20
       - items:
           - enum:
               - st,stm32mp1-dwmac
@@ -121,8 +126,9 @@ allOf:
         compatible:
           contains:
             enum:
-              - st,stm32mp1-dwmac
               - st,stm32-dwmac
+              - st,stm32mp1-dwmac
+              - st,stm32mp25-dwmac
     then:
       properties:
         st,syscon:
-- 
cgit v1.2.3-58-ga151


From 5bcc1afd021989593b40562819e314d299fde849 Mon Sep 17 00:00:00 2001
From: Christophe Roullier <christophe.roullier@foss.st.com>
Date: Mon, 24 Jun 2024 09:10:52 +0200
Subject: net: stmmac: dwmac-stm32: stm32: add management of stm32mp25 for
 stm32

Add Ethernet support for STM32MP25.
STM32MP25 is STM32 SOC with 2 GMACs instances.
GMAC IP version is SNPS 5.3x.
GMAC IP configure with 2 RX and 4 TX queue.
DMA HW capability register supported
RX Checksum Offload Engine supported
TX Checksum insertion supported
Wake-Up On Lan supported
TSO supported

Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 80 ++++++++++++++++++++++-
 1 file changed, 77 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
index b2db0e26c4e4..23cf0a5b047f 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
@@ -53,7 +53,18 @@
 #define SYSCFG_MCU_ETH_SEL_MII		0
 #define SYSCFG_MCU_ETH_SEL_RMII		1
 
-/* STM32MP1 register definitions
+/* STM32MP2 register definitions */
+#define SYSCFG_MP2_ETH_MASK		GENMASK(31, 0)
+
+#define SYSCFG_ETHCR_ETH_PTP_CLK_SEL	BIT(2)
+#define SYSCFG_ETHCR_ETH_CLK_SEL	BIT(1)
+#define SYSCFG_ETHCR_ETH_REF_CLK_SEL	BIT(0)
+
+#define SYSCFG_ETHCR_ETH_SEL_MII	0
+#define SYSCFG_ETHCR_ETH_SEL_RGMII	BIT(4)
+#define SYSCFG_ETHCR_ETH_SEL_RMII	BIT(6)
+
+/* STM32MPx register definitions
  *
  * Below table summarizes the clock requirement and clock sources for
  * supported phy interface modes.
@@ -104,7 +115,7 @@ struct stm32_ops {
 	int (*parse_data)(struct stm32_dwmac *dwmac,
 			  struct device *dev);
 	bool clk_rx_enable_in_suspend;
-	bool is_mp13;
+	bool is_mp13, is_mp2;
 	u32 syscfg_clr_off;
 };
 
@@ -277,8 +288,55 @@ static int stm32mp1_configure_pmcr(struct plat_stmmacenet_data *plat_dat)
 				 dwmac->mode_mask, val);
 }
 
+static int stm32mp2_configure_syscfg(struct plat_stmmacenet_data *plat_dat)
+{
+	struct stm32_dwmac *dwmac = plat_dat->bsp_priv;
+	u32 reg = dwmac->mode_reg;
+	int val = 0;
+
+	switch (plat_dat->mac_interface) {
+	case PHY_INTERFACE_MODE_MII:
+		/* ETH_REF_CLK_SEL bit in SYSCFG register is not applicable in MII mode */
+		break;
+	case PHY_INTERFACE_MODE_RMII:
+		val = SYSCFG_ETHCR_ETH_SEL_RMII;
+		if (dwmac->enable_eth_ck) {
+			/* Internal clock ETH_CLK of 50MHz from RCC is used */
+			val |= SYSCFG_ETHCR_ETH_REF_CLK_SEL;
+		}
+		break;
+	case PHY_INTERFACE_MODE_RGMII:
+	case PHY_INTERFACE_MODE_RGMII_ID:
+	case PHY_INTERFACE_MODE_RGMII_RXID:
+	case PHY_INTERFACE_MODE_RGMII_TXID:
+		val = SYSCFG_ETHCR_ETH_SEL_RGMII;
+		fallthrough;
+	case PHY_INTERFACE_MODE_GMII:
+		if (dwmac->enable_eth_ck) {
+			/* Internal clock ETH_CLK of 125MHz from RCC is used */
+			val |= SYSCFG_ETHCR_ETH_CLK_SEL;
+		}
+		break;
+	default:
+		dev_err(dwmac->dev, "Mode %s not supported",
+			phy_modes(plat_dat->mac_interface));
+		/* Do not manage others interfaces */
+		return -EINVAL;
+	}
+
+	dev_dbg(dwmac->dev, "Mode %s", phy_modes(plat_dat->mac_interface));
+
+	/* Select PTP (IEEE1588) clock selection from RCC (ck_ker_ethxptp) */
+	val |= SYSCFG_ETHCR_ETH_PTP_CLK_SEL;
+
+	/* Update ETHCR (set register) */
+	return regmap_update_bits(dwmac->regmap, reg,
+				 SYSCFG_MP2_ETH_MASK, val);
+}
+
 static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
 {
+	struct stm32_dwmac *dwmac = plat_dat->bsp_priv;
 	int ret;
 
 	ret = stm32mp1_select_ethck_external(plat_dat);
@@ -289,7 +347,10 @@ static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat)
 	if (ret)
 		return ret;
 
-	return stm32mp1_configure_pmcr(plat_dat);
+	if (!dwmac->ops->is_mp2)
+		return stm32mp1_configure_pmcr(plat_dat);
+	else
+		return stm32mp2_configure_syscfg(plat_dat);
 }
 
 static int stm32mcu_set_mode(struct plat_stmmacenet_data *plat_dat)
@@ -365,6 +426,9 @@ static int stm32_dwmac_parse_data(struct stm32_dwmac *dwmac,
 		return err;
 	}
 
+	if (dwmac->ops->is_mp2)
+		return 0;
+
 	dwmac->mode_mask = SYSCFG_MP1_ETH_MASK;
 	err = of_property_read_u32_index(np, "st,syscon", 2, &dwmac->mode_mask);
 	if (err) {
@@ -586,10 +650,20 @@ static struct stm32_ops stm32mp13_dwmac_data = {
 	.clk_rx_enable_in_suspend = true
 };
 
+static struct stm32_ops stm32mp25_dwmac_data = {
+	.set_mode = stm32mp1_set_mode,
+	.suspend = stm32mp1_suspend,
+	.resume = stm32mp1_resume,
+	.parse_data = stm32mp1_parse_data,
+	.is_mp2 = true,
+	.clk_rx_enable_in_suspend = true
+};
+
 static const struct of_device_id stm32_dwmac_match[] = {
 	{ .compatible = "st,stm32-dwmac", .data = &stm32mcu_dwmac_data},
 	{ .compatible = "st,stm32mp1-dwmac", .data = &stm32mp1_dwmac_data},
 	{ .compatible = "st,stm32mp13-dwmac", .data = &stm32mp13_dwmac_data},
+	{ .compatible = "st,stm32mp25-dwmac", .data = &stm32mp25_dwmac_data},
 	{ }
 };
 MODULE_DEVICE_TABLE(of, stm32_dwmac_match);
-- 
cgit v1.2.3-58-ga151


From 9783e0407421a4dfc25e3a3001d0e0cff4236806 Mon Sep 17 00:00:00 2001
From: Rameshkumar Sundaram <quic_ramess@quicinc.com>
Date: Wed, 26 Jun 2024 08:12:57 +0530
Subject: wifi: ath12k: fix driver initialization for WoW unsupported devices

Commit 4a3c212eee0e ("wifi: ath12k: add basic WoW functionalities") broke
driver initialization, now mac registration is allowed only for devices that
advertise WMI_TLV_SERVICE_WOW, but QCN9274 doesn't support WoW and hence mac
registration is aborted and driver is de-initialized.

Allow mac registration to proceed without WoW Support for devices
that don't advertise WMI_TLV_SERVICE_WOW.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1

Fixes: 4a3c212eee0e ("wifi: ath12k: add basic WoW functionalities")
Signed-off-by: Rameshkumar Sundaram <quic_ramess@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240626024257.709460-1-quic_ramess@quicinc.com
---
 drivers/net/wireless/ath/ath12k/wow.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/wow.c b/drivers/net/wireless/ath/ath12k/wow.c
index 685e8e98d845..c5cba825a84a 100644
--- a/drivers/net/wireless/ath/ath12k/wow.c
+++ b/drivers/net/wireless/ath/ath12k/wow.c
@@ -1001,8 +1001,8 @@ exit:
 
 int ath12k_wow_init(struct ath12k *ar)
 {
-	if (WARN_ON(!test_bit(WMI_TLV_SERVICE_WOW, ar->wmi->wmi_ab->svc_map)))
-		return -EINVAL;
+	if (!test_bit(WMI_TLV_SERVICE_WOW, ar->wmi->wmi_ab->svc_map))
+		return 0;
 
 	ar->wow.wowlan_support = ath12k_wowlan_support;
 
-- 
cgit v1.2.3-58-ga151


From 2a901623f0056ccdb64dbfd3d98f9072f704f065 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 26 Jun 2024 13:12:34 -0700
Subject: tools: ynl: use display hints for formatting of scalar attrs

Use display hints for formatting scalar attrs. This is specifically
useful for formatting IPv4 addresses carried typically as u32.

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20240626201234.2572964-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/lib/ynl.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/net/ynl/lib/ynl.py b/tools/net/ynl/lib/ynl.py
index 35e666928119..d42c1d605969 100644
--- a/tools/net/ynl/lib/ynl.py
+++ b/tools/net/ynl/lib/ynl.py
@@ -743,6 +743,8 @@ class YnlFamily(SpecFamily):
                 decoded = attr.as_scalar(attr_spec['type'], attr_spec.byte_order)
                 if 'enum' in attr_spec:
                     decoded = self._decode_enum(decoded, attr_spec)
+                elif attr_spec.display_hint:
+                    decoded = self._formatted_string(decoded, attr_spec.display_hint)
             elif attr_spec["type"] == 'indexed-array':
                 decoded = self._decode_array_attr(attr, attr_spec)
             elif attr_spec["type"] == 'bitfield32':
-- 
cgit v1.2.3-58-ga151


From 346a03e5fbdb868c5d36052b82bb8eacd1cb83b7 Mon Sep 17 00:00:00 2001
From: Jeff Johnson <quic_jjohnson@quicinc.com>
Date: Tue, 25 Jun 2024 09:35:41 -0700
Subject: s390/lcs: add missing MODULE_DESCRIPTION() macro

With ARCH=s390, make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/s390/net/lcs.o

Add the missing invocation of the MODULE_DESCRIPTION() macro.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Acked-by: Alexandra Winter <wintera@linux.ibm.com>
Link: https://patch.msgid.link/20240625-md-s390-drivers-s390-net-v2-1-5a8a2b2f2ae3@quicinc.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/s390/net/lcs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/s390/net/lcs.c b/drivers/s390/net/lcs.c
index 25d4e6376591..88db8378325a 100644
--- a/drivers/s390/net/lcs.c
+++ b/drivers/s390/net/lcs.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0+
 /*
- *  Linux for S/390 Lan Channel Station Network Driver
+ *  Linux for S/390 LAN channel station device driver
  *
  *  Copyright IBM Corp. 1999, 2009
  *  Author(s): Original Code written by
@@ -2380,5 +2380,6 @@ module_init(lcs_init_module);
 module_exit(lcs_cleanup_module);
 
 MODULE_AUTHOR("Frank Pavlic <fpavlic@de.ibm.com>");
+MODULE_DESCRIPTION("S/390 LAN channel station device driver");
 MODULE_LICENSE("GPL");
 
-- 
cgit v1.2.3-58-ga151


From f94ecbc920925d5922d68a434f76aa8ef8d7f21e Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Tue, 25 Jun 2024 13:22:39 -0400
Subject: selftests: openvswitch: Support explicit tunnel port creation.

The OVS module can operate in conjunction with various types of
tunnel ports.  These are created as either explicit tunnel vport
types, OR by creating a tunnel interface which acts as an anchor
for the lightweight tunnel support.

This patch adds the ability to add tunnel ports to an OVS
datapath for testing various scenarios with tunnel ports.  With
this addition, the vswitch "plumbing" will at least be able to
push packets around using the tunnel vports.  Future patches
will add support for setting required tunnel metadata for lwts
in the datapath.  The end goal will be to push packets via these
tunnels, and will be used in an upcoming commit for testing the
path MTU.

Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Simon Horman <horms@kernel.org>
Signed-off-by: Aaron Conole <aconole@redhat.com>
Link: https://patch.msgid.link/20240625172245.233874-2-aconole@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../testing/selftests/net/openvswitch/ovs-dpctl.py | 81 ++++++++++++++++++++--
 1 file changed, 75 insertions(+), 6 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 9f8dec2f6539..02e44c6f0503 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -10,6 +10,7 @@ import ipaddress
 import logging
 import multiprocessing
 import re
+import socket
 import struct
 import sys
 import time
@@ -29,6 +30,7 @@ try:
     from pyroute2.netlink.exceptions import NetlinkError
     from pyroute2.netlink.generic import GenericNetlinkSocket
     import pyroute2
+    import pyroute2.iproute
 
 except ModuleNotFoundError:
     print("Need to install the python pyroute2 package >= 0.6.")
@@ -1617,7 +1619,7 @@ class OvsVport(GenericNetlinkSocket):
             ("OVS_VPORT_ATTR_PORT_NO", "uint32"),
             ("OVS_VPORT_ATTR_TYPE", "uint32"),
             ("OVS_VPORT_ATTR_NAME", "asciiz"),
-            ("OVS_VPORT_ATTR_OPTIONS", "none"),
+            ("OVS_VPORT_ATTR_OPTIONS", "vportopts"),
             ("OVS_VPORT_ATTR_UPCALL_PID", "array(uint32)"),
             ("OVS_VPORT_ATTR_STATS", "vportstats"),
             ("OVS_VPORT_ATTR_PAD", "none"),
@@ -1625,6 +1627,13 @@ class OvsVport(GenericNetlinkSocket):
             ("OVS_VPORT_ATTR_NETNSID", "uint32"),
         )
 
+        class vportopts(nla):
+            nla_map = (
+                ("OVS_TUNNEL_ATTR_UNSPEC", "none"),
+                ("OVS_TUNNEL_ATTR_DST_PORT", "uint16"),
+                ("OVS_TUNNEL_ATTR_EXTENSION", "none"),
+            )
+
         class vportstats(nla):
             fields = (
                 ("rx_packets", "=Q"),
@@ -1693,7 +1702,7 @@ class OvsVport(GenericNetlinkSocket):
                 raise ne
         return reply
 
-    def attach(self, dpindex, vport_ifname, ptype):
+    def attach(self, dpindex, vport_ifname, ptype, dport, lwt):
         msg = OvsVport.ovs_vport_msg()
 
         msg["cmd"] = OVS_VPORT_CMD_NEW
@@ -1702,12 +1711,43 @@ class OvsVport(GenericNetlinkSocket):
         msg["dpifindex"] = dpindex
         port_type = OvsVport.str_to_type(ptype)
 
-        msg["attrs"].append(["OVS_VPORT_ATTR_TYPE", port_type])
         msg["attrs"].append(["OVS_VPORT_ATTR_NAME", vport_ifname])
         msg["attrs"].append(
             ["OVS_VPORT_ATTR_UPCALL_PID", [self.upcall_packet.epid]]
         )
 
+        TUNNEL_DEFAULTS = [("geneve", 6081),
+                           ("vxlan", 4789)]
+
+        for tnl in TUNNEL_DEFAULTS:
+            if ptype == tnl[0]:
+                if not dport:
+                    dport = tnl[1]
+
+                if not lwt:
+                    vportopt = OvsVport.ovs_vport_msg.vportopts()
+                    vportopt["attrs"].append(
+                        ["OVS_TUNNEL_ATTR_DST_PORT", socket.htons(dport)]
+                    )
+                    msg["attrs"].append(
+                        ["OVS_VPORT_ATTR_OPTIONS", vportopt]
+                    )
+                else:
+                    port_type = OvsVport.OVS_VPORT_TYPE_NETDEV
+                    ipr = pyroute2.iproute.IPRoute()
+
+                    if tnl[0] == "geneve":
+                        ipr.link("add", ifname=vport_ifname, kind=tnl[0],
+                                 geneve_port=dport,
+                                 geneve_collect_metadata=True,
+                                 geneve_udp_zero_csum6_rx=1)
+                    elif tnl[0] == "vxlan":
+                        ipr.link("add", ifname=vport_ifname, kind=tnl[0],
+                                 vxlan_learning=0, vxlan_collect_metadata=1,
+                                 vxlan_udp_zero_csum6_rx=1, vxlan_port=dport)
+                break
+        msg["attrs"].append(["OVS_VPORT_ATTR_TYPE", port_type])
+
         try:
             reply = self.nlm_request(
                 msg, msg_type=self.prid, msg_flags=NLM_F_REQUEST | NLM_F_ACK
@@ -2053,12 +2093,19 @@ def print_ovsdp_full(dp_lookup_rep, ifindex, ndb=NDB(), vpl=OvsVport()):
     for iface in ndb.interfaces:
         rep = vpl.info(iface.ifname, ifindex)
         if rep is not None:
+            opts = ""
+            vpo = rep.get_attr("OVS_VPORT_ATTR_OPTIONS")
+            if vpo:
+                dpo = vpo.get_attr("OVS_TUNNEL_ATTR_DST_PORT")
+                if dpo:
+                    opts += " tnl-dport:%s" % socket.ntohs(dpo)
             print(
-                "  port %d: %s (%s)"
+                "  port %d: %s (%s%s)"
                 % (
                     rep.get_attr("OVS_VPORT_ATTR_PORT_NO"),
                     rep.get_attr("OVS_VPORT_ATTR_NAME"),
                     OvsVport.type_to_str(rep.get_attr("OVS_VPORT_ATTR_TYPE")),
+                    opts,
                 )
             )
 
@@ -2120,12 +2167,30 @@ def main(argv):
         "--ptype",
         type=str,
         default="netdev",
-        choices=["netdev", "internal"],
+        choices=["netdev", "internal", "geneve", "vxlan"],
         help="Interface type (default netdev)",
     )
+    addifcmd.add_argument(
+        "-p",
+        "--dport",
+        type=int,
+        default=0,
+        help="Destination port (0 for default)"
+    )
+    addifcmd.add_argument(
+        "-l",
+        "--lwt",
+        type=bool,
+        default=True,
+        help="Use LWT infrastructure instead of vport (default true)."
+    )
     delifcmd = subparsers.add_parser("del-if")
     delifcmd.add_argument("dpname", help="Datapath Name")
     delifcmd.add_argument("delif", help="Interface name for adding")
+    delifcmd.add_argument("-d",
+                          "--dellink",
+                          type=bool, default=False,
+                          help="Delete the link as well.")
 
     dumpflcmd = subparsers.add_parser("dump-flows")
     dumpflcmd.add_argument("dumpdp", help="Datapath Name")
@@ -2186,7 +2251,8 @@ def main(argv):
             print("DP '%s' not found." % args.dpname)
             return 1
         dpindex = rep["dpifindex"]
-        rep = ovsvp.attach(rep["dpifindex"], args.addif, args.ptype)
+        rep = ovsvp.attach(rep["dpifindex"], args.addif, args.ptype,
+                           args.dport, args.lwt)
         msg = "vport '%s'" % args.addif
         if rep and rep["header"]["error"] is None:
             msg += " added."
@@ -2207,6 +2273,9 @@ def main(argv):
             msg += " removed."
         else:
             msg += " failed to remove."
+        if args.dellink:
+            ipr = pyroute2.iproute.IPRoute()
+            ipr.link("del", index=ipr.link_lookup(ifname=args.delif)[0])
     elif hasattr(args, "dumpdp"):
         rep = ovsdp.info(args.dumpdp, 0)
         if rep is None:
-- 
cgit v1.2.3-58-ga151


From 37de65a764ed329408a7ce339ee6283a4a5b682c Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Tue, 25 Jun 2024 13:22:40 -0400
Subject: selftests: openvswitch: Refactor actions parsing.

Until recently, the ovs-dpctl utility was used with a limited actions set
and didn't need to have support for multiple similar actions.  However,
when adding support for tunnels, it will be important to support multiple
set() actions in a single flow.  When printing these actions, the existing
code will be unable to print all of the sets - it will only print the
first.

Refactor this code to be easier to read and support multiple actions of the
same type in an action list.

Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Simon Horman <horms@kernel.org>
Signed-off-by: Aaron Conole <aconole@redhat.com>
Link: https://patch.msgid.link/20240625172245.233874-3-aconole@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../testing/selftests/net/openvswitch/ovs-dpctl.py | 45 +++++++++++-----------
 1 file changed, 23 insertions(+), 22 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 02e44c6f0503..6bbe15daca74 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -439,32 +439,30 @@ class ovsactions(nla):
     def dpstr(self, more=False):
         print_str = ""
 
-        for field in self.nla_map:
+        for field in self["attrs"]:
             if field[1] == "none" or self.get_attr(field[0]) is None:
                 continue
             if print_str != "":
                 print_str += ","
 
-            if field[1] == "uint32":
-                if field[0] == "OVS_ACTION_ATTR_OUTPUT":
-                    print_str += "%d" % int(self.get_attr(field[0]))
-                elif field[0] == "OVS_ACTION_ATTR_RECIRC":
-                    print_str += "recirc(0x%x)" % int(self.get_attr(field[0]))
-                elif field[0] == "OVS_ACTION_ATTR_TRUNC":
-                    print_str += "trunc(%d)" % int(self.get_attr(field[0]))
-                elif field[0] == "OVS_ACTION_ATTR_DROP":
-                    print_str += "drop(%d)" % int(self.get_attr(field[0]))
-            elif field[1] == "flag":
-                if field[0] == "OVS_ACTION_ATTR_CT_CLEAR":
-                    print_str += "ct_clear"
-                elif field[0] == "OVS_ACTION_ATTR_POP_VLAN":
-                    print_str += "pop_vlan"
-                elif field[0] == "OVS_ACTION_ATTR_POP_ETH":
-                    print_str += "pop_eth"
-                elif field[0] == "OVS_ACTION_ATTR_POP_NSH":
-                    print_str += "pop_nsh"
-                elif field[0] == "OVS_ACTION_ATTR_POP_MPLS":
-                    print_str += "pop_mpls"
+            if field[0] == "OVS_ACTION_ATTR_OUTPUT":
+                print_str += "%d" % int(self.get_attr(field[0]))
+            elif field[0] == "OVS_ACTION_ATTR_RECIRC":
+                print_str += "recirc(0x%x)" % int(self.get_attr(field[0]))
+            elif field[0] == "OVS_ACTION_ATTR_TRUNC":
+                print_str += "trunc(%d)" % int(self.get_attr(field[0]))
+            elif field[0] == "OVS_ACTION_ATTR_DROP":
+                print_str += "drop(%d)" % int(self.get_attr(field[0]))
+            elif field[0] == "OVS_ACTION_ATTR_CT_CLEAR":
+                print_str += "ct_clear"
+            elif field[0] == "OVS_ACTION_ATTR_POP_VLAN":
+                print_str += "pop_vlan"
+            elif field[0] == "OVS_ACTION_ATTR_POP_ETH":
+                print_str += "pop_eth"
+            elif field[0] == "OVS_ACTION_ATTR_POP_NSH":
+                print_str += "pop_nsh"
+            elif field[0] == "OVS_ACTION_ATTR_POP_MPLS":
+                print_str += "pop_mpls"
             else:
                 datum = self.get_attr(field[0])
                 if field[0] == "OVS_ACTION_ATTR_CLONE":
@@ -472,7 +470,10 @@ class ovsactions(nla):
                     print_str += datum.dpstr(more)
                     print_str += ")"
                 else:
-                    print_str += datum.dpstr(more)
+                    try:
+                        print_str += datum.dpstr(more)
+                    except:
+                        print_str += "{ATTR: %s not decoded}" % field[0]
 
         return print_str
 
-- 
cgit v1.2.3-58-ga151


From a4126f90a35f53c4bce1f18bde6e477fa4d81f85 Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Tue, 25 Jun 2024 13:22:41 -0400
Subject: selftests: openvswitch: Add set() and set_masked() support.

These will be used in upcoming commits to set specific attributes for
interacting with tunnels.  Since set() will use the key parsing routine, we
also make sure to prepend it with an open paren, for the action parsing to
properly understand it.

Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Simon Horman <horms@kernel.org>
Signed-off-by: Aaron Conole <aconole@redhat.com>
Link: https://patch.msgid.link/20240625172245.233874-4-aconole@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../testing/selftests/net/openvswitch/ovs-dpctl.py | 37 ++++++++++++++++++++--
 1 file changed, 34 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 6bbe15daca74..00a8d6c0163b 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -284,7 +284,7 @@ class ovsactions(nla):
         ("OVS_ACTION_ATTR_UNSPEC", "none"),
         ("OVS_ACTION_ATTR_OUTPUT", "uint32"),
         ("OVS_ACTION_ATTR_USERSPACE", "userspace"),
-        ("OVS_ACTION_ATTR_SET", "none"),
+        ("OVS_ACTION_ATTR_SET", "ovskey"),
         ("OVS_ACTION_ATTR_PUSH_VLAN", "none"),
         ("OVS_ACTION_ATTR_POP_VLAN", "flag"),
         ("OVS_ACTION_ATTR_SAMPLE", "none"),
@@ -292,7 +292,7 @@ class ovsactions(nla):
         ("OVS_ACTION_ATTR_HASH", "none"),
         ("OVS_ACTION_ATTR_PUSH_MPLS", "none"),
         ("OVS_ACTION_ATTR_POP_MPLS", "flag"),
-        ("OVS_ACTION_ATTR_SET_MASKED", "none"),
+        ("OVS_ACTION_ATTR_SET_MASKED", "ovskey"),
         ("OVS_ACTION_ATTR_CT", "ctact"),
         ("OVS_ACTION_ATTR_TRUNC", "uint32"),
         ("OVS_ACTION_ATTR_PUSH_ETH", "none"),
@@ -469,6 +469,18 @@ class ovsactions(nla):
                     print_str += "clone("
                     print_str += datum.dpstr(more)
                     print_str += ")"
+                elif field[0] == "OVS_ACTION_ATTR_SET" or \
+                     field[0] == "OVS_ACTION_ATTR_SET_MASKED":
+                    print_str += "set"
+                    field = datum
+                    mask = None
+                    if field[0] == "OVS_ACTION_ATTR_SET_MASKED":
+                        print_str += "_masked"
+                        field = datum[0]
+                        mask = datum[1]
+                    print_str += "("
+                    print_str += field.dpstr(mask, more)
+                    print_str += ")"
                 else:
                     try:
                         print_str += datum.dpstr(more)
@@ -547,6 +559,25 @@ class ovsactions(nla):
                 self["attrs"].append(("OVS_ACTION_ATTR_CLONE", subacts))
                 actstr = actstr[parsedLen:]
                 parsed = True
+            elif parse_starts_block(actstr, "set(", False):
+                parencount += 1
+                k = ovskey()
+                actstr = actstr[len("set("):]
+                actstr = k.parse(actstr, None)
+                self["attrs"].append(("OVS_ACTION_ATTR_SET", k))
+                if not actstr.startswith(")"):
+                    actstr = ")" + actstr
+                parsed = True
+            elif parse_starts_block(actstr, "set_masked(", False):
+                parencount += 1
+                k = ovskey()
+                m = ovskey()
+                actstr = actstr[len("set_masked("):]
+                actstr = k.parse(actstr, m)
+                self["attrs"].append(("OVS_ACTION_ATTR_SET_MASKED", [k, m]))
+                if not actstr.startswith(")"):
+                    actstr = ")" + actstr
+                parsed = True
             elif parse_starts_block(actstr, "ct(", False):
                 parencount += 1
                 actstr = actstr[len("ct(") :]
@@ -1312,7 +1343,7 @@ class ovskey(nla):
                 mask["attrs"].append([field[0], m])
             self["attrs"].append([field[0], k])
 
-            flowstr = flowstr[strspn(flowstr, "),") :]
+            flowstr = flowstr[strspn(flowstr, "), ") :]
 
         return flowstr
 
-- 
cgit v1.2.3-58-ga151


From fefe3b7d6bec12a4018ea98469f569087d1d7980 Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Tue, 25 Jun 2024 13:22:42 -0400
Subject: selftests: openvswitch: Add support for tunnel() key.

This will be used when setting details about the tunnel to use as
transport.  There is a difference between the ODP format between tunnel():
the 'key' flag is not actually a flag field, so we don't support it in the
same way that the vswitchd userspace supports displaying it.

Signed-off-by: Aaron Conole <aconole@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240625172245.233874-5-aconole@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../testing/selftests/net/openvswitch/ovs-dpctl.py | 167 ++++++++++++++++++++-
 1 file changed, 166 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 00a8d6c0163b..db8163103406 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -709,7 +709,7 @@ class ovskey(nla):
         ("OVS_KEY_ATTR_ARP", "ovs_key_arp"),
         ("OVS_KEY_ATTR_ND", "ovs_key_nd"),
         ("OVS_KEY_ATTR_SKB_MARK", "uint32"),
-        ("OVS_KEY_ATTR_TUNNEL", "none"),
+        ("OVS_KEY_ATTR_TUNNEL", "ovs_key_tunnel"),
         ("OVS_KEY_ATTR_SCTP", "ovs_key_sctp"),
         ("OVS_KEY_ATTR_TCP_FLAGS", "be16"),
         ("OVS_KEY_ATTR_DP_HASH", "uint32"),
@@ -1269,6 +1269,163 @@ class ovskey(nla):
                 init=init,
             )
 
+    class ovs_key_tunnel(nla):
+        nla_flags = NLA_F_NESTED
+
+        nla_map = (
+            ("OVS_TUNNEL_KEY_ATTR_ID", "be64"),
+            ("OVS_TUNNEL_KEY_ATTR_IPV4_SRC", "ipaddr"),
+            ("OVS_TUNNEL_KEY_ATTR_IPV4_DST", "ipaddr"),
+            ("OVS_TUNNEL_KEY_ATTR_TOS", "uint8"),
+            ("OVS_TUNNEL_KEY_ATTR_TTL", "uint8"),
+            ("OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT", "flag"),
+            ("OVS_TUNNEL_KEY_ATTR_CSUM", "flag"),
+            ("OVS_TUNNEL_KEY_ATTR_OAM", "flag"),
+            ("OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS", "array(uint32)"),
+            ("OVS_TUNNEL_KEY_ATTR_TP_SRC", "be16"),
+            ("OVS_TUNNEL_KEY_ATTR_TP_DST", "be16"),
+            ("OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS", "none"),
+            ("OVS_TUNNEL_KEY_ATTR_IPV6_SRC", "ipaddr"),
+            ("OVS_TUNNEL_KEY_ATTR_IPV6_DST", "ipaddr"),
+            ("OVS_TUNNEL_KEY_ATTR_PAD", "none"),
+            ("OVS_TUNNEL_KEY_ATTR_ERSPAN_OPTS", "none"),
+            ("OVS_TUNNEL_KEY_ATTR_IPV4_INFO_BRIDGE", "flag"),
+        )
+
+        def parse(self, flowstr, mask=None):
+            if not flowstr.startswith("tunnel("):
+                return None, None
+
+            k = ovskey.ovs_key_tunnel()
+            if mask is not None:
+                mask = ovskey.ovs_key_tunnel()
+
+            flowstr = flowstr[len("tunnel("):]
+
+            v6_address = None
+
+            fields = [
+                ("tun_id=", r"(\d+)", int, "OVS_TUNNEL_KEY_ATTR_ID",
+                 0xffffffffffffffff, None, None),
+
+                ("src=", r"([0-9a-fA-F\.]+)", str,
+                 "OVS_TUNNEL_KEY_ATTR_IPV4_SRC", "255.255.255.255", "0.0.0.0",
+                 False),
+                ("dst=", r"([0-9a-fA-F\.]+)", str,
+                 "OVS_TUNNEL_KEY_ATTR_IPV4_DST", "255.255.255.255", "0.0.0.0",
+                 False),
+
+                ("ipv6_src=", r"([0-9a-fA-F:]+)", str,
+                 "OVS_TUNNEL_KEY_ATTR_IPV6_SRC",
+                 "ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff", "::", True),
+                ("ipv6_dst=", r"([0-9a-fA-F:]+)", str,
+                 "OVS_TUNNEL_KEY_ATTR_IPV6_DST",
+                 "ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff", "::", True),
+
+                ("tos=", r"(\d+)", int, "OVS_TUNNEL_KEY_ATTR_TOS", 255, 0,
+                 None),
+                ("ttl=", r"(\d+)", int, "OVS_TUNNEL_KEY_ATTR_TTL", 255, 0,
+                 None),
+
+                ("tp_src=", r"(\d+)", int, "OVS_TUNNEL_KEY_ATTR_TP_SRC",
+                 65535, 0, None),
+                ("tp_dst=", r"(\d+)", int, "OVS_TUNNEL_KEY_ATTR_TP_DST",
+                 65535, 0, None),
+            ]
+
+            forced_include = ["OVS_TUNNEL_KEY_ATTR_TTL"]
+
+            for prefix, regex, typ, attr_name, mask_val, default_val, v46_flag in fields:
+                flowstr, value = parse_extract_field(flowstr, prefix, regex, typ, False)
+                if not attr_name:
+                    raise Exception("Bad list value in tunnel fields")
+
+                if value is None and attr_name in forced_include:
+                    value = default_val
+                    mask_val = default_val
+
+                if value is not None:
+                    if v46_flag is not None:
+                        if v6_address is None:
+                            v6_address = v46_flag
+                        if v46_flag != v6_address:
+                            raise ValueError("Cannot mix v6 and v4 addresses")
+                    k["attrs"].append([attr_name, value])
+                    if mask is not None:
+                        mask["attrs"].append([attr_name, mask_val])
+                else:
+                    if v46_flag is not None:
+                        if v6_address is None or v46_flag != v6_address:
+                            continue
+                    if mask is not None:
+                        mask["attrs"].append([attr_name, default_val])
+
+            if k["attrs"][0][0] != "OVS_TUNNEL_KEY_ATTR_ID":
+                raise ValueError("Needs a tunid set")
+
+            if flowstr.startswith("flags("):
+                flowstr = flowstr[len("flags("):]
+                flagspos = flowstr.find(")")
+                flags = flowstr[:flagspos]
+                flowstr = flowstr[flagspos + 1:]
+
+                flag_attrs = {
+                    "df": "OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT",
+                    "csum": "OVS_TUNNEL_KEY_ATTR_CSUM",
+                    "oam": "OVS_TUNNEL_KEY_ATTR_OAM"
+                }
+
+                for flag in flags.split("|"):
+                    if flag in flag_attrs:
+                        k["attrs"].append([flag_attrs[flag], True])
+                        if mask is not None:
+                            mask["attrs"].append([flag_attrs[flag], True])
+
+            flowstr = flowstr[strspn(flowstr, ", ") :]
+            return flowstr, k, mask
+
+        def dpstr(self, mask=None, more=False):
+            print_str = "tunnel("
+
+            flagsattrs = []
+            for k in self["attrs"]:
+                noprint = False
+                if k[0] == "OVS_TUNNEL_KEY_ATTR_ID":
+                    print_str += "tun_id=%d" % k[1]
+                elif k[0] == "OVS_TUNNEL_KEY_ATTR_IPV4_SRC":
+                    print_str += "src=%s" % k[1]
+                elif k[0] == "OVS_TUNNEL_KEY_ATTR_IPV4_DST":
+                    print_str += "dst=%s" % k[1]
+                elif k[0] == "OVS_TUNNEL_KEY_ATTR_IPV6_SRC":
+                    print_str += "ipv6_src=%s" % k[1]
+                elif k[0] == "OVS_TUNNEL_KEY_ATTR_IPV6_DST":
+                    print_str += "ipv6_dst=%s" % k[1]
+                elif k[0] == "OVS_TUNNEL_KEY_ATTR_TOS":
+                    print_str += "tos=%d" % k[1]
+                elif k[0] == "OVS_TUNNEL_KEY_ATTR_TTL":
+                    print_str += "ttl=%d" % k[1]
+                elif k[0] == "OVS_TUNNEL_KEY_ATTR_TP_SRC":
+                    print_str += "tp_src=%d" % k[1]
+                elif k[0] == "OVS_TUNNEL_KEY_ATTR_TP_DST":
+                    print_str += "tp_dst=%d" % k[1]
+                elif k[0] == "OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT":
+                    noprint = True
+                    flagsattrs.append("df")
+                elif k[0] == "OVS_TUNNEL_KEY_ATTR_CSUM":
+                    noprint = True
+                    flagsattrs.append("csum")
+                elif k[0] == "OVS_TUNNEL_KEY_ATTR_OAM":
+                    noprint = True
+                    flagsattrs.append("oam")
+
+                if not noprint:
+                    print_str += ","
+
+            if len(flagsattrs):
+                print_str += "flags(" + "|".join(flagsattrs) + ")"
+            print_str += ")"
+            return print_str
+
     class ovs_key_mpls(nla):
         fields = (("lse", ">I"),)
 
@@ -1277,6 +1434,7 @@ class ovskey(nla):
             ("OVS_KEY_ATTR_PRIORITY", "skb_priority", intparse),
             ("OVS_KEY_ATTR_SKB_MARK", "skb_mark", intparse),
             ("OVS_KEY_ATTR_RECIRC_ID", "recirc_id", intparse),
+            ("OVS_KEY_ATTR_TUNNEL", "tunnel", ovskey.ovs_key_tunnel),
             ("OVS_KEY_ATTR_DP_HASH", "dp_hash", intparse),
             ("OVS_KEY_ATTR_CT_STATE", "ct_state", parse_ct_state),
             ("OVS_KEY_ATTR_CT_ZONE", "ct_zone", intparse),
@@ -1379,6 +1537,13 @@ class ovskey(nla):
                 lambda x: False,
                 True,
             ),
+            (
+                "OVS_KEY_ATTR_TUNNEL",
+                "tunnel",
+                None,
+                False,
+                False,
+            ),
             (
                 "OVS_KEY_ATTR_CT_STATE",
                 "ct_state",
-- 
cgit v1.2.3-58-ga151


From 51458e1084d0f3938a5b96e5d85974d9c5ee1f21 Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Tue, 25 Jun 2024 13:22:43 -0400
Subject: selftests: openvswitch: Support implicit ipv6 arguments.

The current iteration of IPv6 support requires explicit fields to be set
in addition to not properly support the actual IPv6 addresses properly.
With this change, make it so that the ipv6() bare option is usable to
create wildcarded flows to match broad swaths of ipv6 traffic.

Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Simon Horman <horms@kernel.org>
Signed-off-by: Aaron Conole <aconole@redhat.com>
Link: https://patch.msgid.link/20240625172245.233874-6-aconole@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../testing/selftests/net/openvswitch/ovs-dpctl.py | 36 ++++++++++++++--------
 1 file changed, 24 insertions(+), 12 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index db8163103406..cb3b910bbc4a 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -200,6 +200,18 @@ def convert_ipv4(data):
 
     return int(ipaddress.IPv4Address(ip)), int(ipaddress.IPv4Address(mask))
 
+def convert_ipv6(data):
+    ip, _, mask = data.partition('/')
+
+    if not ip:
+        ip = mask = 0
+    elif not mask:
+        mask = 'ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff'
+    elif mask.isdigit():
+        mask = ipaddress.IPv6Network("::/" + mask).hostmask
+
+    return ipaddress.IPv6Address(ip).packed, ipaddress.IPv6Address(mask).packed
+
 def convert_int(size):
     def convert_int_sized(data):
         value, _, mask = data.partition('/')
@@ -941,21 +953,21 @@ class ovskey(nla):
                 "src",
                 "src",
                 lambda x: str(ipaddress.IPv6Address(x)),
-                lambda x: int.from_bytes(x, "big"),
-                lambda x: ipaddress.IPv6Address(x),
+                lambda x: ipaddress.IPv6Address(x).packed if x else 0,
+                convert_ipv6,
             ),
             (
                 "dst",
                 "dst",
                 lambda x: str(ipaddress.IPv6Address(x)),
-                lambda x: int.from_bytes(x, "big"),
-                lambda x: ipaddress.IPv6Address(x),
+                lambda x: ipaddress.IPv6Address(x).packed if x else 0,
+                convert_ipv6,
             ),
-            ("label", "label", "%d", int),
-            ("proto", "proto", "%d", int),
-            ("tclass", "tclass", "%d", int),
-            ("hlimit", "hlimit", "%d", int),
-            ("frag", "frag", "%d", int),
+            ("label", "label", "%d", lambda x: int(x) if x else 0),
+            ("proto", "proto", "%d", lambda x: int(x) if x else 0),
+            ("tclass", "tclass", "%d", lambda x: int(x) if x else 0),
+            ("hlimit", "hlimit", "%d", lambda x: int(x) if x else 0),
+            ("frag", "frag", "%d", lambda x: int(x) if x else 0),
         )
 
         def __init__(
@@ -1153,7 +1165,7 @@ class ovskey(nla):
                 "target",
                 "target",
                 lambda x: str(ipaddress.IPv6Address(x)),
-                lambda x: int.from_bytes(x, "big"),
+                convert_ipv6,
             ),
             ("sll", "sll", macstr, lambda x: int.from_bytes(x, "big")),
             ("tll", "tll", macstr, lambda x: int.from_bytes(x, "big")),
@@ -1238,13 +1250,13 @@ class ovskey(nla):
                 "src",
                 "src",
                 lambda x: str(ipaddress.IPv6Address(x)),
-                lambda x: int.from_bytes(x, "big", convertmac),
+                convert_ipv6,
             ),
             (
                 "dst",
                 "dst",
                 lambda x: str(ipaddress.IPv6Address(x)),
-                lambda x: int.from_bytes(x, "big"),
+                convert_ipv6,
             ),
             ("tp_src", "tp_src", "%d", int),
             ("tp_dst", "tp_dst", "%d", int),
-- 
cgit v1.2.3-58-ga151


From b7ce46fc614d4d31d438878ade1e4d0c3ef890c7 Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Tue, 25 Jun 2024 13:22:44 -0400
Subject: selftests: net: Use the provided dpctl rather than the vswitchd for
 tests.

The current pmtu test infrastucture requires an installed copy of the
ovs-vswitchd userspace.  This means that any automated or constrained
environments may not have the requisite tools to run the tests.  However,
the pmtu tests don't require any special classifier processing.  Indeed
they are only using the vswitchd in the most basic mode - as a NORMAL
switch.

However, the ovs-dpctl kernel utility can now program all the needed basic
flows to allow traffic to traverse the tunnels and provide support for at
least testing some basic pmtu scenarios.  More complicated flow pipelines
can be added to the internal ovs test infrastructure, but that is work for
the future.  For now, enable the most common cases - wide mega flows with
no other prerequisites.

Enhance the pmtu testing to try testing using the internal utility, first.
As a fallback, if the internal utility isn't running, then try with the
ovs-vswitchd userspace tools.

Additionally, make sure that when the pyroute2 package is not available
the ovs-dpctl utility will error out to properly signal an error has
occurred and skip using the internal utility.

Reviewed-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: Aaron Conole <aconole@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240625172245.233874-7-aconole@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../testing/selftests/net/openvswitch/ovs-dpctl.py |   2 +-
 tools/testing/selftests/net/pmtu.sh                | 145 +++++++++++++++++----
 2 files changed, 124 insertions(+), 23 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index cb3b910bbc4a..182a09975975 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -34,7 +34,7 @@ try:
 
 except ModuleNotFoundError:
     print("Need to install the python pyroute2 package >= 0.6.")
-    sys.exit(0)
+    sys.exit(1)
 
 
 OVS_DATAPATH_FAMILY = "ovs_datapath"
diff --git a/tools/testing/selftests/net/pmtu.sh b/tools/testing/selftests/net/pmtu.sh
index cfc84958025a..5175c0c83a23 100755
--- a/tools/testing/selftests/net/pmtu.sh
+++ b/tools/testing/selftests/net/pmtu.sh
@@ -842,25 +842,97 @@ setup_bridge() {
 	run_cmd ${ns_a} ip link set veth_A-C master br0
 }
 
+setup_ovs_via_internal_utility() {
+	type="${1}"
+	a_addr="${2}"
+	b_addr="${3}"
+	dport="${4}"
+
+	run_cmd python3 ./openvswitch/ovs-dpctl.py add-if ovs_br0 ${type}_a -t ${type} || return 1
+
+	ports=$(python3 ./openvswitch/ovs-dpctl.py show)
+	br0_port=$(echo "$ports" | grep -E "\sovs_br0" | sed -e 's@port @@' | cut -d: -f1 | xargs)
+	type_a_port=$(echo "$ports" | grep ${type}_a | sed -e 's@port @@' | cut -d: -f1 | xargs)
+	veth_a_port=$(echo "$ports" | grep veth_A | sed -e 's@port @@' | cut -d: -f1 | xargs)
+
+	v4_a_tun="${prefix4}.${a_r1}.1"
+	v4_b_tun="${prefix4}.${b_r1}.1"
+
+	v6_a_tun="${prefix6}:${a_r1}::1"
+	v6_b_tun="${prefix6}:${b_r1}::1"
+
+	if [ "${v4_a_tun}" = "${a_addr}" ]; then
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x0800),ipv4()" \
+		    "set(tunnel(tun_id=1,dst=${v4_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}"
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x86dd),ipv6()" \
+		    "set(tunnel(tun_id=1,dst=${v4_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}"
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),tunnel(tun_id=1,src=${v4_b_tun},dst=${v4_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x0800),ipv4()" \
+		    "${veth_a_port}"
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),tunnel(tun_id=1,src=${v4_b_tun},dst=${v4_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x86dd),ipv6()" \
+		    "${veth_a_port}"
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),tunnel(tun_id=1,src=${v4_b_tun},dst=${v4_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x0806),arp()" \
+		    "${veth_a_port}"
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x0806),arp(sip=${veth4_c_addr},tip=${tunnel4_b_addr})" \
+		    "set(tunnel(tun_id=1,dst=${v4_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}"
+	else
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x0800),ipv4()" \
+		    "set(tunnel(tun_id=1,ipv6_dst=${v6_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}"
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x86dd),ipv6()" \
+		    "set(tunnel(tun_id=1,ipv6_dst=${v6_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}"
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),tunnel(tun_id=1,ipv6_src=${v6_b_tun},ipv6_dst=${v6_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x0800),ipv4()" \
+		    "${veth_a_port}"
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),tunnel(tun_id=1,ipv6_src=${v6_b_tun},ipv6_dst=${v6_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x86dd),ipv6()" \
+		    "${veth_a_port}"
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),tunnel(tun_id=1,ipv6_src=${v6_b_tun},ipv6_dst=${v6_a_tun}),in_port(${type_a_port}),eth(),eth_type(0x0806),arp()" \
+		    "${veth_a_port}"
+		run_cmd python3 ./openvswitch/ovs-dpctl.py add-flow ovs_br0 \
+		    "recirc_id(0),in_port(${veth_a_port}),eth(),eth_type(0x0806),arp(sip=${veth4_c_addr},tip=${tunnel4_b_addr})" \
+		    "set(tunnel(tun_id=1,ipv6_dst=${v6_b_tun},ttl=64,tp_dst=${dport},flags(df|csum))),${type_a_port}"
+	fi
+}
+
+setup_ovs_via_vswitchd() {
+	type="${1}"
+	b_addr="${2}"
+
+	run_cmd ovs-vsctl add-port ovs_br0 ${type}_a -- \
+		set interface ${type}_a type=${type} \
+		options:remote_ip=${b_addr} options:key=1 options:csum=true || return 1
+}
+
 setup_ovs_vxlan_or_geneve() {
 	type="${1}"
 	a_addr="${2}"
 	b_addr="${3}"
+	dport="6081"
 
 	if [ "${type}" = "vxlan" ]; then
+		dport="4789"
 		opts="${opts} ttl 64 dstport 4789"
 		opts_b="local ${b_addr}"
 	fi
 
-	run_cmd ovs-vsctl add-port ovs_br0 ${type}_a -- \
-		set interface ${type}_a type=${type} \
-		options:remote_ip=${b_addr} options:key=1 options:csum=true || return 1
+	setup_ovs_via_internal_utility "${type}" "${a_addr}" "${b_addr}" \
+				       "${dport}" || \
+	    setup_ovs_via_vswitchd "${type}" "${b_addr}" || return 1
 
 	run_cmd ${ns_b} ip link add ${type}_b type ${type} id 1 ${opts_b} remote ${a_addr} ${opts} || return 1
 
 	run_cmd ${ns_b} ip addr add ${tunnel4_b_addr}/${tunnel4_mask} dev ${type}_b
 	run_cmd ${ns_b} ip addr add ${tunnel6_b_addr}/${tunnel6_mask} dev ${type}_b
 
+	run_cmd ip link set ${type}_a up
 	run_cmd ${ns_b} ip link set ${type}_b up
 }
 
@@ -880,8 +952,24 @@ setup_ovs_vxlan6() {
 	setup_ovs_vxlan_or_geneve vxlan  ${prefix6}:${a_r1}::1 ${prefix6}:${b_r1}::1
 }
 
+setup_ovs_br_internal() {
+	run_cmd python3 ./openvswitch/ovs-dpctl.py add-dp ovs_br0 || \
+		return 1
+}
+
+setup_ovs_br_vswitchd() {
+	run_cmd ovs-vsctl add-br ovs_br0 || return 1
+}
+
+setup_ovs_add_if() {
+	ifname="${1}"
+	run_cmd python3 ./openvswitch/ovs-dpctl.py add-if ovs_br0 \
+		"${ifname}" || \
+		run_cmd ovs-vsctl add-port ovs_br0 "${ifname}"
+}
+
 setup_ovs_bridge() {
-	run_cmd ovs-vsctl add-br ovs_br0 || return $ksft_skip
+	setup_ovs_br_internal || setup_ovs_br_vswitchd || return $ksft_skip
 	run_cmd ip link set ovs_br0 up
 
 	run_cmd ${ns_c} ip link add veth_C-A type veth peer name veth_A-C
@@ -891,7 +979,7 @@ setup_ovs_bridge() {
 	run_cmd ${ns_c} ip link set veth_C-A up
 	run_cmd ${ns_c} ip addr add ${veth4_c_addr}/${veth4_mask} dev veth_C-A
 	run_cmd ${ns_c} ip addr add ${veth6_c_addr}/${veth6_mask} dev veth_C-A
-	run_cmd ovs-vsctl add-port ovs_br0 veth_A-C
+	setup_ovs_add_if veth_A-C
 
 	# Move veth_A-R1 to init
 	run_cmd ${ns_a} ip link set veth_A-R1 netns 1
@@ -922,6 +1010,18 @@ trace() {
 	sleep 1
 }
 
+cleanup_del_ovs_internal() {
+	# squelch the output of the del-if commands since it can be wordy
+	python3 ./openvswitch/ovs-dpctl.py del-if ovs_br0 -d true vxlan_a	>/dev/null	2>&1
+	python3 ./openvswitch/ovs-dpctl.py del-if ovs_br0 -d true geneve_a	>/dev/null	2>&1
+	python3 ./openvswitch/ovs-dpctl.py del-dp ovs_br0			>/dev/null	2>&1
+}
+
+cleanup_del_ovs_vswitchd() {
+	ovs-vsctl --if-exists del-port vxlan_a	2>/dev/null
+	ovs-vsctl --if-exists del-br ovs_br0	2>/dev/null
+}
+
 cleanup() {
 	for pid in ${tcpdump_pids}; do
 		kill ${pid}
@@ -940,10 +1040,10 @@ cleanup() {
 
 	cleanup_all_ns
 
-	ip link del veth_A-C			2>/dev/null
-	ip link del veth_A-R1			2>/dev/null
-	ovs-vsctl --if-exists del-port vxlan_a	2>/dev/null
-	ovs-vsctl --if-exists del-br ovs_br0	2>/dev/null
+	ip link del veth_A-C		2>/dev/null
+	ip link del veth_A-R1		2>/dev/null
+	cleanup_del_ovs_internal
+	cleanup_del_ovs_vswitchd
 	rm -f "$tmpoutfile"
 }
 
@@ -1397,6 +1497,12 @@ test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception() {
 	outer_family=${3}
 	ll_mtu=4000
 
+	if [ "${type}" = "vxlan" ]; then
+		tun_a="vxlan_sys_4789"
+	elif [ "${type}" = "geneve" ]; then
+		tun_a="genev_sys_6081"
+	fi
+
 	if [ ${outer_family} -eq 4 ]; then
 		setup namespaces routing ovs_bridge ovs_${type}4 || return $ksft_skip
 		#                      IPv4 header   UDP header   VXLAN/GENEVE header   Ethernet header
@@ -1407,17 +1513,11 @@ test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception() {
 		exp_mtu=$((${ll_mtu} - 40          - 8          - 8                   - 14))
 	fi
 
-	if [ "${type}" = "vxlan" ]; then
-		tun_a="vxlan_sys_4789"
-	elif [ "${type}" = "geneve" ]; then
-		tun_a="genev_sys_6081"
-	fi
-
-	trace ""        "${tun_a}"  "${ns_b}"  ${type}_b \
-	      ""        veth_A-R1   "${ns_r1}" veth_R1-A \
-	      "${ns_b}" veth_B-R1   "${ns_r1}" veth_R1-B \
-	      ""        ovs_br0     ""         veth-A-C  \
-	      "${ns_c}" veth_C-A
+	trace ""        ${type}_a    "${ns_b}"  ${type}_b \
+	      ""        veth_A-R1    "${ns_r1}" veth_R1-A \
+	      "${ns_b}" veth_B-R1    "${ns_r1}" veth_R1-B \
+	      ""        ovs_br0      ""         veth-A_C  \
+	      "${ns_c}" veth_C-A     ""         "${tun_a}"
 
 	if [ ${family} -eq 4 ]; then
 		ping=ping
@@ -1436,8 +1536,9 @@ test_pmtu_ipvX_over_ovs_vxlanY_or_geneveY_exception() {
 	mtu "${ns_b}"  veth_B-R1 ${ll_mtu}
 	mtu "${ns_r1}" veth_R1-B ${ll_mtu}
 
-	mtu ""        ${tun_a}  $((${ll_mtu} + 1000))
-	mtu "${ns_b}" ${type}_b $((${ll_mtu} + 1000))
+	mtu ""        ${tun_a}  $((${ll_mtu} + 1000)) 2>/dev/null || \
+		mtu ""        ${type}_a  $((${ll_mtu} + 1000)) 2>/dev/null
+	mtu "${ns_b}" ${type}_b  $((${ll_mtu} + 1000))
 
 	run_cmd ${ns_c} ${ping} -q -M want -i 0.1 -c 20 -s $((${ll_mtu} + 500)) ${dst} || return 1
 
-- 
cgit v1.2.3-58-ga151


From 6f437f5c91b03b105c0efa128a687746254049e4 Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Tue, 25 Jun 2024 13:22:45 -0400
Subject: selftests: net: add config for openvswitch

The pmtu testing will require that the OVS module is installed,
so do that.

Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Simon Horman <horms@kernel.org>
Signed-off-by: Aaron Conole <aconole@redhat.com>
Link: https://patch.msgid.link/20240625172245.233874-8-aconole@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/config | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/tools/testing/selftests/net/config b/tools/testing/selftests/net/config
index f46e27cd1e70..5b9baf708950 100644
--- a/tools/testing/selftests/net/config
+++ b/tools/testing/selftests/net/config
@@ -74,7 +74,12 @@ CONFIG_NET_SCH_ETF=m
 CONFIG_NET_SCH_NETEM=y
 CONFIG_NET_SCH_PRIO=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NF_CONNTRACK_OVS=y
 CONFIG_NF_FLOW_TABLE=m
+CONFIG_OPENVSWITCH=m
+CONFIG_OPENVSWITCH_GENEVE=m
+CONFIG_OPENVSWITCH_GRE=m
+CONFIG_OPENVSWITCH_VXLAN=m
 CONFIG_PSAMPLE=m
 CONFIG_TCP_MD5SIG=y
 CONFIG_TEST_BLACKHOLE_DEV=m
-- 
cgit v1.2.3-58-ga151


From 2d5f6801db8ec0ce97bc520ecd51a7be06a35042 Mon Sep 17 00:00:00 2001
From: Sagi Grimberg <sagi@grimberg.me>
Date: Wed, 26 Jun 2024 10:01:53 +0300
Subject: Revert "net: micro-optimize skb_datagram_iter"

This reverts commit 934c29999b57b835d65442da6f741d5e27f3b584.
This triggered a usercopy BUG() in systems with HIGHMEM, reported
by the test robot in:
 https://lore.kernel.org/oe-lkp/202406161539.b5ff7b20-oliver.sang@intel.com

Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Link: https://patch.msgid.link/20240626070153.759257-1-sagi@grimberg.me
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/datagram.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/core/datagram.c b/net/core/datagram.c
index 95f242591fd2..e614cfd8e14a 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -417,14 +417,14 @@ static int __skb_datagram_iter(const struct sk_buff *skb, int offset,
 		end = start + skb_frag_size(frag);
 		if ((copy = end - offset) > 0) {
 			struct page *page = skb_frag_page(frag);
-			u8 *vaddr = kmap_local_page(page);
+			u8 *vaddr = kmap(page);
 
 			if (copy > len)
 				copy = len;
 			n = INDIRECT_CALL_1(cb, simple_copy_to_iter,
 					vaddr + skb_frag_off(frag) + offset - start,
 					copy, data, to);
-			kunmap_local(vaddr);
+			kunmap(page);
 			offset += n;
 			if (n != copy)
 				goto short_copy;
-- 
cgit v1.2.3-58-ga151


From 94833addfaba89d12e5dbd82e350a692c00648ab Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Wed, 26 Jun 2024 10:35:02 -0700
Subject: net: thunderx: Unembed netdev structure

Embedding net_device into structures prohibits the usage of flexible
arrays in the net_device structure. For more details, see the discussion
at [1].

Un-embed the net_devices from struct lmac by converting them
into pointers, and allocating them dynamically. Use the leverage
alloc_netdev() to allocate the net_device object at
bgx_lmac_enable().

The free of the device occurs at bgx_lmac_disable().

 Do not free_netdevice() if bgx_lmac_enable() fails after lmac->netdev
is allocated, since bgx_lmac_disable() is called if bgx_lmac_enable()
fails, and lmac->netdev will be freed there (similarly to lmac->dmacs).

Link: https://lore.kernel.org/all/20240229225910.79e224cf@kernel.org/ [1]
Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20240626173503.87636-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cavium/thunder/thunder_bgx.c | 21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
index a317feb8decb..a40c266c37f2 100644
--- a/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
+++ b/drivers/net/ethernet/cavium/thunder/thunder_bgx.c
@@ -54,7 +54,7 @@ struct lmac {
 	bool			link_up;
 	int			lmacid; /* ID within BGX */
 	int			lmacid_bd; /* ID on board */
-	struct net_device       netdev;
+	struct net_device       *netdev;
 	struct phy_device       *phydev;
 	unsigned int            last_duplex;
 	unsigned int            last_link;
@@ -590,10 +590,12 @@ static void bgx_sgmii_change_link_state(struct lmac *lmac)
 
 static void bgx_lmac_handler(struct net_device *netdev)
 {
-	struct lmac *lmac = container_of(netdev, struct lmac, netdev);
 	struct phy_device *phydev;
+	struct lmac *lmac, **priv;
 	int link_changed = 0;
 
+	priv = netdev_priv(netdev);
+	lmac = *priv;
 	phydev = lmac->phydev;
 
 	if (!phydev->link && lmac->last_link)
@@ -1052,12 +1054,18 @@ static int phy_interface_mode(u8 lmac_type)
 
 static int bgx_lmac_enable(struct bgx *bgx, u8 lmacid)
 {
-	struct lmac *lmac;
+	struct lmac *lmac, **priv;
 	u64 cfg;
 
 	lmac = &bgx->lmac[lmacid];
 	lmac->bgx = bgx;
 
+	lmac->netdev = alloc_netdev_dummy(sizeof(struct lmac *));
+	if (!lmac->netdev)
+		return -ENOMEM;
+	priv = netdev_priv(lmac->netdev);
+	*priv = lmac;
+
 	if ((lmac->lmac_type == BGX_MODE_SGMII) ||
 	    (lmac->lmac_type == BGX_MODE_QSGMII) ||
 	    (lmac->lmac_type == BGX_MODE_RGMII)) {
@@ -1116,7 +1124,7 @@ static int bgx_lmac_enable(struct bgx *bgx, u8 lmacid)
 		}
 		lmac->phydev->dev_flags = 0;
 
-		if (phy_connect_direct(&lmac->netdev, lmac->phydev,
+		if (phy_connect_direct(lmac->netdev, lmac->phydev,
 				       bgx_lmac_handler,
 				       phy_interface_mode(lmac->lmac_type)))
 			return -ENODEV;
@@ -1183,6 +1191,7 @@ static void bgx_lmac_disable(struct bgx *bgx, u8 lmacid)
 	    (lmac->lmac_type != BGX_MODE_10G_KR) && lmac->phydev)
 		phy_disconnect(lmac->phydev);
 
+	free_netdev(lmac->netdev);
 	lmac->phydev = NULL;
 }
 
@@ -1414,7 +1423,7 @@ static acpi_status bgx_acpi_register_phy(acpi_handle handle,
 
 	acpi_get_mac_address(dev, adev, bgx->lmac[bgx->acpi_lmac_idx].mac);
 
-	SET_NETDEV_DEV(&bgx->lmac[bgx->acpi_lmac_idx].netdev, dev);
+	SET_NETDEV_DEV(bgx->lmac[bgx->acpi_lmac_idx].netdev, dev);
 
 	bgx->lmac[bgx->acpi_lmac_idx].lmacid = bgx->acpi_lmac_idx;
 	bgx->acpi_lmac_idx++; /* move to next LMAC */
@@ -1483,7 +1492,7 @@ static int bgx_init_of_phy(struct bgx *bgx)
 
 		of_get_mac_address(node, bgx->lmac[lmac].mac);
 
-		SET_NETDEV_DEV(&bgx->lmac[lmac].netdev, &bgx->pdev->dev);
+		SET_NETDEV_DEV(bgx->lmac[lmac].netdev, &bgx->pdev->dev);
 		bgx->lmac[lmac].lmacid = lmac;
 
 		phy_np = of_parse_phandle(node, "phy-handle", 0);
-- 
cgit v1.2.3-58-ga151


From dd20d16dae83733d7c98c167b8d935e7b63db7c7 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Wed, 29 May 2024 11:12:13 +0200
Subject: can: rcar_canfd: Simplify clock handling

The main CAN clock is either the internal CANFD clock, or the external
CAN clock.  Hence replace the two-valued enum by a simple boolean flag.
Consolidate all CANFD clock handling inside a single branch.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Link: https://lore.kernel.org/all/2cf38c10b83c8e5c04d68b17a930b6d9dbf66f40.1716973640.git.geert+renesas@glider.be
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/rcar/rcar_canfd.c | 24 +++++++-----------------
 1 file changed, 7 insertions(+), 17 deletions(-)

diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c
index b82842718735..474840b58e8f 100644
--- a/drivers/net/can/rcar/rcar_canfd.c
+++ b/drivers/net/can/rcar/rcar_canfd.c
@@ -508,12 +508,6 @@
  */
 #define RCANFD_CFFIFO_IDX		0
 
-/* fCAN clock select register settings */
-enum rcar_canfd_fcanclk {
-	RCANFD_CANFDCLK = 0,		/* CANFD clock */
-	RCANFD_EXTCLK,			/* Externally input clock */
-};
-
 struct rcar_canfd_global;
 
 struct rcar_canfd_hw_info {
@@ -545,8 +539,8 @@ struct rcar_canfd_global {
 	struct platform_device *pdev;	/* Respective platform device */
 	struct clk *clkp;		/* Peripheral clock */
 	struct clk *can_clk;		/* fCAN clock */
-	enum rcar_canfd_fcanclk fcan;	/* CANFD or Ext clock */
 	unsigned long channels_mask;	/* Enabled channels mask */
+	bool extclk;			/* CANFD or Ext clock */
 	bool fdmode;			/* CAN FD or Classical CAN only mode */
 	struct reset_control *rstc1;
 	struct reset_control *rstc2;
@@ -777,7 +771,7 @@ static void rcar_canfd_configure_controller(struct rcar_canfd_global *gpriv)
 		cfg |= RCANFD_GCFG_CMPOC;
 
 	/* Set External Clock if selected */
-	if (gpriv->fcan != RCANFD_CANFDCLK)
+	if (gpriv->extclk)
 		cfg |= RCANFD_GCFG_DCS;
 
 	rcar_canfd_set_bit(gpriv->base, RCANFD_GCFG, cfg);
@@ -1941,16 +1935,12 @@ static int rcar_canfd_probe(struct platform_device *pdev)
 			return dev_err_probe(dev, PTR_ERR(gpriv->can_clk),
 					     "cannot get canfd clock\n");
 
-		gpriv->fcan = RCANFD_CANFDCLK;
-
+		/* CANFD clock may be further divided within the IP */
+		fcan_freq = clk_get_rate(gpriv->can_clk) / info->postdiv;
 	} else {
-		gpriv->fcan = RCANFD_EXTCLK;
+		fcan_freq = clk_get_rate(gpriv->can_clk);
+		gpriv->extclk = true;
 	}
-	fcan_freq = clk_get_rate(gpriv->can_clk);
-
-	if (gpriv->fcan == RCANFD_CANFDCLK)
-		/* CANFD clock is further divided by (1/2) within the IP */
-		fcan_freq /= info->postdiv;
 
 	addr = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(addr)) {
@@ -2060,7 +2050,7 @@ static int rcar_canfd_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, gpriv);
 	dev_info(dev, "global operational state (clk %d, fdmode %d)\n",
-		 gpriv->fcan, gpriv->fdmode);
+		 gpriv->extclk, gpriv->fdmode);
 	return 0;
 
 fail_channel:
-- 
cgit v1.2.3-58-ga151


From 0c1d0a69c5e7314f3787bc4272e286d800dbfc56 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Wed, 29 May 2024 11:12:14 +0200
Subject: can: rcar_canfd: Improve printing of global operational state

Replace the printing of internal numerical values by the printing of
strings reflecting their meaning, to make the message self-explanatory.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Wolfram Sang <wsa@kernel.org>
Reviewed-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Link: https://lore.kernel.org/all/14c8c5ce026e9fec128404706d1c73c8ffa11ced.1716973640.git.geert+renesas@glider.be
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/rcar/rcar_canfd.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c
index 474840b58e8f..c2c1c47bcc7a 100644
--- a/drivers/net/can/rcar/rcar_canfd.c
+++ b/drivers/net/can/rcar/rcar_canfd.c
@@ -2049,8 +2049,9 @@ static int rcar_canfd_probe(struct platform_device *pdev)
 	}
 
 	platform_set_drvdata(pdev, gpriv);
-	dev_info(dev, "global operational state (clk %d, fdmode %d)\n",
-		 gpriv->extclk, gpriv->fdmode);
+	dev_info(dev, "global operational state (%s clk, %s mode)\n",
+		 gpriv->extclk ? "ext" : "canfd",
+		 gpriv->fdmode ? "fd" : "classical");
 	return 0;
 
 fail_channel:
-- 
cgit v1.2.3-58-ga151


From f9a83965d40ef162bab1e3af338523b8da69cb82 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Wed, 29 May 2024 11:12:15 +0200
Subject: can: rcar_canfd: Remove superfluous parentheses in address
 calculations

There is no need to wrap simple variables or multiplications inside
parentheses.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Vincent Mailhol <mailhol.vincent@wanadoo.fr>
Link: https://lore.kernel.org/all/b5aee80895fa029070fd37d1d837cf1c0ecb52dc.1716973640.git.geert+renesas@glider.be
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/rcar/rcar_canfd.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c
index c2c1c47bcc7a..c919668bbe7a 100644
--- a/drivers/net/can/rcar/rcar_canfd.c
+++ b/drivers/net/can/rcar/rcar_canfd.c
@@ -627,28 +627,28 @@ static inline void rcar_canfd_update(u32 mask, u32 val, u32 __iomem *reg)
 
 static inline u32 rcar_canfd_read(void __iomem *base, u32 offset)
 {
-	return readl(base + (offset));
+	return readl(base + offset);
 }
 
 static inline void rcar_canfd_write(void __iomem *base, u32 offset, u32 val)
 {
-	writel(val, base + (offset));
+	writel(val, base + offset);
 }
 
 static void rcar_canfd_set_bit(void __iomem *base, u32 reg, u32 val)
 {
-	rcar_canfd_update(val, val, base + (reg));
+	rcar_canfd_update(val, val, base + reg);
 }
 
 static void rcar_canfd_clear_bit(void __iomem *base, u32 reg, u32 val)
 {
-	rcar_canfd_update(val, 0, base + (reg));
+	rcar_canfd_update(val, 0, base + reg);
 }
 
 static void rcar_canfd_update_bit(void __iomem *base, u32 reg,
 				  u32 mask, u32 val)
 {
-	rcar_canfd_update(mask, val, base + (reg));
+	rcar_canfd_update(mask, val, base + reg);
 }
 
 static void rcar_canfd_get_data(struct rcar_canfd_channel *priv,
@@ -659,7 +659,7 @@ static void rcar_canfd_get_data(struct rcar_canfd_channel *priv,
 	lwords = DIV_ROUND_UP(cf->len, sizeof(u32));
 	for (i = 0; i < lwords; i++)
 		*((u32 *)cf->data + i) =
-			rcar_canfd_read(priv->base, off + (i * sizeof(u32)));
+			rcar_canfd_read(priv->base, off + i * sizeof(u32));
 }
 
 static void rcar_canfd_put_data(struct rcar_canfd_channel *priv,
@@ -669,7 +669,7 @@ static void rcar_canfd_put_data(struct rcar_canfd_channel *priv,
 
 	lwords = DIV_ROUND_UP(cf->len, sizeof(u32));
 	for (i = 0; i < lwords; i++)
-		rcar_canfd_write(priv->base, off + (i * sizeof(u32)),
+		rcar_canfd_write(priv->base, off + i * sizeof(u32),
 				 *((u32 *)cf->data + i));
 }
 
-- 
cgit v1.2.3-58-ga151


From 62d73261a0cfa309fd07d0da41a74b3f47f25b6f Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sun, 23 Jun 2024 22:01:50 +0200
Subject: can: m_can: Constify struct m_can_ops

'struct m_can_ops' is not modified in these drivers.

Constifying this structure moves some data to a read-only section, so
increase overall security.

On a x86_64, with allmodconfig, as an example:
Before:
======
   text	   data	    bss	    dec	    hex	filename
   4806	    520	      0	   5326	   14ce	drivers/net/can/m_can/m_can_pci.o

After:
=====
   text	   data	    bss	    dec	    hex	filename
   4862	    464	      0	   5326	   14ce	drivers/net/can/m_can/m_can_pci.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Link: https://lore.kernel.org/all/a17b96d1be5341c11f263e1e45c9de1cb754e416.1719172843.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/m_can/m_can.h          | 2 +-
 drivers/net/can/m_can/m_can_pci.c      | 2 +-
 drivers/net/can/m_can/m_can_platform.c | 2 +-
 drivers/net/can/m_can/tcan4x5x-core.c  | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/can/m_can/m_can.h b/drivers/net/can/m_can/m_can.h
index 3a9edc292593..92b2bd8628e6 100644
--- a/drivers/net/can/m_can/m_can.h
+++ b/drivers/net/can/m_can/m_can.h
@@ -91,7 +91,7 @@ struct m_can_classdev {
 
 	ktime_t irq_timer_wait;
 
-	struct m_can_ops *ops;
+	const struct m_can_ops *ops;
 
 	int version;
 	u32 irqstatus;
diff --git a/drivers/net/can/m_can/m_can_pci.c b/drivers/net/can/m_can/m_can_pci.c
index 45400de4163d..d72fe771dfc7 100644
--- a/drivers/net/can/m_can/m_can_pci.c
+++ b/drivers/net/can/m_can/m_can_pci.c
@@ -77,7 +77,7 @@ static int iomap_write_fifo(struct m_can_classdev *cdev, int offset,
 	return 0;
 }
 
-static struct m_can_ops m_can_pci_ops = {
+static const struct m_can_ops m_can_pci_ops = {
 	.read_reg = iomap_read_reg,
 	.write_reg = iomap_write_reg,
 	.write_fifo = iomap_write_fifo,
diff --git a/drivers/net/can/m_can/m_can_platform.c b/drivers/net/can/m_can/m_can_platform.c
index df0367124b4c..983ab80260dd 100644
--- a/drivers/net/can/m_can/m_can_platform.c
+++ b/drivers/net/can/m_can/m_can_platform.c
@@ -68,7 +68,7 @@ static int iomap_write_fifo(struct m_can_classdev *cdev, int offset,
 	return 0;
 }
 
-static struct m_can_ops m_can_plat_ops = {
+static const struct m_can_ops m_can_plat_ops = {
 	.read_reg = iomap_read_reg,
 	.write_reg = iomap_write_reg,
 	.write_fifo = iomap_write_fifo,
diff --git a/drivers/net/can/m_can/tcan4x5x-core.c b/drivers/net/can/m_can/tcan4x5x-core.c
index d723206ac7c9..2f73bf3abad8 100644
--- a/drivers/net/can/m_can/tcan4x5x-core.c
+++ b/drivers/net/can/m_can/tcan4x5x-core.c
@@ -357,7 +357,7 @@ static int tcan4x5x_get_gpios(struct m_can_classdev *cdev,
 	return 0;
 }
 
-static struct m_can_ops tcan4x5x_ops = {
+static const struct m_can_ops tcan4x5x_ops = {
 	.init = tcan4x5x_init,
 	.read_reg = tcan4x5x_read_reg,
 	.write_reg = tcan4x5x_write_reg,
-- 
cgit v1.2.3-58-ga151


From a7e5793035792cc46a1a4b0a783655ffa897dfe9 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 27 Jun 2024 10:44:11 +0200
Subject: wifi: nl80211: don't give key data to userspace

When a key is requested by userspace, there's really no need
to include the key data, the sequence counter is really what
userspace needs in this case. The fact that it's included is
just a historic quirk.

Remove the key data.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240627104411.b6a4f097e4ea.I7e6cc976cb9e8a80ef25a3351330f313373b4578@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/nl80211.c | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 19d81200a2a2..674368d028f3 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -4487,10 +4487,7 @@ static void get_key_callback(void *c, struct key_params *params)
 	struct nlattr *key;
 	struct get_key_cookie *cookie = c;
 
-	if ((params->key &&
-	     nla_put(cookie->msg, NL80211_ATTR_KEY_DATA,
-		     params->key_len, params->key)) ||
-	    (params->seq &&
+	if ((params->seq &&
 	     nla_put(cookie->msg, NL80211_ATTR_KEY_SEQ,
 		     params->seq_len, params->seq)) ||
 	    (params->cipher &&
@@ -4502,10 +4499,7 @@ static void get_key_callback(void *c, struct key_params *params)
 	if (!key)
 		goto nla_put_failure;
 
-	if ((params->key &&
-	     nla_put(cookie->msg, NL80211_KEY_DATA,
-		     params->key_len, params->key)) ||
-	    (params->seq &&
+	if ((params->seq &&
 	     nla_put(cookie->msg, NL80211_KEY_SEQ,
 		     params->seq_len, params->seq)) ||
 	    (params->cipher &&
-- 
cgit v1.2.3-58-ga151


From 0a3d5991438fcf89c3aebf55c30231faee31567e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 27 Jun 2024 10:44:12 +0200
Subject: wifi: mac80211: remove key data from get_key callback

This is now unused in nl80211, so there's no need to set
the pointer/length.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240627104411.f8ac49a8cfe0.Ic01f775903f0acd10b9ba77eb39d4ed50e709173@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 7eb2e5bedb6f..3d49b3ee3a2b 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -742,9 +742,6 @@ static int ieee80211_get_key(struct wiphy *wiphy, struct net_device *dev,
 		break;
 	}
 
-	params.key = key->conf.key;
-	params.key_len = key->conf.keylen;
-
 	callback(cookie, &params);
 	err = 0;
 
-- 
cgit v1.2.3-58-ga151


From 8fda53719a596fa2a2880c42b5fa4126fbbbfc3d Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Tue, 25 Jun 2024 20:42:28 +0200
Subject: dt-bindings: net: realtek,rtl82xx: Document known PHY IDs as
 compatible strings

Extract known PHY IDs from Linux kernel realtek PHY driver
and convert them into supported compatible string list for
this DT binding document.

Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../devicetree/bindings/net/realtek,rtl82xx.yaml   | 23 ++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/realtek,rtl82xx.yaml b/Documentation/devicetree/bindings/net/realtek,rtl82xx.yaml
index bb94a2388520..18ee72f5c74a 100644
--- a/Documentation/devicetree/bindings/net/realtek,rtl82xx.yaml
+++ b/Documentation/devicetree/bindings/net/realtek,rtl82xx.yaml
@@ -18,6 +18,29 @@ allOf:
   - $ref: ethernet-phy.yaml#
 
 properties:
+  compatible:
+    enum:
+      - ethernet-phy-id001c.c800
+      - ethernet-phy-id001c.c816
+      - ethernet-phy-id001c.c838
+      - ethernet-phy-id001c.c840
+      - ethernet-phy-id001c.c848
+      - ethernet-phy-id001c.c849
+      - ethernet-phy-id001c.c84a
+      - ethernet-phy-id001c.c862
+      - ethernet-phy-id001c.c878
+      - ethernet-phy-id001c.c880
+      - ethernet-phy-id001c.c910
+      - ethernet-phy-id001c.c912
+      - ethernet-phy-id001c.c913
+      - ethernet-phy-id001c.c914
+      - ethernet-phy-id001c.c915
+      - ethernet-phy-id001c.c916
+      - ethernet-phy-id001c.c942
+      - ethernet-phy-id001c.c961
+      - ethernet-phy-id001c.cad0
+      - ethernet-phy-id001c.cb00
+
   realtek,clkout-disable:
     type: boolean
     description:
-- 
cgit v1.2.3-58-ga151


From 3f5d7ff7c533af6f06a39becb77a7afbfaa8b782 Mon Sep 17 00:00:00 2001
From: Ilan Peer <ilan.peer@intel.com>
Date: Thu, 27 Jun 2024 10:46:00 +0200
Subject: wifi: mac80211: Use the link BSS configuration for beacon processing

The beacon processing should be fully done in the context of the link.
This also resolves a bug with CQM handling with MLO as in such a case
the RSSI thresholds configuration is maintained in the link context and
not in the interface context.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Link: https://patch.msgid.link/20240627104600.bb2f0f697881.I675b6a8a186b717f3eef79113c27361fd1a7622c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/mlme.c | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 3d207d79d11f..80a10cd49565 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -6696,7 +6696,7 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 {
 	struct ieee80211_sub_if_data *sdata = link->sdata;
 	struct ieee80211_if_managed *ifmgd = &sdata->u.mgd;
-	struct ieee80211_bss_conf *bss_conf = &sdata->vif.bss_conf;
+	struct ieee80211_bss_conf *bss_conf = link->conf;
 	struct ieee80211_vif_cfg *vif_cfg = &sdata->vif.cfg;
 	struct ieee80211_mgmt *mgmt = (void *) hdr;
 	size_t baselen;
@@ -6740,7 +6740,7 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 	parse_params.len = len - baselen;
 
 	rcu_read_lock();
-	chanctx_conf = rcu_dereference(link->conf->chanctx_conf);
+	chanctx_conf = rcu_dereference(bss_conf->chanctx_conf);
 	if (!chanctx_conf) {
 		rcu_read_unlock();
 		return;
@@ -6770,11 +6770,11 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 		ifmgd->assoc_data->need_beacon = false;
 		if (ieee80211_hw_check(&local->hw, TIMING_BEACON_ONLY) &&
 		    !ieee80211_is_s1g_beacon(hdr->frame_control)) {
-			link->conf->sync_tsf =
+			bss_conf->sync_tsf =
 				le64_to_cpu(mgmt->u.beacon.timestamp);
-			link->conf->sync_device_ts =
+			bss_conf->sync_device_ts =
 				rx_status->device_timestamp;
-			link->conf->sync_dtim_count = elems->dtim_count;
+			bss_conf->sync_dtim_count = elems->dtim_count;
 		}
 
 		if (elems->mbssid_config_ie)
@@ -6798,7 +6798,7 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 	}
 
 	if (!ifmgd->associated ||
-	    !ieee80211_rx_our_beacon(bssid, link->conf->bss))
+	    !ieee80211_rx_our_beacon(bssid, bss_conf->bss))
 		return;
 	bssid = link->u.mgd.bssid;
 
@@ -6825,7 +6825,7 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 	 */
 	if (!ieee80211_is_s1g_beacon(hdr->frame_control))
 		ncrc = crc32_be(0, (void *)&mgmt->u.beacon.beacon_int, 4);
-	parse_params.bss = link->conf->bss;
+	parse_params.bss = bss_conf->bss;
 	parse_params.filter = care_about_ies;
 	parse_params.crc = ncrc;
 	elems = ieee802_11_parse_elems_full(&parse_params);
@@ -6906,11 +6906,11 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 	 */
 	if (ieee80211_hw_check(&local->hw, TIMING_BEACON_ONLY) &&
 	    !ieee80211_is_s1g_beacon(hdr->frame_control)) {
-		link->conf->sync_tsf =
+		bss_conf->sync_tsf =
 			le64_to_cpu(mgmt->u.beacon.timestamp);
-		link->conf->sync_device_ts =
+		bss_conf->sync_device_ts =
 			rx_status->device_timestamp;
-		link->conf->sync_dtim_count = elems->dtim_count;
+		bss_conf->sync_dtim_count = elems->dtim_count;
 	}
 
 	if ((ncrc == link->u.mgd.beacon_crc && link->u.mgd.beacon_crc_valid) ||
@@ -6973,10 +6973,10 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_link_data *link,
 		goto free;
 	}
 
-	if (WARN_ON(!link->conf->chanreq.oper.chan))
+	if (WARN_ON(!bss_conf->chanreq.oper.chan))
 		goto free;
 
-	sband = local->hw.wiphy->bands[link->conf->chanreq.oper.chan->band];
+	sband = local->hw.wiphy->bands[bss_conf->chanreq.oper.chan->band];
 
 	changed |= ieee80211_recalc_twt_req(sdata, sband, link, link_sta, elems);
 
-- 
cgit v1.2.3-58-ga151


From 4314bb46cb56851f13bda9db356ebc3cf750b7f1 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 27 Jun 2024 10:46:22 +0200
Subject: wifi: mac80211_hwsim: fix kernel-doc

Some kernel-doc in the header file was mismatched, fix that.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240627104621.3b2dd62adb2c.I8a4c469b674e5eac737378a1ed2358b57d01613f@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.h b/drivers/net/wireless/virtual/mac80211_hwsim.h
index 21b1afd83dc1..28c1db482e79 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.h
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.h
@@ -257,7 +257,7 @@ enum hwsim_tx_rate_flags {
 };
 
 /**
- * struct hwsim_tx_rate - rate selection/status
+ * struct hwsim_tx_rate_flag - rate selection/status
  *
  * @idx: rate index to attempt to send with
  * @flags: the rate flags according to &enum hwsim_tx_rate_flags
@@ -295,7 +295,7 @@ enum hwsim_vqs {
 };
 
 /**
- * enum hwsim_rate_info -- bitrate information.
+ * enum hwsim_rate_info_attributes - bitrate information.
  *
  * Information about a receiving or transmitting bitrate
  * that can be mapped to struct rate_info
-- 
cgit v1.2.3-58-ga151


From 8c62617295d3c4cd03f1a02c3b9bf9d4e6d6e0c6 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Thu, 27 Jun 2024 13:25:27 +0200
Subject: wifi: mac80211: remove DEAUTH_NEED_MGD_TX_PREP

This flag is annoying because it puts a lot of logic into mac80211
that could just as well be in the driver (only iwlmvm uses it) and
the implementation is also broken for MLO.

Remove the flag in favour of calling drv_mgd_prepare_tx() without
any conditions even for the deauth-while-assoc case. The drivers
that implement it can take the appropriate actions, which for the
only user of DEAUTH_NEED_MGD_TX_PREP (iwlmvm) is a bit more tricky
than the implementation in mac80211 is anyway, and all others have
no need and can just exit if info->was_assoc is set.

Reviewed-by: Miriam Rachel Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240627132527.94924bcc9c9e.I328a219e45f2e2724cd52e75bb9feee3bf21a463@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c  | 12 +++++-----
 .../net/wireless/intel/iwlwifi/mvm/mld-mac80211.c  |  2 ++
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h       |  4 ++++
 .../net/wireless/intel/iwlwifi/mvm/time-event.c    |  2 ++
 include/net/mac80211.h                             | 22 +++++-------------
 net/mac80211/debugfs.c                             |  1 -
 net/mac80211/main.c                                |  3 ---
 net/mac80211/mlme.c                                | 26 +++-------------------
 8 files changed, 23 insertions(+), 49 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 90cf0ec69682..60bfe42d5386 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -383,12 +383,6 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 	if (!mvm->mld_api_is_used)
 		ieee80211_hw_set(hw, TIMING_BEACON_ONLY);
 
-	/* We should probably have this, but mac80211
-	 * currently doesn't support it for MLO.
-	 */
-	if (!(hw->wiphy->flags & WIPHY_FLAG_SUPPORTS_MLO))
-		ieee80211_hw_set(hw, DEAUTH_NEED_MGD_TX_PREP);
-
 	/*
 	 * On older devices, enabling TX A-MSDU occasionally leads to
 	 * something getting messed up, the command read from the FIFO
@@ -2853,6 +2847,8 @@ static void iwl_mvm_bss_info_changed_station(struct iwl_mvm *mvm,
 
 	if (changes & BSS_CHANGED_ASSOC) {
 		if (vif->cfg.assoc) {
+			mvmvif->session_prot_connection_loss = false;
+
 			/* clear statistics to get clean beacon counter */
 			iwl_mvm_request_statistics(mvm, true);
 			for_each_mvm_vif_valid_link(mvmvif, i)
@@ -4268,8 +4264,12 @@ void iwl_mvm_mac_mgd_prepare_tx(struct ieee80211_hw *hw,
 				struct ieee80211_vif *vif,
 				struct ieee80211_prep_tx_info *info)
 {
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 
+	if (info->was_assoc && !mvmvif->session_prot_connection_loss)
+		return;
+
 	guard(mvm)(mvm);
 	iwl_mvm_protect_assoc(mvm, vif, info->duration, info->link_id);
 }
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
index ebf313e161f4..3c99396ad369 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mld-mac80211.c
@@ -873,6 +873,8 @@ static void iwl_mvm_mld_vif_cfg_changed_station(struct iwl_mvm *mvm,
 
 	if (changes & BSS_CHANGED_ASSOC) {
 		if (vif->cfg.assoc) {
+			mvmvif->session_prot_connection_loss = false;
+
 			/* clear statistics to get clean beacon counter */
 			iwl_mvm_request_statistics(mvm, true);
 			iwl_mvm_sf_update(mvm, vif, false);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 362973fdeac0..393ce424c196 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -450,6 +450,9 @@ struct iwl_mvm_esr_exit {
  * @unblock_esr_tpt_wk: work for unblocking EMLSR when tpt is high enough.
  * @roc_activity: currently running ROC activity for this vif (or
  *	ROC_NUM_ACTIVITIES if no activity is running).
+ * @session_prot_connection_loss: the connection was lost due to session
+ *	protection ending without receiving a beacon, so we need to now
+ *	protect the deauth separately
  */
 struct iwl_mvm_vif {
 	struct iwl_mvm *mvm;
@@ -463,6 +466,7 @@ struct iwl_mvm_vif {
 	bool pm_enabled;
 	bool monitor_active;
 	bool esr_active;
+	bool session_prot_connection_loss;
 
 	u8 low_latency: 6;
 	u8 low_latency_actual: 1;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
index 2773dfa8baa9..77b0cae8566f 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
@@ -222,6 +222,8 @@ static bool iwl_mvm_te_check_disconnect(struct iwl_mvm *mvm,
 		iwl_dbg_tlv_time_point(&mvm->fwrt,
 				       IWL_FW_INI_TIME_POINT_ASSOC_FAILED,
 				       NULL);
+
+		mvmvif->session_prot_connection_loss = true;
 	}
 
 	iwl_mvm_connection_loss(mvm, vif, errmsg);
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index 9c96e8ae9ef7..bd0f8aefa797 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -2767,14 +2767,6 @@ struct ieee80211_txq {
  * @IEEE80211_HW_SUPPORTS_TDLS_BUFFER_STA: Hardware supports buffer STA on
  *	TDLS links.
  *
- * @IEEE80211_HW_DEAUTH_NEED_MGD_TX_PREP: The driver requires the
- *	mgd_prepare_tx() callback to be called before transmission of a
- *	deauthentication frame in case the association was completed but no
- *	beacon was heard. This is required in multi-channel scenarios, where the
- *	virtual interface might not be given air time for the transmission of
- *	the frame, as it is not synced with the AP/P2P GO yet, and thus the
- *	deauthentication frame might not be transmitted.
- *
  * @IEEE80211_HW_DOESNT_SUPPORT_QOS_NDP: The driver (or firmware) doesn't
  *	support QoS NDP for AP probing - that's most likely a driver bug.
  *
@@ -2874,7 +2866,6 @@ enum ieee80211_hw_flags {
 	IEEE80211_HW_REPORTS_LOW_ACK,
 	IEEE80211_HW_SUPPORTS_TX_FRAG,
 	IEEE80211_HW_SUPPORTS_TDLS_BUFFER_STA,
-	IEEE80211_HW_DEAUTH_NEED_MGD_TX_PREP,
 	IEEE80211_HW_DOESNT_SUPPORT_QOS_NDP,
 	IEEE80211_HW_BUFF_MMPDU_TXQ,
 	IEEE80211_HW_SUPPORTS_VHT_EXT_NSS_BW,
@@ -3787,13 +3778,15 @@ enum ieee80211_reconfig_type {
  * @success: whether the frame exchange was successful, only
  *	used with the mgd_complete_tx() method, and then only
  *	valid for auth and (re)assoc.
+ * @was_assoc: set if this call is due to deauth/disassoc
+ *	while just having been associated
  * @link_id: the link id on which the frame will be TX'ed.
  *	Only used with the mgd_prepare_tx() method.
  */
 struct ieee80211_prep_tx_info {
 	u16 duration;
 	u16 subtype;
-	u8 success:1;
+	u8 success:1, was_assoc:1;
 	int link_id;
 };
 
@@ -4242,12 +4235,9 @@ struct ieee80211_prep_tx_info {
  *	yet it need not necessarily be given airtime, in particular since any
  *	transmission to a P2P GO needs to be synchronized against the GO's
  *	powersave state. mac80211 will call this function before transmitting a
- *	management frame prior to having successfully associated to allow the
- *	driver to give it channel time for the transmission, to get a response
- *	and to be able to synchronize with the GO.
- *	For drivers that set %IEEE80211_HW_DEAUTH_NEED_MGD_TX_PREP, mac80211
- *	would also call this function before transmitting a deauthentication
- *	frame in case that no beacon was heard from the AP/P2P GO.
+ *	management frame prior to transmitting that frame to allow the driver
+ *	to give it channel time for the transmission, to get a response and be
+ *	able to synchronize with the GO.
  *	The callback will be called before each transmission and upon return
  *	mac80211 will transmit the frame right away.
  *	Additional information is passed in the &struct ieee80211_prep_tx_info
diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c
index 98310188f330..02b5476a4376 100644
--- a/net/mac80211/debugfs.c
+++ b/net/mac80211/debugfs.c
@@ -483,7 +483,6 @@ static const char *hw_flag_names[] = {
 	FLAG(REPORTS_LOW_ACK),
 	FLAG(SUPPORTS_TX_FRAG),
 	FLAG(SUPPORTS_TDLS_BUFFER_STA),
-	FLAG(DEAUTH_NEED_MGD_TX_PREP),
 	FLAG(DOESNT_SUPPORT_QOS_NDP),
 	FLAG(BUFF_MMPDU_TXQ),
 	FLAG(SUPPORTS_VHT_EXT_NSS_BW),
diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index a9aefc83d30a..7578ea56c12f 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -1161,9 +1161,6 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 
 		if (WARN_ON(!ieee80211_hw_check(hw, AP_LINK_PS)))
 			return -EINVAL;
-
-		if (WARN_ON(ieee80211_hw_check(hw, DEAUTH_NEED_MGD_TX_PREP)))
-			return -EINVAL;
 	}
 
 #ifdef CONFIG_PM
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index 80a10cd49565..4779a18ab75d 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -3521,6 +3521,8 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata,
 	u64 changed = 0;
 	struct ieee80211_prep_tx_info info = {
 		.subtype = stype,
+		.was_assoc = true,
+		.link_id = ffs(sdata->vif.active_links) - 1,
 	};
 
 	lockdep_assert_wiphy(local->hw.wiphy);
@@ -3569,29 +3571,7 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata,
 
 	/* deauthenticate/disassociate now */
 	if (tx || frame_buf) {
-		/*
-		 * In multi channel scenarios guarantee that the virtual
-		 * interface is granted immediate airtime to transmit the
-		 * deauthentication frame by calling mgd_prepare_tx, if the
-		 * driver requested so.
-		 */
-		if (ieee80211_hw_check(&local->hw, DEAUTH_NEED_MGD_TX_PREP)) {
-			for (link_id = 0; link_id < ARRAY_SIZE(sdata->link);
-			     link_id++) {
-				struct ieee80211_link_data *link;
-
-				link = sdata_dereference(sdata->link[link_id],
-							 sdata);
-				if (!link)
-					continue;
-				if (link->u.mgd.have_beacon)
-					break;
-			}
-			if (link_id == IEEE80211_MLD_MAX_NUM_LINKS) {
-				info.link_id = ffs(sdata->vif.active_links) - 1;
-				drv_mgd_prepare_tx(sdata->local, sdata, &info);
-			}
-		}
+		drv_mgd_prepare_tx(sdata->local, sdata, &info);
 
 		ieee80211_send_deauth_disassoc(sdata, sdata->vif.cfg.ap_addr,
 					       sdata->vif.cfg.ap_addr, stype,
-- 
cgit v1.2.3-58-ga151


From 69540b7987ef05d1dff36981d4cc7c31e9716b36 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Thu, 27 Jun 2024 17:08:48 +0300
Subject: ethtool: Add ethtool operation to write to a transceiver module
 EEPROM

Ethtool can already retrieve information from a transceiver module
EEPROM by invoking the ethtool_ops::get_module_eeprom_by_page operation.
Add a corresponding operation that allows ethtool to write to a
transceiver module EEPROM.

The new write operation is purely an in-kernel API and is not exposed to
user space.

The purpose of this operation is not to enable arbitrary read / write
access, but to allow the kernel to write to specific addresses as part
of transceiver module firmware flashing. In the future, more
functionality can be implemented on top of these read / write
operations.

Adjust the comments of the 'ethtool_module_eeprom' structure as it is
no longer used only for read access.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/ethtool.h | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 959196af7f5a..c7f6f2bc9cac 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -506,17 +506,16 @@ struct ethtool_ts_stats {
 #define ETH_MODULE_MAX_I2C_ADDRESS	0x7f
 
 /**
- * struct ethtool_module_eeprom - EEPROM dump from specified page
- * @offset: Offset within the specified EEPROM page to begin read, in bytes.
- * @length: Number of bytes to read.
- * @page: Page number to read from.
- * @bank: Page bank number to read from, if applicable by EEPROM spec.
+ * struct ethtool_module_eeprom - plug-in module EEPROM read / write parameters
+ * @offset: When @offset is 0-127, it is used as an address to the Lower Memory
+ *	(@page must be 0). Otherwise, it is used as an address to the
+ *	Upper Memory.
+ * @length: Number of bytes to read / write.
+ * @page: Page number.
+ * @bank: Bank number, if supported by EEPROM spec.
  * @i2c_address: I2C address of a page. Value less than 0x7f expected. Most
  *	EEPROMs use 0x50 or 0x51.
  * @data: Pointer to buffer with EEPROM data of @length size.
- *
- * This can be used to manage pages during EEPROM dump in ethtool and pass
- * required information to the driver.
  */
 struct ethtool_module_eeprom {
 	u32	offset;
@@ -824,6 +823,8 @@ struct ethtool_rxfh_param {
  * @get_module_eeprom_by_page: Get a region of plug-in module EEPROM data from
  *	specified page. Returns a negative error code or the amount of bytes
  *	read.
+ * @set_module_eeprom_by_page: Write to a region of plug-in module EEPROM,
+ *	from kernel space only. Returns a negative error code or zero.
  * @get_eth_phy_stats: Query some of the IEEE 802.3 PHY statistics.
  * @get_eth_mac_stats: Query some of the IEEE 802.3 MAC statistics.
  * @get_eth_ctrl_stats: Query some of the IEEE 802.3 MAC Ctrl statistics.
@@ -958,6 +959,9 @@ struct ethtool_ops {
 	int	(*get_module_eeprom_by_page)(struct net_device *dev,
 					     const struct ethtool_module_eeprom *page,
 					     struct netlink_ext_ack *extack);
+	int	(*set_module_eeprom_by_page)(struct net_device *dev,
+					     const struct ethtool_module_eeprom *page,
+					     struct netlink_ext_ack *extack);
 	void	(*get_eth_phy_stats)(struct net_device *dev,
 				     struct ethtool_eth_phy_stats *phy_stats);
 	void	(*get_eth_mac_stats)(struct net_device *dev,
-- 
cgit v1.2.3-58-ga151


From 1983a8007032321327a2c31b40d45dab544fcdf9 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Thu, 27 Jun 2024 17:08:49 +0300
Subject: mlxsw: Implement ethtool operation to write to a transceiver module
 EEPROM

Implement the ethtool_ops::set_module_eeprom_by_page operation to allow
ethtool to write to a transceiver module EEPROM, in a similar fashion to
the ethtool_ops::get_module_eeprom_by_page operation.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/mellanox/mlxsw/core_env.c     | 57 ++++++++++++++++++++++
 drivers/net/ethernet/mellanox/mlxsw/core_env.h     |  6 +++
 drivers/net/ethernet/mellanox/mlxsw/minimal.c      | 15 ++++++
 .../net/ethernet/mellanox/mlxsw/spectrum_ethtool.c | 15 ++++++
 4 files changed, 93 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_env.c b/drivers/net/ethernet/mellanox/mlxsw/core_env.c
index 6c06b0592760..294e758f1067 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_env.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_env.c
@@ -513,6 +513,63 @@ mlxsw_env_get_module_eeprom_by_page(struct mlxsw_core *mlxsw_core,
 }
 EXPORT_SYMBOL(mlxsw_env_get_module_eeprom_by_page);
 
+int
+mlxsw_env_set_module_eeprom_by_page(struct mlxsw_core *mlxsw_core,
+				    u8 slot_index, u8 module,
+				    const struct ethtool_module_eeprom *page,
+				    struct netlink_ext_ack *extack)
+{
+	struct mlxsw_env *mlxsw_env = mlxsw_core_env(mlxsw_core);
+	u32 bytes_written = 0;
+	u16 device_addr;
+	int err;
+
+	if (!mlxsw_env_linecard_is_active(mlxsw_env, slot_index)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Cannot write to EEPROM of a module on an inactive line card");
+		return -EIO;
+	}
+
+	err = mlxsw_env_validate_module_type(mlxsw_core, slot_index, module);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "EEPROM is not equipped on port module type");
+		return err;
+	}
+
+	device_addr = page->offset;
+
+	while (bytes_written < page->length) {
+		char mcia_pl[MLXSW_REG_MCIA_LEN];
+		char eeprom_tmp[128] = {};
+		u8 size;
+
+		size = min_t(u8, page->length - bytes_written,
+			     mlxsw_env->max_eeprom_len);
+
+		mlxsw_reg_mcia_pack(mcia_pl, slot_index, module, page->page,
+				    device_addr + bytes_written, size,
+				    page->i2c_address);
+		mlxsw_reg_mcia_bank_number_set(mcia_pl, page->bank);
+		memcpy(eeprom_tmp, page->data + bytes_written, size);
+		mlxsw_reg_mcia_eeprom_memcpy_to(mcia_pl, eeprom_tmp);
+
+		err = mlxsw_reg_write(mlxsw_core, MLXSW_REG(mcia), mcia_pl);
+		if (err) {
+			NL_SET_ERR_MSG_MOD(extack, "Failed to access module's EEPROM");
+			return err;
+		}
+
+		err = mlxsw_env_mcia_status_process(mcia_pl, extack);
+		if (err)
+			return err;
+
+		bytes_written += size;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(mlxsw_env_set_module_eeprom_by_page);
+
 static int mlxsw_env_module_reset(struct mlxsw_core *mlxsw_core, u8 slot_index,
 				  u8 module)
 {
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_env.h b/drivers/net/ethernet/mellanox/mlxsw/core_env.h
index a197e3ae069c..e4ff17869400 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_env.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_env.h
@@ -28,6 +28,12 @@ mlxsw_env_get_module_eeprom_by_page(struct mlxsw_core *mlxsw_core,
 				    const struct ethtool_module_eeprom *page,
 				    struct netlink_ext_ack *extack);
 
+int
+mlxsw_env_set_module_eeprom_by_page(struct mlxsw_core *mlxsw_core,
+				    u8 slot_index, u8 module,
+				    const struct ethtool_module_eeprom *page,
+				    struct netlink_ext_ack *extack);
+
 int mlxsw_env_reset_module(struct net_device *netdev,
 			   struct mlxsw_core *mlxsw_core, u8 slot_index,
 			   u8 module, u32 *flags);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/minimal.c b/drivers/net/ethernet/mellanox/mlxsw/minimal.c
index 431accdc6213..828c65036a4c 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/minimal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/minimal.c
@@ -140,6 +140,20 @@ mlxsw_m_get_module_eeprom_by_page(struct net_device *netdev,
 						   page, extack);
 }
 
+static int
+mlxsw_m_set_module_eeprom_by_page(struct net_device *netdev,
+				  const struct ethtool_module_eeprom *page,
+				  struct netlink_ext_ack *extack)
+{
+	struct mlxsw_m_port *mlxsw_m_port = netdev_priv(netdev);
+	struct mlxsw_core *core = mlxsw_m_port->mlxsw_m->core;
+
+	return mlxsw_env_set_module_eeprom_by_page(core,
+						   mlxsw_m_port->slot_index,
+						   mlxsw_m_port->module,
+						   page, extack);
+}
+
 static int mlxsw_m_reset(struct net_device *netdev, u32 *flags)
 {
 	struct mlxsw_m_port *mlxsw_m_port = netdev_priv(netdev);
@@ -181,6 +195,7 @@ static const struct ethtool_ops mlxsw_m_port_ethtool_ops = {
 	.get_module_info	= mlxsw_m_get_module_info,
 	.get_module_eeprom	= mlxsw_m_get_module_eeprom,
 	.get_module_eeprom_by_page = mlxsw_m_get_module_eeprom_by_page,
+	.set_module_eeprom_by_page = mlxsw_m_set_module_eeprom_by_page,
 	.reset			= mlxsw_m_reset,
 	.get_module_power_mode	= mlxsw_m_get_module_power_mode,
 	.set_module_power_mode	= mlxsw_m_set_module_power_mode,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
index a755b0a901d3..c79da1411d33 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
@@ -1067,6 +1067,20 @@ mlxsw_sp_get_module_eeprom_by_page(struct net_device *dev,
 						   module, page, extack);
 }
 
+static int
+mlxsw_sp_set_module_eeprom_by_page(struct net_device *dev,
+				   const struct ethtool_module_eeprom *page,
+				   struct netlink_ext_ack *extack)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	u8 slot_index = mlxsw_sp_port->mapping.slot_index;
+	u8 module = mlxsw_sp_port->mapping.module;
+
+	return mlxsw_env_set_module_eeprom_by_page(mlxsw_sp->core, slot_index,
+						   module, page, extack);
+}
+
 static int
 mlxsw_sp_get_ts_info(struct net_device *netdev, struct ethtool_ts_info *info)
 {
@@ -1256,6 +1270,7 @@ const struct ethtool_ops mlxsw_sp_port_ethtool_ops = {
 	.get_module_info		= mlxsw_sp_get_module_info,
 	.get_module_eeprom		= mlxsw_sp_get_module_eeprom,
 	.get_module_eeprom_by_page	= mlxsw_sp_get_module_eeprom_by_page,
+	.set_module_eeprom_by_page	= mlxsw_sp_set_module_eeprom_by_page,
 	.get_ts_info			= mlxsw_sp_get_ts_info,
 	.get_eth_phy_stats		= mlxsw_sp_get_eth_phy_stats,
 	.get_eth_mac_stats		= mlxsw_sp_get_eth_mac_stats,
-- 
cgit v1.2.3-58-ga151


From 46fb3ba95b93d1887e6dfa02a535e0526062de95 Mon Sep 17 00:00:00 2001
From: Danielle Ratson <danieller@nvidia.com>
Date: Thu, 27 Jun 2024 17:08:50 +0300
Subject: ethtool: Add an interface for flashing transceiver modules' firmware

CMIS compliant modules such as QSFP-DD might be running a firmware that
can be updated in a vendor-neutral way by exchanging messages between
the host and the module as described in section 7.3.1 of revision 5.2 of
the CMIS standard.

Add a pair of new ethtool messages that allow:

* User space to trigger firmware update of transceiver modules

* The kernel to notify user space about the progress of the process

The user interface is designed to be asynchronous in order to avoid
RTNL being held for too long and to allow several modules to be
updated simultaneously. The interface is designed with CMIS compliant
modules in mind, but kept generic enough to accommodate future use
cases, if these arise.

Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/netlink/specs/ethtool.yaml     | 55 ++++++++++++++++++++++
 Documentation/networking/ethtool-netlink.rst | 70 ++++++++++++++++++++++++++++
 include/uapi/linux/ethtool.h                 | 18 +++++++
 include/uapi/linux/ethtool_netlink.h         | 19 ++++++++
 tools/net/ynl/Makefile.deps                  |  3 +-
 5 files changed, 164 insertions(+), 1 deletion(-)

diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml
index 683f5c3f30ad..949e2722505d 100644
--- a/Documentation/netlink/specs/ethtool.yaml
+++ b/Documentation/netlink/specs/ethtool.yaml
@@ -20,6 +20,10 @@ definitions:
     name: header-flags
     type: flags
     entries: [ compact-bitsets, omit-reply, stats ]
+  -
+    name: module-fw-flash-status
+    type: enum
+    entries: [ started, in_progress, completed, error ]
 
 attribute-sets:
   -
@@ -1004,6 +1008,32 @@ attribute-sets:
       -
         name: burst-tmr
         type: u32
+  -
+    name: module-fw-flash
+    attributes:
+      -
+        name: header
+        type: nest
+        nested-attributes: header
+      -
+        name: file-name
+        type: string
+      -
+        name: password
+        type: u32
+      -
+        name: status
+        type: u32
+        enum: module-fw-flash-status
+      -
+        name: status-msg
+        type: string
+      -
+        name: done
+        type: uint
+      -
+        name: total
+        type: uint
 
 operations:
   enum-model: directional
@@ -1764,3 +1794,28 @@ operations:
       name: mm-ntf
       doc: Notification for change in MAC Merge configuration.
       notify: mm-get
+    -
+      name: module-fw-flash-act
+      doc: Flash transceiver module firmware.
+
+      attribute-set: module-fw-flash
+
+      do:
+        request:
+          attributes:
+            - header
+            - file-name
+            - password
+    -
+      name: module-fw-flash-ntf
+      doc: Notification for firmware flashing progress and status.
+
+      attribute-set: module-fw-flash
+
+      event:
+        attributes:
+          - header
+          - status
+          - status-msg
+          - done
+          - total
diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 7ec08e903bab..bfe2eda8580d 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -228,6 +228,7 @@ Userspace to kernel:
   ``ETHTOOL_MSG_PLCA_GET_STATUS``       get PLCA RS status
   ``ETHTOOL_MSG_MM_GET``                get MAC merge layer state
   ``ETHTOOL_MSG_MM_SET``                set MAC merge layer parameters
+  ``ETHTOOL_MSG_MODULE_FW_FLASH_ACT``   flash transceiver module firmware
   ===================================== =================================
 
 Kernel to userspace:
@@ -274,6 +275,7 @@ Kernel to userspace:
   ``ETHTOOL_MSG_PLCA_GET_STATUS_REPLY``    PLCA RS status
   ``ETHTOOL_MSG_PLCA_NTF``                 PLCA RS parameters
   ``ETHTOOL_MSG_MM_GET_REPLY``             MAC merge layer status
+  ``ETHTOOL_MSG_MODULE_FW_FLASH_NTF``      transceiver module flash updates
   ======================================== =================================
 
 ``GET`` requests are sent by userspace applications to retrieve device
@@ -2041,6 +2043,73 @@ The attributes are propagated to the driver through the following structure:
 .. kernel-doc:: include/linux/ethtool.h
     :identifiers: ethtool_mm_cfg
 
+MODULE_FW_FLASH_ACT
+===================
+
+Flashes transceiver module firmware.
+
+Request contents:
+
+  =======================================  ======  ===========================
+  ``ETHTOOL_A_MODULE_FW_FLASH_HEADER``     nested  request header
+  ``ETHTOOL_A_MODULE_FW_FLASH_FILE_NAME``  string  firmware image file name
+  ``ETHTOOL_A_MODULE_FW_FLASH_PASSWORD``   u32     transceiver module password
+  =======================================  ======  ===========================
+
+The firmware update process consists of three logical steps:
+
+1. Downloading a firmware image to the transceiver module and validating it.
+2. Running the firmware image.
+3. Committing the firmware image so that it is run upon reset.
+
+When flash command is given, those three steps are taken in that order.
+
+This message merely schedules the update process and returns immediately
+without blocking. The process then runs asynchronously.
+Since it can take several minutes to complete, during the update process
+notifications are emitted from the kernel to user space updating it about
+the status and progress.
+
+The ``ETHTOOL_A_MODULE_FW_FLASH_FILE_NAME`` attribute encodes the firmware
+image file name. The firmware image is downloaded to the transceiver module,
+validated, run and committed.
+
+The optional ``ETHTOOL_A_MODULE_FW_FLASH_PASSWORD`` attribute encodes a password
+that might be required as part of the transceiver module firmware update
+process.
+
+The firmware update process can take several minutes to complete. Therefore,
+during the update process notifications are emitted from the kernel to user
+space updating it about the status and progress.
+
+
+
+Notification contents:
+
+ +---------------------------------------------------+--------+----------------+
+ | ``ETHTOOL_A_MODULE_FW_FLASH_HEADER``              | nested | reply header   |
+ +---------------------------------------------------+--------+----------------+
+ | ``ETHTOOL_A_MODULE_FW_FLASH_STATUS``              | u32    | status         |
+ +---------------------------------------------------+--------+----------------+
+ | ``ETHTOOL_A_MODULE_FW_FLASH_STATUS_MSG``          | string | status message |
+ +---------------------------------------------------+--------+----------------+
+ | ``ETHTOOL_A_MODULE_FW_FLASH_DONE``                | uint   | progress       |
+ +---------------------------------------------------+--------+----------------+
+ | ``ETHTOOL_A_MODULE_FW_FLASH_TOTAL``               | uint   | total          |
+ +---------------------------------------------------+--------+----------------+
+
+The ``ETHTOOL_A_MODULE_FW_FLASH_STATUS`` attribute encodes the current status
+of the firmware update process. Possible values are:
+
+.. kernel-doc:: include/uapi/linux/ethtool.h
+    :identifiers: ethtool_module_fw_flash_status
+
+The ``ETHTOOL_A_MODULE_FW_FLASH_STATUS_MSG`` attribute encodes a status message
+string.
+
+The ``ETHTOOL_A_MODULE_FW_FLASH_DONE`` and ``ETHTOOL_A_MODULE_FW_FLASH_TOTAL``
+attributes encode the completed and total amount of work, respectively.
+
 Request translation
 ===================
 
@@ -2147,4 +2216,5 @@ are netlink only.
   n/a                                 ``ETHTOOL_MSG_PLCA_GET_STATUS``
   n/a                                 ``ETHTOOL_MSG_MM_GET``
   n/a                                 ``ETHTOOL_MSG_MM_SET``
+  n/a                                 ``ETHTOOL_MSG_MODULE_FW_FLASH_ACT``
   =================================== =====================================
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 8733a3117902..e011384c915c 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -877,6 +877,24 @@ enum ethtool_mm_verify_status {
 	ETHTOOL_MM_VERIFY_STATUS_DISABLED,
 };
 
+/**
+ * enum ethtool_module_fw_flash_status - plug-in module firmware flashing status
+ * @ETHTOOL_MODULE_FW_FLASH_STATUS_STARTED: The firmware flashing process has
+ *	started.
+ * @ETHTOOL_MODULE_FW_FLASH_STATUS_IN_PROGRESS: The firmware flashing process
+ *	is in progress.
+ * @ETHTOOL_MODULE_FW_FLASH_STATUS_COMPLETED: The firmware flashing process was
+ *	completed successfully.
+ * @ETHTOOL_MODULE_FW_FLASH_STATUS_ERROR: The firmware flashing process was
+ *	stopped due to an error.
+ */
+enum ethtool_module_fw_flash_status {
+	ETHTOOL_MODULE_FW_FLASH_STATUS_STARTED = 1,
+	ETHTOOL_MODULE_FW_FLASH_STATUS_IN_PROGRESS,
+	ETHTOOL_MODULE_FW_FLASH_STATUS_COMPLETED,
+	ETHTOOL_MODULE_FW_FLASH_STATUS_ERROR,
+};
+
 /**
  * struct ethtool_gstrings - string set for data tagging
  * @cmd: Command number = %ETHTOOL_GSTRINGS
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index d15856c7e001..840dabdc9d88 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -57,6 +57,7 @@ enum {
 	ETHTOOL_MSG_PLCA_GET_STATUS,
 	ETHTOOL_MSG_MM_GET,
 	ETHTOOL_MSG_MM_SET,
+	ETHTOOL_MSG_MODULE_FW_FLASH_ACT,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
@@ -109,6 +110,7 @@ enum {
 	ETHTOOL_MSG_PLCA_NTF,
 	ETHTOOL_MSG_MM_GET_REPLY,
 	ETHTOOL_MSG_MM_NTF,
+	ETHTOOL_MSG_MODULE_FW_FLASH_NTF,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
@@ -1018,6 +1020,23 @@ enum {
 	ETHTOOL_A_MM_MAX = (__ETHTOOL_A_MM_CNT - 1)
 };
 
+/* MODULE_FW_FLASH */
+
+enum {
+	ETHTOOL_A_MODULE_FW_FLASH_UNSPEC,
+	ETHTOOL_A_MODULE_FW_FLASH_HEADER,		/* nest - _A_HEADER_* */
+	ETHTOOL_A_MODULE_FW_FLASH_FILE_NAME,		/* string */
+	ETHTOOL_A_MODULE_FW_FLASH_PASSWORD,		/* u32 */
+	ETHTOOL_A_MODULE_FW_FLASH_STATUS,		/* u32 */
+	ETHTOOL_A_MODULE_FW_FLASH_STATUS_MSG,		/* string */
+	ETHTOOL_A_MODULE_FW_FLASH_DONE,			/* uint */
+	ETHTOOL_A_MODULE_FW_FLASH_TOTAL,		/* uint */
+
+	/* add new constants above here */
+	__ETHTOOL_A_MODULE_FW_FLASH_CNT,
+	ETHTOOL_A_MODULE_FW_FLASH_MAX = (__ETHTOOL_A_MODULE_FW_FLASH_CNT - 1)
+};
+
 /* generic netlink info */
 #define ETHTOOL_GENL_NAME "ethtool"
 #define ETHTOOL_GENL_VERSION 1
diff --git a/tools/net/ynl/Makefile.deps b/tools/net/ynl/Makefile.deps
index f4e8eb79c1b8..dbdca32a1c61 100644
--- a/tools/net/ynl/Makefile.deps
+++ b/tools/net/ynl/Makefile.deps
@@ -16,7 +16,8 @@ get_hdr_inc=-D$(1) -include $(UAPI_PATH)/linux/$(2)
 
 CFLAGS_devlink:=$(call get_hdr_inc,_LINUX_DEVLINK_H_,devlink.h)
 CFLAGS_dpll:=$(call get_hdr_inc,_LINUX_DPLL_H,dpll.h)
-CFLAGS_ethtool:=$(call get_hdr_inc,_LINUX_ETHTOOL_NETLINK_H_,ethtool_netlink.h)
+CFLAGS_ethtool:=$(call get_hdr_inc,_LINUX_ETHTOOL_H,ethtool.h) \
+		$(call get_hdr_inc,_LINUX_ETHTOOL_NETLINK_H_,ethtool_netlink.h)
 CFLAGS_handshake:=$(call get_hdr_inc,_LINUX_HANDSHAKE_H,handshake.h)
 CFLAGS_mptcp_pm:=$(call get_hdr_inc,_LINUX_MPTCP_PM_H,mptcp_pm.h)
 CFLAGS_netdev:=$(call get_hdr_inc,_LINUX_NETDEV_H,netdev.h)
-- 
cgit v1.2.3-58-ga151


From d7d4cfc4c97c7cf49cb2893ef60e8ab59dcac047 Mon Sep 17 00:00:00 2001
From: Danielle Ratson <danieller@nvidia.com>
Date: Thu, 27 Jun 2024 17:08:51 +0300
Subject: ethtool: Add flashing transceiver modules' firmware notifications
 ability

Add progress notifications ability to user space while flashing modules'
firmware by implementing the interface between the user space and the
kernel.

Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/module.c    | 117 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/ethtool/module_fw.h |  31 +++++++++++++
 net/ethtool/netlink.c   |   5 +++
 net/ethtool/netlink.h   |   1 +
 4 files changed, 154 insertions(+)
 create mode 100644 net/ethtool/module_fw.h

diff --git a/net/ethtool/module.c b/net/ethtool/module.c
index ceb575efc290..ba728b4a38a1 100644
--- a/net/ethtool/module.c
+++ b/net/ethtool/module.c
@@ -5,6 +5,7 @@
 #include "netlink.h"
 #include "common.h"
 #include "bitset.h"
+#include "module_fw.h"
 
 struct module_req_info {
 	struct ethnl_req_info base;
@@ -158,3 +159,119 @@ const struct ethnl_request_ops ethnl_module_request_ops = {
 	.set			= ethnl_set_module,
 	.set_ntf_cmd		= ETHTOOL_MSG_MODULE_NTF,
 };
+
+/* MODULE_FW_FLASH_NTF */
+
+static int
+ethnl_module_fw_flash_ntf_put_err(struct sk_buff *skb, char *err_msg,
+				  char *sub_err_msg)
+{
+	int err_msg_len, sub_err_msg_len, total_len;
+	struct nlattr *attr;
+
+	if (!err_msg)
+		return 0;
+
+	err_msg_len = strlen(err_msg);
+	total_len = err_msg_len + 2; /* For period and NUL. */
+
+	if (sub_err_msg) {
+		sub_err_msg_len = strlen(sub_err_msg);
+		total_len += sub_err_msg_len + 2; /* For ", ". */
+	}
+
+	attr = nla_reserve(skb, ETHTOOL_A_MODULE_FW_FLASH_STATUS_MSG,
+			   total_len);
+	if (!attr)
+		return -ENOMEM;
+
+	if (sub_err_msg)
+		sprintf(nla_data(attr), "%s, %s.", err_msg, sub_err_msg);
+	else
+		sprintf(nla_data(attr), "%s.", err_msg);
+
+	return 0;
+}
+
+static void
+ethnl_module_fw_flash_ntf(struct net_device *dev,
+			  enum ethtool_module_fw_flash_status status,
+			  struct ethnl_module_fw_flash_ntf_params *ntf_params,
+			  char *err_msg, char *sub_err_msg,
+			  u64 done, u64 total)
+{
+	struct sk_buff *skb;
+	void *hdr;
+	int ret;
+
+	if (ntf_params->closed_sock)
+		return;
+
+	skb = genlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
+	if (!skb)
+		return;
+
+	hdr = ethnl_unicast_put(skb, ntf_params->portid, ntf_params->seq,
+				ETHTOOL_MSG_MODULE_FW_FLASH_NTF);
+	if (!hdr)
+		goto err_skb;
+
+	ret = ethnl_fill_reply_header(skb, dev,
+				      ETHTOOL_A_MODULE_FW_FLASH_HEADER);
+	if (ret < 0)
+		goto err_skb;
+
+	if (nla_put_u32(skb, ETHTOOL_A_MODULE_FW_FLASH_STATUS, status))
+		goto err_skb;
+
+	ret = ethnl_module_fw_flash_ntf_put_err(skb, err_msg, sub_err_msg);
+	if (ret < 0)
+		goto err_skb;
+
+	if (nla_put_uint(skb, ETHTOOL_A_MODULE_FW_FLASH_DONE, done))
+		goto err_skb;
+
+	if (nla_put_uint(skb, ETHTOOL_A_MODULE_FW_FLASH_TOTAL, total))
+		goto err_skb;
+
+	genlmsg_end(skb, hdr);
+	genlmsg_unicast(dev_net(dev), skb, ntf_params->portid);
+	return;
+
+err_skb:
+	nlmsg_free(skb);
+}
+
+void ethnl_module_fw_flash_ntf_err(struct net_device *dev,
+				   struct ethnl_module_fw_flash_ntf_params *params,
+				   char *err_msg, char *sub_err_msg)
+{
+	ethnl_module_fw_flash_ntf(dev, ETHTOOL_MODULE_FW_FLASH_STATUS_ERROR,
+				  params, err_msg, sub_err_msg, 0, 0);
+}
+
+void
+ethnl_module_fw_flash_ntf_start(struct net_device *dev,
+				struct ethnl_module_fw_flash_ntf_params *params)
+{
+	ethnl_module_fw_flash_ntf(dev, ETHTOOL_MODULE_FW_FLASH_STATUS_STARTED,
+				  params, NULL, NULL, 0, 0);
+}
+
+void
+ethnl_module_fw_flash_ntf_complete(struct net_device *dev,
+				   struct ethnl_module_fw_flash_ntf_params *params)
+{
+	ethnl_module_fw_flash_ntf(dev, ETHTOOL_MODULE_FW_FLASH_STATUS_COMPLETED,
+				  params, NULL, NULL, 0, 0);
+}
+
+void
+ethnl_module_fw_flash_ntf_in_progress(struct net_device *dev,
+				      struct ethnl_module_fw_flash_ntf_params *params,
+				      u64 done, u64 total)
+{
+	ethnl_module_fw_flash_ntf(dev,
+				  ETHTOOL_MODULE_FW_FLASH_STATUS_IN_PROGRESS,
+				  params, NULL, NULL, done, total);
+}
diff --git a/net/ethtool/module_fw.h b/net/ethtool/module_fw.h
new file mode 100644
index 000000000000..ee4a291ac1d4
--- /dev/null
+++ b/net/ethtool/module_fw.h
@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#include <uapi/linux/ethtool.h>
+
+/**
+ * struct ethnl_module_fw_flash_ntf_params - module firmware flashing
+ *						notifications parameters
+ * @portid: Netlink portid of sender.
+ * @seq: Sequence number of sender.
+ * @closed_sock: Indicates whether the socket was closed from user space.
+ */
+struct ethnl_module_fw_flash_ntf_params {
+	u32 portid;
+	u32 seq;
+	bool closed_sock;
+};
+
+void
+ethnl_module_fw_flash_ntf_err(struct net_device *dev,
+			      struct ethnl_module_fw_flash_ntf_params *params,
+			      char *err_msg, char *sub_err_msg);
+void
+ethnl_module_fw_flash_ntf_start(struct net_device *dev,
+				struct ethnl_module_fw_flash_ntf_params *params);
+void
+ethnl_module_fw_flash_ntf_complete(struct net_device *dev,
+				   struct ethnl_module_fw_flash_ntf_params *params);
+void
+ethnl_module_fw_flash_ntf_in_progress(struct net_device *dev,
+				      struct ethnl_module_fw_flash_ntf_params *params,
+				      u64 done, u64 total);
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index bd04f28d5cf4..393ce668fb04 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -239,6 +239,11 @@ void *ethnl_bcastmsg_put(struct sk_buff *skb, u8 cmd)
 			   cmd);
 }
 
+void *ethnl_unicast_put(struct sk_buff *skb, u32 portid, u32 seq, u8 cmd)
+{
+	return genlmsg_put(skb, portid, seq, &ethtool_genl_family, 0, cmd);
+}
+
 int ethnl_multicast(struct sk_buff *skb, struct net_device *dev)
 {
 	return genlmsg_multicast_netns(&ethtool_genl_family, dev_net(dev), skb,
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 9a333a8d04c1..5e6c6a7b7adc 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -21,6 +21,7 @@ struct sk_buff *ethnl_reply_init(size_t payload, struct net_device *dev, u8 cmd,
 				 void **ehdrp);
 void *ethnl_dump_put(struct sk_buff *skb, struct netlink_callback *cb, u8 cmd);
 void *ethnl_bcastmsg_put(struct sk_buff *skb, u8 cmd);
+void *ethnl_unicast_put(struct sk_buff *skb, u32 portid, u32 seq, u8 cmd);
 int ethnl_multicast(struct sk_buff *skb, struct net_device *dev);
 
 /**
-- 
cgit v1.2.3-58-ga151


From 31e0aa99dc02b2b038a270b0670fc8201b69ec8a Mon Sep 17 00:00:00 2001
From: Danielle Ratson <danieller@nvidia.com>
Date: Thu, 27 Jun 2024 17:08:52 +0300
Subject: ethtool: Veto some operations during firmware flashing process

Some operations cannot be performed during the firmware flashing
process.

For example:

- Port must be down during the whole flashing process to avoid packet loss
  while committing reset for example.

- Writing to EEPROM interrupts the flashing process, so operations like
  ethtool dump, module reset, get and set power mode should be vetoed.

- Split port firmware flashing should be vetoed.

In order to veto those scenarios, add a flag in 'struct net_device' that
indicates when a firmware flash is taking place on the module and use it
to prevent interruptions during the process.

Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netdevice.h |  4 +++-
 net/ethtool/eeprom.c      |  6 ++++++
 net/ethtool/ioctl.c       | 12 ++++++++++++
 net/ethtool/netlink.c     | 12 ++++++++++++
 4 files changed, 33 insertions(+), 1 deletion(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index cc18acd3c58b..1e3401093c13 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1990,6 +1990,8 @@ enum netdev_reg_state {
  *
  *	@threaded:	napi threaded mode is enabled
  *
+ *	@module_fw_flash_in_progress:	Module firmware flashing is in progress.
+ *
  *	@net_notifier_list:	List of per-net netdev notifier block
  *				that follow this device when it is moved
  *				to another network namespace.
@@ -2374,7 +2376,7 @@ struct net_device {
 	bool			proto_down;
 	bool			threaded;
 	unsigned		wol_enabled:1;
-
+	unsigned		module_fw_flash_in_progress:1;
 	struct list_head	net_notifier_list;
 
 #if IS_ENABLED(CONFIG_MACSEC)
diff --git a/net/ethtool/eeprom.c b/net/ethtool/eeprom.c
index 6209c3a9c8f7..f36811b3ecf1 100644
--- a/net/ethtool/eeprom.c
+++ b/net/ethtool/eeprom.c
@@ -91,6 +91,12 @@ static int get_module_eeprom_by_page(struct net_device *dev,
 {
 	const struct ethtool_ops *ops = dev->ethtool_ops;
 
+	if (dev->module_fw_flash_in_progress) {
+		NL_SET_ERR_MSG(extack,
+			       "Module firmware flashing is in progress");
+		return -EBUSY;
+	}
+
 	if (dev->sfp_bus)
 		return sfp_get_module_eeprom_by_page(dev->sfp_bus, page_data, extack);
 
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index e645d751a5e8..1cca372c0d80 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -658,6 +658,9 @@ static int ethtool_get_settings(struct net_device *dev, void __user *useraddr)
 	if (!dev->ethtool_ops->get_link_ksettings)
 		return -EOPNOTSUPP;
 
+	if (dev->module_fw_flash_in_progress)
+		return -EBUSY;
+
 	memset(&link_ksettings, 0, sizeof(link_ksettings));
 	err = dev->ethtool_ops->get_link_ksettings(dev, &link_ksettings);
 	if (err < 0)
@@ -1449,6 +1452,9 @@ static int ethtool_reset(struct net_device *dev, char __user *useraddr)
 	if (!dev->ethtool_ops->reset)
 		return -EOPNOTSUPP;
 
+	if (dev->module_fw_flash_in_progress)
+		return -EBUSY;
+
 	if (copy_from_user(&reset, useraddr, sizeof(reset)))
 		return -EFAULT;
 
@@ -2462,6 +2468,9 @@ int ethtool_get_module_info_call(struct net_device *dev,
 	const struct ethtool_ops *ops = dev->ethtool_ops;
 	struct phy_device *phydev = dev->phydev;
 
+	if (dev->module_fw_flash_in_progress)
+		return -EBUSY;
+
 	if (dev->sfp_bus)
 		return sfp_get_module_info(dev->sfp_bus, modinfo);
 
@@ -2499,6 +2508,9 @@ int ethtool_get_module_eeprom_call(struct net_device *dev,
 	const struct ethtool_ops *ops = dev->ethtool_ops;
 	struct phy_device *phydev = dev->phydev;
 
+	if (dev->module_fw_flash_in_progress)
+		return -EBUSY;
+
 	if (dev->sfp_bus)
 		return sfp_get_module_eeprom(dev->sfp_bus, ee, data);
 
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 393ce668fb04..a5907bbde427 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -765,10 +765,22 @@ static void ethnl_notify_features(struct netdev_notifier_info *info)
 static int ethnl_netdev_event(struct notifier_block *this, unsigned long event,
 			      void *ptr)
 {
+	struct netdev_notifier_info *info = ptr;
+	struct netlink_ext_ack *extack;
+	struct net_device *dev;
+
+	dev = netdev_notifier_info_to_dev(info);
+	extack = netdev_notifier_info_to_extack(info);
+
 	switch (event) {
 	case NETDEV_FEAT_CHANGE:
 		ethnl_notify_features(ptr);
 		break;
+	case NETDEV_PRE_UP:
+		if (dev->module_fw_flash_in_progress) {
+			NL_SET_ERR_MSG(extack, "Can't set port up while flashing module firmware");
+			return NOTIFY_BAD;
+		}
 	}
 
 	return NOTIFY_DONE;
-- 
cgit v1.2.3-58-ga151


From e4f91936993ce4d0954688ec3cb80b3471b9eda5 Mon Sep 17 00:00:00 2001
From: Danielle Ratson <danieller@nvidia.com>
Date: Thu, 27 Jun 2024 17:08:53 +0300
Subject: net: sfp: Add more extended compliance codes

SFF-8024 is used to define various constants re-used in several SFF
SFP-related specifications.

Add SFF-8024 extended compliance code definitions for CMIS compliant
modules and use them in the next patch to determine the firmware flashing
work.

Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/sfp.h | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/include/linux/sfp.h b/include/linux/sfp.h
index a45da7eef9a2..b14be59550e3 100644
--- a/include/linux/sfp.h
+++ b/include/linux/sfp.h
@@ -284,6 +284,12 @@ enum {
 	SFF8024_ID_QSFP_8438		= 0x0c,
 	SFF8024_ID_QSFP_8436_8636	= 0x0d,
 	SFF8024_ID_QSFP28_8636		= 0x11,
+	SFF8024_ID_QSFP_DD		= 0x18,
+	SFF8024_ID_OSFP			= 0x19,
+	SFF8024_ID_DSFP			= 0x1B,
+	SFF8024_ID_QSFP_PLUS_CMIS	= 0x1E,
+	SFF8024_ID_SFP_DD_CMIS		= 0x1F,
+	SFF8024_ID_SFP_PLUS_CMIS	= 0x20,
 
 	SFF8024_ENCODING_UNSPEC		= 0x00,
 	SFF8024_ENCODING_8B10B		= 0x01,
-- 
cgit v1.2.3-58-ga151


From a39c84d796254e6b1662ca0c46dbc313379e9291 Mon Sep 17 00:00:00 2001
From: Danielle Ratson <danieller@nvidia.com>
Date: Thu, 27 Jun 2024 17:08:54 +0300
Subject: ethtool: cmis_cdb: Add a layer for supporting CDB commands

CDB (Command Data Block Message Communication) reads and writes are
performed on memory map pages 9Fh-AFh according to the CMIS standard,
section 8.20 of revision 5.2.
Page 9Fh is used to specify the CDB command to be executed and also
provides an area for a local payload (LPL).

According to the CMIS standard, the firmware update process is done using
a CDB commands sequence that will be implemented in the next patch.

The kernel interface that will implement the firmware update using CDB
command will include 2 layers that will be added under ethtool:

* The upper layer that will be triggered from the module layer, is
  cmis_fw_update.
* The lower one is cmis_cdb.

In the future there might be more operations to implement using CDB
commands. Therefore, the idea is to keep the CDB interface clean and the
cmis_fw_update specific to the CDB commands handling it.

These two layers will communicate using the API the consists of three
functions:

- struct ethtool_cmis_cdb *
  ethtool_cmis_cdb_init(struct net_device *dev,
			struct ethtool_module_fw_flash_params *params);
- void ethtool_cmis_cdb_fini(struct ethtool_cmis_cdb *cdb);
- int ethtool_cmis_cdb_execute_cmd(struct net_device *dev,
				   struct ethtool_cmis_cdb_cmd_args *args);

Add the CDB layer to support initializing, finishing and executing CDB
commands:

* The initialization process will include creating of an ethtool_cmis_cdb
  instance, querying the module CDB support, entering and validating the
  password from user space (CMD 0x0000) and querying the module features
  (CMD 0x0040).

* The finishing API will simply free the ethtool_cmis_cdb instance.

* The executing process will write the CDB command to EEPROM using
  set_module_eeprom_by_page() that was presented earlier, and will
  process the reply from EEPROM.

Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/Makefile    |   2 +-
 net/ethtool/cmis.h      | 117 ++++++++++
 net/ethtool/cmis_cdb.c  | 602 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/ethtool/module_fw.h |  10 +
 4 files changed, 730 insertions(+), 1 deletion(-)
 create mode 100644 net/ethtool/cmis.h
 create mode 100644 net/ethtool/cmis_cdb.c

diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index 504f954a1b28..38806b3ecf83 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -8,4 +8,4 @@ ethtool_nl-y	:= netlink.o bitset.o strset.o linkinfo.o linkmodes.o rss.o \
 		   linkstate.o debug.o wol.o features.o privflags.o rings.o \
 		   channels.o coalesce.o pause.o eee.o tsinfo.o cabletest.o \
 		   tunnels.o fec.o eeprom.o stats.o phc_vclocks.o mm.o \
-		   module.o pse-pd.o plca.o mm.o
+		   module.o cmis_cdb.o pse-pd.o plca.o mm.o
diff --git a/net/ethtool/cmis.h b/net/ethtool/cmis.h
new file mode 100644
index 000000000000..295f5d0df915
--- /dev/null
+++ b/net/ethtool/cmis.h
@@ -0,0 +1,117 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#define ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH		120
+#define ETHTOOL_CMIS_CDB_CMD_PAGE			0x9F
+#define ETHTOOL_CMIS_CDB_PAGE_I2C_ADDR			0x50
+
+/**
+ * struct ethtool_cmis_cdb - CDB commands parameters
+ * @cmis_rev: CMIS revision major.
+ * @read_write_len_ext: Allowable additional number of byte octets to the LPL
+ *			in a READ or a WRITE CDB commands.
+ * @max_completion_time:  Maximum CDB command completion time in msec.
+ */
+struct ethtool_cmis_cdb {
+	u8	cmis_rev;
+	u8      read_write_len_ext;
+	u16     max_completion_time;
+};
+
+enum ethtool_cmis_cdb_cmd_id {
+	ETHTOOL_CMIS_CDB_CMD_QUERY_STATUS		= 0x0000,
+	ETHTOOL_CMIS_CDB_CMD_MODULE_FEATURES		= 0x0040,
+};
+
+/**
+ * struct ethtool_cmis_cdb_request - CDB commands request fields as decribed in
+ *				the CMIS standard
+ * @id: Command ID.
+ * @epl_len: EPL memory length.
+ * @lpl_len: LPL memory length.
+ * @chk_code: Check code for the previous field and the payload.
+ * @resv1: Added to match the CMIS standard request continuity.
+ * @resv2: Added to match the CMIS standard request continuity.
+ * @payload: Payload for the CDB commands.
+ */
+struct ethtool_cmis_cdb_request {
+	__be16 id;
+	struct_group(body,
+		__be16 epl_len;
+		u8 lpl_len;
+		u8 chk_code;
+		u8 resv1;
+		u8 resv2;
+		u8 payload[ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH];
+	);
+};
+
+#define CDB_F_COMPLETION_VALID		BIT(0)
+#define CDB_F_STATUS_VALID		BIT(1)
+
+/**
+ * struct ethtool_cmis_cdb_cmd_args - CDB commands execution arguments
+ * @req: CDB command fields as described in the CMIS standard.
+ * @max_duration: Maximum duration time for command completion in msec.
+ * @read_write_len_ext: Allowable additional number of byte octets to the LPL
+ *			in a READ or a WRITE commands.
+ * @msleep_pre_rpl: Waiting time before checking reply in msec.
+ * @rpl_exp_len: Expected reply length in bytes.
+ * @flags: Validation flags for CDB commands.
+ * @err_msg: Error message to be sent to user space.
+ */
+struct ethtool_cmis_cdb_cmd_args {
+	struct ethtool_cmis_cdb_request req;
+	u16				max_duration;
+	u8				read_write_len_ext;
+	u8				msleep_pre_rpl;
+	u8                              rpl_exp_len;
+	u8				flags;
+	char				*err_msg;
+};
+
+/**
+ * struct ethtool_cmis_cdb_rpl_hdr - CDB commands reply header arguments
+ * @rpl_len: Reply length.
+ * @rpl_chk_code: Reply check code.
+ */
+struct ethtool_cmis_cdb_rpl_hdr {
+	u8 rpl_len;
+	u8 rpl_chk_code;
+};
+
+/**
+ * struct ethtool_cmis_cdb_rpl - CDB commands reply arguments
+ * @hdr: CDB commands reply header arguments.
+ * @payload: Payload for the CDB commands reply.
+ */
+struct ethtool_cmis_cdb_rpl {
+	struct ethtool_cmis_cdb_rpl_hdr hdr;
+	u8 payload[ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH];
+};
+
+u32 ethtool_cmis_get_max_payload_size(u8 num_of_byte_octs);
+
+void ethtool_cmis_cdb_compose_args(struct ethtool_cmis_cdb_cmd_args *args,
+				   enum ethtool_cmis_cdb_cmd_id cmd, u8 *pl,
+				   u8 lpl_len, u16 max_duration,
+				   u8 read_write_len_ext, u16 msleep_pre_rpl,
+				   u8 rpl_exp_len, u8 flags);
+
+void ethtool_cmis_cdb_check_completion_flag(u8 cmis_rev, u8 *flags);
+
+void ethtool_cmis_page_init(struct ethtool_module_eeprom *page_data,
+			    u8 page, u32 offset, u32 length);
+void ethtool_cmis_page_fini(struct ethtool_module_eeprom *page_data);
+
+struct ethtool_cmis_cdb *
+ethtool_cmis_cdb_init(struct net_device *dev,
+		      const struct ethtool_module_fw_flash_params *params,
+		      struct ethnl_module_fw_flash_ntf_params *ntf_params);
+void ethtool_cmis_cdb_fini(struct ethtool_cmis_cdb *cdb);
+
+int ethtool_cmis_wait_for_cond(struct net_device *dev, u8 flags, u8 flag,
+			       u16 max_duration, u32 offset,
+			       bool (*cond_success)(u8), bool (*cond_fail)(u8), u8 *state);
+
+int ethtool_cmis_cdb_execute_cmd(struct net_device *dev,
+				 struct ethtool_cmis_cdb_cmd_args *args);
diff --git a/net/ethtool/cmis_cdb.c b/net/ethtool/cmis_cdb.c
new file mode 100644
index 000000000000..1bb08783b60d
--- /dev/null
+++ b/net/ethtool/cmis_cdb.c
@@ -0,0 +1,602 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/ethtool.h>
+#include <linux/jiffies.h>
+
+#include "common.h"
+#include "module_fw.h"
+#include "cmis.h"
+
+/* For accessing the LPL field on page 9Fh, the allowable length extension is
+ * min(i, 15) byte octets where i specifies the allowable additional number of
+ * byte octets in a READ or a WRITE.
+ */
+u32 ethtool_cmis_get_max_payload_size(u8 num_of_byte_octs)
+{
+	return 8 * (1 + min_t(u8, num_of_byte_octs, 15));
+}
+
+void ethtool_cmis_cdb_compose_args(struct ethtool_cmis_cdb_cmd_args *args,
+				   enum ethtool_cmis_cdb_cmd_id cmd, u8 *pl,
+				   u8 lpl_len, u16 max_duration,
+				   u8 read_write_len_ext, u16 msleep_pre_rpl,
+				   u8 rpl_exp_len, u8 flags)
+{
+	args->req.id = cpu_to_be16(cmd);
+	args->req.lpl_len = lpl_len;
+	if (pl)
+		memcpy(args->req.payload, pl, args->req.lpl_len);
+
+	args->max_duration = max_duration;
+	args->read_write_len_ext =
+		ethtool_cmis_get_max_payload_size(read_write_len_ext);
+	args->msleep_pre_rpl = msleep_pre_rpl;
+	args->rpl_exp_len = rpl_exp_len;
+	args->flags = flags;
+	args->err_msg = NULL;
+}
+
+void ethtool_cmis_page_init(struct ethtool_module_eeprom *page_data,
+			    u8 page, u32 offset, u32 length)
+{
+	page_data->page = page;
+	page_data->offset = offset;
+	page_data->length = length;
+	page_data->i2c_address = ETHTOOL_CMIS_CDB_PAGE_I2C_ADDR;
+}
+
+#define CMIS_REVISION_PAGE	0x00
+#define CMIS_REVISION_OFFSET	0x01
+
+struct cmis_rev_rpl {
+	u8 rev;
+};
+
+static u8 cmis_rev_rpl_major(struct cmis_rev_rpl *rpl)
+{
+	return rpl->rev >> 4;
+}
+
+static int cmis_rev_major_get(struct net_device *dev, u8 *rev_major)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct ethtool_module_eeprom page_data = {0};
+	struct netlink_ext_ack extack = {};
+	struct cmis_rev_rpl rpl = {};
+	int err;
+
+	ethtool_cmis_page_init(&page_data, CMIS_REVISION_PAGE,
+			       CMIS_REVISION_OFFSET, sizeof(rpl));
+	page_data.data = (u8 *)&rpl;
+
+	err = ops->get_module_eeprom_by_page(dev, &page_data, &extack);
+	if (err < 0) {
+		if (extack._msg)
+			netdev_err(dev, "%s\n", extack._msg);
+		return err;
+	}
+
+	*rev_major = cmis_rev_rpl_major(&rpl);
+
+	return 0;
+}
+
+#define CMIS_CDB_ADVERTISEMENT_PAGE	0x01
+#define CMIS_CDB_ADVERTISEMENT_OFFSET	0xA3
+
+/* Based on section 8.4.11 "CDB Messaging Support Advertisement" in CMIS
+ * standard revision 5.2.
+ */
+struct cmis_cdb_advert_rpl {
+	u8	inst_supported;
+	u8	read_write_len_ext;
+	u8	resv1;
+	u8	resv2;
+};
+
+static u8 cmis_cdb_advert_rpl_inst_supported(struct cmis_cdb_advert_rpl *rpl)
+{
+	return rpl->inst_supported >> 6;
+}
+
+static int cmis_cdb_advertisement_get(struct ethtool_cmis_cdb *cdb,
+				      struct net_device *dev)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct ethtool_module_eeprom page_data = {};
+	struct cmis_cdb_advert_rpl rpl = {};
+	struct netlink_ext_ack extack = {};
+	int err;
+
+	ethtool_cmis_page_init(&page_data, CMIS_CDB_ADVERTISEMENT_PAGE,
+			       CMIS_CDB_ADVERTISEMENT_OFFSET, sizeof(rpl));
+	page_data.data = (u8 *)&rpl;
+
+	err = ops->get_module_eeprom_by_page(dev, &page_data, &extack);
+	if (err < 0) {
+		if (extack._msg)
+			netdev_err(dev, "%s\n", extack._msg);
+		return err;
+	}
+
+	if (!cmis_cdb_advert_rpl_inst_supported(&rpl))
+		return -EOPNOTSUPP;
+
+	cdb->read_write_len_ext = rpl.read_write_len_ext;
+
+	return 0;
+}
+
+#define CMIS_PASSWORD_ENTRY_PAGE	0x00
+#define CMIS_PASSWORD_ENTRY_OFFSET	0x7A
+
+struct cmis_password_entry_pl {
+	__be32 password;
+};
+
+/* See section 9.3.1 "CMD 0000h: Query Status" in CMIS standard revision 5.2.
+ * struct cmis_cdb_query_status_pl and struct cmis_cdb_query_status_rpl are
+ * structured layouts of the flat arrays,
+ * struct ethtool_cmis_cdb_request::payload and
+ * struct ethtool_cmis_cdb_rpl::payload respectively.
+ */
+struct cmis_cdb_query_status_pl {
+	u16 response_delay;
+};
+
+struct cmis_cdb_query_status_rpl {
+	u8 length;
+	u8 status;
+};
+
+static int
+cmis_cdb_validate_password(struct ethtool_cmis_cdb *cdb,
+			   struct net_device *dev,
+			   const struct ethtool_module_fw_flash_params *params,
+			   struct ethnl_module_fw_flash_ntf_params *ntf_params)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct cmis_cdb_query_status_pl qs_pl = {0};
+	struct ethtool_module_eeprom page_data = {};
+	struct ethtool_cmis_cdb_cmd_args args = {};
+	struct cmis_password_entry_pl pe_pl = {};
+	struct cmis_cdb_query_status_rpl *rpl;
+	struct netlink_ext_ack extack = {};
+	int err;
+
+	ethtool_cmis_page_init(&page_data, CMIS_PASSWORD_ENTRY_PAGE,
+			       CMIS_PASSWORD_ENTRY_OFFSET, sizeof(pe_pl));
+	page_data.data = (u8 *)&pe_pl;
+
+	pe_pl = *((struct cmis_password_entry_pl *)page_data.data);
+	pe_pl.password = params->password;
+	err = ops->set_module_eeprom_by_page(dev, &page_data, &extack);
+	if (err < 0) {
+		if (extack._msg)
+			netdev_err(dev, "%s\n", extack._msg);
+		return err;
+	}
+
+	ethtool_cmis_cdb_compose_args(&args, ETHTOOL_CMIS_CDB_CMD_QUERY_STATUS,
+				      (u8 *)&qs_pl, sizeof(qs_pl), 0,
+				      cdb->read_write_len_ext, 1000,
+				      sizeof(*rpl),
+				      CDB_F_COMPLETION_VALID | CDB_F_STATUS_VALID);
+
+	err = ethtool_cmis_cdb_execute_cmd(dev, &args);
+	if (err < 0) {
+		ethnl_module_fw_flash_ntf_err(dev, ntf_params,
+					      "Query Status command failed",
+					      args.err_msg);
+		return err;
+	}
+
+	rpl = (struct cmis_cdb_query_status_rpl *)args.req.payload;
+	if (!rpl->length || !rpl->status) {
+		ethnl_module_fw_flash_ntf_err(dev, ntf_params,
+					      "Password was not accepted",
+					      NULL);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/* Some CDB commands asserts the CDB completion flag only from CMIS
+ * revision 5. Therefore, check the relevant validity flag only when
+ * the revision supports it.
+ */
+void ethtool_cmis_cdb_check_completion_flag(u8 cmis_rev, u8 *flags)
+{
+	*flags |= cmis_rev >= 5 ? CDB_F_COMPLETION_VALID : 0;
+}
+
+#define CMIS_CDB_MODULE_FEATURES_RESV_DATA	34
+
+/* See section 9.4.1 "CMD 0040h: Module Features" in CMIS standard revision 5.2.
+ * struct cmis_cdb_module_features_rpl is structured layout of the flat
+ * array, ethtool_cmis_cdb_rpl::payload.
+ */
+struct cmis_cdb_module_features_rpl {
+	u8	resv1[CMIS_CDB_MODULE_FEATURES_RESV_DATA];
+	__be16	max_completion_time;
+};
+
+static u16
+cmis_cdb_module_features_completion_time(struct cmis_cdb_module_features_rpl *rpl)
+{
+	return be16_to_cpu(rpl->max_completion_time);
+}
+
+static int cmis_cdb_module_features_get(struct ethtool_cmis_cdb *cdb,
+					struct net_device *dev,
+					struct ethnl_module_fw_flash_ntf_params *ntf_params)
+{
+	struct ethtool_cmis_cdb_cmd_args args = {};
+	struct cmis_cdb_module_features_rpl *rpl;
+	u8 flags = CDB_F_STATUS_VALID;
+	int err;
+
+	ethtool_cmis_cdb_check_completion_flag(cdb->cmis_rev, &flags);
+	ethtool_cmis_cdb_compose_args(&args,
+				      ETHTOOL_CMIS_CDB_CMD_MODULE_FEATURES,
+				      NULL, 0, 0, cdb->read_write_len_ext,
+				      1000, sizeof(*rpl), flags);
+
+	err = ethtool_cmis_cdb_execute_cmd(dev, &args);
+	if (err < 0) {
+		ethnl_module_fw_flash_ntf_err(dev, ntf_params,
+					      "Module Features command failed",
+					      args.err_msg);
+		return err;
+	}
+
+	rpl = (struct cmis_cdb_module_features_rpl *)args.req.payload;
+	cdb->max_completion_time =
+		cmis_cdb_module_features_completion_time(rpl);
+
+	return 0;
+}
+
+struct ethtool_cmis_cdb *
+ethtool_cmis_cdb_init(struct net_device *dev,
+		      const struct ethtool_module_fw_flash_params *params,
+		      struct ethnl_module_fw_flash_ntf_params *ntf_params)
+{
+	struct ethtool_cmis_cdb *cdb;
+	int err;
+
+	cdb = kzalloc(sizeof(*cdb), GFP_KERNEL);
+	if (!cdb)
+		return ERR_PTR(-ENOMEM);
+
+	err = cmis_rev_major_get(dev, &cdb->cmis_rev);
+	if (err < 0)
+		goto err;
+
+	if (cdb->cmis_rev < 4) {
+		ethnl_module_fw_flash_ntf_err(dev, ntf_params,
+					      "CMIS revision doesn't support module firmware flashing",
+					      NULL);
+		err = -EOPNOTSUPP;
+		goto err;
+	}
+
+	err = cmis_cdb_advertisement_get(cdb, dev);
+	if (err < 0)
+		goto err;
+
+	if (params->password_valid) {
+		err = cmis_cdb_validate_password(cdb, dev, params, ntf_params);
+		if (err < 0)
+			goto err;
+	}
+
+	err = cmis_cdb_module_features_get(cdb, dev, ntf_params);
+	if (err < 0)
+		goto err;
+
+	return cdb;
+
+err:
+	ethtool_cmis_cdb_fini(cdb);
+	return ERR_PTR(err);
+}
+
+void ethtool_cmis_cdb_fini(struct ethtool_cmis_cdb *cdb)
+{
+	kfree(cdb);
+}
+
+static bool is_completed(u8 data)
+{
+	return !!(data & 0x40);
+}
+
+#define CMIS_CDB_STATUS_SUCCESS	0x01
+
+static bool status_success(u8 data)
+{
+	return data == CMIS_CDB_STATUS_SUCCESS;
+}
+
+#define CMIS_CDB_STATUS_FAIL	0x40
+
+static bool status_fail(u8 data)
+{
+	return data & CMIS_CDB_STATUS_FAIL;
+}
+
+struct cmis_wait_for_cond_rpl {
+	u8 state;
+};
+
+static int
+ethtool_cmis_module_poll(struct net_device *dev,
+			 struct cmis_wait_for_cond_rpl *rpl, u32 offset,
+			 bool (*cond_success)(u8), bool (*cond_fail)(u8))
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct ethtool_module_eeprom page_data = {0};
+	struct netlink_ext_ack extack = {};
+	int err;
+
+	ethtool_cmis_page_init(&page_data, 0, offset, sizeof(rpl));
+	page_data.data = (u8 *)rpl;
+
+	err = ops->get_module_eeprom_by_page(dev, &page_data, &extack);
+	if (err < 0) {
+		if (extack._msg)
+			netdev_err_once(dev, "%s\n", extack._msg);
+		return -EBUSY;
+	}
+
+	if ((*cond_success)(rpl->state))
+		return 0;
+
+	if (*cond_fail && (*cond_fail)(rpl->state))
+		return -EIO;
+
+	return -EBUSY;
+}
+
+int ethtool_cmis_wait_for_cond(struct net_device *dev, u8 flags, u8 flag,
+			       u16 max_duration, u32 offset,
+			       bool (*cond_success)(u8), bool (*cond_fail)(u8),
+			       u8 *state)
+{
+	struct cmis_wait_for_cond_rpl rpl = {};
+	unsigned long end;
+	int err;
+
+	if (!(flags & flag))
+		return 0;
+
+	if (max_duration == 0)
+		max_duration = U16_MAX;
+
+	end = jiffies + msecs_to_jiffies(max_duration);
+	do {
+		err = ethtool_cmis_module_poll(dev, &rpl, offset, cond_success,
+					       cond_fail);
+		if (err != -EBUSY)
+			goto out;
+
+		msleep(20);
+	} while (time_before(jiffies, end));
+
+	err = ethtool_cmis_module_poll(dev, &rpl, offset, cond_success,
+				       cond_fail);
+	if (err == -EBUSY)
+		err = -ETIMEDOUT;
+
+out:
+	*state = rpl.state;
+	return err;
+}
+
+#define CMIS_CDB_COMPLETION_FLAG_OFFSET	0x08
+
+static int cmis_cdb_wait_for_completion(struct net_device *dev,
+					struct ethtool_cmis_cdb_cmd_args *args)
+{
+	u8 flag;
+	int err;
+
+	/* Some vendors demand waiting time before checking completion flag
+	 * in some CDB commands.
+	 */
+	msleep(args->msleep_pre_rpl);
+
+	err = ethtool_cmis_wait_for_cond(dev, args->flags,
+					 CDB_F_COMPLETION_VALID,
+					 args->max_duration,
+					 CMIS_CDB_COMPLETION_FLAG_OFFSET,
+					 is_completed, NULL, &flag);
+	if (err < 0)
+		args->err_msg = "Completion Flag did not set on time";
+
+	return err;
+}
+
+#define CMIS_CDB_STATUS_OFFSET	0x25
+
+static void cmis_cdb_status_fail_msg_get(u8 status, char **err_msg)
+{
+	switch (status) {
+	case 0b10000001:
+		*err_msg = "CDB Status is in progress: Busy capturing command";
+		break;
+	case 0b10000010:
+		*err_msg =
+			"CDB Status is in progress: Busy checking/validating command";
+		break;
+	case 0b10000011:
+		*err_msg = "CDB Status is in progress: Busy executing";
+		break;
+	case 0b01000000:
+		*err_msg = "CDB status failed: no specific failure";
+		break;
+	case 0b01000010:
+		*err_msg =
+			"CDB status failed: Parameter range error or parameter not supported";
+		break;
+	case 0b01000101:
+		*err_msg = "CDB status failed: CdbChkCode error";
+		break;
+	default:
+		*err_msg = "Unknown failure reason";
+	}
+};
+
+static int cmis_cdb_wait_for_status(struct net_device *dev,
+				    struct ethtool_cmis_cdb_cmd_args *args)
+{
+	u8 status;
+	int err;
+
+	/* Some vendors demand waiting time before checking status in some
+	 * CDB commands.
+	 */
+	msleep(args->msleep_pre_rpl);
+
+	err = ethtool_cmis_wait_for_cond(dev, args->flags, CDB_F_STATUS_VALID,
+					 args->max_duration,
+					 CMIS_CDB_STATUS_OFFSET,
+					 status_success, status_fail, &status);
+	if (err < 0 && !args->err_msg)
+		cmis_cdb_status_fail_msg_get(status, &args->err_msg);
+
+	return err;
+}
+
+#define CMIS_CDB_REPLY_OFFSET	0x86
+
+static int cmis_cdb_process_reply(struct net_device *dev,
+				  struct ethtool_module_eeprom *page_data,
+				  struct ethtool_cmis_cdb_cmd_args *args)
+{
+	u8 rpl_hdr_len = sizeof(struct ethtool_cmis_cdb_rpl_hdr);
+	u8 rpl_exp_len = args->rpl_exp_len + rpl_hdr_len;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct netlink_ext_ack extack = {};
+	struct ethtool_cmis_cdb_rpl *rpl;
+	int err;
+
+	if (!args->rpl_exp_len)
+		return 0;
+
+	ethtool_cmis_page_init(page_data, ETHTOOL_CMIS_CDB_CMD_PAGE,
+			       CMIS_CDB_REPLY_OFFSET, rpl_exp_len);
+	page_data->data = kmalloc(page_data->length, GFP_KERNEL);
+	if (!page_data->data)
+		return -ENOMEM;
+
+	err = ops->get_module_eeprom_by_page(dev, page_data, &extack);
+	if (err < 0) {
+		if (extack._msg)
+			netdev_err(dev, "%s\n", extack._msg);
+		goto out;
+	}
+
+	rpl = (struct ethtool_cmis_cdb_rpl *)page_data->data;
+	if ((args->rpl_exp_len > rpl->hdr.rpl_len + rpl_hdr_len) ||
+	    !rpl->hdr.rpl_chk_code) {
+		err = -EIO;
+		goto out;
+	}
+
+	args->req.lpl_len = rpl->hdr.rpl_len;
+	memcpy(args->req.payload, rpl->payload, args->req.lpl_len);
+
+out:
+	kfree(page_data->data);
+	return err;
+}
+
+static int
+__ethtool_cmis_cdb_execute_cmd(struct net_device *dev,
+			       struct ethtool_module_eeprom *page_data,
+			       u8 page, u32 offset, u32 length, void *data)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct netlink_ext_ack extack = {};
+	int err;
+
+	ethtool_cmis_page_init(page_data, page, offset, length);
+	page_data->data = kmemdup(data, page_data->length, GFP_KERNEL);
+	if (!page_data->data)
+		return -ENOMEM;
+
+	err = ops->set_module_eeprom_by_page(dev, page_data, &extack);
+	if (err < 0) {
+		if (extack._msg)
+			netdev_err(dev, "%s\n", extack._msg);
+	}
+
+	kfree(page_data->data);
+	return err;
+}
+
+static u8 cmis_cdb_calc_checksum(const void *data, size_t size)
+{
+	const u8 *bytes = (const u8 *)data;
+	u8 checksum = 0;
+
+	for (size_t i = 0; i < size; i++)
+		checksum += bytes[i];
+
+	return ~checksum;
+}
+
+#define CMIS_CDB_CMD_ID_OFFSET	0x80
+
+int ethtool_cmis_cdb_execute_cmd(struct net_device *dev,
+				 struct ethtool_cmis_cdb_cmd_args *args)
+{
+	struct ethtool_module_eeprom page_data = {};
+	u32 offset;
+	int err;
+
+	args->req.chk_code =
+		cmis_cdb_calc_checksum(&args->req, sizeof(args->req));
+
+	if (args->req.lpl_len > args->read_write_len_ext) {
+		args->err_msg = "LPL length is longer than CDB read write length extension allows";
+		return -EINVAL;
+	}
+
+	/* According to the CMIS standard, there are two options to trigger the
+	 * CDB commands. The default option is triggering the command by writing
+	 * the CMDID bytes. Therefore, the command will be split to 2 calls:
+	 * First, with everything except the CMDID field and then the CMDID
+	 * field.
+	 */
+	offset = CMIS_CDB_CMD_ID_OFFSET +
+		offsetof(struct ethtool_cmis_cdb_request, body);
+	err = __ethtool_cmis_cdb_execute_cmd(dev, &page_data,
+					     ETHTOOL_CMIS_CDB_CMD_PAGE, offset,
+					     sizeof(args->req.body),
+					     &args->req.body);
+	if (err < 0)
+		return err;
+
+	offset = CMIS_CDB_CMD_ID_OFFSET +
+		offsetof(struct ethtool_cmis_cdb_request, id);
+	err = __ethtool_cmis_cdb_execute_cmd(dev, &page_data,
+					     ETHTOOL_CMIS_CDB_CMD_PAGE, offset,
+					     sizeof(args->req.id),
+					     &args->req.id);
+	if (err < 0)
+		return err;
+
+	err = cmis_cdb_wait_for_completion(dev, args);
+	if (err < 0)
+		return err;
+
+	err = cmis_cdb_wait_for_status(dev, args);
+	if (err < 0)
+		return err;
+
+	return cmis_cdb_process_reply(dev, &page_data, args);
+}
diff --git a/net/ethtool/module_fw.h b/net/ethtool/module_fw.h
index ee4a291ac1d4..6c86d05ab6cf 100644
--- a/net/ethtool/module_fw.h
+++ b/net/ethtool/module_fw.h
@@ -15,6 +15,16 @@ struct ethnl_module_fw_flash_ntf_params {
 	bool closed_sock;
 };
 
+/**
+ * struct ethtool_module_fw_flash_params - module firmware flashing parameters
+ * @password: Module password. Only valid when @pass_valid is set.
+ * @password_valid: Whether the module password is valid or not.
+ */
+struct ethtool_module_fw_flash_params {
+	__be32 password;
+	u8 password_valid:1;
+};
+
 void
 ethnl_module_fw_flash_ntf_err(struct net_device *dev,
 			      struct ethnl_module_fw_flash_ntf_params *params,
-- 
cgit v1.2.3-58-ga151


From c4f78134d45c9619339c96b4bea380b1d0699788 Mon Sep 17 00:00:00 2001
From: Danielle Ratson <danieller@nvidia.com>
Date: Thu, 27 Jun 2024 17:08:55 +0300
Subject: ethtool: cmis_fw_update: add a layer for supporting firmware update
 using CDB

According to the CMIS standard, the firmware update process is done using
a CDB commands sequence.

Implement a work that will be triggered from the module layer in the
next patch the will initiate and execute all the CDB commands in order, to
eventually complete the firmware update process.

This flashing process includes, writing the firmware image, running the new
firmware image and committing it after testing, so that it will run upon
reset.

This work will also notify user space about the progress of the firmware
update process.

Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/Makefile         |   2 +-
 net/ethtool/cmis.h           |   7 +
 net/ethtool/cmis_fw_update.c | 399 +++++++++++++++++++++++++++++++++++++++++++
 net/ethtool/module_fw.h      |  31 ++++
 4 files changed, 438 insertions(+), 1 deletion(-)
 create mode 100644 net/ethtool/cmis_fw_update.c

diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index 38806b3ecf83..9a190635fe95 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -8,4 +8,4 @@ ethtool_nl-y	:= netlink.o bitset.o strset.o linkinfo.o linkmodes.o rss.o \
 		   linkstate.o debug.o wol.o features.o privflags.o rings.o \
 		   channels.o coalesce.o pause.o eee.o tsinfo.o cabletest.o \
 		   tunnels.o fec.o eeprom.o stats.o phc_vclocks.o mm.o \
-		   module.o cmis_cdb.o pse-pd.o plca.o mm.o
+		   module.o cmis_fw_update.o cmis_cdb.o pse-pd.o plca.o mm.o
diff --git a/net/ethtool/cmis.h b/net/ethtool/cmis.h
index 295f5d0df915..e71cc3e1b7eb 100644
--- a/net/ethtool/cmis.h
+++ b/net/ethtool/cmis.h
@@ -20,6 +20,12 @@ struct ethtool_cmis_cdb {
 enum ethtool_cmis_cdb_cmd_id {
 	ETHTOOL_CMIS_CDB_CMD_QUERY_STATUS		= 0x0000,
 	ETHTOOL_CMIS_CDB_CMD_MODULE_FEATURES		= 0x0040,
+	ETHTOOL_CMIS_CDB_CMD_FW_MANAGMENT_FEATURES	= 0x0041,
+	ETHTOOL_CMIS_CDB_CMD_START_FW_DOWNLOAD		= 0x0101,
+	ETHTOOL_CMIS_CDB_CMD_WRITE_FW_BLOCK_LPL		= 0x0103,
+	ETHTOOL_CMIS_CDB_CMD_COMPLETE_FW_DOWNLOAD	= 0x0107,
+	ETHTOOL_CMIS_CDB_CMD_RUN_FW_IMAGE		= 0x0109,
+	ETHTOOL_CMIS_CDB_CMD_COMMIT_FW_IMAGE		= 0x010A,
 };
 
 /**
@@ -47,6 +53,7 @@ struct ethtool_cmis_cdb_request {
 
 #define CDB_F_COMPLETION_VALID		BIT(0)
 #define CDB_F_STATUS_VALID		BIT(1)
+#define CDB_F_MODULE_STATE_VALID	BIT(2)
 
 /**
  * struct ethtool_cmis_cdb_cmd_args - CDB commands execution arguments
diff --git a/net/ethtool/cmis_fw_update.c b/net/ethtool/cmis_fw_update.c
new file mode 100644
index 000000000000..ae4b4b28a601
--- /dev/null
+++ b/net/ethtool/cmis_fw_update.c
@@ -0,0 +1,399 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/ethtool.h>
+#include <linux/firmware.h>
+
+#include "common.h"
+#include "module_fw.h"
+#include "cmis.h"
+
+struct cmis_fw_update_fw_mng_features {
+	u8	start_cmd_payload_size;
+	u16	max_duration_start;
+	u16	max_duration_write;
+	u16	max_duration_complete;
+};
+
+/* See section 9.4.2 "CMD 0041h: Firmware Management Features" in CMIS standard
+ * revision 5.2.
+ * struct cmis_cdb_fw_mng_features_rpl is a structured layout of the flat
+ * array, ethtool_cmis_cdb_rpl::payload.
+ */
+struct cmis_cdb_fw_mng_features_rpl {
+	u8	resv1;
+	u8	resv2;
+	u8	start_cmd_payload_size;
+	u8	resv3;
+	u8	read_write_len_ext;
+	u8	write_mechanism;
+	u8	resv4;
+	u8	resv5;
+	__be16	max_duration_start;
+	__be16	resv6;
+	__be16	max_duration_write;
+	__be16	max_duration_complete;
+	__be16	resv7;
+};
+
+#define CMIS_CDB_FW_WRITE_MECHANISM_LPL	0x01
+
+static int
+cmis_fw_update_fw_mng_features_get(struct ethtool_cmis_cdb *cdb,
+				   struct net_device *dev,
+				   struct cmis_fw_update_fw_mng_features *fw_mng,
+				   struct ethnl_module_fw_flash_ntf_params *ntf_params)
+{
+	struct ethtool_cmis_cdb_cmd_args args = {};
+	struct cmis_cdb_fw_mng_features_rpl *rpl;
+	u8 flags = CDB_F_STATUS_VALID;
+	int err;
+
+	ethtool_cmis_cdb_check_completion_flag(cdb->cmis_rev, &flags);
+	ethtool_cmis_cdb_compose_args(&args,
+				      ETHTOOL_CMIS_CDB_CMD_FW_MANAGMENT_FEATURES,
+				      NULL, 0, cdb->max_completion_time,
+				      cdb->read_write_len_ext, 1000,
+				      sizeof(*rpl), flags);
+
+	err = ethtool_cmis_cdb_execute_cmd(dev, &args);
+	if (err < 0) {
+		ethnl_module_fw_flash_ntf_err(dev, ntf_params,
+					      "FW Management Features command failed",
+					      args.err_msg);
+		return err;
+	}
+
+	rpl = (struct cmis_cdb_fw_mng_features_rpl *)args.req.payload;
+	if (!(rpl->write_mechanism == CMIS_CDB_FW_WRITE_MECHANISM_LPL)) {
+		ethnl_module_fw_flash_ntf_err(dev, ntf_params,
+					      "Write LPL is not supported",
+					      NULL);
+		return  -EOPNOTSUPP;
+	}
+
+	/* Above, we used read_write_len_ext that we got from CDB
+	 * advertisement. Update it with the value that we got from module
+	 * features query, which is specific for Firmware Management Commands
+	 * (IDs 0100h-01FFh).
+	 */
+	cdb->read_write_len_ext = rpl->read_write_len_ext;
+	fw_mng->start_cmd_payload_size = rpl->start_cmd_payload_size;
+	fw_mng->max_duration_start = be16_to_cpu(rpl->max_duration_start);
+	fw_mng->max_duration_write = be16_to_cpu(rpl->max_duration_write);
+	fw_mng->max_duration_complete = be16_to_cpu(rpl->max_duration_complete);
+
+	return 0;
+}
+
+/* See section 9.7.2 "CMD 0101h: Start Firmware Download" in CMIS standard
+ * revision 5.2.
+ * struct cmis_cdb_start_fw_download_pl is a structured layout of the
+ * flat array, ethtool_cmis_cdb_request::payload.
+ */
+struct cmis_cdb_start_fw_download_pl {
+	__struct_group(cmis_cdb_start_fw_download_pl_h, head, /* no attrs */,
+			__be32	image_size;
+			__be32	resv1;
+	);
+	u8 vendor_data[ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH -
+		sizeof(struct cmis_cdb_start_fw_download_pl_h)];
+};
+
+static int
+cmis_fw_update_start_download(struct ethtool_cmis_cdb *cdb,
+			      struct ethtool_cmis_fw_update_params *fw_update,
+			      struct cmis_fw_update_fw_mng_features *fw_mng)
+{
+	u8 vendor_data_size = fw_mng->start_cmd_payload_size;
+	struct cmis_cdb_start_fw_download_pl pl = {};
+	struct ethtool_cmis_cdb_cmd_args args = {};
+	u8 lpl_len;
+	int err;
+
+	pl.image_size = cpu_to_be32(fw_update->fw->size);
+	memcpy(pl.vendor_data, fw_update->fw->data, vendor_data_size);
+
+	lpl_len = offsetof(struct cmis_cdb_start_fw_download_pl,
+			   vendor_data[vendor_data_size]);
+
+	ethtool_cmis_cdb_compose_args(&args,
+				      ETHTOOL_CMIS_CDB_CMD_START_FW_DOWNLOAD,
+				      (u8 *)&pl, lpl_len,
+				      fw_mng->max_duration_start,
+				      cdb->read_write_len_ext, 1000, 0,
+				      CDB_F_COMPLETION_VALID | CDB_F_STATUS_VALID);
+
+	err = ethtool_cmis_cdb_execute_cmd(fw_update->dev, &args);
+	if (err < 0)
+		ethnl_module_fw_flash_ntf_err(fw_update->dev,
+					      &fw_update->ntf_params,
+					      "Start FW download command failed",
+					      args.err_msg);
+
+	return err;
+}
+
+/* See section 9.7.4 "CMD 0103h: Write Firmware Block LPL" in CMIS standard
+ * revision 5.2.
+ * struct cmis_cdb_write_fw_block_lpl_pl is a structured layout of the
+ * flat array, ethtool_cmis_cdb_request::payload.
+ */
+struct cmis_cdb_write_fw_block_lpl_pl {
+	__be32	block_address;
+	u8 fw_block[ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH - sizeof(__be32)];
+};
+
+static int
+cmis_fw_update_write_image(struct ethtool_cmis_cdb *cdb,
+			   struct ethtool_cmis_fw_update_params *fw_update,
+			   struct cmis_fw_update_fw_mng_features *fw_mng)
+{
+	u8 start = fw_mng->start_cmd_payload_size;
+	u32 offset, max_block_size, max_lpl_len;
+	u32 image_size = fw_update->fw->size;
+	int err;
+
+	max_lpl_len = min_t(u32,
+			    ethtool_cmis_get_max_payload_size(cdb->read_write_len_ext),
+			    ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH);
+	max_block_size =
+		max_lpl_len - sizeof_field(struct cmis_cdb_write_fw_block_lpl_pl,
+					   block_address);
+
+	for (offset = start; offset < image_size; offset += max_block_size) {
+		struct cmis_cdb_write_fw_block_lpl_pl pl = {
+			.block_address = cpu_to_be32(offset - start),
+		};
+		struct ethtool_cmis_cdb_cmd_args args = {};
+		u32 block_size, lpl_len;
+
+		ethnl_module_fw_flash_ntf_in_progress(fw_update->dev,
+						      &fw_update->ntf_params,
+						      offset - start,
+						      image_size);
+		block_size = min_t(u32, max_block_size, image_size - offset);
+		memcpy(pl.fw_block, &fw_update->fw->data[offset], block_size);
+		lpl_len = block_size +
+			sizeof_field(struct cmis_cdb_write_fw_block_lpl_pl,
+				     block_address);
+
+		ethtool_cmis_cdb_compose_args(&args,
+					      ETHTOOL_CMIS_CDB_CMD_WRITE_FW_BLOCK_LPL,
+					      (u8 *)&pl, lpl_len,
+					      fw_mng->max_duration_write,
+					      cdb->read_write_len_ext, 1, 0,
+					      CDB_F_COMPLETION_VALID | CDB_F_STATUS_VALID);
+
+		err = ethtool_cmis_cdb_execute_cmd(fw_update->dev, &args);
+		if (err < 0) {
+			ethnl_module_fw_flash_ntf_err(fw_update->dev,
+						      &fw_update->ntf_params,
+						      "Write FW block LPL command failed",
+						      args.err_msg);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static int
+cmis_fw_update_complete_download(struct ethtool_cmis_cdb *cdb,
+				 struct net_device *dev,
+				 struct cmis_fw_update_fw_mng_features *fw_mng,
+				 struct ethnl_module_fw_flash_ntf_params *ntf_params)
+{
+	struct ethtool_cmis_cdb_cmd_args args = {};
+	int err;
+
+	ethtool_cmis_cdb_compose_args(&args,
+				      ETHTOOL_CMIS_CDB_CMD_COMPLETE_FW_DOWNLOAD,
+				      NULL, 0, fw_mng->max_duration_complete,
+				      cdb->read_write_len_ext, 1000, 0,
+				      CDB_F_COMPLETION_VALID | CDB_F_STATUS_VALID);
+
+	err = ethtool_cmis_cdb_execute_cmd(dev, &args);
+	if (err < 0)
+		ethnl_module_fw_flash_ntf_err(dev, ntf_params,
+					      "Complete FW download command failed",
+					      args.err_msg);
+
+	return err;
+}
+
+static int
+cmis_fw_update_download_image(struct ethtool_cmis_cdb *cdb,
+			      struct ethtool_cmis_fw_update_params *fw_update,
+			      struct cmis_fw_update_fw_mng_features *fw_mng)
+{
+	int err;
+
+	err = cmis_fw_update_start_download(cdb, fw_update, fw_mng);
+	if (err < 0)
+		return err;
+
+	err = cmis_fw_update_write_image(cdb, fw_update, fw_mng);
+	if (err < 0)
+		return err;
+
+	err = cmis_fw_update_complete_download(cdb, fw_update->dev, fw_mng,
+					       &fw_update->ntf_params);
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
+enum {
+	CMIS_MODULE_LOW_PWR	= 1,
+	CMIS_MODULE_READY	= 3,
+};
+
+static bool module_is_ready(u8 data)
+{
+	u8 state = (data >> 1) & 7;
+
+	return state == CMIS_MODULE_READY || state == CMIS_MODULE_LOW_PWR;
+}
+
+#define CMIS_MODULE_READY_MAX_DURATION_MSEC	1000
+#define CMIS_MODULE_STATE_OFFSET		3
+
+static int
+cmis_fw_update_wait_for_module_state(struct net_device *dev, u8 flags)
+{
+	u8 state;
+
+	return ethtool_cmis_wait_for_cond(dev, flags, CDB_F_MODULE_STATE_VALID,
+					  CMIS_MODULE_READY_MAX_DURATION_MSEC,
+					  CMIS_MODULE_STATE_OFFSET,
+					  module_is_ready, NULL, &state);
+}
+
+/* See section 9.7.10 "CMD 0109h: Run Firmware Image" in CMIS standard
+ * revision 5.2.
+ * struct cmis_cdb_run_fw_image_pl is a structured layout of the flat
+ * array, ethtool_cmis_cdb_request::payload.
+ */
+struct cmis_cdb_run_fw_image_pl {
+	u8 resv1;
+	u8 image_to_run;
+	u16 delay_to_reset;
+};
+
+static int
+cmis_fw_update_run_image(struct ethtool_cmis_cdb *cdb, struct net_device *dev,
+			 struct ethnl_module_fw_flash_ntf_params *ntf_params)
+{
+	struct ethtool_cmis_cdb_cmd_args args = {};
+	struct cmis_cdb_run_fw_image_pl pl = {0};
+	int err;
+
+	ethtool_cmis_cdb_compose_args(&args, ETHTOOL_CMIS_CDB_CMD_RUN_FW_IMAGE,
+				      (u8 *)&pl, sizeof(pl),
+				      cdb->max_completion_time,
+				      cdb->read_write_len_ext, 1000, 0,
+				      CDB_F_MODULE_STATE_VALID);
+
+	err = ethtool_cmis_cdb_execute_cmd(dev, &args);
+	if (err < 0) {
+		ethnl_module_fw_flash_ntf_err(dev, ntf_params,
+					      "Run image command failed",
+					      args.err_msg);
+		return err;
+	}
+
+	err = cmis_fw_update_wait_for_module_state(dev, args.flags);
+	if (err < 0)
+		ethnl_module_fw_flash_ntf_err(dev, ntf_params,
+					      "Module is not ready on time after reset",
+					      NULL);
+
+	return err;
+}
+
+static int
+cmis_fw_update_commit_image(struct ethtool_cmis_cdb *cdb,
+			    struct net_device *dev,
+			    struct ethnl_module_fw_flash_ntf_params *ntf_params)
+{
+	struct ethtool_cmis_cdb_cmd_args args = {};
+	int err;
+
+	ethtool_cmis_cdb_compose_args(&args,
+				      ETHTOOL_CMIS_CDB_CMD_COMMIT_FW_IMAGE,
+				      NULL, 0, cdb->max_completion_time,
+				      cdb->read_write_len_ext, 1000, 0,
+				      CDB_F_COMPLETION_VALID | CDB_F_STATUS_VALID);
+
+	err = ethtool_cmis_cdb_execute_cmd(dev, &args);
+	if (err < 0)
+		ethnl_module_fw_flash_ntf_err(dev, ntf_params,
+					      "Commit image command failed",
+					      args.err_msg);
+
+	return err;
+}
+
+static int cmis_fw_update_reset(struct net_device *dev)
+{
+	__u32 reset_data = ETH_RESET_PHY;
+
+	return dev->ethtool_ops->reset(dev, &reset_data);
+}
+
+void
+ethtool_cmis_fw_update(struct ethtool_cmis_fw_update_params *fw_update)
+{
+	struct ethnl_module_fw_flash_ntf_params *ntf_params =
+						&fw_update->ntf_params;
+	struct cmis_fw_update_fw_mng_features fw_mng = {0};
+	struct net_device *dev = fw_update->dev;
+	struct ethtool_cmis_cdb *cdb;
+	int err;
+
+	cdb = ethtool_cmis_cdb_init(dev, &fw_update->params, ntf_params);
+	if (IS_ERR(cdb))
+		goto err_send_ntf;
+
+	ethnl_module_fw_flash_ntf_start(dev, ntf_params);
+
+	err = cmis_fw_update_fw_mng_features_get(cdb, dev, &fw_mng, ntf_params);
+	if (err < 0)
+		goto err_cdb_fini;
+
+	err = cmis_fw_update_download_image(cdb, fw_update, &fw_mng);
+	if (err < 0)
+		goto err_cdb_fini;
+
+	err = cmis_fw_update_run_image(cdb, dev, ntf_params);
+	if (err < 0)
+		goto err_cdb_fini;
+
+	/* The CDB command "Run Firmware Image" resets the firmware, so the new
+	 * one might have different settings.
+	 * Free the old CDB instance, and init a new one.
+	 */
+	ethtool_cmis_cdb_fini(cdb);
+
+	cdb = ethtool_cmis_cdb_init(dev, &fw_update->params, ntf_params);
+	if (IS_ERR(cdb))
+		goto err_send_ntf;
+
+	err = cmis_fw_update_commit_image(cdb, dev, ntf_params);
+	if (err < 0)
+		goto err_cdb_fini;
+
+	err = cmis_fw_update_reset(dev);
+	if (err < 0)
+		goto err_cdb_fini;
+
+	ethnl_module_fw_flash_ntf_complete(dev, ntf_params);
+	ethtool_cmis_cdb_fini(cdb);
+	return;
+
+err_cdb_fini:
+	ethtool_cmis_cdb_fini(cdb);
+err_send_ntf:
+	ethnl_module_fw_flash_ntf_err(dev, ntf_params, NULL, NULL);
+}
diff --git a/net/ethtool/module_fw.h b/net/ethtool/module_fw.h
index 6c86d05ab6cf..d0fc2529b60e 100644
--- a/net/ethtool/module_fw.h
+++ b/net/ethtool/module_fw.h
@@ -25,6 +25,35 @@ struct ethtool_module_fw_flash_params {
 	u8 password_valid:1;
 };
 
+/**
+ * struct ethtool_cmis_fw_update_params - CMIS firmware update specific
+ *						parameters
+ * @dev: Pointer to the net_device to be flashed.
+ * @params: Module firmware flashing parameters.
+ * @ntf_params: Module firmware flashing notification parameters.
+ * @fw: Firmware to flash.
+ */
+struct ethtool_cmis_fw_update_params {
+	struct net_device *dev;
+	struct ethtool_module_fw_flash_params params;
+	struct ethnl_module_fw_flash_ntf_params ntf_params;
+	const struct firmware *fw;
+};
+
+/**
+ * struct ethtool_module_fw_flash - module firmware flashing
+ * @list: List node for &module_fw_flash_work_list.
+ * @dev_tracker: Refcount tracker for @dev.
+ * @work: The flashing firmware work.
+ * @fw_update: CMIS firmware update specific parameters.
+ */
+struct ethtool_module_fw_flash {
+	struct list_head list;
+	netdevice_tracker dev_tracker;
+	struct work_struct work;
+	struct ethtool_cmis_fw_update_params fw_update;
+};
+
 void
 ethnl_module_fw_flash_ntf_err(struct net_device *dev,
 			      struct ethnl_module_fw_flash_ntf_params *params,
@@ -39,3 +68,5 @@ void
 ethnl_module_fw_flash_ntf_in_progress(struct net_device *dev,
 				      struct ethnl_module_fw_flash_ntf_params *params,
 				      u64 done, u64 total);
+
+void ethtool_cmis_fw_update(struct ethtool_cmis_fw_update_params *params);
-- 
cgit v1.2.3-58-ga151


From 32b4c8b53ee7799e34a2b1634d32d3ce1e36c44e Mon Sep 17 00:00:00 2001
From: Danielle Ratson <danieller@nvidia.com>
Date: Thu, 27 Jun 2024 17:08:56 +0300
Subject: ethtool: Add ability to flash transceiver modules' firmware

Add the ability to flash the modules' firmware by implementing the
interface between the user space and the kernel.

Example from a succeeding implementation:

 # ethtool --flash-module-firmware swp40 file test.bin

 Transceiver module firmware flashing started for device swp40
 Transceiver module firmware flashing in progress for device swp40
 Progress: 99%
 Transceiver module firmware flashing completed for device swp40

In addition, add infrastructure that allows modules to set socket-specific
private data. This ensures that when a socket is closed from user space
during the flashing process, the right socket halts sending notifications
to user space until the work item is completed.

Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/module.c    | 277 ++++++++++++++++++++++++++++++++++++++++++++++++
 net/ethtool/module_fw.h |   3 +
 net/ethtool/netlink.c   |  39 +++++++
 net/ethtool/netlink.h   |  15 +++
 4 files changed, 334 insertions(+)

diff --git a/net/ethtool/module.c b/net/ethtool/module.c
index ba728b4a38a1..6b7448df08d5 100644
--- a/net/ethtool/module.c
+++ b/net/ethtool/module.c
@@ -1,6 +1,9 @@
 // SPDX-License-Identifier: GPL-2.0-only
 
 #include <linux/ethtool.h>
+#include <linux/firmware.h>
+#include <linux/sfp.h>
+#include <net/devlink.h>
 
 #include "netlink.h"
 #include "common.h"
@@ -34,6 +37,12 @@ static int module_get_power_mode(struct net_device *dev,
 	if (!ops->get_module_power_mode)
 		return 0;
 
+	if (dev->module_fw_flash_in_progress) {
+		NL_SET_ERR_MSG(extack,
+			       "Module firmware flashing is in progress");
+		return -EBUSY;
+	}
+
 	return ops->get_module_power_mode(dev, &data->power, extack);
 }
 
@@ -110,6 +119,12 @@ ethnl_set_module_validate(struct ethnl_req_info *req_info,
 	if (!tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY])
 		return 0;
 
+	if (req_info->dev->module_fw_flash_in_progress) {
+		NL_SET_ERR_MSG(info->extack,
+			       "Module firmware flashing is in progress");
+		return -EBUSY;
+	}
+
 	if (!ops->get_module_power_mode || !ops->set_module_power_mode) {
 		NL_SET_ERR_MSG_ATTR(info->extack,
 				    tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY],
@@ -160,6 +175,268 @@ const struct ethnl_request_ops ethnl_module_request_ops = {
 	.set_ntf_cmd		= ETHTOOL_MSG_MODULE_NTF,
 };
 
+/* MODULE_FW_FLASH_ACT */
+
+const struct nla_policy
+ethnl_module_fw_flash_act_policy[ETHTOOL_A_MODULE_FW_FLASH_PASSWORD + 1] = {
+	[ETHTOOL_A_MODULE_FW_FLASH_HEADER] =
+		NLA_POLICY_NESTED(ethnl_header_policy),
+	[ETHTOOL_A_MODULE_FW_FLASH_FILE_NAME] = { .type = NLA_NUL_STRING },
+	[ETHTOOL_A_MODULE_FW_FLASH_PASSWORD] = { .type = NLA_U32 },
+};
+
+static LIST_HEAD(module_fw_flash_work_list);
+static DEFINE_SPINLOCK(module_fw_flash_work_list_lock);
+
+static int
+module_flash_fw_work_list_add(struct ethtool_module_fw_flash *module_fw,
+			      struct genl_info *info)
+{
+	struct ethtool_module_fw_flash *work;
+
+	/* First, check if already registered. */
+	spin_lock(&module_fw_flash_work_list_lock);
+	list_for_each_entry(work, &module_fw_flash_work_list, list) {
+		if (work->fw_update.ntf_params.portid == info->snd_portid &&
+		    work->fw_update.dev == module_fw->fw_update.dev) {
+			spin_unlock(&module_fw_flash_work_list_lock);
+			return -EALREADY;
+		}
+	}
+
+	list_add_tail(&module_fw->list, &module_fw_flash_work_list);
+	spin_unlock(&module_fw_flash_work_list_lock);
+
+	return 0;
+}
+
+static void module_flash_fw_work_list_del(struct list_head *list)
+{
+	spin_lock(&module_fw_flash_work_list_lock);
+	list_del(list);
+	spin_unlock(&module_fw_flash_work_list_lock);
+}
+
+static void module_flash_fw_work(struct work_struct *work)
+{
+	struct ethtool_module_fw_flash *module_fw;
+
+	module_fw = container_of(work, struct ethtool_module_fw_flash, work);
+
+	ethtool_cmis_fw_update(&module_fw->fw_update);
+
+	module_flash_fw_work_list_del(&module_fw->list);
+	module_fw->fw_update.dev->module_fw_flash_in_progress = false;
+	netdev_put(module_fw->fw_update.dev, &module_fw->dev_tracker);
+	release_firmware(module_fw->fw_update.fw);
+	kfree(module_fw);
+}
+
+#define MODULE_EEPROM_PHYS_ID_PAGE	0
+#define MODULE_EEPROM_PHYS_ID_I2C_ADDR	0x50
+
+static int module_flash_fw_work_init(struct ethtool_module_fw_flash *module_fw,
+				     struct net_device *dev,
+				     struct netlink_ext_ack *extack)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	struct ethtool_module_eeprom page_data = {};
+	u8 phys_id;
+	int err;
+
+	/* Fetch the SFF-8024 Identifier Value. For all supported standards, it
+	 * is located at I2C address 0x50, byte 0. See section 4.1 in SFF-8024,
+	 * revision 4.9.
+	 */
+	page_data.page = MODULE_EEPROM_PHYS_ID_PAGE;
+	page_data.offset = SFP_PHYS_ID;
+	page_data.length = sizeof(phys_id);
+	page_data.i2c_address = MODULE_EEPROM_PHYS_ID_I2C_ADDR;
+	page_data.data = &phys_id;
+
+	err = ops->get_module_eeprom_by_page(dev, &page_data, extack);
+	if (err < 0)
+		return err;
+
+	switch (phys_id) {
+	case SFF8024_ID_QSFP_DD:
+	case SFF8024_ID_OSFP:
+	case SFF8024_ID_DSFP:
+	case SFF8024_ID_QSFP_PLUS_CMIS:
+	case SFF8024_ID_SFP_DD_CMIS:
+	case SFF8024_ID_SFP_PLUS_CMIS:
+		INIT_WORK(&module_fw->work, module_flash_fw_work);
+		break;
+	default:
+		NL_SET_ERR_MSG(extack,
+			       "Module type does not support firmware flashing");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+void ethnl_module_fw_flash_sock_destroy(struct ethnl_sock_priv *sk_priv)
+{
+	struct ethtool_module_fw_flash *work;
+
+	spin_lock(&module_fw_flash_work_list_lock);
+	list_for_each_entry(work, &module_fw_flash_work_list, list) {
+		if (work->fw_update.dev == sk_priv->dev &&
+		    work->fw_update.ntf_params.portid == sk_priv->portid) {
+			work->fw_update.ntf_params.closed_sock = true;
+			break;
+		}
+	}
+	spin_unlock(&module_fw_flash_work_list_lock);
+}
+
+static int
+module_flash_fw_schedule(struct net_device *dev, const char *file_name,
+			 struct ethtool_module_fw_flash_params *params,
+			 struct sk_buff *skb, struct genl_info *info)
+{
+	struct ethtool_cmis_fw_update_params *fw_update;
+	struct ethtool_module_fw_flash *module_fw;
+	int err;
+
+	module_fw = kzalloc(sizeof(*module_fw), GFP_KERNEL);
+	if (!module_fw)
+		return -ENOMEM;
+
+	fw_update = &module_fw->fw_update;
+	fw_update->params = *params;
+	err = request_firmware_direct(&fw_update->fw,
+				      file_name, &dev->dev);
+	if (err) {
+		NL_SET_ERR_MSG(info->extack,
+			       "Failed to request module firmware image");
+		goto err_free;
+	}
+
+	err = module_flash_fw_work_init(module_fw, dev, info->extack);
+	if (err < 0)
+		goto err_release_firmware;
+
+	dev->module_fw_flash_in_progress = true;
+	netdev_hold(dev, &module_fw->dev_tracker, GFP_KERNEL);
+	fw_update->dev = dev;
+	fw_update->ntf_params.portid = info->snd_portid;
+	fw_update->ntf_params.seq = info->snd_seq;
+	fw_update->ntf_params.closed_sock = false;
+
+	err = ethnl_sock_priv_set(skb, dev, fw_update->ntf_params.portid,
+				  ETHTOOL_SOCK_TYPE_MODULE_FW_FLASH);
+	if (err < 0)
+		goto err_release_firmware;
+
+	err = module_flash_fw_work_list_add(module_fw, info);
+	if (err < 0)
+		goto err_release_firmware;
+
+	schedule_work(&module_fw->work);
+
+	return 0;
+
+err_release_firmware:
+	release_firmware(fw_update->fw);
+err_free:
+	kfree(module_fw);
+	return err;
+}
+
+static int module_flash_fw(struct net_device *dev, struct nlattr **tb,
+			   struct sk_buff *skb, struct genl_info *info)
+{
+	struct ethtool_module_fw_flash_params params = {};
+	const char *file_name;
+	struct nlattr *attr;
+
+	if (GENL_REQ_ATTR_CHECK(info, ETHTOOL_A_MODULE_FW_FLASH_FILE_NAME))
+		return -EINVAL;
+
+	file_name = nla_data(tb[ETHTOOL_A_MODULE_FW_FLASH_FILE_NAME]);
+
+	attr = tb[ETHTOOL_A_MODULE_FW_FLASH_PASSWORD];
+	if (attr) {
+		params.password = cpu_to_be32(nla_get_u32(attr));
+		params.password_valid = true;
+	}
+
+	return module_flash_fw_schedule(dev, file_name, &params, skb, info);
+}
+
+static int ethnl_module_fw_flash_validate(struct net_device *dev,
+					  struct netlink_ext_ack *extack)
+{
+	struct devlink_port *devlink_port = dev->devlink_port;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+
+	if (!ops->set_module_eeprom_by_page ||
+	    !ops->get_module_eeprom_by_page) {
+		NL_SET_ERR_MSG(extack,
+			       "Flashing module firmware is not supported by this device");
+		return -EOPNOTSUPP;
+	}
+
+	if (!ops->reset) {
+		NL_SET_ERR_MSG(extack,
+			       "Reset module is not supported by this device, so flashing is not permitted");
+		return -EOPNOTSUPP;
+	}
+
+	if (dev->module_fw_flash_in_progress) {
+		NL_SET_ERR_MSG(extack, "Module firmware flashing already in progress");
+		return -EBUSY;
+	}
+
+	if (dev->flags & IFF_UP) {
+		NL_SET_ERR_MSG(extack, "Netdevice is up, so flashing is not permitted");
+		return -EBUSY;
+	}
+
+	if (devlink_port && devlink_port->attrs.split) {
+		NL_SET_ERR_MSG(extack, "Can't perform firmware flashing on a split port");
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+int ethnl_act_module_fw_flash(struct sk_buff *skb, struct genl_info *info)
+{
+	struct ethnl_req_info req_info = {};
+	struct nlattr **tb = info->attrs;
+	struct net_device *dev;
+	int ret;
+
+	ret = ethnl_parse_header_dev_get(&req_info,
+					 tb[ETHTOOL_A_MODULE_FW_FLASH_HEADER],
+					 genl_info_net(info), info->extack,
+					 true);
+	if (ret < 0)
+		return ret;
+	dev = req_info.dev;
+
+	rtnl_lock();
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		goto out_rtnl;
+
+	ret = ethnl_module_fw_flash_validate(dev, info->extack);
+	if (ret < 0)
+		goto out_rtnl;
+
+	ret = module_flash_fw(dev, tb, skb, info);
+
+	ethnl_ops_complete(dev);
+
+out_rtnl:
+	rtnl_unlock();
+	ethnl_parse_header_dev_put(&req_info);
+	return ret;
+}
+
 /* MODULE_FW_FLASH_NTF */
 
 static int
diff --git a/net/ethtool/module_fw.h b/net/ethtool/module_fw.h
index d0fc2529b60e..634543a12d0c 100644
--- a/net/ethtool/module_fw.h
+++ b/net/ethtool/module_fw.h
@@ -1,6 +1,7 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 
 #include <uapi/linux/ethtool.h>
+#include "netlink.h"
 
 /**
  * struct ethnl_module_fw_flash_ntf_params - module firmware flashing
@@ -54,6 +55,8 @@ struct ethtool_module_fw_flash {
 	struct ethtool_cmis_fw_update_params fw_update;
 };
 
+void ethnl_module_fw_flash_sock_destroy(struct ethnl_sock_priv *sk_priv);
+
 void
 ethnl_module_fw_flash_ntf_err(struct net_device *dev,
 			      struct ethnl_module_fw_flash_ntf_params *params,
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index a5907bbde427..81fe2e5b95f6 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -4,6 +4,7 @@
 #include <linux/ethtool_netlink.h>
 #include <linux/pm_runtime.h>
 #include "netlink.h"
+#include "module_fw.h"
 
 static struct genl_family ethtool_genl_family;
 
@@ -30,6 +31,35 @@ const struct nla_policy ethnl_header_policy_stats[] = {
 							  ETHTOOL_FLAGS_STATS),
 };
 
+int ethnl_sock_priv_set(struct sk_buff *skb, struct net_device *dev, u32 portid,
+			enum ethnl_sock_type type)
+{
+	struct ethnl_sock_priv *sk_priv;
+
+	sk_priv = genl_sk_priv_get(&ethtool_genl_family, NETLINK_CB(skb).sk);
+	if (IS_ERR(sk_priv))
+		return PTR_ERR(sk_priv);
+
+	sk_priv->dev = dev;
+	sk_priv->portid = portid;
+	sk_priv->type = type;
+
+	return 0;
+}
+
+static void ethnl_sock_priv_destroy(void *priv)
+{
+	struct ethnl_sock_priv *sk_priv = priv;
+
+	switch (sk_priv->type) {
+	case ETHTOOL_SOCK_TYPE_MODULE_FW_FLASH:
+		ethnl_module_fw_flash_sock_destroy(sk_priv);
+		break;
+	default:
+		break;
+	}
+}
+
 int ethnl_ops_begin(struct net_device *dev)
 {
 	int ret;
@@ -1142,6 +1172,13 @@ static const struct genl_ops ethtool_genl_ops[] = {
 		.policy = ethnl_mm_set_policy,
 		.maxattr = ARRAY_SIZE(ethnl_mm_set_policy) - 1,
 	},
+	{
+		.cmd	= ETHTOOL_MSG_MODULE_FW_FLASH_ACT,
+		.flags	= GENL_UNS_ADMIN_PERM,
+		.doit	= ethnl_act_module_fw_flash,
+		.policy	= ethnl_module_fw_flash_act_policy,
+		.maxattr = ARRAY_SIZE(ethnl_module_fw_flash_act_policy) - 1,
+	},
 };
 
 static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
@@ -1158,6 +1195,8 @@ static struct genl_family ethtool_genl_family __ro_after_init = {
 	.resv_start_op	= ETHTOOL_MSG_MODULE_GET + 1,
 	.mcgrps		= ethtool_nl_mcgrps,
 	.n_mcgrps	= ARRAY_SIZE(ethtool_nl_mcgrps),
+	.sock_priv_size		= sizeof(struct ethnl_sock_priv),
+	.sock_priv_destroy	= ethnl_sock_priv_destroy,
 };
 
 /* module setup */
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 5e6c6a7b7adc..46ec273a87c5 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -284,6 +284,19 @@ struct ethnl_reply_data {
 int ethnl_ops_begin(struct net_device *dev);
 void ethnl_ops_complete(struct net_device *dev);
 
+enum ethnl_sock_type {
+	ETHTOOL_SOCK_TYPE_MODULE_FW_FLASH,
+};
+
+struct ethnl_sock_priv {
+	struct net_device *dev;
+	u32 portid;
+	enum ethnl_sock_type type;
+};
+
+int ethnl_sock_priv_set(struct sk_buff *skb, struct net_device *dev, u32 portid,
+			enum ethnl_sock_type type);
+
 /**
  * struct ethnl_request_ops - unified handling of GET and SET requests
  * @request_cmd:      command id for request (GET)
@@ -442,6 +455,7 @@ extern const struct nla_policy ethnl_plca_set_cfg_policy[ETHTOOL_A_PLCA_MAX + 1]
 extern const struct nla_policy ethnl_plca_get_status_policy[ETHTOOL_A_PLCA_HEADER + 1];
 extern const struct nla_policy ethnl_mm_get_policy[ETHTOOL_A_MM_HEADER + 1];
 extern const struct nla_policy ethnl_mm_set_policy[ETHTOOL_A_MM_MAX + 1];
+extern const struct nla_policy ethnl_module_fw_flash_act_policy[ETHTOOL_A_MODULE_FW_FLASH_PASSWORD + 1];
 
 int ethnl_set_features(struct sk_buff *skb, struct genl_info *info);
 int ethnl_act_cable_test(struct sk_buff *skb, struct genl_info *info);
@@ -449,6 +463,7 @@ int ethnl_act_cable_test_tdr(struct sk_buff *skb, struct genl_info *info);
 int ethnl_tunnel_info_doit(struct sk_buff *skb, struct genl_info *info);
 int ethnl_tunnel_info_start(struct netlink_callback *cb);
 int ethnl_tunnel_info_dumpit(struct sk_buff *skb, struct netlink_callback *cb);
+int ethnl_act_module_fw_flash(struct sk_buff *skb, struct genl_info *info);
 
 extern const char stats_std_names[__ETHTOOL_STATS_CNT][ETH_GSTRING_LEN];
 extern const char stats_eth_phy_names[__ETHTOOL_A_STATS_ETH_PHY_CNT][ETH_GSTRING_LEN];
-- 
cgit v1.2.3-58-ga151


From d5fbb2eb33c287450dadd6e2b5f6b698f0243a14 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:38 +0200
Subject: selftests: libs: Expand "$@" where possible

In some functions, argument-forwarding through "$@" without listing the
individual arguments explicitly is fundamental to the operation of a
function. E.g. xfail_on_veth() should be able to run various tests in the
fail-to-xfail regime, and usage of "$@" is appropriate as an abstraction
mechanism. For functions such as simple_if_init(), $@ is a handy way to
pass an array.

In other functions, it's merely a mechanism to save some typing, which
however ends up obscuring the real arguments and makes life hard for those
that end up reading the code.

This patch adds some of the implicit function arguments and correspondingly
expands $@'s. In several cases this will come in handy as following patches
adjust the parameter lists.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/forwarding/lib.sh      | 67 ++++++++++++++++++----
 .../selftests/net/forwarding/mirror_gre_flower.sh  | 13 ++++-
 .../selftests/net/forwarding/mirror_gre_lib.sh     | 63 +++++++++++++++++---
 .../testing/selftests/net/forwarding/mirror_lib.sh | 63 +++++++++++++++++---
 4 files changed, 176 insertions(+), 30 deletions(-)

diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh
index 9086d2015296..df7cacb29160 100644
--- a/tools/testing/selftests/net/forwarding/lib.sh
+++ b/tools/testing/selftests/net/forwarding/lib.sh
@@ -1257,22 +1257,34 @@ __icmp_capture_add_del()
 
 icmp_capture_install()
 {
-	__icmp_capture_add_del add 100 "" "$@"
+	local tundev=$1; shift
+	local filter=$1; shift
+
+	__icmp_capture_add_del add 100 "" "$tundev" "$filter"
 }
 
 icmp_capture_uninstall()
 {
-	__icmp_capture_add_del del 100 "" "$@"
+	local tundev=$1; shift
+	local filter=$1; shift
+
+	__icmp_capture_add_del del 100 "" "$tundev" "$filter"
 }
 
 icmp6_capture_install()
 {
-	__icmp_capture_add_del add 100 v6 "$@"
+	local tundev=$1; shift
+	local filter=$1; shift
+
+	__icmp_capture_add_del add 100 v6 "$tundev" "$filter"
 }
 
 icmp6_capture_uninstall()
 {
-	__icmp_capture_add_del del 100 v6 "$@"
+	local tundev=$1; shift
+	local filter=$1; shift
+
+	__icmp_capture_add_del del 100 v6 "$tundev" "$filter"
 }
 
 __vlan_capture_add_del()
@@ -1290,12 +1302,18 @@ __vlan_capture_add_del()
 
 vlan_capture_install()
 {
-	__vlan_capture_add_del add 100 "$@"
+	local dev=$1; shift
+	local filter=$1; shift
+
+	__vlan_capture_add_del add 100 "$dev" "$filter"
 }
 
 vlan_capture_uninstall()
 {
-	__vlan_capture_add_del del 100 "$@"
+	local dev=$1; shift
+	local filter=$1; shift
+
+	__vlan_capture_add_del del 100 "$dev" "$filter"
 }
 
 __dscp_capture_add_del()
@@ -1655,34 +1673,61 @@ __start_traffic()
 	local sip=$1; shift
 	local dip=$1; shift
 	local dmac=$1; shift
+	local -a mz_args=("$@")
 
 	$MZ $h_in -p $pktsize -A $sip -B $dip -c 0 \
-		-a own -b $dmac -t "$proto" -q "$@" &
+		-a own -b $dmac -t "$proto" -q "${mz_args[@]}" &
 	sleep 1
 }
 
 start_traffic_pktsize()
 {
 	local pktsize=$1; shift
+	local h_in=$1; shift
+	local sip=$1; shift
+	local dip=$1; shift
+	local dmac=$1; shift
+	local -a mz_args=("$@")
 
-	__start_traffic $pktsize udp "$@"
+	__start_traffic $pktsize udp "$h_in" "$sip" "$dip" "$dmac" \
+			"${mz_args[@]}"
 }
 
 start_tcp_traffic_pktsize()
 {
 	local pktsize=$1; shift
+	local h_in=$1; shift
+	local sip=$1; shift
+	local dip=$1; shift
+	local dmac=$1; shift
+	local -a mz_args=("$@")
 
-	__start_traffic $pktsize tcp "$@"
+	__start_traffic $pktsize tcp "$h_in" "$sip" "$dip" "$dmac" \
+			"${mz_args[@]}"
 }
 
 start_traffic()
 {
-	start_traffic_pktsize 8000 "$@"
+	local h_in=$1; shift
+	local sip=$1; shift
+	local dip=$1; shift
+	local dmac=$1; shift
+	local -a mz_args=("$@")
+
+	start_traffic_pktsize 8000 "$h_in" "$sip" "$dip" "$dmac" \
+			      "${mz_args[@]}"
 }
 
 start_tcp_traffic()
 {
-	start_tcp_traffic_pktsize 8000 "$@"
+	local h_in=$1; shift
+	local sip=$1; shift
+	local dip=$1; shift
+	local dmac=$1; shift
+	local -a mz_args=("$@")
+
+	start_tcp_traffic_pktsize 8000 "$h_in" "$sip" "$dip" "$dmac" \
+				  "${mz_args[@]}"
 }
 
 stop_traffic()
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh b/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh
index 09389f3b9369..6efdd944f59f 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh
@@ -64,12 +64,21 @@ cleanup()
 
 test_span_gre_dir_acl()
 {
-	test_span_gre_dir_ips "$@" 192.0.2.3 192.0.2.4
+	local tundev=$1; shift
+	local direction=$1; shift
+	local forward_type=$1; shift
+	local backward_type=$1; shift
+
+	test_span_gre_dir_ips "$tundev" "$direction" "$forward_type" \
+			      "$backward_type" 192.0.2.3 192.0.2.4
 }
 
 fail_test_span_gre_dir_acl()
 {
-	fail_test_span_gre_dir_ips "$@" 192.0.2.3 192.0.2.4
+	local tundev=$1; shift
+	local direction=$1; shift
+
+	fail_test_span_gre_dir_ips "$tundev" "$direction" 192.0.2.3 192.0.2.4
 }
 
 full_test_span_gre_dir_acl()
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
index 0c36546e131e..da2fd028e2e3 100644
--- a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
@@ -5,22 +5,34 @@ source "$net_forwarding_dir/mirror_lib.sh"
 quick_test_span_gre_dir_ips()
 {
 	local tundev=$1; shift
+	local direction=$1; shift
+	local ip1=$1; shift
+	local ip2=$1; shift
 
-	do_test_span_dir_ips 10 h3-$tundev "$@"
+	do_test_span_dir_ips 10 h3-$tundev "$direction" "$ip1" "$ip2"
 }
 
 fail_test_span_gre_dir_ips()
 {
 	local tundev=$1; shift
+	local direction=$1; shift
+	local ip1=$1; shift
+	local ip2=$1; shift
 
-	do_test_span_dir_ips 0 h3-$tundev "$@"
+	do_test_span_dir_ips 0 h3-$tundev "$direction" "$ip1" "$ip2"
 }
 
 test_span_gre_dir_ips()
 {
 	local tundev=$1; shift
+	local direction=$1; shift
+	local forward_type=$1; shift
+	local backward_type=$1; shift
+	local ip1=$1; shift
+	local ip2=$1; shift
 
-	test_span_dir_ips h3-$tundev "$@"
+	test_span_dir_ips h3-$tundev "$direction" "$forward_type" \
+			  "$backward_type" "$ip1" "$ip2"
 }
 
 full_test_span_gre_dir_ips()
@@ -74,27 +86,55 @@ full_test_span_gre_dir_vlan_ips()
 
 quick_test_span_gre_dir()
 {
-	quick_test_span_gre_dir_ips "$@" 192.0.2.1 192.0.2.2
+	local tundev=$1; shift
+	local direction=$1; shift
+
+	quick_test_span_gre_dir_ips "$tundev" "$direction" 192.0.2.1 192.0.2.2
 }
 
 fail_test_span_gre_dir()
 {
-	fail_test_span_gre_dir_ips "$@" 192.0.2.1 192.0.2.2
+	local tundev=$1; shift
+	local direction=$1; shift
+
+	fail_test_span_gre_dir_ips "$tundev" "$direction" 192.0.2.1 192.0.2.2
 }
 
 test_span_gre_dir()
 {
-	test_span_gre_dir_ips "$@" 192.0.2.1 192.0.2.2
+	local tundev=$1; shift
+	local direction=$1; shift
+	local forward_type=$1; shift
+	local backward_type=$1; shift
+
+	test_span_gre_dir_ips "$tundev" "$direction" "$forward_type" \
+			      "$backward_type" 192.0.2.1 192.0.2.2
 }
 
 full_test_span_gre_dir()
 {
-	full_test_span_gre_dir_ips "$@" 192.0.2.1 192.0.2.2
+	local tundev=$1; shift
+	local direction=$1; shift
+	local forward_type=$1; shift
+	local backward_type=$1; shift
+	local what=$1; shift
+
+	full_test_span_gre_dir_ips "$tundev" "$direction" "$forward_type" \
+				   "$backward_type" "$what" 192.0.2.1 192.0.2.2
 }
 
 full_test_span_gre_dir_vlan()
 {
-	full_test_span_gre_dir_vlan_ips "$@" 192.0.2.1 192.0.2.2
+	local tundev=$1; shift
+	local direction=$1; shift
+	local vlan_match=$1; shift
+	local forward_type=$1; shift
+	local backward_type=$1; shift
+	local what=$1; shift
+
+	full_test_span_gre_dir_vlan_ips "$tundev" "$direction" "$vlan_match" \
+					"$forward_type" "$backward_type" \
+					"$what" 192.0.2.1 192.0.2.2
 }
 
 full_test_span_gre_stp_ips()
@@ -126,5 +166,10 @@ full_test_span_gre_stp_ips()
 
 full_test_span_gre_stp()
 {
-	full_test_span_gre_stp_ips "$@" 192.0.2.1 192.0.2.2
+	local tundev=$1; shift
+	local nbpdev=$1; shift
+	local what=$1; shift
+
+	full_test_span_gre_stp_ips "$tundev" "$nbpdev" "$what" \
+				   192.0.2.1 192.0.2.2
 }
diff --git a/tools/testing/selftests/net/forwarding/mirror_lib.sh b/tools/testing/selftests/net/forwarding/mirror_lib.sh
index 3e8ebeff3019..b5b0a6545d75 100644
--- a/tools/testing/selftests/net/forwarding/mirror_lib.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_lib.sh
@@ -71,12 +71,22 @@ do_test_span_dir_ips()
 
 quick_test_span_dir_ips()
 {
-	do_test_span_dir_ips 10 "$@"
+	local dev=$1; shift
+	local direction=$1; shift
+	local ip1=$1; shift
+	local ip2=$1; shift
+
+	do_test_span_dir_ips 10 "$dev" "$direction" "$ip1" "$ip2"
 }
 
 fail_test_span_dir_ips()
 {
-	do_test_span_dir_ips 0 "$@"
+	local dev=$1; shift
+	local direction=$1; shift
+	local ip1=$1; shift
+	local ip2=$1; shift
+
+	do_test_span_dir_ips 0 "$dev" "$direction" "$ip1" "$ip2"
 }
 
 test_span_dir_ips()
@@ -101,12 +111,21 @@ test_span_dir_ips()
 
 fail_test_span_dir()
 {
-	fail_test_span_dir_ips "$@" 192.0.2.1 192.0.2.2
+	local dev=$1; shift
+	local direction=$1; shift
+
+	fail_test_span_dir_ips "$dev" "$direction" 192.0.2.1 192.0.2.2
 }
 
 test_span_dir()
 {
-	test_span_dir_ips "$@" 192.0.2.1 192.0.2.2
+	local dev=$1; shift
+	local direction=$1; shift
+	local forward_type=$1; shift
+	local backward_type=$1; shift
+
+	test_span_dir_ips "$dev" "$direction" "$forward_type" "$backward_type" \
+			  192.0.2.1 192.0.2.2
 }
 
 do_test_span_vlan_dir_ips()
@@ -130,20 +149,48 @@ do_test_span_vlan_dir_ips()
 
 quick_test_span_vlan_dir_ips()
 {
-	do_test_span_vlan_dir_ips 10 "$@"
+	local dev=$1; shift
+	local vid=$1; shift
+	local direction=$1; shift
+	local ul_proto=$1; shift
+	local ip1=$1; shift
+	local ip2=$1; shift
+
+	do_test_span_vlan_dir_ips 10 "$dev" "$vid" "$direction" "$ul_proto" \
+				  "$ip1" "$ip2"
 }
 
 fail_test_span_vlan_dir_ips()
 {
-	do_test_span_vlan_dir_ips 0 "$@"
+	local dev=$1; shift
+	local vid=$1; shift
+	local direction=$1; shift
+	local ul_proto=$1; shift
+	local ip1=$1; shift
+	local ip2=$1; shift
+
+	do_test_span_vlan_dir_ips 0 "$dev" "$vid" "$direction" "$ul_proto" \
+				  "$ip1" "$ip2"
 }
 
 quick_test_span_vlan_dir()
 {
-	quick_test_span_vlan_dir_ips "$@" 192.0.2.1 192.0.2.2
+	local dev=$1; shift
+	local vid=$1; shift
+	local direction=$1; shift
+	local ul_proto=$1; shift
+
+	quick_test_span_vlan_dir_ips "$dev" "$vid" "$direction" "$ul_proto" \
+				     192.0.2.1 192.0.2.2
 }
 
 fail_test_span_vlan_dir()
 {
-	fail_test_span_vlan_dir_ips "$@" 192.0.2.1 192.0.2.2
+	local dev=$1; shift
+	local vid=$1; shift
+	local direction=$1; shift
+	local ul_proto=$1; shift
+
+	fail_test_span_vlan_dir_ips "$dev" "$vid" "$direction" "$ul_proto" \
+				    192.0.2.1 192.0.2.2
 }
-- 
cgit v1.2.3-58-ga151


From 28e67746b73da72f57609c7250d514051e787e65 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:39 +0200
Subject: selftests: mirror: Drop direction argument from several functions

The argument is not used by these functions except to propagate it for
ultimately no purpose.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/drivers/net/mlxsw/mirror_gre.sh      | 12 ++++-----
 .../testing/selftests/net/forwarding/mirror_gre.sh | 12 ++++-----
 .../selftests/net/forwarding/mirror_gre_changes.sh | 24 ++++++++---------
 .../selftests/net/forwarding/mirror_gre_flower.sh  | 13 ++++------
 .../selftests/net/forwarding/mirror_gre_lib.sh     | 25 ++++++++----------
 .../selftests/net/forwarding/mirror_gre_neigh.sh   |  4 +--
 .../selftests/net/forwarding/mirror_gre_nh.sh      |  6 ++---
 .../net/forwarding/mirror_gre_vlan_bridge_1q.sh    | 30 +++++++++++-----------
 .../testing/selftests/net/forwarding/mirror_lib.sh | 25 +++++-------------
 .../selftests/net/forwarding/mirror_vlan.sh        |  8 +++---
 10 files changed, 69 insertions(+), 90 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh b/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
index 76f1ab4898d9..47be3b269088 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
@@ -119,11 +119,11 @@ test_span_gre_ttl_inherit()
 
 	ip link set dev $tundev type $type ttl inherit
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	fail_test_span_gre_dir $tundev ingress
+	fail_test_span_gre_dir $tundev
 
 	ip link set dev $tundev type $type ttl 100
 
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
 	log_test "$what: no offload on TTL of inherit ($tcflags)"
@@ -139,10 +139,10 @@ test_span_gre_tos_fixed()
 
 	ip link set dev $tundev type $type tos 0x10
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	fail_test_span_gre_dir $tundev ingress
+	fail_test_span_gre_dir $tundev
 
 	ip link set dev $tundev type $type tos inherit
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
 	log_test "$what: no offload on a fixed TOS ($tcflags)"
@@ -158,9 +158,9 @@ test_span_failable()
 
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
 	if ((should_fail)); then
-	    fail_test_span_gre_dir  $tundev ingress
+	    fail_test_span_gre_dir  $tundev
 	else
-	    quick_test_span_gre_dir $tundev ingress
+	    quick_test_span_gre_dir $tundev
 	fi
 	mirror_uninstall $swp1 ingress
 
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre.sh b/tools/testing/selftests/net/forwarding/mirror_gre.sh
index 0266443601bc..00def4b04145 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre.sh
@@ -91,17 +91,17 @@ test_two_spans()
 
 	mirror_install $swp1 ingress gt4 "matchall $tcflags"
 	mirror_install $swp1 egress gt6 "matchall $tcflags"
-	quick_test_span_gre_dir gt4 ingress
-	quick_test_span_gre_dir gt6 egress
+	quick_test_span_gre_dir gt4
+	quick_test_span_gre_dir gt6
 
 	mirror_uninstall $swp1 ingress
-	fail_test_span_gre_dir gt4 ingress
-	quick_test_span_gre_dir gt6 egress
+	fail_test_span_gre_dir gt4
+	quick_test_span_gre_dir gt6
 
 	mirror_install $swp1 ingress gt4 "matchall $tcflags"
 	mirror_uninstall $swp1 egress
-	quick_test_span_gre_dir gt4 ingress
-	fail_test_span_gre_dir gt6 egress
+	quick_test_span_gre_dir gt4
+	fail_test_span_gre_dir gt6
 
 	mirror_uninstall $swp1 ingress
 	log_test "two simultaneously configured mirrors ($tcflags)"
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh b/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh
index 5ea9d63915f7..3796f4256c9c 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh
@@ -99,11 +99,11 @@ test_span_gre_tun_up()
 
 	ip link set dev $tundev down
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	fail_test_span_gre_dir $tundev ingress
+	fail_test_span_gre_dir $tundev
 
 	ip link set dev $tundev up
 
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
 	log_test "$what: tunnel down/up ($tcflags)"
@@ -119,7 +119,7 @@ test_span_gre_egress_up()
 
 	ip link set dev $swp3 down
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	fail_test_span_gre_dir $tundev ingress
+	fail_test_span_gre_dir $tundev
 
 	# After setting the device up, wait for neighbor to get resolved so that
 	# we can expect mirroring to work.
@@ -127,7 +127,7 @@ test_span_gre_egress_up()
 	setup_wait_dev $swp3
 	ping -c 1 -I $swp3 $remote_ip &>/dev/null
 
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
 	log_test "$what: egress down/up ($tcflags)"
@@ -145,10 +145,10 @@ test_span_gre_remote_ip()
 
 	ip link set dev $tundev type $type remote $wrong_ip
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	fail_test_span_gre_dir $tundev ingress
+	fail_test_span_gre_dir $tundev
 
 	ip link set dev $tundev type $type remote $correct_ip
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
 	log_test "$what: remote address change ($tcflags)"
@@ -166,9 +166,9 @@ test_span_gre_tun_del()
 	RET=0
 
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 	ip link del dev $tundev
-	fail_test_span_gre_dir $tundev ingress
+	fail_test_span_gre_dir $tundev
 
 	tunnel_create $tundev $type $local_ip $remote_ip \
 		      ttl 100 tos inherit $flags
@@ -177,7 +177,7 @@ test_span_gre_tun_del()
 	# and verify it works for the follow-up tests.
 	mirror_uninstall $swp1 ingress
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
 	log_test "$what: tunnel deleted ($tcflags)"
@@ -193,13 +193,13 @@ test_span_gre_route_del()
 	RET=0
 
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 
 	ip route del $route dev $edev
-	fail_test_span_gre_dir $tundev ingress
+	fail_test_span_gre_dir $tundev
 
 	ip route add $route dev $edev
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 
 	mirror_uninstall $swp1 ingress
 
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh b/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh
index 6efdd944f59f..c7d532e6f31e 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh
@@ -65,20 +65,18 @@ cleanup()
 test_span_gre_dir_acl()
 {
 	local tundev=$1; shift
-	local direction=$1; shift
 	local forward_type=$1; shift
 	local backward_type=$1; shift
 
-	test_span_gre_dir_ips "$tundev" "$direction" "$forward_type" \
+	test_span_gre_dir_ips "$tundev" "$forward_type" \
 			      "$backward_type" 192.0.2.3 192.0.2.4
 }
 
 fail_test_span_gre_dir_acl()
 {
 	local tundev=$1; shift
-	local direction=$1; shift
 
-	fail_test_span_gre_dir_ips "$tundev" "$direction" 192.0.2.3 192.0.2.4
+	fail_test_span_gre_dir_ips "$tundev" 192.0.2.3 192.0.2.4
 }
 
 full_test_span_gre_dir_acl()
@@ -94,13 +92,12 @@ full_test_span_gre_dir_acl()
 
 	mirror_install $swp1 $direction $tundev \
 		       "protocol ip flower $tcflags dst_ip $match_dip"
-	fail_test_span_gre_dir $tundev $direction
-	test_span_gre_dir_acl "$tundev" "$direction" \
-			  "$forward_type" "$backward_type"
+	fail_test_span_gre_dir $tundev
+	test_span_gre_dir_acl "$tundev" "$forward_type" "$backward_type"
 	mirror_uninstall $swp1 $direction
 
 	# Test lack of mirroring after ACL mirror is uninstalled.
-	fail_test_span_gre_dir_acl "$tundev" "$direction"
+	fail_test_span_gre_dir_acl "$tundev"
 
 	log_test "$direction $what ($tcflags)"
 }
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
index da2fd028e2e3..2bed2a4013aa 100644
--- a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
@@ -5,33 +5,30 @@ source "$net_forwarding_dir/mirror_lib.sh"
 quick_test_span_gre_dir_ips()
 {
 	local tundev=$1; shift
-	local direction=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
 
-	do_test_span_dir_ips 10 h3-$tundev "$direction" "$ip1" "$ip2"
+	do_test_span_dir_ips 10 h3-$tundev "$ip1" "$ip2"
 }
 
 fail_test_span_gre_dir_ips()
 {
 	local tundev=$1; shift
-	local direction=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
 
-	do_test_span_dir_ips 0 h3-$tundev "$direction" "$ip1" "$ip2"
+	do_test_span_dir_ips 0 h3-$tundev "$ip1" "$ip2"
 }
 
 test_span_gre_dir_ips()
 {
 	local tundev=$1; shift
-	local direction=$1; shift
 	local forward_type=$1; shift
 	local backward_type=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
 
-	test_span_dir_ips h3-$tundev "$direction" "$forward_type" \
+	test_span_dir_ips h3-$tundev "$forward_type" \
 			  "$backward_type" "$ip1" "$ip2"
 }
 
@@ -48,7 +45,7 @@ full_test_span_gre_dir_ips()
 	RET=0
 
 	mirror_install $swp1 $direction $tundev "matchall $tcflags"
-	test_span_dir_ips "h3-$tundev" "$direction" "$forward_type" \
+	test_span_dir_ips "h3-$tundev" "$forward_type" \
 			  "$backward_type" "$ip1" "$ip2"
 	mirror_uninstall $swp1 $direction
 
@@ -70,7 +67,7 @@ full_test_span_gre_dir_vlan_ips()
 
 	mirror_install $swp1 $direction $tundev "matchall $tcflags"
 
-	test_span_dir_ips "h3-$tundev" "$direction" "$forward_type" \
+	test_span_dir_ips "h3-$tundev" "$forward_type" \
 			  "$backward_type" "$ip1" "$ip2"
 
 	tc filter add dev $h3 ingress pref 77 prot 802.1q \
@@ -87,17 +84,15 @@ full_test_span_gre_dir_vlan_ips()
 quick_test_span_gre_dir()
 {
 	local tundev=$1; shift
-	local direction=$1; shift
 
-	quick_test_span_gre_dir_ips "$tundev" "$direction" 192.0.2.1 192.0.2.2
+	quick_test_span_gre_dir_ips "$tundev" 192.0.2.1 192.0.2.2
 }
 
 fail_test_span_gre_dir()
 {
 	local tundev=$1; shift
-	local direction=$1; shift
 
-	fail_test_span_gre_dir_ips "$tundev" "$direction" 192.0.2.1 192.0.2.2
+	fail_test_span_gre_dir_ips "$tundev" 192.0.2.1 192.0.2.2
 }
 
 test_span_gre_dir()
@@ -149,15 +144,15 @@ full_test_span_gre_stp_ips()
 	RET=0
 
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	quick_test_span_gre_dir_ips $tundev ingress $ip1 $ip2
+	quick_test_span_gre_dir_ips $tundev $ip1 $ip2
 
 	bridge link set dev $nbpdev state disabled
 	sleep 1
-	fail_test_span_gre_dir_ips $tundev ingress $ip1 $ip2
+	fail_test_span_gre_dir_ips $tundev $ip1 $ip2
 
 	bridge link set dev $nbpdev state forwarding
 	sleep 1
-	quick_test_span_gre_dir_ips $tundev ingress $ip1 $ip2
+	quick_test_span_gre_dir_ips $tundev $ip1 $ip2
 
 	mirror_uninstall $swp1 ingress
 
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh b/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh
index fc0508e40fca..0f2c38eca2a3 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh
@@ -66,9 +66,9 @@ test_span_gre_neigh()
 
 	ip neigh replace dev $swp3 $addr lladdr 00:11:22:33:44:55
 	mirror_install $swp1 $direction $tundev "matchall $tcflags"
-	fail_test_span_gre_dir $tundev ingress
+	fail_test_span_gre_dir $tundev
 	ip neigh del dev $swp3 $addr
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 $direction
 
 	log_test "$direction $what: neighbor change ($tcflags)"
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_nh.sh b/tools/testing/selftests/net/forwarding/mirror_gre_nh.sh
index 6f9ef1820e93..2011dad06cf4 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_nh.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_nh.sh
@@ -81,10 +81,10 @@ test_gretap()
 	# the traffic to tunnel remote address. Then add it and test that
 	# mirroring starts. For IPv6 we can't test this due to the limitation
 	# that routes for locally-specified IPv6 addresses can't be added.
-	fail_test_span_gre_dir gt4 ingress
+	fail_test_span_gre_dir gt4
 
 	ip route add 192.0.2.130/32 via 192.0.2.162
-	quick_test_span_gre_dir gt4 ingress
+	quick_test_span_gre_dir gt4
 	ip route del 192.0.2.130/32 via 192.0.2.162
 
 	mirror_uninstall $swp1 ingress
@@ -96,7 +96,7 @@ test_ip6gretap()
 	RET=0
 
 	mirror_install $swp1 ingress gt6 "matchall $tcflags"
-	quick_test_span_gre_dir gt6 ingress
+	quick_test_span_gre_dir gt6
 	mirror_uninstall $swp1 ingress
 
 	log_test "mirror to ip6gre with next-hop remote ($tcflags)"
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_vlan_bridge_1q.sh b/tools/testing/selftests/net/forwarding/mirror_gre_vlan_bridge_1q.sh
index c8a9b5bd841f..b49b70416e32 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_vlan_bridge_1q.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_vlan_bridge_1q.sh
@@ -154,16 +154,16 @@ test_span_gre_forbidden_cpu()
 
 	# Run the pass-test first, to prime neighbor table.
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 
 	# Now forbid the VLAN at the bridge and see it fail.
 	bridge vlan del dev br1 vid 555 self
 	sleep 1
-	fail_test_span_gre_dir $tundev ingress
+	fail_test_span_gre_dir $tundev
 
 	bridge vlan add dev br1 vid 555 self
 	sleep 1
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 
 	mirror_uninstall $swp1 ingress
 
@@ -188,17 +188,17 @@ test_span_gre_forbidden_egress()
 	RET=0
 
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 
 	bridge vlan del dev $swp3 vid 555
 	sleep 1
-	fail_test_span_gre_dir $tundev ingress
+	fail_test_span_gre_dir $tundev
 
 	bridge vlan add dev $swp3 vid 555
 	# Re-prime FDB
 	$ARPING -I br1.555 192.0.2.130 -fqc 1
 	sleep 1
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 
 	mirror_uninstall $swp1 ingress
 
@@ -225,24 +225,24 @@ test_span_gre_untagged_egress()
 
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
 
-	quick_test_span_gre_dir $tundev ingress
-	quick_test_span_vlan_dir $h3 555 ingress "$ul_proto"
+	quick_test_span_gre_dir $tundev
+	quick_test_span_vlan_dir $h3 555 "$ul_proto"
 
 	h3_addr_add_del del $h3.555
 	bridge vlan add dev $swp3 vid 555 pvid untagged
 	h3_addr_add_del add $h3
 	sleep 5
 
-	quick_test_span_gre_dir $tundev ingress
-	fail_test_span_vlan_dir $h3 555 ingress "$ul_proto"
+	quick_test_span_gre_dir $tundev
+	fail_test_span_vlan_dir $h3 555 "$ul_proto"
 
 	h3_addr_add_del del $h3
 	bridge vlan add dev $swp3 vid 555
 	h3_addr_add_del add $h3.555
 	sleep 5
 
-	quick_test_span_gre_dir $tundev ingress
-	quick_test_span_vlan_dir $h3 555 ingress "$ul_proto"
+	quick_test_span_gre_dir $tundev
+	quick_test_span_vlan_dir $h3 555 "$ul_proto"
 
 	mirror_uninstall $swp1 ingress
 
@@ -268,13 +268,13 @@ test_span_gre_fdb_roaming()
 	RET=0
 
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 
 	while ((RET == 0)); do
 		bridge fdb del dev $swp3 $h3mac vlan 555 master 2>/dev/null
 		bridge fdb add dev $swp2 $h3mac vlan 555 master static
 		sleep 1
-		fail_test_span_gre_dir $tundev ingress
+		fail_test_span_gre_dir $tundev
 
 		if ! bridge fdb sh dev $swp2 vlan 555 master \
 		    | grep -q $h3mac; then
@@ -292,7 +292,7 @@ test_span_gre_fdb_roaming()
 	# Re-prime FDB
 	$ARPING -I br1.555 192.0.2.130 -fqc 1
 	sleep 1
-	quick_test_span_gre_dir $tundev ingress
+	quick_test_span_gre_dir $tundev
 
 	mirror_uninstall $swp1 ingress
 
diff --git a/tools/testing/selftests/net/forwarding/mirror_lib.sh b/tools/testing/selftests/net/forwarding/mirror_lib.sh
index b5b0a6545d75..ee0fd71ad2fd 100644
--- a/tools/testing/selftests/net/forwarding/mirror_lib.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_lib.sh
@@ -59,7 +59,6 @@ do_test_span_dir_ips()
 {
 	local expect=$1; shift
 	local dev=$1; shift
-	local direction=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
 
@@ -72,11 +71,10 @@ do_test_span_dir_ips()
 quick_test_span_dir_ips()
 {
 	local dev=$1; shift
-	local direction=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
 
-	do_test_span_dir_ips 10 "$dev" "$direction" "$ip1" "$ip2"
+	do_test_span_dir_ips 10 "$dev" "$ip1" "$ip2"
 }
 
 fail_test_span_dir_ips()
@@ -92,13 +90,12 @@ fail_test_span_dir_ips()
 test_span_dir_ips()
 {
 	local dev=$1; shift
-	local direction=$1; shift
 	local forward_type=$1; shift
 	local backward_type=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
 
-	quick_test_span_dir_ips "$dev" "$direction" "$ip1" "$ip2"
+	quick_test_span_dir_ips "$dev" "$ip1" "$ip2"
 
 	icmp_capture_install $dev "type $forward_type"
 	mirror_test v$h1 $ip1 $ip2 $dev 100 10
@@ -120,11 +117,10 @@ fail_test_span_dir()
 test_span_dir()
 {
 	local dev=$1; shift
-	local direction=$1; shift
 	local forward_type=$1; shift
 	local backward_type=$1; shift
 
-	test_span_dir_ips "$dev" "$direction" "$forward_type" "$backward_type" \
+	test_span_dir_ips "$dev" "$forward_type" "$backward_type" \
 			  192.0.2.1 192.0.2.2
 }
 
@@ -133,7 +129,6 @@ do_test_span_vlan_dir_ips()
 	local expect=$1; shift
 	local dev=$1; shift
 	local vid=$1; shift
-	local direction=$1; shift
 	local ul_proto=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
@@ -151,36 +146,31 @@ quick_test_span_vlan_dir_ips()
 {
 	local dev=$1; shift
 	local vid=$1; shift
-	local direction=$1; shift
 	local ul_proto=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
 
-	do_test_span_vlan_dir_ips 10 "$dev" "$vid" "$direction" "$ul_proto" \
-				  "$ip1" "$ip2"
+	do_test_span_vlan_dir_ips 10 "$dev" "$vid" "$ul_proto" "$ip1" "$ip2"
 }
 
 fail_test_span_vlan_dir_ips()
 {
 	local dev=$1; shift
 	local vid=$1; shift
-	local direction=$1; shift
 	local ul_proto=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
 
-	do_test_span_vlan_dir_ips 0 "$dev" "$vid" "$direction" "$ul_proto" \
-				  "$ip1" "$ip2"
+	do_test_span_vlan_dir_ips 0 "$dev" "$vid" "$ul_proto" "$ip1" "$ip2"
 }
 
 quick_test_span_vlan_dir()
 {
 	local dev=$1; shift
 	local vid=$1; shift
-	local direction=$1; shift
 	local ul_proto=$1; shift
 
-	quick_test_span_vlan_dir_ips "$dev" "$vid" "$direction" "$ul_proto" \
+	quick_test_span_vlan_dir_ips "$dev" "$vid" "$ul_proto" \
 				     192.0.2.1 192.0.2.2
 }
 
@@ -188,9 +178,8 @@ fail_test_span_vlan_dir()
 {
 	local dev=$1; shift
 	local vid=$1; shift
-	local direction=$1; shift
 	local ul_proto=$1; shift
 
-	fail_test_span_vlan_dir_ips "$dev" "$vid" "$direction" "$ul_proto" \
+	fail_test_span_vlan_dir_ips "$dev" "$vid" "$ul_proto" \
 				    192.0.2.1 192.0.2.2
 }
diff --git a/tools/testing/selftests/net/forwarding/mirror_vlan.sh b/tools/testing/selftests/net/forwarding/mirror_vlan.sh
index 0b44e148235e..074165f46a9d 100755
--- a/tools/testing/selftests/net/forwarding/mirror_vlan.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_vlan.sh
@@ -64,7 +64,7 @@ test_vlan_dir()
 	RET=0
 
 	mirror_install $swp1 $direction $swp3.555 "matchall $tcflags"
-	test_span_dir "$h3.555" "$direction" "$forward_type" "$backward_type"
+	test_span_dir "$h3.555" "$forward_type" "$backward_type"
 	mirror_uninstall $swp1 $direction
 
 	log_test "$direction mirror to vlan ($tcflags)"
@@ -85,10 +85,8 @@ test_tagged_vlan_dir()
 	RET=0
 
 	mirror_install $swp1 $direction $swp3.555 "matchall $tcflags"
-	do_test_span_vlan_dir_ips 10 "$h3.555" 111 "$direction" ip \
-				  192.0.2.17 192.0.2.18
-	do_test_span_vlan_dir_ips  0 "$h3.555" 555 "$direction" ip \
-				  192.0.2.17 192.0.2.18
+	do_test_span_vlan_dir_ips 10 "$h3.555" 111 ip 192.0.2.17 192.0.2.18
+	do_test_span_vlan_dir_ips  0 "$h3.555" 555 ip 192.0.2.17 192.0.2.18
 	mirror_uninstall $swp1 $direction
 
 	log_test "$direction mirror tagged to vlan ($tcflags)"
-- 
cgit v1.2.3-58-ga151


From 9b5d5f272654ca4b97f39b6ef809e4edef794fdc Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:40 +0200
Subject: selftests: lib: tc_rule_stats_get(): Move default to argument
 definition

The argument $dir has a fallback value of "ingress". Move the fallback from
the usage site to the argument definition block to make the fact clearer.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/lib.sh | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/net/lib.sh b/tools/testing/selftests/net/lib.sh
index 915f319bcc8b..d0219032f773 100644
--- a/tools/testing/selftests/net/lib.sh
+++ b/tools/testing/selftests/net/lib.sh
@@ -199,10 +199,10 @@ tc_rule_stats_get()
 {
 	local dev=$1; shift
 	local pref=$1; shift
-	local dir=$1; shift
+	local dir=${1:-ingress}; shift
 	local selector=${1:-.packets}; shift
 
-	tc -j -s filter show dev $dev ${dir:-ingress} pref $pref \
+	tc -j -s filter show dev $dev $dir pref $pref \
 	    | jq ".[1].options.actions[].stats$selector"
 }
 
-- 
cgit v1.2.3-58-ga151


From 95e7b860e16dbdfb07a0f36c9f954764799a3632 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:41 +0200
Subject: selftests: mirror_gre_lag_lacp: Check counters at tunnel

The test works by sending packets through a tunnel, whence they are
forwarded to a LAG. One of the LAG children is removed from the LAG prior
to the exercise, and the test then counts how many packets pass through the
other one. The issue with this is that it counts all packets, not just the
encapsulated ones.

So instead add a second gretap endpoint to receive the sent packets, and
check reception counters there.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../net/forwarding/mirror_gre_lag_lacp.sh          | 37 +++++++++++++---------
 1 file changed, 22 insertions(+), 15 deletions(-)

diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh b/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh
index 9edf4cb104a8..f521648682bf 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh
@@ -37,8 +37,14 @@
 # |    \                                              /                       |
 # |     \____________________________________________/                        |
 # |                           |                                               |
-# |                           + lag2 (team)                                   |
-# |                             192.0.2.130/28                                |
+# |                           + lag2 (team)  ------>   + gt4-dst (gretap)     |
+# |                             192.0.2.130/28           loc=192.0.2.130      |
+# |                                                      rem=192.0.2.129      |
+# |                                                      ttl=100              |
+# |                                                      tos=inherit          |
+# |                                                                           |
+# |                                                                           |
+# |                                                                           |
 # |                                                                           |
 # +---------------------------------------------------------------------------+
 
@@ -122,16 +128,21 @@ h3_create()
 {
 	vrf_create vrf-h3
 	ip link set dev vrf-h3 up
-	tc qdisc add dev $h3 clsact
-	tc qdisc add dev $h4 clsact
 	h3_create_team
+
+	tunnel_create gt4-dst gretap 192.0.2.130 192.0.2.129 \
+		      ttl 100 tos inherit
+	ip link set dev gt4-dst master vrf-h3
+	tc qdisc add dev gt4-dst clsact
 }
 
 h3_destroy()
 {
+	tc qdisc del dev gt4-dst clsact
+	ip link set dev gt4-dst nomaster
+	tunnel_destroy gt4-dst
+
 	h3_destroy_team
-	tc qdisc del dev $h4 clsact
-	tc qdisc del dev $h3 clsact
 	ip link set dev vrf-h3 down
 	vrf_destroy vrf-h3
 }
@@ -188,18 +199,12 @@ setup_prepare()
 	h2_create
 	h3_create
 	switch_create
-
-	trap_install $h3 ingress
-	trap_install $h4 ingress
 }
 
 cleanup()
 {
 	pre_cleanup
 
-	trap_uninstall $h4 ingress
-	trap_uninstall $h3 ingress
-
 	switch_destroy
 	h3_destroy
 	h2_destroy
@@ -219,6 +224,7 @@ test_lag_slave()
 
 	mirror_install $swp1 ingress gt4 \
 		       "proto 802.1q flower vlan_id 333 $tcflags"
+	vlan_capture_install gt4-dst "vlan_ethtype ipv4 ip_proto icmp type 8"
 
 	# Move $down_dev away from the team. That will prompt change in
 	# txability of the connected device, without changing its upness. The
@@ -226,13 +232,14 @@ test_lag_slave()
 	# other slave.
 	ip link set dev $down_dev nomaster
 	sleep 2
-	mirror_test vrf-h1 192.0.2.1 192.0.2.18 $up_dev 1 10
+	mirror_test vrf-h1 192.0.2.1 192.0.2.18 gt4-dst 100 10
 
 	# Test lack of connectivity when neither slave is txable.
 	ip link set dev $up_dev nomaster
 	sleep 2
-	mirror_test vrf-h1 192.0.2.1 192.0.2.18 $h3 1 0
-	mirror_test vrf-h1 192.0.2.1 192.0.2.18 $h4 1 0
+	mirror_test vrf-h1 192.0.2.1 192.0.2.18 gt4-dst 100 0
+
+	vlan_capture_uninstall gt4-dst
 	mirror_uninstall $swp1 ingress
 
 	# Recreate H3's team device, because mlxsw, which this test is
-- 
cgit v1.2.3-58-ga151


From 833415358f341941f35b6ffc53c569db057b2ab5 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:42 +0200
Subject: selftests: mirror: do_test_span_dir_ips(): Install accurate taps

The mirroring selftests work by sending ICMP traffic between two hosts.
Along the way, this traffic is mirrored to a gretap netdevice, and counter
taps are then installed strategically along the path of the mirrored
traffic to verify the mirroring took place.

The problem with this is that besides mirroring the primary traffic, any
other service traffic is mirrored as well. At the same time, because the
tests need to work in HW-offloaded scenarios, the ability of the device to
do arbitrary packet inspection should not be taken for granted. Most tests
therefore simply use matchall, one uses flower to match on IP address.

As a result, the selftests are noisy, because besides the primary ICMP
traffic, any amount of other service traffic is mirrored as well.

However, often the counter tap is installed at the remote end of the gretap
tunnel. Since this is a SW-datapath scenario anyway, we can make the filter
arbitrarily accurate.

Thus in this patch, add parameters forward_type and backward_type to
several mirroring test helpers, as some other helpers already have. Then
change do_test_span_dir_ips() to instead of installing one generic tap and
using it for test in both directions, install the tap for each direction
separately, matching on the ICMP type given by these parameters.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../testing/selftests/net/forwarding/mirror_gre.sh | 12 +++++------
 .../selftests/net/forwarding/mirror_gre_lib.sh     | 23 +++++++++++++++++-----
 .../selftests/net/forwarding/mirror_gre_neigh.sh   | 14 +++++++------
 .../testing/selftests/net/forwarding/mirror_lib.sh | 15 +++++++++++---
 4 files changed, 44 insertions(+), 20 deletions(-)

diff --git a/tools/testing/selftests/net/forwarding/mirror_gre.sh b/tools/testing/selftests/net/forwarding/mirror_gre.sh
index 00def4b04145..b645cb8d696a 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre.sh
@@ -91,17 +91,17 @@ test_two_spans()
 
 	mirror_install $swp1 ingress gt4 "matchall $tcflags"
 	mirror_install $swp1 egress gt6 "matchall $tcflags"
-	quick_test_span_gre_dir gt4
-	quick_test_span_gre_dir gt6
+	quick_test_span_gre_dir gt4 8 0
+	quick_test_span_gre_dir gt6 0 8
 
 	mirror_uninstall $swp1 ingress
-	fail_test_span_gre_dir gt4
-	quick_test_span_gre_dir gt6
+	fail_test_span_gre_dir gt4 8 0
+	quick_test_span_gre_dir gt6 0 8
 
 	mirror_install $swp1 ingress gt4 "matchall $tcflags"
 	mirror_uninstall $swp1 egress
-	quick_test_span_gre_dir gt4
-	fail_test_span_gre_dir gt6
+	quick_test_span_gre_dir gt4 8 0
+	fail_test_span_gre_dir gt6 0 8
 
 	mirror_uninstall $swp1 ingress
 	log_test "two simultaneously configured mirrors ($tcflags)"
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
index 2bed2a4013aa..e49535ce1cdd 100644
--- a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
@@ -7,8 +7,11 @@ quick_test_span_gre_dir_ips()
 	local tundev=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
+	local forward_type=$1; shift
+	local backward_type=$1; shift
 
-	do_test_span_dir_ips 10 h3-$tundev "$ip1" "$ip2"
+	do_test_span_dir_ips 10 h3-$tundev "$ip1" "$ip2" \
+			     "$forward_type" "$backward_type"
 }
 
 fail_test_span_gre_dir_ips()
@@ -84,8 +87,11 @@ full_test_span_gre_dir_vlan_ips()
 quick_test_span_gre_dir()
 {
 	local tundev=$1; shift
+	local forward_type=${1-8}; shift
+	local backward_type=${1-0}; shift
 
-	quick_test_span_gre_dir_ips "$tundev" 192.0.2.1 192.0.2.2
+	quick_test_span_gre_dir_ips "$tundev" 192.0.2.1 192.0.2.2 \
+				    "$forward_type" "$backward_type"
 }
 
 fail_test_span_gre_dir()
@@ -139,12 +145,15 @@ full_test_span_gre_stp_ips()
 	local what=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
+	local forward_type=$1; shift
+	local backward_type=$1; shift
 	local h3mac=$(mac_get $h3)
 
 	RET=0
 
 	mirror_install $swp1 ingress $tundev "matchall $tcflags"
-	quick_test_span_gre_dir_ips $tundev $ip1 $ip2
+	quick_test_span_gre_dir_ips $tundev $ip1 $ip2 \
+				    "$forward_type" "$backward_type"
 
 	bridge link set dev $nbpdev state disabled
 	sleep 1
@@ -152,7 +161,8 @@ full_test_span_gre_stp_ips()
 
 	bridge link set dev $nbpdev state forwarding
 	sleep 1
-	quick_test_span_gre_dir_ips $tundev $ip1 $ip2
+	quick_test_span_gre_dir_ips $tundev $ip1 $ip2 \
+				    "$forward_type" "$backward_type"
 
 	mirror_uninstall $swp1 ingress
 
@@ -164,7 +174,10 @@ full_test_span_gre_stp()
 	local tundev=$1; shift
 	local nbpdev=$1; shift
 	local what=$1; shift
+	local forward_type=${1-8}; shift
+	local backward_type=${1-0}; shift
 
 	full_test_span_gre_stp_ips "$tundev" "$nbpdev" "$what" \
-				   192.0.2.1 192.0.2.2
+				   192.0.2.1 192.0.2.2 \
+				   "$forward_type" "$backward_type"
 }
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh b/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh
index 0f2c38eca2a3..0615f9244406 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh
@@ -60,15 +60,17 @@ test_span_gre_neigh()
 	local addr=$1; shift
 	local tundev=$1; shift
 	local direction=$1; shift
+	local forward_type=$1; shift
+	local backward_type=$1; shift
 	local what=$1; shift
 
 	RET=0
 
 	ip neigh replace dev $swp3 $addr lladdr 00:11:22:33:44:55
 	mirror_install $swp1 $direction $tundev "matchall $tcflags"
-	fail_test_span_gre_dir $tundev
+	fail_test_span_gre_dir $tundev "$forward_type" "$backward_type"
 	ip neigh del dev $swp3 $addr
-	quick_test_span_gre_dir $tundev
+	quick_test_span_gre_dir $tundev "$forward_type" "$backward_type"
 	mirror_uninstall $swp1 $direction
 
 	log_test "$direction $what: neighbor change ($tcflags)"
@@ -76,14 +78,14 @@ test_span_gre_neigh()
 
 test_gretap()
 {
-	test_span_gre_neigh 192.0.2.130 gt4 ingress "mirror to gretap"
-	test_span_gre_neigh 192.0.2.130 gt4 egress "mirror to gretap"
+	test_span_gre_neigh 192.0.2.130 gt4 ingress 8 0 "mirror to gretap"
+	test_span_gre_neigh 192.0.2.130 gt4 egress 0 8 "mirror to gretap"
 }
 
 test_ip6gretap()
 {
-	test_span_gre_neigh 2001:db8:2::2 gt6 ingress "mirror to ip6gretap"
-	test_span_gre_neigh 2001:db8:2::2 gt6 egress "mirror to ip6gretap"
+	test_span_gre_neigh 2001:db8:2::2 gt6 ingress 8 0 "mirror to ip6gretap"
+	test_span_gre_neigh 2001:db8:2::2 gt6 egress 0 8 "mirror to ip6gretap"
 }
 
 test_all()
diff --git a/tools/testing/selftests/net/forwarding/mirror_lib.sh b/tools/testing/selftests/net/forwarding/mirror_lib.sh
index ee0fd71ad2fd..6a0c66d7ba7f 100644
--- a/tools/testing/selftests/net/forwarding/mirror_lib.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_lib.sh
@@ -61,9 +61,14 @@ do_test_span_dir_ips()
 	local dev=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
+	local forward_type=${1-8}; shift
+	local backward_type=${1-0}; shift
 
-	icmp_capture_install $dev
+	icmp_capture_install $dev "type $forward_type"
 	mirror_test v$h1 $ip1 $ip2 $dev 100 $expect
+	icmp_capture_uninstall $dev
+
+	icmp_capture_install $dev "type $backward_type"
 	mirror_test v$h2 $ip2 $ip1 $dev 100 $expect
 	icmp_capture_uninstall $dev
 }
@@ -73,8 +78,11 @@ quick_test_span_dir_ips()
 	local dev=$1; shift
 	local ip1=$1; shift
 	local ip2=$1; shift
+	local forward_type=${1-8}; shift
+	local backward_type=${1-0}; shift
 
-	do_test_span_dir_ips 10 "$dev" "$ip1" "$ip2"
+	do_test_span_dir_ips 10 "$dev" "$ip1" "$ip2" \
+			     "$forward_type" "$backward_type"
 }
 
 fail_test_span_dir_ips()
@@ -95,7 +103,8 @@ test_span_dir_ips()
 	local ip1=$1; shift
 	local ip2=$1; shift
 
-	quick_test_span_dir_ips "$dev" "$ip1" "$ip2"
+	quick_test_span_dir_ips "$dev" "$ip1" "$ip2" \
+				"$forward_type" "$backward_type"
 
 	icmp_capture_install $dev "type $forward_type"
 	mirror_test v$h1 $ip1 $ip2 $dev 100 10
-- 
cgit v1.2.3-58-ga151


From a86e0df9ce25d33e84ba26fa2876a2e67b8d48ff Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:43 +0200
Subject: selftests: mirror: mirror_test(): Allow exact count of packets

The mirroring selftests work by sending ICMP traffic between two hosts.
Along the way, this traffic is mirrored to a gretap netdevice, and counter
taps are then installed strategically along the path of the mirrored
traffic to verify the mirroring took place.

The problem with this is that besides mirroring the primary traffic, any
other service traffic is mirrored as well. At the same time, because the
tests need to work in HW-offloaded scenarios, the ability of the device to
do arbitrary packet inspection should not be taken for granted. Most tests
therefore simply use matchall, one uses flower to match on IP address.

As a result, the selftests are noisy, because besides the primary ICMP
traffic, any amount of other service traffic is mirrored as well.

mirror_test() accommodated this noisiness by giving the counters an
allowance of several packets. But in the previous patch, where possible,
counter taps were changed to match only on an exact ICMP message. At least
in those cases, we can demand an exact number of packets to match.

Where the tap is installed on a connective netdevice, the exact matching is
not practical (though with u32, anything is possible). In those places,
there should still be some leeway -- and probably bigger than before,
because experience shows that these tests are very noisy.

To that end, change mirror_test() so that it can be either called with an
exact number to expect, or with an expression. Where leeway is needed,
adjust callers to pass a ">= 10" instead of mere 10.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/net/forwarding/mirror_gre_bridge_1q_lag.sh   |  2 +-
 .../testing/selftests/net/forwarding/mirror_gre_changes.sh |  2 +-
 tools/testing/selftests/net/forwarding/mirror_gre_lib.sh   |  2 +-
 tools/testing/selftests/net/forwarding/mirror_lib.sh       | 14 +++++++++-----
 tools/testing/selftests/net/forwarding/mirror_vlan.sh      |  2 +-
 5 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q_lag.sh b/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q_lag.sh
index c53148b1dc63..b719d24a4ae5 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q_lag.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q_lag.sh
@@ -239,7 +239,7 @@ test_lag_slave()
 	setup_wait_dev $host_dev
 	$ARPING -I br1 192.0.2.130 -qfc 1
 	sleep 2
-	mirror_test vrf-h1 192.0.2.1 192.0.2.18 $host_dev 1 10
+	mirror_test vrf-h1 192.0.2.1 192.0.2.18 $host_dev 1 ">= 10"
 
 	# Test lack of connectivity when both slaves are down.
 	ip link set dev $up_dev down
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh b/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh
index 3796f4256c9c..b57fb9f069f4 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh
@@ -81,7 +81,7 @@ test_span_gre_ttl()
 
 	ip link set dev $tundev type $type ttl 50
 	sleep 2
-	mirror_test v$h1 192.0.2.1 192.0.2.2 $h3 77 10
+	mirror_test v$h1 192.0.2.1 192.0.2.2 $h3 77 ">= 10"
 
 	ip link set dev $tundev type $type ttl 100
 	tc filter del dev $h3 ingress pref 77
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
index e49535ce1cdd..cc3a0a3f83ba 100644
--- a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
@@ -76,7 +76,7 @@ full_test_span_gre_dir_vlan_ips()
 	tc filter add dev $h3 ingress pref 77 prot 802.1q \
 		flower $vlan_match \
 		action pass
-	mirror_test v$h1 $ip1 $ip2 $h3 77 10
+	mirror_test v$h1 $ip1 $ip2 $h3 77 '>= 10'
 	tc filter del dev $h3 ingress pref 77
 
 	mirror_uninstall $swp1 $direction
diff --git a/tools/testing/selftests/net/forwarding/mirror_lib.sh b/tools/testing/selftests/net/forwarding/mirror_lib.sh
index 6a0c66d7ba7f..1fc8545da6e0 100644
--- a/tools/testing/selftests/net/forwarding/mirror_lib.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_lib.sh
@@ -44,14 +44,17 @@ mirror_test()
 		local type="icmp echoreq"
 	fi
 
+	if [[ -z ${expect//[[:digit:]]/} ]]; then
+		expect="== $expect"
+	fi
+
 	local t0=$(tc_rule_stats_get $dev $pref)
 	$MZ $proto $vrf_name ${sip:+-A $sip} -B $dip -a own -b bc -q \
 	    -c 10 -d 100msec -t $type
 	sleep 0.5
 	local t1=$(tc_rule_stats_get $dev $pref)
 	local delta=$((t1 - t0))
-	# Tolerate a couple stray extra packets.
-	((expect <= delta && delta <= expect + 2))
+	((delta $expect))
 	check_err $? "Expected to capture $expect packets, got $delta."
 }
 
@@ -146,8 +149,8 @@ do_test_span_vlan_dir_ips()
 	# The traffic is meant for local box anyway, so will be trapped to
 	# kernel.
 	vlan_capture_install $dev "skip_hw vlan_id $vid vlan_ethtype $ul_proto"
-	mirror_test v$h1 $ip1 $ip2 $dev 100 $expect
-	mirror_test v$h2 $ip2 $ip1 $dev 100 $expect
+	mirror_test v$h1 $ip1 $ip2 $dev 100 "$expect"
+	mirror_test v$h2 $ip2 $ip1 $dev 100 "$expect"
 	vlan_capture_uninstall $dev
 }
 
@@ -159,7 +162,8 @@ quick_test_span_vlan_dir_ips()
 	local ip1=$1; shift
 	local ip2=$1; shift
 
-	do_test_span_vlan_dir_ips 10 "$dev" "$vid" "$ul_proto" "$ip1" "$ip2"
+	do_test_span_vlan_dir_ips '>= 10' "$dev" "$vid" "$ul_proto" \
+				  "$ip1" "$ip2"
 }
 
 fail_test_span_vlan_dir_ips()
diff --git a/tools/testing/selftests/net/forwarding/mirror_vlan.sh b/tools/testing/selftests/net/forwarding/mirror_vlan.sh
index 074165f46a9d..bc3297846511 100755
--- a/tools/testing/selftests/net/forwarding/mirror_vlan.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_vlan.sh
@@ -85,7 +85,7 @@ test_tagged_vlan_dir()
 	RET=0
 
 	mirror_install $swp1 $direction $swp3.555 "matchall $tcflags"
-	do_test_span_vlan_dir_ips 10 "$h3.555" 111 ip 192.0.2.17 192.0.2.18
+	do_test_span_vlan_dir_ips '>= 10' "$h3.555" 111 ip 192.0.2.17 192.0.2.18
 	do_test_span_vlan_dir_ips  0 "$h3.555" 555 ip 192.0.2.17 192.0.2.18
 	mirror_uninstall $swp1 $direction
 
-- 
cgit v1.2.3-58-ga151


From d361d78fe2cc8a7a60e8c71825f84e70d34938af Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:44 +0200
Subject: selftests: mirror: Drop dual SW/HW testing

The mirroring tests are currently run in a skip_hw and optionally a skip_sw
mode. The former tests the SW datapath, the latter the HW datapath, if
available. In order to be able to test SW datapath on HW loopbacks, traps
are installed on ingress to get traffic from the HW datapath to the SW one.
This adds an unnecessary complexity when it would be much simpler to just
use a veth-based topology to test the SW datapath. Thus drop all the code
that supports this dual testing.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/drivers/net/mlxsw/mirror_gre.sh      | 33 +++------------
 .../drivers/net/mlxsw/mirror_gre_scale.sh          | 18 +--------
 .../testing/selftests/net/forwarding/mirror_gre.sh | 33 ++++-----------
 .../selftests/net/forwarding/mirror_gre_bound.sh   | 23 +----------
 .../net/forwarding/mirror_gre_bridge_1d.sh         | 21 +---------
 .../net/forwarding/mirror_gre_bridge_1d_vlan.sh    | 21 +---------
 .../net/forwarding/mirror_gre_bridge_1q.sh         | 21 +---------
 .../net/forwarding/mirror_gre_bridge_1q_lag.sh     | 27 ++-----------
 .../selftests/net/forwarding/mirror_gre_changes.sh | 47 +++++++---------------
 .../selftests/net/forwarding/mirror_gre_flower.sh  | 25 ++----------
 .../net/forwarding/mirror_gre_lag_lacp.sh          | 25 ++----------
 .../selftests/net/forwarding/mirror_gre_lib.sh     | 12 +++---
 .../selftests/net/forwarding/mirror_gre_neigh.sh   | 25 ++----------
 .../selftests/net/forwarding/mirror_gre_nh.sh      | 29 +++----------
 .../selftests/net/forwarding/mirror_gre_vlan.sh    | 21 +---------
 .../net/forwarding/mirror_gre_vlan_bridge_1q.sh    | 39 +++++-------------
 .../selftests/net/forwarding/mirror_vlan.sh        | 35 +++++-----------
 17 files changed, 77 insertions(+), 378 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh b/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
index 47be3b269088..8cf06fcd724c 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
@@ -118,7 +118,7 @@ test_span_gre_ttl_inherit()
 	RET=0
 
 	ip link set dev $tundev type $type ttl inherit
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	fail_test_span_gre_dir $tundev
 
 	ip link set dev $tundev type $type ttl 100
@@ -126,7 +126,7 @@ test_span_gre_ttl_inherit()
 	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: no offload on TTL of inherit ($tcflags)"
+	log_test "$what: no offload on TTL of inherit"
 }
 
 test_span_gre_tos_fixed()
@@ -138,14 +138,14 @@ test_span_gre_tos_fixed()
 	RET=0
 
 	ip link set dev $tundev type $type tos 0x10
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	fail_test_span_gre_dir $tundev
 
 	ip link set dev $tundev type $type tos inherit
 	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: no offload on a fixed TOS ($tcflags)"
+	log_test "$what: no offload on a fixed TOS"
 }
 
 test_span_failable()
@@ -156,7 +156,7 @@ test_span_failable()
 
 	RET=0
 
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	if ((should_fail)); then
 	    fail_test_span_gre_dir  $tundev
 	else
@@ -164,7 +164,7 @@ test_span_failable()
 	fi
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: should_fail=$should_fail ($tcflags)"
+	log_test "$what: should_fail=$should_fail"
 }
 
 test_failable()
@@ -175,17 +175,6 @@ test_failable()
 	test_span_failable $should_fail gt6-soft "mirror to gretap w/ soft underlay"
 }
 
-test_sw()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	test_failable 0
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 test_hw()
 {
 	test_failable 1
@@ -202,16 +191,6 @@ trap cleanup EXIT
 setup_prepare
 setup_wait
 
-if ! tc_offload_check; then
-    check_err 1 "Could not test offloaded functionality"
-    log_test "mlxsw-specific tests for mirror to gretap"
-    exit
-fi
-
-tcflags="skip_hw"
-test_sw
-
-tcflags="skip_sw"
 test_hw
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/drivers/net/mlxsw/mirror_gre_scale.sh b/tools/testing/selftests/drivers/net/mlxsw/mirror_gre_scale.sh
index e5589e2fca85..d43093310e23 100644
--- a/tools/testing/selftests/drivers/net/mlxsw/mirror_gre_scale.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/mirror_gre_scale.sh
@@ -79,7 +79,7 @@ mirror_gre_tunnels_create()
 		cat >> $MIRROR_GRE_BATCH_FILE <<-EOF
 			filter add dev $swp1 ingress pref 1000 \
 				protocol ipv6 \
-				flower $tcflags dst_ip $match_dip \
+				flower skip_sw dst_ip $match_dip \
 				action mirred egress mirror dev $tun
 		EOF
 	done
@@ -107,7 +107,7 @@ mirror_gre_tunnels_destroy()
 	done
 }
 
-__mirror_gre_test()
+mirror_gre_test()
 {
 	local count=$1; shift
 	local should_fail=$1; shift
@@ -131,20 +131,6 @@ __mirror_gre_test()
 	done
 }
 
-mirror_gre_test()
-{
-	local count=$1; shift
-	local should_fail=$1; shift
-
-	if ! tc_offload_check $TC_FLOWER_NUM_NETIFS; then
-		check_err 1 "Could not test offloaded functionality"
-		return
-	fi
-
-	tcflags="skip_sw"
-	__mirror_gre_test $count $should_fail
-}
-
 mirror_gre_setup_prepare()
 {
 	h1=${NETIFS[p1]}
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre.sh b/tools/testing/selftests/net/forwarding/mirror_gre.sh
index b645cb8d696a..921c733ee04f 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre.sh
@@ -74,7 +74,7 @@ test_span_gre_mac()
 
 	RET=0
 
-	mirror_install $swp1 $direction $tundev "matchall $tcflags"
+	mirror_install $swp1 $direction $tundev "matchall"
 	icmp_capture_install h3-${tundev} "src_mac $src_mac dst_mac $dst_mac"
 
 	mirror_test v$h1 192.0.2.1 192.0.2.2 h3-${tundev} 100 10
@@ -82,15 +82,15 @@ test_span_gre_mac()
 	icmp_capture_uninstall h3-${tundev}
 	mirror_uninstall $swp1 $direction
 
-	log_test "$direction $what: envelope MAC ($tcflags)"
+	log_test "$direction $what: envelope MAC"
 }
 
 test_two_spans()
 {
 	RET=0
 
-	mirror_install $swp1 ingress gt4 "matchall $tcflags"
-	mirror_install $swp1 egress gt6 "matchall $tcflags"
+	mirror_install $swp1 ingress gt4 "matchall"
+	mirror_install $swp1 egress gt6 "matchall"
 	quick_test_span_gre_dir gt4 8 0
 	quick_test_span_gre_dir gt6 0 8
 
@@ -98,13 +98,13 @@ test_two_spans()
 	fail_test_span_gre_dir gt4 8 0
 	quick_test_span_gre_dir gt6 0 8
 
-	mirror_install $swp1 ingress gt4 "matchall $tcflags"
+	mirror_install $swp1 ingress gt4 "matchall"
 	mirror_uninstall $swp1 egress
 	quick_test_span_gre_dir gt4 8 0
 	fail_test_span_gre_dir gt6 0 8
 
 	mirror_uninstall $swp1 ingress
-	log_test "two simultaneously configured mirrors ($tcflags)"
+	log_test "two simultaneously configured mirrors"
 }
 
 test_gretap()
@@ -131,30 +131,11 @@ test_ip6gretap_mac()
 	test_span_gre_mac gt6 egress "mirror to ip6gretap"
 }
 
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_bound.sh b/tools/testing/selftests/net/forwarding/mirror_gre_bound.sh
index 6c257ec03756..e3cd48e18eeb 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_bound.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_bound.sh
@@ -196,32 +196,11 @@ test_ip6gretap()
 	full_test_span_gre_dir gt6 egress  0 8 "mirror to ip6gretap w/ UL"
 }
 
-test_all()
-{
-	RET=0
-
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1d.sh b/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1d.sh
index 04fd14b0a9b7..6c7bd33332c2 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1d.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1d.sh
@@ -108,30 +108,11 @@ test_ip6gretap()
 	full_test_span_gre_dir gt6 egress 0 8 "mirror to ip6gretap"
 }
 
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1d_vlan.sh b/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1d_vlan.sh
index f35313c76fac..909ec956a5e5 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1d_vlan.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1d_vlan.sh
@@ -104,30 +104,11 @@ test_ip6gretap_stp()
 	full_test_span_gre_stp gt6 $swp3.555 "mirror to ip6gretap"
 }
 
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q.sh b/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q.sh
index 0cf4c47a46f9..40ac9dd3aff1 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q.sh
@@ -104,30 +104,11 @@ test_ip6gretap()
 	full_test_span_gre_dir gt6 egress 0 8 "mirror to ip6gretap"
 }
 
-tests()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-tests
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	tests
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q_lag.sh b/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q_lag.sh
index b719d24a4ae5..fe4d7c906a70 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q_lag.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_bridge_1q_lag.sh
@@ -227,10 +227,10 @@ test_lag_slave()
 	RET=0
 
 	tc filter add dev $swp1 ingress pref 999 \
-		proto 802.1q flower vlan_ethtype arp $tcflags \
+		proto 802.1q flower vlan_ethtype arp \
 		action pass
 	mirror_install $swp1 ingress gt4 \
-		"proto 802.1q flower vlan_id 333 $tcflags"
+		"proto 802.1q flower vlan_id 333"
 
 	# Test connectivity through $up_dev when $down_dev is set down.
 	ip link set dev $down_dev down
@@ -252,7 +252,7 @@ test_lag_slave()
 	mirror_uninstall $swp1 ingress
 	tc filter del dev $swp1 ingress pref 999
 
-	log_test "$what ($tcflags)"
+	log_test "$what"
 }
 
 test_mirror_gretap_first()
@@ -265,30 +265,11 @@ test_mirror_gretap_second()
 	test_lag_slave $h4 $swp4 $swp3 "mirror to gretap: LAG second slave"
 }
 
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh b/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh
index b57fb9f069f4..65ae9d960c18 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_changes.sh
@@ -73,7 +73,7 @@ test_span_gre_ttl()
 	RET=0
 
 	mirror_install $swp1 ingress $tundev \
-		"prot ip flower $tcflags ip_prot icmp"
+		"prot ip flower ip_prot icmp"
 	tc filter add dev $h3 ingress pref 77 prot $prot \
 		flower skip_hw ip_ttl 50 action pass
 
@@ -87,7 +87,7 @@ test_span_gre_ttl()
 	tc filter del dev $h3 ingress pref 77
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: TTL change ($tcflags)"
+	log_test "$what: TTL change"
 }
 
 test_span_gre_tun_up()
@@ -98,7 +98,7 @@ test_span_gre_tun_up()
 	RET=0
 
 	ip link set dev $tundev down
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	fail_test_span_gre_dir $tundev
 
 	ip link set dev $tundev up
@@ -106,7 +106,7 @@ test_span_gre_tun_up()
 	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: tunnel down/up ($tcflags)"
+	log_test "$what: tunnel down/up"
 }
 
 test_span_gre_egress_up()
@@ -118,7 +118,7 @@ test_span_gre_egress_up()
 	RET=0
 
 	ip link set dev $swp3 down
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	fail_test_span_gre_dir $tundev
 
 	# After setting the device up, wait for neighbor to get resolved so that
@@ -130,7 +130,7 @@ test_span_gre_egress_up()
 	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: egress down/up ($tcflags)"
+	log_test "$what: egress down/up"
 }
 
 test_span_gre_remote_ip()
@@ -144,14 +144,14 @@ test_span_gre_remote_ip()
 	RET=0
 
 	ip link set dev $tundev type $type remote $wrong_ip
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	fail_test_span_gre_dir $tundev
 
 	ip link set dev $tundev type $type remote $correct_ip
 	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: remote address change ($tcflags)"
+	log_test "$what: remote address change"
 }
 
 test_span_gre_tun_del()
@@ -165,7 +165,7 @@ test_span_gre_tun_del()
 
 	RET=0
 
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	quick_test_span_gre_dir $tundev
 	ip link del dev $tundev
 	fail_test_span_gre_dir $tundev
@@ -176,11 +176,11 @@ test_span_gre_tun_del()
 	# Recreating the tunnel doesn't reestablish mirroring, so reinstall it
 	# and verify it works for the follow-up tests.
 	mirror_uninstall $swp1 ingress
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	quick_test_span_gre_dir $tundev
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: tunnel deleted ($tcflags)"
+	log_test "$what: tunnel deleted"
 }
 
 test_span_gre_route_del()
@@ -192,7 +192,7 @@ test_span_gre_route_del()
 
 	RET=0
 
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	quick_test_span_gre_dir $tundev
 
 	ip route del $route dev $edev
@@ -203,7 +203,7 @@ test_span_gre_route_del()
 
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: underlay route removal ($tcflags)"
+	log_test "$what: underlay route removal"
 }
 
 test_ttl()
@@ -244,30 +244,11 @@ test_route_del()
 	test_span_gre_route_del gt6 $swp3 2001:db8:2::/64 "mirror to ip6gretap"
 }
 
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh b/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh
index c7d532e6f31e..3a84f3ab5856 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_flower.sh
@@ -91,7 +91,7 @@ full_test_span_gre_dir_acl()
 	RET=0
 
 	mirror_install $swp1 $direction $tundev \
-		       "protocol ip flower $tcflags dst_ip $match_dip"
+		       "protocol ip flower dst_ip $match_dip"
 	fail_test_span_gre_dir $tundev
 	test_span_gre_dir_acl "$tundev" "$forward_type" "$backward_type"
 	mirror_uninstall $swp1 $direction
@@ -99,7 +99,7 @@ full_test_span_gre_dir_acl()
 	# Test lack of mirroring after ACL mirror is uninstalled.
 	fail_test_span_gre_dir_acl "$tundev"
 
-	log_test "$direction $what ($tcflags)"
+	log_test "$direction $what"
 }
 
 test_gretap()
@@ -114,30 +114,11 @@ test_ip6gretap()
 	full_test_span_gre_dir_acl gt6 egress 0 8 192.0.2.3 "ACL mirror to ip6gretap"
 }
 
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh b/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh
index f521648682bf..1e4f0887724d 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh
@@ -223,7 +223,7 @@ test_lag_slave()
 	RET=0
 
 	mirror_install $swp1 ingress gt4 \
-		       "proto 802.1q flower vlan_id 333 $tcflags"
+		       "proto 802.1q flower vlan_id 333"
 	vlan_capture_install gt4-dst "vlan_ethtype ipv4 ip_proto icmp type 8"
 
 	# Move $down_dev away from the team. That will prompt change in
@@ -250,7 +250,7 @@ test_lag_slave()
 	# Wait for ${h,swp}{3,4}.
 	setup_wait
 
-	log_test "$what ($tcflags)"
+	log_test "$what"
 }
 
 test_mirror_gretap_first()
@@ -263,30 +263,11 @@ test_mirror_gretap_second()
 	test_lag_slave $h4 $h3 "mirror to gretap: LAG second slave"
 }
 
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
index cc3a0a3f83ba..aa96644e3866 100644
--- a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
@@ -47,12 +47,12 @@ full_test_span_gre_dir_ips()
 
 	RET=0
 
-	mirror_install $swp1 $direction $tundev "matchall $tcflags"
+	mirror_install $swp1 $direction $tundev "matchall"
 	test_span_dir_ips "h3-$tundev" "$forward_type" \
 			  "$backward_type" "$ip1" "$ip2"
 	mirror_uninstall $swp1 $direction
 
-	log_test "$direction $what ($tcflags)"
+	log_test "$direction $what"
 }
 
 full_test_span_gre_dir_vlan_ips()
@@ -68,7 +68,7 @@ full_test_span_gre_dir_vlan_ips()
 
 	RET=0
 
-	mirror_install $swp1 $direction $tundev "matchall $tcflags"
+	mirror_install $swp1 $direction $tundev "matchall"
 
 	test_span_dir_ips "h3-$tundev" "$forward_type" \
 			  "$backward_type" "$ip1" "$ip2"
@@ -81,7 +81,7 @@ full_test_span_gre_dir_vlan_ips()
 
 	mirror_uninstall $swp1 $direction
 
-	log_test "$direction $what ($tcflags)"
+	log_test "$direction $what"
 }
 
 quick_test_span_gre_dir()
@@ -151,7 +151,7 @@ full_test_span_gre_stp_ips()
 
 	RET=0
 
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	quick_test_span_gre_dir_ips $tundev $ip1 $ip2 \
 				    "$forward_type" "$backward_type"
 
@@ -166,7 +166,7 @@ full_test_span_gre_stp_ips()
 
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: STP state ($tcflags)"
+	log_test "$what: STP state"
 }
 
 full_test_span_gre_stp()
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh b/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh
index 0615f9244406..2cbfbecf25c8 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_neigh.sh
@@ -67,13 +67,13 @@ test_span_gre_neigh()
 	RET=0
 
 	ip neigh replace dev $swp3 $addr lladdr 00:11:22:33:44:55
-	mirror_install $swp1 $direction $tundev "matchall $tcflags"
+	mirror_install $swp1 $direction $tundev "matchall"
 	fail_test_span_gre_dir $tundev "$forward_type" "$backward_type"
 	ip neigh del dev $swp3 $addr
 	quick_test_span_gre_dir $tundev "$forward_type" "$backward_type"
 	mirror_uninstall $swp1 $direction
 
-	log_test "$direction $what: neighbor change ($tcflags)"
+	log_test "$direction $what: neighbor change"
 }
 
 test_gretap()
@@ -88,30 +88,11 @@ test_ip6gretap()
 	test_span_gre_neigh 2001:db8:2::2 gt6 egress 0 8 "mirror to ip6gretap"
 }
 
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_nh.sh b/tools/testing/selftests/net/forwarding/mirror_gre_nh.sh
index 2011dad06cf4..34bc646938e3 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_nh.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_nh.sh
@@ -75,7 +75,7 @@ cleanup()
 test_gretap()
 {
 	RET=0
-	mirror_install $swp1 ingress gt4 "matchall $tcflags"
+	mirror_install $swp1 ingress gt4 "matchall"
 
 	# For IPv4, test that there's no mirroring without the route directing
 	# the traffic to tunnel remote address. Then add it and test that
@@ -88,29 +88,18 @@ test_gretap()
 	ip route del 192.0.2.130/32 via 192.0.2.162
 
 	mirror_uninstall $swp1 ingress
-	log_test "mirror to gre with next-hop remote ($tcflags)"
+	log_test "mirror to gre with next-hop remote"
 }
 
 test_ip6gretap()
 {
 	RET=0
 
-	mirror_install $swp1 ingress gt6 "matchall $tcflags"
+	mirror_install $swp1 ingress gt6 "matchall"
 	quick_test_span_gre_dir gt6
 	mirror_uninstall $swp1 ingress
 
-	log_test "mirror to ip6gre with next-hop remote ($tcflags)"
-}
-
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
+	log_test "mirror to ip6gre with next-hop remote"
 }
 
 trap cleanup EXIT
@@ -118,14 +107,6 @@ trap cleanup EXIT
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_vlan.sh b/tools/testing/selftests/net/forwarding/mirror_gre_vlan.sh
index 88cecdb9a861..63689928cb51 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_vlan.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_vlan.sh
@@ -63,30 +63,11 @@ test_gretap()
 	full_test_span_gre_dir gt4 egress 0 8 "mirror to gretap"
 }
 
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_vlan_bridge_1q.sh b/tools/testing/selftests/net/forwarding/mirror_gre_vlan_bridge_1q.sh
index b49b70416e32..1b902cc579f6 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_vlan_bridge_1q.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_vlan_bridge_1q.sh
@@ -153,7 +153,7 @@ test_span_gre_forbidden_cpu()
 	RET=0
 
 	# Run the pass-test first, to prime neighbor table.
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	quick_test_span_gre_dir $tundev
 
 	# Now forbid the VLAN at the bridge and see it fail.
@@ -167,7 +167,7 @@ test_span_gre_forbidden_cpu()
 
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: vlan forbidden at a bridge ($tcflags)"
+	log_test "$what: vlan forbidden at a bridge"
 }
 
 test_gretap_forbidden_cpu()
@@ -187,7 +187,7 @@ test_span_gre_forbidden_egress()
 
 	RET=0
 
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	quick_test_span_gre_dir $tundev
 
 	bridge vlan del dev $swp3 vid 555
@@ -202,7 +202,7 @@ test_span_gre_forbidden_egress()
 
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: vlan forbidden at a bridge egress ($tcflags)"
+	log_test "$what: vlan forbidden at a bridge egress"
 }
 
 test_gretap_forbidden_egress()
@@ -223,7 +223,7 @@ test_span_gre_untagged_egress()
 
 	RET=0
 
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 
 	quick_test_span_gre_dir $tundev
 	quick_test_span_vlan_dir $h3 555 "$ul_proto"
@@ -246,7 +246,7 @@ test_span_gre_untagged_egress()
 
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: vlan untagged at a bridge egress ($tcflags)"
+	log_test "$what: vlan untagged at a bridge egress"
 }
 
 test_gretap_untagged_egress()
@@ -267,7 +267,7 @@ test_span_gre_fdb_roaming()
 
 	RET=0
 
-	mirror_install $swp1 ingress $tundev "matchall $tcflags"
+	mirror_install $swp1 ingress $tundev "matchall"
 	quick_test_span_gre_dir $tundev
 
 	while ((RET == 0)); do
@@ -279,7 +279,7 @@ test_span_gre_fdb_roaming()
 		if ! bridge fdb sh dev $swp2 vlan 555 master \
 		    | grep -q $h3mac; then
 			printf "TEST: %-60s  [RETRY]\n" \
-				"$what: MAC roaming ($tcflags)"
+				"$what: MAC roaming"
 			# ARP or ND probably reprimed the FDB while the test
 			# was running. We would get a spurious failure.
 			RET=0
@@ -296,7 +296,7 @@ test_span_gre_fdb_roaming()
 
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: MAC roaming ($tcflags)"
+	log_test "$what: MAC roaming"
 }
 
 test_gretap_fdb_roaming()
@@ -319,30 +319,11 @@ test_ip6gretap_stp()
 	full_test_span_gre_stp gt6 $swp3 "mirror to ip6gretap"
 }
 
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-
-	tests_run
-
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
diff --git a/tools/testing/selftests/net/forwarding/mirror_vlan.sh b/tools/testing/selftests/net/forwarding/mirror_vlan.sh
index bc3297846511..2f150a414d38 100755
--- a/tools/testing/selftests/net/forwarding/mirror_vlan.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_vlan.sh
@@ -40,12 +40,16 @@ setup_prepare()
 
 	vlan_create $h2 111 v$h2 192.0.2.18/28
 	bridge vlan add dev $swp2 vid 111
+
+	trap_install $h3 ingress
 }
 
 cleanup()
 {
 	pre_cleanup
 
+	trap_uninstall $h3 ingress
+
 	vlan_destroy $h2 111
 	vlan_destroy $h1 111
 	vlan_destroy $h3 555
@@ -63,11 +67,11 @@ test_vlan_dir()
 
 	RET=0
 
-	mirror_install $swp1 $direction $swp3.555 "matchall $tcflags"
+	mirror_install $swp1 $direction $swp3.555 "matchall"
 	test_span_dir "$h3.555" "$forward_type" "$backward_type"
 	mirror_uninstall $swp1 $direction
 
-	log_test "$direction mirror to vlan ($tcflags)"
+	log_test "$direction mirror to vlan"
 }
 
 test_vlan()
@@ -84,12 +88,12 @@ test_tagged_vlan_dir()
 
 	RET=0
 
-	mirror_install $swp1 $direction $swp3.555 "matchall $tcflags"
+	mirror_install $swp1 $direction $swp3.555 "matchall"
 	do_test_span_vlan_dir_ips '>= 10' "$h3.555" 111 ip 192.0.2.17 192.0.2.18
 	do_test_span_vlan_dir_ips  0 "$h3.555" 555 ip 192.0.2.17 192.0.2.18
 	mirror_uninstall $swp1 $direction
 
-	log_test "$direction mirror tagged to vlan ($tcflags)"
+	log_test "$direction mirror tagged to vlan"
 }
 
 test_tagged_vlan()
@@ -98,32 +102,11 @@ test_tagged_vlan()
 	test_tagged_vlan_dir egress 0 8
 }
 
-test_all()
-{
-	slow_path_trap_install $swp1 ingress
-	slow_path_trap_install $swp1 egress
-	trap_install $h3 ingress
-
-	tests_run
-
-	trap_uninstall $h3 ingress
-	slow_path_trap_uninstall $swp1 egress
-	slow_path_trap_uninstall $swp1 ingress
-}
-
 trap cleanup EXIT
 
 setup_prepare
 setup_wait
 
-tcflags="skip_hw"
-test_all
-
-if ! tc_offload_check; then
-	echo "WARN: Could not test offloaded functionality"
-else
-	tcflags="skip_sw"
-	test_all
-fi
+tests_run
 
 exit $EXIT_STATUS
-- 
cgit v1.2.3-58-ga151


From 388b2d985a136c241d918e0647b1c421de229abb Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:45 +0200
Subject: selftests: mlxsw: mirror_gre: Simplify

After the previous patch, the function test_span_failable() is always
called with should_fail=1. Drop the argument and streamline the code.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh | 17 +++++------------
 1 file changed, 5 insertions(+), 12 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh b/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
index 8cf06fcd724c..b9326baeb43e 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
@@ -150,34 +150,27 @@ test_span_gre_tos_fixed()
 
 test_span_failable()
 {
-	local should_fail=$1; shift
 	local tundev=$1; shift
 	local what=$1; shift
 
 	RET=0
 
 	mirror_install $swp1 ingress $tundev "matchall"
-	if ((should_fail)); then
-	    fail_test_span_gre_dir  $tundev
-	else
-	    quick_test_span_gre_dir $tundev
-	fi
+	fail_test_span_gre_dir  $tundev
 	mirror_uninstall $swp1 ingress
 
-	log_test "$what: should_fail=$should_fail"
+	log_test "fail $what"
 }
 
 test_failable()
 {
-	local should_fail=$1; shift
-
-	test_span_failable $should_fail gt6-key "mirror to keyful gretap"
-	test_span_failable $should_fail gt6-soft "mirror to gretap w/ soft underlay"
+	test_span_failable gt6-key "mirror to keyful gretap"
+	test_span_failable gt6-soft "mirror to gretap w/ soft underlay"
 }
 
 test_hw()
 {
-	test_failable 1
+	test_failable
 
 	test_span_gre_tos_fixed gt4 gretap "mirror to gretap"
 	test_span_gre_tos_fixed gt6 ip6gretap "mirror to ip6gretap"
-- 
cgit v1.2.3-58-ga151


From 95d33989cee5844a4bcc561dcad0f78a72a24a6d Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:46 +0200
Subject: selftests: mirror_gre_lag_lacp: Drop unnecessary code

The selftest does not use functions from mirror_gre_lib, ditch the import.

It does not use arping either, so drop the require_command as well.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh b/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh
index 1e4f0887724d..1261e6f46e34 100755
--- a/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_lag_lacp.sh
@@ -56,9 +56,6 @@ ALL_TESTS="
 NUM_NETIFS=6
 source lib.sh
 source mirror_lib.sh
-source mirror_gre_lib.sh
-
-require_command $ARPING
 
 vlan_host_create()
 {
-- 
cgit v1.2.3-58-ga151


From 4e9cd3d03af26eda951af577ddad72c4a468502a Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:47 +0200
Subject: selftests: libs: Drop slow_path_trap_install()/_uninstall()

These functions are not used anymore.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 tools/testing/selftests/net/forwarding/lib.sh | 16 ----------------
 1 file changed, 16 deletions(-)

diff --git a/tools/testing/selftests/net/forwarding/lib.sh b/tools/testing/selftests/net/forwarding/lib.sh
index df7cacb29160..ff96bb7535ff 100644
--- a/tools/testing/selftests/net/forwarding/lib.sh
+++ b/tools/testing/selftests/net/forwarding/lib.sh
@@ -1225,22 +1225,6 @@ trap_uninstall()
 	tc filter del dev $dev $direction pref 1 flower
 }
 
-slow_path_trap_install()
-{
-	# For slow-path testing, we need to install a trap to get to
-	# slow path the packets that would otherwise be switched in HW.
-	if [ "${tcflags/skip_hw}" != "$tcflags" ]; then
-		trap_install "$@"
-	fi
-}
-
-slow_path_trap_uninstall()
-{
-	if [ "${tcflags/skip_hw}" != "$tcflags" ]; then
-		trap_uninstall "$@"
-	fi
-}
-
 __icmp_capture_add_del()
 {
 	local add_del=$1; shift
-- 
cgit v1.2.3-58-ga151


From 06704a0d5e6789be52dfcb2401216c9a02140789 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:48 +0200
Subject: selftests: libs: Drop unused functions

Nothing calls these.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../testing/selftests/net/forwarding/mirror_gre_lib.sh | 11 -----------
 tools/testing/selftests/net/forwarding/mirror_lib.sh   | 18 ------------------
 2 files changed, 29 deletions(-)

diff --git a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
index aa96644e3866..20078cc55f24 100644
--- a/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_gre_lib.sh
@@ -101,17 +101,6 @@ fail_test_span_gre_dir()
 	fail_test_span_gre_dir_ips "$tundev" 192.0.2.1 192.0.2.2
 }
 
-test_span_gre_dir()
-{
-	local tundev=$1; shift
-	local direction=$1; shift
-	local forward_type=$1; shift
-	local backward_type=$1; shift
-
-	test_span_gre_dir_ips "$tundev" "$direction" "$forward_type" \
-			      "$backward_type" 192.0.2.1 192.0.2.2
-}
-
 full_test_span_gre_dir()
 {
 	local tundev=$1; shift
diff --git a/tools/testing/selftests/net/forwarding/mirror_lib.sh b/tools/testing/selftests/net/forwarding/mirror_lib.sh
index 1fc8545da6e0..6bf9d5ae933c 100644
--- a/tools/testing/selftests/net/forwarding/mirror_lib.sh
+++ b/tools/testing/selftests/net/forwarding/mirror_lib.sh
@@ -88,16 +88,6 @@ quick_test_span_dir_ips()
 			     "$forward_type" "$backward_type"
 }
 
-fail_test_span_dir_ips()
-{
-	local dev=$1; shift
-	local direction=$1; shift
-	local ip1=$1; shift
-	local ip2=$1; shift
-
-	do_test_span_dir_ips 0 "$dev" "$direction" "$ip1" "$ip2"
-}
-
 test_span_dir_ips()
 {
 	local dev=$1; shift
@@ -118,14 +108,6 @@ test_span_dir_ips()
 	icmp_capture_uninstall $dev
 }
 
-fail_test_span_dir()
-{
-	local dev=$1; shift
-	local direction=$1; shift
-
-	fail_test_span_dir_ips "$dev" "$direction" 192.0.2.1 192.0.2.2
-}
-
 test_span_dir()
 {
 	local dev=$1; shift
-- 
cgit v1.2.3-58-ga151


From 098ba97d0e895c204de95f933bfbfad9344206a5 Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Thu, 27 Jun 2024 16:48:49 +0200
Subject: selftests: mlxsw: mirror_gre: Obey TESTS

This test is unusual in that overriding TESTS does not change the tests to
be run. Split the individual tests into several functions and invoke them
through tests_run() as appropriate.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../selftests/drivers/net/mlxsw/mirror_gre.sh      | 23 +++++++++++++++++-----
 1 file changed, 18 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh b/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
index b9326baeb43e..e1ad623146d7 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/mirror_gre.sh
@@ -15,6 +15,13 @@ source $lib_dir/mirror_lib.sh
 source $lib_dir/mirror_gre_lib.sh
 source $lib_dir/mirror_gre_topo_lib.sh
 
+ALL_TESTS="
+	test_keyful
+	test_soft
+	test_tos_fixed
+	test_ttl_inherit
+"
+
 setup_keyful()
 {
 	tunnel_create gt6-key ip6gretap 2001:db8:3::1 2001:db8:3::2 \
@@ -162,19 +169,25 @@ test_span_failable()
 	log_test "fail $what"
 }
 
-test_failable()
+test_keyful()
 {
 	test_span_failable gt6-key "mirror to keyful gretap"
-	test_span_failable gt6-soft "mirror to gretap w/ soft underlay"
 }
 
-test_hw()
+test_soft()
 {
-	test_failable
+	test_span_failable gt6-soft "mirror to gretap w/ soft underlay"
+}
 
+test_tos_fixed()
+{
 	test_span_gre_tos_fixed gt4 gretap "mirror to gretap"
 	test_span_gre_tos_fixed gt6 ip6gretap "mirror to ip6gretap"
+}
 
+
+test_ttl_inherit()
+{
 	test_span_gre_ttl_inherit gt4 gretap "mirror to gretap"
 	test_span_gre_ttl_inherit gt6 ip6gretap "mirror to ip6gretap"
 }
@@ -184,6 +197,6 @@ trap cleanup EXIT
 setup_prepare
 setup_wait
 
-test_hw
+tests_run
 
 exit $EXIT_STATUS
-- 
cgit v1.2.3-58-ga151


From 69e2326a21ef409d6c709cb990565331727b9f27 Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Tue, 25 Jun 2024 16:03:52 +0200
Subject: can: gs_usb: add VID/PID for Xylanta SAINT3 product family

Add support for the Xylanta SAINT3 product family.

Cc: Andy Jackson <andy@xylanta.com>
Cc: Ken Aitchison <ken@xylanta.com>
Tested-by: Andy Jackson <andy@xylanta.com>
Link: https://lore.kernel.org/all/20240625140353.769373-1-mkl@pengutronix.de
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/usb/gs_usb.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/can/usb/gs_usb.c b/drivers/net/can/usb/gs_usb.c
index 65c962f76898..340297e3bec7 100644
--- a/drivers/net/can/usb/gs_usb.c
+++ b/drivers/net/can/usb/gs_usb.c
@@ -40,6 +40,9 @@
 #define USB_ABE_CANDEBUGGER_FD_VENDOR_ID 0x16d0
 #define USB_ABE_CANDEBUGGER_FD_PRODUCT_ID 0x10b8
 
+#define USB_XYLANTA_SAINT3_VENDOR_ID 0x16d0
+#define USB_XYLANTA_SAINT3_PRODUCT_ID 0x0f30
+
 #define GS_USB_ENDPOINT_IN 1
 #define GS_USB_ENDPOINT_OUT 2
 
@@ -1530,6 +1533,8 @@ static const struct usb_device_id gs_usb_table[] = {
 				      USB_CES_CANEXT_FD_PRODUCT_ID, 0) },
 	{ USB_DEVICE_INTERFACE_NUMBER(USB_ABE_CANDEBUGGER_FD_VENDOR_ID,
 				      USB_ABE_CANDEBUGGER_FD_PRODUCT_ID, 0) },
+	{ USB_DEVICE_INTERFACE_NUMBER(USB_XYLANTA_SAINT3_VENDOR_ID,
+				      USB_XYLANTA_SAINT3_PRODUCT_ID, 0) },
 	{} /* Terminating entry */
 };
 
-- 
cgit v1.2.3-58-ga151


From 742ad979f500c7707258b368c413c7215af09ed5 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Tue, 25 Jun 2024 21:07:44 +0200
Subject: selftests: netfilter: nft_queue.sh: add test for disappearing
 listener

If userspace program exits while the queue its subscribed to has packets
those need to be discarded.

commit dc21c6cc3d69 ("netfilter: nfnetlink_queue: acquire rcu_read_lock()
in instance_destroy_rcu()") fixed a (harmless) rcu splat that could be
triggered in this case.

Add a test case to cover this.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 tools/testing/selftests/net/netfilter/nft_queue.sh | 37 ++++++++++++++++++++++
 1 file changed, 37 insertions(+)

diff --git a/tools/testing/selftests/net/netfilter/nft_queue.sh b/tools/testing/selftests/net/netfilter/nft_queue.sh
index 8538f08c64c2..c61d23a8c88d 100755
--- a/tools/testing/selftests/net/netfilter/nft_queue.sh
+++ b/tools/testing/selftests/net/netfilter/nft_queue.sh
@@ -375,6 +375,42 @@ EOF
 	wait 2>/dev/null
 }
 
+test_queue_removal()
+{
+	read tainted_then < /proc/sys/kernel/tainted
+
+	ip netns exec "$ns1" nft -f - <<EOF
+flush ruleset
+table ip filter {
+	chain output {
+		type filter hook output priority 0; policy accept;
+		ip protocol icmp queue num 0
+	}
+}
+EOF
+	ip netns exec "$ns1" ./nf_queue -q 0 -d 30000 -t "$timeout" &
+	local nfqpid=$!
+
+	busywait "$BUSYWAIT_TIMEOUT" nf_queue_wait "$ns1" 0
+
+	ip netns exec "$ns1" ping -w 2 -f -c 10 127.0.0.1 -q >/dev/null
+	kill $nfqpid
+
+	ip netns exec "$ns1" nft flush ruleset
+
+	if [ "$tainted_then" -ne 0 ];then
+		return
+	fi
+
+	read tainted_now < /proc/sys/kernel/tainted
+	if [ "$tainted_now" -eq 0 ];then
+		echo "PASS: queue program exiting while packets queued"
+	else
+		echo "TAINT: queue program exiting while packets queued"
+		ret=1
+	fi
+}
+
 ip netns exec "$nsrouter" sysctl net.ipv6.conf.all.forwarding=1 > /dev/null
 ip netns exec "$nsrouter" sysctl net.ipv4.conf.veth0.forwarding=1 > /dev/null
 ip netns exec "$nsrouter" sysctl net.ipv4.conf.veth1.forwarding=1 > /dev/null
@@ -413,5 +449,6 @@ test_tcp_localhost
 test_tcp_localhost_connectclose
 test_tcp_localhost_requeue
 test_icmp_vrf
+test_queue_removal
 
 exit $ret
-- 
cgit v1.2.3-58-ga151


From f4ebd03496f6b67940b0af92ce885c1d0dc9e121 Mon Sep 17 00:00:00 2001
From: Phil Sutter <phil@nwl.cc>
Date: Thu, 27 Jun 2024 00:35:05 +0200
Subject: netfilter: xt_recent: Lift restrictions on max hitcount value

Support tracking of up to 65535 packets per table entry instead of just
255 to better facilitate longer term tracking or higher throughput
scenarios.

Note how this aligns sizes of struct recent_entry's 'nstamps' and
'index' fields when 'nstamps' was larger before. This is unnecessary as
the value of 'nstamps' grows along with that of 'index' after being
initialized to 1 (see recent_entry_update()). Its value will thus never
exceed that of 'index' and therefore does not need to provide space for
larger values.

Requested-by: Fabio <pedretti.fabio@gmail.com>
Link: https://bugzilla.netfilter.org/show_bug.cgi?id=1745
Signed-off-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/xt_recent.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c
index ef93e0d3bee0..588a5e6ad899 100644
--- a/net/netfilter/xt_recent.c
+++ b/net/netfilter/xt_recent.c
@@ -59,9 +59,9 @@ MODULE_PARM_DESC(ip_list_gid, "default owning group of /proc/net/xt_recent/* fil
 /* retained for backwards compatibility */
 static unsigned int ip_pkt_list_tot __read_mostly;
 module_param(ip_pkt_list_tot, uint, 0400);
-MODULE_PARM_DESC(ip_pkt_list_tot, "number of packets per IP address to remember (max. 255)");
+MODULE_PARM_DESC(ip_pkt_list_tot, "number of packets per IP address to remember (max. 65535)");
 
-#define XT_RECENT_MAX_NSTAMPS	256
+#define XT_RECENT_MAX_NSTAMPS	65536
 
 struct recent_entry {
 	struct list_head	list;
@@ -69,7 +69,7 @@ struct recent_entry {
 	union nf_inet_addr	addr;
 	u_int16_t		family;
 	u_int8_t		ttl;
-	u_int8_t		index;
+	u_int16_t		index;
 	u_int16_t		nstamps;
 	unsigned long		stamps[];
 };
@@ -80,7 +80,7 @@ struct recent_table {
 	union nf_inet_addr	mask;
 	unsigned int		refcnt;
 	unsigned int		entries;
-	u8			nstamps_max_mask;
+	u_int16_t		nstamps_max_mask;
 	struct list_head	lru_list;
 	struct list_head	iphash[];
 };
-- 
cgit v1.2.3-58-ga151


From 28cf7829a77fdd0597edb3978bebeb1f8f245b72 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Mon, 17 Jun 2024 12:07:46 -0700
Subject: MAINTAINERS: update Intel Ethernet maintainers

Since Jesse has moved to a new role, replace him with a new maintainer
to work with Tony on representing Intel networking drivers in the
kernel.

Cc: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Acked-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 MAINTAINERS | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 22328600cfd0..e0b6b419318f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -11051,8 +11051,8 @@ F:	include/drm/xe*
 F:	include/uapi/drm/xe_drm.h
 
 INTEL ETHERNET DRIVERS
-M:	Jesse Brandeburg <jesse.brandeburg@intel.com>
 M:	Tony Nguyen <anthony.l.nguyen@intel.com>
+M:	Przemek Kitszel <przemyslaw.kitszel@intel.com>
 L:	intel-wired-lan@lists.osuosl.org (moderated for non-subscribers)
 S:	Supported
 W:	https://www.intel.com/content/www/us/en/support.html
-- 
cgit v1.2.3-58-ga151


From 4dbb4f9b8fc679fb6c3db79989f64946fc7ff789 Mon Sep 17 00:00:00 2001
From: Karthik Sundaravel <ksundara@redhat.com>
Date: Tue, 28 May 2024 20:32:13 +0530
Subject: ice: Add get/set hw address for VFs using devlink commands

Changing the MAC address of the VFs is currently unsupported via devlink.
Add the function handlers to set and get the HW address for the VFs.

Signed-off-by: Karthik Sundaravel <ksundara@redhat.com>
Tested-by: Rafal Romanowski <rafal.romanowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 .../net/ethernet/intel/ice/devlink/devlink_port.c  | 59 +++++++++++++++++++++-
 drivers/net/ethernet/intel/ice/ice_sriov.c         | 34 +++++++++----
 drivers/net/ethernet/intel/ice/ice_sriov.h         |  8 +++
 3 files changed, 91 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/devlink/devlink_port.c b/drivers/net/ethernet/intel/ice/devlink/devlink_port.c
index c9fbeebf7fb9..00fed5a61d62 100644
--- a/drivers/net/ethernet/intel/ice/devlink/devlink_port.c
+++ b/drivers/net/ethernet/intel/ice/devlink/devlink_port.c
@@ -372,6 +372,62 @@ void ice_devlink_destroy_pf_port(struct ice_pf *pf)
 	devl_port_unregister(&pf->devlink_port);
 }
 
+/**
+ * ice_devlink_port_get_vf_fn_mac - .port_fn_hw_addr_get devlink handler
+ * @port: devlink port structure
+ * @hw_addr: MAC address of the port
+ * @hw_addr_len: length of MAC address
+ * @extack: extended netdev ack structure
+ *
+ * Callback for the devlink .port_fn_hw_addr_get operation
+ * Return: zero on success or an error code on failure.
+ */
+static int ice_devlink_port_get_vf_fn_mac(struct devlink_port *port,
+					  u8 *hw_addr, int *hw_addr_len,
+					  struct netlink_ext_ack *extack)
+{
+	struct ice_vf *vf = container_of(port, struct ice_vf, devlink_port);
+
+	ether_addr_copy(hw_addr, vf->dev_lan_addr);
+	*hw_addr_len = ETH_ALEN;
+
+	return 0;
+}
+
+/**
+ * ice_devlink_port_set_vf_fn_mac - .port_fn_hw_addr_set devlink handler
+ * @port: devlink port structure
+ * @hw_addr: MAC address of the port
+ * @hw_addr_len: length of MAC address
+ * @extack: extended netdev ack structure
+ *
+ * Callback for the devlink .port_fn_hw_addr_set operation
+ * Return: zero on success or an error code on failure.
+ */
+static int ice_devlink_port_set_vf_fn_mac(struct devlink_port *port,
+					  const u8 *hw_addr,
+					  int hw_addr_len,
+					  struct netlink_ext_ack *extack)
+
+{
+	struct devlink_port_attrs *attrs = &port->attrs;
+	struct devlink_port_pci_vf_attrs *pci_vf;
+	struct devlink *devlink = port->devlink;
+	struct ice_pf *pf;
+	u16 vf_id;
+
+	pf = devlink_priv(devlink);
+	pci_vf = &attrs->pci_vf;
+	vf_id = pci_vf->vf;
+
+	return __ice_set_vf_mac(pf, vf_id, hw_addr);
+}
+
+static const struct devlink_port_ops ice_devlink_vf_port_ops = {
+	.port_fn_hw_addr_get = ice_devlink_port_get_vf_fn_mac,
+	.port_fn_hw_addr_set = ice_devlink_port_set_vf_fn_mac,
+};
+
 /**
  * ice_devlink_create_vf_port - Create a devlink port for this VF
  * @vf: the VF to create a port for
@@ -407,7 +463,8 @@ int ice_devlink_create_vf_port(struct ice_vf *vf)
 	devlink_port_attrs_set(devlink_port, &attrs);
 	devlink = priv_to_devlink(pf);
 
-	err = devl_port_register(devlink, devlink_port, vsi->idx);
+	err = devl_port_register_with_ops(devlink, devlink_port, vsi->idx,
+					  &ice_devlink_vf_port_ops);
 	if (err) {
 		dev_err(dev, "Failed to create devlink port for VF %d, error %d\n",
 			vf->vf_id, err);
diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.c b/drivers/net/ethernet/intel/ice/ice_sriov.c
index 067712f4923f..55ef33208456 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.c
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.c
@@ -1416,21 +1416,23 @@ out_put_vf:
 }
 
 /**
- * ice_set_vf_mac
- * @netdev: network interface device structure
+ * __ice_set_vf_mac - program VF MAC address
+ * @pf: PF to be configure
  * @vf_id: VF identifier
  * @mac: MAC address
  *
  * program VF MAC address
+ * Return: zero on success or an error code on failure
  */
-int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac)
+int __ice_set_vf_mac(struct ice_pf *pf, u16 vf_id, const u8 *mac)
 {
-	struct ice_pf *pf = ice_netdev_to_pf(netdev);
+	struct device *dev;
 	struct ice_vf *vf;
 	int ret;
 
+	dev = ice_pf_to_dev(pf);
 	if (is_multicast_ether_addr(mac)) {
-		netdev_err(netdev, "%pM not a valid unicast address\n", mac);
+		dev_err(dev, "%pM not a valid unicast address\n", mac);
 		return -EINVAL;
 	}
 
@@ -1459,13 +1461,13 @@ int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac)
 	if (is_zero_ether_addr(mac)) {
 		/* VF will send VIRTCHNL_OP_ADD_ETH_ADDR message with its MAC */
 		vf->pf_set_mac = false;
-		netdev_info(netdev, "Removing MAC on VF %d. VF driver will be reinitialized\n",
-			    vf->vf_id);
+		dev_info(dev, "Removing MAC on VF %d. VF driver will be reinitialized\n",
+			 vf->vf_id);
 	} else {
 		/* PF will add MAC rule for the VF */
 		vf->pf_set_mac = true;
-		netdev_info(netdev, "Setting MAC %pM on VF %d. VF driver will be reinitialized\n",
-			    mac, vf_id);
+		dev_info(dev, "Setting MAC %pM on VF %d. VF driver will be reinitialized\n",
+			 mac, vf_id);
 	}
 
 	ice_reset_vf(vf, ICE_VF_RESET_NOTIFY);
@@ -1476,6 +1478,20 @@ out_put_vf:
 	return ret;
 }
 
+/**
+ * ice_set_vf_mac - .ndo_set_vf_mac handler
+ * @netdev: network interface device structure
+ * @vf_id: VF identifier
+ * @mac: MAC address
+ *
+ * program VF MAC address
+ * Return: zero on success or an error code on failure
+ */
+int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac)
+{
+	return __ice_set_vf_mac(ice_netdev_to_pf(netdev), vf_id, mac);
+}
+
 /**
  * ice_set_vf_trust
  * @netdev: network interface device structure
diff --git a/drivers/net/ethernet/intel/ice/ice_sriov.h b/drivers/net/ethernet/intel/ice/ice_sriov.h
index 8f22313474d6..96549ca5c52c 100644
--- a/drivers/net/ethernet/intel/ice/ice_sriov.h
+++ b/drivers/net/ethernet/intel/ice/ice_sriov.h
@@ -28,6 +28,7 @@
 #ifdef CONFIG_PCI_IOV
 void ice_process_vflr_event(struct ice_pf *pf);
 int ice_sriov_configure(struct pci_dev *pdev, int num_vfs);
+int __ice_set_vf_mac(struct ice_pf *pf, u16 vf_id, const u8 *mac);
 int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac);
 int
 ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi);
@@ -80,6 +81,13 @@ ice_sriov_configure(struct pci_dev __always_unused *pdev,
 	return -EOPNOTSUPP;
 }
 
+static inline int
+__ice_set_vf_mac(struct ice_pf __always_unused *pf,
+		 u16 __always_unused vf_id, const u8 __always_unused *mac)
+{
+	return -EOPNOTSUPP;
+}
+
 static inline int
 ice_set_vf_mac(struct net_device __always_unused *netdev,
 	       int __always_unused vf_id, u8 __always_unused *mac)
-- 
cgit v1.2.3-58-ga151


From d47bf9a495cf424fad674321d943123dc12b926d Mon Sep 17 00:00:00 2001
From: Eric Joyner <eric.joyner@intel.com>
Date: Mon, 17 Jun 2024 14:46:25 +0200
Subject: ice: Check all ice_vsi_rebuild() errors in function

Check the return value from ice_vsi_rebuild() and prevent the usage of
incorrectly configured VSI.

Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Eric Joyner <eric.joyner@intel.com>
Signed-off-by: Karen Ostrowska <karen.ostrowska@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_main.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 080efb7473aa..f4a39016a675 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -4158,13 +4158,17 @@ int ice_vsi_recfg_qs(struct ice_vsi *vsi, int new_rx, int new_tx, bool locked)
 
 	/* set for the next time the netdev is started */
 	if (!netif_running(vsi->netdev)) {
-		ice_vsi_rebuild(vsi, ICE_VSI_FLAG_NO_INIT);
+		err = ice_vsi_rebuild(vsi, ICE_VSI_FLAG_NO_INIT);
+		if (err)
+			goto rebuild_err;
 		dev_dbg(ice_pf_to_dev(pf), "Link is down, queue count change happens when link is brought up\n");
 		goto done;
 	}
 
 	ice_vsi_close(vsi);
-	ice_vsi_rebuild(vsi, ICE_VSI_FLAG_NO_INIT);
+	err = ice_vsi_rebuild(vsi, ICE_VSI_FLAG_NO_INIT);
+	if (err)
+		goto rebuild_err;
 
 	ice_for_each_traffic_class(i) {
 		if (vsi->tc_cfg.ena_tc & BIT(i))
@@ -4175,6 +4179,11 @@ int ice_vsi_recfg_qs(struct ice_vsi *vsi, int new_rx, int new_tx, bool locked)
 	}
 	ice_pf_dcb_recfg(pf, locked);
 	ice_vsi_open(vsi);
+	goto done;
+
+rebuild_err:
+	dev_err(ice_pf_to_dev(pf), "Error during VSI rebuild: %d. Unload and reload the driver.\n",
+		err);
 done:
 	clear_bit(ICE_CFG_BUSY, pf->state);
 	return err;
-- 
cgit v1.2.3-58-ga151


From 7dfefd0b90488da91de125f7383f966e58bcb034 Mon Sep 17 00:00:00 2001
From: Paul Greenwalt <paul.greenwalt@intel.com>
Date: Thu, 6 Jun 2024 09:48:46 -0400
Subject: ice: Allow different FW API versions based on MAC type

Allow the driver to be compatible with different FW API versions based
on the device's MAC type. Currently, E810 is only compatible with one
FW API version. Now the driver can be compatible with different FW API
versions for both E810 and E830. For example, E810 FW API version is
1.5.0 and E830 is 1.7.0.

Signed-off-by: Paul Greenwalt <paul.greenwalt@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_controlq.c | 11 +++++++----
 drivers/net/ethernet/intel/ice/ice_controlq.h | 15 ++++++++++++---
 2 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.c b/drivers/net/ethernet/intel/ice/ice_controlq.c
index ffe660f34992..ca80b34f2f8a 100644
--- a/drivers/net/ethernet/intel/ice/ice_controlq.c
+++ b/drivers/net/ethernet/intel/ice/ice_controlq.c
@@ -510,16 +510,19 @@ shutdown_sq_out:
  */
 static bool ice_aq_ver_check(struct ice_hw *hw)
 {
-	if (hw->api_maj_ver > EXP_FW_API_VER_MAJOR) {
+	u8 exp_fw_api_ver_major = EXP_FW_API_VER_MAJOR_BY_MAC(hw);
+	u8 exp_fw_api_ver_minor = EXP_FW_API_VER_MINOR_BY_MAC(hw);
+
+	if (hw->api_maj_ver > exp_fw_api_ver_major) {
 		/* Major API version is newer than expected, don't load */
 		dev_warn(ice_hw_to_dev(hw),
 			 "The driver for the device stopped because the NVM image is newer than expected. You must install the most recent version of the network driver.\n");
 		return false;
-	} else if (hw->api_maj_ver == EXP_FW_API_VER_MAJOR) {
-		if (hw->api_min_ver > (EXP_FW_API_VER_MINOR + 2))
+	} else if (hw->api_maj_ver == exp_fw_api_ver_major) {
+		if (hw->api_min_ver > (exp_fw_api_ver_minor + 2))
 			dev_info(ice_hw_to_dev(hw),
 				 "The driver for the device detected a newer version of the NVM image than expected. Please install the most recent version of the network driver.\n");
-		else if ((hw->api_min_ver + 2) < EXP_FW_API_VER_MINOR)
+		else if ((hw->api_min_ver + 2) < exp_fw_api_ver_minor)
 			dev_info(ice_hw_to_dev(hw),
 				 "The driver for the device detected an older version of the NVM image than expected. Please update the NVM image.\n");
 	} else {
diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.h b/drivers/net/ethernet/intel/ice/ice_controlq.h
index 8f2fd1613a95..1d54b1cdb1c5 100644
--- a/drivers/net/ethernet/intel/ice/ice_controlq.h
+++ b/drivers/net/ethernet/intel/ice/ice_controlq.h
@@ -21,9 +21,18 @@
 /* Defines that help manage the driver vs FW API checks.
  * Take a look at ice_aq_ver_check in ice_controlq.c for actual usage.
  */
-#define EXP_FW_API_VER_BRANCH		0x00
-#define EXP_FW_API_VER_MAJOR		0x01
-#define EXP_FW_API_VER_MINOR		0x05
+#define EXP_FW_API_VER_MAJOR_E810	0x01
+#define EXP_FW_API_VER_MINOR_E810	0x05
+
+#define EXP_FW_API_VER_MAJOR_E830	0x01
+#define EXP_FW_API_VER_MINOR_E830	0x07
+
+#define EXP_FW_API_VER_MAJOR_BY_MAC(hw) ((hw)->mac_type == ICE_MAC_E830 ? \
+					 EXP_FW_API_VER_MAJOR_E830 : \
+					 EXP_FW_API_VER_MAJOR_E810)
+#define EXP_FW_API_VER_MINOR_BY_MAC(hw) ((hw)->mac_type == ICE_MAC_E830 ? \
+					 EXP_FW_API_VER_MINOR_E830 : \
+					 EXP_FW_API_VER_MINOR_E810)
 
 /* Different control queue types: These are mainly for SW consumption. */
 enum ice_ctl_q {
-- 
cgit v1.2.3-58-ga151


From fdd288e9b7647557c51a28e72fd1003e3a9bf906 Mon Sep 17 00:00:00 2001
From: Piotr Gardocki <piotrx.gardocki@intel.com>
Date: Fri, 14 Jun 2024 12:38:11 +0200
Subject: ice: Distinguish driver reset and removal for AQ shutdown

Admin queue command for shutdown AQ contains a flag to indicate driver
unload. However, the flag is always set in the driver, even for resets. It
can cause the firmware to consider driver as unloaded once the PF reset is
triggered on all ports of device, which could lead to unexpected results.

Add an additional function parameter to functions that shutdown AQ,
indicating whether the driver is actually unloading.

Reviewed-by: Ahmed Zaki <ahmed.zaki@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Piotr Gardocki <piotrx.gardocki@intel.com>
Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_common.h   |  2 +-
 drivers/net/ethernet/intel/ice/ice_controlq.c | 19 +++++++++++--------
 drivers/net/ethernet/intel/ice/ice_main.c     |  6 +++---
 3 files changed, 15 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index 86cc1df469dd..40dc735dc25c 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -23,7 +23,7 @@ int ice_check_reset(struct ice_hw *hw);
 int ice_reset(struct ice_hw *hw, enum ice_reset_req req);
 int ice_create_all_ctrlq(struct ice_hw *hw);
 int ice_init_all_ctrlq(struct ice_hw *hw);
-void ice_shutdown_all_ctrlq(struct ice_hw *hw);
+void ice_shutdown_all_ctrlq(struct ice_hw *hw, bool unloading);
 void ice_destroy_all_ctrlq(struct ice_hw *hw);
 int
 ice_clean_rq_elem(struct ice_hw *hw, struct ice_ctl_q_info *cq,
diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.c b/drivers/net/ethernet/intel/ice/ice_controlq.c
index ca80b34f2f8a..ffaa6511c455 100644
--- a/drivers/net/ethernet/intel/ice/ice_controlq.c
+++ b/drivers/net/ethernet/intel/ice/ice_controlq.c
@@ -687,10 +687,12 @@ struct ice_ctl_q_info *ice_get_sbq(struct ice_hw *hw)
  * ice_shutdown_ctrlq - shutdown routine for any control queue
  * @hw: pointer to the hardware structure
  * @q_type: specific Control queue type
+ * @unloading: is the driver unloading itself
  *
  * NOTE: this function does not destroy the control queue locks.
  */
-static void ice_shutdown_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type)
+static void ice_shutdown_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type,
+			       bool unloading)
 {
 	struct ice_ctl_q_info *cq;
 
@@ -698,7 +700,7 @@ static void ice_shutdown_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type)
 	case ICE_CTL_Q_ADMIN:
 		cq = &hw->adminq;
 		if (ice_check_sq_alive(hw, cq))
-			ice_aq_q_shutdown(hw, true);
+			ice_aq_q_shutdown(hw, unloading);
 		break;
 	case ICE_CTL_Q_SB:
 		cq = &hw->sbq;
@@ -717,20 +719,21 @@ static void ice_shutdown_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type)
 /**
  * ice_shutdown_all_ctrlq - shutdown routine for all control queues
  * @hw: pointer to the hardware structure
+ * @unloading: is the driver unloading itself
  *
  * NOTE: this function does not destroy the control queue locks. The driver
  * may call this at runtime to shutdown and later restart control queues, such
  * as in response to a reset event.
  */
-void ice_shutdown_all_ctrlq(struct ice_hw *hw)
+void ice_shutdown_all_ctrlq(struct ice_hw *hw, bool unloading)
 {
 	/* Shutdown FW admin queue */
-	ice_shutdown_ctrlq(hw, ICE_CTL_Q_ADMIN);
+	ice_shutdown_ctrlq(hw, ICE_CTL_Q_ADMIN, unloading);
 	/* Shutdown PHY Sideband */
 	if (ice_is_sbq_supported(hw))
-		ice_shutdown_ctrlq(hw, ICE_CTL_Q_SB);
+		ice_shutdown_ctrlq(hw, ICE_CTL_Q_SB, unloading);
 	/* Shutdown PF-VF Mailbox */
-	ice_shutdown_ctrlq(hw, ICE_CTL_Q_MAILBOX);
+	ice_shutdown_ctrlq(hw, ICE_CTL_Q_MAILBOX, unloading);
 }
 
 /**
@@ -762,7 +765,7 @@ int ice_init_all_ctrlq(struct ice_hw *hw)
 			break;
 
 		ice_debug(hw, ICE_DBG_AQ_MSG, "Retry Admin Queue init due to FW critical error\n");
-		ice_shutdown_ctrlq(hw, ICE_CTL_Q_ADMIN);
+		ice_shutdown_ctrlq(hw, ICE_CTL_Q_ADMIN, true);
 		msleep(ICE_CTL_Q_ADMIN_INIT_MSEC);
 	} while (retry++ < ICE_CTL_Q_ADMIN_INIT_TIMEOUT);
 
@@ -843,7 +846,7 @@ static void ice_destroy_ctrlq_locks(struct ice_ctl_q_info *cq)
 void ice_destroy_all_ctrlq(struct ice_hw *hw)
 {
 	/* shut down all the control queues first */
-	ice_shutdown_all_ctrlq(hw);
+	ice_shutdown_all_ctrlq(hw, true);
 
 	ice_destroy_ctrlq_locks(&hw->adminq);
 	if (ice_is_sbq_supported(hw))
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index f4a39016a675..14ec4ebcd9af 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -623,7 +623,7 @@ skip:
 	if (hw->port_info)
 		ice_sched_clear_port(hw->port_info);
 
-	ice_shutdown_all_ctrlq(hw);
+	ice_shutdown_all_ctrlq(hw, false);
 
 	set_bit(ICE_PREPARED_FOR_RESET, pf->state);
 }
@@ -5499,7 +5499,7 @@ static void ice_prepare_for_shutdown(struct ice_pf *pf)
 		if (pf->vsi[v])
 			pf->vsi[v]->vsi_num = 0;
 
-	ice_shutdown_all_ctrlq(hw);
+	ice_shutdown_all_ctrlq(hw, true);
 }
 
 /**
@@ -7759,7 +7759,7 @@ err_vsi_rebuild:
 err_sched_init_port:
 	ice_sched_cleanup_all(hw);
 err_init_ctrlq:
-	ice_shutdown_all_ctrlq(hw);
+	ice_shutdown_all_ctrlq(hw, false);
 	set_bit(ICE_RESET_FAILED, pf->state);
 clear_recovery:
 	/* set this bit in PF state to control service task scheduling */
-- 
cgit v1.2.3-58-ga151


From 0f0023c649c7bc50543fbe6e1801eb6357b8bd63 Mon Sep 17 00:00:00 2001
From: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Date: Mon, 17 Jun 2024 15:24:07 +0200
Subject: ice: do not init struct ice_adapter more times than needed

Allocate and initialize struct ice_adapter object only once per physical
card instead of once per port. This is not a big deal by now, but we want
to extend this struct more and more in the near future. Our plans include
PTP stuff and a devlink instance representing whole-device/physical card.

Transactions requiring to be sleep-able (like those doing user (here ice)
memory allocation) must be performed with an additional (on top of xarray)
mutex. Adding it here removes need to xa_lock() manually.

Since this commit is a reimplementation of ice_adapter_get(), a rather new
scoped_guard() wrapper for locking is used to simplify the logic.

It's worth to mention that xa_insert() use gives us both slot reservation
and checks if it is already filled, what simplifies code a tiny bit.

Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Signed-off-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Michal Schmidt <mschmidt@redhat.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_adapter.c | 60 +++++++++++++---------------
 1 file changed, 28 insertions(+), 32 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_adapter.c b/drivers/net/ethernet/intel/ice/ice_adapter.c
index 52d15ef7f4b1..ad84d8ad49a6 100644
--- a/drivers/net/ethernet/intel/ice/ice_adapter.c
+++ b/drivers/net/ethernet/intel/ice/ice_adapter.c
@@ -11,6 +11,7 @@
 #include "ice_adapter.h"
 
 static DEFINE_XARRAY(ice_adapters);
+static DEFINE_MUTEX(ice_adapters_mutex);
 
 /* PCI bus number is 8 bits. Slot is 5 bits. Domain can have the rest. */
 #define INDEX_FIELD_DOMAIN GENMASK(BITS_PER_LONG - 1, 13)
@@ -47,8 +48,6 @@ static void ice_adapter_free(struct ice_adapter *adapter)
 	kfree(adapter);
 }
 
-DEFINE_FREE(ice_adapter_free, struct ice_adapter*, if (_T) ice_adapter_free(_T))
-
 /**
  * ice_adapter_get - Get a shared ice_adapter structure.
  * @pdev: Pointer to the pci_dev whose driver is getting the ice_adapter.
@@ -64,27 +63,26 @@ DEFINE_FREE(ice_adapter_free, struct ice_adapter*, if (_T) ice_adapter_free(_T))
  */
 struct ice_adapter *ice_adapter_get(const struct pci_dev *pdev)
 {
-	struct ice_adapter *ret, __free(ice_adapter_free) *adapter = NULL;
 	unsigned long index = ice_adapter_index(pdev);
-
-	adapter = ice_adapter_new();
-	if (!adapter)
-		return ERR_PTR(-ENOMEM);
-
-	xa_lock(&ice_adapters);
-	ret = __xa_cmpxchg(&ice_adapters, index, NULL, adapter, GFP_KERNEL);
-	if (xa_is_err(ret)) {
-		ret = ERR_PTR(xa_err(ret));
-		goto unlock;
-	}
-	if (ret) {
-		refcount_inc(&ret->refcount);
-		goto unlock;
+	struct ice_adapter *adapter;
+	int err;
+
+	scoped_guard(mutex, &ice_adapters_mutex) {
+		err = xa_insert(&ice_adapters, index, NULL, GFP_KERNEL);
+		if (err == -EBUSY) {
+			adapter = xa_load(&ice_adapters, index);
+			refcount_inc(&adapter->refcount);
+			return adapter;
+		}
+		if (err)
+			return ERR_PTR(err);
+
+		adapter = ice_adapter_new();
+		if (!adapter)
+			return ERR_PTR(-ENOMEM);
+		xa_store(&ice_adapters, index, adapter, GFP_KERNEL);
 	}
-	ret = no_free_ptr(adapter);
-unlock:
-	xa_unlock(&ice_adapters);
-	return ret;
+	return adapter;
 }
 
 /**
@@ -94,23 +92,21 @@ unlock:
  * Releases the reference to ice_adapter previously obtained with
  * ice_adapter_get.
  *
- * Context: Any.
+ * Context: Process, may sleep.
  */
 void ice_adapter_put(const struct pci_dev *pdev)
 {
 	unsigned long index = ice_adapter_index(pdev);
 	struct ice_adapter *adapter;
 
-	xa_lock(&ice_adapters);
-	adapter = xa_load(&ice_adapters, index);
-	if (WARN_ON(!adapter))
-		goto unlock;
+	scoped_guard(mutex, &ice_adapters_mutex) {
+		adapter = xa_load(&ice_adapters, index);
+		if (WARN_ON(!adapter))
+			return;
+		if (!refcount_dec_and_test(&adapter->refcount))
+			return;
 
-	if (!refcount_dec_and_test(&adapter->refcount))
-		goto unlock;
-
-	WARN_ON(__xa_erase(&ice_adapters, index) != adapter);
+		WARN_ON(xa_erase(&ice_adapters, index) != adapter);
+	}
 	ice_adapter_free(adapter);
-unlock:
-	xa_unlock(&ice_adapters);
 }
-- 
cgit v1.2.3-58-ga151


From 51b2a721612236335ddec4f3fb5f59e72a204f3a Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Thu, 25 Apr 2024 10:14:45 +0200
Subject: can: mcp251xfd: properly indent labels

To fix the coding style, remove the whitespace in front of labels.

Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c   | 34 ++++++++++++------------
 drivers/net/can/spi/mcp251xfd/mcp251xfd-dump.c   |  2 +-
 drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c |  2 +-
 drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c    |  2 +-
 4 files changed, 20 insertions(+), 20 deletions(-)

diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
index c116d3255207..61e749f97650 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
@@ -791,7 +791,7 @@ static int mcp251xfd_chip_start(struct mcp251xfd_priv *priv)
 
 	return 0;
 
- out_chip_stop:
+out_chip_stop:
 	mcp251xfd_dump(priv);
 	mcp251xfd_chip_stop(priv, CAN_STATE_STOPPED);
 
@@ -1576,7 +1576,7 @@ static irqreturn_t mcp251xfd_irq(int irq, void *dev_id)
 		handled = IRQ_HANDLED;
 	} while (1);
 
- out_fail:
+out_fail:
 	can_rx_offload_threaded_irq_finish(&priv->offload);
 
 	netdev_err(priv->ndev, "IRQ handler returned %d (intf=0x%08x).\n",
@@ -1641,22 +1641,22 @@ static int mcp251xfd_open(struct net_device *ndev)
 
 	return 0;
 
- out_free_irq:
+out_free_irq:
 	free_irq(spi->irq, priv);
- out_destroy_workqueue:
+out_destroy_workqueue:
 	destroy_workqueue(priv->wq);
- out_can_rx_offload_disable:
+out_can_rx_offload_disable:
 	can_rx_offload_disable(&priv->offload);
 	set_bit(MCP251XFD_FLAGS_DOWN, priv->flags);
 	mcp251xfd_timestamp_stop(priv);
- out_transceiver_disable:
+out_transceiver_disable:
 	mcp251xfd_transceiver_disable(priv);
- out_mcp251xfd_ring_free:
+out_mcp251xfd_ring_free:
 	mcp251xfd_ring_free(priv);
- out_pm_runtime_put:
+out_pm_runtime_put:
 	mcp251xfd_chip_stop(priv, CAN_STATE_STOPPED);
 	pm_runtime_put(ndev->dev.parent);
- out_close_candev:
+out_close_candev:
 	close_candev(ndev);
 
 	return err;
@@ -1820,9 +1820,9 @@ mcp251xfd_register_get_dev_id(const struct mcp251xfd_priv *priv, u32 *dev_id,
 	*effective_speed_hz_slow = xfer[0].effective_speed_hz;
 	*effective_speed_hz_fast = xfer[1].effective_speed_hz;
 
- out_kfree_buf_tx:
+out_kfree_buf_tx:
 	kfree(buf_tx);
- out_kfree_buf_rx:
+out_kfree_buf_rx:
 	kfree(buf_rx);
 
 	return err;
@@ -1936,13 +1936,13 @@ static int mcp251xfd_register(struct mcp251xfd_priv *priv)
 
 	return 0;
 
- out_unregister_candev:
+out_unregister_candev:
 	unregister_candev(ndev);
- out_chip_sleep:
+out_chip_sleep:
 	mcp251xfd_chip_sleep(priv);
- out_runtime_disable:
+out_runtime_disable:
 	pm_runtime_disable(ndev->dev.parent);
- out_runtime_put_noidle:
+out_runtime_put_noidle:
 	pm_runtime_put_noidle(ndev->dev.parent);
 	mcp251xfd_clks_and_vdd_disable(priv);
 
@@ -2155,9 +2155,9 @@ static int mcp251xfd_probe(struct spi_device *spi)
 
 	return 0;
 
- out_can_rx_offload_del:
+out_can_rx_offload_del:
 	can_rx_offload_del(&priv->offload);
- out_free_candev:
+out_free_candev:
 	spi->max_speed_hz = priv->spi_max_speed_hz_orig;
 
 	free_candev(ndev);
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-dump.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-dump.c
index 004eaf96262b..050321345304 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-dump.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-dump.c
@@ -94,7 +94,7 @@ static void mcp251xfd_dump_registers(const struct mcp251xfd_priv *priv,
 		kfree(buf);
 	}
 
- out:
+out:
 	mcp251xfd_dump_header(iter, MCP251XFD_DUMP_OBJECT_TYPE_REG, reg);
 }
 
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c
index 92b7bc7f14b9..65150e762007 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-regmap.c
@@ -397,7 +397,7 @@ mcp251xfd_regmap_crc_read(void *context,
 
 		return err;
 	}
- out:
+out:
 	memcpy(val_buf, buf_rx->data, val_len);
 
 	return 0;
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
index e5bd57b65aaf..ee7028c027b5 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
@@ -216,7 +216,7 @@ int mcp251xfd_handle_tefif(struct mcp251xfd_priv *priv)
 		total_frame_len += frame_len;
 	}
 
- out_netif_wake_queue:
+out_netif_wake_queue:
 	len = i;	/* number of handled goods TEFs */
 	if (len) {
 		struct mcp251xfd_tef_ring *ring = priv->tef;
-- 
cgit v1.2.3-58-ga151


From 71c45e6e0b4297433a8a64a1d24f01823a6a8dd1 Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Sun, 5 May 2024 14:40:10 +0200
Subject: can: mcp251xfd: update errata references

Since the errata references have been added to the driver, new errata
sheets have been published. Update the references for the mcp2517fd
and mcp2518fd. For completeness add references for the mcp251863.

Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
index 61e749f97650..ce1610f240a4 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
@@ -1135,7 +1135,7 @@ mcp251xfd_handle_modif(const struct mcp251xfd_priv *priv, bool *set_normal_mode)
 		return 0;
 	}
 
-	/* According to MCP2517FD errata DS80000792B 1., during a TX
+	/* According to MCP2517FD errata DS80000792C 1., during a TX
 	 * MAB underflow, the controller will transition to Restricted
 	 * Operation Mode or Listen Only Mode (depending on SERR2LOM).
 	 *
@@ -1180,7 +1180,7 @@ static int mcp251xfd_handle_serrif(struct mcp251xfd_priv *priv)
 
 	/* TX MAB underflow
 	 *
-	 * According to MCP2517FD Errata DS80000792B 1. a TX MAB
+	 * According to MCP2517FD Errata DS80000792C 1. a TX MAB
 	 * underflow is indicated by SERRIF and MODIF.
 	 *
 	 * In addition to the effects mentioned in the Errata, there
@@ -1224,7 +1224,7 @@ static int mcp251xfd_handle_serrif(struct mcp251xfd_priv *priv)
 
 	/* RX MAB overflow
 	 *
-	 * According to MCP2517FD Errata DS80000792B 1. a RX MAB
+	 * According to MCP2517FD Errata DS80000792C 1. a RX MAB
 	 * overflow is indicated by SERRIF.
 	 *
 	 * In addition to the effects mentioned in the Errata, (most
@@ -1331,7 +1331,8 @@ mcp251xfd_handle_eccif(struct mcp251xfd_priv *priv, bool set_normal_mode)
 		return err;
 
 	/* Errata Reference:
-	 * mcp2517fd: DS80000789B, mcp2518fd: DS80000792C 2.
+	 * mcp2517fd: DS80000789C 3., mcp2518fd: DS80000792E 2.,
+	 * mcp251863: DS80000984A 2.
 	 *
 	 * ECC single error correction does not work in all cases:
 	 *
@@ -2095,7 +2096,8 @@ static int mcp251xfd_probe(struct spi_device *spi)
 	priv->devtype_data = *(struct mcp251xfd_devtype_data *)spi_get_device_match_data(spi);
 
 	/* Errata Reference:
-	 * mcp2517fd: DS80000792C 5., mcp2518fd: DS80000789C 4.
+	 * mcp2517fd: DS80000792C 5., mcp2518fd: DS80000789E 4.,
+	 * mcp251863: DS80000984A 4.
 	 *
 	 * The SPI can write corrupted data to the RAM at fast SPI
 	 * speeds:
-- 
cgit v1.2.3-58-ga151


From a7801540f325d104de5065850a003f1d9bdc6ad3 Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Wed, 11 Jan 2023 12:10:04 +0100
Subject: can: mcp251xfd: move mcp251xfd_timestamp_start()/stop() into
 mcp251xfd_chip_start/stop()

The mcp251xfd wakes up from Low Power or Sleep Mode when SPI activity
is detected. To avoid this, make sure that the timestamp worker is
stopped before shutting down the chip.

Split the starting of the timestamp worker out of
mcp251xfd_timestamp_init() into the separate function
mcp251xfd_timestamp_start().

Call mcp251xfd_timestamp_init() before mcp251xfd_chip_start(), move
mcp251xfd_timestamp_start() to mcp251xfd_chip_start(). In this way,
mcp251xfd_timestamp_stop() can be called unconditionally by
mcp251xfd_chip_stop().

Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c      | 8 +++++---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-timestamp.c | 7 +++++--
 drivers/net/can/spi/mcp251xfd/mcp251xfd.h           | 1 +
 3 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
index ce1610f240a4..e8e11c32cfda 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
@@ -744,6 +744,7 @@ static void mcp251xfd_chip_stop(struct mcp251xfd_priv *priv,
 
 	mcp251xfd_chip_interrupts_disable(priv);
 	mcp251xfd_chip_rx_int_disable(priv);
+	mcp251xfd_timestamp_stop(priv);
 	mcp251xfd_chip_sleep(priv);
 }
 
@@ -763,6 +764,8 @@ static int mcp251xfd_chip_start(struct mcp251xfd_priv *priv)
 	if (err)
 		goto out_chip_stop;
 
+	mcp251xfd_timestamp_start(priv);
+
 	err = mcp251xfd_set_bittiming(priv);
 	if (err)
 		goto out_chip_stop;
@@ -1611,11 +1614,12 @@ static int mcp251xfd_open(struct net_device *ndev)
 	if (err)
 		goto out_mcp251xfd_ring_free;
 
+	mcp251xfd_timestamp_init(priv);
+
 	err = mcp251xfd_chip_start(priv);
 	if (err)
 		goto out_transceiver_disable;
 
-	mcp251xfd_timestamp_init(priv);
 	clear_bit(MCP251XFD_FLAGS_DOWN, priv->flags);
 	can_rx_offload_enable(&priv->offload);
 
@@ -1649,7 +1653,6 @@ out_destroy_workqueue:
 out_can_rx_offload_disable:
 	can_rx_offload_disable(&priv->offload);
 	set_bit(MCP251XFD_FLAGS_DOWN, priv->flags);
-	mcp251xfd_timestamp_stop(priv);
 out_transceiver_disable:
 	mcp251xfd_transceiver_disable(priv);
 out_mcp251xfd_ring_free:
@@ -1675,7 +1678,6 @@ static int mcp251xfd_stop(struct net_device *ndev)
 	free_irq(ndev->irq, priv);
 	destroy_workqueue(priv->wq);
 	can_rx_offload_disable(&priv->offload);
-	mcp251xfd_timestamp_stop(priv);
 	mcp251xfd_chip_stop(priv, CAN_STATE_STOPPED);
 	mcp251xfd_transceiver_disable(priv);
 	mcp251xfd_ring_free(priv);
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-timestamp.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-timestamp.c
index 712e09186987..7bbf4603038b 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-timestamp.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-timestamp.c
@@ -58,9 +58,12 @@ void mcp251xfd_timestamp_init(struct mcp251xfd_priv *priv)
 	cc->shift = 1;
 	cc->mult = clocksource_hz2mult(priv->can.clock.freq, cc->shift);
 
-	timecounter_init(&priv->tc, &priv->cc, ktime_get_real_ns());
-
 	INIT_DELAYED_WORK(&priv->timestamp, mcp251xfd_timestamp_work);
+}
+
+void mcp251xfd_timestamp_start(struct mcp251xfd_priv *priv)
+{
+	timecounter_init(&priv->tc, &priv->cc, ktime_get_real_ns());
 	schedule_delayed_work(&priv->timestamp,
 			      MCP251XFD_TIMESTAMP_WORK_DELAY_SEC * HZ);
 }
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
index b35bfebd23f2..d6f6b3182e6a 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
@@ -954,6 +954,7 @@ int mcp251xfd_handle_tefif(struct mcp251xfd_priv *priv);
 void mcp251xfd_skb_set_timestamp(const struct mcp251xfd_priv *priv,
 				 struct sk_buff *skb, u32 timestamp);
 void mcp251xfd_timestamp_init(struct mcp251xfd_priv *priv);
+void mcp251xfd_timestamp_start(struct mcp251xfd_priv *priv);
 void mcp251xfd_timestamp_stop(struct mcp251xfd_priv *priv);
 
 void mcp251xfd_tx_obj_write_sync(struct work_struct *work);
-- 
cgit v1.2.3-58-ga151


From e793c724b48ca8cae9693bc3be528e85284c126a Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Wed, 11 Jan 2023 11:48:16 +0100
Subject: can: mcp251xfd: clarify the meaning of timestamp
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The mcp251xfd chip is configured to provide a timestamp with each
received and transmitted CAN frame. The timestamp is derived from the
internal free-running timer, which can also be read from the TBC
register via SPI. The timer is 32 bits wide and is clocked by the
external oscillator (typically 20 or 40 MHz).

To avoid confusion, we call this timestamp "timestamp_raw" or "ts_raw"
for short.

Using the timecounter framework, the "ts_raw" is converted to 64 bit
nanoseconds since the epoch. This is what we call "timestamp".

This is a preparation for the next patches which use the "timestamp"
to work around a bug where so far only the "ts_raw" is used.

Tested-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Tested-by: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c     | 28 +++++++++++-----------
 drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c       |  2 +-
 drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c      |  2 +-
 .../net/can/spi/mcp251xfd/mcp251xfd-timestamp.c    | 22 +++++------------
 drivers/net/can/spi/mcp251xfd/mcp251xfd.h          | 27 ++++++++++++++++-----
 5 files changed, 43 insertions(+), 38 deletions(-)

diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
index e8e11c32cfda..3e7526274e34 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
@@ -2,7 +2,7 @@
 //
 // mcp251xfd - Microchip MCP251xFD Family CAN controller driver
 //
-// Copyright (c) 2019, 2020, 2021 Pengutronix,
+// Copyright (c) 2019, 2020, 2021, 2023 Pengutronix,
 //               Marc Kleine-Budde <kernel@pengutronix.de>
 //
 // Based on:
@@ -870,18 +870,18 @@ static int mcp251xfd_get_berr_counter(const struct net_device *ndev,
 
 static struct sk_buff *
 mcp251xfd_alloc_can_err_skb(struct mcp251xfd_priv *priv,
-			    struct can_frame **cf, u32 *timestamp)
+			    struct can_frame **cf, u32 *ts_raw)
 {
 	struct sk_buff *skb;
 	int err;
 
-	err = mcp251xfd_get_timestamp(priv, timestamp);
+	err = mcp251xfd_get_timestamp_raw(priv, ts_raw);
 	if (err)
 		return NULL;
 
 	skb = alloc_can_err_skb(priv->ndev, cf);
 	if (skb)
-		mcp251xfd_skb_set_timestamp(priv, skb, *timestamp);
+		mcp251xfd_skb_set_timestamp_raw(priv, skb, *ts_raw);
 
 	return skb;
 }
@@ -892,7 +892,7 @@ static int mcp251xfd_handle_rxovif(struct mcp251xfd_priv *priv)
 	struct mcp251xfd_rx_ring *ring;
 	struct sk_buff *skb;
 	struct can_frame *cf;
-	u32 timestamp, rxovif;
+	u32 ts_raw, rxovif;
 	int err, i;
 
 	stats->rx_over_errors++;
@@ -927,14 +927,14 @@ static int mcp251xfd_handle_rxovif(struct mcp251xfd_priv *priv)
 			return err;
 	}
 
-	skb = mcp251xfd_alloc_can_err_skb(priv, &cf, &timestamp);
+	skb = mcp251xfd_alloc_can_err_skb(priv, &cf, &ts_raw);
 	if (!skb)
 		return 0;
 
 	cf->can_id |= CAN_ERR_CRTL;
 	cf->data[1] = CAN_ERR_CRTL_RX_OVERFLOW;
 
-	err = can_rx_offload_queue_timestamp(&priv->offload, skb, timestamp);
+	err = can_rx_offload_queue_timestamp(&priv->offload, skb, ts_raw);
 	if (err)
 		stats->rx_fifo_errors++;
 
@@ -951,12 +951,12 @@ static int mcp251xfd_handle_txatif(struct mcp251xfd_priv *priv)
 static int mcp251xfd_handle_ivmif(struct mcp251xfd_priv *priv)
 {
 	struct net_device_stats *stats = &priv->ndev->stats;
-	u32 bdiag1, timestamp;
+	u32 bdiag1, ts_raw;
 	struct sk_buff *skb;
 	struct can_frame *cf = NULL;
 	int err;
 
-	err = mcp251xfd_get_timestamp(priv, &timestamp);
+	err = mcp251xfd_get_timestamp_raw(priv, &ts_raw);
 	if (err)
 		return err;
 
@@ -1038,8 +1038,8 @@ static int mcp251xfd_handle_ivmif(struct mcp251xfd_priv *priv)
 	if (!cf)
 		return 0;
 
-	mcp251xfd_skb_set_timestamp(priv, skb, timestamp);
-	err = can_rx_offload_queue_timestamp(&priv->offload, skb, timestamp);
+	mcp251xfd_skb_set_timestamp_raw(priv, skb, ts_raw);
+	err = can_rx_offload_queue_timestamp(&priv->offload, skb, ts_raw);
 	if (err)
 		stats->rx_fifo_errors++;
 
@@ -1052,7 +1052,7 @@ static int mcp251xfd_handle_cerrif(struct mcp251xfd_priv *priv)
 	struct sk_buff *skb;
 	struct can_frame *cf = NULL;
 	enum can_state new_state, rx_state, tx_state;
-	u32 trec, timestamp;
+	u32 trec, ts_raw;
 	int err;
 
 	err = regmap_read(priv->map_reg, MCP251XFD_REG_TREC, &trec);
@@ -1082,7 +1082,7 @@ static int mcp251xfd_handle_cerrif(struct mcp251xfd_priv *priv)
 	/* The skb allocation might fail, but can_change_state()
 	 * handles cf == NULL.
 	 */
-	skb = mcp251xfd_alloc_can_err_skb(priv, &cf, &timestamp);
+	skb = mcp251xfd_alloc_can_err_skb(priv, &cf, &ts_raw);
 	can_change_state(priv->ndev, cf, tx_state, rx_state);
 
 	if (new_state == CAN_STATE_BUS_OFF) {
@@ -1113,7 +1113,7 @@ static int mcp251xfd_handle_cerrif(struct mcp251xfd_priv *priv)
 		cf->data[7] = bec.rxerr;
 	}
 
-	err = can_rx_offload_queue_timestamp(&priv->offload, skb, timestamp);
+	err = can_rx_offload_queue_timestamp(&priv->offload, skb, ts_raw);
 	if (err)
 		stats->rx_fifo_errors++;
 
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
index ced8d9c81f8c..6d66b97709b8 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
@@ -149,7 +149,7 @@ mcp251xfd_hw_rx_obj_to_skb(const struct mcp251xfd_priv *priv,
 	if (!(hw_rx_obj->flags & MCP251XFD_OBJ_FLAGS_RTR))
 		memcpy(cfd->data, hw_rx_obj->data, cfd->len);
 
-	mcp251xfd_skb_set_timestamp(priv, skb, hw_rx_obj->ts);
+	mcp251xfd_skb_set_timestamp_raw(priv, skb, hw_rx_obj->ts);
 }
 
 static int
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
index ee7028c027b5..4bc8d71fc491 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
@@ -109,7 +109,7 @@ mcp251xfd_handle_tefif_one(struct mcp251xfd_priv *priv,
 	tef_tail = mcp251xfd_get_tef_tail(priv);
 	skb = priv->can.echo_skb[tef_tail];
 	if (skb)
-		mcp251xfd_skb_set_timestamp(priv, skb, hw_tef_obj->ts);
+		mcp251xfd_skb_set_timestamp_raw(priv, skb, hw_tef_obj->ts);
 	stats->tx_bytes +=
 		can_rx_offload_get_echo_skb_queue_timestamp(&priv->offload,
 							    tef_tail, hw_tef_obj->ts,
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-timestamp.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-timestamp.c
index 7bbf4603038b..202ca0d24d03 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-timestamp.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-timestamp.c
@@ -2,7 +2,7 @@
 //
 // mcp251xfd - Microchip MCP251xFD Family CAN controller driver
 //
-// Copyright (c) 2021 Pengutronix,
+// Copyright (c) 2021, 2023 Pengutronix,
 //               Marc Kleine-Budde <kernel@pengutronix.de>
 //
 
@@ -11,20 +11,20 @@
 
 #include "mcp251xfd.h"
 
-static u64 mcp251xfd_timestamp_read(const struct cyclecounter *cc)
+static u64 mcp251xfd_timestamp_raw_read(const struct cyclecounter *cc)
 {
 	const struct mcp251xfd_priv *priv;
-	u32 timestamp = 0;
+	u32 ts_raw = 0;
 	int err;
 
 	priv = container_of(cc, struct mcp251xfd_priv, cc);
-	err = mcp251xfd_get_timestamp(priv, &timestamp);
+	err = mcp251xfd_get_timestamp_raw(priv, &ts_raw);
 	if (err)
 		netdev_err(priv->ndev,
 			   "Error %d while reading timestamp. HW timestamps may be inaccurate.",
 			   err);
 
-	return timestamp;
+	return ts_raw;
 }
 
 static void mcp251xfd_timestamp_work(struct work_struct *work)
@@ -39,21 +39,11 @@ static void mcp251xfd_timestamp_work(struct work_struct *work)
 			      MCP251XFD_TIMESTAMP_WORK_DELAY_SEC * HZ);
 }
 
-void mcp251xfd_skb_set_timestamp(const struct mcp251xfd_priv *priv,
-				 struct sk_buff *skb, u32 timestamp)
-{
-	struct skb_shared_hwtstamps *hwtstamps = skb_hwtstamps(skb);
-	u64 ns;
-
-	ns = timecounter_cyc2time(&priv->tc, timestamp);
-	hwtstamps->hwtstamp = ns_to_ktime(ns);
-}
-
 void mcp251xfd_timestamp_init(struct mcp251xfd_priv *priv)
 {
 	struct cyclecounter *cc = &priv->cc;
 
-	cc->read = mcp251xfd_timestamp_read;
+	cc->read = mcp251xfd_timestamp_raw_read;
 	cc->mask = CYCLECOUNTER_MASK(32);
 	cc->shift = 1;
 	cc->mult = clocksource_hz2mult(priv->can.clock.freq, cc->shift);
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
index d6f6b3182e6a..4d31689d73c8 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
@@ -2,7 +2,7 @@
  *
  * mcp251xfd - Microchip MCP251xFD Family CAN controller driver
  *
- * Copyright (c) 2019, 2020, 2021 Pengutronix,
+ * Copyright (c) 2019, 2020, 2021, 2023 Pengutronix,
  *               Marc Kleine-Budde <kernel@pengutronix.de>
  * Copyright (c) 2019 Martin Sperl <kernel@martin.sperl.org>
  */
@@ -809,10 +809,27 @@ mcp251xfd_spi_cmd_write(const struct mcp251xfd_priv *priv,
 	return data;
 }
 
-static inline int mcp251xfd_get_timestamp(const struct mcp251xfd_priv *priv,
-					  u32 *timestamp)
+static inline int mcp251xfd_get_timestamp_raw(const struct mcp251xfd_priv *priv,
+					      u32 *ts_raw)
 {
-	return regmap_read(priv->map_reg, MCP251XFD_REG_TBC, timestamp);
+	return regmap_read(priv->map_reg, MCP251XFD_REG_TBC, ts_raw);
+}
+
+static inline void mcp251xfd_skb_set_timestamp(struct sk_buff *skb, u64 ns)
+{
+	struct skb_shared_hwtstamps *hwtstamps = skb_hwtstamps(skb);
+
+	hwtstamps->hwtstamp = ns_to_ktime(ns);
+}
+
+static inline
+void mcp251xfd_skb_set_timestamp_raw(const struct mcp251xfd_priv *priv,
+				     struct sk_buff *skb, u32 ts_raw)
+{
+	u64 ns;
+
+	ns = timecounter_cyc2time(&priv->tc, ts_raw);
+	mcp251xfd_skb_set_timestamp(skb, ns);
 }
 
 static inline u16 mcp251xfd_get_tef_obj_addr(u8 n)
@@ -951,8 +968,6 @@ void mcp251xfd_ring_free(struct mcp251xfd_priv *priv);
 int mcp251xfd_ring_alloc(struct mcp251xfd_priv *priv);
 int mcp251xfd_handle_rxif(struct mcp251xfd_priv *priv);
 int mcp251xfd_handle_tefif(struct mcp251xfd_priv *priv);
-void mcp251xfd_skb_set_timestamp(const struct mcp251xfd_priv *priv,
-				 struct sk_buff *skb, u32 timestamp);
 void mcp251xfd_timestamp_init(struct mcp251xfd_priv *priv);
 void mcp251xfd_timestamp_start(struct mcp251xfd_priv *priv);
 void mcp251xfd_timestamp_stop(struct mcp251xfd_priv *priv);
-- 
cgit v1.2.3-58-ga151


From d49184b7b585f9da7ee546b744525f62117019f6 Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Wed, 11 Jan 2023 20:25:48 +0100
Subject: can: mcp251xfd: mcp251xfd_handle_rxif_ring_uinc(): factor out in
 separate function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This is a preparation patch.

Sending the UINC messages followed by incrementing the tail pointer
will be called in more than one place in upcoming patches, so factor
this out into a separate function.

Also make mcp251xfd_handle_rxif_ring_uinc() safe to be called with a
"len" of 0.

Tested-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Tested-by: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c | 48 ++++++++++++++++++----------
 1 file changed, 32 insertions(+), 16 deletions(-)

diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
index 6d66b97709b8..2c2b057b7ac7 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
@@ -197,6 +197,37 @@ mcp251xfd_rx_obj_read(const struct mcp251xfd_priv *priv,
 	return err;
 }
 
+static int
+mcp251xfd_handle_rxif_ring_uinc(const struct mcp251xfd_priv *priv,
+				struct mcp251xfd_rx_ring *ring,
+				u8 len)
+{
+	int offset;
+	int err;
+
+	if (!len)
+		return 0;
+
+	/* Increment the RX FIFO tail pointer 'len' times in a
+	 * single SPI message.
+	 *
+	 * Note:
+	 * Calculate offset, so that the SPI transfer ends on
+	 * the last message of the uinc_xfer array, which has
+	 * "cs_change == 0", to properly deactivate the chip
+	 * select.
+	 */
+	offset = ARRAY_SIZE(ring->uinc_xfer) - len;
+	err = spi_sync_transfer(priv->spi,
+				ring->uinc_xfer + offset, len);
+	if (err)
+		return err;
+
+	ring->tail += len;
+
+	return 0;
+}
+
 static int
 mcp251xfd_handle_rxif_ring(struct mcp251xfd_priv *priv,
 			   struct mcp251xfd_rx_ring *ring)
@@ -210,8 +241,6 @@ mcp251xfd_handle_rxif_ring(struct mcp251xfd_priv *priv,
 		return err;
 
 	while ((len = mcp251xfd_get_rx_linear_len(ring))) {
-		int offset;
-
 		rx_tail = mcp251xfd_get_rx_tail(ring);
 
 		err = mcp251xfd_rx_obj_read(priv, ring, hw_rx_obj,
@@ -227,22 +256,9 @@ mcp251xfd_handle_rxif_ring(struct mcp251xfd_priv *priv,
 				return err;
 		}
 
-		/* Increment the RX FIFO tail pointer 'len' times in a
-		 * single SPI message.
-		 *
-		 * Note:
-		 * Calculate offset, so that the SPI transfer ends on
-		 * the last message of the uinc_xfer array, which has
-		 * "cs_change == 0", to properly deactivate the chip
-		 * select.
-		 */
-		offset = ARRAY_SIZE(ring->uinc_xfer) - len;
-		err = spi_sync_transfer(priv->spi,
-					ring->uinc_xfer + offset, len);
+		err = mcp251xfd_handle_rxif_ring_uinc(priv, ring, len);
 		if (err)
 			return err;
-
-		ring->tail += len;
 	}
 
 	return 0;
-- 
cgit v1.2.3-58-ga151


From 85505e585637a737e4713c1386c30e37c325b82e Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Wed, 11 Jan 2023 21:07:03 +0100
Subject: can: mcp251xfd: rx: prepare to workaround broken RX FIFO head index
 erratum
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This is a preparatory patch to work around erratum DS80000789E 6 of
the mcp2518fd, the other variants of the chip family (mcp2517fd and
mcp251863) are probably also affected.

When handling the RX interrupt, the driver iterates over all pending
FIFOs (which are implemented as ring buffers in hardware) and reads
the FIFO header index from the RX FIFO STA register of the chip.

In the bad case, the driver reads a too large head index. In the
original code, the driver always trusted the read value, which caused
old CAN frames that were already processed, or new, incompletely
written CAN frames to be (re-)processed.

Instead of reading and trusting the head index, read the head index
and calculate the number of CAN frames that were supposedly received -
replace mcp251xfd_rx_ring_update() with mcp251xfd_get_rx_len().

The mcp251xfd_handle_rxif_ring() function reads the received CAN
frames from the chip, iterates over them and pushes them into the
network stack. Prepare that the iteration can be stopped if an old CAN
frame is detected. The actual code to detect old or incomplete frames
and abort will be added in the next patch.

Link: https://lore.kernel.org/all/BL3PR11MB64844C1C95CA3BDADAE4D8CCFBC99@BL3PR11MB6484.namprd11.prod.outlook.com
Reported-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Closes: https://lore.kernel.org/all/FR0P281MB1966273C216630B120ABB6E197E89@FR0P281MB1966.DEUP281.PROD.OUTLOOK.COM
Tested-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Tested-by: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c |  2 +
 drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c   | 89 +++++++++++++++-----------
 drivers/net/can/spi/mcp251xfd/mcp251xfd.h      | 12 +---
 3 files changed, 56 insertions(+), 47 deletions(-)

diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c
index bfe4caa0c99d..b308a375e26e 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c
@@ -507,6 +507,8 @@ int mcp251xfd_ring_alloc(struct mcp251xfd_priv *priv)
 		}
 
 		rx_ring->obj_num = rx_obj_num;
+		rx_ring->obj_num_shift_to_u8 = BITS_PER_TYPE(rx_ring->obj_num_shift_to_u8) -
+			ilog2(rx_obj_num);
 		rx_ring->obj_size = rx_obj_size;
 		priv->rx[i] = rx_ring;
 	}
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
index 2c2b057b7ac7..a79e6c661ecc 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
@@ -2,7 +2,7 @@
 //
 // mcp251xfd - Microchip MCP251xFD Family CAN controller driver
 //
-// Copyright (c) 2019, 2020, 2021 Pengutronix,
+// Copyright (c) 2019, 2020, 2021, 2023 Pengutronix,
 //               Marc Kleine-Budde <kernel@pengutronix.de>
 //
 // Based on:
@@ -16,23 +16,14 @@
 
 #include "mcp251xfd.h"
 
-static inline int
-mcp251xfd_rx_head_get_from_chip(const struct mcp251xfd_priv *priv,
-				const struct mcp251xfd_rx_ring *ring,
-				u8 *rx_head, bool *fifo_empty)
+static inline bool mcp251xfd_rx_fifo_sta_empty(const u32 fifo_sta)
 {
-	u32 fifo_sta;
-	int err;
-
-	err = regmap_read(priv->map_reg, MCP251XFD_REG_FIFOSTA(ring->fifo_nr),
-			  &fifo_sta);
-	if (err)
-		return err;
-
-	*rx_head = FIELD_GET(MCP251XFD_REG_FIFOSTA_FIFOCI_MASK, fifo_sta);
-	*fifo_empty = !(fifo_sta & MCP251XFD_REG_FIFOSTA_TFNRFNIF);
+	return !(fifo_sta & MCP251XFD_REG_FIFOSTA_TFNRFNIF);
+}
 
-	return 0;
+static inline bool mcp251xfd_rx_fifo_sta_full(const u32 fifo_sta)
+{
+	return fifo_sta & MCP251XFD_REG_FIFOSTA_TFERFFIF;
 }
 
 static inline int
@@ -80,29 +71,49 @@ mcp251xfd_check_rx_tail(const struct mcp251xfd_priv *priv,
 }
 
 static int
-mcp251xfd_rx_ring_update(const struct mcp251xfd_priv *priv,
-			 struct mcp251xfd_rx_ring *ring)
+mcp251xfd_get_rx_len(const struct mcp251xfd_priv *priv,
+		     const struct mcp251xfd_rx_ring *ring,
+		     u8 *len_p)
 {
-	u32 new_head;
-	u8 chip_rx_head;
-	bool fifo_empty;
+	const u8 shift = ring->obj_num_shift_to_u8;
+	u8 chip_head, tail, len;
+	u32 fifo_sta;
 	int err;
 
-	err = mcp251xfd_rx_head_get_from_chip(priv, ring, &chip_rx_head,
-					      &fifo_empty);
-	if (err || fifo_empty)
+	err = regmap_read(priv->map_reg, MCP251XFD_REG_FIFOSTA(ring->fifo_nr),
+			  &fifo_sta);
+	if (err)
+		return err;
+
+	if (mcp251xfd_rx_fifo_sta_empty(fifo_sta)) {
+		*len_p = 0;
+		return 0;
+	}
+
+	if (mcp251xfd_rx_fifo_sta_full(fifo_sta)) {
+		*len_p = ring->obj_num;
+		return 0;
+	}
+
+	chip_head = FIELD_GET(MCP251XFD_REG_FIFOSTA_FIFOCI_MASK, fifo_sta);
+
+	err =  mcp251xfd_check_rx_tail(priv, ring);
+	if (err)
 		return err;
+	tail = mcp251xfd_get_rx_tail(ring);
 
-	/* chip_rx_head, is the next RX-Object filled by the HW.
-	 * The new RX head must be >= the old head.
+	/* First shift to full u8. The subtraction works on signed
+	 * values, that keeps the difference steady around the u8
+	 * overflow. The right shift acts on len, which is an u8.
 	 */
-	new_head = round_down(ring->head, ring->obj_num) + chip_rx_head;
-	if (new_head <= ring->head)
-		new_head += ring->obj_num;
+	BUILD_BUG_ON(sizeof(ring->obj_num) != sizeof(chip_head));
+	BUILD_BUG_ON(sizeof(ring->obj_num) != sizeof(tail));
+	BUILD_BUG_ON(sizeof(ring->obj_num) != sizeof(len));
 
-	ring->head = new_head;
+	len = (chip_head << shift) - (tail << shift);
+	*len_p = len >> shift;
 
-	return mcp251xfd_check_rx_tail(priv, ring);
+	return 0;
 }
 
 static void
@@ -208,6 +219,8 @@ mcp251xfd_handle_rxif_ring_uinc(const struct mcp251xfd_priv *priv,
 	if (!len)
 		return 0;
 
+	ring->head += len;
+
 	/* Increment the RX FIFO tail pointer 'len' times in a
 	 * single SPI message.
 	 *
@@ -233,22 +246,22 @@ mcp251xfd_handle_rxif_ring(struct mcp251xfd_priv *priv,
 			   struct mcp251xfd_rx_ring *ring)
 {
 	struct mcp251xfd_hw_rx_obj_canfd *hw_rx_obj = ring->obj;
-	u8 rx_tail, len;
+	u8 rx_tail, len, l;
 	int err, i;
 
-	err = mcp251xfd_rx_ring_update(priv, ring);
+	err = mcp251xfd_get_rx_len(priv, ring, &len);
 	if (err)
 		return err;
 
-	while ((len = mcp251xfd_get_rx_linear_len(ring))) {
+	while ((l = mcp251xfd_get_rx_linear_len(ring, len))) {
 		rx_tail = mcp251xfd_get_rx_tail(ring);
 
 		err = mcp251xfd_rx_obj_read(priv, ring, hw_rx_obj,
-					    rx_tail, len);
+					    rx_tail, l);
 		if (err)
 			return err;
 
-		for (i = 0; i < len; i++) {
+		for (i = 0; i < l; i++) {
 			err = mcp251xfd_handle_rxif_one(priv, ring,
 							(void *)hw_rx_obj +
 							i * ring->obj_size);
@@ -256,9 +269,11 @@ mcp251xfd_handle_rxif_ring(struct mcp251xfd_priv *priv,
 				return err;
 		}
 
-		err = mcp251xfd_handle_rxif_ring_uinc(priv, ring, len);
+		err = mcp251xfd_handle_rxif_ring_uinc(priv, ring, l);
 		if (err)
 			return err;
+
+		len -= l;
 	}
 
 	return 0;
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
index 4d31689d73c8..453a2b193c75 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
@@ -556,6 +556,7 @@ struct mcp251xfd_rx_ring {
 	u8 nr;
 	u8 fifo_nr;
 	u8 obj_num;
+	u8 obj_num_shift_to_u8;
 	u8 obj_size;
 
 	union mcp251xfd_write_reg_buf irq_enable_buf;
@@ -931,18 +932,9 @@ static inline u8 mcp251xfd_get_rx_tail(const struct mcp251xfd_rx_ring *ring)
 	return ring->tail & (ring->obj_num - 1);
 }
 
-static inline u8 mcp251xfd_get_rx_len(const struct mcp251xfd_rx_ring *ring)
-{
-	return ring->head - ring->tail;
-}
-
 static inline u8
-mcp251xfd_get_rx_linear_len(const struct mcp251xfd_rx_ring *ring)
+mcp251xfd_get_rx_linear_len(const struct mcp251xfd_rx_ring *ring, u8 len)
 {
-	u8 len;
-
-	len = mcp251xfd_get_rx_len(ring);
-
 	return min_t(u8, len, ring->obj_num - mcp251xfd_get_rx_tail(ring));
 }
 
-- 
cgit v1.2.3-58-ga151


From 24436be590c6fbb05f6161b0dfba7d9da60214aa Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Wed, 11 Jan 2023 11:53:50 +0100
Subject: can: mcp251xfd: rx: add workaround for erratum DS80000789E 6 of
 mcp2518fd
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch tries to works around erratum DS80000789E 6 of the
mcp2518fd, the other variants of the chip family (mcp2517fd and
mcp251863) are probably also affected.

In the bad case, the driver reads a too large head index. In the
original code, the driver always trusted the read value, which caused
old, already processed CAN frames or new, incompletely written CAN
frames to be (re-)processed.

To work around this issue, keep a per FIFO timestamp [1] of the last
valid received CAN frame and compare against the timestamp of every
received CAN frame. If an old CAN frame is detected, abort the
iteration and mark the number of valid CAN frames as processed in the
chip by incrementing the FIFO's tail index.

Further tests showed that this workaround can recognize old CAN
frames, but a small time window remains in which partially written CAN
frames [2] are not recognized but then processed. These CAN frames
have the correct data and time stamps, but the DLC has not yet been
updated.

[1] As the raw timestamp overflows every 107 seconds (at the usual
    clock rate of 40 MHz) convert it to nanoseconds with the
    timecounter framework and use this to detect stale CAN frames.

Link: https://lore.kernel.org/all/BL3PR11MB64844C1C95CA3BDADAE4D8CCFBC99@BL3PR11MB6484.namprd11.prod.outlook.com [2]
Reported-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Closes: https://lore.kernel.org/all/FR0P281MB1966273C216630B120ABB6E197E89@FR0P281MB1966.DEUP281.PROD.OUTLOOK.COM
Tested-by: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Tested-by: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c |  1 +
 drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c   | 32 +++++++++++++++++++++++---
 drivers/net/can/spi/mcp251xfd/mcp251xfd.h      |  3 +++
 3 files changed, 33 insertions(+), 3 deletions(-)

diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c
index b308a375e26e..8464fc4f37d0 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c
@@ -206,6 +206,7 @@ mcp251xfd_ring_init_rx(struct mcp251xfd_priv *priv, u16 *base, u8 *fifo_nr)
 	int i, j;
 
 	mcp251xfd_for_each_rx_ring(priv, rx_ring, i) {
+		rx_ring->last_valid = timecounter_read(&priv->tc);
 		rx_ring->head = 0;
 		rx_ring->tail = 0;
 		rx_ring->base = *base;
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
index a79e6c661ecc..fe897f3e4c12 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-rx.c
@@ -159,8 +159,6 @@ mcp251xfd_hw_rx_obj_to_skb(const struct mcp251xfd_priv *priv,
 
 	if (!(hw_rx_obj->flags & MCP251XFD_OBJ_FLAGS_RTR))
 		memcpy(cfd->data, hw_rx_obj->data, cfd->len);
-
-	mcp251xfd_skb_set_timestamp_raw(priv, skb, hw_rx_obj->ts);
 }
 
 static int
@@ -171,8 +169,26 @@ mcp251xfd_handle_rxif_one(struct mcp251xfd_priv *priv,
 	struct net_device_stats *stats = &priv->ndev->stats;
 	struct sk_buff *skb;
 	struct canfd_frame *cfd;
+	u64 timestamp;
 	int err;
 
+	/* According to mcp2518fd erratum DS80000789E 6. the FIFOCI
+	 * bits of a FIFOSTA register, here the RX FIFO head index
+	 * might be corrupted and we might process past the RX FIFO's
+	 * head into old CAN frames.
+	 *
+	 * Compare the timestamp of currently processed CAN frame with
+	 * last valid frame received. Abort with -EBADMSG if an old
+	 * CAN frame is detected.
+	 */
+	timestamp = timecounter_cyc2time(&priv->tc, hw_rx_obj->ts);
+	if (timestamp <= ring->last_valid) {
+		stats->rx_fifo_errors++;
+
+		return -EBADMSG;
+	}
+	ring->last_valid = timestamp;
+
 	if (hw_rx_obj->flags & MCP251XFD_OBJ_FLAGS_FDF)
 		skb = alloc_canfd_skb(priv->ndev, &cfd);
 	else
@@ -183,6 +199,7 @@ mcp251xfd_handle_rxif_one(struct mcp251xfd_priv *priv,
 		return 0;
 	}
 
+	mcp251xfd_skb_set_timestamp(skb, timestamp);
 	mcp251xfd_hw_rx_obj_to_skb(priv, hw_rx_obj, skb);
 	err = can_rx_offload_queue_timestamp(&priv->offload, skb, hw_rx_obj->ts);
 	if (err)
@@ -265,7 +282,16 @@ mcp251xfd_handle_rxif_ring(struct mcp251xfd_priv *priv,
 			err = mcp251xfd_handle_rxif_one(priv, ring,
 							(void *)hw_rx_obj +
 							i * ring->obj_size);
-			if (err)
+
+			/* -EBADMSG means we're affected by mcp2518fd
+			 * erratum DS80000789E 6., i.e. the timestamp
+			 * in the RX object is older that the last
+			 * valid received CAN frame. Don't process any
+			 * further and mark processed frames as good.
+			 */
+			if (err == -EBADMSG)
+				return mcp251xfd_handle_rxif_ring_uinc(priv, ring, i);
+			else if (err)
 				return err;
 		}
 
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
index 453a2b193c75..d32ece3d7aee 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
@@ -552,6 +552,9 @@ struct mcp251xfd_rx_ring {
 	unsigned int head;
 	unsigned int tail;
 
+	/* timestamp of the last valid received CAN frame */
+	u64 last_valid;
+
 	u16 base;
 	u8 nr;
 	u8 fifo_nr;
-- 
cgit v1.2.3-58-ga151


From b8e0ddd36ce9536ad7478dd27df06c9ae92370ba Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Sun, 22 Jan 2023 21:30:41 +0100
Subject: can: mcp251xfd: tef: prepare to workaround broken TEF FIFO tail index
 erratum
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This is a preparatory patch to work around a problem similar to
erratum DS80000789E 6 of the mcp2518fd, the other variants of the chip
family (mcp2517fd and mcp251863) are probably also affected.

Erratum DS80000789E 6 says "reading of the FIFOCI bits in the FIFOSTA
register for an RX FIFO may be corrupted". However observation shows
that this problem is not limited to RX FIFOs but also effects the TEF
FIFO.

When handling the TEF interrupt, the driver reads the FIFO header
index from the TEF FIFO STA register of the chip.

In the bad case, the driver reads a too large head index. In the
original code, the driver always trusted the read value, which caused
old CAN transmit complete events that were already processed to be
re-processed.

Instead of reading and trusting the head index, read the head index
and calculate the number of CAN frames that were supposedly received -
replace mcp251xfd_tef_ring_update() with mcp251xfd_get_tef_len().

The mcp251xfd_handle_tefif() function reads the CAN transmit complete
events from the chip, iterates over them and pushes them into the
network stack. The original driver already contains code to detect old
CAN transmit complete events, that will be updated in the next patch.

Cc: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Cc: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c |  2 +
 drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c  | 54 ++++++++++++++++++--------
 drivers/net/can/spi/mcp251xfd/mcp251xfd.h      | 13 ++-----
 3 files changed, 43 insertions(+), 26 deletions(-)

diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c
index 8464fc4f37d0..7bd2bcb5cf87 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-ring.c
@@ -486,6 +486,8 @@ int mcp251xfd_ring_alloc(struct mcp251xfd_priv *priv)
 		clear_bit(MCP251XFD_FLAGS_FD_MODE, priv->flags);
 	}
 
+	tx_ring->obj_num_shift_to_u8 = BITS_PER_TYPE(tx_ring->obj_num) -
+		ilog2(tx_ring->obj_num);
 	tx_ring->obj_size = tx_obj_size;
 
 	rem = priv->rx_obj_num;
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
index 4bc8d71fc491..3d9c348ad868 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
@@ -2,7 +2,7 @@
 //
 // mcp251xfd - Microchip MCP251xFD Family CAN controller driver
 //
-// Copyright (c) 2019, 2020, 2021 Pengutronix,
+// Copyright (c) 2019, 2020, 2021, 2023 Pengutronix,
 //               Marc Kleine-Budde <kernel@pengutronix.de>
 //
 // Based on:
@@ -16,6 +16,11 @@
 
 #include "mcp251xfd.h"
 
+static inline bool mcp251xfd_tx_fifo_sta_full(u32 fifo_sta)
+{
+	return !(fifo_sta & MCP251XFD_REG_FIFOSTA_TFNRFNIF);
+}
+
 static inline int
 mcp251xfd_tef_tail_get_from_chip(const struct mcp251xfd_priv *priv,
 				 u8 *tef_tail)
@@ -120,28 +125,44 @@ mcp251xfd_handle_tefif_one(struct mcp251xfd_priv *priv,
 	return 0;
 }
 
-static int mcp251xfd_tef_ring_update(struct mcp251xfd_priv *priv)
+static int
+mcp251xfd_get_tef_len(struct mcp251xfd_priv *priv, u8 *len_p)
 {
 	const struct mcp251xfd_tx_ring *tx_ring = priv->tx;
-	unsigned int new_head;
-	u8 chip_tx_tail;
+	const u8 shift = tx_ring->obj_num_shift_to_u8;
+	u8 chip_tx_tail, tail, len;
+	u32 fifo_sta;
 	int err;
 
-	err = mcp251xfd_tx_tail_get_from_chip(priv, &chip_tx_tail);
+	err = regmap_read(priv->map_reg, MCP251XFD_REG_FIFOSTA(priv->tx->fifo_nr),
+			  &fifo_sta);
 	if (err)
 		return err;
 
-	/* chip_tx_tail, is the next TX-Object send by the HW.
-	 * The new TEF head must be >= the old head, ...
+	if (mcp251xfd_tx_fifo_sta_full(fifo_sta)) {
+		*len_p = tx_ring->obj_num;
+		return 0;
+	}
+
+	chip_tx_tail = FIELD_GET(MCP251XFD_REG_FIFOSTA_FIFOCI_MASK, fifo_sta);
+
+	err =  mcp251xfd_check_tef_tail(priv);
+	if (err)
+		return err;
+	tail = mcp251xfd_get_tef_tail(priv);
+
+	/* First shift to full u8. The subtraction works on signed
+	 * values, that keeps the difference steady around the u8
+	 * overflow. The right shift acts on len, which is an u8.
 	 */
-	new_head = round_down(priv->tef->head, tx_ring->obj_num) + chip_tx_tail;
-	if (new_head <= priv->tef->head)
-		new_head += tx_ring->obj_num;
+	BUILD_BUG_ON(sizeof(tx_ring->obj_num) != sizeof(chip_tx_tail));
+	BUILD_BUG_ON(sizeof(tx_ring->obj_num) != sizeof(tail));
+	BUILD_BUG_ON(sizeof(tx_ring->obj_num) != sizeof(len));
 
-	/* ... but it cannot exceed the TX head. */
-	priv->tef->head = min(new_head, tx_ring->head);
+	len = (chip_tx_tail << shift) - (tail << shift);
+	*len_p = len >> shift;
 
-	return mcp251xfd_check_tef_tail(priv);
+	return 0;
 }
 
 static inline int
@@ -182,13 +203,12 @@ int mcp251xfd_handle_tefif(struct mcp251xfd_priv *priv)
 	u8 tef_tail, len, l;
 	int err, i;
 
-	err = mcp251xfd_tef_ring_update(priv);
+	err = mcp251xfd_get_tef_len(priv, &len);
 	if (err)
 		return err;
 
 	tef_tail = mcp251xfd_get_tef_tail(priv);
-	len = mcp251xfd_get_tef_len(priv);
-	l = mcp251xfd_get_tef_linear_len(priv);
+	l = mcp251xfd_get_tef_linear_len(priv, len);
 	err = mcp251xfd_tef_obj_read(priv, hw_tef_obj, tef_tail, l);
 	if (err)
 		return err;
@@ -223,6 +243,8 @@ out_netif_wake_queue:
 		struct mcp251xfd_tx_ring *tx_ring = priv->tx;
 		int offset;
 
+		ring->head += len;
+
 		/* Increment the TEF FIFO tail pointer 'len' times in
 		 * a single SPI message.
 		 *
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
index d32ece3d7aee..dcbbd2b2fae8 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd.h
@@ -524,6 +524,7 @@ struct mcp251xfd_tef_ring {
 
 	/* u8 obj_num equals tx_ring->obj_num */
 	/* u8 obj_size equals sizeof(struct mcp251xfd_hw_tef_obj) */
+	/* u8 obj_num_shift_to_u8 equals tx_ring->obj_num_shift_to_u8 */
 
 	union mcp251xfd_write_reg_buf irq_enable_buf;
 	struct spi_transfer irq_enable_xfer;
@@ -542,6 +543,7 @@ struct mcp251xfd_tx_ring {
 	u8 nr;
 	u8 fifo_nr;
 	u8 obj_num;
+	u8 obj_num_shift_to_u8;
 	u8 obj_size;
 
 	struct mcp251xfd_tx_obj obj[MCP251XFD_TX_OBJ_NUM_MAX];
@@ -882,17 +884,8 @@ static inline u8 mcp251xfd_get_tef_tail(const struct mcp251xfd_priv *priv)
 	return priv->tef->tail & (priv->tx->obj_num - 1);
 }
 
-static inline u8 mcp251xfd_get_tef_len(const struct mcp251xfd_priv *priv)
-{
-	return priv->tef->head - priv->tef->tail;
-}
-
-static inline u8 mcp251xfd_get_tef_linear_len(const struct mcp251xfd_priv *priv)
+static inline u8 mcp251xfd_get_tef_linear_len(const struct mcp251xfd_priv *priv, u8 len)
 {
-	u8 len;
-
-	len = mcp251xfd_get_tef_len(priv);
-
 	return min_t(u8, len, priv->tx->obj_num - mcp251xfd_get_tef_tail(priv));
 }
 
-- 
cgit v1.2.3-58-ga151


From 3a0a88fcbaf9e027ecca3fe8775be9700b4d6460 Mon Sep 17 00:00:00 2001
From: Marc Kleine-Budde <mkl@pengutronix.de>
Date: Sun, 22 Jan 2023 22:35:03 +0100
Subject: can: mcp251xfd: tef: update workaround for erratum DS80000789E 6 of
 mcp2518fd
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch updates the workaround for a problem similar to erratum
DS80000789E 6 of the mcp2518fd, the other variants of the chip
family (mcp2517fd and mcp251863) are probably also affected.

Erratum DS80000789E 6 says "reading of the FIFOCI bits in the FIFOSTA
register for an RX FIFO may be corrupted". However observation shows
that this problem is not limited to RX FIFOs but also effects the TEF
FIFO.

In the bad case, the driver reads a too large head index. As the FIFO
is implemented as a ring buffer, this results in re-handling old CAN
transmit complete events.

Every transmit complete event contains with a sequence number that
equals to the sequence number of the corresponding TX request. This
way old TX complete events can be detected.

If the original driver detects a non matching sequence number, it
prints an info message and tries again later. As wrong sequence
numbers can be explained by the erratum DS80000789E 6, demote the info
message to debug level, streamline the code and update the comments.

Keep the behavior: If an old CAN TX complete event is detected, abort
the iteration and mark the number of valid CAN TX complete events as
processed in the chip by incrementing the FIFO's tail index.

Cc: Stefan Althöfer <Stefan.Althoefer@janztec.com>
Cc: Thomas Kopp <thomas.kopp@microchip.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c | 71 ++++++++++-----------------
 1 file changed, 27 insertions(+), 44 deletions(-)

diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
index 3d9c348ad868..f732556d233a 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c
@@ -60,56 +60,39 @@ static int mcp251xfd_check_tef_tail(const struct mcp251xfd_priv *priv)
 	return 0;
 }
 
-static int
-mcp251xfd_handle_tefif_recover(const struct mcp251xfd_priv *priv, const u32 seq)
-{
-	const struct mcp251xfd_tx_ring *tx_ring = priv->tx;
-	u32 tef_sta;
-	int err;
-
-	err = regmap_read(priv->map_reg, MCP251XFD_REG_TEFSTA, &tef_sta);
-	if (err)
-		return err;
-
-	if (tef_sta & MCP251XFD_REG_TEFSTA_TEFOVIF) {
-		netdev_err(priv->ndev,
-			   "Transmit Event FIFO buffer overflow.\n");
-		return -ENOBUFS;
-	}
-
-	netdev_info(priv->ndev,
-		    "Transmit Event FIFO buffer %s. (seq=0x%08x, tef_tail=0x%08x, tef_head=0x%08x, tx_head=0x%08x).\n",
-		    tef_sta & MCP251XFD_REG_TEFSTA_TEFFIF ?
-		    "full" : tef_sta & MCP251XFD_REG_TEFSTA_TEFNEIF ?
-		    "not empty" : "empty",
-		    seq, priv->tef->tail, priv->tef->head, tx_ring->head);
-
-	/* The Sequence Number in the TEF doesn't match our tef_tail. */
-	return -EAGAIN;
-}
-
 static int
 mcp251xfd_handle_tefif_one(struct mcp251xfd_priv *priv,
 			   const struct mcp251xfd_hw_tef_obj *hw_tef_obj,
 			   unsigned int *frame_len_ptr)
 {
 	struct net_device_stats *stats = &priv->ndev->stats;
+	u32 seq, tef_tail_masked, tef_tail;
 	struct sk_buff *skb;
-	u32 seq, seq_masked, tef_tail_masked, tef_tail;
 
-	seq = FIELD_GET(MCP251XFD_OBJ_FLAGS_SEQ_MCP2518FD_MASK,
+	 /* Use the MCP2517FD mask on the MCP2518FD, too. We only
+	  * compare 7 bits, this is enough to detect old TEF objects.
+	  */
+	seq = FIELD_GET(MCP251XFD_OBJ_FLAGS_SEQ_MCP2517FD_MASK,
 			hw_tef_obj->flags);
-
-	/* Use the MCP2517FD mask on the MCP2518FD, too. We only
-	 * compare 7 bits, this should be enough to detect
-	 * net-yet-completed, i.e. old TEF objects.
-	 */
-	seq_masked = seq &
-		field_mask(MCP251XFD_OBJ_FLAGS_SEQ_MCP2517FD_MASK);
 	tef_tail_masked = priv->tef->tail &
 		field_mask(MCP251XFD_OBJ_FLAGS_SEQ_MCP2517FD_MASK);
-	if (seq_masked != tef_tail_masked)
-		return mcp251xfd_handle_tefif_recover(priv, seq);
+
+	/* According to mcp2518fd erratum DS80000789E 6. the FIFOCI
+	 * bits of a FIFOSTA register, here the TX FIFO tail index
+	 * might be corrupted and we might process past the TEF FIFO's
+	 * head into old CAN frames.
+	 *
+	 * Compare the sequence number of the currently processed CAN
+	 * frame with the expected sequence number. Abort with
+	 * -EBADMSG if an old CAN frame is detected.
+	 */
+	if (seq != tef_tail_masked) {
+		netdev_dbg(priv->ndev, "%s: chip=0x%02x ring=0x%02x\n", __func__,
+			   seq, tef_tail_masked);
+		stats->tx_fifo_errors++;
+
+		return -EBADMSG;
+	}
 
 	tef_tail = mcp251xfd_get_tef_tail(priv);
 	skb = priv->can.echo_skb[tef_tail];
@@ -223,12 +206,12 @@ int mcp251xfd_handle_tefif(struct mcp251xfd_priv *priv)
 		unsigned int frame_len = 0;
 
 		err = mcp251xfd_handle_tefif_one(priv, &hw_tef_obj[i], &frame_len);
-		/* -EAGAIN means the Sequence Number in the TEF
-		 * doesn't match our tef_tail. This can happen if we
-		 * read the TEF objects too early. Leave loop let the
-		 * interrupt handler call us again.
+		/* -EBADMSG means we're affected by mcp2518fd erratum
+		 * DS80000789E 6., i.e. the Sequence Number in the TEF
+		 * doesn't match our tef_tail. Don't process any
+		 * further and mark processed frames as good.
 		 */
-		if (err == -EAGAIN)
+		if (err == -EBADMSG)
 			goto out_netif_wake_queue;
 		if (err)
 			return err;
-- 
cgit v1.2.3-58-ga151


From 10154dbded6d6a2fecaebdfda206609de0f121a9 Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Wed, 26 Jun 2024 19:51:26 +0200
Subject: udp: Allow GSO transmit from devices with no checksum offload

Today sending a UDP GSO packet from a TUN device results in an EIO error:

  import fcntl, os, struct
  from socket import *

  TUNSETIFF = 0x400454CA
  IFF_TUN = 0x0001
  IFF_NO_PI = 0x1000
  UDP_SEGMENT = 103

  tun_fd = os.open("/dev/net/tun", os.O_RDWR)
  ifr = struct.pack("16sH", b"tun0", IFF_TUN | IFF_NO_PI)
  fcntl.ioctl(tun_fd, TUNSETIFF, ifr)

  os.system("ip addr add 192.0.2.1/24 dev tun0")
  os.system("ip link set dev tun0 up")

  s = socket(AF_INET, SOCK_DGRAM)
  s.setsockopt(SOL_UDP, UDP_SEGMENT, 1200)
  s.sendto(b"x" * 3000, ("192.0.2.2", 9)) # EIO

This is due to a check in the udp stack if the egress device offers
checksum offload. While TUN/TAP devices, by default, don't advertise this
capability because it requires support from the TUN/TAP reader.

However, the GSO stack has a software fallback for checksum calculation,
which we can use. This way we don't force UDP_SEGMENT users to handle the
EIO error and implement a segmentation fallback.

Lift the restriction so that UDP_SEGMENT can be used with any egress
device. We also need to adjust the UDP GSO code to match the GSO stack
expectation about ip_summed field, as set in commit 8d63bee643f1 ("net:
avoid skb_warn_bad_offload false positives on UFO"). Otherwise we will hit
the bad offload check.

Users should, however, expect a potential performance impact when
batch-sending packets with UDP_SEGMENT without checksum offload on the
egress device. In such case the packet payload is read twice: first during
the sendmsg syscall when copying data from user memory, and then in the GSO
stack for checksum computation. This double memory read can be less
efficient than a regular sendmsg where the checksum is calculated during
the initial data copy from user memory.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240626-linux-udpgso-v2-1-422dfcbd6b48@cloudflare.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/udp.c         | 3 +--
 net/ipv4/udp_offload.c | 8 ++++++++
 net/ipv6/udp.c         | 3 +--
 3 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index d08bf16d476d..ed97df6af14d 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -938,8 +938,7 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4,
 			kfree_skb(skb);
 			return -EINVAL;
 		}
-		if (skb->ip_summed != CHECKSUM_PARTIAL || is_udplite ||
-		    dst_xfrm(skb_dst(skb))) {
+		if (is_udplite || dst_xfrm(skb_dst(skb))) {
 			kfree_skb(skb);
 			return -EIO;
 		}
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index 59448a2dbf2c..aa2e0a28ca61 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -357,6 +357,14 @@ struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb,
 	else
 		uh->check = gso_make_checksum(seg, ~check) ? : CSUM_MANGLED_0;
 
+	/* On the TX path, CHECKSUM_NONE and CHECKSUM_UNNECESSARY have the same
+	 * meaning. However, check for bad offloads in the GSO stack expects the
+	 * latter, if the checksum was calculated in software. To vouch for the
+	 * segment skbs we actually need to set it on the gso_skb.
+	 */
+	if (gso_skb->ip_summed == CHECKSUM_NONE)
+		gso_skb->ip_summed = CHECKSUM_UNNECESSARY;
+
 	/* update refcount for the packet */
 	if (copy_dtor) {
 		int delta = sum_truesize - gso_skb->truesize;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index b56f0b9f4307..b5456394cc67 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1257,8 +1257,7 @@ static int udp_v6_send_skb(struct sk_buff *skb, struct flowi6 *fl6,
 			kfree_skb(skb);
 			return -EINVAL;
 		}
-		if (skb->ip_summed != CHECKSUM_PARTIAL || is_udplite ||
-		    dst_xfrm(skb_dst(skb))) {
+		if (is_udplite || dst_xfrm(skb_dst(skb))) {
 			kfree_skb(skb);
 			return -EIO;
 		}
-- 
cgit v1.2.3-58-ga151


From 3e400219c04dcfcada224303c8efa2c91beaae0b Mon Sep 17 00:00:00 2001
From: Jakub Sitnicki <jakub@cloudflare.com>
Date: Wed, 26 Jun 2024 19:51:27 +0200
Subject: selftests/net: Add test coverage for UDP GSO software fallback

Extend the existing test to exercise UDP GSO egress through devices with
various offload capabilities, including lack of checksum offload, which is
the default case for TUN/TAP devices.

Test against a dummy device because it is simpler to set up then TUN/TAP.

Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240626-linux-udpgso-v2-2-422dfcbd6b48@cloudflare.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/udpgso.c  | 15 +++++++++---
 tools/testing/selftests/net/udpgso.sh | 43 +++++++++++++++++++++++++++++++++++
 2 files changed, 55 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/net/udpgso.c b/tools/testing/selftests/net/udpgso.c
index 85b3baa3f7f3..3e74cfa1a2bf 100644
--- a/tools/testing/selftests/net/udpgso.c
+++ b/tools/testing/selftests/net/udpgso.c
@@ -53,6 +53,7 @@ static bool		cfg_do_ipv6;
 static bool		cfg_do_connected;
 static bool		cfg_do_connectionless;
 static bool		cfg_do_msgmore;
+static bool		cfg_do_recv = true;
 static bool		cfg_do_setsockopt;
 static int		cfg_specific_test_id = -1;
 
@@ -414,6 +415,9 @@ static void run_one(struct testcase *test, int fdt, int fdr,
 	if (!sent)
 		return;
 
+	if (!cfg_do_recv)
+		return;
+
 	if (test->gso_len)
 		mss = test->gso_len;
 	else
@@ -464,8 +468,10 @@ static void run_test(struct sockaddr *addr, socklen_t alen)
 	if (fdr == -1)
 		error(1, errno, "socket r");
 
-	if (bind(fdr, addr, alen))
-		error(1, errno, "bind");
+	if (cfg_do_recv) {
+		if (bind(fdr, addr, alen))
+			error(1, errno, "bind");
+	}
 
 	/* Have tests fail quickly instead of hang */
 	if (setsockopt(fdr, SOL_SOCKET, SO_RCVTIMEO, &tv, sizeof(tv)))
@@ -524,7 +530,7 @@ static void parse_opts(int argc, char **argv)
 {
 	int c;
 
-	while ((c = getopt(argc, argv, "46cCmst:")) != -1) {
+	while ((c = getopt(argc, argv, "46cCmRst:")) != -1) {
 		switch (c) {
 		case '4':
 			cfg_do_ipv4 = true;
@@ -541,6 +547,9 @@ static void parse_opts(int argc, char **argv)
 		case 'm':
 			cfg_do_msgmore = true;
 			break;
+		case 'R':
+			cfg_do_recv = false;
+			break;
 		case 's':
 			cfg_do_setsockopt = true;
 			break;
diff --git a/tools/testing/selftests/net/udpgso.sh b/tools/testing/selftests/net/udpgso.sh
index 6c63178086b0..85d1fa3c1ff7 100755
--- a/tools/testing/selftests/net/udpgso.sh
+++ b/tools/testing/selftests/net/udpgso.sh
@@ -27,6 +27,31 @@ test_route_mtu() {
 	ip route add local fd00::1/128 table local dev lo mtu 1500
 }
 
+setup_dummy_sink() {
+	ip link add name sink mtu 1500 type dummy
+	ip addr add dev sink 10.0.0.0/24
+	ip addr add dev sink fd00::2/64 nodad
+	ip link set dev sink up
+}
+
+test_hw_gso_hw_csum() {
+	setup_dummy_sink
+	ethtool -K sink tx-checksum-ip-generic on >/dev/null
+	ethtool -K sink tx-udp-segmentation on >/dev/null
+}
+
+test_sw_gso_hw_csum() {
+	setup_dummy_sink
+	ethtool -K sink tx-checksum-ip-generic on >/dev/null
+	ethtool -K sink tx-udp-segmentation off >/dev/null
+}
+
+test_sw_gso_sw_csum() {
+	setup_dummy_sink
+	ethtool -K sink tx-checksum-ip-generic off >/dev/null
+	ethtool -K sink tx-udp-segmentation off >/dev/null
+}
+
 if [ "$#" -gt 0 ]; then
 	"$1"
 	shift 2 # pop "test_*" arg and "--" delimiter
@@ -56,3 +81,21 @@ echo "ipv4 msg_more"
 
 echo "ipv6 msg_more"
 ./in_netns.sh "$0" test_dev_mtu -- ./udpgso -6 -C -m
+
+echo "ipv4 hw-gso hw-csum"
+./in_netns.sh "$0" test_hw_gso_hw_csum -- ./udpgso -4 -C -R
+
+echo "ipv6 hw-gso hw-csum"
+./in_netns.sh "$0" test_hw_gso_hw_csum -- ./udpgso -6 -C -R
+
+echo "ipv4 sw-gso hw-csum"
+./in_netns.sh "$0" test_sw_gso_hw_csum -- ./udpgso -4 -C -R
+
+echo "ipv6 sw-gso hw-csum"
+./in_netns.sh "$0" test_sw_gso_hw_csum -- ./udpgso -6 -C -R
+
+echo "ipv4 sw-gso sw-csum"
+./in_netns.sh "$0" test_sw_gso_sw_csum -- ./udpgso -4 -C -R
+
+echo "ipv6 sw-gso sw-csum"
+./in_netns.sh "$0" test_sw_gso_sw_csum -- ./udpgso -6 -C -R
-- 
cgit v1.2.3-58-ga151


From bf7bb7b43097d8c2b8673c5ea8a6b64a1f6090b0 Mon Sep 17 00:00:00 2001
From: Jon Kohler <jon@nutanix.com>
Date: Thu, 27 Jun 2024 13:20:13 -0700
Subject: enic: add ethtool get_channel support

Add .get_channel to enic_ethtool_ops to enable basic ethtool -l
support to get the current channel configuration.

Note that the driver does not support dynamically changing queue
configuration, so .set_channel is intentionally unused. Instead, users
should use Cisco's hardware management tools (UCSM/IMC) to modify
virtual interface card configuration out of band.

Signed-off-by: Jon Kohler <jon@nutanix.com>
Link: https://patch.msgid.link/20240627202013.2398217-1-jon@nutanix.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cisco/enic/enic_ethtool.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/drivers/net/ethernet/cisco/enic/enic_ethtool.c b/drivers/net/ethernet/cisco/enic/enic_ethtool.c
index 241906697019..a42f3f280f3e 100644
--- a/drivers/net/ethernet/cisco/enic/enic_ethtool.c
+++ b/drivers/net/ethernet/cisco/enic/enic_ethtool.c
@@ -608,6 +608,28 @@ static int enic_get_ts_info(struct net_device *netdev,
 	return 0;
 }
 
+static void enic_get_channels(struct net_device *netdev,
+			      struct ethtool_channels *channels)
+{
+	struct enic *enic = netdev_priv(netdev);
+
+	switch (vnic_dev_get_intr_mode(enic->vdev)) {
+	case VNIC_DEV_INTR_MODE_MSIX:
+		channels->max_rx = ENIC_RQ_MAX;
+		channels->max_tx = ENIC_WQ_MAX;
+		channels->rx_count = enic->rq_count;
+		channels->tx_count = enic->wq_count;
+		break;
+	case VNIC_DEV_INTR_MODE_MSI:
+	case VNIC_DEV_INTR_MODE_INTX:
+		channels->max_combined = 1;
+		channels->combined_count = 1;
+		break;
+	default:
+		break;
+	}
+}
+
 static const struct ethtool_ops enic_ethtool_ops = {
 	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
 				     ETHTOOL_COALESCE_USE_ADAPTIVE_RX |
@@ -632,6 +654,7 @@ static const struct ethtool_ops enic_ethtool_ops = {
 	.set_rxfh = enic_set_rxfh,
 	.get_link_ksettings = enic_get_ksettings,
 	.get_ts_info = enic_get_ts_info,
+	.get_channels = enic_get_channels,
 };
 
 void enic_set_ethtool_ops(struct net_device *netdev)
-- 
cgit v1.2.3-58-ga151


From 147997afaad0ce4835233e7fddacc9e2be7eb087 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 27 Jun 2024 11:55:00 -0700
Subject: selftests: net: ksft: avoid continue when handling results

Exception handlers print the result and use continue
to skip the non-exception result printing. This makes
inserting common post-test code hard. Refactor to
avoid the continues and have only one ktap_result() call.

Reviewed-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/20240627185502.3069139-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/lib/py/ksft.py | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/tools/testing/selftests/net/lib/py/ksft.py b/tools/testing/selftests/net/lib/py/ksft.py
index b6ce3f33d41e..789433262dc7 100644
--- a/tools/testing/selftests/net/lib/py/ksft.py
+++ b/tools/testing/selftests/net/lib/py/ksft.py
@@ -130,29 +130,29 @@ def ksft_run(cases=None, globs=None, case_pfx=None, args=()):
     for case in cases:
         KSFT_RESULT = True
         cnt += 1
+        comment = ""
+        cnt_key = ""
+
         try:
             case(*args)
         except KsftSkipEx as e:
-            ktap_result(True, cnt, case, comment="SKIP " + str(e))
-            totals['skip'] += 1
-            continue
+            comment = "SKIP " + str(e)
+            cnt_key = 'skip'
         except KsftXfailEx as e:
-            ktap_result(True, cnt, case, comment="XFAIL " + str(e))
-            totals['xfail'] += 1
-            continue
+            comment = "XFAIL " + str(e)
+            cnt_key = 'xfail'
         except Exception as e:
             tb = traceback.format_exc()
             for line in tb.strip().split('\n'):
                 ksft_pr("Exception|", line)
-            ktap_result(False, cnt, case)
-            totals['fail'] += 1
-            continue
-
-        ktap_result(KSFT_RESULT, cnt, case)
-        if KSFT_RESULT:
-            totals['pass'] += 1
-        else:
-            totals['fail'] += 1
+            KSFT_RESULT = False
+            cnt_key = 'fail'
+
+        if not cnt_key:
+            cnt_key = 'pass' if KSFT_RESULT else 'fail'
+
+        ktap_result(KSFT_RESULT, cnt, case, comment=comment)
+        totals[cnt_key] += 1
 
     print(
         f"# Totals: pass:{totals['pass']} fail:{totals['fail']} xfail:{totals['xfail']} xpass:0 skip:{totals['skip']} error:0"
-- 
cgit v1.2.3-58-ga151


From 8510801a9dbd9f0d64079d7061d3452efc752550 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 27 Jun 2024 11:55:01 -0700
Subject: selftests: drv-net: add ability to schedule cleanup with defer()

This implements what I was describing in [1]. When writing a test
author can schedule cleanup / undo actions right after the creation
completes, eg:

  cmd("touch /tmp/file")
  defer(cmd, "rm /tmp/file")

defer() takes the function name as first argument, and the rest are
arguments for that function. defer()red functions are called in
inverse order after test exits. It's also possible to capture them
and execute earlier (in which case they get automatically de-queued).

  undo = defer(cmd, "rm /tmp/file")
  # ... some unsafe code ...
  undo.exec()

As a nice safety all exceptions from defer()ed calls are captured,
printed, and ignored (they do make the test fail, however).
This addresses the common problem of exceptions in cleanup paths
often being unhandled, leading to potential leaks.

There is a global action queue, flushed by ksft_run(). We could support
function level defers too, I guess, but there's no immediate need..

Link: https://lore.kernel.org/all/877cedb2ki.fsf@nvidia.com/ # [1]
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/20240627185502.3069139-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/lib/py/ksft.py  | 21 ++++++++++++++++++
 tools/testing/selftests/net/lib/py/utils.py | 34 +++++++++++++++++++++++++++++
 2 files changed, 55 insertions(+)

diff --git a/tools/testing/selftests/net/lib/py/ksft.py b/tools/testing/selftests/net/lib/py/ksft.py
index 789433262dc7..3aaa2748a58e 100644
--- a/tools/testing/selftests/net/lib/py/ksft.py
+++ b/tools/testing/selftests/net/lib/py/ksft.py
@@ -6,6 +6,7 @@ import sys
 import time
 import traceback
 from .consts import KSFT_MAIN_NAME
+from .utils import global_defer_queue
 
 KSFT_RESULT = None
 KSFT_RESULT_ALL = True
@@ -108,6 +109,24 @@ def ktap_result(ok, cnt=1, case="", comment=""):
     print(res)
 
 
+def ksft_flush_defer():
+    global KSFT_RESULT
+
+    i = 0
+    qlen_start = len(global_defer_queue)
+    while global_defer_queue:
+        i += 1
+        entry = global_defer_queue.pop()
+        try:
+            entry.exec_only()
+        except:
+            ksft_pr(f"Exception while handling defer / cleanup (callback {i} of {qlen_start})!")
+            tb = traceback.format_exc()
+            for line in tb.strip().split('\n'):
+                ksft_pr("Defer Exception|", line)
+            KSFT_RESULT = False
+
+
 def ksft_run(cases=None, globs=None, case_pfx=None, args=()):
     cases = cases or []
 
@@ -148,6 +167,8 @@ def ksft_run(cases=None, globs=None, case_pfx=None, args=()):
             KSFT_RESULT = False
             cnt_key = 'fail'
 
+        ksft_flush_defer()
+
         if not cnt_key:
             cnt_key = 'pass' if KSFT_RESULT else 'fail'
 
diff --git a/tools/testing/selftests/net/lib/py/utils.py b/tools/testing/selftests/net/lib/py/utils.py
index 405aa510aaf2..72590c3f90f1 100644
--- a/tools/testing/selftests/net/lib/py/utils.py
+++ b/tools/testing/selftests/net/lib/py/utils.py
@@ -66,6 +66,40 @@ class bkg(cmd):
         return self.process(terminate=self.terminate, fail=self.check_fail)
 
 
+global_defer_queue = []
+
+
+class defer:
+    def __init__(self, func, *args, **kwargs):
+        global global_defer_queue
+
+        if not callable(func):
+            raise Exception("defer created with un-callable object, did you call the function instead of passing its name?")
+
+        self.func = func
+        self.args = args
+        self.kwargs = kwargs
+
+        self._queue =  global_defer_queue
+        self._queue.append(self)
+
+    def __enter__(self):
+        return self
+
+    def __exit__(self, ex_type, ex_value, ex_tb):
+        return self.exec()
+
+    def exec_only(self):
+        self.func(*self.args, **self.kwargs)
+
+    def cancel(self):
+        self._queue.remove(self)
+
+    def exec(self):
+        self.cancel()
+        self.exec_only()
+
+
 def tool(name, args, json=None, ns=None, host=None):
     cmd_str = name + ' '
     if json:
-- 
cgit v1.2.3-58-ga151


From 0759356bf5fadbd16794bec982a9af48351a02ff Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 27 Jun 2024 11:55:02 -0700
Subject: selftests: drv-net: rss_ctx: convert to defer()

Use just added defer().

Reviewed-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/20240627185502.3069139-4-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/hw/rss_ctx.py | 235 ++++++++++------------
 1 file changed, 103 insertions(+), 132 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
index 68c7d40214eb..3a804e41f7cb 100755
--- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py
+++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
@@ -8,7 +8,7 @@ from lib.py import NetDrvEpEnv
 from lib.py import NetdevFamily
 from lib.py import KsftSkipEx
 from lib.py import rand_port
-from lib.py import ethtool, ip, GenerateTraffic, CmdExitFailure
+from lib.py import ethtool, ip, defer, GenerateTraffic, CmdExitFailure
 
 
 def _rss_key_str(key):
@@ -127,64 +127,56 @@ def test_rss_context(cfg, ctx_cnt=1, create_with_cfg=None):
 
     # Try to allocate more queues when necessary
     qcnt = len(_get_rx_cnts(cfg))
-    if qcnt >= 2 + 2 * ctx_cnt:
-        qcnt = None
-    else:
+    if qcnt < 2 + 2 * ctx_cnt:
         try:
             ksft_pr(f"Increasing queue count {qcnt} -> {2 + 2 * ctx_cnt}")
             ethtool(f"-L {cfg.ifname} combined {2 + 2 * ctx_cnt}")
+            defer(ethtool, f"-L {cfg.ifname} combined {qcnt}")
         except:
             raise KsftSkipEx("Not enough queues for the test")
 
-    ntuple = []
-    ctx_id = []
     ports = []
-    try:
-        # Use queues 0 and 1 for normal traffic
-        ethtool(f"-X {cfg.ifname} equal 2")
 
-        for i in range(ctx_cnt):
-            want_cfg = f"start {2 + i * 2} equal 2"
-            create_cfg = want_cfg if create_with_cfg else ""
-
-            try:
-                ctx_id.append(ethtool_create(cfg, "-X", f"context new {create_cfg}"))
-            except CmdExitFailure:
-                # try to carry on and skip at the end
-                if i == 0:
-                    raise
-                ksft_pr(f"Failed to create context {i + 1}, trying to test what we got")
-                ctx_cnt = i
-                break
-
-            if not create_with_cfg:
-                ethtool(f"-X {cfg.ifname} context {ctx_id[i]} {want_cfg}")
-
-            # Sanity check the context we just created
-            data = get_rss(cfg, ctx_id[i])
-            ksft_eq(min(data['rss-indirection-table']), 2 + i * 2, "Unexpected context cfg: " + str(data))
-            ksft_eq(max(data['rss-indirection-table']), 2 + i * 2 + 1, "Unexpected context cfg: " + str(data))
-
-            ports.append(rand_port())
-            flow = f"flow-type tcp{cfg.addr_ipver} dst-port {ports[i]} context {ctx_id[i]}"
-            ntuple.append(ethtool_create(cfg, "-N", flow))
+    # Use queues 0 and 1 for normal traffic
+    ethtool(f"-X {cfg.ifname} equal 2")
+    defer(ethtool, f"-X {cfg.ifname} default")
 
-        for i in range(ctx_cnt):
-            cnts = _get_rx_cnts(cfg)
-            GenerateTraffic(cfg, port=ports[i]).wait_pkts_and_stop(20000)
-            cnts = _get_rx_cnts(cfg, prev=cnts)
+    for i in range(ctx_cnt):
+        want_cfg = f"start {2 + i * 2} equal 2"
+        create_cfg = want_cfg if create_with_cfg else ""
 
-            ksft_lt(sum(cnts[ :2]), 10000, "traffic on main context:" + str(cnts))
-            ksft_ge(sum(cnts[2+i*2:4+i*2]), 20000, f"traffic on context {i}: " + str(cnts))
-            ksft_eq(sum(cnts[2:2+i*2] + cnts[4+i*2:]), 0, "traffic on other contexts: " + str(cnts))
-    finally:
-        for nid in ntuple:
-            ethtool(f"-N {cfg.ifname} delete {nid}")
-        for cid in ctx_id:
-            ethtool(f"-X {cfg.ifname} context {cid} delete")
-        ethtool(f"-X {cfg.ifname} default")
-        if qcnt:
-            ethtool(f"-L {cfg.ifname} combined {qcnt}")
+        try:
+            ctx_id = ethtool_create(cfg, "-X", f"context new {create_cfg}")
+            defer(ethtool, f"-X {cfg.ifname} context {ctx_id} delete")
+        except CmdExitFailure:
+            # try to carry on and skip at the end
+            if i == 0:
+                raise
+            ksft_pr(f"Failed to create context {i + 1}, trying to test what we got")
+            ctx_cnt = i
+            break
+
+        if not create_with_cfg:
+            ethtool(f"-X {cfg.ifname} context {ctx_id} {want_cfg}")
+
+        # Sanity check the context we just created
+        data = get_rss(cfg, ctx_id)
+        ksft_eq(min(data['rss-indirection-table']), 2 + i * 2, "Unexpected context cfg: " + str(data))
+        ksft_eq(max(data['rss-indirection-table']), 2 + i * 2 + 1, "Unexpected context cfg: " + str(data))
+
+        ports.append(rand_port())
+        flow = f"flow-type tcp{cfg.addr_ipver} dst-port {ports[i]} context {ctx_id}"
+        ntuple = ethtool_create(cfg, "-N", flow)
+        defer(ethtool, f"-N {cfg.ifname} delete {ntuple}")
+
+    for i in range(ctx_cnt):
+        cnts = _get_rx_cnts(cfg)
+        GenerateTraffic(cfg, port=ports[i]).wait_pkts_and_stop(20000)
+        cnts = _get_rx_cnts(cfg, prev=cnts)
+
+        ksft_lt(sum(cnts[ :2]), 10000, "traffic on main context:" + str(cnts))
+        ksft_ge(sum(cnts[2+i*2:4+i*2]), 20000, f"traffic on context {i}: " + str(cnts))
+        ksft_eq(sum(cnts[2:2+i*2] + cnts[4+i*2:]), 0, "traffic on other contexts: " + str(cnts))
 
     if requested_ctx_cnt != ctx_cnt:
         raise KsftSkipEx(f"Tested only {ctx_cnt} contexts, wanted {requested_ctx_cnt}")
@@ -216,24 +208,23 @@ def test_rss_context_out_of_order(cfg, ctx_cnt=4):
 
     # Try to allocate more queues when necessary
     qcnt = len(_get_rx_cnts(cfg))
-    if qcnt >= 2 + 2 * ctx_cnt:
-        qcnt = None
-    else:
+    if qcnt < 2 + 2 * ctx_cnt:
         try:
             ksft_pr(f"Increasing queue count {qcnt} -> {2 + 2 * ctx_cnt}")
             ethtool(f"-L {cfg.ifname} combined {2 + 2 * ctx_cnt}")
+            defer(ethtool, f"-L {cfg.ifname} combined {qcnt}")
         except:
             raise KsftSkipEx("Not enough queues for the test")
 
     ntuple = []
-    ctx_id = []
+    ctx = []
     ports = []
 
     def remove_ctx(idx):
-        ethtool(f"-N {cfg.ifname} delete {ntuple[idx]}")
+        ntuple[idx].exec()
         ntuple[idx] = None
-        ethtool(f"-X {cfg.ifname} context {ctx_id[idx]} delete")
-        ctx_id[idx] = None
+        ctx[idx].exec()
+        ctx[idx] = None
 
     def check_traffic():
         for i in range(ctx_cnt):
@@ -241,7 +232,7 @@ def test_rss_context_out_of_order(cfg, ctx_cnt=4):
             GenerateTraffic(cfg, port=ports[i]).wait_pkts_and_stop(20000)
             cnts = _get_rx_cnts(cfg, prev=cnts)
 
-            if ctx_id[i] is None:
+            if ctx[i]:
                 ksft_lt(sum(cnts[ :2]), 10000, "traffic on main context:" + str(cnts))
                 ksft_ge(sum(cnts[2+i*2:4+i*2]), 20000, f"traffic on context {i}: " + str(cnts))
                 ksft_eq(sum(cnts[2:2+i*2] + cnts[4+i*2:]), 0, "traffic on other contexts: " + str(cnts))
@@ -249,41 +240,32 @@ def test_rss_context_out_of_order(cfg, ctx_cnt=4):
                 ksft_ge(sum(cnts[ :2]), 20000, "traffic on main context:" + str(cnts))
                 ksft_eq(sum(cnts[2: ]),     0, "traffic on other contexts: " + str(cnts))
 
-    try:
-        # Use queues 0 and 1 for normal traffic
-        ethtool(f"-X {cfg.ifname} equal 2")
-
-        for i in range(ctx_cnt):
-            ctx_id.append(ethtool_create(cfg, "-X", f"context new start {2 + i * 2} equal 2"))
+    # Use queues 0 and 1 for normal traffic
+    ethtool(f"-X {cfg.ifname} equal 2")
+    defer(ethtool, f"-X {cfg.ifname} default")
 
-            ports.append(rand_port())
-            flow = f"flow-type tcp{cfg.addr_ipver} dst-port {ports[i]} context {ctx_id[i]}"
-            ntuple.append(ethtool_create(cfg, "-N", flow))
+    for i in range(ctx_cnt):
+        ctx_id = ethtool_create(cfg, "-X", f"context new start {2 + i * 2} equal 2")
+        ctx.append(defer(ethtool, f"-X {cfg.ifname} context {ctx_id} delete"))
 
-        check_traffic()
+        ports.append(rand_port())
+        flow = f"flow-type tcp{cfg.addr_ipver} dst-port {ports[i]} context {ctx_id}"
+        ntuple_id = ethtool_create(cfg, "-N", flow)
+        ntuple.append(defer(ethtool, f"-N {cfg.ifname} delete {ntuple_id}"))
 
-        # Remove middle context
-        remove_ctx(ctx_cnt // 2)
-        check_traffic()
+    check_traffic()
 
-        # Remove first context
-        remove_ctx(0)
-        check_traffic()
+    # Remove middle context
+    remove_ctx(ctx_cnt // 2)
+    check_traffic()
 
-        # Remove last context
-        remove_ctx(-1)
-        check_traffic()
+    # Remove first context
+    remove_ctx(0)
+    check_traffic()
 
-    finally:
-        for nid in ntuple:
-            if nid is not None:
-                ethtool(f"-N {cfg.ifname} delete {nid}")
-        for cid in ctx_id:
-            if cid is not None:
-                ethtool(f"-X {cfg.ifname} context {cid} delete")
-        ethtool(f"-X {cfg.ifname} default")
-        if qcnt:
-            ethtool(f"-L {cfg.ifname} combined {qcnt}")
+    # Remove last context
+    remove_ctx(-1)
+    check_traffic()
 
     if requested_ctx_cnt != ctx_cnt:
         raise KsftSkipEx(f"Tested only {ctx_cnt} contexts, wanted {requested_ctx_cnt}")
@@ -298,69 +280,58 @@ def test_rss_context_overlap(cfg, other_ctx=0):
     require_ntuple(cfg)
 
     queue_cnt = len(_get_rx_cnts(cfg))
-    if queue_cnt >= 4:
-        queue_cnt = None
-    else:
+    if queue_cnt < 4:
         try:
             ksft_pr(f"Increasing queue count {queue_cnt} -> 4")
             ethtool(f"-L {cfg.ifname} combined 4")
+            defer(ethtool, f"-L {cfg.ifname} combined {queue_cnt}")
         except:
             raise KsftSkipEx("Not enough queues for the test")
 
-    ctx_id = None
-    ntuple = None
     if other_ctx == 0:
         ethtool(f"-X {cfg.ifname} equal 4")
+        defer(ethtool, f"-X {cfg.ifname} default")
     else:
         other_ctx = ethtool_create(cfg, "-X", "context new")
         ethtool(f"-X {cfg.ifname} context {other_ctx} equal 4")
+        defer(ethtool, f"-X {cfg.ifname} context {other_ctx} delete")
 
-    try:
-        ctx_id = ethtool_create(cfg, "-X", "context new")
-        ethtool(f"-X {cfg.ifname} context {ctx_id} start 2 equal 2")
+    ctx_id = ethtool_create(cfg, "-X", "context new")
+    ethtool(f"-X {cfg.ifname} context {ctx_id} start 2 equal 2")
+    defer(ethtool, f"-X {cfg.ifname} context {ctx_id} delete")
 
-        port = rand_port()
-        if other_ctx:
-            flow = f"flow-type tcp{cfg.addr_ipver} dst-port {port} context {other_ctx}"
-            ntuple = ethtool_create(cfg, "-N", flow)
+    port = rand_port()
+    if other_ctx:
+        flow = f"flow-type tcp{cfg.addr_ipver} dst-port {port} context {other_ctx}"
+        ntuple_id = ethtool_create(cfg, "-N", flow)
+        ntuple = defer(ethtool, f"-N {cfg.ifname} delete {ntuple_id}")
 
-        # Test the main context
-        cnts = _get_rx_cnts(cfg)
-        GenerateTraffic(cfg, port=port).wait_pkts_and_stop(20000)
-        cnts = _get_rx_cnts(cfg, prev=cnts)
+    # Test the main context
+    cnts = _get_rx_cnts(cfg)
+    GenerateTraffic(cfg, port=port).wait_pkts_and_stop(20000)
+    cnts = _get_rx_cnts(cfg, prev=cnts)
 
-        ksft_ge(sum(cnts[ :4]), 20000, "traffic on main context: " + str(cnts))
-        ksft_ge(sum(cnts[ :2]),  7000, "traffic on main context (1/2): " + str(cnts))
-        ksft_ge(sum(cnts[2:4]),  7000, "traffic on main context (2/2): " + str(cnts))
-        if other_ctx == 0:
-            ksft_eq(sum(cnts[4: ]),     0, "traffic on other queues: " + str(cnts))
-
-        # Now create a rule for context 1 and make sure traffic goes to a subset
-        if other_ctx:
-            ethtool(f"-N {cfg.ifname} delete {ntuple}")
-            ntuple = None
-        flow = f"flow-type tcp{cfg.addr_ipver} dst-port {port} context {ctx_id}"
-        ntuple = ethtool_create(cfg, "-N", flow)
+    ksft_ge(sum(cnts[ :4]), 20000, "traffic on main context: " + str(cnts))
+    ksft_ge(sum(cnts[ :2]),  7000, "traffic on main context (1/2): " + str(cnts))
+    ksft_ge(sum(cnts[2:4]),  7000, "traffic on main context (2/2): " + str(cnts))
+    if other_ctx == 0:
+        ksft_eq(sum(cnts[4: ]),     0, "traffic on other queues: " + str(cnts))
 
-        cnts = _get_rx_cnts(cfg)
-        GenerateTraffic(cfg, port=port).wait_pkts_and_stop(20000)
-        cnts = _get_rx_cnts(cfg, prev=cnts)
+    # Now create a rule for context 1 and make sure traffic goes to a subset
+    if other_ctx:
+        ntuple.exec()
+    flow = f"flow-type tcp{cfg.addr_ipver} dst-port {port} context {ctx_id}"
+    ntuple_id = ethtool_create(cfg, "-N", flow)
+    defer(ethtool, f"-N {cfg.ifname} delete {ntuple_id}")
 
-        ksft_lt(sum(cnts[ :2]),  7000, "traffic on main context: " + str(cnts))
-        ksft_ge(sum(cnts[2:4]), 20000, "traffic on extra context: " + str(cnts))
-        if other_ctx == 0:
-            ksft_eq(sum(cnts[4: ]),     0, "traffic on other queues: " + str(cnts))
-    finally:
-        if ntuple is not None:
-            ethtool(f"-N {cfg.ifname} delete {ntuple}")
-        if ctx_id:
-            ethtool(f"-X {cfg.ifname} context {ctx_id} delete")
-        if other_ctx == 0:
-            ethtool(f"-X {cfg.ifname} default")
-        else:
-            ethtool(f"-X {cfg.ifname} context {other_ctx} delete")
-        if queue_cnt:
-            ethtool(f"-L {cfg.ifname} combined {queue_cnt}")
+    cnts = _get_rx_cnts(cfg)
+    GenerateTraffic(cfg, port=port).wait_pkts_and_stop(20000)
+    cnts = _get_rx_cnts(cfg, prev=cnts)
+
+    ksft_lt(sum(cnts[ :2]),  7000, "traffic on main context: " + str(cnts))
+    ksft_ge(sum(cnts[2:4]), 20000, "traffic on extra context: " + str(cnts))
+    if other_ctx == 0:
+        ksft_eq(sum(cnts[4: ]),     0, "traffic on other queues: " + str(cnts))
 
 
 def test_rss_context_overlap2(cfg):
-- 
cgit v1.2.3-58-ga151


From 3ebbd9f6de7ec6d538639ebb657246f629ace81e Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Thu, 27 Jun 2024 16:33:46 +0100
Subject: net: move ethtool-related netdev state into its own struct

net_dev->ethtool is a pointer to new struct ethtool_netdev_state, which
 currently contains only the wol_enabled field.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/293a562278371de7534ed1eb17531838ca090633.1719502239.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/realtek/r8169_main.c        | 4 ++--
 drivers/net/ethernet/wangxun/ngbe/ngbe_ethtool.c | 4 ++--
 drivers/net/ethernet/wangxun/ngbe/ngbe_main.c    | 2 +-
 drivers/net/phy/phy.c                            | 2 +-
 drivers/net/phy/phy_device.c                     | 5 +++--
 drivers/net/phy/phylink.c                        | 2 +-
 include/linux/ethtool.h                          | 8 ++++++++
 include/linux/netdevice.h                        | 8 +++++---
 net/core/dev.c                                   | 4 ++++
 net/ethtool/ioctl.c                              | 2 +-
 net/ethtool/wol.c                                | 2 +-
 11 files changed, 29 insertions(+), 14 deletions(-)

diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c
index 9246ea2118ff..714d2e804694 100644
--- a/drivers/net/ethernet/realtek/r8169_main.c
+++ b/drivers/net/ethernet/realtek/r8169_main.c
@@ -1608,7 +1608,7 @@ static void __rtl8169_set_wol(struct rtl8169_private *tp, u32 wolopts)
 
 	if (!tp->dash_enabled) {
 		rtl_set_d3_pll_down(tp, !wolopts);
-		tp->dev->wol_enabled = wolopts ? 1 : 0;
+		tp->dev->ethtool->wol_enabled = wolopts ? 1 : 0;
 	}
 }
 
@@ -5478,7 +5478,7 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		rtl_set_d3_pll_down(tp, true);
 	} else {
 		rtl_set_d3_pll_down(tp, false);
-		dev->wol_enabled = 1;
+		dev->ethtool->wol_enabled = 1;
 	}
 
 	jumbo_max = rtl_jumbo_max(tp);
diff --git a/drivers/net/ethernet/wangxun/ngbe/ngbe_ethtool.c b/drivers/net/ethernet/wangxun/ngbe/ngbe_ethtool.c
index 46a5a3e95202..e868f7ef4920 100644
--- a/drivers/net/ethernet/wangxun/ngbe/ngbe_ethtool.c
+++ b/drivers/net/ethernet/wangxun/ngbe/ngbe_ethtool.c
@@ -37,9 +37,9 @@ static int ngbe_set_wol(struct net_device *netdev,
 	wx->wol = 0;
 	if (wol->wolopts & WAKE_MAGIC)
 		wx->wol = WX_PSR_WKUP_CTL_MAG;
-	netdev->wol_enabled = !!(wx->wol);
+	netdev->ethtool->wol_enabled = !!(wx->wol);
 	wr32(wx, WX_PSR_WKUP_CTL, wx->wol);
-	device_set_wakeup_enable(&pdev->dev, netdev->wol_enabled);
+	device_set_wakeup_enable(&pdev->dev, netdev->ethtool->wol_enabled);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
index e894e01d030d..a8119de60deb 100644
--- a/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
+++ b/drivers/net/ethernet/wangxun/ngbe/ngbe_main.c
@@ -650,7 +650,7 @@ static int ngbe_probe(struct pci_dev *pdev,
 	if (wx->wol_hw_supported)
 		wx->wol = NGBE_PSR_WKUP_CTL_MAG;
 
-	netdev->wol_enabled = !!(wx->wol);
+	netdev->ethtool->wol_enabled = !!(wx->wol);
 	wr32(wx, NGBE_PSR_WKUP_CTL, wx->wol);
 	device_set_wakeup_enable(&pdev->dev, wx->wol);
 
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index c4236564c1cd..785182fa5fe0 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -1309,7 +1309,7 @@ static irqreturn_t phy_interrupt(int irq, void *phy_dat)
 		if (netdev) {
 			struct device *parent = netdev->dev.parent;
 
-			if (netdev->wol_enabled)
+			if (netdev->ethtool->wol_enabled)
 				pm_system_wakeup();
 			else if (device_may_wakeup(&netdev->dev))
 				pm_wakeup_dev_event(&netdev->dev, 0, true);
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 6c6ec9475709..473cbc1d497b 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -296,7 +296,7 @@ static bool mdio_bus_phy_may_suspend(struct phy_device *phydev)
 	if (!netdev)
 		goto out;
 
-	if (netdev->wol_enabled)
+	if (netdev->ethtool->wol_enabled)
 		return false;
 
 	/* As long as not all affected network drivers support the
@@ -1984,7 +1984,8 @@ int phy_suspend(struct phy_device *phydev)
 		return 0;
 
 	phy_ethtool_get_wol(phydev, &wol);
-	phydev->wol_enabled = wol.wolopts || (netdev && netdev->wol_enabled);
+	phydev->wol_enabled = wol.wolopts ||
+			      (netdev && netdev->ethtool->wol_enabled);
 	/* If the device has WOL enabled, we cannot suspend the PHY */
 	if (phydev->wol_enabled && !(phydrv->flags & PHY_ALWAYS_CALL_SUSPEND))
 		return -EBUSY;
diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 6c24c48dcf0f..51c526d227fa 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -2282,7 +2282,7 @@ void phylink_suspend(struct phylink *pl, bool mac_wol)
 {
 	ASSERT_RTNL();
 
-	if (mac_wol && (!pl->netdev || pl->netdev->wol_enabled)) {
+	if (mac_wol && (!pl->netdev || pl->netdev->ethtool->wol_enabled)) {
 		/* Wake-on-Lan enabled, MAC handling */
 		mutex_lock(&pl->state_mutex);
 
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index c7f6f2bc9cac..374639e661d1 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -1004,6 +1004,14 @@ int ethtool_virtdev_set_link_ksettings(struct net_device *dev,
 				       const struct ethtool_link_ksettings *cmd,
 				       u32 *dev_speed, u8 *dev_duplex);
 
+/**
+ * struct ethtool_netdev_state - per-netdevice state for ethtool features
+ * @wol_enabled:	Wake-on-LAN is enabled
+ */
+struct ethtool_netdev_state {
+	unsigned		wol_enabled:1;
+};
+
 struct phy_device;
 struct phy_tdr_config;
 struct phy_plca_cfg;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 1e3401093c13..3c719f0d5f5a 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -80,6 +80,7 @@ struct xdp_buff;
 struct xdp_frame;
 struct xdp_metadata_ops;
 struct xdp_md;
+struct ethtool_netdev_state;
 
 typedef u32 xdp_features_t;
 
@@ -1986,8 +1987,6 @@ enum netdev_reg_state {
  *			switch driver and used to set the phys state of the
  *			switch port.
  *
- *	@wol_enabled:	Wake-on-LAN is enabled
- *
  *	@threaded:	napi threaded mode is enabled
  *
  *	@module_fw_flash_in_progress:	Module firmware flashing is in progress.
@@ -2001,6 +2000,7 @@ enum netdev_reg_state {
  *	@udp_tunnel_nic_info:	static structure describing the UDP tunnel
  *				offload capabilities of the device
  *	@udp_tunnel_nic:	UDP tunnel offload state
+ *	@ethtool:	ethtool related state
  *	@xdp_state:		stores info on attached XDP BPF programs
  *
  *	@nested_level:	Used as a parameter of spin_lock_nested() of
@@ -2375,7 +2375,7 @@ struct net_device {
 	struct lock_class_key	*qdisc_tx_busylock;
 	bool			proto_down;
 	bool			threaded;
-	unsigned		wol_enabled:1;
+
 	unsigned		module_fw_flash_in_progress:1;
 	struct list_head	net_notifier_list;
 
@@ -2386,6 +2386,8 @@ struct net_device {
 	const struct udp_tunnel_nic_info	*udp_tunnel_nic_info;
 	struct udp_tunnel_nic	*udp_tunnel_nic;
 
+	struct ethtool_netdev_state *ethtool;
+
 	/* protected by rtnl_lock */
 	struct bpf_xdp_entity	xdp_state[__MAX_XDP_MODE];
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 0a23d7da7fbc..c0c6e200e17f 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -11116,6 +11116,9 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
 	dev->real_num_rx_queues = rxqs;
 	if (netif_alloc_rx_queues(dev))
 		goto free_all;
+	dev->ethtool = kzalloc(sizeof(*dev->ethtool), GFP_KERNEL_ACCOUNT);
+	if (!dev->ethtool)
+		goto free_all;
 
 	strcpy(dev->name, name);
 	dev->name_assign_type = name_assign_type;
@@ -11166,6 +11169,7 @@ void free_netdev(struct net_device *dev)
 		return;
 	}
 
+	kfree(dev->ethtool);
 	netif_free_tx_queues(dev);
 	netif_free_rx_queues(dev);
 
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 1cca372c0d80..94059ce9e5f2 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -1509,7 +1509,7 @@ static int ethtool_set_wol(struct net_device *dev, char __user *useraddr)
 	if (ret)
 		return ret;
 
-	dev->wol_enabled = !!wol.wolopts;
+	dev->ethtool->wol_enabled = !!wol.wolopts;
 	ethtool_notify(dev, ETHTOOL_MSG_WOL_NTF, NULL);
 
 	return 0;
diff --git a/net/ethtool/wol.c b/net/ethtool/wol.c
index 0ed56c9ac1bc..a39d8000d808 100644
--- a/net/ethtool/wol.c
+++ b/net/ethtool/wol.c
@@ -137,7 +137,7 @@ ethnl_set_wol(struct ethnl_req_info *req_info, struct genl_info *info)
 	ret = dev->ethtool_ops->set_wol(dev, &wol);
 	if (ret)
 		return ret;
-	dev->wol_enabled = !!wol.wolopts;
+	dev->ethtool->wol_enabled = !!wol.wolopts;
 	return 1;
 }
 
-- 
cgit v1.2.3-58-ga151


From 6ad2962f8adfd53fca52dce7f830783e95d99ce7 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Thu, 27 Jun 2024 16:33:47 +0100
Subject: net: ethtool: attach an XArray of custom RSS contexts to a netdevice

Each context stores the RXFH settings (indir, key, and hfunc) as well
 as optionally some driver private data.
Delete any still-existing contexts at netdev unregister time.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/cbd1c402cec38f2e03124f2ab65b4ae4e08bd90d.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/ethtool.h | 42 ++++++++++++++++++++++++++++++++++++++++++
 net/core/dev.c          | 27 +++++++++++++++++++++++++++
 2 files changed, 69 insertions(+)

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 374639e661d1..c741d6403364 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -159,6 +159,46 @@ static inline u32 ethtool_rxfh_indir_default(u32 index, u32 n_rx_rings)
 	return index % n_rx_rings;
 }
 
+/**
+ * struct ethtool_rxfh_context - a custom RSS context configuration
+ * @indir_size: Number of u32 entries in indirection table
+ * @key_size: Size of hash key, in bytes
+ * @priv_size: Size of driver private data, in bytes
+ * @hfunc: RSS hash function identifier.  One of the %ETH_RSS_HASH_*
+ * @input_xfrm: Defines how the input data is transformed. Valid values are one
+ *	of %RXH_XFRM_*.
+ * @indir_configured: indir has been specified (at create time or subsequently)
+ * @key_configured: hkey has been specified (at create time or subsequently)
+ */
+struct ethtool_rxfh_context {
+	u32 indir_size;
+	u32 key_size;
+	u16 priv_size;
+	u8 hfunc;
+	u8 input_xfrm;
+	u8 indir_configured:1;
+	u8 key_configured:1;
+	/* private: driver private data, indirection table, and hash key are
+	 * stored sequentially in @data area.  Use below helpers to access.
+	 */
+	u8 data[] __aligned(sizeof(void *));
+};
+
+static inline void *ethtool_rxfh_context_priv(struct ethtool_rxfh_context *ctx)
+{
+	return ctx->data;
+}
+
+static inline u32 *ethtool_rxfh_context_indir(struct ethtool_rxfh_context *ctx)
+{
+	return (u32 *)(ctx->data + ALIGN(ctx->priv_size, sizeof(u32)));
+}
+
+static inline u8 *ethtool_rxfh_context_key(struct ethtool_rxfh_context *ctx)
+{
+	return (u8 *)(ethtool_rxfh_context_indir(ctx) + ctx->indir_size);
+}
+
 /* declare a link mode bitmap */
 #define __ETHTOOL_DECLARE_LINK_MODE_MASK(name)		\
 	DECLARE_BITMAP(name, __ETHTOOL_LINK_MODE_MASK_NBITS)
@@ -1006,9 +1046,11 @@ int ethtool_virtdev_set_link_ksettings(struct net_device *dev,
 
 /**
  * struct ethtool_netdev_state - per-netdevice state for ethtool features
+ * @rss_ctx:		XArray of custom RSS contexts
  * @wol_enabled:	Wake-on-LAN is enabled
  */
 struct ethtool_netdev_state {
+	struct xarray		rss_ctx;
 	unsigned		wol_enabled:1;
 };
 
diff --git a/net/core/dev.c b/net/core/dev.c
index c0c6e200e17f..bce810f8ae53 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -10336,6 +10336,9 @@ int register_netdevice(struct net_device *dev)
 	if (ret)
 		return ret;
 
+	/* rss ctx ID 0 is reserved for the default context, start from 1 */
+	xa_init_flags(&dev->ethtool->rss_ctx, XA_FLAGS_ALLOC1);
+
 	spin_lock_init(&dev->addr_list_lock);
 	netdev_set_addr_lockdep_class(dev);
 
@@ -11235,6 +11238,28 @@ void synchronize_net(void)
 }
 EXPORT_SYMBOL(synchronize_net);
 
+static void netdev_rss_contexts_free(struct net_device *dev)
+{
+	struct ethtool_rxfh_context *ctx;
+	unsigned long context;
+
+	xa_for_each(&dev->ethtool->rss_ctx, context, ctx) {
+		struct ethtool_rxfh_param rxfh;
+
+		rxfh.indir = ethtool_rxfh_context_indir(ctx);
+		rxfh.key = ethtool_rxfh_context_key(ctx);
+		rxfh.hfunc = ctx->hfunc;
+		rxfh.input_xfrm = ctx->input_xfrm;
+		rxfh.rss_context = context;
+		rxfh.rss_delete = true;
+
+		xa_erase(&dev->ethtool->rss_ctx, context);
+		dev->ethtool_ops->set_rxfh(dev, &rxfh, NULL);
+		kfree(ctx);
+	}
+	xa_destroy(&dev->ethtool->rss_ctx);
+}
+
 /**
  *	unregister_netdevice_queue - remove device from the kernel
  *	@dev: device
@@ -11338,6 +11363,8 @@ void unregister_netdevice_many_notify(struct list_head *head,
 		netdev_name_node_alt_flush(dev);
 		netdev_name_node_free(dev->name_node);
 
+		netdev_rss_contexts_free(dev);
+
 		call_netdevice_notifiers(NETDEV_PRE_UNINIT, dev);
 
 		if (dev->netdev_ops->ndo_uninit)
-- 
cgit v1.2.3-58-ga151


From eac9122f0c41b832065e01977c34946ec8e76c24 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Thu, 27 Jun 2024 16:33:48 +0100
Subject: net: ethtool: record custom RSS contexts in the XArray

Since drivers are still choosing the context IDs, we have to force the
 XArray to use the ID they've chosen rather than picking one ourselves,
 and handle the case where they give us an ID that's already in use.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/801f5faa4cec87c65b2c6e27fb220c944bce593a.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/ethtool.h | 14 ++++++++++
 net/ethtool/ioctl.c     | 74 ++++++++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 87 insertions(+), 1 deletion(-)

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index c741d6403364..43a2a143034f 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -199,6 +199,17 @@ static inline u8 *ethtool_rxfh_context_key(struct ethtool_rxfh_context *ctx)
 	return (u8 *)(ethtool_rxfh_context_indir(ctx) + ctx->indir_size);
 }
 
+static inline size_t ethtool_rxfh_context_size(u32 indir_size, u32 key_size,
+					       u16 priv_size)
+{
+	size_t indir_bytes = array_size(indir_size, sizeof(u32));
+	size_t flex_len;
+
+	flex_len = size_add(size_add(indir_bytes, key_size),
+			    ALIGN(priv_size, sizeof(u32)));
+	return struct_size_t(struct ethtool_rxfh_context, data, flex_len);
+}
+
 /* declare a link mode bitmap */
 #define __ETHTOOL_DECLARE_LINK_MODE_MASK(name)		\
 	DECLARE_BITMAP(name, __ETHTOOL_LINK_MODE_MASK_NBITS)
@@ -710,6 +721,8 @@ struct ethtool_rxfh_param {
  *	contexts.
  * @cap_rss_sym_xor_supported: indicates if the driver supports symmetric-xor
  *	RSS.
+ * @rxfh_priv_size: size of the driver private data area the core should
+ *	allocate for an RSS context (in &struct ethtool_rxfh_context).
  * @supported_coalesce_params: supported types of interrupt coalescing.
  * @supported_ring_params: supported ring params.
  * @get_drvinfo: Report driver/device information. Modern drivers no
@@ -895,6 +908,7 @@ struct ethtool_ops {
 	u32     cap_link_lanes_supported:1;
 	u32     cap_rss_ctx_supported:1;
 	u32	cap_rss_sym_xor_supported:1;
+	u16	rxfh_priv_size;
 	u32	supported_coalesce_params;
 	u32	supported_ring_params;
 	void	(*get_drvinfo)(struct net_device *, struct ethtool_drvinfo *);
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 94059ce9e5f2..e901a20e97f5 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -1281,10 +1281,12 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 	const struct ethtool_ops *ops = dev->ethtool_ops;
 	u32 dev_indir_size = 0, dev_key_size = 0, i;
 	struct ethtool_rxfh_param rxfh_dev = {};
+	struct ethtool_rxfh_context *ctx = NULL;
 	struct netlink_ext_ack *extack = NULL;
 	struct ethtool_rxnfc rx_rings;
 	struct ethtool_rxfh rxfh;
 	u32 indir_bytes = 0;
+	bool create = false;
 	u8 *rss_config;
 	int ret;
 
@@ -1312,6 +1314,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 	if ((rxfh.input_xfrm & RXH_XFRM_SYM_XOR) &&
 	    !ops->cap_rss_sym_xor_supported)
 		return -EOPNOTSUPP;
+	create = rxfh.rss_context == ETH_RXFH_CONTEXT_ALLOC;
 
 	/* If either indir, hash key or function is valid, proceed further.
 	 * Must request at least one change: indir size, hash key, function
@@ -1377,13 +1380,42 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 		}
 	}
 
+	if (create) {
+		if (rxfh_dev.rss_delete) {
+			ret = -EINVAL;
+			goto out;
+		}
+		ctx = kzalloc(ethtool_rxfh_context_size(dev_indir_size,
+							dev_key_size,
+							ops->rxfh_priv_size),
+			      GFP_KERNEL_ACCOUNT);
+		if (!ctx) {
+			ret = -ENOMEM;
+			goto out;
+		}
+		ctx->indir_size = dev_indir_size;
+		ctx->key_size = dev_key_size;
+		ctx->hfunc = rxfh.hfunc;
+		ctx->input_xfrm = rxfh.input_xfrm;
+		ctx->priv_size = ops->rxfh_priv_size;
+	} else if (rxfh.rss_context) {
+		ctx = xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context);
+		if (!ctx) {
+			ret = -ENOENT;
+			goto out;
+		}
+	}
 	rxfh_dev.hfunc = rxfh.hfunc;
 	rxfh_dev.rss_context = rxfh.rss_context;
 	rxfh_dev.input_xfrm = rxfh.input_xfrm;
 
 	ret = ops->set_rxfh(dev, &rxfh_dev, extack);
-	if (ret)
+	if (ret) {
+		if (create)
+			/* failed to create, free our new tracking entry */
+			kfree(ctx);
 		goto out;
+	}
 
 	if (copy_to_user(useraddr + offsetof(struct ethtool_rxfh, rss_context),
 			 &rxfh_dev.rss_context, sizeof(rxfh_dev.rss_context)))
@@ -1396,6 +1428,46 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 		else if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE)
 			dev->priv_flags |= IFF_RXFH_CONFIGURED;
 	}
+	/* Update rss_ctx tracking */
+	if (create) {
+		/* Ideally this should happen before calling the driver,
+		 * so that we can fail more cleanly; but we don't have the
+		 * context ID until the driver picks it, so we have to
+		 * wait until after.
+		 */
+		if (WARN_ON(xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context))) {
+			/* context ID reused, our tracking is screwed */
+			kfree(ctx);
+			goto out;
+		}
+		/* Allocate the exact ID the driver gave us */
+		if (xa_is_err(xa_store(&dev->ethtool->rss_ctx, rxfh.rss_context,
+				       ctx, GFP_KERNEL))) {
+			kfree(ctx);
+			goto out;
+		}
+		ctx->indir_configured = rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE;
+		ctx->key_configured = !!rxfh.key_size;
+	}
+	if (rxfh_dev.rss_delete) {
+		WARN_ON(xa_erase(&dev->ethtool->rss_ctx, rxfh.rss_context) != ctx);
+		kfree(ctx);
+	} else if (ctx) {
+		if (rxfh_dev.indir) {
+			for (i = 0; i < dev_indir_size; i++)
+				ethtool_rxfh_context_indir(ctx)[i] = rxfh_dev.indir[i];
+			ctx->indir_configured = 1;
+		}
+		if (rxfh_dev.key) {
+			memcpy(ethtool_rxfh_context_key(ctx), rxfh_dev.key,
+			       dev_key_size);
+			ctx->key_configured = 1;
+		}
+		if (rxfh_dev.hfunc != ETH_RSS_HASH_NO_CHANGE)
+			ctx->hfunc = rxfh_dev.hfunc;
+		if (rxfh_dev.input_xfrm != RXH_XFRM_NO_CHANGE)
+			ctx->input_xfrm = rxfh_dev.input_xfrm;
+	}
 
 out:
 	kfree(rss_config);
-- 
cgit v1.2.3-58-ga151


From 847a8ab186767be6ee95643f9739fa9d0f839589 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Thu, 27 Jun 2024 16:33:49 +0100
Subject: net: ethtool: let the core choose RSS context IDs

Add a new API to create/modify/remove RSS contexts, that passes in the
 newly-chosen context ID (not as a pointer) rather than leaving the
 driver to choose it on create.  Also pass in the ctx, allowing drivers
 to easily use its private data area to store their hardware-specific
 state.
Keep the existing .set_rxfh API for now as a fallback, but deprecate it
 for custom contexts (rss_context != 0).

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/45f1fe61df2163c091ec394c9f52000c8b16cc3b.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/ethtool.h | 40 ++++++++++++++++++++++++++++++++++++++++
 net/core/dev.c          |  6 +++++-
 net/ethtool/ioctl.c     | 46 ++++++++++++++++++++++++++++++++++------------
 3 files changed, 79 insertions(+), 13 deletions(-)

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 43a2a143034f..4292a25b2427 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -723,6 +723,10 @@ struct ethtool_rxfh_param {
  *	RSS.
  * @rxfh_priv_size: size of the driver private data area the core should
  *	allocate for an RSS context (in &struct ethtool_rxfh_context).
+ * @rxfh_max_context_id: maximum (exclusive) supported RSS context ID.  If this
+ *	is zero then the core may choose any (nonzero) ID, otherwise the core
+ *	will only use IDs strictly less than this value, as the @rss_context
+ *	argument to @create_rxfh_context and friends.
  * @supported_coalesce_params: supported types of interrupt coalescing.
  * @supported_ring_params: supported ring params.
  * @get_drvinfo: Report driver/device information. Modern drivers no
@@ -819,6 +823,32 @@ struct ethtool_rxfh_param {
  *	will remain unchanged.
  *	Returns a negative error code or zero. An error code must be returned
  *	if at least one unsupported change was requested.
+ * @create_rxfh_context: Create a new RSS context with the specified RX flow
+ *	hash indirection table, hash key, and hash function.
+ *	The &struct ethtool_rxfh_context for this context is passed in @ctx;
+ *	note that the indir table, hkey and hfunc are not yet populated as
+ *	of this call.  The driver does not need to update these; the core
+ *	will do so if this op succeeds.
+ *	However, if @rxfh.indir is set to %NULL, the driver must update the
+ *	indir table in @ctx with the (default or inherited) table actually in
+ *	use; similarly, if @rxfh.key is %NULL, @rxfh.hfunc is
+ *	%ETH_RSS_HASH_NO_CHANGE, or @rxfh.input_xfrm is %RXH_XFRM_NO_CHANGE,
+ *	the driver should update the corresponding information in @ctx.
+ *	If the driver provides this method, it must also provide
+ *	@modify_rxfh_context and @remove_rxfh_context.
+ *	Returns a negative error code or zero.
+ * @modify_rxfh_context: Reconfigure the specified RSS context.  Allows setting
+ *	the contents of the RX flow hash indirection table, hash key, and/or
+ *	hash function associated with the given context.
+ *	Parameters which are set to %NULL or zero will remain unchanged.
+ *	The &struct ethtool_rxfh_context for this context is passed in @ctx;
+ *	note that it will still contain the *old* settings.  The driver does
+ *	not need to update these; the core will do so if this op succeeds.
+ *	Returns a negative error code or zero. An error code must be returned
+ *	if at least one unsupported change was requested.
+ * @remove_rxfh_context: Remove the specified RSS context.
+ *	The &struct ethtool_rxfh_context for this context is passed in @ctx.
+ *	Returns a negative error code or zero.
  * @get_channels: Get number of channels.
  * @set_channels: Set number of channels.  Returns a negative error code or
  *	zero.
@@ -909,6 +939,7 @@ struct ethtool_ops {
 	u32     cap_rss_ctx_supported:1;
 	u32	cap_rss_sym_xor_supported:1;
 	u16	rxfh_priv_size;
+	u32	rxfh_max_context_id;
 	u32	supported_coalesce_params;
 	u32	supported_ring_params;
 	void	(*get_drvinfo)(struct net_device *, struct ethtool_drvinfo *);
@@ -971,6 +1002,15 @@ struct ethtool_ops {
 	int	(*get_rxfh)(struct net_device *, struct ethtool_rxfh_param *);
 	int	(*set_rxfh)(struct net_device *, struct ethtool_rxfh_param *,
 			    struct netlink_ext_ack *extack);
+	int	(*create_rxfh_context)(struct net_device *,
+				       struct ethtool_rxfh_context *ctx,
+				       const struct ethtool_rxfh_param *rxfh);
+	int	(*modify_rxfh_context)(struct net_device *,
+				       struct ethtool_rxfh_context *ctx,
+				       const struct ethtool_rxfh_param *rxfh);
+	int	(*remove_rxfh_context)(struct net_device *,
+				       struct ethtool_rxfh_context *ctx,
+				       u32 rss_context);
 	void	(*get_channels)(struct net_device *, struct ethtool_channels *);
 	int	(*set_channels)(struct net_device *, struct ethtool_channels *);
 	int	(*get_dump_flag)(struct net_device *, struct ethtool_dump *);
diff --git a/net/core/dev.c b/net/core/dev.c
index bce810f8ae53..8481dc4f2196 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -11254,7 +11254,11 @@ static void netdev_rss_contexts_free(struct net_device *dev)
 		rxfh.rss_delete = true;
 
 		xa_erase(&dev->ethtool->rss_ctx, context);
-		dev->ethtool_ops->set_rxfh(dev, &rxfh, NULL);
+		if (dev->ethtool_ops->create_rxfh_context)
+			dev->ethtool_ops->remove_rxfh_context(dev, ctx,
+							      context);
+		else
+			dev->ethtool_ops->set_rxfh(dev, &rxfh, NULL);
 		kfree(ctx);
 	}
 	xa_destroy(&dev->ethtool->rss_ctx);
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index e901a20e97f5..46b50ddde013 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -1395,9 +1395,24 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 		}
 		ctx->indir_size = dev_indir_size;
 		ctx->key_size = dev_key_size;
-		ctx->hfunc = rxfh.hfunc;
-		ctx->input_xfrm = rxfh.input_xfrm;
 		ctx->priv_size = ops->rxfh_priv_size;
+		/* Initialise to an empty context */
+		ctx->hfunc = ETH_RSS_HASH_NO_CHANGE;
+		ctx->input_xfrm = RXH_XFRM_NO_CHANGE;
+		if (ops->create_rxfh_context) {
+			u32 limit = ops->rxfh_max_context_id ?: U32_MAX;
+			u32 ctx_id;
+
+			/* driver uses new API, core allocates ID */
+			ret = xa_alloc(&dev->ethtool->rss_ctx, &ctx_id, ctx,
+				       XA_LIMIT(1, limit), GFP_KERNEL_ACCOUNT);
+			if (ret < 0) {
+				kfree(ctx);
+				goto out;
+			}
+			WARN_ON(!ctx_id); /* can't happen */
+			rxfh.rss_context = ctx_id;
+		}
 	} else if (rxfh.rss_context) {
 		ctx = xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context);
 		if (!ctx) {
@@ -1409,11 +1424,24 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 	rxfh_dev.rss_context = rxfh.rss_context;
 	rxfh_dev.input_xfrm = rxfh.input_xfrm;
 
-	ret = ops->set_rxfh(dev, &rxfh_dev, extack);
-	if (ret) {
+	if (rxfh.rss_context && ops->create_rxfh_context) {
 		if (create)
+			ret = ops->create_rxfh_context(dev, ctx, &rxfh_dev);
+		else if (rxfh_dev.rss_delete)
+			ret = ops->remove_rxfh_context(dev, ctx,
+						       rxfh.rss_context);
+		else
+			ret = ops->modify_rxfh_context(dev, ctx, &rxfh_dev);
+	} else {
+		ret = ops->set_rxfh(dev, &rxfh_dev, extack);
+	}
+	if (ret) {
+		if (create) {
 			/* failed to create, free our new tracking entry */
+			if (ops->create_rxfh_context)
+				xa_erase(&dev->ethtool->rss_ctx, rxfh.rss_context);
 			kfree(ctx);
+		}
 		goto out;
 	}
 
@@ -1429,12 +1457,8 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 			dev->priv_flags |= IFF_RXFH_CONFIGURED;
 	}
 	/* Update rss_ctx tracking */
-	if (create) {
-		/* Ideally this should happen before calling the driver,
-		 * so that we can fail more cleanly; but we don't have the
-		 * context ID until the driver picks it, so we have to
-		 * wait until after.
-		 */
+	if (create && !ops->create_rxfh_context) {
+		/* driver uses old API, it chose context ID */
 		if (WARN_ON(xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context))) {
 			/* context ID reused, our tracking is screwed */
 			kfree(ctx);
@@ -1446,8 +1470,6 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 			kfree(ctx);
 			goto out;
 		}
-		ctx->indir_configured = rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE;
-		ctx->key_configured = !!rxfh.key_size;
 	}
 	if (rxfh_dev.rss_delete) {
 		WARN_ON(xa_erase(&dev->ethtool->rss_ctx, rxfh.rss_context) != ctx);
-- 
cgit v1.2.3-58-ga151


From 30a32cdf6b130356805b3193a6208de25cbb2015 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Thu, 27 Jun 2024 16:33:50 +0100
Subject: net: ethtool: add an extack parameter to new rxfh_context APIs

Currently passed as NULL, but will allow drivers to report back errors
 when ethnl support for these ops is added.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/6e0012347d175fdd1280363d7bfa76a2f2777e17.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/ethtool.h | 9 ++++++---
 net/core/dev.c          | 2 +-
 net/ethtool/ioctl.c     | 9 ++++++---
 3 files changed, 13 insertions(+), 7 deletions(-)

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 4292a25b2427..9cdbc8e3ed5c 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -1004,13 +1004,16 @@ struct ethtool_ops {
 			    struct netlink_ext_ack *extack);
 	int	(*create_rxfh_context)(struct net_device *,
 				       struct ethtool_rxfh_context *ctx,
-				       const struct ethtool_rxfh_param *rxfh);
+				       const struct ethtool_rxfh_param *rxfh,
+				       struct netlink_ext_ack *extack);
 	int	(*modify_rxfh_context)(struct net_device *,
 				       struct ethtool_rxfh_context *ctx,
-				       const struct ethtool_rxfh_param *rxfh);
+				       const struct ethtool_rxfh_param *rxfh,
+				       struct netlink_ext_ack *extack);
 	int	(*remove_rxfh_context)(struct net_device *,
 				       struct ethtool_rxfh_context *ctx,
-				       u32 rss_context);
+				       u32 rss_context,
+				       struct netlink_ext_ack *extack);
 	void	(*get_channels)(struct net_device *, struct ethtool_channels *);
 	int	(*set_channels)(struct net_device *, struct ethtool_channels *);
 	int	(*get_dump_flag)(struct net_device *, struct ethtool_dump *);
diff --git a/net/core/dev.c b/net/core/dev.c
index 8481dc4f2196..2daed4464c08 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -11256,7 +11256,7 @@ static void netdev_rss_contexts_free(struct net_device *dev)
 		xa_erase(&dev->ethtool->rss_ctx, context);
 		if (dev->ethtool_ops->create_rxfh_context)
 			dev->ethtool_ops->remove_rxfh_context(dev, ctx,
-							      context);
+							      context, NULL);
 		else
 			dev->ethtool_ops->set_rxfh(dev, &rxfh, NULL);
 		kfree(ctx);
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 46b50ddde013..82c610e9e6b2 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -1426,12 +1426,15 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 
 	if (rxfh.rss_context && ops->create_rxfh_context) {
 		if (create)
-			ret = ops->create_rxfh_context(dev, ctx, &rxfh_dev);
+			ret = ops->create_rxfh_context(dev, ctx, &rxfh_dev,
+						       extack);
 		else if (rxfh_dev.rss_delete)
 			ret = ops->remove_rxfh_context(dev, ctx,
-						       rxfh.rss_context);
+						       rxfh.rss_context,
+						       extack);
 		else
-			ret = ops->modify_rxfh_context(dev, ctx, &rxfh_dev);
+			ret = ops->modify_rxfh_context(dev, ctx, &rxfh_dev,
+						       extack);
 	} else {
 		ret = ops->set_rxfh(dev, &rxfh_dev, extack);
 	}
-- 
cgit v1.2.3-58-ga151


From 87925151191b64d9623e63ccf11e517eacc99d7d Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Thu, 27 Jun 2024 16:33:51 +0100
Subject: net: ethtool: add a mutex protecting RSS contexts

While this is not needed to serialise the ethtool entry points (which
 are all under RTNL), drivers may have cause to asynchronously access
 dev->ethtool->rss_ctx; taking dev->ethtool->rss_lock allows them to
 do this safely without needing to take the RTNL.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/7f9c15eb7525bf87af62c275dde3a8570ee8bf0a.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/ethtool.h | 3 +++
 net/core/dev.c          | 5 +++++
 net/ethtool/ioctl.c     | 7 +++++++
 3 files changed, 15 insertions(+)

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 9cdbc8e3ed5c..f74bb0cf8ed1 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -1104,10 +1104,13 @@ int ethtool_virtdev_set_link_ksettings(struct net_device *dev,
 /**
  * struct ethtool_netdev_state - per-netdevice state for ethtool features
  * @rss_ctx:		XArray of custom RSS contexts
+ * @rss_lock:		Protects entries in @rss_ctx.  May be taken from
+ *			within RTNL.
  * @wol_enabled:	Wake-on-LAN is enabled
  */
 struct ethtool_netdev_state {
 	struct xarray		rss_ctx;
+	struct mutex		rss_lock;
 	unsigned		wol_enabled:1;
 };
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 2daed4464c08..385c4091aa77 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -10338,6 +10338,7 @@ int register_netdevice(struct net_device *dev)
 
 	/* rss ctx ID 0 is reserved for the default context, start from 1 */
 	xa_init_flags(&dev->ethtool->rss_ctx, XA_FLAGS_ALLOC1);
+	mutex_init(&dev->ethtool->rss_lock);
 
 	spin_lock_init(&dev->addr_list_lock);
 	netdev_set_addr_lockdep_class(dev);
@@ -11243,6 +11244,7 @@ static void netdev_rss_contexts_free(struct net_device *dev)
 	struct ethtool_rxfh_context *ctx;
 	unsigned long context;
 
+	mutex_lock(&dev->ethtool->rss_lock);
 	xa_for_each(&dev->ethtool->rss_ctx, context, ctx) {
 		struct ethtool_rxfh_param rxfh;
 
@@ -11262,6 +11264,7 @@ static void netdev_rss_contexts_free(struct net_device *dev)
 		kfree(ctx);
 	}
 	xa_destroy(&dev->ethtool->rss_ctx);
+	mutex_unlock(&dev->ethtool->rss_lock);
 }
 
 /**
@@ -11374,6 +11377,8 @@ void unregister_netdevice_many_notify(struct list_head *head,
 		if (dev->netdev_ops->ndo_uninit)
 			dev->netdev_ops->ndo_uninit(dev);
 
+		mutex_destroy(&dev->ethtool->rss_lock);
+
 		if (skb)
 			rtmsg_ifinfo_send(skb, dev, GFP_KERNEL, portid, nlh);
 
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 82c610e9e6b2..939ccd106fe1 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -1285,6 +1285,7 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 	struct netlink_ext_ack *extack = NULL;
 	struct ethtool_rxnfc rx_rings;
 	struct ethtool_rxfh rxfh;
+	bool locked = false; /* dev->ethtool->rss_lock taken */
 	u32 indir_bytes = 0;
 	bool create = false;
 	u8 *rss_config;
@@ -1380,6 +1381,10 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 		}
 	}
 
+	if (rxfh.rss_context) {
+		mutex_lock(&dev->ethtool->rss_lock);
+		locked = true;
+	}
 	if (create) {
 		if (rxfh_dev.rss_delete) {
 			ret = -EINVAL;
@@ -1495,6 +1500,8 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 	}
 
 out:
+	if (locked)
+		mutex_unlock(&dev->ethtool->rss_lock);
 	kfree(rss_config);
 	return ret;
 }
-- 
cgit v1.2.3-58-ga151


From a9ee8d4a97d8ea0004615fe1f2c1ce6aff5c1a14 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Thu, 27 Jun 2024 16:33:52 +0100
Subject: sfc: use new rxfh_context API

The core is now responsible for allocating IDs and a memory region for
 us to store our state (struct efx_rss_context_priv), so we no longer
 need efx_alloc_rss_context_entry() and friends.
Since the contexts are now maintained by the core, use the core's lock
 (net_dev->ethtool->rss_lock), rather than our own mutex (efx->rss_lock),
 to serialise access against changes; and remove the now-unused
 efx->rss_lock from struct efx_nic.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/150274740ea8cc137fef5502541ce573d32fb319.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/ef10.c           |   2 +-
 drivers/net/ethernet/sfc/ef100_ethtool.c  |   4 +
 drivers/net/ethernet/sfc/efx.c            |   2 +-
 drivers/net/ethernet/sfc/efx.h            |   2 +-
 drivers/net/ethernet/sfc/efx_common.c     |  10 +-
 drivers/net/ethernet/sfc/ethtool.c        |   4 +
 drivers/net/ethernet/sfc/ethtool_common.c | 152 +++++++++++++++++-------------
 drivers/net/ethernet/sfc/ethtool_common.h |  12 +++
 drivers/net/ethernet/sfc/mcdi_filters.c   | 135 +++++++++++++-------------
 drivers/net/ethernet/sfc/mcdi_filters.h   |   8 +-
 drivers/net/ethernet/sfc/net_driver.h     |  28 +++---
 drivers/net/ethernet/sfc/rx_common.c      |  64 ++-----------
 drivers/net/ethernet/sfc/rx_common.h      |   8 +-
 13 files changed, 212 insertions(+), 219 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index 8fa6c0e9195b..7d69302ffa0a 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -1396,7 +1396,7 @@ static void efx_ef10_table_reset_mc_allocations(struct efx_nic *efx)
 	efx_mcdi_filter_table_reset_mc_allocations(efx);
 	nic_data->must_restore_piobufs = true;
 	efx_ef10_forget_old_piobufs(efx);
-	efx->rss_context.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
+	efx->rss_context.priv.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 
 	/* Driver-created vswitches and vports must be re-created */
 	nic_data->must_probe_vswitching = true;
diff --git a/drivers/net/ethernet/sfc/ef100_ethtool.c b/drivers/net/ethernet/sfc/ef100_ethtool.c
index cf55202b3a7b..896ffca4aee2 100644
--- a/drivers/net/ethernet/sfc/ef100_ethtool.c
+++ b/drivers/net/ethernet/sfc/ef100_ethtool.c
@@ -59,8 +59,12 @@ const struct ethtool_ops ef100_ethtool_ops = {
 
 	.get_rxfh_indir_size	= efx_ethtool_get_rxfh_indir_size,
 	.get_rxfh_key_size	= efx_ethtool_get_rxfh_key_size,
+	.rxfh_priv_size		= sizeof(struct efx_rss_context_priv),
 	.get_rxfh		= efx_ethtool_get_rxfh,
 	.set_rxfh		= efx_ethtool_set_rxfh,
+	.create_rxfh_context	= efx_ethtool_create_rxfh_context,
+	.modify_rxfh_context	= efx_ethtool_modify_rxfh_context,
+	.remove_rxfh_context	= efx_ethtool_remove_rxfh_context,
 
 	.get_module_info	= efx_ethtool_get_module_info,
 	.get_module_eeprom	= efx_ethtool_get_module_eeprom,
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index e9d9de8e648a..6f1a01ded7d4 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -299,7 +299,7 @@ static int efx_probe_nic(struct efx_nic *efx)
 	if (efx->n_channels > 1)
 		netdev_rss_key_fill(efx->rss_context.rx_hash_key,
 				    sizeof(efx->rss_context.rx_hash_key));
-	efx_set_default_rx_indir_table(efx, &efx->rss_context);
+	efx_set_default_rx_indir_table(efx, efx->rss_context.rx_indir_table);
 
 	/* Initialise the interrupt moderation settings */
 	efx->irq_mod_step_us = DIV_ROUND_UP(efx->timer_quantum_ns, 1000);
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index 48d3623735ba..7a6cab883d66 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -158,7 +158,7 @@ static inline s32 efx_filter_get_rx_ids(struct efx_nic *efx,
 }
 
 /* RSS contexts */
-static inline bool efx_rss_active(struct efx_rss_context *ctx)
+static inline bool efx_rss_active(struct efx_rss_context_priv *ctx)
 {
 	return ctx->context_id != EFX_MCDI_RSS_CONTEXT_INVALID;
 }
diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c
index 4ebd5ae23eca..13cf647051af 100644
--- a/drivers/net/ethernet/sfc/efx_common.c
+++ b/drivers/net/ethernet/sfc/efx_common.c
@@ -714,7 +714,7 @@ void efx_reset_down(struct efx_nic *efx, enum reset_type method)
 
 	mutex_lock(&efx->mac_lock);
 	down_write(&efx->filter_sem);
-	mutex_lock(&efx->rss_lock);
+	mutex_lock(&efx->net_dev->ethtool->rss_lock);
 	efx->type->fini(efx);
 }
 
@@ -777,7 +777,7 @@ int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok)
 
 	if (efx->type->rx_restore_rss_contexts)
 		efx->type->rx_restore_rss_contexts(efx);
-	mutex_unlock(&efx->rss_lock);
+	mutex_unlock(&efx->net_dev->ethtool->rss_lock);
 	efx->type->filter_table_restore(efx);
 	up_write(&efx->filter_sem);
 
@@ -793,7 +793,7 @@ int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok)
 fail:
 	efx->port_initialized = false;
 
-	mutex_unlock(&efx->rss_lock);
+	mutex_unlock(&efx->net_dev->ethtool->rss_lock);
 	up_write(&efx->filter_sem);
 	mutex_unlock(&efx->mac_lock);
 
@@ -1000,9 +1000,7 @@ int efx_init_struct(struct efx_nic *efx, struct pci_dev *pci_dev)
 		efx->type->rx_hash_offset - efx->type->rx_prefix_size;
 	efx->rx_packet_ts_offset =
 		efx->type->rx_ts_offset - efx->type->rx_prefix_size;
-	INIT_LIST_HEAD(&efx->rss_context.list);
-	efx->rss_context.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
-	mutex_init(&efx->rss_lock);
+	efx->rss_context.priv.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 	efx->vport_id = EVB_PORT_ID_ASSIGNED;
 	spin_lock_init(&efx->stats_lock);
 	efx->vi_stride = EFX_DEFAULT_VI_STRIDE;
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index 37c69c8d90b1..0f5c68b8bab7 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -268,8 +268,12 @@ const struct ethtool_ops efx_ethtool_ops = {
 	.set_rxnfc		= efx_ethtool_set_rxnfc,
 	.get_rxfh_indir_size	= efx_ethtool_get_rxfh_indir_size,
 	.get_rxfh_key_size	= efx_ethtool_get_rxfh_key_size,
+	.rxfh_priv_size		= sizeof(struct efx_rss_context_priv),
 	.get_rxfh		= efx_ethtool_get_rxfh,
 	.set_rxfh		= efx_ethtool_set_rxfh,
+	.create_rxfh_context	= efx_ethtool_create_rxfh_context,
+	.modify_rxfh_context	= efx_ethtool_modify_rxfh_context,
+	.remove_rxfh_context	= efx_ethtool_remove_rxfh_context,
 	.get_ts_info		= efx_ethtool_get_ts_info,
 	.get_module_info	= efx_ethtool_get_module_info,
 	.get_module_eeprom	= efx_ethtool_get_module_eeprom,
diff --git a/drivers/net/ethernet/sfc/ethtool_common.c b/drivers/net/ethernet/sfc/ethtool_common.c
index 7d5e5db4eac5..0a8d2c9ffce6 100644
--- a/drivers/net/ethernet/sfc/ethtool_common.c
+++ b/drivers/net/ethernet/sfc/ethtool_common.c
@@ -820,10 +820,10 @@ int efx_ethtool_get_rxnfc(struct net_device *net_dev,
 		return 0;
 
 	case ETHTOOL_GRXFH: {
-		struct efx_rss_context *ctx = &efx->rss_context;
+		struct efx_rss_context_priv *ctx = &efx->rss_context.priv;
 		__u64 data;
 
-		mutex_lock(&efx->rss_lock);
+		mutex_lock(&net_dev->ethtool->rss_lock);
 		if (info->flow_type & FLOW_RSS && info->rss_context) {
 			ctx = efx_find_rss_context_entry(efx, info->rss_context);
 			if (!ctx) {
@@ -864,7 +864,7 @@ int efx_ethtool_get_rxnfc(struct net_device *net_dev,
 out_setdata_unlock:
 		info->data = data;
 out_unlock:
-		mutex_unlock(&efx->rss_lock);
+		mutex_unlock(&net_dev->ethtool->rss_lock);
 		return rc;
 	}
 
@@ -1167,31 +1167,33 @@ static int efx_ethtool_get_rxfh_context(struct net_device *net_dev,
 					struct ethtool_rxfh_param *rxfh)
 {
 	struct efx_nic *efx = efx_netdev_priv(net_dev);
-	struct efx_rss_context *ctx;
+	struct efx_rss_context_priv *ctx_priv;
+	struct efx_rss_context ctx;
 	int rc = 0;
 
 	if (!efx->type->rx_pull_rss_context_config)
 		return -EOPNOTSUPP;
 
-	mutex_lock(&efx->rss_lock);
-	ctx = efx_find_rss_context_entry(efx, rxfh->rss_context);
-	if (!ctx) {
+	mutex_lock(&net_dev->ethtool->rss_lock);
+	ctx_priv = efx_find_rss_context_entry(efx, rxfh->rss_context);
+	if (!ctx_priv) {
 		rc = -ENOENT;
 		goto out_unlock;
 	}
-	rc = efx->type->rx_pull_rss_context_config(efx, ctx);
+	ctx.priv = *ctx_priv;
+	rc = efx->type->rx_pull_rss_context_config(efx, &ctx);
 	if (rc)
 		goto out_unlock;
 
 	rxfh->hfunc = ETH_RSS_HASH_TOP;
 	if (rxfh->indir)
-		memcpy(rxfh->indir, ctx->rx_indir_table,
-		       sizeof(ctx->rx_indir_table));
+		memcpy(rxfh->indir, ctx.rx_indir_table,
+		       sizeof(ctx.rx_indir_table));
 	if (rxfh->key)
-		memcpy(rxfh->key, ctx->rx_hash_key,
+		memcpy(rxfh->key, ctx.rx_hash_key,
 		       efx->type->rx_hash_key_size);
 out_unlock:
-	mutex_unlock(&efx->rss_lock);
+	mutex_unlock(&net_dev->ethtool->rss_lock);
 	return rc;
 }
 
@@ -1218,68 +1220,85 @@ int efx_ethtool_get_rxfh(struct net_device *net_dev,
 	return 0;
 }
 
-static int efx_ethtool_set_rxfh_context(struct net_device *net_dev,
-					struct ethtool_rxfh_param *rxfh,
-					struct netlink_ext_ack *extack)
+int efx_ethtool_modify_rxfh_context(struct net_device *net_dev,
+				    struct ethtool_rxfh_context *ctx,
+				    const struct ethtool_rxfh_param *rxfh,
+				    struct netlink_ext_ack *extack)
 {
 	struct efx_nic *efx = efx_netdev_priv(net_dev);
-	u32 *rss_context = &rxfh->rss_context;
-	struct efx_rss_context *ctx;
-	u32 *indir = rxfh->indir;
-	bool allocated = false;
-	u8 *key = rxfh->key;
-	int rc;
+	struct efx_rss_context_priv *priv;
+	const u32 *indir = rxfh->indir;
+	const u8 *key = rxfh->key;
 
-	if (!efx->type->rx_push_rss_context_config)
+	if (!efx->type->rx_push_rss_context_config) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "NIC type does not support custom contexts");
 		return -EOPNOTSUPP;
-
-	mutex_lock(&efx->rss_lock);
-
-	if (*rss_context == ETH_RXFH_CONTEXT_ALLOC) {
-		if (rxfh->rss_delete) {
-			/* alloc + delete == Nothing to do */
-			rc = -EINVAL;
-			goto out_unlock;
-		}
-		ctx = efx_alloc_rss_context_entry(efx);
-		if (!ctx) {
-			rc = -ENOMEM;
-			goto out_unlock;
-		}
-		ctx->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
-		/* Initialise indir table and key to defaults */
-		efx_set_default_rx_indir_table(efx, ctx);
-		netdev_rss_key_fill(ctx->rx_hash_key, sizeof(ctx->rx_hash_key));
-		allocated = true;
-	} else {
-		ctx = efx_find_rss_context_entry(efx, *rss_context);
-		if (!ctx) {
-			rc = -ENOENT;
-			goto out_unlock;
-		}
 	}
-
-	if (rxfh->rss_delete) {
-		/* delete this context */
-		rc = efx->type->rx_push_rss_context_config(efx, ctx, NULL, NULL);
-		if (!rc)
-			efx_free_rss_context_entry(ctx);
-		goto out_unlock;
+	/* Hash function is Toeplitz, cannot be changed */
+	if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	    rxfh->hfunc != ETH_RSS_HASH_TOP) {
+		NL_SET_ERR_MSG_MOD(extack, "Only Toeplitz hash is supported");
+		return -EOPNOTSUPP;
 	}
 
+	priv = ethtool_rxfh_context_priv(ctx);
+
 	if (!key)
-		key = ctx->rx_hash_key;
+		key = ethtool_rxfh_context_key(ctx);
 	if (!indir)
-		indir = ctx->rx_indir_table;
+		indir = ethtool_rxfh_context_indir(ctx);
 
-	rc = efx->type->rx_push_rss_context_config(efx, ctx, indir, key);
-	if (rc && allocated)
-		efx_free_rss_context_entry(ctx);
-	else
-		*rss_context = ctx->user_id;
-out_unlock:
-	mutex_unlock(&efx->rss_lock);
-	return rc;
+	return efx->type->rx_push_rss_context_config(efx, priv, indir, key,
+						     false);
+}
+
+int efx_ethtool_create_rxfh_context(struct net_device *net_dev,
+				    struct ethtool_rxfh_context *ctx,
+				    const struct ethtool_rxfh_param *rxfh,
+				    struct netlink_ext_ack *extack)
+{
+	struct efx_nic *efx = efx_netdev_priv(net_dev);
+	struct efx_rss_context_priv *priv;
+
+	priv = ethtool_rxfh_context_priv(ctx);
+
+	priv->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
+	priv->rx_hash_udp_4tuple = false;
+	/* Generate default indir table and/or key if not specified.
+	 * We use ctx as a place to store these; this is fine because
+	 * we're doing a create, so if we fail then the ctx will just
+	 * be deleted.
+	 */
+	if (!rxfh->indir)
+		efx_set_default_rx_indir_table(efx, ethtool_rxfh_context_indir(ctx));
+	if (!rxfh->key)
+		netdev_rss_key_fill(ethtool_rxfh_context_key(ctx),
+				    ctx->key_size);
+	if (rxfh->hfunc == ETH_RSS_HASH_NO_CHANGE)
+		ctx->hfunc = ETH_RSS_HASH_TOP;
+	if (rxfh->input_xfrm == RXH_XFRM_NO_CHANGE)
+		ctx->input_xfrm = 0;
+	return efx_ethtool_modify_rxfh_context(net_dev, ctx, rxfh, extack);
+}
+
+int efx_ethtool_remove_rxfh_context(struct net_device *net_dev,
+				    struct ethtool_rxfh_context *ctx,
+				    u32 rss_context,
+				    struct netlink_ext_ack *extack)
+{
+	struct efx_nic *efx = efx_netdev_priv(net_dev);
+	struct efx_rss_context_priv *priv;
+
+	if (!efx->type->rx_push_rss_context_config) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "NIC type does not support custom contexts");
+		return -EOPNOTSUPP;
+	}
+
+	priv = ethtool_rxfh_context_priv(ctx);
+	return efx->type->rx_push_rss_context_config(efx, priv, NULL, NULL,
+						     true);
 }
 
 int efx_ethtool_set_rxfh(struct net_device *net_dev,
@@ -1295,8 +1314,9 @@ int efx_ethtool_set_rxfh(struct net_device *net_dev,
 	    rxfh->hfunc != ETH_RSS_HASH_TOP)
 		return -EOPNOTSUPP;
 
-	if (rxfh->rss_context)
-		return efx_ethtool_set_rxfh_context(net_dev, rxfh, extack);
+	/* Custom contexts should use new API */
+	if (WARN_ON_ONCE(rxfh->rss_context))
+		return -EIO;
 
 	if (!indir && !key)
 		return 0;
diff --git a/drivers/net/ethernet/sfc/ethtool_common.h b/drivers/net/ethernet/sfc/ethtool_common.h
index a680e5980213..fc52e891637d 100644
--- a/drivers/net/ethernet/sfc/ethtool_common.h
+++ b/drivers/net/ethernet/sfc/ethtool_common.h
@@ -49,6 +49,18 @@ int efx_ethtool_get_rxfh(struct net_device *net_dev,
 int efx_ethtool_set_rxfh(struct net_device *net_dev,
 			 struct ethtool_rxfh_param *rxfh,
 			 struct netlink_ext_ack *extack);
+int efx_ethtool_create_rxfh_context(struct net_device *net_dev,
+				    struct ethtool_rxfh_context *ctx,
+				    const struct ethtool_rxfh_param *rxfh,
+				    struct netlink_ext_ack *extack);
+int efx_ethtool_modify_rxfh_context(struct net_device *net_dev,
+				    struct ethtool_rxfh_context *ctx,
+				    const struct ethtool_rxfh_param *rxfh,
+				    struct netlink_ext_ack *extack);
+int efx_ethtool_remove_rxfh_context(struct net_device *net_dev,
+				    struct ethtool_rxfh_context *ctx,
+				    u32 rss_context,
+				    struct netlink_ext_ack *extack);
 int efx_ethtool_reset(struct net_device *net_dev, u32 *flags);
 int efx_ethtool_get_module_eeprom(struct net_device *net_dev,
 				  struct ethtool_eeprom *ee,
diff --git a/drivers/net/ethernet/sfc/mcdi_filters.c b/drivers/net/ethernet/sfc/mcdi_filters.c
index 4ff6586116ee..6ef96292909a 100644
--- a/drivers/net/ethernet/sfc/mcdi_filters.c
+++ b/drivers/net/ethernet/sfc/mcdi_filters.c
@@ -194,7 +194,7 @@ efx_mcdi_filter_push_prep_set_match_fields(struct efx_nic *efx,
 static void efx_mcdi_filter_push_prep(struct efx_nic *efx,
 				      const struct efx_filter_spec *spec,
 				      efx_dword_t *inbuf, u64 handle,
-				      struct efx_rss_context *ctx,
+				      struct efx_rss_context_priv *ctx,
 				      bool replacing)
 {
 	u32 flags = spec->flags;
@@ -245,7 +245,7 @@ static void efx_mcdi_filter_push_prep(struct efx_nic *efx,
 
 static int efx_mcdi_filter_push(struct efx_nic *efx,
 				const struct efx_filter_spec *spec, u64 *handle,
-				struct efx_rss_context *ctx, bool replacing)
+				struct efx_rss_context_priv *ctx, bool replacing)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_FILTER_OP_EXT_IN_LEN);
 	MCDI_DECLARE_BUF(outbuf, MC_CMD_FILTER_OP_EXT_OUT_LEN);
@@ -345,9 +345,9 @@ static s32 efx_mcdi_filter_insert_locked(struct efx_nic *efx,
 					 bool replace_equal)
 {
 	DECLARE_BITMAP(mc_rem_map, EFX_EF10_FILTER_SEARCH_LIMIT);
+	struct efx_rss_context_priv *ctx = NULL;
 	struct efx_mcdi_filter_table *table;
 	struct efx_filter_spec *saved_spec;
-	struct efx_rss_context *ctx = NULL;
 	unsigned int match_pri, hash;
 	unsigned int priv_flags;
 	bool rss_locked = false;
@@ -380,12 +380,12 @@ static s32 efx_mcdi_filter_insert_locked(struct efx_nic *efx,
 		bitmap_zero(mc_rem_map, EFX_EF10_FILTER_SEARCH_LIMIT);
 
 	if (spec->flags & EFX_FILTER_FLAG_RX_RSS) {
-		mutex_lock(&efx->rss_lock);
+		mutex_lock(&efx->net_dev->ethtool->rss_lock);
 		rss_locked = true;
 		if (spec->rss_context)
 			ctx = efx_find_rss_context_entry(efx, spec->rss_context);
 		else
-			ctx = &efx->rss_context;
+			ctx = &efx->rss_context.priv;
 		if (!ctx) {
 			rc = -ENOENT;
 			goto out_unlock;
@@ -548,7 +548,7 @@ static s32 efx_mcdi_filter_insert_locked(struct efx_nic *efx,
 
 out_unlock:
 	if (rss_locked)
-		mutex_unlock(&efx->rss_lock);
+		mutex_unlock(&efx->net_dev->ethtool->rss_lock);
 	up_write(&table->lock);
 	return rc;
 }
@@ -611,13 +611,13 @@ static int efx_mcdi_filter_remove_internal(struct efx_nic *efx,
 
 		new_spec.priority = EFX_FILTER_PRI_AUTO;
 		new_spec.flags = (EFX_FILTER_FLAG_RX |
-				  (efx_rss_active(&efx->rss_context) ?
+				  (efx_rss_active(&efx->rss_context.priv) ?
 				   EFX_FILTER_FLAG_RX_RSS : 0));
 		new_spec.dmaq_id = 0;
 		new_spec.rss_context = 0;
 		rc = efx_mcdi_filter_push(efx, &new_spec,
 					  &table->entry[filter_idx].handle,
-					  &efx->rss_context,
+					  &efx->rss_context.priv,
 					  true);
 
 		if (rc == 0)
@@ -764,7 +764,7 @@ static int efx_mcdi_filter_insert_addr_list(struct efx_nic *efx,
 		ids = vlan->uc;
 	}
 
-	filter_flags = efx_rss_active(&efx->rss_context) ? EFX_FILTER_FLAG_RX_RSS : 0;
+	filter_flags = efx_rss_active(&efx->rss_context.priv) ? EFX_FILTER_FLAG_RX_RSS : 0;
 
 	/* Insert/renew filters */
 	for (i = 0; i < addr_count; i++) {
@@ -833,7 +833,7 @@ static int efx_mcdi_filter_insert_def(struct efx_nic *efx,
 	int rc;
 	u16 *id;
 
-	filter_flags = efx_rss_active(&efx->rss_context) ? EFX_FILTER_FLAG_RX_RSS : 0;
+	filter_flags = efx_rss_active(&efx->rss_context.priv) ? EFX_FILTER_FLAG_RX_RSS : 0;
 
 	efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0);
 
@@ -1375,8 +1375,8 @@ void efx_mcdi_filter_table_restore(struct efx_nic *efx)
 	struct efx_mcdi_filter_table *table = efx->filter_state;
 	unsigned int invalid_filters = 0, failed = 0;
 	struct efx_mcdi_filter_vlan *vlan;
+	struct efx_rss_context_priv *ctx;
 	struct efx_filter_spec *spec;
-	struct efx_rss_context *ctx;
 	unsigned int filter_idx;
 	u32 mcdi_flags;
 	int match_pri;
@@ -1388,7 +1388,7 @@ void efx_mcdi_filter_table_restore(struct efx_nic *efx)
 		return;
 
 	down_write(&table->lock);
-	mutex_lock(&efx->rss_lock);
+	mutex_lock(&efx->net_dev->ethtool->rss_lock);
 
 	for (filter_idx = 0; filter_idx < EFX_MCDI_FILTER_TBL_ROWS; filter_idx++) {
 		spec = efx_mcdi_filter_entry_spec(table, filter_idx);
@@ -1407,7 +1407,7 @@ void efx_mcdi_filter_table_restore(struct efx_nic *efx)
 		if (spec->rss_context)
 			ctx = efx_find_rss_context_entry(efx, spec->rss_context);
 		else
-			ctx = &efx->rss_context;
+			ctx = &efx->rss_context.priv;
 		if (spec->flags & EFX_FILTER_FLAG_RX_RSS) {
 			if (!ctx) {
 				netif_warn(efx, drv, efx->net_dev,
@@ -1444,7 +1444,7 @@ not_restored:
 		}
 	}
 
-	mutex_unlock(&efx->rss_lock);
+	mutex_unlock(&efx->net_dev->ethtool->rss_lock);
 	up_write(&table->lock);
 
 	/*
@@ -1861,7 +1861,8 @@ out_unlock:
 					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV6_RSS_MODE_LBN |\
 					 RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_OTHER_IPV6_RSS_MODE_LBN)
 
-int efx_mcdi_get_rss_context_flags(struct efx_nic *efx, u32 context, u32 *flags)
+static int efx_mcdi_get_rss_context_flags(struct efx_nic *efx, u32 context,
+					  u32 *flags)
 {
 	/*
 	 * Firmware had a bug (sfc bug 61952) where it would not actually
@@ -1909,8 +1910,8 @@ int efx_mcdi_get_rss_context_flags(struct efx_nic *efx, u32 context, u32 *flags)
  * Defaults are 4-tuple for TCP and 2-tuple for UDP and other-IP, so we
  * just need to set the UDP ports flags (for both IP versions).
  */
-void efx_mcdi_set_rss_context_flags(struct efx_nic *efx,
-				    struct efx_rss_context *ctx)
+static void efx_mcdi_set_rss_context_flags(struct efx_nic *efx,
+					   struct efx_rss_context_priv *ctx)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_LEN);
 	u32 flags;
@@ -1931,7 +1932,7 @@ void efx_mcdi_set_rss_context_flags(struct efx_nic *efx,
 }
 
 static int efx_mcdi_filter_alloc_rss_context(struct efx_nic *efx, bool exclusive,
-					     struct efx_rss_context *ctx,
+					     struct efx_rss_context_priv *ctx,
 					     unsigned *context_size)
 {
 	MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_ALLOC_IN_LEN);
@@ -2032,25 +2033,26 @@ void efx_mcdi_rx_free_indir_table(struct efx_nic *efx)
 {
 	int rc;
 
-	if (efx->rss_context.context_id != EFX_MCDI_RSS_CONTEXT_INVALID) {
-		rc = efx_mcdi_filter_free_rss_context(efx, efx->rss_context.context_id);
+	if (efx->rss_context.priv.context_id != EFX_MCDI_RSS_CONTEXT_INVALID) {
+		rc = efx_mcdi_filter_free_rss_context(efx, efx->rss_context.priv.context_id);
 		WARN_ON(rc != 0);
 	}
-	efx->rss_context.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
+	efx->rss_context.priv.context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 }
 
 static int efx_mcdi_filter_rx_push_shared_rss_config(struct efx_nic *efx,
 					      unsigned *context_size)
 {
 	struct efx_mcdi_filter_table *table = efx->filter_state;
-	int rc = efx_mcdi_filter_alloc_rss_context(efx, false, &efx->rss_context,
-					    context_size);
+	int rc = efx_mcdi_filter_alloc_rss_context(efx, false,
+						   &efx->rss_context.priv,
+						   context_size);
 
 	if (rc != 0)
 		return rc;
 
 	table->rx_rss_context_exclusive = false;
-	efx_set_default_rx_indir_table(efx, &efx->rss_context);
+	efx_set_default_rx_indir_table(efx, efx->rss_context.rx_indir_table);
 	return 0;
 }
 
@@ -2058,26 +2060,27 @@ static int efx_mcdi_filter_rx_push_exclusive_rss_config(struct efx_nic *efx,
 						 const u32 *rx_indir_table,
 						 const u8 *key)
 {
+	u32 old_rx_rss_context = efx->rss_context.priv.context_id;
 	struct efx_mcdi_filter_table *table = efx->filter_state;
-	u32 old_rx_rss_context = efx->rss_context.context_id;
 	int rc;
 
-	if (efx->rss_context.context_id == EFX_MCDI_RSS_CONTEXT_INVALID ||
+	if (efx->rss_context.priv.context_id == EFX_MCDI_RSS_CONTEXT_INVALID ||
 	    !table->rx_rss_context_exclusive) {
-		rc = efx_mcdi_filter_alloc_rss_context(efx, true, &efx->rss_context,
-						NULL);
+		rc = efx_mcdi_filter_alloc_rss_context(efx, true,
+						       &efx->rss_context.priv,
+						       NULL);
 		if (rc == -EOPNOTSUPP)
 			return rc;
 		else if (rc != 0)
 			goto fail1;
 	}
 
-	rc = efx_mcdi_filter_populate_rss_table(efx, efx->rss_context.context_id,
-					 rx_indir_table, key);
+	rc = efx_mcdi_filter_populate_rss_table(efx, efx->rss_context.priv.context_id,
+						rx_indir_table, key);
 	if (rc != 0)
 		goto fail2;
 
-	if (efx->rss_context.context_id != old_rx_rss_context &&
+	if (efx->rss_context.priv.context_id != old_rx_rss_context &&
 	    old_rx_rss_context != EFX_MCDI_RSS_CONTEXT_INVALID)
 		WARN_ON(efx_mcdi_filter_free_rss_context(efx, old_rx_rss_context) != 0);
 	table->rx_rss_context_exclusive = true;
@@ -2091,9 +2094,9 @@ static int efx_mcdi_filter_rx_push_exclusive_rss_config(struct efx_nic *efx,
 	return 0;
 
 fail2:
-	if (old_rx_rss_context != efx->rss_context.context_id) {
-		WARN_ON(efx_mcdi_filter_free_rss_context(efx, efx->rss_context.context_id) != 0);
-		efx->rss_context.context_id = old_rx_rss_context;
+	if (old_rx_rss_context != efx->rss_context.priv.context_id) {
+		WARN_ON(efx_mcdi_filter_free_rss_context(efx, efx->rss_context.priv.context_id) != 0);
+		efx->rss_context.priv.context_id = old_rx_rss_context;
 	}
 fail1:
 	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
@@ -2101,33 +2104,28 @@ fail1:
 }
 
 int efx_mcdi_rx_push_rss_context_config(struct efx_nic *efx,
-					struct efx_rss_context *ctx,
+					struct efx_rss_context_priv *ctx,
 					const u32 *rx_indir_table,
-					const u8 *key)
+					const u8 *key, bool delete)
 {
 	int rc;
 
-	WARN_ON(!mutex_is_locked(&efx->rss_lock));
+	WARN_ON(!mutex_is_locked(&efx->net_dev->ethtool->rss_lock));
 
 	if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID) {
+		if (delete)
+			/* already wasn't in HW, nothing to do */
+			return 0;
 		rc = efx_mcdi_filter_alloc_rss_context(efx, true, ctx, NULL);
 		if (rc)
 			return rc;
 	}
 
-	if (!rx_indir_table) /* Delete this context */
+	if (delete) /* Delete this context */
 		return efx_mcdi_filter_free_rss_context(efx, ctx->context_id);
 
-	rc = efx_mcdi_filter_populate_rss_table(efx, ctx->context_id,
-					 rx_indir_table, key);
-	if (rc)
-		return rc;
-
-	memcpy(ctx->rx_indir_table, rx_indir_table,
-	       sizeof(efx->rss_context.rx_indir_table));
-	memcpy(ctx->rx_hash_key, key, efx->type->rx_hash_key_size);
-
-	return 0;
+	return efx_mcdi_filter_populate_rss_table(efx, ctx->context_id,
+						  rx_indir_table, key);
 }
 
 int efx_mcdi_rx_pull_rss_context_config(struct efx_nic *efx,
@@ -2139,16 +2137,16 @@ int efx_mcdi_rx_pull_rss_context_config(struct efx_nic *efx,
 	size_t outlen;
 	int rc, i;
 
-	WARN_ON(!mutex_is_locked(&efx->rss_lock));
+	WARN_ON(!mutex_is_locked(&efx->net_dev->ethtool->rss_lock));
 
 	BUILD_BUG_ON(MC_CMD_RSS_CONTEXT_GET_TABLE_IN_LEN !=
 		     MC_CMD_RSS_CONTEXT_GET_KEY_IN_LEN);
 
-	if (ctx->context_id == EFX_MCDI_RSS_CONTEXT_INVALID)
+	if (ctx->priv.context_id == EFX_MCDI_RSS_CONTEXT_INVALID)
 		return -ENOENT;
 
 	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_TABLE_IN_RSS_CONTEXT_ID,
-		       ctx->context_id);
+		       ctx->priv.context_id);
 	BUILD_BUG_ON(ARRAY_SIZE(ctx->rx_indir_table) !=
 		     MC_CMD_RSS_CONTEXT_GET_TABLE_OUT_INDIRECTION_TABLE_LEN);
 	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_GET_TABLE, inbuf, sizeof(inbuf),
@@ -2164,7 +2162,7 @@ int efx_mcdi_rx_pull_rss_context_config(struct efx_nic *efx,
 				RSS_CONTEXT_GET_TABLE_OUT_INDIRECTION_TABLE)[i];
 
 	MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_KEY_IN_RSS_CONTEXT_ID,
-		       ctx->context_id);
+		       ctx->priv.context_id);
 	BUILD_BUG_ON(ARRAY_SIZE(ctx->rx_hash_key) !=
 		     MC_CMD_RSS_CONTEXT_SET_KEY_IN_TOEPLITZ_KEY_LEN);
 	rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_GET_KEY, inbuf, sizeof(inbuf),
@@ -2186,35 +2184,42 @@ int efx_mcdi_rx_pull_rss_config(struct efx_nic *efx)
 {
 	int rc;
 
-	mutex_lock(&efx->rss_lock);
+	mutex_lock(&efx->net_dev->ethtool->rss_lock);
 	rc = efx_mcdi_rx_pull_rss_context_config(efx, &efx->rss_context);
-	mutex_unlock(&efx->rss_lock);
+	mutex_unlock(&efx->net_dev->ethtool->rss_lock);
 	return rc;
 }
 
 void efx_mcdi_rx_restore_rss_contexts(struct efx_nic *efx)
 {
 	struct efx_mcdi_filter_table *table = efx->filter_state;
-	struct efx_rss_context *ctx;
+	struct ethtool_rxfh_context *ctx;
+	unsigned long context;
 	int rc;
 
-	WARN_ON(!mutex_is_locked(&efx->rss_lock));
+	WARN_ON(!mutex_is_locked(&efx->net_dev->ethtool->rss_lock));
 
 	if (!table->must_restore_rss_contexts)
 		return;
 
-	list_for_each_entry(ctx, &efx->rss_context.list, list) {
+	xa_for_each(&efx->net_dev->ethtool->rss_ctx, context, ctx) {
+		struct efx_rss_context_priv *priv;
+		u32 *indir;
+		u8 *key;
+
+		priv = ethtool_rxfh_context_priv(ctx);
 		/* previous NIC RSS context is gone */
-		ctx->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
+		priv->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
 		/* so try to allocate a new one */
-		rc = efx_mcdi_rx_push_rss_context_config(efx, ctx,
-							 ctx->rx_indir_table,
-							 ctx->rx_hash_key);
+		indir = ethtool_rxfh_context_indir(ctx);
+		key = ethtool_rxfh_context_key(ctx);
+		rc = efx_mcdi_rx_push_rss_context_config(efx, priv, indir, key,
+							 false);
 		if (rc)
 			netif_warn(efx, probe, efx->net_dev,
-				   "failed to restore RSS context %u, rc=%d"
+				   "failed to restore RSS context %lu, rc=%d"
 				   "; RSS filters may fail to be applied\n",
-				   ctx->user_id, rc);
+				   context, rc);
 	}
 	table->must_restore_rss_contexts = false;
 }
@@ -2276,7 +2281,7 @@ int efx_mcdi_vf_rx_push_rss_config(struct efx_nic *efx, bool user,
 {
 	if (user)
 		return -EOPNOTSUPP;
-	if (efx->rss_context.context_id != EFX_MCDI_RSS_CONTEXT_INVALID)
+	if (efx->rss_context.priv.context_id != EFX_MCDI_RSS_CONTEXT_INVALID)
 		return 0;
 	return efx_mcdi_filter_rx_push_shared_rss_config(efx, NULL);
 }
@@ -2295,7 +2300,7 @@ int efx_mcdi_push_default_indir_table(struct efx_nic *efx,
 
 	efx_mcdi_rx_free_indir_table(efx);
 	if (rss_spread > 1) {
-		efx_set_default_rx_indir_table(efx, &efx->rss_context);
+		efx_set_default_rx_indir_table(efx, efx->rss_context.rx_indir_table);
 		rc = efx->type->rx_push_rss_config(efx, false,
 				   efx->rss_context.rx_indir_table, NULL);
 	}
diff --git a/drivers/net/ethernet/sfc/mcdi_filters.h b/drivers/net/ethernet/sfc/mcdi_filters.h
index c0d6558b9fd2..11b9f87ed9e1 100644
--- a/drivers/net/ethernet/sfc/mcdi_filters.h
+++ b/drivers/net/ethernet/sfc/mcdi_filters.h
@@ -145,9 +145,9 @@ void efx_mcdi_filter_del_vlan(struct efx_nic *efx, u16 vid);
 
 void efx_mcdi_rx_free_indir_table(struct efx_nic *efx);
 int efx_mcdi_rx_push_rss_context_config(struct efx_nic *efx,
-					struct efx_rss_context *ctx,
+					struct efx_rss_context_priv *ctx,
 					const u32 *rx_indir_table,
-					const u8 *key);
+					const u8 *key, bool delete);
 int efx_mcdi_pf_rx_push_rss_config(struct efx_nic *efx, bool user,
 				   const u32 *rx_indir_table,
 				   const u8 *key);
@@ -161,10 +161,6 @@ int efx_mcdi_push_default_indir_table(struct efx_nic *efx,
 int efx_mcdi_rx_pull_rss_config(struct efx_nic *efx);
 int efx_mcdi_rx_pull_rss_context_config(struct efx_nic *efx,
 					struct efx_rss_context *ctx);
-int efx_mcdi_get_rss_context_flags(struct efx_nic *efx, u32 context,
-				   u32 *flags);
-void efx_mcdi_set_rss_context_flags(struct efx_nic *efx,
-				    struct efx_rss_context *ctx);
 void efx_mcdi_rx_restore_rss_contexts(struct efx_nic *efx);
 
 static inline void efx_mcdi_update_rx_scatter(struct efx_nic *efx)
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index f2dd7feb0e0c..b85c51cbe7f9 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -737,21 +737,24 @@ struct vfdi_status;
 /* The reserved RSS context value */
 #define EFX_MCDI_RSS_CONTEXT_INVALID	0xffffffff
 /**
- * struct efx_rss_context - A user-defined RSS context for filtering
- * @list: node of linked list on which this struct is stored
+ * struct efx_rss_context_priv - driver private data for an RSS context
  * @context_id: the RSS_CONTEXT_ID returned by MC firmware, or
  *	%EFX_MCDI_RSS_CONTEXT_INVALID if this context is not present on the NIC.
- *	For Siena, 0 if RSS is active, else %EFX_MCDI_RSS_CONTEXT_INVALID.
- * @user_id: the rss_context ID exposed to userspace over ethtool.
  * @rx_hash_udp_4tuple: UDP 4-tuple hashing enabled
+ */
+struct efx_rss_context_priv {
+	u32 context_id;
+	bool rx_hash_udp_4tuple;
+};
+
+/**
+ * struct efx_rss_context - an RSS context
+ * @priv: hardware-specific state
  * @rx_hash_key: Toeplitz hash key for this RSS context
  * @indir_table: Indirection table for this RSS context
  */
 struct efx_rss_context {
-	struct list_head list;
-	u32 context_id;
-	u32 user_id;
-	bool rx_hash_udp_4tuple;
+	struct efx_rss_context_priv priv;
 	u8 rx_hash_key[40];
 	u32 rx_indir_table[128];
 };
@@ -883,9 +886,7 @@ struct efx_mae;
  * @rx_packet_ts_offset: Offset of timestamp from start of packet data
  *	(valid only if channel->sync_timestamps_enabled; always negative)
  * @rx_scatter: Scatter mode enabled for receives
- * @rss_context: Main RSS context.  Its @list member is the head of the list of
- *	RSS contexts created by user requests
- * @rss_lock: Protects custom RSS context software state in @rss_context.list
+ * @rss_context: Main RSS context.
  * @vport_id: The function's vport ID, only relevant for PFs
  * @int_error_count: Number of internal errors seen recently
  * @int_error_expire: Time at which error count will be expired
@@ -1052,7 +1053,6 @@ struct efx_nic {
 	int rx_packet_ts_offset;
 	bool rx_scatter;
 	struct efx_rss_context rss_context;
-	struct mutex rss_lock;
 	u32 vport_id;
 
 	unsigned int_error_count;
@@ -1416,9 +1416,9 @@ struct efx_nic_type {
 				  const u32 *rx_indir_table, const u8 *key);
 	int (*rx_pull_rss_config)(struct efx_nic *efx);
 	int (*rx_push_rss_context_config)(struct efx_nic *efx,
-					  struct efx_rss_context *ctx,
+					  struct efx_rss_context_priv *ctx,
 					  const u32 *rx_indir_table,
-					  const u8 *key);
+					  const u8 *key, bool delete);
 	int (*rx_pull_rss_context_config)(struct efx_nic *efx,
 					  struct efx_rss_context *ctx);
 	void (*rx_restore_rss_contexts)(struct efx_nic *efx);
diff --git a/drivers/net/ethernet/sfc/rx_common.c b/drivers/net/ethernet/sfc/rx_common.c
index dcd901eccfc8..0b7dc75c40f9 100644
--- a/drivers/net/ethernet/sfc/rx_common.c
+++ b/drivers/net/ethernet/sfc/rx_common.c
@@ -557,69 +557,25 @@ efx_rx_packet_gro(struct efx_channel *channel, struct efx_rx_buffer *rx_buf,
 	napi_gro_frags(napi);
 }
 
-/* RSS contexts.  We're using linked lists and crappy O(n) algorithms, because
- * (a) this is an infrequent control-plane operation and (b) n is small (max 64)
- */
-struct efx_rss_context *efx_alloc_rss_context_entry(struct efx_nic *efx)
+struct efx_rss_context_priv *efx_find_rss_context_entry(struct efx_nic *efx,
+							u32 id)
 {
-	struct list_head *head = &efx->rss_context.list;
-	struct efx_rss_context *ctx, *new;
-	u32 id = 1; /* Don't use zero, that refers to the master RSS context */
-
-	WARN_ON(!mutex_is_locked(&efx->rss_lock));
+	struct ethtool_rxfh_context *ctx;
 
-	/* Search for first gap in the numbering */
-	list_for_each_entry(ctx, head, list) {
-		if (ctx->user_id != id)
-			break;
-		id++;
-		/* Check for wrap.  If this happens, we have nearly 2^32
-		 * allocated RSS contexts, which seems unlikely.
-		 */
-		if (WARN_ON_ONCE(!id))
-			return NULL;
-	}
+	WARN_ON(!mutex_is_locked(&efx->net_dev->ethtool->rss_lock));
 
-	/* Create the new entry */
-	new = kmalloc(sizeof(*new), GFP_KERNEL);
-	if (!new)
+	ctx = xa_load(&efx->net_dev->ethtool->rss_ctx, id);
+	if (!ctx)
 		return NULL;
-	new->context_id = EFX_MCDI_RSS_CONTEXT_INVALID;
-	new->rx_hash_udp_4tuple = false;
-
-	/* Insert the new entry into the gap */
-	new->user_id = id;
-	list_add_tail(&new->list, &ctx->list);
-	return new;
-}
-
-struct efx_rss_context *efx_find_rss_context_entry(struct efx_nic *efx, u32 id)
-{
-	struct list_head *head = &efx->rss_context.list;
-	struct efx_rss_context *ctx;
-
-	WARN_ON(!mutex_is_locked(&efx->rss_lock));
-
-	list_for_each_entry(ctx, head, list)
-		if (ctx->user_id == id)
-			return ctx;
-	return NULL;
-}
-
-void efx_free_rss_context_entry(struct efx_rss_context *ctx)
-{
-	list_del(&ctx->list);
-	kfree(ctx);
+	return ethtool_rxfh_context_priv(ctx);
 }
 
-void efx_set_default_rx_indir_table(struct efx_nic *efx,
-				    struct efx_rss_context *ctx)
+void efx_set_default_rx_indir_table(struct efx_nic *efx, u32 *indir)
 {
 	size_t i;
 
-	for (i = 0; i < ARRAY_SIZE(ctx->rx_indir_table); i++)
-		ctx->rx_indir_table[i] =
-			ethtool_rxfh_indir_default(i, efx->rss_spread);
+	for (i = 0; i < ARRAY_SIZE(efx->rss_context.rx_indir_table); i++)
+		indir[i] = ethtool_rxfh_indir_default(i, efx->rss_spread);
 }
 
 /**
diff --git a/drivers/net/ethernet/sfc/rx_common.h b/drivers/net/ethernet/sfc/rx_common.h
index fbd2769307f9..75fa84192362 100644
--- a/drivers/net/ethernet/sfc/rx_common.h
+++ b/drivers/net/ethernet/sfc/rx_common.h
@@ -84,11 +84,9 @@ void
 efx_rx_packet_gro(struct efx_channel *channel, struct efx_rx_buffer *rx_buf,
 		  unsigned int n_frags, u8 *eh, __wsum csum);
 
-struct efx_rss_context *efx_alloc_rss_context_entry(struct efx_nic *efx);
-struct efx_rss_context *efx_find_rss_context_entry(struct efx_nic *efx, u32 id);
-void efx_free_rss_context_entry(struct efx_rss_context *ctx);
-void efx_set_default_rx_indir_table(struct efx_nic *efx,
-				    struct efx_rss_context *ctx);
+struct efx_rss_context_priv *efx_find_rss_context_entry(struct efx_nic *efx,
+							u32 id);
+void efx_set_default_rx_indir_table(struct efx_nic *efx, u32 *indir);
 
 bool efx_filter_is_mc_recipient(const struct efx_filter_spec *spec);
 bool efx_filter_spec_equal(const struct efx_filter_spec *left,
-- 
cgit v1.2.3-58-ga151


From 7964e7884643d56d4ecd0e2b92190f8a361cea48 Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Thu, 27 Jun 2024 16:33:53 +0100
Subject: net: ethtool: use the tracking array for get_rxfh on custom RSS
 contexts

On 'ethtool -x' with rss_context != 0, instead of calling the driver to
 read the RSS settings for the context, just get the settings from the
 rss_ctx xarray, and return them to the user with no driver involvement.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/2d0190fa29638f307ea720f882ebd41f6f867694.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ethtool/ioctl.c | 26 +++++++++++++++++++++-----
 1 file changed, 21 insertions(+), 5 deletions(-)

diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 939ccd106fe1..d8795ed07ba3 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -1202,6 +1202,7 @@ static noinline_for_stack int ethtool_get_rxfh(struct net_device *dev,
 	const struct ethtool_ops *ops = dev->ethtool_ops;
 	struct ethtool_rxfh_param rxfh_dev = {};
 	u32 user_indir_size, user_key_size;
+	struct ethtool_rxfh_context *ctx;
 	struct ethtool_rxfh rxfh;
 	u32 indir_bytes;
 	u8 *rss_config;
@@ -1249,11 +1250,26 @@ static noinline_for_stack int ethtool_get_rxfh(struct net_device *dev,
 	if (user_key_size)
 		rxfh_dev.key = rss_config + indir_bytes;
 
-	rxfh_dev.rss_context = rxfh.rss_context;
-
-	ret = dev->ethtool_ops->get_rxfh(dev, &rxfh_dev);
-	if (ret)
-		goto out;
+	if (rxfh.rss_context) {
+		ctx = xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context);
+		if (!ctx) {
+			ret = -ENOENT;
+			goto out;
+		}
+		if (rxfh_dev.indir)
+			memcpy(rxfh_dev.indir, ethtool_rxfh_context_indir(ctx),
+			       indir_bytes);
+		if (rxfh_dev.key)
+			memcpy(rxfh_dev.key, ethtool_rxfh_context_key(ctx),
+			       user_key_size);
+		rxfh_dev.hfunc = ctx->hfunc;
+		rxfh_dev.input_xfrm = ctx->input_xfrm;
+		ret = 0;
+	} else {
+		ret = dev->ethtool_ops->get_rxfh(dev, &rxfh_dev);
+		if (ret)
+			goto out;
+	}
 
 	if (copy_to_user(useraddr + offsetof(struct ethtool_rxfh, hfunc),
 			 &rxfh_dev.hfunc, sizeof(rxfh.hfunc))) {
-- 
cgit v1.2.3-58-ga151


From b859316e8218107f03dab83f1455fa46e949cd7c Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Thu, 27 Jun 2024 16:33:54 +0100
Subject: sfc: remove get_rxfh_context dead code

The core now always satisfies 'ethtool -x context nonzero' from its own
 tracking, so our lookup code for that case is never called.  Remove it.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/b426fcc416dedc8f203e52eebef6891eccebe4c1.1719502240.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/ethtool_common.c | 38 ++-----------------------------
 1 file changed, 2 insertions(+), 36 deletions(-)

diff --git a/drivers/net/ethernet/sfc/ethtool_common.c b/drivers/net/ethernet/sfc/ethtool_common.c
index 0a8d2c9ffce6..6ded44b86052 100644
--- a/drivers/net/ethernet/sfc/ethtool_common.c
+++ b/drivers/net/ethernet/sfc/ethtool_common.c
@@ -1163,48 +1163,14 @@ u32 efx_ethtool_get_rxfh_key_size(struct net_device *net_dev)
 	return efx->type->rx_hash_key_size;
 }
 
-static int efx_ethtool_get_rxfh_context(struct net_device *net_dev,
-					struct ethtool_rxfh_param *rxfh)
-{
-	struct efx_nic *efx = efx_netdev_priv(net_dev);
-	struct efx_rss_context_priv *ctx_priv;
-	struct efx_rss_context ctx;
-	int rc = 0;
-
-	if (!efx->type->rx_pull_rss_context_config)
-		return -EOPNOTSUPP;
-
-	mutex_lock(&net_dev->ethtool->rss_lock);
-	ctx_priv = efx_find_rss_context_entry(efx, rxfh->rss_context);
-	if (!ctx_priv) {
-		rc = -ENOENT;
-		goto out_unlock;
-	}
-	ctx.priv = *ctx_priv;
-	rc = efx->type->rx_pull_rss_context_config(efx, &ctx);
-	if (rc)
-		goto out_unlock;
-
-	rxfh->hfunc = ETH_RSS_HASH_TOP;
-	if (rxfh->indir)
-		memcpy(rxfh->indir, ctx.rx_indir_table,
-		       sizeof(ctx.rx_indir_table));
-	if (rxfh->key)
-		memcpy(rxfh->key, ctx.rx_hash_key,
-		       efx->type->rx_hash_key_size);
-out_unlock:
-	mutex_unlock(&net_dev->ethtool->rss_lock);
-	return rc;
-}
-
 int efx_ethtool_get_rxfh(struct net_device *net_dev,
 			 struct ethtool_rxfh_param *rxfh)
 {
 	struct efx_nic *efx = efx_netdev_priv(net_dev);
 	int rc;
 
-	if (rxfh->rss_context)
-		return efx_ethtool_get_rxfh_context(net_dev, rxfh);
+	if (rxfh->rss_context) /* core should never call us for these */
+		return -EINVAL;
 
 	rc = efx->type->rx_pull_rss_config(efx);
 	if (rc)
-- 
cgit v1.2.3-58-ga151


From 17784801d888238571a0c4101b9ac4401fffeaa0 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Tue, 25 Jun 2024 22:42:17 +0200
Subject: net: phy: realtek: Add support for PHY LEDs on RTL8211F

Realtek RTL8211F Ethernet PHY supports 3 LED pins which are used to
indicate link status and activity. Add minimal LED controller driver
supporting the most common uses with the 'netdev' trigger.

Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/realtek.c | 106 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 106 insertions(+)

diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c
index 2174893c974f..bed839237fb5 100644
--- a/drivers/net/phy/realtek.c
+++ b/drivers/net/phy/realtek.c
@@ -32,6 +32,15 @@
 #define RTL8211F_PHYCR2				0x19
 #define RTL8211F_INSR				0x1d
 
+#define RTL8211F_LEDCR				0x10
+#define RTL8211F_LEDCR_MODE			BIT(15)
+#define RTL8211F_LEDCR_ACT_TXRX			BIT(4)
+#define RTL8211F_LEDCR_LINK_1000		BIT(3)
+#define RTL8211F_LEDCR_LINK_100			BIT(1)
+#define RTL8211F_LEDCR_LINK_10			BIT(0)
+#define RTL8211F_LEDCR_MASK			GENMASK(4, 0)
+#define RTL8211F_LEDCR_SHIFT			5
+
 #define RTL8211F_TX_DELAY			BIT(8)
 #define RTL8211F_RX_DELAY			BIT(3)
 
@@ -87,6 +96,8 @@
 #define RTL_8221B_VN_CG				0x001cc84a
 #define RTL_8251B				0x001cc862
 
+#define RTL8211F_LED_COUNT			3
+
 MODULE_DESCRIPTION("Realtek PHY driver");
 MODULE_AUTHOR("Johnson Leung");
 MODULE_LICENSE("GPL");
@@ -476,6 +487,98 @@ static int rtl821x_resume(struct phy_device *phydev)
 	return 0;
 }
 
+static int rtl8211f_led_hw_is_supported(struct phy_device *phydev, u8 index,
+					unsigned long rules)
+{
+	const unsigned long mask = BIT(TRIGGER_NETDEV_LINK_10) |
+				   BIT(TRIGGER_NETDEV_LINK_100) |
+				   BIT(TRIGGER_NETDEV_LINK_1000) |
+				   BIT(TRIGGER_NETDEV_RX) |
+				   BIT(TRIGGER_NETDEV_TX);
+
+	/* The RTL8211F PHY supports these LED settings on up to three LEDs:
+	 * - Link: Configurable subset of 10/100/1000 link rates
+	 * - Active: Blink on activity, RX or TX is not differentiated
+	 * The Active option has two modes, A and B:
+	 * - A: Link and Active indication at configurable, but matching,
+	 *      subset of 10/100/1000 link rates
+	 * - B: Link indication at configurable subset of 10/100/1000 link
+	 *      rates and Active indication always at all three 10+100+1000
+	 *      link rates.
+	 * This code currently uses mode B only.
+	 */
+
+	if (index >= RTL8211F_LED_COUNT)
+		return -EINVAL;
+
+	/* Filter out any other unsupported triggers. */
+	if (rules & ~mask)
+		return -EOPNOTSUPP;
+
+	/* RX and TX are not differentiated, either both are set or not set. */
+	if (!(rules & BIT(TRIGGER_NETDEV_RX)) ^ !(rules & BIT(TRIGGER_NETDEV_TX)))
+		return -EOPNOTSUPP;
+
+	return 0;
+}
+
+static int rtl8211f_led_hw_control_get(struct phy_device *phydev, u8 index,
+				       unsigned long *rules)
+{
+	int val;
+
+	val = phy_read_paged(phydev, 0xd04, RTL8211F_LEDCR);
+	if (val < 0)
+		return val;
+
+	val >>= RTL8211F_LEDCR_SHIFT * index;
+	val &= RTL8211F_LEDCR_MASK;
+
+	if (val & RTL8211F_LEDCR_LINK_10)
+		set_bit(TRIGGER_NETDEV_LINK_10, rules);
+
+	if (val & RTL8211F_LEDCR_LINK_100)
+		set_bit(TRIGGER_NETDEV_LINK_100, rules);
+
+	if (val & RTL8211F_LEDCR_LINK_1000)
+		set_bit(TRIGGER_NETDEV_LINK_1000, rules);
+
+	if (val & RTL8211F_LEDCR_ACT_TXRX) {
+		set_bit(TRIGGER_NETDEV_RX, rules);
+		set_bit(TRIGGER_NETDEV_TX, rules);
+	}
+
+	return 0;
+}
+
+static int rtl8211f_led_hw_control_set(struct phy_device *phydev, u8 index,
+				       unsigned long rules)
+{
+	const u16 mask = RTL8211F_LEDCR_MASK << (RTL8211F_LEDCR_SHIFT * index);
+	u16 reg = RTL8211F_LEDCR_MODE;	/* Mode B */
+
+	if (index >= RTL8211F_LED_COUNT)
+		return -EINVAL;
+
+	if (test_bit(TRIGGER_NETDEV_LINK_10, &rules))
+		reg |= RTL8211F_LEDCR_LINK_10;
+
+	if (test_bit(TRIGGER_NETDEV_LINK_100, &rules))
+		reg |= RTL8211F_LEDCR_LINK_100;
+
+	if (test_bit(TRIGGER_NETDEV_LINK_1000, &rules))
+		reg |= RTL8211F_LEDCR_LINK_1000;
+
+	if (test_bit(TRIGGER_NETDEV_RX, &rules) ||
+	    test_bit(TRIGGER_NETDEV_TX, &rules)) {
+		reg |= RTL8211F_LEDCR_ACT_TXRX;
+	}
+
+	reg <<= RTL8211F_LEDCR_SHIFT * index;
+
+	return phy_modify_paged(phydev, 0xd04, RTL8211F_LEDCR, mask, reg);
+}
+
 static int rtl8211e_config_init(struct phy_device *phydev)
 {
 	int ret = 0, oldpage;
@@ -1192,6 +1295,9 @@ static struct phy_driver realtek_drvs[] = {
 		.read_page	= rtl821x_read_page,
 		.write_page	= rtl821x_write_page,
 		.flags		= PHY_ALWAYS_CALL_SUSPEND,
+		.led_hw_is_supported = rtl8211f_led_hw_is_supported,
+		.led_hw_control_get = rtl8211f_led_hw_control_get,
+		.led_hw_control_set = rtl8211f_led_hw_control_set,
 	}, {
 		PHY_ID_MATCH_EXACT(RTL_8211FVD_PHYID),
 		.name		= "RTL8211F-VD Gigabit Ethernet",
-- 
cgit v1.2.3-58-ga151


From 7c8110057b1bb9ab9f47ac0efc554d5c3a53b693 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 27 Jun 2024 14:35:50 -0700
Subject: tcp_metrics: add UAPI to the header guard

tcp_metrics' header lacks the customary _UAPI in the header guard.
This makes YNL build rules work less seamlessly.
We can easily fix that on YNL side, but this could also be
problematic if we ever needed to create a kernel-only tcp_metrics.h.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/uapi/linux/tcp_metrics.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/uapi/linux/tcp_metrics.h b/include/uapi/linux/tcp_metrics.h
index 7cb4a172feed..c48841076998 100644
--- a/include/uapi/linux/tcp_metrics.h
+++ b/include/uapi/linux/tcp_metrics.h
@@ -1,8 +1,8 @@
 /* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
 /* tcp_metrics.h - TCP Metrics Interface */
 
-#ifndef _LINUX_TCP_METRICS_H
-#define _LINUX_TCP_METRICS_H
+#ifndef _UAPI_LINUX_TCP_METRICS_H
+#define _UAPI_LINUX_TCP_METRICS_H
 
 #include <linux/types.h>
 
@@ -58,4 +58,4 @@ enum {
 
 #define TCP_METRICS_CMD_MAX	(__TCP_METRICS_CMD_MAX - 1)
 
-#endif /* _LINUX_TCP_METRICS_H */
+#endif /* _UAPI_LINUX_TCP_METRICS_H */
-- 
cgit v1.2.3-58-ga151


From 85674625e0bc25be4dbaa165a556ef6037328379 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 27 Jun 2024 14:35:51 -0700
Subject: tcp_metrics: add netlink protocol spec in YAML

Add a protocol spec for tcp_metrics, so that it's accessible via YNL.
Useful at the very least for testing fixes.

In this episode of "10,000 ways to complicate netlink" the metric
nest has defines which are off by 1. iproute2 does:

        struct rtattr *m[TCP_METRIC_MAX + 1 + 1];

        parse_rtattr_nested(m, TCP_METRIC_MAX + 1, a);

        for (i = 0; i < TCP_METRIC_MAX + 1; i++) {
                // ...
                attr = m[i + 1];

This is too weird to support in YNL, add a new set of defines
with _correct_ values to the official kernel header.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/netlink/specs/tcp_metrics.yaml | 169 +++++++++++++++++++++++++++
 include/uapi/linux/tcp_metrics.h             |  16 +++
 tools/net/ynl/Makefile.deps                  |   1 +
 3 files changed, 186 insertions(+)
 create mode 100644 Documentation/netlink/specs/tcp_metrics.yaml

diff --git a/Documentation/netlink/specs/tcp_metrics.yaml b/Documentation/netlink/specs/tcp_metrics.yaml
new file mode 100644
index 000000000000..1bd94f43e526
--- /dev/null
+++ b/Documentation/netlink/specs/tcp_metrics.yaml
@@ -0,0 +1,169 @@
+# SPDX-License-Identifier: ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause)
+
+name: tcp_metrics
+
+protocol: genetlink-legacy
+
+doc: |
+  Management interface for TCP metrics.
+
+c-family-name: tcp-metrics-genl-name
+c-version-name: tcp-metrics-genl-version
+max-by-define: true
+kernel-policy: global
+
+definitions:
+  -
+    name: tcp-fastopen-cookie-max
+    type: const
+    value: 16
+
+attribute-sets:
+  -
+    name: tcp-metrics
+    name-prefix: tcp-metrics-attr-
+    attributes:
+      -
+        name: addr-ipv4
+        type: u32
+        byte-order: big-endian
+        display-hint: ipv4
+      -
+        name: addr-ipv6
+        type: binary
+        checks:
+          min-len: 16
+        byte-order: big-endian
+        display-hint: ipv6
+      -
+        name: age
+        type: u64
+      -
+        name: tw-tsval
+        type: u32
+        doc: unused
+      -
+        name: tw-ts-stamp
+        type: s32
+        doc: unused
+      -
+        name: vals
+        type: nest
+        nested-attributes: metrics
+      -
+        name: fopen-mss
+        type: u16
+      -
+        name: fopen-syn-drops
+        type: u16
+      -
+        name: fopen-syn-drop-ts
+        type: u64
+      -
+        name: fopen-cookie
+        type: binary
+        checks:
+          min-len: tcp-fastopen-cookie-max
+      -
+        name: saddr-ipv4
+        type: u32
+        byte-order: big-endian
+        display-hint: ipv4
+      -
+        name: saddr-ipv6
+        type: binary
+        checks:
+          min-len: 16
+        byte-order: big-endian
+        display-hint: ipv6
+      -
+        name: pad
+        type: pad
+
+  -
+    name: metrics
+    # Intentionally don't define the name-prefix, see below.
+    doc: |
+      Attributes with metrics. Note that the values here do not match
+      the TCP_METRIC_* defines in the kernel, because kernel defines
+      are off-by one (e.g. rtt is defined as enum 0, while netlink carries
+      attribute type 1).
+    attributes:
+      -
+        name: rtt
+        type: u32
+        doc: |
+          Round Trip Time (RTT), in msecs with 3 bits fractional
+          (left-shift by 3 to get the msec value).
+      -
+        name: rttvar
+        type: u32
+        doc: |
+          Round Trip Time VARiance (RTT), in msecs with 2 bits fractional
+          (left-shift by 2 to get the msec value).
+      -
+        name: ssthresh
+        type: u32
+        doc: Slow Start THRESHold.
+      -
+        name: cwnd
+        type: u32
+        doc: Congestion Window.
+      -
+        name: reodering
+        type: u32
+        doc: Reodering metric.
+      -
+        name: rtt-us
+        type: u32
+        doc: |
+          Round Trip Time (RTT), in usecs, with 3 bits fractional
+          (left-shift by 3 to get the msec value).
+      -
+        name: rttvar-us
+        type: u32
+        doc: |
+          Round Trip Time (RTT), in usecs, with 2 bits fractional
+          (left-shift by 3 to get the msec value).
+
+operations:
+  list:
+    -
+      name: get
+      doc: Retrieve metrics.
+      attribute-set: tcp-metrics
+
+      dont-validate: [ strict, dump ]
+
+      do:
+        request: &sel_attrs
+          attributes:
+            - addr-ipv4
+            - addr-ipv6
+            - saddr-ipv4
+            - saddr-ipv6
+        reply: &all_attrs
+          attributes:
+            - addr-ipv4
+            - addr-ipv6
+            - saddr-ipv4
+            - saddr-ipv6
+            - age
+            - vals
+            - fopen-mss
+            - fopen-syn-drops
+            - fopen-syn-drop-ts
+            - fopen-cookie
+      dump:
+        reply: *all_attrs
+
+    -
+      name: del
+      doc: Delete metrics.
+      attribute-set: tcp-metrics
+
+      dont-validate: [ strict, dump ]
+      flags: [ admin-perm ]
+
+      do:
+        request: *sel_attrs
diff --git a/include/uapi/linux/tcp_metrics.h b/include/uapi/linux/tcp_metrics.h
index c48841076998..927c735a5b0e 100644
--- a/include/uapi/linux/tcp_metrics.h
+++ b/include/uapi/linux/tcp_metrics.h
@@ -27,6 +27,22 @@ enum tcp_metric_index {
 
 #define TCP_METRIC_MAX	(__TCP_METRIC_MAX - 1)
 
+/* Re-define enum tcp_metric_index, again, using the values carried
+ * as netlink attribute types.
+ */
+enum {
+	TCP_METRICS_A_METRICS_RTT = 1,
+	TCP_METRICS_A_METRICS_RTTVAR,
+	TCP_METRICS_A_METRICS_SSTHRESH,
+	TCP_METRICS_A_METRICS_CWND,
+	TCP_METRICS_A_METRICS_REODERING,
+	TCP_METRICS_A_METRICS_RTT_US,
+	TCP_METRICS_A_METRICS_RTTVAR_US,
+
+	__TCP_METRICS_A_METRICS_MAX
+};
+#define TCP_METRICS_A_METRICS_MAX (__TCP_METRICS_A_METRICS_MAX - 1)
+
 enum {
 	TCP_METRICS_ATTR_UNSPEC,
 	TCP_METRICS_ATTR_ADDR_IPV4,		/* u32 */
diff --git a/tools/net/ynl/Makefile.deps b/tools/net/ynl/Makefile.deps
index dbdca32a1c61..0712b5e82eb7 100644
--- a/tools/net/ynl/Makefile.deps
+++ b/tools/net/ynl/Makefile.deps
@@ -26,3 +26,4 @@ CFLAGS_nfsd:=$(call get_hdr_inc,_LINUX_NFSD_NETLINK_H,nfsd_netlink.h)
 CFLAGS_ovs_datapath:=$(call get_hdr_inc,__LINUX_OPENVSWITCH_H,openvswitch.h)
 CFLAGS_ovs_flow:=$(call get_hdr_inc,__LINUX_OPENVSWITCH_H,openvswitch.h)
 CFLAGS_ovs_vport:=$(call get_hdr_inc,__LINUX_OPENVSWITCH_H,openvswitch.h)
+CFLAGS_tcp_metrics:=$(call get_hdr_inc,_LINUX_TCP_METRICS_H,tcp_metrics.h)
-- 
cgit v1.2.3-58-ga151


From 7433d034ac3c2ea745ee345228122dbfbeca5b20 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@gmail.com>
Date: Fri, 28 Jun 2024 22:41:16 +0900
Subject: net: tn40xx: add initial ethtool_ops support

Call phylink_ethtool_ksettings_get() for get_link_ksettings method and
ethtool_op_get_link() for get_link method.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/tehuti/tn40.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/drivers/net/ethernet/tehuti/tn40.c b/drivers/net/ethernet/tehuti/tn40.c
index 11db9fde11fe..565b72537efa 100644
--- a/drivers/net/ethernet/tehuti/tn40.c
+++ b/drivers/net/ethernet/tehuti/tn40.c
@@ -1571,6 +1571,19 @@ static const struct net_device_ops tn40_netdev_ops = {
 	.ndo_vlan_rx_kill_vid = tn40_vlan_rx_kill_vid,
 };
 
+static int tn40_ethtool_get_link_ksettings(struct net_device *ndev,
+					   struct ethtool_link_ksettings *cmd)
+{
+	struct tn40_priv *priv = netdev_priv(ndev);
+
+	return phylink_ethtool_ksettings_get(priv->phylink, cmd);
+}
+
+static const struct ethtool_ops tn40_ethtool_ops = {
+	.get_link		= ethtool_op_get_link,
+	.get_link_ksettings	= tn40_ethtool_get_link_ksettings,
+};
+
 static int tn40_priv_init(struct tn40_priv *priv)
 {
 	int ret;
@@ -1599,6 +1612,7 @@ static struct net_device *tn40_netdev_alloc(struct pci_dev *pdev)
 	if (!ndev)
 		return NULL;
 	ndev->netdev_ops = &tn40_netdev_ops;
+	ndev->ethtool_ops = &tn40_ethtool_ops;
 	ndev->tx_queue_len = TN40_NDEV_TXQ_LEN;
 	ndev->mem_start = pci_resource_start(pdev, 0);
 	ndev->mem_end = pci_resource_end(pdev, 0);
-- 
cgit v1.2.3-58-ga151


From 42c45ac1419cad82f7f6ced6f57950ff40dfa689 Mon Sep 17 00:00:00 2001
From: Nithin Dabilpuram <ndabilpuram@marvell.com>
Date: Fri, 28 Jun 2024 22:01:26 +0530
Subject: octeontx2-af: Sync NIX and NPA contexts from NDC to LLC/DRAM

Octeontx2 hardware uses Near Data Cache(NDC) block to cache
contexts in it so that access to LLC/DRAM can be avoided.
It is recommended in HRM to sync the NDC contents before
releasing/resetting LF resources. Hence implement NDC_SYNC
mailbox and sync contexts during driver teardown.

Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/marvell/octeontx2/af/mbox.h   |  8 +++
 drivers/net/ethernet/marvell/octeontx2/af/rvu.c    | 66 ++++++++++++++++++++++
 drivers/net/ethernet/marvell/octeontx2/af/rvu.h    |  1 +
 .../net/ethernet/marvell/octeontx2/af/rvu_nix.c    |  4 +-
 .../net/ethernet/marvell/octeontx2/af/rvu_reg.h    |  2 +
 .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c   | 24 ++++++++
 6 files changed, 102 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
index 4a77f6fe2622..ebf84c240d6d 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
@@ -139,6 +139,7 @@ M(MSIX_OFFSET,		0x005, msix_offset, msg_req, msix_offset_rsp)	\
 M(VF_FLR,		0x006, vf_flr, msg_req, msg_rsp)		\
 M(PTP_OP,		0x007, ptp_op, ptp_req, ptp_rsp)		\
 M(GET_HW_CAP,		0x008, get_hw_cap, msg_req, get_hw_cap_rsp)	\
+M(NDC_SYNC_OP,		0x009, ndc_sync_op, ndc_sync_op, msg_rsp)	\
 M(LMTST_TBL_SETUP,	0x00a, lmtst_tbl_setup, lmtst_tbl_setup_req,    \
 				msg_rsp)				\
 M(SET_VF_PERM,		0x00b, set_vf_perm, set_vf_perm, msg_rsp)	\
@@ -1716,6 +1717,13 @@ struct lmtst_tbl_setup_req {
 	u64 rsvd[4];
 };
 
+struct ndc_sync_op {
+	struct mbox_msghdr hdr;
+	u8 nix_lf_tx_sync;
+	u8 nix_lf_rx_sync;
+	u8 npa_lf_sync;
+};
+
 /* CPT mailbox error codes
  * Range 901 - 1000.
  */
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
index ff78251f92d4..5a4f774aed64 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
@@ -2014,6 +2014,13 @@ int rvu_mbox_handler_vf_flr(struct rvu *rvu, struct msg_req *req,
 	return 0;
 }
 
+int rvu_ndc_sync(struct rvu *rvu, int lfblkaddr, int lfidx, u64 lfoffset)
+{
+	/* Sync cached info for this LF in NDC to LLC/DRAM */
+	rvu_write64(rvu, lfblkaddr, lfoffset, BIT_ULL(12) | lfidx);
+	return rvu_poll_reg(rvu, lfblkaddr, lfoffset, BIT_ULL(12), true);
+}
+
 int rvu_mbox_handler_get_hw_cap(struct rvu *rvu, struct msg_req *req,
 				struct get_hw_cap_rsp *rsp)
 {
@@ -2068,6 +2075,65 @@ int rvu_mbox_handler_set_vf_perm(struct rvu *rvu, struct set_vf_perm *req,
 	return 0;
 }
 
+int rvu_mbox_handler_ndc_sync_op(struct rvu *rvu,
+				 struct ndc_sync_op *req,
+				 struct msg_rsp *rsp)
+{
+	struct rvu_hwinfo *hw = rvu->hw;
+	u16 pcifunc = req->hdr.pcifunc;
+	int err, lfidx, lfblkaddr;
+
+	if (req->npa_lf_sync) {
+		/* Get NPA LF data */
+		lfblkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPA, pcifunc);
+		if (lfblkaddr < 0)
+			return NPA_AF_ERR_AF_LF_INVALID;
+
+		lfidx = rvu_get_lf(rvu, &hw->block[lfblkaddr], pcifunc, 0);
+		if (lfidx < 0)
+			return NPA_AF_ERR_AF_LF_INVALID;
+
+		/* Sync NPA NDC */
+		err = rvu_ndc_sync(rvu, lfblkaddr,
+				   lfidx, NPA_AF_NDC_SYNC);
+		if (err)
+			dev_err(rvu->dev,
+				"NDC-NPA sync failed for LF %u\n", lfidx);
+	}
+
+	if (!req->nix_lf_tx_sync && !req->nix_lf_rx_sync)
+		return 0;
+
+	/* Get NIX LF data */
+	lfblkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc);
+	if (lfblkaddr < 0)
+		return NIX_AF_ERR_AF_LF_INVALID;
+
+	lfidx = rvu_get_lf(rvu, &hw->block[lfblkaddr], pcifunc, 0);
+	if (lfidx < 0)
+		return NIX_AF_ERR_AF_LF_INVALID;
+
+	if (req->nix_lf_tx_sync) {
+		/* Sync NIX TX NDC */
+		err = rvu_ndc_sync(rvu, lfblkaddr,
+				   lfidx, NIX_AF_NDC_TX_SYNC);
+		if (err)
+			dev_err(rvu->dev,
+				"NDC-NIX-TX sync fail for LF %u\n", lfidx);
+	}
+
+	if (req->nix_lf_rx_sync) {
+		/* Sync NIX RX NDC */
+		err = rvu_ndc_sync(rvu, lfblkaddr,
+				   lfidx, NIX_AF_NDC_RX_SYNC);
+		if (err)
+			dev_err(rvu->dev,
+				"NDC-NIX-RX sync failed for LF %u\n", lfidx);
+	}
+
+	return 0;
+}
+
 static int rvu_process_mbox_msg(struct otx2_mbox *mbox, int devid,
 				struct mbox_msghdr *req)
 {
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
index 3063a84a45ef..03ee93fd9e94 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
@@ -800,6 +800,7 @@ int rvu_lf_reset(struct rvu *rvu, struct rvu_block *block, int lf);
 int rvu_get_blkaddr(struct rvu *rvu, int blktype, u16 pcifunc);
 int rvu_poll_reg(struct rvu *rvu, u64 block, u64 offset, u64 mask, bool zero);
 int rvu_get_num_lbk_chans(void);
+int rvu_ndc_sync(struct rvu *rvu, int lfblkid, int lfidx, u64 lfoffset);
 int rvu_get_blkaddr_from_slot(struct rvu *rvu, int blktype, u16 pcifunc,
 			      u16 global_slot, u16 *slot_in_block);
 
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
index 00af8888e329..4dbbaa719cbf 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
@@ -2497,9 +2497,7 @@ static int nix_txschq_free(struct rvu *rvu, u16 pcifunc)
 	}
 	mutex_unlock(&rvu->rsrc_lock);
 
-	/* Sync cached info for this LF in NDC-TX to LLC/DRAM */
-	rvu_write64(rvu, blkaddr, NIX_AF_NDC_TX_SYNC, BIT_ULL(12) | nixlf);
-	err = rvu_poll_reg(rvu, blkaddr, NIX_AF_NDC_TX_SYNC, BIT_ULL(12), true);
+	err = rvu_ndc_sync(rvu, blkaddr, nixlf, NIX_AF_NDC_TX_SYNC);
 	if (err)
 		dev_err(rvu->dev, "NDC-TX sync failed for NIXLF %d\n", nixlf);
 
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h
index 5ec92654e7ad..d56be5fb7eb4 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_reg.h
@@ -121,6 +121,7 @@
 #define NPA_AF_LF_RST                   (0x0020)
 #define NPA_AF_GEN_CFG                  (0x0030)
 #define NPA_AF_NDC_CFG                  (0x0040)
+#define NPA_AF_NDC_SYNC                 (0x0050)
 #define NPA_AF_INP_CTL                  (0x00D0)
 #define NPA_AF_ACTIVE_CYCLES_PC         (0x00F0)
 #define NPA_AF_AVG_DELAY                (0x0100)
@@ -239,6 +240,7 @@
 #define NIX_AF_RX_CPTX_INST_ADDR	(0x0310)
 #define NIX_AF_RX_CPTX_INST_QSEL(a)	(0x0320ull | (uint64_t)(a) << 3)
 #define NIX_AF_RX_CPTX_CREDIT(a)	(0x0360ull | (uint64_t)(a) << 3)
+#define NIX_AF_NDC_RX_SYNC		(0x03E0)
 #define NIX_AF_NDC_TX_SYNC		(0x03F0)
 #define NIX_AF_AQ_CFG			(0x0400)
 #define NIX_AF_AQ_BASE			(0x0410)
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index ff05ea20409a..5492dea547a1 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -3245,6 +3245,29 @@ static int otx2_sriov_configure(struct pci_dev *pdev, int numvfs)
 		return otx2_sriov_enable(pdev, numvfs);
 }
 
+static void otx2_ndc_sync(struct otx2_nic *pf)
+{
+	struct mbox *mbox = &pf->mbox;
+	struct ndc_sync_op *req;
+
+	mutex_lock(&mbox->lock);
+
+	req = otx2_mbox_alloc_msg_ndc_sync_op(mbox);
+	if (!req) {
+		mutex_unlock(&mbox->lock);
+		return;
+	}
+
+	req->nix_lf_tx_sync = 1;
+	req->nix_lf_rx_sync = 1;
+	req->npa_lf_sync = 1;
+
+	if (!otx2_sync_mbox_msg(mbox))
+		dev_err(pf->dev, "NDC sync operation failed\n");
+
+	mutex_unlock(&mbox->lock);
+}
+
 static void otx2_remove(struct pci_dev *pdev)
 {
 	struct net_device *netdev = pci_get_drvdata(pdev);
@@ -3293,6 +3316,7 @@ static void otx2_remove(struct pci_dev *pdev)
 	otx2_mcam_flow_del(pf);
 	otx2_shutdown_tc(pf);
 	otx2_shutdown_qos(pf);
+	otx2_ndc_sync(pf);
 	otx2_detach_resources(&pf->mbox);
 	if (pf->hw.lmt_info)
 		free_percpu(pf->hw.lmt_info);
-- 
cgit v1.2.3-58-ga151


From be6b7ca3c2ae47a7b1a51cd9bc15d638bc348dad Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Fri, 28 Jun 2024 12:29:56 -0700
Subject: bnxt_en: Add new TX timestamp completion definitions

The new BCM5760X chips will generate this new TX timestamp completion
when a TX packet's timestamp has been taken right before transmission.
The driver logic to retrieve the timestamp will be added in the next
few patches.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.h | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 9cf0acfa04e5..d3ad73d4c00a 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -181,6 +181,32 @@ struct tx_cmp {
 #define TX_CMP_SQ_CONS_IDX(txcmp)					\
 	(le32_to_cpu((txcmp)->sq_cons_idx) & TX_CMP_SQ_CONS_IDX_MASK)
 
+struct tx_ts_cmp {
+	__le32 tx_ts_cmp_flags_type;
+	#define TX_TS_CMP_FLAGS_ERROR				(1 << 6)
+	#define TX_TS_CMP_FLAGS_TS_TYPE				(1 << 7)
+	 #define TX_TS_CMP_FLAGS_TS_TYPE_PM			 (0 << 7)
+	 #define TX_TS_CMP_FLAGS_TS_TYPE_PA			 (1 << 7)
+	#define TX_TS_CMP_FLAGS_TS_FALLBACK			(1 << 8)
+	#define TX_TS_CMP_TS_SUB_NS				(0xf << 12)
+	#define TX_TS_CMP_TS_NS_MID				(0xffff << 16)
+	#define TX_TS_CMP_TS_NS_MID_SFT				16
+	u32 tx_ts_cmp_opaque;
+	__le32 tx_ts_cmp_errors_v;
+	#define TX_TS_CMP_V					(1 << 0)
+	#define TX_TS_CMP_TS_INVALID_ERR			(1 << 10)
+	__le32 tx_ts_cmp_ts_ns_lo;
+};
+
+#define BNXT_GET_TX_TS_48B_NS(tscmp)					\
+	(le32_to_cpu((tscmp)->tx_ts_cmp_ts_ns_lo) |			\
+	 ((u64)(le32_to_cpu((tscmp)->tx_ts_cmp_flags_type) &		\
+	  TX_TS_CMP_TS_NS_MID) << TX_TS_CMP_TS_NS_MID_SFT))
+
+#define BNXT_TX_TS_ERR(tscmp)						\
+	(((tscmp)->tx_ts_cmp_flags_type & cpu_to_le32(TX_TS_CMP_FLAGS_ERROR)) &&\
+	 ((tscmp)->tx_ts_cmp_errors_v & cpu_to_le32(TX_TS_CMP_TS_INVALID_ERR)))
+
 struct rx_cmp {
 	__le32 rx_cmp_len_flags_type;
 	#define RX_CMP_CMP_TYPE					(0x3f << 0)
-- 
cgit v1.2.3-58-ga151


From 449da97512f3bac1520a4d649038f7357f394e5b Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Fri, 28 Jun 2024 12:29:57 -0700
Subject: bnxt_en: Add is_ts_pkt field to struct bnxt_sw_tx_bd

Remove the unused is_gso field and add the is_ts_pkt field to struct
bnxt_sw_tx_bd.  This field will mark the TX BD that has requested
HW TX timestamp.  The field needs to be cleared if the timestamp packet
is later aborted.  This field will be useful when processing the
new TX timestamp completion from the hardware in the next patches.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 7 ++++++-
 drivers/net/ethernet/broadcom/bnxt/bnxt.h | 2 +-
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 1bd0c5973252..af064dcc6142 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -522,6 +522,7 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 				if (vlan_tag_flags)
 					ptp->tx_hdr_off += VLAN_HLEN;
 				lflags |= cpu_to_le32(TX_BD_FLAGS_STAMP);
+				tx_buf->is_ts_pkt = 1;
 				skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 			} else {
 				atomic_inc(&bp->ptp_cfg->tx_avail);
@@ -758,6 +759,7 @@ tx_free:
 	dev_kfree_skb_any(skb);
 tx_kick_pending:
 	if (BNXT_TX_PTP_IS_SET(lflags)) {
+		txr->tx_buf_ring[txr->tx_prod].is_ts_pkt = 0;
 		atomic64_inc(&bp->ptp_cfg->stats.ts_err);
 		atomic_inc(&bp->ptp_cfg->tx_avail);
 	}
@@ -781,6 +783,7 @@ static void __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
 	while (RING_TX(bp, cons) != hw_cons) {
 		struct bnxt_sw_tx_bd *tx_buf;
 		struct sk_buff *skb;
+		bool is_ts_pkt;
 		int j, last;
 
 		tx_buf = &txr->tx_buf_ring[RING_TX(bp, cons)];
@@ -800,6 +803,8 @@ static void __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
 			tx_buf->is_push = 0;
 			goto next_tx_int;
 		}
+		is_ts_pkt = tx_buf->is_ts_pkt;
+		tx_buf->is_ts_pkt = 0;
 
 		dma_unmap_single(&pdev->dev, dma_unmap_addr(tx_buf, mapping),
 				 skb_headlen(skb), DMA_TO_DEVICE);
@@ -814,7 +819,7 @@ static void __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
 				skb_frag_size(&skb_shinfo(skb)->frags[j]),
 				DMA_TO_DEVICE);
 		}
-		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) {
+		if (unlikely(is_ts_pkt)) {
 			if (BNXT_CHIP_P5(bp)) {
 				/* PTP worker takes ownership of the skb */
 				if (!bnxt_get_tx_ts_p5(bp, skb)) {
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index d3ad73d4c00a..00976e8a1e6a 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -874,7 +874,7 @@ struct bnxt_sw_tx_bd {
 	DEFINE_DMA_UNMAP_ADDR(mapping);
 	DEFINE_DMA_UNMAP_LEN(len);
 	struct page		*page;
-	u8			is_gso;
+	u8			is_ts_pkt;
 	u8			is_push;
 	u8			action;
 	unsigned short		nr_frags;
-- 
cgit v1.2.3-58-ga151


From ba0155f1e9fca8e5c59c840cf5451101b8360fe6 Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Fri, 28 Jun 2024 12:29:58 -0700
Subject: bnxt_en: Allow some TX packets to be unprocessed in NAPI

The driver's current logic will always free all the TX SKBs up to
txr->tx_hw_cons within NAPI.  In the next patches, we'll be adding
logic to handle TX timestamp completion and we may need to hold
some remaining TX SKBs if we don't have the timestamp completions
yet.

Modify __bnxt_poll_work_done() to clear each event bit separately to
allow bnapi->tx_int() to decide whether to clear BNXT_TX_CMP_EVENT or
not.  bnapi->tx_int() will not clear BNXT_TX_CMP_EVENT if some TX
SKBs are held waiting for TX timestamps.  Note that legacy chips will
never hold any SKBs this way.  The SKB is always deferred to the PTP
worker slow path to retrieve the timestamp from firmware.  On the new
P7 chips, the timestamp is returned by the hardware directly and we
can retrieve it directly from NAPI.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Andy Gospodarek <andrew.gospodarek@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index af064dcc6142..9dbf2967df1c 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -770,7 +770,8 @@ tx_kick_pending:
 	return NETDEV_TX_OK;
 }
 
-static void __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
+/* Returns true if some remaining TX packets not processed. */
+static bool __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
 			  int budget)
 {
 	struct netdev_queue *txq = netdev_get_tx_queue(bp->dev, txr->txq_index);
@@ -793,7 +794,7 @@ static void __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
 
 		if (unlikely(!skb)) {
 			bnxt_sched_reset_txr(bp, txr, cons);
-			return;
+			return false;
 		}
 
 		tx_pkts++;
@@ -842,18 +843,22 @@ next_tx_int:
 	__netif_txq_completed_wake(txq, tx_pkts, tx_bytes,
 				   bnxt_tx_avail(bp, txr), bp->tx_wake_thresh,
 				   READ_ONCE(txr->dev_state) == BNXT_DEV_STATE_CLOSING);
+
+	return false;
 }
 
 static void bnxt_tx_int(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)
 {
 	struct bnxt_tx_ring_info *txr;
+	bool more = false;
 	int i;
 
 	bnxt_for_each_napi_tx(i, bnapi, txr) {
 		if (txr->tx_hw_cons != RING_TX(bp, txr->tx_cons))
-			__bnxt_tx_int(bp, txr, budget);
+			more |= __bnxt_tx_int(bp, txr, budget);
 	}
-	bnapi->events &= ~BNXT_TX_CMP_EVENT;
+	if (!more)
+		bnapi->events &= ~BNXT_TX_CMP_EVENT;
 }
 
 static struct page *__bnxt_alloc_rx_page(struct bnxt *bp, dma_addr_t *mapping,
@@ -2950,8 +2955,10 @@ static int __bnxt_poll_work(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 		}
 	}
 
-	if (event & BNXT_REDIRECT_EVENT)
+	if (event & BNXT_REDIRECT_EVENT) {
 		xdp_do_flush();
+		event &= ~BNXT_REDIRECT_EVENT;
+	}
 
 	if (event & BNXT_TX_EVENT) {
 		struct bnxt_tx_ring_info *txr = bnapi->tx_ring[0];
@@ -2961,6 +2968,7 @@ static int __bnxt_poll_work(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 		wmb();
 
 		bnxt_db_write_relaxed(bp, &txr->tx_db, prod);
+		event &= ~BNXT_TX_EVENT;
 	}
 
 	cpr->cp_raw_cons = raw_cons;
@@ -2978,13 +2986,14 @@ static void __bnxt_poll_work_done(struct bnxt *bp, struct bnxt_napi *bnapi,
 		struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
 
 		bnxt_db_write(bp, &rxr->rx_db, rxr->rx_prod);
+		bnapi->events &= ~BNXT_RX_EVENT;
 	}
 	if (bnapi->events & BNXT_AGG_EVENT) {
 		struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
 
 		bnxt_db_write(bp, &rxr->rx_agg_db, rxr->rx_agg_prod);
+		bnapi->events &= ~BNXT_AGG_EVENT;
 	}
-	bnapi->events &= BNXT_TX_CMP_EVENT;
 }
 
 static int bnxt_poll_work(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
-- 
cgit v1.2.3-58-ga151


From 1d294b4f903fc3beb18f8f20da3cec46d333daab Mon Sep 17 00:00:00 2001
From: Michael Chan <michael.chan@broadcom.com>
Date: Fri, 28 Jun 2024 12:29:59 -0700
Subject: bnxt_en: Add TX timestamp completion logic

The new BCM5760X chips will return the timestamp of TX packets in a
new completion.  Add logic in __bnxt_poll_work() to handle this
completion type to retrieve the timestamp.  This feature eliminates
the limit on the number of in-flight PTP TX packets.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 40 +++++++++++++++++++--------
 drivers/net/ethernet/broadcom/bnxt/bnxt.h     |  1 +
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c | 32 +++++++++++++++++++++
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h |  2 ++
 4 files changed, 64 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 9dbf2967df1c..e6b0a937d9b1 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -456,6 +456,7 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	dma_addr_t mapping;
 	unsigned int length, pad = 0;
 	u32 len, free_size, vlan_tag_flags, cfa_action, flags;
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
 	u16 prod, last_frag;
 	struct pci_dev *pdev = bp->pdev;
 	struct bnxt_tx_ring_info *txr;
@@ -509,10 +510,13 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			vlan_tag_flags |= 1 << TX_BD_CFA_META_TPID_SHIFT;
 	}
 
-	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) {
-		struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
-
-		if (ptp && ptp->tx_tstamp_en && !skb_is_gso(skb)) {
+	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) && ptp &&
+	    ptp->tx_tstamp_en) {
+		if (bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP) {
+			lflags |= cpu_to_le32(TX_BD_FLAGS_STAMP);
+			tx_buf->is_ts_pkt = 1;
+			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+		} else if (!skb_is_gso(skb)) {
 			if (atomic_dec_if_positive(&ptp->tx_avail) < 0) {
 				atomic64_inc(&ptp->stats.ts_err);
 				goto tx_no_ts;
@@ -780,6 +784,7 @@ static bool __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
 	unsigned int tx_bytes = 0;
 	u16 cons = txr->tx_cons;
 	int tx_pkts = 0;
+	bool rc = false;
 
 	while (RING_TX(bp, cons) != hw_cons) {
 		struct bnxt_sw_tx_bd *tx_buf;
@@ -788,24 +793,29 @@ static bool __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
 		int j, last;
 
 		tx_buf = &txr->tx_buf_ring[RING_TX(bp, cons)];
-		cons = NEXT_TX(cons);
 		skb = tx_buf->skb;
-		tx_buf->skb = NULL;
 
 		if (unlikely(!skb)) {
 			bnxt_sched_reset_txr(bp, txr, cons);
-			return false;
+			return rc;
+		}
+
+		is_ts_pkt = tx_buf->is_ts_pkt;
+		if (is_ts_pkt && (bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP)) {
+			rc = true;
+			break;
 		}
 
+		cons = NEXT_TX(cons);
 		tx_pkts++;
 		tx_bytes += skb->len;
+		tx_buf->skb = NULL;
+		tx_buf->is_ts_pkt = 0;
 
 		if (tx_buf->is_push) {
 			tx_buf->is_push = 0;
 			goto next_tx_int;
 		}
-		is_ts_pkt = tx_buf->is_ts_pkt;
-		tx_buf->is_ts_pkt = 0;
 
 		dma_unmap_single(&pdev->dev, dma_unmap_addr(tx_buf, mapping),
 				 skb_headlen(skb), DMA_TO_DEVICE);
@@ -844,7 +854,7 @@ next_tx_int:
 				   bnxt_tx_avail(bp, txr), bp->tx_wake_thresh,
 				   READ_ONCE(txr->dev_state) == BNXT_DEV_STATE_CLOSING);
 
-	return false;
+	return rc;
 }
 
 static void bnxt_tx_int(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)
@@ -2924,6 +2934,8 @@ static int __bnxt_poll_work(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 					cpr->has_more_work = 1;
 				break;
 			}
+		} else if (cmp_type == CMP_TYPE_TX_L2_PKT_TS_CMP) {
+			bnxt_tx_ts_cmp(bp, bnapi, (struct tx_ts_cmp *)txcmp);
 		} else if (cmp_type >= CMP_TYPE_RX_L2_CMP &&
 			   cmp_type <= CMP_TYPE_RX_L2_TPA_START_V3_CMP) {
 			if (likely(budget))
@@ -6802,6 +6814,7 @@ static int hwrm_ring_alloc_send_msg(struct bnxt *bp,
 	switch (ring_type) {
 	case HWRM_RING_ALLOC_TX: {
 		struct bnxt_tx_ring_info *txr;
+		u16 flags = 0;
 
 		txr = container_of(ring, struct bnxt_tx_ring_info,
 				   tx_ring_struct);
@@ -6815,6 +6828,9 @@ static int hwrm_ring_alloc_send_msg(struct bnxt *bp,
 		if (bp->flags & BNXT_FLAG_TX_COAL_CMPL)
 			req->cmpl_coal_cnt =
 				RING_ALLOC_REQ_CMPL_COAL_CNT_COAL_64;
+		if ((bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP) && bp->ptp_cfg)
+			flags |= RING_ALLOC_REQ_FLAGS_TX_PKT_TS_CMPL_ENABLE;
+		req->flags = cpu_to_le16(flags);
 		break;
 	}
 	case HWRM_RING_ALLOC_RX:
@@ -9109,6 +9125,8 @@ static int __bnxt_hwrm_func_qcaps(struct bnxt *bp)
 		bp->fw_cap |= BNXT_FW_CAP_RX_ALL_PKT_TS;
 	if (flags_ext2 & FUNC_QCAPS_RESP_FLAGS_EXT2_UDP_GSO_SUPPORTED)
 		bp->flags |= BNXT_FLAG_UDP_GSO_CAP;
+	if (flags_ext2 & FUNC_QCAPS_RESP_FLAGS_EXT2_TX_PKT_TS_CMPL_SUPPORTED)
+		bp->fw_cap |= BNXT_FW_CAP_TX_TS_CMP;
 
 	bp->tx_push_thresh = 0;
 	if ((flags & FUNC_QCAPS_RESP_FLAGS_PUSH_MODE_SUPPORTED) &&
@@ -12150,7 +12168,7 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
 	/* VF-reps may need to be re-opened after the PF is re-opened */
 	if (BNXT_PF(bp))
 		bnxt_vf_reps_open(bp);
-	if (bp->ptp_cfg)
+	if (bp->ptp_cfg && !(bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP))
 		atomic_set(&bp->ptp_cfg->tx_avail, BNXT_MAX_TX_TS);
 	bnxt_ptp_init_rtc(bp, true);
 	bnxt_ptp_cfg_tstamp_filters(bp);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 00976e8a1e6a..12d6d17936a9 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -2410,6 +2410,7 @@ struct bnxt {
 	#define BNXT_FW_CAP_CFA_RFS_RING_TBL_IDX_V2	BIT_ULL(16)
 	#define BNXT_FW_CAP_PCIE_STATS_SUPPORTED	BIT_ULL(17)
 	#define BNXT_FW_CAP_EXT_STATS_SUPPORTED		BIT_ULL(18)
+	#define BNXT_FW_CAP_TX_TS_CMP			BIT_ULL(19)
 	#define BNXT_FW_CAP_ERR_RECOVER_RELOAD		BIT_ULL(20)
 	#define BNXT_FW_CAP_HOT_RESET			BIT_ULL(21)
 	#define BNXT_FW_CAP_PTP_RTC			BIT_ULL(22)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
index a14d46b9bfdf..0ca7bc75616b 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
@@ -768,6 +768,38 @@ int bnxt_get_rx_ts_p5(struct bnxt *bp, u64 *ts, u32 pkt_ts)
 	return 0;
 }
 
+void bnxt_tx_ts_cmp(struct bnxt *bp, struct bnxt_napi *bnapi,
+		    struct tx_ts_cmp *tscmp)
+{
+	struct skb_shared_hwtstamps timestamp = {};
+	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	u32 opaque = tscmp->tx_ts_cmp_opaque;
+	struct bnxt_tx_ring_info *txr;
+	struct bnxt_sw_tx_bd *tx_buf;
+	u64 ts, ns;
+	u16 cons;
+
+	txr = bnapi->tx_ring[TX_OPAQUE_RING(opaque)];
+	ts = BNXT_GET_TX_TS_48B_NS(tscmp);
+	cons = TX_OPAQUE_IDX(opaque);
+	tx_buf = &txr->tx_buf_ring[RING_TX(bp, cons)];
+	if (tx_buf->is_ts_pkt) {
+		if (BNXT_TX_TS_ERR(tscmp)) {
+			netdev_err(bp->dev,
+				   "timestamp completion error 0x%x 0x%x\n",
+				   le32_to_cpu(tscmp->tx_ts_cmp_flags_type),
+				   le32_to_cpu(tscmp->tx_ts_cmp_errors_v));
+		} else {
+			spin_lock_bh(&ptp->ptp_lock);
+			ns = timecounter_cyc2time(&ptp->tc, ts);
+			spin_unlock_bh(&ptp->ptp_lock);
+			timestamp.hwtstamp = ns_to_ktime(ns);
+			skb_tstamp_tx(tx_buf->skb, &timestamp);
+		}
+		tx_buf->is_ts_pkt = 0;
+	}
+}
+
 static const struct ptp_clock_info bnxt_ptp_caps = {
 	.owner		= THIS_MODULE,
 	.name		= "bnxt clock",
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
index 8c30b428a428..d38c3500827f 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
@@ -156,6 +156,8 @@ int bnxt_hwtstamp_set(struct net_device *dev, struct ifreq *ifr);
 int bnxt_hwtstamp_get(struct net_device *dev, struct ifreq *ifr);
 int bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb);
 int bnxt_get_rx_ts_p5(struct bnxt *bp, u64 *ts, u32 pkt_ts);
+void bnxt_tx_ts_cmp(struct bnxt *bp, struct bnxt_napi *bnapi,
+		    struct tx_ts_cmp *tscmp);
 void bnxt_ptp_rtc_timecounter_init(struct bnxt_ptp_cfg *ptp, u64 ns);
 int bnxt_ptp_init_rtc(struct bnxt *bp, bool phc_cfg);
 int bnxt_ptp_init(struct bnxt *bp, bool phc_cfg);
-- 
cgit v1.2.3-58-ga151


From 4d588d32b0321aaacf20bf90e5bbcc0f3a587b5e Mon Sep 17 00:00:00 2001
From: Pavan Chebbi <pavan.chebbi@broadcom.com>
Date: Fri, 28 Jun 2024 12:30:00 -0700
Subject: bnxt_en: Add BCM5760X specific PHC registers mapping

BCM5760X firmware will advertise direct 64-bit PHC registers access
for the driver from BAR0.

Make the necessary changes in handling HWRM_PORT_MAC_PTP_QCFG's
response and PHC register mapping for 5760X chips.

Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 12 ++++++++----
 drivers/net/ethernet/broadcom/bnxt/bnxt.h     |  8 ++++++++
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c | 10 +++++++++-
 3 files changed, 25 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index e6b0a937d9b1..f900bbbb7498 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -9011,7 +9011,7 @@ static int __bnxt_hwrm_ptp_qcfg(struct bnxt *bp)
 	u8 flags;
 	int rc;
 
-	if (bp->hwrm_spec_code < 0x10801 || !BNXT_CHIP_P5(bp)) {
+	if (bp->hwrm_spec_code < 0x10801 || !BNXT_CHIP_P5_PLUS(bp)) {
 		rc = -ENODEV;
 		goto no_ptp;
 	}
@@ -9027,7 +9027,8 @@ static int __bnxt_hwrm_ptp_qcfg(struct bnxt *bp)
 		goto exit;
 
 	flags = resp->flags;
-	if (!(flags & PORT_MAC_PTP_QCFG_RESP_FLAGS_HWRM_ACCESS)) {
+	if (BNXT_CHIP_P5_AND_MINUS(bp) &&
+	    !(flags & PORT_MAC_PTP_QCFG_RESP_FLAGS_HWRM_ACCESS)) {
 		rc = -ENODEV;
 		goto exit;
 	}
@@ -9040,10 +9041,13 @@ static int __bnxt_hwrm_ptp_qcfg(struct bnxt *bp)
 		ptp->bp = bp;
 		bp->ptp_cfg = ptp;
 	}
-	if (flags & PORT_MAC_PTP_QCFG_RESP_FLAGS_PARTIAL_DIRECT_ACCESS_REF_CLOCK) {
+
+	if (flags &
+	    (PORT_MAC_PTP_QCFG_RESP_FLAGS_PARTIAL_DIRECT_ACCESS_REF_CLOCK |
+	     PORT_MAC_PTP_QCFG_RESP_FLAGS_64B_PHC_TIME)) {
 		ptp->refclk_regs[0] = le32_to_cpu(resp->ts_ref_clock_reg_lower);
 		ptp->refclk_regs[1] = le32_to_cpu(resp->ts_ref_clock_reg_upper);
-	} else if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
+	} else if (BNXT_CHIP_P5(bp)) {
 		ptp->refclk_regs[0] = BNXT_TS_REG_TIMESYNC_TS0_LOWER;
 		ptp->refclk_regs[1] = BNXT_TS_REG_TIMESYNC_TS0_UPPER;
 	} else {
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 12d6d17936a9..82b05641953f 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -2263,9 +2263,17 @@ struct bnxt {
 	 (BNXT_CHIP_NUM_58700((bp)->chip_num) &&	\
 	  !BNXT_CHIP_TYPE_NITRO_A0(bp)))
 
+/* Chip class phase 3.x */
+#define BNXT_CHIP_P3(bp)			\
+	(BNXT_CHIP_NUM_57X0X((bp)->chip_num) ||	\
+	 BNXT_CHIP_TYPE_NITRO_A0(bp))
+
 #define BNXT_CHIP_P4_PLUS(bp)			\
 	(BNXT_CHIP_P4(bp) || BNXT_CHIP_P5_PLUS(bp))
 
+#define BNXT_CHIP_P5_AND_MINUS(bp)		\
+	(BNXT_CHIP_P3(bp) || BNXT_CHIP_P4(bp) || BNXT_CHIP_P5(bp))
+
 	struct bnxt_aux_priv	*aux_priv;
 	struct bnxt_en_dev	*edev;
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
index 0ca7bc75616b..a3795ef8887d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
@@ -656,6 +656,14 @@ static int bnxt_map_ptp_regs(struct bnxt *bp)
 				(ptp->refclk_regs[i] & BNXT_GRC_OFFSET_MASK);
 		return 0;
 	}
+	if (bp->flags & BNXT_FLAG_CHIP_P7) {
+		for (i = 0; i < 2; i++) {
+			if (reg_arr[i] & BNXT_GRC_BASE_MASK)
+				return -EINVAL;
+			ptp->refclk_mapped_regs[i] = reg_arr[i];
+		}
+		return 0;
+	}
 	return -ENODEV;
 }
 
@@ -1018,7 +1026,7 @@ int bnxt_ptp_init(struct bnxt *bp, bool phc_cfg)
 	ptp->stats.ts_lost = 0;
 	atomic64_set(&ptp->stats.ts_err, 0);
 
-	if (BNXT_CHIP_P5(bp)) {
+	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
 		spin_lock_bh(&ptp->ptp_lock);
 		bnxt_refclk_read(bp, NULL, &ptp->current_time);
 		WRITE_ONCE(ptp->old_time, ptp->current_time);
-- 
cgit v1.2.3-58-ga151


From 92595a0c022387dd855806c7d9dc0ad4a8b1accd Mon Sep 17 00:00:00 2001
From: Pavan Chebbi <pavan.chebbi@broadcom.com>
Date: Fri, 28 Jun 2024 12:30:01 -0700
Subject: bnxt_en: Refactor all PTP TX timestamp fields into a struct

On the older 5750X (P5) chips, we currently support only 1 TX PTP
packet in-flight waiting for the timestamp.  Refactor the
datastructures to prepare to support up to 4 TX PTP packets.

Combine all fields required for PTP TX timestamp query into one
structure.  An array of this structure will be added in follow-on
patches to support multiple outstanding TX timestamps.

Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 10 +++++--
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c | 43 ++++++++++++++-------------
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h | 15 ++++++----
 3 files changed, 40 insertions(+), 28 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index f900bbbb7498..a9741faa960c 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -517,14 +517,18 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			tx_buf->is_ts_pkt = 1;
 			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 		} else if (!skb_is_gso(skb)) {
+			u16 seq_id, hdr_off;
+
 			if (atomic_dec_if_positive(&ptp->tx_avail) < 0) {
 				atomic64_inc(&ptp->stats.ts_err);
 				goto tx_no_ts;
 			}
-			if (!bnxt_ptp_parse(skb, &ptp->tx_seqid,
-					    &ptp->tx_hdr_off)) {
+
+			if (!bnxt_ptp_parse(skb, &seq_id, &hdr_off)) {
 				if (vlan_tag_flags)
-					ptp->tx_hdr_off += VLAN_HLEN;
+					hdr_off += VLAN_HLEN;
+				ptp->txts_req.tx_seqid = seq_id;
+				ptp->txts_req.tx_hdr_off = hdr_off;
 				lflags |= cpu_to_le32(TX_BD_FLAGS_STAMP);
 				tx_buf->is_ts_pkt = 1;
 				skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
index a3795ef8887d..8431cd0ed9e9 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
@@ -123,11 +123,12 @@ static int bnxt_hwrm_port_ts_query(struct bnxt *bp, u32 flags, u64 *ts,
 	req->flags = cpu_to_le32(flags);
 	if ((flags & PORT_TS_QUERY_REQ_FLAGS_PATH) ==
 	    PORT_TS_QUERY_REQ_FLAGS_PATH_TX) {
+		struct bnxt_ptp_tx_req *txts_req = &bp->ptp_cfg->txts_req;
 		u32 tmo_us = txts_tmo * 1000;
 
 		req->enables = cpu_to_le16(BNXT_PTP_QTS_TX_ENABLES);
-		req->ptp_seq_id = cpu_to_le32(bp->ptp_cfg->tx_seqid);
-		req->ptp_hdr_offset = cpu_to_le16(bp->ptp_cfg->tx_hdr_off);
+		req->ptp_seq_id = cpu_to_le32(txts_req->tx_seqid);
+		req->ptp_hdr_offset = cpu_to_le16(txts_req->tx_hdr_off);
 		if (!tmo_us)
 			tmo_us = BNXT_PTP_QTS_TIMEOUT;
 		tmo_us = min(tmo_us, BNXT_PTP_QTS_MAX_TMO_US);
@@ -686,15 +687,17 @@ static void bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
 {
 	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
 	struct skb_shared_hwtstamps timestamp;
+	struct bnxt_ptp_tx_req *txts_req;
 	unsigned long now = jiffies;
 	u64 ts = 0, ns = 0;
 	u32 tmo = 0;
 	int rc;
 
-	if (!ptp->txts_pending)
-		ptp->abs_txts_tmo = now + msecs_to_jiffies(ptp->txts_tmo);
-	if (!time_after_eq(now, ptp->abs_txts_tmo))
-		tmo = jiffies_to_msecs(ptp->abs_txts_tmo - now);
+	txts_req = &ptp->txts_req;
+	if (!txts_req->txts_pending)
+		txts_req->abs_txts_tmo = now + msecs_to_jiffies(ptp->txts_tmo);
+	if (!time_after_eq(now, txts_req->abs_txts_tmo))
+		tmo = jiffies_to_msecs(txts_req->abs_txts_tmo - now);
 	rc = bnxt_hwrm_port_ts_query(bp, PORT_TS_QUERY_REQ_FLAGS_PATH_TX, &ts,
 				     tmo);
 	if (!rc) {
@@ -703,11 +706,11 @@ static void bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
 		ns = timecounter_cyc2time(&ptp->tc, ts);
 		spin_unlock_bh(&ptp->ptp_lock);
 		timestamp.hwtstamp = ns_to_ktime(ns);
-		skb_tstamp_tx(ptp->tx_skb, &timestamp);
+		skb_tstamp_tx(txts_req->tx_skb, &timestamp);
 		ptp->stats.ts_pkts++;
 	} else {
-		if (!time_after_eq(jiffies, ptp->abs_txts_tmo)) {
-			ptp->txts_pending = true;
+		if (!time_after_eq(jiffies, txts_req->abs_txts_tmo)) {
+			txts_req->txts_pending = true;
 			return;
 		}
 		ptp->stats.ts_lost++;
@@ -715,10 +718,10 @@ static void bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
 				 "TS query for TX timer failed rc = %x\n", rc);
 	}
 
-	dev_kfree_skb_any(ptp->tx_skb);
-	ptp->tx_skb = NULL;
+	dev_kfree_skb_any(txts_req->tx_skb);
+	txts_req->tx_skb = NULL;
 	atomic_inc(&ptp->tx_avail);
-	ptp->txts_pending = false;
+	txts_req->txts_pending = false;
 }
 
 static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
@@ -728,8 +731,8 @@ static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
 	unsigned long now = jiffies;
 	struct bnxt *bp = ptp->bp;
 
-	if (ptp->tx_skb)
-		bnxt_stamp_tx_skb(bp, ptp->tx_skb);
+	if (ptp->txts_req.tx_skb)
+		bnxt_stamp_tx_skb(bp, ptp->txts_req.tx_skb);
 
 	if (!time_after_eq(now, ptp->next_period))
 		return ptp->next_period - now;
@@ -742,7 +745,7 @@ static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
 		spin_unlock_bh(&ptp->ptp_lock);
 		ptp->next_overflow_check = now + BNXT_PHC_OVERFLOW_PERIOD;
 	}
-	if (ptp->txts_pending)
+	if (ptp->txts_req.txts_pending)
 		return 0;
 	return HZ;
 }
@@ -751,11 +754,11 @@ int bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb)
 {
 	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
 
-	if (ptp->tx_skb) {
+	if (ptp->txts_req.tx_skb) {
 		netdev_err(bp->dev, "deferring skb:one SKB is still outstanding\n");
 		return -EBUSY;
 	}
-	ptp->tx_skb = skb;
+	ptp->txts_req.tx_skb = skb;
 	ptp_schedule_worker(ptp->ptp_clock, 0);
 	return 0;
 }
@@ -1056,9 +1059,9 @@ void bnxt_ptp_clear(struct bnxt *bp)
 	kfree(ptp->ptp_info.pin_config);
 	ptp->ptp_info.pin_config = NULL;
 
-	if (ptp->tx_skb) {
-		dev_kfree_skb_any(ptp->tx_skb);
-		ptp->tx_skb = NULL;
+	if (ptp->txts_req.tx_skb) {
+		dev_kfree_skb_any(ptp->txts_req.tx_skb);
+		ptp->txts_req.tx_skb = NULL;
 	}
 
 	bnxt_unmap_ptp_regs(bp);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
index d38c3500827f..90f1418211e9 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
@@ -85,6 +85,14 @@ struct bnxt_ptp_stats {
 	atomic64_t	ts_err;
 };
 
+struct bnxt_ptp_tx_req {
+	struct sk_buff		*tx_skb;
+	u16			tx_seqid;
+	u16			tx_hdr_off;
+	u8			txts_pending:1;
+	unsigned long		abs_txts_tmo;
+};
+
 struct bnxt_ptp_cfg {
 	struct ptp_clock_info	ptp_info;
 	struct ptp_clock	*ptp_clock;
@@ -93,7 +101,6 @@ struct bnxt_ptp_cfg {
 	struct bnxt_pps		pps_info;
 	/* serialize timecounter access */
 	spinlock_t		ptp_lock;
-	struct sk_buff		*tx_skb;
 	u64			current_time;
 	u64			old_time;
 	unsigned long		next_period;
@@ -102,8 +109,8 @@ struct bnxt_ptp_cfg {
 	/* a 23b shift cyclecounter will overflow in ~36 mins.  Check overflow every 18 mins. */
 	#define BNXT_PHC_OVERFLOW_PERIOD	(18 * 60 * HZ)
 
-	u16			tx_seqid;
-	u16			tx_hdr_off;
+	struct bnxt_ptp_tx_req	txts_req;
+
 	struct bnxt		*bp;
 	atomic_t		tx_avail;
 #define BNXT_MAX_TX_TS	1
@@ -123,14 +130,12 @@ struct bnxt_ptp_cfg {
 					 BNXT_PTP_MSG_PDELAY_REQ |	\
 					 BNXT_PTP_MSG_PDELAY_RESP)
 	u8			tx_tstamp_en:1;
-	u8			txts_pending:1;
 	int			rx_filter;
 	u32			tstamp_filters;
 
 	u32			refclk_regs[2];
 	u32			refclk_mapped_regs[2];
 	u32			txts_tmo;
-	unsigned long		abs_txts_tmo;
 
 	struct bnxt_ptp_stats	stats;
 };
-- 
cgit v1.2.3-58-ga151


From 573f2a4bfcd49e0073aed51093291c7203b2d406 Mon Sep 17 00:00:00 2001
From: Pavan Chebbi <pavan.chebbi@broadcom.com>
Date: Fri, 28 Jun 2024 12:30:02 -0700
Subject: bnxt_en: Remove an impossible condition check for PTP TX pending SKB

In the current 5750X PTP code paths, there is always at most one TX
SKB requested for timestamp and we won't accept another one until we
have retrieved the timestamp or it has timed out.  Remove the
unnecessary check in bnxt_get_tx_ts_p5() for a pending SKB and change
the function to void.

Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 8 ++------
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c | 7 +------
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h | 2 +-
 3 files changed, 4 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index a9741faa960c..78ccfac0644e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -837,12 +837,8 @@ static bool __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
 		if (unlikely(is_ts_pkt)) {
 			if (BNXT_CHIP_P5(bp)) {
 				/* PTP worker takes ownership of the skb */
-				if (!bnxt_get_tx_ts_p5(bp, skb)) {
-					skb = NULL;
-				} else {
-					atomic64_inc(&bp->ptp_cfg->stats.ts_err);
-					atomic_inc(&bp->ptp_cfg->tx_avail);
-				}
+				bnxt_get_tx_ts_p5(bp, skb);
+				skb = NULL;
 			}
 		}
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
index 8431cd0ed9e9..baf191959b13 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
@@ -750,17 +750,12 @@ static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
 	return HZ;
 }
 
-int bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb)
+void bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb)
 {
 	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
 
-	if (ptp->txts_req.tx_skb) {
-		netdev_err(bp->dev, "deferring skb:one SKB is still outstanding\n");
-		return -EBUSY;
-	}
 	ptp->txts_req.tx_skb = skb;
 	ptp_schedule_worker(ptp->ptp_clock, 0);
-	return 0;
 }
 
 int bnxt_get_rx_ts_p5(struct bnxt *bp, u64 *ts, u32 pkt_ts)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
index 90f1418211e9..ee1709cda47e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
@@ -159,7 +159,7 @@ int bnxt_ptp_cfg_tstamp_filters(struct bnxt *bp);
 void bnxt_ptp_reapply_pps(struct bnxt *bp);
 int bnxt_hwtstamp_set(struct net_device *dev, struct ifreq *ifr);
 int bnxt_hwtstamp_get(struct net_device *dev, struct ifreq *ifr);
-int bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb);
+void bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb);
 int bnxt_get_rx_ts_p5(struct bnxt *bp, u64 *ts, u32 pkt_ts);
 void bnxt_tx_ts_cmp(struct bnxt *bp, struct bnxt_napi *bnapi,
 		    struct tx_ts_cmp *tscmp);
-- 
cgit v1.2.3-58-ga151


From 9bf688d40d66e86310d241e77d6c51967300cc1c Mon Sep 17 00:00:00 2001
From: Pavan Chebbi <pavan.chebbi@broadcom.com>
Date: Fri, 28 Jun 2024 12:30:03 -0700
Subject: bnxt_en: Let bnxt_stamp_tx_skb() return error code

Change the function bnxt_stamp_tx_skb() to return 0 for suceess
or -EAGAIN if the timestamp is still pending in firmware.  The
calling PTP aux worker will reschedule based on the return code.

Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
index baf191959b13..bd1e270307ec 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
@@ -683,7 +683,7 @@ static u64 bnxt_cc_read(const struct cyclecounter *cc)
 	return ns;
 }
 
-static void bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
+static int bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
 {
 	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
 	struct skb_shared_hwtstamps timestamp;
@@ -711,7 +711,7 @@ static void bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
 	} else {
 		if (!time_after_eq(jiffies, txts_req->abs_txts_tmo)) {
 			txts_req->txts_pending = true;
-			return;
+			return -EAGAIN;
 		}
 		ptp->stats.ts_lost++;
 		netdev_warn_once(bp->dev,
@@ -722,6 +722,8 @@ static void bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
 	txts_req->tx_skb = NULL;
 	atomic_inc(&ptp->tx_avail);
 	txts_req->txts_pending = false;
+
+	return 0;
 }
 
 static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
@@ -730,12 +732,16 @@ static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
 						ptp_info);
 	unsigned long now = jiffies;
 	struct bnxt *bp = ptp->bp;
+	int rc = 0;
 
 	if (ptp->txts_req.tx_skb)
-		bnxt_stamp_tx_skb(bp, ptp->txts_req.tx_skb);
+		rc = bnxt_stamp_tx_skb(bp, ptp->txts_req.tx_skb);
 
-	if (!time_after_eq(now, ptp->next_period))
+	if (!time_after_eq(now, ptp->next_period)) {
+		if (rc == -EAGAIN)
+			return 0;
 		return ptp->next_period - now;
+	}
 
 	bnxt_ptp_get_current_time(bp);
 	ptp->next_period = now + HZ;
@@ -745,7 +751,7 @@ static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
 		spin_unlock_bh(&ptp->ptp_lock);
 		ptp->next_overflow_check = now + BNXT_PHC_OVERFLOW_PERIOD;
 	}
-	if (ptp->txts_req.txts_pending)
+	if (rc == -EAGAIN)
 		return 0;
 	return HZ;
 }
-- 
cgit v1.2.3-58-ga151


From 8aa2a79e9b95eadec5c2a97662adb5b0a8f333ad Mon Sep 17 00:00:00 2001
From: Pavan Chebbi <pavan.chebbi@broadcom.com>
Date: Fri, 28 Jun 2024 12:30:04 -0700
Subject: bnxt_en: Increase the max total outstanding PTP TX packets to 4

Start accepting up to 4 TX TS requests on BCM5750X (P5) chips.
These PTP TX packets will be queued in the ptp->txts_req[] array
waiting for the TX timestamp to complete.  The entries in the
array will be managed by a producer and consumer index.  The
producer index is updated under spinlock since multiple TX rings
can try to send PTP packets at the same time.

Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 20 ++++++---
 drivers/net/ethernet/broadcom/bnxt/bnxt.h     |  5 ++-
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c | 60 ++++++++++++++++++---------
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h | 13 ++++--
 4 files changed, 68 insertions(+), 30 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 78ccfac0644e..9cb81b74fac9 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -457,8 +457,8 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	unsigned int length, pad = 0;
 	u32 len, free_size, vlan_tag_flags, cfa_action, flags;
 	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
-	u16 prod, last_frag;
 	struct pci_dev *pdev = bp->pdev;
+	u16 prod, last_frag, txts_prod;
 	struct bnxt_tx_ring_info *txr;
 	struct bnxt_sw_tx_bd *tx_buf;
 	__le32 lflags = 0;
@@ -527,11 +527,19 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			if (!bnxt_ptp_parse(skb, &seq_id, &hdr_off)) {
 				if (vlan_tag_flags)
 					hdr_off += VLAN_HLEN;
-				ptp->txts_req.tx_seqid = seq_id;
-				ptp->txts_req.tx_hdr_off = hdr_off;
 				lflags |= cpu_to_le32(TX_BD_FLAGS_STAMP);
 				tx_buf->is_ts_pkt = 1;
 				skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
+
+				spin_lock_bh(&ptp->ptp_tx_lock);
+				txts_prod = ptp->txts_prod;
+				ptp->txts_prod = NEXT_TXTS(txts_prod);
+				spin_unlock_bh(&ptp->ptp_tx_lock);
+
+				ptp->txts_req[txts_prod].tx_seqid = seq_id;
+				ptp->txts_req[txts_prod].tx_hdr_off = hdr_off;
+				tx_buf->txts_prod = txts_prod;
+
 			} else {
 				atomic_inc(&bp->ptp_cfg->tx_avail);
 			}
@@ -769,7 +777,9 @@ tx_kick_pending:
 	if (BNXT_TX_PTP_IS_SET(lflags)) {
 		txr->tx_buf_ring[txr->tx_prod].is_ts_pkt = 0;
 		atomic64_inc(&bp->ptp_cfg->stats.ts_err);
-		atomic_inc(&bp->ptp_cfg->tx_avail);
+		if (!(bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP))
+			/* set SKB to err so PTP worker will clean up */
+			ptp->txts_req[txts_prod].tx_skb = ERR_PTR(-EIO);
 	}
 	if (txr->kick_pending)
 		bnxt_txr_db_kick(bp, txr, txr->tx_prod);
@@ -837,7 +847,7 @@ static bool __bnxt_tx_int(struct bnxt *bp, struct bnxt_tx_ring_info *txr,
 		if (unlikely(is_ts_pkt)) {
 			if (BNXT_CHIP_P5(bp)) {
 				/* PTP worker takes ownership of the skb */
-				bnxt_get_tx_ts_p5(bp, skb);
+				bnxt_get_tx_ts_p5(bp, skb, tx_buf->txts_prod);
 				skb = NULL;
 			}
 		}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 82b05641953f..e46bd11e52b0 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -878,7 +878,10 @@ struct bnxt_sw_tx_bd {
 	u8			is_push;
 	u8			action;
 	unsigned short		nr_frags;
-	u16			rx_prod;
+	union {
+		u16			rx_prod;
+		u16			txts_prod;
+	};
 };
 
 struct bnxt_sw_rx_bd {
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
index bd1e270307ec..9e93dc8b2b57 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
@@ -110,7 +110,7 @@ static void bnxt_ptp_get_current_time(struct bnxt *bp)
 }
 
 static int bnxt_hwrm_port_ts_query(struct bnxt *bp, u32 flags, u64 *ts,
-				   u32 txts_tmo)
+				   u32 txts_tmo, int slot)
 {
 	struct hwrm_port_ts_query_output *resp;
 	struct hwrm_port_ts_query_input *req;
@@ -123,7 +123,7 @@ static int bnxt_hwrm_port_ts_query(struct bnxt *bp, u32 flags, u64 *ts,
 	req->flags = cpu_to_le32(flags);
 	if ((flags & PORT_TS_QUERY_REQ_FLAGS_PATH) ==
 	    PORT_TS_QUERY_REQ_FLAGS_PATH_TX) {
-		struct bnxt_ptp_tx_req *txts_req = &bp->ptp_cfg->txts_req;
+		struct bnxt_ptp_tx_req *txts_req = &bp->ptp_cfg->txts_req[slot];
 		u32 tmo_us = txts_tmo * 1000;
 
 		req->enables = cpu_to_le16(BNXT_PTP_QTS_TX_ENABLES);
@@ -683,7 +683,7 @@ static u64 bnxt_cc_read(const struct cyclecounter *cc)
 	return ns;
 }
 
-static int bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
+static int bnxt_stamp_tx_skb(struct bnxt *bp, int slot)
 {
 	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
 	struct skb_shared_hwtstamps timestamp;
@@ -693,13 +693,13 @@ static int bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
 	u32 tmo = 0;
 	int rc;
 
-	txts_req = &ptp->txts_req;
-	if (!txts_req->txts_pending)
-		txts_req->abs_txts_tmo = now + msecs_to_jiffies(ptp->txts_tmo);
+	txts_req = &ptp->txts_req[slot];
+	/* make sure bnxt_get_tx_ts_p5() has updated abs_txts_tmo */
+	smp_rmb();
 	if (!time_after_eq(now, txts_req->abs_txts_tmo))
 		tmo = jiffies_to_msecs(txts_req->abs_txts_tmo - now);
 	rc = bnxt_hwrm_port_ts_query(bp, PORT_TS_QUERY_REQ_FLAGS_PATH_TX, &ts,
-				     tmo);
+				     tmo, slot);
 	if (!rc) {
 		memset(&timestamp, 0, sizeof(timestamp));
 		spin_lock_bh(&ptp->ptp_lock);
@@ -709,10 +709,9 @@ static int bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
 		skb_tstamp_tx(txts_req->tx_skb, &timestamp);
 		ptp->stats.ts_pkts++;
 	} else {
-		if (!time_after_eq(jiffies, txts_req->abs_txts_tmo)) {
-			txts_req->txts_pending = true;
+		if (!time_after_eq(jiffies, txts_req->abs_txts_tmo))
 			return -EAGAIN;
-		}
+
 		ptp->stats.ts_lost++;
 		netdev_warn_once(bp->dev,
 				 "TS query for TX timer failed rc = %x\n", rc);
@@ -720,8 +719,6 @@ static int bnxt_stamp_tx_skb(struct bnxt *bp, struct sk_buff *skb)
 
 	dev_kfree_skb_any(txts_req->tx_skb);
 	txts_req->tx_skb = NULL;
-	atomic_inc(&ptp->tx_avail);
-	txts_req->txts_pending = false;
 
 	return 0;
 }
@@ -732,10 +729,24 @@ static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
 						ptp_info);
 	unsigned long now = jiffies;
 	struct bnxt *bp = ptp->bp;
+	u16 cons = ptp->txts_cons;
+	u8 num_requests;
 	int rc = 0;
 
-	if (ptp->txts_req.tx_skb)
-		rc = bnxt_stamp_tx_skb(bp, ptp->txts_req.tx_skb);
+	num_requests = BNXT_MAX_TX_TS - atomic_read(&ptp->tx_avail);
+	while (num_requests--) {
+		if (IS_ERR(ptp->txts_req[cons].tx_skb))
+			goto next_slot;
+		if (!ptp->txts_req[cons].tx_skb)
+			break;
+		rc = bnxt_stamp_tx_skb(bp, cons);
+		if (rc == -EAGAIN)
+			break;
+next_slot:
+		atomic_inc(&ptp->tx_avail);
+		cons = NEXT_TXTS(cons);
+	}
+	ptp->txts_cons = cons;
 
 	if (!time_after_eq(now, ptp->next_period)) {
 		if (rc == -EAGAIN)
@@ -756,11 +767,16 @@ static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
 	return HZ;
 }
 
-void bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb)
+void bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb, u16 prod)
 {
 	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	struct bnxt_ptp_tx_req *txts_req;
 
-	ptp->txts_req.tx_skb = skb;
+	txts_req = &ptp->txts_req[prod];
+	txts_req->abs_txts_tmo = jiffies + msecs_to_jiffies(ptp->txts_tmo);
+	/* make sure abs_txts_tmo is written first */
+	smp_wmb();
+	txts_req->tx_skb = skb;
 	ptp_schedule_worker(ptp->ptp_clock, 0);
 }
 
@@ -958,7 +974,7 @@ int bnxt_ptp_init_rtc(struct bnxt *bp, bool phc_cfg)
 			return rc;
 	} else {
 		rc = bnxt_hwrm_port_ts_query(bp, PORT_TS_QUERY_REQ_FLAGS_CURRENT_TIME,
-					     &ns, 0);
+					     &ns, 0, 0);
 		if (rc)
 			return rc;
 	}
@@ -1000,6 +1016,7 @@ int bnxt_ptp_init(struct bnxt *bp, bool phc_cfg)
 
 	atomic_set(&ptp->tx_avail, BNXT_MAX_TX_TS);
 	spin_lock_init(&ptp->ptp_lock);
+	spin_lock_init(&ptp->ptp_tx_lock);
 
 	if (BNXT_PTP_USE_RTC(bp)) {
 		bnxt_ptp_timecounter_init(bp, false);
@@ -1049,6 +1066,7 @@ out:
 void bnxt_ptp_clear(struct bnxt *bp)
 {
 	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
+	int i;
 
 	if (!ptp)
 		return;
@@ -1060,9 +1078,11 @@ void bnxt_ptp_clear(struct bnxt *bp)
 	kfree(ptp->ptp_info.pin_config);
 	ptp->ptp_info.pin_config = NULL;
 
-	if (ptp->txts_req.tx_skb) {
-		dev_kfree_skb_any(ptp->txts_req.tx_skb);
-		ptp->txts_req.tx_skb = NULL;
+	for (i = 0; i < BNXT_MAX_TX_TS; i++) {
+		if (ptp->txts_req[i].tx_skb) {
+			dev_kfree_skb_any(ptp->txts_req[i].tx_skb);
+			ptp->txts_req[i].tx_skb = NULL;
+		}
 	}
 
 	bnxt_unmap_ptp_regs(bp);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
index ee1709cda47e..a1910ce86cbb 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
@@ -85,11 +85,13 @@ struct bnxt_ptp_stats {
 	atomic64_t	ts_err;
 };
 
+#define BNXT_MAX_TX_TS		4
+#define NEXT_TXTS(idx)		(((idx) + 1) & (BNXT_MAX_TX_TS - 1))
+
 struct bnxt_ptp_tx_req {
 	struct sk_buff		*tx_skb;
 	u16			tx_seqid;
 	u16			tx_hdr_off;
-	u8			txts_pending:1;
 	unsigned long		abs_txts_tmo;
 };
 
@@ -101,6 +103,8 @@ struct bnxt_ptp_cfg {
 	struct bnxt_pps		pps_info;
 	/* serialize timecounter access */
 	spinlock_t		ptp_lock;
+	/* serialize ts tx request queuing */
+	spinlock_t		ptp_tx_lock;
 	u64			current_time;
 	u64			old_time;
 	unsigned long		next_period;
@@ -109,11 +113,10 @@ struct bnxt_ptp_cfg {
 	/* a 23b shift cyclecounter will overflow in ~36 mins.  Check overflow every 18 mins. */
 	#define BNXT_PHC_OVERFLOW_PERIOD	(18 * 60 * HZ)
 
-	struct bnxt_ptp_tx_req	txts_req;
+	struct bnxt_ptp_tx_req	txts_req[BNXT_MAX_TX_TS];
 
 	struct bnxt		*bp;
 	atomic_t		tx_avail;
-#define BNXT_MAX_TX_TS	1
 	u16			rxctl;
 #define BNXT_PTP_MSG_SYNC			(1 << 0)
 #define BNXT_PTP_MSG_DELAY_REQ			(1 << 1)
@@ -136,6 +139,8 @@ struct bnxt_ptp_cfg {
 	u32			refclk_regs[2];
 	u32			refclk_mapped_regs[2];
 	u32			txts_tmo;
+	u16			txts_prod;
+	u16			txts_cons;
 
 	struct bnxt_ptp_stats	stats;
 };
@@ -159,7 +164,7 @@ int bnxt_ptp_cfg_tstamp_filters(struct bnxt *bp);
 void bnxt_ptp_reapply_pps(struct bnxt *bp);
 int bnxt_hwtstamp_set(struct net_device *dev, struct ifreq *ifr);
 int bnxt_hwtstamp_get(struct net_device *dev, struct ifreq *ifr);
-void bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb);
+void bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb, u16 prod);
 int bnxt_get_rx_ts_p5(struct bnxt *bp, u64 *ts, u32 pkt_ts);
 void bnxt_tx_ts_cmp(struct bnxt *bp, struct bnxt_napi *bnapi,
 		    struct tx_ts_cmp *tscmp);
-- 
cgit v1.2.3-58-ga151


From 060338390787ccbfd71f0796d420b7cd063bcac6 Mon Sep 17 00:00:00 2001
From: Pavan Chebbi <pavan.chebbi@broadcom.com>
Date: Fri, 28 Jun 2024 12:30:05 -0700
Subject: bnxt_en: Remove atomic operations on ptp->tx_avail

Now that we require the spinlock to protect ptp->txts_prod, change
ptp->tx_avail to non-atomic and protect it under the same spinlock.
Add a new helper function bnxt_ptp_get_txts_prod() to decrement
ptp->tx_avail under spinlock and return the producer.

Signed-off-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 20 +++-----------------
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c | 23 +++++++++++++++++++----
 drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h | 10 +++++++++-
 3 files changed, 31 insertions(+), 22 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 9cb81b74fac9..a1690207d793 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -519,34 +519,20 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		} else if (!skb_is_gso(skb)) {
 			u16 seq_id, hdr_off;
 
-			if (atomic_dec_if_positive(&ptp->tx_avail) < 0) {
-				atomic64_inc(&ptp->stats.ts_err);
-				goto tx_no_ts;
-			}
-
-			if (!bnxt_ptp_parse(skb, &seq_id, &hdr_off)) {
+			if (!bnxt_ptp_parse(skb, &seq_id, &hdr_off) &&
+			    !bnxt_ptp_get_txts_prod(ptp, &txts_prod)) {
 				if (vlan_tag_flags)
 					hdr_off += VLAN_HLEN;
 				lflags |= cpu_to_le32(TX_BD_FLAGS_STAMP);
 				tx_buf->is_ts_pkt = 1;
 				skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 
-				spin_lock_bh(&ptp->ptp_tx_lock);
-				txts_prod = ptp->txts_prod;
-				ptp->txts_prod = NEXT_TXTS(txts_prod);
-				spin_unlock_bh(&ptp->ptp_tx_lock);
-
 				ptp->txts_req[txts_prod].tx_seqid = seq_id;
 				ptp->txts_req[txts_prod].tx_hdr_off = hdr_off;
 				tx_buf->txts_prod = txts_prod;
-
-			} else {
-				atomic_inc(&bp->ptp_cfg->tx_avail);
 			}
 		}
 	}
-
-tx_no_ts:
 	if (unlikely(skb->no_fcs))
 		lflags |= cpu_to_le32(TX_BD_FLAGS_NO_CRC);
 
@@ -12183,7 +12169,7 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
 	if (BNXT_PF(bp))
 		bnxt_vf_reps_open(bp);
 	if (bp->ptp_cfg && !(bp->fw_cap & BNXT_FW_CAP_TX_TS_CMP))
-		atomic_set(&bp->ptp_cfg->tx_avail, BNXT_MAX_TX_TS);
+		WRITE_ONCE(bp->ptp_cfg->tx_avail, BNXT_MAX_TX_TS);
 	bnxt_ptp_init_rtc(bp, true);
 	bnxt_ptp_cfg_tstamp_filters(bp);
 	if (BNXT_SUPPORTS_MULTI_RSS_CTX(bp))
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
index 9e93dc8b2b57..37d42423459c 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.c
@@ -730,10 +730,10 @@ static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
 	unsigned long now = jiffies;
 	struct bnxt *bp = ptp->bp;
 	u16 cons = ptp->txts_cons;
-	u8 num_requests;
+	u32 num_requests;
 	int rc = 0;
 
-	num_requests = BNXT_MAX_TX_TS - atomic_read(&ptp->tx_avail);
+	num_requests = BNXT_MAX_TX_TS - READ_ONCE(ptp->tx_avail);
 	while (num_requests--) {
 		if (IS_ERR(ptp->txts_req[cons].tx_skb))
 			goto next_slot;
@@ -743,7 +743,7 @@ static long bnxt_ptp_ts_aux_work(struct ptp_clock_info *ptp_info)
 		if (rc == -EAGAIN)
 			break;
 next_slot:
-		atomic_inc(&ptp->tx_avail);
+		BNXT_PTP_INC_TX_AVAIL(ptp);
 		cons = NEXT_TXTS(cons);
 	}
 	ptp->txts_cons = cons;
@@ -767,6 +767,21 @@ next_slot:
 	return HZ;
 }
 
+int bnxt_ptp_get_txts_prod(struct bnxt_ptp_cfg *ptp, u16 *prod)
+{
+	spin_lock_bh(&ptp->ptp_tx_lock);
+	if (ptp->tx_avail) {
+		*prod = ptp->txts_prod;
+		ptp->txts_prod = NEXT_TXTS(*prod);
+		ptp->tx_avail--;
+		spin_unlock_bh(&ptp->ptp_tx_lock);
+		return 0;
+	}
+	spin_unlock_bh(&ptp->ptp_tx_lock);
+	atomic64_inc(&ptp->stats.ts_err);
+	return -ENOSPC;
+}
+
 void bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb, u16 prod)
 {
 	struct bnxt_ptp_cfg *ptp = bp->ptp_cfg;
@@ -1014,7 +1029,7 @@ int bnxt_ptp_init(struct bnxt *bp, bool phc_cfg)
 
 	bnxt_ptp_free(bp);
 
-	atomic_set(&ptp->tx_avail, BNXT_MAX_TX_TS);
+	WRITE_ONCE(ptp->tx_avail, BNXT_MAX_TX_TS);
 	spin_lock_init(&ptp->ptp_lock);
 	spin_lock_init(&ptp->ptp_tx_lock);
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
index a1910ce86cbb..a9a2f9a18c9c 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ptp.h
@@ -116,7 +116,7 @@ struct bnxt_ptp_cfg {
 	struct bnxt_ptp_tx_req	txts_req[BNXT_MAX_TX_TS];
 
 	struct bnxt		*bp;
-	atomic_t		tx_avail;
+	u32			tx_avail;
 	u16			rxctl;
 #define BNXT_PTP_MSG_SYNC			(1 << 0)
 #define BNXT_PTP_MSG_DELAY_REQ			(1 << 1)
@@ -157,6 +157,13 @@ do {						\
 	((dst) = READ_ONCE(src))
 #endif
 
+#define BNXT_PTP_INC_TX_AVAIL(ptp)		\
+do {						\
+	spin_lock_bh(&(ptp)->ptp_tx_lock);	\
+	(ptp)->tx_avail++;			\
+	spin_unlock_bh(&(ptp)->ptp_tx_lock);	\
+} while (0)
+
 int bnxt_ptp_parse(struct sk_buff *skb, u16 *seq_id, u16 *hdr_off);
 void bnxt_ptp_update_current_time(struct bnxt *bp);
 void bnxt_ptp_pps_event(struct bnxt *bp, u32 data1, u32 data2);
@@ -164,6 +171,7 @@ int bnxt_ptp_cfg_tstamp_filters(struct bnxt *bp);
 void bnxt_ptp_reapply_pps(struct bnxt *bp);
 int bnxt_hwtstamp_set(struct net_device *dev, struct ifreq *ifr);
 int bnxt_hwtstamp_get(struct net_device *dev, struct ifreq *ifr);
+int bnxt_ptp_get_txts_prod(struct bnxt_ptp_cfg *ptp, u16 *prod);
 void bnxt_get_tx_ts_p5(struct bnxt *bp, struct sk_buff *skb, u16 prod);
 int bnxt_get_rx_ts_p5(struct bnxt *bp, u64 *ts, u32 pkt_ts);
 void bnxt_tx_ts_cmp(struct bnxt *bp, struct bnxt_napi *bnapi,
-- 
cgit v1.2.3-58-ga151


From 2d5317753e5f02a66e6d0afb9b25105d0beab1be Mon Sep 17 00:00:00 2001
From: Steffen Klassert <steffen.klassert@secunet.com>
Date: Fri, 28 Jun 2024 10:46:25 +0200
Subject: xfrm: Export symbol xfrm_dev_state_delete.

This fixes a build failure if xfrm_user is build as a module.

Fixes: 07b87f9eea0c ("xfrm: Fix unregister netdevice hang on hardware offload.")
Reported-by: Mark Brown <broonie@kernel.org>
Tested-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/xfrm/xfrm_state.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index d531d2a1fae2..936f9348e5f6 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -698,6 +698,7 @@ void xfrm_dev_state_delete(struct xfrm_state *x)
 		spin_unlock_bh(&xfrm_state_dev_gc_lock);
 	}
 }
+EXPORT_SYMBOL_GPL(xfrm_dev_state_delete);
 
 void xfrm_dev_state_free(struct xfrm_state *x)
 {
-- 
cgit v1.2.3-58-ga151


From 74d6529b78f7a440a10aa7f4904ca9f27d1d2f3c Mon Sep 17 00:00:00 2001
From: Heng Qi <hengqi@linux.alibaba.com>
Date: Fri, 28 Jun 2024 12:40:18 +0800
Subject: net: ethtool: Fix the panic caused by dev being null when dumping
 coalesce

syzbot reported a general protection fault caused by a null pointer
dereference in coalesce_fill_reply(). The issue occurs when req_base->dev
is null, leading to an invalid memory access.

This panic occurs if dumping coalesce when no device name is specified.

Fixes: f750dfe825b9 ("ethtool: provide customized dim profile management")
Reported-by: syzbot+e77327e34cdc8c36b7d3@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=e77327e34cdc8c36b7d3
Signed-off-by: Heng Qi <hengqi@linux.alibaba.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethtool/coalesce.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/net/ethtool/coalesce.c b/net/ethtool/coalesce.c
index 759b16e3d134..3e18ca1ccc5e 100644
--- a/net/ethtool/coalesce.c
+++ b/net/ethtool/coalesce.c
@@ -211,9 +211,9 @@ static int coalesce_fill_reply(struct sk_buff *skb,
 {
 	const struct coalesce_reply_data *data = COALESCE_REPDATA(reply_base);
 	const struct kernel_ethtool_coalesce *kcoal = &data->kernel_coalesce;
-	struct dim_irq_moder *moder = req_base->dev->irq_moder;
 	const struct ethtool_coalesce *coal = &data->coalesce;
 	u32 supported = data->supported_params;
+	struct dim_irq_moder *moder;
 	int ret = 0;
 
 	if (coalesce_put_u32(skb, ETHTOOL_A_COALESCE_RX_USECS,
@@ -272,9 +272,10 @@ static int coalesce_fill_reply(struct sk_buff *skb,
 			     kcoal->tx_aggr_time_usecs, supported))
 		return -EMSGSIZE;
 
-	if (!moder)
+	if (!req_base->dev || !req_base->dev->irq_moder)
 		return 0;
 
+	moder = req_base->dev->irq_moder;
 	rcu_read_lock();
 	if (moder->profile_flags & DIM_PROFILE_RX) {
 		ret = coalesce_put_profile(skb, ETHTOOL_A_COALESCE_RX_PROFILE,
-- 
cgit v1.2.3-58-ga151


From 89cc8f1c5f22568142b7ad118c738204708e4207 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Sun, 30 Jun 2024 00:26:48 +0200
Subject: netfilter: nf_tables: Add flowtable map for xdp offload

This adds a small internal mapping table so that a new bpf (xdp) kfunc
can perform lookups in a flowtable.

As-is, xdp program has access to the device pointer, but no way to do a
lookup in a flowtable -- there is no way to obtain the needed struct
without questionable stunts.

This allows to obtain an nf_flowtable pointer given a net_device
structure.

In order to keep backward compatibility, the infrastructure allows the
user to add a given device to multiple flowtables, but it will always
return the first added mapping performing the lookup since it assumes
the right configuration is 1:1 mapping between flowtables and net_devices.

Co-developed-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Pablo Neira Ayuso <pablo@netfilter.org>
Link: https://lore.kernel.org/bpf/9f20e2c36f494b3bf177328718367f636bb0b2ab.1719698275.git.lorenzo@kernel.org
---
 include/net/netfilter/nf_flow_table.h |   5 ++
 net/netfilter/Makefile                |   2 +-
 net/netfilter/nf_flow_table_offload.c |   2 +-
 net/netfilter/nf_flow_table_xdp.c     | 147 ++++++++++++++++++++++++++++++++++
 4 files changed, 154 insertions(+), 2 deletions(-)
 create mode 100644 net/netfilter/nf_flow_table_xdp.c

diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h
index 9abb7ee40d72..d845745207d2 100644
--- a/include/net/netfilter/nf_flow_table.h
+++ b/include/net/netfilter/nf_flow_table.h
@@ -305,6 +305,11 @@ struct flow_ports {
 	__be16 source, dest;
 };
 
+struct nf_flowtable *nf_flowtable_by_dev(const struct net_device *dev);
+int nf_flow_offload_xdp_setup(struct nf_flowtable *flowtable,
+			      struct net_device *dev,
+			      enum flow_block_command cmd);
+
 unsigned int nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb,
 				     const struct nf_hook_state *state);
 unsigned int nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb,
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 614815a3ed73..18046872a38a 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -142,7 +142,7 @@ obj-$(CONFIG_NFT_FWD_NETDEV)	+= nft_fwd_netdev.o
 # flow table infrastructure
 obj-$(CONFIG_NF_FLOW_TABLE)	+= nf_flow_table.o
 nf_flow_table-objs		:= nf_flow_table_core.o nf_flow_table_ip.o \
-				   nf_flow_table_offload.o
+				   nf_flow_table_offload.o nf_flow_table_xdp.o
 nf_flow_table-$(CONFIG_NF_FLOW_TABLE_PROCFS) += nf_flow_table_procfs.o
 
 obj-$(CONFIG_NF_FLOW_TABLE_INET) += nf_flow_table_inet.o
diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
index a010b25076ca..ff1a4e36c2b5 100644
--- a/net/netfilter/nf_flow_table_offload.c
+++ b/net/netfilter/nf_flow_table_offload.c
@@ -1192,7 +1192,7 @@ int nf_flow_table_offload_setup(struct nf_flowtable *flowtable,
 	int err;
 
 	if (!nf_flowtable_hw_offload(flowtable))
-		return 0;
+		return nf_flow_offload_xdp_setup(flowtable, dev, cmd);
 
 	if (dev->netdev_ops->ndo_setup_tc)
 		err = nf_flow_table_offload_cmd(&bo, flowtable, dev, cmd,
diff --git a/net/netfilter/nf_flow_table_xdp.c b/net/netfilter/nf_flow_table_xdp.c
new file mode 100644
index 000000000000..e1252d042699
--- /dev/null
+++ b/net/netfilter/nf_flow_table_xdp.c
@@ -0,0 +1,147 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/netfilter.h>
+#include <linux/rhashtable.h>
+#include <linux/netdevice.h>
+#include <net/flow_offload.h>
+#include <net/netfilter/nf_flow_table.h>
+
+struct flow_offload_xdp_ft {
+	struct list_head head;
+	struct nf_flowtable *ft;
+	struct rcu_head rcuhead;
+};
+
+struct flow_offload_xdp {
+	struct hlist_node hnode;
+	unsigned long net_device_addr;
+	struct list_head head;
+};
+
+#define NF_XDP_HT_BITS	4
+static DEFINE_HASHTABLE(nf_xdp_hashtable, NF_XDP_HT_BITS);
+static DEFINE_MUTEX(nf_xdp_hashtable_lock);
+
+/* caller must hold rcu read lock */
+struct nf_flowtable *nf_flowtable_by_dev(const struct net_device *dev)
+{
+	unsigned long key = (unsigned long)dev;
+	struct flow_offload_xdp *iter;
+
+	hash_for_each_possible_rcu(nf_xdp_hashtable, iter, hnode, key) {
+		if (key == iter->net_device_addr) {
+			struct flow_offload_xdp_ft *ft_elem;
+
+			/* The user is supposed to insert a given net_device
+			 * just into a single nf_flowtable so we always return
+			 * the first element here.
+			 */
+			ft_elem = list_first_or_null_rcu(&iter->head,
+							 struct flow_offload_xdp_ft,
+							 head);
+			return ft_elem ? ft_elem->ft : NULL;
+		}
+	}
+
+	return NULL;
+}
+
+static int nf_flowtable_by_dev_insert(struct nf_flowtable *ft,
+				      const struct net_device *dev)
+{
+	struct flow_offload_xdp *iter, *elem = NULL;
+	unsigned long key = (unsigned long)dev;
+	struct flow_offload_xdp_ft *ft_elem;
+
+	ft_elem = kzalloc(sizeof(*ft_elem), GFP_KERNEL_ACCOUNT);
+	if (!ft_elem)
+		return -ENOMEM;
+
+	ft_elem->ft = ft;
+
+	mutex_lock(&nf_xdp_hashtable_lock);
+
+	hash_for_each_possible(nf_xdp_hashtable, iter, hnode, key) {
+		if (key == iter->net_device_addr) {
+			elem = iter;
+			break;
+		}
+	}
+
+	if (!elem) {
+		elem = kzalloc(sizeof(*elem), GFP_KERNEL_ACCOUNT);
+		if (!elem)
+			goto err_unlock;
+
+		elem->net_device_addr = key;
+		INIT_LIST_HEAD(&elem->head);
+		hash_add_rcu(nf_xdp_hashtable, &elem->hnode, key);
+	}
+	list_add_tail_rcu(&ft_elem->head, &elem->head);
+
+	mutex_unlock(&nf_xdp_hashtable_lock);
+
+	return 0;
+
+err_unlock:
+	mutex_unlock(&nf_xdp_hashtable_lock);
+	kfree(ft_elem);
+
+	return -ENOMEM;
+}
+
+static void nf_flowtable_by_dev_remove(struct nf_flowtable *ft,
+				       const struct net_device *dev)
+{
+	struct flow_offload_xdp *iter, *elem = NULL;
+	unsigned long key = (unsigned long)dev;
+
+	mutex_lock(&nf_xdp_hashtable_lock);
+
+	hash_for_each_possible(nf_xdp_hashtable, iter, hnode, key) {
+		if (key == iter->net_device_addr) {
+			elem = iter;
+			break;
+		}
+	}
+
+	if (elem) {
+		struct flow_offload_xdp_ft *ft_elem, *ft_next;
+
+		list_for_each_entry_safe(ft_elem, ft_next, &elem->head, head) {
+			if (ft_elem->ft == ft) {
+				list_del_rcu(&ft_elem->head);
+				kfree_rcu(ft_elem, rcuhead);
+			}
+		}
+
+		if (list_empty(&elem->head))
+			hash_del_rcu(&elem->hnode);
+		else
+			elem = NULL;
+	}
+
+	mutex_unlock(&nf_xdp_hashtable_lock);
+
+	if (elem) {
+		synchronize_rcu();
+		kfree(elem);
+	}
+}
+
+int nf_flow_offload_xdp_setup(struct nf_flowtable *flowtable,
+			      struct net_device *dev,
+			      enum flow_block_command cmd)
+{
+	switch (cmd) {
+	case FLOW_BLOCK_BIND:
+		return nf_flowtable_by_dev_insert(flowtable, dev);
+	case FLOW_BLOCK_UNBIND:
+		nf_flowtable_by_dev_remove(flowtable, dev);
+		return 0;
+	}
+
+	WARN_ON_ONCE(1);
+	return 0;
+}
-- 
cgit v1.2.3-58-ga151


From 391bb6594fd3a567efb1cd3efc8136c78c4c9e31 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Sun, 30 Jun 2024 00:26:49 +0200
Subject: netfilter: Add bpf_xdp_flow_lookup kfunc

Introduce bpf_xdp_flow_lookup kfunc in order to perform the lookup
of a given flowtable entry based on a fib tuple of incoming traffic.
bpf_xdp_flow_lookup can be used as building block to offload in xdp
the processing of sw flowtable when hw flowtable is not available.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Acked-by: Pablo Neira Ayuso <pablo@netfilter.org>
Link: https://lore.kernel.org/bpf/55d38a4e5856f6d1509d823ff4e98aaa6d356097.1719698275.git.lorenzo@kernel.org
---
 include/net/netfilter/nf_flow_table.h |  10 +++
 net/netfilter/Makefile                |   5 ++
 net/netfilter/nf_flow_table_bpf.c     | 121 ++++++++++++++++++++++++++++++++++
 net/netfilter/nf_flow_table_inet.c    |   2 +-
 4 files changed, 137 insertions(+), 1 deletion(-)
 create mode 100644 net/netfilter/nf_flow_table_bpf.c

diff --git a/include/net/netfilter/nf_flow_table.h b/include/net/netfilter/nf_flow_table.h
index d845745207d2..b63d53bb9dd6 100644
--- a/include/net/netfilter/nf_flow_table.h
+++ b/include/net/netfilter/nf_flow_table.h
@@ -315,6 +315,16 @@ unsigned int nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb,
 unsigned int nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb,
 				       const struct nf_hook_state *state);
 
+#if (IS_BUILTIN(CONFIG_NF_FLOW_TABLE) && IS_ENABLED(CONFIG_DEBUG_INFO_BTF)) || \
+    (IS_MODULE(CONFIG_NF_FLOW_TABLE) && IS_ENABLED(CONFIG_DEBUG_INFO_BTF_MODULES))
+extern int nf_flow_register_bpf(void);
+#else
+static inline int nf_flow_register_bpf(void)
+{
+	return 0;
+}
+#endif
+
 #define MODULE_ALIAS_NF_FLOWTABLE(family)	\
 	MODULE_ALIAS("nf-flowtable-" __stringify(family))
 
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 18046872a38a..f0aa4d7ef499 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -144,6 +144,11 @@ obj-$(CONFIG_NF_FLOW_TABLE)	+= nf_flow_table.o
 nf_flow_table-objs		:= nf_flow_table_core.o nf_flow_table_ip.o \
 				   nf_flow_table_offload.o nf_flow_table_xdp.o
 nf_flow_table-$(CONFIG_NF_FLOW_TABLE_PROCFS) += nf_flow_table_procfs.o
+ifeq ($(CONFIG_NF_FLOW_TABLE),m)
+nf_flow_table-$(CONFIG_DEBUG_INFO_BTF_MODULES) += nf_flow_table_bpf.o
+else ifeq ($(CONFIG_NF_FLOW_TABLE),y)
+nf_flow_table-$(CONFIG_DEBUG_INFO_BTF) += nf_flow_table_bpf.o
+endif
 
 obj-$(CONFIG_NF_FLOW_TABLE_INET) += nf_flow_table_inet.o
 
diff --git a/net/netfilter/nf_flow_table_bpf.c b/net/netfilter/nf_flow_table_bpf.c
new file mode 100644
index 000000000000..4a5f5195f2d2
--- /dev/null
+++ b/net/netfilter/nf_flow_table_bpf.c
@@ -0,0 +1,121 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Unstable Flow Table Helpers for XDP hook
+ *
+ * These are called from the XDP programs.
+ * Note that it is allowed to break compatibility for these functions since
+ * the interface they are exposed through to BPF programs is explicitly
+ * unstable.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <net/netfilter/nf_flow_table.h>
+#include <linux/bpf.h>
+#include <linux/btf.h>
+#include <net/xdp.h>
+
+/* bpf_flowtable_opts - options for bpf flowtable helpers
+ * @error: out parameter, set for any encountered error
+ */
+struct bpf_flowtable_opts {
+	s32 error;
+};
+
+enum {
+	NF_BPF_FLOWTABLE_OPTS_SZ = 4,
+};
+
+__diag_push();
+__diag_ignore_all("-Wmissing-prototypes",
+		  "Global functions as their definitions will be in nf_flow_table BTF");
+
+__bpf_kfunc_start_defs();
+
+static struct flow_offload_tuple_rhash *
+bpf_xdp_flow_tuple_lookup(struct net_device *dev,
+			  struct flow_offload_tuple *tuple, __be16 proto)
+{
+	struct flow_offload_tuple_rhash *tuplehash;
+	struct nf_flowtable *nf_flow_table;
+	struct flow_offload *nf_flow;
+
+	nf_flow_table = nf_flowtable_by_dev(dev);
+	if (!nf_flow_table)
+		return ERR_PTR(-ENOENT);
+
+	tuplehash = flow_offload_lookup(nf_flow_table, tuple);
+	if (!tuplehash)
+		return ERR_PTR(-ENOENT);
+
+	nf_flow = container_of(tuplehash, struct flow_offload,
+			       tuplehash[tuplehash->tuple.dir]);
+	flow_offload_refresh(nf_flow_table, nf_flow, false);
+
+	return tuplehash;
+}
+
+__bpf_kfunc struct flow_offload_tuple_rhash *
+bpf_xdp_flow_lookup(struct xdp_md *ctx, struct bpf_fib_lookup *fib_tuple,
+		    struct bpf_flowtable_opts *opts, u32 opts_len)
+{
+	struct xdp_buff *xdp = (struct xdp_buff *)ctx;
+	struct flow_offload_tuple tuple = {
+		.iifidx = fib_tuple->ifindex,
+		.l3proto = fib_tuple->family,
+		.l4proto = fib_tuple->l4_protocol,
+		.src_port = fib_tuple->sport,
+		.dst_port = fib_tuple->dport,
+	};
+	struct flow_offload_tuple_rhash *tuplehash;
+	__be16 proto;
+
+	if (opts_len != NF_BPF_FLOWTABLE_OPTS_SZ) {
+		opts->error = -EINVAL;
+		return NULL;
+	}
+
+	switch (fib_tuple->family) {
+	case AF_INET:
+		tuple.src_v4.s_addr = fib_tuple->ipv4_src;
+		tuple.dst_v4.s_addr = fib_tuple->ipv4_dst;
+		proto = htons(ETH_P_IP);
+		break;
+	case AF_INET6:
+		tuple.src_v6 = *(struct in6_addr *)&fib_tuple->ipv6_src;
+		tuple.dst_v6 = *(struct in6_addr *)&fib_tuple->ipv6_dst;
+		proto = htons(ETH_P_IPV6);
+		break;
+	default:
+		opts->error = -EAFNOSUPPORT;
+		return NULL;
+	}
+
+	tuplehash = bpf_xdp_flow_tuple_lookup(xdp->rxq->dev, &tuple, proto);
+	if (IS_ERR(tuplehash)) {
+		opts->error = PTR_ERR(tuplehash);
+		return NULL;
+	}
+
+	return tuplehash;
+}
+
+__diag_pop()
+
+__bpf_kfunc_end_defs();
+
+BTF_KFUNCS_START(nf_ft_kfunc_set)
+BTF_ID_FLAGS(func, bpf_xdp_flow_lookup, KF_TRUSTED_ARGS | KF_RET_NULL)
+BTF_KFUNCS_END(nf_ft_kfunc_set)
+
+static const struct btf_kfunc_id_set nf_flow_kfunc_set = {
+	.owner = THIS_MODULE,
+	.set   = &nf_ft_kfunc_set,
+};
+
+int nf_flow_register_bpf(void)
+{
+	return register_btf_kfunc_id_set(BPF_PROG_TYPE_XDP,
+					 &nf_flow_kfunc_set);
+}
+EXPORT_SYMBOL_GPL(nf_flow_register_bpf);
diff --git a/net/netfilter/nf_flow_table_inet.c b/net/netfilter/nf_flow_table_inet.c
index 6eef15648b7b..88787b45e30d 100644
--- a/net/netfilter/nf_flow_table_inet.c
+++ b/net/netfilter/nf_flow_table_inet.c
@@ -98,7 +98,7 @@ static int __init nf_flow_inet_module_init(void)
 	nft_register_flowtable_type(&flowtable_ipv6);
 	nft_register_flowtable_type(&flowtable_inet);
 
-	return 0;
+	return nf_flow_register_bpf();
 }
 
 static void __exit nf_flow_inet_module_exit(void)
-- 
cgit v1.2.3-58-ga151


From c77e572d3a8c0e21c5dca4cc2883c7cd8cbe981f Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Sun, 30 Jun 2024 00:26:50 +0200
Subject: selftests/bpf: Add selftest for bpf_xdp_flow_lookup kfunc

Introduce e2e selftest for bpf_xdp_flow_lookup kfunc through
xdp_flowtable utility.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/b74393fb4539aecbbd5ac7883605f86a95fb0b6b.1719698275.git.lorenzo@kernel.org
---
 tools/testing/selftests/bpf/config                 |  13 ++
 .../selftests/bpf/prog_tests/xdp_flowtable.c       | 168 +++++++++++++++++++++
 tools/testing/selftests/bpf/progs/xdp_flowtable.c  | 144 ++++++++++++++++++
 3 files changed, 325 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/xdp_flowtable.c
 create mode 100644 tools/testing/selftests/bpf/progs/xdp_flowtable.c

diff --git a/tools/testing/selftests/bpf/config b/tools/testing/selftests/bpf/config
index 2fb16da78dce..5291e97df749 100644
--- a/tools/testing/selftests/bpf/config
+++ b/tools/testing/selftests/bpf/config
@@ -83,6 +83,19 @@ CONFIG_NF_CONNTRACK_MARK=y
 CONFIG_NF_CONNTRACK_ZONES=y
 CONFIG_NF_DEFRAG_IPV4=y
 CONFIG_NF_DEFRAG_IPV6=y
+CONFIG_NF_TABLES=y
+CONFIG_NF_TABLES_INET=y
+CONFIG_NF_TABLES_NETDEV=y
+CONFIG_NF_TABLES_IPV4=y
+CONFIG_NF_TABLES_IPV6=y
+CONFIG_NETFILTER_INGRESS=y
+CONFIG_NF_FLOW_TABLE=y
+CONFIG_NF_FLOW_TABLE_INET=y
+CONFIG_NETFILTER_NETLINK=y
+CONFIG_NFT_FLOW_OFFLOAD=y
+CONFIG_IP_NF_IPTABLES=y
+CONFIG_IP6_NF_IPTABLES=y
+CONFIG_IP6_NF_FILTER=y
 CONFIG_NF_NAT=y
 CONFIG_RC_CORE=y
 CONFIG_SECURITY=y
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_flowtable.c b/tools/testing/selftests/bpf/prog_tests/xdp_flowtable.c
new file mode 100644
index 000000000000..e1bf141d3401
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_flowtable.c
@@ -0,0 +1,168 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <test_progs.h>
+#include <network_helpers.h>
+#include <bpf/btf.h>
+#include <linux/if_link.h>
+#include <linux/udp.h>
+#include <net/if.h>
+#include <unistd.h>
+
+#include "xdp_flowtable.skel.h"
+
+#define TX_NETNS_NAME	"ns0"
+#define RX_NETNS_NAME	"ns1"
+
+#define TX_NAME		"v0"
+#define FORWARD_NAME	"v1"
+#define RX_NAME		"d0"
+
+#define TX_MAC		"00:00:00:00:00:01"
+#define FORWARD_MAC	"00:00:00:00:00:02"
+#define RX_MAC		"00:00:00:00:00:03"
+#define DST_MAC		"00:00:00:00:00:04"
+
+#define TX_ADDR		"10.0.0.1"
+#define FORWARD_ADDR	"10.0.0.2"
+#define RX_ADDR		"20.0.0.1"
+#define DST_ADDR	"20.0.0.2"
+
+#define PREFIX_LEN	"8"
+#define N_PACKETS	10
+#define UDP_PORT	12345
+#define UDP_PORT_STR	"12345"
+
+static int send_udp_traffic(void)
+{
+	struct sockaddr_storage addr;
+	int i, sock;
+
+	if (make_sockaddr(AF_INET, DST_ADDR, UDP_PORT, &addr, NULL))
+		return -EINVAL;
+
+	sock = socket(AF_INET, SOCK_DGRAM, 0);
+	if (sock < 0)
+		return sock;
+
+	for (i = 0; i < N_PACKETS; i++) {
+		unsigned char buf[] = { 0xaa, 0xbb, 0xcc };
+		int n;
+
+		n = sendto(sock, buf, sizeof(buf), MSG_NOSIGNAL | MSG_CONFIRM,
+			   (struct sockaddr *)&addr, sizeof(addr));
+		if (n != sizeof(buf)) {
+			close(sock);
+			return -EINVAL;
+		}
+
+		usleep(50000); /* 50ms */
+	}
+	close(sock);
+
+	return 0;
+}
+
+void test_xdp_flowtable(void)
+{
+	struct xdp_flowtable *skel = NULL;
+	struct nstoken *tok = NULL;
+	int iifindex, stats_fd;
+	__u32 value, key = 0;
+	struct bpf_link *link;
+
+	if (SYS_NOFAIL("nft -v")) {
+		fprintf(stdout, "Missing required nft tool\n");
+		test__skip();
+		return;
+	}
+
+	SYS(out, "ip netns add " TX_NETNS_NAME);
+	SYS(out, "ip netns add " RX_NETNS_NAME);
+
+	tok = open_netns(RX_NETNS_NAME);
+	if (!ASSERT_OK_PTR(tok, "setns"))
+		goto out;
+
+	SYS(out, "sysctl -qw net.ipv4.conf.all.forwarding=1");
+
+	SYS(out, "ip link add " TX_NAME " type veth peer " FORWARD_NAME);
+	SYS(out, "ip link set " TX_NAME " netns " TX_NETNS_NAME);
+	SYS(out, "ip link set dev " FORWARD_NAME " address " FORWARD_MAC);
+	SYS(out,
+	    "ip addr add " FORWARD_ADDR "/" PREFIX_LEN " dev " FORWARD_NAME);
+	SYS(out, "ip link set dev " FORWARD_NAME " up");
+
+	SYS(out, "ip link add " RX_NAME " type dummy");
+	SYS(out, "ip link set dev " RX_NAME " address " RX_MAC);
+	SYS(out, "ip addr add " RX_ADDR "/" PREFIX_LEN " dev " RX_NAME);
+	SYS(out, "ip link set dev " RX_NAME " up");
+
+	/* configure the flowtable */
+	SYS(out, "nft add table ip filter");
+	SYS(out,
+	    "nft add flowtable ip filter f { hook ingress priority 0\\; "
+	    "devices = { " FORWARD_NAME ", " RX_NAME " }\\; }");
+	SYS(out,
+	    "nft add chain ip filter forward "
+	    "{ type filter hook forward priority 0\\; }");
+	SYS(out,
+	    "nft add rule ip filter forward ip protocol udp th dport "
+	    UDP_PORT_STR " flow add @f");
+
+	/* Avoid ARP calls */
+	SYS(out,
+	    "ip -4 neigh add " DST_ADDR " lladdr " DST_MAC " dev " RX_NAME);
+
+	close_netns(tok);
+	tok = open_netns(TX_NETNS_NAME);
+	if (!ASSERT_OK_PTR(tok, "setns"))
+		goto out;
+
+	SYS(out, "ip addr add " TX_ADDR "/" PREFIX_LEN " dev " TX_NAME);
+	SYS(out, "ip link set dev " TX_NAME " address " TX_MAC);
+	SYS(out, "ip link set dev " TX_NAME " up");
+	SYS(out, "ip route add default via " FORWARD_ADDR);
+
+	close_netns(tok);
+	tok = open_netns(RX_NETNS_NAME);
+	if (!ASSERT_OK_PTR(tok, "setns"))
+		goto out;
+
+	iifindex = if_nametoindex(FORWARD_NAME);
+	if (!ASSERT_NEQ(iifindex, 0, "iifindex"))
+		goto out;
+
+	skel = xdp_flowtable__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "skel"))
+		goto out;
+
+	link = bpf_program__attach_xdp(skel->progs.xdp_flowtable_do_lookup,
+				       iifindex);
+	if (!ASSERT_OK_PTR(link, "prog_attach"))
+		goto out;
+
+	close_netns(tok);
+	tok = open_netns(TX_NETNS_NAME);
+	if (!ASSERT_OK_PTR(tok, "setns"))
+		goto out;
+
+	if (!ASSERT_OK(send_udp_traffic(), "send udp"))
+		goto out;
+
+	close_netns(tok);
+	tok = open_netns(RX_NETNS_NAME);
+	if (!ASSERT_OK_PTR(tok, "setns"))
+		goto out;
+
+	stats_fd = bpf_map__fd(skel->maps.stats);
+	if (!ASSERT_OK(bpf_map_lookup_elem(stats_fd, &key, &value),
+		       "bpf_map_update_elem stats"))
+		goto out;
+
+	ASSERT_GE(value, N_PACKETS - 2, "bpf_xdp_flow_lookup failed");
+out:
+	xdp_flowtable__destroy(skel);
+	if (tok)
+		close_netns(tok);
+	SYS_NOFAIL("ip netns del " TX_NETNS_NAME);
+	SYS_NOFAIL("ip netns del " RX_NETNS_NAME);
+}
diff --git a/tools/testing/selftests/bpf/progs/xdp_flowtable.c b/tools/testing/selftests/bpf/progs/xdp_flowtable.c
new file mode 100644
index 000000000000..15209650f73b
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/xdp_flowtable.c
@@ -0,0 +1,144 @@
+// SPDX-License-Identifier: GPL-2.0
+#define BPF_NO_KFUNC_PROTOTYPES
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
+
+#define ETH_P_IP	0x0800
+#define ETH_P_IPV6	0x86dd
+#define IP_MF		0x2000	/* "More Fragments" */
+#define IP_OFFSET	0x1fff	/* "Fragment Offset" */
+#define AF_INET		2
+#define AF_INET6	10
+
+struct bpf_flowtable_opts___local {
+	s32 error;
+};
+
+struct flow_offload_tuple_rhash *
+bpf_xdp_flow_lookup(struct xdp_md *, struct bpf_fib_lookup *,
+		    struct bpf_flowtable_opts___local *, u32) __ksym;
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__type(key, __u32);
+	__type(value, __u32);
+	__uint(max_entries, 1);
+} stats SEC(".maps");
+
+static bool xdp_flowtable_offload_check_iphdr(struct iphdr *iph)
+{
+	/* ip fragmented traffic */
+	if (iph->frag_off & bpf_htons(IP_MF | IP_OFFSET))
+		return false;
+
+	/* ip options */
+	if (iph->ihl * 4 != sizeof(*iph))
+		return false;
+
+	if (iph->ttl <= 1)
+		return false;
+
+	return true;
+}
+
+static bool xdp_flowtable_offload_check_tcp_state(void *ports, void *data_end,
+						  u8 proto)
+{
+	if (proto == IPPROTO_TCP) {
+		struct tcphdr *tcph = ports;
+
+		if (tcph + 1 > data_end)
+			return false;
+
+		if (tcph->fin || tcph->rst)
+			return false;
+	}
+
+	return true;
+}
+
+SEC("xdp.frags")
+int xdp_flowtable_do_lookup(struct xdp_md *ctx)
+{
+	void *data_end = (void *)(long)ctx->data_end;
+	struct bpf_flowtable_opts___local opts = {};
+	struct flow_offload_tuple_rhash *tuplehash;
+	struct bpf_fib_lookup tuple = {
+		.ifindex = ctx->ingress_ifindex,
+	};
+	void *data = (void *)(long)ctx->data;
+	struct ethhdr *eth = data;
+	struct flow_ports *ports;
+	__u32 *val, key = 0;
+
+	if (eth + 1 > data_end)
+		return XDP_DROP;
+
+	switch (eth->h_proto) {
+	case bpf_htons(ETH_P_IP): {
+		struct iphdr *iph = data + sizeof(*eth);
+
+		ports = (struct flow_ports *)(iph + 1);
+		if (ports + 1 > data_end)
+			return XDP_PASS;
+
+		/* sanity check on ip header */
+		if (!xdp_flowtable_offload_check_iphdr(iph))
+			return XDP_PASS;
+
+		if (!xdp_flowtable_offload_check_tcp_state(ports, data_end,
+							   iph->protocol))
+			return XDP_PASS;
+
+		tuple.family		= AF_INET;
+		tuple.tos		= iph->tos;
+		tuple.l4_protocol	= iph->protocol;
+		tuple.tot_len		= bpf_ntohs(iph->tot_len);
+		tuple.ipv4_src		= iph->saddr;
+		tuple.ipv4_dst		= iph->daddr;
+		tuple.sport		= ports->source;
+		tuple.dport		= ports->dest;
+		break;
+	}
+	case bpf_htons(ETH_P_IPV6): {
+		struct in6_addr *src = (struct in6_addr *)tuple.ipv6_src;
+		struct in6_addr *dst = (struct in6_addr *)tuple.ipv6_dst;
+		struct ipv6hdr *ip6h = data + sizeof(*eth);
+
+		ports = (struct flow_ports *)(ip6h + 1);
+		if (ports + 1 > data_end)
+			return XDP_PASS;
+
+		if (ip6h->hop_limit <= 1)
+			return XDP_PASS;
+
+		if (!xdp_flowtable_offload_check_tcp_state(ports, data_end,
+							   ip6h->nexthdr))
+			return XDP_PASS;
+
+		tuple.family		= AF_INET6;
+		tuple.l4_protocol	= ip6h->nexthdr;
+		tuple.tot_len		= bpf_ntohs(ip6h->payload_len);
+		*src			= ip6h->saddr;
+		*dst			= ip6h->daddr;
+		tuple.sport		= ports->source;
+		tuple.dport		= ports->dest;
+		break;
+	}
+	default:
+		return XDP_PASS;
+	}
+
+	tuplehash = bpf_xdp_flow_lookup(ctx, &tuple, &opts, sizeof(opts));
+	if (!tuplehash)
+		return XDP_PASS;
+
+	val = bpf_map_lookup_elem(&stats, &key);
+	if (val)
+		__sync_add_and_fetch(val, 1);
+
+	return XDP_PASS;
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-58-ga151


From 5b747c23f17d791e08fdf4baa7e14b704625518c Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Sat, 29 Jun 2024 11:00:58 +0100
Subject: libbpf: Fix error handling in btf__distill_base()

Coverity points out that after calling btf__new_empty_split() the wrong
value is checked for error.

Fixes: 58e185a0dc35 ("libbpf: Add btf__distill_base() creating split BTF with distilled base BTF")
Reported-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240629100058.2866763-1-alan.maguire@oracle.com
---
 tools/lib/bpf/btf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index cd5dd6619214..32c00db3b91b 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -5431,7 +5431,7 @@ int btf__distill_base(const struct btf *src_btf, struct btf **new_base_btf,
 	 * BTF available.
 	 */
 	new_split = btf__new_empty_split(new_base);
-	if (!new_split_btf) {
+	if (!new_split) {
 		err = -errno;
 		goto done;
 	}
-- 
cgit v1.2.3-58-ga151


From d1a426171d76b2cdf3dea5d52f6266090e4aa254 Mon Sep 17 00:00:00 2001
From: Pu Lehui <pulehui@huawei.com>
Date: Sat, 22 Jun 2024 03:04:35 +0000
Subject: bpf: Use precise image size for struct_ops trampoline
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For trampoline using bpf_prog_pack, we need to generate a rw_image
buffer with size of (image_end - image). For regular trampoline, we use
the precise image size generated by arch_bpf_trampoline_size to allocate
rw_image. But for struct_ops trampoline, we allocate rw_image directly
using close to PAGE_SIZE size. We do not need to allocate for that much,
as the patch size is usually much smaller than PAGE_SIZE. Let's use
precise image size for it too.

Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Björn Töpel <bjorn@rivosinc.com> #riscv
Acked-by: Song Liu <song@kernel.org>
Link: https://lore.kernel.org/bpf/20240622030437.3973492-2-pulehui@huaweicloud.com
---
 kernel/bpf/bpf_struct_ops.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index a2cf31b14be4..0d515ec57aa5 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -573,7 +573,7 @@ int bpf_struct_ops_prepare_trampoline(struct bpf_tramp_links *tlinks,
 	}
 
 	size = arch_prepare_bpf_trampoline(NULL, image + image_off,
-					   image + PAGE_SIZE,
+					   image + image_off + size,
 					   model, flags, tlinks, stub_func);
 	if (size <= 0) {
 		if (image != *_image)
-- 
cgit v1.2.3-58-ga151


From 9f1e16fb1fc9826001c69e0551d51fbbcd2d74e9 Mon Sep 17 00:00:00 2001
From: Pu Lehui <pulehui@huawei.com>
Date: Sat, 22 Jun 2024 03:04:36 +0000
Subject: riscv, bpf: Fix out-of-bounds issue when preparing trampoline image

We get the size of the trampoline image during the dry run phase and
allocate memory based on that size. The allocated image will then be
populated with instructions during the real patch phase. But after
commit 26ef208c209a ("bpf: Use arch_bpf_trampoline_size"), the `im`
argument is inconsistent in the dry run and real patch phase. This may
cause emit_imm in RV64 to generate a different number of instructions
when generating the 'im' address, potentially causing out-of-bounds
issues. Let's emit the maximum number of instructions for the "im"
address during dry run to fix this problem.

Fixes: 26ef208c209a ("bpf: Use arch_bpf_trampoline_size")
Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240622030437.3973492-3-pulehui@huaweicloud.com
---
 arch/riscv/net/bpf_jit_comp64.c | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)

diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c
index d5cebb0b0afe..e6d690657f3e 100644
--- a/arch/riscv/net/bpf_jit_comp64.c
+++ b/arch/riscv/net/bpf_jit_comp64.c
@@ -16,6 +16,8 @@
 #include "bpf_jit.h"
 
 #define RV_FENTRY_NINSNS 2
+/* imm that allows emit_imm to emit max count insns */
+#define RV_MAX_COUNT_IMM 0x7FFF7FF7FF7FF7FF
 
 #define RV_REG_TCC RV_REG_A6
 #define RV_REG_TCC_SAVED RV_REG_S6 /* Store A6 in S6 if program do calls */
@@ -916,7 +918,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 		orig_call += RV_FENTRY_NINSNS * 4;
 
 	if (flags & BPF_TRAMP_F_CALL_ORIG) {
-		emit_imm(RV_REG_A0, (const s64)im, ctx);
+		emit_imm(RV_REG_A0, ctx->insns ? (const s64)im : RV_MAX_COUNT_IMM, ctx);
 		ret = emit_call((const u64)__bpf_tramp_enter, true, ctx);
 		if (ret)
 			return ret;
@@ -977,7 +979,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 
 	if (flags & BPF_TRAMP_F_CALL_ORIG) {
 		im->ip_epilogue = ctx->insns + ctx->ninsns;
-		emit_imm(RV_REG_A0, (const s64)im, ctx);
+		emit_imm(RV_REG_A0, ctx->insns ? (const s64)im : RV_MAX_COUNT_IMM, ctx);
 		ret = emit_call((const u64)__bpf_tramp_exit, true, ctx);
 		if (ret)
 			goto out;
@@ -1046,6 +1048,7 @@ int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image,
 {
 	int ret;
 	struct rv_jit_context ctx;
+	u32 size = image_end - image;
 
 	ctx.ninsns = 0;
 	/*
@@ -1059,11 +1062,16 @@ int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image,
 	ctx.ro_insns = image;
 	ret = __arch_prepare_bpf_trampoline(im, m, tlinks, func_addr, flags, &ctx);
 	if (ret < 0)
-		return ret;
+		goto out;
 
-	bpf_flush_icache(ctx.insns, ctx.insns + ctx.ninsns);
+	if (WARN_ON(size < ninsns_rvoff(ctx.ninsns))) {
+		ret = -E2BIG;
+		goto out;
+	}
 
-	return ninsns_rvoff(ret);
+	bpf_flush_icache(image, image_end);
+out:
+	return ret < 0 ? ret : size;
 }
 
 int bpf_jit_emit_insn(const struct bpf_insn *insn, struct rv_jit_context *ctx,
-- 
cgit v1.2.3-58-ga151


From 2382a405c581ae8f39f898055654e2000e7dd0d3 Mon Sep 17 00:00:00 2001
From: Pu Lehui <pulehui@huawei.com>
Date: Sat, 22 Jun 2024 03:04:37 +0000
Subject: riscv, bpf: Use bpf_prog_pack for RV64 bpf trampoline
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We used bpf_prog_pack to aggregate bpf programs into huge page to
relieve the iTLB pressure on the system. We can apply it to bpf
trampoline, as Song had been implemented it in core and x86 [0]. This
patch is going to use bpf_prog_pack to RV64 bpf trampoline. Since Song
and Puranjay have done a lot of work for bpf_prog_pack on RV64,
implementing this function will be easy.

Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Björn Töpel <bjorn@rivosinc.com> #riscv
Link: https://lore.kernel.org/all/20231206224054.492250-1-song@kernel.org [0]
Link: https://lore.kernel.org/bpf/20240622030437.3973492-4-pulehui@huaweicloud.com
---
 arch/riscv/net/bpf_jit_comp64.c | 43 +++++++++++++++++++++++++++--------------
 1 file changed, 29 insertions(+), 14 deletions(-)

diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c
index e6d690657f3e..351e1484205e 100644
--- a/arch/riscv/net/bpf_jit_comp64.c
+++ b/arch/riscv/net/bpf_jit_comp64.c
@@ -957,7 +957,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 			goto out;
 		emit_sd(RV_REG_FP, -retval_off, RV_REG_A0, ctx);
 		emit_sd(RV_REG_FP, -(retval_off - 8), regmap[BPF_REG_0], ctx);
-		im->ip_after_call = ctx->insns + ctx->ninsns;
+		im->ip_after_call = ctx->ro_insns + ctx->ninsns;
 		/* 2 nops reserved for auipc+jalr pair */
 		emit(rv_nop(), ctx);
 		emit(rv_nop(), ctx);
@@ -978,7 +978,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 	}
 
 	if (flags & BPF_TRAMP_F_CALL_ORIG) {
-		im->ip_epilogue = ctx->insns + ctx->ninsns;
+		im->ip_epilogue = ctx->ro_insns + ctx->ninsns;
 		emit_imm(RV_REG_A0, ctx->insns ? (const s64)im : RV_MAX_COUNT_IMM, ctx);
 		ret = emit_call((const u64)__bpf_tramp_exit, true, ctx);
 		if (ret)
@@ -1041,25 +1041,33 @@ int arch_bpf_trampoline_size(const struct btf_func_model *m, u32 flags,
 	return ret < 0 ? ret : ninsns_rvoff(ctx.ninsns);
 }
 
-int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image,
-				void *image_end, const struct btf_func_model *m,
+void *arch_alloc_bpf_trampoline(unsigned int size)
+{
+	return bpf_prog_pack_alloc(size, bpf_fill_ill_insns);
+}
+
+void arch_free_bpf_trampoline(void *image, unsigned int size)
+{
+	bpf_prog_pack_free(image, size);
+}
+
+int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *ro_image,
+				void *ro_image_end, const struct btf_func_model *m,
 				u32 flags, struct bpf_tramp_links *tlinks,
 				void *func_addr)
 {
 	int ret;
+	void *image, *res;
 	struct rv_jit_context ctx;
-	u32 size = image_end - image;
+	u32 size = ro_image_end - ro_image;
+
+	image = kvmalloc(size, GFP_KERNEL);
+	if (!image)
+		return -ENOMEM;
 
 	ctx.ninsns = 0;
-	/*
-	 * The bpf_int_jit_compile() uses a RW buffer (ctx.insns) to write the
-	 * JITed instructions and later copies it to a RX region (ctx.ro_insns).
-	 * It also uses ctx.ro_insns to calculate offsets for jumps etc. As the
-	 * trampoline image uses the same memory area for writing and execution,
-	 * both ctx.insns and ctx.ro_insns can be set to image.
-	 */
 	ctx.insns = image;
-	ctx.ro_insns = image;
+	ctx.ro_insns = ro_image;
 	ret = __arch_prepare_bpf_trampoline(im, m, tlinks, func_addr, flags, &ctx);
 	if (ret < 0)
 		goto out;
@@ -1069,8 +1077,15 @@ int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image,
 		goto out;
 	}
 
-	bpf_flush_icache(image, image_end);
+	res = bpf_arch_text_copy(ro_image, image, size);
+	if (IS_ERR(res)) {
+		ret = PTR_ERR(res);
+		goto out;
+	}
+
+	bpf_flush_icache(ro_image, ro_image_end);
 out:
+	kvfree(image);
 	return ret < 0 ? ret : size;
 }
 
-- 
cgit v1.2.3-58-ga151


From 03922e97bc305c6b2e8bc4b7cc765959ca63b05d Mon Sep 17 00:00:00 2001
From: Zhu Jun <zhujun2@cmss.chinamobile.com>
Date: Wed, 26 Jun 2024 20:19:05 -0700
Subject: selftests/bpf: Delete extra blank lines in test_sockmap

Delete extra blank lines inside of test_selftest().

Signed-off-by: Zhu Jun <zhujun2@cmss.chinamobile.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240627031905.7133-1-zhujun2@cmss.chinamobile.com
---
 tools/testing/selftests/bpf/test_sockmap.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index 9cba4ec844a5..3e02d7267de8 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -1936,7 +1936,6 @@ static void test_selftests_ktls(int cg_fd, struct sockmap_options *opt)
 
 static int test_selftest(int cg_fd, struct sockmap_options *opt)
 {
-
 	test_selftests_sockmap(cg_fd, opt);
 	test_selftests_sockhash(cg_fd, opt);
 	test_selftests_ktls(cg_fd, opt);
-- 
cgit v1.2.3-58-ga151


From 3f73c24f28b317f22df7870c25ff82f1d625c6c2 Mon Sep 17 00:00:00 2001
From: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Date: Wed, 26 Jun 2024 14:28:51 +0530
Subject: wifi: ath12k: Add support to enable debugfs_htt_stats

Create debugfs_htt_stats file when ath12k debugfs support is enabled.
Add basic ath12k_debugfs_htt_stats_register and handle htt_stats_type
file operations.

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Co-developed-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Signed-off-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240626085854.2500681-2-quic_rgnanase@quicinc.com
---
 drivers/net/wireless/ath/ath12k/Makefile           |  2 +-
 drivers/net/wireless/ath/ath12k/core.h             |  7 ++
 drivers/net/wireless/ath/ath12k/debugfs.c          |  3 +
 .../net/wireless/ath/ath12k/debugfs_htt_stats.c    | 81 ++++++++++++++++++++++
 .../net/wireless/ath/ath12k/debugfs_htt_stats.h    | 20 ++++++
 5 files changed, 112 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
 create mode 100644 drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h

diff --git a/drivers/net/wireless/ath/ath12k/Makefile b/drivers/net/wireless/ath/ath12k/Makefile
index f0dcc933aa95..5a1ed20d730e 100644
--- a/drivers/net/wireless/ath/ath12k/Makefile
+++ b/drivers/net/wireless/ath/ath12k/Makefile
@@ -23,7 +23,7 @@ ath12k-y += core.o \
 	    fw.o \
 	    p2p.o
 
-ath12k-$(CONFIG_ATH12K_DEBUGFS) += debugfs.o
+ath12k-$(CONFIG_ATH12K_DEBUGFS) += debugfs.o debugfs_htt_stats.o
 ath12k-$(CONFIG_ACPI) += acpi.o
 ath12k-$(CONFIG_ATH12K_TRACING) += trace.o
 ath12k-$(CONFIG_PM) += wow.o
diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index 9de3df924a1c..7173c7e174a7 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -29,6 +29,7 @@
 #include "fw.h"
 #include "acpi.h"
 #include "wow.h"
+#include "debugfs_htt_stats.h"
 
 #define SM(_v, _f) (((_v) << _f##_LSB) & _f##_MASK)
 
@@ -481,9 +482,15 @@ struct ath12k_fw_stats {
 	struct list_head bcn;
 };
 
+struct ath12k_dbg_htt_stats {
+	enum ath12k_dbg_htt_ext_stats_type type;
+	u32 cfg_param[4];
+};
+
 struct ath12k_debug {
 	struct dentry *debugfs_pdev;
 	struct dentry *debugfs_pdev_symlink;
+	struct ath12k_dbg_htt_stats htt_stats;
 };
 
 struct ath12k_per_peer_tx_stats {
diff --git a/drivers/net/wireless/ath/ath12k/debugfs.c b/drivers/net/wireless/ath/ath12k/debugfs.c
index 751a9b04386a..2a977c36af00 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs.c
+++ b/drivers/net/wireless/ath/ath12k/debugfs.c
@@ -6,6 +6,7 @@
 
 #include "core.h"
 #include "debugfs.h"
+#include "debugfs_htt_stats.h"
 
 static ssize_t ath12k_write_simulate_radar(struct file *file,
 					   const char __user *user_buf,
@@ -89,6 +90,8 @@ void ath12k_debugfs_register(struct ath12k *ar)
 				    ar->debug.debugfs_pdev, ar,
 				    &fops_simulate_radar);
 	}
+
+	ath12k_debugfs_htt_stats_register(ar);
 }
 
 void ath12k_debugfs_unregister(struct ath12k *ar)
diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
new file mode 100644
index 000000000000..d1d469ae6d2b
--- /dev/null
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
@@ -0,0 +1,81 @@
+// SPDX-License-Identifier: BSD-3-Clause-Clear
+/*
+ * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved.
+ * Copyright (c) 2021-2024 Qualcomm Innovation Center, Inc. All rights reserved.
+ */
+
+#include <linux/vmalloc.h>
+#include "core.h"
+#include "debug.h"
+#include "debugfs_htt_stats.h"
+
+static ssize_t ath12k_read_htt_stats_type(struct file *file,
+					  char __user *user_buf,
+					  size_t count, loff_t *ppos)
+{
+	struct ath12k *ar = file->private_data;
+	enum ath12k_dbg_htt_ext_stats_type type;
+	char buf[32];
+	size_t len;
+
+	mutex_lock(&ar->conf_mutex);
+	type = ar->debug.htt_stats.type;
+	mutex_unlock(&ar->conf_mutex);
+
+	len = scnprintf(buf, sizeof(buf), "%u\n", type);
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, len);
+}
+
+static ssize_t ath12k_write_htt_stats_type(struct file *file,
+					   const char __user *user_buf,
+					   size_t count, loff_t *ppos)
+{
+	struct ath12k *ar = file->private_data;
+	enum ath12k_dbg_htt_ext_stats_type type;
+	unsigned int cfg_param[4] = {0};
+	const int size = 32;
+	int num_args;
+
+	char *buf __free(kfree) = kzalloc(size, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	if (copy_from_user(buf, user_buf, count))
+		return -EFAULT;
+
+	num_args = sscanf(buf, "%u %u %u %u %u\n", &type, &cfg_param[0],
+			  &cfg_param[1], &cfg_param[2], &cfg_param[3]);
+	if (!num_args || num_args > 5)
+		return -EINVAL;
+
+	if (type == ATH12K_DBG_HTT_EXT_STATS_RESET ||
+	    type >= ATH12K_DBG_HTT_NUM_EXT_STATS)
+		return -EINVAL;
+
+	mutex_lock(&ar->conf_mutex);
+
+	ar->debug.htt_stats.type = type;
+	ar->debug.htt_stats.cfg_param[0] = cfg_param[0];
+	ar->debug.htt_stats.cfg_param[1] = cfg_param[1];
+	ar->debug.htt_stats.cfg_param[2] = cfg_param[2];
+	ar->debug.htt_stats.cfg_param[3] = cfg_param[3];
+
+	mutex_unlock(&ar->conf_mutex);
+
+	return count;
+}
+
+static const struct file_operations fops_htt_stats_type = {
+	.read = ath12k_read_htt_stats_type,
+	.write = ath12k_write_htt_stats_type,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+void ath12k_debugfs_htt_stats_register(struct ath12k *ar)
+{
+	debugfs_create_file("htt_stats_type", 0600, ar->debug.debugfs_pdev,
+			    ar, &fops_htt_stats_type);
+}
diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
new file mode 100644
index 000000000000..8610db89289a
--- /dev/null
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: BSD-3-Clause-Clear */
+/*
+ * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved.
+ * Copyright (c) 2021-2024 Qualcomm Innovation Center, Inc. All rights reserved.
+ */
+
+#ifndef DEBUG_HTT_STATS_H
+#define DEBUG_HTT_STATS_H
+
+void ath12k_debugfs_htt_stats_register(struct ath12k *ar);
+
+/* htt_dbg_ext_stats_type */
+enum ath12k_dbg_htt_ext_stats_type {
+	ATH12K_DBG_HTT_EXT_STATS_RESET		= 0,
+
+	/* keep this last */
+	ATH12K_DBG_HTT_NUM_EXT_STATS,
+};
+
+#endif
-- 
cgit v1.2.3-58-ga151


From ad201b35d3ec48d81da6c873b9a50586afd8defa Mon Sep 17 00:00:00 2001
From: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Date: Thu, 27 Jun 2024 19:36:50 +0300
Subject: wifi: ath12k: Add htt_stats_dump file ops support

Add dump_htt_stats file operation to dump the stats value requested
for the requested stats_type.
Stats sent from firmware will be cumulative. Hence add debugfs to reset
the requested stats type.

Example with one ath12k device:

ath12k
`-- pci-0000:06:00.0
    |-- mac0
        `-- htt_stats
        |-- htt_stats_type
        |-- htt_stats_reset

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Co-developed-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Signed-off-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240626085854.2500681-3-quic_rgnanase@quicinc.com
---
 drivers/net/wireless/ath/ath12k/core.h             |   2 +
 .../net/wireless/ath/ath12k/debugfs_htt_stats.c    | 205 +++++++++++++++++++++
 .../net/wireless/ath/ath12k/debugfs_htt_stats.h    |  30 +++
 drivers/net/wireless/ath/ath12k/dp_tx.c            |   4 +-
 4 files changed, 240 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/core.h b/drivers/net/wireless/ath/ath12k/core.h
index 7173c7e174a7..cdfd43a7321a 100644
--- a/drivers/net/wireless/ath/ath12k/core.h
+++ b/drivers/net/wireless/ath/ath12k/core.h
@@ -485,6 +485,8 @@ struct ath12k_fw_stats {
 struct ath12k_dbg_htt_stats {
 	enum ath12k_dbg_htt_ext_stats_type type;
 	u32 cfg_param[4];
+	u8 reset;
+	struct debug_htt_stats_req *stats_req;
 };
 
 struct ath12k_debug {
diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
index d1d469ae6d2b..9f44285caa58 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
@@ -8,6 +8,7 @@
 #include "core.h"
 #include "debug.h"
 #include "debugfs_htt_stats.h"
+#include "dp_tx.h"
 
 static ssize_t ath12k_read_htt_stats_type(struct file *file,
 					  char __user *user_buf,
@@ -74,8 +75,212 @@ static const struct file_operations fops_htt_stats_type = {
 	.llseek = default_llseek,
 };
 
+static int ath12k_debugfs_htt_stats_req(struct ath12k *ar)
+{
+	struct debug_htt_stats_req *stats_req = ar->debug.htt_stats.stats_req;
+	enum ath12k_dbg_htt_ext_stats_type type = stats_req->type;
+	u64 cookie;
+	int ret, pdev_id;
+	struct htt_ext_stats_cfg_params cfg_params = { 0 };
+
+	lockdep_assert_held(&ar->conf_mutex);
+
+	init_completion(&stats_req->htt_stats_rcvd);
+
+	pdev_id = ath12k_mac_get_target_pdev_id(ar);
+	stats_req->done = false;
+	stats_req->pdev_id = pdev_id;
+
+	cookie = u64_encode_bits(ATH12K_HTT_STATS_MAGIC_VALUE,
+				 ATH12K_HTT_STATS_COOKIE_MSB);
+	cookie |= u64_encode_bits(pdev_id, ATH12K_HTT_STATS_COOKIE_LSB);
+
+	if (stats_req->override_cfg_param) {
+		cfg_params.cfg0 = stats_req->cfg_param[0];
+		cfg_params.cfg1 = stats_req->cfg_param[1];
+		cfg_params.cfg2 = stats_req->cfg_param[2];
+		cfg_params.cfg3 = stats_req->cfg_param[3];
+	}
+
+	ret = ath12k_dp_tx_htt_h2t_ext_stats_req(ar, type, &cfg_params, cookie);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to send htt stats request: %d\n", ret);
+		return ret;
+	}
+	if (!wait_for_completion_timeout(&stats_req->htt_stats_rcvd, 3 * HZ)) {
+		spin_lock_bh(&ar->data_lock);
+		if (!stats_req->done) {
+			stats_req->done = true;
+			spin_unlock_bh(&ar->data_lock);
+			ath12k_warn(ar->ab, "stats request timed out\n");
+			return -ETIMEDOUT;
+		}
+		spin_unlock_bh(&ar->data_lock);
+	}
+
+	return 0;
+}
+
+static int ath12k_open_htt_stats(struct inode *inode,
+				 struct file *file)
+{
+	struct ath12k *ar = inode->i_private;
+	struct debug_htt_stats_req *stats_req;
+	enum ath12k_dbg_htt_ext_stats_type type = ar->debug.htt_stats.type;
+	struct ath12k_hw *ah = ath12k_ar_to_ah(ar);
+	int ret;
+
+	if (type == ATH12K_DBG_HTT_EXT_STATS_RESET)
+		return -EPERM;
+
+	mutex_lock(&ar->conf_mutex);
+
+	if (ah->state != ATH12K_HW_STATE_ON) {
+		ret = -ENETDOWN;
+		goto err_unlock;
+	}
+
+	if (ar->debug.htt_stats.stats_req) {
+		ret = -EAGAIN;
+		goto err_unlock;
+	}
+
+	stats_req = kzalloc(sizeof(*stats_req) + ATH12K_HTT_STATS_BUF_SIZE, GFP_KERNEL);
+	if (!stats_req) {
+		ret = -ENOMEM;
+		goto err_unlock;
+	}
+
+	ar->debug.htt_stats.stats_req = stats_req;
+	stats_req->type = type;
+	stats_req->cfg_param[0] = ar->debug.htt_stats.cfg_param[0];
+	stats_req->cfg_param[1] = ar->debug.htt_stats.cfg_param[1];
+	stats_req->cfg_param[2] = ar->debug.htt_stats.cfg_param[2];
+	stats_req->cfg_param[3] = ar->debug.htt_stats.cfg_param[3];
+	stats_req->override_cfg_param = !!stats_req->cfg_param[0] ||
+					!!stats_req->cfg_param[1] ||
+					!!stats_req->cfg_param[2] ||
+					!!stats_req->cfg_param[3];
+
+	ret = ath12k_debugfs_htt_stats_req(ar);
+	if (ret < 0)
+		goto out;
+
+	file->private_data = stats_req;
+
+	mutex_unlock(&ar->conf_mutex);
+
+	return 0;
+out:
+	kfree(stats_req);
+	ar->debug.htt_stats.stats_req = NULL;
+err_unlock:
+	mutex_unlock(&ar->conf_mutex);
+
+	return ret;
+}
+
+static int ath12k_release_htt_stats(struct inode *inode,
+				    struct file *file)
+{
+	struct ath12k *ar = inode->i_private;
+
+	mutex_lock(&ar->conf_mutex);
+	kfree(file->private_data);
+	ar->debug.htt_stats.stats_req = NULL;
+	mutex_unlock(&ar->conf_mutex);
+
+	return 0;
+}
+
+static ssize_t ath12k_read_htt_stats(struct file *file,
+				     char __user *user_buf,
+				     size_t count, loff_t *ppos)
+{
+	struct debug_htt_stats_req *stats_req = file->private_data;
+	char *buf;
+	u32 length;
+
+	buf = stats_req->buf;
+	length = min_t(u32, stats_req->buf_len, ATH12K_HTT_STATS_BUF_SIZE);
+	return simple_read_from_buffer(user_buf, count, ppos, buf, length);
+}
+
+static const struct file_operations fops_dump_htt_stats = {
+	.open = ath12k_open_htt_stats,
+	.release = ath12k_release_htt_stats,
+	.read = ath12k_read_htt_stats,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
+static ssize_t ath12k_write_htt_stats_reset(struct file *file,
+					    const char __user *user_buf,
+					    size_t count, loff_t *ppos)
+{
+	struct ath12k *ar = file->private_data;
+	enum ath12k_dbg_htt_ext_stats_type type;
+	struct htt_ext_stats_cfg_params cfg_params = { 0 };
+	u8 param_pos;
+	int ret;
+
+	ret = kstrtou32_from_user(user_buf, count, 0, &type);
+	if (ret)
+		return ret;
+
+	if (type >= ATH12K_DBG_HTT_NUM_EXT_STATS ||
+	    type == ATH12K_DBG_HTT_EXT_STATS_RESET)
+		return -E2BIG;
+
+	mutex_lock(&ar->conf_mutex);
+	cfg_params.cfg0 = HTT_STAT_DEFAULT_RESET_START_OFFSET;
+	param_pos = (type >> 5) + 1;
+
+	switch (param_pos) {
+	case ATH12K_HTT_STATS_RESET_PARAM_CFG_32_BYTES:
+		cfg_params.cfg1 = 1 << (cfg_params.cfg0 + type);
+		break;
+	case ATH12K_HTT_STATS_RESET_PARAM_CFG_64_BYTES:
+		cfg_params.cfg2 = ATH12K_HTT_STATS_RESET_BITMAP32_BIT(cfg_params.cfg0 +
+								      type);
+		break;
+	case ATH12K_HTT_STATS_RESET_PARAM_CFG_128_BYTES:
+		cfg_params.cfg3 = ATH12K_HTT_STATS_RESET_BITMAP64_BIT(cfg_params.cfg0 +
+								      type);
+		break;
+	default:
+		break;
+	}
+
+	ret = ath12k_dp_tx_htt_h2t_ext_stats_req(ar,
+						 ATH12K_DBG_HTT_EXT_STATS_RESET,
+						 &cfg_params,
+						 0ULL);
+	if (ret) {
+		ath12k_warn(ar->ab, "failed to send htt stats request: %d\n", ret);
+		mutex_unlock(&ar->conf_mutex);
+		return ret;
+	}
+
+	ar->debug.htt_stats.reset = type;
+	mutex_unlock(&ar->conf_mutex);
+
+	return count;
+}
+
+static const struct file_operations fops_htt_stats_reset = {
+	.write = ath12k_write_htt_stats_reset,
+	.open = simple_open,
+	.owner = THIS_MODULE,
+	.llseek = default_llseek,
+};
+
 void ath12k_debugfs_htt_stats_register(struct ath12k *ar)
 {
 	debugfs_create_file("htt_stats_type", 0600, ar->debug.debugfs_pdev,
 			    ar, &fops_htt_stats_type);
+	debugfs_create_file("htt_stats", 0400, ar->debug.debugfs_pdev,
+			    ar, &fops_dump_htt_stats);
+	debugfs_create_file("htt_stats_reset", 0200, ar->debug.debugfs_pdev,
+			    ar, &fops_htt_stats_reset);
 }
diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
index 8610db89289a..252c59f5eac8 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
@@ -7,6 +7,18 @@
 #ifndef DEBUG_HTT_STATS_H
 #define DEBUG_HTT_STATS_H
 
+#define ATH12K_HTT_STATS_BUF_SIZE		(1024 * 512)
+#define ATH12K_HTT_STATS_COOKIE_LSB		GENMASK_ULL(31, 0)
+#define ATH12K_HTT_STATS_COOKIE_MSB		GENMASK_ULL(63, 32)
+#define ATH12K_HTT_STATS_MAGIC_VALUE		0xF0F0F0F0
+
+#define ATH12K_HTT_STATS_RESET_BITMAP32_OFFSET(_idx)	((_idx) & 0x1f)
+#define ATH12K_HTT_STATS_RESET_BITMAP64_OFFSET(_idx)	((_idx) & 0x3f)
+#define ATH12K_HTT_STATS_RESET_BITMAP32_BIT(_idx)	(1 << \
+		ATH12K_HTT_STATS_RESET_BITMAP32_OFFSET(_idx))
+#define ATH12K_HTT_STATS_RESET_BITMAP64_BIT(_idx)	(1 << \
+		ATH12K_HTT_STATS_RESET_BITMAP64_OFFSET(_idx))
+
 void ath12k_debugfs_htt_stats_register(struct ath12k *ar);
 
 /* htt_dbg_ext_stats_type */
@@ -17,4 +29,22 @@ enum ath12k_dbg_htt_ext_stats_type {
 	ATH12K_DBG_HTT_NUM_EXT_STATS,
 };
 
+enum ath12k_htt_stats_reset_cfg_param_alloc_pos {
+	ATH12K_HTT_STATS_RESET_PARAM_CFG_32_BYTES = 1,
+	ATH12K_HTT_STATS_RESET_PARAM_CFG_64_BYTES,
+	ATH12K_HTT_STATS_RESET_PARAM_CFG_128_BYTES,
+};
+
+struct debug_htt_stats_req {
+	bool done;
+	bool override_cfg_param;
+	u8 pdev_id;
+	enum ath12k_dbg_htt_ext_stats_type type;
+	u32 cfg_param[4];
+	u8 peer_addr[ETH_ALEN];
+	struct completion htt_stats_rcvd;
+	u32 buf_len;
+	u8 buf[];
+};
+
 #endif
diff --git a/drivers/net/wireless/ath/ath12k/dp_tx.c b/drivers/net/wireless/ath/ath12k/dp_tx.c
index c4cfa7cf7cb9..d08c04343e90 100644
--- a/drivers/net/wireless/ath/ath12k/dp_tx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_tx.c
@@ -1086,6 +1086,7 @@ ath12k_dp_tx_htt_h2t_ext_stats_req(struct ath12k *ar, u8 type,
 	struct htt_ext_stats_cfg_cmd *cmd;
 	int len = sizeof(*cmd);
 	int ret;
+	u32 pdev_id;
 
 	skb = ath12k_htc_alloc_skb(ab, len);
 	if (!skb)
@@ -1097,7 +1098,8 @@ ath12k_dp_tx_htt_h2t_ext_stats_req(struct ath12k *ar, u8 type,
 	memset(cmd, 0, sizeof(*cmd));
 	cmd->hdr.msg_type = HTT_H2T_MSG_TYPE_EXT_STATS_CFG;
 
-	cmd->hdr.pdev_mask = 1 << ar->pdev->pdev_id;
+	pdev_id = ath12k_mac_get_target_pdev_id(ar);
+	cmd->hdr.pdev_mask = 1 << pdev_id;
 
 	cmd->hdr.stats_type = type;
 	cmd->cfg_param0 = cpu_to_le32(cfg_params->cfg0);
-- 
cgit v1.2.3-58-ga151


From adf6df963c0325c6f3848ade1b73d2afc4948351 Mon Sep 17 00:00:00 2001
From: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Date: Thu, 27 Jun 2024 19:36:56 +0300
Subject: wifi: ath12k: Add support to parse requested stats_type

Add extended htt stats parser and print the corresponding TLVs associated
with the requested htt_stats_type.
Add support for TX PDEV related htt stats.

Sample output:
--------------
echo 1 > /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats_type
cat /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats

HTT_TX_PDEV_STATS_CMN_TLV:
mac_id = 0
comp_delivered = 0
self_triggers = 256
hw_queued = 275
hw_reaped = 275
underrun = 241
hw_paused = 0
hw_flush = 0
hw_filt = 1
tx_abort = 0
ppdu_ok = 246
mpdu_requeued = 0
tx_xretry = 0
data_rc = 3
mpdu_dropped_xretry = 0
illegal_rate_phy_err = 0
cont_xretry = 0
tx_timeout = 0
tx_time_dur_data = 0
pdev_resets = 0
phy_underrun = 0
txop_ovf = 0
seq_posted = 247
seq_failed_queueing = 0
seq_completed = 247
seq_restarted = 0
seq_txop_repost_stop = 0
next_seq_cancel = 0
dl_mu_mimo_seq_posted = 0
dl_mu_ofdma_seq_posted = 0
ul_mu_mimo_seq_posted = 0
ul_mu_ofdma_seq_posted = 0
mu_mimo_peer_blacklisted = 0
seq_qdepth_repost_stop = 0
seq_min_msdu_repost_stop = 0
mu_seq_min_msdu_repost_stop = 0
seq_switch_hw_paused = 0
next_seq_posted_dsr = 0
seq_posted_isr = 0
seq_ctrl_cached = 0
mpdu_count_tqm = 0
msdu_count_tqm = 0
mpdu_removed_tqm = 0
msdu_removed_tqm = 0
remove_mpdus_max_retries = 0
mpdus_sw_flush = 0
mpdus_hw_filter = 0
mpdus_truncated = 0
mpdus_ack_failed = 0
mpdus_expired = 0
mpdus_seq_hw_retry = 0
ack_tlv_proc = 0
coex_abort_mpdu_cnt_valid = 0
coex_abort_mpdu_cnt = 5
num_total_ppdus_tried_ota = 5
num_data_ppdus_tried_ota = 0
local_ctrl_mgmt_enqued = 247
local_ctrl_mgmt_freed = 247
local_data_enqued = 0
local_data_freed = 0
mpdu_tried = 0
isr_wait_seq_posted = 0
tx_active_dur_us_low = 0
tx_active_dur_us_high = 0
fes_offsets_err_cnt = 0

HTT_TX_PDEV_STATS_URRN_TLV:
urrn_stats =  0:0, 1:241, 2:0

HTT_TX_PDEV_STATS_SIFS_TLV:
sifs_status =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0

HTT_TX_PDEV_STATS_FLUSH_TLV:
flush_errs =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0, 10:0,
11:0, 12:0, 13:0, 14:0, 15:0, 16:0, 17:0, 18:0, 19:0, 20:0, 21:0, 22:0,
23:0, 24:0, 25:0, 26:0, 27:0, 28:0, 29:0, 30:0, 31:0, 32:0, 33:0, 34:0,
35:0, 36:0, 37:0, 38:0, 39:0, 40:0, 41:0, 42:0, 43:0, 44:0, 45:0, 46:0,
47:0, 48:0, 49:0, 50:0, 51:0, 52:0, 53:0, 54:0, 55:0, 56:0, 57:0, 58:0,
59:0, 60:0, 61:0, 62:0, 63:0, 64:0, 65:0, 66:0, 67:0, 68:0, 69:0, 70:0,
71:0, 72:0, 73:0, 74:0, 75:0, 76:0, 77:0, 78:0, 79:0, 80:0, 81:0, 82:0,
83:0, 84:0, 85:0, 86:0, 87:0, 88:0, 89:0, 90:0, 91:0, 92:0, 93:0, 94:0,
95:0, 96:0, 97:0, 98:0, 99:0, 100:0, 101:0, 102:0, 103:0, 104:0, 105:0,
106:0, 107:0, 108:0, 109:0, 110:0, 111:0, 112:0, 113:0, 114:0, 115:0,
116:0, 117:0, 118:0, 119:0, 120:0, 121:0, 122:0, 123:0, 124:0, 125:0,
126:0, 127:0

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Co-developed-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Signed-off-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240626085854.2500681-4-quic_rgnanase@quicinc.com
---
 .../net/wireless/ath/ath12k/debugfs_htt_stats.c    | 328 +++++++++++++++++++++
 .../net/wireless/ath/ath12k/debugfs_htt_stats.h    | 211 +++++++++++++
 drivers/net/wireless/ath/ath12k/dp_rx.c            |  10 +-
 drivers/net/wireless/ath/ath12k/dp_rx.h            |   4 +
 4 files changed, 549 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
index 9f44285caa58..0c1d8644c1f6 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
@@ -9,6 +9,334 @@
 #include "debug.h"
 #include "debugfs_htt_stats.h"
 #include "dp_tx.h"
+#include "dp_rx.h"
+
+static u32
+print_array_to_buf(u8 *buf, u32 offset, const char *header,
+		   const __le32 *array, u32 array_len, const char *footer)
+{
+	int index = 0;
+	u8 i;
+
+	if (header) {
+		index += scnprintf(buf + offset,
+				   ATH12K_HTT_STATS_BUF_SIZE - offset,
+				   "%s = ", header);
+	}
+	for (i = 0; i < array_len; i++) {
+		index += scnprintf(buf + offset + index,
+				   (ATH12K_HTT_STATS_BUF_SIZE - offset) - index,
+				   " %u:%u,", i, le32_to_cpu(array[i]));
+	}
+	/* To overwrite the last trailing comma */
+	index--;
+	*(buf + offset + index) = '\0';
+
+	if (footer) {
+		index += scnprintf(buf + offset + index,
+				   (ATH12K_HTT_STATS_BUF_SIZE - offset) - index,
+				   "%s", footer);
+	}
+	return index;
+}
+
+static void
+htt_print_tx_pdev_stats_cmn_tlv(const void *tag_buf, u16 tag_len,
+				struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_pdev_stats_cmn_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u32 mac_id_word;
+
+	if (tag_len < sizeof(*htt_stats_buf))
+		return;
+
+	mac_id_word = le32_to_cpu(htt_stats_buf->mac_id__word);
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_TX_PDEV_STATS_CMN_TLV:\n");
+	len += scnprintf(buf + len, buf_len - len, "mac_id = %u\n",
+			 u32_get_bits(mac_id_word, ATH12K_HTT_STATS_MAC_ID));
+	len += scnprintf(buf + len, buf_len - len, "comp_delivered = %u\n",
+			 le32_to_cpu(htt_stats_buf->comp_delivered));
+	len += scnprintf(buf + len, buf_len - len, "self_triggers = %u\n",
+			 le32_to_cpu(htt_stats_buf->self_triggers));
+	len += scnprintf(buf + len, buf_len - len, "hw_queued = %u\n",
+			 le32_to_cpu(htt_stats_buf->hw_queued));
+	len += scnprintf(buf + len, buf_len - len, "hw_reaped = %u\n",
+			 le32_to_cpu(htt_stats_buf->hw_reaped));
+	len += scnprintf(buf + len, buf_len - len, "underrun = %u\n",
+			 le32_to_cpu(htt_stats_buf->underrun));
+	len += scnprintf(buf + len, buf_len - len, "hw_paused = %u\n",
+			 le32_to_cpu(htt_stats_buf->hw_paused));
+	len += scnprintf(buf + len, buf_len - len, "hw_flush = %u\n",
+			 le32_to_cpu(htt_stats_buf->hw_flush));
+	len += scnprintf(buf + len, buf_len - len, "hw_filt = %u\n",
+			 le32_to_cpu(htt_stats_buf->hw_filt));
+	len += scnprintf(buf + len, buf_len - len, "tx_abort = %u\n",
+			 le32_to_cpu(htt_stats_buf->tx_abort));
+	len += scnprintf(buf + len, buf_len - len, "ppdu_ok = %u\n",
+			 le32_to_cpu(htt_stats_buf->ppdu_ok));
+	len += scnprintf(buf + len, buf_len - len, "mpdu_requeued = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdu_requed));
+	len += scnprintf(buf + len, buf_len - len, "tx_xretry = %u\n",
+			 le32_to_cpu(htt_stats_buf->tx_xretry));
+	len += scnprintf(buf + len, buf_len - len, "data_rc = %u\n",
+			 le32_to_cpu(htt_stats_buf->data_rc));
+	len += scnprintf(buf + len, buf_len - len, "mpdu_dropped_xretry = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdu_dropped_xretry));
+	len += scnprintf(buf + len, buf_len - len, "illegal_rate_phy_err = %u\n",
+			 le32_to_cpu(htt_stats_buf->illgl_rate_phy_err));
+	len += scnprintf(buf + len, buf_len - len, "cont_xretry = %u\n",
+			 le32_to_cpu(htt_stats_buf->cont_xretry));
+	len += scnprintf(buf + len, buf_len - len, "tx_timeout = %u\n",
+			 le32_to_cpu(htt_stats_buf->tx_timeout));
+	len += scnprintf(buf + len, buf_len - len, "tx_time_dur_data = %u\n",
+			 le32_to_cpu(htt_stats_buf->tx_time_dur_data));
+	len += scnprintf(buf + len, buf_len - len, "pdev_resets = %u\n",
+			 le32_to_cpu(htt_stats_buf->pdev_resets));
+	len += scnprintf(buf + len, buf_len - len, "phy_underrun = %u\n",
+			 le32_to_cpu(htt_stats_buf->phy_underrun));
+	len += scnprintf(buf + len, buf_len - len, "txop_ovf = %u\n",
+			 le32_to_cpu(htt_stats_buf->txop_ovf));
+	len += scnprintf(buf + len, buf_len - len, "seq_posted = %u\n",
+			 le32_to_cpu(htt_stats_buf->seq_posted));
+	len += scnprintf(buf + len, buf_len - len, "seq_failed_queueing = %u\n",
+			 le32_to_cpu(htt_stats_buf->seq_failed_queueing));
+	len += scnprintf(buf + len, buf_len - len, "seq_completed = %u\n",
+			 le32_to_cpu(htt_stats_buf->seq_completed));
+	len += scnprintf(buf + len, buf_len - len, "seq_restarted = %u\n",
+			 le32_to_cpu(htt_stats_buf->seq_restarted));
+	len += scnprintf(buf + len, buf_len - len, "seq_txop_repost_stop = %u\n",
+			 le32_to_cpu(htt_stats_buf->seq_txop_repost_stop));
+	len += scnprintf(buf + len, buf_len - len, "next_seq_cancel = %u\n",
+			 le32_to_cpu(htt_stats_buf->next_seq_cancel));
+	len += scnprintf(buf + len, buf_len - len, "dl_mu_mimo_seq_posted = %u\n",
+			 le32_to_cpu(htt_stats_buf->mu_seq_posted));
+	len += scnprintf(buf + len, buf_len - len, "dl_mu_ofdma_seq_posted = %u\n",
+			 le32_to_cpu(htt_stats_buf->mu_ofdma_seq_posted));
+	len += scnprintf(buf + len, buf_len - len, "ul_mu_mimo_seq_posted = %u\n",
+			 le32_to_cpu(htt_stats_buf->ul_mumimo_seq_posted));
+	len += scnprintf(buf + len, buf_len - len, "ul_mu_ofdma_seq_posted = %u\n",
+			 le32_to_cpu(htt_stats_buf->ul_ofdma_seq_posted));
+	len += scnprintf(buf + len, buf_len - len, "mu_mimo_peer_blacklisted = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_mu_peer_blacklisted));
+	len += scnprintf(buf + len, buf_len - len, "seq_qdepth_repost_stop = %u\n",
+			 le32_to_cpu(htt_stats_buf->seq_qdepth_repost_stop));
+	len += scnprintf(buf + len, buf_len - len, "seq_min_msdu_repost_stop = %u\n",
+			 le32_to_cpu(htt_stats_buf->seq_min_msdu_repost_stop));
+	len += scnprintf(buf + len, buf_len - len, "mu_seq_min_msdu_repost_stop = %u\n",
+			 le32_to_cpu(htt_stats_buf->mu_seq_min_msdu_repost_stop));
+	len += scnprintf(buf + len, buf_len - len, "seq_switch_hw_paused = %u\n",
+			 le32_to_cpu(htt_stats_buf->seq_switch_hw_paused));
+	len += scnprintf(buf + len, buf_len - len, "next_seq_posted_dsr = %u\n",
+			 le32_to_cpu(htt_stats_buf->next_seq_posted_dsr));
+	len += scnprintf(buf + len, buf_len - len, "seq_posted_isr = %u\n",
+			 le32_to_cpu(htt_stats_buf->seq_posted_isr));
+	len += scnprintf(buf + len, buf_len - len, "seq_ctrl_cached = %u\n",
+			 le32_to_cpu(htt_stats_buf->seq_ctrl_cached));
+	len += scnprintf(buf + len, buf_len - len, "mpdu_count_tqm = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdu_count_tqm));
+	len += scnprintf(buf + len, buf_len - len, "msdu_count_tqm = %u\n",
+			 le32_to_cpu(htt_stats_buf->msdu_count_tqm));
+	len += scnprintf(buf + len, buf_len - len, "mpdu_removed_tqm = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdu_removed_tqm));
+	len += scnprintf(buf + len, buf_len - len, "msdu_removed_tqm = %u\n",
+			 le32_to_cpu(htt_stats_buf->msdu_removed_tqm));
+	len += scnprintf(buf + len, buf_len - len, "remove_mpdus_max_retries = %u\n",
+			 le32_to_cpu(htt_stats_buf->remove_mpdus_max_retries));
+	len += scnprintf(buf + len, buf_len - len, "mpdus_sw_flush = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdus_sw_flush));
+	len += scnprintf(buf + len, buf_len - len, "mpdus_hw_filter = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdus_hw_filter));
+	len += scnprintf(buf + len, buf_len - len, "mpdus_truncated = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdus_truncated));
+	len += scnprintf(buf + len, buf_len - len, "mpdus_ack_failed = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdus_ack_failed));
+	len += scnprintf(buf + len, buf_len - len, "mpdus_expired = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdus_expired));
+	len += scnprintf(buf + len, buf_len - len, "mpdus_seq_hw_retry = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdus_seq_hw_retry));
+	len += scnprintf(buf + len, buf_len - len, "ack_tlv_proc = %u\n",
+			 le32_to_cpu(htt_stats_buf->ack_tlv_proc));
+	len += scnprintf(buf + len, buf_len - len, "coex_abort_mpdu_cnt_valid = %u\n",
+			 le32_to_cpu(htt_stats_buf->coex_abort_mpdu_cnt_valid));
+	len += scnprintf(buf + len, buf_len - len, "coex_abort_mpdu_cnt = %u\n",
+			 le32_to_cpu(htt_stats_buf->coex_abort_mpdu_cnt));
+	len += scnprintf(buf + len, buf_len - len, "num_total_ppdus_tried_ota = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_total_ppdus_tried_ota));
+	len += scnprintf(buf + len, buf_len - len, "num_data_ppdus_tried_ota = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_data_ppdus_tried_ota));
+	len += scnprintf(buf + len, buf_len - len, "local_ctrl_mgmt_enqued = %u\n",
+			 le32_to_cpu(htt_stats_buf->local_ctrl_mgmt_enqued));
+	len += scnprintf(buf + len, buf_len - len, "local_ctrl_mgmt_freed = %u\n",
+			 le32_to_cpu(htt_stats_buf->local_ctrl_mgmt_freed));
+	len += scnprintf(buf + len, buf_len - len, "local_data_enqued = %u\n",
+			 le32_to_cpu(htt_stats_buf->local_data_enqued));
+	len += scnprintf(buf + len, buf_len - len, "local_data_freed = %u\n",
+			 le32_to_cpu(htt_stats_buf->local_data_freed));
+	len += scnprintf(buf + len, buf_len - len, "mpdu_tried = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdu_tried));
+	len += scnprintf(buf + len, buf_len - len, "isr_wait_seq_posted = %u\n",
+			 le32_to_cpu(htt_stats_buf->isr_wait_seq_posted));
+	len += scnprintf(buf + len, buf_len - len, "tx_active_dur_us_low = %u\n",
+			 le32_to_cpu(htt_stats_buf->tx_active_dur_us_low));
+	len += scnprintf(buf + len, buf_len - len, "tx_active_dur_us_high = %u\n",
+			 le32_to_cpu(htt_stats_buf->tx_active_dur_us_high));
+	len += scnprintf(buf + len, buf_len - len, "fes_offsets_err_cnt = %u\n\n",
+			 le32_to_cpu(htt_stats_buf->fes_offsets_err_cnt));
+
+	stats_req->buf_len = len;
+}
+
+static void
+htt_print_tx_pdev_stats_urrn_tlv(const void *tag_buf,
+				 u16 tag_len,
+				 struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_pdev_stats_urrn_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u16 num_elems = min_t(u16, (tag_len >> 2),
+			      HTT_TX_PDEV_MAX_URRN_STATS);
+
+	len += scnprintf(buf + len, buf_len - len,
+			"HTT_TX_PDEV_STATS_URRN_TLV:\n");
+
+	len += print_array_to_buf(buf, len, "urrn_stats", htt_stats_buf->urrn_stats,
+				  num_elems, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
+static void
+htt_print_tx_pdev_stats_flush_tlv(const void *tag_buf,
+				  u16 tag_len,
+				  struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_pdev_stats_flush_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u16 num_elems = min_t(u16, (tag_len >> 2),
+			      ATH12K_HTT_TX_PDEV_MAX_FLUSH_REASON_STATS);
+
+	len += scnprintf(buf + len, buf_len - len,
+			 "HTT_TX_PDEV_STATS_FLUSH_TLV:\n");
+
+	len += print_array_to_buf(buf, len, "flush_errs", htt_stats_buf->flush_errs,
+				  num_elems, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
+static void
+htt_print_tx_pdev_stats_sifs_tlv(const void *tag_buf,
+				 u16 tag_len,
+				 struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_pdev_stats_sifs_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u16 num_elems = min_t(u16, (tag_len >> 2),
+			      ATH12K_HTT_TX_PDEV_MAX_SIFS_BURST_STATS);
+
+	len += scnprintf(buf + len, buf_len - len,
+			 "HTT_TX_PDEV_STATS_SIFS_TLV:\n");
+
+	len += print_array_to_buf(buf, len, "sifs_status", htt_stats_buf->sifs_status,
+				  num_elems, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
+static int ath12k_dbg_htt_ext_stats_parse(struct ath12k_base *ab,
+					  u16 tag, u16 len, const void *tag_buf,
+					  void *user_data)
+{
+	struct debug_htt_stats_req *stats_req = user_data;
+
+	switch (tag) {
+	case HTT_STATS_TX_PDEV_CMN_TAG:
+		htt_print_tx_pdev_stats_cmn_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_PDEV_UNDERRUN_TAG:
+		htt_print_tx_pdev_stats_urrn_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_PDEV_SIFS_TAG:
+		htt_print_tx_pdev_stats_sifs_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_PDEV_FLUSH_TAG:
+		htt_print_tx_pdev_stats_flush_tlv(tag_buf, len, stats_req);
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+void ath12k_debugfs_htt_ext_stats_handler(struct ath12k_base *ab,
+					  struct sk_buff *skb)
+{
+	struct ath12k_htt_extd_stats_msg *msg;
+	struct debug_htt_stats_req *stats_req;
+	struct ath12k *ar;
+	u32 len, pdev_id, stats_info;
+	u64 cookie;
+	int ret;
+	bool send_completion = false;
+
+	msg = (struct ath12k_htt_extd_stats_msg *)skb->data;
+	cookie = le64_to_cpu(msg->cookie);
+
+	if (u64_get_bits(cookie, ATH12K_HTT_STATS_COOKIE_MSB) !=
+			 ATH12K_HTT_STATS_MAGIC_VALUE) {
+		ath12k_warn(ab, "received invalid htt ext stats event\n");
+		return;
+	}
+
+	pdev_id = u64_get_bits(cookie, ATH12K_HTT_STATS_COOKIE_LSB);
+	rcu_read_lock();
+	ar = ath12k_mac_get_ar_by_pdev_id(ab, pdev_id);
+	if (!ar) {
+		ath12k_warn(ab, "failed to get ar for pdev_id %d\n", pdev_id);
+		goto exit;
+	}
+
+	stats_req = ar->debug.htt_stats.stats_req;
+	if (!stats_req)
+		goto exit;
+
+	spin_lock_bh(&ar->data_lock);
+
+	stats_info = le32_to_cpu(msg->info1);
+	stats_req->done = u32_get_bits(stats_info, ATH12K_HTT_T2H_EXT_STATS_INFO1_DONE);
+	if (stats_req->done)
+		send_completion = true;
+
+	spin_unlock_bh(&ar->data_lock);
+
+	len = u32_get_bits(stats_info, ATH12K_HTT_T2H_EXT_STATS_INFO1_LENGTH);
+	if (len > skb->len) {
+		ath12k_warn(ab, "invalid length %d for HTT stats", len);
+		goto exit;
+	}
+
+	ret = ath12k_dp_htt_tlv_iter(ab, msg->data, len,
+				     ath12k_dbg_htt_ext_stats_parse,
+				     stats_req);
+	if (ret)
+		ath12k_warn(ab, "Failed to parse tlv %d\n", ret);
+
+	if (send_completion)
+		complete(&stats_req->htt_stats_rcvd);
+exit:
+	rcu_read_unlock();
+}
 
 static ssize_t ath12k_read_htt_stats_type(struct file *file,
 					  char __user *user_buf,
diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
index 252c59f5eac8..83060c54d00b 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
@@ -21,14 +21,134 @@
 
 void ath12k_debugfs_htt_stats_register(struct ath12k *ar);
 
+#ifdef CONFIG_ATH12K_DEBUGFS
+void ath12k_debugfs_htt_ext_stats_handler(struct ath12k_base *ab,
+					  struct sk_buff *skb);
+#else /* CONFIG_ATH12K_DEBUGFS */
+static inline void ath12k_debugfs_htt_ext_stats_handler(struct ath12k_base *ab,
+							struct sk_buff *skb)
+{
+}
+#endif
+
+/**
+ * DOC: target -> host extended statistics upload
+ *
+ * The following field definitions describe the format of the HTT
+ * target to host stats upload confirmation message.
+ * The message contains a cookie echoed from the HTT host->target stats
+ * upload request, which identifies which request the confirmation is
+ * for, and a single stats can span over multiple HTT stats indication
+ * due to the HTT message size limitation so every HTT ext stats
+ * indication will have tag-length-value stats information elements.
+ * The tag-length header for each HTT stats IND message also includes a
+ * status field, to indicate whether the request for the stat type in
+ * question was fully met, partially met, unable to be met, or invalid
+ * (if the stat type in question is disabled in the target).
+ * A Done bit 1's indicate the end of the of stats info elements.
+ *
+ *
+ * |31                         16|15    12|11|10 8|7   5|4       0|
+ * |--------------------------------------------------------------|
+ * |                   reserved                   |    msg type   |
+ * |--------------------------------------------------------------|
+ * |                         cookie LSBs                          |
+ * |--------------------------------------------------------------|
+ * |                         cookie MSBs                          |
+ * |--------------------------------------------------------------|
+ * |      stats entry length     | rsvd   | D|  S |   stat type   |
+ * |--------------------------------------------------------------|
+ * |                   type-specific stats info                   |
+ * |                      (see debugfs_htt_stats.h)               |
+ * |--------------------------------------------------------------|
+ * Header fields:
+ *  - MSG_TYPE
+ *    Bits 7:0
+ *    Purpose: Identifies this is a extended statistics upload confirmation
+ *             message.
+ *    Value: 0x1c
+ *  - COOKIE_LSBS
+ *    Bits 31:0
+ *    Purpose: Provide a mechanism to match a target->host stats confirmation
+ *        message with its preceding host->target stats request message.
+ *    Value: MSBs of the opaque cookie specified by the host-side requestor
+ *  - COOKIE_MSBS
+ *    Bits 31:0
+ *    Purpose: Provide a mechanism to match a target->host stats confirmation
+ *        message with its preceding host->target stats request message.
+ *    Value: MSBs of the opaque cookie specified by the host-side requestor
+ *
+ * Stats Information Element tag-length header fields:
+ *  - STAT_TYPE
+ *    Bits 7:0
+ *    Purpose: identifies the type of statistics info held in the
+ *        following information element
+ *    Value: ath12k_dbg_htt_ext_stats_type
+ *  - STATUS
+ *    Bits 10:8
+ *    Purpose: indicate whether the requested stats are present
+ *    Value:
+ *       0 -> The requested stats have been delivered in full
+ *       1 -> The requested stats have been delivered in part
+ *       2 -> The requested stats could not be delivered (error case)
+ *       3 -> The requested stat type is either not recognized (invalid)
+ *  - DONE
+ *    Bits 11
+ *    Purpose:
+ *        Indicates the completion of the stats entry, this will be the last
+ *        stats conf HTT segment for the requested stats type.
+ *    Value:
+ *        0 -> the stats retrieval is ongoing
+ *        1 -> the stats retrieval is complete
+ *  - LENGTH
+ *    Bits 31:16
+ *    Purpose: indicate the stats information size
+ *    Value: This field specifies the number of bytes of stats information
+ *       that follows the element tag-length header.
+ *       It is expected but not required that this length is a multiple of
+ *       4 bytes.
+ */
+
+#define ATH12K_HTT_T2H_EXT_STATS_INFO1_DONE		BIT(11)
+#define ATH12K_HTT_T2H_EXT_STATS_INFO1_LENGTH		GENMASK(31, 16)
+
+struct ath12k_htt_extd_stats_msg {
+	__le32 info0;
+	__le64 cookie;
+	__le32 info1;
+	u8 data[];
+} __packed;
+
 /* htt_dbg_ext_stats_type */
 enum ath12k_dbg_htt_ext_stats_type {
 	ATH12K_DBG_HTT_EXT_STATS_RESET		= 0,
+	ATH12K_DBG_HTT_EXT_STATS_PDEV_TX	= 1,
 
 	/* keep this last */
 	ATH12K_DBG_HTT_NUM_EXT_STATS,
 };
 
+enum ath12k_dbg_htt_tlv_tag {
+	HTT_STATS_TX_PDEV_CMN_TAG			= 0,
+	HTT_STATS_TX_PDEV_UNDERRUN_TAG			= 1,
+	HTT_STATS_TX_PDEV_SIFS_TAG			= 2,
+	HTT_STATS_TX_PDEV_FLUSH_TAG			= 3,
+
+	HTT_STATS_MAX_TAG,
+};
+
+#define ATH12K_HTT_STATS_MAC_ID				GENMASK(7, 0)
+
+#define ATH12K_HTT_TX_PDEV_MAX_SIFS_BURST_STATS		9
+#define ATH12K_HTT_TX_PDEV_MAX_FLUSH_REASON_STATS	150
+
+enum ath12k_htt_tx_pdev_underrun_enum {
+	HTT_STATS_TX_PDEV_NO_DATA_UNDERRUN		= 0,
+	HTT_STATS_TX_PDEV_DATA_UNDERRUN_BETWEEN_MPDU	= 1,
+	HTT_STATS_TX_PDEV_DATA_UNDERRUN_WITHIN_MPDU	= 2,
+	HTT_TX_PDEV_MAX_URRN_STATS			= 3,
+};
+
 enum ath12k_htt_stats_reset_cfg_param_alloc_pos {
 	ATH12K_HTT_STATS_RESET_PARAM_CFG_32_BYTES = 1,
 	ATH12K_HTT_STATS_RESET_PARAM_CFG_64_BYTES,
@@ -47,4 +167,95 @@ struct debug_htt_stats_req {
 	u8 buf[];
 };
 
+struct ath12k_htt_tx_pdev_stats_cmn_tlv {
+	__le32 mac_id__word;
+	__le32 hw_queued;
+	__le32 hw_reaped;
+	__le32 underrun;
+	__le32 hw_paused;
+	__le32 hw_flush;
+	__le32 hw_filt;
+	__le32 tx_abort;
+	__le32 mpdu_requed;
+	__le32 tx_xretry;
+	__le32 data_rc;
+	__le32 mpdu_dropped_xretry;
+	__le32 illgl_rate_phy_err;
+	__le32 cont_xretry;
+	__le32 tx_timeout;
+	__le32 pdev_resets;
+	__le32 phy_underrun;
+	__le32 txop_ovf;
+	__le32 seq_posted;
+	__le32 seq_failed_queueing;
+	__le32 seq_completed;
+	__le32 seq_restarted;
+	__le32 mu_seq_posted;
+	__le32 seq_switch_hw_paused;
+	__le32 next_seq_posted_dsr;
+	__le32 seq_posted_isr;
+	__le32 seq_ctrl_cached;
+	__le32 mpdu_count_tqm;
+	__le32 msdu_count_tqm;
+	__le32 mpdu_removed_tqm;
+	__le32 msdu_removed_tqm;
+	__le32 mpdus_sw_flush;
+	__le32 mpdus_hw_filter;
+	__le32 mpdus_truncated;
+	__le32 mpdus_ack_failed;
+	__le32 mpdus_expired;
+	__le32 mpdus_seq_hw_retry;
+	__le32 ack_tlv_proc;
+	__le32 coex_abort_mpdu_cnt_valid;
+	__le32 coex_abort_mpdu_cnt;
+	__le32 num_total_ppdus_tried_ota;
+	__le32 num_data_ppdus_tried_ota;
+	__le32 local_ctrl_mgmt_enqued;
+	__le32 local_ctrl_mgmt_freed;
+	__le32 local_data_enqued;
+	__le32 local_data_freed;
+	__le32 mpdu_tried;
+	__le32 isr_wait_seq_posted;
+
+	__le32 tx_active_dur_us_low;
+	__le32 tx_active_dur_us_high;
+	__le32 remove_mpdus_max_retries;
+	__le32 comp_delivered;
+	__le32 ppdu_ok;
+	__le32 self_triggers;
+	__le32 tx_time_dur_data;
+	__le32 seq_qdepth_repost_stop;
+	__le32 mu_seq_min_msdu_repost_stop;
+	__le32 seq_min_msdu_repost_stop;
+	__le32 seq_txop_repost_stop;
+	__le32 next_seq_cancel;
+	__le32 fes_offsets_err_cnt;
+	__le32 num_mu_peer_blacklisted;
+	__le32 mu_ofdma_seq_posted;
+	__le32 ul_mumimo_seq_posted;
+	__le32 ul_ofdma_seq_posted;
+
+	__le32 thermal_suspend_cnt;
+	__le32 dfs_suspend_cnt;
+	__le32 tx_abort_suspend_cnt;
+	__le32 tgt_specific_opaque_txq_suspend_info;
+	__le32 last_suspend_reason;
+} __packed;
+
+struct ath12k_htt_tx_pdev_stats_urrn_tlv {
+	DECLARE_FLEX_ARRAY(__le32, urrn_stats);
+} __packed;
+
+struct ath12k_htt_tx_pdev_stats_flush_tlv {
+	DECLARE_FLEX_ARRAY(__le32, flush_errs);
+} __packed;
+
+struct ath12k_htt_tx_pdev_stats_phy_err_tlv {
+	DECLARE_FLEX_ARRAY(__le32, phy_errs);
+} __packed;
+
+struct ath12k_htt_tx_pdev_stats_sifs_tlv {
+	DECLARE_FLEX_ARRAY(__le32, sifs_status);
+} __packed;
+
 #endif
diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.c b/drivers/net/wireless/ath/ath12k/dp_rx.c
index d7bb412ad391..14236d0a0c89 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.c
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.c
@@ -17,6 +17,7 @@
 #include "dp_tx.h"
 #include "peer.h"
 #include "dp_mon.h"
+#include "debugfs_htt_stats.h"
 
 #define ATH12K_DP_RX_FRAGMENT_TIMEOUT_MS (2 * HZ)
 
@@ -1268,10 +1269,10 @@ static int ath12k_htt_tlv_ppdu_stats_parse(struct ath12k_base *ab,
 	return 0;
 }
 
-static int ath12k_dp_htt_tlv_iter(struct ath12k_base *ab, const void *ptr, size_t len,
-				  int (*iter)(struct ath12k_base *ar, u16 tag, u16 len,
-					      const void *ptr, void *data),
-				  void *data)
+int ath12k_dp_htt_tlv_iter(struct ath12k_base *ab, const void *ptr, size_t len,
+			   int (*iter)(struct ath12k_base *ar, u16 tag, u16 len,
+				       const void *ptr, void *data),
+			   void *data)
 {
 	const struct htt_tlv *tlv;
 	const void *begin = ptr;
@@ -1741,6 +1742,7 @@ void ath12k_dp_htt_htc_t2h_msg_handler(struct ath12k_base *ab,
 		ath12k_htt_pull_ppdu_stats(ab, skb);
 		break;
 	case HTT_T2H_MSG_TYPE_EXT_STATS_CONF:
+		ath12k_debugfs_htt_ext_stats_handler(ab, skb);
 		break;
 	case HTT_T2H_MSG_TYPE_MLO_TIMESTAMP_OFFSET_IND:
 		ath12k_htt_mlo_offset_event_handler(ab, skb);
diff --git a/drivers/net/wireless/ath/ath12k/dp_rx.h b/drivers/net/wireless/ath/ath12k/dp_rx.h
index 2ff421160181..eb1f92559179 100644
--- a/drivers/net/wireless/ath/ath12k/dp_rx.h
+++ b/drivers/net/wireless/ath/ath12k/dp_rx.h
@@ -139,4 +139,8 @@ ath12k_dp_rx_h_find_peer(struct ath12k_base *ab, struct sk_buff *msdu);
 int ath12k_dp_rxdma_ring_sel_config_qcn9274(struct ath12k_base *ab);
 int ath12k_dp_rxdma_ring_sel_config_wcn7850(struct ath12k_base *ab);
 
+int ath12k_dp_htt_tlv_iter(struct ath12k_base *ab, const void *ptr, size_t len,
+			   int (*iter)(struct ath12k_base *ar, u16 tag, u16 len,
+				       const void *ptr, void *data),
+			   void *data);
 #endif /* ATH12K_DP_RX_H */
-- 
cgit v1.2.3-58-ga151


From a9f2976f0e9ebc4a579cdd507b153a191692a72c Mon Sep 17 00:00:00 2001
From: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Date: Thu, 27 Jun 2024 19:37:02 +0300
Subject: wifi: ath12k: Dump additional Tx PDEV HTT stats

Support to dump additional Tx PDEV stats through HTT stats debugfs.
Following stats dump are supported:
        1. PDEV control path stat to dump Tx management frame count
        2. Tx PDEV SIFS histogram stats
        3. Tx MU MIMO PPDU stats for 802.11ac, 802.11ax and 802.11be

Sample Output:
---------------
echo 1 > /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats_type
cat /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats

HTT_TX_PDEV_STATS_CMN_TLV:
mac_id = 0
comp_delivered = 0
self_triggers = 13
......
......
HTT_TX_PDEV_STATS_CTRL_PATH_TX_STATS:
fw_tx_mgmt_subtype =  0:1, 1:0, 2:0, 3:0, 4:38, 5:0, 6:0, 7:0, 8:0, 9:0, 10:0, 11:1, 12:0, 13:7, 14:0, 15:0

HTT_TX_PDEV_STATS_SIFS_HIST_TLV:
sifs_hist_status =  0:237, 1:185, 2:1, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0

HTT_TX_PDEV_AC_MU_PPDU_DISTRIBUTION_STATS:
ac_mu_mimo_num_seq_posted_nr4 = 0
ac_mu_mimo_num_ppdu_posted_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
ac_mu_mimo_num_ppdu_completed_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
ac_mu_mimo_num_seq_term_status_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0

ac_mu_mimo_num_seq_posted_nr8 = 0
ac_mu_mimo_num_ppdu_posted_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
ac_mu_mimo_num_ppdu_completed_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
ac_mu_mimo_num_seq_term_status_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0

HTT_TX_PDEV_AX_MU_PPDU_DISTRIBUTION_STATS:
ax_mu_mimo_num_seq_posted_nr4 = 0
ax_mu_mimo_num_ppdu_posted_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
ax_mu_mimo_num_ppdu_completed_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
ax_mu_mimo_num_seq_term_status_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0

ax_mu_mimo_num_seq_posted_nr8 = 0
ax_mu_mimo_num_ppdu_posted_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
ax_mu_mimo_num_ppdu_completed_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
ax_mu_mimo_num_seq_term_status_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0

HTT_TX_PDEV_BE_MU_PPDU_DISTRIBUTION_STATS:
be_mu_mimo_num_seq_posted_nr4 = 0
be_mu_mimo_num_ppdu_posted_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
be_mu_mimo_num_ppdu_completed_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
be_mu_mimo_num_seq_term_status_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0

be_mu_mimo_num_seq_posted_nr8 = 0
be_mu_mimo_num_ppdu_posted_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
be_mu_mimo_num_ppdu_completed_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0
be_mu_mimo_num_seq_term_status_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3

Signed-off-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240626085854.2500681-5-quic_rgnanase@quicinc.com
---
 .../net/wireless/ath/ath12k/debugfs_htt_stats.c    | 142 +++++++++++++++++++++
 .../net/wireless/ath/ath12k/debugfs_htt_stats.h    |  39 ++++++
 2 files changed, 181 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
index 0c1d8644c1f6..fd04f057598d 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
@@ -253,6 +253,139 @@ htt_print_tx_pdev_stats_sifs_tlv(const void *tag_buf,
 	stats_req->buf_len = len;
 }
 
+static void
+htt_print_tx_pdev_mu_ppdu_dist_stats_tlv(const void *tag_buf, u16 tag_len,
+					 struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_pdev_mu_ppdu_dist_stats_tlv *htt_stats_buf = tag_buf;
+	char *mode;
+	u8 j, hw_mode, i, str_buf_len;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u32 stats_value;
+	u8 max_ppdu = ATH12K_HTT_STATS_MAX_NUM_MU_PPDU_PER_BURST;
+	u8 max_sched = ATH12K_HTT_STATS_MAX_NUM_SCHED_STATUS;
+	char str_buf[ATH12K_HTT_MAX_STRING_LEN];
+
+	if (tag_len < sizeof(*htt_stats_buf))
+		return;
+
+	hw_mode = le32_to_cpu(htt_stats_buf->hw_mode);
+
+	switch (hw_mode) {
+	case ATH12K_HTT_STATS_HWMODE_AC:
+		len += scnprintf(buf + len, buf_len - len,
+				 "HTT_TX_PDEV_AC_MU_PPDU_DISTRIBUTION_STATS:\n");
+		mode = "ac";
+		break;
+	case ATH12K_HTT_STATS_HWMODE_AX:
+		len += scnprintf(buf + len, buf_len - len,
+				 "HTT_TX_PDEV_AX_MU_PPDU_DISTRIBUTION_STATS:\n");
+		mode = "ax";
+		break;
+	case ATH12K_HTT_STATS_HWMODE_BE:
+		len += scnprintf(buf + len, buf_len - len,
+				 "HTT_TX_PDEV_BE_MU_PPDU_DISTRIBUTION_STATS:\n");
+		mode = "be";
+		break;
+	default:
+		return;
+	}
+
+	for (i = 0; i < ATH12K_HTT_STATS_NUM_NR_BINS ; i++) {
+		len += scnprintf(buf + len, buf_len - len,
+				 "%s_mu_mimo_num_seq_posted_nr%u = %u\n", mode,
+				 ((i + 1) * 4), htt_stats_buf->num_seq_posted[i]);
+		str_buf_len = 0;
+		memset(str_buf, 0x0, sizeof(str_buf));
+		for (j = 0; j < ATH12K_HTT_STATS_MAX_NUM_MU_PPDU_PER_BURST ; j++) {
+			stats_value = le32_to_cpu(htt_stats_buf->num_ppdu_posted_per_burst
+						  [i * max_ppdu + j]);
+			str_buf_len += scnprintf(&str_buf[str_buf_len],
+						ATH12K_HTT_MAX_STRING_LEN - str_buf_len,
+						" %u:%u,", j, stats_value);
+		}
+		/* To overwrite the last trailing comma */
+		str_buf[str_buf_len - 1] = '\0';
+		len += scnprintf(buf + len, buf_len - len,
+				 "%s_mu_mimo_num_ppdu_posted_per_burst_nr%u = %s\n",
+				 mode, ((i + 1) * 4), str_buf);
+		str_buf_len = 0;
+		memset(str_buf, 0x0, sizeof(str_buf));
+		for (j = 0; j < ATH12K_HTT_STATS_MAX_NUM_MU_PPDU_PER_BURST ; j++) {
+			stats_value = le32_to_cpu(htt_stats_buf->num_ppdu_cmpl_per_burst
+						  [i * max_ppdu + j]);
+			str_buf_len += scnprintf(&str_buf[str_buf_len],
+						ATH12K_HTT_MAX_STRING_LEN - str_buf_len,
+						" %u:%u,", j, stats_value);
+		}
+		/* To overwrite the last trailing comma */
+		str_buf[str_buf_len - 1] = '\0';
+		len += scnprintf(buf + len, buf_len - len,
+				 "%s_mu_mimo_num_ppdu_completed_per_burst_nr%u = %s\n",
+				 mode, ((i + 1) * 4), str_buf);
+		str_buf_len = 0;
+		memset(str_buf, 0x0, sizeof(str_buf));
+		for (j = 0; j < ATH12K_HTT_STATS_MAX_NUM_SCHED_STATUS ; j++) {
+			stats_value = le32_to_cpu(htt_stats_buf->num_seq_term_status
+						  [i * max_sched + j]);
+			str_buf_len += scnprintf(&str_buf[str_buf_len],
+						ATH12K_HTT_MAX_STRING_LEN - str_buf_len,
+						" %u:%u,", j, stats_value);
+		}
+		/* To overwrite the last trailing comma */
+		str_buf[str_buf_len - 1] = '\0';
+		len += scnprintf(buf + len, buf_len - len,
+				 "%s_mu_mimo_num_seq_term_status_nr%u = %s\n\n",
+				 mode, ((i + 1) * 4), str_buf);
+	}
+
+	stats_req->buf_len = len;
+}
+
+static void
+htt_print_tx_pdev_stats_sifs_hist_tlv(const void *tag_buf,
+				      u16 tag_len,
+				      struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_pdev_stats_sifs_hist_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u16 num_elems = min_t(u16, (tag_len >> 2),
+			      ATH12K_HTT_TX_PDEV_MAX_SIFS_BURST_HIST_STATS);
+
+	len += scnprintf(buf + len, buf_len - len,
+			 "HTT_TX_PDEV_STATS_SIFS_HIST_TLV:\n");
+
+	len += print_array_to_buf(buf, len, "sifs_hist_status",
+				  htt_stats_buf->sifs_hist_status, num_elems, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
+static void
+htt_print_pdev_ctrl_path_tx_stats_tlv(const void *tag_buf, u16 tag_len,
+				      struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_pdev_ctrl_path_tx_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+
+	if (len < sizeof(*htt_stats_buf))
+		return;
+
+	len += scnprintf(buf + len, buf_len - len,
+			 "HTT_TX_PDEV_STATS_CTRL_PATH_TX_STATS:\n");
+	len += print_array_to_buf(buf, len, "fw_tx_mgmt_subtype",
+				 htt_stats_buf->fw_tx_mgmt_subtype,
+				 ATH12K_HTT_STATS_SUBTYPE_MAX, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
 static int ath12k_dbg_htt_ext_stats_parse(struct ath12k_base *ab,
 					  u16 tag, u16 len, const void *tag_buf,
 					  void *user_data)
@@ -272,6 +405,15 @@ static int ath12k_dbg_htt_ext_stats_parse(struct ath12k_base *ab,
 	case HTT_STATS_TX_PDEV_FLUSH_TAG:
 		htt_print_tx_pdev_stats_flush_tlv(tag_buf, len, stats_req);
 		break;
+	case HTT_STATS_TX_PDEV_SIFS_HIST_TAG:
+		htt_print_tx_pdev_stats_sifs_hist_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_PDEV_CTRL_PATH_TX_STATS_TAG:
+		htt_print_pdev_ctrl_path_tx_stats_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_MU_PPDU_DIST_TAG:
+		htt_print_tx_pdev_mu_ppdu_dist_stats_tlv(tag_buf, len, stats_req);
+		break;
 	default:
 		break;
 	}
diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
index 83060c54d00b..878d9cf6e7cb 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
@@ -11,6 +11,8 @@
 #define ATH12K_HTT_STATS_COOKIE_LSB		GENMASK_ULL(31, 0)
 #define ATH12K_HTT_STATS_COOKIE_MSB		GENMASK_ULL(63, 32)
 #define ATH12K_HTT_STATS_MAGIC_VALUE		0xF0F0F0F0
+#define ATH12K_HTT_STATS_SUBTYPE_MAX		16
+#define ATH12K_HTT_MAX_STRING_LEN		256
 
 #define ATH12K_HTT_STATS_RESET_BITMAP32_OFFSET(_idx)	((_idx) & 0x1f)
 #define ATH12K_HTT_STATS_RESET_BITMAP64_OFFSET(_idx)	((_idx) & 0x3f)
@@ -133,6 +135,9 @@ enum ath12k_dbg_htt_tlv_tag {
 	HTT_STATS_TX_PDEV_UNDERRUN_TAG			= 1,
 	HTT_STATS_TX_PDEV_SIFS_TAG			= 2,
 	HTT_STATS_TX_PDEV_FLUSH_TAG			= 3,
+	HTT_STATS_TX_PDEV_SIFS_HIST_TAG			= 67,
+	HTT_STATS_PDEV_CTRL_PATH_TX_STATS_TAG		= 102,
+	HTT_STATS_MU_PPDU_DIST_TAG			= 129,
 
 	HTT_STATS_MAX_TAG,
 };
@@ -142,6 +147,18 @@ enum ath12k_dbg_htt_tlv_tag {
 #define ATH12K_HTT_TX_PDEV_MAX_SIFS_BURST_STATS		9
 #define ATH12K_HTT_TX_PDEV_MAX_FLUSH_REASON_STATS	150
 
+/* MU MIMO distribution stats is a 2-dimensional array
+ * with dimension one denoting stats for nr4[0] or nr8[1]
+ */
+#define ATH12K_HTT_STATS_NUM_NR_BINS			2
+#define ATH12K_HTT_STATS_MAX_NUM_MU_PPDU_PER_BURST	10
+#define ATH12K_HTT_TX_PDEV_MAX_SIFS_BURST_HIST_STATS	10
+#define ATH12K_HTT_STATS_MAX_NUM_SCHED_STATUS		9
+#define ATH12K_HTT_STATS_NUM_SCHED_STATUS_WORDS		\
+	(ATH12K_HTT_STATS_NUM_NR_BINS * ATH12K_HTT_STATS_MAX_NUM_SCHED_STATUS)
+#define ATH12K_HTT_STATS_MU_PPDU_PER_BURST_WORDS	\
+	(ATH12K_HTT_STATS_NUM_NR_BINS * ATH12K_HTT_STATS_MAX_NUM_MU_PPDU_PER_BURST)
+
 enum ath12k_htt_tx_pdev_underrun_enum {
 	HTT_STATS_TX_PDEV_NO_DATA_UNDERRUN		= 0,
 	HTT_STATS_TX_PDEV_DATA_UNDERRUN_BETWEEN_MPDU	= 1,
@@ -258,4 +275,26 @@ struct ath12k_htt_tx_pdev_stats_sifs_tlv {
 	DECLARE_FLEX_ARRAY(__le32, sifs_status);
 } __packed;
 
+struct ath12k_htt_pdev_ctrl_path_tx_stats_tlv {
+	__le32 fw_tx_mgmt_subtype[ATH12K_HTT_STATS_SUBTYPE_MAX];
+} __packed;
+
+struct ath12k_htt_tx_pdev_stats_sifs_hist_tlv {
+	DECLARE_FLEX_ARRAY(__le32, sifs_hist_status);
+} __packed;
+
+enum ath12k_htt_stats_hw_mode {
+	ATH12K_HTT_STATS_HWMODE_AC = 0,
+	ATH12K_HTT_STATS_HWMODE_AX = 1,
+	ATH12K_HTT_STATS_HWMODE_BE = 2,
+};
+
+struct ath12k_htt_tx_pdev_mu_ppdu_dist_stats_tlv {
+	__le32 hw_mode;
+	__le32 num_seq_term_status[ATH12K_HTT_STATS_NUM_SCHED_STATUS_WORDS];
+	__le32 num_ppdu_cmpl_per_burst[ATH12K_HTT_STATS_MU_PPDU_PER_BURST_WORDS];
+	__le32 num_seq_posted[ATH12K_HTT_STATS_NUM_NR_BINS];
+	__le32 num_ppdu_posted_per_burst[ATH12K_HTT_STATS_MU_PPDU_PER_BURST_WORDS];
+} __packed;
+
 #endif
-- 
cgit v1.2.3-58-ga151


From 4b6dfc758c2f6ce61336637b22bc0db57fa6d649 Mon Sep 17 00:00:00 2001
From: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Date: Tue, 25 Jun 2024 09:52:15 +0530
Subject: wifi: ath12k: Support Transmit Scheduler stats

Add support to request transmission scheduler stats from firmware through
HTT stats type 4. This stat gives information such as pdev stats
scheduled per transmission queue, transmission queue command posted and
reaped, scheduler order, scheduler ineligibility and supercycle triggers.

Sample output:
-------------
echo 4 > /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats_type
cat /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats
HTT_STATS_TX_SCHED_CMN_TLV:
mac_id = 0
current_timestamp = 952546828

HTT_TX_PDEV_STATS_SCHED_PER_TXQ_TLV:
mac_id = 0
txq_id = 14
sched_policy = 2
.....

HTT_SCHED_TXQ_CMD_POSTED_TLV:
sched_cmd_posted =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
10:0, 11:0, 12:0

HTT_SCHED_TXQ_CMD_REAPED_TLV:
sched_cmd_reaped =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
10:0, 11:0, 12:0

HTT_SCHED_TXQ_SCHED_ORDER_SU_TLV:
sched_order_su =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0, 10:0,
11:0, 12:0, 13:0, 14:0, 15:0, 16:0, 17:0, 18:0, 19:0

HTT_SCHED_TXQ_SCHED_INELIGIBILITY:
sched_ineligibility =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
10:0, 11:0, 12:0, 13:0, 14:0, 15:0, 16:0, 17:0, 18:0, 19:0, 20:0, 21:0,
22:0, 23:0, 24:0, 25:0, 26:0, 27:0, 28:0, 29:0, 30:0, 31:0, 32:0, 33:0,
34:0, 35:0

HTT_SCHED_TXQ_SUPERCYCLE_TRIGGER:
supercycle_triggers =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Signed-off-by: Roopni Devanathan <quic_rdevanat@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240625042217.1303332-2-quic_rdevanat@quicinc.com
---
 .../net/wireless/ath/ath12k/debugfs_htt_stats.c    | 233 +++++++++++++++++++++
 .../net/wireless/ath/ath12k/debugfs_htt_stats.h    |  81 +++++++
 2 files changed, 314 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
index fd04f057598d..586f002df15e 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
@@ -386,6 +386,216 @@ htt_print_pdev_ctrl_path_tx_stats_tlv(const void *tag_buf, u16 tag_len,
 	stats_req->buf_len = len;
 }
 
+static void
+ath12k_htt_print_stats_tx_sched_cmn_tlv(const void *tag_buf,
+					u16 tag_len,
+					struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_stats_tx_sched_cmn_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u32 mac_id_word;
+
+	if (tag_len < sizeof(*htt_stats_buf))
+		return;
+
+	mac_id_word = __le32_to_cpu(htt_stats_buf->mac_id__word);
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_STATS_TX_SCHED_CMN_TLV:\n");
+	len += scnprintf(buf + len, buf_len - len, "mac_id = %u\n",
+			 u32_get_bits(mac_id_word, ATH12K_HTT_STATS_MAC_ID));
+	len += scnprintf(buf + len, buf_len - len, "current_timestamp = %u\n\n",
+			 le32_to_cpu(htt_stats_buf->current_timestamp));
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_tx_pdev_stats_sched_per_txq_tlv(const void *tag_buf,
+						 u16 tag_len,
+						 struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_pdev_stats_sched_per_txq_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u32 mac_id_word;
+
+	if (tag_len < sizeof(*htt_stats_buf))
+		return;
+
+	mac_id_word = __le32_to_cpu(htt_stats_buf->mac_id__word);
+
+	len += scnprintf(buf + len, buf_len - len,
+			 "HTT_TX_PDEV_STATS_SCHED_PER_TXQ_TLV:\n");
+	len += scnprintf(buf + len, buf_len - len, "mac_id = %u\n",
+			u32_get_bits(mac_id_word,
+				     ATH12K_HTT_TX_PDEV_STATS_SCHED_PER_TXQ_MAC_ID));
+	len += scnprintf(buf + len, buf_len - len, "txq_id = %u\n",
+			 u32_get_bits(mac_id_word,
+				      ATH12K_HTT_TX_PDEV_STATS_SCHED_PER_TXQ_ID));
+	len += scnprintf(buf + len, buf_len - len, "sched_policy = %u\n",
+			 le32_to_cpu(htt_stats_buf->sched_policy));
+	len += scnprintf(buf + len, buf_len - len,
+			 "last_sched_cmd_posted_timestamp = %u\n",
+			 le32_to_cpu(htt_stats_buf->last_sched_cmd_posted_timestamp));
+	len += scnprintf(buf + len, buf_len - len,
+			 "last_sched_cmd_compl_timestamp = %u\n",
+			 le32_to_cpu(htt_stats_buf->last_sched_cmd_compl_timestamp));
+	len += scnprintf(buf + len, buf_len - len, "sched_2_tac_lwm_count = %u\n",
+			 le32_to_cpu(htt_stats_buf->sched_2_tac_lwm_count));
+	len += scnprintf(buf + len, buf_len - len, "sched_2_tac_ring_full = %u\n",
+			 le32_to_cpu(htt_stats_buf->sched_2_tac_ring_full));
+	len += scnprintf(buf + len, buf_len - len, "sched_cmd_post_failure = %u\n",
+			 le32_to_cpu(htt_stats_buf->sched_cmd_post_failure));
+	len += scnprintf(buf + len, buf_len - len, "num_active_tids = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_active_tids));
+	len += scnprintf(buf + len, buf_len - len, "num_ps_schedules = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_ps_schedules));
+	len += scnprintf(buf + len, buf_len - len, "sched_cmds_pending = %u\n",
+			 le32_to_cpu(htt_stats_buf->sched_cmds_pending));
+	len += scnprintf(buf + len, buf_len - len, "num_tid_register = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_tid_register));
+	len += scnprintf(buf + len, buf_len - len, "num_tid_unregister = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_tid_unregister));
+	len += scnprintf(buf + len, buf_len - len, "num_qstats_queried = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_qstats_queried));
+	len += scnprintf(buf + len, buf_len - len, "qstats_update_pending = %u\n",
+			 le32_to_cpu(htt_stats_buf->qstats_update_pending));
+	len += scnprintf(buf + len, buf_len - len, "last_qstats_query_timestamp = %u\n",
+			 le32_to_cpu(htt_stats_buf->last_qstats_query_timestamp));
+	len += scnprintf(buf + len, buf_len - len, "num_tqm_cmdq_full = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_tqm_cmdq_full));
+	len += scnprintf(buf + len, buf_len - len, "num_de_sched_algo_trigger = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_de_sched_algo_trigger));
+	len += scnprintf(buf + len, buf_len - len, "num_rt_sched_algo_trigger = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_rt_sched_algo_trigger));
+	len += scnprintf(buf + len, buf_len - len, "num_tqm_sched_algo_trigger = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_tqm_sched_algo_trigger));
+	len += scnprintf(buf + len, buf_len - len, "notify_sched = %u\n",
+			 le32_to_cpu(htt_stats_buf->notify_sched));
+	len += scnprintf(buf + len, buf_len - len, "dur_based_sendn_term = %u\n",
+			 le32_to_cpu(htt_stats_buf->dur_based_sendn_term));
+	len += scnprintf(buf + len, buf_len - len, "su_notify2_sched = %u\n",
+			 le32_to_cpu(htt_stats_buf->su_notify2_sched));
+	len += scnprintf(buf + len, buf_len - len, "su_optimal_queued_msdus_sched = %u\n",
+			 le32_to_cpu(htt_stats_buf->su_optimal_queued_msdus_sched));
+	len += scnprintf(buf + len, buf_len - len, "su_delay_timeout_sched = %u\n",
+			 le32_to_cpu(htt_stats_buf->su_delay_timeout_sched));
+	len += scnprintf(buf + len, buf_len - len, "su_min_txtime_sched_delay = %u\n",
+			 le32_to_cpu(htt_stats_buf->su_min_txtime_sched_delay));
+	len += scnprintf(buf + len, buf_len - len, "su_no_delay = %u\n",
+			 le32_to_cpu(htt_stats_buf->su_no_delay));
+	len += scnprintf(buf + len, buf_len - len, "num_supercycles = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_supercycles));
+	len += scnprintf(buf + len, buf_len - len, "num_subcycles_with_sort = %u\n",
+			 le32_to_cpu(htt_stats_buf->num_subcycles_with_sort));
+	len += scnprintf(buf + len, buf_len - len, "num_subcycles_no_sort = %u\n\n",
+			 le32_to_cpu(htt_stats_buf->num_subcycles_no_sort));
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_sched_txq_cmd_posted_tlv(const void *tag_buf,
+					  u16 tag_len,
+					  struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_sched_txq_cmd_posted_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u16 num_elements = tag_len >> 2;
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_SCHED_TXQ_CMD_POSTED_TLV:\n");
+	len += print_array_to_buf(buf, len, "sched_cmd_posted",
+				  htt_stats_buf->sched_cmd_posted, num_elements, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_sched_txq_cmd_reaped_tlv(const void *tag_buf,
+					  u16 tag_len,
+					  struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_sched_txq_cmd_reaped_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u16 num_elements = tag_len >> 2;
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_SCHED_TXQ_CMD_REAPED_TLV:\n");
+	len += print_array_to_buf(buf, len, "sched_cmd_reaped",
+				  htt_stats_buf->sched_cmd_reaped, num_elements, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_sched_txq_sched_order_su_tlv(const void *tag_buf,
+					      u16 tag_len,
+					      struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_sched_txq_sched_order_su_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u32 sched_order_su_num_entries = min_t(u32, (tag_len >> 2),
+					       ATH12K_HTT_TX_PDEV_NUM_SCHED_ORDER_LOG);
+
+	len += scnprintf(buf + len, buf_len - len,
+			 "HTT_SCHED_TXQ_SCHED_ORDER_SU_TLV:\n");
+	len += print_array_to_buf(buf, len, "sched_order_su",
+				  htt_stats_buf->sched_order_su,
+				  sched_order_su_num_entries, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_sched_txq_sched_ineligibility_tlv(const void *tag_buf,
+						   u16 tag_len,
+						   struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_sched_txq_sched_ineligibility_tlv *htt_stats_buf =
+		     tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u32 sched_ineligibility_num_entries = tag_len >> 2;
+
+	len += scnprintf(buf + len, buf_len - len,
+			 "HTT_SCHED_TXQ_SCHED_INELIGIBILITY:\n");
+	len += print_array_to_buf(buf, len, "sched_ineligibility",
+				  htt_stats_buf->sched_ineligibility,
+				  sched_ineligibility_num_entries, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_sched_txq_supercycle_trigger_tlv(const void *tag_buf,
+						  u16 tag_len,
+						  struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_sched_txq_supercycle_triggers_tlv *htt_stats_buf =
+		     tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u16 num_elems = min_t(u16, (tag_len >> 2),
+			      ATH12K_HTT_SCHED_SUPERCYCLE_TRIGGER_MAX);
+
+	len += scnprintf(buf + len, buf_len - len,
+			 "HTT_SCHED_TXQ_SUPERCYCLE_TRIGGER:\n");
+	len += print_array_to_buf(buf, len, "supercycle_triggers",
+				  htt_stats_buf->supercycle_triggers, num_elems, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
 static int ath12k_dbg_htt_ext_stats_parse(struct ath12k_base *ab,
 					  u16 tag, u16 len, const void *tag_buf,
 					  void *user_data)
@@ -414,6 +624,29 @@ static int ath12k_dbg_htt_ext_stats_parse(struct ath12k_base *ab,
 	case HTT_STATS_MU_PPDU_DIST_TAG:
 		htt_print_tx_pdev_mu_ppdu_dist_stats_tlv(tag_buf, len, stats_req);
 		break;
+	case HTT_STATS_TX_SCHED_CMN_TAG:
+		ath12k_htt_print_stats_tx_sched_cmn_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_PDEV_SCHEDULER_TXQ_STATS_TAG:
+		ath12k_htt_print_tx_pdev_stats_sched_per_txq_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_SCHED_TXQ_CMD_POSTED_TAG:
+		ath12k_htt_print_sched_txq_cmd_posted_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_SCHED_TXQ_CMD_REAPED_TAG:
+		ath12k_htt_print_sched_txq_cmd_reaped_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_SCHED_TXQ_SCHED_ORDER_SU_TAG:
+		ath12k_htt_print_sched_txq_sched_order_su_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_SCHED_TXQ_SCHED_INELIGIBILITY_TAG:
+		ath12k_htt_print_sched_txq_sched_ineligibility_tlv(tag_buf, len,
+								   stats_req);
+		break;
+	case HTT_STATS_SCHED_TXQ_SUPERCYCLE_TRIGGER_TAG:
+		ath12k_htt_print_sched_txq_supercycle_trigger_tlv(tag_buf, len,
+								  stats_req);
+		break;
 	default:
 		break;
 	}
diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
index 878d9cf6e7cb..ee90d620fc1d 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
@@ -125,6 +125,7 @@ struct ath12k_htt_extd_stats_msg {
 enum ath12k_dbg_htt_ext_stats_type {
 	ATH12K_DBG_HTT_EXT_STATS_RESET		= 0,
 	ATH12K_DBG_HTT_EXT_STATS_PDEV_TX	= 1,
+	ATH12K_DBG_HTT_EXT_STATS_PDEV_TX_SCHED	= 4,
 
 	/* keep this last */
 	ATH12K_DBG_HTT_NUM_EXT_STATS,
@@ -135,7 +136,14 @@ enum ath12k_dbg_htt_tlv_tag {
 	HTT_STATS_TX_PDEV_UNDERRUN_TAG			= 1,
 	HTT_STATS_TX_PDEV_SIFS_TAG			= 2,
 	HTT_STATS_TX_PDEV_FLUSH_TAG			= 3,
+	HTT_STATS_TX_PDEV_SCHEDULER_TXQ_STATS_TAG	= 36,
+	HTT_STATS_TX_SCHED_CMN_TAG			= 37,
+	HTT_STATS_SCHED_TXQ_CMD_POSTED_TAG		= 39,
+	HTT_STATS_SCHED_TXQ_CMD_REAPED_TAG		= 44,
 	HTT_STATS_TX_PDEV_SIFS_HIST_TAG			= 67,
+	HTT_STATS_SCHED_TXQ_SCHED_ORDER_SU_TAG		= 86,
+	HTT_STATS_SCHED_TXQ_SCHED_INELIGIBILITY_TAG	= 87,
+	HTT_STATS_SCHED_TXQ_SUPERCYCLE_TRIGGER_TAG	= 100,
 	HTT_STATS_PDEV_CTRL_PATH_TX_STATS_TAG		= 102,
 	HTT_STATS_MU_PPDU_DIST_TAG			= 129,
 
@@ -297,4 +305,77 @@ struct ath12k_htt_tx_pdev_mu_ppdu_dist_stats_tlv {
 	__le32 num_ppdu_posted_per_burst[ATH12K_HTT_STATS_MU_PPDU_PER_BURST_WORDS];
 } __packed;
 
+#define ATH12K_HTT_TX_PDEV_STATS_SCHED_PER_TXQ_MAC_ID	GENMASK(7, 0)
+#define ATH12K_HTT_TX_PDEV_STATS_SCHED_PER_TXQ_ID	GENMASK(15, 8)
+
+#define ATH12K_HTT_TX_PDEV_NUM_SCHED_ORDER_LOG	20
+
+struct ath12k_htt_stats_tx_sched_cmn_tlv {
+	__le32 mac_id__word;
+	__le32 current_timestamp;
+} __packed;
+
+struct ath12k_htt_tx_pdev_stats_sched_per_txq_tlv {
+	__le32 mac_id__word;
+	__le32 sched_policy;
+	__le32 last_sched_cmd_posted_timestamp;
+	__le32 last_sched_cmd_compl_timestamp;
+	__le32 sched_2_tac_lwm_count;
+	__le32 sched_2_tac_ring_full;
+	__le32 sched_cmd_post_failure;
+	__le32 num_active_tids;
+	__le32 num_ps_schedules;
+	__le32 sched_cmds_pending;
+	__le32 num_tid_register;
+	__le32 num_tid_unregister;
+	__le32 num_qstats_queried;
+	__le32 qstats_update_pending;
+	__le32 last_qstats_query_timestamp;
+	__le32 num_tqm_cmdq_full;
+	__le32 num_de_sched_algo_trigger;
+	__le32 num_rt_sched_algo_trigger;
+	__le32 num_tqm_sched_algo_trigger;
+	__le32 notify_sched;
+	__le32 dur_based_sendn_term;
+	__le32 su_notify2_sched;
+	__le32 su_optimal_queued_msdus_sched;
+	__le32 su_delay_timeout_sched;
+	__le32 su_min_txtime_sched_delay;
+	__le32 su_no_delay;
+	__le32 num_supercycles;
+	__le32 num_subcycles_with_sort;
+	__le32 num_subcycles_no_sort;
+} __packed;
+
+struct ath12k_htt_sched_txq_cmd_posted_tlv {
+	DECLARE_FLEX_ARRAY(__le32, sched_cmd_posted);
+} __packed;
+
+struct ath12k_htt_sched_txq_cmd_reaped_tlv {
+	DECLARE_FLEX_ARRAY(__le32, sched_cmd_reaped);
+} __packed;
+
+struct ath12k_htt_sched_txq_sched_order_su_tlv {
+	DECLARE_FLEX_ARRAY(__le32, sched_order_su);
+} __packed;
+
+struct ath12k_htt_sched_txq_sched_ineligibility_tlv {
+	DECLARE_FLEX_ARRAY(__le32, sched_ineligibility);
+} __packed;
+
+enum ath12k_htt_sched_txq_supercycle_triggers_tlv_enum {
+	ATH12K_HTT_SCHED_SUPERCYCLE_TRIGGER_NONE = 0,
+	ATH12K_HTT_SCHED_SUPERCYCLE_TRIGGER_FORCED,
+	ATH12K_HTT_SCHED_SUPERCYCLE_TRIGGER_LESS_NUM_TIDQ_ENTRIES,
+	ATH12K_HTT_SCHED_SUPERCYCLE_TRIGGER_LESS_NUM_ACTIVE_TIDS,
+	ATH12K_HTT_SCHED_SUPERCYCLE_TRIGGER_MAX_ITR_REACHED,
+	ATH12K_HTT_SCHED_SUPERCYCLE_TRIGGER_DUR_THRESHOLD_REACHED,
+	ATH12K_HTT_SCHED_SUPERCYCLE_TRIGGER_TWT_TRIGGER,
+	ATH12K_HTT_SCHED_SUPERCYCLE_TRIGGER_MAX,
+};
+
+struct ath12k_htt_sched_txq_supercycle_triggers_tlv {
+	DECLARE_FLEX_ARRAY(__le32, supercycle_triggers);
+} __packed;
+
 #endif
-- 
cgit v1.2.3-58-ga151


From cf1187b3a683feb718a50ee19178dc6e90a73a64 Mon Sep 17 00:00:00 2001
From: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Date: Tue, 25 Jun 2024 09:52:16 +0530
Subject: wifi: ath12k: Support pdev error stats

Add support to request pdev error stats from firmware through HTT stats
type 5. This stats type gives information such as list of pdev errors,
reasons for reset, logging information about recovery, drain and war
stats. etc.

Sample output:
-------------
echo 5 > /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats_type
cat /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats

HTT_HW_STATS_PDEV_ERRS_TLV:
mac_id = 0
tx_abort = 0
tx_abort_fail_count = 0
rx_abort = 53
.....

PDEV_PHY_WARM_RESET_REASONS:
phy_warm_reset_reason_phy_m3 = 0
phy_warm_reset_reason_tx_hw_stuck = 0
phy_warm_reset_reason_num_cca_rx_frame_stuck = 0
.....

WAL_RX_RECOVERY_STATS:
wal_rx_recovery_rst_mac_hang_count = 0
wal_rx_recovery_rst_known_sig_count = 0
wal_rx_recovery_rst_no_rx_count = 0
.....

HTT_RX_DEST_DRAIN_STATS:
rx_dest_drain_rx_descs_leak_prevention_done = 0
rx_dest_drain_rx_descs_saved_cnt = 0
rx_dest_drain_rxdma2reo_leak_detected = 0
.....

HTT_HW_STATS_INTR_MISC_TLV:
hw_intr_name = AMPI
mask = 0
count = 2

HTT_HW_STATS_WHAL_TX_TLV:
mac_id = 0
last_unpause_ppdu_id = 0
hwsch_unpause_wait_tqm_write = 0
.....

HTT_HW_WAR_STATS_TLV:
mac_id = 0
hw_war 0 = 0
hw_war 1 = 0
.....

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Signed-off-by: Roopni Devanathan <quic_rdevanat@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240625042217.1303332-3-quic_rdevanat@quicinc.com
---
 .../net/wireless/ath/ath12k/debugfs_htt_stats.c    | 262 +++++++++++++++++++++
 .../net/wireless/ath/ath12k/debugfs_htt_stats.h    |  84 +++++++
 2 files changed, 346 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
index 586f002df15e..05bb0ea68e15 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
@@ -596,6 +596,256 @@ ath12k_htt_print_sched_txq_supercycle_trigger_tlv(const void *tag_buf,
 	stats_req->buf_len = len;
 }
 
+static void
+ath12k_htt_print_hw_stats_pdev_errs_tlv(const void *tag_buf, u16 tag_len,
+					struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_hw_stats_pdev_errs_tlv *htt_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u32 mac_id_word;
+
+	if (tag_len < sizeof(*htt_buf))
+		return;
+
+	mac_id_word = le32_to_cpu(htt_buf->mac_id__word);
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_HW_STATS_PDEV_ERRS_TLV:\n");
+	len += scnprintf(buf + len, buf_len - len, "mac_id = %u\n",
+			 u32_get_bits(mac_id_word, ATH12K_HTT_STATS_MAC_ID));
+	len += scnprintf(buf + len, buf_len - len, "tx_abort = %u\n",
+			 le32_to_cpu(htt_buf->tx_abort));
+	len += scnprintf(buf + len, buf_len - len, "tx_abort_fail_count = %u\n",
+			 le32_to_cpu(htt_buf->tx_abort_fail_count));
+	len += scnprintf(buf + len, buf_len - len, "rx_abort = %u\n",
+			 le32_to_cpu(htt_buf->rx_abort));
+	len += scnprintf(buf + len, buf_len - len, "rx_abort_fail_count = %u\n",
+			 le32_to_cpu(htt_buf->rx_abort_fail_count));
+	len += scnprintf(buf + len, buf_len - len, "rx_flush_cnt = %u\n",
+			 le32_to_cpu(htt_buf->rx_flush_cnt));
+	len += scnprintf(buf + len, buf_len - len, "warm_reset = %u\n",
+			 le32_to_cpu(htt_buf->warm_reset));
+	len += scnprintf(buf + len, buf_len - len, "cold_reset = %u\n",
+			 le32_to_cpu(htt_buf->cold_reset));
+	len += scnprintf(buf + len, buf_len - len, "mac_cold_reset_restore_cal = %u\n",
+			 le32_to_cpu(htt_buf->mac_cold_reset_restore_cal));
+	len += scnprintf(buf + len, buf_len - len, "mac_cold_reset = %u\n",
+			 le32_to_cpu(htt_buf->mac_cold_reset));
+	len += scnprintf(buf + len, buf_len - len, "mac_warm_reset = %u\n",
+			 le32_to_cpu(htt_buf->mac_warm_reset));
+	len += scnprintf(buf + len, buf_len - len, "mac_only_reset = %u\n",
+			 le32_to_cpu(htt_buf->mac_only_reset));
+	len += scnprintf(buf + len, buf_len - len, "phy_warm_reset = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset));
+	len += scnprintf(buf + len, buf_len - len, "phy_warm_reset_ucode_trig = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_ucode_trig));
+	len += scnprintf(buf + len, buf_len - len, "mac_warm_reset_restore_cal = %u\n",
+			 le32_to_cpu(htt_buf->mac_warm_reset_restore_cal));
+	len += scnprintf(buf + len, buf_len - len, "mac_sfm_reset = %u\n",
+			 le32_to_cpu(htt_buf->mac_sfm_reset));
+	len += scnprintf(buf + len, buf_len - len, "phy_warm_reset_m3_ssr = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_m3_ssr));
+	len += scnprintf(buf + len, buf_len - len, "fw_rx_rings_reset = %u\n",
+			 le32_to_cpu(htt_buf->fw_rx_rings_reset));
+	len += scnprintf(buf + len, buf_len - len, "tx_flush = %u\n",
+			 le32_to_cpu(htt_buf->tx_flush));
+	len += scnprintf(buf + len, buf_len - len, "tx_glb_reset = %u\n",
+			 le32_to_cpu(htt_buf->tx_glb_reset));
+	len += scnprintf(buf + len, buf_len - len, "tx_txq_reset = %u\n",
+			 le32_to_cpu(htt_buf->tx_txq_reset));
+	len += scnprintf(buf + len, buf_len - len, "rx_timeout_reset = %u\n\n",
+			 le32_to_cpu(htt_buf->rx_timeout_reset));
+
+	len += scnprintf(buf + len, buf_len - len, "PDEV_PHY_WARM_RESET_REASONS:\n");
+	len += scnprintf(buf + len, buf_len - len, "phy_warm_reset_reason_phy_m3 = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_reason_phy_m3));
+	len += scnprintf(buf + len, buf_len - len,
+			 "phy_warm_reset_reason_tx_hw_stuck = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_reason_tx_hw_stuck));
+	len += scnprintf(buf + len, buf_len - len,
+			 "phy_warm_reset_reason_num_cca_rx_frame_stuck = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_reason_num_rx_frame_stuck));
+	len += scnprintf(buf + len, buf_len - len,
+			 "phy_warm_reset_reason_wal_rx_recovery_rst_rx_busy = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_reason_wal_rx_rec_rx_busy));
+	len += scnprintf(buf + len, buf_len - len,
+			 "phy_warm_reset_reason_wal_rx_recovery_rst_mac_hang = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_reason_wal_rx_rec_mac_hng));
+	len += scnprintf(buf + len, buf_len - len,
+			 "phy_warm_reset_reason_mac_reset_converted_phy_reset = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_reason_mac_conv_phy_reset));
+	len += scnprintf(buf + len, buf_len - len,
+			 "phy_warm_reset_reason_tx_lifetime_expiry_cca_stuck = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_reason_tx_exp_cca_stuck));
+	len += scnprintf(buf + len, buf_len - len,
+			 "phy_warm_reset_reason_tx_consecutive_flush9_war = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_reason_tx_consec_flsh_war));
+	len += scnprintf(buf + len, buf_len - len,
+			 "phy_warm_reset_reason_tx_hwsch_reset_war = %u\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_reason_tx_hwsch_reset_war));
+	len += scnprintf(buf + len, buf_len - len,
+			 "phy_warm_reset_reason_hwsch_wdog_or_cca_wdog_war = %u\n\n",
+			 le32_to_cpu(htt_buf->phy_warm_reset_reason_hwsch_cca_wdog_war));
+
+	len += scnprintf(buf + len, buf_len - len, "WAL_RX_RECOVERY_STATS:\n");
+	len += scnprintf(buf + len, buf_len - len,
+			 "wal_rx_recovery_rst_mac_hang_count = %u\n",
+			 le32_to_cpu(htt_buf->wal_rx_recovery_rst_mac_hang_cnt));
+	len += scnprintf(buf + len, buf_len - len,
+			 "wal_rx_recovery_rst_known_sig_count = %u\n",
+			 le32_to_cpu(htt_buf->wal_rx_recovery_rst_known_sig_cnt));
+	len += scnprintf(buf + len, buf_len - len,
+			 "wal_rx_recovery_rst_no_rx_count = %u\n",
+			 le32_to_cpu(htt_buf->wal_rx_recovery_rst_no_rx_cnt));
+	len += scnprintf(buf + len, buf_len - len,
+			 "wal_rx_recovery_rst_no_rx_consecutive_count = %u\n",
+			 le32_to_cpu(htt_buf->wal_rx_recovery_rst_no_rx_consec_cnt));
+	len += scnprintf(buf + len, buf_len - len,
+			 "wal_rx_recovery_rst_rx_busy_count = %u\n",
+			 le32_to_cpu(htt_buf->wal_rx_recovery_rst_rx_busy_cnt));
+	len += scnprintf(buf + len, buf_len - len,
+			 "wal_rx_recovery_rst_phy_mac_hang_count = %u\n\n",
+			 le32_to_cpu(htt_buf->wal_rx_recovery_rst_phy_mac_hang_cnt));
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_RX_DEST_DRAIN_STATS:\n");
+	len += scnprintf(buf + len, buf_len - len,
+			 "rx_dest_drain_rx_descs_leak_prevention_done = %u\n",
+			 le32_to_cpu(htt_buf->rx_dest_drain_rx_descs_leak_prevented));
+	len += scnprintf(buf + len, buf_len - len,
+			 "rx_dest_drain_rx_descs_saved_cnt = %u\n",
+			 le32_to_cpu(htt_buf->rx_dest_drain_rx_descs_saved_cnt));
+	len += scnprintf(buf + len, buf_len - len,
+			 "rx_dest_drain_rxdma2reo_leak_detected = %u\n",
+			 le32_to_cpu(htt_buf->rx_dest_drain_rxdma2reo_leak_detected));
+	len += scnprintf(buf + len, buf_len - len,
+			 "rx_dest_drain_rxdma2fw_leak_detected = %u\n",
+			 le32_to_cpu(htt_buf->rx_dest_drain_rxdma2fw_leak_detected));
+	len += scnprintf(buf + len, buf_len - len,
+			 "rx_dest_drain_rxdma2wbm_leak_detected = %u\n",
+			 le32_to_cpu(htt_buf->rx_dest_drain_rxdma2wbm_leak_detected));
+	len += scnprintf(buf + len, buf_len - len,
+			 "rx_dest_drain_rxdma1_2sw_leak_detected = %u\n",
+			 le32_to_cpu(htt_buf->rx_dest_drain_rxdma1_2sw_leak_detected));
+	len += scnprintf(buf + len, buf_len - len,
+			 "rx_dest_drain_rx_drain_ok_mac_idle = %u\n",
+			 le32_to_cpu(htt_buf->rx_dest_drain_rx_drain_ok_mac_idle));
+	len += scnprintf(buf + len, buf_len - len,
+			 "rx_dest_drain_ok_mac_not_idle = %u\n",
+			 le32_to_cpu(htt_buf->rx_dest_drain_ok_mac_not_idle));
+	len += scnprintf(buf + len, buf_len - len,
+			 "rx_dest_drain_prerequisite_invld = %u\n",
+			 le32_to_cpu(htt_buf->rx_dest_drain_prerequisite_invld));
+	len += scnprintf(buf + len, buf_len - len,
+			 "rx_dest_drain_skip_for_non_lmac_reset = %u\n",
+			 le32_to_cpu(htt_buf->rx_dest_drain_skip_non_lmac_reset));
+	len += scnprintf(buf + len, buf_len - len,
+			 "rx_dest_drain_hw_fifo_not_empty_post_drain_wait = %u\n\n",
+			 le32_to_cpu(htt_buf->rx_dest_drain_hw_fifo_notempty_post_wait));
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_hw_stats_intr_misc_tlv(const void *tag_buf, u16 tag_len,
+					struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_hw_stats_intr_misc_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+
+	if (tag_len < sizeof(*htt_stats_buf))
+		return;
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_HW_STATS_INTR_MISC_TLV:\n");
+	len += scnprintf(buf + len, buf_len - len, "hw_intr_name = %s\n",
+			 htt_stats_buf->hw_intr_name);
+	len += scnprintf(buf + len, buf_len - len, "mask = %u\n",
+			 le32_to_cpu(htt_stats_buf->mask));
+	len += scnprintf(buf + len, buf_len - len, "count = %u\n\n",
+			 le32_to_cpu(htt_stats_buf->count));
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_hw_stats_whal_tx_tlv(const void *tag_buf, u16 tag_len,
+				      struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_hw_stats_whal_tx_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u32 mac_id_word;
+
+	if (tag_len < sizeof(*htt_stats_buf))
+		return;
+
+	mac_id_word = __le32_to_cpu(htt_stats_buf->mac_id__word);
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_HW_STATS_WHAL_TX_TLV:\n");
+	len += scnprintf(buf + len, buf_len - len, "mac_id = %u\n",
+			 u32_get_bits(mac_id_word, ATH12K_HTT_STATS_MAC_ID));
+	len += scnprintf(buf + len, buf_len - len, "last_unpause_ppdu_id = %u\n",
+			 le32_to_cpu(htt_stats_buf->last_unpause_ppdu_id));
+	len += scnprintf(buf + len, buf_len - len, "hwsch_unpause_wait_tqm_write = %u\n",
+			 le32_to_cpu(htt_stats_buf->hwsch_unpause_wait_tqm_write));
+	len += scnprintf(buf + len, buf_len - len, "hwsch_dummy_tlv_skipped = %u\n",
+			 le32_to_cpu(htt_stats_buf->hwsch_dummy_tlv_skipped));
+	len += scnprintf(buf + len, buf_len - len,
+			 "hwsch_misaligned_offset_received = %u\n",
+			 le32_to_cpu(htt_stats_buf->hwsch_misaligned_offset_received));
+	len += scnprintf(buf + len, buf_len - len, "hwsch_reset_count = %u\n",
+			 le32_to_cpu(htt_stats_buf->hwsch_reset_count));
+	len += scnprintf(buf + len, buf_len - len, "hwsch_dev_reset_war = %u\n",
+			 le32_to_cpu(htt_stats_buf->hwsch_dev_reset_war));
+	len += scnprintf(buf + len, buf_len - len, "hwsch_delayed_pause = %u\n",
+			 le32_to_cpu(htt_stats_buf->hwsch_delayed_pause));
+	len += scnprintf(buf + len, buf_len - len, "hwsch_long_delayed_pause = %u\n",
+			 le32_to_cpu(htt_stats_buf->hwsch_long_delayed_pause));
+	len += scnprintf(buf + len, buf_len - len, "sch_rx_ppdu_no_response = %u\n",
+			 le32_to_cpu(htt_stats_buf->sch_rx_ppdu_no_response));
+	len += scnprintf(buf + len, buf_len - len, "sch_selfgen_response = %u\n",
+			 le32_to_cpu(htt_stats_buf->sch_selfgen_response));
+	len += scnprintf(buf + len, buf_len - len, "sch_rx_sifs_resp_trigger= %u\n\n",
+			 le32_to_cpu(htt_stats_buf->sch_rx_sifs_resp_trigger));
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_hw_war_tlv(const void *tag_buf, u16 tag_len,
+			    struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_hw_war_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u16 fixed_len, array_len;
+	u8 i, array_words;
+	u32 mac_id;
+
+	if (tag_len < sizeof(*htt_stats_buf))
+		return;
+
+	mac_id = __le32_to_cpu(htt_stats_buf->mac_id__word);
+	fixed_len = sizeof(*htt_stats_buf);
+	array_len = tag_len - fixed_len;
+	array_words = array_len >> 2;
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_HW_WAR_STATS_TLV:\n");
+	len += scnprintf(buf + len, buf_len - len, "mac_id = %u\n",
+			 u32_get_bits(mac_id, ATH12K_HTT_STATS_MAC_ID));
+
+	for (i = 0; i < array_words; i++) {
+		len += scnprintf(buf + len, buf_len - len, "hw_war %u = %u\n\n",
+				 i, le32_to_cpu(htt_stats_buf->hw_wars[i]));
+	}
+
+	stats_req->buf_len = len;
+}
+
 static int ath12k_dbg_htt_ext_stats_parse(struct ath12k_base *ab,
 					  u16 tag, u16 len, const void *tag_buf,
 					  void *user_data)
@@ -647,6 +897,18 @@ static int ath12k_dbg_htt_ext_stats_parse(struct ath12k_base *ab,
 		ath12k_htt_print_sched_txq_supercycle_trigger_tlv(tag_buf, len,
 								  stats_req);
 		break;
+	case HTT_STATS_HW_PDEV_ERRS_TAG:
+		ath12k_htt_print_hw_stats_pdev_errs_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_HW_INTR_MISC_TAG:
+		ath12k_htt_print_hw_stats_intr_misc_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_WHAL_TX_TAG:
+		ath12k_htt_print_hw_stats_whal_tx_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_HW_WAR_TAG:
+		ath12k_htt_print_hw_war_tlv(tag_buf, len, stats_req);
+		break;
 	default:
 		break;
 	}
diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
index ee90d620fc1d..777a2e067a04 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
@@ -126,6 +126,7 @@ enum ath12k_dbg_htt_ext_stats_type {
 	ATH12K_DBG_HTT_EXT_STATS_RESET		= 0,
 	ATH12K_DBG_HTT_EXT_STATS_PDEV_TX	= 1,
 	ATH12K_DBG_HTT_EXT_STATS_PDEV_TX_SCHED	= 4,
+	ATH12K_DBG_HTT_EXT_STATS_PDEV_ERROR	= 5,
 
 	/* keep this last */
 	ATH12K_DBG_HTT_NUM_EXT_STATS,
@@ -140,9 +141,13 @@ enum ath12k_dbg_htt_tlv_tag {
 	HTT_STATS_TX_SCHED_CMN_TAG			= 37,
 	HTT_STATS_SCHED_TXQ_CMD_POSTED_TAG		= 39,
 	HTT_STATS_SCHED_TXQ_CMD_REAPED_TAG		= 44,
+	HTT_STATS_HW_INTR_MISC_TAG			= 54,
+	HTT_STATS_HW_PDEV_ERRS_TAG			= 56,
+	HTT_STATS_WHAL_TX_TAG				= 66,
 	HTT_STATS_TX_PDEV_SIFS_HIST_TAG			= 67,
 	HTT_STATS_SCHED_TXQ_SCHED_ORDER_SU_TAG		= 86,
 	HTT_STATS_SCHED_TXQ_SCHED_INELIGIBILITY_TAG	= 87,
+	HTT_STATS_HW_WAR_TAG				= 89,
 	HTT_STATS_SCHED_TXQ_SUPERCYCLE_TRIGGER_TAG	= 100,
 	HTT_STATS_PDEV_CTRL_PATH_TX_STATS_TAG		= 102,
 	HTT_STATS_MU_PPDU_DIST_TAG			= 129,
@@ -378,4 +383,83 @@ struct ath12k_htt_sched_txq_supercycle_triggers_tlv {
 	DECLARE_FLEX_ARRAY(__le32, supercycle_triggers);
 } __packed;
 
+struct ath12k_htt_hw_stats_pdev_errs_tlv {
+	__le32 mac_id__word;
+	__le32 tx_abort;
+	__le32 tx_abort_fail_count;
+	__le32 rx_abort;
+	__le32 rx_abort_fail_count;
+	__le32 warm_reset;
+	__le32 cold_reset;
+	__le32 tx_flush;
+	__le32 tx_glb_reset;
+	__le32 tx_txq_reset;
+	__le32 rx_timeout_reset;
+	__le32 mac_cold_reset_restore_cal;
+	__le32 mac_cold_reset;
+	__le32 mac_warm_reset;
+	__le32 mac_only_reset;
+	__le32 phy_warm_reset;
+	__le32 phy_warm_reset_ucode_trig;
+	__le32 mac_warm_reset_restore_cal;
+	__le32 mac_sfm_reset;
+	__le32 phy_warm_reset_m3_ssr;
+	__le32 phy_warm_reset_reason_phy_m3;
+	__le32 phy_warm_reset_reason_tx_hw_stuck;
+	__le32 phy_warm_reset_reason_num_rx_frame_stuck;
+	__le32 phy_warm_reset_reason_wal_rx_rec_rx_busy;
+	__le32 phy_warm_reset_reason_wal_rx_rec_mac_hng;
+	__le32 phy_warm_reset_reason_mac_conv_phy_reset;
+	__le32 wal_rx_recovery_rst_mac_hang_cnt;
+	__le32 wal_rx_recovery_rst_known_sig_cnt;
+	__le32 wal_rx_recovery_rst_no_rx_cnt;
+	__le32 wal_rx_recovery_rst_no_rx_consec_cnt;
+	__le32 wal_rx_recovery_rst_rx_busy_cnt;
+	__le32 wal_rx_recovery_rst_phy_mac_hang_cnt;
+	__le32 rx_flush_cnt;
+	__le32 phy_warm_reset_reason_tx_exp_cca_stuck;
+	__le32 phy_warm_reset_reason_tx_consec_flsh_war;
+	__le32 phy_warm_reset_reason_tx_hwsch_reset_war;
+	__le32 phy_warm_reset_reason_hwsch_cca_wdog_war;
+	__le32 fw_rx_rings_reset;
+	__le32 rx_dest_drain_rx_descs_leak_prevented;
+	__le32 rx_dest_drain_rx_descs_saved_cnt;
+	__le32 rx_dest_drain_rxdma2reo_leak_detected;
+	__le32 rx_dest_drain_rxdma2fw_leak_detected;
+	__le32 rx_dest_drain_rxdma2wbm_leak_detected;
+	__le32 rx_dest_drain_rxdma1_2sw_leak_detected;
+	__le32 rx_dest_drain_rx_drain_ok_mac_idle;
+	__le32 rx_dest_drain_ok_mac_not_idle;
+	__le32 rx_dest_drain_prerequisite_invld;
+	__le32 rx_dest_drain_skip_non_lmac_reset;
+	__le32 rx_dest_drain_hw_fifo_notempty_post_wait;
+} __packed;
+
+#define ATH12K_HTT_STATS_MAX_HW_INTR_NAME_LEN 8
+struct ath12k_htt_hw_stats_intr_misc_tlv {
+	u8 hw_intr_name[ATH12K_HTT_STATS_MAX_HW_INTR_NAME_LEN];
+	__le32 mask;
+	__le32 count;
+} __packed;
+
+struct ath12k_htt_hw_stats_whal_tx_tlv {
+	__le32 mac_id__word;
+	__le32 last_unpause_ppdu_id;
+	__le32 hwsch_unpause_wait_tqm_write;
+	__le32 hwsch_dummy_tlv_skipped;
+	__le32 hwsch_misaligned_offset_received;
+	__le32 hwsch_reset_count;
+	__le32 hwsch_dev_reset_war;
+	__le32 hwsch_delayed_pause;
+	__le32 hwsch_long_delayed_pause;
+	__le32 sch_rx_ppdu_no_response;
+	__le32 sch_selfgen_response;
+	__le32 sch_rx_sifs_resp_trigger;
+} __packed;
+
+struct ath12k_htt_hw_war_stats_tlv {
+	__le32 mac_id__word;
+	DECLARE_FLEX_ARRAY(__le32, hw_wars);
+} __packed;
+
 #endif
-- 
cgit v1.2.3-58-ga151


From 5344fc76f8944249026884157f846f88489edfc0 Mon Sep 17 00:00:00 2001
From: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Date: Tue, 25 Jun 2024 09:52:17 +0530
Subject: wifi: ath12k: Support TQM stats

Add support to request pdev TQM stats from firmware through HTT stats
type 6. This stat type gives information such as TQM error, MPDU
related information and TQM pdev stats.

Sample output:
-------------
echo 6 > /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats_type
cat /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats

HTT_TX_TQM_ERROR_STATS_TLV:
q_empty_failure = 0
q_not_empty_failure = 0
add_msdu_failure = 0

TQM_ERROR_RESET_STATS:
tqm_cache_ctl_err = 0
tqm_soft_reset = 0
tqm_reset_total_num_in_use_link_descs = 0
.....

HTT_TX_TQM_GEN_MPDU_STATS_TLV:
gen_mpdu_end_reason =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
10:0, 11:0, 12:0, 13:0, 14:0, 15:0, 16:0

HTT_TX_TQM_LIST_MPDU_STATS_TLV:
list_mpdu_end_reason =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
10:0, 11:0, 12:0

HTT_TX_TQM_LIST_MPDU_CNT_TLV_V:
list_mpdu_cnt_hist =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0

HTT_TX_TQM_PDEV_STATS_TLV_V:
msdu_count = 0
mpdu_count = 0
remove_msdu = 0

Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4

Signed-off-by: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Signed-off-by: Roopni Devanathan <quic_rdevanat@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://patch.msgid.link/20240625042217.1303332-4-quic_rdevanat@quicinc.com
---
 .../net/wireless/ath/ath12k/debugfs_htt_stats.c    | 289 +++++++++++++++++++++
 .../net/wireless/ath/ath12k/debugfs_htt_stats.h    | 102 ++++++++
 2 files changed, 391 insertions(+)

diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
index 05bb0ea68e15..ce80e7b5175b 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.c
@@ -846,6 +846,277 @@ ath12k_htt_print_hw_war_tlv(const void *tag_buf, u16 tag_len,
 	stats_req->buf_len = len;
 }
 
+static void
+ath12k_htt_print_tx_tqm_cmn_stats_tlv(const void *tag_buf, u16 tag_len,
+				      struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_tqm_cmn_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u32 mac_id_word;
+
+	if (tag_len < sizeof(*htt_stats_buf))
+		return;
+
+	mac_id_word = __le32_to_cpu(htt_stats_buf->mac_id__word);
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_TX_TQM_CMN_STATS_TLV:\n");
+	len += scnprintf(buf + len, buf_len - len, "mac_id = %u\n",
+			u32_get_bits(mac_id_word, ATH12K_HTT_STATS_MAC_ID));
+	len += scnprintf(buf + len, buf_len - len, "max_cmdq_id = %u\n",
+			 le32_to_cpu(htt_stats_buf->max_cmdq_id));
+	len += scnprintf(buf + len, buf_len - len, "list_mpdu_cnt_hist_intvl = %u\n",
+			 le32_to_cpu(htt_stats_buf->list_mpdu_cnt_hist_intvl));
+	len += scnprintf(buf + len, buf_len - len, "add_msdu = %u\n",
+			 le32_to_cpu(htt_stats_buf->add_msdu));
+	len += scnprintf(buf + len, buf_len - len, "q_empty = %u\n",
+			 le32_to_cpu(htt_stats_buf->q_empty));
+	len += scnprintf(buf + len, buf_len - len, "q_not_empty = %u\n",
+			 le32_to_cpu(htt_stats_buf->q_not_empty));
+	len += scnprintf(buf + len, buf_len - len, "drop_notification = %u\n",
+			 le32_to_cpu(htt_stats_buf->drop_notification));
+	len += scnprintf(buf + len, buf_len - len, "desc_threshold = %u\n",
+			 le32_to_cpu(htt_stats_buf->desc_threshold));
+	len += scnprintf(buf + len, buf_len - len, "hwsch_tqm_invalid_status = %u\n",
+			 le32_to_cpu(htt_stats_buf->hwsch_tqm_invalid_status));
+	len += scnprintf(buf + len, buf_len - len, "missed_tqm_gen_mpdus = %u\n",
+			 le32_to_cpu(htt_stats_buf->missed_tqm_gen_mpdus));
+	len += scnprintf(buf + len, buf_len - len,
+			 "total_msduq_timestamp_updates = %u\n",
+			 le32_to_cpu(htt_stats_buf->msduq_timestamp_updates));
+	len += scnprintf(buf + len, buf_len - len,
+			 "total_msduq_timestamp_updates_by_get_mpdu_head_info_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->msduq_updates_mpdu_head_info_cmd));
+	len += scnprintf(buf + len, buf_len - len,
+			 "total_msduq_timestamp_updates_by_emp_to_nonemp_status = %u\n",
+			 le32_to_cpu(htt_stats_buf->msduq_updates_emp_to_nonemp_status));
+	len += scnprintf(buf + len, buf_len - len,
+			 "total_get_mpdu_head_info_cmds_by_sched_algo_la_query = %u\n",
+			 le32_to_cpu(htt_stats_buf->get_mpdu_head_info_cmds_by_query));
+	len += scnprintf(buf + len, buf_len - len,
+			 "total_get_mpdu_head_info_cmds_by_tac = %u\n",
+			 le32_to_cpu(htt_stats_buf->get_mpdu_head_info_cmds_by_tac));
+	len += scnprintf(buf + len, buf_len - len,
+			 "total_gen_mpdu_cmds_by_sched_algo_la_query = %u\n",
+			 le32_to_cpu(htt_stats_buf->gen_mpdu_cmds_by_query));
+	len += scnprintf(buf + len, buf_len - len, "active_tqm_tids = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_active_tids));
+	len += scnprintf(buf + len, buf_len - len, "inactive_tqm_tids = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_inactive_tids));
+	len += scnprintf(buf + len, buf_len - len, "tqm_active_msduq_flows = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_active_msduq_flows));
+	len += scnprintf(buf + len, buf_len - len, "hi_prio_q_not_empty = %u\n\n",
+			 le32_to_cpu(htt_stats_buf->high_prio_q_not_empty));
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_tx_tqm_error_stats_tlv(const void *tag_buf, u16 tag_len,
+					struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_tqm_error_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+
+	if (tag_len < sizeof(*htt_stats_buf))
+		return;
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_TX_TQM_ERROR_STATS_TLV:\n");
+	len += scnprintf(buf + len, buf_len - len, "q_empty_failure = %u\n",
+			 le32_to_cpu(htt_stats_buf->q_empty_failure));
+	len += scnprintf(buf + len, buf_len - len, "q_not_empty_failure = %u\n",
+			 le32_to_cpu(htt_stats_buf->q_not_empty_failure));
+	len += scnprintf(buf + len, buf_len - len, "add_msdu_failure = %u\n\n",
+			 le32_to_cpu(htt_stats_buf->add_msdu_failure));
+
+	len += scnprintf(buf + len, buf_len - len, "TQM_ERROR_RESET_STATS:\n");
+	len += scnprintf(buf + len, buf_len - len, "tqm_cache_ctl_err = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_cache_ctl_err));
+	len += scnprintf(buf + len, buf_len - len, "tqm_soft_reset = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_soft_reset));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_total_num_in_use_link_descs = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_num_in_use_link_descs));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_worst_case_num_lost_link_descs = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_num_lost_link_descs));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_worst_case_num_lost_host_tx_bufs_count = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_num_lost_host_tx_buf_cnt));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_num_in_use_link_descs_internal_tqm = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_num_in_use_internal_tqm));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_num_in_use_link_descs_wbm_idle_link_ring = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_num_in_use_idle_link_rng));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_time_to_tqm_hang_delta_ms = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_time_to_tqm_hang_delta_ms));
+	len += scnprintf(buf + len, buf_len - len, "tqm_reset_recovery_time_ms = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_recovery_time_ms));
+	len += scnprintf(buf + len, buf_len - len, "tqm_reset_num_peers_hdl = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_num_peers_hdl));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_cumm_dirty_hw_mpduq_proc_cnt = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_cumm_dirty_hw_mpduq_cnt));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_cumm_dirty_hw_msduq_proc = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_cumm_dirty_hw_msduq_proc));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_flush_cache_cmd_su_cnt = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_flush_cache_cmd_su_cnt));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_flush_cache_cmd_other_cnt = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_flush_cache_cmd_other_cnt));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_flush_cache_cmd_trig_type = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_flush_cache_cmd_trig_type));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_flush_cache_cmd_trig_cfg = %u\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_flush_cache_cmd_trig_cfg));
+	len += scnprintf(buf + len, buf_len - len,
+			 "tqm_reset_flush_cache_cmd_skip_cmd_status_null = %u\n\n",
+			 le32_to_cpu(htt_stats_buf->tqm_reset_flush_cmd_skp_status_null));
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_tx_tqm_gen_mpdu_stats_tlv(const void *tag_buf, u16 tag_len,
+					   struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_tqm_gen_mpdu_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u16 num_elements = tag_len >> 2;
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_TX_TQM_GEN_MPDU_STATS_TLV:\n");
+	len += print_array_to_buf(buf, len, "gen_mpdu_end_reason",
+				  htt_stats_buf->gen_mpdu_end_reason, num_elements,
+				  "\n\n");
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_tx_tqm_list_mpdu_stats_tlv(const void *tag_buf, u16 tag_len,
+					    struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_tqm_list_mpdu_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u16 num_elems = min_t(u16, (tag_len >> 2),
+			      ATH12K_HTT_TX_TQM_MAX_LIST_MPDU_END_REASON);
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_TX_TQM_LIST_MPDU_STATS_TLV:\n");
+	len += print_array_to_buf(buf, len, "list_mpdu_end_reason",
+				  htt_stats_buf->list_mpdu_end_reason, num_elems, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_tx_tqm_list_mpdu_cnt_tlv(const void *tag_buf, u16 tag_len,
+					  struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_tqm_list_mpdu_cnt_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+	u16 num_elems = min_t(u16, (tag_len >> 2),
+			      ATH12K_HTT_TX_TQM_MAX_LIST_MPDU_CNT_HISTOGRAM_BINS);
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_TX_TQM_LIST_MPDU_CNT_TLV_V:\n");
+	len += print_array_to_buf(buf, len, "list_mpdu_cnt_hist",
+				  htt_stats_buf->list_mpdu_cnt_hist, num_elems, "\n\n");
+
+	stats_req->buf_len = len;
+}
+
+static void
+ath12k_htt_print_tx_tqm_pdev_stats_tlv(const void *tag_buf, u16 tag_len,
+				       struct debug_htt_stats_req *stats_req)
+{
+	const struct ath12k_htt_tx_tqm_pdev_stats_tlv *htt_stats_buf = tag_buf;
+	u8 *buf = stats_req->buf;
+	u32 len = stats_req->buf_len;
+	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
+
+	if (tag_len < sizeof(*htt_stats_buf))
+		return;
+
+	len += scnprintf(buf + len, buf_len - len, "HTT_TX_TQM_PDEV_STATS_TLV_V:\n");
+	len += scnprintf(buf + len, buf_len - len, "msdu_count = %u\n",
+			 le32_to_cpu(htt_stats_buf->msdu_count));
+	len += scnprintf(buf + len, buf_len - len, "mpdu_count = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdu_count));
+	len += scnprintf(buf + len, buf_len - len, "remove_msdu = %u\n",
+			 le32_to_cpu(htt_stats_buf->remove_msdu));
+	len += scnprintf(buf + len, buf_len - len, "remove_mpdu = %u\n",
+			 le32_to_cpu(htt_stats_buf->remove_mpdu));
+	len += scnprintf(buf + len, buf_len - len, "remove_msdu_ttl = %u\n",
+			 le32_to_cpu(htt_stats_buf->remove_msdu_ttl));
+	len += scnprintf(buf + len, buf_len - len, "send_bar = %u\n",
+			 le32_to_cpu(htt_stats_buf->send_bar));
+	len += scnprintf(buf + len, buf_len - len, "bar_sync = %u\n",
+			 le32_to_cpu(htt_stats_buf->bar_sync));
+	len += scnprintf(buf + len, buf_len - len, "notify_mpdu = %u\n",
+			 le32_to_cpu(htt_stats_buf->notify_mpdu));
+	len += scnprintf(buf + len, buf_len - len, "sync_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->sync_cmd));
+	len += scnprintf(buf + len, buf_len - len, "write_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->write_cmd));
+	len += scnprintf(buf + len, buf_len - len, "hwsch_trigger = %u\n",
+			 le32_to_cpu(htt_stats_buf->hwsch_trigger));
+	len += scnprintf(buf + len, buf_len - len, "ack_tlv_proc = %u\n",
+			 le32_to_cpu(htt_stats_buf->ack_tlv_proc));
+	len += scnprintf(buf + len, buf_len - len, "gen_mpdu_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->gen_mpdu_cmd));
+	len += scnprintf(buf + len, buf_len - len, "gen_list_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->gen_list_cmd));
+	len += scnprintf(buf + len, buf_len - len, "remove_mpdu_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->remove_mpdu_cmd));
+	len += scnprintf(buf + len, buf_len - len, "remove_mpdu_tried_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->remove_mpdu_tried_cmd));
+	len += scnprintf(buf + len, buf_len - len, "mpdu_queue_stats_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdu_queue_stats_cmd));
+	len += scnprintf(buf + len, buf_len - len, "mpdu_head_info_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->mpdu_head_info_cmd));
+	len += scnprintf(buf + len, buf_len - len, "msdu_flow_stats_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->msdu_flow_stats_cmd));
+	len += scnprintf(buf + len, buf_len - len, "remove_msdu_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->remove_msdu_cmd));
+	len += scnprintf(buf + len, buf_len - len, "remove_msdu_ttl_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->remove_msdu_ttl_cmd));
+	len += scnprintf(buf + len, buf_len - len, "flush_cache_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->flush_cache_cmd));
+	len += scnprintf(buf + len, buf_len - len, "update_mpduq_cmd = %u\n",
+			 le32_to_cpu(htt_stats_buf->update_mpduq_cmd));
+	len += scnprintf(buf + len, buf_len - len, "enqueue = %u\n",
+			 le32_to_cpu(htt_stats_buf->enqueue));
+	len += scnprintf(buf + len, buf_len - len, "enqueue_notify = %u\n",
+			 le32_to_cpu(htt_stats_buf->enqueue_notify));
+	len += scnprintf(buf + len, buf_len - len, "notify_mpdu_at_head = %u\n",
+			 le32_to_cpu(htt_stats_buf->notify_mpdu_at_head));
+	len += scnprintf(buf + len, buf_len - len, "notify_mpdu_state_valid = %u\n",
+			 le32_to_cpu(htt_stats_buf->notify_mpdu_state_valid));
+	len += scnprintf(buf + len, buf_len - len, "sched_udp_notify1 = %u\n",
+			 le32_to_cpu(htt_stats_buf->sched_udp_notify1));
+	len += scnprintf(buf + len, buf_len - len, "sched_udp_notify2 = %u\n",
+			 le32_to_cpu(htt_stats_buf->sched_udp_notify2));
+	len += scnprintf(buf + len, buf_len - len, "sched_nonudp_notify1 = %u\n",
+			 le32_to_cpu(htt_stats_buf->sched_nonudp_notify1));
+	len += scnprintf(buf + len, buf_len - len, "sched_nonudp_notify2 = %u\n\n",
+			 le32_to_cpu(htt_stats_buf->sched_nonudp_notify2));
+
+	stats_req->buf_len = len;
+}
+
 static int ath12k_dbg_htt_ext_stats_parse(struct ath12k_base *ab,
 					  u16 tag, u16 len, const void *tag_buf,
 					  void *user_data)
@@ -909,6 +1180,24 @@ static int ath12k_dbg_htt_ext_stats_parse(struct ath12k_base *ab,
 	case HTT_STATS_HW_WAR_TAG:
 		ath12k_htt_print_hw_war_tlv(tag_buf, len, stats_req);
 		break;
+	case HTT_STATS_TX_TQM_CMN_TAG:
+		ath12k_htt_print_tx_tqm_cmn_stats_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_TQM_ERROR_STATS_TAG:
+		ath12k_htt_print_tx_tqm_error_stats_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_TQM_GEN_MPDU_TAG:
+		ath12k_htt_print_tx_tqm_gen_mpdu_stats_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_TQM_LIST_MPDU_TAG:
+		ath12k_htt_print_tx_tqm_list_mpdu_stats_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_TQM_LIST_MPDU_CNT_TAG:
+		ath12k_htt_print_tx_tqm_list_mpdu_cnt_tlv(tag_buf, len, stats_req);
+		break;
+	case HTT_STATS_TX_TQM_PDEV_TAG:
+		ath12k_htt_print_tx_tqm_pdev_stats_tlv(tag_buf, len, stats_req);
+		break;
 	default:
 		break;
 	}
diff --git a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
index 777a2e067a04..6294a082cf8a 100644
--- a/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
+++ b/drivers/net/wireless/ath/ath12k/debugfs_htt_stats.h
@@ -127,6 +127,7 @@ enum ath12k_dbg_htt_ext_stats_type {
 	ATH12K_DBG_HTT_EXT_STATS_PDEV_TX	= 1,
 	ATH12K_DBG_HTT_EXT_STATS_PDEV_TX_SCHED	= 4,
 	ATH12K_DBG_HTT_EXT_STATS_PDEV_ERROR	= 5,
+	ATH12K_DBG_HTT_EXT_STATS_PDEV_TQM	= 6,
 
 	/* keep this last */
 	ATH12K_DBG_HTT_NUM_EXT_STATS,
@@ -137,9 +138,15 @@ enum ath12k_dbg_htt_tlv_tag {
 	HTT_STATS_TX_PDEV_UNDERRUN_TAG			= 1,
 	HTT_STATS_TX_PDEV_SIFS_TAG			= 2,
 	HTT_STATS_TX_PDEV_FLUSH_TAG			= 3,
+	HTT_STATS_TX_TQM_GEN_MPDU_TAG			= 11,
+	HTT_STATS_TX_TQM_LIST_MPDU_TAG			= 12,
+	HTT_STATS_TX_TQM_LIST_MPDU_CNT_TAG		= 13,
+	HTT_STATS_TX_TQM_CMN_TAG			= 14,
+	HTT_STATS_TX_TQM_PDEV_TAG			= 15,
 	HTT_STATS_TX_PDEV_SCHEDULER_TXQ_STATS_TAG	= 36,
 	HTT_STATS_TX_SCHED_CMN_TAG			= 37,
 	HTT_STATS_SCHED_TXQ_CMD_POSTED_TAG		= 39,
+	HTT_STATS_TX_TQM_ERROR_STATS_TAG                = 43,
 	HTT_STATS_SCHED_TXQ_CMD_REAPED_TAG		= 44,
 	HTT_STATS_HW_INTR_MISC_TAG			= 54,
 	HTT_STATS_HW_PDEV_ERRS_TAG			= 56,
@@ -462,4 +469,99 @@ struct ath12k_htt_hw_war_stats_tlv {
 	DECLARE_FLEX_ARRAY(__le32, hw_wars);
 } __packed;
 
+struct ath12k_htt_tx_tqm_cmn_stats_tlv {
+	__le32 mac_id__word;
+	__le32 max_cmdq_id;
+	__le32 list_mpdu_cnt_hist_intvl;
+	__le32 add_msdu;
+	__le32 q_empty;
+	__le32 q_not_empty;
+	__le32 drop_notification;
+	__le32 desc_threshold;
+	__le32 hwsch_tqm_invalid_status;
+	__le32 missed_tqm_gen_mpdus;
+	__le32 tqm_active_tids;
+	__le32 tqm_inactive_tids;
+	__le32 tqm_active_msduq_flows;
+	__le32 msduq_timestamp_updates;
+	__le32 msduq_updates_mpdu_head_info_cmd;
+	__le32 msduq_updates_emp_to_nonemp_status;
+	__le32 get_mpdu_head_info_cmds_by_query;
+	__le32 get_mpdu_head_info_cmds_by_tac;
+	__le32 gen_mpdu_cmds_by_query;
+	__le32 high_prio_q_not_empty;
+} __packed;
+
+struct ath12k_htt_tx_tqm_error_stats_tlv {
+	__le32 q_empty_failure;
+	__le32 q_not_empty_failure;
+	__le32 add_msdu_failure;
+	__le32 tqm_cache_ctl_err;
+	__le32 tqm_soft_reset;
+	__le32 tqm_reset_num_in_use_link_descs;
+	__le32 tqm_reset_num_lost_link_descs;
+	__le32 tqm_reset_num_lost_host_tx_buf_cnt;
+	__le32 tqm_reset_num_in_use_internal_tqm;
+	__le32 tqm_reset_num_in_use_idle_link_rng;
+	__le32 tqm_reset_time_to_tqm_hang_delta_ms;
+	__le32 tqm_reset_recovery_time_ms;
+	__le32 tqm_reset_num_peers_hdl;
+	__le32 tqm_reset_cumm_dirty_hw_mpduq_cnt;
+	__le32 tqm_reset_cumm_dirty_hw_msduq_proc;
+	__le32 tqm_reset_flush_cache_cmd_su_cnt;
+	__le32 tqm_reset_flush_cache_cmd_other_cnt;
+	__le32 tqm_reset_flush_cache_cmd_trig_type;
+	__le32 tqm_reset_flush_cache_cmd_trig_cfg;
+	__le32 tqm_reset_flush_cmd_skp_status_null;
+} __packed;
+
+struct ath12k_htt_tx_tqm_gen_mpdu_stats_tlv {
+	DECLARE_FLEX_ARRAY(__le32, gen_mpdu_end_reason);
+} __packed;
+
+#define ATH12K_HTT_TX_TQM_MAX_LIST_MPDU_END_REASON		16
+#define ATH12K_HTT_TX_TQM_MAX_LIST_MPDU_CNT_HISTOGRAM_BINS	16
+
+struct ath12k_htt_tx_tqm_list_mpdu_stats_tlv {
+	DECLARE_FLEX_ARRAY(__le32, list_mpdu_end_reason);
+} __packed;
+
+struct ath12k_htt_tx_tqm_list_mpdu_cnt_tlv {
+	DECLARE_FLEX_ARRAY(__le32, list_mpdu_cnt_hist);
+} __packed;
+
+struct ath12k_htt_tx_tqm_pdev_stats_tlv {
+	__le32 msdu_count;
+	__le32 mpdu_count;
+	__le32 remove_msdu;
+	__le32 remove_mpdu;
+	__le32 remove_msdu_ttl;
+	__le32 send_bar;
+	__le32 bar_sync;
+	__le32 notify_mpdu;
+	__le32 sync_cmd;
+	__le32 write_cmd;
+	__le32 hwsch_trigger;
+	__le32 ack_tlv_proc;
+	__le32 gen_mpdu_cmd;
+	__le32 gen_list_cmd;
+	__le32 remove_mpdu_cmd;
+	__le32 remove_mpdu_tried_cmd;
+	__le32 mpdu_queue_stats_cmd;
+	__le32 mpdu_head_info_cmd;
+	__le32 msdu_flow_stats_cmd;
+	__le32 remove_msdu_cmd;
+	__le32 remove_msdu_ttl_cmd;
+	__le32 flush_cache_cmd;
+	__le32 update_mpduq_cmd;
+	__le32 enqueue;
+	__le32 enqueue_notify;
+	__le32 notify_mpdu_at_head;
+	__le32 notify_mpdu_state_valid;
+	__le32 sched_udp_notify1;
+	__le32 sched_udp_notify2;
+	__le32 sched_nonudp_notify1;
+	__le32 sched_nonudp_notify2;
+} __packed;
+
 #endif
-- 
cgit v1.2.3-58-ga151


From 19e6ad2c75782bd15b3df24df7982da457d702c5 Mon Sep 17 00:00:00 2001
From: "Russell King (Oracle)" <rmk+kernel@armlinux.org.uk>
Date: Fri, 28 Jun 2024 11:32:11 +0100
Subject: net: phy: fix potential use of NULL pointer in phy_suspend()

phy_suspend() checks the WoL status, and then dereferences
phydrv->flags if (and only if) we decided that WoL has been enabled
on either the PHY or the netdev.

We then check whether phydrv was NULL, but we've potentially already
dereferenced the pointer.

If phydrv is NULL, then phy_ethtool_get_wol() will return an error
and leave wol.wolopts set to zero. However, if netdev->wol_enabled
is true, then we would dereference a NULL pointer.

Checking the PHY drivers, the only place that phydev->wol_enabled is
checked by them is in their suspend/resume callbacks and nowhere else
(which is correct, because phylib only updates this in phy_suspend()).

So, move the NULL pointer check earlier to avoid a NULL pointer
dereference. Leave the check for phydrv->suspend in place as a driver
may populate the .resume method but not the .suspend method.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/E1sN8tn-00GDCZ-Jj@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/phy_device.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 473cbc1d497b..70b07e621fb2 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -1980,7 +1980,7 @@ int phy_suspend(struct phy_device *phydev)
 	const struct phy_driver *phydrv = phydev->drv;
 	int ret;
 
-	if (phydev->suspended)
+	if (phydev->suspended || !phydrv)
 		return 0;
 
 	phy_ethtool_get_wol(phydev, &wol);
@@ -1990,7 +1990,7 @@ int phy_suspend(struct phy_device *phydev)
 	if (phydev->wol_enabled && !(phydrv->flags & PHY_ALWAYS_CALL_SUSPEND))
 		return -EBUSY;
 
-	if (!phydrv || !phydrv->suspend)
+	if (!phydrv->suspend)
 		return 0;
 
 	ret = phydrv->suspend(phydev);
-- 
cgit v1.2.3-58-ga151


From 9e2db9d3993e270b24fbc4ce1ca7e09756e8df25 Mon Sep 17 00:00:00 2001
From: Pavel Begunkov <asml.silence@gmail.com>
Date: Thu, 27 Jun 2024 13:59:41 +0100
Subject: net: always try to set ubuf in skb_zerocopy_iter_stream

skb_zcopy_set() does nothing if there is already a ubuf_info associated
with an skb, and since ->link_skb should have set it several lines above
the check here essentially does nothing and can be removed. It's also
safer this way, because even if the callback is faulty we'll
have it set.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/skbuff.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index eb9a7e65b5c8..52986e1ce13e 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1899,8 +1899,7 @@ int skb_zerocopy_iter_stream(struct sock *sk, struct sk_buff *skb,
 		return err;
 	}
 
-	if (!uarg->ops->link_skb)
-		skb_zcopy_set(skb, uarg, NULL);
+	skb_zcopy_set(skb, uarg, NULL);
 	return skb->len - orig_len;
 }
 EXPORT_SYMBOL_GPL(skb_zerocopy_iter_stream);
-- 
cgit v1.2.3-58-ga151


From 7fb05423fed41686ccc1a76c20d486728f62023f Mon Sep 17 00:00:00 2001
From: Pavel Begunkov <asml.silence@gmail.com>
Date: Thu, 27 Jun 2024 13:59:42 +0100
Subject: net: split __zerocopy_sg_from_iter()

Split a function out of __zerocopy_sg_from_iter() that only cares about
the traditional path with refcounted pages and doesn't need to know
about ->sg_from_iter. A preparation patch, we'll improve on the function
later.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/datagram.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

diff --git a/net/core/datagram.c b/net/core/datagram.c
index e614cfd8e14a..ef81d6ecbe1e 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -610,16 +610,10 @@ fault:
 }
 EXPORT_SYMBOL(skb_copy_datagram_from_iter);
 
-int __zerocopy_sg_from_iter(struct msghdr *msg, struct sock *sk,
-			    struct sk_buff *skb, struct iov_iter *from,
-			    size_t length)
+static int zerocopy_fill_skb_from_iter(struct sock *sk, struct sk_buff *skb,
+					struct iov_iter *from, size_t length)
 {
-	int frag;
-
-	if (msg && msg->msg_ubuf && msg->sg_from_iter)
-		return msg->sg_from_iter(sk, skb, from, length);
-
-	frag = skb_shinfo(skb)->nr_frags;
+	int frag = skb_shinfo(skb)->nr_frags;
 
 	while (length && iov_iter_count(from)) {
 		struct page *head, *last_head = NULL;
@@ -692,6 +686,16 @@ int __zerocopy_sg_from_iter(struct msghdr *msg, struct sock *sk,
 	}
 	return 0;
 }
+
+int __zerocopy_sg_from_iter(struct msghdr *msg, struct sock *sk,
+			    struct sk_buff *skb, struct iov_iter *from,
+			    size_t length)
+{
+	if (msg && msg->msg_ubuf && msg->sg_from_iter)
+		return msg->sg_from_iter(sk, skb, from, length);
+	else
+		return zerocopy_fill_skb_from_iter(sk, skb, from, length);
+}
 EXPORT_SYMBOL(__zerocopy_sg_from_iter);
 
 /**
-- 
cgit v1.2.3-58-ga151


From aeb320fc05c74e1d3b429aa0e3a777b8a931c189 Mon Sep 17 00:00:00 2001
From: Pavel Begunkov <asml.silence@gmail.com>
Date: Thu, 27 Jun 2024 13:59:43 +0100
Subject: net: batch zerocopy_fill_skb_from_iter accounting

Instead of accounting every page range against the socket separately, do
it in batch based on the change in skb->truesize. It's also moved into
__zerocopy_sg_from_iter(), so that zerocopy_fill_skb_from_iter() is
simpler and responsible for setting frags but not the accounting.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/datagram.c | 31 ++++++++++++++++++-------------
 1 file changed, 18 insertions(+), 13 deletions(-)

diff --git a/net/core/datagram.c b/net/core/datagram.c
index ef81d6ecbe1e..b0dccefd4a09 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -610,7 +610,7 @@ fault:
 }
 EXPORT_SYMBOL(skb_copy_datagram_from_iter);
 
-static int zerocopy_fill_skb_from_iter(struct sock *sk, struct sk_buff *skb,
+static int zerocopy_fill_skb_from_iter(struct sk_buff *skb,
 					struct iov_iter *from, size_t length)
 {
 	int frag = skb_shinfo(skb)->nr_frags;
@@ -621,7 +621,6 @@ static int zerocopy_fill_skb_from_iter(struct sock *sk, struct sk_buff *skb,
 		int refs, order, n = 0;
 		size_t start;
 		ssize_t copied;
-		unsigned long truesize;
 
 		if (frag == MAX_SKB_FRAGS)
 			return -EMSGSIZE;
@@ -633,17 +632,9 @@ static int zerocopy_fill_skb_from_iter(struct sock *sk, struct sk_buff *skb,
 
 		length -= copied;
 
-		truesize = PAGE_ALIGN(copied + start);
 		skb->data_len += copied;
 		skb->len += copied;
-		skb->truesize += truesize;
-		if (sk && sk->sk_type == SOCK_STREAM) {
-			sk_wmem_queued_add(sk, truesize);
-			if (!skb_zcopy_pure(skb))
-				sk_mem_charge(sk, truesize);
-		} else {
-			refcount_add(truesize, &skb->sk->sk_wmem_alloc);
-		}
+		skb->truesize += PAGE_ALIGN(copied + start);
 
 		head = compound_head(pages[n]);
 		order = compound_order(head);
@@ -691,10 +682,24 @@ int __zerocopy_sg_from_iter(struct msghdr *msg, struct sock *sk,
 			    struct sk_buff *skb, struct iov_iter *from,
 			    size_t length)
 {
+	unsigned long orig_size = skb->truesize;
+	unsigned long truesize;
+	int ret;
+
 	if (msg && msg->msg_ubuf && msg->sg_from_iter)
 		return msg->sg_from_iter(sk, skb, from, length);
-	else
-		return zerocopy_fill_skb_from_iter(sk, skb, from, length);
+
+	ret = zerocopy_fill_skb_from_iter(skb, from, length);
+	truesize = skb->truesize - orig_size;
+
+	if (sk && sk->sk_type == SOCK_STREAM) {
+		sk_wmem_queued_add(sk, truesize);
+		if (!skb_zcopy_pure(skb))
+			sk_mem_charge(sk, truesize);
+	} else {
+		refcount_add(truesize, &skb->sk->sk_wmem_alloc);
+	}
+	return ret;
 }
 EXPORT_SYMBOL(__zerocopy_sg_from_iter);
 
-- 
cgit v1.2.3-58-ga151


From 060f4ba6e40338a70932603a3564903acf5f5734 Mon Sep 17 00:00:00 2001
From: Pavel Begunkov <asml.silence@gmail.com>
Date: Thu, 27 Jun 2024 13:59:44 +0100
Subject: io_uring/net: move charging socket out of zc io_uring

Currently, io_uring's io_sg_from_iter() duplicates the part of
__zerocopy_sg_from_iter() charging pages to the socket. It'd be too easy
to miss while changing it in net/, the chunk is not the most
straightforward for outside users and full of internal implementation
details. io_uring is not a good place to keep it, deduplicate it by
moving out of the callback into __zerocopy_sg_from_iter().

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/linux/skbuff.h |  3 +++
 include/linux/socket.h |  2 +-
 io_uring/net.c         | 16 ++++------------
 net/core/datagram.c    | 10 +++++-----
 4 files changed, 13 insertions(+), 18 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index f4cda3fbdb75..9c29bdd5596d 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1703,6 +1703,9 @@ int __zerocopy_sg_from_iter(struct msghdr *msg, struct sock *sk,
 			    struct sk_buff *skb, struct iov_iter *from,
 			    size_t length);
 
+int zerocopy_fill_skb_from_iter(struct sk_buff *skb,
+				struct iov_iter *from, size_t length);
+
 static inline int skb_zerocopy_iter_dgram(struct sk_buff *skb,
 					  struct msghdr *msg, int len)
 {
diff --git a/include/linux/socket.h b/include/linux/socket.h
index 89d16b90370b..2a1ff91d1914 100644
--- a/include/linux/socket.h
+++ b/include/linux/socket.h
@@ -76,7 +76,7 @@ struct msghdr {
 	__kernel_size_t	msg_controllen;	/* ancillary data buffer length */
 	struct kiocb	*msg_iocb;	/* ptr to iocb for async requests */
 	struct ubuf_info *msg_ubuf;
-	int (*sg_from_iter)(struct sock *sk, struct sk_buff *skb,
+	int (*sg_from_iter)(struct sk_buff *skb,
 			    struct iov_iter *from, size_t length);
 };
 
diff --git a/io_uring/net.c b/io_uring/net.c
index 7c98c4d50946..84a7602bcef1 100644
--- a/io_uring/net.c
+++ b/io_uring/net.c
@@ -1265,14 +1265,14 @@ int io_send_zc_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	return io_sendmsg_prep_setup(req, req->opcode == IORING_OP_SENDMSG_ZC);
 }
 
-static int io_sg_from_iter_iovec(struct sock *sk, struct sk_buff *skb,
+static int io_sg_from_iter_iovec(struct sk_buff *skb,
 				 struct iov_iter *from, size_t length)
 {
 	skb_zcopy_downgrade_managed(skb);
-	return __zerocopy_sg_from_iter(NULL, sk, skb, from, length);
+	return zerocopy_fill_skb_from_iter(skb, from, length);
 }
 
-static int io_sg_from_iter(struct sock *sk, struct sk_buff *skb,
+static int io_sg_from_iter(struct sk_buff *skb,
 			   struct iov_iter *from, size_t length)
 {
 	struct skb_shared_info *shinfo = skb_shinfo(skb);
@@ -1285,7 +1285,7 @@ static int io_sg_from_iter(struct sock *sk, struct sk_buff *skb,
 	if (!frag)
 		shinfo->flags |= SKBFL_MANAGED_FRAG_REFS;
 	else if (unlikely(!skb_zcopy_managed(skb)))
-		return __zerocopy_sg_from_iter(NULL, sk, skb, from, length);
+		return zerocopy_fill_skb_from_iter(skb, from, length);
 
 	bi.bi_size = min(from->count, length);
 	bi.bi_bvec_done = from->iov_offset;
@@ -1312,14 +1312,6 @@ static int io_sg_from_iter(struct sock *sk, struct sk_buff *skb,
 	skb->data_len += copied;
 	skb->len += copied;
 	skb->truesize += truesize;
-
-	if (sk && sk->sk_type == SOCK_STREAM) {
-		sk_wmem_queued_add(sk, truesize);
-		if (!skb_zcopy_pure(skb))
-			sk_mem_charge(sk, truesize);
-	} else {
-		refcount_add(truesize, &skb->sk->sk_wmem_alloc);
-	}
 	return ret;
 }
 
diff --git a/net/core/datagram.c b/net/core/datagram.c
index b0dccefd4a09..16507b7cc4fb 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -610,8 +610,8 @@ fault:
 }
 EXPORT_SYMBOL(skb_copy_datagram_from_iter);
 
-static int zerocopy_fill_skb_from_iter(struct sk_buff *skb,
-					struct iov_iter *from, size_t length)
+int zerocopy_fill_skb_from_iter(struct sk_buff *skb,
+				struct iov_iter *from, size_t length)
 {
 	int frag = skb_shinfo(skb)->nr_frags;
 
@@ -687,11 +687,11 @@ int __zerocopy_sg_from_iter(struct msghdr *msg, struct sock *sk,
 	int ret;
 
 	if (msg && msg->msg_ubuf && msg->sg_from_iter)
-		return msg->sg_from_iter(sk, skb, from, length);
+		ret = msg->sg_from_iter(skb, from, length);
+	else
+		ret = zerocopy_fill_skb_from_iter(skb, from, length);
 
-	ret = zerocopy_fill_skb_from_iter(skb, from, length);
 	truesize = skb->truesize - orig_size;
-
 	if (sk && sk->sk_type == SOCK_STREAM) {
 		sk_wmem_queued_add(sk, truesize);
 		if (!skb_zcopy_pure(skb))
-- 
cgit v1.2.3-58-ga151


From 2ca58ed21cefdda45520a0a2b1980c008efe9874 Mon Sep 17 00:00:00 2001
From: Pavel Begunkov <asml.silence@gmail.com>
Date: Thu, 27 Jun 2024 13:59:45 +0100
Subject: net: limit scope of a skb_zerocopy_iter_stream var

skb_zerocopy_iter_stream() only uses @orig_uarg in the !link_skb path,
and we can move the local variable in the appropriate block.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/skbuff.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 52986e1ce13e..0ed4d00d258c 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1871,7 +1871,6 @@ int skb_zerocopy_iter_stream(struct sock *sk, struct sk_buff *skb,
 			     struct msghdr *msg, int len,
 			     struct ubuf_info *uarg)
 {
-	struct ubuf_info *orig_uarg = skb_zcopy(skb);
 	int err, orig_len = skb->len;
 
 	if (uarg->ops->link_skb) {
@@ -1879,6 +1878,8 @@ int skb_zerocopy_iter_stream(struct sock *sk, struct sk_buff *skb,
 		if (err)
 			return err;
 	} else {
+		struct ubuf_info *orig_uarg = skb_zcopy(skb);
+
 		/* An skb can only point to one uarg. This edge case happens
 		 * when TCP appends to an skb, but zerocopy_realloc triggered
 		 * a new alloc.
-- 
cgit v1.2.3-58-ga151


From d2f299a4c9983022907ff8ec4d4ace49a409ce40 Mon Sep 17 00:00:00 2001
From: Ching-Te Ku <ku920601@realtek.com>
Date: Wed, 26 Jun 2024 10:31:29 +0800
Subject: wifi: rtw89: coex: Add coexistence policy for hardware scan

Add additional policy option to let Bluetooth audio can play well
during Wi-Fi is hardware scanning.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240626023129.7776-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/coex.c | 26 +++++++++++++++++++++++++-
 drivers/net/wireless/realtek/rtw89/core.h |  1 +
 2 files changed, 26 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
index d8fc93318a51..24929ef534e0 100644
--- a/drivers/net/wireless/realtek/rtw89/coex.c
+++ b/drivers/net/wireless/realtek/rtw89/coex.c
@@ -91,7 +91,7 @@ static const struct rtw89_btc_fbtc_slot s_def[] = {
 	[CXST_BLK]	= __DEF_FBTC_SLOT(500, 0x55555555, SLOT_MIX),
 	[CXST_E2G]	= __DEF_FBTC_SLOT(0,   0xea5a5a5a, SLOT_MIX),
 	[CXST_E5G]	= __DEF_FBTC_SLOT(0,   0xffffffff, SLOT_ISO),
-	[CXST_EBT]	= __DEF_FBTC_SLOT(0,   0xe5555555, SLOT_MIX),
+	[CXST_EBT]	= __DEF_FBTC_SLOT(5,   0xe5555555, SLOT_MIX),
 	[CXST_ENULL]	= __DEF_FBTC_SLOT(0,   0xaaaaaaaa, SLOT_ISO),
 	[CXST_WLK]	= __DEF_FBTC_SLOT(250, 0xea5a5a5a, SLOT_MIX),
 	[CXST_W1FDD]	= __DEF_FBTC_SLOT(50,  0xffffffff, SLOT_ISO),
@@ -3617,6 +3617,7 @@ void rtw89_btc_set_policy_v1(struct rtw89_dev *rtwdev, u16 policy_type)
 	struct rtw89_btc_wl_info *wl = &btc->cx.wl;
 	u8 type, null_role;
 	u32 tbl_w1, tbl_b1, tbl_b4;
+	u16 dur_2;
 
 	type = FIELD_GET(BTC_CXP_MASK, policy_type);
 
@@ -3727,7 +3728,21 @@ void rtw89_btc_set_policy_v1(struct rtw89_dev *rtwdev, u16 policy_type)
 		if (hid->exist || hfp->exist)
 			tbl_w1 = cxtbl[16];
 
+		dur_2 = dm->e2g_slot_limit;
+
 		switch (policy_type) {
+		case BTC_CXP_OFFE_2GBWISOB: /* for normal-case */
+			_slot_set(btc, CXST_E2G, 0, tbl_w1, SLOT_ISO);
+			_slot_set_le(btc, CXST_EBT, s_def[CXST_EBT].dur,
+				     s_def[CXST_EBT].cxtbl, s_def[CXST_EBT].cxtype);
+			_slot_set_dur(btc, CXST_EBT, dur_2);
+			break;
+		case BTC_CXP_OFFE_2GISOB: /* for bt no-link */
+			_slot_set(btc, CXST_E2G, 0, cxtbl[1], SLOT_ISO);
+			_slot_set_le(btc, CXST_EBT, s_def[CXST_EBT].dur,
+				     s_def[CXST_EBT].cxtbl, s_def[CXST_EBT].cxtype);
+			_slot_set_dur(btc, CXST_EBT, dur_2);
+			break;
 		case BTC_CXP_OFFE_DEF:
 			_slot_set_le(btc, CXST_E2G, s_def[CXST_E2G].dur,
 				     s_def[CXST_E2G].cxtbl, s_def[CXST_E2G].cxtype);
@@ -3747,6 +3762,15 @@ void rtw89_btc_set_policy_v1(struct rtw89_dev *rtwdev, u16 policy_type)
 			_slot_set_le(btc, CXST_ENULL, s_def[CXST_ENULL].dur,
 				     s_def[CXST_ENULL].cxtbl, s_def[CXST_ENULL].cxtype);
 			break;
+		case BTC_CXP_OFFE_2GBWMIXB:
+			_slot_set(btc, CXST_E2G, 0, 0x55555555, SLOT_MIX);
+			_slot_set_le(btc, CXST_EBT, s_def[CXST_EBT].dur,
+				     cpu_to_le32(0x55555555), s_def[CXST_EBT].cxtype);
+			break;
+		case BTC_CXP_OFFE_WL: /* for 4-way */
+			_slot_set(btc, CXST_E2G, 0, cxtbl[1], SLOT_MIX);
+			_slot_set(btc, CXST_EBT, 0, cxtbl[1], SLOT_MIX);
+			break;
 		default:
 			break;
 		}
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index e1bcac7e1324..b20a0d5c9e66 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -1330,6 +1330,7 @@ struct rtw89_btc_wl_smap {
 	u32 scan: 1;
 	u32 connecting: 1;
 	u32 roaming: 1;
+	u32 dbccing: 1;
 	u32 transacting: 1;
 	u32 _4way: 1;
 	u32 rf_off: 1;
-- 
cgit v1.2.3-58-ga151


From dcf920ff8c16868ce0f9738fb1dadf455f86bd94 Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Wed, 26 Jun 2024 10:32:37 +0800
Subject: wifi: rtw89: constrain TX power according to Transmit Power Envelope

Calculate a TX power constraint based on content of ieee80211 Transmit
Power Envelope (TPE). Since HW control registers aren't designed as many
as all kinds of TPE fields, we strictly intersect all TPE inputs in driver.
Then, according to result, constrain TX power via TX power limit/limit_RU.

Besides, extend dbgfs txpwr_table to show info about 6 GHz regulatory.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240626023237.7901-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/core.c     |  10 +-
 drivers/net/wireless/realtek/rtw89/core.h     |  14 +-
 drivers/net/wireless/realtek/rtw89/debug.c    |  25 +++-
 drivers/net/wireless/realtek/rtw89/mac80211.c |   3 +
 drivers/net/wireless/realtek/rtw89/phy.c      |  38 +++++-
 drivers/net/wireless/realtek/rtw89/regd.c     | 190 ++++++++++++++++++++++++--
 6 files changed, 262 insertions(+), 18 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/core.c b/drivers/net/wireless/realtek/rtw89/core.c
index 4acd4301d3c2..7019f7d482a8 100644
--- a/drivers/net/wireless/realtek/rtw89/core.c
+++ b/drivers/net/wireless/realtek/rtw89/core.c
@@ -3376,8 +3376,12 @@ int rtw89_core_sta_add(struct rtw89_dev *rtwdev,
 	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls) {
 		/* for station mode, assign the mac_id from itself */
 		rtwsta->mac_id = rtwvif->mac_id;
-		/* must do rtw89_reg_6ghz_power_recalc() before rfk channel */
-		rtw89_reg_6ghz_power_recalc(rtwdev, rtwvif, true);
+
+		/* must do rtw89_reg_6ghz_recalc() before rfk channel */
+		ret = rtw89_reg_6ghz_recalc(rtwdev, rtwvif, true);
+		if (ret)
+			return ret;
+
 		rtw89_btc_ntfy_role_info(rtwdev, rtwvif, rtwsta,
 					 BTC_ROLE_MSTS_STA_CONN_START);
 		rtw89_chip_rfk_channel(rtwdev);
@@ -3565,7 +3569,7 @@ int rtw89_core_sta_remove(struct rtw89_dev *rtwdev,
 	int ret;
 
 	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls) {
-		rtw89_reg_6ghz_power_recalc(rtwdev, rtwvif, false);
+		rtw89_reg_6ghz_recalc(rtwdev, rtwvif, false);
 		rtw89_btc_ntfy_role_info(rtwdev, rtwvif, rtwsta,
 					 BTC_ROLE_MSTS_STA_DIS_CONN);
 	} else if (vif->type == NL80211_IFTYPE_AP || sta->tdls) {
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index b20a0d5c9e66..001da3ad149e 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -746,6 +746,14 @@ enum rtw89_reg_6ghz_power {
 	RTW89_REG_6GHZ_POWER_DFLT = RTW89_REG_6GHZ_POWER_VLP,
 };
 
+#define RTW89_MIN_VALID_POWER_CONSTRAINT (-10) /* unit: dBm */
+
+/* calculate based on ieee80211 Transmit Power Envelope */
+struct rtw89_reg_6ghz_tpe {
+	bool valid;
+	s8 constraint; /* unit: dBm */
+};
+
 enum rtw89_fw_pkt_ofld_type {
 	RTW89_PKT_OFLD_TYPE_PROBE_RSP = 0,
 	RTW89_PKT_OFLD_TYPE_PS_POLL = 1,
@@ -3397,6 +3405,7 @@ struct rtw89_vif {
 	bool chanctx_assigned; /* only valid when running with chanctx_ops */
 	enum rtw89_sub_entity_idx sub_entity_idx;
 	enum rtw89_reg_6ghz_power reg_6ghz_power;
+	struct rtw89_reg_6ghz_tpe reg_6ghz_tpe;
 
 	u8 mac_id;
 	u8 port;
@@ -4941,6 +4950,7 @@ struct rtw89_regd {
 struct rtw89_regulatory_info {
 	const struct rtw89_regd *regd;
 	enum rtw89_reg_6ghz_power reg_6ghz_power;
+	struct rtw89_reg_6ghz_tpe reg_6ghz_tpe;
 	DECLARE_BITMAP(block_unii4, RTW89_REGD_MAX_COUNTRY_NUM);
 	DECLARE_BITMAP(block_6ghz, RTW89_REGD_MAX_COUNTRY_NUM);
 	DECLARE_BITMAP(block_6ghz_sp, RTW89_REGD_MAX_COUNTRY_NUM);
@@ -6535,8 +6545,8 @@ void rtw89_core_scan_start(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
 			   const u8 *mac_addr, bool hw_scan);
 void rtw89_core_scan_complete(struct rtw89_dev *rtwdev,
 			      struct ieee80211_vif *vif, bool hw_scan);
-void rtw89_reg_6ghz_power_recalc(struct rtw89_dev *rtwdev,
-				 struct rtw89_vif *rtwvif, bool active);
+int rtw89_reg_6ghz_recalc(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
+			  bool active);
 void rtw89_core_update_p2p_ps(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif);
 void rtw89_core_ntfy_btc_event(struct rtw89_dev *rtwdev, enum rtw89_btc_hmsg event);
 
diff --git a/drivers/net/wireless/realtek/rtw89/debug.c b/drivers/net/wireless/realtek/rtw89/debug.c
index 49bbbd049316..9cbf136aa2ac 100644
--- a/drivers/net/wireless/realtek/rtw89/debug.c
+++ b/drivers/net/wireless/realtek/rtw89/debug.c
@@ -818,6 +818,28 @@ static const struct dbgfs_txpwr_table *dbgfs_txpwr_tables[RTW89_CHIP_GEN_NUM] =
 	[RTW89_CHIP_BE] = &dbgfs_txpwr_table_be,
 };
 
+static
+void rtw89_debug_priv_txpwr_table_get_regd(struct seq_file *m,
+					   struct rtw89_dev *rtwdev,
+					   const struct rtw89_chan *chan)
+{
+	const struct rtw89_regulatory_info *regulatory = &rtwdev->regulatory;
+	const struct rtw89_reg_6ghz_tpe *tpe6 = &regulatory->reg_6ghz_tpe;
+
+	seq_printf(m, "[Chanctx] band %u, ch %u, bw %u\n",
+		   chan->band_type, chan->channel, chan->band_width);
+
+	seq_puts(m, "[Regulatory] ");
+	__print_regd(m, rtwdev, chan);
+
+	if (chan->band_type == RTW89_BAND_6G) {
+		seq_printf(m, "[reg6_pwr_type] %u\n", regulatory->reg_6ghz_power);
+
+		if (tpe6->valid)
+			seq_printf(m, "[TPE] %d dBm\n", tpe6->constraint);
+	}
+}
+
 static int rtw89_debug_priv_txpwr_table_get(struct seq_file *m, void *v)
 {
 	struct rtw89_debugfs_priv *debugfs_priv = m->private;
@@ -831,8 +853,7 @@ static int rtw89_debug_priv_txpwr_table_get(struct seq_file *m, void *v)
 	rtw89_leave_ps_mode(rtwdev);
 	chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
 
-	seq_puts(m, "[Regulatory] ");
-	__print_regd(m, rtwdev, chan);
+	rtw89_debug_priv_txpwr_table_get_regd(m, rtwdev, chan);
 
 	seq_puts(m, "[SAR]\n");
 	rtw89_print_sar(m, rtwdev, chan->freq);
diff --git a/drivers/net/wireless/realtek/rtw89/mac80211.c b/drivers/net/wireless/realtek/rtw89/mac80211.c
index 41b286da3d59..c8a0ad0cffc9 100644
--- a/drivers/net/wireless/realtek/rtw89/mac80211.c
+++ b/drivers/net/wireless/realtek/rtw89/mac80211.c
@@ -487,6 +487,9 @@ static void rtw89_ops_link_info_changed(struct ieee80211_hw *hw,
 	if (changed & BSS_CHANGED_CQM)
 		rtw89_fw_h2c_set_bcn_fltr_cfg(rtwdev, vif, true);
 
+	if (changed & BSS_CHANGED_TPE)
+		rtw89_reg_6ghz_recalc(rtwdev, rtwvif, true);
+
 	mutex_unlock(&rtwdev->mutex);
 }
 
diff --git a/drivers/net/wireless/realtek/rtw89/phy.c b/drivers/net/wireless/realtek/rtw89/phy.c
index 60f20e596321..e025f0ed9f59 100644
--- a/drivers/net/wireless/realtek/rtw89/phy.c
+++ b/drivers/net/wireless/realtek/rtw89/phy.c
@@ -1847,6 +1847,36 @@ static s8 rtw89_phy_txpwr_rf_to_mac(struct rtw89_dev *rtwdev, s8 txpwr_rf)
 	return txpwr_rf >> (chip->txpwr_factor_rf - chip->txpwr_factor_mac);
 }
 
+static s8 rtw89_phy_txpwr_dbm_to_mac(struct rtw89_dev *rtwdev, s8 dbm)
+{
+	const struct rtw89_chip_info *chip = rtwdev->chip;
+
+	return clamp_t(s16, dbm << chip->txpwr_factor_mac, -64, 63);
+}
+
+static s8 rtw89_phy_txpwr_dbm_without_tolerance(s8 dbm)
+{
+	const u8 tssi_deviation_point = 0;
+	const u8 tssi_max_deviation = 2;
+
+	if (dbm <= tssi_deviation_point)
+		dbm -= tssi_max_deviation;
+
+	return dbm;
+}
+
+static s8 rtw89_phy_get_tpe_constraint(struct rtw89_dev *rtwdev, u8 band)
+{
+	struct rtw89_regulatory_info *regulatory = &rtwdev->regulatory;
+	const struct rtw89_reg_6ghz_tpe *tpe = &regulatory->reg_6ghz_tpe;
+	s8 cstr = S8_MAX;
+
+	if (band == RTW89_BAND_6G && tpe->valid)
+		cstr = rtw89_phy_txpwr_dbm_without_tolerance(tpe->constraint);
+
+	return rtw89_phy_txpwr_dbm_to_mac(rtwdev, cstr);
+}
+
 s8 rtw89_phy_read_txpwr_byrate(struct rtw89_dev *rtwdev, u8 band, u8 bw,
 			       const struct rtw89_rate_desc *rate_desc)
 {
@@ -1921,6 +1951,7 @@ s8 rtw89_phy_read_txpwr_limit(struct rtw89_dev *rtwdev, u8 band,
 	u8 regd = rtw89_regd_get(rtwdev, band);
 	u8 reg6 = regulatory->reg_6ghz_power;
 	s8 lmt = 0, sar;
+	s8 cstr;
 
 	switch (band) {
 	case RTW89_BAND_2G:
@@ -1953,8 +1984,9 @@ s8 rtw89_phy_read_txpwr_limit(struct rtw89_dev *rtwdev, u8 band,
 
 	lmt = rtw89_phy_txpwr_rf_to_mac(rtwdev, lmt);
 	sar = rtw89_query_sar(rtwdev, freq);
+	cstr = rtw89_phy_get_tpe_constraint(rtwdev, band);
 
-	return min(lmt, sar);
+	return min3(lmt, sar, cstr);
 }
 EXPORT_SYMBOL(rtw89_phy_read_txpwr_limit);
 
@@ -2178,6 +2210,7 @@ s8 rtw89_phy_read_txpwr_limit_ru(struct rtw89_dev *rtwdev, u8 band,
 	u8 regd = rtw89_regd_get(rtwdev, band);
 	u8 reg6 = regulatory->reg_6ghz_power;
 	s8 lmt_ru = 0, sar;
+	s8 cstr;
 
 	switch (band) {
 	case RTW89_BAND_2G:
@@ -2210,8 +2243,9 @@ s8 rtw89_phy_read_txpwr_limit_ru(struct rtw89_dev *rtwdev, u8 band,
 
 	lmt_ru = rtw89_phy_txpwr_rf_to_mac(rtwdev, lmt_ru);
 	sar = rtw89_query_sar(rtwdev, freq);
+	cstr = rtw89_phy_get_tpe_constraint(rtwdev, band);
 
-	return min(lmt_ru, sar);
+	return min3(lmt_ru, sar, cstr);
 }
 
 static void
diff --git a/drivers/net/wireless/realtek/rtw89/regd.c b/drivers/net/wireless/realtek/rtw89/regd.c
index 1a133914f673..a251b0e3b16e 100644
--- a/drivers/net/wireless/realtek/rtw89/regd.c
+++ b/drivers/net/wireless/realtek/rtw89/regd.c
@@ -714,7 +714,154 @@ exit:
 	mutex_unlock(&rtwdev->mutex);
 }
 
-static void __rtw89_reg_6ghz_power_recalc(struct rtw89_dev *rtwdev)
+/* Maximum Transmit Power field (@raw) can be EIRP or PSD.
+ * Both units are 0.5 dB-based. Return a constraint in dB.
+ */
+static s8 tpe_get_constraint(s8 raw)
+{
+	const u8 hw_deviation = 3; /* unit: 0.5 dB */
+	const u8 antenna_gain = 10; /* unit: 0.5 dB */
+	const u8 array_gain = 6; /* unit: 0.5 dB */
+	const u8 offset = hw_deviation + antenna_gain + array_gain;
+
+	return (raw - offset) / 2;
+}
+
+static void tpe_intersect_constraint(struct rtw89_reg_6ghz_tpe *tpe, s8 cstr)
+{
+	if (tpe->valid) {
+		tpe->constraint = min(tpe->constraint, cstr);
+		return;
+	}
+
+	tpe->constraint = cstr;
+	tpe->valid = true;
+}
+
+static void tpe_deal_with_eirp(struct rtw89_reg_6ghz_tpe *tpe,
+			       const struct ieee80211_parsed_tpe_eirp *eirp)
+{
+	unsigned int i;
+	s8 cstr;
+
+	if (!eirp->valid)
+		return;
+
+	for (i = 0; i < eirp->count; i++) {
+		cstr = tpe_get_constraint(eirp->power[i]);
+		tpe_intersect_constraint(tpe, cstr);
+	}
+}
+
+static s8 tpe_convert_psd_to_eirp(s8 psd)
+{
+	static const unsigned int mlog20 = 1301;
+
+	return psd + 10 * mlog20 / 1000;
+}
+
+static void tpe_deal_with_psd(struct rtw89_reg_6ghz_tpe *tpe,
+			      const struct ieee80211_parsed_tpe_psd *psd)
+{
+	unsigned int i;
+	s8 cstr_psd;
+	s8 cstr;
+
+	if (!psd->valid)
+		return;
+
+	for (i = 0; i < psd->count; i++) {
+		cstr_psd = tpe_get_constraint(psd->power[i]);
+		cstr = tpe_convert_psd_to_eirp(cstr_psd);
+		tpe_intersect_constraint(tpe, cstr);
+	}
+}
+
+static void rtw89_calculate_tpe(struct rtw89_dev *rtwdev,
+				struct rtw89_reg_6ghz_tpe *result_tpe,
+				const struct ieee80211_parsed_tpe *parsed_tpe)
+{
+	static const u8 category = IEEE80211_TPE_CAT_6GHZ_DEFAULT;
+
+	tpe_deal_with_eirp(result_tpe, &parsed_tpe->max_local[category]);
+	tpe_deal_with_eirp(result_tpe, &parsed_tpe->max_reg_client[category]);
+	tpe_deal_with_psd(result_tpe, &parsed_tpe->psd_local[category]);
+	tpe_deal_with_psd(result_tpe, &parsed_tpe->psd_reg_client[category]);
+}
+
+static bool __rtw89_reg_6ghz_tpe_recalc(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_regulatory_info *regulatory = &rtwdev->regulatory;
+	struct rtw89_reg_6ghz_tpe new = {};
+	struct rtw89_vif *rtwvif;
+	bool changed = false;
+
+	rtw89_for_each_rtwvif(rtwdev, rtwvif) {
+		const struct rtw89_reg_6ghz_tpe *tmp;
+		const struct rtw89_chan *chan;
+
+		chan = rtw89_chan_get(rtwdev, rtwvif->sub_entity_idx);
+		if (chan->band_type != RTW89_BAND_6G)
+			continue;
+
+		tmp = &rtwvif->reg_6ghz_tpe;
+		if (!tmp->valid)
+			continue;
+
+		tpe_intersect_constraint(&new, tmp->constraint);
+	}
+
+	if (memcmp(&regulatory->reg_6ghz_tpe, &new,
+		   sizeof(regulatory->reg_6ghz_tpe)) != 0)
+		changed = true;
+
+	if (changed) {
+		if (new.valid)
+			rtw89_debug(rtwdev, RTW89_DBG_REGD,
+				    "recalc 6 GHz reg TPE to %d dBm\n",
+				    new.constraint);
+		else
+			rtw89_debug(rtwdev, RTW89_DBG_REGD,
+				    "recalc 6 GHz reg TPE to none\n");
+
+		regulatory->reg_6ghz_tpe = new;
+	}
+
+	return changed;
+}
+
+static int rtw89_reg_6ghz_tpe_recalc(struct rtw89_dev *rtwdev,
+				     struct rtw89_vif *rtwvif, bool active,
+				     unsigned int *changed)
+{
+	struct ieee80211_vif *vif = rtwvif_to_vif(rtwvif);
+	struct ieee80211_bss_conf *bss_conf = &vif->bss_conf;
+	struct rtw89_reg_6ghz_tpe *tpe = &rtwvif->reg_6ghz_tpe;
+
+	memset(tpe, 0, sizeof(*tpe));
+
+	if (!active || rtwvif->reg_6ghz_power != RTW89_REG_6GHZ_POWER_STD)
+		goto bottom;
+
+	rtw89_calculate_tpe(rtwdev, tpe, &bss_conf->tpe);
+	if (!tpe->valid)
+		goto bottom;
+
+	if (tpe->constraint < RTW89_MIN_VALID_POWER_CONSTRAINT) {
+		rtw89_err(rtwdev,
+			  "%s: constraint %d dBm is less than min valid val\n",
+			  __func__, tpe->constraint);
+
+		tpe->valid = false;
+		return -EINVAL;
+	}
+
+bottom:
+	*changed += __rtw89_reg_6ghz_tpe_recalc(rtwdev);
+	return 0;
+}
+
+static bool __rtw89_reg_6ghz_power_recalc(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_regulatory_info *regulatory = &rtwdev->regulatory;
 	const struct rtw89_regd *regd = regulatory->regd;
@@ -751,23 +898,21 @@ static void __rtw89_reg_6ghz_power_recalc(struct rtw89_dev *rtwdev)
 	}
 
 	if (regulatory->reg_6ghz_power == sel)
-		return;
+		return false;
 
 	rtw89_debug(rtwdev, RTW89_DBG_REGD,
 		    "recalc 6 GHz reg power type to %d\n", sel);
 
 	regulatory->reg_6ghz_power = sel;
-
-	rtw89_core_set_chip_txpwr(rtwdev);
+	return true;
 }
 
-void rtw89_reg_6ghz_power_recalc(struct rtw89_dev *rtwdev,
-				 struct rtw89_vif *rtwvif, bool active)
+static int rtw89_reg_6ghz_power_recalc(struct rtw89_dev *rtwdev,
+				       struct rtw89_vif *rtwvif, bool active,
+				       unsigned int *changed)
 {
 	struct ieee80211_vif *vif = rtwvif_to_vif(rtwvif);
 
-	lockdep_assert_held(&rtwdev->mutex);
-
 	if (active) {
 		switch (vif->bss_conf.power_type) {
 		case IEEE80211_REG_VLP_AP:
@@ -787,5 +932,32 @@ void rtw89_reg_6ghz_power_recalc(struct rtw89_dev *rtwdev,
 		rtwvif->reg_6ghz_power = RTW89_REG_6GHZ_POWER_DFLT;
 	}
 
-	__rtw89_reg_6ghz_power_recalc(rtwdev);
+	*changed += __rtw89_reg_6ghz_power_recalc(rtwdev);
+	return 0;
+}
+
+int rtw89_reg_6ghz_recalc(struct rtw89_dev *rtwdev, struct rtw89_vif *rtwvif,
+			  bool active)
+{
+	unsigned int changed = 0;
+	int ret;
+
+	lockdep_assert_held(&rtwdev->mutex);
+
+	/* The result of reg_6ghz_tpe may depend on reg_6ghz_power type,
+	 * so must do reg_6ghz_tpe_recalc() after reg_6ghz_power_recalc().
+	 */
+
+	ret = rtw89_reg_6ghz_power_recalc(rtwdev, rtwvif, active, &changed);
+	if (ret)
+		return ret;
+
+	ret = rtw89_reg_6ghz_tpe_recalc(rtwdev, rtwvif, active, &changed);
+	if (ret)
+		return ret;
+
+	if (changed)
+		rtw89_core_set_chip_txpwr(rtwdev);
+
+	return 0;
 }
-- 
cgit v1.2.3-58-ga151


From f77c5179bc9f6f22d6f1b064bda2e87a567ca7d5 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 27 Jun 2024 10:58:46 +0800
Subject: wifi: rtw89: 8852bt: rfk: add IQK

IQ signal calibration is a very important calibration to yield good RF
performance. We do this calibration once we are going to an AP. During
scanning phase, without this calibration RF performance is still
acceptable because it transmits and receives with low data rate at
this phase.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240627025849.25198-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/reg.h           |   1 +
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c | 892 +++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h |   1 +
 3 files changed, 894 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index 177981a108e0..24e8d244534e 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -8659,6 +8659,7 @@
 #define R_P0_RFCTM 0x5864
 #define B_P0_CLKG_FORCE GENMASK(31, 30)
 #define B_P0_RFCTM_EN BIT(29)
+#define B_P0_GOT_TXRX GENMASK(28, 27)
 #define B_P0_RFCTM_VAL GENMASK(25, 20)
 #define R_P0_RFCTM_RDY BIT(26)
 #define R_P0_TRSW 0x5868
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
index d0a40eabfe2d..4a638e2444db 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
@@ -13,6 +13,7 @@
 #include "rtw8852bt_rfk_table.h"
 #include "rtw8852b_common.h"
 
+#define RTW8852BT_IQK_VER 0x2a
 #define RTW8852BT_SS 2
 #define RTW8852BT_TSSI_PATH_NR 2
 #define RTW8852BT_DPK_VER 0x06
@@ -67,6 +68,23 @@ enum dpk_agc_step {
 	DPK_AGC_STEP_SET_TX_GAIN,
 };
 
+enum rtw8852bt_iqk_type {
+	ID_TXAGC = 0x0,
+	ID_FLOK_COARSE = 0x1,
+	ID_FLOK_FINE = 0x2,
+	ID_TXK = 0x3,
+	ID_RXAGC = 0x4,
+	ID_RXK = 0x5,
+	ID_NBTXK = 0x6,
+	ID_NBRXK = 0x7,
+	ID_FLOK_VBUFFER = 0x8,
+	ID_A_FLOK_COARSE = 0x9,
+	ID_G_FLOK_COARSE = 0xa,
+	ID_A_FLOK_FINE = 0xb,
+	ID_G_FLOK_FINE = 0xc,
+	ID_IQK_RESTORE = 0x10,
+};
+
 enum adc_ck {
 	ADC_NA = 0,
 	ADC_480M = 1,
@@ -347,6 +365,813 @@ static void _rfk_bb_afe_restore(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy
 	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_MV_AVG, B_P1_TXPW_RSTB, 0x2);
 }
 
+static bool _iqk_check_cal(struct rtw89_dev *rtwdev, u8 path, u8 ktype)
+{
+	bool notready = false;
+	u32 val;
+	int ret;
+
+	ret = read_poll_timeout_atomic(rtw89_phy_read32_mask, val, val == 0x55,
+				       10, 8200, false,
+				       rtwdev, R_RFK_ST, MASKBYTE0);
+	if (ret)
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]NCTL1 IQK timeout!!!\n");
+
+	udelay(10);
+
+	ret = read_poll_timeout_atomic(rtw89_phy_read32_mask, val, val == 0x8000,
+				       10, 400, false,
+				       rtwdev, R_RPT_COM, B_RPT_COM_RDY);
+	if (ret) {
+		notready = true;
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]NCTL2 IQK timeout!!!\n");
+	}
+
+	udelay(10);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, MASKBYTE0, 0x0);
+
+	return notready;
+}
+
+static bool _iqk_one_shot(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx,
+			  u8 path, u8 ktype)
+{
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+	u32 iqk_cmd;
+	bool fail;
+
+	switch (ktype) {
+	case ID_TXAGC:
+		iqk_cmd = 0x008 | (1 << (4 + path)) | (path << 1);
+		break;
+	case ID_FLOK_COARSE:
+		iqk_cmd = 0x108 | (1 << (4 + path));
+		break;
+	case ID_FLOK_FINE:
+		iqk_cmd = 0x208 | (1 << (4 + path));
+		break;
+	case ID_FLOK_VBUFFER:
+		iqk_cmd = 0x308 | (1 << (4 + path));
+		break;
+	case ID_TXK:
+		iqk_cmd = 0x008 | (1 << (path + 4)) |
+			  (((0x8 + iqk_info->iqk_bw[path]) & 0xf) << 8);
+		break;
+	case ID_RXAGC:
+		iqk_cmd = 0x508 | (1 << (4 + path)) | (path << 1);
+		break;
+	case ID_RXK:
+		iqk_cmd = 0x008 | (1 << (path + 4)) |
+			  (((0xb + iqk_info->iqk_bw[path]) & 0xf) << 8);
+		break;
+	case ID_NBTXK:
+		iqk_cmd = 0x408 | (1 << (4 + path));
+		break;
+	case ID_NBRXK:
+		iqk_cmd = 0x608 | (1 << (4 + path));
+		break;
+	default:
+		return false;
+	}
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]===>%s, iqk_cmd = %x\n",
+		    __func__, iqk_cmd + 1);
+
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_CFG, MASKDWORD, iqk_cmd + 1);
+	fail = _iqk_check_cal(rtwdev, path, ktype);
+
+	return fail;
+}
+
+static void _iqk_txk_setting(struct rtw89_dev *rtwdev, u8 path)
+{
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+
+	switch (iqk_info->iqk_band[path]) {
+	case RTW89_BAND_2G:
+		rtw89_write_rf(rtwdev, path, RR_TXG1, RR_TXG1_ATT2, 0x0);
+		rtw89_write_rf(rtwdev, path, RR_TXG1, RR_TXG1_ATT1, 0x0);
+		rtw89_write_rf(rtwdev, path, RR_TXG2, RR_TXG2_ATT0, 0x1);
+		rtw89_write_rf(rtwdev, path, RR_TXGA, RR_TXGA_LOK_EXT, 0x0);
+		rtw89_write_rf(rtwdev, path, RR_LUTWE, RR_LUTWE_LOK, 0x1);
+		rtw89_write_rf(rtwdev, path, RR_LUTWA, RR_LUTWA_M1, 0x00);
+		rtw89_write_rf(rtwdev, path, RR_MOD, RR_MOD_IQK, 0x403e);
+		rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_GR0, 0x0);
+		rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_GR1, 0x5);
+		udelay(1);
+		break;
+	case RTW89_BAND_5G:
+		rtw89_write_rf(rtwdev, path, RR_BIASA, RR_BIASA_A, 0x1);
+		rtw89_write_rf(rtwdev, path, RR_TXGA, RR_TXGA_LOK_EXT, 0x0);
+		rtw89_write_rf(rtwdev, path, RR_LUTWE, RR_LUTWE_LOK, 0x1);
+		rtw89_write_rf(rtwdev, path, RR_LUTWA, RR_LUTWA_M1, 0x80);
+		rtw89_write_rf(rtwdev, path, RR_MOD, RR_MOD_IQK, 0x403e);
+		rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_GR0, 0x0);
+		rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_GR1, 0x4);
+		udelay(1);
+		break;
+	default:
+		break;
+	}
+}
+
+static bool _iqk_2g_lok(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx, u8 path)
+{
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]===>%s\n", __func__);
+
+	rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_TG, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8), B_KIP_IQP_IQSW, 0x09);
+	rtw89_phy_write32_mask(rtwdev, R_IQK_DIF4, B_IQK_DIF4_TXT, 0x021);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_CFG, MASKDWORD, 0x00000119 + (path << 4));
+
+	_iqk_check_cal(rtwdev, path, ID_FLOK_COARSE);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+	rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_TG, 0x12);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8), B_KIP_IQP_IQSW, 0x24);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_CFG, MASKDWORD, 0x00000319 + (path << 4));
+
+	_iqk_check_cal(rtwdev, path, ID_FLOK_VBUFFER);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+	rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_TG, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8), B_KIP_IQP_IQSW, 0x09);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_CFG, MASKDWORD, 0x00000219 + (path << 4));
+
+	_iqk_check_cal(rtwdev, path, ID_FLOK_COARSE);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+	rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_TG, 0x12);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8), B_KIP_IQP_IQSW, 0x24);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_CFG, MASKDWORD, 0x00000319 + (path << 4));
+
+	_iqk_check_cal(rtwdev, path, ID_FLOK_VBUFFER);
+
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+	return false;
+}
+
+static bool _iqk_5g_lok(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx, u8 path)
+{
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]===>%s\n", __func__);
+
+	rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_TG, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8), B_KIP_IQP_IQSW, 0x09);
+	rtw89_phy_write32_mask(rtwdev, R_IQK_DIF4, B_IQK_DIF4_TXT, 0x021);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_CFG, MASKDWORD, 0x00000119 + (path << 4));
+
+	_iqk_check_cal(rtwdev, path, ID_FLOK_COARSE);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+	rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_TG, 0x12);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8), B_KIP_IQP_IQSW, 0x24);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_CFG, MASKDWORD, 0x00000319 + (path << 4));
+
+	_iqk_check_cal(rtwdev, path, ID_FLOK_VBUFFER);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+	rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_TG, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8), B_KIP_IQP_IQSW, 0x09);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_CFG, MASKDWORD, 0x00000219 + (path << 4));
+
+	_iqk_check_cal(rtwdev, path, ID_FLOK_COARSE);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+	rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_TG, 0x12);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8), B_KIP_IQP_IQSW, 0x24);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_CFG, MASKDWORD, 0x00000319 + (path << 4));
+
+	_iqk_check_cal(rtwdev, path, ID_FLOK_VBUFFER);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+	return false;
+}
+
+static bool _iqk_2g_tx(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx, u8 path)
+{
+	static const u32 g_power_range[4] = {0x0, 0x0, 0x0, 0x0};
+	static const u32 g_track_range[4] = {0x4, 0x4, 0x6, 0x6};
+	static const u32 g_gain_bb[4] = {0x08, 0x0e, 0x08, 0x0e};
+	static const u32 g_itqt[4] = {0x09, 0x12, 0x1b, 0x24};
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+	bool notready = false;
+	bool kfail = false;
+	u8 gp;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]===>%s\n", __func__);
+
+	for (gp = 0x0; gp < 0x4; gp++) {
+		rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_GR0,
+			       g_power_range[gp]);
+		rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_GR1,
+			       g_track_range[gp]);
+		rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_TG,
+			       g_gain_bb[gp]);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000100, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000010, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000004, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000003, gp);
+		rtw89_phy_write32_mask(rtwdev, R_IQK_DIF4, B_IQK_DIF4_TXT,
+				       0x009);
+		rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8),
+				       B_KIP_IQP_IQSW, g_itqt[gp]);
+		notready = _iqk_one_shot(rtwdev, phy_idx, path, ID_NBTXK);
+		iqk_info->nb_txcfir[path] =
+			rtw89_phy_read32_mask(rtwdev, R_TXIQC + (path << 8), MASKDWORD);
+
+		if (iqk_info->is_nbiqk)
+			break;
+
+		rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8),
+				       B_KIP_IQP_IQSW, g_itqt[gp]);
+		notready = _iqk_one_shot(rtwdev, phy_idx, path, ID_TXK);
+		rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[IQK]S%x, gp = 0x%x, 0x8%x38 = 0x%x\n",
+			    path, gp, 1 << path, iqk_info->nb_txcfir[path]);
+	}
+
+	if (!notready)
+		kfail = !!rtw89_phy_read32_mask(rtwdev, R_NCTL_RPT, B_NCTL_RPT_FLG);
+
+	if (kfail) {
+		iqk_info->nb_txcfir[path] = 0x40000002;
+		rtw89_phy_write32_mask(rtwdev, R_IQK_RES + (path << 8),
+				       B_IQK_RES_TXCFIR, 0x0);
+	}
+
+	return kfail;
+}
+
+static bool _iqk_5g_tx(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx, u8 path)
+{
+	static const u32 a_power_range[4] = {0x0, 0x0, 0x0, 0x0};
+	static const u32 a_track_range[4] = {0x3, 0x3, 0x6, 0x6};
+	static const u32 a_gain_bb[4] = {0x08, 0x10, 0x08, 0x0e};
+	static const u32 a_itqt[4] = {0x09, 0x12, 0x1b, 0x24};
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+	bool notready = false;
+	bool kfail = false;
+	u8 gp;
+
+	for (gp = 0x0; gp < 0x4; gp++) {
+		rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_GR0, a_power_range[gp]);
+		rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_GR1, a_track_range[gp]);
+		rtw89_write_rf(rtwdev, path, RR_TXIG, RR_TXIG_TG, a_gain_bb[gp]);
+
+		rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8),
+				       MASKDWORD, a_itqt[gp]);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000100, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000010, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000004, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000003, gp);
+		rtw89_phy_write32_mask(rtwdev, R_IQK_DIF4, B_IQK_DIF4_TXT,
+				       0x009);
+		rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8),
+				       B_KIP_IQP_IQSW, a_itqt[gp]);
+
+		notready = _iqk_one_shot(rtwdev, phy_idx, path, ID_NBTXK);
+		iqk_info->nb_txcfir[path] =
+			rtw89_phy_read32_mask(rtwdev, R_TXIQC + (path << 8), MASKDWORD);
+
+		if (iqk_info->is_nbiqk)
+			break;
+
+		rtw89_phy_write32_mask(rtwdev, R_KIP_IQP + (path << 8),
+				       B_KIP_IQP_IQSW, a_itqt[gp]);
+		notready = _iqk_one_shot(rtwdev, phy_idx, path, ID_TXK);
+		rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[IQK]S%x, gp = 0x%x, 0x8%x38 = 0x%x\n",
+			    path, gp, 1 << path, iqk_info->nb_txcfir[path]);
+	}
+
+	if (!notready)
+		kfail = !!rtw89_phy_read32_mask(rtwdev, R_NCTL_RPT, B_NCTL_RPT_FLG);
+
+	if (kfail) {
+		iqk_info->nb_txcfir[path] = 0x40000002;
+		rtw89_phy_write32_mask(rtwdev, R_IQK_RES + (path << 8),
+				       B_IQK_RES_TXCFIR, 0x0);
+	}
+
+	return kfail;
+}
+
+static void _iqk_adc_fifo_rst(struct rtw89_dev *rtwdev,
+			      enum rtw89_phy_idx phy_idx, u8 path)
+{
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x0303);
+	udelay(10);
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x3333);
+}
+
+static void _iqk_rxclk_setting(struct rtw89_dev *rtwdev, u8 path)
+{
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]===>%s\n", __func__);
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x0303);
+
+	if (iqk_info->iqk_bw[path] == RTW89_CHANNEL_WIDTH_80) {
+		_rxck_force(rtwdev, RF_PATH_A, true, ADC_960M);
+		_rxck_force(rtwdev, RF_PATH_B, true, ADC_960M);
+		udelay(1);
+
+		rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC,
+				       B_UPD_CLK_ADC_ON, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC,
+				       B_UPD_CLK_ADC_VAL, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_SAMPL_DLY_T_V1,
+				       B_PATH0_SAMPL_DLY_T_MSK_V1, 0x2);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_SAMPL_DLY_T_V1,
+				       B_PATH1_SAMPL_DLY_T_MSK_V1, 0x2);
+		rtw89_phy_write32_mask(rtwdev, R_P0_CFCH_BW1, B_P0_CFCH_BW1, 0x8);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_BW_SEL_V1,
+				       B_PATH1_BW_SEL_MSK_V1, 0x8);
+	} else {
+		_rxck_force(rtwdev, RF_PATH_A, true, ADC_480M);
+		_rxck_force(rtwdev, RF_PATH_B, true, ADC_480M);
+		udelay(1);
+
+		rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC,
+				       B_UPD_CLK_ADC_ON, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC,
+				       B_UPD_CLK_ADC_VAL, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_PATH0_SAMPL_DLY_T_V1,
+				       B_PATH0_SAMPL_DLY_T_MSK_V1, 0x3);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_SAMPL_DLY_T_V1,
+				       B_PATH1_SAMPL_DLY_T_MSK_V1, 0x3);
+		rtw89_phy_write32_mask(rtwdev, R_P0_CFCH_BW1, B_P0_CFCH_BW1, 0xf);
+		rtw89_phy_write32_mask(rtwdev, R_PATH1_BW_SEL_V1,
+				       B_PATH1_BW_SEL_MSK_V1, 0xf);
+	}
+
+	rtw89_phy_write32_mask(rtwdev, R_PATH0_SAMPL_DLY_T_V1, 0x00000780, 0x8);
+	rtw89_phy_write32_mask(rtwdev, R_PATH1_SAMPL_DLY_T_V1, 0x00000780, 0x8);
+	rtw89_phy_write32_mask(rtwdev, R_PATH0_SAMPL_DLY_T_V1, 0x00007800, 0x2);
+	rtw89_phy_write32_mask(rtwdev, R_PATH1_SAMPL_DLY_T_V1, 0x00007800, 0x2);
+	rtw89_phy_write32_mask(rtwdev, R_DRCK, B_DRCK_MUL, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_DBG, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_DBGMOD, B_P1_DBGMOD_ON, 0x1);
+	udelay(1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, B_ANAPAR_PW15, 0x0f);
+	udelay(1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, B_ANAPAR_PW15, 0x03);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_15, 0xa001);
+	udelay(1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_15, 0xa041);
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x3333);
+}
+
+static bool _iqk_2g_rx(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx, u8 path)
+{
+	static const u32 g_idxrxgain[2] = {0x212, 0x310};
+	static const u32 g_idxattc2[2] = {0x00, 0x20};
+	static const u32 g_idxattc1[2] = {0x3, 0x2};
+	static const u32 g_idxrxagc[2] = {0x0, 0x2};
+	static const u32 g_idx[2] = {0x0, 0x2};
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+	bool notready = false;
+	bool kfail = false;
+	u32 rf_18, tmp;
+	u8 gp;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]===>%s\n", __func__);
+
+	rtw89_write_rf(rtwdev, path, RR_MOD, RR_MOD_MASK, 0xc);
+	rtw89_write_rf(rtwdev, path, RR_RXK, RR_RXK_SEL2G, 0x1);
+	rf_18 = rtw89_read_rf(rtwdev, path, RR_CFGCH, RFREG_MASK);
+	rtw89_write_rf(rtwdev, path, RR_RSV4, RFREG_MASK, rf_18);
+
+	for (gp = 0x0; gp < 0x2; gp++) {
+		rtw89_write_rf(rtwdev, path, RR_MOD, RR_MOD_RGM, g_idxrxgain[gp]);
+		rtw89_write_rf(rtwdev, path, RR_RXBB, RR_RXBB_C2G, g_idxattc2[gp]);
+		rtw89_write_rf(rtwdev, path, RR_RXBB, RR_RXBB_C1G, g_idxattc1[gp]);
+
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000100, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000010, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000007, g_idx[gp]);
+		rtw89_write_rf(rtwdev, path, RR_RXKPLL, RFREG_MASK, 0x80013);
+		udelay(100);
+		udelay(100);
+
+		tmp = rtw89_read_rf(rtwdev, path, RR_MOD, RFREG_MASK);
+		rtw89_phy_write32_mask(rtwdev, R_IQK_DIF2, B_IQK_DIF2_RXPI, tmp);
+		rtw89_phy_write32_mask(rtwdev, R_IQK_RXA, B_IQK_RXAGC, g_idxrxagc[gp]);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_IQK_DIF4, B_IQK_DIF4_RXT, 0x11);
+
+		notready = _iqk_one_shot(rtwdev, phy_idx, path, ID_RXAGC);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]S%x, rf rxbb  = %x\n", path,
+			    rtw89_read_rf(rtwdev, path, RR_MOD, 0x003c0));
+
+		rtw89_write_rf(rtwdev, path, RR_RXKPLL, RFREG_MASK, 0x80013);
+		udelay(100);
+		udelay(100);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_IQK_DIF4, B_IQK_DIF4_RXT, 0x011);
+
+		notready = _iqk_one_shot(rtwdev, phy_idx, path, ID_NBRXK);
+		iqk_info->nb_rxcfir[path] =
+			rtw89_phy_read32_mask(rtwdev, R_RXIQC + (path << 8),
+					      MASKDWORD) | 0x2;
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[IQK]S%x, gp = 0x%x, 0x8%x3c = 0x%x\n", path,
+			    g_idx[gp], 1 << path, iqk_info->nb_rxcfir[path]);
+
+		rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+		if (iqk_info->is_nbiqk)
+			break;
+
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+		notready = _iqk_one_shot(rtwdev, phy_idx, path, ID_RXK);
+		rtw89_phy_write32_mask(rtwdev, R_NCTL_N1,  B_NCTL_N1_CIP, 0x00);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+	}
+
+	if (!notready)
+		kfail = !!rtw89_phy_read32_mask(rtwdev, R_NCTL_RPT, B_NCTL_RPT_FLG);
+
+	if (kfail) {
+		iqk_info->nb_txcfir[path] = 0x40000002;
+		rtw89_phy_write32_mask(rtwdev, R_IQK_RES + (path << 8),
+				       B_IQK_RES_RXCFIR, 0x0);
+	}
+	rtw89_write_rf(rtwdev, path, RR_RXK, RR_RXK_SEL2G, 0x0);
+
+	return kfail;
+}
+
+static bool _iqk_5g_rx(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx, u8 path)
+{
+	static const u32 a_idxrxgain[2] = {0x110, 0x290};
+	static const u32 a_idxattc2[2] = {0x0f, 0x0f};
+	static const u32 a_idxattc1[2] = {0x2, 0x2};
+	static const u32 a_idxrxagc[2] = {0x4, 0x6};
+	static const u32 a_idx[2] = {0x0, 0x2};
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+	bool notready = false;
+	bool kfail = false;
+	u32 rf_18, tmp;
+	u8 gp;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]===>%s\n", __func__);
+
+	rtw89_write_rf(rtwdev, path, RR_MOD, RR_MOD_MASK, 0xc);
+	rtw89_write_rf(rtwdev, path, RR_RXK, RR_RXK_SEL5G, 0x1);
+	rf_18 = rtw89_read_rf(rtwdev, path, RR_CFGCH, RFREG_MASK);
+	rtw89_write_rf(rtwdev, path, RR_RSV4, RFREG_MASK, rf_18);
+
+	for (gp = 0x0; gp < 0x2; gp++) {
+		rtw89_write_rf(rtwdev, path, RR_MOD, RR_MOD_RGM, a_idxrxgain[gp]);
+		rtw89_write_rf(rtwdev, path, RR_RXA2, RR_RXA2_HATT, a_idxattc2[gp]);
+		rtw89_write_rf(rtwdev, path, RR_RXA2, RR_RXA2_CC2, a_idxattc1[gp]);
+
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000100, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000010, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8),
+				       0x00000007, a_idx[gp]);
+		rtw89_write_rf(rtwdev, path, RR_RXKPLL, RFREG_MASK, 0x80013);
+		udelay(100);
+		udelay(100);
+
+		tmp = rtw89_read_rf(rtwdev, path, RR_MOD, RFREG_MASK);
+		rtw89_phy_write32_mask(rtwdev, R_IQK_DIF2, B_IQK_DIF2_RXPI, tmp);
+		rtw89_phy_write32_mask(rtwdev, R_IQK_RXA, B_IQK_RXAGC, a_idxrxagc[gp]);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_IQK_DIF4, B_IQK_DIF4_RXT, 0x11);
+
+		notready = _iqk_one_shot(rtwdev, phy_idx, path, ID_RXAGC);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]S%x, rf rxbb  = %x\n", path,
+			    rtw89_read_rf(rtwdev, path, RR_MOD, 0x003c0));
+
+		rtw89_write_rf(rtwdev, path, RR_RXKPLL, RFREG_MASK, 0x80013);
+		udelay(200);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+		rtw89_phy_write32_mask(rtwdev, R_IQK_DIF4, B_IQK_DIF4_RXT, 0x011);
+		notready = _iqk_one_shot(rtwdev, phy_idx, path, ID_NBRXK);
+		iqk_info->nb_rxcfir[path] =
+			rtw89_phy_read32_mask(rtwdev, R_RXIQC + (path << 8),
+					      MASKDWORD) | 0x2;
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[IQK]S%x, gp = 0x%x, 0x8%x3c = 0x%x\n",
+			    path, a_idx[gp], 1 << path, iqk_info->nb_rxcfir[path]);
+		rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+
+		if (iqk_info->is_nbiqk)
+			break;
+
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x1);
+		notready = _iqk_one_shot(rtwdev, phy_idx, path, ID_RXK);
+		rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+		rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_RFCTM_EN, 0x0);
+	}
+
+	if (!notready)
+		kfail = !!rtw89_phy_read32_mask(rtwdev, R_NCTL_RPT, B_NCTL_RPT_FLG);
+
+	if (kfail) {
+		iqk_info->nb_txcfir[path] = 0x40000002;
+		rtw89_phy_write32_mask(rtwdev, R_IQK_RES + (path << 8),
+				       B_IQK_RES_RXCFIR, 0x0);
+	}
+	rtw89_write_rf(rtwdev, path, RR_RXK, RR_RXK_SEL2G, 0x0);
+
+	return kfail;
+}
+
+static void _iqk_by_path(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx, u8 path)
+{
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+	bool lok_result = false;
+	bool txk_result = false;
+	bool rxk_result = false;
+	u8 i;
+
+	for (i = 0; i < 3; i++) {
+		_iqk_txk_setting(rtwdev, path);
+		if (iqk_info->iqk_band[path] == RTW89_BAND_2G)
+			lok_result = _iqk_2g_lok(rtwdev, phy_idx, path);
+		else
+			lok_result = _iqk_5g_lok(rtwdev, phy_idx, path);
+
+		if (!lok_result)
+			break;
+	}
+
+	if (lok_result) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "[IQK]!!!!!!!!!!LOK by Pass !!!!!!!!!!!\n");
+		rtw89_write_rf(rtwdev, path, RR_DTXLOK, RFREG_MASK, 0x80200);
+		rtw89_write_rf(rtwdev, path, RR_RSV2, RFREG_MASK, 0x80200);
+		rtw89_write_rf(rtwdev, path, RR_LOKVB, RFREG_MASK, 0x80200);
+	}
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]RF_0x08[00:19] = 0x%x\n",
+		    rtw89_read_rf(rtwdev, path, RR_DTXLOK, RFREG_MASK));
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]RF_0x09[00:19] = 0x%x\n",
+		    rtw89_read_rf(rtwdev, path, RR_RSV2, RFREG_MASK));
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]RF_0x0a[00:19] = 0x%x\n",
+		    rtw89_read_rf(rtwdev, path, RR_LOKVB, RFREG_MASK));
+
+	if (iqk_info->iqk_band[path] == RTW89_BAND_2G)
+		txk_result = _iqk_2g_tx(rtwdev, phy_idx, path);
+	else
+		txk_result = _iqk_5g_tx(rtwdev, phy_idx, path);
+
+	_iqk_rxclk_setting(rtwdev, path);
+	_iqk_adc_fifo_rst(rtwdev, phy_idx, path);
+
+	if (iqk_info->iqk_band[path] == RTW89_BAND_2G)
+		rxk_result = _iqk_2g_rx(rtwdev, phy_idx, path);
+	else
+		rxk_result = _iqk_5g_rx(rtwdev, phy_idx, path);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[IQK]result  : lok_= %x, txk_= %x, rxk_= %x\n",
+		    lok_result, txk_result, rxk_result);
+}
+
+static void _iqk_get_ch_info(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy, u8 path)
+{
+	const struct rtw89_chan *chan = rtw89_chan_get(rtwdev, RTW89_SUB_ENTITY_0);
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+	u8 get_empty_table = false;
+	u32 reg_rf18;
+	u32 reg_35c;
+	u8 idx;
+
+	for (idx = 0; idx < RTW89_IQK_CHS_NR; idx++) {
+		if (iqk_info->iqk_mcc_ch[idx][path] == 0) {
+			get_empty_table = true;
+			break;
+		}
+	}
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK] (1)idx = %x\n", idx);
+
+	if (!get_empty_table) {
+		idx = iqk_info->iqk_table_idx[path] + 1;
+		if (idx > 1)
+			idx = 0;
+	}
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK] (2)idx = %x\n", idx);
+
+	reg_rf18 = rtw89_read_rf(rtwdev, path, RR_CFGCH, RFREG_MASK);
+	reg_35c = rtw89_phy_read32_mask(rtwdev, R_CIRST, B_CIRST_SYN);
+
+	iqk_info->iqk_band[path] = chan->band_type;
+	iqk_info->iqk_bw[path] = chan->band_width;
+	iqk_info->iqk_ch[path] = chan->channel;
+	iqk_info->iqk_mcc_ch[idx][path] = chan->channel;
+	iqk_info->iqk_table_idx[path] = idx;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]S%x, 0x18= 0x%x, idx = %x\n",
+		    path, reg_rf18, idx);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]S%x, 0x18= 0x%x\n",
+		    path, reg_rf18);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]S%x, 0x35c= 0x%x\n",
+		    path, reg_35c);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]times = 0x%x, ch =%x\n",
+		    iqk_info->iqk_times, idx);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]iqk_mcc_ch[%x][%x] = 0x%x\n",
+		    idx, path, iqk_info->iqk_mcc_ch[idx][path]);
+}
+
+static void _iqk_start_iqk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx, u8 path)
+{
+	_iqk_by_path(rtwdev, phy_idx, path);
+}
+
+static void _iqk_restore(struct rtw89_dev *rtwdev, u8 path)
+{
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "===> %s\n", __func__);
+
+	if (iqk_info->is_nbiqk) {
+		rtw89_phy_write32_mask(rtwdev, R_TXIQC + (path << 8),
+				       MASKDWORD, iqk_info->nb_txcfir[path]);
+		rtw89_phy_write32_mask(rtwdev, R_RXIQC + (path << 8),
+				       MASKDWORD, iqk_info->nb_rxcfir[path]);
+	} else {
+		rtw89_phy_write32_mask(rtwdev, R_TXIQC + (path << 8),
+				       MASKDWORD, 0x40000000);
+		rtw89_phy_write32_mask(rtwdev, R_RXIQC + (path << 8),
+				       MASKDWORD, 0x40000000);
+	}
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_CFG, MASKDWORD,
+			       0x00000e19 + (path << 4));
+
+	_iqk_check_cal(rtwdev, path, 0x0);
+
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_N1, B_NCTL_N1_CIP, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_RPT, MASKDWORD, 0x00000000);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_SYSCFG, MASKDWORD, 0x80000000);
+
+	rtw89_phy_write32_mask(rtwdev, R_KIP_CLK, MASKDWORD, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_IQRSN, B_IQRSN_K2, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_CFIR_SYS + (path << 8), BIT(28), 0x0);
+
+	rtw89_write_rf(rtwdev, path, RR_LUTWE, RR_LUTWE_LOK, 0x0);
+	rtw89_write_rf(rtwdev, path, RR_LUTWE, RR_LUTWE_LOK, 0x0);
+	rtw89_write_rf(rtwdev, path, RR_MOD, RR_MOD_MASK, 0x3);
+	rtw89_write_rf(rtwdev, path, RR_RSV1, RR_RSV1_RST, 0x1);
+	rtw89_write_rf(rtwdev, path, RR_BBDC, RR_BBDC_SEL, 0x1);
+}
+
+static void _iqk_afebb_restore(struct rtw89_dev *rtwdev,
+			       enum rtw89_phy_idx phy_idx, u8 path)
+{
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "===> %s\n", __func__);
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x0303);
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_DBG, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P1_DBGMOD, B_P1_DBGMOD_ON, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_CLKG_FORCE, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P1_RFCTM, B_P1_CLKG_FORCE, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_TXCKEN_FORCE, B_TXCKEN_FORCE_ALL, 0x0000000);
+	rtw89_phy_write32_mask(rtwdev, R_FAHM, B_RXTD_CKEN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC, B_UPD_GEN_ON, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_TX_COLLISION_T2R_ST, 0x0000001f, 0x03);
+	rtw89_phy_write32_mask(rtwdev, R_TX_COLLISION_T2R_ST, 0x000003e0, 0x03);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RXCK, B_P0_TXCK_ALL, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_P1_RXCK, B_P1_TXCK_ALL, 0x00);
+	rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC,
+			       B_UPD_CLK_ADC_VAL | B_UPD_CLK_ADC_ON, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_PD_CTRL, B_PD_HIT_DIS, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_RXCCA, B_RXCCA_DIS, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x0000);
+	rtw89_phy_write32_mask(rtwdev, R_P0_ADCFF_EN, B_P0_ADCFF_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P1_ADCFF_EN, B_P1_ADCFF_EN, 0x0);
+}
+
+static void _iqk_preset(struct rtw89_dev *rtwdev, u8 path)
+{
+	u8 idx = 0;
+
+	rtw89_phy_write32_mask(rtwdev, R_COEF_SEL + (path << 8), 0x00000001, idx);
+	rtw89_phy_write32_mask(rtwdev, R_CFIR_LUT + (path << 8), 0x00000008, idx);
+	rtw89_phy_write32_mask(rtwdev, R_TXIQC + (path << 8), MASKDWORD, 0x40000000);
+	rtw89_phy_write32_mask(rtwdev, R_RXIQC + (path << 8), MASKDWORD, 0x40000000);
+
+	rtw89_write_rf(rtwdev, path, RR_RSV1, RR_RSV1_RST, 0x0);
+	rtw89_write_rf(rtwdev, path, RR_BBDC, RR_BBDC_SEL, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_NCTL_RPT, MASKDWORD, 0x00000080);
+	rtw89_phy_write32_mask(rtwdev, R_KIP_SYSCFG, MASKDWORD, 0x81ff010a);
+}
+
+static void _iqk_macbb_setting(struct rtw89_dev *rtwdev,
+			       enum rtw89_phy_idx phy_idx, u8 path)
+{
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x0303);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_GOT_TXRX, 0x3);
+	rtw89_phy_write32_mask(rtwdev, R_P1_RFCTM, B_P1_GOT_TXRX, 0x3);
+	rtw89_phy_write32_mask(rtwdev, R_P0_ADCFF_EN, B_P0_ADCFF_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_ADCFF_EN, B_P1_ADCFF_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P0_RFCTM, B_P0_CLKG_FORCE, 0x3);
+	rtw89_phy_write32_mask(rtwdev, R_P1_RFCTM, B_P0_CLKG_FORCE, 0x3);
+	rtw89_phy_write32_mask(rtwdev, R_TXCKEN_FORCE, B_TXCKEN_FORCE_ALL, 0x1ffffff);
+	rtw89_phy_write32_mask(rtwdev, R_FAHM, B_RXTD_CKEN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC, B_UPD_GEN_ON, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_TX_COLLISION_T2R_ST, B_TXRX_FORCE_VAL, 0x3ff);
+	rtw89_phy_write32_mask(rtwdev, R_IOQ_IQK_DPK, B_IOQ_IQK_DPK_CLKEN,  0x3);
+	rtw89_phy_write32_mask(rtwdev, R_IQK_DPK_RST, B_IQK_DPK_RST, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P0_PATH_RST, B_P0_PATH_RST, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_PATH_RST, B_P1_PATH_RST, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_PD_CTRL, B_PD_HIT_DIS, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_RXCCA, B_RXCCA_DIS, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_DCFO_WEIGHT, B_DAC_CLK_IDX, 0x1);
+
+	_txck_force(rtwdev, RF_PATH_A, true, DAC_960M);
+	_txck_force(rtwdev, RF_PATH_B, true, DAC_960M);
+	_rxck_force(rtwdev, RF_PATH_A, true, ADC_1920M);
+	_rxck_force(rtwdev, RF_PATH_B, true, ADC_1920M);
+
+	rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC, B_UPD_CLK_ADC_ON, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_UPD_CLK_ADC, B_UPD_CLK_ADC_VAL, 0x2);
+
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_DBG, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_P1_DBGMOD, B_P1_DBGMOD_ON, 0x1);
+	udelay(10);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, B_ANAPAR_PW15, 0x1f);
+	udelay(10);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, B_ANAPAR_PW15, 0x13);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_15, 0x0001);
+	udelay(10);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_15, 0x0041);
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_RSTB, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ADC_FIFO, B_ADC_FIFO_RXK, 0x3333);
+}
+
+static void _iqk_init(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+	u8 idx, path;
+
+	rtw89_phy_write32_mask(rtwdev, R_IQKINF, MASKDWORD, 0x0);
+
+	if (iqk_info->is_iqk_init)
+		return;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]===>%s\n", __func__);
+	iqk_info->is_iqk_init = true;
+	iqk_info->is_nbiqk = false;
+	iqk_info->iqk_fft_en = false;
+	iqk_info->iqk_sram_en = false;
+	iqk_info->iqk_cfir_en = false;
+	iqk_info->iqk_xym_en = false;
+	iqk_info->iqk_times = 0x0;
+
+	for (idx = 0; idx < RTW89_IQK_CHS_NR; idx++) {
+		iqk_info->iqk_channel[idx] = 0x0;
+		for (path = 0; path < RTW8852BT_SS; path++) {
+			iqk_info->lok_cor_fail[idx][path] = false;
+			iqk_info->lok_fin_fail[idx][path] = false;
+			iqk_info->iqk_tx_fail[idx][path] = false;
+			iqk_info->iqk_rx_fail[idx][path] = false;
+			iqk_info->iqk_mcc_ch[idx][path] = 0x0;
+			iqk_info->iqk_table_idx[path] = 0x0;
+		}
+	}
+}
+
 static void _wait_rx_mode(struct rtw89_dev *rtwdev, u8 kpath)
 {
 	u32 rf_mode;
@@ -374,6 +1199,57 @@ static void _tmac_tx_pause(struct rtw89_dev *rtwdev, enum rtw89_phy_idx band_idx
 	_wait_rx_mode(rtwdev, _kpath(rtwdev, band_idx));
 }
 
+static void _doiqk(struct rtw89_dev *rtwdev, bool force,
+		   enum rtw89_phy_idx phy_idx, u8 path)
+{
+	struct rtw89_iqk_info *iqk_info = &rtwdev->iqk;
+	u32 backup_bb_val[BACKUP_BB_REGS_NR];
+	u32 backup_rf_val[RTW8852BT_SS][BACKUP_RF_REGS_NR];
+	u8 phy_map = rtw89_btc_phymap(rtwdev, phy_idx, RF_AB);
+
+	rtw89_btc_ntfy_wl_rfk(rtwdev, phy_map, BTC_WRFKT_IQK, BTC_WRFK_ONESHOT_START);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[IQK]==========IQK start!!!!!==========\n");
+	iqk_info->iqk_times++;
+	iqk_info->version = RTW8852BT_IQK_VER;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[IQK]Test Ver 0x%x\n", iqk_info->version);
+	_iqk_get_ch_info(rtwdev, phy_idx, path);
+
+	_rfk_backup_bb_reg(rtwdev, backup_bb_val);
+	_rfk_backup_rf_reg(rtwdev, backup_rf_val[path], path);
+	_iqk_macbb_setting(rtwdev, phy_idx, path);
+	_iqk_preset(rtwdev, path);
+	_iqk_start_iqk(rtwdev, phy_idx, path);
+	_iqk_restore(rtwdev, path);
+	_iqk_afebb_restore(rtwdev, phy_idx, path);
+	_rfk_reload_bb_reg(rtwdev, backup_bb_val);
+	_rfk_reload_rf_reg(rtwdev, backup_rf_val[path], path);
+
+	rtw89_btc_ntfy_wl_rfk(rtwdev, phy_map, BTC_WRFKT_IQK, BTC_WRFK_ONESHOT_STOP);
+}
+
+static void _iqk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx, bool force)
+{
+	u8 kpath = _kpath(rtwdev, phy_idx);
+
+	switch (kpath) {
+	case RF_A:
+		_doiqk(rtwdev, force, phy_idx, RF_PATH_A);
+		break;
+	case RF_B:
+		_doiqk(rtwdev, force, phy_idx, RF_PATH_B);
+		break;
+	case RF_AB:
+		_doiqk(rtwdev, force, phy_idx, RF_PATH_A);
+		_doiqk(rtwdev, force, phy_idx, RF_PATH_B);
+		break;
+	default:
+		break;
+	}
+}
+
 static void _dpk_onoff(struct rtw89_dev *rtwdev, enum rtw89_rf_path path, bool off)
 {
 	struct rtw89_dpk_info *dpk = &rtwdev->dpk;
@@ -2394,6 +3270,22 @@ void rtw8852bt_dpk_init(struct rtw89_dev *rtwdev)
 	_set_dpd_backoff(rtwdev, RTW89_PHY_0);
 }
 
+void rtw8852bt_iqk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
+{
+	u8 phy_map = rtw89_btc_phymap(rtwdev, phy_idx, 0);
+	u32 tx_en;
+
+	rtw89_btc_ntfy_wl_rfk(rtwdev, phy_map, BTC_WRFKT_IQK, BTC_WRFK_START);
+	rtw89_chip_stop_sch_tx(rtwdev, phy_idx, &tx_en, RTW89_SCH_TX_SEL_ALL);
+	_wait_rx_mode(rtwdev, _kpath(rtwdev, phy_idx));
+
+	_iqk_init(rtwdev);
+	_iqk(rtwdev, phy_idx, false);
+
+	rtw89_chip_resume_sch_tx(rtwdev, phy_idx, tx_en);
+	rtw89_btc_ntfy_wl_rfk(rtwdev, phy_map, BTC_WRFKT_IQK, BTC_WRFK_STOP);
+}
+
 void rtw8852bt_dpk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
 {
 	rtw89_debug(rtwdev, RTW89_DBG_RFK,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
index 89640b42d4d0..22c0c9c3d173 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
@@ -7,6 +7,7 @@
 
 #include "core.h"
 
+void rtw8852bt_iqk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx);
 void rtw8852bt_dpk_init(struct rtw89_dev *rtwdev);
 void rtw8852bt_dpk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy);
 void rtw8852bt_dpk_track(struct rtw89_dev *rtwdev);
-- 
cgit v1.2.3-58-ga151


From c3d9705de5e7442afe8ce121b86f99b4b7de4ac8 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 27 Jun 2024 10:58:47 +0800
Subject: wifi: rtw89: 8852bt: rfk: add RX DCK

RX DCK stands for receiver DC calibration. With this calibration, we have
proper DC offset to reflect correct received signal strength indicator.
Do this calibration when bringing up interface and going to connect.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240627025849.25198-3-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c | 57 ++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h |  1 +
 2 files changed, 58 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
index 4a638e2444db..52dfbe0b3d6b 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
@@ -13,6 +13,7 @@
 #include "rtw8852bt_rfk_table.h"
 #include "rtw8852b_common.h"
 
+#define RTW8852BT_RXDCK_VER 0x1
 #define RTW8852BT_IQK_VER 0x2a
 #define RTW8852BT_SS 2
 #define RTW8852BT_TSSI_PATH_NR 2
@@ -365,6 +366,47 @@ static void _rfk_bb_afe_restore(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy
 	rtw89_phy_write32_mask(rtwdev, R_P1_TSSI_MV_AVG, B_P1_TXPW_RSTB, 0x2);
 }
 
+static void _set_rx_dck(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy,
+			enum rtw89_rf_path path)
+{
+	rtw89_write_rf(rtwdev, path, RR_DCK1, RR_DCK1_CLR, 0x0);
+	rtw89_write_rf(rtwdev, path, RR_DCK, RR_DCK_LV, 0x0);
+	rtw89_write_rf(rtwdev, path, RR_DCK, RR_DCK_LV, 0x1);
+	mdelay(1);
+}
+
+static void _rx_dck(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
+{
+	u8 path, dck_tune;
+	u32 rf_reg5;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[RX_DCK] ****** RXDCK Start (Ver: 0x%x, CV : 0x%x) ******\n",
+		    RTW8852BT_RXDCK_VER, rtwdev->hal.cv);
+
+	for (path = 0; path < RF_PATH_NUM_8852BT; path++) {
+		rf_reg5 = rtw89_read_rf(rtwdev, path, RR_RSV1, RFREG_MASK);
+		dck_tune = rtw89_read_rf(rtwdev, path, RR_DCK, RR_DCK_FINE);
+
+		if (rtwdev->is_tssi_mode[path])
+			rtw89_phy_write32_mask(rtwdev,
+					       R_P0_TSSI_TRK + (path << 13),
+					       B_P0_TSSI_TRK_EN, 0x1);
+
+		rtw89_write_rf(rtwdev, path, RR_RSV1, RR_RSV1_RST, 0x0);
+		rtw89_write_rf(rtwdev, path, RR_DCK, RR_DCK_FINE, 0x0);
+		rtw89_write_rf(rtwdev, path, RR_MOD, RR_MOD_MASK, RR_MOD_V_RX);
+		_set_rx_dck(rtwdev, phy, path);
+		rtw89_write_rf(rtwdev, path, RR_DCK, RR_DCK_FINE, dck_tune);
+		rtw89_write_rf(rtwdev, path, RR_RSV1, RFREG_MASK, rf_reg5);
+
+		if (rtwdev->is_tssi_mode[path])
+			rtw89_phy_write32_mask(rtwdev,
+					       R_P0_TSSI_TRK + (path << 13),
+					       B_P0_TSSI_TRK_EN, 0x0);
+	}
+}
+
 static bool _iqk_check_cal(struct rtw89_dev *rtwdev, u8 path, u8 ktype)
 {
 	bool notready = false;
@@ -3286,6 +3328,21 @@ void rtw8852bt_iqk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
 	rtw89_btc_ntfy_wl_rfk(rtwdev, phy_map, BTC_WRFKT_IQK, BTC_WRFK_STOP);
 }
 
+void rtw8852bt_rx_dck(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
+{
+	u8 phy_map = rtw89_btc_phymap(rtwdev, phy_idx, 0);
+	u32 tx_en;
+
+	rtw89_btc_ntfy_wl_rfk(rtwdev, phy_map, BTC_WRFKT_RXDCK, BTC_WRFK_START);
+	rtw89_chip_stop_sch_tx(rtwdev, phy_idx, &tx_en, RTW89_SCH_TX_SEL_ALL);
+	_wait_rx_mode(rtwdev, _kpath(rtwdev, phy_idx));
+
+	_rx_dck(rtwdev, phy_idx);
+
+	rtw89_chip_resume_sch_tx(rtwdev, phy_idx, tx_en);
+	rtw89_btc_ntfy_wl_rfk(rtwdev, phy_map, BTC_WRFKT_RXDCK, BTC_WRFK_STOP);
+}
+
 void rtw8852bt_dpk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
 {
 	rtw89_debug(rtwdev, RTW89_DBG_RFK,
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
index 22c0c9c3d173..3ac63a2a63f7 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
@@ -8,6 +8,7 @@
 #include "core.h"
 
 void rtw8852bt_iqk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx);
+void rtw8852bt_rx_dck(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx);
 void rtw8852bt_dpk_init(struct rtw89_dev *rtwdev);
 void rtw8852bt_dpk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy);
 void rtw8852bt_dpk_track(struct rtw89_dev *rtwdev);
-- 
cgit v1.2.3-58-ga151


From 39f0123e032192f13c843344a83e3132bff28f3c Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 27 Jun 2024 10:58:48 +0800
Subject: wifi: rtw89: 8852bt: rfk: add DACK

DACK (digital-to-analog converters calibration) is used to calibrate DAC
to output signals as expected.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240627025849.25198-4-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/reg.h           |  15 +
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c | 487 +++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h |   1 +
 3 files changed, 503 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/reg.h b/drivers/net/wireless/realtek/rtw89/reg.h
index 24e8d244534e..7df36f3bff0b 100644
--- a/drivers/net/wireless/realtek/rtw89/reg.h
+++ b/drivers/net/wireless/realtek/rtw89/reg.h
@@ -7836,6 +7836,7 @@
 #define B_ANAPAR_PW15_H2 GENMASK(27, 26)
 #define R_ANAPAR 0x032C
 #define B_ANAPAR_15 GENMASK(31, 16)
+#define B_ANAPAR_EN1 BIT(31)
 #define B_ANAPAR_ADCCLK BIT(30)
 #define B_ANAPAR_FLTRST BIT(22)
 #define B_ANAPAR_CRXBB GENMASK(18, 16)
@@ -9080,11 +9081,13 @@
 #define B_DCOF0_RST BIT(17)
 #define B_DCOF0_V GENMASK(4, 1)
 #define R_DCOF1 0xC004
+#define B_DCOF1_VAL GENMASK(31, 20)
 #define B_DCOF1_RST BIT(17)
 #define B_DCOF1_S BIT(0)
 #define R_DCOF8 0xC020
 #define B_DCOF8_V GENMASK(4, 1)
 #define R_DCOF9 0xC024
+#define B_DCOF9_VAL GENMASK(31, 20)
 #define B_DCOF9_RST BIT(17)
 #define R_DACK_S0P0 0xC040
 #define B_DACK_S0P0_OK BIT(31)
@@ -9158,11 +9161,18 @@
 #define B_ADDCKR0_DC GENMASK(15, 4)
 #define B_ADDCKR0_A1 GENMASK(9, 0)
 #define R_DACK10 0xC100
+#define B_DACK10_RST BIT(17)
 #define B_DACK10 GENMASK(4, 1)
 #define R_DACK1_K 0xc104
+#define B_DACK1_VAL GENMASK(31, 20)
+#define B_DACK1_RST BIT(17)
 #define B_DACK1_EN BIT(0)
 #define R_DACK11 0xC120
 #define B_DACK11 GENMASK(4, 1)
+#define R_DACK2_K 0xC124
+#define B_DACK2_VAL GENMASK(31, 20)
+#define B_DACK2_RST BIT(17)
+#define B_DACK2_EN BIT(0)
 #define R_DACK_S1P0 0xC140
 #define B_DACK_S1P0_OK BIT(31)
 #define R_DACK_BIAS10 0xC148
@@ -9211,6 +9221,11 @@
 #define B_DACKN0_V GENMASK(21, 14)
 #define R_DACKN1_CTL 0xC224
 #define B_DACKN1_V GENMASK(21, 14)
+#define B_DACKN1_ON BIT(0)
+#define R_DACKN2_CTL 0xC238
+#define B_DACKN2_ON BIT(0)
+#define R_DACKN3_CTL 0xC24C
+#define B_DACKN3_ON BIT(0)
 #define R_GAIN_MAP0 0xE44C
 #define B_GAIN_MAP0_EN BIT(0)
 #define R_GAIN_MAP1 0xE54C
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
index 52dfbe0b3d6b..6c2fecb58940 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
@@ -407,6 +407,484 @@ static void _rx_dck(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
 	}
 }
 
+static void _drck(struct rtw89_dev *rtwdev)
+{
+	u32 rck_d;
+	u32 val;
+	int ret;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]Ddie RCK start!!!\n");
+	rtw89_phy_write32_mask(rtwdev, R_DRCK, B_DRCK_EN, 0x1);
+
+	ret = read_poll_timeout_atomic(rtw89_phy_read32_mask, val, val,
+				       1, 10000, false,
+				       rtwdev, R_DRCK_RES, B_DRCK_POL);
+	if (ret)
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]DRCK timeout\n");
+
+	rtw89_phy_write32_mask(rtwdev, R_DRCK, B_DRCK_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_DRCK_FH, B_DRCK_LAT, 0x1);
+	udelay(1);
+	rtw89_phy_write32_mask(rtwdev, R_DRCK_FH, B_DRCK_LAT, 0x0);
+
+	rck_d = rtw89_phy_read32_mask(rtwdev, R_DRCK_RES, 0x7c00);
+	rtw89_phy_write32_mask(rtwdev, R_DRCK, B_DRCK_IDLE, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_DRCK, B_DRCK_VAL, rck_d);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]0xc0c4 = 0x%x\n",
+		    rtw89_phy_read32_mask(rtwdev, R_DRCK, MASKDWORD));
+}
+
+static void _dack_backup_s0(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_dack_info *dack = &rtwdev->dack;
+	u8 i;
+
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_DBG, 0x1);
+
+	for (i = 0; i < 0x10; i++) {
+		rtw89_phy_write32_mask(rtwdev, R_DCOF0, B_DCOF0_V, i);
+		dack->msbk_d[0][0][i] =
+			rtw89_phy_read32_mask(rtwdev, R_DACK_S0P2, B_DACK_S0M0);
+
+		rtw89_phy_write32_mask(rtwdev, R_DCOF8, B_DCOF8_V, i);
+		dack->msbk_d[0][1][i] =
+			rtw89_phy_read32_mask(rtwdev, R_DACK_S0P3, B_DACK_S0M1);
+	}
+
+	dack->biask_d[0][0] =
+		rtw89_phy_read32_mask(rtwdev, R_DACK_BIAS00, B_DACK_BIAS00);
+	dack->biask_d[0][1] =
+		rtw89_phy_read32_mask(rtwdev, R_DACK_BIAS01, B_DACK_BIAS01);
+
+	dack->dadck_d[0][0] =
+		rtw89_phy_read32_mask(rtwdev, R_DACK_DADCK00, B_DACK_DADCK00);
+	dack->dadck_d[0][1] =
+		rtw89_phy_read32_mask(rtwdev, R_DACK_DADCK01, B_DACK_DADCK01);
+}
+
+static void _dack_backup_s1(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_dack_info *dack = &rtwdev->dack;
+	u8 i;
+
+	rtw89_phy_write32_mask(rtwdev, R_P1_DBGMOD, B_P1_DBGMOD_ON, 0x1);
+
+	for (i = 0; i < 0x10; i++) {
+		rtw89_phy_write32_mask(rtwdev, R_DACK10, B_DACK10, i);
+		dack->msbk_d[1][0][i] =
+			rtw89_phy_read32_mask(rtwdev, R_DACK10S, B_DACK10S);
+
+		rtw89_phy_write32_mask(rtwdev, R_DACK11, B_DACK11, i);
+		dack->msbk_d[1][1][i] =
+			rtw89_phy_read32_mask(rtwdev, R_DACK11S, B_DACK11S);
+	}
+
+	dack->biask_d[1][0] =
+		rtw89_phy_read32_mask(rtwdev, R_DACK_BIAS10, B_DACK_BIAS10);
+	dack->biask_d[1][1] =
+		rtw89_phy_read32_mask(rtwdev, R_DACK_BIAS11, B_DACK_BIAS11);
+
+	dack->dadck_d[1][0] =
+		rtw89_phy_read32_mask(rtwdev, R_DACK_DADCK10, B_DACK_DADCK10);
+	dack->dadck_d[1][1] =
+		rtw89_phy_read32_mask(rtwdev, R_DACK_DADCK11, B_DACK_DADCK11);
+}
+
+static
+void _dack_reset(struct rtw89_dev *rtwdev, enum rtw89_rf_path path)
+{
+	if (path == RF_PATH_A) {
+		rtw89_phy_write32_mask(rtwdev, R_DCOF0, B_DCOF0_RST, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_DCOF0, B_DCOF0_RST, 0x1);
+	} else {
+		rtw89_phy_write32_mask(rtwdev, R_DACK10, B_DACK10_RST, 0x0);
+		rtw89_phy_write32_mask(rtwdev, R_DACK10, B_DACK10_RST, 0x1);
+	}
+}
+
+static
+void _dack_reload_by_path(struct rtw89_dev *rtwdev, u8 path, u8 index)
+{
+	struct rtw89_dack_info *dack = &rtwdev->dack;
+	u32 tmp, tmp_offset, tmp_reg;
+	u32 idx_offset, path_offset;
+	u8 i;
+
+	if (index == 0)
+		idx_offset = 0;
+	else
+		idx_offset = 0x14;
+
+	if (path == RF_PATH_A)
+		path_offset = 0;
+	else
+		path_offset = 0x28;
+
+	tmp_offset = idx_offset + path_offset;
+
+	rtw89_phy_write32_mask(rtwdev, R_DCOF1, B_DCOF1_RST, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_DCOF9, B_DCOF9_RST, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_DACK1_K, B_DACK1_RST, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_DACK2_K, B_DACK2_RST, 0x1);
+
+	/* msbk_d: 15/14/13/12 */
+	tmp = 0x0;
+	for (i = 0; i < 4; i++)
+		tmp |= dack->msbk_d[path][index][i + 12] << (i * 8);
+	tmp_reg = 0xc200 + tmp_offset;
+	rtw89_phy_write32(rtwdev, tmp_reg, tmp);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]0x%x=0x%x\n", tmp_reg,
+		    rtw89_phy_read32_mask(rtwdev, tmp_reg, MASKDWORD));
+
+	/* msbk_d: 11/10/9/8 */
+	tmp = 0x0;
+	for (i = 0; i < 4; i++)
+		tmp |= dack->msbk_d[path][index][i + 8] << (i * 8);
+	tmp_reg = 0xc204 + tmp_offset;
+	rtw89_phy_write32(rtwdev, tmp_reg, tmp);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]0x%x=0x%x\n", tmp_reg,
+		    rtw89_phy_read32_mask(rtwdev, tmp_reg, MASKDWORD));
+
+	/* msbk_d: 7/6/5/4 */
+	tmp = 0x0;
+	for (i = 0; i < 4; i++)
+		tmp |= dack->msbk_d[path][index][i + 4] << (i * 8);
+	tmp_reg = 0xc208 + tmp_offset;
+	rtw89_phy_write32(rtwdev, tmp_reg, tmp);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]0x%x=0x%x\n", tmp_reg,
+		    rtw89_phy_read32_mask(rtwdev, tmp_reg, MASKDWORD));
+
+	/* msbk_d: 3/2/1/0 */
+	tmp = 0x0;
+	for (i = 0; i < 4; i++)
+		tmp |= dack->msbk_d[path][index][i] << (i * 8);
+	tmp_reg = 0xc20c + tmp_offset;
+	rtw89_phy_write32(rtwdev, tmp_reg, tmp);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]0x%x=0x%x\n", tmp_reg,
+		    rtw89_phy_read32_mask(rtwdev, tmp_reg, MASKDWORD));
+
+	/* dadak_d/biask_d */
+	tmp = (dack->biask_d[path][index] << 22) |
+	      (dack->dadck_d[path][index] << 14);
+	tmp_reg = 0xc210 + tmp_offset;
+	rtw89_phy_write32(rtwdev, tmp_reg, tmp);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]0x%x=0x%x\n", tmp_reg,
+		    rtw89_phy_read32_mask(rtwdev, tmp_reg, MASKDWORD));
+
+	/* enable DACK result from reg */
+	rtw89_phy_write32_mask(rtwdev, R_DACKN0_CTL + tmp_offset, B_DACKN0_EN, 0x1);
+}
+
+static
+void _dack_reload(struct rtw89_dev *rtwdev, enum rtw89_rf_path path)
+{
+	u8 i;
+
+	for (i = 0; i < 2; i++)
+		_dack_reload_by_path(rtwdev, path, i);
+}
+
+static bool _dack_s0_poll(struct rtw89_dev *rtwdev)
+{
+	if (rtw89_phy_read32_mask(rtwdev, R_DACK_S0P0, B_DACK_S0P0_OK) == 0 ||
+	    rtw89_phy_read32_mask(rtwdev, R_DACK_S0P1, B_DACK_S0P1_OK) == 0 ||
+	    rtw89_phy_read32_mask(rtwdev, R_DACK_S0P2, B_DACK_S0P2_OK) == 0 ||
+	    rtw89_phy_read32_mask(rtwdev, R_DACK_S0P3, B_DACK_S0P3_OK) == 0)
+		return false;
+
+	return true;
+}
+
+static void _dack_s0(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_dack_info *dack = &rtwdev->dack;
+	bool done;
+	int ret;
+
+	_txck_force(rtwdev, RF_PATH_A, true, DAC_160M);
+
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_DBG, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, BIT(28), 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_EN1, 0x0);
+	udelay(100);
+	rtw89_phy_write32_mask(rtwdev, R_DCOF1, B_DCOF1_VAL, 0x30);
+	rtw89_phy_write32_mask(rtwdev, R_DCOF9, B_DCOF9_VAL, 0x30);
+
+	_dack_reset(rtwdev, RF_PATH_A);
+
+	rtw89_phy_write32_mask(rtwdev, R_DCOF1, B_DCOF1_S, 0x1);
+	udelay(1);
+
+	dack->msbk_timeout[0] = false;
+
+	ret = read_poll_timeout_atomic(_dack_s0_poll, done, done,
+				       1, 20000, false, rtwdev);
+	if (ret) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]S0 DACK timeout\n");
+		dack->msbk_timeout[0] = true;
+	}
+
+	rtw89_phy_write32_mask(rtwdev, R_DCOF1, B_DCOF1_S, 0x0);
+
+	_txck_force(rtwdev, RF_PATH_A, false, DAC_960M);
+	_dack_backup_s0(rtwdev);
+	_dack_reload(rtwdev, RF_PATH_A);
+
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_DBG, 0x0);
+}
+
+static bool _dack_s1_poll(struct rtw89_dev *rtwdev)
+{
+	if (rtw89_phy_read32_mask(rtwdev, R_DACK_S1P0, B_DACK_S1P0_OK) == 0 ||
+	    rtw89_phy_read32_mask(rtwdev, R_DACK_S1P1, B_DACK_S1P1_OK) == 0 ||
+	    rtw89_phy_read32_mask(rtwdev, R_DACK_S1P2, B_DACK_S1P2_OK) == 0 ||
+	    rtw89_phy_read32_mask(rtwdev, R_DACK_S1P3, B_DACK_S1P3_OK) == 0)
+		return false;
+
+	return true;
+}
+
+static void _dack_s1(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_dack_info *dack = &rtwdev->dack;
+	bool done;
+	int ret;
+
+	_txck_force(rtwdev, RF_PATH_B, true, DAC_160M);
+
+	rtw89_phy_write32_mask(rtwdev, R_P1_DBGMOD, B_P1_DBGMOD_ON, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, BIT(28), 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_EN1, 0x0);
+	udelay(100);
+	rtw89_phy_write32_mask(rtwdev, R_DACK1_K, B_DACK1_VAL, 0x30);
+	rtw89_phy_write32_mask(rtwdev, R_DACK2_K, B_DACK2_VAL, 0x30);
+
+	_dack_reset(rtwdev, RF_PATH_B);
+
+	rtw89_phy_write32_mask(rtwdev, R_DACK1_K, B_DACK1_EN, 0x1);
+	udelay(1);
+
+	dack->msbk_timeout[1] = false;
+
+	ret = read_poll_timeout_atomic(_dack_s1_poll, done, done,
+				       1, 10000, false, rtwdev);
+	if (ret) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]S1 DACK timeout\n");
+		dack->msbk_timeout[1] = true;
+	}
+
+	rtw89_phy_write32_mask(rtwdev, R_DACK1_K, B_DACK1_EN, 0x0);
+
+	_txck_force(rtwdev, RF_PATH_B, false, DAC_960M);
+	_dack_backup_s1(rtwdev);
+	_dack_reload(rtwdev, RF_PATH_B);
+
+	rtw89_phy_write32_mask(rtwdev, R_P1_DBGMOD, B_P1_DBGMOD_ON, 0x0);
+}
+
+static void _dack(struct rtw89_dev *rtwdev)
+{
+	_dack_s0(rtwdev);
+	_dack_s1(rtwdev);
+}
+
+static void _dack_dump(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_dack_info *dack = &rtwdev->dack;
+	u8 i;
+	u8 t;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DACK]S0 ADC_DCK ic = 0x%x, qc = 0x%x\n",
+		    dack->addck_d[0][0], dack->addck_d[0][1]);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DACK]S1 ADC_DCK ic = 0x%x, qc = 0x%x\n",
+		    dack->addck_d[1][0], dack->addck_d[1][1]);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DACK]S0 DAC_DCK ic = 0x%x, qc = 0x%x\n",
+		    dack->dadck_d[0][0], dack->dadck_d[0][1]);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DACK]S1 DAC_DCK ic = 0x%x, qc = 0x%x\n",
+		    dack->dadck_d[1][0], dack->dadck_d[1][1]);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DACK]S0 biask ic = 0x%x, qc = 0x%x\n",
+		    dack->biask_d[0][0], dack->biask_d[0][1]);
+	rtw89_debug(rtwdev, RTW89_DBG_RFK,
+		    "[DACK]S1 biask ic = 0x%x, qc = 0x%x\n",
+		    dack->biask_d[1][0], dack->biask_d[1][1]);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]S0 MSBK ic:\n");
+	for (i = 0; i < 0x10; i++) {
+		t = dack->msbk_d[0][0][i];
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]0x%x\n", t);
+	}
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]S0 MSBK qc:\n");
+	for (i = 0; i < RTW89_DACK_MSBK_NR; i++) {
+		t = dack->msbk_d[0][1][i];
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]0x%x\n", t);
+	}
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]S1 MSBK ic:\n");
+	for (i = 0; i < RTW89_DACK_MSBK_NR; i++) {
+		t = dack->msbk_d[1][0][i];
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]0x%x\n", t);
+	}
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]S1 MSBK qc:\n");
+	for (i = 0; i < RTW89_DACK_MSBK_NR; i++) {
+		t = dack->msbk_d[1][1][i];
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]0x%x\n", t);
+	}
+}
+
+static void _addck_ori(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_dack_info *dack = &rtwdev->dack;
+	u32 val;
+	int ret;
+
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK0, B_ADDCK0_MAN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK1, B_ADDCK1_MAN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_DBG, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_ADCCLK, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_FLTRST, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_FLTRST, 0x1);
+
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, B_ANAPAR_PW15_H, 0xf);
+	udelay(100);
+
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_PATH0_SAMPL_DLY_T_V1, BIT(4), 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, B_ANAPAR_PW15_H, 0x3);
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK0, B_ADDCK0_TRG, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK0, B_ADDCK0_TRG, 0x0);
+	udelay(1);
+
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK0, B_ADDCK0, 0x1);
+	dack->addck_timeout[0] = false;
+
+	ret = read_poll_timeout_atomic(rtw89_phy_read32_mask, val, val,
+				       1, 10000, false,
+				       rtwdev, R_ADDCKR0, BIT(0));
+	if (ret) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]S0 ADDCK timeout\n");
+		dack->addck_timeout[0] = true;
+	}
+
+	rtw89_phy_write32_mask(rtwdev, R_PATH0_SAMPL_DLY_T_V1, BIT(4), 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, B_ANAPAR_PW15_H, 0xc);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_ADCCLK, 0x1);
+
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK0, B_ADDCK0, 0x0);
+	dack->addck_d[0][0] =
+		rtw89_phy_read32_mask(rtwdev, R_ADDCKR0, B_ADDCKR0_A0);
+	dack->addck_d[0][1] =
+		rtw89_phy_read32_mask(rtwdev, R_ADDCKR0, B_ADDCKR0_A1);
+	rtw89_phy_write32_mask(rtwdev, R_P0_NRBW, B_P0_NRBW_DBG, 0x0);
+
+	rtw89_phy_write32_mask(rtwdev, R_P1_DBGMOD, B_P1_DBGMOD_ON, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_ADCCLK, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_FLTRST, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_FLTRST, 0x1);
+
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, B_ANAPAR_PW15_H, 0xf);
+	udelay(100);
+
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_PATH1_SAMPL_DLY_T_V1, BIT(4), 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, B_ANAPAR_PW15_H, 0x3);
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK1, B_ADDCK1_TRG, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK1, B_ADDCK1_TRG, 0x0);
+	udelay(1);
+
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK1, B_ADDCK1, 0x1);
+	dack->addck_timeout[1] = false;
+
+	ret = read_poll_timeout_atomic(rtw89_phy_read32_mask, val, val,
+				       1, 10000, false,
+				       rtwdev, R_ADDCKR1, BIT(0));
+	if (ret) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]S1 ADDCK timeout\n");
+		dack->addck_timeout[1] = true;
+	}
+
+	rtw89_phy_write32_mask(rtwdev, R_PATH1_SAMPL_DLY_T_V1, BIT(4), 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_EN, 0x1);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR_PW15, B_ANAPAR_PW15_H, 0xc);
+	rtw89_phy_write32_mask(rtwdev, R_ANAPAR, B_ANAPAR_ADCCLK, 0x1);
+
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK1, B_ADDCK1, 0x0);
+	dack->addck_d[1][0] =
+		rtw89_phy_read32_mask(rtwdev, R_ADDCKR1, B_ADDCKR1_A0);
+	dack->addck_d[1][1] =
+		rtw89_phy_read32_mask(rtwdev, R_ADDCKR1, B_ADDCKR1_A1);
+
+	rtw89_phy_write32_mask(rtwdev, R_P1_DBGMOD, B_P1_DBGMOD_ON, 0x0);
+}
+
+static void _addck_reload(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_dack_info *dack = &rtwdev->dack;
+
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK0_RL, B_ADDCK0_RL1, dack->addck_d[0][0]);
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK0_RL, B_ADDCK0_RL0, dack->addck_d[0][1]);
+
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK0_RL, B_ADDCK0_RLS, 0x3);
+
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK1_RL, B_ADDCK1_RL1, dack->addck_d[1][0]);
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK1_RL, B_ADDCK1_RL0, dack->addck_d[1][1]);
+
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK1_RL, B_ADDCK1_RLS, 0x3);
+}
+
+static void _dack_manual_off(struct rtw89_dev *rtwdev)
+{
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK0_RL, B_ADDCK0_RLS, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_ADDCK1_RL, B_ADDCK1_RLS, 0x0);
+
+	rtw89_phy_write32_mask(rtwdev, R_DACKN0_CTL, B_DACKN0_EN, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_DACKN1_CTL, B_DACKN1_ON, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_DACKN2_CTL, B_DACKN2_ON, 0x0);
+	rtw89_phy_write32_mask(rtwdev, R_DACKN3_CTL, B_DACKN3_ON, 0x0);
+}
+
+static void _dac_cal(struct rtw89_dev *rtwdev, bool force)
+{
+	struct rtw89_dack_info *dack = &rtwdev->dack;
+
+	dack->dack_done = false;
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]DACK start!!!\n");
+
+	_drck(rtwdev);
+	_dack_manual_off(rtwdev);
+	rtw89_write_rf(rtwdev, RF_PATH_A, RR_RSV1, RFREG_MASK, 0x0);
+	rtw89_write_rf(rtwdev, RF_PATH_B, RR_RSV1, RFREG_MASK, 0x0);
+	rtw89_write_rf(rtwdev, RF_PATH_A, RR_MOD, RFREG_MASK, 0x337e1);
+	rtw89_write_rf(rtwdev, RF_PATH_B, RR_MOD, RFREG_MASK, 0x337e1);
+	_rxck_force(rtwdev, RF_PATH_A, true, ADC_960M);
+	_rxck_force(rtwdev, RF_PATH_B, true, ADC_960M);
+	_addck_ori(rtwdev);
+
+	_rxck_force(rtwdev, RF_PATH_A, false, ADC_960M);
+	_rxck_force(rtwdev, RF_PATH_B, false, ADC_960M);
+	_addck_reload(rtwdev);
+
+	rtw89_write_rf(rtwdev, RF_PATH_A, RR_MODOPT, RFREG_MASK, 0x0);
+	rtw89_write_rf(rtwdev, RF_PATH_B, RR_MODOPT, RFREG_MASK, 0x0);
+
+	_dack(rtwdev);
+	_dack_dump(rtwdev);
+	dack->dack_done = true;
+	rtw89_write_rf(rtwdev, RF_PATH_A, RR_RSV1, RFREG_MASK, 0x1);
+	rtw89_write_rf(rtwdev, RF_PATH_B, RR_RSV1, RFREG_MASK, 0x1);
+
+	dack->dack_cnt++;
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[DACK]DACK finish!!!\n");
+}
+
 static bool _iqk_check_cal(struct rtw89_dev *rtwdev, u8 path, u8 ktype)
 {
 	bool notready = false;
@@ -3312,6 +3790,15 @@ void rtw8852bt_dpk_init(struct rtw89_dev *rtwdev)
 	_set_dpd_backoff(rtwdev, RTW89_PHY_0);
 }
 
+void rtw8852bt_dack(struct rtw89_dev *rtwdev)
+{
+	u8 phy_map = rtw89_btc_phymap(rtwdev, RTW89_PHY_0, 0);
+
+	rtw89_btc_ntfy_wl_rfk(rtwdev, phy_map, BTC_WRFKT_DACK, BTC_WRFK_START);
+	_dac_cal(rtwdev, false);
+	rtw89_btc_ntfy_wl_rfk(rtwdev, phy_map, BTC_WRFKT_DACK, BTC_WRFK_STOP);
+}
+
 void rtw8852bt_iqk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
 {
 	u8 phy_map = rtw89_btc_phymap(rtwdev, phy_idx, 0);
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
index 3ac63a2a63f7..772a5b099377 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
@@ -7,6 +7,7 @@
 
 #include "core.h"
 
+void rtw8852bt_dack(struct rtw89_dev *rtwdev);
 void rtw8852bt_iqk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx);
 void rtw8852bt_rx_dck(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx);
 void rtw8852bt_dpk_init(struct rtw89_dev *rtwdev);
-- 
cgit v1.2.3-58-ga151


From 746f4ae52a38459dbed792adb58cadaed370dcc2 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Thu, 27 Jun 2024 10:58:49 +0800
Subject: wifi: rtw89: 8852bt: rfk: add RCK

RCK is synchronize RC calibration. Driver triggers this calibration and
writes the result to registers.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240627025849.25198-5-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c | 43 ++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h |  1 +
 2 files changed, 44 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
index 6c2fecb58940..fa0e49d58112 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.c
@@ -407,6 +407,41 @@ static void _rx_dck(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy)
 	}
 }
 
+static void _rck(struct rtw89_dev *rtwdev, enum rtw89_rf_path path)
+{
+	u32 rf_reg5;
+	u32 rck_val;
+	u32 val;
+	int ret;
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[RCK] ====== S%d RCK ======\n", path);
+
+	rf_reg5 = rtw89_read_rf(rtwdev, path, RR_RSV1, RFREG_MASK);
+
+	rtw89_write_rf(rtwdev, path, RR_RSV1, RR_RSV1_RST, 0x0);
+	rtw89_write_rf(rtwdev, path, RR_MOD, RR_MOD_MASK, RR_MOD_V_RX);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[RCK] RF0x00 = 0x%05x\n",
+		    rtw89_read_rf(rtwdev, path, RR_MOD, RFREG_MASK));
+
+	/* RCK trigger */
+	rtw89_write_rf(rtwdev, path, RR_RCKC, RFREG_MASK, 0x00240);
+
+	ret = read_poll_timeout_atomic(rtw89_read_rf, val, val, 2, 30,
+				       false, rtwdev, path, RR_RCKS, BIT(3));
+
+	rck_val = rtw89_read_rf(rtwdev, path, RR_RCKC, RR_RCKC_CA);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[RCK] rck_val = 0x%x, ret = %d\n",
+		    rck_val, ret);
+
+	rtw89_write_rf(rtwdev, path, RR_RCKC, RFREG_MASK, rck_val);
+	rtw89_write_rf(rtwdev, path, RR_RSV1, RFREG_MASK, rf_reg5);
+
+	rtw89_debug(rtwdev, RTW89_DBG_RFK, "[RCK] RF 0x1b = 0x%x\n",
+		    rtw89_read_rf(rtwdev, path, RR_RCKC, RFREG_MASK));
+}
+
 static void _drck(struct rtw89_dev *rtwdev)
 {
 	u32 rck_d;
@@ -3790,6 +3825,14 @@ void rtw8852bt_dpk_init(struct rtw89_dev *rtwdev)
 	_set_dpd_backoff(rtwdev, RTW89_PHY_0);
 }
 
+void rtw8852bt_rck(struct rtw89_dev *rtwdev)
+{
+	u8 path;
+
+	for (path = 0; path < RF_PATH_NUM_8852BT; path++)
+		_rck(rtwdev, path);
+}
+
 void rtw8852bt_dack(struct rtw89_dev *rtwdev)
 {
 	u8 phy_map = rtw89_btc_phymap(rtwdev, RTW89_PHY_0, 0);
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
index 772a5b099377..09918835c6e8 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852bt_rfk.h
@@ -7,6 +7,7 @@
 
 #include "core.h"
 
+void rtw8852bt_rck(struct rtw89_dev *rtwdev);
 void rtw8852bt_dack(struct rtw89_dev *rtwdev);
 void rtw8852bt_iqk(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx);
 void rtw8852bt_rx_dck(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx);
-- 
cgit v1.2.3-58-ga151


From d7f39aee79f04eeaa42085728423501b33ac5be5 Mon Sep 17 00:00:00 2001
From: David Wei <dw@davidwei.uk>
Date: Wed, 26 Jun 2024 20:01:59 -0700
Subject: page_pool: export page_pool_disable_direct_recycling()

56ef27e3 unexported page_pool_unlink_napi() and renamed it to
page_pool_disable_direct_recycling(). This is because there was no
in-tree user of page_pool_unlink_napi().

Since then Rx queue API and an implementation in bnxt got merged. In the
bnxt implementation, it broadly follows the following steps: allocate
new queue memory + page pool, stop old rx queue, swap, then destroy old
queue memory + page pool.

The existing NAPI instance is re-used so when the old page pool that is
no longer used but still linked to this shared NAPI instance is
destroyed, it will trigger warnings.

In my initial patches I unlinked a page pool from a NAPI instance
directly. Instead, export page_pool_disable_direct_recycling() and call
that instead to avoid having a driver touch a core struct.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David Wei <dw@davidwei.uk>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/page_pool/types.h | 1 +
 net/core/page_pool.c          | 3 ++-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/include/net/page_pool/types.h b/include/net/page_pool/types.h
index 7e8477057f3d..9093a964fc33 100644
--- a/include/net/page_pool/types.h
+++ b/include/net/page_pool/types.h
@@ -229,6 +229,7 @@ struct page_pool *page_pool_create_percpu(const struct page_pool_params *params,
 struct xdp_mem_info;
 
 #ifdef CONFIG_PAGE_POOL
+void page_pool_disable_direct_recycling(struct page_pool *pool);
 void page_pool_destroy(struct page_pool *pool);
 void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *),
 			   const struct xdp_mem_info *mem);
diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index 3927a0a7fa9a..5f1ed6f2ca8f 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -1014,7 +1014,7 @@ void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *),
 	pool->xdp_mem_id = mem->id;
 }
 
-static void page_pool_disable_direct_recycling(struct page_pool *pool)
+void page_pool_disable_direct_recycling(struct page_pool *pool)
 {
 	/* Disable direct recycling based on pool->cpuid.
 	 * Paired with READ_ONCE() in page_pool_napi_local().
@@ -1032,6 +1032,7 @@ static void page_pool_disable_direct_recycling(struct page_pool *pool)
 
 	WRITE_ONCE(pool->p.napi, NULL);
 }
+EXPORT_SYMBOL(page_pool_disable_direct_recycling);
 
 void page_pool_destroy(struct page_pool *pool)
 {
-- 
cgit v1.2.3-58-ga151


From 40eca00ae605d77b6d784824a6ce54c5b42dfce6 Mon Sep 17 00:00:00 2001
From: David Wei <dw@davidwei.uk>
Date: Wed, 26 Jun 2024 20:02:00 -0700
Subject: bnxt_en: unlink page pool when stopping Rx queue

Have bnxt call page_pool_disable_direct_recycling() to unlink the old
page pool when resetting a queue prior to destroying it, instead of
touching a netdev core struct directly.

Signed-off-by: David Wei <dw@davidwei.uk>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index a1690207d793..6fc34ccb86e3 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -15081,11 +15081,6 @@ static void bnxt_queue_mem_free(struct net_device *dev, void *qmem)
 	bnxt_free_one_rx_ring(bp, rxr);
 	bnxt_free_one_rx_agg_ring(bp, rxr);
 
-	/* At this point, this NAPI instance has another page pool associated
-	 * with it. Disconnect here before freeing the old page pool to avoid
-	 * warnings.
-	 */
-	rxr->page_pool->p.napi = NULL;
 	page_pool_destroy(rxr->page_pool);
 	rxr->page_pool = NULL;
 
@@ -15205,6 +15200,7 @@ static int bnxt_queue_stop(struct net_device *dev, void *qmem, int idx)
 	bnxt_hwrm_rx_ring_free(bp, rxr, false);
 	bnxt_hwrm_rx_agg_ring_free(bp, rxr, false);
 	rxr->rx_next_cons = 0;
+	page_pool_disable_direct_recycling(rxr->page_pool);
 
 	memcpy(qmem, rxr, sizeof(*rxr));
 	bnxt_init_rx_ring_struct(bp, qmem);
-- 
cgit v1.2.3-58-ga151


From d80d61ab0609f7f7168d59ec82ee5f055a4b6be7 Mon Sep 17 00:00:00 2001
From: Tushar Vyavahare <tushar.vyavahare@intel.com>
Date: Tue, 2 Jul 2024 05:59:15 +0000
Subject: selftests/xsk: Ensure traffic validation proceeds after ring size
 adjustment in xskxceiver

Previously, HW_SW_MIN_RING_SIZE and HW_SW_MAX_RING_SIZE test cases were
not validating Tx/Rx traffic at all due to early return after changing HW
ring size in testapp_validate_traffic().

Fix the flow by checking return value of set_ring_size() and act upon it
rather than terminating the test case there.

Signed-off-by: Tushar Vyavahare <tushar.vyavahare@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/bpf/20240702055916.48071-2-tushar.vyavahare@intel.com
---
 tools/testing/selftests/bpf/xskxceiver.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/bpf/xskxceiver.c b/tools/testing/selftests/bpf/xskxceiver.c
index 2eac0895b0a1..088df53869e8 100644
--- a/tools/testing/selftests/bpf/xskxceiver.c
+++ b/tools/testing/selftests/bpf/xskxceiver.c
@@ -1899,11 +1899,15 @@ static int testapp_validate_traffic(struct test_spec *test)
 	}
 
 	if (test->set_ring) {
-		if (ifobj_tx->hw_ring_size_supp)
-			return set_ring_size(ifobj_tx);
-
-	ksft_test_result_skip("Changing HW ring size not supported.\n");
-	return TEST_SKIP;
+		if (ifobj_tx->hw_ring_size_supp) {
+			if (set_ring_size(ifobj_tx)) {
+				ksft_test_result_skip("Failed to change HW ring size.\n");
+				return TEST_FAILURE;
+			}
+		} else {
+			ksft_test_result_skip("Changing HW ring size not supported.\n");
+			return TEST_SKIP;
+		}
 	}
 
 	xsk_attach_xdp_progs(test, ifobj_rx, ifobj_tx);
-- 
cgit v1.2.3-58-ga151


From e4a195e2b95e4602c667ed19a20f71218df138c2 Mon Sep 17 00:00:00 2001
From: Tushar Vyavahare <tushar.vyavahare@intel.com>
Date: Tue, 2 Jul 2024 05:59:16 +0000
Subject: selftests/xsk: Enhance batch size support with dynamic configurations

Introduce dynamic adjustment capabilities for fill_size and comp_size
parameters to support larger batch sizes beyond the previous 2K limit.

Update HW_SW_MAX_RING_SIZE test cases to evaluate AF_XDP's robustness by
pushing hardware and software ring sizes to their limits. This test
ensures AF_XDP's reliability amidst potential producer/consumer throttling
due to maximum ring utilization.

Signed-off-by: Tushar Vyavahare <tushar.vyavahare@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/bpf/20240702055916.48071-3-tushar.vyavahare@intel.com
---
 tools/testing/selftests/bpf/xskxceiver.c | 26 ++++++++++++++++++++------
 tools/testing/selftests/bpf/xskxceiver.h |  2 ++
 2 files changed, 22 insertions(+), 6 deletions(-)

diff --git a/tools/testing/selftests/bpf/xskxceiver.c b/tools/testing/selftests/bpf/xskxceiver.c
index 088df53869e8..8144fd145237 100644
--- a/tools/testing/selftests/bpf/xskxceiver.c
+++ b/tools/testing/selftests/bpf/xskxceiver.c
@@ -196,6 +196,12 @@ static int xsk_configure_umem(struct ifobject *ifobj, struct xsk_umem_info *umem
 	};
 	int ret;
 
+	if (umem->fill_size)
+		cfg.fill_size = umem->fill_size;
+
+	if (umem->comp_size)
+		cfg.comp_size = umem->comp_size;
+
 	if (umem->unaligned_mode)
 		cfg.flags |= XDP_UMEM_UNALIGNED_CHUNK_FLAG;
 
@@ -265,6 +271,10 @@ static int __xsk_configure_socket(struct xsk_socket_info *xsk, struct xsk_umem_i
 		cfg.bind_flags |= XDP_SHARED_UMEM;
 	if (ifobject->mtu > MAX_ETH_PKT_SIZE)
 		cfg.bind_flags |= XDP_USE_SG;
+	if (umem->comp_size)
+		cfg.tx_size = umem->comp_size;
+	if (umem->fill_size)
+		cfg.rx_size = umem->fill_size;
 
 	txr = ifobject->tx_on ? &xsk->tx : NULL;
 	rxr = ifobject->rx_on ? &xsk->rx : NULL;
@@ -1616,7 +1626,7 @@ static void xsk_populate_fill_ring(struct xsk_umem_info *umem, struct pkt_stream
 	if (umem->num_frames < XSK_RING_PROD__DEFAULT_NUM_DESCS)
 		buffers_to_fill = umem->num_frames;
 	else
-		buffers_to_fill = XSK_RING_PROD__DEFAULT_NUM_DESCS;
+		buffers_to_fill = umem->fill_size;
 
 	ret = xsk_ring_prod__reserve(&umem->fq, buffers_to_fill, &idx);
 	if (ret != buffers_to_fill)
@@ -2445,7 +2455,7 @@ static int testapp_hw_sw_min_ring_size(struct test_spec *test)
 
 static int testapp_hw_sw_max_ring_size(struct test_spec *test)
 {
-	u32 max_descs = XSK_RING_PROD__DEFAULT_NUM_DESCS * 2;
+	u32 max_descs = XSK_RING_PROD__DEFAULT_NUM_DESCS * 4;
 	int ret;
 
 	test->set_ring = true;
@@ -2453,7 +2463,8 @@ static int testapp_hw_sw_max_ring_size(struct test_spec *test)
 	test->ifobj_tx->ring.tx_pending = test->ifobj_tx->ring.tx_max_pending;
 	test->ifobj_tx->ring.rx_pending  = test->ifobj_tx->ring.rx_max_pending;
 	test->ifobj_rx->umem->num_frames = max_descs;
-	test->ifobj_rx->xsk->rxqsize = max_descs;
+	test->ifobj_rx->umem->fill_size = max_descs;
+	test->ifobj_rx->umem->comp_size = max_descs;
 	test->ifobj_tx->xsk->batch_size = XSK_RING_PROD__DEFAULT_NUM_DESCS;
 	test->ifobj_rx->xsk->batch_size = XSK_RING_PROD__DEFAULT_NUM_DESCS;
 
@@ -2461,9 +2472,12 @@ static int testapp_hw_sw_max_ring_size(struct test_spec *test)
 	if (ret)
 		return ret;
 
-	/* Set batch_size to 4095 */
-	test->ifobj_tx->xsk->batch_size = max_descs - 1;
-	test->ifobj_rx->xsk->batch_size = max_descs - 1;
+	/* Set batch_size to 8152 for testing, as the ice HW ignores the 3 lowest bits when
+	 * updating the Rx HW tail register.
+	 */
+	test->ifobj_tx->xsk->batch_size = test->ifobj_tx->ring.tx_max_pending - 8;
+	test->ifobj_rx->xsk->batch_size = test->ifobj_tx->ring.tx_max_pending - 8;
+	pkt_stream_replace(test, max_descs, MIN_PKT_SIZE);
 	return testapp_validate_traffic(test);
 }
 
diff --git a/tools/testing/selftests/bpf/xskxceiver.h b/tools/testing/selftests/bpf/xskxceiver.h
index 906de5fab7a3..885c948c5d83 100644
--- a/tools/testing/selftests/bpf/xskxceiver.h
+++ b/tools/testing/selftests/bpf/xskxceiver.h
@@ -80,6 +80,8 @@ struct xsk_umem_info {
 	void *buffer;
 	u32 frame_size;
 	u32 base_addr;
+	u32 fill_size;
+	u32 comp_size;
 	bool unaligned_mode;
 };
 
-- 
cgit v1.2.3-58-ga151


From 2896624be30b049601ec3ef9b08df184d0c70495 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Fri, 28 Jun 2024 12:18:54 +0200
Subject: net: Remove task_struct::bpf_net_context init on fork.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There is no clone() invocation within a bpf_net_ctx_…() block. Therefore
the task_struct::bpf_net_context has always to be NULL and an explicit
initialisation is not required.

Remove the NULL assignment in the clone() path.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 kernel/fork.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/kernel/fork.c b/kernel/fork.c
index f314bdd7e610..99076dbe27d8 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -2355,7 +2355,6 @@ __latent_entropy struct task_struct *copy_process(
 	RCU_INIT_POINTER(p->bpf_storage, NULL);
 	p->bpf_ctx = NULL;
 #endif
-	p->bpf_net_context =  NULL;
 
 	/* Perform scheduler related setup. Assign this task to a CPU. */
 	retval = sched_fork(clone_flags, p);
-- 
cgit v1.2.3-58-ga151


From d839a73179ae91c07f5f2f97ccb9c69b2b7c3306 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Fri, 28 Jun 2024 12:18:55 +0200
Subject: net: Optimize xdp_do_flush() with bpf_net_context infos.

Every NIC driver utilizing XDP should invoke xdp_do_flush() after
processing all packages. With the introduction of the bpf_net_context
logic the flush lists (for dev, CPU-map and xsk) are lazy initialized
only if used. However xdp_do_flush() tries to flush all three of them so
all three lists are always initialized and the likely empty lists are
"iterated".
Without the usage of XDP but with CONFIG_DEBUG_NET the lists are also
initialized due to xdp_do_check_flushed().

Jakub suggest to utilize the hints in bpf_net_context and avoid invoking
the flush function. This will also avoiding initializing the lists which
are otherwise unused.

Introduce bpf_net_ctx_get_all_used_flush_lists() to return the
individual list if not-empty. Use the logic in xdp_do_flush() and
xdp_do_check_flushed(). Remove the not needed .*_check_flush().

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/linux/bpf.h    | 10 ++++------
 include/linux/filter.h | 27 +++++++++++++++++++++++++++
 include/net/xdp_sock.h | 14 ++------------
 kernel/bpf/cpumap.c    | 13 +------------
 kernel/bpf/devmap.c    | 13 +------------
 net/core/filter.c      | 33 +++++++++++++++++++++++++--------
 net/xdp/xsk.c          | 13 +------------
 7 files changed, 61 insertions(+), 62 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index a834f4b761bc..f5c6bc9093a6 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -2494,7 +2494,7 @@ struct sk_buff;
 struct bpf_dtab_netdev;
 struct bpf_cpu_map_entry;
 
-void __dev_flush(void);
+void __dev_flush(struct list_head *flush_list);
 int dev_xdp_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
 		    struct net_device *dev_rx);
 int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_frame *xdpf,
@@ -2507,7 +2507,7 @@ int dev_map_redirect_multi(struct net_device *dev, struct sk_buff *skb,
 			   struct bpf_prog *xdp_prog, struct bpf_map *map,
 			   bool exclude_ingress);
 
-void __cpu_map_flush(void);
+void __cpu_map_flush(struct list_head *flush_list);
 int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_frame *xdpf,
 		    struct net_device *dev_rx);
 int cpu_map_generic_redirect(struct bpf_cpu_map_entry *rcpu,
@@ -2644,8 +2644,6 @@ void bpf_dynptr_init(struct bpf_dynptr_kern *ptr, void *data,
 void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr);
 void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr);
 
-bool dev_check_flush(void);
-bool cpu_map_check_flush(void);
 #else /* !CONFIG_BPF_SYSCALL */
 static inline struct bpf_prog *bpf_prog_get(u32 ufd)
 {
@@ -2738,7 +2736,7 @@ static inline struct bpf_token *bpf_token_get_from_fd(u32 ufd)
 	return ERR_PTR(-EOPNOTSUPP);
 }
 
-static inline void __dev_flush(void)
+static inline void __dev_flush(struct list_head *flush_list)
 {
 }
 
@@ -2784,7 +2782,7 @@ int dev_map_redirect_multi(struct net_device *dev, struct sk_buff *skb,
 	return 0;
 }
 
-static inline void __cpu_map_flush(void)
+static inline void __cpu_map_flush(struct list_head *flush_list)
 {
 }
 
diff --git a/include/linux/filter.h b/include/linux/filter.h
index c0349522de8f..02ddcfdf94c4 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -829,6 +829,33 @@ static inline struct list_head *bpf_net_ctx_get_xskmap_flush_list(void)
 	return &bpf_net_ctx->xskmap_map_flush_list;
 }
 
+static inline void bpf_net_ctx_get_all_used_flush_lists(struct list_head **lh_map,
+							struct list_head **lh_dev,
+							struct list_head **lh_xsk)
+{
+	struct bpf_net_context *bpf_net_ctx = bpf_net_ctx_get();
+	u32 kern_flags = bpf_net_ctx->ri.kern_flags;
+	struct list_head *lh;
+
+	*lh_map = *lh_dev = *lh_xsk = NULL;
+
+	if (!IS_ENABLED(CONFIG_BPF_SYSCALL))
+		return;
+
+	lh = &bpf_net_ctx->dev_map_flush_list;
+	if (kern_flags & BPF_RI_F_DEV_MAP_INIT && !list_empty(lh))
+		*lh_dev = lh;
+
+	lh = &bpf_net_ctx->cpu_map_flush_list;
+	if (kern_flags & BPF_RI_F_CPU_MAP_INIT && !list_empty(lh))
+		*lh_map = lh;
+
+	lh = &bpf_net_ctx->xskmap_map_flush_list;
+	if (IS_ENABLED(CONFIG_XDP_SOCKETS) &&
+	    kern_flags & BPF_RI_F_XSK_MAP_INIT && !list_empty(lh))
+		*lh_xsk = lh;
+}
+
 /* Compute the linear packet data range [data, data_end) which
  * will be accessed by various program types (cls_bpf, act_bpf,
  * lwt, ...). Subsystems allowing direct data access must (!)
diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h
index 3d54de168a6d..bfe625b55d55 100644
--- a/include/net/xdp_sock.h
+++ b/include/net/xdp_sock.h
@@ -121,7 +121,7 @@ struct xsk_tx_metadata_ops {
 
 int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp);
 int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp);
-void __xsk_map_flush(void);
+void __xsk_map_flush(struct list_head *flush_list);
 
 /**
  *  xsk_tx_metadata_to_compl - Save enough relevant metadata information
@@ -206,7 +206,7 @@ static inline int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp)
 	return -EOPNOTSUPP;
 }
 
-static inline void __xsk_map_flush(void)
+static inline void __xsk_map_flush(struct list_head *flush_list)
 {
 }
 
@@ -228,14 +228,4 @@ static inline void xsk_tx_metadata_complete(struct xsk_tx_metadata_compl *compl,
 }
 
 #endif /* CONFIG_XDP_SOCKETS */
-
-#if defined(CONFIG_XDP_SOCKETS) && defined(CONFIG_DEBUG_NET)
-bool xsk_map_check_flush(void);
-#else
-static inline bool xsk_map_check_flush(void)
-{
-	return false;
-}
-#endif
-
 #endif /* _LINUX_XDP_SOCK_H */
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index 068e994ed781..4acf90cd79eb 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -757,9 +757,8 @@ trace:
 	return ret;
 }
 
-void __cpu_map_flush(void)
+void __cpu_map_flush(struct list_head *flush_list)
 {
-	struct list_head *flush_list = bpf_net_ctx_get_cpu_map_flush_list();
 	struct xdp_bulk_queue *bq, *tmp;
 
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node) {
@@ -769,13 +768,3 @@ void __cpu_map_flush(void)
 		wake_up_process(bq->obj->kthread);
 	}
 }
-
-#ifdef CONFIG_DEBUG_NET
-bool cpu_map_check_flush(void)
-{
-	if (list_empty(bpf_net_ctx_get_cpu_map_flush_list()))
-		return false;
-	__cpu_map_flush();
-	return true;
-}
-#endif
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 317ac2d66ebd..9ca47eaacdd5 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -412,9 +412,8 @@ out:
  * driver before returning from its napi->poll() routine. See the comment above
  * xdp_do_flush() in filter.c.
  */
-void __dev_flush(void)
+void __dev_flush(struct list_head *flush_list)
 {
-	struct list_head *flush_list = bpf_net_ctx_get_dev_flush_list();
 	struct xdp_dev_bulk_queue *bq, *tmp;
 
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node) {
@@ -425,16 +424,6 @@ void __dev_flush(void)
 	}
 }
 
-#ifdef CONFIG_DEBUG_NET
-bool dev_check_flush(void)
-{
-	if (list_empty(bpf_net_ctx_get_dev_flush_list()))
-		return false;
-	__dev_flush();
-	return true;
-}
-#endif
-
 /* Elements are kept alive by RCU; either by rcu_read_lock() (from syscall) or
  * by local_bh_disable() (from XDP calls inside NAPI). The
  * rcu_read_lock_bh_held() below makes lockdep accept both.
diff --git a/net/core/filter.c b/net/core/filter.c
index eb1c4425c06f..403d23faf22e 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -4277,22 +4277,39 @@ static const struct bpf_func_proto bpf_xdp_adjust_meta_proto = {
  */
 void xdp_do_flush(void)
 {
-	__dev_flush();
-	__cpu_map_flush();
-	__xsk_map_flush();
+	struct list_head *lh_map, *lh_dev, *lh_xsk;
+
+	bpf_net_ctx_get_all_used_flush_lists(&lh_map, &lh_dev, &lh_xsk);
+	if (lh_dev)
+		__dev_flush(lh_dev);
+	if (lh_map)
+		__cpu_map_flush(lh_map);
+	if (lh_xsk)
+		__xsk_map_flush(lh_xsk);
 }
 EXPORT_SYMBOL_GPL(xdp_do_flush);
 
 #if defined(CONFIG_DEBUG_NET) && defined(CONFIG_BPF_SYSCALL)
 void xdp_do_check_flushed(struct napi_struct *napi)
 {
-	bool ret;
+	struct list_head *lh_map, *lh_dev, *lh_xsk;
+	bool missed = false;
 
-	ret = dev_check_flush();
-	ret |= cpu_map_check_flush();
-	ret |= xsk_map_check_flush();
+	bpf_net_ctx_get_all_used_flush_lists(&lh_map, &lh_dev, &lh_xsk);
+	if (lh_dev) {
+		__dev_flush(lh_dev);
+		missed = true;
+	}
+	if (lh_map) {
+		__cpu_map_flush(lh_map);
+		missed = true;
+	}
+	if (lh_xsk) {
+		__xsk_map_flush(lh_xsk);
+		missed = true;
+	}
 
-	WARN_ONCE(ret, "Missing xdp_do_flush() invocation after NAPI by %ps\n",
+	WARN_ONCE(missed, "Missing xdp_do_flush() invocation after NAPI by %ps\n",
 		  napi->poll);
 }
 #endif
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index ed062e038389..de9c0322bc29 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -383,9 +383,8 @@ int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp)
 	return 0;
 }
 
-void __xsk_map_flush(void)
+void __xsk_map_flush(struct list_head *flush_list)
 {
-	struct list_head *flush_list = bpf_net_ctx_get_xskmap_flush_list();
 	struct xdp_sock *xs, *tmp;
 
 	list_for_each_entry_safe(xs, tmp, flush_list, flush_node) {
@@ -394,16 +393,6 @@ void __xsk_map_flush(void)
 	}
 }
 
-#ifdef CONFIG_DEBUG_NET
-bool xsk_map_check_flush(void)
-{
-	if (list_empty(bpf_net_ctx_get_xskmap_flush_list()))
-		return false;
-	__xsk_map_flush();
-	return true;
-}
-#endif
-
 void xsk_tx_completed(struct xsk_buff_pool *pool, u32 nb_entries)
 {
 	xskq_prod_submit_n(pool->cq, nb_entries);
-- 
cgit v1.2.3-58-ga151


From e3d69f585d651aba877e18866de7e8cfa2476caa Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Fri, 28 Jun 2024 12:18:56 +0200
Subject: net: Move flush list retrieval to where it is used.

The bpf_net_ctx_get_.*_flush_list() are used at the top of the function.
This means the variable is always assigned even if unused. By moving the
function to where it is used, it is possible to delay the initialisation
until it is unavoidable.
Not sure how much this gains in reality but by looking at bq_enqueue()
(in devmap.c) gcc pushes one register less to the stack. \o/.

 Move flush list retrieval to where it is used.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Acked-by: Jesper Dangaard Brouer <hawk@kernel.org>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 kernel/bpf/cpumap.c | 6 ++++--
 kernel/bpf/devmap.c | 3 ++-
 net/xdp/xsk.c       | 6 ++++--
 3 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index 4acf90cd79eb..fbdf5a1aabfe 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -707,7 +707,6 @@ static void bq_flush_to_queue(struct xdp_bulk_queue *bq)
  */
 static void bq_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_frame *xdpf)
 {
-	struct list_head *flush_list = bpf_net_ctx_get_cpu_map_flush_list();
 	struct xdp_bulk_queue *bq = this_cpu_ptr(rcpu->bulkq);
 
 	if (unlikely(bq->count == CPU_MAP_BULK_SIZE))
@@ -724,8 +723,11 @@ static void bq_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_frame *xdpf)
 	 */
 	bq->q[bq->count++] = xdpf;
 
-	if (!bq->flush_node.prev)
+	if (!bq->flush_node.prev) {
+		struct list_head *flush_list = bpf_net_ctx_get_cpu_map_flush_list();
+
 		list_add(&bq->flush_node, flush_list);
+	}
 }
 
 int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_frame *xdpf,
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 9ca47eaacdd5..b18d4a14a0a7 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -448,7 +448,6 @@ static void *__dev_map_lookup_elem(struct bpf_map *map, u32 key)
 static void bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
 		       struct net_device *dev_rx, struct bpf_prog *xdp_prog)
 {
-	struct list_head *flush_list = bpf_net_ctx_get_dev_flush_list();
 	struct xdp_dev_bulk_queue *bq = this_cpu_ptr(dev->xdp_bulkq);
 
 	if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
@@ -462,6 +461,8 @@ static void bq_enqueue(struct net_device *dev, struct xdp_frame *xdpf,
 	 * are only ever modified together.
 	 */
 	if (!bq->dev_rx) {
+		struct list_head *flush_list = bpf_net_ctx_get_dev_flush_list();
+
 		bq->dev_rx = dev_rx;
 		bq->xdp_prog = xdp_prog;
 		list_add(&bq->flush_node, flush_list);
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index de9c0322bc29..7e16336044b2 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -370,15 +370,17 @@ static int xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 
 int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp)
 {
-	struct list_head *flush_list = bpf_net_ctx_get_xskmap_flush_list();
 	int err;
 
 	err = xsk_rcv(xs, xdp);
 	if (err)
 		return err;
 
-	if (!xs->flush_node.prev)
+	if (!xs->flush_node.prev) {
+		struct list_head *flush_list = bpf_net_ctx_get_xskmap_flush_list();
+
 		list_add(&xs->flush_node, flush_list);
+	}
 
 	return 0;
 }
-- 
cgit v1.2.3-58-ga151


From d01e0e98de31f9cc5cf328f5eb7ae3d7a13c1021 Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Fri, 28 Jun 2024 18:45:12 +0300
Subject: dt-bindings: net: dwmac: Validate PBL for all IP-cores

Indeed the maximum DMA burst length can be programmed not only for DW
xGMACs, Allwinner EMACs and Spear SoC GMAC, but in accordance with
[1, 2, 3] for Generic DW *MAC IP-cores. Moreover the STMMAC driver parses
the property and then apply the configuration for all supported DW MAC
devices. All of that makes the property being available for all IP-cores
the bindings supports. Let's make sure the PBL-related properties are
validated for all of them by the common DW *MAC DT schema.

[1] DesignWare Cores Ethernet MAC Universal Databook, Revision 3.73a,
    October 2013, p.378.

[2] DesignWare Cores Ethernet Quality-of-Service Databook, Revision 5.10a,
    December 2017, p.1223.

[3] DesignWare Cores XGMAC - 10G Ethernet MAC Databook, Revision 2.11a,
    September 2015, p.469-473.

Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://patch.msgid.link/20240628154515.8783-1-fancer.lancer@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 .../devicetree/bindings/net/snps,dwmac.yaml        | 80 +++++++---------------
 1 file changed, 26 insertions(+), 54 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
index 3bab4e1f3fbf..0ab124324eec 100644
--- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
@@ -436,6 +436,32 @@ properties:
     description:
       Use Address-Aligned Beats
 
+  snps,pbl:
+    description:
+      Programmable Burst Length (tx and rx)
+    $ref: /schemas/types.yaml#/definitions/uint32
+    enum: [1, 2, 4, 8, 16, 32]
+
+  snps,txpbl:
+    description:
+      Tx Programmable Burst Length. If set, DMA tx will use this
+      value rather than snps,pbl.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    enum: [1, 2, 4, 8, 16, 32]
+
+  snps,rxpbl:
+    description:
+      Rx Programmable Burst Length. If set, DMA rx will use this
+      value rather than snps,pbl.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    enum: [1, 2, 4, 8, 16, 32]
+
+  snps,no-pbl-x8:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description:
+      Don\'t multiply the pbl/txpbl/rxpbl values by 8. For core
+      rev < 3.50, don\'t multiply the values by 4.
+
   snps,fixed-burst:
     $ref: /schemas/types.yaml#/definitions/flag
     description:
@@ -566,60 +592,6 @@ dependencies:
 
 allOf:
   - $ref: ethernet-controller.yaml#
-  - if:
-      properties:
-        compatible:
-          contains:
-            enum:
-              - allwinner,sun7i-a20-gmac
-              - allwinner,sun8i-a83t-emac
-              - allwinner,sun8i-h3-emac
-              - allwinner,sun8i-r40-gmac
-              - allwinner,sun8i-v3s-emac
-              - allwinner,sun50i-a64-emac
-              - ingenic,jz4775-mac
-              - ingenic,x1000-mac
-              - ingenic,x1600-mac
-              - ingenic,x1830-mac
-              - ingenic,x2000-mac
-              - qcom,sa8775p-ethqos
-              - qcom,sc8280xp-ethqos
-              - snps,dwmac-3.50a
-              - snps,dwmac-4.10a
-              - snps,dwmac-4.20a
-              - snps,dwmac-5.20
-              - snps,dwxgmac
-              - snps,dwxgmac-2.10
-              - st,spear600-gmac
-
-    then:
-      properties:
-        snps,pbl:
-          description:
-            Programmable Burst Length (tx and rx)
-          $ref: /schemas/types.yaml#/definitions/uint32
-          enum: [1, 2, 4, 8, 16, 32]
-
-        snps,txpbl:
-          description:
-            Tx Programmable Burst Length. If set, DMA tx will use this
-            value rather than snps,pbl.
-          $ref: /schemas/types.yaml#/definitions/uint32
-          enum: [1, 2, 4, 8, 16, 32]
-
-        snps,rxpbl:
-          description:
-            Rx Programmable Burst Length. If set, DMA rx will use this
-            value rather than snps,pbl.
-          $ref: /schemas/types.yaml#/definitions/uint32
-          enum: [1, 2, 4, 8, 16, 32]
-
-        snps,no-pbl-x8:
-          $ref: /schemas/types.yaml#/definitions/flag
-          description:
-            Don\'t multiply the pbl/txpbl/rxpbl values by 8. For core
-            rev < 3.50, don\'t multiply the values by 4.
-
   - if:
       properties:
         compatible:
-- 
cgit v1.2.3-58-ga151


From 3cd1d4651cebe9776a0142ade36ff9f2e3545436 Mon Sep 17 00:00:00 2001
From: Christophe Roullier <christophe.roullier@foss.st.com>
Date: Mon, 1 Jul 2024 08:48:37 +0200
Subject: net: stmmac: dwmac-stm32: Add test to verify if ETHCK is used before
 checking clk rate

When we want to use clock from RCC to clock Ethernet PHY (with ETHCK)
we need to check if value of clock rate is authorized.
If ETHCK is unused, the ETHCK frequency is 0Hz and validation fails.
It makes no sense to validate unused ETHCK, so skip the validation.

Fixes: 582ac134963e ("net: stmmac: dwmac-stm32: Separate out external clock rate validation")
Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Tested-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
index 23cf0a5b047f..d6a268237ca1 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
@@ -206,6 +206,9 @@ static int stm32mp1_validate_ethck_rate(struct plat_stmmacenet_data *plat_dat)
 	struct stm32_dwmac *dwmac = plat_dat->bsp_priv;
 	const u32 clk_rate = clk_get_rate(dwmac->clk_eth_ck);
 
+	if (!dwmac->enable_eth_ck)
+		return 0;
+
 	switch (plat_dat->mac_interface) {
 	case PHY_INTERFACE_MODE_MII:
 	case PHY_INTERFACE_MODE_GMII:
-- 
cgit v1.2.3-58-ga151


From f8dbe58e2f1a3c091531b3f8ef86b393ceee67d1 Mon Sep 17 00:00:00 2001
From: Christophe Roullier <christophe.roullier@foss.st.com>
Date: Mon, 1 Jul 2024 08:48:38 +0200
Subject: net: stmmac: dwmac-stm32: update err status in case different of
 stm32mp13

The mask parameter of syscfg property is mandatory for MP13 but
optional for all other cases.
The function should not return error code because for non-MP13
the missing syscfg phandle in DT is not considered an error.
So reset err to 0 in that case to support existing DTs without
syscfg phandle.

Fixes: 50bbc0393114 ("net: stmmac: dwmac-stm32: add management of stm32mp13 for stm32")
Signed-off-by: Christophe Roullier <christophe.roullier@foss.st.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Tested-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
index d6a268237ca1..c1732955a697 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
@@ -435,10 +435,12 @@ static int stm32_dwmac_parse_data(struct stm32_dwmac *dwmac,
 	dwmac->mode_mask = SYSCFG_MP1_ETH_MASK;
 	err = of_property_read_u32_index(np, "st,syscon", 2, &dwmac->mode_mask);
 	if (err) {
-		if (dwmac->ops->is_mp13)
+		if (dwmac->ops->is_mp13) {
 			dev_err(dev, "Sysconfig register mask must be set (%d)\n", err);
-		else
+		} else {
 			dev_dbg(dev, "Warning sysconfig register mask not set\n");
+			err = 0;
+		}
 	}
 
 	return err;
-- 
cgit v1.2.3-58-ga151


From 6801b0aef79db475591c3146a701ea373e4663b7 Mon Sep 17 00:00:00 2001
From: Pu Lehui <pulehui@huawei.com>
Date: Tue, 2 Jul 2024 12:19:42 +0000
Subject: riscv, bpf: Add 12-argument support for RV64 bpf trampoline
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch adds 12 function arguments support for riscv64 bpf trampoline.
The current bpf trampoline supports <= sizeof(u64) bytes scalar arguments [0]
and <= 16 bytes struct arguments [1]. Therefore, we focus on the situation
where scalars are at most XLEN bits and aggregates whose total size does not
exceed 2×XLEN bits in the riscv calling convention [2].

Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Björn Töpel <bjorn@rivosinc.com>
Acked-by: Björn Töpel <bjorn@kernel.org>
Acked-by: Puranjay Mohan <puranjay@kernel.org>
Link: https://elixir.bootlin.com/linux/v6.8/source/kernel/bpf/btf.c#L6184 [0]
Link: https://elixir.bootlin.com/linux/v6.8/source/kernel/bpf/btf.c#L6769 [1]
Link: https://github.com/riscv-non-isa/riscv-elf-psabi-doc/releases/download/draft-20230929-e5c800e661a53efe3c2678d71a306323b60eb13b/riscv-abi.pdf [2]
Link: https://lore.kernel.org/bpf/20240702121944.1091530-2-pulehui@huaweicloud.com
---
 arch/riscv/net/bpf_jit_comp64.c | 66 +++++++++++++++++++++++++++++------------
 1 file changed, 47 insertions(+), 19 deletions(-)

diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c
index 351e1484205e..685c7389ae7e 100644
--- a/arch/riscv/net/bpf_jit_comp64.c
+++ b/arch/riscv/net/bpf_jit_comp64.c
@@ -15,6 +15,7 @@
 #include <asm/percpu.h>
 #include "bpf_jit.h"
 
+#define RV_MAX_REG_ARGS 8
 #define RV_FENTRY_NINSNS 2
 /* imm that allows emit_imm to emit max count insns */
 #define RV_MAX_COUNT_IMM 0x7FFF7FF7FF7FF7FF
@@ -692,26 +693,45 @@ int bpf_arch_text_poke(void *ip, enum bpf_text_poke_type poke_type,
 	return ret;
 }
 
-static void store_args(int nregs, int args_off, struct rv_jit_context *ctx)
+static void store_args(int nr_arg_slots, int args_off, struct rv_jit_context *ctx)
 {
 	int i;
 
-	for (i = 0; i < nregs; i++) {
-		emit_sd(RV_REG_FP, -args_off, RV_REG_A0 + i, ctx);
+	for (i = 0; i < nr_arg_slots; i++) {
+		if (i < RV_MAX_REG_ARGS) {
+			emit_sd(RV_REG_FP, -args_off, RV_REG_A0 + i, ctx);
+		} else {
+			/* skip slots for T0 and FP of traced function */
+			emit_ld(RV_REG_T1, 16 + (i - RV_MAX_REG_ARGS) * 8, RV_REG_FP, ctx);
+			emit_sd(RV_REG_FP, -args_off, RV_REG_T1, ctx);
+		}
 		args_off -= 8;
 	}
 }
 
-static void restore_args(int nregs, int args_off, struct rv_jit_context *ctx)
+static void restore_args(int nr_reg_args, int args_off, struct rv_jit_context *ctx)
 {
 	int i;
 
-	for (i = 0; i < nregs; i++) {
+	for (i = 0; i < nr_reg_args; i++) {
 		emit_ld(RV_REG_A0 + i, -args_off, RV_REG_FP, ctx);
 		args_off -= 8;
 	}
 }
 
+static void restore_stack_args(int nr_stack_args, int args_off, int stk_arg_off,
+			       struct rv_jit_context *ctx)
+{
+	int i;
+
+	for (i = 0; i < nr_stack_args; i++) {
+		emit_ld(RV_REG_T1, -(args_off - RV_MAX_REG_ARGS * 8), RV_REG_FP, ctx);
+		emit_sd(RV_REG_FP, -stk_arg_off, RV_REG_T1, ctx);
+		args_off -= 8;
+		stk_arg_off -= 8;
+	}
+}
+
 static int invoke_bpf_prog(struct bpf_tramp_link *l, int args_off, int retval_off,
 			   int run_ctx_off, bool save_ret, struct rv_jit_context *ctx)
 {
@@ -784,8 +804,8 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 {
 	int i, ret, offset;
 	int *branches_off = NULL;
-	int stack_size = 0, nregs = m->nr_args;
-	int retval_off, args_off, nregs_off, ip_off, run_ctx_off, sreg_off;
+	int stack_size = 0, nr_arg_slots = 0;
+	int retval_off, args_off, nregs_off, ip_off, run_ctx_off, sreg_off, stk_arg_off;
 	struct bpf_tramp_links *fentry = &tlinks[BPF_TRAMP_FENTRY];
 	struct bpf_tramp_links *fexit = &tlinks[BPF_TRAMP_FEXIT];
 	struct bpf_tramp_links *fmod_ret = &tlinks[BPF_TRAMP_MODIFY_RETURN];
@@ -831,20 +851,21 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 	 * FP - sreg_off    [ callee saved reg	]
 	 *
 	 *		    [ pads              ] pads for 16 bytes alignment
+	 *
+	 *		    [ stack_argN        ]
+	 *		    [ ...               ]
+	 * FP - stk_arg_off [ stack_arg1        ] BPF_TRAMP_F_CALL_ORIG
 	 */
 
 	if (flags & (BPF_TRAMP_F_ORIG_STACK | BPF_TRAMP_F_SHARE_IPMODIFY))
 		return -ENOTSUPP;
 
-	/* extra regiters for struct arguments */
-	for (i = 0; i < m->nr_args; i++)
-		if (m->arg_flags[i] & BTF_FMODEL_STRUCT_ARG)
-			nregs += round_up(m->arg_size[i], 8) / 8 - 1;
-
-	/* 8 arguments passed by registers */
-	if (nregs > 8)
+	if (m->nr_args > MAX_BPF_FUNC_ARGS)
 		return -ENOTSUPP;
 
+	for (i = 0; i < m->nr_args; i++)
+		nr_arg_slots += round_up(m->arg_size[i], 8) / 8;
+
 	/* room of trampoline frame to store return address and frame pointer */
 	stack_size += 16;
 
@@ -854,7 +875,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 		retval_off = stack_size;
 	}
 
-	stack_size += nregs * 8;
+	stack_size += nr_arg_slots * 8;
 	args_off = stack_size;
 
 	stack_size += 8;
@@ -871,8 +892,14 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 	stack_size += 8;
 	sreg_off = stack_size;
 
+	if (nr_arg_slots - RV_MAX_REG_ARGS > 0)
+		stack_size += (nr_arg_slots - RV_MAX_REG_ARGS) * 8;
+
 	stack_size = round_up(stack_size, STACK_ALIGN);
 
+	/* room for args on stack must be at the top of stack */
+	stk_arg_off = stack_size;
+
 	if (!is_struct_ops) {
 		/* For the trampoline called from function entry,
 		 * the frame of traced function and the frame of
@@ -908,10 +935,10 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 		emit_sd(RV_REG_FP, -ip_off, RV_REG_T1, ctx);
 	}
 
-	emit_li(RV_REG_T1, nregs, ctx);
+	emit_li(RV_REG_T1, nr_arg_slots, ctx);
 	emit_sd(RV_REG_FP, -nregs_off, RV_REG_T1, ctx);
 
-	store_args(nregs, args_off, ctx);
+	store_args(nr_arg_slots, args_off, ctx);
 
 	/* skip to actual body of traced function */
 	if (flags & BPF_TRAMP_F_SKIP_FRAME)
@@ -951,7 +978,8 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 	}
 
 	if (flags & BPF_TRAMP_F_CALL_ORIG) {
-		restore_args(nregs, args_off, ctx);
+		restore_args(min_t(int, nr_arg_slots, RV_MAX_REG_ARGS), args_off, ctx);
+		restore_stack_args(nr_arg_slots - RV_MAX_REG_ARGS, args_off, stk_arg_off, ctx);
 		ret = emit_call((const u64)orig_call, true, ctx);
 		if (ret)
 			goto out;
@@ -986,7 +1014,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 	}
 
 	if (flags & BPF_TRAMP_F_RESTORE_REGS)
-		restore_args(nregs, args_off, ctx);
+		restore_args(min_t(int, nr_arg_slots, RV_MAX_REG_ARGS), args_off, ctx);
 
 	if (save_ret) {
 		emit_ld(RV_REG_A0, -retval_off, RV_REG_FP, ctx);
-- 
cgit v1.2.3-58-ga151


From 5d52ad36683af64f04da295d67fb943f94658929 Mon Sep 17 00:00:00 2001
From: Pu Lehui <pulehui@huawei.com>
Date: Tue, 2 Jul 2024 12:19:43 +0000
Subject: selftests/bpf: Factor out many args tests from tracing_struct

Factor out many args tests from tracing_struct and rename some function names
to make more sense. Meanwhile, remove unnecessary skeleton detach operation
as it will be covered by skeleton destroy operation.

Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240702121944.1091530-3-pulehui@huaweicloud.com
---
 .../selftests/bpf/prog_tests/tracing_struct.c      | 30 +++++++++--
 tools/testing/selftests/bpf/progs/tracing_struct.c | 54 -------------------
 .../selftests/bpf/progs/tracing_struct_many_args.c | 60 ++++++++++++++++++++++
 3 files changed, 86 insertions(+), 58 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/progs/tracing_struct_many_args.c

diff --git a/tools/testing/selftests/bpf/prog_tests/tracing_struct.c b/tools/testing/selftests/bpf/prog_tests/tracing_struct.c
index fe0fb0c9849a..cb2a95da2617 100644
--- a/tools/testing/selftests/bpf/prog_tests/tracing_struct.c
+++ b/tools/testing/selftests/bpf/prog_tests/tracing_struct.c
@@ -3,8 +3,9 @@
 
 #include <test_progs.h>
 #include "tracing_struct.skel.h"
+#include "tracing_struct_many_args.skel.h"
 
-static void test_fentry(void)
+static void test_struct_args(void)
 {
 	struct tracing_struct *skel;
 	int err;
@@ -55,6 +56,25 @@ static void test_fentry(void)
 
 	ASSERT_EQ(skel->bss->t6, 1, "t6 ret");
 
+destroy_skel:
+	tracing_struct__destroy(skel);
+}
+
+static void test_struct_many_args(void)
+{
+	struct tracing_struct_many_args *skel;
+	int err;
+
+	skel = tracing_struct_many_args__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "tracing_struct_many_args__open_and_load"))
+		return;
+
+	err = tracing_struct_many_args__attach(skel);
+	if (!ASSERT_OK(err, "tracing_struct_many_args__attach"))
+		goto destroy_skel;
+
+	ASSERT_OK(trigger_module_test_read(256), "trigger_read");
+
 	ASSERT_EQ(skel->bss->t7_a, 16, "t7:a");
 	ASSERT_EQ(skel->bss->t7_b, 17, "t7:b");
 	ASSERT_EQ(skel->bss->t7_c, 18, "t7:c");
@@ -74,12 +94,14 @@ static void test_fentry(void)
 	ASSERT_EQ(skel->bss->t8_g, 23, "t8:g");
 	ASSERT_EQ(skel->bss->t8_ret, 156, "t8 ret");
 
-	tracing_struct__detach(skel);
 destroy_skel:
-	tracing_struct__destroy(skel);
+	tracing_struct_many_args__destroy(skel);
 }
 
 void test_tracing_struct(void)
 {
-	test_fentry();
+	if (test__start_subtest("struct_args"))
+		test_struct_args();
+	if (test__start_subtest("struct_many_args"))
+		test_struct_many_args();
 }
diff --git a/tools/testing/selftests/bpf/progs/tracing_struct.c b/tools/testing/selftests/bpf/progs/tracing_struct.c
index 515daef3c84b..c435a3a8328a 100644
--- a/tools/testing/selftests/bpf/progs/tracing_struct.c
+++ b/tools/testing/selftests/bpf/progs/tracing_struct.c
@@ -18,11 +18,6 @@ struct bpf_testmod_struct_arg_3 {
 	int b[];
 };
 
-struct bpf_testmod_struct_arg_4 {
-	u64 a;
-	int b;
-};
-
 long t1_a_a, t1_a_b, t1_b, t1_c, t1_ret, t1_nregs;
 __u64 t1_reg0, t1_reg1, t1_reg2, t1_reg3;
 long t2_a, t2_b_a, t2_b_b, t2_c, t2_ret;
@@ -30,9 +25,6 @@ long t3_a, t3_b, t3_c_a, t3_c_b, t3_ret;
 long t4_a_a, t4_b, t4_c, t4_d, t4_e_a, t4_e_b, t4_ret;
 long t5_ret;
 int t6;
-long t7_a, t7_b, t7_c, t7_d, t7_e, t7_f_a, t7_f_b, t7_ret;
-long t8_a, t8_b, t8_c, t8_d, t8_e, t8_f_a, t8_f_b, t8_g, t8_ret;
-
 
 SEC("fentry/bpf_testmod_test_struct_arg_1")
 int BPF_PROG2(test_struct_arg_1, struct bpf_testmod_struct_arg_2, a, int, b, int, c)
@@ -138,50 +130,4 @@ int BPF_PROG2(test_struct_arg_11, struct bpf_testmod_struct_arg_3 *, a)
 	return 0;
 }
 
-SEC("fentry/bpf_testmod_test_struct_arg_7")
-int BPF_PROG2(test_struct_arg_12, __u64, a, void *, b, short, c, int, d,
-	      void *, e, struct bpf_testmod_struct_arg_4, f)
-{
-	t7_a = a;
-	t7_b = (long)b;
-	t7_c = c;
-	t7_d = d;
-	t7_e = (long)e;
-	t7_f_a = f.a;
-	t7_f_b = f.b;
-	return 0;
-}
-
-SEC("fexit/bpf_testmod_test_struct_arg_7")
-int BPF_PROG2(test_struct_arg_13, __u64, a, void *, b, short, c, int, d,
-	      void *, e, struct bpf_testmod_struct_arg_4, f, int, ret)
-{
-	t7_ret = ret;
-	return 0;
-}
-
-SEC("fentry/bpf_testmod_test_struct_arg_8")
-int BPF_PROG2(test_struct_arg_14, __u64, a, void *, b, short, c, int, d,
-	      void *, e, struct bpf_testmod_struct_arg_4, f, int, g)
-{
-	t8_a = a;
-	t8_b = (long)b;
-	t8_c = c;
-	t8_d = d;
-	t8_e = (long)e;
-	t8_f_a = f.a;
-	t8_f_b = f.b;
-	t8_g = g;
-	return 0;
-}
-
-SEC("fexit/bpf_testmod_test_struct_arg_8")
-int BPF_PROG2(test_struct_arg_15, __u64, a, void *, b, short, c, int, d,
-	      void *, e, struct bpf_testmod_struct_arg_4, f, int, g,
-	      int, ret)
-{
-	t8_ret = ret;
-	return 0;
-}
-
 char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/tracing_struct_many_args.c b/tools/testing/selftests/bpf/progs/tracing_struct_many_args.c
new file mode 100644
index 000000000000..3de4bb918178
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/tracing_struct_many_args.c
@@ -0,0 +1,60 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <vmlinux.h>
+#include <bpf/bpf_tracing.h>
+#include <bpf/bpf_helpers.h>
+
+struct bpf_testmod_struct_arg_4 {
+	u64 a;
+	int b;
+};
+
+long t7_a, t7_b, t7_c, t7_d, t7_e, t7_f_a, t7_f_b, t7_ret;
+long t8_a, t8_b, t8_c, t8_d, t8_e, t8_f_a, t8_f_b, t8_g, t8_ret;
+
+SEC("fentry/bpf_testmod_test_struct_arg_7")
+int BPF_PROG2(test_struct_many_args_1, __u64, a, void *, b, short, c, int, d,
+	      void *, e, struct bpf_testmod_struct_arg_4, f)
+{
+	t7_a = a;
+	t7_b = (long)b;
+	t7_c = c;
+	t7_d = d;
+	t7_e = (long)e;
+	t7_f_a = f.a;
+	t7_f_b = f.b;
+	return 0;
+}
+
+SEC("fexit/bpf_testmod_test_struct_arg_7")
+int BPF_PROG2(test_struct_many_args_2, __u64, a, void *, b, short, c, int, d,
+	      void *, e, struct bpf_testmod_struct_arg_4, f, int, ret)
+{
+	t7_ret = ret;
+	return 0;
+}
+
+SEC("fentry/bpf_testmod_test_struct_arg_8")
+int BPF_PROG2(test_struct_many_args_3, __u64, a, void *, b, short, c, int, d,
+	      void *, e, struct bpf_testmod_struct_arg_4, f, int, g)
+{
+	t8_a = a;
+	t8_b = (long)b;
+	t8_c = c;
+	t8_d = d;
+	t8_e = (long)e;
+	t8_f_a = f.a;
+	t8_f_b = f.b;
+	t8_g = g;
+	return 0;
+}
+
+SEC("fexit/bpf_testmod_test_struct_arg_8")
+int BPF_PROG2(test_struct_many_args_4, __u64, a, void *, b, short, c, int, d,
+	      void *, e, struct bpf_testmod_struct_arg_4, f, int, g,
+	      int, ret)
+{
+	t8_ret = ret;
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-58-ga151


From 9474f72cd6573ee788013147e3590be4a28e085a Mon Sep 17 00:00:00 2001
From: Pu Lehui <pulehui@huawei.com>
Date: Tue, 2 Jul 2024 12:19:44 +0000
Subject: selftests/bpf: Add testcase where 7th argment is struct
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add testcase where 7th argument is struct for architectures with 8 argument
registers, and increase the complexity of the struct.

Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Björn Töpel <bjorn@rivosinc.com>
Acked-by: Björn Töpel <bjorn@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240702121944.1091530-4-pulehui@huaweicloud.com
---
 tools/testing/selftests/bpf/DENYLIST.aarch64       |  1 +
 .../selftests/bpf/bpf_testmod/bpf_testmod.c        | 19 ++++++++++++
 .../selftests/bpf/prog_tests/tracing_struct.c      | 14 +++++++++
 .../selftests/bpf/progs/tracing_struct_many_args.c | 35 ++++++++++++++++++++++
 4 files changed, 69 insertions(+)

diff --git a/tools/testing/selftests/bpf/DENYLIST.aarch64 b/tools/testing/selftests/bpf/DENYLIST.aarch64
index 0445ac38bc07..3c7c3e79aa93 100644
--- a/tools/testing/selftests/bpf/DENYLIST.aarch64
+++ b/tools/testing/selftests/bpf/DENYLIST.aarch64
@@ -6,6 +6,7 @@ kprobe_multi_test                                # needs CONFIG_FPROBE
 module_attach                                    # prog 'kprobe_multi': failed to auto-attach: -95
 fentry_test/fentry_many_args                     # fentry_many_args:FAIL:fentry_many_args_attach unexpected error: -524
 fexit_test/fexit_many_args                       # fexit_many_args:FAIL:fexit_many_args_attach unexpected error: -524
+tracing_struct/struct_many_args                  # struct_many_args:FAIL:tracing_struct_many_args__attach unexpected error: -524
 fill_link_info/kprobe_multi_link_info            # bpf_program__attach_kprobe_multi_opts unexpected error: -95
 fill_link_info/kretprobe_multi_link_info         # bpf_program__attach_kprobe_multi_opts unexpected error: -95
 fill_link_info/kprobe_multi_invalid_ubuff        # bpf_program__attach_kprobe_multi_opts unexpected error: -95
diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index d8bd01d8560b..f8962a1dd397 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -53,6 +53,13 @@ struct bpf_testmod_struct_arg_4 {
 	int b;
 };
 
+struct bpf_testmod_struct_arg_5 {
+	char a;
+	short b;
+	int c;
+	long d;
+};
+
 __bpf_hook_start();
 
 noinline int
@@ -110,6 +117,15 @@ bpf_testmod_test_struct_arg_8(u64 a, void *b, short c, int d, void *e,
 	return bpf_testmod_test_struct_arg_result;
 }
 
+noinline int
+bpf_testmod_test_struct_arg_9(u64 a, void *b, short c, int d, void *e, char f,
+			      short g, struct bpf_testmod_struct_arg_5 h, long i)
+{
+	bpf_testmod_test_struct_arg_result = a + (long)b + c + d + (long)e +
+		f + g + h.a + h.b + h.c + h.d + i;
+	return bpf_testmod_test_struct_arg_result;
+}
+
 noinline int
 bpf_testmod_test_arg_ptr_to_struct(struct bpf_testmod_struct_arg_1 *a) {
 	bpf_testmod_test_struct_arg_result = a->a;
@@ -305,6 +321,7 @@ bpf_testmod_test_read(struct file *file, struct kobject *kobj,
 	struct bpf_testmod_struct_arg_2 struct_arg2 = {2, 3};
 	struct bpf_testmod_struct_arg_3 *struct_arg3;
 	struct bpf_testmod_struct_arg_4 struct_arg4 = {21, 22};
+	struct bpf_testmod_struct_arg_5 struct_arg5 = {23, 24, 25, 26};
 	int i = 1;
 
 	while (bpf_testmod_return_ptr(i))
@@ -319,6 +336,8 @@ bpf_testmod_test_read(struct file *file, struct kobject *kobj,
 					    (void *)20, struct_arg4);
 	(void)bpf_testmod_test_struct_arg_8(16, (void *)17, 18, 19,
 					    (void *)20, struct_arg4, 23);
+	(void)bpf_testmod_test_struct_arg_9(16, (void *)17, 18, 19, (void *)20,
+					    21, 22, struct_arg5, 27);
 
 	(void)bpf_testmod_test_arg_ptr_to_struct(&struct_arg1_2);
 
diff --git a/tools/testing/selftests/bpf/prog_tests/tracing_struct.c b/tools/testing/selftests/bpf/prog_tests/tracing_struct.c
index cb2a95da2617..19e68d4b3532 100644
--- a/tools/testing/selftests/bpf/prog_tests/tracing_struct.c
+++ b/tools/testing/selftests/bpf/prog_tests/tracing_struct.c
@@ -94,6 +94,20 @@ static void test_struct_many_args(void)
 	ASSERT_EQ(skel->bss->t8_g, 23, "t8:g");
 	ASSERT_EQ(skel->bss->t8_ret, 156, "t8 ret");
 
+	ASSERT_EQ(skel->bss->t9_a, 16, "t9:a");
+	ASSERT_EQ(skel->bss->t9_b, 17, "t9:b");
+	ASSERT_EQ(skel->bss->t9_c, 18, "t9:c");
+	ASSERT_EQ(skel->bss->t9_d, 19, "t9:d");
+	ASSERT_EQ(skel->bss->t9_e, 20, "t9:e");
+	ASSERT_EQ(skel->bss->t9_f, 21, "t9:f");
+	ASSERT_EQ(skel->bss->t9_g, 22, "t9:f");
+	ASSERT_EQ(skel->bss->t9_h_a, 23, "t9:h.a");
+	ASSERT_EQ(skel->bss->t9_h_b, 24, "t9:h.b");
+	ASSERT_EQ(skel->bss->t9_h_c, 25, "t9:h.c");
+	ASSERT_EQ(skel->bss->t9_h_d, 26, "t9:h.d");
+	ASSERT_EQ(skel->bss->t9_i, 27, "t9:i");
+	ASSERT_EQ(skel->bss->t9_ret, 258, "t9 ret");
+
 destroy_skel:
 	tracing_struct_many_args__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/progs/tracing_struct_many_args.c b/tools/testing/selftests/bpf/progs/tracing_struct_many_args.c
index 3de4bb918178..4742012ace06 100644
--- a/tools/testing/selftests/bpf/progs/tracing_struct_many_args.c
+++ b/tools/testing/selftests/bpf/progs/tracing_struct_many_args.c
@@ -8,8 +8,16 @@ struct bpf_testmod_struct_arg_4 {
 	int b;
 };
 
+struct bpf_testmod_struct_arg_5 {
+	char a;
+	short b;
+	int c;
+	long d;
+};
+
 long t7_a, t7_b, t7_c, t7_d, t7_e, t7_f_a, t7_f_b, t7_ret;
 long t8_a, t8_b, t8_c, t8_d, t8_e, t8_f_a, t8_f_b, t8_g, t8_ret;
+long t9_a, t9_b, t9_c, t9_d, t9_e, t9_f, t9_g, t9_h_a, t9_h_b, t9_h_c, t9_h_d, t9_i, t9_ret;
 
 SEC("fentry/bpf_testmod_test_struct_arg_7")
 int BPF_PROG2(test_struct_many_args_1, __u64, a, void *, b, short, c, int, d,
@@ -57,4 +65,31 @@ int BPF_PROG2(test_struct_many_args_4, __u64, a, void *, b, short, c, int, d,
 	return 0;
 }
 
+SEC("fentry/bpf_testmod_test_struct_arg_9")
+int BPF_PROG2(test_struct_many_args_5, __u64, a, void *, b, short, c, int, d, void *, e,
+	      char, f, short, g, struct bpf_testmod_struct_arg_5, h, long, i)
+{
+	t9_a = a;
+	t9_b = (long)b;
+	t9_c = c;
+	t9_d = d;
+	t9_e = (long)e;
+	t9_f = f;
+	t9_g = g;
+	t9_h_a = h.a;
+	t9_h_b = h.b;
+	t9_h_c = h.c;
+	t9_h_d = h.d;
+	t9_i = i;
+	return 0;
+}
+
+SEC("fexit/bpf_testmod_test_struct_arg_9")
+int BPF_PROG2(test_struct_many_args_6, __u64, a, void *, b, short, c, int, d, void *, e,
+	      char, f, short, g, struct bpf_testmod_struct_arg_5, h, long, i, int, ret)
+{
+	t9_ret = ret;
+	return 0;
+}
+
 char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-58-ga151


From da5f8fd1f0d393d5eaaba9ad8c22d1c26bb2bf9b Mon Sep 17 00:00:00 2001
From: Tao Chen <chen.dylane@gmail.com>
Date: Tue, 2 Jul 2024 21:11:50 +0800
Subject: bpftool: Mount bpffs when pinmaps path not under the bpffs

As Quentin said [0], BPF map pinning will fail if the pinmaps path is not
under the bpffs, like:

  libbpf: specified path /home/ubuntu/test/sock_ops_map is not on BPF FS
  Error: failed to pin all maps

  [0] https://github.com/libbpf/bpftool/issues/146

Fixes: 3767a94b3253 ("bpftool: add pinmaps argument to the load/loadall")
Signed-off-by: Tao Chen <chen.dylane@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Quentin Monnet <qmo@kernel.org>
Reviewed-by: Quentin Monnet <qmo@kernel.org>
Link: https://lore.kernel.org/bpf/20240702131150.15622-1-chen.dylane@gmail.com
---
 tools/bpf/bpftool/prog.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index 1a501cf09e78..40ea743d139f 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -1813,6 +1813,10 @@ offload_dev:
 	}
 
 	if (pinmaps) {
+		err = create_and_mount_bpffs_dir(pinmaps);
+		if (err)
+			goto err_unpin;
+
 		err = bpf_object__pin_maps(obj, pinmaps);
 		if (err) {
 			p_err("failed to pin all maps");
-- 
cgit v1.2.3-58-ga151


From df34ec9db6f521118895f22795da49f2ec01f8cf Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:19 +0200
Subject: bpf: Fix atomic probe zero-extension

Zero-extending results of atomic probe operations fails with:

    verifier bug. zext_dst is set, but no reg is defined

The problem is that insn_def_regno() handles BPF_ATOMICs, but not
BPF_PROBE_ATOMICs. Fix by adding the missing condition.

Fixes: d503a04f8bc0 ("bpf: Add support for certain atomics in bpf_arena to x86 JIT")
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-2-iii@linux.ibm.com
---
 kernel/bpf/verifier.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index d3927d819465..e25ad5fb9115 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -3217,7 +3217,8 @@ static int insn_def_regno(const struct bpf_insn *insn)
 	case BPF_ST:
 		return -1;
 	case BPF_STX:
-		if (BPF_MODE(insn->code) == BPF_ATOMIC &&
+		if ((BPF_MODE(insn->code) == BPF_ATOMIC ||
+		     BPF_MODE(insn->code) == BPF_PROBE_ATOMIC) &&
 		    (insn->imm & BPF_FETCH)) {
 			if (insn->imm == BPF_CMPXCHG)
 				return BPF_REG_0;
-- 
cgit v1.2.3-58-ga151


From d0736d8c491ddc7d31c7f839d281c907366e2562 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:20 +0200
Subject: s390/bpf: Factor out emitting probe nops

The upcoming arena support for the loop-based BPF_XCHG implementation
requires emitting nop and extable entries separately. Move nop handling
into a separate function, and keep track of the nop offset.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-3-iii@linux.ibm.com
---
 arch/s390/net/bpf_jit_comp.c | 62 ++++++++++++++++++++++++++++----------------
 1 file changed, 40 insertions(+), 22 deletions(-)

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 4be8f5cadd02..260e7009784b 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -693,24 +693,52 @@ bool ex_handler_bpf(const struct exception_table_entry *x, struct pt_regs *regs)
 	return true;
 }
 
+/*
+ * A single BPF probe instruction
+ */
+struct bpf_jit_probe {
+	int prg;	/* JITed instruction offset */
+	int nop_prg;	/* JITed nop offset */
+};
+
+static void bpf_jit_probe_init(struct bpf_jit_probe *probe)
+{
+	probe->prg = -1;
+	probe->nop_prg = -1;
+}
+
+/*
+ * Handlers of certain exceptions leave psw.addr pointing to the instruction
+ * directly after the failing one. Therefore, create two exception table
+ * entries and also add a nop in case two probing instructions come directly
+ * after each other.
+ */
+static void bpf_jit_probe_emit_nop(struct bpf_jit *jit,
+				   struct bpf_jit_probe *probe)
+{
+	probe->nop_prg = jit->prg;
+	/* bcr 0,%0 */
+	_EMIT2(0x0700);
+}
+
 static int bpf_jit_probe_mem(struct bpf_jit *jit, struct bpf_prog *fp,
-			     int probe_prg, int nop_prg)
+			     struct bpf_jit_probe *probe)
 {
 	struct exception_table_entry *ex;
-	int reg, prg;
+	int i, prg, reg;
 	s64 delta;
 	u8 *insn;
-	int i;
 
+	bpf_jit_probe_emit_nop(jit, probe);
 	if (!fp->aux->extable)
 		/* Do nothing during early JIT passes. */
 		return 0;
-	insn = jit->prg_buf + probe_prg;
+	insn = jit->prg_buf + probe->prg;
 	reg = get_probe_mem_regno(insn);
 	if (WARN_ON_ONCE(reg < 0))
 		/* JIT bug - unexpected probe instruction. */
 		return -1;
-	if (WARN_ON_ONCE(probe_prg + insn_length(*insn) != nop_prg))
+	if (WARN_ON_ONCE(probe->prg + insn_length(*insn) != probe->nop_prg))
 		/* JIT bug - gap between probe and nop instructions. */
 		return -1;
 	for (i = 0; i < 2; i++) {
@@ -719,7 +747,7 @@ static int bpf_jit_probe_mem(struct bpf_jit *jit, struct bpf_prog *fp,
 			return -1;
 		ex = &fp->aux->extable[jit->excnt];
 		/* Add extable entries for probe and nop instructions. */
-		prg = i == 0 ? probe_prg : nop_prg;
+		prg = i == 0 ? probe->prg : probe->nop_prg;
 		delta = jit->prg_buf + prg - (u8 *)&ex->insn;
 		if (WARN_ON_ONCE(delta < INT_MIN || delta > INT_MAX))
 			/* JIT bug - code and extable must be close. */
@@ -729,7 +757,7 @@ static int bpf_jit_probe_mem(struct bpf_jit *jit, struct bpf_prog *fp,
 		 * Always land on the nop. Note that extable infrastructure
 		 * ignores fixup field, it is handled by ex_handler_bpf().
 		 */
-		delta = jit->prg_buf + nop_prg - (u8 *)&ex->fixup;
+		delta = jit->prg_buf + probe->nop_prg - (u8 *)&ex->fixup;
 		if (WARN_ON_ONCE(delta < INT_MIN || delta > INT_MAX))
 			/* JIT bug - landing pad and extable must be close. */
 			return -1;
@@ -782,19 +810,19 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 	s32 branch_oc_off = insn->off;
 	u32 dst_reg = insn->dst_reg;
 	u32 src_reg = insn->src_reg;
+	struct bpf_jit_probe probe;
 	int last, insn_count = 1;
 	u32 *addrs = jit->addrs;
 	s32 imm = insn->imm;
 	s16 off = insn->off;
-	int probe_prg = -1;
 	unsigned int mask;
-	int nop_prg;
 	int err;
 
+	bpf_jit_probe_init(&probe);
 	if (BPF_CLASS(insn->code) == BPF_LDX &&
 	    (BPF_MODE(insn->code) == BPF_PROBE_MEM ||
 	     BPF_MODE(insn->code) == BPF_PROBE_MEMSX))
-		probe_prg = jit->prg;
+		probe.prg = jit->prg;
 
 	switch (insn->code) {
 	/*
@@ -1897,18 +1925,8 @@ branch_oc:
 		return -1;
 	}
 
-	if (probe_prg != -1) {
-		/*
-		 * Handlers of certain exceptions leave psw.addr pointing to
-		 * the instruction directly after the failing one. Therefore,
-		 * create two exception table entries and also add a nop in
-		 * case two probing instructions come directly after each
-		 * other.
-		 */
-		nop_prg = jit->prg;
-		/* bcr 0,%0 */
-		_EMIT2(0x0700);
-		err = bpf_jit_probe_mem(jit, fp, probe_prg, nop_prg);
+	if (probe.prg != -1) {
+		err = bpf_jit_probe_mem(jit, fp, &probe);
 		if (err < 0)
 			return err;
 	}
-- 
cgit v1.2.3-58-ga151


From 9a048587269174f218e8d8d737ebfa628589358f Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:21 +0200
Subject: s390/bpf: Get rid of get_probe_mem_regno()

Commit 7fc8c362e782 ("s390/bpf: encode register within extable entry")
introduced explicit passing of the number of the register to be cleared
to ex_handler_bpf(), which replaced deducing it from the respective
native load instruction using get_probe_mem_regno().

Replace the second and last usage in the same manner, and remove this
function.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-4-iii@linux.ibm.com
---
 arch/s390/net/bpf_jit_comp.c | 33 +++++++--------------------------
 1 file changed, 7 insertions(+), 26 deletions(-)

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 260e7009784b..d9d79aa2be1b 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -667,25 +667,6 @@ static void bpf_jit_epilogue(struct bpf_jit *jit, u32 stack_depth)
 	jit->prg += sizeof(struct bpf_plt);
 }
 
-static int get_probe_mem_regno(const u8 *insn)
-{
-	/*
-	 * insn must point to llgc, llgh, llgf, lg, lgb, lgh or lgf, which have
-	 * destination register at the same position.
-	 */
-	if (insn[0] != 0xe3) /* common prefix */
-		return -1;
-	if (insn[5] != 0x90 && /* llgc */
-	    insn[5] != 0x91 && /* llgh */
-	    insn[5] != 0x16 && /* llgf */
-	    insn[5] != 0x04 && /* lg */
-	    insn[5] != 0x77 && /* lgb */
-	    insn[5] != 0x15 && /* lgh */
-	    insn[5] != 0x14) /* lgf */
-		return -1;
-	return insn[1] >> 4;
-}
-
 bool ex_handler_bpf(const struct exception_table_entry *x, struct pt_regs *regs)
 {
 	regs->psw.addr = extable_fixup(x);
@@ -699,12 +680,14 @@ bool ex_handler_bpf(const struct exception_table_entry *x, struct pt_regs *regs)
 struct bpf_jit_probe {
 	int prg;	/* JITed instruction offset */
 	int nop_prg;	/* JITed nop offset */
+	int reg;	/* Register to clear on exception */
 };
 
 static void bpf_jit_probe_init(struct bpf_jit_probe *probe)
 {
 	probe->prg = -1;
 	probe->nop_prg = -1;
+	probe->reg = -1;
 }
 
 /*
@@ -725,7 +708,7 @@ static int bpf_jit_probe_mem(struct bpf_jit *jit, struct bpf_prog *fp,
 			     struct bpf_jit_probe *probe)
 {
 	struct exception_table_entry *ex;
-	int i, prg, reg;
+	int i, prg;
 	s64 delta;
 	u8 *insn;
 
@@ -734,10 +717,6 @@ static int bpf_jit_probe_mem(struct bpf_jit *jit, struct bpf_prog *fp,
 		/* Do nothing during early JIT passes. */
 		return 0;
 	insn = jit->prg_buf + probe->prg;
-	reg = get_probe_mem_regno(insn);
-	if (WARN_ON_ONCE(reg < 0))
-		/* JIT bug - unexpected probe instruction. */
-		return -1;
 	if (WARN_ON_ONCE(probe->prg + insn_length(*insn) != probe->nop_prg))
 		/* JIT bug - gap between probe and nop instructions. */
 		return -1;
@@ -763,7 +742,7 @@ static int bpf_jit_probe_mem(struct bpf_jit *jit, struct bpf_prog *fp,
 			return -1;
 		ex->fixup = delta;
 		ex->type = EX_TYPE_BPF;
-		ex->data = reg;
+		ex->data = probe->reg;
 		jit->excnt++;
 	}
 	return 0;
@@ -821,8 +800,10 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 	bpf_jit_probe_init(&probe);
 	if (BPF_CLASS(insn->code) == BPF_LDX &&
 	    (BPF_MODE(insn->code) == BPF_PROBE_MEM ||
-	     BPF_MODE(insn->code) == BPF_PROBE_MEMSX))
+	     BPF_MODE(insn->code) == BPF_PROBE_MEMSX)) {
 		probe.prg = jit->prg;
+		probe.reg = reg2hex[dst_reg];
+	}
 
 	switch (insn->code) {
 	/*
-- 
cgit v1.2.3-58-ga151


From 89b933a2013794d8272d432591a2a7a9c41f6351 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:22 +0200
Subject: s390/bpf: Introduce pre- and post- probe functions

Currently probe insns are handled by two "if" statements at the
beginning and at the end of bpf_jit_insn(). The first one needs to be
in sync with the huge insn->code statement that follows it, which was
not a problem so far, since the check is small.

The introduction of arena will make it significantly larger, and it
will no longer be obvious whether it is in sync with the opcode switch.

Move these statements to the new bpf_jit_probe_load_pre() and
bpf_jit_probe_post() functions, and call them only from cases that need
them.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-5-iii@linux.ibm.com
---
 arch/s390/net/bpf_jit_comp.c | 58 +++++++++++++++++++++++++++++++++-----------
 1 file changed, 44 insertions(+), 14 deletions(-)

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index d9d79aa2be1b..582fa3830772 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -704,14 +704,28 @@ static void bpf_jit_probe_emit_nop(struct bpf_jit *jit,
 	_EMIT2(0x0700);
 }
 
-static int bpf_jit_probe_mem(struct bpf_jit *jit, struct bpf_prog *fp,
-			     struct bpf_jit_probe *probe)
+static void bpf_jit_probe_load_pre(struct bpf_jit *jit, struct bpf_insn *insn,
+				   struct bpf_jit_probe *probe)
+{
+	if (BPF_MODE(insn->code) != BPF_PROBE_MEM &&
+	    BPF_MODE(insn->code) != BPF_PROBE_MEMSX)
+		return;
+
+	probe->prg = jit->prg;
+	probe->reg = reg2hex[insn->dst_reg];
+}
+
+static int bpf_jit_probe_post(struct bpf_jit *jit, struct bpf_prog *fp,
+			      struct bpf_jit_probe *probe)
 {
 	struct exception_table_entry *ex;
 	int i, prg;
 	s64 delta;
 	u8 *insn;
 
+	if (probe->prg == -1)
+		/* The probe is not armed. */
+		return 0;
 	bpf_jit_probe_emit_nop(jit, probe);
 	if (!fp->aux->extable)
 		/* Do nothing during early JIT passes. */
@@ -798,12 +812,6 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 	int err;
 
 	bpf_jit_probe_init(&probe);
-	if (BPF_CLASS(insn->code) == BPF_LDX &&
-	    (BPF_MODE(insn->code) == BPF_PROBE_MEM ||
-	     BPF_MODE(insn->code) == BPF_PROBE_MEMSX)) {
-		probe.prg = jit->prg;
-		probe.reg = reg2hex[dst_reg];
-	}
 
 	switch (insn->code) {
 	/*
@@ -1497,51 +1505,79 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 	 */
 	case BPF_LDX | BPF_MEM | BPF_B: /* dst = *(u8 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEM | BPF_B:
+		bpf_jit_probe_load_pre(jit, insn, &probe);
 		/* llgc %dst,0(off,%src) */
 		EMIT6_DISP_LH(0xe3000000, 0x0090, dst_reg, src_reg, REG_0, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		if (insn_is_zext(&insn[1]))
 			insn_count = 2;
 		break;
 	case BPF_LDX | BPF_MEMSX | BPF_B: /* dst = *(s8 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEMSX | BPF_B:
+		bpf_jit_probe_load_pre(jit, insn, &probe);
 		/* lgb %dst,0(off,%src) */
 		EMIT6_DISP_LH(0xe3000000, 0x0077, dst_reg, src_reg, REG_0, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		break;
 	case BPF_LDX | BPF_MEM | BPF_H: /* dst = *(u16 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEM | BPF_H:
+		bpf_jit_probe_load_pre(jit, insn, &probe);
 		/* llgh %dst,0(off,%src) */
 		EMIT6_DISP_LH(0xe3000000, 0x0091, dst_reg, src_reg, REG_0, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		if (insn_is_zext(&insn[1]))
 			insn_count = 2;
 		break;
 	case BPF_LDX | BPF_MEMSX | BPF_H: /* dst = *(s16 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEMSX | BPF_H:
+		bpf_jit_probe_load_pre(jit, insn, &probe);
 		/* lgh %dst,0(off,%src) */
 		EMIT6_DISP_LH(0xe3000000, 0x0015, dst_reg, src_reg, REG_0, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		break;
 	case BPF_LDX | BPF_MEM | BPF_W: /* dst = *(u32 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEM | BPF_W:
+		bpf_jit_probe_load_pre(jit, insn, &probe);
 		/* llgf %dst,off(%src) */
 		jit->seen |= SEEN_MEM;
 		EMIT6_DISP_LH(0xe3000000, 0x0016, dst_reg, src_reg, REG_0, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		if (insn_is_zext(&insn[1]))
 			insn_count = 2;
 		break;
 	case BPF_LDX | BPF_MEMSX | BPF_W: /* dst = *(s32 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEMSX | BPF_W:
+		bpf_jit_probe_load_pre(jit, insn, &probe);
 		/* lgf %dst,off(%src) */
 		jit->seen |= SEEN_MEM;
 		EMIT6_DISP_LH(0xe3000000, 0x0014, dst_reg, src_reg, REG_0, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		break;
 	case BPF_LDX | BPF_MEM | BPF_DW: /* dst = *(u64 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEM | BPF_DW:
+		bpf_jit_probe_load_pre(jit, insn, &probe);
 		/* lg %dst,0(off,%src) */
 		jit->seen |= SEEN_MEM;
 		EMIT6_DISP_LH(0xe3000000, 0x0004, dst_reg, src_reg, REG_0, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		break;
 	/*
 	 * BPF_JMP / CALL
@@ -1906,12 +1942,6 @@ branch_oc:
 		return -1;
 	}
 
-	if (probe.prg != -1) {
-		err = bpf_jit_probe_mem(jit, fp, &probe);
-		if (err < 0)
-			return err;
-	}
-
 	return insn_count;
 }
 
-- 
cgit v1.2.3-58-ga151


From a1c04bcc41f9638460a9c68f894fb770596380de Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:23 +0200
Subject: s390/bpf: Land on the next JITed instruction after exception

Currently we land on the nop, which is unnecessary: we can just as well
begin executing the next instruction. Furthermore, the upcoming arena
support for the loop-based BPF_XCHG implementation will require landing
on an instruction that comes after the loop.

So land on the next JITed instruction, which covers both cases.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-6-iii@linux.ibm.com
---
 arch/s390/net/bpf_jit_comp.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 582fa3830772..ecd53f8f0602 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -747,10 +747,11 @@ static int bpf_jit_probe_post(struct bpf_jit *jit, struct bpf_prog *fp,
 			return -1;
 		ex->insn = delta;
 		/*
-		 * Always land on the nop. Note that extable infrastructure
-		 * ignores fixup field, it is handled by ex_handler_bpf().
+		 * Land on the current instruction. Note that the extable
+		 * infrastructure ignores the fixup field; it is handled by
+		 * ex_handler_bpf().
 		 */
-		delta = jit->prg_buf + probe->nop_prg - (u8 *)&ex->fixup;
+		delta = jit->prg_buf + jit->prg - (u8 *)&ex->fixup;
 		if (WARN_ON_ONCE(delta < INT_MIN || delta > INT_MAX))
 			/* JIT bug - landing pad and extable must be close. */
 			return -1;
-- 
cgit v1.2.3-58-ga151


From 4d3a453b434fd2f389960890ae6d767f8d50c403 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:24 +0200
Subject: s390/bpf: Support BPF_PROBE_MEM32

BPF_PROBE_MEM32 is a new mode for LDX, ST and STX instructions. The JIT
is supposed to add the start address of the kernel arena mapping to the
%dst register, and use a probing variant of the respective memory
access.

Reuse the existing probing infrastructure for that. Put the arena
address into the literal pool, load it into %r1 and use that as an
index register. Do not clear any registers in ex_handler_bpf() for
failing ST and STX instructions.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-7-iii@linux.ibm.com
---
 arch/s390/net/bpf_jit_comp.c | 137 ++++++++++++++++++++++++++++++++++---------
 1 file changed, 110 insertions(+), 27 deletions(-)

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index ecd53f8f0602..4b62b5162dfb 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -53,6 +53,7 @@ struct bpf_jit {
 	int excnt;		/* Number of exception table entries */
 	int prologue_plt_ret;	/* Return address for prologue hotpatch PLT */
 	int prologue_plt;	/* Start of prologue hotpatch PLT */
+	int kern_arena;		/* Pool offset of kernel arena address */
 };
 
 #define SEEN_MEM	BIT(0)		/* use mem[] for temporary storage */
@@ -670,7 +671,8 @@ static void bpf_jit_epilogue(struct bpf_jit *jit, u32 stack_depth)
 bool ex_handler_bpf(const struct exception_table_entry *x, struct pt_regs *regs)
 {
 	regs->psw.addr = extable_fixup(x);
-	regs->gprs[x->data] = 0;
+	if (x->data != -1)
+		regs->gprs[x->data] = 0;
 	return true;
 }
 
@@ -681,6 +683,7 @@ struct bpf_jit_probe {
 	int prg;	/* JITed instruction offset */
 	int nop_prg;	/* JITed nop offset */
 	int reg;	/* Register to clear on exception */
+	int arena_reg;	/* Register to use for arena addressing */
 };
 
 static void bpf_jit_probe_init(struct bpf_jit_probe *probe)
@@ -688,6 +691,7 @@ static void bpf_jit_probe_init(struct bpf_jit_probe *probe)
 	probe->prg = -1;
 	probe->nop_prg = -1;
 	probe->reg = -1;
+	probe->arena_reg = REG_0;
 }
 
 /*
@@ -708,13 +712,31 @@ static void bpf_jit_probe_load_pre(struct bpf_jit *jit, struct bpf_insn *insn,
 				   struct bpf_jit_probe *probe)
 {
 	if (BPF_MODE(insn->code) != BPF_PROBE_MEM &&
-	    BPF_MODE(insn->code) != BPF_PROBE_MEMSX)
+	    BPF_MODE(insn->code) != BPF_PROBE_MEMSX &&
+	    BPF_MODE(insn->code) != BPF_PROBE_MEM32)
 		return;
 
+	if (BPF_MODE(insn->code) == BPF_PROBE_MEM32) {
+		/* lgrl %r1,kern_arena */
+		EMIT6_PCREL_RILB(0xc4080000, REG_W1, jit->kern_arena);
+		probe->arena_reg = REG_W1;
+	}
 	probe->prg = jit->prg;
 	probe->reg = reg2hex[insn->dst_reg];
 }
 
+static void bpf_jit_probe_store_pre(struct bpf_jit *jit, struct bpf_insn *insn,
+				    struct bpf_jit_probe *probe)
+{
+	if (BPF_MODE(insn->code) != BPF_PROBE_MEM32)
+		return;
+
+	/* lgrl %r1,kern_arena */
+	EMIT6_PCREL_RILB(0xc4080000, REG_W1, jit->kern_arena);
+	probe->arena_reg = REG_W1;
+	probe->prg = jit->prg;
+}
+
 static int bpf_jit_probe_post(struct bpf_jit *jit, struct bpf_prog *fp,
 			      struct bpf_jit_probe *probe)
 {
@@ -1384,51 +1406,99 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 	 * BPF_ST(X)
 	 */
 	case BPF_STX | BPF_MEM | BPF_B: /* *(u8 *)(dst + off) = src_reg */
-		/* stcy %src,off(%dst) */
-		EMIT6_DISP_LH(0xe3000000, 0x0072, src_reg, dst_reg, REG_0, off);
+	case BPF_STX | BPF_PROBE_MEM32 | BPF_B:
+		bpf_jit_probe_store_pre(jit, insn, &probe);
+		/* stcy %src,off(%dst,%arena) */
+		EMIT6_DISP_LH(0xe3000000, 0x0072, src_reg, dst_reg,
+			      probe.arena_reg, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		break;
 	case BPF_STX | BPF_MEM | BPF_H: /* (u16 *)(dst + off) = src */
-		/* sthy %src,off(%dst) */
-		EMIT6_DISP_LH(0xe3000000, 0x0070, src_reg, dst_reg, REG_0, off);
+	case BPF_STX | BPF_PROBE_MEM32 | BPF_H:
+		bpf_jit_probe_store_pre(jit, insn, &probe);
+		/* sthy %src,off(%dst,%arena) */
+		EMIT6_DISP_LH(0xe3000000, 0x0070, src_reg, dst_reg,
+			      probe.arena_reg, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		break;
 	case BPF_STX | BPF_MEM | BPF_W: /* *(u32 *)(dst + off) = src */
-		/* sty %src,off(%dst) */
-		EMIT6_DISP_LH(0xe3000000, 0x0050, src_reg, dst_reg, REG_0, off);
+	case BPF_STX | BPF_PROBE_MEM32 | BPF_W:
+		bpf_jit_probe_store_pre(jit, insn, &probe);
+		/* sty %src,off(%dst,%arena) */
+		EMIT6_DISP_LH(0xe3000000, 0x0050, src_reg, dst_reg,
+			      probe.arena_reg, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		break;
 	case BPF_STX | BPF_MEM | BPF_DW: /* (u64 *)(dst + off) = src */
-		/* stg %src,off(%dst) */
-		EMIT6_DISP_LH(0xe3000000, 0x0024, src_reg, dst_reg, REG_0, off);
+	case BPF_STX | BPF_PROBE_MEM32 | BPF_DW:
+		bpf_jit_probe_store_pre(jit, insn, &probe);
+		/* stg %src,off(%dst,%arena) */
+		EMIT6_DISP_LH(0xe3000000, 0x0024, src_reg, dst_reg,
+			      probe.arena_reg, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		break;
 	case BPF_ST | BPF_MEM | BPF_B: /* *(u8 *)(dst + off) = imm */
+	case BPF_ST | BPF_PROBE_MEM32 | BPF_B:
 		/* lhi %w0,imm */
 		EMIT4_IMM(0xa7080000, REG_W0, (u8) imm);
-		/* stcy %w0,off(dst) */
-		EMIT6_DISP_LH(0xe3000000, 0x0072, REG_W0, dst_reg, REG_0, off);
+		bpf_jit_probe_store_pre(jit, insn, &probe);
+		/* stcy %w0,off(%dst,%arena) */
+		EMIT6_DISP_LH(0xe3000000, 0x0072, REG_W0, dst_reg,
+			      probe.arena_reg, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		break;
 	case BPF_ST | BPF_MEM | BPF_H: /* (u16 *)(dst + off) = imm */
+	case BPF_ST | BPF_PROBE_MEM32 | BPF_H:
 		/* lhi %w0,imm */
 		EMIT4_IMM(0xa7080000, REG_W0, (u16) imm);
-		/* sthy %w0,off(dst) */
-		EMIT6_DISP_LH(0xe3000000, 0x0070, REG_W0, dst_reg, REG_0, off);
+		bpf_jit_probe_store_pre(jit, insn, &probe);
+		/* sthy %w0,off(%dst,%arena) */
+		EMIT6_DISP_LH(0xe3000000, 0x0070, REG_W0, dst_reg,
+			      probe.arena_reg, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		break;
 	case BPF_ST | BPF_MEM | BPF_W: /* *(u32 *)(dst + off) = imm */
+	case BPF_ST | BPF_PROBE_MEM32 | BPF_W:
 		/* llilf %w0,imm  */
 		EMIT6_IMM(0xc00f0000, REG_W0, (u32) imm);
-		/* sty %w0,off(%dst) */
-		EMIT6_DISP_LH(0xe3000000, 0x0050, REG_W0, dst_reg, REG_0, off);
+		bpf_jit_probe_store_pre(jit, insn, &probe);
+		/* sty %w0,off(%dst,%arena) */
+		EMIT6_DISP_LH(0xe3000000, 0x0050, REG_W0, dst_reg,
+			      probe.arena_reg, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		break;
 	case BPF_ST | BPF_MEM | BPF_DW: /* *(u64 *)(dst + off) = imm */
+	case BPF_ST | BPF_PROBE_MEM32 | BPF_DW:
 		/* lgfi %w0,imm */
 		EMIT6_IMM(0xc0010000, REG_W0, imm);
-		/* stg %w0,off(%dst) */
-		EMIT6_DISP_LH(0xe3000000, 0x0024, REG_W0, dst_reg, REG_0, off);
+		bpf_jit_probe_store_pre(jit, insn, &probe);
+		/* stg %w0,off(%dst,%arena) */
+		EMIT6_DISP_LH(0xe3000000, 0x0024, REG_W0, dst_reg,
+			      probe.arena_reg, off);
+		err = bpf_jit_probe_post(jit, fp, &probe);
+		if (err < 0)
+			return err;
 		jit->seen |= SEEN_MEM;
 		break;
 	/*
@@ -1506,9 +1576,11 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 	 */
 	case BPF_LDX | BPF_MEM | BPF_B: /* dst = *(u8 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEM | BPF_B:
+	case BPF_LDX | BPF_PROBE_MEM32 | BPF_B:
 		bpf_jit_probe_load_pre(jit, insn, &probe);
-		/* llgc %dst,0(off,%src) */
-		EMIT6_DISP_LH(0xe3000000, 0x0090, dst_reg, src_reg, REG_0, off);
+		/* llgc %dst,off(%src,%arena) */
+		EMIT6_DISP_LH(0xe3000000, 0x0090, dst_reg, src_reg,
+			      probe.arena_reg, off);
 		err = bpf_jit_probe_post(jit, fp, &probe);
 		if (err < 0)
 			return err;
@@ -1519,7 +1591,7 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 	case BPF_LDX | BPF_MEMSX | BPF_B: /* dst = *(s8 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEMSX | BPF_B:
 		bpf_jit_probe_load_pre(jit, insn, &probe);
-		/* lgb %dst,0(off,%src) */
+		/* lgb %dst,off(%src) */
 		EMIT6_DISP_LH(0xe3000000, 0x0077, dst_reg, src_reg, REG_0, off);
 		err = bpf_jit_probe_post(jit, fp, &probe);
 		if (err < 0)
@@ -1528,9 +1600,11 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 		break;
 	case BPF_LDX | BPF_MEM | BPF_H: /* dst = *(u16 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEM | BPF_H:
+	case BPF_LDX | BPF_PROBE_MEM32 | BPF_H:
 		bpf_jit_probe_load_pre(jit, insn, &probe);
-		/* llgh %dst,0(off,%src) */
-		EMIT6_DISP_LH(0xe3000000, 0x0091, dst_reg, src_reg, REG_0, off);
+		/* llgh %dst,off(%src,%arena) */
+		EMIT6_DISP_LH(0xe3000000, 0x0091, dst_reg, src_reg,
+			      probe.arena_reg, off);
 		err = bpf_jit_probe_post(jit, fp, &probe);
 		if (err < 0)
 			return err;
@@ -1541,7 +1615,7 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 	case BPF_LDX | BPF_MEMSX | BPF_H: /* dst = *(s16 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEMSX | BPF_H:
 		bpf_jit_probe_load_pre(jit, insn, &probe);
-		/* lgh %dst,0(off,%src) */
+		/* lgh %dst,off(%src) */
 		EMIT6_DISP_LH(0xe3000000, 0x0015, dst_reg, src_reg, REG_0, off);
 		err = bpf_jit_probe_post(jit, fp, &probe);
 		if (err < 0)
@@ -1550,10 +1624,12 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 		break;
 	case BPF_LDX | BPF_MEM | BPF_W: /* dst = *(u32 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEM | BPF_W:
+	case BPF_LDX | BPF_PROBE_MEM32 | BPF_W:
 		bpf_jit_probe_load_pre(jit, insn, &probe);
 		/* llgf %dst,off(%src) */
 		jit->seen |= SEEN_MEM;
-		EMIT6_DISP_LH(0xe3000000, 0x0016, dst_reg, src_reg, REG_0, off);
+		EMIT6_DISP_LH(0xe3000000, 0x0016, dst_reg, src_reg,
+			      probe.arena_reg, off);
 		err = bpf_jit_probe_post(jit, fp, &probe);
 		if (err < 0)
 			return err;
@@ -1572,10 +1648,12 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 		break;
 	case BPF_LDX | BPF_MEM | BPF_DW: /* dst = *(u64 *)(ul) (src + off) */
 	case BPF_LDX | BPF_PROBE_MEM | BPF_DW:
+	case BPF_LDX | BPF_PROBE_MEM32 | BPF_DW:
 		bpf_jit_probe_load_pre(jit, insn, &probe);
-		/* lg %dst,0(off,%src) */
+		/* lg %dst,off(%src,%arena) */
 		jit->seen |= SEEN_MEM;
-		EMIT6_DISP_LH(0xe3000000, 0x0004, dst_reg, src_reg, REG_0, off);
+		EMIT6_DISP_LH(0xe3000000, 0x0004, dst_reg, src_reg,
+			      probe.arena_reg, off);
 		err = bpf_jit_probe_post(jit, fp, &probe);
 		if (err < 0)
 			return err;
@@ -1988,12 +2066,17 @@ static int bpf_jit_prog(struct bpf_jit *jit, struct bpf_prog *fp,
 			bool extra_pass, u32 stack_depth)
 {
 	int i, insn_count, lit32_size, lit64_size;
+	u64 kern_arena;
 
 	jit->lit32 = jit->lit32_start;
 	jit->lit64 = jit->lit64_start;
 	jit->prg = 0;
 	jit->excnt = 0;
 
+	kern_arena = bpf_arena_get_kern_vm_start(fp->aux->arena);
+	if (kern_arena)
+		jit->kern_arena = _EMIT_CONST_U64(kern_arena);
+
 	bpf_jit_prologue(jit, fp, stack_depth);
 	if (bpf_set_addr(jit, 0) < 0)
 		return -1;
-- 
cgit v1.2.3-58-ga151


From 555469cc9be4a7f52c0ad07a4a237d63e8c5c5f4 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:25 +0200
Subject: s390/bpf: Support address space cast instruction

The new address cast instruction translates arena offsets to userspace
addresses. NULL pointers must not be translated.

The common code sets up the mappings in such a way that it's enough to
replace the higher 32 bits to achieve the desired result. s390x has
just an instruction for this: INSERT IMMEDIATE.

Implement the sequence using 3 instruction: LOAD AND TEST, BRANCH
RELATIVE ON CONDITION and INSERT IMMEDIATE.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-8-iii@linux.ibm.com
---
 arch/s390/net/bpf_jit_comp.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 4b62b5162dfb..39c1d9aa7f1e 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -54,6 +54,7 @@ struct bpf_jit {
 	int prologue_plt_ret;	/* Return address for prologue hotpatch PLT */
 	int prologue_plt;	/* Start of prologue hotpatch PLT */
 	int kern_arena;		/* Pool offset of kernel arena address */
+	u64 user_arena;		/* User arena address */
 };
 
 #define SEEN_MEM	BIT(0)		/* use mem[] for temporary storage */
@@ -863,6 +864,22 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 		}
 		break;
 	case BPF_ALU64 | BPF_MOV | BPF_X:
+		if (insn_is_cast_user(insn)) {
+			int patch_brc;
+
+			/* ltgr %dst,%src */
+			EMIT4(0xb9020000, dst_reg, src_reg);
+			/* brc 8,0f */
+			patch_brc = jit->prg;
+			EMIT4_PCREL_RIC(0xa7040000, 8, 0);
+			/* iihf %dst,user_arena>>32 */
+			EMIT6_IMM(0xc0080000, dst_reg, jit->user_arena >> 32);
+			/* 0: */
+			if (jit->prg_buf)
+				*(u16 *)(jit->prg_buf + patch_brc + 2) =
+					(jit->prg - patch_brc) >> 1;
+			break;
+		}
 		switch (insn->off) {
 		case 0: /* DST = SRC */
 			/* lgr %dst,%src */
@@ -2076,6 +2093,7 @@ static int bpf_jit_prog(struct bpf_jit *jit, struct bpf_prog *fp,
 	kern_arena = bpf_arena_get_kern_vm_start(fp->aux->arena);
 	if (kern_arena)
 		jit->kern_arena = _EMIT_CONST_U64(kern_arena);
+	jit->user_arena = bpf_arena_get_user_vm_start(fp->aux->arena);
 
 	bpf_jit_prologue(jit, fp, stack_depth);
 	if (bpf_set_addr(jit, 0) < 0)
-- 
cgit v1.2.3-58-ga151


From 1e36027e39b8b3fa567ce3d743dbda5954dc0a56 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:26 +0200
Subject: s390/bpf: Enable arena

Now that BPF_PROBE_MEM32 and address space cast instructions are
implemented, tell the verifier that the JIT supports arena.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-9-iii@linux.ibm.com
---
 arch/s390/net/bpf_jit_comp.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 39c1d9aa7f1e..1dd359c25ada 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -2820,3 +2820,8 @@ bool bpf_jit_supports_subprog_tailcalls(void)
 {
 	return true;
 }
+
+bool bpf_jit_supports_arena(void)
+{
+	return true;
+}
-- 
cgit v1.2.3-58-ga151


From 2f9469484a3b52c66b799de73bd1ca75617bc8d5 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:27 +0200
Subject: s390/bpf: Support arena atomics

s390x supports most BPF atomics using single instructions, which
makes implementing arena support a matter of adding arena address to
the base register (unfortunately atomics do not support index
registers), and wrapping the respective native instruction in probing
sequences.

An exception is BPF_XCHG, which is implemented using two different
memory accesses and a loop. Make sure there is enough extable entries
for both instructions. Compute the base address once for both memory
accesses. Since on exception we need to land after the loop, emit the
nops manually.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-10-iii@linux.ibm.com
---
 arch/s390/net/bpf_jit_comp.c | 104 ++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 94 insertions(+), 10 deletions(-)

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 1dd359c25ada..ddfc0e99872e 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -704,6 +704,10 @@ static void bpf_jit_probe_init(struct bpf_jit_probe *probe)
 static void bpf_jit_probe_emit_nop(struct bpf_jit *jit,
 				   struct bpf_jit_probe *probe)
 {
+	if (probe->prg == -1 || probe->nop_prg != -1)
+		/* The probe is not armed or nop is already emitted. */
+		return;
+
 	probe->nop_prg = jit->prg;
 	/* bcr 0,%0 */
 	_EMIT2(0x0700);
@@ -738,6 +742,21 @@ static void bpf_jit_probe_store_pre(struct bpf_jit *jit, struct bpf_insn *insn,
 	probe->prg = jit->prg;
 }
 
+static void bpf_jit_probe_atomic_pre(struct bpf_jit *jit,
+				     struct bpf_insn *insn,
+				     struct bpf_jit_probe *probe)
+{
+	if (BPF_MODE(insn->code) != BPF_PROBE_ATOMIC)
+		return;
+
+	/* lgrl %r1,kern_arena */
+	EMIT6_PCREL_RILB(0xc4080000, REG_W1, jit->kern_arena);
+	/* agr %r1,%dst */
+	EMIT4(0xb9080000, REG_W1, insn->dst_reg);
+	probe->arena_reg = REG_W1;
+	probe->prg = jit->prg;
+}
+
 static int bpf_jit_probe_post(struct bpf_jit *jit, struct bpf_prog *fp,
 			      struct bpf_jit_probe *probe)
 {
@@ -1523,15 +1542,30 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 	 */
 	case BPF_STX | BPF_ATOMIC | BPF_DW:
 	case BPF_STX | BPF_ATOMIC | BPF_W:
+	case BPF_STX | BPF_PROBE_ATOMIC | BPF_DW:
+	case BPF_STX | BPF_PROBE_ATOMIC | BPF_W:
 	{
 		bool is32 = BPF_SIZE(insn->code) == BPF_W;
 
+		/*
+		 * Unlike loads and stores, atomics have only a base register,
+		 * but no index register. For the non-arena case, simply use
+		 * %dst as a base. For the arena case, use the work register
+		 * %r1: first, load the arena base into it, and then add %dst
+		 * to it.
+		 */
+		probe.arena_reg = dst_reg;
+
 		switch (insn->imm) {
-/* {op32|op64} {%w0|%src},%src,off(%dst) */
 #define EMIT_ATOMIC(op32, op64) do {					\
+	bpf_jit_probe_atomic_pre(jit, insn, &probe);			\
+	/* {op32|op64} {%w0|%src},%src,off(%arena) */			\
 	EMIT6_DISP_LH(0xeb000000, is32 ? (op32) : (op64),		\
 		      (insn->imm & BPF_FETCH) ? src_reg : REG_W0,	\
-		      src_reg, dst_reg, off);				\
+		      src_reg, probe.arena_reg, off);			\
+	err = bpf_jit_probe_post(jit, fp, &probe);			\
+	if (err < 0)							\
+		return err;						\
 	if (insn->imm & BPF_FETCH) {					\
 		/* bcr 14,0 - see atomic_fetch_{add,and,or,xor}() */	\
 		_EMIT2(0x07e0);						\
@@ -1560,25 +1594,50 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 			EMIT_ATOMIC(0x00f7, 0x00e7);
 			break;
 #undef EMIT_ATOMIC
-		case BPF_XCHG:
-			/* {ly|lg} %w0,off(%dst) */
+		case BPF_XCHG: {
+			struct bpf_jit_probe load_probe = probe;
+			int loop_start;
+
+			bpf_jit_probe_atomic_pre(jit, insn, &load_probe);
+			/* {ly|lg} %w0,off(%arena) */
 			EMIT6_DISP_LH(0xe3000000,
 				      is32 ? 0x0058 : 0x0004, REG_W0, REG_0,
-				      dst_reg, off);
-			/* 0: {csy|csg} %w0,%src,off(%dst) */
+				      load_probe.arena_reg, off);
+			bpf_jit_probe_emit_nop(jit, &load_probe);
+			/* Reuse {ly|lg}'s arena_reg for {csy|csg}. */
+			if (load_probe.prg != -1) {
+				probe.prg = jit->prg;
+				probe.arena_reg = load_probe.arena_reg;
+			}
+			loop_start = jit->prg;
+			/* 0: {csy|csg} %w0,%src,off(%arena) */
 			EMIT6_DISP_LH(0xeb000000, is32 ? 0x0014 : 0x0030,
-				      REG_W0, src_reg, dst_reg, off);
+				      REG_W0, src_reg, probe.arena_reg, off);
+			bpf_jit_probe_emit_nop(jit, &probe);
 			/* brc 4,0b */
-			EMIT4_PCREL_RIC(0xa7040000, 4, jit->prg - 6);
+			EMIT4_PCREL_RIC(0xa7040000, 4, loop_start);
 			/* {llgfr|lgr} %src,%w0 */
 			EMIT4(is32 ? 0xb9160000 : 0xb9040000, src_reg, REG_W0);
+			/* Both probes should land here on exception. */
+			err = bpf_jit_probe_post(jit, fp, &load_probe);
+			if (err < 0)
+				return err;
+			err = bpf_jit_probe_post(jit, fp, &probe);
+			if (err < 0)
+				return err;
 			if (is32 && insn_is_zext(&insn[1]))
 				insn_count = 2;
 			break;
+		}
 		case BPF_CMPXCHG:
-			/* 0: {csy|csg} %b0,%src,off(%dst) */
+			bpf_jit_probe_atomic_pre(jit, insn, &probe);
+			/* 0: {csy|csg} %b0,%src,off(%arena) */
 			EMIT6_DISP_LH(0xeb000000, is32 ? 0x0014 : 0x0030,
-				      BPF_REG_0, src_reg, dst_reg, off);
+				      BPF_REG_0, src_reg,
+				      probe.arena_reg, off);
+			err = bpf_jit_probe_post(jit, fp, &probe);
+			if (err < 0)
+				return err;
 			break;
 		default:
 			pr_err("Unknown atomic operation %02x\n", insn->imm);
@@ -2142,9 +2201,25 @@ static struct bpf_binary_header *bpf_jit_alloc(struct bpf_jit *jit,
 					       struct bpf_prog *fp)
 {
 	struct bpf_binary_header *header;
+	struct bpf_insn *insn;
 	u32 extable_size;
 	u32 code_size;
+	int i;
 
+	for (i = 0; i < fp->len; i++) {
+		insn = &fp->insnsi[i];
+
+		if (BPF_CLASS(insn->code) == BPF_STX &&
+		    BPF_MODE(insn->code) == BPF_PROBE_ATOMIC &&
+		    (BPF_SIZE(insn->code) == BPF_DW ||
+		     BPF_SIZE(insn->code) == BPF_W) &&
+		    insn->imm == BPF_XCHG)
+			/*
+			 * bpf_jit_insn() emits a load and a compare-and-swap,
+			 * both of which need to be probed.
+			 */
+			fp->aux->num_exentries += 1;
+	}
 	/* We need two entries per insn. */
 	fp->aux->num_exentries *= 2;
 
@@ -2825,3 +2900,12 @@ bool bpf_jit_supports_arena(void)
 {
 	return true;
 }
+
+bool bpf_jit_supports_insn(struct bpf_insn *insn, bool in_arena)
+{
+	/*
+	 * Currently the verifier uses this function only to check which
+	 * atomic stores to arena are supported, and they all are.
+	 */
+	return true;
+}
-- 
cgit v1.2.3-58-ga151


From b6349fd3448cf349af327f90585a712d60265429 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:28 +0200
Subject: selftests/bpf: Introduce __arena_global

While clang uses __attribute__((address_space(1))) both for defining
arena pointers and arena globals, GCC requires different syntax for
both. While __arena covers the first use case, introduce __arena_global
to cover the second one.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-11-iii@linux.ibm.com
---
 tools/testing/selftests/bpf/bpf_arena_common.h    |  2 +
 tools/testing/selftests/bpf/progs/arena_atomics.c | 67 ++++++++++-------------
 2 files changed, 32 insertions(+), 37 deletions(-)

diff --git a/tools/testing/selftests/bpf/bpf_arena_common.h b/tools/testing/selftests/bpf/bpf_arena_common.h
index 567491f3e1b5..68a51dcc0669 100644
--- a/tools/testing/selftests/bpf/bpf_arena_common.h
+++ b/tools/testing/selftests/bpf/bpf_arena_common.h
@@ -34,10 +34,12 @@
 
 #if defined(__BPF_FEATURE_ADDR_SPACE_CAST) && !defined(BPF_ARENA_FORCE_ASM)
 #define __arena __attribute__((address_space(1)))
+#define __arena_global __attribute__((address_space(1)))
 #define cast_kern(ptr) /* nop for bpf prog. emitted by LLVM */
 #define cast_user(ptr) /* nop for bpf prog. emitted by LLVM */
 #else
 #define __arena
+#define __arena_global SEC(".addr_space.1")
 #define cast_kern(ptr) bpf_addr_space_cast(ptr, 0, 1)
 #define cast_user(ptr) bpf_addr_space_cast(ptr, 1, 0)
 #endif
diff --git a/tools/testing/selftests/bpf/progs/arena_atomics.c b/tools/testing/selftests/bpf/progs/arena_atomics.c
index 55f10563208d..77a4dfa9cdf9 100644
--- a/tools/testing/selftests/bpf/progs/arena_atomics.c
+++ b/tools/testing/selftests/bpf/progs/arena_atomics.c
@@ -25,20 +25,13 @@ bool skip_tests = true;
 
 __u32 pid = 0;
 
-#undef __arena
-#if defined(__BPF_FEATURE_ADDR_SPACE_CAST)
-#define __arena __attribute__((address_space(1)))
-#else
-#define __arena SEC(".addr_space.1")
-#endif
-
-__u64 __arena add64_value = 1;
-__u64 __arena add64_result = 0;
-__u32 __arena add32_value = 1;
-__u32 __arena add32_result = 0;
-__u64 __arena add_stack_value_copy = 0;
-__u64 __arena add_stack_result = 0;
-__u64 __arena add_noreturn_value = 1;
+__u64 __arena_global add64_value = 1;
+__u64 __arena_global add64_result = 0;
+__u32 __arena_global add32_value = 1;
+__u32 __arena_global add32_result = 0;
+__u64 __arena_global add_stack_value_copy = 0;
+__u64 __arena_global add_stack_result = 0;
+__u64 __arena_global add_noreturn_value = 1;
 
 SEC("raw_tp/sys_enter")
 int add(const void *ctx)
@@ -58,13 +51,13 @@ int add(const void *ctx)
 	return 0;
 }
 
-__s64 __arena sub64_value = 1;
-__s64 __arena sub64_result = 0;
-__s32 __arena sub32_value = 1;
-__s32 __arena sub32_result = 0;
-__s64 __arena sub_stack_value_copy = 0;
-__s64 __arena sub_stack_result = 0;
-__s64 __arena sub_noreturn_value = 1;
+__s64 __arena_global sub64_value = 1;
+__s64 __arena_global sub64_result = 0;
+__s32 __arena_global sub32_value = 1;
+__s32 __arena_global sub32_result = 0;
+__s64 __arena_global sub_stack_value_copy = 0;
+__s64 __arena_global sub_stack_result = 0;
+__s64 __arena_global sub_noreturn_value = 1;
 
 SEC("raw_tp/sys_enter")
 int sub(const void *ctx)
@@ -84,8 +77,8 @@ int sub(const void *ctx)
 	return 0;
 }
 
-__u64 __arena and64_value = (0x110ull << 32);
-__u32 __arena and32_value = 0x110;
+__u64 __arena_global and64_value = (0x110ull << 32);
+__u32 __arena_global and32_value = 0x110;
 
 SEC("raw_tp/sys_enter")
 int and(const void *ctx)
@@ -101,8 +94,8 @@ int and(const void *ctx)
 	return 0;
 }
 
-__u32 __arena or32_value = 0x110;
-__u64 __arena or64_value = (0x110ull << 32);
+__u32 __arena_global or32_value = 0x110;
+__u64 __arena_global or64_value = (0x110ull << 32);
 
 SEC("raw_tp/sys_enter")
 int or(const void *ctx)
@@ -117,8 +110,8 @@ int or(const void *ctx)
 	return 0;
 }
 
-__u64 __arena xor64_value = (0x110ull << 32);
-__u32 __arena xor32_value = 0x110;
+__u64 __arena_global xor64_value = (0x110ull << 32);
+__u32 __arena_global xor32_value = 0x110;
 
 SEC("raw_tp/sys_enter")
 int xor(const void *ctx)
@@ -133,12 +126,12 @@ int xor(const void *ctx)
 	return 0;
 }
 
-__u32 __arena cmpxchg32_value = 1;
-__u32 __arena cmpxchg32_result_fail = 0;
-__u32 __arena cmpxchg32_result_succeed = 0;
-__u64 __arena cmpxchg64_value = 1;
-__u64 __arena cmpxchg64_result_fail = 0;
-__u64 __arena cmpxchg64_result_succeed = 0;
+__u32 __arena_global cmpxchg32_value = 1;
+__u32 __arena_global cmpxchg32_result_fail = 0;
+__u32 __arena_global cmpxchg32_result_succeed = 0;
+__u64 __arena_global cmpxchg64_value = 1;
+__u64 __arena_global cmpxchg64_result_fail = 0;
+__u64 __arena_global cmpxchg64_result_succeed = 0;
 
 SEC("raw_tp/sys_enter")
 int cmpxchg(const void *ctx)
@@ -156,10 +149,10 @@ int cmpxchg(const void *ctx)
 	return 0;
 }
 
-__u64 __arena xchg64_value = 1;
-__u64 __arena xchg64_result = 0;
-__u32 __arena xchg32_value = 1;
-__u32 __arena xchg32_result = 0;
+__u64 __arena_global xchg64_value = 1;
+__u64 __arena_global xchg64_result = 0;
+__u32 __arena_global xchg32_value = 1;
+__u32 __arena_global xchg32_result = 0;
 
 SEC("raw_tp/sys_enter")
 int xchg(const void *ctx)
-- 
cgit v1.2.3-58-ga151


From 490c99d4ed99bb01dac8bf2896e27941403549c4 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:29 +0200
Subject: selftests/bpf: Add UAF tests for arena atomics

Check that __sync_*() functions don't cause kernel panics when handling
freed arena pages.

x86_64 does not support some arena atomics yet, and aarch64 may or may
not support them, based on the availability of LSE atomics at run time.
Do not enable this test for these architectures for simplicity.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-12-iii@linux.ibm.com
---
 .../selftests/bpf/prog_tests/arena_atomics.c       | 18 +++++
 tools/testing/selftests/bpf/progs/arena_atomics.c  | 76 ++++++++++++++++++++++
 2 files changed, 94 insertions(+)

diff --git a/tools/testing/selftests/bpf/prog_tests/arena_atomics.c b/tools/testing/selftests/bpf/prog_tests/arena_atomics.c
index 0807a48a58ee..26e7c06c6cb4 100644
--- a/tools/testing/selftests/bpf/prog_tests/arena_atomics.c
+++ b/tools/testing/selftests/bpf/prog_tests/arena_atomics.c
@@ -146,6 +146,22 @@ static void test_xchg(struct arena_atomics *skel)
 	ASSERT_EQ(skel->arena->xchg32_result, 1, "xchg32_result");
 }
 
+static void test_uaf(struct arena_atomics *skel)
+{
+	LIBBPF_OPTS(bpf_test_run_opts, topts);
+	int err, prog_fd;
+
+	/* No need to attach it, just run it directly */
+	prog_fd = bpf_program__fd(skel->progs.uaf);
+	err = bpf_prog_test_run_opts(prog_fd, &topts);
+	if (!ASSERT_OK(err, "test_run_opts err"))
+		return;
+	if (!ASSERT_OK(topts.retval, "test_run_opts retval"))
+		return;
+
+	ASSERT_EQ(skel->arena->uaf_recovery_fails, 0, "uaf_recovery_fails");
+}
+
 void test_arena_atomics(void)
 {
 	struct arena_atomics *skel;
@@ -180,6 +196,8 @@ void test_arena_atomics(void)
 		test_cmpxchg(skel);
 	if (test__start_subtest("xchg"))
 		test_xchg(skel);
+	if (test__start_subtest("uaf"))
+		test_uaf(skel);
 
 cleanup:
 	arena_atomics__destroy(skel);
diff --git a/tools/testing/selftests/bpf/progs/arena_atomics.c b/tools/testing/selftests/bpf/progs/arena_atomics.c
index 77a4dfa9cdf9..bb0acd79d28a 100644
--- a/tools/testing/selftests/bpf/progs/arena_atomics.c
+++ b/tools/testing/selftests/bpf/progs/arena_atomics.c
@@ -169,3 +169,79 @@ int xchg(const void *ctx)
 
 	return 0;
 }
+
+__u64 __arena_global uaf_sink;
+volatile __u64 __arena_global uaf_recovery_fails;
+
+SEC("syscall")
+int uaf(const void *ctx)
+{
+	if (pid != (bpf_get_current_pid_tgid() >> 32))
+		return 0;
+#if defined(ENABLE_ATOMICS_TESTS) && !defined(__TARGET_ARCH_arm64) && \
+    !defined(__TARGET_ARCH_x86)
+	__u32 __arena *page32;
+	__u64 __arena *page64;
+	void __arena *page;
+
+	page = bpf_arena_alloc_pages(&arena, NULL, 1, NUMA_NO_NODE, 0);
+	bpf_arena_free_pages(&arena, page, 1);
+	uaf_recovery_fails = 24;
+
+	page32 = (__u32 __arena *)page;
+	uaf_sink += __sync_fetch_and_add(page32, 1);
+	uaf_recovery_fails -= 1;
+	__sync_add_and_fetch(page32, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_fetch_and_sub(page32, 1);
+	uaf_recovery_fails -= 1;
+	__sync_sub_and_fetch(page32, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_fetch_and_and(page32, 1);
+	uaf_recovery_fails -= 1;
+	__sync_and_and_fetch(page32, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_fetch_and_or(page32, 1);
+	uaf_recovery_fails -= 1;
+	__sync_or_and_fetch(page32, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_fetch_and_xor(page32, 1);
+	uaf_recovery_fails -= 1;
+	__sync_xor_and_fetch(page32, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_val_compare_and_swap(page32, 0, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_lock_test_and_set(page32, 1);
+	uaf_recovery_fails -= 1;
+
+	page64 = (__u64 __arena *)page;
+	uaf_sink += __sync_fetch_and_add(page64, 1);
+	uaf_recovery_fails -= 1;
+	__sync_add_and_fetch(page64, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_fetch_and_sub(page64, 1);
+	uaf_recovery_fails -= 1;
+	__sync_sub_and_fetch(page64, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_fetch_and_and(page64, 1);
+	uaf_recovery_fails -= 1;
+	__sync_and_and_fetch(page64, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_fetch_and_or(page64, 1);
+	uaf_recovery_fails -= 1;
+	__sync_or_and_fetch(page64, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_fetch_and_xor(page64, 1);
+	uaf_recovery_fails -= 1;
+	__sync_xor_and_fetch(page64, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_val_compare_and_swap(page64, 0, 1);
+	uaf_recovery_fails -= 1;
+	uaf_sink += __sync_lock_test_and_set(page64, 1);
+	uaf_recovery_fails -= 1;
+#endif
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
-- 
cgit v1.2.3-58-ga151


From 69716e44a74af464060faa68fa2b54f3af03c16a Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Tue, 2 Jul 2024 01:40:30 +0200
Subject: selftests/bpf: Remove arena tests from DENYLIST.s390x

Now that the s390x JIT supports arena, remove the respective tests from
the denylist.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240701234304.14336-13-iii@linux.ibm.com
---
 tools/testing/selftests/bpf/DENYLIST.s390x | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/DENYLIST.s390x b/tools/testing/selftests/bpf/DENYLIST.s390x
index c34adf39eeb2..cb810a98e78f 100644
--- a/tools/testing/selftests/bpf/DENYLIST.s390x
+++ b/tools/testing/selftests/bpf/DENYLIST.s390x
@@ -4,6 +4,3 @@ exceptions				 # JIT does not support calling kfunc bpf_throw				       (excepti
 get_stack_raw_tp                         # user_stack corrupted user stack                                             (no backchain userspace)
 stacktrace_build_id                      # compare_map_keys stackid_hmap vs. stackmap err -2 errno 2                   (?)
 verifier_iterating_callbacks
-verifier_arena                           # JIT does not support arena
-arena_htab                               # JIT does not support arena
-arena_atomics
-- 
cgit v1.2.3-58-ga151


From fd8db07705c55a995c42b1e71afc42faad675b0b Mon Sep 17 00:00:00 2001
From: Florian Lehner <dev@der-flo.net>
Date: Sat, 15 Jun 2024 12:11:58 +0200
Subject: bpf, devmap: Add .map_alloc_check

Use the .map_allock_check callback to perform allocation checks before
allocating memory for the devmap.

Signed-off-by: Florian Lehner <dev@der-flo.net>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240615101158.57889-1-dev@der-flo.net
---
 kernel/bpf/devmap.c | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 7f3b34452243..da1fec906b96 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -107,7 +107,7 @@ static inline struct hlist_head *dev_map_index_hash(struct bpf_dtab *dtab,
 	return &dtab->dev_index_head[idx & (dtab->n_buckets - 1)];
 }
 
-static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
+static int dev_map_alloc_check(union bpf_attr *attr)
 {
 	u32 valsize = attr->value_size;
 
@@ -121,23 +121,28 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 	    attr->map_flags & ~DEV_CREATE_FLAG_MASK)
 		return -EINVAL;
 
+	if (attr->map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
+		/* Hash table size must be power of 2; roundup_pow_of_two()
+		 * can overflow into UB on 32-bit arches
+		 */
+		if (attr->max_entries > 1UL << 31)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
+{
 	/* Lookup returns a pointer straight to dev->ifindex, so make sure the
 	 * verifier prevents writes from the BPF side
 	 */
 	attr->map_flags |= BPF_F_RDONLY_PROG;
-
-
 	bpf_map_init_from_attr(&dtab->map, attr);
 
 	if (attr->map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
-		/* hash table size must be power of 2; roundup_pow_of_two() can
-		 * overflow into UB on 32-bit arches, so check that first
-		 */
-		if (dtab->map.max_entries > 1UL << 31)
-			return -EINVAL;
-
+		/* Hash table size must be power of 2 */
 		dtab->n_buckets = roundup_pow_of_two(dtab->map.max_entries);
-
 		dtab->dev_index_head = dev_map_create_hash(dtab->n_buckets,
 							   dtab->map.numa_node);
 		if (!dtab->dev_index_head)
@@ -1040,6 +1045,7 @@ static u64 dev_map_mem_usage(const struct bpf_map *map)
 BTF_ID_LIST_SINGLE(dev_map_btf_ids, struct, bpf_dtab)
 const struct bpf_map_ops dev_map_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
+	.map_alloc_check = dev_map_alloc_check,
 	.map_alloc = dev_map_alloc,
 	.map_free = dev_map_free,
 	.map_get_next_key = dev_map_get_next_key,
@@ -1054,6 +1060,7 @@ const struct bpf_map_ops dev_map_ops = {
 
 const struct bpf_map_ops dev_map_hash_ops = {
 	.map_meta_equal = bpf_map_meta_equal,
+	.map_alloc_check = dev_map_alloc_check,
 	.map_alloc = dev_map_alloc,
 	.map_free = dev_map_free,
 	.map_get_next_key = dev_map_hash_get_next_key,
-- 
cgit v1.2.3-58-ga151


From 4dec64c52e24c2c9a15f81c115f1be5ea35121cb Mon Sep 17 00:00:00 2001
From: Mina Almasry <almasrymina@google.com>
Date: Fri, 28 Jun 2024 00:32:42 +0000
Subject: page_pool: convert to use netmem

Abstract the memory type from the page_pool so we can later add support
for new memory types. Convert the page_pool to use the new netmem type
abstraction, rather than use struct page directly.

As of this patch the netmem type is a no-op abstraction: it's always a
struct page underneath. All the page pool internals are converted to
use struct netmem instead of struct page, and the page pool now exports
2 APIs:

1. The existing struct page API.
2. The new struct netmem API.

Keeping the existing API is transitional; we do not want to refactor all
the current drivers using the page pool at once.

The netmem abstraction is currently a no-op. The page_pool uses
page_to_netmem() to convert allocated pages to netmem, and uses
netmem_to_page() to convert the netmem back to pages to pass to mm APIs,

Follow up patches to this series add non-paged netmem support to the
page_pool. This change is factored out on its own to limit the code
churn to this 1 patch, for ease of code review.

Signed-off-by: Mina Almasry <almasrymina@google.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://patch.msgid.link/20240628003253.1694510-6-almasrymina@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/skbuff_ref.h       |   4 +-
 include/net/netmem.h             |  15 ++
 include/net/page_pool/helpers.h  |  91 +++++++++---
 include/net/page_pool/types.h    |  14 +-
 include/trace/events/page_pool.h |  30 ++--
 net/bpf/test_run.c               |   5 +-
 net/core/page_pool.c             | 304 +++++++++++++++++++++------------------
 net/core/skbuff.c                |   8 +-
 8 files changed, 287 insertions(+), 184 deletions(-)

diff --git a/include/linux/skbuff_ref.h b/include/linux/skbuff_ref.h
index 11f0a4063403..16c241a23472 100644
--- a/include/linux/skbuff_ref.h
+++ b/include/linux/skbuff_ref.h
@@ -32,13 +32,13 @@ static inline void skb_frag_ref(struct sk_buff *skb, int f)
 	__skb_frag_ref(&skb_shinfo(skb)->frags[f]);
 }
 
-bool napi_pp_put_page(struct page *page);
+bool napi_pp_put_page(netmem_ref netmem);
 
 static inline void
 skb_page_unref(struct page *page, bool recycle)
 {
 #ifdef CONFIG_PAGE_POOL
-	if (recycle && napi_pp_put_page(page))
+	if (recycle && napi_pp_put_page(page_to_netmem(page)))
 		return;
 #endif
 	put_page(page);
diff --git a/include/net/netmem.h b/include/net/netmem.h
index d8b810245c1d..46cc9b89ac79 100644
--- a/include/net/netmem.h
+++ b/include/net/netmem.h
@@ -38,4 +38,19 @@ static inline netmem_ref page_to_netmem(struct page *page)
 	return (__force netmem_ref)page;
 }
 
+static inline int netmem_ref_count(netmem_ref netmem)
+{
+	return page_ref_count(netmem_to_page(netmem));
+}
+
+static inline unsigned long netmem_to_pfn(netmem_ref netmem)
+{
+	return page_to_pfn(netmem_to_page(netmem));
+}
+
+static inline netmem_ref netmem_compound_head(netmem_ref netmem)
+{
+	return page_to_netmem(compound_head(netmem_to_page(netmem)));
+}
+
 #endif /* _NET_NETMEM_H */
diff --git a/include/net/page_pool/helpers.h b/include/net/page_pool/helpers.h
index 873631c79ab1..2b43a893c619 100644
--- a/include/net/page_pool/helpers.h
+++ b/include/net/page_pool/helpers.h
@@ -55,6 +55,8 @@
 #include <linux/dma-mapping.h>
 
 #include <net/page_pool/types.h>
+#include <net/net_debug.h>
+#include <net/netmem.h>
 
 #ifdef CONFIG_PAGE_POOL_STATS
 /* Deprecated driver-facing API, use netlink instead */
@@ -212,6 +214,11 @@ page_pool_get_dma_dir(const struct page_pool *pool)
 	return pool->p.dma_dir;
 }
 
+static inline void page_pool_fragment_netmem(netmem_ref netmem, long nr)
+{
+	atomic_long_set(&netmem_to_page(netmem)->pp_ref_count, nr);
+}
+
 /**
  * page_pool_fragment_page() - split a fresh page into fragments
  * @page:	page to split
@@ -232,11 +239,12 @@ page_pool_get_dma_dir(const struct page_pool *pool)
  */
 static inline void page_pool_fragment_page(struct page *page, long nr)
 {
-	atomic_long_set(&page->pp_ref_count, nr);
+	page_pool_fragment_netmem(page_to_netmem(page), nr);
 }
 
-static inline long page_pool_unref_page(struct page *page, long nr)
+static inline long page_pool_unref_netmem(netmem_ref netmem, long nr)
 {
+	struct page *page = netmem_to_page(netmem);
 	long ret;
 
 	/* If nr == pp_ref_count then we have cleared all remaining
@@ -279,15 +287,41 @@ static inline long page_pool_unref_page(struct page *page, long nr)
 	return ret;
 }
 
+static inline long page_pool_unref_page(struct page *page, long nr)
+{
+	return page_pool_unref_netmem(page_to_netmem(page), nr);
+}
+
+static inline void page_pool_ref_netmem(netmem_ref netmem)
+{
+	atomic_long_inc(&netmem_to_page(netmem)->pp_ref_count);
+}
+
 static inline void page_pool_ref_page(struct page *page)
 {
-	atomic_long_inc(&page->pp_ref_count);
+	page_pool_ref_netmem(page_to_netmem(page));
 }
 
-static inline bool page_pool_is_last_ref(struct page *page)
+static inline bool page_pool_is_last_ref(netmem_ref netmem)
 {
 	/* If page_pool_unref_page() returns 0, we were the last user */
-	return page_pool_unref_page(page, 1) == 0;
+	return page_pool_unref_netmem(netmem, 1) == 0;
+}
+
+static inline void page_pool_put_netmem(struct page_pool *pool,
+					netmem_ref netmem,
+					unsigned int dma_sync_size,
+					bool allow_direct)
+{
+	/* When page_pool isn't compiled-in, net/core/xdp.c doesn't
+	 * allow registering MEM_TYPE_PAGE_POOL, but shield linker.
+	 */
+#ifdef CONFIG_PAGE_POOL
+	if (!page_pool_is_last_ref(netmem))
+		return;
+
+	page_pool_put_unrefed_netmem(pool, netmem, dma_sync_size, allow_direct);
+#endif
 }
 
 /**
@@ -308,15 +342,15 @@ static inline void page_pool_put_page(struct page_pool *pool,
 				      unsigned int dma_sync_size,
 				      bool allow_direct)
 {
-	/* When page_pool isn't compiled-in, net/core/xdp.c doesn't
-	 * allow registering MEM_TYPE_PAGE_POOL, but shield linker.
-	 */
-#ifdef CONFIG_PAGE_POOL
-	if (!page_pool_is_last_ref(page))
-		return;
+	page_pool_put_netmem(pool, page_to_netmem(page), dma_sync_size,
+			     allow_direct);
+}
 
-	page_pool_put_unrefed_page(pool, page, dma_sync_size, allow_direct);
-#endif
+static inline void page_pool_put_full_netmem(struct page_pool *pool,
+					     netmem_ref netmem,
+					     bool allow_direct)
+{
+	page_pool_put_netmem(pool, netmem, -1, allow_direct);
 }
 
 /**
@@ -331,7 +365,7 @@ static inline void page_pool_put_page(struct page_pool *pool,
 static inline void page_pool_put_full_page(struct page_pool *pool,
 					   struct page *page, bool allow_direct)
 {
-	page_pool_put_page(pool, page, -1, allow_direct);
+	page_pool_put_netmem(pool, page_to_netmem(page), -1, allow_direct);
 }
 
 /**
@@ -365,6 +399,18 @@ static inline void page_pool_free_va(struct page_pool *pool, void *va,
 	page_pool_put_page(pool, virt_to_head_page(va), -1, allow_direct);
 }
 
+static inline dma_addr_t page_pool_get_dma_addr_netmem(netmem_ref netmem)
+{
+	struct page *page = netmem_to_page(netmem);
+
+	dma_addr_t ret = page->dma_addr;
+
+	if (PAGE_POOL_32BIT_ARCH_WITH_64BIT_DMA)
+		ret <<= PAGE_SHIFT;
+
+	return ret;
+}
+
 /**
  * page_pool_get_dma_addr() - Retrieve the stored DMA address.
  * @page:	page allocated from a page pool
@@ -374,16 +420,14 @@ static inline void page_pool_free_va(struct page_pool *pool, void *va,
  */
 static inline dma_addr_t page_pool_get_dma_addr(const struct page *page)
 {
-	dma_addr_t ret = page->dma_addr;
-
-	if (PAGE_POOL_32BIT_ARCH_WITH_64BIT_DMA)
-		ret <<= PAGE_SHIFT;
-
-	return ret;
+	return page_pool_get_dma_addr_netmem(page_to_netmem((struct page *)page));
 }
 
-static inline bool page_pool_set_dma_addr(struct page *page, dma_addr_t addr)
+static inline bool page_pool_set_dma_addr_netmem(netmem_ref netmem,
+						 dma_addr_t addr)
 {
+	struct page *page = netmem_to_page(netmem);
+
 	if (PAGE_POOL_32BIT_ARCH_WITH_64BIT_DMA) {
 		page->dma_addr = addr >> PAGE_SHIFT;
 
@@ -419,6 +463,11 @@ static inline void page_pool_dma_sync_for_cpu(const struct page_pool *pool,
 				      page_pool_get_dma_dir(pool));
 }
 
+static inline bool page_pool_set_dma_addr(struct page *page, dma_addr_t addr)
+{
+	return page_pool_set_dma_addr_netmem(page_to_netmem(page), addr);
+}
+
 static inline bool page_pool_put(struct page_pool *pool)
 {
 	return refcount_dec_and_test(&pool->user_cnt);
diff --git a/include/net/page_pool/types.h b/include/net/page_pool/types.h
index 9093a964fc33..b70bcc14ceda 100644
--- a/include/net/page_pool/types.h
+++ b/include/net/page_pool/types.h
@@ -6,6 +6,7 @@
 #include <linux/dma-direction.h>
 #include <linux/ptr_ring.h>
 #include <linux/types.h>
+#include <net/netmem.h>
 
 #define PP_FLAG_DMA_MAP		BIT(0) /* Should page_pool do the DMA
 					* map/unmap
@@ -40,7 +41,7 @@
 #define PP_ALLOC_CACHE_REFILL	64
 struct pp_alloc_cache {
 	u32 count;
-	struct page *cache[PP_ALLOC_CACHE_SIZE];
+	netmem_ref cache[PP_ALLOC_CACHE_SIZE];
 };
 
 /**
@@ -73,7 +74,7 @@ struct page_pool_params {
 		struct net_device *netdev;
 		unsigned int	flags;
 /* private: used by test code only */
-		void (*init_callback)(struct page *page, void *arg);
+		void (*init_callback)(netmem_ref netmem, void *arg);
 		void *init_arg;
 	);
 };
@@ -151,7 +152,7 @@ struct page_pool {
 	 */
 	__cacheline_group_begin(frag) __aligned(4 * sizeof(long));
 	long frag_users;
-	struct page *frag_page;
+	netmem_ref frag_page;
 	unsigned int frag_offset;
 	__cacheline_group_end(frag);
 
@@ -220,8 +221,12 @@ struct page_pool {
 };
 
 struct page *page_pool_alloc_pages(struct page_pool *pool, gfp_t gfp);
+netmem_ref page_pool_alloc_netmem(struct page_pool *pool, gfp_t gfp);
 struct page *page_pool_alloc_frag(struct page_pool *pool, unsigned int *offset,
 				  unsigned int size, gfp_t gfp);
+netmem_ref page_pool_alloc_frag_netmem(struct page_pool *pool,
+				       unsigned int *offset, unsigned int size,
+				       gfp_t gfp);
 struct page_pool *page_pool_create(const struct page_pool_params *params);
 struct page_pool *page_pool_create_percpu(const struct page_pool_params *params,
 					  int cpuid);
@@ -252,6 +257,9 @@ static inline void page_pool_put_page_bulk(struct page_pool *pool, void **data,
 }
 #endif
 
+void page_pool_put_unrefed_netmem(struct page_pool *pool, netmem_ref netmem,
+				  unsigned int dma_sync_size,
+				  bool allow_direct);
 void page_pool_put_unrefed_page(struct page_pool *pool, struct page *page,
 				unsigned int dma_sync_size,
 				bool allow_direct);
diff --git a/include/trace/events/page_pool.h b/include/trace/events/page_pool.h
index 6834356b2d2a..543e54e432a1 100644
--- a/include/trace/events/page_pool.h
+++ b/include/trace/events/page_pool.h
@@ -42,51 +42,53 @@ TRACE_EVENT(page_pool_release,
 TRACE_EVENT(page_pool_state_release,
 
 	TP_PROTO(const struct page_pool *pool,
-		 const struct page *page, u32 release),
+		 netmem_ref netmem, u32 release),
 
-	TP_ARGS(pool, page, release),
+	TP_ARGS(pool, netmem, release),
 
 	TP_STRUCT__entry(
 		__field(const struct page_pool *,	pool)
-		__field(const struct page *,		page)
+		__field(unsigned long,			netmem)
 		__field(u32,				release)
 		__field(unsigned long,			pfn)
 	),
 
 	TP_fast_assign(
 		__entry->pool		= pool;
-		__entry->page		= page;
+		__entry->netmem		= (__force unsigned long)netmem;
 		__entry->release	= release;
-		__entry->pfn		= page_to_pfn(page);
+		__entry->pfn		= netmem_to_pfn(netmem);
 	),
 
-	TP_printk("page_pool=%p page=%p pfn=0x%lx release=%u",
-		  __entry->pool, __entry->page, __entry->pfn, __entry->release)
+	TP_printk("page_pool=%p netmem=%p pfn=0x%lx release=%u",
+		  __entry->pool, (void *)__entry->netmem,
+		  __entry->pfn, __entry->release)
 );
 
 TRACE_EVENT(page_pool_state_hold,
 
 	TP_PROTO(const struct page_pool *pool,
-		 const struct page *page, u32 hold),
+		 netmem_ref netmem, u32 hold),
 
-	TP_ARGS(pool, page, hold),
+	TP_ARGS(pool, netmem, hold),
 
 	TP_STRUCT__entry(
 		__field(const struct page_pool *,	pool)
-		__field(const struct page *,		page)
+		__field(unsigned long,			netmem)
 		__field(u32,				hold)
 		__field(unsigned long,			pfn)
 	),
 
 	TP_fast_assign(
 		__entry->pool	= pool;
-		__entry->page	= page;
+		__entry->netmem	= (__force unsigned long)netmem;
 		__entry->hold	= hold;
-		__entry->pfn	= page_to_pfn(page);
+		__entry->pfn	= netmem_to_pfn(netmem);
 	),
 
-	TP_printk("page_pool=%p page=%p pfn=0x%lx hold=%u",
-		  __entry->pool, __entry->page, __entry->pfn, __entry->hold)
+	TP_printk("page_pool=%p netmem=%p pfn=0x%lx hold=%u",
+		  __entry->pool, (void *)__entry->netmem,
+		  __entry->pfn, __entry->hold)
 );
 
 TRACE_EVENT(page_pool_update_nid,
diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index a6d7f790cdda..26417ab34ff4 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -127,9 +127,10 @@ struct xdp_test_data {
 #define TEST_XDP_FRAME_SIZE (PAGE_SIZE - sizeof(struct xdp_page_head))
 #define TEST_XDP_MAX_BATCH 256
 
-static void xdp_test_run_init_page(struct page *page, void *arg)
+static void xdp_test_run_init_page(netmem_ref netmem, void *arg)
 {
-	struct xdp_page_head *head = phys_to_virt(page_to_phys(page));
+	struct xdp_page_head *head =
+		phys_to_virt(page_to_phys(netmem_to_page(netmem)));
 	struct xdp_buff *new_ctx, *orig_ctx;
 	u32 headroom = XDP_PACKET_HEADROOM;
 	struct xdp_test_data *xdp = arg;
diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index 5f1ed6f2ca8f..dc52f078adde 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -327,19 +327,18 @@ struct page_pool *page_pool_create(const struct page_pool_params *params)
 }
 EXPORT_SYMBOL(page_pool_create);
 
-static void page_pool_return_page(struct page_pool *pool, struct page *page);
+static void page_pool_return_page(struct page_pool *pool, netmem_ref netmem);
 
-noinline
-static struct page *page_pool_refill_alloc_cache(struct page_pool *pool)
+static noinline netmem_ref page_pool_refill_alloc_cache(struct page_pool *pool)
 {
 	struct ptr_ring *r = &pool->ring;
-	struct page *page;
+	netmem_ref netmem;
 	int pref_nid; /* preferred NUMA node */
 
 	/* Quicker fallback, avoid locks when ring is empty */
 	if (__ptr_ring_empty(r)) {
 		alloc_stat_inc(pool, empty);
-		return NULL;
+		return 0;
 	}
 
 	/* Softirq guarantee CPU and thus NUMA node is stable. This,
@@ -354,57 +353,57 @@ static struct page *page_pool_refill_alloc_cache(struct page_pool *pool)
 
 	/* Refill alloc array, but only if NUMA match */
 	do {
-		page = __ptr_ring_consume(r);
-		if (unlikely(!page))
+		netmem = (__force netmem_ref)__ptr_ring_consume(r);
+		if (unlikely(!netmem))
 			break;
 
-		if (likely(page_to_nid(page) == pref_nid)) {
-			pool->alloc.cache[pool->alloc.count++] = page;
+		if (likely(page_to_nid(netmem_to_page(netmem)) == pref_nid)) {
+			pool->alloc.cache[pool->alloc.count++] = netmem;
 		} else {
 			/* NUMA mismatch;
 			 * (1) release 1 page to page-allocator and
 			 * (2) break out to fallthrough to alloc_pages_node.
 			 * This limit stress on page buddy alloactor.
 			 */
-			page_pool_return_page(pool, page);
+			page_pool_return_page(pool, netmem);
 			alloc_stat_inc(pool, waive);
-			page = NULL;
+			netmem = 0;
 			break;
 		}
 	} while (pool->alloc.count < PP_ALLOC_CACHE_REFILL);
 
 	/* Return last page */
 	if (likely(pool->alloc.count > 0)) {
-		page = pool->alloc.cache[--pool->alloc.count];
+		netmem = pool->alloc.cache[--pool->alloc.count];
 		alloc_stat_inc(pool, refill);
 	}
 
-	return page;
+	return netmem;
 }
 
 /* fast path */
-static struct page *__page_pool_get_cached(struct page_pool *pool)
+static netmem_ref __page_pool_get_cached(struct page_pool *pool)
 {
-	struct page *page;
+	netmem_ref netmem;
 
 	/* Caller MUST guarantee safe non-concurrent access, e.g. softirq */
 	if (likely(pool->alloc.count)) {
 		/* Fast-path */
-		page = pool->alloc.cache[--pool->alloc.count];
+		netmem = pool->alloc.cache[--pool->alloc.count];
 		alloc_stat_inc(pool, fast);
 	} else {
-		page = page_pool_refill_alloc_cache(pool);
+		netmem = page_pool_refill_alloc_cache(pool);
 	}
 
-	return page;
+	return netmem;
 }
 
 static void __page_pool_dma_sync_for_device(const struct page_pool *pool,
-					    const struct page *page,
+					    netmem_ref netmem,
 					    u32 dma_sync_size)
 {
 #if defined(CONFIG_HAS_DMA) && defined(CONFIG_DMA_NEED_SYNC)
-	dma_addr_t dma_addr = page_pool_get_dma_addr(page);
+	dma_addr_t dma_addr = page_pool_get_dma_addr_netmem(netmem);
 
 	dma_sync_size = min(dma_sync_size, pool->p.max_len);
 	__dma_sync_single_for_device(pool->p.dev, dma_addr + pool->p.offset,
@@ -414,14 +413,14 @@ static void __page_pool_dma_sync_for_device(const struct page_pool *pool,
 
 static __always_inline void
 page_pool_dma_sync_for_device(const struct page_pool *pool,
-			      const struct page *page,
+			      netmem_ref netmem,
 			      u32 dma_sync_size)
 {
 	if (pool->dma_sync && dma_dev_need_sync(pool->p.dev))
-		__page_pool_dma_sync_for_device(pool, page, dma_sync_size);
+		__page_pool_dma_sync_for_device(pool, netmem, dma_sync_size);
 }
 
-static bool page_pool_dma_map(struct page_pool *pool, struct page *page)
+static bool page_pool_dma_map(struct page_pool *pool, netmem_ref netmem)
 {
 	dma_addr_t dma;
 
@@ -430,17 +429,17 @@ static bool page_pool_dma_map(struct page_pool *pool, struct page *page)
 	 * into page private data (i.e 32bit cpu with 64bit DMA caps)
 	 * This mapping is kept for lifetime of page, until leaving pool.
 	 */
-	dma = dma_map_page_attrs(pool->p.dev, page, 0,
-				 (PAGE_SIZE << pool->p.order),
-				 pool->p.dma_dir, DMA_ATTR_SKIP_CPU_SYNC |
-						  DMA_ATTR_WEAK_ORDERING);
+	dma = dma_map_page_attrs(pool->p.dev, netmem_to_page(netmem), 0,
+				 (PAGE_SIZE << pool->p.order), pool->p.dma_dir,
+				 DMA_ATTR_SKIP_CPU_SYNC |
+					 DMA_ATTR_WEAK_ORDERING);
 	if (dma_mapping_error(pool->p.dev, dma))
 		return false;
 
-	if (page_pool_set_dma_addr(page, dma))
+	if (page_pool_set_dma_addr_netmem(netmem, dma))
 		goto unmap_failed;
 
-	page_pool_dma_sync_for_device(pool, page, pool->p.max_len);
+	page_pool_dma_sync_for_device(pool, netmem, pool->p.max_len);
 
 	return true;
 
@@ -452,9 +451,10 @@ unmap_failed:
 	return false;
 }
 
-static void page_pool_set_pp_info(struct page_pool *pool,
-				  struct page *page)
+static void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem)
 {
+	struct page *page = netmem_to_page(netmem);
+
 	page->pp = pool;
 	page->pp_magic |= PP_SIGNATURE;
 
@@ -464,13 +464,15 @@ static void page_pool_set_pp_info(struct page_pool *pool,
 	 * is dirtying the same cache line as the page->pp_magic above, so
 	 * the overhead is negligible.
 	 */
-	page_pool_fragment_page(page, 1);
+	page_pool_fragment_netmem(netmem, 1);
 	if (pool->has_init_callback)
-		pool->slow.init_callback(page, pool->slow.init_arg);
+		pool->slow.init_callback(netmem, pool->slow.init_arg);
 }
 
-static void page_pool_clear_pp_info(struct page *page)
+static void page_pool_clear_pp_info(netmem_ref netmem)
 {
+	struct page *page = netmem_to_page(netmem);
+
 	page->pp_magic = 0;
 	page->pp = NULL;
 }
@@ -485,34 +487,34 @@ static struct page *__page_pool_alloc_page_order(struct page_pool *pool,
 	if (unlikely(!page))
 		return NULL;
 
-	if (pool->dma_map && unlikely(!page_pool_dma_map(pool, page))) {
+	if (pool->dma_map && unlikely(!page_pool_dma_map(pool, page_to_netmem(page)))) {
 		put_page(page);
 		return NULL;
 	}
 
 	alloc_stat_inc(pool, slow_high_order);
-	page_pool_set_pp_info(pool, page);
+	page_pool_set_pp_info(pool, page_to_netmem(page));
 
 	/* Track how many pages are held 'in-flight' */
 	pool->pages_state_hold_cnt++;
-	trace_page_pool_state_hold(pool, page, pool->pages_state_hold_cnt);
+	trace_page_pool_state_hold(pool, page_to_netmem(page),
+				   pool->pages_state_hold_cnt);
 	return page;
 }
 
 /* slow path */
-noinline
-static struct page *__page_pool_alloc_pages_slow(struct page_pool *pool,
-						 gfp_t gfp)
+static noinline netmem_ref __page_pool_alloc_pages_slow(struct page_pool *pool,
+							gfp_t gfp)
 {
 	const int bulk = PP_ALLOC_CACHE_REFILL;
 	unsigned int pp_order = pool->p.order;
 	bool dma_map = pool->dma_map;
-	struct page *page;
+	netmem_ref netmem;
 	int i, nr_pages;
 
 	/* Don't support bulk alloc for high-order pages */
 	if (unlikely(pp_order))
-		return __page_pool_alloc_page_order(pool, gfp);
+		return page_to_netmem(__page_pool_alloc_page_order(pool, gfp));
 
 	/* Unnecessary as alloc cache is empty, but guarantees zero count */
 	if (unlikely(pool->alloc.count > 0))
@@ -521,56 +523,63 @@ static struct page *__page_pool_alloc_pages_slow(struct page_pool *pool,
 	/* Mark empty alloc.cache slots "empty" for alloc_pages_bulk_array */
 	memset(&pool->alloc.cache, 0, sizeof(void *) * bulk);
 
-	nr_pages = alloc_pages_bulk_array_node(gfp, pool->p.nid, bulk,
-					       pool->alloc.cache);
+	nr_pages = alloc_pages_bulk_array_node(gfp,
+					       pool->p.nid, bulk,
+					       (struct page **)pool->alloc.cache);
 	if (unlikely(!nr_pages))
-		return NULL;
+		return 0;
 
 	/* Pages have been filled into alloc.cache array, but count is zero and
 	 * page element have not been (possibly) DMA mapped.
 	 */
 	for (i = 0; i < nr_pages; i++) {
-		page = pool->alloc.cache[i];
-		if (dma_map && unlikely(!page_pool_dma_map(pool, page))) {
-			put_page(page);
+		netmem = pool->alloc.cache[i];
+		if (dma_map && unlikely(!page_pool_dma_map(pool, netmem))) {
+			put_page(netmem_to_page(netmem));
 			continue;
 		}
 
-		page_pool_set_pp_info(pool, page);
-		pool->alloc.cache[pool->alloc.count++] = page;
+		page_pool_set_pp_info(pool, netmem);
+		pool->alloc.cache[pool->alloc.count++] = netmem;
 		/* Track how many pages are held 'in-flight' */
 		pool->pages_state_hold_cnt++;
-		trace_page_pool_state_hold(pool, page,
+		trace_page_pool_state_hold(pool, netmem,
 					   pool->pages_state_hold_cnt);
 	}
 
 	/* Return last page */
 	if (likely(pool->alloc.count > 0)) {
-		page = pool->alloc.cache[--pool->alloc.count];
+		netmem = pool->alloc.cache[--pool->alloc.count];
 		alloc_stat_inc(pool, slow);
 	} else {
-		page = NULL;
+		netmem = 0;
 	}
 
 	/* When page just alloc'ed is should/must have refcnt 1. */
-	return page;
+	return netmem;
 }
 
 /* For using page_pool replace: alloc_pages() API calls, but provide
  * synchronization guarantee for allocation side.
  */
-struct page *page_pool_alloc_pages(struct page_pool *pool, gfp_t gfp)
+netmem_ref page_pool_alloc_netmem(struct page_pool *pool, gfp_t gfp)
 {
-	struct page *page;
+	netmem_ref netmem;
 
 	/* Fast-path: Get a page from cache */
-	page = __page_pool_get_cached(pool);
-	if (page)
-		return page;
+	netmem = __page_pool_get_cached(pool);
+	if (netmem)
+		return netmem;
 
 	/* Slow-path: cache empty, do real allocation */
-	page = __page_pool_alloc_pages_slow(pool, gfp);
-	return page;
+	netmem = __page_pool_alloc_pages_slow(pool, gfp);
+	return netmem;
+}
+EXPORT_SYMBOL(page_pool_alloc_netmem);
+
+struct page *page_pool_alloc_pages(struct page_pool *pool, gfp_t gfp)
+{
+	return netmem_to_page(page_pool_alloc_netmem(pool, gfp));
 }
 EXPORT_SYMBOL(page_pool_alloc_pages);
 ALLOW_ERROR_INJECTION(page_pool_alloc_pages, NULL);
@@ -599,8 +608,8 @@ s32 page_pool_inflight(const struct page_pool *pool, bool strict)
 	return inflight;
 }
 
-static __always_inline
-void __page_pool_release_page_dma(struct page_pool *pool, struct page *page)
+static __always_inline void __page_pool_release_page_dma(struct page_pool *pool,
+							 netmem_ref netmem)
 {
 	dma_addr_t dma;
 
@@ -610,13 +619,13 @@ void __page_pool_release_page_dma(struct page_pool *pool, struct page *page)
 		 */
 		return;
 
-	dma = page_pool_get_dma_addr(page);
+	dma = page_pool_get_dma_addr_netmem(netmem);
 
 	/* When page is unmapped, it cannot be returned to our pool */
 	dma_unmap_page_attrs(pool->p.dev, dma,
 			     PAGE_SIZE << pool->p.order, pool->p.dma_dir,
 			     DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING);
-	page_pool_set_dma_addr(page, 0);
+	page_pool_set_dma_addr_netmem(netmem, 0);
 }
 
 /* Disconnects a page (from a page_pool).  API users can have a need
@@ -624,35 +633,34 @@ void __page_pool_release_page_dma(struct page_pool *pool, struct page *page)
  * a regular page (that will eventually be returned to the normal
  * page-allocator via put_page).
  */
-void page_pool_return_page(struct page_pool *pool, struct page *page)
+void page_pool_return_page(struct page_pool *pool, netmem_ref netmem)
 {
 	int count;
 
-	__page_pool_release_page_dma(pool, page);
-
-	page_pool_clear_pp_info(page);
+	__page_pool_release_page_dma(pool, netmem);
 
 	/* This may be the last page returned, releasing the pool, so
 	 * it is not safe to reference pool afterwards.
 	 */
 	count = atomic_inc_return_relaxed(&pool->pages_state_release_cnt);
-	trace_page_pool_state_release(pool, page, count);
+	trace_page_pool_state_release(pool, netmem, count);
 
-	put_page(page);
+	page_pool_clear_pp_info(netmem);
+	put_page(netmem_to_page(netmem));
 	/* An optimization would be to call __free_pages(page, pool->p.order)
 	 * knowing page is not part of page-cache (thus avoiding a
 	 * __page_cache_release() call).
 	 */
 }
 
-static bool page_pool_recycle_in_ring(struct page_pool *pool, struct page *page)
+static bool page_pool_recycle_in_ring(struct page_pool *pool, netmem_ref netmem)
 {
 	int ret;
 	/* BH protection not needed if current is softirq */
 	if (in_softirq())
-		ret = ptr_ring_produce(&pool->ring, page);
+		ret = ptr_ring_produce(&pool->ring, (__force void *)netmem);
 	else
-		ret = ptr_ring_produce_bh(&pool->ring, page);
+		ret = ptr_ring_produce_bh(&pool->ring, (__force void *)netmem);
 
 	if (!ret) {
 		recycle_stat_inc(pool, ring);
@@ -667,7 +675,7 @@ static bool page_pool_recycle_in_ring(struct page_pool *pool, struct page *page)
  *
  * Caller must provide appropriate safe context.
  */
-static bool page_pool_recycle_in_cache(struct page *page,
+static bool page_pool_recycle_in_cache(netmem_ref netmem,
 				       struct page_pool *pool)
 {
 	if (unlikely(pool->alloc.count == PP_ALLOC_CACHE_SIZE)) {
@@ -676,14 +684,15 @@ static bool page_pool_recycle_in_cache(struct page *page,
 	}
 
 	/* Caller MUST have verified/know (page_ref_count(page) == 1) */
-	pool->alloc.cache[pool->alloc.count++] = page;
+	pool->alloc.cache[pool->alloc.count++] = netmem;
 	recycle_stat_inc(pool, cached);
 	return true;
 }
 
-static bool __page_pool_page_can_be_recycled(const struct page *page)
+static bool __page_pool_page_can_be_recycled(netmem_ref netmem)
 {
-	return page_ref_count(page) == 1 && !page_is_pfmemalloc(page);
+	return page_ref_count(netmem_to_page(netmem)) == 1 &&
+	       !page_is_pfmemalloc(netmem_to_page(netmem));
 }
 
 /* If the page refcnt == 1, this will try to recycle the page.
@@ -692,8 +701,8 @@ static bool __page_pool_page_can_be_recycled(const struct page *page)
  * If the page refcnt != 1, then the page will be returned to memory
  * subsystem.
  */
-static __always_inline struct page *
-__page_pool_put_page(struct page_pool *pool, struct page *page,
+static __always_inline netmem_ref
+__page_pool_put_page(struct page_pool *pool, netmem_ref netmem,
 		     unsigned int dma_sync_size, bool allow_direct)
 {
 	lockdep_assert_no_hardirq();
@@ -707,16 +716,16 @@ __page_pool_put_page(struct page_pool *pool, struct page *page,
 	 * page is NOT reusable when allocated when system is under
 	 * some pressure. (page_is_pfmemalloc)
 	 */
-	if (likely(__page_pool_page_can_be_recycled(page))) {
+	if (likely(__page_pool_page_can_be_recycled(netmem))) {
 		/* Read barrier done in page_ref_count / READ_ONCE */
 
-		page_pool_dma_sync_for_device(pool, page, dma_sync_size);
+		page_pool_dma_sync_for_device(pool, netmem, dma_sync_size);
 
-		if (allow_direct && page_pool_recycle_in_cache(page, pool))
-			return NULL;
+		if (allow_direct && page_pool_recycle_in_cache(netmem, pool))
+			return 0;
 
 		/* Page found as candidate for recycling */
-		return page;
+		return netmem;
 	}
 	/* Fallback/non-XDP mode: API user have elevated refcnt.
 	 *
@@ -732,9 +741,9 @@ __page_pool_put_page(struct page_pool *pool, struct page *page,
 	 * will be invoking put_page.
 	 */
 	recycle_stat_inc(pool, released_refcnt);
-	page_pool_return_page(pool, page);
+	page_pool_return_page(pool, netmem);
 
-	return NULL;
+	return 0;
 }
 
 static bool page_pool_napi_local(const struct page_pool *pool)
@@ -760,19 +769,28 @@ static bool page_pool_napi_local(const struct page_pool *pool)
 	return napi && READ_ONCE(napi->list_owner) == cpuid;
 }
 
-void page_pool_put_unrefed_page(struct page_pool *pool, struct page *page,
-				unsigned int dma_sync_size, bool allow_direct)
+void page_pool_put_unrefed_netmem(struct page_pool *pool, netmem_ref netmem,
+				  unsigned int dma_sync_size, bool allow_direct)
 {
 	if (!allow_direct)
 		allow_direct = page_pool_napi_local(pool);
 
-	page = __page_pool_put_page(pool, page, dma_sync_size, allow_direct);
-	if (page && !page_pool_recycle_in_ring(pool, page)) {
+	netmem =
+		__page_pool_put_page(pool, netmem, dma_sync_size, allow_direct);
+	if (netmem && !page_pool_recycle_in_ring(pool, netmem)) {
 		/* Cache full, fallback to free pages */
 		recycle_stat_inc(pool, ring_full);
-		page_pool_return_page(pool, page);
+		page_pool_return_page(pool, netmem);
 	}
 }
+EXPORT_SYMBOL(page_pool_put_unrefed_netmem);
+
+void page_pool_put_unrefed_page(struct page_pool *pool, struct page *page,
+				unsigned int dma_sync_size, bool allow_direct)
+{
+	page_pool_put_unrefed_netmem(pool, page_to_netmem(page), dma_sync_size,
+				     allow_direct);
+}
 EXPORT_SYMBOL(page_pool_put_unrefed_page);
 
 /**
@@ -800,16 +818,16 @@ void page_pool_put_page_bulk(struct page_pool *pool, void **data,
 	allow_direct = page_pool_napi_local(pool);
 
 	for (i = 0; i < count; i++) {
-		struct page *page = virt_to_head_page(data[i]);
+		netmem_ref netmem = page_to_netmem(virt_to_head_page(data[i]));
 
 		/* It is not the last user for the page frag case */
-		if (!page_pool_is_last_ref(page))
+		if (!page_pool_is_last_ref(netmem))
 			continue;
 
-		page = __page_pool_put_page(pool, page, -1, allow_direct);
+		netmem = __page_pool_put_page(pool, netmem, -1, allow_direct);
 		/* Approved for bulk recycling in ptr_ring cache */
-		if (page)
-			data[bulk_len++] = page;
+		if (netmem)
+			data[bulk_len++] = (__force void *)netmem;
 	}
 
 	if (!bulk_len)
@@ -835,98 +853,106 @@ void page_pool_put_page_bulk(struct page_pool *pool, void **data,
 	 * since put_page() with refcnt == 1 can be an expensive operation
 	 */
 	for (; i < bulk_len; i++)
-		page_pool_return_page(pool, data[i]);
+		page_pool_return_page(pool, (__force netmem_ref)data[i]);
 }
 EXPORT_SYMBOL(page_pool_put_page_bulk);
 
-static struct page *page_pool_drain_frag(struct page_pool *pool,
-					 struct page *page)
+static netmem_ref page_pool_drain_frag(struct page_pool *pool,
+				       netmem_ref netmem)
 {
 	long drain_count = BIAS_MAX - pool->frag_users;
 
 	/* Some user is still using the page frag */
-	if (likely(page_pool_unref_page(page, drain_count)))
-		return NULL;
+	if (likely(page_pool_unref_netmem(netmem, drain_count)))
+		return 0;
 
-	if (__page_pool_page_can_be_recycled(page)) {
-		page_pool_dma_sync_for_device(pool, page, -1);
-		return page;
+	if (__page_pool_page_can_be_recycled(netmem)) {
+		page_pool_dma_sync_for_device(pool, netmem, -1);
+		return netmem;
 	}
 
-	page_pool_return_page(pool, page);
-	return NULL;
+	page_pool_return_page(pool, netmem);
+	return 0;
 }
 
 static void page_pool_free_frag(struct page_pool *pool)
 {
 	long drain_count = BIAS_MAX - pool->frag_users;
-	struct page *page = pool->frag_page;
+	netmem_ref netmem = pool->frag_page;
 
-	pool->frag_page = NULL;
+	pool->frag_page = 0;
 
-	if (!page || page_pool_unref_page(page, drain_count))
+	if (!netmem || page_pool_unref_netmem(netmem, drain_count))
 		return;
 
-	page_pool_return_page(pool, page);
+	page_pool_return_page(pool, netmem);
 }
 
-struct page *page_pool_alloc_frag(struct page_pool *pool,
-				  unsigned int *offset,
-				  unsigned int size, gfp_t gfp)
+netmem_ref page_pool_alloc_frag_netmem(struct page_pool *pool,
+				       unsigned int *offset, unsigned int size,
+				       gfp_t gfp)
 {
 	unsigned int max_size = PAGE_SIZE << pool->p.order;
-	struct page *page = pool->frag_page;
+	netmem_ref netmem = pool->frag_page;
 
 	if (WARN_ON(size > max_size))
-		return NULL;
+		return 0;
 
 	size = ALIGN(size, dma_get_cache_alignment());
 	*offset = pool->frag_offset;
 
-	if (page && *offset + size > max_size) {
-		page = page_pool_drain_frag(pool, page);
-		if (page) {
+	if (netmem && *offset + size > max_size) {
+		netmem = page_pool_drain_frag(pool, netmem);
+		if (netmem) {
 			alloc_stat_inc(pool, fast);
 			goto frag_reset;
 		}
 	}
 
-	if (!page) {
-		page = page_pool_alloc_pages(pool, gfp);
-		if (unlikely(!page)) {
-			pool->frag_page = NULL;
-			return NULL;
+	if (!netmem) {
+		netmem = page_pool_alloc_netmem(pool, gfp);
+		if (unlikely(!netmem)) {
+			pool->frag_page = 0;
+			return 0;
 		}
 
-		pool->frag_page = page;
+		pool->frag_page = netmem;
 
 frag_reset:
 		pool->frag_users = 1;
 		*offset = 0;
 		pool->frag_offset = size;
-		page_pool_fragment_page(page, BIAS_MAX);
-		return page;
+		page_pool_fragment_netmem(netmem, BIAS_MAX);
+		return netmem;
 	}
 
 	pool->frag_users++;
 	pool->frag_offset = *offset + size;
 	alloc_stat_inc(pool, fast);
-	return page;
+	return netmem;
+}
+EXPORT_SYMBOL(page_pool_alloc_frag_netmem);
+
+struct page *page_pool_alloc_frag(struct page_pool *pool, unsigned int *offset,
+				  unsigned int size, gfp_t gfp)
+{
+	return netmem_to_page(page_pool_alloc_frag_netmem(pool, offset, size,
+							  gfp));
 }
 EXPORT_SYMBOL(page_pool_alloc_frag);
 
 static void page_pool_empty_ring(struct page_pool *pool)
 {
-	struct page *page;
+	netmem_ref netmem;
 
 	/* Empty recycle ring */
-	while ((page = ptr_ring_consume_bh(&pool->ring))) {
+	while ((netmem = (__force netmem_ref)ptr_ring_consume_bh(&pool->ring))) {
 		/* Verify the refcnt invariant of cached pages */
-		if (!(page_ref_count(page) == 1))
+		if (!(page_ref_count(netmem_to_page(netmem)) == 1))
 			pr_crit("%s() page_pool refcnt %d violation\n",
-				__func__, page_ref_count(page));
+				__func__, netmem_ref_count(netmem));
 
-		page_pool_return_page(pool, page);
+		page_pool_return_page(pool, netmem);
 	}
 }
 
@@ -942,7 +968,7 @@ static void __page_pool_destroy(struct page_pool *pool)
 
 static void page_pool_empty_alloc_cache_once(struct page_pool *pool)
 {
-	struct page *page;
+	netmem_ref netmem;
 
 	if (pool->destroy_cnt)
 		return;
@@ -952,8 +978,8 @@ static void page_pool_empty_alloc_cache_once(struct page_pool *pool)
 	 * call concurrently.
 	 */
 	while (pool->alloc.count) {
-		page = pool->alloc.cache[--pool->alloc.count];
-		page_pool_return_page(pool, page);
+		netmem = pool->alloc.cache[--pool->alloc.count];
+		page_pool_return_page(pool, netmem);
 	}
 }
 
@@ -1060,15 +1086,15 @@ EXPORT_SYMBOL(page_pool_destroy);
 /* Caller must provide appropriate safe context, e.g. NAPI. */
 void page_pool_update_nid(struct page_pool *pool, int new_nid)
 {
-	struct page *page;
+	netmem_ref netmem;
 
 	trace_page_pool_update_nid(pool, new_nid);
 	pool->p.nid = new_nid;
 
 	/* Flush pool alloc cache, as refill will check NUMA node */
 	while (pool->alloc.count) {
-		page = pool->alloc.cache[--pool->alloc.count];
-		page_pool_return_page(pool, page);
+		netmem = pool->alloc.cache[--pool->alloc.count];
+		page_pool_return_page(pool, netmem);
 	}
 }
 EXPORT_SYMBOL(page_pool_update_nid);
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 0ed4d00d258c..83f8cd8aa2d1 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1015,8 +1015,10 @@ int skb_cow_data_for_xdp(struct page_pool *pool, struct sk_buff **pskb,
 EXPORT_SYMBOL(skb_cow_data_for_xdp);
 
 #if IS_ENABLED(CONFIG_PAGE_POOL)
-bool napi_pp_put_page(struct page *page)
+bool napi_pp_put_page(netmem_ref netmem)
 {
+	struct page *page = netmem_to_page(netmem);
+
 	page = compound_head(page);
 
 	/* page->pp_magic is OR'ed with PP_SIGNATURE after the allocation
@@ -1029,7 +1031,7 @@ bool napi_pp_put_page(struct page *page)
 	if (unlikely(!is_pp_page(page)))
 		return false;
 
-	page_pool_put_full_page(page->pp, page, false);
+	page_pool_put_full_netmem(page->pp, page_to_netmem(page), false);
 
 	return true;
 }
@@ -1040,7 +1042,7 @@ static bool skb_pp_recycle(struct sk_buff *skb, void *data)
 {
 	if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle)
 		return false;
-	return napi_pp_put_page(virt_to_page(data));
+	return napi_pp_put_page(page_to_netmem(virt_to_page(data)));
 }
 
 /**
-- 
cgit v1.2.3-58-ga151


From 07c3cc51a0853986af0fc52d0544b1b67b391ea8 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Fri, 28 Jun 2024 00:32:50 +0000
Subject: tools: net: package libynl for use in selftests

Support building the C YNL userspace library into one big static file.
We can then link selftests against it for easy to use C netlink
interface.

Signed-off-by: Mina Almasry <almasrymina@google.com>
Link: https://patch.msgid.link/20240628003253.1694510-14-almasrymina@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/Makefile             |  6 +++++-
 tools/net/ynl/lib/Makefile         |  4 +++-
 tools/testing/selftests/net/ynl.mk | 21 +++++++++++++++++++++
 3 files changed, 29 insertions(+), 2 deletions(-)
 create mode 100644 tools/testing/selftests/net/ynl.mk

diff --git a/tools/net/ynl/Makefile b/tools/net/ynl/Makefile
index 8e9e09d84e26..d1cdf2a8f826 100644
--- a/tools/net/ynl/Makefile
+++ b/tools/net/ynl/Makefile
@@ -2,9 +2,12 @@
 
 SUBDIRS = lib generated samples
 
-all: $(SUBDIRS)
+all: $(SUBDIRS) libynl.a
 
 samples: | lib generated
+libynl.a: | lib generated
+	@echo -e "\tAR $@"
+	@ar rcs $@ lib/ynl.o generated/*-user.o
 
 $(SUBDIRS):
 	@if [ -f "$@/Makefile" ] ; then \
@@ -17,5 +20,6 @@ clean distclean:
 			$(MAKE) -C $$dir $@; \
 		fi \
 	done
+	rm -f libynl.a
 
 .PHONY: all clean distclean $(SUBDIRS)
diff --git a/tools/net/ynl/lib/Makefile b/tools/net/ynl/lib/Makefile
index dfff3ecd1cba..2887cc5de530 100644
--- a/tools/net/ynl/lib/Makefile
+++ b/tools/net/ynl/lib/Makefile
@@ -14,7 +14,9 @@ include $(wildcard *.d)
 all: ynl.a
 
 ynl.a: $(OBJS)
-	ar rcs $@ $(OBJS)
+	@echo -e "\tAR $@"
+	@ar rcs $@ $(OBJS)
+
 clean:
 	rm -f *.o *.d *~
 	rm -rf __pycache__
diff --git a/tools/testing/selftests/net/ynl.mk b/tools/testing/selftests/net/ynl.mk
new file mode 100644
index 000000000000..59cb26cf3f73
--- /dev/null
+++ b/tools/testing/selftests/net/ynl.mk
@@ -0,0 +1,21 @@
+# SPDX-License-Identifier: GPL-2.0
+
+# YNL selftest build snippet
+
+# Inputs:
+#
+# YNL_GENS:      families we need in the selftests
+# YNL_PROGS:     TEST_PROGS which need YNL (TODO, none exist, yet)
+# YNL_GEN_FILES: TEST_GEN_FILES which need YNL
+
+YNL_OUTPUTS := $(patsubst %,$(OUTPUT)/%,$(YNL_GEN_FILES))
+
+$(YNL_OUTPUTS): $(OUTPUT)/libynl.a
+$(YNL_OUTPUTS): CFLAGS += \
+	-I$(top_srcdir)/usr/include/ $(KHDR_INCLUDES) \
+	-I$(top_srcdir)/tools/net/ynl/lib/ \
+	-I$(top_srcdir)/tools/net/ynl/generated/
+
+$(OUTPUT)/libynl.a:
+	$(Q)$(MAKE) -C $(top_srcdir)/tools/net/ynl GENS="$(YNL_GENS)" libynl.a
+	$(Q)cp $(top_srcdir)/tools/net/ynl/libynl.a $(OUTPUT)/libynl.a
-- 
cgit v1.2.3-58-ga151


From 8d7330b3a9c6db53d4462820df566c0b1e77d831 Mon Sep 17 00:00:00 2001
From: Lucas Stach <l.stach@pengutronix.de>
Date: Mon, 1 Jul 2024 10:53:41 +0200
Subject: net: dsa: microchip: lan9371/2: add 100BaseTX PHY support

On the LAN9371 and LAN9372, the 4th internal PHY is a 100BaseTX PHY
instead of a 100BaseT1 PHY. The 100BaseTX PHYs have a different base
register offset.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Acked-by: Arun Ramadoss <arun.ramadoss@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/microchip/ksz_common.h   | 1 +
 drivers/net/dsa/microchip/lan937x_main.c | 4 ++++
 drivers/net/dsa/microchip/lan937x_reg.h  | 1 +
 3 files changed, 6 insertions(+)

diff --git a/drivers/net/dsa/microchip/ksz_common.h b/drivers/net/dsa/microchip/ksz_common.h
index ee7db46e469d..c4a4664c0385 100644
--- a/drivers/net/dsa/microchip/ksz_common.h
+++ b/drivers/net/dsa/microchip/ksz_common.h
@@ -22,6 +22,7 @@
 /* all KSZ switches count ports from 1 */
 #define KSZ_PORT_1 0
 #define KSZ_PORT_2 1
+#define KSZ_PORT_4 3
 
 struct ksz_device;
 struct ksz_port;
diff --git a/drivers/net/dsa/microchip/lan937x_main.c b/drivers/net/dsa/microchip/lan937x_main.c
index b479a628b1ae..eaa862eb6b26 100644
--- a/drivers/net/dsa/microchip/lan937x_main.c
+++ b/drivers/net/dsa/microchip/lan937x_main.c
@@ -55,6 +55,10 @@ static int lan937x_vphy_ind_addr_wr(struct ksz_device *dev, int addr, int reg)
 	u16 addr_base = REG_PORT_T1_PHY_CTRL_BASE;
 	u16 temp;
 
+	if ((dev->info->chip_id == LAN9371_CHIP_ID ||
+	     dev->info->chip_id == LAN9372_CHIP_ID) && addr == KSZ_PORT_4)
+		addr_base = REG_PORT_TX_PHY_CTRL_BASE;
+
 	/* get register address based on the logical port */
 	temp = PORT_CTRL_ADDR(addr, (addr_base + (reg << 2)));
 
diff --git a/drivers/net/dsa/microchip/lan937x_reg.h b/drivers/net/dsa/microchip/lan937x_reg.h
index 45b606b6429f..7ecada924023 100644
--- a/drivers/net/dsa/microchip/lan937x_reg.h
+++ b/drivers/net/dsa/microchip/lan937x_reg.h
@@ -147,6 +147,7 @@
 
 /* 1 - Phy */
 #define REG_PORT_T1_PHY_CTRL_BASE	0x0100
+#define REG_PORT_TX_PHY_CTRL_BASE	0x0280
 
 /* 3 - xMII */
 #define PORT_SGMII_SEL			BIT(7)
-- 
cgit v1.2.3-58-ga151


From c3db39468a42fec7e83add335765f8d81e1864c3 Mon Sep 17 00:00:00 2001
From: Lucas Stach <l.stach@pengutronix.de>
Date: Mon, 1 Jul 2024 10:53:42 +0200
Subject: net: dsa: microchip: lan937x: disable in-band status support for
 RGMII interfaces

This driver do not support in-band mode and in case of CPU<->Switch
link, this mode is not working any way. So, disable it otherwise ingress
path of the switch MAC will stay disabled.

Note: lan9372 manual do not document 0xN301 BIT(2) for the RGMII mode
and recommend[1] to disable in-band link status update for the RGMII RX
path by clearing 0xN302 BIT(0). But, 0xN301 BIT(2) seems to work too, so
keep it unified with other KSZ switches.

[1] https://microchip.my.site.com/s/article/LAN937X-The-required-configuration-for-the-external-MAC-port-to-operate-at-RGMII-to-RGMII-1Gbps-link-speed

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Acked-by: Arun Ramadoss <arun.ramadoss@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/microchip/ksz_common.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c
index baa1eeb9a1b0..b074b4bb0629 100644
--- a/drivers/net/dsa/microchip/ksz_common.c
+++ b/drivers/net/dsa/microchip/ksz_common.c
@@ -3116,7 +3116,8 @@ static void ksz_set_xmii(struct ksz_device *dev, int port,
 		/* On KSZ9893, disable RGMII in-band status support */
 		if (dev->chip_id == KSZ9893_CHIP_ID ||
 		    dev->chip_id == KSZ8563_CHIP_ID ||
-		    dev->chip_id == KSZ9563_CHIP_ID)
+		    dev->chip_id == KSZ9563_CHIP_ID ||
+		    is_lan937x(dev))
 			data8 &= ~P_MII_MAC_MODE;
 		break;
 	default:
-- 
cgit v1.2.3-58-ga151


From 2e3ed20c17e719cbe7b13feaa3e7c46cf6a85887 Mon Sep 17 00:00:00 2001
From: Lucas Stach <l.stach@pengutronix.de>
Date: Mon, 1 Jul 2024 10:53:43 +0200
Subject: net: dsa: microchip: lan937x: disable VPHY support

As described by the microchip article "LAN937X - The required
configuration for the external MAC port to operate at RGMII-to-RGMII
1Gbps link speed." [1]:

"When VPHY is enabled, the auto-negotiation process following IEEE 802.3
standard will be triggered and will result in RGMII-to-RGMII signal
failure on the interface because VPHY will try to poll the PHY status
that is not available in the scenario of RGMII-to-RGMII connection
(normally the link partner is usually an external processor).

Note that when VPHY fails on accessing PHY registers, it will fall back
to 100Mbps speed, it indicates disabling VPHY is optional if you only
need the port to link at 100Mbps speed.

Again, VPHY must and can only be disabled by writing VPHY_DISABLE bit in
the register below as there is no strapping pin for the control."

This patch was tested on LAN9372, so far it seems to not to affect VPHY
based clock crossing optimization for the ports with integrated PHYs.

[1]: https://microchip.my.site.com/s/article/LAN937X-The-required-configuration-for-the-external-MAC-port-to-operate-at-RGMII-to-RGMII-1Gbps-link-speed

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/microchip/lan937x_main.c | 3 +++
 drivers/net/dsa/microchip/lan937x_reg.h  | 4 ++++
 2 files changed, 7 insertions(+)

diff --git a/drivers/net/dsa/microchip/lan937x_main.c b/drivers/net/dsa/microchip/lan937x_main.c
index eaa862eb6b26..0606796b1485 100644
--- a/drivers/net/dsa/microchip/lan937x_main.c
+++ b/drivers/net/dsa/microchip/lan937x_main.c
@@ -390,6 +390,9 @@ int lan937x_setup(struct dsa_switch *ds)
 	lan937x_cfg(dev, REG_SW_GLOBAL_OUTPUT_CTRL__1,
 		    (SW_CLK125_ENB | SW_CLK25_ENB), true);
 
+	/* Disable global VPHY support. Related to CPU interface only? */
+	ksz_rmw32(dev, REG_SW_CFG_STRAP_OVR, SW_VPHY_DISABLE, SW_VPHY_DISABLE);
+
 	return 0;
 }
 
diff --git a/drivers/net/dsa/microchip/lan937x_reg.h b/drivers/net/dsa/microchip/lan937x_reg.h
index 7ecada924023..2f22a9d01de3 100644
--- a/drivers/net/dsa/microchip/lan937x_reg.h
+++ b/drivers/net/dsa/microchip/lan937x_reg.h
@@ -37,6 +37,10 @@
 #define SW_CLK125_ENB			BIT(1)
 #define SW_CLK25_ENB			BIT(0)
 
+/* 2 - PHY Control */
+#define REG_SW_CFG_STRAP_OVR		0x0214
+#define SW_VPHY_DISABLE			BIT(31)
+
 /* 3 - Operation Control */
 #define REG_SW_OPERATION		0x0300
 
-- 
cgit v1.2.3-58-ga151


From cda91d5b911a5a168a1c6e6917afda43b0e458c8 Mon Sep 17 00:00:00 2001
From: Xin Long <lucien.xin@gmail.com>
Date: Mon, 1 Jul 2024 13:48:49 -0400
Subject: sctp: cancel a blocking accept when shutdown a listen socket

As David Laight noticed,

"In a multithreaded program it is reasonable to have a thread blocked in
 accept(). With TCP a subsequent shutdown(listen_fd, SHUT_RDWR) causes
 the accept to fail. But nothing happens for SCTP."

sctp_disconnect() is eventually called when shutdown a listen socket,
but nothing is done in this function. This patch sets RCV_SHUTDOWN
flag in sk->sk_shutdown there, and adds the check (sk->sk_shutdown &
RCV_SHUTDOWN) to break and return in sctp_accept().

Note that shutdown() is only supported on TCP-style SCTP socket.

Reported-by: David Laight <David.Laight@aculab.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sctp/socket.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index c009383369b2..32f76f1298da 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -4834,10 +4834,14 @@ int sctp_inet_connect(struct socket *sock, struct sockaddr *uaddr,
 	return sctp_connect(sock->sk, uaddr, addr_len, flags);
 }
 
-/* FIXME: Write comments. */
+/* Only called when shutdown a listening SCTP socket. */
 static int sctp_disconnect(struct sock *sk, int flags)
 {
-	return -EOPNOTSUPP; /* STUB */
+	if (!sctp_style(sk, TCP))
+		return -EOPNOTSUPP;
+
+	sk->sk_shutdown |= RCV_SHUTDOWN;
+	return 0;
 }
 
 /* 4.1.4 accept() - TCP Style Syntax
@@ -4866,7 +4870,8 @@ static struct sock *sctp_accept(struct sock *sk, struct proto_accept_arg *arg)
 		goto out;
 	}
 
-	if (!sctp_sstate(sk, LISTENING)) {
+	if (!sctp_sstate(sk, LISTENING) ||
+	    (sk->sk_shutdown & RCV_SHUTDOWN)) {
 		error = -EINVAL;
 		goto out;
 	}
@@ -9393,7 +9398,8 @@ static int sctp_wait_for_accept(struct sock *sk, long timeo)
 		}
 
 		err = -EINVAL;
-		if (!sctp_sstate(sk, LISTENING))
+		if (!sctp_sstate(sk, LISTENING) ||
+		    (sk->sk_shutdown & RCV_SHUTDOWN))
 			break;
 
 		err = 0;
-- 
cgit v1.2.3-58-ga151


From 8c5a9f290e37123d28cda7728b53705643a2916f Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Tue, 2 Jul 2024 16:29:31 +0200
Subject: tools: ynl: use ident name for Family, too.

This allow consistent naming convention between Family and others
element's name.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Link: https://patch.msgid.link/9bbcab3094970b371bd47aa18481ae6ca5a93687.1719930479.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/net/ynl/ynl-gen-c.py | 52 +++++++++++++++++++++++-----------------------
 1 file changed, 26 insertions(+), 26 deletions(-)

diff --git a/tools/net/ynl/ynl-gen-c.py b/tools/net/ynl/ynl-gen-c.py
index 374ca5e86e24..51529fabd517 100755
--- a/tools/net/ynl/ynl-gen-c.py
+++ b/tools/net/ynl/ynl-gen-c.py
@@ -59,9 +59,9 @@ class Type(SpecAttr):
         if 'nested-attributes' in attr:
             self.nested_attrs = attr['nested-attributes']
             if self.nested_attrs == family.name:
-                self.nested_render_name = c_lower(f"{family.name}")
+                self.nested_render_name = c_lower(f"{family.ident_name}")
             else:
-                self.nested_render_name = c_lower(f"{family.name}_{self.nested_attrs}")
+                self.nested_render_name = c_lower(f"{family.ident_name}_{self.nested_attrs}")
 
             if self.nested_attrs in self.family.consts:
                 self.nested_struct_type = 'struct ' + self.nested_render_name + '_'
@@ -693,9 +693,9 @@ class Struct:
 
         self.nested = type_list is None
         if family.name == c_lower(space_name):
-            self.render_name = c_lower(family.name)
+            self.render_name = c_lower(family.ident_name)
         else:
-            self.render_name = c_lower(family.name + '-' + space_name)
+            self.render_name = c_lower(family.ident_name + '-' + space_name)
         self.struct_name = 'struct ' + self.render_name
         if self.nested and space_name in family.consts:
             self.struct_name += '_'
@@ -761,7 +761,7 @@ class EnumEntry(SpecEnumEntry):
 
 class EnumSet(SpecEnumSet):
     def __init__(self, family, yaml):
-        self.render_name = c_lower(family.name + '-' + yaml['name'])
+        self.render_name = c_lower(family.ident_name + '-' + yaml['name'])
 
         if 'enum-name' in yaml:
             if yaml['enum-name']:
@@ -777,7 +777,7 @@ class EnumSet(SpecEnumSet):
         else:
             self.user_type = 'int'
 
-        self.value_pfx = yaml.get('name-prefix', f"{family.name}-{yaml['name']}-")
+        self.value_pfx = yaml.get('name-prefix', f"{family.ident_name}-{yaml['name']}-")
 
         super().__init__(family, yaml)
 
@@ -802,9 +802,9 @@ class AttrSet(SpecAttrSet):
             if 'name-prefix' in yaml:
                 pfx = yaml['name-prefix']
             elif self.name == family.name:
-                pfx = family.name + '-a-'
+                pfx = family.ident_name + '-a-'
             else:
-                pfx = f"{family.name}-a-{self.name}-"
+                pfx = f"{family.ident_name}-a-{self.name}-"
             self.name_prefix = c_upper(pfx)
             self.max_name = c_upper(self.yaml.get('attr-max-name', f"{self.name_prefix}max"))
             self.cnt_name = c_upper(self.yaml.get('attr-cnt-name', f"__{self.name_prefix}max"))
@@ -861,7 +861,7 @@ class Operation(SpecOperation):
     def __init__(self, family, yaml, req_value, rsp_value):
         super().__init__(family, yaml, req_value, rsp_value)
 
-        self.render_name = c_lower(family.name + '_' + self.name)
+        self.render_name = c_lower(family.ident_name + '_' + self.name)
 
         self.dual_policy = ('do' in yaml and 'request' in yaml['do']) and \
                          ('dump' in yaml and 'request' in yaml['dump'])
@@ -911,11 +911,11 @@ class Family(SpecFamily):
         if 'uapi-header' in self.yaml:
             self.uapi_header = self.yaml['uapi-header']
         else:
-            self.uapi_header = f"linux/{self.name}.h"
+            self.uapi_header = f"linux/{self.ident_name}.h"
         if self.uapi_header.startswith("linux/") and self.uapi_header.endswith('.h'):
             self.uapi_header_name = self.uapi_header[6:-2]
         else:
-            self.uapi_header_name = self.name
+            self.uapi_header_name = self.ident_name
 
     def resolve(self):
         self.resolve_up(super())
@@ -923,7 +923,7 @@ class Family(SpecFamily):
         if self.yaml.get('protocol', 'genetlink') not in {'genetlink', 'genetlink-c', 'genetlink-legacy'}:
             raise Exception("Codegen only supported for genetlink")
 
-        self.c_name = c_lower(self.name)
+        self.c_name = c_lower(self.ident_name)
         if 'name-prefix' in self.yaml['operations']:
             self.op_prefix = c_upper(self.yaml['operations']['name-prefix'])
         else:
@@ -2173,7 +2173,7 @@ def print_kernel_op_table_fwd(family, cw, terminate):
     exported = not kernel_can_gen_family_struct(family)
 
     if not terminate or exported:
-        cw.p(f"/* Ops table for {family.name} */")
+        cw.p(f"/* Ops table for {family.ident_name} */")
 
         pol_to_struct = {'global': 'genl_small_ops',
                          'per-op': 'genl_ops',
@@ -2225,12 +2225,12 @@ def print_kernel_op_table_fwd(family, cw, terminate):
             continue
 
         if 'do' in op:
-            name = c_lower(f"{family.name}-nl-{op_name}-doit")
+            name = c_lower(f"{family.ident_name}-nl-{op_name}-doit")
             cw.write_func_prot('int', name,
                                ['struct sk_buff *skb', 'struct genl_info *info'], suffix=';')
 
         if 'dump' in op:
-            name = c_lower(f"{family.name}-nl-{op_name}-dumpit")
+            name = c_lower(f"{family.ident_name}-nl-{op_name}-dumpit")
             cw.write_func_prot('int', name,
                                ['struct sk_buff *skb', 'struct netlink_callback *cb'], suffix=';')
     cw.nl()
@@ -2256,13 +2256,13 @@ def print_kernel_op_table(family, cw):
                                             for x in op['dont-validate']])), )
             for op_mode in ['do', 'dump']:
                 if op_mode in op:
-                    name = c_lower(f"{family.name}-nl-{op_name}-{op_mode}it")
+                    name = c_lower(f"{family.ident_name}-nl-{op_name}-{op_mode}it")
                     members.append((op_mode + 'it', name))
             if family.kernel_policy == 'per-op':
                 struct = Struct(family, op['attribute-set'],
                                 type_list=op['do']['request']['attributes'])
 
-                name = c_lower(f"{family.name}-{op_name}-nl-policy")
+                name = c_lower(f"{family.ident_name}-{op_name}-nl-policy")
                 members.append(('policy', name))
                 members.append(('maxattr', struct.attr_max_val.enum_name))
             if 'flags' in op:
@@ -2294,7 +2294,7 @@ def print_kernel_op_table(family, cw):
                         members.append(('validate',
                                         ' | '.join([c_upper('genl-dont-validate-' + x)
                                                     for x in dont_validate])), )
-                name = c_lower(f"{family.name}-nl-{op_name}-{op_mode}it")
+                name = c_lower(f"{family.ident_name}-nl-{op_name}-{op_mode}it")
                 if 'pre' in op[op_mode]:
                     members.append((cb_names[op_mode]['pre'], c_lower(op[op_mode]['pre'])))
                 members.append((op_mode + 'it', name))
@@ -2305,9 +2305,9 @@ def print_kernel_op_table(family, cw):
                                     type_list=op[op_mode]['request']['attributes'])
 
                     if op.dual_policy:
-                        name = c_lower(f"{family.name}-{op_name}-{op_mode}-nl-policy")
+                        name = c_lower(f"{family.ident_name}-{op_name}-{op_mode}-nl-policy")
                     else:
-                        name = c_lower(f"{family.name}-{op_name}-nl-policy")
+                        name = c_lower(f"{family.ident_name}-{op_name}-nl-policy")
                     members.append(('policy', name))
                     members.append(('maxattr', struct.attr_max_val.enum_name))
                 flags = (op['flags'] if 'flags' in op else []) + ['cmd-cap-' + op_mode]
@@ -2326,7 +2326,7 @@ def print_kernel_mcgrp_hdr(family, cw):
 
     cw.block_start('enum')
     for grp in family.mcgrps['list']:
-        grp_id = c_upper(f"{family.name}-nlgrp-{grp['name']},")
+        grp_id = c_upper(f"{family.ident_name}-nlgrp-{grp['name']},")
         cw.p(grp_id)
     cw.block_end(';')
     cw.nl()
@@ -2339,7 +2339,7 @@ def print_kernel_mcgrp_src(family, cw):
     cw.block_start('static const struct genl_multicast_group ' + family.c_name + '_nl_mcgrps[] =')
     for grp in family.mcgrps['list']:
         name = grp['name']
-        grp_id = c_upper(f"{family.name}-nlgrp-{name}")
+        grp_id = c_upper(f"{family.ident_name}-nlgrp-{name}")
         cw.p('[' + grp_id + '] = { "' + name + '", },')
     cw.block_end(';')
     cw.nl()
@@ -2361,7 +2361,7 @@ def print_kernel_family_struct_src(family, cw):
     if not kernel_can_gen_family_struct(family):
         return
 
-    cw.block_start(f"struct genl_family {family.name}_nl_family __ro_after_init =")
+    cw.block_start(f"struct genl_family {family.ident_name}_nl_family __ro_after_init =")
     cw.p('.name\t\t= ' + family.fam_key + ',')
     cw.p('.version\t= ' + family.ver_key + ',')
     cw.p('.netnsok\t= true,')
@@ -2429,7 +2429,7 @@ def render_uapi(family, cw):
                 cw.p(' */')
 
             uapi_enum_start(family, cw, const, 'name')
-            name_pfx = const.get('name-prefix', f"{family.name}-{const['name']}-")
+            name_pfx = const.get('name-prefix', f"{family.ident_name}-{const['name']}-")
             for entry in enum.entries.values():
                 suffix = ','
                 if entry.value_change:
@@ -2451,7 +2451,7 @@ def render_uapi(family, cw):
             cw.nl()
         elif const['type'] == 'const':
             defines.append([c_upper(family.get('c-define-name',
-                                               f"{family.name}-{const['name']}")),
+                                               f"{family.ident_name}-{const['name']}")),
                             const['value']])
 
     if defines:
@@ -2529,7 +2529,7 @@ def render_uapi(family, cw):
     defines = []
     for grp in family.mcgrps['list']:
         name = grp['name']
-        defines.append([c_upper(grp.get('c-define-name', f"{family.name}-mcgrp-{name}")),
+        defines.append([c_upper(grp.get('c-define-name', f"{family.ident_name}-mcgrp-{name}")),
                         f'{name}'])
     cw.nl()
     if defines:
-- 
cgit v1.2.3-58-ga151


From 0b8774586be5b2f3aa9a0e665846c985e15f621e Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 2 Jul 2024 16:37:28 -0700
Subject: selftests: drv-net: rss_ctx: allow more noise on default context

As predicted by David running the test on a machine with a single
interface is a bit unreliable. We try to send 20k packets with
iperf and expect fewer than 10k packets on the default context.
The test isn't very quick, iperf will usually send 100k packets
by the time we stop it. So we're off by 5x on the number of iperf
packets but still expect default context to only get the hardcoded
10k. The intent is to make sure we get noticeably less traffic
on the default context. Use half of the resulting iperf traffic
instead of the hard coded 10k.

Link: https://patch.msgid.link/20240702233728.4183387-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/hw/rss_ctx.py | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
index 3a804e41f7cb..475f2a63fcd5 100755
--- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py
+++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
@@ -174,8 +174,10 @@ def test_rss_context(cfg, ctx_cnt=1, create_with_cfg=None):
         GenerateTraffic(cfg, port=ports[i]).wait_pkts_and_stop(20000)
         cnts = _get_rx_cnts(cfg, prev=cnts)
 
-        ksft_lt(sum(cnts[ :2]), 10000, "traffic on main context:" + str(cnts))
-        ksft_ge(sum(cnts[2+i*2:4+i*2]), 20000, f"traffic on context {i}: " + str(cnts))
+        directed = sum(cnts[2+i*2:4+i*2])
+
+        ksft_lt(sum(cnts[ :2]), directed / 2, "traffic on main context:" + str(cnts))
+        ksft_ge(directed, 20000, f"traffic on context {i}: " + str(cnts))
         ksft_eq(sum(cnts[2:2+i*2] + cnts[4+i*2:]), 0, "traffic on other contexts: " + str(cnts))
 
     if requested_ctx_cnt != ctx_cnt:
@@ -233,8 +235,9 @@ def test_rss_context_out_of_order(cfg, ctx_cnt=4):
             cnts = _get_rx_cnts(cfg, prev=cnts)
 
             if ctx[i]:
-                ksft_lt(sum(cnts[ :2]), 10000, "traffic on main context:" + str(cnts))
-                ksft_ge(sum(cnts[2+i*2:4+i*2]), 20000, f"traffic on context {i}: " + str(cnts))
+                directed = sum(cnts[2+i*2:4+i*2])
+                ksft_lt(sum(cnts[ :2]), directed / 2, "traffic on main context:" + str(cnts))
+                ksft_ge(directed, 20000, f"traffic on context {i}: " + str(cnts))
                 ksft_eq(sum(cnts[2:2+i*2] + cnts[4+i*2:]), 0, "traffic on other contexts: " + str(cnts))
             else:
                 ksft_ge(sum(cnts[ :2]), 20000, "traffic on main context:" + str(cnts))
@@ -328,8 +331,9 @@ def test_rss_context_overlap(cfg, other_ctx=0):
     GenerateTraffic(cfg, port=port).wait_pkts_and_stop(20000)
     cnts = _get_rx_cnts(cfg, prev=cnts)
 
-    ksft_lt(sum(cnts[ :2]),  7000, "traffic on main context: " + str(cnts))
-    ksft_ge(sum(cnts[2:4]), 20000, "traffic on extra context: " + str(cnts))
+    directed = sum(cnts[2:4])
+    ksft_lt(sum(cnts[ :2]), directed / 2, "traffic on main context: " + str(cnts))
+    ksft_ge(directed, 20000, "traffic on extra context: " + str(cnts))
     if other_ctx == 0:
         ksft_eq(sum(cnts[4: ]),     0, "traffic on other queues: " + str(cnts))
 
-- 
cgit v1.2.3-58-ga151


From 1a16cdf77e0d7de0fb640e65d65c0898b38c1b4b Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 2 Jul 2024 09:41:57 -0700
Subject: net: ethtool: fix compat with old RSS context API

Device driver gets access to rxfh_dev, while rxfh is just a local
copy of user space params. We need to check what RSS context ID
driver assigned in rxfh_dev, not rxfh.

Using rxfh leads to trying to store all contexts at index 0xffffffff.
From the user perspective it leads to "driver chose duplicate ID"
warnings when second context is added and inability to access any
contexts even tho they were successfully created - xa_load() for
the actual context ID will return NULL, and syscall will return -ENOENT.

Looks like a rebasing mistake, since rxfh_dev was added relatively
recently by commit fb6e30a72539 ("net: ethtool: pass a pointer to
parameters to get/set_rxfh ethtool ops").

Fixes: eac9122f0c41 ("net: ethtool: record custom RSS contexts in the XArray")
Reviewed-by: Edward Cree <ecree.xilinx@gmail.com>
Link: https://patch.msgid.link/20240702164157.4018425-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ethtool/ioctl.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index d8795ed07ba3..46f0497ae6bc 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -1483,13 +1483,13 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 	/* Update rss_ctx tracking */
 	if (create && !ops->create_rxfh_context) {
 		/* driver uses old API, it chose context ID */
-		if (WARN_ON(xa_load(&dev->ethtool->rss_ctx, rxfh.rss_context))) {
+		if (WARN_ON(xa_load(&dev->ethtool->rss_ctx, rxfh_dev.rss_context))) {
 			/* context ID reused, our tracking is screwed */
 			kfree(ctx);
 			goto out;
 		}
 		/* Allocate the exact ID the driver gave us */
-		if (xa_is_err(xa_store(&dev->ethtool->rss_ctx, rxfh.rss_context,
+		if (xa_is_err(xa_store(&dev->ethtool->rss_ctx, rxfh_dev.rss_context,
 				       ctx, GFP_KERNEL))) {
 			kfree(ctx);
 			goto out;
-- 
cgit v1.2.3-58-ga151


From ff015706fc7385c51e8418ced2484b98b239a3a8 Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Tue, 2 Jul 2024 09:28:28 -0400
Subject: selftests: openvswitch: Bump timeout to 15 minutes.

We found that since some tests rely on the TCP SYN timeouts to cause flow
misses, the default test suite timeout of 45 seconds is quick to be
exceeded.  Bump the timeout to 15 minutes.

Signed-off-by: Aaron Conole <aconole@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240702132830.213384-2-aconole@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/openvswitch/settings | 1 +
 1 file changed, 1 insertion(+)
 create mode 100644 tools/testing/selftests/net/openvswitch/settings

diff --git a/tools/testing/selftests/net/openvswitch/settings b/tools/testing/selftests/net/openvswitch/settings
new file mode 100644
index 000000000000..e2206265f67c
--- /dev/null
+++ b/tools/testing/selftests/net/openvswitch/settings
@@ -0,0 +1 @@
+timeout=900
-- 
cgit v1.2.3-58-ga151


From 818481db3df41e58396d13fd9fad77542a0706be Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Tue, 2 Jul 2024 09:28:29 -0400
Subject: selftests: openvswitch: Attempt to autoload module.

Previously, the openvswitch.sh test suites would not attempt to autoload
the openvswitch module.  The idea was that a user who is manually running
tests might not even have the OVS module loaded or configured for their
own development.  However, if the kernel module is configured, and the
module can be autoloaded then we should just attempt to load it and run
the tests.  This is especially true in the CI environments, where the CI
tests should be able to rely on auto loading to get the test suite running.

Signed-off-by: Aaron Conole <aconole@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240702132830.213384-3-aconole@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/openvswitch/openvswitch.sh | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/openvswitch.sh b/tools/testing/selftests/net/openvswitch/openvswitch.sh
index 15bca0708717..0bd0425848d9 100755
--- a/tools/testing/selftests/net/openvswitch/openvswitch.sh
+++ b/tools/testing/selftests/net/openvswitch/openvswitch.sh
@@ -613,16 +613,20 @@ run_test() {
 	tname="$1"
 	tdesc="$2"
 
-	if ! lsmod | grep openvswitch >/dev/null 2>&1; then
-		stdbuf -o0 printf "TEST: %-60s  [NOMOD]\n" "${tdesc}"
-		return $ksft_skip
-	fi
-
 	if python3 ovs-dpctl.py -h 2>&1 | \
 	     grep -E "Need to (install|upgrade) the python" >/dev/null 2>&1; then
 		stdbuf -o0 printf "TEST: %-60s  [PYLIB]\n" "${tdesc}"
 		return $ksft_skip
 	fi
+
+	python3 ovs-dpctl.py show >/dev/null 2>&1 || \
+		echo "[DPCTL] show exception."
+
+	if ! lsmod | grep openvswitch >/dev/null 2>&1; then
+		stdbuf -o0 printf "TEST: %-60s  [NOMOD]\n" "${tdesc}"
+		return $ksft_skip
+	fi
+
 	printf "TEST: %-60s  [START]\n" "${tname}"
 
 	unset IFS
-- 
cgit v1.2.3-58-ga151


From 7abfd8ecb7858006edceeeb49c82a1eece2982f5 Mon Sep 17 00:00:00 2001
From: Aaron Conole <aconole@redhat.com>
Date: Tue, 2 Jul 2024 09:28:30 -0400
Subject: selftests: openvswitch: Be more verbose with selftest debugging.

The openvswitch selftest is difficult to debug for anyone that isn't
directly familiar with the openvswitch module and the specifics of the
test cases.  Many times when something fails, the debug log will be
sparsely populated and it takes some time to understand where a failure
occured.

Increase the amount of details logged to the debug log by trapping all
'info' logs, and all 'ovs_sbx' commands.

Signed-off-by: Aaron Conole <aconole@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240702132830.213384-4-aconole@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/openvswitch/openvswitch.sh | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/openvswitch.sh b/tools/testing/selftests/net/openvswitch/openvswitch.sh
index 0bd0425848d9..531951086d9c 100755
--- a/tools/testing/selftests/net/openvswitch/openvswitch.sh
+++ b/tools/testing/selftests/net/openvswitch/openvswitch.sh
@@ -23,7 +23,9 @@ tests="
 	drop_reason				drop: test drop reasons are emitted"
 
 info() {
-    [ $VERBOSE = 0 ] || echo $*
+	[ "${ovs_dir}" != "" ] &&
+		echo "`date +"[%m-%d %H:%M:%S]"` $*" >> ${ovs_dir}/debug.log
+	[ $VERBOSE = 0 ] || echo $*
 }
 
 ovs_base=`pwd`
@@ -65,7 +67,8 @@ ovs_setenv() {
 
 ovs_sbx() {
 	if test "X$2" != X; then
-		(ovs_setenv $1; shift; "$@" >> ${ovs_dir}/debug.log)
+		(ovs_setenv $1; shift;
+		 info "run cmd: $@"; "$@" >> ${ovs_dir}/debug.log)
 	else
 		ovs_setenv $1
 	fi
@@ -139,7 +142,7 @@ ovs_add_flow () {
 	info "Adding flow to DP: sbx:$1 br:$2 flow:$3 act:$4"
 	ovs_sbx "$1" python3 $ovs_base/ovs-dpctl.py add-flow "$2" "$3" "$4"
 	if [ $? -ne 0 ]; then
-		echo "Flow [ $3 : $4 ] failed" >> ${ovs_dir}/debug.log
+		info "Flow [ $3 : $4 ] failed"
 		return 1
 	fi
 	return 0
-- 
cgit v1.2.3-58-ga151


From 47c130130de2fd68d9e4f591b0ea25975bdad68a Mon Sep 17 00:00:00 2001
From: Thorsten Blum <thorsten.blum@toblux.com>
Date: Wed, 3 Jul 2024 08:11:48 +0200
Subject: l2tp: Remove duplicate included header file trace.h

Remove duplicate included header file trace.h and the following warning
reported by make includecheck:

  trace.h is included more than once

Compile-tested only.

Signed-off-by: Thorsten Blum <thorsten.blum@toblux.com>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Link: https://patch.msgid.link/20240703061147.691973-2-thorsten.blum@toblux.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/l2tp/l2tp_core.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index 64f446f0930b..a99032076e04 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -60,7 +60,6 @@
 #include <linux/atomic.h>
 
 #include "l2tp_core.h"
-#include "trace.h"
 
 #define CREATE_TRACE_POINTS
 #include "trace.h"
-- 
cgit v1.2.3-58-ga151


From aa77b112801613d394f893509d8187f1a332c1a4 Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Wed, 3 Jul 2024 10:38:20 +0200
Subject: net: dsa: microchip: lan937x: Add error handling in lan937x_setup

Introduce error handling for lan937x_cfg function calls in lan937x_setup.
This change ensures that if any lan937x_cfg or ksz_rmw32 calls fails, the
function will return the appropriate error code.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Acked-by: Arun Ramadoss <arun.ramadoss@microchip.com>
Link: https://patch.msgid.link/20240703083820.3152100-1-o.rempel@pengutronix.de
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/dsa/microchip/lan937x_main.c | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/drivers/net/dsa/microchip/lan937x_main.c b/drivers/net/dsa/microchip/lan937x_main.c
index 0606796b1485..83ac33fede3f 100644
--- a/drivers/net/dsa/microchip/lan937x_main.c
+++ b/drivers/net/dsa/microchip/lan937x_main.c
@@ -374,26 +374,33 @@ int lan937x_setup(struct dsa_switch *ds)
 	ds->vlan_filtering_is_global = true;
 
 	/* Enable aggressive back off for half duplex & UNH mode */
-	lan937x_cfg(dev, REG_SW_MAC_CTRL_0,
-		    (SW_PAUSE_UNH_MODE | SW_NEW_BACKOFF | SW_AGGR_BACKOFF),
-		    true);
+	ret = lan937x_cfg(dev, REG_SW_MAC_CTRL_0, (SW_PAUSE_UNH_MODE |
+						   SW_NEW_BACKOFF |
+						   SW_AGGR_BACKOFF), true);
+	if (ret < 0)
+		return ret;
 
 	/* If NO_EXC_COLLISION_DROP bit is set, the switch will not drop
 	 * packets when 16 or more collisions occur
 	 */
-	lan937x_cfg(dev, REG_SW_MAC_CTRL_1, NO_EXC_COLLISION_DROP, true);
+	ret = lan937x_cfg(dev, REG_SW_MAC_CTRL_1, NO_EXC_COLLISION_DROP, true);
+	if (ret < 0)
+		return ret;
 
 	/* enable global MIB counter freeze function */
-	lan937x_cfg(dev, REG_SW_MAC_CTRL_6, SW_MIB_COUNTER_FREEZE, true);
+	ret = lan937x_cfg(dev, REG_SW_MAC_CTRL_6, SW_MIB_COUNTER_FREEZE, true);
+	if (ret < 0)
+		return ret;
 
 	/* disable CLK125 & CLK25, 1: disable, 0: enable */
-	lan937x_cfg(dev, REG_SW_GLOBAL_OUTPUT_CTRL__1,
-		    (SW_CLK125_ENB | SW_CLK25_ENB), true);
+	ret = lan937x_cfg(dev, REG_SW_GLOBAL_OUTPUT_CTRL__1,
+			  (SW_CLK125_ENB | SW_CLK25_ENB), true);
+	if (ret < 0)
+		return ret;
 
 	/* Disable global VPHY support. Related to CPU interface only? */
-	ksz_rmw32(dev, REG_SW_CFG_STRAP_OVR, SW_VPHY_DISABLE, SW_VPHY_DISABLE);
-
-	return 0;
+	return ksz_rmw32(dev, REG_SW_CFG_STRAP_OVR, SW_VPHY_DISABLE,
+			 SW_VPHY_DISABLE);
 }
 
 void lan937x_teardown(struct dsa_switch *ds)
-- 
cgit v1.2.3-58-ga151


From 4278d88fedcfde1b91a5cfa96acb0e06d7742e12 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 2 Jul 2024 13:00:00 +0200
Subject: wifi: iwlwifi: mvm: fix interface combinations

The NAN code referenced in this commit isn't actually
present in the driver (any more), and the commit didn't
add the extra NAN entry. Thus, the -1 is incorrect.

Reported-by: Alexander Wetzel <Alexander@wetzel-home.de>
Closes: https://lore.kernel.org/20240702104128.26394-1-Alexander@wetzel-home.de
Fixes: 5c38bedac16a ("wifi: iwlwifi: mvm: unify and fix interface combinations")
Link: https://patch.msgid.link/20240702130001.8c871a3f0b5a.I08a6542f52f63c5bd66bf3feb09e1998ce7c60e5@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index 60bfe42d5386..e40f993b17fd 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -59,13 +59,13 @@ static const struct ieee80211_iface_combination iwl_mvm_iface_combinations[] = {
 		.num_different_channels = 2,
 		.max_interfaces = 3,
 		.limits = iwl_mvm_limits,
-		.n_limits = ARRAY_SIZE(iwl_mvm_limits) - 1,
+		.n_limits = ARRAY_SIZE(iwl_mvm_limits),
 	},
 	{
 		.num_different_channels = 1,
 		.max_interfaces = 3,
 		.limits = iwl_mvm_limits_ap,
-		.n_limits = ARRAY_SIZE(iwl_mvm_limits_ap) - 1,
+		.n_limits = ARRAY_SIZE(iwl_mvm_limits_ap),
 	},
 };
 
-- 
cgit v1.2.3-58-ga151


From 187accaa328dc4de98064eef176841b8a4716f96 Mon Sep 17 00:00:00 2001
From: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Date: Wed, 3 Jul 2024 06:43:17 +0300
Subject: wifi: iwlwifi: mvm: don't send an ROC command with max_delay = 0

The firmware can't handle that (it will crash with ASSERT 300A).
This happened because we looked at vif->bss_conf which is not
the right bss_conf to look at in case of an MLD connection.
Fix iwl_mvm_roc_duration_and_delay to iterate on the active links to
get the right value for the dtim_interval.

Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://patch.msgid.link/20240703064027.e12f8d84c8fd.I3dd9f720c678c06ec7a5bf7ca56e21cf0b614c8c@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/time-event.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
index 77b0cae8566f..a8c42ce3b630 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/time-event.c
@@ -1057,12 +1057,21 @@ void iwl_mvm_roc_duration_and_delay(struct ieee80211_vif *vif,
 				    u32 *duration_tu,
 				    u32 *delay)
 {
-	u32 dtim_interval = vif->bss_conf.dtim_period *
-		vif->bss_conf.beacon_int;
+	struct ieee80211_bss_conf *link_conf;
+	unsigned int link_id;
+	u32 dtim_interval = 0;
 
 	*delay = AUX_ROC_MIN_DELAY;
 	*duration_tu = MSEC_TO_TU(duration_ms);
 
+	rcu_read_lock();
+	for_each_vif_active_link(vif, link_conf, link_id) {
+		dtim_interval =
+			max_t(u32, dtim_interval,
+			      link_conf->dtim_period * link_conf->beacon_int);
+	}
+	rcu_read_unlock();
+
 	/*
 	 * If we are associated we want the delay time to be at least one
 	 * dtim interval so that the FW can wait until after the DTIM and
@@ -1071,8 +1080,10 @@ void iwl_mvm_roc_duration_and_delay(struct ieee80211_vif *vif,
 	 * Since we want to use almost a whole dtim interval we would also
 	 * like the delay to be for 2-3 dtim intervals, in case there are
 	 * other time events with higher priority.
+	 * dtim_interval should never be 0, it can be 1 if we don't know it
+	 * (we haven't heard any beacon yet).
 	 */
-	if (vif->cfg.assoc) {
+	if (vif->cfg.assoc && !WARN_ON(!dtim_interval)) {
 		*delay = min_t(u32, dtim_interval * 3, AUX_ROC_MAX_DELAY);
 		/* We cannot remain off-channel longer than the DTIM interval */
 		if (dtim_interval <= *duration_tu) {
-- 
cgit v1.2.3-58-ga151


From eb801f4fabdf877d38c4746333e3454c4e7603cc Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 3 Jul 2024 12:58:49 +0300
Subject: wifi: iwlwifi: mvm: clean up reorder buffer data

We really don't need to maintain the buffer size per
queue buffer, it's the same for the whole BA session.
Also, we no longer use the mvm pointer inside each
queue's data structure. Clean that up.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.64ea1ba75379.I2a25af040061efaf82379e96a84a76c5fb65c677@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h  |  6 ++----
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c |  6 +++---
 drivers/net/wireless/intel/iwlwifi/mvm/sta.c  | 11 +++++------
 3 files changed, 10 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 393ce424c196..227463522b7c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -739,24 +739,20 @@ struct iwl_mvm_tcm {
  * struct iwl_mvm_reorder_buffer - per ra/tid/queue reorder buffer
  * @head_sn: reorder window head sn
  * @num_stored: number of mpdus stored in the buffer
- * @buf_size: the reorder buffer size as set by the last addba request
  * @queue: queue of this reorder buffer
  * @last_amsdu: track last ASMDU SN for duplication detection
  * @last_sub_index: track ASMDU sub frame index for duplication detection
  * @valid: reordering is valid for this queue
  * @lock: protect reorder buffer internal state
- * @mvm: mvm pointer, needed for frame timer context
  */
 struct iwl_mvm_reorder_buffer {
 	u16 head_sn;
 	u16 num_stored;
-	u16 buf_size;
 	int queue;
 	u16 last_amsdu;
 	u8 last_sub_index;
 	bool valid;
 	spinlock_t lock;
-	struct iwl_mvm *mvm;
 } ____cacheline_aligned_in_smp;
 
 /**
@@ -778,6 +774,7 @@ __aligned(roundup_pow_of_two(sizeof(struct sk_buff_head)))
  * @tid: tid of the session
  * @baid: baid of the session
  * @timeout: the timeout set in the addba request
+ * @buf_size: the reorder buffer size as set by the last addba request
  * @entries_per_queue: # of buffers per queue, this actually gets
  *	aligned up to avoid cache line sharing between queues
  * @last_rx: last rx jiffies, updated only if timeout passed from last update
@@ -794,6 +791,7 @@ struct iwl_mvm_baid_data {
 	u8 tid;
 	u8 baid;
 	u16 timeout;
+	u16 buf_size;
 	u16 entries_per_queue;
 	unsigned long last_rx;
 	struct timer_list session_timer;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
index 5370580f6210..1a210d0c22b3 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c
@@ -566,7 +566,7 @@ static void iwl_mvm_release_frames(struct iwl_mvm *mvm,
 	lockdep_assert_held(&reorder_buf->lock);
 
 	while (ieee80211_sn_less(ssn, nssn)) {
-		int index = ssn % reorder_buf->buf_size;
+		int index = ssn % baid_data->buf_size;
 		struct sk_buff_head *skb_list = &entries[index].frames;
 		struct sk_buff *skb;
 
@@ -617,7 +617,7 @@ static void iwl_mvm_del_ba(struct iwl_mvm *mvm, int queue,
 	spin_lock_bh(&reorder_buf->lock);
 	iwl_mvm_release_frames(mvm, sta, NULL, ba_data, reorder_buf,
 			       ieee80211_sn_add(reorder_buf->head_sn,
-						reorder_buf->buf_size));
+						ba_data->buf_size));
 	spin_unlock_bh(&reorder_buf->lock);
 
 out:
@@ -839,7 +839,7 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm,
 	}
 
 	/* put in reorder buffer */
-	index = sn % buffer->buf_size;
+	index = sn % baid_data->buf_size;
 	__skb_queue_tail(&entries[index].frames, skb);
 	buffer->num_stored++;
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
index c57755e5ff5e..15e64d94d6ea 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c
@@ -2743,7 +2743,7 @@ static void iwl_mvm_free_reorder(struct iwl_mvm *mvm,
 		 */
 		WARN_ON(1);
 
-		for (j = 0; j < reorder_buf->buf_size; j++)
+		for (j = 0; j < data->buf_size; j++)
 			__skb_queue_purge(&entries[j].frames);
 
 		spin_unlock_bh(&reorder_buf->lock);
@@ -2752,7 +2752,7 @@ static void iwl_mvm_free_reorder(struct iwl_mvm *mvm,
 
 static void iwl_mvm_init_reorder_buffer(struct iwl_mvm *mvm,
 					struct iwl_mvm_baid_data *data,
-					u16 ssn, u16 buf_size)
+					u16 ssn)
 {
 	int i;
 
@@ -2765,12 +2765,10 @@ static void iwl_mvm_init_reorder_buffer(struct iwl_mvm *mvm,
 
 		reorder_buf->num_stored = 0;
 		reorder_buf->head_sn = ssn;
-		reorder_buf->buf_size = buf_size;
 		spin_lock_init(&reorder_buf->lock);
-		reorder_buf->mvm = mvm;
 		reorder_buf->queue = i;
 		reorder_buf->valid = false;
-		for (j = 0; j < reorder_buf->buf_size; j++)
+		for (j = 0; j < data->buf_size; j++)
 			__skb_queue_head_init(&entries[j].frames);
 	}
 }
@@ -2979,13 +2977,14 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 		baid_data->mvm = mvm;
 		baid_data->tid = tid;
 		baid_data->sta_mask = iwl_mvm_sta_fw_id_mask(mvm, sta, -1);
+		baid_data->buf_size = buf_size;
 
 		mvm_sta->tid_to_baid[tid] = baid;
 		if (timeout)
 			mod_timer(&baid_data->session_timer,
 				  TU_TO_EXP_TIME(timeout * 2));
 
-		iwl_mvm_init_reorder_buffer(mvm, baid_data, ssn, buf_size);
+		iwl_mvm_init_reorder_buffer(mvm, baid_data, ssn);
 		/*
 		 * protect the BA data with RCU to cover a case where our
 		 * internal RX sync mechanism will timeout (not that it's
-- 
cgit v1.2.3-58-ga151


From d81bf4b63006330d1f70a59414eae61e1a60e677 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 3 Jul 2024 12:58:50 +0300
Subject: wifi: iwlwifi: mvm: align reorder buffer entries to cacheline

The entries[] array needs to be cacheline aligned to avoid false
sharing between different queues, each queue has a set of entries
in it that it writes to.

While it is aligned in practice today given that each array entry
before it is aligned, it's still clearer to explicitly require it
to be aligned, so add the annotation for that.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.3bc7a55ac867.Id3c1df6d40e92c3de9caededcbc32d0e57e4423d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 227463522b7c..056fdc42e253 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -798,7 +798,7 @@ struct iwl_mvm_baid_data {
 	struct iwl_mvm_baid_data __rcu **rcu_ptr;
 	struct iwl_mvm *mvm;
 	struct iwl_mvm_reorder_buffer reorder_buf[IWL_MAX_RX_HW_QUEUES];
-	struct iwl_mvm_reorder_buf_entry entries[];
+	struct iwl_mvm_reorder_buf_entry entries[] ____cacheline_aligned_in_smp;
 };
 
 static inline struct iwl_mvm_baid_data *
-- 
cgit v1.2.3-58-ga151


From f27579ff8b9344226db63a902dc9e9cbddf07b74 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 3 Jul 2024 12:58:51 +0300
Subject: wifi: iwlwifi: mvm: simplify EMLSR blocking

If EMLSR is already blocked for the same reason that
it's blocked for again, there's no need to actually
do any work, so exit early from the function. Also,
print the state after modifying it, so it's clearer.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.6995464f0bac.Iac9fe3546ca0a0d6bc6666c822a667ab257419a9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/link.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 73527781f89a..4e702e7a4b82 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -1033,15 +1033,17 @@ void iwl_mvm_block_esr(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	if (WARN_ON(!(reason & IWL_MVM_BLOCK_ESR_REASONS)))
 		return;
 
-	if (!(mvmvif->esr_disable_reason & reason)) {
-		IWL_DEBUG_INFO(mvm,
-			       "Blocking EMLSR mode. reason = %s (0x%x)\n",
-			       iwl_get_esr_state_string(reason), reason);
-		iwl_mvm_print_esr_state(mvm, mvmvif->esr_disable_reason);
-	}
+	if (mvmvif->esr_disable_reason & reason)
+		return;
+
+	IWL_DEBUG_INFO(mvm,
+		       "Blocking EMLSR mode. reason = %s (0x%x)\n",
+		       iwl_get_esr_state_string(reason), reason);
 
 	mvmvif->esr_disable_reason |= reason;
 
+	iwl_mvm_print_esr_state(mvm, mvmvif->esr_disable_reason);
+
 	iwl_mvm_exit_esr(mvm, vif, reason, link_to_keep);
 }
 
-- 
cgit v1.2.3-58-ga151


From d023a228e927c142c5c418dae346f63d1c8ba105 Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Wed, 3 Jul 2024 12:58:52 +0300
Subject: wifi: iwlwifi: return a new allocation for hdr page space

Instead of returning the pointer to the structure describing the header
page, return the pointer to the newly allocated area. This disentangles
the user from the allocation within the page as it does not need to
advance the position itself.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://patch.msgid.link/20240703125541.044f2cb373f1.I52a807ac6f311b89530e18deacc7452638a6f5d8@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h |  4 +-
 drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c  | 35 +++++++--------
 drivers/net/wireless/intel/iwlwifi/pcie/tx.c       | 50 +++++++++++-----------
 3 files changed, 44 insertions(+), 45 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index f2bb34270ccf..754928062921 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -623,8 +623,8 @@ void iwl_trans_pcie_tx_reset(struct iwl_trans *trans);
 int iwl_pcie_txq_alloc(struct iwl_trans *trans, struct iwl_txq *txq,
 		       int slots_num, bool cmd_queue);
 
-struct iwl_tso_hdr_page *iwl_pcie_get_page_hdr(struct iwl_trans *trans,
-					       size_t len, struct sk_buff *skb);
+void *iwl_pcie_get_page_hdr(struct iwl_trans *trans, size_t len,
+			    struct sk_buff *skb);
 void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb);
 
 static inline dma_addr_t
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
index b897de1b9226..543efb52b084 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
@@ -130,7 +130,6 @@ static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 	unsigned int mss = skb_shinfo(skb)->gso_size;
 	u16 length, amsdu_pad;
 	u8 *start_hdr;
-	struct iwl_tso_hdr_page *hdr_page;
 	struct tso_t tso;
 
 	trace_iwlwifi_dev_tx(trans->dev, skb, tfd, sizeof(*tfd),
@@ -146,12 +145,10 @@ static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 		(3 + snap_ip_tcp_hdrlen + sizeof(struct ethhdr));
 
 	/* Our device supports 9 segments at most, it will fit in 1 page */
-	hdr_page = iwl_pcie_get_page_hdr(trans, hdr_room, skb);
-	if (!hdr_page)
+	start_hdr = iwl_pcie_get_page_hdr(trans, hdr_room, skb);
+	if (!start_hdr)
 		return -ENOMEM;
 
-	start_hdr = hdr_page->pos;
-
 	/*
 	 * Pull the ieee80211 header to be able to use TSO core,
 	 * we will restore it for the tx_status flow.
@@ -172,32 +169,32 @@ static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 		unsigned int data_left = min_t(unsigned int, mss, total_len);
 		unsigned int tb_len;
 		dma_addr_t tb_phys;
-		u8 *subf_hdrs_start = hdr_page->pos;
+		u8 *pos_hdr = start_hdr;
 
 		total_len -= data_left;
 
-		memset(hdr_page->pos, 0, amsdu_pad);
-		hdr_page->pos += amsdu_pad;
+		memset(pos_hdr, 0, amsdu_pad);
+		pos_hdr += amsdu_pad;
 		amsdu_pad = (4 - (sizeof(struct ethhdr) + snap_ip_tcp_hdrlen +
 				  data_left)) & 0x3;
-		ether_addr_copy(hdr_page->pos, ieee80211_get_DA(hdr));
-		hdr_page->pos += ETH_ALEN;
-		ether_addr_copy(hdr_page->pos, ieee80211_get_SA(hdr));
-		hdr_page->pos += ETH_ALEN;
+		ether_addr_copy(pos_hdr, ieee80211_get_DA(hdr));
+		pos_hdr += ETH_ALEN;
+		ether_addr_copy(pos_hdr, ieee80211_get_SA(hdr));
+		pos_hdr += ETH_ALEN;
 
 		length = snap_ip_tcp_hdrlen + data_left;
-		*((__be16 *)hdr_page->pos) = cpu_to_be16(length);
-		hdr_page->pos += sizeof(length);
+		*((__be16 *)pos_hdr) = cpu_to_be16(length);
+		pos_hdr += sizeof(length);
 
 		/*
 		 * This will copy the SNAP as well which will be considered
 		 * as MAC header.
 		 */
-		tso_build_hdr(skb, hdr_page->pos, &tso, data_left, !total_len);
+		tso_build_hdr(skb, pos_hdr, &tso, data_left, !total_len);
 
-		hdr_page->pos += snap_ip_tcp_hdrlen;
+		pos_hdr += snap_ip_tcp_hdrlen;
 
-		tb_len = hdr_page->pos - start_hdr;
+		tb_len = pos_hdr - start_hdr;
 		tb_phys = dma_map_single(trans->dev, start_hdr,
 					 tb_len, DMA_TO_DEVICE);
 		if (unlikely(dma_mapping_error(trans->dev, tb_phys)))
@@ -211,10 +208,10 @@ static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 		trace_iwlwifi_dev_tx_tb(trans->dev, skb, start_hdr,
 					tb_phys, tb_len);
 		/* add this subframe's headers' length to the tx_cmd */
-		le16_add_cpu(&tx_cmd->len, hdr_page->pos - subf_hdrs_start);
+		le16_add_cpu(&tx_cmd->len, tb_len);
 
 		/* prepare the start_hdr for the next subframe */
-		start_hdr = hdr_page->pos;
+		start_hdr = pos_hdr;
 
 		/* put the payload */
 		while (data_left) {
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index f4452732417d..d67ad40634f9 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -1702,12 +1702,13 @@ static int iwl_fill_data_tbs(struct iwl_trans *trans, struct sk_buff *skb,
 }
 
 #ifdef CONFIG_INET
-struct iwl_tso_hdr_page *iwl_pcie_get_page_hdr(struct iwl_trans *trans,
-					       size_t len, struct sk_buff *skb)
+void *iwl_pcie_get_page_hdr(struct iwl_trans *trans,
+			    size_t len, struct sk_buff *skb)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_tso_hdr_page *p = this_cpu_ptr(trans_pcie->txqs.tso_hdr_page);
 	struct page **page_ptr;
+	void *ret;
 
 	page_ptr = (void *)((u8 *)skb->cb + trans_pcie->txqs.page_offs);
 
@@ -1744,7 +1745,10 @@ alloc:
 out:
 	*page_ptr = p->page;
 	get_page(p->page);
-	return p;
+	ret = p->pos;
+	p->pos += len;
+
+	return ret;
 }
 
 static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
@@ -1759,8 +1763,7 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 	unsigned int snap_ip_tcp_hdrlen, ip_hdrlen, total_len, hdr_room;
 	unsigned int mss = skb_shinfo(skb)->gso_size;
 	u16 length, iv_len, amsdu_pad;
-	u8 *start_hdr;
-	struct iwl_tso_hdr_page *hdr_page;
+	u8 *start_hdr, *pos_hdr;
 	struct tso_t tso;
 
 	/* if the packet is protected, then it must be CCMP or GCMP */
@@ -1783,13 +1786,12 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 		(3 + snap_ip_tcp_hdrlen + sizeof(struct ethhdr)) + iv_len;
 
 	/* Our device supports 9 segments at most, it will fit in 1 page */
-	hdr_page = iwl_pcie_get_page_hdr(trans, hdr_room, skb);
-	if (!hdr_page)
+	pos_hdr = start_hdr = iwl_pcie_get_page_hdr(trans, hdr_room, skb);
+	if (!start_hdr)
 		return -ENOMEM;
 
-	start_hdr = hdr_page->pos;
-	memcpy(hdr_page->pos, skb->data + hdr_len, iv_len);
-	hdr_page->pos += iv_len;
+	memcpy(pos_hdr, skb->data + hdr_len, iv_len);
+	pos_hdr += iv_len;
 
 	/*
 	 * Pull the ieee80211 header + IV to be able to use TSO core,
@@ -1812,32 +1814,32 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 			min_t(unsigned int, mss, total_len);
 		unsigned int hdr_tb_len;
 		dma_addr_t hdr_tb_phys;
-		u8 *subf_hdrs_start = hdr_page->pos;
+		u8 *subf_hdrs_start = pos_hdr;
 
 		total_len -= data_left;
 
-		memset(hdr_page->pos, 0, amsdu_pad);
-		hdr_page->pos += amsdu_pad;
+		memset(pos_hdr, 0, amsdu_pad);
+		pos_hdr += amsdu_pad;
 		amsdu_pad = (4 - (sizeof(struct ethhdr) + snap_ip_tcp_hdrlen +
 				  data_left)) & 0x3;
-		ether_addr_copy(hdr_page->pos, ieee80211_get_DA(hdr));
-		hdr_page->pos += ETH_ALEN;
-		ether_addr_copy(hdr_page->pos, ieee80211_get_SA(hdr));
-		hdr_page->pos += ETH_ALEN;
+		ether_addr_copy(pos_hdr, ieee80211_get_DA(hdr));
+		pos_hdr += ETH_ALEN;
+		ether_addr_copy(pos_hdr, ieee80211_get_SA(hdr));
+		pos_hdr += ETH_ALEN;
 
 		length = snap_ip_tcp_hdrlen + data_left;
-		*((__be16 *)hdr_page->pos) = cpu_to_be16(length);
-		hdr_page->pos += sizeof(length);
+		*((__be16 *)pos_hdr) = cpu_to_be16(length);
+		pos_hdr += sizeof(length);
 
 		/*
 		 * This will copy the SNAP as well which will be considered
 		 * as MAC header.
 		 */
-		tso_build_hdr(skb, hdr_page->pos, &tso, data_left, !total_len);
+		tso_build_hdr(skb, pos_hdr, &tso, data_left, !total_len);
 
-		hdr_page->pos += snap_ip_tcp_hdrlen;
+		pos_hdr += snap_ip_tcp_hdrlen;
 
-		hdr_tb_len = hdr_page->pos - start_hdr;
+		hdr_tb_len = pos_hdr - start_hdr;
 		hdr_tb_phys = dma_map_single(trans->dev, start_hdr,
 					     hdr_tb_len, DMA_TO_DEVICE);
 		if (unlikely(dma_mapping_error(trans->dev, hdr_tb_phys)))
@@ -1847,10 +1849,10 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 		trace_iwlwifi_dev_tx_tb(trans->dev, skb, start_hdr,
 					hdr_tb_phys, hdr_tb_len);
 		/* add this subframe's headers' length to the tx_cmd */
-		le16_add_cpu(&tx_cmd->len, hdr_page->pos - subf_hdrs_start);
+		le16_add_cpu(&tx_cmd->len, pos_hdr - subf_hdrs_start);
 
 		/* prepare the start_hdr for the next subframe */
-		start_hdr = hdr_page->pos;
+		start_hdr = pos_hdr;
 
 		/* put the payload */
 		while (data_left) {
-- 
cgit v1.2.3-58-ga151


From 7f5e3038f02902a1409c9ad0f07e3b11c581c113 Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Wed, 3 Jul 2024 12:58:53 +0300
Subject: wifi: iwlwifi: map entire SKB when sending AMSDUs

This adds logic to map the entire SKB for AMSDUs. The required scatter
gather list is allocated together with the space for TSO headers.
Unmapping happens again when free'ing the TSO header page.

For now the mapping is unused, this will be changed in a later commit.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.96c6006f40ff.I55b74bc97c4026761397a7513a559c88a10b6489@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h     |   5 +-
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h |  10 +-
 drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c  |  20 ++--
 drivers/net/wireless/intel/iwlwifi/pcie/tx.c       | 114 +++++++++++++++++++--
 4 files changed, 128 insertions(+), 21 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index ecf94ec2c2b0..015f02122df6 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -120,6 +120,7 @@ enum CMD_MODE {
 	CMD_BLOCK_TXQS		= BIT(3),
 	CMD_SEND_IN_D3          = BIT(4),
 };
+#define CMD_MODE_BITS 5
 
 #define DEF_CMD_PAYLOAD_SIZE 320
 
@@ -712,7 +713,9 @@ struct iwl_dma_ptr {
 struct iwl_cmd_meta {
 	/* only for SYNC commands, iff the reply skb is wanted */
 	struct iwl_host_cmd *source;
-	u32 flags;
+	u32 flags: CMD_MODE_BITS;
+	/* sg_offset is valid if it is non-zero */
+	u32 sg_offset: PAGE_SHIFT;
 	u32 tbs;
 };
 
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index 754928062921..d63c1c284f70 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -623,9 +623,13 @@ void iwl_trans_pcie_tx_reset(struct iwl_trans *trans);
 int iwl_pcie_txq_alloc(struct iwl_trans *trans, struct iwl_txq *txq,
 		       int slots_num, bool cmd_queue);
 
-void *iwl_pcie_get_page_hdr(struct iwl_trans *trans, size_t len,
-			    struct sk_buff *skb);
-void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb);
+dma_addr_t iwl_pcie_get_sgt_tb_phys(struct sg_table *sgt, void *addr);
+struct sg_table *iwl_pcie_prep_tso(struct iwl_trans *trans, struct sk_buff *skb,
+				   struct iwl_cmd_meta *cmd_meta,
+				   u8 **hdr, unsigned int hdr_room);
+
+void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb,
+			    struct iwl_cmd_meta *cmd_meta);
 
 static inline dma_addr_t
 iwl_txq_get_first_tb_dma(struct iwl_txq *txq, int idx)
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
index 543efb52b084..3dcce6a8da50 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
@@ -119,7 +119,9 @@ trace:
 
 static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 				    struct sk_buff *skb,
-				    struct iwl_tfh_tfd *tfd, int start_len,
+				    struct iwl_tfh_tfd *tfd,
+				    struct iwl_cmd_meta *out_meta,
+				    int start_len,
 				    u8 hdr_len,
 				    struct iwl_device_tx_cmd *dev_cmd)
 {
@@ -130,6 +132,7 @@ static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 	unsigned int mss = skb_shinfo(skb)->gso_size;
 	u16 length, amsdu_pad;
 	u8 *start_hdr;
+	struct sg_table *sgt;
 	struct tso_t tso;
 
 	trace_iwlwifi_dev_tx(trans->dev, skb, tfd, sizeof(*tfd),
@@ -145,8 +148,8 @@ static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 		(3 + snap_ip_tcp_hdrlen + sizeof(struct ethhdr));
 
 	/* Our device supports 9 segments at most, it will fit in 1 page */
-	start_hdr = iwl_pcie_get_page_hdr(trans, hdr_room, skb);
-	if (!start_hdr)
+	sgt = iwl_pcie_prep_tso(trans, skb, out_meta, &start_hdr, hdr_room);
+	if (!sgt)
 		return -ENOMEM;
 
 	/*
@@ -287,8 +290,8 @@ iwl_tfh_tfd *iwl_txq_gen2_build_tx_amsdu(struct iwl_trans *trans,
 	 */
 	iwl_txq_gen2_set_tb(trans, tfd, tb_phys, len);
 
-	if (iwl_txq_gen2_build_amsdu(trans, skb, tfd, len + IWL_FIRST_TB_SIZE,
-				     hdr_len, dev_cmd))
+	if (iwl_txq_gen2_build_amsdu(trans, skb, tfd, out_meta,
+				     len + IWL_FIRST_TB_SIZE, hdr_len, dev_cmd))
 		goto out_err;
 
 	/* building the A-MSDU might have changed this data, memcpy it now */
@@ -719,7 +722,7 @@ int iwl_txq_gen2_tx(struct iwl_trans *trans, struct sk_buff *skb,
 
 	/* Set up first empty entry in queue's array of Tx/cmd buffers */
 	out_meta = &txq->entries[idx].meta;
-	out_meta->flags = 0;
+	memset(out_meta, 0, sizeof(*out_meta));
 
 	tfd = iwl_txq_gen2_build_tfd(trans, txq, dev_cmd, skb, out_meta);
 	if (!tfd) {
@@ -775,10 +778,11 @@ static void iwl_txq_gen2_unmap(struct iwl_trans *trans, int txq_id)
 
 		if (txq_id != trans_pcie->txqs.cmd.q_id) {
 			int idx = iwl_txq_get_cmd_index(txq, txq->read_ptr);
+			struct iwl_cmd_meta *cmd_meta = &txq->entries[idx].meta;
 			struct sk_buff *skb = txq->entries[idx].skb;
 
 			if (!WARN_ON_ONCE(!skb))
-				iwl_pcie_free_tso_page(trans, skb);
+				iwl_pcie_free_tso_page(trans, skb, cmd_meta);
 		}
 		iwl_txq_gen2_free_tfd(trans, txq);
 		txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr);
@@ -1247,7 +1251,7 @@ int iwl_pcie_gen2_enqueue_hcmd(struct iwl_trans *trans,
 	out_cmd = txq->entries[idx].cmd;
 	out_meta = &txq->entries[idx].meta;
 
-	/* re-initialize to NULL */
+	/* re-initialize, this also marks the SG list as unused */
 	memset(out_meta, 0, sizeof(*out_meta));
 	if (cmd->flags & CMD_WANT_SKB)
 		out_meta->source = cmd;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index d67ad40634f9..ac545a39ad2a 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -209,7 +209,8 @@ static void iwl_pcie_clear_cmd_in_flight(struct iwl_trans *trans)
 	spin_unlock(&trans_pcie->reg_lock);
 }
 
-void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb)
+void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb,
+			    struct iwl_cmd_meta *cmd_meta)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct page **page_ptr;
@@ -224,6 +225,17 @@ void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb)
 
 		next = *(void **)((u8 *)page_address(next) + PAGE_SIZE -
 				  sizeof(void *));
+
+		/* Unmap the scatter gather list that is on the last page */
+		if (!next && cmd_meta->sg_offset) {
+			struct sg_table *sgt;
+
+			sgt = (void *)((u8 *)page_address(tmp) +
+				       cmd_meta->sg_offset);
+
+			dma_unmap_sgtable(trans->dev, sgt, DMA_TO_DEVICE, 0);
+		}
+
 		__free_page(tmp);
 	}
 }
@@ -363,11 +375,13 @@ static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id)
 
 		if (txq_id != trans_pcie->txqs.cmd.q_id) {
 			struct sk_buff *skb = txq->entries[txq->read_ptr].skb;
+			struct iwl_cmd_meta *cmd_meta =
+				&txq->entries[txq->read_ptr].meta;
 
 			if (WARN_ON_ONCE(!skb))
 				continue;
 
-			iwl_pcie_free_tso_page(trans, skb);
+			iwl_pcie_free_tso_page(trans, skb, cmd_meta);
 		}
 		iwl_txq_free_tfd(trans, txq);
 		txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr);
@@ -1420,7 +1434,8 @@ int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans,
 	out_cmd = txq->entries[idx].cmd;
 	out_meta = &txq->entries[idx].meta;
 
-	memset(out_meta, 0, sizeof(*out_meta));	/* re-initialize to NULL */
+	/* re-initialize, this also marks the SG list as unused */
+	memset(out_meta, 0, sizeof(*out_meta));
 	if (cmd->flags & CMD_WANT_SKB)
 		out_meta->source = cmd;
 
@@ -1702,8 +1717,8 @@ static int iwl_fill_data_tbs(struct iwl_trans *trans, struct sk_buff *skb,
 }
 
 #ifdef CONFIG_INET
-void *iwl_pcie_get_page_hdr(struct iwl_trans *trans,
-			    size_t len, struct sk_buff *skb)
+static void *iwl_pcie_get_page_hdr(struct iwl_trans *trans,
+				   size_t len, struct sk_buff *skb)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_tso_hdr_page *p = this_cpu_ptr(trans_pcie->txqs.tso_hdr_page);
@@ -1751,6 +1766,84 @@ out:
 	return ret;
 }
 
+/**
+ * iwl_pcie_get_sgt_tb_phys - Find TB address in mapped SG list
+ * @sgt: scatter gather table
+ * @addr: Virtual address
+ *
+ * Find the entry that includes the address for the given address and return
+ * correct physical address for the TB entry.
+ *
+ * Returns: Address for TB entry
+ */
+dma_addr_t iwl_pcie_get_sgt_tb_phys(struct sg_table *sgt, void *addr)
+{
+	struct scatterlist *sg;
+	int i;
+
+	for_each_sgtable_dma_sg(sgt, sg, i) {
+		if (addr >= sg_virt(sg) &&
+		    (u8 *)addr < (u8 *)sg_virt(sg) + sg_dma_len(sg))
+			return sg_dma_address(sg) +
+			       ((unsigned long)addr - (unsigned long)sg_virt(sg));
+	}
+
+	WARN_ON_ONCE(1);
+
+	return DMA_MAPPING_ERROR;
+}
+
+/**
+ * iwl_pcie_prep_tso - Prepare TSO page and SKB for sending
+ * @trans: transport private data
+ * @skb: the SKB to map
+ * @cmd_meta: command meta to store the scatter list information for unmapping
+ * @hdr: output argument for TSO headers
+ * @hdr_room: requested length for TSO headers
+ *
+ * Allocate space for a scatter gather list and TSO headers and map the SKB
+ * using the scatter gather list. The SKB is unmapped again when the page is
+ * free'ed again at the end of the operation.
+ *
+ * Returns: newly allocated and mapped scatter gather table with list
+ */
+struct sg_table *iwl_pcie_prep_tso(struct iwl_trans *trans, struct sk_buff *skb,
+				   struct iwl_cmd_meta *cmd_meta,
+				   u8 **hdr, unsigned int hdr_room)
+{
+	struct sg_table *sgt;
+
+	if (WARN_ON_ONCE(skb_has_frag_list(skb)))
+		return NULL;
+
+	*hdr = iwl_pcie_get_page_hdr(trans,
+				     hdr_room + __alignof__(struct sg_table) +
+				     sizeof(struct sg_table) +
+				     (skb_shinfo(skb)->nr_frags + 1) *
+				     sizeof(struct scatterlist),
+				     skb);
+	if (!*hdr)
+		return NULL;
+
+	sgt = (void *)PTR_ALIGN(*hdr + hdr_room, __alignof__(struct sg_table));
+	sgt->sgl = (void *)(sgt + 1);
+
+	sg_init_table(sgt->sgl, skb_shinfo(skb)->nr_frags + 1);
+
+	sgt->orig_nents = skb_to_sgvec(skb, sgt->sgl, 0, skb->len);
+	if (WARN_ON_ONCE(sgt->orig_nents <= 0))
+		return NULL;
+
+	/* And map the entire SKB */
+	if (dma_map_sgtable(trans->dev, sgt, DMA_TO_DEVICE, 0) < 0)
+		return NULL;
+
+	/* Store non-zero (i.e. valid) offset for unmapping */
+	cmd_meta->sg_offset = (unsigned long) sgt & ~PAGE_MASK;
+
+	return sgt;
+}
+
 static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 				   struct iwl_txq *txq, u8 hdr_len,
 				   struct iwl_cmd_meta *out_meta,
@@ -1764,6 +1857,7 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 	unsigned int mss = skb_shinfo(skb)->gso_size;
 	u16 length, iv_len, amsdu_pad;
 	u8 *start_hdr, *pos_hdr;
+	struct sg_table *sgt;
 	struct tso_t tso;
 
 	/* if the packet is protected, then it must be CCMP or GCMP */
@@ -1786,10 +1880,11 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 		(3 + snap_ip_tcp_hdrlen + sizeof(struct ethhdr)) + iv_len;
 
 	/* Our device supports 9 segments at most, it will fit in 1 page */
-	pos_hdr = start_hdr = iwl_pcie_get_page_hdr(trans, hdr_room, skb);
-	if (!start_hdr)
+	sgt = iwl_pcie_prep_tso(trans, skb, out_meta, &start_hdr, hdr_room);
+	if (!sgt)
 		return -ENOMEM;
 
+	pos_hdr = start_hdr;
 	memcpy(pos_hdr, skb->data + hdr_len, iv_len);
 	pos_hdr += iv_len;
 
@@ -2029,7 +2124,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb,
 
 	/* Set up first empty entry in queue's array of Tx/cmd buffers */
 	out_meta = &txq->entries[txq->write_ptr].meta;
-	out_meta->flags = 0;
+	memset(out_meta, 0, sizeof(*out_meta));
 
 	/*
 	 * The second TB (tb1) points to the remainder of the TX command
@@ -2228,13 +2323,14 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 	     read_ptr != tfd_num;
 	     txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr),
 	     read_ptr = iwl_txq_get_cmd_index(txq, txq->read_ptr)) {
+		struct iwl_cmd_meta *cmd_meta = &txq->entries[read_ptr].meta;
 		struct sk_buff *skb = txq->entries[read_ptr].skb;
 
 		if (WARN_ONCE(!skb, "no SKB at %d (%d) on queue %d\n",
 			      read_ptr, txq->read_ptr, txq_id))
 			continue;
 
-		iwl_pcie_free_tso_page(trans, skb);
+		iwl_pcie_free_tso_page(trans, skb, cmd_meta);
 
 		__skb_queue_tail(skbs, skb);
 
-- 
cgit v1.2.3-58-ga151


From adc902ceada26add77ad75426805e973a7c67f01 Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Wed, 3 Jul 2024 12:58:54 +0300
Subject: wifi: iwlwifi: keep the TSO and workaround pages mapped

Map the pages when allocating them so that we will not need to map each
of the used fragments at a later point.

For now the mapping is not used, this will be changed in a later commit.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://patch.msgid.link/20240703125541.7ced468fe431.Ibb109867dc680c37fe8d891e9ab9ef64ed5c5d2d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h | 30 ++++++++++-
 drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c  | 22 ++++++--
 drivers/net/wireless/intel/iwlwifi/pcie/tx.c       | 60 +++++++++++++++++-----
 3 files changed, 95 insertions(+), 17 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index d63c1c284f70..b59de4f80b4b 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -603,6 +603,22 @@ struct iwl_tso_hdr_page {
 	u8 *pos;
 };
 
+/*
+ * Note that we put this struct *last* in the page. By doing that, we ensure
+ * that no TB referencing this page can trigger the 32-bit boundary hardware
+ * bug.
+ */
+struct iwl_tso_page_info {
+	dma_addr_t dma_addr;
+	struct page *next;
+	refcount_t use_count;
+};
+
+#define IWL_TSO_PAGE_DATA_SIZE	(PAGE_SIZE - sizeof(struct iwl_tso_page_info))
+#define IWL_TSO_PAGE_INFO(addr)	\
+	((struct iwl_tso_page_info *)(((unsigned long)addr & PAGE_MASK) + \
+				      IWL_TSO_PAGE_DATA_SIZE))
+
 int iwl_pcie_tx_init(struct iwl_trans *trans);
 void iwl_pcie_tx_start(struct iwl_trans *trans, u32 scd_base_addr);
 int iwl_pcie_tx_stop(struct iwl_trans *trans);
@@ -628,8 +644,18 @@ struct sg_table *iwl_pcie_prep_tso(struct iwl_trans *trans, struct sk_buff *skb,
 				   struct iwl_cmd_meta *cmd_meta,
 				   u8 **hdr, unsigned int hdr_room);
 
-void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb,
-			    struct iwl_cmd_meta *cmd_meta);
+void iwl_pcie_free_tso_pages(struct iwl_trans *trans, struct sk_buff *skb,
+			     struct iwl_cmd_meta *cmd_meta);
+
+static inline dma_addr_t iwl_pcie_get_tso_page_phys(void *addr)
+{
+	dma_addr_t res;
+
+	res = IWL_TSO_PAGE_INFO(addr)->dma_addr;
+	res += (unsigned long)addr & ~PAGE_MASK;
+
+	return res;
+}
 
 static inline dma_addr_t
 iwl_txq_get_first_tb_dma(struct iwl_txq *txq, int idx)
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
index 3dcce6a8da50..10ee2c328458 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
@@ -19,8 +19,10 @@ static struct page *get_workaround_page(struct iwl_trans *trans,
 					struct sk_buff *skb)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_tso_page_info *info;
 	struct page **page_ptr;
 	struct page *ret;
+	dma_addr_t phys;
 
 	page_ptr = (void *)((u8 *)skb->cb + trans_pcie->txqs.page_offs);
 
@@ -28,8 +30,22 @@ static struct page *get_workaround_page(struct iwl_trans *trans,
 	if (!ret)
 		return NULL;
 
+	info = IWL_TSO_PAGE_INFO(page_address(ret));
+
+	/* Create a DMA mapping for the page */
+	phys = dma_map_page_attrs(trans->dev, ret, 0, PAGE_SIZE,
+				  DMA_TO_DEVICE, DMA_ATTR_SKIP_CPU_SYNC);
+	if (unlikely(dma_mapping_error(trans->dev, phys))) {
+		__free_page(ret);
+		return NULL;
+	}
+
+	/* Store physical address and set use count */
+	info->dma_addr = phys;
+	refcount_set(&info->use_count, 1);
+
 	/* set the chaining pointer to the previous page if there */
-	*(void **)((u8 *)page_address(ret) + PAGE_SIZE - sizeof(void *)) = *page_ptr;
+	info->next = *page_ptr;
 	*page_ptr = ret;
 
 	return ret;
@@ -76,7 +92,7 @@ static int iwl_txq_gen2_set_tb_with_wa(struct iwl_trans *trans,
 	 * a new mapping for it so the device will not fail.
 	 */
 
-	if (WARN_ON(len > PAGE_SIZE - sizeof(void *))) {
+	if (WARN_ON(len > IWL_TSO_PAGE_DATA_SIZE)) {
 		ret = -ENOBUFS;
 		goto unmap;
 	}
@@ -782,7 +798,7 @@ static void iwl_txq_gen2_unmap(struct iwl_trans *trans, int txq_id)
 			struct sk_buff *skb = txq->entries[idx].skb;
 
 			if (!WARN_ON_ONCE(!skb))
-				iwl_pcie_free_tso_page(trans, skb, cmd_meta);
+				iwl_pcie_free_tso_pages(trans, skb, cmd_meta);
 		}
 		iwl_txq_gen2_free_tfd(trans, txq);
 		txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index ac545a39ad2a..e00d85866de9 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -209,8 +209,22 @@ static void iwl_pcie_clear_cmd_in_flight(struct iwl_trans *trans)
 	spin_unlock(&trans_pcie->reg_lock);
 }
 
-void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb,
-			    struct iwl_cmd_meta *cmd_meta)
+static void iwl_pcie_free_and_unmap_tso_page(struct iwl_trans *trans,
+					     struct page *page)
+{
+	struct iwl_tso_page_info *info = IWL_TSO_PAGE_INFO(page_address(page));
+
+	/* Decrease internal use count and unmap/free page if needed */
+	if (refcount_dec_and_test(&info->use_count)) {
+		dma_unmap_page(trans->dev, info->dma_addr, PAGE_SIZE,
+			       DMA_TO_DEVICE);
+
+		__free_page(page);
+	}
+}
+
+void iwl_pcie_free_tso_pages(struct iwl_trans *trans, struct sk_buff *skb,
+			     struct iwl_cmd_meta *cmd_meta)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct page **page_ptr;
@@ -221,10 +235,11 @@ void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb,
 	*page_ptr = NULL;
 
 	while (next) {
+		struct iwl_tso_page_info *info;
 		struct page *tmp = next;
 
-		next = *(void **)((u8 *)page_address(next) + PAGE_SIZE -
-				  sizeof(void *));
+		info = IWL_TSO_PAGE_INFO(page_address(next));
+		next = info->next;
 
 		/* Unmap the scatter gather list that is on the last page */
 		if (!next && cmd_meta->sg_offset) {
@@ -236,7 +251,7 @@ void iwl_pcie_free_tso_page(struct iwl_trans *trans, struct sk_buff *skb,
 			dma_unmap_sgtable(trans->dev, sgt, DMA_TO_DEVICE, 0);
 		}
 
-		__free_page(tmp);
+		iwl_pcie_free_and_unmap_tso_page(trans, tmp);
 	}
 }
 
@@ -381,7 +396,7 @@ static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id)
 			if (WARN_ON_ONCE(!skb))
 				continue;
 
-			iwl_pcie_free_tso_page(trans, skb, cmd_meta);
+			iwl_pcie_free_tso_pages(trans, skb, cmd_meta);
 		}
 		iwl_txq_free_tfd(trans, txq);
 		txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr);
@@ -1722,7 +1737,9 @@ static void *iwl_pcie_get_page_hdr(struct iwl_trans *trans,
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_tso_hdr_page *p = this_cpu_ptr(trans_pcie->txqs.tso_hdr_page);
+	struct iwl_tso_page_info *info;
 	struct page **page_ptr;
+	dma_addr_t phys;
 	void *ret;
 
 	page_ptr = (void *)((u8 *)skb->cb + trans_pcie->txqs.page_offs);
@@ -1743,23 +1760,42 @@ static void *iwl_pcie_get_page_hdr(struct iwl_trans *trans,
 	 *
 	 * (see also get_workaround_page() in tx-gen2.c)
 	 */
-	if (p->pos + len < (u8 *)page_address(p->page) + PAGE_SIZE -
-			   sizeof(void *))
+	if (((unsigned long)p->pos & ~PAGE_MASK) + len < IWL_TSO_PAGE_DATA_SIZE) {
+		info = IWL_TSO_PAGE_INFO(page_address(ret));
 		goto out;
+	}
 
 	/* We don't have enough room on this page, get a new one. */
-	__free_page(p->page);
+	iwl_pcie_free_and_unmap_tso_page(trans, p->page);
 
 alloc:
 	p->page = alloc_page(GFP_ATOMIC);
 	if (!p->page)
 		return NULL;
 	p->pos = page_address(p->page);
+
+	info = IWL_TSO_PAGE_INFO(page_address(ret));
+
 	/* set the chaining pointer to NULL */
-	*(void **)((u8 *)page_address(p->page) + PAGE_SIZE - sizeof(void *)) = NULL;
+	info->next = NULL;
+
+	/* Create a DMA mapping for the page */
+	phys = dma_map_page_attrs(trans->dev, p->page, 0, PAGE_SIZE,
+				  DMA_TO_DEVICE, DMA_ATTR_SKIP_CPU_SYNC);
+	if (unlikely(dma_mapping_error(trans->dev, phys))) {
+		__free_page(p->page);
+		p->page = NULL;
+
+		return NULL;
+	}
+
+	/* Store physical address and set use count */
+	info->dma_addr = phys;
+	refcount_set(&info->use_count, 1);
 out:
 	*page_ptr = p->page;
-	get_page(p->page);
+	/* Return an internal reference for the caller */
+	refcount_inc(&info->use_count);
 	ret = p->pos;
 	p->pos += len;
 
@@ -2330,7 +2366,7 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 			      read_ptr, txq->read_ptr, txq_id))
 			continue;
 
-		iwl_pcie_free_tso_page(trans, skb, cmd_meta);
+		iwl_pcie_free_tso_pages(trans, skb, cmd_meta);
 
 		__skb_queue_tail(skbs, skb);
 
-- 
cgit v1.2.3-58-ga151


From 90db50755228252f94b143bbf2380242688b1104 Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Wed, 3 Jul 2024 12:58:55 +0300
Subject: wifi: iwlwifi: use already mapped data when TXing an AMSDU

The previous commits added mappings for the SKB and TSO page. This
switches the code to use these mappings instead of creating new ones.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Link: https://patch.msgid.link/20240703125541.35d89c5e4ae8.I4feb8d34e7b30768d21365ec22c944bacc274d0b@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c | 64 +++++++++++++++++------
 drivers/net/wireless/intel/iwlwifi/pcie/tx.c      | 21 +++++---
 2 files changed, 63 insertions(+), 22 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
index 10ee2c328458..21c3998a76c8 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
@@ -61,7 +61,8 @@ static int iwl_txq_gen2_set_tb_with_wa(struct iwl_trans *trans,
 				       struct sk_buff *skb,
 				       struct iwl_tfh_tfd *tfd,
 				       dma_addr_t phys, void *virt,
-				       u16 len, struct iwl_cmd_meta *meta)
+				       u16 len, struct iwl_cmd_meta *meta,
+				       bool unmap)
 {
 	dma_addr_t oldphys = phys;
 	struct page *page;
@@ -105,10 +106,27 @@ static int iwl_txq_gen2_set_tb_with_wa(struct iwl_trans *trans,
 
 	memcpy(page_address(page), virt, len);
 
-	phys = dma_map_single(trans->dev, page_address(page), len,
-			      DMA_TO_DEVICE);
-	if (unlikely(dma_mapping_error(trans->dev, phys)))
-		return -ENOMEM;
+	/*
+	 * This is a bit odd, but performance does not matter here, what
+	 * matters are the expectations of the calling code and TB cleanup
+	 * function.
+	 *
+	 * As such, if unmap is set, then create another mapping for the TB
+	 * entry as it will be unmapped later. On the other hand, if it is not
+	 * set, then the TB entry will not be unmapped and instead we simply
+	 * reference and sync the mapping that get_workaround_page() created.
+	 */
+	if (unmap) {
+		phys = dma_map_single(trans->dev, page_address(page), len,
+				      DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(trans->dev, phys)))
+			return -ENOMEM;
+	} else {
+		phys = iwl_pcie_get_tso_page_phys(page_address(page));
+		dma_sync_single_for_device(trans->dev, phys, len,
+					   DMA_TO_DEVICE);
+	}
+
 	ret = iwl_txq_gen2_set_tb(trans, tfd, phys, len);
 	if (ret < 0) {
 		/* unmap the new allocation as single */
@@ -116,6 +134,7 @@ static int iwl_txq_gen2_set_tb_with_wa(struct iwl_trans *trans,
 		meta = NULL;
 		goto unmap;
 	}
+
 	IWL_DEBUG_TX(trans,
 		     "TB bug workaround: copied %d bytes from 0x%llx to 0x%llx\n",
 		     len, (unsigned long long)oldphys,
@@ -123,6 +142,9 @@ static int iwl_txq_gen2_set_tb_with_wa(struct iwl_trans *trans,
 
 	ret = 0;
 unmap:
+	if (!unmap)
+		goto trace;
+
 	if (meta)
 		dma_unmap_page(trans->dev, oldphys, len, DMA_TO_DEVICE);
 	else
@@ -146,6 +168,7 @@ static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 	struct ieee80211_hdr *hdr = (void *)skb->data;
 	unsigned int snap_ip_tcp_hdrlen, ip_hdrlen, total_len, hdr_room;
 	unsigned int mss = skb_shinfo(skb)->gso_size;
+	dma_addr_t start_hdr_phys;
 	u16 length, amsdu_pad;
 	u8 *start_hdr;
 	struct sg_table *sgt;
@@ -168,6 +191,8 @@ static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 	if (!sgt)
 		return -ENOMEM;
 
+	start_hdr_phys = iwl_pcie_get_tso_page_phys(start_hdr);
+
 	/*
 	 * Pull the ieee80211 header to be able to use TSO core,
 	 * we will restore it for the tx_status flow.
@@ -214,10 +239,8 @@ static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 		pos_hdr += snap_ip_tcp_hdrlen;
 
 		tb_len = pos_hdr - start_hdr;
-		tb_phys = dma_map_single(trans->dev, start_hdr,
-					 tb_len, DMA_TO_DEVICE);
-		if (unlikely(dma_mapping_error(trans->dev, tb_phys)))
-			goto out_err;
+		tb_phys = iwl_pcie_get_tso_page_phys(start_hdr);
+
 		/*
 		 * No need for _with_wa, this is from the TSO page and
 		 * we leave some space at the end of it so can't hit
@@ -237,11 +260,14 @@ static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 			int ret;
 
 			tb_len = min_t(unsigned int, tso.size, data_left);
-			tb_phys = dma_map_single(trans->dev, tso.data,
-						 tb_len, DMA_TO_DEVICE);
+			tb_phys = iwl_pcie_get_sgt_tb_phys(sgt, tso.data);
+			/* Not a real mapping error, use direct comparison */
+			if (unlikely(tb_phys == DMA_MAPPING_ERROR))
+				goto out_err;
+
 			ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd,
 							  tb_phys, tso.data,
-							  tb_len, NULL);
+							  tb_len, NULL, false);
 			if (ret)
 				goto out_err;
 
@@ -250,6 +276,9 @@ static int iwl_txq_gen2_build_amsdu(struct iwl_trans *trans,
 		}
 	}
 
+	dma_sync_single_for_device(trans->dev, start_hdr_phys, hdr_room,
+				   DMA_TO_DEVICE);
+
 	/* re -add the WiFi header */
 	skb_push(skb, hdr_len);
 
@@ -339,7 +368,7 @@ static int iwl_txq_gen2_tx_add_frags(struct iwl_trans *trans,
 					   fragsz, DMA_TO_DEVICE);
 		ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd, tb_phys,
 						  skb_frag_address(frag),
-						  fragsz, out_meta);
+						  fragsz, out_meta, true);
 		if (ret)
 			return ret;
 	}
@@ -413,7 +442,7 @@ iwl_tfh_tfd *iwl_txq_gen2_build_tx(struct iwl_trans *trans,
 					 tb2_len, DMA_TO_DEVICE);
 		ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd, tb_phys,
 						  skb->data + hdr_len, tb2_len,
-						  NULL);
+						  NULL, true);
 		if (ret)
 			goto out_err;
 	}
@@ -428,7 +457,8 @@ iwl_tfh_tfd *iwl_txq_gen2_build_tx(struct iwl_trans *trans,
 					 skb_headlen(frag), DMA_TO_DEVICE);
 		ret = iwl_txq_gen2_set_tb_with_wa(trans, skb, tfd, tb_phys,
 						  frag->data,
-						  skb_headlen(frag), NULL);
+						  skb_headlen(frag), NULL,
+						  true);
 		if (ret)
 			goto out_err;
 		if (iwl_txq_gen2_tx_add_frags(trans, frag, tfd, out_meta))
@@ -623,6 +653,10 @@ void iwl_txq_gen2_tfd_unmap(struct iwl_trans *trans,
 		return;
 	}
 
+	/* TB1 is mapped directly, the rest is the TSO page and SG list. */
+	if (meta->sg_offset)
+		num_tbs = 2;
+
 	/* first TB is never freed - it's the bidirectional DMA data */
 	for (i = 1; i < num_tbs; i++) {
 		if (meta->tbs & BIT(i))
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index e00d85866de9..8afb5fc1972e 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -303,6 +303,10 @@ static void iwl_txq_gen1_tfd_unmap(struct iwl_trans *trans,
 		return;
 	}
 
+	/* TB1 is mapped directly, the rest is the TSO page and SG list. */
+	if (meta->sg_offset)
+		num_tbs = 2;
+
 	/* first TB is never freed - it's the bidirectional DMA data */
 
 	for (i = 1; i < num_tbs; i++) {
@@ -1892,6 +1896,7 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 	unsigned int snap_ip_tcp_hdrlen, ip_hdrlen, total_len, hdr_room;
 	unsigned int mss = skb_shinfo(skb)->gso_size;
 	u16 length, iv_len, amsdu_pad;
+	dma_addr_t start_hdr_phys;
 	u8 *start_hdr, *pos_hdr;
 	struct sg_table *sgt;
 	struct tso_t tso;
@@ -1920,6 +1925,7 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 	if (!sgt)
 		return -ENOMEM;
 
+	start_hdr_phys = iwl_pcie_get_tso_page_phys(start_hdr);
 	pos_hdr = start_hdr;
 	memcpy(pos_hdr, skb->data + hdr_len, iv_len);
 	pos_hdr += iv_len;
@@ -1971,10 +1977,8 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 		pos_hdr += snap_ip_tcp_hdrlen;
 
 		hdr_tb_len = pos_hdr - start_hdr;
-		hdr_tb_phys = dma_map_single(trans->dev, start_hdr,
-					     hdr_tb_len, DMA_TO_DEVICE);
-		if (unlikely(dma_mapping_error(trans->dev, hdr_tb_phys)))
-			return -EINVAL;
+		hdr_tb_phys = iwl_pcie_get_tso_page_phys(start_hdr);
+
 		iwl_pcie_txq_build_tfd(trans, txq, hdr_tb_phys,
 				       hdr_tb_len, false);
 		trace_iwlwifi_dev_tx_tb(trans->dev, skb, start_hdr,
@@ -1991,9 +1995,9 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 						  data_left);
 			dma_addr_t tb_phys;
 
-			tb_phys = dma_map_single(trans->dev, tso.data,
-						 size, DMA_TO_DEVICE);
-			if (unlikely(dma_mapping_error(trans->dev, tb_phys)))
+			tb_phys = iwl_pcie_get_sgt_tb_phys(sgt, tso.data);
+			/* Not a real mapping error, use direct comparison */
+			if (unlikely(tb_phys == DMA_MAPPING_ERROR))
 				return -EINVAL;
 
 			iwl_pcie_txq_build_tfd(trans, txq, tb_phys,
@@ -2006,6 +2010,9 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb,
 		}
 	}
 
+	dma_sync_single_for_device(trans->dev, start_hdr_phys, hdr_room,
+				   DMA_TO_DEVICE);
+
 	/* re -add the WiFi header and IV */
 	skb_push(skb, hdr_len + iv_len);
 
-- 
cgit v1.2.3-58-ga151


From 1a3364e963989dbcd1465dd8e94d9c5babd3474a Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Wed, 3 Jul 2024 12:58:56 +0300
Subject: wifi: iwlwifi: keep BHs disabled when TXing from reclaim

During reclaim, we may release the txq->lock spinlock in order to call
iwl_trans_tx to queue new frames. The iwl_trans_tx function expects to
be called with BHs disabled and iwl_pcie_reclaim is most of the times
called with BHs disabled already. However, reclaim can also happen after
flushing a STA and in that case BHs will not be disabled.

Solve this corner case by only releasing the spinlock but keeping BHs
disabled.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.5d12e0e54e9f.Ic53a7ff75f1163eb38bdcf5d66b503e91e6ce5ca@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/tx.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index 8afb5fc1972e..1f6db6b90f6f 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -2412,7 +2412,7 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 		 * have tx as well. Bottom line, we can unlock and re-lock
 		 * later.
 		 */
-		spin_unlock_bh(&txq->lock);
+		spin_unlock(&txq->lock);
 
 		while ((skb = __skb_dequeue(&overflow_skbs))) {
 			struct iwl_device_tx_cmd *dev_cmd_ptr;
@@ -2431,7 +2431,7 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 		if (iwl_txq_space(trans, txq) > txq->low_mark)
 			iwl_trans_pcie_wake_queue(trans, txq);
 
-		spin_lock_bh(&txq->lock);
+		spin_lock(&txq->lock);
 		txq->overflow_tx = false;
 	}
 
-- 
cgit v1.2.3-58-ga151


From a2ed933dfefac8976ac915e1b1b60028b9c344b3 Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Wed, 3 Jul 2024 12:58:57 +0300
Subject: wifi: iwlwifi: release TXQ lock during reclaim

Much of the work during reclaim can be done without holding the TXQ
lock and releasing the lock means that command submission can happen at
the same time.

Add a new reclaim_lock to prevent parallel cleanup. Release the lock
while working with an internal copy of the txq->read_ptr and only take
the lock again when updating the read pointer after the cleanup is done.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Reviewed-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.2a81021d49ac.I53698ae92fb75a0461d41176db115462cf8be1cd@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-trans.h    |  3 +
 drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c |  6 +-
 drivers/net/wireless/intel/iwlwifi/pcie/tx.c      | 78 ++++++++++++++---------
 3 files changed, 54 insertions(+), 33 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
index 015f02122df6..6148acbac6af 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h
@@ -752,6 +752,7 @@ struct iwl_pcie_first_tb_buf {
  * @first_tb_dma: DMA address for the first_tb_bufs start
  * @entries: transmit entries (driver state)
  * @lock: queue lock
+ * @reclaim_lock: reclaim lock
  * @stuck_timer: timer that fires if queue gets stuck
  * @trans: pointer back to transport (for timer)
  * @need_update: indicates need to update read/write index
@@ -794,6 +795,8 @@ struct iwl_txq {
 	struct iwl_pcie_txq_entry *entries;
 	/* lock for syncing changes on the queue */
 	spinlock_t lock;
+	/* lock to prevent concurrent reclaim */
+	spinlock_t reclaim_lock;
 	unsigned long frozen_expiry_remainder;
 	struct timer_list stuck_timer;
 	struct iwl_trans *trans;
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
index 21c3998a76c8..2e780fb2da42 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
@@ -821,7 +821,8 @@ static void iwl_txq_gen2_unmap(struct iwl_trans *trans, int txq_id)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
 
-	spin_lock_bh(&txq->lock);
+	spin_lock_bh(&txq->reclaim_lock);
+	spin_lock(&txq->lock);
 	while (txq->write_ptr != txq->read_ptr) {
 		IWL_DEBUG_TX_REPLY(trans, "Q %d Free %d\n",
 				   txq_id, txq->read_ptr);
@@ -844,7 +845,8 @@ static void iwl_txq_gen2_unmap(struct iwl_trans *trans, int txq_id)
 		iwl_op_mode_free_skb(trans->op_mode, skb);
 	}
 
-	spin_unlock_bh(&txq->lock);
+	spin_unlock(&txq->lock);
+	spin_unlock_bh(&txq->reclaim_lock);
 
 	/* just in case - this queue may have been stopped */
 	iwl_trans_pcie_wake_queue(trans, txq);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index 1f6db6b90f6f..748772fa6b3e 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -333,20 +333,21 @@ static void iwl_txq_gen1_tfd_unmap(struct iwl_trans *trans,
  * iwl_txq_free_tfd - Free all chunks referenced by TFD [txq->q.read_ptr]
  * @trans: transport private data
  * @txq: tx queue
+ * @read_ptr: the TXQ read_ptr to free
  *
  * Does NOT advance any TFD circular buffer read/write indexes
  * Does NOT free the TFD itself (which is within circular buffer)
  */
-static void iwl_txq_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq)
+static void iwl_txq_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq,
+			     int read_ptr)
 {
 	/* rd_ptr is bounded by TFD_QUEUE_SIZE_MAX and
 	 * idx is bounded by n_window
 	 */
-	int rd_ptr = txq->read_ptr;
-	int idx = iwl_txq_get_cmd_index(txq, rd_ptr);
+	int idx = iwl_txq_get_cmd_index(txq, read_ptr);
 	struct sk_buff *skb;
 
-	lockdep_assert_held(&txq->lock);
+	lockdep_assert_held(&txq->reclaim_lock);
 
 	if (!txq->entries)
 		return;
@@ -356,10 +357,10 @@ static void iwl_txq_free_tfd(struct iwl_trans *trans, struct iwl_txq *txq)
 	 */
 	if (trans->trans_cfg->gen2)
 		iwl_txq_gen2_tfd_unmap(trans, &txq->entries[idx].meta,
-				       iwl_txq_get_tfd(trans, txq, rd_ptr));
+				       iwl_txq_get_tfd(trans, txq, read_ptr));
 	else
 		iwl_txq_gen1_tfd_unmap(trans, &txq->entries[idx].meta,
-				       txq, rd_ptr);
+				       txq, read_ptr);
 
 	/* free SKB */
 	skb = txq->entries[idx].skb;
@@ -387,7 +388,8 @@ static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id)
 		return;
 	}
 
-	spin_lock_bh(&txq->lock);
+	spin_lock_bh(&txq->reclaim_lock);
+	spin_lock(&txq->lock);
 	while (txq->write_ptr != txq->read_ptr) {
 		IWL_DEBUG_TX_REPLY(trans, "Q %d Free %d\n",
 				   txq_id, txq->read_ptr);
@@ -402,7 +404,7 @@ static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id)
 
 			iwl_pcie_free_tso_pages(trans, skb, cmd_meta);
 		}
-		iwl_txq_free_tfd(trans, txq);
+		iwl_txq_free_tfd(trans, txq, txq->read_ptr);
 		txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr);
 
 		if (txq->read_ptr == txq->write_ptr &&
@@ -416,7 +418,8 @@ static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id)
 		iwl_op_mode_free_skb(trans->op_mode, skb);
 	}
 
-	spin_unlock_bh(&txq->lock);
+	spin_unlock(&txq->lock);
+	spin_unlock_bh(&txq->reclaim_lock);
 
 	/* just in case - this queue may have been stopped */
 	iwl_trans_pcie_wake_queue(trans, txq);
@@ -921,6 +924,7 @@ int iwl_txq_init(struct iwl_trans *trans, struct iwl_txq *txq,
 		return ret;
 
 	spin_lock_init(&txq->lock);
+	spin_lock_init(&txq->reclaim_lock);
 
 	if (cmd_queue) {
 		static struct lock_class_key iwl_txq_cmd_queue_lock_class;
@@ -1055,11 +1059,12 @@ static void iwl_txq_progress(struct iwl_txq *txq)
 		mod_timer(&txq->stuck_timer, jiffies + txq->wd_timeout);
 }
 
-static inline bool iwl_txq_used(const struct iwl_txq *q, int i)
+static inline bool iwl_txq_used(const struct iwl_txq *q, int i,
+				int read_ptr, int write_ptr)
 {
 	int index = iwl_txq_get_cmd_index(q, i);
-	int r = iwl_txq_get_cmd_index(q, q->read_ptr);
-	int w = iwl_txq_get_cmd_index(q, q->write_ptr);
+	int r = iwl_txq_get_cmd_index(q, read_ptr);
+	int w = iwl_txq_get_cmd_index(q, write_ptr);
 
 	return w >= r ?
 		(index >= r && index < w) :
@@ -1086,7 +1091,7 @@ static void iwl_pcie_cmdq_reclaim(struct iwl_trans *trans, int txq_id, int idx)
 	r = iwl_txq_get_cmd_index(txq, txq->read_ptr);
 
 	if (idx >= trans->trans_cfg->base_params->max_tfd_queue_size ||
-	    (!iwl_txq_used(txq, idx))) {
+	    (!iwl_txq_used(txq, idx, txq->read_ptr, txq->write_ptr))) {
 		WARN_ONCE(test_bit(txq_id, trans_pcie->txqs.queue_used),
 			  "%s: Read index for DMA queue txq id (%d), index %d is out of range [0-%d] %d %d.\n",
 			  __func__, txq_id, idx,
@@ -2284,12 +2289,12 @@ out_err:
 }
 
 static void iwl_txq_gen1_inval_byte_cnt_tbl(struct iwl_trans *trans,
-					    struct iwl_txq *txq)
+					    struct iwl_txq *txq,
+					    int read_ptr)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwlagn_scd_bc_tbl *scd_bc_tbl = trans_pcie->txqs.scd_bc_tbls.addr;
 	int txq_id = txq->id;
-	int read_ptr = txq->read_ptr;
 	u8 sta_id = 0;
 	__le16 bc_ent;
 	struct iwl_device_tx_cmd *dev_cmd = txq->entries[read_ptr].cmd;
@@ -2316,6 +2321,7 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_txq *txq = trans_pcie->txqs.txq[txq_id];
 	int tfd_num, read_ptr, last_to_free;
+	int txq_read_ptr, txq_write_ptr;
 
 	/* This function is not meant to release cmd queue*/
 	if (WARN_ON(txq_id == trans_pcie->txqs.cmd.q_id))
@@ -2326,8 +2332,14 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 
 	tfd_num = iwl_txq_get_cmd_index(txq, ssn);
 
-	spin_lock_bh(&txq->lock);
-	read_ptr = iwl_txq_get_cmd_index(txq, txq->read_ptr);
+	spin_lock_bh(&txq->reclaim_lock);
+
+	spin_lock(&txq->lock);
+	txq_read_ptr = txq->read_ptr;
+	txq_write_ptr = txq->write_ptr;
+	spin_unlock(&txq->lock);
+
+	read_ptr = iwl_txq_get_cmd_index(txq, txq_read_ptr);
 
 	if (!test_bit(txq_id, trans_pcie->txqs.queue_used)) {
 		IWL_DEBUG_TX_QUEUES(trans, "Q %d inactive - ignoring idx %d\n",
@@ -2339,19 +2351,19 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 		goto out;
 
 	IWL_DEBUG_TX_REPLY(trans, "[Q %d] %d (%d) -> %d (%d)\n",
-			   txq_id, read_ptr, txq->read_ptr, tfd_num, ssn);
+			   txq_id, read_ptr, txq_read_ptr, tfd_num, ssn);
 
 	/* Since we free until index _not_ inclusive, the one before index is
 	 * the last we will free. This one must be used
 	 */
 	last_to_free = iwl_txq_dec_wrap(trans, tfd_num);
 
-	if (!iwl_txq_used(txq, last_to_free)) {
+	if (!iwl_txq_used(txq, last_to_free, txq_read_ptr, txq_write_ptr)) {
 		IWL_ERR(trans,
 			"%s: Read index for txq id (%d), last_to_free %d is out of range [0-%d] %d %d.\n",
 			__func__, txq_id, last_to_free,
 			trans->trans_cfg->base_params->max_tfd_queue_size,
-			txq->write_ptr, txq->read_ptr);
+			txq_write_ptr, txq_read_ptr);
 
 		iwl_op_mode_time_point(trans->op_mode,
 				       IWL_FW_INI_TIME_POINT_FAKE_TX,
@@ -2364,13 +2376,13 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 
 	for (;
 	     read_ptr != tfd_num;
-	     txq->read_ptr = iwl_txq_inc_wrap(trans, txq->read_ptr),
-	     read_ptr = iwl_txq_get_cmd_index(txq, txq->read_ptr)) {
+	     txq_read_ptr = iwl_txq_inc_wrap(trans, txq_read_ptr),
+	     read_ptr = iwl_txq_get_cmd_index(txq, txq_read_ptr)) {
 		struct iwl_cmd_meta *cmd_meta = &txq->entries[read_ptr].meta;
 		struct sk_buff *skb = txq->entries[read_ptr].skb;
 
 		if (WARN_ONCE(!skb, "no SKB at %d (%d) on queue %d\n",
-			      read_ptr, txq->read_ptr, txq_id))
+			      read_ptr, txq_read_ptr, txq_id))
 			continue;
 
 		iwl_pcie_free_tso_pages(trans, skb, cmd_meta);
@@ -2380,11 +2392,15 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 		txq->entries[read_ptr].skb = NULL;
 
 		if (!trans->trans_cfg->gen2)
-			iwl_txq_gen1_inval_byte_cnt_tbl(trans, txq);
+			iwl_txq_gen1_inval_byte_cnt_tbl(trans, txq,
+							txq_read_ptr);
 
-		iwl_txq_free_tfd(trans, txq);
+		iwl_txq_free_tfd(trans, txq, txq_read_ptr);
 	}
 
+	spin_lock(&txq->lock);
+	txq->read_ptr = txq_read_ptr;
+
 	iwl_txq_progress(txq);
 
 	if (iwl_txq_space(trans, txq) > txq->low_mark &&
@@ -2406,11 +2422,10 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 		txq->overflow_tx = true;
 
 		/*
-		 * This is tricky: we are in reclaim path which is non
-		 * re-entrant, so noone will try to take the access the
-		 * txq data from that path. We stopped tx, so we can't
-		 * have tx as well. Bottom line, we can unlock and re-lock
-		 * later.
+		 * This is tricky: we are in reclaim path and are holding
+		 * reclaim_lock, so noone will try to access the txq data
+		 * from that path. We stopped tx, so we can't have tx as well.
+		 * Bottom line, we can unlock and re-lock later.
 		 */
 		spin_unlock(&txq->lock);
 
@@ -2435,8 +2450,9 @@ void iwl_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn,
 		txq->overflow_tx = false;
 	}
 
+	spin_unlock(&txq->lock);
 out:
-	spin_unlock_bh(&txq->lock);
+	spin_unlock_bh(&txq->reclaim_lock);
 }
 
 /* Set wr_ptr of specific device and txq  */
-- 
cgit v1.2.3-58-ga151


From 5fdbde79eafe45888f41f6947132f63e8d39f485 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 3 Jul 2024 12:58:58 +0300
Subject: wifi: iwlwifi: mvm: add missing string for ROC EMLSR block

This should be labeled for the debug output, add the missing
string.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.03d428f4ff4d.I858f17a5173fe8337cea4e7665fec00dbb15e514@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/link.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/link.c b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
index 4e702e7a4b82..a9929aa49913 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/link.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/link.c
@@ -11,6 +11,7 @@
 	HOW(BLOCKED_TPT)		\
 	HOW(BLOCKED_FW)			\
 	HOW(BLOCKED_NON_BSS)		\
+	HOW(BLOCKED_ROC)		\
 	HOW(EXIT_MISSED_BEACON)		\
 	HOW(EXIT_LOW_RSSI)		\
 	HOW(EXIT_COEX)			\
-- 
cgit v1.2.3-58-ga151


From 3a84161cd3fb0756f640d842cb1278713808f887 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 3 Jul 2024 12:58:59 +0300
Subject: wifi: iwlwifi: fw: api: fix memory region kernel-doc

Add the missing special_mem union member and use constant
formatting (%) for the type constants.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.9344b2b94d45.Id770b4f1893308ba43fc039a8851c526312ad2b5@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 .../net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h    | 34 ++++++++++++----------
 1 file changed, 18 insertions(+), 16 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h b/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h
index e71b3bc20253..855cd13a181e 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/dbg-tlv.h
@@ -147,32 +147,34 @@ struct iwl_fw_ini_region_internal_buffer {
  * Configures parameters for region data collection
  *
  * @hdr: debug header
- * @id: region id. Max id is &IWL_FW_INI_MAX_REGION_ID
+ * @id: region id. Max id is %IWL_FW_INI_MAX_REGION_ID
  * @type: region type. One of &enum iwl_fw_ini_region_type
  * @sub_type: region sub type
  * @sub_type_ver: region sub type version
  * @reserved: not in use
  * @name: region name
  * @dev_addr: device address configuration. Used by
- *	&IWL_FW_INI_REGION_DEVICE_MEMORY, &IWL_FW_INI_REGION_PERIPHERY_MAC,
- *	&IWL_FW_INI_REGION_PERIPHERY_PHY, &IWL_FW_INI_REGION_PERIPHERY_AUX,
- *	&IWL_FW_INI_REGION_PAGING, &IWL_FW_INI_REGION_CSR,
- *	&IWL_FW_INI_REGION_DRAM_IMR and &IWL_FW_INI_REGION_PCI_IOSF_CONFIG
- *	&IWL_FW_INI_REGION_DBGI_SRAM, &FW_TLV_DEBUG_REGION_TYPE_DBGI_SRAM,
- *	&IWL_FW_INI_REGION_PERIPHERY_SNPS_DPHYIP,
+ *	%IWL_FW_INI_REGION_DEVICE_MEMORY, %IWL_FW_INI_REGION_PERIPHERY_MAC,
+ *	%IWL_FW_INI_REGION_PERIPHERY_PHY, %IWL_FW_INI_REGION_PERIPHERY_AUX,
+ *	%IWL_FW_INI_REGION_PAGING, %IWL_FW_INI_REGION_CSR,
+ *	%IWL_FW_INI_REGION_DRAM_IMR and %IWL_FW_INI_REGION_PCI_IOSF_CONFIG
+ *	%IWL_FW_INI_REGION_DBGI_SRAM, %FW_TLV_DEBUG_REGION_TYPE_DBGI_SRAM,
+ *	%IWL_FW_INI_REGION_PERIPHERY_SNPS_DPHYIP,
  * @dev_addr_range: device address range configuration. Used by
- *	&IWL_FW_INI_REGION_PERIPHERY_MAC_RANGE and
- *	&IWL_FW_INI_REGION_PERIPHERY_PHY_RANGE
- * @fifos: fifos configuration. Used by &IWL_FW_INI_REGION_TXF and
- *	&IWL_FW_INI_REGION_RXF
+ *	%IWL_FW_INI_REGION_PERIPHERY_MAC_RANGE and
+ *	%IWL_FW_INI_REGION_PERIPHERY_PHY_RANGE
+ * @fifos: fifos configuration. Used by %IWL_FW_INI_REGION_TXF and
+ *	%IWL_FW_INI_REGION_RXF
  * @err_table: error table configuration. Used by
- *	IWL_FW_INI_REGION_LMAC_ERROR_TABLE and
- *	IWL_FW_INI_REGION_UMAC_ERROR_TABLE
+ *	%IWL_FW_INI_REGION_LMAC_ERROR_TABLE and
+ *	%IWL_FW_INI_REGION_UMAC_ERROR_TABLE
  * @internal_buffer: internal monitor buffer configuration. Used by
- *	&IWL_FW_INI_REGION_INTERNAL_BUFFER
+ *	%IWL_FW_INI_REGION_INTERNAL_BUFFER
+ * @special_mem: special device memory region, used by
+ *	%IWL_FW_INI_REGION_SPECIAL_DEVICE_MEMORY
  * @dram_alloc_id: dram allocation id. One of &enum iwl_fw_ini_allocation_id.
- *	Used by &IWL_FW_INI_REGION_DRAM_BUFFER
- * @tlv_mask: tlv collection mask. Used by &IWL_FW_INI_REGION_TLV
+ *	Used by %IWL_FW_INI_REGION_DRAM_BUFFER
+ * @tlv_mask: tlv collection mask. Used by %IWL_FW_INI_REGION_TLV
  * @addrs: array of addresses attached to the end of the region tlv
  */
 struct iwl_fw_ini_region_tlv {
-- 
cgit v1.2.3-58-ga151


From 6009de8014a85e952e09d6c1944bfe8409114d94 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 3 Jul 2024 12:59:00 +0300
Subject: wifi: iwlwifi: fw: api: mac: fix kernel-doc

Fix kernel-doc for MAC context APIs.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.a7179f1b0e7e.Idd7ce381960707978ff0b16035101a2a725a4fd9@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/mac.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/mac.h b/drivers/net/wireless/intel/iwlwifi/fw/api/mac.h
index 545826973a80..bcbbf8c4a297 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/mac.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/mac.h
@@ -310,6 +310,13 @@ struct iwl_ac_qos {
  * @filter_flags: combination of &enum iwl_mac_filter_flags
  * @qos_flags: from &enum iwl_mac_qos_flags
  * @ac: one iwl_mac_qos configuration for each AC
+ * @ap: AP specific config data, see &struct iwl_mac_data_ap
+ * @go: GO specific config data, see &struct iwl_mac_data_go
+ * @sta: BSS client specific config data, see &struct iwl_mac_data_sta
+ * @p2p_sta: P2P client specific config data, see &struct iwl_mac_data_p2p_sta
+ * @p2p_dev: P2P-device specific config data, see &struct iwl_mac_data_p2p_dev
+ * @pibss: Pseudo-IBSS specific data, unused; see struct iwl_mac_data_pibss
+ * @ibss: IBSS specific config data, see &struct iwl_mac_data_ibss
  */
 struct iwl_mac_ctx_cmd {
 	/* COMMON_INDEX_HDR_API_S_VER_1 */
-- 
cgit v1.2.3-58-ga151


From 94c97e1178d5714d2c02f73da426655adbc24937 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 3 Jul 2024 12:59:01 +0300
Subject: wifi: iwlwifi: fw: api: add puncturing to PHY context docs

Document the puncture_mask field in the PHY context.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.8ba6536ea36b.I181d8da205a14f4fcbf1d8cc6011dcf194a1638a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/phy-ctxt.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/phy-ctxt.h b/drivers/net/wireless/intel/iwlwifi/fw/api/phy-ctxt.h
index f192e02e4ba8..4d8a12799c4d 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/phy-ctxt.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/phy-ctxt.h
@@ -144,6 +144,7 @@ struct iwl_phy_context_cmd_v1 {
  * @rxchain_info: ???
  * @sbb_bandwidth: 0 disabled, 1 - 40Mhz ... 4 - 320MHz
  * @sbb_ctrl_channel_loc: location of the control channel
+ * @puncture_mask: bitmap of punctured subchannels
  * @dsp_cfg_flags: set to 0
  * @reserved: reserved to align to 64 bit
  */
-- 
cgit v1.2.3-58-ga151


From 137579d65db426dbc2e026c1e42cc34ff5ceed85 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 3 Jul 2024 12:59:02 +0300
Subject: wifi: iwlwifi: document PPAG table command union correctly

This is not documented correctly, fix it.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.2ff5ee61e9ec.I3a1299061e472490c38a9fff4bea319ba20f313a@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/power.h | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/power.h b/drivers/net/wireless/intel/iwlwifi/fw/api/power.h
index 19eb6008fe4b..6e6a92d173cc 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/power.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/power.h
@@ -569,9 +569,12 @@ enum iwl_ppag_flags {
  * @v2: version 2
  * version 3, 4, 5 and 6 are the same structure as v2,
  *	but has a different format of the flags bitmap
- * @flags: values from &enum iwl_ppag_flags
- * @gain: table of antenna gain values per chain and sub-band
- * @reserved: reserved
+ * @v1.flags: values from &enum iwl_ppag_flags
+ * @v1.gain: table of antenna gain values per chain and sub-band
+ * @v1.reserved: reserved
+ * @v2.flags: values from &enum iwl_ppag_flags
+ * @v2.gain: table of antenna gain values per chain and sub-band
+ * @v2.reserved: reserved
  */
 union iwl_ppag_table_cmd {
 	struct {
-- 
cgit v1.2.3-58-ga151


From 6188855fbb76a3b9b7c2e5f9744988b3e8b3e641 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 3 Jul 2024 12:59:03 +0300
Subject: wifi: iwlwifi: fw: api: fix missing RX descriptor kernel-doc

The v1/v3 remaining bits are not annotated in kernel-doc,
fix that.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.d7adf8b235fe.I91f75e292d1648f61e5e341e1fe58096f858853d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/rx.h | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
index 8bd85ef90052..691c879cb90d 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/rx.h
@@ -710,7 +710,15 @@ struct iwl_rx_mpdu_desc {
 	__le32 reorder_data;
 
 	union {
+		/**
+		 * @v1: version 1 of the remaining RX descriptor,
+		 *	see &struct iwl_rx_mpdu_desc_v1
+		 */
 		struct iwl_rx_mpdu_desc_v1 v1;
+		/**
+		 * @v3: version 3 of the remaining RX descriptor,
+		 *	see &struct iwl_rx_mpdu_desc_v3
+		 */
 		struct iwl_rx_mpdu_desc_v3 v3;
 	};
 } __packed; /* RX_MPDU_RES_START_API_S_VER_3,
-- 
cgit v1.2.3-58-ga151


From 8d4f5969f7c7d3a8f0e93d37a1a2fc23048dd12f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Wed, 3 Jul 2024 12:59:04 +0300
Subject: wifi: iwlwifi: mvm: document remaining mvm data

Many iwl_mvm_vif members are not documented, add that.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.371664e5e8cd.I593ebee1ab984554b6d269dc2dddc67fbf3bb537@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h | 29 ++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 056fdc42e253..22f48b66d79c 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -453,6 +453,35 @@ struct iwl_mvm_esr_exit {
  * @session_prot_connection_loss: the connection was lost due to session
  *	protection ending without receiving a beacon, so we need to now
  *	protect the deauth separately
+ * @ap_early_keys: The firmware cannot install keys before stations etc.,
+ *	but higher layers work differently, so we store the keys here for
+ *	later installation.
+ * @ap_sta: pointer to the AP STA data structure
+ * @csa_count: CSA counter (old CSA implementation w/o firmware)
+ * @csa_misbehave: CSA AP misbehaviour flag (old implementation)
+ * @csa_target_freq: CSA target channel frequency (old implementation)
+ * @csa_work: CSA work (old implementation)
+ * @dbgfs_bf: beamforming debugfs data
+ * @dbgfs_dir: debugfs directory for this vif
+ * @dbgfs_pm: power management debugfs data
+ * @dbgfs_quota_min: debugfs value for minimal quota
+ * @dbgfs_slink: debugfs symlink for this interface
+ * @ftm_unprotected: unprotected FTM debugfs override
+ * @hs_time_event_data: hotspot/AUX ROC time event data
+ * @mac_pwr_cmd: debugfs override for MAC power command
+ * @target_ipv6_addrs: IPv6 addresses on this interface for offload
+ * @num_target_ipv6_addrs: number of @target_ipv6_addrs
+ * @tentative_addrs: bitmap of tentative IPv6 addresses in @target_ipv6_addrs
+ * @rekey_data: rekeying data for WoWLAN GTK rekey offload
+ * @seqno: storage for seqno for older firmware D0/D3 transition
+ * @seqno_valid: indicates @seqno is valid
+ * @time_event_data: session protection time event data
+ * @tsf_id: the TSF resource ID assigned in firmware (for firmware needing that)
+ * @tx_key_idx: WEP transmit key index for D3
+ * @uapsd_misbehaving_ap_addr: MLD address/BSSID of U-APSD misbehaving AP, to
+ *	not use U-APSD on reconnection
+ * @uapsd_nonagg_detected_wk: worker for handling detection of no aggregation
+ *	in U-APSD
  */
 struct iwl_mvm_vif {
 	struct iwl_mvm *mvm;
-- 
cgit v1.2.3-58-ga151


From 1895712afd1cab549b143717d6f1b6534383a9e0 Mon Sep 17 00:00:00 2001
From: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Date: Wed, 3 Jul 2024 12:59:05 +0300
Subject: wifi: iwlwifi: Remove debug message

Debug logs related to reset_fw are logged with all
notification/response and polluting the trace.

Remove the debug message related to reset_fw setting
when dump is collected.

Signed-off-by: Mukesh Sisodiya <mukesh.sisodiya@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.8fc59cb17526.Ibb5d68b2fe5f7df709db3570de55a566d5af3f24@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c | 13 +------------
 1 file changed, 1 insertion(+), 12 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
index 9810a7f4a591..08d990ba8a79 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-dbg-tlv.c
@@ -1240,12 +1240,6 @@ iwl_dbg_tlv_tp_trigger(struct iwl_fw_runtime *fwrt, bool sync,
 		}
 
 		fwrt->trans->dbg.restart_required = false;
-		IWL_DEBUG_FW(fwrt, "WRT: tp %d, reset_fw %d\n",
-			     tp, dump_data.trig->reset_fw);
-		IWL_DEBUG_FW(fwrt,
-			     "WRT: restart_required %d, last_tp_resetfw %d\n",
-			     fwrt->trans->dbg.restart_required,
-			     fwrt->trans->dbg.last_tp_resetfw);
 
 		if (fwrt->trans->trans_cfg->device_family ==
 		    IWL_DEVICE_FAMILY_9000) {
@@ -1255,22 +1249,17 @@ iwl_dbg_tlv_tp_trigger(struct iwl_fw_runtime *fwrt, bool sync,
 			   IWL_FW_INI_RESET_FW_MODE_STOP_FW_ONLY) {
 			fwrt->trans->dbg.restart_required = false;
 			fwrt->trans->dbg.last_tp_resetfw = 0xFF;
-			IWL_DEBUG_FW(fwrt, "WRT: FW_ASSERT due to reset_fw_mode-no restart\n");
 		} else if (le32_to_cpu(dump_data.trig->reset_fw) ==
 			   IWL_FW_INI_RESET_FW_MODE_STOP_AND_RELOAD_FW) {
-			IWL_DEBUG_FW(fwrt, "WRT: stop and reload firmware\n");
 			fwrt->trans->dbg.restart_required = true;
 		} else if (le32_to_cpu(dump_data.trig->reset_fw) ==
 			   IWL_FW_INI_RESET_FW_MODE_STOP_FW_ONLY) {
-			IWL_DEBUG_FW(fwrt,
-				     "WRT: stop only and no reload firmware\n");
 			fwrt->trans->dbg.restart_required = false;
 			fwrt->trans->dbg.last_tp_resetfw =
 				le32_to_cpu(dump_data.trig->reset_fw);
 		} else if (le32_to_cpu(dump_data.trig->reset_fw) ==
 			   IWL_FW_INI_RESET_FW_MODE_NOTHING) {
-			IWL_DEBUG_FW(fwrt,
-				     "WRT: nothing need to be done after debug collection\n");
+			/* nothing */
 		} else {
 			IWL_ERR(fwrt, "WRT: wrong resetfw %d\n",
 				le32_to_cpu(dump_data.trig->reset_fw));
-- 
cgit v1.2.3-58-ga151


From ef7ddf4e2f949633bd7b2b01c2e0f023aefb6987 Mon Sep 17 00:00:00 2001
From: Anjaneyulu <pagadala.yesu.anjaneyulu@intel.com>
Date: Wed, 3 Jul 2024 12:59:06 +0300
Subject: wifi: iwlwifi: Add support for LARI_CONFIG_CHANGE_CMD v12

Add support for activate/deactivate unii4 in USA, Canada and WW by
reading DSM function 8 from UEFI or ACPI and sending it to the FW.

Signed-off-by: Anjaneyulu <pagadala.yesu.anjaneyulu@intel.com>
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703125541.674604cbb6d1.Ibb946ae8ce7a760940a3c9d101e7f4f1808c43e4@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h | 6 +++++-
 drivers/net/wireless/intel/iwlwifi/fw/regulatory.c  | 4 ++++
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h b/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h
index 9ff5d7e538fd..d424d0126367 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h
+++ b/drivers/net/wireless/intel/iwlwifi/fw/api/nvm-reg.h
@@ -704,6 +704,8 @@ struct iwl_lari_config_change_cmd_v10 {
  *	Each bit represents a country or region, and a band to activate
  *	according to the BIOS definitions.
  *	For LARI cmd version 11 - bits 0:4 are supported.
+ *	For LARI cmd version 12 - bits 0:6 are supported and bits 7:31 are
+ *	reserved. No need to mask out the reserved bits.
  * @force_disable_channels_bitmap: Bitmap of disabled bands/channels.
  *	Each bit represents a set of channels in a specific band that should be
  *	disabled
@@ -731,9 +733,11 @@ struct iwl_lari_config_change_cmd {
 	__le32 oem_11be_allow_bitmap;
 } __packed;
 /* LARI_CHANGE_CONF_CMD_S_VER_11 */
+/* LARI_CHANGE_CONF_CMD_S_VER_12 */
 
 /* Activate UNII-1 (5.2GHz) for World Wide */
-#define ACTIVATE_5G2_IN_WW_MASK	BIT(4)
+#define ACTIVATE_5G2_IN_WW_MASK			BIT(4)
+#define CHAN_STATE_ACTIVE_BITMAP_CMD_V11	0x1F
 
 /**
  * struct iwl_pnvm_init_complete_ntfy - PNVM initialization complete
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c b/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c
index 5228b837a9ef..560a91998cc4 100644
--- a/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c
+++ b/drivers/net/wireless/intel/iwlwifi/fw/regulatory.c
@@ -497,6 +497,7 @@ static size_t iwl_get_lari_config_cmd_size(u8 cmd_ver)
 	size_t cmd_size;
 
 	switch (cmd_ver) {
+	case 12:
 	case 11:
 		cmd_size = sizeof(struct iwl_lari_config_change_cmd);
 		break;
@@ -563,6 +564,9 @@ int iwl_fill_lari_config(struct iwl_fw_runtime *fwrt,
 	if (!ret) {
 		if (cmd_ver < 8)
 			value &= ~ACTIVATE_5G2_IN_WW_MASK;
+		if (cmd_ver < 12)
+			value &= CHAN_STATE_ACTIVE_BITMAP_CMD_V11;
+
 		cmd->chan_state_active_bitmap = cpu_to_le32(value);
 	}
 
-- 
cgit v1.2.3-58-ga151


From 1b431ba4ef9a760e7643d6fbc53bf522d59650f3 Mon Sep 17 00:00:00 2001
From: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Date: Wed, 3 Jul 2024 13:27:37 +0300
Subject: wifi: iwlwifi: mvm: re-enable MLO

MLO was temporarily disabled by
commit 5f404005055 ("wifi: iwlwifi: mvm: disable MLO for the time being"),
until it will stabilize.
Now, that all the bugs were fixed and the minimum FW version was bumped
to a stable one, we can re-enable MLO back.

Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Link: https://patch.msgid.link/20240703132713.8f77a71c3902.Ib302054cbd8fba82db97eb5298b2aaf8bbe106df@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
index e40f993b17fd..41aba087c1a6 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c
@@ -372,7 +372,7 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm)
 	if (mvm->mld_api_is_used && mvm->nvm_data->sku_cap_11be_enable &&
 	    !iwlwifi_mod_params.disable_11ax &&
 	    !iwlwifi_mod_params.disable_11be) {
-		hw->wiphy->flags |= WIPHY_FLAG_DISABLE_WEXT;
+		hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_MLO;
 		/* we handle this already earlier, but need it for MLO */
 		ieee80211_hw_set(hw, HANDLES_QUIET_CSA);
 	}
-- 
cgit v1.2.3-58-ga151


From 9b5c33b1a3b7593c0e1d44d92c0093cc44e635fe Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Tue, 2 Jul 2024 11:55:51 -0700
Subject: crypto: caam: Avoid unused imx8m_machine_match variable

If caam module is built without OF support, the compiler returns the
following warning:

	drivers/crypto/caam/ctrl.c:83:34: warning: 'imx8m_machine_match' defined but not used [-Wunused-const-variable=]

imx8m_machine_match is only referenced by of_match_node(), which is set
to NULL if CONFIG_OF is not set, as of commit 5762c20593b6b ("dt: Add
empty of_match_node() macro"):

	#define of_match_node(_matches, _node)  NULL

Do not create imx8m_machine_match if CONFIG_OF is not set.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202407011309.cpTuOGdg-lkp@intel.com/
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20240702185557.3699991-2-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/crypto/caam/ctrl.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/crypto/caam/ctrl.c b/drivers/crypto/caam/ctrl.c
index bd418dea586d..d4b39184dbdb 100644
--- a/drivers/crypto/caam/ctrl.c
+++ b/drivers/crypto/caam/ctrl.c
@@ -80,6 +80,7 @@ static void build_deinstantiation_desc(u32 *desc, int handle)
 	append_jump(desc, JUMP_CLASS_CLASS1 | JUMP_TYPE_HALT);
 }
 
+#ifdef CONFIG_OF
 static const struct of_device_id imx8m_machine_match[] = {
 	{ .compatible = "fsl,imx8mm", },
 	{ .compatible = "fsl,imx8mn", },
@@ -88,6 +89,7 @@ static const struct of_device_id imx8m_machine_match[] = {
 	{ .compatible = "fsl,imx8ulp", },
 	{ }
 };
+#endif
 
 /*
  * run_descriptor_deco0 - runs a descriptor on DECO0, under direct control of
-- 
cgit v1.2.3-58-ga151


From beba3771d9e049dd552ce42a31f263e098392407 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Tue, 2 Jul 2024 11:55:52 -0700
Subject: crypto: caam: Make CRYPTO_DEV_FSL_CAAM dependent of COMPILE_TEST

As most of the drivers that depend on ARCH_LAYERSCAPE, make
CRYPTO_DEV_FSL_CAAM depend on COMPILE_TEST for compilation and testing.

    # grep -r depends.\*ARCH_LAYERSCAPE.\*COMPILE_TEST | wc -l
    29

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20240702185557.3699991-3-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/crypto/caam/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/crypto/caam/Kconfig b/drivers/crypto/caam/Kconfig
index c631f99e415f..05210a0edb8a 100644
--- a/drivers/crypto/caam/Kconfig
+++ b/drivers/crypto/caam/Kconfig
@@ -10,7 +10,7 @@ config CRYPTO_DEV_FSL_CAAM_AHASH_API_DESC
 
 config CRYPTO_DEV_FSL_CAAM
 	tristate "Freescale CAAM-Multicore platform driver backend"
-	depends on FSL_SOC || ARCH_MXC || ARCH_LAYERSCAPE
+	depends on FSL_SOC || ARCH_MXC || ARCH_LAYERSCAPE || COMPILE_TEST
 	select SOC_BUS
 	select CRYPTO_DEV_FSL_CAAM_COMMON
 	imply FSL_MC_BUS
-- 
cgit v1.2.3-58-ga151


From 82c81e740defbe2cd9889ba76da15f99d7cba252 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Tue, 2 Jul 2024 11:55:53 -0700
Subject: crypto: caam: Unembed net_dev structure from qi

Embedding net_device into structures prohibits the usage of flexible
arrays in the net_device structure. For more details, see the discussion
at [1].

Un-embed the net_devices from struct caam_qi_pcpu_priv by converting them
into pointers, and allocating them dynamically. Use the leverage
alloc_netdev_dummy() to allocate the net_device object at
caam_qi_init().

The free of the device occurs at caam_qi_shutdown().

Link: https://lore.kernel.org/all/20240229225910.79e224cf@kernel.org/ [1]
Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20240702185557.3699991-4-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/crypto/caam/qi.c | 43 +++++++++++++++++++++++++++++++++++--------
 1 file changed, 35 insertions(+), 8 deletions(-)

diff --git a/drivers/crypto/caam/qi.c b/drivers/crypto/caam/qi.c
index 46a083849a8e..ba8fb5d8a7b2 100644
--- a/drivers/crypto/caam/qi.c
+++ b/drivers/crypto/caam/qi.c
@@ -57,7 +57,7 @@ struct caam_napi {
  */
 struct caam_qi_pcpu_priv {
 	struct caam_napi caam_napi;
-	struct net_device net_dev;
+	struct net_device *net_dev;
 	struct qman_fq *rsp_fq;
 } ____cacheline_aligned;
 
@@ -144,7 +144,7 @@ static void caam_fq_ern_cb(struct qman_portal *qm, struct qman_fq *fq,
 {
 	const struct qm_fd *fd;
 	struct caam_drv_req *drv_req;
-	struct device *qidev = &(raw_cpu_ptr(&pcpu_qipriv)->net_dev.dev);
+	struct device *qidev = &(raw_cpu_ptr(&pcpu_qipriv)->net_dev->dev);
 	struct caam_drv_private *priv = dev_get_drvdata(qidev);
 
 	fd = &msg->ern.fd;
@@ -530,6 +530,7 @@ static void caam_qi_shutdown(void *data)
 
 		if (kill_fq(qidev, per_cpu(pcpu_qipriv.rsp_fq, i)))
 			dev_err(qidev, "Rsp FQ kill failed, cpu: %d\n", i);
+		free_netdev(per_cpu(pcpu_qipriv.net_dev, i));
 	}
 
 	qman_delete_cgr_safe(&priv->cgr);
@@ -573,7 +574,7 @@ static enum qman_cb_dqrr_result caam_rsp_fq_dqrr_cb(struct qman_portal *p,
 	struct caam_napi *caam_napi = raw_cpu_ptr(&pcpu_qipriv.caam_napi);
 	struct caam_drv_req *drv_req;
 	const struct qm_fd *fd;
-	struct device *qidev = &(raw_cpu_ptr(&pcpu_qipriv)->net_dev.dev);
+	struct device *qidev = &(raw_cpu_ptr(&pcpu_qipriv)->net_dev->dev);
 	struct caam_drv_private *priv = dev_get_drvdata(qidev);
 	u32 status;
 
@@ -718,12 +719,24 @@ static void free_rsp_fqs(void)
 		kfree(per_cpu(pcpu_qipriv.rsp_fq, i));
 }
 
+static void free_caam_qi_pcpu_netdev(const cpumask_t *cpus)
+{
+	struct caam_qi_pcpu_priv *priv;
+	int i;
+
+	for_each_cpu(i, cpus) {
+		priv = per_cpu_ptr(&pcpu_qipriv, i);
+		free_netdev(priv->net_dev);
+	}
+}
+
 int caam_qi_init(struct platform_device *caam_pdev)
 {
 	int err, i;
 	struct device *ctrldev = &caam_pdev->dev, *qidev;
 	struct caam_drv_private *ctrlpriv;
 	const cpumask_t *cpus = qman_affine_cpus();
+	cpumask_t clean_mask;
 
 	ctrlpriv = dev_get_drvdata(ctrldev);
 	qidev = ctrldev;
@@ -743,6 +756,8 @@ int caam_qi_init(struct platform_device *caam_pdev)
 		return err;
 	}
 
+	cpumask_clear(&clean_mask);
+
 	/*
 	 * Enable the NAPI contexts on each of the core which has an affine
 	 * portal.
@@ -751,10 +766,16 @@ int caam_qi_init(struct platform_device *caam_pdev)
 		struct caam_qi_pcpu_priv *priv = per_cpu_ptr(&pcpu_qipriv, i);
 		struct caam_napi *caam_napi = &priv->caam_napi;
 		struct napi_struct *irqtask = &caam_napi->irqtask;
-		struct net_device *net_dev = &priv->net_dev;
+		struct net_device *net_dev;
 
+		net_dev = alloc_netdev_dummy(0);
+		if (!net_dev) {
+			err = -ENOMEM;
+			goto fail;
+		}
+		cpumask_set_cpu(i, &clean_mask);
+		priv->net_dev = net_dev;
 		net_dev->dev = *qidev;
-		INIT_LIST_HEAD(&net_dev->napi_list);
 
 		netif_napi_add_tx_weight(net_dev, irqtask, caam_qi_poll,
 					 CAAM_NAPI_WEIGHT);
@@ -766,16 +787,22 @@ int caam_qi_init(struct platform_device *caam_pdev)
 				     dma_get_cache_alignment(), 0, NULL);
 	if (!qi_cache) {
 		dev_err(qidev, "Can't allocate CAAM cache\n");
-		free_rsp_fqs();
-		return -ENOMEM;
+		err = -ENOMEM;
+		goto fail2;
 	}
 
 	caam_debugfs_qi_init(ctrlpriv);
 
 	err = devm_add_action_or_reset(qidev, caam_qi_shutdown, ctrlpriv);
 	if (err)
-		return err;
+		goto fail2;
 
 	dev_info(qidev, "Linux CAAM Queue I/F driver initialised\n");
 	return 0;
+
+fail2:
+	free_rsp_fqs();
+fail:
+	free_caam_qi_pcpu_netdev(&clean_mask);
+	return err;
 }
-- 
cgit v1.2.3-58-ga151


From 0e1a4d427f58b843864d835ccb79c32a118c6a77 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Tue, 2 Jul 2024 11:55:54 -0700
Subject: crypto: caam: Unembed net_dev structure in dpaa2

Embedding net_device into structures prohibits the usage of flexible
arrays in the net_device structure. For more details, see the discussion
at [1].

Un-embed the net_devices from struct dpaa2_caam_priv_per_cpu by
converting them into pointers, and allocating them dynamically. Use the
leverage alloc_netdev_dummy() to allocate the net_device object at
dpaa2_dpseci_setup().

The free of the device occurs at dpaa2_dpseci_disable().

Link: https://lore.kernel.org/all/20240229225910.79e224cf@kernel.org/ [1]
Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20240702185557.3699991-5-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/crypto/caam/caamalg_qi2.c | 28 +++++++++++++++++++++++++---
 drivers/crypto/caam/caamalg_qi2.h |  2 +-
 2 files changed, 26 insertions(+), 4 deletions(-)

diff --git a/drivers/crypto/caam/caamalg_qi2.c b/drivers/crypto/caam/caamalg_qi2.c
index a4f6884416a0..207dc422785a 100644
--- a/drivers/crypto/caam/caamalg_qi2.c
+++ b/drivers/crypto/caam/caamalg_qi2.c
@@ -4990,11 +4990,23 @@ err_dma_map:
 	return err;
 }
 
+static void free_dpaa2_pcpu_netdev(struct dpaa2_caam_priv *priv, const cpumask_t *cpus)
+{
+	struct dpaa2_caam_priv_per_cpu *ppriv;
+	int i;
+
+	for_each_cpu(i, cpus) {
+		ppriv = per_cpu_ptr(priv->ppriv, i);
+		free_netdev(ppriv->net_dev);
+	}
+}
+
 static int __cold dpaa2_dpseci_setup(struct fsl_mc_device *ls_dev)
 {
 	struct device *dev = &ls_dev->dev;
 	struct dpaa2_caam_priv *priv;
 	struct dpaa2_caam_priv_per_cpu *ppriv;
+	cpumask_t clean_mask;
 	int err, cpu;
 	u8 i;
 
@@ -5073,6 +5085,7 @@ static int __cold dpaa2_dpseci_setup(struct fsl_mc_device *ls_dev)
 		}
 	}
 
+	cpumask_clear(&clean_mask);
 	i = 0;
 	for_each_online_cpu(cpu) {
 		u8 j;
@@ -5096,15 +5109,23 @@ static int __cold dpaa2_dpseci_setup(struct fsl_mc_device *ls_dev)
 			priv->rx_queue_attr[j].fqid,
 			priv->tx_queue_attr[j].fqid);
 
-		ppriv->net_dev.dev = *dev;
-		INIT_LIST_HEAD(&ppriv->net_dev.napi_list);
-		netif_napi_add_tx_weight(&ppriv->net_dev, &ppriv->napi,
+		ppriv->net_dev = alloc_netdev_dummy(0);
+		if (!ppriv->net_dev) {
+			err = -ENOMEM;
+			goto err_alloc_netdev;
+		}
+		cpumask_set_cpu(cpu, &clean_mask);
+		ppriv->net_dev->dev = *dev;
+
+		netif_napi_add_tx_weight(ppriv->net_dev, &ppriv->napi,
 					 dpaa2_dpseci_poll,
 					 DPAA2_CAAM_NAPI_WEIGHT);
 	}
 
 	return 0;
 
+err_alloc_netdev:
+	free_dpaa2_pcpu_netdev(priv, &clean_mask);
 err_get_rx_queue:
 	dpaa2_dpseci_congestion_free(priv);
 err_get_vers:
@@ -5153,6 +5174,7 @@ static int __cold dpaa2_dpseci_disable(struct dpaa2_caam_priv *priv)
 		ppriv = per_cpu_ptr(priv->ppriv, i);
 		napi_disable(&ppriv->napi);
 		netif_napi_del(&ppriv->napi);
+		free_netdev(ppriv->net_dev);
 	}
 
 	return 0;
diff --git a/drivers/crypto/caam/caamalg_qi2.h b/drivers/crypto/caam/caamalg_qi2.h
index abb502bb675c..61d1219a202f 100644
--- a/drivers/crypto/caam/caamalg_qi2.h
+++ b/drivers/crypto/caam/caamalg_qi2.h
@@ -81,7 +81,7 @@ struct dpaa2_caam_priv {
  */
 struct dpaa2_caam_priv_per_cpu {
 	struct napi_struct napi;
-	struct net_device net_dev;
+	struct net_device *net_dev;
 	int req_fqid;
 	int rsp_fqid;
 	int prio;
-- 
cgit v1.2.3-58-ga151


From caa93b7c25945d302689de07bd404655db93ae6e Mon Sep 17 00:00:00 2001
From: Edward Cree <ecree.xilinx@gmail.com>
Date: Wed, 3 Jul 2024 13:18:49 +0100
Subject: ethtool: move firmware flashing flag to struct ethtool_netdev_state

Commit 31e0aa99dc02 ("ethtool: Veto some operations during firmware flashing process")
 added a flag module_fw_flash_in_progress to struct net_device.  As
 this is ethtool related state, move it to the recently created
 struct ethtool_netdev_state, accessed via the 'ethtool' member of
 struct net_device.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20240703121849.652893-1-edward.cree@amd.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/ethtool.h   |  2 ++
 include/linux/netdevice.h |  3 ---
 net/ethtool/eeprom.c      |  2 +-
 net/ethtool/ioctl.c       |  8 ++++----
 net/ethtool/module.c      | 10 +++++-----
 net/ethtool/netlink.c     |  2 +-
 6 files changed, 13 insertions(+), 14 deletions(-)

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index f74bb0cf8ed1..3a99238ef895 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -1107,11 +1107,13 @@ int ethtool_virtdev_set_link_ksettings(struct net_device *dev,
  * @rss_lock:		Protects entries in @rss_ctx.  May be taken from
  *			within RTNL.
  * @wol_enabled:	Wake-on-LAN is enabled
+ * @module_fw_flash_in_progress: Module firmware flashing is in progress.
  */
 struct ethtool_netdev_state {
 	struct xarray		rss_ctx;
 	struct mutex		rss_lock;
 	unsigned		wol_enabled:1;
+	unsigned		module_fw_flash_in_progress:1;
 };
 
 struct phy_device;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 3c719f0d5f5a..93558645c6d0 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1989,8 +1989,6 @@ enum netdev_reg_state {
  *
  *	@threaded:	napi threaded mode is enabled
  *
- *	@module_fw_flash_in_progress:	Module firmware flashing is in progress.
- *
  *	@net_notifier_list:	List of per-net netdev notifier block
  *				that follow this device when it is moved
  *				to another network namespace.
@@ -2376,7 +2374,6 @@ struct net_device {
 	bool			proto_down;
 	bool			threaded;
 
-	unsigned		module_fw_flash_in_progress:1;
 	struct list_head	net_notifier_list;
 
 #if IS_ENABLED(CONFIG_MACSEC)
diff --git a/net/ethtool/eeprom.c b/net/ethtool/eeprom.c
index f36811b3ecf1..3b8209e930fd 100644
--- a/net/ethtool/eeprom.c
+++ b/net/ethtool/eeprom.c
@@ -91,7 +91,7 @@ static int get_module_eeprom_by_page(struct net_device *dev,
 {
 	const struct ethtool_ops *ops = dev->ethtool_ops;
 
-	if (dev->module_fw_flash_in_progress) {
+	if (dev->ethtool->module_fw_flash_in_progress) {
 		NL_SET_ERR_MSG(extack,
 			       "Module firmware flashing is in progress");
 		return -EBUSY;
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 46f0497ae6bc..d72b0fec89af 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -658,7 +658,7 @@ static int ethtool_get_settings(struct net_device *dev, void __user *useraddr)
 	if (!dev->ethtool_ops->get_link_ksettings)
 		return -EOPNOTSUPP;
 
-	if (dev->module_fw_flash_in_progress)
+	if (dev->ethtool->module_fw_flash_in_progress)
 		return -EBUSY;
 
 	memset(&link_ksettings, 0, sizeof(link_ksettings));
@@ -1572,7 +1572,7 @@ static int ethtool_reset(struct net_device *dev, char __user *useraddr)
 	if (!dev->ethtool_ops->reset)
 		return -EOPNOTSUPP;
 
-	if (dev->module_fw_flash_in_progress)
+	if (dev->ethtool->module_fw_flash_in_progress)
 		return -EBUSY;
 
 	if (copy_from_user(&reset, useraddr, sizeof(reset)))
@@ -2588,7 +2588,7 @@ int ethtool_get_module_info_call(struct net_device *dev,
 	const struct ethtool_ops *ops = dev->ethtool_ops;
 	struct phy_device *phydev = dev->phydev;
 
-	if (dev->module_fw_flash_in_progress)
+	if (dev->ethtool->module_fw_flash_in_progress)
 		return -EBUSY;
 
 	if (dev->sfp_bus)
@@ -2628,7 +2628,7 @@ int ethtool_get_module_eeprom_call(struct net_device *dev,
 	const struct ethtool_ops *ops = dev->ethtool_ops;
 	struct phy_device *phydev = dev->phydev;
 
-	if (dev->module_fw_flash_in_progress)
+	if (dev->ethtool->module_fw_flash_in_progress)
 		return -EBUSY;
 
 	if (dev->sfp_bus)
diff --git a/net/ethtool/module.c b/net/ethtool/module.c
index 6b7448df08d5..aba78436d350 100644
--- a/net/ethtool/module.c
+++ b/net/ethtool/module.c
@@ -37,7 +37,7 @@ static int module_get_power_mode(struct net_device *dev,
 	if (!ops->get_module_power_mode)
 		return 0;
 
-	if (dev->module_fw_flash_in_progress) {
+	if (dev->ethtool->module_fw_flash_in_progress) {
 		NL_SET_ERR_MSG(extack,
 			       "Module firmware flashing is in progress");
 		return -EBUSY;
@@ -119,7 +119,7 @@ ethnl_set_module_validate(struct ethnl_req_info *req_info,
 	if (!tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY])
 		return 0;
 
-	if (req_info->dev->module_fw_flash_in_progress) {
+	if (req_info->dev->ethtool->module_fw_flash_in_progress) {
 		NL_SET_ERR_MSG(info->extack,
 			       "Module firmware flashing is in progress");
 		return -EBUSY;
@@ -226,7 +226,7 @@ static void module_flash_fw_work(struct work_struct *work)
 	ethtool_cmis_fw_update(&module_fw->fw_update);
 
 	module_flash_fw_work_list_del(&module_fw->list);
-	module_fw->fw_update.dev->module_fw_flash_in_progress = false;
+	module_fw->fw_update.dev->ethtool->module_fw_flash_in_progress = false;
 	netdev_put(module_fw->fw_update.dev, &module_fw->dev_tracker);
 	release_firmware(module_fw->fw_update.fw);
 	kfree(module_fw);
@@ -318,7 +318,7 @@ module_flash_fw_schedule(struct net_device *dev, const char *file_name,
 	if (err < 0)
 		goto err_release_firmware;
 
-	dev->module_fw_flash_in_progress = true;
+	dev->ethtool->module_fw_flash_in_progress = true;
 	netdev_hold(dev, &module_fw->dev_tracker, GFP_KERNEL);
 	fw_update->dev = dev;
 	fw_update->ntf_params.portid = info->snd_portid;
@@ -385,7 +385,7 @@ static int ethnl_module_fw_flash_validate(struct net_device *dev,
 		return -EOPNOTSUPP;
 	}
 
-	if (dev->module_fw_flash_in_progress) {
+	if (dev->ethtool->module_fw_flash_in_progress) {
 		NL_SET_ERR_MSG(extack, "Module firmware flashing already in progress");
 		return -EBUSY;
 	}
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 81fe2e5b95f6..cb1eea00e349 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -807,7 +807,7 @@ static int ethnl_netdev_event(struct notifier_block *this, unsigned long event,
 		ethnl_notify_features(ptr);
 		break;
 	case NETDEV_PRE_UP:
-		if (dev->module_fw_flash_in_progress) {
+		if (dev->ethtool->module_fw_flash_in_progress) {
 			NL_SET_ERR_MSG(extack, "Can't set port up while flashing module firmware");
 			return NOTIFY_BAD;
 		}
-- 
cgit v1.2.3-58-ga151


From adc539784c98a7cc602cbf557debfc2e7b9be8b3 Mon Sep 17 00:00:00 2001
From: Marcin Ślusarz <mslusarz@renau.com>
Date: Tue, 28 May 2024 13:02:46 +0200
Subject: wifi: rtw88: usb: schedule rx work after everything is set up
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Right now it's possible to hit NULL pointer dereference in
rtw_rx_fill_rx_status on hw object and/or its fields because
initialization routine can start getting USB replies before
rtw_dev is fully setup.

The stack trace looks like this:

rtw_rx_fill_rx_status
rtw8821c_query_rx_desc
rtw_usb_rx_handler
...
queue_work
rtw_usb_read_port_complete
...
usb_submit_urb
rtw_usb_rx_resubmit
rtw_usb_init_rx
rtw_usb_probe

So while we do the async stuff rtw_usb_probe continues and calls
rtw_register_hw, which does all kinds of initialization (e.g.
via ieee80211_register_hw) that rtw_rx_fill_rx_status relies on.

Fix this by moving the first usb_submit_urb after everything
is set up.

For me, this bug manifested as:
[    8.893177] rtw_8821cu 1-1:1.2: band wrong, packet dropped
[    8.910904] rtw_8821cu 1-1:1.2: hw->conf.chandef.chan NULL in rtw_rx_fill_rx_status
because I'm using Larry's backport of rtw88 driver with the NULL
checks in rtw_rx_fill_rx_status.

Link: https://lore.kernel.org/linux-wireless/CA+shoWQ7P49jhQasofDcTdQhiuarPTjYEDa--NiVVx494WcuQw@mail.gmail.com/
Signed-off-by: Marcin Ślusarz <mslusarz@renau.com>
Cc: Tim K <tpkuester@gmail.com>
Cc: Ping-Ke Shih <pkshih@realtek.com>
Cc: Larry Finger <Larry.Finger@lwfinger.net>
Cc: Kalle Valo <kvalo@kernel.org>
Cc: linux-wireless@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240528110246.477321-1-marcin.slusarz@gmail.com
---
 drivers/net/wireless/realtek/rtw88/usb.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw88/usb.c b/drivers/net/wireless/realtek/rtw88/usb.c
index 057c0ffbe944..a55ca5a24227 100644
--- a/drivers/net/wireless/realtek/rtw88/usb.c
+++ b/drivers/net/wireless/realtek/rtw88/usb.c
@@ -742,7 +742,6 @@ static struct rtw_hci_ops rtw_usb_ops = {
 static int rtw_usb_init_rx(struct rtw_dev *rtwdev)
 {
 	struct rtw_usb *rtwusb = rtw_get_usb_priv(rtwdev);
-	int i;
 
 	rtwusb->rxwq = create_singlethread_workqueue("rtw88_usb: rx wq");
 	if (!rtwusb->rxwq) {
@@ -754,13 +753,19 @@ static int rtw_usb_init_rx(struct rtw_dev *rtwdev)
 
 	INIT_WORK(&rtwusb->rx_work, rtw_usb_rx_handler);
 
+	return 0;
+}
+
+static void rtw_usb_setup_rx(struct rtw_dev *rtwdev)
+{
+	struct rtw_usb *rtwusb = rtw_get_usb_priv(rtwdev);
+	int i;
+
 	for (i = 0; i < RTW_USB_RXCB_NUM; i++) {
 		struct rx_usb_ctrl_block *rxcb = &rtwusb->rx_cb[i];
 
 		rtw_usb_rx_resubmit(rtwusb, rxcb);
 	}
-
-	return 0;
 }
 
 static void rtw_usb_deinit_rx(struct rtw_dev *rtwdev)
@@ -897,6 +902,8 @@ int rtw_usb_probe(struct usb_interface *intf, const struct usb_device_id *id)
 		goto err_destroy_rxwq;
 	}
 
+	rtw_usb_setup_rx(rtwdev);
+
 	return 0;
 
 err_destroy_rxwq:
-- 
cgit v1.2.3-58-ga151


From 0e557c5c1a61ff4d896a21b00e4e7397e3c123b5 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Mon, 1 Jul 2024 09:46:18 +0800
Subject: wifi: rtw89: 8852bx: move BTC common code from 8852b to 8852b_common

The BT coexistence part of 8852B and 8852BT are similar, so move shared
code into common module.

Don't change logic for existing RTL8852BE.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240701014619.7300-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852b.c      | 173 +--------------------
 .../net/wireless/realtek/rtw89/rtw8852b_common.c   | 168 ++++++++++++++++++++
 .../net/wireless/realtek/rtw89/rtw8852b_common.h   |  42 +++++
 3 files changed, 216 insertions(+), 167 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b.c b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
index 1a9c4951060a..a22847a311ad 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b.c
@@ -633,86 +633,6 @@ static void rtw8852b_btc_set_rfe(struct rtw89_dev *rtwdev)
 	}
 }
 
-static
-void rtw8852b_set_trx_mask(struct rtw89_dev *rtwdev, u8 path, u8 group, u32 val)
-{
-	rtw89_write_rf(rtwdev, path, RR_LUTWE, RFREG_MASK, 0x20000);
-	rtw89_write_rf(rtwdev, path, RR_LUTWA, RFREG_MASK, group);
-	rtw89_write_rf(rtwdev, path, RR_LUTWD0, RFREG_MASK, val);
-	rtw89_write_rf(rtwdev, path, RR_LUTWE, RFREG_MASK, 0x0);
-}
-
-static void rtw8852b_btc_init_cfg(struct rtw89_dev *rtwdev)
-{
-	struct rtw89_btc *btc = &rtwdev->btc;
-	const struct rtw89_chip_info *chip = rtwdev->chip;
-	const struct rtw89_mac_ax_coex coex_params = {
-		.pta_mode = RTW89_MAC_AX_COEX_RTK_MODE,
-		.direction = RTW89_MAC_AX_COEX_INNER,
-	};
-
-	/* PTA init  */
-	rtw89_mac_coex_init(rtwdev, &coex_params);
-
-	/* set WL Tx response = Hi-Pri */
-	chip->ops->btc_set_wl_pri(rtwdev, BTC_PRI_MASK_TX_RESP, true);
-	chip->ops->btc_set_wl_pri(rtwdev, BTC_PRI_MASK_BEACON, true);
-
-	/* set rf gnt debug off */
-	rtw89_write_rf(rtwdev, RF_PATH_A, RR_WLSEL, RFREG_MASK, 0x0);
-	rtw89_write_rf(rtwdev, RF_PATH_B, RR_WLSEL, RFREG_MASK, 0x0);
-
-	/* set WL Tx thru in TRX mask table if GNT_WL = 0 && BT_S1 = ss group */
-	if (btc->ant_type == BTC_ANT_SHARED) {
-		rtw8852b_set_trx_mask(rtwdev, RF_PATH_A, BTC_BT_SS_GROUP, 0x5ff);
-		rtw8852b_set_trx_mask(rtwdev, RF_PATH_B, BTC_BT_SS_GROUP, 0x5ff);
-		/* set path-A(S0) Tx/Rx no-mask if GNT_WL=0 && BT_S1=tx group */
-		rtw8852b_set_trx_mask(rtwdev, RF_PATH_A, BTC_BT_TX_GROUP, 0x5ff);
-		rtw8852b_set_trx_mask(rtwdev, RF_PATH_B, BTC_BT_TX_GROUP, 0x55f);
-	} else { /* set WL Tx stb if GNT_WL = 0 && BT_S1 = ss group for 3-ant */
-		rtw8852b_set_trx_mask(rtwdev, RF_PATH_A, BTC_BT_SS_GROUP, 0x5df);
-		rtw8852b_set_trx_mask(rtwdev, RF_PATH_B, BTC_BT_SS_GROUP, 0x5df);
-		rtw8852b_set_trx_mask(rtwdev, RF_PATH_A, BTC_BT_TX_GROUP, 0x5ff);
-		rtw8852b_set_trx_mask(rtwdev, RF_PATH_B, BTC_BT_TX_GROUP, 0x5ff);
-	}
-
-	/* set PTA break table */
-	rtw89_write32(rtwdev, R_BTC_BREAK_TABLE, BTC_BREAK_PARAM);
-
-	 /* enable BT counter 0xda40[16,2] = 2b'11 */
-	rtw89_write32_set(rtwdev, R_AX_CSR_MODE, B_AX_BT_CNT_RST | B_AX_STATIS_BT_EN);
-	btc->cx.wl.status.map.init_ok = true;
-}
-
-static
-void rtw8852b_btc_set_wl_pri(struct rtw89_dev *rtwdev, u8 map, bool state)
-{
-	u32 bitmap;
-	u32 reg;
-
-	switch (map) {
-	case BTC_PRI_MASK_TX_RESP:
-		reg = R_BTC_BT_COEX_MSK_TABLE;
-		bitmap = B_BTC_PRI_MASK_TX_RESP_V1;
-		break;
-	case BTC_PRI_MASK_BEACON:
-		reg = R_AX_WL_PRI_MSK;
-		bitmap = B_AX_PTA_WL_PRI_MASK_BCNQ;
-		break;
-	case BTC_PRI_MASK_RX_CCK:
-		reg = R_BTC_BT_COEX_MSK_TABLE;
-		bitmap = B_BTC_PRI_MASK_RXCCK_V1;
-		break;
-	default:
-		return;
-	}
-
-	if (state)
-		rtw89_write32_set(rtwdev, reg, bitmap);
-	else
-		rtw89_write32_clr(rtwdev, reg, bitmap);
-}
-
 union rtw8852b_btc_wl_txpwr_ctrl {
 	u32 txpwr_val;
 	struct {
@@ -780,87 +700,6 @@ do {								\
 #undef __write_ctrl
 }
 
-static
-s8 rtw8852b_btc_get_bt_rssi(struct rtw89_dev *rtwdev, s8 val)
-{
-	/* +6 for compensate offset */
-	return clamp_t(s8, val + 6, -100, 0) + 100;
-}
-
-static
-void rtw8852b_btc_update_bt_cnt(struct rtw89_dev *rtwdev)
-{
-	/* Feature move to firmware */
-}
-
-static void rtw8852b_btc_wl_s1_standby(struct rtw89_dev *rtwdev, bool state)
-{
-	rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x80000);
-	rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x1);
-	rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD1, RFREG_MASK, 0x31);
-
-	/* set WL standby = Rx for GNT_BT_Tx = 1->0 settle issue */
-	if (state)
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x179);
-	else
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x20);
-
-	rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x0);
-}
-
-static void rtw8852b_btc_set_wl_lna2(struct rtw89_dev *rtwdev, u8 level)
-{
-	switch (level) {
-	case 0: /* default */
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x1000);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x0);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x1);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x17);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x2);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x3);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x17);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x0);
-		break;
-	case 1: /* Fix LNA2=5  */
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x1000);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x0);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x1);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x5);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x2);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x3);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x5);
-		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x0);
-		break;
-	}
-}
-
-static void rtw8852b_btc_set_wl_rx_gain(struct rtw89_dev *rtwdev, u32 level)
-{
-	struct rtw89_btc *btc = &rtwdev->btc;
-
-	switch (level) {
-	case 0: /* original */
-	default:
-		rtw8852bx_ctrl_nbtg_bt_tx(rtwdev, false, RTW89_PHY_0);
-		btc->dm.wl_lna2 = 0;
-		break;
-	case 1: /* for FDD free-run */
-		rtw8852bx_ctrl_nbtg_bt_tx(rtwdev, true, RTW89_PHY_0);
-		btc->dm.wl_lna2 = 0;
-		break;
-	case 2: /* for BTG Co-Rx*/
-		rtw8852bx_ctrl_nbtg_bt_tx(rtwdev, false, RTW89_PHY_0);
-		btc->dm.wl_lna2 = 1;
-		break;
-	}
-
-	rtw8852b_btc_set_wl_lna2(rtwdev, btc->dm.wl_lna2);
-}
-
 static const struct rtw89_chip_ops rtw8852b_chip_ops = {
 	.enable_bb_rf		= rtw8852bx_mac_enable_bb_rf,
 	.disable_bb_rf		= rtw8852bx_mac_disable_bb_rf,
@@ -911,13 +750,13 @@ static const struct rtw89_chip_ops rtw8852b_chip_ops = {
 	.h2c_ba_cam		= rtw89_fw_h2c_ba_cam,
 
 	.btc_set_rfe		= rtw8852b_btc_set_rfe,
-	.btc_init_cfg		= rtw8852b_btc_init_cfg,
-	.btc_set_wl_pri		= rtw8852b_btc_set_wl_pri,
+	.btc_init_cfg		= rtw8852bx_btc_init_cfg,
+	.btc_set_wl_pri		= rtw8852bx_btc_set_wl_pri,
 	.btc_set_wl_txpwr_ctrl	= rtw8852b_btc_set_wl_txpwr_ctrl,
-	.btc_get_bt_rssi	= rtw8852b_btc_get_bt_rssi,
-	.btc_update_bt_cnt	= rtw8852b_btc_update_bt_cnt,
-	.btc_wl_s1_standby	= rtw8852b_btc_wl_s1_standby,
-	.btc_set_wl_rx_gain	= rtw8852b_btc_set_wl_rx_gain,
+	.btc_get_bt_rssi	= rtw8852bx_btc_get_bt_rssi,
+	.btc_update_bt_cnt	= rtw8852bx_btc_update_bt_cnt,
+	.btc_wl_s1_standby	= rtw8852bx_btc_wl_s1_standby,
+	.btc_set_wl_rx_gain	= rtw8852bx_btc_set_wl_rx_gain,
 	.btc_set_policy		= rtw89_btc_set_policy_v1,
 };
 
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
index 60e9bd5b1376..f183dce16398 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
@@ -2,6 +2,7 @@
 /* Copyright(c) 2024  Realtek Corporation
  */
 
+#include "coex.h"
 #include "debug.h"
 #include "mac.h"
 #include "phy.h"
@@ -1749,6 +1750,167 @@ static u8 __rtw8852bx_get_thermal(struct rtw89_dev *rtwdev, enum rtw89_rf_path r
 	return rtw89_read_rf(rtwdev, rf_path, RR_TM, RR_TM_VAL);
 }
 
+static
+void rtw8852bx_set_trx_mask(struct rtw89_dev *rtwdev, u8 path, u8 group, u32 val)
+{
+	rtw89_write_rf(rtwdev, path, RR_LUTWE, RFREG_MASK, 0x20000);
+	rtw89_write_rf(rtwdev, path, RR_LUTWA, RFREG_MASK, group);
+	rtw89_write_rf(rtwdev, path, RR_LUTWD0, RFREG_MASK, val);
+	rtw89_write_rf(rtwdev, path, RR_LUTWE, RFREG_MASK, 0x0);
+}
+
+static void __rtw8852bx_btc_init_cfg(struct rtw89_dev *rtwdev)
+{
+	struct rtw89_btc *btc = &rtwdev->btc;
+	const struct rtw89_chip_info *chip = rtwdev->chip;
+	const struct rtw89_mac_ax_coex coex_params = {
+		.pta_mode = RTW89_MAC_AX_COEX_RTK_MODE,
+		.direction = RTW89_MAC_AX_COEX_INNER,
+	};
+
+	/* PTA init  */
+	rtw89_mac_coex_init(rtwdev, &coex_params);
+
+	/* set WL Tx response = Hi-Pri */
+	chip->ops->btc_set_wl_pri(rtwdev, BTC_PRI_MASK_TX_RESP, true);
+	chip->ops->btc_set_wl_pri(rtwdev, BTC_PRI_MASK_BEACON, true);
+
+	/* set rf gnt debug off */
+	rtw89_write_rf(rtwdev, RF_PATH_A, RR_WLSEL, RFREG_MASK, 0x0);
+	rtw89_write_rf(rtwdev, RF_PATH_B, RR_WLSEL, RFREG_MASK, 0x0);
+
+	/* set WL Tx thru in TRX mask table if GNT_WL = 0 && BT_S1 = ss group */
+	if (btc->ant_type == BTC_ANT_SHARED) {
+		rtw8852bx_set_trx_mask(rtwdev, RF_PATH_A, BTC_BT_SS_GROUP, 0x5ff);
+		rtw8852bx_set_trx_mask(rtwdev, RF_PATH_B, BTC_BT_SS_GROUP, 0x5ff);
+		/* set path-A(S0) Tx/Rx no-mask if GNT_WL=0 && BT_S1=tx group */
+		rtw8852bx_set_trx_mask(rtwdev, RF_PATH_A, BTC_BT_TX_GROUP, 0x5ff);
+		rtw8852bx_set_trx_mask(rtwdev, RF_PATH_B, BTC_BT_TX_GROUP, 0x55f);
+	} else { /* set WL Tx stb if GNT_WL = 0 && BT_S1 = ss group for 3-ant */
+		rtw8852bx_set_trx_mask(rtwdev, RF_PATH_A, BTC_BT_SS_GROUP, 0x5df);
+		rtw8852bx_set_trx_mask(rtwdev, RF_PATH_B, BTC_BT_SS_GROUP, 0x5df);
+		rtw8852bx_set_trx_mask(rtwdev, RF_PATH_A, BTC_BT_TX_GROUP, 0x5ff);
+		rtw8852bx_set_trx_mask(rtwdev, RF_PATH_B, BTC_BT_TX_GROUP, 0x5ff);
+	}
+
+	/* set PTA break table */
+	rtw89_write32(rtwdev, R_BTC_BREAK_TABLE, BTC_BREAK_PARAM);
+
+	 /* enable BT counter 0xda40[16,2] = 2b'11 */
+	rtw89_write32_set(rtwdev, R_AX_CSR_MODE, B_AX_BT_CNT_RST | B_AX_STATIS_BT_EN);
+	btc->cx.wl.status.map.init_ok = true;
+}
+
+static
+void __rtw8852bx_btc_set_wl_pri(struct rtw89_dev *rtwdev, u8 map, bool state)
+{
+	u32 bitmap;
+	u32 reg;
+
+	switch (map) {
+	case BTC_PRI_MASK_TX_RESP:
+		reg = R_BTC_BT_COEX_MSK_TABLE;
+		bitmap = B_BTC_PRI_MASK_TX_RESP_V1;
+		break;
+	case BTC_PRI_MASK_BEACON:
+		reg = R_AX_WL_PRI_MSK;
+		bitmap = B_AX_PTA_WL_PRI_MASK_BCNQ;
+		break;
+	case BTC_PRI_MASK_RX_CCK:
+		reg = R_BTC_BT_COEX_MSK_TABLE;
+		bitmap = B_BTC_PRI_MASK_RXCCK_V1;
+		break;
+	default:
+		return;
+	}
+
+	if (state)
+		rtw89_write32_set(rtwdev, reg, bitmap);
+	else
+		rtw89_write32_clr(rtwdev, reg, bitmap);
+}
+
+static
+s8 __rtw8852bx_btc_get_bt_rssi(struct rtw89_dev *rtwdev, s8 val)
+{
+	/* +6 for compensate offset */
+	return clamp_t(s8, val + 6, -100, 0) + 100;
+}
+
+static
+void __rtw8852bx_btc_update_bt_cnt(struct rtw89_dev *rtwdev)
+{
+	/* Feature move to firmware */
+}
+
+static void __rtw8852bx_btc_wl_s1_standby(struct rtw89_dev *rtwdev, bool state)
+{
+	rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x80000);
+	rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x1);
+	rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD1, RFREG_MASK, 0x31);
+
+	/* set WL standby = Rx for GNT_BT_Tx = 1->0 settle issue */
+	if (state)
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x179);
+	else
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x20);
+
+	rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x0);
+}
+
+static void rtw8852bx_btc_set_wl_lna2(struct rtw89_dev *rtwdev, u8 level)
+{
+	switch (level) {
+	case 0: /* default */
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x1000);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x0);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x1);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x17);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x2);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x3);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x17);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x0);
+		break;
+	case 1: /* Fix LNA2=5  */
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x1000);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x0);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x1);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x5);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x2);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x15);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWA, RFREG_MASK, 0x3);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWD0, RFREG_MASK, 0x5);
+		rtw89_write_rf(rtwdev, RF_PATH_B, RR_LUTWE, RFREG_MASK, 0x0);
+		break;
+	}
+}
+
+static void __rtw8852bx_btc_set_wl_rx_gain(struct rtw89_dev *rtwdev, u32 level)
+{
+	struct rtw89_btc *btc = &rtwdev->btc;
+
+	switch (level) {
+	case 0: /* original */
+	default:
+		rtw8852bx_ctrl_nbtg_bt_tx(rtwdev, false, RTW89_PHY_0);
+		btc->dm.wl_lna2 = 0;
+		break;
+	case 1: /* for FDD free-run */
+		rtw8852bx_ctrl_nbtg_bt_tx(rtwdev, true, RTW89_PHY_0);
+		btc->dm.wl_lna2 = 0;
+		break;
+	case 2: /* for BTG Co-Rx*/
+		rtw8852bx_ctrl_nbtg_bt_tx(rtwdev, false, RTW89_PHY_0);
+		btc->dm.wl_lna2 = 1;
+		break;
+	}
+
+	rtw8852bx_btc_set_wl_lna2(rtwdev, btc->dm.wl_lna2);
+}
+
 static void rtw8852bx_fill_freq_with_ppdu(struct rtw89_dev *rtwdev,
 					  struct rtw89_rx_phy_ppdu *phy_ppdu,
 					  struct ieee80211_rx_status *status)
@@ -1872,6 +2034,12 @@ const struct rtw8852bx_info rtw8852bx_info = {
 	.set_txpwr_ul_tb_offset = __rtw8852bx_set_txpwr_ul_tb_offset,
 	.get_thermal = __rtw8852bx_get_thermal,
 	.adc_cfg = rtw8852bt_adc_cfg,
+	.btc_init_cfg = __rtw8852bx_btc_init_cfg,
+	.btc_set_wl_pri = __rtw8852bx_btc_set_wl_pri,
+	.btc_get_bt_rssi = __rtw8852bx_btc_get_bt_rssi,
+	.btc_update_bt_cnt = __rtw8852bx_btc_update_bt_cnt,
+	.btc_wl_s1_standby = __rtw8852bx_btc_wl_s1_standby,
+	.btc_set_wl_rx_gain = __rtw8852bx_btc_set_wl_rx_gain,
 };
 EXPORT_SYMBOL(rtw8852bx_info);
 
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.h b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.h
index b8e3f20ba189..801e7ab9f4fa 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.h
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.h
@@ -159,6 +159,12 @@ struct rtw8852bx_info {
 				       s8 pw_ofst, enum rtw89_mac_idx mac_idx);
 	u8 (*get_thermal)(struct rtw89_dev *rtwdev, enum rtw89_rf_path rf_path);
 	void (*adc_cfg)(struct rtw89_dev *rtwdev, u8 bw, u8 path);
+	void (*btc_init_cfg)(struct rtw89_dev *rtwdev);
+	void (*btc_set_wl_pri)(struct rtw89_dev *rtwdev, u8 map, bool state);
+	s8 (*btc_get_bt_rssi)(struct rtw89_dev *rtwdev, s8 val);
+	void (*btc_update_bt_cnt)(struct rtw89_dev *rtwdev);
+	void (*btc_wl_s1_standby)(struct rtw89_dev *rtwdev, bool state);
+	void (*btc_set_wl_rx_gain)(struct rtw89_dev *rtwdev, u32 level);
 };
 
 extern const struct rtw8852bx_info rtw8852bx_info;
@@ -343,4 +349,40 @@ void rtw8852bx_adc_cfg(struct rtw89_dev *rtwdev, u8 bw, u8 path)
 	rtw8852bx_info.adc_cfg(rtwdev, bw, path);
 }
 
+static inline
+void rtw8852bx_btc_init_cfg(struct rtw89_dev *rtwdev)
+{
+	rtw8852bx_info.btc_init_cfg(rtwdev);
+}
+
+static inline
+void rtw8852bx_btc_set_wl_pri(struct rtw89_dev *rtwdev, u8 map, bool state)
+{
+	rtw8852bx_info.btc_set_wl_pri(rtwdev, map, state);
+}
+
+static inline
+s8 rtw8852bx_btc_get_bt_rssi(struct rtw89_dev *rtwdev, s8 val)
+{
+	return rtw8852bx_info.btc_get_bt_rssi(rtwdev, val);
+}
+
+static inline
+void rtw8852bx_btc_update_bt_cnt(struct rtw89_dev *rtwdev)
+{
+	rtw8852bx_info.btc_update_bt_cnt(rtwdev);
+}
+
+static inline
+void rtw8852bx_btc_wl_s1_standby(struct rtw89_dev *rtwdev, bool state)
+{
+	rtw8852bx_info.btc_wl_s1_standby(rtwdev, state);
+}
+
+static inline
+void rtw8852bx_btc_set_wl_rx_gain(struct rtw89_dev *rtwdev, u32 level)
+{
+	rtw8852bx_info.btc_set_wl_rx_gain(rtwdev, level);
+}
+
 #endif
-- 
cgit v1.2.3-58-ga151


From 52bc83ad2ed5b115c33e3acc19cbf636dd2f37c9 Mon Sep 17 00:00:00 2001
From: Ping-Ke Shih <pkshih@realtek.com>
Date: Mon, 1 Jul 2024 09:46:19 +0800
Subject: wifi: rtw89: 8852bx: add extra handles of BTC for 8852BT in
 8852b_common

For 8852BT, the initial settings of BT-coexistence is a little bit
different, so add the extra handles.

Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240701014619.7300-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/rtw8852b_common.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
index f183dce16398..1745c2882acf 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852b_common.c
@@ -1793,6 +1793,11 @@ static void __rtw8852bx_btc_init_cfg(struct rtw89_dev *rtwdev)
 		rtw8852bx_set_trx_mask(rtwdev, RF_PATH_B, BTC_BT_TX_GROUP, 0x5ff);
 	}
 
+	if (rtwdev->chip->chip_id == RTL8852BT) {
+		rtw8852bx_set_trx_mask(rtwdev, RF_PATH_A, BTC_BT_RX_GROUP, 0x5df);
+		rtw8852bx_set_trx_mask(rtwdev, RF_PATH_B, BTC_BT_RX_GROUP, 0x5df);
+	}
+
 	/* set PTA break table */
 	rtw89_write32(rtwdev, R_BTC_BREAK_TABLE, BTC_BREAK_PARAM);
 
-- 
cgit v1.2.3-58-ga151


From 1e71be6a344628d3a6041dfcf26220ad5d66e699 Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Tue, 2 Jul 2024 20:44:51 +0800
Subject: wifi: rtw89: mac: parse MRC C2H failure report

MRC (multi-role concurrency) has a C2H event for status report. Newer
FW will report some kinds of failures. We parse them now and show by
debug log.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240702124452.18747-1-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/mac.c | 40 ++++++++++++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/mac.h | 10 ++++++++
 2 files changed, 50 insertions(+)

diff --git a/drivers/net/wireless/realtek/rtw89/mac.c b/drivers/net/wireless/realtek/rtw89/mac.c
index 73462f3343e3..e2399796aeb1 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.c
+++ b/drivers/net/wireless/realtek/rtw89/mac.c
@@ -5202,6 +5202,46 @@ rtw89_mac_c2h_mrc_status_rpt(struct rtw89_dev *rtwdev, struct sk_buff *c2h, u32
 	case RTW89_MAC_MRC_DEL_SCH_OK:
 		func = H2C_FUNC_DEL_MRC;
 		break;
+	case RTW89_MAC_MRC_EMPTY_SCH_FAIL:
+		rtw89_debug(rtwdev, RTW89_DBG_CHAN,
+			    "MRC C2H STS RPT: empty sch fail\n");
+		return;
+	case RTW89_MAC_MRC_ROLE_NOT_EXIST_FAIL:
+		rtw89_debug(rtwdev, RTW89_DBG_CHAN,
+			    "MRC C2H STS RPT: role not exist fail\n");
+		return;
+	case RTW89_MAC_MRC_DATA_NOT_FOUND_FAIL:
+		rtw89_debug(rtwdev, RTW89_DBG_CHAN,
+			    "MRC C2H STS RPT: data not found fail\n");
+		return;
+	case RTW89_MAC_MRC_GET_NEXT_SLOT_FAIL:
+		rtw89_debug(rtwdev, RTW89_DBG_CHAN,
+			    "MRC C2H STS RPT: get next slot fail\n");
+		return;
+	case RTW89_MAC_MRC_ALT_ROLE_FAIL:
+		rtw89_debug(rtwdev, RTW89_DBG_CHAN,
+			    "MRC C2H STS RPT: alt role fail\n");
+		return;
+	case RTW89_MAC_MRC_ADD_PSTIMER_FAIL:
+		rtw89_debug(rtwdev, RTW89_DBG_CHAN,
+			    "MRC C2H STS RPT: add ps timer fail\n");
+		return;
+	case RTW89_MAC_MRC_MALLOC_FAIL:
+		rtw89_debug(rtwdev, RTW89_DBG_CHAN,
+			    "MRC C2H STS RPT: malloc fail\n");
+		return;
+	case RTW89_MAC_MRC_SWITCH_CH_FAIL:
+		rtw89_debug(rtwdev, RTW89_DBG_CHAN,
+			    "MRC C2H STS RPT: switch ch fail\n");
+		return;
+	case RTW89_MAC_MRC_TXNULL0_FAIL:
+		rtw89_debug(rtwdev, RTW89_DBG_CHAN,
+			    "MRC C2H STS RPT: tx null-0 fail\n");
+		return;
+	case RTW89_MAC_MRC_PORT_FUNC_EN_FAIL:
+		rtw89_debug(rtwdev, RTW89_DBG_CHAN,
+			    "MRC C2H STS RPT: port func en fail\n");
+		return;
 	default:
 		rtw89_debug(rtwdev, RTW89_DBG_CHAN,
 			    "invalid MRC C2H STS RPT: status %d\n", status);
diff --git a/drivers/net/wireless/realtek/rtw89/mac.h b/drivers/net/wireless/realtek/rtw89/mac.h
index cc9f914d440c..d5895516b3ed 100644
--- a/drivers/net/wireless/realtek/rtw89/mac.h
+++ b/drivers/net/wireless/realtek/rtw89/mac.h
@@ -466,6 +466,16 @@ enum rtw89_mac_mrc_status {
 	RTW89_MAC_MRC_START_SCH_OK = 0,
 	RTW89_MAC_MRC_STOP_SCH_OK = 1,
 	RTW89_MAC_MRC_DEL_SCH_OK = 2,
+	RTW89_MAC_MRC_EMPTY_SCH_FAIL = 16,
+	RTW89_MAC_MRC_ROLE_NOT_EXIST_FAIL = 17,
+	RTW89_MAC_MRC_DATA_NOT_FOUND_FAIL = 18,
+	RTW89_MAC_MRC_GET_NEXT_SLOT_FAIL = 19,
+	RTW89_MAC_MRC_ALT_ROLE_FAIL = 20,
+	RTW89_MAC_MRC_ADD_PSTIMER_FAIL = 21,
+	RTW89_MAC_MRC_MALLOC_FAIL = 22,
+	RTW89_MAC_MRC_SWITCH_CH_FAIL = 23,
+	RTW89_MAC_MRC_TXNULL0_FAIL = 24,
+	RTW89_MAC_MRC_PORT_FUNC_EN_FAIL = 25,
 };
 
 struct rtw89_mac_ax_coex {
-- 
cgit v1.2.3-58-ga151


From 8095364696d38092f34bbd3989b1a73b180a3aaf Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Tue, 2 Jul 2024 20:44:52 +0800
Subject: wifi: rtw89: unify the selection logic of RFK table when MCC

Driver will notify FW the target index of RFK table to use at some
moments. When MCC (multi-channel concurrent), the correctness of the
notification is especially important.

We now unify the selection logic of RFK table as below among chips.
1. check each table if it matches target channel
2. check all tables if any is idle by iterating active channels
3. replace the first table if all are busy unexpectedly

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240702124452.18747-2-pkshih@realtek.com
---
 drivers/net/wireless/realtek/rtw89/chan.c         | 22 ++++++++
 drivers/net/wireless/realtek/rtw89/chan.h         |  4 ++
 drivers/net/wireless/realtek/rtw89/core.h         |  7 ++-
 drivers/net/wireless/realtek/rtw89/phy.c          | 69 +++++++++++++++++++++++
 drivers/net/wireless/realtek/rtw89/phy.h          | 16 ++++++
 drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c | 32 +++--------
 drivers/net/wireless/realtek/rtw89/rtw8922a_rfk.c | 17 +++++-
 7 files changed, 141 insertions(+), 26 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtw89/chan.c b/drivers/net/wireless/realtek/rtw89/chan.c
index 3b1997223cc5..7f90d93dcdc0 100644
--- a/drivers/net/wireless/realtek/rtw89/chan.c
+++ b/drivers/net/wireless/realtek/rtw89/chan.c
@@ -141,6 +141,28 @@ bool rtw89_assign_entity_chan(struct rtw89_dev *rtwdev,
 	return band_changed;
 }
 
+int rtw89_iterate_entity_chan(struct rtw89_dev *rtwdev,
+			      int (*iterator)(const struct rtw89_chan *chan,
+					      void *data),
+			      void *data)
+{
+	struct rtw89_hal *hal = &rtwdev->hal;
+	const struct rtw89_chan *chan;
+	int ret;
+	u8 idx;
+
+	lockdep_assert_held(&rtwdev->mutex);
+
+	for_each_set_bit(idx,  hal->entity_map, NUM_OF_RTW89_SUB_ENTITY) {
+		chan = rtw89_chan_get(rtwdev, idx);
+		ret = iterator(chan, data);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static void __rtw89_config_entity_chandef(struct rtw89_dev *rtwdev,
 					  enum rtw89_sub_entity_idx idx,
 					  const struct cfg80211_chan_def *chandef,
diff --git a/drivers/net/wireless/realtek/rtw89/chan.h b/drivers/net/wireless/realtek/rtw89/chan.h
index ffa412f281f3..5278ff8c513b 100644
--- a/drivers/net/wireless/realtek/rtw89/chan.h
+++ b/drivers/net/wireless/realtek/rtw89/chan.h
@@ -78,6 +78,10 @@ void rtw89_chan_create(struct rtw89_chan *chan, u8 center_chan, u8 primary_chan,
 bool rtw89_assign_entity_chan(struct rtw89_dev *rtwdev,
 			      enum rtw89_sub_entity_idx idx,
 			      const struct rtw89_chan *new);
+int rtw89_iterate_entity_chan(struct rtw89_dev *rtwdev,
+			      int (*iterator)(const struct rtw89_chan *chan,
+					      void *data),
+			      void *data);
 void rtw89_config_entity_chandef(struct rtw89_dev *rtwdev,
 				 enum rtw89_sub_entity_idx idx,
 				 const struct cfg80211_chan_def *chandef);
diff --git a/drivers/net/wireless/realtek/rtw89/core.h b/drivers/net/wireless/realtek/rtw89/core.h
index 001da3ad149e..11fa003a9788 100644
--- a/drivers/net/wireless/realtek/rtw89/core.h
+++ b/drivers/net/wireless/realtek/rtw89/core.h
@@ -4695,7 +4695,12 @@ struct rtw89_dack_info {
 	bool msbk_timeout[RTW89_DACK_PATH_NR];
 };
 
-#define RTW89_RFK_CHS_NR 3
+enum rtw89_rfk_chs_nrs {
+	__RTW89_RFK_CHS_NR_V0 = 2,
+	__RTW89_RFK_CHS_NR_V1 = 3,
+
+	RTW89_RFK_CHS_NR = __RTW89_RFK_CHS_NR_V1,
+};
 
 struct rtw89_rfk_mcc_info {
 	u8 ch[RTW89_RFK_CHS_NR];
diff --git a/drivers/net/wireless/realtek/rtw89/phy.c b/drivers/net/wireless/realtek/rtw89/phy.c
index e025f0ed9f59..ad11d1414874 100644
--- a/drivers/net/wireless/realtek/rtw89/phy.c
+++ b/drivers/net/wireless/realtek/rtw89/phy.c
@@ -2,6 +2,7 @@
 /* Copyright(c) 2019-2020  Realtek Corporation
  */
 
+#include "chan.h"
 #include "coex.h"
 #include "debug.h"
 #include "fw.h"
@@ -6003,6 +6004,74 @@ void rtw89_phy_set_bss_color(struct rtw89_dev *rtwdev, struct ieee80211_vif *vif
 			      vif->cfg.aid, phy_idx);
 }
 
+static bool rfk_chan_validate_desc(const struct rtw89_rfk_chan_desc *desc)
+{
+	return desc->ch != 0;
+}
+
+static bool rfk_chan_is_equivalent(const struct rtw89_rfk_chan_desc *desc,
+				   const struct rtw89_chan *chan)
+{
+	if (!rfk_chan_validate_desc(desc))
+		return false;
+
+	if (desc->ch != chan->channel)
+		return false;
+
+	if (desc->has_band && desc->band != chan->band_type)
+		return false;
+
+	if (desc->has_bw && desc->bw != chan->band_width)
+		return false;
+
+	return true;
+}
+
+struct rfk_chan_iter_data {
+	const struct rtw89_rfk_chan_desc desc;
+	unsigned int found;
+};
+
+static int rfk_chan_iter_search(const struct rtw89_chan *chan, void *data)
+{
+	struct rfk_chan_iter_data *iter_data = data;
+
+	if (rfk_chan_is_equivalent(&iter_data->desc, chan))
+		iter_data->found++;
+
+	return 0;
+}
+
+u8 rtw89_rfk_chan_lookup(struct rtw89_dev *rtwdev,
+			 const struct rtw89_rfk_chan_desc *desc, u8 desc_nr,
+			 const struct rtw89_chan *target_chan)
+{
+	int sel = -1;
+	u8 i;
+
+	for (i = 0; i < desc_nr; i++) {
+		struct rfk_chan_iter_data iter_data = {
+			.desc = desc[i],
+		};
+
+		if (rfk_chan_is_equivalent(&desc[i], target_chan))
+			return i;
+
+		rtw89_iterate_entity_chan(rtwdev, rfk_chan_iter_search, &iter_data);
+		if (!iter_data.found && sel == -1)
+			sel = i;
+	}
+
+	if (sel == -1) {
+		rtw89_debug(rtwdev, RTW89_DBG_RFK,
+			    "no idle rfk entry; force replace the first\n");
+		sel = 0;
+	}
+
+	return sel;
+}
+EXPORT_SYMBOL(rtw89_rfk_chan_lookup);
+
 static void
 _rfk_write_rf(struct rtw89_dev *rtwdev, const struct rtw89_reg5_def *def)
 {
diff --git a/drivers/net/wireless/realtek/rtw89/phy.h b/drivers/net/wireless/realtek/rtw89/phy.h
index 6e53ac9140fb..d8df553b9cb0 100644
--- a/drivers/net/wireless/realtek/rtw89/phy.h
+++ b/drivers/net/wireless/realtek/rtw89/phy.h
@@ -715,6 +715,19 @@ enum rtw89_phy_gain_band_be rtw89_subband_to_gain_band_be(enum rtw89_subband sub
 	}
 }
 
+struct rtw89_rfk_chan_desc {
+	/* desc is valid iff ch is non-zero */
+	u8 ch;
+
+	/* To avoid us from extending old chip code every time, each new
+	 * field must be defined along with a bool flag in positivte way.
+	 */
+	bool has_band;
+	u8 band;
+	bool has_bw;
+	u8 bw;
+};
+
 enum rtw89_rfk_flag {
 	RTW89_RFK_F_WRF = 0,
 	RTW89_RFK_F_WM = 1,
@@ -950,5 +963,8 @@ enum rtw89_rf_path_bit rtw89_phy_get_kpath(struct rtw89_dev *rtwdev,
 					   enum rtw89_phy_idx phy_idx);
 enum rtw89_rf_path rtw89_phy_get_syn_sel(struct rtw89_dev *rtwdev,
 					 enum rtw89_phy_idx phy_idx);
+u8 rtw89_rfk_chan_lookup(struct rtw89_dev *rtwdev,
+			 const struct rtw89_rfk_chan_desc *desc, u8 desc_nr,
+			 const struct rtw89_chan *target_chan);
 
 #endif
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c
index 654e3e5507cb..743f7014bf3e 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8852c_rfk.c
@@ -4070,12 +4070,11 @@ void rtw8852c_set_channel_rf(struct rtw89_dev *rtwdev,
 void rtw8852c_mcc_get_ch_info(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_idx)
 {
 	struct rtw89_rfk_mcc_info *rfk_mcc = &rtwdev->rfk_mcc;
-	DECLARE_BITMAP(map, RTW89_IQK_CHS_NR) = {};
+	struct rtw89_rfk_chan_desc desc[__RTW89_RFK_CHS_NR_V0] = {};
 	const struct rtw89_chan *chan;
 	enum rtw89_entity_mode mode;
 	u8 chan_idx;
 	u8 idx;
-	u8 i;
 
 	mode = rtw89_get_entity_mode(rtwdev);
 	switch (mode) {
@@ -4087,34 +4086,21 @@ void rtw8852c_mcc_get_ch_info(struct rtw89_dev *rtwdev, enum rtw89_phy_idx phy_i
 		break;
 	}
 
-	for (i = 0; i <= chan_idx; i++) {
-		chan = rtw89_chan_get(rtwdev, i);
+	chan = rtw89_chan_get(rtwdev, chan_idx);
 
-		for (idx = 0; idx < RTW89_IQK_CHS_NR; idx++) {
-			if (rfk_mcc->ch[idx] == chan->channel &&
-			    rfk_mcc->band[idx] == chan->band_type) {
-				if (i != chan_idx) {
-					set_bit(idx, map);
-					break;
-				}
+	for (idx = 0; idx < ARRAY_SIZE(desc); idx++) {
+		struct rtw89_rfk_chan_desc *p = &desc[idx];
 
-				goto bottom;
-			}
-		}
-	}
+		p->ch = rfk_mcc->ch[idx];
 
-	idx = find_first_zero_bit(map, RTW89_IQK_CHS_NR);
-	if (idx == RTW89_IQK_CHS_NR) {
-		rtw89_debug(rtwdev, RTW89_DBG_RFK,
-			    "%s: no empty rfk table; force replace the first\n",
-			    __func__);
-		idx = 0;
+		p->has_band = true;
+		p->band = rfk_mcc->band[idx];
 	}
 
+	idx = rtw89_rfk_chan_lookup(rtwdev, desc, ARRAY_SIZE(desc), chan);
+
 	rfk_mcc->ch[idx] = chan->channel;
 	rfk_mcc->band[idx] = chan->band_type;
-
-bottom:
 	rfk_mcc->table_idx = idx;
 }
 
diff --git a/drivers/net/wireless/realtek/rtw89/rtw8922a_rfk.c b/drivers/net/wireless/realtek/rtw89/rtw8922a_rfk.c
index 2a371829268c..0ebcb06ae848 100644
--- a/drivers/net/wireless/realtek/rtw89/rtw8922a_rfk.c
+++ b/drivers/net/wireless/realtek/rtw89/rtw8922a_rfk.c
@@ -255,6 +255,7 @@ static void rtw8922a_chlk_ktbl_sel(struct rtw89_dev *rtwdev, u8 kpath, u8 idx)
 static void rtw8922a_chlk_reload(struct rtw89_dev *rtwdev)
 {
 	struct rtw89_rfk_mcc_info *rfk_mcc = &rtwdev->rfk_mcc;
+	struct rtw89_rfk_chan_desc desc[__RTW89_RFK_CHS_NR_V1] = {};
 	enum rtw89_sub_entity_idx sub_entity_idx;
 	const struct rtw89_chan *chan;
 	enum rtw89_entity_mode mode;
@@ -265,16 +266,28 @@ static void rtw8922a_chlk_reload(struct rtw89_dev *rtwdev)
 	switch (mode) {
 	case RTW89_ENTITY_MODE_MCC_PREPARE:
 		sub_entity_idx = RTW89_SUB_ENTITY_1;
-		tbl_sel = 1;
 		break;
 	default:
 		sub_entity_idx = RTW89_SUB_ENTITY_0;
-		tbl_sel = 0;
 		break;
 	}
 
 	chan = rtw89_chan_get(rtwdev, sub_entity_idx);
 
+	for (tbl_sel = 0; tbl_sel < ARRAY_SIZE(desc); tbl_sel++) {
+		struct rtw89_rfk_chan_desc *p = &desc[tbl_sel];
+
+		p->ch = rfk_mcc->ch[tbl_sel];
+
+		p->has_band = true;
+		p->band = rfk_mcc->band[tbl_sel];
+
+		p->has_bw = true;
+		p->bw = rfk_mcc->bw[tbl_sel];
+	}
+
+	tbl_sel = rtw89_rfk_chan_lookup(rtwdev, desc, ARRAY_SIZE(desc), chan);
+
 	rfk_mcc->ch[tbl_sel] = chan->channel;
 	rfk_mcc->band[tbl_sel] = chan->band_type;
 	rfk_mcc->bw[tbl_sel] = chan->band_width;
-- 
cgit v1.2.3-58-ga151


From a1e7eafd12c415a6ac2a65d8ddf32569acd1ecf0 Mon Sep 17 00:00:00 2001
From: Fredrik Lönnegren <fredrik@frelon.se>
Date: Wed, 3 Jul 2024 09:06:23 +0200
Subject: wifi: rtlwifi: fix default typo
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Change 'defult' to 'default' in comments in several rtlwifi drivers.

Signed-off-by: Fredrik Lönnegren <fredrik@frelon.se>
Acked-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240703070627.135328-1-fredrik@frelon.se
---
 drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.c | 2 +-
 drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.c | 2 +-
 drivers/net/wireless/realtek/rtlwifi/rtl8192de/sw.c | 2 +-
 drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.c | 2 +-
 drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.c | 2 +-
 drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c | 2 +-
 drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.c | 2 +-
 drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.c | 2 +-
 8 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.c
index 37bb59fa8bfa..35875cda30fc 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/sw.c
@@ -27,7 +27,7 @@ static void rtl88e_init_aspm_vars(struct ieee80211_hw *hw)
 	 * 2 - Enable ASPM with Clock Req,
 	 * 3 - Alwyas Enable ASPM with Clock Req,
 	 * 4 - Always Enable ASPM without Clock Req.
-	 * set defult to RTL8192CE:3 RTL8192E:2
+	 * set default to RTL8192CE:3 RTL8192E:2
 	 */
 	rtlpci->const_pci_aspm = 3;
 
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.c
index e20f2bec45c4..ce7c28d9c874 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/sw.c
@@ -31,7 +31,7 @@ static void rtl92c_init_aspm_vars(struct ieee80211_hw *hw)
 	 * 2 - Enable ASPM with Clock Req,
 	 * 3 - Alwyas Enable ASPM with Clock Req,
 	 * 4 - Always Enable ASPM without Clock Req.
-	 * set defult to RTL8192CE:3 RTL8192E:2
+	 * set default to RTL8192CE:3 RTL8192E:2
 	 * */
 	rtlpci->const_pci_aspm = 3;
 
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/sw.c
index f5ce4889523e..e36e4aeb9a95 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/sw.c
@@ -32,7 +32,7 @@ static void rtl92d_init_aspm_vars(struct ieee80211_hw *hw)
 	 * 2 - Enable ASPM with Clock Req,
 	 * 3 - Alwyas Enable ASPM with Clock Req,
 	 * 4 - Always Enable ASPM without Clock Req.
-	 * set defult to RTL8192CE:3 RTL8192E:2
+	 * set default to RTL8192CE:3 RTL8192E:2
 	 * */
 	rtlpci->const_pci_aspm = 3;
 
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.c
index 7bde20fdbeab..162e734d5b08 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/sw.c
@@ -31,7 +31,7 @@ static void rtl92ee_init_aspm_vars(struct ieee80211_hw *hw)
 	 * 2 - Enable ASPM with Clock Req,
 	 * 3 - Alwyas Enable ASPM with Clock Req,
 	 * 4 - Always Enable ASPM without Clock Req.
-	 * set defult to RTL8192CE:3 RTL8192E:2
+	 * set default to RTL8192CE:3 RTL8192E:2
 	 */
 	rtlpci->const_pci_aspm = 3;
 
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.c
index 675bdd32feb1..bbf8ff63dced 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/sw.c
@@ -27,7 +27,7 @@ static void rtl92s_init_aspm_vars(struct ieee80211_hw *hw)
 	 * 2 - Enable ASPM with Clock Req,
 	 * 3 - Alwyas Enable ASPM with Clock Req,
 	 * 4 - Always Enable ASPM without Clock Req.
-	 * set defult to RTL8192CE:3 RTL8192E:2
+	 * set default to RTL8192CE:3 RTL8192E:2
 	 * */
 	rtlpci->const_pci_aspm = 2;
 
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c
index dd7505e2f22c..1b144fbd4d26 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c
@@ -33,7 +33,7 @@ static void rtl8723e_init_aspm_vars(struct ieee80211_hw *hw)
 	 * 2 - Enable ASPM with Clock Req,
 	 * 3 - Alwyas Enable ASPM with Clock Req,
 	 * 4 - Always Enable ASPM without Clock Req.
-	 * set defult to RTL8192CE:3 RTL8192E:2
+	 * set default to RTL8192CE:3 RTL8192E:2
 	 */
 	rtlpci->const_pci_aspm = 3;
 
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.c
index 162c34f0e9b7..0a92d0325098 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/sw.c
@@ -32,7 +32,7 @@ static void rtl8723be_init_aspm_vars(struct ieee80211_hw *hw)
 	 * 2 - Enable ASPM with Clock Req,
 	 * 3 - Alwyas Enable ASPM with Clock Req,
 	 * 4 - Always Enable ASPM without Clock Req.
-	 * set defult to RTL8192CE:3 RTL8192E:2
+	 * set default to RTL8192CE:3 RTL8192E:2
 	 */
 	rtlpci->const_pci_aspm = 3;
 
diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.c
index 7b911695db33..a65503c5ae5a 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/sw.c
@@ -30,7 +30,7 @@ static void rtl8821ae_init_aspm_vars(struct ieee80211_hw *hw)
 	 * 2 - Enable ASPM with Clock Req,
 	 * 3 - Alwyas Enable ASPM with Clock Req,
 	 * 4 - Always Enable ASPM without Clock Req.
-	 * set defult to RTL8192CE:3 RTL8192E:2
+	 * set default to RTL8192CE:3 RTL8192E:2
 	 */
 	rtlpci->const_pci_aspm = 3;
 
-- 
cgit v1.2.3-58-ga151


From 85099c7ce4f9e64c66aa397cd9a37473637ab891 Mon Sep 17 00:00:00 2001
From: Aleksandr Mishin <amishin@t-argos.ru>
Date: Thu, 4 Jul 2024 00:05:10 +0300
Subject: wifi: rtw89: Fix array index mistake in rtw89_sta_info_get_iter()

In rtw89_sta_info_get_iter() 'status->he_gi' is compared to array size.
But then 'rate->he_gi' is used as array index instead of 'status->he_gi'.
This can lead to go beyond array boundaries in case of 'rate->he_gi' is
not equal to 'status->he_gi' and is bigger than array size. Looks like
"copy-paste" mistake.

Fix this mistake by replacing 'rate->he_gi' with 'status->he_gi'.

Found by Linux Verification Center (linuxtesting.org) with SVACE.

Fixes: e3ec7017f6a2 ("rtw89: add Realtek 802.11ax driver")
Signed-off-by: Aleksandr Mishin <amishin@t-argos.ru>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Link: https://patch.msgid.link/20240703210510.11089-1-amishin@t-argos.ru
---
 drivers/net/wireless/realtek/rtw89/debug.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/realtek/rtw89/debug.c b/drivers/net/wireless/realtek/rtw89/debug.c
index 9cbf136aa2ac..9e1353cce9cc 100644
--- a/drivers/net/wireless/realtek/rtw89/debug.c
+++ b/drivers/net/wireless/realtek/rtw89/debug.c
@@ -3552,7 +3552,7 @@ static void rtw89_sta_info_get_iter(void *data, struct ieee80211_sta *sta)
 	case RX_ENC_HE:
 		seq_printf(m, "HE %dSS MCS-%d GI:%s", status->nss, status->rate_idx,
 			   status->he_gi <= NL80211_RATE_INFO_HE_GI_3_2 ?
-			   he_gi_str[rate->he_gi] : "N/A");
+			   he_gi_str[status->he_gi] : "N/A");
 		break;
 	case RX_ENC_EHT:
 		seq_printf(m, "EHT %dSS MCS-%d GI:%s", status->nss, status->rate_idx,
-- 
cgit v1.2.3-58-ga151


From 390b14b5e9f6e987381a5dc96b3d636000a44b53 Mon Sep 17 00:00:00 2001
From: "Rob Herring (Arm)" <robh@kernel.org>
Date: Wed, 3 Jul 2024 13:58:27 -0600
Subject: dt-bindings: net: Define properties at top-level

Convention is DT schemas should define all properties at the top-level
and not inside of if/then schemas. That minimizes the if/then schemas
and is more future proof.

Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Serge Semin <fancer.lancer@gmail.com>
Link: https://patch.msgid.link/20240703195827.1670594-2-robh@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../devicetree/bindings/net/mediatek,net.yaml      | 28 +++++----
 .../devicetree/bindings/net/snps,dwmac.yaml        | 67 +++++++++++-----------
 2 files changed, 52 insertions(+), 43 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/mediatek,net.yaml b/Documentation/devicetree/bindings/net/mediatek,net.yaml
index 3202dc7967c5..686b5c2fae40 100644
--- a/Documentation/devicetree/bindings/net/mediatek,net.yaml
+++ b/Documentation/devicetree/bindings/net/mediatek,net.yaml
@@ -68,6 +68,17 @@ properties:
       Phandle to the syscon node that handles the path from GMAC to
       PHY variants.
 
+  mediatek,pcie-mirror:
+    $ref: /schemas/types.yaml#/definitions/phandle
+    description:
+      Phandle to the mediatek pcie-mirror controller.
+
+  mediatek,pctl:
+    $ref: /schemas/types.yaml#/definitions/phandle
+    description:
+      Phandle to the syscon node that handles the ports slew rate and
+      driver current.
+
   mediatek,sgmiisys:
     $ref: /schemas/types.yaml#/definitions/phandle-array
     minItems: 1
@@ -131,15 +142,12 @@ allOf:
 
         mediatek,infracfg: false
 
-        mediatek,pctl:
-          $ref: /schemas/types.yaml#/definitions/phandle
-          description:
-            Phandle to the syscon node that handles the ports slew rate and
-            driver current.
-
         mediatek,wed: false
 
         mediatek,wed-pcie: false
+    else:
+      properties:
+        mediatek,pctl: false
 
   - if:
       properties:
@@ -201,12 +209,10 @@ allOf:
           minItems: 1
           maxItems: 1
 
-        mediatek,pcie-mirror:
-          $ref: /schemas/types.yaml#/definitions/phandle
-          description:
-            Phandle to the mediatek pcie-mirror controller.
-
         mediatek,wed-pcie: false
+    else:
+      properties:
+        mediatek,pcie-mirror: false
 
   - if:
       properties:
diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
index 0ab124324eec..3eb65e63fdae 100644
--- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml
+++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml
@@ -512,6 +512,12 @@ properties:
     description:
       Frequency division factor for MDC clock.
 
+  snps,tso:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description:
+      Enables the TSO feature otherwise it will be managed by MAC HW capability
+      register.
+
   mdio:
     $ref: mdio.yaml#
     unevaluatedProperties: false
@@ -595,41 +601,38 @@ allOf:
   - if:
       properties:
         compatible:
-          contains:
-            enum:
-              - allwinner,sun7i-a20-gmac
-              - allwinner,sun8i-a83t-emac
-              - allwinner,sun8i-h3-emac
-              - allwinner,sun8i-r40-gmac
-              - allwinner,sun8i-v3s-emac
-              - allwinner,sun50i-a64-emac
-              - loongson,ls2k-dwmac
-              - loongson,ls7a-dwmac
-              - ingenic,jz4775-mac
-              - ingenic,x1000-mac
-              - ingenic,x1600-mac
-              - ingenic,x1830-mac
-              - ingenic,x2000-mac
-              - qcom,qcs404-ethqos
-              - qcom,sa8775p-ethqos
-              - qcom,sc8280xp-ethqos
-              - qcom,sm8150-ethqos
-              - snps,dwmac-4.00
-              - snps,dwmac-4.10a
-              - snps,dwmac-4.20a
-              - snps,dwmac-5.10a
-              - snps,dwmac-5.20
-              - snps,dwxgmac
-              - snps,dwxgmac-2.10
-              - st,spear600-gmac
+          not:
+            contains:
+              enum:
+                - allwinner,sun7i-a20-gmac
+                - allwinner,sun8i-a83t-emac
+                - allwinner,sun8i-h3-emac
+                - allwinner,sun8i-r40-gmac
+                - allwinner,sun8i-v3s-emac
+                - allwinner,sun50i-a64-emac
+                - loongson,ls2k-dwmac
+                - loongson,ls7a-dwmac
+                - ingenic,jz4775-mac
+                - ingenic,x1000-mac
+                - ingenic,x1600-mac
+                - ingenic,x1830-mac
+                - ingenic,x2000-mac
+                - qcom,qcs404-ethqos
+                - qcom,sa8775p-ethqos
+                - qcom,sc8280xp-ethqos
+                - qcom,sm8150-ethqos
+                - snps,dwmac-4.00
+                - snps,dwmac-4.10a
+                - snps,dwmac-4.20a
+                - snps,dwmac-5.10a
+                - snps,dwmac-5.20
+                - snps,dwxgmac
+                - snps,dwxgmac-2.10
+                - st,spear600-gmac
 
     then:
       properties:
-        snps,tso:
-          $ref: /schemas/types.yaml#/definitions/flag
-          description:
-            Enables the TSO feature otherwise it will be managed by
-            MAC HW capability register.
+        snps,tso: false
 
 additionalProperties: true
 
-- 
cgit v1.2.3-58-ga151


From f37bee9508885ce5edd4875c744e1af28cfac76c Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Mon, 1 Jul 2024 21:28:32 +0300
Subject: net: pcs: xpcs: Move native device ID macro to linux/pcs/pcs-xpcs.h

One of the next commits will alter the DW XPCS driver to support setting a
custom device ID for the particular MDIO-device detected on the platform.
The generic DW XPCS ID can be used as a custom ID as well in case if the
DW XPCS-device was erroneously synthesized with no or some undefined ID.
In addition to that having all supported DW XPCS device IDs defined in a
single place will improve the code maintainability and readability.

Note while at it rename the macros to being shorter and looking alike to
the already defined NXP XPCS ID macro.

Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/pcs/pcs-xpcs.c   | 8 ++++----
 drivers/net/pcs/pcs-xpcs.h   | 3 ---
 include/linux/pcs/pcs-xpcs.h | 2 ++
 3 files changed, 6 insertions(+), 7 deletions(-)

diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c
index 31525fe9c32e..99adbf15ab36 100644
--- a/drivers/net/pcs/pcs-xpcs.c
+++ b/drivers/net/pcs/pcs-xpcs.c
@@ -1343,16 +1343,16 @@ static const struct xpcs_compat nxp_sja1110_xpcs_compat[DW_XPCS_INTERFACE_MAX] =
 
 static const struct xpcs_id xpcs_id_list[] = {
 	{
-		.id = SYNOPSYS_XPCS_ID,
-		.mask = SYNOPSYS_XPCS_MASK,
+		.id = DW_XPCS_ID,
+		.mask = DW_XPCS_ID_MASK,
 		.compat = synopsys_xpcs_compat,
 	}, {
 		.id = NXP_SJA1105_XPCS_ID,
-		.mask = SYNOPSYS_XPCS_MASK,
+		.mask = DW_XPCS_ID_MASK,
 		.compat = nxp_sja1105_xpcs_compat,
 	}, {
 		.id = NXP_SJA1110_XPCS_ID,
-		.mask = SYNOPSYS_XPCS_MASK,
+		.mask = DW_XPCS_ID_MASK,
 		.compat = nxp_sja1110_xpcs_compat,
 	},
 };
diff --git a/drivers/net/pcs/pcs-xpcs.h b/drivers/net/pcs/pcs-xpcs.h
index 96c36b32ca99..369e9196f45a 100644
--- a/drivers/net/pcs/pcs-xpcs.h
+++ b/drivers/net/pcs/pcs-xpcs.h
@@ -6,9 +6,6 @@
  * Author: Jose Abreu <Jose.Abreu@synopsys.com>
  */
 
-#define SYNOPSYS_XPCS_ID		0x7996ced0
-#define SYNOPSYS_XPCS_MASK		0xffffffff
-
 /* Vendor regs access */
 #define DW_VENDOR			BIT(15)
 
diff --git a/include/linux/pcs/pcs-xpcs.h b/include/linux/pcs/pcs-xpcs.h
index da3a6c30f6d2..8dfe90295f12 100644
--- a/include/linux/pcs/pcs-xpcs.h
+++ b/include/linux/pcs/pcs-xpcs.h
@@ -12,6 +12,8 @@
 
 #define NXP_SJA1105_XPCS_ID		0x00000010
 #define NXP_SJA1110_XPCS_ID		0x00000020
+#define DW_XPCS_ID			0x7996ced0
+#define DW_XPCS_ID_MASK			0xffffffff
 
 /* AN mode */
 #define DW_AN_C73			1
-- 
cgit v1.2.3-58-ga151


From 03b3be07c69a5fd71e4c01a35b511c13d2adb3c5 Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Mon, 1 Jul 2024 21:28:33 +0300
Subject: net: pcs: xpcs: Split up xpcs_create() body to sub-functions

As an initial preparation before adding the fwnode-based DW XPCS device
support let's split the xpcs_create() function code up to a set of the
small sub-functions. Thus the xpcs_create() implementation will get to
look simpler and turn to be more coherent. Further updates will just touch
the new sub-functions a bit: add platform-specific device info, add the
reference clock getting and enabling.

The xpcs_create() method will now contain the next static methods calls:

xpcs_create_data() - create the DW XPCS device descriptor, pre-initialize
it' fields and increase the mdio device refcount-er;

xpcs_init_id() - find XPCS ID instance and save it in the device
descriptor;

xpcs_init_iface() - find MAC/PCS interface descriptor and perform
basic initialization specific to it: soft-reset, disable polling.

The update doesn't imply any semantic change but merely makes the code
looking simpler and more ready for adding new features support.

Note the xpcs_destroy() has been moved to being defined below the
xpcs_create_mdiodev() function as the driver now implies having the
protagonist-then-antagonist functions definition order.

Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/pcs/pcs-xpcs.c | 102 ++++++++++++++++++++++++++++++---------------
 1 file changed, 69 insertions(+), 33 deletions(-)

diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c
index 99adbf15ab36..2dcfd0ff069a 100644
--- a/drivers/net/pcs/pcs-xpcs.c
+++ b/drivers/net/pcs/pcs-xpcs.c
@@ -1365,12 +1365,9 @@ static const struct phylink_pcs_ops xpcs_phylink_ops = {
 	.pcs_link_up = xpcs_link_up,
 };
 
-static struct dw_xpcs *xpcs_create(struct mdio_device *mdiodev,
-				   phy_interface_t interface)
+static struct dw_xpcs *xpcs_create_data(struct mdio_device *mdiodev)
 {
 	struct dw_xpcs *xpcs;
-	u32 xpcs_id;
-	int i, ret;
 
 	xpcs = kzalloc(sizeof(*xpcs), GFP_KERNEL);
 	if (!xpcs)
@@ -1378,59 +1375,89 @@ static struct dw_xpcs *xpcs_create(struct mdio_device *mdiodev,
 
 	mdio_device_get(mdiodev);
 	xpcs->mdiodev = mdiodev;
+	xpcs->pcs.ops = &xpcs_phylink_ops;
+	xpcs->pcs.neg_mode = true;
+	xpcs->pcs.poll = true;
+
+	return xpcs;
+}
+
+static void xpcs_free_data(struct dw_xpcs *xpcs)
+{
+	mdio_device_put(xpcs->mdiodev);
+	kfree(xpcs);
+}
+
+static int xpcs_init_id(struct dw_xpcs *xpcs)
+{
+	u32 xpcs_id;
+	int i, ret;
 
 	xpcs_id = xpcs_get_id(xpcs);
 
 	for (i = 0; i < ARRAY_SIZE(xpcs_id_list); i++) {
 		const struct xpcs_id *entry = &xpcs_id_list[i];
-		const struct xpcs_compat *compat;
 
 		if ((xpcs_id & entry->mask) != entry->id)
 			continue;
 
 		xpcs->id = entry;
 
-		compat = xpcs_find_compat(entry, interface);
-		if (!compat) {
-			ret = -ENODEV;
-			goto out;
-		}
+		break;
+	}
 
-		ret = xpcs_dev_flag(xpcs);
-		if (ret)
-			goto out;
+	if (!xpcs->id)
+		return -ENODEV;
 
-		xpcs->pcs.ops = &xpcs_phylink_ops;
-		xpcs->pcs.neg_mode = true;
+	ret = xpcs_dev_flag(xpcs);
+	if (ret < 0)
+		return ret;
 
-		if (xpcs->dev_flag != DW_DEV_TXGBE) {
-			xpcs->pcs.poll = true;
+	return 0;
+}
 
-			ret = xpcs_soft_reset(xpcs, compat);
-			if (ret)
-				goto out;
-		}
+static int xpcs_init_iface(struct dw_xpcs *xpcs, phy_interface_t interface)
+{
+	const struct xpcs_compat *compat;
 
-		return xpcs;
+	compat = xpcs_find_compat(xpcs->id, interface);
+	if (!compat)
+		return -EINVAL;
+
+	if (xpcs->dev_flag == DW_DEV_TXGBE) {
+		xpcs->pcs.poll = false;
+		return 0;
 	}
 
-	ret = -ENODEV;
+	return xpcs_soft_reset(xpcs, compat);
+}
+
+static struct dw_xpcs *xpcs_create(struct mdio_device *mdiodev,
+				   phy_interface_t interface)
+{
+	struct dw_xpcs *xpcs;
+	int ret;
+
+	xpcs = xpcs_create_data(mdiodev);
+	if (IS_ERR(xpcs))
+		return xpcs;
+
+	ret = xpcs_init_id(xpcs);
+	if (ret)
+		goto out;
+
+	ret = xpcs_init_iface(xpcs, interface);
+	if (ret)
+		goto out;
+
+	return xpcs;
 
 out:
-	mdio_device_put(mdiodev);
-	kfree(xpcs);
+	xpcs_free_data(xpcs);
 
 	return ERR_PTR(ret);
 }
 
-void xpcs_destroy(struct dw_xpcs *xpcs)
-{
-	if (xpcs)
-		mdio_device_put(xpcs->mdiodev);
-	kfree(xpcs);
-}
-EXPORT_SYMBOL_GPL(xpcs_destroy);
-
 struct dw_xpcs *xpcs_create_mdiodev(struct mii_bus *bus, int addr,
 				    phy_interface_t interface)
 {
@@ -1455,5 +1482,14 @@ struct dw_xpcs *xpcs_create_mdiodev(struct mii_bus *bus, int addr,
 }
 EXPORT_SYMBOL_GPL(xpcs_create_mdiodev);
 
+void xpcs_destroy(struct dw_xpcs *xpcs)
+{
+	if (!xpcs)
+		return;
+
+	xpcs_free_data(xpcs);
+}
+EXPORT_SYMBOL_GPL(xpcs_destroy);
+
 MODULE_DESCRIPTION("Synopsys DesignWare XPCS library");
 MODULE_LICENSE("GPL v2");
-- 
cgit v1.2.3-58-ga151


From 71b200b388ef2ff1b547114f17dc1fd088bfc2c6 Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Mon, 1 Jul 2024 21:28:34 +0300
Subject: net: pcs: xpcs: Convert xpcs_id to dw_xpcs_desc

A structure with the PCS/PMA MMD IDs data is being introduced in one of
the next commits. In order to prevent the names ambiguity let's convert
the xpcs_id structure name to dw_xpcs_desc. The later version is more
suitable since the structure content is indeed the device descriptor
containing the data and callbacks required for the driver to correctly set
the device up.

Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/pcs/pcs-xpcs.c   | 32 ++++++++++++++++----------------
 include/linux/pcs/pcs-xpcs.h |  4 ++--
 2 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c
index 2dcfd0ff069a..4ed29df8c963 100644
--- a/drivers/net/pcs/pcs-xpcs.c
+++ b/drivers/net/pcs/pcs-xpcs.c
@@ -151,19 +151,19 @@ struct xpcs_compat {
 	int (*pma_config)(struct dw_xpcs *xpcs);
 };
 
-struct xpcs_id {
+struct dw_xpcs_desc {
 	u32 id;
 	u32 mask;
 	const struct xpcs_compat *compat;
 };
 
-static const struct xpcs_compat *xpcs_find_compat(const struct xpcs_id *id,
-						  phy_interface_t interface)
+static const struct xpcs_compat *
+xpcs_find_compat(const struct dw_xpcs_desc *desc, phy_interface_t interface)
 {
 	int i, j;
 
 	for (i = 0; i < DW_XPCS_INTERFACE_MAX; i++) {
-		const struct xpcs_compat *compat = &id->compat[i];
+		const struct xpcs_compat *compat = &desc->compat[i];
 
 		for (j = 0; j < compat->num_interfaces; j++)
 			if (compat->interface[j] == interface)
@@ -177,7 +177,7 @@ int xpcs_get_an_mode(struct dw_xpcs *xpcs, phy_interface_t interface)
 {
 	const struct xpcs_compat *compat;
 
-	compat = xpcs_find_compat(xpcs->id, interface);
+	compat = xpcs_find_compat(xpcs->desc, interface);
 	if (!compat)
 		return -ENODEV;
 
@@ -612,7 +612,7 @@ static int xpcs_validate(struct phylink_pcs *pcs, unsigned long *supported,
 	int i;
 
 	xpcs = phylink_pcs_to_xpcs(pcs);
-	compat = xpcs_find_compat(xpcs->id, state->interface);
+	compat = xpcs_find_compat(xpcs->desc, state->interface);
 	if (!compat)
 		return -EINVAL;
 
@@ -633,7 +633,7 @@ void xpcs_get_interfaces(struct dw_xpcs *xpcs, unsigned long *interfaces)
 	int i, j;
 
 	for (i = 0; i < DW_XPCS_INTERFACE_MAX; i++) {
-		const struct xpcs_compat *compat = &xpcs->id->compat[i];
+		const struct xpcs_compat *compat = &xpcs->desc->compat[i];
 
 		for (j = 0; j < compat->num_interfaces; j++)
 			__set_bit(compat->interface[j], interfaces);
@@ -853,7 +853,7 @@ int xpcs_do_config(struct dw_xpcs *xpcs, phy_interface_t interface,
 	const struct xpcs_compat *compat;
 	int ret;
 
-	compat = xpcs_find_compat(xpcs->id, interface);
+	compat = xpcs_find_compat(xpcs->desc, interface);
 	if (!compat)
 		return -ENODEV;
 
@@ -1118,7 +1118,7 @@ static void xpcs_get_state(struct phylink_pcs *pcs,
 	const struct xpcs_compat *compat;
 	int ret;
 
-	compat = xpcs_find_compat(xpcs->id, state->interface);
+	compat = xpcs_find_compat(xpcs->desc, state->interface);
 	if (!compat)
 		return;
 
@@ -1341,7 +1341,7 @@ static const struct xpcs_compat nxp_sja1110_xpcs_compat[DW_XPCS_INTERFACE_MAX] =
 	},
 };
 
-static const struct xpcs_id xpcs_id_list[] = {
+static const struct dw_xpcs_desc xpcs_desc_list[] = {
 	{
 		.id = DW_XPCS_ID,
 		.mask = DW_XPCS_ID_MASK,
@@ -1395,18 +1395,18 @@ static int xpcs_init_id(struct dw_xpcs *xpcs)
 
 	xpcs_id = xpcs_get_id(xpcs);
 
-	for (i = 0; i < ARRAY_SIZE(xpcs_id_list); i++) {
-		const struct xpcs_id *entry = &xpcs_id_list[i];
+	for (i = 0; i < ARRAY_SIZE(xpcs_desc_list); i++) {
+		const struct dw_xpcs_desc *desc = &xpcs_desc_list[i];
 
-		if ((xpcs_id & entry->mask) != entry->id)
+		if ((xpcs_id & desc->mask) != desc->id)
 			continue;
 
-		xpcs->id = entry;
+		xpcs->desc = desc;
 
 		break;
 	}
 
-	if (!xpcs->id)
+	if (!xpcs->desc)
 		return -ENODEV;
 
 	ret = xpcs_dev_flag(xpcs);
@@ -1420,7 +1420,7 @@ static int xpcs_init_iface(struct dw_xpcs *xpcs, phy_interface_t interface)
 {
 	const struct xpcs_compat *compat;
 
-	compat = xpcs_find_compat(xpcs->id, interface);
+	compat = xpcs_find_compat(xpcs->desc, interface);
 	if (!compat)
 		return -EINVAL;
 
diff --git a/include/linux/pcs/pcs-xpcs.h b/include/linux/pcs/pcs-xpcs.h
index 8dfe90295f12..e706bd16b986 100644
--- a/include/linux/pcs/pcs-xpcs.h
+++ b/include/linux/pcs/pcs-xpcs.h
@@ -28,11 +28,11 @@
 /* dev_flag */
 #define DW_DEV_TXGBE			BIT(0)
 
-struct xpcs_id;
+struct dw_xpcs_desc;
 
 struct dw_xpcs {
+	const struct dw_xpcs_desc *desc;
 	struct mdio_device *mdiodev;
-	const struct xpcs_id *id;
 	struct phylink_pcs pcs;
 	phy_interface_t interface;
 	int dev_flag;
-- 
cgit v1.2.3-58-ga151


From 410232ab3c07f999674a7f254b33decdd5492c46 Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Mon, 1 Jul 2024 21:28:35 +0300
Subject: net: pcs: xpcs: Convert xpcs_compat to dw_xpcs_compat

The xpcs_compat structure has been left as the only dw-prefix-less
structure since the previous commit. Let's unify at least the structures
naming in the driver by adding the dw_-prefix to it.

Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/pcs/pcs-xpcs.c | 38 +++++++++++++++++++-------------------
 1 file changed, 19 insertions(+), 19 deletions(-)

diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c
index 4ed29df8c963..eefb5e1dbe21 100644
--- a/drivers/net/pcs/pcs-xpcs.c
+++ b/drivers/net/pcs/pcs-xpcs.c
@@ -143,7 +143,7 @@ enum {
 	DW_XPCS_INTERFACE_MAX,
 };
 
-struct xpcs_compat {
+struct dw_xpcs_compat {
 	const int *supported;
 	const phy_interface_t *interface;
 	int num_interfaces;
@@ -154,16 +154,16 @@ struct xpcs_compat {
 struct dw_xpcs_desc {
 	u32 id;
 	u32 mask;
-	const struct xpcs_compat *compat;
+	const struct dw_xpcs_compat *compat;
 };
 
-static const struct xpcs_compat *
+static const struct dw_xpcs_compat *
 xpcs_find_compat(const struct dw_xpcs_desc *desc, phy_interface_t interface)
 {
 	int i, j;
 
 	for (i = 0; i < DW_XPCS_INTERFACE_MAX; i++) {
-		const struct xpcs_compat *compat = &desc->compat[i];
+		const struct dw_xpcs_compat *compat = &desc->compat[i];
 
 		for (j = 0; j < compat->num_interfaces; j++)
 			if (compat->interface[j] == interface)
@@ -175,7 +175,7 @@ xpcs_find_compat(const struct dw_xpcs_desc *desc, phy_interface_t interface)
 
 int xpcs_get_an_mode(struct dw_xpcs *xpcs, phy_interface_t interface)
 {
-	const struct xpcs_compat *compat;
+	const struct dw_xpcs_compat *compat;
 
 	compat = xpcs_find_compat(xpcs->desc, interface);
 	if (!compat)
@@ -185,7 +185,7 @@ int xpcs_get_an_mode(struct dw_xpcs *xpcs, phy_interface_t interface)
 }
 EXPORT_SYMBOL_GPL(xpcs_get_an_mode);
 
-static bool __xpcs_linkmode_supported(const struct xpcs_compat *compat,
+static bool __xpcs_linkmode_supported(const struct dw_xpcs_compat *compat,
 				      enum ethtool_link_mode_bit_indices linkmode)
 {
 	int i;
@@ -277,7 +277,7 @@ static int xpcs_poll_reset(struct dw_xpcs *xpcs, int dev)
 }
 
 static int xpcs_soft_reset(struct dw_xpcs *xpcs,
-			   const struct xpcs_compat *compat)
+			   const struct dw_xpcs_compat *compat)
 {
 	int ret, dev;
 
@@ -418,7 +418,7 @@ out:
 }
 
 static int _xpcs_config_aneg_c73(struct dw_xpcs *xpcs,
-				 const struct xpcs_compat *compat)
+				 const struct dw_xpcs_compat *compat)
 {
 	int ret, adv;
 
@@ -463,7 +463,7 @@ static int _xpcs_config_aneg_c73(struct dw_xpcs *xpcs,
 }
 
 static int xpcs_config_aneg_c73(struct dw_xpcs *xpcs,
-				const struct xpcs_compat *compat)
+				const struct dw_xpcs_compat *compat)
 {
 	int ret;
 
@@ -482,7 +482,7 @@ static int xpcs_config_aneg_c73(struct dw_xpcs *xpcs,
 
 static int xpcs_aneg_done_c73(struct dw_xpcs *xpcs,
 			      struct phylink_link_state *state,
-			      const struct xpcs_compat *compat, u16 an_stat1)
+			      const struct dw_xpcs_compat *compat, u16 an_stat1)
 {
 	int ret;
 
@@ -607,7 +607,7 @@ static int xpcs_validate(struct phylink_pcs *pcs, unsigned long *supported,
 			 const struct phylink_link_state *state)
 {
 	__ETHTOOL_DECLARE_LINK_MODE_MASK(xpcs_supported) = { 0, };
-	const struct xpcs_compat *compat;
+	const struct dw_xpcs_compat *compat;
 	struct dw_xpcs *xpcs;
 	int i;
 
@@ -633,7 +633,7 @@ void xpcs_get_interfaces(struct dw_xpcs *xpcs, unsigned long *interfaces)
 	int i, j;
 
 	for (i = 0; i < DW_XPCS_INTERFACE_MAX; i++) {
-		const struct xpcs_compat *compat = &xpcs->desc->compat[i];
+		const struct dw_xpcs_compat *compat = &xpcs->desc->compat[i];
 
 		for (j = 0; j < compat->num_interfaces; j++)
 			__set_bit(compat->interface[j], interfaces);
@@ -850,7 +850,7 @@ static int xpcs_config_2500basex(struct dw_xpcs *xpcs)
 int xpcs_do_config(struct dw_xpcs *xpcs, phy_interface_t interface,
 		   const unsigned long *advertising, unsigned int neg_mode)
 {
-	const struct xpcs_compat *compat;
+	const struct dw_xpcs_compat *compat;
 	int ret;
 
 	compat = xpcs_find_compat(xpcs->desc, interface);
@@ -915,7 +915,7 @@ static int xpcs_config(struct phylink_pcs *pcs, unsigned int neg_mode,
 
 static int xpcs_get_state_c73(struct dw_xpcs *xpcs,
 			      struct phylink_link_state *state,
-			      const struct xpcs_compat *compat)
+			      const struct dw_xpcs_compat *compat)
 {
 	bool an_enabled;
 	int pcs_stat1;
@@ -1115,7 +1115,7 @@ static void xpcs_get_state(struct phylink_pcs *pcs,
 			   struct phylink_link_state *state)
 {
 	struct dw_xpcs *xpcs = phylink_pcs_to_xpcs(pcs);
-	const struct xpcs_compat *compat;
+	const struct dw_xpcs_compat *compat;
 	int ret;
 
 	compat = xpcs_find_compat(xpcs->desc, state->interface);
@@ -1269,7 +1269,7 @@ static u32 xpcs_get_id(struct dw_xpcs *xpcs)
 	return 0xffffffff;
 }
 
-static const struct xpcs_compat synopsys_xpcs_compat[DW_XPCS_INTERFACE_MAX] = {
+static const struct dw_xpcs_compat synopsys_xpcs_compat[DW_XPCS_INTERFACE_MAX] = {
 	[DW_XPCS_USXGMII] = {
 		.supported = xpcs_usxgmii_features,
 		.interface = xpcs_usxgmii_interfaces,
@@ -1314,7 +1314,7 @@ static const struct xpcs_compat synopsys_xpcs_compat[DW_XPCS_INTERFACE_MAX] = {
 	},
 };
 
-static const struct xpcs_compat nxp_sja1105_xpcs_compat[DW_XPCS_INTERFACE_MAX] = {
+static const struct dw_xpcs_compat nxp_sja1105_xpcs_compat[DW_XPCS_INTERFACE_MAX] = {
 	[DW_XPCS_SGMII] = {
 		.supported = xpcs_sgmii_features,
 		.interface = xpcs_sgmii_interfaces,
@@ -1324,7 +1324,7 @@ static const struct xpcs_compat nxp_sja1105_xpcs_compat[DW_XPCS_INTERFACE_MAX] =
 	},
 };
 
-static const struct xpcs_compat nxp_sja1110_xpcs_compat[DW_XPCS_INTERFACE_MAX] = {
+static const struct dw_xpcs_compat nxp_sja1110_xpcs_compat[DW_XPCS_INTERFACE_MAX] = {
 	[DW_XPCS_SGMII] = {
 		.supported = xpcs_sgmii_features,
 		.interface = xpcs_sgmii_interfaces,
@@ -1418,7 +1418,7 @@ static int xpcs_init_id(struct dw_xpcs *xpcs)
 
 static int xpcs_init_iface(struct dw_xpcs *xpcs, phy_interface_t interface)
 {
-	const struct xpcs_compat *compat;
+	const struct dw_xpcs_compat *compat;
 
 	compat = xpcs_find_compat(xpcs->desc, interface);
 	if (!compat)
-- 
cgit v1.2.3-58-ga151


From bcac735cf653ef8dc6d7c08ed311e0a77f2665f0 Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Mon, 1 Jul 2024 21:28:36 +0300
Subject: net: pcs: xpcs: Introduce DW XPCS info structure

The being introduced structure will preserve the PCS and PMA IDs retrieved
from the respective DW XPCS MMDs or potentially pre-defined by the client
drivers. (The later change will be introduced later in the framework of
the commit adding the memory-mapped DW XPCS devices support.)

The structure fields are filled in in the xpcs_get_id() function, which
used to be responsible for the PCS Device ID getting only. Besides of the
PCS ID the method now fetches the PMA/PMD IDs too from MMD 1, which used
to be done in xpcs_dev_flag(). The retrieved PMA ID will be from now
utilized for the PMA-specific tweaks like it was introduced for the
Wangxun TxGBE PCS in the commit f629acc6f210 ("net: pcs: xpcs: support to
switch mode for Wangxun NICs").

Note 1. The xpcs_get_id() error-handling semantics has been changed. From
now the error number will be returned from the function. There is no point
in the next IOs or saving 0xffs and then looping over the actual device
IDs if device couldn't be reached. -ENODEV will be returned if the very
first IO operation failed thus indicating that no device could be found.

Note 2. The PCS and PMA IDs macros have been converted to enum'es. The
enum'es will be populated later in another commit with the virtual IDs
identifying the DW XPCS devices which have some platform-specifics, but
have been synthesized with the default PCS/PMA ID.

Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/pcs/pcs-xpcs.c   | 104 +++++++++++++++++++++----------------------
 include/linux/pcs/pcs-xpcs.h |  28 +++++++-----
 2 files changed, 67 insertions(+), 65 deletions(-)

diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c
index eefb5e1dbe21..6397dac82d1f 100644
--- a/drivers/net/pcs/pcs-xpcs.c
+++ b/drivers/net/pcs/pcs-xpcs.c
@@ -237,29 +237,6 @@ int xpcs_write_vpcs(struct dw_xpcs *xpcs, int reg, u16 val)
 	return xpcs_write_vendor(xpcs, MDIO_MMD_PCS, reg, val);
 }
 
-static int xpcs_dev_flag(struct dw_xpcs *xpcs)
-{
-	int ret, oui;
-
-	ret = xpcs_read(xpcs, MDIO_MMD_PMAPMD, MDIO_DEVID1);
-	if (ret < 0)
-		return ret;
-
-	oui = ret;
-
-	ret = xpcs_read(xpcs, MDIO_MMD_PMAPMD, MDIO_DEVID2);
-	if (ret < 0)
-		return ret;
-
-	ret = (ret >> 10) & 0x3F;
-	oui |= ret << 16;
-
-	if (oui == DW_OUI_WX)
-		xpcs->dev_flag = DW_DEV_TXGBE;
-
-	return 0;
-}
-
 static int xpcs_poll_reset(struct dw_xpcs *xpcs, int dev)
 {
 	/* Poll until the reset bit clears (50ms per retry == 0.6 sec) */
@@ -684,7 +661,7 @@ static int xpcs_config_aneg_c37_sgmii(struct dw_xpcs *xpcs,
 {
 	int ret, mdio_ctrl, tx_conf;
 
-	if (xpcs->dev_flag == DW_DEV_TXGBE)
+	if (xpcs->info.pma == WX_TXGBE_XPCS_PMA_10G_ID)
 		xpcs_write_vpcs(xpcs, DW_VR_XS_PCS_DIG_CTRL1, DW_CL37_BP | DW_EN_VSMMD1);
 
 	/* For AN for C37 SGMII mode, the settings are :-
@@ -722,7 +699,7 @@ static int xpcs_config_aneg_c37_sgmii(struct dw_xpcs *xpcs,
 	ret |= (DW_VR_MII_PCS_MODE_C37_SGMII <<
 		DW_VR_MII_AN_CTRL_PCS_MODE_SHIFT &
 		DW_VR_MII_PCS_MODE_MASK);
-	if (xpcs->dev_flag == DW_DEV_TXGBE) {
+	if (xpcs->info.pma == WX_TXGBE_XPCS_PMA_10G_ID) {
 		ret |= DW_VR_MII_AN_CTRL_8BIT;
 		/* Hardware requires it to be PHY side SGMII */
 		tx_conf = DW_VR_MII_TX_CONFIG_PHY_SIDE_SGMII;
@@ -744,7 +721,7 @@ static int xpcs_config_aneg_c37_sgmii(struct dw_xpcs *xpcs,
 	else
 		ret &= ~DW_VR_MII_DIG_CTRL1_MAC_AUTO_SW;
 
-	if (xpcs->dev_flag == DW_DEV_TXGBE)
+	if (xpcs->info.pma == WX_TXGBE_XPCS_PMA_10G_ID)
 		ret |= DW_VR_MII_DIG_CTRL1_PHY_MODE_CTRL;
 
 	ret = xpcs_write(xpcs, MDIO_MMD_VEND2, DW_VR_MII_DIG_CTRL1, ret);
@@ -766,7 +743,7 @@ static int xpcs_config_aneg_c37_1000basex(struct dw_xpcs *xpcs,
 	int ret, mdio_ctrl, adv;
 	bool changed = 0;
 
-	if (xpcs->dev_flag == DW_DEV_TXGBE)
+	if (xpcs->info.pma == WX_TXGBE_XPCS_PMA_10G_ID)
 		xpcs_write_vpcs(xpcs, DW_VR_XS_PCS_DIG_CTRL1, DW_CL37_BP | DW_EN_VSMMD1);
 
 	/* According to Chap 7.12, to set 1000BASE-X C37 AN, AN must
@@ -857,7 +834,7 @@ int xpcs_do_config(struct dw_xpcs *xpcs, phy_interface_t interface,
 	if (!compat)
 		return -ENODEV;
 
-	if (xpcs->dev_flag == DW_DEV_TXGBE) {
+	if (xpcs->info.pma == WX_TXGBE_XPCS_PMA_10G_ID) {
 		ret = txgbe_xpcs_switch_mode(xpcs, interface);
 		if (ret)
 			return ret;
@@ -1229,44 +1206,66 @@ static void xpcs_an_restart(struct phylink_pcs *pcs)
 	}
 }
 
-static u32 xpcs_get_id(struct dw_xpcs *xpcs)
+static int xpcs_get_id(struct dw_xpcs *xpcs)
 {
 	int ret;
 	u32 id;
 
-	/* First, search C73 PCS using PCS MMD */
+	/* First, search C73 PCS using PCS MMD 3. Return ENODEV if communication
+	 * failed indicating that device couldn't be reached.
+	 */
 	ret = xpcs_read(xpcs, MDIO_MMD_PCS, MII_PHYSID1);
 	if (ret < 0)
-		return 0xffffffff;
+		return -ENODEV;
 
 	id = ret << 16;
 
 	ret = xpcs_read(xpcs, MDIO_MMD_PCS, MII_PHYSID2);
 	if (ret < 0)
-		return 0xffffffff;
+		return ret;
 
-	/* If Device IDs are not all zeros or all ones,
-	 * we found C73 AN-type device
+	id |= ret;
+
+	/* If Device IDs are not all zeros or ones, then 10GBase-X/R or C73
+	 * KR/KX4 PCS found. Otherwise fallback to detecting 1000Base-X or C37
+	 * PCS in MII MMD 31.
 	 */
-	if ((id | ret) && (id | ret) != 0xffffffff)
-		return id | ret;
+	if (!id || id == 0xffffffff) {
+		ret = xpcs_read(xpcs, MDIO_MMD_VEND2, MII_PHYSID1);
+		if (ret < 0)
+			return ret;
+
+		id = ret << 16;
+
+		ret = xpcs_read(xpcs, MDIO_MMD_VEND2, MII_PHYSID2);
+		if (ret < 0)
+			return ret;
 
-	/* Next, search C37 PCS using Vendor-Specific MII MMD */
-	ret = xpcs_read(xpcs, MDIO_MMD_VEND2, MII_PHYSID1);
+		id |= ret;
+	}
+
+	xpcs->info.pcs = id;
+
+	/* Find out PMA/PMD ID from MMD 1 device ID registers */
+	ret = xpcs_read(xpcs, MDIO_MMD_PMAPMD, MDIO_DEVID1);
 	if (ret < 0)
-		return 0xffffffff;
+		return ret;
 
-	id = ret << 16;
+	id = ret;
 
-	ret = xpcs_read(xpcs, MDIO_MMD_VEND2, MII_PHYSID2);
+	ret = xpcs_read(xpcs, MDIO_MMD_PMAPMD, MDIO_DEVID2);
 	if (ret < 0)
-		return 0xffffffff;
+		return ret;
+
+	/* Note the inverted dword order and masked out Model/Revision numbers
+	 * with respect to what is done with the PCS ID...
+	 */
+	ret = (ret >> 10) & 0x3F;
+	id |= ret << 16;
 
-	/* If Device IDs are not all zeros, we found C37 AN-type device */
-	if (id | ret)
-		return id | ret;
+	xpcs->info.pma = id;
 
-	return 0xffffffff;
+	return 0;
 }
 
 static const struct dw_xpcs_compat synopsys_xpcs_compat[DW_XPCS_INTERFACE_MAX] = {
@@ -1390,15 +1389,16 @@ static void xpcs_free_data(struct dw_xpcs *xpcs)
 
 static int xpcs_init_id(struct dw_xpcs *xpcs)
 {
-	u32 xpcs_id;
 	int i, ret;
 
-	xpcs_id = xpcs_get_id(xpcs);
+	ret = xpcs_get_id(xpcs);
+	if (ret < 0)
+		return ret;
 
 	for (i = 0; i < ARRAY_SIZE(xpcs_desc_list); i++) {
 		const struct dw_xpcs_desc *desc = &xpcs_desc_list[i];
 
-		if ((xpcs_id & desc->mask) != desc->id)
+		if ((xpcs->info.pcs & desc->mask) != desc->id)
 			continue;
 
 		xpcs->desc = desc;
@@ -1409,10 +1409,6 @@ static int xpcs_init_id(struct dw_xpcs *xpcs)
 	if (!xpcs->desc)
 		return -ENODEV;
 
-	ret = xpcs_dev_flag(xpcs);
-	if (ret < 0)
-		return ret;
-
 	return 0;
 }
 
@@ -1424,7 +1420,7 @@ static int xpcs_init_iface(struct dw_xpcs *xpcs, phy_interface_t interface)
 	if (!compat)
 		return -EINVAL;
 
-	if (xpcs->dev_flag == DW_DEV_TXGBE) {
+	if (xpcs->info.pma == WX_TXGBE_XPCS_PMA_10G_ID) {
 		xpcs->pcs.poll = false;
 		return 0;
 	}
diff --git a/include/linux/pcs/pcs-xpcs.h b/include/linux/pcs/pcs-xpcs.h
index e706bd16b986..1dc60f5e653f 100644
--- a/include/linux/pcs/pcs-xpcs.h
+++ b/include/linux/pcs/pcs-xpcs.h
@@ -9,11 +9,7 @@
 
 #include <linux/phy.h>
 #include <linux/phylink.h>
-
-#define NXP_SJA1105_XPCS_ID		0x00000010
-#define NXP_SJA1110_XPCS_ID		0x00000020
-#define DW_XPCS_ID			0x7996ced0
-#define DW_XPCS_ID_MASK			0xffffffff
+#include <linux/types.h>
 
 /* AN mode */
 #define DW_AN_C73			1
@@ -22,20 +18,30 @@
 #define DW_AN_C37_1000BASEX		4
 #define DW_10GBASER			5
 
-/* device vendor OUI */
-#define DW_OUI_WX			0x0018fc80
+struct dw_xpcs_desc;
 
-/* dev_flag */
-#define DW_DEV_TXGBE			BIT(0)
+enum dw_xpcs_pcs_id {
+	NXP_SJA1105_XPCS_ID = 0x00000010,
+	NXP_SJA1110_XPCS_ID = 0x00000020,
+	DW_XPCS_ID = 0x7996ced0,
+	DW_XPCS_ID_MASK = 0xffffffff,
+};
 
-struct dw_xpcs_desc;
+enum dw_xpcs_pma_id {
+	WX_TXGBE_XPCS_PMA_10G_ID = 0x0018fc80,
+};
+
+struct dw_xpcs_info {
+	u32 pcs;
+	u32 pma;
+};
 
 struct dw_xpcs {
+	struct dw_xpcs_info info;
 	const struct dw_xpcs_desc *desc;
 	struct mdio_device *mdiodev;
 	struct phylink_pcs pcs;
 	phy_interface_t interface;
-	int dev_flag;
 };
 
 int xpcs_get_an_mode(struct dw_xpcs *xpcs, phy_interface_t interface);
-- 
cgit v1.2.3-58-ga151


From 664690eb08f71290c2e67c5a1d234075077b4f5b Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Mon, 1 Jul 2024 21:28:37 +0300
Subject: dt-bindings: net: Add Synopsys DW xPCS bindings

Synopsys DesignWare XPCS IP-core is a Physical Coding Sublayer (PCS) layer
providing an interface between the Media Access Control (MAC) and Physical
Medium Attachment Sublayer (PMA) through a Media independent interface.
From software point of view it exposes IEEE std. Clause 45 CSR space and
can be accessible either by MDIO or MCI/APB3 bus interfaces. In the former
case the PCS device is supposed to be defined under the respective MDIO
bus DT-node. In the later case the DW xPCS will be just a normal IO
memory-mapped device.

Besides of that DW XPCS DT-nodes can have an interrupt signal and clock
source properties specified. The former one indicates the Clause 73/37
auto-negotiation events like: negotiation page received, AN is completed
or incompatible link partner. The clock DT-properties can describe up to
three clock sources: peripheral bus clock source, internal reference clock
and the externally connected reference clock.

Finally the DW XPCS IP-core can be optionally synthesized with a
vendor-specific interface connected to the Synopsys PMA (also called
DesignWare Consumer/Enterprise PHY). Alas that isn't auto-detectable in a
portable way. So if the DW XPCS device has the respective PMA attached
then it should be reflected in the DT-node compatible string so the driver
would be aware of the PMA-specific device capabilities (mainly connected
with CSRs available for the fine-tunings).

Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 .../devicetree/bindings/net/pcs/snps,dw-xpcs.yaml  | 136 +++++++++++++++++++++
 1 file changed, 136 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/pcs/snps,dw-xpcs.yaml

diff --git a/Documentation/devicetree/bindings/net/pcs/snps,dw-xpcs.yaml b/Documentation/devicetree/bindings/net/pcs/snps,dw-xpcs.yaml
new file mode 100644
index 000000000000..e77eec9ac9ee
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/pcs/snps,dw-xpcs.yaml
@@ -0,0 +1,136 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/pcs/snps,dw-xpcs.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Synopsys DesignWare Ethernet PCS
+
+maintainers:
+  - Serge Semin <fancer.lancer@gmail.com>
+
+description:
+  Synopsys DesignWare Ethernet Physical Coding Sublayer provides an interface
+  between Media Access Control and Physical Medium Attachment Sublayer through
+  the Media Independent Interface (XGMII, USXGMII, XLGMII, GMII, etc)
+  controlled by means of the IEEE std. Clause 45 registers set. The PCS can be
+  optionally synthesized with a vendor-specific interface connected to
+  Synopsys PMA (also called DesignWare Consumer/Enterprise PHY) although in
+  general it can be used to communicate with any compatible PHY.
+
+  The PCS CSRs can be accessible either over the Ethernet MDIO bus or directly
+  by means of the APB3/MCI interfaces. In the later case the XPCS can be mapped
+  right to the system IO memory space.
+
+properties:
+  compatible:
+    oneOf:
+      - description: Synopsys DesignWare XPCS with none or unknown PMA
+        const: snps,dw-xpcs
+      - description: Synopsys DesignWare XPCS with Consumer Gen1 3G PMA
+        const: snps,dw-xpcs-gen1-3g
+      - description: Synopsys DesignWare XPCS with Consumer Gen2 3G PMA
+        const: snps,dw-xpcs-gen2-3g
+      - description: Synopsys DesignWare XPCS with Consumer Gen2 6G PMA
+        const: snps,dw-xpcs-gen2-6g
+      - description: Synopsys DesignWare XPCS with Consumer Gen4 3G PMA
+        const: snps,dw-xpcs-gen4-3g
+      - description: Synopsys DesignWare XPCS with Consumer Gen4 6G PMA
+        const: snps,dw-xpcs-gen4-6g
+      - description: Synopsys DesignWare XPCS with Consumer Gen5 10G PMA
+        const: snps,dw-xpcs-gen5-10g
+      - description: Synopsys DesignWare XPCS with Consumer Gen5 12G PMA
+        const: snps,dw-xpcs-gen5-12g
+
+  reg:
+    items:
+      - description:
+          In case of the MDIO management interface this just a 5-bits ID
+          of the MDIO bus device. If DW XPCS CSRs space is accessed over the
+          MCI or APB3 management interfaces, then the space mapping can be
+          either 'direct' or 'indirect'. In the former case all Clause 45
+          registers are contiguously mapped within the address space
+          MMD '[20:16]', Reg '[15:0]'. In the later case the space is divided
+          to the multiple 256 register sets. There is a special viewport CSR
+          which is responsible for the set selection. The upper part of
+          the CSR address MMD+REG[20:8] is supposed to be written in there
+          so the corresponding subset would be mapped to the lowest 255 CSRs.
+
+  reg-names:
+    items:
+      - enum: [ direct, indirect ]
+
+  reg-io-width:
+    description:
+      The way the CSRs are mapped to the memory is platform depended. Since
+      each Clause 45 CSR is of 16-bits wide the access instructions must be
+      two bytes aligned at least.
+    default: 2
+    enum: [ 2, 4 ]
+
+  interrupts:
+    description:
+      System interface interrupt output (sbd_intr_o) indicating Clause 73/37
+      auto-negotiation events':' Page received, AN is completed or incompatible
+      link partner.
+    maxItems: 1
+
+  clocks:
+    description:
+      The MCI and APB3 interfaces are supposed to be equipped with a clock
+      source connected to the clk_csr_i line.
+
+      PCS/PMA layer can be clocked by an internal reference clock source
+      (phyN_core_refclk) or by an externally connected (phyN_pad_refclk) clock
+      generator. Both clocks can be supplied at a time.
+    minItems: 1
+    maxItems: 3
+
+  clock-names:
+    oneOf:
+      - minItems: 1
+        items: # MDIO
+          - enum: [core, pad]
+          - const: pad
+      - minItems: 1
+        items: # MCI or APB
+          - const: csr
+          - enum: [core, pad]
+          - const: pad
+
+required:
+  - compatible
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/irq.h>
+
+    ethernet-pcs@1f05d000 {
+      compatible = "snps,dw-xpcs";
+      reg = <0x1f05d000 0x1000>;
+      reg-names = "indirect";
+
+      reg-io-width = <4>;
+
+      interrupts = <79 IRQ_TYPE_LEVEL_HIGH>;
+
+      clocks = <&ccu_pclk>, <&ccu_core>, <&ccu_pad>;
+      clock-names = "csr", "core", "pad";
+    };
+  - |
+    mdio-bus {
+      #address-cells = <1>;
+      #size-cells = <0>;
+
+      ethernet-pcs@0 {
+        compatible = "snps,dw-xpcs";
+        reg = <0>;
+
+        clocks = <&ccu_core>, <&ccu_pad>;
+        clock-names = "core", "pad";
+      };
+    };
+...
-- 
cgit v1.2.3-58-ga151


From f6bb3e9d98c2e8d70587d5ddaf9426ef30d7865c Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Mon, 1 Jul 2024 21:28:38 +0300
Subject: net: pcs: xpcs: Add Synopsys DW xPCS platform device driver

Synopsys DesignWare XPCS IP-core can be synthesized with the device CSRs
being accessible over the MCI or APB3 interface instead of the MDIO bus
(see the CSR_INTERFACE HDL parameter). Thus all the PCS registers can be
just memory mapped and be a subject of the standard MMIO operations of
course taking into account the peculiarities of the Clause C45 CSRs
mapping. From that perspective the DW XPCS devices would look as just
normal platform devices for the kernel.

On the other hand in order to have the DW XPCS devices handled by the
pcs-xpcs.c driver they need to be registered in the framework of the
MDIO-subsystem. So the suggested change is about providing a DW XPCS
platform device driver registering a virtual MDIO-bus with a single
MDIO-device representing the DW XPCS device.

DW XPCS platform device is supposed to be described by the respective
compatible string "snps,dw-xpcs" (or with the PMA-specific compatible
string), CSRs memory space and optional peripheral bus and reference clock
sources. Depending on the INDIRECT_ACCESS IP-core synthesize parameter the
memory-mapped reg-space can be represented as either directly or
indirectly mapped Clause 45 space. In the former case the particular
address is determined based on the MMD device and the registers offset (5
+ 16 bits all together) within the device reg-space. In the later case
there is only 8 lower address bits are utilized for the registers mapping
(255 CSRs). The upper bits are supposed to be written into the respective
viewport CSR in order to select the respective MMD sub-page.

Note, only the peripheral bus clock source is requested in the platform
device probe procedure. The core and pad clocks handling has been
implemented in the framework of the xpcs_create() method intentionally
since the clocks-related setups are supposed to be performed later, during
the DW XPCS main configuration procedures. (For instance they will be
required for the DW Gen5 10G PMA configuration.)

Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/pcs/Kconfig         |   6 +-
 drivers/net/pcs/Makefile        |   3 +-
 drivers/net/pcs/pcs-xpcs-plat.c | 460 ++++++++++++++++++++++++++++++++++++++++
 drivers/net/pcs/pcs-xpcs.c      |  63 +++++-
 drivers/net/pcs/pcs-xpcs.h      |   6 +
 include/linux/pcs/pcs-xpcs.h    |  18 ++
 6 files changed, 547 insertions(+), 9 deletions(-)
 create mode 100644 drivers/net/pcs/pcs-xpcs-plat.c

diff --git a/drivers/net/pcs/Kconfig b/drivers/net/pcs/Kconfig
index 87cf308fc6d8..f6aa437473de 100644
--- a/drivers/net/pcs/Kconfig
+++ b/drivers/net/pcs/Kconfig
@@ -6,11 +6,11 @@
 menu "PCS device drivers"
 
 config PCS_XPCS
-	tristate
+	tristate "Synopsys DesignWare Ethernet XPCS"
 	select PHYLINK
 	help
-	  This module provides helper functions for Synopsys DesignWare XPCS
-	  controllers.
+	  This module provides a driver and helper functions for Synopsys
+	  DesignWare XPCS controllers.
 
 config PCS_LYNX
 	tristate
diff --git a/drivers/net/pcs/Makefile b/drivers/net/pcs/Makefile
index fb1694192ae6..4f7920618b90 100644
--- a/drivers/net/pcs/Makefile
+++ b/drivers/net/pcs/Makefile
@@ -1,7 +1,8 @@
 # SPDX-License-Identifier: GPL-2.0
 # Makefile for Linux PCS drivers
 
-pcs_xpcs-$(CONFIG_PCS_XPCS)	:= pcs-xpcs.o pcs-xpcs-nxp.o pcs-xpcs-wx.o
+pcs_xpcs-$(CONFIG_PCS_XPCS)	:= pcs-xpcs.o pcs-xpcs-plat.o \
+				   pcs-xpcs-nxp.o pcs-xpcs-wx.o
 
 obj-$(CONFIG_PCS_XPCS)		+= pcs_xpcs.o
 obj-$(CONFIG_PCS_LYNX)		+= pcs-lynx.o
diff --git a/drivers/net/pcs/pcs-xpcs-plat.c b/drivers/net/pcs/pcs-xpcs-plat.c
new file mode 100644
index 000000000000..629315f1e57c
--- /dev/null
+++ b/drivers/net/pcs/pcs-xpcs-plat.c
@@ -0,0 +1,460 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Synopsys DesignWare XPCS platform device driver
+ *
+ * Copyright (C) 2024 Serge Semin
+ */
+
+#include <linux/atomic.h>
+#include <linux/bitfield.h>
+#include <linux/clk.h>
+#include <linux/device.h>
+#include <linux/kernel.h>
+#include <linux/mdio.h>
+#include <linux/module.h>
+#include <linux/pcs/pcs-xpcs.h>
+#include <linux/phy.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/property.h>
+#include <linux/sizes.h>
+
+#include "pcs-xpcs.h"
+
+/* Page select register for the indirect MMIO CSRs access */
+#define DW_VR_CSR_VIEWPORT		0xff
+
+struct dw_xpcs_plat {
+	struct platform_device *pdev;
+	struct mii_bus *bus;
+	bool reg_indir;
+	int reg_width;
+	void __iomem *reg_base;
+	struct clk *cclk;
+};
+
+static ptrdiff_t xpcs_mmio_addr_format(int dev, int reg)
+{
+	return FIELD_PREP(0x1f0000, dev) | FIELD_PREP(0xffff, reg);
+}
+
+static u16 xpcs_mmio_addr_page(ptrdiff_t csr)
+{
+	return FIELD_GET(0x1fff00, csr);
+}
+
+static ptrdiff_t xpcs_mmio_addr_offset(ptrdiff_t csr)
+{
+	return FIELD_GET(0xff, csr);
+}
+
+static int xpcs_mmio_read_reg_indirect(struct dw_xpcs_plat *pxpcs,
+				       int dev, int reg)
+{
+	ptrdiff_t csr, ofs;
+	u16 page;
+	int ret;
+
+	csr = xpcs_mmio_addr_format(dev, reg);
+	page = xpcs_mmio_addr_page(csr);
+	ofs = xpcs_mmio_addr_offset(csr);
+
+	ret = pm_runtime_resume_and_get(&pxpcs->pdev->dev);
+	if (ret)
+		return ret;
+
+	switch (pxpcs->reg_width) {
+	case 4:
+		writel(page, pxpcs->reg_base + (DW_VR_CSR_VIEWPORT << 2));
+		ret = readl(pxpcs->reg_base + (ofs << 2));
+		break;
+	default:
+		writew(page, pxpcs->reg_base + (DW_VR_CSR_VIEWPORT << 1));
+		ret = readw(pxpcs->reg_base + (ofs << 1));
+		break;
+	}
+
+	pm_runtime_put(&pxpcs->pdev->dev);
+
+	return ret;
+}
+
+static int xpcs_mmio_write_reg_indirect(struct dw_xpcs_plat *pxpcs,
+					int dev, int reg, u16 val)
+{
+	ptrdiff_t csr, ofs;
+	u16 page;
+	int ret;
+
+	csr = xpcs_mmio_addr_format(dev, reg);
+	page = xpcs_mmio_addr_page(csr);
+	ofs = xpcs_mmio_addr_offset(csr);
+
+	ret = pm_runtime_resume_and_get(&pxpcs->pdev->dev);
+	if (ret)
+		return ret;
+
+	switch (pxpcs->reg_width) {
+	case 4:
+		writel(page, pxpcs->reg_base + (DW_VR_CSR_VIEWPORT << 2));
+		writel(val, pxpcs->reg_base + (ofs << 2));
+		break;
+	default:
+		writew(page, pxpcs->reg_base + (DW_VR_CSR_VIEWPORT << 1));
+		writew(val, pxpcs->reg_base + (ofs << 1));
+		break;
+	}
+
+	pm_runtime_put(&pxpcs->pdev->dev);
+
+	return 0;
+}
+
+static int xpcs_mmio_read_reg_direct(struct dw_xpcs_plat *pxpcs,
+				     int dev, int reg)
+{
+	ptrdiff_t csr;
+	int ret;
+
+	csr = xpcs_mmio_addr_format(dev, reg);
+
+	ret = pm_runtime_resume_and_get(&pxpcs->pdev->dev);
+	if (ret)
+		return ret;
+
+	switch (pxpcs->reg_width) {
+	case 4:
+		ret = readl(pxpcs->reg_base + (csr << 2));
+		break;
+	default:
+		ret = readw(pxpcs->reg_base + (csr << 1));
+		break;
+	}
+
+	pm_runtime_put(&pxpcs->pdev->dev);
+
+	return ret;
+}
+
+static int xpcs_mmio_write_reg_direct(struct dw_xpcs_plat *pxpcs,
+				      int dev, int reg, u16 val)
+{
+	ptrdiff_t csr;
+	int ret;
+
+	csr = xpcs_mmio_addr_format(dev, reg);
+
+	ret = pm_runtime_resume_and_get(&pxpcs->pdev->dev);
+	if (ret)
+		return ret;
+
+	switch (pxpcs->reg_width) {
+	case 4:
+		writel(val, pxpcs->reg_base + (csr << 2));
+		break;
+	default:
+		writew(val, pxpcs->reg_base + (csr << 1));
+		break;
+	}
+
+	pm_runtime_put(&pxpcs->pdev->dev);
+
+	return 0;
+}
+
+static int xpcs_mmio_read_c22(struct mii_bus *bus, int addr, int reg)
+{
+	struct dw_xpcs_plat *pxpcs = bus->priv;
+
+	if (addr != 0)
+		return -ENODEV;
+
+	if (pxpcs->reg_indir)
+		return xpcs_mmio_read_reg_indirect(pxpcs, MDIO_MMD_VEND2, reg);
+	else
+		return xpcs_mmio_read_reg_direct(pxpcs, MDIO_MMD_VEND2, reg);
+}
+
+static int xpcs_mmio_write_c22(struct mii_bus *bus, int addr, int reg, u16 val)
+{
+	struct dw_xpcs_plat *pxpcs = bus->priv;
+
+	if (addr != 0)
+		return -ENODEV;
+
+	if (pxpcs->reg_indir)
+		return xpcs_mmio_write_reg_indirect(pxpcs, MDIO_MMD_VEND2, reg, val);
+	else
+		return xpcs_mmio_write_reg_direct(pxpcs, MDIO_MMD_VEND2, reg, val);
+}
+
+static int xpcs_mmio_read_c45(struct mii_bus *bus, int addr, int dev, int reg)
+{
+	struct dw_xpcs_plat *pxpcs = bus->priv;
+
+	if (addr != 0)
+		return -ENODEV;
+
+	if (pxpcs->reg_indir)
+		return xpcs_mmio_read_reg_indirect(pxpcs, dev, reg);
+	else
+		return xpcs_mmio_read_reg_direct(pxpcs, dev, reg);
+}
+
+static int xpcs_mmio_write_c45(struct mii_bus *bus, int addr, int dev,
+			       int reg, u16 val)
+{
+	struct dw_xpcs_plat *pxpcs = bus->priv;
+
+	if (addr != 0)
+		return -ENODEV;
+
+	if (pxpcs->reg_indir)
+		return xpcs_mmio_write_reg_indirect(pxpcs, dev, reg, val);
+	else
+		return xpcs_mmio_write_reg_direct(pxpcs, dev, reg, val);
+}
+
+static struct dw_xpcs_plat *xpcs_plat_create_data(struct platform_device *pdev)
+{
+	struct dw_xpcs_plat *pxpcs;
+
+	pxpcs = devm_kzalloc(&pdev->dev, sizeof(*pxpcs), GFP_KERNEL);
+	if (!pxpcs)
+		return ERR_PTR(-ENOMEM);
+
+	pxpcs->pdev = pdev;
+
+	dev_set_drvdata(&pdev->dev, pxpcs);
+
+	return pxpcs;
+}
+
+static int xpcs_plat_init_res(struct dw_xpcs_plat *pxpcs)
+{
+	struct platform_device *pdev = pxpcs->pdev;
+	struct device *dev = &pdev->dev;
+	resource_size_t spc_size;
+	struct resource *res;
+
+	if (!device_property_read_u32(dev, "reg-io-width", &pxpcs->reg_width)) {
+		if (pxpcs->reg_width != 2 && pxpcs->reg_width != 4) {
+			dev_err(dev, "Invalid reg-space data width\n");
+			return -EINVAL;
+		}
+	} else {
+		pxpcs->reg_width = 2;
+	}
+
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "direct") ?:
+	      platform_get_resource_byname(pdev, IORESOURCE_MEM, "indirect");
+	if (!res) {
+		dev_err(dev, "No reg-space found\n");
+		return -EINVAL;
+	}
+
+	if (!strcmp(res->name, "indirect"))
+		pxpcs->reg_indir = true;
+
+	if (pxpcs->reg_indir)
+		spc_size = pxpcs->reg_width * SZ_256;
+	else
+		spc_size = pxpcs->reg_width * SZ_2M;
+
+	if (resource_size(res) < spc_size) {
+		dev_err(dev, "Invalid reg-space size\n");
+		return -EINVAL;
+	}
+
+	pxpcs->reg_base = devm_ioremap_resource(dev, res);
+	if (IS_ERR(pxpcs->reg_base)) {
+		dev_err(dev, "Failed to map reg-space\n");
+		return PTR_ERR(pxpcs->reg_base);
+	}
+
+	return 0;
+}
+
+static int xpcs_plat_init_clk(struct dw_xpcs_plat *pxpcs)
+{
+	struct device *dev = &pxpcs->pdev->dev;
+	int ret;
+
+	pxpcs->cclk = devm_clk_get(dev, "csr");
+	if (IS_ERR(pxpcs->cclk))
+		return dev_err_probe(dev, PTR_ERR(pxpcs->cclk),
+				     "Failed to get CSR clock\n");
+
+	pm_runtime_set_active(dev);
+	ret = devm_pm_runtime_enable(dev);
+	if (ret) {
+		dev_err(dev, "Failed to enable runtime-PM\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int xpcs_plat_init_bus(struct dw_xpcs_plat *pxpcs)
+{
+	struct device *dev = &pxpcs->pdev->dev;
+	static atomic_t id = ATOMIC_INIT(-1);
+	int ret;
+
+	pxpcs->bus = devm_mdiobus_alloc_size(dev, 0);
+	if (!pxpcs->bus)
+		return -ENOMEM;
+
+	pxpcs->bus->name = "DW XPCS MCI/APB3";
+	pxpcs->bus->read = xpcs_mmio_read_c22;
+	pxpcs->bus->write = xpcs_mmio_write_c22;
+	pxpcs->bus->read_c45 = xpcs_mmio_read_c45;
+	pxpcs->bus->write_c45 = xpcs_mmio_write_c45;
+	pxpcs->bus->phy_mask = ~0;
+	pxpcs->bus->parent = dev;
+	pxpcs->bus->priv = pxpcs;
+
+	snprintf(pxpcs->bus->id, MII_BUS_ID_SIZE,
+		 "dwxpcs-%x", atomic_inc_return(&id));
+
+	/* MDIO-bus here serves as just a back-end engine abstracting out
+	 * the MDIO and MCI/APB3 IO interfaces utilized for the DW XPCS CSRs
+	 * access.
+	 */
+	ret = devm_mdiobus_register(dev, pxpcs->bus);
+	if (ret) {
+		dev_err(dev, "Failed to create MDIO bus\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+/* Note there is no need in the next function antagonist because the MDIO-bus
+ * de-registration will effectively remove and destroy all the MDIO-devices
+ * registered on the bus.
+ */
+static int xpcs_plat_init_dev(struct dw_xpcs_plat *pxpcs)
+{
+	struct device *dev = &pxpcs->pdev->dev;
+	struct mdio_device *mdiodev;
+	int ret;
+
+	/* There is a single memory-mapped DW XPCS device */
+	mdiodev = mdio_device_create(pxpcs->bus, 0);
+	if (IS_ERR(mdiodev))
+		return PTR_ERR(mdiodev);
+
+	/* Associate the FW-node with the device structure so it can be looked
+	 * up later. Make sure DD-core is aware of the OF-node being re-used.
+	 */
+	device_set_node(&mdiodev->dev, fwnode_handle_get(dev_fwnode(dev)));
+	mdiodev->dev.of_node_reused = true;
+
+	/* Pass the data further so the DW XPCS driver core could use it */
+	mdiodev->dev.platform_data = (void *)device_get_match_data(dev);
+
+	ret = mdio_device_register(mdiodev);
+	if (ret) {
+		dev_err(dev, "Failed to register MDIO device\n");
+		goto err_clean_data;
+	}
+
+	return 0;
+
+err_clean_data:
+	mdiodev->dev.platform_data = NULL;
+
+	fwnode_handle_put(dev_fwnode(&mdiodev->dev));
+	device_set_node(&mdiodev->dev, NULL);
+
+	mdio_device_free(mdiodev);
+
+	return ret;
+}
+
+static int xpcs_plat_probe(struct platform_device *pdev)
+{
+	struct dw_xpcs_plat *pxpcs;
+	int ret;
+
+	pxpcs = xpcs_plat_create_data(pdev);
+	if (IS_ERR(pxpcs))
+		return PTR_ERR(pxpcs);
+
+	ret = xpcs_plat_init_res(pxpcs);
+	if (ret)
+		return ret;
+
+	ret = xpcs_plat_init_clk(pxpcs);
+	if (ret)
+		return ret;
+
+	ret = xpcs_plat_init_bus(pxpcs);
+	if (ret)
+		return ret;
+
+	ret = xpcs_plat_init_dev(pxpcs);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int __maybe_unused xpcs_plat_pm_runtime_suspend(struct device *dev)
+{
+	struct dw_xpcs_plat *pxpcs = dev_get_drvdata(dev);
+
+	clk_disable_unprepare(pxpcs->cclk);
+
+	return 0;
+}
+
+static int __maybe_unused xpcs_plat_pm_runtime_resume(struct device *dev)
+{
+	struct dw_xpcs_plat *pxpcs = dev_get_drvdata(dev);
+
+	return clk_prepare_enable(pxpcs->cclk);
+}
+
+static const struct dev_pm_ops xpcs_plat_pm_ops = {
+	SET_RUNTIME_PM_OPS(xpcs_plat_pm_runtime_suspend,
+			   xpcs_plat_pm_runtime_resume,
+			   NULL)
+};
+
+DW_XPCS_INFO_DECLARE(xpcs_generic, DW_XPCS_ID_NATIVE, DW_XPCS_PMA_ID_NATIVE);
+DW_XPCS_INFO_DECLARE(xpcs_pma_gen1_3g, DW_XPCS_ID_NATIVE, DW_XPCS_PMA_GEN1_3G_ID);
+DW_XPCS_INFO_DECLARE(xpcs_pma_gen2_3g, DW_XPCS_ID_NATIVE, DW_XPCS_PMA_GEN2_3G_ID);
+DW_XPCS_INFO_DECLARE(xpcs_pma_gen2_6g, DW_XPCS_ID_NATIVE, DW_XPCS_PMA_GEN2_6G_ID);
+DW_XPCS_INFO_DECLARE(xpcs_pma_gen4_3g, DW_XPCS_ID_NATIVE, DW_XPCS_PMA_GEN4_3G_ID);
+DW_XPCS_INFO_DECLARE(xpcs_pma_gen4_6g, DW_XPCS_ID_NATIVE, DW_XPCS_PMA_GEN4_6G_ID);
+DW_XPCS_INFO_DECLARE(xpcs_pma_gen5_10g, DW_XPCS_ID_NATIVE, DW_XPCS_PMA_GEN5_10G_ID);
+DW_XPCS_INFO_DECLARE(xpcs_pma_gen5_12g, DW_XPCS_ID_NATIVE, DW_XPCS_PMA_GEN5_12G_ID);
+
+static const struct of_device_id xpcs_of_ids[] = {
+	{ .compatible = "snps,dw-xpcs", .data = &xpcs_generic },
+	{ .compatible = "snps,dw-xpcs-gen1-3g", .data = &xpcs_pma_gen1_3g },
+	{ .compatible = "snps,dw-xpcs-gen2-3g", .data = &xpcs_pma_gen2_3g },
+	{ .compatible = "snps,dw-xpcs-gen2-6g", .data = &xpcs_pma_gen2_6g },
+	{ .compatible = "snps,dw-xpcs-gen4-3g", .data = &xpcs_pma_gen4_3g },
+	{ .compatible = "snps,dw-xpcs-gen4-6g", .data = &xpcs_pma_gen4_6g },
+	{ .compatible = "snps,dw-xpcs-gen5-10g", .data = &xpcs_pma_gen5_10g },
+	{ .compatible = "snps,dw-xpcs-gen5-12g", .data = &xpcs_pma_gen5_12g },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, xpcs_of_ids);
+
+static struct platform_driver xpcs_plat_driver = {
+	.probe = xpcs_plat_probe,
+	.driver = {
+		.name = "dwxpcs",
+		.pm = &xpcs_plat_pm_ops,
+		.of_match_table = xpcs_of_ids,
+	},
+};
+module_platform_driver(xpcs_plat_driver);
+
+MODULE_DESCRIPTION("Synopsys DesignWare XPCS platform device driver");
+MODULE_AUTHOR("Signed-off-by: Serge Semin <fancer.lancer@gmail.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c
index 6397dac82d1f..f4425a7c74d5 100644
--- a/drivers/net/pcs/pcs-xpcs.c
+++ b/drivers/net/pcs/pcs-xpcs.c
@@ -6,6 +6,7 @@
  * Author: Jose Abreu <Jose.Abreu@synopsys.com>
  */
 
+#include <linux/clk.h>
 #include <linux/delay.h>
 #include <linux/pcs/pcs-xpcs.h>
 #include <linux/mdio.h>
@@ -1244,7 +1245,9 @@ static int xpcs_get_id(struct dw_xpcs *xpcs)
 		id |= ret;
 	}
 
-	xpcs->info.pcs = id;
+	/* Set the PCS ID if it hasn't been pre-initialized */
+	if (xpcs->info.pcs == DW_XPCS_ID_NATIVE)
+		xpcs->info.pcs = id;
 
 	/* Find out PMA/PMD ID from MMD 1 device ID registers */
 	ret = xpcs_read(xpcs, MDIO_MMD_PMAPMD, MDIO_DEVID1);
@@ -1263,7 +1266,9 @@ static int xpcs_get_id(struct dw_xpcs *xpcs)
 	ret = (ret >> 10) & 0x3F;
 	id |= ret << 16;
 
-	xpcs->info.pma = id;
+	/* Set the PMA ID if it hasn't been pre-initialized */
+	if (xpcs->info.pma == DW_XPCS_PMA_ID_NATIVE)
+		xpcs->info.pma = id;
 
 	return 0;
 }
@@ -1387,10 +1392,49 @@ static void xpcs_free_data(struct dw_xpcs *xpcs)
 	kfree(xpcs);
 }
 
+static int xpcs_init_clks(struct dw_xpcs *xpcs)
+{
+	static const char *ids[DW_XPCS_NUM_CLKS] = {
+		[DW_XPCS_CORE_CLK] = "core",
+		[DW_XPCS_PAD_CLK] = "pad",
+	};
+	struct device *dev = &xpcs->mdiodev->dev;
+	int ret, i;
+
+	for (i = 0; i < DW_XPCS_NUM_CLKS; ++i)
+		xpcs->clks[i].id = ids[i];
+
+	ret = clk_bulk_get_optional(dev, DW_XPCS_NUM_CLKS, xpcs->clks);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to get clocks\n");
+
+	ret = clk_bulk_prepare_enable(DW_XPCS_NUM_CLKS, xpcs->clks);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to enable clocks\n");
+
+	return 0;
+}
+
+static void xpcs_clear_clks(struct dw_xpcs *xpcs)
+{
+	clk_bulk_disable_unprepare(DW_XPCS_NUM_CLKS, xpcs->clks);
+
+	clk_bulk_put(DW_XPCS_NUM_CLKS, xpcs->clks);
+}
+
 static int xpcs_init_id(struct dw_xpcs *xpcs)
 {
+	const struct dw_xpcs_info *info;
 	int i, ret;
 
+	info = dev_get_platdata(&xpcs->mdiodev->dev);
+	if (!info) {
+		xpcs->info.pcs = DW_XPCS_ID_NATIVE;
+		xpcs->info.pma = DW_XPCS_PMA_ID_NATIVE;
+	} else {
+		xpcs->info = *info;
+	}
+
 	ret = xpcs_get_id(xpcs);
 	if (ret < 0)
 		return ret;
@@ -1438,17 +1482,24 @@ static struct dw_xpcs *xpcs_create(struct mdio_device *mdiodev,
 	if (IS_ERR(xpcs))
 		return xpcs;
 
+	ret = xpcs_init_clks(xpcs);
+	if (ret)
+		goto out_free_data;
+
 	ret = xpcs_init_id(xpcs);
 	if (ret)
-		goto out;
+		goto out_clear_clks;
 
 	ret = xpcs_init_iface(xpcs, interface);
 	if (ret)
-		goto out;
+		goto out_clear_clks;
 
 	return xpcs;
 
-out:
+out_clear_clks:
+	xpcs_clear_clks(xpcs);
+
+out_free_data:
 	xpcs_free_data(xpcs);
 
 	return ERR_PTR(ret);
@@ -1483,6 +1534,8 @@ void xpcs_destroy(struct dw_xpcs *xpcs)
 	if (!xpcs)
 		return;
 
+	xpcs_clear_clks(xpcs);
+
 	xpcs_free_data(xpcs);
 }
 EXPORT_SYMBOL_GPL(xpcs_destroy);
diff --git a/drivers/net/pcs/pcs-xpcs.h b/drivers/net/pcs/pcs-xpcs.h
index 369e9196f45a..fa05adfae220 100644
--- a/drivers/net/pcs/pcs-xpcs.h
+++ b/drivers/net/pcs/pcs-xpcs.h
@@ -6,6 +6,9 @@
  * Author: Jose Abreu <Jose.Abreu@synopsys.com>
  */
 
+#include <linux/bits.h>
+#include <linux/pcs/pcs-xpcs.h>
+
 /* Vendor regs access */
 #define DW_VENDOR			BIT(15)
 
@@ -117,6 +120,9 @@
 /* VR MII EEE Control 1 defines */
 #define DW_VR_MII_EEE_TRN_LPI		BIT(0)	/* Transparent Mode Enable */
 
+#define DW_XPCS_INFO_DECLARE(_name, _pcs, _pma)				\
+	static const struct dw_xpcs_info _name = { .pcs = _pcs, .pma = _pma }
+
 int xpcs_read(struct dw_xpcs *xpcs, int dev, u32 reg);
 int xpcs_write(struct dw_xpcs *xpcs, int dev, u32 reg, u16 val);
 int xpcs_read_vpcs(struct dw_xpcs *xpcs, int reg);
diff --git a/include/linux/pcs/pcs-xpcs.h b/include/linux/pcs/pcs-xpcs.h
index 1dc60f5e653f..813be644647f 100644
--- a/include/linux/pcs/pcs-xpcs.h
+++ b/include/linux/pcs/pcs-xpcs.h
@@ -7,6 +7,8 @@
 #ifndef __LINUX_PCS_XPCS_H
 #define __LINUX_PCS_XPCS_H
 
+#include <linux/clk.h>
+#include <linux/mdio.h>
 #include <linux/phy.h>
 #include <linux/phylink.h>
 #include <linux/types.h>
@@ -21,6 +23,7 @@
 struct dw_xpcs_desc;
 
 enum dw_xpcs_pcs_id {
+	DW_XPCS_ID_NATIVE = 0,
 	NXP_SJA1105_XPCS_ID = 0x00000010,
 	NXP_SJA1110_XPCS_ID = 0x00000020,
 	DW_XPCS_ID = 0x7996ced0,
@@ -28,6 +31,14 @@ enum dw_xpcs_pcs_id {
 };
 
 enum dw_xpcs_pma_id {
+	DW_XPCS_PMA_ID_NATIVE = 0,
+	DW_XPCS_PMA_GEN1_3G_ID,
+	DW_XPCS_PMA_GEN2_3G_ID,
+	DW_XPCS_PMA_GEN2_6G_ID,
+	DW_XPCS_PMA_GEN4_3G_ID,
+	DW_XPCS_PMA_GEN4_6G_ID,
+	DW_XPCS_PMA_GEN5_10G_ID,
+	DW_XPCS_PMA_GEN5_12G_ID,
 	WX_TXGBE_XPCS_PMA_10G_ID = 0x0018fc80,
 };
 
@@ -36,10 +47,17 @@ struct dw_xpcs_info {
 	u32 pma;
 };
 
+enum dw_xpcs_clock {
+	DW_XPCS_CORE_CLK,
+	DW_XPCS_PAD_CLK,
+	DW_XPCS_NUM_CLKS,
+};
+
 struct dw_xpcs {
 	struct dw_xpcs_info info;
 	const struct dw_xpcs_desc *desc;
 	struct mdio_device *mdiodev;
+	struct clk_bulk_data clks[DW_XPCS_NUM_CLKS];
 	struct phylink_pcs pcs;
 	phy_interface_t interface;
 };
-- 
cgit v1.2.3-58-ga151


From 9cad7275463ab9e52aeae8d2ecb169afee6876f3 Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Mon, 1 Jul 2024 21:28:39 +0300
Subject: net: pcs: xpcs: Add fwnode-based descriptor creation method

It's now possible to have the DW XPCS device defined as a standard
platform device for instance in the platform DT-file. Although that
functionality is useless unless there is a way to have the device found by
the client drivers (STMMAC/DW *MAC, NXP SJA1105 Eth Switch, etc). Provide
such ability by means of the xpcs_create_fwnode() method. It needs to be
called with the device DW XPCS fwnode instance passed. That node will be
then used to find the MDIO-device instance in order to create the DW XPCS
descriptor.

Note the method semantics and name is similar to what has been recently
introduced in the Lynx PCS driver.

Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/pcs/pcs-xpcs.c   | 50 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/pcs/pcs-xpcs.h |  3 +++
 2 files changed, 53 insertions(+)

diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c
index f4425a7c74d5..82463f9d50c8 100644
--- a/drivers/net/pcs/pcs-xpcs.c
+++ b/drivers/net/pcs/pcs-xpcs.c
@@ -10,7 +10,9 @@
 #include <linux/delay.h>
 #include <linux/pcs/pcs-xpcs.h>
 #include <linux/mdio.h>
+#include <linux/phy.h>
 #include <linux/phylink.h>
+#include <linux/property.h>
 
 #include "pcs-xpcs.h"
 
@@ -1505,6 +1507,16 @@ out_free_data:
 	return ERR_PTR(ret);
 }
 
+/**
+ * xpcs_create_mdiodev() - create a DW xPCS instance with the MDIO @addr
+ * @bus: pointer to the MDIO-bus descriptor for the device to be looked at
+ * @addr: device MDIO-bus ID
+ * @interface: requested PHY interface
+ *
+ * Return: a pointer to the DW XPCS handle if successful, otherwise -ENODEV if
+ * the PCS device couldn't be found on the bus and other negative errno related
+ * to the data allocation and MDIO-bus communications.
+ */
 struct dw_xpcs *xpcs_create_mdiodev(struct mii_bus *bus, int addr,
 				    phy_interface_t interface)
 {
@@ -1529,6 +1541,44 @@ struct dw_xpcs *xpcs_create_mdiodev(struct mii_bus *bus, int addr,
 }
 EXPORT_SYMBOL_GPL(xpcs_create_mdiodev);
 
+/**
+ * xpcs_create_fwnode() - Create a DW xPCS instance from @fwnode
+ * @fwnode: fwnode handle poining to the DW XPCS device
+ * @interface: requested PHY interface
+ *
+ * Return: a pointer to the DW XPCS handle if successful, otherwise -ENODEV if
+ * the fwnode device is unavailable or the PCS device couldn't be found on the
+ * bus, -EPROBE_DEFER if the respective MDIO-device instance couldn't be found,
+ * other negative errno related to the data allocations and MDIO-bus
+ * communications.
+ */
+struct dw_xpcs *xpcs_create_fwnode(struct fwnode_handle *fwnode,
+				   phy_interface_t interface)
+{
+	struct mdio_device *mdiodev;
+	struct dw_xpcs *xpcs;
+
+	if (!fwnode_device_is_available(fwnode))
+		return ERR_PTR(-ENODEV);
+
+	mdiodev = fwnode_mdio_find_device(fwnode);
+	if (!mdiodev)
+		return ERR_PTR(-EPROBE_DEFER);
+
+	xpcs = xpcs_create(mdiodev, interface);
+
+	/* xpcs_create() has taken a refcount on the mdiodev if it was
+	 * successful. If xpcs_create() fails, this will free the mdio
+	 * device here. In any case, we don't need to hold our reference
+	 * anymore, and putting it here will allow mdio_device_put() in
+	 * xpcs_destroy() to automatically free the mdio device.
+	 */
+	mdio_device_put(mdiodev);
+
+	return xpcs;
+}
+EXPORT_SYMBOL_GPL(xpcs_create_fwnode);
+
 void xpcs_destroy(struct dw_xpcs *xpcs)
 {
 	if (!xpcs)
diff --git a/include/linux/pcs/pcs-xpcs.h b/include/linux/pcs/pcs-xpcs.h
index 813be644647f..b4a4eb6c8866 100644
--- a/include/linux/pcs/pcs-xpcs.h
+++ b/include/linux/pcs/pcs-xpcs.h
@@ -8,6 +8,7 @@
 #define __LINUX_PCS_XPCS_H
 
 #include <linux/clk.h>
+#include <linux/fwnode.h>
 #include <linux/mdio.h>
 #include <linux/phy.h>
 #include <linux/phylink.h>
@@ -72,6 +73,8 @@ int xpcs_config_eee(struct dw_xpcs *xpcs, int mult_fact_100ns,
 		    int enable);
 struct dw_xpcs *xpcs_create_mdiodev(struct mii_bus *bus, int addr,
 				    phy_interface_t interface);
+struct dw_xpcs *xpcs_create_fwnode(struct fwnode_handle *fwnode,
+				   phy_interface_t interface);
 void xpcs_destroy(struct dw_xpcs *xpcs);
 
 #endif /* __LINUX_PCS_XPCS_H */
-- 
cgit v1.2.3-58-ga151


From 351066bad6ade5ad0f5f90fde2dc759759959969 Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Mon, 1 Jul 2024 21:28:40 +0300
Subject: net: stmmac: Create DW XPCS device with particular address

Currently the only STMMAC platform driver using the DW XPCS code is the
Intel mGBE device driver. (It can be determined by finding all the drivers
having the stmmac_mdio_bus_data::has_xpcs flag set.) At the same time the
low-level platform driver masks out the DW XPCS MDIO-address from being
auto-detected as PHY by the MDIO subsystem core. Seeing the PCS MDIO ID is
known the procedure of the DW XPCS device creation can be simplified by
dropping the loop over all the MDIO IDs. From now the DW XPCS device
descriptor will be created for the MDIO-bus address pre-defined by the
platform drivers via the stmmac_mdio_bus_data::pcs_mask field.

Note besides this shall speed up a bit the Intel mGBE probing.

Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c |  2 +-
 drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c | 19 ++++++-------------
 include/linux/stmmac.h                            |  2 +-
 3 files changed, 8 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
index 094d34c4193c..2dbfbca606af 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c
@@ -595,7 +595,7 @@ static int intel_mgbe_common_data(struct pci_dev *pdev,
 	/* Intel mgbe SGMII interface uses pcs-xcps */
 	if (plat->phy_interface == PHY_INTERFACE_MODE_SGMII ||
 	    plat->phy_interface == PHY_INTERFACE_MODE_1000BASEX) {
-		plat->mdio_bus_data->has_xpcs = true;
+		plat->mdio_bus_data->pcs_mask = BIT(INTEL_MGBE_XPCS_ADDR);
 		plat->mdio_bus_data->default_an_inband = true;
 		plat->select_pcs = intel_mgbe_select_pcs;
 	}
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
index aa43117134d3..74de6ec00bbf 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
@@ -499,8 +499,7 @@ int stmmac_pcs_setup(struct net_device *ndev)
 {
 	struct dw_xpcs *xpcs = NULL;
 	struct stmmac_priv *priv;
-	int ret = -ENODEV;
-	int mode, addr;
+	int addr, mode, ret;
 
 	priv = netdev_priv(ndev);
 	mode = priv->plat->phy_interface;
@@ -508,16 +507,10 @@ int stmmac_pcs_setup(struct net_device *ndev)
 	if (priv->plat->pcs_init) {
 		ret = priv->plat->pcs_init(priv);
 	} else if (priv->plat->mdio_bus_data &&
-		   priv->plat->mdio_bus_data->has_xpcs) {
-		/* Try to probe the XPCS by scanning all addresses */
-		for (addr = 0; addr < PHY_MAX_ADDR; addr++) {
-			xpcs = xpcs_create_mdiodev(priv->mii, addr, mode);
-			if (IS_ERR(xpcs))
-				continue;
-
-			ret = 0;
-			break;
-		}
+		   priv->plat->mdio_bus_data->pcs_mask) {
+		addr = ffs(priv->plat->mdio_bus_data->pcs_mask) - 1;
+		xpcs = xpcs_create_mdiodev(priv->mii, addr, mode);
+		ret = PTR_ERR_OR_ZERO(xpcs);
 	} else {
 		return 0;
 	}
@@ -610,7 +603,7 @@ int stmmac_mdio_register(struct net_device *ndev)
 	snprintf(new_bus->id, MII_BUS_ID_SIZE, "%s-%x",
 		 new_bus->name, priv->plat->bus_id);
 	new_bus->priv = ndev;
-	new_bus->phy_mask = mdio_bus_data->phy_mask;
+	new_bus->phy_mask = mdio_bus_data->phy_mask | mdio_bus_data->pcs_mask;
 	new_bus->parent = priv->device;
 
 	err = of_mdiobus_register(new_bus, mdio_node);
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index 9c54f82901a1..84e13bd5df28 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -82,7 +82,7 @@ struct stmmac_priv;
 
 struct stmmac_mdio_bus_data {
 	unsigned int phy_mask;
-	unsigned int has_xpcs;
+	unsigned int pcs_mask;
 	unsigned int default_an_inband;
 	int *irqs;
 	int probed_phy_irq;
-- 
cgit v1.2.3-58-ga151


From 357768c7e79242995f71ff51d2bc49623be8fb7d Mon Sep 17 00:00:00 2001
From: Serge Semin <fancer.lancer@gmail.com>
Date: Mon, 1 Jul 2024 21:28:41 +0300
Subject: net: stmmac: Add DW XPCS specified via "pcs-handle" support

Recently the DW XPCS DT-bindings have been introduced and the DW XPCS
driver has been altered to support the DW XPCS registered as a platform
device. In order to have the DW XPCS DT-device accessed from the STMMAC
driver let's alter the STMMAC PCS-setup procedure to support the
"pcs-handle" property containing the phandle reference to the DW XPCS
device DT-node. The respective fwnode will be then passed to the
xpcs_create_fwnode() function which in its turn will create the DW XPCS
descriptor utilized in the main driver for the PCS-related setups.

Signed-off-by: Serge Semin <fancer.lancer@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
index 74de6ec00bbf..03f90676b3ad 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
@@ -497,15 +497,22 @@ int stmmac_mdio_reset(struct mii_bus *bus)
 
 int stmmac_pcs_setup(struct net_device *ndev)
 {
+	struct fwnode_handle *devnode, *pcsnode;
 	struct dw_xpcs *xpcs = NULL;
 	struct stmmac_priv *priv;
 	int addr, mode, ret;
 
 	priv = netdev_priv(ndev);
 	mode = priv->plat->phy_interface;
+	devnode = priv->plat->port_node;
 
 	if (priv->plat->pcs_init) {
 		ret = priv->plat->pcs_init(priv);
+	} else if (fwnode_property_present(devnode, "pcs-handle")) {
+		pcsnode = fwnode_find_reference(devnode, "pcs-handle", 0);
+		xpcs = xpcs_create_fwnode(pcsnode, mode);
+		fwnode_handle_put(pcsnode);
+		ret = PTR_ERR_OR_ZERO(xpcs);
 	} else if (priv->plat->mdio_bus_data &&
 		   priv->plat->mdio_bus_data->pcs_mask) {
 		addr = ffs(priv->plat->mdio_bus_data->pcs_mask) - 1;
@@ -515,10 +522,8 @@ int stmmac_pcs_setup(struct net_device *ndev)
 		return 0;
 	}
 
-	if (ret) {
-		dev_warn(priv->device, "No xPCS found\n");
-		return ret;
-	}
+	if (ret)
+		return dev_err_probe(priv->device, ret, "No xPCS found\n");
 
 	priv->hw->xpcs = xpcs;
 
-- 
cgit v1.2.3-58-ga151


From c7f79f2620b7776586c626edf21eb6ed6ed3d1eb Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Wed, 3 Jul 2024 12:46:34 +0200
Subject: openvswitch: prepare for stolen verdict coming from conntrack and nat
 engine

At this time, conntrack either returns NF_ACCEPT or NF_DROP.
To improve debuging it would be nice to be able to replace NF_DROP
verdict with NF_DROP_REASON() helper,

This helper releases the skb instantly (so drop_monitor can pinpoint
precise location) and returns NF_STOLEN.

Prepare call sites to deal with this before introducing such changes
in conntrack and nat core.

Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Aaron Conole <aconole@redhat.om>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/openvswitch/conntrack.c | 47 +++++++++++++++++++++++++++++++++++----------
 1 file changed, 37 insertions(+), 10 deletions(-)

diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c
index 3b980bf2770b..8eb1d644b741 100644
--- a/net/openvswitch/conntrack.c
+++ b/net/openvswitch/conntrack.c
@@ -679,6 +679,8 @@ static int ovs_ct_nat(struct net *net, struct sw_flow_key *key,
 		action |= BIT(NF_NAT_MANIP_DST);
 
 	err = nf_ct_nat(skb, ct, ctinfo, &action, &info->range, info->commit);
+	if (err != NF_ACCEPT)
+		return err;
 
 	if (action & BIT(NF_NAT_MANIP_SRC))
 		ovs_nat_update_key(key, skb, NF_NAT_MANIP_SRC);
@@ -697,6 +699,22 @@ static int ovs_ct_nat(struct net *net, struct sw_flow_key *key,
 }
 #endif
 
+static int verdict_to_errno(unsigned int verdict)
+{
+	switch (verdict & NF_VERDICT_MASK) {
+	case NF_ACCEPT:
+		return 0;
+	case NF_DROP:
+		return -EINVAL;
+	case NF_STOLEN:
+		return -EINPROGRESS;
+	default:
+		break;
+	}
+
+	return -EINVAL;
+}
+
 /* Pass 'skb' through conntrack in 'net', using zone configured in 'info', if
  * not done already.  Update key with new CT state after passing the packet
  * through conntrack.
@@ -735,7 +753,7 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 
 		err = nf_conntrack_in(skb, &state);
 		if (err != NF_ACCEPT)
-			return -ENOENT;
+			return verdict_to_errno(err);
 
 		/* Clear CT state NAT flags to mark that we have not yet done
 		 * NAT after the nf_conntrack_in() call.  We can actually clear
@@ -762,9 +780,12 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 		 * the key->ct_state.
 		 */
 		if (info->nat && !(key->ct_state & OVS_CS_F_NAT_MASK) &&
-		    (nf_ct_is_confirmed(ct) || info->commit) &&
-		    ovs_ct_nat(net, key, info, skb, ct, ctinfo) != NF_ACCEPT) {
-			return -EINVAL;
+		    (nf_ct_is_confirmed(ct) || info->commit)) {
+			int err = ovs_ct_nat(net, key, info, skb, ct, ctinfo);
+
+			err = verdict_to_errno(err);
+			if (err)
+				return err;
 		}
 
 		/* Userspace may decide to perform a ct lookup without a helper
@@ -795,9 +816,12 @@ static int __ovs_ct_lookup(struct net *net, struct sw_flow_key *key,
 		 * - When committing an unconfirmed connection.
 		 */
 		if ((nf_ct_is_confirmed(ct) ? !cached || add_helper :
-					      info->commit) &&
-		    nf_ct_helper(skb, ct, ctinfo, info->family) != NF_ACCEPT) {
-			return -EINVAL;
+					      info->commit)) {
+			int err = nf_ct_helper(skb, ct, ctinfo, info->family);
+
+			err = verdict_to_errno(err);
+			if (err)
+				return err;
 		}
 
 		if (nf_ct_protonum(ct) == IPPROTO_TCP &&
@@ -1001,10 +1025,9 @@ static int ovs_ct_commit(struct net *net, struct sw_flow_key *key,
 	/* This will take care of sending queued events even if the connection
 	 * is already confirmed.
 	 */
-	if (nf_conntrack_confirm(skb) != NF_ACCEPT)
-		return -EINVAL;
+	err = nf_conntrack_confirm(skb);
 
-	return 0;
+	return verdict_to_errno(err);
 }
 
 /* Returns 0 on success, -EINPROGRESS if 'skb' is stolen, or other nonzero
@@ -1039,6 +1062,10 @@ int ovs_ct_execute(struct net *net, struct sk_buff *skb,
 	else
 		err = ovs_ct_lookup(net, key, info, skb);
 
+	/* conntrack core returned NF_STOLEN */
+	if (err == -EINPROGRESS)
+		return err;
+
 	skb_push_rcsum(skb, nh_ofs);
 	if (err)
 		ovs_kfree_skb_reason(skb, OVS_DROP_CONNTRACK);
-- 
cgit v1.2.3-58-ga151


From 3b2aef99221d395ce37efa426d7b50e7dcd621d6 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Mon, 1 Jul 2024 20:28:14 +0100
Subject: net: ethernet: mediatek: Allow gaps in MAC allocation

Some devices with MediaTek SoCs don't use the first but only the second
MAC in the chip. Especially with MT7981 which got a built-in 1GE PHY
connected to the second MAC this is quite common.
Make sure to reset and enable PSE also in those cases by skipping gaps
using 'continue' instead of aborting the loop using 'break'.

Fixes: dee4dd10c79a ("net: ethernet: mtk_eth_soc: ppe: add support for multiple PPEs")
Suggested-by: Elad Yifee <eladwf@gmail.com>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/379ae584cea112db60f4ada79c7e5ba4f3364a64.1719862038.git.daniel@makrotopia.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 13d78d9b3197..2529b5b607c8 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -3396,7 +3396,7 @@ static int mtk_open(struct net_device *dev)
 
 		for (i = 0; i < MTK_MAX_DEVS; i++) {
 			if (!eth->netdev[i])
-				break;
+				continue;
 
 			target_mac = netdev_priv(eth->netdev[i]);
 			if (!soc->offload_version) {
-- 
cgit v1.2.3-58-ga151


From fecef4cd42c689a200bdd39e6fffa71475904bc1 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 4 Jul 2024 16:48:15 +0200
Subject: tun: Assign missing bpf_net_context.

During the introduction of struct bpf_net_context handling for
XDP-redirect, the tun driver has been missed.
Jakub also pointed out that there is another call chain to
do_xdp_generic() originating from netif_receive_skb() and drivers may
use it outside from the NAPI context.

Set the bpf_net_context before invoking BPF XDP program within the TUN
driver. Set the bpf_net_context also in do_xdp_generic() if a xdp
program is available.

Reported-by: syzbot+0b5c75599f1d872bea6f@syzkaller.appspotmail.com
Reported-by: syzbot+5ae46b237278e2369cac@syzkaller.appspotmail.com
Reported-by: syzbot+c1e04a422bbc0f0f2921@syzkaller.appspotmail.com
Fixes: 401cb7dae8130 ("net: Reference bpf_redirect_info via task_struct on PREEMPT_RT.")
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20240704144815.j8xQda5r@linutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/tun.c | 7 +++++++
 net/core/dev.c    | 5 +++++
 2 files changed, 12 insertions(+)

diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 9254bca2813d..9b24861464bc 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -1661,6 +1661,7 @@ static struct sk_buff *tun_build_skb(struct tun_struct *tun,
 				     int len, int *skb_xdp)
 {
 	struct page_frag *alloc_frag = &current->task_frag;
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 	struct bpf_prog *xdp_prog;
 	int buflen = SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
 	char *buf;
@@ -1700,6 +1701,7 @@ static struct sk_buff *tun_build_skb(struct tun_struct *tun,
 
 	local_bh_disable();
 	rcu_read_lock();
+	bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
 	xdp_prog = rcu_dereference(tun->xdp_prog);
 	if (xdp_prog) {
 		struct xdp_buff xdp;
@@ -1728,12 +1730,14 @@ static struct sk_buff *tun_build_skb(struct tun_struct *tun,
 		pad = xdp.data - xdp.data_hard_start;
 		len = xdp.data_end - xdp.data;
 	}
+	bpf_net_ctx_clear(bpf_net_ctx);
 	rcu_read_unlock();
 	local_bh_enable();
 
 	return __tun_build_skb(tfile, alloc_frag, buf, buflen, len, pad);
 
 out:
+	bpf_net_ctx_clear(bpf_net_ctx);
 	rcu_read_unlock();
 	local_bh_enable();
 	return NULL;
@@ -2566,6 +2570,7 @@ static int tun_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
 
 	if (m->msg_controllen == sizeof(struct tun_msg_ctl) &&
 	    ctl && ctl->type == TUN_MSG_PTR) {
+		struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
 		struct tun_page tpage;
 		int n = ctl->num;
 		int flush = 0, queued = 0;
@@ -2574,6 +2579,7 @@ static int tun_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
 
 		local_bh_disable();
 		rcu_read_lock();
+		bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
 
 		for (i = 0; i < n; i++) {
 			xdp = &((struct xdp_buff *)ctl->ptr)[i];
@@ -2588,6 +2594,7 @@ static int tun_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len)
 		if (tfile->napi_enabled && queued > 0)
 			napi_schedule(&tfile->napi);
 
+		bpf_net_ctx_clear(bpf_net_ctx);
 		rcu_read_unlock();
 		local_bh_enable();
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 385c4091aa77..73e5af6943c3 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5126,11 +5126,14 @@ static DEFINE_STATIC_KEY_FALSE(generic_xdp_needed_key);
 
 int do_xdp_generic(struct bpf_prog *xdp_prog, struct sk_buff **pskb)
 {
+	struct bpf_net_context __bpf_net_ctx, *bpf_net_ctx;
+
 	if (xdp_prog) {
 		struct xdp_buff xdp;
 		u32 act;
 		int err;
 
+		bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
 		act = netif_receive_generic_xdp(pskb, &xdp, xdp_prog);
 		if (act != XDP_PASS) {
 			switch (act) {
@@ -5144,11 +5147,13 @@ int do_xdp_generic(struct bpf_prog *xdp_prog, struct sk_buff **pskb)
 				generic_xdp_tx(*pskb, xdp_prog);
 				break;
 			}
+			bpf_net_ctx_clear(bpf_net_ctx);
 			return XDP_DROP;
 		}
 	}
 	return XDP_PASS;
 out_redir:
+	bpf_net_ctx_clear(bpf_net_ctx);
 	kfree_skb_reason(*pskb, SKB_DROP_REASON_XDP);
 	return XDP_DROP;
 }
-- 
cgit v1.2.3-58-ga151


From 8dfbb068a4e14a3ff497661d211bfb4b2acc3563 Mon Sep 17 00:00:00 2001
From: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Date: Wed, 3 Jul 2024 23:46:33 +0200
Subject: net: dsa: qca8k: constify struct regmap_config

`qca8k_regmap_config` is not modified and can be declared as const to
move its data to a read-only section.

Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240703-net-const-regmap-v1-1-ff4aeceda02c@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/qca/qca8k-8xxx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/dsa/qca/qca8k-8xxx.c b/drivers/net/dsa/qca/qca8k-8xxx.c
index b3c27cf538e8..f8d8c70642c4 100644
--- a/drivers/net/dsa/qca/qca8k-8xxx.c
+++ b/drivers/net/dsa/qca/qca8k-8xxx.c
@@ -565,7 +565,7 @@ qca8k_regmap_update_bits(void *ctx, uint32_t reg, uint32_t mask, uint32_t write_
 	return qca8k_regmap_update_bits_mii(priv, reg, mask, write_val);
 }
 
-static struct regmap_config qca8k_regmap_config = {
+static const struct regmap_config qca8k_regmap_config = {
 	.reg_bits = 16,
 	.val_bits = 32,
 	.reg_stride = 4,
-- 
cgit v1.2.3-58-ga151


From 7f805f6396d496e9b7b7e92389c987178afeef7b Mon Sep 17 00:00:00 2001
From: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Date: Wed, 3 Jul 2024 23:46:34 +0200
Subject: net: ti: icss-iep: constify struct regmap_config

`am654_icss_iep_regmap_config` is only assigned to a pointer that passes
the data as read-only.

Add the const modifier to the struct and pointer to move the data to a
read-only section.

Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Reviewed-by: MD Danish Anwar <danishanwar@ti.com>
Link: https://patch.msgid.link/20240703-net-const-regmap-v1-2-ff4aeceda02c@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/icssg/icss_iep.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/ti/icssg/icss_iep.c b/drivers/net/ethernet/ti/icssg/icss_iep.c
index 003668dee738..75c294ce6fb6 100644
--- a/drivers/net/ethernet/ti/icssg/icss_iep.c
+++ b/drivers/net/ethernet/ti/icssg/icss_iep.c
@@ -95,7 +95,7 @@ enum {
  * @flags: Flags to represent IEP properties
  */
 struct icss_iep_plat_data {
-	struct regmap_config *config;
+	const struct regmap_config *config;
 	u32 reg_offs[ICSS_IEP_MAX_REGS];
 	u32 flags;
 };
@@ -952,7 +952,7 @@ static int icss_iep_regmap_read(void *context, unsigned int reg,
 	return 0;
 }
 
-static struct regmap_config am654_icss_iep_regmap_config = {
+static const struct regmap_config am654_icss_iep_regmap_config = {
 	.name = "icss iep",
 	.reg_stride = 1,
 	.reg_write = icss_iep_regmap_write,
-- 
cgit v1.2.3-58-ga151


From 9969163c447229af8d144913500ef14e57b68235 Mon Sep 17 00:00:00 2001
From: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Date: Wed, 3 Jul 2024 23:46:35 +0200
Subject: net: encx24j600: constify struct regmap_bus/regmap_config

`regmap_encx24j600`, `phycfg` and `phymap_encx24j600` are not modified
and can be declared as const to move their data to a read-only section.

Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Link: https://patch.msgid.link/20240703-net-const-regmap-v1-3-ff4aeceda02c@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/microchip/encx24j600-regmap.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/microchip/encx24j600-regmap.c b/drivers/net/ethernet/microchip/encx24j600-regmap.c
index 3885d6fbace1..26b00e66d912 100644
--- a/drivers/net/ethernet/microchip/encx24j600-regmap.c
+++ b/drivers/net/ethernet/microchip/encx24j600-regmap.c
@@ -474,13 +474,13 @@ static struct regmap_config regcfg = {
 	.unlock = regmap_unlock_mutex,
 };
 
-static struct regmap_bus regmap_encx24j600 = {
+static const struct regmap_bus regmap_encx24j600 = {
 	.write = regmap_encx24j600_write,
 	.read = regmap_encx24j600_read,
 	.reg_update_bits = regmap_encx24j600_reg_update_bits,
 };
 
-static struct regmap_config phycfg = {
+static const struct regmap_config phycfg = {
 	.name = "phy",
 	.reg_bits = 8,
 	.val_bits = 16,
@@ -492,7 +492,7 @@ static struct regmap_config phycfg = {
 	.volatile_reg = encx24j600_phymap_volatile,
 };
 
-static struct regmap_bus phymap_encx24j600 = {
+static const struct regmap_bus phymap_encx24j600 = {
 	.reg_write = regmap_encx24j600_phy_reg_write,
 	.reg_read = regmap_encx24j600_phy_reg_read,
 };
-- 
cgit v1.2.3-58-ga151


From 3b05c7995cae66c263547084ed2f7771f2d3b3a4 Mon Sep 17 00:00:00 2001
From: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Date: Wed, 3 Jul 2024 23:46:36 +0200
Subject: net: dsa: ar9331: constify struct regmap_bus

`ar9331_sw_bus` is not modified and can be declared as const to
move its data to a read-only section.

Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240703-net-const-regmap-v1-4-ff4aeceda02c@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/qca/ar9331.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/dsa/qca/ar9331.c b/drivers/net/dsa/qca/ar9331.c
index 968cb81088bf..e9f2c67bc15f 100644
--- a/drivers/net/dsa/qca/ar9331.c
+++ b/drivers/net/dsa/qca/ar9331.c
@@ -1021,7 +1021,7 @@ static const struct regmap_config ar9331_mdio_regmap_config = {
 	.cache_type = REGCACHE_MAPLE,
 };
 
-static struct regmap_bus ar9331_sw_bus = {
+static const struct regmap_bus ar9331_sw_bus = {
 	.reg_format_endian_default = REGMAP_ENDIAN_NATIVE,
 	.val_format_endian_default = REGMAP_ENDIAN_NATIVE,
 	.read = ar9331_mdio_read,
-- 
cgit v1.2.3-58-ga151


From ca18300e00d584d5693127eb60c108b84883b8ac Mon Sep 17 00:00:00 2001
From: Shengyu Qu <wiagn233@outlook.com>
Date: Fri, 5 Jul 2024 01:26:26 +0800
Subject: net: ethernet: mtk_ppe: Change PPE entries number to 16K

MT7981,7986 and 7988 all supports 32768 PPE entries, and MT7621/MT7620
supports 16384 PPE entries, but only set to 8192 entries in driver. So
incrase max entries to 16384 instead.

Signed-off-by: Elad Yifee <eladwf@gmail.com>
Signed-off-by: Shengyu Qu <wiagn233@outlook.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/TY3P286MB261103F937DE4EEB0F88437D98DE2@TY3P286MB2611.JPNP286.PROD.OUTLOOK.COM
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mediatek/mtk_ppe.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.h b/drivers/net/ethernet/mediatek/mtk_ppe.h
index 691806bca372..223f709e2704 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.h
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.h
@@ -8,7 +8,7 @@
 #include <linux/bitfield.h>
 #include <linux/rhashtable.h>
 
-#define MTK_PPE_ENTRIES_SHIFT		3
+#define MTK_PPE_ENTRIES_SHIFT		4
 #define MTK_PPE_ENTRIES			(1024 << MTK_PPE_ENTRIES_SHIFT)
 #define MTK_PPE_HASH_MASK		(MTK_PPE_ENTRIES - 1)
 #define MTK_PPE_WAIT_TIMEOUT_US		1000000
-- 
cgit v1.2.3-58-ga151


From 064fbc4e9b5a6dbda7fe7b67dc7e9e95d31f8d75 Mon Sep 17 00:00:00 2001
From: Daniel Golle <daniel@makrotopia.org>
Date: Thu, 4 Jul 2024 11:14:55 +0100
Subject: net: ethernet: mtk_eth_soc: implement .{get,set}_pauseparam ethtool
 ops

Implement operations to get and set flow-control link parameters.
Both is done by simply calling phylink_ethtool_{get,set}_pauseparam().
Fix whitespace in mtk_ethtool_ops while at it.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Tested-by: Rui Salvaterra <rsalvaterra@gmail.com>
Link: https://patch.msgid.link/e3ece47323444631d6cb479f32af0dfd6d145be0.1720088047.git.daniel@makrotopia.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 2529b5b607c8..0cc2dd85652f 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -4464,6 +4464,20 @@ static int mtk_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
 	return ret;
 }
 
+static void mtk_get_pauseparam(struct net_device *dev, struct ethtool_pauseparam *pause)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+
+	phylink_ethtool_get_pauseparam(mac->phylink, pause);
+}
+
+static int mtk_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam *pause)
+{
+	struct mtk_mac *mac = netdev_priv(dev);
+
+	return phylink_ethtool_set_pauseparam(mac->phylink, pause);
+}
+
 static u16 mtk_select_queue(struct net_device *dev, struct sk_buff *skb,
 			    struct net_device *sb_dev)
 {
@@ -4492,8 +4506,10 @@ static const struct ethtool_ops mtk_ethtool_ops = {
 	.get_strings		= mtk_get_strings,
 	.get_sset_count		= mtk_get_sset_count,
 	.get_ethtool_stats	= mtk_get_ethtool_stats,
+	.get_pauseparam		= mtk_get_pauseparam,
+	.set_pauseparam		= mtk_set_pauseparam,
 	.get_rxnfc		= mtk_get_rxnfc,
-	.set_rxnfc              = mtk_set_rxnfc,
+	.set_rxnfc		= mtk_set_rxnfc,
 };
 
 static const struct net_device_ops mtk_netdev_ops = {
-- 
cgit v1.2.3-58-ga151


From 093b0f366567aa3fed85c316f832607069202b23 Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Thu, 4 Jul 2024 10:56:52 +0200
Subject: net: psample: add user cookie

Add a user cookie to the sample metadata so that sample emitters can
provide more contextual information to samples.

If present, send the user cookie in a new attribute:
PSAMPLE_ATTR_USER_COOKIE.

Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Link: https://patch.msgid.link/20240704085710.353845-2-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/psample.h        | 2 ++
 include/uapi/linux/psample.h | 1 +
 net/psample/psample.c        | 9 ++++++++-
 3 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/include/net/psample.h b/include/net/psample.h
index 0509d2d6be67..2ac71260a546 100644
--- a/include/net/psample.h
+++ b/include/net/psample.h
@@ -25,6 +25,8 @@ struct psample_metadata {
 	   out_tc_occ_valid:1,
 	   latency_valid:1,
 	   unused:5;
+	const u8 *user_cookie;
+	u32 user_cookie_len;
 };
 
 struct psample_group *psample_group_get(struct net *net, u32 group_num);
diff --git a/include/uapi/linux/psample.h b/include/uapi/linux/psample.h
index e585db5bf2d2..e80637e1d97b 100644
--- a/include/uapi/linux/psample.h
+++ b/include/uapi/linux/psample.h
@@ -19,6 +19,7 @@ enum {
 	PSAMPLE_ATTR_LATENCY,		/* u64, nanoseconds */
 	PSAMPLE_ATTR_TIMESTAMP,		/* u64, nanoseconds */
 	PSAMPLE_ATTR_PROTO,		/* u16 */
+	PSAMPLE_ATTR_USER_COOKIE,	/* binary, user provided data */
 
 	__PSAMPLE_ATTR_MAX
 };
diff --git a/net/psample/psample.c b/net/psample/psample.c
index a5d9b8446f77..b37488f426bc 100644
--- a/net/psample/psample.c
+++ b/net/psample/psample.c
@@ -386,7 +386,9 @@ void psample_sample_packet(struct psample_group *group, struct sk_buff *skb,
 		   nla_total_size(sizeof(u32)) +	/* group_num */
 		   nla_total_size(sizeof(u32)) +	/* seq */
 		   nla_total_size_64bit(sizeof(u64)) +	/* timestamp */
-		   nla_total_size(sizeof(u16));		/* protocol */
+		   nla_total_size(sizeof(u16)) +	/* protocol */
+		   (md->user_cookie_len ?
+		    nla_total_size(md->user_cookie_len) : 0); /* user cookie */
 
 #ifdef CONFIG_INET
 	tun_info = skb_tunnel_info(skb);
@@ -486,6 +488,11 @@ void psample_sample_packet(struct psample_group *group, struct sk_buff *skb,
 	}
 #endif
 
+	if (md->user_cookie && md->user_cookie_len &&
+	    nla_put(nl_skb, PSAMPLE_ATTR_USER_COOKIE, md->user_cookie_len,
+		    md->user_cookie))
+		goto error;
+
 	genlmsg_end(nl_skb, data);
 	genlmsg_multicast_netns(&psample_nl_family, group->net, nl_skb, 0,
 				PSAMPLE_NL_MCGRP_SAMPLE, GFP_ATOMIC);
-- 
cgit v1.2.3-58-ga151


From 03448444ae5c6e1de1ae7ff9fa19dc320eeb2d49 Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Thu, 4 Jul 2024 10:56:53 +0200
Subject: net: sched: act_sample: add action cookie to sample

If the action has a user_cookie, pass it along to the sample so it can
be easily identified.

Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Reviewed-by: Aaron Conole <aconole@redhat.com>
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Link: https://patch.msgid.link/20240704085710.353845-3-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/act_sample.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/net/sched/act_sample.c b/net/sched/act_sample.c
index a69b53d54039..2ceb4d141b71 100644
--- a/net/sched/act_sample.c
+++ b/net/sched/act_sample.c
@@ -167,7 +167,9 @@ TC_INDIRECT_SCOPE int tcf_sample_act(struct sk_buff *skb,
 {
 	struct tcf_sample *s = to_sample(a);
 	struct psample_group *psample_group;
+	u8 cookie_data[TC_COOKIE_MAX_SIZE];
 	struct psample_metadata md = {};
+	struct tc_cookie *user_cookie;
 	int retval;
 
 	tcf_lastuse_update(&s->tcf_tm);
@@ -189,6 +191,16 @@ TC_INDIRECT_SCOPE int tcf_sample_act(struct sk_buff *skb,
 		if (skb_at_tc_ingress(skb) && tcf_sample_dev_ok_push(skb->dev))
 			skb_push(skb, skb->mac_len);
 
+		rcu_read_lock();
+		user_cookie = rcu_dereference(a->user_cookie);
+		if (user_cookie) {
+			memcpy(cookie_data, user_cookie->data,
+			       user_cookie->len);
+			md.user_cookie = cookie_data;
+			md.user_cookie_len = user_cookie->len;
+		}
+		rcu_read_unlock();
+
 		md.trunc_size = s->truncate ? s->trunc_size : skb->len;
 		psample_sample_packet(psample_group, skb, s->rate, &md);
 
-- 
cgit v1.2.3-58-ga151


From c35d86a23029f1186e3c7a65df7c38b762fb0434 Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Thu, 4 Jul 2024 10:56:54 +0200
Subject: net: psample: skip packet copy if no listeners

If nobody is listening on the multicast group, generating the sample,
which involves copying packet data, seems completely unnecessary.

Return fast in this case.

Reviewed-by: Aaron Conole <aconole@redhat.com>
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Link: https://patch.msgid.link/20240704085710.353845-4-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/psample/psample.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/net/psample/psample.c b/net/psample/psample.c
index b37488f426bc..1c76f3e48dcd 100644
--- a/net/psample/psample.c
+++ b/net/psample/psample.c
@@ -376,6 +376,10 @@ void psample_sample_packet(struct psample_group *group, struct sk_buff *skb,
 	void *data;
 	int ret;
 
+	if (!genl_has_listeners(&psample_nl_family, group->net,
+				PSAMPLE_NL_MCGRP_SAMPLE))
+		return;
+
 	meta_len = (in_ifindex ? nla_total_size(sizeof(u16)) : 0) +
 		   (out_ifindex ? nla_total_size(sizeof(u16)) : 0) +
 		   (md->out_tc_valid ? nla_total_size(sizeof(u16)) : 0) +
-- 
cgit v1.2.3-58-ga151


From 7b1b2b60c63f070e0dfbe072ccaae13168b38d01 Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Thu, 4 Jul 2024 10:56:55 +0200
Subject: net: psample: allow using rate as probability

Although not explicitly documented in the psample module itself, the
definition of PSAMPLE_ATTR_SAMPLE_RATE seems inherited from act_sample.

Quoting tc-sample(8):
"RATE of 100 will lead to an average of one sampled packet out of every
100 observed."

With this semantics, the rates that we can express with an unsigned
32-bits number are very unevenly distributed and concentrated towards
"sampling few packets".
For example, we can express a probability of 2.32E-8% but we
cannot express anything between 100% and 50%.

For sampling applications that are capable of sampling a decent
amount of packets, this sampling rate semantics is not very useful.

Add a new flag to the uAPI that indicates that the sampling rate is
expressed in scaled probability, this is:
- 0 is 0% probability, no packets get sampled.
- U32_MAX is 100% probability, all packets get sampled.

Reviewed-by: Aaron Conole <aconole@redhat.com>
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Link: https://patch.msgid.link/20240704085710.353845-5-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/psample.h        |  3 ++-
 include/uapi/linux/psample.h | 10 +++++++++-
 net/psample/psample.c        |  3 +++
 3 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/include/net/psample.h b/include/net/psample.h
index 2ac71260a546..c52e9ebd88dd 100644
--- a/include/net/psample.h
+++ b/include/net/psample.h
@@ -24,7 +24,8 @@ struct psample_metadata {
 	u8 out_tc_valid:1,
 	   out_tc_occ_valid:1,
 	   latency_valid:1,
-	   unused:5;
+	   rate_as_probability:1,
+	   unused:4;
 	const u8 *user_cookie;
 	u32 user_cookie_len;
 };
diff --git a/include/uapi/linux/psample.h b/include/uapi/linux/psample.h
index e80637e1d97b..b765f0e81f20 100644
--- a/include/uapi/linux/psample.h
+++ b/include/uapi/linux/psample.h
@@ -8,7 +8,11 @@ enum {
 	PSAMPLE_ATTR_ORIGSIZE,
 	PSAMPLE_ATTR_SAMPLE_GROUP,
 	PSAMPLE_ATTR_GROUP_SEQ,
-	PSAMPLE_ATTR_SAMPLE_RATE,
+	PSAMPLE_ATTR_SAMPLE_RATE,	/* u32, ratio between observed and
+					 * sampled packets or scaled probability
+					 * if PSAMPLE_ATTR_SAMPLE_PROBABILITY
+					 * is set.
+					 */
 	PSAMPLE_ATTR_DATA,
 	PSAMPLE_ATTR_GROUP_REFCOUNT,
 	PSAMPLE_ATTR_TUNNEL,
@@ -20,6 +24,10 @@ enum {
 	PSAMPLE_ATTR_TIMESTAMP,		/* u64, nanoseconds */
 	PSAMPLE_ATTR_PROTO,		/* u16 */
 	PSAMPLE_ATTR_USER_COOKIE,	/* binary, user provided data */
+	PSAMPLE_ATTR_SAMPLE_PROBABILITY,/* no argument, interpret rate in
+					 * PSAMPLE_ATTR_SAMPLE_RATE as a
+					 * probability scaled 0 - U32_MAX.
+					 */
 
 	__PSAMPLE_ATTR_MAX
 };
diff --git a/net/psample/psample.c b/net/psample/psample.c
index 1c76f3e48dcd..f48b5b9cd409 100644
--- a/net/psample/psample.c
+++ b/net/psample/psample.c
@@ -497,6 +497,9 @@ void psample_sample_packet(struct psample_group *group, struct sk_buff *skb,
 		    md->user_cookie))
 		goto error;
 
+	if (md->rate_as_probability)
+		nla_put_flag(skb, PSAMPLE_ATTR_SAMPLE_PROBABILITY);
+
 	genlmsg_end(nl_skb, data);
 	genlmsg_multicast_netns(&psample_nl_family, group->net, nl_skb, 0,
 				PSAMPLE_NL_MCGRP_SAMPLE, GFP_ATOMIC);
-- 
cgit v1.2.3-58-ga151


From aae0b82b46cb5004bdf82a000c004d69a0885c33 Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Thu, 4 Jul 2024 10:56:56 +0200
Subject: net: openvswitch: add psample action

Add support for a new action: psample.

This action accepts a u32 group id and a variable-length cookie and uses
the psample multicast group to make the packet available for
observability.

The maximum length of the user-defined cookie is set to 16, same as
tc_cookie, to discourage using cookies that will not be offloadable.

Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Reviewed-by: Aaron Conole <aconole@redhat.com>
Reviewed-by: Ilya Maximets <i.maximets@ovn.org>
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Link: https://patch.msgid.link/20240704085710.353845-6-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/ovs_flow.yaml | 17 +++++++++++
 include/uapi/linux/openvswitch.h          | 28 ++++++++++++++++++
 net/openvswitch/Kconfig                   |  1 +
 net/openvswitch/actions.c                 | 48 +++++++++++++++++++++++++++++++
 net/openvswitch/flow_netlink.c            | 32 ++++++++++++++++++++-
 5 files changed, 125 insertions(+), 1 deletion(-)

diff --git a/Documentation/netlink/specs/ovs_flow.yaml b/Documentation/netlink/specs/ovs_flow.yaml
index 4fdfc6b5cae9..46f5d1cd8a5f 100644
--- a/Documentation/netlink/specs/ovs_flow.yaml
+++ b/Documentation/netlink/specs/ovs_flow.yaml
@@ -727,6 +727,12 @@ attribute-sets:
         name: dec-ttl
         type: nest
         nested-attributes: dec-ttl-attrs
+      -
+        name: psample
+        type: nest
+        nested-attributes: psample-attrs
+        doc: |
+          Sends a packet sample to psample for external observation.
   -
     name: tunnel-key-attrs
     enum-name: ovs-tunnel-key-attr
@@ -938,6 +944,17 @@ attribute-sets:
       -
         name: gbp
         type: u32
+  -
+    name: psample-attrs
+    enum-name: ovs-psample-attr
+    name-prefix: ovs-psample-attr-
+    attributes:
+      -
+        name: group
+        type: u32
+      -
+        name: cookie
+        type: binary
 
 operations:
   name-prefix: ovs-flow-cmd-
diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h
index efc82c318fa2..3dd653748725 100644
--- a/include/uapi/linux/openvswitch.h
+++ b/include/uapi/linux/openvswitch.h
@@ -914,6 +914,31 @@ struct check_pkt_len_arg {
 };
 #endif
 
+#define OVS_PSAMPLE_COOKIE_MAX_SIZE 16
+/**
+ * enum ovs_psample_attr - Attributes for %OVS_ACTION_ATTR_PSAMPLE
+ * action.
+ *
+ * @OVS_PSAMPLE_ATTR_GROUP: 32-bit number to identify the source of the
+ * sample.
+ * @OVS_PSAMPLE_ATTR_COOKIE: An optional variable-length binary cookie that
+ * contains user-defined metadata. The maximum length is
+ * OVS_PSAMPLE_COOKIE_MAX_SIZE bytes.
+ *
+ * Sends the packet to the psample multicast group with the specified group and
+ * cookie. It is possible to combine this action with the
+ * %OVS_ACTION_ATTR_TRUNC action to limit the size of the sample.
+ */
+enum ovs_psample_attr {
+	OVS_PSAMPLE_ATTR_GROUP = 1,	/* u32 number. */
+	OVS_PSAMPLE_ATTR_COOKIE,	/* Optional, user specified cookie. */
+
+	/* private: */
+	__OVS_PSAMPLE_ATTR_MAX
+};
+
+#define OVS_PSAMPLE_ATTR_MAX (__OVS_PSAMPLE_ATTR_MAX - 1)
+
 /**
  * enum ovs_action_attr - Action types.
  *
@@ -966,6 +991,8 @@ struct check_pkt_len_arg {
  * of l3 tunnel flag in the tun_flags field of OVS_ACTION_ATTR_ADD_MPLS
  * argument.
  * @OVS_ACTION_ATTR_DROP: Explicit drop action.
+ * @OVS_ACTION_ATTR_PSAMPLE: Send a sample of the packet to external observers
+ * via psample.
  *
  * Only a single header can be set with a single %OVS_ACTION_ATTR_SET.  Not all
  * fields within a header are modifiable, e.g. the IPv4 protocol and fragment
@@ -1004,6 +1031,7 @@ enum ovs_action_attr {
 	OVS_ACTION_ATTR_ADD_MPLS,     /* struct ovs_action_add_mpls. */
 	OVS_ACTION_ATTR_DEC_TTL,      /* Nested OVS_DEC_TTL_ATTR_*. */
 	OVS_ACTION_ATTR_DROP,         /* u32 error code. */
+	OVS_ACTION_ATTR_PSAMPLE,      /* Nested OVS_PSAMPLE_ATTR_*. */
 
 	__OVS_ACTION_ATTR_MAX,	      /* Nothing past this will be accepted
 				       * from userspace. */
diff --git a/net/openvswitch/Kconfig b/net/openvswitch/Kconfig
index 29a7081858cd..2535f3f9f462 100644
--- a/net/openvswitch/Kconfig
+++ b/net/openvswitch/Kconfig
@@ -10,6 +10,7 @@ config OPENVSWITCH
 		   (NF_CONNTRACK && ((!NF_DEFRAG_IPV6 || NF_DEFRAG_IPV6) && \
 				     (!NF_NAT || NF_NAT) && \
 				     (!NETFILTER_CONNCOUNT || NETFILTER_CONNCOUNT)))
+	depends on PSAMPLE || !PSAMPLE
 	select LIBCRC32C
 	select MPLS
 	select NET_MPLS_GSO
diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c
index 964225580824..892d7e48fc5b 100644
--- a/net/openvswitch/actions.c
+++ b/net/openvswitch/actions.c
@@ -24,6 +24,11 @@
 #include <net/checksum.h>
 #include <net/dsfield.h>
 #include <net/mpls.h>
+
+#if IS_ENABLED(CONFIG_PSAMPLE)
+#include <net/psample.h>
+#endif
+
 #include <net/sctp/checksum.h>
 
 #include "datapath.h"
@@ -1299,6 +1304,40 @@ static int execute_dec_ttl(struct sk_buff *skb, struct sw_flow_key *key)
 	return 0;
 }
 
+#if IS_ENABLED(CONFIG_PSAMPLE)
+static void execute_psample(struct datapath *dp, struct sk_buff *skb,
+			    const struct nlattr *attr)
+{
+	struct psample_group psample_group = {};
+	struct psample_metadata md = {};
+	const struct nlattr *a;
+	int rem;
+
+	nla_for_each_attr(a, nla_data(attr), nla_len(attr), rem) {
+		switch (nla_type(a)) {
+		case OVS_PSAMPLE_ATTR_GROUP:
+			psample_group.group_num = nla_get_u32(a);
+			break;
+
+		case OVS_PSAMPLE_ATTR_COOKIE:
+			md.user_cookie = nla_data(a);
+			md.user_cookie_len = nla_len(a);
+			break;
+		}
+	}
+
+	psample_group.net = ovs_dp_get_net(dp);
+	md.in_ifindex = OVS_CB(skb)->input_vport->dev->ifindex;
+	md.trunc_size = skb->len - OVS_CB(skb)->cutlen;
+
+	psample_sample_packet(&psample_group, skb, 0, &md);
+}
+#else
+static void execute_psample(struct datapath *dp, struct sk_buff *skb,
+			    const struct nlattr *attr)
+{}
+#endif
+
 /* Execute a list of actions against 'skb'. */
 static int do_execute_actions(struct datapath *dp, struct sk_buff *skb,
 			      struct sw_flow_key *key,
@@ -1502,6 +1541,15 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb,
 			ovs_kfree_skb_reason(skb, reason);
 			return 0;
 		}
+
+		case OVS_ACTION_ATTR_PSAMPLE:
+			execute_psample(dp, skb, a);
+			OVS_CB(skb)->cutlen = 0;
+			if (nla_is_last(a, rem)) {
+				consume_skb(skb);
+				return 0;
+			}
+			break;
 		}
 
 		if (unlikely(err)) {
diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c
index f224d9bcea5e..c92bdc4dfe19 100644
--- a/net/openvswitch/flow_netlink.c
+++ b/net/openvswitch/flow_netlink.c
@@ -64,6 +64,7 @@ static bool actions_may_change_flow(const struct nlattr *actions)
 		case OVS_ACTION_ATTR_TRUNC:
 		case OVS_ACTION_ATTR_USERSPACE:
 		case OVS_ACTION_ATTR_DROP:
+		case OVS_ACTION_ATTR_PSAMPLE:
 			break;
 
 		case OVS_ACTION_ATTR_CT:
@@ -2409,7 +2410,7 @@ static void ovs_nla_free_nested_actions(const struct nlattr *actions, int len)
 	/* Whenever new actions are added, the need to update this
 	 * function should be considered.
 	 */
-	BUILD_BUG_ON(OVS_ACTION_ATTR_MAX != 24);
+	BUILD_BUG_ON(OVS_ACTION_ATTR_MAX != 25);
 
 	if (!actions)
 		return;
@@ -3157,6 +3158,28 @@ static int validate_and_copy_check_pkt_len(struct net *net,
 	return 0;
 }
 
+static int validate_psample(const struct nlattr *attr)
+{
+	static const struct nla_policy policy[OVS_PSAMPLE_ATTR_MAX + 1] = {
+		[OVS_PSAMPLE_ATTR_GROUP] = { .type = NLA_U32 },
+		[OVS_PSAMPLE_ATTR_COOKIE] = {
+			.type = NLA_BINARY,
+			.len = OVS_PSAMPLE_COOKIE_MAX_SIZE,
+		},
+	};
+	struct nlattr *a[OVS_PSAMPLE_ATTR_MAX + 1];
+	int err;
+
+	if (!IS_ENABLED(CONFIG_PSAMPLE))
+		return -EOPNOTSUPP;
+
+	err = nla_parse_nested(a, OVS_PSAMPLE_ATTR_MAX, attr, policy, NULL);
+	if (err)
+		return err;
+
+	return a[OVS_PSAMPLE_ATTR_GROUP] ? 0 : -EINVAL;
+}
+
 static int copy_action(const struct nlattr *from,
 		       struct sw_flow_actions **sfa, bool log)
 {
@@ -3212,6 +3235,7 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 			[OVS_ACTION_ATTR_ADD_MPLS] = sizeof(struct ovs_action_add_mpls),
 			[OVS_ACTION_ATTR_DEC_TTL] = (u32)-1,
 			[OVS_ACTION_ATTR_DROP] = sizeof(u32),
+			[OVS_ACTION_ATTR_PSAMPLE] = (u32)-1,
 		};
 		const struct ovs_action_push_vlan *vlan;
 		int type = nla_type(a);
@@ -3490,6 +3514,12 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr,
 				return -EINVAL;
 			break;
 
+		case OVS_ACTION_ATTR_PSAMPLE:
+			err = validate_psample(a);
+			if (err)
+				return err;
+			break;
+
 		default:
 			OVS_NLERR(log, "Unknown Action type %d", type);
 			return -EINVAL;
-- 
cgit v1.2.3-58-ga151


From 71763d8a8203c28178d7be7f18af73d4dddb36ba Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Thu, 4 Jul 2024 10:56:57 +0200
Subject: net: openvswitch: store sampling probability in cb.

When a packet sample is observed, the sampling rate that was used is
important to estimate the real frequency of such event.

Store the probability of the parent sample action in the skb's cb area
and use it in psample action to pass it down to psample module.

Reviewed-by: Aaron Conole <aconole@redhat.com>
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Reviewed-by: Ilya Maximets <i.maximets@ovn.org>
Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Link: https://patch.msgid.link/20240704085710.353845-7-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/uapi/linux/openvswitch.h |  3 ++-
 net/openvswitch/actions.c        | 20 +++++++++++++++++---
 net/openvswitch/datapath.h       |  3 +++
 net/openvswitch/vport.c          |  1 +
 4 files changed, 23 insertions(+), 4 deletions(-)

diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h
index 3dd653748725..3a701bd1f31b 100644
--- a/include/uapi/linux/openvswitch.h
+++ b/include/uapi/linux/openvswitch.h
@@ -649,7 +649,8 @@ enum ovs_flow_attr {
  * Actions are passed as nested attributes.
  *
  * Executes the specified actions with the given probability on a per-packet
- * basis.
+ * basis. Nested actions will be able to access the probability value of the
+ * parent @OVS_ACTION_ATTR_SAMPLE.
  */
 enum ovs_sample_attr {
 	OVS_SAMPLE_ATTR_UNSPEC,
diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c
index 892d7e48fc5b..101f9a23792c 100644
--- a/net/openvswitch/actions.c
+++ b/net/openvswitch/actions.c
@@ -1048,12 +1048,15 @@ static int sample(struct datapath *dp, struct sk_buff *skb,
 	struct nlattr *sample_arg;
 	int rem = nla_len(attr);
 	const struct sample_arg *arg;
+	u32 init_probability;
 	bool clone_flow_key;
+	int err;
 
 	/* The first action is always 'OVS_SAMPLE_ATTR_ARG'. */
 	sample_arg = nla_data(attr);
 	arg = nla_data(sample_arg);
 	actions = nla_next(sample_arg, &rem);
+	init_probability = OVS_CB(skb)->probability;
 
 	if ((arg->probability != U32_MAX) &&
 	    (!arg->probability || get_random_u32() > arg->probability)) {
@@ -1062,9 +1065,16 @@ static int sample(struct datapath *dp, struct sk_buff *skb,
 		return 0;
 	}
 
+	OVS_CB(skb)->probability = arg->probability;
+
 	clone_flow_key = !arg->exec;
-	return clone_execute(dp, skb, key, 0, actions, rem, last,
-			     clone_flow_key);
+	err = clone_execute(dp, skb, key, 0, actions, rem, last,
+			    clone_flow_key);
+
+	if (!last)
+		OVS_CB(skb)->probability = init_probability;
+
+	return err;
 }
 
 /* When 'last' is true, clone() should always consume the 'skb'.
@@ -1311,6 +1321,7 @@ static void execute_psample(struct datapath *dp, struct sk_buff *skb,
 	struct psample_group psample_group = {};
 	struct psample_metadata md = {};
 	const struct nlattr *a;
+	u32 rate;
 	int rem;
 
 	nla_for_each_attr(a, nla_data(attr), nla_len(attr), rem) {
@@ -1329,8 +1340,11 @@ static void execute_psample(struct datapath *dp, struct sk_buff *skb,
 	psample_group.net = ovs_dp_get_net(dp);
 	md.in_ifindex = OVS_CB(skb)->input_vport->dev->ifindex;
 	md.trunc_size = skb->len - OVS_CB(skb)->cutlen;
+	md.rate_as_probability = 1;
+
+	rate = OVS_CB(skb)->probability ? OVS_CB(skb)->probability : U32_MAX;
 
-	psample_sample_packet(&psample_group, skb, 0, &md);
+	psample_sample_packet(&psample_group, skb, rate, &md);
 }
 #else
 static void execute_psample(struct datapath *dp, struct sk_buff *skb,
diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h
index 0cd29971a907..9ca6231ea647 100644
--- a/net/openvswitch/datapath.h
+++ b/net/openvswitch/datapath.h
@@ -115,12 +115,15 @@ struct datapath {
  * fragmented.
  * @acts_origlen: The netlink size of the flow actions applied to this skb.
  * @cutlen: The number of bytes from the packet end to be removed.
+ * @probability: The sampling probability that was applied to this skb; 0 means
+ * no sampling has occurred; U32_MAX means 100% probability.
  */
 struct ovs_skb_cb {
 	struct vport		*input_vport;
 	u16			mru;
 	u16			acts_origlen;
 	u32			cutlen;
+	u32			probability;
 };
 #define OVS_CB(skb) ((struct ovs_skb_cb *)(skb)->cb)
 
diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c
index 972ae01a70f7..8732f6e51ae5 100644
--- a/net/openvswitch/vport.c
+++ b/net/openvswitch/vport.c
@@ -500,6 +500,7 @@ int ovs_vport_receive(struct vport *vport, struct sk_buff *skb,
 	OVS_CB(skb)->input_vport = vport;
 	OVS_CB(skb)->mru = 0;
 	OVS_CB(skb)->cutlen = 0;
+	OVS_CB(skb)->probability = 0;
 	if (unlikely(dev_net(skb->dev) != ovs_dp_get_net(vport->dp))) {
 		u32 mark;
 
-- 
cgit v1.2.3-58-ga151


From 60ccf62d3ceb37c89391e60d5ba402f52b720b62 Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Thu, 4 Jul 2024 10:56:58 +0200
Subject: selftests: openvswitch: add psample action

Add sample and psample action support to ovs-dpctl.py.

Refactor common attribute parsing logic into an external function.

Reviewed-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Link: https://patch.msgid.link/20240704085710.353845-8-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../testing/selftests/net/openvswitch/ovs-dpctl.py | 162 ++++++++++++++++++++-
 1 file changed, 161 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 182a09975975..dcc400a21a22 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -8,6 +8,7 @@ import argparse
 import errno
 import ipaddress
 import logging
+import math
 import multiprocessing
 import re
 import socket
@@ -60,6 +61,7 @@ OVS_FLOW_CMD_DEL = 2
 OVS_FLOW_CMD_GET = 3
 OVS_FLOW_CMD_SET = 4
 
+UINT32_MAX = 0xFFFFFFFF
 
 def macstr(mac):
     outstr = ":".join(["%02X" % i for i in mac])
@@ -281,6 +283,75 @@ def parse_extract_field(
     return str_skipped, data
 
 
+def parse_attrs(actstr, attr_desc):
+    """Parses the given action string and returns a list of netlink
+    attributes based on a list of attribute descriptions.
+
+    Each element in the attribute description list is a tuple such as:
+        (name, attr_name, parse_func)
+    where:
+        name: is the string representing the attribute
+        attr_name: is the name of the attribute as defined in the uAPI.
+        parse_func: is a callable accepting a string and returning either
+            a single object (the parsed attribute value) or a tuple of
+            two values (the parsed attribute value and the remaining string)
+
+    Returns a list of attributes and the remaining string.
+    """
+    def parse_attr(actstr, key, func):
+        actstr = actstr[len(key) :]
+
+        if not func:
+            return None, actstr
+
+        delim = actstr[0]
+        actstr = actstr[1:]
+
+        if delim == "=":
+            pos = strcspn(actstr, ",)")
+            ret = func(actstr[:pos])
+        else:
+            ret = func(actstr)
+
+        if isinstance(ret, tuple):
+            (datum, actstr) = ret
+        else:
+            datum = ret
+            actstr = actstr[strcspn(actstr, ",)"):]
+
+        if delim == "(":
+            if not actstr or actstr[0] != ")":
+                raise ValueError("Action contains unbalanced parentheses")
+
+            actstr = actstr[1:]
+
+        actstr = actstr[strspn(actstr, ", ") :]
+
+        return datum, actstr
+
+    attrs = []
+    attr_desc = list(attr_desc)
+    while actstr and actstr[0] != ")" and attr_desc:
+        found = False
+        for i, (key, attr, func) in enumerate(attr_desc):
+            if actstr.startswith(key):
+                datum, actstr = parse_attr(actstr, key, func)
+                attrs.append([attr, datum])
+                found = True
+                del attr_desc[i]
+
+        if not found:
+            raise ValueError("Unknown attribute: '%s'" % actstr)
+
+        actstr = actstr[strspn(actstr, ", ") :]
+
+    if actstr[0] != ")":
+        raise ValueError("Action string contains extra garbage or has "
+                         "unbalanced parenthesis: '%s'" % actstr)
+
+    return attrs, actstr[1:]
+
+
 class ovs_dp_msg(genlmsg):
     # include the OVS version
     # We need a custom header rather than just being able to rely on
@@ -299,7 +370,7 @@ class ovsactions(nla):
         ("OVS_ACTION_ATTR_SET", "ovskey"),
         ("OVS_ACTION_ATTR_PUSH_VLAN", "none"),
         ("OVS_ACTION_ATTR_POP_VLAN", "flag"),
-        ("OVS_ACTION_ATTR_SAMPLE", "none"),
+        ("OVS_ACTION_ATTR_SAMPLE", "sample"),
         ("OVS_ACTION_ATTR_RECIRC", "uint32"),
         ("OVS_ACTION_ATTR_HASH", "none"),
         ("OVS_ACTION_ATTR_PUSH_MPLS", "none"),
@@ -318,8 +389,85 @@ class ovsactions(nla):
         ("OVS_ACTION_ATTR_ADD_MPLS", "none"),
         ("OVS_ACTION_ATTR_DEC_TTL", "none"),
         ("OVS_ACTION_ATTR_DROP", "uint32"),
+        ("OVS_ACTION_ATTR_PSAMPLE", "psample"),
     )
 
+    class psample(nla):
+        nla_flags = NLA_F_NESTED
+
+        nla_map = (
+            ("OVS_PSAMPLE_ATTR_UNSPEC", "none"),
+            ("OVS_PSAMPLE_ATTR_GROUP", "uint32"),
+            ("OVS_PSAMPLE_ATTR_COOKIE", "array(uint8)"),
+        )
+
+        def dpstr(self, more=False):
+            args = "group=%d" % self.get_attr("OVS_PSAMPLE_ATTR_GROUP")
+
+            cookie = self.get_attr("OVS_PSAMPLE_ATTR_COOKIE")
+            if cookie:
+                args += ",cookie(%s)" % \
+                        "".join(format(x, "02x") for x in cookie)
+
+            return "psample(%s)" % args
+
+        def parse(self, actstr):
+            desc = (
+                ("group", "OVS_PSAMPLE_ATTR_GROUP", int),
+                ("cookie", "OVS_PSAMPLE_ATTR_COOKIE",
+                    lambda x: list(bytearray.fromhex(x)))
+            )
+
+            attrs, actstr = parse_attrs(actstr, desc)
+
+            for attr in attrs:
+                self["attrs"].append(attr)
+
+            return actstr
+
+    class sample(nla):
+        nla_flags = NLA_F_NESTED
+
+        nla_map = (
+            ("OVS_SAMPLE_ATTR_UNSPEC", "none"),
+            ("OVS_SAMPLE_ATTR_PROBABILITY", "uint32"),
+            ("OVS_SAMPLE_ATTR_ACTIONS", "ovsactions"),
+        )
+
+        def dpstr(self, more=False):
+            args = []
+
+            args.append("sample={:.2f}%".format(
+                100 * self.get_attr("OVS_SAMPLE_ATTR_PROBABILITY") /
+                UINT32_MAX))
+
+            actions = self.get_attr("OVS_SAMPLE_ATTR_ACTIONS")
+            if actions:
+                args.append("actions(%s)" % actions.dpstr(more))
+
+            return "sample(%s)" % ",".join(args)
+
+        def parse(self, actstr):
+            def parse_nested_actions(actstr):
+                subacts = ovsactions()
+                parsed_len = subacts.parse(actstr)
+                return subacts, actstr[parsed_len :]
+
+            def percent_to_rate(percent):
+                percent = float(percent.strip('%'))
+                return int(math.floor(UINT32_MAX * (percent / 100.0) + .5))
+
+            desc = (
+                ("sample", "OVS_SAMPLE_ATTR_PROBABILITY", percent_to_rate),
+                ("actions", "OVS_SAMPLE_ATTR_ACTIONS", parse_nested_actions),
+            )
+            attrs, actstr = parse_attrs(actstr, desc)
+
+            for attr in attrs:
+                self["attrs"].append(attr)
+
+            return actstr
+
     class ctact(nla):
         nla_flags = NLA_F_NESTED
 
@@ -683,6 +831,18 @@ class ovsactions(nla):
                 self["attrs"].append(["OVS_ACTION_ATTR_CT", ctact])
                 parsed = True
 
+            elif parse_starts_block(actstr, "sample(", False):
+                sampleact = self.sample()
+                actstr = sampleact.parse(actstr[len("sample(") : ])
+                self["attrs"].append(["OVS_ACTION_ATTR_SAMPLE", sampleact])
+                parsed = True
+
+            elif parse_starts_block(actstr, "psample(", False):
+                psampleact = self.psample()
+                actstr = psampleact.parse(actstr[len("psample(") : ])
+                self["attrs"].append(["OVS_ACTION_ATTR_PSAMPLE", psampleact])
+                parsed = True
+
             actstr = actstr[strspn(actstr, ", ") :]
             while parencount > 0:
                 parencount -= 1
-- 
cgit v1.2.3-58-ga151


From c7815abbea457afa2a1ab3ded46d29c19da377d2 Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Thu, 4 Jul 2024 10:56:59 +0200
Subject: selftests: openvswitch: add userspace parsing

The userspace action lacks parsing support plus it contains a bug in the
name of one of its attributes.

This patch makes userspace action work.

Reviewed-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Link: https://patch.msgid.link/20240704085710.353845-9-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../testing/selftests/net/openvswitch/ovs-dpctl.py | 24 ++++++++++++++++++++--
 1 file changed, 22 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index dcc400a21a22..4ccf26f96327 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -589,13 +589,27 @@ class ovsactions(nla):
                 print_str += "userdata="
                 for f in self.get_attr("OVS_USERSPACE_ATTR_USERDATA"):
                     print_str += "%x." % f
-            if self.get_attr("OVS_USERSPACE_ATTR_TUN_PORT") is not None:
+            if self.get_attr("OVS_USERSPACE_ATTR_EGRESS_TUN_PORT") is not None:
                 print_str += "egress_tun_port=%d" % self.get_attr(
-                    "OVS_USERSPACE_ATTR_TUN_PORT"
+                    "OVS_USERSPACE_ATTR_EGRESS_TUN_PORT"
                 )
             print_str += ")"
             return print_str
 
+        def parse(self, actstr):
+            attrs_desc = (
+                ("pid", "OVS_USERSPACE_ATTR_PID", int),
+                ("userdata", "OVS_USERSPACE_ATTR_USERDATA",
+                    lambda x: list(bytearray.fromhex(x))),
+                ("egress_tun_port", "OVS_USERSPACE_ATTR_EGRESS_TUN_PORT", int)
+            )
+
+            attrs, actstr = parse_attrs(actstr, attrs_desc)
+            for attr in attrs:
+                self["attrs"].append(attr)
+
+            return actstr
+
     def dpstr(self, more=False):
         print_str = ""
 
@@ -843,6 +857,12 @@ class ovsactions(nla):
                 self["attrs"].append(["OVS_ACTION_ATTR_PSAMPLE", psampleact])
                 parsed = True
 
+            elif parse_starts_block(actstr, "userspace(", False):
+                uact = self.userspace()
+                actstr = uact.parse(actstr[len("userspace(") : ])
+                self["attrs"].append(["OVS_ACTION_ATTR_USERSPACE", uact])
+                parsed = True
+
             actstr = actstr[strspn(actstr, ", ") :]
             while parencount > 0:
                 parencount -= 1
-- 
cgit v1.2.3-58-ga151


From b192bf12dbb015d37bb904d61751cad431ceb578 Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Thu, 4 Jul 2024 10:57:00 +0200
Subject: selftests: openvswitch: parse trunc action

The trunc action was supported decode-able but not parse-able. Add
support for parsing the action string.

Reviewed-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Link: https://patch.msgid.link/20240704085710.353845-10-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/openvswitch/ovs-dpctl.py | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 4ccf26f96327..e8dc9af10d4d 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -863,6 +863,19 @@ class ovsactions(nla):
                 self["attrs"].append(["OVS_ACTION_ATTR_USERSPACE", uact])
                 parsed = True
 
+            elif parse_starts_block(actstr, "trunc(", False):
+                parencount += 1
+                actstr, val = parse_extract_field(
+                    actstr,
+                    "trunc(",
+                    r"([0-9]+)",
+                    int,
+                    False,
+                    None,
+                )
+                self["attrs"].append(["OVS_ACTION_ATTR_TRUNC", val])
+                parsed = True
+
             actstr = actstr[strspn(actstr, ", ") :]
             while parencount > 0:
                 parencount -= 1
-- 
cgit v1.2.3-58-ga151


From 30d772a03582bd10447af1bcdd1c731c89b54174 Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Thu, 4 Jul 2024 10:57:01 +0200
Subject: selftests: openvswitch: add psample test

Add a test to verify sampling packets via psample works.

In order to do that, create a subcommand in ovs-dpctl.py to listen to
on the psample multicast group and print samples.

Reviewed-by: Aaron Conole <aconole@redhat.com>
Tested-by: Ilya Maximets <i.maximets@ovn.org>
Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Link: https://patch.msgid.link/20240704085710.353845-11-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../selftests/net/openvswitch/openvswitch.sh       | 115 ++++++++++++++++++++-
 .../testing/selftests/net/openvswitch/ovs-dpctl.py |  73 ++++++++++++-
 2 files changed, 182 insertions(+), 6 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/openvswitch.sh b/tools/testing/selftests/net/openvswitch/openvswitch.sh
index 531951086d9c..bc71dbc18b21 100755
--- a/tools/testing/selftests/net/openvswitch/openvswitch.sh
+++ b/tools/testing/selftests/net/openvswitch/openvswitch.sh
@@ -20,7 +20,8 @@ tests="
 	nat_related_v4				ip4-nat-related: ICMP related matches work with SNAT
 	netlink_checks				ovsnl: validate netlink attrs and settings
 	upcall_interfaces			ovs: test the upcall interfaces
-	drop_reason				drop: test drop reasons are emitted"
+	drop_reason				drop: test drop reasons are emitted
+	psample					psample: Sampling packets with psample"
 
 info() {
 	[ "${ovs_dir}" != "" ] &&
@@ -105,12 +106,21 @@ ovs_netns_spawn_daemon() {
 	shift
 	netns=$1
 	shift
-	info "spawning cmd: $*"
-	ip netns exec $netns $*  >> $ovs_dir/stdout  2>> $ovs_dir/stderr &
+	if [ "$netns" == "_default" ]; then
+		$*  >> $ovs_dir/stdout  2>> $ovs_dir/stderr &
+	else
+		ip netns exec $netns $*  >> $ovs_dir/stdout  2>> $ovs_dir/stderr &
+	fi
 	pid=$!
 	ovs_sbx "$sbx" on_exit "kill -TERM $pid 2>/dev/null"
 }
 
+ovs_spawn_daemon() {
+	sbx=$1
+	shift
+	ovs_netns_spawn_daemon $sbx "_default" $*
+}
+
 ovs_add_netns_and_veths () {
 	info "Adding netns attached: sbx:$1 dp:$2 {$3, $4, $5}"
 	ovs_sbx "$1" ip netns add "$3" || return 1
@@ -173,6 +183,19 @@ ovs_drop_reason_count()
 	return `echo "$perf_output" | grep "$pattern" | wc -l`
 }
 
+ovs_test_flow_fails () {
+	ERR_MSG="Flow actions may not be safe on all matching packets"
+
+	PRE_TEST=$(dmesg | grep -c "${ERR_MSG}")
+	ovs_add_flow $@ &> /dev/null $@ && return 1
+	POST_TEST=$(dmesg | grep -c "${ERR_MSG}")
+
+	if [ "$PRE_TEST" == "$POST_TEST" ]; then
+		return 1
+	fi
+	return 0
+}
+
 usage() {
 	echo
 	echo "$0 [OPTIONS] [TEST]..."
@@ -187,6 +210,92 @@ usage() {
 	exit 1
 }
 
+
+# psample test
+# - use psample to observe packets
+test_psample() {
+	sbx_add "test_psample" || return $?
+
+	# Add a datapath with per-vport dispatching.
+	ovs_add_dp "test_psample" psample -V 2:1 || return 1
+
+	info "create namespaces"
+	ovs_add_netns_and_veths "test_psample" "psample" \
+		client c0 c1 172.31.110.10/24 -u || return 1
+	ovs_add_netns_and_veths "test_psample" "psample" \
+		server s0 s1 172.31.110.20/24 -u || return 1
+
+	# Check if psample actions can be configured.
+	ovs_add_flow "test_psample" psample \
+	'in_port(1),eth(),eth_type(0x0806),arp()' 'psample(group=1)' &> /dev/null
+	if [ $? == 1 ]; then
+		info "no support for psample - skipping"
+		ovs_exit_sig
+		return $ksft_skip
+	fi
+
+	ovs_del_flows "test_psample" psample
+
+	# Test action verification.
+	OLDIFS=$IFS
+	IFS='*'
+	min_key='in_port(1),eth(),eth_type(0x0800),ipv4()'
+	for testcase in \
+		"cookie to large"*"psample(group=1,cookie=1615141312111009080706050403020100)" \
+		"no group with cookie"*"psample(cookie=abcd)" \
+		"no group"*"psample()";
+	do
+		set -- $testcase;
+		ovs_test_flow_fails "test_psample" psample $min_key $2
+		if [ $? == 1 ]; then
+			info "failed - $1"
+			return 1
+		fi
+	done
+	IFS=$OLDIFS
+
+	ovs_del_flows "test_psample" psample
+	# Allow ARP
+	ovs_add_flow "test_psample" psample \
+		'in_port(1),eth(),eth_type(0x0806),arp()' '2' || return 1
+	ovs_add_flow "test_psample" psample \
+		'in_port(2),eth(),eth_type(0x0806),arp()' '1' || return 1
+
+	# Sample first 14 bytes of all traffic.
+	ovs_add_flow "test_psample" psample \
+	    "in_port(1),eth(),eth_type(0x0800),ipv4()" \
+            "trunc(14),psample(group=1,cookie=c0ffee),2"
+
+	# Sample all traffic. In this case, use a sample() action with both
+	# psample and an upcall emulating simultaneous local sampling and
+	# sFlow / IPFIX.
+	nlpid=$(grep -E "listening on upcall packet handler" \
+            $ovs_dir/s0.out | cut -d ":" -f 2 | tr -d ' ')
+
+	ovs_add_flow "test_psample" psample \
+            "in_port(2),eth(),eth_type(0x0800),ipv4()" \
+            "sample(sample=100%,actions(psample(group=2,cookie=eeff0c),userspace(pid=${nlpid},userdata=eeff0c))),1"
+
+	# Record psample data.
+	ovs_spawn_daemon "test_psample" python3 $ovs_base/ovs-dpctl.py psample-events
+
+	# Send a single ping.
+	sleep 1
+	ovs_sbx "test_psample" ip netns exec client ping -I c1 172.31.110.20 -c 1 || return 1
+	sleep 1
+
+	# We should have received one userspace action upcall and 2 psample packets.
+	grep -E "userspace action command" $ovs_dir/s0.out >/dev/null 2>&1 || return 1
+
+	# client -> server samples should only contain the first 14 bytes of the packet.
+	grep -E "rate:4294967295,group:1,cookie:c0ffee data:[0-9a-f]{28}$" \
+			 $ovs_dir/stdout >/dev/null 2>&1 || return 1
+	grep -E "rate:4294967295,group:2,cookie:eeff0c" \
+			 $ovs_dir/stdout >/dev/null 2>&1 || return 1
+
+	return 0
+}
+
 # drop_reason test
 # - drop packets and verify the right drop reason is reported
 test_drop_reason() {
diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index e8dc9af10d4d..1e15b0818074 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -28,8 +28,10 @@ try:
     from pyroute2.netlink import genlmsg
     from pyroute2.netlink import nla
     from pyroute2.netlink import nlmsg_atoms
+    from pyroute2.netlink.event import EventSocket
     from pyroute2.netlink.exceptions import NetlinkError
     from pyroute2.netlink.generic import GenericNetlinkSocket
+    from pyroute2.netlink.nlsocket import Marshal
     import pyroute2
     import pyroute2.iproute
 
@@ -2460,10 +2462,70 @@ class OvsFlow(GenericNetlinkSocket):
         print("MISS upcall[%d/%s]: %s" % (seq, pktpres, keystr), flush=True)
 
     def execute(self, packetmsg):
-        print("userspace execute command")
+        print("userspace execute command", flush=True)
 
     def action(self, packetmsg):
-        print("userspace action command")
+        print("userspace action command", flush=True)
+
+
+class psample_sample(genlmsg):
+    nla_map = (
+        ("PSAMPLE_ATTR_IIFINDEX", "none"),
+        ("PSAMPLE_ATTR_OIFINDEX", "none"),
+        ("PSAMPLE_ATTR_ORIGSIZE", "none"),
+        ("PSAMPLE_ATTR_SAMPLE_GROUP", "uint32"),
+        ("PSAMPLE_ATTR_GROUP_SEQ", "none"),
+        ("PSAMPLE_ATTR_SAMPLE_RATE", "uint32"),
+        ("PSAMPLE_ATTR_DATA", "array(uint8)"),
+        ("PSAMPLE_ATTR_GROUP_REFCOUNT", "none"),
+        ("PSAMPLE_ATTR_TUNNEL", "none"),
+        ("PSAMPLE_ATTR_PAD", "none"),
+        ("PSAMPLE_ATTR_OUT_TC", "none"),
+        ("PSAMPLE_ATTR_OUT_TC_OCC", "none"),
+        ("PSAMPLE_ATTR_LATENCY", "none"),
+        ("PSAMPLE_ATTR_TIMESTAMP", "none"),
+        ("PSAMPLE_ATTR_PROTO", "none"),
+        ("PSAMPLE_ATTR_USER_COOKIE", "array(uint8)"),
+    )
+
+    def dpstr(self):
+        fields = []
+        data = ""
+        for (attr, value) in self["attrs"]:
+            if attr == "PSAMPLE_ATTR_SAMPLE_GROUP":
+                fields.append("group:%d" % value)
+            if attr == "PSAMPLE_ATTR_SAMPLE_RATE":
+                fields.append("rate:%d" % value)
+            if attr == "PSAMPLE_ATTR_USER_COOKIE":
+                value = "".join(format(x, "02x") for x in value)
+                fields.append("cookie:%s" % value)
+            if attr == "PSAMPLE_ATTR_DATA" and len(value) > 0:
+                data = "data:%s" % "".join(format(x, "02x") for x in value)
+
+        return ("%s %s" % (",".join(fields), data)).strip()
+
+
+class psample_msg(Marshal):
+    PSAMPLE_CMD_SAMPLE = 0
+    PSAMPLE_CMD_GET_GROUP = 1
+    PSAMPLE_CMD_NEW_GROUP = 2
+    PSAMPLE_CMD_DEL_GROUP = 3
+    PSAMPLE_CMD_SET_FILTER = 4
+    msg_map = {PSAMPLE_CMD_SAMPLE: psample_sample}
+
+
+class PsampleEvent(EventSocket):
+    genl_family = "psample"
+    mcast_groups = ["packets"]
+    marshal_class = psample_msg
+
+    def read_samples(self):
+        while True:
+            try:
+                for msg in self.get():
+                    print(msg.dpstr(), flush=True)
+            except NetlinkError as ne:
+                raise ne
 
 
 def print_ovsdp_full(dp_lookup_rep, ifindex, ndb=NDB(), vpl=OvsVport()):
@@ -2530,7 +2592,7 @@ def main(argv):
         help="Increment 'verbose' output counter.",
         default=0,
     )
-    subparsers = parser.add_subparsers()
+    subparsers = parser.add_subparsers(dest="subcommand")
 
     showdpcmd = subparsers.add_parser("show")
     showdpcmd.add_argument(
@@ -2605,6 +2667,8 @@ def main(argv):
     delfscmd = subparsers.add_parser("del-flows")
     delfscmd.add_argument("flsbr", help="Datapath name")
 
+    subparsers.add_parser("psample-events")
+
     args = parser.parse_args()
 
     if args.verbose > 0:
@@ -2619,6 +2683,9 @@ def main(argv):
 
     sys.setrecursionlimit(100000)
 
+    if args.subcommand == "psample-events":
+        PsampleEvent().read_samples()
+
     if hasattr(args, "showdp"):
         found = False
         for iface in ndb.interfaces:
-- 
cgit v1.2.3-58-ga151


From e46296002113b4556baffd5dc68c4f9e22dae13a Mon Sep 17 00:00:00 2001
From: "Kory Maincent (Dent Project)" <kory.maincent@bootlin.com>
Date: Thu, 4 Jul 2024 10:11:56 +0200
Subject: net: ethtool: pse-pd: Expand C33 PSE status with class, power and
 extended state

This update expands the status information provided by ethtool for PSE c33.
It includes details such as the detected class, current power delivered,
and extended state information.

Reviewed-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240704-feature_poe_power_cap-v6-1-320003204264@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/ethtool-netlink.rst |  37 ++++++
 include/linux/ethtool.h                      |  15 +++
 include/linux/pse-pd/pse.h                   |   8 ++
 include/uapi/linux/ethtool.h                 | 191 +++++++++++++++++++++++++++
 include/uapi/linux/ethtool_netlink.h         |   4 +
 net/ethtool/pse-pd.c                         |  32 ++++-
 6 files changed, 286 insertions(+), 1 deletion(-)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index bfe2eda8580d..0656ad4be000 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -1740,6 +1740,13 @@ Kernel response contents:
                                                   PSE functions.
   ``ETHTOOL_A_C33_PSE_PW_D_STATUS``          u32  power detection status of the
                                                   PoE PSE.
+  ``ETHTOOL_A_C33_PSE_PW_CLASS``             u32  power class of the PoE PSE.
+  ``ETHTOOL_A_C33_PSE_ACTUAL_PW``            u32  actual power drawn on the
+                                                  PoE PSE.
+  ``ETHTOOL_A_C33_PSE_EXT_STATE``            u32  power extended state of the
+                                                  PoE PSE.
+  ``ETHTOOL_A_C33_PSE_EXT_SUBSTATE``         u32  power extended substatus of
+                                                  the PoE PSE.
   ======================================  ======  =============================
 
 When set, the optional ``ETHTOOL_A_PODL_PSE_ADMIN_STATE`` attribute identifies
@@ -1772,6 +1779,36 @@ The same goes for ``ETHTOOL_A_C33_PSE_ADMIN_PW_D_STATUS`` implementing
 .. kernel-doc:: include/uapi/linux/ethtool.h
     :identifiers: ethtool_c33_pse_pw_d_status
 
+When set, the optional ``ETHTOOL_A_C33_PSE_PW_CLASS`` attribute identifies
+the power class of the C33 PSE. It depends on the class negotiated between
+the PSE and the PD. This option is corresponding to ``IEEE 802.3-2022``
+30.9.1.1.8 aPSEPowerClassification.
+
+When set, the optional ``ETHTOOL_A_C33_PSE_ACTUAL_PW`` attribute identifies
+This option is corresponding to ``IEEE 802.3-2022`` 30.9.1.1.23 aPSEActualPower.
+Actual power is reported in mW.
+
+When set, the optional ``ETHTOOL_A_C33_PSE_EXT_STATE`` attribute identifies
+the extended error state of the C33 PSE. Possible values are:
+
+.. kernel-doc:: include/uapi/linux/ethtool.h
+    :identifiers: ethtool_c33_pse_ext_state
+
+When set, the optional ``ETHTOOL_A_C33_PSE_EXT_SUBSTATE`` attribute identifies
+the extended error state of the C33 PSE. Possible values are:
+Possible values are:
+
+.. kernel-doc:: include/uapi/linux/ethtool.h
+    :identifiers: ethtool_c33_pse_ext_substate_class_num_events
+		  ethtool_c33_pse_ext_substate_error_condition
+		  ethtool_c33_pse_ext_substate_mr_pse_enable
+		  ethtool_c33_pse_ext_substate_option_detect_ted
+		  ethtool_c33_pse_ext_substate_option_vport_lim
+		  ethtool_c33_pse_ext_substate_ovld_detected
+		  ethtool_c33_pse_ext_substate_pd_dll_power_type
+		  ethtool_c33_pse_ext_substate_power_not_available
+		  ethtool_c33_pse_ext_substate_short_detected
+
 PSE_SET
 =======
 
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 3a99238ef895..3e70f5d9e0bb 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -1273,4 +1273,19 @@ struct ethtool_forced_speed_map {
 
 void
 ethtool_forced_speed_maps_init(struct ethtool_forced_speed_map *maps, u32 size);
+
+/* C33 PSE extended state and substate. */
+struct ethtool_c33_pse_ext_state_info {
+	enum ethtool_c33_pse_ext_state c33_pse_ext_state;
+	union {
+		enum ethtool_c33_pse_ext_substate_error_condition error_condition;
+		enum ethtool_c33_pse_ext_substate_mr_pse_enable mr_pse_enable;
+		enum ethtool_c33_pse_ext_substate_option_detect_ted option_detect_ted;
+		enum ethtool_c33_pse_ext_substate_option_vport_lim option_vport_lim;
+		enum ethtool_c33_pse_ext_substate_ovld_detected ovld_detected;
+		enum ethtool_c33_pse_ext_substate_power_not_available power_not_available;
+		enum ethtool_c33_pse_ext_substate_short_detected short_detected;
+		u32 __c33_pse_ext_substate;
+	};
+};
 #endif /* _LINUX_ETHTOOL_H */
diff --git a/include/linux/pse-pd/pse.h b/include/linux/pse-pd/pse.h
index 6eec24ffa866..38b9308e5e7a 100644
--- a/include/linux/pse-pd/pse.h
+++ b/include/linux/pse-pd/pse.h
@@ -36,12 +36,20 @@ struct pse_control_config {
  *	functions. IEEE 802.3-2022 30.9.1.1.2 aPSEAdminState
  * @c33_pw_status: power detection status of the PSE.
  *	IEEE 802.3-2022 30.9.1.1.5 aPSEPowerDetectionStatus:
+ * @c33_pw_class: detected class of a powered PD
+ *	IEEE 802.3-2022 30.9.1.1.8 aPSEPowerClassification
+ * @c33_actual_pw: power currently delivered by the PSE in mW
+ *	IEEE 802.3-2022 30.9.1.1.23 aPSEActualPower
+ * @c33_ext_state_info: extended state information of the PSE
  */
 struct pse_control_status {
 	enum ethtool_podl_pse_admin_state podl_admin_state;
 	enum ethtool_podl_pse_pw_d_status podl_pw_status;
 	enum ethtool_c33_pse_admin_state c33_admin_state;
 	enum ethtool_c33_pse_pw_d_status c33_pw_status;
+	u32 c33_pw_class;
+	u32 c33_actual_pw;
+	struct ethtool_c33_pse_ext_state_info c33_ext_state_info;
 };
 
 /**
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index e011384c915c..230110b97029 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -752,6 +752,197 @@ enum ethtool_module_power_mode {
 	ETHTOOL_MODULE_POWER_MODE_HIGH,
 };
 
+/**
+ * enum ethtool_c33_pse_ext_state - groups of PSE extended states
+ *      functions. IEEE 802.3-2022 33.2.4.4 Variables
+ *
+ * @ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION: Group of error_condition states
+ * @ETHTOOL_C33_PSE_EXT_STATE_MR_MPS_VALID: Group of mr_mps_valid states
+ * @ETHTOOL_C33_PSE_EXT_STATE_MR_PSE_ENABLE: Group of mr_pse_enable states
+ * @ETHTOOL_C33_PSE_EXT_STATE_OPTION_DETECT_TED: Group of option_detect_ted
+ *	states
+ * @ETHTOOL_C33_PSE_EXT_STATE_OPTION_VPORT_LIM: Group of option_vport_lim states
+ * @ETHTOOL_C33_PSE_EXT_STATE_OVLD_DETECTED: Group of ovld_detected states
+ * @ETHTOOL_C33_PSE_EXT_STATE_PD_DLL_POWER_TYPE: Group of pd_dll_power_type
+ *	states
+ * @ETHTOOL_C33_PSE_EXT_STATE_POWER_NOT_AVAILABLE: Group of power_not_available
+ *	states
+ * @ETHTOOL_C33_PSE_EXT_STATE_SHORT_DETECTED: Group of short_detected states
+ */
+enum ethtool_c33_pse_ext_state {
+	ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION = 1,
+	ETHTOOL_C33_PSE_EXT_STATE_MR_MPS_VALID,
+	ETHTOOL_C33_PSE_EXT_STATE_MR_PSE_ENABLE,
+	ETHTOOL_C33_PSE_EXT_STATE_OPTION_DETECT_TED,
+	ETHTOOL_C33_PSE_EXT_STATE_OPTION_VPORT_LIM,
+	ETHTOOL_C33_PSE_EXT_STATE_OVLD_DETECTED,
+	ETHTOOL_C33_PSE_EXT_STATE_PD_DLL_POWER_TYPE,
+	ETHTOOL_C33_PSE_EXT_STATE_POWER_NOT_AVAILABLE,
+	ETHTOOL_C33_PSE_EXT_STATE_SHORT_DETECTED,
+};
+
+/**
+ * enum ethtool_c33_pse_ext_substate_mr_mps_valid - mr_mps_valid states
+ *      functions. IEEE 802.3-2022 33.2.4.4 Variables
+ *
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_MR_MPS_VALID_DETECTED_UNDERLOAD: Underload
+ *	state
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_MR_MPS_VALID_CONNECTION_OPEN: Port is not
+ *	connected
+ *
+ * The PSE monitors either the DC or AC Maintain Power Signature
+ * (MPS, see 33.2.9.1). This variable indicates the presence or absence of
+ * a valid MPS.
+ */
+enum ethtool_c33_pse_ext_substate_mr_mps_valid {
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_MR_MPS_VALID_DETECTED_UNDERLOAD = 1,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_MR_MPS_VALID_CONNECTION_OPEN,
+};
+
+/**
+ * enum ethtool_c33_pse_ext_substate_error_condition - error_condition states
+ *      functions. IEEE 802.3-2022 33.2.4.4 Variables
+ *
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_NON_EXISTING_PORT: Non-existing
+ *	port number
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_UNDEFINED_PORT: Undefined port
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_INTERNAL_HW_FAULT: Internal
+ *	hardware fault
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_COMM_ERROR_AFTER_FORCE_ON:
+ *	Communication error after force on
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_UNKNOWN_PORT_STATUS: Unknown
+ *	port status
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_HOST_CRASH_TURN_OFF: Host
+ *	crash turn off
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_HOST_CRASH_FORCE_SHUTDOWN:
+ *	Host crash force shutdown
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_CONFIG_CHANGE: Configuration
+ *	change
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_DETECTED_OVER_TEMP: Over
+ *	temperature detected
+ *
+ * error_condition is a variable indicating the status of
+ * implementation-specific fault conditions or optionally other system faults
+ * that prevent the PSE from meeting the specifications in Table 33–11 and that
+ * require the PSE not to source power. These error conditions are different
+ * from those monitored by the state diagrams in Figure 33–10.
+ */
+enum ethtool_c33_pse_ext_substate_error_condition {
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_NON_EXISTING_PORT = 1,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_UNDEFINED_PORT,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_INTERNAL_HW_FAULT,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_COMM_ERROR_AFTER_FORCE_ON,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_UNKNOWN_PORT_STATUS,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_HOST_CRASH_TURN_OFF,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_HOST_CRASH_FORCE_SHUTDOWN,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_CONFIG_CHANGE,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_DETECTED_OVER_TEMP,
+};
+
+/**
+ * enum ethtool_c33_pse_ext_substate_mr_pse_enable - mr_pse_enable states
+ *      functions. IEEE 802.3-2022 33.2.4.4 Variables
+ *
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_MR_PSE_ENABLE_DISABLE_PIN_ACTIVE: Disable
+ *	pin active
+ *
+ * mr_pse_enable is control variable that selects PSE operation and test
+ * functions.
+ */
+enum ethtool_c33_pse_ext_substate_mr_pse_enable {
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_MR_PSE_ENABLE_DISABLE_PIN_ACTIVE = 1,
+};
+
+/**
+ * enum ethtool_c33_pse_ext_substate_option_detect_ted - option_detect_ted
+ *	states functions. IEEE 802.3-2022 33.2.4.4 Variables
+ *
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_DETECT_TED_DET_IN_PROCESS: Detection
+ *	in process
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_DETECT_TED_CONNECTION_CHECK_ERROR:
+ *	Connection check error
+ *
+ * option_detect_ted is a variable indicating if detection can be performed
+ * by the PSE during the ted_timer interval.
+ */
+enum ethtool_c33_pse_ext_substate_option_detect_ted {
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_DETECT_TED_DET_IN_PROCESS = 1,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_DETECT_TED_CONNECTION_CHECK_ERROR,
+};
+
+/**
+ * enum ethtool_c33_pse_ext_substate_option_vport_lim - option_vport_lim states
+ *      functions. IEEE 802.3-2022 33.2.4.4 Variables
+ *
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_VPORT_LIM_HIGH_VOLTAGE: Main supply
+ *	voltage is high
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_VPORT_LIM_LOW_VOLTAGE: Main supply
+ *	voltage is low
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_VPORT_LIM_VOLTAGE_INJECTION: Voltage
+ *	injection into the port
+ *
+ * option_vport_lim is an optional variable indicates if VPSE is out of the
+ * operating range during normal operating state.
+ */
+enum ethtool_c33_pse_ext_substate_option_vport_lim {
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_VPORT_LIM_HIGH_VOLTAGE = 1,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_VPORT_LIM_LOW_VOLTAGE,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_VPORT_LIM_VOLTAGE_INJECTION,
+};
+
+/**
+ * enum ethtool_c33_pse_ext_substate_ovld_detected - ovld_detected states
+ *      functions. IEEE 802.3-2022 33.2.4.4 Variables
+ *
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_OVLD_DETECTED_OVERLOAD: Overload state
+ *
+ * ovld_detected is a variable indicating if the PSE output current has been
+ * in an overload condition (see 33.2.7.6) for at least TCUT of a one-second
+ * sliding time.
+ */
+enum ethtool_c33_pse_ext_substate_ovld_detected {
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_OVLD_DETECTED_OVERLOAD = 1,
+};
+
+/**
+ * enum ethtool_c33_pse_ext_substate_power_not_available - power_not_available
+ *	states functions. IEEE 802.3-2022 33.2.4.4 Variables
+ *
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_BUDGET_EXCEEDED: Power
+ *	budget exceeded for the controller
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_PORT_PW_LIMIT_EXCEEDS_CONTROLLER_BUDGET:
+ *	Configured port power limit exceeded controller power budget
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_PD_REQUEST_EXCEEDS_PORT_LIMIT:
+ *	Power request from PD exceeds port limit
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_HW_PW_LIMIT: Power
+ *	denied due to Hardware power limit
+ *
+ * power_not_available is a variable that is asserted in an
+ * implementation-dependent manner when the PSE is no longer capable of
+ * sourcing sufficient power to support the attached PD. Sufficient power
+ * is defined by classification; see 33.2.6.
+ */
+enum ethtool_c33_pse_ext_substate_power_not_available {
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_BUDGET_EXCEEDED =  1,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_PORT_PW_LIMIT_EXCEEDS_CONTROLLER_BUDGET,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_PD_REQUEST_EXCEEDS_PORT_LIMIT,
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_HW_PW_LIMIT,
+};
+
+/**
+ * enum ethtool_c33_pse_ext_substate_short_detected - short_detected states
+ *      functions. IEEE 802.3-2022 33.2.4.4 Variables
+ *
+ * @ETHTOOL_C33_PSE_EXT_SUBSTATE_SHORT_DETECTED_SHORT_CONDITION: Short
+ *	condition was detected
+ *
+ * short_detected is a variable indicating if the PSE output current has been
+ * in a short circuit condition for TLIM within a sliding window (see 33.2.7.7).
+ */
+enum ethtool_c33_pse_ext_substate_short_detected {
+	ETHTOOL_C33_PSE_EXT_SUBSTATE_SHORT_DETECTED_SHORT_CONDITION = 1,
+};
+
 /**
  * enum ethtool_pse_types - Types of PSE controller.
  * @ETHTOOL_PSE_UNKNOWN: Type of PSE controller is unknown
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 840dabdc9d88..b8895da001bc 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -939,6 +939,10 @@ enum {
 	ETHTOOL_A_C33_PSE_ADMIN_STATE,		/* u32 */
 	ETHTOOL_A_C33_PSE_ADMIN_CONTROL,	/* u32 */
 	ETHTOOL_A_C33_PSE_PW_D_STATUS,		/* u32 */
+	ETHTOOL_A_C33_PSE_PW_CLASS,		/* u32 */
+	ETHTOOL_A_C33_PSE_ACTUAL_PW,		/* u32 */
+	ETHTOOL_A_C33_PSE_EXT_STATE,		/* u32 */
+	ETHTOOL_A_C33_PSE_EXT_SUBSTATE,		/* u32 */
 
 	/* add new constants above here */
 	__ETHTOOL_A_PSE_CNT,
diff --git a/net/ethtool/pse-pd.c b/net/ethtool/pse-pd.c
index 2c981d443f27..d2a1c14d789f 100644
--- a/net/ethtool/pse-pd.c
+++ b/net/ethtool/pse-pd.c
@@ -86,7 +86,16 @@ static int pse_reply_size(const struct ethnl_req_info *req_base,
 		len += nla_total_size(sizeof(u32)); /* _C33_PSE_ADMIN_STATE */
 	if (st->c33_pw_status > 0)
 		len += nla_total_size(sizeof(u32)); /* _C33_PSE_PW_D_STATUS */
-
+	if (st->c33_pw_class > 0)
+		len += nla_total_size(sizeof(u32)); /* _C33_PSE_PW_CLASS */
+	if (st->c33_actual_pw > 0)
+		len += nla_total_size(sizeof(u32)); /* _C33_PSE_ACTUAL_PW */
+	if (st->c33_ext_state_info.c33_pse_ext_state > 0) {
+		len += nla_total_size(sizeof(u32)); /* _C33_PSE_EXT_STATE */
+		if (st->c33_ext_state_info.__c33_pse_ext_substate > 0)
+			/* _C33_PSE_EXT_SUBSTATE */
+			len += nla_total_size(sizeof(u32));
+	}
 	return len;
 }
 
@@ -117,6 +126,27 @@ static int pse_fill_reply(struct sk_buff *skb,
 			st->c33_pw_status))
 		return -EMSGSIZE;
 
+	if (st->c33_pw_class > 0 &&
+	    nla_put_u32(skb, ETHTOOL_A_C33_PSE_PW_CLASS,
+			st->c33_pw_class))
+		return -EMSGSIZE;
+
+	if (st->c33_actual_pw > 0 &&
+	    nla_put_u32(skb, ETHTOOL_A_C33_PSE_ACTUAL_PW,
+			st->c33_actual_pw))
+		return -EMSGSIZE;
+
+	if (st->c33_ext_state_info.c33_pse_ext_state > 0) {
+		if (nla_put_u32(skb, ETHTOOL_A_C33_PSE_EXT_STATE,
+				st->c33_ext_state_info.c33_pse_ext_state))
+			return -EMSGSIZE;
+
+		if (st->c33_ext_state_info.__c33_pse_ext_substate > 0 &&
+		    nla_put_u32(skb, ETHTOOL_A_C33_PSE_EXT_SUBSTATE,
+				st->c33_ext_state_info.__c33_pse_ext_substate))
+			return -EMSGSIZE;
+	}
+
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From c8149739af864227485abe4f79184c25a601211f Mon Sep 17 00:00:00 2001
From: "Kory Maincent (Dent Project)" <kory.maincent@bootlin.com>
Date: Thu, 4 Jul 2024 10:11:57 +0200
Subject: netlink: specs: Expand the PSE netlink command with C33 new features

Expand the c33 PSE attributes with PSE class, extended state information
and power consumption.

./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do pse-get
	     --json '{"header":{"dev-name":"eth0"}}'
{'c33-pse-actual-pw': 1700,
 'c33-pse-admin-state': 3,
 'c33-pse-pw-class': 4,
 'c33-pse-pw-d-status': 4,
 'header': {'dev-index': 4, 'dev-name': 'eth0'}}

./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do pse-get
	     --json '{"header":{"dev-name":"eth0"}}'
{'c33-pse-admin-state': 3,
 'c33-pse-ext-state': 'mr-mps-valid',
 'c33-pse-ext-substate': 2,
 'c33-pse-pw-d-status': 2,
 'header': {'dev-index': 4, 'dev-name': 'eth0'}}

Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240704-feature_poe_power_cap-v6-2-320003204264@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/ethtool.yaml | 36 ++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml
index 949e2722505d..3c7744acc6c7 100644
--- a/Documentation/netlink/specs/ethtool.yaml
+++ b/Documentation/netlink/specs/ethtool.yaml
@@ -24,6 +24,21 @@ definitions:
     name: module-fw-flash-status
     type: enum
     entries: [ started, in_progress, completed, error ]
+  -
+    name: c33-pse-ext-state
+    enum-name:
+    type: enum
+    name-prefix: ethtool-c33-pse-ext-state-
+    entries:
+        - none
+        - error-condition
+        - mr-mps-valid
+        - mr-pse-enable
+        - option-detect-ted
+        - option-vport-lim
+        - ovld-detected
+        - power-not-available
+        - short-detected
 
 attribute-sets:
   -
@@ -955,6 +970,23 @@ attribute-sets:
         name: c33-pse-pw-d-status
         type: u32
         name-prefix: ethtool-a-
+      -
+        name: c33-pse-pw-class
+        type: u32
+        name-prefix: ethtool-a-
+      -
+        name: c33-pse-actual-pw
+        type: u32
+        name-prefix: ethtool-a-
+      -
+        name: c33-pse-ext-state
+        type: u32
+        name-prefix: ethtool-a-
+        enum: c33-pse-ext-state
+      -
+        name: c33-pse-ext-substate
+        type: u32
+        name-prefix: ethtool-a-
   -
     name: rss
     attributes:
@@ -1672,6 +1704,10 @@ operations:
             - c33-pse-admin-state
             - c33-pse-admin-control
             - c33-pse-pw-d-status
+            - c33-pse-pw-class
+            - c33-pse-actual-pw
+            - c33-pse-ext-state
+            - c33-pse-ext-substate
       dump: *pse-get-op
     -
       name: pse-set
-- 
cgit v1.2.3-58-ga151


From ae37dc574259f2750e9e2666591b752678fcd7cc Mon Sep 17 00:00:00 2001
From: "Kory Maincent (Dent Project)" <kory.maincent@bootlin.com>
Date: Thu, 4 Jul 2024 10:11:58 +0200
Subject: net: pse-pd: pd692x0: Expand ethtool status message

This update expands pd692x0_ethtool_get_status() callback with newly
introduced details such as the detected class, current power delivered,
and extended state information.

Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240704-feature_poe_power_cap-v6-3-320003204264@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/pse-pd/pd692x0.c | 101 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 101 insertions(+)

diff --git a/drivers/net/pse-pd/pd692x0.c b/drivers/net/pse-pd/pd692x0.c
index 820358b71f0f..cce1b7ce044b 100644
--- a/drivers/net/pse-pd/pd692x0.c
+++ b/drivers/net/pse-pd/pd692x0.c
@@ -73,6 +73,7 @@ enum {
 	PD692X0_MSG_SET_PORT_PARAM,
 	PD692X0_MSG_GET_PORT_STATUS,
 	PD692X0_MSG_DOWNLOAD_CMD,
+	PD692X0_MSG_GET_PORT_CLASS,
 
 	/* add new message above here */
 	PD692X0_MSG_CNT
@@ -149,6 +150,12 @@ static const struct pd692x0_msg pd692x0_msg_template_list[PD692X0_MSG_CNT] = {
 		.data = {0x16, 0x16, 0x99, 0x4e,
 			 0x4e, 0x4e, 0x4e, 0x4e},
 	},
+	[PD692X0_MSG_GET_PORT_CLASS] = {
+		.key = PD692X0_KEY_REQ,
+		.sub = {0x05, 0xc4},
+		.data = {0x4e, 0x4e, 0x4e, 0x4e,
+			 0x4e, 0x4e, 0x4e, 0x4e},
+	},
 };
 
 static u8 pd692x0_build_msg(struct pd692x0_msg *msg, u8 echo)
@@ -435,6 +442,84 @@ static int pd692x0_pi_is_enabled(struct pse_controller_dev *pcdev, int id)
 	}
 }
 
+struct pd692x0_pse_ext_state_mapping {
+	u32 status_code;
+	enum ethtool_c33_pse_ext_state pse_ext_state;
+	u32 pse_ext_substate;
+};
+
+static const struct pd692x0_pse_ext_state_mapping
+pd692x0_pse_ext_state_map[] = {
+	{0x06, ETHTOOL_C33_PSE_EXT_STATE_OPTION_VPORT_LIM,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_VPORT_LIM_HIGH_VOLTAGE},
+	{0x07, ETHTOOL_C33_PSE_EXT_STATE_OPTION_VPORT_LIM,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_VPORT_LIM_LOW_VOLTAGE},
+	{0x08, ETHTOOL_C33_PSE_EXT_STATE_MR_PSE_ENABLE,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_MR_PSE_ENABLE_DISABLE_PIN_ACTIVE},
+	{0x0C, ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_NON_EXISTING_PORT},
+	{0x11, ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_UNDEFINED_PORT},
+	{0x12, ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_INTERNAL_HW_FAULT},
+	{0x1B, ETHTOOL_C33_PSE_EXT_STATE_OPTION_DETECT_TED,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_DETECT_TED_DET_IN_PROCESS},
+	{0x1C, ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_UNKNOWN_PORT_STATUS},
+	{0x1E, ETHTOOL_C33_PSE_EXT_STATE_MR_MPS_VALID,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_MR_MPS_VALID_DETECTED_UNDERLOAD},
+	{0x1F, ETHTOOL_C33_PSE_EXT_STATE_OVLD_DETECTED,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_OVLD_DETECTED_OVERLOAD},
+	{0x20, ETHTOOL_C33_PSE_EXT_STATE_POWER_NOT_AVAILABLE,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_BUDGET_EXCEEDED},
+	{0x21, ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_INTERNAL_HW_FAULT},
+	{0x22, ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_CONFIG_CHANGE},
+	{0x24, ETHTOOL_C33_PSE_EXT_STATE_OPTION_VPORT_LIM,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_VPORT_LIM_VOLTAGE_INJECTION},
+	{0x25, ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_UNKNOWN_PORT_STATUS},
+	{0x34, ETHTOOL_C33_PSE_EXT_STATE_SHORT_DETECTED,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_SHORT_DETECTED_SHORT_CONDITION},
+	{0x35, ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_DETECTED_OVER_TEMP},
+	{0x36, ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_DETECTED_OVER_TEMP},
+	{0x37, ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_UNKNOWN_PORT_STATUS},
+	{0x3C, ETHTOOL_C33_PSE_EXT_STATE_POWER_NOT_AVAILABLE,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_PORT_PW_LIMIT_EXCEEDS_CONTROLLER_BUDGET},
+	{0x3D, ETHTOOL_C33_PSE_EXT_STATE_POWER_NOT_AVAILABLE,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_PD_REQUEST_EXCEEDS_PORT_LIMIT},
+	{0x41, ETHTOOL_C33_PSE_EXT_STATE_POWER_NOT_AVAILABLE,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_POWER_NOT_AVAILABLE_HW_PW_LIMIT},
+	{0x43, ETHTOOL_C33_PSE_EXT_STATE_ERROR_CONDITION,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_ERROR_CONDITION_UNKNOWN_PORT_STATUS},
+	{0xA7, ETHTOOL_C33_PSE_EXT_STATE_OPTION_DETECT_TED,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_OPTION_DETECT_TED_CONNECTION_CHECK_ERROR},
+	{0xA8, ETHTOOL_C33_PSE_EXT_STATE_MR_MPS_VALID,
+		ETHTOOL_C33_PSE_EXT_SUBSTATE_MR_MPS_VALID_CONNECTION_OPEN},
+	{ /* sentinel */ }
+};
+
+static void
+pd692x0_get_ext_state(struct ethtool_c33_pse_ext_state_info *c33_ext_state_info,
+		      u32 status_code)
+{
+	const struct pd692x0_pse_ext_state_mapping *ext_state_map;
+
+	ext_state_map = pd692x0_pse_ext_state_map;
+	while (ext_state_map->status_code) {
+		if (ext_state_map->status_code == status_code) {
+			c33_ext_state_info->c33_pse_ext_state = ext_state_map->pse_ext_state;
+			c33_ext_state_info->__c33_pse_ext_substate = ext_state_map->pse_ext_substate;
+			return;
+		}
+		ext_state_map++;
+	}
+}
+
 static int pd692x0_ethtool_get_status(struct pse_controller_dev *pcdev,
 				      unsigned long id,
 				      struct netlink_ext_ack *extack,
@@ -442,6 +527,7 @@ static int pd692x0_ethtool_get_status(struct pse_controller_dev *pcdev,
 {
 	struct pd692x0_priv *priv = to_pd692x0_priv(pcdev);
 	struct pd692x0_msg msg, buf = {0};
+	u32 class;
 	int ret;
 
 	ret = pd692x0_fw_unavailable(priv);
@@ -471,6 +557,21 @@ static int pd692x0_ethtool_get_status(struct pse_controller_dev *pcdev,
 
 	priv->admin_state[id] = status->c33_admin_state;
 
+	pd692x0_get_ext_state(&status->c33_ext_state_info, buf.sub[0]);
+
+	status->c33_actual_pw = (buf.data[0] << 4 | buf.data[1]) * 100;
+
+	memset(&buf, 0, sizeof(buf));
+	msg = pd692x0_msg_template_list[PD692X0_MSG_GET_PORT_CLASS];
+	msg.sub[2] = id;
+	ret = pd692x0_sendrecv_msg(priv, &msg, &buf);
+	if (ret < 0)
+		return ret;
+
+	class = buf.data[3] >> 4;
+	if (class <= 8)
+		status->c33_pw_class = class;
+
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From 4a83abcef5f4f36be4e04ba18edd64226f6f4a19 Mon Sep 17 00:00:00 2001
From: "Kory Maincent (Dent Project)" <kory.maincent@bootlin.com>
Date: Thu, 4 Jul 2024 10:11:59 +0200
Subject: net: pse-pd: Add new power limit get and set c33 features

This patch add a way to get and set the power limit of a PSE PI.
For that it uses regulator API callbacks wrapper like get_voltage() and
get/set_current_limit() as power is simply V * I.
We used mW unit as defined by the IEEE 802.3-2022 standards.

set_current_limit() uses the voltage return by get_voltage() and the
desired power limit to calculate the current limit. get_voltage() callback
is then mandatory to set the power limit.

get_current_limit() callback is by default looking at a driver callback
and fallback to extracting the current limit from _pse_ethtool_get_status()
if the driver does not set its callback. We prefer let the user the choice
because ethtool_get_status return much more information than the current
limit.

expand pse status with c33_pw_limit_ranges to return the ranges available
to configure the power limit.

Reviewed-by: Sai Krishna <saikrishnag@marvell.com>
Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240704-feature_poe_power_cap-v6-4-320003204264@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/pse-pd/pse_core.c | 172 +++++++++++++++++++++++++++++++++++++++---
 include/linux/pse-pd/pse.h    |  43 +++++++++++
 2 files changed, 204 insertions(+), 11 deletions(-)

diff --git a/drivers/net/pse-pd/pse_core.c b/drivers/net/pse-pd/pse_core.c
index 795ab264eaf2..16f364b3473b 100644
--- a/drivers/net/pse-pd/pse_core.c
+++ b/drivers/net/pse-pd/pse_core.c
@@ -265,10 +265,113 @@ static int pse_pi_disable(struct regulator_dev *rdev)
 	return ret;
 }
 
+static int _pse_pi_get_voltage(struct regulator_dev *rdev)
+{
+	struct pse_controller_dev *pcdev = rdev_get_drvdata(rdev);
+	const struct pse_controller_ops *ops;
+	int id;
+
+	ops = pcdev->ops;
+	if (!ops->pi_get_voltage)
+		return -EOPNOTSUPP;
+
+	id = rdev_get_id(rdev);
+	return ops->pi_get_voltage(pcdev, id);
+}
+
+static int pse_pi_get_voltage(struct regulator_dev *rdev)
+{
+	struct pse_controller_dev *pcdev = rdev_get_drvdata(rdev);
+	int ret;
+
+	mutex_lock(&pcdev->lock);
+	ret = _pse_pi_get_voltage(rdev);
+	mutex_unlock(&pcdev->lock);
+
+	return ret;
+}
+
+static int _pse_ethtool_get_status(struct pse_controller_dev *pcdev,
+				   int id,
+				   struct netlink_ext_ack *extack,
+				   struct pse_control_status *status);
+
+static int pse_pi_get_current_limit(struct regulator_dev *rdev)
+{
+	struct pse_controller_dev *pcdev = rdev_get_drvdata(rdev);
+	const struct pse_controller_ops *ops;
+	struct netlink_ext_ack extack = {};
+	struct pse_control_status st = {};
+	int id, uV, ret;
+	s64 tmp_64;
+
+	ops = pcdev->ops;
+	id = rdev_get_id(rdev);
+	mutex_lock(&pcdev->lock);
+	if (ops->pi_get_current_limit) {
+		ret = ops->pi_get_current_limit(pcdev, id);
+		goto out;
+	}
+
+	/* If pi_get_current_limit() callback not populated get voltage
+	 * from pi_get_voltage() and power limit from ethtool_get_status()
+	 *  to calculate current limit.
+	 */
+	ret = _pse_pi_get_voltage(rdev);
+	if (!ret) {
+		dev_err(pcdev->dev, "Voltage null\n");
+		ret = -ERANGE;
+		goto out;
+	}
+	if (ret < 0)
+		goto out;
+	uV = ret;
+
+	ret = _pse_ethtool_get_status(pcdev, id, &extack, &st);
+	if (ret)
+		goto out;
+
+	if (!st.c33_avail_pw_limit) {
+		ret = -ENODATA;
+		goto out;
+	}
+
+	tmp_64 = st.c33_avail_pw_limit;
+	tmp_64 *= 1000000000ull;
+	/* uA = mW * 1000000000 / uV */
+	ret = DIV_ROUND_CLOSEST_ULL(tmp_64, uV);
+
+out:
+	mutex_unlock(&pcdev->lock);
+	return ret;
+}
+
+static int pse_pi_set_current_limit(struct regulator_dev *rdev, int min_uA,
+				    int max_uA)
+{
+	struct pse_controller_dev *pcdev = rdev_get_drvdata(rdev);
+	const struct pse_controller_ops *ops;
+	int id, ret;
+
+	ops = pcdev->ops;
+	if (!ops->pi_set_current_limit)
+		return -EOPNOTSUPP;
+
+	id = rdev_get_id(rdev);
+	mutex_lock(&pcdev->lock);
+	ret = ops->pi_set_current_limit(pcdev, id, max_uA);
+	mutex_unlock(&pcdev->lock);
+
+	return ret;
+}
+
 static const struct regulator_ops pse_pi_ops = {
 	.is_enabled = pse_pi_is_enabled,
 	.enable = pse_pi_enable,
 	.disable = pse_pi_disable,
+	.get_voltage = pse_pi_get_voltage,
+	.get_current_limit = pse_pi_get_current_limit,
+	.set_current_limit = pse_pi_set_current_limit,
 };
 
 static int
@@ -298,7 +401,9 @@ devm_pse_pi_regulator_register(struct pse_controller_dev *pcdev,
 	rdesc->ops = &pse_pi_ops;
 	rdesc->owner = pcdev->owner;
 
-	rinit_data->constraints.valid_ops_mask = REGULATOR_CHANGE_STATUS;
+	rinit_data->constraints.valid_ops_mask = REGULATOR_CHANGE_STATUS |
+						 REGULATOR_CHANGE_CURRENT;
+	rinit_data->constraints.max_uA = MAX_PI_CURRENT;
 	rinit_data->supply_regulator = "vpwr";
 
 	rconfig.dev = pcdev->dev;
@@ -626,6 +731,23 @@ out:
 }
 EXPORT_SYMBOL_GPL(of_pse_control_get);
 
+static int _pse_ethtool_get_status(struct pse_controller_dev *pcdev,
+				   int id,
+				   struct netlink_ext_ack *extack,
+				   struct pse_control_status *status)
+{
+	const struct pse_controller_ops *ops;
+
+	ops = pcdev->ops;
+	if (!ops->ethtool_get_status) {
+		NL_SET_ERR_MSG(extack,
+			       "PSE driver does not support status report");
+		return -EOPNOTSUPP;
+	}
+
+	return ops->ethtool_get_status(pcdev, id, extack, status);
+}
+
 /**
  * pse_ethtool_get_status - get status of PSE control
  * @psec: PSE control pointer
@@ -638,19 +760,10 @@ int pse_ethtool_get_status(struct pse_control *psec,
 			   struct netlink_ext_ack *extack,
 			   struct pse_control_status *status)
 {
-	const struct pse_controller_ops *ops;
 	int err;
 
-	ops = psec->pcdev->ops;
-
-	if (!ops->ethtool_get_status) {
-		NL_SET_ERR_MSG(extack,
-			       "PSE driver does not support status report");
-		return -EOPNOTSUPP;
-	}
-
 	mutex_lock(&psec->pcdev->lock);
-	err = ops->ethtool_get_status(psec->pcdev, psec->id, extack, status);
+	err = _pse_ethtool_get_status(psec->pcdev, psec->id, extack, status);
 	mutex_unlock(&psec->pcdev->lock);
 
 	return err;
@@ -732,6 +845,43 @@ int pse_ethtool_set_config(struct pse_control *psec,
 }
 EXPORT_SYMBOL_GPL(pse_ethtool_set_config);
 
+/**
+ * pse_ethtool_set_pw_limit - set PSE control power limit
+ * @psec: PSE control pointer
+ * @extack: extack for reporting useful error messages
+ * @pw_limit: power limit value in mW
+ *
+ * Return: 0 on success and failure value on error
+ */
+int pse_ethtool_set_pw_limit(struct pse_control *psec,
+			     struct netlink_ext_ack *extack,
+			     const unsigned int pw_limit)
+{
+	int uV, uA, ret;
+	s64 tmp_64;
+
+	ret = regulator_get_voltage(psec->ps);
+	if (!ret) {
+		NL_SET_ERR_MSG(extack,
+			       "Can't calculate the current, PSE voltage read is 0");
+		return -ERANGE;
+	}
+	if (ret < 0) {
+		NL_SET_ERR_MSG(extack,
+			       "Error reading PSE voltage");
+		return ret;
+	}
+	uV = ret;
+
+	tmp_64 = pw_limit;
+	tmp_64 *= 1000000000ull;
+	/* uA = mW * 1000000000 / uV */
+	uA = DIV_ROUND_CLOSEST_ULL(tmp_64, uV);
+
+	return regulator_set_current_limit(psec->ps, 0, uA);
+}
+EXPORT_SYMBOL_GPL(pse_ethtool_set_pw_limit);
+
 bool pse_has_podl(struct pse_control *psec)
 {
 	return psec->pcdev->types & ETHTOOL_PSE_PODL;
diff --git a/include/linux/pse-pd/pse.h b/include/linux/pse-pd/pse.h
index 38b9308e5e7a..591a53e082e6 100644
--- a/include/linux/pse-pd/pse.h
+++ b/include/linux/pse-pd/pse.h
@@ -9,6 +9,9 @@
 #include <linux/list.h>
 #include <uapi/linux/ethtool.h>
 
+/* Maximum current in uA according to IEEE 802.3-2022 Table 145-1 */
+#define MAX_PI_CURRENT 1920000
+
 struct phy_device;
 struct pse_controller_dev;
 
@@ -41,6 +44,12 @@ struct pse_control_config {
  * @c33_actual_pw: power currently delivered by the PSE in mW
  *	IEEE 802.3-2022 30.9.1.1.23 aPSEActualPower
  * @c33_ext_state_info: extended state information of the PSE
+ * @c33_avail_pw_limit: available power limit of the PSE in mW
+ *	IEEE 802.3-2022 145.2.5.4 pse_avail_pwr
+ * @c33_pw_limit_ranges: supported power limit configuration range. The driver
+ *	is in charge of the memory allocation.
+ * @c33_pw_limit_nb_ranges: number of supported power limit configuration
+ *	ranges
  */
 struct pse_control_status {
 	enum ethtool_podl_pse_admin_state podl_admin_state;
@@ -50,6 +59,9 @@ struct pse_control_status {
 	u32 c33_pw_class;
 	u32 c33_actual_pw;
 	struct ethtool_c33_pse_ext_state_info c33_ext_state_info;
+	u32 c33_avail_pw_limit;
+	struct ethtool_c33_pse_pw_limit_range *c33_pw_limit_ranges;
+	u32 c33_pw_limit_nb_ranges;
 };
 
 /**
@@ -61,6 +73,14 @@ struct pse_control_status {
  *		   May also return negative errno.
  * @pi_enable: Configure the PSE PI as enabled.
  * @pi_disable: Configure the PSE PI as disabled.
+ * @pi_get_voltage: Return voltage similarly to get_voltage regulator
+ *		    callback.
+ * @pi_get_current_limit: Get the configured current limit similarly to
+ *			  get_current_limit regulator callback.
+ * @pi_set_current_limit: Configure the current limit similarly to
+ *			  set_current_limit regulator callback.
+ *			  Should not return an error in case of MAX_PI_CURRENT
+ *			  current value set.
  */
 struct pse_controller_ops {
 	int (*ethtool_get_status)(struct pse_controller_dev *pcdev,
@@ -70,6 +90,11 @@ struct pse_controller_ops {
 	int (*pi_is_enabled)(struct pse_controller_dev *pcdev, int id);
 	int (*pi_enable)(struct pse_controller_dev *pcdev, int id);
 	int (*pi_disable)(struct pse_controller_dev *pcdev, int id);
+	int (*pi_get_voltage)(struct pse_controller_dev *pcdev, int id);
+	int (*pi_get_current_limit)(struct pse_controller_dev *pcdev,
+				    int id);
+	int (*pi_set_current_limit)(struct pse_controller_dev *pcdev,
+				    int id, int max_uA);
 };
 
 struct module;
@@ -156,6 +181,11 @@ int pse_ethtool_get_status(struct pse_control *psec,
 int pse_ethtool_set_config(struct pse_control *psec,
 			   struct netlink_ext_ack *extack,
 			   const struct pse_control_config *config);
+int pse_ethtool_set_pw_limit(struct pse_control *psec,
+			     struct netlink_ext_ack *extack,
+			     const unsigned int pw_limit);
+int pse_ethtool_get_pw_limit(struct pse_control *psec,
+			     struct netlink_ext_ack *extack);
 
 bool pse_has_podl(struct pse_control *psec);
 bool pse_has_c33(struct pse_control *psec);
@@ -185,6 +215,19 @@ static inline int pse_ethtool_set_config(struct pse_control *psec,
 	return -EOPNOTSUPP;
 }
 
+static inline int pse_ethtool_set_pw_limit(struct pse_control *psec,
+					   struct netlink_ext_ack *extack,
+					   const unsigned int pw_limit)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int pse_ethtool_get_pw_limit(struct pse_control *psec,
+					   struct netlink_ext_ack *extack)
+{
+	return -EOPNOTSUPP;
+}
+
 static inline bool pse_has_podl(struct pse_control *psec)
 {
 	return false;
-- 
cgit v1.2.3-58-ga151


From 30d7b6727724ce3729f2cb5b8be985d2d1931d2b Mon Sep 17 00:00:00 2001
From: "Kory Maincent (Dent Project)" <kory.maincent@bootlin.com>
Date: Thu, 4 Jul 2024 10:12:00 +0200
Subject: net: ethtool: Add new power limit get and set features

This patch expands the status information provided by ethtool for PSE c33
with available power limit and available power limit ranges. It also adds
a call to pse_ethtool_set_pw_limit() to configure the PSE control power
limit.

Reviewed-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240704-feature_poe_power_cap-v6-5-320003204264@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/ethtool-netlink.rst | 64 ++++++++++++++------
 include/linux/ethtool.h                      |  5 ++
 include/uapi/linux/ethtool_netlink.h         |  8 +++
 net/ethtool/pse-pd.c                         | 89 +++++++++++++++++++++++++---
 4 files changed, 141 insertions(+), 25 deletions(-)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 0656ad4be000..3ab423b80e91 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -1730,24 +1730,28 @@ Request contents:
 
 Kernel response contents:
 
-  ======================================  ======  =============================
-  ``ETHTOOL_A_PSE_HEADER``                nested  reply header
-  ``ETHTOOL_A_PODL_PSE_ADMIN_STATE``         u32  Operational state of the PoDL
-                                                  PSE functions
-  ``ETHTOOL_A_PODL_PSE_PW_D_STATUS``         u32  power detection status of the
-                                                  PoDL PSE.
-  ``ETHTOOL_A_C33_PSE_ADMIN_STATE``          u32  Operational state of the PoE
-                                                  PSE functions.
-  ``ETHTOOL_A_C33_PSE_PW_D_STATUS``          u32  power detection status of the
-                                                  PoE PSE.
-  ``ETHTOOL_A_C33_PSE_PW_CLASS``             u32  power class of the PoE PSE.
-  ``ETHTOOL_A_C33_PSE_ACTUAL_PW``            u32  actual power drawn on the
-                                                  PoE PSE.
-  ``ETHTOOL_A_C33_PSE_EXT_STATE``            u32  power extended state of the
-                                                  PoE PSE.
-  ``ETHTOOL_A_C33_PSE_EXT_SUBSTATE``         u32  power extended substatus of
-                                                  the PoE PSE.
-  ======================================  ======  =============================
+  ==========================================  ======  =============================
+  ``ETHTOOL_A_PSE_HEADER``                    nested  reply header
+  ``ETHTOOL_A_PODL_PSE_ADMIN_STATE``             u32  Operational state of the PoDL
+                                                      PSE functions
+  ``ETHTOOL_A_PODL_PSE_PW_D_STATUS``             u32  power detection status of the
+                                                      PoDL PSE.
+  ``ETHTOOL_A_C33_PSE_ADMIN_STATE``              u32  Operational state of the PoE
+                                                      PSE functions.
+  ``ETHTOOL_A_C33_PSE_PW_D_STATUS``              u32  power detection status of the
+                                                      PoE PSE.
+  ``ETHTOOL_A_C33_PSE_PW_CLASS``                 u32  power class of the PoE PSE.
+  ``ETHTOOL_A_C33_PSE_ACTUAL_PW``                u32  actual power drawn on the
+                                                      PoE PSE.
+  ``ETHTOOL_A_C33_PSE_EXT_STATE``                u32  power extended state of the
+                                                      PoE PSE.
+  ``ETHTOOL_A_C33_PSE_EXT_SUBSTATE``             u32  power extended substatus of
+                                                      the PoE PSE.
+  ``ETHTOOL_A_C33_PSE_AVAIL_PW_LIMIT``           u32  currently configured power
+                                                      limit of the PoE PSE.
+  ``ETHTOOL_A_C33_PSE_PW_LIMIT_RANGES``       nested  Supported power limit
+                                                      configuration ranges.
+  ==========================================  ======  =============================
 
 When set, the optional ``ETHTOOL_A_PODL_PSE_ADMIN_STATE`` attribute identifies
 the operational state of the PoDL PSE functions.  The operational state of the
@@ -1809,6 +1813,16 @@ Possible values are:
 		  ethtool_c33_pse_ext_substate_power_not_available
 		  ethtool_c33_pse_ext_substate_short_detected
 
+When set, the optional ``ETHTOOL_A_C33_PSE_AVAIL_PW_LIMIT`` attribute
+identifies the C33 PSE power limit in mW.
+
+When set the optional ``ETHTOOL_A_C33_PSE_PW_LIMIT_RANGES`` nested attribute
+identifies the C33 PSE power limit ranges through
+``ETHTOOL_A_C33_PSE_PWR_VAL_LIMIT_RANGE_MIN`` and
+``ETHTOOL_A_C33_PSE_PWR_VAL_LIMIT_RANGE_MAX``.
+If the controller works with fixed classes, the min and max values will be
+equal.
+
 PSE_SET
 =======
 
@@ -1820,6 +1834,8 @@ Request contents:
   ``ETHTOOL_A_PSE_HEADER``                nested  request header
   ``ETHTOOL_A_PODL_PSE_ADMIN_CONTROL``       u32  Control PoDL PSE Admin state
   ``ETHTOOL_A_C33_PSE_ADMIN_CONTROL``        u32  Control PSE Admin state
+  ``ETHTOOL_A_C33_PSE_AVAIL_PWR_LIMIT``      u32  Control PoE PSE available
+                                                  power limit
   ======================================  ======  =============================
 
 When set, the optional ``ETHTOOL_A_PODL_PSE_ADMIN_CONTROL`` attribute is used
@@ -1830,6 +1846,18 @@ to control PoDL PSE Admin functions. This option is implementing
 The same goes for ``ETHTOOL_A_C33_PSE_ADMIN_CONTROL`` implementing
 ``IEEE 802.3-2022`` 30.9.1.2.1 acPSEAdminControl.
 
+When set, the optional ``ETHTOOL_A_C33_PSE_AVAIL_PWR_LIMIT`` attribute is
+used to control the available power value limit for C33 PSE in milliwatts.
+This attribute corresponds  to the `pse_available_power` variable described in
+``IEEE 802.3-2022`` 33.2.4.4 Variables  and `pse_avail_pwr` in 145.2.5.4
+Variables, which are described in power classes.
+
+It was decided to use milliwatts for this interface to unify it with other
+power monitoring interfaces, which also use milliwatts, and to align with
+various existing products that document power consumption in watts rather than
+classes. If power limit configuration based on classes is needed, the
+conversion can be done in user space, for example by ethtool.
+
 RSS_GET
 =======
 
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 3e70f5d9e0bb..e213b5508da6 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -1288,4 +1288,9 @@ struct ethtool_c33_pse_ext_state_info {
 		u32 __c33_pse_ext_substate;
 	};
 };
+
+struct ethtool_c33_pse_pw_limit_range {
+	u32 min;
+	u32 max;
+};
 #endif /* _LINUX_ETHTOOL_H */
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index b8895da001bc..6d5bdcc67631 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -930,6 +930,12 @@ enum {
 };
 
 /* Power Sourcing Equipment */
+enum {
+	ETHTOOL_A_C33_PSE_PW_LIMIT_UNSPEC,
+	ETHTOOL_A_C33_PSE_PW_LIMIT_MIN,	/* u32 */
+	ETHTOOL_A_C33_PSE_PW_LIMIT_MAX,	/* u32 */
+};
+
 enum {
 	ETHTOOL_A_PSE_UNSPEC,
 	ETHTOOL_A_PSE_HEADER,			/* nest - _A_HEADER_* */
@@ -943,6 +949,8 @@ enum {
 	ETHTOOL_A_C33_PSE_ACTUAL_PW,		/* u32 */
 	ETHTOOL_A_C33_PSE_EXT_STATE,		/* u32 */
 	ETHTOOL_A_C33_PSE_EXT_SUBSTATE,		/* u32 */
+	ETHTOOL_A_C33_PSE_AVAIL_PW_LIMIT,	/* u32 */
+	ETHTOOL_A_C33_PSE_PW_LIMIT_RANGES,	/* nest - _C33_PSE_PW_LIMIT_* */
 
 	/* add new constants above here */
 	__ETHTOOL_A_PSE_CNT,
diff --git a/net/ethtool/pse-pd.c b/net/ethtool/pse-pd.c
index d2a1c14d789f..ba46c9c8b12d 100644
--- a/net/ethtool/pse-pd.c
+++ b/net/ethtool/pse-pd.c
@@ -96,9 +96,46 @@ static int pse_reply_size(const struct ethnl_req_info *req_base,
 			/* _C33_PSE_EXT_SUBSTATE */
 			len += nla_total_size(sizeof(u32));
 	}
+	if (st->c33_avail_pw_limit > 0)
+		/* _C33_AVAIL_PSE_PW_LIMIT */
+		len += nla_total_size(sizeof(u32));
+	if (st->c33_pw_limit_nb_ranges > 0)
+		/* _C33_PSE_PW_LIMIT_RANGES */
+		len += st->c33_pw_limit_nb_ranges *
+		       (nla_total_size(0) +
+			nla_total_size(sizeof(u32)) * 2);
+
 	return len;
 }
 
+static int pse_put_pw_limit_ranges(struct sk_buff *skb,
+				   const struct pse_control_status *st)
+{
+	const struct ethtool_c33_pse_pw_limit_range *pw_limit_ranges;
+	int i;
+
+	pw_limit_ranges = st->c33_pw_limit_ranges;
+	for (i = 0; i < st->c33_pw_limit_nb_ranges; i++) {
+		struct nlattr *nest;
+
+		nest = nla_nest_start(skb, ETHTOOL_A_C33_PSE_PW_LIMIT_RANGES);
+		if (!nest)
+			return -EMSGSIZE;
+
+		if (nla_put_u32(skb, ETHTOOL_A_C33_PSE_PW_LIMIT_MIN,
+				pw_limit_ranges->min) ||
+		    nla_put_u32(skb, ETHTOOL_A_C33_PSE_PW_LIMIT_MAX,
+				pw_limit_ranges->max)) {
+			nla_nest_cancel(skb, nest);
+			return -EMSGSIZE;
+		}
+		nla_nest_end(skb, nest);
+		pw_limit_ranges++;
+	}
+
+	return 0;
+}
+
 static int pse_fill_reply(struct sk_buff *skb,
 			  const struct ethnl_req_info *req_base,
 			  const struct ethnl_reply_data *reply_base)
@@ -147,9 +184,25 @@ static int pse_fill_reply(struct sk_buff *skb,
 			return -EMSGSIZE;
 	}
 
+	if (st->c33_avail_pw_limit > 0 &&
+	    nla_put_u32(skb, ETHTOOL_A_C33_PSE_AVAIL_PW_LIMIT,
+			st->c33_avail_pw_limit))
+		return -EMSGSIZE;
+
+	if (st->c33_pw_limit_nb_ranges > 0 &&
+	    pse_put_pw_limit_ranges(skb, st))
+		return -EMSGSIZE;
+
 	return 0;
 }
 
+static void pse_cleanup_data(struct ethnl_reply_data *reply_base)
+{
+	const struct pse_reply_data *data = PSE_REPDATA(reply_base);
+
+	kfree(data->status.c33_pw_limit_ranges);
+}
+
 /* PSE_SET */
 
 const struct nla_policy ethnl_pse_set_policy[ETHTOOL_A_PSE_MAX + 1] = {
@@ -160,6 +213,7 @@ const struct nla_policy ethnl_pse_set_policy[ETHTOOL_A_PSE_MAX + 1] = {
 	[ETHTOOL_A_C33_PSE_ADMIN_CONTROL] =
 		NLA_POLICY_RANGE(NLA_U32, ETHTOOL_C33_PSE_ADMIN_STATE_DISABLED,
 				 ETHTOOL_C33_PSE_ADMIN_STATE_ENABLED),
+	[ETHTOOL_A_C33_PSE_AVAIL_PW_LIMIT] = { .type = NLA_U32 },
 };
 
 static int
@@ -202,19 +256,39 @@ static int
 ethnl_set_pse(struct ethnl_req_info *req_info, struct genl_info *info)
 {
 	struct net_device *dev = req_info->dev;
-	struct pse_control_config config = {};
 	struct nlattr **tb = info->attrs;
 	struct phy_device *phydev;
+	int ret = 0;
 
 	phydev = dev->phydev;
+
+	if (tb[ETHTOOL_A_C33_PSE_AVAIL_PW_LIMIT]) {
+		unsigned int pw_limit;
+
+		pw_limit = nla_get_u32(tb[ETHTOOL_A_C33_PSE_AVAIL_PW_LIMIT]);
+		ret = pse_ethtool_set_pw_limit(phydev->psec, info->extack,
+					       pw_limit);
+		if (ret)
+			return ret;
+	}
+
 	/* These values are already validated by the ethnl_pse_set_policy */
-	if (pse_has_podl(phydev->psec))
-		config.podl_admin_control = nla_get_u32(tb[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL]);
-	if (pse_has_c33(phydev->psec))
-		config.c33_admin_control = nla_get_u32(tb[ETHTOOL_A_C33_PSE_ADMIN_CONTROL]);
+	if (tb[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL] ||
+	    tb[ETHTOOL_A_C33_PSE_ADMIN_CONTROL]) {
+		struct pse_control_config config = {};
+
+		if (pse_has_podl(phydev->psec))
+			config.podl_admin_control = nla_get_u32(tb[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL]);
+		if (pse_has_c33(phydev->psec))
+			config.c33_admin_control = nla_get_u32(tb[ETHTOOL_A_C33_PSE_ADMIN_CONTROL]);
+
+		ret = pse_ethtool_set_config(phydev->psec, info->extack,
+					     &config);
+		if (ret)
+			return ret;
+	}
 
-	/* Return errno directly - PSE has no notification */
-	return pse_ethtool_set_config(phydev->psec, info->extack, &config);
+	return ret;
 }
 
 const struct ethnl_request_ops ethnl_pse_request_ops = {
@@ -227,6 +301,7 @@ const struct ethnl_request_ops ethnl_pse_request_ops = {
 	.prepare_data		= pse_prepare_data,
 	.reply_size		= pse_reply_size,
 	.fill_reply		= pse_fill_reply,
+	.cleanup_data		= pse_cleanup_data,
 
 	.set_validate		= ethnl_set_pse_validate,
 	.set			= ethnl_set_pse,
-- 
cgit v1.2.3-58-ga151


From dac3de193095a0cb579e5d31c237e3e447e4a9b6 Mon Sep 17 00:00:00 2001
From: "Kory Maincent (Dent Project)" <kory.maincent@bootlin.com>
Date: Thu, 4 Jul 2024 10:12:01 +0200
Subject: netlink: specs: Expand the PSE netlink command with C33 pw-limit
 attributes

Expand the c33 PSE attributes with power limit to be able to set and get
the PSE Power Interface power limit.

./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do pse-get
             --json '{"header":{"dev-name":"eth1"}}'
{'c33-pse-actual-pw': 1700,
 'c33-pse-admin-state': 3,
 'c33-pse-avail-pw-limit': 97500,
 'c33-pse-pw-class': 4,
 'c33-pse-pw-d-status': 4,
 'c33-pse-pw-limit-ranges': [{'max': 18100, 'min': 15000},
                             {'max': 38000, 'min': 30000},
                             {'max': 65000, 'min': 60000},
                             {'max': 97500, 'min': 90000}],
 'header': {'dev-index': 5, 'dev-name': 'eth1'}}

./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do pse-set
             --json '{"header":{"dev-name":"eth1"},
                      "c33-pse-avail-pw-limit":19000}'
None

Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240704-feature_poe_power_cap-v6-6-320003204264@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/ethtool.yaml | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml
index 3c7744acc6c7..495e35fcfb21 100644
--- a/Documentation/netlink/specs/ethtool.yaml
+++ b/Documentation/netlink/specs/ethtool.yaml
@@ -939,6 +939,15 @@ attribute-sets:
       -
         name: power-mode
         type: u8
+  -
+    name: c33-pse-pw-limit
+    attributes:
+      -
+        name: min
+        type: u32
+      -
+        name: max
+        type: u32
   -
     name: pse
     attributes:
@@ -987,6 +996,16 @@ attribute-sets:
         name: c33-pse-ext-substate
         type: u32
         name-prefix: ethtool-a-
+      -
+        name: c33-pse-avail-pw-limit
+        type: u32
+        name-prefix: ethtool-a-
+      -
+        name: c33-pse-pw-limit-ranges
+        name-prefix: ethtool-a-
+        type: nest
+        multi-attr: true
+        nested-attributes: c33-pse-pw-limit
   -
     name: rss
     attributes:
@@ -1708,6 +1727,8 @@ operations:
             - c33-pse-actual-pw
             - c33-pse-ext-state
             - c33-pse-ext-substate
+            - c33-pse-avail-pw-limit
+            - c33-pse-pw-limit-ranges
       dump: *pse-get-op
     -
       name: pse-set
@@ -1721,6 +1742,7 @@ operations:
             - header
             - podl-pse-admin-control
             - c33-pse-admin-control
+            - c33-pse-avail-pw-limit
     -
       name: rss-get
       doc: Get RSS params.
-- 
cgit v1.2.3-58-ga151


From a87e699c9d3315f2253626c8c66e75dec1e551b7 Mon Sep 17 00:00:00 2001
From: "Kory Maincent (Dent Project)" <kory.maincent@bootlin.com>
Date: Thu, 4 Jul 2024 10:12:02 +0200
Subject: net: pse-pd: pd692x0: Enhance with new current limit and voltage read
 callbacks

This patch expands PSE callbacks with newly introduced
pi_get/set_current_limit() and pi_get_voltage() callback.
It also add the power limit ranges description in the status returned.
The only way to set ps692x0 port power limit is by configure the power
class plus a small power supplement which maximum depends on each class.

Acked-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240704-feature_poe_power_cap-v6-7-320003204264@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/pse-pd/pd692x0.c | 218 ++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 216 insertions(+), 2 deletions(-)

diff --git a/drivers/net/pse-pd/pd692x0.c b/drivers/net/pse-pd/pd692x0.c
index cce1b7ce044b..29cc76a66c13 100644
--- a/drivers/net/pse-pd/pd692x0.c
+++ b/drivers/net/pse-pd/pd692x0.c
@@ -74,6 +74,8 @@ enum {
 	PD692X0_MSG_GET_PORT_STATUS,
 	PD692X0_MSG_DOWNLOAD_CMD,
 	PD692X0_MSG_GET_PORT_CLASS,
+	PD692X0_MSG_GET_PORT_MEAS,
+	PD692X0_MSG_GET_PORT_PARAM,
 
 	/* add new message above here */
 	PD692X0_MSG_CNT
@@ -135,7 +137,7 @@ static const struct pd692x0_msg pd692x0_msg_template_list[PD692X0_MSG_CNT] = {
 	[PD692X0_MSG_SET_PORT_PARAM] = {
 		.key = PD692X0_KEY_CMD,
 		.sub = {0x05, 0xc0},
-		.data = {   0, 0xff, 0xff, 0xff,
+		.data = { 0xf, 0xff, 0xff, 0xff,
 			 0x4e, 0x4e, 0x4e, 0x4e},
 	},
 	[PD692X0_MSG_GET_PORT_STATUS] = {
@@ -156,6 +158,18 @@ static const struct pd692x0_msg pd692x0_msg_template_list[PD692X0_MSG_CNT] = {
 		.data = {0x4e, 0x4e, 0x4e, 0x4e,
 			 0x4e, 0x4e, 0x4e, 0x4e},
 	},
+	[PD692X0_MSG_GET_PORT_MEAS] = {
+		.key = PD692X0_KEY_REQ,
+		.sub = {0x05, 0xc5},
+		.data = {0x4e, 0x4e, 0x4e, 0x4e,
+			 0x4e, 0x4e, 0x4e, 0x4e},
+	},
+	[PD692X0_MSG_GET_PORT_PARAM] = {
+		.key = PD692X0_KEY_REQ,
+		.sub = {0x05, 0xc0},
+		.data = {0x4e, 0x4e, 0x4e, 0x4e,
+			 0x4e, 0x4e, 0x4e, 0x4e},
+	},
 };
 
 static u8 pd692x0_build_msg(struct pd692x0_msg *msg, u8 echo)
@@ -520,6 +534,106 @@ pd692x0_get_ext_state(struct ethtool_c33_pse_ext_state_info *c33_ext_state_info,
 	}
 }
 
+struct pd692x0_class_pw {
+	int class;
+	int class_cfg_value;
+	int class_pw;
+	int max_added_class_pw;
+};
+
+#define PD692X0_CLASS_PW_TABLE_SIZE 4
+/* 4/2 pairs class configuration power table in compliance mode.
+ * Need to be arranged in ascending order of power support.
+ */
+static const struct pd692x0_class_pw
+pd692x0_class_pw_table[PD692X0_CLASS_PW_TABLE_SIZE] = {
+	{.class = 3, .class_cfg_value = 0x3, .class_pw = 15000, .max_added_class_pw = 3100},
+	{.class = 4, .class_cfg_value = 0x2, .class_pw = 30000, .max_added_class_pw = 8000},
+	{.class = 6, .class_cfg_value = 0x1, .class_pw = 60000, .max_added_class_pw = 5000},
+	{.class = 8, .class_cfg_value = 0x0, .class_pw = 90000, .max_added_class_pw = 7500},
+};
+
+static int pd692x0_pi_get_pw_from_table(int op_mode, int added_pw)
+{
+	const struct pd692x0_class_pw *pw_table;
+	int i;
+
+	pw_table = pd692x0_class_pw_table;
+	for (i = 0; i < PD692X0_CLASS_PW_TABLE_SIZE; i++, pw_table++) {
+		if (pw_table->class_cfg_value == op_mode)
+			return pw_table->class_pw + added_pw * 100;
+	}
+
+	return -ERANGE;
+}
+
+static int pd692x0_pi_set_pw_from_table(struct device *dev,
+					struct pd692x0_msg *msg, int pw)
+{
+	const struct pd692x0_class_pw *pw_table;
+	int i;
+
+	pw_table = pd692x0_class_pw_table;
+	if (pw < pw_table->class_pw) {
+		dev_err(dev,
+			"Power limit %dmW not supported. Ranges minimal available: [%d-%d]\n",
+			pw,
+			pw_table->class_pw,
+			pw_table->class_pw + pw_table->max_added_class_pw);
+		return -ERANGE;
+	}
+
+	for (i = 0; i < PD692X0_CLASS_PW_TABLE_SIZE; i++, pw_table++) {
+		if (pw > (pw_table->class_pw + pw_table->max_added_class_pw))
+			continue;
+
+		if (pw < pw_table->class_pw) {
+			dev_err(dev,
+				"Power limit %dmW not supported. Ranges availables: [%d-%d] or [%d-%d]\n",
+				pw,
+				(pw_table - 1)->class_pw,
+				(pw_table - 1)->class_pw + (pw_table - 1)->max_added_class_pw,
+				pw_table->class_pw,
+				pw_table->class_pw + pw_table->max_added_class_pw);
+			return -ERANGE;
+		}
+
+		msg->data[2] = pw_table->class_cfg_value;
+		msg->data[3] = (pw - pw_table->class_pw) / 100;
+		return 0;
+	}
+
+	pw_table--;
+	dev_warn(dev,
+		 "Power limit %dmW not supported. Set to highest power limit %dmW\n",
+		 pw, pw_table->class_pw + pw_table->max_added_class_pw);
+	msg->data[2] = pw_table->class_cfg_value;
+	msg->data[3] = pw_table->max_added_class_pw / 100;
+	return 0;
+}
+
+static int
+pd692x0_pi_get_pw_ranges(struct pse_control_status *st)
+{
+	const struct pd692x0_class_pw *pw_table;
+	int i;
+
+	pw_table = pd692x0_class_pw_table;
+	st->c33_pw_limit_ranges = kcalloc(PD692X0_CLASS_PW_TABLE_SIZE,
+					  sizeof(struct ethtool_c33_pse_pw_limit_range),
+					  GFP_KERNEL);
+	if (!st->c33_pw_limit_ranges)
+		return -ENOMEM;
+
+	for (i = 0; i < PD692X0_CLASS_PW_TABLE_SIZE; i++, pw_table++) {
+		st->c33_pw_limit_ranges[i].min = pw_table->class_pw;
+		st->c33_pw_limit_ranges[i].max = pw_table->class_pw + pw_table->max_added_class_pw;
+	}
+
+	st->c33_pw_limit_nb_ranges = i;
+	return 0;
+}
+
 static int pd692x0_ethtool_get_status(struct pse_controller_dev *pcdev,
 				      unsigned long id,
 				      struct netlink_ext_ack *extack,
@@ -558,9 +672,20 @@ static int pd692x0_ethtool_get_status(struct pse_controller_dev *pcdev,
 	priv->admin_state[id] = status->c33_admin_state;
 
 	pd692x0_get_ext_state(&status->c33_ext_state_info, buf.sub[0]);
-
 	status->c33_actual_pw = (buf.data[0] << 4 | buf.data[1]) * 100;
 
+	msg = pd692x0_msg_template_list[PD692X0_MSG_GET_PORT_PARAM];
+	msg.sub[2] = id;
+	memset(&buf, 0, sizeof(buf));
+	ret = pd692x0_sendrecv_msg(priv, &msg, &buf);
+	if (ret < 0)
+		return ret;
+
+	ret = pd692x0_pi_get_pw_from_table(buf.data[0], buf.data[1]);
+	if (ret < 0)
+		return ret;
+	status->c33_avail_pw_limit = ret;
+
 	memset(&buf, 0, sizeof(buf));
 	msg = pd692x0_msg_template_list[PD692X0_MSG_GET_PORT_CLASS];
 	msg.sub[2] = id;
@@ -572,6 +697,10 @@ static int pd692x0_ethtool_get_status(struct pse_controller_dev *pcdev,
 	if (class <= 8)
 		status->c33_pw_class = class;
 
+	ret = pd692x0_pi_get_pw_ranges(status);
+	if (ret < 0)
+		return ret;
+
 	return 0;
 }
 
@@ -850,12 +979,97 @@ out:
 	return ret;
 }
 
+static int pd692x0_pi_get_voltage(struct pse_controller_dev *pcdev, int id)
+{
+	struct pd692x0_priv *priv = to_pd692x0_priv(pcdev);
+	struct pd692x0_msg msg, buf = {0};
+	int ret;
+
+	ret = pd692x0_fw_unavailable(priv);
+	if (ret)
+		return ret;
+
+	msg = pd692x0_msg_template_list[PD692X0_MSG_GET_PORT_MEAS];
+	msg.sub[2] = id;
+	ret = pd692x0_sendrecv_msg(priv, &msg, &buf);
+	if (ret < 0)
+		return ret;
+
+	/* Convert 0.1V unit to uV */
+	return (buf.sub[0] << 8 | buf.sub[1]) * 100000;
+}
+
+static int pd692x0_pi_get_current_limit(struct pse_controller_dev *pcdev,
+					int id)
+{
+	struct pd692x0_priv *priv = to_pd692x0_priv(pcdev);
+	struct pd692x0_msg msg, buf = {0};
+	int mW, uV, uA, ret;
+	s64 tmp_64;
+
+	msg = pd692x0_msg_template_list[PD692X0_MSG_GET_PORT_PARAM];
+	msg.sub[2] = id;
+	ret = pd692x0_sendrecv_msg(priv, &msg, &buf);
+	if (ret < 0)
+		return ret;
+
+	ret = pd692x0_pi_get_pw_from_table(buf.data[2], buf.data[3]);
+	if (ret < 0)
+		return ret;
+	mW = ret;
+
+	ret = pd692x0_pi_get_voltage(pcdev, id);
+	if (ret < 0)
+		return ret;
+	uV = ret;
+
+	tmp_64 = mW;
+	tmp_64 *= 1000000000ull;
+	/* uA = mW * 1000000000 / uV */
+	uA = DIV_ROUND_CLOSEST_ULL(tmp_64, uV);
+	return uA;
+}
+
+static int pd692x0_pi_set_current_limit(struct pse_controller_dev *pcdev,
+					int id, int max_uA)
+{
+	struct pd692x0_priv *priv = to_pd692x0_priv(pcdev);
+	struct device *dev = &priv->client->dev;
+	struct pd692x0_msg msg, buf = {0};
+	int uV, ret, mW;
+	s64 tmp_64;
+
+	ret = pd692x0_fw_unavailable(priv);
+	if (ret)
+		return ret;
+
+	ret = pd692x0_pi_get_voltage(pcdev, id);
+	if (ret < 0)
+		return ret;
+	uV = ret;
+
+	msg = pd692x0_msg_template_list[PD692X0_MSG_SET_PORT_PARAM];
+	msg.sub[2] = id;
+	tmp_64 = uV;
+	tmp_64 *= max_uA;
+	/* mW = uV * uA / 1000000000 */
+	mW = DIV_ROUND_CLOSEST_ULL(tmp_64, 1000000000);
+	ret = pd692x0_pi_set_pw_from_table(dev, &msg, mW);
+	if (ret)
+		return ret;
+
+	return pd692x0_sendrecv_msg(priv, &msg, &buf);
+}
+
 static const struct pse_controller_ops pd692x0_ops = {
 	.setup_pi_matrix = pd692x0_setup_pi_matrix,
 	.ethtool_get_status = pd692x0_ethtool_get_status,
 	.pi_enable = pd692x0_pi_enable,
 	.pi_disable = pd692x0_pi_disable,
 	.pi_is_enabled = pd692x0_pi_is_enabled,
+	.pi_get_voltage = pd692x0_pi_get_voltage,
+	.pi_get_current_limit = pd692x0_pi_get_current_limit,
+	.pi_set_current_limit = pd692x0_pi_set_current_limit,
 };
 
 #define PD692X0_FW_LINE_MAX_SZ 0xff
-- 
cgit v1.2.3-58-ga151


From 3abbd7ed8b768ff37798e81910b47c2e05156d49 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 4 Jul 2024 13:29:20 +0200
Subject: act_ct: prepare for stolen verdict coming from conntrack and nat
 engine

At this time, conntrack either returns NF_ACCEPT or NF_DROP.
To improve debuging it would be nice to be able to replace NF_DROP verdict
with NF_DROP_REASON() helper,

This helper releases the skb instantly (so drop_monitor can pinpoint
exact location) and returns NF_STOLEN.

Prepare call sites to deal with this before introducing such changes
in conntrack and nat core.

Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/act_ct.c | 31 +++++++++++++++++++++++++------
 1 file changed, 25 insertions(+), 6 deletions(-)

diff --git a/net/sched/act_ct.c b/net/sched/act_ct.c
index 2a96d9c1db65..a6b7c514a181 100644
--- a/net/sched/act_ct.c
+++ b/net/sched/act_ct.c
@@ -944,6 +944,8 @@ static int tcf_ct_act_nat(struct sk_buff *skb,
 		action |= BIT(NF_NAT_MANIP_DST);
 
 	err = nf_ct_nat(skb, ct, ctinfo, &action, range, commit);
+	if (err != NF_ACCEPT)
+		return err & NF_VERDICT_MASK;
 
 	if (action & BIT(NF_NAT_MANIP_SRC))
 		tc_skb_cb(skb)->post_ct_snat = 1;
@@ -1035,7 +1037,7 @@ TC_INDIRECT_SCOPE int tcf_ct_act(struct sk_buff *skb, const struct tc_action *a,
 		state.pf = family;
 		err = nf_conntrack_in(skb, &state);
 		if (err != NF_ACCEPT)
-			goto out_push;
+			goto nf_error;
 	}
 
 do_nat:
@@ -1047,7 +1049,7 @@ do_nat:
 
 	err = tcf_ct_act_nat(skb, ct, ctinfo, p->ct_action, &p->range, commit);
 	if (err != NF_ACCEPT)
-		goto drop;
+		goto nf_error;
 
 	if (!nf_ct_is_confirmed(ct) && commit && p->helper && !nfct_help(ct)) {
 		err = __nf_ct_try_assign_helper(ct, p->tmpl, GFP_ATOMIC);
@@ -1061,8 +1063,9 @@ do_nat:
 	}
 
 	if (nf_ct_is_confirmed(ct) ? ((!cached && !skip_add) || add_helper) : commit) {
-		if (nf_ct_helper(skb, ct, ctinfo, family) != NF_ACCEPT)
-			goto drop;
+		err = nf_ct_helper(skb, ct, ctinfo, family);
+		if (err != NF_ACCEPT)
+			goto nf_error;
 	}
 
 	if (commit) {
@@ -1075,8 +1078,9 @@ do_nat:
 		/* This will take care of sending queued events
 		 * even if the connection is already confirmed.
 		 */
-		if (nf_conntrack_confirm(skb) != NF_ACCEPT)
-			goto drop;
+		err = nf_conntrack_confirm(skb);
+		if (err != NF_ACCEPT)
+			goto nf_error;
 	}
 
 	if (!skip_add)
@@ -1100,6 +1104,21 @@ out_frag:
 drop:
 	tcf_action_inc_drop_qstats(&c->common);
 	return TC_ACT_SHOT;
+
+nf_error:
+	/* some verdicts store extra data in upper bits, such
+	 * as errno or queue number.
+	 */
+	switch (err & NF_VERDICT_MASK) {
+	case NF_DROP:
+		goto drop;
+	case NF_STOLEN:
+		tcf_action_inc_drop_qstats(&c->common);
+		return TC_ACT_CONSUMED;
+	default:
+		DEBUG_NET_WARN_ON_ONCE(1);
+		goto drop;
+	}
 }
 
 static const struct nla_policy ct_policy[TCA_CT_MAX + 1] = {
-- 
cgit v1.2.3-58-ga151


From 9199b915e9fad7f5eff6160d24ff6b38e970107d Mon Sep 17 00:00:00 2001
From: Jianbo Liu <jianbol@nvidia.com>
Date: Mon, 8 Jul 2024 09:58:11 +0300
Subject: xfrm: fix netdev reference count imbalance

In cited commit, netdev_tracker_alloc() is called for the newly
allocated xfrm state, but dev_hold() is missed, which causes netdev
reference count imbalance, because netdev_put() is called when the
state is freed in xfrm_dev_state_free(). Fix the issue by replacing
netdev_tracker_alloc() with netdev_hold().

Fixes: f8a70afafc17 ("xfrm: add TX datapath support for IPsec packet offload mode")
Signed-off-by: Jianbo Liu <jianbol@nvidia.com>
Reviewed-by: Cosmin Ratiu <cratiu@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/xfrm/xfrm_state.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 936f9348e5f6..67b2a399a48a 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -1331,8 +1331,7 @@ found:
 			xso->dev = xdo->dev;
 			xso->real_dev = xdo->real_dev;
 			xso->flags = XFRM_DEV_OFFLOAD_FLAG_ACQ;
-			netdev_tracker_alloc(xso->dev, &xso->dev_tracker,
-					     GFP_ATOMIC);
+			netdev_hold(xso->dev, &xso->dev_tracker, GFP_ATOMIC);
 			error = xso->dev->xfrmdev_ops->xdo_dev_state_add(x, NULL);
 			if (error) {
 				xso->dir = 0;
-- 
cgit v1.2.3-58-ga151


From 89a2aefe4b084686c2ffc1ee939585111ea4fc0f Mon Sep 17 00:00:00 2001
From: Jianbo Liu <jianbol@nvidia.com>
Date: Mon, 8 Jul 2024 09:58:12 +0300
Subject: xfrm: call xfrm_dev_policy_delete when kill policy

xfrm_policy_kill() is called at different places to delete xfrm
policy. It will call xfrm_pol_put(). But xfrm_dev_policy_delete() is
not called to free the policy offloaded to hardware.

The three commits cited here are to handle this issue by calling
xfrm_dev_policy_delete() outside xfrm_get_policy(). But they didn't
cover all the cases. An example, which is not handled for now, is
xfrm_policy_insert(). It is called when XFRM_MSG_UPDPOLICY request is
received. Old policy is replaced by new one, but the offloaded policy
is not deleted, so driver doesn't have the chance to release hardware
resources.

To resolve this issue for all cases, move xfrm_dev_policy_delete()
into xfrm_policy_kill(), so the offloaded policy can be deleted from
hardware when it is called, which avoids hardware resources leakage.

Fixes: 919e43fad516 ("xfrm: add an interface to offload policy")
Fixes: bf06fcf4be0f ("xfrm: add missed call to delete offloaded policies")
Fixes: 982c3aca8bac ("xfrm: delete offloaded policy")
Signed-off-by: Jianbo Liu <jianbol@nvidia.com>
Reviewed-by: Cosmin Ratiu <cratiu@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/xfrm/xfrm_policy.c | 5 ++---
 net/xfrm/xfrm_user.c   | 1 -
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 475b904fe68b..10f68d572885 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -452,6 +452,8 @@ EXPORT_SYMBOL(xfrm_policy_destroy);
 
 static void xfrm_policy_kill(struct xfrm_policy *policy)
 {
+	xfrm_dev_policy_delete(policy);
+
 	write_lock_bh(&policy->lock);
 	policy->walk.dead = 1;
 	write_unlock_bh(&policy->lock);
@@ -1850,7 +1852,6 @@ again:
 
 		__xfrm_policy_unlink(pol, dir);
 		spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
-		xfrm_dev_policy_delete(pol);
 		cnt++;
 		xfrm_audit_policy_delete(pol, 1, task_valid);
 		xfrm_policy_kill(pol);
@@ -1891,7 +1892,6 @@ again:
 
 		__xfrm_policy_unlink(pol, dir);
 		spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
-		xfrm_dev_policy_delete(pol);
 		cnt++;
 		xfrm_audit_policy_delete(pol, 1, task_valid);
 		xfrm_policy_kill(pol);
@@ -2342,7 +2342,6 @@ int xfrm_policy_delete(struct xfrm_policy *pol, int dir)
 	pol = __xfrm_policy_unlink(pol, dir);
 	spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
 	if (pol) {
-		xfrm_dev_policy_delete(pol);
 		xfrm_policy_kill(pol);
 		return 0;
 	}
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index e83c687bd64e..77355422ce82 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -2455,7 +2455,6 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh,
 					    NETLINK_CB(skb).portid);
 		}
 	} else {
-		xfrm_dev_policy_delete(xp);
 		xfrm_audit_policy_delete(xp, err ? 0 : 1, true);
 
 		if (err != 0)
-- 
cgit v1.2.3-58-ga151


From 5483cbfd863f01126c08703898a9e71d991a9bfe Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Fri, 5 Jul 2024 10:47:15 +0200
Subject: net: dsa: microchip: lan9371/2: update MAC capabilities for port 4

Set proper MAC capabilities for port 4 on LAN9371 and LAN9372 switches with
integrated 100BaseTX PHY. And introduce the is_lan937x_tx_phy() function to
reuse it where applicable.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Acked-by: Arun Ramadoss <arun.ramadoss@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/dsa/microchip/ksz_common.h   | 6 ++++++
 drivers/net/dsa/microchip/lan937x_main.c | 6 ++++--
 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/microchip/ksz_common.h b/drivers/net/dsa/microchip/ksz_common.h
index c4a4664c0385..5f0a628b9849 100644
--- a/drivers/net/dsa/microchip/ksz_common.h
+++ b/drivers/net/dsa/microchip/ksz_common.h
@@ -638,6 +638,12 @@ static inline int is_lan937x(struct ksz_device *dev)
 		dev->chip_id == LAN9374_CHIP_ID;
 }
 
+static inline bool is_lan937x_tx_phy(struct ksz_device *dev, int port)
+{
+	return (dev->chip_id == LAN9371_CHIP_ID ||
+		dev->chip_id == LAN9372_CHIP_ID) && port == KSZ_PORT_4;
+}
+
 /* STP State Defines */
 #define PORT_TX_ENABLE			BIT(2)
 #define PORT_RX_ENABLE			BIT(1)
diff --git a/drivers/net/dsa/microchip/lan937x_main.c b/drivers/net/dsa/microchip/lan937x_main.c
index 83ac33fede3f..824d9309a3d3 100644
--- a/drivers/net/dsa/microchip/lan937x_main.c
+++ b/drivers/net/dsa/microchip/lan937x_main.c
@@ -55,8 +55,7 @@ static int lan937x_vphy_ind_addr_wr(struct ksz_device *dev, int addr, int reg)
 	u16 addr_base = REG_PORT_T1_PHY_CTRL_BASE;
 	u16 temp;
 
-	if ((dev->info->chip_id == LAN9371_CHIP_ID ||
-	     dev->info->chip_id == LAN9372_CHIP_ID) && addr == KSZ_PORT_4)
+	if (is_lan937x_tx_phy(dev, addr))
 		addr_base = REG_PORT_TX_PHY_CTRL_BASE;
 
 	/* get register address based on the logical port */
@@ -324,6 +323,9 @@ void lan937x_phylink_get_caps(struct ksz_device *dev, int port,
 		/* MII/RMII/RGMII ports */
 		config->mac_capabilities |= MAC_ASYM_PAUSE | MAC_SYM_PAUSE |
 					    MAC_100HD | MAC_10 | MAC_1000FD;
+	} else if (is_lan937x_tx_phy(dev, port)) {
+		config->mac_capabilities |= MAC_ASYM_PAUSE | MAC_SYM_PAUSE |
+					    MAC_100HD | MAC_10;
 	}
 }
 
-- 
cgit v1.2.3-58-ga151


From a5912c37faf723c0812a6a2c9dd18ffbd34a233b Mon Sep 17 00:00:00 2001
From: Puranjay Mohan <puranjay@kernel.org>
Date: Mon, 8 Jul 2024 11:47:58 +0000
Subject: riscv, bpf: Optimize stack usage of trampoline

When BPF_TRAMP_F_CALL_ORIG is not set, stack space for passing arguments
on stack doesn't need to be reserved because the original function is
not called.

Only reserve space for stacked arguments when BPF_TRAMP_F_CALL_ORIG is
set.

Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Pu Lehui <pulehui@huawei.com>
Link: https://lore.kernel.org/bpf/20240708114758.64414-1-puranjay@kernel.org
---
 arch/riscv/net/bpf_jit_comp64.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c
index 685c7389ae7e..0795efdd3519 100644
--- a/arch/riscv/net/bpf_jit_comp64.c
+++ b/arch/riscv/net/bpf_jit_comp64.c
@@ -892,7 +892,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 	stack_size += 8;
 	sreg_off = stack_size;
 
-	if (nr_arg_slots - RV_MAX_REG_ARGS > 0)
+	if ((flags & BPF_TRAMP_F_CALL_ORIG) && (nr_arg_slots - RV_MAX_REG_ARGS > 0))
 		stack_size += (nr_arg_slots - RV_MAX_REG_ARGS) * 8;
 
 	stack_size = round_up(stack_size, STACK_ALIGN);
-- 
cgit v1.2.3-58-ga151


From bc239eb271e5f35fcade689c6782c962ef3704c8 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Thu, 4 Jul 2024 10:19:19 -0500
Subject: bpf: Remove unnecessary loop in task_file_seq_get_next()

After commit 0ede61d8589c ("file: convert to SLAB_TYPESAFE_BY_RCU") this
loop always iterates exactly one time.  Delete the for statement and pull
the code in a tab.

Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/bpf/ZoWJF51D4zWb6f5t@stanley.mountain
---
 kernel/bpf/task_iter.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/kernel/bpf/task_iter.c b/kernel/bpf/task_iter.c
index ec4e97c61eef..02aa9db8d796 100644
--- a/kernel/bpf/task_iter.c
+++ b/kernel/bpf/task_iter.c
@@ -261,6 +261,7 @@ task_file_seq_get_next(struct bpf_iter_seq_task_file_info *info)
 	u32 saved_tid = info->tid;
 	struct task_struct *curr_task;
 	unsigned int curr_fd = info->fd;
+	struct file *f;
 
 	/* If this function returns a non-NULL file object,
 	 * it held a reference to the task/file.
@@ -286,12 +287,8 @@ again:
 	}
 
 	rcu_read_lock();
-	for (;; curr_fd++) {
-		struct file *f;
-		f = task_lookup_next_fdget_rcu(curr_task, &curr_fd);
-		if (!f)
-			break;
-
+	f = task_lookup_next_fdget_rcu(curr_task, &curr_fd);
+	if (f) {
 		/* set info->fd */
 		info->fd = curr_fd;
 		rcu_read_unlock();
-- 
cgit v1.2.3-58-ga151


From 7ba4f43e16de351fe9821de80e15d88c884b2967 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Wed, 3 Jul 2024 02:48:47 +0200
Subject: s390/bpf: Change seen_reg to a mask

Using a mask instead of an array saves a small amount of memory and
allows marking multiple registers as seen with a simple "or". Another
positive side-effect is that it speeds up verification with jitterbug.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240703005047.40915-2-iii@linux.ibm.com
---
 arch/s390/net/bpf_jit_comp.c | 32 ++++++++++++++++----------------
 1 file changed, 16 insertions(+), 16 deletions(-)

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index ddfc0e99872e..945f2ee6511b 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -35,7 +35,7 @@
 
 struct bpf_jit {
 	u32 seen;		/* Flags to remember seen eBPF instructions */
-	u32 seen_reg[16];	/* Array to remember which registers are used */
+	u16 seen_regs;		/* Mask to remember which registers are used */
 	u32 *addrs;		/* Array with relative instruction addresses */
 	u8 *prg_buf;		/* Start of program */
 	int size;		/* Size of program and literal pool */
@@ -120,8 +120,8 @@ static inline void reg_set_seen(struct bpf_jit *jit, u32 b1)
 {
 	u32 r1 = reg2hex[b1];
 
-	if (r1 >= 6 && r1 <= 15 && !jit->seen_reg[r1])
-		jit->seen_reg[r1] = 1;
+	if (r1 >= 6 && r1 <= 15)
+		jit->seen_regs |= (1 << r1);
 }
 
 #define REG_SET_SEEN(b1)					\
@@ -129,8 +129,6 @@ static inline void reg_set_seen(struct bpf_jit *jit, u32 b1)
 	reg_set_seen(jit, b1);					\
 })
 
-#define REG_SEEN(b1) jit->seen_reg[reg2hex[(b1)]]
-
 /*
  * EMIT macros for code generation
  */
@@ -438,12 +436,12 @@ static void restore_regs(struct bpf_jit *jit, u32 rs, u32 re, u32 stack_depth)
 /*
  * Return first seen register (from start)
  */
-static int get_start(struct bpf_jit *jit, int start)
+static int get_start(u16 seen_regs, int start)
 {
 	int i;
 
 	for (i = start; i <= 15; i++) {
-		if (jit->seen_reg[i])
+		if (seen_regs & (1 << i))
 			return i;
 	}
 	return 0;
@@ -452,15 +450,15 @@ static int get_start(struct bpf_jit *jit, int start)
 /*
  * Return last seen register (from start) (gap >= 2)
  */
-static int get_end(struct bpf_jit *jit, int start)
+static int get_end(u16 seen_regs, int start)
 {
 	int i;
 
 	for (i = start; i < 15; i++) {
-		if (!jit->seen_reg[i] && !jit->seen_reg[i + 1])
+		if (!(seen_regs & (3 << i)))
 			return i - 1;
 	}
-	return jit->seen_reg[15] ? 15 : 14;
+	return (seen_regs & (1 << 15)) ? 15 : 14;
 }
 
 #define REGS_SAVE	1
@@ -469,8 +467,10 @@ static int get_end(struct bpf_jit *jit, int start)
  * Save and restore clobbered registers (6-15) on stack.
  * We save/restore registers in chunks with gap >= 2 registers.
  */
-static void save_restore_regs(struct bpf_jit *jit, int op, u32 stack_depth)
+static void save_restore_regs(struct bpf_jit *jit, int op, u32 stack_depth,
+			      u16 extra_regs)
 {
+	u16 seen_regs = jit->seen_regs | extra_regs;
 	const int last = 15, save_restore_size = 6;
 	int re = 6, rs;
 
@@ -484,10 +484,10 @@ static void save_restore_regs(struct bpf_jit *jit, int op, u32 stack_depth)
 	}
 
 	do {
-		rs = get_start(jit, re);
+		rs = get_start(seen_regs, re);
 		if (!rs)
 			break;
-		re = get_end(jit, rs + 1);
+		re = get_end(seen_regs, rs + 1);
 		if (op == REGS_SAVE)
 			save_regs(jit, rs, re);
 		else
@@ -573,7 +573,7 @@ static void bpf_jit_prologue(struct bpf_jit *jit, struct bpf_prog *fp,
 	/* Tail calls have to skip above initialization */
 	jit->tail_call_start = jit->prg;
 	/* Save registers */
-	save_restore_regs(jit, REGS_SAVE, stack_depth);
+	save_restore_regs(jit, REGS_SAVE, stack_depth, 0);
 	/* Setup literal pool */
 	if (is_first_pass(jit) || (jit->seen & SEEN_LITERAL)) {
 		if (!is_first_pass(jit) &&
@@ -649,7 +649,7 @@ static void bpf_jit_epilogue(struct bpf_jit *jit, u32 stack_depth)
 	/* Load exit code: lgr %r2,%b0 */
 	EMIT4(0xb9040000, REG_2, BPF_REG_0);
 	/* Restore registers */
-	save_restore_regs(jit, REGS_RESTORE, stack_depth);
+	save_restore_regs(jit, REGS_RESTORE, stack_depth, 0);
 	if (nospec_uses_trampoline()) {
 		jit->r14_thunk_ip = jit->prg;
 		/* Generate __s390_indirect_jump_r14 thunk */
@@ -1847,7 +1847,7 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 		/*
 		 * Restore registers before calling function
 		 */
-		save_restore_regs(jit, REGS_RESTORE, stack_depth);
+		save_restore_regs(jit, REGS_RESTORE, stack_depth, 0);
 
 		/*
 		 * goto *(prog->bpf_func + tail_call_start);
-- 
cgit v1.2.3-58-ga151


From fa7bd4b000a7ae32eb6fc049125943561e5b46f3 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Wed, 3 Jul 2024 02:48:48 +0200
Subject: s390/bpf: Implement exceptions

Implement the following three pieces required from the JIT:

- A "top-level" BPF prog (exception_boundary) must save all
  non-volatile registers, and not only the ones that it clobbers.
- A "handler" BPF prog (exception_cb) must switch stack to that of
  exception_boundary, and restore the registers that exception_boundary
  saved.
- arch_bpf_stack_walk() must unwind the stack and provide the results
  in a way that satisfies both bpf_throw() and exception_cb.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240703005047.40915-3-iii@linux.ibm.com
---
 arch/s390/net/bpf_jit_comp.c | 55 ++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 53 insertions(+), 2 deletions(-)

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 945f2ee6511b..9d440a0b729e 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -31,6 +31,7 @@
 #include <asm/nospec-branch.h>
 #include <asm/set_memory.h>
 #include <asm/text-patching.h>
+#include <asm/unwind.h>
 #include "bpf_jit.h"
 
 struct bpf_jit {
@@ -62,6 +63,8 @@ struct bpf_jit {
 #define SEEN_FUNC	BIT(2)		/* calls C functions */
 #define SEEN_STACK	(SEEN_FUNC | SEEN_MEM)
 
+#define NVREGS		0xffc0		/* %r6-%r15 */
+
 /*
  * s390 registers
  */
@@ -572,8 +575,21 @@ static void bpf_jit_prologue(struct bpf_jit *jit, struct bpf_prog *fp,
 	}
 	/* Tail calls have to skip above initialization */
 	jit->tail_call_start = jit->prg;
-	/* Save registers */
-	save_restore_regs(jit, REGS_SAVE, stack_depth, 0);
+	if (fp->aux->exception_cb) {
+		/*
+		 * Switch stack, the new address is in the 2nd parameter.
+		 *
+		 * Arrange the restoration of %r6-%r15 in the epilogue.
+		 * Do not restore them now, the prog does not need them.
+		 */
+		/* lgr %r15,%r3 */
+		EMIT4(0xb9040000, REG_15, REG_3);
+		jit->seen_regs |= NVREGS;
+	} else {
+		/* Save registers */
+		save_restore_regs(jit, REGS_SAVE, stack_depth,
+				  fp->aux->exception_boundary ? NVREGS : 0);
+	}
 	/* Setup literal pool */
 	if (is_first_pass(jit) || (jit->seen & SEEN_LITERAL)) {
 		if (!is_first_pass(jit) &&
@@ -2909,3 +2925,38 @@ bool bpf_jit_supports_insn(struct bpf_insn *insn, bool in_arena)
 	 */
 	return true;
 }
+
+bool bpf_jit_supports_exceptions(void)
+{
+	/*
+	 * Exceptions require unwinding support, which is always available,
+	 * because the kernel is always built with backchain.
+	 */
+	return true;
+}
+
+void arch_bpf_stack_walk(bool (*consume_fn)(void *, u64, u64, u64),
+			 void *cookie)
+{
+	unsigned long addr, prev_addr = 0;
+	struct unwind_state state;
+
+	unwind_for_each_frame(&state, NULL, NULL, 0) {
+		addr = unwind_get_return_address(&state);
+		if (!addr)
+			break;
+		/*
+		 * addr is a return address and state.sp is the value of %r15
+		 * at this address. exception_cb needs %r15 at entry to the
+		 * function containing addr, so take the next state.sp.
+		 *
+		 * There is no bp, and the exception_cb prog does not need one
+		 * to perform a quasi-longjmp. The common code requires a
+		 * non-zero bp, so pass sp there as well.
+		 */
+		if (prev_addr && !consume_fn(cookie, prev_addr, state.sp,
+					     state.sp))
+			break;
+		prev_addr = addr;
+	}
+}
-- 
cgit v1.2.3-58-ga151


From 02480fe8a6a6d44c16900b1d3a2a66d140d0a005 Mon Sep 17 00:00:00 2001
From: Ilya Leoshkevich <iii@linux.ibm.com>
Date: Wed, 3 Jul 2024 02:48:49 +0200
Subject: selftests/bpf: Remove exceptions tests from DENYLIST.s390x

Now that the s390x JIT supports exceptions, remove the respective tests
from the denylist.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240703005047.40915-4-iii@linux.ibm.com
---
 tools/testing/selftests/bpf/DENYLIST.s390x | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/DENYLIST.s390x b/tools/testing/selftests/bpf/DENYLIST.s390x
index cb810a98e78f..3ebd77206f98 100644
--- a/tools/testing/selftests/bpf/DENYLIST.s390x
+++ b/tools/testing/selftests/bpf/DENYLIST.s390x
@@ -1,6 +1,5 @@
 # TEMPORARY
 # Alphabetical order
-exceptions				 # JIT does not support calling kfunc bpf_throw				       (exceptions)
 get_stack_raw_tp                         # user_stack corrupted user stack                                             (no backchain userspace)
 stacktrace_build_id                      # compare_map_keys stackid_hmap vs. stackmap err -2 errno 2                   (?)
 verifier_iterating_callbacks
-- 
cgit v1.2.3-58-ga151


From 34ce9c8b8ada0c03ea8f2a99fcc7a1b297ccaa95 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 2 Jul 2024 13:35:56 +0200
Subject: wifi: nl80211: split helper function from
 nl80211_put_iface_combinations

Create a helper function that puts the data from struct
ieee80211_iface_combination to a nl80211 message.
This will be used for adding per-radio interface combination data.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://patch.msgid.link/22a0eee19dbcf98627239328bc66decd3395122c.1719919832.git-series.nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/wireless/nl80211.c | 111 ++++++++++++++++++++++++++-----------------------
 1 file changed, 59 insertions(+), 52 deletions(-)

diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 674368d028f3..deacd5f3f256 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -1631,71 +1631,78 @@ nla_put_failure:
 	return -ENOBUFS;
 }
 
-static int nl80211_put_iface_combinations(struct wiphy *wiphy,
-					  struct sk_buff *msg,
-					  bool large)
+static int nl80211_put_ifcomb_data(struct sk_buff *msg, bool large, int idx,
+				   const struct ieee80211_iface_combination *c)
 {
-	struct nlattr *nl_combis;
-	int i, j;
+	struct nlattr *nl_combi, *nl_limits;
+	int i;
 
-	nl_combis = nla_nest_start_noflag(msg,
-					  NL80211_ATTR_INTERFACE_COMBINATIONS);
-	if (!nl_combis)
+	nl_combi = nla_nest_start_noflag(msg, idx);
+	if (!nl_combi)
 		goto nla_put_failure;
 
-	for (i = 0; i < wiphy->n_iface_combinations; i++) {
-		const struct ieee80211_iface_combination *c;
-		struct nlattr *nl_combi, *nl_limits;
+	nl_limits = nla_nest_start_noflag(msg, NL80211_IFACE_COMB_LIMITS);
+	if (!nl_limits)
+		goto nla_put_failure;
 
-		c = &wiphy->iface_combinations[i];
+	for (i = 0; i < c->n_limits; i++) {
+		struct nlattr *nl_limit;
 
-		nl_combi = nla_nest_start_noflag(msg, i + 1);
-		if (!nl_combi)
+		nl_limit = nla_nest_start_noflag(msg, i + 1);
+		if (!nl_limit)
 			goto nla_put_failure;
-
-		nl_limits = nla_nest_start_noflag(msg,
-						  NL80211_IFACE_COMB_LIMITS);
-		if (!nl_limits)
+		if (nla_put_u32(msg, NL80211_IFACE_LIMIT_MAX, c->limits[i].max))
 			goto nla_put_failure;
+		if (nl80211_put_iftypes(msg, NL80211_IFACE_LIMIT_TYPES,
+					c->limits[i].types))
+			goto nla_put_failure;
+		nla_nest_end(msg, nl_limit);
+	}
 
-		for (j = 0; j < c->n_limits; j++) {
-			struct nlattr *nl_limit;
+	nla_nest_end(msg, nl_limits);
 
-			nl_limit = nla_nest_start_noflag(msg, j + 1);
-			if (!nl_limit)
-				goto nla_put_failure;
-			if (nla_put_u32(msg, NL80211_IFACE_LIMIT_MAX,
-					c->limits[j].max))
-				goto nla_put_failure;
-			if (nl80211_put_iftypes(msg, NL80211_IFACE_LIMIT_TYPES,
-						c->limits[j].types))
-				goto nla_put_failure;
-			nla_nest_end(msg, nl_limit);
-		}
+	if (c->beacon_int_infra_match &&
+	    nla_put_flag(msg, NL80211_IFACE_COMB_STA_AP_BI_MATCH))
+		goto nla_put_failure;
+	if (nla_put_u32(msg, NL80211_IFACE_COMB_NUM_CHANNELS,
+			c->num_different_channels) ||
+	    nla_put_u32(msg, NL80211_IFACE_COMB_MAXNUM,
+			c->max_interfaces))
+		goto nla_put_failure;
+	if (large &&
+	    (nla_put_u32(msg, NL80211_IFACE_COMB_RADAR_DETECT_WIDTHS,
+			c->radar_detect_widths) ||
+	     nla_put_u32(msg, NL80211_IFACE_COMB_RADAR_DETECT_REGIONS,
+			c->radar_detect_regions)))
+		goto nla_put_failure;
+	if (c->beacon_int_min_gcd &&
+	    nla_put_u32(msg, NL80211_IFACE_COMB_BI_MIN_GCD,
+			c->beacon_int_min_gcd))
+		goto nla_put_failure;
 
-		nla_nest_end(msg, nl_limits);
+	nla_nest_end(msg, nl_combi);
 
-		if (c->beacon_int_infra_match &&
-		    nla_put_flag(msg, NL80211_IFACE_COMB_STA_AP_BI_MATCH))
-			goto nla_put_failure;
-		if (nla_put_u32(msg, NL80211_IFACE_COMB_NUM_CHANNELS,
-				c->num_different_channels) ||
-		    nla_put_u32(msg, NL80211_IFACE_COMB_MAXNUM,
-				c->max_interfaces))
-			goto nla_put_failure;
-		if (large &&
-		    (nla_put_u32(msg, NL80211_IFACE_COMB_RADAR_DETECT_WIDTHS,
-				c->radar_detect_widths) ||
-		     nla_put_u32(msg, NL80211_IFACE_COMB_RADAR_DETECT_REGIONS,
-				c->radar_detect_regions)))
-			goto nla_put_failure;
-		if (c->beacon_int_min_gcd &&
-		    nla_put_u32(msg, NL80211_IFACE_COMB_BI_MIN_GCD,
-				c->beacon_int_min_gcd))
-			goto nla_put_failure;
+	return 0;
+nla_put_failure:
+	return -ENOBUFS;
+}
 
-		nla_nest_end(msg, nl_combi);
-	}
+static int nl80211_put_iface_combinations(struct wiphy *wiphy,
+					  struct sk_buff *msg,
+					  bool large)
+{
+	struct nlattr *nl_combis;
+	int i;
+
+	nl_combis = nla_nest_start_noflag(msg,
+					  NL80211_ATTR_INTERFACE_COMBINATIONS);
+	if (!nl_combis)
+		goto nla_put_failure;
+
+	for (i = 0; i < wiphy->n_iface_combinations; i++)
+		if (nl80211_put_ifcomb_data(msg, large, i + 1,
+					    &wiphy->iface_combinations[i]))
+			goto nla_put_failure;
 
 	nla_nest_end(msg, nl_combis);
 
-- 
cgit v1.2.3-58-ga151


From cedc12c5b57f7efa6dbebfb2b140e8675f5a2616 Mon Sep 17 00:00:00 2001
From: Andreas Ziegler <ziegler.andreas@siemens.com>
Date: Wed, 3 Jul 2024 10:34:36 +0200
Subject: libbpf: Add NULL checks to bpf_object__{prev_map,next_map}

In the current state, an erroneous call to
bpf_object__find_map_by_name(NULL, ...) leads to a segmentation
fault through the following call chain:

  bpf_object__find_map_by_name(obj = NULL, ...)
  -> bpf_object__for_each_map(pos, obj = NULL)
  -> bpf_object__next_map((obj = NULL), NULL)
  -> return (obj = NULL)->maps

While calling bpf_object__find_map_by_name with obj = NULL is
obviously incorrect, this should not lead to a segmentation
fault but rather be handled gracefully.

As __bpf_map__iter already handles this situation correctly, we
can delegate the check for the regular case there and only add
a check in case the prev or next parameter is NULL.

Signed-off-by: Andreas Ziegler <ziegler.andreas@siemens.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240703083436.505124-1-ziegler.andreas@siemens.com
---
 tools/lib/bpf/libbpf.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 4a28fac4908a..30f121754d83 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -10375,7 +10375,7 @@ __bpf_map__iter(const struct bpf_map *m, const struct bpf_object *obj, int i)
 struct bpf_map *
 bpf_object__next_map(const struct bpf_object *obj, const struct bpf_map *prev)
 {
-	if (prev == NULL)
+	if (prev == NULL && obj != NULL)
 		return obj->maps;
 
 	return __bpf_map__iter(prev, obj, 1);
@@ -10384,7 +10384,7 @@ bpf_object__next_map(const struct bpf_object *obj, const struct bpf_map *prev)
 struct bpf_map *
 bpf_object__prev_map(const struct bpf_object *obj, const struct bpf_map *next)
 {
-	if (next == NULL) {
+	if (next == NULL && obj != NULL) {
 		if (!obj->nr_maps)
 			return NULL;
 		return obj->maps + obj->nr_maps - 1;
-- 
cgit v1.2.3-58-ga151


From 574e609c4e6a0843a9ed53de79e00da8fb3e7437 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Thu, 4 Jul 2024 15:09:47 +0200
Subject: wifi: mac80211: clear vif drv_priv after remove_interface when
 stopping

Avoid reusing stale driver data when an interface is brought down and up
again. In order to avoid having to duplicate the memset in every single
driver, do it here.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://patch.msgid.link/20240704130947.48609-1-nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/iface.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 6d969d9f1ac9..97aee0a1a39a 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -689,8 +689,12 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, bool going_do
 
 		fallthrough;
 	default:
-		if (going_down)
-			drv_remove_interface(local, sdata);
+		if (!going_down)
+			break;
+		drv_remove_interface(local, sdata);
+
+		/* Clear private driver data to prevent reuse */
+		memset(sdata->vif.drv_priv, 0, local->hw.vif_data_size);
 	}
 
 	ieee80211_recalc_ps(local);
-- 
cgit v1.2.3-58-ga151


From b5d14b0c6716fad7f0c94ac6e1d6f60a49f985c7 Mon Sep 17 00:00:00 2001
From: En-Wei Wu <en-wei.wu@canonical.com>
Date: Fri, 5 Jul 2024 10:37:56 +0800
Subject: wifi: virt_wifi: avoid reporting connection success with wrong SSID

When user issues a connection with a different SSID than the one
virt_wifi has advertised, the __cfg80211_connect_result() will
trigger the warning: WARN_ON(bss_not_found).

The issue is because the connection code in virt_wifi does not
check the SSID from user space (it only checks the BSSID), and
virt_wifi will call cfg80211_connect_result() with WLAN_STATUS_SUCCESS
even if the SSID is different from the one virt_wifi has advertised.
Eventually cfg80211 won't be able to find the cfg80211_bss and generate
the warning.

Fixed it by checking the SSID (from user space) in the connection code.

Fixes: c7cdba31ed8b ("mac80211-next: rtnetlink wifi simulation device")
Reported-by: syzbot+d6eb9cee2885ec06f5e3@syzkaller.appspotmail.com
Signed-off-by: En-Wei Wu <en-wei.wu@canonical.com>
Link: https://patch.msgid.link/20240705023756.10954-1-en-wei.wu@canonical.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/virt_wifi.c | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/virtual/virt_wifi.c b/drivers/net/wireless/virtual/virt_wifi.c
index 6a84ec58d618..d2271a7644a0 100644
--- a/drivers/net/wireless/virtual/virt_wifi.c
+++ b/drivers/net/wireless/virtual/virt_wifi.c
@@ -136,6 +136,8 @@ static struct ieee80211_supported_band band_5ghz = {
 /* Assigned at module init. Guaranteed locally-administered and unicast. */
 static u8 fake_router_bssid[ETH_ALEN] __ro_after_init = {};
 
+#define VIRT_WIFI_SSID "VirtWifi"
+
 static void virt_wifi_inform_bss(struct wiphy *wiphy)
 {
 	u64 tsf = div_u64(ktime_get_boottime_ns(), 1000);
@@ -146,8 +148,8 @@ static void virt_wifi_inform_bss(struct wiphy *wiphy)
 		u8 ssid[8];
 	} __packed ssid = {
 		.tag = WLAN_EID_SSID,
-		.len = 8,
-		.ssid = "VirtWifi",
+		.len = strlen(VIRT_WIFI_SSID),
+		.ssid = VIRT_WIFI_SSID,
 	};
 
 	informed_bss = cfg80211_inform_bss(wiphy, &channel_5ghz,
@@ -213,6 +215,8 @@ struct virt_wifi_netdev_priv {
 	struct net_device *upperdev;
 	u32 tx_packets;
 	u32 tx_failed;
+	u32 connect_requested_ssid_len;
+	u8 connect_requested_ssid[IEEE80211_MAX_SSID_LEN];
 	u8 connect_requested_bss[ETH_ALEN];
 	bool is_up;
 	bool is_connected;
@@ -229,6 +233,12 @@ static int virt_wifi_connect(struct wiphy *wiphy, struct net_device *netdev,
 	if (priv->being_deleted || !priv->is_up)
 		return -EBUSY;
 
+	if (!sme->ssid)
+		return -EINVAL;
+
+	priv->connect_requested_ssid_len = sme->ssid_len;
+	memcpy(priv->connect_requested_ssid, sme->ssid, sme->ssid_len);
+
 	could_schedule = schedule_delayed_work(&priv->connect, HZ * 2);
 	if (!could_schedule)
 		return -EBUSY;
@@ -252,12 +262,15 @@ static void virt_wifi_connect_complete(struct work_struct *work)
 		container_of(work, struct virt_wifi_netdev_priv, connect.work);
 	u8 *requested_bss = priv->connect_requested_bss;
 	bool right_addr = ether_addr_equal(requested_bss, fake_router_bssid);
+	bool right_ssid = priv->connect_requested_ssid_len == strlen(VIRT_WIFI_SSID) &&
+			  !memcmp(priv->connect_requested_ssid, VIRT_WIFI_SSID,
+				  priv->connect_requested_ssid_len);
 	u16 status = WLAN_STATUS_SUCCESS;
 
 	if (is_zero_ether_addr(requested_bss))
 		requested_bss = NULL;
 
-	if (!priv->is_up || (requested_bss && !right_addr))
+	if (!priv->is_up || (requested_bss && !right_addr) || !right_ssid)
 		status = WLAN_STATUS_UNSPECIFIED_FAILURE;
 	else
 		priv->is_connected = true;
-- 
cgit v1.2.3-58-ga151


From 7cd445635562a38c1e3534e3e691c25d2737be54 Mon Sep 17 00:00:00 2001
From: Michael-CY Lee <michael-cy.lee@mediatek.com>
Date: Fri, 5 Jul 2024 15:43:26 +0800
Subject: wifi: mac80211: cancel color change finalize work when link is
 stopped

The color change finalize work might be called after the link is
stopped, which might lead to a kernel crash.

Signed-off-by: Michael-CY Lee <michael-cy.lee@mediatek.com>
Link: https://patch.msgid.link/20240705074326.11172-1-michael-cy.lee@mediatek.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/link.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/mac80211/link.c b/net/mac80211/link.c
index 2e6e92defbca..1a211b8d4057 100644
--- a/net/mac80211/link.c
+++ b/net/mac80211/link.c
@@ -71,6 +71,8 @@ void ieee80211_link_stop(struct ieee80211_link_data *link)
 		ieee80211_mgd_stop_link(link);
 
 	cancel_delayed_work_sync(&link->color_collision_detect_work);
+	wiphy_work_cancel(link->sdata->local->hw.wiphy,
+			  &link->color_change_finalize_work);
 	wiphy_work_cancel(link->sdata->local->hw.wiphy,
 			  &link->csa.finalize_work);
 	ieee80211_link_release_channel(link);
-- 
cgit v1.2.3-58-ga151


From 4044b23781104801f70c4a4ec3ca090730a161c5 Mon Sep 17 00:00:00 2001
From: Michael-CY Lee <michael-cy.lee@mediatek.com>
Date: Fri, 5 Jul 2024 15:43:46 +0800
Subject: wifi: mac80211: do not check BSS color collision in certain cases

Do not check BSS color collision in following cases
1. already under a color change
2. color change is disabled

Signed-off-by: Michael-CY Lee <michael-cy.lee@mediatek.com>
Link: https://patch.msgid.link/20240705074346.11228-1-michael-cy.lee@mediatek.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/rx.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 0ff9062a130c..59ad24a71141 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -3358,6 +3358,7 @@ static void
 ieee80211_rx_check_bss_color_collision(struct ieee80211_rx_data *rx)
 {
 	struct ieee80211_mgmt *mgmt = (void *)rx->skb->data;
+	struct ieee80211_bss_conf *bss_conf;
 	const struct element *ie;
 	size_t baselen;
 
@@ -3368,7 +3369,9 @@ ieee80211_rx_check_bss_color_collision(struct ieee80211_rx_data *rx)
 	if (ieee80211_hw_check(&rx->local->hw, DETECTS_COLOR_COLLISION))
 		return;
 
-	if (rx->link->conf->csa_active)
+	bss_conf = rx->link->conf;
+	if (bss_conf->csa_active || bss_conf->color_change_active ||
+	    !bss_conf->he_bss_color.enabled)
 		return;
 
 	baselen = mgmt->u.beacon.variable - rx->skb->data;
@@ -3380,7 +3383,6 @@ ieee80211_rx_check_bss_color_collision(struct ieee80211_rx_data *rx)
 				    rx->skb->len - baselen);
 	if (ie && ie->datalen >= sizeof(struct ieee80211_he_operation) &&
 	    ie->datalen >= ieee80211_he_oper_size(ie->data + 1)) {
-		struct ieee80211_bss_conf *bss_conf = rx->link->conf;
 		const struct ieee80211_he_operation *he_oper;
 		u8 color;
 
-- 
cgit v1.2.3-58-ga151


From b3603133d3b727b0b07a7094ab74cf27c8aee3bd Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 5 Jul 2024 13:39:21 +0200
Subject: wifi: mac80211_hwsim: fix warning

WARN_ON("string") will unconditionally trigger a warning, but
not really do what it may look like. Use WARN(1, ...) instead
and add the mode number as well.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://patch.msgid.link/20240705133921.a50aa5b15ece.I9a25b7448b0498c0c2e503986978dae165c8bdf8@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index 8491eb32f760..d60c780d4e8a 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -3261,7 +3261,7 @@ static int mac80211_hwsim_switch_vif_chanctx(struct ieee80211_hw *hw,
 			hwsim_clear_chanctx_magic(vifs[i].old_ctx);
 			break;
 		default:
-			WARN_ON("Invalid mode");
+			WARN(1, "Invalid mode %d\n", mode);
 		}
 	}
 	return 0;
-- 
cgit v1.2.3-58-ga151


From f56f4d541eab1ae060a46b56dd6ec9130d6e3a98 Mon Sep 17 00:00:00 2001
From: Benjamin Tissoires <bentiss@kernel.org>
Date: Mon, 8 Jul 2024 11:52:57 +0200
Subject: bpf: helpers: fix bpf_wq_set_callback_impl signature

I realized this while having a map containing both a struct bpf_timer and
a struct bpf_wq: the third argument provided to the bpf_wq callback is
not the struct bpf_wq pointer itself, but the pointer to the value in
the map.

Which means that the users need to double cast the provided "value" as
this is not a struct bpf_wq *.

This is a change of API, but there doesn't seem to be much users of bpf_wq
right now, so we should be able to go with this right now.

Fixes: 81f1d7a583fa ("bpf: wq: add bpf_wq_set_callback_impl")
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>
Link: https://lore.kernel.org/r/20240708-fix-wq-v2-1-667e5c9fbd99@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/helpers.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 229396172026..5241ba671c5a 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -2734,7 +2734,7 @@ __bpf_kfunc int bpf_wq_start(struct bpf_wq *wq, unsigned int flags)
 }
 
 __bpf_kfunc int bpf_wq_set_callback_impl(struct bpf_wq *wq,
-					 int (callback_fn)(void *map, int *key, struct bpf_wq *wq),
+					 int (callback_fn)(void *map, int *key, void *value),
 					 unsigned int flags,
 					 void *aux__ign)
 {
-- 
cgit v1.2.3-58-ga151


From 16e86f2e8199cdb8789573c8784eb5c1cd478f13 Mon Sep 17 00:00:00 2001
From: Benjamin Tissoires <bentiss@kernel.org>
Date: Mon, 8 Jul 2024 11:52:58 +0200
Subject: selftests/bpf: amend for wrong bpf_wq_set_callback_impl signature

See the previous patch: the API was wrong, we were provided the pointer
to the value, not the actual struct bpf_wq *.

Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>
Link: https://lore.kernel.org/r/20240708-fix-wq-v2-2-667e5c9fbd99@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/bpf_experimental.h  |  2 +-
 tools/testing/selftests/bpf/progs/wq.c          | 19 ++++++++++++++-----
 tools/testing/selftests/bpf/progs/wq_failures.c |  4 ++--
 3 files changed, 17 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/bpf/bpf_experimental.h b/tools/testing/selftests/bpf/bpf_experimental.h
index eede6fc2ccb4..828556cdc2f0 100644
--- a/tools/testing/selftests/bpf/bpf_experimental.h
+++ b/tools/testing/selftests/bpf/bpf_experimental.h
@@ -552,7 +552,7 @@ extern void bpf_iter_css_destroy(struct bpf_iter_css *it) __weak __ksym;
 extern int bpf_wq_init(struct bpf_wq *wq, void *p__map, unsigned int flags) __weak __ksym;
 extern int bpf_wq_start(struct bpf_wq *wq, unsigned int flags) __weak __ksym;
 extern int bpf_wq_set_callback_impl(struct bpf_wq *wq,
-		int (callback_fn)(void *map, int *key, struct bpf_wq *wq),
+		int (callback_fn)(void *map, int *key, void *value),
 		unsigned int flags__k, void *aux__ign) __ksym;
 #define bpf_wq_set_callback(timer, cb, flags) \
 	bpf_wq_set_callback_impl(timer, cb, flags, NULL)
diff --git a/tools/testing/selftests/bpf/progs/wq.c b/tools/testing/selftests/bpf/progs/wq.c
index 49e712acbf60..f8d3ae0c29ae 100644
--- a/tools/testing/selftests/bpf/progs/wq.c
+++ b/tools/testing/selftests/bpf/progs/wq.c
@@ -32,6 +32,7 @@ struct {
 } hmap_malloc SEC(".maps");
 
 struct elem {
+	int ok_offset;
 	struct bpf_wq w;
 };
 
@@ -53,7 +54,7 @@ __u32 ok;
 __u32 ok_sleepable;
 
 static int test_elem_callback(void *map, int *key,
-		int (callback_fn)(void *map, int *key, struct bpf_wq *wq))
+		int (callback_fn)(void *map, int *key, void *value))
 {
 	struct elem init = {}, *val;
 	struct bpf_wq *wq;
@@ -70,6 +71,8 @@ static int test_elem_callback(void *map, int *key,
 	if (!val)
 		return -2;
 
+	val->ok_offset = *key;
+
 	wq = &val->w;
 	if (bpf_wq_init(wq, map, 0) != 0)
 		return -3;
@@ -84,7 +87,7 @@ static int test_elem_callback(void *map, int *key,
 }
 
 static int test_hmap_elem_callback(void *map, int *key,
-		int (callback_fn)(void *map, int *key, struct bpf_wq *wq))
+		int (callback_fn)(void *map, int *key, void *value))
 {
 	struct hmap_elem init = {}, *val;
 	struct bpf_wq *wq;
@@ -114,7 +117,7 @@ static int test_hmap_elem_callback(void *map, int *key,
 }
 
 /* callback for non sleepable workqueue */
-static int wq_callback(void *map, int *key, struct bpf_wq *work)
+static int wq_callback(void *map, int *key, void *value)
 {
 	bpf_kfunc_common_test();
 	ok |= (1 << *key);
@@ -122,10 +125,16 @@ static int wq_callback(void *map, int *key, struct bpf_wq *work)
 }
 
 /* callback for sleepable workqueue */
-static int wq_cb_sleepable(void *map, int *key, struct bpf_wq *work)
+static int wq_cb_sleepable(void *map, int *key, void *value)
 {
+	struct elem *data = (struct elem *)value;
+	int offset = data->ok_offset;
+
+	if (*key != offset)
+		return 0;
+
 	bpf_kfunc_call_test_sleepable();
-	ok_sleepable |= (1 << *key);
+	ok_sleepable |= (1 << offset);
 	return 0;
 }
 
diff --git a/tools/testing/selftests/bpf/progs/wq_failures.c b/tools/testing/selftests/bpf/progs/wq_failures.c
index 4cbdb425f223..25b51a72fe0f 100644
--- a/tools/testing/selftests/bpf/progs/wq_failures.c
+++ b/tools/testing/selftests/bpf/progs/wq_failures.c
@@ -28,14 +28,14 @@ struct {
 } lru SEC(".maps");
 
 /* callback for non sleepable workqueue */
-static int wq_callback(void *map, int *key, struct bpf_wq *work)
+static int wq_callback(void *map, int *key, void *value)
 {
 	bpf_kfunc_common_test();
 	return 0;
 }
 
 /* callback for sleepable workqueue */
-static int wq_cb_sleepable(void *map, int *key, struct bpf_wq *work)
+static int wq_cb_sleepable(void *map, int *key, void *value)
 {
 	bpf_kfunc_call_test_sleepable();
 	return 0;
-- 
cgit v1.2.3-58-ga151


From 90dc946059b7d346f077b870a8d8aaf03b4d0772 Mon Sep 17 00:00:00 2001
From: Puranjay Mohan <puranjay@kernel.org>
Date: Fri, 5 Jul 2024 14:50:09 +0000
Subject: selftests/bpf: DENYLIST.aarch64: Remove fexit_sleep

fexit_sleep test runs successfully now on the BPF CI so remove it
from the deny list. ftrace direct calls was blocking tracing programs
on arm64 but it has been resolved by now. For more details see also
discussion in [*].

Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240705145009.32340-1-puranjay@kernel.org [*]
---
 tools/testing/selftests/bpf/DENYLIST.aarch64 | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/DENYLIST.aarch64 b/tools/testing/selftests/bpf/DENYLIST.aarch64
index 3c7c3e79aa93..901349da680f 100644
--- a/tools/testing/selftests/bpf/DENYLIST.aarch64
+++ b/tools/testing/selftests/bpf/DENYLIST.aarch64
@@ -1,6 +1,5 @@
 bpf_cookie/multi_kprobe_attach_api               # kprobe_multi_link_api_subtest:FAIL:fentry_raw_skel_load unexpected error: -3
 bpf_cookie/multi_kprobe_link_api                 # kprobe_multi_link_api_subtest:FAIL:fentry_raw_skel_load unexpected error: -3
-fexit_sleep                                      # The test never returns. The remaining tests cannot start.
 kprobe_multi_bench_attach                        # needs CONFIG_FPROBE
 kprobe_multi_test                                # needs CONFIG_FPROBE
 module_attach                                    # prog 'kprobe_multi': failed to auto-attach: -95
-- 
cgit v1.2.3-58-ga151


From e0ee68a8bef9cf27e324a017691ee64b235c310e Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 4 Jul 2024 18:52:22 -0700
Subject: selftests: net: ksft: interrupt cleanly on KeyboardInterrupt

It's very useful to be able to interrupt the tests during development.
Detect KeyboardInterrupt, run the cleanups and exit.

Link: https://patch.msgid.link/20240705015222.675840-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/lib/py/ksft.py | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/net/lib/py/ksft.py b/tools/testing/selftests/net/lib/py/ksft.py
index 3aaa2748a58e..f26c20df9db4 100644
--- a/tools/testing/selftests/net/lib/py/ksft.py
+++ b/tools/testing/selftests/net/lib/py/ksft.py
@@ -146,6 +146,7 @@ def ksft_run(cases=None, globs=None, case_pfx=None, args=()):
 
     global KSFT_RESULT
     cnt = 0
+    stop = False
     for case in cases:
         KSFT_RESULT = True
         cnt += 1
@@ -160,10 +161,13 @@ def ksft_run(cases=None, globs=None, case_pfx=None, args=()):
         except KsftXfailEx as e:
             comment = "XFAIL " + str(e)
             cnt_key = 'xfail'
-        except Exception as e:
+        except BaseException as e:
+            stop |= isinstance(e, KeyboardInterrupt)
             tb = traceback.format_exc()
             for line in tb.strip().split('\n'):
                 ksft_pr("Exception|", line)
+            if stop:
+                ksft_pr("Stopping tests due to KeyboardInterrupt.")
             KSFT_RESULT = False
             cnt_key = 'fail'
 
@@ -175,6 +179,9 @@ def ksft_run(cases=None, globs=None, case_pfx=None, args=()):
         ktap_result(KSFT_RESULT, cnt, case, comment=comment)
         totals[cnt_key] += 1
 
+        if stop:
+            break
+
     print(
         f"# Totals: pass:{totals['pass']} fail:{totals['fail']} xfail:{totals['xfail']} xpass:0 skip:{totals['skip']} error:0"
     )
-- 
cgit v1.2.3-58-ga151


From 946b6c48cca48591fb495508c5dbfade767173d0 Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Fri, 5 Jul 2024 13:42:06 +0200
Subject: net: page_pool: fix warning code

WARN_ON_ONCE("string") doesn't really do what appears to
be intended, so fix that.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Fixes: 90de47f020db ("page_pool: fragment API support for 32-bit arch with 64-bit DMA")
Link: https://patch.msgid.link/20240705134221.2f4de205caa1.I28496dc0f2ced580282d1fb892048017c4491e21@changeid
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/page_pool.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index dc52f078adde..855271a6cad2 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -444,7 +444,7 @@ static bool page_pool_dma_map(struct page_pool *pool, netmem_ref netmem)
 	return true;
 
 unmap_failed:
-	WARN_ON_ONCE("unexpected DMA address, please report to netdev@");
+	WARN_ONCE(1, "unexpected DMA address, please report to netdev@");
 	dma_unmap_page_attrs(pool->p.dev, dma,
 			     PAGE_SIZE << pool->p.order, pool->p.dma_dir,
 			     DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING);
-- 
cgit v1.2.3-58-ga151


From 61e9be0efbe88b5cb6be6a859167a0f3724b6305 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Wed, 3 Jul 2024 20:14:58 +0200
Subject: net: stmmac: qcom-ethqos: add support for 2.5G BASEX mode

Add support for 2.5G speed in 2500BASEX mode to the QCom ethqos driver.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20240703181500.28491-2-brgl@bgdev.pl
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
index e8a1701cdb7c..3de4140fb41e 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
@@ -665,6 +665,14 @@ static int ethqos_configure_sgmii(struct qcom_ethqos *ethqos)
 	return val;
 }
 
+static void qcom_ethqos_speed_mode_2500(struct net_device *ndev, void *data)
+{
+	struct stmmac_priv *priv = netdev_priv(ndev);
+
+	priv->plat->max_speed = 2500;
+	priv->plat->phy_interface = PHY_INTERFACE_MODE_2500BASEX;
+}
+
 static int ethqos_configure(struct qcom_ethqos *ethqos)
 {
 	return ethqos->configure_func(ethqos);
@@ -787,6 +795,9 @@ static int qcom_ethqos_probe(struct platform_device *pdev)
 	case PHY_INTERFACE_MODE_RGMII_TXID:
 		ethqos->configure_func = ethqos_configure_rgmii;
 		break;
+	case PHY_INTERFACE_MODE_2500BASEX:
+		plat_dat->speed_mode_2500 = qcom_ethqos_speed_mode_2500;
+		fallthrough;
 	case PHY_INTERFACE_MODE_SGMII:
 		ethqos->configure_func = ethqos_configure_sgmii;
 		break;
-- 
cgit v1.2.3-58-ga151


From 3c466d6537b99f801b3f68af3d8124d4312437a0 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Wed, 3 Jul 2024 20:14:59 +0200
Subject: net: stmmac: qcom-ethqos: enable SGMII loopback during DMA reset on
 sa8775p-ride-r3

On sa8775p-ride-r3 the RX clocks from the AQR115C PHY are not available at
the time of the DMA reset. We can however extract the RX clock from the
internal SERDES block. Once the link is up, we can revert to the
previous state.

The AQR115C PHY doesn't support in-band signalling so we can count on
getting the link up notification and safely reuse existing callbacks
which are already used by another HW quirk workaround which enables the
functional clock to avoid a DMA reset due to timeout.

Only enable loopback on revision 3 of the board - check the phy_mode to
make sure.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20240703181500.28491-3-brgl@bgdev.pl
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c    | 23 ++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
index 3de4140fb41e..901a3c1959fa 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
@@ -21,6 +21,7 @@
 #define RGMII_IO_MACRO_CONFIG2		0x1C
 #define RGMII_IO_MACRO_DEBUG1		0x20
 #define EMAC_SYSTEM_LOW_POWER_DEBUG	0x28
+#define EMAC_WRAPPER_SGMII_PHY_CNTRL1	0xf4
 
 /* RGMII_IO_MACRO_CONFIG fields */
 #define RGMII_CONFIG_FUNC_CLK_EN		BIT(30)
@@ -79,6 +80,9 @@
 #define ETHQOS_MAC_CTRL_SPEED_MODE		BIT(14)
 #define ETHQOS_MAC_CTRL_PORT_SEL		BIT(15)
 
+/* EMAC_WRAPPER_SGMII_PHY_CNTRL1 bits */
+#define SGMII_PHY_CNTRL1_SGMII_TX_TO_RX_LOOPBACK_EN	BIT(3)
+
 #define SGMII_10M_RX_CLK_DVDR			0x31
 
 struct ethqos_emac_por {
@@ -95,6 +99,7 @@ struct ethqos_emac_driver_data {
 	bool has_integrated_pcs;
 	u32 dma_addr_width;
 	struct dwmac4_addrs dwmac4_addrs;
+	bool needs_sgmii_loopback;
 };
 
 struct qcom_ethqos {
@@ -114,6 +119,7 @@ struct qcom_ethqos {
 	unsigned int num_por;
 	bool rgmii_config_loopback_en;
 	bool has_emac_ge_3;
+	bool needs_sgmii_loopback;
 };
 
 static int rgmii_readl(struct qcom_ethqos *ethqos, unsigned int offset)
@@ -191,8 +197,22 @@ ethqos_update_link_clk(struct qcom_ethqos *ethqos, unsigned int speed)
 	clk_set_rate(ethqos->link_clk, ethqos->link_clk_rate);
 }
 
+static void
+qcom_ethqos_set_sgmii_loopback(struct qcom_ethqos *ethqos, bool enable)
+{
+	if (!ethqos->needs_sgmii_loopback ||
+	    ethqos->phy_mode != PHY_INTERFACE_MODE_2500BASEX)
+		return;
+
+	rgmii_updatel(ethqos,
+		      SGMII_PHY_CNTRL1_SGMII_TX_TO_RX_LOOPBACK_EN,
+		      enable ? SGMII_PHY_CNTRL1_SGMII_TX_TO_RX_LOOPBACK_EN : 0,
+		      EMAC_WRAPPER_SGMII_PHY_CNTRL1);
+}
+
 static void ethqos_set_func_clk_en(struct qcom_ethqos *ethqos)
 {
+	qcom_ethqos_set_sgmii_loopback(ethqos, true);
 	rgmii_updatel(ethqos, RGMII_CONFIG_FUNC_CLK_EN,
 		      RGMII_CONFIG_FUNC_CLK_EN, RGMII_IO_MACRO_CONFIG);
 }
@@ -277,6 +297,7 @@ static const struct ethqos_emac_driver_data emac_v4_0_0_data = {
 	.has_emac_ge_3 = true,
 	.link_clk_name = "phyaux",
 	.has_integrated_pcs = true,
+	.needs_sgmii_loopback = true,
 	.dma_addr_width = 36,
 	.dwmac4_addrs = {
 		.dma_chan = 0x00008100,
@@ -682,6 +703,7 @@ static void ethqos_fix_mac_speed(void *priv, unsigned int speed, unsigned int mo
 {
 	struct qcom_ethqos *ethqos = priv;
 
+	qcom_ethqos_set_sgmii_loopback(ethqos, false);
 	ethqos->speed = speed;
 	ethqos_update_link_clk(ethqos, speed);
 	ethqos_configure(ethqos);
@@ -820,6 +842,7 @@ static int qcom_ethqos_probe(struct platform_device *pdev)
 	ethqos->num_por = data->num_por;
 	ethqos->rgmii_config_loopback_en = data->rgmii_config_loopback_en;
 	ethqos->has_emac_ge_3 = data->has_emac_ge_3;
+	ethqos->needs_sgmii_loopback = data->needs_sgmii_loopback;
 
 	ethqos->link_clk = devm_clk_get(dev, data->link_clk_name ?: "rgmii");
 	if (IS_ERR(ethqos->link_clk))
-- 
cgit v1.2.3-58-ga151


From 6e909f489191b365364e9d636dec33b5dfd4e5eb Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 9 Jul 2024 08:34:09 +0200
Subject: wifi: virt_wifi: don't use strlen() in const context

Looks like not all compilers allow strlen(constant) as
a constant, so don't do that. Instead, revert back to
defining the length as the first submission had it.

Fixes: b5d14b0c6716 ("wifi: virt_wifi: avoid reporting connection success with wrong SSID")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202407090934.NnR1TUbW-lkp@intel.com/
Closes: https://lore.kernel.org/oe-kbuild-all/202407090944.mpwLHGt9-lkp@intel.com/
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/virt_wifi.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/virtual/virt_wifi.c b/drivers/net/wireless/virtual/virt_wifi.c
index d2271a7644a0..4ee374080466 100644
--- a/drivers/net/wireless/virtual/virt_wifi.c
+++ b/drivers/net/wireless/virtual/virt_wifi.c
@@ -137,6 +137,7 @@ static struct ieee80211_supported_band band_5ghz = {
 static u8 fake_router_bssid[ETH_ALEN] __ro_after_init = {};
 
 #define VIRT_WIFI_SSID "VirtWifi"
+#define VIRT_WIFI_SSID_LEN 8
 
 static void virt_wifi_inform_bss(struct wiphy *wiphy)
 {
@@ -148,7 +149,7 @@ static void virt_wifi_inform_bss(struct wiphy *wiphy)
 		u8 ssid[8];
 	} __packed ssid = {
 		.tag = WLAN_EID_SSID,
-		.len = strlen(VIRT_WIFI_SSID),
+		.len = VIRT_WIFI_SSID_LEN,
 		.ssid = VIRT_WIFI_SSID,
 	};
 
@@ -262,7 +263,7 @@ static void virt_wifi_connect_complete(struct work_struct *work)
 		container_of(work, struct virt_wifi_netdev_priv, connect.work);
 	u8 *requested_bss = priv->connect_requested_bss;
 	bool right_addr = ether_addr_equal(requested_bss, fake_router_bssid);
-	bool right_ssid = priv->connect_requested_ssid_len == strlen(VIRT_WIFI_SSID) &&
+	bool right_ssid = priv->connect_requested_ssid_len == VIRT_WIFI_SSID_LEN &&
 			  !memcmp(priv->connect_requested_ssid, VIRT_WIFI_SSID,
 				  priv->connect_requested_ssid_len);
 	u16 status = WLAN_STATUS_SUCCESS;
-- 
cgit v1.2.3-58-ga151


From f8ad00f3fb2af98f29aacd7ceb4ecdd5ad3c9a7f Mon Sep 17 00:00:00 2001
From: James Chapman <jchapman@katalix.com>
Date: Thu, 4 Jul 2024 16:25:08 +0100
Subject: l2tp: fix possible UAF when cleaning up tunnels

syzbot reported a UAF caused by a race when the L2TP work queue closes a
tunnel at the same time as a userspace thread closes a session in that
tunnel.

Tunnel cleanup is handled by a work queue which iterates through the
sessions contained within a tunnel, and closes them in turn.

Meanwhile, a userspace thread may arbitrarily close a session via
either netlink command or by closing the pppox socket in the case of
l2tp_ppp.

The race condition may occur when l2tp_tunnel_closeall walks the list
of sessions in the tunnel and deletes each one.  Currently this is
implemented using list_for_each_safe, but because the list spinlock is
dropped in the loop body it's possible for other threads to manipulate
the list during list_for_each_safe's list walk.  This can lead to the
list iterator being corrupted, leading to list_for_each_safe spinning.
One sequence of events which may lead to this is as follows:

 * A tunnel is created, containing two sessions A and B.
 * A thread closes the tunnel, triggering tunnel cleanup via the work
   queue.
 * l2tp_tunnel_closeall runs in the context of the work queue.  It
   removes session A from the tunnel session list, then drops the list
   lock.  At this point the list_for_each_safe temporary variable is
   pointing to the other session on the list, which is session B, and
   the list can be manipulated by other threads since the list lock has
   been released.
 * Userspace closes session B, which removes the session from its parent
   tunnel via l2tp_session_delete.  Since l2tp_tunnel_closeall has
   released the tunnel list lock, l2tp_session_delete is able to call
   list_del_init on the session B list node.
 * Back on the work queue, l2tp_tunnel_closeall resumes execution and
   will now spin forever on the same list entry until the underlying
   session structure is freed, at which point UAF occurs.

The solution is to iterate over the tunnel's session list using
list_first_entry_not_null to avoid the possibility of the list
iterator pointing at a list item which may be removed during the walk.

Also, have l2tp_tunnel_closeall ref each session while it processes it
to prevent another thread from freeing it.

	cpu1				cpu2
	---				---
					pppol2tp_release()

	spin_lock_bh(&tunnel->list_lock);
	for (;;) {
		session = list_first_entry_or_null(&tunnel->session_list,
						   struct l2tp_session, list);
		if (!session)
			break;
		list_del_init(&session->list);
		spin_unlock_bh(&tunnel->list_lock);

 					l2tp_session_delete(session);

		l2tp_session_delete(session);
		spin_lock_bh(&tunnel->list_lock);
	}
	spin_unlock_bh(&tunnel->list_lock);

Calling l2tp_session_delete on the same session twice isn't a problem
per-se, but if cpu2 manages to destruct the socket and unref the
session to zero before cpu1 progresses then it would lead to UAF.

Reported-by: syzbot+b471b7c936301a59745b@syzkaller.appspotmail.com
Reported-by: syzbot+c041b4ce3a6dfd1e63e2@syzkaller.appspotmail.com
Fixes: d18d3f0a24fc ("l2tp: replace hlist with simple list for per-tunnel session list")
Signed-off-by: James Chapman <jchapman@katalix.com>
Signed-off-by: Tom Parkin <tparkin@katalix.com>
Link: https://patch.msgid.link/20240704152508.1923908-1-jchapman@katalix.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/l2tp/l2tp_core.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index a99032076e04..29dfbd70c79c 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -1289,17 +1289,20 @@ static void l2tp_session_unhash(struct l2tp_session *session)
 static void l2tp_tunnel_closeall(struct l2tp_tunnel *tunnel)
 {
 	struct l2tp_session *session;
-	struct list_head *pos;
-	struct list_head *tmp;
 
 	spin_lock_bh(&tunnel->list_lock);
 	tunnel->acpt_newsess = false;
-	list_for_each_safe(pos, tmp, &tunnel->session_list) {
-		session = list_entry(pos, struct l2tp_session, list);
+	for (;;) {
+		session = list_first_entry_or_null(&tunnel->session_list,
+						   struct l2tp_session, list);
+		if (!session)
+			break;
+		l2tp_session_inc_refcount(session);
 		list_del_init(&session->list);
 		spin_unlock_bh(&tunnel->list_lock);
 		l2tp_session_delete(session);
 		spin_lock_bh(&tunnel->list_lock);
+		l2tp_session_dec_refcount(session);
 	}
 	spin_unlock_bh(&tunnel->list_lock);
 }
-- 
cgit v1.2.3-58-ga151


From 19b815ed71aadee9a2d31b7a700ef61ae8048010 Mon Sep 17 00:00:00 2001
From: Zong-Zhe Yang <kevin_yang@realtek.com>
Date: Tue, 9 Jul 2024 15:35:31 +0800
Subject: wifi: mac80211: chanctx emulation set CHANGE_CHANNEL when in_reconfig

Chanctx emulation didn't info IEEE80211_CONF_CHANGE_CHANNEL to drivers
during ieee80211_restart_hw (ieee80211_emulate_add_chanctx). It caused
non-chanctx drivers to not stand on the correct channel after recovery.
RX then behaved abnormally. Finally, disconnection/reconnection occurred.

So, set IEEE80211_CONF_CHANGE_CHANNEL when in_reconfig.

Signed-off-by: Zong-Zhe Yang <kevin_yang@realtek.com>
Link: https://patch.msgid.link/20240709073531.30565-1-kevin_yang@realtek.com
Cc: stable@vger.kernel.org
Fixes: 0a44dfc07074 ("wifi: mac80211: simplify non-chanctx drivers")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 7578ea56c12f..85a267bdb3e3 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -148,7 +148,7 @@ static u32 ieee80211_calc_hw_conf_chan(struct ieee80211_local *local,
 	offchannel_flag ^= local->hw.conf.flags & IEEE80211_CONF_OFFCHANNEL;
 
 	/* force it also for scanning, since drivers might config differently */
-	if (offchannel_flag || local->scanning ||
+	if (offchannel_flag || local->scanning || local->in_reconfig ||
 	    !cfg80211_chandef_identical(&local->hw.conf.chandef, &chandef)) {
 		local->hw.conf.chandef = chandef;
 		changed |= IEEE80211_CONF_CHANGE_CHANNEL;
-- 
cgit v1.2.3-58-ga151


From e6c06ca8f21d1cdb444c708e385d86a54bc5fc60 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 9 Jul 2024 10:38:30 +0200
Subject: wifi: cfg80211: add support for advertising multiple radios belonging
 to a wiphy

The prerequisite for MLO support in cfg80211/mac80211 is that all the links
participating in MLO must be from the same wiphy/ieee80211_hw. To meet this
expectation, some drivers may need to group multiple discrete hardware each
acting as a link in MLO under single wiphy.

With this change, supported frequencies and interface combinations of each
individual radio are reported to user space. This allows user space to figure
out the limitations of what combination of channels can be used concurrently.

Even for non-MLO devices, this improves support for devices capable of
running on multiple channels at the same time.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://patch.msgid.link/18a88f9ce82b1c9f7c12f1672430eaf2bb0be295.1720514221.git-series.nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h       |  42 +++++++++++++++-
 include/uapi/linux/nl80211.h |  65 +++++++++++++++++++++++++
 net/wireless/nl80211.c       | 113 ++++++++++++++++++++++++++++++++++++++-----
 3 files changed, 208 insertions(+), 12 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 6f992aff74ae..a00cf80e61dc 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -5046,7 +5046,9 @@ struct ieee80211_iface_limit {
  * struct ieee80211_iface_combination - possible interface combination
  *
  * With this structure the driver can describe which interface
- * combinations it supports concurrently.
+ * combinations it supports concurrently. When set in a struct wiphy_radio,
+ * the combinations refer to combinations of interfaces currently active on
+ * that radio.
  *
  * Examples:
  *
@@ -5406,6 +5408,38 @@ struct wiphy_iftype_akm_suites {
 	int n_akm_suites;
 };
 
+/**
+ * struct wiphy_radio_freq_range - wiphy frequency range
+ * @start_freq:  start range edge frequency (kHz)
+ * @end_freq:    end range edge frequency (kHz)
+ */
+struct wiphy_radio_freq_range {
+	u32 start_freq;
+	u32 end_freq;
+};
+
+
+/**
+ * struct wiphy_radio - physical radio of a wiphy
+ * This structure describes a physical radio belonging to a wiphy.
+ * It is used to describe concurrent-channel capabilities. Only one channel
+ * can be active on the radio described by struct wiphy_radio.
+ *
+ * @freq_range: frequency range that the radio can operate on.
+ * @n_freq_range: number of elements in @freq_range
+ *
+ * @iface_combinations: Valid interface combinations array, should not
+ *	list single interface types.
+ * @n_iface_combinations: number of entries in @iface_combinations array.
+ */
+struct wiphy_radio {
+	const struct wiphy_radio_freq_range *freq_range;
+	int n_freq_range;
+
+	const struct ieee80211_iface_combination *iface_combinations;
+	int n_iface_combinations;
+};
+
 #define CFG80211_HW_TIMESTAMP_ALL_PEERS	0xffff
 
 /**
@@ -5624,6 +5658,9 @@ struct wiphy_iftype_akm_suites {
  *	A value of %CFG80211_HW_TIMESTAMP_ALL_PEERS indicates the driver
  *	supports enabling HW timestamping for all peers (i.e. no need to
  *	specify a mac address).
+ *
+ * @radio: radios belonging to this wiphy
+ * @n_radio: number of radios
  */
 struct wiphy {
 	struct mutex mtx;
@@ -5774,6 +5811,9 @@ struct wiphy {
 
 	u16 hw_timestamp_max_peers;
 
+	int n_radio;
+	const struct wiphy_radio *radio;
+
 	char priv[] __aligned(NETDEV_ALIGN);
 };
 
diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index 6ae3997061b6..f97f5adc8d51 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -2052,6 +2052,10 @@ enum nl80211_commands {
  * @NL80211_ATTR_INTERFACE_COMBINATIONS: Nested attribute listing the supported
  *	interface combinations. In each nested item, it contains attributes
  *	defined in &enum nl80211_if_combination_attrs.
+ *	If the wiphy uses multiple radios (@NL80211_ATTR_WIPHY_RADIOS is set),
+ *	this attribute contains the interface combinations of the first radio.
+ *	See @NL80211_ATTR_WIPHY_INTERFACE_COMBINATIONS for the global wiphy
+ *	combinations for the sum of all radios.
  * @NL80211_ATTR_SOFTWARE_IFTYPES: Nested attribute (just like
  *	%NL80211_ATTR_SUPPORTED_IFTYPES) containing the interface types that
  *	are managed in software: interfaces of these types aren't subject to
@@ -2856,6 +2860,14 @@ enum nl80211_commands {
  *	%NL80211_CMD_ASSOCIATE indicating the SPP A-MSDUs
  *	are used on this connection
  *
+ * @NL80211_ATTR_WIPHY_RADIOS: Nested attribute describing physical radios
+ *	belonging to this wiphy. See &enum nl80211_wiphy_radio_attrs.
+ *
+ * @NL80211_ATTR_WIPHY_INTERFACE_COMBINATIONS: Nested attribute listing the
+ *	supported interface combinations for all radios combined. In each
+ *	nested item, it contains attributes defined in
+ *	&enum nl80211_if_combination_attrs.
+ *
  * @NUM_NL80211_ATTR: total number of nl80211_attrs available
  * @NL80211_ATTR_MAX: highest attribute number currently defined
  * @__NL80211_ATTR_AFTER_LAST: internal use
@@ -3401,6 +3413,9 @@ enum nl80211_attrs {
 
 	NL80211_ATTR_ASSOC_SPP_AMSDU,
 
+	NL80211_ATTR_WIPHY_RADIOS,
+	NL80211_ATTR_WIPHY_INTERFACE_COMBINATIONS,
+
 	/* add attributes here, update the policy in nl80211.c */
 
 	__NL80211_ATTR_AFTER_LAST,
@@ -8005,4 +8020,54 @@ enum nl80211_ap_settings_flags {
 	NL80211_AP_SETTINGS_SA_QUERY_OFFLOAD_SUPPORT	= 1 << 1,
 };
 
+/**
+ * enum nl80211_wiphy_radio_attrs - wiphy radio attributes
+ *
+ * @__NL80211_WIPHY_RADIO_ATTR_INVALID: Invalid
+ *
+ * @NL80211_WIPHY_RADIO_ATTR_INDEX: Index of this radio (u32)
+ * @NL80211_WIPHY_RADIO_ATTR_FREQ_RANGE: Frequency range supported by this
+ *	radio. Attribute may be present multiple times.
+ * @NL80211_WIPHY_RADIO_ATTR_INTERFACE_COMBINATION: Supported interface
+ *	combination for this radio. Attribute may be present multiple times
+ *	and contains attributes defined in &enum nl80211_if_combination_attrs.
+ *
+ * @__NL80211_WIPHY_RADIO_ATTR_LAST: Internal
+ * @NL80211_WIPHY_RADIO_ATTR_MAX: Highest attribute
+ */
+enum nl80211_wiphy_radio_attrs {
+	__NL80211_WIPHY_RADIO_ATTR_INVALID,
+
+	NL80211_WIPHY_RADIO_ATTR_INDEX,
+	NL80211_WIPHY_RADIO_ATTR_FREQ_RANGE,
+	NL80211_WIPHY_RADIO_ATTR_INTERFACE_COMBINATION,
+
+	/* keep last */
+	__NL80211_WIPHY_RADIO_ATTR_LAST,
+	NL80211_WIPHY_RADIO_ATTR_MAX = __NL80211_WIPHY_RADIO_ATTR_LAST - 1,
+};
+
+/**
+ * enum nl80211_wiphy_radio_freq_range - wiphy radio frequency range
+ *
+ * @__NL80211_WIPHY_RADIO_FREQ_ATTR_INVALID: Invalid
+ *
+ * @NL80211_WIPHY_RADIO_FREQ_ATTR_START: Frequency range start (u32).
+ *	The unit is kHz.
+ * @NL80211_WIPHY_RADIO_FREQ_ATTR_END: Frequency range end (u32).
+ *	The unit is kHz.
+ *
+ * @__NL80211_WIPHY_RADIO_FREQ_ATTR_LAST: Internal
+ * @NL80211_WIPHY_RADIO_FREQ_ATTR_MAX: Highest attribute
+ */
+enum nl80211_wiphy_radio_freq_range {
+	__NL80211_WIPHY_RADIO_FREQ_ATTR_INVALID,
+
+	NL80211_WIPHY_RADIO_FREQ_ATTR_START,
+	NL80211_WIPHY_RADIO_FREQ_ATTR_END,
+
+	__NL80211_WIPHY_RADIO_FREQ_ATTR_LAST,
+	NL80211_WIPHY_RADIO_FREQ_ATTR_MAX = __NL80211_WIPHY_RADIO_FREQ_ATTR_LAST - 1,
+};
+
 #endif /* __LINUX_NL80211_H */
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index deacd5f3f256..7397a372c78e 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -1632,16 +1632,18 @@ nla_put_failure:
 }
 
 static int nl80211_put_ifcomb_data(struct sk_buff *msg, bool large, int idx,
-				   const struct ieee80211_iface_combination *c)
+				   const struct ieee80211_iface_combination *c,
+				   u16 nested)
 {
 	struct nlattr *nl_combi, *nl_limits;
 	int i;
 
-	nl_combi = nla_nest_start_noflag(msg, idx);
+	nl_combi = nla_nest_start_noflag(msg, idx | nested);
 	if (!nl_combi)
 		goto nla_put_failure;
 
-	nl_limits = nla_nest_start_noflag(msg, NL80211_IFACE_COMB_LIMITS);
+	nl_limits = nla_nest_start_noflag(msg, NL80211_IFACE_COMB_LIMITS |
+					       nested);
 	if (!nl_limits)
 		goto nla_put_failure;
 
@@ -1689,19 +1691,26 @@ nla_put_failure:
 
 static int nl80211_put_iface_combinations(struct wiphy *wiphy,
 					  struct sk_buff *msg,
-					  bool large)
+					  int attr, int radio,
+					  bool large, u16 nested)
 {
+	const struct ieee80211_iface_combination *c;
 	struct nlattr *nl_combis;
-	int i;
+	int i, n;
 
-	nl_combis = nla_nest_start_noflag(msg,
-					  NL80211_ATTR_INTERFACE_COMBINATIONS);
+	nl_combis = nla_nest_start_noflag(msg, attr | nested);
 	if (!nl_combis)
 		goto nla_put_failure;
 
-	for (i = 0; i < wiphy->n_iface_combinations; i++)
-		if (nl80211_put_ifcomb_data(msg, large, i + 1,
-					    &wiphy->iface_combinations[i]))
+	if (radio >= 0) {
+		c = wiphy->radio[0].iface_combinations;
+		n = wiphy->radio[0].n_iface_combinations;
+	} else {
+		c = wiphy->iface_combinations;
+		n = wiphy->n_iface_combinations;
+	}
+	for (i = 0; i < n; i++)
+		if (nl80211_put_ifcomb_data(msg, large, i + 1, &c[i], nested))
 			goto nla_put_failure;
 
 	nla_nest_end(msg, nl_combis);
@@ -2408,6 +2417,80 @@ fail:
 	return -ENOBUFS;
 }
 
+static int nl80211_put_radio(struct wiphy *wiphy, struct sk_buff *msg, int idx)
+{
+	const struct wiphy_radio *r = &wiphy->radio[idx];
+	struct nlattr *radio, *freq;
+	int i;
+
+	radio = nla_nest_start(msg, idx);
+	if (!radio)
+		return -ENOBUFS;
+
+	if (nla_put_u32(msg, NL80211_WIPHY_RADIO_ATTR_INDEX, idx))
+		goto nla_put_failure;
+
+	for (i = 0; i < r->n_freq_range; i++) {
+		const struct wiphy_radio_freq_range *range = &r->freq_range[i];
+
+		freq = nla_nest_start(msg, NL80211_WIPHY_RADIO_ATTR_FREQ_RANGE);
+		if (!freq)
+			goto nla_put_failure;
+
+		if (nla_put_u32(msg, NL80211_WIPHY_RADIO_FREQ_ATTR_START,
+				range->start_freq) ||
+		    nla_put_u32(msg, NL80211_WIPHY_RADIO_FREQ_ATTR_END,
+				range->end_freq))
+			goto nla_put_failure;
+
+		nla_nest_end(msg, freq);
+	}
+
+	for (i = 0; i < r->n_iface_combinations; i++)
+		if (nl80211_put_ifcomb_data(msg, true,
+					    NL80211_WIPHY_RADIO_ATTR_INTERFACE_COMBINATION,
+					    &r->iface_combinations[i],
+					    NLA_F_NESTED))
+			goto nla_put_failure;
+
+	nla_nest_end(msg, radio);
+
+	return 0;
+
+nla_put_failure:
+	return -ENOBUFS;
+}
+
+static int nl80211_put_radios(struct wiphy *wiphy, struct sk_buff *msg)
+{
+	struct nlattr *radios;
+	int i;
+
+	if (!wiphy->n_radio)
+		return 0;
+
+	radios = nla_nest_start(msg, NL80211_ATTR_WIPHY_RADIOS);
+	if (!radios)
+		return -ENOBUFS;
+
+	for (i = 0; i < wiphy->n_radio; i++)
+		if (nl80211_put_radio(wiphy, msg, i))
+			goto fail;
+
+	nla_nest_end(msg, radios);
+
+	if (nl80211_put_iface_combinations(wiphy, msg,
+					   NL80211_ATTR_WIPHY_INTERFACE_COMBINATIONS,
+					   -1, true, NLA_F_NESTED))
+		return -ENOBUFS;
+
+	return 0;
+
+fail:
+	nla_nest_cancel(msg, radios);
+	return -ENOBUFS;
+}
+
 struct nl80211_dump_wiphy_state {
 	s64 filter_wiphy;
 	long start;
@@ -2703,7 +2786,9 @@ static int nl80211_send_wiphy(struct cfg80211_registered_device *rdev,
 			goto nla_put_failure;
 
 		if (nl80211_put_iface_combinations(&rdev->wiphy, msg,
-						   state->split))
+						   NL80211_ATTR_INTERFACE_COMBINATIONS,
+						   rdev->wiphy.n_radio ? 0 : -1,
+						   state->split, 0))
 			goto nla_put_failure;
 
 		state->split_start++;
@@ -3017,6 +3102,12 @@ static int nl80211_send_wiphy(struct cfg80211_registered_device *rdev,
 				rdev->wiphy.hw_timestamp_max_peers))
 			goto nla_put_failure;
 
+		state->split_start++;
+		break;
+	case 17:
+		if (nl80211_put_radios(&rdev->wiphy, msg))
+			goto nla_put_failure;
+
 		/* done */
 		state->split_start = 0;
 		break;
-- 
cgit v1.2.3-58-ga151


From abb4cfe3661aa05426916b21164f88ca5a405a3a Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 9 Jul 2024 10:38:31 +0200
Subject: wifi: cfg80211: extend interface combination check for multi-radio

Add a field in struct iface_combination_params to check per-radio
interface combinations instead of per-wiphy ones.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://patch.msgid.link/32b28da89c2d759b0324deeefe2be4cee91de18e.1720514221.git-series.nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h  |  5 +++++
 net/mac80211/util.c     |  5 ++++-
 net/wireless/rdev-ops.h | 12 ++++++++++++
 net/wireless/util.c     | 33 ++++++++++++++++++++++++++-------
 4 files changed, 47 insertions(+), 8 deletions(-)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index a00cf80e61dc..4767e2c76b01 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -1598,6 +1598,7 @@ struct cfg80211_color_change_settings {
  *
  * Used to pass interface combination parameters
  *
+ * @radio_idx: wiphy radio index or -1 for global
  * @num_different_channels: the number of different channels we want
  *	to use for verification
  * @radar_detect: a bitmap where each bit corresponds to a channel
@@ -1611,6 +1612,7 @@ struct cfg80211_color_change_settings {
  *	the verification
  */
 struct iface_combination_params {
+	int radio_idx;
 	int num_different_channels;
 	u8 radar_detect;
 	int iftype_num[NUM_NL80211_IFTYPES];
@@ -4580,6 +4582,8 @@ struct mgmt_frame_regs {
  *
  * @set_hw_timestamp: Enable/disable HW timestamping of TM/FTM frames.
  * @set_ttlm: set the TID to link mapping.
+ * @get_radio_mask: get bitmask of radios in use.
+ *	(invoked with the wiphy mutex held)
  */
 struct cfg80211_ops {
 	int	(*suspend)(struct wiphy *wiphy, struct cfg80211_wowlan *wow);
@@ -4941,6 +4945,7 @@ struct cfg80211_ops {
 				    struct cfg80211_set_hw_timestamp *hwts);
 	int	(*set_ttlm)(struct wiphy *wiphy, struct net_device *dev,
 			    struct cfg80211_ttlm_params *params);
+	u32	(*get_radio_mask)(struct wiphy *wiphy, struct net_device *dev);
 };
 
 /*
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index c6d5f73119d8..27f0db2e9796 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -3944,6 +3944,7 @@ int ieee80211_check_combinations(struct ieee80211_sub_if_data *sdata,
 	int total = 1;
 	struct iface_combination_params params = {
 		.radar_detect = radar_detect,
+		.radio_idx = -1,
 	};
 
 	lockdep_assert_wiphy(local->hw.wiphy);
@@ -4034,7 +4035,9 @@ int ieee80211_max_num_channels(struct ieee80211_local *local)
 	struct ieee80211_chanctx *ctx;
 	u32 max_num_different_channels = 1;
 	int err;
-	struct iface_combination_params params = {0};
+	struct iface_combination_params params = {
+		.radio_idx = -1,
+	};
 
 	lockdep_assert_wiphy(local->hw.wiphy);
 
diff --git a/net/wireless/rdev-ops.h b/net/wireless/rdev-ops.h
index 8f15658002ee..ec3f4aa1c807 100644
--- a/net/wireless/rdev-ops.h
+++ b/net/wireless/rdev-ops.h
@@ -1532,4 +1532,16 @@ rdev_set_ttlm(struct cfg80211_registered_device *rdev,
 
 	return ret;
 }
+
+static inline u32
+rdev_get_radio_mask(struct cfg80211_registered_device *rdev,
+		    struct net_device *dev)
+{
+	struct wiphy *wiphy = &rdev->wiphy;
+
+	if (!rdev->ops->get_radio_mask)
+		return 0;
+
+	return rdev->ops->get_radio_mask(wiphy, dev);
+}
 #endif /* __CFG80211_RDEV_OPS */
diff --git a/net/wireless/util.c b/net/wireless/util.c
index af6ec719567f..2492f259621f 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -2307,13 +2307,16 @@ static int cfg80211_wdev_bi(struct wireless_dev *wdev)
 
 static void cfg80211_calculate_bi_data(struct wiphy *wiphy, u32 new_beacon_int,
 				       u32 *beacon_int_gcd,
-				       bool *beacon_int_different)
+				       bool *beacon_int_different,
+				       int radio_idx)
 {
+	struct cfg80211_registered_device *rdev;
 	struct wireless_dev *wdev;
 
 	*beacon_int_gcd = 0;
 	*beacon_int_different = false;
 
+	rdev = wiphy_to_rdev(wiphy);
 	list_for_each_entry(wdev, &wiphy->wdev_list, list) {
 		int wdev_bi;
 
@@ -2321,6 +2324,11 @@ static void cfg80211_calculate_bi_data(struct wiphy *wiphy, u32 new_beacon_int,
 		if (wdev->valid_links)
 			continue;
 
+		/* skip wdevs not active on the given wiphy radio */
+		if (radio_idx >= 0 &&
+		    !(rdev_get_radio_mask(rdev, wdev->netdev) & BIT(radio_idx)))
+			continue;
+
 		wdev_bi = cfg80211_wdev_bi(wdev);
 
 		if (!wdev_bi)
@@ -2368,14 +2376,19 @@ int cfg80211_iter_combinations(struct wiphy *wiphy,
 					    void *data),
 			       void *data)
 {
+	const struct wiphy_radio *radio = NULL;
+	const struct ieee80211_iface_combination *c, *cs;
 	const struct ieee80211_regdomain *regdom;
 	enum nl80211_dfs_regions region = 0;
-	int i, j, iftype;
+	int i, j, n, iftype;
 	int num_interfaces = 0;
 	u32 used_iftypes = 0;
 	u32 beacon_int_gcd;
 	bool beacon_int_different;
 
+	if (params->radio_idx >= 0)
+		radio = &wiphy->radio[params->radio_idx];
+
 	/*
 	 * This is a bit strange, since the iteration used to rely only on
 	 * the data given by the driver, but here it now relies on context,
@@ -2387,7 +2400,8 @@ int cfg80211_iter_combinations(struct wiphy *wiphy,
 	 * interfaces (while being brought up) and channel/radar data.
 	 */
 	cfg80211_calculate_bi_data(wiphy, params->new_beacon_int,
-				   &beacon_int_gcd, &beacon_int_different);
+				   &beacon_int_gcd, &beacon_int_different,
+				   params->radio_idx);
 
 	if (params->radar_detect) {
 		rcu_read_lock();
@@ -2404,13 +2418,18 @@ int cfg80211_iter_combinations(struct wiphy *wiphy,
 			used_iftypes |= BIT(iftype);
 	}
 
-	for (i = 0; i < wiphy->n_iface_combinations; i++) {
-		const struct ieee80211_iface_combination *c;
+	if (radio) {
+		cs = radio->iface_combinations;
+		n = radio->n_iface_combinations;
+	} else {
+		cs = wiphy->iface_combinations;
+		n = wiphy->n_iface_combinations;
+	}
+	for (i = 0; i < n; i++) {
 		struct ieee80211_iface_limit *limits;
 		u32 all_iftypes = 0;
 
-		c = &wiphy->iface_combinations[i];
-
+		c = &cs[i];
 		if (num_interfaces > c->max_interfaces)
 			continue;
 		if (params->num_different_channels > c->num_different_channels)
-- 
cgit v1.2.3-58-ga151


From 417d88189ccf645a157a3dc3da7e894c1f2829d1 Mon Sep 17 00:00:00 2001
From: Thorsten Blum <thorsten.blum@toblux.com>
Date: Thu, 4 Jul 2024 22:25:59 +0200
Subject: sctp: Fix typos and improve comments

Fix typos s/steam/stream/ and spell out Schedule/Unschedule in the
comments.

Compile-tested only.

Signed-off-by: Thorsten Blum <thorsten.blum@toblux.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240704202558.62704-2-thorsten.blum@toblux.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/sctp/stream_sched.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/net/sctp/stream_sched.h b/include/net/sctp/stream_sched.h
index 572d73fdcd5e..8034bf5febbe 100644
--- a/include/net/sctp/stream_sched.h
+++ b/include/net/sctp/stream_sched.h
@@ -35,10 +35,10 @@ struct sctp_sched_ops {
 	struct sctp_chunk *(*dequeue)(struct sctp_outq *q);
 	/* Called only if the chunk fit the packet */
 	void (*dequeue_done)(struct sctp_outq *q, struct sctp_chunk *chunk);
-	/* Sched all chunks already enqueued */
-	void (*sched_all)(struct sctp_stream *steam);
-	/* Unched all chunks already enqueued */
-	void (*unsched_all)(struct sctp_stream *steam);
+	/* Schedule all chunks already enqueued */
+	void (*sched_all)(struct sctp_stream *stream);
+	/* Unschedule all chunks already enqueued */
+	void (*unsched_all)(struct sctp_stream *stream);
 };
 
 int sctp_sched_set_sched(struct sctp_association *asoc,
-- 
cgit v1.2.3-58-ga151


From 510dba80ed669d6123901ccf0476706122b008b1 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 9 Jul 2024 10:38:32 +0200
Subject: wifi: cfg80211: add helper for checking if a chandef is valid on a
 radio

Check if the full channel width is in the radio's frequency range.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://patch.msgid.link/7c8ea146feb6f37cee62e5ba6be5370403695797.1720514221.git-series.nbd@nbd.name
[add missing Return: documentation]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/cfg80211.h | 11 +++++++++++
 net/wireless/util.c    | 35 +++++++++++++++++++++++++++++++++++
 2 files changed, 46 insertions(+)

diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index 4767e2c76b01..192d72c8b465 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -6508,6 +6508,17 @@ static inline bool cfg80211_channel_is_psc(struct ieee80211_channel *chan)
 	return ieee80211_frequency_to_channel(chan->center_freq) % 16 == 5;
 }
 
+/**
+ * cfg80211_radio_chandef_valid - Check if the radio supports the chandef
+ *
+ * @radio: wiphy radio
+ * @chandef: chandef for current channel
+ *
+ * Return: whether or not the given chandef is valid for the given radio
+ */
+bool cfg80211_radio_chandef_valid(const struct wiphy_radio *radio,
+				  const struct cfg80211_chan_def *chandef);
+
 /**
  * ieee80211_get_response_rate - get basic rate for a given rate
  *
diff --git a/net/wireless/util.c b/net/wireless/util.c
index 2492f259621f..9a7c3adc8a3b 100644
--- a/net/wireless/util.c
+++ b/net/wireless/util.c
@@ -2886,3 +2886,38 @@ cfg80211_get_iftype_ext_capa(struct wiphy *wiphy, enum nl80211_iftype type)
 	return NULL;
 }
 EXPORT_SYMBOL(cfg80211_get_iftype_ext_capa);
+
+static bool
+ieee80211_radio_freq_range_valid(const struct wiphy_radio *radio,
+				 u32 freq, u32 width)
+{
+	const struct wiphy_radio_freq_range *r;
+	int i;
+
+	for (i = 0; i < radio->n_freq_range; i++) {
+		r = &radio->freq_range[i];
+		if (freq - width / 2 >= r->start_freq &&
+		    freq + width / 2 <= r->end_freq)
+			return true;
+	}
+
+	return false;
+}
+
+bool cfg80211_radio_chandef_valid(const struct wiphy_radio *radio,
+				  const struct cfg80211_chan_def *chandef)
+{
+	u32 freq, width;
+
+	freq = ieee80211_chandef_to_khz(chandef);
+	width = nl80211_chan_width_to_mhz(chandef->width);
+	if (!ieee80211_radio_freq_range_valid(radio, freq, width))
+		return false;
+
+	freq = MHZ_TO_KHZ(chandef->center_freq2);
+	if (freq && !ieee80211_radio_freq_range_valid(radio, freq, width))
+		return false;
+
+	return true;
+}
+EXPORT_SYMBOL(cfg80211_radio_chandef_valid);
-- 
cgit v1.2.3-58-ga151


From a01b1e9f9955679fa04adead06c100022e313e55 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 9 Jul 2024 10:38:33 +0200
Subject: wifi: mac80211: add support for DFS with multiple radios

DFS can be supported with multi-channel combinations, as long as each DFS
capable radio only supports one channel.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://patch.msgid.link/4d27a4adca99fa832af1f7cda4f2e71016bd9fda.1720514221.git-series.nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/main.c | 50 +++++++++++++++++++++++++++++++-------------------
 1 file changed, 31 insertions(+), 19 deletions(-)

diff --git a/net/mac80211/main.c b/net/mac80211/main.c
index 85a267bdb3e3..a3104b6ea6f0 100644
--- a/net/mac80211/main.c
+++ b/net/mac80211/main.c
@@ -1091,6 +1091,27 @@ static int ieee80211_init_cipher_suites(struct ieee80211_local *local)
 	return 0;
 }
 
+static bool
+ieee80211_ifcomb_check(const struct ieee80211_iface_combination *c, int n_comb)
+{
+	int i, j;
+
+	for (i = 0; i < n_comb; i++, c++) {
+		/* DFS is not supported with multi-channel combinations yet */
+		if (c->radar_detect_widths &&
+		    c->num_different_channels > 1)
+			return false;
+
+		/* mac80211 doesn't support more than one IBSS interface */
+		for (j = 0; j < c->n_limits; j++)
+			if ((c->limits[j].types & BIT(NL80211_IFTYPE_ADHOC)) &&
+			    c->limits[j].max > 1)
+				return false;
+	}
+
+	return true;
+}
+
 int ieee80211_register_hw(struct ieee80211_hw *hw)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
@@ -1177,17 +1198,20 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 			if (comb->num_different_channels > 1)
 				return -EINVAL;
 		}
-	} else {
-		/* DFS is not supported with multi-channel combinations yet */
-		for (i = 0; i < local->hw.wiphy->n_iface_combinations; i++) {
-			const struct ieee80211_iface_combination *comb;
+	}
 
-			comb = &local->hw.wiphy->iface_combinations[i];
+	if (hw->wiphy->n_radio) {
+		for (i = 0; i < hw->wiphy->n_radio; i++) {
+			const struct wiphy_radio *radio = &hw->wiphy->radio[i];
 
-			if (comb->radar_detect_widths &&
-			    comb->num_different_channels > 1)
+			if (!ieee80211_ifcomb_check(radio->iface_combinations,
+						    radio->n_iface_combinations))
 				return -EINVAL;
 		}
+	} else {
+		if (!ieee80211_ifcomb_check(hw->wiphy->iface_combinations,
+					    hw->wiphy->n_iface_combinations))
+			return -EINVAL;
 	}
 
 	/* Only HW csum features are currently compatible with mac80211 */
@@ -1317,18 +1341,6 @@ int ieee80211_register_hw(struct ieee80211_hw *hw)
 	hw->wiphy->interface_modes |= BIT(NL80211_IFTYPE_MONITOR);
 	hw->wiphy->software_iftypes |= BIT(NL80211_IFTYPE_MONITOR);
 
-	/* mac80211 doesn't support more than one IBSS interface right now */
-	for (i = 0; i < hw->wiphy->n_iface_combinations; i++) {
-		const struct ieee80211_iface_combination *c;
-		int j;
-
-		c = &hw->wiphy->iface_combinations[i];
-
-		for (j = 0; j < c->n_limits; j++)
-			if ((c->limits[j].types & BIT(NL80211_IFTYPE_ADHOC)) &&
-			    c->limits[j].max > 1)
-				return -EINVAL;
-	}
 
 	local->int_scan_req = kzalloc(sizeof(*local->int_scan_req) +
 				      sizeof(void *) * channels, GFP_KERNEL);
-- 
cgit v1.2.3-58-ga151


From 2920bc8d916d30b5273ec16e6878f13b24e3851f Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 9 Jul 2024 10:38:34 +0200
Subject: wifi: mac80211: add radio index to ieee80211_chanctx_conf

Will be used to explicitly assign a channel context to a wiphy radio.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://patch.msgid.link/59f76f57d935f155099276be22badfa671d5bfd9.1720514221.git-series.nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 include/net/mac80211.h | 2 ++
 net/mac80211/chan.c    | 8 +++++---
 2 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index bd0f8aefa797..e78ccbe38d6d 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -250,6 +250,7 @@ struct ieee80211_chan_req {
  * @min_def: the minimum channel definition currently required.
  * @ap: the channel definition the AP actually is operating as,
  *	for use with (wider bandwidth) OFDMA
+ * @radio_idx: index of the wiphy radio used used for this channel
  * @rx_chains_static: The number of RX chains that must always be
  *	active on the channel to receive MIMO transmissions
  * @rx_chains_dynamic: The number of RX chains that must be enabled
@@ -264,6 +265,7 @@ struct ieee80211_chanctx_conf {
 	struct cfg80211_chan_def min_def;
 	struct cfg80211_chan_def ap;
 
+	int radio_idx;
 	u8 rx_chains_static, rx_chains_dynamic;
 
 	bool radar_enabled;
diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c
index 06a65dc6f6c6..6c4d02cb07bb 100644
--- a/net/mac80211/chan.c
+++ b/net/mac80211/chan.c
@@ -656,7 +656,8 @@ ieee80211_chanctx_radar_required(struct ieee80211_local *local,
 static struct ieee80211_chanctx *
 ieee80211_alloc_chanctx(struct ieee80211_local *local,
 			const struct ieee80211_chan_req *chanreq,
-			enum ieee80211_chanctx_mode mode)
+			enum ieee80211_chanctx_mode mode,
+			int radio_idx)
 {
 	struct ieee80211_chanctx *ctx;
 
@@ -674,6 +675,7 @@ ieee80211_alloc_chanctx(struct ieee80211_local *local,
 	ctx->conf.rx_chains_dynamic = 1;
 	ctx->mode = mode;
 	ctx->conf.radar_enabled = false;
+	ctx->conf.radio_idx = radio_idx;
 	_ieee80211_recalc_chanctx_min_def(local, ctx, NULL, false);
 
 	return ctx;
@@ -714,7 +716,7 @@ ieee80211_new_chanctx(struct ieee80211_local *local,
 
 	lockdep_assert_wiphy(local->hw.wiphy);
 
-	ctx = ieee80211_alloc_chanctx(local, chanreq, mode);
+	ctx = ieee80211_alloc_chanctx(local, chanreq, mode, -1);
 	if (!ctx)
 		return ERR_PTR(-ENOMEM);
 
@@ -1155,7 +1157,7 @@ int ieee80211_link_reserve_chanctx(struct ieee80211_link_data *link,
 			    !list_empty(&curr_ctx->reserved_links))
 				return -EBUSY;
 
-			new_ctx = ieee80211_alloc_chanctx(local, chanreq, mode);
+			new_ctx = ieee80211_alloc_chanctx(local, chanreq, mode, -1);
 			if (!new_ctx)
 				return -ENOMEM;
 
-- 
cgit v1.2.3-58-ga151


From 0874bcd0e1c97db0bada32df0934d5cf2507bedd Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 9 Jul 2024 10:38:35 +0200
Subject: wifi: mac80211: extend ifcomb check functions for multi-radio

Add support for counting global and per-radio max/current number of
channels, as well as checking radio-specific interface combinations.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://patch.msgid.link/e76307f8ce562a91a74faab274ae01f6a5ba0a2e.1720514221.git-series.nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/cfg.c         |   7 ++-
 net/mac80211/chan.c        |  17 ++++--
 net/mac80211/ibss.c        |   2 +-
 net/mac80211/ieee80211_i.h |   5 +-
 net/mac80211/iface.c       |   2 +-
 net/mac80211/util.c        | 146 +++++++++++++++++++++++++++++----------------
 6 files changed, 114 insertions(+), 65 deletions(-)

diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c
index 3d49b3ee3a2b..85cb71de370f 100644
--- a/net/mac80211/cfg.c
+++ b/net/mac80211/cfg.c
@@ -263,7 +263,7 @@ static int ieee80211_start_p2p_device(struct wiphy *wiphy,
 
 	lockdep_assert_wiphy(sdata->local->hw.wiphy);
 
-	ret = ieee80211_check_combinations(sdata, NULL, 0, 0);
+	ret = ieee80211_check_combinations(sdata, NULL, 0, 0, -1);
 	if (ret < 0)
 		return ret;
 
@@ -285,7 +285,7 @@ static int ieee80211_start_nan(struct wiphy *wiphy,
 
 	lockdep_assert_wiphy(sdata->local->hw.wiphy);
 
-	ret = ieee80211_check_combinations(sdata, NULL, 0, 0);
+	ret = ieee80211_check_combinations(sdata, NULL, 0, 0, -1);
 	if (ret < 0)
 		return ret;
 
@@ -4008,7 +4008,7 @@ __ieee80211_channel_switch(struct wiphy *wiphy, struct net_device *dev,
 		goto out;
 
 	/* if reservation is invalid then this will fail */
-	err = ieee80211_check_combinations(sdata, NULL, chanctx->mode, 0);
+	err = ieee80211_check_combinations(sdata, NULL, chanctx->mode, 0, -1);
 	if (err) {
 		ieee80211_link_unreserve_chanctx(link_data);
 		goto out;
@@ -5203,4 +5203,5 @@ const struct cfg80211_ops mac80211_config_ops = {
 	.del_link_station = ieee80211_del_link_station,
 	.set_hw_timestamp = ieee80211_set_hw_timestamp,
 	.set_ttlm = ieee80211_set_ttlm,
+	.get_radio_mask = ieee80211_get_radio_mask,
 };
diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c
index 6c4d02cb07bb..12bad02da561 100644
--- a/net/mac80211/chan.c
+++ b/net/mac80211/chan.c
@@ -47,24 +47,29 @@ int ieee80211_chanctx_refcount(struct ieee80211_local *local,
 	       ieee80211_chanctx_num_reserved(local, ctx);
 }
 
-static int ieee80211_num_chanctx(struct ieee80211_local *local)
+static int ieee80211_num_chanctx(struct ieee80211_local *local, int radio_idx)
 {
 	struct ieee80211_chanctx *ctx;
 	int num = 0;
 
 	lockdep_assert_wiphy(local->hw.wiphy);
 
-	list_for_each_entry(ctx, &local->chanctx_list, list)
+	list_for_each_entry(ctx, &local->chanctx_list, list) {
+		if (radio_idx >= 0 && ctx->conf.radio_idx != radio_idx)
+			continue;
 		num++;
+	}
 
 	return num;
 }
 
-static bool ieee80211_can_create_new_chanctx(struct ieee80211_local *local)
+static bool ieee80211_can_create_new_chanctx(struct ieee80211_local *local,
+					     int radio_idx)
 {
 	lockdep_assert_wiphy(local->hw.wiphy);
 
-	return ieee80211_num_chanctx(local) < ieee80211_max_num_channels(local);
+	return ieee80211_num_chanctx(local, radio_idx) <
+	       ieee80211_max_num_channels(local, radio_idx);
 }
 
 static struct ieee80211_chanctx *
@@ -1101,7 +1106,7 @@ int ieee80211_link_reserve_chanctx(struct ieee80211_link_data *link,
 
 	new_ctx = ieee80211_find_reservation_chanctx(local, chanreq, mode);
 	if (!new_ctx) {
-		if (ieee80211_can_create_new_chanctx(local)) {
+		if (ieee80211_can_create_new_chanctx(local, -1)) {
 			new_ctx = ieee80211_new_chanctx(local, chanreq, mode,
 							false);
 			if (IS_ERR(new_ctx))
@@ -1822,7 +1827,7 @@ int _ieee80211_link_use_channel(struct ieee80211_link_data *link,
 	link->radar_required = ret;
 
 	ret = ieee80211_check_combinations(sdata, &chanreq->oper, mode,
-					   radar_detect_width);
+					   radar_detect_width, -1);
 	if (ret < 0)
 		goto out;
 
diff --git a/net/mac80211/ibss.c b/net/mac80211/ibss.c
index 7db4c3ee7e6d..3f74bbceeca5 100644
--- a/net/mac80211/ibss.c
+++ b/net/mac80211/ibss.c
@@ -1746,7 +1746,7 @@ int ieee80211_ibss_join(struct ieee80211_sub_if_data *sdata,
 		IEEE80211_CHANCTX_SHARED : IEEE80211_CHANCTX_EXCLUSIVE;
 
 	ret = ieee80211_check_combinations(sdata, &params->chandef, chanmode,
-					   radar_detect_width);
+					   radar_detect_width, -1);
 	if (ret < 0)
 		return ret;
 
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 6349552e62a8..a3485e4c6132 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -2640,8 +2640,9 @@ void ieee80211_recalc_dtim(struct ieee80211_local *local,
 int ieee80211_check_combinations(struct ieee80211_sub_if_data *sdata,
 				 const struct cfg80211_chan_def *chandef,
 				 enum ieee80211_chanctx_mode chanmode,
-				 u8 radar_detect);
-int ieee80211_max_num_channels(struct ieee80211_local *local);
+				 u8 radar_detect, int radio_idx);
+int ieee80211_max_num_channels(struct ieee80211_local *local, int radio_idx);
+u32 ieee80211_get_radio_mask(struct wiphy *wiphy, struct net_device *dev);
 void ieee80211_recalc_chanctx_chantype(struct ieee80211_local *local,
 				       struct ieee80211_chanctx *ctx);
 
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 97aee0a1a39a..b4ad66af3af3 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -397,7 +397,7 @@ static int ieee80211_check_concurrent_iface(struct ieee80211_sub_if_data *sdata,
 		}
 	}
 
-	return ieee80211_check_combinations(sdata, NULL, 0, 0);
+	return ieee80211_check_combinations(sdata, NULL, 0, 0, -1);
 }
 
 static int ieee80211_check_queues(struct ieee80211_sub_if_data *sdata,
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 27f0db2e9796..ced19ce7c51a 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -3932,20 +3932,103 @@ static u8 ieee80211_chanctx_radar_detect(struct ieee80211_local *local,
 	return radar_detect;
 }
 
+static u32
+__ieee80211_get_radio_mask(struct ieee80211_sub_if_data *sdata)
+{
+	struct ieee80211_bss_conf *link_conf;
+	struct ieee80211_chanctx_conf *conf;
+	unsigned int link_id;
+	u32 mask = 0;
+
+	for_each_vif_active_link(&sdata->vif, link_conf, link_id) {
+		conf = sdata_dereference(link_conf->chanctx_conf, sdata);
+		if (!conf || conf->radio_idx < 0)
+			continue;
+
+		mask |= BIT(conf->radio_idx);
+	}
+
+	return mask;
+}
+
+u32 ieee80211_get_radio_mask(struct wiphy *wiphy, struct net_device *dev)
+{
+	struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev);
+
+	return __ieee80211_get_radio_mask(sdata);
+}
+
+static bool
+ieee80211_sdata_uses_radio(struct ieee80211_sub_if_data *sdata, int radio_idx)
+{
+	if (radio_idx < 0)
+		return true;
+
+	return __ieee80211_get_radio_mask(sdata) & BIT(radio_idx);
+}
+
+static int
+ieee80211_fill_ifcomb_params(struct ieee80211_local *local,
+			     struct iface_combination_params *params,
+			     const struct cfg80211_chan_def *chandef,
+			     struct ieee80211_sub_if_data *sdata)
+{
+	struct ieee80211_sub_if_data *sdata_iter;
+	struct ieee80211_chanctx *ctx;
+	int total = !!sdata;
+
+	list_for_each_entry(ctx, &local->chanctx_list, list) {
+		if (ctx->replace_state == IEEE80211_CHANCTX_WILL_BE_REPLACED)
+			continue;
+
+		if (params->radio_idx >= 0 &&
+		    ctx->conf.radio_idx != params->radio_idx)
+			continue;
+
+		params->radar_detect |=
+			ieee80211_chanctx_radar_detect(local, ctx);
+
+		if (chandef && ctx->mode != IEEE80211_CHANCTX_EXCLUSIVE &&
+		    cfg80211_chandef_compatible(chandef, &ctx->conf.def))
+			continue;
+
+		params->num_different_channels++;
+	}
+
+	list_for_each_entry(sdata_iter, &local->interfaces, list) {
+		struct wireless_dev *wdev_iter;
+
+		wdev_iter = &sdata_iter->wdev;
+
+		if (sdata_iter == sdata ||
+		    !ieee80211_sdata_running(sdata_iter) ||
+		    cfg80211_iftype_allowed(local->hw.wiphy,
+					    wdev_iter->iftype, 0, 1))
+			continue;
+
+		if (!ieee80211_sdata_uses_radio(sdata_iter, params->radio_idx))
+			continue;
+
+		params->iftype_num[wdev_iter->iftype]++;
+		total++;
+	}
+
+	return total;
+}
+
 int ieee80211_check_combinations(struct ieee80211_sub_if_data *sdata,
 				 const struct cfg80211_chan_def *chandef,
 				 enum ieee80211_chanctx_mode chanmode,
-				 u8 radar_detect)
+				 u8 radar_detect, int radio_idx)
 {
+	bool shared = chanmode == IEEE80211_CHANCTX_SHARED;
 	struct ieee80211_local *local = sdata->local;
-	struct ieee80211_sub_if_data *sdata_iter;
 	enum nl80211_iftype iftype = sdata->wdev.iftype;
-	struct ieee80211_chanctx *ctx;
-	int total = 1;
 	struct iface_combination_params params = {
 		.radar_detect = radar_detect,
-		.radio_idx = -1,
+		.radio_idx = radio_idx,
 	};
+	int total;
 
 	lockdep_assert_wiphy(local->hw.wiphy);
 
@@ -3982,37 +4065,9 @@ int ieee80211_check_combinations(struct ieee80211_sub_if_data *sdata,
 	if (iftype != NL80211_IFTYPE_UNSPECIFIED)
 		params.iftype_num[iftype] = 1;
 
-	list_for_each_entry(ctx, &local->chanctx_list, list) {
-		if (ctx->replace_state == IEEE80211_CHANCTX_WILL_BE_REPLACED)
-			continue;
-		params.radar_detect |=
-			ieee80211_chanctx_radar_detect(local, ctx);
-		if (ctx->mode == IEEE80211_CHANCTX_EXCLUSIVE) {
-			params.num_different_channels++;
-			continue;
-		}
-		if (chandef && chanmode == IEEE80211_CHANCTX_SHARED &&
-		    cfg80211_chandef_compatible(chandef,
-						&ctx->conf.def))
-			continue;
-		params.num_different_channels++;
-	}
-
-	list_for_each_entry_rcu(sdata_iter, &local->interfaces, list) {
-		struct wireless_dev *wdev_iter;
-
-		wdev_iter = &sdata_iter->wdev;
-
-		if (sdata_iter == sdata ||
-		    !ieee80211_sdata_running(sdata_iter) ||
-		    cfg80211_iftype_allowed(local->hw.wiphy,
-					    wdev_iter->iftype, 0, 1))
-			continue;
-
-		params.iftype_num[wdev_iter->iftype]++;
-		total++;
-	}
-
+	total = ieee80211_fill_ifcomb_params(local, &params,
+					     shared ? chandef : NULL,
+					     sdata);
 	if (total == 1 && !params.radar_detect)
 		return 0;
 
@@ -4029,30 +4084,17 @@ ieee80211_iter_max_chans(const struct ieee80211_iface_combination *c,
 					  c->num_different_channels);
 }
 
-int ieee80211_max_num_channels(struct ieee80211_local *local)
+int ieee80211_max_num_channels(struct ieee80211_local *local, int radio_idx)
 {
-	struct ieee80211_sub_if_data *sdata;
-	struct ieee80211_chanctx *ctx;
 	u32 max_num_different_channels = 1;
 	int err;
 	struct iface_combination_params params = {
-		.radio_idx = -1,
+		.radio_idx = radio_idx,
 	};
 
 	lockdep_assert_wiphy(local->hw.wiphy);
 
-	list_for_each_entry(ctx, &local->chanctx_list, list) {
-		if (ctx->replace_state == IEEE80211_CHANCTX_WILL_BE_REPLACED)
-			continue;
-
-		params.num_different_channels++;
-
-		params.radar_detect |=
-			ieee80211_chanctx_radar_detect(local, ctx);
-	}
-
-	list_for_each_entry_rcu(sdata, &local->interfaces, list)
-		params.iftype_num[sdata->wdev.iftype]++;
+	ieee80211_fill_ifcomb_params(local, &params, NULL, NULL);
 
 	err = cfg80211_iter_combinations(local->hw.wiphy, &params,
 					 ieee80211_iter_max_chans,
-- 
cgit v1.2.3-58-ga151


From 6265c67f2668047c97834c5c434f6abcf86e3406 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 9 Jul 2024 10:38:36 +0200
Subject: wifi: mac80211: move code in ieee80211_link_reserve_chanctx to a
 helper

Reduces indentation in preparation for further changes

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://patch.msgid.link/cce95007092336254d51570f4a27e05a6f150a53.1720514221.git-series.nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/chan.c | 141 +++++++++++++++++++++++++++-------------------------
 1 file changed, 72 insertions(+), 69 deletions(-)

diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c
index 12bad02da561..9aa3b9e25228 100644
--- a/net/mac80211/chan.c
+++ b/net/mac80211/chan.c
@@ -1089,6 +1089,71 @@ int ieee80211_link_unreserve_chanctx(struct ieee80211_link_data *link)
 	return 0;
 }
 
+static struct ieee80211_chanctx *
+ieee80211_replace_chanctx(struct ieee80211_local *local,
+			  const struct ieee80211_chan_req *chanreq,
+			  enum ieee80211_chanctx_mode mode,
+			  struct ieee80211_chanctx *curr_ctx)
+{
+	struct ieee80211_chanctx *new_ctx, *ctx;
+
+	if (!curr_ctx || (curr_ctx->replace_state ==
+			  IEEE80211_CHANCTX_WILL_BE_REPLACED) ||
+	    !list_empty(&curr_ctx->reserved_links)) {
+		/*
+		 * Another link already requested this context for a
+		 * reservation. Find another one hoping all links assigned
+		 * to it will also switch soon enough.
+		 *
+		 * TODO: This needs a little more work as some cases
+		 * (more than 2 chanctx capable devices) may fail which could
+		 * otherwise succeed provided some channel context juggling was
+		 * performed.
+		 *
+		 * Consider ctx1..3, link1..6, each ctx has 2 links. link1 and
+		 * link2 from ctx1 request new different chandefs starting 2
+		 * in-place reserations with ctx4 and ctx5 replacing ctx1 and
+		 * ctx2 respectively. Next link5 and link6 from ctx3 reserve
+		 * ctx4. If link3 and link4 remain on ctx2 as they are then this
+		 * fails unless `replace_ctx` from ctx5 is replaced with ctx3.
+		 */
+		list_for_each_entry(ctx, &local->chanctx_list, list) {
+			if (ctx->replace_state !=
+			    IEEE80211_CHANCTX_REPLACE_NONE)
+				continue;
+
+			if (!list_empty(&ctx->reserved_links))
+				continue;
+
+			curr_ctx = ctx;
+			break;
+		}
+	}
+
+	/*
+	 * If that's true then all available contexts already have reservations
+	 * and cannot be used.
+	 */
+	if (!curr_ctx || (curr_ctx->replace_state ==
+			  IEEE80211_CHANCTX_WILL_BE_REPLACED) ||
+	    !list_empty(&curr_ctx->reserved_links))
+		return ERR_PTR(-EBUSY);
+
+	new_ctx = ieee80211_alloc_chanctx(local, chanreq, mode, -1);
+	if (!new_ctx)
+		return ERR_PTR(-ENOMEM);
+
+	new_ctx->replace_ctx = curr_ctx;
+	new_ctx->replace_state = IEEE80211_CHANCTX_REPLACES_OTHER;
+
+	curr_ctx->replace_ctx = new_ctx;
+	curr_ctx->replace_state = IEEE80211_CHANCTX_WILL_BE_REPLACED;
+
+	list_add_rcu(&new_ctx->list, &local->chanctx_list);
+
+	return new_ctx;
+}
+
 int ieee80211_link_reserve_chanctx(struct ieee80211_link_data *link,
 				   const struct ieee80211_chan_req *chanreq,
 				   enum ieee80211_chanctx_mode mode,
@@ -1096,7 +1161,7 @@ int ieee80211_link_reserve_chanctx(struct ieee80211_link_data *link,
 {
 	struct ieee80211_sub_if_data *sdata = link->sdata;
 	struct ieee80211_local *local = sdata->local;
-	struct ieee80211_chanctx *new_ctx, *curr_ctx, *ctx;
+	struct ieee80211_chanctx *new_ctx, *curr_ctx;
 
 	lockdep_assert_wiphy(local->hw.wiphy);
 
@@ -1106,76 +1171,14 @@ int ieee80211_link_reserve_chanctx(struct ieee80211_link_data *link,
 
 	new_ctx = ieee80211_find_reservation_chanctx(local, chanreq, mode);
 	if (!new_ctx) {
-		if (ieee80211_can_create_new_chanctx(local, -1)) {
+		if (ieee80211_can_create_new_chanctx(local, -1))
 			new_ctx = ieee80211_new_chanctx(local, chanreq, mode,
 							false);
-			if (IS_ERR(new_ctx))
-				return PTR_ERR(new_ctx);
-		} else {
-			if (!curr_ctx ||
-			    (curr_ctx->replace_state ==
-			     IEEE80211_CHANCTX_WILL_BE_REPLACED) ||
-			    !list_empty(&curr_ctx->reserved_links)) {
-				/*
-				 * Another link already requested this context
-				 * for a reservation. Find another one hoping
-				 * all links assigned to it will also switch
-				 * soon enough.
-				 *
-				 * TODO: This needs a little more work as some
-				 * cases (more than 2 chanctx capable devices)
-				 * may fail which could otherwise succeed
-				 * provided some channel context juggling was
-				 * performed.
-				 *
-				 * Consider ctx1..3, link1..6, each ctx has 2
-				 * links. link1 and link2 from ctx1 request new
-				 * different chandefs starting 2 in-place
-				 * reserations with ctx4 and ctx5 replacing
-				 * ctx1 and ctx2 respectively. Next link5 and
-				 * link6 from ctx3 reserve ctx4. If link3 and
-				 * link4 remain on ctx2 as they are then this
-				 * fails unless `replace_ctx` from ctx5 is
-				 * replaced with ctx3.
-				 */
-				list_for_each_entry(ctx, &local->chanctx_list,
-						    list) {
-					if (ctx->replace_state !=
-					    IEEE80211_CHANCTX_REPLACE_NONE)
-						continue;
-
-					if (!list_empty(&ctx->reserved_links))
-						continue;
-
-					curr_ctx = ctx;
-					break;
-				}
-			}
-
-			/*
-			 * If that's true then all available contexts already
-			 * have reservations and cannot be used.
-			 */
-			if (!curr_ctx ||
-			    (curr_ctx->replace_state ==
-			     IEEE80211_CHANCTX_WILL_BE_REPLACED) ||
-			    !list_empty(&curr_ctx->reserved_links))
-				return -EBUSY;
-
-			new_ctx = ieee80211_alloc_chanctx(local, chanreq, mode, -1);
-			if (!new_ctx)
-				return -ENOMEM;
-
-			new_ctx->replace_ctx = curr_ctx;
-			new_ctx->replace_state =
-					IEEE80211_CHANCTX_REPLACES_OTHER;
-
-			curr_ctx->replace_ctx = new_ctx;
-			curr_ctx->replace_state =
-					IEEE80211_CHANCTX_WILL_BE_REPLACED;
-
-			list_add_rcu(&new_ctx->list, &local->chanctx_list);
-		}
+		else
+			new_ctx = ieee80211_replace_chanctx(local, chanreq,
+							    mode, curr_ctx);
+		if (IS_ERR(new_ctx))
+			return PTR_ERR(new_ctx);
 	}
 
 	list_add(&link->reserved_chanctx_list, &new_ctx->reserved_links);
-- 
cgit v1.2.3-58-ga151


From 27d4c03441eb951142c79fff0a25dd8ba3263875 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 9 Jul 2024 10:38:37 +0200
Subject: wifi: mac80211: add wiphy radio assignment and validation

Validate number of channels and interface combinations per radio.
Assign each channel context to a radio.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://patch.msgid.link/1d3e9ba70a30ce18aaff337f0a76d7aeb311bafb.1720514221.git-series.nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/chan.c | 52 +++++++++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 47 insertions(+), 5 deletions(-)

diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c
index 9aa3b9e25228..e8567723e94d 100644
--- a/net/mac80211/chan.c
+++ b/net/mac80211/chan.c
@@ -714,14 +714,15 @@ static struct ieee80211_chanctx *
 ieee80211_new_chanctx(struct ieee80211_local *local,
 		      const struct ieee80211_chan_req *chanreq,
 		      enum ieee80211_chanctx_mode mode,
-		      bool assign_on_failure)
+		      bool assign_on_failure,
+		      int radio_idx)
 {
 	struct ieee80211_chanctx *ctx;
 	int err;
 
 	lockdep_assert_wiphy(local->hw.wiphy);
 
-	ctx = ieee80211_alloc_chanctx(local, chanreq, mode, -1);
+	ctx = ieee80211_alloc_chanctx(local, chanreq, mode, radio_idx);
 	if (!ctx)
 		return ERR_PTR(-ENOMEM);
 
@@ -1096,6 +1097,8 @@ ieee80211_replace_chanctx(struct ieee80211_local *local,
 			  struct ieee80211_chanctx *curr_ctx)
 {
 	struct ieee80211_chanctx *new_ctx, *ctx;
+	struct wiphy *wiphy = local->hw.wiphy;
+	const struct wiphy_radio *radio;
 
 	if (!curr_ctx || (curr_ctx->replace_state ==
 			  IEEE80211_CHANCTX_WILL_BE_REPLACED) ||
@@ -1125,6 +1128,12 @@ ieee80211_replace_chanctx(struct ieee80211_local *local,
 			if (!list_empty(&ctx->reserved_links))
 				continue;
 
+			if (ctx->conf.radio_idx >= 0) {
+				radio = &wiphy->radio[ctx->conf.radio_idx];
+				if (!cfg80211_radio_chandef_valid(radio, &chanreq->oper))
+					continue;
+			}
+
 			curr_ctx = ctx;
 			break;
 		}
@@ -1154,6 +1163,34 @@ ieee80211_replace_chanctx(struct ieee80211_local *local,
 	return new_ctx;
 }
 
+static bool
+ieee80211_find_available_radio(struct ieee80211_local *local,
+			       const struct ieee80211_chan_req *chanreq,
+			       int *radio_idx)
+{
+	struct wiphy *wiphy = local->hw.wiphy;
+	const struct wiphy_radio *radio;
+	int i;
+
+	*radio_idx = -1;
+	if (!wiphy->n_radio)
+		return true;
+
+	for (i = 0; i < wiphy->n_radio; i++) {
+		radio = &wiphy->radio[i];
+		if (!cfg80211_radio_chandef_valid(radio, &chanreq->oper))
+			continue;
+
+		if (!ieee80211_can_create_new_chanctx(local, i))
+			continue;
+
+		*radio_idx = i;
+		return true;
+	}
+
+	return false;
+}
+
 int ieee80211_link_reserve_chanctx(struct ieee80211_link_data *link,
 				   const struct ieee80211_chan_req *chanreq,
 				   enum ieee80211_chanctx_mode mode,
@@ -1162,6 +1199,7 @@ int ieee80211_link_reserve_chanctx(struct ieee80211_link_data *link,
 	struct ieee80211_sub_if_data *sdata = link->sdata;
 	struct ieee80211_local *local = sdata->local;
 	struct ieee80211_chanctx *new_ctx, *curr_ctx;
+	int radio_idx;
 
 	lockdep_assert_wiphy(local->hw.wiphy);
 
@@ -1171,9 +1209,10 @@ int ieee80211_link_reserve_chanctx(struct ieee80211_link_data *link,
 
 	new_ctx = ieee80211_find_reservation_chanctx(local, chanreq, mode);
 	if (!new_ctx) {
-		if (ieee80211_can_create_new_chanctx(local, -1))
+		if (ieee80211_can_create_new_chanctx(local, -1) &&
+		    ieee80211_find_available_radio(local, chanreq, &radio_idx))
 			new_ctx = ieee80211_new_chanctx(local, chanreq, mode,
-							false);
+							false, radio_idx);
 		else
 			new_ctx = ieee80211_replace_chanctx(local, chanreq,
 							    mode, curr_ctx);
@@ -1810,6 +1849,7 @@ int _ieee80211_link_use_channel(struct ieee80211_link_data *link,
 	struct ieee80211_chanctx *ctx;
 	u8 radar_detect_width = 0;
 	bool reserved = false;
+	int radio_idx;
 	int ret;
 
 	lockdep_assert_wiphy(local->hw.wiphy);
@@ -1840,9 +1880,11 @@ int _ieee80211_link_use_channel(struct ieee80211_link_data *link,
 	/* Note: context is now reserved */
 	if (ctx)
 		reserved = true;
+	else if (!ieee80211_find_available_radio(local, chanreq, &radio_idx))
+		ctx = ERR_PTR(-EBUSY);
 	else
 		ctx = ieee80211_new_chanctx(local, chanreq, mode,
-					    assign_on_failure);
+					    assign_on_failure, radio_idx);
 	if (IS_ERR(ctx)) {
 		ret = PTR_ERR(ctx);
 		goto out;
-- 
cgit v1.2.3-58-ga151


From d2601e34a1024cb2466e6bd13bd943e19fd54fa6 Mon Sep 17 00:00:00 2001
From: Felix Fietkau <nbd@nbd.name>
Date: Tue, 9 Jul 2024 10:38:38 +0200
Subject: wifi: mac80211_hwsim: add support for multi-radio wiphy

This registers one wiphy radio per supported band. Number of different
channels is set per radio.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Link: https://patch.msgid.link/3a16838bb7a7d1a072bd7c9d586d17f70fcd8a60.1720514221.git-series.nbd@nbd.name
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 74 +++++++++++++++++++++++----
 drivers/net/wireless/virtual/mac80211_hwsim.h |  4 ++
 2 files changed, 67 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c b/drivers/net/wireless/virtual/mac80211_hwsim.c
index d60c780d4e8a..714e1f04b0cb 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.c
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
@@ -69,6 +69,10 @@ static bool mlo;
 module_param(mlo, bool, 0444);
 MODULE_PARM_DESC(mlo, "Support MLO");
 
+static bool multi_radio;
+module_param(multi_radio, bool, 0444);
+MODULE_PARM_DESC(mlo, "Support Multiple Radios per wiphy");
+
 /**
  * enum hwsim_regtest - the type of regulatory tests we offer
  *
@@ -669,6 +673,10 @@ struct mac80211_hwsim_data {
 	struct ieee80211_iface_limit if_limits[3];
 	int n_if_limits;
 
+	struct ieee80211_iface_combination if_combination_radio;
+	struct wiphy_radio_freq_range radio_range[NUM_NL80211_BANDS];
+	struct wiphy_radio radio[NUM_NL80211_BANDS];
+
 	u32 ciphers[ARRAY_SIZE(hwsim_ciphers)];
 
 	struct mac_address addresses[2];
@@ -917,6 +925,7 @@ static const struct nla_policy hwsim_genl_policy[HWSIM_ATTR_MAX + 1] = {
 	[HWSIM_ATTR_MLO_SUPPORT] = { .type = NLA_FLAG },
 	[HWSIM_ATTR_PMSR_SUPPORT] = NLA_POLICY_NESTED(hwsim_pmsr_capa_policy),
 	[HWSIM_ATTR_PMSR_RESULT] = NLA_POLICY_NESTED(hwsim_pmsr_peers_result_policy),
+	[HWSIM_ATTR_MULTI_RADIO] = { .type = NLA_FLAG },
 };
 
 #if IS_REACHABLE(CONFIG_VIRTIO)
@@ -4018,6 +4027,7 @@ struct hwsim_new_radio_params {
 	bool reg_strict;
 	bool p2p_device;
 	bool use_chanctx;
+	bool multi_radio;
 	bool destroy_on_close;
 	const char *hwname;
 	bool no_vif;
@@ -4094,6 +4104,12 @@ static int append_radio_msg(struct sk_buff *skb, int id,
 			return ret;
 	}
 
+	if (param->multi_radio) {
+		ret = nla_put_flag(skb, HWSIM_ATTR_MULTI_RADIO);
+		if (ret < 0)
+			return ret;
+	}
+
 	if (param->hwname) {
 		ret = nla_put(skb, HWSIM_ATTR_RADIO_NAME,
 			      strlen(param->hwname), param->hwname);
@@ -5114,6 +5130,7 @@ static int mac80211_hwsim_new_radio(struct genl_info *info,
 	struct net *net;
 	int idx, i;
 	int n_limits = 0;
+	int n_bands = 0;
 
 	if (WARN_ON(param->channels > 1 && !param->use_chanctx))
 		return -EINVAL;
@@ -5217,22 +5234,22 @@ static int mac80211_hwsim_new_radio(struct genl_info *info,
 		n_limits++;
 	}
 
+	data->if_combination.radar_detect_widths =
+				BIT(NL80211_CHAN_WIDTH_5) |
+				BIT(NL80211_CHAN_WIDTH_10) |
+				BIT(NL80211_CHAN_WIDTH_20_NOHT) |
+				BIT(NL80211_CHAN_WIDTH_20) |
+				BIT(NL80211_CHAN_WIDTH_40) |
+				BIT(NL80211_CHAN_WIDTH_80) |
+				BIT(NL80211_CHAN_WIDTH_160);
+
 	if (data->use_chanctx) {
 		hw->wiphy->max_scan_ssids = 255;
 		hw->wiphy->max_scan_ie_len = IEEE80211_MAX_DATA_LEN;
 		hw->wiphy->max_remain_on_channel_duration = 1000;
-		data->if_combination.radar_detect_widths = 0;
 		data->if_combination.num_different_channels = data->channels;
 	} else {
 		data->if_combination.num_different_channels = 1;
-		data->if_combination.radar_detect_widths =
-					BIT(NL80211_CHAN_WIDTH_5) |
-					BIT(NL80211_CHAN_WIDTH_10) |
-					BIT(NL80211_CHAN_WIDTH_20_NOHT) |
-					BIT(NL80211_CHAN_WIDTH_20) |
-					BIT(NL80211_CHAN_WIDTH_40) |
-					BIT(NL80211_CHAN_WIDTH_80) |
-					BIT(NL80211_CHAN_WIDTH_160);
 	}
 
 	if (!n_limits) {
@@ -5350,6 +5367,9 @@ static int mac80211_hwsim_new_radio(struct genl_info *info,
 
 	for (band = NL80211_BAND_2GHZ; band < NUM_NL80211_BANDS; band++) {
 		struct ieee80211_supported_band *sband = &data->bands[band];
+		struct wiphy_radio_freq_range *radio_range;
+		const struct ieee80211_channel *c;
+		struct wiphy_radio *radio;
 
 		sband->band = band;
 
@@ -5423,8 +5443,36 @@ static int mac80211_hwsim_new_radio(struct genl_info *info,
 		mac80211_hwsim_sband_capab(sband);
 
 		hw->wiphy->bands[band] = sband;
+
+		if (!param->multi_radio)
+			continue;
+
+		c = sband->channels;
+		radio_range = &data->radio_range[n_bands];
+		radio_range->start_freq = ieee80211_channel_to_khz(c) - 10000;
+
+		c += sband->n_channels - 1;
+		radio_range->end_freq = ieee80211_channel_to_khz(c) + 10000;
+
+		radio = &data->radio[n_bands++];
+		radio->freq_range = radio_range;
+		radio->n_freq_range = 1;
+		radio->iface_combinations = &data->if_combination_radio;
+		radio->n_iface_combinations = 1;
 	}
 
+	if (param->multi_radio) {
+		hw->wiphy->radio = data->radio;
+		hw->wiphy->n_radio = n_bands;
+
+		memcpy(&data->if_combination_radio, &data->if_combination,
+		       sizeof(data->if_combination));
+		data->if_combination.num_different_channels *= n_bands;
+	}
+
+	if (data->use_chanctx)
+		data->if_combination.radar_detect_widths = 0;
+
 	/* By default all radios belong to the first group */
 	data->group = 1;
 	mutex_init(&data->mutex);
@@ -6042,6 +6090,9 @@ static int hwsim_new_radio_nl(struct sk_buff *msg, struct genl_info *info)
 	else
 		param.use_chanctx = (param.channels > 1);
 
+	if (info->attrs[HWSIM_ATTR_MULTI_RADIO])
+		param.multi_radio = true;
+
 	if (info->attrs[HWSIM_ATTR_REG_HINT_ALPHA2])
 		param.reg_alpha2 =
 			nla_data(info->attrs[HWSIM_ATTR_REG_HINT_ALPHA2]);
@@ -6122,7 +6173,7 @@ static int hwsim_new_radio_nl(struct sk_buff *msg, struct genl_info *info)
 
 	param.mlo = info->attrs[HWSIM_ATTR_MLO_SUPPORT];
 
-	if (param.mlo)
+	if (param.mlo || param.multi_radio)
 		param.use_chanctx = true;
 
 	if (info->attrs[HWSIM_ATTR_RADIO_NAME]) {
@@ -6815,7 +6866,8 @@ static int __init init_mac80211_hwsim(void)
 
 		param.p2p_device = support_p2p_device;
 		param.mlo = mlo;
-		param.use_chanctx = channels > 1 || mlo;
+		param.multi_radio = multi_radio;
+		param.use_chanctx = channels > 1 || mlo || multi_radio;
 		param.iftypes = HWSIM_IFTYPE_SUPPORT_MASK;
 		if (param.p2p_device)
 			param.iftypes |= BIT(NL80211_IFTYPE_P2P_DEVICE);
diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.h b/drivers/net/wireless/virtual/mac80211_hwsim.h
index 28c1db482e79..f32fc3a492b0 100644
--- a/drivers/net/wireless/virtual/mac80211_hwsim.h
+++ b/drivers/net/wireless/virtual/mac80211_hwsim.h
@@ -157,6 +157,9 @@ enum hwsim_commands {
  *	to provide details about peer measurement request (nl80211_peer_measurement_attrs)
  * @HWSIM_ATTR_PMSR_RESULT: nested attributed used with %HWSIM_CMD_REPORT_PMSR
  *	to provide peer measurement result (nl80211_peer_measurement_attrs)
+ * @HWSIM_ATTR_MULTI_RADIO: Register multiple wiphy radios (flag).
+ *	Adds one radio for each band. Number of supported channels will be set for
+ *	each radio instead of for the wiphy.
  * @__HWSIM_ATTR_MAX: enum limit
  */
 enum hwsim_attrs {
@@ -189,6 +192,7 @@ enum hwsim_attrs {
 	HWSIM_ATTR_PMSR_SUPPORT,
 	HWSIM_ATTR_PMSR_REQUEST,
 	HWSIM_ATTR_PMSR_RESULT,
+	HWSIM_ATTR_MULTI_RADIO,
 	__HWSIM_ATTR_MAX,
 };
 #define HWSIM_ATTR_MAX (__HWSIM_ATTR_MAX - 1)
-- 
cgit v1.2.3-58-ga151


From c145eea2f75ff7949392aebecf7ef0a81c1f6c14 Mon Sep 17 00:00:00 2001
From: Sascha Hauer <s.hauer@pengutronix.de>
Date: Wed, 3 Jul 2024 09:24:09 +0200
Subject: wifi: mwifiex: Do not return unused priv in mwifiex_get_priv_by_id()

mwifiex_get_priv_by_id() returns the priv pointer corresponding to
the bss_num and bss_type, but without checking if the priv is actually
currently in use.
Unused priv pointers do not have a wiphy attached to them which can
lead to NULL pointer dereferences further down the callstack.  Fix
this by returning only used priv pointers which have priv->bss_mode
set to something else than NL80211_IFTYPE_UNSPECIFIED.

Said NULL pointer dereference happened when an Accesspoint was started
with wpa_supplicant -i mlan0 with this config:

network={
        ssid="somessid"
        mode=2
        frequency=2412
        key_mgmt=WPA-PSK WPA-PSK-SHA256
        proto=RSN
        group=CCMP
        pairwise=CCMP
        psk="12345678"
}

When waiting for the AP to be established, interrupting wpa_supplicant
with <ctrl-c> and starting it again this happens:

| Unable to handle kernel NULL pointer dereference at virtual address 0000000000000140
| Mem abort info:
|   ESR = 0x0000000096000004
|   EC = 0x25: DABT (current EL), IL = 32 bits
|   SET = 0, FnV = 0
|   EA = 0, S1PTW = 0
|   FSC = 0x04: level 0 translation fault
| Data abort info:
|   ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
|   CM = 0, WnR = 0, TnD = 0, TagAccess = 0
|   GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
| user pgtable: 4k pages, 48-bit VAs, pgdp=0000000046d96000
| [0000000000000140] pgd=0000000000000000, p4d=0000000000000000
| Internal error: Oops: 0000000096000004 [#1] PREEMPT SMP
| Modules linked in: caam_jr caamhash_desc spidev caamalg_desc crypto_engine authenc libdes mwifiex_sdio
+mwifiex crct10dif_ce cdc_acm onboard_usb_hub fsl_imx8_ddr_perf imx8m_ddrc rtc_ds1307 lm75 rtc_snvs
+imx_sdma caam imx8mm_thermal spi_imx error imx_cpufreq_dt fuse ip_tables x_tables ipv6
| CPU: 0 PID: 8 Comm: kworker/0:1 Not tainted 6.9.0-00007-g937242013fce-dirty #18
| Hardware name: somemachine (DT)
| Workqueue: events sdio_irq_work
| pstate: 00000005 (nzcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
| pc : mwifiex_get_cfp+0xd8/0x15c [mwifiex]
| lr : mwifiex_get_cfp+0x34/0x15c [mwifiex]
| sp : ffff8000818b3a70
| x29: ffff8000818b3a70 x28: ffff000006bfd8a5 x27: 0000000000000004
| x26: 000000000000002c x25: 0000000000001511 x24: 0000000002e86bc9
| x23: ffff000006bfd996 x22: 0000000000000004 x21: ffff000007bec000
| x20: 000000000000002c x19: 0000000000000000 x18: 0000000000000000
| x17: 000000040044ffff x16: 00500072b5503510 x15: ccc283740681e517
| x14: 0201000101006d15 x13: 0000000002e8ff43 x12: 002c01000000ffb1
| x11: 0100000000000000 x10: 02e8ff43002c0100 x9 : 0000ffb100100157
| x8 : ffff000003d20000 x7 : 00000000000002f1 x6 : 00000000ffffe124
| x5 : 0000000000000001 x4 : 0000000000000003 x3 : 0000000000000000
| x2 : 0000000000000000 x1 : 0001000000011001 x0 : 0000000000000000
| Call trace:
|  mwifiex_get_cfp+0xd8/0x15c [mwifiex]
|  mwifiex_parse_single_response_buf+0x1d0/0x504 [mwifiex]
|  mwifiex_handle_event_ext_scan_report+0x19c/0x2f8 [mwifiex]
|  mwifiex_process_sta_event+0x298/0xf0c [mwifiex]
|  mwifiex_process_event+0x110/0x238 [mwifiex]
|  mwifiex_main_process+0x428/0xa44 [mwifiex]
|  mwifiex_sdio_interrupt+0x64/0x12c [mwifiex_sdio]
|  process_sdio_pending_irqs+0x64/0x1b8
|  sdio_irq_work+0x4c/0x7c
|  process_one_work+0x148/0x2a0
|  worker_thread+0x2fc/0x40c
|  kthread+0x110/0x114
|  ret_from_fork+0x10/0x20
| Code: a94153f3 a8c37bfd d50323bf d65f03c0 (f940a000)
| ---[ end trace 0000000000000000 ]---

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Brian Norris <briannorris@chromium.org>
Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://patch.msgid.link/20240703072409.556618-1-s.hauer@pengutronix.de
---
 drivers/net/wireless/marvell/mwifiex/main.h | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h
index 175882485a19..c5164ae41b54 100644
--- a/drivers/net/wireless/marvell/mwifiex/main.h
+++ b/drivers/net/wireless/marvell/mwifiex/main.h
@@ -1287,6 +1287,9 @@ mwifiex_get_priv_by_id(struct mwifiex_adapter *adapter,
 
 	for (i = 0; i < adapter->priv_num; i++) {
 		if (adapter->priv[i]) {
+			if (adapter->priv[i]->bss_mode == NL80211_IFTYPE_UNSPECIFIED)
+				continue;
+
 			if ((adapter->priv[i]->bss_num == bss_num) &&
 			    (adapter->priv[i]->bss_type == bss_type))
 				break;
-- 
cgit v1.2.3-58-ga151


From 3588e6438a1c2e353616959ce16bce06aef0e424 Mon Sep 17 00:00:00 2001
From: Chen Ni <nichen@iscas.ac.cn>
Date: Thu, 4 Jul 2024 17:06:22 +0800
Subject: wifi: ipw2x00: Use kzalloc() instead of kmalloc()/memset()

Replace kmalloc() + memset() to kzalloc() for
better code readability and simplicity.

Signed-off-by: Chen Ni <nichen@iscas.ac.cn>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://patch.msgid.link/20240704090622.2260102-1-nichen@iscas.ac.cn
---
 drivers/net/wireless/intel/ipw2x00/libipw_tx.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/ipw2x00/libipw_tx.c b/drivers/net/wireless/intel/ipw2x00/libipw_tx.c
index 4aec1fce1ae2..e22a6732a4c3 100644
--- a/drivers/net/wireless/intel/ipw2x00/libipw_tx.c
+++ b/drivers/net/wireless/intel/ipw2x00/libipw_tx.c
@@ -180,11 +180,10 @@ static struct libipw_txb *libipw_alloc_txb(int nr_frags, int txb_size,
 	struct libipw_txb *txb;
 	int i;
 
-	txb = kmalloc(struct_size(txb, fragments, nr_frags), gfp_mask);
+	txb = kzalloc(struct_size(txb, fragments, nr_frags), gfp_mask);
 	if (!txb)
 		return NULL;
 
-	memset(txb, 0, sizeof(struct libipw_txb));
 	txb->nr_frags = nr_frags;
 	txb->frag_size = txb_size;
 
-- 
cgit v1.2.3-58-ga151


From 6c2a4c2f70e5b6c9c57d6d8adc893c55ba318613 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@gmail.com>
Date: Sat, 6 Jul 2024 15:43:24 +0900
Subject: net: tn40xx: add per queue netdev-genl stats support

Add support for the netdev-genl per queue stats API.

./tools/net/ynl/cli.py --spec Documentation/netlink/specs/netdev.yaml \
--dump qstats-get --json '{"scope":"queue"}'
[{'ifindex': 4,
  'queue-id': 0,
  'queue-type': 'rx',
  'rx-alloc-fail': 0,
  'rx-bytes': 266613,
  'rx-packets': 3325},
 {'ifindex': 4,
  'queue-id': 0,
  'queue-type': 'tx',
  'tx-bytes': 142823367,
  'tx-packets': 2387}]

Signed-off-by: FUJITA Tomonori <fujita.tomonori@gmail.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Link: https://patch.msgid.link/20240706064324.137574-1-fujita.tomonori@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/ethernet/tehuti/tn40.c | 54 ++++++++++++++++++++++++++++++++++++--
 drivers/net/ethernet/tehuti/tn40.h |  1 +
 2 files changed, 53 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/tehuti/tn40.c b/drivers/net/ethernet/tehuti/tn40.c
index 565b72537efa..259bdac24cf2 100644
--- a/drivers/net/ethernet/tehuti/tn40.c
+++ b/drivers/net/ethernet/tehuti/tn40.c
@@ -10,6 +10,7 @@
 #include <linux/pci.h>
 #include <linux/phylink.h>
 #include <linux/vmalloc.h>
+#include <net/netdev_queues.h>
 #include <net/page_pool/helpers.h>
 
 #include "tn40.h"
@@ -378,6 +379,7 @@ static int tn40_rx_receive(struct tn40_priv *priv, int budget)
 		if (!skb) {
 			u64_stats_update_begin(&priv->syncp);
 			priv->stats.rx_dropped++;
+			priv->alloc_fail++;
 			u64_stats_update_end(&priv->syncp);
 			tn40_recycle_rx_buffer(priv, rxdd);
 			break;
@@ -1580,8 +1582,55 @@ static int tn40_ethtool_get_link_ksettings(struct net_device *ndev,
 }
 
 static const struct ethtool_ops tn40_ethtool_ops = {
-	.get_link		= ethtool_op_get_link,
-	.get_link_ksettings	= tn40_ethtool_get_link_ksettings,
+	.get_link = ethtool_op_get_link,
+	.get_link_ksettings = tn40_ethtool_get_link_ksettings,
+};
+
+static void tn40_get_queue_stats_rx(struct net_device *ndev, int idx,
+				    struct netdev_queue_stats_rx *stats)
+{
+	struct tn40_priv *priv = netdev_priv(ndev);
+	unsigned int start;
+
+	do {
+		start = u64_stats_fetch_begin(&priv->syncp);
+
+		stats->packets = priv->stats.rx_packets;
+		stats->bytes = priv->stats.rx_bytes;
+		stats->alloc_fail = priv->alloc_fail;
+	} while (u64_stats_fetch_retry(&priv->syncp, start));
+}
+
+static void tn40_get_queue_stats_tx(struct net_device *ndev, int idx,
+				    struct netdev_queue_stats_tx *stats)
+{
+	struct tn40_priv *priv = netdev_priv(ndev);
+	unsigned int start;
+
+	do {
+		start = u64_stats_fetch_begin(&priv->syncp);
+
+		stats->packets = priv->stats.tx_packets;
+		stats->bytes = priv->stats.tx_bytes;
+	} while (u64_stats_fetch_retry(&priv->syncp, start));
+}
+
+static void tn40_get_base_stats(struct net_device *ndev,
+				struct netdev_queue_stats_rx *rx,
+				struct netdev_queue_stats_tx *tx)
+{
+	rx->packets = 0;
+	rx->bytes = 0;
+	rx->alloc_fail = 0;
+
+	tx->packets = 0;
+	tx->bytes = 0;
+}
+
+static const struct netdev_stat_ops tn40_stat_ops = {
+	.get_queue_stats_rx = tn40_get_queue_stats_rx,
+	.get_queue_stats_tx = tn40_get_queue_stats_tx,
+	.get_base_stats = tn40_get_base_stats,
 };
 
 static int tn40_priv_init(struct tn40_priv *priv)
@@ -1613,6 +1662,7 @@ static struct net_device *tn40_netdev_alloc(struct pci_dev *pdev)
 		return NULL;
 	ndev->netdev_ops = &tn40_netdev_ops;
 	ndev->ethtool_ops = &tn40_ethtool_ops;
+	ndev->stat_ops = &tn40_stat_ops;
 	ndev->tx_queue_len = TN40_NDEV_TXQ_LEN;
 	ndev->mem_start = pci_resource_start(pdev, 0);
 	ndev->mem_end = pci_resource_end(pdev, 0);
diff --git a/drivers/net/ethernet/tehuti/tn40.h b/drivers/net/ethernet/tehuti/tn40.h
index 10368264f7b7..490781fe5120 100644
--- a/drivers/net/ethernet/tehuti/tn40.h
+++ b/drivers/net/ethernet/tehuti/tn40.h
@@ -123,6 +123,7 @@ struct tn40_priv {
 
 	int stats_flag;
 	struct rtnl_link_stats64 stats;
+	u64 alloc_fail;
 	struct u64_stats_sync syncp;
 
 	u8 txd_size;
-- 
cgit v1.2.3-58-ga151


From 0787ab206f802deb7edbae32ba021d6b1101472a Mon Sep 17 00:00:00 2001
From: Thorsten Blum <thorsten.blum@toblux.com>
Date: Sat, 6 Jul 2024 09:11:33 +0200
Subject: udp: Remove duplicate included header file trace/events/udp.h

Remove duplicate included header file trace/events/udp.h and the
following warning reported by make includecheck:

  trace/events/udp.h is included more than once

Compile-tested only.

Signed-off-by: Thorsten Blum <thorsten.blum@toblux.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240706071132.274352-2-thorsten.blum@toblux.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/ipv6/udp.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index b5456394cc67..6602a2e9cdb5 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -46,7 +46,6 @@
 #include <net/tcp_states.h>
 #include <net/ip6_checksum.h>
 #include <net/ip6_tunnel.h>
-#include <trace/events/udp.h>
 #include <net/xfrm.h>
 #include <net/inet_hashtables.h>
 #include <net/inet6_hashtables.h>
-- 
cgit v1.2.3-58-ga151


From 870a1dbcbc2ebd2114d5f18bb0bd88a7ff07540f Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Sat, 6 Jul 2024 17:42:01 +0200
Subject: net: phy: microchip: lan937x: add support for 100BaseTX PHY

Add support of 100BaseTX PHY build in to LAN9371 and LAN9372 switches.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Michal Kubiak <michal.kubiak@intel.com>
Acked-by: Arun Ramadoss <arun.ramadoss@microchip.com>
Link: https://patch.msgid.link/20240706154201.1456098-1-o.rempel@pengutronix.de
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 drivers/net/phy/microchip.c | 126 +++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 125 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/microchip.c b/drivers/net/phy/microchip.c
index 0b88635f4fbc..d3273bc0da4a 100644
--- a/drivers/net/phy/microchip.c
+++ b/drivers/net/phy/microchip.c
@@ -12,8 +12,14 @@
 #include <linux/of.h>
 #include <dt-bindings/net/microchip-lan78xx.h>
 
+#define PHY_ID_LAN937X_TX			0x0007c190
+
+#define LAN937X_MODE_CTRL_STATUS_REG		0x11
+#define LAN937X_AUTOMDIX_EN			BIT(7)
+#define LAN937X_MDI_MODE			BIT(6)
+
 #define DRIVER_AUTHOR	"WOOJUNG HUH <woojung.huh@microchip.com>"
-#define DRIVER_DESC	"Microchip LAN88XX PHY driver"
+#define DRIVER_DESC	"Microchip LAN88XX/LAN937X TX PHY driver"
 
 struct lan88xx_priv {
 	int	chip_id;
@@ -373,6 +379,115 @@ static void lan88xx_link_change_notify(struct phy_device *phydev)
 	}
 }
 
+/**
+ * lan937x_tx_read_mdix_status - Read the MDIX status for the LAN937x TX PHY.
+ * @phydev: Pointer to the phy_device structure.
+ *
+ * This function reads the MDIX status of the LAN937x TX PHY and sets the
+ * mdix_ctrl and mdix fields of the phy_device structure accordingly.
+ * Note that MDIX status is not supported in AUTO mode, and will be set
+ * to invalid in such cases.
+ *
+ * Return: 0 on success, a negative error code on failure.
+ */
+static int lan937x_tx_read_mdix_status(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = phy_read(phydev, LAN937X_MODE_CTRL_STATUS_REG);
+	if (ret < 0)
+		return ret;
+
+	if (ret & LAN937X_AUTOMDIX_EN) {
+		phydev->mdix_ctrl = ETH_TP_MDI_AUTO;
+		/* MDI/MDIX status is unknown */
+		phydev->mdix = ETH_TP_MDI_INVALID;
+	} else if (ret & LAN937X_MDI_MODE) {
+		phydev->mdix_ctrl = ETH_TP_MDI_X;
+		phydev->mdix = ETH_TP_MDI_X;
+	} else {
+		phydev->mdix_ctrl = ETH_TP_MDI;
+		phydev->mdix = ETH_TP_MDI;
+	}
+
+	return 0;
+}
+
+/**
+ * lan937x_tx_read_status - Read the status for the LAN937x TX PHY.
+ * @phydev: Pointer to the phy_device structure.
+ *
+ * This function reads the status of the LAN937x TX PHY and updates the
+ * phy_device structure accordingly.
+ *
+ * Return: 0 on success, a negative error code on failure.
+ */
+static int lan937x_tx_read_status(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = genphy_read_status(phydev);
+	if (ret < 0)
+		return ret;
+
+	return lan937x_tx_read_mdix_status(phydev);
+}
+
+/**
+ * lan937x_tx_set_mdix - Set the MDIX mode for the LAN937x TX PHY.
+ * @phydev: Pointer to the phy_device structure.
+ *
+ * This function configures the MDIX mode of the LAN937x TX PHY based on the
+ * mdix_ctrl field of the phy_device structure. The MDIX mode can be set to
+ * MDI (straight-through), MDIX (crossover), or AUTO (auto-MDIX). If the mode
+ * is not recognized, it returns 0 without making any changes.
+ *
+ * Return: 0 on success, a negative error code on failure.
+ */
+static int lan937x_tx_set_mdix(struct phy_device *phydev)
+{
+	u16 val;
+
+	switch (phydev->mdix_ctrl) {
+	case ETH_TP_MDI:
+		val = 0;
+		break;
+	case ETH_TP_MDI_X:
+		val = LAN937X_MDI_MODE;
+		break;
+	case ETH_TP_MDI_AUTO:
+		val = LAN937X_AUTOMDIX_EN;
+		break;
+	default:
+		return 0;
+	}
+
+	return phy_modify(phydev, LAN937X_MODE_CTRL_STATUS_REG,
+			  LAN937X_AUTOMDIX_EN | LAN937X_MDI_MODE, val);
+}
+
+/**
+ * lan937x_tx_config_aneg - Configure auto-negotiation and fixed modes for the
+ *                          LAN937x TX PHY.
+ * @phydev: Pointer to the phy_device structure.
+ *
+ * This function configures the MDIX mode for the LAN937x TX PHY and then
+ * proceeds to configure the auto-negotiation or fixed mode settings
+ * based on the phy_device structure.
+ *
+ * Return: 0 on success, a negative error code on failure.
+ */
+static int lan937x_tx_config_aneg(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = lan937x_tx_set_mdix(phydev);
+	if (ret < 0)
+		return ret;
+
+	return genphy_config_aneg(phydev);
+}
+
 static struct phy_driver microchip_phy_driver[] = {
 {
 	.phy_id		= 0x0007c132,
@@ -400,12 +515,21 @@ static struct phy_driver microchip_phy_driver[] = {
 	.set_wol	= lan88xx_set_wol,
 	.read_page	= lan88xx_read_page,
 	.write_page	= lan88xx_write_page,
+},
+{
+	PHY_ID_MATCH_MODEL(PHY_ID_LAN937X_TX),
+	.name		= "Microchip LAN937x TX",
+	.suspend	= genphy_suspend,
+	.resume		= genphy_resume,
+	.config_aneg	= lan937x_tx_config_aneg,
+	.read_status	= lan937x_tx_read_status,
 } };
 
 module_phy_driver(microchip_phy_driver);
 
 static struct mdio_device_id __maybe_unused microchip_tbl[] = {
 	{ 0x0007c132, 0xfffffff2 },
+	{ PHY_ID_MATCH_MODEL(PHY_ID_LAN937X_TX) },
 	{ }
 };
 
-- 
cgit v1.2.3-58-ga151


From 3699e57aae8817b232c311229ab531ee0ec39fc8 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Sun, 7 Jul 2024 12:54:58 +0300
Subject: selftests: forwarding: Make vxlan-bridge-1d pass on debug kernels

The ageing time used by the test is too short for debug kernels and
results in entries being aged out prematurely [1].

Fix by increasing the ageing time.

The same change was done for the VLAN-aware version of the test in
commit dfbab74044be ("selftests: forwarding: Make vxlan-bridge-1q pass
on debug kernels").

[1]
 # ./vxlan_bridge_1d.sh
 [...]
 # TEST: VXLAN: flood before learning                              [ OK ]
 # TEST: VXLAN: show learned FDB entry                             [ OK ]
 # TEST: VXLAN: learned FDB entry                                  [FAIL]
 # veth3: Expected to capture 0 packets, got 4.
 # RTNETLINK answers: No such file or directory
 # TEST: VXLAN: deletion of learned FDB entry                      [ OK ]
 # TEST: VXLAN: Ageing of learned FDB entry                        [FAIL]
 # veth3: Expected to capture 0 packets, got 2.
 [...]

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240707095458.2870260-1-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/forwarding/vxlan_bridge_1d.sh | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/net/forwarding/vxlan_bridge_1d.sh b/tools/testing/selftests/net/forwarding/vxlan_bridge_1d.sh
index 6f0a2e452ba1..3f9d50f1ef9e 100755
--- a/tools/testing/selftests/net/forwarding/vxlan_bridge_1d.sh
+++ b/tools/testing/selftests/net/forwarding/vxlan_bridge_1d.sh
@@ -680,9 +680,9 @@ test_learning()
 	local mac=de:ad:be:ef:13:37
 	local dst=192.0.2.100
 
-	# Enable learning on the VxLAN device and set ageing time to 10 seconds
-	ip link set dev br1 type bridge ageing_time 1000
-	ip link set dev vx1 type vxlan ageing 10
+	# Enable learning on the VxLAN device and set ageing time to 30 seconds
+	ip link set dev br1 type bridge ageing_time 3000
+	ip link set dev vx1 type vxlan ageing 30
 	ip link set dev vx1 type vxlan learning
 	reapply_config
 
@@ -740,7 +740,7 @@ test_learning()
 
 	vxlan_flood_test $mac $dst 0 10 0
 
-	sleep 20
+	sleep 60
 
 	bridge fdb show brport vx1 | grep $mac | grep -q self
 	check_fail $?
-- 
cgit v1.2.3-58-ga151


From 0d9e699d3421b5db83eb0cb7a708882440ee78ef Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Mon, 8 Jul 2024 08:27:19 +0100
Subject: net: tls: Pass union tls_crypto_context pointer to memzero_explicit

Pass union tls_crypto_context pointer, rather than struct
tls_crypto_info pointer, to memzero_explicit().

The address of the pointer is the same before and after.
But the new construct means that the size of the dereferenced pointer type
matches the size being zeroed. Which aids static analysis.

As reported by Smatch:

  .../tls_main.c:842 do_tls_setsockopt_conf() error: memzero_explicit() 'crypto_info' too small (4 vs 56)

No functional change intended.
Compile tested only.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240708-tls-memzero-v2-1-9694eaf31b79@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/tls/tls_main.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c
index 90b7f253d363..6b4b9f2749a6 100644
--- a/net/tls/tls_main.c
+++ b/net/tls/tls_main.c
@@ -616,6 +616,7 @@ static int do_tls_setsockopt_conf(struct sock *sk, sockptr_t optval,
 	struct tls_crypto_info *alt_crypto_info;
 	struct tls_context *ctx = tls_get_ctx(sk);
 	const struct tls_cipher_desc *cipher_desc;
+	union tls_crypto_context *crypto_ctx;
 	int rc = 0;
 	int conf;
 
@@ -623,13 +624,15 @@ static int do_tls_setsockopt_conf(struct sock *sk, sockptr_t optval,
 		return -EINVAL;
 
 	if (tx) {
-		crypto_info = &ctx->crypto_send.info;
+		crypto_ctx = &ctx->crypto_send;
 		alt_crypto_info = &ctx->crypto_recv.info;
 	} else {
-		crypto_info = &ctx->crypto_recv.info;
+		crypto_ctx = &ctx->crypto_recv;
 		alt_crypto_info = &ctx->crypto_send.info;
 	}
 
+	crypto_info = &crypto_ctx->info;
+
 	/* Currently we don't support set crypto info more than one time */
 	if (TLS_CRYPTO_INFO_READY(crypto_info))
 		return -EBUSY;
@@ -710,7 +713,7 @@ static int do_tls_setsockopt_conf(struct sock *sk, sockptr_t optval,
 	return 0;
 
 err_crypto_info:
-	memzero_explicit(crypto_info, sizeof(union tls_crypto_context));
+	memzero_explicit(crypto_ctx, sizeof(*crypto_ctx));
 	return rc;
 }
 
-- 
cgit v1.2.3-58-ga151


From dd84d831ef27036031f9e11a67df1efebef78916 Mon Sep 17 00:00:00 2001
From: Frank Li <Frank.Li@nxp.com>
Date: Mon, 8 Jul 2024 14:09:48 -0400
Subject: dt-bindings: net: fsl,fman: allow dma-coherent property

Add dma-coherent property to fix below warning.
arch/arm64/boot/dts/freescale/fsl-ls1046a-rdb.dtb: fman@1a00000: 'dma-coherent', 'ptimer-handle' do not match any of the regexes: '^ethernet@[a-f0-9]+$', '^mdio@[a-f0-9]+$', '^muram@[a-f0-9]+$', '^phc@[a-f0-9]+$', '^port@[a-f0-9]+$', 'pinctrl-[0-9]+'
	from schema $id: http://devicetree.org/schemas/net/fsl,fman.yaml#

Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20240708180949.1898495-1-Frank.Li@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/fsl,fman.yaml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/fsl,fman.yaml b/Documentation/devicetree/bindings/net/fsl,fman.yaml
index 7908f67413de..f0261861f3cb 100644
--- a/Documentation/devicetree/bindings/net/fsl,fman.yaml
+++ b/Documentation/devicetree/bindings/net/fsl,fman.yaml
@@ -78,6 +78,8 @@ properties:
       - description: The first element is associated with the event interrupts.
       - description: the second element is associated with the error interrupts.
 
+  dma-coherent: true
+
   fsl,qman-channel-range:
     $ref: /schemas/types.yaml#/definitions/uint32-array
     description:
-- 
cgit v1.2.3-58-ga151


From 5618ced0197932d8a87760f430e53534ec169919 Mon Sep 17 00:00:00 2001
From: Frank Li <Frank.Li@nxp.com>
Date: Mon, 8 Jul 2024 14:09:49 -0400
Subject: dt-bindings: net: fsl,fman: add ptimer-handle property

Add ptimer-handle property to link to ptp-timer node handle.
Fix below warning:
arch/arm64/boot/dts/freescale/fsl-ls1043a-rdb.dtb: fman@1a00000: 'ptimer-handle' do not match any of the regexes: '^ethernet@[a-f0-9]+$', '^mdio@[a-f0-9]+$', '^muram@[a-f0-9]+$', '^phc@[a-f0-9]+$', '^port@[a-f0-9]+$', 'pinctrl-[0-9]+'

Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20240708180949.1898495-2-Frank.Li@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/fsl,fman.yaml | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/fsl,fman.yaml b/Documentation/devicetree/bindings/net/fsl,fman.yaml
index f0261861f3cb..9bbf39ef31a2 100644
--- a/Documentation/devicetree/bindings/net/fsl,fman.yaml
+++ b/Documentation/devicetree/bindings/net/fsl,fman.yaml
@@ -80,6 +80,10 @@ properties:
 
   dma-coherent: true
 
+  ptimer-handle:
+    $ref: /schemas/types.yaml#/definitions/phandle
+    description: see ptp/fsl,ptp.yaml
+
   fsl,qman-channel-range:
     $ref: /schemas/types.yaml#/definitions/uint32-array
     description:
-- 
cgit v1.2.3-58-ga151


From d694711355748d9b835d29f4678ae011ed4ce9b6 Mon Sep 17 00:00:00 2001
From: Guillaume La Roque <glaroque@baylibre.com>
Date: Mon, 8 Jul 2024 16:31:34 +0200
Subject: net: ti: icssg-prueth: add missing deps

Add missing dependency on NET_SWITCHDEV.

Fixes: abd5576b9c57 ("net: ti: icssg-prueth: Add support for ICSSG switch firmware")
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Guillaume La Roque <glaroque@baylibre.com>
Reviewed-by: MD Danish Anwar <danishanwar@ti.com>
Link: https://patch.msgid.link/20240708-net-deps-v2-1-b22fb74da2a3@baylibre.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/ti/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/ti/Kconfig b/drivers/net/ethernet/ti/Kconfig
index f160a3b71499..0d5a862cd78a 100644
--- a/drivers/net/ethernet/ti/Kconfig
+++ b/drivers/net/ethernet/ti/Kconfig
@@ -188,6 +188,7 @@ config TI_ICSSG_PRUETH
 	select TI_ICSS_IEP
 	select TI_K3_CPPI_DESC_POOL
 	depends on PRU_REMOTEPROC
+	depends on NET_SWITCHDEV
 	depends on ARCH_K3 && OF && TI_K3_UDMA_GLUE_LAYER
 	depends on PTP_1588_CLOCK_OPTIONAL
 	help
-- 
cgit v1.2.3-58-ga151


From 30e89baeb01fc765824ed931b33b0faf4d8d1461 Mon Sep 17 00:00:00 2001
From: Deren Wu <deren.wu@mediatek.com>
Date: Wed, 12 Jun 2024 20:01:55 -0700
Subject: wifi: mt76: mt792x: add struct mt792x_bss_conf

We are introducing a new structure, mt792x_bss_conf, to manage per-link
configurations in preparation for future STA support of Multi-Link
Operation (MLO).

The patch does not include any logic changes, only structural changes.

We have moved relevant parameters from the mt7921x_vif structure such
as to mt76 of mt76_vif structure, rssi and Tx queue parameters to
mt7921x_bss_conf structure. We can access those members to configure the
BSS for this interface, whether it is our own BSS or the one we are
associated with.

However, we have not yet created the per-link BSS configuration indexed by
link ID for Multi-Link Device (MLD) support. This step needs to be
extended before adding MLD support for AP mode.

Co-developed-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-2-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/main.c | 42 +++++++++---------
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.c  | 42 +++++++++---------
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 56 ++++++++++++------------
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c  | 56 ++++++++++++------------
 drivers/net/wireless/mediatek/mt76/mt792x.h      | 12 ++---
 drivers/net/wireless/mediatek/mt76/mt792x_core.c | 18 ++++----
 drivers/net/wireless/mediatek/mt76/mt792x_mac.c  |  2 +-
 7 files changed, 115 insertions(+), 113 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/main.c b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
index 4f30426afbb7..77999d778a38 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
@@ -295,30 +295,30 @@ mt7921_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 
 	mt792x_mutex_acquire(dev);
 
-	mvif->mt76.idx = __ffs64(~dev->mt76.vif_mask);
-	if (mvif->mt76.idx >= MT792x_MAX_INTERFACES) {
+	mvif->bss_conf.mt76.idx = __ffs64(~dev->mt76.vif_mask);
+	if (mvif->bss_conf.mt76.idx >= MT792x_MAX_INTERFACES) {
 		ret = -ENOSPC;
 		goto out;
 	}
 
-	mvif->mt76.omac_idx = mvif->mt76.idx;
+	mvif->bss_conf.mt76.omac_idx = mvif->bss_conf.mt76.idx;
 	mvif->phy = phy;
-	mvif->mt76.band_idx = 0;
-	mvif->mt76.wmm_idx = mvif->mt76.idx % MT76_CONNAC_MAX_WMM_SETS;
+	mvif->bss_conf.mt76.band_idx = 0;
+	mvif->bss_conf.mt76.wmm_idx = mvif->bss_conf.mt76.idx % MT76_CONNAC_MAX_WMM_SETS;
 
 	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.wcid,
 					  true);
 	if (ret)
 		goto out;
 
-	dev->mt76.vif_mask |= BIT_ULL(mvif->mt76.idx);
-	phy->omac_mask |= BIT_ULL(mvif->mt76.omac_idx);
+	dev->mt76.vif_mask |= BIT_ULL(mvif->bss_conf.mt76.idx);
+	phy->omac_mask |= BIT_ULL(mvif->bss_conf.mt76.omac_idx);
 
-	idx = MT792x_WTBL_RESERVED - mvif->mt76.idx;
+	idx = MT792x_WTBL_RESERVED - mvif->bss_conf.mt76.idx;
 
 	INIT_LIST_HEAD(&mvif->sta.wcid.poll_list);
 	mvif->sta.wcid.idx = idx;
-	mvif->sta.wcid.phy_idx = mvif->mt76.band_idx;
+	mvif->sta.wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
 	mvif->sta.wcid.hw_key_idx = -1;
 	mvif->sta.wcid.tx_info |= MT_WCID_TX_INFO_SET;
 	mt76_wcid_init(&mvif->sta.wcid);
@@ -326,7 +326,7 @@ mt7921_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 	mt7921_mac_wtbl_update(dev, idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
-	ewma_rssi_init(&mvif->rssi);
+	ewma_rssi_init(&mvif->bss_conf.rssi);
 
 	rcu_assign_pointer(dev->mt76.wcid[idx], &mvif->sta.wcid);
 	if (vif->txq) {
@@ -718,7 +718,7 @@ static void mt7921_bss_info_changed(struct ieee80211_hw *hw,
 	if (changed & BSS_CHANGED_ARP_FILTER) {
 		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 
-		mt76_connac_mcu_update_arp_filter(&dev->mt76, &mvif->mt76,
+		mt76_connac_mcu_update_arp_filter(&dev->mt76, &mvif->bss_conf.mt76,
 						  info);
 	}
 
@@ -803,7 +803,7 @@ int mt7921_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	msta->vif = mvif;
 	msta->wcid.sta = 1;
 	msta->wcid.idx = idx;
-	msta->wcid.phy_idx = mvif->mt76.band_idx;
+	msta->wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
 	msta->wcid.tx_info |= MT_WCID_TX_INFO_SET;
 	msta->last_txs = jiffies;
 
@@ -841,7 +841,7 @@ void mt7921_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 
 	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
 		mt76_connac_mcu_uni_add_bss(&dev->mphy, vif, &mvif->sta.wcid,
-					    true, mvif->mt76.ctx);
+					    true, mvif->bss_conf.mt76.ctx);
 
 	ewma_avg_signal_init(&msta->avg_ack_signal);
 
@@ -872,11 +872,11 @@ void mt7921_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 
 		mvif->wep_sta = NULL;
-		ewma_rssi_init(&mvif->rssi);
+		ewma_rssi_init(&mvif->bss_conf.rssi);
 		if (!sta->tdls)
 			mt76_connac_mcu_uni_add_bss(&dev->mphy, vif,
 						    &mvif->sta.wcid, false,
-						    mvif->mt76.ctx);
+						    mvif->bss_conf.mt76.ctx);
 	}
 
 	spin_lock_bh(&dev->mt76.sta_poll_lock);
@@ -1196,7 +1196,7 @@ static void mt7921_ipv6_addr_change(struct ieee80211_hw *hw,
 		struct mt76_connac_arpns_tlv arpns;
 	} req_hdr = {
 		.hdr = {
-			.bss_idx = mvif->mt76.idx,
+			.bss_idx = mvif->bss_conf.mt76.idx,
 		},
 		.arpns = {
 			.tag = cpu_to_le16(UNI_OFFLOAD_OFFLOAD_ND),
@@ -1295,7 +1295,7 @@ mt7921_start_ap(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	mt792x_mutex_acquire(dev);
 
 	err = mt76_connac_mcu_uni_add_bss(phy->mt76, vif, &mvif->sta.wcid,
-					  true, mvif->mt76.ctx);
+					  true, mvif->bss_conf.mt76.ctx);
 	if (err)
 		goto out;
 
@@ -1327,7 +1327,7 @@ mt7921_stop_ap(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		goto out;
 
 	mt76_connac_mcu_uni_add_bss(phy->mt76, vif, &mvif->sta.wcid, false,
-				    mvif->mt76.ctx);
+				    mvif->bss_conf.mt76.ctx);
 
 out:
 	mt792x_mutex_release(dev);
@@ -1352,13 +1352,13 @@ static void mt7921_ctx_iter(void *priv, u8 *mac,
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct ieee80211_chanctx_conf *ctx = priv;
 
-	if (ctx != mvif->mt76.ctx)
+	if (ctx != mvif->bss_conf.mt76.ctx)
 		return;
 
 	if (vif->type == NL80211_IFTYPE_MONITOR)
 		mt7921_mcu_config_sniffer(mvif, ctx);
 	else
-		mt76_connac_mcu_uni_set_chctx(mvif->phy->mt76, &mvif->mt76, ctx);
+		mt76_connac_mcu_uni_set_chctx(mvif->phy->mt76, &mvif->bss_conf.mt76, ctx);
 }
 
 static void
@@ -1385,7 +1385,7 @@ static void mt7921_mgd_prepare_tx(struct ieee80211_hw *hw,
 		       jiffies_to_msecs(HZ);
 
 	mt792x_mutex_acquire(dev);
-	mt7921_set_roc(mvif->phy, mvif, mvif->mt76.ctx->def.chan, duration,
+	mt7921_set_roc(mvif->phy, mvif, mvif->bss_conf.mt76.ctx->def.chan, duration,
 		       MT7921_ROC_REQ_JOIN);
 	mt792x_mutex_release(dev);
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
index bdd8b5f19b24..0453af012016 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
@@ -105,7 +105,7 @@ mt7921_mcu_set_ipv6_ns_filter(struct mt76_dev *dev,
 		struct mt76_connac_arpns_tlv arpns;
 	} req = {
 		.hdr = {
-			.bss_idx = mvif->mt76.idx,
+			.bss_idx = mvif->bss_conf.mt76.idx,
 		},
 		.arpns = {
 			.tag = cpu_to_le16(UNI_OFFLOAD_OFFLOAD_ND),
@@ -260,7 +260,7 @@ mt7921_mcu_rssi_monitor_iter(void *priv, u8 *mac,
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt76_connac_rssi_notify_event *event = priv;
 	enum nl80211_cqm_rssi_threshold_event nl_event;
-	s32 rssi = le32_to_cpu(event->rssi[mvif->mt76.idx]);
+	s32 rssi = le32_to_cpu(event->rssi[mvif->bss_conf.mt76.idx]);
 
 	if (!rssi)
 		return;
@@ -388,7 +388,7 @@ int mt7921_mcu_uni_tx_ba(struct mt792x_dev *dev,
 	if (enable && !params->amsdu)
 		msta->wcid.amsdu = false;
 
-	return mt76_connac_mcu_sta_ba(&dev->mt76, &msta->vif->mt76, params,
+	return mt76_connac_mcu_sta_ba(&dev->mt76, &msta->vif->bss_conf.mt76, params,
 				      MCU_UNI_CMD(STA_REC_UPDATE),
 				      enable, true);
 }
@@ -399,7 +399,7 @@ int mt7921_mcu_uni_rx_ba(struct mt792x_dev *dev,
 {
 	struct mt792x_sta *msta = (struct mt792x_sta *)params->sta->drv_priv;
 
-	return mt76_connac_mcu_sta_ba(&dev->mt76, &msta->vif->mt76, params,
+	return mt76_connac_mcu_sta_ba(&dev->mt76, &msta->vif->bss_conf.mt76, params,
 				      MCU_UNI_CMD(STA_REC_UPDATE),
 				      enable, false);
 }
@@ -678,9 +678,9 @@ int mt7921_mcu_set_tx(struct mt792x_dev *dev, struct ieee80211_vif *vif)
 		u8 wmm_idx;
 		u8 pad;
 	} __packed req = {
-		.bss_idx = mvif->mt76.idx,
+		.bss_idx = mvif->bss_conf.mt76.idx,
 		.qos = vif->bss_conf.qos,
-		.wmm_idx = mvif->mt76.wmm_idx,
+		.wmm_idx = mvif->bss_conf.mt76.wmm_idx,
 	};
 	struct mu_edca {
 		u8 cw_min;
@@ -701,15 +701,15 @@ int mt7921_mcu_set_tx(struct mt792x_dev *dev, struct ieee80211_vif *vif)
 		struct mu_edca edca[IEEE80211_NUM_ACS];
 		u8 pad3[32];
 	} __packed req_mu = {
-		.bss_idx = mvif->mt76.idx,
+		.bss_idx = mvif->bss_conf.mt76.idx,
 		.qos = vif->bss_conf.qos,
-		.wmm_idx = mvif->mt76.wmm_idx,
+		.wmm_idx = mvif->bss_conf.mt76.wmm_idx,
 	};
 	static const int to_aci[] = { 1, 0, 2, 3 };
 	int ac, ret;
 
 	for (ac = 0; ac < IEEE80211_NUM_ACS; ac++) {
-		struct ieee80211_tx_queue_params *q = &mvif->queue_params[ac];
+		struct ieee80211_tx_queue_params *q = &mvif->bss_conf.queue_params[ac];
 		struct edca *e = &req.edca[to_aci[ac]];
 
 		e->aifs = cpu_to_le16(q->aifs);
@@ -738,10 +738,10 @@ int mt7921_mcu_set_tx(struct mt792x_dev *dev, struct ieee80211_vif *vif)
 		struct ieee80211_he_mu_edca_param_ac_rec *q;
 		struct mu_edca *e;
 
-		if (!mvif->queue_params[ac].mu_edca)
+		if (!mvif->bss_conf.queue_params[ac].mu_edca)
 			break;
 
-		q = &mvif->queue_params[ac].mu_edca_param_rec;
+		q = &mvif->bss_conf.queue_params[ac].mu_edca_param_rec;
 		e = &(req_mu.edca[to_aci[ac]]);
 
 		e->cw_min = q->ecw_min_max & 0xf;
@@ -790,7 +790,7 @@ int mt7921_mcu_set_roc(struct mt792x_phy *phy, struct mt792x_vif *vif,
 			.tokenid = token_id,
 			.reqtype = type,
 			.maxinterval = cpu_to_le32(duration),
-			.bss_idx = vif->mt76.idx,
+			.bss_idx = vif->bss_conf.mt76.idx,
 			.control_channel = chan->hw_value,
 			.bw = CMD_CBW_20MHZ,
 			.bw_from_ap = CMD_CBW_20MHZ,
@@ -842,7 +842,7 @@ int mt7921_mcu_abort_roc(struct mt792x_phy *phy, struct mt792x_vif *vif,
 			.tag = cpu_to_le16(UNI_ROC_ABORT),
 			.len = cpu_to_le16(sizeof(struct roc_abort_tlv)),
 			.tokenid = token_id,
-			.bss_idx = vif->mt76.idx,
+			.bss_idx = vif->bss_conf.mt76.idx,
 			.dbdcband = 0xff, /* auto*/
 		},
 	};
@@ -947,7 +947,7 @@ int mt7921_mcu_uni_bss_ps(struct mt792x_dev *dev, struct ieee80211_vif *vif)
 		} __packed ps;
 	} __packed ps_req = {
 		.hdr = {
-			.bss_idx = mvif->mt76.idx,
+			.bss_idx = mvif->bss_conf.mt76.idx,
 		},
 		.ps = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_PS),
@@ -982,7 +982,7 @@ mt7921_mcu_uni_bss_bcnft(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 		} __packed bcnft;
 	} __packed bcnft_req = {
 		.hdr = {
-			.bss_idx = mvif->mt76.idx,
+			.bss_idx = mvif->bss_conf.mt76.idx,
 		},
 		.bcnft = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_BCNFT),
@@ -1015,7 +1015,7 @@ mt7921_mcu_set_bss_pm(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 		u8 bmc_triggered_ac;
 		u8 pad;
 	} req = {
-		.bss_idx = mvif->mt76.idx,
+		.bss_idx = mvif->bss_conf.mt76.idx,
 		.aid = cpu_to_le16(vif->cfg.aid),
 		.dtim_period = vif->bss_conf.dtim_period,
 		.bcn_interval = cpu_to_le16(vif->bss_conf.beacon_int),
@@ -1024,7 +1024,7 @@ mt7921_mcu_set_bss_pm(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 		u8 bss_idx;
 		u8 pad[3];
 	} req_hdr = {
-		.bss_idx = mvif->mt76.idx,
+		.bss_idx = mvif->bss_conf.mt76.idx,
 	};
 	int err;
 
@@ -1042,7 +1042,7 @@ int mt7921_mcu_sta_update(struct mt792x_dev *dev, struct ieee80211_sta *sta,
 			  enum mt76_sta_info_state state)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	int rssi = -ewma_rssi_read(&mvif->rssi);
+	int rssi = -ewma_rssi_read(&mvif->bss_conf.rssi);
 	struct mt76_sta_cmd_info info = {
 		.sta = sta,
 		.vif = vif,
@@ -1190,7 +1190,7 @@ int mt7921_mcu_config_sniffer(struct mt792x_vif *vif,
 		} __packed tlv;
 	} __packed req = {
 		.hdr = {
-			.band_idx = vif->mt76.band_idx,
+			.band_idx = vif->bss_conf.mt76.band_idx,
 		},
 		.tlv = {
 			.tag = cpu_to_le16(1),
@@ -1251,7 +1251,7 @@ mt7921_mcu_uni_add_beacon_offload(struct mt792x_dev *dev,
 		} __packed beacon_tlv;
 	} req = {
 		.hdr = {
-			.bss_idx = mvif->mt76.idx,
+			.bss_idx = mvif->bss_conf.mt76.idx,
 		},
 		.beacon_tlv = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_BCN_CONTENT),
@@ -1460,7 +1460,7 @@ int mt7921_mcu_set_rssimonitor(struct mt792x_dev *dev, struct ieee80211_vif *vif
 		.enable = vif->cfg.assoc,
 		.cqm_rssi_high = vif->bss_conf.cqm_rssi_thold + vif->bss_conf.cqm_rssi_hyst,
 		.cqm_rssi_low = vif->bss_conf.cqm_rssi_thold - vif->bss_conf.cqm_rssi_hyst,
-		.bss_idx = mvif->mt76.idx,
+		.bss_idx = mvif->bss_conf.mt76.idx,
 	};
 
 	return mt76_mcu_send_msg(&dev->mt76, MCU_CE_CMD(RSSI_MONITOR),
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 6179798a8845..dfc606cc9bdd 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -328,35 +328,35 @@ mt7925_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 
 	mt792x_mutex_acquire(dev);
 
-	mvif->mt76.idx = __ffs64(~dev->mt76.vif_mask);
-	if (mvif->mt76.idx >= MT792x_MAX_INTERFACES) {
+	mvif->bss_conf.mt76.idx = __ffs64(~dev->mt76.vif_mask);
+	if (mvif->bss_conf.mt76.idx >= MT792x_MAX_INTERFACES) {
 		ret = -ENOSPC;
 		goto out;
 	}
 
-	mvif->mt76.omac_idx = mvif->mt76.idx;
+	mvif->bss_conf.mt76.omac_idx = mvif->bss_conf.mt76.idx;
 	mvif->phy = phy;
-	mvif->mt76.band_idx = 0;
-	mvif->mt76.wmm_idx = mvif->mt76.idx % MT76_CONNAC_MAX_WMM_SETS;
+	mvif->bss_conf.mt76.band_idx = 0;
+	mvif->bss_conf.mt76.wmm_idx = mvif->bss_conf.mt76.idx % MT76_CONNAC_MAX_WMM_SETS;
 
 	if (phy->mt76->chandef.chan->band != NL80211_BAND_2GHZ)
-		mvif->mt76.basic_rates_idx = MT792x_BASIC_RATES_TBL + 4;
+		mvif->bss_conf.mt76.basic_rates_idx = MT792x_BASIC_RATES_TBL + 4;
 	else
-		mvif->mt76.basic_rates_idx = MT792x_BASIC_RATES_TBL;
+		mvif->bss_conf.mt76.basic_rates_idx = MT792x_BASIC_RATES_TBL;
 
 	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.wcid,
 					  true);
 	if (ret)
 		goto out;
 
-	dev->mt76.vif_mask |= BIT_ULL(mvif->mt76.idx);
-	phy->omac_mask |= BIT_ULL(mvif->mt76.omac_idx);
+	dev->mt76.vif_mask |= BIT_ULL(mvif->bss_conf.mt76.idx);
+	phy->omac_mask |= BIT_ULL(mvif->bss_conf.mt76.omac_idx);
 
-	idx = MT792x_WTBL_RESERVED - mvif->mt76.idx;
+	idx = MT792x_WTBL_RESERVED - mvif->bss_conf.mt76.idx;
 
 	INIT_LIST_HEAD(&mvif->sta.wcid.poll_list);
 	mvif->sta.wcid.idx = idx;
-	mvif->sta.wcid.phy_idx = mvif->mt76.band_idx;
+	mvif->sta.wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
 	mvif->sta.wcid.hw_key_idx = -1;
 	mvif->sta.wcid.tx_info |= MT_WCID_TX_INFO_SET;
 	mvif->sta.vif = mvif;
@@ -365,7 +365,7 @@ mt7925_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 	mt7925_mac_wtbl_update(dev, idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
-	ewma_rssi_init(&mvif->rssi);
+	ewma_rssi_init(&mvif->bss_conf.rssi);
 
 	rcu_assign_pointer(dev->mt76.wcid[idx], &mvif->sta.wcid);
 	if (vif->txq) {
@@ -524,11 +524,11 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 
 	mt792x_mutex_acquire(dev);
 
-	if (cmd == SET_KEY && !mvif->mt76.cipher) {
+	if (cmd == SET_KEY && !mvif->bss_conf.mt76.cipher) {
 		struct mt792x_phy *phy = mt792x_hw_phy(hw);
 
-		mvif->mt76.cipher = mt7925_mcu_get_cipher(key->cipher);
-		mt7925_mcu_add_bss_info(phy, mvif->mt76.ctx, vif, sta, true);
+		mvif->bss_conf.mt76.cipher = mt7925_mcu_get_cipher(key->cipher);
+		mt7925_mcu_add_bss_info(phy, mvif->bss_conf.mt76.ctx, vif, sta, true);
 	}
 
 	if (cmd == SET_KEY)
@@ -748,7 +748,7 @@ static void mt7925_bss_info_changed(struct ieee80211_hw *hw,
 	if (changed & BSS_CHANGED_ARP_FILTER) {
 		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 
-		mt7925_mcu_update_arp_filter(&dev->mt76, &mvif->mt76, info);
+		mt7925_mcu_update_arp_filter(&dev->mt76, &mvif->bss_conf.mt76, info);
 	}
 
 	mt792x_mutex_release(dev);
@@ -770,7 +770,7 @@ int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	msta->vif = mvif;
 	msta->wcid.sta = 1;
 	msta->wcid.idx = idx;
-	msta->wcid.phy_idx = mvif->mt76.band_idx;
+	msta->wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
 	msta->wcid.tx_info |= MT_WCID_TX_INFO_SET;
 	msta->last_txs = jiffies;
 
@@ -786,7 +786,7 @@ int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 
 	/* should update bss info before STA add */
 	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
-		mt7925_mcu_add_bss_info(&dev->phy, mvif->mt76.ctx, vif, sta,
+		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, vif, sta,
 					false);
 
 	ret = mt7925_mcu_sta_update(dev, sta, vif, true,
@@ -810,7 +810,7 @@ void mt7925_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	mt792x_mutex_acquire(dev);
 
 	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
-		mt7925_mcu_add_bss_info(&dev->phy, mvif->mt76.ctx, vif, sta,
+		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, vif, sta,
 					true);
 
 	ewma_avg_signal_init(&msta->avg_ack_signal);
@@ -842,9 +842,9 @@ void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 
 		mvif->wep_sta = NULL;
-		ewma_rssi_init(&mvif->rssi);
+		ewma_rssi_init(&mvif->bss_conf.rssi);
 		if (!sta->tdls)
-			mt7925_mcu_add_bss_info(&dev->phy, mvif->mt76.ctx, vif, sta,
+			mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, vif, sta,
 						false);
 	}
 
@@ -1190,7 +1190,7 @@ static void mt7925_ipv6_addr_change(struct ieee80211_hw *hw,
 		struct in6_addr ns_addrs[IEEE80211_BSS_ARP_ADDR_LIST_LEN];
 	} req_hdr = {
 		.hdr = {
-			.bss_idx = mvif->mt76.idx,
+			.bss_idx = mvif->bss_conf.mt76.idx,
 		},
 		.arpns = {
 			.tag = cpu_to_le16(UNI_OFFLOAD_OFFLOAD_ND),
@@ -1288,7 +1288,7 @@ mt7925_conf_tx(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	};
 
 	/* firmware uses access class index */
-	mvif->queue_params[mq_to_aci[queue]] = *params;
+	mvif->bss_conf.queue_params[mq_to_aci[queue]] = *params;
 
 	return 0;
 }
@@ -1303,7 +1303,7 @@ mt7925_start_ap(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 
 	mt792x_mutex_acquire(dev);
 
-	err = mt7925_mcu_add_bss_info(&dev->phy, mvif->mt76.ctx, vif, NULL,
+	err = mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, vif, NULL,
 				      true);
 	if (err)
 		goto out;
@@ -1334,7 +1334,7 @@ mt7925_stop_ap(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	if (err)
 		goto out;
 
-	mt7925_mcu_add_bss_info(&dev->phy, mvif->mt76.ctx, vif, NULL,
+	mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, vif, NULL,
 				false);
 
 out:
@@ -1360,14 +1360,14 @@ static void mt7925_ctx_iter(void *priv, u8 *mac,
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct ieee80211_chanctx_conf *ctx = priv;
 
-	if (ctx != mvif->mt76.ctx)
+	if (ctx != mvif->bss_conf.mt76.ctx)
 		return;
 
 	if (vif->type == NL80211_IFTYPE_MONITOR) {
 		mt7925_mcu_set_sniffer(mvif->phy->dev, vif, true);
 		mt7925_mcu_config_sniffer(mvif, ctx);
 	} else {
-		mt7925_mcu_set_chctx(mvif->phy->mt76, &mvif->mt76, ctx);
+		mt7925_mcu_set_chctx(mvif->phy->mt76, &mvif->bss_conf.mt76, ctx);
 	}
 }
 
@@ -1395,7 +1395,7 @@ static void mt7925_mgd_prepare_tx(struct ieee80211_hw *hw,
 		       jiffies_to_msecs(HZ);
 
 	mt792x_mutex_acquire(dev);
-	mt7925_set_roc(mvif->phy, mvif, mvif->mt76.ctx->def.chan, duration,
+	mt7925_set_roc(mvif->phy, mvif, mvif->bss_conf.mt76.ctx->def.chan, duration,
 		       MT7925_ROC_REQ_JOIN);
 	mt792x_mutex_release(dev);
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 652a9accc43c..c259aedb96e0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -546,7 +546,7 @@ int mt7925_mcu_uni_tx_ba(struct mt792x_dev *dev,
 	if (enable && !params->amsdu)
 		msta->wcid.amsdu = false;
 
-	return mt7925_mcu_sta_ba(&dev->mt76, &mvif->mt76, params,
+	return mt7925_mcu_sta_ba(&dev->mt76, &mvif->bss_conf.mt76, params,
 				 enable, true);
 }
 
@@ -557,7 +557,7 @@ int mt7925_mcu_uni_rx_ba(struct mt792x_dev *dev,
 	struct mt792x_sta *msta = (struct mt792x_sta *)params->sta->drv_priv;
 	struct mt792x_vif *mvif = msta->vif;
 
-	return mt7925_mcu_sta_ba(&dev->mt76, &mvif->mt76, params,
+	return mt7925_mcu_sta_ba(&dev->mt76, &mvif->bss_conf.mt76, params,
 				 enable, false);
 }
 
@@ -889,7 +889,7 @@ int mt7925_mcu_wtbl_update_hdr_trans(struct mt792x_dev *dev,
 
 	msta = sta ? (struct mt792x_sta *)sta->drv_priv : &mvif->sta;
 
-	skb = __mt76_connac_mcu_alloc_sta_req(&dev->mt76, &mvif->mt76,
+	skb = __mt76_connac_mcu_alloc_sta_req(&dev->mt76, &mvif->bss_conf.mt76,
 					      &msta->wcid,
 					      MT7925_STA_UPDATE_MAX_SIZE);
 	if (IS_ERR(skb))
@@ -915,7 +915,7 @@ int mt7925_mcu_set_tx(struct mt792x_dev *dev, struct ieee80211_vif *vif)
 		u8 bss_idx;
 		u8 __rsv[3];
 	} __packed hdr = {
-		.bss_idx = mvif->mt76.idx,
+		.bss_idx = mvif->bss_conf.mt76.idx,
 	};
 	struct sk_buff *skb;
 	int len = sizeof(hdr) + IEEE80211_NUM_ACS * sizeof(struct edca);
@@ -928,7 +928,7 @@ int mt7925_mcu_set_tx(struct mt792x_dev *dev, struct ieee80211_vif *vif)
 	skb_put_data(skb, &hdr, sizeof(hdr));
 
 	for (ac = 0; ac < IEEE80211_NUM_ACS; ac++) {
-		struct ieee80211_tx_queue_params *q = &mvif->queue_params[ac];
+		struct ieee80211_tx_queue_params *q = &mvif->bss_conf.queue_params[ac];
 		struct edca *e;
 		struct tlv *tlv;
 
@@ -976,7 +976,7 @@ mt7925_mcu_sta_key_tlv(struct mt76_wcid *wcid,
 
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_KEY_V3, sizeof(*sec));
 	sec = (struct sta_rec_sec_uni *)tlv;
-	sec->bss_idx = mvif->mt76.idx;
+	sec->bss_idx = mvif->bss_conf.mt76.idx;
 	sec->is_authenticator = 0;
 	sec->mgmt_prot = 0;
 	sec->wlan_idx = (u8)wcid->idx;
@@ -1085,7 +1085,7 @@ int mt7925_mcu_set_roc(struct mt792x_phy *phy, struct mt792x_vif *vif,
 			.tokenid = token_id,
 			.reqtype = type,
 			.maxinterval = cpu_to_le32(duration),
-			.bss_idx = vif->mt76.idx,
+			.bss_idx = vif->bss_conf.mt76.idx,
 			.control_channel = chan->hw_value,
 			.bw = CMD_CBW_20MHZ,
 			.bw_from_ap = CMD_CBW_20MHZ,
@@ -1137,7 +1137,7 @@ int mt7925_mcu_abort_roc(struct mt792x_phy *phy, struct mt792x_vif *vif,
 			.tag = cpu_to_le16(UNI_ROC_ABORT),
 			.len = cpu_to_le16(sizeof(struct roc_abort_tlv)),
 			.tokenid = token_id,
-			.bss_idx = vif->mt76.idx,
+			.bss_idx = vif->bss_conf.mt76.idx,
 			.dbdcband = 0xff, /* auto*/
 		},
 	};
@@ -1263,7 +1263,7 @@ int mt7925_mcu_uni_bss_ps(struct mt792x_dev *dev, struct ieee80211_vif *vif)
 		} __packed ps;
 	} __packed ps_req = {
 		.hdr = {
-			.bss_idx = mvif->mt76.idx,
+			.bss_idx = mvif->bss_conf.mt76.idx,
 		},
 		.ps = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_PS),
@@ -1300,7 +1300,7 @@ mt7925_mcu_uni_bss_bcnft(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 		} __packed bcnft;
 	} __packed bcnft_req = {
 		.hdr = {
-			.bss_idx = mvif->mt76.idx,
+			.bss_idx = mvif->bss_conf.mt76.idx,
 		},
 		.bcnft = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_BCNFT),
@@ -1338,7 +1338,7 @@ mt7925_mcu_set_bss_pm(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 		} __packed enable;
 	} req = {
 		.hdr = {
-			.bss_idx = mvif->mt76.idx,
+			.bss_idx = mvif->bss_conf.mt76.idx,
 		},
 		.enable = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_BCNFT),
@@ -1358,7 +1358,7 @@ mt7925_mcu_set_bss_pm(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 		} __packed disable;
 	} req1 = {
 		.hdr = {
-			.bss_idx = mvif->mt76.idx,
+			.bss_idx = mvif->bss_conf.mt76.idx,
 		},
 		.disable = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_PM_DISABLE),
@@ -1502,7 +1502,7 @@ mt7925_mcu_sta_phy_tlv(struct sk_buff *skb,
 		       struct ieee80211_vif *vif, struct ieee80211_sta *sta)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	struct cfg80211_chan_def *chandef = &mvif->mt76.ctx->def;
+	struct cfg80211_chan_def *chandef = &mvif->bss_conf.mt76.ctx->def;
 	struct sta_rec_phy *phy;
 	struct tlv *tlv;
 	u8 af = 0, mm = 0;
@@ -1569,7 +1569,7 @@ mt7925_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb,
 			     struct ieee80211_vif *vif, struct ieee80211_sta *sta)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	struct cfg80211_chan_def *chandef = &mvif->mt76.ctx->def;
+	struct cfg80211_chan_def *chandef = &mvif->bss_conf.mt76.ctx->def;
 	enum nl80211_band band = chandef->chan->band;
 	struct sta_rec_ra_info *ra_info;
 	struct tlv *tlv;
@@ -1655,7 +1655,7 @@ int mt7925_mcu_sta_update(struct mt792x_dev *dev, struct ieee80211_sta *sta,
 			  enum mt76_sta_info_state state)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	int rssi = -ewma_rssi_read(&mvif->rssi);
+	int rssi = -ewma_rssi_read(&mvif->bss_conf.rssi);
 	struct mt76_sta_cmd_info info = {
 		.sta = sta,
 		.vif = vif,
@@ -1746,7 +1746,7 @@ int mt7925_mcu_set_sniffer(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 		} __packed enable;
 	} __packed req = {
 		.hdr = {
-			.band_idx = mvif->mt76.band_idx,
+			.band_idx = mvif->bss_conf.mt76.band_idx,
 		},
 		.enable = {
 			.tag = cpu_to_le16(UNI_SNIFFER_ENABLE),
@@ -1805,7 +1805,7 @@ int mt7925_mcu_config_sniffer(struct mt792x_vif *vif,
 		} __packed tlv;
 	} __packed req = {
 		.hdr = {
-			.band_idx = vif->mt76.band_idx,
+			.band_idx = vif->bss_conf.mt76.band_idx,
 		},
 		.tlv = {
 			.tag = cpu_to_le16(UNI_SNIFFER_CONFIG),
@@ -1866,7 +1866,7 @@ mt7925_mcu_uni_add_beacon_offload(struct mt792x_dev *dev,
 		} __packed beacon_tlv;
 	} req = {
 		.hdr = {
-			.bss_idx = mvif->mt76.idx,
+			.bss_idx = mvif->bss_conf.mt76.idx,
 		},
 		.beacon_tlv = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_BCN_CONTENT),
@@ -2080,8 +2080,8 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 	tlv = mt76_connac_mcu_add_tlv(skb, UNI_BSS_INFO_BASIC, sizeof(*basic_req));
 	basic_req = (struct mt76_connac_bss_basic_tlv *)tlv;
 
-	idx = mvif->mt76.omac_idx > EXT_BSSID_START ? HW_BSSID_0 :
-						      mvif->mt76.omac_idx;
+	idx = mvif->bss_conf.mt76.omac_idx > EXT_BSSID_START ? HW_BSSID_0 :
+						      mvif->bss_conf.mt76.omac_idx;
 	basic_req->hw_bss_idx = idx;
 
 	basic_req->phymode_ext = mt7925_get_phy_mode_ext(phy, vif, band, sta);
@@ -2097,9 +2097,9 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 	basic_req->dtim_period = vif->bss_conf.dtim_period;
 	basic_req->bmc_tx_wlan_idx = cpu_to_le16(wlan_idx);
 	basic_req->sta_idx = cpu_to_le16(msta->wcid.idx);
-	basic_req->omac_idx = mvif->mt76.omac_idx;
-	basic_req->band_idx = mvif->mt76.band_idx;
-	basic_req->wmm_idx = mvif->mt76.wmm_idx;
+	basic_req->omac_idx = mvif->bss_conf.mt76.omac_idx;
+	basic_req->band_idx = mvif->bss_conf.mt76.band_idx;
+	basic_req->wmm_idx = mvif->bss_conf.mt76.wmm_idx;
 	basic_req->conn_state = !enable;
 
 	switch (vif->type) {
@@ -2217,8 +2217,8 @@ mt7925_mcu_bss_mld_tlv(struct sk_buff *skb,
 	mld = (struct bss_mld_tlv *)tlv;
 
 	mld->link_id = sta ? (is_mld ? vif->bss_conf.link_id : 0) : 0xff;
-	mld->group_mld_id = is_mld ? mvif->mt76.idx : 0xff;
-	mld->own_mld_id = mvif->mt76.idx + 32;
+	mld->group_mld_id = is_mld ? mvif->bss_conf.mt76.idx : 0xff;
+	mld->own_mld_id = mvif->bss_conf.mt76.idx + 32;
 	mld->remap_idx = 0xff;
 
 	if (sta)
@@ -2301,7 +2301,7 @@ int mt7925_mcu_set_timing(struct mt792x_phy *phy,
 	struct mt792x_dev *dev = phy->dev;
 	struct sk_buff *skb;
 
-	skb = __mt7925_mcu_alloc_bss_req(&dev->mt76, &mvif->mt76,
+	skb = __mt7925_mcu_alloc_bss_req(&dev->mt76, &mvif->bss_conf.mt76,
 					 MT7925_BSS_UPDATE_MAX_SIZE);
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
@@ -2323,7 +2323,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 	struct sk_buff *skb;
 	int err;
 
-	skb = __mt7925_mcu_alloc_bss_req(&dev->mt76, &mvif->mt76,
+	skb = __mt7925_mcu_alloc_bss_req(&dev->mt76, &mvif->bss_conf.mt76,
 					 MT7925_BSS_UPDATE_MAX_SIZE);
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
@@ -2348,7 +2348,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 	if (err < 0)
 		return err;
 
-	return mt7925_mcu_set_chctx(phy->mt76, &mvif->mt76, ctx);
+	return mt7925_mcu_set_chctx(phy->mt76, &mvif->bss_conf.mt76, ctx);
 }
 
 int mt7925_mcu_set_dbdc(struct mt76_phy *phy)
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index cf14a38c5e72..27bfda04d0f2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -98,17 +98,19 @@ struct mt792x_sta {
 
 DECLARE_EWMA(rssi, 10, 8);
 
-struct mt792x_vif {
+struct mt792x_bss_conf {
 	struct mt76_vif mt76; /* must be first */
+	struct ewma_rssi rssi;
+	struct ieee80211_tx_queue_params queue_params[IEEE80211_NUM_ACS];
+};
+
+struct mt792x_vif {
+	struct mt792x_bss_conf bss_conf; /* must be first */
 
 	struct mt792x_sta sta;
 	struct mt792x_sta *wep_sta;
 
 	struct mt792x_phy *phy;
-
-	struct ewma_rssi rssi;
-
-	struct ieee80211_tx_queue_params queue_params[IEEE80211_NUM_ACS];
 };
 
 struct mt792x_phy {
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_core.c b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
index 4adca99eb9b8..9eeaf3344217 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_core.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
@@ -128,8 +128,8 @@ void mt792x_remove_interface(struct ieee80211_hw *hw,
 
 	rcu_assign_pointer(dev->mt76.wcid[idx], NULL);
 
-	dev->mt76.vif_mask &= ~BIT_ULL(mvif->mt76.idx);
-	phy->omac_mask &= ~BIT_ULL(mvif->mt76.omac_idx);
+	dev->mt76.vif_mask &= ~BIT_ULL(mvif->bss_conf.mt76.idx);
+	phy->omac_mask &= ~BIT_ULL(mvif->bss_conf.mt76.omac_idx);
 	mt792x_mutex_release(dev);
 
 	spin_lock_bh(&dev->mt76.sta_poll_lock);
@@ -149,7 +149,7 @@ int mt792x_conf_tx(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 
 	/* no need to update right away, we'll get BSS_CHANGED_QOS */
 	queue = mt76_connac_lmac_mapping(queue);
-	mvif->queue_params[queue] = *params;
+	mvif->bss_conf.queue_params[queue] = *params;
 
 	return 0;
 }
@@ -178,7 +178,7 @@ u64 mt792x_get_tsf(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
-	u8 omac_idx = mvif->mt76.omac_idx;
+	u8 omac_idx = mvif->bss_conf.mt76.omac_idx;
 	union {
 		u64 t64;
 		u32 t32[2];
@@ -204,7 +204,7 @@ void mt792x_set_tsf(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
-	u8 omac_idx = mvif->mt76.omac_idx;
+	u8 omac_idx = mvif->bss_conf.mt76.omac_idx;
 	union {
 		u64 t64;
 		u32 t32[2];
@@ -265,7 +265,7 @@ int mt792x_assign_vif_chanctx(struct ieee80211_hw *hw,
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
 
 	mutex_lock(&dev->mt76.mutex);
-	mvif->mt76.ctx = ctx;
+	mvif->bss_conf.mt76.ctx = ctx;
 	mutex_unlock(&dev->mt76.mutex);
 
 	return 0;
@@ -281,7 +281,7 @@ void mt792x_unassign_vif_chanctx(struct ieee80211_hw *hw,
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
 
 	mutex_lock(&dev->mt76.mutex);
-	mvif->mt76.ctx = NULL;
+	mvif->bss_conf.mt76.ctx = NULL;
 	mutex_unlock(&dev->mt76.mutex);
 }
 EXPORT_SYMBOL_GPL(mt792x_unassign_vif_chanctx);
@@ -405,7 +405,7 @@ mt792x_ethtool_worker(void *wi_data, struct ieee80211_sta *sta)
 	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 	struct mt76_ethtool_worker_info *wi = wi_data;
 
-	if (msta->vif->mt76.idx != wi->idx)
+	if (msta->vif->bss_conf.mt76.idx != wi->idx)
 		return;
 
 	mt76_ethtool_worker(wi, &msta->wcid.stats, true);
@@ -421,7 +421,7 @@ void mt792x_get_et_stats(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	struct mt76_mib_stats *mib = &phy->mib;
 	struct mt76_ethtool_worker_info wi = {
 		.data = data,
-		.idx = mvif->mt76.idx,
+		.idx = mvif->bss_conf.mt76.idx,
 	};
 	int i, ei = 0;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_mac.c b/drivers/net/wireless/mediatek/mt76/mt792x_mac.c
index eb29434abee1..45dd94ff386d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_mac.c
@@ -173,7 +173,7 @@ mt792x_mac_rssi_iter(void *priv, u8 *mac, struct ieee80211_vif *vif)
 	if (!ether_addr_equal(vif->addr, hdr->addr1))
 		return;
 
-	ewma_rssi_add(&mvif->rssi, -status->signal);
+	ewma_rssi_add(&mvif->bss_conf.rssi, -status->signal);
 }
 
 void mt792x_mac_assoc_rssi(struct mt792x_dev *dev, struct sk_buff *skb)
-- 
cgit v1.2.3-58-ga151


From 4c28c0976ed83c3e8240e80ea918c4d1a3f9c5bc Mon Sep 17 00:00:00 2001
From: Deren Wu <deren.wu@mediatek.com>
Date: Wed, 12 Jun 2024 20:01:56 -0700
Subject: wifi: mt76: mt792x: add struct mt792x_link_sta

We are introducing a new structure, mt792x_link_sta, to manage per-link
configurations in preparation for future station (STA) support of
Multi-Link Operation (MLO).

This patch only includes structural changes and does not involve any
logic changes.

We have moved relevant parameters, such as the wcid from the mt76_wcid
structure, from the mt7921x_sta structure to the mt792x_link_sta structure.

For current drivers that do not support MLO, there is only one link STA,
and link information is accessed via the deflink member.

However, we have not yet created the per-link BSS configuration indexed
by link ID for Multi-Link Device (MLD) support in mt7921x_sta.
This step needs to be completed before adding MLD support for STA mode.

Co-developed-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-3-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/mac.c    | 65 +++++++++---------
 drivers/net/wireless/mediatek/mt76/mt7921/main.c   | 76 +++++++++++-----------
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.c    |  4 +-
 .../net/wireless/mediatek/mt76/mt7921/pci_mac.c    |  4 +-
 drivers/net/wireless/mediatek/mt76/mt7925/mac.c    | 66 ++++++++++---------
 drivers/net/wireless/mediatek/mt76/mt7925/main.c   | 66 +++++++++----------
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    | 17 ++---
 .../net/wireless/mediatek/mt76/mt7925/pci_mac.c    |  4 +-
 drivers/net/wireless/mediatek/mt76/mt792x.h        | 10 ++-
 drivers/net/wireless/mediatek/mt76/mt792x_core.c   | 28 ++++----
 drivers/net/wireless/mediatek/mt76/mt792x_mac.c    |  6 +-
 11 files changed, 180 insertions(+), 166 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
index 73e42ef42983..9ae9590999f4 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
@@ -39,6 +39,7 @@ static void mt7921_mac_sta_poll(struct mt792x_dev *dev)
 	};
 	struct ieee80211_sta *sta;
 	struct mt792x_sta *msta;
+	struct mt792x_link_sta *mlink;
 	u32 tx_time[IEEE80211_NUM_ACS], rx_time[IEEE80211_NUM_ACS];
 	LIST_HEAD(sta_poll_list);
 	struct rate_info *rate;
@@ -60,23 +61,25 @@ static void mt7921_mac_sta_poll(struct mt792x_dev *dev)
 			spin_unlock_bh(&dev->mt76.sta_poll_lock);
 			break;
 		}
-		msta = list_first_entry(&sta_poll_list,
-					struct mt792x_sta, wcid.poll_list);
-		list_del_init(&msta->wcid.poll_list);
+		mlink = list_first_entry(&sta_poll_list,
+					 struct mt792x_link_sta,
+					 wcid.poll_list);
+		msta = container_of(mlink, struct mt792x_sta, deflink);
+		list_del_init(&mlink->wcid.poll_list);
 		spin_unlock_bh(&dev->mt76.sta_poll_lock);
 
-		idx = msta->wcid.idx;
+		idx = mlink->wcid.idx;
 		addr = mt7921_mac_wtbl_lmac_addr(idx, MT_WTBL_AC0_CTT_OFFSET);
 
 		for (i = 0; i < IEEE80211_NUM_ACS; i++) {
-			u32 tx_last = msta->airtime_ac[i];
-			u32 rx_last = msta->airtime_ac[i + 4];
+			u32 tx_last = mlink->airtime_ac[i];
+			u32 rx_last = mlink->airtime_ac[i + 4];
 
-			msta->airtime_ac[i] = mt76_rr(dev, addr);
-			msta->airtime_ac[i + 4] = mt76_rr(dev, addr + 4);
+			mlink->airtime_ac[i] = mt76_rr(dev, addr);
+			mlink->airtime_ac[i + 4] = mt76_rr(dev, addr + 4);
 
-			tx_time[i] = msta->airtime_ac[i] - tx_last;
-			rx_time[i] = msta->airtime_ac[i + 4] - rx_last;
+			tx_time[i] = mlink->airtime_ac[i] - tx_last;
+			rx_time[i] = mlink->airtime_ac[i + 4] - rx_last;
 
 			if ((tx_last | rx_last) & BIT(30))
 				clear = true;
@@ -87,10 +90,10 @@ static void mt7921_mac_sta_poll(struct mt792x_dev *dev)
 		if (clear) {
 			mt7921_mac_wtbl_update(dev, idx,
 					       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
-			memset(msta->airtime_ac, 0, sizeof(msta->airtime_ac));
+			memset(mlink->airtime_ac, 0, sizeof(mlink->airtime_ac));
 		}
 
-		if (!msta->wcid.sta)
+		if (!mlink->wcid.sta)
 			continue;
 
 		sta = container_of((void *)msta, struct ieee80211_sta,
@@ -113,7 +116,7 @@ static void mt7921_mac_sta_poll(struct mt792x_dev *dev)
 		 * we need to make sure that flags match so polling GI
 		 * from per-sta counters directly.
 		 */
-		rate = &msta->wcid.rate;
+		rate = &mlink->wcid.rate;
 		addr = mt7921_mac_wtbl_lmac_addr(idx,
 						 MT_WTBL_TXRX_CAP_RATE_OFFSET);
 		val = mt76_rr(dev, addr);
@@ -154,10 +157,10 @@ static void mt7921_mac_sta_poll(struct mt792x_dev *dev)
 		rssi[2] = to_rssi(GENMASK(23, 16), val);
 		rssi[3] = to_rssi(GENMASK(31, 14), val);
 
-		msta->ack_signal =
+		mlink->ack_signal =
 			mt76_rx_signal(msta->vif->phy->mt76->antenna_mask, rssi);
 
-		ewma_avg_signal_add(&msta->avg_ack_signal, -msta->ack_signal);
+		ewma_avg_signal_add(&mlink->avg_ack_signal, -mlink->ack_signal);
 	}
 }
 
@@ -180,6 +183,7 @@ mt7921_mac_fill_rx(struct mt792x_dev *dev, struct sk_buff *skb)
 	u32 rxd3 = le32_to_cpu(rxd[3]);
 	u32 rxd4 = le32_to_cpu(rxd[4]);
 	struct mt792x_sta *msta = NULL;
+	struct mt792x_link_sta *mlink;
 	u16 seq_ctrl = 0;
 	__le16 fc = 0;
 	u8 mode = 0;
@@ -210,10 +214,11 @@ mt7921_mac_fill_rx(struct mt792x_dev *dev, struct sk_buff *skb)
 	status->wcid = mt792x_rx_get_wcid(dev, idx, unicast);
 
 	if (status->wcid) {
-		msta = container_of(status->wcid, struct mt792x_sta, wcid);
+		mlink = container_of(status->wcid, struct mt792x_link_sta, wcid);
+		msta = container_of(mlink, struct mt792x_sta, deflink);
 		spin_lock_bh(&dev->mt76.sta_poll_lock);
-		if (list_empty(&msta->wcid.poll_list))
-			list_add_tail(&msta->wcid.poll_list,
+		if (list_empty(&mlink->wcid.poll_list))
+			list_add_tail(&mlink->wcid.poll_list,
 				      &dev->mt76.sta_poll_list);
 		spin_unlock_bh(&dev->mt76.sta_poll_lock);
 	}
@@ -444,7 +449,7 @@ mt7921_mac_fill_rx(struct mt792x_dev *dev, struct sk_buff *skb)
 
 void mt7921_mac_add_txs(struct mt792x_dev *dev, void *data)
 {
-	struct mt792x_sta *msta = NULL;
+	struct mt792x_link_sta *mlink;
 	struct mt76_wcid *wcid;
 	__le32 *txs_data = data;
 	u16 wcidx;
@@ -468,15 +473,15 @@ void mt7921_mac_add_txs(struct mt792x_dev *dev, void *data)
 	if (!wcid)
 		goto out;
 
-	msta = container_of(wcid, struct mt792x_sta, wcid);
+	mlink = container_of(wcid, struct mt792x_link_sta, wcid);
 
 	mt76_connac2_mac_add_txs_skb(&dev->mt76, wcid, pid, txs_data);
 	if (!wcid->sta)
 		goto out;
 
 	spin_lock_bh(&dev->mt76.sta_poll_lock);
-	if (list_empty(&msta->wcid.poll_list))
-		list_add_tail(&msta->wcid.poll_list, &dev->mt76.sta_poll_list);
+	if (list_empty(&mlink->wcid.poll_list))
+		list_add_tail(&mlink->wcid.poll_list, &dev->mt76.sta_poll_list);
 	spin_unlock_bh(&dev->mt76.sta_poll_lock);
 
 out:
@@ -513,7 +518,7 @@ static void mt7921_mac_tx_free(struct mt792x_dev *dev, void *data, int len)
 		 * 1'b0: msdu_id with the same 'wcid pair' as above.
 		 */
 		if (info & MT_TX_FREE_PAIR) {
-			struct mt792x_sta *msta;
+			struct mt792x_link_sta *mlink;
 			u16 idx;
 
 			count++;
@@ -523,10 +528,10 @@ static void mt7921_mac_tx_free(struct mt792x_dev *dev, void *data, int len)
 			if (!sta)
 				continue;
 
-			msta = container_of(wcid, struct mt792x_sta, wcid);
+			mlink = container_of(wcid, struct mt792x_link_sta, wcid);
 			spin_lock_bh(&mdev->sta_poll_lock);
-			if (list_empty(&msta->wcid.poll_list))
-				list_add_tail(&msta->wcid.poll_list,
+			if (list_empty(&mlink->wcid.poll_list))
+				list_add_tail(&mlink->wcid.poll_list,
 					      &mdev->sta_poll_list);
 			spin_unlock_bh(&mdev->sta_poll_lock);
 			continue;
@@ -641,11 +646,11 @@ mt7921_vif_connect_iter(void *priv, u8 *mac,
 	if (vif->type == NL80211_IFTYPE_STATION)
 		ieee80211_disconnect(vif, true);
 
-	mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.wcid, true);
+	mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.deflink.wcid, true);
 	mt7921_mcu_set_tx(dev, vif);
 
 	if (vif->type == NL80211_IFTYPE_AP) {
-		mt76_connac_mcu_uni_add_bss(dev->phy.mt76, vif, &mvif->sta.wcid,
+		mt76_connac_mcu_uni_add_bss(dev->phy.mt76, vif, &mvif->sta.deflink.wcid,
 					    true, NULL);
 		mt7921_mcu_sta_update(dev, NULL, vif, true,
 				      MT76_STA_INFO_STATE_NONE);
@@ -786,9 +791,9 @@ int mt7921_usb_sdio_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr,
 	if (sta) {
 		struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 
-		if (time_after(jiffies, msta->last_txs + HZ / 4)) {
+		if (time_after(jiffies, msta->deflink.last_txs + HZ / 4)) {
 			info->flags |= IEEE80211_TX_CTL_REQ_TX_STATUS;
-			msta->last_txs = jiffies;
+			msta->deflink.last_txs = jiffies;
 		}
 	}
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/main.c b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
index 77999d778a38..f83cb45957d2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
@@ -306,7 +306,7 @@ mt7921_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 	mvif->bss_conf.mt76.band_idx = 0;
 	mvif->bss_conf.mt76.wmm_idx = mvif->bss_conf.mt76.idx % MT76_CONNAC_MAX_WMM_SETS;
 
-	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.wcid,
+	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.deflink.wcid,
 					  true);
 	if (ret)
 		goto out;
@@ -316,19 +316,19 @@ mt7921_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 
 	idx = MT792x_WTBL_RESERVED - mvif->bss_conf.mt76.idx;
 
-	INIT_LIST_HEAD(&mvif->sta.wcid.poll_list);
-	mvif->sta.wcid.idx = idx;
-	mvif->sta.wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
-	mvif->sta.wcid.hw_key_idx = -1;
-	mvif->sta.wcid.tx_info |= MT_WCID_TX_INFO_SET;
-	mt76_wcid_init(&mvif->sta.wcid);
+	INIT_LIST_HEAD(&mvif->sta.deflink.wcid.poll_list);
+	mvif->sta.deflink.wcid.idx = idx;
+	mvif->sta.deflink.wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
+	mvif->sta.deflink.wcid.hw_key_idx = -1;
+	mvif->sta.deflink.wcid.tx_info |= MT_WCID_TX_INFO_SET;
+	mt76_wcid_init(&mvif->sta.deflink.wcid);
 
 	mt7921_mac_wtbl_update(dev, idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
 	ewma_rssi_init(&mvif->bss_conf.rssi);
 
-	rcu_assign_pointer(dev->mt76.wcid[idx], &mvif->sta.wcid);
+	rcu_assign_pointer(dev->mt76.wcid[idx], &mvif->sta.deflink.wcid);
 	if (vif->txq) {
 		mtxq = (struct mt76_txq *)vif->txq->drv_priv;
 		mtxq->wcid = idx;
@@ -494,7 +494,7 @@ static int mt7921_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_sta *msta = sta ? (struct mt792x_sta *)sta->drv_priv :
 				  &mvif->sta;
-	struct mt76_wcid *wcid = &msta->wcid;
+	struct mt76_wcid *wcid = &msta->deflink.wcid;
 	u8 *wcid_keyidx = &wcid->hw_key_idx;
 	int idx = key->keyidx, err = 0;
 
@@ -541,18 +541,18 @@ static int mt7921_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	}
 
 	mt76_wcid_key_setup(&dev->mt76, wcid, key);
-	err = mt76_connac_mcu_add_key(&dev->mt76, vif, &msta->bip,
+	err = mt76_connac_mcu_add_key(&dev->mt76, vif, &msta->deflink.bip,
 				      key, MCU_UNI_CMD(STA_REC_UPDATE),
-				      &msta->wcid, cmd);
+				      &msta->deflink.wcid, cmd);
 	if (err)
 		goto out;
 
 	if (key->cipher == WLAN_CIPHER_SUITE_WEP104 ||
 	    key->cipher == WLAN_CIPHER_SUITE_WEP40)
 		err = mt76_connac_mcu_add_key(&dev->mt76, vif,
-					      &mvif->wep_sta->bip,
+					      &mvif->wep_sta->deflink.bip,
 					      key, MCU_UNI_CMD(STA_REC_UPDATE),
-					      &mvif->wep_sta->wcid, cmd);
+					      &mvif->wep_sta->deflink.wcid, cmd);
 out:
 	mt792x_mutex_release(dev);
 
@@ -799,13 +799,13 @@ int mt7921_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	if (idx < 0)
 		return -ENOSPC;
 
-	INIT_LIST_HEAD(&msta->wcid.poll_list);
+	INIT_LIST_HEAD(&msta->deflink.wcid.poll_list);
 	msta->vif = mvif;
-	msta->wcid.sta = 1;
-	msta->wcid.idx = idx;
-	msta->wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
-	msta->wcid.tx_info |= MT_WCID_TX_INFO_SET;
-	msta->last_txs = jiffies;
+	msta->deflink.wcid.sta = 1;
+	msta->deflink.wcid.idx = idx;
+	msta->deflink.wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
+	msta->deflink.wcid.tx_info |= MT_WCID_TX_INFO_SET;
+	msta->deflink.last_txs = jiffies;
 
 	ret = mt76_connac_pm_wake(&dev->mphy, &dev->pm);
 	if (ret)
@@ -840,14 +840,14 @@ void mt7921_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	mt792x_mutex_acquire(dev);
 
 	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
-		mt76_connac_mcu_uni_add_bss(&dev->mphy, vif, &mvif->sta.wcid,
+		mt76_connac_mcu_uni_add_bss(&dev->mphy, vif, &mvif->sta.deflink.wcid,
 					    true, mvif->bss_conf.mt76.ctx);
 
-	ewma_avg_signal_init(&msta->avg_ack_signal);
+	ewma_avg_signal_init(&msta->deflink.avg_ack_signal);
 
-	mt7921_mac_wtbl_update(dev, msta->wcid.idx,
+	mt7921_mac_wtbl_update(dev, msta->deflink.wcid.idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
-	memset(msta->airtime_ac, 0, sizeof(msta->airtime_ac));
+	memset(msta->deflink.airtime_ac, 0, sizeof(msta->deflink.airtime_ac));
 
 	mt7921_mcu_sta_update(dev, sta, vif, true, MT76_STA_INFO_STATE_ASSOC);
 
@@ -861,11 +861,11 @@ void mt7921_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
 	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 
-	mt76_connac_free_pending_tx_skbs(&dev->pm, &msta->wcid);
+	mt76_connac_free_pending_tx_skbs(&dev->pm, &msta->deflink.wcid);
 	mt76_connac_pm_wake(&dev->mphy, &dev->pm);
 
 	mt7921_mcu_sta_update(dev, sta, vif, false, MT76_STA_INFO_STATE_NONE);
-	mt7921_mac_wtbl_update(dev, msta->wcid.idx,
+	mt7921_mac_wtbl_update(dev, msta->deflink.wcid.idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
 	if (vif->type == NL80211_IFTYPE_STATION) {
@@ -875,13 +875,13 @@ void mt7921_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 		ewma_rssi_init(&mvif->bss_conf.rssi);
 		if (!sta->tdls)
 			mt76_connac_mcu_uni_add_bss(&dev->mphy, vif,
-						    &mvif->sta.wcid, false,
+						    &mvif->sta.deflink.wcid, false,
 						    mvif->bss_conf.mt76.ctx);
 	}
 
 	spin_lock_bh(&dev->mt76.sta_poll_lock);
-	if (!list_empty(&msta->wcid.poll_list))
-		list_del_init(&msta->wcid.poll_list);
+	if (!list_empty(&msta->deflink.wcid.poll_list))
+		list_del_init(&msta->deflink.wcid.poll_list);
 	spin_unlock_bh(&dev->mt76.sta_poll_lock);
 
 	mt7921_regd_set_6ghz_power_type(vif, false);
@@ -923,12 +923,12 @@ mt7921_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	mt792x_mutex_acquire(dev);
 	switch (action) {
 	case IEEE80211_AMPDU_RX_START:
-		mt76_rx_aggr_start(&dev->mt76, &msta->wcid, tid, ssn,
+		mt76_rx_aggr_start(&dev->mt76, &msta->deflink.wcid, tid, ssn,
 				   params->buf_size);
 		mt7921_mcu_uni_rx_ba(dev, params, true);
 		break;
 	case IEEE80211_AMPDU_RX_STOP:
-		mt76_rx_aggr_stop(&dev->mt76, &msta->wcid, tid);
+		mt76_rx_aggr_stop(&dev->mt76, &msta->deflink.wcid, tid);
 		mt7921_mcu_uni_rx_ba(dev, params, false);
 		break;
 	case IEEE80211_AMPDU_TX_OPERATIONAL:
@@ -939,16 +939,16 @@ mt7921_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	case IEEE80211_AMPDU_TX_STOP_FLUSH:
 	case IEEE80211_AMPDU_TX_STOP_FLUSH_CONT:
 		mtxq->aggr = false;
-		clear_bit(tid, &msta->wcid.ampdu_state);
+		clear_bit(tid, &msta->deflink.wcid.ampdu_state);
 		mt7921_mcu_uni_tx_ba(dev, params, false);
 		break;
 	case IEEE80211_AMPDU_TX_START:
-		set_bit(tid, &msta->wcid.ampdu_state);
+		set_bit(tid, &msta->deflink.wcid.ampdu_state);
 		ret = IEEE80211_AMPDU_TX_START_IMMEDIATE;
 		break;
 	case IEEE80211_AMPDU_TX_STOP_CONT:
 		mtxq->aggr = false;
-		clear_bit(tid, &msta->wcid.ampdu_state);
+		clear_bit(tid, &msta->deflink.wcid.ampdu_state);
 		mt7921_mcu_uni_tx_ba(dev, params, false);
 		ieee80211_stop_tx_ba_cb_irqsafe(vif, sta->addr, tid);
 		break;
@@ -1166,11 +1166,11 @@ static void mt7921_sta_set_decap_offload(struct ieee80211_hw *hw,
 	mt792x_mutex_acquire(dev);
 
 	if (enabled)
-		set_bit(MT_WCID_FLAG_HDR_TRANS, &msta->wcid.flags);
+		set_bit(MT_WCID_FLAG_HDR_TRANS, &msta->deflink.wcid.flags);
 	else
-		clear_bit(MT_WCID_FLAG_HDR_TRANS, &msta->wcid.flags);
+		clear_bit(MT_WCID_FLAG_HDR_TRANS, &msta->deflink.wcid.flags);
 
-	mt76_connac_mcu_sta_update_hdr_trans(&dev->mt76, vif, &msta->wcid,
+	mt76_connac_mcu_sta_update_hdr_trans(&dev->mt76, vif, &msta->deflink.wcid,
 					     MCU_UNI_CMD(STA_REC_UPDATE));
 
 	mt792x_mutex_release(dev);
@@ -1294,7 +1294,7 @@ mt7921_start_ap(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 
 	mt792x_mutex_acquire(dev);
 
-	err = mt76_connac_mcu_uni_add_bss(phy->mt76, vif, &mvif->sta.wcid,
+	err = mt76_connac_mcu_uni_add_bss(phy->mt76, vif, &mvif->sta.deflink.wcid,
 					  true, mvif->bss_conf.mt76.ctx);
 	if (err)
 		goto out;
@@ -1326,7 +1326,7 @@ mt7921_stop_ap(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	if (err)
 		goto out;
 
-	mt76_connac_mcu_uni_add_bss(phy->mt76, vif, &mvif->sta.wcid, false,
+	mt76_connac_mcu_uni_add_bss(phy->mt76, vif, &mvif->sta.deflink.wcid, false,
 				    mvif->bss_conf.mt76.ctx);
 
 out:
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
index 0453af012016..394fcd799345 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mcu.c
@@ -386,7 +386,7 @@ int mt7921_mcu_uni_tx_ba(struct mt792x_dev *dev,
 	struct mt792x_sta *msta = (struct mt792x_sta *)params->sta->drv_priv;
 
 	if (enable && !params->amsdu)
-		msta->wcid.amsdu = false;
+		msta->deflink.wcid.amsdu = false;
 
 	return mt76_connac_mcu_sta_ba(&dev->mt76, &msta->vif->bss_conf.mt76, params,
 				      MCU_UNI_CMD(STA_REC_UPDATE),
@@ -1055,7 +1055,7 @@ int mt7921_mcu_sta_update(struct mt792x_dev *dev, struct ieee80211_sta *sta,
 	struct mt792x_sta *msta;
 
 	msta = sta ? (struct mt792x_sta *)sta->drv_priv : NULL;
-	info.wcid = msta ? &msta->wcid : &mvif->sta.wcid;
+	info.wcid = msta ? &msta->deflink.wcid : &mvif->sta.deflink.wcid;
 	info.newly = msta ? state != MT76_STA_INFO_STATE_ASSOC : true;
 
 	return mt76_connac_mcu_sta_cmd(&dev->mphy, &info);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c
index 031ba9aaa4e2..2452b1a2d118 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci_mac.c
@@ -34,9 +34,9 @@ int mt7921e_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr,
 	if (sta) {
 		struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 
-		if (time_after(jiffies, msta->last_txs + HZ / 4)) {
+		if (time_after(jiffies, msta->deflink.last_txs + HZ / 4)) {
 			info->flags |= IEEE80211_TX_CTL_REQ_TX_STATUS;
-			msta->last_txs = jiffies;
+			msta->deflink.last_txs = jiffies;
 		}
 	}
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
index c2460ef4993d..88e7b2ff4593 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
@@ -28,6 +28,7 @@ static void mt7925_mac_sta_poll(struct mt792x_dev *dev)
 	};
 	struct ieee80211_sta *sta;
 	struct mt792x_sta *msta;
+	struct mt792x_link_sta *mlink;
 	u32 tx_time[IEEE80211_NUM_ACS], rx_time[IEEE80211_NUM_ACS];
 	LIST_HEAD(sta_poll_list);
 	struct rate_info *rate;
@@ -46,24 +47,25 @@ static void mt7925_mac_sta_poll(struct mt792x_dev *dev)
 
 		if (list_empty(&sta_poll_list))
 			break;
-		msta = list_first_entry(&sta_poll_list,
-					struct mt792x_sta, wcid.poll_list);
+		mlink = list_first_entry(&sta_poll_list,
+					 struct mt792x_link_sta, wcid.poll_list);
+		msta = container_of(mlink, struct mt792x_sta, deflink);
 		spin_lock_bh(&dev->mt76.sta_poll_lock);
-		list_del_init(&msta->wcid.poll_list);
+		list_del_init(&mlink->wcid.poll_list);
 		spin_unlock_bh(&dev->mt76.sta_poll_lock);
 
-		idx = msta->wcid.idx;
+		idx = mlink->wcid.idx;
 		addr = mt7925_mac_wtbl_lmac_addr(dev, idx, MT_WTBL_AC0_CTT_OFFSET);
 
 		for (i = 0; i < IEEE80211_NUM_ACS; i++) {
-			u32 tx_last = msta->airtime_ac[i];
-			u32 rx_last = msta->airtime_ac[i + 4];
+			u32 tx_last = mlink->airtime_ac[i];
+			u32 rx_last = mlink->airtime_ac[i + 4];
 
-			msta->airtime_ac[i] = mt76_rr(dev, addr);
-			msta->airtime_ac[i + 4] = mt76_rr(dev, addr + 4);
+			mlink->airtime_ac[i] = mt76_rr(dev, addr);
+			mlink->airtime_ac[i + 4] = mt76_rr(dev, addr + 4);
 
-			tx_time[i] = msta->airtime_ac[i] - tx_last;
-			rx_time[i] = msta->airtime_ac[i + 4] - rx_last;
+			tx_time[i] = mlink->airtime_ac[i] - tx_last;
+			rx_time[i] = mlink->airtime_ac[i + 4] - rx_last;
 
 			if ((tx_last | rx_last) & BIT(30))
 				clear = true;
@@ -74,10 +76,10 @@ static void mt7925_mac_sta_poll(struct mt792x_dev *dev)
 		if (clear) {
 			mt7925_mac_wtbl_update(dev, idx,
 					       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
-			memset(msta->airtime_ac, 0, sizeof(msta->airtime_ac));
+			memset(mlink->airtime_ac, 0, sizeof(mlink->airtime_ac));
 		}
 
-		if (!msta->wcid.sta)
+		if (!mlink->wcid.sta)
 			continue;
 
 		sta = container_of((void *)msta, struct ieee80211_sta,
@@ -100,7 +102,7 @@ static void mt7925_mac_sta_poll(struct mt792x_dev *dev)
 		 * we need to make sure that flags match so polling GI
 		 * from per-sta counters directly.
 		 */
-		rate = &msta->wcid.rate;
+		rate = &mlink->wcid.rate;
 
 		switch (rate->bw) {
 		case RATE_INFO_BW_160:
@@ -144,10 +146,10 @@ static void mt7925_mac_sta_poll(struct mt792x_dev *dev)
 		rssi[2] = to_rssi(GENMASK(23, 16), val);
 		rssi[3] = to_rssi(GENMASK(31, 14), val);
 
-		msta->ack_signal =
+		mlink->ack_signal =
 			mt76_rx_signal(msta->vif->phy->mt76->antenna_mask, rssi);
 
-		ewma_avg_signal_add(&msta->avg_ack_signal, -msta->ack_signal);
+		ewma_avg_signal_add(&mlink->avg_ack_signal, -mlink->ack_signal);
 	}
 }
 
@@ -365,7 +367,7 @@ mt7925_mac_fill_rx(struct mt792x_dev *dev, struct sk_buff *skb)
 	u32 rxd2 = le32_to_cpu(rxd[2]);
 	u32 rxd3 = le32_to_cpu(rxd[3]);
 	u32 rxd4 = le32_to_cpu(rxd[4]);
-	struct mt792x_sta *msta = NULL;
+	struct mt792x_link_sta *mlink;
 	u8 mode = 0; /* , band_idx; */
 	u16 seq_ctrl = 0;
 	__le16 fc = 0;
@@ -393,10 +395,10 @@ mt7925_mac_fill_rx(struct mt792x_dev *dev, struct sk_buff *skb)
 	status->wcid = mt792x_rx_get_wcid(dev, idx, unicast);
 
 	if (status->wcid) {
-		msta = container_of(status->wcid, struct mt792x_sta, wcid);
+		mlink = container_of(status->wcid, struct mt792x_link_sta, wcid);
 		spin_lock_bh(&dev->mt76.sta_poll_lock);
-		if (list_empty(&msta->wcid.poll_list))
-			list_add_tail(&msta->wcid.poll_list,
+		if (list_empty(&mlink->wcid.poll_list))
+			list_add_tail(&mlink->wcid.poll_list,
 				      &dev->mt76.sta_poll_list);
 		spin_unlock_bh(&dev->mt76.sta_poll_lock);
 	}
@@ -851,7 +853,7 @@ static void mt7925_tx_check_aggr(struct ieee80211_sta *sta, __le32 *txwi)
 		return;
 
 	msta = (struct mt792x_sta *)sta->drv_priv;
-	if (!test_and_set_bit(tid, &msta->wcid.ampdu_state))
+	if (!test_and_set_bit(tid, &msta->deflink.wcid.ampdu_state))
 		ieee80211_start_tx_ba_session(sta, tid, 0);
 }
 
@@ -991,7 +993,7 @@ out_no_skb:
 
 void mt7925_mac_add_txs(struct mt792x_dev *dev, void *data)
 {
-	struct mt792x_sta *msta = NULL;
+	struct mt792x_link_sta *mlink = NULL;
 	struct mt76_wcid *wcid;
 	__le32 *txs_data = data;
 	u16 wcidx;
@@ -1015,15 +1017,15 @@ void mt7925_mac_add_txs(struct mt792x_dev *dev, void *data)
 	if (!wcid)
 		goto out;
 
-	msta = container_of(wcid, struct mt792x_sta, wcid);
+	mlink = container_of(wcid, struct mt792x_link_sta, wcid);
 
 	mt7925_mac_add_txs_skb(dev, wcid, pid, txs_data);
 	if (!wcid->sta)
 		goto out;
 
 	spin_lock_bh(&dev->mt76.sta_poll_lock);
-	if (list_empty(&msta->wcid.poll_list))
-		list_add_tail(&msta->wcid.poll_list, &dev->mt76.sta_poll_list);
+	if (list_empty(&mlink->wcid.poll_list))
+		list_add_tail(&mlink->wcid.poll_list, &dev->mt76.sta_poll_list);
 	spin_unlock_bh(&dev->mt76.sta_poll_lock);
 
 out:
@@ -1094,7 +1096,7 @@ mt7925_mac_tx_free(struct mt792x_dev *dev, void *data, int len)
 		 */
 		info = le32_to_cpu(*cur_info);
 		if (info & MT_TXFREE_INFO_PAIR) {
-			struct mt792x_sta *msta;
+			struct mt792x_link_sta *mlink;
 			u16 idx;
 
 			idx = FIELD_GET(MT_TXFREE_INFO_WLAN_ID, info);
@@ -1103,10 +1105,10 @@ mt7925_mac_tx_free(struct mt792x_dev *dev, void *data, int len)
 			if (!sta)
 				continue;
 
-			msta = container_of(wcid, struct mt792x_sta, wcid);
+			mlink = container_of(wcid, struct mt792x_link_sta, wcid);
 			spin_lock_bh(&mdev->sta_poll_lock);
-			if (list_empty(&msta->wcid.poll_list))
-				list_add_tail(&msta->wcid.poll_list,
+			if (list_empty(&mlink->wcid.poll_list))
+				list_add_tail(&mlink->wcid.poll_list,
 					      &mdev->sta_poll_list);
 			spin_unlock_bh(&mdev->sta_poll_lock);
 			continue;
@@ -1241,11 +1243,11 @@ mt7925_vif_connect_iter(void *priv, u8 *mac,
 	if (vif->type == NL80211_IFTYPE_STATION)
 		ieee80211_disconnect(vif, true);
 
-	mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.wcid, true);
+	mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.deflink.wcid, true);
 	mt7925_mcu_set_tx(dev, vif);
 
 	if (vif->type == NL80211_IFTYPE_AP) {
-		mt76_connac_mcu_uni_add_bss(dev->phy.mt76, vif, &mvif->sta.wcid,
+		mt76_connac_mcu_uni_add_bss(dev->phy.mt76, vif, &mvif->sta.deflink.wcid,
 					    true, NULL);
 		mt7925_mcu_sta_update(dev, NULL, vif, true,
 				      MT76_STA_INFO_STATE_NONE);
@@ -1380,9 +1382,9 @@ int mt7925_usb_sdio_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr,
 	if (sta) {
 		struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 
-		if (time_after(jiffies, msta->last_txs + HZ / 4)) {
+		if (time_after(jiffies, msta->deflink.last_txs + HZ / 4)) {
 			info->flags |= IEEE80211_TX_CTL_REQ_TX_STATUS;
-			msta->last_txs = jiffies;
+			msta->deflink.last_txs = jiffies;
 		}
 	}
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index dfc606cc9bdd..504c2ecc3845 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -344,7 +344,7 @@ mt7925_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 	else
 		mvif->bss_conf.mt76.basic_rates_idx = MT792x_BASIC_RATES_TBL;
 
-	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.wcid,
+	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.deflink.wcid,
 					  true);
 	if (ret)
 		goto out;
@@ -354,20 +354,20 @@ mt7925_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 
 	idx = MT792x_WTBL_RESERVED - mvif->bss_conf.mt76.idx;
 
-	INIT_LIST_HEAD(&mvif->sta.wcid.poll_list);
-	mvif->sta.wcid.idx = idx;
-	mvif->sta.wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
-	mvif->sta.wcid.hw_key_idx = -1;
-	mvif->sta.wcid.tx_info |= MT_WCID_TX_INFO_SET;
+	INIT_LIST_HEAD(&mvif->sta.deflink.wcid.poll_list);
+	mvif->sta.deflink.wcid.idx = idx;
+	mvif->sta.deflink.wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
+	mvif->sta.deflink.wcid.hw_key_idx = -1;
+	mvif->sta.deflink.wcid.tx_info |= MT_WCID_TX_INFO_SET;
 	mvif->sta.vif = mvif;
-	mt76_wcid_init(&mvif->sta.wcid);
+	mt76_wcid_init(&mvif->sta.deflink.wcid);
 
 	mt7925_mac_wtbl_update(dev, idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
 	ewma_rssi_init(&mvif->bss_conf.rssi);
 
-	rcu_assign_pointer(dev->mt76.wcid[idx], &mvif->sta.wcid);
+	rcu_assign_pointer(dev->mt76.wcid[idx], &mvif->sta.deflink.wcid);
 	if (vif->txq) {
 		mtxq = (struct mt76_txq *)vif->txq->drv_priv;
 		mtxq->wcid = idx;
@@ -486,7 +486,7 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_sta *msta = sta ? (struct mt792x_sta *)sta->drv_priv :
 				  &mvif->sta;
-	struct mt76_wcid *wcid = &msta->wcid;
+	struct mt76_wcid *wcid = &msta->deflink.wcid;
 	u8 *wcid_keyidx = &wcid->hw_key_idx;
 	int idx = key->keyidx, err = 0;
 
@@ -541,18 +541,18 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	mt76_wcid_key_setup(&dev->mt76, wcid,
 			    cmd == SET_KEY ? key : NULL);
 
-	err = mt7925_mcu_add_key(&dev->mt76, vif, &msta->bip,
+	err = mt7925_mcu_add_key(&dev->mt76, vif, &msta->deflink.bip,
 				 key, MCU_UNI_CMD(STA_REC_UPDATE),
-				 &msta->wcid, cmd);
+				 &msta->deflink.wcid, cmd);
 
 	if (err)
 		goto out;
 
 	if (key->cipher == WLAN_CIPHER_SUITE_WEP104 ||
 	    key->cipher == WLAN_CIPHER_SUITE_WEP40)
-		err = mt7925_mcu_add_key(&dev->mt76, vif, &mvif->wep_sta->bip,
+		err = mt7925_mcu_add_key(&dev->mt76, vif, &mvif->wep_sta->deflink.bip,
 					 key, MCU_WMWA_UNI_CMD(STA_REC_UPDATE),
-					 &mvif->wep_sta->wcid, cmd);
+					 &mvif->wep_sta->deflink.wcid, cmd);
 
 out:
 	mt792x_mutex_release(dev);
@@ -766,13 +766,13 @@ int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	if (idx < 0)
 		return -ENOSPC;
 
-	INIT_LIST_HEAD(&msta->wcid.poll_list);
+	INIT_LIST_HEAD(&msta->deflink.wcid.poll_list);
 	msta->vif = mvif;
-	msta->wcid.sta = 1;
-	msta->wcid.idx = idx;
-	msta->wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
-	msta->wcid.tx_info |= MT_WCID_TX_INFO_SET;
-	msta->last_txs = jiffies;
+	msta->deflink.wcid.sta = 1;
+	msta->deflink.wcid.idx = idx;
+	msta->deflink.wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
+	msta->deflink.wcid.tx_info |= MT_WCID_TX_INFO_SET;
+	msta->deflink.last_txs = jiffies;
 
 	ret = mt76_connac_pm_wake(&dev->mphy, &dev->pm);
 	if (ret)
@@ -813,11 +813,11 @@ void mt7925_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, vif, sta,
 					true);
 
-	ewma_avg_signal_init(&msta->avg_ack_signal);
+	ewma_avg_signal_init(&msta->deflink.avg_ack_signal);
 
-	mt7925_mac_wtbl_update(dev, msta->wcid.idx,
+	mt7925_mac_wtbl_update(dev, msta->deflink.wcid.idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
-	memset(msta->airtime_ac, 0, sizeof(msta->airtime_ac));
+	memset(msta->deflink.airtime_ac, 0, sizeof(msta->deflink.airtime_ac));
 
 	mt7925_mcu_sta_update(dev, sta, vif, true, MT76_STA_INFO_STATE_ASSOC);
 
@@ -831,11 +831,11 @@ void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
 	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 
-	mt76_connac_free_pending_tx_skbs(&dev->pm, &msta->wcid);
+	mt76_connac_free_pending_tx_skbs(&dev->pm, &msta->deflink.wcid);
 	mt76_connac_pm_wake(&dev->mphy, &dev->pm);
 
 	mt7925_mcu_sta_update(dev, sta, vif, false, MT76_STA_INFO_STATE_NONE);
-	mt7925_mac_wtbl_update(dev, msta->wcid.idx,
+	mt7925_mac_wtbl_update(dev, msta->deflink.wcid.idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
 	if (vif->type == NL80211_IFTYPE_STATION) {
@@ -849,8 +849,8 @@ void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	}
 
 	spin_lock_bh(&mdev->sta_poll_lock);
-	if (!list_empty(&msta->wcid.poll_list))
-		list_del_init(&msta->wcid.poll_list);
+	if (!list_empty(&msta->deflink.wcid.poll_list))
+		list_del_init(&msta->deflink.wcid.poll_list);
 	spin_unlock_bh(&mdev->sta_poll_lock);
 
 	mt76_connac_power_save_sched(&dev->mphy, &dev->pm);
@@ -890,12 +890,12 @@ mt7925_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	mt792x_mutex_acquire(dev);
 	switch (action) {
 	case IEEE80211_AMPDU_RX_START:
-		mt76_rx_aggr_start(&dev->mt76, &msta->wcid, tid, ssn,
+		mt76_rx_aggr_start(&dev->mt76, &msta->deflink.wcid, tid, ssn,
 				   params->buf_size);
 		mt7925_mcu_uni_rx_ba(dev, params, true);
 		break;
 	case IEEE80211_AMPDU_RX_STOP:
-		mt76_rx_aggr_stop(&dev->mt76, &msta->wcid, tid);
+		mt76_rx_aggr_stop(&dev->mt76, &msta->deflink.wcid, tid);
 		mt7925_mcu_uni_rx_ba(dev, params, false);
 		break;
 	case IEEE80211_AMPDU_TX_OPERATIONAL:
@@ -906,16 +906,16 @@ mt7925_ampdu_action(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	case IEEE80211_AMPDU_TX_STOP_FLUSH:
 	case IEEE80211_AMPDU_TX_STOP_FLUSH_CONT:
 		mtxq->aggr = false;
-		clear_bit(tid, &msta->wcid.ampdu_state);
+		clear_bit(tid, &msta->deflink.wcid.ampdu_state);
 		mt7925_mcu_uni_tx_ba(dev, params, false);
 		break;
 	case IEEE80211_AMPDU_TX_START:
-		set_bit(tid, &msta->wcid.ampdu_state);
+		set_bit(tid, &msta->deflink.wcid.ampdu_state);
 		ret = IEEE80211_AMPDU_TX_START_IMMEDIATE;
 		break;
 	case IEEE80211_AMPDU_TX_STOP_CONT:
 		mtxq->aggr = false;
-		clear_bit(tid, &msta->wcid.ampdu_state);
+		clear_bit(tid, &msta->deflink.wcid.ampdu_state);
 		mt7925_mcu_uni_tx_ba(dev, params, false);
 		ieee80211_stop_tx_ba_cb_irqsafe(vif, sta->addr, tid);
 		break;
@@ -1161,9 +1161,9 @@ static void mt7925_sta_set_decap_offload(struct ieee80211_hw *hw,
 	mt792x_mutex_acquire(dev);
 
 	if (enabled)
-		set_bit(MT_WCID_FLAG_HDR_TRANS, &msta->wcid.flags);
+		set_bit(MT_WCID_FLAG_HDR_TRANS, &msta->deflink.wcid.flags);
 	else
-		clear_bit(MT_WCID_FLAG_HDR_TRANS, &msta->wcid.flags);
+		clear_bit(MT_WCID_FLAG_HDR_TRANS, &msta->deflink.wcid.flags);
 
 	mt7925_mcu_wtbl_update_hdr_trans(dev, vif, sta);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index c259aedb96e0..67c05f785190 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -544,7 +544,7 @@ int mt7925_mcu_uni_tx_ba(struct mt792x_dev *dev,
 	struct mt792x_vif *mvif = msta->vif;
 
 	if (enable && !params->amsdu)
-		msta->wcid.amsdu = false;
+		msta->deflink.wcid.amsdu = false;
 
 	return mt7925_mcu_sta_ba(&dev->mt76, &mvif->bss_conf.mt76, params,
 				 enable, true);
@@ -867,7 +867,7 @@ mt7925_mcu_sta_hdr_trans_tlv(struct sk_buff *skb,
 	if (sta)
 		wcid = (struct mt76_wcid *)sta->drv_priv;
 	else
-		wcid = &mvif->sta.wcid;
+		wcid = &mvif->sta.deflink.wcid;
 
 	if (!wcid)
 		return;
@@ -890,7 +890,7 @@ int mt7925_mcu_wtbl_update_hdr_trans(struct mt792x_dev *dev,
 	msta = sta ? (struct mt792x_sta *)sta->drv_priv : &mvif->sta;
 
 	skb = __mt76_connac_mcu_alloc_sta_req(&dev->mt76, &mvif->bss_conf.mt76,
-					      &msta->wcid,
+					      &msta->deflink.wcid,
 					      MT7925_STA_UPDATE_MAX_SIZE);
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
@@ -962,7 +962,8 @@ mt7925_mcu_sta_key_tlv(struct mt76_wcid *wcid,
 		       struct ieee80211_key_conf *key,
 		       enum set_key_cmd cmd)
 {
-	struct mt792x_sta *msta = container_of(wcid, struct mt792x_sta, wcid);
+	struct mt792x_link_sta *mlink = container_of(wcid, struct mt792x_link_sta, wcid);
+	struct mt792x_sta *msta = container_of(mlink, struct mt792x_sta, deflink);
 	struct sta_rec_sec_uni *sec;
 	struct mt792x_vif *mvif = msta->vif;
 	struct ieee80211_sta *sta;
@@ -1480,7 +1481,7 @@ mt7925_mcu_sta_amsdu_tlv(struct sk_buff *skb,
 	amsdu = (struct sta_rec_amsdu *)tlv;
 	amsdu->max_amsdu_num = 8;
 	amsdu->amsdu_en = true;
-	msta->wcid.amsdu = true;
+	msta->deflink.wcid.amsdu = true;
 
 	switch (sta->deflink.agg.max_amsdu_len) {
 	case IEEE80211_MAX_MPDU_LEN_VHT_11454:
@@ -1668,7 +1669,7 @@ int mt7925_mcu_sta_update(struct mt792x_dev *dev, struct ieee80211_sta *sta,
 	struct mt792x_sta *msta;
 
 	msta = sta ? (struct mt792x_sta *)sta->drv_priv : NULL;
-	info.wcid = msta ? &msta->wcid : &mvif->sta.wcid;
+	info.wcid = msta ? &msta->deflink.wcid : &mvif->sta.deflink.wcid;
 	info.newly = msta ? state != MT76_STA_INFO_STATE_ASSOC : true;
 
 	return mt7925_mcu_sta_cmd(&dev->mphy, &info);
@@ -2096,7 +2097,7 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 	basic_req->bcn_interval = cpu_to_le16(vif->bss_conf.beacon_int);
 	basic_req->dtim_period = vif->bss_conf.dtim_period;
 	basic_req->bmc_tx_wlan_idx = cpu_to_le16(wlan_idx);
-	basic_req->sta_idx = cpu_to_le16(msta->wcid.idx);
+	basic_req->sta_idx = cpu_to_le16(msta->deflink.wcid.idx);
 	basic_req->omac_idx = mvif->bss_conf.mt76.omac_idx;
 	basic_req->band_idx = mvif->bss_conf.mt76.band_idx;
 	basic_req->wmm_idx = mvif->bss_conf.mt76.wmm_idx;
@@ -2330,7 +2331,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 
 	/* bss_basic must be first */
 	mt7925_mcu_bss_basic_tlv(skb, vif, sta, ctx, phy->mt76,
-				 mvif->sta.wcid.idx, enable);
+				 mvif->sta.deflink.wcid.idx, enable);
 	mt7925_mcu_bss_sec_tlv(skb, vif);
 
 	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, vif, sta);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c b/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c
index 9fca887977d2..ef631a68005d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c
@@ -34,9 +34,9 @@ int mt7925e_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr,
 	if (sta) {
 		struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 
-		if (time_after(jiffies, msta->last_txs + HZ / 4)) {
+		if (time_after(jiffies, msta->deflink.last_txs + HZ / 4)) {
 			info->flags |= IEEE80211_TX_CTL_REQ_TX_STATUS;
-			msta->last_txs = jiffies;
+			msta->deflink.last_txs = jiffies;
 		}
 	}
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index 27bfda04d0f2..97865404444d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -81,11 +81,9 @@ enum mt792x_reg_power_type {
 
 DECLARE_EWMA(avg_signal, 10, 8)
 
-struct mt792x_sta {
+struct mt792x_link_sta {
 	struct mt76_wcid wcid; /* must be first */
 
-	struct mt792x_vif *vif;
-
 	u32 airtime_ac[8];
 
 	int ack_signal;
@@ -96,6 +94,12 @@ struct mt792x_sta {
 	struct mt76_connac_sta_key_conf bip;
 };
 
+struct mt792x_sta {
+	struct mt792x_link_sta deflink; /* must be first */
+
+	struct mt792x_vif *vif;
+};
+
 DECLARE_EWMA(rssi, 10, 8);
 
 struct mt792x_bss_conf {
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_core.c b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
index 9eeaf3344217..dbe98d4dbf94 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_core.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
@@ -65,14 +65,14 @@ void mt792x_tx(struct ieee80211_hw *hw, struct ieee80211_tx_control *control,
 		struct mt792x_sta *sta;
 
 		sta = (struct mt792x_sta *)control->sta->drv_priv;
-		wcid = &sta->wcid;
+		wcid = &sta->deflink.wcid;
 	}
 
 	if (vif && !control->sta) {
 		struct mt792x_vif *mvif;
 
 		mvif = (struct mt792x_vif *)vif->drv_priv;
-		wcid = &mvif->sta.wcid;
+		wcid = &mvif->sta.deflink.wcid;
 	}
 
 	if (mt76_connac_pm_ref(mphy, &dev->pm)) {
@@ -120,11 +120,11 @@ void mt792x_remove_interface(struct ieee80211_hw *hw,
 	struct mt792x_sta *msta = &mvif->sta;
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
 	struct mt792x_phy *phy = mt792x_hw_phy(hw);
-	int idx = msta->wcid.idx;
+	int idx = msta->deflink.wcid.idx;
 
 	mt792x_mutex_acquire(dev);
-	mt76_connac_free_pending_tx_skbs(&dev->pm, &msta->wcid);
-	mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.wcid, false);
+	mt76_connac_free_pending_tx_skbs(&dev->pm, &msta->deflink.wcid);
+	mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.deflink.wcid, false);
 
 	rcu_assign_pointer(dev->mt76.wcid[idx], NULL);
 
@@ -133,11 +133,11 @@ void mt792x_remove_interface(struct ieee80211_hw *hw,
 	mt792x_mutex_release(dev);
 
 	spin_lock_bh(&dev->mt76.sta_poll_lock);
-	if (!list_empty(&msta->wcid.poll_list))
-		list_del_init(&msta->wcid.poll_list);
+	if (!list_empty(&msta->deflink.wcid.poll_list))
+		list_del_init(&msta->deflink.wcid.poll_list);
 	spin_unlock_bh(&dev->mt76.sta_poll_lock);
 
-	mt76_wcid_cleanup(&dev->mt76, &msta->wcid);
+	mt76_wcid_cleanup(&dev->mt76, &msta->deflink.wcid);
 }
 EXPORT_SYMBOL_GPL(mt792x_remove_interface);
 
@@ -408,7 +408,7 @@ mt792x_ethtool_worker(void *wi_data, struct ieee80211_sta *sta)
 	if (msta->vif->bss_conf.mt76.idx != wi->idx)
 		return;
 
-	mt76_ethtool_worker(wi, &msta->wcid.stats, true);
+	mt76_ethtool_worker(wi, &msta->deflink.wcid.stats, true);
 }
 
 void mt792x_get_et_stats(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
@@ -487,7 +487,7 @@ void mt792x_sta_statistics(struct ieee80211_hw *hw,
 			   struct station_info *sinfo)
 {
 	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
-	struct rate_info *txrate = &msta->wcid.rate;
+	struct rate_info *txrate = &msta->deflink.wcid.rate;
 
 	if (!txrate->legacy && !txrate->flags)
 		return;
@@ -502,19 +502,19 @@ void mt792x_sta_statistics(struct ieee80211_hw *hw,
 		sinfo->txrate.he_dcm = txrate->he_dcm;
 		sinfo->txrate.he_ru_alloc = txrate->he_ru_alloc;
 	}
-	sinfo->tx_failed = msta->wcid.stats.tx_failed;
+	sinfo->tx_failed = msta->deflink.wcid.stats.tx_failed;
 	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_TX_FAILED);
 
-	sinfo->tx_retries = msta->wcid.stats.tx_retries;
+	sinfo->tx_retries = msta->deflink.wcid.stats.tx_retries;
 	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_TX_RETRIES);
 
 	sinfo->txrate.flags = txrate->flags;
 	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_TX_BITRATE);
 
-	sinfo->ack_signal = (s8)msta->ack_signal;
+	sinfo->ack_signal = (s8)msta->deflink.ack_signal;
 	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_ACK_SIGNAL);
 
-	sinfo->avg_ack_signal = -(s8)ewma_avg_signal_read(&msta->avg_ack_signal);
+	sinfo->avg_ack_signal = -(s8)ewma_avg_signal_read(&msta->deflink.avg_ack_signal);
 	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_ACK_SIGNAL_AVG);
 }
 EXPORT_SYMBOL_GPL(mt792x_sta_statistics);
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_mac.c b/drivers/net/wireless/mediatek/mt76/mt792x_mac.c
index 45dd94ff386d..106273935b26 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_mac.c
@@ -138,6 +138,7 @@ EXPORT_SYMBOL_GPL(mt792x_mac_update_mib_stats);
 struct mt76_wcid *mt792x_rx_get_wcid(struct mt792x_dev *dev, u16 idx,
 				     bool unicast)
 {
+	struct mt792x_link_sta *link;
 	struct mt792x_sta *sta;
 	struct mt76_wcid *wcid;
 
@@ -151,11 +152,12 @@ struct mt76_wcid *mt792x_rx_get_wcid(struct mt792x_dev *dev, u16 idx,
 	if (!wcid->sta)
 		return NULL;
 
-	sta = container_of(wcid, struct mt792x_sta, wcid);
+	link = container_of(wcid, struct mt792x_link_sta, wcid);
+	sta = container_of(link, struct mt792x_sta, deflink);
 	if (!sta->vif)
 		return NULL;
 
-	return &sta->vif->sta.wcid;
+	return &sta->vif->sta.deflink.wcid;
 }
 EXPORT_SYMBOL_GPL(mt792x_rx_get_wcid);
 
-- 
cgit v1.2.3-58-ga151


From f3898da2e86e43b83165347acf9194914a9146e0 Mon Sep 17 00:00:00 2001
From: Deren Wu <deren.wu@mediatek.com>
Date: Wed, 12 Jun 2024 20:01:57 -0700
Subject: wifi: mt76: mt792x: add struct mt792x_chanctx

We are introducing a new structure, mt792x_chanctx, to quickly identify
the linked BSS it is working on. This eliminates the need to search in
mt7921_ctx_iter() or mt7925_ctx_iter() when the channel context changes.

Co-developed-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-4-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7921/main.c | 31 ++++++++++--------------
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 31 ++++++++++--------------
 drivers/net/wireless/mediatek/mt76/mt792x.h      |  4 +++
 drivers/net/wireless/mediatek/mt76/mt792x_core.c |  5 ++++
 4 files changed, 35 insertions(+), 36 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/main.c b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
index f83cb45957d2..5dd9a7507f0b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
@@ -1346,32 +1346,27 @@ mt7921_remove_chanctx(struct ieee80211_hw *hw,
 {
 }
 
-static void mt7921_ctx_iter(void *priv, u8 *mac,
-			    struct ieee80211_vif *vif)
-{
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	struct ieee80211_chanctx_conf *ctx = priv;
-
-	if (ctx != mvif->bss_conf.mt76.ctx)
-		return;
-
-	if (vif->type == NL80211_IFTYPE_MONITOR)
-		mt7921_mcu_config_sniffer(mvif, ctx);
-	else
-		mt76_connac_mcu_uni_set_chctx(mvif->phy->mt76, &mvif->bss_conf.mt76, ctx);
-}
-
 static void
 mt7921_change_chanctx(struct ieee80211_hw *hw,
 		      struct ieee80211_chanctx_conf *ctx,
 		      u32 changed)
 {
+	struct mt792x_chanctx *mctx = (struct mt792x_chanctx *)ctx->drv_priv;
 	struct mt792x_phy *phy = mt792x_hw_phy(hw);
+	struct ieee80211_vif *vif;
+	struct mt792x_vif *mvif;
+
+	if (!mctx->bss_conf)
+		return;
+
+	mvif = container_of(mctx->bss_conf, struct mt792x_vif, bss_conf);
+	vif = container_of((void *)mvif, struct ieee80211_vif, drv_priv);
 
 	mt792x_mutex_acquire(phy->dev);
-	ieee80211_iterate_active_interfaces(phy->mt76->hw,
-					    IEEE80211_IFACE_ITER_ACTIVE,
-					    mt7921_ctx_iter, ctx);
+	if (vif->type == NL80211_IFTYPE_MONITOR)
+		mt7921_mcu_config_sniffer(mvif, ctx);
+	else
+		mt76_connac_mcu_uni_set_chctx(mvif->phy->mt76, &mvif->bss_conf.mt76, ctx);
 	mt792x_mutex_release(phy->dev);
 }
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 504c2ecc3845..91edbe076c22 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1354,34 +1354,29 @@ mt7925_remove_chanctx(struct ieee80211_hw *hw,
 {
 }
 
-static void mt7925_ctx_iter(void *priv, u8 *mac,
-			    struct ieee80211_vif *vif)
+static void
+mt7925_change_chanctx(struct ieee80211_hw *hw,
+		      struct ieee80211_chanctx_conf *ctx,
+		      u32 changed)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	struct ieee80211_chanctx_conf *ctx = priv;
+	struct mt792x_chanctx *mctx = (struct mt792x_chanctx *)ctx->drv_priv;
+	struct mt792x_phy *phy = mt792x_hw_phy(hw);
+	struct ieee80211_vif *vif;
+	struct mt792x_vif *mvif;
 
-	if (ctx != mvif->bss_conf.mt76.ctx)
+	if (!mctx->bss_conf)
 		return;
 
+	mvif = container_of(mctx->bss_conf, struct mt792x_vif, bss_conf);
+	vif = container_of((void *)mvif, struct ieee80211_vif, drv_priv);
+
+	mt792x_mutex_acquire(phy->dev);
 	if (vif->type == NL80211_IFTYPE_MONITOR) {
 		mt7925_mcu_set_sniffer(mvif->phy->dev, vif, true);
 		mt7925_mcu_config_sniffer(mvif, ctx);
 	} else {
 		mt7925_mcu_set_chctx(mvif->phy->mt76, &mvif->bss_conf.mt76, ctx);
 	}
-}
-
-static void
-mt7925_change_chanctx(struct ieee80211_hw *hw,
-		      struct ieee80211_chanctx_conf *ctx,
-		      u32 changed)
-{
-	struct mt792x_phy *phy = mt792x_hw_phy(hw);
-
-	mt792x_mutex_acquire(phy->dev);
-	ieee80211_iterate_active_interfaces(phy->mt76->hw,
-					    IEEE80211_IFACE_ITER_ACTIVE,
-					    mt7925_ctx_iter, ctx);
 	mt792x_mutex_release(phy->dev);
 }
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index 97865404444d..eee872aa0108 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -102,6 +102,10 @@ struct mt792x_sta {
 
 DECLARE_EWMA(rssi, 10, 8);
 
+struct mt792x_chanctx {
+	struct mt792x_bss_conf *bss_conf;
+};
+
 struct mt792x_bss_conf {
 	struct mt76_vif mt76; /* must be first */
 	struct ewma_rssi rssi;
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_core.c b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
index dbe98d4dbf94..4d0e2def1c1d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_core.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
@@ -261,11 +261,13 @@ int mt792x_assign_vif_chanctx(struct ieee80211_hw *hw,
 			      struct ieee80211_bss_conf *link_conf,
 			      struct ieee80211_chanctx_conf *ctx)
 {
+	struct mt792x_chanctx *mctx = (struct mt792x_chanctx *)ctx->drv_priv;
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
 
 	mutex_lock(&dev->mt76.mutex);
 	mvif->bss_conf.mt76.ctx = ctx;
+	mctx->bss_conf = &mvif->bss_conf;
 	mutex_unlock(&dev->mt76.mutex);
 
 	return 0;
@@ -277,10 +279,12 @@ void mt792x_unassign_vif_chanctx(struct ieee80211_hw *hw,
 				 struct ieee80211_bss_conf *link_conf,
 				 struct ieee80211_chanctx_conf *ctx)
 {
+	struct mt792x_chanctx *mctx = (struct mt792x_chanctx *)ctx->drv_priv;
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
 
 	mutex_lock(&dev->mt76.mutex);
+	mctx->bss_conf = NULL;
 	mvif->bss_conf.mt76.ctx = NULL;
 	mutex_unlock(&dev->mt76.mutex);
 }
@@ -556,6 +560,7 @@ int mt792x_init_wiphy(struct ieee80211_hw *hw)
 
 	hw->sta_data_size = sizeof(struct mt792x_sta);
 	hw->vif_data_size = sizeof(struct mt792x_vif);
+	hw->chanctx_data_size = sizeof(struct mt792x_chanctx);
 
 	if (dev->fw_features & MT792x_FW_CAP_CNM) {
 		wiphy->flags |= WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL;
-- 
cgit v1.2.3-58-ga151


From 43626f0e0c991593ff196a886ccb9455701e4e2e Mon Sep 17 00:00:00 2001
From: Deren Wu <deren.wu@mediatek.com>
Date: Wed, 12 Jun 2024 20:01:58 -0700
Subject: wifi: mt76: mt7925: support for split bss_info_changed method

The .bss_info_changed method is being split into .vif_cfg_changed and
.link_info_changed to support MLD devices, but it will still support
non-MLD devices.

Co-developed-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-5-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 132 ++++++++++++-----------
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c  |   7 +-
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.h  |   3 +-
 3 files changed, 76 insertions(+), 66 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 91edbe076c22..397776ddf01a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -695,65 +695,6 @@ mt7925_get_rates_table(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	return mvif->basic_rates_idx;
 }
 
-static void mt7925_bss_info_changed(struct ieee80211_hw *hw,
-				    struct ieee80211_vif *vif,
-				    struct ieee80211_bss_conf *info,
-				    u64 changed)
-{
-	struct mt76_vif *mvif = (struct mt76_vif *)vif->drv_priv;
-	struct mt792x_phy *phy = mt792x_hw_phy(hw);
-	struct mt792x_dev *dev = mt792x_hw_dev(hw);
-
-	mt792x_mutex_acquire(dev);
-
-	if (changed & BSS_CHANGED_ERP_SLOT) {
-		int slottime = info->use_short_slot ? 9 : 20;
-
-		if (slottime != phy->slottime) {
-			phy->slottime = slottime;
-			mt7925_mcu_set_timing(phy, vif);
-		}
-	}
-
-	if (changed & BSS_CHANGED_MCAST_RATE)
-		mvif->mcast_rates_idx =
-				mt7925_get_rates_table(hw, vif, false, true);
-
-	if (changed & BSS_CHANGED_BASIC_RATES)
-		mvif->basic_rates_idx =
-				mt7925_get_rates_table(hw, vif, false, false);
-
-	if (changed & (BSS_CHANGED_BEACON |
-		       BSS_CHANGED_BEACON_ENABLED)) {
-		mvif->beacon_rates_idx =
-				mt7925_get_rates_table(hw, vif, true, false);
-
-		mt7925_mcu_uni_add_beacon_offload(dev, hw, vif,
-						  info->enable_beacon);
-	}
-
-	/* ensure that enable txcmd_mode after bss_info */
-	if (changed & (BSS_CHANGED_QOS | BSS_CHANGED_BEACON_ENABLED))
-		mt7925_mcu_set_tx(dev, vif);
-
-	if (changed & BSS_CHANGED_PS)
-		mt7925_mcu_uni_bss_ps(dev, vif);
-
-	if (changed & BSS_CHANGED_ASSOC) {
-		mt7925_mcu_sta_update(dev, NULL, vif, true,
-				      MT76_STA_INFO_STATE_ASSOC);
-		mt7925_mcu_set_beacon_filter(dev, vif, vif->cfg.assoc);
-	}
-
-	if (changed & BSS_CHANGED_ARP_FILTER) {
-		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-
-		mt7925_mcu_update_arp_filter(&dev->mt76, &mvif->bss_conf.mt76, info);
-	}
-
-	mt792x_mutex_release(dev);
-}
-
 int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta)
 {
@@ -1404,6 +1345,76 @@ static void mt7925_mgd_complete_tx(struct ieee80211_hw *hw,
 	mt7925_abort_roc(mvif->phy, mvif);
 }
 
+static void mt7925_vif_cfg_changed(struct ieee80211_hw *hw,
+				   struct ieee80211_vif *vif,
+				   u64 changed)
+{
+	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+
+	mt792x_mutex_acquire(dev);
+
+	if (changed & BSS_CHANGED_ASSOC) {
+		mt7925_mcu_sta_update(dev, NULL, vif, true,
+				      MT76_STA_INFO_STATE_ASSOC);
+		mt7925_mcu_set_beacon_filter(dev, vif, vif->cfg.assoc);
+	}
+
+	if (changed & BSS_CHANGED_ARP_FILTER) {
+		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+
+		mt7925_mcu_update_arp_filter(&dev->mt76, &mvif->bss_conf.mt76);
+	}
+
+	if (changed & BSS_CHANGED_PS)
+		mt7925_mcu_uni_bss_ps(dev, vif);
+
+	mt792x_mutex_release(dev);
+}
+
+static void mt7925_link_info_changed(struct ieee80211_hw *hw,
+				     struct ieee80211_vif *vif,
+				     struct ieee80211_bss_conf *info,
+				     u64 changed)
+{
+	struct mt76_vif *mvif = (struct mt76_vif *)vif->drv_priv;
+	struct mt792x_phy *phy = mt792x_hw_phy(hw);
+	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+
+	mt792x_mutex_acquire(dev);
+
+	if (changed & BSS_CHANGED_ERP_SLOT) {
+		int slottime = info->use_short_slot ? 9 : 20;
+
+		if (slottime != phy->slottime) {
+			phy->slottime = slottime;
+			mt7925_mcu_set_timing(phy, vif);
+		}
+	}
+
+	if (changed & BSS_CHANGED_MCAST_RATE)
+		mvif->mcast_rates_idx =
+				mt7925_get_rates_table(hw, vif, false, true);
+
+	if (changed & BSS_CHANGED_BASIC_RATES)
+		mvif->basic_rates_idx =
+				mt7925_get_rates_table(hw, vif, false, false);
+
+	if (changed & (BSS_CHANGED_BEACON |
+		       BSS_CHANGED_BEACON_ENABLED)) {
+		mvif->beacon_rates_idx =
+				mt7925_get_rates_table(hw, vif, true, false);
+
+		mt7925_mcu_uni_add_beacon_offload(dev, hw, vif,
+						  info->enable_beacon);
+	}
+
+	/* ensure that enable txcmd_mode after bss_info */
+	if (changed & (BSS_CHANGED_QOS | BSS_CHANGED_BEACON_ENABLED))
+		mt7925_mcu_set_tx(dev, vif);
+
+	mt792x_mutex_release(dev);
+}
+
 const struct ieee80211_ops mt7925_ops = {
 	.tx = mt792x_tx,
 	.start = mt7925_start,
@@ -1413,7 +1424,6 @@ const struct ieee80211_ops mt7925_ops = {
 	.config = mt7925_config,
 	.conf_tx = mt7925_conf_tx,
 	.configure_filter = mt7925_configure_filter,
-	.bss_info_changed = mt7925_bss_info_changed,
 	.start_ap = mt7925_start_ap,
 	.stop_ap = mt7925_stop_ap,
 	.sta_state = mt76_sta_state,
@@ -1461,6 +1471,8 @@ const struct ieee80211_ops mt7925_ops = {
 	.unassign_vif_chanctx = mt792x_unassign_vif_chanctx,
 	.mgd_prepare_tx = mt7925_mgd_prepare_tx,
 	.mgd_complete_tx = mt7925_mgd_complete_tx,
+	.vif_cfg_changed = mt7925_vif_cfg_changed,
+	.link_info_changed = mt7925_link_info_changed,
 };
 EXPORT_SYMBOL_GPL(mt7925_ops);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 67c05f785190..5e72f7278e8b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -121,11 +121,10 @@ int mt7925_mcu_regval(struct mt792x_dev *dev, u32 regidx, u32 *val, bool set)
 EXPORT_SYMBOL_GPL(mt7925_mcu_regval);
 
 int mt7925_mcu_update_arp_filter(struct mt76_dev *dev,
-				 struct mt76_vif *vif,
-				 struct ieee80211_bss_conf *info)
+				 struct mt76_vif *vif)
 {
-	struct ieee80211_vif *mvif = container_of(info, struct ieee80211_vif,
-						  bss_conf);
+	struct ieee80211_vif *mvif = container_of((void *)vif, struct ieee80211_vif,
+						  drv_priv);
 	struct sk_buff *skb;
 	int i, len = min_t(int, mvif->cfg.arp_addr_cnt,
 			   IEEE80211_BSS_ARP_ADDR_LIST_LEN);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
index b8315a89f4a9..b499b1d0531a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
@@ -590,6 +590,5 @@ int mt7925_mcu_set_chctx(struct mt76_phy *phy, struct mt76_vif *mvif,
 			 struct ieee80211_chanctx_conf *ctx);
 int mt7925_mcu_set_rate_txpower(struct mt76_phy *phy);
 int mt7925_mcu_update_arp_filter(struct mt76_dev *dev,
-				 struct mt76_vif *vif,
-				 struct ieee80211_bss_conf *info);
+				 struct mt76_vif *vif);
 #endif
-- 
cgit v1.2.3-58-ga151


From 7cebb6a66ac6a08b5fc80c1e6cfb337473f4f4d2 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:01:59 -0700
Subject: wifi: mt76: mt792x: extend mt76_connac_mcu_uni_add_dev for per-link
 BSS

Extend mt76_connac_mcu_uni_add_dev with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

We also extend link_idx field in mt76_connac_bss_basic_tlv for the firmware
to able to identify the link index in the MLO mode that is not harmful
for the current non-MLO mode.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-6-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7615/mcu.c      |  4 ++--
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 12 +++++++-----
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h |  4 ++--
 drivers/net/wireless/mediatek/mt76/mt7921/mac.c      |  3 ++-
 drivers/net/wireless/mediatek/mt76/mt7921/main.c     |  4 ++--
 drivers/net/wireless/mediatek/mt76/mt7925/mac.c      |  3 ++-
 drivers/net/wireless/mediatek/mt76/mt7925/main.c     |  4 ++--
 drivers/net/wireless/mediatek/mt76/mt792x_core.c     |  2 +-
 8 files changed, 20 insertions(+), 16 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
index c807bd8d928d..792ae41f7c69 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
@@ -1109,8 +1109,8 @@ mt7615_mcu_uni_add_dev(struct mt7615_phy *phy, struct ieee80211_vif *vif,
 {
 	struct mt7615_vif *mvif = (struct mt7615_vif *)vif->drv_priv;
 
-	return mt76_connac_mcu_uni_add_dev(phy->mt76, vif, &mvif->sta.wcid,
-					   enable);
+	return mt76_connac_mcu_uni_add_dev(phy->mt76, &vif->bss_conf,
+					   &mvif->sta.wcid, enable);
 }
 
 static int
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
index 162c57fb7954..a059532f383a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
@@ -1132,11 +1132,11 @@ void mt76_connac_mcu_wtbl_ba_tlv(struct mt76_dev *dev, struct sk_buff *skb,
 EXPORT_SYMBOL_GPL(mt76_connac_mcu_wtbl_ba_tlv);
 
 int mt76_connac_mcu_uni_add_dev(struct mt76_phy *phy,
-				struct ieee80211_vif *vif,
+				struct ieee80211_bss_conf *bss_conf,
 				struct mt76_wcid *wcid,
 				bool enable)
 {
-	struct mt76_vif *mvif = (struct mt76_vif *)vif->drv_priv;
+	struct mt76_vif *mvif = (struct mt76_vif *)bss_conf->vif->drv_priv;
 	struct mt76_dev *dev = phy->dev;
 	struct {
 		struct {
@@ -1148,7 +1148,7 @@ int mt76_connac_mcu_uni_add_dev(struct mt76_phy *phy,
 			__le16 tag;
 			__le16 len;
 			u8 active;
-			u8 pad;
+			u8 link_idx; /* not link_id */
 			u8 omac_addr[ETH_ALEN];
 		} __packed tlv;
 	} dev_req = {
@@ -1160,6 +1160,7 @@ int mt76_connac_mcu_uni_add_dev(struct mt76_phy *phy,
 			.tag = cpu_to_le16(DEV_INFO_ACTIVE),
 			.len = cpu_to_le16(sizeof(struct req_tlv)),
 			.active = enable,
+			.link_idx = mvif->idx,
 		},
 	};
 	struct {
@@ -1182,12 +1183,13 @@ int mt76_connac_mcu_uni_add_dev(struct mt76_phy *phy,
 			.bmc_tx_wlan_idx = cpu_to_le16(wcid->idx),
 			.sta_idx = cpu_to_le16(wcid->idx),
 			.conn_state = 1,
+			.link_idx = mvif->idx,
 		},
 	};
 	int err, idx, cmd, len;
 	void *data;
 
-	switch (vif->type) {
+	switch (bss_conf->vif->type) {
 	case NL80211_IFTYPE_MESH_POINT:
 	case NL80211_IFTYPE_MONITOR:
 	case NL80211_IFTYPE_AP:
@@ -1207,7 +1209,7 @@ int mt76_connac_mcu_uni_add_dev(struct mt76_phy *phy,
 	idx = mvif->omac_idx > EXT_BSSID_START ? HW_BSSID_0 : mvif->omac_idx;
 	basic_req.basic.hw_bss_idx = idx;
 
-	memcpy(dev_req.tlv.omac_addr, vif->addr, ETH_ALEN);
+	memcpy(dev_req.tlv.omac_addr, bss_conf->vif->addr, ETH_ALEN);
 
 	cmd = enable ? MCU_UNI_CMD(DEV_INFO_UPDATE) : MCU_UNI_CMD(BSS_INFO_UPDATE);
 	data = enable ? (void *)&dev_req : (void *)&basic_req;
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index 6873ce14d299..5c7c42c95de1 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -1443,7 +1443,7 @@ struct mt76_connac_bss_basic_tlv {
 	__le16 sta_idx;
 	__le16 nonht_basic_phy;
 	u8 phymode_ext; /* bit(0) AX_6G */
-	u8 pad[1];
+	u8 link_idx;
 } __packed;
 
 struct mt76_connac_bss_qos_tlv {
@@ -1917,7 +1917,7 @@ void mt76_connac_mcu_sta_ba_tlv(struct sk_buff *skb,
 				struct ieee80211_ampdu_params *params,
 				bool enable, bool tx);
 int mt76_connac_mcu_uni_add_dev(struct mt76_phy *phy,
-				struct ieee80211_vif *vif,
+				struct ieee80211_bss_conf *bss_conf,
 				struct mt76_wcid *wcid,
 				bool enable);
 int mt76_connac_mcu_sta_ba(struct mt76_dev *dev, struct mt76_vif *mvif,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
index 9ae9590999f4..047106b65d2b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/mac.c
@@ -646,7 +646,8 @@ mt7921_vif_connect_iter(void *priv, u8 *mac,
 	if (vif->type == NL80211_IFTYPE_STATION)
 		ieee80211_disconnect(vif, true);
 
-	mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.deflink.wcid, true);
+	mt76_connac_mcu_uni_add_dev(&dev->mphy, &vif->bss_conf,
+				    &mvif->sta.deflink.wcid, true);
 	mt7921_mcu_set_tx(dev, vif);
 
 	if (vif->type == NL80211_IFTYPE_AP) {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/main.c b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
index 5dd9a7507f0b..2e6268cb06c0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/main.c
@@ -306,8 +306,8 @@ mt7921_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 	mvif->bss_conf.mt76.band_idx = 0;
 	mvif->bss_conf.mt76.wmm_idx = mvif->bss_conf.mt76.idx % MT76_CONNAC_MAX_WMM_SETS;
 
-	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.deflink.wcid,
-					  true);
+	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, &vif->bss_conf,
+					  &mvif->sta.deflink.wcid, true);
 	if (ret)
 		goto out;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
index 88e7b2ff4593..48425fc4e7fe 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
@@ -1243,7 +1243,8 @@ mt7925_vif_connect_iter(void *priv, u8 *mac,
 	if (vif->type == NL80211_IFTYPE_STATION)
 		ieee80211_disconnect(vif, true);
 
-	mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.deflink.wcid, true);
+	mt76_connac_mcu_uni_add_dev(&dev->mphy, &vif->bss_conf,
+				    &mvif->sta.deflink.wcid, true);
 	mt7925_mcu_set_tx(dev, vif);
 
 	if (vif->type == NL80211_IFTYPE_AP) {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 397776ddf01a..b0c8b090d2ce 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -344,8 +344,8 @@ mt7925_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 	else
 		mvif->bss_conf.mt76.basic_rates_idx = MT792x_BASIC_RATES_TBL;
 
-	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.deflink.wcid,
-					  true);
+	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, &vif->bss_conf,
+					  &mvif->sta.deflink.wcid, true);
 	if (ret)
 		goto out;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_core.c b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
index 4d0e2def1c1d..b1c51e3f39d0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_core.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
@@ -124,7 +124,7 @@ void mt792x_remove_interface(struct ieee80211_hw *hw,
 
 	mt792x_mutex_acquire(dev);
 	mt76_connac_free_pending_tx_skbs(&dev->pm, &msta->deflink.wcid);
-	mt76_connac_mcu_uni_add_dev(&dev->mphy, vif, &mvif->sta.deflink.wcid, false);
+	mt76_connac_mcu_uni_add_dev(&dev->mphy, &vif->bss_conf, &mvif->sta.deflink.wcid, false);
 
 	rcu_assign_pointer(dev->mt76.wcid[idx], NULL);
 
-- 
cgit v1.2.3-58-ga151


From acdfc3e79899871ef8146efd9be74a9b78858b67 Mon Sep 17 00:00:00 2001
From: Deren Wu <deren.wu@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:00 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_set_tx with for per-link BSS

Extend mt7925_mcu_set_tx with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-7-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mac.c    |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7925/main.c   |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    |  9 ++++----
 drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h |  2 +-
 drivers/net/wireless/mediatek/mt76/mt792x.h        | 25 ++++++++++++++++++++++
 5 files changed, 33 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
index 48425fc4e7fe..2be2ab914411 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
@@ -1245,7 +1245,7 @@ mt7925_vif_connect_iter(void *priv, u8 *mac,
 
 	mt76_connac_mcu_uni_add_dev(&dev->mphy, &vif->bss_conf,
 				    &mvif->sta.deflink.wcid, true);
-	mt7925_mcu_set_tx(dev, vif);
+	mt7925_mcu_set_tx(dev, &vif->bss_conf);
 
 	if (vif->type == NL80211_IFTYPE_AP) {
 		mt76_connac_mcu_uni_add_bss(dev->phy.mt76, vif, &mvif->sta.deflink.wcid,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index b0c8b090d2ce..d11919688bae 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1410,7 +1410,7 @@ static void mt7925_link_info_changed(struct ieee80211_hw *hw,
 
 	/* ensure that enable txcmd_mode after bss_info */
 	if (changed & (BSS_CHANGED_QOS | BSS_CHANGED_BEACON_ENABLED))
-		mt7925_mcu_set_tx(dev, vif);
+		mt7925_mcu_set_tx(dev, info);
 
 	mt792x_mutex_release(dev);
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 5e72f7278e8b..1c262a024dc8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -900,7 +900,8 @@ int mt7925_mcu_wtbl_update_hdr_trans(struct mt792x_dev *dev,
 				     MCU_WMWA_UNI_CMD(STA_REC_UPDATE), true);
 }
 
-int mt7925_mcu_set_tx(struct mt792x_dev *dev, struct ieee80211_vif *vif)
+int mt7925_mcu_set_tx(struct mt792x_dev *dev,
+		      struct ieee80211_bss_conf *bss_conf)
 {
 #define MCU_EDCA_AC_PARAM	0
 #define WMM_AIFS_SET		BIT(0)
@@ -909,12 +910,12 @@ int mt7925_mcu_set_tx(struct mt792x_dev *dev, struct ieee80211_vif *vif)
 #define WMM_TXOP_SET		BIT(3)
 #define WMM_PARAM_SET		(WMM_AIFS_SET | WMM_CW_MIN_SET | \
 				 WMM_CW_MAX_SET | WMM_TXOP_SET)
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(bss_conf);
 	struct {
 		u8 bss_idx;
 		u8 __rsv[3];
 	} __packed hdr = {
-		.bss_idx = mvif->bss_conf.mt76.idx,
+		.bss_idx = mconf->mt76.idx,
 	};
 	struct sk_buff *skb;
 	int len = sizeof(hdr) + IEEE80211_NUM_ACS * sizeof(struct edca);
@@ -927,7 +928,7 @@ int mt7925_mcu_set_tx(struct mt792x_dev *dev, struct ieee80211_vif *vif)
 	skb_put_data(skb, &hdr, sizeof(hdr));
 
 	for (ac = 0; ac < IEEE80211_NUM_ACS; ac++) {
-		struct ieee80211_tx_queue_params *q = &mvif->bss_conf.queue_params[ac];
+		struct ieee80211_tx_queue_params *q = &mconf->queue_params[ac];
 		struct edca *e;
 		struct tlv *tlv;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
index 8a4a71f6bcb6..f689f7a359ea 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
@@ -198,7 +198,7 @@ int mt7925_mcu_sta_update(struct mt792x_dev *dev, struct ieee80211_sta *sta,
 			  struct ieee80211_vif *vif, bool enable,
 			  enum mt76_sta_info_state state);
 int mt7925_mcu_set_chan_info(struct mt792x_phy *phy, u16 tag);
-int mt7925_mcu_set_tx(struct mt792x_dev *dev, struct ieee80211_vif *vif);
+int mt7925_mcu_set_tx(struct mt792x_dev *dev, struct ieee80211_bss_conf *bss_conf);
 int mt7925_mcu_set_eeprom(struct mt792x_dev *dev);
 int mt7925_mcu_get_rx_rate(struct mt792x_phy *phy, struct ieee80211_vif *vif,
 			   struct ieee80211_sta *sta, struct rate_info *rate);
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index eee872aa0108..9630514ae25f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -114,6 +114,7 @@ struct mt792x_bss_conf {
 
 struct mt792x_vif {
 	struct mt792x_bss_conf bss_conf; /* must be first */
+	struct mt792x_bss_conf __rcu *link_conf[IEEE80211_MLD_MAX_NUM_LINKS];
 
 	struct mt792x_sta sta;
 	struct mt792x_sta *wep_sta;
@@ -221,6 +222,30 @@ struct mt792x_dev {
 	u32 backup_l2;
 };
 
+static inline struct mt792x_bss_conf *
+mt792x_vif_to_link(struct mt792x_vif *mvif, u8 link_id)
+{
+	struct ieee80211_vif *vif;
+
+	vif = container_of((void *)mvif, struct ieee80211_vif, drv_priv);
+
+	if (!ieee80211_vif_is_mld(vif) ||
+	    link_id >= IEEE80211_LINK_UNSPECIFIED)
+		return &mvif->bss_conf;
+
+	return rcu_dereference_protected(mvif->link_conf[link_id],
+		lockdep_is_held(&mvif->phy->dev->mt76.mutex));
+}
+
+static inline struct mt792x_bss_conf *
+mt792x_link_conf_to_mconf(struct ieee80211_bss_conf *link_conf)
+{
+	struct ieee80211_vif *vif = link_conf->vif;
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+
+	return mt792x_vif_to_link(mvif, link_conf->link_id);
+}
+
 static inline struct mt792x_dev *
 mt792x_hw_dev(struct ieee80211_hw *hw)
 {
-- 
cgit v1.2.3-58-ga151


From fa5f44463f51944a2f33caf92c47e2d02217e3f1 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:01 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_add_bss_info for per-link BSS

Extend mt7925_mcu_add_bss_info with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-8-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 35 ++++++++++++++++--------
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c  | 22 +++++++--------
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.h  |  2 +-
 drivers/net/wireless/mediatek/mt76/mt792x.h      | 10 +++++++
 4 files changed, 46 insertions(+), 23 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index d11919688bae..7b9deb586b34 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -487,9 +487,12 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	struct mt792x_sta *msta = sta ? (struct mt792x_sta *)sta->drv_priv :
 				  &mvif->sta;
 	struct mt76_wcid *wcid = &msta->deflink.wcid;
+	struct ieee80211_bss_conf *link_conf;
 	u8 *wcid_keyidx = &wcid->hw_key_idx;
 	int idx = key->keyidx, err = 0;
 
+	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
+
 	/* The hardware does not support per-STA RX GTK, fallback
 	 * to software mode for these.
 	 */
@@ -528,7 +531,8 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 		struct mt792x_phy *phy = mt792x_hw_phy(hw);
 
 		mvif->bss_conf.mt76.cipher = mt7925_mcu_get_cipher(key->cipher);
-		mt7925_mcu_add_bss_info(phy, mvif->bss_conf.mt76.ctx, vif, sta, true);
+		mt7925_mcu_add_bss_info(phy, mvif->bss_conf.mt76.ctx, link_conf,
+					sta, true);
 	}
 
 	if (cmd == SET_KEY)
@@ -701,6 +705,7 @@ int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
 	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct ieee80211_bss_conf *link_conf;
 	int ret, idx;
 
 	idx = mt76_wcid_alloc(dev->mt76.wcid_mask, MT792x_WTBL_STA - 1);
@@ -725,10 +730,12 @@ int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	mt7925_mac_wtbl_update(dev, idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
+	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
+
 	/* should update bss info before STA add */
 	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
-		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, vif, sta,
-					false);
+		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
+					link_conf, sta, false);
 
 	ret = mt7925_mcu_sta_update(dev, sta, vif, true,
 				    MT76_STA_INFO_STATE_NONE);
@@ -747,12 +754,15 @@ void mt7925_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
 	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct ieee80211_bss_conf *link_conf;
 
 	mt792x_mutex_acquire(dev);
 
+	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
+
 	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
-		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, vif, sta,
-					true);
+		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
+					link_conf, sta, true);
 
 	ewma_avg_signal_init(&msta->deflink.avg_ack_signal);
 
@@ -771,6 +781,7 @@ void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 {
 	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
 	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
+	struct ieee80211_bss_conf *link_conf;
 
 	mt76_connac_free_pending_tx_skbs(&dev->pm, &msta->deflink.wcid);
 	mt76_connac_pm_wake(&dev->mphy, &dev->pm);
@@ -779,14 +790,16 @@ void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	mt7925_mac_wtbl_update(dev, msta->deflink.wcid.idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
+	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
+
 	if (vif->type == NL80211_IFTYPE_STATION) {
 		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 
 		mvif->wep_sta = NULL;
 		ewma_rssi_init(&mvif->bss_conf.rssi);
 		if (!sta->tdls)
-			mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, vif, sta,
-						false);
+			mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
+						link_conf, sta, false);
 	}
 
 	spin_lock_bh(&mdev->sta_poll_lock);
@@ -1244,8 +1257,8 @@ mt7925_start_ap(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 
 	mt792x_mutex_acquire(dev);
 
-	err = mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, vif, NULL,
-				      true);
+	err = mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
+				      link_conf, NULL, true);
 	if (err)
 		goto out;
 
@@ -1275,8 +1288,8 @@ mt7925_stop_ap(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	if (err)
 		goto out;
 
-	mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, vif, NULL,
-				false);
+	mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx, link_conf,
+				NULL, false);
 
 out:
 	mt792x_mutex_release(dev);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 1c262a024dc8..fb261ee525be 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2315,11 +2315,11 @@ int mt7925_mcu_set_timing(struct mt792x_phy *phy,
 
 int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 			    struct ieee80211_chanctx_conf *ctx,
-			    struct ieee80211_vif *vif,
+			    struct ieee80211_bss_conf *link_conf,
 			    struct ieee80211_sta *sta,
 			    int enable)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_vif *mvif = (struct mt792x_vif *)link_conf->vif->drv_priv;
 	struct mt792x_dev *dev = phy->dev;
 	struct sk_buff *skb;
 	int err;
@@ -2330,18 +2330,18 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 		return PTR_ERR(skb);
 
 	/* bss_basic must be first */
-	mt7925_mcu_bss_basic_tlv(skb, vif, sta, ctx, phy->mt76,
+	mt7925_mcu_bss_basic_tlv(skb, link_conf->vif, sta, ctx, phy->mt76,
 				 mvif->sta.deflink.wcid.idx, enable);
-	mt7925_mcu_bss_sec_tlv(skb, vif);
+	mt7925_mcu_bss_sec_tlv(skb, link_conf->vif);
 
-	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, vif, sta);
-	mt7925_mcu_bss_qos_tlv(skb, vif);
-	mt7925_mcu_bss_mld_tlv(skb, vif, sta);
-	mt7925_mcu_bss_ifs_tlv(skb, vif);
+	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf->vif, sta);
+	mt7925_mcu_bss_qos_tlv(skb, link_conf->vif);
+	mt7925_mcu_bss_mld_tlv(skb, link_conf->vif, sta);
+	mt7925_mcu_bss_ifs_tlv(skb, link_conf->vif);
 
-	if (vif->bss_conf.he_support) {
-		mt7925_mcu_bss_he_tlv(skb, vif, phy);
-		mt7925_mcu_bss_color_tlv(skb, vif, enable);
+	if (link_conf->he_support) {
+		mt7925_mcu_bss_he_tlv(skb, link_conf->vif, phy);
+		mt7925_mcu_bss_color_tlv(skb, link_conf->vif, enable);
 	}
 
 	err = mt76_mcu_skb_send_msg(&dev->mt76, skb,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
index b499b1d0531a..c692996e057c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
@@ -578,7 +578,7 @@ int mt7925_mcu_sched_scan_enable(struct mt76_phy *phy,
 				 bool enable);
 int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 			    struct ieee80211_chanctx_conf *ctx,
-			    struct ieee80211_vif *vif,
+			    struct ieee80211_bss_conf *link_conf,
 			    struct ieee80211_sta *sta,
 			    int enable);
 int mt7925_mcu_set_timing(struct mt792x_phy *phy,
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index 9630514ae25f..718c6b185d99 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -246,6 +246,16 @@ mt792x_link_conf_to_mconf(struct ieee80211_bss_conf *link_conf)
 	return mt792x_vif_to_link(mvif, link_conf->link_id);
 }
 
+static inline struct ieee80211_bss_conf *
+mt792x_vif_to_bss_conf(struct ieee80211_vif *vif, unsigned int link_id)
+{
+	if (!ieee80211_vif_is_mld(vif) ||
+	    link_id >= IEEE80211_LINK_UNSPECIFIED)
+		return &vif->bss_conf;
+
+	return link_conf_dereference_protected(vif, link_id);
+}
+
 static inline struct mt792x_dev *
 mt792x_hw_dev(struct ieee80211_hw *hw)
 {
-- 
cgit v1.2.3-58-ga151


From b8b04b6616bacca2ddd3eeefeb54b1bcdb2d7f4b Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:02 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_set_timing for per-link BSS

Extend mt7925_mcu_set_timing with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-9-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 2 +-
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c  | 6 +++---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.h  | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 7b9deb586b34..6a1afff92399 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1400,7 +1400,7 @@ static void mt7925_link_info_changed(struct ieee80211_hw *hw,
 
 		if (slottime != phy->slottime) {
 			phy->slottime = slottime;
-			mt7925_mcu_set_timing(phy, vif);
+			mt7925_mcu_set_timing(phy, info);
 		}
 	}
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index fb261ee525be..b82cc1bd09b6 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2296,9 +2296,9 @@ mt7925_mcu_bss_ifs_tlv(struct sk_buff *skb, struct ieee80211_vif *vif)
 }
 
 int mt7925_mcu_set_timing(struct mt792x_phy *phy,
-			  struct ieee80211_vif *vif)
+			  struct ieee80211_bss_conf *link_conf)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_vif *mvif = (struct mt792x_vif *)link_conf->vif->drv_priv;
 	struct mt792x_dev *dev = phy->dev;
 	struct sk_buff *skb;
 
@@ -2307,7 +2307,7 @@ int mt7925_mcu_set_timing(struct mt792x_phy *phy,
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
 
-	mt7925_mcu_bss_ifs_tlv(skb, vif);
+	mt7925_mcu_bss_ifs_tlv(skb, link_conf->vif);
 
 	return mt76_mcu_skb_send_msg(&dev->mt76, skb,
 				     MCU_UNI_CMD(BSS_INFO_UPDATE), true);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
index c692996e057c..d3584743c266 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
@@ -582,7 +582,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 			    struct ieee80211_sta *sta,
 			    int enable);
 int mt7925_mcu_set_timing(struct mt792x_phy *phy,
-			  struct ieee80211_vif *vif);
+			  struct ieee80211_bss_conf *link_conf);
 int mt7925_mcu_set_deep_sleep(struct mt792x_dev *dev, bool enable);
 int mt7925_mcu_set_channel_domain(struct mt76_phy *phy);
 int mt7925_mcu_set_radio_en(struct mt792x_phy *phy, bool enable);
-- 
cgit v1.2.3-58-ga151


From d62f77e3477819ddda50e62974ee160eb4c7734f Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:03 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_bss_ifs_tlv for per-link BSS

Extend mt7925_mcu_bss_ifs_tlv with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-10-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index b82cc1bd09b6..7cde18e273d8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2282,9 +2282,10 @@ mt7925_mcu_bss_color_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
 }
 
 static void
-mt7925_mcu_bss_ifs_tlv(struct sk_buff *skb, struct ieee80211_vif *vif)
+mt7925_mcu_bss_ifs_tlv(struct sk_buff *skb,
+		       struct ieee80211_bss_conf *link_conf)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_vif *mvif = (struct mt792x_vif *)link_conf->vif->drv_priv;
 	struct mt792x_phy *phy = mvif->phy;
 	struct bss_ifs_time_tlv *ifs_time;
 	struct tlv *tlv;
@@ -2307,7 +2308,7 @@ int mt7925_mcu_set_timing(struct mt792x_phy *phy,
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
 
-	mt7925_mcu_bss_ifs_tlv(skb, link_conf->vif);
+	mt7925_mcu_bss_ifs_tlv(skb, link_conf);
 
 	return mt76_mcu_skb_send_msg(&dev->mt76, skb,
 				     MCU_UNI_CMD(BSS_INFO_UPDATE), true);
@@ -2337,7 +2338,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf->vif, sta);
 	mt7925_mcu_bss_qos_tlv(skb, link_conf->vif);
 	mt7925_mcu_bss_mld_tlv(skb, link_conf->vif, sta);
-	mt7925_mcu_bss_ifs_tlv(skb, link_conf->vif);
+	mt7925_mcu_bss_ifs_tlv(skb, link_conf);
 
 	if (link_conf->he_support) {
 		mt7925_mcu_bss_he_tlv(skb, link_conf->vif, phy);
-- 
cgit v1.2.3-58-ga151


From b698264e1ec20b22dad6f58323f3d53d7680cc2d Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:04 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_bss_color_tlv for per-link BSS

Extend mt7925_mcu_bss_color_tlv with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-11-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 7cde18e273d8..10e38637eb83 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2266,7 +2266,7 @@ mt7925_mcu_bss_he_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
 }
 
 static void
-mt7925_mcu_bss_color_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
+mt7925_mcu_bss_color_tlv(struct sk_buff *skb, struct ieee80211_bss_conf *link_conf,
 			 bool enable)
 {
 	struct bss_info_uni_bss_color *color;
@@ -2276,9 +2276,9 @@ mt7925_mcu_bss_color_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
 	color = (struct bss_info_uni_bss_color *)tlv;
 
 	color->enable = enable ?
-		vif->bss_conf.he_bss_color.enabled : 0;
+		link_conf->he_bss_color.enabled : 0;
 	color->bss_color = enable ?
-		vif->bss_conf.he_bss_color.color : 0;
+		link_conf->he_bss_color.color : 0;
 }
 
 static void
@@ -2342,7 +2342,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 
 	if (link_conf->he_support) {
 		mt7925_mcu_bss_he_tlv(skb, link_conf->vif, phy);
-		mt7925_mcu_bss_color_tlv(skb, link_conf->vif, enable);
+		mt7925_mcu_bss_color_tlv(skb, link_conf, enable);
 	}
 
 	err = mt76_mcu_skb_send_msg(&dev->mt76, skb,
-- 
cgit v1.2.3-58-ga151


From a5c372f77aa7969db9cde1b71c3193bdb126259b Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:05 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_bss_he_tlv for per-link BSS

Extend mt7925_mcu_bss_he_tlv with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-12-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 10e38637eb83..533214403008 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2238,7 +2238,7 @@ mt7925_mcu_bss_qos_tlv(struct sk_buff *skb, struct ieee80211_vif *vif)
 }
 
 static void
-mt7925_mcu_bss_he_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
+mt7925_mcu_bss_he_tlv(struct sk_buff *skb, struct ieee80211_bss_conf *link_conf,
 		      struct mt792x_phy *phy)
 {
 #define DEFAULT_HE_PE_DURATION		4
@@ -2247,16 +2247,16 @@ mt7925_mcu_bss_he_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
 	struct bss_info_uni_he *he;
 	struct tlv *tlv;
 
-	cap = mt76_connac_get_he_phy_cap(phy->mt76, vif);
+	cap = mt76_connac_get_he_phy_cap(phy->mt76, link_conf->vif);
 
 	tlv = mt76_connac_mcu_add_tlv(skb, UNI_BSS_INFO_HE_BASIC, sizeof(*he));
 
 	he = (struct bss_info_uni_he *)tlv;
-	he->he_pe_duration = vif->bss_conf.htc_trig_based_pkt_ext;
+	he->he_pe_duration = link_conf->htc_trig_based_pkt_ext;
 	if (!he->he_pe_duration)
 		he->he_pe_duration = DEFAULT_HE_PE_DURATION;
 
-	he->he_rts_thres = cpu_to_le16(vif->bss_conf.frame_time_rts_th);
+	he->he_rts_thres = cpu_to_le16(link_conf->frame_time_rts_th);
 	if (!he->he_rts_thres)
 		he->he_rts_thres = cpu_to_le16(DEFAULT_HE_DURATION_RTS_THRES);
 
@@ -2341,7 +2341,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 	mt7925_mcu_bss_ifs_tlv(skb, link_conf);
 
 	if (link_conf->he_support) {
-		mt7925_mcu_bss_he_tlv(skb, link_conf->vif, phy);
+		mt7925_mcu_bss_he_tlv(skb, link_conf, phy);
 		mt7925_mcu_bss_color_tlv(skb, link_conf, enable);
 	}
 
-- 
cgit v1.2.3-58-ga151


From eff53d6ee11bb45d7dbcd348fbf4a1734f97d753 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:06 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_bss_qos_tlv for per-link BSS

Extend mt7925_mcu_bss_qos_tlv with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-13-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 533214403008..d6a590061e36 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2227,14 +2227,14 @@ mt7925_mcu_bss_mld_tlv(struct sk_buff *skb,
 }
 
 static void
-mt7925_mcu_bss_qos_tlv(struct sk_buff *skb, struct ieee80211_vif *vif)
+mt7925_mcu_bss_qos_tlv(struct sk_buff *skb, struct ieee80211_bss_conf *link_conf)
 {
 	struct mt76_connac_bss_qos_tlv *qos;
 	struct tlv *tlv;
 
 	tlv = mt76_connac_mcu_add_tlv(skb, UNI_BSS_INFO_QBSS, sizeof(*qos));
 	qos = (struct mt76_connac_bss_qos_tlv *)tlv;
-	qos->qos = vif->bss_conf.qos;
+	qos->qos = link_conf->qos;
 }
 
 static void
@@ -2336,7 +2336,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 	mt7925_mcu_bss_sec_tlv(skb, link_conf->vif);
 
 	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf->vif, sta);
-	mt7925_mcu_bss_qos_tlv(skb, link_conf->vif);
+	mt7925_mcu_bss_qos_tlv(skb, link_conf);
 	mt7925_mcu_bss_mld_tlv(skb, link_conf->vif, sta);
 	mt7925_mcu_bss_ifs_tlv(skb, link_conf);
 
-- 
cgit v1.2.3-58-ga151


From 220865160cf624c8630de76b5a6013562399eb2e Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:07 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_bss_mld_tlv for per-link BSS

Extend mt7925_mcu_bss_mld_tlv with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-14-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index d6a590061e36..93ce018a0801 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2206,20 +2206,20 @@ mt7925_mcu_bss_bmc_tlv(struct sk_buff *skb, struct mt792x_phy *phy,
 
 static void
 mt7925_mcu_bss_mld_tlv(struct sk_buff *skb,
-		       struct ieee80211_vif *vif,
+		       struct ieee80211_bss_conf *link_conf,
 		       struct ieee80211_sta *sta)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	bool is_mld = ieee80211_vif_is_mld(vif);
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
+	bool is_mld = ieee80211_vif_is_mld(link_conf->vif);
 	struct bss_mld_tlv *mld;
 	struct tlv *tlv;
 
 	tlv = mt76_connac_mcu_add_tlv(skb, UNI_BSS_INFO_MLD, sizeof(*mld));
 	mld = (struct bss_mld_tlv *)tlv;
 
-	mld->link_id = sta ? (is_mld ? vif->bss_conf.link_id : 0) : 0xff;
-	mld->group_mld_id = is_mld ? mvif->bss_conf.mt76.idx : 0xff;
-	mld->own_mld_id = mvif->bss_conf.mt76.idx + 32;
+	mld->link_id = sta ? (is_mld ? link_conf->link_id : 0) : 0xff;
+	mld->group_mld_id = is_mld ? mconf->mt76.idx : 0xff;
+	mld->own_mld_id = mconf->mt76.idx + 32;
 	mld->remap_idx = 0xff;
 
 	if (sta)
@@ -2337,7 +2337,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 
 	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf->vif, sta);
 	mt7925_mcu_bss_qos_tlv(skb, link_conf);
-	mt7925_mcu_bss_mld_tlv(skb, link_conf->vif, sta);
+	mt7925_mcu_bss_mld_tlv(skb, link_conf, sta);
 	mt7925_mcu_bss_ifs_tlv(skb, link_conf);
 
 	if (link_conf->he_support) {
-- 
cgit v1.2.3-58-ga151


From 1d1f12689919588050afe89d5577a009774f4439 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:08 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_bss_bmc_tlv for per-link BSS

Extend mt7925_mcu_bss_bmc_tlv with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-15-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 93ce018a0801..48b157f42db9 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2179,12 +2179,13 @@ mt7925_mcu_bss_sec_tlv(struct sk_buff *skb, struct ieee80211_vif *vif)
 static void
 mt7925_mcu_bss_bmc_tlv(struct sk_buff *skb, struct mt792x_phy *phy,
 		       struct ieee80211_chanctx_conf *ctx,
-			   struct ieee80211_vif *vif,
+		       struct ieee80211_bss_conf *link_conf,
 			   struct ieee80211_sta *sta)
 {
 	struct cfg80211_chan_def *chandef = ctx ? &ctx->def : &phy->mt76->chandef;
-	struct mt76_vif *mvif = (struct mt76_vif *)vif->drv_priv;
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
 	enum nl80211_band band = chandef->chan->band;
+	struct mt76_vif *mvif = &mconf->mt76;
 	struct bss_rate_tlv *bmc;
 	struct tlv *tlv;
 	u8 idx = mvif->mcast_rates_idx ?
@@ -2335,7 +2336,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 				 mvif->sta.deflink.wcid.idx, enable);
 	mt7925_mcu_bss_sec_tlv(skb, link_conf->vif);
 
-	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf->vif, sta);
+	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf, sta);
 	mt7925_mcu_bss_qos_tlv(skb, link_conf);
 	mt7925_mcu_bss_mld_tlv(skb, link_conf, sta);
 	mt7925_mcu_bss_ifs_tlv(skb, link_conf);
-- 
cgit v1.2.3-58-ga151


From 0637256a79ecf22d483876263881329b35ad788c Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:09 -0700
Subject: wifi: mt76: mt7925: remove unused parameters in
 mt7925_mcu_bss_bmc_tlv

remove unused parameters in mt7925_mcu_bss_bmc_tlv

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-16-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 48b157f42db9..797d4c936fec 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2179,8 +2179,7 @@ mt7925_mcu_bss_sec_tlv(struct sk_buff *skb, struct ieee80211_vif *vif)
 static void
 mt7925_mcu_bss_bmc_tlv(struct sk_buff *skb, struct mt792x_phy *phy,
 		       struct ieee80211_chanctx_conf *ctx,
-		       struct ieee80211_bss_conf *link_conf,
-			   struct ieee80211_sta *sta)
+		       struct ieee80211_bss_conf *link_conf)
 {
 	struct cfg80211_chan_def *chandef = ctx ? &ctx->def : &phy->mt76->chandef;
 	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
@@ -2336,7 +2335,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 				 mvif->sta.deflink.wcid.idx, enable);
 	mt7925_mcu_bss_sec_tlv(skb, link_conf->vif);
 
-	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf, sta);
+	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf);
 	mt7925_mcu_bss_qos_tlv(skb, link_conf);
 	mt7925_mcu_bss_mld_tlv(skb, link_conf, sta);
 	mt7925_mcu_bss_ifs_tlv(skb, link_conf);
-- 
cgit v1.2.3-58-ga151


From ff5efcd4f8cb264ba86236785f5d32ba604625ef Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:10 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_bss_sec_tlv for per-link BSS

Extend mt7925_mcu_bss_sec_tlv with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-17-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 797d4c936fec..028c94d8c21f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2132,9 +2132,11 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 }
 
 static void
-mt7925_mcu_bss_sec_tlv(struct sk_buff *skb, struct ieee80211_vif *vif)
+mt7925_mcu_bss_sec_tlv(struct sk_buff *skb,
+		       struct ieee80211_bss_conf *link_conf)
 {
-	struct mt76_vif *mvif = (struct mt76_vif *)vif->drv_priv;
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
+	struct mt76_vif *mvif = &mconf->mt76;
 	struct bss_sec_tlv {
 		__le16 tag;
 		__le16 len;
@@ -2333,7 +2335,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 	/* bss_basic must be first */
 	mt7925_mcu_bss_basic_tlv(skb, link_conf->vif, sta, ctx, phy->mt76,
 				 mvif->sta.deflink.wcid.idx, enable);
-	mt7925_mcu_bss_sec_tlv(skb, link_conf->vif);
+	mt7925_mcu_bss_sec_tlv(skb, link_conf);
 
 	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf);
 	mt7925_mcu_bss_qos_tlv(skb, link_conf);
-- 
cgit v1.2.3-58-ga151


From 21760dcd2ab61333533e9d72c1e585f6bf25bb7b Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:11 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_bss_basic_tlv for per-link BSS

Extend mt7925_mcu_bss_basic_tlv with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-18-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c |  1 +
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c  | 25 ++++++++++++------------
 drivers/net/wireless/mediatek/mt76/mt792x.h      |  1 +
 3 files changed, 15 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 6a1afff92399..457e1fa2fde1 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -338,6 +338,7 @@ mt7925_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 	mvif->phy = phy;
 	mvif->bss_conf.mt76.band_idx = 0;
 	mvif->bss_conf.mt76.wmm_idx = mvif->bss_conf.mt76.idx % MT76_CONNAC_MAX_WMM_SETS;
+	mvif->bss_conf.vif = mvif;
 
 	if (phy->mt76->chandef.chan->band != NL80211_BAND_2GHZ)
 		mvif->bss_conf.mt76.basic_rates_idx = MT792x_BASIC_RATES_TBL + 4;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 028c94d8c21f..18e19ffdbb7e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2062,15 +2062,16 @@ mt7925_get_phy_mode_ext(struct mt76_phy *phy, struct ieee80211_vif *vif,
 
 static void
 mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
-			 struct ieee80211_vif *vif,
+			 struct ieee80211_bss_conf *link_conf,
 			 struct ieee80211_sta *sta,
 			 struct ieee80211_chanctx_conf *ctx,
 			 struct mt76_phy *phy, u16 wlan_idx,
 			 bool enable)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct ieee80211_vif *vif = link_conf->vif;
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
 	struct mt792x_sta *msta = sta ? (struct mt792x_sta *)sta->drv_priv :
-				  &mvif->sta;
+				  &mconf->vif->sta;
 	struct cfg80211_chan_def *chandef = ctx ? &ctx->def : &phy->chandef;
 	enum nl80211_band band = chandef->chan->band;
 	struct mt76_connac_bss_basic_tlv *basic_req;
@@ -2081,8 +2082,8 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 	tlv = mt76_connac_mcu_add_tlv(skb, UNI_BSS_INFO_BASIC, sizeof(*basic_req));
 	basic_req = (struct mt76_connac_bss_basic_tlv *)tlv;
 
-	idx = mvif->bss_conf.mt76.omac_idx > EXT_BSSID_START ? HW_BSSID_0 :
-						      mvif->bss_conf.mt76.omac_idx;
+	idx = mconf->mt76.omac_idx > EXT_BSSID_START ? HW_BSSID_0 :
+						      mconf->mt76.omac_idx;
 	basic_req->hw_bss_idx = idx;
 
 	basic_req->phymode_ext = mt7925_get_phy_mode_ext(phy, vif, band, sta);
@@ -2092,15 +2093,15 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 	else
 		basic_req->nonht_basic_phy = cpu_to_le16(PHY_TYPE_OFDM_INDEX);
 
-	memcpy(basic_req->bssid, vif->bss_conf.bssid, ETH_ALEN);
+	memcpy(basic_req->bssid, link_conf->bssid, ETH_ALEN);
 	basic_req->phymode = mt76_connac_get_phy_mode(phy, vif, band, sta);
-	basic_req->bcn_interval = cpu_to_le16(vif->bss_conf.beacon_int);
-	basic_req->dtim_period = vif->bss_conf.dtim_period;
+	basic_req->bcn_interval = cpu_to_le16(link_conf->beacon_int);
+	basic_req->dtim_period = link_conf->dtim_period;
 	basic_req->bmc_tx_wlan_idx = cpu_to_le16(wlan_idx);
 	basic_req->sta_idx = cpu_to_le16(msta->deflink.wcid.idx);
-	basic_req->omac_idx = mvif->bss_conf.mt76.omac_idx;
-	basic_req->band_idx = mvif->bss_conf.mt76.band_idx;
-	basic_req->wmm_idx = mvif->bss_conf.mt76.wmm_idx;
+	basic_req->omac_idx = mconf->mt76.omac_idx;
+	basic_req->band_idx = mconf->mt76.band_idx;
+	basic_req->wmm_idx = mconf->mt76.wmm_idx;
 	basic_req->conn_state = !enable;
 
 	switch (vif->type) {
@@ -2333,7 +2334,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 		return PTR_ERR(skb);
 
 	/* bss_basic must be first */
-	mt7925_mcu_bss_basic_tlv(skb, link_conf->vif, sta, ctx, phy->mt76,
+	mt7925_mcu_bss_basic_tlv(skb, link_conf, sta, ctx, phy->mt76,
 				 mvif->sta.deflink.wcid.idx, enable);
 	mt7925_mcu_bss_sec_tlv(skb, link_conf);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index 718c6b185d99..f36e4aeb79b2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -108,6 +108,7 @@ struct mt792x_chanctx {
 
 struct mt792x_bss_conf {
 	struct mt76_vif mt76; /* must be first */
+	struct mt792x_vif *vif;
 	struct ewma_rssi rssi;
 	struct ieee80211_tx_queue_params queue_params[IEEE80211_NUM_ACS];
 };
-- 
cgit v1.2.3-58-ga151


From b6749ee8b466d5f0fe0a8b80991bc44866ce0be5 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:12 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_set_bss_pm for per-link BSS

Extend mt7925_mcu_set_bss_pm with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-19-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c   |  4 ++--
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    | 15 ++++++++-------
 drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h |  3 ++-
 3 files changed, 12 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 457e1fa2fde1..aeed2c777a20 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1263,7 +1263,7 @@ mt7925_start_ap(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	if (err)
 		goto out;
 
-	err = mt7925_mcu_set_bss_pm(dev, vif, true);
+	err = mt7925_mcu_set_bss_pm(dev, link_conf, true);
 	if (err)
 		goto out;
 
@@ -1285,7 +1285,7 @@ mt7925_stop_ap(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 
 	mt792x_mutex_acquire(dev);
 
-	err = mt7925_mcu_set_bss_pm(dev, vif, false);
+	err = mt7925_mcu_set_bss_pm(dev, link_conf, false);
 	if (err)
 		goto out;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 18e19ffdbb7e..0350f0e34690 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1319,10 +1319,11 @@ mt7925_mcu_uni_bss_bcnft(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 }
 
 int
-mt7925_mcu_set_bss_pm(struct mt792x_dev *dev, struct ieee80211_vif *vif,
+mt7925_mcu_set_bss_pm(struct mt792x_dev *dev,
+		      struct ieee80211_bss_conf *link_conf,
 		      bool enable)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
 	struct {
 		struct {
 			u8 bss_idx;
@@ -1339,13 +1340,13 @@ mt7925_mcu_set_bss_pm(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 		} __packed enable;
 	} req = {
 		.hdr = {
-			.bss_idx = mvif->bss_conf.mt76.idx,
+			.bss_idx = mconf->mt76.idx,
 		},
 		.enable = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_BCNFT),
 			.len = cpu_to_le16(sizeof(struct bcnft_tlv)),
-			.dtim_period = vif->bss_conf.dtim_period,
-			.bcn_interval = cpu_to_le16(vif->bss_conf.beacon_int),
+			.dtim_period = link_conf->dtim_period,
+			.bcn_interval = cpu_to_le16(link_conf->beacon_int),
 		},
 	};
 	struct {
@@ -1359,7 +1360,7 @@ mt7925_mcu_set_bss_pm(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 		} __packed disable;
 	} req1 = {
 		.hdr = {
-			.bss_idx = mvif->bss_conf.mt76.idx,
+			.bss_idx = mconf->mt76.idx,
 		},
 		.disable = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_PM_DISABLE),
@@ -1693,7 +1694,7 @@ int mt7925_mcu_set_beacon_filter(struct mt792x_dev *dev,
 					       MT_WF_RFCR_DROP_OTHER_BEACON);
 	}
 
-	err = mt7925_mcu_set_bss_pm(dev, vif, false);
+	err = mt7925_mcu_set_bss_pm(dev, &vif->bss_conf, false);
 	if (err)
 		return err;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
index f689f7a359ea..07ea03195e4e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
@@ -192,7 +192,8 @@ int __mt7925_start(struct mt792x_phy *phy);
 int mt7925_register_device(struct mt792x_dev *dev);
 void mt7925_unregister_device(struct mt792x_dev *dev);
 int mt7925_run_firmware(struct mt792x_dev *dev);
-int mt7925_mcu_set_bss_pm(struct mt792x_dev *dev, struct ieee80211_vif *vif,
+int mt7925_mcu_set_bss_pm(struct mt792x_dev *dev,
+			  struct ieee80211_bss_conf *link_conf,
 			  bool enable);
 int mt7925_mcu_sta_update(struct mt792x_dev *dev, struct ieee80211_sta *sta,
 			  struct ieee80211_vif *vif, bool enable,
-- 
cgit v1.2.3-58-ga151


From 0aa5e96c6df77aa8c0e5b3f77afa0a4fc84ffe4c Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:13 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_[abort, set]_roc for per-link
 BSS

The channel context has to be bound with the per-link BSS, thus,
we extend mt7925_mcu_[abort, set]_roc to work with the per-link BSS
configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-20-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c   | 23 ++++++++++++----------
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    |  8 ++++----
 drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h |  4 ++--
 3 files changed, 19 insertions(+), 16 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index aeed2c777a20..bba17e51a023 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -387,7 +387,7 @@ static void mt7925_roc_iter(void *priv, u8 *mac,
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_phy *phy = priv;
 
-	mt7925_mcu_abort_roc(phy, mvif, phy->roc_token_id);
+	mt7925_mcu_abort_roc(phy, &mvif->bss_conf, phy->roc_token_id);
 }
 
 void mt7925_roc_work(struct work_struct *work)
@@ -408,7 +408,8 @@ void mt7925_roc_work(struct work_struct *work)
 	ieee80211_remain_on_channel_expired(phy->mt76->hw);
 }
 
-static int mt7925_abort_roc(struct mt792x_phy *phy, struct mt792x_vif *vif)
+static int mt7925_abort_roc(struct mt792x_phy *phy,
+			    struct mt792x_bss_conf *mconf)
 {
 	int err = 0;
 
@@ -417,14 +418,14 @@ static int mt7925_abort_roc(struct mt792x_phy *phy, struct mt792x_vif *vif)
 
 	mt792x_mutex_acquire(phy->dev);
 	if (test_and_clear_bit(MT76_STATE_ROC, &phy->mt76->state))
-		err = mt7925_mcu_abort_roc(phy, vif, phy->roc_token_id);
+		err = mt7925_mcu_abort_roc(phy, mconf, phy->roc_token_id);
 	mt792x_mutex_release(phy->dev);
 
 	return err;
 }
 
 static int mt7925_set_roc(struct mt792x_phy *phy,
-			  struct mt792x_vif *vif,
+			  struct mt792x_bss_conf *mconf,
 			  struct ieee80211_channel *chan,
 			  int duration,
 			  enum mt7925_roc_req type)
@@ -436,7 +437,7 @@ static int mt7925_set_roc(struct mt792x_phy *phy,
 
 	phy->roc_grant = false;
 
-	err = mt7925_mcu_set_roc(phy, vif, chan, duration, type,
+	err = mt7925_mcu_set_roc(phy, mconf, chan, duration, type,
 				 ++phy->roc_token_id);
 	if (err < 0) {
 		clear_bit(MT76_STATE_ROC, &phy->mt76->state);
@@ -444,7 +445,7 @@ static int mt7925_set_roc(struct mt792x_phy *phy,
 	}
 
 	if (!wait_event_timeout(phy->roc_wait, phy->roc_grant, 4 * HZ)) {
-		mt7925_mcu_abort_roc(phy, vif, phy->roc_token_id);
+		mt7925_mcu_abort_roc(phy, mconf, phy->roc_token_id);
 		clear_bit(MT76_STATE_ROC, &phy->mt76->state);
 		err = -ETIMEDOUT;
 	}
@@ -464,7 +465,8 @@ static int mt7925_remain_on_channel(struct ieee80211_hw *hw,
 	int err;
 
 	mt792x_mutex_acquire(phy->dev);
-	err = mt7925_set_roc(phy, mvif, chan, duration, MT7925_ROC_REQ_ROC);
+	err = mt7925_set_roc(phy, &mvif->bss_conf,
+			     chan, duration, MT7925_ROC_REQ_ROC);
 	mt792x_mutex_release(phy->dev);
 
 	return err;
@@ -476,7 +478,7 @@ static int mt7925_cancel_remain_on_channel(struct ieee80211_hw *hw,
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_phy *phy = mt792x_hw_phy(hw);
 
-	return mt7925_abort_roc(phy, mvif);
+	return mt7925_abort_roc(phy, &mvif->bss_conf);
 }
 
 static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
@@ -1345,7 +1347,8 @@ static void mt7925_mgd_prepare_tx(struct ieee80211_hw *hw,
 		       jiffies_to_msecs(HZ);
 
 	mt792x_mutex_acquire(dev);
-	mt7925_set_roc(mvif->phy, mvif, mvif->bss_conf.mt76.ctx->def.chan, duration,
+	mt7925_set_roc(mvif->phy, &mvif->bss_conf,
+		       mvif->bss_conf.mt76.ctx->def.chan, duration,
 		       MT7925_ROC_REQ_JOIN);
 	mt792x_mutex_release(dev);
 }
@@ -1356,7 +1359,7 @@ static void mt7925_mgd_complete_tx(struct ieee80211_hw *hw,
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 
-	mt7925_abort_roc(mvif->phy, mvif);
+	mt7925_abort_roc(mvif->phy, &mvif->bss_conf);
 }
 
 static void mt7925_vif_cfg_changed(struct ieee80211_hw *hw,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 0350f0e34690..a40ad1578212 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1050,7 +1050,7 @@ int mt7925_mcu_add_key(struct mt76_dev *dev, struct ieee80211_vif *vif,
 	return mt76_mcu_skb_send_msg(dev, skb, mcu_cmd, true);
 }
 
-int mt7925_mcu_set_roc(struct mt792x_phy *phy, struct mt792x_vif *vif,
+int mt7925_mcu_set_roc(struct mt792x_phy *phy, struct mt792x_bss_conf *mconf,
 		       struct ieee80211_channel *chan, int duration,
 		       enum mt7925_roc_req type, u8 token_id)
 {
@@ -1086,7 +1086,7 @@ int mt7925_mcu_set_roc(struct mt792x_phy *phy, struct mt792x_vif *vif,
 			.tokenid = token_id,
 			.reqtype = type,
 			.maxinterval = cpu_to_le32(duration),
-			.bss_idx = vif->bss_conf.mt76.idx,
+			.bss_idx = mconf->mt76.idx,
 			.control_channel = chan->hw_value,
 			.bw = CMD_CBW_20MHZ,
 			.bw_from_ap = CMD_CBW_20MHZ,
@@ -1117,7 +1117,7 @@ int mt7925_mcu_set_roc(struct mt792x_phy *phy, struct mt792x_vif *vif,
 				 &req, sizeof(req), false);
 }
 
-int mt7925_mcu_abort_roc(struct mt792x_phy *phy, struct mt792x_vif *vif,
+int mt7925_mcu_abort_roc(struct mt792x_phy *phy, struct mt792x_bss_conf *mconf,
 			 u8 token_id)
 {
 	struct mt792x_dev *dev = phy->dev;
@@ -1138,7 +1138,7 @@ int mt7925_mcu_abort_roc(struct mt792x_phy *phy, struct mt792x_vif *vif,
 			.tag = cpu_to_le16(UNI_ROC_ABORT),
 			.len = cpu_to_le16(sizeof(struct roc_abort_tlv)),
 			.tokenid = token_id,
-			.bss_idx = vif->bss_conf.mt76.idx,
+			.bss_idx = mconf->mt76.idx,
 			.dbdcband = 0xff, /* auto*/
 		},
 	};
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
index 07ea03195e4e..8ab2bb01ba73 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
@@ -292,10 +292,10 @@ int mt7925_set_tx_sar_pwr(struct ieee80211_hw *hw,
 int mt7925_mcu_regval(struct mt792x_dev *dev, u32 regidx, u32 *val, bool set);
 int mt7925_mcu_set_clc(struct mt792x_dev *dev, u8 *alpha2,
 		       enum environment_cap env_cap);
-int mt7925_mcu_set_roc(struct mt792x_phy *phy, struct mt792x_vif *vif,
+int mt7925_mcu_set_roc(struct mt792x_phy *phy, struct mt792x_bss_conf *mconf,
 		       struct ieee80211_channel *chan, int duration,
 		       enum mt7925_roc_req type, u8 token_id);
-int mt7925_mcu_abort_roc(struct mt792x_phy *phy, struct mt792x_vif *vif,
+int mt7925_mcu_abort_roc(struct mt792x_phy *phy, struct mt792x_bss_conf *mconf,
 			 u8 token_id);
 int mt7925_mcu_fill_message(struct mt76_dev *mdev, struct sk_buff *skb,
 			    int cmd, int *wait_seq);
-- 
cgit v1.2.3-58-ga151


From d1c5f3719faa782efe7b0dd26b8f263345e1d314 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:14 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_uni_bss_bcnft for per-link BSS

Extend mt7925_mcu_uni_bss_bcnft with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-21-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index a40ad1578212..8950df09a025 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1281,10 +1281,10 @@ int mt7925_mcu_uni_bss_ps(struct mt792x_dev *dev, struct ieee80211_vif *vif)
 }
 
 static int
-mt7925_mcu_uni_bss_bcnft(struct mt792x_dev *dev, struct ieee80211_vif *vif,
-			 bool enable)
+mt7925_mcu_uni_bss_bcnft(struct mt792x_dev *dev,
+			 struct ieee80211_bss_conf *link_conf, bool enable)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
 	struct {
 		struct {
 			u8 bss_idx;
@@ -1301,17 +1301,17 @@ mt7925_mcu_uni_bss_bcnft(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 		} __packed bcnft;
 	} __packed bcnft_req = {
 		.hdr = {
-			.bss_idx = mvif->bss_conf.mt76.idx,
+			.bss_idx = mconf->mt76.idx,
 		},
 		.bcnft = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_BCNFT),
 			.len = cpu_to_le16(sizeof(struct bcnft_tlv)),
-			.bcn_interval = cpu_to_le16(vif->bss_conf.beacon_int),
-			.dtim_period = vif->bss_conf.dtim_period,
+			.bcn_interval = cpu_to_le16(link_conf->beacon_int),
+			.dtim_period = link_conf->dtim_period,
 		},
 	};
 
-	if (vif->type != NL80211_IFTYPE_STATION)
+	if (link_conf->vif->type != NL80211_IFTYPE_STATION)
 		return 0;
 
 	return mt76_mcu_send_msg(&dev->mt76, MCU_UNI_CMD(BSS_INFO_UPDATE),
@@ -1685,7 +1685,7 @@ int mt7925_mcu_set_beacon_filter(struct mt792x_dev *dev,
 	int err = 0;
 
 	if (enable) {
-		err = mt7925_mcu_uni_bss_bcnft(dev, vif, true);
+		err = mt7925_mcu_uni_bss_bcnft(dev, &vif->bss_conf, true);
 		if (err)
 			return err;
 
-- 
cgit v1.2.3-58-ga151


From 3a0c8d2908297b63f0e5a3364721ffd32e5a0c0c Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:15 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_uni_bss_ps for per-link BSS

Extend mt7925_mcu_uni_bss_ps with per-link BSS configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-22-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c   |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    | 11 ++++++-----
 drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h |  3 ++-
 3 files changed, 9 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index bba17e51a023..cd20b85109b9 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1383,7 +1383,7 @@ static void mt7925_vif_cfg_changed(struct ieee80211_hw *hw,
 	}
 
 	if (changed & BSS_CHANGED_PS)
-		mt7925_mcu_uni_bss_ps(dev, vif);
+		mt7925_mcu_uni_bss_ps(dev, &vif->bss_conf);
 
 	mt792x_mutex_release(dev);
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 8950df09a025..7624ec23e064 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1243,9 +1243,10 @@ int mt7925_mcu_set_eeprom(struct mt792x_dev *dev)
 }
 EXPORT_SYMBOL_GPL(mt7925_mcu_set_eeprom);
 
-int mt7925_mcu_uni_bss_ps(struct mt792x_dev *dev, struct ieee80211_vif *vif)
+int mt7925_mcu_uni_bss_ps(struct mt792x_dev *dev,
+			  struct ieee80211_bss_conf *link_conf)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
 	struct {
 		struct {
 			u8 bss_idx;
@@ -1264,16 +1265,16 @@ int mt7925_mcu_uni_bss_ps(struct mt792x_dev *dev, struct ieee80211_vif *vif)
 		} __packed ps;
 	} __packed ps_req = {
 		.hdr = {
-			.bss_idx = mvif->bss_conf.mt76.idx,
+			.bss_idx = mconf->mt76.idx,
 		},
 		.ps = {
 			.tag = cpu_to_le16(UNI_BSS_INFO_PS),
 			.len = cpu_to_le16(sizeof(struct ps_tlv)),
-			.ps_state = vif->cfg.ps ? 2 : 0,
+			.ps_state = link_conf->vif->cfg.ps ? 2 : 0,
 		},
 	};
 
-	if (vif->type != NL80211_IFTYPE_STATION)
+	if (link_conf->vif->type != NL80211_IFTYPE_STATION)
 		return -EOPNOTSUPP;
 
 	return mt76_mcu_send_msg(&dev->mt76, MCU_UNI_CMD(BSS_INFO_UPDATE),
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
index 8ab2bb01ba73..ef37951f69d5 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
@@ -242,7 +242,8 @@ int mt7925_mcu_uni_rx_ba(struct mt792x_dev *dev,
 			 bool enable);
 void mt7925_scan_work(struct work_struct *work);
 void mt7925_roc_work(struct work_struct *work);
-int mt7925_mcu_uni_bss_ps(struct mt792x_dev *dev, struct ieee80211_vif *vif);
+int mt7925_mcu_uni_bss_ps(struct mt792x_dev *dev,
+			  struct ieee80211_bss_conf *link_conf);
 void mt7925_coredump_work(struct work_struct *work);
 int mt7925_get_txpwr_info(struct mt792x_dev *dev, u8 band_idx,
 			  struct mt7925_txpwr *txpwr);
-- 
cgit v1.2.3-58-ga151


From ca64503a8f06ec712c372a7436ffb2b371e4d93c Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:16 -0700
Subject: wifi: mt76: mt7925: add mt7925_mcu_bss_rlm_tlv to constitue the RLM
 TLV

add mt7925_mcu_bss_rlm_tlv to maintain consistency with the other TLVs that
constitute the command for creating the per-link BSS.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-23-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 64 ++++++++++++++++++++++---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.h | 16 +++++++
 2 files changed, 74 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 7624ec23e064..96b7a9df17d8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1921,6 +1921,59 @@ mt7925_mcu_uni_add_beacon_offload(struct mt792x_dev *dev,
 				 &req, sizeof(req), true);
 }
 
+static
+void mt7925_mcu_bss_rlm_tlv(struct sk_buff *skb, struct mt76_phy *phy,
+			    struct ieee80211_chanctx_conf *ctx)
+{
+	struct cfg80211_chan_def *chandef = ctx ? &ctx->def : &phy->chandef;
+	int freq1 = chandef->center_freq1, freq2 = chandef->center_freq2;
+	enum nl80211_band band = chandef->chan->band;
+	struct bss_rlm_tlv *req;
+	struct tlv *tlv;
+
+	tlv = mt76_connac_mcu_add_tlv(skb, UNI_BSS_INFO_RLM, sizeof(*req));
+	req = (struct bss_rlm_tlv *)tlv;
+	req->control_channel = chandef->chan->hw_value,
+	req->center_chan = ieee80211_frequency_to_channel(freq1),
+	req->center_chan2 = ieee80211_frequency_to_channel(freq2),
+	req->tx_streams = hweight8(phy->antenna_mask),
+	req->ht_op_info = 4, /* set HT 40M allowed */
+	req->rx_streams = hweight8(phy->antenna_mask),
+	req->band = band;
+
+	switch (chandef->width) {
+	case NL80211_CHAN_WIDTH_40:
+		req->bw = CMD_CBW_40MHZ;
+		break;
+	case NL80211_CHAN_WIDTH_80:
+		req->bw = CMD_CBW_80MHZ;
+		break;
+	case NL80211_CHAN_WIDTH_80P80:
+		req->bw = CMD_CBW_8080MHZ;
+		break;
+	case NL80211_CHAN_WIDTH_160:
+		req->bw = CMD_CBW_160MHZ;
+		break;
+	case NL80211_CHAN_WIDTH_5:
+		req->bw = CMD_CBW_5MHZ;
+		break;
+	case NL80211_CHAN_WIDTH_10:
+		req->bw = CMD_CBW_10MHZ;
+		break;
+	case NL80211_CHAN_WIDTH_20_NOHT:
+	case NL80211_CHAN_WIDTH_20:
+	default:
+		req->bw = CMD_CBW_20MHZ;
+		req->ht_op_info = 0;
+		break;
+	}
+
+	if (req->control_channel < req->center_chan)
+		req->sco = 1; /* SCA */
+	else if (req->control_channel > req->center_chan)
+		req->sco = 3; /* SCB */
+}
+
 int mt7925_mcu_set_chctx(struct mt76_phy *phy, struct mt76_vif *mvif,
 			 struct ieee80211_chanctx_conf *ctx)
 {
@@ -2326,9 +2379,10 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 			    int enable)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)link_conf->vif->drv_priv;
+	struct mt792x_bss_conf *mconf = mt792x_vif_to_link(mvif,
+							   link_conf->link_id);
 	struct mt792x_dev *dev = phy->dev;
 	struct sk_buff *skb;
-	int err;
 
 	skb = __mt7925_mcu_alloc_bss_req(&dev->mt76, &mvif->bss_conf.mt76,
 					 MT7925_BSS_UPDATE_MAX_SIZE);
@@ -2350,12 +2404,10 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 		mt7925_mcu_bss_color_tlv(skb, link_conf, enable);
 	}
 
-	err = mt76_mcu_skb_send_msg(&dev->mt76, skb,
-				    MCU_UNI_CMD(BSS_INFO_UPDATE), true);
-	if (err < 0)
-		return err;
+	mt7925_mcu_bss_rlm_tlv(skb, phy->mt76, mconf->mt76.ctx);
 
-	return mt7925_mcu_set_chctx(phy->mt76, &mvif->bss_conf.mt76, ctx);
+	return mt76_mcu_skb_send_msg(&dev->mt76, skb,
+				     MCU_UNI_CMD(BSS_INFO_UPDATE), true);
 }
 
 int mt7925_mcu_set_dbdc(struct mt76_phy *phy)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
index d3584743c266..c5197577cdc8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
@@ -456,6 +456,21 @@ struct bss_ifs_time_tlv {
 	__le16 eifs_cck_time;
 } __packed;
 
+struct bss_rlm_tlv {
+	__le16 tag;
+	__le16 len;
+	u8 control_channel;
+	u8 center_chan;
+	u8 center_chan2;
+	u8 bw;
+	u8 tx_streams;
+	u8 rx_streams;
+	u8 ht_op_info;
+	u8 sco;
+	u8 band;
+	u8 pad[3];
+} __packed;
+
 #define MT7925_STA_UPDATE_MAX_SIZE	(sizeof(struct sta_req_hdr) +		\
 					 sizeof(struct sta_rec_basic) +		\
 					 sizeof(struct sta_rec_bf) +		\
@@ -484,6 +499,7 @@ struct bss_ifs_time_tlv {
 					 sizeof(struct bss_info_uni_he) +		\
 					 sizeof(struct bss_info_uni_bss_color) +	\
 					 sizeof(struct bss_ifs_time_tlv) +		\
+					 sizeof(struct bss_rlm_tlv) +		\
 					 sizeof(struct tlv))
 
 #define MT_CONNAC3_SKU_POWER_LIMIT      449
-- 
cgit v1.2.3-58-ga151


From 22d66ef6653bb62c7b3eb6974b86435ff55af3a7 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:17 -0700
Subject: wifi: mt76: mt7925: mt7925_mcu_set_chctx rely on
 mt7925_mcu_bss_rlm_tlv

make mt7925_mcu_set_chctx rely on mt7925_mcu_bss_rlm_tlv to get rid of
the redundant code.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-24-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 95 +++++--------------------
 1 file changed, 16 insertions(+), 79 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 96b7a9df17d8..9d56013b2f55 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1974,85 +1974,6 @@ void mt7925_mcu_bss_rlm_tlv(struct sk_buff *skb, struct mt76_phy *phy,
 		req->sco = 3; /* SCB */
 }
 
-int mt7925_mcu_set_chctx(struct mt76_phy *phy, struct mt76_vif *mvif,
-			 struct ieee80211_chanctx_conf *ctx)
-{
-	struct cfg80211_chan_def *chandef = ctx ? &ctx->def : &phy->chandef;
-	int freq1 = chandef->center_freq1, freq2 = chandef->center_freq2;
-	enum nl80211_band band = chandef->chan->band;
-	struct mt76_dev *mdev = phy->dev;
-	struct {
-		struct {
-			u8 bss_idx;
-			u8 pad[3];
-		} __packed hdr;
-		struct rlm_tlv {
-			__le16 tag;
-			__le16 len;
-			u8 control_channel;
-			u8 center_chan;
-			u8 center_chan2;
-			u8 bw;
-			u8 tx_streams;
-			u8 rx_streams;
-			u8 ht_op_info;
-			u8 sco;
-			u8 band;
-			u8 pad[3];
-		} __packed rlm;
-	} __packed rlm_req = {
-		.hdr = {
-			.bss_idx = mvif->idx,
-		},
-		.rlm = {
-			.tag = cpu_to_le16(UNI_BSS_INFO_RLM),
-			.len = cpu_to_le16(sizeof(struct rlm_tlv)),
-			.control_channel = chandef->chan->hw_value,
-			.center_chan = ieee80211_frequency_to_channel(freq1),
-			.center_chan2 = ieee80211_frequency_to_channel(freq2),
-			.tx_streams = hweight8(phy->antenna_mask),
-			.ht_op_info = 4, /* set HT 40M allowed */
-			.rx_streams = hweight8(phy->antenna_mask),
-			.band = band,
-		},
-	};
-
-	switch (chandef->width) {
-	case NL80211_CHAN_WIDTH_40:
-		rlm_req.rlm.bw = CMD_CBW_40MHZ;
-		break;
-	case NL80211_CHAN_WIDTH_80:
-		rlm_req.rlm.bw = CMD_CBW_80MHZ;
-		break;
-	case NL80211_CHAN_WIDTH_80P80:
-		rlm_req.rlm.bw = CMD_CBW_8080MHZ;
-		break;
-	case NL80211_CHAN_WIDTH_160:
-		rlm_req.rlm.bw = CMD_CBW_160MHZ;
-		break;
-	case NL80211_CHAN_WIDTH_5:
-		rlm_req.rlm.bw = CMD_CBW_5MHZ;
-		break;
-	case NL80211_CHAN_WIDTH_10:
-		rlm_req.rlm.bw = CMD_CBW_10MHZ;
-		break;
-	case NL80211_CHAN_WIDTH_20_NOHT:
-	case NL80211_CHAN_WIDTH_20:
-	default:
-		rlm_req.rlm.bw = CMD_CBW_20MHZ;
-		rlm_req.rlm.ht_op_info = 0;
-		break;
-	}
-
-	if (rlm_req.rlm.control_channel < rlm_req.rlm.center_chan)
-		rlm_req.rlm.sco = 1; /* SCA */
-	else if (rlm_req.rlm.control_channel > rlm_req.rlm.center_chan)
-		rlm_req.rlm.sco = 3; /* SCB */
-
-	return mt76_mcu_send_msg(mdev, MCU_UNI_CMD(BSS_INFO_UPDATE), &rlm_req,
-				 sizeof(rlm_req), true);
-}
-
 static struct sk_buff *
 __mt7925_mcu_alloc_bss_req(struct mt76_dev *dev, struct mt76_vif *mvif, int len)
 {
@@ -2070,6 +1991,22 @@ __mt7925_mcu_alloc_bss_req(struct mt76_dev *dev, struct mt76_vif *mvif, int len)
 	return skb;
 }
 
+int mt7925_mcu_set_chctx(struct mt76_phy *phy, struct mt76_vif *mvif,
+			 struct ieee80211_chanctx_conf *ctx)
+{
+	struct sk_buff *skb;
+
+	skb = __mt7925_mcu_alloc_bss_req(phy->dev, mvif,
+					 MT7925_BSS_UPDATE_MAX_SIZE);
+	if (IS_ERR(skb))
+		return PTR_ERR(skb);
+
+	mt7925_mcu_bss_rlm_tlv(skb, phy, ctx);
+
+	return mt76_mcu_skb_send_msg(phy->dev, skb,
+				     MCU_UNI_CMD(BSS_INFO_UPDATE), true);
+}
+
 static u8
 mt7925_get_phy_mode_ext(struct mt76_phy *phy, struct ieee80211_vif *vif,
 			enum nl80211_band band, struct ieee80211_sta *sta)
-- 
cgit v1.2.3-58-ga151


From f7cc8944039c0066f1e7b78a7f8c02bf558d098f Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:18 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_sta_update for per-link STA

Extend mt7925_mcu_sta_update with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-25-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 .../net/wireless/mediatek/mt76/mt76_connac_mcu.h   |  5 +-
 drivers/net/wireless/mediatek/mt76/mt7925/main.c   |  6 +--
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    | 53 +++++++++++++---------
 drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h |  3 +-
 drivers/net/wireless/mediatek/mt76/mt792x.h        | 16 +++++++
 5 files changed, 57 insertions(+), 26 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index 5c7c42c95de1..994136285815 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -1733,7 +1733,10 @@ enum mt76_sta_info_state {
 };
 
 struct mt76_sta_cmd_info {
-	struct ieee80211_sta *sta;
+	union {
+		struct ieee80211_sta *sta;
+		struct ieee80211_link_sta *link_sta;
+	};
 	struct mt76_wcid *wcid;
 
 	struct ieee80211_vif *vif;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index cd20b85109b9..f3a32fe9a754 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -740,7 +740,7 @@ int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
 					link_conf, sta, false);
 
-	ret = mt7925_mcu_sta_update(dev, sta, vif, true,
+	ret = mt7925_mcu_sta_update(dev, &sta->deflink, vif, true,
 				    MT76_STA_INFO_STATE_NONE);
 	if (ret)
 		return ret;
@@ -773,7 +773,7 @@ void mt7925_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 	memset(msta->deflink.airtime_ac, 0, sizeof(msta->deflink.airtime_ac));
 
-	mt7925_mcu_sta_update(dev, sta, vif, true, MT76_STA_INFO_STATE_ASSOC);
+	mt7925_mcu_sta_update(dev, &sta->deflink, vif, true, MT76_STA_INFO_STATE_ASSOC);
 
 	mt792x_mutex_release(dev);
 }
@@ -789,7 +789,7 @@ void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	mt76_connac_free_pending_tx_skbs(&dev->pm, &msta->deflink.wcid);
 	mt76_connac_pm_wake(&dev->mphy, &dev->pm);
 
-	mt7925_mcu_sta_update(dev, sta, vif, false, MT76_STA_INFO_STATE_NONE);
+	mt7925_mcu_sta_update(dev, &sta->deflink, vif, false, MT76_STA_INFO_STATE_NONE);
 	mt7925_mac_wtbl_update(dev, msta->deflink.wcid.idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 9d56013b2f55..9b69b4b178de 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1628,39 +1628,46 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
 
-	if (info->sta || !info->offload_fw)
-		mt76_connac_mcu_sta_basic_tlv(dev, skb, info->vif, info->sta,
+	if (info->link_sta || !info->offload_fw)
+		mt76_connac_mcu_sta_basic_tlv(dev, skb, info->vif,
+					      info->link_sta->sta,
 					      info->enable, info->newly);
-	if (info->sta && info->enable) {
-		mt7925_mcu_sta_phy_tlv(skb, info->vif, info->sta);
-		mt7925_mcu_sta_ht_tlv(skb, info->sta);
-		mt7925_mcu_sta_vht_tlv(skb, info->sta);
-		mt76_connac_mcu_sta_uapsd(skb, info->vif, info->sta);
-		mt7925_mcu_sta_amsdu_tlv(skb, info->vif, info->sta);
-		mt7925_mcu_sta_he_tlv(skb, info->sta);
-		mt7925_mcu_sta_he_6g_tlv(skb, info->sta);
-		mt7925_mcu_sta_eht_tlv(skb, info->sta);
-		mt7925_mcu_sta_rate_ctrl_tlv(skb, info->vif, info->sta);
-		mt7925_mcu_sta_state_v2_tlv(phy, skb, info->sta,
+	if (info->link_sta && info->enable) {
+		mt7925_mcu_sta_phy_tlv(skb, info->vif, info->link_sta->sta);
+		mt7925_mcu_sta_ht_tlv(skb, info->link_sta->sta);
+		mt7925_mcu_sta_vht_tlv(skb, info->link_sta->sta);
+		mt76_connac_mcu_sta_uapsd(skb, info->vif, info->link_sta->sta);
+		mt7925_mcu_sta_amsdu_tlv(skb, info->vif, info->link_sta->sta);
+		mt7925_mcu_sta_he_tlv(skb, info->link_sta->sta);
+		mt7925_mcu_sta_he_6g_tlv(skb, info->link_sta->sta);
+		mt7925_mcu_sta_eht_tlv(skb, info->link_sta->sta);
+		mt7925_mcu_sta_rate_ctrl_tlv(skb, info->vif,
+					     info->link_sta->sta);
+		mt7925_mcu_sta_state_v2_tlv(phy, skb, info->link_sta->sta,
 					    info->vif, info->rcpi,
 					    info->state);
-		mt7925_mcu_sta_mld_tlv(skb, info->vif, info->sta);
+		mt7925_mcu_sta_mld_tlv(skb, info->vif, info->link_sta->sta);
 	}
 
-	if (info->enable)
-		mt7925_mcu_sta_hdr_trans_tlv(skb, info->vif, info->sta);
+	if (info->enable) {
+		struct ieee80211_sta *sta = info->link_sta ?
+			info->link_sta->sta : NULL;
+
+		mt7925_mcu_sta_hdr_trans_tlv(skb, info->vif, sta);
+	}
 
 	return mt76_mcu_skb_send_msg(dev, skb, info->cmd, true);
 }
 
-int mt7925_mcu_sta_update(struct mt792x_dev *dev, struct ieee80211_sta *sta,
+int mt7925_mcu_sta_update(struct mt792x_dev *dev,
+			  struct ieee80211_link_sta *link_sta,
 			  struct ieee80211_vif *vif, bool enable,
 			  enum mt76_sta_info_state state)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	int rssi = -ewma_rssi_read(&mvif->bss_conf.rssi);
 	struct mt76_sta_cmd_info info = {
-		.sta = sta,
+		.link_sta = link_sta,
 		.vif = vif,
 		.enable = enable,
 		.cmd = MCU_UNI_CMD(STA_REC_UPDATE),
@@ -1669,10 +1676,14 @@ int mt7925_mcu_sta_update(struct mt792x_dev *dev, struct ieee80211_sta *sta,
 		.rcpi = to_rcpi(rssi),
 	};
 	struct mt792x_sta *msta;
+	struct mt792x_link_sta *mlink;
 
-	msta = sta ? (struct mt792x_sta *)sta->drv_priv : NULL;
-	info.wcid = msta ? &msta->deflink.wcid : &mvif->sta.deflink.wcid;
-	info.newly = msta ? state != MT76_STA_INFO_STATE_ASSOC : true;
+	if (link_sta) {
+		msta = (struct mt792x_sta *)link_sta->sta->drv_priv;
+		mlink = mt792x_sta_to_link(msta, link_sta->link_id);
+	}
+	info.wcid = link_sta ? &mlink->wcid : &mvif->sta.deflink.wcid;
+	info.newly = link_sta ? state != MT76_STA_INFO_STATE_ASSOC : true;
 
 	return mt7925_mcu_sta_cmd(&dev->mphy, &info);
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
index ef37951f69d5..8ec5a3a20202 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
@@ -195,7 +195,8 @@ int mt7925_run_firmware(struct mt792x_dev *dev);
 int mt7925_mcu_set_bss_pm(struct mt792x_dev *dev,
 			  struct ieee80211_bss_conf *link_conf,
 			  bool enable);
-int mt7925_mcu_sta_update(struct mt792x_dev *dev, struct ieee80211_sta *sta,
+int mt7925_mcu_sta_update(struct mt792x_dev *dev,
+			  struct ieee80211_link_sta *link_sta,
 			  struct ieee80211_vif *vif, bool enable,
 			  enum mt76_sta_info_state state);
 int mt7925_mcu_set_chan_info(struct mt792x_phy *phy, u16 tag);
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index f36e4aeb79b2..69eb8dac0b70 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -96,6 +96,7 @@ struct mt792x_link_sta {
 
 struct mt792x_sta {
 	struct mt792x_link_sta deflink; /* must be first */
+	struct mt792x_link_sta __rcu *link[IEEE80211_MLD_MAX_NUM_LINKS];
 
 	struct mt792x_vif *vif;
 };
@@ -238,6 +239,21 @@ mt792x_vif_to_link(struct mt792x_vif *mvif, u8 link_id)
 		lockdep_is_held(&mvif->phy->dev->mt76.mutex));
 }
 
+static inline struct mt792x_link_sta *
+mt792x_sta_to_link(struct mt792x_sta *msta, u8 link_id)
+{
+	struct ieee80211_vif *vif;
+
+	vif = container_of((void *)msta->vif, struct ieee80211_vif, drv_priv);
+
+	if (!ieee80211_vif_is_mld(vif) ||
+	    link_id >= IEEE80211_LINK_UNSPECIFIED)
+		return &msta->deflink;
+
+	return rcu_dereference_protected(msta->link[link_id],
+		lockdep_is_held(&msta->vif->phy->dev->mt76.mutex));
+}
+
 static inline struct mt792x_bss_conf *
 mt792x_link_conf_to_mconf(struct ieee80211_bss_conf *link_conf)
 {
-- 
cgit v1.2.3-58-ga151


From ec8a86b2f09ca14d8492a625f637c9fbf7264f40 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:19 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_sta_state_v2_tlv for per-link
 STA

Extend mt7925_mcu_sta_state_v2_tlv with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-26-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 9b69b4b178de..45437603763e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1540,7 +1540,7 @@ mt7925_mcu_sta_phy_tlv(struct sk_buff *skb,
 
 static void
 mt7925_mcu_sta_state_v2_tlv(struct mt76_phy *mphy, struct sk_buff *skb,
-			    struct ieee80211_sta *sta,
+			    struct ieee80211_link_sta *link_sta,
 			    struct ieee80211_vif *vif,
 			    u8 rcpi, u8 sta_state)
 {
@@ -1560,9 +1560,9 @@ mt7925_mcu_sta_state_v2_tlv(struct mt76_phy *mphy, struct sk_buff *skb,
 	state = (struct sta_rec_state_v2 *)tlv;
 	state->state = sta_state;
 
-	if (sta->deflink.vht_cap.vht_supported) {
-		state->vht_opmode = sta->deflink.bandwidth;
-		state->vht_opmode |= sta->deflink.rx_nss <<
+	if (link_sta->vht_cap.vht_supported) {
+		state->vht_opmode = link_sta->bandwidth;
+		state->vht_opmode |= link_sta->rx_nss <<
 			IEEE80211_OPMODE_NOTIF_RX_NSS_SHIFT;
 	}
 }
@@ -1643,7 +1643,7 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 		mt7925_mcu_sta_eht_tlv(skb, info->link_sta->sta);
 		mt7925_mcu_sta_rate_ctrl_tlv(skb, info->vif,
 					     info->link_sta->sta);
-		mt7925_mcu_sta_state_v2_tlv(phy, skb, info->link_sta->sta,
+		mt7925_mcu_sta_state_v2_tlv(phy, skb, info->link_sta,
 					    info->vif, info->rcpi,
 					    info->state);
 		mt7925_mcu_sta_mld_tlv(skb, info->vif, info->link_sta->sta);
-- 
cgit v1.2.3-58-ga151


From 22e177d2a72e8e8cbdb62722795e36769964b6eb Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:20 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_sta_rate_ctrl_tlv with per-link
 STA

Extend mt7925_mcu_sta_rate_ctrl_tlv with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-27-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 45437603763e..266010c0e00a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1569,7 +1569,8 @@ mt7925_mcu_sta_state_v2_tlv(struct mt76_phy *mphy, struct sk_buff *skb,
 
 static void
 mt7925_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb,
-			     struct ieee80211_vif *vif, struct ieee80211_sta *sta)
+			     struct ieee80211_vif *vif,
+			     struct ieee80211_link_sta *link_sta)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct cfg80211_chan_def *chandef = &mvif->bss_conf.mt76.ctx->def;
@@ -1581,7 +1582,7 @@ mt7925_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb,
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_RA, sizeof(*ra_info));
 	ra_info = (struct sta_rec_ra_info *)tlv;
 
-	supp_rates = sta->deflink.supp_rates[band];
+	supp_rates = link_sta->supp_rates[band];
 	if (band == NL80211_BAND_2GHZ)
 		supp_rates = FIELD_PREP(RA_LEGACY_OFDM, supp_rates >> 4) |
 			     FIELD_PREP(RA_LEGACY_CCK, supp_rates & 0xf);
@@ -1590,9 +1591,9 @@ mt7925_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb,
 
 	ra_info->legacy = cpu_to_le16(supp_rates);
 
-	if (sta->deflink.ht_cap.ht_supported)
+	if (link_sta->ht_cap.ht_supported)
 		memcpy(ra_info->rx_mcs_bitmask,
-		       sta->deflink.ht_cap.mcs.rx_mask,
+		       link_sta->ht_cap.mcs.rx_mask,
 		       HT_MCS_MASK_NUM);
 }
 
@@ -1642,7 +1643,7 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 		mt7925_mcu_sta_he_6g_tlv(skb, info->link_sta->sta);
 		mt7925_mcu_sta_eht_tlv(skb, info->link_sta->sta);
 		mt7925_mcu_sta_rate_ctrl_tlv(skb, info->vif,
-					     info->link_sta->sta);
+					     info->link_sta);
 		mt7925_mcu_sta_state_v2_tlv(phy, skb, info->link_sta,
 					    info->vif, info->rcpi,
 					    info->state);
-- 
cgit v1.2.3-58-ga151


From b2f597730619202a9d11f65d1b642a771d6c2658 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:21 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_sta_eht_tlv for per-link STA

Extend mt7925_mcu_sta_eht_tlv with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-28-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 266010c0e00a..0c9cebd3ee71 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1404,18 +1404,18 @@ mt7925_mcu_sta_he_6g_tlv(struct sk_buff *skb, struct ieee80211_sta *sta)
 }
 
 static void
-mt7925_mcu_sta_eht_tlv(struct sk_buff *skb, struct ieee80211_sta *sta)
+mt7925_mcu_sta_eht_tlv(struct sk_buff *skb, struct ieee80211_link_sta *link_sta)
 {
 	struct ieee80211_eht_mcs_nss_supp *mcs_map;
 	struct ieee80211_eht_cap_elem_fixed *elem;
 	struct sta_rec_eht *eht;
 	struct tlv *tlv;
 
-	if (!sta->deflink.eht_cap.has_eht)
+	if (!link_sta->eht_cap.has_eht)
 		return;
 
-	mcs_map = &sta->deflink.eht_cap.eht_mcs_nss_supp;
-	elem = &sta->deflink.eht_cap.eht_cap_elem;
+	mcs_map = &link_sta->eht_cap.eht_mcs_nss_supp;
+	elem = &link_sta->eht_cap.eht_cap_elem;
 
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_EHT, sizeof(*eht));
 
@@ -1425,7 +1425,7 @@ mt7925_mcu_sta_eht_tlv(struct sk_buff *skb, struct ieee80211_sta *sta)
 	eht->phy_cap = cpu_to_le64(*(u64 *)elem->phy_cap_info);
 	eht->phy_cap_ext = cpu_to_le64(elem->phy_cap_info[8]);
 
-	if (sta->deflink.bandwidth == IEEE80211_STA_RX_BW_20)
+	if (link_sta->bandwidth == IEEE80211_STA_RX_BW_20)
 		memcpy(eht->mcs_map_bw20, &mcs_map->only_20mhz, sizeof(eht->mcs_map_bw20));
 	memcpy(eht->mcs_map_bw80, &mcs_map->bw._80, sizeof(eht->mcs_map_bw80));
 	memcpy(eht->mcs_map_bw160, &mcs_map->bw._160, sizeof(eht->mcs_map_bw160));
@@ -1641,7 +1641,7 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 		mt7925_mcu_sta_amsdu_tlv(skb, info->vif, info->link_sta->sta);
 		mt7925_mcu_sta_he_tlv(skb, info->link_sta->sta);
 		mt7925_mcu_sta_he_6g_tlv(skb, info->link_sta->sta);
-		mt7925_mcu_sta_eht_tlv(skb, info->link_sta->sta);
+		mt7925_mcu_sta_eht_tlv(skb, info->link_sta);
 		mt7925_mcu_sta_rate_ctrl_tlv(skb, info->vif,
 					     info->link_sta);
 		mt7925_mcu_sta_state_v2_tlv(phy, skb, info->link_sta,
-- 
cgit v1.2.3-58-ga151


From 6805b190d2e13f6ddbdfe3bfd434ae2a1aba1d89 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:22 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_sta_he_6g_tlv for per-link STA

Extend mt7925_mcu_sta_he_6g_tlv with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-29-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 0c9cebd3ee71..e1dc8452468c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1389,18 +1389,19 @@ mt7925_mcu_sta_he_tlv(struct sk_buff *skb, struct ieee80211_sta *sta)
 }
 
 static void
-mt7925_mcu_sta_he_6g_tlv(struct sk_buff *skb, struct ieee80211_sta *sta)
+mt7925_mcu_sta_he_6g_tlv(struct sk_buff *skb,
+			 struct ieee80211_link_sta *link_sta)
 {
 	struct sta_rec_he_6g_capa *he_6g;
 	struct tlv *tlv;
 
-	if (!sta->deflink.he_6ghz_capa.capa)
+	if (!link_sta->he_6ghz_capa.capa)
 		return;
 
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_HE_6G, sizeof(*he_6g));
 
 	he_6g = (struct sta_rec_he_6g_capa *)tlv;
-	he_6g->capa = sta->deflink.he_6ghz_capa.capa;
+	he_6g->capa = link_sta->he_6ghz_capa.capa;
 }
 
 static void
@@ -1640,7 +1641,7 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 		mt76_connac_mcu_sta_uapsd(skb, info->vif, info->link_sta->sta);
 		mt7925_mcu_sta_amsdu_tlv(skb, info->vif, info->link_sta->sta);
 		mt7925_mcu_sta_he_tlv(skb, info->link_sta->sta);
-		mt7925_mcu_sta_he_6g_tlv(skb, info->link_sta->sta);
+		mt7925_mcu_sta_he_6g_tlv(skb, info->link_sta);
 		mt7925_mcu_sta_eht_tlv(skb, info->link_sta);
 		mt7925_mcu_sta_rate_ctrl_tlv(skb, info->vif,
 					     info->link_sta);
-- 
cgit v1.2.3-58-ga151


From e2f10f5bdd687c0ebfdb5dc9f452c05121a45f86 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:23 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_sta_he_tlv for per-link STA

Extend mt7925_mcu_sta_he_tlv with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-30-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index e1dc8452468c..a57620aed63b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1380,12 +1380,12 @@ mt7925_mcu_set_bss_pm(struct mt792x_dev *dev,
 }
 
 static void
-mt7925_mcu_sta_he_tlv(struct sk_buff *skb, struct ieee80211_sta *sta)
+mt7925_mcu_sta_he_tlv(struct sk_buff *skb, struct ieee80211_link_sta *link_sta)
 {
-	if (!sta->deflink.he_cap.has_he)
+	if (!link_sta->he_cap.has_he)
 		return;
 
-	mt76_connac_mcu_sta_he_tlv_v2(skb, sta);
+	mt76_connac_mcu_sta_he_tlv_v2(skb, link_sta->sta);
 }
 
 static void
@@ -1640,7 +1640,7 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 		mt7925_mcu_sta_vht_tlv(skb, info->link_sta->sta);
 		mt76_connac_mcu_sta_uapsd(skb, info->vif, info->link_sta->sta);
 		mt7925_mcu_sta_amsdu_tlv(skb, info->vif, info->link_sta->sta);
-		mt7925_mcu_sta_he_tlv(skb, info->link_sta->sta);
+		mt7925_mcu_sta_he_tlv(skb, info->link_sta);
 		mt7925_mcu_sta_he_6g_tlv(skb, info->link_sta);
 		mt7925_mcu_sta_eht_tlv(skb, info->link_sta);
 		mt7925_mcu_sta_rate_ctrl_tlv(skb, info->vif,
-- 
cgit v1.2.3-58-ga151


From 8ebb1a61b41e3c803c99b2828953c21eca181fcd Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:24 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_sta_amsdu_tlv for per-link STA

Extend mt7925_mcu_sta_amsdu_tlv with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-31-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index a57620aed63b..57b89eca4e21 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1467,9 +1467,11 @@ mt7925_mcu_sta_vht_tlv(struct sk_buff *skb, struct ieee80211_sta *sta)
 
 static void
 mt7925_mcu_sta_amsdu_tlv(struct sk_buff *skb,
-			 struct ieee80211_vif *vif, struct ieee80211_sta *sta)
+			 struct ieee80211_vif *vif,
+			 struct ieee80211_link_sta *link_sta)
 {
-	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
+	struct mt792x_sta *msta = (struct mt792x_sta *)link_sta->sta->drv_priv;
+	struct mt792x_link_sta *mlink;
 	struct sta_rec_amsdu *amsdu;
 	struct tlv *tlv;
 
@@ -1477,16 +1479,18 @@ mt7925_mcu_sta_amsdu_tlv(struct sk_buff *skb,
 	    vif->type != NL80211_IFTYPE_AP)
 		return;
 
-	if (!sta->deflink.agg.max_amsdu_len)
+	if (!link_sta->agg.max_amsdu_len)
 		return;
 
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_HW_AMSDU, sizeof(*amsdu));
 	amsdu = (struct sta_rec_amsdu *)tlv;
 	amsdu->max_amsdu_num = 8;
 	amsdu->amsdu_en = true;
-	msta->deflink.wcid.amsdu = true;
 
-	switch (sta->deflink.agg.max_amsdu_len) {
+	mlink = mt792x_sta_to_link(msta, link_sta->link_id);
+	mlink->wcid.amsdu = true;
+
+	switch (link_sta->agg.max_amsdu_len) {
 	case IEEE80211_MAX_MPDU_LEN_VHT_11454:
 		amsdu->max_mpdu_size =
 			IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454;
@@ -1639,7 +1643,7 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 		mt7925_mcu_sta_ht_tlv(skb, info->link_sta->sta);
 		mt7925_mcu_sta_vht_tlv(skb, info->link_sta->sta);
 		mt76_connac_mcu_sta_uapsd(skb, info->vif, info->link_sta->sta);
-		mt7925_mcu_sta_amsdu_tlv(skb, info->vif, info->link_sta->sta);
+		mt7925_mcu_sta_amsdu_tlv(skb, info->vif, info->link_sta);
 		mt7925_mcu_sta_he_tlv(skb, info->link_sta);
 		mt7925_mcu_sta_he_6g_tlv(skb, info->link_sta);
 		mt7925_mcu_sta_eht_tlv(skb, info->link_sta);
-- 
cgit v1.2.3-58-ga151


From f75b6ee03aecfe5b74885a1042a5f80189b47144 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:25 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_sta_vht_tlv for per-link STA

Extend mt7925_mcu_sta_vht_tlv with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-32-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 57b89eca4e21..1b66dd26bd1e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1448,21 +1448,21 @@ mt7925_mcu_sta_ht_tlv(struct sk_buff *skb, struct ieee80211_sta *sta)
 }
 
 static void
-mt7925_mcu_sta_vht_tlv(struct sk_buff *skb, struct ieee80211_sta *sta)
+mt7925_mcu_sta_vht_tlv(struct sk_buff *skb, struct ieee80211_link_sta *link_sta)
 {
 	struct sta_rec_vht *vht;
 	struct tlv *tlv;
 
 	/* For 6G band, this tlv is necessary to let hw work normally */
-	if (!sta->deflink.he_6ghz_capa.capa && !sta->deflink.vht_cap.vht_supported)
+	if (!link_sta->he_6ghz_capa.capa && !link_sta->vht_cap.vht_supported)
 		return;
 
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_VHT, sizeof(*vht));
 
 	vht = (struct sta_rec_vht *)tlv;
-	vht->vht_cap = cpu_to_le32(sta->deflink.vht_cap.cap);
-	vht->vht_rx_mcs_map = sta->deflink.vht_cap.vht_mcs.rx_mcs_map;
-	vht->vht_tx_mcs_map = sta->deflink.vht_cap.vht_mcs.tx_mcs_map;
+	vht->vht_cap = cpu_to_le32(link_sta->vht_cap.cap);
+	vht->vht_rx_mcs_map = link_sta->vht_cap.vht_mcs.rx_mcs_map;
+	vht->vht_tx_mcs_map = link_sta->vht_cap.vht_mcs.tx_mcs_map;
 }
 
 static void
@@ -1641,7 +1641,7 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 	if (info->link_sta && info->enable) {
 		mt7925_mcu_sta_phy_tlv(skb, info->vif, info->link_sta->sta);
 		mt7925_mcu_sta_ht_tlv(skb, info->link_sta->sta);
-		mt7925_mcu_sta_vht_tlv(skb, info->link_sta->sta);
+		mt7925_mcu_sta_vht_tlv(skb, info->link_sta);
 		mt76_connac_mcu_sta_uapsd(skb, info->vif, info->link_sta->sta);
 		mt7925_mcu_sta_amsdu_tlv(skb, info->vif, info->link_sta);
 		mt7925_mcu_sta_he_tlv(skb, info->link_sta);
-- 
cgit v1.2.3-58-ga151


From 872b5a03369c97f8e8c51d018618c25c04442d77 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:26 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_sta_ht_tlv for per-link STA

Extend mt7925_mcu_sta_ht_tlv with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-33-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 1b66dd26bd1e..66d45d676739 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1433,18 +1433,18 @@ mt7925_mcu_sta_eht_tlv(struct sk_buff *skb, struct ieee80211_link_sta *link_sta)
 }
 
 static void
-mt7925_mcu_sta_ht_tlv(struct sk_buff *skb, struct ieee80211_sta *sta)
+mt7925_mcu_sta_ht_tlv(struct sk_buff *skb, struct ieee80211_link_sta *link_sta)
 {
 	struct sta_rec_ht *ht;
 	struct tlv *tlv;
 
-	if (!sta->deflink.ht_cap.ht_supported)
+	if (!link_sta->ht_cap.ht_supported)
 		return;
 
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_HT, sizeof(*ht));
 
 	ht = (struct sta_rec_ht *)tlv;
-	ht->ht_cap = cpu_to_le16(sta->deflink.ht_cap.cap);
+	ht->ht_cap = cpu_to_le16(link_sta->ht_cap.cap);
 }
 
 static void
@@ -1640,7 +1640,7 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 					      info->enable, info->newly);
 	if (info->link_sta && info->enable) {
 		mt7925_mcu_sta_phy_tlv(skb, info->vif, info->link_sta->sta);
-		mt7925_mcu_sta_ht_tlv(skb, info->link_sta->sta);
+		mt7925_mcu_sta_ht_tlv(skb, info->link_sta);
 		mt7925_mcu_sta_vht_tlv(skb, info->link_sta);
 		mt76_connac_mcu_sta_uapsd(skb, info->vif, info->link_sta->sta);
 		mt7925_mcu_sta_amsdu_tlv(skb, info->vif, info->link_sta);
-- 
cgit v1.2.3-58-ga151


From e7a9853e6b8ae7f3da4134364e010894cbb9a754 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:27 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_sta_phy_tlv for per-link STA

Extend mt7925_mcu_sta_phy_tlv with per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-34-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 35 ++++++++++++++++---------
 1 file changed, 22 insertions(+), 13 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 66d45d676739..a0fd02dd9551 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1507,34 +1507,43 @@ mt7925_mcu_sta_amsdu_tlv(struct sk_buff *skb,
 
 static void
 mt7925_mcu_sta_phy_tlv(struct sk_buff *skb,
-		       struct ieee80211_vif *vif, struct ieee80211_sta *sta)
+		       struct ieee80211_vif *vif,
+		       struct ieee80211_link_sta *link_sta)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	struct cfg80211_chan_def *chandef = &mvif->bss_conf.mt76.ctx->def;
+	struct ieee80211_bss_conf *link_conf;
+	struct cfg80211_chan_def *chandef;
+	struct mt792x_bss_conf *mconf;
 	struct sta_rec_phy *phy;
 	struct tlv *tlv;
 	u8 af = 0, mm = 0;
 
+	link_conf = mt792x_vif_to_bss_conf(vif, link_sta->link_id);
+	mconf = mt792x_vif_to_link(mvif, link_sta->link_id);
+	chandef = &mconf->mt76.ctx->def;
+
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_PHY, sizeof(*phy));
 	phy = (struct sta_rec_phy *)tlv;
-	phy->phy_type = mt76_connac_get_phy_mode_v2(mvif->phy->mt76, vif, chandef->chan->band, sta);
-	phy->basic_rate = cpu_to_le16((u16)vif->bss_conf.basic_rates);
-	if (sta->deflink.ht_cap.ht_supported) {
-		af = sta->deflink.ht_cap.ampdu_factor;
-		mm = sta->deflink.ht_cap.ampdu_density;
+	phy->phy_type = mt76_connac_get_phy_mode_v2(mvif->phy->mt76, vif,
+						    chandef->chan->band,
+						    link_sta->sta);
+	phy->basic_rate = cpu_to_le16((u16)link_conf->basic_rates);
+	if (link_sta->ht_cap.ht_supported) {
+		af = link_sta->ht_cap.ampdu_factor;
+		mm = link_sta->ht_cap.ampdu_density;
 	}
 
-	if (sta->deflink.vht_cap.vht_supported) {
+	if (link_sta->vht_cap.vht_supported) {
 		u8 vht_af = FIELD_GET(IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK,
-				      sta->deflink.vht_cap.cap);
+				      link_sta->vht_cap.cap);
 
 		af = max_t(u8, af, vht_af);
 	}
 
-	if (sta->deflink.he_6ghz_capa.capa) {
-		af = le16_get_bits(sta->deflink.he_6ghz_capa.capa,
+	if (link_sta->he_6ghz_capa.capa) {
+		af = le16_get_bits(link_sta->he_6ghz_capa.capa,
 				   IEEE80211_HE_6GHZ_CAP_MAX_AMPDU_LEN_EXP);
-		mm = le16_get_bits(sta->deflink.he_6ghz_capa.capa,
+		mm = le16_get_bits(link_sta->he_6ghz_capa.capa,
 				   IEEE80211_HE_6GHZ_CAP_MIN_MPDU_START);
 	}
 
@@ -1639,7 +1648,7 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 					      info->link_sta->sta,
 					      info->enable, info->newly);
 	if (info->link_sta && info->enable) {
-		mt7925_mcu_sta_phy_tlv(skb, info->vif, info->link_sta->sta);
+		mt7925_mcu_sta_phy_tlv(skb, info->vif, info->link_sta);
 		mt7925_mcu_sta_ht_tlv(skb, info->link_sta);
 		mt7925_mcu_sta_vht_tlv(skb, info->link_sta);
 		mt76_connac_mcu_sta_uapsd(skb, info->vif, info->link_sta->sta);
-- 
cgit v1.2.3-58-ga151


From 15ced2a624a534df601d7e74e512177dca3901d5 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:28 -0700
Subject: wifi: mt76: mt7925: extend mt7925_get_phy_mode_ext for per-link STA

Extend mt7925_get_phy_mode_ext with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-35-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index a0fd02dd9551..c6902009fd28 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2035,16 +2035,17 @@ int mt7925_mcu_set_chctx(struct mt76_phy *phy, struct mt76_vif *mvif,
 
 static u8
 mt7925_get_phy_mode_ext(struct mt76_phy *phy, struct ieee80211_vif *vif,
-			enum nl80211_band band, struct ieee80211_sta *sta)
+			enum nl80211_band band,
+			struct ieee80211_link_sta *link_sta)
 {
 	struct ieee80211_he_6ghz_capa *he_6ghz_capa;
 	const struct ieee80211_sta_eht_cap *eht_cap;
 	__le16 capa = 0;
 	u8 mode = 0;
 
-	if (sta) {
-		he_6ghz_capa = &sta->deflink.he_6ghz_capa;
-		eht_cap = &sta->deflink.eht_cap;
+	if (link_sta) {
+		he_6ghz_capa = &link_sta->he_6ghz_capa;
+		eht_cap = &link_sta->eht_cap;
 	} else {
 		struct ieee80211_supported_band *sband;
 
@@ -2090,6 +2091,7 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
 	struct mt792x_sta *msta = sta ? (struct mt792x_sta *)sta->drv_priv :
 				  &mconf->vif->sta;
+	struct ieee80211_link_sta *link_sta = sta ? &sta->deflink : NULL;
 	struct cfg80211_chan_def *chandef = ctx ? &ctx->def : &phy->chandef;
 	enum nl80211_band band = chandef->chan->band;
 	struct mt76_connac_bss_basic_tlv *basic_req;
@@ -2104,7 +2106,8 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 						      mconf->mt76.omac_idx;
 	basic_req->hw_bss_idx = idx;
 
-	basic_req->phymode_ext = mt7925_get_phy_mode_ext(phy, vif, band, sta);
+	basic_req->phymode_ext = mt7925_get_phy_mode_ext(phy, vif, band,
+							 link_sta);
 
 	if (band == NL80211_BAND_2GHZ)
 		basic_req->nonht_basic_phy = cpu_to_le16(PHY_TYPE_ERP_INDEX);
-- 
cgit v1.2.3-58-ga151


From 9d345e15ef58f7ac6e3ad0f1b99169f6b0bcd521 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:29 -0700
Subject: wifi: mt76: mt7925: extend mt7925_get_phy_mode for per-link STA

Extend mt7925_get_phy_mode with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-36-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 11 ++++++-----
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h |  3 ++-
 drivers/net/wireless/mediatek/mt76/mt7915/mcu.c      |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c      |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c      |  2 +-
 5 files changed, 11 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
index a059532f383a..a1a07bbca7b3 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
@@ -1307,7 +1307,8 @@ int mt76_connac_mcu_sta_ba(struct mt76_dev *dev, struct mt76_vif *mvif,
 EXPORT_SYMBOL_GPL(mt76_connac_mcu_sta_ba);
 
 u8 mt76_connac_get_phy_mode(struct mt76_phy *phy, struct ieee80211_vif *vif,
-			    enum nl80211_band band, struct ieee80211_sta *sta)
+			    enum nl80211_band band,
+			    struct ieee80211_link_sta *link_sta)
 {
 	struct mt76_dev *dev = phy->dev;
 	const struct ieee80211_sta_he_cap *he_cap;
@@ -1318,10 +1319,10 @@ u8 mt76_connac_get_phy_mode(struct mt76_phy *phy, struct ieee80211_vif *vif,
 	if (is_connac_v1(dev))
 		return 0x38;
 
-	if (sta) {
-		ht_cap = &sta->deflink.ht_cap;
-		vht_cap = &sta->deflink.vht_cap;
-		he_cap = &sta->deflink.he_cap;
+	if (link_sta) {
+		ht_cap = &link_sta->ht_cap;
+		vht_cap = &link_sta->vht_cap;
+		he_cap = &link_sta->he_cap;
 	} else {
 		struct ieee80211_supported_band *sband;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index 994136285815..b56896cdd4d8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -1995,7 +1995,8 @@ mt76_connac_get_he_phy_cap(struct mt76_phy *phy, struct ieee80211_vif *vif);
 const struct ieee80211_sta_eht_cap *
 mt76_connac_get_eht_phy_cap(struct mt76_phy *phy, struct ieee80211_vif *vif);
 u8 mt76_connac_get_phy_mode(struct mt76_phy *phy, struct ieee80211_vif *vif,
-			    enum nl80211_band band, struct ieee80211_sta *sta);
+			    enum nl80211_band band,
+			    struct ieee80211_link_sta *sta);
 u8 mt76_connac_get_phy_mode_ext(struct mt76_phy *phy, struct ieee80211_vif *vif,
 				enum nl80211_band band);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
index 9599adf104b1..590ee510eb7c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
@@ -1503,7 +1503,7 @@ mt7915_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb, struct mt7915_dev *dev,
 
 	ra->valid = true;
 	ra->auto_rate = true;
-	ra->phy_mode = mt76_connac_get_phy_mode(mphy, vif, band, sta);
+	ra->phy_mode = mt76_connac_get_phy_mode(mphy, vif, band, &sta->deflink);
 	ra->channel = chandef->chan->hw_value;
 	ra->bw = sta->deflink.bandwidth;
 	ra->phy.bw = sta->deflink.bandwidth;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index c6902009fd28..c19c6a0be597 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2115,7 +2115,7 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 		basic_req->nonht_basic_phy = cpu_to_le16(PHY_TYPE_OFDM_INDEX);
 
 	memcpy(basic_req->bssid, link_conf->bssid, ETH_ALEN);
-	basic_req->phymode = mt76_connac_get_phy_mode(phy, vif, band, sta);
+	basic_req->phymode = mt76_connac_get_phy_mode(phy, vif, band, link_sta);
 	basic_req->bcn_interval = cpu_to_le16(link_conf->beacon_int);
 	basic_req->dtim_period = link_conf->dtim_period;
 	basic_req->bmc_tx_wlan_idx = cpu_to_le16(wlan_idx);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index 2c8578677800..15905c0bfa8a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -2002,7 +2002,7 @@ mt7996_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb, struct mt7996_dev *dev,
 
 	ra->valid = true;
 	ra->auto_rate = true;
-	ra->phy_mode = mt76_connac_get_phy_mode(mphy, vif, band, sta);
+	ra->phy_mode = mt76_connac_get_phy_mode(mphy, vif, band, &sta->deflink);
 	ra->channel = chandef->chan->hw_value;
 	ra->bw = (sta->deflink.bandwidth == IEEE80211_STA_RX_BW_320) ?
 		 CMD_CBW_320MHZ : sta->deflink.bandwidth;
-- 
cgit v1.2.3-58-ga151


From 45020d10d276f4e9678572207962fcf6d32f625b Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:30 -0700
Subject: wifi: mt76: mt792x: extend mt76_connac_get_phy_mode_v2 for per-link
 STA

Extend mt76_connac_get_phy_mode_v2 with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-37-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 16 +++++++++-------
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h |  3 ++-
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c      |  2 +-
 3 files changed, 12 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
index a1a07bbca7b3..8e9078342429 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
@@ -793,7 +793,8 @@ EXPORT_SYMBOL_GPL(mt76_connac_mcu_sta_he_tlv_v2);
 
 u8
 mt76_connac_get_phy_mode_v2(struct mt76_phy *mphy, struct ieee80211_vif *vif,
-			    enum nl80211_band band, struct ieee80211_sta *sta)
+			    enum nl80211_band band,
+			    struct ieee80211_link_sta *link_sta)
 {
 	struct ieee80211_sta_ht_cap *ht_cap;
 	struct ieee80211_sta_vht_cap *vht_cap;
@@ -801,11 +802,11 @@ mt76_connac_get_phy_mode_v2(struct mt76_phy *mphy, struct ieee80211_vif *vif,
 	const struct ieee80211_sta_eht_cap *eht_cap;
 	u8 mode = 0;
 
-	if (sta) {
-		ht_cap = &sta->deflink.ht_cap;
-		vht_cap = &sta->deflink.vht_cap;
-		he_cap = &sta->deflink.he_cap;
-		eht_cap = &sta->deflink.eht_cap;
+	if (link_sta) {
+		ht_cap = &link_sta->ht_cap;
+		vht_cap = &link_sta->vht_cap;
+		he_cap = &link_sta->he_cap;
+		eht_cap = &link_sta->eht_cap;
 	} else {
 		struct ieee80211_supported_band *sband;
 
@@ -911,7 +912,8 @@ void mt76_connac_mcu_sta_tlv(struct mt76_phy *mphy, struct sk_buff *skb,
 
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_PHY, sizeof(*phy));
 	phy = (struct sta_rec_phy *)tlv;
-	phy->phy_type = mt76_connac_get_phy_mode_v2(mphy, vif, band, sta);
+	phy->phy_type = mt76_connac_get_phy_mode_v2(mphy, vif, band,
+						    &sta->deflink);
 	phy->basic_rate = cpu_to_le16((u16)vif->bss_conf.basic_rates);
 	phy->rcpi = rcpi;
 	phy->ampdu = FIELD_PREP(IEEE80211_HT_AMPDU_PARM_FACTOR,
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index b56896cdd4d8..d4488b9cca46 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -1901,7 +1901,8 @@ int mt76_connac_mcu_sta_update_hdr_trans(struct mt76_dev *dev,
 					 struct mt76_wcid *wcid, int cmd);
 void mt76_connac_mcu_sta_he_tlv_v2(struct sk_buff *skb, struct ieee80211_sta *sta);
 u8 mt76_connac_get_phy_mode_v2(struct mt76_phy *mphy, struct ieee80211_vif *vif,
-			       enum nl80211_band band, struct ieee80211_sta *sta);
+			       enum nl80211_band band,
+			       struct ieee80211_link_sta *link_sta);
 int mt76_connac_mcu_wtbl_update_hdr_trans(struct mt76_dev *dev,
 					  struct ieee80211_vif *vif,
 					  struct ieee80211_sta *sta);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index c19c6a0be597..80b735cfbdc7 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1526,7 +1526,7 @@ mt7925_mcu_sta_phy_tlv(struct sk_buff *skb,
 	phy = (struct sta_rec_phy *)tlv;
 	phy->phy_type = mt76_connac_get_phy_mode_v2(mvif->phy->mt76, vif,
 						    chandef->chan->band,
-						    link_sta->sta);
+						    link_sta);
 	phy->basic_rate = cpu_to_le16((u16)link_conf->basic_rates);
 	if (link_sta->ht_cap.ht_supported) {
 		af = link_sta->ht_cap.ampdu_factor;
-- 
cgit v1.2.3-58-ga151


From 6e226666efc06dadc0fc92c63c8fe9fa17898b8b Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:31 -0700
Subject: wifi: mt76: mt762x: extend mt76_connac_mcu_sta_basic_tlv for per-link
 STA

Extend mt76_connac_mcu_sta_basic_tlv with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-38-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7615/mcu.c      |  6 ++++--
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 19 +++++++++++--------
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h |  4 ++--
 drivers/net/wireless/mediatek/mt76/mt7915/mcu.c      |  4 +++-
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c      |  2 +-
 drivers/net/wireless/mediatek/mt76/mt7996/mcu.c      |  5 ++++-
 6 files changed, 25 insertions(+), 15 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
index 792ae41f7c69..d50d967828be 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7615/mcu.c
@@ -842,6 +842,7 @@ mt7615_mcu_wtbl_sta_add(struct mt7615_phy *phy, struct ieee80211_vif *vif,
 {
 	struct mt7615_vif *mvif = (struct mt7615_vif *)vif->drv_priv;
 	struct sk_buff *skb, *sskb, *wskb = NULL;
+	struct ieee80211_link_sta *link_sta;
 	struct mt7615_dev *dev = phy->dev;
 	struct wtbl_req_hdr *wtbl_hdr;
 	struct mt7615_sta *msta;
@@ -849,6 +850,7 @@ mt7615_mcu_wtbl_sta_add(struct mt7615_phy *phy, struct ieee80211_vif *vif,
 	int cmd, err;
 
 	msta = sta ? (struct mt7615_sta *)sta->drv_priv : &mvif->sta;
+	link_sta = sta ? &sta->deflink : NULL;
 
 	sskb = mt76_connac_mcu_alloc_sta_req(&dev->mt76, &mvif->mt76,
 					     &msta->wcid);
@@ -861,8 +863,8 @@ mt7615_mcu_wtbl_sta_add(struct mt7615_phy *phy, struct ieee80211_vif *vif,
 		else
 			mvif->sta_added = true;
 	}
-	mt76_connac_mcu_sta_basic_tlv(&dev->mt76, sskb, vif, sta, enable,
-				      new_entry);
+	mt76_connac_mcu_sta_basic_tlv(&dev->mt76, sskb, vif, link_sta,
+				      enable, new_entry);
 	if (enable && sta)
 		mt76_connac_mcu_sta_tlv(phy->mt76, sskb, sta, vif, 0,
 					MT76_STA_INFO_STATE_ASSOC);
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
index 8e9078342429..4e3c8af98fe7 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
@@ -370,7 +370,7 @@ EXPORT_SYMBOL_GPL(mt76_connac_mcu_bss_omac_tlv);
 
 void mt76_connac_mcu_sta_basic_tlv(struct mt76_dev *dev, struct sk_buff *skb,
 				   struct ieee80211_vif *vif,
-				   struct ieee80211_sta *sta,
+				   struct ieee80211_link_sta *link_sta,
 				   bool enable, bool newly)
 {
 	struct sta_rec_basic *basic;
@@ -390,7 +390,7 @@ void mt76_connac_mcu_sta_basic_tlv(struct mt76_dev *dev, struct sk_buff *skb,
 		basic->conn_state = CONN_STATE_DISCONNECT;
 	}
 
-	if (!sta) {
+	if (!link_sta) {
 		basic->conn_type = cpu_to_le32(CONNECTION_INFRA_BC);
 
 		if (vif->type == NL80211_IFTYPE_STATION &&
@@ -411,7 +411,7 @@ void mt76_connac_mcu_sta_basic_tlv(struct mt76_dev *dev, struct sk_buff *skb,
 		else
 			conn_type = CONNECTION_INFRA_STA;
 		basic->conn_type = cpu_to_le32(conn_type);
-		basic->aid = cpu_to_le16(sta->aid);
+		basic->aid = cpu_to_le16(link_sta->sta->aid);
 		break;
 	case NL80211_IFTYPE_STATION:
 		if (vif->p2p && !is_mt7921(dev))
@@ -423,15 +423,15 @@ void mt76_connac_mcu_sta_basic_tlv(struct mt76_dev *dev, struct sk_buff *skb,
 		break;
 	case NL80211_IFTYPE_ADHOC:
 		basic->conn_type = cpu_to_le32(CONNECTION_IBSS_ADHOC);
-		basic->aid = cpu_to_le16(sta->aid);
+		basic->aid = cpu_to_le16(link_sta->sta->aid);
 		break;
 	default:
 		WARN_ON(1);
 		break;
 	}
 
-	memcpy(basic->peer_addr, sta->addr, ETH_ALEN);
-	basic->qos = sta->wme;
+	memcpy(basic->peer_addr, link_sta->addr, ETH_ALEN);
+	basic->qos = link_sta->sta->wme;
 }
 EXPORT_SYMBOL_GPL(mt76_connac_mcu_sta_basic_tlv);
 
@@ -1046,6 +1046,7 @@ int mt76_connac_mcu_sta_cmd(struct mt76_phy *phy,
 			    struct mt76_sta_cmd_info *info)
 {
 	struct mt76_vif *mvif = (struct mt76_vif *)info->vif->drv_priv;
+	struct ieee80211_link_sta *link_sta;
 	struct mt76_dev *dev = phy->dev;
 	struct wtbl_req_hdr *wtbl_hdr;
 	struct tlv *sta_wtbl;
@@ -1055,9 +1056,11 @@ int mt76_connac_mcu_sta_cmd(struct mt76_phy *phy,
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
 
+	link_sta = info->sta ? &info->sta->deflink : NULL;
 	if (info->sta || !info->offload_fw)
-		mt76_connac_mcu_sta_basic_tlv(dev, skb, info->vif, info->sta,
-					      info->enable, info->newly);
+		mt76_connac_mcu_sta_basic_tlv(dev, skb, info->vif,
+					      link_sta, info->enable,
+					      info->newly);
 	if (info->sta && info->enable)
 		mt76_connac_mcu_sta_tlv(phy, skb, info->sta,
 					info->vif, info->rcpi,
diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index d4488b9cca46..e636b033b87e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -1886,8 +1886,8 @@ int mt76_connac_mcu_set_channel_domain(struct mt76_phy *phy);
 int mt76_connac_mcu_set_vif_ps(struct mt76_dev *dev, struct ieee80211_vif *vif);
 void mt76_connac_mcu_sta_basic_tlv(struct mt76_dev *dev, struct sk_buff *skb,
 				   struct ieee80211_vif *vif,
-				   struct ieee80211_sta *sta, bool enable,
-				   bool newly);
+				   struct ieee80211_link_sta *link_sta,
+				   bool enable, bool newly);
 void mt76_connac_mcu_wtbl_generic_tlv(struct mt76_dev *dev, struct sk_buff *skb,
 				      struct ieee80211_vif *vif,
 				      struct ieee80211_sta *sta, void *sta_wtbl,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
index 590ee510eb7c..2185cd24e2e1 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c
@@ -1656,11 +1656,13 @@ int mt7915_mcu_add_sta(struct mt7915_dev *dev, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta, bool enable)
 {
 	struct mt7915_vif *mvif = (struct mt7915_vif *)vif->drv_priv;
+	struct ieee80211_link_sta *link_sta;
 	struct mt7915_sta *msta;
 	struct sk_buff *skb;
 	int ret;
 
 	msta = sta ? (struct mt7915_sta *)sta->drv_priv : &mvif->sta;
+	link_sta = sta ? &sta->deflink : NULL;
 
 	skb = mt76_connac_mcu_alloc_sta_req(&dev->mt76, &mvif->mt76,
 					    &msta->wcid);
@@ -1668,7 +1670,7 @@ int mt7915_mcu_add_sta(struct mt7915_dev *dev, struct ieee80211_vif *vif,
 		return PTR_ERR(skb);
 
 	/* starec basic */
-	mt76_connac_mcu_sta_basic_tlv(&dev->mt76, skb, vif, sta, enable,
+	mt76_connac_mcu_sta_basic_tlv(&dev->mt76, skb, vif, link_sta, enable,
 				      !rcu_access_pointer(dev->mt76.wcid[msta->wcid.idx]));
 	if (!enable)
 		goto out;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 80b735cfbdc7..bf2afaf4c0dc 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1645,7 +1645,7 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 
 	if (info->link_sta || !info->offload_fw)
 		mt76_connac_mcu_sta_basic_tlv(dev, skb, info->vif,
-					      info->link_sta->sta,
+					      info->link_sta,
 					      info->enable, info->newly);
 	if (info->link_sta && info->enable) {
 		mt7925_mcu_sta_phy_tlv(skb, info->vif, info->link_sta);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
index 15905c0bfa8a..2e4fa9f48dfb 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c
@@ -2157,11 +2157,13 @@ int mt7996_mcu_add_sta(struct mt7996_dev *dev, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta, bool enable, bool newly)
 {
 	struct mt7996_vif *mvif = (struct mt7996_vif *)vif->drv_priv;
+	struct ieee80211_link_sta *link_sta;
 	struct mt7996_sta *msta;
 	struct sk_buff *skb;
 	int ret;
 
 	msta = sta ? (struct mt7996_sta *)sta->drv_priv : &mvif->sta;
+	link_sta = sta ? &sta->deflink : NULL;
 
 	skb = __mt76_connac_mcu_alloc_sta_req(&dev->mt76, &mvif->mt76,
 					      &msta->wcid,
@@ -2170,7 +2172,8 @@ int mt7996_mcu_add_sta(struct mt7996_dev *dev, struct ieee80211_vif *vif,
 		return PTR_ERR(skb);
 
 	/* starec basic */
-	mt76_connac_mcu_sta_basic_tlv(&dev->mt76, skb, vif, sta, enable, newly);
+	mt76_connac_mcu_sta_basic_tlv(&dev->mt76, skb, vif, link_sta,
+				      enable, newly);
 
 	if (!enable)
 		goto out;
-- 
cgit v1.2.3-58-ga151


From 612004dd7480b0a039909f3959a73cbc59523f65 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:32 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_sta_hdr_trans_tlv for per-link
 STA

Extend mt7925_mcu_sta_hdr_trans_tlv with per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-39-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index bf2afaf4c0dc..901a5d1f4f04 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -847,7 +847,7 @@ EXPORT_SYMBOL_GPL(mt7925_run_firmware);
 static void
 mt7925_mcu_sta_hdr_trans_tlv(struct sk_buff *skb,
 			     struct ieee80211_vif *vif,
-			     struct ieee80211_sta *sta)
+			     struct ieee80211_link_sta *link_sta)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct sta_rec_hdr_trans *hdr_trans;
@@ -863,8 +863,8 @@ mt7925_mcu_sta_hdr_trans_tlv(struct sk_buff *skb,
 	else
 		hdr_trans->from_ds = true;
 
-	if (sta)
-		wcid = (struct mt76_wcid *)sta->drv_priv;
+	if (link_sta)
+		wcid = (struct mt76_wcid *)link_sta->sta->drv_priv;
 	else
 		wcid = &mvif->sta.deflink.wcid;
 
@@ -883,6 +883,7 @@ int mt7925_mcu_wtbl_update_hdr_trans(struct mt792x_dev *dev,
 				     struct ieee80211_sta *sta)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct ieee80211_link_sta *link_sta = sta ? &sta->deflink : NULL;
 	struct mt792x_sta *msta;
 	struct sk_buff *skb;
 
@@ -895,7 +896,7 @@ int mt7925_mcu_wtbl_update_hdr_trans(struct mt792x_dev *dev,
 		return PTR_ERR(skb);
 
 	/* starec hdr trans */
-	mt7925_mcu_sta_hdr_trans_tlv(skb, vif, sta);
+	mt7925_mcu_sta_hdr_trans_tlv(skb, vif, link_sta);
 	return mt76_mcu_skb_send_msg(&dev->mt76, skb,
 				     MCU_WMWA_UNI_CMD(STA_REC_UPDATE), true);
 }
@@ -1664,12 +1665,8 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 		mt7925_mcu_sta_mld_tlv(skb, info->vif, info->link_sta->sta);
 	}
 
-	if (info->enable) {
-		struct ieee80211_sta *sta = info->link_sta ?
-			info->link_sta->sta : NULL;
-
-		mt7925_mcu_sta_hdr_trans_tlv(skb, info->vif, sta);
-	}
+	if (info->enable)
+		mt7925_mcu_sta_hdr_trans_tlv(skb, info->vif, info->link_sta);
 
 	return mt76_mcu_skb_send_msg(dev, skb, info->cmd, true);
 }
-- 
cgit v1.2.3-58-ga151


From 064a5955aa2732844ca1e30b93d66c9ce179e811 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:33 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_add_bss_info for per-link STA

Extend mt7925_mcu_add_bss_info with the per-link STA configuration.

The patch we created is a prerequisite to enable the MLO function in the
driver. It is purely a refactoring patch so the functionality should
remain unchanged.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-40-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 9 +++++----
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c  | 6 +++---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.h  | 2 +-
 3 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index f3a32fe9a754..e701d40569fb 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -489,6 +489,7 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_sta *msta = sta ? (struct mt792x_sta *)sta->drv_priv :
 				  &mvif->sta;
+	struct ieee80211_link_sta *link_sta = sta ? &sta->deflink : NULL;
 	struct mt76_wcid *wcid = &msta->deflink.wcid;
 	struct ieee80211_bss_conf *link_conf;
 	u8 *wcid_keyidx = &wcid->hw_key_idx;
@@ -535,7 +536,7 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 
 		mvif->bss_conf.mt76.cipher = mt7925_mcu_get_cipher(key->cipher);
 		mt7925_mcu_add_bss_info(phy, mvif->bss_conf.mt76.ctx, link_conf,
-					sta, true);
+					link_sta, true);
 	}
 
 	if (cmd == SET_KEY)
@@ -738,7 +739,7 @@ int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 	/* should update bss info before STA add */
 	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
 		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
-					link_conf, sta, false);
+					link_conf, &sta->deflink, false);
 
 	ret = mt7925_mcu_sta_update(dev, &sta->deflink, vif, true,
 				    MT76_STA_INFO_STATE_NONE);
@@ -765,7 +766,7 @@ void mt7925_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 
 	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
 		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
-					link_conf, sta, true);
+					link_conf, &sta->deflink, true);
 
 	ewma_avg_signal_init(&msta->deflink.avg_ack_signal);
 
@@ -802,7 +803,7 @@ void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 		ewma_rssi_init(&mvif->bss_conf.rssi);
 		if (!sta->tdls)
 			mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
-						link_conf, sta, false);
+						link_conf, &sta->deflink, false);
 	}
 
 	spin_lock_bh(&mdev->sta_poll_lock);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 901a5d1f4f04..03cce9fca0c3 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2338,7 +2338,7 @@ int mt7925_mcu_set_timing(struct mt792x_phy *phy,
 int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 			    struct ieee80211_chanctx_conf *ctx,
 			    struct ieee80211_bss_conf *link_conf,
-			    struct ieee80211_sta *sta,
+			    struct ieee80211_link_sta *link_sta,
 			    int enable)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)link_conf->vif->drv_priv;
@@ -2353,13 +2353,13 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 		return PTR_ERR(skb);
 
 	/* bss_basic must be first */
-	mt7925_mcu_bss_basic_tlv(skb, link_conf, sta, ctx, phy->mt76,
+	mt7925_mcu_bss_basic_tlv(skb, link_conf, link_sta->sta, ctx, phy->mt76,
 				 mvif->sta.deflink.wcid.idx, enable);
 	mt7925_mcu_bss_sec_tlv(skb, link_conf);
 
 	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf);
 	mt7925_mcu_bss_qos_tlv(skb, link_conf);
-	mt7925_mcu_bss_mld_tlv(skb, link_conf, sta);
+	mt7925_mcu_bss_mld_tlv(skb, link_conf, link_sta->sta);
 	mt7925_mcu_bss_ifs_tlv(skb, link_conf);
 
 	if (link_conf->he_support) {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
index c5197577cdc8..da4e84f468d2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
@@ -595,7 +595,7 @@ int mt7925_mcu_sched_scan_enable(struct mt76_phy *phy,
 int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 			    struct ieee80211_chanctx_conf *ctx,
 			    struct ieee80211_bss_conf *link_conf,
-			    struct ieee80211_sta *sta,
+			    struct ieee80211_link_sta *link_sta,
 			    int enable);
 int mt7925_mcu_set_timing(struct mt792x_phy *phy,
 			  struct ieee80211_bss_conf *link_conf);
-- 
cgit v1.2.3-58-ga151


From 523f5d9c256079fe21087b3dfc4c3c0c1fffab53 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:34 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_bss_mld_tlv for per-link STA

Extend mt7925_mcu_bss_mld_tlv with the per-link STA configuration.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-41-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 03cce9fca0c3..2a96ff89ef67 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2228,7 +2228,7 @@ mt7925_mcu_bss_bmc_tlv(struct sk_buff *skb, struct mt792x_phy *phy,
 static void
 mt7925_mcu_bss_mld_tlv(struct sk_buff *skb,
 		       struct ieee80211_bss_conf *link_conf,
-		       struct ieee80211_sta *sta)
+		       struct ieee80211_link_sta *link_sta)
 {
 	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
 	bool is_mld = ieee80211_vif_is_mld(link_conf->vif);
@@ -2238,13 +2238,13 @@ mt7925_mcu_bss_mld_tlv(struct sk_buff *skb,
 	tlv = mt76_connac_mcu_add_tlv(skb, UNI_BSS_INFO_MLD, sizeof(*mld));
 	mld = (struct bss_mld_tlv *)tlv;
 
-	mld->link_id = sta ? (is_mld ? link_conf->link_id : 0) : 0xff;
+	mld->link_id = link_sta ? (is_mld ? link_conf->link_id : 0) : 0xff;
 	mld->group_mld_id = is_mld ? mconf->mt76.idx : 0xff;
 	mld->own_mld_id = mconf->mt76.idx + 32;
 	mld->remap_idx = 0xff;
 
-	if (sta)
-		memcpy(mld->mac_addr, sta->addr, ETH_ALEN);
+	if (link_sta)
+		memcpy(mld->mac_addr, link_sta->addr, ETH_ALEN);
 }
 
 static void
@@ -2359,7 +2359,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 
 	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf);
 	mt7925_mcu_bss_qos_tlv(skb, link_conf);
-	mt7925_mcu_bss_mld_tlv(skb, link_conf, link_sta->sta);
+	mt7925_mcu_bss_mld_tlv(skb, link_conf, link_sta);
 	mt7925_mcu_bss_ifs_tlv(skb, link_conf);
 
 	if (link_conf->he_support) {
-- 
cgit v1.2.3-58-ga151


From c1c98696de70e378a70b5b34ecf6193a8ccad4e5 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:35 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_bss_basic_tlv for per-link STA

Extend mt7925_mcu_bss_basic_tlv with the per-link STA configuration.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-42-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 2a96ff89ef67..09ffa7abe8aa 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2079,19 +2079,17 @@ mt7925_get_phy_mode_ext(struct mt76_phy *phy, struct ieee80211_vif *vif,
 static void
 mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 			 struct ieee80211_bss_conf *link_conf,
-			 struct ieee80211_sta *sta,
+			 struct ieee80211_link_sta *link_sta,
 			 struct ieee80211_chanctx_conf *ctx,
 			 struct mt76_phy *phy, u16 wlan_idx,
 			 bool enable)
 {
 	struct ieee80211_vif *vif = link_conf->vif;
 	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
-	struct mt792x_sta *msta = sta ? (struct mt792x_sta *)sta->drv_priv :
-				  &mconf->vif->sta;
-	struct ieee80211_link_sta *link_sta = sta ? &sta->deflink : NULL;
 	struct cfg80211_chan_def *chandef = ctx ? &ctx->def : &phy->chandef;
 	enum nl80211_band band = chandef->chan->band;
 	struct mt76_connac_bss_basic_tlv *basic_req;
+	struct mt792x_link_sta *mlink;
 	struct tlv *tlv;
 	int conn_type;
 	u8 idx;
@@ -2116,7 +2114,18 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 	basic_req->bcn_interval = cpu_to_le16(link_conf->beacon_int);
 	basic_req->dtim_period = link_conf->dtim_period;
 	basic_req->bmc_tx_wlan_idx = cpu_to_le16(wlan_idx);
-	basic_req->sta_idx = cpu_to_le16(msta->deflink.wcid.idx);
+
+	if (link_sta) {
+		struct mt792x_sta *msta;
+
+		msta = (struct mt792x_sta *)link_sta->sta->drv_priv;
+		mlink = mt792x_sta_to_link(msta, link_sta->link_id);
+
+	} else {
+		mlink = &mconf->vif->sta.deflink;
+	}
+
+	basic_req->sta_idx = cpu_to_le16(mlink->wcid.idx);
 	basic_req->omac_idx = mconf->mt76.omac_idx;
 	basic_req->band_idx = mconf->mt76.band_idx;
 	basic_req->wmm_idx = mconf->mt76.wmm_idx;
@@ -2353,7 +2362,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 		return PTR_ERR(skb);
 
 	/* bss_basic must be first */
-	mt7925_mcu_bss_basic_tlv(skb, link_conf, link_sta->sta, ctx, phy->mt76,
+	mt7925_mcu_bss_basic_tlv(skb, link_conf, link_sta, ctx, phy->mt76,
 				 mvif->sta.deflink.wcid.idx, enable);
 	mt7925_mcu_bss_sec_tlv(skb, link_conf);
 
-- 
cgit v1.2.3-58-ga151


From 5029871583c3cf42f3320b7709820ad44a1313fc Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:36 -0700
Subject: wifi: mt76: mt7925: add mt7925_mac_link_sta_add to create per-link
 STA

The mt7925_mac_link_sta_add function currently adds the per-link STA.
We will extend this function when we implement the MLO functionality.

This patch only includes structural changes and does not involve any
logic changes.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-43-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 47 ++++++++++++++++--------
 1 file changed, 31 insertions(+), 16 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index e701d40569fb..3872b5debdd8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -703,45 +703,46 @@ mt7925_get_rates_table(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	return mvif->basic_rates_idx;
 }
 
-int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
-		       struct ieee80211_sta *sta)
+static int mt7925_mac_link_sta_add(struct mt76_dev *mdev,
+				   struct ieee80211_vif *vif,
+				   struct ieee80211_link_sta *link_sta)
 {
 	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
-	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct ieee80211_bss_conf *link_conf;
+	struct mt792x_link_sta *mlink;
+	struct mt792x_sta *msta;
 	int ret, idx;
 
+	msta = (struct mt792x_sta *)link_sta->sta->drv_priv;
+	mlink = mt792x_sta_to_link(msta, link_sta->link_id);
+
 	idx = mt76_wcid_alloc(dev->mt76.wcid_mask, MT792x_WTBL_STA - 1);
 	if (idx < 0)
 		return -ENOSPC;
 
-	INIT_LIST_HEAD(&msta->deflink.wcid.poll_list);
-	msta->vif = mvif;
-	msta->deflink.wcid.sta = 1;
-	msta->deflink.wcid.idx = idx;
-	msta->deflink.wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
-	msta->deflink.wcid.tx_info |= MT_WCID_TX_INFO_SET;
-	msta->deflink.last_txs = jiffies;
+	INIT_LIST_HEAD(&mlink->wcid.poll_list);
+	mlink->wcid.sta = 1;
+	mlink->wcid.idx = idx;
+	mlink->wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
+	mlink->wcid.tx_info |= MT_WCID_TX_INFO_SET;
+	mlink->last_txs = jiffies;
 
 	ret = mt76_connac_pm_wake(&dev->mphy, &dev->pm);
 	if (ret)
 		return ret;
 
-	if (vif->type == NL80211_IFTYPE_STATION)
-		mvif->wep_sta = msta;
-
 	mt7925_mac_wtbl_update(dev, idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
 	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
 
 	/* should update bss info before STA add */
-	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
+	if (vif->type == NL80211_IFTYPE_STATION && !link_sta->sta->tdls)
 		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
-					link_conf, &sta->deflink, false);
+					link_conf, link_sta, false);
 
-	ret = mt7925_mcu_sta_update(dev, &sta->deflink, vif, true,
+	ret = mt7925_mcu_sta_update(dev, link_sta, vif, true,
 				    MT76_STA_INFO_STATE_NONE);
 	if (ret)
 		return ret;
@@ -750,6 +751,20 @@ int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 
 	return 0;
 }
+
+int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
+		       struct ieee80211_sta *sta)
+{
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
+
+	msta->vif = mvif;
+
+	if (vif->type == NL80211_IFTYPE_STATION)
+		mvif->wep_sta = msta;
+
+	return mt7925_mac_link_sta_add(mdev, vif, &sta->deflink);
+}
 EXPORT_SYMBOL_GPL(mt7925_mac_sta_add);
 
 void mt7925_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
-- 
cgit v1.2.3-58-ga151


From 89397bccc882a446373874cd8c4164e1612d8494 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:37 -0700
Subject: wifi: mt76: mt7925: add mt7925_mac_link_sta_assoc to associate
 per-link STA

The mt7925_mac_link_sta_add function currently associates the per-link STA.
We will extend this function when we implement the MLO functionality.

This patch only includes structural changes and does not involve any
logic changes.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-44-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 29 ++++++++++++++++--------
 1 file changed, 20 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 3872b5debdd8..202bd84716ab 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -767,32 +767,43 @@ int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 }
 EXPORT_SYMBOL_GPL(mt7925_mac_sta_add);
 
-void mt7925_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
-			  struct ieee80211_sta *sta)
+static void mt7925_mac_link_sta_assoc(struct mt76_dev *mdev,
+				      struct ieee80211_vif *vif,
+				      struct ieee80211_link_sta *link_sta)
 {
 	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
-	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct ieee80211_bss_conf *link_conf;
+	struct mt792x_link_sta *mlink;
+	struct mt792x_sta *msta;
+
+	msta = (struct mt792x_sta *)link_sta->sta->drv_priv;
+	mlink = mt792x_sta_to_link(msta, link_sta->link_id);
 
 	mt792x_mutex_acquire(dev);
 
 	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
 
-	if (vif->type == NL80211_IFTYPE_STATION && !sta->tdls)
+	if (vif->type == NL80211_IFTYPE_STATION && !link_sta->sta->tdls)
 		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
-					link_conf, &sta->deflink, true);
+					link_conf, link_sta, true);
 
-	ewma_avg_signal_init(&msta->deflink.avg_ack_signal);
+	ewma_avg_signal_init(&mlink->avg_ack_signal);
 
-	mt7925_mac_wtbl_update(dev, msta->deflink.wcid.idx,
+	mt7925_mac_wtbl_update(dev, mlink->wcid.idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
-	memset(msta->deflink.airtime_ac, 0, sizeof(msta->deflink.airtime_ac));
+	memset(mlink->airtime_ac, 0, sizeof(mlink->airtime_ac));
 
-	mt7925_mcu_sta_update(dev, &sta->deflink, vif, true, MT76_STA_INFO_STATE_ASSOC);
+	mt7925_mcu_sta_update(dev, link_sta, vif, true, MT76_STA_INFO_STATE_ASSOC);
 
 	mt792x_mutex_release(dev);
 }
+
+void mt7925_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
+			  struct ieee80211_sta *sta)
+{
+	mt7925_mac_link_sta_assoc(mdev, vif, &sta->deflink);
+}
 EXPORT_SYMBOL_GPL(mt7925_mac_sta_assoc);
 
 void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
-- 
cgit v1.2.3-58-ga151


From 5f5b6a745c697c1a80cfd9eacf1f91613d29b050 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:38 -0700
Subject: wifi: mt76: mt7925: add mt7925_mac_link_sta_remove to remove per-link
 STA

The mt7925_mac_link_sta_remove function currently removes the per-link STA.
We will extend this function when we implement the MLO functionality.

This patch only includes structural changes and does not involve any
logic changes.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-45-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 46 ++++++++++++++++--------
 1 file changed, 31 insertions(+), 15 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 202bd84716ab..21a2d1bd8d18 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -806,39 +806,55 @@ void mt7925_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 }
 EXPORT_SYMBOL_GPL(mt7925_mac_sta_assoc);
 
-void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
-			   struct ieee80211_sta *sta)
+static void mt7925_mac_link_sta_remove(struct mt76_dev *mdev,
+				       struct ieee80211_vif *vif,
+				       struct ieee80211_link_sta *link_sta)
 {
 	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
-	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
 	struct ieee80211_bss_conf *link_conf;
+	struct mt792x_link_sta *mlink;
+	struct mt792x_sta *msta;
 
-	mt76_connac_free_pending_tx_skbs(&dev->pm, &msta->deflink.wcid);
+	msta = (struct mt792x_sta *)link_sta->sta->drv_priv;
+	mlink = mt792x_sta_to_link(msta, link_sta->link_id);
+
+	mt76_connac_free_pending_tx_skbs(&dev->pm, &mlink->wcid);
 	mt76_connac_pm_wake(&dev->mphy, &dev->pm);
 
-	mt7925_mcu_sta_update(dev, &sta->deflink, vif, false, MT76_STA_INFO_STATE_NONE);
-	mt7925_mac_wtbl_update(dev, msta->deflink.wcid.idx,
+	mt7925_mcu_sta_update(dev, link_sta, vif, false,
+			      MT76_STA_INFO_STATE_NONE);
+	mt7925_mac_wtbl_update(dev, mlink->wcid.idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
 	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
 
-	if (vif->type == NL80211_IFTYPE_STATION) {
+	if (vif->type == NL80211_IFTYPE_STATION && !link_sta->sta->tdls) {
 		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-
-		mvif->wep_sta = NULL;
-		ewma_rssi_init(&mvif->bss_conf.rssi);
-		if (!sta->tdls)
-			mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
-						link_conf, &sta->deflink, false);
+		mt7925_mcu_add_bss_info(&dev->phy,
+					mvif->bss_conf.mt76.ctx, link_conf,
+					link_sta, false);
 	}
 
 	spin_lock_bh(&mdev->sta_poll_lock);
-	if (!list_empty(&msta->deflink.wcid.poll_list))
-		list_del_init(&msta->deflink.wcid.poll_list);
+	if (!list_empty(&mlink->wcid.poll_list))
+		list_del_init(&mlink->wcid.poll_list);
 	spin_unlock_bh(&mdev->sta_poll_lock);
 
 	mt76_connac_power_save_sched(&dev->mphy, &dev->pm);
 }
+
+void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
+			   struct ieee80211_sta *sta)
+{
+	mt7925_mac_link_sta_remove(mdev, vif, &sta->deflink);
+
+	if (vif->type == NL80211_IFTYPE_STATION) {
+		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+
+		mvif->wep_sta = NULL;
+		ewma_rssi_init(&mvif->bss_conf.rssi);
+	}
+}
 EXPORT_SYMBOL_GPL(mt7925_mac_sta_remove);
 
 static int mt7925_set_rts_threshold(struct ieee80211_hw *hw, u32 val)
-- 
cgit v1.2.3-58-ga151


From 95977b1d10cfeb12b67a76dc2eefa86300c396d1 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:39 -0700
Subject: wifi: mt76: mt7925: add mt7925_mac_link_bss_add to create per-link
 BSS

The mt7925_mac_link_bss_add function currently creates the per-link BSS.
We will extend this function when we implement the MLO functionality.

This patch only includes structural changes and does not involve any
logic changes.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-46-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 80 +++++++++++++++---------
 1 file changed, 49 insertions(+), 31 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 21a2d1bd8d18..8b10b63bc98d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -317,63 +317,81 @@ static int mt7925_start(struct ieee80211_hw *hw)
 	return err;
 }
 
-static int
-mt7925_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
+static int mt7925_mac_link_bss_add(struct mt792x_dev *dev,
+				   struct ieee80211_bss_conf *link_conf,
+				   struct mt792x_link_sta *mlink)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	struct mt792x_dev *dev = mt792x_hw_dev(hw);
-	struct mt792x_phy *phy = mt792x_hw_phy(hw);
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
+	struct ieee80211_vif *vif = link_conf->vif;
+	struct mt792x_vif *mvif = mconf->vif;
 	struct mt76_txq *mtxq;
 	int idx, ret = 0;
 
-	mt792x_mutex_acquire(dev);
-
-	mvif->bss_conf.mt76.idx = __ffs64(~dev->mt76.vif_mask);
-	if (mvif->bss_conf.mt76.idx >= MT792x_MAX_INTERFACES) {
+	mconf->mt76.idx = __ffs64(~dev->mt76.vif_mask);
+	if (mconf->mt76.idx >= MT792x_MAX_INTERFACES) {
 		ret = -ENOSPC;
 		goto out;
 	}
 
-	mvif->bss_conf.mt76.omac_idx = mvif->bss_conf.mt76.idx;
-	mvif->phy = phy;
-	mvif->bss_conf.mt76.band_idx = 0;
-	mvif->bss_conf.mt76.wmm_idx = mvif->bss_conf.mt76.idx % MT76_CONNAC_MAX_WMM_SETS;
-	mvif->bss_conf.vif = mvif;
+	mconf->mt76.omac_idx = mconf->mt76.idx;
+	mconf->mt76.band_idx = 0;
+	mconf->mt76.wmm_idx = mconf->mt76.idx % MT76_CONNAC_MAX_WMM_SETS;
 
-	if (phy->mt76->chandef.chan->band != NL80211_BAND_2GHZ)
-		mvif->bss_conf.mt76.basic_rates_idx = MT792x_BASIC_RATES_TBL + 4;
+	if (mvif->phy->mt76->chandef.chan->band != NL80211_BAND_2GHZ)
+		mconf->mt76.basic_rates_idx = MT792x_BASIC_RATES_TBL + 4;
 	else
-		mvif->bss_conf.mt76.basic_rates_idx = MT792x_BASIC_RATES_TBL;
+		mconf->mt76.basic_rates_idx = MT792x_BASIC_RATES_TBL;
 
-	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, &vif->bss_conf,
-					  &mvif->sta.deflink.wcid, true);
+	ret = mt76_connac_mcu_uni_add_dev(&dev->mphy, link_conf,
+					  &mlink->wcid, true);
 	if (ret)
 		goto out;
 
-	dev->mt76.vif_mask |= BIT_ULL(mvif->bss_conf.mt76.idx);
-	phy->omac_mask |= BIT_ULL(mvif->bss_conf.mt76.omac_idx);
+	dev->mt76.vif_mask |= BIT_ULL(mconf->mt76.idx);
+	mvif->phy->omac_mask |= BIT_ULL(mconf->mt76.omac_idx);
 
-	idx = MT792x_WTBL_RESERVED - mvif->bss_conf.mt76.idx;
+	idx = MT792x_WTBL_RESERVED - mconf->mt76.idx;
 
-	INIT_LIST_HEAD(&mvif->sta.deflink.wcid.poll_list);
-	mvif->sta.deflink.wcid.idx = idx;
-	mvif->sta.deflink.wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
-	mvif->sta.deflink.wcid.hw_key_idx = -1;
-	mvif->sta.deflink.wcid.tx_info |= MT_WCID_TX_INFO_SET;
-	mvif->sta.vif = mvif;
-	mt76_wcid_init(&mvif->sta.deflink.wcid);
+	INIT_LIST_HEAD(&mlink->wcid.poll_list);
+	mlink->wcid.idx = idx;
+	mlink->wcid.phy_idx = mconf->mt76.band_idx;
+	mlink->wcid.hw_key_idx = -1;
+	mlink->wcid.tx_info |= MT_WCID_TX_INFO_SET;
+	mt76_wcid_init(&mlink->wcid);
 
 	mt7925_mac_wtbl_update(dev, idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
-	ewma_rssi_init(&mvif->bss_conf.rssi);
+	ewma_rssi_init(&mconf->rssi);
 
-	rcu_assign_pointer(dev->mt76.wcid[idx], &mvif->sta.deflink.wcid);
+	rcu_assign_pointer(dev->mt76.wcid[idx], &mlink->wcid);
 	if (vif->txq) {
 		mtxq = (struct mt76_txq *)vif->txq->drv_priv;
 		mtxq->wcid = idx;
 	}
 
+out:
+	return ret;
+}
+
+static int
+mt7925_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
+{
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+	struct mt792x_phy *phy = mt792x_hw_phy(hw);
+	int ret = 0;
+
+	mt792x_mutex_acquire(dev);
+
+	mvif->phy = phy;
+	mvif->bss_conf.vif = mvif;
+	mvif->sta.vif = mvif;
+
+	ret = mt7925_mac_link_bss_add(dev, &vif->bss_conf, &mvif->sta.deflink);
+	if (ret < 0)
+		goto out;
+
 	vif->driver_flags |= IEEE80211_VIF_BEACON_FILTER;
 out:
 	mt792x_mutex_release(dev);
-- 
cgit v1.2.3-58-ga151


From 1541d63c5fe2cebce85b2af84a2850a302ffda9c Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:40 -0700
Subject: wifi: mt76: mt7925: add mt7925_mac_link_bss_remove to remove per-link
 BSS

The mt7925_mac_link_bss_remove function currently removes the per-link BSS.
We will extend this function when we implement the MLO functionality.

This patch only includes structural changes and does not involve any
logic changes.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-47-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt792x_core.c | 35 ++++++++++++++----------
 1 file changed, 21 insertions(+), 14 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_core.c b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
index b1c51e3f39d0..813296fad0ed 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_core.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
@@ -113,31 +113,38 @@ void mt792x_stop(struct ieee80211_hw *hw, bool suspend)
 }
 EXPORT_SYMBOL_GPL(mt792x_stop);
 
-void mt792x_remove_interface(struct ieee80211_hw *hw,
-			     struct ieee80211_vif *vif)
+static void mt792x_mac_link_bss_remove(struct mt792x_dev *dev,
+				       struct ieee80211_bss_conf *link_conf,
+				       struct mt792x_link_sta *mlink)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	struct mt792x_sta *msta = &mvif->sta;
-	struct mt792x_dev *dev = mt792x_hw_dev(hw);
-	struct mt792x_phy *phy = mt792x_hw_phy(hw);
-	int idx = msta->deflink.wcid.idx;
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
+	int idx = mlink->wcid.idx;
 
 	mt792x_mutex_acquire(dev);
-	mt76_connac_free_pending_tx_skbs(&dev->pm, &msta->deflink.wcid);
-	mt76_connac_mcu_uni_add_dev(&dev->mphy, &vif->bss_conf, &mvif->sta.deflink.wcid, false);
+	mt76_connac_free_pending_tx_skbs(&dev->pm, &mlink->wcid);
+	mt76_connac_mcu_uni_add_dev(&dev->mphy, link_conf, &mlink->wcid, false);
 
 	rcu_assign_pointer(dev->mt76.wcid[idx], NULL);
 
-	dev->mt76.vif_mask &= ~BIT_ULL(mvif->bss_conf.mt76.idx);
-	phy->omac_mask &= ~BIT_ULL(mvif->bss_conf.mt76.omac_idx);
+	dev->mt76.vif_mask &= ~BIT_ULL(mconf->mt76.idx);
+	mconf->vif->phy->omac_mask &= ~BIT_ULL(mconf->mt76.omac_idx);
 	mt792x_mutex_release(dev);
 
 	spin_lock_bh(&dev->mt76.sta_poll_lock);
-	if (!list_empty(&msta->deflink.wcid.poll_list))
-		list_del_init(&msta->deflink.wcid.poll_list);
+	if (!list_empty(&mlink->wcid.poll_list))
+		list_del_init(&mlink->wcid.poll_list);
 	spin_unlock_bh(&dev->mt76.sta_poll_lock);
 
-	mt76_wcid_cleanup(&dev->mt76, &msta->deflink.wcid);
+	mt76_wcid_cleanup(&dev->mt76, &mlink->wcid);
+}
+
+void mt792x_remove_interface(struct ieee80211_hw *hw,
+			     struct ieee80211_vif *vif)
+{
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+
+	mt792x_mac_link_bss_remove(dev, &vif->bss_conf, &mvif->sta.deflink);
 }
 EXPORT_SYMBOL_GPL(mt792x_remove_interface);
 
-- 
cgit v1.2.3-58-ga151


From 8b7eb93b7a75f453d2b35c5b904895b628949593 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 12 Jun 2024 20:02:41 -0700
Subject: wifi: mt76: mt7925: simpify mt7925_mcu_sta_cmd logic by removing
 fw_offload

The mt7925 only supports offload firmware making the fw_offload check
redundant. This update simplifies the logic in "mt7925_mcu_sta_cmd"
by removing the unnecessary check for fw_offoad.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240613030241.5771-48-sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 09ffa7abe8aa..a67d8dd6075f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1644,7 +1644,7 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
 
-	if (info->link_sta || !info->offload_fw)
+	if (info->link_sta)
 		mt76_connac_mcu_sta_basic_tlv(dev, skb, info->vif,
 					      info->link_sta,
 					      info->enable, info->newly);
-- 
cgit v1.2.3-58-ga151


From 2938b50917a5431efa8c00c9c8b2ee89c406cf05 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 3 Jul 2024 14:42:06 -0700
Subject: wifi: mt76: mt7925: update mt76_connac_mcu_uni_add_dev for MLO

Set OMAC address with the per-link BSS. The change remains compatible with
the non-MLO mode and the older firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/6b2502d0c3a53093f83ae33329e9d782ba2465b3.1720042294.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
index 4e3c8af98fe7..4dce03ddbfa4 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c
@@ -1214,7 +1214,7 @@ int mt76_connac_mcu_uni_add_dev(struct mt76_phy *phy,
 	idx = mvif->omac_idx > EXT_BSSID_START ? HW_BSSID_0 : mvif->omac_idx;
 	basic_req.basic.hw_bss_idx = idx;
 
-	memcpy(dev_req.tlv.omac_addr, bss_conf->vif->addr, ETH_ALEN);
+	memcpy(dev_req.tlv.omac_addr, bss_conf->addr, ETH_ALEN);
 
 	cmd = enable ? MCU_UNI_CMD(DEV_INFO_UPDATE) : MCU_UNI_CMD(BSS_INFO_UPDATE);
 	data = enable ? (void *)&dev_req : (void *)&basic_req;
-- 
cgit v1.2.3-58-ga151


From 3878b4333602498a6af3df75e78d5fe255fa5b17 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:42 -0700
Subject: wifi: mt76: mt7925: update mt7925_mac_link_sta_[add, assoc, remove]
 for MLO

[add, assoc, remove] the per-link STA from the corresponding to per-link
BSS with the same link id. The change remains compatible with the non-MLO
mode.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/2410cc73eec3849c2185b5cfa613b7e903e68abb.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 32 +++++++++++++++---------
 1 file changed, 20 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 8b10b63bc98d..a043d93f3e1d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -728,21 +728,24 @@ static int mt7925_mac_link_sta_add(struct mt76_dev *mdev,
 	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct ieee80211_bss_conf *link_conf;
+	struct mt792x_bss_conf *mconf;
+	u8 link_id = link_sta->link_id;
 	struct mt792x_link_sta *mlink;
 	struct mt792x_sta *msta;
 	int ret, idx;
 
 	msta = (struct mt792x_sta *)link_sta->sta->drv_priv;
-	mlink = mt792x_sta_to_link(msta, link_sta->link_id);
+	mlink = mt792x_sta_to_link(msta, link_id);
 
 	idx = mt76_wcid_alloc(dev->mt76.wcid_mask, MT792x_WTBL_STA - 1);
 	if (idx < 0)
 		return -ENOSPC;
 
+	mconf = mt792x_vif_to_link(mvif, link_id);
 	INIT_LIST_HEAD(&mlink->wcid.poll_list);
 	mlink->wcid.sta = 1;
 	mlink->wcid.idx = idx;
-	mlink->wcid.phy_idx = mvif->bss_conf.mt76.band_idx;
+	mlink->wcid.phy_idx = mconf->mt76.band_idx;
 	mlink->wcid.tx_info |= MT_WCID_TX_INFO_SET;
 	mlink->last_txs = jiffies;
 
@@ -753,11 +756,11 @@ static int mt7925_mac_link_sta_add(struct mt76_dev *mdev,
 	mt7925_mac_wtbl_update(dev, idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
-	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
+	link_conf = mt792x_vif_to_bss_conf(vif, link_id);
 
 	/* should update bss info before STA add */
 	if (vif->type == NL80211_IFTYPE_STATION && !link_sta->sta->tdls)
-		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
+		mt7925_mcu_add_bss_info(&dev->phy, mconf->mt76.ctx,
 					link_conf, link_sta, false);
 
 	ret = mt7925_mcu_sta_update(dev, link_sta, vif, true,
@@ -790,7 +793,6 @@ static void mt7925_mac_link_sta_assoc(struct mt76_dev *mdev,
 				      struct ieee80211_link_sta *link_sta)
 {
 	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct ieee80211_bss_conf *link_conf;
 	struct mt792x_link_sta *mlink;
 	struct mt792x_sta *msta;
@@ -802,9 +804,13 @@ static void mt7925_mac_link_sta_assoc(struct mt76_dev *mdev,
 
 	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
 
-	if (vif->type == NL80211_IFTYPE_STATION && !link_sta->sta->tdls)
-		mt7925_mcu_add_bss_info(&dev->phy, mvif->bss_conf.mt76.ctx,
+	if (vif->type == NL80211_IFTYPE_STATION && !link_sta->sta->tdls) {
+		struct mt792x_bss_conf *mconf;
+
+		mconf = mt792x_link_conf_to_mconf(link_conf);
+		mt7925_mcu_add_bss_info(&dev->phy, mconf->mt76.ctx,
 					link_conf, link_sta, true);
+	}
 
 	ewma_avg_signal_init(&mlink->avg_ack_signal);
 
@@ -830,11 +836,12 @@ static void mt7925_mac_link_sta_remove(struct mt76_dev *mdev,
 {
 	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
 	struct ieee80211_bss_conf *link_conf;
+	u8 link_id = link_sta->link_id;
 	struct mt792x_link_sta *mlink;
 	struct mt792x_sta *msta;
 
 	msta = (struct mt792x_sta *)link_sta->sta->drv_priv;
-	mlink = mt792x_sta_to_link(msta, link_sta->link_id);
+	mlink = mt792x_sta_to_link(msta, link_id);
 
 	mt76_connac_free_pending_tx_skbs(&dev->pm, &mlink->wcid);
 	mt76_connac_pm_wake(&dev->mphy, &dev->pm);
@@ -844,12 +851,13 @@ static void mt7925_mac_link_sta_remove(struct mt76_dev *mdev,
 	mt7925_mac_wtbl_update(dev, mlink->wcid.idx,
 			       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
 
-	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
+	link_conf = mt792x_vif_to_bss_conf(vif, link_id);
 
 	if (vif->type == NL80211_IFTYPE_STATION && !link_sta->sta->tdls) {
-		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-		mt7925_mcu_add_bss_info(&dev->phy,
-					mvif->bss_conf.mt76.ctx, link_conf,
+		struct mt792x_bss_conf *mconf;
+
+		mconf = mt792x_link_conf_to_mconf(link_conf);
+		mt7925_mcu_add_bss_info(&dev->phy, mconf->mt76.ctx, link_conf,
 					link_sta, false);
 	}
 
-- 
cgit v1.2.3-58-ga151


From 882b97c16ab861d7e267262df48b32d6db4121f1 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:43 -0700
Subject: wifi: mt76: mt7925: set Tx queue parameters according to link id

Configure TX queue parameters according to link id.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/84a401891b8942f66d1ffbd0a3f10866c94d1bc0.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index a043d93f3e1d..da70c203d618 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1306,6 +1306,7 @@ mt7925_conf_tx(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	       const struct ieee80211_tx_queue_params *params)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_bss_conf *mconf = mt792x_vif_to_link(mvif, link_id);
 	static const u8 mq_to_aci[] = {
 		    [IEEE80211_AC_VO] = 3,
 		    [IEEE80211_AC_VI] = 2,
@@ -1314,7 +1315,7 @@ mt7925_conf_tx(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 	};
 
 	/* firmware uses access class index */
-	mvif->bss_conf.queue_params[mq_to_aci[queue]] = *params;
+	mconf->queue_params[mq_to_aci[queue]] = *params;
 
 	return 0;
 }
-- 
cgit v1.2.3-58-ga151


From 00e1ca0cad2e072df173ac6e8290109288352307 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:44 -0700
Subject: wifi: mt76: mt7925: set mt7925_mcu_sta_key_tlv according to link id

configure mt7925_mcu_sta_key_tlv according to link id

We created a link id field in the common structure mt76_wcid so that
mt7925_mcu_sta_key_tlv can access per-link STA/BSS. Additionally,
.link_id will be required when reporting the link information of Rx status
to mac80211. We will submit the changes for reporting the Rx status in
a separate patch.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/16d65aae010113f9be3c4e6e759b4226d396a1be.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76.h        |  1 +
 drivers/net/wireless/mediatek/mt76/mt7925/main.c |  1 +
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c  | 15 +++++++++++----
 drivers/net/wireless/mediatek/mt76/mt792x.h      | 11 +++++++++++
 4 files changed, 24 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h
index 15f83b5adac7..daf1e43e304e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76.h
@@ -349,6 +349,7 @@ struct mt76_wcid {
 	u8 sta:1;
 	u8 amsdu:1;
 	u8 phy_idx:2;
+	u8 link_id:4;
 
 	u8 rx_check_pn;
 	u8 rx_key_pn[IEEE80211_NUM_TIDS + 1][6];
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index da70c203d618..59b1e920bb4a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -748,6 +748,7 @@ static int mt7925_mac_link_sta_add(struct mt76_dev *mdev,
 	mlink->wcid.phy_idx = mconf->mt76.band_idx;
 	mlink->wcid.tx_info |= MT_WCID_TX_INFO_SET;
 	mlink->last_txs = jiffies;
+	mlink->wcid.link_id = link_sta->link_id;
 
 	ret = mt76_connac_pm_wake(&dev->mphy, &dev->pm);
 	if (ret)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index a67d8dd6075f..ba91c8e37dff 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -967,6 +967,7 @@ mt7925_mcu_sta_key_tlv(struct mt76_wcid *wcid,
 	struct mt792x_sta *msta = container_of(mlink, struct mt792x_sta, deflink);
 	struct sta_rec_sec_uni *sec;
 	struct mt792x_vif *mvif = msta->vif;
+	struct mt792x_bss_conf *mconf = mt792x_vif_to_link(mvif, wcid->link_id);
 	struct ieee80211_sta *sta;
 	struct ieee80211_vif *vif;
 	struct tlv *tlv;
@@ -978,17 +979,23 @@ mt7925_mcu_sta_key_tlv(struct mt76_wcid *wcid,
 
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_KEY_V3, sizeof(*sec));
 	sec = (struct sta_rec_sec_uni *)tlv;
-	sec->bss_idx = mvif->bss_conf.mt76.idx;
+	sec->bss_idx = mconf->mt76.idx;
 	sec->is_authenticator = 0;
-	sec->mgmt_prot = 0;
+	sec->mgmt_prot = 1; /* only used in MLO mode */
 	sec->wlan_idx = (u8)wcid->idx;
 
 	if (sta) {
+		struct ieee80211_link_sta *link_sta;
+
 		sec->tx_key = 1;
 		sec->key_type = 1;
-		memcpy(sec->peer_addr, sta->addr, ETH_ALEN);
+		link_sta = mt792x_sta_to_link_sta(vif, sta, wcid->link_id);
+		memcpy(sec->peer_addr, link_sta->addr, ETH_ALEN);
 	} else {
-		memcpy(sec->peer_addr, vif->bss_conf.bssid, ETH_ALEN);
+		struct ieee80211_bss_conf *link_conf;
+
+		link_conf = mt792x_vif_to_bss_conf(vif, wcid->link_id);
+		memcpy(sec->peer_addr, link_conf->bssid, ETH_ALEN);
 	}
 
 	if (cmd == SET_KEY) {
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index 69eb8dac0b70..fe5a6c932705 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -273,6 +273,17 @@ mt792x_vif_to_bss_conf(struct ieee80211_vif *vif, unsigned int link_id)
 	return link_conf_dereference_protected(vif, link_id);
 }
 
+static inline struct ieee80211_link_sta *
+mt792x_sta_to_link_sta(struct ieee80211_vif *vif, struct ieee80211_sta *sta,
+		       unsigned int link_id)
+{
+	if (!ieee80211_vif_is_mld(vif) ||
+	    link_id >= IEEE80211_LINK_UNSPECIFIED)
+		return &sta->deflink;
+
+	return link_sta_dereference_protected(sta, link_id);
+}
+
 static inline struct mt792x_dev *
 mt792x_hw_dev(struct ieee80211_hw *hw)
 {
-- 
cgit v1.2.3-58-ga151


From 7cf7785a922b920a048fe8edbe9f30d4ea5b9ffe Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:45 -0700
Subject: wifi: mt76: mt7925: add mt7925_set_link_key

add mt7925_set_link_key to set up the key according to the link id

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/e8cd7d37484be238a4eb9e500ef2b8aa46e43667.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c   | 70 ++++++++++++++--------
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    | 25 ++++----
 drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h |  3 +-
 3 files changed, 61 insertions(+), 37 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 59b1e920bb4a..7c7452729710 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -499,31 +499,28 @@ static int mt7925_cancel_remain_on_channel(struct ieee80211_hw *hw,
 	return mt7925_abort_roc(phy, &mvif->bss_conf);
 }
 
-static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
-			  struct ieee80211_vif *vif, struct ieee80211_sta *sta,
-			  struct ieee80211_key_conf *key)
+static int mt7925_set_link_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
+			       struct ieee80211_vif *vif, struct ieee80211_sta *sta,
+			       struct ieee80211_key_conf *key, int link_id)
 {
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_sta *msta = sta ? (struct mt792x_sta *)sta->drv_priv :
 				  &mvif->sta;
-	struct ieee80211_link_sta *link_sta = sta ? &sta->deflink : NULL;
-	struct mt76_wcid *wcid = &msta->deflink.wcid;
 	struct ieee80211_bss_conf *link_conf;
-	u8 *wcid_keyidx = &wcid->hw_key_idx;
+	struct ieee80211_link_sta *link_sta;
 	int idx = key->keyidx, err = 0;
+	struct mt792x_link_sta *mlink;
+	struct mt792x_bss_conf *mconf;
+	struct mt76_wcid *wcid;
+	u8 *wcid_keyidx;
 
-	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
-
-	/* The hardware does not support per-STA RX GTK, fallback
-	 * to software mode for these.
-	 */
-	if ((vif->type == NL80211_IFTYPE_ADHOC ||
-	     vif->type == NL80211_IFTYPE_MESH_POINT) &&
-	    (key->cipher == WLAN_CIPHER_SUITE_TKIP ||
-	     key->cipher == WLAN_CIPHER_SUITE_CCMP) &&
-	    !(key->flags & IEEE80211_KEY_FLAG_PAIRWISE))
-		return -EOPNOTSUPP;
+	link_conf = mt792x_vif_to_bss_conf(vif, link_id);
+	link_sta = sta ? mt792x_sta_to_link_sta(vif, sta, link_id) : NULL;
+	mconf = mt792x_vif_to_link(mvif, link_id);
+	mlink = mt792x_sta_to_link(msta, link_id);
+	wcid = &mlink->wcid;
+	wcid_keyidx = &wcid->hw_key_idx;
 
 	/* fall back to sw encryption for unsupported ciphers */
 	switch (key->cipher) {
@@ -547,13 +544,11 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 		return -EOPNOTSUPP;
 	}
 
-	mt792x_mutex_acquire(dev);
-
-	if (cmd == SET_KEY && !mvif->bss_conf.mt76.cipher) {
+	if (cmd == SET_KEY && !mconf->mt76.cipher) {
 		struct mt792x_phy *phy = mt792x_hw_phy(hw);
 
-		mvif->bss_conf.mt76.cipher = mt7925_mcu_get_cipher(key->cipher);
-		mt7925_mcu_add_bss_info(phy, mvif->bss_conf.mt76.ctx, link_conf,
+		mconf->mt76.cipher = mt7925_mcu_get_cipher(key->cipher);
+		mt7925_mcu_add_bss_info(phy, mconf->mt76.ctx, link_conf,
 					link_sta, true);
 	}
 
@@ -567,9 +562,9 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	mt76_wcid_key_setup(&dev->mt76, wcid,
 			    cmd == SET_KEY ? key : NULL);
 
-	err = mt7925_mcu_add_key(&dev->mt76, vif, &msta->deflink.bip,
+	err = mt7925_mcu_add_key(&dev->mt76, vif, &mlink->bip,
 				 key, MCU_UNI_CMD(STA_REC_UPDATE),
-				 &msta->deflink.wcid, cmd);
+				 &mlink->wcid, cmd, msta);
 
 	if (err)
 		goto out;
@@ -578,9 +573,32 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 	    key->cipher == WLAN_CIPHER_SUITE_WEP40)
 		err = mt7925_mcu_add_key(&dev->mt76, vif, &mvif->wep_sta->deflink.bip,
 					 key, MCU_WMWA_UNI_CMD(STA_REC_UPDATE),
-					 &mvif->wep_sta->deflink.wcid, cmd);
-
+					 &mvif->wep_sta->deflink.wcid, cmd, msta);
 out:
+	return err;
+}
+
+static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
+			  struct ieee80211_vif *vif, struct ieee80211_sta *sta,
+			  struct ieee80211_key_conf *key)
+{
+	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+	int err;
+
+	/* The hardware does not support per-STA RX GTK, fallback
+	 * to software mode for these.
+	 */
+	if ((vif->type == NL80211_IFTYPE_ADHOC ||
+	     vif->type == NL80211_IFTYPE_MESH_POINT) &&
+	    (key->cipher == WLAN_CIPHER_SUITE_TKIP ||
+	     key->cipher == WLAN_CIPHER_SUITE_CCMP) &&
+	    !(key->flags & IEEE80211_KEY_FLAG_PAIRWISE))
+		return -EOPNOTSUPP;
+
+	mt792x_mutex_acquire(dev);
+
+	err = mt7925_set_link_key(hw, cmd, vif, sta, key, vif->bss_conf.link_id);
+
 	mt792x_mutex_release(dev);
 
 	return err;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index ba91c8e37dff..1db81cf4d71d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -961,13 +961,12 @@ mt7925_mcu_sta_key_tlv(struct mt76_wcid *wcid,
 		       struct mt76_connac_sta_key_conf *sta_key_conf,
 		       struct sk_buff *skb,
 		       struct ieee80211_key_conf *key,
-		       enum set_key_cmd cmd)
+		       enum set_key_cmd cmd,
+		       struct mt792x_sta *msta)
 {
-	struct mt792x_link_sta *mlink = container_of(wcid, struct mt792x_link_sta, wcid);
-	struct mt792x_sta *msta = container_of(mlink, struct mt792x_sta, deflink);
-	struct sta_rec_sec_uni *sec;
 	struct mt792x_vif *mvif = msta->vif;
 	struct mt792x_bss_conf *mconf = mt792x_vif_to_link(mvif, wcid->link_id);
+	struct sta_rec_sec_uni *sec;
 	struct ieee80211_sta *sta;
 	struct ieee80211_vif *vif;
 	struct tlv *tlv;
@@ -990,12 +989,16 @@ mt7925_mcu_sta_key_tlv(struct mt76_wcid *wcid,
 		sec->tx_key = 1;
 		sec->key_type = 1;
 		link_sta = mt792x_sta_to_link_sta(vif, sta, wcid->link_id);
-		memcpy(sec->peer_addr, link_sta->addr, ETH_ALEN);
+
+		if (link_sta)
+			memcpy(sec->peer_addr, link_sta->addr, ETH_ALEN);
 	} else {
 		struct ieee80211_bss_conf *link_conf;
 
 		link_conf = mt792x_vif_to_bss_conf(vif, wcid->link_id);
-		memcpy(sec->peer_addr, link_conf->bssid, ETH_ALEN);
+
+		if (link_conf)
+			memcpy(sec->peer_addr, link_conf->bssid, ETH_ALEN);
 	}
 
 	if (cmd == SET_KEY) {
@@ -1040,18 +1043,20 @@ mt7925_mcu_sta_key_tlv(struct mt76_wcid *wcid,
 int mt7925_mcu_add_key(struct mt76_dev *dev, struct ieee80211_vif *vif,
 		       struct mt76_connac_sta_key_conf *sta_key_conf,
 		       struct ieee80211_key_conf *key, int mcu_cmd,
-		       struct mt76_wcid *wcid, enum set_key_cmd cmd)
+		       struct mt76_wcid *wcid, enum set_key_cmd cmd,
+		       struct mt792x_sta *msta)
 {
-	struct mt76_vif *mvif = (struct mt76_vif *)vif->drv_priv;
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_bss_conf *mconf = mt792x_vif_to_link(mvif, wcid->link_id);
 	struct sk_buff *skb;
 	int ret;
 
-	skb = __mt76_connac_mcu_alloc_sta_req(dev, mvif, wcid,
+	skb = __mt76_connac_mcu_alloc_sta_req(dev, &mconf->mt76, wcid,
 					      MT7925_STA_UPDATE_MAX_SIZE);
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
 
-	ret = mt7925_mcu_sta_key_tlv(wcid, sta_key_conf, skb, key, cmd);
+	ret = mt7925_mcu_sta_key_tlv(wcid, sta_key_conf, skb, key, cmd, msta);
 	if (ret)
 		return ret;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
index 8ec5a3a20202..7750049c42f2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
@@ -304,7 +304,8 @@ int mt7925_mcu_fill_message(struct mt76_dev *mdev, struct sk_buff *skb,
 int mt7925_mcu_add_key(struct mt76_dev *dev, struct ieee80211_vif *vif,
 		       struct mt76_connac_sta_key_conf *sta_key_conf,
 		       struct ieee80211_key_conf *key, int mcu_cmd,
-		       struct mt76_wcid *wcid, enum set_key_cmd cmd);
+		       struct mt76_wcid *wcid, enum set_key_cmd cmd,
+		       struct mt792x_sta *msta);
 int mt7925_mcu_set_rts_thresh(struct mt792x_phy *phy, u32 val);
 int mt7925_mcu_wtbl_update_hdr_trans(struct mt792x_dev *dev,
 				     struct ieee80211_vif *vif,
-- 
cgit v1.2.3-58-ga151


From 49681949b99a3a4bb1eae08e139752b2ffa05df2 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:46 -0700
Subject: wifi: mt76: mt7925: extend mt7925_mcu_uni_roc_event

We extended the function to be able multiple pieces of information
in a single event, supporting the future MLO-enabled firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/4be7b34904034d521dab147883b898b5402fd29f.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    | 31 ++++++++++++++++++----
 drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h |  1 +
 2 files changed, 27 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 1db81cf4d71d..cf4b46ebfc92 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -313,16 +313,14 @@ mt7925_mcu_roc_iter(void *priv, u8 *mac, struct ieee80211_vif *vif)
 	mvif->band_idx = grant->dbdcband;
 }
 
-static void
-mt7925_mcu_uni_roc_event(struct mt792x_dev *dev, struct sk_buff *skb)
+static void mt7925_mcu_roc_handle_grant(struct mt792x_dev *dev,
+					struct tlv *tlv)
 {
 	struct ieee80211_hw *hw = dev->mt76.hw;
 	struct mt7925_roc_grant_tlv *grant;
-	struct mt7925_mcu_rxd *rxd;
 	int duration;
 
-	rxd = (struct mt7925_mcu_rxd *)skb->data;
-	grant = (struct mt7925_roc_grant_tlv *)(rxd->tlv + 4);
+	grant = (struct mt7925_roc_grant_tlv *)tlv;
 
 	/* should never happen */
 	WARN_ON_ONCE((le16_to_cpu(grant->tag) != UNI_EVENT_ROC_GRANT));
@@ -340,6 +338,29 @@ mt7925_mcu_uni_roc_event(struct mt792x_dev *dev, struct sk_buff *skb)
 		  jiffies + msecs_to_jiffies(duration));
 }
 
+static void
+mt7925_mcu_uni_roc_event(struct mt792x_dev *dev, struct sk_buff *skb)
+{
+	struct tlv *tlv;
+	int i = 0;
+
+	skb_pull(skb, sizeof(struct mt7925_mcu_rxd) + 4);
+
+	while (i < skb->len) {
+		tlv = (struct tlv *)(skb->data + i);
+
+		switch (le16_to_cpu(tlv->tag)) {
+		case UNI_EVENT_ROC_GRANT:
+			mt7925_mcu_roc_handle_grant(dev, tlv);
+			break;
+		case UNI_EVENT_ROC_GRANT_SUB_LINK:
+			break;
+		}
+
+		i += le16_to_cpu(tlv->len);
+	}
+}
+
 static void
 mt7925_mcu_scan_event(struct mt792x_dev *dev, struct sk_buff *skb)
 {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
index 7750049c42f2..b7f8802b9cd6 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
@@ -41,6 +41,7 @@ enum mt7925_roc_req {
 
 enum {
 	UNI_EVENT_ROC_GRANT = 0,
+	UNI_EVENT_ROC_GRANT_SUB_LINK = 4,
 	UNI_EVENT_ROC_TAG_NUM
 };
 
-- 
cgit v1.2.3-58-ga151


From 69acd6d910b0c83842bd45c36224d4f8fe59d1d4 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:47 -0700
Subject: wifi: mt76: mt7925: add mt7925_change_vif_links

Add mt7925_change_vif_links to change the valid links on an interface,
supporting the MLO-enabled firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/31c9de30c3a18592c8e100f909845383b66f8677.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c   | 130 +++++++++++++++++++++
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    | 117 ++++++++++++++++---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.h    |  20 ++++
 drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h |   6 +
 drivers/net/wireless/mediatek/mt76/mt792x.h        |   5 +
 drivers/net/wireless/mediatek/mt76/mt792x_core.c   |  23 ++--
 6 files changed, 275 insertions(+), 26 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 7c7452729710..fcf7bd1a06bc 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -472,6 +472,33 @@ out:
 	return err;
 }
 
+static int mt7925_set_mlo_roc(struct mt792x_phy *phy,
+			      struct mt792x_bss_conf *mconf,
+			      u16 sel_links)
+{
+	int err;
+
+	if (WARN_ON_ONCE(test_and_set_bit(MT76_STATE_ROC, &phy->mt76->state)))
+		return -EBUSY;
+
+	phy->roc_grant = false;
+
+	err = mt7925_mcu_set_mlo_roc(mconf, sel_links, 5, ++phy->roc_token_id);
+	if (err < 0) {
+		clear_bit(MT76_STATE_ROC, &phy->mt76->state);
+		goto out;
+	}
+
+	if (!wait_event_timeout(phy->roc_wait, phy->roc_grant, 4 * HZ)) {
+		mt7925_mcu_abort_roc(phy, mconf, phy->roc_token_id);
+		clear_bit(MT76_STATE_ROC, &phy->mt76->state);
+		err = -ETIMEDOUT;
+	}
+
+out:
+	return err;
+}
+
 static int mt7925_remain_on_channel(struct ieee80211_hw *hw,
 				    struct ieee80211_vif *vif,
 				    struct ieee80211_channel *chan,
@@ -1521,6 +1548,108 @@ static void mt7925_link_info_changed(struct ieee80211_hw *hw,
 	mt792x_mutex_release(dev);
 }
 
+static int
+mt7925_change_vif_links(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			u16 old_links, u16 new_links,
+			struct ieee80211_bss_conf *old[IEEE80211_MLD_MAX_NUM_LINKS])
+{
+	struct mt792x_bss_conf *mconfs[IEEE80211_MLD_MAX_NUM_LINKS] = {}, *mconf;
+	struct mt792x_link_sta *mlinks[IEEE80211_MLD_MAX_NUM_LINKS] = {}, *mlink;
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	unsigned long add = new_links & ~old_links;
+	unsigned long rem = old_links & ~new_links;
+	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+	struct mt792x_phy *phy = mt792x_hw_phy(hw);
+	struct ieee80211_bss_conf *link_conf;
+	unsigned int link_id;
+	int err;
+
+	if (old_links == new_links)
+		return 0;
+
+	mt792x_mutex_acquire(dev);
+
+	for_each_set_bit(link_id, &rem, IEEE80211_MLD_MAX_NUM_LINKS) {
+		mconf = mt792x_vif_to_link(mvif, link_id);
+		mlink = mt792x_sta_to_link(&mvif->sta, link_id);
+
+		if (!mconf || !mlink)
+			continue;
+
+		if (mconf != &mvif->bss_conf) {
+			mt792x_mac_link_bss_remove(dev, mconf, mlink);
+			devm_kfree(dev->mt76.dev, mconf);
+			devm_kfree(dev->mt76.dev, mlink);
+		}
+
+		rcu_assign_pointer(mvif->link_conf[link_id], NULL);
+		rcu_assign_pointer(mvif->sta.link[link_id], NULL);
+	}
+
+	for_each_set_bit(link_id, &add, IEEE80211_MLD_MAX_NUM_LINKS) {
+		if (!old_links) {
+			mconf = &mvif->bss_conf;
+			mlink = &mvif->sta.deflink;
+		} else {
+			mconf = devm_kzalloc(dev->mt76.dev, sizeof(*mconf),
+					     GFP_KERNEL);
+			mlink = devm_kzalloc(dev->mt76.dev, sizeof(*mlink),
+					     GFP_KERNEL);
+		}
+
+		mconfs[link_id] = mconf;
+		mlinks[link_id] = mlink;
+		mconf->link_id = link_id;
+		mconf->vif = mvif;
+		mlink->wcid.link_id = link_id;
+	}
+
+	if (hweight16(mvif->valid_links) == 0)
+		mt792x_mac_link_bss_remove(dev, &mvif->bss_conf,
+					   &mvif->sta.deflink);
+
+	for_each_set_bit(link_id, &add, IEEE80211_MLD_MAX_NUM_LINKS) {
+		mconf = mconfs[link_id];
+		mlink = mlinks[link_id];
+		link_conf = mt792x_vif_to_bss_conf(vif, link_id);
+
+		rcu_assign_pointer(mvif->link_conf[link_id], mconf);
+		rcu_assign_pointer(mvif->sta.link[link_id], mlink);
+
+		err = mt7925_mac_link_bss_add(dev, link_conf, mlink);
+		if (err < 0)
+			goto free;
+
+		if (mconf != &mvif->bss_conf) {
+			err = mt7925_set_mlo_roc(phy, &mvif->bss_conf,
+						 vif->active_links);
+			if (err < 0)
+				goto free;
+		}
+	}
+
+	mvif->valid_links = new_links;
+
+	mt792x_mutex_release(dev);
+
+	return 0;
+
+free:
+	for_each_set_bit(link_id, &add, IEEE80211_MLD_MAX_NUM_LINKS) {
+		rcu_assign_pointer(mvif->link_conf[link_id], NULL);
+		rcu_assign_pointer(mvif->sta.link[link_id], NULL);
+
+		if (mconf != &mvif->bss_conf)
+			devm_kfree(dev->mt76.dev, mconfs[link_id]);
+		if (mlink != &mvif->sta.deflink)
+			devm_kfree(dev->mt76.dev, mlinks[link_id]);
+	}
+
+	mt792x_mutex_release(dev);
+
+	return err;
+}
+
 const struct ieee80211_ops mt7925_ops = {
 	.tx = mt792x_tx,
 	.start = mt7925_start,
@@ -1579,6 +1708,7 @@ const struct ieee80211_ops mt7925_ops = {
 	.mgd_complete_tx = mt7925_mgd_complete_tx,
 	.vif_cfg_changed = mt7925_vif_cfg_changed,
 	.link_info_changed = mt7925_link_info_changed,
+	.change_vif_links = mt7925_change_vif_links,
 };
 EXPORT_SYMBOL_GPL(mt7925_ops);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index cf4b46ebfc92..d87c23911119 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -307,6 +307,9 @@ mt7925_mcu_roc_iter(void *priv, u8 *mac, struct ieee80211_vif *vif)
 	struct mt76_vif *mvif = (struct mt76_vif *)vif->drv_priv;
 	struct mt7925_roc_grant_tlv *grant = priv;
 
+	if (ieee80211_vif_is_mld(vif) && vif->type == NL80211_IFTYPE_STATION)
+		return;
+
 	if (mvif->idx != grant->bss_idx)
 		return;
 
@@ -1084,6 +1087,100 @@ int mt7925_mcu_add_key(struct mt76_dev *dev, struct ieee80211_vif *vif,
 	return mt76_mcu_skb_send_msg(dev, skb, mcu_cmd, true);
 }
 
+int mt7925_mcu_set_mlo_roc(struct mt792x_bss_conf *mconf, u16 sel_links,
+			   int duration, u8 token_id)
+{
+	struct mt792x_vif *mvif = mconf->vif;
+	struct ieee80211_vif *vif = container_of((void *)mvif,
+						 struct ieee80211_vif, drv_priv);
+	struct ieee80211_bss_conf *link_conf;
+	struct ieee80211_channel *chan;
+	const u8 ch_band[] = {
+		[NL80211_BAND_2GHZ] = 1,
+		[NL80211_BAND_5GHZ] = 2,
+		[NL80211_BAND_6GHZ] = 3,
+	};
+	enum mt7925_roc_req type;
+	int center_ch, i = 0;
+	bool is_AG_band = false;
+	struct {
+		u8 id;
+		u8 bss_idx;
+		u16 tag;
+		struct mt792x_bss_conf *mconf;
+		struct ieee80211_channel *chan;
+	} links[2];
+
+	struct {
+		struct {
+			u8 rsv[4];
+		} __packed hdr;
+		struct roc_acquire_tlv roc[2];
+	} __packed req;
+
+	if (!mconf || hweight16(vif->valid_links) < 2 ||
+	    hweight16(sel_links) != 2)
+		return -EPERM;
+
+	for (i = 0; i < ARRAY_SIZE(links); i++) {
+		links[i].id = i ? __ffs(~BIT(mconf->link_id) & sel_links) :
+				 mconf->link_id;
+		link_conf = mt792x_vif_to_bss_conf(vif, links[i].id);
+		if (WARN_ON_ONCE(!link_conf))
+			return -EPERM;
+
+		links[i].chan = link_conf->chanreq.oper.chan;
+		if (WARN_ON_ONCE(!links[i].chan))
+			return -EPERM;
+
+		links[i].mconf = mt792x_vif_to_link(mvif, links[i].id);
+		links[i].tag = links[i].id == mconf->link_id ?
+			       UNI_ROC_ACQUIRE : UNI_ROC_SUB_LINK;
+
+		is_AG_band |= links[i].chan->band == NL80211_BAND_2GHZ;
+	}
+
+	if (vif->cfg.eml_cap & IEEE80211_EML_CAP_EMLSR_SUPP)
+		type = is_AG_band ? MT7925_ROC_REQ_MLSR_AG :
+				    MT7925_ROC_REQ_MLSR_AA;
+	else
+		type = MT7925_ROC_REQ_JOIN;
+
+	for (i = 0; i < ARRAY_SIZE(links) && i < hweight16(vif->active_links); i++) {
+		if (WARN_ON_ONCE(!links[i].mconf || !links[i].chan))
+			continue;
+
+		chan = links[i].chan;
+		center_ch = ieee80211_frequency_to_channel(chan->center_freq);
+		req.roc[i].len = cpu_to_le16(sizeof(struct roc_acquire_tlv));
+		req.roc[i].tag = cpu_to_le16(links[i].tag);
+		req.roc[i].tokenid = token_id;
+		req.roc[i].reqtype = type;
+		req.roc[i].maxinterval = cpu_to_le32(duration);
+		req.roc[i].bss_idx = links[i].mconf->mt76.idx;
+		req.roc[i].control_channel = chan->hw_value;
+		req.roc[i].bw = CMD_CBW_20MHZ;
+		req.roc[i].bw_from_ap = CMD_CBW_20MHZ;
+		req.roc[i].center_chan = center_ch;
+		req.roc[i].center_chan_from_ap = center_ch;
+
+		/* STR : 0xfe indicates BAND_ALL with enabling DBDC
+		 * EMLSR : 0xff indicates (BAND_AUTO) without DBDC
+		 */
+		req.roc[i].dbdcband = type == MT7925_ROC_REQ_JOIN ? 0xfe : 0xff;
+
+		if (chan->hw_value < center_ch)
+			req.roc[i].sco = 1; /* SCA */
+		else if (chan->hw_value > center_ch)
+			req.roc[i].sco = 3; /* SCB */
+
+		req.roc[i].band = ch_band[chan->band];
+	}
+
+	return mt76_mcu_send_msg(&mvif->phy->dev->mt76, MCU_UNI_CMD(ROC),
+				 &req, sizeof(req), false);
+}
+
 int mt7925_mcu_set_roc(struct mt792x_phy *phy, struct mt792x_bss_conf *mconf,
 		       struct ieee80211_channel *chan, int duration,
 		       enum mt7925_roc_req type, u8 token_id)
@@ -1094,25 +1191,7 @@ int mt7925_mcu_set_roc(struct mt792x_phy *phy, struct mt792x_bss_conf *mconf,
 		struct {
 			u8 rsv[4];
 		} __packed hdr;
-		struct roc_acquire_tlv {
-			__le16 tag;
-			__le16 len;
-			u8 bss_idx;
-			u8 tokenid;
-			u8 control_channel;
-			u8 sco;
-			u8 band;
-			u8 bw;
-			u8 center_chan;
-			u8 center_chan2;
-			u8 bw_from_ap;
-			u8 center_chan_from_ap;
-			u8 center_chan2_from_ap;
-			u8 reqtype;
-			__le32 maxinterval;
-			u8 dbdcband;
-			u8 rsv[3];
-		} __packed roc;
+		struct roc_acquire_tlv roc;
 	} __packed req = {
 		.roc = {
 			.tag = cpu_to_le16(UNI_ROC_ACQUIRE),
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
index da4e84f468d2..fefb6ab384bb 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
@@ -554,6 +554,26 @@ struct mt7925_wow_pattern_tlv {
 	u8 rsv[7];
 } __packed;
 
+struct roc_acquire_tlv {
+	__le16 tag;
+	__le16 len;
+	u8 bss_idx;
+	u8 tokenid;
+	u8 control_channel;
+	u8 sco;
+	u8 band;
+	u8 bw;
+	u8 center_chan;
+	u8 center_chan2;
+	u8 bw_from_ap;
+	u8 center_chan_from_ap;
+	u8 center_chan2_from_ap;
+	u8 reqtype;
+	__le32 maxinterval;
+	u8 dbdcband;
+	u8 rsv[3];
+} __packed;
+
 static inline enum connac3_mcu_cipher_type
 mt7925_mcu_get_cipher(int cipher)
 {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
index b7f8802b9cd6..dbc9e3d426e9 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
@@ -30,12 +30,16 @@
 enum {
 	UNI_ROC_ACQUIRE,
 	UNI_ROC_ABORT,
+	UNI_ROC_SUB_LINK = 3,
 	UNI_ROC_NUM
 };
 
 enum mt7925_roc_req {
 	MT7925_ROC_REQ_JOIN,
 	MT7925_ROC_REQ_ROC,
+	MT7925_ROC_REQ_SUB_LINK,
+	MT7925_ROC_REQ_MLSR_AG = 10,
+	MT7925_ROC_REQ_MLSR_AA,
 	MT7925_ROC_REQ_NUM
 };
 
@@ -295,6 +299,8 @@ int mt7925_set_tx_sar_pwr(struct ieee80211_hw *hw,
 int mt7925_mcu_regval(struct mt792x_dev *dev, u32 regidx, u32 *val, bool set);
 int mt7925_mcu_set_clc(struct mt792x_dev *dev, u8 *alpha2,
 		       enum environment_cap env_cap);
+int mt7925_mcu_set_mlo_roc(struct mt792x_bss_conf *mconf, u16 sel_links,
+			   int duration, u8 token_id);
 int mt7925_mcu_set_roc(struct mt792x_phy *phy, struct mt792x_bss_conf *mconf,
 		       struct ieee80211_channel *chan, int duration,
 		       enum mt7925_roc_req type, u8 token_id);
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index fe5a6c932705..9a7443a8a951 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -112,6 +112,7 @@ struct mt792x_bss_conf {
 	struct mt792x_vif *vif;
 	struct ewma_rssi rssi;
 	struct ieee80211_tx_queue_params queue_params[IEEE80211_NUM_ACS];
+	unsigned int link_id;
 };
 
 struct mt792x_vif {
@@ -122,6 +123,7 @@ struct mt792x_vif {
 	struct mt792x_sta *wep_sta;
 
 	struct mt792x_phy *phy;
+	u16 valid_links;
 };
 
 struct mt792x_phy {
@@ -398,6 +400,9 @@ mt792x_get_mac80211_ops(struct device *dev,
 int mt792x_init_wcid(struct mt792x_dev *dev);
 int mt792x_mcu_drv_pmctrl(struct mt792x_dev *dev);
 int mt792x_mcu_fw_pmctrl(struct mt792x_dev *dev);
+void mt792x_mac_link_bss_remove(struct mt792x_dev *dev,
+				struct mt792x_bss_conf *mconf,
+				struct mt792x_link_sta *mlink);
 
 static inline char *mt792x_ram_name(struct mt792x_dev *dev)
 {
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_core.c b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
index 813296fad0ed..62f471b5498f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_core.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
@@ -113,14 +113,17 @@ void mt792x_stop(struct ieee80211_hw *hw, bool suspend)
 }
 EXPORT_SYMBOL_GPL(mt792x_stop);
 
-static void mt792x_mac_link_bss_remove(struct mt792x_dev *dev,
-				       struct ieee80211_bss_conf *link_conf,
-				       struct mt792x_link_sta *mlink)
+void mt792x_mac_link_bss_remove(struct mt792x_dev *dev,
+				struct mt792x_bss_conf *mconf,
+				struct mt792x_link_sta *mlink)
 {
-	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
+	struct ieee80211_vif *vif = container_of((void *)mconf->vif,
+						 struct ieee80211_vif, drv_priv);
+	struct ieee80211_bss_conf *link_conf;
 	int idx = mlink->wcid.idx;
 
-	mt792x_mutex_acquire(dev);
+	link_conf = mt792x_vif_to_bss_conf(vif, mconf->link_id);
+
 	mt76_connac_free_pending_tx_skbs(&dev->pm, &mlink->wcid);
 	mt76_connac_mcu_uni_add_dev(&dev->mphy, link_conf, &mlink->wcid, false);
 
@@ -128,7 +131,6 @@ static void mt792x_mac_link_bss_remove(struct mt792x_dev *dev,
 
 	dev->mt76.vif_mask &= ~BIT_ULL(mconf->mt76.idx);
 	mconf->vif->phy->omac_mask &= ~BIT_ULL(mconf->mt76.omac_idx);
-	mt792x_mutex_release(dev);
 
 	spin_lock_bh(&dev->mt76.sta_poll_lock);
 	if (!list_empty(&mlink->wcid.poll_list))
@@ -137,14 +139,21 @@ static void mt792x_mac_link_bss_remove(struct mt792x_dev *dev,
 
 	mt76_wcid_cleanup(&dev->mt76, &mlink->wcid);
 }
+EXPORT_SYMBOL_GPL(mt792x_mac_link_bss_remove);
 
 void mt792x_remove_interface(struct ieee80211_hw *hw,
 			     struct ieee80211_vif *vif)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+	struct mt792x_bss_conf *mconf;
 
-	mt792x_mac_link_bss_remove(dev, &vif->bss_conf, &mvif->sta.deflink);
+	mt792x_mutex_acquire(dev);
+
+	mconf = mt792x_link_conf_to_mconf(&vif->bss_conf);
+	mt792x_mac_link_bss_remove(dev, mconf, &mvif->sta.deflink);
+
+	mt792x_mutex_release(dev);
 }
 EXPORT_SYMBOL_GPL(mt792x_remove_interface);
 
-- 
cgit v1.2.3-58-ga151


From 43aa4033eb18f46332860db061b135bd9e51be2c Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:48 -0700
Subject: wifi: mt76: mt7925: add mt7925_change_sta_links

add mt7925_change_sta_links to change the valid links of a station,
supporting the MLO-enabled firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/eeedec88f0576315791a8b1b453464173a4addbe.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 142 ++++++++++++++++++++++-
 drivers/net/wireless/mediatek/mt76/mt792x.h      |   7 ++
 2 files changed, 145 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index fcf7bd1a06bc..afb58a9206b0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -809,16 +809,79 @@ static int mt7925_mac_link_sta_add(struct mt76_dev *mdev,
 		mt7925_mcu_add_bss_info(&dev->phy, mconf->mt76.ctx,
 					link_conf, link_sta, false);
 
-	ret = mt7925_mcu_sta_update(dev, link_sta, vif, true,
-				    MT76_STA_INFO_STATE_NONE);
-	if (ret)
-		return ret;
+	if (ieee80211_vif_is_mld(vif) &&
+	    link_sta == mlink->pri_link) {
+		ret = mt7925_mcu_sta_update(dev, link_sta, vif, true,
+					    MT76_STA_INFO_STATE_NONE);
+		if (ret)
+			return ret;
+	} else if (ieee80211_vif_is_mld(vif) &&
+		   link_sta != mlink->pri_link) {
+		ret = mt7925_mcu_sta_update(dev, mlink->pri_link, vif,
+					    true, MT76_STA_INFO_STATE_ASSOC);
+		if (ret)
+			return ret;
+
+		ret = mt7925_mcu_sta_update(dev, link_sta, vif, true,
+					    MT76_STA_INFO_STATE_ASSOC);
+		if (ret)
+			return ret;
+	} else {
+		ret = mt7925_mcu_sta_update(dev, link_sta, vif, true,
+					    MT76_STA_INFO_STATE_NONE);
+		if (ret)
+			return ret;
+	}
 
 	mt76_connac_power_save_sched(&dev->mphy, &dev->pm);
 
 	return 0;
 }
 
+static int
+mt7925_mac_sta_add_links(struct mt792x_dev *dev, struct ieee80211_vif *vif,
+			 struct ieee80211_sta *sta, unsigned long new_links)
+{
+	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
+	struct mt76_wcid *wcid;
+	unsigned int link_id;
+	int err = 0;
+
+	for_each_set_bit(link_id, &new_links, IEEE80211_MLD_MAX_NUM_LINKS) {
+		struct ieee80211_link_sta *link_sta;
+		struct mt792x_link_sta *mlink;
+
+		if (msta->deflink_id == IEEE80211_LINK_UNSPECIFIED) {
+			mlink = &msta->deflink;
+			msta->deflink_id = link_id;
+		} else {
+			mlink = devm_kzalloc(dev->mt76.dev, sizeof(*mlink), GFP_KERNEL);
+			if (!mlink) {
+				err = -ENOMEM;
+				break;
+			}
+
+			wcid = &mlink->wcid;
+			ewma_signal_init(&wcid->rssi);
+			rcu_assign_pointer(dev->mt76.wcid[wcid->idx], wcid);
+			mt76_wcid_init(wcid);
+			ewma_avg_signal_init(&mlink->avg_ack_signal);
+			memset(mlink->airtime_ac, 0,
+			       sizeof(msta->deflink.airtime_ac));
+		}
+
+		msta->valid_links |= BIT(link_id);
+		rcu_assign_pointer(msta->link[link_id], mlink);
+		mlink->sta = msta;
+		mlink->pri_link = &sta->deflink;
+
+		link_sta = mt792x_sta_to_link_sta(vif, sta, link_id);
+		mt7925_mac_link_sta_add(&dev->mt76, vif, link_sta);
+	}
+
+	return err;
+}
+
 int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta)
 {
@@ -915,6 +978,48 @@ static void mt7925_mac_link_sta_remove(struct mt76_dev *mdev,
 	mt76_connac_power_save_sched(&dev->mphy, &dev->pm);
 }
 
+static int
+mt7925_mac_sta_remove_links(struct mt792x_dev *dev, struct ieee80211_vif *vif,
+			    struct ieee80211_sta *sta, unsigned long old_links)
+{
+	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
+	struct mt76_dev *mdev = &dev->mt76;
+	struct mt76_wcid *wcid;
+	unsigned int link_id;
+
+	for_each_set_bit(link_id, &old_links, IEEE80211_MLD_MAX_NUM_LINKS) {
+		struct ieee80211_link_sta *link_sta;
+		struct mt792x_link_sta *mlink;
+
+		link_sta = mt792x_sta_to_link_sta(vif, sta, link_id);
+		if (!link_sta)
+			continue;
+
+		mlink = mt792x_sta_to_link(msta, link_id);
+		if (!mlink)
+			continue;
+
+		mt7925_mac_link_sta_remove(&dev->mt76, vif, link_sta);
+
+		wcid = &mlink->wcid;
+		rcu_assign_pointer(msta->link[link_id], NULL);
+		msta->valid_links &= ~BIT(link_id);
+		mlink->sta = NULL;
+		mlink->pri_link = NULL;
+
+		if (link_sta != mlink->pri_link) {
+			mt76_wcid_cleanup(mdev, wcid);
+			mt76_wcid_mask_clear(mdev->wcid_mask, wcid->idx);
+			mt76_wcid_mask_clear(mdev->wcid_phy_mask, wcid->idx);
+		}
+
+		if (msta->deflink_id == link_id)
+			msta->deflink_id = IEEE80211_LINK_UNSPECIFIED;
+	}
+
+	return 0;
+}
+
 void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 			   struct ieee80211_sta *sta)
 {
@@ -1650,6 +1755,34 @@ free:
 	return err;
 }
 
+static int
+mt7925_change_sta_links(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
+			struct ieee80211_sta *sta, u16 old_links, u16 new_links)
+{
+	unsigned long add = new_links & ~old_links;
+	unsigned long rem = old_links & ~new_links;
+	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+	int err = 0;
+
+	if (old_links == new_links)
+		return 0;
+
+	mt792x_mutex_acquire(dev);
+
+	err = mt7925_mac_sta_remove_links(dev, vif, sta, rem);
+	if (err < 0)
+		goto out;
+
+	err = mt7925_mac_sta_add_links(dev, vif, sta, add);
+	if (err < 0)
+		goto out;
+
+out:
+	mt792x_mutex_release(dev);
+
+	return err;
+}
+
 const struct ieee80211_ops mt7925_ops = {
 	.tx = mt792x_tx,
 	.start = mt7925_start,
@@ -1709,6 +1842,7 @@ const struct ieee80211_ops mt7925_ops = {
 	.vif_cfg_changed = mt7925_vif_cfg_changed,
 	.link_info_changed = mt7925_link_info_changed,
 	.change_vif_links = mt7925_change_vif_links,
+	.change_sta_links = mt7925_change_sta_links,
 };
 EXPORT_SYMBOL_GPL(mt7925_ops);
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index 9a7443a8a951..af9a103dc7f4 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -92,6 +92,10 @@ struct mt792x_link_sta {
 	unsigned long last_txs;
 
 	struct mt76_connac_sta_key_conf bip;
+
+	struct mt792x_sta *sta;
+
+	struct ieee80211_link_sta *pri_link;
 };
 
 struct mt792x_sta {
@@ -99,6 +103,9 @@ struct mt792x_sta {
 	struct mt792x_link_sta __rcu *link[IEEE80211_MLD_MAX_NUM_LINKS];
 
 	struct mt792x_vif *vif;
+
+	u16 valid_links;
+	u8 deflink_id;
 };
 
 DECLARE_EWMA(rssi, 10, 8);
-- 
cgit v1.2.3-58-ga151


From 61c92cc3adcfd84d76dfe3f5889206ab968055e2 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:49 -0700
Subject: wifi: mt76: mt7925: add link handling in mt7925_mac_sta_add

Added link handling to mt7925_mac_sta_add to support MLD devices.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/632faefe8c82cd5609c2af00aabee15ccc9616ed.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index afb58a9206b0..042c9fe5a2d9 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -885,15 +885,25 @@ mt7925_mac_sta_add_links(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 		       struct ieee80211_sta *sta)
 {
+	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
+	int err;
 
 	msta->vif = mvif;
 
 	if (vif->type == NL80211_IFTYPE_STATION)
 		mvif->wep_sta = msta;
 
-	return mt7925_mac_link_sta_add(mdev, vif, &sta->deflink);
+	if (ieee80211_vif_is_mld(vif)) {
+		msta->deflink_id = IEEE80211_LINK_UNSPECIFIED;
+
+		err = mt7925_mac_sta_add_links(dev, vif, sta, sta->valid_links);
+	} else {
+		err = mt7925_mac_link_sta_add(mdev, vif, &sta->deflink);
+	}
+
+	return err;
 }
 EXPORT_SYMBOL_GPL(mt7925_mac_sta_add);
 
-- 
cgit v1.2.3-58-ga151


From 39d80be9951434755566272b401c1619eab37d4d Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:50 -0700
Subject: wifi: mt76: mt7925: add link handling in mt7925_mac_sta_remove

Added link handling to mt7925_mac_sta_remove, supporting the MLO-enabled
firmware. The change remains comptabile with the non-MLO mode.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20d53511fe1f64984d893f4d3ec87cb9f87d3070.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 042c9fe5a2d9..79b43101237c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1033,7 +1033,13 @@ mt7925_mac_sta_remove_links(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 void mt7925_mac_sta_remove(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 			   struct ieee80211_sta *sta)
 {
-	mt7925_mac_link_sta_remove(mdev, vif, &sta->deflink);
+	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
+	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
+	unsigned long rem;
+
+	rem = ieee80211_vif_is_mld(vif) ? msta->valid_links : BIT(0);
+
+	mt7925_mac_sta_remove_links(dev, vif, sta, rem);
 
 	if (vif->type == NL80211_IFTYPE_STATION) {
 		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-- 
cgit v1.2.3-58-ga151


From ebb1406813c6ea60d2d95efc69ce6d53bbe43b31 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:51 -0700
Subject: wifi: mt76: mt7925: add link handling to txwi

add link handling to mt792x_tx and writing txwi.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/c0a7c0b33ea2640bcb4f13a5f79a8976b73fa8f2.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mac.c  | 12 ++++++++---
 drivers/net/wireless/mediatek/mt76/mt792x_core.c | 26 ++++++++++++++++++++++--
 2 files changed, 33 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
index 2be2ab914411..6f5f2130420e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
@@ -740,8 +740,12 @@ mt7925_mac_write_txwi(struct mt76_dev *dev, __le32 *txwi,
 				    BSS_CHANGED_BEACON_ENABLED));
 	bool inband_disc = !!(changed & (BSS_CHANGED_UNSOL_BCAST_PROBE_RESP |
 					 BSS_CHANGED_FILS_DISCOVERY));
+	struct mt792x_bss_conf *mconf;
+
+	mconf = vif ? mt792x_vif_to_link((struct mt792x_vif *)vif->drv_priv,
+					 wcid->link_id) : NULL;
+	mvif = mconf ? (struct mt76_vif *)&mconf->mt76 : NULL;
 
-	mvif = vif ? (struct mt76_vif *)vif->drv_priv : NULL;
 	if (mvif) {
 		omac_idx = mvif->omac_idx;
 		wmm_idx = mvif->wmm_idx;
@@ -802,8 +806,10 @@ mt7925_mac_write_txwi(struct mt76_dev *dev, __le32 *txwi,
 
 	txwi[5] = cpu_to_le32(val);
 
-	val = MT_TXD6_DIS_MAT | MT_TXD6_DAS |
-	      FIELD_PREP(MT_TXD6_MSDU_CNT, 1);
+	val = MT_TXD6_DAS | FIELD_PREP(MT_TXD6_MSDU_CNT, 1);
+	if (!ieee80211_vif_is_mld(vif) ||
+	    (q_idx >= MT_LMAC_ALTX0 && q_idx <= MT_LMAC_BCN0))
+		val |= MT_TXD6_DIS_MAT;
 	txwi[6] = cpu_to_le32(val);
 	txwi[7] = 0;
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_core.c b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
index 62f471b5498f..ae2ac146a7f7 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_core.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
@@ -59,13 +59,17 @@ void mt792x_tx(struct ieee80211_hw *hw, struct ieee80211_tx_control *control,
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct ieee80211_vif *vif = info->control.vif;
 	struct mt76_wcid *wcid = &dev->mt76.global_wcid;
+	u8 link_id;
 	int qid;
 
 	if (control->sta) {
+		struct mt792x_link_sta *mlink;
 		struct mt792x_sta *sta;
-
+		link_id = u32_get_bits(info->control.flags,
+				       IEEE80211_TX_CTRL_MLO_LINK);
 		sta = (struct mt792x_sta *)control->sta->drv_priv;
-		wcid = &sta->deflink.wcid;
+		mlink = mt792x_sta_to_link(sta, link_id);
+		wcid = &mlink->wcid;
 	}
 
 	if (vif && !control->sta) {
@@ -75,6 +79,24 @@ void mt792x_tx(struct ieee80211_hw *hw, struct ieee80211_tx_control *control,
 		wcid = &mvif->sta.deflink.wcid;
 	}
 
+	if (vif && control->sta && ieee80211_vif_is_mld(vif)) {
+		struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+		struct ieee80211_link_sta *link_sta;
+		struct ieee80211_bss_conf *conf;
+
+		link_id = wcid->link_id;
+		rcu_read_lock();
+		conf = rcu_dereference(vif->link_conf[link_id]);
+		memcpy(hdr->addr2, conf->addr, ETH_ALEN);
+
+		link_sta = rcu_dereference(control->sta->link[link_id]);
+		memcpy(hdr->addr1, link_sta->addr, ETH_ALEN);
+
+		if (vif->type == NL80211_IFTYPE_STATION)
+			memcpy(hdr->addr3, conf->bssid, ETH_ALEN);
+		rcu_read_unlock();
+	}
+
 	if (mt76_connac_pm_ref(mphy, &dev->pm)) {
 		mt76_tx(mphy, control->sta, wcid, skb);
 		mt76_connac_pm_unref(mphy, &dev->pm);
-- 
cgit v1.2.3-58-ga151


From 3e758e9f777cde8e9d0106e7b36c6eca389e5856 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:52 -0700
Subject: wifi: mt76: mt7925: add link handling in mt7925_set_key

add link handling in mt7925_set_key to support MLO-enabled
firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/164608d96ce10a40e673bc6f252189c94bc48da7.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 79b43101237c..a2cf05fdfc3a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -610,6 +610,9 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 			  struct ieee80211_key_conf *key)
 {
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_sta *msta = sta ? (struct mt792x_sta *)sta->drv_priv :
+				  &mvif->sta;
 	int err;
 
 	/* The hardware does not support per-STA RX GTK, fallback
@@ -624,7 +627,20 @@ static int mt7925_set_key(struct ieee80211_hw *hw, enum set_key_cmd cmd,
 
 	mt792x_mutex_acquire(dev);
 
-	err = mt7925_set_link_key(hw, cmd, vif, sta, key, vif->bss_conf.link_id);
+	if (ieee80211_vif_is_mld(vif)) {
+		unsigned int link_id;
+		unsigned long add;
+
+		add = key->link_id != -1 ? BIT(key->link_id) : msta->valid_links;
+
+		for_each_set_bit(link_id, &add, IEEE80211_MLD_MAX_NUM_LINKS) {
+			err = mt7925_set_link_key(hw, cmd, vif, sta, key, link_id);
+			if (err < 0)
+				break;
+		}
+	} else {
+		err = mt7925_set_link_key(hw, cmd, vif, sta, key, vif->bss_conf.link_id);
+	}
 
 	mt792x_mutex_release(dev);
 
-- 
cgit v1.2.3-58-ga151


From b880cb3caa3e19610701322451b8bdb6f1f6813b Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:53 -0700
Subject: wifi: mt76: mt7925: add link handling to mt7925_change_chanctx

add link handling to mt7925_change_chanctx

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/49f2a2f8230979a4e123bc1dab70a8496875cc6c.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 22 ++++++++++++++++++++--
 1 file changed, 20 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index a2cf05fdfc3a..957e6bcaf9c1 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1571,13 +1571,15 @@ mt7925_change_chanctx(struct ieee80211_hw *hw,
 {
 	struct mt792x_chanctx *mctx = (struct mt792x_chanctx *)ctx->drv_priv;
 	struct mt792x_phy *phy = mt792x_hw_phy(hw);
+	struct mt792x_bss_conf *mconf;
 	struct ieee80211_vif *vif;
 	struct mt792x_vif *mvif;
 
 	if (!mctx->bss_conf)
 		return;
 
-	mvif = container_of(mctx->bss_conf, struct mt792x_vif, bss_conf);
+	mconf = mctx->bss_conf;
+	mvif = mconf->vif;
 	vif = container_of((void *)mvif, struct ieee80211_vif, drv_priv);
 
 	mt792x_mutex_acquire(phy->dev);
@@ -1585,8 +1587,24 @@ mt7925_change_chanctx(struct ieee80211_hw *hw,
 		mt7925_mcu_set_sniffer(mvif->phy->dev, vif, true);
 		mt7925_mcu_config_sniffer(mvif, ctx);
 	} else {
-		mt7925_mcu_set_chctx(mvif->phy->mt76, &mvif->bss_conf.mt76, ctx);
+		if (ieee80211_vif_is_mld(vif)) {
+			unsigned long valid = mvif->valid_links;
+			u8 i;
+
+			for_each_set_bit(i, &valid, IEEE80211_MLD_MAX_NUM_LINKS) {
+				mconf = mt792x_vif_to_link(mvif, i);
+				if (mconf && mconf->mt76.ctx == ctx)
+					break;
+			}
+
+		} else {
+			mconf = &mvif->bss_conf;
+		}
+
+		if (mconf)
+			mt7925_mcu_set_chctx(mvif->phy->mt76, &mconf->mt76, ctx);
 	}
+
 	mt792x_mutex_release(phy->dev);
 }
 
-- 
cgit v1.2.3-58-ga151


From a03abb5045c4f47d4b70bf8b8bb1328bffbb2a74 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:54 -0700
Subject: wifi: mt76: mt7925: add link handling in the BSS_CHANGED_PS handler

Added link handling in the BSS_CHANGED_PS handler.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/542ca29261af0cac007d2cb821ed36284d51c679.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 957e6bcaf9c1..938909ab8d1c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1637,6 +1637,7 @@ static void mt7925_vif_cfg_changed(struct ieee80211_hw *hw,
 				   struct ieee80211_vif *vif,
 				   u64 changed)
 {
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
 
 	mt792x_mutex_acquire(dev);
@@ -1653,8 +1654,17 @@ static void mt7925_vif_cfg_changed(struct ieee80211_hw *hw,
 		mt7925_mcu_update_arp_filter(&dev->mt76, &mvif->bss_conf.mt76);
 	}
 
-	if (changed & BSS_CHANGED_PS)
-		mt7925_mcu_uni_bss_ps(dev, &vif->bss_conf);
+	if (changed & BSS_CHANGED_PS) {
+		unsigned long valid = ieee80211_vif_is_mld(vif) ?
+					      mvif->valid_links : BIT(0);
+		struct ieee80211_bss_conf *bss_conf;
+		int i;
+
+		for_each_set_bit(i, &valid, IEEE80211_MLD_MAX_NUM_LINKS) {
+			bss_conf = mt792x_vif_to_bss_conf(vif, i);
+			mt7925_mcu_uni_bss_ps(dev, bss_conf);
+		}
+	}
 
 	mt792x_mutex_release(dev);
 }
-- 
cgit v1.2.3-58-ga151


From acc7a4a6257fb88734125ab89fd6ca5857d77682 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:55 -0700
Subject: wifi: mt76: mt7925: add link handling in mt7925_mcu_set_beacon_filter

add link handling in mt7925_mcu_set_beacon_filter

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/b814718a40ad5df6100e63d15c87ba95a318bc22.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 23 +++++++++++++++++------
 1 file changed, 17 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index d87c23911119..76ead51fdef2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1818,21 +1818,32 @@ int mt7925_mcu_set_beacon_filter(struct mt792x_dev *dev,
 {
 #define MT7925_FIF_BIT_CLR		BIT(1)
 #define MT7925_FIF_BIT_SET		BIT(0)
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	unsigned long valid = ieee80211_vif_is_mld(vif) ?
+				      mvif->valid_links : BIT(0);
+	struct ieee80211_bss_conf *bss_conf;
 	int err = 0;
+	int i;
 
 	if (enable) {
-		err = mt7925_mcu_uni_bss_bcnft(dev, &vif->bss_conf, true);
-		if (err)
-			return err;
+		for_each_set_bit(i, &valid, IEEE80211_MLD_MAX_NUM_LINKS) {
+			bss_conf = mt792x_vif_to_bss_conf(vif, i);
+			err = mt7925_mcu_uni_bss_bcnft(dev, bss_conf, true);
+			if (err < 0)
+				return err;
+		}
 
 		return mt7925_mcu_set_rxfilter(dev, 0,
 					       MT7925_FIF_BIT_SET,
 					       MT_WF_RFCR_DROP_OTHER_BEACON);
 	}
 
-	err = mt7925_mcu_set_bss_pm(dev, &vif->bss_conf, false);
-	if (err)
-		return err;
+	for_each_set_bit(i, &valid, IEEE80211_MLD_MAX_NUM_LINKS) {
+		bss_conf = mt792x_vif_to_bss_conf(vif, i);
+		err = mt7925_mcu_set_bss_pm(dev, bss_conf, false);
+		if (err)
+			return err;
+	}
 
 	return mt7925_mcu_set_rxfilter(dev, 0,
 				       MT7925_FIF_BIT_CLR,
-- 
cgit v1.2.3-58-ga151


From 44eb173bdd4f1775e13fce483d9e44ea8929717f Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:56 -0700
Subject: wifi: mt76: mt7925: add link handling in mt7925_txwi_free

add link handling in mt7925_txwi_free.

MT7996 should have the similar the logic, we try to copy from there to
ensure consistency between both.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/f8084e19d388c6b0807618ac6d1bb8ac633dc8cd.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mac.c    | 54 ++++++++++++++++------
 drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h |  2 +-
 .../net/wireless/mediatek/mt76/mt7925/pci_mac.c    |  2 +-
 3 files changed, 41 insertions(+), 17 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
index 6f5f2130420e..0cb9031e9fdd 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
@@ -839,27 +839,53 @@ mt7925_mac_write_txwi(struct mt76_dev *dev, __le32 *txwi,
 }
 EXPORT_SYMBOL_GPL(mt7925_mac_write_txwi);
 
-static void mt7925_tx_check_aggr(struct ieee80211_sta *sta, __le32 *txwi)
+static void mt7925_tx_check_aggr(struct ieee80211_sta *sta, struct sk_buff *skb,
+				 struct mt76_wcid *wcid)
 {
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ieee80211_link_sta *link_sta;
+	struct mt792x_link_sta *mlink;
 	struct mt792x_sta *msta;
+	bool is_8023;
 	u16 fc, tid;
-	u32 val;
 
-	if (!sta || !(sta->deflink.ht_cap.ht_supported || sta->deflink.he_cap.has_he))
+	link_sta = rcu_dereference(sta->link[wcid->link_id]);
+	if (!link_sta)
 		return;
 
-	tid = le32_get_bits(txwi[1], MT_TXD1_TID);
-	if (tid >= 6) /* skip VO queue */
+	if (!sta || !(link_sta->ht_cap.ht_supported || link_sta->he_cap.has_he))
 		return;
 
-	val = le32_to_cpu(txwi[2]);
-	fc = FIELD_GET(MT_TXD2_FRAME_TYPE, val) << 2 |
-	     FIELD_GET(MT_TXD2_SUB_TYPE, val) << 4;
+	tid = skb->priority & IEEE80211_QOS_CTL_TID_MASK;
+	is_8023 = info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP;
+
+	if (is_8023) {
+		fc = IEEE80211_FTYPE_DATA |
+		     (sta->wme ? IEEE80211_STYPE_QOS_DATA :
+		      IEEE80211_STYPE_DATA);
+	} else {
+		/* No need to get precise TID for Action/Management Frame,
+		 * since it will not meet the following Frame Control
+		 * condition anyway.
+		 */
+
+		struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+
+		fc = le16_to_cpu(hdr->frame_control) &
+		     (IEEE80211_FCTL_FTYPE | IEEE80211_FCTL_STYPE);
+	}
+
 	if (unlikely(fc != (IEEE80211_FTYPE_DATA | IEEE80211_STYPE_QOS_DATA)))
 		return;
 
 	msta = (struct mt792x_sta *)sta->drv_priv;
-	if (!test_and_set_bit(tid, &msta->deflink.wcid.ampdu_state))
+
+	if (sta->mlo && msta->deflink_id != IEEE80211_LINK_UNSPECIFIED)
+		mlink = rcu_dereference(msta->link[msta->deflink_id]);
+	else
+		mlink = &msta->deflink;
+
+	if (!test_and_set_bit(tid, &mlink->wcid.ampdu_state))
 		ieee80211_start_tx_ba_session(sta, tid, 0);
 }
 
@@ -1039,7 +1065,7 @@ out:
 }
 
 void mt7925_txwi_free(struct mt792x_dev *dev, struct mt76_txwi_cache *t,
-		      struct ieee80211_sta *sta, bool clear_status,
+		      struct ieee80211_sta *sta, struct mt76_wcid *wcid,
 		      struct list_head *free_list)
 {
 	struct mt76_dev *mdev = &dev->mt76;
@@ -1052,10 +1078,8 @@ void mt7925_txwi_free(struct mt792x_dev *dev, struct mt76_txwi_cache *t,
 
 	txwi = (__le32 *)mt76_get_txwi_ptr(mdev, t);
 	if (sta) {
-		struct mt76_wcid *wcid = (struct mt76_wcid *)sta->drv_priv;
-
 		if (likely(t->skb->protocol != cpu_to_be16(ETH_P_PAE)))
-			mt7925_tx_check_aggr(sta, txwi);
+			mt7925_tx_check_aggr(sta, t->skb, wcid);
 
 		wcid_idx = wcid->idx;
 	} else {
@@ -1140,7 +1164,7 @@ mt7925_mac_tx_free(struct mt792x_dev *dev, void *data, int len)
 			if (!txwi)
 				continue;
 
-			mt7925_txwi_free(dev, txwi, sta, 0, &free_list);
+			mt7925_txwi_free(dev, txwi, sta, wcid, &free_list);
 		}
 	}
 
@@ -1426,7 +1450,7 @@ void mt7925_usb_sdio_tx_complete_skb(struct mt76_dev *mdev,
 	sta = wcid_to_sta(wcid);
 
 	if (sta && likely(e->skb->protocol != cpu_to_be16(ETH_P_PAE)))
-		mt7925_tx_check_aggr(sta, txwi);
+		mt76_connac2_tx_check_aggr(sta, txwi);
 
 	skb_pull(e->skb, headroom);
 	mt76_tx_complete_skb(mdev, e->wcid, e->skb);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
index dbc9e3d426e9..bcf8541fa31d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
@@ -260,7 +260,7 @@ void mt7925_mac_write_txwi(struct mt76_dev *dev, __le32 *txwi,
 			   struct ieee80211_key_conf *key, int pid,
 			   enum mt76_txq_id qid, u32 changed);
 void mt7925_txwi_free(struct mt792x_dev *dev, struct mt76_txwi_cache *t,
-		      struct ieee80211_sta *sta, bool clear_status,
+		      struct ieee80211_sta *sta, struct mt76_wcid *wcid,
 		      struct list_head *free_list);
 int mt7925_mcu_parse_response(struct mt76_dev *mdev, int cmd,
 			      struct sk_buff *skb, int seq);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c b/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c
index ef631a68005d..faedbf766d1a 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/pci_mac.c
@@ -60,7 +60,7 @@ void mt7925_tx_token_put(struct mt792x_dev *dev)
 
 	spin_lock_bh(&dev->mt76.token_lock);
 	idr_for_each_entry(&dev->mt76.token, txwi, id) {
-		mt7925_txwi_free(dev, txwi, NULL, false, NULL);
+		mt7925_txwi_free(dev, txwi, NULL, NULL, NULL);
 		dev->mt76.token_count--;
 	}
 	spin_unlock_bh(&dev->mt76.token_lock);
-- 
cgit v1.2.3-58-ga151


From 9f89f05bef1a57352e908910e0065b15d542038c Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:57 -0700
Subject: wifi: mt76: mt7925: add link handling in mt7925_mac_sta_assoc

add links handling in mt7925_mac_sta_assoc

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/0ddf1834783b111d626b05d298478f520d7c0caa.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 104 ++++++++++++++++++++++-
 drivers/net/wireless/mediatek/mt76/mt792x.h      |   1 +
 2 files changed, 103 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 938909ab8d1c..68a615bbec76 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -387,6 +387,7 @@ mt7925_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)
 	mvif->phy = phy;
 	mvif->bss_conf.vif = mvif;
 	mvif->sta.vif = mvif;
+	mvif->deflink_id = IEEE80211_LINK_UNSPECIFIED;
 
 	ret = mt7925_mac_link_bss_add(dev, &vif->bss_conf, &mvif->sta.deflink);
 	if (ret < 0)
@@ -923,6 +924,89 @@ int mt7925_mac_sta_add(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 }
 EXPORT_SYMBOL_GPL(mt7925_mac_sta_add);
 
+static u16
+mt7925_mac_select_links(struct mt76_dev *mdev, struct ieee80211_vif *vif)
+{
+	unsigned long usable_links = ieee80211_vif_usable_links(vif);
+	struct  {
+		u8 link_id;
+		enum nl80211_band band;
+	} data[IEEE80211_MLD_MAX_NUM_LINKS];
+	u8 link_id, i, j, n_data = 0;
+	u16 sel_links = 0;
+
+	if (!ieee80211_vif_is_mld(vif))
+		return 0;
+
+	if (vif->active_links == usable_links)
+		return vif->active_links;
+
+	rcu_read_lock();
+	for_each_set_bit(link_id, &usable_links, IEEE80211_MLD_MAX_NUM_LINKS) {
+		struct ieee80211_bss_conf *link_conf =
+			rcu_dereference(vif->link_conf[link_id]);
+
+		if (WARN_ON_ONCE(!link_conf))
+			continue;
+
+		data[n_data].link_id = link_id;
+		data[n_data].band = link_conf->chanreq.oper.chan->band;
+		n_data++;
+	}
+	rcu_read_unlock();
+
+	for (i = 0; i < n_data; i++) {
+		if (!(BIT(data[i].link_id) & vif->active_links))
+			continue;
+
+		sel_links = BIT(data[i].link_id);
+
+		for (j = 0; j < n_data; j++) {
+			if (data[i].band != data[j].band) {
+				sel_links |= BIT(data[j].link_id);
+				break;
+			}
+		}
+
+		break;
+	}
+
+	return sel_links;
+}
+
+static void
+mt7925_mac_set_links(struct mt76_dev *mdev, struct ieee80211_vif *vif)
+{
+	struct mt792x_dev *dev = container_of(mdev, struct mt792x_dev, mt76);
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct ieee80211_bss_conf *link_conf =
+		mt792x_vif_to_bss_conf(vif, mvif->deflink_id);
+	struct cfg80211_chan_def *chandef = &link_conf->chanreq.oper;
+	enum nl80211_band band = chandef->chan->band, secondary_band;
+
+	u16 sel_links = mt7925_mac_select_links(mdev, vif);
+	u8 secondary_link_id = __ffs(~BIT(mvif->deflink_id) & sel_links);
+
+	if (!ieee80211_vif_is_mld(vif) || hweight16(sel_links) < 2)
+		return;
+
+	link_conf = mt792x_vif_to_bss_conf(vif, secondary_link_id);
+	secondary_band = link_conf->chanreq.oper.chan->band;
+
+	if (band == NL80211_BAND_2GHZ ||
+	    (band == NL80211_BAND_5GHZ && secondary_band == NL80211_BAND_6GHZ)) {
+		mt7925_abort_roc(mvif->phy, &mvif->bss_conf);
+
+		mt792x_mutex_acquire(dev);
+
+		mt7925_set_mlo_roc(mvif->phy, &mvif->bss_conf, sel_links);
+
+		mt792x_mutex_release(dev);
+	}
+
+	ieee80211_set_active_links_async(vif, sel_links);
+}
+
 static void mt7925_mac_link_sta_assoc(struct mt76_dev *mdev,
 				      struct ieee80211_vif *vif,
 				      struct ieee80211_link_sta *link_sta)
@@ -937,7 +1021,11 @@ static void mt7925_mac_link_sta_assoc(struct mt76_dev *mdev,
 
 	mt792x_mutex_acquire(dev);
 
-	link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
+	if (ieee80211_vif_is_mld(vif)) {
+		link_conf = mt792x_vif_to_bss_conf(vif, msta->deflink_id);
+	} else {
+		link_conf = mt792x_vif_to_bss_conf(vif, vif->bss_conf.link_id);
+	}
 
 	if (vif->type == NL80211_IFTYPE_STATION && !link_sta->sta->tdls) {
 		struct mt792x_bss_conf *mconf;
@@ -961,7 +1049,18 @@ static void mt7925_mac_link_sta_assoc(struct mt76_dev *mdev,
 void mt7925_mac_sta_assoc(struct mt76_dev *mdev, struct ieee80211_vif *vif,
 			  struct ieee80211_sta *sta)
 {
-	mt7925_mac_link_sta_assoc(mdev, vif, &sta->deflink);
+	if (ieee80211_vif_is_mld(vif)) {
+		struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
+		struct ieee80211_link_sta *link_sta;
+
+		link_sta = mt792x_sta_to_link_sta(vif, sta, msta->deflink_id);
+
+		mt7925_mac_set_links(mdev, vif);
+
+		mt7925_mac_link_sta_assoc(mdev, vif, link_sta);
+	} else {
+		mt7925_mac_link_sta_assoc(mdev, vif, &sta->deflink);
+	}
 }
 EXPORT_SYMBOL_GPL(mt7925_mac_sta_assoc);
 
@@ -1753,6 +1852,7 @@ mt7925_change_vif_links(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 
 	for_each_set_bit(link_id, &add, IEEE80211_MLD_MAX_NUM_LINKS) {
 		if (!old_links) {
+			mvif->deflink_id = link_id;
 			mconf = &mvif->bss_conf;
 			mlink = &mvif->sta.deflink;
 		} else {
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index af9a103dc7f4..5ede24116748 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -131,6 +131,7 @@ struct mt792x_vif {
 
 	struct mt792x_phy *phy;
 	u16 valid_links;
+	u8 deflink_id;
 };
 
 struct mt792x_phy {
-- 
cgit v1.2.3-58-ga151


From b859ad65309a5f1654e8b284de582831fc88e2d8 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:58 -0700
Subject: wifi: mt76: mt7925: add link handling in mt7925_sta_set_decap_offload

add link handling in mt7925_sta_set_decap_offload

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/df2524e3a63a93699bde441e4ab70a70bbac53ef.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c   | 21 ++++++++++++++-----
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    | 24 ++++++++++++++++------
 drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h |  3 ++-
 3 files changed, 36 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 68a615bbec76..0b61695c9b65 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1464,16 +1464,27 @@ static void mt7925_sta_set_decap_offload(struct ieee80211_hw *hw,
 					 bool enabled)
 {
 	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+	unsigned long valid = mvif->valid_links;
+	u8 i;
 
 	mt792x_mutex_acquire(dev);
 
-	if (enabled)
-		set_bit(MT_WCID_FLAG_HDR_TRANS, &msta->deflink.wcid.flags);
-	else
-		clear_bit(MT_WCID_FLAG_HDR_TRANS, &msta->deflink.wcid.flags);
+	valid = ieee80211_vif_is_mld(vif) ? mvif->valid_links : BIT(0);
+
+	for_each_set_bit(i, &valid, IEEE80211_MLD_MAX_NUM_LINKS) {
+		struct mt792x_link_sta *mlink;
 
-	mt7925_mcu_wtbl_update_hdr_trans(dev, vif, sta);
+		mlink = mt792x_sta_to_link(msta, i);
+
+		if (enabled)
+			set_bit(MT_WCID_FLAG_HDR_TRANS, &mlink->wcid.flags);
+		else
+			clear_bit(MT_WCID_FLAG_HDR_TRANS, &mlink->wcid.flags);
+
+		mt7925_mcu_wtbl_update_hdr_trans(dev, vif, sta, i);
+	}
 
 	mt792x_mutex_release(dev);
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 76ead51fdef2..20b118cd450c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -887,10 +887,15 @@ mt7925_mcu_sta_hdr_trans_tlv(struct sk_buff *skb,
 	else
 		hdr_trans->from_ds = true;
 
-	if (link_sta)
-		wcid = (struct mt76_wcid *)link_sta->sta->drv_priv;
-	else
+	if (link_sta) {
+		struct mt792x_sta *msta = (struct mt792x_sta *)link_sta->sta->drv_priv;
+		struct mt792x_link_sta *mlink;
+
+		mlink = mt792x_sta_to_link(msta, link_sta->link_id);
+		wcid = &mlink->wcid;
+	} else {
 		wcid = &mvif->sta.deflink.wcid;
+	}
 
 	if (!wcid)
 		return;
@@ -904,17 +909,24 @@ mt7925_mcu_sta_hdr_trans_tlv(struct sk_buff *skb,
 
 int mt7925_mcu_wtbl_update_hdr_trans(struct mt792x_dev *dev,
 				     struct ieee80211_vif *vif,
-				     struct ieee80211_sta *sta)
+				     struct ieee80211_sta *sta,
+				     int link_id)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct ieee80211_link_sta *link_sta = sta ? &sta->deflink : NULL;
+	struct mt792x_link_sta *mlink;
+	struct mt792x_bss_conf *mconf;
 	struct mt792x_sta *msta;
 	struct sk_buff *skb;
 
 	msta = sta ? (struct mt792x_sta *)sta->drv_priv : &mvif->sta;
 
-	skb = __mt76_connac_mcu_alloc_sta_req(&dev->mt76, &mvif->bss_conf.mt76,
-					      &msta->deflink.wcid,
+	mlink = mt792x_sta_to_link(msta, link_id);
+	link_sta = mt792x_sta_to_link_sta(vif, sta, link_id);
+	mconf = mt792x_vif_to_link(mvif, link_id);
+
+	skb = __mt76_connac_mcu_alloc_sta_req(&dev->mt76, &mconf->mt76,
+					      &mlink->wcid,
 					      MT7925_STA_UPDATE_MAX_SIZE);
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
index bcf8541fa31d..1ae66629e39f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
@@ -316,6 +316,7 @@ int mt7925_mcu_add_key(struct mt76_dev *dev, struct ieee80211_vif *vif,
 int mt7925_mcu_set_rts_thresh(struct mt792x_phy *phy, u32 val);
 int mt7925_mcu_wtbl_update_hdr_trans(struct mt792x_dev *dev,
 				     struct ieee80211_vif *vif,
-				     struct ieee80211_sta *sta);
+				     struct ieee80211_sta *sta,
+				     int link_id);
 
 #endif
-- 
cgit v1.2.3-58-ga151


From 14061994184df6bb89cba31657ee1db24480e92a Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:27:59 -0700
Subject: wifi: mt76: mt7925: add link handling in mt7925_vif_connect_iter

add link handling in mt7925_vif_connect_iter

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/c8d66043a4d31801249d53dd23f695c28a97a933.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mac.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
index 0cb9031e9fdd..cf36750cf709 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mac.c
@@ -1267,15 +1267,23 @@ mt7925_vif_connect_iter(void *priv, u8 *mac,
 			struct ieee80211_vif *vif)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	unsigned long valid = ieee80211_vif_is_mld(vif) ?
+			      mvif->valid_links : BIT(0);
 	struct mt792x_dev *dev = mvif->phy->dev;
 	struct ieee80211_hw *hw = mt76_hw(dev);
+	struct ieee80211_bss_conf *bss_conf;
+	int i;
 
 	if (vif->type == NL80211_IFTYPE_STATION)
 		ieee80211_disconnect(vif, true);
 
-	mt76_connac_mcu_uni_add_dev(&dev->mphy, &vif->bss_conf,
-				    &mvif->sta.deflink.wcid, true);
-	mt7925_mcu_set_tx(dev, &vif->bss_conf);
+	for_each_set_bit(i, &valid, IEEE80211_MLD_MAX_NUM_LINKS) {
+		bss_conf = mt792x_vif_to_bss_conf(vif, i);
+
+		mt76_connac_mcu_uni_add_dev(&dev->mphy, bss_conf,
+					    &mvif->sta.deflink.wcid, true);
+		mt7925_mcu_set_tx(dev, bss_conf);
+	}
 
 	if (vif->type == NL80211_IFTYPE_AP) {
 		mt76_connac_mcu_uni_add_bss(dev->phy.mt76, vif, &mvif->sta.deflink.wcid,
-- 
cgit v1.2.3-58-ga151


From 147324292979fdd0e612199f898db7162ae083c6 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:00 -0700
Subject: wifi: mt76: mt7925: add link handling in the BSS_CHANGED_ARP_FILTER
 handler

add link handling in the BSS_CHANGED_ARP_FILTER handler.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/6e26e2ce253630ab29702693a7822f9961de6a18.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 16 ++++++++--------
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c  |  8 +++++---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.h  |  2 +-
 3 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 0b61695c9b65..2329e1a19781 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1749,6 +1749,10 @@ static void mt7925_vif_cfg_changed(struct ieee80211_hw *hw,
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+	unsigned long valid = ieee80211_vif_is_mld(vif) ?
+				      mvif->valid_links : BIT(0);
+	struct ieee80211_bss_conf *bss_conf;
+	int i;
 
 	mt792x_mutex_acquire(dev);
 
@@ -1759,17 +1763,13 @@ static void mt7925_vif_cfg_changed(struct ieee80211_hw *hw,
 	}
 
 	if (changed & BSS_CHANGED_ARP_FILTER) {
-		struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-
-		mt7925_mcu_update_arp_filter(&dev->mt76, &mvif->bss_conf.mt76);
+		for_each_set_bit(i, &valid, IEEE80211_MLD_MAX_NUM_LINKS) {
+			bss_conf = mt792x_vif_to_bss_conf(vif, i);
+			mt7925_mcu_update_arp_filter(&dev->mt76, bss_conf);
+		}
 	}
 
 	if (changed & BSS_CHANGED_PS) {
-		unsigned long valid = ieee80211_vif_is_mld(vif) ?
-					      mvif->valid_links : BIT(0);
-		struct ieee80211_bss_conf *bss_conf;
-		int i;
-
 		for_each_set_bit(i, &valid, IEEE80211_MLD_MAX_NUM_LINKS) {
 			bss_conf = mt792x_vif_to_bss_conf(vif, i);
 			mt7925_mcu_uni_bss_ps(dev, bss_conf);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 20b118cd450c..2dbe9362efcf 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -121,10 +121,12 @@ int mt7925_mcu_regval(struct mt792x_dev *dev, u32 regidx, u32 *val, bool set)
 EXPORT_SYMBOL_GPL(mt7925_mcu_regval);
 
 int mt7925_mcu_update_arp_filter(struct mt76_dev *dev,
-				 struct mt76_vif *vif)
+				 struct ieee80211_bss_conf *link_conf)
 {
-	struct ieee80211_vif *mvif = container_of((void *)vif, struct ieee80211_vif,
+	struct ieee80211_vif *mvif = container_of((void *)link_conf->vif,
+						  struct ieee80211_vif,
 						  drv_priv);
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
 	struct sk_buff *skb;
 	int i, len = min_t(int, mvif->cfg.arp_addr_cnt,
 			   IEEE80211_BSS_ARP_ADDR_LIST_LEN);
@@ -136,7 +138,7 @@ int mt7925_mcu_update_arp_filter(struct mt76_dev *dev,
 		struct mt7925_arpns_tlv arp;
 	} req = {
 		.hdr = {
-			.bss_idx = vif->idx,
+			.bss_idx = mconf->mt76.idx,
 		},
 		.arp = {
 			.tag = cpu_to_le16(UNI_OFFLOAD_OFFLOAD_ARP),
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
index fefb6ab384bb..7fb46cad9157 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
@@ -626,5 +626,5 @@ int mt7925_mcu_set_chctx(struct mt76_phy *phy, struct mt76_vif *mvif,
 			 struct ieee80211_chanctx_conf *ctx);
 int mt7925_mcu_set_rate_txpower(struct mt76_phy *phy);
 int mt7925_mcu_update_arp_filter(struct mt76_dev *dev,
-				 struct mt76_vif *vif);
+				 struct ieee80211_bss_conf *link_conf);
 #endif
-- 
cgit v1.2.3-58-ga151


From 9f8f587cab696e68804e2c4460f4ac1f1a686c8b Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:01 -0700
Subject: wifi: mt76: mt7925: add link handling in the mt7925_ipv6_addr_change

add link handling in the mt7925_ipv6_addr_change

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/ed429b1896cc10c8db25dc2ade3e59bec78d7131.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 29 +++++++++++++++++++-----
 1 file changed, 23 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 2329e1a19781..bb8a89e08d7d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1490,12 +1490,12 @@ static void mt7925_sta_set_decap_offload(struct ieee80211_hw *hw,
 }
 
 #if IS_ENABLED(CONFIG_IPV6)
-static void mt7925_ipv6_addr_change(struct ieee80211_hw *hw,
-				    struct ieee80211_vif *vif,
-				    struct inet6_dev *idev)
+static void __mt7925_ipv6_addr_change(struct ieee80211_hw *hw,
+				      struct ieee80211_bss_conf *link_conf,
+				      struct inet6_dev *idev)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	struct mt792x_dev *dev = mvif->phy->dev;
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
+	struct mt792x_dev *dev = mt792x_hw_dev(hw);
 	struct inet6_ifaddr *ifa;
 	struct sk_buff *skb;
 	u8 idx = 0;
@@ -1509,7 +1509,7 @@ static void mt7925_ipv6_addr_change(struct ieee80211_hw *hw,
 		struct in6_addr ns_addrs[IEEE80211_BSS_ARP_ADDR_LIST_LEN];
 	} req_hdr = {
 		.hdr = {
-			.bss_idx = mvif->bss_conf.mt76.idx,
+			.bss_idx = mconf->mt76.idx,
 		},
 		.arpns = {
 			.tag = cpu_to_le16(UNI_OFFLOAD_OFFLOAD_ND),
@@ -1544,6 +1544,23 @@ static void mt7925_ipv6_addr_change(struct ieee80211_hw *hw,
 
 	ieee80211_queue_work(dev->mt76.hw, &dev->ipv6_ns_work);
 }
+
+static void mt7925_ipv6_addr_change(struct ieee80211_hw *hw,
+				    struct ieee80211_vif *vif,
+				    struct inet6_dev *idev)
+{
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	unsigned long valid = ieee80211_vif_is_mld(vif) ?
+			      mvif->valid_links : BIT(0);
+	struct ieee80211_bss_conf *bss_conf;
+	int i;
+
+	for_each_set_bit(i, &valid, IEEE80211_MLD_MAX_NUM_LINKS) {
+		bss_conf = mt792x_vif_to_bss_conf(vif, i);
+		__mt7925_ipv6_addr_change(hw, bss_conf, idev);
+	}
+}
+
 #endif
 
 int mt7925_set_tx_sar_pwr(struct ieee80211_hw *hw,
-- 
cgit v1.2.3-58-ga151


From 3fa01f055c41d20ec11485cd12fa0f82ee5557c1 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:02 -0700
Subject: wifi: mt76: mt7925: update rate index according to link id

Update rate index according to link id.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/e0bbc0932edbd9225fe7b7736693c137cf433a52.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index bb8a89e08d7d..3d23a87e9a74 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1801,9 +1801,12 @@ static void mt7925_link_info_changed(struct ieee80211_hw *hw,
 				     struct ieee80211_bss_conf *info,
 				     u64 changed)
 {
-	struct mt76_vif *mvif = (struct mt76_vif *)vif->drv_priv;
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
 	struct mt792x_phy *phy = mt792x_hw_phy(hw);
 	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+	struct mt792x_bss_conf *mconf;
+
+	mconf = mt792x_vif_to_link(mvif, info->link_id);
 
 	mt792x_mutex_acquire(dev);
 
@@ -1817,16 +1820,16 @@ static void mt7925_link_info_changed(struct ieee80211_hw *hw,
 	}
 
 	if (changed & BSS_CHANGED_MCAST_RATE)
-		mvif->mcast_rates_idx =
+		mconf->mt76.mcast_rates_idx =
 				mt7925_get_rates_table(hw, vif, false, true);
 
 	if (changed & BSS_CHANGED_BASIC_RATES)
-		mvif->basic_rates_idx =
+		mconf->mt76.basic_rates_idx =
 				mt7925_get_rates_table(hw, vif, false, false);
 
 	if (changed & (BSS_CHANGED_BEACON |
 		       BSS_CHANGED_BEACON_ENABLED)) {
-		mvif->beacon_rates_idx =
+		mconf->mt76.beacon_rates_idx =
 				mt7925_get_rates_table(hw, vif, true, false);
 
 		mt7925_mcu_uni_add_beacon_offload(dev, hw, vif,
-- 
cgit v1.2.3-58-ga151


From 4f0f33d2747f207ba552dd065042c2d7f3063402 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:03 -0700
Subject: wifi: mt76: mt7925: report link information in rx status

report link information in rx status

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/c7cdc17d94375a76802a5c4a825ea3e51cdb53af.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mac80211.c    | 5 +++++
 drivers/net/wireless/mediatek/mt76/mt76.h        | 1 +
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 2 ++
 3 files changed, 8 insertions(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mac80211.c b/drivers/net/wireless/mediatek/mt76/mac80211.c
index e8ba2e4e8484..bb291fe314fb 100644
--- a/drivers/net/wireless/mediatek/mt76/mac80211.c
+++ b/drivers/net/wireless/mediatek/mt76/mac80211.c
@@ -1125,6 +1125,11 @@ mt76_rx_convert(struct mt76_dev *dev, struct sk_buff *skb,
 	memcpy(status->chain_signal, mstat.chain_signal,
 	       sizeof(mstat.chain_signal));
 
+	if (mstat.wcid) {
+		status->link_valid = mstat.wcid->link_valid;
+		status->link_id = mstat.wcid->link_id;
+	}
+
 	*sta = wcid_to_sta(mstat.wcid);
 	*hw = mt76_phy_hw(dev, mstat.phy_idx);
 }
diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h
index daf1e43e304e..e1d30326f5af 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76.h
@@ -350,6 +350,7 @@ struct mt76_wcid {
 	u8 amsdu:1;
 	u8 phy_idx:2;
 	u8 link_id:4;
+	bool link_valid;
 
 	u8 rx_check_pn;
 	u8 rx_key_pn[IEEE80211_NUM_TIDS + 1][6];
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 3d23a87e9a74..be0a336f601e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -811,6 +811,7 @@ static int mt7925_mac_link_sta_add(struct mt76_dev *mdev,
 	mlink->wcid.tx_info |= MT_WCID_TX_INFO_SET;
 	mlink->last_txs = jiffies;
 	mlink->wcid.link_id = link_sta->link_id;
+	mlink->wcid.link_valid = !!link_sta->sta->valid_links;
 
 	ret = mt76_connac_pm_wake(&dev->mphy, &dev->pm);
 	if (ret)
@@ -1898,6 +1899,7 @@ mt7925_change_vif_links(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		mconf->link_id = link_id;
 		mconf->vif = mvif;
 		mlink->wcid.link_id = link_id;
+		mlink->wcid.link_valid = !!vif->valid_links;
 	}
 
 	if (hweight16(mvif->valid_links) == 0)
-- 
cgit v1.2.3-58-ga151


From b1d21403c0cfe712ce7ad4020f40c5ca6a013ec5 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:04 -0700
Subject: wifi: mt76: add def_wcid to struct mt76_wcid

add def_wcid to struct mt76_wcid to allow per-link wcid to retrieve the
struct ieee80211_sta structure.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/d5188ca633666eaad7e90a8061d1a0585487e128.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76.h        | 5 +++++
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 2 ++
 2 files changed, 7 insertions(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h
index e1d30326f5af..5f8773364d51 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76.h
@@ -368,6 +368,8 @@ struct mt76_wcid {
 	struct mt76_sta_stats stats;
 
 	struct list_head poll_list;
+
+	struct mt76_wcid *def_wcid;
 };
 
 struct mt76_txq {
@@ -1258,6 +1260,9 @@ wcid_to_sta(struct mt76_wcid *wcid)
 	if (!wcid || !wcid->sta)
 		return NULL;
 
+	if (wcid->def_wcid)
+		ptr = wcid->def_wcid;
+
 	return container_of(ptr, struct ieee80211_sta, drv_priv);
 }
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index be0a336f601e..c25822e3a943 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -892,6 +892,7 @@ mt7925_mac_sta_add_links(struct mt792x_dev *dev, struct ieee80211_vif *vif,
 		rcu_assign_pointer(msta->link[link_id], mlink);
 		mlink->sta = msta;
 		mlink->pri_link = &sta->deflink;
+		mlink->wcid.def_wcid = &msta->deflink.wcid;
 
 		link_sta = mt792x_sta_to_link_sta(vif, sta, link_id);
 		mt7925_mac_link_sta_add(&dev->mt76, vif, link_sta);
@@ -1900,6 +1901,7 @@ mt7925_change_vif_links(struct ieee80211_hw *hw, struct ieee80211_vif *vif,
 		mconf->vif = mvif;
 		mlink->wcid.link_id = link_id;
 		mlink->wcid.link_valid = !!vif->valid_links;
+		mlink->wcid.def_wcid = &mvif->sta.deflink.wcid;
 	}
 
 	if (hweight16(mvif->valid_links) == 0)
-- 
cgit v1.2.3-58-ga151


From 621e9e22b8d031da5c026b913de903f2e2180d79 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:05 -0700
Subject: wifi: mt76: mt7925: add mt7925_[assign,unassign]_vif_chanctx

add mt7925_[assign,unassign]_vif_chanctx to assign and unassign
chanctx to the specific link configuration. If the chctx is not
pass in the parameter, we will look up the channel information
from link_conf->chanreq.oper.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/15261879a16cb12674d6dea7703410baa6883799.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 75 ++++++++++++++++++++++--
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c  | 17 +++---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.h  |  1 +
 3 files changed, 82 insertions(+), 11 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index c25822e3a943..0627ef208641 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -1730,8 +1730,13 @@ mt7925_change_chanctx(struct ieee80211_hw *hw,
 			mconf = &mvif->bss_conf;
 		}
 
-		if (mconf)
-			mt7925_mcu_set_chctx(mvif->phy->mt76, &mconf->mt76, ctx);
+		if (mconf) {
+			struct ieee80211_bss_conf *link_conf;
+
+			link_conf = mt792x_vif_to_bss_conf(vif, mconf->link_id);
+			mt7925_mcu_set_chctx(mvif->phy->mt76, &mconf->mt76,
+					     link_conf, ctx);
+		}
 	}
 
 	mt792x_mutex_release(phy->dev);
@@ -1978,6 +1983,68 @@ out:
 	return err;
 }
 
+static int mt7925_assign_vif_chanctx(struct ieee80211_hw *hw,
+				     struct ieee80211_vif *vif,
+				     struct ieee80211_bss_conf *link_conf,
+				     struct ieee80211_chanctx_conf *ctx)
+{
+	struct mt792x_chanctx *mctx = (struct mt792x_chanctx *)ctx->drv_priv;
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+	struct ieee80211_bss_conf *pri_link_conf;
+	struct mt792x_bss_conf *mconf;
+
+	mutex_lock(&dev->mt76.mutex);
+
+	if (ieee80211_vif_is_mld(vif)) {
+		mconf = mt792x_vif_to_link(mvif, link_conf->link_id);
+		pri_link_conf = mt792x_vif_to_bss_conf(vif, mvif->deflink_id);
+
+		if (vif->type == NL80211_IFTYPE_STATION &&
+		    mconf == &mvif->bss_conf)
+			mt7925_mcu_add_bss_info(&dev->phy, NULL, pri_link_conf,
+						NULL, true);
+	} else {
+		mconf = &mvif->bss_conf;
+	}
+
+	mconf->mt76.ctx = ctx;
+	mctx->bss_conf = mconf;
+	mutex_unlock(&dev->mt76.mutex);
+
+	return 0;
+}
+
+static void mt7925_unassign_vif_chanctx(struct ieee80211_hw *hw,
+					struct ieee80211_vif *vif,
+					struct ieee80211_bss_conf *link_conf,
+					struct ieee80211_chanctx_conf *ctx)
+{
+	struct mt792x_chanctx *mctx = (struct mt792x_chanctx *)ctx->drv_priv;
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_dev *dev = mt792x_hw_dev(hw);
+	struct ieee80211_bss_conf *pri_link_conf;
+	struct mt792x_bss_conf *mconf;
+
+	mutex_lock(&dev->mt76.mutex);
+
+	if (ieee80211_vif_is_mld(vif)) {
+		mconf = mt792x_vif_to_link(mvif, link_conf->link_id);
+		pri_link_conf = mt792x_vif_to_bss_conf(vif, mvif->deflink_id);
+
+		if (vif->type == NL80211_IFTYPE_STATION &&
+		    mconf == &mvif->bss_conf)
+			mt7925_mcu_add_bss_info(&dev->phy, NULL, pri_link_conf,
+						NULL, false);
+	} else {
+		mconf = &mvif->bss_conf;
+	}
+
+	mctx->bss_conf = NULL;
+	mconf->mt76.ctx = NULL;
+	mutex_unlock(&dev->mt76.mutex);
+}
+
 const struct ieee80211_ops mt7925_ops = {
 	.tx = mt792x_tx,
 	.start = mt7925_start,
@@ -2030,8 +2097,8 @@ const struct ieee80211_ops mt7925_ops = {
 	.add_chanctx = mt7925_add_chanctx,
 	.remove_chanctx = mt7925_remove_chanctx,
 	.change_chanctx = mt7925_change_chanctx,
-	.assign_vif_chanctx = mt792x_assign_vif_chanctx,
-	.unassign_vif_chanctx = mt792x_unassign_vif_chanctx,
+	.assign_vif_chanctx = mt7925_assign_vif_chanctx,
+	.unassign_vif_chanctx = mt7925_unassign_vif_chanctx,
 	.mgd_prepare_tx = mt7925_mgd_prepare_tx,
 	.mgd_complete_tx = mt7925_mgd_complete_tx,
 	.vif_cfg_changed = mt7925_vif_cfg_changed,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 2dbe9362efcf..da3192bd82cd 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2083,9 +2083,11 @@ mt7925_mcu_uni_add_beacon_offload(struct mt792x_dev *dev,
 
 static
 void mt7925_mcu_bss_rlm_tlv(struct sk_buff *skb, struct mt76_phy *phy,
+			    struct ieee80211_bss_conf *link_conf,
 			    struct ieee80211_chanctx_conf *ctx)
 {
-	struct cfg80211_chan_def *chandef = ctx ? &ctx->def : &phy->chandef;
+	struct cfg80211_chan_def *chandef = ctx ? &ctx->def :
+						  &link_conf->chanreq.oper;
 	int freq1 = chandef->center_freq1, freq2 = chandef->center_freq2;
 	enum nl80211_band band = chandef->chan->band;
 	struct bss_rlm_tlv *req;
@@ -2152,6 +2154,7 @@ __mt7925_mcu_alloc_bss_req(struct mt76_dev *dev, struct mt76_vif *mvif, int len)
 }
 
 int mt7925_mcu_set_chctx(struct mt76_phy *phy, struct mt76_vif *mvif,
+			 struct ieee80211_bss_conf *link_conf,
 			 struct ieee80211_chanctx_conf *ctx)
 {
 	struct sk_buff *skb;
@@ -2161,7 +2164,7 @@ int mt7925_mcu_set_chctx(struct mt76_phy *phy, struct mt76_vif *mvif,
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
 
-	mt7925_mcu_bss_rlm_tlv(skb, phy, ctx);
+	mt7925_mcu_bss_rlm_tlv(skb, phy, link_conf, ctx);
 
 	return mt76_mcu_skb_send_msg(phy->dev, skb,
 				     MCU_UNI_CMD(BSS_INFO_UPDATE), true);
@@ -2223,7 +2226,8 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 {
 	struct ieee80211_vif *vif = link_conf->vif;
 	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
-	struct cfg80211_chan_def *chandef = ctx ? &ctx->def : &phy->chandef;
+	struct cfg80211_chan_def *chandef = ctx ? &ctx->def :
+						  &link_conf->chanreq.oper;
 	enum nl80211_band band = chandef->chan->band;
 	struct mt76_connac_bss_basic_tlv *basic_req;
 	struct mt792x_link_sta *mlink;
@@ -2348,7 +2352,8 @@ mt7925_mcu_bss_bmc_tlv(struct sk_buff *skb, struct mt792x_phy *phy,
 		       struct ieee80211_chanctx_conf *ctx,
 		       struct ieee80211_bss_conf *link_conf)
 {
-	struct cfg80211_chan_def *chandef = ctx ? &ctx->def : &phy->mt76->chandef;
+	struct cfg80211_chan_def *chandef = ctx ? &ctx->def :
+						  &link_conf->chanreq.oper;
 	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
 	enum nl80211_band band = chandef->chan->band;
 	struct mt76_vif *mvif = &mconf->mt76;
@@ -2488,8 +2493,6 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 			    int enable)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)link_conf->vif->drv_priv;
-	struct mt792x_bss_conf *mconf = mt792x_vif_to_link(mvif,
-							   link_conf->link_id);
 	struct mt792x_dev *dev = phy->dev;
 	struct sk_buff *skb;
 
@@ -2513,7 +2516,7 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 		mt7925_mcu_bss_color_tlv(skb, link_conf, enable);
 	}
 
-	mt7925_mcu_bss_rlm_tlv(skb, phy->mt76, mconf->mt76.ctx);
+	mt7925_mcu_bss_rlm_tlv(skb, phy->mt76, link_conf, ctx);
 
 	return mt76_mcu_skb_send_msg(&dev->mt76, skb,
 				     MCU_UNI_CMD(BSS_INFO_UPDATE), true);
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
index 7fb46cad9157..6ae5286385d9 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
@@ -623,6 +623,7 @@ int mt7925_mcu_set_deep_sleep(struct mt792x_dev *dev, bool enable);
 int mt7925_mcu_set_channel_domain(struct mt76_phy *phy);
 int mt7925_mcu_set_radio_en(struct mt792x_phy *phy, bool enable);
 int mt7925_mcu_set_chctx(struct mt76_phy *phy, struct mt76_vif *mvif,
+			 struct ieee80211_bss_conf *link_conf,
 			 struct ieee80211_chanctx_conf *ctx);
 int mt7925_mcu_set_rate_txpower(struct mt76_phy *phy);
 int mt7925_mcu_update_arp_filter(struct mt76_dev *dev,
-- 
cgit v1.2.3-58-ga151


From 254f777a9e842e2325f2ee8ba3d0cf45625c71b7 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:06 -0700
Subject: wifi: mt76: mt7925: update mt7925_mcu_sta_mld_tlv for MLO

Update mt7925_mcu_sta_mld_tlv for the MLO-enabled firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/25ea3ffd3875083577ab440ec63f3dcd9dc233f4.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 29 +++++++++++++++++++------
 1 file changed, 22 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index da3192bd82cd..59ae1fbc238b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1742,19 +1742,34 @@ static void
 mt7925_mcu_sta_mld_tlv(struct sk_buff *skb,
 		       struct ieee80211_vif *vif, struct ieee80211_sta *sta)
 {
-	struct mt76_wcid *wcid = (struct mt76_wcid *)sta->drv_priv;
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct mt792x_sta *msta = (struct mt792x_sta *)sta->drv_priv;
+	unsigned long valid = mvif->valid_links;
+	struct mt792x_bss_conf *mconf;
+	struct mt792x_link_sta *mlink;
 	struct sta_rec_mld *mld;
 	struct tlv *tlv;
+	int i, cnt = 0;
 
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_MLD, sizeof(*mld));
 	mld = (struct sta_rec_mld *)tlv;
-	memcpy(mld->mac_addr, vif->addr, ETH_ALEN);
-	mld->primary_id = cpu_to_le16(wcid->idx);
-	mld->wlan_id = cpu_to_le16(wcid->idx);
+	memcpy(mld->mac_addr, sta->addr, ETH_ALEN);
+	mld->primary_id = cpu_to_le16(msta->deflink.wcid.idx);
+	mld->wlan_id = cpu_to_le16(msta->deflink.wcid.idx);
+	mld->link_num = min_t(u8, hweight16(mvif->valid_links), 2);
+
+	for_each_set_bit(i, &valid, IEEE80211_MLD_MAX_NUM_LINKS) {
+		if (cnt == mld->link_num)
+			break;
 
-	/* TODO: 0 means deflink only, add secondary link(1) later */
-	mld->link_num = !!(hweight8(vif->active_links) > 1);
-	WARN_ON_ONCE(mld->link_num);
+		mconf = mt792x_vif_to_link(mvif, i);
+		mlink = mt792x_sta_to_link(msta, i);
+		mld->link[cnt].wlan_id = cpu_to_le16(mlink->wcid.idx);
+		mld->link[cnt++].bss_idx = mconf->mt76.idx;
+
+		if (mlink != &msta->deflink)
+			mld->secondary_id = cpu_to_le16(mlink->wcid.idx);
+	}
 }
 
 static int
-- 
cgit v1.2.3-58-ga151


From f8a80a4eed42499f5b923c60d6a51d00e3b6b7d4 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:07 -0700
Subject: wifi: mt76: mt7925: update mt7925_mcu_bss_mld_tlv for MLO

update mt7925_mcu_bss_mld_tlv for the MLO-enabled firmware,
the change remains compatible with the non-MLO mode and the
older firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/2d89b93bcfd92e7c6987aaa0a3c4637e605ac154.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 22 +++++++++++++---------
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.h |  5 ++++-
 2 files changed, 17 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 59ae1fbc238b..7a0225127135 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2393,24 +2393,29 @@ mt7925_mcu_bss_bmc_tlv(struct sk_buff *skb, struct mt792x_phy *phy,
 
 static void
 mt7925_mcu_bss_mld_tlv(struct sk_buff *skb,
-		       struct ieee80211_bss_conf *link_conf,
-		       struct ieee80211_link_sta *link_sta)
+		       struct ieee80211_bss_conf *link_conf)
 {
 	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
-	bool is_mld = ieee80211_vif_is_mld(link_conf->vif);
+	struct mt792x_vif *mvif = (struct mt792x_vif *)link_conf->vif->drv_priv;
 	struct bss_mld_tlv *mld;
 	struct tlv *tlv;
+	bool is_mld;
+
+	is_mld = ieee80211_vif_is_mld(link_conf->vif) ||
+		 (hweight16(mvif->valid_links) > 1);
 
 	tlv = mt76_connac_mcu_add_tlv(skb, UNI_BSS_INFO_MLD, sizeof(*mld));
 	mld = (struct bss_mld_tlv *)tlv;
 
-	mld->link_id = link_sta ? (is_mld ? link_conf->link_id : 0) : 0xff;
-	mld->group_mld_id = is_mld ? mconf->mt76.idx : 0xff;
+	mld->link_id = is_mld ? link_conf->link_id : 0xff;
+	/* apply the index of the primary link */
+	mld->group_mld_id = is_mld ? mvif->bss_conf.mt76.idx : 0xff;
 	mld->own_mld_id = mconf->mt76.idx + 32;
 	mld->remap_idx = 0xff;
+	mld->eml_enable = !!(link_conf->vif->cfg.eml_cap &
+			     IEEE80211_EML_CAP_EMLSR_SUPP);
 
-	if (link_sta)
-		memcpy(mld->mac_addr, link_sta->addr, ETH_ALEN);
+	memcpy(mld->mac_addr, link_conf->addr, ETH_ALEN);
 }
 
 static void
@@ -2520,10 +2525,9 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 	mt7925_mcu_bss_basic_tlv(skb, link_conf, link_sta, ctx, phy->mt76,
 				 mvif->sta.deflink.wcid.idx, enable);
 	mt7925_mcu_bss_sec_tlv(skb, link_conf);
-
 	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf);
 	mt7925_mcu_bss_qos_tlv(skb, link_conf);
-	mt7925_mcu_bss_mld_tlv(skb, link_conf, link_sta);
+	mt7925_mcu_bss_mld_tlv(skb, link_conf);
 	mt7925_mcu_bss_ifs_tlv(skb, link_conf);
 
 	if (link_conf->he_support) {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
index 6ae5286385d9..7b0c6937f918 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
@@ -366,7 +366,10 @@ struct bss_mld_tlv {
 	u8 mac_addr[ETH_ALEN];
 	u8 remap_idx;
 	u8 link_id;
-	u8 __rsv[2];
+	u8 eml_enable;
+	u8 max_link_num;
+	u8 hybrid_mode;
+	u8 __rsv[3];
 } __packed;
 
 struct sta_rec_ba_uni {
-- 
cgit v1.2.3-58-ga151


From 790e3e1ec8aa232abdbfaa26401167580e789366 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:08 -0700
Subject: wifi: mt76: mt7925: update mt7925_mcu_add_bss_info for MLO

Update mt7925_mcu_bss_mld_tlv for the MLO-enabled firmware

The change remains compatible with the non-MLO mode and the
older firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/a3d03b0a1ca916b2b8b2e7c0afcdcd7e258d97c3.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 7a0225127135..094fa4dde5fe 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2513,17 +2513,21 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 			    int enable)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)link_conf->vif->drv_priv;
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
 	struct mt792x_dev *dev = phy->dev;
+	struct mt792x_link_sta *mlink_bc;
 	struct sk_buff *skb;
 
-	skb = __mt7925_mcu_alloc_bss_req(&dev->mt76, &mvif->bss_conf.mt76,
+	skb = __mt7925_mcu_alloc_bss_req(&dev->mt76, &mconf->mt76,
 					 MT7925_BSS_UPDATE_MAX_SIZE);
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
 
+	mlink_bc = mt792x_sta_to_link(&mvif->sta, mconf->link_id);
+
 	/* bss_basic must be first */
 	mt7925_mcu_bss_basic_tlv(skb, link_conf, link_sta, ctx, phy->mt76,
-				 mvif->sta.deflink.wcid.idx, enable);
+				 mlink_bc->wcid.idx, enable);
 	mt7925_mcu_bss_sec_tlv(skb, link_conf);
 	mt7925_mcu_bss_bmc_tlv(skb, phy, ctx, link_conf);
 	mt7925_mcu_bss_qos_tlv(skb, link_conf);
@@ -2535,7 +2539,8 @@ int mt7925_mcu_add_bss_info(struct mt792x_phy *phy,
 		mt7925_mcu_bss_color_tlv(skb, link_conf, enable);
 	}
 
-	mt7925_mcu_bss_rlm_tlv(skb, phy->mt76, link_conf, ctx);
+	if (enable)
+		mt7925_mcu_bss_rlm_tlv(skb, phy->mt76, link_conf, ctx);
 
 	return mt76_mcu_skb_send_msg(&dev->mt76, skb,
 				     MCU_UNI_CMD(BSS_INFO_UPDATE), true);
-- 
cgit v1.2.3-58-ga151


From a5942a3979d6553c09790553c9f0a6daa986ad58 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:09 -0700
Subject: wifi: mt76: mt7925: update mt7925_mcu_sta_update for MLO

update mt7925_mcu_sta_update for the MLO-enabled firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/2e25bfc2d8aa503ecd1ba8099c6e54dce0c27b99.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 .../net/wireless/mediatek/mt76/mt76_connac_mcu.h   |  9 +++
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    | 68 +++++++++++++++++++++-
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.h    |  3 +-
 3 files changed, 78 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index e636b033b87e..9329c959674f 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -545,6 +545,13 @@ struct sta_rec_muru {
 	} mimo_ul;
 } __packed;
 
+struct sta_rec_remove {
+	__le16 tag;
+	__le16 len;
+	u8 action;
+	u8 pad[3];
+} __packed;
+
 struct sta_phy {
 	u8 type;
 	u8 flag;
@@ -814,6 +821,8 @@ enum {
 	STA_REC_HE_V2 = 0x19,
 	STA_REC_MLD = 0x20,
 	STA_REC_EHT = 0x22,
+	STA_REC_MLD_OFF = 0x23,
+	STA_REC_REMOVE = 0x25,
 	STA_REC_PN_INFO = 0x26,
 	STA_REC_KEY_V3 = 0x27,
 	STA_REC_HDRT = 0x28,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 094fa4dde5fe..159d253ba5b8 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1812,6 +1812,66 @@ mt7925_mcu_sta_cmd(struct mt76_phy *phy,
 	return mt76_mcu_skb_send_msg(dev, skb, info->cmd, true);
 }
 
+static void
+mt7925_mcu_sta_remove_tlv(struct sk_buff *skb)
+{
+	struct sta_rec_remove *rem;
+	struct tlv *tlv;
+
+	tlv = mt76_connac_mcu_add_tlv(skb, 0x25, sizeof(*rem));
+	rem = (struct sta_rec_remove *)tlv;
+	rem->action = 0;
+}
+
+static int
+mt7925_mcu_mlo_sta_cmd(struct mt76_phy *phy,
+		       struct mt76_sta_cmd_info *info)
+{
+	struct mt792x_vif *mvif = (struct mt792x_vif *)info->vif->drv_priv;
+	struct mt76_dev *dev = phy->dev;
+	struct mt792x_bss_conf *mconf;
+	struct sk_buff *skb;
+
+	mconf = mt792x_vif_to_link(mvif, info->wcid->link_id);
+
+	skb = __mt76_connac_mcu_alloc_sta_req(dev, &mconf->mt76, info->wcid,
+					      MT7925_STA_UPDATE_MAX_SIZE);
+	if (IS_ERR(skb))
+		return PTR_ERR(skb);
+
+	if (info->enable)
+		mt76_connac_mcu_sta_basic_tlv(dev, skb, info->vif,
+					      info->link_sta,
+					      info->enable, info->newly);
+
+	if (info->enable && info->link_sta) {
+		mt7925_mcu_sta_phy_tlv(skb, info->vif, info->link_sta);
+		mt7925_mcu_sta_ht_tlv(skb, info->link_sta);
+		mt7925_mcu_sta_vht_tlv(skb, info->link_sta);
+		mt76_connac_mcu_sta_uapsd(skb, info->vif, info->link_sta->sta);
+		mt7925_mcu_sta_amsdu_tlv(skb, info->vif, info->link_sta);
+		mt7925_mcu_sta_he_tlv(skb, info->link_sta);
+		mt7925_mcu_sta_he_6g_tlv(skb, info->link_sta);
+		mt7925_mcu_sta_eht_tlv(skb, info->link_sta);
+		mt7925_mcu_sta_rate_ctrl_tlv(skb, info->vif,
+					     info->link_sta);
+		mt7925_mcu_sta_state_v2_tlv(phy, skb, info->link_sta,
+					    info->vif, info->rcpi,
+					    info->state);
+		if (info->state != MT76_STA_INFO_STATE_NONE)
+			mt7925_mcu_sta_mld_tlv(skb, info->vif, info->link_sta->sta);
+		mt7925_mcu_sta_hdr_trans_tlv(skb, info->vif, info->link_sta);
+	}
+
+	if (!info->enable) {
+		mt7925_mcu_sta_remove_tlv(skb);
+		mt76_connac_mcu_add_tlv(skb, STA_REC_MLD_OFF,
+					sizeof(struct tlv));
+	}
+
+	return mt76_mcu_skb_send_msg(dev, skb, info->cmd, true);
+}
+
 int mt7925_mcu_sta_update(struct mt792x_dev *dev,
 			  struct ieee80211_link_sta *link_sta,
 			  struct ieee80211_vif *vif, bool enable,
@@ -1830,6 +1890,7 @@ int mt7925_mcu_sta_update(struct mt792x_dev *dev,
 	};
 	struct mt792x_sta *msta;
 	struct mt792x_link_sta *mlink;
+	int err;
 
 	if (link_sta) {
 		msta = (struct mt792x_sta *)link_sta->sta->drv_priv;
@@ -1838,7 +1899,12 @@ int mt7925_mcu_sta_update(struct mt792x_dev *dev,
 	info.wcid = link_sta ? &mlink->wcid : &mvif->sta.deflink.wcid;
 	info.newly = link_sta ? state != MT76_STA_INFO_STATE_ASSOC : true;
 
-	return mt7925_mcu_sta_cmd(&dev->mphy, &info);
+	if (ieee80211_vif_is_mld(vif))
+		err = mt7925_mcu_mlo_sta_cmd(&dev->mphy, &info);
+	else
+		err = mt7925_mcu_sta_cmd(&dev->mphy, &info);
+
+	return err;
 }
 
 int mt7925_mcu_set_beacon_filter(struct mt792x_dev *dev,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
index 7b0c6937f918..a97e5b539ab9 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
@@ -492,7 +492,8 @@ struct bss_rlm_tlv {
 					 sizeof(struct sta_rec_eht) +		\
 					 sizeof(struct sta_rec_hdr_trans) +	\
 					 sizeof(struct sta_rec_mld) +		\
-					 sizeof(struct tlv))
+					 sizeof(struct tlv) * 2 +		\
+					 sizeof(struct sta_rec_remove))
 
 #define MT7925_BSS_UPDATE_MAX_SIZE	(sizeof(struct bss_req_hdr) +		\
 					 sizeof(struct mt76_connac_bss_basic_tlv) +	\
-- 
cgit v1.2.3-58-ga151


From 6db8639bc813a7670b365b8ed2b77e93ef96cf67 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:10 -0700
Subject: wifi: mt76: mt7925: add mt7925_mcu_sta_eht_mld_tlv for MLO

add mt7925_mcu_sta_eht_mld_tlv for the MLO-enabled firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/a1001592eeef4e4ee2c3e15dc94cca0815d64e59.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 .../net/wireless/mediatek/mt76/mt76_connac_mcu.h   |  1 +
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    | 42 +++++++++++++++++++++-
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.h    | 11 ++++++
 3 files changed, 53 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index 9329c959674f..d6882c9fd6bc 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -820,6 +820,7 @@ enum {
 	STA_REC_HE_6G = 0x17,
 	STA_REC_HE_V2 = 0x19,
 	STA_REC_MLD = 0x20,
+	STA_REC_EHT_MLD = 0x21,
 	STA_REC_EHT = 0x22,
 	STA_REC_MLD_OFF = 0x23,
 	STA_REC_REMOVE = 0x25,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 159d253ba5b8..199070bff0f0 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1738,6 +1738,42 @@ mt7925_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb,
 		       HT_MCS_MASK_NUM);
 }
 
+static void
+mt7925_mcu_sta_eht_mld_tlv(struct sk_buff *skb,
+			   struct ieee80211_vif *vif, struct ieee80211_sta *sta)
+{
+	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
+	struct wiphy *wiphy = mvif->phy->mt76->hw->wiphy;
+	const struct wiphy_iftype_ext_capab *ext_capa;
+	struct sta_rec_eht_mld *eht_mld;
+	struct tlv *tlv;
+	u16 eml_cap;
+
+	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_EHT_MLD, sizeof(*eht_mld));
+	eht_mld = (struct sta_rec_eht_mld *)tlv;
+	eht_mld->mld_type = 0xff;
+
+	if (!ieee80211_vif_is_mld(vif))
+		return;
+
+	ext_capa = cfg80211_get_iftype_ext_capa(wiphy,
+						ieee80211_vif_type_p2p(vif));
+	if (!ext_capa)
+		return;
+
+	eml_cap = (vif->cfg.eml_cap & (IEEE80211_EML_CAP_EMLSR_SUPP |
+				       IEEE80211_EML_CAP_TRANSITION_TIMEOUT)) |
+		  (ext_capa->eml_capabilities & (IEEE80211_EML_CAP_EMLSR_PADDING_DELAY |
+						IEEE80211_EML_CAP_EMLSR_TRANSITION_DELAY));
+
+	if (eml_cap & IEEE80211_EML_CAP_EMLSR_SUPP) {
+		eht_mld->eml_cap[0] = u16_get_bits(eml_cap, GENMASK(7, 0));
+		eht_mld->eml_cap[1] = u16_get_bits(eml_cap, GENMASK(15, 8));
+	} else {
+		eht_mld->str_cap[0] = BIT(1);
+	}
+}
+
 static void
 mt7925_mcu_sta_mld_tlv(struct sk_buff *skb,
 		       struct ieee80211_vif *vif, struct ieee80211_sta *sta)
@@ -1858,8 +1894,12 @@ mt7925_mcu_mlo_sta_cmd(struct mt76_phy *phy,
 		mt7925_mcu_sta_state_v2_tlv(phy, skb, info->link_sta,
 					    info->vif, info->rcpi,
 					    info->state);
-		if (info->state != MT76_STA_INFO_STATE_NONE)
+
+		if (info->state != MT76_STA_INFO_STATE_NONE) {
 			mt7925_mcu_sta_mld_tlv(skb, info->vif, info->link_sta->sta);
+			mt7925_mcu_sta_eht_mld_tlv(skb, info->vif, info->link_sta->sta);
+		}
+
 		mt7925_mcu_sta_hdr_trans_tlv(skb, info->vif, info->link_sta);
 	}
 
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
index a97e5b539ab9..ac53bdc99332 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.h
@@ -443,6 +443,17 @@ struct sta_rec_mld {
 	} __packed link[2];
 } __packed;
 
+struct sta_rec_eht_mld {
+	__le16 tag;
+	__le16 len;
+	u8 nsep;
+	u8 mld_type;
+	u8 __rsv1[1];
+	u8 str_cap[3];
+	u8 eml_cap[3];
+	u8 __rsv2[3];
+} __packed;
+
 struct bss_ifs_time_tlv {
 	__le16 tag;
 	__le16 len;
-- 
cgit v1.2.3-58-ga151


From 677f19e161afb1096f30d993ae28140aa61a9847 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:11 -0700
Subject: wifi: mt76: mt7925: update mt7925_mcu_sta_rate_ctrl_tlv for MLO

Get band information from the per-link BSS.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/4f6ca2d082627bb93f79d5a4df80dc08998ee5d2.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 199070bff0f0..fbf43dda0cd6 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1714,12 +1714,20 @@ mt7925_mcu_sta_rate_ctrl_tlv(struct sk_buff *skb,
 			     struct ieee80211_link_sta *link_sta)
 {
 	struct mt792x_vif *mvif = (struct mt792x_vif *)vif->drv_priv;
-	struct cfg80211_chan_def *chandef = &mvif->bss_conf.mt76.ctx->def;
-	enum nl80211_band band = chandef->chan->band;
+	struct ieee80211_bss_conf *link_conf;
+	struct cfg80211_chan_def *chandef;
 	struct sta_rec_ra_info *ra_info;
+	struct mt792x_bss_conf *mconf;
+	enum nl80211_band band;
 	struct tlv *tlv;
 	u16 supp_rates;
 
+	link_conf = mt792x_vif_to_bss_conf(vif, link_sta->link_id);
+	mconf = mt792x_vif_to_link(mvif, link_sta->link_id);
+	chandef = mconf->mt76.ctx ? &mconf->mt76.ctx->def :
+				    &link_conf->chanreq.oper;
+	band = chandef->chan->band;
+
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_RA, sizeof(*ra_info));
 	ra_info = (struct sta_rec_ra_info *)tlv;
 
-- 
cgit v1.2.3-58-ga151


From beef8d1643663d501ccbd9637fc314ea79ee09ac Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:12 -0700
Subject: wifi: mt76: mt7925: update mt7925_mcu_sta_phy_tlv for MLO

when we serve MLO during the multiple-linked connection, the chanctx may
be not available in the internal driver, we should get the proper the
band information from link_conf for the MLO-enabled firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/6620e00fbeef686fb2041556deba346411eab0f1.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index fbf43dda0cd6..caf2ac238b4d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1647,7 +1647,8 @@ mt7925_mcu_sta_phy_tlv(struct sk_buff *skb,
 
 	link_conf = mt792x_vif_to_bss_conf(vif, link_sta->link_id);
 	mconf = mt792x_vif_to_link(mvif, link_sta->link_id);
-	chandef = &mconf->mt76.ctx->def;
+	chandef = mconf->mt76.ctx ? &mconf->mt76.ctx->def :
+				    &link_conf->chanreq.oper;
 
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_PHY, sizeof(*phy));
 	phy = (struct sta_rec_phy *)tlv;
-- 
cgit v1.2.3-58-ga151


From cfbe7f78346e00d33a6461f3f8d94f72c6256acb Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:13 -0700
Subject: wifi: mt76: mt7925: update mt7925_mcu_set_timing for MLO

Set the BSS index from the per-linked BSS for the firmware to identify
which link the command is operating on.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/dde970c748586d03fbecbccb10cb9903284719aa.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index caf2ac238b4d..bc6d6d957765 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2606,11 +2606,11 @@ mt7925_mcu_bss_ifs_tlv(struct sk_buff *skb,
 int mt7925_mcu_set_timing(struct mt792x_phy *phy,
 			  struct ieee80211_bss_conf *link_conf)
 {
-	struct mt792x_vif *mvif = (struct mt792x_vif *)link_conf->vif->drv_priv;
+	struct mt792x_bss_conf *mconf = mt792x_link_conf_to_mconf(link_conf);
 	struct mt792x_dev *dev = phy->dev;
 	struct sk_buff *skb;
 
-	skb = __mt7925_mcu_alloc_bss_req(&dev->mt76, &mvif->bss_conf.mt76,
+	skb = __mt7925_mcu_alloc_bss_req(&dev->mt76, &mconf->mt76,
 					 MT7925_BSS_UPDATE_MAX_SIZE);
 	if (IS_ERR(skb))
 		return PTR_ERR(skb);
-- 
cgit v1.2.3-58-ga151


From 696d9f3baf03f59584c7c46b62ae28dfd1a731ae Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:14 -0700
Subject: wifi: mt76: mt7925: update mt7925_mcu_bss_basic_tlv for MLO

update mt7925_mcu_bss_basic_tlv for the MLO-enabled firmware.

The change remains compatible with the non-MLO mode and the
older firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/583a3ac10583cfef0028f9a0b5a4802fc4070b99.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index bc6d6d957765..7bfb9c51388c 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -2385,6 +2385,7 @@ mt7925_mcu_bss_basic_tlv(struct sk_buff *skb,
 	basic_req->bcn_interval = cpu_to_le16(link_conf->beacon_int);
 	basic_req->dtim_period = link_conf->dtim_period;
 	basic_req->bmc_tx_wlan_idx = cpu_to_le16(wlan_idx);
+	basic_req->link_idx = mconf->mt76.idx;
 
 	if (link_sta) {
 		struct mt792x_sta *msta;
-- 
cgit v1.2.3-58-ga151


From 0cb349d742d17707b5b42f0db1df7f3963fc3a8a Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:15 -0700
Subject: wifi: mt76: mt7925: update mt7925_mac_link_bss_add for MLO

Update band_idx in per-link BSS to be auto for the MLO-enabled firmware
and the MLO-enabled firmware only supports omac index in 0.

The change remains compatible with the non-MLO mode and the older firmware.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/37fcf3ea7b4883840d61f97231277b297e1b15a3.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/main.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 0627ef208641..1385ef1f1c79 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -333,8 +333,9 @@ static int mt7925_mac_link_bss_add(struct mt792x_dev *dev,
 		goto out;
 	}
 
-	mconf->mt76.omac_idx = mconf->mt76.idx;
-	mconf->mt76.band_idx = 0;
+	mconf->mt76.omac_idx = ieee80211_vif_is_mld(vif) ?
+			       0 : mconf->mt76.idx;
+	mconf->mt76.band_idx = 0xff;
 	mconf->mt76.wmm_idx = mconf->mt76.idx % MT76_CONNAC_MAX_WMM_SETS;
 
 	if (mvif->phy->mt76->chandef.chan->band != NL80211_BAND_2GHZ)
-- 
cgit v1.2.3-58-ga151


From 67e9847b0208ae02efbe42dbcda4c1a9a1a853cd Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:16 -0700
Subject: wifi: mt76: mt7925: remove the unused mt7925_mcu_set_chan_info

remove the unused function mt7925_mcu_set_chan_info

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/0aa0abe7fc661380ae812c6bc879a7705de401b2.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c | 74 -------------------------
 1 file changed, 74 deletions(-)

diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 7bfb9c51388c..26f78c0a491e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -1274,80 +1274,6 @@ int mt7925_mcu_abort_roc(struct mt792x_phy *phy, struct mt792x_bss_conf *mconf,
 				 &req, sizeof(req), false);
 }
 
-int mt7925_mcu_set_chan_info(struct mt792x_phy *phy, u16 tag)
-{
-	static const u8 ch_band[] = {
-		[NL80211_BAND_2GHZ] = 0,
-		[NL80211_BAND_5GHZ] = 1,
-		[NL80211_BAND_6GHZ] = 2,
-	};
-	struct mt792x_dev *dev = phy->dev;
-	struct cfg80211_chan_def *chandef = &phy->mt76->chandef;
-	int freq1 = chandef->center_freq1;
-	u8 band_idx = chandef->chan->band != NL80211_BAND_2GHZ;
-	struct {
-		/* fixed field */
-		u8 __rsv[4];
-
-		__le16 tag;
-		__le16 len;
-		u8 control_ch;
-		u8 center_ch;
-		u8 bw;
-		u8 tx_path_num;
-		u8 rx_path;	/* mask or num */
-		u8 switch_reason;
-		u8 band_idx;
-		u8 center_ch2;	/* for 80+80 only */
-		__le16 cac_case;
-		u8 channel_band;
-		u8 rsv0;
-		__le32 outband_freq;
-		u8 txpower_drop;
-		u8 ap_bw;
-		u8 ap_center_ch;
-		u8 rsv1[53];
-	} __packed req = {
-		.tag = cpu_to_le16(tag),
-		.len = cpu_to_le16(sizeof(req) - 4),
-		.control_ch = chandef->chan->hw_value,
-		.center_ch = ieee80211_frequency_to_channel(freq1),
-		.bw = mt76_connac_chan_bw(chandef),
-		.tx_path_num = hweight8(phy->mt76->antenna_mask),
-		.rx_path = phy->mt76->antenna_mask,
-		.band_idx = band_idx,
-		.channel_band = ch_band[chandef->chan->band],
-	};
-
-	if (chandef->chan->band == NL80211_BAND_6GHZ)
-		req.channel_band = 2;
-	else
-		req.channel_band = chandef->chan->band;
-
-	if (tag == UNI_CHANNEL_RX_PATH ||
-	    dev->mt76.hw->conf.flags & IEEE80211_CONF_MONITOR)
-		req.switch_reason = CH_SWITCH_NORMAL;
-	else if (phy->mt76->hw->conf.flags & IEEE80211_CONF_OFFCHANNEL)
-		req.switch_reason = CH_SWITCH_SCAN_BYPASS_DPD;
-	else if (!cfg80211_reg_can_beacon(phy->mt76->hw->wiphy, chandef,
-					  NL80211_IFTYPE_AP))
-		req.switch_reason = CH_SWITCH_DFS;
-	else
-		req.switch_reason = CH_SWITCH_NORMAL;
-
-	if (tag == UNI_CHANNEL_SWITCH)
-		req.rx_path = hweight8(req.rx_path);
-
-	if (chandef->width == NL80211_CHAN_WIDTH_80P80) {
-		int freq2 = chandef->center_freq2;
-
-		req.center_ch2 = ieee80211_frequency_to_channel(freq2);
-	}
-
-	return mt76_mcu_send_msg(&dev->mt76, MCU_UNI_CMD(CHANNEL_SWITCH),
-				 &req, sizeof(req), true);
-}
-
 int mt7925_mcu_set_eeprom(struct mt792x_dev *dev)
 {
 	struct {
-- 
cgit v1.2.3-58-ga151


From 86c051f2c418abd7eefd8c545a01bdffd220cb95 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Sat, 6 Jul 2024 01:28:17 -0700
Subject: wifi: mt76: mt7925: enabling MLO when the firmware supports it

Register MLD capability for the firmware supporting MLO.

Co-developed-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Co-developed-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Deren Wu <deren.wu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/49c796b101e792c84bc2c0d74753022b75fd3355.1720248331.git.sean.wang@kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 .../net/wireless/mediatek/mt76/mt76_connac_mcu.h   |  1 +
 drivers/net/wireless/mediatek/mt76/mt7925/init.c   |  6 +++++
 drivers/net/wireless/mediatek/mt76/mt7925/main.c   | 29 ++++++++++++++++++++++
 drivers/net/wireless/mediatek/mt76/mt7925/mcu.c    | 20 +++++++++++++++
 drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h |  1 +
 drivers/net/wireless/mediatek/mt76/mt792x.h        |  2 ++
 6 files changed, 59 insertions(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
index d6882c9fd6bc..4242d436de26 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h
@@ -1402,6 +1402,7 @@ enum {
 	MT_NIC_CAP_WFDMA_REALLOC,
 	MT_NIC_CAP_6G,
 	MT_NIC_CAP_CHIP_CAP = 0x20,
+	MT_NIC_CAP_EML_CAP = 0x22,
 };
 
 #define UNI_WOW_DETECT_TYPE_MAGIC		BIT(0)
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/init.c b/drivers/net/wireless/mediatek/mt76/mt7925/init.c
index c4cbc8976046..039949b344b9 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/init.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/init.c
@@ -179,6 +179,12 @@ static void mt7925_init_work(struct work_struct *work)
 	mt76_set_stream_caps(&dev->mphy, true);
 	mt7925_set_stream_he_eht_caps(&dev->phy);
 
+	ret = mt7925_init_mlo_caps(&dev->phy);
+	if (ret) {
+		dev_err(dev->mt76.dev, "MLO init failed\n");
+		return;
+	}
+
 	ret = mt76_register_device(&dev->mt76, true, mt76_rates,
 				   ARRAY_SIZE(mt76_rates));
 	if (ret) {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/main.c b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
index 1385ef1f1c79..8c0768bf9343 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/main.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/main.c
@@ -236,6 +236,35 @@ mt7925_init_eht_caps(struct mt792x_phy *phy, enum nl80211_band band,
 	eht_nss->bw._160.rx_tx_mcs13_max_nss = val;
 }
 
+int mt7925_init_mlo_caps(struct mt792x_phy *phy)
+{
+	struct wiphy *wiphy = phy->mt76->hw->wiphy;
+	static const u8 ext_capa_sta[] = {
+		[7] = WLAN_EXT_CAPA8_OPMODE_NOTIF,
+	};
+	static struct wiphy_iftype_ext_capab ext_capab[] = {
+		{
+			.iftype = NL80211_IFTYPE_STATION,
+			.extended_capabilities = ext_capa_sta,
+			.extended_capabilities_mask = ext_capa_sta,
+			.extended_capabilities_len = sizeof(ext_capa_sta),
+		},
+	};
+
+	if (!(phy->chip_cap & MT792x_CHIP_CAP_MLO_EVT_EN))
+		return 0;
+
+	ext_capab[0].eml_capabilities = phy->eml_cap;
+	ext_capab[0].mld_capa_and_ops =
+		u16_encode_bits(1, IEEE80211_MLD_CAP_OP_MAX_SIMUL_LINKS);
+
+	wiphy->flags |= WIPHY_FLAG_SUPPORTS_MLO;
+	wiphy->iftype_ext_capab = ext_capab;
+	wiphy->num_iftype_ext_capab = ARRAY_SIZE(ext_capab);
+
+	return 0;
+}
+
 static void
 __mt7925_set_stream_he_eht_caps(struct mt792x_phy *phy,
 				struct ieee80211_supported_band *sband,
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
index 26f78c0a491e..9dc22fbe25d3 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mcu.c
@@ -751,6 +751,20 @@ mt7925_mcu_parse_phy_cap(struct mt792x_dev *dev, char *data)
 	dev->has_eht = cap->eht;
 }
 
+static void
+mt7925_mcu_parse_eml_cap(struct mt792x_dev *dev, char *data)
+{
+	struct mt7925_mcu_eml_cap {
+		u8 rsv[4];
+		__le16 eml_cap;
+		u8 rsv2[6];
+	} __packed * cap;
+
+	cap = (struct mt7925_mcu_eml_cap *)data;
+
+	dev->phy.eml_cap = le16_to_cpu(cap->eml_cap);
+}
+
 static int
 mt7925_mcu_get_nic_capability(struct mt792x_dev *dev)
 {
@@ -805,6 +819,12 @@ mt7925_mcu_get_nic_capability(struct mt792x_dev *dev)
 		case MT_NIC_CAP_PHY:
 			mt7925_mcu_parse_phy_cap(dev, tlv->data);
 			break;
+		case MT_NIC_CAP_CHIP_CAP:
+			memcpy(&dev->phy.chip_cap, (void *)skb->data, sizeof(u64));
+			break;
+		case MT_NIC_CAP_EML_CAP:
+			mt7925_mcu_parse_eml_cap(dev, tlv->data);
+			break;
 		default:
 			break;
 		}
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
index 1ae66629e39f..669f3a079d04 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/mt7925.h
@@ -235,6 +235,7 @@ void mt7925_queue_rx_skb(struct mt76_dev *mdev, enum mt76_rxq_id q,
 			 struct sk_buff *skb, u32 *info);
 void mt7925_stats_work(struct work_struct *work);
 void mt7925_set_stream_he_eht_caps(struct mt792x_phy *phy);
+int mt7925_init_mlo_caps(struct mt792x_phy *phy);
 int mt7925_init_debugfs(struct mt792x_dev *dev);
 
 int mt7925_mcu_set_beacon_filter(struct mt792x_dev *dev,
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index 5ede24116748..30635aeba363 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -27,6 +27,7 @@
 
 #define MT792x_CHIP_CAP_CLC_EVT_EN BIT(0)
 #define MT792x_CHIP_CAP_RSSI_NOTIFY_EVT_EN BIT(1)
+#define MT792x_CHIP_CAP_MLO_EVT_EN BIT(2)
 
 /* NOTE: used to map mt76_rates. idx may change if firmware expands table */
 #define MT792x_BASIC_RATES_TBL	11
@@ -163,6 +164,7 @@ struct mt792x_phy {
 #endif
 	void *clc[MT792x_CLC_MAX_NUM];
 	u64 chip_cap;
+	u16 eml_cap;
 
 	struct work_struct roc_work;
 	struct timer_list roc_timer;
-- 
cgit v1.2.3-58-ga151


From d53ab629cff57e450fe69fc90eb1ddc372e8da2d Mon Sep 17 00:00:00 2001
From: Michael Lo <michael.lo@mediatek.com>
Date: Thu, 23 May 2024 19:21:31 +0800
Subject: wifi: mt76: mt792x: fix scheduler interference in drv own process

Add some time to wait for LP engine to complete its operation
before polling pmctrl register.

Signed-off-by: Michael Lo <michael.lo@mediatek.com>
Signed-off-by: Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
Tested-by: David Ruth <druth@chromium.org>
Acked-by: Sean Wang <sean.wang@mediatek.com>
Link: https://patch.msgid.link/20240523112131.31437-1-mingyen.hsieh@mediatek.com
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/wireless/mediatek/mt76/mt76.h        |  1 +
 drivers/net/wireless/mediatek/mt76/mt7921/pci.c  |  3 +++
 drivers/net/wireless/mediatek/mt76/mt7925/pci.c  |  3 +++
 drivers/net/wireless/mediatek/mt76/mt792x.h      |  1 +
 drivers/net/wireless/mediatek/mt76/mt792x_core.c |  4 ++++
 drivers/net/wireless/mediatek/mt76/pci.c         | 23 +++++++++++++++++++++++
 6 files changed, 35 insertions(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h
index 5f8773364d51..4a58a78d5ed2 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76.h
@@ -1085,6 +1085,7 @@ bool ____mt76_poll_msec(struct mt76_dev *dev, u32 offset, u32 mask, u32 val,
 
 void mt76_mmio_init(struct mt76_dev *dev, void __iomem *regs);
 void mt76_pci_disable_aspm(struct pci_dev *pdev);
+bool mt76_pci_aspm_supported(struct pci_dev *pdev);
 
 static inline u16 mt76_chip(struct mt76_dev *dev)
 {
diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
index e75e7b6d3aaf..a7430216a80d 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
@@ -339,6 +339,9 @@ static int mt7921_pci_probe(struct pci_dev *pdev,
 	bus_ops->rmw = mt7921_rmw;
 	dev->mt76.bus = bus_ops;
 
+	if (!mt7921_disable_aspm && mt76_pci_aspm_supported(pdev))
+		dev->aspm_supported = true;
+
 	ret = mt792xe_mcu_fw_pmctrl(dev);
 	if (ret)
 		goto err_free_dev;
diff --git a/drivers/net/wireless/mediatek/mt76/mt7925/pci.c b/drivers/net/wireless/mediatek/mt76/mt7925/pci.c
index 577574fb7a1e..6e4f4e78c350 100644
--- a/drivers/net/wireless/mediatek/mt76/mt7925/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt7925/pci.c
@@ -373,6 +373,9 @@ static int mt7925_pci_probe(struct pci_dev *pdev,
 	bus_ops->rmw = mt7925_rmw;
 	dev->mt76.bus = bus_ops;
 
+	if (!mt7925_disable_aspm && mt76_pci_aspm_supported(pdev))
+		dev->aspm_supported = true;
+
 	ret = __mt792x_mcu_fw_pmctrl(dev);
 	if (ret)
 		goto err_free_dev;
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x.h b/drivers/net/wireless/mediatek/mt76/mt792x.h
index 30635aeba363..7fa74d59cc48 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x.h
+++ b/drivers/net/wireless/mediatek/mt76/mt792x.h
@@ -215,6 +215,7 @@ struct mt792x_dev {
 	bool fw_assert:1;
 	bool has_eht:1;
 	bool regd_in_progress:1;
+	bool aspm_supported:1;
 	wait_queue_head_t wait;
 
 	struct work_struct init_work;
diff --git a/drivers/net/wireless/mediatek/mt76/mt792x_core.c b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
index ae2ac146a7f7..78fe37c2e07b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt792x_core.c
+++ b/drivers/net/wireless/mediatek/mt76/mt792x_core.c
@@ -809,6 +809,10 @@ int __mt792xe_mcu_drv_pmctrl(struct mt792x_dev *dev)
 
 	for (i = 0; i < MT792x_DRV_OWN_RETRY_COUNT; i++) {
 		mt76_wr(dev, MT_CONN_ON_LPCTL, PCIE_LPCR_HOST_CLR_OWN);
+
+		if (dev->aspm_supported)
+			usleep_range(2000, 3000);
+
 		if (mt76_poll_msec_tick(dev, MT_CONN_ON_LPCTL,
 					PCIE_LPCR_HOST_OWN_SYNC, 0, 50, 1))
 			break;
diff --git a/drivers/net/wireless/mediatek/mt76/pci.c b/drivers/net/wireless/mediatek/mt76/pci.c
index 4c1c159fbb62..b5031ca7f73f 100644
--- a/drivers/net/wireless/mediatek/mt76/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/pci.c
@@ -45,3 +45,26 @@ void mt76_pci_disable_aspm(struct pci_dev *pdev)
 					   aspm_conf);
 }
 EXPORT_SYMBOL_GPL(mt76_pci_disable_aspm);
+
+bool mt76_pci_aspm_supported(struct pci_dev *pdev)
+{
+	struct pci_dev *parent = pdev->bus->self;
+	u16 aspm_conf, parent_aspm_conf = 0;
+	bool result = true;
+
+	pcie_capability_read_word(pdev, PCI_EXP_LNKCTL, &aspm_conf);
+	aspm_conf &= PCI_EXP_LNKCTL_ASPMC;
+	if (parent) {
+		pcie_capability_read_word(parent, PCI_EXP_LNKCTL,
+					  &parent_aspm_conf);
+		parent_aspm_conf &= PCI_EXP_LNKCTL_ASPMC;
+	}
+
+	if (!aspm_conf && (!parent || !parent_aspm_conf)) {
+		/* aspm already disabled */
+		result = false;
+	}
+
+	return result;
+}
+EXPORT_SYMBOL_GPL(mt76_pci_aspm_supported);
-- 
cgit v1.2.3-58-ga151


From a0aab7d7c8605f53ea77dfaafec1dcc0bfb1f232 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 8 Jul 2024 14:36:23 -0700
Subject: selftests: drv-net: rss_ctx: fix cleanup in the basic test

The basic test may fail without resetting the RSS indir table.
Use the .exec() method to run cleanup early since we re-test
with traffic that returning to default state works.
While at it reformat the doc a tiny bit.

Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240708213627.226025-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/hw/rss_ctx.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
index 475f2a63fcd5..de2a55c0f35c 100755
--- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py
+++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
@@ -64,9 +64,8 @@ def _get_rx_cnts(cfg, prev=None):
 
 
 def test_rss_key_indir(cfg):
-    """
-    Test basics like updating the main RSS key and indirection table.
-    """
+    """Test basics like updating the main RSS key and indirection table."""
+
     if len(_get_rx_cnts(cfg)) < 2:
         KsftSkipEx("Device has only one queue (or doesn't support queue stats)")
 
@@ -89,6 +88,7 @@ def test_rss_key_indir(cfg):
 
     # Set the indirection table
     ethtool(f"-X {cfg.ifname} equal 2")
+    reset_indir = defer(ethtool, f"-X {cfg.ifname} default")
     data = get_rss(cfg)
     ksft_eq(0, min(data['rss-indirection-table']))
     ksft_eq(1, max(data['rss-indirection-table']))
@@ -104,7 +104,7 @@ def test_rss_key_indir(cfg):
     ksft_eq(sum(cnts[2:]), 0, "traffic on unused queues: " + str(cnts))
 
     # Restore, and check traffic gets spread again
-    ethtool(f"-X {cfg.ifname} default")
+    reset_indir.exec()
 
     cnts = _get_rx_cnts(cfg)
     GenerateTraffic(cfg).wait_pkts_and_stop(20000)
-- 
cgit v1.2.3-58-ga151


From 847aa551fa7897dbe457bfd8badd4ca22d44add1 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 8 Jul 2024 14:36:24 -0700
Subject: selftests: drv-net: rss_ctx: factor out send traffic and check

Wrap up sending traffic and checking in which queues it landed
in a helper.

The method used for testing is to send a lot of iperf traffic
and check which queues received the most packets. Those should
be the queues where we expect iperf to land - either because we
installed a filter for the port iperf uses, or we didn't and
expect it to use context 0.

Contexts get disjoint queue sets, but the main context (AKA context 0)
may receive some background traffic (noise).

Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240708213627.226025-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/hw/rss_ctx.py | 58 +++++++++++++++--------
 1 file changed, 39 insertions(+), 19 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
index de2a55c0f35c..ede8eee1d9a9 100755
--- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py
+++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
@@ -63,6 +63,30 @@ def _get_rx_cnts(cfg, prev=None):
     return queue_stats
 
 
+def _send_traffic_check(cfg, port, name, params):
+    # params is a dict with 3 possible keys:
+    #  - "target": required, which queues we expect to get iperf traffic
+    #  - "empty": optional, which queues should see no traffic at all
+    #  - "noise": optional, which queues we expect to see low traffic;
+    #             used for queues of the main context, since some background
+    #             OS activity may use those queues while we're testing
+    # the value for each is a list, or some other iterable containing queue ids.
+
+    cnts = _get_rx_cnts(cfg)
+    GenerateTraffic(cfg, port=port).wait_pkts_and_stop(20000)
+    cnts = _get_rx_cnts(cfg, prev=cnts)
+
+    directed = sum(cnts[i] for i in params['target'])
+
+    ksft_ge(directed, 20000, f"traffic on {name}: " + str(cnts))
+    if params.get('noise'):
+        ksft_lt(sum(cnts[i] for i in params['noise']), directed / 2,
+                "traffic on other queues:" + str(cnts))
+    if params.get('empty'):
+        ksft_eq(sum(cnts[i] for i in params['empty']), 0,
+                "traffic on inactive queues: " + str(cnts))
+
+
 def test_rss_key_indir(cfg):
     """Test basics like updating the main RSS key and indirection table."""
 
@@ -170,15 +194,10 @@ def test_rss_context(cfg, ctx_cnt=1, create_with_cfg=None):
         defer(ethtool, f"-N {cfg.ifname} delete {ntuple}")
 
     for i in range(ctx_cnt):
-        cnts = _get_rx_cnts(cfg)
-        GenerateTraffic(cfg, port=ports[i]).wait_pkts_and_stop(20000)
-        cnts = _get_rx_cnts(cfg, prev=cnts)
-
-        directed = sum(cnts[2+i*2:4+i*2])
-
-        ksft_lt(sum(cnts[ :2]), directed / 2, "traffic on main context:" + str(cnts))
-        ksft_ge(directed, 20000, f"traffic on context {i}: " + str(cnts))
-        ksft_eq(sum(cnts[2:2+i*2] + cnts[4+i*2:]), 0, "traffic on other contexts: " + str(cnts))
+        _send_traffic_check(cfg, ports[i], f"context {i}",
+                            { 'target': (2+i*2, 3+i*2),
+                              'noise': (0, 1),
+                              'empty': list(range(2, 2+i*2)) + list(range(4+i*2, 2+2*ctx_cnt)) })
 
     if requested_ctx_cnt != ctx_cnt:
         raise KsftSkipEx(f"Tested only {ctx_cnt} contexts, wanted {requested_ctx_cnt}")
@@ -230,18 +249,19 @@ def test_rss_context_out_of_order(cfg, ctx_cnt=4):
 
     def check_traffic():
         for i in range(ctx_cnt):
-            cnts = _get_rx_cnts(cfg)
-            GenerateTraffic(cfg, port=ports[i]).wait_pkts_and_stop(20000)
-            cnts = _get_rx_cnts(cfg, prev=cnts)
-
             if ctx[i]:
-                directed = sum(cnts[2+i*2:4+i*2])
-                ksft_lt(sum(cnts[ :2]), directed / 2, "traffic on main context:" + str(cnts))
-                ksft_ge(directed, 20000, f"traffic on context {i}: " + str(cnts))
-                ksft_eq(sum(cnts[2:2+i*2] + cnts[4+i*2:]), 0, "traffic on other contexts: " + str(cnts))
+                expected = {
+                    'target': (2+i*2, 3+i*2),
+                    'noise': (0, 1),
+                    'empty': list(range(2, 2+i*2)) + list(range(4+i*2, 2+2*ctx_cnt))
+                }
             else:
-                ksft_ge(sum(cnts[ :2]), 20000, "traffic on main context:" + str(cnts))
-                ksft_eq(sum(cnts[2: ]),     0, "traffic on other contexts: " + str(cnts))
+                expected = {
+                    'target': (0, 1),
+                    'empty':  range(2, 2+2*ctx_cnt)
+                }
+
+            _send_traffic_check(cfg, ports[i], f"context {i}", expected)
 
     # Use queues 0 and 1 for normal traffic
     ethtool(f"-X {cfg.ifname} equal 2")
-- 
cgit v1.2.3-58-ga151


From e2c9703d424e7a7a60f325cb16ddcd45b2213c70 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 8 Jul 2024 14:36:25 -0700
Subject: selftests: drv-net: rss_ctx: test queue changes vs user RSS config

By default main RSS table should change to include all queues.
When user sets a specific RSS config the driver should preserve it,
even when queue count changes. Driver should refuse to deactivate
queues used in the user-set RSS config.

For additional contexts driver should still refuse to deactivate
queues in use. Whether the contexts should get resized like
context 0 when queue count increases is a bit unclear. I anticipate
most drivers today don't do that. Since main use case for additional
contexts is to set the indir table - it doesn't seem worthwhile to
care about behavior of the default table too much. Don't test that.

Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240708213627.226025-4-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/hw/rss_ctx.py | 81 ++++++++++++++++++++++-
 1 file changed, 80 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
index ede8eee1d9a9..177abfd06412 100755
--- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py
+++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
@@ -137,6 +137,80 @@ def test_rss_key_indir(cfg):
     ksft_lt(sum(cnts[:2]), sum(cnts[2:]), "traffic distributed: " + str(cnts))
 
 
+def test_rss_queue_reconfigure(cfg, main_ctx=True):
+    """Make sure queue changes can't override requested RSS config.
+
+    By default main RSS table should change to include all queues.
+    When user sets a specific RSS config the driver should preserve it,
+    even when queue count changes. Driver should refuse to deactivate
+    queues used in the user-set RSS config.
+    """
+
+    if not main_ctx:
+        require_ntuple(cfg)
+
+    # Start with 4 queues, an arbitrary known number.
+    try:
+        qcnt = len(_get_rx_cnts(cfg))
+        ethtool(f"-L {cfg.ifname} combined 4")
+        defer(ethtool, f"-L {cfg.ifname} combined {qcnt}")
+    except:
+        raise KsftSkipEx("Not enough queues for the test or qstat not supported")
+
+    if main_ctx:
+        ctx_id = 0
+        ctx_ref = ""
+    else:
+        ctx_id = ethtool_create(cfg, "-X", "context new")
+        ctx_ref = f"context {ctx_id}"
+        defer(ethtool, f"-X {cfg.ifname} {ctx_ref} delete")
+
+    # Indirection table should be distributing to all queues.
+    data = get_rss(cfg, context=ctx_id)
+    ksft_eq(0, min(data['rss-indirection-table']))
+    ksft_eq(3, max(data['rss-indirection-table']))
+
+    # Increase queues, indirection table should be distributing to all queues.
+    # It's unclear whether tables of additional contexts should be reset, too.
+    if main_ctx:
+        ethtool(f"-L {cfg.ifname} combined 5")
+        data = get_rss(cfg)
+        ksft_eq(0, min(data['rss-indirection-table']))
+        ksft_eq(4, max(data['rss-indirection-table']))
+        ethtool(f"-L {cfg.ifname} combined 4")
+
+    # Configure the table explicitly
+    port = rand_port()
+    ethtool(f"-X {cfg.ifname} {ctx_ref} weight 1 0 0 1")
+    if main_ctx:
+        other_key = 'empty'
+        defer(ethtool, f"-X {cfg.ifname} default")
+    else:
+        other_key = 'noise'
+        flow = f"flow-type tcp{cfg.addr_ipver} dst-port {port} context {ctx_id}"
+        ntuple = ethtool_create(cfg, "-N", flow)
+        defer(ethtool, f"-N {cfg.ifname} delete {ntuple}")
+
+    _send_traffic_check(cfg, port, ctx_ref, { 'target': (0, 3),
+                                              other_key: (1, 2) })
+
+    # We should be able to increase queues, but table should be left untouched
+    ethtool(f"-L {cfg.ifname} combined 5")
+    data = get_rss(cfg, context=ctx_id)
+    ksft_eq({0, 3}, set(data['rss-indirection-table']))
+
+    _send_traffic_check(cfg, port, ctx_ref, { 'target': (0, 3),
+                                              other_key: (1, 2, 4) })
+
+    # Setting queue count to 3 should fail, queue 3 is used
+    try:
+        ethtool(f"-L {cfg.ifname} combined 3")
+    except CmdExitFailure:
+        pass
+    else:
+        raise Exception(f"Driver didn't prevent us from deactivating a used queue (context {ctx_id})")
+
+
 def test_rss_context(cfg, ctx_cnt=1, create_with_cfg=None):
     """
     Test separating traffic into RSS contexts.
@@ -215,6 +289,10 @@ def test_rss_context4_create_with_cfg(cfg):
     test_rss_context(cfg, 4, create_with_cfg=True)
 
 
+def test_rss_context_queue_reconfigure(cfg):
+    test_rss_queue_reconfigure(cfg, main_ctx=False)
+
+
 def test_rss_context_out_of_order(cfg, ctx_cnt=4):
     """
     Test separating traffic into RSS contexts.
@@ -366,8 +444,9 @@ def main() -> None:
     with NetDrvEpEnv(__file__, nsim_test=False) as cfg:
         cfg.netdevnl = NetdevFamily()
 
-        ksft_run([test_rss_key_indir,
+        ksft_run([test_rss_key_indir, test_rss_queue_reconfigure,
                   test_rss_context, test_rss_context4, test_rss_context32,
+                  test_rss_context_queue_reconfigure,
                   test_rss_context_overlap, test_rss_context_overlap2,
                   test_rss_context_out_of_order, test_rss_context4_create_with_cfg],
                  args=(cfg, ))
-- 
cgit v1.2.3-58-ga151


From 7e3e5b0bc51d03c452e1f752b7656bbc4df87ee2 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 8 Jul 2024 14:36:26 -0700
Subject: selftests: drv-net: rss_ctx: check behavior of indirection table
 resizing

Some devices dynamically increase and decrease the size of the RSS
indirection table based on the number of enabled queues.
When that happens driver must maintain the balance of entries
(preferably duplicating the smaller table).

Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240708213627.226025-5-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/hw/rss_ctx.py | 37 ++++++++++++++++++++++-
 1 file changed, 36 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
index 177abfd06412..38a871220bff 100755
--- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py
+++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
@@ -5,7 +5,7 @@ import datetime
 import random
 from lib.py import ksft_run, ksft_pr, ksft_exit, ksft_eq, ksft_ge, ksft_lt
 from lib.py import NetDrvEpEnv
-from lib.py import NetdevFamily
+from lib.py import EthtoolFamily, NetdevFamily
 from lib.py import KsftSkipEx
 from lib.py import rand_port
 from lib.py import ethtool, ip, defer, GenerateTraffic, CmdExitFailure
@@ -211,6 +211,39 @@ def test_rss_queue_reconfigure(cfg, main_ctx=True):
         raise Exception(f"Driver didn't prevent us from deactivating a used queue (context {ctx_id})")
 
 
+def test_rss_resize(cfg):
+    """Test resizing of the RSS table.
+
+    Some devices dynamically increase and decrease the size of the RSS
+    indirection table based on the number of enabled queues.
+    When that happens driver must maintain the balance of entries
+    (preferably duplicating the smaller table).
+    """
+
+    channels = cfg.ethnl.channels_get({'header': {'dev-index': cfg.ifindex}})
+    ch_max = channels['combined-max']
+    qcnt = channels['combined-count']
+
+    if ch_max < 2:
+        raise KsftSkipEx(f"Not enough queues for the test: {ch_max}")
+
+    ethtool(f"-L {cfg.ifname} combined 2")
+    defer(ethtool, f"-L {cfg.ifname} combined {qcnt}")
+
+    ethtool(f"-X {cfg.ifname} weight 1 7")
+    defer(ethtool, f"-X {cfg.ifname} default")
+
+    ethtool(f"-L {cfg.ifname} combined {ch_max}")
+    data = get_rss(cfg)
+    ksft_eq(0, min(data['rss-indirection-table']))
+    ksft_eq(1, max(data['rss-indirection-table']))
+
+    ksft_eq(7,
+            data['rss-indirection-table'].count(1) /
+            data['rss-indirection-table'].count(0),
+            f"Table imbalance after resize: {data['rss-indirection-table']}")
+
+
 def test_rss_context(cfg, ctx_cnt=1, create_with_cfg=None):
     """
     Test separating traffic into RSS contexts.
@@ -442,9 +475,11 @@ def test_rss_context_overlap2(cfg):
 
 def main() -> None:
     with NetDrvEpEnv(__file__, nsim_test=False) as cfg:
+        cfg.ethnl = EthtoolFamily()
         cfg.netdevnl = NetdevFamily()
 
         ksft_run([test_rss_key_indir, test_rss_queue_reconfigure,
+                  test_rss_resize,
                   test_rss_context, test_rss_context4, test_rss_context32,
                   test_rss_context_queue_reconfigure,
                   test_rss_context_overlap, test_rss_context_overlap2,
-- 
cgit v1.2.3-58-ga151


From 933048fec4dda42a5e3fdee26653db6d5904d727 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 8 Jul 2024 14:36:27 -0700
Subject: selftests: drv-net: rss_ctx: test flow rehashing without impacting
 traffic

Some workloads may want to rehash the flows in response to an imbalance.
Most effective way to do that is changing the RSS key. Check that changing
the key does not cause link flaps or traffic disruption.

Disrupting traffic for key update is not incorrect, but makes the key
update unusable for rehashing under load.

Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240708213627.226025-6-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/drivers/net/hw/rss_ctx.py | 32 ++++++++++++++++++++++-
 1 file changed, 31 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
index 38a871220bff..931dbc36ca43 100755
--- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py
+++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py
@@ -244,6 +244,36 @@ def test_rss_resize(cfg):
             f"Table imbalance after resize: {data['rss-indirection-table']}")
 
 
+def test_hitless_key_update(cfg):
+    """Test that flows may be rehashed without impacting traffic.
+
+    Some workloads may want to rehash the flows in response to an imbalance.
+    Most effective way to do that is changing the RSS key. Check that changing
+    the key does not cause link flaps or traffic disruption.
+
+    Disrupting traffic for key update is not a bug, but makes the key
+    update unusable for rehashing under load.
+    """
+    data = get_rss(cfg)
+    key_len = len(data['rss-hash-key'])
+
+    key = _rss_key_rand(key_len)
+
+    tgen = GenerateTraffic(cfg)
+    try:
+        errors0, carrier0 = get_drop_err_sum(cfg)
+        t0 = datetime.datetime.now()
+        ethtool(f"-X {cfg.ifname} hkey " + _rss_key_str(key))
+        t1 = datetime.datetime.now()
+        errors1, carrier1 = get_drop_err_sum(cfg)
+    finally:
+        tgen.wait_pkts_and_stop(5000)
+
+    ksft_lt((t1 - t0).total_seconds(), 0.2)
+    ksft_eq(errors1 - errors1, 0)
+    ksft_eq(carrier1 - carrier0, 0)
+
+
 def test_rss_context(cfg, ctx_cnt=1, create_with_cfg=None):
     """
     Test separating traffic into RSS contexts.
@@ -479,7 +509,7 @@ def main() -> None:
         cfg.netdevnl = NetdevFamily()
 
         ksft_run([test_rss_key_indir, test_rss_queue_reconfigure,
-                  test_rss_resize,
+                  test_rss_resize, test_hitless_key_update,
                   test_rss_context, test_rss_context4, test_rss_context32,
                   test_rss_context_queue_reconfigure,
                   test_rss_context_overlap, test_rss_context_overlap2,
-- 
cgit v1.2.3-58-ga151


From b45c76e5f43fa66868e874a42a29bd0288a5ddce Mon Sep 17 00:00:00 2001
From: Petr Machata <petrm@nvidia.com>
Date: Mon, 8 Jul 2024 16:23:40 +0200
Subject: mlxsw: Warn about invalid accesses to array fields

A forgotten or buggy variable initialization can cause out-of-bounds access
to a register or other item array field. For an overflow, such access would
mangle adjacent parts of the register payload. For an underflow, due to all
variables being unsigned, the access would likely trample unrelated memory.
Since neither is correct, replace these accesses with accesses at the index
of 0, and warn about the issue.

Suggested-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/b988fb265c2f6c1206fe12d5bfdcfa188b7672d1.1720447210.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/item.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/item.h b/drivers/net/ethernet/mellanox/mlxsw/item.h
index cfafbeb42586..a619a0736bd1 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/item.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/item.h
@@ -218,6 +218,10 @@ __mlxsw_item_bit_array_offset(const struct mlxsw_item *item,
 	}
 
 	max_index = (item->size.bytes << 3) / item->element_size - 1;
+	if (WARN_ONCE(index > max_index,
+		      "name=%s,index=%u,max_index=%u\n", item->name, index,
+		      max_index))
+		index = 0;
 	be_index = max_index - index;
 	offset = be_index * item->element_size >> 3;
 	in_byte_index  = index % (BITS_PER_BYTE / item->element_size);
-- 
cgit v1.2.3-58-ga151


From a22f3bc80075c10c3835b9ec1c7da17496386904 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Mon, 8 Jul 2024 16:23:41 +0200
Subject: mlxsw: core_thermal: Report valid current state during cooling device
 registration

Commit 31a0fa0019b0 ("thermal/debugfs: Pass cooling device state to
thermal_debug_cdev_add()") changed the thermal core to read the current
state of the cooling device as part of the cooling device's
registration. This is incompatible with the current implementation of
the cooling device operations in mlxsw, leading to initialization
failure with errors such as:

mlxsw_spectrum 0000:01:00.0: Failed to register cooling device
mlxsw_spectrum 0000:01:00.0: cannot register bus device

The reason for the failure is that when the get current state operation
is invoked the driver tries to derive the index of the cooling device by
walking a per thermal zone array and looking for the matching cooling
device pointer. However, the pointer is returned from the registration
function and therefore only set in the array after the registration.

The issue was later fixed by commit 1af89dedc8a5 ("thermal: core: Do not
fail cdev registration because of invalid initial state") by not failing
the registration of the cooling device if it cannot report a valid
current state during registration, although drivers are responsible for
ensuring that this will not happen.

Therefore, make sure the driver is able to report a valid current state
for the cooling device during registration by passing to the
registration function a per cooling device private data that already has
the cooling device index populated.

While at it, call thermal_cooling_device_unregister() unconditionally
since the function returns immediately if the cooling device pointer is
NULL.

Reviewed-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/c823c4678b6b7afb902c35b3551c81a053afd110.1720447210.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/core_thermal.c | 51 +++++++++++-----------
 1 file changed, 25 insertions(+), 26 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
index 5c511e1a8efa..d61478c0c632 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
@@ -100,6 +100,12 @@ static const struct mlxsw_cooling_states default_cooling_states[] = {
 
 struct mlxsw_thermal;
 
+struct mlxsw_thermal_cooling_device {
+	struct mlxsw_thermal *thermal;
+	struct thermal_cooling_device *cdev;
+	unsigned int idx;
+};
+
 struct mlxsw_thermal_module {
 	struct mlxsw_thermal *parent;
 	struct thermal_zone_device *tzdev;
@@ -123,7 +129,7 @@ struct mlxsw_thermal {
 	const struct mlxsw_bus_info *bus_info;
 	struct thermal_zone_device *tzdev;
 	int polling_delay;
-	struct thermal_cooling_device *cdevs[MLXSW_MFCR_PWMS_MAX];
+	struct mlxsw_thermal_cooling_device cdevs[MLXSW_MFCR_PWMS_MAX];
 	struct thermal_trip trips[MLXSW_THERMAL_NUM_TRIPS];
 	struct mlxsw_cooling_states cooling_states[MLXSW_THERMAL_NUM_TRIPS];
 	struct mlxsw_thermal_area line_cards[];
@@ -147,7 +153,7 @@ static int mlxsw_get_cooling_device_idx(struct mlxsw_thermal *thermal,
 	int i;
 
 	for (i = 0; i < MLXSW_MFCR_PWMS_MAX; i++)
-		if (thermal->cdevs[i] == cdev)
+		if (thermal->cdevs[i].cdev == cdev)
 			return i;
 
 	/* Allow mlxsw thermal zone binding to an external cooling device */
@@ -352,17 +358,14 @@ static int mlxsw_thermal_get_cur_state(struct thermal_cooling_device *cdev,
 				       unsigned long *p_state)
 
 {
-	struct mlxsw_thermal *thermal = cdev->devdata;
+	struct mlxsw_thermal_cooling_device *mlxsw_cdev = cdev->devdata;
+	struct mlxsw_thermal *thermal = mlxsw_cdev->thermal;
 	struct device *dev = thermal->bus_info->dev;
 	char mfsc_pl[MLXSW_REG_MFSC_LEN];
-	int err, idx;
 	u8 duty;
+	int err;
 
-	idx = mlxsw_get_cooling_device_idx(thermal, cdev);
-	if (idx < 0)
-		return idx;
-
-	mlxsw_reg_mfsc_pack(mfsc_pl, idx, 0);
+	mlxsw_reg_mfsc_pack(mfsc_pl, mlxsw_cdev->idx, 0);
 	err = mlxsw_reg_query(thermal->core, MLXSW_REG(mfsc), mfsc_pl);
 	if (err) {
 		dev_err(dev, "Failed to query PWM duty\n");
@@ -378,22 +381,19 @@ static int mlxsw_thermal_set_cur_state(struct thermal_cooling_device *cdev,
 				       unsigned long state)
 
 {
-	struct mlxsw_thermal *thermal = cdev->devdata;
+	struct mlxsw_thermal_cooling_device *mlxsw_cdev = cdev->devdata;
+	struct mlxsw_thermal *thermal = mlxsw_cdev->thermal;
 	struct device *dev = thermal->bus_info->dev;
 	char mfsc_pl[MLXSW_REG_MFSC_LEN];
-	int idx;
 	int err;
 
 	if (state > MLXSW_THERMAL_MAX_STATE)
 		return -EINVAL;
 
-	idx = mlxsw_get_cooling_device_idx(thermal, cdev);
-	if (idx < 0)
-		return idx;
-
 	/* Normalize the state to the valid speed range. */
 	state = max_t(unsigned long, MLXSW_THERMAL_MIN_STATE, state);
-	mlxsw_reg_mfsc_pack(mfsc_pl, idx, mlxsw_state_to_duty(state));
+	mlxsw_reg_mfsc_pack(mfsc_pl, mlxsw_cdev->idx,
+			    mlxsw_state_to_duty(state));
 	err = mlxsw_reg_write(thermal->core, MLXSW_REG(mfsc), mfsc_pl);
 	if (err) {
 		dev_err(dev, "Failed to write PWM duty\n");
@@ -753,17 +753,21 @@ int mlxsw_thermal_init(struct mlxsw_core *core,
 	}
 	for (i = 0; i < MLXSW_MFCR_PWMS_MAX; i++) {
 		if (pwm_active & BIT(i)) {
+			struct mlxsw_thermal_cooling_device *mlxsw_cdev;
 			struct thermal_cooling_device *cdev;
 
+			mlxsw_cdev = &thermal->cdevs[i];
+			mlxsw_cdev->thermal = thermal;
+			mlxsw_cdev->idx = i;
 			cdev = thermal_cooling_device_register("mlxsw_fan",
-							       thermal,
+							       mlxsw_cdev,
 							       &mlxsw_cooling_ops);
 			if (IS_ERR(cdev)) {
 				err = PTR_ERR(cdev);
 				dev_err(dev, "Failed to register cooling device\n");
 				goto err_thermal_cooling_device_register;
 			}
-			thermal->cdevs[i] = cdev;
+			mlxsw_cdev->cdev = cdev;
 		}
 	}
 
@@ -824,8 +828,7 @@ err_thermal_modules_init:
 err_thermal_zone_device_register:
 err_thermal_cooling_device_register:
 	for (i = 0; i < MLXSW_MFCR_PWMS_MAX; i++)
-		if (thermal->cdevs[i])
-			thermal_cooling_device_unregister(thermal->cdevs[i]);
+		thermal_cooling_device_unregister(thermal->cdevs[i].cdev);
 err_reg_write:
 err_reg_query:
 	kfree(thermal);
@@ -847,12 +850,8 @@ void mlxsw_thermal_fini(struct mlxsw_thermal *thermal)
 		thermal->tzdev = NULL;
 	}
 
-	for (i = 0; i < MLXSW_MFCR_PWMS_MAX; i++) {
-		if (thermal->cdevs[i]) {
-			thermal_cooling_device_unregister(thermal->cdevs[i]);
-			thermal->cdevs[i] = NULL;
-		}
-	}
+	for (i = 0; i < MLXSW_MFCR_PWMS_MAX; i++)
+		thermal_cooling_device_unregister(thermal->cdevs[i].cdev);
 
 	kfree(thermal);
 }
-- 
cgit v1.2.3-58-ga151


From 0970836c348b6bc2ea77ce4348a136d6febfd440 Mon Sep 17 00:00:00 2001
From: Ido Schimmel <idosch@nvidia.com>
Date: Mon, 8 Jul 2024 16:23:42 +0200
Subject: mlxsw: pci: Lock configuration space of upstream bridge during reset

The driver triggers a "Secondary Bus Reset" (SBR) by calling
__pci_reset_function_locked() which asserts the SBR bit in the "Bridge
Control Register" in the configuration space of the upstream bridge for
2ms. This is done without locking the configuration space of the
upstream bridge port, allowing user space to access it concurrently.

Linux 6.11 will start warning about such unlocked resets [1][2]:

pcieport 0000:00:01.0: unlocked secondary bus reset via: pci_reset_bus_function+0x51c/0x6a0

Avoid the warning and the concurrent access by locking the configuration
space of the upstream bridge prior to the reset and unlocking it
afterwards.

[1] https://lore.kernel.org/all/171711746953.1628941.4692125082286867825.stgit@dwillia2-xfh.jf.intel.com/
[2] https://lore.kernel.org/all/20240531213150.GA610983@bhelgaas/

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Link: https://patch.msgid.link/9937b0afdb50f2f2825945393c94c093c04a5897.1720447210.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlxsw/pci.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index 0320dabd1380..060e5b939211 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -1784,6 +1784,7 @@ static int mlxsw_pci_reset_at_pci_disable(struct mlxsw_pci *mlxsw_pci,
 {
 	struct pci_dev *pdev = mlxsw_pci->pdev;
 	char mrsr_pl[MLXSW_REG_MRSR_LEN];
+	struct pci_dev *bridge;
 	int err;
 
 	if (!pci_reset_sbr_supported) {
@@ -1800,6 +1801,9 @@ static int mlxsw_pci_reset_at_pci_disable(struct mlxsw_pci *mlxsw_pci,
 sbr:
 	device_lock_assert(&pdev->dev);
 
+	bridge = pci_upstream_bridge(pdev);
+	if (bridge)
+		pci_cfg_access_lock(bridge);
 	pci_cfg_access_lock(pdev);
 	pci_save_state(pdev);
 
@@ -1809,6 +1813,8 @@ sbr:
 
 	pci_restore_state(pdev);
 	pci_cfg_access_unlock(pdev);
+	if (bridge)
+		pci_cfg_access_unlock(bridge);
 
 	return err;
 }
-- 
cgit v1.2.3-58-ga151


From e829a331ec28d194f1e2d7ea218c6be0a6be0e56 Mon Sep 17 00:00:00 2001
From: Yevgeny Kliteynik <kliteyn@nvidia.com>
Date: Mon, 8 Jul 2024 11:00:23 +0300
Subject: net/mlx5: DR, Remove definer functions from SW Steering API

No need to expose definer get/put functions as part of
SW Steering API - they are internal functions.

Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20240708080025.1593555-9-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h | 5 +++++
 drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h   | 5 -----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 81eff6c410ce..7618c6147f86 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -1379,6 +1379,11 @@ int mlx5dr_cmd_create_modify_header_arg(struct mlx5_core_dev *dev,
 void mlx5dr_cmd_destroy_modify_header_arg(struct mlx5_core_dev *dev,
 					  u32 obj_id);
 
+int mlx5dr_definer_get(struct mlx5dr_domain *dmn, u16 format_id,
+		       u8 *dw_selectors, u8 *byte_selectors,
+		       u8 *match_mask, u32 *definer_id);
+void mlx5dr_definer_put(struct mlx5dr_domain *dmn, u32 definer_id);
+
 struct mlx5dr_icm_pool *mlx5dr_icm_pool_create(struct mlx5dr_domain *dmn,
 					       enum mlx5dr_icm_type icm_type);
 void mlx5dr_icm_pool_destroy(struct mlx5dr_icm_pool *pool);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
index 89fced86936f..3ac7dc67509f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
@@ -153,11 +153,6 @@ int mlx5dr_action_destroy(struct mlx5dr_action *action);
 
 u32 mlx5dr_action_get_pkt_reformat_id(struct mlx5dr_action *action);
 
-int mlx5dr_definer_get(struct mlx5dr_domain *dmn, u16 format_id,
-		       u8 *dw_selectors, u8 *byte_selectors,
-		       u8 *match_mask, u32 *definer_id);
-void mlx5dr_definer_put(struct mlx5dr_domain *dmn, u32 definer_id);
-
 static inline bool
 mlx5dr_is_supported(struct mlx5_core_dev *dev)
 {
-- 
cgit v1.2.3-58-ga151


From 7204730b330460a3622bdbe038c492272d4d9a00 Mon Sep 17 00:00:00 2001
From: Dragos Tatulea <dtatulea@nvidia.com>
Date: Mon, 8 Jul 2024 11:00:24 +0300
Subject: net/mlx5e: SHAMPO, Add missing aggregate counter

When the rx_hds_nodata_packets/bytes counters were added, the aggregate
counters were omitted. This patch adds them.

Fixes: e95c5b9e8912 ("net/mlx5e: SHAMPO, Add header-only ethtool counters for header data split")
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20240708080025.1593555-10-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index db1cac68292f..e7a3290a708a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -142,6 +142,8 @@ static const struct counter_desc sw_stats_desc[] = {
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_gro_bytes) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_gro_skbs) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_gro_large_hds) },
+	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_hds_nodata_packets) },
+	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_hds_nodata_bytes) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_ecn_mark) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_removed_vlan_packets) },
 	{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_csum_unnecessary) },
-- 
cgit v1.2.3-58-ga151


From f1ac0b7dcd49edf3ea5cc9345fdae70b6c2b199b Mon Sep 17 00:00:00 2001
From: Cosmin Ratiu <cratiu@nvidia.com>
Date: Mon, 8 Jul 2024 11:00:25 +0300
Subject: net/mlx5e: CT: Initialize err to 0 to avoid warning

It is theoretically possible to return bogus uninitialized values from
mlx5_tc_ct_entry_replace_rules, even though in practice this will never
be the case as the flow rule will be part of at least the regular ct
table or the ct nat table, if not both.

But to reduce noise, initialize err to 0.

Fixes: 49d37d05f216 ("net/mlx5: CT: Separate CT and CT-NAT tuple entries")
Signed-off-by: Cosmin Ratiu <cratiu@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20240708080025.1593555-11-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
index b49d87a51f21..8cf8ba2622f2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
@@ -1145,7 +1145,7 @@ mlx5_tc_ct_entry_replace_rules(struct mlx5_tc_ct_priv *ct_priv,
 			       struct mlx5_ct_entry *entry,
 			       u8 zone_restore_id)
 {
-	int err;
+	int err = 0;
 
 	if (mlx5_tc_ct_entry_in_ct_table(entry)) {
 		err = mlx5_tc_ct_entry_replace_rule(ct_priv, flow_rule, entry, false,
-- 
cgit v1.2.3-58-ga151


From 40ab9e0dc86535acebe78ee3d38c576bc932df78 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Mon, 8 Jul 2024 11:25:14 +0200
Subject: netxen_nic: Use {low,upp}er_32_bits() helpers

Use the existing {low,upp}er_32_bits() helpers instead of defining
custom variants.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/319d4a5313ac75f7bbbb6b230b6802b18075c3e0.1720430602.git.geert+renesas@glider.be
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/qlogic/netxen/netxen_nic_ctx.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_ctx.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_ctx.c
index 2fcbcecb41d1..fef4b2b0b1f2 100644
--- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_ctx.c
+++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_ctx.c
@@ -571,9 +571,6 @@ static u64 ctx_addr_sig_regs[][3] = {
 #define CRB_CTX_ADDR_REG_HI(FUNC_ID)	(ctx_addr_sig_regs[FUNC_ID][2])
 #define CRB_CTX_SIGNATURE_REG(FUNC_ID)	(ctx_addr_sig_regs[FUNC_ID][1])
 
-#define lower32(x)	((u32)((x) & 0xffffffff))
-#define upper32(x)	((u32)(((u64)(x) >> 32) & 0xffffffff))
-
 static struct netxen_recv_crb recv_crb_registers[] = {
 	/* Instance 0 */
 	{
@@ -723,9 +720,9 @@ netxen_init_old_ctx(struct netxen_adapter *adapter)
 		NETXEN_CTX_SIGNATURE_V2 : NETXEN_CTX_SIGNATURE;
 
 	NXWR32(adapter, CRB_CTX_ADDR_REG_LO(port),
-			lower32(recv_ctx->phys_addr));
+			lower_32_bits(recv_ctx->phys_addr));
 	NXWR32(adapter, CRB_CTX_ADDR_REG_HI(port),
-			upper32(recv_ctx->phys_addr));
+			upper_32_bits(recv_ctx->phys_addr));
 	NXWR32(adapter, CRB_CTX_SIGNATURE_REG(port),
 			signature | port);
 	return 0;
-- 
cgit v1.2.3-58-ga151


From 06e71ad534881d2a09ced7509d2ab0daedac4c96 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 8 Jul 2024 13:45:38 -0700
Subject: bpftool: improve skeleton backwards compat with old buggy libbpfs

Old versions of libbpf don't handle varying sizes of bpf_map_skeleton
struct correctly. As such, BPF skeleton generated by newest bpftool
might not be compatible with older libbpf (though only when libbpf is
used as a shared library), even though it, by design, should.

Going forward libbpf will be fixed, plus we'll release bug fixed
versions of relevant old libbpfs, but meanwhile try to mitigate from
bpftool side by conservatively assuming older and smaller definition of
bpf_map_skeleton, if possible. Meaning, if there are no struct_ops maps.

If there are struct_ops, then presumably user would like to have
auto-attaching logic and struct_ops map link placeholders, so use the
full bpf_map_skeleton definition in that case.

Acked-by: Quentin Monnet <qmo@kernel.org>
Co-developed-by: Mykyta Yatsenko <yatsenko@meta.com>
Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20240708204540.4188946-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/bpf/bpftool/gen.c | 46 ++++++++++++++++++++++++++++++++--------------
 1 file changed, 32 insertions(+), 14 deletions(-)

diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index 51eaed76db97..5a4d3240689e 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -852,24 +852,41 @@ codegen_maps_skeleton(struct bpf_object *obj, size_t map_cnt, bool mmaped, bool
 {
 	struct bpf_map *map;
 	char ident[256];
-	size_t i;
+	size_t i, map_sz;
 
 	if (!map_cnt)
 		return;
 
+	/* for backward compatibility with old libbpf versions that don't
+	 * handle new BPF skeleton with new struct bpf_map_skeleton definition
+	 * that includes link field, avoid specifying new increased size,
+	 * unless we absolutely have to (i.e., if there are struct_ops maps
+	 * present)
+	 */
+	map_sz = offsetof(struct bpf_map_skeleton, link);
+	if (populate_links) {
+		bpf_object__for_each_map(map, obj) {
+			if (bpf_map__type(map) == BPF_MAP_TYPE_STRUCT_OPS) {
+				map_sz = sizeof(struct bpf_map_skeleton);
+				break;
+			}
+		}
+	}
+
 	codegen("\
 		\n\
-									\n\
+								    \n\
 			/* maps */				    \n\
 			s->map_cnt = %zu;			    \n\
-			s->map_skel_sz = sizeof(*s->maps);	    \n\
-			s->maps = (struct bpf_map_skeleton *)calloc(s->map_cnt, s->map_skel_sz);\n\
+			s->map_skel_sz = %zu;			    \n\
+			s->maps = (struct bpf_map_skeleton *)calloc(s->map_cnt,\n\
+					sizeof(*s->maps) > %zu ? sizeof(*s->maps) : %zu);\n\
 			if (!s->maps) {				    \n\
 				err = -ENOMEM;			    \n\
 				goto err;			    \n\
 			}					    \n\
 		",
-		map_cnt
+		map_cnt, map_sz, map_sz, map_sz
 	);
 	i = 0;
 	bpf_object__for_each_map(map, obj) {
@@ -878,23 +895,22 @@ codegen_maps_skeleton(struct bpf_object *obj, size_t map_cnt, bool mmaped, bool
 
 		codegen("\
 			\n\
-									\n\
-				s->maps[%zu].name = \"%s\";	    \n\
-				s->maps[%zu].map = &obj->maps.%s;   \n\
+								    \n\
+				map = (struct bpf_map_skeleton *)((char *)s->maps + %zu * s->map_skel_sz);\n\
+				map->name = \"%s\";		    \n\
+				map->map = &obj->maps.%s;	    \n\
 			",
-			i, bpf_map__name(map), i, ident);
+			i, bpf_map__name(map), ident);
 		/* memory-mapped internal maps */
 		if (mmaped && is_mmapable_map(map, ident, sizeof(ident))) {
-			printf("\ts->maps[%zu].mmaped = (void **)&obj->%s;\n",
-				i, ident);
+			printf("\tmap->mmaped = (void **)&obj->%s;\n", ident);
 		}
 
 		if (populate_links && bpf_map__type(map) == BPF_MAP_TYPE_STRUCT_OPS) {
 			codegen("\
 				\n\
-					s->maps[%zu].link = &obj->links.%s;\n\
-				",
-				i, ident);
+					map->link = &obj->links.%s; \n\
+				", ident);
 		}
 		i++;
 	}
@@ -1463,6 +1479,7 @@ static int do_skeleton(int argc, char **argv)
 		%1$s__create_skeleton(struct %1$s *obj)			    \n\
 		{							    \n\
 			struct bpf_object_skeleton *s;			    \n\
+			struct bpf_map_skeleton *map __attribute__((unused));\n\
 			int err;					    \n\
 									    \n\
 			s = (struct bpf_object_skeleton *)calloc(1, sizeof(*s));\n\
@@ -1753,6 +1770,7 @@ static int do_subskeleton(int argc, char **argv)
 		{							    \n\
 			struct %1$s *obj;				    \n\
 			struct bpf_object_subskeleton *s;		    \n\
+			struct bpf_map_skeleton *map __attribute__((unused));\n\
 			int err;					    \n\
 									    \n\
 			obj = (struct %1$s *)calloc(1, sizeof(*obj));	    \n\
-- 
cgit v1.2.3-58-ga151


From 99fb9531886d8ffa0aa9a693089784c7338518a3 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 8 Jul 2024 13:45:39 -0700
Subject: libbpf: fix BPF skeleton forward/backward compat handling

BPF skeleton was designed from day one to be extensible. Generated BPF
skeleton code specifies actual sizes of map/prog/variable skeletons for
that reason and libbpf is supposed to work with newer/older versions
correctly.

Unfortunately, it was missed that we implicitly embed hard-coded most
up-to-date (according to libbpf's version of libbpf.h header used to
compile BPF skeleton header) sizes of those structs, which can differ
from the actual sizes at runtime when libbpf is used as a shared
library.

We have a few places were we just index array of maps/progs/vars, which
implicitly uses these potentially invalid sizes of structs.

This patch aims to fix this problem going forward. Once this lands,
we'll backport these changes in Github repo to create patched releases
for older libbpfs.

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Reviewed-by: Alan Maguire <alan.maguire@oracle.com>
Fixes: d66562fba1ce ("libbpf: Add BPF object skeleton support")
Fixes: 430025e5dca5 ("libbpf: Add subskeleton scaffolding")
Fixes: 08ac454e258e ("libbpf: Auto-attach struct_ops BPF maps in BPF skeleton")
Co-developed-by: Mykyta Yatsenko <yatsenko@meta.com>
Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20240708204540.4188946-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/lib/bpf/libbpf.c | 47 +++++++++++++++++++++++++++--------------------
 1 file changed, 27 insertions(+), 20 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 30f121754d83..8625c948f60a 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -13712,14 +13712,15 @@ int libbpf_num_possible_cpus(void)
 
 static int populate_skeleton_maps(const struct bpf_object *obj,
 				  struct bpf_map_skeleton *maps,
-				  size_t map_cnt)
+				  size_t map_cnt, size_t map_skel_sz)
 {
 	int i;
 
 	for (i = 0; i < map_cnt; i++) {
-		struct bpf_map **map = maps[i].map;
-		const char *name = maps[i].name;
-		void **mmaped = maps[i].mmaped;
+		struct bpf_map_skeleton *map_skel = (void *)maps + i * map_skel_sz;
+		struct bpf_map **map = map_skel->map;
+		const char *name = map_skel->name;
+		void **mmaped = map_skel->mmaped;
 
 		*map = bpf_object__find_map_by_name(obj, name);
 		if (!*map) {
@@ -13736,13 +13737,14 @@ static int populate_skeleton_maps(const struct bpf_object *obj,
 
 static int populate_skeleton_progs(const struct bpf_object *obj,
 				   struct bpf_prog_skeleton *progs,
-				   size_t prog_cnt)
+				   size_t prog_cnt, size_t prog_skel_sz)
 {
 	int i;
 
 	for (i = 0; i < prog_cnt; i++) {
-		struct bpf_program **prog = progs[i].prog;
-		const char *name = progs[i].name;
+		struct bpf_prog_skeleton *prog_skel = (void *)progs + i * prog_skel_sz;
+		struct bpf_program **prog = prog_skel->prog;
+		const char *name = prog_skel->name;
 
 		*prog = bpf_object__find_program_by_name(obj, name);
 		if (!*prog) {
@@ -13783,13 +13785,13 @@ int bpf_object__open_skeleton(struct bpf_object_skeleton *s,
 	}
 
 	*s->obj = obj;
-	err = populate_skeleton_maps(obj, s->maps, s->map_cnt);
+	err = populate_skeleton_maps(obj, s->maps, s->map_cnt, s->map_skel_sz);
 	if (err) {
 		pr_warn("failed to populate skeleton maps for '%s': %d\n", s->name, err);
 		return libbpf_err(err);
 	}
 
-	err = populate_skeleton_progs(obj, s->progs, s->prog_cnt);
+	err = populate_skeleton_progs(obj, s->progs, s->prog_cnt, s->prog_skel_sz);
 	if (err) {
 		pr_warn("failed to populate skeleton progs for '%s': %d\n", s->name, err);
 		return libbpf_err(err);
@@ -13819,20 +13821,20 @@ int bpf_object__open_subskeleton(struct bpf_object_subskeleton *s)
 		return libbpf_err(-errno);
 	}
 
-	err = populate_skeleton_maps(s->obj, s->maps, s->map_cnt);
+	err = populate_skeleton_maps(s->obj, s->maps, s->map_cnt, s->map_skel_sz);
 	if (err) {
 		pr_warn("failed to populate subskeleton maps: %d\n", err);
 		return libbpf_err(err);
 	}
 
-	err = populate_skeleton_progs(s->obj, s->progs, s->prog_cnt);
+	err = populate_skeleton_progs(s->obj, s->progs, s->prog_cnt, s->prog_skel_sz);
 	if (err) {
 		pr_warn("failed to populate subskeleton maps: %d\n", err);
 		return libbpf_err(err);
 	}
 
 	for (var_idx = 0; var_idx < s->var_cnt; var_idx++) {
-		var_skel = &s->vars[var_idx];
+		var_skel = (void *)s->vars + var_idx * s->var_skel_sz;
 		map = *var_skel->map;
 		map_type_id = bpf_map__btf_value_type_id(map);
 		map_type = btf__type_by_id(btf, map_type_id);
@@ -13879,10 +13881,11 @@ int bpf_object__load_skeleton(struct bpf_object_skeleton *s)
 	}
 
 	for (i = 0; i < s->map_cnt; i++) {
-		struct bpf_map *map = *s->maps[i].map;
+		struct bpf_map_skeleton *map_skel = (void *)s->maps + i * s->map_skel_sz;
+		struct bpf_map *map = *map_skel->map;
 		size_t mmap_sz = bpf_map_mmap_sz(map);
 		int prot, map_fd = map->fd;
-		void **mmaped = s->maps[i].mmaped;
+		void **mmaped = map_skel->mmaped;
 
 		if (!mmaped)
 			continue;
@@ -13930,8 +13933,9 @@ int bpf_object__attach_skeleton(struct bpf_object_skeleton *s)
 	int i, err;
 
 	for (i = 0; i < s->prog_cnt; i++) {
-		struct bpf_program *prog = *s->progs[i].prog;
-		struct bpf_link **link = s->progs[i].link;
+		struct bpf_prog_skeleton *prog_skel = (void *)s->progs + i * s->prog_skel_sz;
+		struct bpf_program *prog = *prog_skel->prog;
+		struct bpf_link **link = prog_skel->link;
 
 		if (!prog->autoload || !prog->autoattach)
 			continue;
@@ -13970,8 +13974,9 @@ int bpf_object__attach_skeleton(struct bpf_object_skeleton *s)
 		return 0;
 
 	for (i = 0; i < s->map_cnt; i++) {
-		struct bpf_map *map = *s->maps[i].map;
-		struct bpf_link **link = s->maps[i].link;
+		struct bpf_map_skeleton *map_skel = (void *)s->maps + i * s->map_skel_sz;
+		struct bpf_map *map = *map_skel->map;
+		struct bpf_link **link = map_skel->link;
 
 		if (!map->autocreate || !map->autoattach)
 			continue;
@@ -14000,7 +14005,8 @@ void bpf_object__detach_skeleton(struct bpf_object_skeleton *s)
 	int i;
 
 	for (i = 0; i < s->prog_cnt; i++) {
-		struct bpf_link **link = s->progs[i].link;
+		struct bpf_prog_skeleton *prog_skel = (void *)s->progs + i * s->prog_skel_sz;
+		struct bpf_link **link = prog_skel->link;
 
 		bpf_link__destroy(*link);
 		*link = NULL;
@@ -14010,7 +14016,8 @@ void bpf_object__detach_skeleton(struct bpf_object_skeleton *s)
 		return;
 
 	for (i = 0; i < s->map_cnt; i++) {
-		struct bpf_link **link = s->maps[i].link;
+		struct bpf_map_skeleton *map_skel = (void *)s->maps + i * s->map_skel_sz;
+		struct bpf_link **link = map_skel->link;
 
 		if (link) {
 			bpf_link__destroy(*link);
-- 
cgit v1.2.3-58-ga151


From a459f4bb27f2e2730039c57786b82288742c8c74 Mon Sep 17 00:00:00 2001
From: Andrii Nakryiko <andrii@kernel.org>
Date: Mon, 8 Jul 2024 13:45:40 -0700
Subject: libbpf: improve old BPF skeleton handling for map auto-attach

Improve how we handle old BPF skeletons when it comes to BPF map
auto-attachment. Emit one warn-level message per each struct_ops map
that could have been auto-attached, if user provided recent enough BPF
skeleton version. Don't spam log if there are no relevant struct_ops
maps, though.

This should help users realize that they probably need to regenerate BPF
skeleton header with more recent bpftool/libbpf-cargo (or whatever other
means of BPF skeleton generation).

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20240708204540.4188946-4-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/lib/bpf/libbpf.c | 26 ++++++++++++++------------
 1 file changed, 14 insertions(+), 12 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 8625c948f60a..a3be6f8fac09 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -13967,32 +13967,34 @@ int bpf_object__attach_skeleton(struct bpf_object_skeleton *s)
 		 */
 	}
 
-	/* Skeleton is created with earlier version of bpftool
-	 * which does not support auto-attachment
-	 */
-	if (s->map_skel_sz < sizeof(struct bpf_map_skeleton))
-		return 0;
 
 	for (i = 0; i < s->map_cnt; i++) {
 		struct bpf_map_skeleton *map_skel = (void *)s->maps + i * s->map_skel_sz;
 		struct bpf_map *map = *map_skel->map;
-		struct bpf_link **link = map_skel->link;
+		struct bpf_link **link;
 
 		if (!map->autocreate || !map->autoattach)
 			continue;
 
-		if (*link)
-			continue;
-
 		/* only struct_ops maps can be attached */
 		if (!bpf_map__is_struct_ops(map))
 			continue;
-		*link = bpf_map__attach_struct_ops(map);
 
+		/* skeleton is created with earlier version of bpftool, notify user */
+		if (s->map_skel_sz < offsetofend(struct bpf_map_skeleton, link)) {
+			pr_warn("map '%s': BPF skeleton version is old, skipping map auto-attachment...\n",
+				bpf_map__name(map));
+			continue;
+		}
+
+		link = map_skel->link;
+		if (*link)
+			continue;
+
+		*link = bpf_map__attach_struct_ops(map);
 		if (!*link) {
 			err = -errno;
-			pr_warn("map '%s': failed to auto-attach: %d\n",
-				bpf_map__name(map), err);
+			pr_warn("map '%s': failed to auto-attach: %d\n", bpf_map__name(map), err);
 			return libbpf_err(err);
 		}
 	}
-- 
cgit v1.2.3-58-ga151


From 605c96997d89c01c11bbddb4db820ede570581c7 Mon Sep 17 00:00:00 2001
From: Matt Bobrowski <mattbobrowski@google.com>
Date: Tue, 9 Jul 2024 21:09:39 +0000
Subject: bpf: relax zero fixed offset constraint on KF_TRUSTED_ARGS/KF_RCU

Currently, BPF kfuncs which accept trusted pointer arguments
i.e. those flagged as KF_TRUSTED_ARGS, KF_RCU, or KF_RELEASE, all
require an original/unmodified trusted pointer argument to be supplied
to them. By original/unmodified, it means that the backing register
holding the trusted pointer argument that is to be supplied to the BPF
kfunc must have its fixed offset set to zero, or else the BPF verifier
will outright reject the BPF program load. However, this zero fixed
offset constraint that is currently enforced by the BPF verifier onto
BPF kfuncs specifically flagged to accept KF_TRUSTED_ARGS or KF_RCU
trusted pointer arguments is rather unnecessary, and can limit their
usability in practice. Specifically, it completely eliminates the
possibility of constructing a derived trusted pointer from an original
trusted pointer. To put it simply, a derived pointer is a pointer
which points to one of the nested member fields of the object being
pointed to by the original trusted pointer.

This patch relaxes the zero fixed offset constraint that is enforced
upon BPF kfuncs which specifically accept KF_TRUSTED_ARGS, or KF_RCU
arguments. Although, the zero fixed offset constraint technically also
applies to BPF kfuncs accepting KF_RELEASE arguments, relaxing this
constraint for such BPF kfuncs has subtle and unwanted
side-effects. This was discovered by experimenting a little further
with an initial version of this patch series [0]. The primary issue
with relaxing the zero fixed offset constraint on BPF kfuncs accepting
KF_RELEASE arguments is that it'd would open up the opportunity for
BPF programs to supply both trusted pointers and derived trusted
pointers to them. For KF_RELEASE BPF kfuncs specifically, this could
be problematic as resources associated with the backing pointer could
be released by the backing BPF kfunc and cause instabilities for the
rest of the kernel.

With this new fixed offset semantic in-place for BPF kfuncs accepting
KF_TRUSTED_ARGS and KF_RCU arguments, we now have more flexibility
when it comes to the BPF kfuncs that we're able to introduce moving
forward.

Early discussions covering the possibility of relaxing the zero fixed
offset constraint can be found using the link below. This will provide
more context on where all this has stemmed from [1].

Notably, pre-existing tests have been updated such that they provide
coverage for the updated zero fixed offset
functionality. Specifically, the nested offset test was converted from
a negative to positive test as it was already designed to assert zero
fixed offset semantics of a KF_TRUSTED_ARGS BPF kfunc.

[0] https://lore.kernel.org/bpf/ZnA9ndnXKtHOuYMe@google.com/
[1] https://lore.kernel.org/bpf/ZhkbrM55MKQ0KeIV@google.com/

Signed-off-by: Matt Bobrowski <mattbobrowski@google.com>
Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20240709210939.1544011-1-mattbobrowski@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c                                    | 9 +++------
 tools/testing/selftests/bpf/progs/nested_trust_failure.c | 8 --------
 tools/testing/selftests/bpf/progs/nested_trust_success.c | 8 ++++++++
 tools/testing/selftests/bpf/verifier/calls.c             | 2 +-
 4 files changed, 12 insertions(+), 15 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 3d6306c363b7..c0263fb5ca4b 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -11335,7 +11335,9 @@ static int process_kf_arg_ptr_to_btf_id(struct bpf_verifier_env *env,
 	    btf_type_ids_nocast_alias(&env->log, reg_btf, reg_ref_id, meta->btf, ref_id))
 		strict_type_match = true;
 
-	WARN_ON_ONCE(is_kfunc_trusted_args(meta) && reg->off);
+	WARN_ON_ONCE(is_kfunc_release(meta) &&
+		     (reg->off || !tnum_is_const(reg->var_off) ||
+		      reg->var_off.value));
 
 	reg_ref_t = btf_type_skip_modifiers(reg_btf, reg_ref_id, &reg_ref_id);
 	reg_ref_tname = btf_name_by_offset(reg_btf, reg_ref_t->name_off);
@@ -11917,12 +11919,8 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 					return -EINVAL;
 				}
 			}
-
 			fallthrough;
 		case KF_ARG_PTR_TO_CTX:
-			/* Trusted arguments have the same offset checks as release arguments */
-			arg_type |= OBJ_RELEASE;
-			break;
 		case KF_ARG_PTR_TO_DYNPTR:
 		case KF_ARG_PTR_TO_ITER:
 		case KF_ARG_PTR_TO_LIST_HEAD:
@@ -11935,7 +11933,6 @@ static int check_kfunc_args(struct bpf_verifier_env *env, struct bpf_kfunc_call_
 		case KF_ARG_PTR_TO_REFCOUNTED_KPTR:
 		case KF_ARG_PTR_TO_CONST_STR:
 		case KF_ARG_PTR_TO_WORKQUEUE:
-			/* Trusted by default */
 			break;
 		default:
 			WARN_ON_ONCE(1);
diff --git a/tools/testing/selftests/bpf/progs/nested_trust_failure.c b/tools/testing/selftests/bpf/progs/nested_trust_failure.c
index ea39497f11ed..3568ec450100 100644
--- a/tools/testing/selftests/bpf/progs/nested_trust_failure.c
+++ b/tools/testing/selftests/bpf/progs/nested_trust_failure.c
@@ -31,14 +31,6 @@ int BPF_PROG(test_invalid_nested_user_cpus, struct task_struct *task, u64 clone_
 	return 0;
 }
 
-SEC("tp_btf/task_newtask")
-__failure __msg("R1 must have zero offset when passed to release func or trusted arg to kfunc")
-int BPF_PROG(test_invalid_nested_offset, struct task_struct *task, u64 clone_flags)
-{
-	bpf_cpumask_first_zero(&task->cpus_mask);
-	return 0;
-}
-
 /* Although R2 is of type sk_buff but sock_common is expected, we will hit untrusted ptr first. */
 SEC("tp_btf/tcp_probe")
 __failure __msg("R2 type=untrusted_ptr_ expected=ptr_, trusted_ptr_, rcu_ptr_")
diff --git a/tools/testing/selftests/bpf/progs/nested_trust_success.c b/tools/testing/selftests/bpf/progs/nested_trust_success.c
index 833840bffd3b..2b66953ca82e 100644
--- a/tools/testing/selftests/bpf/progs/nested_trust_success.c
+++ b/tools/testing/selftests/bpf/progs/nested_trust_success.c
@@ -32,3 +32,11 @@ int BPF_PROG(test_skb_field, struct sock *sk, struct sk_buff *skb)
 	bpf_sk_storage_get(&sk_storage_map, skb->sk, 0, 0);
 	return 0;
 }
+
+SEC("tp_btf/task_newtask")
+__success
+int BPF_PROG(test_nested_offset, struct task_struct *task, u64 clone_flags)
+{
+	bpf_cpumask_first_zero(&task->cpus_mask);
+	return 0;
+}
diff --git a/tools/testing/selftests/bpf/verifier/calls.c b/tools/testing/selftests/bpf/verifier/calls.c
index ab25a81fd3a1..d76ef2018859 100644
--- a/tools/testing/selftests/bpf/verifier/calls.c
+++ b/tools/testing/selftests/bpf/verifier/calls.c
@@ -76,7 +76,7 @@
 	},
 	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
 	.result = REJECT,
-	.errstr = "R1 must have zero offset when passed to release func or trusted arg to kfunc",
+	.errstr = "arg#0 expected pointer to ctx, but got PTR",
 	.fixup_kfunc_btf_id = {
 		{ "bpf_kfunc_call_test_pass_ctx", 2 },
 	},
-- 
cgit v1.2.3-58-ga151


From 663117327a39d1ad8be79d3c6451173c37e97c56 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Mon, 8 Jul 2024 09:50:20 +0200
Subject: net: phy: aquantia: rename and export aqr107_wait_reset_complete()

This function is quite generic in this driver and not limited to aqr107.
We will use it outside its current compilation unit soon so rename it
and declare it in the header.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/aquantia/aquantia.h      | 2 ++
 drivers/net/phy/aquantia/aquantia_main.c | 6 +++---
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/net/phy/aquantia/aquantia.h b/drivers/net/phy/aquantia/aquantia.h
index 0bca28ec3332..2465345081f8 100644
--- a/drivers/net/phy/aquantia/aquantia.h
+++ b/drivers/net/phy/aquantia/aquantia.h
@@ -201,4 +201,6 @@ int aqr_phy_led_hw_control_set(struct phy_device *phydev, u8 index,
 int aqr_phy_led_active_low_set(struct phy_device *phydev, int index, bool enable);
 int aqr_phy_led_polarity_set(struct phy_device *phydev, int index,
 			     unsigned long modes);
+int aqr_wait_reset_complete(struct phy_device *phydev);
+
 #endif /* AQUANTIA_H */
diff --git a/drivers/net/phy/aquantia/aquantia_main.c b/drivers/net/phy/aquantia/aquantia_main.c
index 6c14355744b7..974795bd0860 100644
--- a/drivers/net/phy/aquantia/aquantia_main.c
+++ b/drivers/net/phy/aquantia/aquantia_main.c
@@ -441,7 +441,7 @@ static int aqr107_set_tunable(struct phy_device *phydev,
  * The chip also provides a "reset completed" bit, but it's cleared after
  * read. Therefore function would time out if called again.
  */
-static int aqr107_wait_reset_complete(struct phy_device *phydev)
+int aqr_wait_reset_complete(struct phy_device *phydev)
 {
 	int val;
 
@@ -494,7 +494,7 @@ static int aqr107_config_init(struct phy_device *phydev)
 	WARN(phydev->interface == PHY_INTERFACE_MODE_XGMII,
 	     "Your devicetree is out of date, please update it. The AQR107 family doesn't support XGMII, maybe you mean USXGMII.\n");
 
-	ret = aqr107_wait_reset_complete(phydev);
+	ret = aqr_wait_reset_complete(phydev);
 	if (!ret)
 		aqr107_chip_info(phydev);
 
@@ -522,7 +522,7 @@ static int aqcs109_config_init(struct phy_device *phydev)
 	    phydev->interface != PHY_INTERFACE_MODE_2500BASEX)
 		return -ENODEV;
 
-	ret = aqr107_wait_reset_complete(phydev);
+	ret = aqr_wait_reset_complete(phydev);
 	if (!ret)
 		aqr107_chip_info(phydev);
 
-- 
cgit v1.2.3-58-ga151


From ad649a1fac37e390de5b4211b902ec666c4cd202 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Mon, 8 Jul 2024 09:50:21 +0200
Subject: net: phy: aquantia: wait for FW reset before checking the vendor ID

Checking the firmware register before it complete the boot process makes
no sense, it will report 0 even if FW is available from internal memory.
Always wait for FW to boot before continuing or we'll unnecessarily try
to load it from nvmem/filesystem and fail.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/aquantia/aquantia_firmware.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/phy/aquantia/aquantia_firmware.c b/drivers/net/phy/aquantia/aquantia_firmware.c
index 0c9640ef153b..524627a36c6f 100644
--- a/drivers/net/phy/aquantia/aquantia_firmware.c
+++ b/drivers/net/phy/aquantia/aquantia_firmware.c
@@ -353,6 +353,10 @@ int aqr_firmware_load(struct phy_device *phydev)
 {
 	int ret;
 
+	ret = aqr_wait_reset_complete(phydev);
+	if (ret)
+		return ret;
+
 	/* Check if the firmware is not already loaded by pooling
 	 * the current version returned by the PHY. If 0 is returned,
 	 * no firmware is loaded.
-- 
cgit v1.2.3-58-ga151


From 708405f3e56e3bee086f46a43398ee300c36a1d8 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Mon, 8 Jul 2024 09:50:22 +0200
Subject: net: phy: aquantia: wait for the GLOBAL_CFG to start returning real
 values

When the PHY is first coming up (or resuming from suspend), it's
possible that although the FW status shows as running, we still see
zeroes in the GLOBAL_CFG set of registers and cannot determine available
modes. Since all models support 10M, add a poll and wait the config to
become available.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/aquantia/aquantia_main.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/aquantia/aquantia_main.c b/drivers/net/phy/aquantia/aquantia_main.c
index 974795bd0860..2c8ba2725a91 100644
--- a/drivers/net/phy/aquantia/aquantia_main.c
+++ b/drivers/net/phy/aquantia/aquantia_main.c
@@ -652,7 +652,13 @@ static int aqr107_fill_interface_modes(struct phy_device *phydev)
 	unsigned long *possible = phydev->possible_interfaces;
 	unsigned int serdes_mode, rate_adapt;
 	phy_interface_t interface;
-	int i, val;
+	int i, val, ret;
+
+	ret = phy_read_mmd_poll_timeout(phydev, MDIO_MMD_VEND1,
+					VEND1_GLOBAL_CFG_10M, val, val != 0,
+					1000, 100000, false);
+	if (ret)
+		return ret;
 
 	/* Walk the media-speed configuration registers to determine which
 	 * host-side serdes modes may be used by the PHY depending on the
-- 
cgit v1.2.3-58-ga151


From 0ebc581f8a4b7e992bd3bab4d3ad4783245ad09c Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Mon, 8 Jul 2024 09:50:23 +0200
Subject: net: phy: aquantia: add support for aqr115c

Add support for a new model to the Aquantia driver. This PHY supports
2.5 gigabit speeds. The PHY mode is referred to by the manufacturer as
Overclocked SGMII (OCSGMII) but this actually is just 2500BASEX without
in-band signalling so reuse the existing mode to avoid changing the
uAPI.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/phy/aquantia/aquantia_main.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/drivers/net/phy/aquantia/aquantia_main.c b/drivers/net/phy/aquantia/aquantia_main.c
index 2c8ba2725a91..d12e35374231 100644
--- a/drivers/net/phy/aquantia/aquantia_main.c
+++ b/drivers/net/phy/aquantia/aquantia_main.c
@@ -29,6 +29,7 @@
 #define PHY_ID_AQR113	0x31c31c40
 #define PHY_ID_AQR113C	0x31c31c12
 #define PHY_ID_AQR114C	0x31c31c22
+#define PHY_ID_AQR115C	0x31c31c33
 #define PHY_ID_AQR813	0x31c31cb2
 
 #define MDIO_PHYXS_VEND_IF_STATUS		0xe812
@@ -1005,6 +1006,30 @@ static struct phy_driver aqr_driver[] = {
 	.led_hw_control_get = aqr_phy_led_hw_control_get,
 	.led_polarity_set = aqr_phy_led_polarity_set,
 },
+{
+	PHY_ID_MATCH_MODEL(PHY_ID_AQR115C),
+	.name           = "Aquantia AQR115C",
+	.probe          = aqr107_probe,
+	.get_rate_matching = aqr107_get_rate_matching,
+	.config_init    = aqr113c_config_init,
+	.config_aneg    = aqr_config_aneg,
+	.config_intr    = aqr_config_intr,
+	.handle_interrupt = aqr_handle_interrupt,
+	.read_status    = aqr107_read_status,
+	.get_tunable    = aqr107_get_tunable,
+	.set_tunable    = aqr107_set_tunable,
+	.suspend        = aqr107_suspend,
+	.resume         = aqr107_resume,
+	.get_sset_count = aqr107_get_sset_count,
+	.get_strings    = aqr107_get_strings,
+	.get_stats      = aqr107_get_stats,
+	.link_change_notify = aqr107_link_change_notify,
+	.led_brightness_set = aqr_phy_led_brightness_set,
+	.led_hw_is_supported = aqr_phy_led_hw_is_supported,
+	.led_hw_control_set = aqr_phy_led_hw_control_set,
+	.led_hw_control_get = aqr_phy_led_hw_control_get,
+	.led_polarity_set = aqr_phy_led_polarity_set,
+},
 {
 	PHY_ID_MATCH_MODEL(PHY_ID_AQR813),
 	.name		= "Aquantia AQR813",
@@ -1048,6 +1073,7 @@ static struct mdio_device_id __maybe_unused aqr_tbl[] = {
 	{ PHY_ID_MATCH_MODEL(PHY_ID_AQR113) },
 	{ PHY_ID_MATCH_MODEL(PHY_ID_AQR113C) },
 	{ PHY_ID_MATCH_MODEL(PHY_ID_AQR114C) },
+	{ PHY_ID_MATCH_MODEL(PHY_ID_AQR115C) },
 	{ PHY_ID_MATCH_MODEL(PHY_ID_AQR813) },
 	{ }
 };
-- 
cgit v1.2.3-58-ga151


From 003eae5a28c6c9d50290a4ac9b955be912f24c9f Mon Sep 17 00:00:00 2001
From: Benjamin Berg <benjamin.berg@intel.com>
Date: Tue, 9 Jul 2024 14:31:49 +0200
Subject: wifi: iwlwifi: correctly reference TSO page information

The code got copied from get_workaround_page, but here p->page is the
correct way to reference the page.

Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
Fixes: adc902ceada2 ("wifi: iwlwifi: keep the TSO and workaround pages mapped")
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/r/202407062135.NNjnmMdR-lkp@intel.com/
Link: https://patch.msgid.link/20240709123149.1848315-1-benjamin@sipsolutions.net
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 drivers/net/wireless/intel/iwlwifi/pcie/tx.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
index 748772fa6b3e..22d482ae53d9 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
@@ -1770,7 +1770,7 @@ static void *iwl_pcie_get_page_hdr(struct iwl_trans *trans,
 	 * (see also get_workaround_page() in tx-gen2.c)
 	 */
 	if (((unsigned long)p->pos & ~PAGE_MASK) + len < IWL_TSO_PAGE_DATA_SIZE) {
-		info = IWL_TSO_PAGE_INFO(page_address(ret));
+		info = IWL_TSO_PAGE_INFO(page_address(p->page));
 		goto out;
 	}
 
@@ -1783,7 +1783,7 @@ alloc:
 		return NULL;
 	p->pos = page_address(p->page);
 
-	info = IWL_TSO_PAGE_INFO(page_address(ret));
+	info = IWL_TSO_PAGE_INFO(page_address(p->page));
 
 	/* set the chaining pointer to NULL */
 	info->next = NULL;
-- 
cgit v1.2.3-58-ga151


From 408ac28c62f0b869d5477be12dd9eddc0c37242e Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes.berg@intel.com>
Date: Tue, 9 Jul 2024 16:08:50 +0200
Subject: wifi: mac80211: fix AP chandef capturing in CSA

When the CSA is announced with only HT elements, the AP
chandef isn't captured correctly, leading to crashes in
the later code that checks for TPE changes during CSA.

Capture the AP chandef correctly in both cases to fix
this.

Reported-by: Jouni Malinen <j@w1.fi>
Fixes: 4540568136fe ("wifi: mac80211: handle TPE element during CSA")
Link: https://patch.msgid.link/20240709160851.47805f24624d.I024091f701447f7921e93bb23b46e01c2f46347d@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
---
 net/mac80211/spectmgmt.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/net/mac80211/spectmgmt.c b/net/mac80211/spectmgmt.c
index e91ca4ccdd37..073ff9e0f397 100644
--- a/net/mac80211/spectmgmt.c
+++ b/net/mac80211/spectmgmt.c
@@ -343,6 +343,9 @@ int ieee80211_parse_ch_switch_ie(struct ieee80211_sub_if_data *sdata,
 		break;
 	}
 
+	/* capture the AP configuration */
+	csa_ie->chanreq.ap = csa_ie->chanreq.oper;
+
 	/* parse one of the Elements to build a new chandef */
 	memset(&new_chandef, 0, sizeof(new_chandef));
 	new_chandef.chan = new_chan;
@@ -369,11 +372,11 @@ int ieee80211_parse_ch_switch_ie(struct ieee80211_sub_if_data *sdata,
 		validate_chandef_by_ht_vht_oper(sdata, conn, vht_cap_info,
 						&new_chandef);
 
-	/* capture the AP chandef before (potential) downgrading */
-	csa_ie->chanreq.ap = new_chandef;
-
 	/* if data is there validate the bandwidth & use it */
 	if (new_chandef.chan) {
+		/* capture the AP chandef before (potential) downgrading */
+		csa_ie->chanreq.ap = new_chandef;
+
 		if (conn->bw_limit < IEEE80211_CONN_BW_LIMIT_320 &&
 		    new_chandef.width == NL80211_CHAN_WIDTH_320)
 			ieee80211_chandef_downgrade(&new_chandef, NULL);
-- 
cgit v1.2.3-58-ga151


From c13fda93aca118b8e5cd202e339046728ee7dddb Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Wed, 10 Jul 2024 16:16:31 +0200
Subject: bpf: Remove tst_run from lwt_seg6local_prog_ops.

The syzbot reported that the lwt_seg6 related BPF ops can be invoked
via bpf_test_run() without without entering input_action_end_bpf()
first.

Martin KaFai Lau said that self test for BPF_PROG_TYPE_LWT_SEG6LOCAL
probably didn't work since it was introduced in commit 04d4b274e2a
("ipv6: sr: Add seg6local action End.BPF"). The reason is that the
per-CPU variable seg6_bpf_srh_states::srh is never assigned in the self
test case but each BPF function expects it.

Remove test_run for BPF_PROG_TYPE_LWT_SEG6LOCAL.

Suggested-by: Martin KaFai Lau <martin.lau@linux.dev>
Reported-by: syzbot+608a2acde8c5a101d07d@syzkaller.appspotmail.com
Fixes: d1542d4ae4df ("seg6: Use nested-BH locking for seg6_bpf_srh_states.")
Fixes: 004d4b274e2a ("ipv6: sr: Add seg6local action End.BPF")
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/r/20240710141631.FbmHcQaX@linutronix.de
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 net/core/filter.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/core/filter.c b/net/core/filter.c
index d767880c276d..4cf1d34f7617 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -11053,7 +11053,6 @@ const struct bpf_verifier_ops lwt_seg6local_verifier_ops = {
 };
 
 const struct bpf_prog_ops lwt_seg6local_prog_ops = {
-	.test_run		= bpf_prog_test_run_skb,
 };
 
 const struct bpf_verifier_ops cg_sock_verifier_ops = {
-- 
cgit v1.2.3-58-ga151


From 2cb13dec8c5e5e104fd2f71c2dee761d6ed9a333 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:34 +0200
Subject: cache: add __cacheline_group_{begin, end}_aligned() (+ couple more)

__cacheline_group_begin(), unfortunately, doesn't align the group
anyhow. If it is wanted, then you need to do something like

__cacheline_group_begin(grp) __aligned(ALIGN)

which isn't really convenient nor compact.
Add the _aligned() counterparts to align the groups automatically to
either the specified alignment (optional) or ``SMP_CACHE_BYTES``.
Note that the actual struct layout will then be (on x64 with 64-byte CL):

struct x {
	u32 y;				// offset 0, size 4, padding 56
	__cacheline_group_begin__grp;	// offset 64, size 0
	u32 z;				// offset 64, size 4, padding 4
	__cacheline_group_end__grp;	// offset 72, size 0
	__cacheline_group_pad__grp;	// offset 72, size 0, padding 56
	u32 w;				// offset 128
};

The end marker is aligned to long, so that you can assert the struct
size more strictly, but the offset of the next field in the structure
will be aligned to the group alignment, so that the next field won't
fall into the group it's not intended to.

Add __LARGEST_ALIGN definition and LARGEST_ALIGN() macro.
__LARGEST_ALIGN is the value to which the compilers align fields when
__aligned_largest is specified. Sometimes, it might be needed to get
this value outside of variable definitions. LARGEST_ALIGN() is macro
which just aligns a value to __LARGEST_ALIGN.
Also add SMP_CACHE_ALIGN(), similar to L1_CACHE_ALIGN(), but using
``SMP_CACHE_BYTES`` instead of ``L1_CACHE_BYTES`` as the former
also accounts L2, needed in some cases.

Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 include/linux/cache.h | 59 +++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 59 insertions(+)

diff --git a/include/linux/cache.h b/include/linux/cache.h
index 0ecb17bb6883..ca2a05682a54 100644
--- a/include/linux/cache.h
+++ b/include/linux/cache.h
@@ -13,6 +13,32 @@
 #define SMP_CACHE_BYTES L1_CACHE_BYTES
 #endif
 
+/**
+ * SMP_CACHE_ALIGN - align a value to the L2 cacheline size
+ * @x: value to align
+ *
+ * On some architectures, L2 ("SMP") CL size is bigger than L1, and sometimes,
+ * this needs to be accounted.
+ *
+ * Return: aligned value.
+ */
+#ifndef SMP_CACHE_ALIGN
+#define SMP_CACHE_ALIGN(x)	ALIGN(x, SMP_CACHE_BYTES)
+#endif
+
+/*
+ * ``__aligned_largest`` aligns a field to the value most optimal for the
+ * target architecture to perform memory operations. Get the actual value
+ * to be able to use it anywhere else.
+ */
+#ifndef __LARGEST_ALIGN
+#define __LARGEST_ALIGN		sizeof(struct { long x; } __aligned_largest)
+#endif
+
+#ifndef LARGEST_ALIGN
+#define LARGEST_ALIGN(x)	ALIGN(x, __LARGEST_ALIGN)
+#endif
+
 /*
  * __read_mostly is used to keep rarely changing variables out of frequently
  * updated cachelines. Its use should be reserved for data that is used
@@ -95,6 +121,39 @@
 	__u8 __cacheline_group_end__##GROUP[0]
 #endif
 
+/**
+ * __cacheline_group_begin_aligned - declare an aligned group start
+ * @GROUP: name of the group
+ * @...: optional group alignment
+ *
+ * The following block inside a struct:
+ *
+ *	__cacheline_group_begin_aligned(grp);
+ *	field a;
+ *	field b;
+ *	__cacheline_group_end_aligned(grp);
+ *
+ * will always be aligned to either the specified alignment or
+ * ``SMP_CACHE_BYTES``.
+ */
+#define __cacheline_group_begin_aligned(GROUP, ...)		\
+	__cacheline_group_begin(GROUP)				\
+	__aligned((__VA_ARGS__ + 0) ? : SMP_CACHE_BYTES)
+
+/**
+ * __cacheline_group_end_aligned - declare an aligned group end
+ * @GROUP: name of the group
+ * @...: optional alignment (same as was in __cacheline_group_begin_aligned())
+ *
+ * Note that the end marker is aligned to sizeof(long) to allow more precise
+ * size assertion. It also declares a padding at the end to avoid next field
+ * falling into this cacheline.
+ */
+#define __cacheline_group_end_aligned(GROUP, ...)		\
+	__cacheline_group_end(GROUP) __aligned(sizeof(long));	\
+	struct { } __cacheline_group_pad__##GROUP		\
+	__aligned((__VA_ARGS__ + 0) ? : SMP_CACHE_BYTES)
+
 #ifndef CACHELINE_ASSERT_GROUP_MEMBER
 #define CACHELINE_ASSERT_GROUP_MEMBER(TYPE, GROUP, MEMBER) \
 	BUILD_BUG_ON(!(offsetof(TYPE, MEMBER) >= \
-- 
cgit v1.2.3-58-ga151


From 39daa09d34ada1bc7227d68def63e0a2105b5496 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:35 +0200
Subject: page_pool: use __cacheline_group_{begin, end}_aligned()

Instead of doing __cacheline_group_begin() __aligned(), use the new
__cacheline_group_{begin,end}_aligned(), so that it will take care
of the group alignment itself.
Also replace open-coded `4 * sizeof(long)` in two places with
a definition.

Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 include/net/page_pool/types.h | 22 ++++++++++++----------
 net/core/page_pool.c          |  3 ++-
 2 files changed, 14 insertions(+), 11 deletions(-)

diff --git a/include/net/page_pool/types.h b/include/net/page_pool/types.h
index b70bcc14ceda..50569fed7868 100644
--- a/include/net/page_pool/types.h
+++ b/include/net/page_pool/types.h
@@ -129,6 +129,16 @@ struct page_pool_stats {
 };
 #endif
 
+/* The whole frag API block must stay within one cacheline. On 32-bit systems,
+ * sizeof(long) == sizeof(int), so that the block size is ``3 * sizeof(long)``.
+ * On 64-bit systems, the actual size is ``2 * sizeof(long) + sizeof(int)``.
+ * The closest pow-2 to both of them is ``4 * sizeof(long)``, so just use that
+ * one for simplicity.
+ * Having it aligned to a cacheline boundary may be excessive and doesn't bring
+ * any good.
+ */
+#define PAGE_POOL_FRAG_GROUP_ALIGN	(4 * sizeof(long))
+
 struct page_pool {
 	struct page_pool_params_fast p;
 
@@ -142,19 +152,11 @@ struct page_pool {
 	bool system:1;			/* This is a global percpu pool */
 #endif
 
-	/* The following block must stay within one cacheline. On 32-bit
-	 * systems, sizeof(long) == sizeof(int), so that the block size is
-	 * ``3 * sizeof(long)``. On 64-bit systems, the actual size is
-	 * ``2 * sizeof(long) + sizeof(int)``. The closest pow-2 to both of
-	 * them is ``4 * sizeof(long)``, so just use that one for simplicity.
-	 * Having it aligned to a cacheline boundary may be excessive and
-	 * doesn't bring any good.
-	 */
-	__cacheline_group_begin(frag) __aligned(4 * sizeof(long));
+	__cacheline_group_begin_aligned(frag, PAGE_POOL_FRAG_GROUP_ALIGN);
 	long frag_users;
 	netmem_ref frag_page;
 	unsigned int frag_offset;
-	__cacheline_group_end(frag);
+	__cacheline_group_end_aligned(frag, PAGE_POOL_FRAG_GROUP_ALIGN);
 
 	struct delayed_work release_dw;
 	void (*disconnect)(void *pool);
diff --git a/net/core/page_pool.c b/net/core/page_pool.c
index 855271a6cad2..2abe6e919224 100644
--- a/net/core/page_pool.c
+++ b/net/core/page_pool.c
@@ -178,7 +178,8 @@ static void page_pool_struct_check(void)
 	CACHELINE_ASSERT_GROUP_MEMBER(struct page_pool, frag, frag_users);
 	CACHELINE_ASSERT_GROUP_MEMBER(struct page_pool, frag, frag_page);
 	CACHELINE_ASSERT_GROUP_MEMBER(struct page_pool, frag, frag_offset);
-	CACHELINE_ASSERT_GROUP_SIZE(struct page_pool, frag, 4 * sizeof(long));
+	CACHELINE_ASSERT_GROUP_SIZE(struct page_pool, frag,
+				    PAGE_POOL_FRAG_GROUP_ALIGN);
 }
 
 static int page_pool_init(struct page_pool *pool,
-- 
cgit v1.2.3-58-ga151


From 62c884256ea1f898b00f20729a306a4eeb1840b1 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:36 +0200
Subject: libeth: add cacheline / struct layout assertion helpers

Add helpers to assert struct field layout, a bit more crazy and
networking-specific than in <linux/cache.h>. They assume you have
3 CL-aligned groups (read-mostly, read-write, cold) in a struct
you want to assert, and nothing besides them.
For 64-bit with 64-byte cachelines, the assertions are as strict
as possible, as the size can then be easily predicted.
For the rest, make sure they don't cross the specified bound.

Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 include/net/libeth/cache.h | 66 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 66 insertions(+)
 create mode 100644 include/net/libeth/cache.h

diff --git a/include/net/libeth/cache.h b/include/net/libeth/cache.h
new file mode 100644
index 000000000000..bdb0c043ce61
--- /dev/null
+++ b/include/net/libeth/cache.h
@@ -0,0 +1,66 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright (C) 2024 Intel Corporation */
+
+#ifndef __LIBETH_CACHE_H
+#define __LIBETH_CACHE_H
+
+#include <linux/cache.h>
+
+/**
+ * libeth_cacheline_group_assert - make sure cacheline group size is expected
+ * @type: type of the structure containing the group
+ * @grp: group name inside the struct
+ * @sz: expected group size
+ */
+#if defined(CONFIG_64BIT) && SMP_CACHE_BYTES == 64
+#define libeth_cacheline_group_assert(type, grp, sz)			      \
+	static_assert(offsetof(type, __cacheline_group_end__##grp) -	      \
+		      offsetofend(type, __cacheline_group_begin__##grp) ==    \
+		      (sz))
+#define __libeth_cacheline_struct_assert(type, sz)			      \
+	static_assert(sizeof(type) == (sz))
+#else /* !CONFIG_64BIT || SMP_CACHE_BYTES != 64 */
+#define libeth_cacheline_group_assert(type, grp, sz)			      \
+	static_assert(offsetof(type, __cacheline_group_end__##grp) -	      \
+		      offsetofend(type, __cacheline_group_begin__##grp) <=    \
+		      (sz))
+#define __libeth_cacheline_struct_assert(type, sz)			      \
+	static_assert(sizeof(type) <= (sz))
+#endif /* !CONFIG_64BIT || SMP_CACHE_BYTES != 64 */
+
+#define __libeth_cls1(sz1)	SMP_CACHE_ALIGN(sz1)
+#define __libeth_cls2(sz1, sz2)	(SMP_CACHE_ALIGN(sz1) + SMP_CACHE_ALIGN(sz2))
+#define __libeth_cls3(sz1, sz2, sz3)					      \
+	(SMP_CACHE_ALIGN(sz1) + SMP_CACHE_ALIGN(sz2) + SMP_CACHE_ALIGN(sz3))
+#define __libeth_cls(...)						      \
+	CONCATENATE(__libeth_cls, COUNT_ARGS(__VA_ARGS__))(__VA_ARGS__)
+
+/**
+ * libeth_cacheline_struct_assert - make sure CL-based struct size is expected
+ * @type: type of the struct
+ * @...: from 1 to 3 CL group sizes (read-mostly, read-write, cold)
+ *
+ * When a struct contains several CL groups, it's difficult to predict its size
+ * on different architectures. The macro instead takes sizes of all of the
+ * groups the structure contains and generates the final struct size.
+ */
+#define libeth_cacheline_struct_assert(type, ...)			      \
+	__libeth_cacheline_struct_assert(type, __libeth_cls(__VA_ARGS__));    \
+	static_assert(__alignof(type) >= SMP_CACHE_BYTES)
+
+/**
+ * libeth_cacheline_set_assert - make sure CL-based struct layout is expected
+ * @type: type of the struct
+ * @ro: expected size of the read-mostly group
+ * @rw: expected size of the read-write group
+ * @c: expected size of the cold group
+ *
+ * Check that each group size is expected and then do final struct size check.
+ */
+#define libeth_cacheline_set_assert(type, ro, rw, c)			      \
+	libeth_cacheline_group_assert(type, read_mostly, ro);		      \
+	libeth_cacheline_group_assert(type, read_write, rw);		      \
+	libeth_cacheline_group_assert(type, cold, c);			      \
+	libeth_cacheline_struct_assert(type, ro, rw, c)
+
+#endif /* __LIBETH_CACHE_H */
-- 
cgit v1.2.3-58-ga151


From 66c27e3b19d5aae58d7f0145113de61d6fba5e09 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:37 +0200
Subject: idpf: stop using macros for accessing queue descriptors

In C, we have structures and unions.
Casting `void *` via macros is not only error-prone, but also looks
confusing and awful in general.
In preparation for splitting the queue structs, replace it with a
union and direct array dereferences.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/idpf/idpf.h             |  1 -
 drivers/net/ethernet/intel/idpf/idpf_lan_txrx.h    |  2 +
 .../net/ethernet/intel/idpf/idpf_singleq_txrx.c    | 20 ++++-----
 drivers/net/ethernet/intel/idpf/idpf_txrx.c        | 32 +++++++--------
 drivers/net/ethernet/intel/idpf/idpf_txrx.h        | 47 ++++++++++++----------
 5 files changed, 52 insertions(+), 50 deletions(-)

diff --git a/drivers/net/ethernet/intel/idpf/idpf.h b/drivers/net/ethernet/intel/idpf/idpf.h
index e7a036538246..0b26dd9b8a51 100644
--- a/drivers/net/ethernet/intel/idpf/idpf.h
+++ b/drivers/net/ethernet/intel/idpf/idpf.h
@@ -20,7 +20,6 @@ struct idpf_vport_max_q;
 #include <linux/dim.h>
 
 #include "virtchnl2.h"
-#include "idpf_lan_txrx.h"
 #include "idpf_txrx.h"
 #include "idpf_controlq.h"
 
diff --git a/drivers/net/ethernet/intel/idpf/idpf_lan_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_lan_txrx.h
index a5752dcab888..8c7f8ef8f1a1 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_lan_txrx.h
+++ b/drivers/net/ethernet/intel/idpf/idpf_lan_txrx.h
@@ -4,6 +4,8 @@
 #ifndef _IDPF_LAN_TXRX_H_
 #define _IDPF_LAN_TXRX_H_
 
+#include <linux/bits.h>
+
 enum idpf_rss_hash {
 	IDPF_HASH_INVALID			= 0,
 	/* Values 1 - 28 are reserved for future use */
diff --git a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
index 27b93592c4ba..b17d88e15000 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
@@ -205,7 +205,7 @@ static void idpf_tx_singleq_map(struct idpf_queue *tx_q,
 	data_len = skb->data_len;
 	size = skb_headlen(skb);
 
-	tx_desc = IDPF_BASE_TX_DESC(tx_q, i);
+	tx_desc = &tx_q->base_tx[i];
 
 	dma = dma_map_single(tx_q->dev, skb->data, size, DMA_TO_DEVICE);
 
@@ -239,7 +239,7 @@ static void idpf_tx_singleq_map(struct idpf_queue *tx_q,
 			i++;
 
 			if (i == tx_q->desc_count) {
-				tx_desc = IDPF_BASE_TX_DESC(tx_q, 0);
+				tx_desc = &tx_q->base_tx[0];
 				i = 0;
 			}
 
@@ -259,7 +259,7 @@ static void idpf_tx_singleq_map(struct idpf_queue *tx_q,
 		i++;
 
 		if (i == tx_q->desc_count) {
-			tx_desc = IDPF_BASE_TX_DESC(tx_q, 0);
+			tx_desc = &tx_q->base_tx[0];
 			i = 0;
 		}
 
@@ -307,7 +307,7 @@ idpf_tx_singleq_get_ctx_desc(struct idpf_queue *txq)
 	memset(&txq->tx_buf[ntu], 0, sizeof(struct idpf_tx_buf));
 	txq->tx_buf[ntu].ctx_entry = true;
 
-	ctx_desc = IDPF_BASE_TX_CTX_DESC(txq, ntu);
+	ctx_desc = &txq->base_ctx[ntu];
 
 	IDPF_SINGLEQ_BUMP_RING_IDX(txq, ntu);
 	txq->next_to_use = ntu;
@@ -455,7 +455,7 @@ static bool idpf_tx_singleq_clean(struct idpf_queue *tx_q, int napi_budget,
 	struct netdev_queue *nq;
 	bool dont_wake;
 
-	tx_desc = IDPF_BASE_TX_DESC(tx_q, ntc);
+	tx_desc = &tx_q->base_tx[ntc];
 	tx_buf = &tx_q->tx_buf[ntc];
 	ntc -= tx_q->desc_count;
 
@@ -517,7 +517,7 @@ static bool idpf_tx_singleq_clean(struct idpf_queue *tx_q, int napi_budget,
 			if (unlikely(!ntc)) {
 				ntc -= tx_q->desc_count;
 				tx_buf = tx_q->tx_buf;
-				tx_desc = IDPF_BASE_TX_DESC(tx_q, 0);
+				tx_desc = &tx_q->base_tx[0];
 			}
 
 			/* unmap any remaining paged data */
@@ -540,7 +540,7 @@ fetch_next_txq_desc:
 		if (unlikely(!ntc)) {
 			ntc -= tx_q->desc_count;
 			tx_buf = tx_q->tx_buf;
-			tx_desc = IDPF_BASE_TX_DESC(tx_q, 0);
+			tx_desc = &tx_q->base_tx[0];
 		}
 	} while (likely(budget));
 
@@ -895,7 +895,7 @@ bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_queue *rx_q,
 	if (!cleaned_count)
 		return false;
 
-	desc = IDPF_SINGLEQ_RX_BUF_DESC(rx_q, nta);
+	desc = &rx_q->single_buf[nta];
 	buf = &rx_q->rx_buf.buf[nta];
 
 	do {
@@ -915,7 +915,7 @@ bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_queue *rx_q,
 		buf++;
 		nta++;
 		if (unlikely(nta == rx_q->desc_count)) {
-			desc = IDPF_SINGLEQ_RX_BUF_DESC(rx_q, 0);
+			desc = &rx_q->single_buf[0];
 			buf = rx_q->rx_buf.buf;
 			nta = 0;
 		}
@@ -1016,7 +1016,7 @@ static int idpf_rx_singleq_clean(struct idpf_queue *rx_q, int budget)
 		struct idpf_rx_buf *rx_buf;
 
 		/* get the Rx desc from Rx queue based on 'next_to_clean' */
-		rx_desc = IDPF_RX_DESC(rx_q, ntc);
+		rx_desc = &rx_q->rx[ntc];
 
 		/* status_error_ptype_len will always be zero for unused
 		 * descriptors because it's cleared in cleanup, and overlaps
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
index b023704bbbda..01301e640fba 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
@@ -531,7 +531,7 @@ static bool idpf_rx_post_buf_desc(struct idpf_queue *bufq, u16 buf_id)
 	struct idpf_rx_buf *buf;
 	dma_addr_t addr;
 
-	splitq_rx_desc = IDPF_SPLITQ_RX_BUF_DESC(bufq, nta);
+	splitq_rx_desc = &bufq->split_buf[nta];
 	buf = &bufq->rx_buf.buf[buf_id];
 
 	if (bufq->rx_hsplit_en) {
@@ -1584,7 +1584,7 @@ do {								\
 	if (unlikely(!(ntc))) {					\
 		ntc -= (txq)->desc_count;			\
 		buf = (txq)->tx_buf;				\
-		desc = IDPF_FLEX_TX_DESC(txq, 0);		\
+		desc = &(txq)->flex_tx[0];			\
 	} else {						\
 		(buf)++;					\
 		(desc)++;					\
@@ -1617,8 +1617,8 @@ static void idpf_tx_splitq_clean(struct idpf_queue *tx_q, u16 end,
 	s16 ntc = tx_q->next_to_clean;
 	struct idpf_tx_buf *tx_buf;
 
-	tx_desc = IDPF_FLEX_TX_DESC(tx_q, ntc);
-	next_pending_desc = IDPF_FLEX_TX_DESC(tx_q, end);
+	tx_desc = &tx_q->flex_tx[ntc];
+	next_pending_desc = &tx_q->flex_tx[end];
 	tx_buf = &tx_q->tx_buf[ntc];
 	ntc -= tx_q->desc_count;
 
@@ -1814,7 +1814,7 @@ static bool idpf_tx_clean_complq(struct idpf_queue *complq, int budget,
 	int i;
 
 	complq_budget = vport->compln_clean_budget;
-	tx_desc = IDPF_SPLITQ_TX_COMPLQ_DESC(complq, ntc);
+	tx_desc = &complq->comp[ntc];
 	ntc -= complq->desc_count;
 
 	do {
@@ -1879,7 +1879,7 @@ fetch_next_desc:
 		ntc++;
 		if (unlikely(!ntc)) {
 			ntc -= complq->desc_count;
-			tx_desc = IDPF_SPLITQ_TX_COMPLQ_DESC(complq, 0);
+			tx_desc = &complq->comp[0];
 			change_bit(__IDPF_Q_GEN_CHK, complq->flags);
 		}
 
@@ -2143,7 +2143,7 @@ void idpf_tx_dma_map_error(struct idpf_queue *txq, struct sk_buff *skb,
 		 * used one additional descriptor for a context
 		 * descriptor. Reset that here.
 		 */
-		tx_desc = IDPF_FLEX_TX_DESC(txq, idx);
+		tx_desc = &txq->flex_tx[idx];
 		memset(tx_desc, 0, sizeof(struct idpf_flex_tx_ctx_desc));
 		if (idx == 0)
 			idx = txq->desc_count;
@@ -2202,7 +2202,7 @@ static void idpf_tx_splitq_map(struct idpf_queue *tx_q,
 	data_len = skb->data_len;
 	size = skb_headlen(skb);
 
-	tx_desc = IDPF_FLEX_TX_DESC(tx_q, i);
+	tx_desc = &tx_q->flex_tx[i];
 
 	dma = dma_map_single(tx_q->dev, skb->data, size, DMA_TO_DEVICE);
 
@@ -2275,7 +2275,7 @@ static void idpf_tx_splitq_map(struct idpf_queue *tx_q,
 			i++;
 
 			if (i == tx_q->desc_count) {
-				tx_desc = IDPF_FLEX_TX_DESC(tx_q, 0);
+				tx_desc = &tx_q->flex_tx[0];
 				i = 0;
 				tx_q->compl_tag_cur_gen =
 					IDPF_TX_ADJ_COMPL_TAG_GEN(tx_q);
@@ -2320,7 +2320,7 @@ static void idpf_tx_splitq_map(struct idpf_queue *tx_q,
 		i++;
 
 		if (i == tx_q->desc_count) {
-			tx_desc = IDPF_FLEX_TX_DESC(tx_q, 0);
+			tx_desc = &tx_q->flex_tx[0];
 			i = 0;
 			tx_q->compl_tag_cur_gen = IDPF_TX_ADJ_COMPL_TAG_GEN(tx_q);
 		}
@@ -2553,7 +2553,7 @@ idpf_tx_splitq_get_ctx_desc(struct idpf_queue *txq)
 	txq->tx_buf[i].compl_tag = IDPF_SPLITQ_TX_INVAL_COMPL_TAG;
 
 	/* grab the next descriptor */
-	desc = IDPF_FLEX_TX_CTX_DESC(txq, i);
+	desc = &txq->flex_ctx[i];
 	txq->next_to_use = idpf_tx_splitq_bump_ntu(txq, i);
 
 	return desc;
@@ -3128,7 +3128,6 @@ static int idpf_rx_splitq_clean(struct idpf_queue *rxq, int budget)
 		struct idpf_sw_queue *refillq = NULL;
 		struct idpf_rxq_set *rxq_set = NULL;
 		struct idpf_rx_buf *rx_buf = NULL;
-		union virtchnl2_rx_desc *desc;
 		unsigned int pkt_len = 0;
 		unsigned int hdr_len = 0;
 		u16 gen_id, buf_id = 0;
@@ -3138,8 +3137,7 @@ static int idpf_rx_splitq_clean(struct idpf_queue *rxq, int budget)
 		u8 rxdid;
 
 		/* get the Rx desc from Rx queue based on 'next_to_clean' */
-		desc = IDPF_RX_DESC(rxq, ntc);
-		rx_desc = (struct virtchnl2_rx_flex_desc_adv_nic_3 *)desc;
+		rx_desc = &rxq->rx[ntc].flex_adv_nic_3_wb;
 
 		/* This memory barrier is needed to keep us from reading
 		 * any other fields out of the rx_desc
@@ -3320,11 +3318,11 @@ static void idpf_rx_clean_refillq(struct idpf_queue *bufq,
 	int cleaned = 0;
 	u16 gen;
 
-	buf_desc = IDPF_SPLITQ_RX_BUF_DESC(bufq, bufq_nta);
+	buf_desc = &bufq->split_buf[bufq_nta];
 
 	/* make sure we stop at ring wrap in the unlikely case ring is full */
 	while (likely(cleaned < refillq->desc_count)) {
-		u16 refill_desc = IDPF_SPLITQ_RX_BI_DESC(refillq, ntc);
+		u16 refill_desc = refillq->ring[ntc];
 		bool failure;
 
 		gen = FIELD_GET(IDPF_RX_BI_GEN_M, refill_desc);
@@ -3342,7 +3340,7 @@ static void idpf_rx_clean_refillq(struct idpf_queue *bufq,
 		}
 
 		if (unlikely(++bufq_nta == bufq->desc_count)) {
-			buf_desc = IDPF_SPLITQ_RX_BUF_DESC(bufq, 0);
+			buf_desc = &bufq->split_buf[0];
 			bufq_nta = 0;
 		} else {
 			buf_desc++;
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
index 551391e20464..6dce14483215 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
@@ -8,6 +8,7 @@
 #include <net/tcp.h>
 #include <net/netdev_queues.h>
 
+#include "idpf_lan_txrx.h"
 #include "virtchnl2_lan_desc.h"
 
 #define IDPF_LARGE_MAX_Q			256
@@ -117,24 +118,6 @@ do {								\
 #define IDPF_RXD_EOF_SPLITQ		VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_EOF_M
 #define IDPF_RXD_EOF_SINGLEQ		VIRTCHNL2_RX_BASE_DESC_STATUS_EOF_M
 
-#define IDPF_SINGLEQ_RX_BUF_DESC(rxq, i)	\
-	(&(((struct virtchnl2_singleq_rx_buf_desc *)((rxq)->desc_ring))[i]))
-#define IDPF_SPLITQ_RX_BUF_DESC(rxq, i)	\
-	(&(((struct virtchnl2_splitq_rx_buf_desc *)((rxq)->desc_ring))[i]))
-#define IDPF_SPLITQ_RX_BI_DESC(rxq, i) ((((rxq)->ring))[i])
-
-#define IDPF_BASE_TX_DESC(txq, i)	\
-	(&(((struct idpf_base_tx_desc *)((txq)->desc_ring))[i]))
-#define IDPF_BASE_TX_CTX_DESC(txq, i) \
-	(&(((struct idpf_base_tx_ctx_desc *)((txq)->desc_ring))[i]))
-#define IDPF_SPLITQ_TX_COMPLQ_DESC(txcq, i)	\
-	(&(((struct idpf_splitq_tx_compl_desc *)((txcq)->desc_ring))[i]))
-
-#define IDPF_FLEX_TX_DESC(txq, i) \
-	(&(((union idpf_tx_flex_desc *)((txq)->desc_ring))[i]))
-#define IDPF_FLEX_TX_CTX_DESC(txq, i)	\
-	(&(((struct idpf_flex_tx_ctx_desc *)((txq)->desc_ring))[i]))
-
 #define IDPF_DESC_UNUSED(txq)     \
 	((((txq)->next_to_clean > (txq)->next_to_use) ? 0 : (txq)->desc_count) + \
 	(txq)->next_to_clean - (txq)->next_to_use - 1)
@@ -317,8 +300,6 @@ struct idpf_rx_extracted {
 
 #define IDPF_RX_DMA_ATTR \
 	(DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING)
-#define IDPF_RX_DESC(rxq, i)	\
-	(&(((union virtchnl2_rx_desc *)((rxq)->desc_ring))[i]))
 
 struct idpf_rx_buf {
 	struct page *page;
@@ -655,7 +636,15 @@ union idpf_queue_stats {
  * @q_vector: Backreference to associated vector
  * @size: Length of descriptor ring in bytes
  * @dma: Physical address of ring
- * @desc_ring: Descriptor ring memory
+ * @rx: universal receive descriptor array
+ * @single_buf: Rx buffer descriptor array in singleq
+ * @split_buf: Rx buffer descriptor array in splitq
+ * @base_tx: basic Tx descriptor array
+ * @base_ctx: basic Tx context descriptor array
+ * @flex_tx: flex Tx descriptor array
+ * @flex_ctx: flex Tx context descriptor array
+ * @comp: completion descriptor array
+ * @desc_ring: virtual descriptor ring address
  * @tx_max_bufs: Max buffers that can be transmitted with scatter-gather
  * @tx_min_pkt_len: Min supported packet length
  * @num_completions: Only relevant for TX completion queue. It tracks the
@@ -733,7 +722,21 @@ struct idpf_queue {
 	struct idpf_q_vector *q_vector;
 	unsigned int size;
 	dma_addr_t dma;
-	void *desc_ring;
+	union {
+		union virtchnl2_rx_desc *rx;
+
+		struct virtchnl2_singleq_rx_buf_desc *single_buf;
+		struct virtchnl2_splitq_rx_buf_desc *split_buf;
+
+		struct idpf_base_tx_desc *base_tx;
+		struct idpf_base_tx_ctx_desc *base_ctx;
+		union idpf_tx_flex_desc *flex_tx;
+		struct idpf_flex_tx_ctx_desc *flex_ctx;
+
+		struct idpf_splitq_tx_compl_desc *comp;
+
+		void *desc_ring;
+	};
 
 	u16 tx_max_bufs;
 	u8 tx_min_pkt_len;
-- 
cgit v1.2.3-58-ga151


From e4891e4687c8dd136d80d6c1b857a02931ed6fc8 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:38 +0200
Subject: idpf: split &idpf_queue into 4 strictly-typed queue structures

Currently, sizeof(struct idpf_queue) is 32 Kb.
This is due to the 12-bit hashtable declaration at the end of the queue.
This HT is needed only for Tx queues when the flow scheduling mode is
enabled. But &idpf_queue is unified for all of the queue types,
provoking excessive memory usage.
The unified structure in general makes the code less effective via
suboptimal fields placement. You can't avoid that unless you make unions
each 2 fields. Even then, different field alignment etc., doesn't allow
you to optimize things to the limit.
Split &idpf_queue into 4 structures corresponding to the queue types:
RQ (Rx queue), SQ (Tx queue), FQ (buffer queue), and CQ (completion
queue). Place only needed fields there and shortcuts handy for hotpath.
Allocate the abovementioned hashtable dynamically and only when needed,
keeping &idpf_tx_queue relatively short (192 bytes, same as Rx). This HT
is used only for OOO completions, which aren't really hotpath anyway.
Note that this change must be done atomically, otherwise it's really
easy to get lost and miss something.

Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/idpf/idpf.h             |   3 +-
 drivers/net/ethernet/intel/idpf/idpf_ethtool.c     | 125 +--
 drivers/net/ethernet/intel/idpf/idpf_lib.c         |  46 +-
 .../net/ethernet/intel/idpf/idpf_singleq_txrx.c    | 144 ++--
 drivers/net/ethernet/intel/idpf/idpf_txrx.c        | 915 ++++++++++++---------
 drivers/net/ethernet/intel/idpf/idpf_txrx.h        | 440 ++++++----
 drivers/net/ethernet/intel/idpf/idpf_virtchnl.c    |  73 +-
 7 files changed, 1018 insertions(+), 728 deletions(-)

diff --git a/drivers/net/ethernet/intel/idpf/idpf.h b/drivers/net/ethernet/intel/idpf/idpf.h
index 0b26dd9b8a51..f9e43d171f17 100644
--- a/drivers/net/ethernet/intel/idpf/idpf.h
+++ b/drivers/net/ethernet/intel/idpf/idpf.h
@@ -17,7 +17,6 @@ struct idpf_vport_max_q;
 #include <linux/sctp.h>
 #include <linux/ethtool_netlink.h>
 #include <net/gro.h>
-#include <linux/dim.h>
 
 #include "virtchnl2.h"
 #include "idpf_txrx.h"
@@ -301,7 +300,7 @@ struct idpf_vport {
 	u16 num_txq_grp;
 	struct idpf_txq_group *txq_grps;
 	u32 txq_model;
-	struct idpf_queue **txqs;
+	struct idpf_tx_queue **txqs;
 	bool crc_enable;
 
 	u16 num_rxq;
diff --git a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c
index 1885ba618981..e933fed16c7e 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c
@@ -437,22 +437,24 @@ struct idpf_stats {
 	.stat_offset = offsetof(_type, _stat) \
 }
 
-/* Helper macro for defining some statistics related to queues */
-#define IDPF_QUEUE_STAT(_name, _stat) \
-	IDPF_STAT(struct idpf_queue, _name, _stat)
+/* Helper macros for defining some statistics related to queues */
+#define IDPF_RX_QUEUE_STAT(_name, _stat) \
+	IDPF_STAT(struct idpf_rx_queue, _name, _stat)
+#define IDPF_TX_QUEUE_STAT(_name, _stat) \
+	IDPF_STAT(struct idpf_tx_queue, _name, _stat)
 
 /* Stats associated with a Tx queue */
 static const struct idpf_stats idpf_gstrings_tx_queue_stats[] = {
-	IDPF_QUEUE_STAT("pkts", q_stats.tx.packets),
-	IDPF_QUEUE_STAT("bytes", q_stats.tx.bytes),
-	IDPF_QUEUE_STAT("lso_pkts", q_stats.tx.lso_pkts),
+	IDPF_TX_QUEUE_STAT("pkts", q_stats.packets),
+	IDPF_TX_QUEUE_STAT("bytes", q_stats.bytes),
+	IDPF_TX_QUEUE_STAT("lso_pkts", q_stats.lso_pkts),
 };
 
 /* Stats associated with an Rx queue */
 static const struct idpf_stats idpf_gstrings_rx_queue_stats[] = {
-	IDPF_QUEUE_STAT("pkts", q_stats.rx.packets),
-	IDPF_QUEUE_STAT("bytes", q_stats.rx.bytes),
-	IDPF_QUEUE_STAT("rx_gro_hw_pkts", q_stats.rx.rsc_pkts),
+	IDPF_RX_QUEUE_STAT("pkts", q_stats.packets),
+	IDPF_RX_QUEUE_STAT("bytes", q_stats.bytes),
+	IDPF_RX_QUEUE_STAT("rx_gro_hw_pkts", q_stats.rsc_pkts),
 };
 
 #define IDPF_TX_QUEUE_STATS_LEN		ARRAY_SIZE(idpf_gstrings_tx_queue_stats)
@@ -633,7 +635,7 @@ static int idpf_get_sset_count(struct net_device *netdev, int sset)
  * Copies the stat data defined by the pointer and stat structure pair into
  * the memory supplied as data. If the pointer is null, data will be zero'd.
  */
-static void idpf_add_one_ethtool_stat(u64 *data, void *pstat,
+static void idpf_add_one_ethtool_stat(u64 *data, const void *pstat,
 				      const struct idpf_stats *stat)
 {
 	char *p;
@@ -671,6 +673,7 @@ static void idpf_add_one_ethtool_stat(u64 *data, void *pstat,
  * idpf_add_queue_stats - copy queue statistics into supplied buffer
  * @data: ethtool stats buffer
  * @q: the queue to copy
+ * @type: type of the queue
  *
  * Queue statistics must be copied while protected by u64_stats_fetch_begin,
  * so we can't directly use idpf_add_ethtool_stats. Assumes that queue stats
@@ -681,19 +684,23 @@ static void idpf_add_one_ethtool_stat(u64 *data, void *pstat,
  *
  * This function expects to be called while under rcu_read_lock().
  */
-static void idpf_add_queue_stats(u64 **data, struct idpf_queue *q)
+static void idpf_add_queue_stats(u64 **data, const void *q,
+				 enum virtchnl2_queue_type type)
 {
+	const struct u64_stats_sync *stats_sync;
 	const struct idpf_stats *stats;
 	unsigned int start;
 	unsigned int size;
 	unsigned int i;
 
-	if (q->q_type == VIRTCHNL2_QUEUE_TYPE_RX) {
+	if (type == VIRTCHNL2_QUEUE_TYPE_RX) {
 		size = IDPF_RX_QUEUE_STATS_LEN;
 		stats = idpf_gstrings_rx_queue_stats;
+		stats_sync = &((const struct idpf_rx_queue *)q)->stats_sync;
 	} else {
 		size = IDPF_TX_QUEUE_STATS_LEN;
 		stats = idpf_gstrings_tx_queue_stats;
+		stats_sync = &((const struct idpf_tx_queue *)q)->stats_sync;
 	}
 
 	/* To avoid invalid statistics values, ensure that we keep retrying
@@ -701,10 +708,10 @@ static void idpf_add_queue_stats(u64 **data, struct idpf_queue *q)
 	 * u64_stats_fetch_retry.
 	 */
 	do {
-		start = u64_stats_fetch_begin(&q->stats_sync);
+		start = u64_stats_fetch_begin(stats_sync);
 		for (i = 0; i < size; i++)
 			idpf_add_one_ethtool_stat(&(*data)[i], q, &stats[i]);
-	} while (u64_stats_fetch_retry(&q->stats_sync, start));
+	} while (u64_stats_fetch_retry(stats_sync, start));
 
 	/* Once we successfully copy the stats in, update the data pointer */
 	*data += size;
@@ -793,7 +800,7 @@ static void idpf_collect_queue_stats(struct idpf_vport *vport)
 		for (j = 0; j < num_rxq; j++) {
 			u64 hw_csum_err, hsplit, hsplit_hbo, bad_descs;
 			struct idpf_rx_queue_stats *stats;
-			struct idpf_queue *rxq;
+			struct idpf_rx_queue *rxq;
 			unsigned int start;
 
 			if (idpf_is_queue_model_split(vport->rxq_model))
@@ -807,7 +814,7 @@ static void idpf_collect_queue_stats(struct idpf_vport *vport)
 			do {
 				start = u64_stats_fetch_begin(&rxq->stats_sync);
 
-				stats = &rxq->q_stats.rx;
+				stats = &rxq->q_stats;
 				hw_csum_err = u64_stats_read(&stats->hw_csum_err);
 				hsplit = u64_stats_read(&stats->hsplit_pkts);
 				hsplit_hbo = u64_stats_read(&stats->hsplit_buf_ovf);
@@ -828,7 +835,7 @@ static void idpf_collect_queue_stats(struct idpf_vport *vport)
 
 		for (j = 0; j < txq_grp->num_txq; j++) {
 			u64 linearize, qbusy, skb_drops, dma_map_errs;
-			struct idpf_queue *txq = txq_grp->txqs[j];
+			struct idpf_tx_queue *txq = txq_grp->txqs[j];
 			struct idpf_tx_queue_stats *stats;
 			unsigned int start;
 
@@ -838,7 +845,7 @@ static void idpf_collect_queue_stats(struct idpf_vport *vport)
 			do {
 				start = u64_stats_fetch_begin(&txq->stats_sync);
 
-				stats = &txq->q_stats.tx;
+				stats = &txq->q_stats;
 				linearize = u64_stats_read(&stats->linearize);
 				qbusy = u64_stats_read(&stats->q_busy);
 				skb_drops = u64_stats_read(&stats->skb_drops);
@@ -896,12 +903,12 @@ static void idpf_get_ethtool_stats(struct net_device *netdev,
 		qtype = VIRTCHNL2_QUEUE_TYPE_TX;
 
 		for (j = 0; j < txq_grp->num_txq; j++, total++) {
-			struct idpf_queue *txq = txq_grp->txqs[j];
+			struct idpf_tx_queue *txq = txq_grp->txqs[j];
 
 			if (!txq)
 				idpf_add_empty_queue_stats(&data, qtype);
 			else
-				idpf_add_queue_stats(&data, txq);
+				idpf_add_queue_stats(&data, txq, qtype);
 		}
 	}
 
@@ -929,7 +936,7 @@ static void idpf_get_ethtool_stats(struct net_device *netdev,
 			num_rxq = rxq_grp->singleq.num_rxq;
 
 		for (j = 0; j < num_rxq; j++, total++) {
-			struct idpf_queue *rxq;
+			struct idpf_rx_queue *rxq;
 
 			if (is_splitq)
 				rxq = &rxq_grp->splitq.rxq_sets[j]->rxq;
@@ -938,7 +945,7 @@ static void idpf_get_ethtool_stats(struct net_device *netdev,
 			if (!rxq)
 				idpf_add_empty_queue_stats(&data, qtype);
 			else
-				idpf_add_queue_stats(&data, rxq);
+				idpf_add_queue_stats(&data, rxq, qtype);
 
 			/* In splitq mode, don't get page pool stats here since
 			 * the pools are attached to the buffer queues
@@ -953,7 +960,7 @@ static void idpf_get_ethtool_stats(struct net_device *netdev,
 
 	for (i = 0; i < vport->num_rxq_grp; i++) {
 		for (j = 0; j < vport->num_bufqs_per_qgrp; j++) {
-			struct idpf_queue *rxbufq =
+			struct idpf_buf_queue *rxbufq =
 				&vport->rxq_grps[i].splitq.bufq_sets[j].bufq;
 
 			page_pool_get_stats(rxbufq->pp, &pp_stats);
@@ -971,60 +978,64 @@ static void idpf_get_ethtool_stats(struct net_device *netdev,
 }
 
 /**
- * idpf_find_rxq - find rxq from q index
+ * idpf_find_rxq_vec - find rxq vector from q index
  * @vport: virtual port associated to queue
  * @q_num: q index used to find queue
  *
- * returns pointer to rx queue
+ * returns pointer to rx vector
  */
-static struct idpf_queue *idpf_find_rxq(struct idpf_vport *vport, int q_num)
+static struct idpf_q_vector *idpf_find_rxq_vec(const struct idpf_vport *vport,
+					       int q_num)
 {
 	int q_grp, q_idx;
 
 	if (!idpf_is_queue_model_split(vport->rxq_model))
-		return vport->rxq_grps->singleq.rxqs[q_num];
+		return vport->rxq_grps->singleq.rxqs[q_num]->q_vector;
 
 	q_grp = q_num / IDPF_DFLT_SPLITQ_RXQ_PER_GROUP;
 	q_idx = q_num % IDPF_DFLT_SPLITQ_RXQ_PER_GROUP;
 
-	return &vport->rxq_grps[q_grp].splitq.rxq_sets[q_idx]->rxq;
+	return vport->rxq_grps[q_grp].splitq.rxq_sets[q_idx]->rxq.q_vector;
 }
 
 /**
- * idpf_find_txq - find txq from q index
+ * idpf_find_txq_vec - find txq vector from q index
  * @vport: virtual port associated to queue
  * @q_num: q index used to find queue
  *
- * returns pointer to tx queue
+ * returns pointer to tx vector
  */
-static struct idpf_queue *idpf_find_txq(struct idpf_vport *vport, int q_num)
+static struct idpf_q_vector *idpf_find_txq_vec(const struct idpf_vport *vport,
+					       int q_num)
 {
 	int q_grp;
 
 	if (!idpf_is_queue_model_split(vport->txq_model))
-		return vport->txqs[q_num];
+		return vport->txqs[q_num]->q_vector;
 
 	q_grp = q_num / IDPF_DFLT_SPLITQ_TXQ_PER_GROUP;
 
-	return vport->txq_grps[q_grp].complq;
+	return vport->txq_grps[q_grp].complq->q_vector;
 }
 
 /**
  * __idpf_get_q_coalesce - get ITR values for specific queue
  * @ec: ethtool structure to fill with driver's coalesce settings
- * @q: quuee of Rx or Tx
+ * @q_vector: queue vector corresponding to this queue
+ * @type: queue type
  */
 static void __idpf_get_q_coalesce(struct ethtool_coalesce *ec,
-				  struct idpf_queue *q)
+				  const struct idpf_q_vector *q_vector,
+				  enum virtchnl2_queue_type type)
 {
-	if (q->q_type == VIRTCHNL2_QUEUE_TYPE_RX) {
+	if (type == VIRTCHNL2_QUEUE_TYPE_RX) {
 		ec->use_adaptive_rx_coalesce =
-				IDPF_ITR_IS_DYNAMIC(q->q_vector->rx_intr_mode);
-		ec->rx_coalesce_usecs = q->q_vector->rx_itr_value;
+				IDPF_ITR_IS_DYNAMIC(q_vector->rx_intr_mode);
+		ec->rx_coalesce_usecs = q_vector->rx_itr_value;
 	} else {
 		ec->use_adaptive_tx_coalesce =
-				IDPF_ITR_IS_DYNAMIC(q->q_vector->tx_intr_mode);
-		ec->tx_coalesce_usecs = q->q_vector->tx_itr_value;
+				IDPF_ITR_IS_DYNAMIC(q_vector->tx_intr_mode);
+		ec->tx_coalesce_usecs = q_vector->tx_itr_value;
 	}
 }
 
@@ -1040,8 +1051,8 @@ static int idpf_get_q_coalesce(struct net_device *netdev,
 			       struct ethtool_coalesce *ec,
 			       u32 q_num)
 {
-	struct idpf_netdev_priv *np = netdev_priv(netdev);
-	struct idpf_vport *vport;
+	const struct idpf_netdev_priv *np = netdev_priv(netdev);
+	const struct idpf_vport *vport;
 	int err = 0;
 
 	idpf_vport_ctrl_lock(netdev);
@@ -1056,10 +1067,12 @@ static int idpf_get_q_coalesce(struct net_device *netdev,
 	}
 
 	if (q_num < vport->num_rxq)
-		__idpf_get_q_coalesce(ec, idpf_find_rxq(vport, q_num));
+		__idpf_get_q_coalesce(ec, idpf_find_rxq_vec(vport, q_num),
+				      VIRTCHNL2_QUEUE_TYPE_RX);
 
 	if (q_num < vport->num_txq)
-		__idpf_get_q_coalesce(ec, idpf_find_txq(vport, q_num));
+		__idpf_get_q_coalesce(ec, idpf_find_txq_vec(vport, q_num),
+				      VIRTCHNL2_QUEUE_TYPE_TX);
 
 unlock_mutex:
 	idpf_vport_ctrl_unlock(netdev);
@@ -1103,16 +1116,15 @@ static int idpf_get_per_q_coalesce(struct net_device *netdev, u32 q_num,
 /**
  * __idpf_set_q_coalesce - set ITR values for specific queue
  * @ec: ethtool structure from user to update ITR settings
- * @q: queue for which itr values has to be set
+ * @qv: queue vector for which itr values has to be set
  * @is_rxq: is queue type rx
  *
  * Returns 0 on success, negative otherwise.
  */
-static int __idpf_set_q_coalesce(struct ethtool_coalesce *ec,
-				 struct idpf_queue *q, bool is_rxq)
+static int __idpf_set_q_coalesce(const struct ethtool_coalesce *ec,
+				 struct idpf_q_vector *qv, bool is_rxq)
 {
 	u32 use_adaptive_coalesce, coalesce_usecs;
-	struct idpf_q_vector *qv = q->q_vector;
 	bool is_dim_ena = false;
 	u16 itr_val;
 
@@ -1128,7 +1140,7 @@ static int __idpf_set_q_coalesce(struct ethtool_coalesce *ec,
 		itr_val = qv->tx_itr_value;
 	}
 	if (coalesce_usecs != itr_val && use_adaptive_coalesce) {
-		netdev_err(q->vport->netdev, "Cannot set coalesce usecs if adaptive enabled\n");
+		netdev_err(qv->vport->netdev, "Cannot set coalesce usecs if adaptive enabled\n");
 
 		return -EINVAL;
 	}
@@ -1137,7 +1149,7 @@ static int __idpf_set_q_coalesce(struct ethtool_coalesce *ec,
 		return 0;
 
 	if (coalesce_usecs > IDPF_ITR_MAX) {
-		netdev_err(q->vport->netdev,
+		netdev_err(qv->vport->netdev,
 			   "Invalid value, %d-usecs range is 0-%d\n",
 			   coalesce_usecs, IDPF_ITR_MAX);
 
@@ -1146,7 +1158,7 @@ static int __idpf_set_q_coalesce(struct ethtool_coalesce *ec,
 
 	if (coalesce_usecs % 2) {
 		coalesce_usecs--;
-		netdev_info(q->vport->netdev,
+		netdev_info(qv->vport->netdev,
 			    "HW only supports even ITR values, ITR rounded to %d\n",
 			    coalesce_usecs);
 	}
@@ -1185,15 +1197,16 @@ static int __idpf_set_q_coalesce(struct ethtool_coalesce *ec,
  *
  * Return 0 on success, and negative on failure
  */
-static int idpf_set_q_coalesce(struct idpf_vport *vport,
-			       struct ethtool_coalesce *ec,
+static int idpf_set_q_coalesce(const struct idpf_vport *vport,
+			       const struct ethtool_coalesce *ec,
 			       int q_num, bool is_rxq)
 {
-	struct idpf_queue *q;
+	struct idpf_q_vector *qv;
 
-	q = is_rxq ? idpf_find_rxq(vport, q_num) : idpf_find_txq(vport, q_num);
+	qv = is_rxq ? idpf_find_rxq_vec(vport, q_num) :
+		      idpf_find_txq_vec(vport, q_num);
 
-	if (q && __idpf_set_q_coalesce(ec, q, is_rxq))
+	if (qv && __idpf_set_q_coalesce(ec, qv, is_rxq))
 		return -EINVAL;
 
 	return 0;
diff --git a/drivers/net/ethernet/intel/idpf/idpf_lib.c b/drivers/net/ethernet/intel/idpf/idpf_lib.c
index f1ee5584e8fa..a5612120ad51 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_lib.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_lib.c
@@ -1318,14 +1318,14 @@ static void idpf_rx_init_buf_tail(struct idpf_vport *vport)
 
 		if (idpf_is_queue_model_split(vport->rxq_model)) {
 			for (j = 0; j < vport->num_bufqs_per_qgrp; j++) {
-				struct idpf_queue *q =
+				const struct idpf_buf_queue *q =
 					&grp->splitq.bufq_sets[j].bufq;
 
 				writel(q->next_to_alloc, q->tail);
 			}
 		} else {
 			for (j = 0; j < grp->singleq.num_rxq; j++) {
-				struct idpf_queue *q =
+				const struct idpf_rx_queue *q =
 					grp->singleq.rxqs[j];
 
 				writel(q->next_to_alloc, q->tail);
@@ -1855,7 +1855,7 @@ int idpf_initiate_soft_reset(struct idpf_vport *vport,
 	enum idpf_vport_state current_state = np->state;
 	struct idpf_adapter *adapter = vport->adapter;
 	struct idpf_vport *new_vport;
-	int err, i;
+	int err;
 
 	/* If the system is low on memory, we can end up in bad state if we
 	 * free all the memory for queue resources and try to allocate them
@@ -1929,46 +1929,6 @@ int idpf_initiate_soft_reset(struct idpf_vport *vport,
 	 */
 	memcpy(vport, new_vport, offsetof(struct idpf_vport, link_speed_mbps));
 
-	/* Since idpf_vport_queues_alloc was called with new_port, the queue
-	 * back pointers are currently pointing to the local new_vport. Reset
-	 * the backpointers to the original vport here
-	 */
-	for (i = 0; i < vport->num_txq_grp; i++) {
-		struct idpf_txq_group *tx_qgrp = &vport->txq_grps[i];
-		int j;
-
-		tx_qgrp->vport = vport;
-		for (j = 0; j < tx_qgrp->num_txq; j++)
-			tx_qgrp->txqs[j]->vport = vport;
-
-		if (idpf_is_queue_model_split(vport->txq_model))
-			tx_qgrp->complq->vport = vport;
-	}
-
-	for (i = 0; i < vport->num_rxq_grp; i++) {
-		struct idpf_rxq_group *rx_qgrp = &vport->rxq_grps[i];
-		struct idpf_queue *q;
-		u16 num_rxq;
-		int j;
-
-		rx_qgrp->vport = vport;
-		for (j = 0; j < vport->num_bufqs_per_qgrp; j++)
-			rx_qgrp->splitq.bufq_sets[j].bufq.vport = vport;
-
-		if (idpf_is_queue_model_split(vport->rxq_model))
-			num_rxq = rx_qgrp->splitq.num_rxq_sets;
-		else
-			num_rxq = rx_qgrp->singleq.num_rxq;
-
-		for (j = 0; j < num_rxq; j++) {
-			if (idpf_is_queue_model_split(vport->rxq_model))
-				q = &rx_qgrp->splitq.rxq_sets[j]->rxq;
-			else
-				q = rx_qgrp->singleq.rxqs[j];
-			q->vport = vport;
-		}
-	}
-
 	if (reset_cause == IDPF_SR_Q_CHANGE)
 		idpf_vport_alloc_vec_indexes(vport);
 
diff --git a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
index b17d88e15000..9864a3992f0c 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
@@ -186,7 +186,7 @@ static int idpf_tx_singleq_csum(struct sk_buff *skb,
  * and gets a physical address for each memory location and programs
  * it and the length into the transmit base mode descriptor.
  */
-static void idpf_tx_singleq_map(struct idpf_queue *tx_q,
+static void idpf_tx_singleq_map(struct idpf_tx_queue *tx_q,
 				struct idpf_tx_buf *first,
 				struct idpf_tx_offload_params *offloads)
 {
@@ -210,7 +210,7 @@ static void idpf_tx_singleq_map(struct idpf_queue *tx_q,
 	dma = dma_map_single(tx_q->dev, skb->data, size, DMA_TO_DEVICE);
 
 	/* write each descriptor with CRC bit */
-	if (tx_q->vport->crc_enable)
+	if (idpf_queue_has(CRC_EN, tx_q))
 		td_cmd |= IDPF_TX_DESC_CMD_ICRC;
 
 	for (frag = &skb_shinfo(skb)->frags[0];; frag++) {
@@ -285,7 +285,7 @@ static void idpf_tx_singleq_map(struct idpf_queue *tx_q,
 	/* set next_to_watch value indicating a packet is present */
 	first->next_to_watch = tx_desc;
 
-	nq = netdev_get_tx_queue(tx_q->vport->netdev, tx_q->idx);
+	nq = netdev_get_tx_queue(tx_q->netdev, tx_q->idx);
 	netdev_tx_sent_queue(nq, first->bytecount);
 
 	idpf_tx_buf_hw_update(tx_q, i, netdev_xmit_more());
@@ -299,7 +299,7 @@ static void idpf_tx_singleq_map(struct idpf_queue *tx_q,
  * ring entry to reflect that this index is a context descriptor
  */
 static struct idpf_base_tx_ctx_desc *
-idpf_tx_singleq_get_ctx_desc(struct idpf_queue *txq)
+idpf_tx_singleq_get_ctx_desc(struct idpf_tx_queue *txq)
 {
 	struct idpf_base_tx_ctx_desc *ctx_desc;
 	int ntu = txq->next_to_use;
@@ -320,7 +320,7 @@ idpf_tx_singleq_get_ctx_desc(struct idpf_queue *txq)
  * @txq: queue to send buffer on
  * @offload: offload parameter structure
  **/
-static void idpf_tx_singleq_build_ctx_desc(struct idpf_queue *txq,
+static void idpf_tx_singleq_build_ctx_desc(struct idpf_tx_queue *txq,
 					   struct idpf_tx_offload_params *offload)
 {
 	struct idpf_base_tx_ctx_desc *desc = idpf_tx_singleq_get_ctx_desc(txq);
@@ -333,7 +333,7 @@ static void idpf_tx_singleq_build_ctx_desc(struct idpf_queue *txq,
 		qw1 |= FIELD_PREP(IDPF_TXD_CTX_QW1_MSS_M, offload->mss);
 
 		u64_stats_update_begin(&txq->stats_sync);
-		u64_stats_inc(&txq->q_stats.tx.lso_pkts);
+		u64_stats_inc(&txq->q_stats.lso_pkts);
 		u64_stats_update_end(&txq->stats_sync);
 	}
 
@@ -352,7 +352,7 @@ static void idpf_tx_singleq_build_ctx_desc(struct idpf_queue *txq,
  * Returns NETDEV_TX_OK if sent, else an error code
  */
 static netdev_tx_t idpf_tx_singleq_frame(struct sk_buff *skb,
-					 struct idpf_queue *tx_q)
+					 struct idpf_tx_queue *tx_q)
 {
 	struct idpf_tx_offload_params offload = { };
 	struct idpf_tx_buf *first;
@@ -419,7 +419,7 @@ netdev_tx_t idpf_tx_singleq_start(struct sk_buff *skb,
 				  struct net_device *netdev)
 {
 	struct idpf_vport *vport = idpf_netdev_to_vport(netdev);
-	struct idpf_queue *tx_q;
+	struct idpf_tx_queue *tx_q;
 
 	tx_q = vport->txqs[skb_get_queue_mapping(skb)];
 
@@ -442,16 +442,15 @@ netdev_tx_t idpf_tx_singleq_start(struct sk_buff *skb,
  * @cleaned: returns number of packets cleaned
  *
  */
-static bool idpf_tx_singleq_clean(struct idpf_queue *tx_q, int napi_budget,
+static bool idpf_tx_singleq_clean(struct idpf_tx_queue *tx_q, int napi_budget,
 				  int *cleaned)
 {
-	unsigned int budget = tx_q->vport->compln_clean_budget;
 	unsigned int total_bytes = 0, total_pkts = 0;
 	struct idpf_base_tx_desc *tx_desc;
+	u32 budget = tx_q->clean_budget;
 	s16 ntc = tx_q->next_to_clean;
 	struct idpf_netdev_priv *np;
 	struct idpf_tx_buf *tx_buf;
-	struct idpf_vport *vport;
 	struct netdev_queue *nq;
 	bool dont_wake;
 
@@ -550,16 +549,15 @@ fetch_next_txq_desc:
 	*cleaned += total_pkts;
 
 	u64_stats_update_begin(&tx_q->stats_sync);
-	u64_stats_add(&tx_q->q_stats.tx.packets, total_pkts);
-	u64_stats_add(&tx_q->q_stats.tx.bytes, total_bytes);
+	u64_stats_add(&tx_q->q_stats.packets, total_pkts);
+	u64_stats_add(&tx_q->q_stats.bytes, total_bytes);
 	u64_stats_update_end(&tx_q->stats_sync);
 
-	vport = tx_q->vport;
-	np = netdev_priv(vport->netdev);
-	nq = netdev_get_tx_queue(vport->netdev, tx_q->idx);
+	np = netdev_priv(tx_q->netdev);
+	nq = netdev_get_tx_queue(tx_q->netdev, tx_q->idx);
 
 	dont_wake = np->state != __IDPF_VPORT_UP ||
-		    !netif_carrier_ok(vport->netdev);
+		    !netif_carrier_ok(tx_q->netdev);
 	__netif_txq_completed_wake(nq, total_pkts, total_bytes,
 				   IDPF_DESC_UNUSED(tx_q), IDPF_TX_WAKE_THRESH,
 				   dont_wake);
@@ -584,7 +582,7 @@ static bool idpf_tx_singleq_clean_all(struct idpf_q_vector *q_vec, int budget,
 
 	budget_per_q = num_txq ? max(budget / num_txq, 1) : 0;
 	for (i = 0; i < num_txq; i++) {
-		struct idpf_queue *q;
+		struct idpf_tx_queue *q;
 
 		q = q_vec->tx[i];
 		clean_complete &= idpf_tx_singleq_clean(q, budget_per_q,
@@ -614,14 +612,9 @@ static bool idpf_rx_singleq_test_staterr(const union virtchnl2_rx_desc *rx_desc,
 
 /**
  * idpf_rx_singleq_is_non_eop - process handling of non-EOP buffers
- * @rxq: Rx ring being processed
  * @rx_desc: Rx descriptor for current buffer
- * @skb: Current socket buffer containing buffer in progress
- * @ntc: next to clean
  */
-static bool idpf_rx_singleq_is_non_eop(struct idpf_queue *rxq,
-				       union virtchnl2_rx_desc *rx_desc,
-				       struct sk_buff *skb, u16 ntc)
+static bool idpf_rx_singleq_is_non_eop(const union virtchnl2_rx_desc *rx_desc)
 {
 	/* if we are the last buffer then there is nothing else to do */
 	if (likely(idpf_rx_singleq_test_staterr(rx_desc, IDPF_RXD_EOF_SINGLEQ)))
@@ -639,7 +632,7 @@ static bool idpf_rx_singleq_is_non_eop(struct idpf_queue *rxq,
  *
  * skb->protocol must be set before this function is called
  */
-static void idpf_rx_singleq_csum(struct idpf_queue *rxq, struct sk_buff *skb,
+static void idpf_rx_singleq_csum(struct idpf_rx_queue *rxq, struct sk_buff *skb,
 				 struct idpf_rx_csum_decoded *csum_bits,
 				 u16 ptype)
 {
@@ -647,14 +640,14 @@ static void idpf_rx_singleq_csum(struct idpf_queue *rxq, struct sk_buff *skb,
 	bool ipv4, ipv6;
 
 	/* check if Rx checksum is enabled */
-	if (unlikely(!(rxq->vport->netdev->features & NETIF_F_RXCSUM)))
+	if (unlikely(!(rxq->netdev->features & NETIF_F_RXCSUM)))
 		return;
 
 	/* check if HW has decoded the packet and checksum */
 	if (unlikely(!(csum_bits->l3l4p)))
 		return;
 
-	decoded = rxq->vport->rx_ptype_lkup[ptype];
+	decoded = rxq->rx_ptype_lkup[ptype];
 	if (unlikely(!(decoded.known && decoded.outer_ip)))
 		return;
 
@@ -707,7 +700,7 @@ static void idpf_rx_singleq_csum(struct idpf_queue *rxq, struct sk_buff *skb,
 
 checksum_fail:
 	u64_stats_update_begin(&rxq->stats_sync);
-	u64_stats_inc(&rxq->q_stats.rx.hw_csum_err);
+	u64_stats_inc(&rxq->q_stats.hw_csum_err);
 	u64_stats_update_end(&rxq->stats_sync);
 }
 
@@ -721,9 +714,9 @@ checksum_fail:
  * This function only operates on the VIRTCHNL2_RXDID_1_32B_BASE_M base 32byte
  * descriptor writeback format.
  **/
-static void idpf_rx_singleq_base_csum(struct idpf_queue *rx_q,
+static void idpf_rx_singleq_base_csum(struct idpf_rx_queue *rx_q,
 				      struct sk_buff *skb,
-				      union virtchnl2_rx_desc *rx_desc,
+				      const union virtchnl2_rx_desc *rx_desc,
 				      u16 ptype)
 {
 	struct idpf_rx_csum_decoded csum_bits;
@@ -761,9 +754,9 @@ static void idpf_rx_singleq_base_csum(struct idpf_queue *rx_q,
  * This function only operates on the VIRTCHNL2_RXDID_2_FLEX_SQ_NIC flexible
  * descriptor writeback format.
  **/
-static void idpf_rx_singleq_flex_csum(struct idpf_queue *rx_q,
+static void idpf_rx_singleq_flex_csum(struct idpf_rx_queue *rx_q,
 				      struct sk_buff *skb,
-				      union virtchnl2_rx_desc *rx_desc,
+				      const union virtchnl2_rx_desc *rx_desc,
 				      u16 ptype)
 {
 	struct idpf_rx_csum_decoded csum_bits;
@@ -801,14 +794,14 @@ static void idpf_rx_singleq_flex_csum(struct idpf_queue *rx_q,
  * This function only operates on the VIRTCHNL2_RXDID_1_32B_BASE_M base 32byte
  * descriptor writeback format.
  **/
-static void idpf_rx_singleq_base_hash(struct idpf_queue *rx_q,
+static void idpf_rx_singleq_base_hash(struct idpf_rx_queue *rx_q,
 				      struct sk_buff *skb,
-				      union virtchnl2_rx_desc *rx_desc,
+				      const union virtchnl2_rx_desc *rx_desc,
 				      struct idpf_rx_ptype_decoded *decoded)
 {
 	u64 mask, qw1;
 
-	if (unlikely(!(rx_q->vport->netdev->features & NETIF_F_RXHASH)))
+	if (unlikely(!(rx_q->netdev->features & NETIF_F_RXHASH)))
 		return;
 
 	mask = VIRTCHNL2_RX_BASE_DESC_FLTSTAT_RSS_HASH_M;
@@ -831,12 +824,12 @@ static void idpf_rx_singleq_base_hash(struct idpf_queue *rx_q,
  * This function only operates on the VIRTCHNL2_RXDID_2_FLEX_SQ_NIC flexible
  * descriptor writeback format.
  **/
-static void idpf_rx_singleq_flex_hash(struct idpf_queue *rx_q,
+static void idpf_rx_singleq_flex_hash(struct idpf_rx_queue *rx_q,
 				      struct sk_buff *skb,
-				      union virtchnl2_rx_desc *rx_desc,
+				      const union virtchnl2_rx_desc *rx_desc,
 				      struct idpf_rx_ptype_decoded *decoded)
 {
-	if (unlikely(!(rx_q->vport->netdev->features & NETIF_F_RXHASH)))
+	if (unlikely(!(rx_q->netdev->features & NETIF_F_RXHASH)))
 		return;
 
 	if (FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_STATUS0_RSS_VALID_M,
@@ -857,16 +850,16 @@ static void idpf_rx_singleq_flex_hash(struct idpf_queue *rx_q,
  * order to populate the hash, checksum, VLAN, protocol, and
  * other fields within the skb.
  */
-static void idpf_rx_singleq_process_skb_fields(struct idpf_queue *rx_q,
-					       struct sk_buff *skb,
-					       union virtchnl2_rx_desc *rx_desc,
-					       u16 ptype)
+static void
+idpf_rx_singleq_process_skb_fields(struct idpf_rx_queue *rx_q,
+				   struct sk_buff *skb,
+				   const union virtchnl2_rx_desc *rx_desc,
+				   u16 ptype)
 {
-	struct idpf_rx_ptype_decoded decoded =
-					rx_q->vport->rx_ptype_lkup[ptype];
+	struct idpf_rx_ptype_decoded decoded = rx_q->rx_ptype_lkup[ptype];
 
 	/* modifies the skb - consumes the enet header */
-	skb->protocol = eth_type_trans(skb, rx_q->vport->netdev);
+	skb->protocol = eth_type_trans(skb, rx_q->netdev);
 
 	/* Check if we're using base mode descriptor IDs */
 	if (rx_q->rxdids == VIRTCHNL2_RXDID_1_32B_BASE_M) {
@@ -878,6 +871,22 @@ static void idpf_rx_singleq_process_skb_fields(struct idpf_queue *rx_q,
 	}
 }
 
+/**
+ * idpf_rx_buf_hw_update - Store the new tail and head values
+ * @rxq: queue to bump
+ * @val: new head index
+ */
+static void idpf_rx_buf_hw_update(struct idpf_rx_queue *rxq, u32 val)
+{
+	rxq->next_to_use = val;
+
+	if (unlikely(!rxq->tail))
+		return;
+
+	/* writel has an implicit memory barrier */
+	writel(val, rxq->tail);
+}
+
 /**
  * idpf_rx_singleq_buf_hw_alloc_all - Replace used receive buffers
  * @rx_q: queue for which the hw buffers are allocated
@@ -885,7 +894,7 @@ static void idpf_rx_singleq_process_skb_fields(struct idpf_queue *rx_q,
  *
  * Returns false if all allocations were successful, true if any fail
  */
-bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_queue *rx_q,
+bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_rx_queue *rx_q,
 				      u16 cleaned_count)
 {
 	struct virtchnl2_singleq_rx_buf_desc *desc;
@@ -896,7 +905,7 @@ bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_queue *rx_q,
 		return false;
 
 	desc = &rx_q->single_buf[nta];
-	buf = &rx_q->rx_buf.buf[nta];
+	buf = &rx_q->rx_buf[nta];
 
 	do {
 		dma_addr_t addr;
@@ -916,7 +925,7 @@ bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_queue *rx_q,
 		nta++;
 		if (unlikely(nta == rx_q->desc_count)) {
 			desc = &rx_q->single_buf[0];
-			buf = rx_q->rx_buf.buf;
+			buf = rx_q->rx_buf;
 			nta = 0;
 		}
 
@@ -933,7 +942,6 @@ bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_queue *rx_q,
 
 /**
  * idpf_rx_singleq_extract_base_fields - Extract fields from the Rx descriptor
- * @rx_q: Rx descriptor queue
  * @rx_desc: the descriptor to process
  * @fields: storage for extracted values
  *
@@ -943,9 +951,9 @@ bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_queue *rx_q,
  * This function only operates on the VIRTCHNL2_RXDID_1_32B_BASE_M base 32byte
  * descriptor writeback format.
  */
-static void idpf_rx_singleq_extract_base_fields(struct idpf_queue *rx_q,
-						union virtchnl2_rx_desc *rx_desc,
-						struct idpf_rx_extracted *fields)
+static void
+idpf_rx_singleq_extract_base_fields(const union virtchnl2_rx_desc *rx_desc,
+				    struct idpf_rx_extracted *fields)
 {
 	u64 qword;
 
@@ -957,7 +965,6 @@ static void idpf_rx_singleq_extract_base_fields(struct idpf_queue *rx_q,
 
 /**
  * idpf_rx_singleq_extract_flex_fields - Extract fields from the Rx descriptor
- * @rx_q: Rx descriptor queue
  * @rx_desc: the descriptor to process
  * @fields: storage for extracted values
  *
@@ -967,9 +974,9 @@ static void idpf_rx_singleq_extract_base_fields(struct idpf_queue *rx_q,
  * This function only operates on the VIRTCHNL2_RXDID_2_FLEX_SQ_NIC flexible
  * descriptor writeback format.
  */
-static void idpf_rx_singleq_extract_flex_fields(struct idpf_queue *rx_q,
-						union virtchnl2_rx_desc *rx_desc,
-						struct idpf_rx_extracted *fields)
+static void
+idpf_rx_singleq_extract_flex_fields(const union virtchnl2_rx_desc *rx_desc,
+				    struct idpf_rx_extracted *fields)
 {
 	fields->size = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_PKT_LEN_M,
 				 le16_to_cpu(rx_desc->flex_nic_wb.pkt_len));
@@ -984,14 +991,15 @@ static void idpf_rx_singleq_extract_flex_fields(struct idpf_queue *rx_q,
  * @fields: storage for extracted values
  *
  */
-static void idpf_rx_singleq_extract_fields(struct idpf_queue *rx_q,
-					   union virtchnl2_rx_desc *rx_desc,
-					   struct idpf_rx_extracted *fields)
+static void
+idpf_rx_singleq_extract_fields(const struct idpf_rx_queue *rx_q,
+			       const union virtchnl2_rx_desc *rx_desc,
+			       struct idpf_rx_extracted *fields)
 {
 	if (rx_q->rxdids == VIRTCHNL2_RXDID_1_32B_BASE_M)
-		idpf_rx_singleq_extract_base_fields(rx_q, rx_desc, fields);
+		idpf_rx_singleq_extract_base_fields(rx_desc, fields);
 	else
-		idpf_rx_singleq_extract_flex_fields(rx_q, rx_desc, fields);
+		idpf_rx_singleq_extract_flex_fields(rx_desc, fields);
 }
 
 /**
@@ -1001,7 +1009,7 @@ static void idpf_rx_singleq_extract_fields(struct idpf_queue *rx_q,
  *
  * Returns true if there's any budget left (e.g. the clean is finished)
  */
-static int idpf_rx_singleq_clean(struct idpf_queue *rx_q, int budget)
+static int idpf_rx_singleq_clean(struct idpf_rx_queue *rx_q, int budget)
 {
 	unsigned int total_rx_bytes = 0, total_rx_pkts = 0;
 	struct sk_buff *skb = rx_q->skb;
@@ -1036,7 +1044,7 @@ static int idpf_rx_singleq_clean(struct idpf_queue *rx_q, int budget)
 
 		idpf_rx_singleq_extract_fields(rx_q, rx_desc, &fields);
 
-		rx_buf = &rx_q->rx_buf.buf[ntc];
+		rx_buf = &rx_q->rx_buf[ntc];
 		if (!fields.size) {
 			idpf_rx_put_page(rx_buf);
 			goto skip_data;
@@ -1058,7 +1066,7 @@ skip_data:
 		cleaned_count++;
 
 		/* skip if it is non EOP desc */
-		if (idpf_rx_singleq_is_non_eop(rx_q, rx_desc, skb, ntc))
+		if (idpf_rx_singleq_is_non_eop(rx_desc))
 			continue;
 
 #define IDPF_RXD_ERR_S FIELD_PREP(VIRTCHNL2_RX_BASE_DESC_QW1_ERROR_M, \
@@ -1084,7 +1092,7 @@ skip_data:
 						   rx_desc, fields.rx_ptype);
 
 		/* send completed skb up the stack */
-		napi_gro_receive(&rx_q->q_vector->napi, skb);
+		napi_gro_receive(rx_q->pp->p.napi, skb);
 		skb = NULL;
 
 		/* update budget accounting */
@@ -1099,8 +1107,8 @@ skip_data:
 		failure = idpf_rx_singleq_buf_hw_alloc_all(rx_q, cleaned_count);
 
 	u64_stats_update_begin(&rx_q->stats_sync);
-	u64_stats_add(&rx_q->q_stats.rx.packets, total_rx_pkts);
-	u64_stats_add(&rx_q->q_stats.rx.bytes, total_rx_bytes);
+	u64_stats_add(&rx_q->q_stats.packets, total_rx_pkts);
+	u64_stats_add(&rx_q->q_stats.bytes, total_rx_bytes);
 	u64_stats_update_end(&rx_q->stats_sync);
 
 	/* guarantee a trip back through this routine if there was a failure */
@@ -1127,7 +1135,7 @@ static bool idpf_rx_singleq_clean_all(struct idpf_q_vector *q_vec, int budget,
 	 */
 	budget_per_q = num_rxq ? max(budget / num_rxq, 1) : 0;
 	for (i = 0; i < num_rxq; i++) {
-		struct idpf_queue *rxq = q_vec->rx[i];
+		struct idpf_rx_queue *rxq = q_vec->rx[i];
 		int pkts_cleaned_per_q;
 
 		pkts_cleaned_per_q = idpf_rx_singleq_clean(rxq, budget_per_q);
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
index 01301e640fba..7be5a723f558 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
@@ -60,7 +60,8 @@ void idpf_tx_timeout(struct net_device *netdev, unsigned int txqueue)
  * @tx_q: the queue that owns the buffer
  * @tx_buf: the buffer to free
  */
-static void idpf_tx_buf_rel(struct idpf_queue *tx_q, struct idpf_tx_buf *tx_buf)
+static void idpf_tx_buf_rel(struct idpf_tx_queue *tx_q,
+			    struct idpf_tx_buf *tx_buf)
 {
 	if (tx_buf->skb) {
 		if (dma_unmap_len(tx_buf, len))
@@ -86,8 +87,9 @@ static void idpf_tx_buf_rel(struct idpf_queue *tx_q, struct idpf_tx_buf *tx_buf)
  * idpf_tx_buf_rel_all - Free any empty Tx buffers
  * @txq: queue to be cleaned
  */
-static void idpf_tx_buf_rel_all(struct idpf_queue *txq)
+static void idpf_tx_buf_rel_all(struct idpf_tx_queue *txq)
 {
+	struct idpf_buf_lifo *buf_stack;
 	u16 i;
 
 	/* Buffers already cleared, nothing to do */
@@ -101,38 +103,57 @@ static void idpf_tx_buf_rel_all(struct idpf_queue *txq)
 	kfree(txq->tx_buf);
 	txq->tx_buf = NULL;
 
-	if (!txq->buf_stack.bufs)
+	if (!idpf_queue_has(FLOW_SCH_EN, txq))
 		return;
 
-	for (i = 0; i < txq->buf_stack.size; i++)
-		kfree(txq->buf_stack.bufs[i]);
+	buf_stack = &txq->stash->buf_stack;
+	if (!buf_stack->bufs)
+		return;
+
+	for (i = 0; i < buf_stack->size; i++)
+		kfree(buf_stack->bufs[i]);
 
-	kfree(txq->buf_stack.bufs);
-	txq->buf_stack.bufs = NULL;
+	kfree(buf_stack->bufs);
+	buf_stack->bufs = NULL;
 }
 
 /**
  * idpf_tx_desc_rel - Free Tx resources per queue
  * @txq: Tx descriptor ring for a specific queue
- * @bufq: buffer q or completion q
  *
  * Free all transmit software resources
  */
-static void idpf_tx_desc_rel(struct idpf_queue *txq, bool bufq)
+static void idpf_tx_desc_rel(struct idpf_tx_queue *txq)
 {
-	if (bufq)
-		idpf_tx_buf_rel_all(txq);
+	idpf_tx_buf_rel_all(txq);
 
 	if (!txq->desc_ring)
 		return;
 
 	dmam_free_coherent(txq->dev, txq->size, txq->desc_ring, txq->dma);
 	txq->desc_ring = NULL;
-	txq->next_to_alloc = 0;
 	txq->next_to_use = 0;
 	txq->next_to_clean = 0;
 }
 
+/**
+ * idpf_compl_desc_rel - Free completion resources per queue
+ * @complq: completion queue
+ *
+ * Free all completion software resources.
+ */
+static void idpf_compl_desc_rel(struct idpf_compl_queue *complq)
+{
+	if (!complq->comp)
+		return;
+
+	dma_free_coherent(complq->netdev->dev.parent, complq->size,
+			  complq->comp, complq->dma);
+	complq->comp = NULL;
+	complq->next_to_use = 0;
+	complq->next_to_clean = 0;
+}
+
 /**
  * idpf_tx_desc_rel_all - Free Tx Resources for All Queues
  * @vport: virtual port structure
@@ -150,10 +171,10 @@ static void idpf_tx_desc_rel_all(struct idpf_vport *vport)
 		struct idpf_txq_group *txq_grp = &vport->txq_grps[i];
 
 		for (j = 0; j < txq_grp->num_txq; j++)
-			idpf_tx_desc_rel(txq_grp->txqs[j], true);
+			idpf_tx_desc_rel(txq_grp->txqs[j]);
 
 		if (idpf_is_queue_model_split(vport->txq_model))
-			idpf_tx_desc_rel(txq_grp->complq, false);
+			idpf_compl_desc_rel(txq_grp->complq);
 	}
 }
 
@@ -163,8 +184,9 @@ static void idpf_tx_desc_rel_all(struct idpf_vport *vport)
  *
  * Returns 0 on success, negative on failure
  */
-static int idpf_tx_buf_alloc_all(struct idpf_queue *tx_q)
+static int idpf_tx_buf_alloc_all(struct idpf_tx_queue *tx_q)
 {
+	struct idpf_buf_lifo *buf_stack;
 	int buf_size;
 	int i;
 
@@ -180,22 +202,26 @@ static int idpf_tx_buf_alloc_all(struct idpf_queue *tx_q)
 	for (i = 0; i < tx_q->desc_count; i++)
 		tx_q->tx_buf[i].compl_tag = IDPF_SPLITQ_TX_INVAL_COMPL_TAG;
 
+	if (!idpf_queue_has(FLOW_SCH_EN, tx_q))
+		return 0;
+
+	buf_stack = &tx_q->stash->buf_stack;
+
 	/* Initialize tx buf stack for out-of-order completions if
 	 * flow scheduling offload is enabled
 	 */
-	tx_q->buf_stack.bufs =
-		kcalloc(tx_q->desc_count, sizeof(struct idpf_tx_stash *),
-			GFP_KERNEL);
-	if (!tx_q->buf_stack.bufs)
+	buf_stack->bufs = kcalloc(tx_q->desc_count, sizeof(*buf_stack->bufs),
+				  GFP_KERNEL);
+	if (!buf_stack->bufs)
 		return -ENOMEM;
 
-	tx_q->buf_stack.size = tx_q->desc_count;
-	tx_q->buf_stack.top = tx_q->desc_count;
+	buf_stack->size = tx_q->desc_count;
+	buf_stack->top = tx_q->desc_count;
 
 	for (i = 0; i < tx_q->desc_count; i++) {
-		tx_q->buf_stack.bufs[i] = kzalloc(sizeof(*tx_q->buf_stack.bufs[i]),
-						  GFP_KERNEL);
-		if (!tx_q->buf_stack.bufs[i])
+		buf_stack->bufs[i] = kzalloc(sizeof(*buf_stack->bufs[i]),
+					     GFP_KERNEL);
+		if (!buf_stack->bufs[i])
 			return -ENOMEM;
 	}
 
@@ -204,28 +230,22 @@ static int idpf_tx_buf_alloc_all(struct idpf_queue *tx_q)
 
 /**
  * idpf_tx_desc_alloc - Allocate the Tx descriptors
+ * @vport: vport to allocate resources for
  * @tx_q: the tx ring to set up
- * @bufq: buffer or completion queue
  *
  * Returns 0 on success, negative on failure
  */
-static int idpf_tx_desc_alloc(struct idpf_queue *tx_q, bool bufq)
+static int idpf_tx_desc_alloc(const struct idpf_vport *vport,
+			      struct idpf_tx_queue *tx_q)
 {
 	struct device *dev = tx_q->dev;
-	u32 desc_sz;
 	int err;
 
-	if (bufq) {
-		err = idpf_tx_buf_alloc_all(tx_q);
-		if (err)
-			goto err_alloc;
-
-		desc_sz = sizeof(struct idpf_base_tx_desc);
-	} else {
-		desc_sz = sizeof(struct idpf_splitq_tx_compl_desc);
-	}
+	err = idpf_tx_buf_alloc_all(tx_q);
+	if (err)
+		goto err_alloc;
 
-	tx_q->size = tx_q->desc_count * desc_sz;
+	tx_q->size = tx_q->desc_count * sizeof(*tx_q->base_tx);
 
 	/* Allocate descriptors also round up to nearest 4K */
 	tx_q->size = ALIGN(tx_q->size, 4096);
@@ -238,19 +258,43 @@ static int idpf_tx_desc_alloc(struct idpf_queue *tx_q, bool bufq)
 		goto err_alloc;
 	}
 
-	tx_q->next_to_alloc = 0;
 	tx_q->next_to_use = 0;
 	tx_q->next_to_clean = 0;
-	set_bit(__IDPF_Q_GEN_CHK, tx_q->flags);
+	idpf_queue_set(GEN_CHK, tx_q);
 
 	return 0;
 
 err_alloc:
-	idpf_tx_desc_rel(tx_q, bufq);
+	idpf_tx_desc_rel(tx_q);
 
 	return err;
 }
 
+/**
+ * idpf_compl_desc_alloc - allocate completion descriptors
+ * @vport: vport to allocate resources for
+ * @complq: completion queue to set up
+ *
+ * Return: 0 on success, -errno on failure.
+ */
+static int idpf_compl_desc_alloc(const struct idpf_vport *vport,
+				 struct idpf_compl_queue *complq)
+{
+	complq->size = array_size(complq->desc_count, sizeof(*complq->comp));
+
+	complq->comp = dma_alloc_coherent(complq->netdev->dev.parent,
+					  complq->size, &complq->dma,
+					  GFP_KERNEL);
+	if (!complq->comp)
+		return -ENOMEM;
+
+	complq->next_to_use = 0;
+	complq->next_to_clean = 0;
+	idpf_queue_set(GEN_CHK, complq);
+
+	return 0;
+}
+
 /**
  * idpf_tx_desc_alloc_all - allocate all queues Tx resources
  * @vport: virtual port private structure
@@ -259,7 +303,6 @@ err_alloc:
  */
 static int idpf_tx_desc_alloc_all(struct idpf_vport *vport)
 {
-	struct device *dev = &vport->adapter->pdev->dev;
 	int err = 0;
 	int i, j;
 
@@ -268,13 +311,14 @@ static int idpf_tx_desc_alloc_all(struct idpf_vport *vport)
 	 */
 	for (i = 0; i < vport->num_txq_grp; i++) {
 		for (j = 0; j < vport->txq_grps[i].num_txq; j++) {
-			struct idpf_queue *txq = vport->txq_grps[i].txqs[j];
+			struct idpf_tx_queue *txq = vport->txq_grps[i].txqs[j];
 			u8 gen_bits = 0;
 			u16 bufidx_mask;
 
-			err = idpf_tx_desc_alloc(txq, true);
+			err = idpf_tx_desc_alloc(vport, txq);
 			if (err) {
-				dev_err(dev, "Allocation for Tx Queue %u failed\n",
+				pci_err(vport->adapter->pdev,
+					"Allocation for Tx Queue %u failed\n",
 					i);
 				goto err_out;
 			}
@@ -312,9 +356,10 @@ static int idpf_tx_desc_alloc_all(struct idpf_vport *vport)
 			continue;
 
 		/* Setup completion queues */
-		err = idpf_tx_desc_alloc(vport->txq_grps[i].complq, false);
+		err = idpf_compl_desc_alloc(vport, vport->txq_grps[i].complq);
 		if (err) {
-			dev_err(dev, "Allocation for Tx Completion Queue %u failed\n",
+			pci_err(vport->adapter->pdev,
+				"Allocation for Tx Completion Queue %u failed\n",
 				i);
 			goto err_out;
 		}
@@ -329,15 +374,14 @@ err_out:
 
 /**
  * idpf_rx_page_rel - Release an rx buffer page
- * @rxq: the queue that owns the buffer
  * @rx_buf: the buffer to free
  */
-static void idpf_rx_page_rel(struct idpf_queue *rxq, struct idpf_rx_buf *rx_buf)
+static void idpf_rx_page_rel(struct idpf_rx_buf *rx_buf)
 {
 	if (unlikely(!rx_buf->page))
 		return;
 
-	page_pool_put_full_page(rxq->pp, rx_buf->page, false);
+	page_pool_put_full_page(rx_buf->page->pp, rx_buf->page, false);
 
 	rx_buf->page = NULL;
 	rx_buf->page_offset = 0;
@@ -345,54 +389,72 @@ static void idpf_rx_page_rel(struct idpf_queue *rxq, struct idpf_rx_buf *rx_buf)
 
 /**
  * idpf_rx_hdr_buf_rel_all - Release header buffer memory
- * @rxq: queue to use
+ * @bufq: queue to use
+ * @dev: device to free DMA memory
  */
-static void idpf_rx_hdr_buf_rel_all(struct idpf_queue *rxq)
+static void idpf_rx_hdr_buf_rel_all(struct idpf_buf_queue *bufq,
+				    struct device *dev)
 {
-	struct idpf_adapter *adapter = rxq->vport->adapter;
-
-	dma_free_coherent(&adapter->pdev->dev,
-			  rxq->desc_count * IDPF_HDR_BUF_SIZE,
-			  rxq->rx_buf.hdr_buf_va,
-			  rxq->rx_buf.hdr_buf_pa);
-	rxq->rx_buf.hdr_buf_va = NULL;
+	dma_free_coherent(dev, bufq->desc_count * IDPF_HDR_BUF_SIZE,
+			  bufq->rx_buf.hdr_buf_va, bufq->rx_buf.hdr_buf_pa);
+	bufq->rx_buf.hdr_buf_va = NULL;
 }
 
 /**
- * idpf_rx_buf_rel_all - Free all Rx buffer resources for a queue
- * @rxq: queue to be cleaned
+ * idpf_rx_buf_rel_bufq - Free all Rx buffer resources for a buffer queue
+ * @bufq: queue to be cleaned
+ * @dev: device to free DMA memory
  */
-static void idpf_rx_buf_rel_all(struct idpf_queue *rxq)
+static void idpf_rx_buf_rel_bufq(struct idpf_buf_queue *bufq,
+				 struct device *dev)
 {
-	u16 i;
-
 	/* queue already cleared, nothing to do */
-	if (!rxq->rx_buf.buf)
+	if (!bufq->rx_buf.buf)
 		return;
 
 	/* Free all the bufs allocated and given to hw on Rx queue */
-	for (i = 0; i < rxq->desc_count; i++)
-		idpf_rx_page_rel(rxq, &rxq->rx_buf.buf[i]);
+	for (u32 i = 0; i < bufq->desc_count; i++)
+		idpf_rx_page_rel(&bufq->rx_buf.buf[i]);
+
+	if (idpf_queue_has(HSPLIT_EN, bufq))
+		idpf_rx_hdr_buf_rel_all(bufq, dev);
+
+	page_pool_destroy(bufq->pp);
+	bufq->pp = NULL;
+
+	kfree(bufq->rx_buf.buf);
+	bufq->rx_buf.buf = NULL;
+}
 
-	if (rxq->rx_hsplit_en)
-		idpf_rx_hdr_buf_rel_all(rxq);
+/**
+ * idpf_rx_buf_rel_all - Free all Rx buffer resources for a receive queue
+ * @rxq: queue to be cleaned
+ */
+static void idpf_rx_buf_rel_all(struct idpf_rx_queue *rxq)
+{
+	if (!rxq->rx_buf)
+		return;
+
+	for (u32 i = 0; i < rxq->desc_count; i++)
+		idpf_rx_page_rel(&rxq->rx_buf[i]);
 
 	page_pool_destroy(rxq->pp);
 	rxq->pp = NULL;
 
-	kfree(rxq->rx_buf.buf);
-	rxq->rx_buf.buf = NULL;
+	kfree(rxq->rx_buf);
+	rxq->rx_buf = NULL;
 }
 
 /**
  * idpf_rx_desc_rel - Free a specific Rx q resources
  * @rxq: queue to clean the resources from
- * @bufq: buffer q or completion q
- * @q_model: single or split q model
+ * @dev: device to free DMA memory
+ * @model: single or split queue model
  *
  * Free a specific rx queue resources
  */
-static void idpf_rx_desc_rel(struct idpf_queue *rxq, bool bufq, s32 q_model)
+static void idpf_rx_desc_rel(struct idpf_rx_queue *rxq, struct device *dev,
+			     u32 model)
 {
 	if (!rxq)
 		return;
@@ -402,7 +464,7 @@ static void idpf_rx_desc_rel(struct idpf_queue *rxq, bool bufq, s32 q_model)
 		rxq->skb = NULL;
 	}
 
-	if (bufq || !idpf_is_queue_model_split(q_model))
+	if (!idpf_is_queue_model_split(model))
 		idpf_rx_buf_rel_all(rxq);
 
 	rxq->next_to_alloc = 0;
@@ -411,10 +473,34 @@ static void idpf_rx_desc_rel(struct idpf_queue *rxq, bool bufq, s32 q_model)
 	if (!rxq->desc_ring)
 		return;
 
-	dmam_free_coherent(rxq->dev, rxq->size, rxq->desc_ring, rxq->dma);
+	dmam_free_coherent(dev, rxq->size, rxq->desc_ring, rxq->dma);
 	rxq->desc_ring = NULL;
 }
 
+/**
+ * idpf_rx_desc_rel_bufq - free buffer queue resources
+ * @bufq: buffer queue to clean the resources from
+ * @dev: device to free DMA memory
+ */
+static void idpf_rx_desc_rel_bufq(struct idpf_buf_queue *bufq,
+				  struct device *dev)
+{
+	if (!bufq)
+		return;
+
+	idpf_rx_buf_rel_bufq(bufq, dev);
+
+	bufq->next_to_alloc = 0;
+	bufq->next_to_clean = 0;
+	bufq->next_to_use = 0;
+
+	if (!bufq->split_buf)
+		return;
+
+	dma_free_coherent(dev, bufq->size, bufq->split_buf, bufq->dma);
+	bufq->split_buf = NULL;
+}
+
 /**
  * idpf_rx_desc_rel_all - Free Rx Resources for All Queues
  * @vport: virtual port structure
@@ -423,6 +509,7 @@ static void idpf_rx_desc_rel(struct idpf_queue *rxq, bool bufq, s32 q_model)
  */
 static void idpf_rx_desc_rel_all(struct idpf_vport *vport)
 {
+	struct device *dev = &vport->adapter->pdev->dev;
 	struct idpf_rxq_group *rx_qgrp;
 	u16 num_rxq;
 	int i, j;
@@ -435,15 +522,15 @@ static void idpf_rx_desc_rel_all(struct idpf_vport *vport)
 
 		if (!idpf_is_queue_model_split(vport->rxq_model)) {
 			for (j = 0; j < rx_qgrp->singleq.num_rxq; j++)
-				idpf_rx_desc_rel(rx_qgrp->singleq.rxqs[j],
-						 false, vport->rxq_model);
+				idpf_rx_desc_rel(rx_qgrp->singleq.rxqs[j], dev,
+						 VIRTCHNL2_QUEUE_MODEL_SINGLE);
 			continue;
 		}
 
 		num_rxq = rx_qgrp->splitq.num_rxq_sets;
 		for (j = 0; j < num_rxq; j++)
 			idpf_rx_desc_rel(&rx_qgrp->splitq.rxq_sets[j]->rxq,
-					 false, vport->rxq_model);
+					 dev, VIRTCHNL2_QUEUE_MODEL_SPLIT);
 
 		if (!rx_qgrp->splitq.bufq_sets)
 			continue;
@@ -452,44 +539,40 @@ static void idpf_rx_desc_rel_all(struct idpf_vport *vport)
 			struct idpf_bufq_set *bufq_set =
 				&rx_qgrp->splitq.bufq_sets[j];
 
-			idpf_rx_desc_rel(&bufq_set->bufq, true,
-					 vport->rxq_model);
+			idpf_rx_desc_rel_bufq(&bufq_set->bufq, dev);
 		}
 	}
 }
 
 /**
  * idpf_rx_buf_hw_update - Store the new tail and head values
- * @rxq: queue to bump
+ * @bufq: queue to bump
  * @val: new head index
  */
-void idpf_rx_buf_hw_update(struct idpf_queue *rxq, u32 val)
+static void idpf_rx_buf_hw_update(struct idpf_buf_queue *bufq, u32 val)
 {
-	rxq->next_to_use = val;
+	bufq->next_to_use = val;
 
-	if (unlikely(!rxq->tail))
+	if (unlikely(!bufq->tail))
 		return;
 
 	/* writel has an implicit memory barrier */
-	writel(val, rxq->tail);
+	writel(val, bufq->tail);
 }
 
 /**
  * idpf_rx_hdr_buf_alloc_all - Allocate memory for header buffers
- * @rxq: ring to use
+ * @bufq: ring to use
  *
  * Returns 0 on success, negative on failure.
  */
-static int idpf_rx_hdr_buf_alloc_all(struct idpf_queue *rxq)
+static int idpf_rx_hdr_buf_alloc_all(struct idpf_buf_queue *bufq)
 {
-	struct idpf_adapter *adapter = rxq->vport->adapter;
-
-	rxq->rx_buf.hdr_buf_va =
-		dma_alloc_coherent(&adapter->pdev->dev,
-				   IDPF_HDR_BUF_SIZE * rxq->desc_count,
-				   &rxq->rx_buf.hdr_buf_pa,
-				   GFP_KERNEL);
-	if (!rxq->rx_buf.hdr_buf_va)
+	bufq->rx_buf.hdr_buf_va =
+		dma_alloc_coherent(bufq->q_vector->vport->netdev->dev.parent,
+				   IDPF_HDR_BUF_SIZE * bufq->desc_count,
+				   &bufq->rx_buf.hdr_buf_pa, GFP_KERNEL);
+	if (!bufq->rx_buf.hdr_buf_va)
 		return -ENOMEM;
 
 	return 0;
@@ -502,19 +585,20 @@ static int idpf_rx_hdr_buf_alloc_all(struct idpf_queue *rxq)
  */
 static void idpf_rx_post_buf_refill(struct idpf_sw_queue *refillq, u16 buf_id)
 {
-	u16 nta = refillq->next_to_alloc;
+	u32 nta = refillq->next_to_use;
 
 	/* store the buffer ID and the SW maintained GEN bit to the refillq */
 	refillq->ring[nta] =
 		FIELD_PREP(IDPF_RX_BI_BUFID_M, buf_id) |
 		FIELD_PREP(IDPF_RX_BI_GEN_M,
-			   test_bit(__IDPF_Q_GEN_CHK, refillq->flags));
+			   idpf_queue_has(GEN_CHK, refillq));
 
 	if (unlikely(++nta == refillq->desc_count)) {
 		nta = 0;
-		change_bit(__IDPF_Q_GEN_CHK, refillq->flags);
+		idpf_queue_change(GEN_CHK, refillq);
 	}
-	refillq->next_to_alloc = nta;
+
+	refillq->next_to_use = nta;
 }
 
 /**
@@ -524,7 +608,7 @@ static void idpf_rx_post_buf_refill(struct idpf_sw_queue *refillq, u16 buf_id)
  *
  * Returns false if buffer could not be allocated, true otherwise.
  */
-static bool idpf_rx_post_buf_desc(struct idpf_queue *bufq, u16 buf_id)
+static bool idpf_rx_post_buf_desc(struct idpf_buf_queue *bufq, u16 buf_id)
 {
 	struct virtchnl2_splitq_rx_buf_desc *splitq_rx_desc = NULL;
 	u16 nta = bufq->next_to_alloc;
@@ -534,11 +618,10 @@ static bool idpf_rx_post_buf_desc(struct idpf_queue *bufq, u16 buf_id)
 	splitq_rx_desc = &bufq->split_buf[nta];
 	buf = &bufq->rx_buf.buf[buf_id];
 
-	if (bufq->rx_hsplit_en) {
+	if (idpf_queue_has(HSPLIT_EN, bufq))
 		splitq_rx_desc->hdr_addr =
 			cpu_to_le64(bufq->rx_buf.hdr_buf_pa +
 				    (u32)buf_id * IDPF_HDR_BUF_SIZE);
-	}
 
 	addr = idpf_alloc_page(bufq->pp, buf, bufq->rx_buf_size);
 	if (unlikely(addr == DMA_MAPPING_ERROR))
@@ -562,7 +645,8 @@ static bool idpf_rx_post_buf_desc(struct idpf_queue *bufq, u16 buf_id)
  *
  * Returns true if @working_set bufs were posted successfully, false otherwise.
  */
-static bool idpf_rx_post_init_bufs(struct idpf_queue *bufq, u16 working_set)
+static bool idpf_rx_post_init_bufs(struct idpf_buf_queue *bufq,
+				   u16 working_set)
 {
 	int i;
 
@@ -571,26 +655,28 @@ static bool idpf_rx_post_init_bufs(struct idpf_queue *bufq, u16 working_set)
 			return false;
 	}
 
-	idpf_rx_buf_hw_update(bufq,
-			      bufq->next_to_alloc & ~(bufq->rx_buf_stride - 1));
+	idpf_rx_buf_hw_update(bufq, ALIGN_DOWN(bufq->next_to_alloc,
+					       IDPF_RX_BUF_STRIDE));
 
 	return true;
 }
 
 /**
  * idpf_rx_create_page_pool - Create a page pool
- * @rxbufq: RX queue to create page pool for
+ * @napi: NAPI of the associated queue vector
+ * @count: queue descriptor count
  *
  * Returns &page_pool on success, casted -errno on failure
  */
-static struct page_pool *idpf_rx_create_page_pool(struct idpf_queue *rxbufq)
+static struct page_pool *idpf_rx_create_page_pool(struct napi_struct *napi,
+						  u32 count)
 {
 	struct page_pool_params pp = {
 		.flags		= PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV,
 		.order		= 0,
-		.pool_size	= rxbufq->desc_count,
+		.pool_size	= count,
 		.nid		= NUMA_NO_NODE,
-		.dev		= rxbufq->vport->netdev->dev.parent,
+		.dev		= napi->dev->dev.parent,
 		.max_len	= PAGE_SIZE,
 		.dma_dir	= DMA_FROM_DEVICE,
 		.offset		= 0,
@@ -599,15 +685,58 @@ static struct page_pool *idpf_rx_create_page_pool(struct idpf_queue *rxbufq)
 	return page_pool_create(&pp);
 }
 
+/**
+ * idpf_rx_buf_alloc_singleq - Allocate memory for all buffer resources
+ * @rxq: queue for which the buffers are allocated
+ *
+ * Return: 0 on success, -ENOMEM on failure.
+ */
+static int idpf_rx_buf_alloc_singleq(struct idpf_rx_queue *rxq)
+{
+	rxq->rx_buf = kcalloc(rxq->desc_count, sizeof(*rxq->rx_buf),
+			      GFP_KERNEL);
+	if (!rxq->rx_buf)
+		return -ENOMEM;
+
+	if (idpf_rx_singleq_buf_hw_alloc_all(rxq, rxq->desc_count - 1))
+		goto err;
+
+	return 0;
+
+err:
+	idpf_rx_buf_rel_all(rxq);
+
+	return -ENOMEM;
+}
+
+/**
+ * idpf_rx_bufs_init_singleq - Initialize page pool and allocate Rx bufs
+ * @rxq: buffer queue to create page pool for
+ *
+ * Return: 0 on success, -errno on failure.
+ */
+static int idpf_rx_bufs_init_singleq(struct idpf_rx_queue *rxq)
+{
+	struct page_pool *pool;
+
+	pool = idpf_rx_create_page_pool(&rxq->q_vector->napi, rxq->desc_count);
+	if (IS_ERR(pool))
+		return PTR_ERR(pool);
+
+	rxq->pp = pool;
+
+	return idpf_rx_buf_alloc_singleq(rxq);
+}
+
 /**
  * idpf_rx_buf_alloc_all - Allocate memory for all buffer resources
- * @rxbufq: queue for which the buffers are allocated; equivalent to
- * rxq when operating in singleq mode
+ * @rxbufq: queue for which the buffers are allocated
  *
  * Returns 0 on success, negative on failure
  */
-static int idpf_rx_buf_alloc_all(struct idpf_queue *rxbufq)
+static int idpf_rx_buf_alloc_all(struct idpf_buf_queue *rxbufq)
 {
+	struct device *dev = rxbufq->q_vector->vport->netdev->dev.parent;
 	int err = 0;
 
 	/* Allocate book keeping buffers */
@@ -618,48 +747,41 @@ static int idpf_rx_buf_alloc_all(struct idpf_queue *rxbufq)
 		goto rx_buf_alloc_all_out;
 	}
 
-	if (rxbufq->rx_hsplit_en) {
+	if (idpf_queue_has(HSPLIT_EN, rxbufq)) {
 		err = idpf_rx_hdr_buf_alloc_all(rxbufq);
 		if (err)
 			goto rx_buf_alloc_all_out;
 	}
 
 	/* Allocate buffers to be given to HW.	 */
-	if (idpf_is_queue_model_split(rxbufq->vport->rxq_model)) {
-		int working_set = IDPF_RX_BUFQ_WORKING_SET(rxbufq);
-
-		if (!idpf_rx_post_init_bufs(rxbufq, working_set))
-			err = -ENOMEM;
-	} else {
-		if (idpf_rx_singleq_buf_hw_alloc_all(rxbufq,
-						     rxbufq->desc_count - 1))
-			err = -ENOMEM;
-	}
+	if (!idpf_rx_post_init_bufs(rxbufq, IDPF_RX_BUFQ_WORKING_SET(rxbufq)))
+		err = -ENOMEM;
 
 rx_buf_alloc_all_out:
 	if (err)
-		idpf_rx_buf_rel_all(rxbufq);
+		idpf_rx_buf_rel_bufq(rxbufq, dev);
 
 	return err;
 }
 
 /**
  * idpf_rx_bufs_init - Initialize page pool, allocate rx bufs, and post to HW
- * @rxbufq: RX queue to create page pool for
+ * @bufq: buffer queue to create page pool for
  *
  * Returns 0 on success, negative on failure
  */
-static int idpf_rx_bufs_init(struct idpf_queue *rxbufq)
+static int idpf_rx_bufs_init(struct idpf_buf_queue *bufq)
 {
 	struct page_pool *pool;
 
-	pool = idpf_rx_create_page_pool(rxbufq);
+	pool = idpf_rx_create_page_pool(&bufq->q_vector->napi,
+					bufq->desc_count);
 	if (IS_ERR(pool))
 		return PTR_ERR(pool);
 
-	rxbufq->pp = pool;
+	bufq->pp = pool;
 
-	return idpf_rx_buf_alloc_all(rxbufq);
+	return idpf_rx_buf_alloc_all(bufq);
 }
 
 /**
@@ -671,7 +793,6 @@ static int idpf_rx_bufs_init(struct idpf_queue *rxbufq)
 int idpf_rx_bufs_init_all(struct idpf_vport *vport)
 {
 	struct idpf_rxq_group *rx_qgrp;
-	struct idpf_queue *q;
 	int i, j, err;
 
 	for (i = 0; i < vport->num_rxq_grp; i++) {
@@ -682,8 +803,10 @@ int idpf_rx_bufs_init_all(struct idpf_vport *vport)
 			int num_rxq = rx_qgrp->singleq.num_rxq;
 
 			for (j = 0; j < num_rxq; j++) {
+				struct idpf_rx_queue *q;
+
 				q = rx_qgrp->singleq.rxqs[j];
-				err = idpf_rx_bufs_init(q);
+				err = idpf_rx_bufs_init_singleq(q);
 				if (err)
 					return err;
 			}
@@ -693,6 +816,8 @@ int idpf_rx_bufs_init_all(struct idpf_vport *vport)
 
 		/* Otherwise, allocate bufs for the buffer queues */
 		for (j = 0; j < vport->num_bufqs_per_qgrp; j++) {
+			struct idpf_buf_queue *q;
+
 			q = &rx_qgrp->splitq.bufq_sets[j].bufq;
 			err = idpf_rx_bufs_init(q);
 			if (err)
@@ -705,22 +830,17 @@ int idpf_rx_bufs_init_all(struct idpf_vport *vport)
 
 /**
  * idpf_rx_desc_alloc - Allocate queue Rx resources
+ * @vport: vport to allocate resources for
  * @rxq: Rx queue for which the resources are setup
- * @bufq: buffer or completion queue
- * @q_model: single or split queue model
  *
  * Returns 0 on success, negative on failure
  */
-static int idpf_rx_desc_alloc(struct idpf_queue *rxq, bool bufq, s32 q_model)
+static int idpf_rx_desc_alloc(const struct idpf_vport *vport,
+			      struct idpf_rx_queue *rxq)
 {
-	struct device *dev = rxq->dev;
+	struct device *dev = &vport->adapter->pdev->dev;
 
-	if (bufq)
-		rxq->size = rxq->desc_count *
-			sizeof(struct virtchnl2_splitq_rx_buf_desc);
-	else
-		rxq->size = rxq->desc_count *
-			sizeof(union virtchnl2_rx_desc);
+	rxq->size = rxq->desc_count * sizeof(union virtchnl2_rx_desc);
 
 	/* Allocate descriptors and also round up to nearest 4K */
 	rxq->size = ALIGN(rxq->size, 4096);
@@ -735,7 +855,35 @@ static int idpf_rx_desc_alloc(struct idpf_queue *rxq, bool bufq, s32 q_model)
 	rxq->next_to_alloc = 0;
 	rxq->next_to_clean = 0;
 	rxq->next_to_use = 0;
-	set_bit(__IDPF_Q_GEN_CHK, rxq->flags);
+	idpf_queue_set(GEN_CHK, rxq);
+
+	return 0;
+}
+
+/**
+ * idpf_bufq_desc_alloc - Allocate buffer queue descriptor ring
+ * @vport: vport to allocate resources for
+ * @bufq: buffer queue for which the resources are set up
+ *
+ * Return: 0 on success, -ENOMEM on failure.
+ */
+static int idpf_bufq_desc_alloc(const struct idpf_vport *vport,
+				struct idpf_buf_queue *bufq)
+{
+	struct device *dev = &vport->adapter->pdev->dev;
+
+	bufq->size = array_size(bufq->desc_count, sizeof(*bufq->split_buf));
+
+	bufq->split_buf = dma_alloc_coherent(dev, bufq->size, &bufq->dma,
+					     GFP_KERNEL);
+	if (!bufq->split_buf)
+		return -ENOMEM;
+
+	bufq->next_to_alloc = 0;
+	bufq->next_to_clean = 0;
+	bufq->next_to_use = 0;
+
+	idpf_queue_set(GEN_CHK, bufq);
 
 	return 0;
 }
@@ -748,9 +896,7 @@ static int idpf_rx_desc_alloc(struct idpf_queue *rxq, bool bufq, s32 q_model)
  */
 static int idpf_rx_desc_alloc_all(struct idpf_vport *vport)
 {
-	struct device *dev = &vport->adapter->pdev->dev;
 	struct idpf_rxq_group *rx_qgrp;
-	struct idpf_queue *q;
 	int i, j, err;
 	u16 num_rxq;
 
@@ -762,13 +908,17 @@ static int idpf_rx_desc_alloc_all(struct idpf_vport *vport)
 			num_rxq = rx_qgrp->singleq.num_rxq;
 
 		for (j = 0; j < num_rxq; j++) {
+			struct idpf_rx_queue *q;
+
 			if (idpf_is_queue_model_split(vport->rxq_model))
 				q = &rx_qgrp->splitq.rxq_sets[j]->rxq;
 			else
 				q = rx_qgrp->singleq.rxqs[j];
-			err = idpf_rx_desc_alloc(q, false, vport->rxq_model);
+
+			err = idpf_rx_desc_alloc(vport, q);
 			if (err) {
-				dev_err(dev, "Memory allocation for Rx Queue %u failed\n",
+				pci_err(vport->adapter->pdev,
+					"Memory allocation for Rx Queue %u failed\n",
 					i);
 				goto err_out;
 			}
@@ -778,10 +928,14 @@ static int idpf_rx_desc_alloc_all(struct idpf_vport *vport)
 			continue;
 
 		for (j = 0; j < vport->num_bufqs_per_qgrp; j++) {
+			struct idpf_buf_queue *q;
+
 			q = &rx_qgrp->splitq.bufq_sets[j].bufq;
-			err = idpf_rx_desc_alloc(q, true, vport->rxq_model);
+
+			err = idpf_bufq_desc_alloc(vport, q);
 			if (err) {
-				dev_err(dev, "Memory allocation for Rx Buffer Queue %u failed\n",
+				pci_err(vport->adapter->pdev,
+					"Memory allocation for Rx Buffer Queue %u failed\n",
 					i);
 				goto err_out;
 			}
@@ -802,11 +956,16 @@ err_out:
  */
 static void idpf_txq_group_rel(struct idpf_vport *vport)
 {
+	bool split, flow_sch_en;
 	int i, j;
 
 	if (!vport->txq_grps)
 		return;
 
+	split = idpf_is_queue_model_split(vport->txq_model);
+	flow_sch_en = !idpf_is_cap_ena(vport->adapter, IDPF_OTHER_CAPS,
+				       VIRTCHNL2_CAP_SPLITQ_QSCHED);
+
 	for (i = 0; i < vport->num_txq_grp; i++) {
 		struct idpf_txq_group *txq_grp = &vport->txq_grps[i];
 
@@ -814,8 +973,15 @@ static void idpf_txq_group_rel(struct idpf_vport *vport)
 			kfree(txq_grp->txqs[j]);
 			txq_grp->txqs[j] = NULL;
 		}
+
+		if (!split)
+			continue;
+
 		kfree(txq_grp->complq);
 		txq_grp->complq = NULL;
+
+		if (flow_sch_en)
+			kfree(txq_grp->stashes);
 	}
 	kfree(vport->txq_grps);
 	vport->txq_grps = NULL;
@@ -919,7 +1085,7 @@ static int idpf_vport_init_fast_path_txqs(struct idpf_vport *vport)
 {
 	int i, j, k = 0;
 
-	vport->txqs = kcalloc(vport->num_txq, sizeof(struct idpf_queue *),
+	vport->txqs = kcalloc(vport->num_txq, sizeof(*vport->txqs),
 			      GFP_KERNEL);
 
 	if (!vport->txqs)
@@ -1137,7 +1303,8 @@ static void idpf_vport_calc_numq_per_grp(struct idpf_vport *vport,
  * @q: rx queue for which descids are set
  *
  */
-static void idpf_rxq_set_descids(struct idpf_vport *vport, struct idpf_queue *q)
+static void idpf_rxq_set_descids(const struct idpf_vport *vport,
+				 struct idpf_rx_queue *q)
 {
 	if (vport->rxq_model == VIRTCHNL2_QUEUE_MODEL_SPLIT) {
 		q->rxdids = VIRTCHNL2_RXDID_2_FLEX_SPLITQ_M;
@@ -1158,20 +1325,22 @@ static void idpf_rxq_set_descids(struct idpf_vport *vport, struct idpf_queue *q)
  */
 static int idpf_txq_group_alloc(struct idpf_vport *vport, u16 num_txq)
 {
-	bool flow_sch_en;
-	int err, i;
+	bool split, flow_sch_en;
+	int i;
 
 	vport->txq_grps = kcalloc(vport->num_txq_grp,
 				  sizeof(*vport->txq_grps), GFP_KERNEL);
 	if (!vport->txq_grps)
 		return -ENOMEM;
 
+	split = idpf_is_queue_model_split(vport->txq_model);
 	flow_sch_en = !idpf_is_cap_ena(vport->adapter, IDPF_OTHER_CAPS,
 				       VIRTCHNL2_CAP_SPLITQ_QSCHED);
 
 	for (i = 0; i < vport->num_txq_grp; i++) {
 		struct idpf_txq_group *tx_qgrp = &vport->txq_grps[i];
 		struct idpf_adapter *adapter = vport->adapter;
+		struct idpf_txq_stash *stashes;
 		int j;
 
 		tx_qgrp->vport = vport;
@@ -1180,45 +1349,62 @@ static int idpf_txq_group_alloc(struct idpf_vport *vport, u16 num_txq)
 		for (j = 0; j < tx_qgrp->num_txq; j++) {
 			tx_qgrp->txqs[j] = kzalloc(sizeof(*tx_qgrp->txqs[j]),
 						   GFP_KERNEL);
-			if (!tx_qgrp->txqs[j]) {
-				err = -ENOMEM;
+			if (!tx_qgrp->txqs[j])
 				goto err_alloc;
-			}
+		}
+
+		if (split && flow_sch_en) {
+			stashes = kcalloc(num_txq, sizeof(*stashes),
+					  GFP_KERNEL);
+			if (!stashes)
+				goto err_alloc;
+
+			tx_qgrp->stashes = stashes;
 		}
 
 		for (j = 0; j < tx_qgrp->num_txq; j++) {
-			struct idpf_queue *q = tx_qgrp->txqs[j];
+			struct idpf_tx_queue *q = tx_qgrp->txqs[j];
 
 			q->dev = &adapter->pdev->dev;
 			q->desc_count = vport->txq_desc_count;
 			q->tx_max_bufs = idpf_get_max_tx_bufs(adapter);
 			q->tx_min_pkt_len = idpf_get_min_tx_pkt_len(adapter);
-			q->vport = vport;
+			q->netdev = vport->netdev;
 			q->txq_grp = tx_qgrp;
-			hash_init(q->sched_buf_hash);
 
-			if (flow_sch_en)
-				set_bit(__IDPF_Q_FLOW_SCH_EN, q->flags);
+			if (!split) {
+				q->clean_budget = vport->compln_clean_budget;
+				idpf_queue_assign(CRC_EN, q,
+						  vport->crc_enable);
+			}
+
+			if (!flow_sch_en)
+				continue;
+
+			if (split) {
+				q->stash = &stashes[j];
+				hash_init(q->stash->sched_buf_hash);
+			}
+
+			idpf_queue_set(FLOW_SCH_EN, q);
 		}
 
-		if (!idpf_is_queue_model_split(vport->txq_model))
+		if (!split)
 			continue;
 
 		tx_qgrp->complq = kcalloc(IDPF_COMPLQ_PER_GROUP,
 					  sizeof(*tx_qgrp->complq),
 					  GFP_KERNEL);
-		if (!tx_qgrp->complq) {
-			err = -ENOMEM;
+		if (!tx_qgrp->complq)
 			goto err_alloc;
-		}
 
-		tx_qgrp->complq->dev = &adapter->pdev->dev;
 		tx_qgrp->complq->desc_count = vport->complq_desc_count;
-		tx_qgrp->complq->vport = vport;
 		tx_qgrp->complq->txq_grp = tx_qgrp;
+		tx_qgrp->complq->netdev = vport->netdev;
+		tx_qgrp->complq->clean_budget = vport->compln_clean_budget;
 
 		if (flow_sch_en)
-			__set_bit(__IDPF_Q_FLOW_SCH_EN, tx_qgrp->complq->flags);
+			idpf_queue_set(FLOW_SCH_EN, tx_qgrp->complq);
 	}
 
 	return 0;
@@ -1226,7 +1412,7 @@ static int idpf_txq_group_alloc(struct idpf_vport *vport, u16 num_txq)
 err_alloc:
 	idpf_txq_group_rel(vport);
 
-	return err;
+	return -ENOMEM;
 }
 
 /**
@@ -1238,8 +1424,6 @@ err_alloc:
  */
 static int idpf_rxq_group_alloc(struct idpf_vport *vport, u16 num_rxq)
 {
-	struct idpf_adapter *adapter = vport->adapter;
-	struct idpf_queue *q;
 	int i, k, err = 0;
 	bool hs;
 
@@ -1292,21 +1476,15 @@ static int idpf_rxq_group_alloc(struct idpf_vport *vport, u16 num_rxq)
 			struct idpf_bufq_set *bufq_set =
 				&rx_qgrp->splitq.bufq_sets[j];
 			int swq_size = sizeof(struct idpf_sw_queue);
+			struct idpf_buf_queue *q;
 
 			q = &rx_qgrp->splitq.bufq_sets[j].bufq;
-			q->dev = &adapter->pdev->dev;
 			q->desc_count = vport->bufq_desc_count[j];
-			q->vport = vport;
-			q->rxq_grp = rx_qgrp;
-			q->idx = j;
 			q->rx_buf_size = vport->bufq_size[j];
 			q->rx_buffer_low_watermark = IDPF_LOW_WATERMARK;
-			q->rx_buf_stride = IDPF_RX_BUF_STRIDE;
 
-			if (hs) {
-				q->rx_hsplit_en = true;
-				q->rx_hbuf_size = IDPF_HDR_BUF_SIZE;
-			}
+			idpf_queue_assign(HSPLIT_EN, q, hs);
+			q->rx_hbuf_size = hs ? IDPF_HDR_BUF_SIZE : 0;
 
 			bufq_set->num_refillqs = num_rxq;
 			bufq_set->refillqs = kcalloc(num_rxq, swq_size,
@@ -1319,13 +1497,12 @@ static int idpf_rxq_group_alloc(struct idpf_vport *vport, u16 num_rxq)
 				struct idpf_sw_queue *refillq =
 					&bufq_set->refillqs[k];
 
-				refillq->dev = &vport->adapter->pdev->dev;
 				refillq->desc_count =
 					vport->bufq_desc_count[j];
-				set_bit(__IDPF_Q_GEN_CHK, refillq->flags);
-				set_bit(__IDPF_RFLQ_GEN_CHK, refillq->flags);
+				idpf_queue_set(GEN_CHK, refillq);
+				idpf_queue_set(RFL_GEN_CHK, refillq);
 				refillq->ring = kcalloc(refillq->desc_count,
-							sizeof(u16),
+							sizeof(*refillq->ring),
 							GFP_KERNEL);
 				if (!refillq->ring) {
 					err = -ENOMEM;
@@ -1336,27 +1513,27 @@ static int idpf_rxq_group_alloc(struct idpf_vport *vport, u16 num_rxq)
 
 skip_splitq_rx_init:
 		for (j = 0; j < num_rxq; j++) {
+			struct idpf_rx_queue *q;
+
 			if (!idpf_is_queue_model_split(vport->rxq_model)) {
 				q = rx_qgrp->singleq.rxqs[j];
 				goto setup_rxq;
 			}
 			q = &rx_qgrp->splitq.rxq_sets[j]->rxq;
-			rx_qgrp->splitq.rxq_sets[j]->refillq0 =
+			rx_qgrp->splitq.rxq_sets[j]->refillq[0] =
 			      &rx_qgrp->splitq.bufq_sets[0].refillqs[j];
 			if (vport->num_bufqs_per_qgrp > IDPF_SINGLE_BUFQ_PER_RXQ_GRP)
-				rx_qgrp->splitq.rxq_sets[j]->refillq1 =
+				rx_qgrp->splitq.rxq_sets[j]->refillq[1] =
 				      &rx_qgrp->splitq.bufq_sets[1].refillqs[j];
 
-			if (hs) {
-				q->rx_hsplit_en = true;
-				q->rx_hbuf_size = IDPF_HDR_BUF_SIZE;
-			}
+			idpf_queue_assign(HSPLIT_EN, q, hs);
+			q->rx_hbuf_size = hs ? IDPF_HDR_BUF_SIZE : 0;
 
 setup_rxq:
-			q->dev = &adapter->pdev->dev;
 			q->desc_count = vport->rxq_desc_count;
-			q->vport = vport;
-			q->rxq_grp = rx_qgrp;
+			q->rx_ptype_lkup = vport->rx_ptype_lkup;
+			q->netdev = vport->netdev;
+			q->bufq_sets = rx_qgrp->splitq.bufq_sets;
 			q->idx = (i * num_rxq) + j;
 			/* In splitq mode, RXQ buffer size should be
 			 * set to that of the first buffer queue
@@ -1445,12 +1622,13 @@ err_out:
  * idpf_tx_handle_sw_marker - Handle queue marker packet
  * @tx_q: tx queue to handle software marker
  */
-static void idpf_tx_handle_sw_marker(struct idpf_queue *tx_q)
+static void idpf_tx_handle_sw_marker(struct idpf_tx_queue *tx_q)
 {
-	struct idpf_vport *vport = tx_q->vport;
+	struct idpf_netdev_priv *priv = netdev_priv(tx_q->netdev);
+	struct idpf_vport *vport = priv->vport;
 	int i;
 
-	clear_bit(__IDPF_Q_SW_MARKER, tx_q->flags);
+	idpf_queue_clear(SW_MARKER, tx_q);
 	/* Hardware must write marker packets to all queues associated with
 	 * completion queues. So check if all queues received marker packets
 	 */
@@ -1458,7 +1636,7 @@ static void idpf_tx_handle_sw_marker(struct idpf_queue *tx_q)
 		/* If we're still waiting on any other TXQ marker completions,
 		 * just return now since we cannot wake up the marker_wq yet.
 		 */
-		if (test_bit(__IDPF_Q_SW_MARKER, vport->txqs[i]->flags))
+		if (idpf_queue_has(SW_MARKER, vport->txqs[i]))
 			return;
 
 	/* Drain complete */
@@ -1474,7 +1652,7 @@ static void idpf_tx_handle_sw_marker(struct idpf_queue *tx_q)
  * @cleaned: pointer to stats struct to track cleaned packets/bytes
  * @napi_budget: Used to determine if we are in netpoll
  */
-static void idpf_tx_splitq_clean_hdr(struct idpf_queue *tx_q,
+static void idpf_tx_splitq_clean_hdr(struct idpf_tx_queue *tx_q,
 				     struct idpf_tx_buf *tx_buf,
 				     struct idpf_cleaned_stats *cleaned,
 				     int napi_budget)
@@ -1505,7 +1683,8 @@ static void idpf_tx_splitq_clean_hdr(struct idpf_queue *tx_q,
  * @cleaned: pointer to stats struct to track cleaned packets/bytes
  * @budget: Used to determine if we are in netpoll
  */
-static void idpf_tx_clean_stashed_bufs(struct idpf_queue *txq, u16 compl_tag,
+static void idpf_tx_clean_stashed_bufs(struct idpf_tx_queue *txq,
+				       u16 compl_tag,
 				       struct idpf_cleaned_stats *cleaned,
 				       int budget)
 {
@@ -1513,7 +1692,7 @@ static void idpf_tx_clean_stashed_bufs(struct idpf_queue *txq, u16 compl_tag,
 	struct hlist_node *tmp_buf;
 
 	/* Buffer completion */
-	hash_for_each_possible_safe(txq->sched_buf_hash, stash, tmp_buf,
+	hash_for_each_possible_safe(txq->stash->sched_buf_hash, stash, tmp_buf,
 				    hlist, compl_tag) {
 		if (unlikely(stash->buf.compl_tag != (int)compl_tag))
 			continue;
@@ -1530,7 +1709,7 @@ static void idpf_tx_clean_stashed_bufs(struct idpf_queue *txq, u16 compl_tag,
 		}
 
 		/* Push shadow buf back onto stack */
-		idpf_buf_lifo_push(&txq->buf_stack, stash);
+		idpf_buf_lifo_push(&txq->stash->buf_stack, stash);
 
 		hash_del(&stash->hlist);
 	}
@@ -1542,7 +1721,7 @@ static void idpf_tx_clean_stashed_bufs(struct idpf_queue *txq, u16 compl_tag,
  * @txq: Tx queue to clean
  * @tx_buf: buffer to store
  */
-static int idpf_stash_flow_sch_buffers(struct idpf_queue *txq,
+static int idpf_stash_flow_sch_buffers(struct idpf_tx_queue *txq,
 				       struct idpf_tx_buf *tx_buf)
 {
 	struct idpf_tx_stash *stash;
@@ -1551,10 +1730,10 @@ static int idpf_stash_flow_sch_buffers(struct idpf_queue *txq,
 		     !dma_unmap_len(tx_buf, len)))
 		return 0;
 
-	stash = idpf_buf_lifo_pop(&txq->buf_stack);
+	stash = idpf_buf_lifo_pop(&txq->stash->buf_stack);
 	if (unlikely(!stash)) {
 		net_err_ratelimited("%s: No out-of-order TX buffers left!\n",
-				    txq->vport->netdev->name);
+				    netdev_name(txq->netdev));
 
 		return -ENOMEM;
 	}
@@ -1568,7 +1747,8 @@ static int idpf_stash_flow_sch_buffers(struct idpf_queue *txq,
 	stash->buf.compl_tag = tx_buf->compl_tag;
 
 	/* Add buffer to buf_hash table to be freed later */
-	hash_add(txq->sched_buf_hash, &stash->hlist, stash->buf.compl_tag);
+	hash_add(txq->stash->sched_buf_hash, &stash->hlist,
+		 stash->buf.compl_tag);
 
 	memset(tx_buf, 0, sizeof(struct idpf_tx_buf));
 
@@ -1607,7 +1787,7 @@ do {								\
  * and the buffers will be cleaned separately. The stats are not updated from
  * this function when using flow-based scheduling.
  */
-static void idpf_tx_splitq_clean(struct idpf_queue *tx_q, u16 end,
+static void idpf_tx_splitq_clean(struct idpf_tx_queue *tx_q, u16 end,
 				 int napi_budget,
 				 struct idpf_cleaned_stats *cleaned,
 				 bool descs_only)
@@ -1703,7 +1883,7 @@ do {							\
  * stashed. Returns the byte/segment count for the cleaned packet associated
  * this completion tag.
  */
-static bool idpf_tx_clean_buf_ring(struct idpf_queue *txq, u16 compl_tag,
+static bool idpf_tx_clean_buf_ring(struct idpf_tx_queue *txq, u16 compl_tag,
 				   struct idpf_cleaned_stats *cleaned,
 				   int budget)
 {
@@ -1772,14 +1952,14 @@ static bool idpf_tx_clean_buf_ring(struct idpf_queue *txq, u16 compl_tag,
  *
  * Returns bytes/packets cleaned
  */
-static void idpf_tx_handle_rs_completion(struct idpf_queue *txq,
+static void idpf_tx_handle_rs_completion(struct idpf_tx_queue *txq,
 					 struct idpf_splitq_tx_compl_desc *desc,
 					 struct idpf_cleaned_stats *cleaned,
 					 int budget)
 {
 	u16 compl_tag;
 
-	if (!test_bit(__IDPF_Q_FLOW_SCH_EN, txq->flags)) {
+	if (!idpf_queue_has(FLOW_SCH_EN, txq)) {
 		u16 head = le16_to_cpu(desc->q_head_compl_tag.q_head);
 
 		return idpf_tx_splitq_clean(txq, head, budget, cleaned, false);
@@ -1802,24 +1982,23 @@ static void idpf_tx_handle_rs_completion(struct idpf_queue *txq,
  *
  * Returns true if there's any budget left (e.g. the clean is finished)
  */
-static bool idpf_tx_clean_complq(struct idpf_queue *complq, int budget,
+static bool idpf_tx_clean_complq(struct idpf_compl_queue *complq, int budget,
 				 int *cleaned)
 {
 	struct idpf_splitq_tx_compl_desc *tx_desc;
-	struct idpf_vport *vport = complq->vport;
 	s16 ntc = complq->next_to_clean;
 	struct idpf_netdev_priv *np;
 	unsigned int complq_budget;
 	bool complq_ok = true;
 	int i;
 
-	complq_budget = vport->compln_clean_budget;
+	complq_budget = complq->clean_budget;
 	tx_desc = &complq->comp[ntc];
 	ntc -= complq->desc_count;
 
 	do {
 		struct idpf_cleaned_stats cleaned_stats = { };
-		struct idpf_queue *tx_q;
+		struct idpf_tx_queue *tx_q;
 		int rel_tx_qid;
 		u16 hw_head;
 		u8 ctype;	/* completion type */
@@ -1828,7 +2007,7 @@ static bool idpf_tx_clean_complq(struct idpf_queue *complq, int budget,
 		/* if the descriptor isn't done, no work yet to do */
 		gen = le16_get_bits(tx_desc->qid_comptype_gen,
 				    IDPF_TXD_COMPLQ_GEN_M);
-		if (test_bit(__IDPF_Q_GEN_CHK, complq->flags) != gen)
+		if (idpf_queue_has(GEN_CHK, complq) != gen)
 			break;
 
 		/* Find necessary info of TX queue to clean buffers */
@@ -1836,8 +2015,7 @@ static bool idpf_tx_clean_complq(struct idpf_queue *complq, int budget,
 					   IDPF_TXD_COMPLQ_QID_M);
 		if (rel_tx_qid >= complq->txq_grp->num_txq ||
 		    !complq->txq_grp->txqs[rel_tx_qid]) {
-			dev_err(&complq->vport->adapter->pdev->dev,
-				"TxQ not found\n");
+			netdev_err(complq->netdev, "TxQ not found\n");
 			goto fetch_next_desc;
 		}
 		tx_q = complq->txq_grp->txqs[rel_tx_qid];
@@ -1860,15 +2038,14 @@ static bool idpf_tx_clean_complq(struct idpf_queue *complq, int budget,
 			idpf_tx_handle_sw_marker(tx_q);
 			break;
 		default:
-			dev_err(&tx_q->vport->adapter->pdev->dev,
-				"Unknown TX completion type: %d\n",
-				ctype);
+			netdev_err(tx_q->netdev,
+				   "Unknown TX completion type: %d\n", ctype);
 			goto fetch_next_desc;
 		}
 
 		u64_stats_update_begin(&tx_q->stats_sync);
-		u64_stats_add(&tx_q->q_stats.tx.packets, cleaned_stats.packets);
-		u64_stats_add(&tx_q->q_stats.tx.bytes, cleaned_stats.bytes);
+		u64_stats_add(&tx_q->q_stats.packets, cleaned_stats.packets);
+		u64_stats_add(&tx_q->q_stats.bytes, cleaned_stats.bytes);
 		tx_q->cleaned_pkts += cleaned_stats.packets;
 		tx_q->cleaned_bytes += cleaned_stats.bytes;
 		complq->num_completions++;
@@ -1880,7 +2057,7 @@ fetch_next_desc:
 		if (unlikely(!ntc)) {
 			ntc -= complq->desc_count;
 			tx_desc = &complq->comp[0];
-			change_bit(__IDPF_Q_GEN_CHK, complq->flags);
+			idpf_queue_change(GEN_CHK, complq);
 		}
 
 		prefetch(tx_desc);
@@ -1896,9 +2073,9 @@ fetch_next_desc:
 		     IDPF_TX_COMPLQ_OVERFLOW_THRESH(complq)))
 		complq_ok = false;
 
-	np = netdev_priv(complq->vport->netdev);
+	np = netdev_priv(complq->netdev);
 	for (i = 0; i < complq->txq_grp->num_txq; ++i) {
-		struct idpf_queue *tx_q = complq->txq_grp->txqs[i];
+		struct idpf_tx_queue *tx_q = complq->txq_grp->txqs[i];
 		struct netdev_queue *nq;
 		bool dont_wake;
 
@@ -1909,11 +2086,11 @@ fetch_next_desc:
 		*cleaned += tx_q->cleaned_pkts;
 
 		/* Update BQL */
-		nq = netdev_get_tx_queue(tx_q->vport->netdev, tx_q->idx);
+		nq = netdev_get_tx_queue(tx_q->netdev, tx_q->idx);
 
 		dont_wake = !complq_ok || IDPF_TX_BUF_RSV_LOW(tx_q) ||
 			    np->state != __IDPF_VPORT_UP ||
-			    !netif_carrier_ok(tx_q->vport->netdev);
+			    !netif_carrier_ok(tx_q->netdev);
 		/* Check if the TXQ needs to and can be restarted */
 		__netif_txq_completed_wake(nq, tx_q->cleaned_pkts, tx_q->cleaned_bytes,
 					   IDPF_DESC_UNUSED(tx_q), IDPF_TX_WAKE_THRESH,
@@ -1976,7 +2153,7 @@ void idpf_tx_splitq_build_flow_desc(union idpf_tx_flex_desc *desc,
  *
  * Returns 0 if stop is not needed
  */
-int idpf_tx_maybe_stop_common(struct idpf_queue *tx_q, unsigned int size)
+int idpf_tx_maybe_stop_common(struct idpf_tx_queue *tx_q, unsigned int size)
 {
 	struct netdev_queue *nq;
 
@@ -1984,10 +2161,10 @@ int idpf_tx_maybe_stop_common(struct idpf_queue *tx_q, unsigned int size)
 		return 0;
 
 	u64_stats_update_begin(&tx_q->stats_sync);
-	u64_stats_inc(&tx_q->q_stats.tx.q_busy);
+	u64_stats_inc(&tx_q->q_stats.q_busy);
 	u64_stats_update_end(&tx_q->stats_sync);
 
-	nq = netdev_get_tx_queue(tx_q->vport->netdev, tx_q->idx);
+	nq = netdev_get_tx_queue(tx_q->netdev, tx_q->idx);
 
 	return netif_txq_maybe_stop(nq, IDPF_DESC_UNUSED(tx_q), size, size);
 }
@@ -1999,7 +2176,7 @@ int idpf_tx_maybe_stop_common(struct idpf_queue *tx_q, unsigned int size)
  *
  * Returns 0 if stop is not needed
  */
-static int idpf_tx_maybe_stop_splitq(struct idpf_queue *tx_q,
+static int idpf_tx_maybe_stop_splitq(struct idpf_tx_queue *tx_q,
 				     unsigned int descs_needed)
 {
 	if (idpf_tx_maybe_stop_common(tx_q, descs_needed))
@@ -2023,9 +2200,9 @@ static int idpf_tx_maybe_stop_splitq(struct idpf_queue *tx_q,
 
 splitq_stop:
 	u64_stats_update_begin(&tx_q->stats_sync);
-	u64_stats_inc(&tx_q->q_stats.tx.q_busy);
+	u64_stats_inc(&tx_q->q_stats.q_busy);
 	u64_stats_update_end(&tx_q->stats_sync);
-	netif_stop_subqueue(tx_q->vport->netdev, tx_q->idx);
+	netif_stop_subqueue(tx_q->netdev, tx_q->idx);
 
 	return -EBUSY;
 }
@@ -2040,12 +2217,12 @@ splitq_stop:
  * to do a register write to update our queue status. We know this can only
  * mean tail here as HW should be owning head for TX.
  */
-void idpf_tx_buf_hw_update(struct idpf_queue *tx_q, u32 val,
+void idpf_tx_buf_hw_update(struct idpf_tx_queue *tx_q, u32 val,
 			   bool xmit_more)
 {
 	struct netdev_queue *nq;
 
-	nq = netdev_get_tx_queue(tx_q->vport->netdev, tx_q->idx);
+	nq = netdev_get_tx_queue(tx_q->netdev, tx_q->idx);
 	tx_q->next_to_use = val;
 
 	idpf_tx_maybe_stop_common(tx_q, IDPF_TX_DESC_NEEDED);
@@ -2069,7 +2246,7 @@ void idpf_tx_buf_hw_update(struct idpf_queue *tx_q, u32 val,
  *
  * Returns number of data descriptors needed for this skb.
  */
-unsigned int idpf_tx_desc_count_required(struct idpf_queue *txq,
+unsigned int idpf_tx_desc_count_required(struct idpf_tx_queue *txq,
 					 struct sk_buff *skb)
 {
 	const struct skb_shared_info *shinfo;
@@ -2102,7 +2279,7 @@ unsigned int idpf_tx_desc_count_required(struct idpf_queue *txq,
 
 		count = idpf_size_to_txd_count(skb->len);
 		u64_stats_update_begin(&txq->stats_sync);
-		u64_stats_inc(&txq->q_stats.tx.linearize);
+		u64_stats_inc(&txq->q_stats.linearize);
 		u64_stats_update_end(&txq->stats_sync);
 	}
 
@@ -2116,11 +2293,11 @@ unsigned int idpf_tx_desc_count_required(struct idpf_queue *txq,
  * @first: original first buffer info buffer for packet
  * @idx: starting point on ring to unwind
  */
-void idpf_tx_dma_map_error(struct idpf_queue *txq, struct sk_buff *skb,
+void idpf_tx_dma_map_error(struct idpf_tx_queue *txq, struct sk_buff *skb,
 			   struct idpf_tx_buf *first, u16 idx)
 {
 	u64_stats_update_begin(&txq->stats_sync);
-	u64_stats_inc(&txq->q_stats.tx.dma_map_errs);
+	u64_stats_inc(&txq->q_stats.dma_map_errs);
 	u64_stats_update_end(&txq->stats_sync);
 
 	/* clear dma mappings for failed tx_buf map */
@@ -2159,7 +2336,7 @@ void idpf_tx_dma_map_error(struct idpf_queue *txq, struct sk_buff *skb,
  * @txq: the tx ring to wrap
  * @ntu: ring index to bump
  */
-static unsigned int idpf_tx_splitq_bump_ntu(struct idpf_queue *txq, u16 ntu)
+static unsigned int idpf_tx_splitq_bump_ntu(struct idpf_tx_queue *txq, u16 ntu)
 {
 	ntu++;
 
@@ -2181,7 +2358,7 @@ static unsigned int idpf_tx_splitq_bump_ntu(struct idpf_queue *txq, u16 ntu)
  * and gets a physical address for each memory location and programs
  * it and the length into the transmit flex descriptor.
  */
-static void idpf_tx_splitq_map(struct idpf_queue *tx_q,
+static void idpf_tx_splitq_map(struct idpf_tx_queue *tx_q,
 			       struct idpf_tx_splitq_params *params,
 			       struct idpf_tx_buf *first)
 {
@@ -2348,7 +2525,7 @@ static void idpf_tx_splitq_map(struct idpf_queue *tx_q,
 	tx_q->txq_grp->num_completions_pending++;
 
 	/* record bytecount for BQL */
-	nq = netdev_get_tx_queue(tx_q->vport->netdev, tx_q->idx);
+	nq = netdev_get_tx_queue(tx_q->netdev, tx_q->idx);
 	netdev_tx_sent_queue(nq, first->bytecount);
 
 	idpf_tx_buf_hw_update(tx_q, i, netdev_xmit_more());
@@ -2544,7 +2721,7 @@ bool idpf_chk_linearize(struct sk_buff *skb, unsigned int max_bufs,
  * ring entry to reflect that this index is a context descriptor
  */
 static struct idpf_flex_tx_ctx_desc *
-idpf_tx_splitq_get_ctx_desc(struct idpf_queue *txq)
+idpf_tx_splitq_get_ctx_desc(struct idpf_tx_queue *txq)
 {
 	struct idpf_flex_tx_ctx_desc *desc;
 	int i = txq->next_to_use;
@@ -2564,10 +2741,10 @@ idpf_tx_splitq_get_ctx_desc(struct idpf_queue *txq)
  * @tx_q: queue to send buffer on
  * @skb: pointer to skb
  */
-netdev_tx_t idpf_tx_drop_skb(struct idpf_queue *tx_q, struct sk_buff *skb)
+netdev_tx_t idpf_tx_drop_skb(struct idpf_tx_queue *tx_q, struct sk_buff *skb)
 {
 	u64_stats_update_begin(&tx_q->stats_sync);
-	u64_stats_inc(&tx_q->q_stats.tx.skb_drops);
+	u64_stats_inc(&tx_q->q_stats.skb_drops);
 	u64_stats_update_end(&tx_q->stats_sync);
 
 	idpf_tx_buf_hw_update(tx_q, tx_q->next_to_use, false);
@@ -2585,7 +2762,7 @@ netdev_tx_t idpf_tx_drop_skb(struct idpf_queue *tx_q, struct sk_buff *skb)
  * Returns NETDEV_TX_OK if sent, else an error code
  */
 static netdev_tx_t idpf_tx_splitq_frame(struct sk_buff *skb,
-					struct idpf_queue *tx_q)
+					struct idpf_tx_queue *tx_q)
 {
 	struct idpf_tx_splitq_params tx_params = { };
 	struct idpf_tx_buf *first;
@@ -2625,7 +2802,7 @@ static netdev_tx_t idpf_tx_splitq_frame(struct sk_buff *skb,
 		ctx_desc->tso.qw0.hdr_len = tx_params.offload.tso_hdr_len;
 
 		u64_stats_update_begin(&tx_q->stats_sync);
-		u64_stats_inc(&tx_q->q_stats.tx.lso_pkts);
+		u64_stats_inc(&tx_q->q_stats.lso_pkts);
 		u64_stats_update_end(&tx_q->stats_sync);
 	}
 
@@ -2642,7 +2819,7 @@ static netdev_tx_t idpf_tx_splitq_frame(struct sk_buff *skb,
 		first->bytecount = max_t(unsigned int, skb->len, ETH_ZLEN);
 	}
 
-	if (test_bit(__IDPF_Q_FLOW_SCH_EN, tx_q->flags)) {
+	if (idpf_queue_has(FLOW_SCH_EN, tx_q)) {
 		tx_params.dtype = IDPF_TX_DESC_DTYPE_FLEX_FLOW_SCHE;
 		tx_params.eop_cmd = IDPF_TXD_FLEX_FLOW_CMD_EOP;
 		/* Set the RE bit to catch any packets that may have not been
@@ -2682,7 +2859,7 @@ netdev_tx_t idpf_tx_splitq_start(struct sk_buff *skb,
 				 struct net_device *netdev)
 {
 	struct idpf_vport *vport = idpf_netdev_to_vport(netdev);
-	struct idpf_queue *tx_q;
+	struct idpf_tx_queue *tx_q;
 
 	if (unlikely(skb_get_queue_mapping(skb) >= vport->num_txq)) {
 		dev_kfree_skb_any(skb);
@@ -2735,13 +2912,14 @@ enum pkt_hash_types idpf_ptype_to_htype(const struct idpf_rx_ptype_decoded *deco
  * @rx_desc: Receive descriptor
  * @decoded: Decoded Rx packet type related fields
  */
-static void idpf_rx_hash(struct idpf_queue *rxq, struct sk_buff *skb,
-			 struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc,
-			 struct idpf_rx_ptype_decoded *decoded)
+static void
+idpf_rx_hash(const struct idpf_rx_queue *rxq, struct sk_buff *skb,
+	     const struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc,
+	     struct idpf_rx_ptype_decoded *decoded)
 {
 	u32 hash;
 
-	if (unlikely(!idpf_is_feature_ena(rxq->vport, NETIF_F_RXHASH)))
+	if (unlikely(!(rxq->netdev->features & NETIF_F_RXHASH)))
 		return;
 
 	hash = le16_to_cpu(rx_desc->hash1) |
@@ -2760,14 +2938,14 @@ static void idpf_rx_hash(struct idpf_queue *rxq, struct sk_buff *skb,
  *
  * skb->protocol must be set before this function is called
  */
-static void idpf_rx_csum(struct idpf_queue *rxq, struct sk_buff *skb,
+static void idpf_rx_csum(struct idpf_rx_queue *rxq, struct sk_buff *skb,
 			 struct idpf_rx_csum_decoded *csum_bits,
 			 struct idpf_rx_ptype_decoded *decoded)
 {
 	bool ipv4, ipv6;
 
 	/* check if Rx checksum is enabled */
-	if (unlikely(!idpf_is_feature_ena(rxq->vport, NETIF_F_RXCSUM)))
+	if (unlikely(!(rxq->netdev->features & NETIF_F_RXCSUM)))
 		return;
 
 	/* check if HW has decoded the packet and checksum */
@@ -2814,7 +2992,7 @@ static void idpf_rx_csum(struct idpf_queue *rxq, struct sk_buff *skb,
 
 checksum_fail:
 	u64_stats_update_begin(&rxq->stats_sync);
-	u64_stats_inc(&rxq->q_stats.rx.hw_csum_err);
+	u64_stats_inc(&rxq->q_stats.hw_csum_err);
 	u64_stats_update_end(&rxq->stats_sync);
 }
 
@@ -2824,8 +3002,9 @@ checksum_fail:
  * @csum: structure to extract checksum fields
  *
  **/
-static void idpf_rx_splitq_extract_csum_bits(struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc,
-					     struct idpf_rx_csum_decoded *csum)
+static void
+idpf_rx_splitq_extract_csum_bits(const struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc,
+				 struct idpf_rx_csum_decoded *csum)
 {
 	u8 qword0, qword1;
 
@@ -2860,8 +3039,8 @@ static void idpf_rx_splitq_extract_csum_bits(struct virtchnl2_rx_flex_desc_adv_n
  * Populate the skb fields with the total number of RSC segments, RSC payload
  * length and packet type.
  */
-static int idpf_rx_rsc(struct idpf_queue *rxq, struct sk_buff *skb,
-		       struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc,
+static int idpf_rx_rsc(struct idpf_rx_queue *rxq, struct sk_buff *skb,
+		       const struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc,
 		       struct idpf_rx_ptype_decoded *decoded)
 {
 	u16 rsc_segments, rsc_seg_len;
@@ -2914,7 +3093,7 @@ static int idpf_rx_rsc(struct idpf_queue *rxq, struct sk_buff *skb,
 	tcp_gro_complete(skb);
 
 	u64_stats_update_begin(&rxq->stats_sync);
-	u64_stats_inc(&rxq->q_stats.rx.rsc_pkts);
+	u64_stats_inc(&rxq->q_stats.rsc_pkts);
 	u64_stats_update_end(&rxq->stats_sync);
 
 	return 0;
@@ -2930,9 +3109,9 @@ static int idpf_rx_rsc(struct idpf_queue *rxq, struct sk_buff *skb,
  * order to populate the hash, checksum, protocol, and
  * other fields within the skb.
  */
-static int idpf_rx_process_skb_fields(struct idpf_queue *rxq,
-				      struct sk_buff *skb,
-				      struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc)
+static int
+idpf_rx_process_skb_fields(struct idpf_rx_queue *rxq, struct sk_buff *skb,
+			   const struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc)
 {
 	struct idpf_rx_csum_decoded csum_bits = { };
 	struct idpf_rx_ptype_decoded decoded;
@@ -2940,19 +3119,13 @@ static int idpf_rx_process_skb_fields(struct idpf_queue *rxq,
 
 	rx_ptype = le16_get_bits(rx_desc->ptype_err_fflags0,
 				 VIRTCHNL2_RX_FLEX_DESC_ADV_PTYPE_M);
-
-	skb->protocol = eth_type_trans(skb, rxq->vport->netdev);
-
-	decoded = rxq->vport->rx_ptype_lkup[rx_ptype];
-	/* If we don't know the ptype we can't do anything else with it. Just
-	 * pass it up the stack as-is.
-	 */
-	if (!decoded.known)
-		return 0;
+	decoded = rxq->rx_ptype_lkup[rx_ptype];
 
 	/* process RSS/hash */
 	idpf_rx_hash(rxq, skb, rx_desc, &decoded);
 
+	skb->protocol = eth_type_trans(skb, rxq->netdev);
+
 	if (le16_get_bits(rx_desc->hdrlen_flags,
 			  VIRTCHNL2_RX_FLEX_DESC_ADV_RSC_M))
 		return idpf_rx_rsc(rxq, skb, rx_desc, &decoded);
@@ -2992,7 +3165,7 @@ void idpf_rx_add_frag(struct idpf_rx_buf *rx_buf, struct sk_buff *skb,
  * data from the current receive descriptor, taking care to set up the
  * skb correctly.
  */
-struct sk_buff *idpf_rx_construct_skb(struct idpf_queue *rxq,
+struct sk_buff *idpf_rx_construct_skb(const struct idpf_rx_queue *rxq,
 				      struct idpf_rx_buf *rx_buf,
 				      unsigned int size)
 {
@@ -3005,7 +3178,7 @@ struct sk_buff *idpf_rx_construct_skb(struct idpf_queue *rxq,
 	/* prefetch first cache line of first page */
 	net_prefetch(va);
 	/* allocate a skb to store the frags */
-	skb = napi_alloc_skb(&rxq->q_vector->napi, IDPF_RX_HDR_SIZE);
+	skb = napi_alloc_skb(rxq->napi, IDPF_RX_HDR_SIZE);
 	if (unlikely(!skb)) {
 		idpf_rx_put_page(rx_buf);
 
@@ -3052,14 +3225,14 @@ struct sk_buff *idpf_rx_construct_skb(struct idpf_queue *rxq,
  * the current receive descriptor, taking care to set up the skb correctly.
  * This specifically uses a header buffer to start building the skb.
  */
-static struct sk_buff *idpf_rx_hdr_construct_skb(struct idpf_queue *rxq,
-						 const void *va,
-						 unsigned int size)
+static struct sk_buff *
+idpf_rx_hdr_construct_skb(const struct idpf_rx_queue *rxq, const void *va,
+			  unsigned int size)
 {
 	struct sk_buff *skb;
 
 	/* allocate a skb to store the frags */
-	skb = napi_alloc_skb(&rxq->q_vector->napi, size);
+	skb = napi_alloc_skb(rxq->napi, size);
 	if (unlikely(!skb))
 		return NULL;
 
@@ -3115,10 +3288,10 @@ static bool idpf_rx_splitq_is_eop(struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_de
  *
  * Returns amount of work completed
  */
-static int idpf_rx_splitq_clean(struct idpf_queue *rxq, int budget)
+static int idpf_rx_splitq_clean(struct idpf_rx_queue *rxq, int budget)
 {
 	int total_rx_bytes = 0, total_rx_pkts = 0;
-	struct idpf_queue *rx_bufq = NULL;
+	struct idpf_buf_queue *rx_bufq = NULL;
 	struct sk_buff *skb = rxq->skb;
 	u16 ntc = rxq->next_to_clean;
 
@@ -3148,7 +3321,7 @@ static int idpf_rx_splitq_clean(struct idpf_queue *rxq, int budget)
 		gen_id = le16_get_bits(rx_desc->pktlen_gen_bufq_id,
 				       VIRTCHNL2_RX_FLEX_DESC_ADV_GEN_M);
 
-		if (test_bit(__IDPF_Q_GEN_CHK, rxq->flags) != gen_id)
+		if (idpf_queue_has(GEN_CHK, rxq) != gen_id)
 			break;
 
 		rxdid = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_RXDID_M,
@@ -3156,7 +3329,7 @@ static int idpf_rx_splitq_clean(struct idpf_queue *rxq, int budget)
 		if (rxdid != VIRTCHNL2_RXDID_2_FLEX_SPLITQ) {
 			IDPF_RX_BUMP_NTC(rxq, ntc);
 			u64_stats_update_begin(&rxq->stats_sync);
-			u64_stats_inc(&rxq->q_stats.rx.bad_descs);
+			u64_stats_inc(&rxq->q_stats.bad_descs);
 			u64_stats_update_end(&rxq->stats_sync);
 			continue;
 		}
@@ -3174,7 +3347,7 @@ static int idpf_rx_splitq_clean(struct idpf_queue *rxq, int budget)
 			 * data/payload buffer.
 			 */
 			u64_stats_update_begin(&rxq->stats_sync);
-			u64_stats_inc(&rxq->q_stats.rx.hsplit_buf_ovf);
+			u64_stats_inc(&rxq->q_stats.hsplit_buf_ovf);
 			u64_stats_update_end(&rxq->stats_sync);
 			goto bypass_hsplit;
 		}
@@ -3187,13 +3360,10 @@ bypass_hsplit:
 					VIRTCHNL2_RX_FLEX_DESC_ADV_BUFQ_ID_M);
 
 		rxq_set = container_of(rxq, struct idpf_rxq_set, rxq);
-		if (!bufq_id)
-			refillq = rxq_set->refillq0;
-		else
-			refillq = rxq_set->refillq1;
+		refillq = rxq_set->refillq[bufq_id];
 
 		/* retrieve buffer from the rxq */
-		rx_bufq = &rxq->rxq_grp->splitq.bufq_sets[bufq_id].bufq;
+		rx_bufq = &rxq->bufq_sets[bufq_id].bufq;
 
 		buf_id = le16_to_cpu(rx_desc->buf_id);
 
@@ -3205,7 +3375,7 @@ bypass_hsplit:
 
 			skb = idpf_rx_hdr_construct_skb(rxq, va, hdr_len);
 			u64_stats_update_begin(&rxq->stats_sync);
-			u64_stats_inc(&rxq->q_stats.rx.hsplit_pkts);
+			u64_stats_inc(&rxq->q_stats.hsplit_pkts);
 			u64_stats_update_end(&rxq->stats_sync);
 		}
 
@@ -3248,7 +3418,7 @@ bypass_hsplit:
 		}
 
 		/* send completed skb up the stack */
-		napi_gro_receive(&rxq->q_vector->napi, skb);
+		napi_gro_receive(rxq->napi, skb);
 		skb = NULL;
 
 		/* update budget accounting */
@@ -3259,8 +3429,8 @@ bypass_hsplit:
 
 	rxq->skb = skb;
 	u64_stats_update_begin(&rxq->stats_sync);
-	u64_stats_add(&rxq->q_stats.rx.packets, total_rx_pkts);
-	u64_stats_add(&rxq->q_stats.rx.bytes, total_rx_bytes);
+	u64_stats_add(&rxq->q_stats.packets, total_rx_pkts);
+	u64_stats_add(&rxq->q_stats.bytes, total_rx_bytes);
 	u64_stats_update_end(&rxq->stats_sync);
 
 	/* guarantee a trip back through this routine if there was a failure */
@@ -3270,19 +3440,16 @@ bypass_hsplit:
 /**
  * idpf_rx_update_bufq_desc - Update buffer queue descriptor
  * @bufq: Pointer to the buffer queue
- * @refill_desc: SW Refill queue descriptor containing buffer ID
+ * @buf_id: buffer ID
  * @buf_desc: Buffer queue descriptor
  *
  * Return 0 on success and negative on failure.
  */
-static int idpf_rx_update_bufq_desc(struct idpf_queue *bufq, u16 refill_desc,
+static int idpf_rx_update_bufq_desc(struct idpf_buf_queue *bufq, u32 buf_id,
 				    struct virtchnl2_splitq_rx_buf_desc *buf_desc)
 {
 	struct idpf_rx_buf *buf;
 	dma_addr_t addr;
-	u16 buf_id;
-
-	buf_id = FIELD_GET(IDPF_RX_BI_BUFID_M, refill_desc);
 
 	buf = &bufq->rx_buf.buf[buf_id];
 
@@ -3293,7 +3460,7 @@ static int idpf_rx_update_bufq_desc(struct idpf_queue *bufq, u16 refill_desc,
 	buf_desc->pkt_addr = cpu_to_le64(addr);
 	buf_desc->qword0.buf_id = cpu_to_le16(buf_id);
 
-	if (!bufq->rx_hsplit_en)
+	if (!idpf_queue_has(HSPLIT_EN, bufq))
 		return 0;
 
 	buf_desc->hdr_addr = cpu_to_le64(bufq->rx_buf.hdr_buf_pa +
@@ -3309,33 +3476,32 @@ static int idpf_rx_update_bufq_desc(struct idpf_queue *bufq, u16 refill_desc,
  *
  * This function takes care of the buffer refill management
  */
-static void idpf_rx_clean_refillq(struct idpf_queue *bufq,
+static void idpf_rx_clean_refillq(struct idpf_buf_queue *bufq,
 				  struct idpf_sw_queue *refillq)
 {
 	struct virtchnl2_splitq_rx_buf_desc *buf_desc;
 	u16 bufq_nta = bufq->next_to_alloc;
 	u16 ntc = refillq->next_to_clean;
 	int cleaned = 0;
-	u16 gen;
 
 	buf_desc = &bufq->split_buf[bufq_nta];
 
 	/* make sure we stop at ring wrap in the unlikely case ring is full */
 	while (likely(cleaned < refillq->desc_count)) {
-		u16 refill_desc = refillq->ring[ntc];
+		u32 buf_id, refill_desc = refillq->ring[ntc];
 		bool failure;
 
-		gen = FIELD_GET(IDPF_RX_BI_GEN_M, refill_desc);
-		if (test_bit(__IDPF_RFLQ_GEN_CHK, refillq->flags) != gen)
+		if (idpf_queue_has(RFL_GEN_CHK, refillq) !=
+		    !!(refill_desc & IDPF_RX_BI_GEN_M))
 			break;
 
-		failure = idpf_rx_update_bufq_desc(bufq, refill_desc,
-						   buf_desc);
+		buf_id = FIELD_GET(IDPF_RX_BI_BUFID_M, refill_desc);
+		failure = idpf_rx_update_bufq_desc(bufq, buf_id, buf_desc);
 		if (failure)
 			break;
 
 		if (unlikely(++ntc == refillq->desc_count)) {
-			change_bit(__IDPF_RFLQ_GEN_CHK, refillq->flags);
+			idpf_queue_change(RFL_GEN_CHK, refillq);
 			ntc = 0;
 		}
 
@@ -3374,7 +3540,7 @@ static void idpf_rx_clean_refillq(struct idpf_queue *bufq,
  * this vector.  Returns true if clean is complete within budget, false
  * otherwise.
  */
-static void idpf_rx_clean_refillq_all(struct idpf_queue *bufq)
+static void idpf_rx_clean_refillq_all(struct idpf_buf_queue *bufq)
 {
 	struct idpf_bufq_set *bufq_set;
 	int i;
@@ -3439,6 +3605,8 @@ void idpf_vport_intr_rel(struct idpf_vport *vport)
 	for (v_idx = 0; v_idx < vport->num_q_vectors; v_idx++) {
 		struct idpf_q_vector *q_vector = &vport->q_vectors[v_idx];
 
+		kfree(q_vector->complq);
+		q_vector->complq = NULL;
 		kfree(q_vector->bufq);
 		q_vector->bufq = NULL;
 		kfree(q_vector->tx);
@@ -3577,13 +3745,13 @@ static void idpf_net_dim(struct idpf_q_vector *q_vector)
 		goto check_rx_itr;
 
 	for (i = 0, packets = 0, bytes = 0; i < q_vector->num_txq; i++) {
-		struct idpf_queue *txq = q_vector->tx[i];
+		struct idpf_tx_queue *txq = q_vector->tx[i];
 		unsigned int start;
 
 		do {
 			start = u64_stats_fetch_begin(&txq->stats_sync);
-			packets += u64_stats_read(&txq->q_stats.tx.packets);
-			bytes += u64_stats_read(&txq->q_stats.tx.bytes);
+			packets += u64_stats_read(&txq->q_stats.packets);
+			bytes += u64_stats_read(&txq->q_stats.bytes);
 		} while (u64_stats_fetch_retry(&txq->stats_sync, start));
 	}
 
@@ -3596,13 +3764,13 @@ check_rx_itr:
 		return;
 
 	for (i = 0, packets = 0, bytes = 0; i < q_vector->num_rxq; i++) {
-		struct idpf_queue *rxq = q_vector->rx[i];
+		struct idpf_rx_queue *rxq = q_vector->rx[i];
 		unsigned int start;
 
 		do {
 			start = u64_stats_fetch_begin(&rxq->stats_sync);
-			packets += u64_stats_read(&rxq->q_stats.rx.packets);
-			bytes += u64_stats_read(&rxq->q_stats.rx.bytes);
+			packets += u64_stats_read(&rxq->q_stats.packets);
+			bytes += u64_stats_read(&rxq->q_stats.bytes);
 		} while (u64_stats_fetch_retry(&rxq->stats_sync, start));
 	}
 
@@ -3844,16 +4012,17 @@ static void idpf_vport_intr_napi_ena_all(struct idpf_vport *vport)
 static bool idpf_tx_splitq_clean_all(struct idpf_q_vector *q_vec,
 				     int budget, int *cleaned)
 {
-	u16 num_txq = q_vec->num_txq;
+	u16 num_complq = q_vec->num_complq;
 	bool clean_complete = true;
 	int i, budget_per_q;
 
-	if (unlikely(!num_txq))
+	if (unlikely(!num_complq))
 		return true;
 
-	budget_per_q = DIV_ROUND_UP(budget, num_txq);
-	for (i = 0; i < num_txq; i++)
-		clean_complete &= idpf_tx_clean_complq(q_vec->tx[i],
+	budget_per_q = DIV_ROUND_UP(budget, num_complq);
+
+	for (i = 0; i < num_complq; i++)
+		clean_complete &= idpf_tx_clean_complq(q_vec->complq[i],
 						       budget_per_q, cleaned);
 
 	return clean_complete;
@@ -3880,7 +4049,7 @@ static bool idpf_rx_splitq_clean_all(struct idpf_q_vector *q_vec, int budget,
 	 */
 	budget_per_q = num_rxq ? max(budget / num_rxq, 1) : 0;
 	for (i = 0; i < num_rxq; i++) {
-		struct idpf_queue *rxq = q_vec->rx[i];
+		struct idpf_rx_queue *rxq = q_vec->rx[i];
 		int pkts_cleaned_per_q;
 
 		pkts_cleaned_per_q = idpf_rx_splitq_clean(rxq, budget_per_q);
@@ -3935,8 +4104,8 @@ static int idpf_vport_splitq_napi_poll(struct napi_struct *napi, int budget)
 	 * queues virtchnl message, as the interrupts will be disabled after
 	 * that
 	 */
-	if (unlikely(q_vector->num_txq && test_bit(__IDPF_Q_POLL_MODE,
-						   q_vector->tx[0]->flags)))
+	if (unlikely(q_vector->num_txq && idpf_queue_has(POLL_MODE,
+							 q_vector->tx[0])))
 		return budget;
 	else
 		return work_done;
@@ -3950,27 +4119,28 @@ static int idpf_vport_splitq_napi_poll(struct napi_struct *napi, int budget)
  */
 static void idpf_vport_intr_map_vector_to_qs(struct idpf_vport *vport)
 {
+	bool split = idpf_is_queue_model_split(vport->rxq_model);
 	u16 num_txq_grp = vport->num_txq_grp;
-	int i, j, qv_idx, bufq_vidx = 0;
 	struct idpf_rxq_group *rx_qgrp;
 	struct idpf_txq_group *tx_qgrp;
-	struct idpf_queue *q, *bufq;
-	u16 q_index;
+	u32 i, qv_idx, q_index;
 
 	for (i = 0, qv_idx = 0; i < vport->num_rxq_grp; i++) {
 		u16 num_rxq;
 
+		if (qv_idx >= vport->num_q_vectors)
+			qv_idx = 0;
+
 		rx_qgrp = &vport->rxq_grps[i];
-		if (idpf_is_queue_model_split(vport->rxq_model))
+		if (split)
 			num_rxq = rx_qgrp->splitq.num_rxq_sets;
 		else
 			num_rxq = rx_qgrp->singleq.num_rxq;
 
-		for (j = 0; j < num_rxq; j++) {
-			if (qv_idx >= vport->num_q_vectors)
-				qv_idx = 0;
+		for (u32 j = 0; j < num_rxq; j++) {
+			struct idpf_rx_queue *q;
 
-			if (idpf_is_queue_model_split(vport->rxq_model))
+			if (split)
 				q = &rx_qgrp->splitq.rxq_sets[j]->rxq;
 			else
 				q = rx_qgrp->singleq.rxqs[j];
@@ -3978,52 +4148,53 @@ static void idpf_vport_intr_map_vector_to_qs(struct idpf_vport *vport)
 			q_index = q->q_vector->num_rxq;
 			q->q_vector->rx[q_index] = q;
 			q->q_vector->num_rxq++;
-			qv_idx++;
+
+			if (split)
+				q->napi = &q->q_vector->napi;
 		}
 
-		if (idpf_is_queue_model_split(vport->rxq_model)) {
-			for (j = 0; j < vport->num_bufqs_per_qgrp; j++) {
+		if (split) {
+			for (u32 j = 0; j < vport->num_bufqs_per_qgrp; j++) {
+				struct idpf_buf_queue *bufq;
+
 				bufq = &rx_qgrp->splitq.bufq_sets[j].bufq;
-				bufq->q_vector = &vport->q_vectors[bufq_vidx];
+				bufq->q_vector = &vport->q_vectors[qv_idx];
 				q_index = bufq->q_vector->num_bufq;
 				bufq->q_vector->bufq[q_index] = bufq;
 				bufq->q_vector->num_bufq++;
 			}
-			if (++bufq_vidx >= vport->num_q_vectors)
-				bufq_vidx = 0;
 		}
+
+		qv_idx++;
 	}
 
+	split = idpf_is_queue_model_split(vport->txq_model);
+
 	for (i = 0, qv_idx = 0; i < num_txq_grp; i++) {
 		u16 num_txq;
 
+		if (qv_idx >= vport->num_q_vectors)
+			qv_idx = 0;
+
 		tx_qgrp = &vport->txq_grps[i];
 		num_txq = tx_qgrp->num_txq;
 
-		if (idpf_is_queue_model_split(vport->txq_model)) {
-			if (qv_idx >= vport->num_q_vectors)
-				qv_idx = 0;
+		for (u32 j = 0; j < num_txq; j++) {
+			struct idpf_tx_queue *q;
 
-			q = tx_qgrp->complq;
+			q = tx_qgrp->txqs[j];
 			q->q_vector = &vport->q_vectors[qv_idx];
-			q_index = q->q_vector->num_txq;
-			q->q_vector->tx[q_index] = q;
-			q->q_vector->num_txq++;
-			qv_idx++;
-		} else {
-			for (j = 0; j < num_txq; j++) {
-				if (qv_idx >= vport->num_q_vectors)
-					qv_idx = 0;
+			q->q_vector->tx[q->q_vector->num_txq++] = q;
+		}
 
-				q = tx_qgrp->txqs[j];
-				q->q_vector = &vport->q_vectors[qv_idx];
-				q_index = q->q_vector->num_txq;
-				q->q_vector->tx[q_index] = q;
-				q->q_vector->num_txq++;
+		if (split) {
+			struct idpf_compl_queue *q = tx_qgrp->complq;
 
-				qv_idx++;
-			}
+			q->q_vector = &vport->q_vectors[qv_idx];
+			q->q_vector->complq[q->q_vector->num_complq++] = q;
 		}
+
+		qv_idx++;
 	}
 }
 
@@ -4099,18 +4270,22 @@ int idpf_vport_intr_alloc(struct idpf_vport *vport)
 {
 	u16 txqs_per_vector, rxqs_per_vector, bufqs_per_vector;
 	struct idpf_q_vector *q_vector;
-	int v_idx, err;
+	u32 complqs_per_vector, v_idx;
 
 	vport->q_vectors = kcalloc(vport->num_q_vectors,
 				   sizeof(struct idpf_q_vector), GFP_KERNEL);
 	if (!vport->q_vectors)
 		return -ENOMEM;
 
-	txqs_per_vector = DIV_ROUND_UP(vport->num_txq, vport->num_q_vectors);
-	rxqs_per_vector = DIV_ROUND_UP(vport->num_rxq, vport->num_q_vectors);
+	txqs_per_vector = DIV_ROUND_UP(vport->num_txq_grp,
+				       vport->num_q_vectors);
+	rxqs_per_vector = DIV_ROUND_UP(vport->num_rxq_grp,
+				       vport->num_q_vectors);
 	bufqs_per_vector = vport->num_bufqs_per_qgrp *
 			   DIV_ROUND_UP(vport->num_rxq_grp,
 					vport->num_q_vectors);
+	complqs_per_vector = DIV_ROUND_UP(vport->num_txq_grp,
+					  vport->num_q_vectors);
 
 	for (v_idx = 0; v_idx < vport->num_q_vectors; v_idx++) {
 		q_vector = &vport->q_vectors[v_idx];
@@ -4124,32 +4299,30 @@ int idpf_vport_intr_alloc(struct idpf_vport *vport)
 		q_vector->rx_intr_mode = IDPF_ITR_DYNAMIC;
 		q_vector->rx_itr_idx = VIRTCHNL2_ITR_IDX_0;
 
-		q_vector->tx = kcalloc(txqs_per_vector,
-				       sizeof(struct idpf_queue *),
+		q_vector->tx = kcalloc(txqs_per_vector, sizeof(*q_vector->tx),
 				       GFP_KERNEL);
-		if (!q_vector->tx) {
-			err = -ENOMEM;
+		if (!q_vector->tx)
 			goto error;
-		}
 
-		q_vector->rx = kcalloc(rxqs_per_vector,
-				       sizeof(struct idpf_queue *),
+		q_vector->rx = kcalloc(rxqs_per_vector, sizeof(*q_vector->rx),
 				       GFP_KERNEL);
-		if (!q_vector->rx) {
-			err = -ENOMEM;
+		if (!q_vector->rx)
 			goto error;
-		}
 
 		if (!idpf_is_queue_model_split(vport->rxq_model))
 			continue;
 
 		q_vector->bufq = kcalloc(bufqs_per_vector,
-					 sizeof(struct idpf_queue *),
+					 sizeof(*q_vector->bufq),
 					 GFP_KERNEL);
-		if (!q_vector->bufq) {
-			err = -ENOMEM;
+		if (!q_vector->bufq)
+			goto error;
+
+		q_vector->complq = kcalloc(complqs_per_vector,
+					   sizeof(*q_vector->complq),
+					   GFP_KERNEL);
+		if (!q_vector->complq)
 			goto error;
-		}
 	}
 
 	return 0;
@@ -4157,7 +4330,7 @@ int idpf_vport_intr_alloc(struct idpf_vport *vport)
 error:
 	idpf_vport_intr_rel(vport);
 
-	return err;
+	return -ENOMEM;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
index 6dce14483215..704aec5c383b 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
@@ -4,6 +4,8 @@
 #ifndef _IDPF_TXRX_H_
 #define _IDPF_TXRX_H_
 
+#include <linux/dim.h>
+
 #include <net/page_pool/helpers.h>
 #include <net/tcp.h>
 #include <net/netdev_queues.h>
@@ -84,7 +86,7 @@
 do {								\
 	if (unlikely(++(ntc) == (rxq)->desc_count)) {		\
 		ntc = 0;					\
-		change_bit(__IDPF_Q_GEN_CHK, (rxq)->flags);	\
+		idpf_queue_change(GEN_CHK, rxq);		\
 	}							\
 } while (0)
 
@@ -111,10 +113,9 @@ do {								\
  */
 #define IDPF_TX_SPLITQ_RE_MIN_GAP	64
 
-#define IDPF_RX_BI_BUFID_S		0
-#define IDPF_RX_BI_BUFID_M		GENMASK(14, 0)
-#define IDPF_RX_BI_GEN_S		15
-#define IDPF_RX_BI_GEN_M		BIT(IDPF_RX_BI_GEN_S)
+#define IDPF_RX_BI_GEN_M		BIT(16)
+#define IDPF_RX_BI_BUFID_M		GENMASK(15, 0)
+
 #define IDPF_RXD_EOF_SPLITQ		VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_EOF_M
 #define IDPF_RXD_EOF_SINGLEQ		VIRTCHNL2_RX_BASE_DESC_STATUS_EOF_M
 
@@ -122,7 +123,7 @@ do {								\
 	((((txq)->next_to_clean > (txq)->next_to_use) ? 0 : (txq)->desc_count) + \
 	(txq)->next_to_clean - (txq)->next_to_use - 1)
 
-#define IDPF_TX_BUF_RSV_UNUSED(txq)	((txq)->buf_stack.top)
+#define IDPF_TX_BUF_RSV_UNUSED(txq)	((txq)->stash->buf_stack.top)
 #define IDPF_TX_BUF_RSV_LOW(txq)	(IDPF_TX_BUF_RSV_UNUSED(txq) < \
 					 (txq)->desc_count >> 2)
 
@@ -433,23 +434,37 @@ struct idpf_rx_ptype_decoded {
  *		      to 1 and knows that reading a gen bit of 1 in any
  *		      descriptor on the initial pass of the ring indicates a
  *		      writeback. It also flips on every ring wrap.
- * @__IDPF_RFLQ_GEN_CHK: Refill queues are SW only, so Q_GEN acts as the HW bit
- *			 and RFLGQ_GEN is the SW bit.
+ * @__IDPF_Q_RFL_GEN_CHK: Refill queues are SW only, so Q_GEN acts as the HW
+ *			  bit and Q_RFL_GEN is the SW bit.
  * @__IDPF_Q_FLOW_SCH_EN: Enable flow scheduling
  * @__IDPF_Q_SW_MARKER: Used to indicate TX queue marker completions
  * @__IDPF_Q_POLL_MODE: Enable poll mode
+ * @__IDPF_Q_CRC_EN: enable CRC offload in singleq mode
+ * @__IDPF_Q_HSPLIT_EN: enable header split on Rx (splitq)
  * @__IDPF_Q_FLAGS_NBITS: Must be last
  */
 enum idpf_queue_flags_t {
 	__IDPF_Q_GEN_CHK,
-	__IDPF_RFLQ_GEN_CHK,
+	__IDPF_Q_RFL_GEN_CHK,
 	__IDPF_Q_FLOW_SCH_EN,
 	__IDPF_Q_SW_MARKER,
 	__IDPF_Q_POLL_MODE,
+	__IDPF_Q_CRC_EN,
+	__IDPF_Q_HSPLIT_EN,
 
 	__IDPF_Q_FLAGS_NBITS,
 };
 
+#define idpf_queue_set(f, q)		__set_bit(__IDPF_Q_##f, (q)->flags)
+#define idpf_queue_clear(f, q)		__clear_bit(__IDPF_Q_##f, (q)->flags)
+#define idpf_queue_change(f, q)		__change_bit(__IDPF_Q_##f, (q)->flags)
+#define idpf_queue_has(f, q)		test_bit(__IDPF_Q_##f, (q)->flags)
+
+#define idpf_queue_has_clear(f, q)			\
+	__test_and_clear_bit(__IDPF_Q_##f, (q)->flags)
+#define idpf_queue_assign(f, q, v)			\
+	__assign_bit(__IDPF_Q_##f, (q)->flags, v)
+
 /**
  * struct idpf_vec_regs
  * @dyn_ctl_reg: Dynamic control interrupt register offset
@@ -495,7 +510,9 @@ struct idpf_intr_reg {
  * @v_idx: Vector index
  * @intr_reg: See struct idpf_intr_reg
  * @num_txq: Number of TX queues
+ * @num_complq: number of completion queues
  * @tx: Array of TX queues to service
+ * @complq: array of completion queues
  * @tx_dim: Data for TX net_dim algorithm
  * @tx_itr_value: TX interrupt throttling rate
  * @tx_intr_mode: Dynamic ITR or not
@@ -519,21 +536,24 @@ struct idpf_q_vector {
 	struct idpf_intr_reg intr_reg;
 
 	u16 num_txq;
-	struct idpf_queue **tx;
+	u16 num_complq;
+	struct idpf_tx_queue **tx;
+	struct idpf_compl_queue **complq;
+
 	struct dim tx_dim;
 	u16 tx_itr_value;
 	bool tx_intr_mode;
 	u32 tx_itr_idx;
 
 	u16 num_rxq;
-	struct idpf_queue **rx;
+	struct idpf_rx_queue **rx;
 	struct dim rx_dim;
 	u16 rx_itr_value;
 	bool rx_intr_mode;
 	u32 rx_itr_idx;
 
 	u16 num_bufq;
-	struct idpf_queue **bufq;
+	struct idpf_buf_queue **bufq;
 
 	u16 total_events;
 	char *name;
@@ -564,11 +584,6 @@ struct idpf_cleaned_stats {
 	u32 bytes;
 };
 
-union idpf_queue_stats {
-	struct idpf_rx_queue_stats rx;
-	struct idpf_tx_queue_stats tx;
-};
-
 #define IDPF_ITR_DYNAMIC	1
 #define IDPF_ITR_MAX		0x1FE0
 #define IDPF_ITR_20K		0x0032
@@ -584,39 +599,114 @@ union idpf_queue_stats {
 #define IDPF_DIM_DEFAULT_PROFILE_IX		1
 
 /**
- * struct idpf_queue
- * @dev: Device back pointer for DMA mapping
- * @vport: Back pointer to associated vport
- * @txq_grp: See struct idpf_txq_group
- * @rxq_grp: See struct idpf_rxq_group
- * @idx: For buffer queue, it is used as group id, either 0 or 1. On clean,
- *	 buffer queue uses this index to determine which group of refill queues
- *	 to clean.
- *	 For TX queue, it is used as index to map between TX queue group and
- *	 hot path TX pointers stored in vport. Used in both singleq/splitq.
- *	 For RX queue, it is used to index to total RX queue across groups and
+ * struct idpf_txq_stash - Tx buffer stash for Flow-based scheduling mode
+ * @buf_stack: Stack of empty buffers to store buffer info for out of order
+ *	       buffer completions. See struct idpf_buf_lifo
+ * @sched_buf_hash: Hash table to store buffers
+ */
+struct idpf_txq_stash {
+	struct idpf_buf_lifo buf_stack;
+	DECLARE_HASHTABLE(sched_buf_hash, 12);
+} ____cacheline_aligned;
+
+/**
+ * struct idpf_rx_queue - software structure representing a receive queue
+ * @rx: universal receive descriptor array
+ * @single_buf: buffer descriptor array in singleq
+ * @desc_ring: virtual descriptor ring address
+ * @bufq_sets: Pointer to the array of buffer queues in splitq mode
+ * @napi: NAPI instance corresponding to this queue (splitq)
+ * @rx_buf: See struct idpf_rx_buf
+ * @pp: Page pool pointer in singleq mode
+ * @netdev: &net_device corresponding to this queue
+ * @tail: Tail offset. Used for both queue models single and split.
+ * @flags: See enum idpf_queue_flags_t
+ * @idx: For RX queue, it is used to index to total RX queue across groups and
  *	 used for skb reporting.
- * @tail: Tail offset. Used for both queue models single and split. In splitq
- *	  model relevant only for TX queue and RX queue.
- * @tx_buf: See struct idpf_tx_buf
- * @rx_buf: Struct with RX buffer related members
- * @rx_buf.buf: See struct idpf_rx_buf
- * @rx_buf.hdr_buf_pa: DMA handle
- * @rx_buf.hdr_buf_va: Virtual address
- * @pp: Page pool pointer
+ * @desc_count: Number of descriptors
+ * @next_to_use: Next descriptor to use
+ * @next_to_clean: Next descriptor to clean
+ * @next_to_alloc: RX buffer to allocate at
+ * @rxdids: Supported RX descriptor ids
+ * @rx_ptype_lkup: LUT of Rx ptypes
  * @skb: Pointer to the skb
- * @q_type: Queue type (TX, RX, TX completion, RX buffer)
+ * @stats_sync: See struct u64_stats_sync
+ * @q_stats: See union idpf_rx_queue_stats
  * @q_id: Queue id
- * @desc_count: Number of descriptors
- * @next_to_use: Next descriptor to use. Relevant in both split & single txq
- *		 and bufq.
- * @next_to_clean: Next descriptor to clean. In split queue model, only
- *		   relevant to TX completion queue and RX queue.
- * @next_to_alloc: RX buffer to allocate at. Used only for RX. In splitq model
- *		   only relevant to RX queue.
+ * @size: Length of descriptor ring in bytes
+ * @dma: Physical address of ring
+ * @q_vector: Backreference to associated vector
+ * @rx_buffer_low_watermark: RX buffer low watermark
+ * @rx_hbuf_size: Header buffer size
+ * @rx_buf_size: Buffer size
+ * @rx_max_pkt_size: RX max packet size
+ */
+struct idpf_rx_queue {
+	union {
+		union virtchnl2_rx_desc *rx;
+		struct virtchnl2_singleq_rx_buf_desc *single_buf;
+
+		void *desc_ring;
+	};
+	union {
+		struct {
+			struct idpf_bufq_set *bufq_sets;
+			struct napi_struct *napi;
+		};
+		struct {
+			struct idpf_rx_buf *rx_buf;
+			struct page_pool *pp;
+		};
+	};
+	struct net_device *netdev;
+	void __iomem *tail;
+
+	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
+	u16 idx;
+	u16 desc_count;
+	u16 next_to_use;
+	u16 next_to_clean;
+	u16 next_to_alloc;
+
+	u32 rxdids;
+
+	const struct idpf_rx_ptype_decoded *rx_ptype_lkup;
+	struct sk_buff *skb;
+
+	struct u64_stats_sync stats_sync;
+	struct idpf_rx_queue_stats q_stats;
+
+	/* Slowpath */
+	u32 q_id;
+	u32 size;
+	dma_addr_t dma;
+
+	struct idpf_q_vector *q_vector;
+
+	u16 rx_buffer_low_watermark;
+	u16 rx_hbuf_size;
+	u16 rx_buf_size;
+	u16 rx_max_pkt_size;
+} ____cacheline_aligned;
+
+/**
+ * struct idpf_tx_queue - software structure representing a transmit queue
+ * @base_tx: base Tx descriptor array
+ * @base_ctx: base Tx context descriptor array
+ * @flex_tx: flex Tx descriptor array
+ * @flex_ctx: flex Tx context descriptor array
+ * @desc_ring: virtual descriptor ring address
+ * @tx_buf: See struct idpf_tx_buf
+ * @txq_grp: See struct idpf_txq_group
+ * @dev: Device back pointer for DMA mapping
+ * @tail: Tail offset. Used for both queue models single and split
  * @flags: See enum idpf_queue_flags_t
- * @q_stats: See union idpf_queue_stats
- * @stats_sync: See struct u64_stats_sync
+ * @idx: For TX queue, it is used as index to map between TX queue group and
+ *	 hot path TX pointers stored in vport. Used in both singleq/splitq.
+ * @desc_count: Number of descriptors
+ * @next_to_use: Next descriptor to use
+ * @next_to_clean: Next descriptor to clean
+ * @netdev: &net_device corresponding to this queue
  * @cleaned_bytes: Splitq only, TXQ only: When a TX completion is received on
  *		   the TX completion queue, it can be for any TXQ associated
  *		   with that completion queue. This means we can clean up to
@@ -625,34 +715,10 @@ union idpf_queue_stats {
  *		   that single call to clean the completion queue. By doing so,
  *		   we can update BQL with aggregate cleaned stats for each TXQ
  *		   only once at the end of the cleaning routine.
+ * @clean_budget: singleq only, queue cleaning budget
  * @cleaned_pkts: Number of packets cleaned for the above said case
- * @rx_hsplit_en: RX headsplit enable
- * @rx_hbuf_size: Header buffer size
- * @rx_buf_size: Buffer size
- * @rx_max_pkt_size: RX max packet size
- * @rx_buf_stride: RX buffer stride
- * @rx_buffer_low_watermark: RX buffer low watermark
- * @rxdids: Supported RX descriptor ids
- * @q_vector: Backreference to associated vector
- * @size: Length of descriptor ring in bytes
- * @dma: Physical address of ring
- * @rx: universal receive descriptor array
- * @single_buf: Rx buffer descriptor array in singleq
- * @split_buf: Rx buffer descriptor array in splitq
- * @base_tx: basic Tx descriptor array
- * @base_ctx: basic Tx context descriptor array
- * @flex_tx: flex Tx descriptor array
- * @flex_ctx: flex Tx context descriptor array
- * @comp: completion descriptor array
- * @desc_ring: virtual descriptor ring address
  * @tx_max_bufs: Max buffers that can be transmitted with scatter-gather
  * @tx_min_pkt_len: Min supported packet length
- * @num_completions: Only relevant for TX completion queue. It tracks the
- *		     number of completions received to compare against the
- *		     number of completions pending, as accumulated by the
- *		     TX queues.
- * @buf_stack: Stack of empty buffers to store buffer info for out of order
- *	       buffer completions. See struct idpf_buf_lifo.
  * @compl_tag_bufid_m: Completion tag buffer id mask
  * @compl_tag_gen_s: Completion tag generation bit
  *	The format of the completion tag will change based on the TXQ
@@ -676,120 +742,188 @@ union idpf_queue_stats {
  *	This gives us 8*8160 = 65280 possible unique values.
  * @compl_tag_cur_gen: Used to keep track of current completion tag generation
  * @compl_tag_gen_max: To determine when compl_tag_cur_gen should be reset
- * @sched_buf_hash: Hash table to stores buffers
+ * @stash: Tx buffer stash for Flow-based scheduling mode
+ * @stats_sync: See struct u64_stats_sync
+ * @q_stats: See union idpf_tx_queue_stats
+ * @q_id: Queue id
+ * @size: Length of descriptor ring in bytes
+ * @dma: Physical address of ring
+ * @q_vector: Backreference to associated vector
  */
-struct idpf_queue {
-	struct device *dev;
-	struct idpf_vport *vport;
+struct idpf_tx_queue {
 	union {
-		struct idpf_txq_group *txq_grp;
-		struct idpf_rxq_group *rxq_grp;
+		struct idpf_base_tx_desc *base_tx;
+		struct idpf_base_tx_ctx_desc *base_ctx;
+		union idpf_tx_flex_desc *flex_tx;
+		struct idpf_flex_tx_ctx_desc *flex_ctx;
+
+		void *desc_ring;
 	};
-	u16 idx;
+	struct idpf_tx_buf *tx_buf;
+	struct idpf_txq_group *txq_grp;
+	struct device *dev;
 	void __iomem *tail;
-	union {
-		struct idpf_tx_buf *tx_buf;
-		struct {
-			struct idpf_rx_buf *buf;
-			dma_addr_t hdr_buf_pa;
-			void *hdr_buf_va;
-		} rx_buf;
-	};
-	struct page_pool *pp;
-	struct sk_buff *skb;
-	u16 q_type;
-	u32 q_id;
-	u16 desc_count;
 
+	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
+	u16 idx;
+	u16 desc_count;
 	u16 next_to_use;
 	u16 next_to_clean;
-	u16 next_to_alloc;
-	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
 
-	union idpf_queue_stats q_stats;
-	struct u64_stats_sync stats_sync;
+	struct net_device *netdev;
 
-	u32 cleaned_bytes;
+	union {
+		u32 cleaned_bytes;
+		u32 clean_budget;
+	};
 	u16 cleaned_pkts;
 
-	bool rx_hsplit_en;
-	u16 rx_hbuf_size;
-	u16 rx_buf_size;
-	u16 rx_max_pkt_size;
-	u16 rx_buf_stride;
-	u8 rx_buffer_low_watermark;
-	u64 rxdids;
-	struct idpf_q_vector *q_vector;
-	unsigned int size;
+	u16 tx_max_bufs;
+	u16 tx_min_pkt_len;
+
+	u16 compl_tag_bufid_m;
+	u16 compl_tag_gen_s;
+
+	u16 compl_tag_cur_gen;
+	u16 compl_tag_gen_max;
+
+	struct idpf_txq_stash *stash;
+
+	struct u64_stats_sync stats_sync;
+	struct idpf_tx_queue_stats q_stats;
+
+	/* Slowpath */
+	u32 q_id;
+	u32 size;
 	dma_addr_t dma;
-	union {
-		union virtchnl2_rx_desc *rx;
 
-		struct virtchnl2_singleq_rx_buf_desc *single_buf;
-		struct virtchnl2_splitq_rx_buf_desc *split_buf;
+	struct idpf_q_vector *q_vector;
+} ____cacheline_aligned;
 
-		struct idpf_base_tx_desc *base_tx;
-		struct idpf_base_tx_ctx_desc *base_ctx;
-		union idpf_tx_flex_desc *flex_tx;
-		struct idpf_flex_tx_ctx_desc *flex_ctx;
+/**
+ * struct idpf_buf_queue - software structure representing a buffer queue
+ * @split_buf: buffer descriptor array
+ * @rx_buf: Struct with RX buffer related members
+ * @rx_buf.buf: See struct idpf_rx_buf
+ * @rx_buf.hdr_buf_pa: DMA handle
+ * @rx_buf.hdr_buf_va: Virtual address
+ * @pp: Page pool pointer
+ * @tail: Tail offset
+ * @flags: See enum idpf_queue_flags_t
+ * @desc_count: Number of descriptors
+ * @next_to_use: Next descriptor to use
+ * @next_to_clean: Next descriptor to clean
+ * @next_to_alloc: RX buffer to allocate at
+ * @q_id: Queue id
+ * @size: Length of descriptor ring in bytes
+ * @dma: Physical address of ring
+ * @q_vector: Backreference to associated vector
+ * @rx_buffer_low_watermark: RX buffer low watermark
+ * @rx_hbuf_size: Header buffer size
+ * @rx_buf_size: Buffer size
+ */
+struct idpf_buf_queue {
+	struct virtchnl2_splitq_rx_buf_desc *split_buf;
+	struct {
+		struct idpf_rx_buf *buf;
+		dma_addr_t hdr_buf_pa;
+		void *hdr_buf_va;
+	} rx_buf;
+	struct page_pool *pp;
+	void __iomem *tail;
 
-		struct idpf_splitq_tx_compl_desc *comp;
+	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
+	u16 desc_count;
+	u16 next_to_use;
+	u16 next_to_clean;
+	u16 next_to_alloc;
 
-		void *desc_ring;
-	};
+	/* Slowpath */
+	u32 q_id;
+	u32 size;
+	dma_addr_t dma;
 
-	u16 tx_max_bufs;
-	u8 tx_min_pkt_len;
+	struct idpf_q_vector *q_vector;
 
-	u32 num_completions;
+	u16 rx_buffer_low_watermark;
+	u16 rx_hbuf_size;
+	u16 rx_buf_size;
+} ____cacheline_aligned;
 
-	struct idpf_buf_lifo buf_stack;
+/**
+ * struct idpf_compl_queue - software structure representing a completion queue
+ * @comp: completion descriptor array
+ * @txq_grp: See struct idpf_txq_group
+ * @flags: See enum idpf_queue_flags_t
+ * @desc_count: Number of descriptors
+ * @next_to_use: Next descriptor to use. Relevant in both split & single txq
+ *		 and bufq.
+ * @next_to_clean: Next descriptor to clean
+ * @netdev: &net_device corresponding to this queue
+ * @clean_budget: queue cleaning budget
+ * @num_completions: Only relevant for TX completion queue. It tracks the
+ *		     number of completions received to compare against the
+ *		     number of completions pending, as accumulated by the
+ *		     TX queues.
+ * @q_id: Queue id
+ * @size: Length of descriptor ring in bytes
+ * @dma: Physical address of ring
+ * @q_vector: Backreference to associated vector
+ */
+struct idpf_compl_queue {
+	struct idpf_splitq_tx_compl_desc *comp;
+	struct idpf_txq_group *txq_grp;
 
-	u16 compl_tag_bufid_m;
-	u16 compl_tag_gen_s;
+	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
+	u16 desc_count;
+	u16 next_to_use;
+	u16 next_to_clean;
 
-	u16 compl_tag_cur_gen;
-	u16 compl_tag_gen_max;
+	struct net_device *netdev;
+	u32 clean_budget;
+	u32 num_completions;
 
-	DECLARE_HASHTABLE(sched_buf_hash, 12);
-} ____cacheline_internodealigned_in_smp;
+	/* Slowpath */
+	u32 q_id;
+	u32 size;
+	dma_addr_t dma;
+
+	struct idpf_q_vector *q_vector;
+} ____cacheline_aligned;
 
 /**
  * struct idpf_sw_queue
- * @next_to_clean: Next descriptor to clean
- * @next_to_alloc: Buffer to allocate at
- * @flags: See enum idpf_queue_flags_t
  * @ring: Pointer to the ring
+ * @flags: See enum idpf_queue_flags_t
  * @desc_count: Descriptor count
- * @dev: Device back pointer for DMA mapping
+ * @next_to_use: Buffer to allocate at
+ * @next_to_clean: Next descriptor to clean
  *
  * Software queues are used in splitq mode to manage buffers between rxq
  * producer and the bufq consumer.  These are required in order to maintain a
  * lockless buffer management system and are strictly software only constructs.
  */
 struct idpf_sw_queue {
-	u16 next_to_clean;
-	u16 next_to_alloc;
+	u32 *ring;
+
 	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
-	u16 *ring;
 	u16 desc_count;
-	struct device *dev;
-} ____cacheline_internodealigned_in_smp;
+	u16 next_to_use;
+	u16 next_to_clean;
+} ____cacheline_aligned;
 
 /**
  * struct idpf_rxq_set
  * @rxq: RX queue
- * @refillq0: Pointer to refill queue 0
- * @refillq1: Pointer to refill queue 1
+ * @refillq: pointers to refill queues
  *
  * Splitq only.  idpf_rxq_set associates an rxq with at an array of refillqs.
  * Each rxq needs a refillq to return used buffers back to the respective bufq.
  * Bufqs then clean these refillqs for buffers to give to hardware.
  */
 struct idpf_rxq_set {
-	struct idpf_queue rxq;
-	struct idpf_sw_queue *refillq0;
-	struct idpf_sw_queue *refillq1;
+	struct idpf_rx_queue rxq;
+	struct idpf_sw_queue *refillq[IDPF_MAX_BUFQS_PER_RXQ_GRP];
 };
 
 /**
@@ -808,7 +942,7 @@ struct idpf_rxq_set {
  * managed by at most two bufqs (depending on performance configuration).
  */
 struct idpf_bufq_set {
-	struct idpf_queue bufq;
+	struct idpf_buf_queue bufq;
 	int num_refillqs;
 	struct idpf_sw_queue *refillqs;
 };
@@ -834,7 +968,7 @@ struct idpf_rxq_group {
 	union {
 		struct {
 			u16 num_rxq;
-			struct idpf_queue *rxqs[IDPF_LARGE_MAX_Q];
+			struct idpf_rx_queue *rxqs[IDPF_LARGE_MAX_Q];
 		} singleq;
 		struct {
 			u16 num_rxq_sets;
@@ -849,6 +983,7 @@ struct idpf_rxq_group {
  * @vport: Vport back pointer
  * @num_txq: Number of TX queues associated
  * @txqs: Array of TX queue pointers
+ * @stashes: array of OOO stashes for the queues
  * @complq: Associated completion queue pointer, split queue only
  * @num_completions_pending: Total number of completions pending for the
  *			     completion queue, acculumated for all TX queues
@@ -862,9 +997,10 @@ struct idpf_txq_group {
 	struct idpf_vport *vport;
 
 	u16 num_txq;
-	struct idpf_queue *txqs[IDPF_LARGE_MAX_Q];
+	struct idpf_tx_queue *txqs[IDPF_LARGE_MAX_Q];
+	struct idpf_txq_stash *stashes;
 
-	struct idpf_queue *complq;
+	struct idpf_compl_queue *complq;
 
 	u32 num_completions_pending;
 };
@@ -1001,28 +1137,26 @@ void idpf_deinit_rss(struct idpf_vport *vport);
 int idpf_rx_bufs_init_all(struct idpf_vport *vport);
 void idpf_rx_add_frag(struct idpf_rx_buf *rx_buf, struct sk_buff *skb,
 		      unsigned int size);
-struct sk_buff *idpf_rx_construct_skb(struct idpf_queue *rxq,
+struct sk_buff *idpf_rx_construct_skb(const struct idpf_rx_queue *rxq,
 				      struct idpf_rx_buf *rx_buf,
 				      unsigned int size);
-bool idpf_init_rx_buf_hw_alloc(struct idpf_queue *rxq, struct idpf_rx_buf *buf);
-void idpf_rx_buf_hw_update(struct idpf_queue *rxq, u32 val);
-void idpf_tx_buf_hw_update(struct idpf_queue *tx_q, u32 val,
+void idpf_tx_buf_hw_update(struct idpf_tx_queue *tx_q, u32 val,
 			   bool xmit_more);
 unsigned int idpf_size_to_txd_count(unsigned int size);
-netdev_tx_t idpf_tx_drop_skb(struct idpf_queue *tx_q, struct sk_buff *skb);
-void idpf_tx_dma_map_error(struct idpf_queue *txq, struct sk_buff *skb,
+netdev_tx_t idpf_tx_drop_skb(struct idpf_tx_queue *tx_q, struct sk_buff *skb);
+void idpf_tx_dma_map_error(struct idpf_tx_queue *txq, struct sk_buff *skb,
 			   struct idpf_tx_buf *first, u16 ring_idx);
-unsigned int idpf_tx_desc_count_required(struct idpf_queue *txq,
+unsigned int idpf_tx_desc_count_required(struct idpf_tx_queue *txq,
 					 struct sk_buff *skb);
 bool idpf_chk_linearize(struct sk_buff *skb, unsigned int max_bufs,
 			unsigned int count);
-int idpf_tx_maybe_stop_common(struct idpf_queue *tx_q, unsigned int size);
+int idpf_tx_maybe_stop_common(struct idpf_tx_queue *tx_q, unsigned int size);
 void idpf_tx_timeout(struct net_device *netdev, unsigned int txqueue);
 netdev_tx_t idpf_tx_splitq_start(struct sk_buff *skb,
 				 struct net_device *netdev);
 netdev_tx_t idpf_tx_singleq_start(struct sk_buff *skb,
 				  struct net_device *netdev);
-bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_queue *rxq,
+bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_rx_queue *rxq,
 				      u16 cleaned_count);
 int idpf_tso(struct sk_buff *skb, struct idpf_tx_offload_params *off);
 
diff --git a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
index a5f9b7a5effe..44602b87cd41 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
@@ -750,7 +750,7 @@ static int idpf_wait_for_marker_event(struct idpf_vport *vport)
 	int i;
 
 	for (i = 0; i < vport->num_txq; i++)
-		set_bit(__IDPF_Q_SW_MARKER, vport->txqs[i]->flags);
+		idpf_queue_set(SW_MARKER, vport->txqs[i]);
 
 	event = wait_event_timeout(vport->sw_marker_wq,
 				   test_and_clear_bit(IDPF_VPORT_SW_MARKER,
@@ -758,7 +758,7 @@ static int idpf_wait_for_marker_event(struct idpf_vport *vport)
 				   msecs_to_jiffies(500));
 
 	for (i = 0; i < vport->num_txq; i++)
-		clear_bit(__IDPF_Q_POLL_MODE, vport->txqs[i]->flags);
+		idpf_queue_clear(POLL_MODE, vport->txqs[i]);
 
 	if (event)
 		return 0;
@@ -1092,7 +1092,6 @@ static int __idpf_queue_reg_init(struct idpf_vport *vport, u32 *reg_vals,
 				 int num_regs, u32 q_type)
 {
 	struct idpf_adapter *adapter = vport->adapter;
-	struct idpf_queue *q;
 	int i, j, k = 0;
 
 	switch (q_type) {
@@ -1111,6 +1110,8 @@ static int __idpf_queue_reg_init(struct idpf_vport *vport, u32 *reg_vals,
 			u16 num_rxq = rx_qgrp->singleq.num_rxq;
 
 			for (j = 0; j < num_rxq && k < num_regs; j++, k++) {
+				struct idpf_rx_queue *q;
+
 				q = rx_qgrp->singleq.rxqs[j];
 				q->tail = idpf_get_reg_addr(adapter,
 							    reg_vals[k]);
@@ -1123,6 +1124,8 @@ static int __idpf_queue_reg_init(struct idpf_vport *vport, u32 *reg_vals,
 			u8 num_bufqs = vport->num_bufqs_per_qgrp;
 
 			for (j = 0; j < num_bufqs && k < num_regs; j++, k++) {
+				struct idpf_buf_queue *q;
+
 				q = &rx_qgrp->splitq.bufq_sets[j].bufq;
 				q->tail = idpf_get_reg_addr(adapter,
 							    reg_vals[k]);
@@ -1449,19 +1452,19 @@ static int idpf_send_config_tx_queues_msg(struct idpf_vport *vport)
 			qi[k].model =
 				cpu_to_le16(vport->txq_model);
 			qi[k].type =
-				cpu_to_le32(tx_qgrp->txqs[j]->q_type);
+				cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_TX);
 			qi[k].ring_len =
 				cpu_to_le16(tx_qgrp->txqs[j]->desc_count);
 			qi[k].dma_ring_addr =
 				cpu_to_le64(tx_qgrp->txqs[j]->dma);
 			if (idpf_is_queue_model_split(vport->txq_model)) {
-				struct idpf_queue *q = tx_qgrp->txqs[j];
+				struct idpf_tx_queue *q = tx_qgrp->txqs[j];
 
 				qi[k].tx_compl_queue_id =
 					cpu_to_le16(tx_qgrp->complq->q_id);
 				qi[k].relative_queue_id = cpu_to_le16(j);
 
-				if (test_bit(__IDPF_Q_FLOW_SCH_EN, q->flags))
+				if (idpf_queue_has(FLOW_SCH_EN, q))
 					qi[k].sched_mode =
 					cpu_to_le16(VIRTCHNL2_TXQ_SCHED_MODE_FLOW);
 				else
@@ -1478,11 +1481,11 @@ static int idpf_send_config_tx_queues_msg(struct idpf_vport *vport)
 
 		qi[k].queue_id = cpu_to_le32(tx_qgrp->complq->q_id);
 		qi[k].model = cpu_to_le16(vport->txq_model);
-		qi[k].type = cpu_to_le32(tx_qgrp->complq->q_type);
+		qi[k].type = cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_TX_COMPLETION);
 		qi[k].ring_len = cpu_to_le16(tx_qgrp->complq->desc_count);
 		qi[k].dma_ring_addr = cpu_to_le64(tx_qgrp->complq->dma);
 
-		if (test_bit(__IDPF_Q_FLOW_SCH_EN, tx_qgrp->complq->flags))
+		if (idpf_queue_has(FLOW_SCH_EN, tx_qgrp->complq))
 			sched_mode = VIRTCHNL2_TXQ_SCHED_MODE_FLOW;
 		else
 			sched_mode = VIRTCHNL2_TXQ_SCHED_MODE_QUEUE;
@@ -1567,17 +1570,18 @@ static int idpf_send_config_rx_queues_msg(struct idpf_vport *vport)
 			goto setup_rxqs;
 
 		for (j = 0; j < vport->num_bufqs_per_qgrp; j++, k++) {
-			struct idpf_queue *bufq =
+			struct idpf_buf_queue *bufq =
 				&rx_qgrp->splitq.bufq_sets[j].bufq;
 
 			qi[k].queue_id = cpu_to_le32(bufq->q_id);
 			qi[k].model = cpu_to_le16(vport->rxq_model);
-			qi[k].type = cpu_to_le32(bufq->q_type);
+			qi[k].type =
+				cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_RX_BUFFER);
 			qi[k].desc_ids = cpu_to_le64(VIRTCHNL2_RXDID_2_FLEX_SPLITQ_M);
 			qi[k].ring_len = cpu_to_le16(bufq->desc_count);
 			qi[k].dma_ring_addr = cpu_to_le64(bufq->dma);
 			qi[k].data_buffer_size = cpu_to_le32(bufq->rx_buf_size);
-			qi[k].buffer_notif_stride = bufq->rx_buf_stride;
+			qi[k].buffer_notif_stride = IDPF_RX_BUF_STRIDE;
 			qi[k].rx_buffer_low_watermark =
 				cpu_to_le16(bufq->rx_buffer_low_watermark);
 			if (idpf_is_feature_ena(vport, NETIF_F_GRO_HW))
@@ -1591,7 +1595,7 @@ setup_rxqs:
 			num_rxq = rx_qgrp->singleq.num_rxq;
 
 		for (j = 0; j < num_rxq; j++, k++) {
-			struct idpf_queue *rxq;
+			struct idpf_rx_queue *rxq;
 
 			if (!idpf_is_queue_model_split(vport->rxq_model)) {
 				rxq = rx_qgrp->singleq.rxqs[j];
@@ -1599,11 +1603,11 @@ setup_rxqs:
 			}
 			rxq = &rx_qgrp->splitq.rxq_sets[j]->rxq;
 			qi[k].rx_bufq1_id =
-			  cpu_to_le16(rxq->rxq_grp->splitq.bufq_sets[0].bufq.q_id);
+			  cpu_to_le16(rxq->bufq_sets[0].bufq.q_id);
 			if (vport->num_bufqs_per_qgrp > IDPF_SINGLE_BUFQ_PER_RXQ_GRP) {
 				qi[k].bufq2_ena = IDPF_BUFQ2_ENA;
 				qi[k].rx_bufq2_id =
-				  cpu_to_le16(rxq->rxq_grp->splitq.bufq_sets[1].bufq.q_id);
+				  cpu_to_le16(rxq->bufq_sets[1].bufq.q_id);
 			}
 			qi[k].rx_buffer_low_watermark =
 				cpu_to_le16(rxq->rx_buffer_low_watermark);
@@ -1611,7 +1615,7 @@ setup_rxqs:
 				qi[k].qflags |= cpu_to_le16(VIRTCHNL2_RXQ_RSC);
 
 common_qi_fields:
-			if (rxq->rx_hsplit_en) {
+			if (idpf_queue_has(HSPLIT_EN, rxq)) {
 				qi[k].qflags |=
 					cpu_to_le16(VIRTCHNL2_RXQ_HDR_SPLIT);
 				qi[k].hdr_buffer_size =
@@ -1619,7 +1623,7 @@ common_qi_fields:
 			}
 			qi[k].queue_id = cpu_to_le32(rxq->q_id);
 			qi[k].model = cpu_to_le16(vport->rxq_model);
-			qi[k].type = cpu_to_le32(rxq->q_type);
+			qi[k].type = cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_RX);
 			qi[k].ring_len = cpu_to_le16(rxq->desc_count);
 			qi[k].dma_ring_addr = cpu_to_le64(rxq->dma);
 			qi[k].max_pkt_size = cpu_to_le32(rxq->rx_max_pkt_size);
@@ -1706,7 +1710,7 @@ static int idpf_send_ena_dis_queues_msg(struct idpf_vport *vport, bool ena)
 		struct idpf_txq_group *tx_qgrp = &vport->txq_grps[i];
 
 		for (j = 0; j < tx_qgrp->num_txq; j++, k++) {
-			qc[k].type = cpu_to_le32(tx_qgrp->txqs[j]->q_type);
+			qc[k].type = cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_TX);
 			qc[k].start_queue_id = cpu_to_le32(tx_qgrp->txqs[j]->q_id);
 			qc[k].num_queues = cpu_to_le32(IDPF_NUMQ_PER_CHUNK);
 		}
@@ -1720,7 +1724,7 @@ static int idpf_send_ena_dis_queues_msg(struct idpf_vport *vport, bool ena)
 	for (i = 0; i < vport->num_txq_grp; i++, k++) {
 		struct idpf_txq_group *tx_qgrp = &vport->txq_grps[i];
 
-		qc[k].type = cpu_to_le32(tx_qgrp->complq->q_type);
+		qc[k].type = cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_TX_COMPLETION);
 		qc[k].start_queue_id = cpu_to_le32(tx_qgrp->complq->q_id);
 		qc[k].num_queues = cpu_to_le32(IDPF_NUMQ_PER_CHUNK);
 	}
@@ -1741,12 +1745,12 @@ setup_rx:
 				qc[k].start_queue_id =
 				cpu_to_le32(rx_qgrp->splitq.rxq_sets[j]->rxq.q_id);
 				qc[k].type =
-				cpu_to_le32(rx_qgrp->splitq.rxq_sets[j]->rxq.q_type);
+				cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_RX);
 			} else {
 				qc[k].start_queue_id =
 				cpu_to_le32(rx_qgrp->singleq.rxqs[j]->q_id);
 				qc[k].type =
-				cpu_to_le32(rx_qgrp->singleq.rxqs[j]->q_type);
+				cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_RX);
 			}
 			qc[k].num_queues = cpu_to_le32(IDPF_NUMQ_PER_CHUNK);
 		}
@@ -1761,10 +1765,11 @@ setup_rx:
 		struct idpf_rxq_group *rx_qgrp = &vport->rxq_grps[i];
 
 		for (j = 0; j < vport->num_bufqs_per_qgrp; j++, k++) {
-			struct idpf_queue *q;
+			const struct idpf_buf_queue *q;
 
 			q = &rx_qgrp->splitq.bufq_sets[j].bufq;
-			qc[k].type = cpu_to_le32(q->q_type);
+			qc[k].type =
+				cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_RX_BUFFER);
 			qc[k].start_queue_id = cpu_to_le32(q->q_id);
 			qc[k].num_queues = cpu_to_le32(IDPF_NUMQ_PER_CHUNK);
 		}
@@ -1849,7 +1854,8 @@ int idpf_send_map_unmap_queue_vector_msg(struct idpf_vport *vport, bool map)
 		struct idpf_txq_group *tx_qgrp = &vport->txq_grps[i];
 
 		for (j = 0; j < tx_qgrp->num_txq; j++, k++) {
-			vqv[k].queue_type = cpu_to_le32(tx_qgrp->txqs[j]->q_type);
+			vqv[k].queue_type =
+				cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_TX);
 			vqv[k].queue_id = cpu_to_le32(tx_qgrp->txqs[j]->q_id);
 
 			if (idpf_is_queue_model_split(vport->txq_model)) {
@@ -1879,14 +1885,15 @@ int idpf_send_map_unmap_queue_vector_msg(struct idpf_vport *vport, bool map)
 			num_rxq = rx_qgrp->singleq.num_rxq;
 
 		for (j = 0; j < num_rxq; j++, k++) {
-			struct idpf_queue *rxq;
+			struct idpf_rx_queue *rxq;
 
 			if (idpf_is_queue_model_split(vport->rxq_model))
 				rxq = &rx_qgrp->splitq.rxq_sets[j]->rxq;
 			else
 				rxq = rx_qgrp->singleq.rxqs[j];
 
-			vqv[k].queue_type = cpu_to_le32(rxq->q_type);
+			vqv[k].queue_type =
+				cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_RX);
 			vqv[k].queue_id = cpu_to_le32(rxq->q_id);
 			vqv[k].vector_id = cpu_to_le16(rxq->q_vector->v_idx);
 			vqv[k].itr_idx = cpu_to_le32(rxq->q_vector->rx_itr_idx);
@@ -1975,7 +1982,7 @@ int idpf_send_disable_queues_msg(struct idpf_vport *vport)
 	 * queues virtchnl message is sent
 	 */
 	for (i = 0; i < vport->num_txq; i++)
-		set_bit(__IDPF_Q_POLL_MODE, vport->txqs[i]->flags);
+		idpf_queue_set(POLL_MODE, vport->txqs[i]);
 
 	/* schedule the napi to receive all the marker packets */
 	local_bh_disable();
@@ -3242,7 +3249,6 @@ static int __idpf_vport_queue_ids_init(struct idpf_vport *vport,
 				       int num_qids,
 				       u32 q_type)
 {
-	struct idpf_queue *q;
 	int i, j, k = 0;
 
 	switch (q_type) {
@@ -3250,11 +3256,8 @@ static int __idpf_vport_queue_ids_init(struct idpf_vport *vport,
 		for (i = 0; i < vport->num_txq_grp; i++) {
 			struct idpf_txq_group *tx_qgrp = &vport->txq_grps[i];
 
-			for (j = 0; j < tx_qgrp->num_txq && k < num_qids; j++, k++) {
+			for (j = 0; j < tx_qgrp->num_txq && k < num_qids; j++, k++)
 				tx_qgrp->txqs[j]->q_id = qids[k];
-				tx_qgrp->txqs[j]->q_type =
-					VIRTCHNL2_QUEUE_TYPE_TX;
-			}
 		}
 		break;
 	case VIRTCHNL2_QUEUE_TYPE_RX:
@@ -3268,12 +3271,13 @@ static int __idpf_vport_queue_ids_init(struct idpf_vport *vport,
 				num_rxq = rx_qgrp->singleq.num_rxq;
 
 			for (j = 0; j < num_rxq && k < num_qids; j++, k++) {
+				struct idpf_rx_queue *q;
+
 				if (idpf_is_queue_model_split(vport->rxq_model))
 					q = &rx_qgrp->splitq.rxq_sets[j]->rxq;
 				else
 					q = rx_qgrp->singleq.rxqs[j];
 				q->q_id = qids[k];
-				q->q_type = VIRTCHNL2_QUEUE_TYPE_RX;
 			}
 		}
 		break;
@@ -3282,8 +3286,6 @@ static int __idpf_vport_queue_ids_init(struct idpf_vport *vport,
 			struct idpf_txq_group *tx_qgrp = &vport->txq_grps[i];
 
 			tx_qgrp->complq->q_id = qids[k];
-			tx_qgrp->complq->q_type =
-				VIRTCHNL2_QUEUE_TYPE_TX_COMPLETION;
 		}
 		break;
 	case VIRTCHNL2_QUEUE_TYPE_RX_BUFFER:
@@ -3292,9 +3294,10 @@ static int __idpf_vport_queue_ids_init(struct idpf_vport *vport,
 			u8 num_bufqs = vport->num_bufqs_per_qgrp;
 
 			for (j = 0; j < num_bufqs && k < num_qids; j++, k++) {
+				struct idpf_buf_queue *q;
+
 				q = &rx_qgrp->splitq.bufq_sets[j].bufq;
 				q->q_id = qids[k];
-				q->q_type = VIRTCHNL2_QUEUE_TYPE_RX_BUFFER;
 			}
 		}
 		break;
-- 
cgit v1.2.3-58-ga151


From bf9bf7042a38ebd2485592467772db50605bd4a2 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:39 +0200
Subject: idpf: avoid bloating &idpf_q_vector with big %NR_CPUS

With CONFIG_MAXSMP, sizeof(cpumask_t) is 1 Kb. The queue vector
structure has them embedded, which means 1 additional Kb of not
really hotpath data.
We have cpumask_var_t, which is either an embedded cpumask or a pointer
for allocating it dynamically when it's big. Use it instead of plain
cpumasks and put &idpf_q_vector on a good diet.
Also remove redundant pointer to the interrupt name from the structure.
request_irq() saves it and free_irq() returns it on deinit, so that you
can free the memory.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/idpf/idpf_lib.c  | 13 ++++++-------
 drivers/net/ethernet/intel/idpf/idpf_txrx.c | 20 +++++++++++++-------
 drivers/net/ethernet/intel/idpf/idpf_txrx.h |  7 +++----
 3 files changed, 22 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/intel/idpf/idpf_lib.c b/drivers/net/ethernet/intel/idpf/idpf_lib.c
index a5612120ad51..fafd279127a3 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_lib.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_lib.c
@@ -69,7 +69,7 @@ static void idpf_deinit_vector_stack(struct idpf_adapter *adapter)
 static void idpf_mb_intr_rel_irq(struct idpf_adapter *adapter)
 {
 	clear_bit(IDPF_MB_INTR_MODE, adapter->flags);
-	free_irq(adapter->msix_entries[0].vector, adapter);
+	kfree(free_irq(adapter->msix_entries[0].vector, adapter));
 	queue_delayed_work(adapter->mbx_wq, &adapter->mbx_task, 0);
 }
 
@@ -124,15 +124,14 @@ static void idpf_mb_irq_enable(struct idpf_adapter *adapter)
  */
 static int idpf_mb_intr_req_irq(struct idpf_adapter *adapter)
 {
-	struct idpf_q_vector *mb_vector = &adapter->mb_vector;
 	int irq_num, mb_vidx = 0, err;
+	char *name;
 
 	irq_num = adapter->msix_entries[mb_vidx].vector;
-	mb_vector->name = kasprintf(GFP_KERNEL, "%s-%s-%d",
-				    dev_driver_string(&adapter->pdev->dev),
-				    "Mailbox", mb_vidx);
-	err = request_irq(irq_num, adapter->irq_mb_handler, 0,
-			  mb_vector->name, adapter);
+	name = kasprintf(GFP_KERNEL, "%s-%s-%d",
+			 dev_driver_string(&adapter->pdev->dev),
+			 "Mailbox", mb_vidx);
+	err = request_irq(irq_num, adapter->irq_mb_handler, 0, name, adapter);
 	if (err) {
 		dev_err(&adapter->pdev->dev,
 			"IRQ request for mailbox failed, error: %d\n", err);
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
index 7be5a723f558..f569ea389b04 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
@@ -3613,6 +3613,8 @@ void idpf_vport_intr_rel(struct idpf_vport *vport)
 		q_vector->tx = NULL;
 		kfree(q_vector->rx);
 		q_vector->rx = NULL;
+
+		free_cpumask_var(q_vector->affinity_mask);
 	}
 
 	/* Clean up the mapping of queues to vectors */
@@ -3661,7 +3663,7 @@ static void idpf_vport_intr_rel_irq(struct idpf_vport *vport)
 
 		/* clear the affinity_mask in the IRQ descriptor */
 		irq_set_affinity_hint(irq_num, NULL);
-		free_irq(irq_num, q_vector);
+		kfree(free_irq(irq_num, q_vector));
 	}
 }
 
@@ -3812,6 +3814,7 @@ static int idpf_vport_intr_req_irq(struct idpf_vport *vport, char *basename)
 
 	for (vector = 0; vector < vport->num_q_vectors; vector++) {
 		struct idpf_q_vector *q_vector = &vport->q_vectors[vector];
+		char *name;
 
 		vidx = vport->q_vector_idxs[vector];
 		irq_num = adapter->msix_entries[vidx].vector;
@@ -3825,18 +3828,18 @@ static int idpf_vport_intr_req_irq(struct idpf_vport *vport, char *basename)
 		else
 			continue;
 
-		q_vector->name = kasprintf(GFP_KERNEL, "%s-%s-%d",
-					   basename, vec_name, vidx);
+		name = kasprintf(GFP_KERNEL, "%s-%s-%d", basename, vec_name,
+				 vidx);
 
 		err = request_irq(irq_num, idpf_vport_intr_clean_queues, 0,
-				  q_vector->name, q_vector);
+				  name, q_vector);
 		if (err) {
 			netdev_err(vport->netdev,
 				   "Request_irq failed, error: %d\n", err);
 			goto free_q_irqs;
 		}
 		/* assign the mask for this irq */
-		irq_set_affinity_hint(irq_num, &q_vector->affinity_mask);
+		irq_set_affinity_hint(irq_num, q_vector->affinity_mask);
 	}
 
 	return 0;
@@ -3845,7 +3848,7 @@ free_q_irqs:
 	while (--vector >= 0) {
 		vidx = vport->q_vector_idxs[vector];
 		irq_num = adapter->msix_entries[vidx].vector;
-		free_irq(irq_num, &vport->q_vectors[vector]);
+		kfree(free_irq(irq_num, &vport->q_vectors[vector]));
 	}
 
 	return err;
@@ -4255,7 +4258,7 @@ static void idpf_vport_intr_napi_add_all(struct idpf_vport *vport)
 
 		/* only set affinity_mask if the CPU is online */
 		if (cpu_online(v_idx))
-			cpumask_set_cpu(v_idx, &q_vector->affinity_mask);
+			cpumask_set_cpu(v_idx, q_vector->affinity_mask);
 	}
 }
 
@@ -4299,6 +4302,9 @@ int idpf_vport_intr_alloc(struct idpf_vport *vport)
 		q_vector->rx_intr_mode = IDPF_ITR_DYNAMIC;
 		q_vector->rx_itr_idx = VIRTCHNL2_ITR_IDX_0;
 
+		if (!zalloc_cpumask_var(&q_vector->affinity_mask, GFP_KERNEL))
+			goto error;
+
 		q_vector->tx = kcalloc(txqs_per_vector, sizeof(*q_vector->tx),
 				       GFP_KERNEL);
 		if (!q_vector->tx)
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
index 704aec5c383b..5daa8f905f86 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
@@ -505,7 +505,6 @@ struct idpf_intr_reg {
 /**
  * struct idpf_q_vector
  * @vport: Vport back pointer
- * @affinity_mask: CPU affinity mask
  * @napi: napi handler
  * @v_idx: Vector index
  * @intr_reg: See struct idpf_intr_reg
@@ -526,11 +525,10 @@ struct idpf_intr_reg {
  * @num_bufq: Number of buffer queues
  * @bufq: Array of buffer queues to service
  * @total_events: Number of interrupts processed
- * @name: Queue vector name
+ * @affinity_mask: CPU affinity mask
  */
 struct idpf_q_vector {
 	struct idpf_vport *vport;
-	cpumask_t affinity_mask;
 	struct napi_struct napi;
 	u16 v_idx;
 	struct idpf_intr_reg intr_reg;
@@ -556,7 +554,8 @@ struct idpf_q_vector {
 	struct idpf_buf_queue **bufq;
 
 	u16 total_events;
-	char *name;
+
+	cpumask_var_t affinity_mask;
 };
 
 struct idpf_rx_queue_stats {
-- 
cgit v1.2.3-58-ga151


From 5a816aae2d463d74882e21672ac5366573b0c511 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:40 +0200
Subject: idpf: strictly assert cachelines of queue and queue vector structures

Now that the queue and queue vector structures are separated and laid
out optimally, group the fields as read-mostly, read-write, and cold
cachelines and add size assertions to make sure new features won't push
something out of its place and provoke perf regression.
Despite looking innocent, this gives up to 2% of perf bump on Rx.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/idpf/idpf_txrx.h | 185 ++++++++++++++++++----------
 1 file changed, 118 insertions(+), 67 deletions(-)

diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
index 5daa8f905f86..c7ae20ab567b 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
@@ -6,6 +6,7 @@
 
 #include <linux/dim.h>
 
+#include <net/libeth/cache.h>
 #include <net/page_pool/helpers.h>
 #include <net/tcp.h>
 #include <net/netdev_queues.h>
@@ -505,58 +506,68 @@ struct idpf_intr_reg {
 /**
  * struct idpf_q_vector
  * @vport: Vport back pointer
- * @napi: napi handler
- * @v_idx: Vector index
- * @intr_reg: See struct idpf_intr_reg
+ * @num_rxq: Number of RX queues
  * @num_txq: Number of TX queues
+ * @num_bufq: Number of buffer queues
  * @num_complq: number of completion queues
+ * @rx: Array of RX queues to service
  * @tx: Array of TX queues to service
+ * @bufq: Array of buffer queues to service
  * @complq: array of completion queues
+ * @intr_reg: See struct idpf_intr_reg
+ * @napi: napi handler
+ * @total_events: Number of interrupts processed
  * @tx_dim: Data for TX net_dim algorithm
  * @tx_itr_value: TX interrupt throttling rate
  * @tx_intr_mode: Dynamic ITR or not
  * @tx_itr_idx: TX ITR index
- * @num_rxq: Number of RX queues
- * @rx: Array of RX queues to service
  * @rx_dim: Data for RX net_dim algorithm
  * @rx_itr_value: RX interrupt throttling rate
  * @rx_intr_mode: Dynamic ITR or not
  * @rx_itr_idx: RX ITR index
- * @num_bufq: Number of buffer queues
- * @bufq: Array of buffer queues to service
- * @total_events: Number of interrupts processed
+ * @v_idx: Vector index
  * @affinity_mask: CPU affinity mask
  */
 struct idpf_q_vector {
+	__cacheline_group_begin_aligned(read_mostly);
 	struct idpf_vport *vport;
-	struct napi_struct napi;
-	u16 v_idx;
-	struct idpf_intr_reg intr_reg;
 
+	u16 num_rxq;
 	u16 num_txq;
+	u16 num_bufq;
 	u16 num_complq;
+	struct idpf_rx_queue **rx;
 	struct idpf_tx_queue **tx;
+	struct idpf_buf_queue **bufq;
 	struct idpf_compl_queue **complq;
 
+	struct idpf_intr_reg intr_reg;
+	__cacheline_group_end_aligned(read_mostly);
+
+	__cacheline_group_begin_aligned(read_write);
+	struct napi_struct napi;
+	u16 total_events;
+
 	struct dim tx_dim;
 	u16 tx_itr_value;
 	bool tx_intr_mode;
 	u32 tx_itr_idx;
 
-	u16 num_rxq;
-	struct idpf_rx_queue **rx;
 	struct dim rx_dim;
 	u16 rx_itr_value;
 	bool rx_intr_mode;
 	u32 rx_itr_idx;
+	__cacheline_group_end_aligned(read_write);
 
-	u16 num_bufq;
-	struct idpf_buf_queue **bufq;
-
-	u16 total_events;
+	__cacheline_group_begin_aligned(cold);
+	u16 v_idx;
 
 	cpumask_var_t affinity_mask;
+	__cacheline_group_end_aligned(cold);
 };
+libeth_cacheline_set_assert(struct idpf_q_vector, 104,
+			    424 + 2 * sizeof(struct dim),
+			    8 + sizeof(cpumask_var_t));
 
 struct idpf_rx_queue_stats {
 	u64_stats_t packets;
@@ -623,11 +634,11 @@ struct idpf_txq_stash {
  * @idx: For RX queue, it is used to index to total RX queue across groups and
  *	 used for skb reporting.
  * @desc_count: Number of descriptors
+ * @rxdids: Supported RX descriptor ids
+ * @rx_ptype_lkup: LUT of Rx ptypes
  * @next_to_use: Next descriptor to use
  * @next_to_clean: Next descriptor to clean
  * @next_to_alloc: RX buffer to allocate at
- * @rxdids: Supported RX descriptor ids
- * @rx_ptype_lkup: LUT of Rx ptypes
  * @skb: Pointer to the skb
  * @stats_sync: See struct u64_stats_sync
  * @q_stats: See union idpf_rx_queue_stats
@@ -641,6 +652,7 @@ struct idpf_txq_stash {
  * @rx_max_pkt_size: RX max packet size
  */
 struct idpf_rx_queue {
+	__cacheline_group_begin_aligned(read_mostly);
 	union {
 		union virtchnl2_rx_desc *rx;
 		struct virtchnl2_singleq_rx_buf_desc *single_buf;
@@ -663,19 +675,23 @@ struct idpf_rx_queue {
 	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
 	u16 idx;
 	u16 desc_count;
+
+	u32 rxdids;
+	const struct idpf_rx_ptype_decoded *rx_ptype_lkup;
+	__cacheline_group_end_aligned(read_mostly);
+
+	__cacheline_group_begin_aligned(read_write);
 	u16 next_to_use;
 	u16 next_to_clean;
 	u16 next_to_alloc;
 
-	u32 rxdids;
-
-	const struct idpf_rx_ptype_decoded *rx_ptype_lkup;
 	struct sk_buff *skb;
 
 	struct u64_stats_sync stats_sync;
 	struct idpf_rx_queue_stats q_stats;
+	__cacheline_group_end_aligned(read_write);
 
-	/* Slowpath */
+	__cacheline_group_begin_aligned(cold);
 	u32 q_id;
 	u32 size;
 	dma_addr_t dma;
@@ -686,7 +702,11 @@ struct idpf_rx_queue {
 	u16 rx_hbuf_size;
 	u16 rx_buf_size;
 	u16 rx_max_pkt_size;
-} ____cacheline_aligned;
+	__cacheline_group_end_aligned(cold);
+};
+libeth_cacheline_set_assert(struct idpf_rx_queue, 64,
+			    72 + sizeof(struct u64_stats_sync),
+			    32);
 
 /**
  * struct idpf_tx_queue - software structure representing a transmit queue
@@ -703,22 +723,7 @@ struct idpf_rx_queue {
  * @idx: For TX queue, it is used as index to map between TX queue group and
  *	 hot path TX pointers stored in vport. Used in both singleq/splitq.
  * @desc_count: Number of descriptors
- * @next_to_use: Next descriptor to use
- * @next_to_clean: Next descriptor to clean
- * @netdev: &net_device corresponding to this queue
- * @cleaned_bytes: Splitq only, TXQ only: When a TX completion is received on
- *		   the TX completion queue, it can be for any TXQ associated
- *		   with that completion queue. This means we can clean up to
- *		   N TXQs during a single call to clean the completion queue.
- *		   cleaned_bytes|pkts tracks the clean stats per TXQ during
- *		   that single call to clean the completion queue. By doing so,
- *		   we can update BQL with aggregate cleaned stats for each TXQ
- *		   only once at the end of the cleaning routine.
- * @clean_budget: singleq only, queue cleaning budget
- * @cleaned_pkts: Number of packets cleaned for the above said case
- * @tx_max_bufs: Max buffers that can be transmitted with scatter-gather
  * @tx_min_pkt_len: Min supported packet length
- * @compl_tag_bufid_m: Completion tag buffer id mask
  * @compl_tag_gen_s: Completion tag generation bit
  *	The format of the completion tag will change based on the TXQ
  *	descriptor ring size so that we can maintain roughly the same level
@@ -739,9 +744,24 @@ struct idpf_rx_queue {
  *	--------------------------------
  *
  *	This gives us 8*8160 = 65280 possible unique values.
+ * @netdev: &net_device corresponding to this queue
+ * @next_to_use: Next descriptor to use
+ * @next_to_clean: Next descriptor to clean
+ * @cleaned_bytes: Splitq only, TXQ only: When a TX completion is received on
+ *		   the TX completion queue, it can be for any TXQ associated
+ *		   with that completion queue. This means we can clean up to
+ *		   N TXQs during a single call to clean the completion queue.
+ *		   cleaned_bytes|pkts tracks the clean stats per TXQ during
+ *		   that single call to clean the completion queue. By doing so,
+ *		   we can update BQL with aggregate cleaned stats for each TXQ
+ *		   only once at the end of the cleaning routine.
+ * @clean_budget: singleq only, queue cleaning budget
+ * @cleaned_pkts: Number of packets cleaned for the above said case
+ * @tx_max_bufs: Max buffers that can be transmitted with scatter-gather
+ * @stash: Tx buffer stash for Flow-based scheduling mode
+ * @compl_tag_bufid_m: Completion tag buffer id mask
  * @compl_tag_cur_gen: Used to keep track of current completion tag generation
  * @compl_tag_gen_max: To determine when compl_tag_cur_gen should be reset
- * @stash: Tx buffer stash for Flow-based scheduling mode
  * @stats_sync: See struct u64_stats_sync
  * @q_stats: See union idpf_tx_queue_stats
  * @q_id: Queue id
@@ -750,6 +770,7 @@ struct idpf_rx_queue {
  * @q_vector: Backreference to associated vector
  */
 struct idpf_tx_queue {
+	__cacheline_group_begin_aligned(read_mostly);
 	union {
 		struct idpf_base_tx_desc *base_tx;
 		struct idpf_base_tx_ctx_desc *base_ctx;
@@ -766,10 +787,16 @@ struct idpf_tx_queue {
 	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
 	u16 idx;
 	u16 desc_count;
-	u16 next_to_use;
-	u16 next_to_clean;
+
+	u16 tx_min_pkt_len;
+	u16 compl_tag_gen_s;
 
 	struct net_device *netdev;
+	__cacheline_group_end_aligned(read_mostly);
+
+	__cacheline_group_begin_aligned(read_write);
+	u16 next_to_use;
+	u16 next_to_clean;
 
 	union {
 		u32 cleaned_bytes;
@@ -778,26 +805,27 @@ struct idpf_tx_queue {
 	u16 cleaned_pkts;
 
 	u16 tx_max_bufs;
-	u16 tx_min_pkt_len;
+	struct idpf_txq_stash *stash;
 
 	u16 compl_tag_bufid_m;
-	u16 compl_tag_gen_s;
-
 	u16 compl_tag_cur_gen;
 	u16 compl_tag_gen_max;
 
-	struct idpf_txq_stash *stash;
-
 	struct u64_stats_sync stats_sync;
 	struct idpf_tx_queue_stats q_stats;
+	__cacheline_group_end_aligned(read_write);
 
-	/* Slowpath */
+	__cacheline_group_begin_aligned(cold);
 	u32 q_id;
 	u32 size;
 	dma_addr_t dma;
 
 	struct idpf_q_vector *q_vector;
-} ____cacheline_aligned;
+	__cacheline_group_end_aligned(cold);
+};
+libeth_cacheline_set_assert(struct idpf_tx_queue, 64,
+			    88 + sizeof(struct u64_stats_sync),
+			    24);
 
 /**
  * struct idpf_buf_queue - software structure representing a buffer queue
@@ -822,6 +850,7 @@ struct idpf_tx_queue {
  * @rx_buf_size: Buffer size
  */
 struct idpf_buf_queue {
+	__cacheline_group_begin_aligned(read_mostly);
 	struct virtchnl2_splitq_rx_buf_desc *split_buf;
 	struct {
 		struct idpf_rx_buf *buf;
@@ -832,12 +861,16 @@ struct idpf_buf_queue {
 	void __iomem *tail;
 
 	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
-	u16 desc_count;
-	u16 next_to_use;
-	u16 next_to_clean;
-	u16 next_to_alloc;
+	u32 desc_count;
+	__cacheline_group_end_aligned(read_mostly);
 
-	/* Slowpath */
+	__cacheline_group_begin_aligned(read_write);
+	u32 next_to_use;
+	u32 next_to_clean;
+	u32 next_to_alloc;
+	__cacheline_group_end_aligned(read_write);
+
+	__cacheline_group_begin_aligned(cold);
 	u32 q_id;
 	u32 size;
 	dma_addr_t dma;
@@ -847,7 +880,9 @@ struct idpf_buf_queue {
 	u16 rx_buffer_low_watermark;
 	u16 rx_hbuf_size;
 	u16 rx_buf_size;
-} ____cacheline_aligned;
+	__cacheline_group_end_aligned(cold);
+};
+libeth_cacheline_set_assert(struct idpf_buf_queue, 64, 16, 32);
 
 /**
  * struct idpf_compl_queue - software structure representing a completion queue
@@ -855,11 +890,11 @@ struct idpf_buf_queue {
  * @txq_grp: See struct idpf_txq_group
  * @flags: See enum idpf_queue_flags_t
  * @desc_count: Number of descriptors
+ * @clean_budget: queue cleaning budget
+ * @netdev: &net_device corresponding to this queue
  * @next_to_use: Next descriptor to use. Relevant in both split & single txq
  *		 and bufq.
  * @next_to_clean: Next descriptor to clean
- * @netdev: &net_device corresponding to this queue
- * @clean_budget: queue cleaning budget
  * @num_completions: Only relevant for TX completion queue. It tracks the
  *		     number of completions received to compare against the
  *		     number of completions pending, as accumulated by the
@@ -870,25 +905,33 @@ struct idpf_buf_queue {
  * @q_vector: Backreference to associated vector
  */
 struct idpf_compl_queue {
+	__cacheline_group_begin_aligned(read_mostly);
 	struct idpf_splitq_tx_compl_desc *comp;
 	struct idpf_txq_group *txq_grp;
 
 	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
-	u16 desc_count;
-	u16 next_to_use;
-	u16 next_to_clean;
+	u32 desc_count;
 
-	struct net_device *netdev;
 	u32 clean_budget;
+	struct net_device *netdev;
+	__cacheline_group_end_aligned(read_mostly);
+
+	__cacheline_group_begin_aligned(read_write);
+	u32 next_to_use;
+	u32 next_to_clean;
+
 	u32 num_completions;
+	__cacheline_group_end_aligned(read_write);
 
-	/* Slowpath */
+	__cacheline_group_begin_aligned(cold);
 	u32 q_id;
 	u32 size;
 	dma_addr_t dma;
 
 	struct idpf_q_vector *q_vector;
-} ____cacheline_aligned;
+	__cacheline_group_end_aligned(cold);
+};
+libeth_cacheline_set_assert(struct idpf_compl_queue, 40, 16, 24);
 
 /**
  * struct idpf_sw_queue
@@ -903,13 +946,21 @@ struct idpf_compl_queue {
  * lockless buffer management system and are strictly software only constructs.
  */
 struct idpf_sw_queue {
+	__cacheline_group_begin_aligned(read_mostly);
 	u32 *ring;
 
 	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
-	u16 desc_count;
-	u16 next_to_use;
-	u16 next_to_clean;
-} ____cacheline_aligned;
+	u32 desc_count;
+	__cacheline_group_end_aligned(read_mostly);
+
+	__cacheline_group_begin_aligned(read_write);
+	u32 next_to_use;
+	u32 next_to_clean;
+	__cacheline_group_end_aligned(read_write);
+};
+libeth_cacheline_group_assert(struct idpf_sw_queue, read_mostly, 24);
+libeth_cacheline_group_assert(struct idpf_sw_queue, read_write, 8);
+libeth_cacheline_struct_assert(struct idpf_sw_queue, 24, 8);
 
 /**
  * struct idpf_rxq_set
-- 
cgit v1.2.3-58-ga151


From 14f662b43bf8c765114f73d184af2702b2280436 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:41 +0200
Subject: idpf: merge singleq and splitq &net_device_ops

It makes no sense to have a second &net_device_ops struct (800 bytes of
rodata) with only one difference in .ndo_start_xmit, which can easily
be just one `if`. This `if` is a drop in the ocean and you won't see
any difference.
Define unified idpf_xmit_start(). The preparation for sending is the
same, just call either idpf_tx_splitq_frame() or idpf_tx_singleq_frame()
depending on the active model to actually map and send the skb.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/idpf/idpf_lib.c         | 26 +++---------------
 .../net/ethernet/intel/idpf/idpf_singleq_txrx.c    | 31 ++--------------------
 drivers/net/ethernet/intel/idpf/idpf_txrx.c        | 17 +++++++-----
 drivers/net/ethernet/intel/idpf/idpf_txrx.h        |  9 +++----
 4 files changed, 20 insertions(+), 63 deletions(-)

diff --git a/drivers/net/ethernet/intel/idpf/idpf_lib.c b/drivers/net/ethernet/intel/idpf/idpf_lib.c
index fafd279127a3..56fdbb2b6fbc 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_lib.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_lib.c
@@ -4,8 +4,7 @@
 #include "idpf.h"
 #include "idpf_virtchnl.h"
 
-static const struct net_device_ops idpf_netdev_ops_splitq;
-static const struct net_device_ops idpf_netdev_ops_singleq;
+static const struct net_device_ops idpf_netdev_ops;
 
 /**
  * idpf_init_vector_stack - Fill the MSIX vector stack with vector index
@@ -764,10 +763,7 @@ static int idpf_cfg_netdev(struct idpf_vport *vport)
 	}
 
 	/* assign netdev_ops */
-	if (idpf_is_queue_model_split(vport->txq_model))
-		netdev->netdev_ops = &idpf_netdev_ops_splitq;
-	else
-		netdev->netdev_ops = &idpf_netdev_ops_singleq;
+	netdev->netdev_ops = &idpf_netdev_ops;
 
 	/* setup watchdog timeout value to be 5 second */
 	netdev->watchdog_timeo = 5 * HZ;
@@ -2352,24 +2348,10 @@ void idpf_free_dma_mem(struct idpf_hw *hw, struct idpf_dma_mem *mem)
 	mem->pa = 0;
 }
 
-static const struct net_device_ops idpf_netdev_ops_splitq = {
-	.ndo_open = idpf_open,
-	.ndo_stop = idpf_stop,
-	.ndo_start_xmit = idpf_tx_splitq_start,
-	.ndo_features_check = idpf_features_check,
-	.ndo_set_rx_mode = idpf_set_rx_mode,
-	.ndo_validate_addr = eth_validate_addr,
-	.ndo_set_mac_address = idpf_set_mac,
-	.ndo_change_mtu = idpf_change_mtu,
-	.ndo_get_stats64 = idpf_get_stats64,
-	.ndo_set_features = idpf_set_features,
-	.ndo_tx_timeout = idpf_tx_timeout,
-};
-
-static const struct net_device_ops idpf_netdev_ops_singleq = {
+static const struct net_device_ops idpf_netdev_ops = {
 	.ndo_open = idpf_open,
 	.ndo_stop = idpf_stop,
-	.ndo_start_xmit = idpf_tx_singleq_start,
+	.ndo_start_xmit = idpf_tx_start,
 	.ndo_features_check = idpf_features_check,
 	.ndo_set_rx_mode = idpf_set_rx_mode,
 	.ndo_validate_addr = eth_validate_addr,
diff --git a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
index 9864a3992f0c..8630db24f63a 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
@@ -351,8 +351,8 @@ static void idpf_tx_singleq_build_ctx_desc(struct idpf_tx_queue *txq,
  *
  * Returns NETDEV_TX_OK if sent, else an error code
  */
-static netdev_tx_t idpf_tx_singleq_frame(struct sk_buff *skb,
-					 struct idpf_tx_queue *tx_q)
+netdev_tx_t idpf_tx_singleq_frame(struct sk_buff *skb,
+				  struct idpf_tx_queue *tx_q)
 {
 	struct idpf_tx_offload_params offload = { };
 	struct idpf_tx_buf *first;
@@ -408,33 +408,6 @@ out_drop:
 	return idpf_tx_drop_skb(tx_q, skb);
 }
 
-/**
- * idpf_tx_singleq_start - Selects the right Tx queue to send buffer
- * @skb: send buffer
- * @netdev: network interface device structure
- *
- * Returns NETDEV_TX_OK if sent, else an error code
- */
-netdev_tx_t idpf_tx_singleq_start(struct sk_buff *skb,
-				  struct net_device *netdev)
-{
-	struct idpf_vport *vport = idpf_netdev_to_vport(netdev);
-	struct idpf_tx_queue *tx_q;
-
-	tx_q = vport->txqs[skb_get_queue_mapping(skb)];
-
-	/* hardware can't handle really short frames, hardware padding works
-	 * beyond this point
-	 */
-	if (skb_put_padto(skb, IDPF_TX_MIN_PKT_LEN)) {
-		idpf_tx_buf_hw_update(tx_q, tx_q->next_to_use, false);
-
-		return NETDEV_TX_OK;
-	}
-
-	return idpf_tx_singleq_frame(skb, tx_q);
-}
-
 /**
  * idpf_tx_singleq_clean - Reclaim resources from queue
  * @tx_q: Tx queue to clean
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
index f569ea389b04..5dd1b1a9e624 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
@@ -4,6 +4,9 @@
 #include "idpf.h"
 #include "idpf_virtchnl.h"
 
+static bool idpf_chk_linearize(struct sk_buff *skb, unsigned int max_bufs,
+			       unsigned int count);
+
 /**
  * idpf_buf_lifo_push - push a buffer pointer onto stack
  * @stack: pointer to stack struct
@@ -2702,8 +2705,8 @@ static bool __idpf_chk_linearize(struct sk_buff *skb, unsigned int max_bufs)
  * E.g.: a packet with 7 fragments can require 9 DMA transactions; 1 for TSO
  * header, 1 for segment payload, and then 7 for the fragments.
  */
-bool idpf_chk_linearize(struct sk_buff *skb, unsigned int max_bufs,
-			unsigned int count)
+static bool idpf_chk_linearize(struct sk_buff *skb, unsigned int max_bufs,
+			       unsigned int count)
 {
 	if (likely(count < max_bufs))
 		return false;
@@ -2849,14 +2852,13 @@ static netdev_tx_t idpf_tx_splitq_frame(struct sk_buff *skb,
 }
 
 /**
- * idpf_tx_splitq_start - Selects the right Tx queue to send buffer
+ * idpf_tx_start - Selects the right Tx queue to send buffer
  * @skb: send buffer
  * @netdev: network interface device structure
  *
  * Returns NETDEV_TX_OK if sent, else an error code
  */
-netdev_tx_t idpf_tx_splitq_start(struct sk_buff *skb,
-				 struct net_device *netdev)
+netdev_tx_t idpf_tx_start(struct sk_buff *skb, struct net_device *netdev)
 {
 	struct idpf_vport *vport = idpf_netdev_to_vport(netdev);
 	struct idpf_tx_queue *tx_q;
@@ -2878,7 +2880,10 @@ netdev_tx_t idpf_tx_splitq_start(struct sk_buff *skb,
 		return NETDEV_TX_OK;
 	}
 
-	return idpf_tx_splitq_frame(skb, tx_q);
+	if (idpf_is_queue_model_split(vport->txq_model))
+		return idpf_tx_splitq_frame(skb, tx_q);
+	else
+		return idpf_tx_singleq_frame(skb, tx_q);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
index c7ae20ab567b..b2bf58146484 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
@@ -1198,14 +1198,11 @@ void idpf_tx_dma_map_error(struct idpf_tx_queue *txq, struct sk_buff *skb,
 			   struct idpf_tx_buf *first, u16 ring_idx);
 unsigned int idpf_tx_desc_count_required(struct idpf_tx_queue *txq,
 					 struct sk_buff *skb);
-bool idpf_chk_linearize(struct sk_buff *skb, unsigned int max_bufs,
-			unsigned int count);
 int idpf_tx_maybe_stop_common(struct idpf_tx_queue *tx_q, unsigned int size);
 void idpf_tx_timeout(struct net_device *netdev, unsigned int txqueue);
-netdev_tx_t idpf_tx_splitq_start(struct sk_buff *skb,
-				 struct net_device *netdev);
-netdev_tx_t idpf_tx_singleq_start(struct sk_buff *skb,
-				  struct net_device *netdev);
+netdev_tx_t idpf_tx_singleq_frame(struct sk_buff *skb,
+				  struct idpf_tx_queue *tx_q);
+netdev_tx_t idpf_tx_start(struct sk_buff *skb, struct net_device *netdev);
 bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_rx_queue *rxq,
 				      u16 cleaned_count);
 int idpf_tso(struct sk_buff *skb, struct idpf_tx_offload_params *off);
-- 
cgit v1.2.3-58-ga151


From f771314d6b75181de7079c3c7d666293e4ed2b22 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:42 +0200
Subject: idpf: compile singleq code only under default-n CONFIG_IDPF_SINGLEQ

Currently, all HW supporting idpf supports the singleq model, but none
of it advertises it by default, as splitq is supported and preferred
for multiple reasons. Still, this almost dead code often times adds
hotpath branches and redundant cacheline accesses.
While it can't currently be removed, add CONFIG_IDPF_SINGLEQ and build
the singleq code only when it's enabled manually. This corresponds to
-10 Kb of object code size and a good bunch of hotpath checks.
idpf_is_queue_model_split() works as a gate and compiles out to `true`
when the config option is disabled.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/Kconfig              | 13 +-----------
 drivers/net/ethernet/intel/idpf/Kconfig         | 27 +++++++++++++++++++++++++
 drivers/net/ethernet/intel/idpf/Makefile        |  3 ++-
 drivers/net/ethernet/intel/idpf/idpf.h          |  3 ++-
 drivers/net/ethernet/intel/idpf/idpf_txrx.c     |  2 +-
 drivers/net/ethernet/intel/idpf/idpf_virtchnl.c | 15 ++++++++++----
 6 files changed, 44 insertions(+), 19 deletions(-)
 create mode 100644 drivers/net/ethernet/intel/idpf/Kconfig

diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig
index e0287fbd501d..0375c7448a57 100644
--- a/drivers/net/ethernet/intel/Kconfig
+++ b/drivers/net/ethernet/intel/Kconfig
@@ -384,17 +384,6 @@ config IGC_LEDS
 	  Optional support for controlling the NIC LED's with the netdev
 	  LED trigger.
 
-config IDPF
-	tristate "Intel(R) Infrastructure Data Path Function Support"
-	depends on PCI_MSI
-	select DIMLIB
-	select PAGE_POOL
-	select PAGE_POOL_STATS
-	help
-	  This driver supports Intel(R) Infrastructure Data Path Function
-	  devices.
-
-	  To compile this driver as a module, choose M here. The module
-	  will be called idpf.
+source "drivers/net/ethernet/intel/idpf/Kconfig"
 
 endif # NET_VENDOR_INTEL
diff --git a/drivers/net/ethernet/intel/idpf/Kconfig b/drivers/net/ethernet/intel/idpf/Kconfig
new file mode 100644
index 000000000000..9082c16edb7e
--- /dev/null
+++ b/drivers/net/ethernet/intel/idpf/Kconfig
@@ -0,0 +1,27 @@
+# SPDX-License-Identifier: GPL-2.0-only
+# Copyright (C) 2024 Intel Corporation
+
+config IDPF
+	tristate "Intel(R) Infrastructure Data Path Function Support"
+	depends on PCI_MSI
+	select DIMLIB
+	select PAGE_POOL
+	select PAGE_POOL_STATS
+	help
+	  This driver supports Intel(R) Infrastructure Data Path Function
+	  devices.
+
+	  To compile this driver as a module, choose M here. The module
+	  will be called idpf.
+
+if IDPF
+
+config IDPF_SINGLEQ
+	bool "idpf singleq support"
+	help
+	  This option enables support for legacy single Rx/Tx queues w/no
+	  completion and fill queues. Only enable if you have hardware which
+	  wants to work in this mode as it increases the driver size and adds
+	  runtme checks on hotpath.
+
+endif # IDPF
diff --git a/drivers/net/ethernet/intel/idpf/Makefile b/drivers/net/ethernet/intel/idpf/Makefile
index 6844ead2f3ac..2ce01a0b5898 100644
--- a/drivers/net/ethernet/intel/idpf/Makefile
+++ b/drivers/net/ethernet/intel/idpf/Makefile
@@ -12,7 +12,8 @@ idpf-y := \
 	idpf_ethtool.o		\
 	idpf_lib.o		\
 	idpf_main.o		\
-	idpf_singleq_txrx.o	\
 	idpf_txrx.o		\
 	idpf_virtchnl.o 	\
 	idpf_vf_dev.o
+
+idpf-$(CONFIG_IDPF_SINGLEQ)	+= idpf_singleq_txrx.o
diff --git a/drivers/net/ethernet/intel/idpf/idpf.h b/drivers/net/ethernet/intel/idpf/idpf.h
index f9e43d171f17..5d9529f5b41b 100644
--- a/drivers/net/ethernet/intel/idpf/idpf.h
+++ b/drivers/net/ethernet/intel/idpf/idpf.h
@@ -599,7 +599,8 @@ struct idpf_adapter {
  */
 static inline int idpf_is_queue_model_split(u16 q_model)
 {
-	return q_model == VIRTCHNL2_QUEUE_MODEL_SPLIT;
+	return !IS_ENABLED(CONFIG_IDPF_SINGLEQ) ||
+	       q_model == VIRTCHNL2_QUEUE_MODEL_SPLIT;
 }
 
 #define idpf_is_cap_ena(adapter, field, flag) \
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
index 5dd1b1a9e624..e79a8f9dfc40 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
@@ -1309,7 +1309,7 @@ static void idpf_vport_calc_numq_per_grp(struct idpf_vport *vport,
 static void idpf_rxq_set_descids(const struct idpf_vport *vport,
 				 struct idpf_rx_queue *q)
 {
-	if (vport->rxq_model == VIRTCHNL2_QUEUE_MODEL_SPLIT) {
+	if (idpf_is_queue_model_split(vport->rxq_model)) {
 		q->rxdids = VIRTCHNL2_RXDID_2_FLEX_SPLITQ_M;
 	} else {
 		if (vport->base_rxd)
diff --git a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
index 44602b87cd41..1aa4770dfe18 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
@@ -1256,12 +1256,12 @@ int idpf_send_create_vport_msg(struct idpf_adapter *adapter,
 	vport_msg->vport_type = cpu_to_le16(VIRTCHNL2_VPORT_TYPE_DEFAULT);
 	vport_msg->vport_index = cpu_to_le16(idx);
 
-	if (adapter->req_tx_splitq)
+	if (adapter->req_tx_splitq || !IS_ENABLED(CONFIG_IDPF_SINGLEQ))
 		vport_msg->txq_model = cpu_to_le16(VIRTCHNL2_QUEUE_MODEL_SPLIT);
 	else
 		vport_msg->txq_model = cpu_to_le16(VIRTCHNL2_QUEUE_MODEL_SINGLE);
 
-	if (adapter->req_rx_splitq)
+	if (adapter->req_rx_splitq || !IS_ENABLED(CONFIG_IDPF_SINGLEQ))
 		vport_msg->rxq_model = cpu_to_le16(VIRTCHNL2_QUEUE_MODEL_SPLIT);
 	else
 		vport_msg->rxq_model = cpu_to_le16(VIRTCHNL2_QUEUE_MODEL_SINGLE);
@@ -1323,10 +1323,17 @@ int idpf_check_supported_desc_ids(struct idpf_vport *vport)
 
 	vport_msg = adapter->vport_params_recvd[vport->idx];
 
+	if (!IS_ENABLED(CONFIG_IDPF_SINGLEQ) &&
+	    (vport_msg->rxq_model == VIRTCHNL2_QUEUE_MODEL_SINGLE ||
+	     vport_msg->txq_model == VIRTCHNL2_QUEUE_MODEL_SINGLE)) {
+		pci_err(adapter->pdev, "singleq mode requested, but not compiled-in\n");
+		return -EOPNOTSUPP;
+	}
+
 	rx_desc_ids = le64_to_cpu(vport_msg->rx_desc_ids);
 	tx_desc_ids = le64_to_cpu(vport_msg->tx_desc_ids);
 
-	if (vport->rxq_model == VIRTCHNL2_QUEUE_MODEL_SPLIT) {
+	if (idpf_is_queue_model_split(vport->rxq_model)) {
 		if (!(rx_desc_ids & VIRTCHNL2_RXDID_2_FLEX_SPLITQ_M)) {
 			dev_info(&adapter->pdev->dev, "Minimum RX descriptor support not provided, using the default\n");
 			vport_msg->rx_desc_ids = cpu_to_le64(VIRTCHNL2_RXDID_2_FLEX_SPLITQ_M);
@@ -1336,7 +1343,7 @@ int idpf_check_supported_desc_ids(struct idpf_vport *vport)
 			vport->base_rxd = true;
 	}
 
-	if (vport->txq_model != VIRTCHNL2_QUEUE_MODEL_SPLIT)
+	if (!idpf_is_queue_model_split(vport->txq_model))
 		return 0;
 
 	if ((tx_desc_ids & MIN_SUPPORT_TXDID) != MIN_SUPPORT_TXDID) {
-- 
cgit v1.2.3-58-ga151


From 1b1b26208515482d8bc3d875267076f61f3355f7 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:43 +0200
Subject: idpf: reuse libeth's definitions of parsed ptype structures

idpf's in-kernel parsed ptype structure is almost identical to the one
used in the previous Intel drivers, which means it can be converted to
use libeth's definitions and even helpers. The only difference is that
it doesn't use a constant table (libie), rather than one obtained from
the device.
Remove the driver counterpart and use libeth's helpers for hashes and
checksums. This slightly optimizes skb fields processing due to faster
checks. Also don't define big static array of ptypes in &idpf_vport --
allocate them dynamically. The pointer to it is anyway cached in
&idpf_rx_queue.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/idpf/Kconfig            |   1 +
 drivers/net/ethernet/intel/idpf/idpf.h             |   2 +-
 drivers/net/ethernet/intel/idpf/idpf_lib.c         |   3 +
 drivers/net/ethernet/intel/idpf/idpf_main.c        |   1 +
 .../net/ethernet/intel/idpf/idpf_singleq_txrx.c    | 113 ++++++++-----------
 drivers/net/ethernet/intel/idpf/idpf_txrx.c        | 125 ++++++++-------------
 drivers/net/ethernet/intel/idpf/idpf_txrx.h        |  88 +--------------
 drivers/net/ethernet/intel/idpf/idpf_virtchnl.c    |  72 +++++++-----
 8 files changed, 153 insertions(+), 252 deletions(-)

diff --git a/drivers/net/ethernet/intel/idpf/Kconfig b/drivers/net/ethernet/intel/idpf/Kconfig
index 9082c16edb7e..638484c5723c 100644
--- a/drivers/net/ethernet/intel/idpf/Kconfig
+++ b/drivers/net/ethernet/intel/idpf/Kconfig
@@ -5,6 +5,7 @@ config IDPF
 	tristate "Intel(R) Infrastructure Data Path Function Support"
 	depends on PCI_MSI
 	select DIMLIB
+	select LIBETH
 	select PAGE_POOL
 	select PAGE_POOL_STATS
 	help
diff --git a/drivers/net/ethernet/intel/idpf/idpf.h b/drivers/net/ethernet/intel/idpf/idpf.h
index 5d9529f5b41b..078340a01757 100644
--- a/drivers/net/ethernet/intel/idpf/idpf.h
+++ b/drivers/net/ethernet/intel/idpf/idpf.h
@@ -312,7 +312,7 @@ struct idpf_vport {
 	u16 num_rxq_grp;
 	struct idpf_rxq_group *rxq_grps;
 	u32 rxq_model;
-	struct idpf_rx_ptype_decoded rx_ptype_lkup[IDPF_RX_MAX_PTYPE];
+	struct libeth_rx_pt *rx_ptype_lkup;
 
 	struct idpf_adapter *adapter;
 	struct net_device *netdev;
diff --git a/drivers/net/ethernet/intel/idpf/idpf_lib.c b/drivers/net/ethernet/intel/idpf/idpf_lib.c
index 56fdbb2b6fbc..5dbf2b4ba1b0 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_lib.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_lib.c
@@ -941,6 +941,9 @@ static void idpf_decfg_netdev(struct idpf_vport *vport)
 {
 	struct idpf_adapter *adapter = vport->adapter;
 
+	kfree(vport->rx_ptype_lkup);
+	vport->rx_ptype_lkup = NULL;
+
 	unregister_netdev(vport->netdev);
 	free_netdev(vport->netdev);
 	vport->netdev = NULL;
diff --git a/drivers/net/ethernet/intel/idpf/idpf_main.c b/drivers/net/ethernet/intel/idpf/idpf_main.c
index f784eea044bd..db476b3314c8 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_main.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_main.c
@@ -8,6 +8,7 @@
 #define DRV_SUMMARY	"Intel(R) Infrastructure Data Path Function Linux Driver"
 
 MODULE_DESCRIPTION(DRV_SUMMARY);
+MODULE_IMPORT_NS(LIBETH);
 MODULE_LICENSE("GPL");
 
 /**
diff --git a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
index 8630db24f63a..6c0794e01922 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
@@ -1,6 +1,8 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /* Copyright (C) 2023 Intel Corporation */
 
+#include <net/libeth/rx.h>
+
 #include "idpf.h"
 
 /**
@@ -601,75 +603,62 @@ static bool idpf_rx_singleq_is_non_eop(const union virtchnl2_rx_desc *rx_desc)
  * @rxq: Rx ring being processed
  * @skb: skb currently being received and modified
  * @csum_bits: checksum bits from descriptor
- * @ptype: the packet type decoded by hardware
+ * @decoded: the packet type decoded by hardware
  *
  * skb->protocol must be set before this function is called
  */
-static void idpf_rx_singleq_csum(struct idpf_rx_queue *rxq, struct sk_buff *skb,
-				 struct idpf_rx_csum_decoded *csum_bits,
-				 u16 ptype)
+static void idpf_rx_singleq_csum(struct idpf_rx_queue *rxq,
+				 struct sk_buff *skb,
+				 struct idpf_rx_csum_decoded csum_bits,
+				 struct libeth_rx_pt decoded)
 {
-	struct idpf_rx_ptype_decoded decoded;
 	bool ipv4, ipv6;
 
 	/* check if Rx checksum is enabled */
-	if (unlikely(!(rxq->netdev->features & NETIF_F_RXCSUM)))
+	if (!libeth_rx_pt_has_checksum(rxq->netdev, decoded))
 		return;
 
 	/* check if HW has decoded the packet and checksum */
-	if (unlikely(!(csum_bits->l3l4p)))
-		return;
-
-	decoded = rxq->rx_ptype_lkup[ptype];
-	if (unlikely(!(decoded.known && decoded.outer_ip)))
+	if (unlikely(!csum_bits.l3l4p))
 		return;
 
-	ipv4 = IDPF_RX_PTYPE_TO_IPV(&decoded, IDPF_RX_PTYPE_OUTER_IPV4);
-	ipv6 = IDPF_RX_PTYPE_TO_IPV(&decoded, IDPF_RX_PTYPE_OUTER_IPV6);
+	ipv4 = libeth_rx_pt_get_ip_ver(decoded) == LIBETH_RX_PT_OUTER_IPV4;
+	ipv6 = libeth_rx_pt_get_ip_ver(decoded) == LIBETH_RX_PT_OUTER_IPV6;
 
 	/* Check if there were any checksum errors */
-	if (unlikely(ipv4 && (csum_bits->ipe || csum_bits->eipe)))
+	if (unlikely(ipv4 && (csum_bits.ipe || csum_bits.eipe)))
 		goto checksum_fail;
 
 	/* Device could not do any checksum offload for certain extension
 	 * headers as indicated by setting IPV6EXADD bit
 	 */
-	if (unlikely(ipv6 && csum_bits->ipv6exadd))
+	if (unlikely(ipv6 && csum_bits.ipv6exadd))
 		return;
 
 	/* check for L4 errors and handle packets that were not able to be
 	 * checksummed due to arrival speed
 	 */
-	if (unlikely(csum_bits->l4e))
+	if (unlikely(csum_bits.l4e))
 		goto checksum_fail;
 
-	if (unlikely(csum_bits->nat && csum_bits->eudpe))
+	if (unlikely(csum_bits.nat && csum_bits.eudpe))
 		goto checksum_fail;
 
 	/* Handle packets that were not able to be checksummed due to arrival
 	 * speed, in this case the stack can compute the csum.
 	 */
-	if (unlikely(csum_bits->pprs))
+	if (unlikely(csum_bits.pprs))
 		return;
 
 	/* If there is an outer header present that might contain a checksum
 	 * we need to bump the checksum level by 1 to reflect the fact that
 	 * we are indicating we validated the inner checksum.
 	 */
-	if (decoded.tunnel_type >= IDPF_RX_PTYPE_TUNNEL_IP_GRENAT)
+	if (decoded.tunnel_type >= LIBETH_RX_PT_TUNNEL_IP_GRENAT)
 		skb->csum_level = 1;
 
-	/* Only report checksum unnecessary for ICMP, TCP, UDP, or SCTP */
-	switch (decoded.inner_prot) {
-	case IDPF_RX_PTYPE_INNER_PROT_ICMP:
-	case IDPF_RX_PTYPE_INNER_PROT_TCP:
-	case IDPF_RX_PTYPE_INNER_PROT_UDP:
-	case IDPF_RX_PTYPE_INNER_PROT_SCTP:
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
-		return;
-	default:
-		return;
-	}
+	skb->ip_summed = CHECKSUM_UNNECESSARY;
+	return;
 
 checksum_fail:
 	u64_stats_update_begin(&rxq->stats_sync);
@@ -679,20 +668,17 @@ checksum_fail:
 
 /**
  * idpf_rx_singleq_base_csum - Indicate in skb if hw indicated a good cksum
- * @rx_q: Rx completion queue
- * @skb: skb currently being received and modified
  * @rx_desc: the receive descriptor
- * @ptype: Rx packet type
  *
  * This function only operates on the VIRTCHNL2_RXDID_1_32B_BASE_M base 32byte
  * descriptor writeback format.
+ *
+ * Return: parsed checksum status.
  **/
-static void idpf_rx_singleq_base_csum(struct idpf_rx_queue *rx_q,
-				      struct sk_buff *skb,
-				      const union virtchnl2_rx_desc *rx_desc,
-				      u16 ptype)
+static struct idpf_rx_csum_decoded
+idpf_rx_singleq_base_csum(const union virtchnl2_rx_desc *rx_desc)
 {
-	struct idpf_rx_csum_decoded csum_bits;
+	struct idpf_rx_csum_decoded csum_bits = { };
 	u32 rx_error, rx_status;
 	u64 qword;
 
@@ -711,28 +697,23 @@ static void idpf_rx_singleq_base_csum(struct idpf_rx_queue *rx_q,
 				    rx_status);
 	csum_bits.ipv6exadd = FIELD_GET(VIRTCHNL2_RX_BASE_DESC_STATUS_IPV6EXADD_M,
 					rx_status);
-	csum_bits.nat = 0;
-	csum_bits.eudpe = 0;
 
-	idpf_rx_singleq_csum(rx_q, skb, &csum_bits, ptype);
+	return csum_bits;
 }
 
 /**
  * idpf_rx_singleq_flex_csum - Indicate in skb if hw indicated a good cksum
- * @rx_q: Rx completion queue
- * @skb: skb currently being received and modified
  * @rx_desc: the receive descriptor
- * @ptype: Rx packet type
  *
  * This function only operates on the VIRTCHNL2_RXDID_2_FLEX_SQ_NIC flexible
  * descriptor writeback format.
+ *
+ * Return: parsed checksum status.
  **/
-static void idpf_rx_singleq_flex_csum(struct idpf_rx_queue *rx_q,
-				      struct sk_buff *skb,
-				      const union virtchnl2_rx_desc *rx_desc,
-				      u16 ptype)
+static struct idpf_rx_csum_decoded
+idpf_rx_singleq_flex_csum(const union virtchnl2_rx_desc *rx_desc)
 {
-	struct idpf_rx_csum_decoded csum_bits;
+	struct idpf_rx_csum_decoded csum_bits = { };
 	u16 rx_status0, rx_status1;
 
 	rx_status0 = le16_to_cpu(rx_desc->flex_nic_wb.status_error0);
@@ -752,9 +733,8 @@ static void idpf_rx_singleq_flex_csum(struct idpf_rx_queue *rx_q,
 					rx_status0);
 	csum_bits.nat = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_STATUS1_NAT_M,
 				  rx_status1);
-	csum_bits.pprs = 0;
 
-	idpf_rx_singleq_csum(rx_q, skb, &csum_bits, ptype);
+	return csum_bits;
 }
 
 /**
@@ -770,11 +750,11 @@ static void idpf_rx_singleq_flex_csum(struct idpf_rx_queue *rx_q,
 static void idpf_rx_singleq_base_hash(struct idpf_rx_queue *rx_q,
 				      struct sk_buff *skb,
 				      const union virtchnl2_rx_desc *rx_desc,
-				      struct idpf_rx_ptype_decoded *decoded)
+				      struct libeth_rx_pt decoded)
 {
 	u64 mask, qw1;
 
-	if (unlikely(!(rx_q->netdev->features & NETIF_F_RXHASH)))
+	if (!libeth_rx_pt_has_hash(rx_q->netdev, decoded))
 		return;
 
 	mask = VIRTCHNL2_RX_BASE_DESC_FLTSTAT_RSS_HASH_M;
@@ -783,7 +763,7 @@ static void idpf_rx_singleq_base_hash(struct idpf_rx_queue *rx_q,
 	if (FIELD_GET(mask, qw1) == mask) {
 		u32 hash = le32_to_cpu(rx_desc->base_wb.qword0.hi_dword.rss);
 
-		skb_set_hash(skb, hash, idpf_ptype_to_htype(decoded));
+		libeth_rx_pt_set_hash(skb, hash, decoded);
 	}
 }
 
@@ -800,15 +780,17 @@ static void idpf_rx_singleq_base_hash(struct idpf_rx_queue *rx_q,
 static void idpf_rx_singleq_flex_hash(struct idpf_rx_queue *rx_q,
 				      struct sk_buff *skb,
 				      const union virtchnl2_rx_desc *rx_desc,
-				      struct idpf_rx_ptype_decoded *decoded)
+				      struct libeth_rx_pt decoded)
 {
-	if (unlikely(!(rx_q->netdev->features & NETIF_F_RXHASH)))
+	if (!libeth_rx_pt_has_hash(rx_q->netdev, decoded))
 		return;
 
 	if (FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_STATUS0_RSS_VALID_M,
-		      le16_to_cpu(rx_desc->flex_nic_wb.status_error0)))
-		skb_set_hash(skb, le32_to_cpu(rx_desc->flex_nic_wb.rss_hash),
-			     idpf_ptype_to_htype(decoded));
+		      le16_to_cpu(rx_desc->flex_nic_wb.status_error0))) {
+		u32 hash = le32_to_cpu(rx_desc->flex_nic_wb.rss_hash);
+
+		libeth_rx_pt_set_hash(skb, hash, decoded);
+	}
 }
 
 /**
@@ -829,19 +811,22 @@ idpf_rx_singleq_process_skb_fields(struct idpf_rx_queue *rx_q,
 				   const union virtchnl2_rx_desc *rx_desc,
 				   u16 ptype)
 {
-	struct idpf_rx_ptype_decoded decoded = rx_q->rx_ptype_lkup[ptype];
+	struct libeth_rx_pt decoded = rx_q->rx_ptype_lkup[ptype];
+	struct idpf_rx_csum_decoded csum_bits;
 
 	/* modifies the skb - consumes the enet header */
 	skb->protocol = eth_type_trans(skb, rx_q->netdev);
 
 	/* Check if we're using base mode descriptor IDs */
 	if (rx_q->rxdids == VIRTCHNL2_RXDID_1_32B_BASE_M) {
-		idpf_rx_singleq_base_hash(rx_q, skb, rx_desc, &decoded);
-		idpf_rx_singleq_base_csum(rx_q, skb, rx_desc, ptype);
+		idpf_rx_singleq_base_hash(rx_q, skb, rx_desc, decoded);
+		csum_bits = idpf_rx_singleq_base_csum(rx_desc);
 	} else {
-		idpf_rx_singleq_flex_hash(rx_q, skb, rx_desc, &decoded);
-		idpf_rx_singleq_flex_csum(rx_q, skb, rx_desc, ptype);
+		idpf_rx_singleq_flex_hash(rx_q, skb, rx_desc, decoded);
+		csum_bits = idpf_rx_singleq_flex_csum(rx_desc);
 	}
+
+	idpf_rx_singleq_csum(rx_q, skb, csum_bits, decoded);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
index e79a8f9dfc40..883e58815f92 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
@@ -1,6 +1,8 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /* Copyright (C) 2023 Intel Corporation */
 
+#include <net/libeth/rx.h>
+
 #include "idpf.h"
 #include "idpf_virtchnl.h"
 
@@ -2886,30 +2888,6 @@ netdev_tx_t idpf_tx_start(struct sk_buff *skb, struct net_device *netdev)
 		return idpf_tx_singleq_frame(skb, tx_q);
 }
 
-/**
- * idpf_ptype_to_htype - get a hash type
- * @decoded: Decoded Rx packet type related fields
- *
- * Returns appropriate hash type (such as PKT_HASH_TYPE_L2/L3/L4) to be used by
- * skb_set_hash based on PTYPE as parsed by HW Rx pipeline and is part of
- * Rx desc.
- */
-enum pkt_hash_types idpf_ptype_to_htype(const struct idpf_rx_ptype_decoded *decoded)
-{
-	if (!decoded->known)
-		return PKT_HASH_TYPE_NONE;
-	if (decoded->payload_layer == IDPF_RX_PTYPE_PAYLOAD_LAYER_PAY2 &&
-	    decoded->inner_prot)
-		return PKT_HASH_TYPE_L4;
-	if (decoded->payload_layer == IDPF_RX_PTYPE_PAYLOAD_LAYER_PAY2 &&
-	    decoded->outer_ip)
-		return PKT_HASH_TYPE_L3;
-	if (decoded->outer_ip == IDPF_RX_PTYPE_OUTER_L2)
-		return PKT_HASH_TYPE_L2;
-
-	return PKT_HASH_TYPE_NONE;
-}
-
 /**
  * idpf_rx_hash - set the hash value in the skb
  * @rxq: Rx descriptor ring packet is being transacted on
@@ -2920,18 +2898,18 @@ enum pkt_hash_types idpf_ptype_to_htype(const struct idpf_rx_ptype_decoded *deco
 static void
 idpf_rx_hash(const struct idpf_rx_queue *rxq, struct sk_buff *skb,
 	     const struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc,
-	     struct idpf_rx_ptype_decoded *decoded)
+	     struct libeth_rx_pt decoded)
 {
 	u32 hash;
 
-	if (unlikely(!(rxq->netdev->features & NETIF_F_RXHASH)))
+	if (!libeth_rx_pt_has_hash(rxq->netdev, decoded))
 		return;
 
 	hash = le16_to_cpu(rx_desc->hash1) |
 	       (rx_desc->ff2_mirrid_hash2.hash2 << 16) |
 	       (rx_desc->hash3 << 24);
 
-	skb_set_hash(skb, hash, idpf_ptype_to_htype(decoded));
+	libeth_rx_pt_set_hash(skb, hash, decoded);
 }
 
 /**
@@ -2944,55 +2922,43 @@ idpf_rx_hash(const struct idpf_rx_queue *rxq, struct sk_buff *skb,
  * skb->protocol must be set before this function is called
  */
 static void idpf_rx_csum(struct idpf_rx_queue *rxq, struct sk_buff *skb,
-			 struct idpf_rx_csum_decoded *csum_bits,
-			 struct idpf_rx_ptype_decoded *decoded)
+			 struct idpf_rx_csum_decoded csum_bits,
+			 struct libeth_rx_pt decoded)
 {
 	bool ipv4, ipv6;
 
 	/* check if Rx checksum is enabled */
-	if (unlikely(!(rxq->netdev->features & NETIF_F_RXCSUM)))
+	if (!libeth_rx_pt_has_checksum(rxq->netdev, decoded))
 		return;
 
 	/* check if HW has decoded the packet and checksum */
-	if (!(csum_bits->l3l4p))
+	if (unlikely(!csum_bits.l3l4p))
 		return;
 
-	ipv4 = IDPF_RX_PTYPE_TO_IPV(decoded, IDPF_RX_PTYPE_OUTER_IPV4);
-	ipv6 = IDPF_RX_PTYPE_TO_IPV(decoded, IDPF_RX_PTYPE_OUTER_IPV6);
+	ipv4 = libeth_rx_pt_get_ip_ver(decoded) == LIBETH_RX_PT_OUTER_IPV4;
+	ipv6 = libeth_rx_pt_get_ip_ver(decoded) == LIBETH_RX_PT_OUTER_IPV6;
 
-	if (ipv4 && (csum_bits->ipe || csum_bits->eipe))
+	if (unlikely(ipv4 && (csum_bits.ipe || csum_bits.eipe)))
 		goto checksum_fail;
 
-	if (ipv6 && csum_bits->ipv6exadd)
+	if (unlikely(ipv6 && csum_bits.ipv6exadd))
 		return;
 
 	/* check for L4 errors and handle packets that were not able to be
 	 * checksummed
 	 */
-	if (csum_bits->l4e)
+	if (unlikely(csum_bits.l4e))
 		goto checksum_fail;
 
-	/* Only report checksum unnecessary for ICMP, TCP, UDP, or SCTP */
-	switch (decoded->inner_prot) {
-	case IDPF_RX_PTYPE_INNER_PROT_ICMP:
-	case IDPF_RX_PTYPE_INNER_PROT_TCP:
-	case IDPF_RX_PTYPE_INNER_PROT_UDP:
-		if (!csum_bits->raw_csum_inv) {
-			u16 csum = csum_bits->raw_csum;
-
-			skb->csum = csum_unfold((__force __sum16)~swab16(csum));
-			skb->ip_summed = CHECKSUM_COMPLETE;
-		} else {
-			skb->ip_summed = CHECKSUM_UNNECESSARY;
-		}
-		break;
-	case IDPF_RX_PTYPE_INNER_PROT_SCTP:
+	if (csum_bits.raw_csum_inv ||
+	    decoded.inner_prot == LIBETH_RX_PT_INNER_SCTP) {
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
-		break;
-	default:
-		break;
+		return;
 	}
 
+	skb->csum = csum_unfold((__force __sum16)~swab16(csum_bits.raw_csum));
+	skb->ip_summed = CHECKSUM_COMPLETE;
+
 	return;
 
 checksum_fail:
@@ -3004,32 +2970,34 @@ checksum_fail:
 /**
  * idpf_rx_splitq_extract_csum_bits - Extract checksum bits from descriptor
  * @rx_desc: receive descriptor
- * @csum: structure to extract checksum fields
  *
+ * Return: parsed checksum status.
  **/
-static void
-idpf_rx_splitq_extract_csum_bits(const struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc,
-				 struct idpf_rx_csum_decoded *csum)
+static struct idpf_rx_csum_decoded
+idpf_rx_splitq_extract_csum_bits(const struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc)
 {
+	struct idpf_rx_csum_decoded csum = { };
 	u8 qword0, qword1;
 
 	qword0 = rx_desc->status_err0_qw0;
 	qword1 = rx_desc->status_err0_qw1;
 
-	csum->ipe = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_XSUM_IPE_M,
+	csum.ipe = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_XSUM_IPE_M,
+			     qword1);
+	csum.eipe = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_XSUM_EIPE_M,
 			      qword1);
-	csum->eipe = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_XSUM_EIPE_M,
+	csum.l4e = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_XSUM_L4E_M,
+			     qword1);
+	csum.l3l4p = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_L3L4P_M,
 			       qword1);
-	csum->l4e = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_XSUM_L4E_M,
-			      qword1);
-	csum->l3l4p = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_L3L4P_M,
-				qword1);
-	csum->ipv6exadd = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_IPV6EXADD_M,
-				    qword0);
-	csum->raw_csum_inv =
+	csum.ipv6exadd = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_IPV6EXADD_M,
+				   qword0);
+	csum.raw_csum_inv =
 		le16_get_bits(rx_desc->ptype_err_fflags0,
 			      VIRTCHNL2_RX_FLEX_DESC_ADV_RAW_CSUM_INV_M);
-	csum->raw_csum = le16_to_cpu(rx_desc->misc.raw_cs);
+	csum.raw_csum = le16_to_cpu(rx_desc->misc.raw_cs);
+
+	return csum;
 }
 
 /**
@@ -3046,21 +3014,22 @@ idpf_rx_splitq_extract_csum_bits(const struct virtchnl2_rx_flex_desc_adv_nic_3 *
  */
 static int idpf_rx_rsc(struct idpf_rx_queue *rxq, struct sk_buff *skb,
 		       const struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc,
-		       struct idpf_rx_ptype_decoded *decoded)
+		       struct libeth_rx_pt decoded)
 {
 	u16 rsc_segments, rsc_seg_len;
 	bool ipv4, ipv6;
 	int len;
 
-	if (unlikely(!decoded->outer_ip))
+	if (unlikely(libeth_rx_pt_get_ip_ver(decoded) ==
+		     LIBETH_RX_PT_OUTER_L2))
 		return -EINVAL;
 
 	rsc_seg_len = le16_to_cpu(rx_desc->misc.rscseglen);
 	if (unlikely(!rsc_seg_len))
 		return -EINVAL;
 
-	ipv4 = IDPF_RX_PTYPE_TO_IPV(decoded, IDPF_RX_PTYPE_OUTER_IPV4);
-	ipv6 = IDPF_RX_PTYPE_TO_IPV(decoded, IDPF_RX_PTYPE_OUTER_IPV6);
+	ipv4 = libeth_rx_pt_get_ip_ver(decoded) == LIBETH_RX_PT_OUTER_IPV4;
+	ipv6 = libeth_rx_pt_get_ip_ver(decoded) == LIBETH_RX_PT_OUTER_IPV6;
 
 	if (unlikely(!(ipv4 ^ ipv6)))
 		return -EINVAL;
@@ -3118,8 +3087,8 @@ static int
 idpf_rx_process_skb_fields(struct idpf_rx_queue *rxq, struct sk_buff *skb,
 			   const struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc)
 {
-	struct idpf_rx_csum_decoded csum_bits = { };
-	struct idpf_rx_ptype_decoded decoded;
+	struct idpf_rx_csum_decoded csum_bits;
+	struct libeth_rx_pt decoded;
 	u16 rx_ptype;
 
 	rx_ptype = le16_get_bits(rx_desc->ptype_err_fflags0,
@@ -3127,16 +3096,16 @@ idpf_rx_process_skb_fields(struct idpf_rx_queue *rxq, struct sk_buff *skb,
 	decoded = rxq->rx_ptype_lkup[rx_ptype];
 
 	/* process RSS/hash */
-	idpf_rx_hash(rxq, skb, rx_desc, &decoded);
+	idpf_rx_hash(rxq, skb, rx_desc, decoded);
 
 	skb->protocol = eth_type_trans(skb, rxq->netdev);
 
 	if (le16_get_bits(rx_desc->hdrlen_flags,
 			  VIRTCHNL2_RX_FLEX_DESC_ADV_RSC_M))
-		return idpf_rx_rsc(rxq, skb, rx_desc, &decoded);
+		return idpf_rx_rsc(rxq, skb, rx_desc, decoded);
 
-	idpf_rx_splitq_extract_csum_bits(rx_desc, &csum_bits);
-	idpf_rx_csum(rxq, skb, &csum_bits, &decoded);
+	csum_bits = idpf_rx_splitq_extract_csum_bits(rx_desc);
+	idpf_rx_csum(rxq, skb, csum_bits, decoded);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
index b2bf58146484..98d47728cc8c 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
@@ -331,72 +331,6 @@ struct idpf_rx_buf {
 #define IDPF_RX_MAX_BASE_PTYPE	256
 #define IDPF_INVALID_PTYPE_ID	0xFFFF
 
-/* Packet type non-ip values */
-enum idpf_rx_ptype_l2 {
-	IDPF_RX_PTYPE_L2_RESERVED	= 0,
-	IDPF_RX_PTYPE_L2_MAC_PAY2	= 1,
-	IDPF_RX_PTYPE_L2_TIMESYNC_PAY2	= 2,
-	IDPF_RX_PTYPE_L2_FIP_PAY2	= 3,
-	IDPF_RX_PTYPE_L2_OUI_PAY2	= 4,
-	IDPF_RX_PTYPE_L2_MACCNTRL_PAY2	= 5,
-	IDPF_RX_PTYPE_L2_LLDP_PAY2	= 6,
-	IDPF_RX_PTYPE_L2_ECP_PAY2	= 7,
-	IDPF_RX_PTYPE_L2_EVB_PAY2	= 8,
-	IDPF_RX_PTYPE_L2_QCN_PAY2	= 9,
-	IDPF_RX_PTYPE_L2_EAPOL_PAY2	= 10,
-	IDPF_RX_PTYPE_L2_ARP		= 11,
-};
-
-enum idpf_rx_ptype_outer_ip {
-	IDPF_RX_PTYPE_OUTER_L2	= 0,
-	IDPF_RX_PTYPE_OUTER_IP	= 1,
-};
-
-#define IDPF_RX_PTYPE_TO_IPV(ptype, ipv)			\
-	(((ptype)->outer_ip == IDPF_RX_PTYPE_OUTER_IP) &&	\
-	 ((ptype)->outer_ip_ver == (ipv)))
-
-enum idpf_rx_ptype_outer_ip_ver {
-	IDPF_RX_PTYPE_OUTER_NONE	= 0,
-	IDPF_RX_PTYPE_OUTER_IPV4	= 1,
-	IDPF_RX_PTYPE_OUTER_IPV6	= 2,
-};
-
-enum idpf_rx_ptype_outer_fragmented {
-	IDPF_RX_PTYPE_NOT_FRAG	= 0,
-	IDPF_RX_PTYPE_FRAG	= 1,
-};
-
-enum idpf_rx_ptype_tunnel_type {
-	IDPF_RX_PTYPE_TUNNEL_NONE		= 0,
-	IDPF_RX_PTYPE_TUNNEL_IP_IP		= 1,
-	IDPF_RX_PTYPE_TUNNEL_IP_GRENAT		= 2,
-	IDPF_RX_PTYPE_TUNNEL_IP_GRENAT_MAC	= 3,
-	IDPF_RX_PTYPE_TUNNEL_IP_GRENAT_MAC_VLAN	= 4,
-};
-
-enum idpf_rx_ptype_tunnel_end_prot {
-	IDPF_RX_PTYPE_TUNNEL_END_NONE	= 0,
-	IDPF_RX_PTYPE_TUNNEL_END_IPV4	= 1,
-	IDPF_RX_PTYPE_TUNNEL_END_IPV6	= 2,
-};
-
-enum idpf_rx_ptype_inner_prot {
-	IDPF_RX_PTYPE_INNER_PROT_NONE		= 0,
-	IDPF_RX_PTYPE_INNER_PROT_UDP		= 1,
-	IDPF_RX_PTYPE_INNER_PROT_TCP		= 2,
-	IDPF_RX_PTYPE_INNER_PROT_SCTP		= 3,
-	IDPF_RX_PTYPE_INNER_PROT_ICMP		= 4,
-	IDPF_RX_PTYPE_INNER_PROT_TIMESYNC	= 5,
-};
-
-enum idpf_rx_ptype_payload_layer {
-	IDPF_RX_PTYPE_PAYLOAD_LAYER_NONE	= 0,
-	IDPF_RX_PTYPE_PAYLOAD_LAYER_PAY2	= 1,
-	IDPF_RX_PTYPE_PAYLOAD_LAYER_PAY3	= 2,
-	IDPF_RX_PTYPE_PAYLOAD_LAYER_PAY4	= 3,
-};
-
 enum idpf_tunnel_state {
 	IDPF_PTYPE_TUNNEL_IP                    = BIT(0),
 	IDPF_PTYPE_TUNNEL_IP_GRENAT             = BIT(1),
@@ -404,22 +338,9 @@ enum idpf_tunnel_state {
 };
 
 struct idpf_ptype_state {
-	bool outer_ip;
-	bool outer_frag;
-	u8 tunnel_state;
-};
-
-struct idpf_rx_ptype_decoded {
-	u32 ptype:10;
-	u32 known:1;
-	u32 outer_ip:1;
-	u32 outer_ip_ver:2;
-	u32 outer_frag:1;
-	u32 tunnel_type:3;
-	u32 tunnel_end_prot:2;
-	u32 tunnel_end_frag:1;
-	u32 inner_prot:4;
-	u32 payload_layer:3;
+	bool outer_ip:1;
+	bool outer_frag:1;
+	u8 tunnel_state:6;
 };
 
 /**
@@ -677,7 +598,7 @@ struct idpf_rx_queue {
 	u16 desc_count;
 
 	u32 rxdids;
-	const struct idpf_rx_ptype_decoded *rx_ptype_lkup;
+	const struct libeth_rx_pt *rx_ptype_lkup;
 	__cacheline_group_end_aligned(read_mostly);
 
 	__cacheline_group_begin_aligned(read_write);
@@ -1180,7 +1101,6 @@ void idpf_vport_intr_update_itr_ena_irq(struct idpf_q_vector *q_vector);
 void idpf_vport_intr_deinit(struct idpf_vport *vport);
 int idpf_vport_intr_init(struct idpf_vport *vport);
 void idpf_vport_intr_ena(struct idpf_vport *vport);
-enum pkt_hash_types idpf_ptype_to_htype(const struct idpf_rx_ptype_decoded *decoded);
 int idpf_config_rss(struct idpf_vport *vport);
 int idpf_init_rss(struct idpf_vport *vport);
 void idpf_deinit_rss(struct idpf_vport *vport);
diff --git a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
index 1aa4770dfe18..83f3543a30e1 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
@@ -1,6 +1,8 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /* Copyright (C) 2023 Intel Corporation */
 
+#include <net/libeth/rx.h>
+
 #include "idpf.h"
 #include "idpf_virtchnl.h"
 
@@ -2483,39 +2485,52 @@ do_memcpy:
  * @frag: fragmentation allowed
  *
  */
-static void idpf_fill_ptype_lookup(struct idpf_rx_ptype_decoded *ptype,
+static void idpf_fill_ptype_lookup(struct libeth_rx_pt *ptype,
 				   struct idpf_ptype_state *pstate,
 				   bool ipv4, bool frag)
 {
 	if (!pstate->outer_ip || !pstate->outer_frag) {
-		ptype->outer_ip = IDPF_RX_PTYPE_OUTER_IP;
 		pstate->outer_ip = true;
 
 		if (ipv4)
-			ptype->outer_ip_ver = IDPF_RX_PTYPE_OUTER_IPV4;
+			ptype->outer_ip = LIBETH_RX_PT_OUTER_IPV4;
 		else
-			ptype->outer_ip_ver = IDPF_RX_PTYPE_OUTER_IPV6;
+			ptype->outer_ip = LIBETH_RX_PT_OUTER_IPV6;
 
 		if (frag) {
-			ptype->outer_frag = IDPF_RX_PTYPE_FRAG;
+			ptype->outer_frag = LIBETH_RX_PT_FRAG;
 			pstate->outer_frag = true;
 		}
 	} else {
-		ptype->tunnel_type = IDPF_RX_PTYPE_TUNNEL_IP_IP;
+		ptype->tunnel_type = LIBETH_RX_PT_TUNNEL_IP_IP;
 		pstate->tunnel_state = IDPF_PTYPE_TUNNEL_IP;
 
 		if (ipv4)
-			ptype->tunnel_end_prot =
-					IDPF_RX_PTYPE_TUNNEL_END_IPV4;
+			ptype->tunnel_end_prot = LIBETH_RX_PT_TUNNEL_END_IPV4;
 		else
-			ptype->tunnel_end_prot =
-					IDPF_RX_PTYPE_TUNNEL_END_IPV6;
+			ptype->tunnel_end_prot = LIBETH_RX_PT_TUNNEL_END_IPV6;
 
 		if (frag)
-			ptype->tunnel_end_frag = IDPF_RX_PTYPE_FRAG;
+			ptype->tunnel_end_frag = LIBETH_RX_PT_FRAG;
 	}
 }
 
+static void idpf_finalize_ptype_lookup(struct libeth_rx_pt *ptype)
+{
+	if (ptype->payload_layer == LIBETH_RX_PT_PAYLOAD_L2 &&
+	    ptype->inner_prot)
+		ptype->payload_layer = LIBETH_RX_PT_PAYLOAD_L4;
+	else if (ptype->payload_layer == LIBETH_RX_PT_PAYLOAD_L2 &&
+		 ptype->outer_ip)
+		ptype->payload_layer = LIBETH_RX_PT_PAYLOAD_L3;
+	else if (ptype->outer_ip == LIBETH_RX_PT_OUTER_L2)
+		ptype->payload_layer = LIBETH_RX_PT_PAYLOAD_L2;
+	else
+		ptype->payload_layer = LIBETH_RX_PT_PAYLOAD_NONE;
+
+	libeth_rx_pt_gen_hash_type(ptype);
+}
+
 /**
  * idpf_send_get_rx_ptype_msg - Send virtchnl for ptype info
  * @vport: virtual port data structure
@@ -2526,7 +2541,7 @@ int idpf_send_get_rx_ptype_msg(struct idpf_vport *vport)
 {
 	struct virtchnl2_get_ptype_info *get_ptype_info __free(kfree) = NULL;
 	struct virtchnl2_get_ptype_info *ptype_info __free(kfree) = NULL;
-	struct idpf_rx_ptype_decoded *ptype_lkup = vport->rx_ptype_lkup;
+	struct libeth_rx_pt *ptype_lkup __free(kfree) = NULL;
 	int max_ptype, ptypes_recvd = 0, ptype_offset;
 	struct idpf_adapter *adapter = vport->adapter;
 	struct idpf_vc_xn_params xn_params = {};
@@ -2534,12 +2549,17 @@ int idpf_send_get_rx_ptype_msg(struct idpf_vport *vport)
 	ssize_t reply_sz;
 	int i, j, k;
 
+	if (vport->rx_ptype_lkup)
+		return 0;
+
 	if (idpf_is_queue_model_split(vport->rxq_model))
 		max_ptype = IDPF_RX_MAX_PTYPE;
 	else
 		max_ptype = IDPF_RX_MAX_BASE_PTYPE;
 
-	memset(vport->rx_ptype_lkup, 0, sizeof(vport->rx_ptype_lkup));
+	ptype_lkup = kcalloc(max_ptype, sizeof(*ptype_lkup), GFP_KERNEL);
+	if (!ptype_lkup)
+		return -ENOMEM;
 
 	get_ptype_info = kzalloc(sizeof(*get_ptype_info), GFP_KERNEL);
 	if (!get_ptype_info)
@@ -2597,16 +2617,13 @@ int idpf_send_get_rx_ptype_msg(struct idpf_vport *vport)
 			/* 0xFFFF indicates end of ptypes */
 			if (le16_to_cpu(ptype->ptype_id_10) ==
 							IDPF_INVALID_PTYPE_ID)
-				return 0;
+				goto out;
 
 			if (idpf_is_queue_model_split(vport->rxq_model))
 				k = le16_to_cpu(ptype->ptype_id_10);
 			else
 				k = ptype->ptype_id_8;
 
-			if (ptype->proto_id_count)
-				ptype_lkup[k].known = 1;
-
 			for (j = 0; j < ptype->proto_id_count; j++) {
 				id = le16_to_cpu(ptype->proto_id[j]);
 				switch (id) {
@@ -2614,18 +2631,18 @@ int idpf_send_get_rx_ptype_msg(struct idpf_vport *vport)
 					if (pstate.tunnel_state ==
 							IDPF_PTYPE_TUNNEL_IP) {
 						ptype_lkup[k].tunnel_type =
-						IDPF_RX_PTYPE_TUNNEL_IP_GRENAT;
+						LIBETH_RX_PT_TUNNEL_IP_GRENAT;
 						pstate.tunnel_state |=
 						IDPF_PTYPE_TUNNEL_IP_GRENAT;
 					}
 					break;
 				case VIRTCHNL2_PROTO_HDR_MAC:
 					ptype_lkup[k].outer_ip =
-						IDPF_RX_PTYPE_OUTER_L2;
+						LIBETH_RX_PT_OUTER_L2;
 					if (pstate.tunnel_state ==
 							IDPF_TUN_IP_GRE) {
 						ptype_lkup[k].tunnel_type =
-						IDPF_RX_PTYPE_TUNNEL_IP_GRENAT_MAC;
+						LIBETH_RX_PT_TUNNEL_IP_GRENAT_MAC;
 						pstate.tunnel_state |=
 						IDPF_PTYPE_TUNNEL_IP_GRENAT_MAC;
 					}
@@ -2652,23 +2669,23 @@ int idpf_send_get_rx_ptype_msg(struct idpf_vport *vport)
 					break;
 				case VIRTCHNL2_PROTO_HDR_UDP:
 					ptype_lkup[k].inner_prot =
-					IDPF_RX_PTYPE_INNER_PROT_UDP;
+					LIBETH_RX_PT_INNER_UDP;
 					break;
 				case VIRTCHNL2_PROTO_HDR_TCP:
 					ptype_lkup[k].inner_prot =
-					IDPF_RX_PTYPE_INNER_PROT_TCP;
+					LIBETH_RX_PT_INNER_TCP;
 					break;
 				case VIRTCHNL2_PROTO_HDR_SCTP:
 					ptype_lkup[k].inner_prot =
-					IDPF_RX_PTYPE_INNER_PROT_SCTP;
+					LIBETH_RX_PT_INNER_SCTP;
 					break;
 				case VIRTCHNL2_PROTO_HDR_ICMP:
 					ptype_lkup[k].inner_prot =
-					IDPF_RX_PTYPE_INNER_PROT_ICMP;
+					LIBETH_RX_PT_INNER_ICMP;
 					break;
 				case VIRTCHNL2_PROTO_HDR_PAY:
 					ptype_lkup[k].payload_layer =
-						IDPF_RX_PTYPE_PAYLOAD_LAYER_PAY2;
+						LIBETH_RX_PT_PAYLOAD_L2;
 					break;
 				case VIRTCHNL2_PROTO_HDR_ICMPV6:
 				case VIRTCHNL2_PROTO_HDR_IPV6_EH:
@@ -2722,9 +2739,14 @@ int idpf_send_get_rx_ptype_msg(struct idpf_vport *vport)
 					break;
 				}
 			}
+
+			idpf_finalize_ptype_lookup(&ptype_lkup[k]);
 		}
 	}
 
+out:
+	vport->rx_ptype_lkup = no_free_ptr(ptype_lkup);
+
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From 4309363f19598999b25a1e55fccf688daa4cc220 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:44 +0200
Subject: idpf: remove legacy Page Pool Ethtool stats

Page Pool Ethtool stats are deprecated since the Netlink Page Pool
interface introduction.
idpf receives big changes in Rx buffer management, including &page_pool
layout, so keeping these deprecated stats does only harm, not speaking
of that CONFIG_IDPF selects CONFIG_PAGE_POOL_STATS unconditionally,
while the latter is often turned off for better performance.
Remove all the references to PP stats from the Ethtool code. The stats
are still available in their full via the generic Netlink interface.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/idpf/Kconfig        |  1 -
 drivers/net/ethernet/intel/idpf/idpf_ethtool.c | 29 +-------------------------
 2 files changed, 1 insertion(+), 29 deletions(-)

diff --git a/drivers/net/ethernet/intel/idpf/Kconfig b/drivers/net/ethernet/intel/idpf/Kconfig
index 638484c5723c..1f071143d992 100644
--- a/drivers/net/ethernet/intel/idpf/Kconfig
+++ b/drivers/net/ethernet/intel/idpf/Kconfig
@@ -7,7 +7,6 @@ config IDPF
 	select DIMLIB
 	select LIBETH
 	select PAGE_POOL
-	select PAGE_POOL_STATS
 	help
 	  This driver supports Intel(R) Infrastructure Data Path Function
 	  devices.
diff --git a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c
index e933fed16c7e..3806ddd3ce4a 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c
@@ -565,8 +565,6 @@ static void idpf_get_stat_strings(struct net_device *netdev, u8 *data)
 	for (i = 0; i < vport_config->max_q.max_rxq; i++)
 		idpf_add_qstat_strings(&data, idpf_gstrings_rx_queue_stats,
 				       "rx", i);
-
-	page_pool_ethtool_stats_get_strings(data);
 }
 
 /**
@@ -600,7 +598,6 @@ static int idpf_get_sset_count(struct net_device *netdev, int sset)
 	struct idpf_netdev_priv *np = netdev_priv(netdev);
 	struct idpf_vport_config *vport_config;
 	u16 max_txq, max_rxq;
-	unsigned int size;
 
 	if (sset != ETH_SS_STATS)
 		return -EINVAL;
@@ -619,11 +616,8 @@ static int idpf_get_sset_count(struct net_device *netdev, int sset)
 	max_txq = vport_config->max_q.max_txq;
 	max_rxq = vport_config->max_q.max_rxq;
 
-	size = IDPF_PORT_STATS_LEN + (IDPF_TX_QUEUE_STATS_LEN * max_txq) +
+	return IDPF_PORT_STATS_LEN + (IDPF_TX_QUEUE_STATS_LEN * max_txq) +
 	       (IDPF_RX_QUEUE_STATS_LEN * max_rxq);
-	size += page_pool_ethtool_stats_get_count();
-
-	return size;
 }
 
 /**
@@ -876,7 +870,6 @@ static void idpf_get_ethtool_stats(struct net_device *netdev,
 {
 	struct idpf_netdev_priv *np = netdev_priv(netdev);
 	struct idpf_vport_config *vport_config;
-	struct page_pool_stats pp_stats = { };
 	struct idpf_vport *vport;
 	unsigned int total = 0;
 	unsigned int i, j;
@@ -946,32 +939,12 @@ static void idpf_get_ethtool_stats(struct net_device *netdev,
 				idpf_add_empty_queue_stats(&data, qtype);
 			else
 				idpf_add_queue_stats(&data, rxq, qtype);
-
-			/* In splitq mode, don't get page pool stats here since
-			 * the pools are attached to the buffer queues
-			 */
-			if (is_splitq)
-				continue;
-
-			if (rxq)
-				page_pool_get_stats(rxq->pp, &pp_stats);
-		}
-	}
-
-	for (i = 0; i < vport->num_rxq_grp; i++) {
-		for (j = 0; j < vport->num_bufqs_per_qgrp; j++) {
-			struct idpf_buf_queue *rxbufq =
-				&vport->rxq_grps[i].splitq.bufq_sets[j].bufq;
-
-			page_pool_get_stats(rxbufq->pp, &pp_stats);
 		}
 	}
 
 	for (; total < vport_config->max_q.max_rxq; total++)
 		idpf_add_empty_queue_stats(&data, VIRTCHNL2_QUEUE_TYPE_RX);
 
-	page_pool_ethtool_stats_get(data, &pp_stats);
-
 	rcu_read_unlock();
 
 	idpf_vport_ctrl_unlock(netdev);
-- 
cgit v1.2.3-58-ga151


From 5aaac1aece4e4db9d620791a33f7a4173c660e65 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:45 +0200
Subject: libeth: support different types of buffers for Rx

Unlike previous generations, idpf requires more buffer types for optimal
performance. This includes: header buffers, short buffers, and
no-overhead buffers (w/o headroom and tailroom, for TCP zerocopy when
the header split is enabled).
Introduce libeth Rx buffer type and calculate page_pool params
accordingly. All the HW-related details like buffer alignment are still
accounted. For the header buffers, pick 256 bytes as in most places in
the kernel (have you ever seen frames with bigger headers?).

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/libeth/rx.c | 132 ++++++++++++++++++++++++++++++---
 include/net/libeth/rx.h                |  19 +++++
 2 files changed, 140 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/intel/libeth/rx.c b/drivers/net/ethernet/intel/libeth/rx.c
index 6221b88c34ac..d0b158b6e55b 100644
--- a/drivers/net/ethernet/intel/libeth/rx.c
+++ b/drivers/net/ethernet/intel/libeth/rx.c
@@ -6,7 +6,7 @@
 /* Rx buffer management */
 
 /**
- * libeth_rx_hw_len - get the actual buffer size to be passed to HW
+ * libeth_rx_hw_len_mtu - get the actual buffer size to be passed to HW
  * @pp: &page_pool_params of the netdev to calculate the size for
  * @max_len: maximum buffer size for a single descriptor
  *
@@ -14,7 +14,7 @@
  * MTU the @dev has, HW required alignment, minimum and maximum allowed values,
  * and system's page size.
  */
-static u32 libeth_rx_hw_len(const struct page_pool_params *pp, u32 max_len)
+static u32 libeth_rx_hw_len_mtu(const struct page_pool_params *pp, u32 max_len)
 {
 	u32 len;
 
@@ -26,6 +26,118 @@ static u32 libeth_rx_hw_len(const struct page_pool_params *pp, u32 max_len)
 	return len;
 }
 
+/**
+ * libeth_rx_hw_len_truesize - get the short buffer size to be passed to HW
+ * @pp: &page_pool_params of the netdev to calculate the size for
+ * @max_len: maximum buffer size for a single descriptor
+ * @truesize: desired truesize for the buffers
+ *
+ * Return: HW-writeable length per one buffer to pass it to the HW ignoring the
+ * MTU and closest to the passed truesize. Can be used for "short" buffer
+ * queues to fragment pages more efficiently.
+ */
+static u32 libeth_rx_hw_len_truesize(const struct page_pool_params *pp,
+				     u32 max_len, u32 truesize)
+{
+	u32 min, len;
+
+	min = SKB_HEAD_ALIGN(pp->offset + LIBETH_RX_BUF_STRIDE);
+	truesize = clamp(roundup_pow_of_two(truesize), roundup_pow_of_two(min),
+			 PAGE_SIZE << LIBETH_RX_PAGE_ORDER);
+
+	len = SKB_WITH_OVERHEAD(truesize - pp->offset);
+	len = ALIGN_DOWN(len, LIBETH_RX_BUF_STRIDE) ? : LIBETH_RX_BUF_STRIDE;
+	len = min3(len, ALIGN_DOWN(max_len ? : U32_MAX, LIBETH_RX_BUF_STRIDE),
+		   pp->max_len);
+
+	return len;
+}
+
+/**
+ * libeth_rx_page_pool_params - calculate params with the stack overhead
+ * @fq: buffer queue to calculate the size for
+ * @pp: &page_pool_params of the netdev
+ *
+ * Set the PP params to will all needed stack overhead (headroom, tailroom) and
+ * both the HW buffer length and the truesize for all types of buffers. For
+ * "short" buffers, truesize never exceeds the "wanted" one; for the rest,
+ * it can be up to the page size.
+ *
+ * Return: true on success, false on invalid input params.
+ */
+static bool libeth_rx_page_pool_params(struct libeth_fq *fq,
+				       struct page_pool_params *pp)
+{
+	pp->offset = LIBETH_SKB_HEADROOM;
+	/* HW-writeable / syncable length per one page */
+	pp->max_len = LIBETH_RX_PAGE_LEN(pp->offset);
+
+	/* HW-writeable length per buffer */
+	switch (fq->type) {
+	case LIBETH_FQE_MTU:
+		fq->buf_len = libeth_rx_hw_len_mtu(pp, fq->buf_len);
+		break;
+	case LIBETH_FQE_SHORT:
+		fq->buf_len = libeth_rx_hw_len_truesize(pp, fq->buf_len,
+							fq->truesize);
+		break;
+	case LIBETH_FQE_HDR:
+		fq->buf_len = ALIGN(LIBETH_MAX_HEAD, LIBETH_RX_BUF_STRIDE);
+		break;
+	default:
+		return false;
+	}
+
+	/* Buffer size to allocate */
+	fq->truesize = roundup_pow_of_two(SKB_HEAD_ALIGN(pp->offset +
+							 fq->buf_len));
+
+	return true;
+}
+
+/**
+ * libeth_rx_page_pool_params_zc - calculate params without the stack overhead
+ * @fq: buffer queue to calculate the size for
+ * @pp: &page_pool_params of the netdev
+ *
+ * Set the PP params to exclude the stack overhead and both the buffer length
+ * and the truesize, which are equal for the data buffers. Note that this
+ * requires separate header buffers to be always active and account the
+ * overhead.
+ * With the MTU == ``PAGE_SIZE``, this allows the kernel to enable the zerocopy
+ * mode.
+ *
+ * Return: true on success, false on invalid input params.
+ */
+static bool libeth_rx_page_pool_params_zc(struct libeth_fq *fq,
+					  struct page_pool_params *pp)
+{
+	u32 mtu, max;
+
+	pp->offset = 0;
+	pp->max_len = PAGE_SIZE << LIBETH_RX_PAGE_ORDER;
+
+	switch (fq->type) {
+	case LIBETH_FQE_MTU:
+		mtu = READ_ONCE(pp->netdev->mtu);
+		break;
+	case LIBETH_FQE_SHORT:
+		mtu = fq->truesize;
+		break;
+	default:
+		return false;
+	}
+
+	mtu = roundup_pow_of_two(mtu);
+	max = min(rounddown_pow_of_two(fq->buf_len ? : U32_MAX),
+		  pp->max_len);
+
+	fq->buf_len = clamp(mtu, LIBETH_RX_BUF_STRIDE, max);
+	fq->truesize = fq->buf_len;
+
+	return true;
+}
+
 /**
  * libeth_rx_fq_create - create a PP with the default libeth settings
  * @fq: buffer queue struct to fill
@@ -44,19 +156,17 @@ int libeth_rx_fq_create(struct libeth_fq *fq, struct napi_struct *napi)
 		.netdev		= napi->dev,
 		.napi		= napi,
 		.dma_dir	= DMA_FROM_DEVICE,
-		.offset		= LIBETH_SKB_HEADROOM,
 	};
 	struct libeth_fqe *fqes;
 	struct page_pool *pool;
+	bool ret;
 
-	/* HW-writeable / syncable length per one page */
-	pp.max_len = LIBETH_RX_PAGE_LEN(pp.offset);
-
-	/* HW-writeable length per buffer */
-	fq->buf_len = libeth_rx_hw_len(&pp, fq->buf_len);
-	/* Buffer size to allocate */
-	fq->truesize = roundup_pow_of_two(SKB_HEAD_ALIGN(pp.offset +
-							 fq->buf_len));
+	if (!fq->hsplit)
+		ret = libeth_rx_page_pool_params(fq, &pp);
+	else
+		ret = libeth_rx_page_pool_params_zc(fq, &pp);
+	if (!ret)
+		return -EINVAL;
 
 	pool = page_pool_create(&pp);
 	if (IS_ERR(pool))
diff --git a/include/net/libeth/rx.h b/include/net/libeth/rx.h
index f29ea3e34c6c..43574bd6612f 100644
--- a/include/net/libeth/rx.h
+++ b/include/net/libeth/rx.h
@@ -17,6 +17,8 @@
 #define LIBETH_MAX_HEADROOM	LIBETH_SKB_HEADROOM
 /* Link layer / L2 overhead: Ethernet, 2 VLAN tags (C + S), FCS */
 #define LIBETH_RX_LL_LEN	(ETH_HLEN + 2 * VLAN_HLEN + ETH_FCS_LEN)
+/* Maximum supported L2-L4 header length */
+#define LIBETH_MAX_HEAD		roundup_pow_of_two(max(MAX_HEADER, 256))
 
 /* Always use order-0 pages */
 #define LIBETH_RX_PAGE_ORDER	0
@@ -43,6 +45,18 @@ struct libeth_fqe {
 	u32			truesize;
 } __aligned_largest;
 
+/**
+ * enum libeth_fqe_type - enum representing types of Rx buffers
+ * @LIBETH_FQE_MTU: buffer size is determined by MTU
+ * @LIBETH_FQE_SHORT: buffer size is smaller than MTU, for short frames
+ * @LIBETH_FQE_HDR: buffer size is ```LIBETH_MAX_HEAD```-sized, for headers
+ */
+enum libeth_fqe_type {
+	LIBETH_FQE_MTU		= 0U,
+	LIBETH_FQE_SHORT,
+	LIBETH_FQE_HDR,
+};
+
 /**
  * struct libeth_fq - structure representing a buffer (fill) queue
  * @fp: hotpath part of the structure
@@ -50,6 +64,8 @@ struct libeth_fqe {
  * @fqes: array of Rx buffers
  * @truesize: size to allocate per buffer, w/overhead
  * @count: number of descriptors/buffers the queue has
+ * @type: type of the buffers this queue has
+ * @hsplit: flag whether header split is enabled
  * @buf_len: HW-writeable length per each buffer
  * @nid: ID of the closest NUMA node with memory
  */
@@ -63,6 +79,9 @@ struct libeth_fq {
 	);
 
 	/* Cold fields */
+	enum libeth_fqe_type	type:2;
+	bool			hsplit:1;
+
 	u32			buf_len;
 	int			nid;
 };
-- 
cgit v1.2.3-58-ga151


From 90912f9f4f2d159e93cf9a4c732506342489a1d5 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:46 +0200
Subject: idpf: convert header split mode to libeth + napi_build_skb()

Currently, idpf uses the following model for the header buffers:

* buffers are allocated via dma_alloc_coherent();
* when receiving, napi_alloc_skb() is called and then the header is
  copied to the newly allocated linear part.

This is far from optimal as DMA coherent zone is slow on many systems
and memcpy() neutralizes the idea and benefits of the header split. Not
speaking of that XDP can't be run on DMA coherent buffers, but at the
same time the idea of allocating an skb to run XDP program is ill.
Instead, use libeth to create page_pools for the header buffers, allocate
them dynamically and then build an skb via napi_build_skb() around them
with no memory copy. With one exception...

When you enable header split, you expect you'll always have a separate
header buffer, so that you could reserve headroom and tailroom only
there and then use full buffers for the data. For example, this is how
TCP zerocopy works -- you have to have the payload aligned to PAGE_SIZE.
The current hardware running idpf does *not* guarantee that you'll
always have headers placed separately. For example, on my setup, even
ICMP packets are written as one piece to the data buffers. You can't
build a valid skb around a data buffer in this case.
To not complicate things and not lose TCP zerocopy etc., when such thing
happens, use the empty header buffer and pull either full frame (if it's
short) or the Ethernet header there and build an skb around it. GRO
layer will pull more from the data buffer later. This W/A will hopefully
be removed one day.

Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 .../net/ethernet/intel/idpf/idpf_singleq_txrx.c    |   1 +
 drivers/net/ethernet/intel/idpf/idpf_txrx.c        | 253 ++++++++++++++-------
 drivers/net/ethernet/intel/idpf/idpf_txrx.h        |  52 +++--
 drivers/net/ethernet/intel/idpf/idpf_virtchnl.c    |  14 +-
 4 files changed, 204 insertions(+), 116 deletions(-)

diff --git a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
index 6c0794e01922..ced1fdf7a1fb 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
@@ -827,6 +827,7 @@ idpf_rx_singleq_process_skb_fields(struct idpf_rx_queue *rx_q,
 	}
 
 	idpf_rx_singleq_csum(rx_q, skb, csum_bits, decoded);
+	skb_record_rx_queue(skb, rx_q->idx);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
index 883e58815f92..ec4a9501d19f 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
@@ -381,7 +381,7 @@ err_out:
  * idpf_rx_page_rel - Release an rx buffer page
  * @rx_buf: the buffer to free
  */
-static void idpf_rx_page_rel(struct idpf_rx_buf *rx_buf)
+static void idpf_rx_page_rel(struct libeth_fqe *rx_buf)
 {
 	if (unlikely(!rx_buf->page))
 		return;
@@ -389,46 +389,50 @@ static void idpf_rx_page_rel(struct idpf_rx_buf *rx_buf)
 	page_pool_put_full_page(rx_buf->page->pp, rx_buf->page, false);
 
 	rx_buf->page = NULL;
-	rx_buf->page_offset = 0;
+	rx_buf->offset = 0;
 }
 
 /**
  * idpf_rx_hdr_buf_rel_all - Release header buffer memory
  * @bufq: queue to use
- * @dev: device to free DMA memory
  */
-static void idpf_rx_hdr_buf_rel_all(struct idpf_buf_queue *bufq,
-				    struct device *dev)
+static void idpf_rx_hdr_buf_rel_all(struct idpf_buf_queue *bufq)
 {
-	dma_free_coherent(dev, bufq->desc_count * IDPF_HDR_BUF_SIZE,
-			  bufq->rx_buf.hdr_buf_va, bufq->rx_buf.hdr_buf_pa);
-	bufq->rx_buf.hdr_buf_va = NULL;
+	struct libeth_fq fq = {
+		.fqes	= bufq->hdr_buf,
+		.pp	= bufq->hdr_pp,
+	};
+
+	for (u32 i = 0; i < bufq->desc_count; i++)
+		idpf_rx_page_rel(&bufq->hdr_buf[i]);
+
+	libeth_rx_fq_destroy(&fq);
+	bufq->hdr_buf = NULL;
+	bufq->hdr_pp = NULL;
 }
 
 /**
  * idpf_rx_buf_rel_bufq - Free all Rx buffer resources for a buffer queue
  * @bufq: queue to be cleaned
- * @dev: device to free DMA memory
  */
-static void idpf_rx_buf_rel_bufq(struct idpf_buf_queue *bufq,
-				 struct device *dev)
+static void idpf_rx_buf_rel_bufq(struct idpf_buf_queue *bufq)
 {
 	/* queue already cleared, nothing to do */
-	if (!bufq->rx_buf.buf)
+	if (!bufq->buf)
 		return;
 
 	/* Free all the bufs allocated and given to hw on Rx queue */
 	for (u32 i = 0; i < bufq->desc_count; i++)
-		idpf_rx_page_rel(&bufq->rx_buf.buf[i]);
+		idpf_rx_page_rel(&bufq->buf[i]);
 
 	if (idpf_queue_has(HSPLIT_EN, bufq))
-		idpf_rx_hdr_buf_rel_all(bufq, dev);
+		idpf_rx_hdr_buf_rel_all(bufq);
 
 	page_pool_destroy(bufq->pp);
 	bufq->pp = NULL;
 
-	kfree(bufq->rx_buf.buf);
-	bufq->rx_buf.buf = NULL;
+	kfree(bufq->buf);
+	bufq->buf = NULL;
 }
 
 /**
@@ -493,7 +497,7 @@ static void idpf_rx_desc_rel_bufq(struct idpf_buf_queue *bufq,
 	if (!bufq)
 		return;
 
-	idpf_rx_buf_rel_bufq(bufq, dev);
+	idpf_rx_buf_rel_bufq(bufq);
 
 	bufq->next_to_alloc = 0;
 	bufq->next_to_clean = 0;
@@ -573,12 +577,21 @@ static void idpf_rx_buf_hw_update(struct idpf_buf_queue *bufq, u32 val)
  */
 static int idpf_rx_hdr_buf_alloc_all(struct idpf_buf_queue *bufq)
 {
-	bufq->rx_buf.hdr_buf_va =
-		dma_alloc_coherent(bufq->q_vector->vport->netdev->dev.parent,
-				   IDPF_HDR_BUF_SIZE * bufq->desc_count,
-				   &bufq->rx_buf.hdr_buf_pa, GFP_KERNEL);
-	if (!bufq->rx_buf.hdr_buf_va)
-		return -ENOMEM;
+	struct libeth_fq fq = {
+		.count	= bufq->desc_count,
+		.type	= LIBETH_FQE_HDR,
+		.nid	= idpf_q_vector_to_mem(bufq->q_vector),
+	};
+	int ret;
+
+	ret = libeth_rx_fq_create(&fq, &bufq->q_vector->napi);
+	if (ret)
+		return ret;
+
+	bufq->hdr_pp = fq.pp;
+	bufq->hdr_buf = fq.fqes;
+	bufq->hdr_truesize = fq.truesize;
+	bufq->rx_hbuf_size = fq.buf_len;
 
 	return 0;
 }
@@ -616,17 +629,27 @@ static void idpf_rx_post_buf_refill(struct idpf_sw_queue *refillq, u16 buf_id)
 static bool idpf_rx_post_buf_desc(struct idpf_buf_queue *bufq, u16 buf_id)
 {
 	struct virtchnl2_splitq_rx_buf_desc *splitq_rx_desc = NULL;
+	struct libeth_fq_fp fq = {
+		.count	= bufq->desc_count,
+	};
 	u16 nta = bufq->next_to_alloc;
 	struct idpf_rx_buf *buf;
 	dma_addr_t addr;
 
 	splitq_rx_desc = &bufq->split_buf[nta];
-	buf = &bufq->rx_buf.buf[buf_id];
+	buf = &bufq->buf[buf_id];
 
-	if (idpf_queue_has(HSPLIT_EN, bufq))
-		splitq_rx_desc->hdr_addr =
-			cpu_to_le64(bufq->rx_buf.hdr_buf_pa +
-				    (u32)buf_id * IDPF_HDR_BUF_SIZE);
+	if (idpf_queue_has(HSPLIT_EN, bufq)) {
+		fq.pp = bufq->hdr_pp;
+		fq.fqes = bufq->hdr_buf;
+		fq.truesize = bufq->hdr_truesize;
+
+		addr = libeth_rx_alloc(&fq, buf_id);
+		if (addr == DMA_MAPPING_ERROR)
+			return false;
+
+		splitq_rx_desc->hdr_addr = cpu_to_le64(addr);
+	}
 
 	addr = idpf_alloc_page(bufq->pp, buf, bufq->rx_buf_size);
 	if (unlikely(addr == DMA_MAPPING_ERROR))
@@ -741,13 +764,12 @@ static int idpf_rx_bufs_init_singleq(struct idpf_rx_queue *rxq)
  */
 static int idpf_rx_buf_alloc_all(struct idpf_buf_queue *rxbufq)
 {
-	struct device *dev = rxbufq->q_vector->vport->netdev->dev.parent;
 	int err = 0;
 
 	/* Allocate book keeping buffers */
-	rxbufq->rx_buf.buf = kcalloc(rxbufq->desc_count,
-				     sizeof(struct idpf_rx_buf), GFP_KERNEL);
-	if (!rxbufq->rx_buf.buf) {
+	rxbufq->buf = kcalloc(rxbufq->desc_count, sizeof(*rxbufq->buf),
+			      GFP_KERNEL);
+	if (!rxbufq->buf) {
 		err = -ENOMEM;
 		goto rx_buf_alloc_all_out;
 	}
@@ -764,7 +786,7 @@ static int idpf_rx_buf_alloc_all(struct idpf_buf_queue *rxbufq)
 
 rx_buf_alloc_all_out:
 	if (err)
-		idpf_rx_buf_rel_bufq(rxbufq, dev);
+		idpf_rx_buf_rel_bufq(rxbufq);
 
 	return err;
 }
@@ -1489,7 +1511,6 @@ static int idpf_rxq_group_alloc(struct idpf_vport *vport, u16 num_rxq)
 			q->rx_buffer_low_watermark = IDPF_LOW_WATERMARK;
 
 			idpf_queue_assign(HSPLIT_EN, q, hs);
-			q->rx_hbuf_size = hs ? IDPF_HDR_BUF_SIZE : 0;
 
 			bufq_set->num_refillqs = num_rxq;
 			bufq_set->refillqs = kcalloc(num_rxq, swq_size,
@@ -1532,7 +1553,6 @@ skip_splitq_rx_init:
 				      &rx_qgrp->splitq.bufq_sets[1].refillqs[j];
 
 			idpf_queue_assign(HSPLIT_EN, q, hs);
-			q->rx_hbuf_size = hs ? IDPF_HDR_BUF_SIZE : 0;
 
 setup_rxq:
 			q->desc_count = vport->rxq_desc_count;
@@ -3107,6 +3127,8 @@ idpf_rx_process_skb_fields(struct idpf_rx_queue *rxq, struct sk_buff *skb,
 	csum_bits = idpf_rx_splitq_extract_csum_bits(rx_desc);
 	idpf_rx_csum(rxq, skb, csum_bits, decoded);
 
+	skb_record_rx_queue(skb, rxq->idx);
+
 	return 0;
 }
 
@@ -3124,7 +3146,7 @@ void idpf_rx_add_frag(struct idpf_rx_buf *rx_buf, struct sk_buff *skb,
 		      unsigned int size)
 {
 	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buf->page,
-			rx_buf->page_offset, size, rx_buf->truesize);
+			rx_buf->offset, size, rx_buf->truesize);
 
 	rx_buf->page = NULL;
 }
@@ -3147,7 +3169,7 @@ struct sk_buff *idpf_rx_construct_skb(const struct idpf_rx_queue *rxq,
 	struct sk_buff *skb;
 	void *va;
 
-	va = page_address(rx_buf->page) + rx_buf->page_offset;
+	va = page_address(rx_buf->page) + rx_buf->offset;
 
 	/* prefetch first cache line of first page */
 	net_prefetch(va);
@@ -3159,7 +3181,6 @@ struct sk_buff *idpf_rx_construct_skb(const struct idpf_rx_queue *rxq,
 		return NULL;
 	}
 
-	skb_record_rx_queue(skb, rxq->idx);
 	skb_mark_for_recycle(skb);
 
 	/* Determine available headroom for copy */
@@ -3178,7 +3199,7 @@ struct sk_buff *idpf_rx_construct_skb(const struct idpf_rx_queue *rxq,
 		return skb;
 	}
 
-	skb_add_rx_frag(skb, 0, rx_buf->page, rx_buf->page_offset + headlen,
+	skb_add_rx_frag(skb, 0, rx_buf->page, rx_buf->offset + headlen,
 			size, rx_buf->truesize);
 
 	/* Since we're giving the page to the stack, clear our reference to it.
@@ -3190,36 +3211,66 @@ struct sk_buff *idpf_rx_construct_skb(const struct idpf_rx_queue *rxq,
 }
 
 /**
- * idpf_rx_hdr_construct_skb - Allocate skb and populate it from header buffer
- * @rxq: Rx descriptor queue
- * @va: Rx buffer to pull data from
+ * idpf_rx_hsplit_wa - handle header buffer overflows and split errors
+ * @hdr: Rx buffer for the headers
+ * @buf: Rx buffer for the payload
+ * @data_len: number of bytes received to the payload buffer
+ *
+ * When a header buffer overflow occurs or the HW was unable do parse the
+ * packet type to perform header split, the whole frame gets placed to the
+ * payload buffer. We can't build a valid skb around a payload buffer when
+ * the header split is active since it doesn't reserve any head- or tailroom.
+ * In that case, copy either the whole frame when it's short or just the
+ * Ethernet header to the header buffer to be able to build an skb and adjust
+ * the data offset in the payload buffer, IOW emulate the header split.
+ *
+ * Return: number of bytes copied to the header buffer.
+ */
+static u32 idpf_rx_hsplit_wa(const struct libeth_fqe *hdr,
+			     struct libeth_fqe *buf, u32 data_len)
+{
+	u32 copy = data_len <= L1_CACHE_BYTES ? data_len : ETH_HLEN;
+	const void *src;
+	void *dst;
+
+	if (!libeth_rx_sync_for_cpu(buf, copy))
+		return 0;
+
+	dst = page_address(hdr->page) + hdr->offset + hdr->page->pp->p.offset;
+	src = page_address(buf->page) + buf->offset + buf->page->pp->p.offset;
+	memcpy(dst, src, LARGEST_ALIGN(copy));
+
+	buf->offset += copy;
+
+	return copy;
+}
+
+/**
+ * idpf_rx_build_skb - Allocate skb and populate it from header buffer
+ * @buf: Rx buffer to pull data from
  * @size: the length of the packet
  *
  * This function allocates an skb. It then populates it with the page data from
  * the current receive descriptor, taking care to set up the skb correctly.
- * This specifically uses a header buffer to start building the skb.
  */
-static struct sk_buff *
-idpf_rx_hdr_construct_skb(const struct idpf_rx_queue *rxq, const void *va,
-			  unsigned int size)
+struct sk_buff *idpf_rx_build_skb(const struct libeth_fqe *buf, u32 size)
 {
+	u32 hr = buf->page->pp->p.offset;
 	struct sk_buff *skb;
+	void *va;
 
-	/* allocate a skb to store the frags */
-	skb = napi_alloc_skb(rxq->napi, size);
+	va = page_address(buf->page) + buf->offset;
+	prefetch(va + hr);
+
+	skb = napi_build_skb(va, buf->truesize);
 	if (unlikely(!skb))
 		return NULL;
 
-	skb_record_rx_queue(skb, rxq->idx);
-
-	memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long)));
-
-	/* More than likely, a payload fragment, which will use a page from
-	 * page_pool will be added to the SKB so mark it for recycle
-	 * preemptively. And if not, it's inconsequential.
-	 */
 	skb_mark_for_recycle(skb);
 
+	skb_reserve(skb, hr);
+	__skb_put(skb, size);
+
 	return skb;
 }
 
@@ -3272,14 +3323,12 @@ static int idpf_rx_splitq_clean(struct idpf_rx_queue *rxq, int budget)
 	/* Process Rx packets bounded by budget */
 	while (likely(total_rx_pkts < budget)) {
 		struct virtchnl2_rx_flex_desc_adv_nic_3 *rx_desc;
+		struct libeth_fqe *hdr, *rx_buf = NULL;
 		struct idpf_sw_queue *refillq = NULL;
 		struct idpf_rxq_set *rxq_set = NULL;
-		struct idpf_rx_buf *rx_buf = NULL;
 		unsigned int pkt_len = 0;
 		unsigned int hdr_len = 0;
 		u16 gen_id, buf_id = 0;
-		 /* Header buffer overflow only valid for header split */
-		bool hbo = false;
 		int bufq_id;
 		u8 rxdid;
 
@@ -3311,25 +3360,6 @@ static int idpf_rx_splitq_clean(struct idpf_rx_queue *rxq, int budget)
 		pkt_len = le16_get_bits(rx_desc->pktlen_gen_bufq_id,
 					VIRTCHNL2_RX_FLEX_DESC_ADV_LEN_PBUF_M);
 
-		hbo = FIELD_GET(VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_HBO_M,
-				rx_desc->status_err0_qw1);
-
-		if (unlikely(hbo)) {
-			/* If a header buffer overflow, occurs, i.e. header is
-			 * too large to fit in the header split buffer, HW will
-			 * put the entire packet, including headers, in the
-			 * data/payload buffer.
-			 */
-			u64_stats_update_begin(&rxq->stats_sync);
-			u64_stats_inc(&rxq->q_stats.hsplit_buf_ovf);
-			u64_stats_update_end(&rxq->stats_sync);
-			goto bypass_hsplit;
-		}
-
-		hdr_len = le16_get_bits(rx_desc->hdrlen_flags,
-					VIRTCHNL2_RX_FLEX_DESC_ADV_LEN_HDR_M);
-
-bypass_hsplit:
 		bufq_id = le16_get_bits(rx_desc->pktlen_gen_bufq_id,
 					VIRTCHNL2_RX_FLEX_DESC_ADV_BUFQ_ID_M);
 
@@ -3341,18 +3371,48 @@ bypass_hsplit:
 
 		buf_id = le16_to_cpu(rx_desc->buf_id);
 
-		rx_buf = &rx_bufq->rx_buf.buf[buf_id];
+		rx_buf = &rx_bufq->buf[buf_id];
 
-		if (hdr_len) {
-			const void *va = (u8 *)rx_bufq->rx_buf.hdr_buf_va +
-						(u32)buf_id * IDPF_HDR_BUF_SIZE;
+		if (!rx_bufq->hdr_pp)
+			goto payload;
+
+#define __HBO_BIT	VIRTCHNL2_RX_FLEX_DESC_ADV_STATUS0_HBO_M
+#define __HDR_LEN_MASK	VIRTCHNL2_RX_FLEX_DESC_ADV_LEN_HDR_M
+		if (likely(!(rx_desc->status_err0_qw1 & __HBO_BIT)))
+			/* If a header buffer overflow, occurs, i.e. header is
+			 * too large to fit in the header split buffer, HW will
+			 * put the entire packet, including headers, in the
+			 * data/payload buffer.
+			 */
+			hdr_len = le16_get_bits(rx_desc->hdrlen_flags,
+						__HDR_LEN_MASK);
+#undef __HDR_LEN_MASK
+#undef __HBO_BIT
+
+		hdr = &rx_bufq->hdr_buf[buf_id];
+
+		if (unlikely(!hdr_len && !skb)) {
+			hdr_len = idpf_rx_hsplit_wa(hdr, rx_buf, pkt_len);
+			pkt_len -= hdr_len;
+
+			u64_stats_update_begin(&rxq->stats_sync);
+			u64_stats_inc(&rxq->q_stats.hsplit_buf_ovf);
+			u64_stats_update_end(&rxq->stats_sync);
+		}
+
+		if (libeth_rx_sync_for_cpu(hdr, hdr_len)) {
+			skb = idpf_rx_build_skb(hdr, hdr_len);
+			if (!skb)
+				break;
 
-			skb = idpf_rx_hdr_construct_skb(rxq, va, hdr_len);
 			u64_stats_update_begin(&rxq->stats_sync);
 			u64_stats_inc(&rxq->q_stats.hsplit_pkts);
 			u64_stats_update_end(&rxq->stats_sync);
 		}
 
+		hdr->page = NULL;
+
+payload:
 		if (pkt_len) {
 			idpf_rx_sync_for_cpu(rx_buf, pkt_len);
 			if (skb)
@@ -3422,10 +3482,13 @@ bypass_hsplit:
 static int idpf_rx_update_bufq_desc(struct idpf_buf_queue *bufq, u32 buf_id,
 				    struct virtchnl2_splitq_rx_buf_desc *buf_desc)
 {
+	struct libeth_fq_fp fq = {
+		.count		= bufq->desc_count,
+	};
 	struct idpf_rx_buf *buf;
 	dma_addr_t addr;
 
-	buf = &bufq->rx_buf.buf[buf_id];
+	buf = &bufq->buf[buf_id];
 
 	addr = idpf_alloc_page(bufq->pp, buf, bufq->rx_buf_size);
 	if (unlikely(addr == DMA_MAPPING_ERROR))
@@ -3437,8 +3500,15 @@ static int idpf_rx_update_bufq_desc(struct idpf_buf_queue *bufq, u32 buf_id,
 	if (!idpf_queue_has(HSPLIT_EN, bufq))
 		return 0;
 
-	buf_desc->hdr_addr = cpu_to_le64(bufq->rx_buf.hdr_buf_pa +
-					 (u32)buf_id * IDPF_HDR_BUF_SIZE);
+	fq.pp = bufq->hdr_pp;
+	fq.fqes = bufq->hdr_buf;
+	fq.truesize = bufq->hdr_truesize;
+
+	addr = libeth_rx_alloc(&fq, buf_id);
+	if (addr == DMA_MAPPING_ERROR)
+		return -ENOMEM;
+
+	buf_desc->hdr_addr = cpu_to_le64(addr);
 
 	return 0;
 }
@@ -3509,16 +3579,20 @@ static void idpf_rx_clean_refillq(struct idpf_buf_queue *bufq,
 /**
  * idpf_rx_clean_refillq_all - Clean all refill queues
  * @bufq: buffer queue with refill queues
+ * @nid: ID of the closest NUMA node with memory
  *
  * Iterates through all refill queues assigned to the buffer queue assigned to
  * this vector.  Returns true if clean is complete within budget, false
  * otherwise.
  */
-static void idpf_rx_clean_refillq_all(struct idpf_buf_queue *bufq)
+static void idpf_rx_clean_refillq_all(struct idpf_buf_queue *bufq, int nid)
 {
 	struct idpf_bufq_set *bufq_set;
 	int i;
 
+	if (bufq->hdr_pp)
+		page_pool_nid_changed(bufq->hdr_pp, nid);
+
 	bufq_set = container_of(bufq, struct idpf_bufq_set, bufq);
 	for (i = 0; i < bufq_set->num_refillqs; i++)
 		idpf_rx_clean_refillq(bufq, &bufq_set->refillqs[i]);
@@ -4020,6 +4094,7 @@ static bool idpf_rx_splitq_clean_all(struct idpf_q_vector *q_vec, int budget,
 	bool clean_complete = true;
 	int pkts_cleaned = 0;
 	int i, budget_per_q;
+	int nid;
 
 	/* We attempt to distribute budget to each Rx queue fairly, but don't
 	 * allow the budget to go below 1 because that would exit polling early.
@@ -4037,8 +4112,10 @@ static bool idpf_rx_splitq_clean_all(struct idpf_q_vector *q_vec, int budget,
 	}
 	*cleaned = pkts_cleaned;
 
+	nid = numa_mem_id();
+
 	for (i = 0; i < q_vec->num_bufq; i++)
-		idpf_rx_clean_refillq_all(q_vec->bufq[i]);
+		idpf_rx_clean_refillq_all(q_vec->bufq[i], nid);
 
 	return clean_complete;
 }
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
index 98d47728cc8c..47d70ad0d6b4 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
@@ -7,7 +7,7 @@
 #include <linux/dim.h>
 
 #include <net/libeth/cache.h>
-#include <net/page_pool/helpers.h>
+#include <net/libeth/rx.h>
 #include <net/tcp.h>
 #include <net/netdev_queues.h>
 
@@ -103,8 +103,6 @@ do {								\
 #define IDPF_RX_BUF_STRIDE			32
 #define IDPF_RX_BUF_POST_STRIDE			16
 #define IDPF_LOW_WATERMARK			64
-/* Size of header buffer specifically for header split */
-#define IDPF_HDR_BUF_SIZE			256
 #define IDPF_PACKET_HDR_PAD	\
 	(ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN * 2)
 #define IDPF_TX_TSO_MIN_MSS			88
@@ -300,14 +298,7 @@ struct idpf_rx_extracted {
 #define IDPF_TX_MAX_DESC_DATA_ALIGNED \
 	ALIGN_DOWN(IDPF_TX_MAX_DESC_DATA, IDPF_TX_MAX_READ_REQ_SIZE)
 
-#define IDPF_RX_DMA_ATTR \
-	(DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING)
-
-struct idpf_rx_buf {
-	struct page *page;
-	unsigned int page_offset;
-	u16 truesize;
-};
+#define idpf_rx_buf libeth_fqe
 
 #define IDPF_RX_MAX_PTYPE_PROTO_IDS    32
 #define IDPF_RX_MAX_PTYPE_SZ	(sizeof(struct virtchnl2_ptype) + \
@@ -751,14 +742,14 @@ libeth_cacheline_set_assert(struct idpf_tx_queue, 64,
 /**
  * struct idpf_buf_queue - software structure representing a buffer queue
  * @split_buf: buffer descriptor array
- * @rx_buf: Struct with RX buffer related members
- * @rx_buf.buf: See struct idpf_rx_buf
- * @rx_buf.hdr_buf_pa: DMA handle
- * @rx_buf.hdr_buf_va: Virtual address
- * @pp: Page pool pointer
+ * @hdr_buf: &libeth_fqe for header buffers
+ * @hdr_pp: &page_pool for header buffers
+ * @buf: &idpf_rx_buf for data buffers
+ * @pp: &page_pool for data buffers
  * @tail: Tail offset
  * @flags: See enum idpf_queue_flags_t
  * @desc_count: Number of descriptors
+ * @hdr_truesize: truesize for buffer headers
  * @next_to_use: Next descriptor to use
  * @next_to_clean: Next descriptor to clean
  * @next_to_alloc: RX buffer to allocate at
@@ -773,16 +764,16 @@ libeth_cacheline_set_assert(struct idpf_tx_queue, 64,
 struct idpf_buf_queue {
 	__cacheline_group_begin_aligned(read_mostly);
 	struct virtchnl2_splitq_rx_buf_desc *split_buf;
-	struct {
-		struct idpf_rx_buf *buf;
-		dma_addr_t hdr_buf_pa;
-		void *hdr_buf_va;
-	} rx_buf;
+	struct libeth_fqe *hdr_buf;
+	struct page_pool *hdr_pp;
+	struct idpf_rx_buf *buf;
 	struct page_pool *pp;
 	void __iomem *tail;
 
 	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
 	u32 desc_count;
+
+	u32 hdr_truesize;
 	__cacheline_group_end_aligned(read_mostly);
 
 	__cacheline_group_begin_aligned(read_write);
@@ -976,6 +967,18 @@ struct idpf_txq_group {
 	u32 num_completions_pending;
 };
 
+static inline int idpf_q_vector_to_mem(const struct idpf_q_vector *q_vector)
+{
+	u32 cpu;
+
+	if (!q_vector)
+		return NUMA_NO_NODE;
+
+	cpu = cpumask_first(q_vector->affinity_mask);
+
+	return cpu < nr_cpu_ids ? cpu_to_mem(cpu) : NUMA_NO_NODE;
+}
+
 /**
  * idpf_size_to_txd_count - Get number of descriptors needed for large Tx frag
  * @size: transmit request size in bytes
@@ -1044,7 +1047,7 @@ static inline dma_addr_t idpf_alloc_page(struct page_pool *pool,
 					 unsigned int buf_size)
 {
 	if (buf_size == IDPF_RX_BUF_2048)
-		buf->page = page_pool_dev_alloc_frag(pool, &buf->page_offset,
+		buf->page = page_pool_dev_alloc_frag(pool, &buf->offset,
 						     buf_size);
 	else
 		buf->page = page_pool_dev_alloc_pages(pool);
@@ -1054,7 +1057,7 @@ static inline dma_addr_t idpf_alloc_page(struct page_pool *pool,
 
 	buf->truesize = buf_size;
 
-	return page_pool_get_dma_addr(buf->page) + buf->page_offset +
+	return page_pool_get_dma_addr(buf->page) + buf->offset +
 	       pool->p.offset;
 }
 
@@ -1081,7 +1084,7 @@ static inline void idpf_rx_sync_for_cpu(struct idpf_rx_buf *rx_buf, u32 len)
 
 	dma_sync_single_range_for_cpu(pp->p.dev,
 				      page_pool_get_dma_addr(page),
-				      rx_buf->page_offset + pp->p.offset, len,
+				      rx_buf->offset + pp->p.offset, len,
 				      page_pool_get_dma_dir(pp));
 }
 
@@ -1110,6 +1113,7 @@ void idpf_rx_add_frag(struct idpf_rx_buf *rx_buf, struct sk_buff *skb,
 struct sk_buff *idpf_rx_construct_skb(const struct idpf_rx_queue *rxq,
 				      struct idpf_rx_buf *rx_buf,
 				      unsigned int size);
+struct sk_buff *idpf_rx_build_skb(const struct libeth_fqe *buf, u32 size);
 void idpf_tx_buf_hw_update(struct idpf_tx_queue *tx_q, u32 val,
 			   bool xmit_more);
 unsigned int idpf_size_to_txd_count(unsigned int size);
diff --git a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
index 83f3543a30e1..50dcb3ab02b1 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
@@ -1604,32 +1604,38 @@ setup_rxqs:
 			num_rxq = rx_qgrp->singleq.num_rxq;
 
 		for (j = 0; j < num_rxq; j++, k++) {
+			const struct idpf_bufq_set *sets;
 			struct idpf_rx_queue *rxq;
 
 			if (!idpf_is_queue_model_split(vport->rxq_model)) {
 				rxq = rx_qgrp->singleq.rxqs[j];
 				goto common_qi_fields;
 			}
+
 			rxq = &rx_qgrp->splitq.rxq_sets[j]->rxq;
-			qi[k].rx_bufq1_id =
-			  cpu_to_le16(rxq->bufq_sets[0].bufq.q_id);
+			sets = rxq->bufq_sets;
+
+			qi[k].rx_bufq1_id = cpu_to_le16(sets[0].bufq.q_id);
 			if (vport->num_bufqs_per_qgrp > IDPF_SINGLE_BUFQ_PER_RXQ_GRP) {
 				qi[k].bufq2_ena = IDPF_BUFQ2_ENA;
 				qi[k].rx_bufq2_id =
-				  cpu_to_le16(rxq->bufq_sets[1].bufq.q_id);
+					cpu_to_le16(sets[1].bufq.q_id);
 			}
 			qi[k].rx_buffer_low_watermark =
 				cpu_to_le16(rxq->rx_buffer_low_watermark);
 			if (idpf_is_feature_ena(vport, NETIF_F_GRO_HW))
 				qi[k].qflags |= cpu_to_le16(VIRTCHNL2_RXQ_RSC);
 
-common_qi_fields:
+			rxq->rx_hbuf_size = sets[0].bufq.rx_hbuf_size;
+
 			if (idpf_queue_has(HSPLIT_EN, rxq)) {
 				qi[k].qflags |=
 					cpu_to_le16(VIRTCHNL2_RXQ_HDR_SPLIT);
 				qi[k].hdr_buffer_size =
 					cpu_to_le16(rxq->rx_hbuf_size);
 			}
+
+common_qi_fields:
 			qi[k].queue_id = cpu_to_le32(rxq->q_id);
 			qi[k].model = cpu_to_le16(vport->rxq_model);
 			qi[k].type = cpu_to_le32(VIRTCHNL2_QUEUE_TYPE_RX);
-- 
cgit v1.2.3-58-ga151


From 74d1412ac8f3719bc8dd08b5775276dd1cbc3997 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Thu, 20 Jun 2024 15:53:47 +0200
Subject: idpf: use libeth Rx buffer management for payload buffer

idpf uses Page Pool for data buffers with hardcoded buffer lengths of
4k for "classic" buffers and 2k for "short" ones. This is not flexible
and does not ensure optimal memory usage. Why would you need 4k buffers
when the MTU is 1500?
Use libeth for the data buffers and don't hardcode any buffer sizes. Let
them be calculated from the MTU for "classics" and then divide the
truesize by 2 for "short" ones. The memory usage is now greatly reduced
and 2 buffer queues starts make sense: on frames <= 1024, you'll recycle
(and resync) a page only after 4 HW writes rather than two.

Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/idpf/Kconfig            |   1 -
 drivers/net/ethernet/intel/idpf/idpf.h             |   2 -
 .../net/ethernet/intel/idpf/idpf_singleq_txrx.c    |  27 +--
 drivers/net/ethernet/intel/idpf/idpf_txrx.c        | 242 ++++++++-------------
 drivers/net/ethernet/intel/idpf/idpf_txrx.h        |  86 ++------
 drivers/net/ethernet/intel/idpf/idpf_virtchnl.c    |   8 +-
 6 files changed, 120 insertions(+), 246 deletions(-)

diff --git a/drivers/net/ethernet/intel/idpf/Kconfig b/drivers/net/ethernet/intel/idpf/Kconfig
index 1f071143d992..1addd663acad 100644
--- a/drivers/net/ethernet/intel/idpf/Kconfig
+++ b/drivers/net/ethernet/intel/idpf/Kconfig
@@ -6,7 +6,6 @@ config IDPF
 	depends on PCI_MSI
 	select DIMLIB
 	select LIBETH
-	select PAGE_POOL
 	help
 	  This driver supports Intel(R) Infrastructure Data Path Function
 	  devices.
diff --git a/drivers/net/ethernet/intel/idpf/idpf.h b/drivers/net/ethernet/intel/idpf/idpf.h
index 078340a01757..2c31ad87587a 100644
--- a/drivers/net/ethernet/intel/idpf/idpf.h
+++ b/drivers/net/ethernet/intel/idpf/idpf.h
@@ -264,7 +264,6 @@ struct idpf_port_stats {
  *		    the worst case.
  * @num_bufqs_per_qgrp: Buffer queues per RX queue in a given grouping
  * @bufq_desc_count: Buffer queue descriptor count
- * @bufq_size: Size of buffers in ring (e.g. 2K, 4K, etc)
  * @num_rxq_grp: Number of RX queues in a group
  * @rxq_grps: Total number of RX groups. Number of groups * number of RX per
  *	      group will yield total number of RX queues.
@@ -308,7 +307,6 @@ struct idpf_vport {
 	u32 rxq_desc_count;
 	u8 num_bufqs_per_qgrp;
 	u32 bufq_desc_count[IDPF_MAX_BUFQS_PER_RXQ_GRP];
-	u32 bufq_size[IDPF_MAX_BUFQS_PER_RXQ_GRP];
 	u16 num_rxq_grp;
 	struct idpf_rxq_group *rxq_grps;
 	u32 rxq_model;
diff --git a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
index ced1fdf7a1fb..fe64febf7436 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_singleq_txrx.c
@@ -857,20 +857,24 @@ bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_rx_queue *rx_q,
 				      u16 cleaned_count)
 {
 	struct virtchnl2_singleq_rx_buf_desc *desc;
+	const struct libeth_fq_fp fq = {
+		.pp		= rx_q->pp,
+		.fqes		= rx_q->rx_buf,
+		.truesize	= rx_q->truesize,
+		.count		= rx_q->desc_count,
+	};
 	u16 nta = rx_q->next_to_alloc;
-	struct idpf_rx_buf *buf;
 
 	if (!cleaned_count)
 		return false;
 
 	desc = &rx_q->single_buf[nta];
-	buf = &rx_q->rx_buf[nta];
 
 	do {
 		dma_addr_t addr;
 
-		addr = idpf_alloc_page(rx_q->pp, buf, rx_q->rx_buf_size);
-		if (unlikely(addr == DMA_MAPPING_ERROR))
+		addr = libeth_rx_alloc(&fq, nta);
+		if (addr == DMA_MAPPING_ERROR)
 			break;
 
 		/* Refresh the desc even if buffer_addrs didn't change
@@ -880,11 +884,9 @@ bool idpf_rx_singleq_buf_hw_alloc_all(struct idpf_rx_queue *rx_q,
 		desc->hdr_addr = 0;
 		desc++;
 
-		buf++;
 		nta++;
 		if (unlikely(nta == rx_q->desc_count)) {
 			desc = &rx_q->single_buf[0];
-			buf = rx_q->rx_buf;
 			nta = 0;
 		}
 
@@ -1004,28 +1006,26 @@ static int idpf_rx_singleq_clean(struct idpf_rx_queue *rx_q, int budget)
 		idpf_rx_singleq_extract_fields(rx_q, rx_desc, &fields);
 
 		rx_buf = &rx_q->rx_buf[ntc];
-		if (!fields.size) {
-			idpf_rx_put_page(rx_buf);
+		if (!libeth_rx_sync_for_cpu(rx_buf, fields.size))
 			goto skip_data;
-		}
 
-		idpf_rx_sync_for_cpu(rx_buf, fields.size);
 		if (skb)
 			idpf_rx_add_frag(rx_buf, skb, fields.size);
 		else
-			skb = idpf_rx_construct_skb(rx_q, rx_buf, fields.size);
+			skb = idpf_rx_build_skb(rx_buf, fields.size);
 
 		/* exit if we failed to retrieve a buffer */
 		if (!skb)
 			break;
 
 skip_data:
-		IDPF_SINGLEQ_BUMP_RING_IDX(rx_q, ntc);
+		rx_buf->page = NULL;
 
+		IDPF_SINGLEQ_BUMP_RING_IDX(rx_q, ntc);
 		cleaned_count++;
 
 		/* skip if it is non EOP desc */
-		if (idpf_rx_singleq_is_non_eop(rx_desc))
+		if (idpf_rx_singleq_is_non_eop(rx_desc) || unlikely(!skb))
 			continue;
 
 #define IDPF_RXD_ERR_S FIELD_PREP(VIRTCHNL2_RX_BASE_DESC_QW1_ERROR_M, \
@@ -1062,6 +1062,7 @@ skip_data:
 
 	rx_q->next_to_clean = ntc;
 
+	page_pool_nid_changed(rx_q->pp, numa_mem_id());
 	if (cleaned_count)
 		failure = idpf_rx_singleq_buf_hw_alloc_all(rx_q, cleaned_count);
 
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
index ec4a9501d19f..af2879f03b8d 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
@@ -417,6 +417,11 @@ static void idpf_rx_hdr_buf_rel_all(struct idpf_buf_queue *bufq)
  */
 static void idpf_rx_buf_rel_bufq(struct idpf_buf_queue *bufq)
 {
+	struct libeth_fq fq = {
+		.fqes	= bufq->buf,
+		.pp	= bufq->pp,
+	};
+
 	/* queue already cleared, nothing to do */
 	if (!bufq->buf)
 		return;
@@ -428,11 +433,9 @@ static void idpf_rx_buf_rel_bufq(struct idpf_buf_queue *bufq)
 	if (idpf_queue_has(HSPLIT_EN, bufq))
 		idpf_rx_hdr_buf_rel_all(bufq);
 
-	page_pool_destroy(bufq->pp);
-	bufq->pp = NULL;
-
-	kfree(bufq->buf);
+	libeth_rx_fq_destroy(&fq);
 	bufq->buf = NULL;
+	bufq->pp = NULL;
 }
 
 /**
@@ -441,17 +444,20 @@ static void idpf_rx_buf_rel_bufq(struct idpf_buf_queue *bufq)
  */
 static void idpf_rx_buf_rel_all(struct idpf_rx_queue *rxq)
 {
+	struct libeth_fq fq = {
+		.fqes	= rxq->rx_buf,
+		.pp	= rxq->pp,
+	};
+
 	if (!rxq->rx_buf)
 		return;
 
 	for (u32 i = 0; i < rxq->desc_count; i++)
 		idpf_rx_page_rel(&rxq->rx_buf[i]);
 
-	page_pool_destroy(rxq->pp);
-	rxq->pp = NULL;
-
-	kfree(rxq->rx_buf);
+	libeth_rx_fq_destroy(&fq);
 	rxq->rx_buf = NULL;
+	rxq->pp = NULL;
 }
 
 /**
@@ -633,11 +639,9 @@ static bool idpf_rx_post_buf_desc(struct idpf_buf_queue *bufq, u16 buf_id)
 		.count	= bufq->desc_count,
 	};
 	u16 nta = bufq->next_to_alloc;
-	struct idpf_rx_buf *buf;
 	dma_addr_t addr;
 
 	splitq_rx_desc = &bufq->split_buf[nta];
-	buf = &bufq->buf[buf_id];
 
 	if (idpf_queue_has(HSPLIT_EN, bufq)) {
 		fq.pp = bufq->hdr_pp;
@@ -651,8 +655,12 @@ static bool idpf_rx_post_buf_desc(struct idpf_buf_queue *bufq, u16 buf_id)
 		splitq_rx_desc->hdr_addr = cpu_to_le64(addr);
 	}
 
-	addr = idpf_alloc_page(bufq->pp, buf, bufq->rx_buf_size);
-	if (unlikely(addr == DMA_MAPPING_ERROR))
+	fq.pp = bufq->pp;
+	fq.fqes = bufq->buf;
+	fq.truesize = bufq->truesize;
+
+	addr = libeth_rx_alloc(&fq, buf_id);
+	if (addr == DMA_MAPPING_ERROR)
 		return false;
 
 	splitq_rx_desc->pkt_addr = cpu_to_le64(addr);
@@ -689,30 +697,6 @@ static bool idpf_rx_post_init_bufs(struct idpf_buf_queue *bufq,
 	return true;
 }
 
-/**
- * idpf_rx_create_page_pool - Create a page pool
- * @napi: NAPI of the associated queue vector
- * @count: queue descriptor count
- *
- * Returns &page_pool on success, casted -errno on failure
- */
-static struct page_pool *idpf_rx_create_page_pool(struct napi_struct *napi,
-						  u32 count)
-{
-	struct page_pool_params pp = {
-		.flags		= PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV,
-		.order		= 0,
-		.pool_size	= count,
-		.nid		= NUMA_NO_NODE,
-		.dev		= napi->dev->dev.parent,
-		.max_len	= PAGE_SIZE,
-		.dma_dir	= DMA_FROM_DEVICE,
-		.offset		= 0,
-	};
-
-	return page_pool_create(&pp);
-}
-
 /**
  * idpf_rx_buf_alloc_singleq - Allocate memory for all buffer resources
  * @rxq: queue for which the buffers are allocated
@@ -721,11 +705,6 @@ static struct page_pool *idpf_rx_create_page_pool(struct napi_struct *napi,
  */
 static int idpf_rx_buf_alloc_singleq(struct idpf_rx_queue *rxq)
 {
-	rxq->rx_buf = kcalloc(rxq->desc_count, sizeof(*rxq->rx_buf),
-			      GFP_KERNEL);
-	if (!rxq->rx_buf)
-		return -ENOMEM;
-
 	if (idpf_rx_singleq_buf_hw_alloc_all(rxq, rxq->desc_count - 1))
 		goto err;
 
@@ -745,13 +724,21 @@ err:
  */
 static int idpf_rx_bufs_init_singleq(struct idpf_rx_queue *rxq)
 {
-	struct page_pool *pool;
+	struct libeth_fq fq = {
+		.count	= rxq->desc_count,
+		.type	= LIBETH_FQE_MTU,
+		.nid	= idpf_q_vector_to_mem(rxq->q_vector),
+	};
+	int ret;
 
-	pool = idpf_rx_create_page_pool(&rxq->q_vector->napi, rxq->desc_count);
-	if (IS_ERR(pool))
-		return PTR_ERR(pool);
+	ret = libeth_rx_fq_create(&fq, &rxq->q_vector->napi);
+	if (ret)
+		return ret;
 
-	rxq->pp = pool;
+	rxq->pp = fq.pp;
+	rxq->rx_buf = fq.fqes;
+	rxq->truesize = fq.truesize;
+	rxq->rx_buf_size = fq.buf_len;
 
 	return idpf_rx_buf_alloc_singleq(rxq);
 }
@@ -766,14 +753,6 @@ static int idpf_rx_buf_alloc_all(struct idpf_buf_queue *rxbufq)
 {
 	int err = 0;
 
-	/* Allocate book keeping buffers */
-	rxbufq->buf = kcalloc(rxbufq->desc_count, sizeof(*rxbufq->buf),
-			      GFP_KERNEL);
-	if (!rxbufq->buf) {
-		err = -ENOMEM;
-		goto rx_buf_alloc_all_out;
-	}
-
 	if (idpf_queue_has(HSPLIT_EN, rxbufq)) {
 		err = idpf_rx_hdr_buf_alloc_all(rxbufq);
 		if (err)
@@ -794,19 +773,30 @@ rx_buf_alloc_all_out:
 /**
  * idpf_rx_bufs_init - Initialize page pool, allocate rx bufs, and post to HW
  * @bufq: buffer queue to create page pool for
+ * @type: type of Rx buffers to allocate
  *
  * Returns 0 on success, negative on failure
  */
-static int idpf_rx_bufs_init(struct idpf_buf_queue *bufq)
+static int idpf_rx_bufs_init(struct idpf_buf_queue *bufq,
+			     enum libeth_fqe_type type)
 {
-	struct page_pool *pool;
+	struct libeth_fq fq = {
+		.truesize	= bufq->truesize,
+		.count		= bufq->desc_count,
+		.type		= type,
+		.hsplit		= idpf_queue_has(HSPLIT_EN, bufq),
+		.nid		= idpf_q_vector_to_mem(bufq->q_vector),
+	};
+	int ret;
 
-	pool = idpf_rx_create_page_pool(&bufq->q_vector->napi,
-					bufq->desc_count);
-	if (IS_ERR(pool))
-		return PTR_ERR(pool);
+	ret = libeth_rx_fq_create(&fq, &bufq->q_vector->napi);
+	if (ret)
+		return ret;
 
-	bufq->pp = pool;
+	bufq->pp = fq.pp;
+	bufq->buf = fq.fqes;
+	bufq->truesize = fq.truesize;
+	bufq->rx_buf_size = fq.buf_len;
 
 	return idpf_rx_buf_alloc_all(bufq);
 }
@@ -819,14 +809,15 @@ static int idpf_rx_bufs_init(struct idpf_buf_queue *bufq)
  */
 int idpf_rx_bufs_init_all(struct idpf_vport *vport)
 {
-	struct idpf_rxq_group *rx_qgrp;
+	bool split = idpf_is_queue_model_split(vport->rxq_model);
 	int i, j, err;
 
 	for (i = 0; i < vport->num_rxq_grp; i++) {
-		rx_qgrp = &vport->rxq_grps[i];
+		struct idpf_rxq_group *rx_qgrp = &vport->rxq_grps[i];
+		u32 truesize = 0;
 
 		/* Allocate bufs for the rxq itself in singleq */
-		if (!idpf_is_queue_model_split(vport->rxq_model)) {
+		if (!split) {
 			int num_rxq = rx_qgrp->singleq.num_rxq;
 
 			for (j = 0; j < num_rxq; j++) {
@@ -843,12 +834,19 @@ int idpf_rx_bufs_init_all(struct idpf_vport *vport)
 
 		/* Otherwise, allocate bufs for the buffer queues */
 		for (j = 0; j < vport->num_bufqs_per_qgrp; j++) {
+			enum libeth_fqe_type type;
 			struct idpf_buf_queue *q;
 
 			q = &rx_qgrp->splitq.bufq_sets[j].bufq;
-			err = idpf_rx_bufs_init(q);
+			q->truesize = truesize;
+
+			type = truesize ? LIBETH_FQE_SHORT : LIBETH_FQE_MTU;
+
+			err = idpf_rx_bufs_init(q, type);
 			if (err)
 				return err;
+
+			truesize = q->truesize >> 1;
 		}
 	}
 
@@ -1160,17 +1158,11 @@ void idpf_vport_init_num_qs(struct idpf_vport *vport,
 	/* Adjust number of buffer queues per Rx queue group. */
 	if (!idpf_is_queue_model_split(vport->rxq_model)) {
 		vport->num_bufqs_per_qgrp = 0;
-		vport->bufq_size[0] = IDPF_RX_BUF_2048;
 
 		return;
 	}
 
 	vport->num_bufqs_per_qgrp = IDPF_MAX_BUFQS_PER_RXQ_GRP;
-	/* Bufq[0] default buffer size is 4K
-	 * Bufq[1] default buffer size is 2K
-	 */
-	vport->bufq_size[0] = IDPF_RX_BUF_4096;
-	vport->bufq_size[1] = IDPF_RX_BUF_2048;
 }
 
 /**
@@ -1507,7 +1499,6 @@ static int idpf_rxq_group_alloc(struct idpf_vport *vport, u16 num_rxq)
 
 			q = &rx_qgrp->splitq.bufq_sets[j].bufq;
 			q->desc_count = vport->bufq_desc_count[j];
-			q->rx_buf_size = vport->bufq_size[j];
 			q->rx_buffer_low_watermark = IDPF_LOW_WATERMARK;
 
 			idpf_queue_assign(HSPLIT_EN, q, hs);
@@ -1560,14 +1551,9 @@ setup_rxq:
 			q->netdev = vport->netdev;
 			q->bufq_sets = rx_qgrp->splitq.bufq_sets;
 			q->idx = (i * num_rxq) + j;
-			/* In splitq mode, RXQ buffer size should be
-			 * set to that of the first buffer queue
-			 * associated with this RXQ
-			 */
-			q->rx_buf_size = vport->bufq_size[0];
 			q->rx_buffer_low_watermark = IDPF_LOW_WATERMARK;
 			q->rx_max_pkt_size = vport->netdev->mtu +
-							IDPF_PACKET_HDR_PAD;
+							LIBETH_RX_LL_LEN;
 			idpf_rxq_set_descids(vport, q);
 		}
 	}
@@ -3145,69 +3131,10 @@ idpf_rx_process_skb_fields(struct idpf_rx_queue *rxq, struct sk_buff *skb,
 void idpf_rx_add_frag(struct idpf_rx_buf *rx_buf, struct sk_buff *skb,
 		      unsigned int size)
 {
-	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buf->page,
-			rx_buf->offset, size, rx_buf->truesize);
-
-	rx_buf->page = NULL;
-}
-
-/**
- * idpf_rx_construct_skb - Allocate skb and populate it
- * @rxq: Rx descriptor queue
- * @rx_buf: Rx buffer to pull data from
- * @size: the length of the packet
- *
- * This function allocates an skb. It then populates it with the page
- * data from the current receive descriptor, taking care to set up the
- * skb correctly.
- */
-struct sk_buff *idpf_rx_construct_skb(const struct idpf_rx_queue *rxq,
-				      struct idpf_rx_buf *rx_buf,
-				      unsigned int size)
-{
-	unsigned int headlen;
-	struct sk_buff *skb;
-	void *va;
-
-	va = page_address(rx_buf->page) + rx_buf->offset;
-
-	/* prefetch first cache line of first page */
-	net_prefetch(va);
-	/* allocate a skb to store the frags */
-	skb = napi_alloc_skb(rxq->napi, IDPF_RX_HDR_SIZE);
-	if (unlikely(!skb)) {
-		idpf_rx_put_page(rx_buf);
-
-		return NULL;
-	}
+	u32 hr = rx_buf->page->pp->p.offset;
 
-	skb_mark_for_recycle(skb);
-
-	/* Determine available headroom for copy */
-	headlen = size;
-	if (headlen > IDPF_RX_HDR_SIZE)
-		headlen = eth_get_headlen(skb->dev, va, IDPF_RX_HDR_SIZE);
-
-	/* align pull length to size of long to optimize memcpy performance */
-	memcpy(__skb_put(skb, headlen), va, ALIGN(headlen, sizeof(long)));
-
-	/* if we exhaust the linear part then add what is left as a frag */
-	size -= headlen;
-	if (!size) {
-		idpf_rx_put_page(rx_buf);
-
-		return skb;
-	}
-
-	skb_add_rx_frag(skb, 0, rx_buf->page, rx_buf->offset + headlen,
-			size, rx_buf->truesize);
-
-	/* Since we're giving the page to the stack, clear our reference to it.
-	 * We'll get a new one during buffer posting.
-	 */
-	rx_buf->page = NULL;
-
-	return skb;
+	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buf->page,
+			rx_buf->offset + hr, size, rx_buf->truesize);
 }
 
 /**
@@ -3413,26 +3340,26 @@ static int idpf_rx_splitq_clean(struct idpf_rx_queue *rxq, int budget)
 		hdr->page = NULL;
 
 payload:
-		if (pkt_len) {
-			idpf_rx_sync_for_cpu(rx_buf, pkt_len);
-			if (skb)
-				idpf_rx_add_frag(rx_buf, skb, pkt_len);
-			else
-				skb = idpf_rx_construct_skb(rxq, rx_buf,
-							    pkt_len);
-		} else {
-			idpf_rx_put_page(rx_buf);
-		}
+		if (!libeth_rx_sync_for_cpu(rx_buf, pkt_len))
+			goto skip_data;
+
+		if (skb)
+			idpf_rx_add_frag(rx_buf, skb, pkt_len);
+		else
+			skb = idpf_rx_build_skb(rx_buf, pkt_len);
 
 		/* exit if we failed to retrieve a buffer */
 		if (!skb)
 			break;
 
-		idpf_rx_post_buf_refill(refillq, buf_id);
+skip_data:
+		rx_buf->page = NULL;
 
+		idpf_rx_post_buf_refill(refillq, buf_id);
 		IDPF_RX_BUMP_NTC(rxq, ntc);
+
 		/* skip if it is non EOP desc */
-		if (!idpf_rx_splitq_is_eop(rx_desc))
+		if (!idpf_rx_splitq_is_eop(rx_desc) || unlikely(!skb))
 			continue;
 
 		/* pad skb if needed (to make valid ethernet frame) */
@@ -3483,15 +3410,15 @@ static int idpf_rx_update_bufq_desc(struct idpf_buf_queue *bufq, u32 buf_id,
 				    struct virtchnl2_splitq_rx_buf_desc *buf_desc)
 {
 	struct libeth_fq_fp fq = {
+		.pp		= bufq->pp,
+		.fqes		= bufq->buf,
+		.truesize	= bufq->truesize,
 		.count		= bufq->desc_count,
 	};
-	struct idpf_rx_buf *buf;
 	dma_addr_t addr;
 
-	buf = &bufq->buf[buf_id];
-
-	addr = idpf_alloc_page(bufq->pp, buf, bufq->rx_buf_size);
-	if (unlikely(addr == DMA_MAPPING_ERROR))
+	addr = libeth_rx_alloc(&fq, buf_id);
+	if (addr == DMA_MAPPING_ERROR)
 		return -ENOMEM;
 
 	buf_desc->pkt_addr = cpu_to_le64(addr);
@@ -3590,6 +3517,7 @@ static void idpf_rx_clean_refillq_all(struct idpf_buf_queue *bufq, int nid)
 	struct idpf_bufq_set *bufq_set;
 	int i;
 
+	page_pool_nid_changed(bufq->pp, nid);
 	if (bufq->hdr_pp)
 		page_pool_nid_changed(bufq->hdr_pp, nid);
 
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
index 47d70ad0d6b4..6215dbee5546 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
@@ -7,7 +7,6 @@
 #include <linux/dim.h>
 
 #include <net/libeth/cache.h>
-#include <net/libeth/rx.h>
 #include <net/tcp.h>
 #include <net/netdev_queues.h>
 
@@ -97,14 +96,10 @@ do {								\
 		idx = 0;					\
 } while (0)
 
-#define IDPF_RX_HDR_SIZE			256
-#define IDPF_RX_BUF_2048			2048
-#define IDPF_RX_BUF_4096			4096
 #define IDPF_RX_BUF_STRIDE			32
 #define IDPF_RX_BUF_POST_STRIDE			16
 #define IDPF_LOW_WATERMARK			64
-#define IDPF_PACKET_HDR_PAD	\
-	(ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN * 2)
+
 #define IDPF_TX_TSO_MIN_MSS			88
 
 /* Minimum number of descriptors between 2 descriptors with the RE bit set;
@@ -538,7 +533,7 @@ struct idpf_txq_stash {
  * @desc_ring: virtual descriptor ring address
  * @bufq_sets: Pointer to the array of buffer queues in splitq mode
  * @napi: NAPI instance corresponding to this queue (splitq)
- * @rx_buf: See struct idpf_rx_buf
+ * @rx_buf: See struct &libeth_fqe
  * @pp: Page pool pointer in singleq mode
  * @netdev: &net_device corresponding to this queue
  * @tail: Tail offset. Used for both queue models single and split.
@@ -552,6 +547,7 @@ struct idpf_txq_stash {
  * @next_to_clean: Next descriptor to clean
  * @next_to_alloc: RX buffer to allocate at
  * @skb: Pointer to the skb
+ * @truesize: data buffer truesize in singleq
  * @stats_sync: See struct u64_stats_sync
  * @q_stats: See union idpf_rx_queue_stats
  * @q_id: Queue id
@@ -577,7 +573,7 @@ struct idpf_rx_queue {
 			struct napi_struct *napi;
 		};
 		struct {
-			struct idpf_rx_buf *rx_buf;
+			struct libeth_fqe *rx_buf;
 			struct page_pool *pp;
 		};
 	};
@@ -598,6 +594,7 @@ struct idpf_rx_queue {
 	u16 next_to_alloc;
 
 	struct sk_buff *skb;
+	u32 truesize;
 
 	struct u64_stats_sync stats_sync;
 	struct idpf_rx_queue_stats q_stats;
@@ -617,7 +614,7 @@ struct idpf_rx_queue {
 	__cacheline_group_end_aligned(cold);
 };
 libeth_cacheline_set_assert(struct idpf_rx_queue, 64,
-			    72 + sizeof(struct u64_stats_sync),
+			    80 + sizeof(struct u64_stats_sync),
 			    32);
 
 /**
@@ -744,15 +741,16 @@ libeth_cacheline_set_assert(struct idpf_tx_queue, 64,
  * @split_buf: buffer descriptor array
  * @hdr_buf: &libeth_fqe for header buffers
  * @hdr_pp: &page_pool for header buffers
- * @buf: &idpf_rx_buf for data buffers
+ * @buf: &libeth_fqe for data buffers
  * @pp: &page_pool for data buffers
  * @tail: Tail offset
  * @flags: See enum idpf_queue_flags_t
  * @desc_count: Number of descriptors
- * @hdr_truesize: truesize for buffer headers
  * @next_to_use: Next descriptor to use
  * @next_to_clean: Next descriptor to clean
  * @next_to_alloc: RX buffer to allocate at
+ * @hdr_truesize: truesize for buffer headers
+ * @truesize: truesize for data buffers
  * @q_id: Queue id
  * @size: Length of descriptor ring in bytes
  * @dma: Physical address of ring
@@ -766,20 +764,21 @@ struct idpf_buf_queue {
 	struct virtchnl2_splitq_rx_buf_desc *split_buf;
 	struct libeth_fqe *hdr_buf;
 	struct page_pool *hdr_pp;
-	struct idpf_rx_buf *buf;
+	struct libeth_fqe *buf;
 	struct page_pool *pp;
 	void __iomem *tail;
 
 	DECLARE_BITMAP(flags, __IDPF_Q_FLAGS_NBITS);
 	u32 desc_count;
-
-	u32 hdr_truesize;
 	__cacheline_group_end_aligned(read_mostly);
 
 	__cacheline_group_begin_aligned(read_write);
 	u32 next_to_use;
 	u32 next_to_clean;
 	u32 next_to_alloc;
+
+	u32 hdr_truesize;
+	u32 truesize;
 	__cacheline_group_end_aligned(read_write);
 
 	__cacheline_group_begin_aligned(cold);
@@ -794,7 +793,7 @@ struct idpf_buf_queue {
 	u16 rx_buf_size;
 	__cacheline_group_end_aligned(cold);
 };
-libeth_cacheline_set_assert(struct idpf_buf_queue, 64, 16, 32);
+libeth_cacheline_set_assert(struct idpf_buf_queue, 64, 24, 32);
 
 /**
  * struct idpf_compl_queue - software structure representing a completion queue
@@ -1034,60 +1033,6 @@ static inline void idpf_tx_splitq_build_desc(union idpf_tx_flex_desc *desc,
 		idpf_tx_splitq_build_flow_desc(desc, params, td_cmd, size);
 }
 
-/**
- * idpf_alloc_page - Allocate a new RX buffer from the page pool
- * @pool: page_pool to allocate from
- * @buf: metadata struct to populate with page info
- * @buf_size: 2K or 4K
- *
- * Returns &dma_addr_t to be passed to HW for Rx, %DMA_MAPPING_ERROR otherwise.
- */
-static inline dma_addr_t idpf_alloc_page(struct page_pool *pool,
-					 struct idpf_rx_buf *buf,
-					 unsigned int buf_size)
-{
-	if (buf_size == IDPF_RX_BUF_2048)
-		buf->page = page_pool_dev_alloc_frag(pool, &buf->offset,
-						     buf_size);
-	else
-		buf->page = page_pool_dev_alloc_pages(pool);
-
-	if (!buf->page)
-		return DMA_MAPPING_ERROR;
-
-	buf->truesize = buf_size;
-
-	return page_pool_get_dma_addr(buf->page) + buf->offset +
-	       pool->p.offset;
-}
-
-/**
- * idpf_rx_put_page - Return RX buffer page to pool
- * @rx_buf: RX buffer metadata struct
- */
-static inline void idpf_rx_put_page(struct idpf_rx_buf *rx_buf)
-{
-	page_pool_put_page(rx_buf->page->pp, rx_buf->page,
-			   rx_buf->truesize, true);
-	rx_buf->page = NULL;
-}
-
-/**
- * idpf_rx_sync_for_cpu - Synchronize DMA buffer
- * @rx_buf: RX buffer metadata struct
- * @len: frame length from descriptor
- */
-static inline void idpf_rx_sync_for_cpu(struct idpf_rx_buf *rx_buf, u32 len)
-{
-	struct page *page = rx_buf->page;
-	struct page_pool *pp = page->pp;
-
-	dma_sync_single_range_for_cpu(pp->p.dev,
-				      page_pool_get_dma_addr(page),
-				      rx_buf->offset + pp->p.offset, len,
-				      page_pool_get_dma_dir(pp));
-}
-
 int idpf_vport_singleq_napi_poll(struct napi_struct *napi, int budget);
 void idpf_vport_init_num_qs(struct idpf_vport *vport,
 			    struct virtchnl2_create_vport *vport_msg);
@@ -1110,9 +1055,6 @@ void idpf_deinit_rss(struct idpf_vport *vport);
 int idpf_rx_bufs_init_all(struct idpf_vport *vport);
 void idpf_rx_add_frag(struct idpf_rx_buf *rx_buf, struct sk_buff *skb,
 		      unsigned int size);
-struct sk_buff *idpf_rx_construct_skb(const struct idpf_rx_queue *rxq,
-				      struct idpf_rx_buf *rx_buf,
-				      unsigned int size);
 struct sk_buff *idpf_rx_build_skb(const struct libeth_fqe *buf, u32 size);
 void idpf_tx_buf_hw_update(struct idpf_tx_queue *tx_q, u32 val,
 			   bool xmit_more);
diff --git a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
index 50dcb3ab02b1..70986e12da28 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c
@@ -1615,6 +1615,12 @@ setup_rxqs:
 			rxq = &rx_qgrp->splitq.rxq_sets[j]->rxq;
 			sets = rxq->bufq_sets;
 
+			/* In splitq mode, RXQ buffer size should be
+			 * set to that of the first buffer queue
+			 * associated with this RXQ.
+			 */
+			rxq->rx_buf_size = sets[0].bufq.rx_buf_size;
+
 			qi[k].rx_bufq1_id = cpu_to_le16(sets[0].bufq.q_id);
 			if (vport->num_bufqs_per_qgrp > IDPF_SINGLE_BUFQ_PER_RXQ_GRP) {
 				qi[k].bufq2_ena = IDPF_BUFQ2_ENA;
@@ -3167,7 +3173,7 @@ void idpf_vport_init(struct idpf_vport *vport, struct idpf_vport_max_q *max_q)
 	rss_data->rss_lut_size = le16_to_cpu(vport_msg->rss_lut_size);
 
 	ether_addr_copy(vport->default_mac_addr, vport_msg->default_mac_addr);
-	vport->max_mtu = le16_to_cpu(vport_msg->max_mtu) - IDPF_PACKET_HDR_PAD;
+	vport->max_mtu = le16_to_cpu(vport_msg->max_mtu) - LIBETH_RX_LL_LEN;
 
 	/* Initialize Tx and Rx profiles for Dynamic Interrupt Moderation */
 	memcpy(vport->rx_itr_profile, rx_itr, IDPF_DIM_PROFILE_SLOTS);
-- 
cgit v1.2.3-58-ga151


From eeb23b54e447ea62b247d89681f0140abab00d7f Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Wed, 10 Jul 2024 16:00:51 +0100
Subject: selftests/bpf: fix compilation failure when CONFIG_NF_FLOW_TABLE=m

In many cases, kernel netfilter functionality is built as modules.
If CONFIG_NF_FLOW_TABLE=m in particular, progs/xdp_flowtable.c
(and hence selftests) will fail to compile, so add a ___local
version of "struct flow_ports".

Fixes: c77e572d3a8c ("selftests/bpf: Add selftest for bpf_xdp_flow_lookup kfunc")
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Link: https://lore.kernel.org/r/20240710150051.192598-1-alan.maguire@oracle.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 tools/testing/selftests/bpf/progs/xdp_flowtable.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/progs/xdp_flowtable.c b/tools/testing/selftests/bpf/progs/xdp_flowtable.c
index 15209650f73b..7fdc7b23ee74 100644
--- a/tools/testing/selftests/bpf/progs/xdp_flowtable.c
+++ b/tools/testing/selftests/bpf/progs/xdp_flowtable.c
@@ -58,6 +58,10 @@ static bool xdp_flowtable_offload_check_tcp_state(void *ports, void *data_end,
 	return true;
 }
 
+struct flow_ports___local {
+	__be16 source, dest;
+} __attribute__((preserve_access_index));
+
 SEC("xdp.frags")
 int xdp_flowtable_do_lookup(struct xdp_md *ctx)
 {
@@ -69,7 +73,7 @@ int xdp_flowtable_do_lookup(struct xdp_md *ctx)
 	};
 	void *data = (void *)(long)ctx->data;
 	struct ethhdr *eth = data;
-	struct flow_ports *ports;
+	struct flow_ports___local *ports;
 	__u32 *val, key = 0;
 
 	if (eth + 1 > data_end)
@@ -79,7 +83,7 @@ int xdp_flowtable_do_lookup(struct xdp_md *ctx)
 	case bpf_htons(ETH_P_IP): {
 		struct iphdr *iph = data + sizeof(*eth);
 
-		ports = (struct flow_ports *)(iph + 1);
+		ports = (struct flow_ports___local *)(iph + 1);
 		if (ports + 1 > data_end)
 			return XDP_PASS;
 
@@ -106,7 +110,7 @@ int xdp_flowtable_do_lookup(struct xdp_md *ctx)
 		struct in6_addr *dst = (struct in6_addr *)tuple.ipv6_dst;
 		struct ipv6hdr *ip6h = data + sizeof(*eth);
 
-		ports = (struct flow_ports *)(ip6h + 1);
+		ports = (struct flow_ports___local *)(ip6h + 1);
 		if (ports + 1 > data_end)
 			return XDP_PASS;
 
-- 
cgit v1.2.3-58-ga151


From a3016a27cea8e6d10b200b9e19c19961c402d106 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Tue, 9 Jul 2024 17:16:17 +0800
Subject: selftests/bpf: Add backlog for network_helper_opts

Some callers expect __start_server() helper to pass their own "backlog"
value to listen() instead of the default of 1. So this patch adds struct
member "backlog" for network_helper_opts to allow callers to set "backlog"
value via start_server_str() helper.

listen(fd, 0 /* backlog */) can be used to enforce syncookie. Meaning
backlog 0 is a legit value.

Using 0 as a default and changing it to 1 here is fine. It makes the test
program easier to write for the common case. Enforcing syncookie mode by
using backlog 0 is a niche use case but it should at least have a way for
the caller to do that. Thus, -ve backlog value is used here for the
syncookie use case. Please see the comment in network_helpers.h for
the details.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/1660229659b66eaad07aa2126e9c9fe217eba0dd.1720515893.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/network_helpers.c |  2 +-
 tools/testing/selftests/bpf/network_helpers.h | 10 ++++++++++
 2 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index 44c2c8fa542a..e0cba4178e41 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -106,7 +106,7 @@ static int __start_server(int type, const struct sockaddr *addr, socklen_t addrl
 	}
 
 	if (type == SOCK_STREAM) {
-		if (listen(fd, 1) < 0) {
+		if (listen(fd, opts->backlog ? MAX(opts->backlog, 0) : 1) < 0) {
 			log_err("Failed to listed on socket");
 			goto error_close;
 		}
diff --git a/tools/testing/selftests/bpf/network_helpers.h b/tools/testing/selftests/bpf/network_helpers.h
index 9ea36524b9db..aac5b94d6379 100644
--- a/tools/testing/selftests/bpf/network_helpers.h
+++ b/tools/testing/selftests/bpf/network_helpers.h
@@ -25,6 +25,16 @@ struct network_helper_opts {
 	int timeout_ms;
 	bool must_fail;
 	int proto;
+	/* +ve: Passed to listen() as-is.
+	 *   0: Default when the test does not set
+	 *      a particular value during the struct init.
+	 *      It is changed to 1 before passing to listen().
+	 *      Most tests only have one on-going connection.
+	 * -ve: It is changed to 0 before passing to listen().
+	 *      It is useful to force syncookie without
+	 *	changing the "tcp_syncookies" sysctl from 1 to 2.
+	 */
+	int backlog;
 	int (*post_socket_cb)(int fd, void *opts);
 	void *cb_opts;
 };
-- 
cgit v1.2.3-58-ga151


From 7046345d48adcc3f519e7b6192184f6049908bdb Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Tue, 9 Jul 2024 17:16:18 +0800
Subject: selftests/bpf: Add ASSERT_OK_FD macro

Add a new dedicated ASSERT macro ASSERT_OK_FD to test whether a socket
FD is valid or not. It can be used to replace macros ASSERT_GT(fd, 0, ""),
ASSERT_NEQ(fd, -1, "") or statements (fd < 0), (fd != -1).

Suggested-by: Martin KaFai Lau <martin.lau@kernel.org>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/ded75be86ac630a3a5099739431854c1ec33f0ea.1720515893.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/test_progs.h | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h
index 0ba5a20b19ba..51341d50213b 100644
--- a/tools/testing/selftests/bpf/test_progs.h
+++ b/tools/testing/selftests/bpf/test_progs.h
@@ -377,6 +377,15 @@ int test__join_cgroup(const char *path);
 	___ok;								\
 })
 
+#define ASSERT_OK_FD(fd, name) ({					\
+	static int duration = 0;					\
+	int ___fd = (fd);						\
+	bool ___ok = ___fd >= 0;					\
+	CHECK(!___ok, (name), "unexpected fd: %d (errno %d)\n",		\
+	      ___fd, errno);						\
+	___ok;								\
+})
+
 #define SYS(goto_label, fmt, ...)					\
 	({								\
 		char cmd[1024];						\
-- 
cgit v1.2.3-58-ga151


From adae187ebedcd95d02f045bc37dfecfd5b29434b Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Tue, 9 Jul 2024 17:16:19 +0800
Subject: selftests/bpf: Close fd in error path in drop_on_reuseport

In the error path when update_lookup_map() fails in drop_on_reuseport in
prog_tests/sk_lookup.c, "server1", the fd of server 1, should be closed.
This patch fixes this by using "goto close_srv1" lable instead of "detach"
to close "server1" in this case.

Fixes: 0ab5539f8584 ("selftests/bpf: Tests for BPF_SK_LOOKUP attach point")
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/86aed33b4b0ea3f04497c757845cff7e8e621a2d.1720515893.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/sk_lookup.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
index 597d0467a926..de2466547efe 100644
--- a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
+++ b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
@@ -994,7 +994,7 @@ static void drop_on_reuseport(const struct test *t)
 
 	err = update_lookup_map(t->sock_map, SERVER_A, server1);
 	if (err)
-		goto detach;
+		goto close_srv1;
 
 	/* second server on destination address we should never reach */
 	server2 = make_server(t->sotype, t->connect_to.ip, t->connect_to.port,
-- 
cgit v1.2.3-58-ga151


From 14fc6fcd35e7dde6d1de062b6711476b3050b22e Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Tue, 9 Jul 2024 17:16:20 +0800
Subject: selftests/bpf: Use start_server_str in sk_lookup

This patch uses public helper start_server_str() to simplify make_server()
in prog_tests/sk_lookup.c.

Add a callback setsockopts() to do all sockopts, set it to post_socket_cb
pointer of struct network_helper_opts. And add a new struct cb_opts to save
the data needed to pass to the callback. Then pass this network_helper_opts
to start_server_str().

Also use ASSERT_OK_FD() to check fd returned by start_server_str().

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/5981539f5591d2c4998c962ef2bf45f34c940548.1720515893.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/sk_lookup.c | 58 +++++++++++++---------
 1 file changed, 34 insertions(+), 24 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
index de2466547efe..20ee5da2c721 100644
--- a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
+++ b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
@@ -77,6 +77,12 @@ struct test {
 	bool reuseport_has_conns; /* Add a connected socket to reuseport group */
 };
 
+struct cb_opts {
+	int family;
+	int sotype;
+	bool reuseport;
+};
+
 static __u32 duration;		/* for CHECK macro */
 
 static bool is_ipv6(const char *ip)
@@ -142,19 +148,14 @@ static int make_socket(int sotype, const char *ip, int port,
 	return fd;
 }
 
-static int make_server(int sotype, const char *ip, int port,
-		       struct bpf_program *reuseport_prog)
+static int setsockopts(int fd, void *opts)
 {
-	struct sockaddr_storage addr = {0};
+	struct cb_opts *co = (struct cb_opts *)opts;
 	const int one = 1;
-	int err, fd = -1;
-
-	fd = make_socket(sotype, ip, port, &addr);
-	if (fd < 0)
-		return -1;
+	int err = 0;
 
 	/* Enabled for UDPv6 sockets for IPv4-mapped IPv6 to work. */
-	if (sotype == SOCK_DGRAM) {
+	if (co->sotype == SOCK_DGRAM) {
 		err = setsockopt(fd, SOL_IP, IP_RECVORIGDSTADDR, &one,
 				 sizeof(one));
 		if (CHECK(err, "setsockopt(IP_RECVORIGDSTADDR)", "failed\n")) {
@@ -163,7 +164,7 @@ static int make_server(int sotype, const char *ip, int port,
 		}
 	}
 
-	if (sotype == SOCK_DGRAM && addr.ss_family == AF_INET6) {
+	if (co->sotype == SOCK_DGRAM && co->family == AF_INET6) {
 		err = setsockopt(fd, SOL_IPV6, IPV6_RECVORIGDSTADDR, &one,
 				 sizeof(one));
 		if (CHECK(err, "setsockopt(IPV6_RECVORIGDSTADDR)", "failed\n")) {
@@ -172,7 +173,7 @@ static int make_server(int sotype, const char *ip, int port,
 		}
 	}
 
-	if (sotype == SOCK_STREAM) {
+	if (co->sotype == SOCK_STREAM) {
 		err = setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, &one,
 				 sizeof(one));
 		if (CHECK(err, "setsockopt(SO_REUSEADDR)", "failed\n")) {
@@ -181,7 +182,7 @@ static int make_server(int sotype, const char *ip, int port,
 		}
 	}
 
-	if (reuseport_prog) {
+	if (co->reuseport) {
 		err = setsockopt(fd, SOL_SOCKET, SO_REUSEPORT, &one,
 				 sizeof(one));
 		if (CHECK(err, "setsockopt(SO_REUSEPORT)", "failed\n")) {
@@ -190,19 +191,28 @@ static int make_server(int sotype, const char *ip, int port,
 		}
 	}
 
-	err = bind(fd, (void *)&addr, inetaddr_len(&addr));
-	if (CHECK(err, "bind", "failed\n")) {
-		log_err("failed to bind listen socket");
-		goto fail;
-	}
+fail:
+	return err;
+}
 
-	if (sotype == SOCK_STREAM) {
-		err = listen(fd, SOMAXCONN);
-		if (CHECK(err, "make_server", "listen")) {
-			log_err("failed to listen on port %d", port);
-			goto fail;
-		}
-	}
+static int make_server(int sotype, const char *ip, int port,
+		       struct bpf_program *reuseport_prog)
+{
+	struct cb_opts cb_opts = {
+		.family = is_ipv6(ip) ? AF_INET6 : AF_INET,
+		.sotype = sotype,
+		.reuseport = reuseport_prog,
+	};
+	struct network_helper_opts opts = {
+		.backlog	= SOMAXCONN,
+		.post_socket_cb = setsockopts,
+		.cb_opts	= &cb_opts,
+	};
+	int err, fd;
+
+	fd = start_server_str(cb_opts.family, sotype, ip, port, &opts);
+	if (!ASSERT_OK_FD(fd, "start_server_str"))
+		return -1;
 
 	/* Late attach reuseport prog so we can have one init path */
 	if (reuseport_prog) {
-- 
cgit v1.2.3-58-ga151


From d9810c43f660fd502c5003244a5e9c181aa7df99 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Tue, 9 Jul 2024 17:16:21 +0800
Subject: selftests/bpf: Use start_server_addr in sk_lookup

This patch uses public helper start_server_addr() in udp_recv_send()
in prog_tests/sk_lookup.c to simplify the code.

And use ASSERT_OK_FD() to check fd returned by start_server_addr().

Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/f11cabfef4a2170ecb66a1e8e2e72116d8f621b3.1720515893.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/sk_lookup.c | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
index 20ee5da2c721..386e482be617 100644
--- a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
+++ b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
@@ -416,18 +416,12 @@ static int udp_recv_send(int server_fd)
 	}
 
 	/* Reply from original destination address. */
-	fd = socket(dst_addr->ss_family, SOCK_DGRAM, 0);
-	if (CHECK(fd < 0, "socket", "failed\n")) {
+	fd = start_server_addr(SOCK_DGRAM, dst_addr, sizeof(*dst_addr), NULL);
+	if (!ASSERT_OK_FD(fd, "start_server_addr")) {
 		log_err("failed to create tx socket");
 		return -1;
 	}
 
-	ret = bind(fd, (struct sockaddr *)dst_addr, sizeof(*dst_addr));
-	if (CHECK(ret, "bind", "failed\n")) {
-		log_err("failed to bind tx socket");
-		goto out;
-	}
-
 	msg.msg_control = NULL;
 	msg.msg_controllen = 0;
 	n = sendmsg(fd, &msg, 0);
-- 
cgit v1.2.3-58-ga151


From 9004054b1629d481fedea2d92b880f79fc6fa81b Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Tue, 9 Jul 2024 17:16:22 +0800
Subject: selftests/bpf: Use connect_fd_to_fd in sk_lookup

This patch uses public helper connect_fd_to_fd() exported in
network_helpers.h instead of using getsockname() + connect() in
run_lookup_prog() in prog_tests/sk_lookup.c. This can simplify
the code.

Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/7077c277cde5a1864cdc244727162fb75c8bb9c5.1720515893.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/sk_lookup.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
index 386e482be617..ae87c00867ba 100644
--- a/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
+++ b/tools/testing/selftests/bpf/prog_tests/sk_lookup.c
@@ -633,9 +633,6 @@ static void run_lookup_prog(const struct test *t)
 	 * BPF socket lookup.
 	 */
 	if (t->reuseport_has_conns) {
-		struct sockaddr_storage addr = {};
-		socklen_t len = sizeof(addr);
-
 		/* Add an extra socket to reuseport group */
 		reuse_conn_fd = make_server(t->sotype, t->listen_at.ip,
 					    t->listen_at.port,
@@ -643,12 +640,9 @@ static void run_lookup_prog(const struct test *t)
 		if (reuse_conn_fd < 0)
 			goto close;
 
-		/* Connect the extra socket to itself */
-		err = getsockname(reuse_conn_fd, (void *)&addr, &len);
-		if (CHECK(err, "getsockname", "errno %d\n", errno))
-			goto close;
-		err = connect(reuse_conn_fd, (void *)&addr, len);
-		if (CHECK(err, "connect", "errno %d\n", errno))
+		 /* Connect the extra socket to itself */
+		err = connect_fd_to_fd(reuse_conn_fd, reuse_conn_fd, 0);
+		if (!ASSERT_OK(err, "connect_fd_to_fd"))
 			goto close;
 	}
 
-- 
cgit v1.2.3-58-ga151


From eef0532e900c20a6760da829e82dac3ee18688c5 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Wed, 10 Jul 2024 21:10:16 +0800
Subject: selftests/bpf: Null checks for links in bpf_tcp_ca

Run bpf_tcp_ca selftests (./test_progs -t bpf_tcp_ca) on a Loongarch
platform, some "Segmentation fault" errors occur:

'''
 test_dctcp:PASS:bpf_dctcp__open_and_load 0 nsec
 test_dctcp:FAIL:bpf_map__attach_struct_ops unexpected error: -524
 #29/1    bpf_tcp_ca/dctcp:FAIL
 test_cubic:PASS:bpf_cubic__open_and_load 0 nsec
 test_cubic:FAIL:bpf_map__attach_struct_ops unexpected error: -524
 #29/2    bpf_tcp_ca/cubic:FAIL
 test_dctcp_fallback:PASS:dctcp_skel 0 nsec
 test_dctcp_fallback:PASS:bpf_dctcp__load 0 nsec
 test_dctcp_fallback:FAIL:dctcp link unexpected error: -524
 #29/4    bpf_tcp_ca/dctcp_fallback:FAIL
 test_write_sk_pacing:PASS:open_and_load 0 nsec
 test_write_sk_pacing:FAIL:attach_struct_ops unexpected error: -524
 #29/6    bpf_tcp_ca/write_sk_pacing:FAIL
 test_update_ca:PASS:open 0 nsec
 test_update_ca:FAIL:attach_struct_ops unexpected error: -524
 settcpca:FAIL:setsockopt unexpected setsockopt: \
					actual -1 == expected -1
 (network_helpers.c:99: errno: No such file or directory) \
					Failed to call post_socket_cb
 start_test:FAIL:start_server_str unexpected start_server_str: \
					actual -1 == expected -1
 test_update_ca:FAIL:ca1_ca1_cnt unexpected ca1_ca1_cnt: \
					actual 0 <= expected 0
 #29/9    bpf_tcp_ca/update_ca:FAIL
 #29      bpf_tcp_ca:FAIL
 Caught signal #11!
 Stack trace:
 ./test_progs(crash_handler+0x28)[0x5555567ed91c]
 linux-vdso.so.1(__vdso_rt_sigreturn+0x0)[0x7ffffee408b0]
 ./test_progs(bpf_link__update_map+0x80)[0x555556824a78]
 ./test_progs(+0x94d68)[0x5555564c4d68]
 ./test_progs(test_bpf_tcp_ca+0xe8)[0x5555564c6a88]
 ./test_progs(+0x3bde54)[0x5555567ede54]
 ./test_progs(main+0x61c)[0x5555567efd54]
 /usr/lib64/libc.so.6(+0x22208)[0x7ffff2aaa208]
 /usr/lib64/libc.so.6(__libc_start_main+0xac)[0x7ffff2aaa30c]
 ./test_progs(_start+0x48)[0x55555646bca8]
 Segmentation fault
'''

This is because BPF trampoline is not implemented on Loongarch yet,
"link" returned by bpf_map__attach_struct_ops() is NULL. test_progs
crashs when this NULL link passes to bpf_link__update_map(). This
patch adds NULL checks for all links in bpf_tcp_ca to fix these errors.
If "link" is NULL, goto the newly added label "out" to destroy the skel.

v2:
 - use "goto out" instead of "return" as Eduard suggested.

Fixes: 06da9f3bd641 ("selftests/bpf: Test switching TCP Congestion Control algorithms.")
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Reviewed-by: Alan Maguire <alan.maguire@oracle.com>
Link: https://lore.kernel.org/r/b4c841492bd4ed97964e4e61e92827ce51bf1dc9.1720615848.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
index bceff5900016..63422f4f3896 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c
@@ -411,7 +411,8 @@ static void test_update_ca(void)
 		return;
 
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
-	ASSERT_OK_PTR(link, "attach_struct_ops");
+	if (!ASSERT_OK_PTR(link, "attach_struct_ops"))
+		goto out;
 
 	do_test(&opts);
 	saved_ca1_cnt = skel->bss->ca1_cnt;
@@ -425,6 +426,7 @@ static void test_update_ca(void)
 	ASSERT_GT(skel->bss->ca2_cnt, 0, "ca2_ca2_cnt");
 
 	bpf_link__destroy(link);
+out:
 	tcp_ca_update__destroy(skel);
 }
 
@@ -447,7 +449,8 @@ static void test_update_wrong(void)
 		return;
 
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
-	ASSERT_OK_PTR(link, "attach_struct_ops");
+	if (!ASSERT_OK_PTR(link, "attach_struct_ops"))
+		goto out;
 
 	do_test(&opts);
 	saved_ca1_cnt = skel->bss->ca1_cnt;
@@ -460,6 +463,7 @@ static void test_update_wrong(void)
 	ASSERT_GT(skel->bss->ca1_cnt, saved_ca1_cnt, "ca2_ca1_cnt");
 
 	bpf_link__destroy(link);
+out:
 	tcp_ca_update__destroy(skel);
 }
 
@@ -481,7 +485,8 @@ static void test_mixed_links(void)
 		return;
 
 	link_nl = bpf_map__attach_struct_ops(skel->maps.ca_no_link);
-	ASSERT_OK_PTR(link_nl, "attach_struct_ops_nl");
+	if (!ASSERT_OK_PTR(link_nl, "attach_struct_ops_nl"))
+		goto out;
 
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_1);
 	ASSERT_OK_PTR(link, "attach_struct_ops");
@@ -494,6 +499,7 @@ static void test_mixed_links(void)
 
 	bpf_link__destroy(link);
 	bpf_link__destroy(link_nl);
+out:
 	tcp_ca_update__destroy(skel);
 }
 
@@ -536,7 +542,8 @@ static void test_link_replace(void)
 	bpf_link__destroy(link);
 
 	link = bpf_map__attach_struct_ops(skel->maps.ca_update_2);
-	ASSERT_OK_PTR(link, "attach_struct_ops_2nd");
+	if (!ASSERT_OK_PTR(link, "attach_struct_ops_2nd"))
+		goto out;
 
 	/* BPF_F_REPLACE with a wrong old map Fd. It should fail!
 	 *
@@ -559,6 +566,7 @@ static void test_link_replace(void)
 
 	bpf_link__destroy(link);
 
+out:
 	tcp_ca_update__destroy(skel);
 }
 
-- 
cgit v1.2.3-58-ga151


From 52b49ec1b2c78deb258596c3b231201445ef5380 Mon Sep 17 00:00:00 2001
From: Geliang Tang <tanggeliang@kylinos.cn>
Date: Wed, 10 Jul 2024 21:10:17 +0800
Subject: selftests/bpf: Close obj in error path in xdp_adjust_tail

If bpf_object__load() fails in test_xdp_adjust_frags_tail_grow(), "obj"
opened before this should be closed. So use "goto out" to close it instead
of using "return" here.

Fixes: 110221081aac ("bpf: selftests: update xdp_adjust_tail selftest to include xdp frags")
Signed-off-by: Geliang Tang <tanggeliang@kylinos.cn>
Link: https://lore.kernel.org/r/f282a1ed2d0e3fb38cceefec8e81cabb69cab260.1720615848.git.tanggeliang@kylinos.cn
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
---
 tools/testing/selftests/bpf/prog_tests/xdp_adjust_tail.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_adjust_tail.c b/tools/testing/selftests/bpf/prog_tests/xdp_adjust_tail.c
index f09505f8b038..53d6ad8c2257 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_adjust_tail.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_adjust_tail.c
@@ -222,7 +222,7 @@ static void test_xdp_adjust_frags_tail_grow(void)
 
 	prog = bpf_object__next_program(obj, NULL);
 	if (bpf_object__load(obj))
-		return;
+		goto out;
 
 	prog_fd = bpf_program__fd(prog);
 
-- 
cgit v1.2.3-58-ga151


From ab896aa62e309b3ceaa35b82dc99b64da7ce6bfa Mon Sep 17 00:00:00 2001
From: Marek Vasut <marex@denx.de>
Date: Mon, 8 Jul 2024 23:16:29 +0200
Subject: dt-bindings: net: realtek,rtl82xx: Document RTL8211F LED support

The RTL8211F PHY does support LED configuration, document support
for LEDs in the binding document.

Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://patch.msgid.link/20240708211649.165793-1-marex@denx.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../devicetree/bindings/net/realtek,rtl82xx.yaml        | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/realtek,rtl82xx.yaml b/Documentation/devicetree/bindings/net/realtek,rtl82xx.yaml
index 18ee72f5c74a..d248a08a2136 100644
--- a/Documentation/devicetree/bindings/net/realtek,rtl82xx.yaml
+++ b/Documentation/devicetree/bindings/net/realtek,rtl82xx.yaml
@@ -14,9 +14,6 @@ maintainers:
 description:
   Bindings for Realtek RTL82xx PHYs
 
-allOf:
-  - $ref: ethernet-phy.yaml#
-
 properties:
   compatible:
     enum:
@@ -41,6 +38,8 @@ properties:
       - ethernet-phy-id001c.cad0
       - ethernet-phy-id001c.cb00
 
+  leds: true
+
   realtek,clkout-disable:
     type: boolean
     description:
@@ -54,6 +53,18 @@ properties:
 
 unevaluatedProperties: false
 
+allOf:
+  - $ref: ethernet-phy.yaml#
+  - if:
+      not:
+        properties:
+          compatible:
+            contains:
+              const: ethernet-phy-id001c.c916
+    then:
+      properties:
+        leds: false
+
 examples:
   - |
     mdio {
-- 
cgit v1.2.3-58-ga151


From d00ba1d734f78c8f0a4826acd1f19ec42bbcb9be Mon Sep 17 00:00:00 2001
From: Frank Li <Frank.Li@nxp.com>
Date: Tue, 9 Jul 2024 17:48:41 -0400
Subject: dt-bindings: net: convert enetc to yaml

Convert enetc device binding file to yaml. Split to 3 yaml files,
'fsl,enetc.yaml', 'fsl,enetc-mdio.yaml', 'fsl,enetc-ierb.yaml'.

Additional Changes:
- Add pci<vendor id>,<production id> in compatible string.
- Ref to common ethernet-controller.yaml and mdio.yaml.
- Add Wei fang, Vladimir and Claudiu as maintainer.
- Update ENETC description.
- Remove fixed-link part.

Signed-off-by: Frank Li <Frank.Li@nxp.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://patch.msgid.link/20240709214841.570154-1-Frank.Li@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../devicetree/bindings/net/fsl,enetc-ierb.yaml    |  38 +++++++
 .../devicetree/bindings/net/fsl,enetc-mdio.yaml    |  57 ++++++++++
 .../devicetree/bindings/net/fsl,enetc.yaml         |  66 ++++++++++++
 .../devicetree/bindings/net/fsl-enetc.txt          | 119 ---------------------
 4 files changed, 161 insertions(+), 119 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/net/fsl,enetc-ierb.yaml
 create mode 100644 Documentation/devicetree/bindings/net/fsl,enetc-mdio.yaml
 create mode 100644 Documentation/devicetree/bindings/net/fsl,enetc.yaml
 delete mode 100644 Documentation/devicetree/bindings/net/fsl-enetc.txt

diff --git a/Documentation/devicetree/bindings/net/fsl,enetc-ierb.yaml b/Documentation/devicetree/bindings/net/fsl,enetc-ierb.yaml
new file mode 100644
index 000000000000..c8a654310b90
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/fsl,enetc-ierb.yaml
@@ -0,0 +1,38 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/fsl,enetc-ierb.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Integrated Endpoint Register Block
+
+description:
+  The fsl_enetc driver can probe on the Integrated Endpoint Register Block,
+  which preconfigures the FIFO limits for the ENETC ports.
+
+maintainers:
+  - Frank Li <Frank.Li@nxp.com>
+  - Vladimir Oltean <vladimir.oltean@nxp.com>
+  - Wei Fang <wei.fang@nxp.com>
+  - Claudiu Manoil <claudiu.manoil@nxp.com>
+
+properties:
+  compatible:
+    enum:
+      - fsl,ls1028a-enetc-ierb
+
+  reg:
+    maxItems: 1
+
+required:
+  - compatible
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    endpoint-config@f0800000 {
+        compatible = "fsl,ls1028a-enetc-ierb";
+        reg = <0xf0800000 0x10000>;
+    };
diff --git a/Documentation/devicetree/bindings/net/fsl,enetc-mdio.yaml b/Documentation/devicetree/bindings/net/fsl,enetc-mdio.yaml
new file mode 100644
index 000000000000..c1dd6aa04321
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/fsl,enetc-mdio.yaml
@@ -0,0 +1,57 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/fsl,enetc-mdio.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: ENETC external MDIO PCIe endpoint device
+
+description:
+  NETC provides an external master MDIO interface (EMDIO) for managing external
+  devices (PHYs). EMDIO supports both Clause 22 and 45 protocols. And the EMDIO
+  provides a means for different software modules to share a single set of MDIO
+  signals to access their PHYs.
+
+maintainers:
+  - Frank Li <Frank.Li@nxp.com>
+  - Vladimir Oltean <vladimir.oltean@nxp.com>
+  - Wei Fang <wei.fang@nxp.com>
+  - Claudiu Manoil <claudiu.manoil@nxp.com>
+
+properties:
+  compatible:
+    items:
+      - enum:
+          - pci1957,ee01
+      - const: fsl,enetc-mdio
+
+  reg:
+    maxItems: 1
+
+required:
+  - compatible
+  - reg
+
+allOf:
+  - $ref: mdio.yaml
+  - $ref: /schemas/pci/pci-device.yaml
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    pcie{
+        #address-cells = <3>;
+        #size-cells = <2>;
+
+        mdio@0,3 {
+            compatible = "pci1957,ee01", "fsl,enetc-mdio";
+            reg = <0x000300 0 0 0 0>;
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            ethernet-phy@2 {
+                reg = <0x2>;
+            };
+        };
+    };
diff --git a/Documentation/devicetree/bindings/net/fsl,enetc.yaml b/Documentation/devicetree/bindings/net/fsl,enetc.yaml
new file mode 100644
index 000000000000..e152c93998fe
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/fsl,enetc.yaml
@@ -0,0 +1,66 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/fsl,enetc.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: The NIC functionality of NXP NETC
+
+description:
+  The NIC functionality in NETC is known as EtherNET Controller (ENETC). ENETC
+  supports virtualization/isolation based on PCIe Single Root IO Virtualization
+  (SR-IOV), advanced QoS with 8 traffic classes and 4 drop resilience levels,
+  and a full range of TSN standards and NIC offload capabilities
+
+maintainers:
+  - Frank Li <Frank.Li@nxp.com>
+  - Vladimir Oltean <vladimir.oltean@nxp.com>
+  - Wei Fang <wei.fang@nxp.com>
+  - Claudiu Manoil <claudiu.manoil@nxp.com>
+
+properties:
+  compatible:
+    items:
+      - enum:
+          - pci1957,e100
+      - const: fsl,enetc
+
+  reg:
+    maxItems: 1
+
+  mdio:
+    $ref: mdio.yaml
+    unevaluatedProperties: false
+    description: Optional child node for ENETC instance, otherwise use NETC EMDIO.
+
+required:
+  - compatible
+  - reg
+
+allOf:
+  - $ref: /schemas/pci/pci-device.yaml
+  - $ref: ethernet-controller.yaml
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    pcie {
+        #address-cells = <3>;
+        #size-cells = <2>;
+
+        ethernet@0,0 {
+            compatible = "pci1957,e100", "fsl,enetc";
+            reg = <0x000000 0 0 0 0>;
+            phy-handle = <&sgmii_phy0>;
+            phy-connection-type = "sgmii";
+
+            mdio {
+                #address-cells = <1>;
+                #size-cells = <0>;
+                phy@2 {
+                    reg = <0x2>;
+                };
+            };
+        };
+    };
diff --git a/Documentation/devicetree/bindings/net/fsl-enetc.txt b/Documentation/devicetree/bindings/net/fsl-enetc.txt
deleted file mode 100644
index 9b9a3f197e2d..000000000000
--- a/Documentation/devicetree/bindings/net/fsl-enetc.txt
+++ /dev/null
@@ -1,119 +0,0 @@
-* ENETC ethernet device tree bindings
-
-Depending on board design and ENETC port type (internal or
-external) there are two supported link modes specified by
-below device tree bindings.
-
-Required properties:
-
-- reg		: Specifies PCIe Device Number and Function
-		  Number of the ENETC endpoint device, according
-		  to parent node bindings.
-- compatible	: Should be "fsl,enetc".
-
-1. The ENETC external port is connected to a MDIO configurable phy
-
-1.1. Using the local ENETC Port MDIO interface
-
-In this case, the ENETC node should include a "mdio" sub-node
-that in turn should contain the "ethernet-phy" node describing the
-external phy.  Below properties are required, their bindings
-already defined in Documentation/devicetree/bindings/net/ethernet.txt or
-Documentation/devicetree/bindings/net/phy.txt.
-
-Required:
-
-- phy-handle		: Phandle to a PHY on the MDIO bus.
-			  Defined in ethernet.txt.
-
-- phy-connection-type	: Defined in ethernet.txt.
-
-- mdio			: "mdio" node, defined in mdio.txt.
-
-- ethernet-phy		: "ethernet-phy" node, defined in phy.txt.
-
-Example:
-
-	ethernet@0,0 {
-		compatible = "fsl,enetc";
-		reg = <0x000000 0 0 0 0>;
-		phy-handle = <&sgmii_phy0>;
-		phy-connection-type = "sgmii";
-
-		mdio {
-			#address-cells = <1>;
-			#size-cells = <0>;
-			sgmii_phy0: ethernet-phy@2 {
-				reg = <0x2>;
-			};
-		};
-	};
-
-1.2. Using the central MDIO PCIe endpoint device
-
-In this case, the mdio node should be defined as another PCIe
-endpoint node, at the same level with the ENETC port nodes.
-
-Required properties:
-
-- reg		: Specifies PCIe Device Number and Function
-		  Number of the ENETC endpoint device, according
-		  to parent node bindings.
-- compatible	: Should be "fsl,enetc-mdio".
-
-The remaining required mdio bus properties are standard, their bindings
-already defined in Documentation/devicetree/bindings/net/mdio.txt.
-
-Example:
-
-	ethernet@0,0 {
-		compatible = "fsl,enetc";
-		reg = <0x000000 0 0 0 0>;
-		phy-handle = <&sgmii_phy0>;
-		phy-connection-type = "sgmii";
-	};
-
-	mdio@0,3 {
-		compatible = "fsl,enetc-mdio";
-		reg = <0x000300 0 0 0 0>;
-		#address-cells = <1>;
-		#size-cells = <0>;
-		sgmii_phy0: ethernet-phy@2 {
-			reg = <0x2>;
-		};
-	};
-
-2. The ENETC port is an internal port or has a fixed-link external
-connection
-
-In this case, the ENETC port node defines a fixed link connection,
-as specified by Documentation/devicetree/bindings/net/fixed-link.txt.
-
-Required:
-
-- fixed-link	: "fixed-link" node, defined in "fixed-link.txt".
-
-Example:
-	ethernet@0,2 {
-		compatible = "fsl,enetc";
-		reg = <0x000200 0 0 0 0>;
-		fixed-link {
-			speed = <1000>;
-			full-duplex;
-		};
-	};
-
-* Integrated Endpoint Register Block bindings
-
-Optionally, the fsl_enetc driver can probe on the Integrated Endpoint Register
-Block, which preconfigures the FIFO limits for the ENETC ports. This is a node
-with the following properties:
-
-- reg		: Specifies the address in the SoC memory space.
-- compatible	: Must be "fsl,ls1028a-enetc-ierb".
-
-Example:
-	ierb@1f0800000 {
-		compatible = "fsl,ls1028a-enetc-ierb";
-		reg = <0x01 0xf0800000 0x0 0x10000>;
-	};
-- 
cgit v1.2.3-58-ga151


From a317f873ea85e1e3abf7963670e88fd6f23f8c68 Mon Sep 17 00:00:00 2001
From: Anil Samal <anil.samal@intel.com>
Date: Tue, 9 Jul 2024 13:29:47 -0700
Subject: ice: Extend Sideband Queue command to support flags

Current driver implementation for Sideband Queue supports a
fixed flag (ICE_AQ_FLAG_RD). To retrieve FEC statistics from
firmware, Sideband Queue command is used with a different flag.

Extend API for Sideband Queue command to use 'flags' as input
argument.

Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Anil Samal <anil.samal@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Link: https://patch.msgid.link/20240709202951.2103115-2-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_common.c |  5 +++--
 drivers/net/ethernet/intel/ice/ice_common.h |  2 +-
 drivers/net/ethernet/intel/ice/ice_ptp_hw.c | 20 ++++++++++----------
 3 files changed, 14 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 9ae61cd8923e..9fa9d914ae46 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -1497,8 +1497,9 @@ ice_sbq_send_cmd(struct ice_hw *hw, struct ice_sbq_cmd_desc *desc,
  * ice_sbq_rw_reg - Fill Sideband Queue command
  * @hw: pointer to the HW struct
  * @in: message info to be filled in descriptor
+ * @flags: control queue descriptor flags
  */
-int ice_sbq_rw_reg(struct ice_hw *hw, struct ice_sbq_msg_input *in)
+int ice_sbq_rw_reg(struct ice_hw *hw, struct ice_sbq_msg_input *in, u16 flags)
 {
 	struct ice_sbq_cmd_desc desc = {0};
 	struct ice_sbq_msg_req msg = {0};
@@ -1522,7 +1523,7 @@ int ice_sbq_rw_reg(struct ice_hw *hw, struct ice_sbq_msg_input *in)
 		 */
 		msg_len -= sizeof(msg.data);
 
-	desc.flags = cpu_to_le16(ICE_AQ_FLAG_RD);
+	desc.flags = cpu_to_le16(flags);
 	desc.opcode = cpu_to_le16(ice_sbq_opc_neigh_dev_req);
 	desc.param0.cmd_len = cpu_to_le16(msg_len);
 	status = ice_sbq_send_cmd(hw, &desc, &msg, msg_len, NULL);
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index 40dc735dc25c..ef9e8c23580d 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -201,7 +201,7 @@ int ice_replay_vsi(struct ice_hw *hw, u16 vsi_handle);
 void ice_replay_post(struct ice_hw *hw);
 struct ice_q_ctx *
 ice_get_lan_q_ctx(struct ice_hw *hw, u16 vsi_handle, u8 tc, u16 q_handle);
-int ice_sbq_rw_reg(struct ice_hw *hw, struct ice_sbq_msg_input *in);
+int ice_sbq_rw_reg(struct ice_hw *hw, struct ice_sbq_msg_input *in, u16 flag);
 int
 ice_aq_get_cgu_abilities(struct ice_hw *hw,
 			 struct ice_aqc_get_cgu_abilities *abilities);
diff --git a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
index 1e9a4ccd0ea2..3a33e6b9b313 100644
--- a/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
+++ b/drivers/net/ethernet/intel/ice/ice_ptp_hw.c
@@ -247,7 +247,7 @@ static int ice_read_cgu_reg_e82x(struct ice_hw *hw, u32 addr, u32 *val)
 	};
 	int err;
 
-	err = ice_sbq_rw_reg(hw, &cgu_msg);
+	err = ice_sbq_rw_reg(hw, &cgu_msg, ICE_AQ_FLAG_RD);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "Failed to read CGU register 0x%04x, err %d\n",
 			  addr, err);
@@ -280,7 +280,7 @@ static int ice_write_cgu_reg_e82x(struct ice_hw *hw, u32 addr, u32 val)
 	};
 	int err;
 
-	err = ice_sbq_rw_reg(hw, &cgu_msg);
+	err = ice_sbq_rw_reg(hw, &cgu_msg, ICE_AQ_FLAG_RD);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "Failed to write CGU register 0x%04x, err %d\n",
 			  addr, err);
@@ -902,7 +902,7 @@ static int ice_write_phy_eth56g(struct ice_hw *hw, u8 phy_idx, u32 addr,
 	phy_msg.data = val;
 	phy_msg.dest_dev = hw->ptp.phy.eth56g.phy_addr[phy_idx];
 
-	err = ice_sbq_rw_reg(hw, &phy_msg);
+	err = ice_sbq_rw_reg(hw, &phy_msg, ICE_AQ_FLAG_RD);
 
 	if (err)
 		ice_debug(hw, ICE_DBG_PTP, "PTP failed to send msg to phy %d\n",
@@ -934,7 +934,7 @@ static int ice_read_phy_eth56g(struct ice_hw *hw, u8 phy_idx, u32 addr,
 	phy_msg.data = 0;
 	phy_msg.dest_dev = hw->ptp.phy.eth56g.phy_addr[phy_idx];
 
-	err = ice_sbq_rw_reg(hw, &phy_msg);
+	err = ice_sbq_rw_reg(hw, &phy_msg, ICE_AQ_FLAG_RD);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "PTP failed to send msg to phy %d\n",
 			  err);
@@ -2855,7 +2855,7 @@ ice_read_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 offset, u32 *val)
 	ice_fill_phy_msg_e82x(hw, &msg, port, offset);
 	msg.opcode = ice_sbq_msg_rd;
 
-	err = ice_sbq_rw_reg(hw, &msg);
+	err = ice_sbq_rw_reg(hw, &msg, ICE_AQ_FLAG_RD);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
 			  err);
@@ -2933,7 +2933,7 @@ ice_write_phy_reg_e82x(struct ice_hw *hw, u8 port, u16 offset, u32 val)
 	msg.opcode = ice_sbq_msg_wr;
 	msg.data = val;
 
-	err = ice_sbq_rw_reg(hw, &msg);
+	err = ice_sbq_rw_reg(hw, &msg, ICE_AQ_FLAG_RD);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
 			  err);
@@ -3094,7 +3094,7 @@ ice_read_quad_reg_e82x(struct ice_hw *hw, u8 quad, u16 offset, u32 *val)
 
 	msg.opcode = ice_sbq_msg_rd;
 
-	err = ice_sbq_rw_reg(hw, &msg);
+	err = ice_sbq_rw_reg(hw, &msg, ICE_AQ_FLAG_RD);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
 			  err);
@@ -3129,7 +3129,7 @@ ice_write_quad_reg_e82x(struct ice_hw *hw, u8 quad, u16 offset, u32 val)
 	msg.opcode = ice_sbq_msg_wr;
 	msg.data = val;
 
-	err = ice_sbq_rw_reg(hw, &msg);
+	err = ice_sbq_rw_reg(hw, &msg, ICE_AQ_FLAG_RD);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
 			  err);
@@ -4780,7 +4780,7 @@ static int ice_read_phy_reg_e810(struct ice_hw *hw, u32 addr, u32 *val)
 	msg.opcode = ice_sbq_msg_rd;
 	msg.dest_dev = rmn_0;
 
-	err = ice_sbq_rw_reg(hw, &msg);
+	err = ice_sbq_rw_reg(hw, &msg, ICE_AQ_FLAG_RD);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
 			  err);
@@ -4811,7 +4811,7 @@ static int ice_write_phy_reg_e810(struct ice_hw *hw, u32 addr, u32 val)
 	msg.dest_dev = rmn_0;
 	msg.data = val;
 
-	err = ice_sbq_rw_reg(hw, &msg);
+	err = ice_sbq_rw_reg(hw, &msg, ICE_AQ_FLAG_RD);
 	if (err) {
 		ice_debug(hw, ICE_DBG_PTP, "Failed to send message to PHY, err %d\n",
 			  err);
-- 
cgit v1.2.3-58-ga151


From ac21add2540e1a6ed27abea47fd2234ab05c71eb Mon Sep 17 00:00:00 2001
From: Anil Samal <anil.samal@intel.com>
Date: Tue, 9 Jul 2024 13:29:48 -0700
Subject: ice: Implement driver functionality to dump fec statistics

To debug link issues in the field, it is paramount to
dump fec corrected/uncorrected block counts from firmware.
Firmware requires PCS quad number and PCS port number to
read FEC statistics. Current driver implementation does
not maintain above physical properties of a port.

Add new driver API to derive physical properties of an input
port.These properties include PCS quad number, PCS port number,
serdes lane count, primary serdes lane number.
Extend ethtool option '--show-fec' to support fec statistics.
The IEEE standard mandates two sets of counters:
 - 30.5.1.1.17 aFECCorrectedBlocks
 - 30.5.1.1.18 aFECUncorrectableBlocks

Standard defines above statistics per lane but current
implementation supports total FEC statistics per port
i.e. sum of all lane per port. Find sample output below

FEC parameters for ens21f0np0:
Supported/Configured FEC encodings: Auto RS BaseR
Active FEC encoding: RS
Statistics:
  corrected_blocks: 0
  uncorrectable_blocks: 0

Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Anil Samal <anil.samal@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Link: https://patch.msgid.link/20240709202951.2103115-3-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_common.c  |  57 +++++
 drivers/net/ethernet/intel/ice/ice_common.h  |  24 +++
 drivers/net/ethernet/intel/ice/ice_ethtool.c | 304 +++++++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_ethtool.h |  10 +
 drivers/net/ethernet/intel/ice/ice_type.h    |   8 +
 5 files changed, 403 insertions(+)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 9fa9d914ae46..a4f175cf5c61 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -3371,6 +3371,63 @@ int ice_update_link_info(struct ice_port_info *pi)
 	return status;
 }
 
+#define FEC_REG_PORT(port) {	\
+	FEC_CORR_LOW_REG_PORT##port,		\
+	FEC_CORR_HIGH_REG_PORT##port,	\
+	FEC_UNCORR_LOW_REG_PORT##port,	\
+	FEC_UNCORR_HIGH_REG_PORT##port,	\
+}
+
+static const u32 fec_reg[][ICE_FEC_MAX] = {
+	FEC_REG_PORT(0),
+	FEC_REG_PORT(1),
+	FEC_REG_PORT(2),
+	FEC_REG_PORT(3)
+};
+
+/**
+ * ice_aq_get_fec_stats - reads fec stats from phy
+ * @hw: pointer to the HW struct
+ * @pcs_quad: represents pcsquad of user input serdes
+ * @pcs_port: represents the pcs port number part of above pcs quad
+ * @fec_type: represents FEC stats type
+ * @output: pointer to the caller-supplied buffer to return requested fec stats
+ *
+ * Return: non-zero status on error and 0 on success.
+ */
+int ice_aq_get_fec_stats(struct ice_hw *hw, u16 pcs_quad, u16 pcs_port,
+			 enum ice_fec_stats_types fec_type, u32 *output)
+{
+	u16 flag = (ICE_AQ_FLAG_RD | ICE_AQ_FLAG_BUF | ICE_AQ_FLAG_SI);
+	struct ice_sbq_msg_input msg = {};
+	u32 receiver_id, reg_offset;
+	int err;
+
+	if (pcs_port > 3)
+		return -EINVAL;
+
+	reg_offset = fec_reg[pcs_port][fec_type];
+
+	if (pcs_quad == 0)
+		receiver_id = FEC_RECEIVER_ID_PCS0;
+	else if (pcs_quad == 1)
+		receiver_id = FEC_RECEIVER_ID_PCS1;
+	else
+		return -EINVAL;
+
+	msg.msg_addr_low = lower_16_bits(reg_offset);
+	msg.msg_addr_high = receiver_id;
+	msg.opcode = ice_sbq_msg_rd;
+	msg.dest_dev = rmn_0;
+
+	err = ice_sbq_rw_reg(hw, &msg, flag);
+	if (err)
+		return err;
+
+	*output = msg.data;
+	return 0;
+}
+
 /**
  * ice_cache_phy_user_req
  * @pi: port information structure
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index ef9e8c23580d..bd814e751cbf 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -17,6 +17,27 @@
 #define ICE_SQ_SEND_DELAY_TIME_MS	10
 #define ICE_SQ_SEND_MAX_EXECUTE		3
 
+#define FEC_REG_SHIFT 2
+#define FEC_RECV_ID_SHIFT 4
+#define FEC_CORR_LOW_REG_PORT0 (0x02 << FEC_REG_SHIFT)
+#define FEC_CORR_HIGH_REG_PORT0 (0x03 << FEC_REG_SHIFT)
+#define FEC_UNCORR_LOW_REG_PORT0 (0x04 << FEC_REG_SHIFT)
+#define FEC_UNCORR_HIGH_REG_PORT0 (0x05 << FEC_REG_SHIFT)
+#define FEC_CORR_LOW_REG_PORT1 (0x42 << FEC_REG_SHIFT)
+#define FEC_CORR_HIGH_REG_PORT1 (0x43 << FEC_REG_SHIFT)
+#define FEC_UNCORR_LOW_REG_PORT1 (0x44 << FEC_REG_SHIFT)
+#define FEC_UNCORR_HIGH_REG_PORT1 (0x45 << FEC_REG_SHIFT)
+#define FEC_CORR_LOW_REG_PORT2 (0x4A << FEC_REG_SHIFT)
+#define FEC_CORR_HIGH_REG_PORT2 (0x4B << FEC_REG_SHIFT)
+#define FEC_UNCORR_LOW_REG_PORT2 (0x4C << FEC_REG_SHIFT)
+#define FEC_UNCORR_HIGH_REG_PORT2 (0x4D << FEC_REG_SHIFT)
+#define FEC_CORR_LOW_REG_PORT3 (0x52 << FEC_REG_SHIFT)
+#define FEC_CORR_HIGH_REG_PORT3 (0x53 << FEC_REG_SHIFT)
+#define FEC_UNCORR_LOW_REG_PORT3 (0x54 << FEC_REG_SHIFT)
+#define FEC_UNCORR_HIGH_REG_PORT3 (0x55 << FEC_REG_SHIFT)
+#define FEC_RECEIVER_ID_PCS0 (0x33 << FEC_RECV_ID_SHIFT)
+#define FEC_RECEIVER_ID_PCS1 (0x34 << FEC_RECV_ID_SHIFT)
+
 int ice_init_hw(struct ice_hw *hw);
 void ice_deinit_hw(struct ice_hw *hw);
 int ice_check_reset(struct ice_hw *hw);
@@ -121,6 +142,9 @@ int
 ice_get_link_default_override(struct ice_link_default_override_tlv *ldo,
 			      struct ice_port_info *pi);
 bool ice_is_phy_caps_an_enabled(struct ice_aqc_get_phy_caps_data *caps);
+int
+ice_aq_get_fec_stats(struct ice_hw *hw, u16 pcs_quad, u16 pcs_port,
+		     enum ice_fec_stats_types fec_type, u32 *output);
 
 enum ice_fc_mode ice_caps_to_fc_mode(u8 caps);
 enum ice_fec_mode ice_caps_to_fec_mode(u8 caps, u8 fec_options);
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 62c8205fceba..cb3fff0ff0b8 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -466,6 +466,221 @@ static int ice_get_regs_len(struct net_device __always_unused *netdev)
 	return sizeof(ice_regs_dump_list);
 }
 
+/**
+ * ice_ethtool_get_maxspeed - Get the max speed for given lport
+ * @hw: pointer to the HW struct
+ * @lport: logical port for which max speed is requested
+ * @max_speed: return max speed for input lport
+ *
+ * Return: 0 on success, negative on failure.
+ */
+static int ice_ethtool_get_maxspeed(struct ice_hw *hw, u8 lport, u8 *max_speed)
+{
+	struct ice_aqc_get_port_options_elem options[ICE_AQC_PORT_OPT_MAX] = {};
+	bool active_valid = false, pending_valid = true;
+	u8 option_count = ICE_AQC_PORT_OPT_MAX;
+	u8 active_idx = 0, pending_idx = 0;
+	int status;
+
+	status = ice_aq_get_port_options(hw, options, &option_count, lport,
+					 true, &active_idx, &active_valid,
+					 &pending_idx, &pending_valid);
+	if (status)
+		return -EIO;
+	if (!active_valid)
+		return -EINVAL;
+
+	*max_speed = options[active_idx].max_lane_speed & ICE_AQC_PORT_OPT_MAX_LANE_M;
+	return 0;
+}
+
+/**
+ * ice_is_serdes_muxed - returns whether serdes is muxed in hardware
+ * @hw: pointer to the HW struct
+ *
+ * Return: true when serdes is muxed, false when serdes is not muxed.
+ */
+static bool ice_is_serdes_muxed(struct ice_hw *hw)
+{
+	u32 reg_value = rd32(hw, GLGEN_SWITCH_MODE_CONFIG);
+
+	return FIELD_GET(GLGEN_SWITCH_MODE_CONFIG_25X4_QUAD_M, reg_value);
+}
+
+static int ice_map_port_topology_for_sfp(struct ice_port_topology *port_topology,
+					 u8 lport, bool is_muxed)
+{
+	switch (lport) {
+	case 0:
+		port_topology->pcs_quad_select = 0;
+		port_topology->pcs_port = 0;
+		port_topology->primary_serdes_lane = 0;
+		break;
+	case 1:
+		port_topology->pcs_quad_select = 1;
+		port_topology->pcs_port = 0;
+		if (is_muxed)
+			port_topology->primary_serdes_lane = 2;
+		else
+			port_topology->primary_serdes_lane = 4;
+		break;
+	case 2:
+		port_topology->pcs_quad_select = 0;
+		port_topology->pcs_port = 1;
+		port_topology->primary_serdes_lane = 1;
+		break;
+	case 3:
+		port_topology->pcs_quad_select = 1;
+		port_topology->pcs_port = 1;
+		if (is_muxed)
+			port_topology->primary_serdes_lane = 3;
+		else
+			port_topology->primary_serdes_lane = 5;
+		break;
+	case 4:
+		port_topology->pcs_quad_select = 0;
+		port_topology->pcs_port = 2;
+		port_topology->primary_serdes_lane = 2;
+		break;
+	case 5:
+		port_topology->pcs_quad_select = 1;
+		port_topology->pcs_port = 2;
+		port_topology->primary_serdes_lane = 6;
+		break;
+	case 6:
+		port_topology->pcs_quad_select = 0;
+		port_topology->pcs_port = 3;
+		port_topology->primary_serdes_lane = 3;
+		break;
+	case 7:
+		port_topology->pcs_quad_select = 1;
+		port_topology->pcs_port = 3;
+		port_topology->primary_serdes_lane = 7;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int ice_map_port_topology_for_qsfp(struct ice_port_topology *port_topology,
+					  u8 lport, bool is_muxed)
+{
+	switch (lport) {
+	case 0:
+		port_topology->pcs_quad_select = 0;
+		port_topology->pcs_port = 0;
+		port_topology->primary_serdes_lane = 0;
+		break;
+	case 1:
+		port_topology->pcs_quad_select = 1;
+		port_topology->pcs_port = 0;
+		if (is_muxed)
+			port_topology->primary_serdes_lane = 2;
+		else
+			port_topology->primary_serdes_lane = 4;
+		break;
+	case 2:
+		port_topology->pcs_quad_select = 0;
+		port_topology->pcs_port = 1;
+		port_topology->primary_serdes_lane = 1;
+		break;
+	case 3:
+		port_topology->pcs_quad_select = 1;
+		port_topology->pcs_port = 1;
+		if (is_muxed)
+			port_topology->primary_serdes_lane = 3;
+		else
+			port_topology->primary_serdes_lane = 5;
+		break;
+	case 4:
+		port_topology->pcs_quad_select = 0;
+		port_topology->pcs_port = 2;
+		port_topology->primary_serdes_lane = 2;
+		break;
+	case 5:
+		port_topology->pcs_quad_select = 1;
+		port_topology->pcs_port = 2;
+		port_topology->primary_serdes_lane = 6;
+		break;
+	case 6:
+		port_topology->pcs_quad_select = 0;
+		port_topology->pcs_port = 3;
+		port_topology->primary_serdes_lane = 3;
+		break;
+	case 7:
+		port_topology->pcs_quad_select = 1;
+		port_topology->pcs_port = 3;
+		port_topology->primary_serdes_lane = 7;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/**
+ * ice_get_port_topology - returns physical topology like pcsquad, pcsport,
+ *                         serdes number
+ * @hw: pointer to the HW struct
+ * @lport: logical port for which physical info requested
+ * @port_topology: buffer to hold port topology
+ *
+ * Return: 0 on success, negative on failure.
+ */
+static int ice_get_port_topology(struct ice_hw *hw, u8 lport,
+				 struct ice_port_topology *port_topology)
+{
+	struct ice_aqc_get_link_topo cmd = {};
+	u16 node_handle = 0;
+	u8 cage_type = 0;
+	bool is_muxed;
+	int err;
+	u8 ctx;
+
+	ctx = ICE_AQC_LINK_TOPO_NODE_TYPE_CAGE << ICE_AQC_LINK_TOPO_NODE_TYPE_S;
+	ctx |= ICE_AQC_LINK_TOPO_NODE_CTX_PORT << ICE_AQC_LINK_TOPO_NODE_CTX_S;
+	cmd.addr.topo_params.node_type_ctx = ctx;
+
+	err = ice_aq_get_netlist_node(hw, &cmd, &cage_type, &node_handle);
+	if (err)
+		return -EINVAL;
+
+	is_muxed = ice_is_serdes_muxed(hw);
+
+	if (cage_type == 0x11 ||	/* SFP+ */
+	    cage_type == 0x12) {	/* SFP28 */
+		port_topology->serdes_lane_count = 1;
+		err = ice_map_port_topology_for_sfp(port_topology, lport, is_muxed);
+		if (err)
+			return err;
+	} else if (cage_type == 0x13 ||	/* QSFP */
+		   cage_type == 0x14) {	/* QSFP28 */
+		u8 max_speed = 0;
+
+		err = ice_ethtool_get_maxspeed(hw, lport, &max_speed);
+		if (err)
+			return err;
+
+		if (max_speed == ICE_AQC_PORT_OPT_MAX_LANE_100G)
+			port_topology->serdes_lane_count = 4;
+		else if (max_speed == ICE_AQC_PORT_OPT_MAX_LANE_50G)
+			port_topology->serdes_lane_count = 2;
+		else
+			port_topology->serdes_lane_count = 1;
+
+		err = ice_map_port_topology_for_qsfp(port_topology, lport, is_muxed);
+		if (err)
+			return err;
+	} else {
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 static void
 ice_get_regs(struct net_device *netdev, struct ethtool_regs *regs, void *p)
 {
@@ -4282,6 +4497,94 @@ ice_get_module_eeprom(struct net_device *netdev,
 	return 0;
 }
 
+/**
+ * ice_get_port_fec_stats - returns FEC correctable, uncorrectable stats per
+ *                          pcsquad, pcsport
+ * @hw: pointer to the HW struct
+ * @pcs_quad: pcsquad for input port
+ * @pcs_port: pcsport for input port
+ * @fec_stats: buffer to hold FEC statistics for given port
+ *
+ * Return: 0 on success, negative on failure.
+ */
+static int ice_get_port_fec_stats(struct ice_hw *hw, u16 pcs_quad, u16 pcs_port,
+				  struct ethtool_fec_stats *fec_stats)
+{
+	u32 fec_uncorr_low_val = 0, fec_uncorr_high_val = 0;
+	u32 fec_corr_low_val = 0, fec_corr_high_val = 0;
+	int err;
+
+	if (pcs_quad > 1 || pcs_port > 3)
+		return -EINVAL;
+
+	err = ice_aq_get_fec_stats(hw, pcs_quad, pcs_port, ICE_FEC_CORR_LOW,
+				   &fec_corr_low_val);
+	if (err)
+		return err;
+
+	err = ice_aq_get_fec_stats(hw, pcs_quad, pcs_port, ICE_FEC_CORR_HIGH,
+				   &fec_corr_high_val);
+	if (err)
+		return err;
+
+	err = ice_aq_get_fec_stats(hw, pcs_quad, pcs_port,
+				   ICE_FEC_UNCORR_LOW,
+				   &fec_uncorr_low_val);
+	if (err)
+		return err;
+
+	err = ice_aq_get_fec_stats(hw, pcs_quad, pcs_port,
+				   ICE_FEC_UNCORR_HIGH,
+				   &fec_uncorr_high_val);
+	if (err)
+		return err;
+
+	fec_stats->uncorrectable_blocks.total = (fec_corr_high_val << 16) +
+						 fec_corr_low_val;
+	fec_stats->corrected_blocks.total = (fec_uncorr_high_val << 16) +
+					     fec_uncorr_low_val;
+	return 0;
+}
+
+/**
+ * ice_get_fec_stats - returns FEC correctable, uncorrectable stats per netdev
+ * @netdev: network interface device structure
+ * @fec_stats: buffer to hold FEC statistics for given port
+ *
+ */
+static void ice_get_fec_stats(struct net_device *netdev,
+			      struct ethtool_fec_stats *fec_stats)
+{
+	struct ice_netdev_priv *np = netdev_priv(netdev);
+	struct ice_port_topology port_topology;
+	struct ice_port_info *pi;
+	struct ice_pf *pf;
+	struct ice_hw *hw;
+	int err;
+
+	pf = np->vsi->back;
+	hw = &pf->hw;
+	pi = np->vsi->port_info;
+
+	/* Serdes parameters are not supported if not the PF VSI */
+	if (np->vsi->type != ICE_VSI_PF || !pi)
+		return;
+
+	err = ice_get_port_topology(hw, pi->lport, &port_topology);
+	if (err) {
+		netdev_info(netdev, "Extended register dump failed Lport %d\n",
+			    pi->lport);
+		return;
+	}
+
+	/* Get FEC correctable, uncorrectable counter */
+	err = ice_get_port_fec_stats(hw, port_topology.pcs_quad_select,
+				     port_topology.pcs_port, fec_stats);
+	if (err)
+		netdev_info(netdev, "FEC stats get failed Lport %d Err %d\n",
+			    pi->lport, err);
+}
+
 static const struct ethtool_ops ice_ethtool_ops = {
 	.cap_rss_ctx_supported  = true,
 	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
@@ -4290,6 +4593,7 @@ static const struct ethtool_ops ice_ethtool_ops = {
 	.cap_rss_sym_xor_supported = true,
 	.get_link_ksettings	= ice_get_link_ksettings,
 	.set_link_ksettings	= ice_set_link_ksettings,
+	.get_fec_stats		= ice_get_fec_stats,
 	.get_drvinfo		= ice_get_drvinfo,
 	.get_regs_len		= ice_get_regs_len,
 	.get_regs		= ice_get_regs,
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.h b/drivers/net/ethernet/intel/ice/ice_ethtool.h
index b88e3da06f13..ffc8ad180e61 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.h
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.h
@@ -9,6 +9,16 @@ struct ice_phy_type_to_ethtool {
 	u8 link_mode;
 };
 
+/* Port topology from lport i.e.
+ * serdes mapping, pcsquad, macport, cage etc...
+ */
+struct ice_port_topology {
+	u16 pcs_port;
+	u16 primary_serdes_lane;
+	u16 serdes_lane_count;
+	u16 pcs_quad_select;
+};
+
 /* Macro to make PHY type to Ethtool link mode table entry.
  * The index is the PHY type.
  */
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index aac59c85a911..e26ae79578ba 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -71,6 +71,14 @@ enum ice_aq_res_ids {
 	ICE_GLOBAL_CFG_LOCK_RES_ID
 };
 
+enum ice_fec_stats_types {
+	ICE_FEC_CORR_LOW,
+	ICE_FEC_CORR_HIGH,
+	ICE_FEC_UNCORR_LOW,
+	ICE_FEC_UNCORR_HIGH,
+	ICE_FEC_MAX
+};
+
 /* FW update timeout definitions are in milliseconds */
 #define ICE_NVM_TIMEOUT			180000
 #define ICE_CHANGE_LOCK_TIMEOUT		1000
-- 
cgit v1.2.3-58-ga151


From 70838938e89c759586eb4f3b5ce29ca252c8240d Mon Sep 17 00:00:00 2001
From: Anil Samal <anil.samal@intel.com>
Date: Tue, 9 Jul 2024 13:29:49 -0700
Subject: ice: Implement driver functionality to dump serdes equalizer values

To debug link issues in the field, serdes Tx/Rx equalizer values
help to determine the health of serdes lane.

Extend 'ethtool -d' option to dump serdes Tx/Rx equalizer.
The following list of equalizer param is supported
    a. rx_equalization_pre2
    b. rx_equalization_pre1
    c. rx_equalization_post1
    d. rx_equalization_bflf
    e. rx_equalization_bfhf
    f. rx_equalization_drate
    g. tx_equalization_pre1
    h. tx_equalization_pre3
    i. tx_equalization_atten
    j. tx_equalization_post1
    k. tx_equalization_pre2

Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Anil Samal <anil.samal@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Link: https://patch.msgid.link/20240709202951.2103115-4-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_adminq_cmd.h |  51 +++++++++
 drivers/net/ethernet/intel/ice/ice_common.c     |  37 +++++++
 drivers/net/ethernet/intel/ice/ice_common.h     |   2 +
 drivers/net/ethernet/intel/ice/ice_ethtool.c    | 138 +++++++++++++++++++++++-
 drivers/net/ethernet/intel/ice/ice_ethtool.h    |  19 ++++
 5 files changed, 245 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index 621a2ca7093e..b70d4ca43443 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -1461,6 +1461,55 @@ struct ice_aqc_get_sensor_reading_resp {
 	} data;
 };
 
+/* DNL call command (indirect 0x0682)
+ * Struct is used for both command and response
+ */
+struct ice_aqc_dnl_call_command {
+	u8 ctx; /* Used in command, reserved in response */
+	u8 reserved;
+	__le16 activity_id;
+#define ICE_AQC_ACT_ID_DNL 0x1129
+	__le32 reserved1;
+	__le32 addr_high;
+	__le32 addr_low;
+};
+
+struct ice_aqc_dnl_equa_param {
+	__le16 data_in;
+#define ICE_AQC_RX_EQU_SHIFT 8
+#define ICE_AQC_RX_EQU_PRE2 (0x10 << ICE_AQC_RX_EQU_SHIFT)
+#define ICE_AQC_RX_EQU_PRE1 (0x11 << ICE_AQC_RX_EQU_SHIFT)
+#define ICE_AQC_RX_EQU_POST1 (0x12 << ICE_AQC_RX_EQU_SHIFT)
+#define ICE_AQC_RX_EQU_BFLF (0x13 << ICE_AQC_RX_EQU_SHIFT)
+#define ICE_AQC_RX_EQU_BFHF (0x14 << ICE_AQC_RX_EQU_SHIFT)
+#define ICE_AQC_RX_EQU_DRATE (0x15 << ICE_AQC_RX_EQU_SHIFT)
+#define ICE_AQC_TX_EQU_PRE1 0x0
+#define ICE_AQC_TX_EQU_PRE3 0x3
+#define ICE_AQC_TX_EQU_ATTEN 0x4
+#define ICE_AQC_TX_EQU_POST1 0x8
+#define ICE_AQC_TX_EQU_PRE2 0xC
+	__le16 op_code_serdes_sel;
+#define ICE_AQC_OP_CODE_SHIFT 4
+#define ICE_AQC_OP_CODE_RX_EQU (0x9 << ICE_AQC_OP_CODE_SHIFT)
+#define ICE_AQC_OP_CODE_TX_EQU (0x10 << ICE_AQC_OP_CODE_SHIFT)
+	__le32 reserved[3];
+};
+
+struct ice_aqc_dnl_equa_respon {
+	/* Equalization value can be negative */
+	int val;
+	__le32 reserved[3];
+};
+
+/* DNL call command/response buffer (indirect 0x0682) */
+struct ice_aqc_dnl_call {
+	union {
+		struct ice_aqc_dnl_equa_param txrx_equa_reqs;
+		__le32 stores[4];
+		struct ice_aqc_dnl_equa_respon txrx_equa_resp;
+	} sto;
+};
+
 struct ice_aqc_link_topo_params {
 	u8 lport_num;
 	u8 lport_num_valid;
@@ -2564,6 +2613,7 @@ struct ice_aq_desc {
 		struct ice_aqc_get_link_status get_link_status;
 		struct ice_aqc_event_lan_overflow lan_overflow;
 		struct ice_aqc_get_link_topo get_link_topo;
+		struct ice_aqc_dnl_call_command dnl_call;
 		struct ice_aqc_i2c read_write_i2c;
 		struct ice_aqc_read_i2c_resp read_i2c_resp;
 		struct ice_aqc_get_set_tx_topo get_set_tx_topo;
@@ -2688,6 +2738,7 @@ enum ice_adminq_opc {
 	ice_aqc_opc_set_phy_rec_clk_out			= 0x0630,
 	ice_aqc_opc_get_phy_rec_clk_out			= 0x0631,
 	ice_aqc_opc_get_sensor_reading			= 0x0632,
+	ice_aqc_opc_dnl_call                            = 0x0682,
 	ice_aqc_opc_get_link_topo			= 0x06E0,
 	ice_aqc_opc_read_i2c				= 0x06E2,
 	ice_aqc_opc_write_i2c				= 0x06E3,
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index a4f175cf5c61..e311a41a74fa 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -3371,6 +3371,43 @@ int ice_update_link_info(struct ice_port_info *pi)
 	return status;
 }
 
+/**
+ * ice_aq_get_phy_equalization - function to read serdes equaliser
+ * value from firmware using admin queue command.
+ * @hw: pointer to the HW struct
+ * @data_in: represents the serdes equalization parameter requested
+ * @op_code: represents the serdes number and flag to represent tx or rx
+ * @serdes_num: represents the serdes number
+ * @output: pointer to the caller-supplied buffer to return serdes equaliser
+ *
+ * Return: non-zero status on error and 0 on success.
+ */
+int ice_aq_get_phy_equalization(struct ice_hw *hw, u16 data_in, u16 op_code,
+				u8 serdes_num, int *output)
+{
+	struct ice_aqc_dnl_call_command *cmd;
+	struct ice_aqc_dnl_call buf = {};
+	struct ice_aq_desc desc;
+	int err;
+
+	buf.sto.txrx_equa_reqs.data_in = cpu_to_le16(data_in);
+	buf.sto.txrx_equa_reqs.op_code_serdes_sel =
+		cpu_to_le16(op_code | (serdes_num & 0xF));
+	cmd = &desc.params.dnl_call;
+	ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_dnl_call);
+	desc.flags |= cpu_to_le16(ICE_AQ_FLAG_BUF |
+				  ICE_AQ_FLAG_RD |
+				  ICE_AQ_FLAG_SI);
+	desc.datalen = cpu_to_le16(sizeof(struct ice_aqc_dnl_call));
+	cmd->activity_id = cpu_to_le16(ICE_AQC_ACT_ID_DNL);
+
+	err = ice_aq_send_cmd(hw, &desc, &buf, sizeof(struct ice_aqc_dnl_call),
+			      NULL);
+	*output = err ? 0 : buf.sto.txrx_equa_resp.val;
+
+	return err;
+}
+
 #define FEC_REG_PORT(port) {	\
 	FEC_CORR_LOW_REG_PORT##port,		\
 	FEC_CORR_HIGH_REG_PORT##port,	\
diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h
index bd814e751cbf..66f29bac783a 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.h
+++ b/drivers/net/ethernet/intel/ice/ice_common.h
@@ -142,6 +142,8 @@ int
 ice_get_link_default_override(struct ice_link_default_override_tlv *ldo,
 			      struct ice_port_info *pi);
 bool ice_is_phy_caps_an_enabled(struct ice_aqc_get_phy_caps_data *caps);
+int ice_aq_get_phy_equalization(struct ice_hw *hw, u16 data_in, u16 op_code,
+				u8 serdes_num, int *output);
 int
 ice_aq_get_fec_stats(struct ice_hw *hw, u16 pcs_quad, u16 pcs_port,
 		     enum ice_fec_stats_types fec_type, u32 *output);
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index cb3fff0ff0b8..2b4f6f07737c 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -463,7 +463,8 @@ ice_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo)
 
 static int ice_get_regs_len(struct net_device __always_unused *netdev)
 {
-	return sizeof(ice_regs_dump_list);
+	return (sizeof(ice_regs_dump_list) +
+		sizeof(struct ice_regdump_to_ethtool));
 }
 
 /**
@@ -681,6 +682,137 @@ static int ice_get_port_topology(struct ice_hw *hw, u8 lport,
 	return 0;
 }
 
+/**
+ * ice_get_tx_rx_equa - read serdes tx rx equaliser param
+ * @hw: pointer to the HW struct
+ * @serdes_num: represents the serdes number
+ * @ptr: structure to read all serdes parameter for given serdes
+ *
+ * Return: all serdes equalization parameter supported per serdes number
+ */
+static int ice_get_tx_rx_equa(struct ice_hw *hw, u8 serdes_num,
+			      struct ice_serdes_equalization_to_ethtool *ptr)
+{
+	int err;
+
+	err = ice_aq_get_phy_equalization(hw, ICE_AQC_TX_EQU_PRE1,
+					  ICE_AQC_OP_CODE_TX_EQU, serdes_num,
+					  &ptr->tx_equalization_pre1);
+	if (err)
+		return err;
+
+	err = ice_aq_get_phy_equalization(hw, ICE_AQC_TX_EQU_PRE3,
+					  ICE_AQC_OP_CODE_TX_EQU, serdes_num,
+					  &ptr->tx_equalization_pre3);
+	if (err)
+		return err;
+
+	err = ice_aq_get_phy_equalization(hw, ICE_AQC_TX_EQU_ATTEN,
+					  ICE_AQC_OP_CODE_TX_EQU, serdes_num,
+					  &ptr->tx_equalization_atten);
+	if (err)
+		return err;
+
+	err = ice_aq_get_phy_equalization(hw, ICE_AQC_TX_EQU_POST1,
+					  ICE_AQC_OP_CODE_TX_EQU, serdes_num,
+					  &ptr->tx_equalization_post1);
+	if (err)
+		return err;
+
+	err = ice_aq_get_phy_equalization(hw, ICE_AQC_TX_EQU_PRE2,
+					  ICE_AQC_OP_CODE_TX_EQU, serdes_num,
+					  &ptr->tx_equalization_pre2);
+	if (err)
+		return err;
+
+	err = ice_aq_get_phy_equalization(hw, ICE_AQC_RX_EQU_PRE2,
+					  ICE_AQC_OP_CODE_RX_EQU, serdes_num,
+					  &ptr->rx_equalization_pre2);
+	if (err)
+		return err;
+
+	err = ice_aq_get_phy_equalization(hw, ICE_AQC_RX_EQU_PRE1,
+					  ICE_AQC_OP_CODE_RX_EQU, serdes_num,
+					  &ptr->rx_equalization_pre1);
+	if (err)
+		return err;
+
+	err = ice_aq_get_phy_equalization(hw, ICE_AQC_RX_EQU_POST1,
+					  ICE_AQC_OP_CODE_RX_EQU, serdes_num,
+					  &ptr->rx_equalization_post1);
+	if (err)
+		return err;
+
+	err = ice_aq_get_phy_equalization(hw, ICE_AQC_RX_EQU_BFLF,
+					  ICE_AQC_OP_CODE_RX_EQU, serdes_num,
+					  &ptr->rx_equalization_bflf);
+	if (err)
+		return err;
+
+	err = ice_aq_get_phy_equalization(hw, ICE_AQC_RX_EQU_BFHF,
+					  ICE_AQC_OP_CODE_RX_EQU, serdes_num,
+					  &ptr->rx_equalization_bfhf);
+	if (err)
+		return err;
+
+	err = ice_aq_get_phy_equalization(hw, ICE_AQC_RX_EQU_DRATE,
+					  ICE_AQC_OP_CODE_RX_EQU, serdes_num,
+					  &ptr->rx_equalization_drate);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+/**
+ * ice_get_extended_regs - returns FEC correctable, uncorrectable stats per
+ *                         pcsquad, pcsport
+ * @netdev: pointer to net device structure
+ * @p: output buffer to fill requested register dump
+ *
+ * Return: 0 on success, negative on failure.
+ */
+static int ice_get_extended_regs(struct net_device *netdev, void *p)
+{
+	struct ice_netdev_priv *np = netdev_priv(netdev);
+	struct ice_regdump_to_ethtool *ice_prv_regs_buf;
+	struct ice_port_topology port_topology = {};
+	struct ice_port_info *pi;
+	struct ice_pf *pf;
+	struct ice_hw *hw;
+	unsigned int i;
+	int err;
+
+	pf = np->vsi->back;
+	hw = &pf->hw;
+	pi = np->vsi->port_info;
+
+	/* Serdes parameters are not supported if not the PF VSI */
+	if (np->vsi->type != ICE_VSI_PF || !pi)
+		return -EINVAL;
+
+	err = ice_get_port_topology(hw, pi->lport, &port_topology);
+	if (err)
+		return -EINVAL;
+	if (port_topology.serdes_lane_count > 4)
+		return -EINVAL;
+
+	ice_prv_regs_buf = p;
+
+	/* Get serdes equalization parameter for available serdes */
+	for (i = 0; i < port_topology.serdes_lane_count; i++) {
+		u8 serdes_num = 0;
+
+		serdes_num = port_topology.primary_serdes_lane + i;
+		err = ice_get_tx_rx_equa(hw, serdes_num,
+					 &ice_prv_regs_buf->equalization[i]);
+		if (err)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
 static void
 ice_get_regs(struct net_device *netdev, struct ethtool_regs *regs, void *p)
 {
@@ -690,10 +822,12 @@ ice_get_regs(struct net_device *netdev, struct ethtool_regs *regs, void *p)
 	u32 *regs_buf = (u32 *)p;
 	unsigned int i;
 
-	regs->version = 1;
+	regs->version = 2;
 
 	for (i = 0; i < ARRAY_SIZE(ice_regs_dump_list); ++i)
 		regs_buf[i] = rd32(hw, ice_regs_dump_list[i]);
+
+	ice_get_extended_regs(netdev, (void *)&regs_buf[i]);
 }
 
 static u32 ice_get_msglevel(struct net_device *netdev)
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.h b/drivers/net/ethernet/intel/ice/ice_ethtool.h
index ffc8ad180e61..9acccae38625 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.h
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.h
@@ -9,6 +9,25 @@ struct ice_phy_type_to_ethtool {
 	u8 link_mode;
 };
 
+struct ice_serdes_equalization_to_ethtool {
+	int rx_equalization_pre2;
+	int rx_equalization_pre1;
+	int rx_equalization_post1;
+	int rx_equalization_bflf;
+	int rx_equalization_bfhf;
+	int rx_equalization_drate;
+	int tx_equalization_pre1;
+	int tx_equalization_pre3;
+	int tx_equalization_atten;
+	int tx_equalization_post1;
+	int tx_equalization_pre2;
+};
+
+struct ice_regdump_to_ethtool {
+	/* A multilane port can have max 4 serdes */
+	struct ice_serdes_equalization_to_ethtool equalization[4];
+};
+
 /* Port topology from lport i.e.
  * serdes mapping, pcsquad, macport, cage etc...
  */
-- 
cgit v1.2.3-58-ga151


From e247267bbeaf0d904b95765b2721495f6581eb65 Mon Sep 17 00:00:00 2001
From: Marcin Szycik <marcin.szycik@linux.intel.com>
Date: Thu, 27 Jun 2024 16:55:41 +0200
Subject: ice: Remove unused struct ice_prot_lkup_ext members

Remove field_off as it's never used.

Remove done bitmap, as its value is only checked and never assigned.
Reusing sub-recipes while creating new root recipes is currently not
supported in the driver.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_protocol_type.h |  4 --
 drivers/net/ethernet/intel/ice/ice_switch.c        | 44 ++++++++++------------
 2 files changed, 19 insertions(+), 29 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_protocol_type.h b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
index 755a9c55267c..c396dabacef4 100644
--- a/drivers/net/ethernet/intel/ice/ice_protocol_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
@@ -452,13 +452,9 @@ struct ice_prot_lkup_ext {
 	u16 prot_type;
 	u8 n_val_words;
 	/* create a buffer to hold max words per recipe */
-	u16 field_off[ICE_MAX_CHAIN_WORDS];
 	u16 field_mask[ICE_MAX_CHAIN_WORDS];
 
 	struct ice_fv_word fv_words[ICE_MAX_CHAIN_WORDS];
-
-	/* Indicate field offsets that have field vector indices assigned */
-	DECLARE_BITMAP(done, ICE_MAX_CHAIN_WORDS);
 };
 
 struct ice_pref_recipe_group {
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index 1191031b2a43..0d58cf185698 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -4918,33 +4918,27 @@ ice_create_first_fit_recp_def(struct ice_hw *hw,
 
 	*recp_cnt = 0;
 
-	/* Walk through every word in the rule to check if it is not done. If so
-	 * then this word needs to be part of a new recipe.
-	 */
-	for (j = 0; j < lkup_exts->n_val_words; j++)
-		if (!test_bit(j, lkup_exts->done)) {
-			if (!grp ||
-			    grp->n_val_pairs == ICE_NUM_WORDS_RECIPE) {
-				struct ice_recp_grp_entry *entry;
-
-				entry = devm_kzalloc(ice_hw_to_dev(hw),
-						     sizeof(*entry),
-						     GFP_KERNEL);
-				if (!entry)
-					return -ENOMEM;
-				list_add(&entry->l_entry, rg_list);
-				grp = &entry->r_group;
-				(*recp_cnt)++;
-			}
-
-			grp->pairs[grp->n_val_pairs].prot_id =
-				lkup_exts->fv_words[j].prot_id;
-			grp->pairs[grp->n_val_pairs].off =
-				lkup_exts->fv_words[j].off;
-			grp->mask[grp->n_val_pairs] = lkup_exts->field_mask[j];
-			grp->n_val_pairs++;
+	for (j = 0; j < lkup_exts->n_val_words; j++) {
+		if (!grp || grp->n_val_pairs == ICE_NUM_WORDS_RECIPE) {
+			struct ice_recp_grp_entry *entry;
+
+			entry = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*entry),
+					     GFP_KERNEL);
+			if (!entry)
+				return -ENOMEM;
+
+			list_add(&entry->l_entry, rg_list);
+			grp = &entry->r_group;
+			(*recp_cnt)++;
 		}
 
+		grp->pairs[grp->n_val_pairs].prot_id =
+			lkup_exts->fv_words[j].prot_id;
+		grp->pairs[grp->n_val_pairs].off = lkup_exts->fv_words[j].off;
+		grp->mask[grp->n_val_pairs] = lkup_exts->field_mask[j];
+		grp->n_val_pairs++;
+	}
+
 	return 0;
 }
 
-- 
cgit v1.2.3-58-ga151


From c563908494e9578da70a11d49f6ad75c56fe8d48 Mon Sep 17 00:00:00 2001
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Date: Thu, 27 Jun 2024 16:55:42 +0200
Subject: ice: Remove reading all recipes before adding a new one

The content of the first read recipe is used as a template when adding a
recipe. It isn't needed - only prune index is directly set from there. Set
it in the code instead. Also, now there's no need to set rid and lookup
indexes to 0, as the whole recipe buffer is initialized to 0.

Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_switch.c | 29 ++---------------------------
 1 file changed, 2 insertions(+), 27 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index 0d58cf185698..da065512889d 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -5079,11 +5079,9 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 {
 	DECLARE_BITMAP(result_idx_bm, ICE_MAX_FV_WORDS);
 	struct ice_aqc_recipe_content *content;
-	struct ice_aqc_recipe_data_elem *tmp;
 	struct ice_aqc_recipe_data_elem *buf;
 	struct ice_recp_grp_entry *entry;
 	u16 free_res_idx;
-	u16 recipe_count;
 	u8 chain_idx;
 	u8 recps = 0;
 	int status;
@@ -5110,10 +5108,6 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 	if (rm->n_grp_count > ICE_MAX_CHAIN_RECIPE)
 		return -ENOSPC;
 
-	tmp = kcalloc(ICE_MAX_NUM_RECIPES, sizeof(*tmp), GFP_KERNEL);
-	if (!tmp)
-		return -ENOMEM;
-
 	buf = devm_kcalloc(ice_hw_to_dev(hw), rm->n_grp_count, sizeof(*buf),
 			   GFP_KERNEL);
 	if (!buf) {
@@ -5122,11 +5116,6 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 	}
 
 	bitmap_zero(rm->r_bitmap, ICE_MAX_NUM_RECIPES);
-	recipe_count = ICE_MAX_NUM_RECIPES;
-	status = ice_aq_get_recipe(hw, tmp, &recipe_count, ICE_SW_LKUP_MAC,
-				   NULL);
-	if (status || recipe_count == 0)
-		goto err_unroll;
 
 	/* Allocate the recipe resources, and configure them according to the
 	 * match fields from protocol headers and extracted field vectors.
@@ -5141,19 +5130,9 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 
 		content = &buf[recps].content;
 
-		/* Clear the result index of the located recipe, as this will be
-		 * updated, if needed, later in the recipe creation process.
-		 */
-		tmp[0].content.result_indx = 0;
-
-		buf[recps] = tmp[0];
 		buf[recps].recipe_indx = (u8)entry->rid;
-		/* if the recipe is a non-root recipe RID should be programmed
-		 * as 0 for the rules to be applied correctly.
-		 */
-		content->rid = 0;
-		memset(&content->lkup_indx, 0,
-		       sizeof(content->lkup_indx));
+
+		buf[recps].content.act_ctrl |= ICE_AQ_RECIPE_ACT_PRUNE_INDX_M;
 
 		/* All recipes use look-up index 0 to match switch ID. */
 		content->lkup_indx[0] = ICE_AQ_SW_ID_LKUP_IDX;
@@ -5192,8 +5171,6 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 		}
 
 		/* fill recipe dependencies */
-		bitmap_zero((unsigned long *)buf[recps].recipe_bitmap,
-			    ICE_MAX_NUM_RECIPES);
 		set_bit(buf[recps].recipe_indx,
 			(unsigned long *)buf[recps].recipe_bitmap);
 		content->act_ctrl_fwd_priority = rm->priority;
@@ -5357,12 +5334,10 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 		recp->recp_created = true;
 	}
 	rm->root_buf = buf;
-	kfree(tmp);
 	return status;
 
 err_unroll:
 err_mem:
-	kfree(tmp);
 	devm_kfree(ice_hw_to_dev(hw), buf);
 	return status;
 }
-- 
cgit v1.2.3-58-ga151


From 3125eb5595904c41f93587e35712acffa51e14c8 Mon Sep 17 00:00:00 2001
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Date: Thu, 27 Jun 2024 16:55:43 +0200
Subject: ice: Simplify bitmap setting in adding recipe

Remove unnecessary size checks when copying bitmaps in ice_add_sw_recipe()
and replace them with compile time assert. Check if the bitmaps are equal
size, as they are copied both ways.

Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Co-developed-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_switch.c | 24 +++++++++---------------
 1 file changed, 9 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index da065512889d..3ee4242f9880 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -5067,6 +5067,10 @@ ice_find_free_recp_res_idx(struct ice_hw *hw, const unsigned long *profiles,
 	return (u16)bitmap_weight(free_idx, ICE_MAX_FV_WORDS);
 }
 
+/* For memcpy in ice_add_sw_recipe. */
+static_assert(sizeof_field(struct ice_aqc_recipe_data_elem, recipe_bitmap) ==
+	      sizeof_field(struct ice_sw_recipe, r_bitmap));
+
 /**
  * ice_add_sw_recipe - function to call AQ calls to create switch recipe
  * @hw: pointer to hardware structure
@@ -5187,13 +5191,9 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 		rm->root_rid = buf[0].recipe_indx;
 		set_bit(buf[0].recipe_indx, rm->r_bitmap);
 		buf[0].content.rid = rm->root_rid | ICE_AQ_RECIPE_ID_IS_ROOT;
-		if (sizeof(buf[0].recipe_bitmap) >= sizeof(rm->r_bitmap)) {
-			memcpy(buf[0].recipe_bitmap, rm->r_bitmap,
-			       sizeof(buf[0].recipe_bitmap));
-		} else {
-			status = -EINVAL;
-			goto err_unroll;
-		}
+		memcpy(buf[0].recipe_bitmap, rm->r_bitmap,
+		       sizeof(buf[0].recipe_bitmap));
+
 		/* Applicable only for ROOT_RECIPE, set the fwd_priority for
 		 * the recipe which is getting created if specified
 		 * by user. Usually any advanced switch filter, which results
@@ -5256,14 +5256,8 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 			set_bit(entry->rid, rm->r_bitmap);
 		}
 		list_add(&last_chain_entry->l_entry, &rm->rg_list);
-		if (sizeof(buf[recps].recipe_bitmap) >=
-		    sizeof(rm->r_bitmap)) {
-			memcpy(buf[recps].recipe_bitmap, rm->r_bitmap,
-			       sizeof(buf[recps].recipe_bitmap));
-		} else {
-			status = -EINVAL;
-			goto err_unroll;
-		}
+		memcpy(buf[recps].recipe_bitmap, rm->r_bitmap,
+		       sizeof(buf[recps].recipe_bitmap));
 		content->act_ctrl_fwd_priority = rm->priority;
 
 		recps++;
-- 
cgit v1.2.3-58-ga151


From 19d3c179a37730caf600a97fed3794feac2b197b Mon Sep 17 00:00:00 2001
From: Puranjay Mohan <puranjay@kernel.org>
Date: Thu, 11 Jul 2024 15:18:38 +0000
Subject: bpf, arm64: Fix trampoline for BPF_TRAMP_F_CALL_ORIG

When BPF_TRAMP_F_CALL_ORIG is set, the trampoline calls
__bpf_tramp_enter() and __bpf_tramp_exit() functions, passing them
the struct bpf_tramp_image *im pointer as an argument in R0.

The trampoline generation code uses emit_addr_mov_i64() to emit
instructions for moving the bpf_tramp_image address into R0, but
emit_addr_mov_i64() assumes the address to be in the vmalloc() space
and uses only 48 bits. Because bpf_tramp_image is allocated using
kzalloc(), its address can use more than 48-bits, in this case the
trampoline will pass an invalid address to __bpf_tramp_enter/exit()
causing a kernel crash.

Fix this by using emit_a64_mov_i64() in place of emit_addr_mov_i64()
as it can work with addresses that are greater than 48-bits.

Fixes: efc9909fdce0 ("bpf, arm64: Add bpf trampoline for arm64")
Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Closes: https://lore.kernel.org/all/SJ0PR15MB461564D3F7E7A763498CA6A8CBDB2@SJ0PR15MB4615.namprd15.prod.outlook.com/
Link: https://lore.kernel.org/bpf/20240711151838.43469-1-puranjay@kernel.org
---
 arch/arm64/net/bpf_jit_comp.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c
index 751331f5ba90..dd0bb069df4b 100644
--- a/arch/arm64/net/bpf_jit_comp.c
+++ b/arch/arm64/net/bpf_jit_comp.c
@@ -2147,7 +2147,7 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 	emit(A64_STR64I(A64_R(20), A64_SP, regs_off + 8), ctx);
 
 	if (flags & BPF_TRAMP_F_CALL_ORIG) {
-		emit_addr_mov_i64(A64_R(0), (const u64)im, ctx);
+		emit_a64_mov_i64(A64_R(0), (const u64)im, ctx);
 		emit_call((const u64)__bpf_tramp_enter, ctx);
 	}
 
@@ -2191,7 +2191,7 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 
 	if (flags & BPF_TRAMP_F_CALL_ORIG) {
 		im->ip_epilogue = ctx->ro_image + ctx->idx;
-		emit_addr_mov_i64(A64_R(0), (const u64)im, ctx);
+		emit_a64_mov_i64(A64_R(0), (const u64)im, ctx);
 		emit_call((const u64)__bpf_tramp_exit, ctx);
 	}
 
-- 
cgit v1.2.3-58-ga151


From 589dd7145a8ea6bcc6faa286205e467ba5403f79 Mon Sep 17 00:00:00 2001
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Date: Thu, 27 Jun 2024 16:55:44 +0200
Subject: ice: remove unused recipe bookkeeping data

Remove root_buf from recipe struct. Its only usage was in ice_find_recp(),
where if recipe had an inverse action, it was skipped, but actually the
driver never adds inverse actions, so effectively it was pointless.

Without root_buf, the recipe data element in ice_add_sw_recipe() does
not need to be persistent and can also be automatically deallocated with
__free, which nicely simplifies unroll.

Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_common.c |  1 -
 drivers/net/ethernet/intel/ice/ice_switch.c | 55 ++++++++---------------------
 drivers/net/ethernet/intel/ice/ice_switch.h |  2 --
 3 files changed, 15 insertions(+), 43 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index e311a41a74fa..17f6626fec7e 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -993,7 +993,6 @@ static void ice_cleanup_fltr_mgmt_struct(struct ice_hw *hw)
 				devm_kfree(ice_hw_to_dev(hw), lst_itr);
 			}
 		}
-		devm_kfree(ice_hw_to_dev(hw), recps[i].root_buf);
 	}
 	ice_rm_all_sw_replay_rule_info(hw);
 	devm_kfree(ice_hw_to_dev(hw), sw->recp_list);
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index 3ee4242f9880..81e5448da1aa 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -2445,13 +2445,6 @@ ice_get_recp_frm_fw(struct ice_hw *hw, struct ice_sw_recipe *recps, u8 rid,
 	lkup_exts->n_val_words = fv_word_idx;
 	recps[rid].big_recp = (num_recps > 1);
 	recps[rid].n_grp_count = (u8)num_recps;
-	recps[rid].root_buf = devm_kmemdup(ice_hw_to_dev(hw), tmp,
-					   recps[rid].n_grp_count * sizeof(*recps[rid].root_buf),
-					   GFP_KERNEL);
-	if (!recps[rid].root_buf) {
-		status = -ENOMEM;
-		goto err_unroll;
-	}
 
 	/* Copy result indexes */
 	bitmap_copy(recps[rid].res_idxs, result_bm, ICE_MAX_FV_WORDS);
@@ -4768,11 +4761,6 @@ ice_find_recp(struct ice_hw *hw, struct ice_prot_lkup_ext *lkup_exts,
 				continue;
 		}
 
-		/* Skip inverse action recipes */
-		if (recp[i].root_buf && recp[i].root_buf->content.act_ctrl &
-		    ICE_AQ_RECIPE_ACT_INV_ACT)
-			continue;
-
 		/* if number of words we are looking for match */
 		if (lkup_exts->n_val_words == recp[i].lkup_exts.n_val_words) {
 			struct ice_fv_word *ar = recp[i].lkup_exts.fv_words;
@@ -5081,9 +5069,9 @@ static int
 ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 		  unsigned long *profiles)
 {
+	struct ice_aqc_recipe_data_elem *buf __free(kfree) = NULL;
 	DECLARE_BITMAP(result_idx_bm, ICE_MAX_FV_WORDS);
 	struct ice_aqc_recipe_content *content;
-	struct ice_aqc_recipe_data_elem *buf;
 	struct ice_recp_grp_entry *entry;
 	u16 free_res_idx;
 	u8 chain_idx;
@@ -5112,12 +5100,9 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 	if (rm->n_grp_count > ICE_MAX_CHAIN_RECIPE)
 		return -ENOSPC;
 
-	buf = devm_kcalloc(ice_hw_to_dev(hw), rm->n_grp_count, sizeof(*buf),
-			   GFP_KERNEL);
-	if (!buf) {
-		status = -ENOMEM;
-		goto err_mem;
-	}
+	buf = kcalloc(rm->n_grp_count, sizeof(*buf), GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
 
 	bitmap_zero(rm->r_bitmap, ICE_MAX_NUM_RECIPES);
 
@@ -5130,7 +5115,7 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 
 		status = ice_alloc_recipe(hw, &entry->rid);
 		if (status)
-			goto err_unroll;
+			return status;
 
 		content = &buf[recps].content;
 
@@ -5160,8 +5145,7 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 			 */
 			if (chain_idx >= ICE_MAX_FV_WORDS) {
 				ice_debug(hw, ICE_DBG_SW, "No chain index available\n");
-				status = -ENOSPC;
-				goto err_unroll;
+				return -ENOSPC;
 			}
 
 			entry->chain_idx = chain_idx;
@@ -5215,7 +5199,7 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 		 */
 		status = ice_alloc_recipe(hw, &rid);
 		if (status)
-			goto err_unroll;
+			return status;
 
 		content = &buf[recps].content;
 
@@ -5228,10 +5212,9 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 		last_chain_entry = devm_kzalloc(ice_hw_to_dev(hw),
 						sizeof(*last_chain_entry),
 						GFP_KERNEL);
-		if (!last_chain_entry) {
-			status = -ENOMEM;
-			goto err_unroll;
-		}
+		if (!last_chain_entry)
+			return -ENOMEM;
+
 		last_chain_entry->rid = rid;
 		memset(&content->lkup_indx, 0, sizeof(content->lkup_indx));
 		/* All recipes use look-up index 0 to match switch ID. */
@@ -5265,12 +5248,12 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 	}
 	status = ice_acquire_change_lock(hw, ICE_RES_WRITE);
 	if (status)
-		goto err_unroll;
+		return status;
 
 	status = ice_aq_add_recipe(hw, buf, rm->n_grp_count, NULL);
 	ice_release_change_lock(hw);
 	if (status)
-		goto err_unroll;
+		return status;
 
 	/* Every recipe that just got created add it to the recipe
 	 * book keeping list
@@ -5288,10 +5271,8 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 				idx_found = true;
 			}
 
-		if (!idx_found) {
-			status = -EIO;
-			goto err_unroll;
-		}
+		if (!idx_found)
+			return -EIO;
 
 		recp = &sw->recp_list[entry->rid];
 		is_root = (rm->root_rid == entry->rid);
@@ -5327,13 +5308,8 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 		recp->allow_pass_l2 = rm->allow_pass_l2;
 		recp->recp_created = true;
 	}
-	rm->root_buf = buf;
-	return status;
 
-err_unroll:
-err_mem:
-	devm_kfree(ice_hw_to_dev(hw), buf);
-	return status;
+	return 0;
 }
 
 /**
@@ -5632,7 +5608,6 @@ err_unroll:
 		devm_kfree(ice_hw_to_dev(hw), fvit);
 	}
 
-	devm_kfree(ice_hw_to_dev(hw), rm->root_buf);
 	kfree(rm);
 
 err_free_lkup_exts:
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.h b/drivers/net/ethernet/intel/ice/ice_switch.h
index ad98e98c812d..0c410ce29700 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.h
+++ b/drivers/net/ethernet/intel/ice/ice_switch.h
@@ -274,8 +274,6 @@ struct ice_sw_recipe {
 
 	struct list_head rg_list;
 
-	/* AQ buffer associated with this recipe */
-	struct ice_aqc_recipe_data_elem *root_buf;
 	/* This struct saves the fv_words for a given lookup */
 	struct ice_prot_lkup_ext lkup_exts;
 };
-- 
cgit v1.2.3-58-ga151


From 2ecdd4ba47fc499ff610d49cb42df7ef27d0292e Mon Sep 17 00:00:00 2001
From: Marcin Szycik <marcin.szycik@linux.intel.com>
Date: Thu, 27 Jun 2024 16:55:45 +0200
Subject: ice: Optimize switch recipe creation

Currently when creating switch recipes, switch ID is always added as the
first word in every recipe. There are only 5 words in a recipe, so one
word is always wasted. This is also true for the last recipe, which stores
result indexes (in case of chain recipes). Therefore the maximum usable
length of a chain recipe is 4 * 4 = 16 words. 4 words in a recipe, 4
recipes that can be chained (using a 5th one for result indexes).

Current max size chained recipe:
0: smmmm
1: smmmm
2: smmmm
3: smmmm
4: srrrr

Where:
s - switch ID
m - regular match (e.g. ipv4 src addr, udp dst port, etc.)
r - result index

Switch ID does not actually need to be present in every recipe, only in one
of them (in case of chained recipe). This frees up to 8 extra words:
3 from recipes in the middle (because first recipe still needs to have
switch ID), and 5 from one extra recipe (because now the last recipe also
does not have switch ID, so it can chain 1 more recipe).

Max size chained recipe after changes:
0: smmmm
1: Mmmmm
2: Mmmmm
3: Mmmmm
4: MMMMM
5: Rrrrr

Extra usable words available after this change are highlighted with capital
letters.

Changing how switch ID is added is not straightforward, because it's not a
regular lookup. Its FV index and mask can't be determined based on protocol
+ offset pair read from package and instead need to be added manually.

Additionally, change how result indexes are added. Currently they are
always inserted in a new recipe at the end. Example for 13 words, (with
above optimization, switch ID being one of the words):
0: smmmm
1: mmmmm
2: mmmxx
3: rrrxx

Where:
x - unused word

In this and some other cases, the result indexes can be moved just after
last matches because there are unused words, saving one recipe. Example
for 13 words after both optimizations:
0: smmmm
1: mmmmm
2: mmmrr

Note how one less result index is needed in this case, because the last
recipe does not need to "link" to itself.

There are cases when adding an additional recipe for result indexes cannot
be avoided. In that cases result indexes are all put in the last recipe.
Example for 14 words after both optimizations:
0: smmmm
1: mmmmm
2: mmmmx
3: rrrxx

With these two changes, recipes/rules are more space efficient, allowing
more to be created in total.

Co-developed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_protocol_type.h |  22 +-
 drivers/net/ethernet/intel/ice/ice_switch.c        | 525 ++++++++-------------
 drivers/net/ethernet/intel/ice/ice_switch.h        |   2 +
 3 files changed, 212 insertions(+), 337 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_protocol_type.h b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
index c396dabacef4..47e405bf1382 100644
--- a/drivers/net/ethernet/intel/ice/ice_protocol_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
@@ -7,18 +7,24 @@
 
 /* Each recipe can match up to 5 different fields. Fields to match can be meta-
  * data, values extracted from packet headers, or results from other recipes.
- * One of the 5 fields is reserved for matching the switch ID. So, up to 4
- * recipes can provide intermediate results to another one through chaining,
- * e.g. recipes 0, 1, 2, and 3 can provide intermediate results to recipe 4.
+ * Therefore, up to 5 recipes can provide intermediate results to another one
+ * through chaining, e.g. recipes 0, 1, 2, 3 and 4 can provide intermediate
+ * results to recipe 5. Note that one of the fields in one of the recipes must
+ * always be reserved for matching the switch ID.
  */
-#define ICE_NUM_WORDS_RECIPE 4
+#define ICE_NUM_WORDS_RECIPE 5
 
-/* Max recipes that can be chained */
+/* Max recipes that can be chained, not including the last one, which combines
+ * intermediate results.
+ */
 #define ICE_MAX_CHAIN_RECIPE 5
 
-/* 1 word reserved for switch ID from allowed 5 words.
- * So a recipe can have max 4 words. And you can chain 5 such recipes
- * together. So maximum words that can be programmed for look up is 5 * 4.
+/* Total max recipes in chain recipe (including intermediate results) */
+#define ICE_MAX_CHAIN_RECIPE_RES (ICE_MAX_CHAIN_RECIPE + 1)
+
+/* A recipe can have max 5 words, and 5 recipes can be chained together (using
+ * the 6th one, which would contain only result indexes). So maximum words that
+ * can be programmed for lookup is 5 * 5 (not including intermediate results).
  */
 #define ICE_MAX_CHAIN_WORDS (ICE_NUM_WORDS_RECIPE * ICE_MAX_CHAIN_RECIPE)
 
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index 81e5448da1aa..5a092dcea172 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -2281,20 +2281,6 @@ static void ice_get_recp_to_prof_map(struct ice_hw *hw)
 	}
 }
 
-/**
- * ice_collect_result_idx - copy result index values
- * @buf: buffer that contains the result index
- * @recp: the recipe struct to copy data into
- */
-static void
-ice_collect_result_idx(struct ice_aqc_recipe_data_elem *buf,
-		       struct ice_sw_recipe *recp)
-{
-	if (buf->content.result_indx & ICE_AQ_RECIPE_RESULT_EN)
-		set_bit(buf->content.result_indx & ~ICE_AQ_RECIPE_RESULT_EN,
-			recp->res_idxs);
-}
-
 /**
  * ice_get_recp_frm_fw - update SW bookkeeping from FW recipe entries
  * @hw: pointer to hardware structure
@@ -2377,11 +2363,11 @@ ice_get_recp_frm_fw(struct ice_hw *hw, struct ice_sw_recipe *recps, u8 rid,
 		prof = find_first_bit(recipe_to_profile[idx],
 				      ICE_MAX_NUM_PROFILES);
 		for (i = 0; i < ICE_NUM_WORDS_RECIPE; i++) {
-			u8 lkup_indx = root_bufs.content.lkup_indx[i + 1];
+			u8 lkup_indx = root_bufs.content.lkup_indx[i];
 
 			rg_entry->fv_idx[i] = lkup_indx;
 			rg_entry->fv_mask[i] =
-				le16_to_cpu(root_bufs.content.mask[i + 1]);
+				le16_to_cpu(root_bufs.content.mask[i]);
 
 			/* If the recipe is a chained recipe then all its
 			 * child recipe's result will have a result index.
@@ -4884,105 +4870,56 @@ ice_fill_valid_words(struct ice_adv_lkup_elem *rule,
 	return ret_val;
 }
 
-/**
- * ice_create_first_fit_recp_def - Create a recipe grouping
- * @hw: pointer to the hardware structure
- * @lkup_exts: an array of protocol header extractions
- * @rg_list: pointer to a list that stores new recipe groups
- * @recp_cnt: pointer to a variable that stores returned number of recipe groups
- *
- * Using first fit algorithm, take all the words that are still not done
- * and start grouping them in 4-word groups. Each group makes up one
- * recipe.
- */
-static int
-ice_create_first_fit_recp_def(struct ice_hw *hw,
-			      struct ice_prot_lkup_ext *lkup_exts,
-			      struct list_head *rg_list,
-			      u8 *recp_cnt)
-{
-	struct ice_pref_recipe_group *grp = NULL;
-	u8 j;
-
-	*recp_cnt = 0;
-
-	for (j = 0; j < lkup_exts->n_val_words; j++) {
-		if (!grp || grp->n_val_pairs == ICE_NUM_WORDS_RECIPE) {
-			struct ice_recp_grp_entry *entry;
-
-			entry = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*entry),
-					     GFP_KERNEL);
-			if (!entry)
-				return -ENOMEM;
-
-			list_add(&entry->l_entry, rg_list);
-			grp = &entry->r_group;
-			(*recp_cnt)++;
-		}
-
-		grp->pairs[grp->n_val_pairs].prot_id =
-			lkup_exts->fv_words[j].prot_id;
-		grp->pairs[grp->n_val_pairs].off = lkup_exts->fv_words[j].off;
-		grp->mask[grp->n_val_pairs] = lkup_exts->field_mask[j];
-		grp->n_val_pairs++;
-	}
-
-	return 0;
-}
-
 /**
  * ice_fill_fv_word_index - fill in the field vector indices for a recipe group
  * @hw: pointer to the hardware structure
- * @fv_list: field vector with the extraction sequence information
- * @rg_list: recipe groupings with protocol-offset pairs
+ * @rm: recipe management list entry
  *
  * Helper function to fill in the field vector indices for protocol-offset
  * pairs. These indexes are then ultimately programmed into a recipe.
  */
 static int
-ice_fill_fv_word_index(struct ice_hw *hw, struct list_head *fv_list,
-		       struct list_head *rg_list)
+ice_fill_fv_word_index(struct ice_hw *hw, struct ice_sw_recipe *rm)
 {
 	struct ice_sw_fv_list_entry *fv;
-	struct ice_recp_grp_entry *rg;
 	struct ice_fv_word *fv_ext;
+	u8 i;
 
-	if (list_empty(fv_list))
-		return 0;
+	if (list_empty(&rm->fv_list))
+		return -EINVAL;
 
-	fv = list_first_entry(fv_list, struct ice_sw_fv_list_entry,
+	fv = list_first_entry(&rm->fv_list, struct ice_sw_fv_list_entry,
 			      list_entry);
 	fv_ext = fv->fv_ptr->ew;
 
-	list_for_each_entry(rg, rg_list, l_entry) {
-		u8 i;
+	/* Add switch id as the first word. */
+	rm->fv_idx[0] = ICE_AQ_SW_ID_LKUP_IDX;
+	rm->fv_mask[0] = ICE_AQ_SW_ID_LKUP_MASK;
+	rm->n_ext_words++;
 
-		for (i = 0; i < rg->r_group.n_val_pairs; i++) {
-			struct ice_fv_word *pr;
-			bool found = false;
-			u16 mask;
-			u8 j;
-
-			pr = &rg->r_group.pairs[i];
-			mask = rg->r_group.mask[i];
-
-			for (j = 0; j < hw->blk[ICE_BLK_SW].es.fvw; j++)
-				if (fv_ext[j].prot_id == pr->prot_id &&
-				    fv_ext[j].off == pr->off) {
-					found = true;
+	for (i = 1; i < rm->n_ext_words; i++) {
+		struct ice_fv_word *fv_word = &rm->ext_words[i - 1];
+		u16 fv_mask = rm->word_masks[i - 1];
+		bool found = false;
+		u8 j;
 
-					/* Store index of field vector */
-					rg->fv_idx[i] = j;
-					rg->fv_mask[i] = mask;
-					break;
-				}
+		for (j = 0; j < hw->blk[ICE_BLK_SW].es.fvw; j++) {
+			if (fv_ext[j].prot_id == fv_word->prot_id &&
+			    fv_ext[j].off == fv_word->off) {
+				found = true;
 
-			/* Protocol/offset could not be found, caller gave an
-			 * invalid pair
-			 */
-			if (!found)
-				return -EINVAL;
+				/* Store index of field vector */
+				rm->fv_idx[i] = j;
+				rm->fv_mask[i] = fv_mask;
+				break;
+			}
 		}
+
+		/* Protocol/offset could not be found, caller gave an invalid
+		 * pair.
+		 */
+		if (!found)
+			return -EINVAL;
 	}
 
 	return 0;
@@ -5055,6 +4992,69 @@ ice_find_free_recp_res_idx(struct ice_hw *hw, const unsigned long *profiles,
 	return (u16)bitmap_weight(free_idx, ICE_MAX_FV_WORDS);
 }
 
+/**
+ * ice_calc_recp_cnt - calculate number of recipes based on word count
+ * @word_cnt: number of lookup words
+ *
+ * Word count should include switch ID word and regular lookup words.
+ * Returns: number of recipes required to fit @word_cnt, including extra recipes
+ * needed for recipe chaining (if needed).
+ */
+static int ice_calc_recp_cnt(u8 word_cnt)
+{
+	/* All words fit in a single recipe, no need for chaining. */
+	if (word_cnt <= ICE_NUM_WORDS_RECIPE)
+		return 1;
+
+	/* Recipe chaining required. Result indexes are fitted right after
+	 * regular lookup words. In some cases a new recipe must be added in
+	 * order to fit result indexes.
+	 *
+	 * While the word count increases, every 5 words an extra recipe needs
+	 * to be added. However, by adding a recipe, one word for its result
+	 * index must also be added, therefore every 4 words recipe count
+	 * increases by 1. This calculation does not apply to word count == 1,
+	 * which is handled above.
+	 */
+	return (word_cnt + 2) / (ICE_NUM_WORDS_RECIPE - 1);
+}
+
+static void fill_recipe_template(struct ice_aqc_recipe_data_elem *recp, u16 rid,
+				 const struct ice_sw_recipe *rm)
+{
+	int i;
+
+	recp->recipe_indx = rid;
+	recp->content.act_ctrl |= ICE_AQ_RECIPE_ACT_PRUNE_INDX_M;
+
+	for (i = 0; i < ICE_NUM_WORDS_RECIPE; i++) {
+		recp->content.lkup_indx[i] = ICE_AQ_RECIPE_LKUP_IGNORE;
+		recp->content.mask[i] = cpu_to_le16(0);
+	}
+
+	set_bit(rid, (unsigned long *)recp->recipe_bitmap);
+	recp->content.act_ctrl_fwd_priority = rm->priority;
+
+	if (rm->need_pass_l2)
+		recp->content.act_ctrl |= ICE_AQ_RECIPE_ACT_NEED_PASS_L2;
+
+	if (rm->allow_pass_l2)
+		recp->content.act_ctrl |= ICE_AQ_RECIPE_ACT_ALLOW_PASS_L2;
+}
+
+static void bookkeep_recipe(struct ice_sw_recipe *recipe,
+			    struct ice_aqc_recipe_data_elem *r,
+			    const struct ice_sw_recipe *rm)
+{
+	memcpy(recipe->r_bitmap, r->recipe_bitmap, sizeof(recipe->r_bitmap));
+
+	recipe->priority = r->content.act_ctrl_fwd_priority;
+	recipe->tun_type = rm->tun_type;
+	recipe->need_pass_l2 = rm->need_pass_l2;
+	recipe->allow_pass_l2 = rm->allow_pass_l2;
+	recipe->recp_created = true;
+}
+
 /* For memcpy in ice_add_sw_recipe. */
 static_assert(sizeof_field(struct ice_aqc_recipe_data_elem, recipe_bitmap) ==
 	      sizeof_field(struct ice_sw_recipe, r_bitmap));
@@ -5071,279 +5071,148 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 {
 	struct ice_aqc_recipe_data_elem *buf __free(kfree) = NULL;
 	DECLARE_BITMAP(result_idx_bm, ICE_MAX_FV_WORDS);
-	struct ice_aqc_recipe_content *content;
-	struct ice_recp_grp_entry *entry;
-	u16 free_res_idx;
-	u8 chain_idx;
-	u8 recps = 0;
+	struct ice_aqc_recipe_data_elem *root;
+	struct ice_sw_recipe *recipe;
+	u16 free_res_idx, rid;
+	int lookup = 0;
+	int recp_cnt;
 	int status;
+	int word;
+	int i;
+
+	recp_cnt = ice_calc_recp_cnt(rm->n_ext_words);
 
-	/* When more than one recipe are required, another recipe is needed to
-	 * chain them together. Matching a tunnel metadata ID takes up one of
-	 * the match fields in the chaining recipe reducing the number of
-	 * chained recipes by one.
-	 */
-	 /* check number of free result indices */
 	bitmap_zero(result_idx_bm, ICE_MAX_FV_WORDS);
+	bitmap_zero(rm->r_bitmap, ICE_MAX_NUM_RECIPES);
+
+	/* Check number of free result indices */
 	free_res_idx = ice_find_free_recp_res_idx(hw, profiles, result_idx_bm);
 
 	ice_debug(hw, ICE_DBG_SW, "Result idx slots: %d, need %d\n",
-		  free_res_idx, rm->n_grp_count);
-
-	if (rm->n_grp_count > 1) {
-		if (rm->n_grp_count > free_res_idx)
-			return -ENOSPC;
+		  free_res_idx, recp_cnt);
 
-		rm->n_grp_count++;
-	}
-
-	if (rm->n_grp_count > ICE_MAX_CHAIN_RECIPE)
+	/* Last recipe doesn't need result index */
+	if (recp_cnt - 1 > free_res_idx)
 		return -ENOSPC;
 
-	buf = kcalloc(rm->n_grp_count, sizeof(*buf), GFP_KERNEL);
+	if (recp_cnt > ICE_MAX_CHAIN_RECIPE_RES)
+		return -E2BIG;
+
+	buf = kcalloc(recp_cnt, sizeof(*buf), GFP_KERNEL);
 	if (!buf)
 		return -ENOMEM;
 
-	bitmap_zero(rm->r_bitmap, ICE_MAX_NUM_RECIPES);
-
-	/* Allocate the recipe resources, and configure them according to the
-	 * match fields from protocol headers and extracted field vectors.
+	/* Setup the non-root subrecipes. These do not contain lookups for other
+	 * subrecipes results. Set associated recipe only to own recipe index.
+	 * Each non-root subrecipe needs a free result index from FV.
+	 *
+	 * Note: only done if there is more than one recipe.
 	 */
-	chain_idx = find_first_bit(result_idx_bm, ICE_MAX_FV_WORDS);
-	list_for_each_entry(entry, &rm->rg_list, l_entry) {
-		u8 i;
+	for (i = 0; i < recp_cnt - 1; i++) {
+		struct ice_aqc_recipe_content *content;
+		u8 result_idx;
 
-		status = ice_alloc_recipe(hw, &entry->rid);
+		status = ice_alloc_recipe(hw, &rid);
 		if (status)
 			return status;
 
-		content = &buf[recps].content;
-
-		buf[recps].recipe_indx = (u8)entry->rid;
-
-		buf[recps].content.act_ctrl |= ICE_AQ_RECIPE_ACT_PRUNE_INDX_M;
+		fill_recipe_template(&buf[i], rid, rm);
 
-		/* All recipes use look-up index 0 to match switch ID. */
-		content->lkup_indx[0] = ICE_AQ_SW_ID_LKUP_IDX;
-		content->mask[0] = cpu_to_le16(ICE_AQ_SW_ID_LKUP_MASK);
-		/* Setup lkup_indx 1..4 to INVALID/ignore and set the mask
-		 * to be 0
+		result_idx = find_first_bit(result_idx_bm, ICE_MAX_FV_WORDS);
+		/* Check if there really is a valid result index that can be
+		 * used.
 		 */
-		for (i = 1; i <= ICE_NUM_WORDS_RECIPE; i++) {
-			content->lkup_indx[i] = 0x80;
-			content->mask[i] = 0;
-		}
-
-		for (i = 0; i < entry->r_group.n_val_pairs; i++) {
-			content->lkup_indx[i + 1] = entry->fv_idx[i];
-			content->mask[i + 1] = cpu_to_le16(entry->fv_mask[i]);
+		if (result_idx >= ICE_MAX_FV_WORDS) {
+			ice_debug(hw, ICE_DBG_SW, "No chain index available\n");
+			return -ENOSPC;
 		}
+		clear_bit(result_idx, result_idx_bm);
 
-		if (rm->n_grp_count > 1) {
-			/* Checks to see if there really is a valid result index
-			 * that can be used.
-			 */
-			if (chain_idx >= ICE_MAX_FV_WORDS) {
-				ice_debug(hw, ICE_DBG_SW, "No chain index available\n");
-				return -ENOSPC;
-			}
+		content = &buf[i].content;
+		content->result_indx = ICE_AQ_RECIPE_RESULT_EN |
+				       FIELD_PREP(ICE_AQ_RECIPE_RESULT_DATA_M,
+						  result_idx);
 
-			entry->chain_idx = chain_idx;
-			content->result_indx =
-				ICE_AQ_RECIPE_RESULT_EN |
-				FIELD_PREP(ICE_AQ_RECIPE_RESULT_DATA_M,
-					   chain_idx);
-			clear_bit(chain_idx, result_idx_bm);
-			chain_idx = find_first_bit(result_idx_bm,
-						   ICE_MAX_FV_WORDS);
-		}
+		/* Set recipe association to be used for root recipe */
+		set_bit(rid, rm->r_bitmap);
 
-		/* fill recipe dependencies */
-		set_bit(buf[recps].recipe_indx,
-			(unsigned long *)buf[recps].recipe_bitmap);
-		content->act_ctrl_fwd_priority = rm->priority;
+		word = 0;
+		while (lookup < rm->n_ext_words &&
+		       word < ICE_NUM_WORDS_RECIPE) {
+			content->lkup_indx[word] = rm->fv_idx[lookup];
+			content->mask[word] = cpu_to_le16(rm->fv_mask[lookup]);
 
-		if (rm->need_pass_l2)
-			content->act_ctrl |= ICE_AQ_RECIPE_ACT_NEED_PASS_L2;
+			lookup++;
+			word++;
+		}
 
-		if (rm->allow_pass_l2)
-			content->act_ctrl |= ICE_AQ_RECIPE_ACT_ALLOW_PASS_L2;
-		recps++;
+		recipe = &hw->switch_info->recp_list[rid];
+		set_bit(result_idx, recipe->res_idxs);
+		bookkeep_recipe(recipe, &buf[i], rm);
 	}
 
-	if (rm->n_grp_count == 1) {
-		rm->root_rid = buf[0].recipe_indx;
-		set_bit(buf[0].recipe_indx, rm->r_bitmap);
-		buf[0].content.rid = rm->root_rid | ICE_AQ_RECIPE_ID_IS_ROOT;
-		memcpy(buf[0].recipe_bitmap, rm->r_bitmap,
-		       sizeof(buf[0].recipe_bitmap));
-
-		/* Applicable only for ROOT_RECIPE, set the fwd_priority for
-		 * the recipe which is getting created if specified
-		 * by user. Usually any advanced switch filter, which results
-		 * into new extraction sequence, ended up creating a new recipe
-		 * of type ROOT and usually recipes are associated with profiles
-		 * Switch rule referreing newly created recipe, needs to have
-		 * either/or 'fwd' or 'join' priority, otherwise switch rule
-		 * evaluation will not happen correctly. In other words, if
-		 * switch rule to be evaluated on priority basis, then recipe
-		 * needs to have priority, otherwise it will be evaluated last.
-		 */
-		buf[0].content.act_ctrl_fwd_priority = rm->priority;
-	} else {
-		struct ice_recp_grp_entry *last_chain_entry;
-		u16 rid, i;
+	/* Setup the root recipe */
+	status = ice_alloc_recipe(hw, &rid);
+	if (status)
+		return status;
 
-		/* Allocate the last recipe that will chain the outcomes of the
-		 * other recipes together
-		 */
-		status = ice_alloc_recipe(hw, &rid);
-		if (status)
-			return status;
+	recipe = &hw->switch_info->recp_list[rid];
+	recipe->is_root = true;
+	root = &buf[recp_cnt - 1];
+	fill_recipe_template(root, rid, rm);
 
-		content = &buf[recps].content;
+	/* Set recipe association, use previously set bitmap and own rid */
+	set_bit(rid, rm->r_bitmap);
+	memcpy(root->recipe_bitmap, rm->r_bitmap, sizeof(root->recipe_bitmap));
 
-		buf[recps].recipe_indx = (u8)rid;
-		content->rid = (u8)rid;
-		content->rid |= ICE_AQ_RECIPE_ID_IS_ROOT;
-		/* the new entry created should also be part of rg_list to
-		 * make sure we have complete recipe
-		 */
-		last_chain_entry = devm_kzalloc(ice_hw_to_dev(hw),
-						sizeof(*last_chain_entry),
-						GFP_KERNEL);
-		if (!last_chain_entry)
-			return -ENOMEM;
+	/* For non-root recipes rid should be 0, for root it should be correct
+	 * rid value ored with 0x80 (is root bit).
+	 */
+	root->content.rid = rid | ICE_AQ_RECIPE_ID_IS_ROOT;
 
-		last_chain_entry->rid = rid;
-		memset(&content->lkup_indx, 0, sizeof(content->lkup_indx));
-		/* All recipes use look-up index 0 to match switch ID. */
-		content->lkup_indx[0] = ICE_AQ_SW_ID_LKUP_IDX;
-		content->mask[0] = cpu_to_le16(ICE_AQ_SW_ID_LKUP_MASK);
-		for (i = 1; i <= ICE_NUM_WORDS_RECIPE; i++) {
-			content->lkup_indx[i] = ICE_AQ_RECIPE_LKUP_IGNORE;
-			content->mask[i] = 0;
-		}
+	/* Fill remaining lookups in root recipe */
+	word = 0;
+	while (lookup < rm->n_ext_words &&
+	       word < ICE_NUM_WORDS_RECIPE /* should always be true */) {
+		root->content.lkup_indx[word] = rm->fv_idx[lookup];
+		root->content.mask[word] = cpu_to_le16(rm->fv_mask[lookup]);
 
-		i = 1;
-		/* update r_bitmap with the recp that is used for chaining */
-		set_bit(rid, rm->r_bitmap);
-		/* this is the recipe that chains all the other recipes so it
-		 * should not have a chaining ID to indicate the same
+		lookup++;
+		word++;
+	}
+
+	/* Fill result indexes as lookups */
+	i = 0;
+	while (i < recp_cnt - 1 &&
+	       word < ICE_NUM_WORDS_RECIPE /* should always be true */) {
+		root->content.lkup_indx[word] = buf[i].content.result_indx &
+						~ICE_AQ_RECIPE_RESULT_EN;
+		root->content.mask[word] = cpu_to_le16(0xffff);
+		/* For bookkeeping, it is needed to mark FV index as used for
+		 * intermediate result.
 		 */
-		last_chain_entry->chain_idx = ICE_INVAL_CHAIN_IND;
-		list_for_each_entry(entry, &rm->rg_list, l_entry) {
-			last_chain_entry->fv_idx[i] = entry->chain_idx;
-			content->lkup_indx[i] = entry->chain_idx;
-			content->mask[i++] = cpu_to_le16(0xFFFF);
-			set_bit(entry->rid, rm->r_bitmap);
-		}
-		list_add(&last_chain_entry->l_entry, &rm->rg_list);
-		memcpy(buf[recps].recipe_bitmap, rm->r_bitmap,
-		       sizeof(buf[recps].recipe_bitmap));
-		content->act_ctrl_fwd_priority = rm->priority;
+		set_bit(root->content.lkup_indx[word], recipe->res_idxs);
 
-		recps++;
-		rm->root_rid = (u8)rid;
+		i++;
+		word++;
 	}
+
+	rm->root_rid = rid;
+	bookkeep_recipe(&hw->switch_info->recp_list[rid], root, rm);
+
+	/* Program the recipe */
 	status = ice_acquire_change_lock(hw, ICE_RES_WRITE);
 	if (status)
 		return status;
 
-	status = ice_aq_add_recipe(hw, buf, rm->n_grp_count, NULL);
+	status = ice_aq_add_recipe(hw, buf, recp_cnt, NULL);
 	ice_release_change_lock(hw);
 	if (status)
 		return status;
 
-	/* Every recipe that just got created add it to the recipe
-	 * book keeping list
-	 */
-	list_for_each_entry(entry, &rm->rg_list, l_entry) {
-		struct ice_switch_info *sw = hw->switch_info;
-		bool is_root, idx_found = false;
-		struct ice_sw_recipe *recp;
-		u16 idx, buf_idx = 0;
-
-		/* find buffer index for copying some data */
-		for (idx = 0; idx < rm->n_grp_count; idx++)
-			if (buf[idx].recipe_indx == entry->rid) {
-				buf_idx = idx;
-				idx_found = true;
-			}
-
-		if (!idx_found)
-			return -EIO;
-
-		recp = &sw->recp_list[entry->rid];
-		is_root = (rm->root_rid == entry->rid);
-		recp->is_root = is_root;
-
-		recp->root_rid = entry->rid;
-		recp->big_recp = (is_root && rm->n_grp_count > 1);
-
-		memcpy(&recp->ext_words, entry->r_group.pairs,
-		       entry->r_group.n_val_pairs * sizeof(struct ice_fv_word));
-
-		memcpy(recp->r_bitmap, buf[buf_idx].recipe_bitmap,
-		       sizeof(recp->r_bitmap));
-
-		/* Copy non-result fv index values and masks to recipe. This
-		 * call will also update the result recipe bitmask.
-		 */
-		ice_collect_result_idx(&buf[buf_idx], recp);
-
-		/* for non-root recipes, also copy to the root, this allows
-		 * easier matching of a complete chained recipe
-		 */
-		if (!is_root)
-			ice_collect_result_idx(&buf[buf_idx],
-					       &sw->recp_list[rm->root_rid]);
-
-		recp->n_ext_words = entry->r_group.n_val_pairs;
-		recp->chain_idx = entry->chain_idx;
-		recp->priority = buf[buf_idx].content.act_ctrl_fwd_priority;
-		recp->n_grp_count = rm->n_grp_count;
-		recp->tun_type = rm->tun_type;
-		recp->need_pass_l2 = rm->need_pass_l2;
-		recp->allow_pass_l2 = rm->allow_pass_l2;
-		recp->recp_created = true;
-	}
-
 	return 0;
 }
 
-/**
- * ice_create_recipe_group - creates recipe group
- * @hw: pointer to hardware structure
- * @rm: recipe management list entry
- * @lkup_exts: lookup elements
- */
-static int
-ice_create_recipe_group(struct ice_hw *hw, struct ice_sw_recipe *rm,
-			struct ice_prot_lkup_ext *lkup_exts)
-{
-	u8 recp_count = 0;
-	int status;
-
-	rm->n_grp_count = 0;
-
-	/* Create recipes for words that are marked not done by packing them
-	 * as best fit.
-	 */
-	status = ice_create_first_fit_recp_def(hw, lkup_exts,
-					       &rm->rg_list, &recp_count);
-	if (!status) {
-		rm->n_grp_count += recp_count;
-		rm->n_ext_words = lkup_exts->n_val_words;
-		memcpy(&rm->ext_words, lkup_exts->fv_words,
-		       sizeof(rm->ext_words));
-		memcpy(rm->word_masks, lkup_exts->field_mask,
-		       sizeof(rm->word_masks));
-	}
-
-	return status;
-}
-
 /* ice_get_compat_fv_bitmap - Get compatible field vector bitmap for rule
  * @hw: pointer to hardware structure
  * @rinfo: other information regarding the rule e.g. priority and action info
@@ -5504,12 +5373,10 @@ ice_add_adv_recipe(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
 	if (status)
 		goto err_unroll;
 
-	/* Group match words into recipes using preferred recipe grouping
-	 * criteria.
-	 */
-	status = ice_create_recipe_group(hw, rm, lkup_exts);
-	if (status)
-		goto err_unroll;
+	/* Copy FV words and masks from lkup_exts to recipe struct. */
+	rm->n_ext_words = lkup_exts->n_val_words;
+	memcpy(rm->ext_words, lkup_exts->fv_words, sizeof(rm->ext_words));
+	memcpy(rm->word_masks, lkup_exts->field_mask, sizeof(rm->word_masks));
 
 	/* set the recipe priority if specified */
 	rm->priority = (u8)rinfo->priority;
@@ -5520,7 +5387,7 @@ ice_add_adv_recipe(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
 	/* Find offsets from the field vector. Pick the first one for all the
 	 * recipes.
 	 */
-	status = ice_fill_fv_word_index(hw, &rm->fv_list, &rm->rg_list);
+	status = ice_fill_fv_word_index(hw, rm);
 	if (status)
 		goto err_unroll;
 
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.h b/drivers/net/ethernet/intel/ice/ice_switch.h
index 0c410ce29700..4bf170b505ae 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.h
+++ b/drivers/net/ethernet/intel/ice/ice_switch.h
@@ -227,6 +227,8 @@ struct ice_sw_recipe {
 	 */
 	struct ice_fv_word ext_words[ICE_MAX_CHAIN_WORDS];
 	u16 word_masks[ICE_MAX_CHAIN_WORDS];
+	u8 fv_idx[ICE_MAX_CHAIN_WORDS];
+	u16 fv_mask[ICE_MAX_CHAIN_WORDS];
 
 	/* if this recipe is a collection of other recipe */
 	u8 big_recp;
-- 
cgit v1.2.3-58-ga151


From 1d2ac128531e381dab5bb7fd21f95cf2d075bff5 Mon Sep 17 00:00:00 2001
From: Marcin Szycik <marcin.szycik@linux.intel.com>
Date: Thu, 27 Jun 2024 16:55:46 +0200
Subject: ice: Remove unused members from switch API

Remove several members of struct ice_sw_recipe and struct
ice_prot_lkup_ext. Remove struct ice_recp_grp_entry and struct
ice_pref_recipe_group, since they are now unused as well.

All of the deleted members were only written to and never read, so it's
pointless to keep them.

Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_common.c        |  7 ---
 drivers/net/ethernet/intel/ice/ice_protocol_type.h | 17 --------
 drivers/net/ethernet/intel/ice/ice_switch.c        | 51 +++++-----------------
 drivers/net/ethernet/intel/ice/ice_switch.h        | 16 -------
 4 files changed, 10 insertions(+), 81 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 17f6626fec7e..e77261704ee4 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -961,14 +961,7 @@ static void ice_cleanup_fltr_mgmt_struct(struct ice_hw *hw)
 	}
 	recps = sw->recp_list;
 	for (i = 0; i < ICE_MAX_NUM_RECIPES; i++) {
-		struct ice_recp_grp_entry *rg_entry, *tmprg_entry;
-
 		recps[i].root_rid = i;
-		list_for_each_entry_safe(rg_entry, tmprg_entry,
-					 &recps[i].rg_list, l_entry) {
-			list_del(&rg_entry->l_entry);
-			devm_kfree(ice_hw_to_dev(hw), rg_entry);
-		}
 
 		if (recps[i].adv_rule) {
 			struct ice_adv_fltr_mgmt_list_entry *tmp_entry;
diff --git a/drivers/net/ethernet/intel/ice/ice_protocol_type.h b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
index 47e405bf1382..7c09ea0f03ba 100644
--- a/drivers/net/ethernet/intel/ice/ice_protocol_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_protocol_type.h
@@ -455,7 +455,6 @@ struct ice_prot_ext_tbl_entry {
 
 /* Extractions to be looked up for a given recipe */
 struct ice_prot_lkup_ext {
-	u16 prot_type;
 	u8 n_val_words;
 	/* create a buffer to hold max words per recipe */
 	u16 field_mask[ICE_MAX_CHAIN_WORDS];
@@ -463,20 +462,4 @@ struct ice_prot_lkup_ext {
 	struct ice_fv_word fv_words[ICE_MAX_CHAIN_WORDS];
 };
 
-struct ice_pref_recipe_group {
-	u8 n_val_pairs;		/* Number of valid pairs */
-	struct ice_fv_word pairs[ICE_NUM_WORDS_RECIPE];
-	u16 mask[ICE_NUM_WORDS_RECIPE];
-};
-
-struct ice_recp_grp_entry {
-	struct list_head l_entry;
-
-#define ICE_INVAL_CHAIN_IND 0xFF
-	u16 rid;
-	u8 chain_idx;
-	u16 fv_idx[ICE_NUM_WORDS_RECIPE];
-	u16 fv_mask[ICE_NUM_WORDS_RECIPE];
-	struct ice_pref_recipe_group r_group;
-};
 #endif /* _ICE_PROTOCOL_TYPE_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index 5a092dcea172..27828cdfe085 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -1471,7 +1471,6 @@ int ice_init_def_sw_recp(struct ice_hw *hw)
 		recps[i].root_rid = i;
 		INIT_LIST_HEAD(&recps[i].filt_rules);
 		INIT_LIST_HEAD(&recps[i].filt_replay_rules);
-		INIT_LIST_HEAD(&recps[i].rg_list);
 		mutex_init(&recps[i].filt_rule_lock);
 	}
 
@@ -2339,18 +2338,10 @@ ice_get_recp_frm_fw(struct ice_hw *hw, struct ice_sw_recipe *recps, u8 rid,
 
 	for (sub_recps = 0; sub_recps < num_recps; sub_recps++) {
 		struct ice_aqc_recipe_data_elem root_bufs = tmp[sub_recps];
-		struct ice_recp_grp_entry *rg_entry;
 		u8 i, prof, idx, prot = 0;
 		bool is_root;
 		u16 off = 0;
 
-		rg_entry = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*rg_entry),
-					GFP_KERNEL);
-		if (!rg_entry) {
-			status = -ENOMEM;
-			goto err_unroll;
-		}
-
 		idx = root_bufs.recipe_indx;
 		is_root = root_bufs.content.rid & ICE_AQ_RECIPE_ID_IS_ROOT;
 
@@ -2364,10 +2355,7 @@ ice_get_recp_frm_fw(struct ice_hw *hw, struct ice_sw_recipe *recps, u8 rid,
 				      ICE_MAX_NUM_PROFILES);
 		for (i = 0; i < ICE_NUM_WORDS_RECIPE; i++) {
 			u8 lkup_indx = root_bufs.content.lkup_indx[i];
-
-			rg_entry->fv_idx[i] = lkup_indx;
-			rg_entry->fv_mask[i] =
-				le16_to_cpu(root_bufs.content.mask[i]);
+			u16 lkup_mask = le16_to_cpu(root_bufs.content.mask[i]);
 
 			/* If the recipe is a chained recipe then all its
 			 * child recipe's result will have a result index.
@@ -2378,26 +2366,21 @@ ice_get_recp_frm_fw(struct ice_hw *hw, struct ice_sw_recipe *recps, u8 rid,
 			 * has ICE_AQ_RECIPE_LKUP_IGNORE or 0 since it isn't a
 			 * valid offset value.
 			 */
-			if (test_bit(rg_entry->fv_idx[i], hw->switch_info->prof_res_bm[prof]) ||
-			    rg_entry->fv_idx[i] & ICE_AQ_RECIPE_LKUP_IGNORE ||
-			    rg_entry->fv_idx[i] == 0)
+			if (!lkup_indx ||
+			    (lkup_indx & ICE_AQ_RECIPE_LKUP_IGNORE) ||
+			    test_bit(lkup_indx,
+				     hw->switch_info->prof_res_bm[prof]))
 				continue;
 
-			ice_find_prot_off(hw, ICE_BLK_SW, prof,
-					  rg_entry->fv_idx[i], &prot, &off);
+			ice_find_prot_off(hw, ICE_BLK_SW, prof, lkup_indx,
+					  &prot, &off);
 			lkup_exts->fv_words[fv_word_idx].prot_id = prot;
 			lkup_exts->fv_words[fv_word_idx].off = off;
-			lkup_exts->field_mask[fv_word_idx] =
-				rg_entry->fv_mask[i];
+			lkup_exts->field_mask[fv_word_idx] = lkup_mask;
 			fv_word_idx++;
 		}
-		/* populate rg_list with the data from the child entry of this
-		 * recipe
-		 */
-		list_add(&rg_entry->l_entry, &recps[rid].rg_list);
 
 		/* Propagate some data to the recipe database */
-		recps[idx].is_root = !!is_root;
 		recps[idx].priority = root_bufs.content.act_ctrl_fwd_priority;
 		recps[idx].need_pass_l2 = root_bufs.content.act_ctrl &
 					  ICE_AQ_RECIPE_ACT_NEED_PASS_L2;
@@ -2405,11 +2388,8 @@ ice_get_recp_frm_fw(struct ice_hw *hw, struct ice_sw_recipe *recps, u8 rid,
 					   ICE_AQ_RECIPE_ACT_ALLOW_PASS_L2;
 		bitmap_zero(recps[idx].res_idxs, ICE_MAX_FV_WORDS);
 		if (root_bufs.content.result_indx & ICE_AQ_RECIPE_RESULT_EN) {
-			recps[idx].chain_idx = root_bufs.content.result_indx &
-				~ICE_AQ_RECIPE_RESULT_EN;
-			set_bit(recps[idx].chain_idx, recps[idx].res_idxs);
-		} else {
-			recps[idx].chain_idx = ICE_INVAL_CHAIN_IND;
+			set_bit(root_bufs.content.result_indx &
+				~ICE_AQ_RECIPE_RESULT_EN, recps[idx].res_idxs);
 		}
 
 		if (!is_root) {
@@ -2429,8 +2409,6 @@ ice_get_recp_frm_fw(struct ice_hw *hw, struct ice_sw_recipe *recps, u8 rid,
 
 	/* Complete initialization of the root recipe entry */
 	lkup_exts->n_val_words = fv_word_idx;
-	recps[rid].big_recp = (num_recps > 1);
-	recps[rid].n_grp_count = (u8)num_recps;
 
 	/* Copy result indexes */
 	bitmap_copy(recps[rid].res_idxs, result_bm, ICE_MAX_FV_WORDS);
@@ -5157,7 +5135,6 @@ ice_add_sw_recipe(struct ice_hw *hw, struct ice_sw_recipe *rm,
 		return status;
 
 	recipe = &hw->switch_info->recp_list[rid];
-	recipe->is_root = true;
 	root = &buf[recp_cnt - 1];
 	fill_recipe_template(root, rid, rm);
 
@@ -5317,9 +5294,7 @@ ice_add_adv_recipe(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
 	DECLARE_BITMAP(fv_bitmap, ICE_MAX_NUM_PROFILES);
 	DECLARE_BITMAP(profiles, ICE_MAX_NUM_PROFILES);
 	struct ice_prot_lkup_ext *lkup_exts;
-	struct ice_recp_grp_entry *r_entry;
 	struct ice_sw_fv_list_entry *fvit;
-	struct ice_recp_grp_entry *r_tmp;
 	struct ice_sw_fv_list_entry *tmp;
 	struct ice_sw_recipe *rm;
 	int status = 0;
@@ -5361,7 +5336,6 @@ ice_add_adv_recipe(struct ice_hw *hw, struct ice_adv_lkup_elem *lkups,
 	 * headers being programmed.
 	 */
 	INIT_LIST_HEAD(&rm->fv_list);
-	INIT_LIST_HEAD(&rm->rg_list);
 
 	/* Get bitmap of field vectors (profiles) that are compatible with the
 	 * rule request; only these will be searched in the subsequent call to
@@ -5465,11 +5439,6 @@ err_free_recipe:
 	}
 
 err_unroll:
-	list_for_each_entry_safe(r_entry, r_tmp, &rm->rg_list, l_entry) {
-		list_del(&r_entry->l_entry);
-		devm_kfree(ice_hw_to_dev(hw), r_entry);
-	}
-
 	list_for_each_entry_safe(fvit, tmp, &rm->fv_list, list_entry) {
 		list_del(&fvit->list_entry);
 		devm_kfree(ice_hw_to_dev(hw), fvit);
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.h b/drivers/net/ethernet/intel/ice/ice_switch.h
index 4bf170b505ae..671d7a5f359f 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.h
+++ b/drivers/net/ethernet/intel/ice/ice_switch.h
@@ -216,7 +216,6 @@ struct ice_sw_recipe {
 	/* For a chained recipe the root recipe is what should be used for
 	 * programming rules
 	 */
-	u8 is_root;
 	u8 root_rid;
 	u8 recp_created;
 
@@ -230,19 +229,6 @@ struct ice_sw_recipe {
 	u8 fv_idx[ICE_MAX_CHAIN_WORDS];
 	u16 fv_mask[ICE_MAX_CHAIN_WORDS];
 
-	/* if this recipe is a collection of other recipe */
-	u8 big_recp;
-
-	/* if this recipe is part of another bigger recipe then chain index
-	 * corresponding to this recipe
-	 */
-	u8 chain_idx;
-
-	/* if this recipe is a collection of other recipe then count of other
-	 * recipes and recipe IDs of those recipes
-	 */
-	u8 n_grp_count;
-
 	/* Bit map specifying the IDs associated with this group of recipe */
 	DECLARE_BITMAP(r_bitmap, ICE_MAX_NUM_RECIPES);
 
@@ -274,8 +260,6 @@ struct ice_sw_recipe {
 	u8 need_pass_l2:1;
 	u8 allow_pass_l2:1;
 
-	struct list_head rg_list;
-
 	/* This struct saves the fv_words for a given lookup */
 	struct ice_prot_lkup_ext lkup_exts;
 };
-- 
cgit v1.2.3-58-ga151


From e10989e56f5231934103d5a4c7c8926c783924ca Mon Sep 17 00:00:00 2001
From: Marcin Szycik <marcin.szycik@linux.intel.com>
Date: Thu, 27 Jun 2024 16:55:47 +0200
Subject: ice: Add tracepoint for adding and removing switch rules

Track the number of rules and recipes added to switch. Add a tracepoint to
ice_aq_sw_rules(), which shows both rule and recipe count. This information
can be helpful when designing a set of rules to program to the hardware, as
it shows where the practical limit is. Actual limits are known (64 recipes,
32k rules), but it's hard to translate these values to how many rules the
*user* can actually create, because of extra metadata being implicitly
added, and recipe/rule chaining. Chaining combines several recipes/rules to
create a larger recipe/rule, so one large rule added by the user might
actually consume multiple rules from hardware perspective.

Rule counter is simply incremented/decremented in ice_aq_sw_rules(), since
all rules are added or removed via it.

Counting recipes is harder, as recipes can't be removed (only overwritten).
Recipes added via ice_aq_add_recipe() could end up being unused, when
there is an error in later stages of rule creation. Instead, track the
allocation and freeing of recipes, which should reflect the actual usage of
recipes (if something fails after recipe(s) were created, caller should
free them). Also, a number of recipes are loaded from NVM by default -
initialize the recipe counter with the number of these recipes on switch
initialization.

Example configuration:
  cd /sys/kernel/tracing
  echo function > current_tracer
  echo ice_aq_sw_rules > set_ftrace_filter
  echo ice_aq_sw_rules > set_event
  echo 1 > tracing_on
  cat trace

Example output:
  tc-4097    [069] ...1.   787.595536: ice_aq_sw_rules <-ice_rem_adv_rule
  tc-4097    [069] .....   787.595705: ice_aq_sw_rules: rules=9 recipes=15
  tc-4098    [057] ...1.   787.652033: ice_aq_sw_rules <-ice_add_adv_rule
  tc-4098    [057] .....   787.652201: ice_aq_sw_rules: rules=10 recipes=16

Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_common.c |  3 +++
 drivers/net/ethernet/intel/ice/ice_switch.c | 22 ++++++++++++++++++++--
 drivers/net/ethernet/intel/ice/ice_trace.h  | 18 ++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_type.h   |  2 ++
 4 files changed, 43 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index e77261704ee4..c1a364174a47 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -934,6 +934,9 @@ static int ice_init_fltr_mgmt_struct(struct ice_hw *hw)
 	INIT_LIST_HEAD(&sw->vsi_list_map_head);
 	sw->prof_res_bm_init = 0;
 
+	/* Initialize recipe count with default recipes read from NVM */
+	sw->recp_cnt = ICE_SW_LKUP_LAST;
+
 	status = ice_init_def_sw_recp(hw);
 	if (status) {
 		devm_kfree(ice_hw_to_dev(hw), hw->switch_info);
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index 27828cdfe085..3caafcdc301f 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -3,6 +3,7 @@
 
 #include "ice_lib.h"
 #include "ice_switch.h"
+#include "ice_trace.h"
 
 #define ICE_ETH_DA_OFFSET		0
 #define ICE_ETH_ETHTYPE_OFFSET		12
@@ -1961,6 +1962,15 @@ ice_aq_sw_rules(struct ice_hw *hw, void *rule_list, u16 rule_list_sz,
 	    hw->adminq.sq_last_status == ICE_AQ_RC_ENOENT)
 		status = -ENOENT;
 
+	if (!status) {
+		if (opc == ice_aqc_opc_add_sw_rules)
+			hw->switch_info->rule_cnt += num_rules;
+		else if (opc == ice_aqc_opc_remove_sw_rules)
+			hw->switch_info->rule_cnt -= num_rules;
+	}
+
+	trace_ice_aq_sw_rules(hw->switch_info);
+
 	return status;
 }
 
@@ -2181,8 +2191,10 @@ int ice_alloc_recipe(struct ice_hw *hw, u16 *rid)
 	sw_buf->res_type = cpu_to_le16(res_type);
 	status = ice_aq_alloc_free_res(hw, sw_buf, buf_len,
 				       ice_aqc_opc_alloc_res);
-	if (!status)
+	if (!status) {
 		*rid = le16_to_cpu(sw_buf->elem[0].e.sw_resp);
+		hw->switch_info->recp_cnt++;
+	}
 
 	return status;
 }
@@ -2196,7 +2208,13 @@ int ice_alloc_recipe(struct ice_hw *hw, u16 *rid)
  */
 static int ice_free_recipe_res(struct ice_hw *hw, u16 rid)
 {
-	return ice_free_hw_res(hw, ICE_AQC_RES_TYPE_RECIPE, 1, &rid);
+	int status;
+
+	status = ice_free_hw_res(hw, ICE_AQC_RES_TYPE_RECIPE, 1, &rid);
+	if (!status)
+		hw->switch_info->recp_cnt--;
+
+	return status;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_trace.h b/drivers/net/ethernet/intel/ice/ice_trace.h
index 244cddd2a9ea..07aab6e130cd 100644
--- a/drivers/net/ethernet/intel/ice/ice_trace.h
+++ b/drivers/net/ethernet/intel/ice/ice_trace.h
@@ -330,6 +330,24 @@ DEFINE_EVENT(ice_esw_br_port_template,
 	     TP_ARGS(port)
 );
 
+DECLARE_EVENT_CLASS(ice_switch_stats_template,
+		    TP_PROTO(struct ice_switch_info *sw_info),
+		    TP_ARGS(sw_info),
+		    TP_STRUCT__entry(__field(u16, rule_cnt)
+				     __field(u8, recp_cnt)),
+		    TP_fast_assign(__entry->rule_cnt = sw_info->rule_cnt;
+				   __entry->recp_cnt = sw_info->recp_cnt;),
+		    TP_printk("rules=%u recipes=%u",
+			      __entry->rule_cnt,
+			      __entry->recp_cnt)
+);
+
+DEFINE_EVENT(ice_switch_stats_template,
+	     ice_aq_sw_rules,
+	     TP_PROTO(struct ice_switch_info *sw_info),
+	     TP_ARGS(sw_info)
+);
+
 /* End tracepoints */
 
 #endif /* _ICE_TRACE_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index e26ae79578ba..cf3cddbae23b 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -761,6 +761,8 @@ struct ice_switch_info {
 	struct ice_sw_recipe *recp_list;
 	u16 prof_res_bm_init;
 	u16 max_used_prof_index;
+	u16 rule_cnt;
+	u8 recp_cnt;
 
 	DECLARE_BITMAP(prof_res_bm[ICE_MAX_NUM_PROFILES], ICE_MAX_FV_WORDS);
 };
-- 
cgit v1.2.3-58-ga151


From a6a9fcb10836105e525ccb8bc1a6af4b20a113be Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Tue, 9 Jul 2024 11:52:22 +0100
Subject: net: pse-pd: pd692x0: Fix spelling mistake "availables" ->
 "available"

There is a spelling mistake in a dev_err message. Fix it.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Acked-by: Kory Maincent <Kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240709105222.168306-1-colin.i.king@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/pse-pd/pd692x0.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/pse-pd/pd692x0.c b/drivers/net/pse-pd/pd692x0.c
index 29cc76a66c13..0af7db80b2f8 100644
--- a/drivers/net/pse-pd/pd692x0.c
+++ b/drivers/net/pse-pd/pd692x0.c
@@ -589,7 +589,7 @@ static int pd692x0_pi_set_pw_from_table(struct device *dev,
 
 		if (pw < pw_table->class_pw) {
 			dev_err(dev,
-				"Power limit %dmW not supported. Ranges availables: [%d-%d] or [%d-%d]\n",
+				"Power limit %dmW not supported. Ranges available: [%d-%d] or [%d-%d]\n",
 				pw,
 				(pw_table - 1)->class_pw,
 				(pw_table - 1)->class_pw + (pw_table - 1)->max_added_class_pw,
-- 
cgit v1.2.3-58-ga151


From bf130ed3aaa37bbd9c7f4b305c42e02128187cd4 Mon Sep 17 00:00:00 2001
From: Tony Nguyen <anthony.l.nguyen@intel.com>
Date: Tue, 2 Jul 2024 14:38:46 -0700
Subject: net: intel: Remove MODULE_AUTHORs

We are moving away from the Sourceforge email address. Rather than
removing or updating the email for the affected entries, remove the
MODULE_AUTHOR altogether as its usage is incorrect [1].

Link: https://lore.kernel.org/netdev/20200626115236.7f36d379@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com/ [1]
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Acked-by: Alexander Lobakin <aleksander.lobakin@intel.com> # libeth, libie
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/e100.c                 | 1 -
 drivers/net/ethernet/intel/e1000/e1000_main.c     | 1 -
 drivers/net/ethernet/intel/e1000e/netdev.c        | 1 -
 drivers/net/ethernet/intel/fm10k/fm10k_main.c     | 1 -
 drivers/net/ethernet/intel/i40e/i40e_main.c       | 1 -
 drivers/net/ethernet/intel/iavf/iavf_main.c       | 1 -
 drivers/net/ethernet/intel/ice/ice_main.c         | 1 -
 drivers/net/ethernet/intel/igb/igb_main.c         | 1 -
 drivers/net/ethernet/intel/igbvf/netdev.c         | 1 -
 drivers/net/ethernet/intel/igc/igc_main.c         | 1 -
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c     | 1 -
 drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c | 1 -
 drivers/net/ethernet/intel/libeth/rx.c            | 1 -
 drivers/net/ethernet/intel/libie/rx.c             | 1 -
 14 files changed, 14 deletions(-)

diff --git a/drivers/net/ethernet/intel/e100.c b/drivers/net/ethernet/intel/e100.c
index 9b068d40778d..aa139b67a55b 100644
--- a/drivers/net/ethernet/intel/e100.c
+++ b/drivers/net/ethernet/intel/e100.c
@@ -161,7 +161,6 @@
 #define FIRMWARE_D102E		"e100/d102e_ucode.bin"
 
 MODULE_DESCRIPTION(DRV_DESCRIPTION);
-MODULE_AUTHOR(DRV_COPYRIGHT);
 MODULE_LICENSE("GPL v2");
 MODULE_FIRMWARE(FIRMWARE_D101M);
 MODULE_FIRMWARE(FIRMWARE_D101S);
diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c
index 60fff9a6c53e..ab7ae418d294 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_main.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_main.c
@@ -187,7 +187,6 @@ static struct pci_driver e1000_driver = {
 	.err_handler = &e1000_err_handler
 };
 
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
 MODULE_DESCRIPTION("Intel(R) PRO/1000 Network Driver");
 MODULE_LICENSE("GPL v2");
 
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index 3cd161c6672b..360ee26557f7 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -7969,7 +7969,6 @@ static void __exit e1000_exit_module(void)
 }
 module_exit(e1000_exit_module);
 
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
 MODULE_DESCRIPTION("Intel(R) PRO/1000 Network Driver");
 MODULE_LICENSE("GPL v2");
 
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
index fc373472e4e1..142f07ca8bc0 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
@@ -17,7 +17,6 @@ static const char fm10k_driver_string[] = DRV_SUMMARY;
 static const char fm10k_copyright[] =
 	"Copyright(c) 2013 - 2019 Intel Corporation.";
 
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
 MODULE_DESCRIPTION(DRV_SUMMARY);
 MODULE_LICENSE("GPL v2");
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 284c3fad5a6e..8535fb5c4e46 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -98,7 +98,6 @@ static int debug = -1;
 module_param(debug, uint, 0);
 MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all), Debug mask (0x8XXXXXXX)");
 
-MODULE_AUTHOR("Intel Corporation, <e1000-devel@lists.sourceforge.net>");
 MODULE_DESCRIPTION("Intel(R) Ethernet Connection XL710 Network Driver");
 MODULE_IMPORT_NS(LIBIE);
 MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c
index c6dff0963053..ff11bafb3b4f 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_main.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_main.c
@@ -45,7 +45,6 @@ static const struct pci_device_id iavf_pci_tbl[] = {
 MODULE_DEVICE_TABLE(pci, iavf_pci_tbl);
 
 MODULE_ALIAS("i40evf");
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
 MODULE_DESCRIPTION("Intel(R) Ethernet Adaptive Virtual Function Network Driver");
 MODULE_IMPORT_NS(LIBETH);
 MODULE_IMPORT_NS(LIBIE);
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 14ec4ebcd9af..bd3a60dd779f 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -35,7 +35,6 @@ static const char ice_copyright[] = "Copyright (c) 2018, Intel Corporation.";
 #define ICE_DDP_PKG_PATH	"intel/ice/ddp/"
 #define ICE_DDP_PKG_FILE	ICE_DDP_PKG_PATH "ice.pkg"
 
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
 MODULE_DESCRIPTION(DRV_SUMMARY);
 MODULE_IMPORT_NS(LIBIE);
 MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 3af03a211c3c..11be39f435f3 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -203,7 +203,6 @@ static const struct pci_error_handlers igb_err_handler = {
 
 static void igb_init_dmac(struct igb_adapter *adapter, u32 pba);
 
-MODULE_AUTHOR("Intel Corporation, <e1000-devel@lists.sourceforge.net>");
 MODULE_DESCRIPTION("Intel(R) Gigabit Ethernet Network Driver");
 MODULE_LICENSE("GPL v2");
 
diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c
index 7661edd7d0f2..925d7286a8ee 100644
--- a/drivers/net/ethernet/intel/igbvf/netdev.c
+++ b/drivers/net/ethernet/intel/igbvf/netdev.c
@@ -3001,7 +3001,6 @@ static void __exit igbvf_exit_module(void)
 }
 module_exit(igbvf_exit_module);
 
-MODULE_AUTHOR("Intel Corporation, <e1000-devel@lists.sourceforge.net>");
 MODULE_DESCRIPTION("Intel(R) Gigabit Virtual Function Network Driver");
 MODULE_LICENSE("GPL v2");
 
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 87b655b839c1..7a7cbed237d3 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -32,7 +32,6 @@
 
 static int debug = -1;
 
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
 MODULE_DESCRIPTION(DRV_SUMMARY);
 MODULE_LICENSE("GPL v2");
 module_param(debug, int, 0);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 094653e81b97..8057cef61f39 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -162,7 +162,6 @@ static int debug = -1;
 module_param(debug, int, 0);
 MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
 
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
 MODULE_DESCRIPTION("Intel(R) 10 Gigabit PCI Express Network Driver");
 MODULE_LICENSE("GPL v2");
 
diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
index b938dc06045d..149911e3002a 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
+++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
@@ -76,7 +76,6 @@ static const struct pci_device_id ixgbevf_pci_tbl[] = {
 };
 MODULE_DEVICE_TABLE(pci, ixgbevf_pci_tbl);
 
-MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
 MODULE_DESCRIPTION("Intel(R) 10 Gigabit Virtual Function Network Driver");
 MODULE_LICENSE("GPL v2");
 
diff --git a/drivers/net/ethernet/intel/libeth/rx.c b/drivers/net/ethernet/intel/libeth/rx.c
index 6221b88c34ac..bd135d6dccca 100644
--- a/drivers/net/ethernet/intel/libeth/rx.c
+++ b/drivers/net/ethernet/intel/libeth/rx.c
@@ -145,6 +145,5 @@ EXPORT_SYMBOL_NS_GPL(libeth_rx_pt_gen_hash_type, LIBETH);
 
 /* Module */
 
-MODULE_AUTHOR("Intel Corporation");
 MODULE_DESCRIPTION("Common Ethernet library");
 MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/intel/libie/rx.c b/drivers/net/ethernet/intel/libie/rx.c
index 38201ee1e891..aceb8d8813c4 100644
--- a/drivers/net/ethernet/intel/libie/rx.c
+++ b/drivers/net/ethernet/intel/libie/rx.c
@@ -118,7 +118,6 @@ const struct libeth_rx_pt libie_rx_pt_lut[LIBIE_RX_PT_NUM] = {
 };
 EXPORT_SYMBOL_NS_GPL(libie_rx_pt_lut, LIBIE);
 
-MODULE_AUTHOR("Intel Corporation");
 MODULE_DESCRIPTION("Intel(R) Ethernet common library");
 MODULE_IMPORT_NS(LIBETH);
 MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-58-ga151


From 4c8c36fe4e3b3f7d321f885b76afdf858962335b Mon Sep 17 00:00:00 2001
From: Simon Horman <horms@kernel.org>
Date: Fri, 5 Jul 2024 14:15:40 +0100
Subject: i40e: correct i40e_addr_to_hkey() name in kdoc

Correct name of i40e_addr_to_hkey() in it's kdoc.

kernel-doc -none reports:

 drivers/net/ethernet/intel/i40e/i40e.h:739: warning: expecting prototype for i40e_mac_to_hkey(). Prototype was for i40e_addr_to_hkey() instead

Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index bca2084cc54b..d546567e0286 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -735,7 +735,7 @@ __i40e_pf_next_veb(struct i40e_pf *pf, int *idx)
 	     _i++, _veb = __i40e_pf_next_veb(_pf, &_i))
 
 /**
- * i40e_mac_to_hkey - Convert a 6-byte MAC Address to a u64 hash key
+ * i40e_addr_to_hkey - Convert a 6-byte MAC Address to a u64 hash key
  * @macaddr: the MAC Address as the base key
  *
  * Simply copies the address and returns it as a u64 for hashing
-- 
cgit v1.2.3-58-ga151


From a59618b98543248adae7e77d60f67974ca243319 Mon Sep 17 00:00:00 2001
From: Pawel Kaminski <pawel.kaminski@intel.com>
Date: Fri, 7 Jun 2024 12:43:49 +0200
Subject: ice: Add support for devlink local_forwarding param

Add support for driver-specific devlink local_forwarding param.
Supported values are "enabled", "disabled" and "prioritized".
Default configuration is set to "enabled".

Add documentation in networking/devlink/ice.rst.

In previous generations of Intel NICs the transmit scheduler was only
limited by PCIe bandwidth when scheduling/assigning hairpin-bandwidth
between VFs. Changes to E810 HW design introduced scheduler limitation,
so that available hairpin-bandwidth is bound to external port speed.
In order to address this limitation and enable NFV services such as
"service chaining" a knob to adjust the scheduler config was created.
Driver can send a configuration message to the FW over admin queue and
internal FW logic will reconfigure HW to prioritize and add more BW to
VF to VF traffic. An end result, for example, 10G port will no longer
limit hairpin-bandwidth to 10G and much higher speeds can be achieved.

Devlink local_forwarding param set to "prioritized" enables higher
hairpin-bandwitdh on related PFs. Configuration is applicable only to
8x10G and 4x25G cards.

Changing local_forwarding configuration will trigger CORER reset in
order to take effect.

Example command to change current value:
devlink dev param set pci/0000:b2:00.3 name local_forwarding \
        value prioritized \
        cmode runtime

Co-developed-by: Michal Wilczynski <michal.wilczynski@intel.com>
Signed-off-by: Michal Wilczynski <michal.wilczynski@intel.com>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Signed-off-by: Pawel Kaminski <pawel.kaminski@intel.com>
Signed-off-by: Wojciech Drewek <wojciech.drewek@intel.com>
Tested-by: Rafal Romanowski <rafal.romanowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 Documentation/networking/devlink/ice.rst         |  25 +++++
 drivers/net/ethernet/intel/ice/devlink/devlink.c | 126 +++++++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_adminq_cmd.h  |  11 +-
 drivers/net/ethernet/intel/ice/ice_common.c      |   4 +
 drivers/net/ethernet/intel/ice/ice_type.h        |   1 +
 5 files changed, 166 insertions(+), 1 deletion(-)

diff --git a/Documentation/networking/devlink/ice.rst b/Documentation/networking/devlink/ice.rst
index 830c04354222..e3972d03cea0 100644
--- a/Documentation/networking/devlink/ice.rst
+++ b/Documentation/networking/devlink/ice.rst
@@ -11,6 +11,7 @@ Parameters
 ==========
 
 .. list-table:: Generic parameters implemented
+   :widths: 5 5 90
 
    * - Name
      - Mode
@@ -68,6 +69,30 @@ Parameters
 
        To verify that value has been set:
        $ devlink dev param show pci/0000:16:00.0 name tx_scheduling_layers
+.. list-table:: Driver specific parameters implemented
+    :widths: 5 5 90
+
+    * - Name
+      - Mode
+      - Description
+    * - ``local_forwarding``
+      - runtime
+      - Controls loopback behavior by tuning scheduler bandwidth.
+        It impacts all kinds of functions: physical, virtual and
+        subfunctions.
+        Supported values are:
+
+        ``enabled`` - loopback traffic is allowed on port
+
+        ``disabled`` - loopback traffic is not allowed on this port
+
+        ``prioritized`` - loopback traffic is prioritized on this port
+
+        Default value of ``local_forwarding`` parameter is ``enabled``.
+        ``prioritized`` provides ability to adjust loopback traffic rate to increase
+        one port capacity at cost of the another. User needs to disable
+        local forwarding on one of the ports in order have increased capacity
+        on the ``prioritized`` port.
 
 Info versions
 =============
diff --git a/drivers/net/ethernet/intel/ice/devlink/devlink.c b/drivers/net/ethernet/intel/ice/devlink/devlink.c
index f774781ab514..810a901d7afd 100644
--- a/drivers/net/ethernet/intel/ice/devlink/devlink.c
+++ b/drivers/net/ethernet/intel/ice/devlink/devlink.c
@@ -1381,9 +1381,129 @@ ice_devlink_enable_iw_validate(struct devlink *devlink, u32 id,
 	return 0;
 }
 
+#define DEVLINK_LOCAL_FWD_DISABLED_STR "disabled"
+#define DEVLINK_LOCAL_FWD_ENABLED_STR "enabled"
+#define DEVLINK_LOCAL_FWD_PRIORITIZED_STR "prioritized"
+
+/**
+ * ice_devlink_local_fwd_mode_to_str - Get string for local_fwd mode.
+ * @mode: local forwarding for mode used in port_info struct.
+ *
+ * Return: Mode respective string or "Invalid".
+ */
+static const char *
+ice_devlink_local_fwd_mode_to_str(enum ice_local_fwd_mode mode)
+{
+	switch (mode) {
+	case ICE_LOCAL_FWD_MODE_ENABLED:
+		return DEVLINK_LOCAL_FWD_ENABLED_STR;
+	case ICE_LOCAL_FWD_MODE_PRIORITIZED:
+		return DEVLINK_LOCAL_FWD_PRIORITIZED_STR;
+	case ICE_LOCAL_FWD_MODE_DISABLED:
+		return DEVLINK_LOCAL_FWD_DISABLED_STR;
+	}
+
+	return "Invalid";
+}
+
+/**
+ * ice_devlink_local_fwd_str_to_mode - Get local_fwd mode from string name.
+ * @mode_str: local forwarding mode string.
+ *
+ * Return: Mode value or negative number if invalid.
+ */
+static int ice_devlink_local_fwd_str_to_mode(const char *mode_str)
+{
+	if (!strcmp(mode_str, DEVLINK_LOCAL_FWD_ENABLED_STR))
+		return ICE_LOCAL_FWD_MODE_ENABLED;
+	else if (!strcmp(mode_str, DEVLINK_LOCAL_FWD_PRIORITIZED_STR))
+		return ICE_LOCAL_FWD_MODE_PRIORITIZED;
+	else if (!strcmp(mode_str, DEVLINK_LOCAL_FWD_DISABLED_STR))
+		return ICE_LOCAL_FWD_MODE_DISABLED;
+
+	return -EINVAL;
+}
+
+/**
+ * ice_devlink_local_fwd_get - Get local_fwd parameter.
+ * @devlink: Pointer to the devlink instance.
+ * @id: The parameter ID to set.
+ * @ctx: Context to store the parameter value.
+ *
+ * Return: Zero.
+ */
+static int ice_devlink_local_fwd_get(struct devlink *devlink, u32 id,
+				     struct devlink_param_gset_ctx *ctx)
+{
+	struct ice_pf *pf = devlink_priv(devlink);
+	struct ice_port_info *pi;
+	const char *mode_str;
+
+	pi = pf->hw.port_info;
+	mode_str = ice_devlink_local_fwd_mode_to_str(pi->local_fwd_mode);
+	snprintf(ctx->val.vstr, sizeof(ctx->val.vstr), "%s", mode_str);
+
+	return 0;
+}
+
+/**
+ * ice_devlink_local_fwd_set - Set local_fwd parameter.
+ * @devlink: Pointer to the devlink instance.
+ * @id: The parameter ID to set.
+ * @ctx: Context to get the parameter value.
+ * @extack: Netlink extended ACK structure.
+ *
+ * Return: Zero.
+ */
+static int ice_devlink_local_fwd_set(struct devlink *devlink, u32 id,
+				     struct devlink_param_gset_ctx *ctx,
+				     struct netlink_ext_ack *extack)
+{
+	int new_local_fwd_mode = ice_devlink_local_fwd_str_to_mode(ctx->val.vstr);
+	struct ice_pf *pf = devlink_priv(devlink);
+	struct device *dev = ice_pf_to_dev(pf);
+	struct ice_port_info *pi;
+
+	pi = pf->hw.port_info;
+	if (pi->local_fwd_mode != new_local_fwd_mode) {
+		pi->local_fwd_mode = new_local_fwd_mode;
+		dev_info(dev, "Setting local_fwd to %s\n", ctx->val.vstr);
+		ice_schedule_reset(pf, ICE_RESET_CORER);
+	}
+
+	return 0;
+}
+
+/**
+ * ice_devlink_local_fwd_validate - Validate passed local_fwd parameter value.
+ * @devlink: Unused pointer to devlink instance.
+ * @id: The parameter ID to validate.
+ * @val: Value to validate.
+ * @extack: Netlink extended ACK structure.
+ *
+ * Supported values are:
+ * "enabled" - local_fwd is enabled, "disabled" - local_fwd is disabled
+ * "prioritized" - local_fwd traffic is prioritized in scheduling.
+ *
+ * Return: Zero when passed parameter value is supported. Negative value on
+ * error.
+ */
+static int ice_devlink_local_fwd_validate(struct devlink *devlink, u32 id,
+					  union devlink_param_value val,
+					  struct netlink_ext_ack *extack)
+{
+	if (ice_devlink_local_fwd_str_to_mode(val.vstr) < 0) {
+		NL_SET_ERR_MSG_MOD(extack, "Error: Requested value is not supported.");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 enum ice_param_id {
 	ICE_DEVLINK_PARAM_ID_BASE = DEVLINK_PARAM_GENERIC_ID_MAX,
 	ICE_DEVLINK_PARAM_ID_TX_SCHED_LAYERS,
+	ICE_DEVLINK_PARAM_ID_LOCAL_FWD,
 };
 
 static const struct devlink_param ice_dvl_rdma_params[] = {
@@ -1405,6 +1525,12 @@ static const struct devlink_param ice_dvl_sched_params[] = {
 			     ice_devlink_tx_sched_layers_get,
 			     ice_devlink_tx_sched_layers_set,
 			     ice_devlink_tx_sched_layers_validate),
+	DEVLINK_PARAM_DRIVER(ICE_DEVLINK_PARAM_ID_LOCAL_FWD,
+			     "local_forwarding", DEVLINK_PARAM_TYPE_STRING,
+			     BIT(DEVLINK_PARAM_CMODE_RUNTIME),
+			     ice_devlink_local_fwd_get,
+			     ice_devlink_local_fwd_set,
+			     ice_devlink_local_fwd_validate),
 };
 
 static void ice_devlink_free(void *devlink_ptr)
diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
index b70d4ca43443..66f02988d549 100644
--- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h
@@ -232,6 +232,13 @@ struct ice_aqc_get_sw_cfg_resp_elem {
 #define ICE_AQC_GET_SW_CONF_RESP_IS_VF		BIT(15)
 };
 
+/* Loopback port parameter mode values. */
+enum ice_local_fwd_mode {
+	ICE_LOCAL_FWD_MODE_ENABLED = 0,
+	ICE_LOCAL_FWD_MODE_DISABLED = 1,
+	ICE_LOCAL_FWD_MODE_PRIORITIZED = 2,
+};
+
 /* Set Port parameters, (direct, 0x0203) */
 struct ice_aqc_set_port_params {
 	__le16 cmd_flags;
@@ -240,7 +247,9 @@ struct ice_aqc_set_port_params {
 	__le16 swid;
 #define ICE_AQC_PORT_SWID_VALID			BIT(15)
 #define ICE_AQC_PORT_SWID_M			0xFF
-	u8 reserved[10];
+	u8 local_fwd_mode;
+#define ICE_AQC_SET_P_PARAMS_LOCAL_FWD_MODE_VALID BIT(2)
+	u8 reserved[9];
 };
 
 /* These resource type defines are used for all switch resource
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index e311a41a74fa..9cd649053ef8 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -1086,6 +1086,7 @@ int ice_init_hw(struct ice_hw *hw)
 		goto err_unroll_cqinit;
 	}
 
+	hw->port_info->local_fwd_mode = ICE_LOCAL_FWD_MODE_ENABLED;
 	/* set the back pointer to HW */
 	hw->port_info->hw = hw;
 
@@ -3071,6 +3072,9 @@ ice_aq_set_port_params(struct ice_port_info *pi, bool double_vlan,
 		cmd_flags |= ICE_AQC_SET_P_PARAMS_DOUBLE_VLAN_ENA;
 	cmd->cmd_flags = cpu_to_le16(cmd_flags);
 
+	cmd->local_fwd_mode = pi->local_fwd_mode |
+				ICE_AQC_SET_P_PARAMS_LOCAL_FWD_MODE_VALID;
+
 	return ice_aq_send_cmd(hw, &desc, NULL, 0, cd);
 }
 
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index e26ae79578ba..f3e376cbdd92 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -738,6 +738,7 @@ struct ice_port_info {
 	u16 sw_id;			/* Initial switch ID belongs to port */
 	u16 pf_vf_num;
 	u8 port_state;
+	u8 local_fwd_mode;
 #define ICE_SCHED_PORT_STATE_INIT	0x0
 #define ICE_SCHED_PORT_STATE_READY	0x1
 	u8 lport;
-- 
cgit v1.2.3-58-ga151


From aff6e32cc7ffb271b07b6390ef1580b7ec576a8f Mon Sep 17 00:00:00 2001
From: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Date: Mon, 24 Jun 2024 10:05:10 +0200
Subject: ice: remove eswitch rebuild

Since the port representors are added one by one there is no need to do
eswitch rebuild. Each port representor is detached and attached in VF
reset path.

Reviewed-by: Wojciech Drewek <wojciech.drewek@intel.com>
Signed-off-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/ice/ice_eswitch.c | 16 ----------------
 drivers/net/ethernet/intel/ice/ice_eswitch.h |  6 ------
 drivers/net/ethernet/intel/ice/ice_main.c    |  2 --
 3 files changed, 24 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_eswitch.c b/drivers/net/ethernet/intel/ice/ice_eswitch.c
index 4f539b1c7781..3cfa071e3718 100644
--- a/drivers/net/ethernet/intel/ice/ice_eswitch.c
+++ b/drivers/net/ethernet/intel/ice/ice_eswitch.c
@@ -536,22 +536,6 @@ void ice_eswitch_detach(struct ice_pf *pf, struct ice_vf *vf)
 	devl_unlock(devlink);
 }
 
-/**
- * ice_eswitch_rebuild - rebuild eswitch
- * @pf: pointer to PF structure
- */
-void ice_eswitch_rebuild(struct ice_pf *pf)
-{
-	struct ice_repr *repr;
-	unsigned long id;
-
-	if (!ice_is_switchdev_running(pf))
-		return;
-
-	xa_for_each(&pf->eswitch.reprs, id, repr)
-		ice_eswitch_detach(pf, repr->vf);
-}
-
 /**
  * ice_eswitch_get_target - get netdev based on src_vsi from descriptor
  * @rx_ring: ring used to receive the packet
diff --git a/drivers/net/ethernet/intel/ice/ice_eswitch.h b/drivers/net/ethernet/intel/ice/ice_eswitch.h
index 09194d514f9b..78fd39a6935d 100644
--- a/drivers/net/ethernet/intel/ice/ice_eswitch.h
+++ b/drivers/net/ethernet/intel/ice/ice_eswitch.h
@@ -10,7 +10,6 @@
 void ice_eswitch_detach(struct ice_pf *pf, struct ice_vf *vf);
 int
 ice_eswitch_attach(struct ice_pf *pf, struct ice_vf *vf);
-void ice_eswitch_rebuild(struct ice_pf *pf);
 
 int ice_eswitch_mode_get(struct devlink *devlink, u16 *mode);
 int
@@ -54,11 +53,6 @@ static inline int ice_eswitch_configure(struct ice_pf *pf)
 	return 0;
 }
 
-static inline int ice_eswitch_rebuild(struct ice_pf *pf)
-{
-	return -EOPNOTSUPP;
-}
-
 static inline int ice_eswitch_mode_get(struct devlink *devlink, u16 *mode)
 {
 	return DEVLINK_ESWITCH_MODE_LEGACY;
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index bd3a60dd779f..ec636be4d17d 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -7702,8 +7702,6 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type)
 		goto err_vsi_rebuild;
 	}
 
-	ice_eswitch_rebuild(pf);
-
 	if (reset_type == ICE_RESET_PFR) {
 		err = ice_rebuild_channels(pf);
 		if (err) {
-- 
cgit v1.2.3-58-ga151


From 1712c9ee36d1f7b00e89bcd9743c18ba00a47f42 Mon Sep 17 00:00:00 2001
From: Sasha Neftin <sasha.neftin@intel.com>
Date: Thu, 13 Jun 2024 08:17:49 +0300
Subject: igc: Remove the internal 'eee_advert' field

Since the kernel's 'ethtool_keee' structure is in use, the internal
'eee_advert' field becomes pointless and can be removed.

This patch comes to clean up this redundant code.

Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Mor Bar-Gabay <morx.bar.gabay@intel.com> (A Contingent Worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/igc/igc.h         | 1 -
 drivers/net/ethernet/intel/igc/igc_ethtool.c | 6 ------
 drivers/net/ethernet/intel/igc/igc_main.c    | 3 ---
 3 files changed, 10 deletions(-)

diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h
index 8b14c029eda1..c38b4d0f00ce 100644
--- a/drivers/net/ethernet/intel/igc/igc.h
+++ b/drivers/net/ethernet/intel/igc/igc.h
@@ -202,7 +202,6 @@ struct igc_adapter {
 	struct net_device *netdev;
 
 	struct ethtool_keee eee;
-	u16 eee_advert;
 
 	unsigned long state;
 	unsigned int flags;
diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c
index 0cd2bd695db1..ab17170fe7e6 100644
--- a/drivers/net/ethernet/intel/igc/igc_ethtool.c
+++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c
@@ -1636,10 +1636,6 @@ static int igc_ethtool_get_eee(struct net_device *netdev,
 	linkmode_set_bit(ETHTOOL_LINK_MODE_100baseT_Full_BIT,
 			 edata->supported);
 
-	if (hw->dev_spec._base.eee_enable)
-		mii_eee_cap1_mod_linkmode_t(edata->advertised,
-					    adapter->eee_advert);
-
 	eeer = rd32(IGC_EEER);
 
 	/* EEE status on negotiated link */
@@ -1700,8 +1696,6 @@ static int igc_ethtool_set_eee(struct net_device *netdev,
 		return -EINVAL;
 	}
 
-	adapter->eee_advert = linkmode_to_mii_eee_cap1_t(edata->advertised);
-
 	if (hw->dev_spec._base.eee_enable != edata->eee_enabled) {
 		hw->dev_spec._base.eee_enable = edata->eee_enabled;
 		adapter->flags |= IGC_FLAG_EEE;
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 7a7cbed237d3..cb5c7b09e8a0 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -4975,9 +4975,6 @@ void igc_up(struct igc_adapter *adapter)
 	/* start the watchdog. */
 	hw->mac.get_link_status = true;
 	schedule_work(&adapter->watchdog_task);
-
-	adapter->eee_advert = MDIO_EEE_100TX | MDIO_EEE_1000T |
-			      MDIO_EEE_2_5GT;
 }
 
 /**
-- 
cgit v1.2.3-58-ga151


From b3470da314fd8018ee237e382000c4154a942420 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Thu, 11 Jul 2024 19:23:21 +0100
Subject: bpf: annotate BTF show functions with __printf

-Werror=suggest-attribute=format warns about two functions
in kernel/bpf/btf.c [1]; add __printf() annotations to silence
these warnings since for CONFIG_WERROR=y they will trigger
build failures.

[1] https://lore.kernel.org/bpf/a8b20c72-6631-4404-9e1f-0410642d7d20@gmail.com/

Fixes: 31d0bc81637d ("bpf: Move to generic BTF show support, apply it to seq files/strings")
Reported-by: Mirsad Todorovac <mtodorovac69@gmail.com>
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Tested-by: Mirsad Todorovac <mtodorovac69@yahoo.com>
Link: https://lore.kernel.org/r/20240711182321.963667-1-alan.maguire@oracle.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/btf.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 4ff11779699e..d5019c4454d6 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -7538,8 +7538,8 @@ static void btf_type_show(const struct btf *btf, u32 type_id, void *obj,
 	btf_type_ops(t)->show(btf, t, type_id, obj, 0, show);
 }
 
-static void btf_seq_show(struct btf_show *show, const char *fmt,
-			 va_list args)
+__printf(2, 0) static void btf_seq_show(struct btf_show *show, const char *fmt,
+					va_list args)
 {
 	seq_vprintf((struct seq_file *)show->target, fmt, args);
 }
@@ -7572,8 +7572,8 @@ struct btf_show_snprintf {
 	int len;		/* length we would have written */
 };
 
-static void btf_snprintf_show(struct btf_show *show, const char *fmt,
-			      va_list args)
+__printf(2, 0) static void btf_snprintf_show(struct btf_show *show, const char *fmt,
+					     va_list args)
 {
 	struct btf_show_snprintf *ssnprintf = (struct btf_show_snprintf *)show;
 	int len;
-- 
cgit v1.2.3-58-ga151


From a808878308a8041ae10a151d69e2d22f94cae9f4 Mon Sep 17 00:00:00 2001
From: Shay Drory <shayd@nvidia.com>
Date: Sun, 14 Apr 2024 11:05:25 +0300
Subject: driver core: auxiliary bus: show auxiliary device IRQs

PCI subfunctions (SF) are anchored on the auxiliary bus. PCI physical
and virtual functions are anchored on the PCI bus. The irq information
of each such function is visible to users via sysfs directory "msi_irqs"
containing files for each irq entry. However, for PCI SFs such
information is unavailable. Due to this users have no visibility on IRQs
used by the SFs.
Secondly, an SF can be multi function device supporting rdma, netdevice
and more. Without irq information at the bus level, the user is unable
to view or use the affinity of the SF IRQs.

Hence to match to the equivalent PCI PFs and VFs, add "irqs" directory,
for supporting auxiliary devices, containing file for each irq entry.

For example:
$ ls /sys/bus/auxiliary/devices/mlx5_core.sf.1/irqs/
50  51  52  53  54  55  56  57  58

Cc: Simon Horman <horms@kernel.org>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Parav Pandit <parav@nvidia.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

---
v9-v10:
- remove Przemek RB
- add name field to auxiliary_irq_info (Greg and Przemek)
- handle bogus IRQ in auxiliary_device_sysfs_irq_remove (Greg)
v8-v9:
- add Przemek RB
- use guard() in auxiliary_irq_dir_prepare (Paolo)
v7-v8:
- use cleanup.h for info and name fields (Greg)
- correct error flow in auxiliary_irq_dir_prepare (Przemek)
- add documentation for new fields of auxiliary_device (Simon)
v6-v7:
- dynamically creating irqs directory when first irq file created (Greg)
- removed irqs flag and simplified the dev_add() API (Greg)
- move sysfs related new code to a new auxiliary_sysfs.c file (Greg)
v5-v6:
- removed concept of shared and exclusive and hence global xarray (Greg)
v4-v5:
- restore global mutex and replace refcount_t with simple integer (Greg)
v3->4:
- remove global mutex (Przemek)
v2->v3:
- fix function declaration in case SYSFS isn't defined
v1->v2:
- move #ifdefs from drivers/base/auxiliary.c to
  include/linux/auxiliary_bus.h (Greg)
- use EXPORT_SYMBOL_GPL instead of EXPORT_SYMBOL (Greg)
- Fix kzalloc(ref) to kzalloc(*ref) (Simon)
- Add return description in auxiliary_device_sysfs_irq_add() kdoc (Simon)
- Fix auxiliary_irq_mode_show doc (kernel test boot)
---
 Documentation/ABI/testing/sysfs-bus-auxiliary |   9 ++
 drivers/base/Makefile                         |   1 +
 drivers/base/auxiliary.c                      |   1 +
 drivers/base/auxiliary_sysfs.c                | 113 ++++++++++++++++++++++++++
 include/linux/auxiliary_bus.h                 |  24 ++++++
 5 files changed, 148 insertions(+)
 create mode 100644 Documentation/ABI/testing/sysfs-bus-auxiliary
 create mode 100644 drivers/base/auxiliary_sysfs.c

diff --git a/Documentation/ABI/testing/sysfs-bus-auxiliary b/Documentation/ABI/testing/sysfs-bus-auxiliary
new file mode 100644
index 000000000000..cc856079690f
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-bus-auxiliary
@@ -0,0 +1,9 @@
+What:		/sys/bus/auxiliary/devices/.../irqs/
+Date:		April, 2024
+Contact:	Shay Drory <shayd@nvidia.com>
+Description:
+		The /sys/devices/.../irqs directory contains a variable set of
+		files, with each file is named as irq number similar to PCI PF
+		or VF's irq number located in msi_irqs directory.
+		These irq files are added and removed dynamically when an IRQ
+		is requested and freed respectively for the PCI SF.
diff --git a/drivers/base/Makefile b/drivers/base/Makefile
index 3079bfe53d04..7fb21768ca36 100644
--- a/drivers/base/Makefile
+++ b/drivers/base/Makefile
@@ -16,6 +16,7 @@ obj-$(CONFIG_NUMA)	+= node.o
 obj-$(CONFIG_MEMORY_HOTPLUG) += memory.o
 ifeq ($(CONFIG_SYSFS),y)
 obj-$(CONFIG_MODULES)	+= module.o
+obj-$(CONFIG_AUXILIARY_BUS) += auxiliary_sysfs.o
 endif
 obj-$(CONFIG_SYS_HYPERVISOR) += hypervisor.o
 obj-$(CONFIG_REGMAP)	+= regmap/
diff --git a/drivers/base/auxiliary.c b/drivers/base/auxiliary.c
index d3a2c40c2f12..3f01f4ec69e5 100644
--- a/drivers/base/auxiliary.c
+++ b/drivers/base/auxiliary.c
@@ -287,6 +287,7 @@ int auxiliary_device_init(struct auxiliary_device *auxdev)
 
 	dev->bus = &auxiliary_bus_type;
 	device_initialize(&auxdev->dev);
+	mutex_init(&auxdev->sysfs.lock);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(auxiliary_device_init);
diff --git a/drivers/base/auxiliary_sysfs.c b/drivers/base/auxiliary_sysfs.c
new file mode 100644
index 000000000000..754f21730afd
--- /dev/null
+++ b/drivers/base/auxiliary_sysfs.c
@@ -0,0 +1,113 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2024, NVIDIA CORPORATION & AFFILIATES
+ */
+
+#include <linux/auxiliary_bus.h>
+#include <linux/slab.h>
+
+#define AUXILIARY_MAX_IRQ_NAME 11
+
+struct auxiliary_irq_info {
+	struct device_attribute sysfs_attr;
+	char name[AUXILIARY_MAX_IRQ_NAME];
+};
+
+static struct attribute *auxiliary_irq_attrs[] = {
+	NULL
+};
+
+static const struct attribute_group auxiliary_irqs_group = {
+	.name = "irqs",
+	.attrs = auxiliary_irq_attrs,
+};
+
+static int auxiliary_irq_dir_prepare(struct auxiliary_device *auxdev)
+{
+	int ret = 0;
+
+	guard(mutex)(&auxdev->sysfs.lock);
+	if (auxdev->sysfs.irq_dir_exists)
+		return 0;
+
+	ret = devm_device_add_group(&auxdev->dev, &auxiliary_irqs_group);
+	if (ret)
+		return ret;
+
+	auxdev->sysfs.irq_dir_exists = true;
+	xa_init(&auxdev->sysfs.irqs);
+	return 0;
+}
+
+/**
+ * auxiliary_device_sysfs_irq_add - add a sysfs entry for the given IRQ
+ * @auxdev: auxiliary bus device to add the sysfs entry.
+ * @irq: The associated interrupt number.
+ *
+ * This function should be called after auxiliary device have successfully
+ * received the irq.
+ * The driver is responsible to add a unique irq for the auxiliary device. The
+ * driver can invoke this function from multiple thread context safely for
+ * unique irqs of the auxiliary devices. The driver must not invoke this API
+ * multiple times if the irq is already added previously.
+ *
+ * Return: zero on success or an error code on failure.
+ */
+int auxiliary_device_sysfs_irq_add(struct auxiliary_device *auxdev, int irq)
+{
+	struct auxiliary_irq_info *info __free(kfree) = NULL;
+	struct device *dev = &auxdev->dev;
+	int ret;
+
+	ret = auxiliary_irq_dir_prepare(auxdev);
+	if (ret)
+		return ret;
+
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	if (!info)
+		return -ENOMEM;
+
+	sysfs_attr_init(&info->sysfs_attr.attr);
+	snprintf(info->name, AUXILIARY_MAX_IRQ_NAME, "%d", irq);
+
+	ret = xa_insert(&auxdev->sysfs.irqs, irq, info, GFP_KERNEL);
+	if (ret)
+		return ret;
+
+	info->sysfs_attr.attr.name = info->name;
+	ret = sysfs_add_file_to_group(&dev->kobj, &info->sysfs_attr.attr,
+				      auxiliary_irqs_group.name);
+	if (ret)
+		goto sysfs_add_err;
+
+	xa_store(&auxdev->sysfs.irqs, irq, no_free_ptr(info), GFP_KERNEL);
+	return 0;
+
+sysfs_add_err:
+	xa_erase(&auxdev->sysfs.irqs, irq);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(auxiliary_device_sysfs_irq_add);
+
+/**
+ * auxiliary_device_sysfs_irq_remove - remove a sysfs entry for the given IRQ
+ * @auxdev: auxiliary bus device to add the sysfs entry.
+ * @irq: the IRQ to remove.
+ *
+ * This function should be called to remove an IRQ sysfs entry.
+ * The driver must invoke this API when IRQ is released by the device.
+ */
+void auxiliary_device_sysfs_irq_remove(struct auxiliary_device *auxdev, int irq)
+{
+	struct auxiliary_irq_info *info __free(kfree) = xa_load(&auxdev->sysfs.irqs, irq);
+	struct device *dev = &auxdev->dev;
+
+	if (!info) {
+		dev_err(&auxdev->dev, "IRQ %d doesn't exist\n", irq);
+		return;
+	}
+	sysfs_remove_file_from_group(&dev->kobj, &info->sysfs_attr.attr,
+				     auxiliary_irqs_group.name);
+	xa_erase(&auxdev->sysfs.irqs, irq);
+}
+EXPORT_SYMBOL_GPL(auxiliary_device_sysfs_irq_remove);
diff --git a/include/linux/auxiliary_bus.h b/include/linux/auxiliary_bus.h
index de21d9d24a95..3ba4487c9cd9 100644
--- a/include/linux/auxiliary_bus.h
+++ b/include/linux/auxiliary_bus.h
@@ -58,6 +58,9 @@
  *       in
  * @name: Match name found by the auxiliary device driver,
  * @id: unique identitier if multiple devices of the same name are exported,
+ * @irqs: irqs xarray contains irq indices which are used by the device,
+ * @lock: Synchronize irq sysfs creation,
+ * @irq_dir_exists: whether "irqs" directory exists,
  *
  * An auxiliary_device represents a part of its parent device's functionality.
  * It is given a name that, combined with the registering drivers
@@ -139,6 +142,11 @@ struct auxiliary_device {
 	struct device dev;
 	const char *name;
 	u32 id;
+	struct {
+		struct xarray irqs;
+		struct mutex lock; /* Synchronize irq sysfs creation */
+		bool irq_dir_exists;
+	} sysfs;
 };
 
 /**
@@ -212,8 +220,24 @@ int auxiliary_device_init(struct auxiliary_device *auxdev);
 int __auxiliary_device_add(struct auxiliary_device *auxdev, const char *modname);
 #define auxiliary_device_add(auxdev) __auxiliary_device_add(auxdev, KBUILD_MODNAME)
 
+#ifdef CONFIG_SYSFS
+int auxiliary_device_sysfs_irq_add(struct auxiliary_device *auxdev, int irq);
+void auxiliary_device_sysfs_irq_remove(struct auxiliary_device *auxdev,
+				       int irq);
+#else /* CONFIG_SYSFS */
+static inline int
+auxiliary_device_sysfs_irq_add(struct auxiliary_device *auxdev, int irq)
+{
+	return 0;
+}
+
+static inline void
+auxiliary_device_sysfs_irq_remove(struct auxiliary_device *auxdev, int irq) {}
+#endif
+
 static inline void auxiliary_device_uninit(struct auxiliary_device *auxdev)
 {
+	mutex_destroy(&auxdev->sysfs.lock);
 	put_device(&auxdev->dev);
 }
 
-- 
cgit v1.2.3-58-ga151


From 0477d5168bbb8767275822830b47112519a8446d Mon Sep 17 00:00:00 2001
From: Shay Drory <shayd@nvidia.com>
Date: Sun, 14 Apr 2024 11:26:07 +0300
Subject: net/mlx5: Expose SFs IRQs

Expose the sysfs files for the IRQs that the mlx5 PCI SFs are using.
These entries are similar to PCI PFs and VFs in 'msi_irqs' directory.

Reviewed-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

---
v8-v9:
- add Przemek RB
v6->v7:
- remove not needed changes to mlx5 sfnum SF sysfs
v5->v6:
- fail IRQ creation in case auxiliary_device_sysfs_irq_add() failed
  (Parav and Przemek)
v2->v3:
- fix mlx5 sfnum SF sysfs
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c           |  6 +++---
 drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c | 18 +++++++++++++++++-
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h    |  6 ++++++
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h     | 12 ++++++++----
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c      | 12 +++++++++---
 5 files changed, 43 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 5693986ae656..5661f047702e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -714,7 +714,7 @@ err2:
 err1:
 	mlx5_cmd_allowed_opcode(dev, CMD_ALLOWED_OPCODE_ALL);
 	mlx5_eq_notifier_unregister(dev, &table->cq_err_nb);
-	mlx5_ctrl_irq_release(table->ctrl_irq);
+	mlx5_ctrl_irq_release(dev, table->ctrl_irq);
 	return err;
 }
 
@@ -730,7 +730,7 @@ static void destroy_async_eqs(struct mlx5_core_dev *dev)
 	cleanup_async_eq(dev, &table->cmd_eq, "cmd");
 	mlx5_cmd_allowed_opcode(dev, CMD_ALLOWED_OPCODE_ALL);
 	mlx5_eq_notifier_unregister(dev, &table->cq_err_nb);
-	mlx5_ctrl_irq_release(table->ctrl_irq);
+	mlx5_ctrl_irq_release(dev, table->ctrl_irq);
 }
 
 struct mlx5_eq *mlx5_get_async_eq(struct mlx5_core_dev *dev)
@@ -918,7 +918,7 @@ static int comp_irq_request_sf(struct mlx5_core_dev *dev, u16 vecidx)
 	af_desc.is_managed = 1;
 	cpumask_copy(&af_desc.mask, cpu_online_mask);
 	cpumask_andnot(&af_desc.mask, &af_desc.mask, &table->used_cpus);
-	irq = mlx5_irq_affinity_request(pool, &af_desc);
+	irq = mlx5_irq_affinity_request(dev, pool, &af_desc);
 	if (IS_ERR(irq))
 		return PTR_ERR(irq);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c b/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c
index 612e666ec263..f7b01b3f0cba 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/irq_affinity.c
@@ -112,15 +112,18 @@ irq_pool_find_least_loaded(struct mlx5_irq_pool *pool, const struct cpumask *req
 
 /**
  * mlx5_irq_affinity_request - request an IRQ according to the given mask.
+ * @dev: mlx5 core device which is requesting the IRQ.
  * @pool: IRQ pool to request from.
  * @af_desc: affinity descriptor for this IRQ.
  *
  * This function returns a pointer to IRQ, or ERR_PTR in case of error.
  */
 struct mlx5_irq *
-mlx5_irq_affinity_request(struct mlx5_irq_pool *pool, struct irq_affinity_desc *af_desc)
+mlx5_irq_affinity_request(struct mlx5_core_dev *dev, struct mlx5_irq_pool *pool,
+			  struct irq_affinity_desc *af_desc)
 {
 	struct mlx5_irq *least_loaded_irq, *new_irq;
+	int ret;
 
 	mutex_lock(&pool->lock);
 	least_loaded_irq = irq_pool_find_least_loaded(pool, &af_desc->mask);
@@ -153,6 +156,16 @@ out:
 			      mlx5_irq_read_locked(least_loaded_irq) / MLX5_EQ_REFS_PER_IRQ);
 unlock:
 	mutex_unlock(&pool->lock);
+	if (mlx5_irq_pool_is_sf_pool(pool)) {
+		ret = auxiliary_device_sysfs_irq_add(mlx5_sf_coredev_to_adev(dev),
+						     mlx5_irq_get_irq(least_loaded_irq));
+		if (ret) {
+			mlx5_core_err(dev, "Failed to create sysfs entry for irq %d, ret = %d\n",
+				      mlx5_irq_get_irq(least_loaded_irq), ret);
+			mlx5_irq_put(least_loaded_irq);
+			least_loaded_irq = ERR_PTR(ret);
+		}
+	}
 	return least_loaded_irq;
 }
 
@@ -164,6 +177,9 @@ void mlx5_irq_affinity_irq_release(struct mlx5_core_dev *dev, struct mlx5_irq *i
 	cpu = cpumask_first(mlx5_irq_get_affinity_mask(irq));
 	synchronize_irq(pci_irq_vector(pool->dev->pdev,
 				       mlx5_irq_get_index(irq)));
+	if (mlx5_irq_pool_is_sf_pool(pool))
+		auxiliary_device_sysfs_irq_remove(mlx5_sf_coredev_to_adev(dev),
+						  mlx5_irq_get_irq(irq));
 	if (mlx5_irq_put(irq))
 		if (pool->irqs_per_cpu)
 			cpu_put(pool, cpu);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index c38342b9f320..e764b720d9b2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -320,6 +320,12 @@ static inline bool mlx5_core_is_sf(const struct mlx5_core_dev *dev)
 	return dev->coredev_type == MLX5_COREDEV_SF;
 }
 
+static inline struct auxiliary_device *
+mlx5_sf_coredev_to_adev(struct mlx5_core_dev *mdev)
+{
+	return container_of(mdev->device, struct auxiliary_device, dev);
+}
+
 int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx);
 void mlx5_mdev_uninit(struct mlx5_core_dev *dev);
 int mlx5_init_one(struct mlx5_core_dev *dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
index 1088114e905d..0881e961d8b1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
@@ -25,7 +25,7 @@ int mlx5_set_msix_vec_count(struct mlx5_core_dev *dev, int devfn,
 int mlx5_get_default_msix_vec_count(struct mlx5_core_dev *dev, int num_vfs);
 
 struct mlx5_irq *mlx5_ctrl_irq_request(struct mlx5_core_dev *dev);
-void mlx5_ctrl_irq_release(struct mlx5_irq *ctrl_irq);
+void mlx5_ctrl_irq_release(struct mlx5_core_dev *dev, struct mlx5_irq *ctrl_irq);
 struct mlx5_irq *mlx5_irq_request(struct mlx5_core_dev *dev, u16 vecidx,
 				  struct irq_affinity_desc *af_desc,
 				  struct cpu_rmap **rmap);
@@ -36,13 +36,15 @@ int mlx5_irq_attach_nb(struct mlx5_irq *irq, struct notifier_block *nb);
 int mlx5_irq_detach_nb(struct mlx5_irq *irq, struct notifier_block *nb);
 struct cpumask *mlx5_irq_get_affinity_mask(struct mlx5_irq *irq);
 int mlx5_irq_get_index(struct mlx5_irq *irq);
+int mlx5_irq_get_irq(const struct mlx5_irq *irq);
 
 struct mlx5_irq_pool;
 #ifdef CONFIG_MLX5_SF
 struct mlx5_irq *mlx5_irq_affinity_irq_request_auto(struct mlx5_core_dev *dev,
 						    struct cpumask *used_cpus, u16 vecidx);
-struct mlx5_irq *mlx5_irq_affinity_request(struct mlx5_irq_pool *pool,
-					   struct irq_affinity_desc *af_desc);
+struct mlx5_irq *
+mlx5_irq_affinity_request(struct mlx5_core_dev *dev, struct mlx5_irq_pool *pool,
+			  struct irq_affinity_desc *af_desc);
 void mlx5_irq_affinity_irq_release(struct mlx5_core_dev *dev, struct mlx5_irq *irq);
 #else
 static inline
@@ -53,7 +55,8 @@ struct mlx5_irq *mlx5_irq_affinity_irq_request_auto(struct mlx5_core_dev *dev,
 }
 
 static inline struct mlx5_irq *
-mlx5_irq_affinity_request(struct mlx5_irq_pool *pool, struct irq_affinity_desc *af_desc)
+mlx5_irq_affinity_request(struct mlx5_core_dev *dev, struct mlx5_irq_pool *pool,
+			  struct irq_affinity_desc *af_desc)
 {
 	return ERR_PTR(-EOPNOTSUPP);
 }
@@ -61,6 +64,7 @@ mlx5_irq_affinity_request(struct mlx5_irq_pool *pool, struct irq_affinity_desc *
 static inline
 void mlx5_irq_affinity_irq_release(struct mlx5_core_dev *dev, struct mlx5_irq *irq)
 {
+	mlx5_irq_release_vector(irq);
 }
 #endif
 #endif /* __MLX5_IRQ_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index fb8787e30d3f..ac7c3a76b4cf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -367,6 +367,11 @@ struct cpumask *mlx5_irq_get_affinity_mask(struct mlx5_irq *irq)
 	return irq->mask;
 }
 
+int mlx5_irq_get_irq(const struct mlx5_irq *irq)
+{
+	return irq->map.virq;
+}
+
 int mlx5_irq_get_index(struct mlx5_irq *irq)
 {
 	return irq->map.index;
@@ -440,11 +445,12 @@ static void _mlx5_irq_release(struct mlx5_irq *irq)
 
 /**
  * mlx5_ctrl_irq_release - release a ctrl IRQ back to the system.
+ * @dev: mlx5 device that releasing the IRQ.
  * @ctrl_irq: ctrl IRQ to be released.
  */
-void mlx5_ctrl_irq_release(struct mlx5_irq *ctrl_irq)
+void mlx5_ctrl_irq_release(struct mlx5_core_dev *dev, struct mlx5_irq *ctrl_irq)
 {
-	_mlx5_irq_release(ctrl_irq);
+	mlx5_irq_affinity_irq_release(dev, ctrl_irq);
 }
 
 /**
@@ -473,7 +479,7 @@ struct mlx5_irq *mlx5_ctrl_irq_request(struct mlx5_core_dev *dev)
 		/* Allocate the IRQ in index 0. The vector was already allocated */
 		irq = irq_pool_request_vector(pool, 0, &af_desc, NULL);
 	} else {
-		irq = mlx5_irq_affinity_request(pool, &af_desc);
+		irq = mlx5_irq_affinity_request(dev, pool, &af_desc);
 	}
 
 	return irq;
-- 
cgit v1.2.3-58-ga151


From 2899d58462ba868287d6ff3acad3675e7adf934f Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 10 Jul 2024 10:40:42 -0700
Subject: ethtool: fail closed if we can't get max channel used in indirection
 tables

Commit 0d1b7d6c9274 ("bnxt: fix crashes when reducing ring count with
active RSS contexts") proves that allowing indirection table to contain
channels with out of bounds IDs may lead to crashes. Currently the
max channel check in the core gets skipped if driver can't fetch
the indirection table or when we can't allocate memory.

Both of those conditions should be extremely rare but if they do
happen we should try to be safe and fail the channel change.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://patch.msgid.link/20240710174043.754664-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ethtool/channels.c |  6 ++----
 net/ethtool/common.c   | 26 +++++++++++++++-----------
 net/ethtool/common.h   |  2 +-
 net/ethtool/ioctl.c    |  4 +---
 4 files changed, 19 insertions(+), 19 deletions(-)

diff --git a/net/ethtool/channels.c b/net/ethtool/channels.c
index 7b4bbd674bae..cee188da54f8 100644
--- a/net/ethtool/channels.c
+++ b/net/ethtool/channels.c
@@ -171,11 +171,9 @@ ethnl_set_channels(struct ethnl_req_info *req_info, struct genl_info *info)
 	 */
 	if (ethtool_get_max_rxnfc_channel(dev, &max_rxnfc_in_use))
 		max_rxnfc_in_use = 0;
-	if (!netif_is_rxfh_configured(dev) ||
-	    ethtool_get_max_rxfh_channel(dev, &max_rxfh_in_use))
-		max_rxfh_in_use = 0;
+	max_rxfh_in_use = ethtool_get_max_rxfh_channel(dev);
 	if (channels.combined_count + channels.rx_count <= max_rxfh_in_use) {
-		GENL_SET_ERR_MSG(info, "requested channel counts are too low for existing indirection table settings");
+		GENL_SET_ERR_MSG_FMT(info, "requested channel counts are too low for existing indirection table (%d)", max_rxfh_in_use);
 		return -EINVAL;
 	}
 	if (channels.combined_count + channels.rx_count <= max_rxnfc_in_use) {
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 6b2a360dcdf0..8a62375ebd1f 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -587,35 +587,39 @@ err_free_info:
 	return err;
 }
 
-int ethtool_get_max_rxfh_channel(struct net_device *dev, u32 *max)
+u32 ethtool_get_max_rxfh_channel(struct net_device *dev)
 {
 	struct ethtool_rxfh_param rxfh = {};
-	u32 dev_size, current_max = 0;
+	u32 dev_size, current_max;
 	int ret;
 
+	if (!netif_is_rxfh_configured(dev))
+		return 0;
+
 	if (!dev->ethtool_ops->get_rxfh_indir_size ||
 	    !dev->ethtool_ops->get_rxfh)
-		return -EOPNOTSUPP;
+		return 0;
 	dev_size = dev->ethtool_ops->get_rxfh_indir_size(dev);
 	if (dev_size == 0)
-		return -EOPNOTSUPP;
+		return 0;
 
 	rxfh.indir = kcalloc(dev_size, sizeof(rxfh.indir[0]), GFP_USER);
 	if (!rxfh.indir)
-		return -ENOMEM;
+		return U32_MAX;
 
 	ret = dev->ethtool_ops->get_rxfh(dev, &rxfh);
-	if (ret)
-		goto out;
+	if (ret) {
+		current_max = U32_MAX;
+		goto out_free;
+	}
 
+	current_max = 0;
 	while (dev_size--)
 		current_max = max(current_max, rxfh.indir[dev_size]);
 
-	*max = current_max;
-
-out:
+out_free:
 	kfree(rxfh.indir);
-	return ret;
+	return current_max;
 }
 
 int ethtool_check_ops(const struct ethtool_ops *ops)
diff --git a/net/ethtool/common.h b/net/ethtool/common.h
index 28b8aaaf9bcb..b55705a9ad5a 100644
--- a/net/ethtool/common.h
+++ b/net/ethtool/common.h
@@ -42,7 +42,7 @@ int __ethtool_get_link(struct net_device *dev);
 bool convert_legacy_settings_to_link_ksettings(
 	struct ethtool_link_ksettings *link_ksettings,
 	const struct ethtool_cmd *legacy_settings);
-int ethtool_get_max_rxfh_channel(struct net_device *dev, u32 *max);
+u32 ethtool_get_max_rxfh_channel(struct net_device *dev);
 int ethtool_get_max_rxnfc_channel(struct net_device *dev, u64 *max);
 int __ethtool_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info);
 
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index d72b0fec89af..615812ff8974 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -2049,9 +2049,7 @@ static noinline_for_stack int ethtool_set_channels(struct net_device *dev,
 	 * indirection table/rxnfc settings */
 	if (ethtool_get_max_rxnfc_channel(dev, &max_rxnfc_in_use))
 		max_rxnfc_in_use = 0;
-	if (!netif_is_rxfh_configured(dev) ||
-	    ethtool_get_max_rxfh_channel(dev, &max_rxfh_in_use))
-		max_rxfh_in_use = 0;
+	max_rxfh_in_use = ethtool_get_max_rxfh_channel(dev);
 	if (channels.combined_count + channels.rx_count <=
 	    max_t(u64, max_rxnfc_in_use, max_rxfh_in_use))
 		return -EINVAL;
-- 
cgit v1.2.3-58-ga151


From 24ac7e5440815bb03bdfa9bc7e43a412b050dbaa Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Wed, 10 Jul 2024 10:40:43 -0700
Subject: ethtool: use the rss context XArray in ring deactivation safety-check

ethtool_get_max_rxfh_channel() gets called when user requests
deactivating Rx channels. Check the additional RSS contexts, too.

While we do track whether RSS context has an indirection
table explicitly set by the user, no driver looks at that bit.
Assume drivers won't auto-regenerate the additional tables,
to be safe.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://patch.msgid.link/20240710174043.754664-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ethtool/common.c | 33 +++++++++++++++++++++++++++++----
 1 file changed, 29 insertions(+), 4 deletions(-)

diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 8a62375ebd1f..7bda9600efcf 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -587,21 +587,47 @@ err_free_info:
 	return err;
 }
 
+static u32 ethtool_get_max_rss_ctx_channel(struct net_device *dev)
+{
+	struct ethtool_rxfh_context *ctx;
+	unsigned long context;
+	u32 max_ring = 0;
+
+	mutex_lock(&dev->ethtool->rss_lock);
+	xa_for_each(&dev->ethtool->rss_ctx, context, ctx) {
+		u32 i, *tbl;
+
+		tbl = ethtool_rxfh_context_indir(ctx);
+		for (i = 0; i < ctx->indir_size; i++)
+			max_ring = max(max_ring, tbl[i]);
+	}
+	mutex_unlock(&dev->ethtool->rss_lock);
+
+	return max_ring;
+}
+
 u32 ethtool_get_max_rxfh_channel(struct net_device *dev)
 {
 	struct ethtool_rxfh_param rxfh = {};
 	u32 dev_size, current_max;
 	int ret;
 
+	/* While we do track whether RSS context has an indirection
+	 * table explicitly set by the user, no driver looks at that bit.
+	 * Assume drivers won't auto-regenerate the additional tables,
+	 * to be safe.
+	 */
+	current_max = ethtool_get_max_rss_ctx_channel(dev);
+
 	if (!netif_is_rxfh_configured(dev))
-		return 0;
+		return current_max;
 
 	if (!dev->ethtool_ops->get_rxfh_indir_size ||
 	    !dev->ethtool_ops->get_rxfh)
-		return 0;
+		return current_max;
 	dev_size = dev->ethtool_ops->get_rxfh_indir_size(dev);
 	if (dev_size == 0)
-		return 0;
+		return current_max;
 
 	rxfh.indir = kcalloc(dev_size, sizeof(rxfh.indir[0]), GFP_USER);
 	if (!rxfh.indir)
@@ -613,7 +639,6 @@ u32 ethtool_get_max_rxfh_channel(struct net_device *dev)
 		goto out_free;
 	}
 
-	current_max = 0;
 	while (dev_size--)
 		current_max = max(current_max, rxfh.indir[dev_size]);
 
-- 
cgit v1.2.3-58-ga151


From b07593edd2faf310f9b758896f4f1c6054515be4 Mon Sep 17 00:00:00 2001
From: Chen Ni <nichen@iscas.ac.cn>
Date: Tue, 9 Jul 2024 15:28:38 +0800
Subject: net/sched: act_skbmod: convert comma to semicolon

Replace a comma between expression statements by a semicolon.

Signed-off-by: Chen Ni <nichen@iscas.ac.cn>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240709072838.1152880-1-nichen@iscas.ac.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/act_skbmod.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/sched/act_skbmod.c b/net/sched/act_skbmod.c
index cd0accaf844a..dc0229693461 100644
--- a/net/sched/act_skbmod.c
+++ b/net/sched/act_skbmod.c
@@ -246,7 +246,7 @@ static int tcf_skbmod_dump(struct sk_buff *skb, struct tc_action *a,
 
 	memset(&opt, 0, sizeof(opt));
 	opt.index   = d->tcf_index;
-	opt.refcnt  = refcount_read(&d->tcf_refcnt) - ref,
+	opt.refcnt  = refcount_read(&d->tcf_refcnt) - ref;
 	opt.bindcnt = atomic_read(&d->tcf_bindcnt) - bind;
 	spin_lock_bh(&d->tcf_lock);
 	opt.action = d->tcf_action;
-- 
cgit v1.2.3-58-ga151


From cef4902b0fadfc4181176ef5713f0b7cf2a40d8f Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Wed, 10 Jul 2024 15:16:53 +0000
Subject: net: reduce rtnetlink_rcv_msg() stack usage

IFLA_MAX is increasing slowly but surely.

Some compilers use more than 512 bytes of stack in rtnetlink_rcv_msg()
because it calls rtnl_calcit() for RTM_GETLINK message.

Use noinline_for_stack attribute to not inline rtnl_calcit(),
and directly use nla_for_each_attr_type() (Jakub suggestion)
because we only care about IFLA_EXT_MASK at this stage.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240710151653.3786604-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/rtnetlink.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index eabfc8290f5e..87e67194f240 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -3969,22 +3969,28 @@ static int rtnl_dellinkprop(struct sk_buff *skb, struct nlmsghdr *nlh,
 	return rtnl_linkprop(RTM_DELLINKPROP, skb, nlh, extack);
 }
 
-static u32 rtnl_calcit(struct sk_buff *skb, struct nlmsghdr *nlh)
+static noinline_for_stack u32 rtnl_calcit(struct sk_buff *skb,
+					  struct nlmsghdr *nlh)
 {
 	struct net *net = sock_net(skb->sk);
 	size_t min_ifinfo_dump_size = 0;
-	struct nlattr *tb[IFLA_MAX+1];
 	u32 ext_filter_mask = 0;
 	struct net_device *dev;
-	int hdrlen;
+	struct nlattr *nla;
+	int hdrlen, rem;
 
 	/* Same kernel<->userspace interface hack as in rtnl_dump_ifinfo. */
 	hdrlen = nlmsg_len(nlh) < sizeof(struct ifinfomsg) ?
 		 sizeof(struct rtgenmsg) : sizeof(struct ifinfomsg);
 
-	if (nlmsg_parse_deprecated(nlh, hdrlen, tb, IFLA_MAX, ifla_policy, NULL) >= 0) {
-		if (tb[IFLA_EXT_MASK])
-			ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
+	if (nlh->nlmsg_len < nlmsg_msg_size(hdrlen))
+		return NLMSG_GOODSIZE;
+
+	nla_for_each_attr_type(nla, IFLA_EXT_MASK,
+			       nlmsg_attrdata(nlh, hdrlen),
+			       nlmsg_attrlen(nlh, hdrlen), rem) {
+		if (nla_len(nla) == sizeof(u32))
+			ext_filter_mask = nla_get_u32(nla);
 	}
 
 	if (!ext_filter_mask)
-- 
cgit v1.2.3-58-ga151


From c9741a03dc8e491e57b95fba0058ab46b7e506da Mon Sep 17 00:00:00 2001
From: Alexey Kodanev <aleksei.kodanev@bell-sw.com>
Date: Mon, 8 Jul 2024 10:50:08 +0000
Subject: bna: adjust 'name' buf size of bna_tcb and bna_ccb structures

To have enough space to write all possible sprintf() args. Currently
'name' size is 16, but the first '%s' specifier may already need at
least 16 characters, since 'bnad->netdev->name' is used there.

For '%d' specifiers, assume that they require:
 * 1 char for 'tx_id + tx_info->tcb[i]->id' sum, BNAD_MAX_TXQ_PER_TX is 8
 * 2 chars for 'rx_id + rx_info->rx_ctrl[i].ccb->id', BNAD_MAX_RXP_PER_RX
   is 16

And replace sprintf with snprintf.

Detected using the static analysis tool - Svace.

Fixes: 8b230ed8ec96 ("bna: Brocade 10Gb Ethernet device driver")
Signed-off-by: Alexey Kodanev <aleksei.kodanev@bell-sw.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/brocade/bna/bna_types.h |  2 +-
 drivers/net/ethernet/brocade/bna/bnad.c      | 11 ++++++-----
 2 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/brocade/bna/bna_types.h b/drivers/net/ethernet/brocade/bna/bna_types.h
index a5ebd7110e07..986f43d27711 100644
--- a/drivers/net/ethernet/brocade/bna/bna_types.h
+++ b/drivers/net/ethernet/brocade/bna/bna_types.h
@@ -416,7 +416,7 @@ struct bna_ib {
 /* Tx object */
 
 /* Tx datapath control structure */
-#define BNA_Q_NAME_SIZE		16
+#define BNA_Q_NAME_SIZE		(IFNAMSIZ + 6)
 struct bna_tcb {
 	/* Fast path */
 	void			**sw_qpt;
diff --git a/drivers/net/ethernet/brocade/bna/bnad.c b/drivers/net/ethernet/brocade/bna/bnad.c
index fe121d36112d..ece6f3b48327 100644
--- a/drivers/net/ethernet/brocade/bna/bnad.c
+++ b/drivers/net/ethernet/brocade/bna/bnad.c
@@ -1534,8 +1534,9 @@ bnad_tx_msix_register(struct bnad *bnad, struct bnad_tx_info *tx_info,
 
 	for (i = 0; i < num_txqs; i++) {
 		vector_num = tx_info->tcb[i]->intr_vector;
-		sprintf(tx_info->tcb[i]->name, "%s TXQ %d", bnad->netdev->name,
-				tx_id + tx_info->tcb[i]->id);
+		snprintf(tx_info->tcb[i]->name, BNA_Q_NAME_SIZE, "%s TXQ %d",
+			 bnad->netdev->name,
+			 tx_id + tx_info->tcb[i]->id);
 		err = request_irq(bnad->msix_table[vector_num].vector,
 				  (irq_handler_t)bnad_msix_tx, 0,
 				  tx_info->tcb[i]->name,
@@ -1585,9 +1586,9 @@ bnad_rx_msix_register(struct bnad *bnad, struct bnad_rx_info *rx_info,
 
 	for (i = 0; i < num_rxps; i++) {
 		vector_num = rx_info->rx_ctrl[i].ccb->intr_vector;
-		sprintf(rx_info->rx_ctrl[i].ccb->name, "%s CQ %d",
-			bnad->netdev->name,
-			rx_id + rx_info->rx_ctrl[i].ccb->id);
+		snprintf(rx_info->rx_ctrl[i].ccb->name, BNA_Q_NAME_SIZE,
+			 "%s CQ %d", bnad->netdev->name,
+			 rx_id + rx_info->rx_ctrl[i].ccb->id);
 		err = request_irq(bnad->msix_table[vector_num].vector,
 				  (irq_handler_t)bnad_msix_rx, 0,
 				  rx_info->rx_ctrl[i].ccb->name,
-- 
cgit v1.2.3-58-ga151


From 8341eee81c794db0d8dd503c2b0ea2f55eba7334 Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Wed, 10 Jul 2024 19:10:04 +0200
Subject: net: psample: fix flag being set in wrong skb

A typo makes PSAMPLE_ATTR_SAMPLE_RATE netlink flag be added to the wrong
sk_buff.

Fix the error and make the input sk_buff pointer "const" so that it
doesn't happen again.

Acked-by: Eelco Chaudron <echaudro@redhat.com>
Fixes: 7b1b2b60c63f ("net: psample: allow using rate as probability")
Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Antoine Tenart <atenart@kernel.org>
Link: https://patch.msgid.link/20240710171004.2164034-1-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/psample.h | 8 +++++---
 net/psample/psample.c | 7 ++++---
 2 files changed, 9 insertions(+), 6 deletions(-)

diff --git a/include/net/psample.h b/include/net/psample.h
index c52e9ebd88dd..5071b5fc2b59 100644
--- a/include/net/psample.h
+++ b/include/net/psample.h
@@ -38,13 +38,15 @@ struct sk_buff;
 
 #if IS_ENABLED(CONFIG_PSAMPLE)
 
-void psample_sample_packet(struct psample_group *group, struct sk_buff *skb,
-			   u32 sample_rate, const struct psample_metadata *md);
+void psample_sample_packet(struct psample_group *group,
+			   const struct sk_buff *skb, u32 sample_rate,
+			   const struct psample_metadata *md);
 
 #else
 
 static inline void psample_sample_packet(struct psample_group *group,
-					 struct sk_buff *skb, u32 sample_rate,
+					 const struct sk_buff *skb,
+					 u32 sample_rate,
 					 const struct psample_metadata *md)
 {
 }
diff --git a/net/psample/psample.c b/net/psample/psample.c
index f48b5b9cd409..a0ddae8a65f9 100644
--- a/net/psample/psample.c
+++ b/net/psample/psample.c
@@ -360,8 +360,9 @@ static int psample_tunnel_meta_len(struct ip_tunnel_info *tun_info)
 }
 #endif
 
-void psample_sample_packet(struct psample_group *group, struct sk_buff *skb,
-			   u32 sample_rate, const struct psample_metadata *md)
+void psample_sample_packet(struct psample_group *group,
+			   const struct sk_buff *skb, u32 sample_rate,
+			   const struct psample_metadata *md)
 {
 	ktime_t tstamp = ktime_get_real();
 	int out_ifindex = md->out_ifindex;
@@ -498,7 +499,7 @@ void psample_sample_packet(struct psample_group *group, struct sk_buff *skb,
 		goto error;
 
 	if (md->rate_as_probability)
-		nla_put_flag(skb, PSAMPLE_ATTR_SAMPLE_PROBABILITY);
+		nla_put_flag(nl_skb, PSAMPLE_ATTR_SAMPLE_PROBABILITY);
 
 	genlmsg_end(nl_skb, data);
 	genlmsg_multicast_netns(&psample_nl_family, group->net, nl_skb, 0,
-- 
cgit v1.2.3-58-ga151


From 13cabc47f8ae69d24653f32c28399d493fde0a56 Mon Sep 17 00:00:00 2001
From: Alexander Lobakin <aleksander.lobakin@intel.com>
Date: Wed, 10 Jul 2024 04:30:28 -0700
Subject: netdevice: define and allocate &net_device _properly_

In fact, this structure contains a flexible array at the end, but
historically its size, alignment etc., is calculated manually.
There are several instances of the structure embedded into other
structures, but also there's ongoing effort to remove them and we
could in the meantime declare &net_device properly.
Declare the array explicitly, use struct_size() and store the array
size inside the structure, so that __counted_by() can be applied.
Don't use PTR_ALIGN(), as SLUB itself tries its best to ensure the
allocated buffer is aligned to what the user expects.
Also, change its alignment from %NETDEV_ALIGN to the cacheline size
as per several suggestions on the netdev ML.

bloat-o-meter for vmlinux:

free_netdev                                  445     440      -5
netdev_freemem                                24       -     -24
alloc_netdev_mqs                            1481    1450     -31

On x86_64 with several NICs of different vendors, I was never able to
get a &net_device pointer not aligned to the cacheline size after the
change.

Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kees Cook <kees@kernel.org>
Link: https://patch.msgid.link/20240710113036.2125584-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/netdevice.h | 15 +++++++++------
 net/core/dev.c            | 30 ++++++------------------------
 net/core/net-sysfs.c      |  2 +-
 3 files changed, 16 insertions(+), 31 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 93558645c6d0..607009150b5f 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1819,7 +1819,8 @@ enum netdev_reg_state {
  *	@priv_flags:	Like 'flags' but invisible to userspace,
  *			see if.h for the definitions
  *	@gflags:	Global flags ( kept as legacy )
- *	@padded:	How much padding added by alloc_netdev()
+ *	@priv_len:	Size of the ->priv flexible array
+ *	@priv:		Flexible array containing private data
  *	@operstate:	RFC2863 operstate
  *	@link_mode:	Mapping policy to operstate
  *	@if_port:	Selectable AUI, TP, ...
@@ -2199,10 +2200,10 @@ struct net_device {
 	unsigned short		neigh_priv_len;
 	unsigned short          dev_id;
 	unsigned short          dev_port;
-	unsigned short		padded;
+	int			irq;
+	u32			priv_len;
 
 	spinlock_t		addr_list_lock;
-	int			irq;
 
 	struct netdev_hw_addr_list	uc;
 	struct netdev_hw_addr_list	mc;
@@ -2406,7 +2407,10 @@ struct net_device {
 
 	/** @irq_moder: dim parameters used if IS_ENABLED(CONFIG_DIMLIB). */
 	struct dim_irq_moder	*irq_moder;
-};
+
+	u8			priv[] ____cacheline_aligned
+				       __counted_by(priv_len);
+} ____cacheline_aligned;
 #define to_net_dev(d) container_of(d, struct net_device, dev)
 
 /*
@@ -2596,7 +2600,7 @@ void dev_net_set(struct net_device *dev, struct net *net)
  */
 static inline void *netdev_priv(const struct net_device *dev)
 {
-	return (char *)dev + ALIGN(sizeof(struct net_device), NETDEV_ALIGN);
+	return (void *)dev->priv;
 }
 
 /* Set the sysfs physical device reference for the network logical device
@@ -3127,7 +3131,6 @@ static inline void unregister_netdevice(struct net_device *dev)
 
 int netdev_refcnt_read(const struct net_device *dev);
 void free_netdev(struct net_device *dev);
-void netdev_freemem(struct net_device *dev);
 void init_dummy_netdev(struct net_device *dev);
 
 struct net_device *netdev_get_xmit_slave(struct net_device *dev,
diff --git a/net/core/dev.c b/net/core/dev.c
index 73e5af6943c3..6ea1d20676fb 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -11006,13 +11006,6 @@ void netdev_sw_irq_coalesce_default_on(struct net_device *dev)
 }
 EXPORT_SYMBOL_GPL(netdev_sw_irq_coalesce_default_on);
 
-void netdev_freemem(struct net_device *dev)
-{
-	char *addr = (char *)dev - dev->padded;
-
-	kvfree(addr);
-}
-
 /**
  * alloc_netdev_mqs - allocate network device
  * @sizeof_priv: size of private data to allocate space for
@@ -11032,8 +11025,6 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
 		unsigned int txqs, unsigned int rxqs)
 {
 	struct net_device *dev;
-	unsigned int alloc_size;
-	struct net_device *p;
 
 	BUG_ON(strlen(name) >= sizeof(dev->name));
 
@@ -11047,21 +11038,12 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
 		return NULL;
 	}
 
-	alloc_size = sizeof(struct net_device);
-	if (sizeof_priv) {
-		/* ensure 32-byte alignment of private area */
-		alloc_size = ALIGN(alloc_size, NETDEV_ALIGN);
-		alloc_size += sizeof_priv;
-	}
-	/* ensure 32-byte alignment of whole construct */
-	alloc_size += NETDEV_ALIGN - 1;
-
-	p = kvzalloc(alloc_size, GFP_KERNEL_ACCOUNT | __GFP_RETRY_MAYFAIL);
-	if (!p)
+	dev = kvzalloc(struct_size(dev, priv, sizeof_priv),
+		       GFP_KERNEL_ACCOUNT | __GFP_RETRY_MAYFAIL);
+	if (!dev)
 		return NULL;
 
-	dev = PTR_ALIGN(p, NETDEV_ALIGN);
-	dev->padded = (char *)dev - (char *)p;
+	dev->priv_len = sizeof_priv;
 
 	ref_tracker_dir_init(&dev->refcnt_tracker, 128, name);
 #ifdef CONFIG_PCPU_DEV_REFCNT
@@ -11148,7 +11130,7 @@ free_pcpu:
 	free_percpu(dev->pcpu_refcnt);
 free_dev:
 #endif
-	netdev_freemem(dev);
+	kvfree(dev);
 	return NULL;
 }
 EXPORT_SYMBOL(alloc_netdev_mqs);
@@ -11203,7 +11185,7 @@ void free_netdev(struct net_device *dev)
 	/*  Compatibility with error handling in drivers */
 	if (dev->reg_state == NETREG_UNINITIALIZED ||
 	    dev->reg_state == NETREG_DUMMY) {
-		netdev_freemem(dev);
+		kvfree(dev);
 		return;
 	}
 
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index 4c27a360c294..0e2084ce7b75 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -2028,7 +2028,7 @@ static void netdev_release(struct device *d)
 	 * device is dead and about to be freed.
 	 */
 	kfree(rcu_access_pointer(dev->ifalias));
-	netdev_freemem(dev);
+	kvfree(dev);
 }
 
 static const void *net_namespace(const struct device *d)
-- 
cgit v1.2.3-58-ga151


From 5e724cb688a207ae7a348d57f9ea77f475998883 Mon Sep 17 00:00:00 2001
From: Adrian Moreno <amorenoz@redhat.com>
Date: Wed, 10 Jul 2024 11:04:59 +0200
Subject: selftests: openvswitch: retry instead of sleep

There are a couple of places where the test script "sleep"s to wait for
some external condition to be met.

This is error prone, specially in slow systems (identified in CI by
"KSFT_MACHINE_SLOW=yes").

To fix this, add a "ovs_wait" function that tries to execute a command
a few times until it succeeds. The timeout used is set to 5s for
"normal" systems and doubled if a slow CI machine is detected.

This should make the following work:

$ vng --build  \
    --config tools/testing/selftests/net/config \
    --config kernel/configs/debug.config

$ vng --run . --user root -- "make -C tools/testing/selftests/ \
    KSFT_MACHINE_SLOW=yes TARGETS=net/openvswitch run_tests"

Signed-off-by: Adrian Moreno <amorenoz@redhat.com>
Reviewed-by: Ilya Maximets <i.maximets@ovn.org>
Link: https://patch.msgid.link/20240710090500.1655212-1-amorenoz@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../selftests/net/openvswitch/openvswitch.sh       | 45 ++++++++++++++++++----
 .../testing/selftests/net/openvswitch/ovs-dpctl.py |  1 +
 2 files changed, 38 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/net/openvswitch/openvswitch.sh b/tools/testing/selftests/net/openvswitch/openvswitch.sh
index bc71dbc18b21..cc0bfae2bafa 100755
--- a/tools/testing/selftests/net/openvswitch/openvswitch.sh
+++ b/tools/testing/selftests/net/openvswitch/openvswitch.sh
@@ -11,6 +11,11 @@ ksft_skip=4
 PAUSE_ON_FAIL=no
 VERBOSE=0
 TRACING=0
+WAIT_TIMEOUT=5
+
+if test "X$KSFT_MACHINE_SLOW" == "Xyes"; then
+	WAIT_TIMEOUT=10
+fi
 
 tests="
 	arp_ping				eth-arp: Basic arp ping between two NS
@@ -29,6 +34,30 @@ info() {
 	[ $VERBOSE = 0 ] || echo $*
 }
 
+ovs_wait() {
+	info "waiting $WAIT_TIMEOUT s for: $@"
+
+	if "$@" ; then
+		info "wait succeeded immediately"
+		return 0
+	fi
+
+	# A quick re-check helps speed up small races in fast systems.
+	# However, fractional sleeps might not necessarily work.
+	local start=0
+	sleep 0.1 || { sleep 1; start=1; }
+
+	for (( i=start; i<WAIT_TIMEOUT; i++ )); do
+		if "$@" ; then
+			info "wait succeeded after $i seconds"
+			return 0
+		fi
+		sleep 1
+	done
+	info "wait failed after $i seconds"
+	return 1
+}
+
 ovs_base=`pwd`
 sbxs=
 sbx_add () {
@@ -278,20 +307,19 @@ test_psample() {
 
 	# Record psample data.
 	ovs_spawn_daemon "test_psample" python3 $ovs_base/ovs-dpctl.py psample-events
+	ovs_wait grep -q "listening for psample events" ${ovs_dir}/stdout
 
 	# Send a single ping.
-	sleep 1
 	ovs_sbx "test_psample" ip netns exec client ping -I c1 172.31.110.20 -c 1 || return 1
-	sleep 1
 
 	# We should have received one userspace action upcall and 2 psample packets.
-	grep -E "userspace action command" $ovs_dir/s0.out >/dev/null 2>&1 || return 1
+	ovs_wait grep -q "userspace action command" $ovs_dir/s0.out || return 1
 
 	# client -> server samples should only contain the first 14 bytes of the packet.
-	grep -E "rate:4294967295,group:1,cookie:c0ffee data:[0-9a-f]{28}$" \
-			 $ovs_dir/stdout >/dev/null 2>&1 || return 1
-	grep -E "rate:4294967295,group:2,cookie:eeff0c" \
-			 $ovs_dir/stdout >/dev/null 2>&1 || return 1
+	ovs_wait grep -qE "rate:4294967295,group:1,cookie:c0ffee data:[0-9a-f]{28}$" \
+		$ovs_dir/stdout || return 1
+
+	ovs_wait grep -q "rate:4294967295,group:2,cookie:eeff0c" $ovs_dir/stdout || return 1
 
 	return 0
 }
@@ -711,7 +739,8 @@ test_upcall_interfaces() {
 	ovs_add_netns_and_veths "test_upcall_interfaces" ui0 upc left0 l0 \
 	    172.31.110.1/24 -u || return 1
 
-	sleep 1
+	ovs_wait grep -q "listening on upcall packet handler" ${ovs_dir}/left0.out
+
 	info "sending arping"
 	ip netns exec upc arping -I l0 172.31.110.20 -c 1 \
 	    >$ovs_dir/arping.stdout 2>$ovs_dir/arping.stderr
diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 1e15b0818074..8a0396bfaf99 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -2520,6 +2520,7 @@ class PsampleEvent(EventSocket):
     marshal_class = psample_msg
 
     def read_samples(self):
+        print("listening for psample events", flush=True)
         while True:
             try:
                 for msg in self.get():
-- 
cgit v1.2.3-58-ga151


From a9359e8b0065a76cc11791f668bce3e9c7bf4fef Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Tue, 9 Jul 2024 07:43:59 -0700
Subject: net: netconsole: Remove unnecessary cast from bool

The 'enabled' variable is already a bool, so casting it to its value
is redundant.

Remove the superfluous cast, improving code clarity without changing
functionality.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20240709144403.544099-2-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/netconsole.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index ab8a0623b1a1..de0d89e4e4e2 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -344,7 +344,7 @@ static ssize_t enabled_store(struct config_item *item,
 		goto out_unlock;
 
 	err = -EINVAL;
-	if ((bool)enabled == nt->enabled) {
+	if (enabled == nt->enabled) {
 		pr_info("network logging has already %s\n",
 			nt->enabled ? "started" : "stopped");
 		goto out_unlock;
-- 
cgit v1.2.3-58-ga151


From 0066623d40085bb5d0af8d645a3bba27fc5cfced Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Tue, 9 Jul 2024 07:44:00 -0700
Subject: net: netconsole: Eliminate redundant setting of enabled field

When disabling a netconsole target, enabled_store() is called with
enabled=false. Currently, this results in updating the nt->enabled
field twice:

1. Inside the if/else block, with the target_list_lock spinlock held
2. Later, without the target_list_lock

This patch eliminates the redundancy by setting the field only once,
improving efficiency and reducing potential race conditions.

Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20240709144403.544099-3-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/netconsole.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index de0d89e4e4e2..5ef680cf994a 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -369,6 +369,7 @@ static ssize_t enabled_store(struct config_item *item,
 		if (err)
 			goto out_unlock;
 
+		nt->enabled = true;
 		pr_info("network logging started\n");
 	} else {	/* false */
 		/* We need to disable the netconsole before cleaning it up
@@ -381,8 +382,6 @@ static ssize_t enabled_store(struct config_item *item,
 		netpoll_cleanup(&nt->np);
 	}
 
-	nt->enabled = enabled;
-
 	mutex_unlock(&dynamic_netconsole_mutex);
 	return strnlen(buf, count);
 out_unlock:
-- 
cgit v1.2.3-58-ga151


From 534ea0a95e2d7b31eb1e1248d04edaffed75e366 Mon Sep 17 00:00:00 2001
From: Shigeru Yoshida <syoshida@redhat.com>
Date: Tue, 9 Jul 2024 23:34:10 +0900
Subject: tipc: Remove unused struct declaration

struct tipc_name_table in core.h is not used. Remove this declaration.

Signed-off-by: Shigeru Yoshida <syoshida@redhat.com>
Reviewed-by: Tung Nguyen <tung.q.nguyen@endava.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/tipc/core.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/tipc/core.h b/net/tipc/core.h
index 7eccd97e0609..7f3fe3401c45 100644
--- a/net/tipc/core.h
+++ b/net/tipc/core.h
@@ -72,7 +72,6 @@ struct tipc_node;
 struct tipc_bearer;
 struct tipc_bc_base;
 struct tipc_link;
-struct tipc_name_table;
 struct tipc_topsrv;
 struct tipc_monitor;
 #ifdef CONFIG_TIPC_CRYPTO
-- 
cgit v1.2.3-58-ga151


From b6c67967897e2e02482f7bbd26232090a6e30ae5 Mon Sep 17 00:00:00 2001
From: Shigeru Yoshida <syoshida@redhat.com>
Date: Tue, 9 Jul 2024 23:36:32 +0900
Subject: tipc: Consolidate redundant functions

link_is_up() and tipc_link_is_up() have the same functionality.
Consolidate these functions.

Signed-off-by: Shigeru Yoshida <syoshida@redhat.com>
Reviewed-by: Tung Nguyen <tung.q.nguyen@endava.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/tipc/link.c | 27 ++++++++++-----------------
 1 file changed, 10 insertions(+), 17 deletions(-)

diff --git a/net/tipc/link.c b/net/tipc/link.c
index 0716eb5c8a31..5c2088a469ce 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -241,13 +241,6 @@ enum {
 	LINK_SYNCHING        = 0xc  << 24
 };
 
-/* Link FSM state checking routines
- */
-static int link_is_up(struct tipc_link *l)
-{
-	return l->state & (LINK_ESTABLISHED | LINK_SYNCHING);
-}
-
 static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 			       struct sk_buff_head *xmitq);
 static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
@@ -274,7 +267,7 @@ static void tipc_link_update_cwin(struct tipc_link *l, int released,
  */
 bool tipc_link_is_up(struct tipc_link *l)
 {
-	return link_is_up(l);
+	return l->state & (LINK_ESTABLISHED | LINK_SYNCHING);
 }
 
 bool tipc_link_peer_is_down(struct tipc_link *l)
@@ -1790,7 +1783,7 @@ int tipc_link_rcv(struct tipc_link *l, struct sk_buff *skb,
 		rcv_nxt = l->rcv_nxt;
 		win_lim = rcv_nxt + TIPC_MAX_LINK_WIN;
 
-		if (unlikely(!link_is_up(l))) {
+		if (unlikely(!tipc_link_is_up(l))) {
 			if (l->state == LINK_ESTABLISHING)
 				rc = TIPC_LINK_UP_EVT;
 			kfree_skb(skb);
@@ -1848,7 +1841,7 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
 	struct tipc_link *bcl = l->bc_rcvlink;
 	struct tipc_msg *hdr;
 	struct sk_buff *skb;
-	bool node_up = link_is_up(bcl);
+	bool node_up = tipc_link_is_up(bcl);
 	u16 glen = 0, bc_rcvgap = 0;
 	int dlen = 0;
 	void *data;
@@ -2163,7 +2156,7 @@ bool tipc_link_validate_msg(struct tipc_link *l, struct tipc_msg *hdr)
 		if (session != curr_session)
 			return false;
 		/* Extra sanity check */
-		if (!link_is_up(l) && msg_ack(hdr))
+		if (!tipc_link_is_up(l) && msg_ack(hdr))
 			return false;
 		if (!(l->peer_caps & TIPC_LINK_PROTO_SEQNO))
 			return true;
@@ -2261,7 +2254,7 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 		}
 
 		/* ACTIVATE_MSG serves as PEER_RESET if link is already down */
-		if (mtyp == RESET_MSG || !link_is_up(l))
+		if (mtyp == RESET_MSG || !tipc_link_is_up(l))
 			rc = tipc_link_fsm_evt(l, LINK_PEER_RESET_EVT);
 
 		/* ACTIVATE_MSG takes up link if it was already locally reset */
@@ -2300,7 +2293,7 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 		if (msg_probe(hdr))
 			l->stats.recv_probes++;
 
-		if (!link_is_up(l)) {
+		if (!tipc_link_is_up(l)) {
 			if (l->state == LINK_ESTABLISHING)
 				rc = TIPC_LINK_UP_EVT;
 			break;
@@ -2387,7 +2380,7 @@ void tipc_link_bc_init_rcv(struct tipc_link *l, struct tipc_msg *hdr)
 	int mtyp = msg_type(hdr);
 	u16 peers_snd_nxt = msg_bc_snd_nxt(hdr);
 
-	if (link_is_up(l))
+	if (tipc_link_is_up(l))
 		return;
 
 	if (msg_user(hdr) == BCAST_PROTOCOL) {
@@ -2415,7 +2408,7 @@ int tipc_link_bc_sync_rcv(struct tipc_link *l, struct tipc_msg *hdr,
 	u16 peers_snd_nxt = msg_bc_snd_nxt(hdr);
 	int rc = 0;
 
-	if (!link_is_up(l))
+	if (!tipc_link_is_up(l))
 		return rc;
 
 	if (!msg_peer_node_is_up(hdr))
@@ -2475,7 +2468,7 @@ int tipc_link_bc_ack_rcv(struct tipc_link *r, u16 acked, u16 gap,
 	bool unused = false;
 	int rc = 0;
 
-	if (!link_is_up(r) || !r->bc_peer_is_up)
+	if (!tipc_link_is_up(r) || !r->bc_peer_is_up)
 		return 0;
 
 	if (gap) {
@@ -2873,7 +2866,7 @@ void tipc_link_set_tolerance(struct tipc_link *l, u32 tol,
 	l->tolerance = tol;
 	if (l->bc_rcvlink)
 		l->bc_rcvlink->tolerance = tol;
-	if (link_is_up(l))
+	if (tipc_link_is_up(l))
 		tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, tol, 0, xmitq);
 }
 
-- 
cgit v1.2.3-58-ga151


From 2146b7dd354c2a1384381ca3cd5751bfff6137d6 Mon Sep 17 00:00:00 2001
From: James Chapman <jchapman@katalix.com>
Date: Tue, 9 Jul 2024 17:28:39 +0100
Subject: l2tp: fix l2tp_session_register with colliding l2tpv3 IDs

When handling colliding L2TPv3 session IDs, we use the existing
session IDR entry and link the new session on that using
session->coll_list. However, when using an existing IDR entry, we must
not do the idr_replace step.

Fixes: aa5e17e1f5ec ("l2tp: store l2tpv3 sessions in per-net IDR")
Signed-off-by: James Chapman <jchapman@katalix.com>
Signed-off-by: Tom Parkin <tparkin@katalix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/l2tp/l2tp_core.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c
index 29dfbd70c79c..1c1decce7f06 100644
--- a/net/l2tp/l2tp_core.c
+++ b/net/l2tp/l2tp_core.c
@@ -436,6 +436,7 @@ int l2tp_session_register(struct l2tp_session *session,
 			  struct l2tp_tunnel *tunnel)
 {
 	struct l2tp_net *pn = l2tp_pernet(tunnel->l2tp_net);
+	struct l2tp_session *other_session = NULL;
 	u32 session_key;
 	int err;
 
@@ -456,11 +457,10 @@ int l2tp_session_register(struct l2tp_session *session,
 		 * support existing userspace which depends on it.
 		 */
 		if (err == -ENOSPC && tunnel->encap == L2TP_ENCAPTYPE_UDP) {
-			struct l2tp_session *session2;
-
-			session2 = idr_find(&pn->l2tp_v3_session_idr,
-					    session_key);
-			err = l2tp_session_collision_add(pn, session, session2);
+			other_session = idr_find(&pn->l2tp_v3_session_idr,
+						 session_key);
+			err = l2tp_session_collision_add(pn, session,
+							 other_session);
 		}
 		spin_unlock_bh(&pn->l2tp_session_idr_lock);
 	} else {
@@ -484,10 +484,12 @@ int l2tp_session_register(struct l2tp_session *session,
 	spin_unlock_bh(&tunnel->list_lock);
 
 	spin_lock_bh(&pn->l2tp_session_idr_lock);
-	if (tunnel->version == L2TP_HDR_VER_3)
-		idr_replace(&pn->l2tp_v3_session_idr, session, session_key);
-	else
+	if (tunnel->version == L2TP_HDR_VER_3) {
+		if (!other_session)
+			idr_replace(&pn->l2tp_v3_session_idr, session, session_key);
+	} else {
 		idr_replace(&pn->l2tp_v2_session_idr, session, session_key);
+	}
 	spin_unlock_bh(&pn->l2tp_session_idr_lock);
 
 	trace_register_session(session);
-- 
cgit v1.2.3-58-ga151


From f7e8542d71e908706855a9f3e9e7844fea1feea0 Mon Sep 17 00:00:00 2001
From: Mike Yu <yumike@google.com>
Date: Fri, 12 Jul 2024 10:51:22 +0800
Subject: xfrm: Support crypto offload for inbound IPv6 ESP packets not in GRO
 path

IPsec crypt offload supports outbound IPv6 ESP packets, but it doesn't
support inbound IPv6 ESP packets.

This change enables the crypto offload for inbound IPv6 ESP packets
that are not handled through GRO code path. If HW drivers add the
offload information to the skb, the packet will be handled in the
crypto offload rx code path.

Apart from the change in crypto offload rx code path, the change
in xfrm_policy_check is also needed.

Exampe of RX data path:

  +-----------+   +-------+
  | HW Driver |-->| wlan0 |--------+
  +-----------+   +-------+        |
                                   v
                             +---------------+   +------+
                     +------>| Network Stack |-->| Apps |
                     |       +---------------+   +------+
                     |             |
                     |             v
                 +--------+   +------------+
                 | ipsec1 |<--| XFRM Stack |
                 +--------+   +------------+

Test: Enabled both in/out IPsec crypto offload, and verified IPv6
      ESP packets on Android device on both wifi/cellular network
Signed-off-by: Mike Yu <yumike@google.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/xfrm/xfrm_input.c  | 2 +-
 net/xfrm/xfrm_policy.c | 5 ++++-
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index d2ea18dcb0cb..ba8deb0235ba 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -471,7 +471,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 	struct xfrm_offload *xo = xfrm_offload(skb);
 	struct sec_path *sp;
 
-	if (encap_type < 0 || (xo && xo->flags & XFRM_GRO)) {
+	if (encap_type < 0 || (xo && (xo->flags & XFRM_GRO || encap_type == 0))) {
 		x = xfrm_input_state(skb);
 
 		if (unlikely(x->dir && x->dir != XFRM_SA_DIR_IN)) {
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 6603d3bd171f..2a9a31f2a9c1 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -3718,12 +3718,15 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 		pol = xfrm_in_fwd_icmp(skb, &fl, family, if_id);
 
 	if (!pol) {
+		const bool is_crypto_offload = sp &&
+			(xfrm_input_state(skb)->xso.type == XFRM_DEV_OFFLOAD_CRYPTO);
+
 		if (net->xfrm.policy_default[dir] == XFRM_USERPOLICY_BLOCK) {
 			XFRM_INC_STATS(net, LINUX_MIB_XFRMINNOPOLS);
 			return 0;
 		}
 
-		if (sp && secpath_has_nontransport(sp, 0, &xerr_idx)) {
+		if (sp && secpath_has_nontransport(sp, 0, &xerr_idx) && !is_crypto_offload) {
 			xfrm_secpath_reject(xerr_idx, skb, &fl);
 			XFRM_INC_STATS(net, LINUX_MIB_XFRMINNOPOLS);
 			return 0;
-- 
cgit v1.2.3-58-ga151


From a10fb4a84a693891b2e205684456d0825363fc24 Mon Sep 17 00:00:00 2001
From: Mike Yu <yumike@google.com>
Date: Fri, 12 Jul 2024 10:51:23 +0800
Subject: xfrm: Allow UDP encapsulation in crypto offload control path

Unblock this limitation so that SAs with encapsulation specified
can be passed to HW drivers. HW drivers can still reject the SA
in their implementation of xdo_dev_state_add if the encapsulation
is not supported.

Test: Verified on Android device
Signed-off-by: Mike Yu <yumike@google.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/xfrm/xfrm_device.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/xfrm/xfrm_device.c b/net/xfrm/xfrm_device.c
index 2455a76a1cff..9a44d363ba62 100644
--- a/net/xfrm/xfrm_device.c
+++ b/net/xfrm/xfrm_device.c
@@ -261,9 +261,9 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x,
 
 	is_packet_offload = xuo->flags & XFRM_OFFLOAD_PACKET;
 
-	/* We don't yet support UDP encapsulation and TFC padding. */
-	if ((!is_packet_offload && x->encap) || x->tfcpad) {
-		NL_SET_ERR_MSG(extack, "Encapsulation and TFC padding can't be offloaded");
+	/* We don't yet support TFC padding. */
+	if (x->tfcpad) {
+		NL_SET_ERR_MSG(extack, "TFC padding can't be offloaded");
 		return -EINVAL;
 	}
 
-- 
cgit v1.2.3-58-ga151


From 4ecbac84b5b8c4ad033b473782582568810baf61 Mon Sep 17 00:00:00 2001
From: Mike Yu <yumike@google.com>
Date: Fri, 12 Jul 2024 10:51:24 +0800
Subject: xfrm: Support crypto offload for inbound IPv4 UDP-encapsulated ESP
 packet

If xfrm_input() is called with UDP_ENCAP_ESPINUDP, the packet is
already processed in UDP layer that removes the UDP header.
Therefore, there should be no much difference to treat it as an
ESP packet in the XFRM stack.

Test: Enabled dir=in IPsec crypto offload, and verified IPv4
      UDP-encapsulated ESP packets on both wifi/cellular network
Signed-off-by: Mike Yu <yumike@google.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/xfrm/xfrm_input.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index ba8deb0235ba..7cee9c0a2cdc 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -471,7 +471,8 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 	struct xfrm_offload *xo = xfrm_offload(skb);
 	struct sec_path *sp;
 
-	if (encap_type < 0 || (xo && (xo->flags & XFRM_GRO || encap_type == 0))) {
+	if (encap_type < 0 || (xo && (xo->flags & XFRM_GRO || encap_type == 0 ||
+				      encap_type == UDP_ENCAP_ESPINUDP))) {
 		x = xfrm_input_state(skb);
 
 		if (unlikely(x->dir && x->dir != XFRM_SA_DIR_IN)) {
-- 
cgit v1.2.3-58-ga151


From 447bc4b1906f100e65c662528b7ae4e1dc2e9b80 Mon Sep 17 00:00:00 2001
From: Mike Yu <yumike@google.com>
Date: Fri, 12 Jul 2024 10:51:25 +0800
Subject: xfrm: Support crypto offload for outbound IPv4 UDP-encapsulated ESP
 packet

esp_xmit() is already able to handle UDP encapsulation through the call to
esp_output_head(). However, the ESP header and the outer IP header
are not correct and need to be corrected.

Test: Enabled both dir=in/out IPsec crypto offload, and verified IPv4
      UDP-encapsulated ESP packets on both wifi/cellular network
Signed-off-by: Mike Yu <yumike@google.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
---
 net/ipv4/esp4.c         |  8 +++++++-
 net/ipv4/esp4_offload.c | 17 ++++++++++++++++-
 2 files changed, 23 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index 3968d3f98e08..73981595f062 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -349,6 +349,7 @@ static struct ip_esp_hdr *esp_output_udp_encap(struct sk_buff *skb,
 {
 	struct udphdr *uh;
 	unsigned int len;
+	struct xfrm_offload *xo = xfrm_offload(skb);
 
 	len = skb->len + esp->tailen - skb_transport_offset(skb);
 	if (len + sizeof(struct iphdr) > IP_MAX_MTU)
@@ -360,7 +361,12 @@ static struct ip_esp_hdr *esp_output_udp_encap(struct sk_buff *skb,
 	uh->len = htons(len);
 	uh->check = 0;
 
-	*skb_mac_header(skb) = IPPROTO_UDP;
+	/* For IPv4 ESP with UDP encapsulation, if xo is not null, the skb is in the crypto offload
+	 * data path, which means that esp_output_udp_encap is called outside of the XFRM stack.
+	 * In this case, the mac header doesn't point to the IPv4 protocol field, so don't set it.
+	 */
+	if (!xo || encap_type != UDP_ENCAP_ESPINUDP)
+		*skb_mac_header(skb) = IPPROTO_UDP;
 
 	return (struct ip_esp_hdr *)(uh + 1);
 }
diff --git a/net/ipv4/esp4_offload.c b/net/ipv4/esp4_offload.c
index b3271957ad9a..a37d18858c72 100644
--- a/net/ipv4/esp4_offload.c
+++ b/net/ipv4/esp4_offload.c
@@ -264,6 +264,7 @@ static int esp_xmit(struct xfrm_state *x, struct sk_buff *skb,  netdev_features_
 	struct esp_info esp;
 	bool hw_offload = true;
 	__u32 seq;
+	int encap_type = 0;
 
 	esp.inplace = true;
 
@@ -296,8 +297,10 @@ static int esp_xmit(struct xfrm_state *x, struct sk_buff *skb,  netdev_features_
 
 	esp.esph = ip_esp_hdr(skb);
 
+	if (x->encap)
+		encap_type = x->encap->encap_type;
 
-	if (!hw_offload || !skb_is_gso(skb)) {
+	if (!hw_offload || !skb_is_gso(skb) || (hw_offload && encap_type == UDP_ENCAP_ESPINUDP)) {
 		esp.nfrags = esp_output_head(x, skb, &esp);
 		if (esp.nfrags < 0)
 			return esp.nfrags;
@@ -324,6 +327,18 @@ static int esp_xmit(struct xfrm_state *x, struct sk_buff *skb,  netdev_features_
 
 	esp.seqno = cpu_to_be64(seq + ((u64)xo->seq.hi << 32));
 
+	if (hw_offload && encap_type == UDP_ENCAP_ESPINUDP) {
+		/* In the XFRM stack, the encapsulation protocol is set to iphdr->protocol by
+		 * setting *skb_mac_header(skb) (see esp_output_udp_encap()) where skb->mac_header
+		 * points to iphdr->protocol (see xfrm4_tunnel_encap_add()).
+		 * However, in esp_xmit(), skb->mac_header doesn't point to iphdr->protocol.
+		 * Therefore, the protocol field needs to be corrected.
+		 */
+		ip_hdr(skb)->protocol = IPPROTO_UDP;
+
+		esph->seq_no = htonl(seq);
+	}
+
 	ip_hdr(skb)->tot_len = htons(skb->len);
 	ip_send_check(ip_hdr(skb));
 
-- 
cgit v1.2.3-58-ga151


From 338a93cf4a18c2036b567e9f613367f7a52f2511 Mon Sep 17 00:00:00 2001
From: Jeremy Kerr <jk@codeconstruct.com.au>
Date: Wed, 10 Jul 2024 10:17:22 +0800
Subject: net: mctp-i2c: invalidate flows immediately on TX errors

If we encounter an error on i2c packet transmit, we won't have a valid
flow anymore; since we didn't transmit a valid packet sequence, we'll
have to wait for the key to timeout instead of dropping it on the reply.

This causes the i2c lock to be held for longer than necessary.

Instead, invalidate the flow on TX error, and release the i2c lock
immediately.

Cc: Bonnie Lo <Bonnie_Lo@wiwynn.com>
Tested-by: Jerry C Chen <Jerry_C_Chen@wiwynn.com>
Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mctp/mctp-i2c.c | 41 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

diff --git a/drivers/net/mctp/mctp-i2c.c b/drivers/net/mctp/mctp-i2c.c
index f9afea25044f..4dc057c121f5 100644
--- a/drivers/net/mctp/mctp-i2c.c
+++ b/drivers/net/mctp/mctp-i2c.c
@@ -442,6 +442,42 @@ static void mctp_i2c_unlock_reset(struct mctp_i2c_dev *midev)
 		i2c_unlock_bus(midev->adapter, I2C_LOCK_SEGMENT);
 }
 
+static void mctp_i2c_invalidate_tx_flow(struct mctp_i2c_dev *midev,
+					struct sk_buff *skb)
+{
+	struct mctp_sk_key *key;
+	struct mctp_flow *flow;
+	unsigned long flags;
+	bool release;
+
+	flow = skb_ext_find(skb, SKB_EXT_MCTP);
+	if (!flow)
+		return;
+
+	key = flow->key;
+	if (!key)
+		return;
+
+	spin_lock_irqsave(&key->lock, flags);
+	if (key->manual_alloc) {
+		/* we don't have control over lifetimes for manually-allocated
+		 * keys, so cannot assume we can invalidate all future flows
+		 * that would use this key.
+		 */
+		release = false;
+	} else {
+		release = key->dev_flow_state == MCTP_I2C_FLOW_STATE_ACTIVE;
+		key->dev_flow_state = MCTP_I2C_FLOW_STATE_INVALID;
+	}
+	spin_unlock_irqrestore(&key->lock, flags);
+
+	/* if we have changed state from active, the flow held a reference on
+	 * the lock; release that now.
+	 */
+	if (release)
+		mctp_i2c_unlock_nest(midev);
+}
+
 static void mctp_i2c_xmit(struct mctp_i2c_dev *midev, struct sk_buff *skb)
 {
 	struct net_device_stats *stats = &midev->ndev->stats;
@@ -500,6 +536,11 @@ static void mctp_i2c_xmit(struct mctp_i2c_dev *midev, struct sk_buff *skb)
 	case MCTP_I2C_TX_FLOW_EXISTING:
 		/* existing flow: we already have the lock; just tx */
 		rc = __i2c_transfer(midev->adapter, &msg, 1);
+
+		/* on tx errors, the flow can no longer be considered valid */
+		if (rc)
+			mctp_i2c_invalidate_tx_flow(midev, skb);
+
 		break;
 
 	case MCTP_I2C_TX_FLOW_INVALID:
-- 
cgit v1.2.3-58-ga151


From 2454075f8e2915cebbe52a1195631bc7efe2b7e1 Mon Sep 17 00:00:00 2001
From: Alan Maguire <alan.maguire@oracle.com>
Date: Fri, 12 Jul 2024 10:28:59 +0100
Subject: bpf: Eliminate remaining "make W=1" warnings in kernel/bpf/btf.o
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

As reported by Mirsad [1] we still see format warnings in kernel/bpf/btf.o
at W=1 warning level:

  CC      kernel/bpf/btf.o
./kernel/bpf/btf.c: In function ‘btf_type_seq_show_flags’:
./kernel/bpf/btf.c:7553:21: warning: assignment left-hand side might be a candidate for a format attribute [-Wsuggest-attribute=format]
 7553 |         sseq.showfn = btf_seq_show;
      |                     ^
./kernel/bpf/btf.c: In function ‘btf_type_snprintf_show’:
./kernel/bpf/btf.c:7604:31: warning: assignment left-hand side might be a candidate for a format attribute [-Wsuggest-attribute=format]
 7604 |         ssnprintf.show.showfn = btf_snprintf_show;
      |                               ^

Combined with CONFIG_WERROR=y these can halt the build.

The fix (annotating the structure field with __printf())
suggested by Mirsad resolves these. Apologies I missed this last time.
No other W=1 warnings were observed in kernel/bpf after this fix.

[1] https://lore.kernel.org/bpf/92c9d047-f058-400c-9c7d-81d4dc1ef71b@gmail.com/

Fixes: b3470da314fd ("bpf: annotate BTF show functions with __printf")
Reported-by: Mirsad Todorovac <mtodorovac69@gmail.com>
Suggested-by: Mirsad Todorovac <mtodorovac69@gmail.com>
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240712092859.1390960-1-alan.maguire@oracle.com
---
 kernel/bpf/btf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index d5019c4454d6..520f49f422fe 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -415,7 +415,7 @@ const char *btf_type_str(const struct btf_type *t)
 struct btf_show {
 	u64 flags;
 	void *target;	/* target of show operation (seq file, buffer) */
-	void (*showfn)(struct btf_show *show, const char *fmt, va_list args);
+	__printf(2, 0) void (*showfn)(struct btf_show *show, const char *fmt, va_list args);
 	const struct btf *btf;
 	/* below are used during iteration */
 	struct {
-- 
cgit v1.2.3-58-ga151


From 4a04b4f0de59dd5c621e78f15803ee0b0544eeb8 Mon Sep 17 00:00:00 2001
From: Shung-Hsi Yu <shung-hsi.yu@suse.com>
Date: Fri, 12 Jul 2024 16:01:24 +0800
Subject: bpf: fix overflow check in adjust_jmp_off()

adjust_jmp_off() incorrectly used the insn->imm field for all overflow check,
which is incorrect as that should only be done or the BPF_JMP32 | BPF_JA case,
not the general jump instruction case. Fix it by using insn->off for overflow
check in the general case.

Fixes: 5337ac4c9b80 ("bpf: Fix the corner case with may_goto and jump to the 1st insn.")
Signed-off-by: Shung-Hsi Yu <shung-hsi.yu@suse.com>
Link: https://lore.kernel.org/r/20240712080127.136608-2-shung-hsi.yu@suse.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index c0263fb5ca4b..cf2eb07ddf28 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -18852,7 +18852,7 @@ static int adjust_jmp_off(struct bpf_prog *prog, u32 tgt_idx, u32 delta)
 		} else {
 			if (i + 1 + insn->off != tgt_idx)
 				continue;
-			if (signed_add16_overflows(insn->imm, delta))
+			if (signed_add16_overflows(insn->off, delta))
 				return -ERANGE;
 			insn->off += delta;
 		}
-- 
cgit v1.2.3-58-ga151


From 28a4411076b254c67842348e3b25c2fb41a94cad Mon Sep 17 00:00:00 2001
From: Shung-Hsi Yu <shung-hsi.yu@suse.com>
Date: Fri, 12 Jul 2024 16:01:25 +0800
Subject: bpf: use check_add_overflow() to check for addition overflows

signed_add*_overflows() was added back when there was no overflow-check
helper. With the introduction of such helpers in commit f0907827a8a91
("compiler.h: enable builtin overflow checkers and add fallback code"), we
can drop signed_add*_overflows() in kernel/bpf/verifier.c and use the
generic check_add_overflow() instead.

This will make future refactoring easier, and takes advantage of
compiler-emitted hardware instructions that efficiently implement these
checks.

After the change GCC 13.3.0 generates cleaner assembly on x86_64:

	err = adjust_scalar_min_max_vals(env, insn, dst_reg, *src_reg);
   13625:	mov    0x28(%rbx),%r9  /*  r9 = src_reg->smin_value */
   13629:	mov    0x30(%rbx),%rcx /* rcx = src_reg->smax_value */
   ...
	if (check_add_overflow(*dst_smin, src_reg->smin_value, dst_smin) ||
   141c1:	mov    %r9,%rax
   141c4:	add    0x28(%r12),%rax
   141c9:	mov    %rax,0x28(%r12)
   141ce:	jo     146e4 <adjust_reg_min_max_vals+0x1294>
	    check_add_overflow(*dst_smax, src_reg->smax_value, dst_smax)) {
   141d4:	add    0x30(%r12),%rcx
   141d9:	mov    %rcx,0x30(%r12)
	if (check_add_overflow(*dst_smin, src_reg->smin_value, dst_smin) ||
   141de:	jo     146e4 <adjust_reg_min_max_vals+0x1294>
   ...
		*dst_smin = S64_MIN;
   146e4:	movabs $0x8000000000000000,%rax
   146ee:	mov    %rax,0x28(%r12)
		*dst_smax = S64_MAX;
   146f3:	sub    $0x1,%rax
   146f7:	mov    %rax,0x30(%r12)

Before the change it gives:

	s64 smin_val = src_reg->smin_value;
     675:	mov    0x28(%rsi),%r8
	s64 smax_val = src_reg->smax_value;
	u64 umin_val = src_reg->umin_value;
	u64 umax_val = src_reg->umax_value;
     679:	mov    %rdi,%rax /* rax = dst_reg */
	if (signed_add_overflows(dst_reg->smin_value, smin_val) ||
     67c:	mov    0x28(%rdi),%rdi /* rdi = dst_reg->smin_value */
	u64 umin_val = src_reg->umin_value;
     680:	mov    0x38(%rsi),%rdx
	u64 umax_val = src_reg->umax_value;
     684:	mov    0x40(%rsi),%rcx
	s64 res = (s64)((u64)a + (u64)b);
     688:	lea    (%r8,%rdi,1),%r9 /* r9 = dst_reg->smin_value + src_reg->smin_value */
	return res < a;
     68c:	cmp    %r9,%rdi
     68f:	setg   %r10b /* r10b = (dst_reg->smin_value + src_reg->smin_value) > dst_reg->smin_value */
	if (b < 0)
     693:	test   %r8,%r8
     696:	js     72b <scalar_min_max_add+0xbb>
	    signed_add_overflows(dst_reg->smax_value, smax_val)) {
		dst_reg->smin_value = S64_MIN;
		dst_reg->smax_value = S64_MAX;
     69c:	movabs $0x7fffffffffffffff,%rdi
	s64 smax_val = src_reg->smax_value;
     6a6:	mov    0x30(%rsi),%r8
		dst_reg->smin_value = S64_MIN;
     6aa:	00 00 00 	movabs $0x8000000000000000,%rsi
	if (signed_add_overflows(dst_reg->smin_value, smin_val) ||
     6b4:	test   %r10b,%r10b /* (dst_reg->smin_value + src_reg->smin_value) > dst_reg->smin_value ? goto 6cb */
     6b7:	jne    6cb <scalar_min_max_add+0x5b>
	    signed_add_overflows(dst_reg->smax_value, smax_val)) {
     6b9:	mov    0x30(%rax),%r10   /* r10 = dst_reg->smax_value */
	s64 res = (s64)((u64)a + (u64)b);
     6bd:	lea    (%r10,%r8,1),%r11 /* r11 = dst_reg->smax_value + src_reg->smax_value */
	if (b < 0)
     6c1:	test   %r8,%r8
     6c4:	js     71e <scalar_min_max_add+0xae>
	if (signed_add_overflows(dst_reg->smin_value, smin_val) ||
     6c6:	cmp    %r11,%r10 /* (dst_reg->smax_value + src_reg->smax_value) <= dst_reg->smax_value ? goto 723 */
     6c9:	jle    723 <scalar_min_max_add+0xb3>
	} else {
		dst_reg->smin_value += smin_val;
		dst_reg->smax_value += smax_val;
	}
     6cb:	mov    %rsi,0x28(%rax)
     ...
     6d5:	mov    %rdi,0x30(%rax)
     ...
	if (signed_add_overflows(dst_reg->smin_value, smin_val) ||
     71e:	cmp    %r11,%r10
     721:	jl     6cb <scalar_min_max_add+0x5b>
		dst_reg->smin_value += smin_val;
     723:	mov    %r9,%rsi
		dst_reg->smax_value += smax_val;
     726:	mov    %r11,%rdi
     729:	jmp    6cb <scalar_min_max_add+0x5b>
		return res > a;
     72b:	cmp    %r9,%rdi
     72e:	setl   %r10b
     732:	jmp    69c <scalar_min_max_add+0x2c>
     737:	nopw   0x0(%rax,%rax,1)

Note: unlike adjust_ptr_min_max_vals() and scalar*_min_max_add(), it is
necessary to introduce intermediate variable in adjust_jmp_off() to keep
the functional behavior unchanged. Without an intermediate variable
imm/off will be altered even on overflow.

Suggested-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Shung-Hsi Yu <shung-hsi.yu@suse.com>
Link: https://lore.kernel.org/r/20240712080127.136608-3-shung-hsi.yu@suse.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 114 +++++++++++++++-----------------------------------
 1 file changed, 34 insertions(+), 80 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index cf2eb07ddf28..0126c9c80c58 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -12726,36 +12726,6 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 	return 0;
 }
 
-static bool signed_add_overflows(s64 a, s64 b)
-{
-	/* Do the add in u64, where overflow is well-defined */
-	s64 res = (s64)((u64)a + (u64)b);
-
-	if (b < 0)
-		return res > a;
-	return res < a;
-}
-
-static bool signed_add32_overflows(s32 a, s32 b)
-{
-	/* Do the add in u32, where overflow is well-defined */
-	s32 res = (s32)((u32)a + (u32)b);
-
-	if (b < 0)
-		return res > a;
-	return res < a;
-}
-
-static bool signed_add16_overflows(s16 a, s16 b)
-{
-	/* Do the add in u16, where overflow is well-defined */
-	s16 res = (s16)((u16)a + (u16)b);
-
-	if (b < 0)
-		return res > a;
-	return res < a;
-}
-
 static bool signed_sub_overflows(s64 a, s64 b)
 {
 	/* Do the sub in u64, where overflow is well-defined */
@@ -13257,21 +13227,15 @@ static int adjust_ptr_min_max_vals(struct bpf_verifier_env *env,
 		 * added into the variable offset, and we copy the fixed offset
 		 * from ptr_reg.
 		 */
-		if (signed_add_overflows(smin_ptr, smin_val) ||
-		    signed_add_overflows(smax_ptr, smax_val)) {
+		if (check_add_overflow(smin_ptr, smin_val, &dst_reg->smin_value) ||
+		    check_add_overflow(smax_ptr, smax_val, &dst_reg->smax_value)) {
 			dst_reg->smin_value = S64_MIN;
 			dst_reg->smax_value = S64_MAX;
-		} else {
-			dst_reg->smin_value = smin_ptr + smin_val;
-			dst_reg->smax_value = smax_ptr + smax_val;
 		}
-		if (umin_ptr + umin_val < umin_ptr ||
-		    umax_ptr + umax_val < umax_ptr) {
+		if (check_add_overflow(umin_ptr, umin_val, &dst_reg->umin_value) ||
+		    check_add_overflow(umax_ptr, umax_val, &dst_reg->umax_value)) {
 			dst_reg->umin_value = 0;
 			dst_reg->umax_value = U64_MAX;
-		} else {
-			dst_reg->umin_value = umin_ptr + umin_val;
-			dst_reg->umax_value = umax_ptr + umax_val;
 		}
 		dst_reg->var_off = tnum_add(ptr_reg->var_off, off_reg->var_off);
 		dst_reg->off = ptr_reg->off;
@@ -13374,52 +13338,40 @@ static int adjust_ptr_min_max_vals(struct bpf_verifier_env *env,
 static void scalar32_min_max_add(struct bpf_reg_state *dst_reg,
 				 struct bpf_reg_state *src_reg)
 {
-	s32 smin_val = src_reg->s32_min_value;
-	s32 smax_val = src_reg->s32_max_value;
-	u32 umin_val = src_reg->u32_min_value;
-	u32 umax_val = src_reg->u32_max_value;
+	s32 *dst_smin = &dst_reg->s32_min_value;
+	s32 *dst_smax = &dst_reg->s32_max_value;
+	u32 *dst_umin = &dst_reg->u32_min_value;
+	u32 *dst_umax = &dst_reg->u32_max_value;
 
-	if (signed_add32_overflows(dst_reg->s32_min_value, smin_val) ||
-	    signed_add32_overflows(dst_reg->s32_max_value, smax_val)) {
-		dst_reg->s32_min_value = S32_MIN;
-		dst_reg->s32_max_value = S32_MAX;
-	} else {
-		dst_reg->s32_min_value += smin_val;
-		dst_reg->s32_max_value += smax_val;
+	if (check_add_overflow(*dst_smin, src_reg->s32_min_value, dst_smin) ||
+	    check_add_overflow(*dst_smax, src_reg->s32_max_value, dst_smax)) {
+		*dst_smin = S32_MIN;
+		*dst_smax = S32_MAX;
 	}
-	if (dst_reg->u32_min_value + umin_val < umin_val ||
-	    dst_reg->u32_max_value + umax_val < umax_val) {
-		dst_reg->u32_min_value = 0;
-		dst_reg->u32_max_value = U32_MAX;
-	} else {
-		dst_reg->u32_min_value += umin_val;
-		dst_reg->u32_max_value += umax_val;
+	if (check_add_overflow(*dst_umin, src_reg->u32_min_value, dst_umin) ||
+	    check_add_overflow(*dst_umax, src_reg->u32_max_value, dst_umax)) {
+		*dst_umin = 0;
+		*dst_umax = U32_MAX;
 	}
 }
 
 static void scalar_min_max_add(struct bpf_reg_state *dst_reg,
 			       struct bpf_reg_state *src_reg)
 {
-	s64 smin_val = src_reg->smin_value;
-	s64 smax_val = src_reg->smax_value;
-	u64 umin_val = src_reg->umin_value;
-	u64 umax_val = src_reg->umax_value;
+	s64 *dst_smin = &dst_reg->smin_value;
+	s64 *dst_smax = &dst_reg->smax_value;
+	u64 *dst_umin = &dst_reg->umin_value;
+	u64 *dst_umax = &dst_reg->umax_value;
 
-	if (signed_add_overflows(dst_reg->smin_value, smin_val) ||
-	    signed_add_overflows(dst_reg->smax_value, smax_val)) {
-		dst_reg->smin_value = S64_MIN;
-		dst_reg->smax_value = S64_MAX;
-	} else {
-		dst_reg->smin_value += smin_val;
-		dst_reg->smax_value += smax_val;
+	if (check_add_overflow(*dst_smin, src_reg->smin_value, dst_smin) ||
+	    check_add_overflow(*dst_smax, src_reg->smax_value, dst_smax)) {
+		*dst_smin = S64_MIN;
+		*dst_smax = S64_MAX;
 	}
-	if (dst_reg->umin_value + umin_val < umin_val ||
-	    dst_reg->umax_value + umax_val < umax_val) {
-		dst_reg->umin_value = 0;
-		dst_reg->umax_value = U64_MAX;
-	} else {
-		dst_reg->umin_value += umin_val;
-		dst_reg->umax_value += umax_val;
+	if (check_add_overflow(*dst_umin, src_reg->umin_value, dst_umin) ||
+	    check_add_overflow(*dst_umax, src_reg->umax_value, dst_umax)) {
+		*dst_umin = 0;
+		*dst_umax = U64_MAX;
 	}
 }
 
@@ -18835,6 +18787,8 @@ static int adjust_jmp_off(struct bpf_prog *prog, u32 tgt_idx, u32 delta)
 {
 	struct bpf_insn *insn = prog->insnsi;
 	u32 insn_cnt = prog->len, i;
+	s32 imm;
+	s16 off;
 
 	for (i = 0; i < insn_cnt; i++, insn++) {
 		u8 code = insn->code;
@@ -18846,15 +18800,15 @@ static int adjust_jmp_off(struct bpf_prog *prog, u32 tgt_idx, u32 delta)
 		if (insn->code == (BPF_JMP32 | BPF_JA)) {
 			if (i + 1 + insn->imm != tgt_idx)
 				continue;
-			if (signed_add32_overflows(insn->imm, delta))
+			if (check_add_overflow(insn->imm, delta, &imm))
 				return -ERANGE;
-			insn->imm += delta;
+			insn->imm = imm;
 		} else {
 			if (i + 1 + insn->off != tgt_idx)
 				continue;
-			if (signed_add16_overflows(insn->off, delta))
+			if (check_add_overflow(insn->off, delta, &off))
 				return -ERANGE;
-			insn->off += delta;
+			insn->off = off;
 		}
 	}
 	return 0;
-- 
cgit v1.2.3-58-ga151


From deac5871eb0751454cb80b3ff6b69e42a6c1bab2 Mon Sep 17 00:00:00 2001
From: Shung-Hsi Yu <shung-hsi.yu@suse.com>
Date: Fri, 12 Jul 2024 16:01:26 +0800
Subject: bpf: use check_sub_overflow() to check for subtraction overflows

Similar to previous patch that drops signed_add*_overflows() and uses
(compiler) builtin-based check_add_overflow(), do the same for
signed_sub*_overflows() and replace them with the generic
check_sub_overflow() to make future refactoring easier and have the
checks implemented more efficiently.

Unsigned overflow check for subtraction does not use helpers and are
simple enough already, so they're left untouched.

After the change GCC 13.3.0 generates cleaner assembly on x86_64:

	if (check_sub_overflow(*dst_smin, src_reg->smax_value, dst_smin) ||
   139bf:	mov    0x28(%r12),%rax
   139c4:	mov    %edx,0x54(%r12)
   139c9:	sub    %r11,%rax
   139cc:	mov    %rax,0x28(%r12)
   139d1:	jo     14627 <adjust_reg_min_max_vals+0x1237>
	    check_sub_overflow(*dst_smax, src_reg->smin_value, dst_smax)) {
   139d7:	mov    0x30(%r12),%rax
   139dc:	sub    %r9,%rax
   139df:	mov    %rax,0x30(%r12)
	if (check_sub_overflow(*dst_smin, src_reg->smax_value, dst_smin) ||
   139e4:	jo     14627 <adjust_reg_min_max_vals+0x1237>
   ...
		*dst_smin = S64_MIN;
   14627:	movabs $0x8000000000000000,%rax
   14631:	mov    %rax,0x28(%r12)
		*dst_smax = S64_MAX;
   14636:	sub    $0x1,%rax
   1463a:	mov    %rax,0x30(%r12)

Before the change it gives:

	if (signed_sub_overflows(dst_reg->smin_value, smax_val) ||
   13a50:	mov    0x28(%r12),%rdi
   13a55:	mov    %edx,0x54(%r12)
		dst_reg->smax_value = S64_MAX;
   13a5a:	movabs $0x7fffffffffffffff,%rdx
   13a64:	mov    %eax,0x50(%r12)
		dst_reg->smin_value = S64_MIN;
   13a69:	movabs $0x8000000000000000,%rax
	s64 res = (s64)((u64)a - (u64)b);
   13a73:	mov    %rdi,%rsi
   13a76:	sub    %rcx,%rsi
	if (b < 0)
   13a79:	test   %rcx,%rcx
   13a7c:	js     145ea <adjust_reg_min_max_vals+0x119a>
	if (signed_sub_overflows(dst_reg->smin_value, smax_val) ||
   13a82:	cmp    %rsi,%rdi
   13a85:	jl     13ac7 <adjust_reg_min_max_vals+0x677>
	    signed_sub_overflows(dst_reg->smax_value, smin_val)) {
   13a87:	mov    0x30(%r12),%r8
	s64 res = (s64)((u64)a - (u64)b);
   13a8c:	mov    %r8,%rax
   13a8f:	sub    %r9,%rax
	return res > a;
   13a92:	cmp    %rax,%r8
   13a95:	setl   %sil
	if (b < 0)
   13a99:	test   %r9,%r9
   13a9c:	js     147d1 <adjust_reg_min_max_vals+0x1381>
		dst_reg->smax_value = S64_MAX;
   13aa2:	movabs $0x7fffffffffffffff,%rdx
		dst_reg->smin_value = S64_MIN;
   13aac:	movabs $0x8000000000000000,%rax
	if (signed_sub_overflows(dst_reg->smin_value, smax_val) ||
   13ab6:	test   %sil,%sil
   13ab9:	jne    13ac7 <adjust_reg_min_max_vals+0x677>
		dst_reg->smin_value -= smax_val;
   13abb:	mov    %rdi,%rax
		dst_reg->smax_value -= smin_val;
   13abe:	mov    %r8,%rdx
		dst_reg->smin_value -= smax_val;
   13ac1:	sub    %rcx,%rax
		dst_reg->smax_value -= smin_val;
   13ac4:	sub    %r9,%rdx
   13ac7:	mov    %rax,0x28(%r12)
   ...
   13ad1:	mov    %rdx,0x30(%r12)
   ...
	if (signed_sub_overflows(dst_reg->smin_value, smax_val) ||
   145ea:	cmp    %rsi,%rdi
   145ed:	jg     13ac7 <adjust_reg_min_max_vals+0x677>
   145f3:	jmp    13a87 <adjust_reg_min_max_vals+0x637>

Suggested-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Shung-Hsi Yu <shung-hsi.yu@suse.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20240712080127.136608-4-shung-hsi.yu@suse.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
 kernel/bpf/verifier.c | 57 +++++++++++++--------------------------------------
 1 file changed, 14 insertions(+), 43 deletions(-)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 0126c9c80c58..8da132a1ef28 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -12726,26 +12726,6 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn,
 	return 0;
 }
 
-static bool signed_sub_overflows(s64 a, s64 b)
-{
-	/* Do the sub in u64, where overflow is well-defined */
-	s64 res = (s64)((u64)a - (u64)b);
-
-	if (b < 0)
-		return res < a;
-	return res > a;
-}
-
-static bool signed_sub32_overflows(s32 a, s32 b)
-{
-	/* Do the sub in u32, where overflow is well-defined */
-	s32 res = (s32)((u32)a - (u32)b);
-
-	if (b < 0)
-		return res < a;
-	return res > a;
-}
-
 static bool check_reg_sane_offset(struct bpf_verifier_env *env,
 				  const struct bpf_reg_state *reg,
 				  enum bpf_reg_type type)
@@ -13278,14 +13258,11 @@ static int adjust_ptr_min_max_vals(struct bpf_verifier_env *env,
 		/* A new variable offset is created.  If the subtrahend is known
 		 * nonnegative, then any reg->range we had before is still good.
 		 */
-		if (signed_sub_overflows(smin_ptr, smax_val) ||
-		    signed_sub_overflows(smax_ptr, smin_val)) {
+		if (check_sub_overflow(smin_ptr, smax_val, &dst_reg->smin_value) ||
+		    check_sub_overflow(smax_ptr, smin_val, &dst_reg->smax_value)) {
 			/* Overflow possible, we know nothing */
 			dst_reg->smin_value = S64_MIN;
 			dst_reg->smax_value = S64_MAX;
-		} else {
-			dst_reg->smin_value = smin_ptr - smax_val;
-			dst_reg->smax_value = smax_ptr - smin_val;
 		}
 		if (umin_ptr < umax_val) {
 			/* Overflow possible, we know nothing */
@@ -13378,19 +13355,16 @@ static void scalar_min_max_add(struct bpf_reg_state *dst_reg,
 static void scalar32_min_max_sub(struct bpf_reg_state *dst_reg,
 				 struct bpf_reg_state *src_reg)
 {
-	s32 smin_val = src_reg->s32_min_value;
-	s32 smax_val = src_reg->s32_max_value;
+	s32 *dst_smin = &dst_reg->s32_min_value;
+	s32 *dst_smax = &dst_reg->s32_max_value;
 	u32 umin_val = src_reg->u32_min_value;
 	u32 umax_val = src_reg->u32_max_value;
 
-	if (signed_sub32_overflows(dst_reg->s32_min_value, smax_val) ||
-	    signed_sub32_overflows(dst_reg->s32_max_value, smin_val)) {
+	if (check_sub_overflow(*dst_smin, src_reg->s32_max_value, dst_smin) ||
+	    check_sub_overflow(*dst_smax, src_reg->s32_min_value, dst_smax)) {
 		/* Overflow possible, we know nothing */
-		dst_reg->s32_min_value = S32_MIN;
-		dst_reg->s32_max_value = S32_MAX;
-	} else {
-		dst_reg->s32_min_value -= smax_val;
-		dst_reg->s32_max_value -= smin_val;
+		*dst_smin = S32_MIN;
+		*dst_smax = S32_MAX;
 	}
 	if (dst_reg->u32_min_value < umax_val) {
 		/* Overflow possible, we know nothing */
@@ -13406,19 +13380,16 @@ static void scalar32_min_max_sub(struct bpf_reg_state *dst_reg,
 static void scalar_min_max_sub(struct bpf_reg_state *dst_reg,
 			       struct bpf_reg_state *src_reg)
 {
-	s64 smin_val = src_reg->smin_value;
-	s64 smax_val = src_reg->smax_value;
+	s64 *dst_smin = &dst_reg->smin_value;
+	s64 *dst_smax = &dst_reg->smax_value;
 	u64 umin_val = src_reg->umin_value;
 	u64 umax_val = src_reg->umax_value;
 
-	if (signed_sub_overflows(dst_reg->smin_value, smax_val) ||
-	    signed_sub_overflows(dst_reg->smax_value, smin_val)) {
+	if (check_sub_overflow(*dst_smin, src_reg->smax_value, dst_smin) ||
+	    check_sub_overflow(*dst_smax, src_reg->smin_value, dst_smax)) {
 		/* Overflow possible, we know nothing */
-		dst_reg->smin_value = S64_MIN;
-		dst_reg->smax_value = S64_MAX;
-	} else {
-		dst_reg->smin_value -= smax_val;
-		dst_reg->smax_value -= smin_val;
+		*dst_smin = S64_MIN;
+		*dst_smax = S64_MAX;
 	}
 	if (dst_reg->umin_value < umax_val) {
 		/* Overflow possible, we know nothing */
-- 
cgit v1.2.3-58-ga151


From 517125f6749402e579f715519147145944f12ad9 Mon Sep 17 00:00:00 2001
From: Daniel Borkmann <daniel@iogearbox.net>
Date: Fri, 12 Jul 2024 18:12:30 +0200
Subject: selftests/bpf: DENYLIST.aarch64: Skip fexit_sleep again

Revert commit 90dc946059b7 ("selftests/bpf: DENYLIST.aarch64: Remove
fexit_sleep") again. The fix in 19d3c179a377 ("bpf, arm64: Fix trampoline
for BPF_TRAMP_F_CALL_ORIG") does not address all of the issues and BPF
CI is still hanging and timing out:

   https://github.com/kernel-patches/bpf/actions/runs/9905842936/job/27366435436

   [...]
   #89/11   fexit_bpf2bpf/func_replace_global_func:OK
   #89/12   fexit_bpf2bpf/fentry_to_cgroup_bpf:OK
   #89/13   fexit_bpf2bpf/func_replace_progmap:OK
   #89      fexit_bpf2bpf:OK
   Error: The operation was canceled.

Thus more investigation work & fixing is needed before the test can be put
in place again.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Puranjay Mohan <puranjay@kernel.org>
Link: https://lore.kernel.org/bpf/20240705145009.32340-1-puranjay@kernel.org
---
 tools/testing/selftests/bpf/DENYLIST.aarch64 | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/testing/selftests/bpf/DENYLIST.aarch64 b/tools/testing/selftests/bpf/DENYLIST.aarch64
index 901349da680f..3c7c3e79aa93 100644
--- a/tools/testing/selftests/bpf/DENYLIST.aarch64
+++ b/tools/testing/selftests/bpf/DENYLIST.aarch64
@@ -1,5 +1,6 @@
 bpf_cookie/multi_kprobe_attach_api               # kprobe_multi_link_api_subtest:FAIL:fentry_raw_skel_load unexpected error: -3
 bpf_cookie/multi_kprobe_link_api                 # kprobe_multi_link_api_subtest:FAIL:fentry_raw_skel_load unexpected error: -3
+fexit_sleep                                      # The test never returns. The remaining tests cannot start.
 kprobe_multi_bench_attach                        # needs CONFIG_FPROBE
 kprobe_multi_test                                # needs CONFIG_FPROBE
 module_attach                                    # prog 'kprobe_multi': failed to auto-attach: -95
-- 
cgit v1.2.3-58-ga151


From f7866c35873377313ff94398f17d425b28b71de1 Mon Sep 17 00:00:00 2001
From: Tengda Wu <wutengda@huaweicloud.com>
Date: Thu, 11 Jul 2024 22:58:18 +0800
Subject: bpf: Fix null pointer dereference in resolve_prog_type() for
 BPF_PROG_TYPE_EXT

When loading a EXT program without specifying `attr->attach_prog_fd`,
the `prog->aux->dst_prog` will be null. At this time, calling
resolve_prog_type() anywhere will result in a null pointer dereference.

Example stack trace:

[    8.107863] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000004
[    8.108262] Mem abort info:
[    8.108384]   ESR = 0x0000000096000004
[    8.108547]   EC = 0x25: DABT (current EL), IL = 32 bits
[    8.108722]   SET = 0, FnV = 0
[    8.108827]   EA = 0, S1PTW = 0
[    8.108939]   FSC = 0x04: level 0 translation fault
[    8.109102] Data abort info:
[    8.109203]   ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
[    8.109399]   CM = 0, WnR = 0, TnD = 0, TagAccess = 0
[    8.109614]   GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[    8.109836] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000101354000
[    8.110011] [0000000000000004] pgd=0000000000000000, p4d=0000000000000000
[    8.112624] Internal error: Oops: 0000000096000004 [#1] PREEMPT SMP
[    8.112783] Modules linked in:
[    8.113120] CPU: 0 PID: 99 Comm: may_access_dire Not tainted 6.10.0-rc3-next-20240613-dirty #1
[    8.113230] Hardware name: linux,dummy-virt (DT)
[    8.113390] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[    8.113429] pc : may_access_direct_pkt_data+0x24/0xa0
[    8.113746] lr : add_subprog_and_kfunc+0x634/0x8e8
[    8.113798] sp : ffff80008283b9f0
[    8.113813] x29: ffff80008283b9f0 x28: ffff800082795048 x27: 0000000000000001
[    8.113881] x26: ffff0000c0bb2600 x25: 0000000000000000 x24: 0000000000000000
[    8.113897] x23: ffff0000c1134000 x22: 000000000001864f x21: ffff0000c1138000
[    8.113912] x20: 0000000000000001 x19: ffff0000c12b8000 x18: ffffffffffffffff
[    8.113929] x17: 0000000000000000 x16: 0000000000000000 x15: 0720072007200720
[    8.113944] x14: 0720072007200720 x13: 0720072007200720 x12: 0720072007200720
[    8.113958] x11: 0720072007200720 x10: 0000000000f9fca4 x9 : ffff80008021f4e4
[    8.113991] x8 : 0101010101010101 x7 : 746f72705f6d656d x6 : 000000001e0e0f5f
[    8.114006] x5 : 000000000001864f x4 : ffff0000c12b8000 x3 : 000000000000001c
[    8.114020] x2 : 0000000000000002 x1 : 0000000000000000 x0 : 0000000000000000
[    8.114126] Call trace:
[    8.114159]  may_access_direct_pkt_data+0x24/0xa0
[    8.114202]  bpf_check+0x3bc/0x28c0
[    8.114214]  bpf_prog_load+0x658/0xa58
[    8.114227]  __sys_bpf+0xc50/0x2250
[    8.114240]  __arm64_sys_bpf+0x28/0x40
[    8.114254]  invoke_syscall.constprop.0+0x54/0xf0
[    8.114273]  do_el0_svc+0x4c/0xd8
[    8.114289]  el0_svc+0x3c/0x140
[    8.114305]  el0t_64_sync_handler+0x134/0x150
[    8.114331]  el0t_64_sync+0x168/0x170
[    8.114477] Code: 7100707f 54000081 f9401c00 f9403800 (b9400403)
[    8.118672] ---[ end trace 0000000000000000 ]---

One way to fix it is by forcing `attach_prog_fd` non-empty when
bpf_prog_load(). But this will lead to `libbpf_probe_bpf_prog_type`
API broken which use verifier log to probe prog type and will log
nothing if we reject invalid EXT prog before bpf_check().

Another way is by adding null check in resolve_prog_type().

The issue was introduced by commit 4a9c7bbe2ed4 ("bpf: Resolve to
prog->aux->dst_prog->type only for BPF_PROG_TYPE_EXT") which wanted
to correct type resolution for BPF_PROG_TYPE_TRACING programs. Before
that, the type resolution of BPF_PROG_TYPE_EXT prog actually follows
the logic below:

  prog->aux->dst_prog ? prog->aux->dst_prog->type : prog->type;

It implies that when EXT program is not yet attached to `dst_prog`,
the prog type should be EXT itself. This code worked fine in the past.
So just keep using it.

Fix this by returning `prog->type` for BPF_PROG_TYPE_EXT if `dst_prog`
is not present in resolve_prog_type().

Fixes: 4a9c7bbe2ed4 ("bpf: Resolve to prog->aux->dst_prog->type only for BPF_PROG_TYPE_EXT")
Signed-off-by: Tengda Wu <wutengda@huaweicloud.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Cc: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20240711145819.254178-2-wutengda@huaweicloud.com
---
 include/linux/bpf_verifier.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index e98ba5a5cf79..6503c85b10a3 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -856,7 +856,7 @@ static inline u32 type_flag(u32 type)
 /* only use after check_attach_btf_id() */
 static inline enum bpf_prog_type resolve_prog_type(const struct bpf_prog *prog)
 {
-	return prog->type == BPF_PROG_TYPE_EXT ?
+	return (prog->type == BPF_PROG_TYPE_EXT && prog->aux->dst_prog) ?
 		prog->aux->dst_prog->type : prog->type;
 }
 
-- 
cgit v1.2.3-58-ga151


From e435b043d89a267bd6eb3d5650d2319805d7924a Mon Sep 17 00:00:00 2001
From: Tengda Wu <wutengda@huaweicloud.com>
Date: Thu, 11 Jul 2024 22:58:19 +0800
Subject: selftests/bpf: Test for null-pointer-deref bugfix in
 resolve_prog_type()

This test verifies that resolve_prog_type() works as expected when
`attach_prog_fd` is not passed in.

`prog->aux->dst_prog` in resolve_prog_type() is assigned by
`attach_prog_fd`, and would be NULL if `attach_prog_fd` is not provided.

Loading EXT prog with bpf_dynptr_from_skb() kfunc call in this way will
lead to null-pointer-deref.

Verify that the null-pointer-deref bug in resolve_prog_type() is fixed.

Signed-off-by: Tengda Wu <wutengda@huaweicloud.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240711145819.254178-3-wutengda@huaweicloud.com
---
 tools/testing/selftests/bpf/verifier/calls.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/tools/testing/selftests/bpf/verifier/calls.c b/tools/testing/selftests/bpf/verifier/calls.c
index d76ef2018859..d0cdd156cd55 100644
--- a/tools/testing/selftests/bpf/verifier/calls.c
+++ b/tools/testing/selftests/bpf/verifier/calls.c
@@ -275,6 +275,19 @@
 	.result_unpriv = REJECT,
 	.result = ACCEPT,
 },
+{
+	"calls: invalid kfunc call: must provide (attach_prog_fd, btf_id) pair when freplace",
+	.insns = {
+	BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, 0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.prog_type = BPF_PROG_TYPE_EXT,
+	.result = REJECT,
+	.errstr = "Tracing programs must provide btf_id",
+	.fixup_kfunc_btf_id = {
+		{ "bpf_dynptr_from_skb", 0 },
+	},
+},
 {
 	"calls: basic sanity",
 	.insns = {
-- 
cgit v1.2.3-58-ga151


From d69ba6bbaf1f606ac354e925571a54d025e32aae Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:07:03 -0700
Subject: net: ethtool: let drivers remove lost RSS contexts

RSS contexts may get lost from a device, in various extreme circumstances.
Specifically if the firmware leaks resources and resets, or crashes and
either recovers in partially working state or the crash causes a
different FW version to run - creating the context again may fail.

Drivers should do their absolute best to prevent this from happening.
When it does, however, telling user that a context exists, when it can't
possibly be used any more is counter productive. Add a helper for
drivers to discard contexts. Print an error, in the future netlink
notification will also be sent.

More robust approaches were proposed, like keeping the contexts
but marking them as "dead" (but possibly resurrected by next reset).
That may be better but it's unclear at this stage whether the
effort is worth the benefits.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20240711220713.283778-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/ethtool.h |  2 ++
 net/ethtool/common.c    | 14 ++++++++++++++
 2 files changed, 16 insertions(+)

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index e213b5508da6..89da0254ccd4 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -210,6 +210,8 @@ static inline size_t ethtool_rxfh_context_size(u32 indir_size, u32 key_size,
 	return struct_size_t(struct ethtool_rxfh_context, data, flex_len);
 }
 
+void ethtool_rxfh_context_lost(struct net_device *dev, u32 context_id);
+
 /* declare a link mode bitmap */
 #define __ETHTOOL_DECLARE_LINK_MODE_MASK(name)		\
 	DECLARE_BITMAP(name, __ETHTOOL_LINK_MODE_MASK_NBITS)
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 7bda9600efcf..67d06cd002a5 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -741,3 +741,17 @@ ethtool_forced_speed_maps_init(struct ethtool_forced_speed_map *maps, u32 size)
 	}
 }
 EXPORT_SYMBOL_GPL(ethtool_forced_speed_maps_init);
+
+void ethtool_rxfh_context_lost(struct net_device *dev, u32 context_id)
+{
+	struct ethtool_rxfh_context *ctx;
+
+	WARN_ONCE(!rtnl_is_locked() &&
+		  !lockdep_is_held_type(&dev->ethtool->rss_lock, -1),
+		  "RSS context lock assertion failed\n");
+
+	netdev_err(dev, "device error, RSS context %d lost\n", context_id);
+	ctx = xa_erase(&dev->ethtool->rss_ctx, context_id);
+	kfree(ctx);
+}
+EXPORT_SYMBOL(ethtool_rxfh_context_lost);
-- 
cgit v1.2.3-58-ga151


From 28c8757a792bbbc76407777bd0303862daa75057 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:07:04 -0700
Subject: net: ethtool: let drivers declare max size of RSS indir table and key

Some drivers (bnxt but I think also mlx5 from ML discussions) change
the size of the indirection table depending on the number of Rx rings.
Decouple the max table size from the size of the currently used table,
so that we can reserve space in the context for table growth.

Static members in ethtool_ops are good enough for now, we can add
callbacks to read the max size more dynamically if someone needs
that.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20240711220713.283778-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/ethtool.h | 20 ++++++++------------
 net/ethtool/ioctl.c     | 46 ++++++++++++++++++++++++++++++++++++----------
 2 files changed, 44 insertions(+), 22 deletions(-)

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 89da0254ccd4..a1ee76936f53 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -181,6 +181,7 @@ struct ethtool_rxfh_context {
 	/* private: driver private data, indirection table, and hash key are
 	 * stored sequentially in @data area.  Use below helpers to access.
 	 */
+	u32 key_off;
 	u8 data[] __aligned(sizeof(void *));
 };
 
@@ -196,18 +197,7 @@ static inline u32 *ethtool_rxfh_context_indir(struct ethtool_rxfh_context *ctx)
 
 static inline u8 *ethtool_rxfh_context_key(struct ethtool_rxfh_context *ctx)
 {
-	return (u8 *)(ethtool_rxfh_context_indir(ctx) + ctx->indir_size);
-}
-
-static inline size_t ethtool_rxfh_context_size(u32 indir_size, u32 key_size,
-					       u16 priv_size)
-{
-	size_t indir_bytes = array_size(indir_size, sizeof(u32));
-	size_t flex_len;
-
-	flex_len = size_add(size_add(indir_bytes, key_size),
-			    ALIGN(priv_size, sizeof(u32)));
-	return struct_size_t(struct ethtool_rxfh_context, data, flex_len);
+	return &ctx->data[ctx->key_off];
 }
 
 void ethtool_rxfh_context_lost(struct net_device *dev, u32 context_id);
@@ -723,6 +713,10 @@ struct ethtool_rxfh_param {
  *	contexts.
  * @cap_rss_sym_xor_supported: indicates if the driver supports symmetric-xor
  *	RSS.
+ * @rxfh_indir_space: max size of RSS indirection tables, if indirection table
+ *	size as returned by @get_rxfh_indir_size may change during lifetime
+ *	of the device. Leave as 0 if the table size is constant.
+ * @rxfh_key_space: same as @rxfh_indir_space, but for the key.
  * @rxfh_priv_size: size of the driver private data area the core should
  *	allocate for an RSS context (in &struct ethtool_rxfh_context).
  * @rxfh_max_context_id: maximum (exclusive) supported RSS context ID.  If this
@@ -940,6 +934,8 @@ struct ethtool_ops {
 	u32     cap_link_lanes_supported:1;
 	u32     cap_rss_ctx_supported:1;
 	u32	cap_rss_sym_xor_supported:1;
+	u32	rxfh_indir_space;
+	u16	rxfh_key_space;
 	u16	rxfh_priv_size;
 	u32	rxfh_max_context_id;
 	u32	supported_coalesce_params;
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index 615812ff8974..0732710a4836 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -1290,6 +1290,40 @@ out:
 	return ret;
 }
 
+static struct ethtool_rxfh_context *
+ethtool_rxfh_ctx_alloc(const struct ethtool_ops *ops,
+		       u32 indir_size, u32 key_size)
+{
+	size_t indir_bytes, flex_len, key_off, size;
+	struct ethtool_rxfh_context *ctx;
+	u32 priv_bytes, indir_max;
+	u16 key_max;
+
+	key_max = max(key_size, ops->rxfh_key_space);
+	indir_max = max(indir_size, ops->rxfh_indir_space);
+
+	priv_bytes = ALIGN(ops->rxfh_priv_size, sizeof(u32));
+	indir_bytes = array_size(indir_max, sizeof(u32));
+
+	key_off = size_add(priv_bytes, indir_bytes);
+	flex_len = size_add(key_off, key_max);
+	size = struct_size_t(struct ethtool_rxfh_context, data, flex_len);
+
+	ctx = kzalloc(size, GFP_KERNEL_ACCOUNT);
+	if (!ctx)
+		return NULL;
+
+	ctx->indir_size = indir_size;
+	ctx->key_size = key_size;
+	ctx->key_off = key_off;
+	ctx->priv_size = ops->rxfh_priv_size;
+
+	ctx->hfunc = ETH_RSS_HASH_NO_CHANGE;
+	ctx->input_xfrm = RXH_XFRM_NO_CHANGE;
+
+	return ctx;
+}
+
 static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 					       void __user *useraddr)
 {
@@ -1406,20 +1440,12 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev,
 			ret = -EINVAL;
 			goto out;
 		}
-		ctx = kzalloc(ethtool_rxfh_context_size(dev_indir_size,
-							dev_key_size,
-							ops->rxfh_priv_size),
-			      GFP_KERNEL_ACCOUNT);
+		ctx = ethtool_rxfh_ctx_alloc(ops, dev_indir_size, dev_key_size);
 		if (!ctx) {
 			ret = -ENOMEM;
 			goto out;
 		}
-		ctx->indir_size = dev_indir_size;
-		ctx->key_size = dev_key_size;
-		ctx->priv_size = ops->rxfh_priv_size;
-		/* Initialise to an empty context */
-		ctx->hfunc = ETH_RSS_HASH_NO_CHANGE;
-		ctx->input_xfrm = RXH_XFRM_NO_CHANGE;
+
 		if (ops->create_rxfh_context) {
 			u32 limit = ops->rxfh_max_context_id ?: U32_MAX;
 			u32 ctx_id;
-- 
cgit v1.2.3-58-ga151


From 667ac333dbb7e265b3f5bc4bc94e236f64682c86 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:07:05 -0700
Subject: eth: bnxt: allow deleting RSS contexts when the device is down

Contexts get deleted from FW when the device is down, but they
are kept in SW and re-added back on open. bnxt_set_rxfh_context()
apparently does not want to deal with complexity of dealing with
both the device down and device up cases. This is perhaps acceptable
for creating new contexts, but not being able to delete contexts
makes core-driven cleanups messy. Specifically with the new RSS
API core will try to delete contexts automatically after bringing
the device down.

Support the delete-while-down case. Skip the FW logic and delete
just the driver state.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20240711220713.283778-4-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         | 10 ++++++----
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c |  6 ++++--
 2 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 80fce0aaad66..e8965cf743fc 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -10210,10 +10210,12 @@ void bnxt_del_one_rss_ctx(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 	struct bnxt_ntuple_filter *ntp_fltr;
 	int i;
 
-	bnxt_hwrm_vnic_free_one(bp, &rss_ctx->vnic);
-	for (i = 0; i < BNXT_MAX_CTX_PER_VNIC; i++) {
-		if (vnic->fw_rss_cos_lb_ctx[i] != INVALID_HW_RING_ID)
-			bnxt_hwrm_vnic_ctx_free_one(bp, vnic, i);
+	if (netif_running(bp->dev)) {
+		bnxt_hwrm_vnic_free_one(bp, &rss_ctx->vnic);
+		for (i = 0; i < BNXT_MAX_CTX_PER_VNIC; i++) {
+			if (vnic->fw_rss_cos_lb_ctx[i] != INVALID_HW_RING_ID)
+				bnxt_hwrm_vnic_ctx_free_one(bp, vnic, i);
+		}
 	}
 	if (!all)
 		return;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 4d53ec7adc61..846a19b5f58d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -1874,6 +1874,7 @@ static int bnxt_set_rxfh_context(struct bnxt *bp,
 	struct bnxt_rss_ctx *rss_ctx;
 	struct bnxt_vnic_info *vnic;
 	bool modify = false;
+	bool delete;
 	int bit_id;
 	int rc;
 
@@ -1882,7 +1883,8 @@ static int bnxt_set_rxfh_context(struct bnxt *bp,
 		return -EOPNOTSUPP;
 	}
 
-	if (!netif_running(bp->dev)) {
+	delete = *rss_context != ETH_RXFH_CONTEXT_ALLOC && rxfh->rss_delete;
+	if (!netif_running(bp->dev) && !delete) {
 		NL_SET_ERR_MSG_MOD(extack, "Unable to set RSS contexts when interface is down");
 		return -EAGAIN;
 	}
@@ -1894,7 +1896,7 @@ static int bnxt_set_rxfh_context(struct bnxt *bp,
 					       *rss_context);
 			return -EINVAL;
 		}
-		if (*rss_context && rxfh->rss_delete) {
+		if (delete) {
 			bnxt_del_one_rss_ctx(bp, rss_ctx, true);
 			return 0;
 		}
-- 
cgit v1.2.3-58-ga151


From 5c466b4d4e75c09a5a3e8d7a79cc20a22bc83809 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:07:06 -0700
Subject: eth: bnxt: move from .set_rxfh to .create_rxfh_context and friends

Use the new ethtool ops for RSS context management. The conversion
is pretty straightforward cut / paste of the right chunks of the
combined handler. Main change is that we let the core pick the IDs
(bitmap will be removed separately for ease of review), so we need
to tell the core when we lose a context.
Since the new API passes rxfh as const, change bnxt_modify_rss()
to also take const.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20240711220713.283778-5-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         |   1 +
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 119 ++++++++++++++--------
 2 files changed, 75 insertions(+), 45 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index e8965cf743fc..bd087c85af5c 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -10257,6 +10257,7 @@ static void bnxt_hwrm_realloc_rss_ctx_vnic(struct bnxt *bp)
 			netdev_err(bp->dev, "Failed to restore RSS ctx %d\n",
 				   rss_ctx->index);
 			bnxt_del_one_rss_ctx(bp, rss_ctx, true);
+			ethtool_rxfh_context_lost(bp->dev, rss_ctx->index);
 		}
 	}
 }
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 846a19b5f58d..afbb7546ee14 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -1841,7 +1841,7 @@ static int bnxt_get_rxfh(struct net_device *dev,
 }
 
 static void bnxt_modify_rss(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
-			    struct ethtool_rxfh_param *rxfh)
+			    const struct ethtool_rxfh_param *rxfh)
 {
 	if (rxfh->key) {
 		if (rss_ctx) {
@@ -1866,44 +1866,35 @@ static void bnxt_modify_rss(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 	}
 }
 
-static int bnxt_set_rxfh_context(struct bnxt *bp,
-				 struct ethtool_rxfh_param *rxfh,
-				 struct netlink_ext_ack *extack)
+static int bnxt_rxfh_context_check(struct bnxt *bp,
+				   struct netlink_ext_ack *extack)
 {
-	u32 *rss_context = &rxfh->rss_context;
-	struct bnxt_rss_ctx *rss_ctx;
-	struct bnxt_vnic_info *vnic;
-	bool modify = false;
-	bool delete;
-	int bit_id;
-	int rc;
-
 	if (!BNXT_SUPPORTS_MULTI_RSS_CTX(bp)) {
 		NL_SET_ERR_MSG_MOD(extack, "RSS contexts not supported");
 		return -EOPNOTSUPP;
 	}
 
-	delete = *rss_context != ETH_RXFH_CONTEXT_ALLOC && rxfh->rss_delete;
-	if (!netif_running(bp->dev) && !delete) {
+	if (!netif_running(bp->dev)) {
 		NL_SET_ERR_MSG_MOD(extack, "Unable to set RSS contexts when interface is down");
 		return -EAGAIN;
 	}
 
-	if (*rss_context != ETH_RXFH_CONTEXT_ALLOC) {
-		rss_ctx = bnxt_get_rss_ctx_from_index(bp, *rss_context);
-		if (!rss_ctx) {
-			NL_SET_ERR_MSG_FMT_MOD(extack, "RSS context %u not found",
-					       *rss_context);
-			return -EINVAL;
-		}
-		if (delete) {
-			bnxt_del_one_rss_ctx(bp, rss_ctx, true);
-			return 0;
-		}
-		modify = true;
-		vnic = &rss_ctx->vnic;
-		goto modify_context;
-	}
+	return 0;
+}
+
+static int bnxt_create_rxfh_context(struct net_device *dev,
+				    struct ethtool_rxfh_context *ctx,
+				    const struct ethtool_rxfh_param *rxfh,
+				    struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_rss_ctx *rss_ctx;
+	struct bnxt_vnic_info *vnic;
+	int rc;
+
+	rc = bnxt_rxfh_context_check(bp, extack);
+	if (rc)
+		return rc;
 
 	if (bp->num_rss_ctx >= BNXT_MAX_ETH_RSS_CTX) {
 		NL_SET_ERR_MSG_FMT_MOD(extack, "Out of RSS contexts, maximum %u",
@@ -1911,6 +1902,11 @@ static int bnxt_set_rxfh_context(struct bnxt *bp,
 		return -EINVAL;
 	}
 
+	if (test_and_set_bit(rxfh->rss_context, bp->rss_ctx_bmap)) {
+		NL_SET_ERR_MSG_MOD(extack, "Context ID conflict");
+		return -EINVAL;
+	}
+
 	if (!bnxt_rfs_capable(bp, true)) {
 		NL_SET_ERR_MSG_MOD(extack, "Out hardware resources");
 		return -ENOMEM;
@@ -1945,33 +1941,65 @@ static int bnxt_set_rxfh_context(struct bnxt *bp,
 		NL_SET_ERR_MSG_MOD(extack, "Unable to setup TPA");
 		goto out;
 	}
-modify_context:
 	bnxt_modify_rss(bp, rss_ctx, rxfh);
 
-	if (modify)
-		return bnxt_hwrm_vnic_rss_cfg_p5(bp, vnic);
-
 	rc = __bnxt_setup_vnic_p5(bp, vnic);
 	if (rc) {
 		NL_SET_ERR_MSG_MOD(extack, "Unable to setup TPA");
 		goto out;
 	}
 
-	bit_id = bitmap_find_free_region(bp->rss_ctx_bmap,
-					 BNXT_RSS_CTX_BMAP_LEN, 0);
-	if (bit_id < 0) {
-		rc = -ENOMEM;
-		goto out;
-	}
-	rss_ctx->index = (u16)bit_id;
-	*rss_context = rss_ctx->index;
-
+	rss_ctx->index = rxfh->rss_context;
 	return 0;
 out:
 	bnxt_del_one_rss_ctx(bp, rss_ctx, true);
 	return rc;
 }
 
+static int bnxt_modify_rxfh_context(struct net_device *dev,
+				    struct ethtool_rxfh_context *ctx,
+				    const struct ethtool_rxfh_param *rxfh,
+				    struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_rss_ctx *rss_ctx;
+	int rc;
+
+	rc = bnxt_rxfh_context_check(bp, extack);
+	if (rc)
+		return rc;
+
+	rss_ctx = bnxt_get_rss_ctx_from_index(bp, rxfh->rss_context);
+	if (!rss_ctx) {
+		NL_SET_ERR_MSG_FMT_MOD(extack, "RSS context %u not found",
+				       rxfh->rss_context);
+		return -EINVAL;
+	}
+
+	bnxt_modify_rss(bp, rss_ctx, rxfh);
+
+	return bnxt_hwrm_vnic_rss_cfg_p5(bp, &rss_ctx->vnic);
+}
+
+static int bnxt_remove_rxfh_context(struct net_device *dev,
+				    struct ethtool_rxfh_context *ctx,
+				    u32 rss_context,
+				    struct netlink_ext_ack *extack)
+{
+	struct bnxt *bp = netdev_priv(dev);
+	struct bnxt_rss_ctx *rss_ctx;
+
+	rss_ctx = bnxt_get_rss_ctx_from_index(bp, rss_context);
+	if (!rss_ctx) {
+		NL_SET_ERR_MSG_FMT_MOD(extack, "RSS context %u not found",
+				       rss_context);
+		return -EINVAL;
+	}
+
+	bnxt_del_one_rss_ctx(bp, rss_ctx, true);
+	return 0;
+}
+
 static int bnxt_set_rxfh(struct net_device *dev,
 			 struct ethtool_rxfh_param *rxfh,
 			 struct netlink_ext_ack *extack)
@@ -1982,9 +2010,6 @@ static int bnxt_set_rxfh(struct net_device *dev,
 	if (rxfh->hfunc && rxfh->hfunc != ETH_RSS_HASH_TOP)
 		return -EOPNOTSUPP;
 
-	if (rxfh->rss_context)
-		return bnxt_set_rxfh_context(bp, rxfh, extack);
-
 	bnxt_modify_rss(bp, NULL, rxfh);
 
 	bnxt_clear_usr_fltrs(bp, false);
@@ -5277,6 +5302,7 @@ void bnxt_ethtool_free(struct bnxt *bp)
 const struct ethtool_ops bnxt_ethtool_ops = {
 	.cap_link_lanes_supported	= 1,
 	.cap_rss_ctx_supported		= 1,
+	.rxfh_max_context_id		= BNXT_MAX_ETH_RSS_CTX,
 	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
 				     ETHTOOL_COALESCE_MAX_FRAMES |
 				     ETHTOOL_COALESCE_USECS_IRQ |
@@ -5314,6 +5340,9 @@ const struct ethtool_ops bnxt_ethtool_ops = {
 	.get_rxfh_key_size      = bnxt_get_rxfh_key_size,
 	.get_rxfh               = bnxt_get_rxfh,
 	.set_rxfh		= bnxt_set_rxfh,
+	.create_rxfh_context	= bnxt_create_rxfh_context,
+	.modify_rxfh_context	= bnxt_modify_rxfh_context,
+	.remove_rxfh_context	= bnxt_remove_rxfh_context,
 	.flash_device		= bnxt_flash_device,
 	.get_eeprom_len         = bnxt_get_eeprom_len,
 	.get_eeprom             = bnxt_get_eeprom,
-- 
cgit v1.2.3-58-ga151


From 1a49a23c034b9c991f01b7034efc555886cee11a Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:07:07 -0700
Subject: eth: bnxt: remove rss_ctx_bmap

Core will allocate IDs for the driver, from the range
[1, BNXT_MAX_ETH_RSS_CTX], no need to track the allocations.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20240711220713.283778-6-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         | 13 ++-----------
 drivers/net/ethernet/broadcom/bnxt/bnxt.h         |  2 --
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c |  5 -----
 3 files changed, 2 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index bd087c85af5c..026ceaa0b329 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -10239,7 +10239,6 @@ void bnxt_del_one_rss_ctx(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 	kfree(rss_ctx->rss_indir_tbl);
 	list_del(&rss_ctx->list);
 	bp->num_rss_ctx--;
-	clear_bit(rss_ctx->index, bp->rss_ctx_bmap);
 	kfree(rss_ctx);
 }
 
@@ -10281,20 +10280,12 @@ void bnxt_clear_rss_ctxs(struct bnxt *bp, bool all)
 
 	list_for_each_entry_safe(rss_ctx, tmp, &bp->rss_ctx_list, list)
 		bnxt_del_one_rss_ctx(bp, rss_ctx, all);
-
-	if (all)
-		bitmap_free(bp->rss_ctx_bmap);
 }
 
 static void bnxt_init_multi_rss_ctx(struct bnxt *bp)
 {
-	bp->rss_ctx_bmap = bitmap_zalloc(BNXT_RSS_CTX_BMAP_LEN, GFP_KERNEL);
-	if (bp->rss_ctx_bmap) {
-		/* burn index 0 since we cannot have context 0 */
-		__set_bit(0, bp->rss_ctx_bmap);
-		INIT_LIST_HEAD(&bp->rss_ctx_list);
-		bp->rss_cap |= BNXT_RSS_CAP_MULTI_RSS_CTX;
-	}
+	INIT_LIST_HEAD(&bp->rss_ctx_list);
+	bp->rss_cap |= BNXT_RSS_CAP_MULTI_RSS_CTX;
 }
 
 /* Allow PF, trusted VFs and VFs with default VLAN to be in promiscuous mode */
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 3c8826875ceb..9b0c6656ce27 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1298,7 +1298,6 @@ struct bnxt_rss_ctx {
 };
 
 #define BNXT_MAX_ETH_RSS_CTX	32
-#define BNXT_RSS_CTX_BMAP_LEN	(BNXT_MAX_ETH_RSS_CTX + 1)
 #define BNXT_VNIC_ID_INVALID	0xffffffff
 
 struct bnxt_hw_rings {
@@ -2332,7 +2331,6 @@ struct bnxt {
 	struct bnxt_ring_grp_info	*grp_info;
 	struct bnxt_vnic_info	*vnic_info;
 	struct list_head	rss_ctx_list;
-	unsigned long		*rss_ctx_bmap;
 	u32			num_rss_ctx;
 	int			nr_vnics;
 	u16			*rss_indir_tbl;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index afbb7546ee14..04855846b5f6 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -1902,11 +1902,6 @@ static int bnxt_create_rxfh_context(struct net_device *dev,
 		return -EINVAL;
 	}
 
-	if (test_and_set_bit(rxfh->rss_context, bp->rss_ctx_bmap)) {
-		NL_SET_ERR_MSG_MOD(extack, "Context ID conflict");
-		return -EINVAL;
-	}
-
 	if (!bnxt_rfs_capable(bp, true)) {
 		NL_SET_ERR_MSG_MOD(extack, "Out hardware resources");
 		return -ENOMEM;
-- 
cgit v1.2.3-58-ga151


From bf30162915f890c113bb2aefcea8be20b3326d00 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:07:08 -0700
Subject: eth: bnxt: depend on core cleaning up RSS contexts

New RSS context API removes old contexts on netdev unregister.
No need to wipe them manually.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20240711220713.283778-7-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         | 11 ++++-------
 drivers/net/ethernet/broadcom/bnxt/bnxt.h         |  2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c |  2 +-
 3 files changed, 6 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 026ceaa0b329..d6a4ce5066c6 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -10274,12 +10274,12 @@ struct bnxt_rss_ctx *bnxt_alloc_rss_ctx(struct bnxt *bp)
 	return rss_ctx;
 }
 
-void bnxt_clear_rss_ctxs(struct bnxt *bp, bool all)
+void bnxt_clear_rss_ctxs(struct bnxt *bp)
 {
 	struct bnxt_rss_ctx *rss_ctx, *tmp;
 
 	list_for_each_entry_safe(rss_ctx, tmp, &bp->rss_ctx_list, list)
-		bnxt_del_one_rss_ctx(bp, rss_ctx, all);
+		bnxt_del_one_rss_ctx(bp, rss_ctx, false);
 }
 
 static void bnxt_init_multi_rss_ctx(struct bnxt *bp)
@@ -12331,7 +12331,7 @@ static void __bnxt_close_nic(struct bnxt *bp, bool irq_re_init,
 		msleep(20);
 
 	if (BNXT_SUPPORTS_MULTI_RSS_CTX(bp))
-		bnxt_clear_rss_ctxs(bp, false);
+		bnxt_clear_rss_ctxs(bp);
 	/* Flush rings and disable interrupts */
 	bnxt_shutdown_nic(bp, irq_re_init);
 
@@ -15246,8 +15246,7 @@ static void bnxt_remove_one(struct pci_dev *pdev)
 
 	bnxt_free_l2_filters(bp, true);
 	bnxt_free_ntp_fltrs(bp, true);
-	if (BNXT_SUPPORTS_MULTI_RSS_CTX(bp))
-		bnxt_clear_rss_ctxs(bp, true);
+	WARN_ON(bp->num_rss_ctx);
 	clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
 	/* Flush any pending tasks */
 	cancel_work_sync(&bp->sp_task);
@@ -15898,8 +15897,6 @@ init_err_dl:
 	bnxt_clear_int_mode(bp);
 
 init_err_pci_clean:
-	if (BNXT_SUPPORTS_MULTI_RSS_CTX(bp))
-		bnxt_clear_rss_ctxs(bp, true);
 	bnxt_hwrm_func_drv_unrgtr(bp);
 	bnxt_free_hwrm_resources(bp);
 	bnxt_hwmon_uninit(bp);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 9b0c6656ce27..df96524a8b8b 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -2847,7 +2847,7 @@ int __bnxt_setup_vnic_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic);
 void bnxt_del_one_rss_ctx(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 			  bool all);
 struct bnxt_rss_ctx *bnxt_alloc_rss_ctx(struct bnxt *bp);
-void bnxt_clear_rss_ctxs(struct bnxt *bp, bool all);
+void bnxt_clear_rss_ctxs(struct bnxt *bp);
 int bnxt_open_nic(struct bnxt *, bool, bool);
 int bnxt_half_open_nic(struct bnxt *bp);
 void bnxt_half_close_nic(struct bnxt *bp);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 04855846b5f6..5a64c0ea56c0 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -976,7 +976,7 @@ static int bnxt_set_channels(struct net_device *dev,
 
 	bnxt_clear_usr_fltrs(bp, true);
 	if (BNXT_SUPPORTS_MULTI_RSS_CTX(bp))
-		bnxt_clear_rss_ctxs(bp, false);
+		bnxt_clear_rss_ctxs(bp);
 	if (netif_running(dev)) {
 		if (BNXT_PF(bp)) {
 			/* TODO CHIMP_FW: Send message to all VF's
-- 
cgit v1.2.3-58-ga151


From 63d4769cf74acfc42a3436ff28e92ef4f70e617c Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:07:09 -0700
Subject: eth: bnxt: use context priv for struct bnxt_rss_ctx

Core can allocate space for per-context driver-private data,
use it for struct bnxt_rss_ctx. Inline bnxt_alloc_rss_ctx()
at this point, most of the init (as in the actions bnxt_del_one_rss_ctx()
will undo) is open coded in bnxt_create_rxfh_context(), anyway.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20240711220713.283778-8-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         | 14 -------------
 drivers/net/ethernet/broadcom/bnxt/bnxt.h         |  1 -
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 24 +++++++++--------------
 3 files changed, 9 insertions(+), 30 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index d6a4ce5066c6..e3cc34712f33 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -10239,7 +10239,6 @@ void bnxt_del_one_rss_ctx(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 	kfree(rss_ctx->rss_indir_tbl);
 	list_del(&rss_ctx->list);
 	bp->num_rss_ctx--;
-	kfree(rss_ctx);
 }
 
 static void bnxt_hwrm_realloc_rss_ctx_vnic(struct bnxt *bp)
@@ -10261,19 +10260,6 @@ static void bnxt_hwrm_realloc_rss_ctx_vnic(struct bnxt *bp)
 	}
 }
 
-struct bnxt_rss_ctx *bnxt_alloc_rss_ctx(struct bnxt *bp)
-{
-	struct bnxt_rss_ctx *rss_ctx = NULL;
-
-	rss_ctx = kzalloc(sizeof(*rss_ctx), GFP_KERNEL);
-	if (rss_ctx) {
-		rss_ctx->vnic.rss_ctx = rss_ctx;
-		list_add_tail(&rss_ctx->list, &bp->rss_ctx_list);
-		bp->num_rss_ctx++;
-	}
-	return rss_ctx;
-}
-
 void bnxt_clear_rss_ctxs(struct bnxt *bp)
 {
 	struct bnxt_rss_ctx *rss_ctx, *tmp;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index df96524a8b8b..c5fd7a4e6681 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -2846,7 +2846,6 @@ int bnxt_hwrm_vnic_rss_cfg_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic);
 int __bnxt_setup_vnic_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic);
 void bnxt_del_one_rss_ctx(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 			  bool all);
-struct bnxt_rss_ctx *bnxt_alloc_rss_ctx(struct bnxt *bp);
 void bnxt_clear_rss_ctxs(struct bnxt *bp);
 int bnxt_open_nic(struct bnxt *, bool, bool);
 int bnxt_half_open_nic(struct bnxt *bp);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 5a64c0ea56c0..de8e13412151 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -1907,11 +1907,13 @@ static int bnxt_create_rxfh_context(struct net_device *dev,
 		return -ENOMEM;
 	}
 
-	rss_ctx = bnxt_alloc_rss_ctx(bp);
-	if (!rss_ctx)
-		return -ENOMEM;
+	rss_ctx = ethtool_rxfh_context_priv(ctx);
+
+	list_add_tail(&rss_ctx->list, &bp->rss_ctx_list);
+	bp->num_rss_ctx++;
 
 	vnic = &rss_ctx->vnic;
+	vnic->rss_ctx = rss_ctx;
 	vnic->flags |= BNXT_VNIC_RSSCTX_FLAG;
 	vnic->vnic_id = BNXT_VNIC_ID_INVALID;
 	rc = bnxt_alloc_rss_ctx_rss_table(bp, rss_ctx);
@@ -1964,12 +1966,7 @@ static int bnxt_modify_rxfh_context(struct net_device *dev,
 	if (rc)
 		return rc;
 
-	rss_ctx = bnxt_get_rss_ctx_from_index(bp, rxfh->rss_context);
-	if (!rss_ctx) {
-		NL_SET_ERR_MSG_FMT_MOD(extack, "RSS context %u not found",
-				       rxfh->rss_context);
-		return -EINVAL;
-	}
+	rss_ctx = ethtool_rxfh_context_priv(ctx);
 
 	bnxt_modify_rss(bp, rss_ctx, rxfh);
 
@@ -1984,12 +1981,7 @@ static int bnxt_remove_rxfh_context(struct net_device *dev,
 	struct bnxt *bp = netdev_priv(dev);
 	struct bnxt_rss_ctx *rss_ctx;
 
-	rss_ctx = bnxt_get_rss_ctx_from_index(bp, rss_context);
-	if (!rss_ctx) {
-		NL_SET_ERR_MSG_FMT_MOD(extack, "RSS context %u not found",
-				       rss_context);
-		return -EINVAL;
-	}
+	rss_ctx = ethtool_rxfh_context_priv(ctx);
 
 	bnxt_del_one_rss_ctx(bp, rss_ctx, true);
 	return 0;
@@ -5298,6 +5290,8 @@ const struct ethtool_ops bnxt_ethtool_ops = {
 	.cap_link_lanes_supported	= 1,
 	.cap_rss_ctx_supported		= 1,
 	.rxfh_max_context_id		= BNXT_MAX_ETH_RSS_CTX,
+	.rxfh_indir_space		= BNXT_MAX_RSS_TABLE_ENTRIES_P5,
+	.rxfh_priv_size			= sizeof(struct bnxt_rss_ctx),
 	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
 				     ETHTOOL_COALESCE_MAX_FRAMES |
 				     ETHTOOL_COALESCE_USECS_IRQ |
-- 
cgit v1.2.3-58-ga151


From 20c8ad72eb7f1510d2a802328cd0809b4db9e18a Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:07:10 -0700
Subject: eth: bnxt: use the RSS context XArray instead of the local list

Core already maintains all RSS contexts in an XArray, no need
to keep a second list in the driver.

Remove bnxt_get_max_rss_ctx_ring() completely since core performs
the same check already.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20240711220713.283778-9-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         | 56 +++++++++--------------
 drivers/net/ethernet/broadcom/bnxt/bnxt.h         |  3 --
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 17 ++-----
 3 files changed, 26 insertions(+), 50 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index e3cc34712f33..f9554f512314 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -5970,17 +5970,20 @@ bnxt_cfg_rfs_ring_tbl_idx(struct bnxt *bp,
 			  struct hwrm_cfa_ntuple_filter_alloc_input *req,
 			  struct bnxt_ntuple_filter *fltr)
 {
-	struct bnxt_rss_ctx *rss_ctx, *tmp;
 	u16 rxq = fltr->base.rxq;
 
 	if (fltr->base.flags & BNXT_ACT_RSS_CTX) {
-		list_for_each_entry_safe(rss_ctx, tmp, &bp->rss_ctx_list, list) {
-			if (rss_ctx->index == fltr->base.fw_vnic_id) {
-				struct bnxt_vnic_info *vnic = &rss_ctx->vnic;
+		struct ethtool_rxfh_context *ctx;
+		struct bnxt_rss_ctx *rss_ctx;
+		struct bnxt_vnic_info *vnic;
 
-				req->dst_id = cpu_to_le16(vnic->fw_vnic_id);
-				break;
-			}
+		ctx = xa_load(&bp->dev->ethtool->rss_ctx,
+			      fltr->base.fw_vnic_id);
+		if (ctx) {
+			rss_ctx = ethtool_rxfh_context_priv(ctx);
+			vnic = &rss_ctx->vnic;
+
+			req->dst_id = cpu_to_le16(vnic->fw_vnic_id);
 		}
 		return;
 	}
@@ -6282,21 +6285,6 @@ static u16 bnxt_get_max_rss_ring(struct bnxt *bp)
 	return max_ring;
 }
 
-u16 bnxt_get_max_rss_ctx_ring(struct bnxt *bp)
-{
-	u16 i, tbl_size, max_ring = 0;
-	struct bnxt_rss_ctx *rss_ctx;
-
-	tbl_size = bnxt_get_rxfh_indir_size(bp->dev);
-
-	list_for_each_entry(rss_ctx, &bp->rss_ctx_list, list) {
-		for (i = 0; i < tbl_size; i++)
-			max_ring = max(max_ring, rss_ctx->rss_indir_tbl[i]);
-	}
-
-	return max_ring;
-}
-
 int bnxt_get_nr_rss_ctxs(struct bnxt *bp, int rx_rings)
 {
 	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS) {
@@ -10237,16 +10225,17 @@ void bnxt_del_one_rss_ctx(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 				  vnic->rss_table,
 				  vnic->rss_table_dma_addr);
 	kfree(rss_ctx->rss_indir_tbl);
-	list_del(&rss_ctx->list);
 	bp->num_rss_ctx--;
 }
 
 static void bnxt_hwrm_realloc_rss_ctx_vnic(struct bnxt *bp)
 {
 	bool set_tpa = !!(bp->flags & BNXT_FLAG_TPA);
-	struct bnxt_rss_ctx *rss_ctx, *tmp;
+	struct ethtool_rxfh_context *ctx;
+	unsigned long context;
 
-	list_for_each_entry_safe(rss_ctx, tmp, &bp->rss_ctx_list, list) {
+	xa_for_each(&bp->dev->ethtool->rss_ctx, context, ctx) {
+		struct bnxt_rss_ctx *rss_ctx = ethtool_rxfh_context_priv(ctx);
 		struct bnxt_vnic_info *vnic = &rss_ctx->vnic;
 
 		if (bnxt_hwrm_vnic_alloc(bp, vnic, 0, bp->rx_nr_rings) ||
@@ -10262,16 +10251,14 @@ static void bnxt_hwrm_realloc_rss_ctx_vnic(struct bnxt *bp)
 
 void bnxt_clear_rss_ctxs(struct bnxt *bp)
 {
-	struct bnxt_rss_ctx *rss_ctx, *tmp;
+	struct ethtool_rxfh_context *ctx;
+	unsigned long context;
 
-	list_for_each_entry_safe(rss_ctx, tmp, &bp->rss_ctx_list, list)
-		bnxt_del_one_rss_ctx(bp, rss_ctx, false);
-}
+	xa_for_each(&bp->dev->ethtool->rss_ctx, context, ctx) {
+		struct bnxt_rss_ctx *rss_ctx = ethtool_rxfh_context_priv(ctx);
 
-static void bnxt_init_multi_rss_ctx(struct bnxt *bp)
-{
-	INIT_LIST_HEAD(&bp->rss_ctx_list);
-	bp->rss_cap |= BNXT_RSS_CAP_MULTI_RSS_CTX;
+		bnxt_del_one_rss_ctx(bp, rss_ctx, false);
+	}
 }
 
 /* Allow PF, trusted VFs and VFs with default VLAN to be in promiscuous mode */
@@ -15859,8 +15846,7 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	INIT_LIST_HEAD(&bp->usr_fltr_list);
 
 	if (BNXT_SUPPORTS_NTUPLE_VNIC(bp))
-		bnxt_init_multi_rss_ctx(bp);
-
+		bp->rss_cap |= BNXT_RSS_CAP_MULTI_RSS_CTX;
 
 	rc = register_netdev(dev);
 	if (rc)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index c5fd7a4e6681..be40e0513777 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1291,7 +1291,6 @@ struct bnxt_vnic_info {
 };
 
 struct bnxt_rss_ctx {
-	struct list_head list;
 	struct bnxt_vnic_info vnic;
 	u16	*rss_indir_tbl;
 	u8	index;
@@ -2330,7 +2329,6 @@ struct bnxt {
 	/* grp_info indexed by completion ring index */
 	struct bnxt_ring_grp_info	*grp_info;
 	struct bnxt_vnic_info	*vnic_info;
-	struct list_head	rss_ctx_list;
 	u32			num_rss_ctx;
 	int			nr_vnics;
 	u16			*rss_indir_tbl;
@@ -2812,7 +2810,6 @@ int bnxt_hwrm_vnic_set_tpa(struct bnxt *bp, struct bnxt_vnic_info *vnic,
 void bnxt_fill_ipv6_mask(__be32 mask[4]);
 int bnxt_alloc_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx);
 void bnxt_set_dflt_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx);
-u16 bnxt_get_max_rss_ctx_ring(struct bnxt *bp);
 int bnxt_get_nr_rss_ctxs(struct bnxt *bp, int rx_rings);
 int bnxt_hwrm_vnic_cfg(struct bnxt *bp, struct bnxt_vnic_info *vnic);
 int bnxt_hwrm_vnic_alloc(struct bnxt *bp, struct bnxt_vnic_info *vnic,
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index de8e13412151..74765583405b 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -961,12 +961,6 @@ static int bnxt_set_channels(struct net_device *dev,
 		return rc;
 	}
 
-	if (req_rx_rings < bp->rx_nr_rings &&
-	    req_rx_rings <= bnxt_get_max_rss_ctx_ring(bp)) {
-		netdev_warn(dev, "Can't deactivate rings used by RSS contexts\n");
-		return -EINVAL;
-	}
-
 	if (bnxt_get_nr_rss_ctxs(bp, req_rx_rings) !=
 	    bnxt_get_nr_rss_ctxs(bp, bp->rx_nr_rings) &&
 	    netif_is_rxfh_configured(dev)) {
@@ -1216,12 +1210,12 @@ fltr_err:
 static struct bnxt_rss_ctx *bnxt_get_rss_ctx_from_index(struct bnxt *bp,
 							u32 index)
 {
-	struct bnxt_rss_ctx *rss_ctx, *tmp;
+	struct ethtool_rxfh_context *ctx;
 
-	list_for_each_entry_safe(rss_ctx, tmp, &bp->rss_ctx_list, list)
-		if (rss_ctx->index == index)
-			return rss_ctx;
-	return NULL;
+	ctx = xa_load(&bp->dev->ethtool->rss_ctx, index);
+	if (!ctx)
+		return NULL;
+	return ethtool_rxfh_context_priv(ctx);
 }
 
 static int bnxt_alloc_rss_ctx_rss_table(struct bnxt *bp,
@@ -1909,7 +1903,6 @@ static int bnxt_create_rxfh_context(struct net_device *dev,
 
 	rss_ctx = ethtool_rxfh_context_priv(ctx);
 
-	list_add_tail(&rss_ctx->list, &bp->rss_ctx_list);
 	bp->num_rss_ctx++;
 
 	vnic = &rss_ctx->vnic;
-- 
cgit v1.2.3-58-ga151


From 9c34c6c28c704fb2f6124d7b6a6837bdbc8e72fc Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:07:11 -0700
Subject: eth: bnxt: pad out the correct indirection table

bnxt allocates tables of max size, and changes the used size
based on number of active rings. The unused entries get padded
out with zeros. bnxt_modify_rss() seems to always pad out
the table of the main / default RSS context, instead of
the table of the modified context.

I haven't observed any behavior change due to this patch,
so I don't think it's a fix. Not entirely sure what role
the padding plays, 0 is a valid queue ID.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20240711220713.283778-10-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 74765583405b..8c30f740a9c5 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -1856,7 +1856,7 @@ static void bnxt_modify_rss(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 			indir_tbl[i] = rxfh->indir[i];
 		pad = bp->rss_indir_tbl_entries - tbl_size;
 		if (pad)
-			memset(&bp->rss_indir_tbl[i], 0, pad * sizeof(u16));
+			memset(&indir_tbl[i], 0, pad * sizeof(u16));
 	}
 }
 
-- 
cgit v1.2.3-58-ga151


From 73afb518af4af2364793849b069716425a665c6c Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:07:12 -0700
Subject: eth: bnxt: bump the entry size in indir tables to u32

Ethtool core stores indirection table with u32 entries, "just to be safe".
Switch the type in the driver, so that it's easier to swap local tables
for the core ones. Memory allocations already use sizeof(*entry), switch
the memset()s as well.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20240711220713.283778-11-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         | 8 ++++----
 drivers/net/ethernet/broadcom/bnxt/bnxt.h         | 4 ++--
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 6 +++---
 3 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index f9554f512314..ed46d74533ed 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -6225,7 +6225,7 @@ static u16 bnxt_cp_ring_for_tx(struct bnxt *bp, struct bnxt_tx_ring_info *txr)
 int bnxt_alloc_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx)
 {
 	int entries;
-	u16 *tbl;
+	u32 *tbl;
 
 	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
 		entries = BNXT_MAX_RSS_TABLE_ENTRIES_P5;
@@ -6248,7 +6248,7 @@ int bnxt_alloc_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx)
 void bnxt_set_dflt_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx)
 {
 	u16 max_rings, max_entries, pad, i;
-	u16 *rss_indir_tbl;
+	u32 *rss_indir_tbl;
 
 	if (!bp->rx_nr_rings)
 		return;
@@ -6269,12 +6269,12 @@ void bnxt_set_dflt_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx)
 
 	pad = bp->rss_indir_tbl_entries - max_entries;
 	if (pad)
-		memset(&rss_indir_tbl[i], 0, pad * sizeof(u16));
+		memset(&rss_indir_tbl[i], 0, pad * sizeof(*rss_indir_tbl));
 }
 
 static u16 bnxt_get_max_rss_ring(struct bnxt *bp)
 {
-	u16 i, tbl_size, max_ring = 0;
+	u32 i, tbl_size, max_ring = 0;
 
 	if (!bp->rss_indir_tbl)
 		return 0;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index be40e0513777..1a33824a32a8 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1292,7 +1292,7 @@ struct bnxt_vnic_info {
 
 struct bnxt_rss_ctx {
 	struct bnxt_vnic_info vnic;
-	u16	*rss_indir_tbl;
+	u32	*rss_indir_tbl;
 	u8	index;
 };
 
@@ -2331,7 +2331,7 @@ struct bnxt {
 	struct bnxt_vnic_info	*vnic_info;
 	u32			num_rss_ctx;
 	int			nr_vnics;
-	u16			*rss_indir_tbl;
+	u32			*rss_indir_tbl;
 	u16			rss_indir_tbl_entries;
 	u32			rss_hash_cfg;
 	u32			rss_hash_delta;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 8c30f740a9c5..edcbb1a4f6c8 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -1804,7 +1804,7 @@ static int bnxt_get_rxfh(struct net_device *dev,
 	u32 rss_context = rxfh->rss_context;
 	struct bnxt_rss_ctx *rss_ctx = NULL;
 	struct bnxt *bp = netdev_priv(dev);
-	u16 *indir_tbl = bp->rss_indir_tbl;
+	u32 *indir_tbl = bp->rss_indir_tbl;
 	struct bnxt_vnic_info *vnic;
 	u32 i, tbl_size;
 
@@ -1848,7 +1848,7 @@ static void bnxt_modify_rss(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 	}
 	if (rxfh->indir) {
 		u32 i, pad, tbl_size = bnxt_get_rxfh_indir_size(bp->dev);
-		u16 *indir_tbl = bp->rss_indir_tbl;
+		u32 *indir_tbl = bp->rss_indir_tbl;
 
 		if (rss_ctx)
 			indir_tbl = rss_ctx->rss_indir_tbl;
@@ -1856,7 +1856,7 @@ static void bnxt_modify_rss(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 			indir_tbl[i] = rxfh->indir[i];
 		pad = bp->rss_indir_tbl_entries - tbl_size;
 		if (pad)
-			memset(&indir_tbl[i], 0, pad * sizeof(u16));
+			memset(&indir_tbl[i], 0, pad * sizeof(*indir_tbl));
 	}
 }
 
-- 
cgit v1.2.3-58-ga151


From 46e457a454de1d95c57394129a2be6eac8bf85e4 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:07:13 -0700
Subject: eth: bnxt: use the indir table from ethtool context

Instead of allocating a separate indir table in the vnic use
the one already present in the RSS context allocated by the core.
This saves some LoC and also we won't have to worry about syncing
the local version back to the core, once core learns how to dump
contexts.

Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Link: https://patch.msgid.link/20240711220713.283778-12-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c         | 23 ++++++---------
 drivers/net/ethernet/broadcom/bnxt/bnxt.h         |  7 ++---
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 36 +++++++++++------------
 3 files changed, 29 insertions(+), 37 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index ed46d74533ed..bb3be33c1bbd 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -6222,10 +6222,9 @@ static u16 bnxt_cp_ring_for_tx(struct bnxt *bp, struct bnxt_tx_ring_info *txr)
 		return bnxt_cp_ring_from_grp(bp, &txr->tx_ring_struct);
 }
 
-int bnxt_alloc_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx)
+static int bnxt_alloc_rss_indir_tbl(struct bnxt *bp)
 {
 	int entries;
-	u32 *tbl;
 
 	if (bp->flags & BNXT_FLAG_CHIP_P5_PLUS)
 		entries = BNXT_MAX_RSS_TABLE_ENTRIES_P5;
@@ -6233,19 +6232,16 @@ int bnxt_alloc_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx)
 		entries = HW_HASH_INDEX_SIZE;
 
 	bp->rss_indir_tbl_entries = entries;
-	tbl = kmalloc_array(entries, sizeof(*bp->rss_indir_tbl), GFP_KERNEL);
-	if (!tbl)
+	bp->rss_indir_tbl =
+		kmalloc_array(entries, sizeof(*bp->rss_indir_tbl), GFP_KERNEL);
+	if (!bp->rss_indir_tbl)
 		return -ENOMEM;
 
-	if (rss_ctx)
-		rss_ctx->rss_indir_tbl = tbl;
-	else
-		bp->rss_indir_tbl = tbl;
-
 	return 0;
 }
 
-void bnxt_set_dflt_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx)
+void bnxt_set_dflt_rss_indir_tbl(struct bnxt *bp,
+				 struct ethtool_rxfh_context *rss_ctx)
 {
 	u16 max_rings, max_entries, pad, i;
 	u32 *rss_indir_tbl;
@@ -6260,7 +6256,7 @@ void bnxt_set_dflt_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx)
 
 	max_entries = bnxt_get_rxfh_indir_size(bp->dev);
 	if (rss_ctx)
-		rss_indir_tbl = &rss_ctx->rss_indir_tbl[0];
+		rss_indir_tbl = ethtool_rxfh_context_indir(rss_ctx);
 	else
 		rss_indir_tbl = &bp->rss_indir_tbl[0];
 
@@ -6326,7 +6322,7 @@ static void bnxt_fill_hw_rss_tbl_p5(struct bnxt *bp,
 		if (vnic->flags & BNXT_VNIC_NTUPLE_FLAG)
 			j = ethtool_rxfh_indir_default(i, bp->rx_nr_rings);
 		else if (vnic->flags & BNXT_VNIC_RSSCTX_FLAG)
-			j = vnic->rss_ctx->rss_indir_tbl[i];
+			j = ethtool_rxfh_context_indir(vnic->rss_ctx)[i];
 		else
 			j = bp->rss_indir_tbl[i];
 		rxr = &bp->rx_ring[j];
@@ -10224,7 +10220,6 @@ void bnxt_del_one_rss_ctx(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 		dma_free_coherent(&bp->pdev->dev, vnic->rss_table_size,
 				  vnic->rss_table,
 				  vnic->rss_table_dma_addr);
-	kfree(rss_ctx->rss_indir_tbl);
 	bp->num_rss_ctx--;
 }
 
@@ -15689,7 +15684,7 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 			bp->flags |= BNXT_FLAG_CHIP_P7;
 	}
 
-	rc = bnxt_alloc_rss_indir_tbl(bp, NULL);
+	rc = bnxt_alloc_rss_indir_tbl(bp);
 	if (rc)
 		goto init_err_pci_clean;
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 1a33824a32a8..6bbdc718c3a7 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1286,13 +1286,12 @@ struct bnxt_vnic_info {
 #define BNXT_VNIC_RFS_NEW_RSS_FLAG	0x10
 #define BNXT_VNIC_NTUPLE_FLAG		0x20
 #define BNXT_VNIC_RSSCTX_FLAG		0x40
-	struct bnxt_rss_ctx	*rss_ctx;
+	struct ethtool_rxfh_context *rss_ctx;
 	u32		vnic_id;
 };
 
 struct bnxt_rss_ctx {
 	struct bnxt_vnic_info vnic;
-	u32	*rss_indir_tbl;
 	u8	index;
 };
 
@@ -2808,8 +2807,8 @@ int bnxt_hwrm_cfa_ntuple_filter_alloc(struct bnxt *bp,
 int bnxt_hwrm_vnic_set_tpa(struct bnxt *bp, struct bnxt_vnic_info *vnic,
 			   u32 tpa_flags);
 void bnxt_fill_ipv6_mask(__be32 mask[4]);
-int bnxt_alloc_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx);
-void bnxt_set_dflt_rss_indir_tbl(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx);
+void bnxt_set_dflt_rss_indir_tbl(struct bnxt *bp,
+				 struct ethtool_rxfh_context *rss_ctx);
 int bnxt_get_nr_rss_ctxs(struct bnxt *bp, int rx_rings);
 int bnxt_hwrm_vnic_cfg(struct bnxt *bp, struct bnxt_vnic_info *vnic);
 int bnxt_hwrm_vnic_alloc(struct bnxt *bp, struct bnxt_vnic_info *vnic,
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index edcbb1a4f6c8..f121a5e9691f 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -1218,11 +1218,10 @@ static struct bnxt_rss_ctx *bnxt_get_rss_ctx_from_index(struct bnxt *bp,
 	return ethtool_rxfh_context_priv(ctx);
 }
 
-static int bnxt_alloc_rss_ctx_rss_table(struct bnxt *bp,
-					struct bnxt_rss_ctx *rss_ctx)
+static int bnxt_alloc_vnic_rss_table(struct bnxt *bp,
+				     struct bnxt_vnic_info *vnic)
 {
 	int size = L1_CACHE_ALIGN(BNXT_MAX_RSS_TABLE_SIZE_P5);
-	struct bnxt_vnic_info *vnic = &rss_ctx->vnic;
 
 	vnic->rss_table_size = size + HW_HASH_KEY_SIZE;
 	vnic->rss_table = dma_alloc_coherent(&bp->pdev->dev,
@@ -1801,7 +1800,6 @@ static u32 bnxt_get_rxfh_key_size(struct net_device *dev)
 static int bnxt_get_rxfh(struct net_device *dev,
 			 struct ethtool_rxfh_param *rxfh)
 {
-	u32 rss_context = rxfh->rss_context;
 	struct bnxt_rss_ctx *rss_ctx = NULL;
 	struct bnxt *bp = netdev_priv(dev);
 	u32 *indir_tbl = bp->rss_indir_tbl;
@@ -1815,10 +1813,13 @@ static int bnxt_get_rxfh(struct net_device *dev,
 
 	vnic = &bp->vnic_info[BNXT_VNIC_DEFAULT];
 	if (rxfh->rss_context) {
-		rss_ctx = bnxt_get_rss_ctx_from_index(bp, rss_context);
-		if (!rss_ctx)
+		struct ethtool_rxfh_context *ctx;
+
+		ctx = xa_load(&bp->dev->ethtool->rss_ctx, rxfh->rss_context);
+		if (!ctx)
 			return -EINVAL;
-		indir_tbl = rss_ctx->rss_indir_tbl;
+		indir_tbl = ethtool_rxfh_context_indir(ctx);
+		rss_ctx = ethtool_rxfh_context_priv(ctx);
 		vnic = &rss_ctx->vnic;
 	}
 
@@ -1834,7 +1835,8 @@ static int bnxt_get_rxfh(struct net_device *dev,
 	return 0;
 }
 
-static void bnxt_modify_rss(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
+static void bnxt_modify_rss(struct bnxt *bp, struct ethtool_rxfh_context *ctx,
+			    struct bnxt_rss_ctx *rss_ctx,
 			    const struct ethtool_rxfh_param *rxfh)
 {
 	if (rxfh->key) {
@@ -1851,7 +1853,7 @@ static void bnxt_modify_rss(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx,
 		u32 *indir_tbl = bp->rss_indir_tbl;
 
 		if (rss_ctx)
-			indir_tbl = rss_ctx->rss_indir_tbl;
+			indir_tbl = ethtool_rxfh_context_indir(ctx);
 		for (i = 0; i < tbl_size; i++)
 			indir_tbl[i] = rxfh->indir[i];
 		pad = bp->rss_indir_tbl_entries - tbl_size;
@@ -1906,18 +1908,14 @@ static int bnxt_create_rxfh_context(struct net_device *dev,
 	bp->num_rss_ctx++;
 
 	vnic = &rss_ctx->vnic;
-	vnic->rss_ctx = rss_ctx;
+	vnic->rss_ctx = ctx;
 	vnic->flags |= BNXT_VNIC_RSSCTX_FLAG;
 	vnic->vnic_id = BNXT_VNIC_ID_INVALID;
-	rc = bnxt_alloc_rss_ctx_rss_table(bp, rss_ctx);
-	if (rc)
-		goto out;
-
-	rc = bnxt_alloc_rss_indir_tbl(bp, rss_ctx);
+	rc = bnxt_alloc_vnic_rss_table(bp, vnic);
 	if (rc)
 		goto out;
 
-	bnxt_set_dflt_rss_indir_tbl(bp, rss_ctx);
+	bnxt_set_dflt_rss_indir_tbl(bp, ctx);
 	memcpy(vnic->rss_hash_key, bp->rss_hash_key, HW_HASH_KEY_SIZE);
 
 	rc = bnxt_hwrm_vnic_alloc(bp, vnic, 0, bp->rx_nr_rings);
@@ -1931,7 +1929,7 @@ static int bnxt_create_rxfh_context(struct net_device *dev,
 		NL_SET_ERR_MSG_MOD(extack, "Unable to setup TPA");
 		goto out;
 	}
-	bnxt_modify_rss(bp, rss_ctx, rxfh);
+	bnxt_modify_rss(bp, ctx, rss_ctx, rxfh);
 
 	rc = __bnxt_setup_vnic_p5(bp, vnic);
 	if (rc) {
@@ -1961,7 +1959,7 @@ static int bnxt_modify_rxfh_context(struct net_device *dev,
 
 	rss_ctx = ethtool_rxfh_context_priv(ctx);
 
-	bnxt_modify_rss(bp, rss_ctx, rxfh);
+	bnxt_modify_rss(bp, ctx, rss_ctx, rxfh);
 
 	return bnxt_hwrm_vnic_rss_cfg_p5(bp, &rss_ctx->vnic);
 }
@@ -1990,7 +1988,7 @@ static int bnxt_set_rxfh(struct net_device *dev,
 	if (rxfh->hfunc && rxfh->hfunc != ETH_RSS_HASH_TOP)
 		return -EOPNOTSUPP;
 
-	bnxt_modify_rss(bp, NULL, rxfh);
+	bnxt_modify_rss(bp, NULL, NULL, rxfh);
 
 	bnxt_clear_usr_fltrs(bp, false);
 	if (netif_running(bp->dev)) {
-- 
cgit v1.2.3-58-ga151


From 8e5f53a684338c04b7f7487c65e1519b83295046 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Fri, 12 Jul 2024 23:42:13 +0000
Subject: MAINTAINERS: add 5 missing tcp-related files

Following files are part of TCP stack:

- net/ipv4/inet_connection_sock.c
- net/ipv4/inet_hashtables.c
- net/ipv4/inet_timewait_sock.c
- net/ipv6/inet6_connection_sock.c
- net/ipv6/inet6_hashtables.c

Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20240712234213.3178593-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 MAINTAINERS | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 204ae640bf13..5adfbc01bd2e 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -15769,8 +15769,13 @@ F:	include/linux/tcp.h
 F:	include/net/tcp.h
 F:	include/trace/events/tcp.h
 F:	include/uapi/linux/tcp.h
+F:	net/ipv4/inet_connection_sock.c
+F:	net/ipv4/inet_hashtables.c
+F:	net/ipv4/inet_timewait_sock.c
 F:	net/ipv4/syncookies.c
 F:	net/ipv4/tcp*.c
+F:	net/ipv6/inet6_connection_sock.c
+F:	net/ipv6/inet6_hashtables.c
 F:	net/ipv6/syncookies.c
 F:	net/ipv6/tcp*.c
 
-- 
cgit v1.2.3-58-ga151


From 42ffe242860c401c34c62aa369c2add341a6eece Mon Sep 17 00:00:00 2001
From: Peng Fan <peng.fan@nxp.com>
Date: Wed, 10 Jul 2024 20:27:28 +0800
Subject: test/vsock: add install target

Add install target for vsock to make Yocto easy to install the images.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Link: https://patch.msgid.link/20240710122728.45044-1-peng.fan@oss.nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/vsock/Makefile | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/tools/testing/vsock/Makefile b/tools/testing/vsock/Makefile
index a7f56a09ca9f..6e0b4e95e230 100644
--- a/tools/testing/vsock/Makefile
+++ b/tools/testing/vsock/Makefile
@@ -13,3 +13,16 @@ CFLAGS += -g -O2 -Werror -Wall -I. -I../../include -I../../../usr/include -Wno-p
 clean:
 	${RM} *.o *.d vsock_test vsock_diag_test vsock_perf vsock_uring_test
 -include *.d
+
+VSOCK_INSTALL_PATH ?=
+
+install: all
+ifdef VSOCK_INSTALL_PATH
+	mkdir -p $(VSOCK_INSTALL_PATH)
+	install -m 744 vsock_test $(VSOCK_INSTALL_PATH)
+	install -m 744 vsock_perf $(VSOCK_INSTALL_PATH)
+	install -m 744 vsock_diag_test $(VSOCK_INSTALL_PATH)
+	install -m 744 vsock_uring_test $(VSOCK_INSTALL_PATH)
+else
+	$(error Error: set VSOCK_INSTALL_PATH to use install)
+endif
-- 
cgit v1.2.3-58-ga151


From 23e89e8ee7be73e21200947885a6d3a109a2c58d Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Wed, 10 Jul 2024 10:12:45 -0700
Subject: tcp: Don't drop SYN+ACK for simultaneous connect().

RFC 9293 states that in the case of simultaneous connect(), the connection
gets established when SYN+ACK is received. [0]

      TCP Peer A                                       TCP Peer B

  1.  CLOSED                                           CLOSED
  2.  SYN-SENT     --> <SEQ=100><CTL=SYN>              ...
  3.  SYN-RECEIVED <-- <SEQ=300><CTL=SYN>              <-- SYN-SENT
  4.               ... <SEQ=100><CTL=SYN>              --> SYN-RECEIVED
  5.  SYN-RECEIVED --> <SEQ=100><ACK=301><CTL=SYN,ACK> ...
  6.  ESTABLISHED  <-- <SEQ=300><ACK=101><CTL=SYN,ACK> <-- SYN-RECEIVED
  7.               ... <SEQ=100><ACK=301><CTL=SYN,ACK> --> ESTABLISHED

However, since commit 0c24604b68fc ("tcp: implement RFC 5961 4.2"), such a
SYN+ACK is dropped in tcp_validate_incoming() and responded with Challenge
ACK.

For example, the write() syscall in the following packetdrill script fails
with -EAGAIN, and wrong SNMP stats get incremented.

   0 socket(..., SOCK_STREAM|SOCK_NONBLOCK, IPPROTO_TCP) = 3
  +0 connect(3, ..., ...) = -1 EINPROGRESS (Operation now in progress)

  +0 > S  0:0(0) <mss 1460,sackOK,TS val 1000 ecr 0,nop,wscale 8>
  +0 < S  0:0(0) win 1000 <mss 1000>
  +0 > S. 0:0(0) ack 1 <mss 1460,sackOK,TS val 3308134035 ecr 0,nop,wscale 8>
  +0 < S. 0:0(0) ack 1 win 1000

  +0 write(3, ..., 100) = 100
  +0 > P. 1:101(100) ack 1

  --

  # packetdrill cross-synack.pkt
  cross-synack.pkt:13: runtime error in write call: Expected result 100 but got -1 with errno 11 (Resource temporarily unavailable)
  # nstat
  ...
  TcpExtTCPChallengeACK           1                  0.0
  TcpExtTCPSYNChallenge           1                  0.0

The problem is that bpf_skops_established() is triggered by the Challenge
ACK instead of SYN+ACK.  This causes the bpf prog to miss the chance to
check if the peer supports a TCP option that is expected to be exchanged
in SYN and SYN+ACK.

Let's accept a bare SYN+ACK for active-open TCP_SYN_RECV sockets to avoid
such a situation.

Note that tcp_ack_snd_check() in tcp_rcv_state_process() is skipped not to
send an unnecessary ACK, but this could be a bit risky for net.git, so this
targets for net-next.

Link: https://www.rfc-editor.org/rfc/rfc9293.html#section-3.5-7 [0]
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20240710171246.87533-2-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/tcp_input.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index e0f54b9be850..ff9ab3d01ced 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -5998,6 +5998,11 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
 	 * RFC 5961 4.2 : Send a challenge ack
 	 */
 	if (th->syn) {
+		if (sk->sk_state == TCP_SYN_RECV && sk->sk_socket && th->ack &&
+		    TCP_SKB_CB(skb)->seq + 1 == TCP_SKB_CB(skb)->end_seq &&
+		    TCP_SKB_CB(skb)->seq + 1 == tp->rcv_nxt &&
+		    TCP_SKB_CB(skb)->ack_seq == tp->snd_nxt)
+			goto pass;
 syn_challenge:
 		if (syn_inerr)
 			TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS);
@@ -6007,6 +6012,7 @@ syn_challenge:
 		goto discard;
 	}
 
+pass:
 	bpf_skops_parse_hdr(sk, skb);
 
 	return true;
@@ -6813,6 +6819,9 @@ tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb)
 		tcp_fast_path_on(tp);
 		if (sk->sk_shutdown & SEND_SHUTDOWN)
 			tcp_shutdown(sk, SEND_SHUTDOWN);
+
+		if (sk->sk_socket)
+			goto consume;
 		break;
 
 	case TCP_FIN_WAIT1: {
-- 
cgit v1.2.3-58-ga151


From b3bb4d23a41b825c78ff4436c9dc8650b1ea5704 Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Wed, 10 Jul 2024 10:12:46 -0700
Subject: selftests: tcp: Remove broken SNMP assumptions for TCP AO
 self-connect tests.

tcp_ao/self-connect.c checked the following SNMP stats before/after
connect() to confirm that the test exercises the simultaneous connect()
path.

  * TCPChallengeACK
  * TCPSYNChallenge

But the stats should not be counted for self-connect in the first place,
and the assumption is no longer true.

Let's remove the check.

Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Dmitry Safonov <dima@arista.com>
Link: https://patch.msgid.link/20240710171246.87533-3-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/tcp_ao/self-connect.c | 18 ------------------
 1 file changed, 18 deletions(-)

diff --git a/tools/testing/selftests/net/tcp_ao/self-connect.c b/tools/testing/selftests/net/tcp_ao/self-connect.c
index e154d9e198a9..a5698b0a3718 100644
--- a/tools/testing/selftests/net/tcp_ao/self-connect.c
+++ b/tools/testing/selftests/net/tcp_ao/self-connect.c
@@ -30,8 +30,6 @@ static void setup_lo_intf(const char *lo_intf)
 static void tcp_self_connect(const char *tst, unsigned int port,
 			     bool different_keyids, bool check_restore)
 {
-	uint64_t before_challenge_ack, after_challenge_ack;
-	uint64_t before_syn_challenge, after_syn_challenge;
 	struct tcp_ao_counters before_ao, after_ao;
 	uint64_t before_aogood, after_aogood;
 	struct netstat *ns_before, *ns_after;
@@ -62,8 +60,6 @@ static void tcp_self_connect(const char *tst, unsigned int port,
 
 	ns_before = netstat_read();
 	before_aogood = netstat_get(ns_before, "TCPAOGood", NULL);
-	before_challenge_ack = netstat_get(ns_before, "TCPChallengeACK", NULL);
-	before_syn_challenge = netstat_get(ns_before, "TCPSYNChallenge", NULL);
 	if (test_get_tcp_ao_counters(sk, &before_ao))
 		test_error("test_get_tcp_ao_counters()");
 
@@ -82,8 +78,6 @@ static void tcp_self_connect(const char *tst, unsigned int port,
 
 	ns_after = netstat_read();
 	after_aogood = netstat_get(ns_after, "TCPAOGood", NULL);
-	after_challenge_ack = netstat_get(ns_after, "TCPChallengeACK", NULL);
-	after_syn_challenge = netstat_get(ns_after, "TCPSYNChallenge", NULL);
 	if (test_get_tcp_ao_counters(sk, &after_ao))
 		test_error("test_get_tcp_ao_counters()");
 	if (!check_restore) {
@@ -98,18 +92,6 @@ static void tcp_self_connect(const char *tst, unsigned int port,
 		close(sk);
 		return;
 	}
-	if (after_challenge_ack <= before_challenge_ack ||
-	    after_syn_challenge <= before_syn_challenge) {
-		/*
-		 * It's also meant to test simultaneous open, so check
-		 * these counters as well.
-		 */
-		test_fail("%s: Didn't challenge SYN or ACK: %zu <= %zu OR %zu <= %zu",
-			  tst, after_challenge_ack, before_challenge_ack,
-			  after_syn_challenge, before_syn_challenge);
-		close(sk);
-		return;
-	}
 
 	if (test_tcp_ao_counters_cmp(tst, &before_ao, &after_ao, TEST_CNT_GOOD)) {
 		close(sk);
-- 
cgit v1.2.3-58-ga151


From 275a63c9fe10f39066782d2d775d7d3efb20b01f Mon Sep 17 00:00:00 2001
From: Danielle Ratson <danieller@nvidia.com>
Date: Thu, 11 Jul 2024 11:09:34 +0300
Subject: net: ethtool: Monotonically increase the message sequence number

Currently, during the module firmware flashing process, unicast
notifications are sent from the kernel using the same sequence number,
making it impossible for user space to track missed notifications.

Monotonically increase the message sequence number, so the order of
notifications could be tracked effectively.

Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Link: https://patch.msgid.link/20240711080934.2071869-1-danieller@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ethtool/module.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ethtool/module.c b/net/ethtool/module.c
index aba78436d350..6988e07bdcd6 100644
--- a/net/ethtool/module.c
+++ b/net/ethtool/module.c
@@ -488,7 +488,7 @@ ethnl_module_fw_flash_ntf(struct net_device *dev,
 	if (!skb)
 		return;
 
-	hdr = ethnl_unicast_put(skb, ntf_params->portid, ntf_params->seq,
+	hdr = ethnl_unicast_put(skb, ntf_params->portid, ++ntf_params->seq,
 				ETHTOOL_MSG_MODULE_FW_FLASH_NTF);
 	if (!hdr)
 		goto err_skb;
-- 
cgit v1.2.3-58-ga151


From f7023b3d697c6a7dfe2d9c70e0d8c2c580ccbd76 Mon Sep 17 00:00:00 2001
From: Thorsten Blum <thorsten.blum@toblux.com>
Date: Thu, 11 Jul 2024 17:47:43 +0200
Subject: net: mvpp2: Improve data types and use min()

Change the data type of the variable freq in mvpp2_rx_time_coal_set()
and mvpp2_tx_time_coal_set() to u32 because port->priv->tclk also has
the data type u32.

Change the data type of the function parameter clk_hz in
mvpp2_usec_to_cycles() and mvpp2_cycles_to_usec() to u32 accordingly
and remove the following Coccinelle/coccicheck warning reported by
do_div.cocci:

  WARNING: do_div() does a 64-by-32 division, please consider using div64_ul instead

Use min() to simplify the code and improve its readability.

Compile-tested only.

Signed-off-by: Thorsten Blum <thorsten.blum@toblux.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240711154741.174745-1-thorsten.blum@toblux.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
index 9adf4301c9b1..1e52256a9ea8 100644
--- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
+++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
@@ -2766,29 +2766,29 @@ static void mvpp2_tx_pkts_coal_set(struct mvpp2_port *port,
 	}
 }
 
-static u32 mvpp2_usec_to_cycles(u32 usec, unsigned long clk_hz)
+static u32 mvpp2_usec_to_cycles(u32 usec, u32 clk_hz)
 {
 	u64 tmp = (u64)clk_hz * usec;
 
 	do_div(tmp, USEC_PER_SEC);
 
-	return tmp > U32_MAX ? U32_MAX : tmp;
+	return min(tmp, U32_MAX);
 }
 
-static u32 mvpp2_cycles_to_usec(u32 cycles, unsigned long clk_hz)
+static u32 mvpp2_cycles_to_usec(u32 cycles, u32 clk_hz)
 {
 	u64 tmp = (u64)cycles * USEC_PER_SEC;
 
 	do_div(tmp, clk_hz);
 
-	return tmp > U32_MAX ? U32_MAX : tmp;
+	return min(tmp, U32_MAX);
 }
 
 /* Set the time delay in usec before Rx interrupt */
 static void mvpp2_rx_time_coal_set(struct mvpp2_port *port,
 				   struct mvpp2_rx_queue *rxq)
 {
-	unsigned long freq = port->priv->tclk;
+	u32 freq = port->priv->tclk;
 	u32 val = mvpp2_usec_to_cycles(rxq->time_coal, freq);
 
 	if (val > MVPP2_MAX_ISR_RX_THRESHOLD) {
@@ -2804,7 +2804,7 @@ static void mvpp2_rx_time_coal_set(struct mvpp2_port *port,
 
 static void mvpp2_tx_time_coal_set(struct mvpp2_port *port)
 {
-	unsigned long freq = port->priv->tclk;
+	u32 freq = port->priv->tclk;
 	u32 val = mvpp2_usec_to_cycles(port->tx_time_coal, freq);
 
 	if (val > MVPP2_MAX_ISR_TX_THRESHOLD) {
-- 
cgit v1.2.3-58-ga151


From 63c6e08eac8e7e2ad2e883c968e58e23e1dc8c70 Mon Sep 17 00:00:00 2001
From: Daniel Jurgens <danielj@nvidia.com>
Date: Thu, 11 Jul 2024 17:33:07 -0700
Subject: net/mlx5: IFC updates for SF max IO EQs

Expose a new cap sf_eq_usage. The vhca_resource_manager can write this
cap, indicating the SF driver should use max_num_eqs_24b to determine
how many EQs to use.

Will be used in the next patch, to indicate to the SF driver from the PF
that the user has set the max io eqs via devlink. So the SF driver can
later query the proper max eq value from the new cap.

devlink port function set pci/0000:08:00.0/32768 max_io_eqs 32

Signed-off-by: Daniel Jurgens <danielj@nvidia.com>
Reviewed-by: William Tu <witu@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://patch.msgid.link/20240712003310.355106-2-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/mlx5/mlx5_ifc.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index fdad0071d599..360d42f041b0 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -1994,7 +1994,9 @@ struct mlx5_ifc_cmd_hca_cap_2_bits {
 	u8	   migration_tracking_state[0x1];
 	u8	   reserved_at_ca[0x6];
 	u8	   migration_in_chunks[0x1];
-	u8	   reserved_at_d1[0xf];
+	u8	   reserved_at_d1[0x1];
+	u8	   sf_eq_usage[0x1];
+	u8	   reserved_at_d3[0xd];
 
 	u8	   cross_vhca_object_to_object_supported[0x20];
 
-- 
cgit v1.2.3-58-ga151


From 2ece6c72ea040b2bec823b11f8e5bad0d92d8af2 Mon Sep 17 00:00:00 2001
From: Daniel Jurgens <danielj@nvidia.com>
Date: Thu, 11 Jul 2024 17:33:08 -0700
Subject: net/mlx5: Set sf_eq_usage for SF max EQs

When setting max_io_eqs for an SF function also set the sf_eq_usage_cap.
This is to indicate to the SF driver from the PF that the user has set
the max io eqs via devlink. So the SF driver can later query the proper
max eq value from the new cap.

Signed-off-by: Daniel Jurgens <danielj@nvidia.com>
Reviewed-by: William Tu <witu@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://patch.msgid.link/20240712003310.355106-3-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 72949cb85244..099a716f1784 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -4676,6 +4676,9 @@ mlx5_devlink_port_fn_max_io_eqs_set(struct devlink_port *port, u32 max_io_eqs,
 	hca_caps = MLX5_ADDR_OF(query_hca_cap_out, query_ctx, capability);
 	MLX5_SET(cmd_hca_cap_2, hca_caps, max_num_eqs_24b, max_eqs);
 
+	if (mlx5_esw_is_sf_vport(esw, vport_num))
+		MLX5_SET(cmd_hca_cap_2, hca_caps, sf_eq_usage, 1);
+
 	err = mlx5_vport_set_other_func_cap(esw->dev, hca_caps, vport_num,
 					    MLX5_SET_HCA_CAP_OP_MOD_GENERAL_DEVICE2);
 	if (err)
-- 
cgit v1.2.3-58-ga151


From 20d80b95a7e4adbd1963c152cf79a58a93b4f577 Mon Sep 17 00:00:00 2001
From: Daniel Jurgens <danielj@nvidia.com>
Date: Thu, 11 Jul 2024 17:33:09 -0700
Subject: net/mlx5: Set default max eqs for SFs

If the user hasn't configured max_io_eqs set a low default. The SF
driver shouldn't try to create more than this, but FW will enforce this
limit.

Signed-off-by: Daniel Jurgens <danielj@nvidia.com>
Reviewed-by: William Tu <witu@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://patch.msgid.link/20240712003310.355106-4-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h          |  3 +++
 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 12 +++++++++++-
 drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c       | 12 ++++++++++++
 3 files changed, 26 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 88745dc6aed5..578466d69f21 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -223,6 +223,7 @@ struct mlx5_vport {
 
 	u16 vport;
 	bool                    enabled;
+	bool max_eqs_set;
 	enum mlx5_eswitch_vport_event enabled_events;
 	int index;
 	struct mlx5_devlink_port *dl_port;
@@ -579,6 +580,8 @@ int mlx5_devlink_port_fn_max_io_eqs_get(struct devlink_port *port,
 int mlx5_devlink_port_fn_max_io_eqs_set(struct devlink_port *port,
 					u32 max_io_eqs,
 					struct netlink_ext_ack *extack);
+int mlx5_devlink_port_fn_max_io_eqs_set_sf_default(struct devlink_port *port,
+						   struct netlink_ext_ack *extack);
 
 void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw, u8 rep_type);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 099a716f1784..768199d2255a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -68,6 +68,7 @@
 #define MLX5_ESW_FT_OFFLOADS_DROP_RULE (1)
 
 #define MLX5_ESW_MAX_CTRL_EQS 4
+#define MLX5_ESW_DEFAULT_SF_COMP_EQS 8
 
 static struct esw_vport_tbl_namespace mlx5_esw_vport_tbl_mirror_ns = {
 	.max_fte = MLX5_ESW_VPORT_TBL_SIZE,
@@ -4683,9 +4684,18 @@ mlx5_devlink_port_fn_max_io_eqs_set(struct devlink_port *port, u32 max_io_eqs,
 					    MLX5_SET_HCA_CAP_OP_MOD_GENERAL_DEVICE2);
 	if (err)
 		NL_SET_ERR_MSG_MOD(extack, "Failed setting HCA caps");
-
+	vport->max_eqs_set = true;
 out:
 	mutex_unlock(&esw->state_lock);
 	kfree(query_ctx);
 	return err;
 }
+
+int
+mlx5_devlink_port_fn_max_io_eqs_set_sf_default(struct devlink_port *port,
+					       struct netlink_ext_ack *extack)
+{
+	return mlx5_devlink_port_fn_max_io_eqs_set(port,
+						   MLX5_ESW_DEFAULT_SF_COMP_EQS,
+						   extack);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
index 6c11e075cab0..a96be98be032 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
@@ -161,6 +161,7 @@ int mlx5_devlink_sf_port_fn_state_get(struct devlink_port *dl_port,
 static int mlx5_sf_activate(struct mlx5_core_dev *dev, struct mlx5_sf *sf,
 			    struct netlink_ext_ack *extack)
 {
+	struct mlx5_vport *vport;
 	int err;
 
 	if (mlx5_sf_is_active(sf))
@@ -170,6 +171,13 @@ static int mlx5_sf_activate(struct mlx5_core_dev *dev, struct mlx5_sf *sf,
 		return -EBUSY;
 	}
 
+	vport = mlx5_devlink_port_vport_get(&sf->dl_port.dl_port);
+	if (!vport->max_eqs_set && MLX5_CAP_GEN_2(dev, max_num_eqs_24b)) {
+		err = mlx5_devlink_port_fn_max_io_eqs_set_sf_default(&sf->dl_port.dl_port,
+								     extack);
+		if (err)
+			return err;
+	}
 	err = mlx5_cmd_sf_enable_hca(dev, sf->hw_fn_id);
 	if (err)
 		return err;
@@ -318,7 +326,11 @@ int mlx5_devlink_sf_port_new(struct devlink *devlink,
 
 static void mlx5_sf_dealloc(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 {
+	struct mlx5_vport *vport;
+
 	mutex_lock(&table->sf_state_lock);
+	vport = mlx5_devlink_port_vport_get(&sf->dl_port.dl_port);
+	vport->max_eqs_set = false;
 
 	mlx5_sf_function_id_erase(table, sf);
 
-- 
cgit v1.2.3-58-ga151


From 4b66be76a6fbe16918a01439bb9023da154d7694 Mon Sep 17 00:00:00 2001
From: Daniel Jurgens <danielj@nvidia.com>
Date: Thu, 11 Jul 2024 17:33:10 -0700
Subject: net/mlx5: Use set number of max EQs

If a maximum number of EQs has been set for an SF, use that amount.

Signed-off-by: Daniel Jurgens <danielj@nvidia.com>
Reviewed-by: William Tu <witu@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://patch.msgid.link/20240712003310.355106-5-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c      |  7 +++++--
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 12 ++++--------
 2 files changed, 9 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index ac1565c0c8af..4326aa42bf2d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -1187,7 +1187,6 @@ static int get_num_eqs(struct mlx5_core_dev *dev)
 {
 	struct mlx5_eq_table *eq_table = dev->priv.eq_table;
 	int max_dev_eqs;
-	int max_eqs_sf;
 	int num_eqs;
 
 	/* If ethernet is disabled we use just a single completion vector to
@@ -1202,7 +1201,11 @@ static int get_num_eqs(struct mlx5_core_dev *dev)
 	num_eqs = min_t(int, mlx5_irq_table_get_num_comp(eq_table->irq_table),
 			max_dev_eqs - MLX5_MAX_ASYNC_EQS);
 	if (mlx5_core_is_sf(dev)) {
-		max_eqs_sf = min_t(int, MLX5_COMP_EQS_PER_SF,
+		int max_eqs_sf = MLX5_CAP_GEN_2(dev, sf_eq_usage) ?
+				 MLX5_CAP_GEN_2(dev, max_num_eqs_24b) :
+				 MLX5_COMP_EQS_PER_SF;
+
+		max_eqs_sf = min_t(int, max_eqs_sf,
 				   mlx5_irq_table_get_sfs_vec(eq_table->irq_table));
 		num_eqs = min_t(int, num_eqs, max_eqs_sf);
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index 401d39069680..86208b86eea8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -16,6 +16,7 @@
 #endif
 
 #define MLX5_SFS_PER_CTRL_IRQ 64
+#define MLX5_MAX_MSIX_PER_SF 256
 #define MLX5_IRQ_CTRL_SF_MAX 8
 /* min num of vectors for SFs to be enabled */
 #define MLX5_IRQ_VEC_COMP_BASE_SF 2
@@ -589,8 +590,6 @@ static void irq_pool_free(struct mlx5_irq_pool *pool)
 static int irq_pools_init(struct mlx5_core_dev *dev, int sf_vec, int pcif_vec)
 {
 	struct mlx5_irq_table *table = dev->priv.irq_table;
-	int num_sf_ctrl_by_msix;
-	int num_sf_ctrl_by_sfs;
 	int num_sf_ctrl;
 	int err;
 
@@ -608,10 +607,8 @@ static int irq_pools_init(struct mlx5_core_dev *dev, int sf_vec, int pcif_vec)
 	}
 
 	/* init sf_ctrl_pool */
-	num_sf_ctrl_by_msix = DIV_ROUND_UP(sf_vec, MLX5_COMP_EQS_PER_SF);
-	num_sf_ctrl_by_sfs = DIV_ROUND_UP(mlx5_sf_max_functions(dev),
-					  MLX5_SFS_PER_CTRL_IRQ);
-	num_sf_ctrl = min_t(int, num_sf_ctrl_by_msix, num_sf_ctrl_by_sfs);
+	num_sf_ctrl = DIV_ROUND_UP(mlx5_sf_max_functions(dev),
+				   MLX5_SFS_PER_CTRL_IRQ);
 	num_sf_ctrl = min_t(int, MLX5_IRQ_CTRL_SF_MAX, num_sf_ctrl);
 	table->sf_ctrl_pool = irq_pool_alloc(dev, pcif_vec, num_sf_ctrl,
 					     "mlx5_sf_ctrl",
@@ -726,8 +723,7 @@ int mlx5_irq_table_create(struct mlx5_core_dev *dev)
 
 	total_vec = pcif_vec;
 	if (mlx5_sf_max_functions(dev))
-		total_vec += MLX5_IRQ_CTRL_SF_MAX +
-			MLX5_COMP_EQS_PER_SF * mlx5_sf_max_functions(dev);
+		total_vec += MLX5_MAX_MSIX_PER_SF * mlx5_sf_max_functions(dev);
 	total_vec = min_t(int, total_vec, pci_msix_vec_count(dev->pdev));
 	pcif_vec = min_t(int, pcif_vec, pci_msix_vec_count(dev->pdev));
 
-- 
cgit v1.2.3-58-ga151


From 464b99e77b3d024a319db0dabe917d04824039db Mon Sep 17 00:00:00 2001
From: "Matthieu Baerts (NGI0)" <matttbe@kernel.org>
Date: Fri, 12 Jul 2024 12:00:15 +0200
Subject: selftests: mptcp: lib: fix shellcheck errors

It looks like we missed these two errors recently:

  - SC2068: Double quote array expansions to avoid re-splitting elements.
  - SC2145: Argument mixes string and array. Use * or separate argument.

Two simple fixes, it is not supposed to change the behaviour as the
variable names should not have any spaces in their names. Still, better
to fix them to easily spot new issues.

Fixes: f265d3119a29 ("selftests: mptcp: lib: use setup/cleanup_ns helpers")
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20240712-upstream-net-next-20240712-selftests-mptcp-fix-shellcheck-v1-1-1cb7180db40a@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/mptcp/mptcp_lib.sh | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/net/mptcp/mptcp_lib.sh b/tools/testing/selftests/net/mptcp/mptcp_lib.sh
index 194c8fc2e55a..438280e68434 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_lib.sh
+++ b/tools/testing/selftests/net/mptcp/mptcp_lib.sh
@@ -428,8 +428,8 @@ mptcp_lib_check_tools() {
 }
 
 mptcp_lib_ns_init() {
-	if ! setup_ns ${@}; then
-		mptcp_lib_pr_fail "Failed to setup namespace ${@}"
+	if ! setup_ns "${@}"; then
+		mptcp_lib_pr_fail "Failed to setup namespaces ${*}"
 		exit ${KSFT_FAIL}
 	fi
 
-- 
cgit v1.2.3-58-ga151


From 3771266bf8418904794175fdc83aaf4732114cba Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Thu, 11 Jul 2024 15:37:22 -0700
Subject: eth: mlx5: expose NETIF_F_NTUPLE when ARFS is compiled out

ARFS depends on NTUPLE filters, but the inverse is not true.
Drivers which don't support ARFS commonly still support NTUPLE
filtering. mlx5 has a Kconfig option to disable ARFS (MLX5_EN_ARFS)
and does not advertise NTUPLE filters as a feature at all when ARFS
is compiled out. That's not correct, ntuple filters indeed still work
just fine (as long as MLX5_EN_RXNFC is enabled).

This is needed to make the RSS test not skip all RSS context
related testing.

Acked-by: Saeed Mahameed <saeedm@nvidia.com>
Link: https://patch.msgid.link/20240711223722.297676-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/fs.h       | 13 +++++++++++++
 drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c  |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_fs.c       |  5 ++---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c     |  8 +++++---
 drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c |  8 +++-----
 5 files changed, 24 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h b/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h
index 4d6225e0eec7..1e8b7d330701 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/fs.h
@@ -154,6 +154,19 @@ struct mlx5e_tc_table *mlx5e_fs_get_tc(struct mlx5e_flow_steering *fs);
 struct mlx5e_l2_table *mlx5e_fs_get_l2(struct mlx5e_flow_steering *fs);
 struct mlx5_flow_namespace *mlx5e_fs_get_ns(struct mlx5e_flow_steering *fs, bool egress);
 void mlx5e_fs_set_ns(struct mlx5e_flow_steering *fs, struct mlx5_flow_namespace *ns, bool egress);
+
+static inline bool mlx5e_fs_has_arfs(struct net_device *netdev)
+{
+	return IS_ENABLED(CONFIG_MLX5_EN_ARFS) &&
+		netdev->hw_features & NETIF_F_NTUPLE;
+}
+
+static inline bool mlx5e_fs_want_arfs(struct net_device *netdev)
+{
+	return IS_ENABLED(CONFIG_MLX5_EN_ARFS) &&
+		netdev->features & NETIF_F_NTUPLE;
+}
+
 #ifdef CONFIG_MLX5_EN_RXNFC
 struct mlx5e_ethtool_steering *mlx5e_fs_get_ethtool(struct mlx5e_flow_steering *fs);
 #endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 3320f12ba2db..5582c93a62f1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -525,7 +525,7 @@ int mlx5e_ethtool_set_channels(struct mlx5e_priv *priv,
 
 	opened = test_bit(MLX5E_STATE_OPENED, &priv->state);
 
-	arfs_enabled = opened && (priv->netdev->features & NETIF_F_NTUPLE);
+	arfs_enabled = opened && mlx5e_fs_want_arfs(priv->netdev);
 	if (arfs_enabled)
 		mlx5e_arfs_disable(priv->fs);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
index 8c5b291a171f..05058710d2c7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c
@@ -1307,8 +1307,7 @@ int mlx5e_create_flow_steering(struct mlx5e_flow_steering *fs,
 		return -EOPNOTSUPP;
 
 	mlx5e_fs_set_ns(fs, ns, false);
-	err = mlx5e_arfs_create_tables(fs, rx_res,
-				       !!(netdev->hw_features & NETIF_F_NTUPLE));
+	err = mlx5e_arfs_create_tables(fs, rx_res, mlx5e_fs_has_arfs(netdev));
 	if (err) {
 		fs_err(fs, "Failed to create arfs tables, err=%d\n", err);
 		netdev->hw_features &= ~NETIF_F_NTUPLE;
@@ -1355,7 +1354,7 @@ err_destroy_ttc_table:
 err_destroy_inner_ttc_table:
 	mlx5e_destroy_inner_ttc_table(fs);
 err_destroy_arfs_tables:
-	mlx5e_arfs_destroy_tables(fs, !!(netdev->hw_features & NETIF_F_NTUPLE));
+	mlx5e_arfs_destroy_tables(fs, mlx5e_fs_has_arfs(netdev));
 
 	return err;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index ff335527c10a..6f686fabed44 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -5556,8 +5556,10 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
 #if IS_ENABLED(CONFIG_MLX5_CLS_ACT)
 		netdev->hw_features      |= NETIF_F_HW_TC;
 #endif
-#ifdef CONFIG_MLX5_EN_ARFS
+#if IS_ENABLED(CONFIG_MLX5_EN_ARFS)
 		netdev->hw_features	 |= NETIF_F_NTUPLE;
+#elif IS_ENABLED(CONFIG_MLX5_EN_RXNFC)
+		netdev->features	 |= NETIF_F_NTUPLE;
 #endif
 	}
 
@@ -5731,7 +5733,7 @@ static int mlx5e_init_nic_rx(struct mlx5e_priv *priv)
 err_tc_nic_cleanup:
 	mlx5e_tc_nic_cleanup(priv);
 err_destroy_flow_steering:
-	mlx5e_destroy_flow_steering(priv->fs, !!(priv->netdev->hw_features & NETIF_F_NTUPLE),
+	mlx5e_destroy_flow_steering(priv->fs, mlx5e_fs_has_arfs(priv->netdev),
 				    priv->profile);
 err_destroy_rx_res:
 	mlx5e_rx_res_destroy(priv->rx_res);
@@ -5747,7 +5749,7 @@ static void mlx5e_cleanup_nic_rx(struct mlx5e_priv *priv)
 {
 	mlx5e_accel_cleanup_rx(priv);
 	mlx5e_tc_nic_cleanup(priv);
-	mlx5e_destroy_flow_steering(priv->fs, !!(priv->netdev->hw_features & NETIF_F_NTUPLE),
+	mlx5e_destroy_flow_steering(priv->fs, mlx5e_fs_has_arfs(priv->netdev),
 				    priv->profile);
 	mlx5e_rx_res_destroy(priv->rx_res);
 	priv->rx_res = NULL;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
index 8e0404c0d1ca..0979d672d47f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c
@@ -372,7 +372,7 @@ static int mlx5i_create_flow_steering(struct mlx5e_priv *priv)
 
 	mlx5e_fs_set_ns(priv->fs, ns, false);
 	err = mlx5e_arfs_create_tables(priv->fs, priv->rx_res,
-				       !!(priv->netdev->hw_features & NETIF_F_NTUPLE));
+				       mlx5e_fs_has_arfs(priv->netdev));
 	if (err) {
 		netdev_err(priv->netdev, "Failed to create arfs tables, err=%d\n",
 			   err);
@@ -391,8 +391,7 @@ static int mlx5i_create_flow_steering(struct mlx5e_priv *priv)
 	return 0;
 
 err_destroy_arfs_tables:
-	mlx5e_arfs_destroy_tables(priv->fs,
-				  !!(priv->netdev->hw_features & NETIF_F_NTUPLE));
+	mlx5e_arfs_destroy_tables(priv->fs, mlx5e_fs_has_arfs(priv->netdev));
 
 	return err;
 }
@@ -400,8 +399,7 @@ err_destroy_arfs_tables:
 static void mlx5i_destroy_flow_steering(struct mlx5e_priv *priv)
 {
 	mlx5e_destroy_ttc_table(priv->fs);
-	mlx5e_arfs_destroy_tables(priv->fs,
-				  !!(priv->netdev->hw_features & NETIF_F_NTUPLE));
+	mlx5e_arfs_destroy_tables(priv->fs, mlx5e_fs_has_arfs(priv->netdev));
 	mlx5e_ethtool_cleanup_steering(priv->fs);
 }
 
-- 
cgit v1.2.3-58-ga151


From e7cdef626f1d7ca69eb3351a8c4ae6a8ec46feed Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Fri, 12 Jul 2024 06:48:16 -0700
Subject: net: dpaa: Fix compilation Warning

Remove variables that are defined and incremented but never read.
This issue appeared in network tests[1] as:

	drivers/net/ethernet/freescale/dpaa/dpaa_eth_sysfs.c:38:6: warning: variable 'i' set but not used [-Wunused-but-set-variable]
	38 |         int i = 0;
	   |             ^

Link: https://netdev.bots.linux.dev/static/nipa/870263/13729811/build_clang/stderr [1]
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://patch.msgid.link/20240712134817.913756-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/dpaa/dpaa_eth_sysfs.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sysfs.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sysfs.c
index 4fee74c024bd..aad470e9caea 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sysfs.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_sysfs.c
@@ -35,7 +35,6 @@ static ssize_t dpaa_eth_show_fqids(struct device *dev,
 	u32 last_fqid = 0;
 	ssize_t bytes = 0;
 	char *str;
-	int i = 0;
 
 	list_for_each_entry_safe(fq, tmp, &priv->dpaa_fq_list, list) {
 		switch (fq->fq_type) {
@@ -85,7 +84,6 @@ static ssize_t dpaa_eth_show_fqids(struct device *dev,
 
 		prev = fq;
 		prevstr = str;
-		i++;
 	}
 
 	if (prev) {
-- 
cgit v1.2.3-58-ga151


From 0cda1acfa235f2ee75597e7b378de54a91d69e99 Mon Sep 17 00:00:00 2001
From: Oleksij Rempel <o.rempel@pengutronix.de>
Date: Fri, 12 Jul 2024 17:28:48 +0200
Subject: net: phy: dp83td510: add cable testing support

This patch implements the TDR test procedure as described in
"Application Note DP83TD510E Cable Diagnostics Toolkit revC", section 3.2.

The procedure was tested with "draka 08 signalkabel 2x0.8mm". The reported
cable length was 5 meters more for each 20 meters of actual cable length.
For instance, a 20-meter cable showed as 25 meters, and a 40-meter cable
showed as 50 meters. Since other parts of the diagnostics provided by this
PHY (e.g., Active Link Cable Diagnostics) require accurate cable
characterization to provide proper results, this tuning can be implemented
in a separate patch/interface.

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
changes v2:
- add comments
- change post silence time to 1000ms
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20240712152848.2479912-1-o.rempel@pengutronix.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/dp83td510.c | 264 ++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 264 insertions(+)

diff --git a/drivers/net/phy/dp83td510.c b/drivers/net/phy/dp83td510.c
index d7616b13c594..551e37786c2d 100644
--- a/drivers/net/phy/dp83td510.c
+++ b/drivers/net/phy/dp83td510.c
@@ -4,6 +4,7 @@
  */
 
 #include <linux/bitfield.h>
+#include <linux/ethtool_netlink.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/phy.h>
@@ -29,6 +30,10 @@
 #define DP83TD510E_INT1_LINK			BIT(13)
 #define DP83TD510E_INT1_LINK_EN			BIT(5)
 
+#define DP83TD510E_CTRL				0x1f
+#define DP83TD510E_CTRL_HW_RESET		BIT(15)
+#define DP83TD510E_CTRL_SW_RESET		BIT(14)
+
 #define DP83TD510E_AN_STAT_1			0x60c
 #define DP83TD510E_MASTER_SLAVE_RESOL_FAIL	BIT(15)
 
@@ -53,6 +58,117 @@ static const u16 dp83td510_mse_sqi_map[] = {
 	0x0000  /* 24dB =< SNR */
 };
 
+/* Time Domain Reflectometry (TDR) Functionality of DP83TD510 PHY
+ *
+ * I assume that this PHY is using a variation of Spread Spectrum Time Domain
+ * Reflectometry (SSTDR) rather than the commonly used TDR found in many PHYs.
+ * Here are the following observations which likely confirm this:
+ * - The DP83TD510 PHY transmits a modulated signal of configurable length
+ *   (default 16000 µs) instead of a single pulse pattern, which is typical
+ *   for traditional TDR.
+ * - The pulse observed on the wire, triggered by the HW RESET register, is not
+ *   part of the cable testing process.
+ *
+ * I assume that SSTDR seems to be a logical choice for the 10BaseT1L
+ * environment due to improved noise resistance, making it suitable for
+ * environments  with significant electrical noise, such as long 10BaseT1L cable
+ * runs.
+ *
+ * Configuration Variables:
+ * The SSTDR variation used in this PHY involves more configuration variables
+ * that can dramatically affect the functionality and precision of cable
+ * testing. Since most of  these configuration options are either not well
+ * documented or documented with minimal details, the following sections
+ * describe my understanding and observations of these variables and their
+ * impact on TDR functionality.
+ *
+ * Timeline:
+ *     ,<--cfg_pre_silence_time
+ *     |            ,<-SSTDR Modulated Transmission
+ *     |	    |            ,<--cfg_post_silence_time
+ *     |	    |            |             ,<--Force Link Mode
+ * |<--'-->|<-------'------->|<--'-->|<--------'------->|
+ *
+ * - cfg_pre_silence_time: Optional silence time before TDR transmission starts.
+ * - SSTDR Modulated Transmission: Transmission duration configured by
+ *   cfg_tdr_tx_duration and amplitude configured by cfg_tdr_tx_type.
+ * - cfg_post_silence_time: Silence time after TDR transmission.
+ * - Force Link Mode: If nothing is configured after cfg_post_silence_time,
+ *   the PHY continues in force link mode without autonegotiation.
+ */
+
+#define DP83TD510E_TDR_CFG				0x1e
+#define DP83TD510E_TDR_START				BIT(15)
+#define DP83TD510E_TDR_DONE				BIT(1)
+#define DP83TD510E_TDR_FAIL				BIT(0)
+
+#define DP83TD510E_TDR_CFG1				0x300
+/* cfg_tdr_tx_type: Transmit voltage level for TDR.
+ * 0 = 1V, 1 = 2.4V
+ * Note: Using different voltage levels may not work
+ * in all configuration variations. For example, setting
+ * 2.4V may give different cable length measurements.
+ * Other settings may be needed to make it work properly.
+ */
+#define DP83TD510E_TDR_TX_TYPE				BIT(12)
+#define DP83TD510E_TDR_TX_TYPE_1V			0
+#define DP83TD510E_TDR_TX_TYPE_2_4V			1
+/* cfg_post_silence_time: Time after the TDR sequence. Since we force master mode
+ * for the TDR will proceed with forced link state after this time. For Linux
+ * it is better to set max value to avoid false link state detection.
+ */
+#define DP83TD510E_TDR_CFG1_POST_SILENCE_TIME		GENMASK(3, 2)
+#define DP83TD510E_TDR_CFG1_POST_SILENCE_TIME_0MS	0
+#define DP83TD510E_TDR_CFG1_POST_SILENCE_TIME_10MS	1
+#define DP83TD510E_TDR_CFG1_POST_SILENCE_TIME_100MS	2
+#define DP83TD510E_TDR_CFG1_POST_SILENCE_TIME_1000MS	3
+/* cfg_pre_silence_time: Time before the TDR sequence. It should be enough to
+ * settle down all pulses and reflections. Since for 10BASE-T1L we have
+ * maximum 2000m cable length, we can set it to 1ms.
+ */
+#define DP83TD510E_TDR_CFG1_PRE_SILENCE_TIME		GENMASK(1, 0)
+#define DP83TD510E_TDR_CFG1_PRE_SILENCE_TIME_0MS	0
+#define DP83TD510E_TDR_CFG1_PRE_SILENCE_TIME_10MS	1
+#define DP83TD510E_TDR_CFG1_PRE_SILENCE_TIME_100MS	2
+#define DP83TD510E_TDR_CFG1_PRE_SILENCE_TIME_1000MS	3
+
+#define DP83TD510E_TDR_CFG2				0x301
+#define DP83TD510E_TDR_END_TAP_INDEX_1			GENMASK(14, 8)
+#define DP83TD510E_TDR_END_TAP_INDEX_1_DEF		36
+#define DP83TD510E_TDR_START_TAP_INDEX_1		GENMASK(6, 0)
+#define DP83TD510E_TDR_START_TAP_INDEX_1_DEF		4
+
+#define DP83TD510E_TDR_CFG3				0x302
+/* cfg_tdr_tx_duration: Duration of the TDR transmission in microseconds.
+ * This value sets the duration of the modulated signal used for TDR
+ * measurements.
+ * - Default: 16000 µs
+ * - Observation: A minimum duration of 6000 µs is recommended to ensure
+ *   accurate detection of cable faults. Durations shorter than 6000 µs may
+ *   result in incomplete data, especially for shorter cables (e.g., 20 meters),
+ *   leading to false "OK" results. Longer durations (e.g., 6000 µs or more)
+ *   provide better accuracy, particularly for detecting open circuits.
+ */
+#define DP83TD510E_TDR_TX_DURATION_US			GENMASK(15, 0)
+#define DP83TD510E_TDR_TX_DURATION_US_DEF		16000
+
+#define DP83TD510E_TDR_FAULT_CFG1			0x303
+#define DP83TD510E_TDR_FLT_LOC_OFFSET_1			GENMASK(14, 8)
+#define DP83TD510E_TDR_FLT_LOC_OFFSET_1_DEF		4
+#define DP83TD510E_TDR_FLT_INIT_1			GENMASK(7, 0)
+#define DP83TD510E_TDR_FLT_INIT_1_DEF			62
+
+#define DP83TD510E_TDR_FAULT_STAT			0x30c
+#define DP83TD510E_TDR_PEAK_DETECT			BIT(11)
+#define DP83TD510E_TDR_PEAK_SIGN			BIT(10)
+#define DP83TD510E_TDR_PEAK_LOCATION			GENMASK(9, 0)
+
+/* Not documented registers and values but recommended according to
+ * "DP83TD510E Cable Diagnostics Toolkit revC"
+ */
+#define DP83TD510E_UNKN_030E				0x30e
+#define DP83TD510E_030E_VAL				0x2520
+
 static int dp83td510_config_intr(struct phy_device *phydev)
 {
 	int ret;
@@ -198,6 +314,151 @@ static int dp83td510_get_sqi_max(struct phy_device *phydev)
 	return DP83TD510_SQI_MAX;
 }
 
+/**
+ * dp83td510_cable_test_start - Start the cable test for the DP83TD510 PHY.
+ * @phydev: Pointer to the phy_device structure.
+ *
+ * This sequence is implemented according to the "Application Note DP83TD510E
+ * Cable Diagnostics Toolkit revC".
+ *
+ * Returns: 0 on success, a negative error code on failure.
+ */
+static int dp83td510_cable_test_start(struct phy_device *phydev)
+{
+	int ret;
+
+	ret = phy_set_bits_mmd(phydev, MDIO_MMD_VEND2, DP83TD510E_CTRL,
+			       DP83TD510E_CTRL_HW_RESET);
+	if (ret)
+		return ret;
+
+	ret = genphy_c45_an_disable_aneg(phydev);
+	if (ret)
+		return ret;
+
+	/* Force master mode */
+	ret = phy_set_bits_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_PMA_PMD_BT1_CTRL,
+			       MDIO_PMA_PMD_BT1_CTRL_CFG_MST);
+	if (ret)
+		return ret;
+
+	/* There is no official recommendation for this register, but it is
+	 * better to use 1V for TDR since other values seems to be optimized
+	 * for this amplitude. Except of amplitude, it is better to configure
+	 * pre TDR silence time to 10ms to avoid false reflections (value 0
+	 * seems to be too short, otherwise we need to implement own silence
+	 * time). Also, post TDR silence time should be set to 1000ms to avoid
+	 * false link state detection, it fits to the polling time of the
+	 * PHY framework. The idea is to wait until
+	 * dp83td510_cable_test_get_status() will be called and reconfigure
+	 * the PHY to the default state within the post silence time window.
+	 */
+	ret = phy_modify_mmd(phydev, MDIO_MMD_VEND2, DP83TD510E_TDR_CFG1,
+			     DP83TD510E_TDR_TX_TYPE |
+			     DP83TD510E_TDR_CFG1_POST_SILENCE_TIME |
+			     DP83TD510E_TDR_CFG1_PRE_SILENCE_TIME,
+			     DP83TD510E_TDR_TX_TYPE_1V |
+			     DP83TD510E_TDR_CFG1_PRE_SILENCE_TIME_10MS |
+			     DP83TD510E_TDR_CFG1_POST_SILENCE_TIME_1000MS);
+	if (ret)
+		return ret;
+
+	ret = phy_write_mmd(phydev, MDIO_MMD_VEND2, DP83TD510E_TDR_CFG2,
+			    FIELD_PREP(DP83TD510E_TDR_END_TAP_INDEX_1,
+				       DP83TD510E_TDR_END_TAP_INDEX_1_DEF) |
+			    FIELD_PREP(DP83TD510E_TDR_START_TAP_INDEX_1,
+				       DP83TD510E_TDR_START_TAP_INDEX_1_DEF));
+	if (ret)
+		return ret;
+
+	ret = phy_write_mmd(phydev, MDIO_MMD_VEND2, DP83TD510E_TDR_FAULT_CFG1,
+			    FIELD_PREP(DP83TD510E_TDR_FLT_LOC_OFFSET_1,
+				       DP83TD510E_TDR_FLT_LOC_OFFSET_1_DEF) |
+			    FIELD_PREP(DP83TD510E_TDR_FLT_INIT_1,
+				       DP83TD510E_TDR_FLT_INIT_1_DEF));
+	if (ret)
+		return ret;
+
+	/* Undocumented register, from the "Application Note DP83TD510E Cable
+	 * Diagnostics Toolkit revC".
+	 */
+	ret = phy_write_mmd(phydev, MDIO_MMD_VEND2, DP83TD510E_UNKN_030E,
+			    DP83TD510E_030E_VAL);
+	if (ret)
+		return ret;
+
+	ret = phy_write_mmd(phydev, MDIO_MMD_VEND2, DP83TD510E_TDR_CFG3,
+			    DP83TD510E_TDR_TX_DURATION_US_DEF);
+	if (ret)
+		return ret;
+
+	ret = phy_set_bits_mmd(phydev, MDIO_MMD_VEND2, DP83TD510E_CTRL,
+			       DP83TD510E_CTRL_SW_RESET);
+	if (ret)
+		return ret;
+
+	return phy_set_bits_mmd(phydev, MDIO_MMD_VEND2, DP83TD510E_TDR_CFG,
+				DP83TD510E_TDR_START);
+}
+
+/**
+ * dp83td510_cable_test_get_status - Get the status of the cable test for the
+ *                                   DP83TD510 PHY.
+ * @phydev: Pointer to the phy_device structure.
+ * @finished: Pointer to a boolean that indicates whether the test is finished.
+ *
+ * The function sets the @finished flag to true if the test is complete.
+ *
+ * Returns: 0 on success or a negative error code on failure.
+ */
+static int dp83td510_cable_test_get_status(struct phy_device *phydev,
+					   bool *finished)
+{
+	int ret, stat;
+
+	*finished = false;
+
+	ret = phy_read_mmd(phydev, MDIO_MMD_VEND2, DP83TD510E_TDR_CFG);
+	if (ret < 0)
+		return ret;
+
+	if (!(ret & DP83TD510E_TDR_DONE))
+		return 0;
+
+	if (!(ret & DP83TD510E_TDR_FAIL)) {
+		int location;
+
+		ret = phy_read_mmd(phydev, MDIO_MMD_VEND2,
+				   DP83TD510E_TDR_FAULT_STAT);
+		if (ret < 0)
+			return ret;
+
+		if (ret & DP83TD510E_TDR_PEAK_DETECT) {
+			if (ret & DP83TD510E_TDR_PEAK_SIGN)
+				stat = ETHTOOL_A_CABLE_RESULT_CODE_OPEN;
+			else
+				stat = ETHTOOL_A_CABLE_RESULT_CODE_SAME_SHORT;
+
+			location = FIELD_GET(DP83TD510E_TDR_PEAK_LOCATION,
+					     ret) * 100;
+			ethnl_cable_test_fault_length(phydev,
+						      ETHTOOL_A_CABLE_PAIR_A,
+						      location);
+		} else {
+			stat = ETHTOOL_A_CABLE_RESULT_CODE_OK;
+		}
+	} else {
+		/* Most probably we have active link partner */
+		stat = ETHTOOL_A_CABLE_RESULT_CODE_UNSPEC;
+	}
+
+	*finished = true;
+
+	ethnl_cable_test_result(phydev, ETHTOOL_A_CABLE_PAIR_A, stat);
+
+	return phy_init_hw(phydev);
+}
+
 static int dp83td510_get_features(struct phy_device *phydev)
 {
 	/* This PHY can't respond on MDIO bus if no RMII clock is enabled.
@@ -221,6 +482,7 @@ static struct phy_driver dp83td510_driver[] = {
 	PHY_ID_MATCH_MODEL(DP83TD510E_PHY_ID),
 	.name		= "TI DP83TD510E",
 
+	.flags          = PHY_POLL_CABLE_TEST,
 	.config_aneg	= dp83td510_config_aneg,
 	.read_status	= dp83td510_read_status,
 	.get_features	= dp83td510_get_features,
@@ -228,6 +490,8 @@ static struct phy_driver dp83td510_driver[] = {
 	.handle_interrupt = dp83td510_handle_interrupt,
 	.get_sqi	= dp83td510_get_sqi,
 	.get_sqi_max	= dp83td510_get_sqi_max,
+	.cable_test_start = dp83td510_cable_test_start,
+	.cable_test_get_status = dp83td510_cable_test_get_status,
 
 	.suspend	= genphy_suspend,
 	.resume		= genphy_resume,
-- 
cgit v1.2.3-58-ga151


From ba88b47816a2089cd270a38fc002bc7df13f8ebe Mon Sep 17 00:00:00 2001
From: Easwar Hariharan <eahariha@linux.microsoft.com>
Date: Thu, 11 Jul 2024 05:27:32 +0000
Subject: sfc: falcon: Make I2C terminology more inclusive

I2C v7, SMBus 3.2, and I3C 1.1.1 specifications have replaced "master/slave"
with more appropriate terms. Inspired by Wolfram's series to fix drivers/i2c/,
fix the terminology for users of I2C_ALGOBIT bitbanging interface, now that
the approved verbiage exists in the specification.

Reviewed-by: Martin Habets <habetsm.xilinx@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Easwar Hariharan <eahariha@linux.microsoft.com>
Link: https://patch.msgid.link/20240711052734.1273652-5-eahariha@linux.microsoft.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/sfc/falcon/falcon.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/sfc/falcon/falcon.c b/drivers/net/ethernet/sfc/falcon/falcon.c
index 7a1c9337081b..36114ce88034 100644
--- a/drivers/net/ethernet/sfc/falcon/falcon.c
+++ b/drivers/net/ethernet/sfc/falcon/falcon.c
@@ -367,7 +367,7 @@ static const struct i2c_algo_bit_data falcon_i2c_bit_operations = {
 	.getsda		= falcon_getsda,
 	.getscl		= falcon_getscl,
 	.udelay		= 5,
-	/* Wait up to 50 ms for slave to let us pull SCL high */
+	/* Wait up to 50 ms for target to let us pull SCL high */
 	.timeout	= DIV_ROUND_UP(HZ, 20),
 };
 
-- 
cgit v1.2.3-58-ga151


From 93c3a96c301f0b1ac0bafb5e74bef58e79937648 Mon Sep 17 00:00:00 2001
From: Kory Maincent <kory.maincent@bootlin.com>
Date: Thu, 11 Jul 2024 15:55:18 +0200
Subject: net: pse-pd: Do not return EOPNOSUPP if config is null

For a PSE supporting both c33 and PoDL, setting config for one type of PoE
leaves the other type's config null. Currently, this case returns
EOPNOTSUPP, which is incorrect. Instead, we should do nothing if the
configuration is empty.

Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Fixes: d83e13761d5b ("net: pse-pd: Use regulator framework within PSE framework")
Link: https://patch.msgid.link/20240711-fix_pse_pd_deref-v3-1-edd78fc4fe42@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/pse-pd/pse_core.c | 4 ++--
 net/ethtool/pse-pd.c          | 4 +++-
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/net/pse-pd/pse_core.c b/drivers/net/pse-pd/pse_core.c
index 795ab264eaf2..513cd7f85933 100644
--- a/drivers/net/pse-pd/pse_core.c
+++ b/drivers/net/pse-pd/pse_core.c
@@ -719,13 +719,13 @@ int pse_ethtool_set_config(struct pse_control *psec,
 {
 	int err = 0;
 
-	if (pse_has_c33(psec)) {
+	if (pse_has_c33(psec) && config->c33_admin_control) {
 		err = pse_ethtool_c33_set_config(psec, config);
 		if (err)
 			return err;
 	}
 
-	if (pse_has_podl(psec))
+	if (pse_has_podl(psec) && config->podl_admin_control)
 		err = pse_ethtool_podl_set_config(psec, config);
 
 	return err;
diff --git a/net/ethtool/pse-pd.c b/net/ethtool/pse-pd.c
index 2c981d443f27..982995ff1628 100644
--- a/net/ethtool/pse-pd.c
+++ b/net/ethtool/pse-pd.c
@@ -183,7 +183,9 @@ ethnl_set_pse(struct ethnl_req_info *req_info, struct genl_info *info)
 	if (pse_has_c33(phydev->psec))
 		config.c33_admin_control = nla_get_u32(tb[ETHTOOL_A_C33_PSE_ADMIN_CONTROL]);
 
-	/* Return errno directly - PSE has no notification */
+	/* Return errno directly - PSE has no notification
+	 * pse_ethtool_set_config() will do nothing if the config is null
+	 */
 	return pse_ethtool_set_config(phydev->psec, info->extack, &config);
 }
 
-- 
cgit v1.2.3-58-ga151


From 4cddb0f15ea9c62f81b4889ea69a99368cc63a86 Mon Sep 17 00:00:00 2001
From: Kory Maincent <kory.maincent@bootlin.com>
Date: Thu, 11 Jul 2024 15:55:19 +0200
Subject: net: ethtool: pse-pd: Fix possible null-deref

Fix a possible null dereference when a PSE supports both c33 and PoDL, but
only one of the netlink attributes is specified. The c33 or PoDL PSE
capabilities are already validated in the ethnl_set_pse_validate() call.

Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Reported-by: Jakub Kicinski <kuba@kernel.org>
Closes: https://lore.kernel.org/netdev/20240705184116.13d8235a@kernel.org/
Fixes: 4d18e3ddf427 ("net: ethtool: pse-pd: Expand pse commands with the PSE PoE interface")
Link: https://patch.msgid.link/20240711-fix_pse_pd_deref-v3-2-edd78fc4fe42@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ethtool/pse-pd.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ethtool/pse-pd.c b/net/ethtool/pse-pd.c
index 982995ff1628..776ac96cdadc 100644
--- a/net/ethtool/pse-pd.c
+++ b/net/ethtool/pse-pd.c
@@ -178,9 +178,9 @@ ethnl_set_pse(struct ethnl_req_info *req_info, struct genl_info *info)
 
 	phydev = dev->phydev;
 	/* These values are already validated by the ethnl_pse_set_policy */
-	if (pse_has_podl(phydev->psec))
+	if (tb[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL])
 		config.podl_admin_control = nla_get_u32(tb[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL]);
-	if (pse_has_c33(phydev->psec))
+	if (tb[ETHTOOL_A_C33_PSE_ADMIN_CONTROL])
 		config.c33_admin_control = nla_get_u32(tb[ETHTOOL_A_C33_PSE_ADMIN_CONTROL]);
 
 	/* Return errno directly - PSE has no notification
-- 
cgit v1.2.3-58-ga151


From f67a90a0c8f5b3d0acc18f10650d90fec44775f9 Mon Sep 17 00:00:00 2001
From: Amit Cohen <amcohen@nvidia.com>
Date: Thu, 11 Jul 2024 17:27:02 +0200
Subject: selftests: forwarding: devlink_lib: Wait for udev events after
 reloading

Lately, an additional locking was added by commit c0a40097f0bc
("drivers: core: synchronize really_probe() and dev_uevent()"). The
locking protects dev_uevent() calling. This function is used to send
messages from the kernel to user space. Uevent messages notify user space
about changes in device states, such as when a device is added, removed,
or changed. These messages are used by udev (or other similar user-space
tools) to apply device-specific rules.

After reloading devlink instance, udev events should be processed. This
locking causes a short delay of udev events handling.

One example for useful udev rule is renaming ports. 'forwading.config'
can be configured to use names after udev rules are applied. Some tests run
devlink_reload() and immediately use the updated names. This worked before
the above mentioned commit was pushed, but now the delay of uevent messages
causes that devlink_reload() returns before udev events are handled and
tests fail.

Adjust devlink_reload() to not assume that udev events are already
processed when devlink reload is done, instead, wait for udev events to
ensure they are processed before returning from the function.

Without this patch:
TESTS='rif_mac_profile' ./resource_scale.sh
TEST: 'rif_mac_profile' 4                                           [ OK ]
sysctl: cannot stat /proc/sys/net/ipv6/conf/swp1/disable_ipv6: No such file or directory
sysctl: cannot stat /proc/sys/net/ipv6/conf/swp1/disable_ipv6: No such file or directory
sysctl: cannot stat /proc/sys/net/ipv6/conf/swp2/disable_ipv6: No such file or directory
sysctl: cannot stat /proc/sys/net/ipv6/conf/swp2/disable_ipv6: No such file or directory
Cannot find device "swp1"
Cannot find device "swp2"
TEST: setup_wait_dev (: Interface swp1 does not come up.) [FAIL]

With this patch:
$ TESTS='rif_mac_profile' ./resource_scale.sh
TEST: 'rif_mac_profile' 4                                           [ OK ]
TEST: 'rif_mac_profile' overflow 5                                  [ OK ]

This is relevant not only for this test.

Fixes: bc7cbb1e9f4c ("selftests: forwarding: Add devlink_lib.sh")
Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/89367666e04b38a8993027f1526801ca327ab96a.1720709333.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/forwarding/devlink_lib.sh | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tools/testing/selftests/net/forwarding/devlink_lib.sh b/tools/testing/selftests/net/forwarding/devlink_lib.sh
index f1de525cfa55..62a05bca1e82 100644
--- a/tools/testing/selftests/net/forwarding/devlink_lib.sh
+++ b/tools/testing/selftests/net/forwarding/devlink_lib.sh
@@ -122,6 +122,8 @@ devlink_reload()
 	still_pending=$(devlink resource show "$DEVLINK_DEV" | \
 			grep -c "size_new")
 	check_err $still_pending "Failed reload - There are still unset sizes"
+
+	udevadm settle
 }
 
 declare -A DEVLINK_ORIG
-- 
cgit v1.2.3-58-ga151


From 6807352353561187a718e87204458999dbcbba1b Mon Sep 17 00:00:00 2001
From: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Date: Wed, 10 Jul 2024 10:14:27 +0200
Subject: ipv4: fix source address selection with route leak

By default, an address assigned to the output interface is selected when
the source address is not specified. This is problematic when a route,
configured in a vrf, uses an interface from another vrf (aka route leak).
The original vrf does not own the selected source address.

Let's add a check against the output interface and call the appropriate
function to select the source address.

CC: stable@vger.kernel.org
Fixes: 8cbb512c923d ("net: Add source address lookup op for VRF")
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://patch.msgid.link/20240710081521.3809742-2-nicolas.dichtel@6wind.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/fib_semantics.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index f669da98d11d..8956026bc0a2 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -2270,6 +2270,15 @@ void fib_select_path(struct net *net, struct fib_result *res,
 		fib_select_default(fl4, res);
 
 check_saddr:
-	if (!fl4->saddr)
-		fl4->saddr = fib_result_prefsrc(net, res);
+	if (!fl4->saddr) {
+		struct net_device *l3mdev;
+
+		l3mdev = dev_get_by_index_rcu(net, fl4->flowi4_l3mdev);
+
+		if (!l3mdev ||
+		    l3mdev_master_dev_rcu(FIB_RES_DEV(*res)) == l3mdev)
+			fl4->saddr = fib_result_prefsrc(net, res);
+		else
+			fl4->saddr = inet_select_addr(l3mdev, 0, RT_SCOPE_LINK);
+	}
 }
-- 
cgit v1.2.3-58-ga151


From 252442f2ae317d109ef0b4b39ce0608c09563042 Mon Sep 17 00:00:00 2001
From: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Date: Wed, 10 Jul 2024 10:14:28 +0200
Subject: ipv6: fix source address selection with route leak

By default, an address assigned to the output interface is selected when
the source address is not specified. This is problematic when a route,
configured in a vrf, uses an interface from another vrf (aka route leak).
The original vrf does not own the selected source address.

Let's add a check against the output interface and call the appropriate
function to select the source address.

CC: stable@vger.kernel.org
Fixes: 0d240e7811c4 ("net: vrf: Implement get_saddr for IPv6")
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Link: https://patch.msgid.link/20240710081521.3809742-3-nicolas.dichtel@6wind.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/ip6_route.h | 20 ++++++++++++++------
 net/ipv6/ip6_output.c   |  1 +
 net/ipv6/route.c        |  2 +-
 3 files changed, 16 insertions(+), 7 deletions(-)

diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index a18ed24fed94..6dbdf60b342f 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -127,18 +127,26 @@ void rt6_age_exceptions(struct fib6_info *f6i, struct fib6_gc_args *gc_args,
 
 static inline int ip6_route_get_saddr(struct net *net, struct fib6_info *f6i,
 				      const struct in6_addr *daddr,
-				      unsigned int prefs,
+				      unsigned int prefs, int l3mdev_index,
 				      struct in6_addr *saddr)
 {
+	struct net_device *l3mdev;
+	struct net_device *dev;
+	bool same_vrf;
 	int err = 0;
 
-	if (f6i && f6i->fib6_prefsrc.plen) {
+	rcu_read_lock();
+
+	l3mdev = dev_get_by_index_rcu(net, l3mdev_index);
+	if (!f6i || !f6i->fib6_prefsrc.plen || l3mdev)
+		dev = f6i ? fib6_info_nh_dev(f6i) : NULL;
+	same_vrf = !l3mdev || l3mdev_master_dev_rcu(dev) == l3mdev;
+	if (f6i && f6i->fib6_prefsrc.plen && same_vrf)
 		*saddr = f6i->fib6_prefsrc.addr;
-	} else {
-		struct net_device *dev = f6i ? fib6_info_nh_dev(f6i) : NULL;
+	else
+		err = ipv6_dev_get_saddr(net, same_vrf ? dev : l3mdev, daddr, prefs, saddr);
 
-		err = ipv6_dev_get_saddr(net, dev, daddr, prefs, saddr);
-	}
+	rcu_read_unlock();
 
 	return err;
 }
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 27d8725445e3..784424ac4147 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1124,6 +1124,7 @@ static int ip6_dst_lookup_tail(struct net *net, const struct sock *sk,
 		from = rt ? rcu_dereference(rt->from) : NULL;
 		err = ip6_route_get_saddr(net, from, &fl6->daddr,
 					  sk ? READ_ONCE(inet6_sk(sk)->srcprefs) : 0,
+					  fl6->flowi6_l3mdev,
 					  &fl6->saddr);
 		rcu_read_unlock();
 
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 8d72ca0b086d..c9a9506b714d 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -5689,7 +5689,7 @@ static int rt6_fill_node(struct net *net, struct sk_buff *skb,
 				goto nla_put_failure;
 	} else if (dest) {
 		struct in6_addr saddr_buf;
-		if (ip6_route_get_saddr(net, rt, dest, 0, &saddr_buf) == 0 &&
+		if (ip6_route_get_saddr(net, rt, dest, 0, 0, &saddr_buf) == 0 &&
 		    nla_put_in6_addr(skb, RTA_PREFSRC, &saddr_buf))
 			goto nla_put_failure;
 	}
-- 
cgit v1.2.3-58-ga151


From abb9a68d2c64dd9b128ae1f2e635e4d805e7ce64 Mon Sep 17 00:00:00 2001
From: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Date: Wed, 10 Jul 2024 10:14:29 +0200
Subject: ipv6: take care of scope when choosing the src addr

When the source address is selected, the scope must be checked. For
example, if a loopback address is assigned to the vrf device, it must not
be chosen for packets sent outside.

CC: stable@vger.kernel.org
Fixes: afbac6010aec ("net: ipv6: Address selection needs to consider L3 domains")
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://patch.msgid.link/20240710081521.3809742-4-nicolas.dichtel@6wind.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv6/addrconf.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 5c424a0e7232..4f2c5cc31015 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -1873,7 +1873,8 @@ int ipv6_dev_get_saddr(struct net *net, const struct net_device *dst_dev,
 							    master, &dst,
 							    scores, hiscore_idx);
 
-			if (scores[hiscore_idx].ifa)
+			if (scores[hiscore_idx].ifa &&
+			    scores[hiscore_idx].scopedist >= 0)
 				goto out;
 		}
 
-- 
cgit v1.2.3-58-ga151


From 39367183aecfc05f006a29b1f96708a6d1a3d8c3 Mon Sep 17 00:00:00 2001
From: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Date: Wed, 10 Jul 2024 10:14:30 +0200
Subject: selftests: vrf_route_leaking: add local test

The goal is to check that the source address selected by the kernel is
routable when a leaking route is used. ICMP, TCP and UDP connections are
tested.
The symmetric topology is enough for this test.

Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://patch.msgid.link/20240710081521.3809742-5-nicolas.dichtel@6wind.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/vrf_route_leaking.sh | 93 +++++++++++++++++++++++-
 1 file changed, 91 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/net/vrf_route_leaking.sh b/tools/testing/selftests/net/vrf_route_leaking.sh
index 2da32f4c479b..152171fb1fc8 100755
--- a/tools/testing/selftests/net/vrf_route_leaking.sh
+++ b/tools/testing/selftests/net/vrf_route_leaking.sh
@@ -59,6 +59,7 @@
 # while it is forwarded between different vrfs.
 
 source lib.sh
+PATH=$PWD:$PWD/tools/testing/selftests/net:$PATH
 VERBOSE=0
 PAUSE_ON_FAIL=no
 DEFAULT_TTYPE=sym
@@ -533,6 +534,86 @@ ipv6_ping_frag_asym()
 	ipv6_ping_frag asym
 }
 
+ipv4_ping_local()
+{
+	log_section "IPv4 (sym route): VRF ICMP local error route lookup ping"
+
+	setup_sym
+
+	check_connectivity || return
+
+	run_cmd ip netns exec $r1 ip vrf exec blue ping -c1 -w1 ${H2_N2_IP}
+	log_test $? 0 "VRF ICMP local IPv4"
+}
+
+ipv4_tcp_local()
+{
+	log_section "IPv4 (sym route): VRF tcp local connection"
+
+	setup_sym
+
+	check_connectivity || return
+
+	run_cmd nettest -s -O "$h2" -l ${H2_N2_IP} -I eth0 -3 eth0 &
+	sleep 1
+	run_cmd nettest -N "$r1" -d blue -r ${H2_N2_IP}
+	log_test $? 0 "VRF tcp local connection IPv4"
+}
+
+ipv4_udp_local()
+{
+	log_section "IPv4 (sym route): VRF udp local connection"
+
+	setup_sym
+
+	check_connectivity || return
+
+	run_cmd nettest -s -D -O "$h2" -l ${H2_N2_IP} -I eth0 -3 eth0 &
+	sleep 1
+	run_cmd nettest -D -N "$r1" -d blue -r ${H2_N2_IP}
+	log_test $? 0 "VRF udp local connection IPv4"
+}
+
+ipv6_ping_local()
+{
+	log_section "IPv6 (sym route): VRF ICMP local error route lookup ping"
+
+	setup_sym
+
+	check_connectivity6 || return
+
+	run_cmd ip netns exec $r1 ip vrf exec blue ${ping6} -c1 -w1 ${H2_N2_IP6}
+	log_test $? 0 "VRF ICMP local IPv6"
+}
+
+ipv6_tcp_local()
+{
+	log_section "IPv6 (sym route): VRF tcp local connection"
+
+	setup_sym
+
+	check_connectivity6 || return
+
+	run_cmd nettest -s -6 -O "$h2" -l ${H2_N2_IP6} -I eth0 -3 eth0 &
+	sleep 1
+	run_cmd nettest -6 -N "$r1" -d blue -r ${H2_N2_IP6}
+	log_test $? 0 "VRF tcp local connection IPv6"
+}
+
+ipv6_udp_local()
+{
+	log_section "IPv6 (sym route): VRF udp local connection"
+
+	setup_sym
+
+	check_connectivity6 || return
+
+	run_cmd nettest -s -6 -D -O "$h2" -l ${H2_N2_IP6} -I eth0 -3 eth0 &
+	sleep 1
+	run_cmd nettest -6 -D -N "$r1" -d blue -r ${H2_N2_IP6}
+	log_test $? 0 "VRF udp local connection IPv6"
+}
+
 ################################################################################
 # usage
 
@@ -555,8 +636,10 @@ EOF
 # Some systems don't have a ping6 binary anymore
 command -v ping6 > /dev/null 2>&1 && ping6=$(command -v ping6) || ping6=$(command -v ping)
 
-TESTS_IPV4="ipv4_ping_ttl ipv4_traceroute ipv4_ping_frag ipv4_ping_ttl_asym ipv4_traceroute_asym"
-TESTS_IPV6="ipv6_ping_ttl ipv6_traceroute ipv6_ping_ttl_asym ipv6_traceroute_asym"
+TESTS_IPV4="ipv4_ping_ttl ipv4_traceroute ipv4_ping_frag ipv4_ping_local ipv4_tcp_local
+ipv4_udp_local ipv4_ping_ttl_asym ipv4_traceroute_asym"
+TESTS_IPV6="ipv6_ping_ttl ipv6_traceroute ipv6_ping_local ipv6_tcp_local ipv6_udp_local
+ipv6_ping_ttl_asym ipv6_traceroute_asym"
 
 ret=0
 nsuccess=0
@@ -594,12 +677,18 @@ do
 	ipv4_traceroute|traceroute)      ipv4_traceroute;;&
 	ipv4_traceroute_asym|traceroute) ipv4_traceroute_asym;;&
 	ipv4_ping_frag|ping)             ipv4_ping_frag;;&
+	ipv4_ping_local|ping)            ipv4_ping_local;;&
+	ipv4_tcp_local)                  ipv4_tcp_local;;&
+	ipv4_udp_local)                  ipv4_udp_local;;&
 
 	ipv6_ping_ttl|ping)              ipv6_ping_ttl;;&
 	ipv6_ping_ttl_asym|ping)         ipv6_ping_ttl_asym;;&
 	ipv6_traceroute|traceroute)      ipv6_traceroute;;&
 	ipv6_traceroute_asym|traceroute) ipv6_traceroute_asym;;&
 	ipv6_ping_frag|ping)             ipv6_ping_frag;;&
+	ipv6_ping_local|ping)            ipv6_ping_local;;&
+	ipv6_tcp_local)                  ipv6_tcp_local;;&
+	ipv6_udp_local)                  ipv6_udp_local;;&
 
 	# setup namespaces and config, but do not run any tests
 	setup_sym|setup)                 setup_sym; exit 0;;
-- 
cgit v1.2.3-58-ga151


From 97d9fba9a812cada5484667a46e14a4c976ca330 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Fri, 12 Jul 2024 07:34:15 -0700
Subject: net: netconsole: Disable target before netpoll cleanup

Currently, netconsole cleans up the netpoll structure before disabling
the target. This approach can lead to race conditions, as message
senders (write_ext_msg() and write_msg()) check if the target is
enabled before using netpoll. The sender can validate that the target is
enabled, but, the netpoll might be de-allocated already, causing
undesired behaviours.

This patch reverses the order of operations:
1. Disable the target
2. Clean up the netpoll structure

This change eliminates the potential race condition, ensuring that
no messages are sent through a partially cleaned-up netpoll structure.

Fixes: 2382b15bcc39 ("netconsole: take care of NETDEV_UNREGISTER event")
Cc: stable@vger.kernel.org
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20240712143415.1141039-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/netconsole.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index d7070dd4fe73..aa66c923790f 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -974,6 +974,7 @@ restart:
 				/* rtnl_lock already held
 				 * we might sleep in __netpoll_cleanup()
 				 */
+				nt->enabled = false;
 				spin_unlock_irqrestore(&target_list_lock, flags);
 
 				__netpoll_cleanup(&nt->np);
@@ -981,7 +982,6 @@ restart:
 				spin_lock_irqsave(&target_list_lock, flags);
 				netdev_put(nt->np.dev, &nt->np.dev_tracker);
 				nt->np.dev = NULL;
-				nt->enabled = false;
 				stopped = true;
 				netconsole_target_put(nt);
 				goto restart;
-- 
cgit v1.2.3-58-ga151


From 6bc8719c9dbf2d107760a238394dd77bb7df9ab9 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Fri, 12 Jul 2024 23:27:57 +0200
Subject: dt-bindings: net: airoha: Add EN7581 ethernet controller

Introduce device-tree binding documentation for Airoha EN7581 ethernet
mac controller.

Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Link: https://patch.msgid.link/7dfecf8aa4e6519562a94455b95c49e1b3c858a0.1720818878.git.lorenzo@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../devicetree/bindings/net/airoha,en7581-eth.yaml | 143 +++++++++++++++++++++
 1 file changed, 143 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/airoha,en7581-eth.yaml

diff --git a/Documentation/devicetree/bindings/net/airoha,en7581-eth.yaml b/Documentation/devicetree/bindings/net/airoha,en7581-eth.yaml
new file mode 100644
index 000000000000..c578637c5826
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/airoha,en7581-eth.yaml
@@ -0,0 +1,143 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/airoha,en7581-eth.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Airoha EN7581 Frame Engine Ethernet controller
+
+maintainers:
+  - Lorenzo Bianconi <lorenzo@kernel.org>
+
+description:
+  The frame engine ethernet controller can be found on Airoha SoCs.
+  These SoCs have multi-GMAC ports.
+
+properties:
+  compatible:
+    enum:
+      - airoha,en7581-eth
+
+  reg:
+    items:
+      - description: Frame engine base address
+      - description: QDMA0 base address
+      - description: QDMA1 base address
+
+  reg-names:
+    items:
+      - const: fe
+      - const: qdma0
+      - const: qdma1
+
+  interrupts:
+    items:
+      - description: QDMA lan irq0
+      - description: QDMA lan irq1
+      - description: QDMA lan irq2
+      - description: QDMA lan irq3
+      - description: QDMA wan irq0
+      - description: QDMA wan irq1
+      - description: QDMA wan irq2
+      - description: QDMA wan irq3
+      - description: FE error irq
+      - description: PDMA irq
+
+  resets:
+    maxItems: 8
+
+  reset-names:
+    items:
+      - const: fe
+      - const: pdma
+      - const: qdma
+      - const: xsi-mac
+      - const: hsi0-mac
+      - const: hsi1-mac
+      - const: hsi-mac
+      - const: xfp-mac
+
+  "#address-cells":
+    const: 1
+
+  "#size-cells":
+    const: 0
+
+patternProperties:
+  "^ethernet@[1-4]$":
+    type: object
+    unevaluatedProperties: false
+    $ref: ethernet-controller.yaml#
+    description:
+      Ethernet GMAC port associated to the MAC controller
+    properties:
+      compatible:
+        const: airoha,eth-mac
+
+      reg:
+        minimum: 1
+        maximum: 4
+        description: GMAC port identifier
+
+    required:
+      - reg
+      - compatible
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - resets
+  - reset-names
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/interrupt-controller/irq.h>
+    #include <dt-bindings/clock/en7523-clk.h>
+
+    soc {
+      #address-cells = <2>;
+      #size-cells = <2>;
+
+      eth: ethernet@1fb50000 {
+        compatible = "airoha,en7581-eth";
+        reg = <0 0x1fb50000 0 0x2600>,
+              <0 0x1fb54000 0 0x2000>,
+              <0 0x1fb56000 0 0x2000>;
+        reg-names = "fe", "qdma0", "qdma1";
+
+        resets = <&scuclk 44>,
+                 <&scuclk 30>,
+                 <&scuclk 31>,
+                 <&scuclk 6>,
+                 <&scuclk 15>,
+                 <&scuclk 16>,
+                 <&scuclk 17>,
+                 <&scuclk 26>;
+        reset-names = "fe", "pdma", "qdma", "xsi-mac",
+                      "hsi0-mac", "hsi1-mac", "hsi-mac",
+                      "xfp-mac";
+
+        interrupts = <GIC_SPI 37 IRQ_TYPE_LEVEL_HIGH>,
+                     <GIC_SPI 55 IRQ_TYPE_LEVEL_HIGH>,
+                     <GIC_SPI 56 IRQ_TYPE_LEVEL_HIGH>,
+                     <GIC_SPI 57 IRQ_TYPE_LEVEL_HIGH>,
+                     <GIC_SPI 38 IRQ_TYPE_LEVEL_HIGH>,
+                     <GIC_SPI 58 IRQ_TYPE_LEVEL_HIGH>,
+                     <GIC_SPI 59 IRQ_TYPE_LEVEL_HIGH>,
+                     <GIC_SPI 60 IRQ_TYPE_LEVEL_HIGH>,
+                     <GIC_SPI 49 IRQ_TYPE_LEVEL_HIGH>,
+                     <GIC_SPI 64 IRQ_TYPE_LEVEL_HIGH>;
+
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        mac: ethernet@1 {
+          compatible = "airoha,eth-mac";
+          reg = <1>;
+        };
+      };
+    };
-- 
cgit v1.2.3-58-ga151


From 23020f04932701d5c8363e60756f12b43b8ed752 Mon Sep 17 00:00:00 2001
From: Lorenzo Bianconi <lorenzo@kernel.org>
Date: Fri, 12 Jul 2024 23:27:58 +0200
Subject: net: airoha: Introduce ethernet support for EN7581 SoC
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add airoha_eth driver in order to introduce ethernet support for
Airoha EN7581 SoC available on EN7581 development board (en7581-evb).
EN7581 mac controller is mainly composed by the Frame Engine (PSE+PPE)
and QoS-DMA (QDMA) modules. FE is used for traffic offloading (just
basic functionalities are currently supported) while QDMA is used for
DMA operations and QOS functionalities between the mac layer and the
external modules conncted to the FE GDM ports (e.g MT7530 DSA switch
or external phys).
A general overview of airoha_eth architecture is reported below:

               ┌───────┐                                     ┌───────┐
               │ QDMA2 │                                     │ QDMA1 │
               └───┬───┘                                     └───┬───┘
                   │                                             │
           ┌───────▼─────────────────────────────────────────────▼────────┐
           │                                                              │
           │       P5                                            P0       │
           │                                                              │
           │                                                              │
           │                                                              │    ┌──────┐
           │                                                           P3 ├────► GDM3 │
           │                                                              │    └──────┘
           │                                                              │
           │                                                              │
┌─────┐    │                                                              │
│ PPE ◄────┤ P4                          PSE                              │
└─────┘    │                                                              │
           │                                                              │
           │                                                              │
           │                                                              │    ┌──────┐
           │                                                           P9 ├────► GDM4 │
           │                                                              │    └──────┘
           │                                                              │
           │                                                              │
           │                                                              │
           │        P2                                           P1       │
           └─────────┬───────────────────────────────────────────┬────────┘
                     │                                           │
                 ┌───▼──┐                                     ┌──▼───┐
                 │ GDM2 │                                     │ GDM1 │
                 └──────┘                                     └──┬───┘
                                                                 │
                                                            ┌────▼─────┐
                                                            │  MT7530  │
                                                            └──────────┘

Currently only hw LAN features (QDMA1+GDM1) are available while hw WAN
(QDMA2+GDM{2,3,4}) ones will be added with subsequent patches introducing
traffic offloading support.

Tested-by: Benjamin Larsson <benjamin.larsson@genexis.eu>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Link: https://patch.msgid.link/274945d2391c195098ab180a46d0617b18b9e42c.1720818878.git.lorenzo@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 MAINTAINERS                                |    9 +
 drivers/net/ethernet/mediatek/Kconfig      |   10 +-
 drivers/net/ethernet/mediatek/Makefile     |    1 +
 drivers/net/ethernet/mediatek/airoha_eth.c | 2730 ++++++++++++++++++++++++++++
 4 files changed, 2749 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/ethernet/mediatek/airoha_eth.c

diff --git a/MAINTAINERS b/MAINTAINERS
index 5adfbc01bd2e..e138c90b1adf 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -682,6 +682,15 @@ S:	Supported
 F:	fs/aio.c
 F:	include/linux/*aio*.h
 
+AIROHA ETHERNET DRIVER
+M:	Lorenzo Bianconi <lorenzo@kernel.org>
+L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
+L:	linux-mediatek@lists.infradead.org (moderated for non-subscribers)
+L:	netdev@vger.kernel.org
+S:	Maintained
+F:	Documentation/devicetree/bindings/net/airoha,en7581-eth.yaml
+F:	drivers/net/ethernet/mediatek/airoha_eth.c
+
 AIROHA SPI SNFI DRIVER
 M:	Lorenzo Bianconi <lorenzo@kernel.org>
 M:	Ray Liu <ray.liu@airoha.com>
diff --git a/drivers/net/ethernet/mediatek/Kconfig b/drivers/net/ethernet/mediatek/Kconfig
index da0db417ab69..95c4405b7d7b 100644
--- a/drivers/net/ethernet/mediatek/Kconfig
+++ b/drivers/net/ethernet/mediatek/Kconfig
@@ -1,12 +1,20 @@
 # SPDX-License-Identifier: GPL-2.0-only
 config NET_VENDOR_MEDIATEK
 	bool "MediaTek devices"
-	depends on ARCH_MEDIATEK || SOC_MT7621 || SOC_MT7620 || COMPILE_TEST
+	depends on ARCH_MEDIATEK || ARCH_AIROHA || SOC_MT7621 || SOC_MT7620 || COMPILE_TEST
 	help
 	  If you have a Mediatek SoC with ethernet, say Y.
 
 if NET_VENDOR_MEDIATEK
 
+config NET_AIROHA
+	tristate "Airoha SoC Gigabit Ethernet support"
+	depends on NET_DSA || !NET_DSA
+	select PAGE_POOL
+	help
+	  This driver supports the gigabit ethernet MACs in the
+	  Airoha SoC family.
+
 config NET_MEDIATEK_SOC_WED
 	depends on ARCH_MEDIATEK || COMPILE_TEST
 	def_bool NET_MEDIATEK_SOC != n
diff --git a/drivers/net/ethernet/mediatek/Makefile b/drivers/net/ethernet/mediatek/Makefile
index 03e008fbc859..ddbb7f4a516c 100644
--- a/drivers/net/ethernet/mediatek/Makefile
+++ b/drivers/net/ethernet/mediatek/Makefile
@@ -11,3 +11,4 @@ mtk_eth-$(CONFIG_NET_MEDIATEK_SOC_WED) += mtk_wed_debugfs.o
 endif
 obj-$(CONFIG_NET_MEDIATEK_SOC_WED) += mtk_wed_ops.o
 obj-$(CONFIG_NET_MEDIATEK_STAR_EMAC) += mtk_star_emac.o
+obj-$(CONFIG_NET_AIROHA) += airoha_eth.o
diff --git a/drivers/net/ethernet/mediatek/airoha_eth.c b/drivers/net/ethernet/mediatek/airoha_eth.c
new file mode 100644
index 000000000000..7967a92803c2
--- /dev/null
+++ b/drivers/net/ethernet/mediatek/airoha_eth.c
@@ -0,0 +1,2730 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2024 AIROHA Inc
+ * Author: Lorenzo Bianconi <lorenzo@kernel.org>
+ */
+#include <linux/etherdevice.h>
+#include <linux/iopoll.h>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/of.h>
+#include <linux/of_net.h>
+#include <linux/platform_device.h>
+#include <linux/reset.h>
+#include <linux/tcp.h>
+#include <linux/u64_stats_sync.h>
+#include <net/dsa.h>
+#include <net/page_pool/helpers.h>
+#include <uapi/linux/ppp_defs.h>
+
+#define AIROHA_MAX_NUM_GDM_PORTS	1
+#define AIROHA_MAX_NUM_RSTS		3
+#define AIROHA_MAX_NUM_XSI_RSTS		5
+#define AIROHA_MAX_MTU			2000
+#define AIROHA_MAX_PACKET_SIZE		2048
+#define AIROHA_NUM_TX_RING		32
+#define AIROHA_NUM_RX_RING		32
+#define AIROHA_FE_MC_MAX_VLAN_TABLE	64
+#define AIROHA_FE_MC_MAX_VLAN_PORT	16
+#define AIROHA_NUM_TX_IRQ		2
+#define HW_DSCP_NUM			2048
+#define IRQ_QUEUE_LEN(_n)		((_n) ? 1024 : 2048)
+#define TX_DSCP_NUM			1024
+#define RX_DSCP_NUM(_n)			\
+	((_n) ==  2 ? 128 :		\
+	 (_n) == 11 ? 128 :		\
+	 (_n) == 15 ? 128 :		\
+	 (_n) ==  0 ? 1024 : 16)
+
+#define PSE_RSV_PAGES			128
+#define PSE_QUEUE_RSV_PAGES		64
+
+/* FE */
+#define PSE_BASE			0x0100
+#define CSR_IFC_BASE			0x0200
+#define CDM1_BASE			0x0400
+#define GDM1_BASE			0x0500
+#define PPE1_BASE			0x0c00
+
+#define CDM2_BASE			0x1400
+#define GDM2_BASE			0x1500
+
+#define GDM3_BASE			0x1100
+#define GDM4_BASE			0x2500
+
+#define GDM_BASE(_n)			\
+	((_n) == 4 ? GDM4_BASE :	\
+	 (_n) == 3 ? GDM3_BASE :	\
+	 (_n) == 2 ? GDM2_BASE : GDM1_BASE)
+
+#define REG_FE_DMA_GLO_CFG		0x0000
+#define FE_DMA_GLO_L2_SPACE_MASK	GENMASK(7, 4)
+#define FE_DMA_GLO_PG_SZ_MASK		BIT(3)
+
+#define REG_FE_RST_GLO_CFG		0x0004
+#define FE_RST_GDM4_MBI_ARB_MASK	BIT(3)
+#define FE_RST_GDM3_MBI_ARB_MASK	BIT(2)
+#define FE_RST_CORE_MASK		BIT(0)
+
+#define REG_FE_LAN_MAC_H		0x0040
+#define REG_FE_LAN_MAC_LMIN		0x0044
+#define REG_FE_LAN_MAC_LMAX		0x0048
+
+#define REG_FE_CDM1_OQ_MAP0		0x0050
+#define REG_FE_CDM1_OQ_MAP1		0x0054
+#define REG_FE_CDM1_OQ_MAP2		0x0058
+#define REG_FE_CDM1_OQ_MAP3		0x005c
+
+#define REG_FE_PCE_CFG			0x0070
+#define PCE_DPI_EN_MASK			BIT(2)
+#define PCE_KA_EN_MASK			BIT(1)
+#define PCE_MC_EN_MASK			BIT(0)
+
+#define REG_FE_PSE_QUEUE_CFG_WR		0x0080
+#define PSE_CFG_PORT_ID_MASK		GENMASK(27, 24)
+#define PSE_CFG_QUEUE_ID_MASK		GENMASK(20, 16)
+#define PSE_CFG_WR_EN_MASK		BIT(8)
+#define PSE_CFG_OQRSV_SEL_MASK		BIT(0)
+
+#define REG_FE_PSE_QUEUE_CFG_VAL	0x0084
+#define PSE_CFG_OQ_RSV_MASK		GENMASK(13, 0)
+
+#define PSE_FQ_CFG			0x008c
+#define PSE_FQ_LIMIT_MASK		GENMASK(14, 0)
+
+#define REG_FE_PSE_BUF_SET		0x0090
+#define PSE_SHARE_USED_LTHD_MASK	GENMASK(31, 16)
+#define PSE_ALLRSV_MASK			GENMASK(14, 0)
+
+#define REG_PSE_SHARE_USED_THD		0x0094
+#define PSE_SHARE_USED_MTHD_MASK	GENMASK(31, 16)
+#define PSE_SHARE_USED_HTHD_MASK	GENMASK(15, 0)
+
+#define REG_GDM_MISC_CFG		0x0148
+#define GDM2_RDM_ACK_WAIT_PREF_MASK	BIT(9)
+#define GDM2_CHN_VLD_MODE_MASK		BIT(5)
+
+#define REG_FE_CSR_IFC_CFG		CSR_IFC_BASE
+#define FE_IFC_EN_MASK			BIT(0)
+
+#define REG_FE_VIP_PORT_EN		0x01f0
+#define REG_FE_IFC_PORT_EN		0x01f4
+
+#define REG_PSE_IQ_REV1			(PSE_BASE + 0x08)
+#define PSE_IQ_RES1_P2_MASK		GENMASK(23, 16)
+
+#define REG_PSE_IQ_REV2			(PSE_BASE + 0x0c)
+#define PSE_IQ_RES2_P5_MASK		GENMASK(15, 8)
+#define PSE_IQ_RES2_P4_MASK		GENMASK(7, 0)
+
+#define REG_FE_VIP_EN(_n)		(0x0300 + ((_n) << 3))
+#define PATN_FCPU_EN_MASK		BIT(7)
+#define PATN_SWP_EN_MASK		BIT(6)
+#define PATN_DP_EN_MASK			BIT(5)
+#define PATN_SP_EN_MASK			BIT(4)
+#define PATN_TYPE_MASK			GENMASK(3, 1)
+#define PATN_EN_MASK			BIT(0)
+
+#define REG_FE_VIP_PATN(_n)		(0x0304 + ((_n) << 3))
+#define PATN_DP_MASK			GENMASK(31, 16)
+#define PATN_SP_MASK			GENMASK(15, 0)
+
+#define REG_CDM1_VLAN_CTRL		CDM1_BASE
+#define CDM1_VLAN_MASK			GENMASK(31, 16)
+
+#define REG_CDM1_FWD_CFG		(CDM1_BASE + 0x08)
+#define CDM1_VIP_QSEL_MASK		GENMASK(24, 20)
+
+#define REG_CDM1_CRSN_QSEL(_n)		(CDM1_BASE + 0x10 + ((_n) << 2))
+#define CDM1_CRSN_QSEL_REASON_MASK(_n)	\
+	GENMASK(4 + (((_n) % 4) << 3),	(((_n) % 4) << 3))
+
+#define REG_CDM2_FWD_CFG		(CDM2_BASE + 0x08)
+#define CDM2_OAM_QSEL_MASK		GENMASK(31, 27)
+#define CDM2_VIP_QSEL_MASK		GENMASK(24, 20)
+
+#define REG_CDM2_CRSN_QSEL(_n)		(CDM2_BASE + 0x10 + ((_n) << 2))
+#define CDM2_CRSN_QSEL_REASON_MASK(_n)	\
+	GENMASK(4 + (((_n) % 4) << 3),	(((_n) % 4) << 3))
+
+#define REG_GDM_FWD_CFG(_n)		GDM_BASE(_n)
+#define GDM_DROP_CRC_ERR		BIT(23)
+#define GDM_IP4_CKSUM			BIT(22)
+#define GDM_TCP_CKSUM			BIT(21)
+#define GDM_UDP_CKSUM			BIT(20)
+#define GDM_UCFQ_MASK			GENMASK(15, 12)
+#define GDM_BCFQ_MASK			GENMASK(11, 8)
+#define GDM_MCFQ_MASK			GENMASK(7, 4)
+#define GDM_OCFQ_MASK			GENMASK(3, 0)
+
+#define REG_GDM_INGRESS_CFG(_n)		(GDM_BASE(_n) + 0x10)
+#define GDM_INGRESS_FC_EN_MASK		BIT(1)
+#define GDM_STAG_EN_MASK		BIT(0)
+
+#define REG_GDM_LEN_CFG(_n)		(GDM_BASE(_n) + 0x14)
+#define GDM_SHORT_LEN_MASK		GENMASK(13, 0)
+#define GDM_LONG_LEN_MASK		GENMASK(29, 16)
+
+#define REG_FE_CPORT_CFG		(GDM1_BASE + 0x40)
+#define FE_CPORT_PAD			BIT(26)
+#define FE_CPORT_PORT_XFC_MASK		BIT(25)
+#define FE_CPORT_QUEUE_XFC_MASK		BIT(24)
+
+#define REG_FE_GDM_MIB_CLEAR(_n)	(GDM_BASE(_n) + 0xf0)
+#define FE_GDM_MIB_RX_CLEAR_MASK	BIT(1)
+#define FE_GDM_MIB_TX_CLEAR_MASK	BIT(0)
+
+#define REG_FE_GDM1_MIB_CFG		(GDM1_BASE + 0xf4)
+#define FE_STRICT_RFC2819_MODE_MASK	BIT(31)
+#define FE_GDM1_TX_MIB_SPLIT_EN_MASK	BIT(17)
+#define FE_GDM1_RX_MIB_SPLIT_EN_MASK	BIT(16)
+#define FE_TX_MIB_ID_MASK		GENMASK(15, 8)
+#define FE_RX_MIB_ID_MASK		GENMASK(7, 0)
+
+#define REG_FE_GDM_TX_OK_PKT_CNT_L(_n)		(GDM_BASE(_n) + 0x104)
+#define REG_FE_GDM_TX_OK_BYTE_CNT_L(_n)		(GDM_BASE(_n) + 0x10c)
+#define REG_FE_GDM_TX_ETH_PKT_CNT_L(_n)		(GDM_BASE(_n) + 0x110)
+#define REG_FE_GDM_TX_ETH_BYTE_CNT_L(_n)	(GDM_BASE(_n) + 0x114)
+#define REG_FE_GDM_TX_ETH_DROP_CNT(_n)		(GDM_BASE(_n) + 0x118)
+#define REG_FE_GDM_TX_ETH_BC_CNT(_n)		(GDM_BASE(_n) + 0x11c)
+#define REG_FE_GDM_TX_ETH_MC_CNT(_n)		(GDM_BASE(_n) + 0x120)
+#define REG_FE_GDM_TX_ETH_RUNT_CNT(_n)		(GDM_BASE(_n) + 0x124)
+#define REG_FE_GDM_TX_ETH_LONG_CNT(_n)		(GDM_BASE(_n) + 0x128)
+#define REG_FE_GDM_TX_ETH_E64_CNT_L(_n)		(GDM_BASE(_n) + 0x12c)
+#define REG_FE_GDM_TX_ETH_L64_CNT_L(_n)		(GDM_BASE(_n) + 0x130)
+#define REG_FE_GDM_TX_ETH_L127_CNT_L(_n)	(GDM_BASE(_n) + 0x134)
+#define REG_FE_GDM_TX_ETH_L255_CNT_L(_n)	(GDM_BASE(_n) + 0x138)
+#define REG_FE_GDM_TX_ETH_L511_CNT_L(_n)	(GDM_BASE(_n) + 0x13c)
+#define REG_FE_GDM_TX_ETH_L1023_CNT_L(_n)	(GDM_BASE(_n) + 0x140)
+
+#define REG_FE_GDM_RX_OK_PKT_CNT_L(_n)		(GDM_BASE(_n) + 0x148)
+#define REG_FE_GDM_RX_FC_DROP_CNT(_n)		(GDM_BASE(_n) + 0x14c)
+#define REG_FE_GDM_RX_RC_DROP_CNT(_n)		(GDM_BASE(_n) + 0x150)
+#define REG_FE_GDM_RX_OVERFLOW_DROP_CNT(_n)	(GDM_BASE(_n) + 0x154)
+#define REG_FE_GDM_RX_ERROR_DROP_CNT(_n)	(GDM_BASE(_n) + 0x158)
+#define REG_FE_GDM_RX_OK_BYTE_CNT_L(_n)		(GDM_BASE(_n) + 0x15c)
+#define REG_FE_GDM_RX_ETH_PKT_CNT_L(_n)		(GDM_BASE(_n) + 0x160)
+#define REG_FE_GDM_RX_ETH_BYTE_CNT_L(_n)	(GDM_BASE(_n) + 0x164)
+#define REG_FE_GDM_RX_ETH_DROP_CNT(_n)		(GDM_BASE(_n) + 0x168)
+#define REG_FE_GDM_RX_ETH_BC_CNT(_n)		(GDM_BASE(_n) + 0x16c)
+#define REG_FE_GDM_RX_ETH_MC_CNT(_n)		(GDM_BASE(_n) + 0x170)
+#define REG_FE_GDM_RX_ETH_CRC_ERR_CNT(_n)	(GDM_BASE(_n) + 0x174)
+#define REG_FE_GDM_RX_ETH_FRAG_CNT(_n)		(GDM_BASE(_n) + 0x178)
+#define REG_FE_GDM_RX_ETH_JABBER_CNT(_n)	(GDM_BASE(_n) + 0x17c)
+#define REG_FE_GDM_RX_ETH_RUNT_CNT(_n)		(GDM_BASE(_n) + 0x180)
+#define REG_FE_GDM_RX_ETH_LONG_CNT(_n)		(GDM_BASE(_n) + 0x184)
+#define REG_FE_GDM_RX_ETH_E64_CNT_L(_n)		(GDM_BASE(_n) + 0x188)
+#define REG_FE_GDM_RX_ETH_L64_CNT_L(_n)		(GDM_BASE(_n) + 0x18c)
+#define REG_FE_GDM_RX_ETH_L127_CNT_L(_n)	(GDM_BASE(_n) + 0x190)
+#define REG_FE_GDM_RX_ETH_L255_CNT_L(_n)	(GDM_BASE(_n) + 0x194)
+#define REG_FE_GDM_RX_ETH_L511_CNT_L(_n)	(GDM_BASE(_n) + 0x198)
+#define REG_FE_GDM_RX_ETH_L1023_CNT_L(_n)	(GDM_BASE(_n) + 0x19c)
+
+#define REG_PPE1_TB_HASH_CFG		(PPE1_BASE + 0x250)
+#define PPE1_SRAM_TABLE_EN_MASK		BIT(0)
+#define PPE1_SRAM_HASH1_EN_MASK		BIT(8)
+#define PPE1_DRAM_TABLE_EN_MASK		BIT(16)
+#define PPE1_DRAM_HASH1_EN_MASK		BIT(24)
+
+#define REG_FE_GDM_TX_OK_PKT_CNT_H(_n)		(GDM_BASE(_n) + 0x280)
+#define REG_FE_GDM_TX_OK_BYTE_CNT_H(_n)		(GDM_BASE(_n) + 0x284)
+#define REG_FE_GDM_TX_ETH_PKT_CNT_H(_n)		(GDM_BASE(_n) + 0x288)
+#define REG_FE_GDM_TX_ETH_BYTE_CNT_H(_n)	(GDM_BASE(_n) + 0x28c)
+
+#define REG_FE_GDM_RX_OK_PKT_CNT_H(_n)		(GDM_BASE(_n) + 0x290)
+#define REG_FE_GDM_RX_OK_BYTE_CNT_H(_n)		(GDM_BASE(_n) + 0x294)
+#define REG_FE_GDM_RX_ETH_PKT_CNT_H(_n)		(GDM_BASE(_n) + 0x298)
+#define REG_FE_GDM_RX_ETH_BYTE_CNT_H(_n)	(GDM_BASE(_n) + 0x29c)
+#define REG_FE_GDM_TX_ETH_E64_CNT_H(_n)		(GDM_BASE(_n) + 0x2b8)
+#define REG_FE_GDM_TX_ETH_L64_CNT_H(_n)		(GDM_BASE(_n) + 0x2bc)
+#define REG_FE_GDM_TX_ETH_L127_CNT_H(_n)	(GDM_BASE(_n) + 0x2c0)
+#define REG_FE_GDM_TX_ETH_L255_CNT_H(_n)	(GDM_BASE(_n) + 0x2c4)
+#define REG_FE_GDM_TX_ETH_L511_CNT_H(_n)	(GDM_BASE(_n) + 0x2c8)
+#define REG_FE_GDM_TX_ETH_L1023_CNT_H(_n)	(GDM_BASE(_n) + 0x2cc)
+#define REG_FE_GDM_RX_ETH_E64_CNT_H(_n)		(GDM_BASE(_n) + 0x2e8)
+#define REG_FE_GDM_RX_ETH_L64_CNT_H(_n)		(GDM_BASE(_n) + 0x2ec)
+#define REG_FE_GDM_RX_ETH_L127_CNT_H(_n)	(GDM_BASE(_n) + 0x2f0)
+#define REG_FE_GDM_RX_ETH_L255_CNT_H(_n)	(GDM_BASE(_n) + 0x2f4)
+#define REG_FE_GDM_RX_ETH_L511_CNT_H(_n)	(GDM_BASE(_n) + 0x2f8)
+#define REG_FE_GDM_RX_ETH_L1023_CNT_H(_n)	(GDM_BASE(_n) + 0x2fc)
+
+#define REG_GDM2_CHN_RLS		(GDM2_BASE + 0x20)
+#define MBI_RX_AGE_SEL_MASK		GENMASK(18, 17)
+#define MBI_TX_AGE_SEL_MASK		GENMASK(18, 17)
+
+#define REG_GDM3_FWD_CFG		GDM3_BASE
+#define GDM3_PAD_EN_MASK		BIT(28)
+
+#define REG_GDM4_FWD_CFG		(GDM4_BASE + 0x100)
+#define GDM4_PAD_EN_MASK		BIT(28)
+#define GDM4_SPORT_OFFSET0_MASK		GENMASK(11, 8)
+
+#define REG_GDM4_SRC_PORT_SET		(GDM4_BASE + 0x33c)
+#define GDM4_SPORT_OFF2_MASK		GENMASK(19, 16)
+#define GDM4_SPORT_OFF1_MASK		GENMASK(15, 12)
+#define GDM4_SPORT_OFF0_MASK		GENMASK(11, 8)
+
+#define REG_IP_FRAG_FP			0x2010
+#define IP_ASSEMBLE_PORT_MASK		GENMASK(24, 21)
+#define IP_ASSEMBLE_NBQ_MASK		GENMASK(20, 16)
+#define IP_FRAGMENT_PORT_MASK		GENMASK(8, 5)
+#define IP_FRAGMENT_NBQ_MASK		GENMASK(4, 0)
+
+#define REG_MC_VLAN_EN			0x2100
+#define MC_VLAN_EN_MASK			BIT(0)
+
+#define REG_MC_VLAN_CFG			0x2104
+#define MC_VLAN_CFG_CMD_DONE_MASK	BIT(31)
+#define MC_VLAN_CFG_TABLE_ID_MASK	GENMASK(21, 16)
+#define MC_VLAN_CFG_PORT_ID_MASK	GENMASK(11, 8)
+#define MC_VLAN_CFG_TABLE_SEL_MASK	BIT(4)
+#define MC_VLAN_CFG_RW_MASK		BIT(0)
+
+#define REG_MC_VLAN_DATA		0x2108
+
+#define REG_CDM5_RX_OQ1_DROP_CNT	0x29d4
+
+/* QDMA */
+#define REG_QDMA_GLOBAL_CFG			0x0004
+#define GLOBAL_CFG_RX_2B_OFFSET_MASK		BIT(31)
+#define GLOBAL_CFG_DMA_PREFERENCE_MASK		GENMASK(30, 29)
+#define GLOBAL_CFG_CPU_TXR_RR_MASK		BIT(28)
+#define GLOBAL_CFG_DSCP_BYTE_SWAP_MASK		BIT(27)
+#define GLOBAL_CFG_PAYLOAD_BYTE_SWAP_MASK	BIT(26)
+#define GLOBAL_CFG_MULTICAST_MODIFY_FP_MASK	BIT(25)
+#define GLOBAL_CFG_OAM_MODIFY_MASK		BIT(24)
+#define GLOBAL_CFG_RESET_MASK			BIT(23)
+#define GLOBAL_CFG_RESET_DONE_MASK		BIT(22)
+#define GLOBAL_CFG_MULTICAST_EN_MASK		BIT(21)
+#define GLOBAL_CFG_IRQ1_EN_MASK			BIT(20)
+#define GLOBAL_CFG_IRQ0_EN_MASK			BIT(19)
+#define GLOBAL_CFG_LOOPCNT_EN_MASK		BIT(18)
+#define GLOBAL_CFG_RD_BYPASS_WR_MASK		BIT(17)
+#define GLOBAL_CFG_QDMA_LOOPBACK_MASK		BIT(16)
+#define GLOBAL_CFG_LPBK_RXQ_SEL_MASK		GENMASK(13, 8)
+#define GLOBAL_CFG_CHECK_DONE_MASK		BIT(7)
+#define GLOBAL_CFG_TX_WB_DONE_MASK		BIT(6)
+#define GLOBAL_CFG_MAX_ISSUE_NUM_MASK		GENMASK(5, 4)
+#define GLOBAL_CFG_RX_DMA_BUSY_MASK		BIT(3)
+#define GLOBAL_CFG_RX_DMA_EN_MASK		BIT(2)
+#define GLOBAL_CFG_TX_DMA_BUSY_MASK		BIT(1)
+#define GLOBAL_CFG_TX_DMA_EN_MASK		BIT(0)
+
+#define REG_FWD_DSCP_BASE			0x0010
+#define REG_FWD_BUF_BASE			0x0014
+
+#define REG_HW_FWD_DSCP_CFG			0x0018
+#define HW_FWD_DSCP_PAYLOAD_SIZE_MASK		GENMASK(29, 28)
+#define HW_FWD_DSCP_SCATTER_LEN_MASK		GENMASK(17, 16)
+#define HW_FWD_DSCP_MIN_SCATTER_LEN_MASK	GENMASK(15, 0)
+
+#define REG_INT_STATUS(_n)		\
+	(((_n) == 4) ? 0x0730 :		\
+	 ((_n) == 3) ? 0x0724 :		\
+	 ((_n) == 2) ? 0x0720 :		\
+	 ((_n) == 1) ? 0x0024 : 0x0020)
+
+#define REG_INT_ENABLE(_n)		\
+	(((_n) == 4) ? 0x0750 :		\
+	 ((_n) == 3) ? 0x0744 :		\
+	 ((_n) == 2) ? 0x0740 :		\
+	 ((_n) == 1) ? 0x002c : 0x0028)
+
+/* QDMA_CSR_INT_ENABLE1 */
+#define RX15_COHERENT_INT_MASK		BIT(31)
+#define RX14_COHERENT_INT_MASK		BIT(30)
+#define RX13_COHERENT_INT_MASK		BIT(29)
+#define RX12_COHERENT_INT_MASK		BIT(28)
+#define RX11_COHERENT_INT_MASK		BIT(27)
+#define RX10_COHERENT_INT_MASK		BIT(26)
+#define RX9_COHERENT_INT_MASK		BIT(25)
+#define RX8_COHERENT_INT_MASK		BIT(24)
+#define RX7_COHERENT_INT_MASK		BIT(23)
+#define RX6_COHERENT_INT_MASK		BIT(22)
+#define RX5_COHERENT_INT_MASK		BIT(21)
+#define RX4_COHERENT_INT_MASK		BIT(20)
+#define RX3_COHERENT_INT_MASK		BIT(19)
+#define RX2_COHERENT_INT_MASK		BIT(18)
+#define RX1_COHERENT_INT_MASK		BIT(17)
+#define RX0_COHERENT_INT_MASK		BIT(16)
+#define TX7_COHERENT_INT_MASK		BIT(15)
+#define TX6_COHERENT_INT_MASK		BIT(14)
+#define TX5_COHERENT_INT_MASK		BIT(13)
+#define TX4_COHERENT_INT_MASK		BIT(12)
+#define TX3_COHERENT_INT_MASK		BIT(11)
+#define TX2_COHERENT_INT_MASK		BIT(10)
+#define TX1_COHERENT_INT_MASK		BIT(9)
+#define TX0_COHERENT_INT_MASK		BIT(8)
+#define CNT_OVER_FLOW_INT_MASK		BIT(7)
+#define IRQ1_FULL_INT_MASK		BIT(5)
+#define IRQ1_INT_MASK			BIT(4)
+#define HWFWD_DSCP_LOW_INT_MASK		BIT(3)
+#define HWFWD_DSCP_EMPTY_INT_MASK	BIT(2)
+#define IRQ0_FULL_INT_MASK		BIT(1)
+#define IRQ0_INT_MASK			BIT(0)
+
+#define TX_DONE_INT_MASK(_n)					\
+	((_n) ? IRQ1_INT_MASK | IRQ1_FULL_INT_MASK		\
+	      : IRQ0_INT_MASK | IRQ0_FULL_INT_MASK)
+
+#define INT_TX_MASK						\
+	(IRQ1_INT_MASK | IRQ1_FULL_INT_MASK |			\
+	 IRQ0_INT_MASK | IRQ0_FULL_INT_MASK)
+
+#define INT_IDX0_MASK						\
+	(TX0_COHERENT_INT_MASK | TX1_COHERENT_INT_MASK |	\
+	 TX2_COHERENT_INT_MASK | TX3_COHERENT_INT_MASK |	\
+	 TX4_COHERENT_INT_MASK | TX5_COHERENT_INT_MASK |	\
+	 TX6_COHERENT_INT_MASK | TX7_COHERENT_INT_MASK |	\
+	 RX0_COHERENT_INT_MASK | RX1_COHERENT_INT_MASK |	\
+	 RX2_COHERENT_INT_MASK | RX3_COHERENT_INT_MASK |	\
+	 RX4_COHERENT_INT_MASK | RX7_COHERENT_INT_MASK |	\
+	 RX8_COHERENT_INT_MASK | RX9_COHERENT_INT_MASK |	\
+	 RX15_COHERENT_INT_MASK | INT_TX_MASK)
+
+/* QDMA_CSR_INT_ENABLE2 */
+#define RX15_NO_CPU_DSCP_INT_MASK	BIT(31)
+#define RX14_NO_CPU_DSCP_INT_MASK	BIT(30)
+#define RX13_NO_CPU_DSCP_INT_MASK	BIT(29)
+#define RX12_NO_CPU_DSCP_INT_MASK	BIT(28)
+#define RX11_NO_CPU_DSCP_INT_MASK	BIT(27)
+#define RX10_NO_CPU_DSCP_INT_MASK	BIT(26)
+#define RX9_NO_CPU_DSCP_INT_MASK	BIT(25)
+#define RX8_NO_CPU_DSCP_INT_MASK	BIT(24)
+#define RX7_NO_CPU_DSCP_INT_MASK	BIT(23)
+#define RX6_NO_CPU_DSCP_INT_MASK	BIT(22)
+#define RX5_NO_CPU_DSCP_INT_MASK	BIT(21)
+#define RX4_NO_CPU_DSCP_INT_MASK	BIT(20)
+#define RX3_NO_CPU_DSCP_INT_MASK	BIT(19)
+#define RX2_NO_CPU_DSCP_INT_MASK	BIT(18)
+#define RX1_NO_CPU_DSCP_INT_MASK	BIT(17)
+#define RX0_NO_CPU_DSCP_INT_MASK	BIT(16)
+#define RX15_DONE_INT_MASK		BIT(15)
+#define RX14_DONE_INT_MASK		BIT(14)
+#define RX13_DONE_INT_MASK		BIT(13)
+#define RX12_DONE_INT_MASK		BIT(12)
+#define RX11_DONE_INT_MASK		BIT(11)
+#define RX10_DONE_INT_MASK		BIT(10)
+#define RX9_DONE_INT_MASK		BIT(9)
+#define RX8_DONE_INT_MASK		BIT(8)
+#define RX7_DONE_INT_MASK		BIT(7)
+#define RX6_DONE_INT_MASK		BIT(6)
+#define RX5_DONE_INT_MASK		BIT(5)
+#define RX4_DONE_INT_MASK		BIT(4)
+#define RX3_DONE_INT_MASK		BIT(3)
+#define RX2_DONE_INT_MASK		BIT(2)
+#define RX1_DONE_INT_MASK		BIT(1)
+#define RX0_DONE_INT_MASK		BIT(0)
+
+#define RX_DONE_INT_MASK					\
+	(RX0_DONE_INT_MASK | RX1_DONE_INT_MASK |		\
+	 RX2_DONE_INT_MASK | RX3_DONE_INT_MASK |		\
+	 RX4_DONE_INT_MASK | RX7_DONE_INT_MASK |		\
+	 RX8_DONE_INT_MASK | RX9_DONE_INT_MASK |		\
+	 RX15_DONE_INT_MASK)
+#define INT_IDX1_MASK						\
+	(RX_DONE_INT_MASK |					\
+	 RX0_NO_CPU_DSCP_INT_MASK | RX1_NO_CPU_DSCP_INT_MASK |	\
+	 RX2_NO_CPU_DSCP_INT_MASK | RX3_NO_CPU_DSCP_INT_MASK |	\
+	 RX4_NO_CPU_DSCP_INT_MASK | RX7_NO_CPU_DSCP_INT_MASK |	\
+	 RX8_NO_CPU_DSCP_INT_MASK | RX9_NO_CPU_DSCP_INT_MASK |	\
+	 RX15_NO_CPU_DSCP_INT_MASK)
+
+/* QDMA_CSR_INT_ENABLE5 */
+#define TX31_COHERENT_INT_MASK		BIT(31)
+#define TX30_COHERENT_INT_MASK		BIT(30)
+#define TX29_COHERENT_INT_MASK		BIT(29)
+#define TX28_COHERENT_INT_MASK		BIT(28)
+#define TX27_COHERENT_INT_MASK		BIT(27)
+#define TX26_COHERENT_INT_MASK		BIT(26)
+#define TX25_COHERENT_INT_MASK		BIT(25)
+#define TX24_COHERENT_INT_MASK		BIT(24)
+#define TX23_COHERENT_INT_MASK		BIT(23)
+#define TX22_COHERENT_INT_MASK		BIT(22)
+#define TX21_COHERENT_INT_MASK		BIT(21)
+#define TX20_COHERENT_INT_MASK		BIT(20)
+#define TX19_COHERENT_INT_MASK		BIT(19)
+#define TX18_COHERENT_INT_MASK		BIT(18)
+#define TX17_COHERENT_INT_MASK		BIT(17)
+#define TX16_COHERENT_INT_MASK		BIT(16)
+#define TX15_COHERENT_INT_MASK		BIT(15)
+#define TX14_COHERENT_INT_MASK		BIT(14)
+#define TX13_COHERENT_INT_MASK		BIT(13)
+#define TX12_COHERENT_INT_MASK		BIT(12)
+#define TX11_COHERENT_INT_MASK		BIT(11)
+#define TX10_COHERENT_INT_MASK		BIT(10)
+#define TX9_COHERENT_INT_MASK		BIT(9)
+#define TX8_COHERENT_INT_MASK		BIT(8)
+
+#define INT_IDX4_MASK						\
+	(TX8_COHERENT_INT_MASK | TX9_COHERENT_INT_MASK |	\
+	 TX10_COHERENT_INT_MASK | TX11_COHERENT_INT_MASK |	\
+	 TX12_COHERENT_INT_MASK | TX13_COHERENT_INT_MASK |	\
+	 TX14_COHERENT_INT_MASK | TX15_COHERENT_INT_MASK |	\
+	 TX16_COHERENT_INT_MASK | TX17_COHERENT_INT_MASK |	\
+	 TX18_COHERENT_INT_MASK | TX19_COHERENT_INT_MASK |	\
+	 TX20_COHERENT_INT_MASK | TX21_COHERENT_INT_MASK |	\
+	 TX22_COHERENT_INT_MASK | TX23_COHERENT_INT_MASK |	\
+	 TX24_COHERENT_INT_MASK | TX25_COHERENT_INT_MASK |	\
+	 TX26_COHERENT_INT_MASK | TX27_COHERENT_INT_MASK |	\
+	 TX28_COHERENT_INT_MASK | TX29_COHERENT_INT_MASK |	\
+	 TX30_COHERENT_INT_MASK | TX31_COHERENT_INT_MASK)
+
+#define REG_TX_IRQ_BASE(_n)		((_n) ? 0x0048 : 0x0050)
+
+#define REG_TX_IRQ_CFG(_n)		((_n) ? 0x004c : 0x0054)
+#define TX_IRQ_THR_MASK			GENMASK(27, 16)
+#define TX_IRQ_DEPTH_MASK		GENMASK(11, 0)
+
+#define REG_IRQ_CLEAR_LEN(_n)		((_n) ? 0x0064 : 0x0058)
+#define IRQ_CLEAR_LEN_MASK		GENMASK(7, 0)
+
+#define REG_IRQ_STATUS(_n)		((_n) ? 0x0068 : 0x005c)
+#define IRQ_ENTRY_LEN_MASK		GENMASK(27, 16)
+#define IRQ_HEAD_IDX_MASK		GENMASK(11, 0)
+
+#define REG_TX_RING_BASE(_n)	\
+	(((_n) < 8) ? 0x0100 + ((_n) << 5) : 0x0b00 + (((_n) - 8) << 5))
+
+#define REG_TX_RING_BLOCKING(_n)	\
+	(((_n) < 8) ? 0x0104 + ((_n) << 5) : 0x0b04 + (((_n) - 8) << 5))
+
+#define TX_RING_IRQ_BLOCKING_MAP_MASK			BIT(6)
+#define TX_RING_IRQ_BLOCKING_CFG_MASK			BIT(4)
+#define TX_RING_IRQ_BLOCKING_TX_DROP_EN_MASK		BIT(2)
+#define TX_RING_IRQ_BLOCKING_MAX_TH_TXRING_EN_MASK	BIT(1)
+#define TX_RING_IRQ_BLOCKING_MIN_TH_TXRING_EN_MASK	BIT(0)
+
+#define REG_TX_CPU_IDX(_n)	\
+	(((_n) < 8) ? 0x0108 + ((_n) << 5) : 0x0b08 + (((_n) - 8) << 5))
+
+#define TX_RING_CPU_IDX_MASK		GENMASK(15, 0)
+
+#define REG_TX_DMA_IDX(_n)	\
+	(((_n) < 8) ? 0x010c + ((_n) << 5) : 0x0b0c + (((_n) - 8) << 5))
+
+#define TX_RING_DMA_IDX_MASK		GENMASK(15, 0)
+
+#define IRQ_RING_IDX_MASK		GENMASK(20, 16)
+#define IRQ_DESC_IDX_MASK		GENMASK(15, 0)
+
+#define REG_RX_RING_BASE(_n)	\
+	(((_n) < 16) ? 0x0200 + ((_n) << 5) : 0x0e00 + (((_n) - 16) << 5))
+
+#define REG_RX_RING_SIZE(_n)	\
+	(((_n) < 16) ? 0x0204 + ((_n) << 5) : 0x0e04 + (((_n) - 16) << 5))
+
+#define RX_RING_THR_MASK		GENMASK(31, 16)
+#define RX_RING_SIZE_MASK		GENMASK(15, 0)
+
+#define REG_RX_CPU_IDX(_n)	\
+	(((_n) < 16) ? 0x0208 + ((_n) << 5) : 0x0e08 + (((_n) - 16) << 5))
+
+#define RX_RING_CPU_IDX_MASK		GENMASK(15, 0)
+
+#define REG_RX_DMA_IDX(_n)	\
+	(((_n) < 16) ? 0x020c + ((_n) << 5) : 0x0e0c + (((_n) - 16) << 5))
+
+#define REG_RX_DELAY_INT_IDX(_n)	\
+	(((_n) < 16) ? 0x0210 + ((_n) << 5) : 0x0e10 + (((_n) - 16) << 5))
+
+#define RX_DELAY_INT_MASK		GENMASK(15, 0)
+
+#define RX_RING_DMA_IDX_MASK		GENMASK(15, 0)
+
+#define REG_INGRESS_TRTCM_CFG		0x0070
+#define INGRESS_TRTCM_EN_MASK		BIT(31)
+#define INGRESS_TRTCM_MODE_MASK		BIT(30)
+#define INGRESS_SLOW_TICK_RATIO_MASK	GENMASK(29, 16)
+#define INGRESS_FAST_TICK_MASK		GENMASK(15, 0)
+
+#define REG_TXQ_DIS_CFG_BASE(_n)	((_n) ? 0x20a0 : 0x00a0)
+#define REG_TXQ_DIS_CFG(_n, _m)		(REG_TXQ_DIS_CFG_BASE((_n)) + (_m) << 2)
+
+#define REG_LMGR_INIT_CFG		0x1000
+#define LMGR_INIT_START			BIT(31)
+#define LMGR_SRAM_MODE_MASK		BIT(30)
+#define HW_FWD_PKTSIZE_OVERHEAD_MASK	GENMASK(27, 20)
+#define HW_FWD_DESC_NUM_MASK		GENMASK(16, 0)
+
+#define REG_FWD_DSCP_LOW_THR		0x1004
+#define FWD_DSCP_LOW_THR_MASK		GENMASK(17, 0)
+
+#define REG_EGRESS_RATE_METER_CFG		0x100c
+#define EGRESS_RATE_METER_EN_MASK		BIT(29)
+#define EGRESS_RATE_METER_EQ_RATE_EN_MASK	BIT(17)
+#define EGRESS_RATE_METER_WINDOW_SZ_MASK	GENMASK(16, 12)
+#define EGRESS_RATE_METER_TIMESLICE_MASK	GENMASK(10, 0)
+
+#define REG_EGRESS_TRTCM_CFG		0x1010
+#define EGRESS_TRTCM_EN_MASK		BIT(31)
+#define EGRESS_TRTCM_MODE_MASK		BIT(30)
+#define EGRESS_SLOW_TICK_RATIO_MASK	GENMASK(29, 16)
+#define EGRESS_FAST_TICK_MASK		GENMASK(15, 0)
+
+#define REG_TXWRR_MODE_CFG		0x1020
+#define TWRR_WEIGHT_SCALE_MASK		BIT(31)
+#define TWRR_WEIGHT_BASE_MASK		BIT(3)
+
+#define REG_PSE_BUF_USAGE_CFG		0x1028
+#define PSE_BUF_ESTIMATE_EN_MASK	BIT(29)
+
+#define REG_GLB_TRTCM_CFG		0x1080
+#define GLB_TRTCM_EN_MASK		BIT(31)
+#define GLB_TRTCM_MODE_MASK		BIT(30)
+#define GLB_SLOW_TICK_RATIO_MASK	GENMASK(29, 16)
+#define GLB_FAST_TICK_MASK		GENMASK(15, 0)
+
+#define REG_TXQ_CNGST_CFG		0x10a0
+#define TXQ_CNGST_DROP_EN		BIT(31)
+#define TXQ_CNGST_DEI_DROP_EN		BIT(30)
+
+#define REG_SLA_TRTCM_CFG		0x1150
+#define SLA_TRTCM_EN_MASK		BIT(31)
+#define SLA_TRTCM_MODE_MASK		BIT(30)
+#define SLA_SLOW_TICK_RATIO_MASK	GENMASK(29, 16)
+#define SLA_FAST_TICK_MASK		GENMASK(15, 0)
+
+/* CTRL */
+#define QDMA_DESC_DONE_MASK		BIT(31)
+#define QDMA_DESC_DROP_MASK		BIT(30) /* tx: drop - rx: overflow */
+#define QDMA_DESC_MORE_MASK		BIT(29) /* more SG elements */
+#define QDMA_DESC_DEI_MASK		BIT(25)
+#define QDMA_DESC_NO_DROP_MASK		BIT(24)
+#define QDMA_DESC_LEN_MASK		GENMASK(15, 0)
+/* DATA */
+#define QDMA_DESC_NEXT_ID_MASK		GENMASK(15, 0)
+/* TX MSG0 */
+#define QDMA_ETH_TXMSG_MIC_IDX_MASK	BIT(30)
+#define QDMA_ETH_TXMSG_SP_TAG_MASK	GENMASK(29, 14)
+#define QDMA_ETH_TXMSG_ICO_MASK		BIT(13)
+#define QDMA_ETH_TXMSG_UCO_MASK		BIT(12)
+#define QDMA_ETH_TXMSG_TCO_MASK		BIT(11)
+#define QDMA_ETH_TXMSG_TSO_MASK		BIT(10)
+#define QDMA_ETH_TXMSG_FAST_MASK	BIT(9)
+#define QDMA_ETH_TXMSG_OAM_MASK		BIT(8)
+#define QDMA_ETH_TXMSG_CHAN_MASK	GENMASK(7, 3)
+#define QDMA_ETH_TXMSG_QUEUE_MASK	GENMASK(2, 0)
+/* TX MSG1 */
+#define QDMA_ETH_TXMSG_NO_DROP		BIT(31)
+#define QDMA_ETH_TXMSG_METER_MASK	GENMASK(30, 24)	/* 0x7f no meters */
+#define QDMA_ETH_TXMSG_FPORT_MASK	GENMASK(23, 20)
+#define QDMA_ETH_TXMSG_NBOQ_MASK	GENMASK(19, 15)
+#define QDMA_ETH_TXMSG_HWF_MASK		BIT(14)
+#define QDMA_ETH_TXMSG_HOP_MASK		BIT(13)
+#define QDMA_ETH_TXMSG_PTP_MASK		BIT(12)
+#define QDMA_ETH_TXMSG_ACNT_G1_MASK	GENMASK(10, 6)	/* 0x1f do not count */
+#define QDMA_ETH_TXMSG_ACNT_G0_MASK	GENMASK(5, 0)	/* 0x3f do not count */
+
+/* RX MSG1 */
+#define QDMA_ETH_RXMSG_DEI_MASK		BIT(31)
+#define QDMA_ETH_RXMSG_IP6_MASK		BIT(30)
+#define QDMA_ETH_RXMSG_IP4_MASK		BIT(29)
+#define QDMA_ETH_RXMSG_IP4F_MASK	BIT(28)
+#define QDMA_ETH_RXMSG_L4_VALID_MASK	BIT(27)
+#define QDMA_ETH_RXMSG_L4F_MASK		BIT(26)
+#define QDMA_ETH_RXMSG_SPORT_MASK	GENMASK(25, 21)
+#define QDMA_ETH_RXMSG_CRSN_MASK	GENMASK(20, 16)
+#define QDMA_ETH_RXMSG_PPE_ENTRY_MASK	GENMASK(15, 0)
+
+struct airoha_qdma_desc {
+	__le32 rsv;
+	__le32 ctrl;
+	__le32 addr;
+	__le32 data;
+	__le32 msg0;
+	__le32 msg1;
+	__le32 msg2;
+	__le32 msg3;
+};
+
+/* CTRL0 */
+#define QDMA_FWD_DESC_CTX_MASK		BIT(31)
+#define QDMA_FWD_DESC_RING_MASK		GENMASK(30, 28)
+#define QDMA_FWD_DESC_IDX_MASK		GENMASK(27, 16)
+#define QDMA_FWD_DESC_LEN_MASK		GENMASK(15, 0)
+/* CTRL1 */
+#define QDMA_FWD_DESC_FIRST_IDX_MASK	GENMASK(15, 0)
+/* CTRL2 */
+#define QDMA_FWD_DESC_MORE_PKT_NUM_MASK	GENMASK(2, 0)
+
+struct airoha_qdma_fwd_desc {
+	__le32 addr;
+	__le32 ctrl0;
+	__le32 ctrl1;
+	__le32 ctrl2;
+	__le32 msg0;
+	__le32 msg1;
+	__le32 rsv0;
+	__le32 rsv1;
+};
+
+enum {
+	QDMA_INT_REG_IDX0,
+	QDMA_INT_REG_IDX1,
+	QDMA_INT_REG_IDX2,
+	QDMA_INT_REG_IDX3,
+	QDMA_INT_REG_IDX4,
+	QDMA_INT_REG_MAX
+};
+
+enum {
+	XSI_PCIE0_PORT,
+	XSI_PCIE1_PORT,
+	XSI_USB_PORT,
+	XSI_AE_PORT,
+	XSI_ETH_PORT,
+};
+
+enum {
+	XSI_PCIE0_VIP_PORT_MASK	= BIT(22),
+	XSI_PCIE1_VIP_PORT_MASK	= BIT(23),
+	XSI_USB_VIP_PORT_MASK	= BIT(25),
+	XSI_ETH_VIP_PORT_MASK	= BIT(24),
+};
+
+enum {
+	DEV_STATE_INITIALIZED,
+};
+
+enum {
+	CDM_CRSN_QSEL_Q1 = 1,
+	CDM_CRSN_QSEL_Q5 = 5,
+	CDM_CRSN_QSEL_Q6 = 6,
+	CDM_CRSN_QSEL_Q15 = 15,
+};
+
+enum {
+	CRSN_08 = 0x8,
+	CRSN_21 = 0x15, /* KA */
+	CRSN_22 = 0x16, /* hit bind and force route to CPU */
+	CRSN_24 = 0x18,
+	CRSN_25 = 0x19,
+};
+
+enum {
+	FE_PSE_PORT_CDM1,
+	FE_PSE_PORT_GDM1,
+	FE_PSE_PORT_GDM2,
+	FE_PSE_PORT_GDM3,
+	FE_PSE_PORT_PPE1,
+	FE_PSE_PORT_CDM2,
+	FE_PSE_PORT_CDM3,
+	FE_PSE_PORT_CDM4,
+	FE_PSE_PORT_PPE2,
+	FE_PSE_PORT_GDM4,
+	FE_PSE_PORT_CDM5,
+	FE_PSE_PORT_DROP = 0xf,
+};
+
+struct airoha_queue_entry {
+	union {
+		void *buf;
+		struct sk_buff *skb;
+	};
+	dma_addr_t dma_addr;
+	u16 dma_len;
+};
+
+struct airoha_queue {
+	struct airoha_eth *eth;
+
+	/* protect concurrent queue accesses */
+	spinlock_t lock;
+	struct airoha_queue_entry *entry;
+	struct airoha_qdma_desc *desc;
+	u16 head;
+	u16 tail;
+
+	int queued;
+	int ndesc;
+	int free_thr;
+	int buf_size;
+
+	struct napi_struct napi;
+	struct page_pool *page_pool;
+};
+
+struct airoha_tx_irq_queue {
+	struct airoha_eth *eth;
+
+	struct napi_struct napi;
+	u32 *q;
+
+	int size;
+	int queued;
+	u16 head;
+};
+
+struct airoha_hw_stats {
+	/* protect concurrent hw_stats accesses */
+	spinlock_t lock;
+	struct u64_stats_sync syncp;
+
+	/* get_stats64 */
+	u64 rx_ok_pkts;
+	u64 tx_ok_pkts;
+	u64 rx_ok_bytes;
+	u64 tx_ok_bytes;
+	u64 rx_multicast;
+	u64 rx_errors;
+	u64 rx_drops;
+	u64 tx_drops;
+	u64 rx_crc_error;
+	u64 rx_over_errors;
+	/* ethtool stats */
+	u64 tx_broadcast;
+	u64 tx_multicast;
+	u64 tx_len[7];
+	u64 rx_broadcast;
+	u64 rx_fragment;
+	u64 rx_jabber;
+	u64 rx_len[7];
+};
+
+struct airoha_gdm_port {
+	struct net_device *dev;
+	struct airoha_eth *eth;
+	int id;
+
+	struct airoha_hw_stats stats;
+};
+
+struct airoha_eth {
+	struct device *dev;
+
+	unsigned long state;
+
+	void __iomem *qdma_regs;
+	void __iomem *fe_regs;
+
+	/* protect concurrent irqmask accesses */
+	spinlock_t irq_lock;
+	u32 irqmask[QDMA_INT_REG_MAX];
+	int irq;
+
+	struct reset_control_bulk_data rsts[AIROHA_MAX_NUM_RSTS];
+	struct reset_control_bulk_data xsi_rsts[AIROHA_MAX_NUM_XSI_RSTS];
+
+	struct airoha_gdm_port *ports[AIROHA_MAX_NUM_GDM_PORTS];
+
+	struct net_device *napi_dev;
+	struct airoha_queue q_tx[AIROHA_NUM_TX_RING];
+	struct airoha_queue q_rx[AIROHA_NUM_RX_RING];
+
+	struct airoha_tx_irq_queue q_tx_irq[AIROHA_NUM_TX_IRQ];
+
+	/* descriptor and packet buffers for qdma hw forward */
+	struct {
+		void *desc;
+		void *q;
+	} hfwd;
+};
+
+static u32 airoha_rr(void __iomem *base, u32 offset)
+{
+	return readl(base + offset);
+}
+
+static void airoha_wr(void __iomem *base, u32 offset, u32 val)
+{
+	writel(val, base + offset);
+}
+
+static u32 airoha_rmw(void __iomem *base, u32 offset, u32 mask, u32 val)
+{
+	val |= (airoha_rr(base, offset) & ~mask);
+	airoha_wr(base, offset, val);
+
+	return val;
+}
+
+#define airoha_fe_rr(eth, offset)				\
+	airoha_rr((eth)->fe_regs, (offset))
+#define airoha_fe_wr(eth, offset, val)				\
+	airoha_wr((eth)->fe_regs, (offset), (val))
+#define airoha_fe_rmw(eth, offset, mask, val)			\
+	airoha_rmw((eth)->fe_regs, (offset), (mask), (val))
+#define airoha_fe_set(eth, offset, val)				\
+	airoha_rmw((eth)->fe_regs, (offset), 0, (val))
+#define airoha_fe_clear(eth, offset, val)			\
+	airoha_rmw((eth)->fe_regs, (offset), (val), 0)
+
+#define airoha_qdma_rr(eth, offset)				\
+	airoha_rr((eth)->qdma_regs, (offset))
+#define airoha_qdma_wr(eth, offset, val)			\
+	airoha_wr((eth)->qdma_regs, (offset), (val))
+#define airoha_qdma_rmw(eth, offset, mask, val)			\
+	airoha_rmw((eth)->qdma_regs, (offset), (mask), (val))
+#define airoha_qdma_set(eth, offset, val)			\
+	airoha_rmw((eth)->qdma_regs, (offset), 0, (val))
+#define airoha_qdma_clear(eth, offset, val)			\
+	airoha_rmw((eth)->qdma_regs, (offset), (val), 0)
+
+static void airoha_qdma_set_irqmask(struct airoha_eth *eth, int index,
+				    u32 clear, u32 set)
+{
+	unsigned long flags;
+
+	if (WARN_ON_ONCE(index >= ARRAY_SIZE(eth->irqmask)))
+		return;
+
+	spin_lock_irqsave(&eth->irq_lock, flags);
+
+	eth->irqmask[index] &= ~clear;
+	eth->irqmask[index] |= set;
+	airoha_qdma_wr(eth, REG_INT_ENABLE(index), eth->irqmask[index]);
+	/* Read irq_enable register in order to guarantee the update above
+	 * completes in the spinlock critical section.
+	 */
+	airoha_qdma_rr(eth, REG_INT_ENABLE(index));
+
+	spin_unlock_irqrestore(&eth->irq_lock, flags);
+}
+
+static void airoha_qdma_irq_enable(struct airoha_eth *eth, int index,
+				   u32 mask)
+{
+	airoha_qdma_set_irqmask(eth, index, 0, mask);
+}
+
+static void airoha_qdma_irq_disable(struct airoha_eth *eth, int index,
+				    u32 mask)
+{
+	airoha_qdma_set_irqmask(eth, index, mask, 0);
+}
+
+static void airoha_set_macaddr(struct airoha_eth *eth, const u8 *addr)
+{
+	u32 val;
+
+	val = (addr[0] << 16) | (addr[1] << 8) | addr[2];
+	airoha_fe_wr(eth, REG_FE_LAN_MAC_H, val);
+
+	val = (addr[3] << 16) | (addr[4] << 8) | addr[5];
+	airoha_fe_wr(eth, REG_FE_LAN_MAC_LMIN, val);
+	airoha_fe_wr(eth, REG_FE_LAN_MAC_LMAX, val);
+}
+
+static void airoha_set_gdm_port_fwd_cfg(struct airoha_eth *eth, u32 addr,
+					u32 val)
+{
+	airoha_fe_rmw(eth, addr, GDM_OCFQ_MASK,
+		      FIELD_PREP(GDM_OCFQ_MASK, val));
+	airoha_fe_rmw(eth, addr, GDM_MCFQ_MASK,
+		      FIELD_PREP(GDM_MCFQ_MASK, val));
+	airoha_fe_rmw(eth, addr, GDM_BCFQ_MASK,
+		      FIELD_PREP(GDM_BCFQ_MASK, val));
+	airoha_fe_rmw(eth, addr, GDM_UCFQ_MASK,
+		      FIELD_PREP(GDM_UCFQ_MASK, val));
+}
+
+static int airoha_set_gdm_port(struct airoha_eth *eth, int port, bool enable)
+{
+	u32 val = enable ? FE_PSE_PORT_PPE1 : FE_PSE_PORT_DROP;
+	u32 vip_port, cfg_addr;
+
+	switch (port) {
+	case XSI_PCIE0_PORT:
+		vip_port = XSI_PCIE0_VIP_PORT_MASK;
+		cfg_addr = REG_GDM_FWD_CFG(3);
+		break;
+	case XSI_PCIE1_PORT:
+		vip_port = XSI_PCIE1_VIP_PORT_MASK;
+		cfg_addr = REG_GDM_FWD_CFG(3);
+		break;
+	case XSI_USB_PORT:
+		vip_port = XSI_USB_VIP_PORT_MASK;
+		cfg_addr = REG_GDM_FWD_CFG(4);
+		break;
+	case XSI_ETH_PORT:
+		vip_port = XSI_ETH_VIP_PORT_MASK;
+		cfg_addr = REG_GDM_FWD_CFG(4);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (enable) {
+		airoha_fe_set(eth, REG_FE_VIP_PORT_EN, vip_port);
+		airoha_fe_set(eth, REG_FE_IFC_PORT_EN, vip_port);
+	} else {
+		airoha_fe_clear(eth, REG_FE_VIP_PORT_EN, vip_port);
+		airoha_fe_clear(eth, REG_FE_IFC_PORT_EN, vip_port);
+	}
+
+	airoha_set_gdm_port_fwd_cfg(eth, cfg_addr, val);
+
+	return 0;
+}
+
+static int airoha_set_gdm_ports(struct airoha_eth *eth, bool enable)
+{
+	const int port_list[] = {
+		XSI_PCIE0_PORT,
+		XSI_PCIE1_PORT,
+		XSI_USB_PORT,
+		XSI_ETH_PORT
+	};
+	int i, err;
+
+	for (i = 0; i < ARRAY_SIZE(port_list); i++) {
+		err = airoha_set_gdm_port(eth, port_list[i], enable);
+		if (err)
+			goto error;
+	}
+
+	return 0;
+
+error:
+	for (i--; i >= 0; i++)
+		airoha_set_gdm_port(eth, port_list[i], false);
+
+	return err;
+}
+
+static void airoha_fe_maccr_init(struct airoha_eth *eth)
+{
+	int p;
+
+	for (p = 1; p <= ARRAY_SIZE(eth->ports); p++) {
+		airoha_fe_set(eth, REG_GDM_FWD_CFG(p),
+			      GDM_TCP_CKSUM | GDM_UDP_CKSUM | GDM_IP4_CKSUM |
+			      GDM_DROP_CRC_ERR);
+		airoha_set_gdm_port_fwd_cfg(eth, REG_GDM_FWD_CFG(p),
+					    FE_PSE_PORT_CDM1);
+		airoha_fe_rmw(eth, REG_GDM_LEN_CFG(p),
+			      GDM_SHORT_LEN_MASK | GDM_LONG_LEN_MASK,
+			      FIELD_PREP(GDM_SHORT_LEN_MASK, 60) |
+			      FIELD_PREP(GDM_LONG_LEN_MASK, 4004));
+	}
+
+	airoha_fe_rmw(eth, REG_CDM1_VLAN_CTRL, CDM1_VLAN_MASK,
+		      FIELD_PREP(CDM1_VLAN_MASK, 0x8100));
+
+	airoha_fe_set(eth, REG_FE_CPORT_CFG, FE_CPORT_PAD);
+}
+
+static void airoha_fe_vip_setup(struct airoha_eth *eth)
+{
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(3), ETH_P_PPP_DISC);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(3), PATN_FCPU_EN_MASK | PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(4), PPP_LCP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(4),
+		     PATN_FCPU_EN_MASK | FIELD_PREP(PATN_TYPE_MASK, 1) |
+		     PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(6), PPP_IPCP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(6),
+		     PATN_FCPU_EN_MASK | FIELD_PREP(PATN_TYPE_MASK, 1) |
+		     PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(7), PPP_CHAP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(7),
+		     PATN_FCPU_EN_MASK | FIELD_PREP(PATN_TYPE_MASK, 1) |
+		     PATN_EN_MASK);
+
+	/* BOOTP (0x43) */
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(8), 0x43);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(8),
+		     PATN_FCPU_EN_MASK | PATN_SP_EN_MASK |
+		     FIELD_PREP(PATN_TYPE_MASK, 4) | PATN_EN_MASK);
+
+	/* BOOTP (0x44) */
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(9), 0x44);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(9),
+		     PATN_FCPU_EN_MASK | PATN_SP_EN_MASK |
+		     FIELD_PREP(PATN_TYPE_MASK, 4) | PATN_EN_MASK);
+
+	/* ISAKMP */
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(10), 0x1f401f4);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(10),
+		     PATN_FCPU_EN_MASK | PATN_DP_EN_MASK | PATN_SP_EN_MASK |
+		     FIELD_PREP(PATN_TYPE_MASK, 4) | PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(11), PPP_IPV6CP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(11),
+		     PATN_FCPU_EN_MASK | FIELD_PREP(PATN_TYPE_MASK, 1) |
+		     PATN_EN_MASK);
+
+	/* DHCPv6 */
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(12), 0x2220223);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(12),
+		     PATN_FCPU_EN_MASK | PATN_DP_EN_MASK | PATN_SP_EN_MASK |
+		     FIELD_PREP(PATN_TYPE_MASK, 4) | PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(19), PPP_PAP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(19),
+		     PATN_FCPU_EN_MASK | FIELD_PREP(PATN_TYPE_MASK, 1) |
+		     PATN_EN_MASK);
+
+	/* ETH->ETH_P_1905 (0x893a) */
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(20), 0x893a);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(20),
+		     PATN_FCPU_EN_MASK | PATN_EN_MASK);
+
+	airoha_fe_wr(eth, REG_FE_VIP_PATN(21), ETH_P_LLDP);
+	airoha_fe_wr(eth, REG_FE_VIP_EN(21),
+		     PATN_FCPU_EN_MASK | PATN_EN_MASK);
+}
+
+static u32 airoha_fe_get_pse_queue_rsv_pages(struct airoha_eth *eth,
+					     u32 port, u32 queue)
+{
+	u32 val;
+
+	airoha_fe_rmw(eth, REG_FE_PSE_QUEUE_CFG_WR,
+		      PSE_CFG_PORT_ID_MASK | PSE_CFG_QUEUE_ID_MASK,
+		      FIELD_PREP(PSE_CFG_PORT_ID_MASK, port) |
+		      FIELD_PREP(PSE_CFG_QUEUE_ID_MASK, queue));
+	val = airoha_fe_rr(eth, REG_FE_PSE_QUEUE_CFG_VAL);
+
+	return FIELD_GET(PSE_CFG_OQ_RSV_MASK, val);
+}
+
+static void airoha_fe_set_pse_queue_rsv_pages(struct airoha_eth *eth,
+					      u32 port, u32 queue, u32 val)
+{
+	airoha_fe_rmw(eth, REG_FE_PSE_QUEUE_CFG_VAL, PSE_CFG_OQ_RSV_MASK,
+		      FIELD_PREP(PSE_CFG_OQ_RSV_MASK, val));
+	airoha_fe_rmw(eth, REG_FE_PSE_QUEUE_CFG_WR,
+		      PSE_CFG_PORT_ID_MASK | PSE_CFG_QUEUE_ID_MASK |
+		      PSE_CFG_WR_EN_MASK | PSE_CFG_OQRSV_SEL_MASK,
+		      FIELD_PREP(PSE_CFG_PORT_ID_MASK, port) |
+		      FIELD_PREP(PSE_CFG_QUEUE_ID_MASK, queue) |
+		      PSE_CFG_WR_EN_MASK | PSE_CFG_OQRSV_SEL_MASK);
+}
+
+static int airoha_fe_set_pse_oq_rsv(struct airoha_eth *eth,
+				    u32 port, u32 queue, u32 val)
+{
+	u32 orig_val, tmp, all_rsv, fq_limit;
+
+	airoha_fe_set_pse_queue_rsv_pages(eth, port, queue, val);
+
+	/* modify all rsv */
+	orig_val = airoha_fe_get_pse_queue_rsv_pages(eth, port, queue);
+	tmp = airoha_fe_rr(eth, REG_FE_PSE_BUF_SET);
+	all_rsv = FIELD_GET(PSE_ALLRSV_MASK, tmp);
+	all_rsv += (val - orig_val);
+	airoha_fe_rmw(eth, REG_FE_PSE_BUF_SET, PSE_ALLRSV_MASK,
+		      FIELD_PREP(PSE_ALLRSV_MASK, all_rsv));
+
+	/* modify hthd */
+	tmp = airoha_fe_rr(eth, PSE_FQ_CFG);
+	fq_limit = FIELD_GET(PSE_FQ_LIMIT_MASK, tmp);
+	tmp = fq_limit - all_rsv - 0x20;
+	airoha_fe_rmw(eth, REG_PSE_SHARE_USED_THD,
+		      PSE_SHARE_USED_HTHD_MASK,
+		      FIELD_PREP(PSE_SHARE_USED_HTHD_MASK, tmp));
+
+	tmp = fq_limit - all_rsv - 0x100;
+	airoha_fe_rmw(eth, REG_PSE_SHARE_USED_THD,
+		      PSE_SHARE_USED_MTHD_MASK,
+		      FIELD_PREP(PSE_SHARE_USED_MTHD_MASK, tmp));
+	tmp = (3 * tmp) >> 2;
+	airoha_fe_rmw(eth, REG_FE_PSE_BUF_SET,
+		      PSE_SHARE_USED_LTHD_MASK,
+		      FIELD_PREP(PSE_SHARE_USED_LTHD_MASK, tmp));
+
+	return 0;
+}
+
+static void airoha_fe_pse_ports_init(struct airoha_eth *eth)
+{
+	const u32 pse_port_num_queues[] = {
+		[FE_PSE_PORT_CDM1] = 6,
+		[FE_PSE_PORT_GDM1] = 6,
+		[FE_PSE_PORT_GDM2] = 32,
+		[FE_PSE_PORT_GDM3] = 6,
+		[FE_PSE_PORT_PPE1] = 4,
+		[FE_PSE_PORT_CDM2] = 6,
+		[FE_PSE_PORT_CDM3] = 8,
+		[FE_PSE_PORT_CDM4] = 10,
+		[FE_PSE_PORT_PPE2] = 4,
+		[FE_PSE_PORT_GDM4] = 2,
+		[FE_PSE_PORT_CDM5] = 2,
+	};
+	int q;
+
+	/* hw misses PPE2 oq rsv */
+	airoha_fe_set(eth, REG_FE_PSE_BUF_SET,
+		      PSE_RSV_PAGES * pse_port_num_queues[FE_PSE_PORT_PPE2]);
+
+	/* CMD1 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_CDM1]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_CDM1, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* GMD1 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_GDM1]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_GDM1, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* GMD2 */
+	for (q = 6; q < pse_port_num_queues[FE_PSE_PORT_GDM2]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_GDM2, q, 0);
+	/* GMD3 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_GDM3]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_GDM3, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* PPE1 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_PPE1]; q++) {
+		if (q < pse_port_num_queues[FE_PSE_PORT_PPE1])
+			airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE1, q,
+						 PSE_QUEUE_RSV_PAGES);
+		else
+			airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE1, q, 0);
+	}
+	/* CDM2 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_CDM2]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_CDM2, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* CDM3 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_CDM3] - 1; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_CDM3, q, 0);
+	/* CDM4 */
+	for (q = 4; q < pse_port_num_queues[FE_PSE_PORT_CDM4]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_CDM4, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* PPE2 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_PPE2]; q++) {
+		if (q < pse_port_num_queues[FE_PSE_PORT_PPE2] / 2)
+			airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE2, q,
+						 PSE_QUEUE_RSV_PAGES);
+		else
+			airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_PPE2, q, 0);
+	}
+	/* GMD4 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_GDM4]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_GDM4, q,
+					 PSE_QUEUE_RSV_PAGES);
+	/* CDM5 */
+	for (q = 0; q < pse_port_num_queues[FE_PSE_PORT_CDM5]; q++)
+		airoha_fe_set_pse_oq_rsv(eth, FE_PSE_PORT_CDM5, q,
+					 PSE_QUEUE_RSV_PAGES);
+}
+
+static int airoha_fe_mc_vlan_clear(struct airoha_eth *eth)
+{
+	int i;
+
+	for (i = 0; i < AIROHA_FE_MC_MAX_VLAN_TABLE; i++) {
+		int err, j;
+		u32 val;
+
+		airoha_fe_wr(eth, REG_MC_VLAN_DATA, 0x0);
+
+		val = FIELD_PREP(MC_VLAN_CFG_TABLE_ID_MASK, i) |
+		      MC_VLAN_CFG_TABLE_SEL_MASK | MC_VLAN_CFG_RW_MASK;
+		airoha_fe_wr(eth, REG_MC_VLAN_CFG, val);
+		err = read_poll_timeout(airoha_fe_rr, val,
+					val & MC_VLAN_CFG_CMD_DONE_MASK,
+					USEC_PER_MSEC, 5 * USEC_PER_MSEC,
+					false, eth, REG_MC_VLAN_CFG);
+		if (err)
+			return err;
+
+		for (j = 0; j < AIROHA_FE_MC_MAX_VLAN_PORT; j++) {
+			airoha_fe_wr(eth, REG_MC_VLAN_DATA, 0x0);
+
+			val = FIELD_PREP(MC_VLAN_CFG_TABLE_ID_MASK, i) |
+			      FIELD_PREP(MC_VLAN_CFG_PORT_ID_MASK, j) |
+			      MC_VLAN_CFG_RW_MASK;
+			airoha_fe_wr(eth, REG_MC_VLAN_CFG, val);
+			err = read_poll_timeout(airoha_fe_rr, val,
+						val & MC_VLAN_CFG_CMD_DONE_MASK,
+						USEC_PER_MSEC,
+						5 * USEC_PER_MSEC, false, eth,
+						REG_MC_VLAN_CFG);
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
+}
+
+static void airoha_fe_crsn_qsel_init(struct airoha_eth *eth)
+{
+	/* CDM1_CRSN_QSEL */
+	airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_22 >> 2),
+		      CDM1_CRSN_QSEL_REASON_MASK(CRSN_22),
+		      FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_22),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_08 >> 2),
+		      CDM1_CRSN_QSEL_REASON_MASK(CRSN_08),
+		      FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_08),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_21 >> 2),
+		      CDM1_CRSN_QSEL_REASON_MASK(CRSN_21),
+		      FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_21),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_24 >> 2),
+		      CDM1_CRSN_QSEL_REASON_MASK(CRSN_24),
+		      FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_24),
+				 CDM_CRSN_QSEL_Q6));
+	airoha_fe_rmw(eth, REG_CDM1_CRSN_QSEL(CRSN_25 >> 2),
+		      CDM1_CRSN_QSEL_REASON_MASK(CRSN_25),
+		      FIELD_PREP(CDM1_CRSN_QSEL_REASON_MASK(CRSN_25),
+				 CDM_CRSN_QSEL_Q1));
+	/* CDM2_CRSN_QSEL */
+	airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_08 >> 2),
+		      CDM2_CRSN_QSEL_REASON_MASK(CRSN_08),
+		      FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_08),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_21 >> 2),
+		      CDM2_CRSN_QSEL_REASON_MASK(CRSN_21),
+		      FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_21),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_22 >> 2),
+		      CDM2_CRSN_QSEL_REASON_MASK(CRSN_22),
+		      FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_22),
+				 CDM_CRSN_QSEL_Q1));
+	airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_24 >> 2),
+		      CDM2_CRSN_QSEL_REASON_MASK(CRSN_24),
+		      FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_24),
+				 CDM_CRSN_QSEL_Q6));
+	airoha_fe_rmw(eth, REG_CDM2_CRSN_QSEL(CRSN_25 >> 2),
+		      CDM2_CRSN_QSEL_REASON_MASK(CRSN_25),
+		      FIELD_PREP(CDM2_CRSN_QSEL_REASON_MASK(CRSN_25),
+				 CDM_CRSN_QSEL_Q1));
+}
+
+static int airoha_fe_init(struct airoha_eth *eth)
+{
+	airoha_fe_maccr_init(eth);
+
+	/* PSE IQ reserve */
+	airoha_fe_rmw(eth, REG_PSE_IQ_REV1, PSE_IQ_RES1_P2_MASK,
+		      FIELD_PREP(PSE_IQ_RES1_P2_MASK, 0x10));
+	airoha_fe_rmw(eth, REG_PSE_IQ_REV2,
+		      PSE_IQ_RES2_P5_MASK | PSE_IQ_RES2_P4_MASK,
+		      FIELD_PREP(PSE_IQ_RES2_P5_MASK, 0x40) |
+		      FIELD_PREP(PSE_IQ_RES2_P4_MASK, 0x34));
+
+	/* enable FE copy engine for MC/KA/DPI */
+	airoha_fe_wr(eth, REG_FE_PCE_CFG,
+		     PCE_DPI_EN_MASK | PCE_KA_EN_MASK | PCE_MC_EN_MASK);
+	/* set vip queue selection to ring 1 */
+	airoha_fe_rmw(eth, REG_CDM1_FWD_CFG, CDM1_VIP_QSEL_MASK,
+		      FIELD_PREP(CDM1_VIP_QSEL_MASK, 0x4));
+	airoha_fe_rmw(eth, REG_CDM2_FWD_CFG, CDM2_VIP_QSEL_MASK,
+		      FIELD_PREP(CDM2_VIP_QSEL_MASK, 0x4));
+	/* set GDM4 source interface offset to 8 */
+	airoha_fe_rmw(eth, REG_GDM4_SRC_PORT_SET,
+		      GDM4_SPORT_OFF2_MASK |
+		      GDM4_SPORT_OFF1_MASK |
+		      GDM4_SPORT_OFF0_MASK,
+		      FIELD_PREP(GDM4_SPORT_OFF2_MASK, 8) |
+		      FIELD_PREP(GDM4_SPORT_OFF1_MASK, 8) |
+		      FIELD_PREP(GDM4_SPORT_OFF0_MASK, 8));
+
+	/* set PSE Page as 128B */
+	airoha_fe_rmw(eth, REG_FE_DMA_GLO_CFG,
+		      FE_DMA_GLO_L2_SPACE_MASK | FE_DMA_GLO_PG_SZ_MASK,
+		      FIELD_PREP(FE_DMA_GLO_L2_SPACE_MASK, 2) |
+		      FE_DMA_GLO_PG_SZ_MASK);
+	airoha_fe_wr(eth, REG_FE_RST_GLO_CFG,
+		     FE_RST_CORE_MASK | FE_RST_GDM3_MBI_ARB_MASK |
+		     FE_RST_GDM4_MBI_ARB_MASK);
+	usleep_range(1000, 2000);
+
+	/* connect RxRing1 and RxRing15 to PSE Port0 OQ-1
+	 * connect other rings to PSE Port0 OQ-0
+	 */
+	airoha_fe_wr(eth, REG_FE_CDM1_OQ_MAP0, BIT(4));
+	airoha_fe_wr(eth, REG_FE_CDM1_OQ_MAP1, BIT(28));
+	airoha_fe_wr(eth, REG_FE_CDM1_OQ_MAP2, BIT(4));
+	airoha_fe_wr(eth, REG_FE_CDM1_OQ_MAP3, BIT(28));
+
+	airoha_fe_vip_setup(eth);
+	airoha_fe_pse_ports_init(eth);
+
+	airoha_fe_set(eth, REG_GDM_MISC_CFG,
+		      GDM2_RDM_ACK_WAIT_PREF_MASK |
+		      GDM2_CHN_VLD_MODE_MASK);
+	airoha_fe_rmw(eth, REG_CDM2_FWD_CFG, CDM2_OAM_QSEL_MASK, 15);
+
+	/* init fragment and assemble Force Port */
+	/* NPU Core-3, NPU Bridge Channel-3 */
+	airoha_fe_rmw(eth, REG_IP_FRAG_FP,
+		      IP_FRAGMENT_PORT_MASK | IP_FRAGMENT_NBQ_MASK,
+		      FIELD_PREP(IP_FRAGMENT_PORT_MASK, 6) |
+		      FIELD_PREP(IP_FRAGMENT_NBQ_MASK, 3));
+	/* QDMA LAN, RX Ring-22 */
+	airoha_fe_rmw(eth, REG_IP_FRAG_FP,
+		      IP_ASSEMBLE_PORT_MASK | IP_ASSEMBLE_NBQ_MASK,
+		      FIELD_PREP(IP_ASSEMBLE_PORT_MASK, 0) |
+		      FIELD_PREP(IP_ASSEMBLE_NBQ_MASK, 22));
+
+	airoha_fe_set(eth, REG_GDM3_FWD_CFG, GDM3_PAD_EN_MASK);
+	airoha_fe_set(eth, REG_GDM4_FWD_CFG, GDM4_PAD_EN_MASK);
+
+	airoha_fe_crsn_qsel_init(eth);
+
+	airoha_fe_clear(eth, REG_FE_CPORT_CFG, FE_CPORT_QUEUE_XFC_MASK);
+	airoha_fe_set(eth, REG_FE_CPORT_CFG, FE_CPORT_PORT_XFC_MASK);
+
+	/* default aging mode for mbi unlock issue */
+	airoha_fe_rmw(eth, REG_GDM2_CHN_RLS,
+		      MBI_RX_AGE_SEL_MASK | MBI_TX_AGE_SEL_MASK,
+		      FIELD_PREP(MBI_RX_AGE_SEL_MASK, 3) |
+		      FIELD_PREP(MBI_TX_AGE_SEL_MASK, 3));
+
+	/* disable IFC by default */
+	airoha_fe_clear(eth, REG_FE_CSR_IFC_CFG, FE_IFC_EN_MASK);
+
+	/* enable 1:N vlan action, init vlan table */
+	airoha_fe_set(eth, REG_MC_VLAN_EN, MC_VLAN_EN_MASK);
+
+	return airoha_fe_mc_vlan_clear(eth);
+}
+
+static int airoha_qdma_fill_rx_queue(struct airoha_queue *q)
+{
+	enum dma_data_direction dir = page_pool_get_dma_dir(q->page_pool);
+	struct airoha_eth *eth = q->eth;
+	int qid = q - &eth->q_rx[0];
+	int nframes = 0;
+
+	while (q->queued < q->ndesc - 1) {
+		struct airoha_queue_entry *e = &q->entry[q->head];
+		struct airoha_qdma_desc *desc = &q->desc[q->head];
+		struct page *page;
+		int offset;
+		u32 val;
+
+		page = page_pool_dev_alloc_frag(q->page_pool, &offset,
+						q->buf_size);
+		if (!page)
+			break;
+
+		q->head = (q->head + 1) % q->ndesc;
+		q->queued++;
+		nframes++;
+
+		e->buf = page_address(page) + offset;
+		e->dma_addr = page_pool_get_dma_addr(page) + offset;
+		e->dma_len = SKB_WITH_OVERHEAD(q->buf_size);
+
+		dma_sync_single_for_device(eth->dev, e->dma_addr, e->dma_len,
+					   dir);
+
+		val = FIELD_PREP(QDMA_DESC_LEN_MASK, e->dma_len);
+		WRITE_ONCE(desc->ctrl, cpu_to_le32(val));
+		WRITE_ONCE(desc->addr, cpu_to_le32(e->dma_addr));
+		val = FIELD_PREP(QDMA_DESC_NEXT_ID_MASK, q->head);
+		WRITE_ONCE(desc->data, cpu_to_le32(val));
+		WRITE_ONCE(desc->msg0, 0);
+		WRITE_ONCE(desc->msg1, 0);
+		WRITE_ONCE(desc->msg2, 0);
+		WRITE_ONCE(desc->msg3, 0);
+
+		airoha_qdma_rmw(eth, REG_RX_CPU_IDX(qid), RX_RING_CPU_IDX_MASK,
+				FIELD_PREP(RX_RING_CPU_IDX_MASK, q->head));
+	}
+
+	return nframes;
+}
+
+static int airoha_qdma_get_gdm_port(struct airoha_eth *eth,
+				    struct airoha_qdma_desc *desc)
+{
+	u32 port, sport, msg1 = le32_to_cpu(desc->msg1);
+
+	sport = FIELD_GET(QDMA_ETH_RXMSG_SPORT_MASK, msg1);
+	switch (sport) {
+	case 0x10 ... 0x13:
+		port = 0;
+		break;
+	case 0x2 ... 0x4:
+		port = sport - 1;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return port >= ARRAY_SIZE(eth->ports) ? -EINVAL : port;
+}
+
+static int airoha_qdma_rx_process(struct airoha_queue *q, int budget)
+{
+	enum dma_data_direction dir = page_pool_get_dma_dir(q->page_pool);
+	struct airoha_eth *eth = q->eth;
+	int qid = q - &eth->q_rx[0];
+	int done = 0;
+
+	while (done < budget) {
+		struct airoha_queue_entry *e = &q->entry[q->tail];
+		struct airoha_qdma_desc *desc = &q->desc[q->tail];
+		dma_addr_t dma_addr = le32_to_cpu(desc->addr);
+		u32 desc_ctrl = le32_to_cpu(desc->ctrl);
+		struct sk_buff *skb;
+		int len, p;
+
+		if (!(desc_ctrl & QDMA_DESC_DONE_MASK))
+			break;
+
+		if (!dma_addr)
+			break;
+
+		len = FIELD_GET(QDMA_DESC_LEN_MASK, desc_ctrl);
+		if (!len)
+			break;
+
+		q->tail = (q->tail + 1) % q->ndesc;
+		q->queued--;
+
+		dma_sync_single_for_cpu(eth->dev, dma_addr,
+					SKB_WITH_OVERHEAD(q->buf_size), dir);
+
+		p = airoha_qdma_get_gdm_port(eth, desc);
+		if (p < 0 || !eth->ports[p]) {
+			page_pool_put_full_page(q->page_pool,
+						virt_to_head_page(e->buf),
+						true);
+			continue;
+		}
+
+		skb = napi_build_skb(e->buf, q->buf_size);
+		if (!skb) {
+			page_pool_put_full_page(q->page_pool,
+						virt_to_head_page(e->buf),
+						true);
+			break;
+		}
+
+		skb_reserve(skb, 2);
+		__skb_put(skb, len);
+		skb_mark_for_recycle(skb);
+		skb->dev = eth->ports[p]->dev;
+		skb->protocol = eth_type_trans(skb, skb->dev);
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+		skb_record_rx_queue(skb, qid);
+		napi_gro_receive(&q->napi, skb);
+
+		done++;
+	}
+	airoha_qdma_fill_rx_queue(q);
+
+	return done;
+}
+
+static int airoha_qdma_rx_napi_poll(struct napi_struct *napi, int budget)
+{
+	struct airoha_queue *q = container_of(napi, struct airoha_queue, napi);
+	struct airoha_eth *eth = q->eth;
+	int cur, done = 0;
+
+	do {
+		cur = airoha_qdma_rx_process(q, budget - done);
+		done += cur;
+	} while (cur && done < budget);
+
+	if (done < budget && napi_complete(napi))
+		airoha_qdma_irq_enable(eth, QDMA_INT_REG_IDX1,
+				       RX_DONE_INT_MASK);
+
+	return done;
+}
+
+static int airoha_qdma_init_rx_queue(struct airoha_eth *eth,
+				     struct airoha_queue *q, int ndesc)
+{
+	const struct page_pool_params pp_params = {
+		.order = 0,
+		.pool_size = 256,
+		.flags = PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV,
+		.dma_dir = DMA_FROM_DEVICE,
+		.max_len = PAGE_SIZE,
+		.nid = NUMA_NO_NODE,
+		.dev = eth->dev,
+		.napi = &q->napi,
+	};
+	int qid = q - &eth->q_rx[0], thr;
+	dma_addr_t dma_addr;
+
+	q->buf_size = PAGE_SIZE / 2;
+	q->ndesc = ndesc;
+	q->eth = eth;
+
+	q->entry = devm_kzalloc(eth->dev, q->ndesc * sizeof(*q->entry),
+				GFP_KERNEL);
+	if (!q->entry)
+		return -ENOMEM;
+
+	q->page_pool = page_pool_create(&pp_params);
+	if (IS_ERR(q->page_pool)) {
+		int err = PTR_ERR(q->page_pool);
+
+		q->page_pool = NULL;
+		return err;
+	}
+
+	q->desc = dmam_alloc_coherent(eth->dev, q->ndesc * sizeof(*q->desc),
+				      &dma_addr, GFP_KERNEL);
+	if (!q->desc)
+		return -ENOMEM;
+
+	netif_napi_add(eth->napi_dev, &q->napi, airoha_qdma_rx_napi_poll);
+
+	airoha_qdma_wr(eth, REG_RX_RING_BASE(qid), dma_addr);
+	airoha_qdma_rmw(eth, REG_RX_RING_SIZE(qid), RX_RING_SIZE_MASK,
+			FIELD_PREP(RX_RING_SIZE_MASK, ndesc));
+
+	thr = clamp(ndesc >> 3, 1, 32);
+	airoha_qdma_rmw(eth, REG_RX_RING_SIZE(qid), RX_RING_THR_MASK,
+			FIELD_PREP(RX_RING_THR_MASK, thr));
+	airoha_qdma_rmw(eth, REG_RX_DMA_IDX(qid), RX_RING_DMA_IDX_MASK,
+			FIELD_PREP(RX_RING_DMA_IDX_MASK, q->head));
+
+	airoha_qdma_fill_rx_queue(q);
+
+	return 0;
+}
+
+static void airoha_qdma_cleanup_rx_queue(struct airoha_queue *q)
+{
+	enum dma_data_direction dir = page_pool_get_dma_dir(q->page_pool);
+	struct airoha_eth *eth = q->eth;
+
+	while (q->queued) {
+		struct airoha_queue_entry *e = &q->entry[q->tail];
+		struct page *page = virt_to_head_page(e->buf);
+
+		dma_sync_single_for_cpu(eth->dev, e->dma_addr, e->dma_len,
+					dir);
+		page_pool_put_full_page(q->page_pool, page, false);
+		q->tail = (q->tail + 1) % q->ndesc;
+		q->queued--;
+	}
+}
+
+static int airoha_qdma_init_rx(struct airoha_eth *eth)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(eth->q_rx); i++) {
+		int err;
+
+		if (!(RX_DONE_INT_MASK & BIT(i))) {
+			/* rx-queue not binded to irq */
+			continue;
+		}
+
+		err = airoha_qdma_init_rx_queue(eth, &eth->q_rx[i],
+						RX_DSCP_NUM(i));
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int airoha_qdma_tx_napi_poll(struct napi_struct *napi, int budget)
+{
+	struct airoha_tx_irq_queue *irq_q;
+	struct airoha_eth *eth;
+	int id, done = 0;
+
+	irq_q = container_of(napi, struct airoha_tx_irq_queue, napi);
+	eth = irq_q->eth;
+	id = irq_q - &eth->q_tx_irq[0];
+
+	while (irq_q->queued > 0 && done < budget) {
+		u32 qid, last, val = irq_q->q[irq_q->head];
+		struct airoha_queue *q;
+
+		if (val == 0xff)
+			break;
+
+		irq_q->q[irq_q->head] = 0xff; /* mark as done */
+		irq_q->head = (irq_q->head + 1) % irq_q->size;
+		irq_q->queued--;
+		done++;
+
+		last = FIELD_GET(IRQ_DESC_IDX_MASK, val);
+		qid = FIELD_GET(IRQ_RING_IDX_MASK, val);
+
+		if (qid >= ARRAY_SIZE(eth->q_tx))
+			continue;
+
+		q = &eth->q_tx[qid];
+		if (!q->ndesc)
+			continue;
+
+		spin_lock_bh(&q->lock);
+
+		while (q->queued > 0) {
+			struct airoha_qdma_desc *desc = &q->desc[q->tail];
+			struct airoha_queue_entry *e = &q->entry[q->tail];
+			u32 desc_ctrl = le32_to_cpu(desc->ctrl);
+			struct sk_buff *skb = e->skb;
+			u16 index = q->tail;
+
+			if (!(desc_ctrl & QDMA_DESC_DONE_MASK) &&
+			    !(desc_ctrl & QDMA_DESC_DROP_MASK))
+				break;
+
+			q->tail = (q->tail + 1) % q->ndesc;
+			q->queued--;
+
+			dma_unmap_single(eth->dev, e->dma_addr, e->dma_len,
+					 DMA_TO_DEVICE);
+
+			WRITE_ONCE(desc->msg0, 0);
+			WRITE_ONCE(desc->msg1, 0);
+
+			if (skb) {
+				struct netdev_queue *txq;
+
+				txq = netdev_get_tx_queue(skb->dev, qid);
+				if (netif_tx_queue_stopped(txq) &&
+				    q->ndesc - q->queued >= q->free_thr)
+					netif_tx_wake_queue(txq);
+
+				dev_kfree_skb_any(skb);
+				e->skb = NULL;
+			}
+
+			if (index == last)
+				break;
+		}
+
+		spin_unlock_bh(&q->lock);
+	}
+
+	if (done) {
+		int i, len = done >> 7;
+
+		for (i = 0; i < len; i++)
+			airoha_qdma_rmw(eth, REG_IRQ_CLEAR_LEN(id),
+					IRQ_CLEAR_LEN_MASK, 0x80);
+		airoha_qdma_rmw(eth, REG_IRQ_CLEAR_LEN(id),
+				IRQ_CLEAR_LEN_MASK, (done & 0x7f));
+	}
+
+	if (done < budget && napi_complete(napi))
+		airoha_qdma_irq_enable(eth, QDMA_INT_REG_IDX0,
+				       TX_DONE_INT_MASK(id));
+
+	return done;
+}
+
+static int airoha_qdma_init_tx_queue(struct airoha_eth *eth,
+				     struct airoha_queue *q, int size)
+{
+	int i, qid = q - &eth->q_tx[0];
+	dma_addr_t dma_addr;
+
+	spin_lock_init(&q->lock);
+	q->ndesc = size;
+	q->eth = eth;
+	q->free_thr = 1 + MAX_SKB_FRAGS;
+
+	q->entry = devm_kzalloc(eth->dev, q->ndesc * sizeof(*q->entry),
+				GFP_KERNEL);
+	if (!q->entry)
+		return -ENOMEM;
+
+	q->desc = dmam_alloc_coherent(eth->dev, q->ndesc * sizeof(*q->desc),
+				      &dma_addr, GFP_KERNEL);
+	if (!q->desc)
+		return -ENOMEM;
+
+	for (i = 0; i < q->ndesc; i++) {
+		u32 val;
+
+		val = FIELD_PREP(QDMA_DESC_DONE_MASK, 1);
+		WRITE_ONCE(q->desc[i].ctrl, cpu_to_le32(val));
+	}
+
+	airoha_qdma_wr(eth, REG_TX_RING_BASE(qid), dma_addr);
+	airoha_qdma_rmw(eth, REG_TX_CPU_IDX(qid), TX_RING_CPU_IDX_MASK,
+			FIELD_PREP(TX_RING_CPU_IDX_MASK, q->head));
+	airoha_qdma_rmw(eth, REG_TX_DMA_IDX(qid), TX_RING_DMA_IDX_MASK,
+			FIELD_PREP(TX_RING_DMA_IDX_MASK, q->head));
+
+	return 0;
+}
+
+static int airoha_qdma_tx_irq_init(struct airoha_eth *eth,
+				   struct airoha_tx_irq_queue *irq_q,
+				   int size)
+{
+	int id = irq_q - &eth->q_tx_irq[0];
+	dma_addr_t dma_addr;
+
+	netif_napi_add_tx(eth->napi_dev, &irq_q->napi,
+			  airoha_qdma_tx_napi_poll);
+	irq_q->q = dmam_alloc_coherent(eth->dev, size * sizeof(u32),
+				       &dma_addr, GFP_KERNEL);
+	if (!irq_q->q)
+		return -ENOMEM;
+
+	memset(irq_q->q, 0xff, size * sizeof(u32));
+	irq_q->size = size;
+	irq_q->eth = eth;
+
+	airoha_qdma_wr(eth, REG_TX_IRQ_BASE(id), dma_addr);
+	airoha_qdma_rmw(eth, REG_TX_IRQ_CFG(id), TX_IRQ_DEPTH_MASK,
+			FIELD_PREP(TX_IRQ_DEPTH_MASK, size));
+	airoha_qdma_rmw(eth, REG_TX_IRQ_CFG(id), TX_IRQ_THR_MASK,
+			FIELD_PREP(TX_IRQ_THR_MASK, 1));
+
+	return 0;
+}
+
+static int airoha_qdma_init_tx(struct airoha_eth *eth)
+{
+	int i, err;
+
+	for (i = 0; i < ARRAY_SIZE(eth->q_tx_irq); i++) {
+		err = airoha_qdma_tx_irq_init(eth, &eth->q_tx_irq[i],
+					      IRQ_QUEUE_LEN(i));
+		if (err)
+			return err;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(eth->q_tx); i++) {
+		err = airoha_qdma_init_tx_queue(eth, &eth->q_tx[i],
+						TX_DSCP_NUM);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static void airoha_qdma_cleanup_tx_queue(struct airoha_queue *q)
+{
+	struct airoha_eth *eth = q->eth;
+
+	spin_lock_bh(&q->lock);
+	while (q->queued) {
+		struct airoha_queue_entry *e = &q->entry[q->tail];
+
+		dma_unmap_single(eth->dev, e->dma_addr, e->dma_len,
+				 DMA_TO_DEVICE);
+		dev_kfree_skb_any(e->skb);
+		e->skb = NULL;
+
+		q->tail = (q->tail + 1) % q->ndesc;
+		q->queued--;
+	}
+	spin_unlock_bh(&q->lock);
+}
+
+static int airoha_qdma_init_hfwd_queues(struct airoha_eth *eth)
+{
+	dma_addr_t dma_addr;
+	u32 status;
+	int size;
+
+	size = HW_DSCP_NUM * sizeof(struct airoha_qdma_fwd_desc);
+	eth->hfwd.desc = dmam_alloc_coherent(eth->dev, size, &dma_addr,
+					     GFP_KERNEL);
+	if (!eth->hfwd.desc)
+		return -ENOMEM;
+
+	airoha_qdma_wr(eth, REG_FWD_DSCP_BASE, dma_addr);
+
+	size = AIROHA_MAX_PACKET_SIZE * HW_DSCP_NUM;
+	eth->hfwd.q = dmam_alloc_coherent(eth->dev, size, &dma_addr,
+					  GFP_KERNEL);
+	if (!eth->hfwd.q)
+		return -ENOMEM;
+
+	airoha_qdma_wr(eth, REG_FWD_BUF_BASE, dma_addr);
+
+	airoha_qdma_rmw(eth, REG_HW_FWD_DSCP_CFG,
+			HW_FWD_DSCP_PAYLOAD_SIZE_MASK,
+			FIELD_PREP(HW_FWD_DSCP_PAYLOAD_SIZE_MASK, 0));
+	airoha_qdma_rmw(eth, REG_FWD_DSCP_LOW_THR, FWD_DSCP_LOW_THR_MASK,
+			FIELD_PREP(FWD_DSCP_LOW_THR_MASK, 128));
+	airoha_qdma_rmw(eth, REG_LMGR_INIT_CFG,
+			LMGR_INIT_START | LMGR_SRAM_MODE_MASK |
+			HW_FWD_DESC_NUM_MASK,
+			FIELD_PREP(HW_FWD_DESC_NUM_MASK, HW_DSCP_NUM) |
+			LMGR_INIT_START);
+
+	return read_poll_timeout(airoha_qdma_rr, status,
+				 !(status & LMGR_INIT_START), USEC_PER_MSEC,
+				 30 * USEC_PER_MSEC, true, eth,
+				 REG_LMGR_INIT_CFG);
+}
+
+static void airoha_qdma_init_qos(struct airoha_eth *eth)
+{
+	airoha_qdma_clear(eth, REG_TXWRR_MODE_CFG, TWRR_WEIGHT_SCALE_MASK);
+	airoha_qdma_set(eth, REG_TXWRR_MODE_CFG, TWRR_WEIGHT_BASE_MASK);
+
+	airoha_qdma_clear(eth, REG_PSE_BUF_USAGE_CFG,
+			  PSE_BUF_ESTIMATE_EN_MASK);
+
+	airoha_qdma_set(eth, REG_EGRESS_RATE_METER_CFG,
+			EGRESS_RATE_METER_EN_MASK |
+			EGRESS_RATE_METER_EQ_RATE_EN_MASK);
+	/* 2047us x 31 = 63.457ms */
+	airoha_qdma_rmw(eth, REG_EGRESS_RATE_METER_CFG,
+			EGRESS_RATE_METER_WINDOW_SZ_MASK,
+			FIELD_PREP(EGRESS_RATE_METER_WINDOW_SZ_MASK, 0x1f));
+	airoha_qdma_rmw(eth, REG_EGRESS_RATE_METER_CFG,
+			EGRESS_RATE_METER_TIMESLICE_MASK,
+			FIELD_PREP(EGRESS_RATE_METER_TIMESLICE_MASK, 0x7ff));
+
+	/* ratelimit init */
+	airoha_qdma_set(eth, REG_GLB_TRTCM_CFG, GLB_TRTCM_EN_MASK);
+	/* fast-tick 25us */
+	airoha_qdma_rmw(eth, REG_GLB_TRTCM_CFG, GLB_FAST_TICK_MASK,
+			FIELD_PREP(GLB_FAST_TICK_MASK, 25));
+	airoha_qdma_rmw(eth, REG_GLB_TRTCM_CFG, GLB_SLOW_TICK_RATIO_MASK,
+			FIELD_PREP(GLB_SLOW_TICK_RATIO_MASK, 40));
+
+	airoha_qdma_set(eth, REG_EGRESS_TRTCM_CFG, EGRESS_TRTCM_EN_MASK);
+	airoha_qdma_rmw(eth, REG_EGRESS_TRTCM_CFG, EGRESS_FAST_TICK_MASK,
+			FIELD_PREP(EGRESS_FAST_TICK_MASK, 25));
+	airoha_qdma_rmw(eth, REG_EGRESS_TRTCM_CFG,
+			EGRESS_SLOW_TICK_RATIO_MASK,
+			FIELD_PREP(EGRESS_SLOW_TICK_RATIO_MASK, 40));
+
+	airoha_qdma_set(eth, REG_INGRESS_TRTCM_CFG, INGRESS_TRTCM_EN_MASK);
+	airoha_qdma_clear(eth, REG_INGRESS_TRTCM_CFG,
+			  INGRESS_TRTCM_MODE_MASK);
+	airoha_qdma_rmw(eth, REG_INGRESS_TRTCM_CFG, INGRESS_FAST_TICK_MASK,
+			FIELD_PREP(INGRESS_FAST_TICK_MASK, 125));
+	airoha_qdma_rmw(eth, REG_INGRESS_TRTCM_CFG,
+			INGRESS_SLOW_TICK_RATIO_MASK,
+			FIELD_PREP(INGRESS_SLOW_TICK_RATIO_MASK, 8));
+
+	airoha_qdma_set(eth, REG_SLA_TRTCM_CFG, SLA_TRTCM_EN_MASK);
+	airoha_qdma_rmw(eth, REG_SLA_TRTCM_CFG, SLA_FAST_TICK_MASK,
+			FIELD_PREP(SLA_FAST_TICK_MASK, 25));
+	airoha_qdma_rmw(eth, REG_SLA_TRTCM_CFG, SLA_SLOW_TICK_RATIO_MASK,
+			FIELD_PREP(SLA_SLOW_TICK_RATIO_MASK, 40));
+}
+
+static int airoha_qdma_hw_init(struct airoha_eth *eth)
+{
+	int i;
+
+	/* clear pending irqs */
+	for (i = 0; i < ARRAY_SIZE(eth->irqmask); i++)
+		airoha_qdma_wr(eth, REG_INT_STATUS(i), 0xffffffff);
+
+	/* setup irqs */
+	airoha_qdma_irq_enable(eth, QDMA_INT_REG_IDX0, INT_IDX0_MASK);
+	airoha_qdma_irq_enable(eth, QDMA_INT_REG_IDX1, INT_IDX1_MASK);
+	airoha_qdma_irq_enable(eth, QDMA_INT_REG_IDX4, INT_IDX4_MASK);
+
+	/* setup irq binding */
+	for (i = 0; i < ARRAY_SIZE(eth->q_tx); i++) {
+		if (!eth->q_tx[i].ndesc)
+			continue;
+
+		if (TX_RING_IRQ_BLOCKING_MAP_MASK & BIT(i))
+			airoha_qdma_set(eth, REG_TX_RING_BLOCKING(i),
+					TX_RING_IRQ_BLOCKING_CFG_MASK);
+		else
+			airoha_qdma_clear(eth, REG_TX_RING_BLOCKING(i),
+					  TX_RING_IRQ_BLOCKING_CFG_MASK);
+	}
+
+	airoha_qdma_wr(eth, REG_QDMA_GLOBAL_CFG,
+		       GLOBAL_CFG_RX_2B_OFFSET_MASK |
+		       FIELD_PREP(GLOBAL_CFG_DMA_PREFERENCE_MASK, 3) |
+		       GLOBAL_CFG_CPU_TXR_RR_MASK |
+		       GLOBAL_CFG_PAYLOAD_BYTE_SWAP_MASK |
+		       GLOBAL_CFG_MULTICAST_MODIFY_FP_MASK |
+		       GLOBAL_CFG_MULTICAST_EN_MASK |
+		       GLOBAL_CFG_IRQ0_EN_MASK | GLOBAL_CFG_IRQ1_EN_MASK |
+		       GLOBAL_CFG_TX_WB_DONE_MASK |
+		       FIELD_PREP(GLOBAL_CFG_MAX_ISSUE_NUM_MASK, 2));
+
+	airoha_qdma_init_qos(eth);
+
+	/* disable qdma rx delay interrupt */
+	for (i = 0; i < ARRAY_SIZE(eth->q_rx); i++) {
+		if (!eth->q_rx[i].ndesc)
+			continue;
+
+		airoha_qdma_clear(eth, REG_RX_DELAY_INT_IDX(i),
+				  RX_DELAY_INT_MASK);
+	}
+
+	airoha_qdma_set(eth, REG_TXQ_CNGST_CFG,
+			TXQ_CNGST_DROP_EN | TXQ_CNGST_DEI_DROP_EN);
+
+	return 0;
+}
+
+static irqreturn_t airoha_irq_handler(int irq, void *dev_instance)
+{
+	struct airoha_eth *eth = dev_instance;
+	u32 intr[ARRAY_SIZE(eth->irqmask)];
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(eth->irqmask); i++) {
+		intr[i] = airoha_qdma_rr(eth, REG_INT_STATUS(i));
+		intr[i] &= eth->irqmask[i];
+		airoha_qdma_wr(eth, REG_INT_STATUS(i), intr[i]);
+	}
+
+	if (!test_bit(DEV_STATE_INITIALIZED, &eth->state))
+		return IRQ_NONE;
+
+	if (intr[1] & RX_DONE_INT_MASK) {
+		airoha_qdma_irq_disable(eth, QDMA_INT_REG_IDX1,
+					RX_DONE_INT_MASK);
+
+		for (i = 0; i < ARRAY_SIZE(eth->q_rx); i++) {
+			if (!eth->q_rx[i].ndesc)
+				continue;
+
+			if (intr[1] & BIT(i))
+				napi_schedule(&eth->q_rx[i].napi);
+		}
+	}
+
+	if (intr[0] & INT_TX_MASK) {
+		for (i = 0; i < ARRAY_SIZE(eth->q_tx_irq); i++) {
+			struct airoha_tx_irq_queue *irq_q = &eth->q_tx_irq[i];
+			u32 status, head;
+
+			if (!(intr[0] & TX_DONE_INT_MASK(i)))
+				continue;
+
+			airoha_qdma_irq_disable(eth, QDMA_INT_REG_IDX0,
+						TX_DONE_INT_MASK(i));
+
+			status = airoha_qdma_rr(eth, REG_IRQ_STATUS(i));
+			head = FIELD_GET(IRQ_HEAD_IDX_MASK, status);
+			irq_q->head = head % irq_q->size;
+			irq_q->queued = FIELD_GET(IRQ_ENTRY_LEN_MASK, status);
+
+			napi_schedule(&eth->q_tx_irq[i].napi);
+		}
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int airoha_qdma_init(struct airoha_eth *eth)
+{
+	int err;
+
+	err = devm_request_irq(eth->dev, eth->irq, airoha_irq_handler,
+			       IRQF_SHARED, KBUILD_MODNAME, eth);
+	if (err)
+		return err;
+
+	err = airoha_qdma_init_rx(eth);
+	if (err)
+		return err;
+
+	err = airoha_qdma_init_tx(eth);
+	if (err)
+		return err;
+
+	err = airoha_qdma_init_hfwd_queues(eth);
+	if (err)
+		return err;
+
+	err = airoha_qdma_hw_init(eth);
+	if (err)
+		return err;
+
+	set_bit(DEV_STATE_INITIALIZED, &eth->state);
+
+	return 0;
+}
+
+static int airoha_hw_init(struct airoha_eth *eth)
+{
+	int err;
+
+	/* disable xsi */
+	reset_control_bulk_assert(ARRAY_SIZE(eth->xsi_rsts), eth->xsi_rsts);
+
+	reset_control_bulk_assert(ARRAY_SIZE(eth->rsts), eth->rsts);
+	msleep(20);
+	reset_control_bulk_deassert(ARRAY_SIZE(eth->rsts), eth->rsts);
+	msleep(20);
+
+	err = airoha_fe_init(eth);
+	if (err)
+		return err;
+
+	return airoha_qdma_init(eth);
+}
+
+static void airoha_hw_cleanup(struct airoha_eth *eth)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(eth->q_rx); i++) {
+		if (!eth->q_rx[i].ndesc)
+			continue;
+
+		napi_disable(&eth->q_rx[i].napi);
+		netif_napi_del(&eth->q_rx[i].napi);
+		airoha_qdma_cleanup_rx_queue(&eth->q_rx[i]);
+		if (eth->q_rx[i].page_pool)
+			page_pool_destroy(eth->q_rx[i].page_pool);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(eth->q_tx_irq); i++) {
+		napi_disable(&eth->q_tx_irq[i].napi);
+		netif_napi_del(&eth->q_tx_irq[i].napi);
+	}
+
+	for (i = 0; i < ARRAY_SIZE(eth->q_tx); i++) {
+		if (!eth->q_tx[i].ndesc)
+			continue;
+
+		airoha_qdma_cleanup_tx_queue(&eth->q_tx[i]);
+	}
+}
+
+static void airoha_qdma_start_napi(struct airoha_eth *eth)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(eth->q_tx_irq); i++)
+		napi_enable(&eth->q_tx_irq[i].napi);
+
+	for (i = 0; i < ARRAY_SIZE(eth->q_rx); i++) {
+		if (!eth->q_rx[i].ndesc)
+			continue;
+
+		napi_enable(&eth->q_rx[i].napi);
+	}
+}
+
+static void airoha_update_hw_stats(struct airoha_gdm_port *port)
+{
+	struct airoha_eth *eth = port->eth;
+	u32 val, i = 0;
+
+	spin_lock(&port->stats.lock);
+	u64_stats_update_begin(&port->stats.syncp);
+
+	/* TX */
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_OK_PKT_CNT_H(port->id));
+	port->stats.tx_ok_pkts += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_OK_PKT_CNT_L(port->id));
+	port->stats.tx_ok_pkts += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_OK_BYTE_CNT_H(port->id));
+	port->stats.tx_ok_bytes += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_OK_BYTE_CNT_L(port->id));
+	port->stats.tx_ok_bytes += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_DROP_CNT(port->id));
+	port->stats.tx_drops += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_BC_CNT(port->id));
+	port->stats.tx_broadcast += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_MC_CNT(port->id));
+	port->stats.tx_multicast += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_RUNT_CNT(port->id));
+	port->stats.tx_len[i] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_E64_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_E64_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L64_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L64_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L127_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L127_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L255_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L255_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L511_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L511_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L1023_CNT_H(port->id));
+	port->stats.tx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_L1023_CNT_L(port->id));
+	port->stats.tx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_TX_ETH_LONG_CNT(port->id));
+	port->stats.tx_len[i++] += val;
+
+	/* RX */
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_OK_PKT_CNT_H(port->id));
+	port->stats.rx_ok_pkts += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_OK_PKT_CNT_L(port->id));
+	port->stats.rx_ok_pkts += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_OK_BYTE_CNT_H(port->id));
+	port->stats.rx_ok_bytes += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_OK_BYTE_CNT_L(port->id));
+	port->stats.rx_ok_bytes += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_DROP_CNT(port->id));
+	port->stats.rx_drops += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_BC_CNT(port->id));
+	port->stats.rx_broadcast += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_MC_CNT(port->id));
+	port->stats.rx_multicast += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ERROR_DROP_CNT(port->id));
+	port->stats.rx_errors += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_CRC_ERR_CNT(port->id));
+	port->stats.rx_crc_error += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_OVERFLOW_DROP_CNT(port->id));
+	port->stats.rx_over_errors += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_FRAG_CNT(port->id));
+	port->stats.rx_fragment += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_JABBER_CNT(port->id));
+	port->stats.rx_jabber += val;
+
+	i = 0;
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_RUNT_CNT(port->id));
+	port->stats.rx_len[i] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_E64_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_E64_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L64_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L64_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L127_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L127_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L255_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L255_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L511_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L511_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L1023_CNT_H(port->id));
+	port->stats.rx_len[i] += ((u64)val << 32);
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_L1023_CNT_L(port->id));
+	port->stats.rx_len[i++] += val;
+
+	val = airoha_fe_rr(eth, REG_FE_GDM_RX_ETH_LONG_CNT(port->id));
+	port->stats.rx_len[i++] += val;
+
+	/* reset mib counters */
+	airoha_fe_set(eth, REG_FE_GDM_MIB_CLEAR(port->id),
+		      FE_GDM_MIB_RX_CLEAR_MASK | FE_GDM_MIB_TX_CLEAR_MASK);
+
+	u64_stats_update_end(&port->stats.syncp);
+	spin_unlock(&port->stats.lock);
+}
+
+static int airoha_dev_open(struct net_device *dev)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_eth *eth = port->eth;
+	int err;
+
+	netif_tx_start_all_queues(dev);
+	err = airoha_set_gdm_ports(eth, true);
+	if (err)
+		return err;
+
+	if (netdev_uses_dsa(dev))
+		airoha_fe_set(eth, REG_GDM_INGRESS_CFG(port->id),
+			      GDM_STAG_EN_MASK);
+	else
+		airoha_fe_clear(eth, REG_GDM_INGRESS_CFG(port->id),
+				GDM_STAG_EN_MASK);
+
+	airoha_qdma_set(eth, REG_QDMA_GLOBAL_CFG, GLOBAL_CFG_TX_DMA_EN_MASK);
+	airoha_qdma_set(eth, REG_QDMA_GLOBAL_CFG, GLOBAL_CFG_RX_DMA_EN_MASK);
+
+	return 0;
+}
+
+static int airoha_dev_stop(struct net_device *dev)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_eth *eth = port->eth;
+	int err;
+
+	netif_tx_disable(dev);
+	err = airoha_set_gdm_ports(eth, false);
+	if (err)
+		return err;
+
+	airoha_qdma_clear(eth, REG_QDMA_GLOBAL_CFG, GLOBAL_CFG_TX_DMA_EN_MASK);
+	airoha_qdma_clear(eth, REG_QDMA_GLOBAL_CFG, GLOBAL_CFG_RX_DMA_EN_MASK);
+
+	return 0;
+}
+
+static int airoha_dev_set_macaddr(struct net_device *dev, void *p)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	int err;
+
+	err = eth_mac_addr(dev, p);
+	if (err)
+		return err;
+
+	airoha_set_macaddr(port->eth, dev->dev_addr);
+
+	return 0;
+}
+
+static int airoha_dev_init(struct net_device *dev)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+
+	airoha_set_macaddr(port->eth, dev->dev_addr);
+
+	return 0;
+}
+
+static void airoha_dev_get_stats64(struct net_device *dev,
+				   struct rtnl_link_stats64 *storage)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	unsigned int start;
+
+	airoha_update_hw_stats(port);
+	do {
+		start = u64_stats_fetch_begin(&port->stats.syncp);
+		storage->rx_packets = port->stats.rx_ok_pkts;
+		storage->tx_packets = port->stats.tx_ok_pkts;
+		storage->rx_bytes = port->stats.rx_ok_bytes;
+		storage->tx_bytes = port->stats.tx_ok_bytes;
+		storage->multicast = port->stats.rx_multicast;
+		storage->rx_errors = port->stats.rx_errors;
+		storage->rx_dropped = port->stats.rx_drops;
+		storage->tx_dropped = port->stats.tx_drops;
+		storage->rx_crc_errors = port->stats.rx_crc_error;
+		storage->rx_over_errors = port->stats.rx_over_errors;
+	} while (u64_stats_fetch_retry(&port->stats.syncp, start));
+}
+
+static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb,
+				   struct net_device *dev)
+{
+	struct skb_shared_info *sinfo = skb_shinfo(skb);
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	u32 msg0 = 0, msg1, len = skb_headlen(skb);
+	int i, qid = skb_get_queue_mapping(skb);
+	struct airoha_eth *eth = port->eth;
+	u32 nr_frags = 1 + sinfo->nr_frags;
+	struct netdev_queue *txq;
+	struct airoha_queue *q;
+	void *data = skb->data;
+	u16 index;
+	u8 fport;
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL)
+		msg0 |= FIELD_PREP(QDMA_ETH_TXMSG_TCO_MASK, 1) |
+			FIELD_PREP(QDMA_ETH_TXMSG_UCO_MASK, 1) |
+			FIELD_PREP(QDMA_ETH_TXMSG_ICO_MASK, 1);
+
+	/* TSO: fill MSS info in tcp checksum field */
+	if (skb_is_gso(skb)) {
+		if (skb_cow_head(skb, 0))
+			goto error;
+
+		if (sinfo->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6)) {
+			__be16 csum = cpu_to_be16(sinfo->gso_size);
+
+			tcp_hdr(skb)->check = (__force __sum16)csum;
+			msg0 |= FIELD_PREP(QDMA_ETH_TXMSG_TSO_MASK, 1);
+		}
+	}
+
+	fport = port->id == 4 ? FE_PSE_PORT_GDM4 : port->id;
+	msg1 = FIELD_PREP(QDMA_ETH_TXMSG_FPORT_MASK, fport) |
+	       FIELD_PREP(QDMA_ETH_TXMSG_METER_MASK, 0x7f);
+
+	q = &eth->q_tx[qid];
+	if (WARN_ON_ONCE(!q->ndesc))
+		goto error;
+
+	spin_lock_bh(&q->lock);
+
+	txq = netdev_get_tx_queue(dev, qid);
+	if (q->queued + nr_frags > q->ndesc) {
+		/* not enough space in the queue */
+		netif_tx_stop_queue(txq);
+		spin_unlock_bh(&q->lock);
+		return NETDEV_TX_BUSY;
+	}
+
+	index = q->head;
+	for (i = 0; i < nr_frags; i++) {
+		struct airoha_qdma_desc *desc = &q->desc[index];
+		struct airoha_queue_entry *e = &q->entry[index];
+		skb_frag_t *frag = &sinfo->frags[i];
+		dma_addr_t addr;
+		u32 val;
+
+		addr = dma_map_single(dev->dev.parent, data, len,
+				      DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(dev->dev.parent, addr)))
+			goto error_unmap;
+
+		index = (index + 1) % q->ndesc;
+
+		val = FIELD_PREP(QDMA_DESC_LEN_MASK, len);
+		if (i < nr_frags - 1)
+			val |= FIELD_PREP(QDMA_DESC_MORE_MASK, 1);
+		WRITE_ONCE(desc->ctrl, cpu_to_le32(val));
+		WRITE_ONCE(desc->addr, cpu_to_le32(addr));
+		val = FIELD_PREP(QDMA_DESC_NEXT_ID_MASK, index);
+		WRITE_ONCE(desc->data, cpu_to_le32(val));
+		WRITE_ONCE(desc->msg0, cpu_to_le32(msg0));
+		WRITE_ONCE(desc->msg1, cpu_to_le32(msg1));
+		WRITE_ONCE(desc->msg2, cpu_to_le32(0xffff));
+
+		e->skb = i ? NULL : skb;
+		e->dma_addr = addr;
+		e->dma_len = len;
+
+		airoha_qdma_rmw(eth, REG_TX_CPU_IDX(qid), TX_RING_CPU_IDX_MASK,
+				FIELD_PREP(TX_RING_CPU_IDX_MASK, index));
+
+		data = skb_frag_address(frag);
+		len = skb_frag_size(frag);
+	}
+
+	q->head = index;
+	q->queued += i;
+
+	skb_tx_timestamp(skb);
+	if (q->ndesc - q->queued < q->free_thr)
+		netif_tx_stop_queue(txq);
+
+	spin_unlock_bh(&q->lock);
+
+	return NETDEV_TX_OK;
+
+error_unmap:
+	for (i--; i >= 0; i++)
+		dma_unmap_single(dev->dev.parent, q->entry[i].dma_addr,
+				 q->entry[i].dma_len, DMA_TO_DEVICE);
+
+	spin_unlock_bh(&q->lock);
+error:
+	dev_kfree_skb_any(skb);
+	dev->stats.tx_dropped++;
+
+	return NETDEV_TX_OK;
+}
+
+static void airoha_ethtool_get_drvinfo(struct net_device *dev,
+				       struct ethtool_drvinfo *info)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_eth *eth = port->eth;
+
+	strscpy(info->driver, eth->dev->driver->name, sizeof(info->driver));
+	strscpy(info->bus_info, dev_name(eth->dev), sizeof(info->bus_info));
+}
+
+static void airoha_ethtool_get_mac_stats(struct net_device *dev,
+					 struct ethtool_eth_mac_stats *stats)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	unsigned int start;
+
+	airoha_update_hw_stats(port);
+	do {
+		start = u64_stats_fetch_begin(&port->stats.syncp);
+		stats->MulticastFramesXmittedOK = port->stats.tx_multicast;
+		stats->BroadcastFramesXmittedOK = port->stats.tx_broadcast;
+		stats->BroadcastFramesReceivedOK = port->stats.rx_broadcast;
+	} while (u64_stats_fetch_retry(&port->stats.syncp, start));
+}
+
+static const struct ethtool_rmon_hist_range airoha_ethtool_rmon_ranges[] = {
+	{    0,    64 },
+	{   65,   127 },
+	{  128,   255 },
+	{  256,   511 },
+	{  512,  1023 },
+	{ 1024,  1518 },
+	{ 1519, 10239 },
+	{},
+};
+
+static void
+airoha_ethtool_get_rmon_stats(struct net_device *dev,
+			      struct ethtool_rmon_stats *stats,
+			      const struct ethtool_rmon_hist_range **ranges)
+{
+	struct airoha_gdm_port *port = netdev_priv(dev);
+	struct airoha_hw_stats *hw_stats = &port->stats;
+	unsigned int start;
+
+	BUILD_BUG_ON(ARRAY_SIZE(airoha_ethtool_rmon_ranges) !=
+		     ARRAY_SIZE(hw_stats->tx_len) + 1);
+	BUILD_BUG_ON(ARRAY_SIZE(airoha_ethtool_rmon_ranges) !=
+		     ARRAY_SIZE(hw_stats->rx_len) + 1);
+
+	*ranges = airoha_ethtool_rmon_ranges;
+	airoha_update_hw_stats(port);
+	do {
+		int i;
+
+		start = u64_stats_fetch_begin(&port->stats.syncp);
+		stats->fragments = hw_stats->rx_fragment;
+		stats->jabbers = hw_stats->rx_jabber;
+		for (i = 0; i < ARRAY_SIZE(airoha_ethtool_rmon_ranges) - 1;
+		     i++) {
+			stats->hist[i] = hw_stats->rx_len[i];
+			stats->hist_tx[i] = hw_stats->tx_len[i];
+		}
+	} while (u64_stats_fetch_retry(&port->stats.syncp, start));
+}
+
+static const struct net_device_ops airoha_netdev_ops = {
+	.ndo_init		= airoha_dev_init,
+	.ndo_open		= airoha_dev_open,
+	.ndo_stop		= airoha_dev_stop,
+	.ndo_start_xmit		= airoha_dev_xmit,
+	.ndo_get_stats64        = airoha_dev_get_stats64,
+	.ndo_set_mac_address	= airoha_dev_set_macaddr,
+};
+
+static const struct ethtool_ops airoha_ethtool_ops = {
+	.get_drvinfo		= airoha_ethtool_get_drvinfo,
+	.get_eth_mac_stats      = airoha_ethtool_get_mac_stats,
+	.get_rmon_stats		= airoha_ethtool_get_rmon_stats,
+};
+
+static int airoha_alloc_gdm_port(struct airoha_eth *eth, struct device_node *np)
+{
+	const __be32 *id_ptr = of_get_property(np, "reg", NULL);
+	struct airoha_gdm_port *port;
+	struct net_device *dev;
+	int err, index;
+	u32 id;
+
+	if (!id_ptr) {
+		dev_err(eth->dev, "missing gdm port id\n");
+		return -EINVAL;
+	}
+
+	id = be32_to_cpup(id_ptr);
+	index = id - 1;
+
+	if (!id || id > ARRAY_SIZE(eth->ports)) {
+		dev_err(eth->dev, "invalid gdm port id: %d\n", id);
+		return -EINVAL;
+	}
+
+	if (eth->ports[index]) {
+		dev_err(eth->dev, "duplicate gdm port id: %d\n", id);
+		return -EINVAL;
+	}
+
+	dev = devm_alloc_etherdev_mqs(eth->dev, sizeof(*port),
+				      AIROHA_NUM_TX_RING, AIROHA_NUM_RX_RING);
+	if (!dev) {
+		dev_err(eth->dev, "alloc_etherdev failed\n");
+		return -ENOMEM;
+	}
+
+	dev->netdev_ops = &airoha_netdev_ops;
+	dev->ethtool_ops = &airoha_ethtool_ops;
+	dev->max_mtu = AIROHA_MAX_MTU;
+	dev->watchdog_timeo = 5 * HZ;
+	dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
+			   NETIF_F_TSO6 | NETIF_F_IPV6_CSUM |
+			   NETIF_F_SG | NETIF_F_TSO;
+	dev->features |= dev->hw_features;
+	dev->dev.of_node = np;
+	SET_NETDEV_DEV(dev, eth->dev);
+
+	err = of_get_ethdev_address(np, dev);
+	if (err) {
+		if (err == -EPROBE_DEFER)
+			return err;
+
+		eth_hw_addr_random(dev);
+		dev_info(eth->dev, "generated random MAC address %pM\n",
+			 dev->dev_addr);
+	}
+
+	port = netdev_priv(dev);
+	u64_stats_init(&port->stats.syncp);
+	spin_lock_init(&port->stats.lock);
+	port->dev = dev;
+	port->eth = eth;
+	port->id = id;
+	eth->ports[index] = port;
+
+	return register_netdev(dev);
+}
+
+static int airoha_probe(struct platform_device *pdev)
+{
+	struct device_node *np;
+	struct airoha_eth *eth;
+	int i, err;
+
+	eth = devm_kzalloc(&pdev->dev, sizeof(*eth), GFP_KERNEL);
+	if (!eth)
+		return -ENOMEM;
+
+	eth->dev = &pdev->dev;
+
+	err = dma_set_mask_and_coherent(eth->dev, DMA_BIT_MASK(32));
+	if (err) {
+		dev_err(eth->dev, "failed configuring DMA mask\n");
+		return err;
+	}
+
+	eth->fe_regs = devm_platform_ioremap_resource_byname(pdev, "fe");
+	if (IS_ERR(eth->fe_regs))
+		return dev_err_probe(eth->dev, PTR_ERR(eth->fe_regs),
+				     "failed to iomap fe regs\n");
+
+	eth->qdma_regs = devm_platform_ioremap_resource_byname(pdev, "qdma0");
+	if (IS_ERR(eth->qdma_regs))
+		return dev_err_probe(eth->dev, PTR_ERR(eth->qdma_regs),
+				     "failed to iomap qdma regs\n");
+
+	eth->rsts[0].id = "fe";
+	eth->rsts[1].id = "pdma";
+	eth->rsts[2].id = "qdma";
+	err = devm_reset_control_bulk_get_exclusive(eth->dev,
+						    ARRAY_SIZE(eth->rsts),
+						    eth->rsts);
+	if (err) {
+		dev_err(eth->dev, "failed to get bulk reset lines\n");
+		return err;
+	}
+
+	eth->xsi_rsts[0].id = "xsi-mac";
+	eth->xsi_rsts[1].id = "hsi0-mac";
+	eth->xsi_rsts[2].id = "hsi1-mac";
+	eth->xsi_rsts[3].id = "hsi-mac";
+	eth->xsi_rsts[4].id = "xfp-mac";
+	err = devm_reset_control_bulk_get_exclusive(eth->dev,
+						    ARRAY_SIZE(eth->xsi_rsts),
+						    eth->xsi_rsts);
+	if (err) {
+		dev_err(eth->dev, "failed to get bulk xsi reset lines\n");
+		return err;
+	}
+
+	spin_lock_init(&eth->irq_lock);
+	eth->irq = platform_get_irq(pdev, 0);
+	if (eth->irq < 0)
+		return eth->irq;
+
+	eth->napi_dev = alloc_netdev_dummy(0);
+	if (!eth->napi_dev)
+		return -ENOMEM;
+
+	/* Enable threaded NAPI by default */
+	eth->napi_dev->threaded = true;
+	strscpy(eth->napi_dev->name, "qdma_eth", sizeof(eth->napi_dev->name));
+	platform_set_drvdata(pdev, eth);
+
+	err = airoha_hw_init(eth);
+	if (err)
+		goto error;
+
+	airoha_qdma_start_napi(eth);
+	for_each_child_of_node(pdev->dev.of_node, np) {
+		if (!of_device_is_compatible(np, "airoha,eth-mac"))
+			continue;
+
+		if (!of_device_is_available(np))
+			continue;
+
+		err = airoha_alloc_gdm_port(eth, np);
+		if (err) {
+			of_node_put(np);
+			goto error;
+		}
+	}
+
+	return 0;
+
+error:
+	airoha_hw_cleanup(eth);
+	for (i = 0; i < ARRAY_SIZE(eth->ports); i++) {
+		struct airoha_gdm_port *port = eth->ports[i];
+
+		if (port && port->dev->reg_state == NETREG_REGISTERED)
+			unregister_netdev(port->dev);
+	}
+	free_netdev(eth->napi_dev);
+	platform_set_drvdata(pdev, NULL);
+
+	return err;
+}
+
+static void airoha_remove(struct platform_device *pdev)
+{
+	struct airoha_eth *eth = platform_get_drvdata(pdev);
+	int i;
+
+	airoha_hw_cleanup(eth);
+	for (i = 0; i < ARRAY_SIZE(eth->ports); i++) {
+		struct airoha_gdm_port *port = eth->ports[i];
+
+		if (!port)
+			continue;
+
+		airoha_dev_stop(port->dev);
+		unregister_netdev(port->dev);
+	}
+	free_netdev(eth->napi_dev);
+
+	platform_set_drvdata(pdev, NULL);
+}
+
+static const struct of_device_id of_airoha_match[] = {
+	{ .compatible = "airoha,en7581-eth" },
+	{ /* sentinel */ }
+};
+
+static struct platform_driver airoha_driver = {
+	.probe = airoha_probe,
+	.remove_new = airoha_remove,
+	.driver = {
+		.name = KBUILD_MODNAME,
+		.of_match_table = of_airoha_match,
+	},
+};
+module_platform_driver(airoha_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Lorenzo Bianconi <lorenzo@kernel.org>");
+MODULE_DESCRIPTION("Ethernet driver for Airoha SoC");
-- 
cgit v1.2.3-58-ga151


From 0ece498c27d8cd2fdad6f49a6abc34b8badd8fbc Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Fri, 10 May 2024 10:36:45 -0400
Subject: Bluetooth: MGMT: Make MGMT_OP_LOAD_CONN_PARAM update existing
 connection

This makes MGMT_OP_LOAD_CONN_PARAM update existing connection by
dectecting the request is just for one connection, parameters already
exists and there is a connection.

Since this is a new behavior the revision is also updated to enable
userspace to detect it.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_sync.h |  3 +++
 net/bluetooth/hci_sync.c         | 18 +++++++++++++++
 net/bluetooth/mgmt.c             | 50 ++++++++++++++++++++++++++++++++++++++--
 3 files changed, 69 insertions(+), 2 deletions(-)

diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h
index 534c3386e714..20168732f20e 100644
--- a/include/net/bluetooth/hci_sync.h
+++ b/include/net/bluetooth/hci_sync.h
@@ -138,6 +138,7 @@ int hci_suspend_sync(struct hci_dev *hdev);
 int hci_resume_sync(struct hci_dev *hdev);
 
 struct hci_conn;
+struct hci_conn_params;
 
 int hci_abort_conn_sync(struct hci_dev *hdev, struct hci_conn *conn, u8 reason);
 
@@ -156,3 +157,5 @@ int hci_connect_acl_sync(struct hci_dev *hdev, struct hci_conn *conn);
 int hci_connect_le_sync(struct hci_dev *hdev, struct hci_conn *conn);
 
 int hci_cancel_connect_sync(struct hci_dev *hdev, struct hci_conn *conn);
+int hci_le_conn_update_sync(struct hci_dev *hdev, struct hci_conn *conn,
+			    struct hci_conn_params *params);
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index eea34e6a236f..82db6092965b 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -6724,3 +6724,21 @@ int hci_cancel_connect_sync(struct hci_dev *hdev, struct hci_conn *conn)
 
 	return -ENOENT;
 }
+
+int hci_le_conn_update_sync(struct hci_dev *hdev, struct hci_conn *conn,
+			    struct hci_conn_params *params)
+{
+	struct hci_cp_le_conn_update cp;
+
+	memset(&cp, 0, sizeof(cp));
+	cp.handle		= cpu_to_le16(conn->handle);
+	cp.conn_interval_min	= cpu_to_le16(params->conn_min_interval);
+	cp.conn_interval_max	= cpu_to_le16(params->conn_max_interval);
+	cp.conn_latency		= cpu_to_le16(params->conn_latency);
+	cp.supervision_timeout	= cpu_to_le16(params->supervision_timeout);
+	cp.min_ce_len		= cpu_to_le16(0x0000);
+	cp.max_ce_len		= cpu_to_le16(0x0000);
+
+	return __hci_cmd_sync_status(hdev, HCI_OP_LE_CONN_UPDATE,
+				     sizeof(cp), &cp, HCI_CMD_TIMEOUT);
+}
diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 80f220b7e19d..20eca8a9c681 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -42,7 +42,7 @@
 #include "aosp.h"
 
 #define MGMT_VERSION	1
-#define MGMT_REVISION	22
+#define MGMT_REVISION	23
 
 static const u16 mgmt_commands[] = {
 	MGMT_OP_READ_INDEX_LIST,
@@ -7813,6 +7813,18 @@ unlock:
 	return err;
 }
 
+static int conn_update_sync(struct hci_dev *hdev, void *data)
+{
+	struct hci_conn_params *params = data;
+	struct hci_conn *conn;
+
+	conn = hci_conn_hash_lookup_le(hdev, &params->addr, params->addr_type);
+	if (!conn)
+		return -ECANCELED;
+
+	return hci_le_conn_update_sync(hdev, conn, params);
+}
+
 static int load_conn_param(struct sock *sk, struct hci_dev *hdev, void *data,
 			   u16 len)
 {
@@ -7846,13 +7858,15 @@ static int load_conn_param(struct sock *sk, struct hci_dev *hdev, void *data,
 
 	hci_dev_lock(hdev);
 
-	hci_conn_params_clear_disabled(hdev);
+	if (param_count > 1)
+		hci_conn_params_clear_disabled(hdev);
 
 	for (i = 0; i < param_count; i++) {
 		struct mgmt_conn_param *param = &cp->params[i];
 		struct hci_conn_params *hci_param;
 		u16 min, max, latency, timeout;
 		u8 addr_type;
+		bool update;
 
 		bt_dev_dbg(hdev, "Adding %pMR (type %u)", &param->addr.bdaddr,
 			   param->addr.type);
@@ -7879,6 +7893,19 @@ static int load_conn_param(struct sock *sk, struct hci_dev *hdev, void *data,
 			continue;
 		}
 
+		/* Detect when the loading is for an existing parameter then
+		 * attempt to trigger the connection update procedure.
+		 */
+		if (!i && param_count == 1) {
+			hci_param = hci_conn_params_lookup(hdev,
+							   &param->addr.bdaddr,
+							   addr_type);
+			if (hci_param)
+				update = true;
+			else
+				hci_conn_params_clear_disabled(hdev);
+		}
+
 		hci_param = hci_conn_params_add(hdev, &param->addr.bdaddr,
 						addr_type);
 		if (!hci_param) {
@@ -7890,6 +7917,25 @@ static int load_conn_param(struct sock *sk, struct hci_dev *hdev, void *data,
 		hci_param->conn_max_interval = max;
 		hci_param->conn_latency = latency;
 		hci_param->supervision_timeout = timeout;
+
+		/* Check if we need to trigger a connection update */
+		if (update) {
+			struct hci_conn *conn;
+
+			/* Lookup for existing connection as central and check
+			 * if parameters match and if they don't then trigger
+			 * a connection update.
+			 */
+			conn = hci_conn_hash_lookup_le(hdev, &hci_param->addr,
+						       addr_type);
+			if (conn && conn->role == HCI_ROLE_MASTER &&
+			    (conn->le_conn_min_interval != min ||
+			     conn->le_conn_max_interval != max ||
+			     conn->le_conn_latency != latency ||
+			     conn->le_supv_timeout != timeout))
+				hci_cmd_sync_queue(hdev, conn_update_sync,
+						   hci_param, NULL);
+		}
 	}
 
 	hci_dev_unlock(hdev);
-- 
cgit v1.2.3-58-ga151


From 56c695a823e4ee1e5294a8340d5afe5de73828ec Mon Sep 17 00:00:00 2001
From: Sven Peter <sven@svenpeter.dev>
Date: Wed, 15 May 2024 18:15:02 +0000
Subject: Bluetooth: hci_bcm4377: Use correct unit for timeouts

BCM4377_TIMEOUT is always used to wait for completitions and their API
expects a timeout in jiffies instead of msecs.

Fixes: 8a06127602de ("Bluetooth: hci_bcm4377: Add new driver for BCM4377 PCIe boards")
Signed-off-by: Sven Peter <sven@svenpeter.dev>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_bcm4377.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bluetooth/hci_bcm4377.c b/drivers/bluetooth/hci_bcm4377.c
index d90858ea2fe5..a77a30fdc630 100644
--- a/drivers/bluetooth/hci_bcm4377.c
+++ b/drivers/bluetooth/hci_bcm4377.c
@@ -32,7 +32,7 @@ enum bcm4377_chip {
 #define BCM4378_DEVICE_ID 0x5f69
 #define BCM4387_DEVICE_ID 0x5f71
 
-#define BCM4377_TIMEOUT 1000
+#define BCM4377_TIMEOUT msecs_to_jiffies(1000)
 
 /*
  * These devices only support DMA transactions inside a 32bit window
-- 
cgit v1.2.3-58-ga151


From 3b867602854058c6d26b146372f3b33c3a5040b9 Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Wed, 15 May 2024 18:15:03 +0000
Subject: Bluetooth: hci_bcm4377: Increase boot timeout

BCM4388 takes over 2 seconds to boot, so increase the timeout.

Signed-off-by: Hector Martin <marcan@marcan.st>
Reviewed-by: Sven Peter <sven@svenpeter.dev>
Signed-off-by: Sven Peter <sven@svenpeter.dev>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_bcm4377.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/bluetooth/hci_bcm4377.c b/drivers/bluetooth/hci_bcm4377.c
index a77a30fdc630..e6644cb0163c 100644
--- a/drivers/bluetooth/hci_bcm4377.c
+++ b/drivers/bluetooth/hci_bcm4377.c
@@ -33,6 +33,7 @@ enum bcm4377_chip {
 #define BCM4387_DEVICE_ID 0x5f71
 
 #define BCM4377_TIMEOUT msecs_to_jiffies(1000)
+#define BCM4377_BOOT_TIMEOUT msecs_to_jiffies(5000)
 
 /*
  * These devices only support DMA transactions inside a 32bit window
@@ -1862,7 +1863,7 @@ static int bcm4377_boot(struct bcm4377_data *bcm4377)
 	dev_dbg(&bcm4377->pdev->dev, "waiting for firmware to boot\n");
 
 	ret = wait_for_completion_interruptible_timeout(&bcm4377->event,
-							BCM4377_TIMEOUT);
+							BCM4377_BOOT_TIMEOUT);
 	if (ret == 0) {
 		ret = -ETIMEDOUT;
 		goto out_dma_free;
-- 
cgit v1.2.3-58-ga151


From c68bbf5e334b35b36ac5b9f0419f1f93f796bad1 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Wed, 15 May 2024 12:36:55 +0530
Subject: Bluetooth: btnxpuart: Fix Null pointer dereference in
 btnxpuart_flush()

This adds a check before freeing the rx->skb in flush and close
functions to handle the kernel crash seen while removing driver after FW
download fails or before FW download completes.

dmesg log:
[   54.634586] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000080
[   54.643398] Mem abort info:
[   54.646204]   ESR = 0x0000000096000004
[   54.649964]   EC = 0x25: DABT (current EL), IL = 32 bits
[   54.655286]   SET = 0, FnV = 0
[   54.658348]   EA = 0, S1PTW = 0
[   54.661498]   FSC = 0x04: level 0 translation fault
[   54.666391] Data abort info:
[   54.669273]   ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
[   54.674768]   CM = 0, WnR = 0, TnD = 0, TagAccess = 0
[   54.674771]   GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[   54.674775] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000048860000
[   54.674780] [0000000000000080] pgd=0000000000000000, p4d=0000000000000000
[   54.703880] Internal error: Oops: 0000000096000004 [#1] PREEMPT SMP
[   54.710152] Modules linked in: btnxpuart(-) overlay fsl_jr_uio caam_jr caamkeyblob_desc caamhash_desc caamalg_desc crypto_engine authenc libdes crct10dif_ce polyval_ce polyval_generic snd_soc_imx_spdif snd_soc_imx_card snd_soc_ak5558 snd_soc_ak4458 caam secvio error snd_soc_fsl_micfil snd_soc_fsl_spdif snd_soc_fsl_sai snd_soc_fsl_utils imx_pcm_dma gpio_ir_recv rc_core sch_fq_codel fuse
[   54.744357] CPU: 3 PID: 72 Comm: kworker/u9:0 Not tainted 6.6.3-otbr-g128004619037 #2
[   54.744364] Hardware name: FSL i.MX8MM EVK board (DT)
[   54.744368] Workqueue: hci0 hci_power_on
[   54.757244] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[   54.757249] pc : kfree_skb_reason+0x18/0xb0
[   54.772299] lr : btnxpuart_flush+0x40/0x58 [btnxpuart]
[   54.782921] sp : ffff8000805ebca0
[   54.782923] x29: ffff8000805ebca0 x28: ffffa5c6cf1869c0 x27: ffffa5c6cf186000
[   54.782931] x26: ffff377b84852400 x25: ffff377b848523c0 x24: ffff377b845e7230
[   54.782938] x23: ffffa5c6ce8dbe08 x22: ffffa5c6ceb65410 x21: 00000000ffffff92
[   54.782945] x20: ffffa5c6ce8dbe98 x19: ffffffffffffffac x18: ffffffffffffffff
[   54.807651] x17: 0000000000000000 x16: ffffa5c6ce2824ec x15: ffff8001005eb857
[   54.821917] x14: 0000000000000000 x13: ffffa5c6cf1a02e0 x12: 0000000000000642
[   54.821924] x11: 0000000000000040 x10: ffffa5c6cf19d690 x9 : ffffa5c6cf19d688
[   54.821931] x8 : ffff377b86000028 x7 : 0000000000000000 x6 : 0000000000000000
[   54.821938] x5 : ffff377b86000000 x4 : 0000000000000000 x3 : 0000000000000000
[   54.843331] x2 : 0000000000000000 x1 : 0000000000000002 x0 : ffffffffffffffac
[   54.857599] Call trace:
[   54.857601]  kfree_skb_reason+0x18/0xb0
[   54.863878]  btnxpuart_flush+0x40/0x58 [btnxpuart]
[   54.863888]  hci_dev_open_sync+0x3a8/0xa04
[   54.872773]  hci_power_on+0x54/0x2e4
[   54.881832]  process_one_work+0x138/0x260
[   54.881842]  worker_thread+0x32c/0x438
[   54.881847]  kthread+0x118/0x11c
[   54.881853]  ret_from_fork+0x10/0x20
[   54.896406] Code: a9be7bfd 910003fd f9000bf3 aa0003f3 (b940d400)
[   54.896410] ---[ end trace 0000000000000000 ]---

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Tested-by: Guillaume Legoupil <guillaume.legoupil@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index 9bfa9a6ad56c..9a4b9009cdfc 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -1253,8 +1253,10 @@ static int btnxpuart_close(struct hci_dev *hdev)
 	ps_wakeup(nxpdev);
 	serdev_device_close(nxpdev->serdev);
 	skb_queue_purge(&nxpdev->txq);
-	kfree_skb(nxpdev->rx_skb);
-	nxpdev->rx_skb = NULL;
+	if (!IS_ERR_OR_NULL(nxpdev->rx_skb)) {
+		kfree_skb(nxpdev->rx_skb);
+		nxpdev->rx_skb = NULL;
+	}
 	clear_bit(BTNXPUART_SERDEV_OPEN, &nxpdev->tx_state);
 	return 0;
 }
@@ -1269,8 +1271,10 @@ static int btnxpuart_flush(struct hci_dev *hdev)
 
 	cancel_work_sync(&nxpdev->tx_work);
 
-	kfree_skb(nxpdev->rx_skb);
-	nxpdev->rx_skb = NULL;
+	if (!IS_ERR_OR_NULL(nxpdev->rx_skb)) {
+		kfree_skb(nxpdev->rx_skb);
+		nxpdev->rx_skb = NULL;
+	}
 
 	return 0;
 }
-- 
cgit v1.2.3-58-ga151


From 2684dd614ccf085b64456d78e2c23911006d503d Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Wed, 15 May 2024 12:36:56 +0530
Subject: Bluetooth: btnxpuart: Enable status prints for firmware download

This enables prints for firmware download which can help automation
tests to verify firmware download functionality.

dmesg logs before:
modprobe btnxpuart
[ 1999.187264] Bluetooth: MGMT ver 1.22

dmesg logs with this patch:
modprobe btnxpuart
[16179.758515] Bluetooth: hci0: ChipID: 7601, Version: 0
[16179.764748] Bluetooth: hci0: Request Firmware: nxp/uartspi_n61x_v1.bin.se
[16181.217490] Bluetooth: hci0: FW Download Complete: 372696 bytes
[16182.701398] Bluetooth: MGMT ver 1.22

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Tested-by: Guillaume Legoupil <guillaume.legoupil@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index 9a4b9009cdfc..81ad7ea72004 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -693,7 +693,7 @@ static int nxp_request_firmware(struct hci_dev *hdev, const char *fw_name)
 	if (!strlen(nxpdev->fw_name)) {
 		snprintf(nxpdev->fw_name, MAX_FW_FILE_NAME_LEN, "%s", fw_name);
 
-		bt_dev_dbg(hdev, "Request Firmware: %s", nxpdev->fw_name);
+		bt_dev_info(hdev, "Request Firmware: %s", nxpdev->fw_name);
 		err = request_firmware(&nxpdev->fw, nxpdev->fw_name, &hdev->dev);
 		if (err < 0) {
 			bt_dev_err(hdev, "Firmware file %s not found", nxpdev->fw_name);
@@ -781,7 +781,7 @@ static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
 	}
 
 	if (!len) {
-		bt_dev_dbg(hdev, "FW Downloaded Successfully: %zu bytes",
+		bt_dev_info(hdev, "FW Download Complete: %zu bytes",
 			   nxpdev->fw->size);
 		if (nxp_data->helper_fw_name && !nxpdev->helper_downloaded) {
 			nxpdev->helper_downloaded = true;
@@ -890,6 +890,7 @@ static int nxp_recv_chip_ver_v3(struct hci_dev *hdev, struct sk_buff *skb)
 
 	chip_id = le16_to_cpu(req->chip_id);
 	loader_ver = req->loader_ver;
+	bt_dev_info(hdev, "ChipID: %04x, Version: %d", chip_id, loader_ver);
 	if (!nxp_request_firmware(hdev, nxp_get_fw_name_from_chipid(hdev,
 								    chip_id, loader_ver)))
 		nxp_send_ack(NXP_ACK_V3, hdev);
@@ -934,7 +935,7 @@ static int nxp_recv_fw_req_v3(struct hci_dev *hdev, struct sk_buff *skb)
 	}
 
 	if (req->len == 0) {
-		bt_dev_dbg(hdev, "FW Downloaded Successfully: %zu bytes",
+		bt_dev_info(hdev, "FW Download Complete: %zu bytes",
 			   nxpdev->fw->size);
 		clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
 		wake_up_interruptible(&nxpdev->fw_dnld_done_wait_q);
@@ -1037,7 +1038,7 @@ static int nxp_setup(struct hci_dev *hdev)
 		if (err < 0)
 			return err;
 	} else {
-		bt_dev_dbg(hdev, "FW already running.");
+		bt_dev_info(hdev, "FW already running.");
 		clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
 	}
 
-- 
cgit v1.2.3-58-ga151


From e3c4891098c875a63ab0c3b31d584f6d4f1895fd Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Wed, 15 May 2024 12:36:57 +0530
Subject: Bluetooth: btnxpuart: Handle FW Download Abort scenario

This adds a new flag BTNXPUART_FW_DOWNLOAD_ABORT which handles the
situation where driver is removed while firmware download is in
progress.

logs:
modprobe btnxpuart
[65239.230431] Bluetooth: hci0: ChipID: 7601, Version: 0
[65239.236670] Bluetooth: hci0: Request Firmware: nxp/uartspi_n61x_v1.bin.se
rmmod btnxpuart
[65241.425300] Bluetooth: hci0: FW Download Aborted

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Tested-by: Guillaume Legoupil <guillaume.legoupil@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 47 ++++++++++++++++++++++++++++++-------------
 1 file changed, 33 insertions(+), 14 deletions(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index 81ad7ea72004..ecefae4f5cd9 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -29,6 +29,7 @@
 #define BTNXPUART_CHECK_BOOT_SIGNATURE	3
 #define BTNXPUART_SERDEV_OPEN		4
 #define BTNXPUART_IR_IN_PROGRESS	5
+#define BTNXPUART_FW_DOWNLOAD_ABORT	6
 
 /* NXP HW err codes */
 #define BTNXPUART_IR_HW_ERR		0xb0
@@ -159,6 +160,7 @@ struct btnxpuart_dev {
 	u8 fw_name[MAX_FW_FILE_NAME_LEN];
 	u32 fw_dnld_v1_offset;
 	u32 fw_v1_sent_bytes;
+	u32 fw_dnld_v3_offset;
 	u32 fw_v3_offset_correction;
 	u32 fw_v1_expected_len;
 	u32 boot_reg_offset;
@@ -550,6 +552,7 @@ static int nxp_download_firmware(struct hci_dev *hdev)
 	nxpdev->fw_v1_sent_bytes = 0;
 	nxpdev->fw_v1_expected_len = HDR_LEN;
 	nxpdev->boot_reg_offset = 0;
+	nxpdev->fw_dnld_v3_offset = 0;
 	nxpdev->fw_v3_offset_correction = 0;
 	nxpdev->baudrate_changed = false;
 	nxpdev->timeout_changed = false;
@@ -564,14 +567,23 @@ static int nxp_download_firmware(struct hci_dev *hdev)
 					       !test_bit(BTNXPUART_FW_DOWNLOADING,
 							 &nxpdev->tx_state),
 					       msecs_to_jiffies(60000));
+
+	release_firmware(nxpdev->fw);
+	memset(nxpdev->fw_name, 0, sizeof(nxpdev->fw_name));
+
 	if (err == 0) {
-		bt_dev_err(hdev, "FW Download Timeout.");
+		bt_dev_err(hdev, "FW Download Timeout. offset: %d",
+				nxpdev->fw_dnld_v1_offset ?
+				nxpdev->fw_dnld_v1_offset :
+				nxpdev->fw_dnld_v3_offset);
 		return -ETIMEDOUT;
 	}
+	if (test_bit(BTNXPUART_FW_DOWNLOAD_ABORT, &nxpdev->tx_state)) {
+		bt_dev_err(hdev, "FW Download Aborted");
+		return -EINTR;
+	}
 
 	serdev_device_set_flow_control(nxpdev->serdev, true);
-	release_firmware(nxpdev->fw);
-	memset(nxpdev->fw_name, 0, sizeof(nxpdev->fw_name));
 
 	/* Allow the downloaded FW to initialize */
 	msleep(1200);
@@ -955,8 +967,9 @@ static int nxp_recv_fw_req_v3(struct hci_dev *hdev, struct sk_buff *skb)
 		goto free_skb;
 	}
 
-	serdev_device_write_buf(nxpdev->serdev, nxpdev->fw->data + offset -
-				nxpdev->fw_v3_offset_correction, len);
+	nxpdev->fw_dnld_v3_offset = offset - nxpdev->fw_v3_offset_correction;
+	serdev_device_write_buf(nxpdev->serdev, nxpdev->fw->data +
+				nxpdev->fw_dnld_v3_offset, len);
 
 free_skb:
 	kfree_skb(skb);
@@ -1390,16 +1403,22 @@ static void nxp_serdev_remove(struct serdev_device *serdev)
 	struct btnxpuart_dev *nxpdev = serdev_device_get_drvdata(serdev);
 	struct hci_dev *hdev = nxpdev->hdev;
 
-	/* Restore FW baudrate to fw_init_baudrate if changed.
-	 * This will ensure FW baudrate is in sync with
-	 * driver baudrate in case this driver is re-inserted.
-	 */
-	if (nxpdev->current_baudrate != nxpdev->fw_init_baudrate) {
-		nxpdev->new_baudrate = nxpdev->fw_init_baudrate;
-		nxp_set_baudrate_cmd(hdev, NULL);
+	if (is_fw_downloading(nxpdev)) {
+		set_bit(BTNXPUART_FW_DOWNLOAD_ABORT, &nxpdev->tx_state);
+		clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
+		wake_up_interruptible(&nxpdev->check_boot_sign_wait_q);
+		wake_up_interruptible(&nxpdev->fw_dnld_done_wait_q);
+	} else {
+		/* Restore FW baudrate to fw_init_baudrate if changed.
+		 * This will ensure FW baudrate is in sync with
+		 * driver baudrate in case this driver is re-inserted.
+		 */
+		if (nxpdev->current_baudrate != nxpdev->fw_init_baudrate) {
+			nxpdev->new_baudrate = nxpdev->fw_init_baudrate;
+			nxp_set_baudrate_cmd(hdev, NULL);
+		}
+		ps_cancel_timer(nxpdev);
 	}
-
-	ps_cancel_timer(nxpdev);
 	hci_unregister_dev(hdev);
 	hci_free_dev(hdev);
 }
-- 
cgit v1.2.3-58-ga151


From 0d0df1e750bac0fdaa77940e711c1625cff08d33 Mon Sep 17 00:00:00 2001
From: Luke Wang <ziniu.wang_1@nxp.com>
Date: Fri, 17 May 2024 19:15:35 +0800
Subject: Bluetooth: btnxpuart: Shutdown timer and prevent rearming when driver
 unloading

When unload the btnxpuart driver, its associated timer will be deleted.
If the timer happens to be modified at this moment, it leads to the
kernel call this timer even after the driver unloaded, resulting in
kernel panic.
Use timer_shutdown_sync() instead of del_timer_sync() to prevent rearming.

panic log:
  Internal error: Oops: 0000000086000007 [#1] PREEMPT SMP
  Modules linked in: algif_hash algif_skcipher af_alg moal(O) mlan(O) crct10dif_ce polyval_ce polyval_generic   snd_soc_imx_card snd_soc_fsl_asoc_card snd_soc_imx_audmux mxc_jpeg_encdec v4l2_jpeg snd_soc_wm8962 snd_soc_fsl_micfil   snd_soc_fsl_sai flexcan snd_soc_fsl_utils ap130x rpmsg_ctrl imx_pcm_dma can_dev rpmsg_char pwm_fan fuse [last unloaded:   btnxpuart]
  CPU: 5 PID: 723 Comm: memtester Tainted: G           O       6.6.23-lts-next-06207-g4aef2658ac28 #1
  Hardware name: NXP i.MX95 19X19 board (DT)
  pstate: 20400009 (nzCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
  pc : 0xffff80007a2cf464
  lr : call_timer_fn.isra.0+0x24/0x80
...
  Call trace:
   0xffff80007a2cf464
   __run_timers+0x234/0x280
   run_timer_softirq+0x20/0x40
   __do_softirq+0x100/0x26c
   ____do_softirq+0x10/0x1c
   call_on_irq_stack+0x24/0x4c
   do_softirq_own_stack+0x1c/0x2c
   irq_exit_rcu+0xc0/0xdc
   el0_interrupt+0x54/0xd8
   __el0_irq_handler_common+0x18/0x24
   el0t_64_irq_handler+0x10/0x1c
   el0t_64_irq+0x190/0x194
  Code: ???????? ???????? ???????? ???????? (????????)
  ---[ end trace 0000000000000000 ]---
  Kernel panic - not syncing: Oops: Fatal exception in interrupt
  SMP: stopping secondary CPUs
  Kernel Offset: disabled
  CPU features: 0x0,c0000000,40028143,1000721b
  Memory Limit: none
  ---[ end Kernel panic - not syncing: Oops: Fatal exception in interrupt ]---

Signed-off-by: Luke Wang <ziniu.wang_1@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index ecefae4f5cd9..50c1578ea3ed 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -330,7 +330,7 @@ static void ps_cancel_timer(struct btnxpuart_dev *nxpdev)
 	struct ps_data *psdata = &nxpdev->psdata;
 
 	flush_work(&psdata->work);
-	del_timer_sync(&psdata->ps_timer);
+	timer_shutdown_sync(&psdata->ps_timer);
 }
 
 static void ps_control(struct hci_dev *hdev, u8 ps_state)
-- 
cgit v1.2.3-58-ga151


From 0a3e2eca1daa5627c8ecd1554e3146de82d61dd2 Mon Sep 17 00:00:00 2001
From: Kiran K <kiran.k@intel.com>
Date: Thu, 16 May 2024 17:54:36 +0530
Subject: Bluetooth: btintel: Refactor btintel_set_ppag()

Current flow iterates the ACPI table associated with Bluetooth
controller looking for PPAG method. Method name can be directly passed
to acpi_evaluate_object function instead of iterating the table.

Fixes: c585a92b2f9c ("Bluetooth: btintel: Set Per Platform Antenna Gain(PPAG)")
Signed-off-by: Kiran K <kiran.k@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btintel.c | 119 +++++++++++++-------------------------------
 1 file changed, 34 insertions(+), 85 deletions(-)

diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c
index 0c855c3ee1c1..7ecc67deecb0 100644
--- a/drivers/bluetooth/btintel.c
+++ b/drivers/bluetooth/btintel.c
@@ -26,21 +26,11 @@
 #define ECDSA_OFFSET		644
 #define ECDSA_HEADER_LEN	320
 
-#define BTINTEL_PPAG_NAME   "PPAG"
-
 enum {
 	DSM_SET_WDISABLE2_DELAY = 1,
 	DSM_SET_RESET_METHOD = 3,
 };
 
-/* structure to store the PPAG data read from ACPI table */
-struct btintel_ppag {
-	u32	domain;
-	u32     mode;
-	acpi_status status;
-	struct hci_dev *hdev;
-};
-
 #define CMD_WRITE_BOOT_PARAMS	0xfc0e
 struct cmd_write_boot_params {
 	__le32 boot_addr;
@@ -1324,65 +1314,6 @@ static int btintel_read_debug_features(struct hci_dev *hdev,
 	return 0;
 }
 
-static acpi_status btintel_ppag_callback(acpi_handle handle, u32 lvl, void *data,
-					 void **ret)
-{
-	acpi_status status;
-	size_t len;
-	struct btintel_ppag *ppag = data;
-	union acpi_object *p, *elements;
-	struct acpi_buffer string = {ACPI_ALLOCATE_BUFFER, NULL};
-	struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL};
-	struct hci_dev *hdev = ppag->hdev;
-
-	status = acpi_get_name(handle, ACPI_FULL_PATHNAME, &string);
-	if (ACPI_FAILURE(status)) {
-		bt_dev_warn(hdev, "PPAG-BT: ACPI Failure: %s", acpi_format_exception(status));
-		return status;
-	}
-
-	len = strlen(string.pointer);
-	if (len < strlen(BTINTEL_PPAG_NAME)) {
-		kfree(string.pointer);
-		return AE_OK;
-	}
-
-	if (strncmp((char *)string.pointer + len - 4, BTINTEL_PPAG_NAME, 4)) {
-		kfree(string.pointer);
-		return AE_OK;
-	}
-	kfree(string.pointer);
-
-	status = acpi_evaluate_object(handle, NULL, NULL, &buffer);
-	if (ACPI_FAILURE(status)) {
-		ppag->status = status;
-		bt_dev_warn(hdev, "PPAG-BT: ACPI Failure: %s", acpi_format_exception(status));
-		return status;
-	}
-
-	p = buffer.pointer;
-	ppag = (struct btintel_ppag *)data;
-
-	if (p->type != ACPI_TYPE_PACKAGE || p->package.count != 2) {
-		kfree(buffer.pointer);
-		bt_dev_warn(hdev, "PPAG-BT: Invalid object type: %d or package count: %d",
-			    p->type, p->package.count);
-		ppag->status = AE_ERROR;
-		return AE_ERROR;
-	}
-
-	elements = p->package.elements;
-
-	/* PPAG table is located at element[1] */
-	p = &elements[1];
-
-	ppag->domain = (u32)p->package.elements[0].integer.value;
-	ppag->mode = (u32)p->package.elements[1].integer.value;
-	ppag->status = AE_OK;
-	kfree(buffer.pointer);
-	return AE_CTRL_TERMINATE;
-}
-
 static int btintel_set_debug_features(struct hci_dev *hdev,
 			       const struct intel_debug_features *features)
 {
@@ -2427,10 +2358,13 @@ error:
 
 static void btintel_set_ppag(struct hci_dev *hdev, struct intel_version_tlv *ver)
 {
-	struct btintel_ppag ppag;
 	struct sk_buff *skb;
 	struct hci_ppag_enable_cmd ppag_cmd;
 	acpi_handle handle;
+	struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL};
+	union acpi_object *p, *elements;
+	u32 domain, mode;
+	acpi_status status;
 
 	/* PPAG is not supported if CRF is HrP2, Jfp2, JfP1 */
 	switch (ver->cnvr_top & 0xFFF) {
@@ -2448,22 +2382,34 @@ static void btintel_set_ppag(struct hci_dev *hdev, struct intel_version_tlv *ver
 		return;
 	}
 
-	memset(&ppag, 0, sizeof(ppag));
-
-	ppag.hdev = hdev;
-	ppag.status = AE_NOT_FOUND;
-	acpi_walk_namespace(ACPI_TYPE_PACKAGE, handle, 1, NULL,
-			    btintel_ppag_callback, &ppag, NULL);
-
-	if (ACPI_FAILURE(ppag.status)) {
-		if (ppag.status == AE_NOT_FOUND) {
+	status = acpi_evaluate_object(handle, "PPAG", NULL, &buffer);
+	if (ACPI_FAILURE(status)) {
+		if (status == AE_NOT_FOUND) {
 			bt_dev_dbg(hdev, "PPAG-BT: ACPI entry not found");
 			return;
 		}
+		bt_dev_warn(hdev, "PPAG-BT: ACPI Failure: %s", acpi_format_exception(status));
+		return;
+	}
+
+	p = buffer.pointer;
+	if (p->type != ACPI_TYPE_PACKAGE || p->package.count != 2) {
+		bt_dev_warn(hdev, "PPAG-BT: Invalid object type: %d or package count: %d",
+			    p->type, p->package.count);
+		kfree(buffer.pointer);
 		return;
 	}
 
-	if (ppag.domain != 0x12) {
+	elements = p->package.elements;
+
+	/* PPAG table is located at element[1] */
+	p = &elements[1];
+
+	domain = (u32)p->package.elements[0].integer.value;
+	mode = (u32)p->package.elements[1].integer.value;
+	kfree(buffer.pointer);
+
+	if (domain != 0x12) {
 		bt_dev_dbg(hdev, "PPAG-BT: Bluetooth domain is disabled in ACPI firmware");
 		return;
 	}
@@ -2474,19 +2420,22 @@ static void btintel_set_ppag(struct hci_dev *hdev, struct intel_version_tlv *ver
 	 * BIT 1 : 0 Disabled in China
 	 *         1 Enabled in China
 	 */
-	if ((ppag.mode & 0x01) != BIT(0) && (ppag.mode & 0x02) != BIT(1)) {
-		bt_dev_dbg(hdev, "PPAG-BT: EU, China mode are disabled in CB/BIOS");
+	mode &= 0x03;
+
+	if (!mode) {
+		bt_dev_dbg(hdev, "PPAG-BT: EU, China mode are disabled in BIOS");
 		return;
 	}
 
-	ppag_cmd.ppag_enable_flags = cpu_to_le32(ppag.mode);
+	ppag_cmd.ppag_enable_flags = cpu_to_le32(mode);
 
-	skb = __hci_cmd_sync(hdev, INTEL_OP_PPAG_CMD, sizeof(ppag_cmd), &ppag_cmd, HCI_CMD_TIMEOUT);
+	skb = __hci_cmd_sync(hdev, INTEL_OP_PPAG_CMD, sizeof(ppag_cmd),
+			     &ppag_cmd, HCI_CMD_TIMEOUT);
 	if (IS_ERR(skb)) {
 		bt_dev_warn(hdev, "Failed to send PPAG Enable (%ld)", PTR_ERR(skb));
 		return;
 	}
-	bt_dev_info(hdev, "PPAG-BT: Enabled (Mode %d)", ppag.mode);
+	bt_dev_info(hdev, "PPAG-BT: Enabled (Mode %d)", mode);
 	kfree_skb(skb);
 }
 
-- 
cgit v1.2.3-58-ga151


From 00f993fdec06c8f036a1b9c8ee6b004c17143bd1 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 15 May 2024 16:15:17 -0700
Subject: Bluetooth: btmtk: add the function to get the fw name

Include a shared function to get the firmware name, to prevent repeating
code for similar chipsets.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btmtk.c | 18 ++++++++++++++++++
 drivers/bluetooth/btmtk.h |  8 ++++++++
 2 files changed, 26 insertions(+)

diff --git a/drivers/bluetooth/btmtk.c b/drivers/bluetooth/btmtk.c
index 812fd2a8f853..a27c251bf56e 100644
--- a/drivers/bluetooth/btmtk.c
+++ b/drivers/bluetooth/btmtk.c
@@ -103,6 +103,24 @@ static void btmtk_coredump_notify(struct hci_dev *hdev, int state)
 	}
 }
 
+void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id, u32 fw_ver,
+			   u32 fw_flavor)
+{
+	if (dev_id == 0x7925)
+		snprintf(buf, size,
+			 "mediatek/mt%04x/BT_RAM_CODE_MT%04x_1_%x_hdr.bin",
+			 dev_id & 0xffff, dev_id & 0xffff, (fw_ver & 0xff) + 1);
+	else if (dev_id == 0x7961 && fw_flavor)
+		snprintf(buf, size,
+			 "mediatek/BT_RAM_CODE_MT%04x_1a_%x_hdr.bin",
+			 dev_id & 0xffff, (fw_ver & 0xff) + 1);
+	else
+		snprintf(buf, size,
+			 "mediatek/BT_RAM_CODE_MT%04x_1_%x_hdr.bin",
+			 dev_id & 0xffff, (fw_ver & 0xff) + 1);
+}
+EXPORT_SYMBOL_GPL(btmtk_fw_get_filename);
+
 int btmtk_setup_firmware_79xx(struct hci_dev *hdev, const char *fwname,
 			      wmt_cmd_sync_func_t wmt_cmd_sync)
 {
diff --git a/drivers/bluetooth/btmtk.h b/drivers/bluetooth/btmtk.h
index cbcdb99a22e6..e76b8a358be8 100644
--- a/drivers/bluetooth/btmtk.h
+++ b/drivers/bluetooth/btmtk.h
@@ -160,6 +160,9 @@ int btmtk_register_coredump(struct hci_dev *hdev, const char *name,
 			    u32 fw_version);
 
 int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb);
+
+void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id, u32 fw_ver,
+			   u32 fw_flavor);
 #else
 
 static inline int btmtk_set_bdaddr(struct hci_dev *hdev,
@@ -194,4 +197,9 @@ static int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	return -EOPNOTSUPP;
 }
+
+static void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id,
+				  u32 fw_ver, u32 fw_flavor)
+{
+}
 #endif
-- 
cgit v1.2.3-58-ga151


From 7f935b21bee4d8a9f26b57ae6c3682161b247809 Mon Sep 17 00:00:00 2001
From: Sean Wang <sean.wang@mediatek.com>
Date: Wed, 15 May 2024 16:15:18 -0700
Subject: Bluetooth: btmtk: apply the common btmtk_fw_get_filename

Apply the common btmtk_fw_get_filename to avoid the similar coding in each
driver.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btmtksdio.c |  3 +++
 drivers/bluetooth/btusb.c     | 14 ++------------
 2 files changed, 5 insertions(+), 12 deletions(-)

diff --git a/drivers/bluetooth/btmtksdio.c b/drivers/bluetooth/btmtksdio.c
index 8ded9ef8089a..b4e560a17aa6 100644
--- a/drivers/bluetooth/btmtksdio.c
+++ b/drivers/bluetooth/btmtksdio.c
@@ -1117,6 +1117,9 @@ static int btmtksdio_setup(struct hci_dev *hdev)
 			return err;
 		}
 
+		btmtk_fw_get_filename(fwname, sizeof(fwname), dev_id,
+				      fw_version, 0);
+
 		snprintf(fwname, sizeof(fwname),
 			 "mediatek/BT_RAM_CODE_MT%04x_1_%x_hdr.bin",
 			 dev_id & 0xffff, (fw_version & 0xff) + 1);
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index e384ef6ff050..611135d1ce01 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -3166,18 +3166,8 @@ static int btusb_mtk_setup(struct hci_dev *hdev)
 	case 0x7922:
 	case 0x7961:
 	case 0x7925:
-		if (dev_id == 0x7925)
-			snprintf(fw_bin_name, sizeof(fw_bin_name),
-				 "mediatek/mt%04x/BT_RAM_CODE_MT%04x_1_%x_hdr.bin",
-				 dev_id & 0xffff, dev_id & 0xffff, (fw_version & 0xff) + 1);
-		else if (dev_id == 0x7961 && fw_flavor)
-			snprintf(fw_bin_name, sizeof(fw_bin_name),
-				 "mediatek/BT_RAM_CODE_MT%04x_1a_%x_hdr.bin",
-				 dev_id & 0xffff, (fw_version & 0xff) + 1);
-		else
-			snprintf(fw_bin_name, sizeof(fw_bin_name),
-				 "mediatek/BT_RAM_CODE_MT%04x_1_%x_hdr.bin",
-				 dev_id & 0xffff, (fw_version & 0xff) + 1);
+		btmtk_fw_get_filename(fw_bin_name, sizeof(fw_bin_name), dev_id,
+				      fw_version, fw_flavor);
 
 		err = btmtk_setup_firmware_79xx(hdev, fw_bin_name,
 						btusb_mtk_hci_wmt_sync);
-- 
cgit v1.2.3-58-ga151


From 4c0c28f2bbec0c51395fd1f13c697da67483964b Mon Sep 17 00:00:00 2001
From: Hao Qin <hao.qin@mediatek.com>
Date: Wed, 15 May 2024 16:15:19 -0700
Subject: Bluetooth: btusb: mediatek: refactor the function btusb_mtk_reset

Extract the function btusb_mtk_subsys_reset from the btusb_mtk_reset
for the future handling of resetting bluetooth controller without
the USB reset.

Co-developed-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Hao Qin <hao.qin@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 45 +++++++++++++++++++++++++++------------------
 1 file changed, 27 insertions(+), 18 deletions(-)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 611135d1ce01..602456e94fdb 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -3033,28 +3033,13 @@ static u32 btusb_mtk_reset_done(struct hci_dev *hdev)
 	return val & MTK_BT_RST_DONE;
 }
 
-static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
+static int btusb_mtk_subsys_reset(struct hci_dev *hdev, u32 dev_id)
 {
 	struct btusb_data *data = hci_get_drvdata(hdev);
-	struct btmediatek_data *mediatek;
 	u32 val;
 	int err;
 
-	/* It's MediaTek specific bluetooth reset mechanism via USB */
-	if (test_and_set_bit(BTUSB_HW_RESET_ACTIVE, &data->flags)) {
-		bt_dev_err(hdev, "last reset failed? Not resetting again");
-		return -EBUSY;
-	}
-
-	err = usb_autopm_get_interface(data->intf);
-	if (err < 0)
-		return err;
-
-	btusb_stop_traffic(data);
-	usb_kill_anchored_urbs(&data->tx_anchor);
-	mediatek = hci_get_priv(hdev);
-
-	if (mediatek->dev_id == 0x7925) {
+	if (dev_id == 0x7925) {
 		btusb_mtk_uhw_reg_read(data, MTK_BT_RESET_REG_CONNV3, &val);
 		val |= (1 << 5);
 		btusb_mtk_uhw_reg_write(data, MTK_BT_RESET_REG_CONNV3, val);
@@ -3098,8 +3083,32 @@ static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
 	if (!val)
 		bt_dev_err(hdev, "Can't get device id, subsys reset fail.");
 
-	usb_queue_reset_device(data->intf);
+	return err;
+}
 
+static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
+{
+	struct btusb_data *data = hci_get_drvdata(hdev);
+	struct btmediatek_data *mtk_data;
+	int err;
+
+	/* It's MediaTek specific bluetooth reset mechanism via USB */
+	if (test_and_set_bit(BTUSB_HW_RESET_ACTIVE, &data->flags)) {
+		bt_dev_err(hdev, "last reset failed? Not resetting again");
+		return -EBUSY;
+	}
+
+	err = usb_autopm_get_interface(data->intf);
+	if (err < 0)
+		return err;
+
+	btusb_stop_traffic(data);
+	usb_kill_anchored_urbs(&data->tx_anchor);
+	mtk_data = hci_get_priv(hdev);
+
+	err = btusb_mtk_subsys_reset(hdev, mtk_data->dev_id);
+
+	usb_queue_reset_device(data->intf);
 	clear_bit(BTUSB_HW_RESET_ACTIVE, &data->flags);
 
 	return err;
-- 
cgit v1.2.3-58-ga151


From ccfc8948d7e4d93cab341a99774b24586717d89a Mon Sep 17 00:00:00 2001
From: Hao Qin <hao.qin@mediatek.com>
Date: Wed, 15 May 2024 16:15:20 -0700
Subject: Bluetooth: btusb: mediatek: reset the controller before downloading
 the fw

Reset the controller before downloading the firmware to improve its
reliability. This includes situations like cold or warm reboots, ensuring
the controller is in its initial state before starting the firmware
download.

Co-developed-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Co-developed-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Hao Qin <hao.qin@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 602456e94fdb..ddd5ca7596e5 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -3175,6 +3175,13 @@ static int btusb_mtk_setup(struct hci_dev *hdev)
 	case 0x7922:
 	case 0x7961:
 	case 0x7925:
+		/* Reset the device to ensure it's in the initial state before
+		 * downloading the firmware to ensure.
+		 */
+
+		if (!test_bit(BTUSB_FIRMWARE_LOADED, &data->flags))
+			btusb_mtk_subsys_reset(hdev, dev_id);
+
 		btmtk_fw_get_filename(fw_bin_name, sizeof(fw_bin_name), dev_id,
 				      fw_version, fw_flavor);
 
@@ -3182,9 +3189,12 @@ static int btusb_mtk_setup(struct hci_dev *hdev)
 						btusb_mtk_hci_wmt_sync);
 		if (err < 0) {
 			bt_dev_err(hdev, "Failed to set up firmware (%d)", err);
+			clear_bit(BTUSB_FIRMWARE_LOADED, &data->flags);
 			return err;
 		}
 
+		set_bit(BTUSB_FIRMWARE_LOADED, &data->flags);
+
 		/* It's Device EndPoint Reset Option Register */
 		btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, MTK_EP_RST_IN_OUT_OPT);
 
-- 
cgit v1.2.3-58-ga151


From 064aee501012eaeeeb40a2acb991598e0a89fac9 Mon Sep 17 00:00:00 2001
From: Hao Qin <hao.qin@mediatek.com>
Date: Wed, 15 May 2024 16:15:21 -0700
Subject: Bluetooth: btusb: mediatek: add MT7922 subsystem reset

Add the support of MT7922 bluetooth subsystem reset that was called the
auto revert to self-recover from the fatal error in the controller like
the host encounters HCI cmd timeout or the controller crashes.

Co-developed-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Hao Qin <hao.qin@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index ddd5ca7596e5..79aefdb3324d 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -3039,7 +3039,16 @@ static int btusb_mtk_subsys_reset(struct hci_dev *hdev, u32 dev_id)
 	u32 val;
 	int err;
 
-	if (dev_id == 0x7925) {
+	if (dev_id == 0x7922) {
+		btusb_mtk_uhw_reg_read(data, MTK_BT_SUBSYS_RST, &val);
+		val |= 0x00002020;
+		btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, val);
+		btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, 0x00010001);
+		btusb_mtk_uhw_reg_read(data, MTK_BT_SUBSYS_RST, &val);
+		val |= BIT(0);
+		btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, val);
+		msleep(100);
+	} else if (dev_id == 0x7925) {
 		btusb_mtk_uhw_reg_read(data, MTK_BT_RESET_REG_CONNV3, &val);
 		val |= (1 << 5);
 		btusb_mtk_uhw_reg_write(data, MTK_BT_RESET_REG_CONNV3, val);
@@ -3079,6 +3088,9 @@ static int btusb_mtk_subsys_reset(struct hci_dev *hdev, u32 dev_id)
 	if (err < 0)
 		bt_dev_err(hdev, "Reset timeout");
 
+	if (dev_id == 0x7922)
+		btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT, 0x000000FF);
+
 	btusb_mtk_id_get(data, 0x70010200, &val);
 	if (!val)
 		bt_dev_err(hdev, "Can't get device id, subsys reset fail.");
-- 
cgit v1.2.3-58-ga151


From 17813af5656b1f47fc3bc2c486ac8014e0026ad8 Mon Sep 17 00:00:00 2001
From: Kiran K <kiran.k@intel.com>
Date: Fri, 17 May 2024 15:24:45 +0530
Subject: Bluetooth: btintel_pcie: Print Firmware Sequencer information

Firmware sequencer (FSEQ) is a common code shared across Bluetooth
and Wifi. Printing FSEQ will help to debug if there is any mismatch
between Bluetooth and Wifi FSEQ.

Make 'btintel_print_fseq_info' public and use it in btintel_pcie.c.

dmesg:

....

[ 5335.695740] Bluetooth: hci0: Device booted in 33872 usecs
[ 5335.695918] Bluetooth: hci0: Found Intel DDC parameters: intel/ibt-0190-0291.ddc
[ 5335.697011] Bluetooth: hci0: Applying Intel DDC parameters completed
[ 5335.697837] Bluetooth: hci0: Firmware timestamp 2024.20 buildtype 0 build 62871
[ 5335.697848] Bluetooth: hci0: Firmware SHA1: 0xeffdce06
[ 5335.698655] Bluetooth: hci0: Fseq status: Success (0x00)
[ 5335.698666] Bluetooth: hci0: Fseq executed: 00.00.04.176
[ 5335.698670] Bluetooth: hci0: Fseq BT Top: 00.00.04.176
[ 5335.750204] Bluetooth: MGMT ver 1.22

Signed-off-by: Kiran K <kiran.k@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btintel.c      | 3 ++-
 drivers/bluetooth/btintel.h      | 5 +++++
 drivers/bluetooth/btintel_pcie.c | 2 ++
 3 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c
index 7ecc67deecb0..27e03951e68b 100644
--- a/drivers/bluetooth/btintel.c
+++ b/drivers/bluetooth/btintel.c
@@ -2662,7 +2662,7 @@ void btintel_set_msft_opcode(struct hci_dev *hdev, u8 hw_variant)
 }
 EXPORT_SYMBOL_GPL(btintel_set_msft_opcode);
 
-static void btintel_print_fseq_info(struct hci_dev *hdev)
+void btintel_print_fseq_info(struct hci_dev *hdev)
 {
 	struct sk_buff *skb;
 	u8 *p;
@@ -2774,6 +2774,7 @@ static void btintel_print_fseq_info(struct hci_dev *hdev)
 
 	kfree_skb(skb);
 }
+EXPORT_SYMBOL_GPL(btintel_print_fseq_info);
 
 static int btintel_setup_combined(struct hci_dev *hdev)
 {
diff --git a/drivers/bluetooth/btintel.h b/drivers/bluetooth/btintel.h
index b5fea735e260..9dbad1a7c47c 100644
--- a/drivers/bluetooth/btintel.h
+++ b/drivers/bluetooth/btintel.h
@@ -244,6 +244,7 @@ int btintel_bootloader_setup_tlv(struct hci_dev *hdev,
 				 struct intel_version_tlv *ver);
 int btintel_shutdown_combined(struct hci_dev *hdev);
 void btintel_hw_error(struct hci_dev *hdev, u8 code);
+void btintel_print_fseq_info(struct hci_dev *hdev);
 #else
 
 static inline int btintel_check_bdaddr(struct hci_dev *hdev)
@@ -373,4 +374,8 @@ static inline int btintel_shutdown_combined(struct hci_dev *hdev)
 static inline void btintel_hw_error(struct hci_dev *hdev, u8 code)
 {
 }
+
+static inline void btintel_print_fseq_info(struct hci_dev *hdev)
+{
+}
 #endif
diff --git a/drivers/bluetooth/btintel_pcie.c b/drivers/bluetooth/btintel_pcie.c
index dd3c0626c72d..4ebd5b22a40f 100644
--- a/drivers/bluetooth/btintel_pcie.c
+++ b/drivers/bluetooth/btintel_pcie.c
@@ -1197,9 +1197,11 @@ static int btintel_pcie_setup(struct hci_dev *hdev)
 		bt_dev_err(hdev, "Unsupported Intel hw variant (%u)",
 			   INTEL_HW_VARIANT(ver_tlv.cnvi_bt));
 		err = -EINVAL;
+		goto exit_error;
 		break;
 	}
 
+	btintel_print_fseq_info(hdev);
 exit_error:
 	kfree_skb(skb);
 
-- 
cgit v1.2.3-58-ga151


From 041677e7aad6f57020aa271aafd4cb9e2af1536f Mon Sep 17 00:00:00 2001
From: Kiran K <kiran.k@intel.com>
Date: Fri, 17 May 2024 15:24:46 +0530
Subject: Bluetooth: btintel_pcie: Fix irq leak

Free irq before releasing irq vector.

Fixes: c2b636b3f788 ("Bluetooth: btintel_pcie: Add support for PCIe transport")
Signed-off-by: Kiran K <kiran.k@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btintel_pcie.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/bluetooth/btintel_pcie.c b/drivers/bluetooth/btintel_pcie.c
index 4ebd5b22a40f..58144c82b1cb 100644
--- a/drivers/bluetooth/btintel_pcie.c
+++ b/drivers/bluetooth/btintel_pcie.c
@@ -1329,6 +1329,12 @@ static void btintel_pcie_remove(struct pci_dev *pdev)
 	data = pci_get_drvdata(pdev);
 
 	btintel_pcie_reset_bt(data);
+	for (int i = 0; i < data->alloc_vecs; i++) {
+		struct msix_entry *msix_entry;
+
+		msix_entry = &data->msix_entries[i];
+		free_irq(msix_entry->vector, msix_entry);
+	}
 
 	pci_free_irq_vectors(pdev);
 
-- 
cgit v1.2.3-58-ga151


From 8f7dfe171c576aaec4911cc59feaed26d79c7c7f Mon Sep 17 00:00:00 2001
From: Erick Archer <erick.archer@outlook.com>
Date: Sat, 18 May 2024 10:30:38 +0200
Subject: Bluetooth: hci_core: Prefer struct_size over open coded arithmetic

This is an effort to get rid of all multiplications from allocation
functions in order to prevent integer overflows [1][2].

As the "dl" variable is a pointer to "struct hci_dev_list_req" and this
structure ends in a flexible array:

struct hci_dev_list_req {
	[...]
	struct hci_dev_req dev_req[];	/* hci_dev_req structures */
};

the preferred way in the kernel is to use the struct_size() helper to
do the arithmetic instead of the calculation "size + count * size" in
the kzalloc() and copy_to_user() functions.

At the same time, prepare for the coming implementation by GCC and Clang
of the __counted_by attribute. Flexible array members annotated with
__counted_by can have their accesses bounds-checked at run-time via
CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for
strcpy/memcpy-family functions).

In this case, it is important to note that the logic needs a little
refactoring to ensure that the "dev_num" member is initialized before
the first access to the flex array. Specifically, add the assignment
before the list_for_each_entry() loop.

Also remove the "size" variable as it is no longer needed.

This way, the code is more readable and safer.

This code was detected with the help of Coccinelle, and audited and
modified manually.

Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#open-coded-arithmetic-in-allocator-arguments [1]
Link: https://github.com/KSPP/linux/issues/160 [2]
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Erick Archer <erick.archer@outlook.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_sock.h |  2 +-
 net/bluetooth/hci_core.c         | 11 ++++-------
 2 files changed, 5 insertions(+), 8 deletions(-)

diff --git a/include/net/bluetooth/hci_sock.h b/include/net/bluetooth/hci_sock.h
index 9949870f7d78..13e8cd4414a1 100644
--- a/include/net/bluetooth/hci_sock.h
+++ b/include/net/bluetooth/hci_sock.h
@@ -144,7 +144,7 @@ struct hci_dev_req {
 
 struct hci_dev_list_req {
 	__u16  dev_num;
-	struct hci_dev_req dev_req[];	/* hci_dev_req structures */
+	struct hci_dev_req dev_req[] __counted_by(dev_num);
 };
 
 struct hci_conn_list_req {
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index c644b30977bd..a2cad8a982f6 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -801,7 +801,7 @@ int hci_get_dev_list(void __user *arg)
 	struct hci_dev *hdev;
 	struct hci_dev_list_req *dl;
 	struct hci_dev_req *dr;
-	int n = 0, size, err;
+	int n = 0, err;
 	__u16 dev_num;
 
 	if (get_user(dev_num, (__u16 __user *) arg))
@@ -810,12 +810,11 @@ int hci_get_dev_list(void __user *arg)
 	if (!dev_num || dev_num > (PAGE_SIZE * 2) / sizeof(*dr))
 		return -EINVAL;
 
-	size = sizeof(*dl) + dev_num * sizeof(*dr);
-
-	dl = kzalloc(size, GFP_KERNEL);
+	dl = kzalloc(struct_size(dl, dev_req, dev_num), GFP_KERNEL);
 	if (!dl)
 		return -ENOMEM;
 
+	dl->dev_num = dev_num;
 	dr = dl->dev_req;
 
 	read_lock(&hci_dev_list_lock);
@@ -838,9 +837,7 @@ int hci_get_dev_list(void __user *arg)
 	read_unlock(&hci_dev_list_lock);
 
 	dl->dev_num = n;
-	size = sizeof(*dl) + n * sizeof(*dr);
-
-	err = copy_to_user(arg, dl, size);
+	err = copy_to_user(arg, dl, struct_size(dl, dev_req, n));
 	kfree(dl);
 
 	return err ? -EFAULT : 0;
-- 
cgit v1.2.3-58-ga151


From c61e41121036aa610e904ef60f8520e10455ee8c Mon Sep 17 00:00:00 2001
From: Erick Archer <erick.archer@outlook.com>
Date: Sat, 18 May 2024 10:30:39 +0200
Subject: Bluetooth: hci_core: Prefer array indexing over pointer arithmetic

Refactor the list_for_each_entry() loop of hci_get_dev_list()
function to use array indexing instead of pointer arithmetic.

This way, the code is more readable and idiomatic.

Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Erick Archer <erick.archer@outlook.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index a2cad8a982f6..55bdc365916f 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -828,8 +828,8 @@ int hci_get_dev_list(void __user *arg)
 		if (hci_dev_test_flag(hdev, HCI_AUTO_OFF))
 			flags &= ~BIT(HCI_UP);
 
-		(dr + n)->dev_id  = hdev->id;
-		(dr + n)->dev_opt = flags;
+		dr[n].dev_id  = hdev->id;
+		dr[n].dev_opt = flags;
 
 		if (++n >= dev_num)
 			break;
-- 
cgit v1.2.3-58-ga151


From 7d2c7ddba6238e6a14cd89ef869878dd22f2a661 Mon Sep 17 00:00:00 2001
From: Erick Archer <erick.archer@outlook.com>
Date: Fri, 17 May 2024 19:21:49 +0200
Subject: tty: rfcomm: prefer struct_size over open coded arithmetic

This is an effort to get rid of all multiplications from allocation
functions in order to prevent integer overflows [1][2].

As the "dl" variable is a pointer to "struct rfcomm_dev_list_req" and
this structure ends in a flexible array:

struct rfcomm_dev_list_req {
	[...]
	struct   rfcomm_dev_info dev_info[];
};

the preferred way in the kernel is to use the struct_size() helper to
do the arithmetic instead of the calculation "size + count * size" in
the kzalloc() and copy_to_user() functions.

At the same time, prepare for the coming implementation by GCC and Clang
of the __counted_by attribute. Flexible array members annotated with
__counted_by can have their accesses bounds-checked at run-time via
CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for
strcpy/memcpy-family functions).

In this case, it is important to note that the logic needs a little
refactoring to ensure that the "dev_num" member is initialized before
the first access to the flex array. Specifically, add the assignment
before the list_for_each_entry() loop.

Also remove the "size" variable as it is no longer needed.

This way, the code is more readable and safer.

This code was detected with the help of Coccinelle, and audited and
modified manually.

Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#open-coded-arithmetic-in-allocator-arguments [1]
Link: https://github.com/KSPP/linux/issues/160 [2]
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Erick Archer <erick.archer@outlook.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/rfcomm.h |  2 +-
 net/bluetooth/rfcomm/tty.c     | 11 ++++-------
 2 files changed, 5 insertions(+), 8 deletions(-)

diff --git a/include/net/bluetooth/rfcomm.h b/include/net/bluetooth/rfcomm.h
index 99d26879b02a..c05882476900 100644
--- a/include/net/bluetooth/rfcomm.h
+++ b/include/net/bluetooth/rfcomm.h
@@ -355,7 +355,7 @@ struct rfcomm_dev_info {
 
 struct rfcomm_dev_list_req {
 	u16      dev_num;
-	struct   rfcomm_dev_info dev_info[];
+	struct   rfcomm_dev_info dev_info[] __counted_by(dev_num);
 };
 
 int  rfcomm_dev_ioctl(struct sock *sk, unsigned int cmd, void __user *arg);
diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c
index 69c75c041fe1..44b781e7569e 100644
--- a/net/bluetooth/rfcomm/tty.c
+++ b/net/bluetooth/rfcomm/tty.c
@@ -504,7 +504,7 @@ static int rfcomm_get_dev_list(void __user *arg)
 	struct rfcomm_dev *dev;
 	struct rfcomm_dev_list_req *dl;
 	struct rfcomm_dev_info *di;
-	int n = 0, size, err;
+	int n = 0, err;
 	u16 dev_num;
 
 	BT_DBG("");
@@ -515,12 +515,11 @@ static int rfcomm_get_dev_list(void __user *arg)
 	if (!dev_num || dev_num > (PAGE_SIZE * 4) / sizeof(*di))
 		return -EINVAL;
 
-	size = sizeof(*dl) + dev_num * sizeof(*di);
-
-	dl = kzalloc(size, GFP_KERNEL);
+	dl = kzalloc(struct_size(dl, dev_info, dev_num), GFP_KERNEL);
 	if (!dl)
 		return -ENOMEM;
 
+	dl->dev_num = dev_num;
 	di = dl->dev_info;
 
 	mutex_lock(&rfcomm_dev_lock);
@@ -542,9 +541,7 @@ static int rfcomm_get_dev_list(void __user *arg)
 	mutex_unlock(&rfcomm_dev_lock);
 
 	dl->dev_num = n;
-	size = sizeof(*dl) + n * sizeof(*di);
-
-	err = copy_to_user(arg, dl, size);
+	err = copy_to_user(arg, dl, struct_size(dl, dev_info, n));
 	kfree(dl);
 
 	return err ? -EFAULT : 0;
-- 
cgit v1.2.3-58-ga151


From b1c7cd6caaf6209ed9f7df59b61a2ac885eb79b3 Mon Sep 17 00:00:00 2001
From: Erick Archer <erick.archer@outlook.com>
Date: Fri, 17 May 2024 19:21:50 +0200
Subject: tty: rfcomm: prefer array indexing over pointer arithmetic

Refactor the list_for_each_entry() loop of rfcomm_get_dev_list()
function to use array indexing instead of pointer arithmetic.

This way, the code is more readable and idiomatic.

Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Erick Archer <erick.archer@outlook.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/rfcomm/tty.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c
index 44b781e7569e..af80d599c337 100644
--- a/net/bluetooth/rfcomm/tty.c
+++ b/net/bluetooth/rfcomm/tty.c
@@ -527,12 +527,12 @@ static int rfcomm_get_dev_list(void __user *arg)
 	list_for_each_entry(dev, &rfcomm_dev_list, list) {
 		if (!tty_port_get(&dev->port))
 			continue;
-		(di + n)->id      = dev->id;
-		(di + n)->flags   = dev->flags;
-		(di + n)->state   = dev->dlc->state;
-		(di + n)->channel = dev->channel;
-		bacpy(&(di + n)->src, &dev->src);
-		bacpy(&(di + n)->dst, &dev->dst);
+		di[n].id      = dev->id;
+		di[n].flags   = dev->flags;
+		di[n].state   = dev->dlc->state;
+		di[n].channel = dev->channel;
+		bacpy(&di[n].src, &dev->src);
+		bacpy(&di[n].dst, &dev->dst);
 		tty_port_put(&dev->port);
 		if (++n >= dev_num)
 			break;
-- 
cgit v1.2.3-58-ga151


From fbcd1492d8c0ae39bf9f1d3d85228e7e279c4d24 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Thu, 9 May 2024 01:11:38 +0100
Subject: Bluetooth/nokia: Remove unused struct 'hci_nokia_radio_hdr'

'hci_nokia_radio_hdr' looks like it was unused since it's
initial commit.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_nokia.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/bluetooth/hci_nokia.c b/drivers/bluetooth/hci_nokia.c
index 97da0b2bfd17..62633d9ba7c4 100644
--- a/drivers/bluetooth/hci_nokia.c
+++ b/drivers/bluetooth/hci_nokia.c
@@ -116,11 +116,6 @@ struct hci_nokia_neg_evt {
 #define SETUP_BAUD_RATE		921600
 #define INIT_BAUD_RATE		120000
 
-struct hci_nokia_radio_hdr {
-	u8	evt;
-	u8	dlen;
-} __packed;
-
 struct nokia_bt_dev {
 	struct hci_uart hu;
 	struct serdev_device *serdev;
-- 
cgit v1.2.3-58-ga151


From 183469bccf68cb05dfb56cc47e73520217854bab Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Fri, 24 May 2024 13:14:02 +0300
Subject: Bluetooth: MGMT: Uninitialized variable in load_conn_param()

The "update" variable needs to be initialized to false.

Fixes: 0ece498c27d8 ("Bluetooth: MGMT: Make MGMT_OP_LOAD_CONN_PARAM update existing connection")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/mgmt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 20eca8a9c681..3ab1558ff391 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -7865,8 +7865,8 @@ static int load_conn_param(struct sock *sk, struct hci_dev *hdev, void *data,
 		struct mgmt_conn_param *param = &cp->params[i];
 		struct hci_conn_params *hci_param;
 		u16 min, max, latency, timeout;
+		bool update = false;
 		u8 addr_type;
-		bool update;
 
 		bt_dev_dbg(hdev, "Adding %pMR (type %u)", &param->addr.bdaddr,
 			   param->addr.type);
-- 
cgit v1.2.3-58-ga151


From 973dd9c4db4746200f88fe46e30eada7054fdbea Mon Sep 17 00:00:00 2001
From: Erick Archer <erick.archer@outlook.com>
Date: Fri, 24 May 2024 19:11:51 +0200
Subject: Bluetooth: Use sizeof(*pointer) instead of sizeof(type)

It is preferred to use sizeof(*pointer) instead of sizeof(type)
due to the type of the variable can change and one needs not
change the former (unlike the latter). This patch has no effect
on runtime behavior.

Signed-off-by: Erick Archer <erick.archer@outlook.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btrtl.c     | 2 +-
 drivers/bluetooth/hci_ldisc.c | 2 +-
 drivers/bluetooth/hci_qca.c   | 5 ++---
 drivers/bluetooth/hci_vhci.c  | 2 +-
 4 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c
index 4f1e37b4f780..f2f37143c454 100644
--- a/drivers/bluetooth/btrtl.c
+++ b/drivers/bluetooth/btrtl.c
@@ -811,7 +811,7 @@ static int rtl_download_firmware(struct hci_dev *hdev,
 	struct sk_buff *skb;
 	struct hci_rp_read_local_version *rp;
 
-	dl_cmd = kmalloc(sizeof(struct rtl_download_cmd), GFP_KERNEL);
+	dl_cmd = kmalloc(sizeof(*dl_cmd), GFP_KERNEL);
 	if (!dl_cmd)
 		return -ENOMEM;
 
diff --git a/drivers/bluetooth/hci_ldisc.c b/drivers/bluetooth/hci_ldisc.c
index 17a2f158a0df..30192bb08354 100644
--- a/drivers/bluetooth/hci_ldisc.c
+++ b/drivers/bluetooth/hci_ldisc.c
@@ -488,7 +488,7 @@ static int hci_uart_tty_open(struct tty_struct *tty)
 	if (tty->ops->write == NULL)
 		return -EOPNOTSUPP;
 
-	hu = kzalloc(sizeof(struct hci_uart), GFP_KERNEL);
+	hu = kzalloc(sizeof(*hu), GFP_KERNEL);
 	if (!hu) {
 		BT_ERR("Can't allocate control structure");
 		return -ENFILE;
diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 9a0bc86f9aac..2f47267508d5 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -569,7 +569,7 @@ static int qca_open(struct hci_uart *hu)
 	if (!hci_uart_has_flow_control(hu))
 		return -EOPNOTSUPP;
 
-	qca = kzalloc(sizeof(struct qca_data), GFP_KERNEL);
+	qca = kzalloc(sizeof(*qca), GFP_KERNEL);
 	if (!qca)
 		return -ENOMEM;
 
@@ -1040,8 +1040,7 @@ static void qca_controller_memdump(struct work_struct *work)
 		}
 
 		if (!qca_memdump) {
-			qca_memdump = kzalloc(sizeof(struct qca_memdump_info),
-					      GFP_ATOMIC);
+			qca_memdump = kzalloc(sizeof(*qca_memdump), GFP_ATOMIC);
 			if (!qca_memdump) {
 				mutex_unlock(&qca->hci_memdump_lock);
 				return;
diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c
index 28750a40f0ed..c4046f8f1985 100644
--- a/drivers/bluetooth/hci_vhci.c
+++ b/drivers/bluetooth/hci_vhci.c
@@ -633,7 +633,7 @@ static int vhci_open(struct inode *inode, struct file *file)
 {
 	struct vhci_data *data;
 
-	data = kzalloc(sizeof(struct vhci_data), GFP_KERNEL);
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
 	if (!data)
 		return -ENOMEM;
 
-- 
cgit v1.2.3-58-ga151


From 67c3bceabb42780f18362d7c5ec72a700051ca36 Mon Sep 17 00:00:00 2001
From: Thorsten Blum <thorsten.blum@toblux.com>
Date: Sat, 1 Jun 2024 01:51:33 +0200
Subject: Bluetooth: btintel_pcie: Remove unnecessary memset(0) calls

Remove memset(0) after dma_alloc_coherent(), which already zeroes out
the memory, and fix the following two Coccinelle/coccicheck warnings
reported by zalloc-simple.cocci:

btintel_pcie.c:837:19-37: WARNING: dma_alloc_coherent used in

	/* Allocate full chunk of data buffer for DMA first and do indexing and
	 * initialization next, so it can be freed easily
	 */
	rxq->buf_v_addr   already zeroes out memory, so memset is not needed

btintel_pcie.c:792:19-37: WARNING: dma_alloc_coherent used in

	/* Allocate full chunk of data buffer for DMA first and do indexing and
	 * initialization next, so it can be freed easily
	 */
	txq->buf_v_addr   already zeroes out memory, so memset is not needed

Fixes: c2b636b3f788 ("Bluetooth: btintel_pcie: Add support for PCIe transport")
Signed-off-by: Thorsten Blum <thorsten.blum@toblux.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btintel_pcie.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/bluetooth/btintel_pcie.c b/drivers/bluetooth/btintel_pcie.c
index 58144c82b1cb..0d1a0415557b 100644
--- a/drivers/bluetooth/btintel_pcie.c
+++ b/drivers/bluetooth/btintel_pcie.c
@@ -797,7 +797,6 @@ static int btintel_pcie_setup_txq_bufs(struct btintel_pcie_data *data,
 		kfree(txq->bufs);
 		return -ENOMEM;
 	}
-	memset(txq->buf_v_addr, 0, txq->count * BTINTEL_PCIE_BUFFER_SIZE);
 
 	/* Setup the allocated DMA buffer to bufs. Each data_buf should
 	 * have virtual address and physical address
@@ -842,7 +841,6 @@ static int btintel_pcie_setup_rxq_bufs(struct btintel_pcie_data *data,
 		kfree(rxq->bufs);
 		return -ENOMEM;
 	}
-	memset(rxq->buf_v_addr, 0, rxq->count * BTINTEL_PCIE_BUFFER_SIZE);
 
 	/* Setup the allocated DMA buffer to bufs. Each data_buf should
 	 * have virtual address and physical address
-- 
cgit v1.2.3-58-ga151


From d4cc4ee48167b4c25abf49ead19412c29a1ad9f4 Mon Sep 17 00:00:00 2001
From: "Dr. David Alan Gilbert" <linux@treblig.org>
Date: Sat, 1 Jun 2024 00:26:14 +0100
Subject: Bluetooth: iso: remove unused struct 'iso_list_data'

'iso_list_data' has been unused since the original
commit ccf74f2390d6 ("Bluetooth: Add BTPROTO_ISO socket type").

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/iso.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c
index 398fb81f7a13..d5e00d0dd1a0 100644
--- a/net/bluetooth/iso.c
+++ b/net/bluetooth/iso.c
@@ -1720,11 +1720,6 @@ static void iso_sock_ready(struct sock *sk)
 	release_sock(sk);
 }
 
-struct iso_list_data {
-	struct hci_conn *hcon;
-	int count;
-};
-
 static bool iso_match_big(struct sock *sk, void *data)
 {
 	struct hci_evt_le_big_sync_estabilished *ev = data;
-- 
cgit v1.2.3-58-ga151


From f25b7fd36cc3a850e006aed686f5bbecd200de1b Mon Sep 17 00:00:00 2001
From: Ying Hsu <yinghsu@chromium.org>
Date: Wed, 29 May 2024 08:00:00 +0000
Subject: Bluetooth: Add vendor-specific packet classification for ISO data

When HCI raw sockets are opened, the Bluetooth kernel module doesn't
track CIS/BIS connections. User-space applications have to identify
ISO data by maintaining connection information and look up the mapping
for each ACL data packet received. Besides, btsnoop log captured in
kernel couldn't tell ISO data from ACL data in this case.

To avoid additional lookups, this patch introduces vendor-specific
packet classification for Intel BT controllers to distinguish
ISO data packets from ACL data packets.

Signed-off-by: Ying Hsu <yinghsu@chromium.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btintel.c      | 25 +++++++++++++++++++++++--
 include/net/bluetooth/hci_core.h |  1 +
 net/bluetooth/hci_core.c         | 16 ++++++++++++++++
 3 files changed, 40 insertions(+), 2 deletions(-)

diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c
index 27e03951e68b..ff33e1aa2929 100644
--- a/drivers/bluetooth/btintel.c
+++ b/drivers/bluetooth/btintel.c
@@ -2549,6 +2549,24 @@ static void btintel_set_dsm_reset_method(struct hci_dev *hdev,
 	data->acpi_reset_method = btintel_acpi_reset_method;
 }
 
+#define BTINTEL_ISODATA_HANDLE_BASE 0x900
+
+static u8 btintel_classify_pkt_type(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	/*
+	 * Distinguish ISO data packets form ACL data packets
+	 * based on their connection handle value range.
+	 */
+	if (hci_skb_pkt_type(skb) == HCI_ACLDATA_PKT) {
+		__u16 handle = __le16_to_cpu(hci_acl_hdr(skb)->handle);
+
+		if (hci_handle(handle) >= BTINTEL_ISODATA_HANDLE_BASE)
+			return HCI_ISODATA_PKT;
+	}
+
+	return hci_skb_pkt_type(skb);
+}
+
 int btintel_bootloader_setup_tlv(struct hci_dev *hdev,
 				 struct intel_version_tlv *ver)
 {
@@ -2989,11 +3007,14 @@ static int btintel_setup_combined(struct hci_dev *hdev)
 		err = btintel_bootloader_setup(hdev, &ver);
 		btintel_register_devcoredump_support(hdev);
 		break;
+	case 0x18: /* GfP2 */
+	case 0x1c: /* GaP */
+		/* Re-classify packet type for controllers with LE audio */
+		hdev->classify_pkt_type = btintel_classify_pkt_type;
+		fallthrough;
 	case 0x17:
-	case 0x18:
 	case 0x19:
 	case 0x1b:
-	case 0x1c:
 	case 0x1e:
 		/* Display version information of TLV type */
 		btintel_version_info_tlv(hdev, &ver_tlv);
diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index c43716edf205..f7de2681d457 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -649,6 +649,7 @@ struct hci_dev {
 	int (*get_codec_config_data)(struct hci_dev *hdev, __u8 type,
 				     struct bt_codec *codec, __u8 *vnd_len,
 				     __u8 **vnd_data);
+	u8 (*classify_pkt_type)(struct hci_dev *hdev, struct sk_buff *skb);
 };
 
 #define HCI_PHY_HANDLE(handle)	(handle & 0xff)
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 55bdc365916f..144e85ebfbdb 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -2909,15 +2909,31 @@ int hci_reset_dev(struct hci_dev *hdev)
 }
 EXPORT_SYMBOL(hci_reset_dev);
 
+static u8 hci_dev_classify_pkt_type(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	if (hdev->classify_pkt_type)
+		return hdev->classify_pkt_type(hdev, skb);
+
+	return hci_skb_pkt_type(skb);
+}
+
 /* Receive frame from HCI drivers */
 int hci_recv_frame(struct hci_dev *hdev, struct sk_buff *skb)
 {
+	u8 dev_pkt_type;
+
 	if (!hdev || (!test_bit(HCI_UP, &hdev->flags)
 		      && !test_bit(HCI_INIT, &hdev->flags))) {
 		kfree_skb(skb);
 		return -ENXIO;
 	}
 
+	/* Check if the driver agree with packet type classification */
+	dev_pkt_type = hci_dev_classify_pkt_type(hdev, skb);
+	if (hci_skb_pkt_type(skb) != dev_pkt_type) {
+		hci_skb_pkt_type(skb) = dev_pkt_type;
+	}
+
 	switch (hci_skb_pkt_type(skb)) {
 	case HCI_EVENT_PKT:
 		break;
-- 
cgit v1.2.3-58-ga151


From 08c22b52262041164990eef4b0e076bce173a3b2 Mon Sep 17 00:00:00 2001
From: Hector Martin <marcan@marcan.st>
Date: Sun, 2 Jun 2024 09:57:05 +0000
Subject: Bluetooth: hci_bcm4377: Add BCM4388 support

This new variant needs a different core2_window1 and always uses
beamforming.
The BAR2 also has an offset (RAM start, presumably), so add that.

Signed-off-by: Hector Martin <marcan@marcan.st>
Reviewed-by: Sven Peter <sven@svenpeter.dev>
[sven: rebased, updated some comments, mentioned 4388 in Kconfig]
Signed-off-by: Sven Peter <sven@svenpeter.dev>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/Kconfig       |  6 ++--
 drivers/bluetooth/hci_bcm4377.c | 61 +++++++++++++++++++++++++++++++----------
 2 files changed, 50 insertions(+), 17 deletions(-)

diff --git a/drivers/bluetooth/Kconfig b/drivers/bluetooth/Kconfig
index 0b5f218ac505..87484f5de8e3 100644
--- a/drivers/bluetooth/Kconfig
+++ b/drivers/bluetooth/Kconfig
@@ -287,12 +287,12 @@ config BT_HCIBCM203X
 
 
 config BT_HCIBCM4377
-	tristate "HCI BCM4377/4378/4387 PCIe driver"
+	tristate "HCI BCM4377/4378/4387/4388 PCIe driver"
 	depends on PCI
 	select FW_LOADER
 	help
-	  Support for Broadcom BCM4377/4378/4387 Bluetooth chipsets attached via
-	  PCIe. These are usually found in Apple machines.
+	  Support for Broadcom BCM4377/4378/4387/4388 Bluetooth chipsets
+	  attached via PCIe. These are usually found in Apple machines.
 
 	  Say Y here to compile support for HCI BCM4377 family devices into the
 	  kernel or say M to compile it as module (hci_bcm4377).
diff --git a/drivers/bluetooth/hci_bcm4377.c b/drivers/bluetooth/hci_bcm4377.c
index e6644cb0163c..77a5454a8721 100644
--- a/drivers/bluetooth/hci_bcm4377.c
+++ b/drivers/bluetooth/hci_bcm4377.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0-only OR MIT
 /*
- * Bluetooth HCI driver for Broadcom 4377/4378/4387 devices attached via PCIe
+ * Bluetooth HCI driver for Broadcom 4377/4378/4387/4388 devices attached via PCIe
  *
  * Copyright (C) The Asahi Linux Contributors
  */
@@ -26,11 +26,13 @@ enum bcm4377_chip {
 	BCM4377 = 0,
 	BCM4378,
 	BCM4387,
+	BCM4388,
 };
 
 #define BCM4377_DEVICE_ID 0x5fa0
 #define BCM4378_DEVICE_ID 0x5f69
 #define BCM4387_DEVICE_ID 0x5f71
+#define BCM4388_DEVICE_ID 0x5f72
 
 #define BCM4377_TIMEOUT msecs_to_jiffies(1000)
 #define BCM4377_BOOT_TIMEOUT msecs_to_jiffies(5000)
@@ -488,6 +490,7 @@ struct bcm4377_data;
  *                     second window in BAR0
  * has_bar0_core2_window2: Set to true if this chip requires the second core's
  *                         second window to be configured
+ * bar2_offset: Offset to the start of the variables in BAR2
  * clear_pciecfg_subsystem_ctrl_bit19: Set to true if bit 19 in the
  *                                     vendor-specific subsystem control
  *                                     register has to be cleared
@@ -511,6 +514,7 @@ struct bcm4377_hw {
 	u32 bar0_window1;
 	u32 bar0_window2;
 	u32 bar0_core2_window2;
+	u32 bar2_offset;
 
 	unsigned long has_bar0_core2_window2 : 1;
 	unsigned long clear_pciecfg_subsystem_ctrl_bit19 : 1;
@@ -836,8 +840,8 @@ static irqreturn_t bcm4377_irq(int irq, void *data)
 	struct bcm4377_data *bcm4377 = data;
 	u32 bootstage, rti_status;
 
-	bootstage = ioread32(bcm4377->bar2 + BCM4377_BAR2_BOOTSTAGE);
-	rti_status = ioread32(bcm4377->bar2 + BCM4377_BAR2_RTI_STATUS);
+	bootstage = ioread32(bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_BOOTSTAGE);
+	rti_status = ioread32(bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_RTI_STATUS);
 
 	if (bootstage != bcm4377->bootstage ||
 	    rti_status != bcm4377->rti_status) {
@@ -1197,6 +1201,14 @@ static int bcm4387_send_calibration(struct bcm4377_data *bcm4377)
 						  bcm4377->taurus_cal_size);
 }
 
+static int bcm4388_send_calibration(struct bcm4377_data *bcm4377)
+{
+	/* BCM4388 always uses beamforming */
+	return __bcm4378_send_calibration(
+		bcm4377, bcm4377->taurus_beamforming_cal_blob,
+		bcm4377->taurus_beamforming_cal_size);
+}
+
 static const struct firmware *bcm4377_request_blob(struct bcm4377_data *bcm4377,
 						   const char *suffix)
 {
@@ -1820,8 +1832,8 @@ static int bcm4377_boot(struct bcm4377_data *bcm4377)
 	int ret = 0;
 	u32 bootstage, rti_status;
 
-	bootstage = ioread32(bcm4377->bar2 + BCM4377_BAR2_BOOTSTAGE);
-	rti_status = ioread32(bcm4377->bar2 + BCM4377_BAR2_RTI_STATUS);
+	bootstage = ioread32(bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_BOOTSTAGE);
+	rti_status = ioread32(bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_RTI_STATUS);
 
 	if (bootstage != 0) {
 		dev_err(&bcm4377->pdev->dev, "bootstage is %d and not 0\n",
@@ -1855,9 +1867,12 @@ static int bcm4377_boot(struct bcm4377_data *bcm4377)
 	iowrite32(BCM4377_DMA_MASK,
 		  bcm4377->bar0 + BCM4377_BAR0_HOST_WINDOW_SIZE);
 
-	iowrite32(lower_32_bits(fw_dma), bcm4377->bar2 + BCM4377_BAR2_FW_LO);
-	iowrite32(upper_32_bits(fw_dma), bcm4377->bar2 + BCM4377_BAR2_FW_HI);
-	iowrite32(fw->size, bcm4377->bar2 + BCM4377_BAR2_FW_SIZE);
+	iowrite32(lower_32_bits(fw_dma),
+		  bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_FW_LO);
+	iowrite32(upper_32_bits(fw_dma),
+		  bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_FW_HI);
+	iowrite32(fw->size,
+		  bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_FW_SIZE);
 	iowrite32(0, bcm4377->bar0 + BCM4377_BAR0_FW_DOORBELL);
 
 	dev_dbg(&bcm4377->pdev->dev, "waiting for firmware to boot\n");
@@ -1914,16 +1929,16 @@ static int bcm4377_setup_rti(struct bcm4377_data *bcm4377)
 	dev_dbg(&bcm4377->pdev->dev, "RTI is in state 1\n");
 
 	/* allow access to the entire IOVA space again */
-	iowrite32(0, bcm4377->bar2 + BCM4377_BAR2_RTI_WINDOW_LO);
-	iowrite32(0, bcm4377->bar2 + BCM4377_BAR2_RTI_WINDOW_HI);
+	iowrite32(0, bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_RTI_WINDOW_LO);
+	iowrite32(0, bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_RTI_WINDOW_HI);
 	iowrite32(BCM4377_DMA_MASK,
-		  bcm4377->bar2 + BCM4377_BAR2_RTI_WINDOW_SIZE);
+		  bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_RTI_WINDOW_SIZE);
 
 	/* setup "Converged IPC" context */
 	iowrite32(lower_32_bits(bcm4377->ctx_dma),
-		  bcm4377->bar2 + BCM4377_BAR2_CONTEXT_ADDR_LO);
+		  bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_CONTEXT_ADDR_LO);
 	iowrite32(upper_32_bits(bcm4377->ctx_dma),
-		  bcm4377->bar2 + BCM4377_BAR2_CONTEXT_ADDR_HI);
+		  bcm4377->bar2 + bcm4377->hw->bar2_offset + BCM4377_BAR2_CONTEXT_ADDR_HI);
 	iowrite32(2, bcm4377->bar0 + BCM4377_BAR0_RTI_CONTROL);
 
 	ret = wait_for_completion_interruptible_timeout(&bcm4377->event,
@@ -2489,6 +2504,21 @@ static const struct bcm4377_hw bcm4377_hw_variants[] = {
 		.send_calibration = bcm4387_send_calibration,
 		.send_ptb = bcm4378_send_ptb,
 	},
+
+	[BCM4388] = {
+		.id = 0x4388,
+		.otp_offset = 0x415c,
+		.bar2_offset = 0x200000,
+		.bar0_window1 = 0x18002000,
+		.bar0_window2 = 0x18109000,
+		.bar0_core2_window2 = 0x18106000,
+		.has_bar0_core2_window2 = true,
+		.broken_mws_transport_config = true,
+		.broken_le_coded = true,
+		.broken_le_ext_adv_report_phy = true,
+		.send_calibration = bcm4388_send_calibration,
+		.send_ptb = bcm4378_send_ptb,
+	},
 };
 
 #define BCM4377_DEVID_ENTRY(id)                                             \
@@ -2502,6 +2532,7 @@ static const struct pci_device_id bcm4377_devid_table[] = {
 	BCM4377_DEVID_ENTRY(4377),
 	BCM4377_DEVID_ENTRY(4378),
 	BCM4377_DEVID_ENTRY(4387),
+	BCM4377_DEVID_ENTRY(4388),
 	{},
 };
 MODULE_DEVICE_TABLE(pci, bcm4377_devid_table);
@@ -2516,7 +2547,7 @@ static struct pci_driver bcm4377_pci_driver = {
 module_pci_driver(bcm4377_pci_driver);
 
 MODULE_AUTHOR("Sven Peter <sven@svenpeter.dev>");
-MODULE_DESCRIPTION("Bluetooth support for Broadcom 4377/4378/4387 devices");
+MODULE_DESCRIPTION("Bluetooth support for Broadcom 4377/4378/4387/4388 devices");
 MODULE_LICENSE("Dual MIT/GPL");
 MODULE_FIRMWARE("brcm/brcmbt4377*.bin");
 MODULE_FIRMWARE("brcm/brcmbt4377*.ptb");
@@ -2524,3 +2555,5 @@ MODULE_FIRMWARE("brcm/brcmbt4378*.bin");
 MODULE_FIRMWARE("brcm/brcmbt4378*.ptb");
 MODULE_FIRMWARE("brcm/brcmbt4387*.bin");
 MODULE_FIRMWARE("brcm/brcmbt4387*.ptb");
+MODULE_FIRMWARE("brcm/brcmbt4388*.bin");
+MODULE_FIRMWARE("brcm/brcmbt4388*.ptb");
-- 
cgit v1.2.3-58-ga151


From 921825453873a15e8c83d6c171f0a6cf128fc963 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Fri, 14 Jun 2024 14:19:40 +0530
Subject: dt-bindings: net: bluetooth: nxp: Add firmware-name property

This adds a new optional device tree property called firware-name.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 Documentation/devicetree/bindings/net/bluetooth/nxp,88w8987-bt.yaml | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/bluetooth/nxp,88w8987-bt.yaml b/Documentation/devicetree/bindings/net/bluetooth/nxp,88w8987-bt.yaml
index f01a3988538c..37a65badb448 100644
--- a/Documentation/devicetree/bindings/net/bluetooth/nxp,88w8987-bt.yaml
+++ b/Documentation/devicetree/bindings/net/bluetooth/nxp,88w8987-bt.yaml
@@ -31,6 +31,9 @@ properties:
       This property depends on the module vendor's
       configuration.
 
+  firmware-name:
+    maxItems: 1
+
 required:
   - compatible
 
@@ -42,5 +45,6 @@ examples:
         bluetooth {
             compatible = "nxp,88w8987-bt";
             fw-init-baudrate = <3000000>;
+            firmware-name = "uartuart8987_bt_v0.bin";
         };
     };
-- 
cgit v1.2.3-58-ga151


From cb24bb7e6f8ae569c699f569ac4b5093606dca49 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Fri, 14 Jun 2024 14:19:41 +0530
Subject: Bluetooth: btnxpuart: Update firmware names

This updates the firmware names of 3 chipsets: w8987, w8997, w9098.
These changes are been done to standardize chip specific firmware
file names to be in sync with firmware names of newer chipsets.

The naming convention for BT-only files would be as follows:
For dual-radio WiFi+BT chipsets:
- <protocol-BT><chip-name>_bt_v<HW-version>.bin
For tri-radio WiFi+BT+15.4 chipsets:
- <protocol-BT><protocol-15.4><chip-name>_bt_v<HW-version>.bin

To maintain backward compatibility, this commit adds a provision to
request older firmware file name, if new firmware file name not found in
/lib/firmware/nxp/.

A new device tree property has been introduced called firmware-name, to
override the hardcoded firmware names (old and new) in the driver.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 67 ++++++++++++++++++++++++++++++++-----------
 1 file changed, 50 insertions(+), 17 deletions(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index 50c1578ea3ed..8c5599ddf0f5 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -34,16 +34,19 @@
 /* NXP HW err codes */
 #define BTNXPUART_IR_HW_ERR		0xb0
 
-#define FIRMWARE_W8987		"nxp/uartuart8987_bt.bin"
-#define FIRMWARE_W8997		"nxp/uartuart8997_bt_v4.bin"
-#define FIRMWARE_W9098		"nxp/uartuart9098_bt_v1.bin"
-#define FIRMWARE_IW416		"nxp/uartiw416_bt_v0.bin"
-#define FIRMWARE_IW612		"nxp/uartspi_n61x_v1.bin.se"
-#define FIRMWARE_IW624		"nxp/uartiw624_bt.bin"
-#define FIRMWARE_SECURE_IW624	"nxp/uartiw624_bt.bin.se"
-#define FIRMWARE_AW693		"nxp/uartaw693_bt.bin"
-#define FIRMWARE_SECURE_AW693	"nxp/uartaw693_bt.bin.se"
-#define FIRMWARE_HELPER		"nxp/helper_uart_3000000.bin"
+#define FIRMWARE_W8987		"uart8987_bt_v0.bin"
+#define FIRMWARE_W8987_OLD	"uartuart8987_bt.bin"
+#define FIRMWARE_W8997		"uart8997_bt_v4.bin"
+#define FIRMWARE_W8997_OLD	"uartuart8997_bt_v4.bin"
+#define FIRMWARE_W9098		"uart9098_bt_v1.bin"
+#define FIRMWARE_W9098_OLD	"uartuart9098_bt_v1.bin"
+#define FIRMWARE_IW416		"uartiw416_bt_v0.bin"
+#define FIRMWARE_IW612		"uartspi_n61x_v1.bin.se"
+#define FIRMWARE_IW624		"uartiw624_bt.bin"
+#define FIRMWARE_SECURE_IW624	"uartiw624_bt.bin.se"
+#define FIRMWARE_AW693		"uartaw693_bt.bin"
+#define FIRMWARE_SECURE_AW693	"uartaw693_bt.bin.se"
+#define FIRMWARE_HELPER		"helper_uart_3000000.bin"
 
 #define CHIP_ID_W9098		0x5c03
 #define CHIP_ID_IW416		0x7201
@@ -145,6 +148,7 @@ struct psmode_cmd_payload {
 struct btnxpuart_data {
 	const char *helper_fw_name;
 	const char *fw_name;
+	const char *fw_name_old;
 };
 
 struct btnxpuart_dev {
@@ -694,19 +698,30 @@ static bool process_boot_signature(struct btnxpuart_dev *nxpdev)
 	return is_fw_downloading(nxpdev);
 }
 
-static int nxp_request_firmware(struct hci_dev *hdev, const char *fw_name)
+static int nxp_request_firmware(struct hci_dev *hdev, const char *fw_name,
+				const char *fw_name_old)
 {
 	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	const char *fw_name_dt;
 	int err = 0;
 
 	if (!fw_name)
 		return -ENOENT;
 
 	if (!strlen(nxpdev->fw_name)) {
-		snprintf(nxpdev->fw_name, MAX_FW_FILE_NAME_LEN, "%s", fw_name);
+		if (strcmp(fw_name, FIRMWARE_HELPER) &&
+		    !device_property_read_string(&nxpdev->serdev->dev,
+						 "firmware-name",
+						 &fw_name_dt))
+			fw_name = fw_name_dt;
+		snprintf(nxpdev->fw_name, MAX_FW_FILE_NAME_LEN, "nxp/%s", fw_name);
+		err = request_firmware_direct(&nxpdev->fw, nxpdev->fw_name, &hdev->dev);
+		if (err < 0 && fw_name_old) {
+			snprintf(nxpdev->fw_name, MAX_FW_FILE_NAME_LEN, "nxp/%s", fw_name_old);
+			err = request_firmware_direct(&nxpdev->fw, nxpdev->fw_name, &hdev->dev);
+		}
 
 		bt_dev_info(hdev, "Request Firmware: %s", nxpdev->fw_name);
-		err = request_firmware(&nxpdev->fw, nxpdev->fw_name, &hdev->dev);
 		if (err < 0) {
 			bt_dev_err(hdev, "Firmware file %s not found", nxpdev->fw_name);
 			clear_bit(BTNXPUART_FW_DOWNLOADING, &nxpdev->tx_state);
@@ -785,10 +800,10 @@ static int nxp_recv_fw_req_v1(struct hci_dev *hdev, struct sk_buff *skb)
 	}
 
 	if (!nxp_data->helper_fw_name || nxpdev->helper_downloaded) {
-		if (nxp_request_firmware(hdev, nxp_data->fw_name))
+		if (nxp_request_firmware(hdev, nxp_data->fw_name, nxp_data->fw_name_old))
 			goto free_skb;
 	} else if (nxp_data->helper_fw_name && !nxpdev->helper_downloaded) {
-		if (nxp_request_firmware(hdev, nxp_data->helper_fw_name))
+		if (nxp_request_firmware(hdev, nxp_data->helper_fw_name, NULL))
 			goto free_skb;
 	}
 
@@ -890,10 +905,25 @@ static char *nxp_get_fw_name_from_chipid(struct hci_dev *hdev, u16 chipid,
 	return fw_name;
 }
 
+static char *nxp_get_old_fw_name_from_chipid(struct hci_dev *hdev, u16 chipid,
+					 u8 loader_ver)
+{
+	char *fw_name_old = NULL;
+
+	switch (chipid) {
+	case CHIP_ID_W9098:
+		fw_name_old = FIRMWARE_W9098_OLD;
+		break;
+	}
+	return fw_name_old;
+}
+
 static int nxp_recv_chip_ver_v3(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct v3_start_ind *req = skb_pull_data(skb, sizeof(*req));
 	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	const char *fw_name;
+	const char *fw_name_old;
 	u16 chip_id;
 	u8 loader_ver;
 
@@ -903,8 +933,9 @@ static int nxp_recv_chip_ver_v3(struct hci_dev *hdev, struct sk_buff *skb)
 	chip_id = le16_to_cpu(req->chip_id);
 	loader_ver = req->loader_ver;
 	bt_dev_info(hdev, "ChipID: %04x, Version: %d", chip_id, loader_ver);
-	if (!nxp_request_firmware(hdev, nxp_get_fw_name_from_chipid(hdev,
-								    chip_id, loader_ver)))
+	fw_name = nxp_get_fw_name_from_chipid(hdev, chip_id, loader_ver);
+	fw_name_old = nxp_get_old_fw_name_from_chipid(hdev, chip_id, loader_ver);
+	if (!nxp_request_firmware(hdev, fw_name, fw_name_old))
 		nxp_send_ack(NXP_ACK_V3, hdev);
 
 free_skb:
@@ -1426,11 +1457,13 @@ static void nxp_serdev_remove(struct serdev_device *serdev)
 static struct btnxpuart_data w8987_data __maybe_unused = {
 	.helper_fw_name = NULL,
 	.fw_name = FIRMWARE_W8987,
+	.fw_name_old = FIRMWARE_W8987_OLD,
 };
 
 static struct btnxpuart_data w8997_data __maybe_unused = {
 	.helper_fw_name = FIRMWARE_HELPER,
 	.fw_name = FIRMWARE_W8997,
+	.fw_name_old = FIRMWARE_W8997_OLD,
 };
 
 static const struct of_device_id nxpuart_of_match_table[] __maybe_unused = {
-- 
cgit v1.2.3-58-ga151


From 27489364299a2ddb5c54cd9f29a3f41bd8d151ab Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Fri, 14 Jun 2024 13:53:38 +0530
Subject: Bluetooth: btnxpuart: Add handling for boot-signature timeout errors

This handles the timeout error codes sent by the chip as part of the
bootloader signatures during firmware download process.

When the bootloader does not receive a response packet from the host
within a specific time, it adds an error code to the bootloader
signature while requesting for the FW chunk from the same offset.

The host is expected to clear this error code with a NAK, and reply to
only those bootloader signatures which have error code 0.

However, the driver was ignoring this error code and replying with the
firmware chunks instead, which is apparently ignored by the chip and the
chip resends the same bootloader signature with the error code again. This
happens in a loop until the error code self clears and firmware download
proceeds ahead, adding a couple of milliseconds to the total firmware
download time.

Commit 689ca16e5232 was an initial implementation which simply printed
the following line during driver debug:
- FW Download received err 0x04 from chip

This commit adds the expected handling to the error codes.

This error handling is valid for data_req bootloader signatures for V3
and future bootloader versions.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Fixes: 689ca16e5232 ("Bluetooth: NXP: Add protocol support for NXP Bluetooth chipsets")
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 52 +++++++++++++++++++++++++++++++++++++++----
 1 file changed, 48 insertions(+), 4 deletions(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index 8c5599ddf0f5..da18fd4f54f3 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -193,6 +193,11 @@ struct btnxpuart_dev {
 #define NXP_NAK_V3		0x7b
 #define NXP_CRC_ERROR_V3	0x7c
 
+/* Bootloader signature error codes */
+#define NXP_ACK_RX_TIMEOUT	0x0002	/* ACK not received from host */
+#define NXP_HDR_RX_TIMEOUT	0x0003	/* FW Header chunk not received */
+#define NXP_DATA_RX_TIMEOUT	0x0004	/* FW Data chunk not received */
+
 #define HDR_LEN			16
 
 #define NXP_RECV_CHIP_VER_V1 \
@@ -283,6 +288,17 @@ struct nxp_bootloader_cmd {
 	__be32 crc;
 } __packed;
 
+struct nxp_v3_rx_timeout_nak {
+	u8 nak;
+	__le32 offset;
+	u8 crc;
+} __packed;
+
+union nxp_v3_rx_timeout_nak_u {
+	struct nxp_v3_rx_timeout_nak pkt;
+	u8 buf[6];
+};
+
 static u8 crc8_table[CRC8_TABLE_SIZE];
 
 /* Default configurations */
@@ -943,6 +959,32 @@ free_skb:
 	return 0;
 }
 
+static void nxp_handle_fw_download_error(struct hci_dev *hdev, struct v3_data_req *req)
+{
+	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
+	__u32 offset = __le32_to_cpu(req->offset);
+	__u16 err = __le16_to_cpu(req->error);
+	union nxp_v3_rx_timeout_nak_u nak_tx_buf;
+
+	switch (err) {
+	case NXP_ACK_RX_TIMEOUT:
+	case NXP_HDR_RX_TIMEOUT:
+	case NXP_DATA_RX_TIMEOUT:
+		nak_tx_buf.pkt.nak = NXP_NAK_V3;
+		nak_tx_buf.pkt.offset = __cpu_to_le32(offset);
+		nak_tx_buf.pkt.crc = crc8(crc8_table, nak_tx_buf.buf,
+				      sizeof(nak_tx_buf) - 1, 0xff);
+		serdev_device_write_buf(nxpdev->serdev, nak_tx_buf.buf,
+					sizeof(nak_tx_buf));
+		break;
+	default:
+		bt_dev_dbg(hdev, "Unknown bootloader error code: %d", err);
+		break;
+
+	}
+
+}
+
 static int nxp_recv_fw_req_v3(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct btnxpuart_dev *nxpdev = hci_get_drvdata(hdev);
@@ -957,7 +999,12 @@ static int nxp_recv_fw_req_v3(struct hci_dev *hdev, struct sk_buff *skb)
 	if (!req || !nxpdev->fw)
 		goto free_skb;
 
-	nxp_send_ack(NXP_ACK_V3, hdev);
+	if (!req->error) {
+		nxp_send_ack(NXP_ACK_V3, hdev);
+	} else {
+		nxp_handle_fw_download_error(hdev, req);
+		goto free_skb;
+	}
 
 	len = __le16_to_cpu(req->len);
 
@@ -984,9 +1031,6 @@ static int nxp_recv_fw_req_v3(struct hci_dev *hdev, struct sk_buff *skb)
 		wake_up_interruptible(&nxpdev->fw_dnld_done_wait_q);
 		goto free_skb;
 	}
-	if (req->error)
-		bt_dev_dbg(hdev, "FW Download received err 0x%02x from chip",
-			   req->error);
 
 	offset = __le32_to_cpu(req->offset);
 	if (offset < nxpdev->fw_v3_offset_correction) {
-- 
cgit v1.2.3-58-ga151


From 5a820b13db3988122080f8de2920721f770c37a0 Mon Sep 17 00:00:00 2001
From: Iulia Tanasescu <iulia.tanasescu@nxp.com>
Date: Mon, 17 Jun 2024 11:52:06 +0300
Subject: Bluetooth: hci_event: Set QoS encryption from BIGInfo report

On a Broadcast Sink, after synchronizing to the PA transimitted by a
Broadcast Source, the BIGInfo advertising reports emitted by the
Controller hold the encryption field, which indicates whether the
Broadcast Source is transmitting encrypted streams.

This updates the PA sync hcon QoS with the encryption value reported
in the BIGInfo report, so that this information is accurate if the
userspace tries to access the QoS struct via getsockopt.

Fixes: 1d11d70d1f6b ("Bluetooth: ISO: Pass BIG encryption info through QoS")
Signed-off-by: Iulia Tanasescu <iulia.tanasescu@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_event.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index 93f7ac905cec..4611a67d7dcc 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -6988,6 +6988,8 @@ static void hci_le_big_info_adv_report_evt(struct hci_dev *hdev, void *data,
 	if (!pa_sync)
 		goto unlock;
 
+	pa_sync->iso_qos.bcast.encryption = ev->encryption;
+
 	/* Notify iso layer */
 	hci_connect_cfm(pa_sync, 0);
 
-- 
cgit v1.2.3-58-ga151


From 295ef07a9dae6182ad4b689aa8c6a7dbba21474c Mon Sep 17 00:00:00 2001
From: Hilda Wu <hildawu@realtek.com>
Date: Mon, 17 Jun 2024 17:05:18 +0800
Subject: Bluetooth: btusb: Add RTL8852BE device 0489:e125 to device tables

Add the support ID 0489:e125 to usb_device_id table for
Realtek RTL8852B chip.

The device info from /sys/kernel/debug/usb/devices as below.

T:  Bus=01 Lev=01 Prnt=01 Port=07 Cnt=03 Dev#=  5 Spd=12   MxCh= 0
D:  Ver= 1.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=0489 ProdID=e125 Rev= 0.00
S:  Manufacturer=Realtek
S:  Product=Bluetooth Radio
S:  SerialNumber=00e04c000001
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=500mA
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  16 Ivl=1ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms

Signed-off-by: Hilda Wu <hildawu@realtek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 79aefdb3324d..2d7d47f9d007 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -555,6 +555,8 @@ static const struct usb_device_id quirks_table[] = {
 						     BTUSB_WIDEBAND_SPEECH },
 	{ USB_DEVICE(0x13d3, 0x3572), .driver_info = BTUSB_REALTEK |
 						     BTUSB_WIDEBAND_SPEECH },
+	{ USB_DEVICE(0x0489, 0xe125), .driver_info = BTUSB_REALTEK |
+						     BTUSB_WIDEBAND_SPEECH },
 
 	/* Realtek 8852BT/8852BE-VT Bluetooth devices */
 	{ USB_DEVICE(0x0bda, 0x8520), .driver_info = BTUSB_REALTEK |
-- 
cgit v1.2.3-58-ga151


From da63f331353c9e1e6dc29e49e28f8f4fe5d642fd Mon Sep 17 00:00:00 2001
From: Dmitry Antipov <dmantipov@yandex.ru>
Date: Tue, 18 Jun 2024 21:59:32 +0300
Subject: Bluetooth: hci_core, hci_sync: cleanup struct discovery_state

After commit 78db544b5d27 ("Bluetooth: hci_core: Remove le_restart_scan
work"), 'scan_start' and 'scan_duration' of 'struct discovery_state'
are still initialized but actually unused. So remove the aforementioned
fields and adjust 'hci_discovery_filter_clear()' and 'le_scan_disable()'
accordingly. Compile tested only.

Fixes: 78db544b5d27 ("Bluetooth: hci_core: Remove le_restart_scan work")
Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_core.h | 4 ----
 net/bluetooth/hci_sync.c         | 2 --
 2 files changed, 6 deletions(-)

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index f7de2681d457..eaeaf3dc07aa 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -91,8 +91,6 @@ struct discovery_state {
 	s8			rssi;
 	u16			uuid_count;
 	u8			(*uuids)[16];
-	unsigned long		scan_start;
-	unsigned long		scan_duration;
 	unsigned long		name_resolve_timeout;
 };
 
@@ -891,8 +889,6 @@ static inline void hci_discovery_filter_clear(struct hci_dev *hdev)
 	hdev->discovery.uuid_count = 0;
 	kfree(hdev->discovery.uuids);
 	hdev->discovery.uuids = NULL;
-	hdev->discovery.scan_start = 0;
-	hdev->discovery.scan_duration = 0;
 }
 
 bool hci_discovery_active(struct hci_dev *hdev);
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 82db6092965b..ccad43f10415 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -371,8 +371,6 @@ static void le_scan_disable(struct work_struct *work)
 		goto _return;
 	}
 
-	hdev->discovery.scan_start = 0;
-
 	/* If we were running LE only scan, change discovery state. If
 	 * we were running both LE and BR/EDR inquiry simultaneously,
 	 * and BR/EDR inquiry is already finished, stop discovery,
-- 
cgit v1.2.3-58-ga151


From 164c62f958f8c7f0bde1e9a5a8677971c6f28205 Mon Sep 17 00:00:00 2001
From: Kiran K <kiran.k@intel.com>
Date: Wed, 19 Jun 2024 15:39:33 +0530
Subject: Bluetooth: btintel: Add firmware ID to firmware name

From BlazarI onwards, driver shall append the firmware ID (usually
represents transport type) while constructing the firmware name.
Firmware ID is returned on Intel Read Version command.

The new firmware file name for operational image and ddc file shall be,
ibt-<cnvi_top type+cnvi_top step>-<cnvr_top type+cnvr_top step-fw_id>.[sfi|ddc]

dmesg snippet from BlazarI pcie  product:

......

[17.098858] Bluetooth: hci0: Found device firmware: intel/ibt-0190-0291-pci.sfi
[17.098871] Bluetooth: hci0: Boot Address: 0x10000800
[17.098872] Bluetooth: hci0: Firmware Version: 214-25.24
[17.158229] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
[17.158236] Bluetooth: BNEP filters: protocol multicast
[17.158241] Bluetooth: BNEP socket layer initialized
[17.468789] Bluetooth: hci0: Waiting for firmware download to complete
[17.468793] Bluetooth: hci0: Firmware loaded in 361262 usecs
[17.468872] Bluetooth: hci0: Waiting for device to boot
[17.504148] Bluetooth: hci0: Device booted in 34512 usecs
[17.504148] Bluetooth: hci0: Malformed MSFT vendor event: 0x02
[17.504682] Bluetooth: hci0: Found Intel DDC parameters: intel/ibt-0190-0291-pci.ddc
[17.505380] Bluetooth: hci0: Applying Intel DDC parameters completed
[17.505622] Bluetooth: hci0: Firmware timestamp 2024.25 buildtype 3 build 64726
[17.505624] Bluetooth: hci0: Firmware SHA1: 0x9f4adddc
[17.505838] Bluetooth: hci0: Fseq status: Success (0x00)
[17.505839] Bluetooth: hci0: Fseq executed: 00.00.04.183
[17.505840] Bluetooth: hci0: Fseq BT Top: 00.00.04.183

dmesg snippet from BlazarI usb product:

.......

[14.212072] Bluetooth: hci0: Found device firmware: intel/ibt-0190-0291-usb.sfi
[14.212091] Bluetooth: hci0: Boot Address: 0x10000800
[14.212093] Bluetooth: hci0: Firmware Version: 79-21.24
[14.262125] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
[14.262129] Bluetooth: BNEP filters: protocol multicast
[14.262133] Bluetooth: BNEP socket layer initialized
[15.865421] Bluetooth: hci0: Waiting for firmware download to complete
[15.865991] Bluetooth: hci0: Firmware loaded in 1615150 usecs
[15.866017] Bluetooth: hci0: Waiting for device to boot
[15.899934] Bluetooth: hci0: Malformed MSFT vendor event: 0x02
[15.899942] Bluetooth: hci0: Device booted in 33139 usecs
[15.900172] Bluetooth: hci0: Found Intel DDC parameters: intel/ibt-0190-0291-usb.ddc
[15.901928] Bluetooth: hci0: Applying Intel DDC parameters completed
[15.904993] Bluetooth: hci0: Firmware timestamp 2024.21 buildtype 3 build 63311
[15.904996] Bluetooth: hci0: Firmware SHA1: 0x8b217cf7
[15.908929] Bluetooth: hci0: Fseq status: Success (0x00)
[15.908934] Bluetooth: hci0: Fseq executed: 00.00.04.180
[15.908935] Bluetooth: hci0: Fseq BT Top: 00.00.04.180

Signed-off-by: Kiran K <kiran.k@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btintel.c | 88 ++++++++++++++++++++++++++++++++++-----------
 drivers/bluetooth/btintel.h |  6 +++-
 2 files changed, 73 insertions(+), 21 deletions(-)

diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c
index ff33e1aa2929..5d735391545a 100644
--- a/drivers/bluetooth/btintel.c
+++ b/drivers/bluetooth/btintel.c
@@ -631,6 +631,10 @@ int btintel_parse_version_tlv(struct hci_dev *hdev,
 		case INTEL_TLV_GIT_SHA1:
 			version->git_sha1 = get_unaligned_le32(tlv->val);
 			break;
+		case INTEL_TLV_FW_ID:
+			snprintf(version->fw_id, sizeof(version->fw_id),
+				 "%s", tlv->val);
+			break;
 		default:
 			/* Ignore rest of information */
 			break;
@@ -2133,30 +2137,61 @@ static void btintel_get_fw_name_tlv(const struct intel_version_tlv *ver,
 				    const char *suffix)
 {
 	const char *format;
-	/* The firmware file name for new generation controllers will be
-	 * ibt-<cnvi_top type+cnvi_top step>-<cnvr_top type+cnvr_top step>
-	 */
-	switch (ver->cnvi_top & 0xfff) {
+	u32 cnvi, cnvr;
+
+	cnvi = INTEL_CNVX_TOP_PACK_SWAB(INTEL_CNVX_TOP_TYPE(ver->cnvi_top),
+					INTEL_CNVX_TOP_STEP(ver->cnvi_top));
+
+	cnvr = INTEL_CNVX_TOP_PACK_SWAB(INTEL_CNVX_TOP_TYPE(ver->cnvr_top),
+					INTEL_CNVX_TOP_STEP(ver->cnvr_top));
+
 	/* Only Blazar  product supports downloading of intermediate loader
 	 * image
 	 */
-	case BTINTEL_CNVI_BLAZARI:
-		if (ver->img_type == BTINTEL_IMG_BOOTLOADER)
+	if ((ver->cnvi_top & 0xfff) >= BTINTEL_CNVI_BLAZARI) {
+		u8 zero[BTINTEL_FWID_MAXLEN];
+
+		if (ver->img_type == BTINTEL_IMG_BOOTLOADER) {
 			format = "intel/ibt-%04x-%04x-iml.%s";
-		else
-			format = "intel/ibt-%04x-%04x.%s";
-		break;
-	default:
-			format = "intel/ibt-%04x-%04x.%s";
-		break;
+			snprintf(fw_name, len, format, cnvi, cnvr, suffix);
+			return;
+		}
+
+		memset(zero, 0, sizeof(zero));
+
+		/* ibt-<cnvi_top type+cnvi_top step>-<cnvr_top type+cnvr_top step-fw_id> */
+		if (memcmp(ver->fw_id, zero, sizeof(zero))) {
+			format = "intel/ibt-%04x-%04x-%s.%s";
+			snprintf(fw_name, len, format, cnvi, cnvr,
+				 ver->fw_id, suffix);
+			return;
+		}
+		/* If firmware id is not present, fallback to legacy naming
+		 * convention
+		 */
 	}
+	/* Fallback to legacy naming convention for other controllers
+	 * ibt-<cnvi_top type+cnvi_top step>-<cnvr_top type+cnvr_top step>
+	 */
+	format = "intel/ibt-%04x-%04x.%s";
+	snprintf(fw_name, len, format, cnvi, cnvr, suffix);
+}
+
+static void btintel_get_iml_tlv(const struct intel_version_tlv *ver,
+				char *fw_name, size_t len,
+				const char *suffix)
+{
+	const char *format;
+	u32 cnvi, cnvr;
+
+	cnvi = INTEL_CNVX_TOP_PACK_SWAB(INTEL_CNVX_TOP_TYPE(ver->cnvi_top),
+					INTEL_CNVX_TOP_STEP(ver->cnvi_top));
 
-	snprintf(fw_name, len, format,
-		 INTEL_CNVX_TOP_PACK_SWAB(INTEL_CNVX_TOP_TYPE(ver->cnvi_top),
-					  INTEL_CNVX_TOP_STEP(ver->cnvi_top)),
-		 INTEL_CNVX_TOP_PACK_SWAB(INTEL_CNVX_TOP_TYPE(ver->cnvr_top),
-					  INTEL_CNVX_TOP_STEP(ver->cnvr_top)),
-		 suffix);
+	cnvr = INTEL_CNVX_TOP_PACK_SWAB(INTEL_CNVX_TOP_TYPE(ver->cnvr_top),
+					INTEL_CNVX_TOP_STEP(ver->cnvr_top));
+
+	format = "intel/ibt-%04x-%04x-iml.%s";
+	snprintf(fw_name, len, format, cnvi, cnvr, suffix);
 }
 
 static int btintel_prepare_fw_download_tlv(struct hci_dev *hdev,
@@ -2164,7 +2199,7 @@ static int btintel_prepare_fw_download_tlv(struct hci_dev *hdev,
 					   u32 *boot_param)
 {
 	const struct firmware *fw;
-	char fwname[64];
+	char fwname[128];
 	int err;
 	ktime_t calltime;
 
@@ -2199,7 +2234,20 @@ static int btintel_prepare_fw_download_tlv(struct hci_dev *hdev,
 		}
 	}
 
-	btintel_get_fw_name_tlv(ver, fwname, sizeof(fwname), "sfi");
+	if (ver->img_type == BTINTEL_IMG_OP) {
+		/* Controller running OP image. In case of FW downgrade,
+		 * FWID TLV may not be present and driver may attempt to load
+		 * firmware image which doesn't exist. Lets compare the version
+		 * of IML image
+		 */
+		if ((ver->cnvi_top & 0xfff) >= BTINTEL_CNVI_BLAZARI)
+			btintel_get_iml_tlv(ver, fwname, sizeof(fwname), "sfi");
+		else
+			btintel_get_fw_name_tlv(ver, fwname, sizeof(fwname), "sfi");
+	} else {
+		btintel_get_fw_name_tlv(ver, fwname, sizeof(fwname), "sfi");
+	}
+
 	err = firmware_request_nowarn(&fw, fwname, &hdev->dev);
 	if (err < 0) {
 		if (!btintel_test_flag(hdev, INTEL_BOOTLOADER)) {
diff --git a/drivers/bluetooth/btintel.h b/drivers/bluetooth/btintel.h
index 9dbad1a7c47c..aa70e4c27416 100644
--- a/drivers/bluetooth/btintel.h
+++ b/drivers/bluetooth/btintel.h
@@ -42,7 +42,8 @@ enum {
 	INTEL_TLV_SBE_TYPE,
 	INTEL_TLV_OTP_BDADDR,
 	INTEL_TLV_UNLOCKED_STATE,
-	INTEL_TLV_GIT_SHA1
+	INTEL_TLV_GIT_SHA1,
+	INTEL_TLV_FW_ID = 0x50
 };
 
 struct intel_tlv {
@@ -57,6 +58,8 @@ struct intel_tlv {
 #define BTINTEL_IMG_IML			0x02	/* Intermediate image */
 #define BTINTEL_IMG_OP			0x03	/* Operational image */
 
+#define BTINTEL_FWID_MAXLEN 64
+
 struct intel_version_tlv {
 	u32	cnvi_top;
 	u32	cnvr_top;
@@ -77,6 +80,7 @@ struct intel_version_tlv {
 	u8	limited_cce;
 	u8	sbe_type;
 	u32	git_sha1;
+	u8	fw_id[BTINTEL_FWID_MAXLEN];
 	bdaddr_t otp_bd_addr;
 };
 
-- 
cgit v1.2.3-58-ga151


From 555a05d84ca2c587e2d4777006e2c2fb3dfbd91d Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sun, 14 Jul 2024 01:53:32 +0300
Subject: net: dpaa: avoid on-stack arrays of NR_CPUS elements

The dpaa-eth driver is written for PowerPC and Arm SoCs which have 1-24
CPUs. It depends on CONFIG_NR_CPUS having a reasonably small value in
Kconfig. Otherwise, there are 2 functions which allocate on-stack arrays
of NR_CPUS elements, and these can quickly explode in size, leading to
warnings such as:

  drivers/net/ethernet/freescale/dpaa/dpaa_eth.c:3280:12: warning:
  stack frame size (16664) exceeds limit (2048) in 'dpaa_eth_probe' [-Wframe-larger-than]

The problem is twofold:
- Reducing the array size to the boot-time num_possible_cpus() (rather
  than the compile-time NR_CPUS) creates a variable-length array,
  which should be avoided in the Linux kernel.
- Using NR_CPUS as an array size makes the driver blow up in stack
  consumption with generic, as opposed to hand-crafted, .config files.

A simple solution is to use dynamic allocation for num_possible_cpus()
elements (aka a small number determined at runtime).

Link: https://lore.kernel.org/all/202406261920.l5pzM1rj-lkp@intel.com/
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Breno Leitao <leitao@debian.org>
Acked-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Link: https://patch.msgid.link/20240713225336.1746343-2-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/dpaa/dpaa_eth.c     | 20 +++++++++++++++-----
 drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c | 10 +++++++++-
 2 files changed, 24 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index baa0b3c2ce6f..946c3d3b69d9 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -931,14 +931,18 @@ static inline void dpaa_setup_egress(const struct dpaa_priv *priv,
 	}
 }
 
-static void dpaa_fq_setup(struct dpaa_priv *priv,
-			  const struct dpaa_fq_cbs *fq_cbs,
-			  struct fman_port *tx_port)
+static int dpaa_fq_setup(struct dpaa_priv *priv,
+			 const struct dpaa_fq_cbs *fq_cbs,
+			 struct fman_port *tx_port)
 {
 	int egress_cnt = 0, conf_cnt = 0, num_portals = 0, portal_cnt = 0, cpu;
 	const cpumask_t *affine_cpus = qman_affine_cpus();
-	u16 channels[NR_CPUS];
 	struct dpaa_fq *fq;
+	u16 *channels;
+
+	channels = kcalloc(num_possible_cpus(), sizeof(u16), GFP_KERNEL);
+	if (!channels)
+		return -ENOMEM;
 
 	for_each_cpu_and(cpu, affine_cpus, cpu_online_mask)
 		channels[num_portals++] = qman_affine_channel(cpu);
@@ -997,6 +1001,10 @@ static void dpaa_fq_setup(struct dpaa_priv *priv,
 				break;
 		}
 	}
+
+	kfree(channels);
+
+	return 0;
 }
 
 static inline int dpaa_tx_fq_to_id(const struct dpaa_priv *priv,
@@ -3416,7 +3424,9 @@ static int dpaa_eth_probe(struct platform_device *pdev)
 	 */
 	dpaa_eth_add_channel(priv->channel, &pdev->dev);
 
-	dpaa_fq_setup(priv, &dpaa_fq_cbs, priv->mac_dev->port[TX]);
+	err = dpaa_fq_setup(priv, &dpaa_fq_cbs, priv->mac_dev->port[TX]);
+	if (err)
+		goto free_dpaa_bps;
 
 	/* Create a congestion group for this netdev, with
 	 * dynamically-allocated CGR ID.
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
index 5bd0b36d1feb..3f8cd4a7d845 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
@@ -457,12 +457,16 @@ static int dpaa_set_coalesce(struct net_device *dev,
 			     struct netlink_ext_ack *extack)
 {
 	const cpumask_t *cpus = qman_affine_cpus();
-	bool needs_revert[NR_CPUS] = {false};
 	struct qman_portal *portal;
 	u32 period, prev_period;
 	u8 thresh, prev_thresh;
+	bool *needs_revert;
 	int cpu, res;
 
+	needs_revert = kcalloc(num_possible_cpus(), sizeof(bool), GFP_KERNEL);
+	if (!needs_revert)
+		return -ENOMEM;
+
 	period = c->rx_coalesce_usecs;
 	thresh = c->rx_max_coalesced_frames;
 
@@ -485,6 +489,8 @@ static int dpaa_set_coalesce(struct net_device *dev,
 		needs_revert[cpu] = true;
 	}
 
+	kfree(needs_revert);
+
 	return 0;
 
 revert_values:
@@ -498,6 +504,8 @@ revert_values:
 		qman_dqrr_set_ithresh(portal, prev_thresh);
 	}
 
+	kfree(needs_revert);
+
 	return res;
 }
 
-- 
cgit v1.2.3-58-ga151


From e7072750bbcb4d74945ed1e22070b96cc3457def Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sun, 14 Jul 2024 01:53:33 +0300
Subject: net: dpaa: eliminate NR_CPUS dependency in egress_fqs[] and
 conf_fqs[]

The driver uses the DPAA_TC_TXQ_NUM and DPAA_ETH_TXQ_NUM macros for TX
queue handling, and they depend on CONFIG_NR_CPUS.

In generic .config files, these can go to very large (8096 CPUs) values
for the systems that DPAA1 is integrated in (1-24 CPUs). We allocate a
lot of resources that will never be used. Those are:
- system memory
- QMan FQIDs as managed by qman_alloc_fqid_range(). This is especially
  painful since currently, when booting with CONFIG_NR_CPUS=8096, a
  LS1046A-RDB system will only manage to probe 3 of its 6 interfaces.
  The rest will run out of FQD ("/reserved-memory/qman-fqd" in the
  device tree) and fail at the qman_create_fq() stage of the probing
  process.
- netdev queues as alloc_etherdev_mq() argument. The high queue indices
  are simply hidden from the network stack after the call to
  netif_set_real_num_tx_queues().

With just a tiny bit more effort, we can replace the NR_CPUS
compile-time constant with the num_possible_cpus() run-time constant,
and dynamically allocate the egress_fqs[] and conf_fqs[] arrays.
Even on a system with a high CONFIG_NR_CPUS, num_possible_cpus() will
remain equal to the number of available cores on the SoC.

The replacement is as follows:
- DPAA_TC_TXQ_NUM -> dpaa_num_txqs_per_tc()
- DPAA_ETH_TXQ_NUM -> dpaa_max_num_txqs()

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Link: https://patch.msgid.link/20240713225336.1746343-3-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 47 ++++++++++++++++++--------
 drivers/net/ethernet/freescale/dpaa/dpaa_eth.h | 20 +++++++----
 2 files changed, 47 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index 946c3d3b69d9..ea1fcee77f9b 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -371,6 +371,7 @@ static int dpaa_setup_tc(struct net_device *net_dev, enum tc_setup_type type,
 			 void *type_data)
 {
 	struct dpaa_priv *priv = netdev_priv(net_dev);
+	int num_txqs_per_tc = dpaa_num_txqs_per_tc();
 	struct tc_mqprio_qopt *mqprio = type_data;
 	u8 num_tc;
 	int i;
@@ -398,12 +399,12 @@ static int dpaa_setup_tc(struct net_device *net_dev, enum tc_setup_type type,
 	netdev_set_num_tc(net_dev, num_tc);
 
 	for (i = 0; i < num_tc; i++)
-		netdev_set_tc_queue(net_dev, i, DPAA_TC_TXQ_NUM,
-				    i * DPAA_TC_TXQ_NUM);
+		netdev_set_tc_queue(net_dev, i, num_txqs_per_tc,
+				    i * num_txqs_per_tc);
 
 out:
 	priv->num_tc = num_tc ? : 1;
-	netif_set_real_num_tx_queues(net_dev, priv->num_tc * DPAA_TC_TXQ_NUM);
+	netif_set_real_num_tx_queues(net_dev, priv->num_tc * num_txqs_per_tc);
 	return 0;
 }
 
@@ -649,7 +650,7 @@ static inline void dpaa_assign_wq(struct dpaa_fq *fq, int idx)
 		fq->wq = 6;
 		break;
 	case FQ_TYPE_TX:
-		switch (idx / DPAA_TC_TXQ_NUM) {
+		switch (idx / dpaa_num_txqs_per_tc()) {
 		case 0:
 			/* Low priority (best effort) */
 			fq->wq = 6;
@@ -667,8 +668,8 @@ static inline void dpaa_assign_wq(struct dpaa_fq *fq, int idx)
 			fq->wq = 0;
 			break;
 		default:
-			WARN(1, "Too many TX FQs: more than %d!\n",
-			     DPAA_ETH_TXQ_NUM);
+			WARN(1, "Too many TX FQs: more than %zu!\n",
+			     dpaa_max_num_txqs());
 		}
 		break;
 	default:
@@ -740,7 +741,8 @@ static int dpaa_alloc_all_fqs(struct device *dev, struct list_head *list,
 
 	port_fqs->rx_pcdq = &dpaa_fq[0];
 
-	if (!dpaa_fq_alloc(dev, 0, DPAA_ETH_TXQ_NUM, list, FQ_TYPE_TX_CONF_MQ))
+	if (!dpaa_fq_alloc(dev, 0, dpaa_max_num_txqs(), list,
+			   FQ_TYPE_TX_CONF_MQ))
 		goto fq_alloc_failed;
 
 	dpaa_fq = dpaa_fq_alloc(dev, 0, 1, list, FQ_TYPE_TX_ERROR);
@@ -755,7 +757,7 @@ static int dpaa_alloc_all_fqs(struct device *dev, struct list_head *list,
 
 	port_fqs->tx_defq = &dpaa_fq[0];
 
-	if (!dpaa_fq_alloc(dev, 0, DPAA_ETH_TXQ_NUM, list, FQ_TYPE_TX))
+	if (!dpaa_fq_alloc(dev, 0, dpaa_max_num_txqs(), list, FQ_TYPE_TX))
 		goto fq_alloc_failed;
 
 	return 0;
@@ -972,7 +974,7 @@ static int dpaa_fq_setup(struct dpaa_priv *priv,
 			/* If we have more Tx queues than the number of cores,
 			 * just ignore the extra ones.
 			 */
-			if (egress_cnt < DPAA_ETH_TXQ_NUM)
+			if (egress_cnt < dpaa_max_num_txqs())
 				priv->egress_fqs[egress_cnt++] = &fq->fq_base;
 			break;
 		case FQ_TYPE_TX_CONF_MQ:
@@ -992,12 +994,12 @@ static int dpaa_fq_setup(struct dpaa_priv *priv,
 	}
 
 	 /* Make sure all CPUs receive a corresponding Tx queue. */
-	while (egress_cnt < DPAA_ETH_TXQ_NUM) {
+	while (egress_cnt < dpaa_max_num_txqs()) {
 		list_for_each_entry(fq, &priv->dpaa_fq_list, list) {
 			if (fq->fq_type != FQ_TYPE_TX)
 				continue;
 			priv->egress_fqs[egress_cnt++] = &fq->fq_base;
-			if (egress_cnt == DPAA_ETH_TXQ_NUM)
+			if (egress_cnt == dpaa_max_num_txqs())
 				break;
 		}
 	}
@@ -1012,7 +1014,7 @@ static inline int dpaa_tx_fq_to_id(const struct dpaa_priv *priv,
 {
 	int i;
 
-	for (i = 0; i < DPAA_ETH_TXQ_NUM; i++)
+	for (i = 0; i < dpaa_max_num_txqs(); i++)
 		if (priv->egress_fqs[i] == tx_fq)
 			return i;
 
@@ -3332,7 +3334,7 @@ static int dpaa_eth_probe(struct platform_device *pdev)
 	/* Allocate this early, so we can store relevant information in
 	 * the private area
 	 */
-	net_dev = alloc_etherdev_mq(sizeof(*priv), DPAA_ETH_TXQ_NUM);
+	net_dev = alloc_etherdev_mq(sizeof(*priv), dpaa_max_num_txqs());
 	if (!net_dev) {
 		dev_err(dev, "alloc_etherdev_mq() failed\n");
 		return -ENOMEM;
@@ -3347,6 +3349,22 @@ static int dpaa_eth_probe(struct platform_device *pdev)
 
 	priv->msg_enable = netif_msg_init(debug, DPAA_MSG_DEFAULT);
 
+	priv->egress_fqs = devm_kcalloc(dev, dpaa_max_num_txqs(),
+					sizeof(*priv->egress_fqs),
+					GFP_KERNEL);
+	if (!priv->egress_fqs) {
+		err = -ENOMEM;
+		goto free_netdev;
+	}
+
+	priv->conf_fqs = devm_kcalloc(dev, dpaa_max_num_txqs(),
+				      sizeof(*priv->conf_fqs),
+				      GFP_KERNEL);
+	if (!priv->conf_fqs) {
+		err = -ENOMEM;
+		goto free_netdev;
+	}
+
 	mac_dev = dpaa_mac_dev_get(pdev);
 	if (IS_ERR(mac_dev)) {
 		netdev_err(net_dev, "dpaa_mac_dev_get() failed\n");
@@ -3472,7 +3490,8 @@ static int dpaa_eth_probe(struct platform_device *pdev)
 	}
 
 	priv->num_tc = 1;
-	netif_set_real_num_tx_queues(net_dev, priv->num_tc * DPAA_TC_TXQ_NUM);
+	netif_set_real_num_tx_queues(net_dev,
+				     priv->num_tc * dpaa_num_txqs_per_tc());
 
 	/* Initialize NAPI */
 	err = dpaa_napi_add(net_dev);
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h
index ac3c8ed57bbe..7ed659eb08de 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.h
@@ -18,10 +18,6 @@
 
 /* Number of prioritised traffic classes */
 #define DPAA_TC_NUM		4
-/* Number of Tx queues per traffic class */
-#define DPAA_TC_TXQ_NUM		NR_CPUS
-/* Total number of Tx queues */
-#define DPAA_ETH_TXQ_NUM	(DPAA_TC_NUM * DPAA_TC_TXQ_NUM)
 
 /* More detailed FQ types - used for fine-grained WQ assignments */
 enum dpaa_fq_type {
@@ -142,8 +138,8 @@ struct dpaa_priv {
 	struct mac_device *mac_dev;
 	struct device *rx_dma_dev;
 	struct device *tx_dma_dev;
-	struct qman_fq *egress_fqs[DPAA_ETH_TXQ_NUM];
-	struct qman_fq *conf_fqs[DPAA_ETH_TXQ_NUM];
+	struct qman_fq **egress_fqs;
+	struct qman_fq **conf_fqs;
 
 	u16 channel;
 	struct list_head dpaa_fq_list;
@@ -185,4 +181,16 @@ extern const struct ethtool_ops dpaa_ethtool_ops;
 /* from dpaa_eth_sysfs.c */
 void dpaa_eth_sysfs_remove(struct device *dev);
 void dpaa_eth_sysfs_init(struct device *dev);
+
+static inline size_t dpaa_num_txqs_per_tc(void)
+{
+	return num_possible_cpus();
+}
+
+/* Total number of Tx queues */
+static inline size_t dpaa_max_num_txqs(void)
+{
+	return DPAA_TC_NUM * dpaa_num_txqs_per_tc();
+}
+
 #endif	/* __DPAA_H */
-- 
cgit v1.2.3-58-ga151


From e3672a6d5e89589b734fdbf1e79528b7bcafd11f Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sun, 14 Jul 2024 01:53:34 +0300
Subject: net: dpaa: stop ignoring TX queues past the number of CPUs

dpaa_fq_setup() iterates through the queues allocated by dpaa_alloc_all_fqs()
and saved in &priv->dpaa_fq_list.

The allocation for FQ_TYPE_TX looks as follows:

	if (!dpaa_fq_alloc(dev, 0, dpaa_max_num_txqs(), list, FQ_TYPE_TX))
		goto fq_alloc_failed;

Thus, iterating again through FQ_TYPE_TX queues in dpaa_fq_setup() and
counting them will never yield an egress_cnt larger than the allocated
size, dpaa_max_num_txqs().

The comparison serves no purpose since it is always true; remove it.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Link: https://patch.msgid.link/20240713225336.1746343-4-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index ea1fcee77f9b..f844c9261571 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -971,11 +971,7 @@ static int dpaa_fq_setup(struct dpaa_priv *priv,
 		case FQ_TYPE_TX:
 			dpaa_setup_egress(priv, fq, tx_port,
 					  &fq_cbs->egress_ern);
-			/* If we have more Tx queues than the number of cores,
-			 * just ignore the extra ones.
-			 */
-			if (egress_cnt < dpaa_max_num_txqs())
-				priv->egress_fqs[egress_cnt++] = &fq->fq_base;
+			priv->egress_fqs[egress_cnt++] = &fq->fq_base;
 			break;
 		case FQ_TYPE_TX_CONF_MQ:
 			priv->conf_fqs[conf_cnt++] = &fq->fq_base;
-- 
cgit v1.2.3-58-ga151


From 6d2338205d78bb27ec2a3e035a4c2867320a0719 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sun, 14 Jul 2024 01:53:35 +0300
Subject: net: dpaa: no need to make sure all CPUs receive a corresponding Tx
 queue

dpaa_fq_setup() iterates through the &priv->dpaa_fq_list elements
allocated by dpaa_alloc_all_fqs(). This includes a call to:

	if (!dpaa_fq_alloc(dev, 0, dpaa_max_num_txqs(), list, FQ_TYPE_TX))
		goto fq_alloc_failed;

which gives us dpaa_max_num_txqs() elements of FQ_TYPE_TX type.

The code block which we are deleting runs after an earlier iteration
through &priv->dpaa_fq_list. So at the end of this iteration (for which
there is no early break), egress_cnt will be unconditionally equal to
dpaa_max_num_txqs().

In other words, dpaa_alloc_all_fqs() has already allocated TX queues for
all possible CPUs and the maximal number of traffic classes, and we've
already iterated once through them all.

The while() condition is dead code, remove it.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Link: https://patch.msgid.link/20240713225336.1746343-5-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 11 -----------
 1 file changed, 11 deletions(-)

diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index f844c9261571..cfe6b57b1da0 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -989,17 +989,6 @@ static int dpaa_fq_setup(struct dpaa_priv *priv,
 		}
 	}
 
-	 /* Make sure all CPUs receive a corresponding Tx queue. */
-	while (egress_cnt < dpaa_max_num_txqs()) {
-		list_for_each_entry(fq, &priv->dpaa_fq_list, list) {
-			if (fq->fq_type != FQ_TYPE_TX)
-				continue;
-			priv->egress_fqs[egress_cnt++] = &fq->fq_base;
-			if (egress_cnt == dpaa_max_num_txqs())
-				break;
-		}
-	}
-
 	kfree(channels);
 
 	return 0;
-- 
cgit v1.2.3-58-ga151


From 782fe08e9861d00ce8aca370453dd9ceb9a23d50 Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Sun, 14 Jul 2024 01:53:36 +0300
Subject: soc: fsl: qbman: FSL_DPAA depends on COMPILE_TEST

As most of the drivers that depend on ARCH_LAYERSCAPE, make FSL_DPAA
depend on COMPILE_TEST for compilation and testing.

	# grep -r depends.\*ARCH_LAYERSCAPE.\*COMPILE_TEST | wc -l
	29

Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Link: https://patch.msgid.link/20240713225336.1746343-6-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/soc/fsl/qbman/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/soc/fsl/qbman/Kconfig b/drivers/soc/fsl/qbman/Kconfig
index bdecb86bb656..27774ec6ff90 100644
--- a/drivers/soc/fsl/qbman/Kconfig
+++ b/drivers/soc/fsl/qbman/Kconfig
@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0-only
 menuconfig FSL_DPAA
 	bool "QorIQ DPAA1 framework support"
-	depends on ((FSL_SOC_BOOKE || ARCH_LAYERSCAPE) && ARCH_DMA_ADDR_T_64BIT)
+	depends on ((FSL_SOC_BOOKE || ARCH_LAYERSCAPE || COMPILE_TEST) && ARCH_DMA_ADDR_T_64BIT)
 	select GENERIC_ALLOCATOR
 	help
 	  The Freescale Data Path Acceleration Architecture (DPAA) is a set of
-- 
cgit v1.2.3-58-ga151


From 79eecf631c14e7f4057186570ac20e2cfac3802e Mon Sep 17 00:00:00 2001
From: Chengen Du <chengen.du@canonical.com>
Date: Sat, 13 Jul 2024 19:47:35 +0800
Subject: af_packet: Handle outgoing VLAN packets without hardware offloading

The issue initially stems from libpcap. The ethertype will be overwritten
as the VLAN TPID if the network interface lacks hardware VLAN offloading.
In the outbound packet path, if hardware VLAN offloading is unavailable,
the VLAN tag is inserted into the payload but then cleared from the sk_buff
struct. Consequently, this can lead to a false negative when checking for
the presence of a VLAN tag, causing the packet sniffing outcome to lack
VLAN tag information (i.e., TCI-TPID). As a result, the packet capturing
tool may be unable to parse packets as expected.

The TCI-TPID is missing because the prb_fill_vlan_info() function does not
modify the tp_vlan_tci/tp_vlan_tpid values, as the information is in the
payload and not in the sk_buff struct. The skb_vlan_tag_present() function
only checks vlan_all in the sk_buff struct. In cooked mode, the L2 header
is stripped, preventing the packet capturing tool from determining the
correct TCI-TPID value. Additionally, the protocol in SLL is incorrect,
which means the packet capturing tool cannot parse the L3 header correctly.

Link: https://github.com/the-tcpdump-group/libpcap/issues/1105
Link: https://lore.kernel.org/netdev/20240520070348.26725-1-chengen.du@canonical.com/T/#u
Fixes: 393e52e33c6c ("packet: deliver VLAN TCI to userspace")
Cc: stable@vger.kernel.org
Signed-off-by: Chengen Du <chengen.du@canonical.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20240713114735.62360-1-chengen.du@canonical.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/packet/af_packet.c | 86 ++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 84 insertions(+), 2 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index ea3ebc160e25..4692a9ef110b 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -538,6 +538,61 @@ static void *packet_current_frame(struct packet_sock *po,
 	return packet_lookup_frame(po, rb, rb->head, status);
 }
 
+static u16 vlan_get_tci(struct sk_buff *skb, struct net_device *dev)
+{
+	u8 *skb_orig_data = skb->data;
+	int skb_orig_len = skb->len;
+	struct vlan_hdr vhdr, *vh;
+	unsigned int header_len;
+
+	if (!dev)
+		return 0;
+
+	/* In the SOCK_DGRAM scenario, skb data starts at the network
+	 * protocol, which is after the VLAN headers. The outer VLAN
+	 * header is at the hard_header_len offset in non-variable
+	 * length link layer headers. If it's a VLAN device, the
+	 * min_header_len should be used to exclude the VLAN header
+	 * size.
+	 */
+	if (dev->min_header_len == dev->hard_header_len)
+		header_len = dev->hard_header_len;
+	else if (is_vlan_dev(dev))
+		header_len = dev->min_header_len;
+	else
+		return 0;
+
+	skb_push(skb, skb->data - skb_mac_header(skb));
+	vh = skb_header_pointer(skb, header_len, sizeof(vhdr), &vhdr);
+	if (skb_orig_data != skb->data) {
+		skb->data = skb_orig_data;
+		skb->len = skb_orig_len;
+	}
+	if (unlikely(!vh))
+		return 0;
+
+	return ntohs(vh->h_vlan_TCI);
+}
+
+static __be16 vlan_get_protocol_dgram(struct sk_buff *skb)
+{
+	__be16 proto = skb->protocol;
+
+	if (unlikely(eth_type_vlan(proto))) {
+		u8 *skb_orig_data = skb->data;
+		int skb_orig_len = skb->len;
+
+		skb_push(skb, skb->data - skb_mac_header(skb));
+		proto = __vlan_get_protocol(skb, proto, NULL);
+		if (skb_orig_data != skb->data) {
+			skb->data = skb_orig_data;
+			skb->len = skb_orig_len;
+		}
+	}
+
+	return proto;
+}
+
 static void prb_del_retire_blk_timer(struct tpacket_kbdq_core *pkc)
 {
 	del_timer_sync(&pkc->retire_blk_timer);
@@ -1007,10 +1062,16 @@ static void prb_clear_rxhash(struct tpacket_kbdq_core *pkc,
 static void prb_fill_vlan_info(struct tpacket_kbdq_core *pkc,
 			struct tpacket3_hdr *ppd)
 {
+	struct packet_sock *po = container_of(pkc, struct packet_sock, rx_ring.prb_bdqc);
+
 	if (skb_vlan_tag_present(pkc->skb)) {
 		ppd->hv1.tp_vlan_tci = skb_vlan_tag_get(pkc->skb);
 		ppd->hv1.tp_vlan_tpid = ntohs(pkc->skb->vlan_proto);
 		ppd->tp_status = TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
+	} else if (unlikely(po->sk.sk_type == SOCK_DGRAM && eth_type_vlan(pkc->skb->protocol))) {
+		ppd->hv1.tp_vlan_tci = vlan_get_tci(pkc->skb, pkc->skb->dev);
+		ppd->hv1.tp_vlan_tpid = ntohs(pkc->skb->protocol);
+		ppd->tp_status = TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
 	} else {
 		ppd->hv1.tp_vlan_tci = 0;
 		ppd->hv1.tp_vlan_tpid = 0;
@@ -2428,6 +2489,10 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 			h.h2->tp_vlan_tci = skb_vlan_tag_get(skb);
 			h.h2->tp_vlan_tpid = ntohs(skb->vlan_proto);
 			status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
+		} else if (unlikely(sk->sk_type == SOCK_DGRAM && eth_type_vlan(skb->protocol))) {
+			h.h2->tp_vlan_tci = vlan_get_tci(skb, skb->dev);
+			h.h2->tp_vlan_tpid = ntohs(skb->protocol);
+			status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
 		} else {
 			h.h2->tp_vlan_tci = 0;
 			h.h2->tp_vlan_tpid = 0;
@@ -2457,7 +2522,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev,
 	sll->sll_halen = dev_parse_header(skb, sll->sll_addr);
 	sll->sll_family = AF_PACKET;
 	sll->sll_hatype = dev->type;
-	sll->sll_protocol = skb->protocol;
+	sll->sll_protocol = (sk->sk_type == SOCK_DGRAM) ?
+		vlan_get_protocol_dgram(skb) : skb->protocol;
 	sll->sll_pkttype = skb->pkt_type;
 	if (unlikely(packet_sock_flag(po, PACKET_SOCK_ORIGDEV)))
 		sll->sll_ifindex = orig_dev->ifindex;
@@ -3482,7 +3548,8 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 		/* Original length was stored in sockaddr_ll fields */
 		origlen = PACKET_SKB_CB(skb)->sa.origlen;
 		sll->sll_family = AF_PACKET;
-		sll->sll_protocol = skb->protocol;
+		sll->sll_protocol = (sock->type == SOCK_DGRAM) ?
+			vlan_get_protocol_dgram(skb) : skb->protocol;
 	}
 
 	sock_recv_cmsgs(msg, sk, skb);
@@ -3539,6 +3606,21 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 			aux.tp_vlan_tci = skb_vlan_tag_get(skb);
 			aux.tp_vlan_tpid = ntohs(skb->vlan_proto);
 			aux.tp_status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
+		} else if (unlikely(sock->type == SOCK_DGRAM && eth_type_vlan(skb->protocol))) {
+			struct sockaddr_ll *sll = &PACKET_SKB_CB(skb)->sa.ll;
+			struct net_device *dev;
+
+			rcu_read_lock();
+			dev = dev_get_by_index_rcu(sock_net(sk), sll->sll_ifindex);
+			if (dev) {
+				aux.tp_vlan_tci = vlan_get_tci(skb, dev);
+				aux.tp_vlan_tpid = ntohs(skb->protocol);
+				aux.tp_status |= TP_STATUS_VLAN_VALID | TP_STATUS_VLAN_TPID_VALID;
+			} else {
+				aux.tp_vlan_tci = 0;
+				aux.tp_vlan_tpid = 0;
+			}
+			rcu_read_unlock();
 		} else {
 			aux.tp_vlan_tci = 0;
 			aux.tp_vlan_tpid = 0;
-- 
cgit v1.2.3-58-ga151


From 41d4a174201e62fc95562dc8e53097baf8568f24 Mon Sep 17 00:00:00 2001
From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Date: Mon, 8 Jul 2024 19:25:28 +0800
Subject: virtio_net: replace VIRTIO_XDP_HEADROOM by XDP_PACKET_HEADROOM

virtio net has VIRTIO_XDP_HEADROOM that is equal to
XDP_PACKET_HEADROOM to calculate the headroom for xdp.

But here we should use the macro XDP_PACKET_HEADROOM from bpf.h to
calculate the headroom for xdp. So here we remove the
VIRTIO_XDP_HEADROOM, and use the XDP_PACKET_HEADROOM to replace it.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-2-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 21 +++++++++------------
 1 file changed, 9 insertions(+), 12 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 0b4747e81464..d99898e44456 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -40,9 +40,6 @@ module_param(napi_tx, bool, 0644);
 
 #define VIRTNET_RX_PAD (NET_IP_ALIGN + NET_SKB_PAD)
 
-/* Amount of XDP headroom to prepend to packets for use by xdp_adjust_head */
-#define VIRTIO_XDP_HEADROOM 256
-
 /* Separating two types of XDP xmit */
 #define VIRTIO_XDP_TX		BIT(0)
 #define VIRTIO_XDP_REDIR	BIT(1)
@@ -1268,7 +1265,7 @@ static int virtnet_xdp_handler(struct bpf_prog *xdp_prog, struct xdp_buff *xdp,
 
 static unsigned int virtnet_get_headroom(struct virtnet_info *vi)
 {
-	return vi->xdp_enabled ? VIRTIO_XDP_HEADROOM : 0;
+	return vi->xdp_enabled ? XDP_PACKET_HEADROOM : 0;
 }
 
 /* We copy the packet for XDP in the following cases:
@@ -1332,7 +1329,7 @@ static struct page *xdp_linearize_page(struct receive_queue *rq,
 	}
 
 	/* Headroom does not contribute to packet length */
-	*len = page_off - VIRTIO_XDP_HEADROOM;
+	*len = page_off - XDP_PACKET_HEADROOM;
 	return page;
 err_buf:
 	__free_pages(page, 0);
@@ -1619,8 +1616,8 @@ static int virtnet_build_xdp_buff_mrg(struct net_device *dev,
 	void *ctx;
 
 	xdp_init_buff(xdp, frame_sz, &rq->xdp_rxq);
-	xdp_prepare_buff(xdp, buf - VIRTIO_XDP_HEADROOM,
-			 VIRTIO_XDP_HEADROOM + vi->hdr_len, len - vi->hdr_len, true);
+	xdp_prepare_buff(xdp, buf - XDP_PACKET_HEADROOM,
+			 XDP_PACKET_HEADROOM + vi->hdr_len, len - vi->hdr_len, true);
 
 	if (!*num_buf)
 		return 0;
@@ -1737,12 +1734,12 @@ static void *mergeable_xdp_get_buf(struct virtnet_info *vi,
 		/* linearize data for XDP */
 		xdp_page = xdp_linearize_page(rq, num_buf,
 					      *page, offset,
-					      VIRTIO_XDP_HEADROOM,
+					      XDP_PACKET_HEADROOM,
 					      len);
 		if (!xdp_page)
 			return NULL;
 	} else {
-		xdp_room = SKB_DATA_ALIGN(VIRTIO_XDP_HEADROOM +
+		xdp_room = SKB_DATA_ALIGN(XDP_PACKET_HEADROOM +
 					  sizeof(struct skb_shared_info));
 		if (*len + xdp_room > PAGE_SIZE)
 			return NULL;
@@ -1751,7 +1748,7 @@ static void *mergeable_xdp_get_buf(struct virtnet_info *vi,
 		if (!xdp_page)
 			return NULL;
 
-		memcpy(page_address(xdp_page) + VIRTIO_XDP_HEADROOM,
+		memcpy(page_address(xdp_page) + XDP_PACKET_HEADROOM,
 		       page_address(*page) + offset, *len);
 	}
 
@@ -1761,7 +1758,7 @@ static void *mergeable_xdp_get_buf(struct virtnet_info *vi,
 
 	*page = xdp_page;
 
-	return page_address(*page) + VIRTIO_XDP_HEADROOM;
+	return page_address(*page) + XDP_PACKET_HEADROOM;
 }
 
 static struct sk_buff *receive_mergeable_xdp(struct net_device *dev,
@@ -4971,7 +4968,7 @@ static int virtnet_restore_guest_offloads(struct virtnet_info *vi)
 static int virtnet_xdp_set(struct net_device *dev, struct bpf_prog *prog,
 			   struct netlink_ext_ack *extack)
 {
-	unsigned int room = SKB_DATA_ALIGN(VIRTIO_XDP_HEADROOM +
+	unsigned int room = SKB_DATA_ALIGN(XDP_PACKET_HEADROOM +
 					   sizeof(struct skb_shared_info));
 	unsigned int max_sz = PAGE_SIZE - room - ETH_HLEN;
 	struct virtnet_info *vi = netdev_priv(dev);
-- 
cgit v1.2.3-58-ga151


From 47879b7322fa0d53c54970099a0ec0aee1d281e2 Mon Sep 17 00:00:00 2001
From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Date: Mon, 8 Jul 2024 19:25:29 +0800
Subject: virtio_net: separate virtnet_rx_resize()

This patch separates two sub-functions from virtnet_rx_resize():

* virtnet_rx_pause
* virtnet_rx_resume

Then the subsequent reset rx for xsk can share these two functions.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-3-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 29 +++++++++++++++++++++--------
 1 file changed, 21 insertions(+), 8 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index d99898e44456..df5b23374c53 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -2665,28 +2665,41 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev)
 	return NETDEV_TX_OK;
 }
 
-static int virtnet_rx_resize(struct virtnet_info *vi,
-			     struct receive_queue *rq, u32 ring_num)
+static void virtnet_rx_pause(struct virtnet_info *vi, struct receive_queue *rq)
 {
 	bool running = netif_running(vi->dev);
-	int err, qindex;
-
-	qindex = rq - vi->rq;
 
 	if (running) {
 		napi_disable(&rq->napi);
 		virtnet_cancel_dim(vi, &rq->dim);
 	}
+}
 
-	err = virtqueue_resize(rq->vq, ring_num, virtnet_rq_unmap_free_buf);
-	if (err)
-		netdev_err(vi->dev, "resize rx fail: rx queue index: %d err: %d\n", qindex, err);
+static void virtnet_rx_resume(struct virtnet_info *vi, struct receive_queue *rq)
+{
+	bool running = netif_running(vi->dev);
 
 	if (!try_fill_recv(vi, rq, GFP_KERNEL))
 		schedule_delayed_work(&vi->refill, 0);
 
 	if (running)
 		virtnet_napi_enable(rq->vq, &rq->napi);
+}
+
+static int virtnet_rx_resize(struct virtnet_info *vi,
+			     struct receive_queue *rq, u32 ring_num)
+{
+	int err, qindex;
+
+	qindex = rq - vi->rq;
+
+	virtnet_rx_pause(vi, rq);
+
+	err = virtqueue_resize(rq->vq, ring_num, virtnet_rq_unmap_free_buf);
+	if (err)
+		netdev_err(vi->dev, "resize rx fail: rx queue index: %d err: %d\n", qindex, err);
+
+	virtnet_rx_resume(vi, rq);
 	return err;
 }
 
-- 
cgit v1.2.3-58-ga151


From 391aa2aad0229d0f34aed6270c5266a0754893a2 Mon Sep 17 00:00:00 2001
From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Date: Mon, 8 Jul 2024 19:25:30 +0800
Subject: virtio_net: separate virtnet_tx_resize()

This patch separates two sub-functions from virtnet_tx_resize():

* virtnet_tx_pause
* virtnet_tx_resume

Then the subsequent virtnet_tx_reset() can share these two functions.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-4-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 35 +++++++++++++++++++++++++++++------
 1 file changed, 29 insertions(+), 6 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index df5b23374c53..7d762614113b 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -2703,12 +2703,11 @@ static int virtnet_rx_resize(struct virtnet_info *vi,
 	return err;
 }
 
-static int virtnet_tx_resize(struct virtnet_info *vi,
-			     struct send_queue *sq, u32 ring_num)
+static void virtnet_tx_pause(struct virtnet_info *vi, struct send_queue *sq)
 {
 	bool running = netif_running(vi->dev);
 	struct netdev_queue *txq;
-	int err, qindex;
+	int qindex;
 
 	qindex = sq - vi->sq;
 
@@ -2729,10 +2728,17 @@ static int virtnet_tx_resize(struct virtnet_info *vi,
 	netif_stop_subqueue(vi->dev, qindex);
 
 	__netif_tx_unlock_bh(txq);
+}
 
-	err = virtqueue_resize(sq->vq, ring_num, virtnet_sq_free_unused_buf);
-	if (err)
-		netdev_err(vi->dev, "resize tx fail: tx queue index: %d err: %d\n", qindex, err);
+static void virtnet_tx_resume(struct virtnet_info *vi, struct send_queue *sq)
+{
+	bool running = netif_running(vi->dev);
+	struct netdev_queue *txq;
+	int qindex;
+
+	qindex = sq - vi->sq;
+
+	txq = netdev_get_tx_queue(vi->dev, qindex);
 
 	__netif_tx_lock_bh(txq);
 	sq->reset = false;
@@ -2741,6 +2747,23 @@ static int virtnet_tx_resize(struct virtnet_info *vi,
 
 	if (running)
 		virtnet_napi_tx_enable(vi, sq->vq, &sq->napi);
+}
+
+static int virtnet_tx_resize(struct virtnet_info *vi, struct send_queue *sq,
+			     u32 ring_num)
+{
+	int qindex, err;
+
+	qindex = sq - vi->sq;
+
+	virtnet_tx_pause(vi, sq);
+
+	err = virtqueue_resize(sq->vq, ring_num, virtnet_sq_free_unused_buf);
+	if (err)
+		netdev_err(vi->dev, "resize tx fail: tx queue index: %d err: %d\n", qindex, err);
+
+	virtnet_tx_resume(vi, sq);
+
 	return err;
 }
 
-- 
cgit v1.2.3-58-ga151


From c86c120fde29dc659a50540f749b6033f7267a62 Mon Sep 17 00:00:00 2001
From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Date: Mon, 8 Jul 2024 19:25:31 +0800
Subject: virtio_net: separate receive_buf

This commit separates the function receive_buf(), then we wrap the logic
of handling the skb to an independent function virtnet_receive_done().
The subsequent commit will reuse it.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-5-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 62 +++++++++++++++++++++++++++---------------------
 1 file changed, 35 insertions(+), 27 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 7d762614113b..abfc84af90ce 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1968,6 +1968,40 @@ static void virtio_skb_set_hash(const struct virtio_net_hdr_v1_hash *hdr_hash,
 	skb_set_hash(skb, __le32_to_cpu(hdr_hash->hash_value), rss_hash_type);
 }
 
+static void virtnet_receive_done(struct virtnet_info *vi, struct receive_queue *rq,
+				 struct sk_buff *skb, u8 flags)
+{
+	struct virtio_net_common_hdr *hdr;
+	struct net_device *dev = vi->dev;
+
+	hdr = skb_vnet_common_hdr(skb);
+	if (dev->features & NETIF_F_RXHASH && vi->has_rss_hash_report)
+		virtio_skb_set_hash(&hdr->hash_v1_hdr, skb);
+
+	if (flags & VIRTIO_NET_HDR_F_DATA_VALID)
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	if (virtio_net_hdr_to_skb(skb, &hdr->hdr,
+				  virtio_is_little_endian(vi->vdev))) {
+		net_warn_ratelimited("%s: bad gso: type: %u, size: %u\n",
+				     dev->name, hdr->hdr.gso_type,
+				     hdr->hdr.gso_size);
+		goto frame_err;
+	}
+
+	skb_record_rx_queue(skb, vq2rxq(rq->vq));
+	skb->protocol = eth_type_trans(skb, dev);
+	pr_debug("Receiving skb proto 0x%04x len %i type %i\n",
+		 ntohs(skb->protocol), skb->len, skb->pkt_type);
+
+	napi_gro_receive(&rq->napi, skb);
+	return;
+
+frame_err:
+	DEV_STATS_INC(dev, rx_frame_errors);
+	dev_kfree_skb(skb);
+}
+
 static void receive_buf(struct virtnet_info *vi, struct receive_queue *rq,
 			void *buf, unsigned int len, void **ctx,
 			unsigned int *xdp_xmit,
@@ -1975,7 +2009,6 @@ static void receive_buf(struct virtnet_info *vi, struct receive_queue *rq,
 {
 	struct net_device *dev = vi->dev;
 	struct sk_buff *skb;
-	struct virtio_net_common_hdr *hdr;
 	u8 flags;
 
 	if (unlikely(len < vi->hdr_len + ETH_HLEN)) {
@@ -2005,32 +2038,7 @@ static void receive_buf(struct virtnet_info *vi, struct receive_queue *rq,
 	if (unlikely(!skb))
 		return;
 
-	hdr = skb_vnet_common_hdr(skb);
-	if (dev->features & NETIF_F_RXHASH && vi->has_rss_hash_report)
-		virtio_skb_set_hash(&hdr->hash_v1_hdr, skb);
-
-	if (flags & VIRTIO_NET_HDR_F_DATA_VALID)
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
-
-	if (virtio_net_hdr_to_skb(skb, &hdr->hdr,
-				  virtio_is_little_endian(vi->vdev))) {
-		net_warn_ratelimited("%s: bad gso: type: %u, size: %u\n",
-				     dev->name, hdr->hdr.gso_type,
-				     hdr->hdr.gso_size);
-		goto frame_err;
-	}
-
-	skb_record_rx_queue(skb, vq2rxq(rq->vq));
-	skb->protocol = eth_type_trans(skb, dev);
-	pr_debug("Receiving skb proto 0x%04x len %i type %i\n",
-		 ntohs(skb->protocol), skb->len, skb->pkt_type);
-
-	napi_gro_receive(&rq->napi, skb);
-	return;
-
-frame_err:
-	DEV_STATS_INC(dev, rx_frame_errors);
-	dev_kfree_skb(skb);
+	virtnet_receive_done(vi, rq, skb, flags);
 }
 
 /* Unlike mergeable buffers, all buffers are allocated to the
-- 
cgit v1.2.3-58-ga151


From 5db481059d79e304596881af0bd195a4418f987a Mon Sep 17 00:00:00 2001
From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Date: Mon, 8 Jul 2024 19:25:32 +0800
Subject: virtio_net: separate receive_mergeable

This commit separates the function receive_mergeable(),
put the logic of appending frag to the skb as an independent function.
The subsequent commit will reuse it.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-6-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 77 +++++++++++++++++++++++++++++-------------------
 1 file changed, 47 insertions(+), 30 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index abfc84af90ce..3c828cdd438b 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1821,6 +1821,49 @@ err_xdp:
 	return NULL;
 }
 
+static struct sk_buff *virtnet_skb_append_frag(struct sk_buff *head_skb,
+					       struct sk_buff *curr_skb,
+					       struct page *page, void *buf,
+					       int len, int truesize)
+{
+	int num_skb_frags;
+	int offset;
+
+	num_skb_frags = skb_shinfo(curr_skb)->nr_frags;
+	if (unlikely(num_skb_frags == MAX_SKB_FRAGS)) {
+		struct sk_buff *nskb = alloc_skb(0, GFP_ATOMIC);
+
+		if (unlikely(!nskb))
+			return NULL;
+
+		if (curr_skb == head_skb)
+			skb_shinfo(curr_skb)->frag_list = nskb;
+		else
+			curr_skb->next = nskb;
+		curr_skb = nskb;
+		head_skb->truesize += nskb->truesize;
+		num_skb_frags = 0;
+	}
+
+	if (curr_skb != head_skb) {
+		head_skb->data_len += len;
+		head_skb->len += len;
+		head_skb->truesize += truesize;
+	}
+
+	offset = buf - page_address(page);
+	if (skb_can_coalesce(curr_skb, num_skb_frags, page, offset)) {
+		put_page(page);
+		skb_coalesce_rx_frag(curr_skb, num_skb_frags - 1,
+				     len, truesize);
+	} else {
+		skb_add_rx_frag(curr_skb, num_skb_frags, page,
+				offset, len, truesize);
+	}
+
+	return curr_skb;
+}
+
 static struct sk_buff *receive_mergeable(struct net_device *dev,
 					 struct virtnet_info *vi,
 					 struct receive_queue *rq,
@@ -1870,8 +1913,6 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
 	if (unlikely(!curr_skb))
 		goto err_skb;
 	while (--num_buf) {
-		int num_skb_frags;
-
 		buf = virtnet_rq_get_buf(rq, &len, &ctx);
 		if (unlikely(!buf)) {
 			pr_debug("%s: rx error: %d buffers out of %d missing\n",
@@ -1896,34 +1937,10 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
 			goto err_skb;
 		}
 
-		num_skb_frags = skb_shinfo(curr_skb)->nr_frags;
-		if (unlikely(num_skb_frags == MAX_SKB_FRAGS)) {
-			struct sk_buff *nskb = alloc_skb(0, GFP_ATOMIC);
-
-			if (unlikely(!nskb))
-				goto err_skb;
-			if (curr_skb == head_skb)
-				skb_shinfo(curr_skb)->frag_list = nskb;
-			else
-				curr_skb->next = nskb;
-			curr_skb = nskb;
-			head_skb->truesize += nskb->truesize;
-			num_skb_frags = 0;
-		}
-		if (curr_skb != head_skb) {
-			head_skb->data_len += len;
-			head_skb->len += len;
-			head_skb->truesize += truesize;
-		}
-		offset = buf - page_address(page);
-		if (skb_can_coalesce(curr_skb, num_skb_frags, page, offset)) {
-			put_page(page);
-			skb_coalesce_rx_frag(curr_skb, num_skb_frags - 1,
-					     len, truesize);
-		} else {
-			skb_add_rx_frag(curr_skb, num_skb_frags, page,
-					offset, len, truesize);
-		}
+		curr_skb  = virtnet_skb_append_frag(head_skb, curr_skb, page,
+						    buf, len, truesize);
+		if (!curr_skb)
+			goto err_skb;
 	}
 
 	ewma_pkt_len_add(&rq->mrg_avg_pkt_len, head_skb->len);
-- 
cgit v1.2.3-58-ga151


From 09d2b3182c8e3a215a9b2a1834f81dd07305989f Mon Sep 17 00:00:00 2001
From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Date: Mon, 8 Jul 2024 19:25:33 +0800
Subject: virtio_net: xsk: bind/unbind xsk for rx

This patch implement the logic of bind/unbind xsk pool to rq.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-7-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 134 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 134 insertions(+)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 3c828cdd438b..c4f5ebd10092 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -25,6 +25,7 @@
 #include <net/net_failover.h>
 #include <net/netdev_rx_queue.h>
 #include <net/netdev_queues.h>
+#include <net/xdp_sock_drv.h>
 
 static int napi_weight = NAPI_POLL_WEIGHT;
 module_param(napi_weight, int, 0444);
@@ -348,6 +349,11 @@ struct receive_queue {
 
 	/* Record the last dma info to free after new pages is allocated. */
 	struct virtnet_rq_dma *last_dma;
+
+	struct xsk_buff_pool *xsk_pool;
+
+	/* xdp rxq used by xsk */
+	struct xdp_rxq_info xsk_rxq_info;
 };
 
 /* This structure can contain rss message with maximum settings for indirection table and keysize
@@ -5026,6 +5032,132 @@ static int virtnet_restore_guest_offloads(struct virtnet_info *vi)
 	return virtnet_set_guest_offloads(vi, offloads);
 }
 
+static int virtnet_rq_bind_xsk_pool(struct virtnet_info *vi, struct receive_queue *rq,
+				    struct xsk_buff_pool *pool)
+{
+	int err, qindex;
+
+	qindex = rq - vi->rq;
+
+	if (pool) {
+		err = xdp_rxq_info_reg(&rq->xsk_rxq_info, vi->dev, qindex, rq->napi.napi_id);
+		if (err < 0)
+			return err;
+
+		err = xdp_rxq_info_reg_mem_model(&rq->xsk_rxq_info,
+						 MEM_TYPE_XSK_BUFF_POOL, NULL);
+		if (err < 0)
+			goto unreg;
+
+		xsk_pool_set_rxq_info(pool, &rq->xsk_rxq_info);
+	}
+
+	virtnet_rx_pause(vi, rq);
+
+	err = virtqueue_reset(rq->vq, virtnet_rq_unmap_free_buf);
+	if (err) {
+		netdev_err(vi->dev, "reset rx fail: rx queue index: %d err: %d\n", qindex, err);
+
+		pool = NULL;
+	}
+
+	rq->xsk_pool = pool;
+
+	virtnet_rx_resume(vi, rq);
+
+	if (pool)
+		return 0;
+
+unreg:
+	xdp_rxq_info_unreg(&rq->xsk_rxq_info);
+	return err;
+}
+
+static int virtnet_xsk_pool_enable(struct net_device *dev,
+				   struct xsk_buff_pool *pool,
+				   u16 qid)
+{
+	struct virtnet_info *vi = netdev_priv(dev);
+	struct receive_queue *rq;
+	struct device *dma_dev;
+	struct send_queue *sq;
+	int err;
+
+	if (vi->hdr_len > xsk_pool_get_headroom(pool))
+		return -EINVAL;
+
+	/* In big_packets mode, xdp cannot work, so there is no need to
+	 * initialize xsk of rq.
+	 */
+	if (vi->big_packets && !vi->mergeable_rx_bufs)
+		return -ENOENT;
+
+	if (qid >= vi->curr_queue_pairs)
+		return -EINVAL;
+
+	sq = &vi->sq[qid];
+	rq = &vi->rq[qid];
+
+	/* xsk assumes that tx and rx must have the same dma device. The af-xdp
+	 * may use one buffer to receive from the rx and reuse this buffer to
+	 * send by the tx. So the dma dev of sq and rq must be the same one.
+	 *
+	 * But vq->dma_dev allows every vq has the respective dma dev. So I
+	 * check the dma dev of vq and sq is the same dev.
+	 */
+	if (virtqueue_dma_dev(rq->vq) != virtqueue_dma_dev(sq->vq))
+		return -EINVAL;
+
+	dma_dev = virtqueue_dma_dev(rq->vq);
+	if (!dma_dev)
+		return -EINVAL;
+
+	err = xsk_pool_dma_map(pool, dma_dev, 0);
+	if (err)
+		goto err_xsk_map;
+
+	err = virtnet_rq_bind_xsk_pool(vi, rq, pool);
+	if (err)
+		goto err_rq;
+
+	return 0;
+
+err_rq:
+	xsk_pool_dma_unmap(pool, 0);
+err_xsk_map:
+	return err;
+}
+
+static int virtnet_xsk_pool_disable(struct net_device *dev, u16 qid)
+{
+	struct virtnet_info *vi = netdev_priv(dev);
+	struct xsk_buff_pool *pool;
+	struct receive_queue *rq;
+	int err;
+
+	if (qid >= vi->curr_queue_pairs)
+		return -EINVAL;
+
+	rq = &vi->rq[qid];
+
+	pool = rq->xsk_pool;
+
+	err = virtnet_rq_bind_xsk_pool(vi, rq, NULL);
+
+	xsk_pool_dma_unmap(pool, 0);
+
+	return err;
+}
+
+static int virtnet_xsk_pool_setup(struct net_device *dev, struct netdev_bpf *xdp)
+{
+	if (xdp->xsk.pool)
+		return virtnet_xsk_pool_enable(dev, xdp->xsk.pool,
+					       xdp->xsk.queue_id);
+	else
+		return virtnet_xsk_pool_disable(dev, xdp->xsk.queue_id);
+}
+
 static int virtnet_xdp_set(struct net_device *dev, struct bpf_prog *prog,
 			   struct netlink_ext_ack *extack)
 {
@@ -5151,6 +5283,8 @@ static int virtnet_xdp(struct net_device *dev, struct netdev_bpf *xdp)
 	switch (xdp->command) {
 	case XDP_SETUP_PROG:
 		return virtnet_xdp_set(dev, xdp->prog, xdp->extack);
+	case XDP_SETUP_XSK_POOL:
+		return virtnet_xsk_pool_setup(dev, xdp);
 	default:
 		return -EINVAL;
 	}
-- 
cgit v1.2.3-58-ga151


From 19a5a7710ee1c60e491e326f9107666435f94044 Mon Sep 17 00:00:00 2001
From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Date: Mon, 8 Jul 2024 19:25:34 +0800
Subject: virtio_net: xsk: support wakeup

xsk wakeup is used to trigger the logic for xsk xmit by xsk framework or
user.

Virtio-net does not support to actively generate an interruption, so it
tries to trigger tx NAPI on the local cpu.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-8-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index c4f5ebd10092..665b4925e545 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1054,6 +1054,29 @@ static void check_sq_full_and_disable(struct virtnet_info *vi,
 	}
 }
 
+static int virtnet_xsk_wakeup(struct net_device *dev, u32 qid, u32 flag)
+{
+	struct virtnet_info *vi = netdev_priv(dev);
+	struct send_queue *sq;
+
+	if (!netif_running(dev))
+		return -ENETDOWN;
+
+	if (qid >= vi->curr_queue_pairs)
+		return -EINVAL;
+
+	sq = &vi->sq[qid];
+
+	if (napi_if_scheduled_mark_missed(&sq->napi))
+		return 0;
+
+	local_bh_disable();
+	virtqueue_napi_schedule(&sq->napi, sq->vq);
+	local_bh_enable();
+
+	return 0;
+}
+
 static int __virtnet_xdp_xmit_one(struct virtnet_info *vi,
 				   struct send_queue *sq,
 				   struct xdp_frame *xdpf)
@@ -5399,6 +5422,7 @@ static const struct net_device_ops virtnet_netdev = {
 	.ndo_vlan_rx_kill_vid = virtnet_vlan_rx_kill_vid,
 	.ndo_bpf		= virtnet_xdp,
 	.ndo_xdp_xmit		= virtnet_xdp_xmit,
+	.ndo_xsk_wakeup         = virtnet_xsk_wakeup,
 	.ndo_features_check	= passthru_features_check,
 	.ndo_get_phys_port_name	= virtnet_get_phys_port_name,
 	.ndo_set_features	= virtnet_set_features,
-- 
cgit v1.2.3-58-ga151


From e9f3962441c0a4d6f16c656e6c8aa02a3ccdd568 Mon Sep 17 00:00:00 2001
From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Date: Mon, 8 Jul 2024 19:25:35 +0800
Subject: virtio_net: xsk: rx: support fill with xsk buffer

Implement the logic of filling rq with XSK buffers.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-9-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 70 +++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 66 insertions(+), 4 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 665b4925e545..a908f5e72677 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -354,6 +354,8 @@ struct receive_queue {
 
 	/* xdp rxq used by xsk */
 	struct xdp_rxq_info xsk_rxq_info;
+
+	struct xdp_buff **xsk_buffs;
 };
 
 /* This structure can contain rss message with maximum settings for indirection table and keysize
@@ -1054,6 +1056,53 @@ static void check_sq_full_and_disable(struct virtnet_info *vi,
 	}
 }
 
+static void sg_fill_dma(struct scatterlist *sg, dma_addr_t addr, u32 len)
+{
+	sg->dma_address = addr;
+	sg->length = len;
+}
+
+static int virtnet_add_recvbuf_xsk(struct virtnet_info *vi, struct receive_queue *rq,
+				   struct xsk_buff_pool *pool, gfp_t gfp)
+{
+	struct xdp_buff **xsk_buffs;
+	dma_addr_t addr;
+	int err = 0;
+	u32 len, i;
+	int num;
+
+	xsk_buffs = rq->xsk_buffs;
+
+	num = xsk_buff_alloc_batch(pool, xsk_buffs, rq->vq->num_free);
+	if (!num)
+		return -ENOMEM;
+
+	len = xsk_pool_get_rx_frame_size(pool) + vi->hdr_len;
+
+	for (i = 0; i < num; ++i) {
+		/* Use the part of XDP_PACKET_HEADROOM as the virtnet hdr space.
+		 * We assume XDP_PACKET_HEADROOM is larger than hdr->len.
+		 * (see function virtnet_xsk_pool_enable)
+		 */
+		addr = xsk_buff_xdp_get_dma(xsk_buffs[i]) - vi->hdr_len;
+
+		sg_init_table(rq->sg, 1);
+		sg_fill_dma(rq->sg, addr, len);
+
+		err = virtqueue_add_inbuf(rq->vq, rq->sg, 1, xsk_buffs[i], gfp);
+		if (err)
+			goto err;
+	}
+
+	return num;
+
+err:
+	for (; i < num; ++i)
+		xsk_buff_free(xsk_buffs[i]);
+
+	return err;
+}
+
 static int virtnet_xsk_wakeup(struct net_device *dev, u32 qid, u32 flag)
 {
 	struct virtnet_info *vi = netdev_priv(dev);
@@ -2245,7 +2294,11 @@ static bool try_fill_recv(struct virtnet_info *vi, struct receive_queue *rq,
 			  gfp_t gfp)
 {
 	int err;
-	bool oom;
+
+	if (rq->xsk_pool) {
+		err = virtnet_add_recvbuf_xsk(vi, rq, rq->xsk_pool, gfp);
+		goto kick;
+	}
 
 	do {
 		if (vi->mergeable_rx_bufs)
@@ -2255,10 +2308,11 @@ static bool try_fill_recv(struct virtnet_info *vi, struct receive_queue *rq,
 		else
 			err = add_recvbuf_small(vi, rq, gfp);
 
-		oom = err == -ENOMEM;
 		if (err)
 			break;
 	} while (rq->vq->num_free);
+
+kick:
 	if (virtqueue_kick_prepare(rq->vq) && virtqueue_notify(rq->vq)) {
 		unsigned long flags;
 
@@ -2267,7 +2321,7 @@ static bool try_fill_recv(struct virtnet_info *vi, struct receive_queue *rq,
 		u64_stats_update_end_irqrestore(&rq->stats.syncp, flags);
 	}
 
-	return !oom;
+	return err != -ENOMEM;
 }
 
 static void skb_recv_done(struct virtqueue *rvq)
@@ -5104,7 +5158,7 @@ static int virtnet_xsk_pool_enable(struct net_device *dev,
 	struct receive_queue *rq;
 	struct device *dma_dev;
 	struct send_queue *sq;
-	int err;
+	int err, size;
 
 	if (vi->hdr_len > xsk_pool_get_headroom(pool))
 		return -EINVAL;
@@ -5135,6 +5189,12 @@ static int virtnet_xsk_pool_enable(struct net_device *dev,
 	if (!dma_dev)
 		return -EINVAL;
 
+	size = virtqueue_get_vring_size(rq->vq);
+
+	rq->xsk_buffs = kvcalloc(size, sizeof(*rq->xsk_buffs), GFP_KERNEL);
+	if (!rq->xsk_buffs)
+		return -ENOMEM;
+
 	err = xsk_pool_dma_map(pool, dma_dev, 0);
 	if (err)
 		goto err_xsk_map;
@@ -5169,6 +5229,8 @@ static int virtnet_xsk_pool_disable(struct net_device *dev, u16 qid)
 
 	xsk_pool_dma_unmap(pool, 0);
 
+	kvfree(rq->xsk_buffs);
+
 	return err;
 }
 
-- 
cgit v1.2.3-58-ga151


From a4e7ba7027012f009f22a68bcfde670f9298d3a4 Mon Sep 17 00:00:00 2001
From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Date: Mon, 8 Jul 2024 19:25:36 +0800
Subject: virtio_net: xsk: rx: support recv small mode

In the process:
1. We may need to copy data to create skb for XDP_PASS.
2. We may need to call xsk_buff_free() to release the buffer.
3. The handle for xdp_buff is difference from the buffer.

If we pushed this logic into existing receive handle(merge and small),
we would have to maintain code scattered inside merge and small (and big).
So I think it is a good choice for us to put the xsk code into an
independent function.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-10-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 176 ++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 168 insertions(+), 8 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index a908f5e72677..c9c65948b71f 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -498,6 +498,12 @@ struct virtio_net_common_hdr {
 };
 
 static void virtnet_sq_free_unused_buf(struct virtqueue *vq, void *buf);
+static int virtnet_xdp_handler(struct bpf_prog *xdp_prog, struct xdp_buff *xdp,
+			       struct net_device *dev,
+			       unsigned int *xdp_xmit,
+			       struct virtnet_rq_stats *stats);
+static void virtnet_receive_done(struct virtnet_info *vi, struct receive_queue *rq,
+				 struct sk_buff *skb, u8 flags);
 
 static bool is_xdp_frame(void *ptr)
 {
@@ -1062,6 +1068,124 @@ static void sg_fill_dma(struct scatterlist *sg, dma_addr_t addr, u32 len)
 	sg->length = len;
 }
 
+static struct xdp_buff *buf_to_xdp(struct virtnet_info *vi,
+				   struct receive_queue *rq, void *buf, u32 len)
+{
+	struct xdp_buff *xdp;
+	u32 bufsize;
+
+	xdp = (struct xdp_buff *)buf;
+
+	bufsize = xsk_pool_get_rx_frame_size(rq->xsk_pool) + vi->hdr_len;
+
+	if (unlikely(len > bufsize)) {
+		pr_debug("%s: rx error: len %u exceeds truesize %u\n",
+			 vi->dev->name, len, bufsize);
+		DEV_STATS_INC(vi->dev, rx_length_errors);
+		xsk_buff_free(xdp);
+		return NULL;
+	}
+
+	xsk_buff_set_size(xdp, len);
+	xsk_buff_dma_sync_for_cpu(xdp);
+
+	return xdp;
+}
+
+static struct sk_buff *xsk_construct_skb(struct receive_queue *rq,
+					 struct xdp_buff *xdp)
+{
+	unsigned int metasize = xdp->data - xdp->data_meta;
+	struct sk_buff *skb;
+	unsigned int size;
+
+	size = xdp->data_end - xdp->data_hard_start;
+	skb = napi_alloc_skb(&rq->napi, size);
+	if (unlikely(!skb)) {
+		xsk_buff_free(xdp);
+		return NULL;
+	}
+
+	skb_reserve(skb, xdp->data_meta - xdp->data_hard_start);
+
+	size = xdp->data_end - xdp->data_meta;
+	memcpy(__skb_put(skb, size), xdp->data_meta, size);
+
+	if (metasize) {
+		__skb_pull(skb, metasize);
+		skb_metadata_set(skb, metasize);
+	}
+
+	xsk_buff_free(xdp);
+
+	return skb;
+}
+
+static struct sk_buff *virtnet_receive_xsk_small(struct net_device *dev, struct virtnet_info *vi,
+						 struct receive_queue *rq, struct xdp_buff *xdp,
+						 unsigned int *xdp_xmit,
+						 struct virtnet_rq_stats *stats)
+{
+	struct bpf_prog *prog;
+	u32 ret;
+
+	ret = XDP_PASS;
+	rcu_read_lock();
+	prog = rcu_dereference(rq->xdp_prog);
+	if (prog)
+		ret = virtnet_xdp_handler(prog, xdp, dev, xdp_xmit, stats);
+	rcu_read_unlock();
+
+	switch (ret) {
+	case XDP_PASS:
+		return xsk_construct_skb(rq, xdp);
+
+	case XDP_TX:
+	case XDP_REDIRECT:
+		return NULL;
+
+	default:
+		/* drop packet */
+		xsk_buff_free(xdp);
+		u64_stats_inc(&stats->drops);
+		return NULL;
+	}
+}
+
+static void virtnet_receive_xsk_buf(struct virtnet_info *vi, struct receive_queue *rq,
+				    void *buf, u32 len,
+				    unsigned int *xdp_xmit,
+				    struct virtnet_rq_stats *stats)
+{
+	struct net_device *dev = vi->dev;
+	struct sk_buff *skb = NULL;
+	struct xdp_buff *xdp;
+	u8 flags;
+
+	len -= vi->hdr_len;
+
+	u64_stats_add(&stats->bytes, len);
+
+	xdp = buf_to_xdp(vi, rq, buf, len);
+	if (!xdp)
+		return;
+
+	if (unlikely(len < ETH_HLEN)) {
+		pr_debug("%s: short packet %i\n", dev->name, len);
+		DEV_STATS_INC(dev, rx_length_errors);
+		xsk_buff_free(xdp);
+		return;
+	}
+
+	flags = ((struct virtio_net_common_hdr *)(xdp->data - vi->hdr_len))->hdr.flags;
+
+	if (!vi->mergeable_rx_bufs)
+		skb = virtnet_receive_xsk_small(dev, vi, rq, xdp, xdp_xmit, stats);
+
+	if (skb)
+		virtnet_receive_done(vi, rq, skb, flags);
+}
+
 static int virtnet_add_recvbuf_xsk(struct virtnet_info *vi, struct receive_queue *rq,
 				   struct xsk_buff_pool *pool, gfp_t gfp)
 {
@@ -2392,32 +2516,68 @@ static void refill_work(struct work_struct *work)
 	}
 }
 
-static int virtnet_receive(struct receive_queue *rq, int budget,
-			   unsigned int *xdp_xmit)
+static int virtnet_receive_xsk_bufs(struct virtnet_info *vi,
+				    struct receive_queue *rq,
+				    int budget,
+				    unsigned int *xdp_xmit,
+				    struct virtnet_rq_stats *stats)
+{
+	unsigned int len;
+	int packets = 0;
+	void *buf;
+
+	while (packets < budget) {
+		buf = virtqueue_get_buf(rq->vq, &len);
+		if (!buf)
+			break;
+
+		virtnet_receive_xsk_buf(vi, rq, buf, len, xdp_xmit, stats);
+		packets++;
+	}
+
+	return packets;
+}
+
+static int virtnet_receive_packets(struct virtnet_info *vi,
+				   struct receive_queue *rq,
+				   int budget,
+				   unsigned int *xdp_xmit,
+				   struct virtnet_rq_stats *stats)
 {
-	struct virtnet_info *vi = rq->vq->vdev->priv;
-	struct virtnet_rq_stats stats = {};
 	unsigned int len;
 	int packets = 0;
 	void *buf;
-	int i;
 
 	if (!vi->big_packets || vi->mergeable_rx_bufs) {
 		void *ctx;
-
 		while (packets < budget &&
 		       (buf = virtnet_rq_get_buf(rq, &len, &ctx))) {
-			receive_buf(vi, rq, buf, len, ctx, xdp_xmit, &stats);
+			receive_buf(vi, rq, buf, len, ctx, xdp_xmit, stats);
 			packets++;
 		}
 	} else {
 		while (packets < budget &&
 		       (buf = virtqueue_get_buf(rq->vq, &len)) != NULL) {
-			receive_buf(vi, rq, buf, len, NULL, xdp_xmit, &stats);
+			receive_buf(vi, rq, buf, len, NULL, xdp_xmit, stats);
 			packets++;
 		}
 	}
 
+	return packets;
+}
+
+static int virtnet_receive(struct receive_queue *rq, int budget,
+			   unsigned int *xdp_xmit)
+{
+	struct virtnet_info *vi = rq->vq->vdev->priv;
+	struct virtnet_rq_stats stats = {};
+	int i, packets;
+
+	if (rq->xsk_pool)
+		packets = virtnet_receive_xsk_bufs(vi, rq, budget, xdp_xmit, &stats);
+	else
+		packets = virtnet_receive_packets(vi, rq, budget, xdp_xmit, &stats);
+
 	if (rq->vq->num_free > min((unsigned int)budget, virtqueue_get_vring_size(rq->vq)) / 2) {
 		if (!try_fill_recv(vi, rq, GFP_ATOMIC)) {
 			spin_lock(&vi->refill_lock);
-- 
cgit v1.2.3-58-ga151


From 99c861b44eb1fb9dfe8776854116a6a9064c19bb Mon Sep 17 00:00:00 2001
From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Date: Mon, 8 Jul 2024 19:25:37 +0800
Subject: virtio_net: xsk: rx: support recv merge mode

Support AF-XDP for merge mode.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-11-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 144 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 144 insertions(+)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index c9c65948b71f..b6323346208d 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -504,6 +504,10 @@ static int virtnet_xdp_handler(struct bpf_prog *xdp_prog, struct xdp_buff *xdp,
 			       struct virtnet_rq_stats *stats);
 static void virtnet_receive_done(struct virtnet_info *vi, struct receive_queue *rq,
 				 struct sk_buff *skb, u8 flags);
+static struct sk_buff *virtnet_skb_append_frag(struct sk_buff *head_skb,
+					       struct sk_buff *curr_skb,
+					       struct page *page, void *buf,
+					       int len, int truesize);
 
 static bool is_xdp_frame(void *ptr)
 {
@@ -984,6 +988,11 @@ static void virtnet_rq_unmap_free_buf(struct virtqueue *vq, void *buf)
 
 	rq = &vi->rq[i];
 
+	if (rq->xsk_pool) {
+		xsk_buff_free((struct xdp_buff *)buf);
+		return;
+	}
+
 	if (!vi->big_packets || vi->mergeable_rx_bufs)
 		virtnet_rq_unmap(rq, buf, 0);
 
@@ -1152,6 +1161,139 @@ static struct sk_buff *virtnet_receive_xsk_small(struct net_device *dev, struct
 	}
 }
 
+static void xsk_drop_follow_bufs(struct net_device *dev,
+				 struct receive_queue *rq,
+				 u32 num_buf,
+				 struct virtnet_rq_stats *stats)
+{
+	struct xdp_buff *xdp;
+	u32 len;
+
+	while (num_buf-- > 1) {
+		xdp = virtqueue_get_buf(rq->vq, &len);
+		if (unlikely(!xdp)) {
+			pr_debug("%s: rx error: %d buffers missing\n",
+				 dev->name, num_buf);
+			DEV_STATS_INC(dev, rx_length_errors);
+			break;
+		}
+		u64_stats_add(&stats->bytes, len);
+		xsk_buff_free(xdp);
+	}
+}
+
+static int xsk_append_merge_buffer(struct virtnet_info *vi,
+				   struct receive_queue *rq,
+				   struct sk_buff *head_skb,
+				   u32 num_buf,
+				   struct virtio_net_hdr_mrg_rxbuf *hdr,
+				   struct virtnet_rq_stats *stats)
+{
+	struct sk_buff *curr_skb;
+	struct xdp_buff *xdp;
+	u32 len, truesize;
+	struct page *page;
+	void *buf;
+
+	curr_skb = head_skb;
+
+	while (--num_buf) {
+		buf = virtqueue_get_buf(rq->vq, &len);
+		if (unlikely(!buf)) {
+			pr_debug("%s: rx error: %d buffers out of %d missing\n",
+				 vi->dev->name, num_buf,
+				 virtio16_to_cpu(vi->vdev,
+						 hdr->num_buffers));
+			DEV_STATS_INC(vi->dev, rx_length_errors);
+			return -EINVAL;
+		}
+
+		u64_stats_add(&stats->bytes, len);
+
+		xdp = buf_to_xdp(vi, rq, buf, len);
+		if (!xdp)
+			goto err;
+
+		buf = napi_alloc_frag(len);
+		if (!buf) {
+			xsk_buff_free(xdp);
+			goto err;
+		}
+
+		memcpy(buf, xdp->data - vi->hdr_len, len);
+
+		xsk_buff_free(xdp);
+
+		page = virt_to_page(buf);
+
+		truesize = len;
+
+		curr_skb  = virtnet_skb_append_frag(head_skb, curr_skb, page,
+						    buf, len, truesize);
+		if (!curr_skb) {
+			put_page(page);
+			goto err;
+		}
+	}
+
+	return 0;
+
+err:
+	xsk_drop_follow_bufs(vi->dev, rq, num_buf, stats);
+	return -EINVAL;
+}
+
+static struct sk_buff *virtnet_receive_xsk_merge(struct net_device *dev, struct virtnet_info *vi,
+						 struct receive_queue *rq, struct xdp_buff *xdp,
+						 unsigned int *xdp_xmit,
+						 struct virtnet_rq_stats *stats)
+{
+	struct virtio_net_hdr_mrg_rxbuf *hdr;
+	struct bpf_prog *prog;
+	struct sk_buff *skb;
+	u32 ret, num_buf;
+
+	hdr = xdp->data - vi->hdr_len;
+	num_buf = virtio16_to_cpu(vi->vdev, hdr->num_buffers);
+
+	ret = XDP_PASS;
+	rcu_read_lock();
+	prog = rcu_dereference(rq->xdp_prog);
+	/* TODO: support multi buffer. */
+	if (prog && num_buf == 1)
+		ret = virtnet_xdp_handler(prog, xdp, dev, xdp_xmit, stats);
+	rcu_read_unlock();
+
+	switch (ret) {
+	case XDP_PASS:
+		skb = xsk_construct_skb(rq, xdp);
+		if (!skb)
+			goto drop_bufs;
+
+		if (xsk_append_merge_buffer(vi, rq, skb, num_buf, hdr, stats)) {
+			dev_kfree_skb(skb);
+			goto drop;
+		}
+
+		return skb;
+
+	case XDP_TX:
+	case XDP_REDIRECT:
+		return NULL;
+
+	default:
+		/* drop packet */
+		xsk_buff_free(xdp);
+	}
+
+drop_bufs:
+	xsk_drop_follow_bufs(dev, rq, num_buf, stats);
+
+drop:
+	u64_stats_inc(&stats->drops);
+	return NULL;
+}
+
 static void virtnet_receive_xsk_buf(struct virtnet_info *vi, struct receive_queue *rq,
 				    void *buf, u32 len,
 				    unsigned int *xdp_xmit,
@@ -1181,6 +1323,8 @@ static void virtnet_receive_xsk_buf(struct virtnet_info *vi, struct receive_queu
 
 	if (!vi->mergeable_rx_bufs)
 		skb = virtnet_receive_xsk_small(dev, vi, rq, xdp, xdp_xmit, stats);
+	else
+		skb = virtnet_receive_xsk_merge(dev, vi, rq, xdp, xdp_xmit, stats);
 
 	if (skb)
 		virtnet_receive_done(vi, rq, skb, flags);
-- 
cgit v1.2.3-58-ga151


From 2c1583290b08c5aa9005178a573be8f329de2976 Mon Sep 17 00:00:00 2001
From: "Kamil Horák (2N)" <kamilh@axis.com>
Date: Fri, 12 Jul 2024 17:07:06 +0200
Subject: net: phy: bcm54811: New link mode for BroadR-Reach
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Introduce a new link mode necessary for 10 MBit single-pair
connection in BroadR-Reach mode on bcm5481x PHY by Broadcom.
This new link mode, 10baseT1BRR, is known as 1BR10 in the Broadcom
terminology. Another link mode to be used is 1BR100 and it is already
present as 100baseT1, because Broadcom's 1BR100 became 100baseT1
(IEEE 802.3bw).

Signed-off-by: Kamil Horák (2N) <kamilh@axis.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20240712150709.3134474-2-kamilh@axis.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/phy-core.c   | 3 ++-
 include/uapi/linux/ethtool.h | 1 +
 net/ethtool/common.c         | 3 +++
 3 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/phy/phy-core.c b/drivers/net/phy/phy-core.c
index a235ea2264a7..1f98b6a96c15 100644
--- a/drivers/net/phy/phy-core.c
+++ b/drivers/net/phy/phy-core.c
@@ -13,7 +13,7 @@
  */
 const char *phy_speed_to_str(int speed)
 {
-	BUILD_BUG_ON_MSG(__ETHTOOL_LINK_MODE_MASK_NBITS != 102,
+	BUILD_BUG_ON_MSG(__ETHTOOL_LINK_MODE_MASK_NBITS != 103,
 		"Enum ethtool_link_mode_bit_indices and phylib are out of sync. "
 		"If a speed or mode has been added please update phy_speed_to_str "
 		"and the PHY settings array.\n");
@@ -266,6 +266,7 @@ static const struct phy_setting settings[] = {
 	PHY_SETTING(     10, FULL,     10baseT1S_Full		),
 	PHY_SETTING(     10, HALF,     10baseT1S_Half		),
 	PHY_SETTING(     10, HALF,     10baseT1S_P2MP_Half	),
+	PHY_SETTING(     10, FULL,     10baseT1BRR_Full		),
 };
 #undef PHY_SETTING
 
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 230110b97029..4a0a6e703483 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -2054,6 +2054,7 @@ enum ethtool_link_mode_bit_indices {
 	ETHTOOL_LINK_MODE_10baseT1S_Full_BIT		 = 99,
 	ETHTOOL_LINK_MODE_10baseT1S_Half_BIT		 = 100,
 	ETHTOOL_LINK_MODE_10baseT1S_P2MP_Half_BIT	 = 101,
+	ETHTOOL_LINK_MODE_10baseT1BRR_Full_BIT		 = 102,
 
 	/* must be last entry */
 	__ETHTOOL_LINK_MODE_MASK_NBITS
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 67d06cd002a5..5f99a32e4fe5 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -211,6 +211,7 @@ const char link_mode_names[][ETH_GSTRING_LEN] = {
 	__DEFINE_LINK_MODE_NAME(10, T1S, Full),
 	__DEFINE_LINK_MODE_NAME(10, T1S, Half),
 	__DEFINE_LINK_MODE_NAME(10, T1S_P2MP, Half),
+	__DEFINE_LINK_MODE_NAME(10, T1BRR, Full),
 };
 static_assert(ARRAY_SIZE(link_mode_names) == __ETHTOOL_LINK_MODE_MASK_NBITS);
 
@@ -251,6 +252,7 @@ static_assert(ARRAY_SIZE(link_mode_names) == __ETHTOOL_LINK_MODE_MASK_NBITS);
 #define __LINK_MODE_LANES_T1S_P2MP	1
 #define __LINK_MODE_LANES_VR8		8
 #define __LINK_MODE_LANES_DR8_2		8
+#define __LINK_MODE_LANES_T1BRR		1
 
 #define __DEFINE_LINK_MODE_PARAMS(_speed, _type, _duplex)	\
 	[ETHTOOL_LINK_MODE(_speed, _type, _duplex)] = {		\
@@ -374,6 +376,7 @@ const struct link_mode_info link_mode_params[] = {
 	__DEFINE_LINK_MODE_PARAMS(10, T1S, Full),
 	__DEFINE_LINK_MODE_PARAMS(10, T1S, Half),
 	__DEFINE_LINK_MODE_PARAMS(10, T1S_P2MP, Half),
+	__DEFINE_LINK_MODE_PARAMS(10, T1BRR, Full),
 };
 static_assert(ARRAY_SIZE(link_mode_params) == __ETHTOOL_LINK_MODE_MASK_NBITS);
 
-- 
cgit v1.2.3-58-ga151


From ff253875ff3bab75214c4e94352956dc02a0d965 Mon Sep 17 00:00:00 2001
From: "Kamil Horák (2N)" <kamilh@axis.com>
Date: Fri, 12 Jul 2024 17:07:07 +0200
Subject: net: phy: bcm54811: Add LRE registers definitions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add the definitions of LRE registers for Broadcom BCM5481x PHY

Signed-off-by: Kamil Horák (2N) <kamilh@axis.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://patch.msgid.link/20240712150709.3134474-3-kamilh@axis.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/brcmphy.h | 88 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 88 insertions(+)

diff --git a/include/linux/brcmphy.h b/include/linux/brcmphy.h
index 1394ba302367..028b3e00378e 100644
--- a/include/linux/brcmphy.h
+++ b/include/linux/brcmphy.h
@@ -271,12 +271,100 @@
 #define BCM5482_SSD_SGMII_SLAVE_EN	0x0002	/* Slave mode enable */
 #define BCM5482_SSD_SGMII_SLAVE_AD	0x0001	/* Slave auto-detection */
 
+/* BroadR-Reach LRE Registers. */
+#define MII_BCM54XX_LRECR		0x00	/* LRE Control Register                    */
+#define MII_BCM54XX_LRESR		0x01	/* LRE Status Register                     */
+#define MII_BCM54XX_LREPHYSID1		0x02	/* LRE PHYS ID 1                           */
+#define MII_BCM54XX_LREPHYSID2		0x03	/* LRE PHYS ID 2                           */
+#define MII_BCM54XX_LREANAA		0x04	/* LDS Auto-Negotiation Advertised Ability */
+#define MII_BCM54XX_LREANAC		0x05	/* LDS Auto-Negotiation Advertised Control */
+#define MII_BCM54XX_LREANPT		0x06	/* LDS Ability Next Page Transmit          */
+#define MII_BCM54XX_LRELPA		0x07	/* LDS Link Partner Ability                */
+#define MII_BCM54XX_LRELPNPM		0x08	/* LDS Link Partner Next Page Message      */
+#define MII_BCM54XX_LRELPNPC		0x09	/* LDS Link Partner Next Page Control      */
+#define MII_BCM54XX_LRELDSE		0x0a	/* LDS Expansion Register                  */
+#define MII_BCM54XX_LREES		0x0f	/* LRE Extended Status                     */
+
+/* LRE control register. */
+#define LRECR_RESET			0x8000	/* Reset to default state      */
+#define LRECR_LOOPBACK			0x4000	/* Internal Loopback           */
+#define LRECR_LDSRES			0x2000	/* Restart LDS Process         */
+#define LRECR_LDSEN			0x1000	/* LDS Enable                  */
+#define LRECR_PDOWN			0x0800	/* Enable low power state      */
+#define LRECR_ISOLATE			0x0400	/* Isolate data paths from MII */
+#define LRECR_SPEED100			0x0200	/* Select 100 Mbps             */
+#define LRECR_SPEED10			0x0000	/* Select 10 Mbps              */
+#define LRECR_4PAIRS			0x0020	/* Select 4 Pairs              */
+#define LRECR_2PAIRS			0x0010	/* Select 2 Pairs              */
+#define LRECR_1PAIR			0x0000	/* Select 1 Pair               */
+#define LRECR_MASTER			0x0008	/* Force Master when LDS disabled */
+#define LRECR_SLAVE			0x0000	/* Force Slave when LDS disabled  */
+
+/* LRE status register. */
+#define LRESR_100_1PAIR			0x2000	/* Can do 100Mbps 1 Pair       */
+#define LRESR_100_4PAIR			0x1000	/* Can do 100Mbps 4 Pairs      */
+#define LRESR_100_2PAIR			0x0800	/* Can do 100Mbps 2 Pairs      */
+#define LRESR_10_2PAIR			0x0400	/* Can do 10Mbps 2 Pairs       */
+#define LRESR_10_1PAIR			0x0200	/* Can do 10Mbps 1 Pair        */
+#define LRESR_ESTATEN			0x0100	/* Extended Status in R15      */
+#define LRESR_RESV			0x0080	/* Unused...                   */
+#define LRESR_MFPS			0x0040	/* Can suppress Management Frames Preamble */
+#define LRESR_LDSCOMPLETE		0x0020	/* LDS Auto-negotiation complete */
+#define LRESR_8023			0x0010	/* Has IEEE 802.3 Support      */
+#define LRESR_LDSABILITY		0x0008	/* LDS auto-negotiation capable */
+#define LRESR_LSTATUS			0x0004	/* Link status                 */
+#define LRESR_JCD			0x0002	/* Jabber detected             */
+#define LRESR_ERCAP			0x0001	/* Ext-reg capability          */
+
+/* LDS Auto-Negotiation Advertised Ability. */
+#define LREANAA_PAUSE_ASYM		0x8000	/* Can pause asymmetrically    */
+#define LREANAA_PAUSE			0x4000	/* Can pause                   */
+#define LREANAA_100_1PAIR		0x0020	/* Can do 100Mbps 1 Pair       */
+#define LREANAA_100_4PAIR		0x0010	/* Can do 100Mbps 4 Pair       */
+#define LREANAA_100_2PAIR		0x0008	/* Can do 100Mbps 2 Pair       */
+#define LREANAA_10_2PAIR		0x0004	/* Can do 10Mbps 2 Pair        */
+#define LREANAA_10_1PAIR		0x0002	/* Can do 10Mbps 1 Pair        */
+
+#define LRE_ADVERTISE_FULL		(LREANAA_100_1PAIR | LREANAA_100_4PAIR | \
+					 LREANAA_100_2PAIR | LREANAA_10_2PAIR | \
+					 LREANAA_10_1PAIR)
+
+#define LRE_ADVERTISE_ALL		LRE_ADVERTISE_FULL
+
+/* LDS Link Partner Ability. */
+#define LRELPA_PAUSE_ASYM		0x8000	/* Supports asymmetric pause   */
+#define LRELPA_PAUSE			0x4000	/* Supports pause capability   */
+#define LRELPA_100_1PAIR		0x0020	/* 100Mbps 1 Pair capable      */
+#define LRELPA_100_4PAIR		0x0010	/* 100Mbps 4 Pair capable      */
+#define LRELPA_100_2PAIR		0x0008	/* 100Mbps 2 Pair capable      */
+#define LRELPA_10_2PAIR			0x0004	/* 10Mbps 2 Pair capable       */
+#define LRELPA_10_1PAIR			0x0002	/* 10Mbps 1 Pair capable       */
+
+/* LDS Expansion register. */
+#define LDSE_DOWNGRADE			0x8000	/* Can do LDS Speed Downgrade  */
+#define LDSE_MASTER			0x4000	/* Master / Slave              */
+#define LDSE_PAIRS_MASK			0x3000	/* Pair Count Mask             */
+#define LDSE_PAIRS_SHIFT		12
+#define LDSE_4PAIRS			(2 << LDSE_PAIRS_SHIFT)	/* 4 Pairs Connection */
+#define LDSE_2PAIRS			(1 << LDSE_PAIRS_SHIFT)	/* 2 Pairs Connection */
+#define LDSE_1PAIR			(0 << LDSE_PAIRS_SHIFT)	/* 1 Pair  Connection */
+#define LDSE_CABLEN_MASK		0x0FFF	/* Cable Length Mask           */
+
 /* BCM54810 Registers */
 #define BCM54810_EXP_BROADREACH_LRE_MISC_CTL	(MII_BCM54XX_EXP_SEL_ER + 0x90)
 #define BCM54810_EXP_BROADREACH_LRE_MISC_CTL_EN	(1 << 0)
 #define BCM54810_SHD_CLK_CTL			0x3
 #define BCM54810_SHD_CLK_CTL_GTXCLK_EN		(1 << 9)
 
+/* BCM54811 Registers */
+#define BCM54811_EXP_BROADREACH_LRE_OVERLAY_CTL	(MII_BCM54XX_EXP_SEL_ER + 0x9A)
+/* Access Control Override Enable */
+#define BCM54811_EXP_BROADREACH_LRE_OVERLAY_CTL_EN		BIT(15)
+/* Access Control Override Value */
+#define BCM54811_EXP_BROADREACH_LRE_OVERLAY_CTL_OVERRIDE_VAL	BIT(14)
+/* Access Control Value */
+#define BCM54811_EXP_BROADREACH_LRE_OVERLAY_CTL_VAL		BIT(13)
+
 /* BCM54612E Registers */
 #define BCM54612E_EXP_SPARE0		(MII_BCM54XX_EXP_SEL_ETC + 0x34)
 #define BCM54612E_LED4_CLK125OUT_EN	(1 << 1)
-- 
cgit v1.2.3-58-ga151


From 775631d7845b34996ceb0d975725c16d9a414430 Mon Sep 17 00:00:00 2001
From: "Kamil Horák (2N)" <kamilh@axis.com>
Date: Fri, 12 Jul 2024 17:07:08 +0200
Subject: dt-bindings: ethernet-phy: add optional brr-mode flag
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There is a group of PHY chips supporting BroadR-Reach link modes in
a manner allowing for more or less identical register usage as standard
Clause 22 PHY.
These chips support standard Ethernet link modes as well, however, the
circuitry is mutually exclusive and cannot be auto-detected.
The link modes in question are 100Base-T1 as defined in IEEE802.3bw,
based on Broadcom's 1BR-100 link mode, and newly defined 10Base-T1BRR
(1BR-10 in Broadcom documents).

Add optional brr-mode flag to switch the PHY to BroadR-Reach mode.

Signed-off-by: Kamil Horák (2N) <kamilh@axis.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://patch.msgid.link/20240712150709.3134474-4-kamilh@axis.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/devicetree/bindings/net/ethernet-phy.yaml | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/ethernet-phy.yaml b/Documentation/devicetree/bindings/net/ethernet-phy.yaml
index 8fb2a6ee7e5b..d9b62741a225 100644
--- a/Documentation/devicetree/bindings/net/ethernet-phy.yaml
+++ b/Documentation/devicetree/bindings/net/ethernet-phy.yaml
@@ -93,6 +93,14 @@ properties:
       the turn around line low at end of the control phase of the
       MDIO transaction.
 
+  brr-mode:
+    $ref: /schemas/types.yaml#/definitions/flag
+    description:
+      If set, indicates the network cable interface is an alternative one as
+      defined in the BroadR-Reach link mode specification under 1BR-100 and
+      1BR-10 names. The PHY must be configured to operate in BroadR-Reach mode
+      by software.
+
   clocks:
     maxItems: 1
     description:
-- 
cgit v1.2.3-58-ga151


From 03ab6c244bb0ced0a67b86399a282987740bb116 Mon Sep 17 00:00:00 2001
From: "Kamil Horák (2N)" <kamilh@axis.com>
Date: Fri, 12 Jul 2024 17:07:09 +0200
Subject: net: phy: bcm-phy-lib: Implement BroadR-Reach link modes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Implement single-pair BroadR-Reach modes on bcm5481x PHY by Broadcom.
Create set of functions alternative to IEEE 802.3 to handle
configuration of these modes on compatible Broadcom PHYs.
There is only subset of capabilities supported because of limited
collection of hardware available for the development.
For BroadR-Reach capable PHYs, the LRE (Long Reach Ethernet)
alternative register set is handled. Only bcm54811 PHY is verified,
for bcm54810, there is some support possible but untested. There
is no auto-negotiation of the link parameters (called LDS in the
Broadcom terminology, Long-Distance Signaling) for bcm54811.
It should be possible to enable LDS for bcm54810.

Signed-off-by: Kamil Horák (2N) <kamilh@axis.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://patch.msgid.link/20240712150709.3134474-5-kamilh@axis.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/bcm-phy-lib.c | 115 ++++++++++++
 drivers/net/phy/bcm-phy-lib.h |   4 +
 drivers/net/phy/broadcom.c    | 403 +++++++++++++++++++++++++++++++++++++++---
 3 files changed, 494 insertions(+), 28 deletions(-)

diff --git a/drivers/net/phy/bcm-phy-lib.c b/drivers/net/phy/bcm-phy-lib.c
index 876f28fd8256..6c52f7dda514 100644
--- a/drivers/net/phy/bcm-phy-lib.c
+++ b/drivers/net/phy/bcm-phy-lib.c
@@ -794,6 +794,49 @@ out:
 	return ret;
 }
 
+static int bcm_setup_lre_forced(struct phy_device *phydev)
+{
+	u16 ctl = 0;
+
+	phydev->pause = 0;
+	phydev->asym_pause = 0;
+
+	if (phydev->speed == SPEED_100)
+		ctl |= LRECR_SPEED100;
+
+	if (phydev->duplex != DUPLEX_FULL)
+		return -EOPNOTSUPP;
+
+	return phy_modify(phydev, MII_BCM54XX_LRECR, LRECR_SPEED100, ctl);
+}
+
+/**
+ * bcm_linkmode_adv_to_lre_adv_t - translate linkmode advertisement to LDS
+ * @advertising: the linkmode advertisement settings
+ * Return: LDS Auto-Negotiation Advertised Ability register value
+ *
+ * A small helper function that translates linkmode advertisement
+ * settings to phy LDS autonegotiation advertisements for the
+ * MII_BCM54XX_LREANAA register of Broadcom PHYs capable of LDS
+ */
+static u32 bcm_linkmode_adv_to_lre_adv_t(unsigned long *advertising)
+{
+	u32 result = 0;
+
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_10baseT1BRR_Full_BIT,
+			      advertising))
+		result |= LREANAA_10_1PAIR;
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_100baseT1_Full_BIT,
+			      advertising))
+		result |= LREANAA_100_1PAIR;
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_Pause_BIT, advertising))
+		result |= LRELPA_PAUSE;
+	if (linkmode_test_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT, advertising))
+		result |= LRELPA_PAUSE_ASYM;
+
+	return result;
+}
+
 int bcm_phy_cable_test_start(struct phy_device *phydev)
 {
 	return _bcm_phy_cable_test_start(phydev, false);
@@ -1066,6 +1109,78 @@ int bcm_phy_led_brightness_set(struct phy_device *phydev,
 }
 EXPORT_SYMBOL_GPL(bcm_phy_led_brightness_set);
 
+int bcm_setup_lre_master_slave(struct phy_device *phydev)
+{
+	u16 ctl = 0;
+
+	switch (phydev->master_slave_set) {
+	case MASTER_SLAVE_CFG_MASTER_PREFERRED:
+	case MASTER_SLAVE_CFG_MASTER_FORCE:
+		ctl = LRECR_MASTER;
+		break;
+	case MASTER_SLAVE_CFG_SLAVE_PREFERRED:
+	case MASTER_SLAVE_CFG_SLAVE_FORCE:
+		break;
+	case MASTER_SLAVE_CFG_UNKNOWN:
+	case MASTER_SLAVE_CFG_UNSUPPORTED:
+		return 0;
+	default:
+		phydev_warn(phydev, "Unsupported Master/Slave mode\n");
+		return -EOPNOTSUPP;
+	}
+
+	return phy_modify_changed(phydev, MII_BCM54XX_LRECR, LRECR_MASTER, ctl);
+}
+EXPORT_SYMBOL_GPL(bcm_setup_lre_master_slave);
+
+int bcm_config_lre_aneg(struct phy_device *phydev, bool changed)
+{
+	int err;
+
+	if (genphy_config_eee_advert(phydev))
+		changed = true;
+
+	err = bcm_setup_lre_master_slave(phydev);
+	if (err < 0)
+		return err;
+	else if (err)
+		changed = true;
+
+	if (phydev->autoneg != AUTONEG_ENABLE)
+		return bcm_setup_lre_forced(phydev);
+
+	err = bcm_config_lre_advert(phydev);
+	if (err < 0)
+		return err;
+	else if (err)
+		changed = true;
+
+	return genphy_check_and_restart_aneg(phydev, changed);
+}
+EXPORT_SYMBOL_GPL(bcm_config_lre_aneg);
+
+/**
+ * bcm_config_lre_advert - sanitize and advertise Long-Distance Signaling
+ *  auto-negotiation parameters
+ * @phydev: target phy_device struct
+ * Return:  0 if the PHY's advertisement hasn't changed, < 0 on error,
+ *          > 0 if it has changed
+ *
+ * Writes MII_BCM54XX_LREANAA with the appropriate values. The values are to be
+ *   sanitized before, to make sure we only advertise what is supported.
+ *  The sanitization is done already in phy_ethtool_ksettings_set()
+ */
+int bcm_config_lre_advert(struct phy_device *phydev)
+{
+	u32 adv = bcm_linkmode_adv_to_lre_adv_t(phydev->advertising);
+
+	/* Setup BroadR-Reach mode advertisement */
+	return phy_modify_changed(phydev, MII_BCM54XX_LREANAA,
+				 LRE_ADVERTISE_ALL | LREANAA_PAUSE |
+				 LREANAA_PAUSE_ASYM, adv);
+}
+EXPORT_SYMBOL_GPL(bcm_config_lre_advert);
+
 MODULE_DESCRIPTION("Broadcom PHY Library");
 MODULE_LICENSE("GPL v2");
 MODULE_AUTHOR("Broadcom Corporation");
diff --git a/drivers/net/phy/bcm-phy-lib.h b/drivers/net/phy/bcm-phy-lib.h
index b52189e45a84..bceddbc860eb 100644
--- a/drivers/net/phy/bcm-phy-lib.h
+++ b/drivers/net/phy/bcm-phy-lib.h
@@ -121,4 +121,8 @@ irqreturn_t bcm_phy_wol_isr(int irq, void *dev_id);
 int bcm_phy_led_brightness_set(struct phy_device *phydev,
 			       u8 index, enum led_brightness value);
 
+int bcm_setup_lre_master_slave(struct phy_device *phydev);
+int bcm_config_lre_aneg(struct phy_device *phydev, bool changed);
+int bcm_config_lre_advert(struct phy_device *phydev);
+
 #endif /* _LINUX_BCM_PHY_LIB_H */
diff --git a/drivers/net/phy/broadcom.c b/drivers/net/phy/broadcom.c
index 370e4ed45098..ddded162c44c 100644
--- a/drivers/net/phy/broadcom.c
+++ b/drivers/net/phy/broadcom.c
@@ -5,6 +5,8 @@
  *	Broadcom BCM5411, BCM5421 and BCM5461 Gigabit Ethernet
  *	transceivers.
  *
+ *	Broadcom BCM54810, BCM54811 BroadR-Reach transceivers.
+ *
  *	Copyright (c) 2006  Maciej W. Rozycki
  *
  *	Inspired by code written by Amy Fong.
@@ -36,6 +38,29 @@ struct bcm54xx_phy_priv {
 	struct bcm_ptp_private *ptp;
 	int	wake_irq;
 	bool	wake_irq_enabled;
+	bool	brr_mode;
+};
+
+/* Link modes for BCM58411 PHY */
+static const int bcm54811_linkmodes[] = {
+	ETHTOOL_LINK_MODE_100baseT1_Full_BIT,
+	ETHTOOL_LINK_MODE_10baseT1BRR_Full_BIT,
+	ETHTOOL_LINK_MODE_1000baseT_Full_BIT,
+	ETHTOOL_LINK_MODE_1000baseX_Full_BIT,
+	ETHTOOL_LINK_MODE_1000baseT_Half_BIT,
+	ETHTOOL_LINK_MODE_100baseT_Full_BIT,
+	ETHTOOL_LINK_MODE_100baseT_Half_BIT,
+	ETHTOOL_LINK_MODE_10baseT_Full_BIT,
+	ETHTOOL_LINK_MODE_10baseT_Half_BIT
+};
+
+/* Long-Distance Signaling (BroadR-Reach mode aneg) relevant linkmode bits */
+static const int lds_br_bits[] = {
+	ETHTOOL_LINK_MODE_Autoneg_BIT,
+	ETHTOOL_LINK_MODE_Pause_BIT,
+	ETHTOOL_LINK_MODE_Asym_Pause_BIT,
+	ETHTOOL_LINK_MODE_10baseT1BRR_Full_BIT,
+	ETHTOOL_LINK_MODE_100baseT1_Full_BIT
 };
 
 static bool bcm54xx_phy_can_wakeup(struct phy_device *phydev)
@@ -347,6 +372,61 @@ static void bcm54xx_ptp_config_init(struct phy_device *phydev)
 		bcm_ptp_config_init(phydev);
 }
 
+static int bcm5481x_set_brrmode(struct phy_device *phydev, bool on)
+{
+	int reg;
+	int err;
+	u16 val;
+
+	reg = bcm_phy_read_exp(phydev, BCM54810_EXP_BROADREACH_LRE_MISC_CTL);
+
+	if (reg < 0)
+		return reg;
+
+	if (on)
+		reg |= BCM54810_EXP_BROADREACH_LRE_MISC_CTL_EN;
+	else
+		reg &= ~BCM54810_EXP_BROADREACH_LRE_MISC_CTL_EN;
+
+	err = bcm_phy_write_exp(phydev,
+				BCM54810_EXP_BROADREACH_LRE_MISC_CTL, reg);
+	if (err)
+		return err;
+
+	/* Ensure LRE or IEEE register set is accessed according to the brr
+	 *  on/off, thus set the override
+	 */
+	val = BCM54811_EXP_BROADREACH_LRE_OVERLAY_CTL_EN;
+	if (!on)
+		val |= BCM54811_EXP_BROADREACH_LRE_OVERLAY_CTL_OVERRIDE_VAL;
+
+	return bcm_phy_write_exp(phydev,
+				 BCM54811_EXP_BROADREACH_LRE_OVERLAY_CTL, val);
+}
+
+static int bcm54811_config_init(struct phy_device *phydev)
+{
+	struct bcm54xx_phy_priv *priv = phydev->priv;
+	int err, reg;
+
+	/* Enable CLK125 MUX on LED4 if ref clock is enabled. */
+	if (!(phydev->dev_flags & PHY_BRCM_RX_REFCLK_UNUSED)) {
+		reg = bcm_phy_read_exp(phydev, BCM54612E_EXP_SPARE0);
+		if (reg < 0)
+			return reg;
+		err = bcm_phy_write_exp(phydev, BCM54612E_EXP_SPARE0,
+					BCM54612E_LED4_CLK125OUT_EN | reg);
+		if (err < 0)
+			return err;
+	}
+
+	/* With BCM54811, BroadR-Reach implies no autoneg */
+	if (priv->brr_mode)
+		phydev->autoneg = 0;
+
+	return bcm5481x_set_brrmode(phydev, priv->brr_mode);
+}
+
 static int bcm54xx_config_init(struct phy_device *phydev)
 {
 	int reg, err, val;
@@ -399,6 +479,9 @@ static int bcm54xx_config_init(struct phy_device *phydev)
 					BCM54810_EXP_BROADREACH_LRE_MISC_CTL,
 					val);
 		break;
+	case PHY_ID_BCM54811:
+		err = bcm54811_config_init(phydev);
+		break;
 	}
 	if (err)
 		return err;
@@ -553,52 +636,117 @@ static int bcm54810_write_mmd(struct phy_device *phydev, int devnum, u16 regnum,
 	return -EOPNOTSUPP;
 }
 
-static int bcm54811_config_init(struct phy_device *phydev)
+
+/**
+ * bcm5481x_read_abilities - read PHY abilities from LRESR or Clause 22
+ * (BMSR) registers, based on whether the PHY is in BroadR-Reach or IEEE mode
+ * @phydev: target phy_device struct
+ *
+ * Description: Reads the PHY's abilities and populates phydev->supported
+ * accordingly. The register to read the abilities from is determined by
+ * the brr mode setting of the PHY as read from the device tree.
+ * Note that the LRE and IEEE sets of abilities are disjunct, in other words,
+ * not only the link modes differ, but also the auto-negotiation and
+ * master-slave setup is controlled differently.
+ *
+ * Returns: 0 on success, < 0 on failure
+ */
+static int bcm5481x_read_abilities(struct phy_device *phydev)
 {
-	int err, reg;
+	struct device_node *np = phydev->mdio.dev.of_node;
+	struct bcm54xx_phy_priv *priv = phydev->priv;
+	int i, val, err;
 
-	/* Disable BroadR-Reach function. */
-	reg = bcm_phy_read_exp(phydev, BCM54810_EXP_BROADREACH_LRE_MISC_CTL);
-	reg &= ~BCM54810_EXP_BROADREACH_LRE_MISC_CTL_EN;
-	err = bcm_phy_write_exp(phydev, BCM54810_EXP_BROADREACH_LRE_MISC_CTL,
-				reg);
-	if (err < 0)
+	for (i = 0; i < ARRAY_SIZE(bcm54811_linkmodes); i++)
+		linkmode_clear_bit(bcm54811_linkmodes[i], phydev->supported);
+
+	priv->brr_mode = of_property_read_bool(np, "brr-mode");
+
+	/* Set BroadR-Reach mode as configured in the DT. */
+	err = bcm5481x_set_brrmode(phydev, priv->brr_mode);
+	if (err)
 		return err;
 
-	err = bcm54xx_config_init(phydev);
+	if (priv->brr_mode) {
+		linkmode_set_bit_array(phy_basic_ports_array,
+				       ARRAY_SIZE(phy_basic_ports_array),
+				       phydev->supported);
 
-	/* Enable CLK125 MUX on LED4 if ref clock is enabled. */
-	if (!(phydev->dev_flags & PHY_BRCM_RX_REFCLK_UNUSED)) {
-		reg = bcm_phy_read_exp(phydev, BCM54612E_EXP_SPARE0);
-		err = bcm_phy_write_exp(phydev, BCM54612E_EXP_SPARE0,
-					BCM54612E_LED4_CLK125OUT_EN | reg);
-		if (err < 0)
-			return err;
+		val = phy_read(phydev, MII_BCM54XX_LRESR);
+		if (val < 0)
+			return val;
+
+		linkmode_mod_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+				 phydev->supported,
+				 val & LRESR_LDSABILITY);
+		linkmode_mod_bit(ETHTOOL_LINK_MODE_100baseT1_Full_BIT,
+				 phydev->supported,
+				 val & LRESR_100_1PAIR);
+		linkmode_mod_bit(ETHTOOL_LINK_MODE_10baseT1BRR_Full_BIT,
+				 phydev->supported,
+				 val & LRESR_10_1PAIR);
+		return 0;
 	}
 
-	return err;
+	return genphy_read_abilities(phydev);
 }
 
-static int bcm5481_config_aneg(struct phy_device *phydev)
+static int bcm5481x_config_delay_swap(struct phy_device *phydev)
 {
 	struct device_node *np = phydev->mdio.dev.of_node;
-	int ret;
-
-	/* Aneg firstly. */
-	ret = genphy_config_aneg(phydev);
 
-	/* Then we can set up the delay. */
+	/* Set up the delay. */
 	bcm54xx_config_clock_delay(phydev);
 
 	if (of_property_read_bool(np, "enet-phy-lane-swap")) {
 		/* Lane Swap - Undocumented register...magic! */
-		ret = bcm_phy_write_exp(phydev, MII_BCM54XX_EXP_SEL_ER + 0x9,
-					0x11B);
+		int ret = bcm_phy_write_exp(phydev,
+					    MII_BCM54XX_EXP_SEL_ER + 0x9,
+					    0x11B);
 		if (ret < 0)
 			return ret;
 	}
 
-	return ret;
+	return 0;
+}
+
+static int bcm5481_config_aneg(struct phy_device *phydev)
+{
+	struct bcm54xx_phy_priv *priv = phydev->priv;
+	int ret;
+
+	/* Aneg firstly. */
+	if (priv->brr_mode)
+		ret = bcm_config_lre_aneg(phydev, false);
+	else
+		ret = genphy_config_aneg(phydev);
+
+	if (ret)
+		return ret;
+
+	/* Then we can set up the delay and swap. */
+	return bcm5481x_config_delay_swap(phydev);
+}
+
+static int bcm54811_config_aneg(struct phy_device *phydev)
+{
+	struct bcm54xx_phy_priv *priv = phydev->priv;
+	int ret;
+
+	/* Aneg firstly. */
+	if (priv->brr_mode) {
+		/* BCM54811 is only capable of autonegotiation in IEEE mode */
+		phydev->autoneg = 0;
+		ret = bcm_config_lre_aneg(phydev, false);
+	} else {
+		ret = genphy_config_aneg(phydev);
+	}
+
+	if (ret)
+		return ret;
+
+	/* Then we can set up the delay and swap. */
+	return bcm5481x_config_delay_swap(phydev);
 }
 
 struct bcm54616s_phy_priv {
@@ -1062,6 +1210,203 @@ static void bcm54xx_link_change_notify(struct phy_device *phydev)
 	bcm_phy_write_exp(phydev, MII_BCM54XX_EXP_EXP08, ret);
 }
 
+static int lre_read_master_slave(struct phy_device *phydev)
+{
+	int cfg = MASTER_SLAVE_CFG_UNKNOWN, state;
+	int val;
+
+	/* In BroadR-Reach mode we are always capable of master-slave
+	 *  and there is no preferred master or slave configuration
+	 */
+	phydev->master_slave_get = MASTER_SLAVE_CFG_UNKNOWN;
+	phydev->master_slave_state = MASTER_SLAVE_STATE_UNKNOWN;
+
+	val = phy_read(phydev, MII_BCM54XX_LRECR);
+	if (val < 0)
+		return val;
+
+	if ((val & LRECR_LDSEN) == 0) {
+		if (val & LRECR_MASTER)
+			cfg = MASTER_SLAVE_CFG_MASTER_FORCE;
+		else
+			cfg = MASTER_SLAVE_CFG_SLAVE_FORCE;
+	}
+
+	val = phy_read(phydev, MII_BCM54XX_LRELDSE);
+	if (val < 0)
+		return val;
+
+	if (val & LDSE_MASTER)
+		state = MASTER_SLAVE_STATE_MASTER;
+	else
+		state = MASTER_SLAVE_STATE_SLAVE;
+
+	phydev->master_slave_get = cfg;
+	phydev->master_slave_state = state;
+
+	return 0;
+}
+
+/* Read LDS Link Partner Ability in BroadR-Reach mode */
+static int lre_read_lpa(struct phy_device *phydev)
+{
+	int i, lrelpa;
+
+	if (phydev->autoneg != AUTONEG_ENABLE) {
+		if (!phydev->autoneg_complete) {
+			/* aneg not yet done, reset all relevant bits */
+			for (i = 0; i < ARRAY_SIZE(lds_br_bits); i++)
+				linkmode_clear_bit(lds_br_bits[i],
+						   phydev->lp_advertising);
+
+			return 0;
+		}
+
+		/* Long-Distance Signaling Link Partner Ability */
+		lrelpa = phy_read(phydev, MII_BCM54XX_LRELPA);
+		if (lrelpa < 0)
+			return lrelpa;
+
+		linkmode_mod_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
+				 phydev->lp_advertising,
+				 lrelpa & LRELPA_PAUSE_ASYM);
+		linkmode_mod_bit(ETHTOOL_LINK_MODE_Pause_BIT,
+				 phydev->lp_advertising,
+				 lrelpa & LRELPA_PAUSE);
+		linkmode_mod_bit(ETHTOOL_LINK_MODE_100baseT1_Full_BIT,
+				 phydev->lp_advertising,
+				 lrelpa & LRELPA_100_1PAIR);
+		linkmode_mod_bit(ETHTOOL_LINK_MODE_10baseT1BRR_Full_BIT,
+				 phydev->lp_advertising,
+				 lrelpa & LRELPA_10_1PAIR);
+	} else {
+		linkmode_zero(phydev->lp_advertising);
+	}
+
+	return 0;
+}
+
+static int lre_read_status_fixed(struct phy_device *phydev)
+{
+	int lrecr = phy_read(phydev, MII_BCM54XX_LRECR);
+
+	if (lrecr < 0)
+		return lrecr;
+
+	phydev->duplex = DUPLEX_FULL;
+
+	if (lrecr & LRECR_SPEED100)
+		phydev->speed = SPEED_100;
+	else
+		phydev->speed = SPEED_10;
+
+	return 0;
+}
+
+/**
+ * lre_update_link - update link status in @phydev
+ * @phydev: target phy_device struct
+ * Return:  0 on success, < 0 on error
+ *
+ * Description: Update the value in phydev->link to reflect the
+ *   current link value.  In order to do this, we need to read
+ *   the status register twice, keeping the second value.
+ *   This is a genphy_update_link modified to work on LRE registers
+ *   of BroadR-Reach PHY
+ */
+static int lre_update_link(struct phy_device *phydev)
+{
+	int status = 0, lrecr;
+
+	lrecr = phy_read(phydev, MII_BCM54XX_LRECR);
+	if (lrecr < 0)
+		return lrecr;
+
+	/* Autoneg is being started, therefore disregard BMSR value and
+	 * report link as down.
+	 */
+	if (lrecr & BMCR_ANRESTART)
+		goto done;
+
+	/* The link state is latched low so that momentary link
+	 * drops can be detected. Do not double-read the status
+	 * in polling mode to detect such short link drops except
+	 * the link was already down.
+	 */
+	if (!phy_polling_mode(phydev) || !phydev->link) {
+		status = phy_read(phydev, MII_BCM54XX_LRESR);
+		if (status < 0)
+			return status;
+		else if (status & LRESR_LSTATUS)
+			goto done;
+	}
+
+	/* Read link and autonegotiation status */
+	status = phy_read(phydev, MII_BCM54XX_LRESR);
+	if (status < 0)
+		return status;
+done:
+	phydev->link = status & LRESR_LSTATUS ? 1 : 0;
+	phydev->autoneg_complete = status & LRESR_LDSCOMPLETE ? 1 : 0;
+
+	/* Consider the case that autoneg was started and "aneg complete"
+	 * bit has been reset, but "link up" bit not yet.
+	 */
+	if (phydev->autoneg == AUTONEG_ENABLE && !phydev->autoneg_complete)
+		phydev->link = 0;
+
+	return 0;
+}
+
+/* Get the status in BroadRReach mode just like genphy_read_status does
+*   in normal mode
+*/
+static int bcm54811_lre_read_status(struct phy_device *phydev)
+{
+	int err, old_link = phydev->link;
+
+	/* Update the link, but return if there was an error */
+	err = lre_update_link(phydev);
+	if (err)
+		return err;
+
+	/* why bother the PHY if nothing can have changed */
+	if (phydev->autoneg ==
+		AUTONEG_ENABLE && old_link && phydev->link)
+		return 0;
+
+	phydev->speed = SPEED_UNKNOWN;
+	phydev->duplex = DUPLEX_UNKNOWN;
+	phydev->pause = 0;
+	phydev->asym_pause = 0;
+
+	err = lre_read_master_slave(phydev);
+	if (err < 0)
+		return err;
+
+	/* Read LDS Link Partner Ability */
+	err = lre_read_lpa(phydev);
+	if (err < 0)
+		return err;
+
+	if (phydev->autoneg == AUTONEG_ENABLE && phydev->autoneg_complete)
+		phy_resolve_aneg_linkmode(phydev);
+	else if (phydev->autoneg == AUTONEG_DISABLE)
+		err = lre_read_status_fixed(phydev);
+
+	return err;
+}
+
+static int bcm54811_read_status(struct phy_device *phydev)
+{
+	struct bcm54xx_phy_priv *priv = phydev->priv;
+
+	if (priv->brr_mode)
+		return  bcm54811_lre_read_status(phydev);
+
+	return genphy_read_status(phydev);
+}
+
 static struct phy_driver broadcom_drivers[] = {
 {
 	.phy_id		= PHY_ID_BCM5411,
@@ -1211,10 +1556,12 @@ static struct phy_driver broadcom_drivers[] = {
 	.get_strings	= bcm_phy_get_strings,
 	.get_stats	= bcm54xx_get_stats,
 	.probe		= bcm54xx_phy_probe,
-	.config_init    = bcm54811_config_init,
-	.config_aneg    = bcm5481_config_aneg,
+	.config_init    = bcm54xx_config_init,
+	.config_aneg    = bcm54811_config_aneg,
 	.config_intr    = bcm_phy_config_intr,
 	.handle_interrupt = bcm_phy_handle_interrupt,
+	.read_status	= bcm54811_read_status,
+	.get_features	= bcm5481x_read_abilities,
 	.suspend	= bcm54xx_suspend,
 	.resume		= bcm54xx_resume,
 	.link_change_notify	= bcm54xx_link_change_notify,
-- 
cgit v1.2.3-58-ga151


From f8321fa75102246d7415a6af441872f6637c93ab Mon Sep 17 00:00:00 2001
From: Breno Leitao <leitao@debian.org>
Date: Fri, 12 Jul 2024 04:53:25 -0700
Subject: virtio_net: Fix napi_skb_cache_put warning

After the commit bdacf3e34945 ("net: Use nested-BH locking for
napi_alloc_cache.") was merged, the following warning began to appear:

	 WARNING: CPU: 5 PID: 1 at net/core/skbuff.c:1451 napi_skb_cache_put+0x82/0x4b0

	  __warn+0x12f/0x340
	  napi_skb_cache_put+0x82/0x4b0
	  napi_skb_cache_put+0x82/0x4b0
	  report_bug+0x165/0x370
	  handle_bug+0x3d/0x80
	  exc_invalid_op+0x1a/0x50
	  asm_exc_invalid_op+0x1a/0x20
	  __free_old_xmit+0x1c8/0x510
	  napi_skb_cache_put+0x82/0x4b0
	  __free_old_xmit+0x1c8/0x510
	  __free_old_xmit+0x1c8/0x510
	  __pfx___free_old_xmit+0x10/0x10

The issue arises because virtio is assuming it's running in NAPI context
even when it's not, such as in the netpoll case.

To resolve this, modify virtnet_poll_tx() to only set NAPI when budget
is available. Same for virtnet_poll_cleantx(), which always assumed that
it was in a NAPI context.

Fixes: df133f3f9625 ("virtio_net: bulk free tx skbs")
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Heng Qi <hengqi@linux.alibaba.com>
Link: https://patch.msgid.link/20240712115325.54175-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/virtio_net.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index b6323346208d..af474cc191d0 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -2750,7 +2750,7 @@ static int virtnet_receive(struct receive_queue *rq, int budget,
 	return packets;
 }
 
-static void virtnet_poll_cleantx(struct receive_queue *rq)
+static void virtnet_poll_cleantx(struct receive_queue *rq, int budget)
 {
 	struct virtnet_info *vi = rq->vq->vdev->priv;
 	unsigned int index = vq2rxq(rq->vq);
@@ -2768,7 +2768,7 @@ static void virtnet_poll_cleantx(struct receive_queue *rq)
 
 		do {
 			virtqueue_disable_cb(sq->vq);
-			free_old_xmit(sq, txq, true);
+			free_old_xmit(sq, txq, !!budget);
 		} while (unlikely(!virtqueue_enable_cb_delayed(sq->vq)));
 
 		if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS) {
@@ -2813,7 +2813,7 @@ static int virtnet_poll(struct napi_struct *napi, int budget)
 	unsigned int xdp_xmit = 0;
 	bool napi_complete;
 
-	virtnet_poll_cleantx(rq);
+	virtnet_poll_cleantx(rq, budget);
 
 	received = virtnet_receive(rq, budget, &xdp_xmit);
 	rq->packets_in_napi += received;
@@ -2935,7 +2935,7 @@ static int virtnet_poll_tx(struct napi_struct *napi, int budget)
 	txq = netdev_get_tx_queue(vi->dev, index);
 	__netif_tx_lock(txq, raw_smp_processor_id());
 	virtqueue_disable_cb(sq->vq);
-	free_old_xmit(sq, txq, true);
+	free_old_xmit(sq, txq, !!budget);
 
 	if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS) {
 		if (netif_tx_queue_stopped(txq)) {
-- 
cgit v1.2.3-58-ga151


From 59a931c5b732ca5fc2ca727f5a72aeabaafa85ec Mon Sep 17 00:00:00 2001
From: Taehee Yoo <ap420073@gmail.com>
Date: Fri, 12 Jul 2024 09:51:16 +0000
Subject: xdp: fix invalid wait context of page_pool_destroy()

If the driver uses a page pool, it creates a page pool with
page_pool_create().
The reference count of page pool is 1 as default.
A page pool will be destroyed only when a reference count reaches 0.
page_pool_destroy() is used to destroy page pool, it decreases a
reference count.
When a page pool is destroyed, ->disconnect() is called, which is
mem_allocator_disconnect().
This function internally acquires mutex_lock().

If the driver uses XDP, it registers a memory model with
xdp_rxq_info_reg_mem_model().
The xdp_rxq_info_reg_mem_model() internally increases a page pool
reference count if a memory model is a page pool.
Now the reference count is 2.

To destroy a page pool, the driver should call both page_pool_destroy()
and xdp_unreg_mem_model().
The xdp_unreg_mem_model() internally calls page_pool_destroy().
Only page_pool_destroy() decreases a reference count.

If a driver calls page_pool_destroy() then xdp_unreg_mem_model(), we
will face an invalid wait context warning.
Because xdp_unreg_mem_model() calls page_pool_destroy() with
rcu_read_lock().
The page_pool_destroy() internally acquires mutex_lock().

Splat looks like:
=============================
[ BUG: Invalid wait context ]
6.10.0-rc6+ #4 Tainted: G W
-----------------------------
ethtool/1806 is trying to lock:
ffffffff90387b90 (mem_id_lock){+.+.}-{4:4}, at: mem_allocator_disconnect+0x73/0x150
other info that might help us debug this:
context-{5:5}
3 locks held by ethtool/1806:
stack backtrace:
CPU: 0 PID: 1806 Comm: ethtool Tainted: G W 6.10.0-rc6+ #4 f916f41f172891c800f2fed
Hardware name: ASUS System Product Name/PRIME Z690-P D4, BIOS 0603 11/01/2021
Call Trace:
<TASK>
dump_stack_lvl+0x7e/0xc0
__lock_acquire+0x1681/0x4de0
? _printk+0x64/0xe0
? __pfx_mark_lock.part.0+0x10/0x10
? __pfx___lock_acquire+0x10/0x10
lock_acquire+0x1b3/0x580
? mem_allocator_disconnect+0x73/0x150
? __wake_up_klogd.part.0+0x16/0xc0
? __pfx_lock_acquire+0x10/0x10
? dump_stack_lvl+0x91/0xc0
__mutex_lock+0x15c/0x1690
? mem_allocator_disconnect+0x73/0x150
? __pfx_prb_read_valid+0x10/0x10
? mem_allocator_disconnect+0x73/0x150
? __pfx_llist_add_batch+0x10/0x10
? console_unlock+0x193/0x1b0
? lockdep_hardirqs_on+0xbe/0x140
? __pfx___mutex_lock+0x10/0x10
? tick_nohz_tick_stopped+0x16/0x90
? __irq_work_queue_local+0x1e5/0x330
? irq_work_queue+0x39/0x50
? __wake_up_klogd.part.0+0x79/0xc0
? mem_allocator_disconnect+0x73/0x150
mem_allocator_disconnect+0x73/0x150
? __pfx_mem_allocator_disconnect+0x10/0x10
? mark_held_locks+0xa5/0xf0
? rcu_is_watching+0x11/0xb0
page_pool_release+0x36e/0x6d0
page_pool_destroy+0xd7/0x440
xdp_unreg_mem_model+0x1a7/0x2a0
? __pfx_xdp_unreg_mem_model+0x10/0x10
? kfree+0x125/0x370
? bnxt_free_ring.isra.0+0x2eb/0x500
? bnxt_free_mem+0x5ac/0x2500
xdp_rxq_info_unreg+0x4a/0xd0
bnxt_free_mem+0x1356/0x2500
bnxt_close_nic+0xf0/0x3b0
? __pfx_bnxt_close_nic+0x10/0x10
? ethnl_parse_bit+0x2c6/0x6d0
? __pfx___nla_validate_parse+0x10/0x10
? __pfx_ethnl_parse_bit+0x10/0x10
bnxt_set_features+0x2a8/0x3e0
__netdev_update_features+0x4dc/0x1370
? ethnl_parse_bitset+0x4ff/0x750
? __pfx_ethnl_parse_bitset+0x10/0x10
? __pfx___netdev_update_features+0x10/0x10
? mark_held_locks+0xa5/0xf0
? _raw_spin_unlock_irqrestore+0x42/0x70
? __pm_runtime_resume+0x7d/0x110
ethnl_set_features+0x32d/0xa20

To fix this problem, it uses rhashtable_lookup_fast() instead of
rhashtable_lookup() with rcu_read_lock().
Using xa without rcu_read_lock() here is safe.
xa is freed by __xdp_mem_allocator_rcu_free() and this is called by
call_rcu() of mem_xa_remove().
The mem_xa_remove() is called by page_pool_destroy() if a reference
count reaches 0.
The xa is already protected by the reference count mechanism well in the
control plane.
So removing rcu_read_lock() for page_pool_destroy() is safe.

Fixes: c3f812cea0d7 ("page_pool: do not release pool until inflight == 0.")
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Link: https://patch.msgid.link/20240712095116.3801586-1-ap420073@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/xdp.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/net/core/xdp.c b/net/core/xdp.c
index 022c12059cf2..bcc5551c6424 100644
--- a/net/core/xdp.c
+++ b/net/core/xdp.c
@@ -127,10 +127,8 @@ void xdp_unreg_mem_model(struct xdp_mem_info *mem)
 		return;
 
 	if (type == MEM_TYPE_PAGE_POOL) {
-		rcu_read_lock();
-		xa = rhashtable_lookup(mem_id_ht, &id, mem_id_rht_params);
+		xa = rhashtable_lookup_fast(mem_id_ht, &id, mem_id_rht_params);
 		page_pool_destroy(xa->page_pool);
-		rcu_read_unlock();
 	}
 }
 EXPORT_SYMBOL_GPL(xdp_unreg_mem_model);
-- 
cgit v1.2.3-58-ga151


From 0a1868b93fad5938dbcca77286b25bf211c49f7a Mon Sep 17 00:00:00 2001
From: Elliot Ayrey <elliot.ayrey@alliedtelesis.co.nz>
Date: Fri, 12 Jul 2024 13:31:33 +1200
Subject: net: bridge: mst: Check vlan state for egress decision

If a port is blocking in the common instance but forwarding in an MST
instance, traffic egressing the bridge will be dropped because the
state of the common instance is overriding that of the MST instance.

Fix this by skipping the port state check in MST mode to allow
checking the vlan state via br_allowed_egress(). This is similar to
what happens in br_handle_frame_finish() when checking ingress
traffic, which was introduced in the change below.

Fixes: ec7328b59176 ("net: bridge: mst: Multiple Spanning Tree (MST) mode")
Signed-off-by: Elliot Ayrey <elliot.ayrey@alliedtelesis.co.nz>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_forward.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index d97064d460dc..e19b583ff2c6 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -25,8 +25,8 @@ static inline int should_deliver(const struct net_bridge_port *p,
 
 	vg = nbp_vlan_group_rcu(p);
 	return ((p->flags & BR_HAIRPIN_MODE) || skb->dev != p->dev) &&
-		p->state == BR_STATE_FORWARDING && br_allowed_egress(vg, skb) &&
-		nbp_switchdev_allowed_egress(p, skb) &&
+		(br_mst_is_enabled(p->br) || p->state == BR_STATE_FORWARDING) &&
+		br_allowed_egress(vg, skb) && nbp_switchdev_allowed_egress(p, skb) &&
 		!br_skb_isolated(p, skb);
 }
 
-- 
cgit v1.2.3-58-ga151


From a8ea8d531d1edf7b29e559713fe1aaab3530d7d9 Mon Sep 17 00:00:00 2001
From: MD Danish Anwar <danishanwar@ti.com>
Date: Fri, 12 Jul 2024 17:36:36 +0530
Subject: net: ti: icssg-prueth: Split out common object into module

icssg_prueth.c and icssg_prueth_sr1.c drivers use multiple common .c
files. These common objects are getting added to multiple modules. As a
result when both drivers are enabled in .config, below warning is seen.

drivers/net/ethernet/ti/Makefile: icssg/icssg_common.o is added to multiple modules: icssg-prueth icssg-prueth-sr1
drivers/net/ethernet/ti/Makefile: icssg/icssg_classifier.o is added to multiple modules: icssg-prueth icssg-prueth-sr1
drivers/net/ethernet/ti/Makefile: icssg/icssg_config.o is added to multiple modules: icssg-prueth icssg-prueth-sr1
drivers/net/ethernet/ti/Makefile: icssg/icssg_mii_cfg.o is added to multiple modules: icssg-prueth icssg-prueth-sr1
drivers/net/ethernet/ti/Makefile: icssg/icssg_stats.o is added to multiple modules: icssg-prueth icssg-prueth-sr1
drivers/net/ethernet/ti/Makefile: icssg/icssg_ethtool.o is added to multiple modules: icssg-prueth icssg-prueth-sr1

Fix this by building a new module (icssg.o) for all the common objects.
Both the driver can then depend on this common module.

Some APIs being exported have emac_ as the prefix which may result into
confusion with other existing APIs with emac_ prefix, to avoid
confusion, rename the APIs being exported with emac_ to icssg_ prefix.

This also fixes below error seen when both drivers are built.
ERROR: modpost: "icssg_queue_pop"
[drivers/net/ethernet/ti/icssg-prueth-sr1.ko] undefined!
ERROR: modpost: "icssg_queue_push"
[drivers/net/ethernet/ti/icssg-prueth-sr1.ko] undefined!

Reported-and-tested-by: Thorsten Leemhuis <linux@leemhuis.info>
Closes: https://lore.kernel.org/oe-kbuild-all/202405182038.ncf1mL7Z-lkp@intel.com/
Fixes: 487f7323f39a ("net: ti: icssg-prueth: Add helper functions to configure FDB")
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
Reviewed-by: Sai Krishna <saikrishnag@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ethernet/ti/Makefile                 | 32 +++++++--------
 drivers/net/ethernet/ti/icssg/icssg_classifier.c |  6 +++
 drivers/net/ethernet/ti/icssg/icssg_common.c     | 50 +++++++++++++++++++-----
 drivers/net/ethernet/ti/icssg/icssg_config.c     | 15 ++++++-
 drivers/net/ethernet/ti/icssg/icssg_ethtool.c    |  1 +
 drivers/net/ethernet/ti/icssg/icssg_mii_cfg.c    |  4 ++
 drivers/net/ethernet/ti/icssg/icssg_prueth.c     | 38 +++++++++---------
 drivers/net/ethernet/ti/icssg/icssg_prueth.h     | 22 +++++------
 drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c | 14 +++----
 drivers/net/ethernet/ti/icssg/icssg_queues.c     |  2 +
 drivers/net/ethernet/ti/icssg/icssg_stats.c      |  3 +-
 drivers/net/ethernet/ti/icssg/icssg_switchdev.c  |  4 +-
 12 files changed, 122 insertions(+), 69 deletions(-)

diff --git a/drivers/net/ethernet/ti/Makefile b/drivers/net/ethernet/ti/Makefile
index 59cd20a38267..cbcf44806924 100644
--- a/drivers/net/ethernet/ti/Makefile
+++ b/drivers/net/ethernet/ti/Makefile
@@ -31,22 +31,18 @@ ti-am65-cpsw-nuss-$(CONFIG_TI_AM65_CPSW_QOS) += am65-cpsw-qos.o
 ti-am65-cpsw-nuss-$(CONFIG_TI_K3_AM65_CPSW_SWITCHDEV) += am65-cpsw-switchdev.o
 obj-$(CONFIG_TI_K3_AM65_CPTS) += am65-cpts.o
 
-obj-$(CONFIG_TI_ICSSG_PRUETH) += icssg-prueth.o
-icssg-prueth-y := icssg/icssg_prueth.o \
-		  icssg/icssg_common.o \
-		  icssg/icssg_classifier.o \
-		  icssg/icssg_queues.o \
-		  icssg/icssg_config.o \
-		  icssg/icssg_mii_cfg.o \
-		  icssg/icssg_stats.o \
-		  icssg/icssg_ethtool.o \
-		  icssg/icssg_switchdev.o
-obj-$(CONFIG_TI_ICSSG_PRUETH_SR1) += icssg-prueth-sr1.o
-icssg-prueth-sr1-y := icssg/icssg_prueth_sr1.o \
-		      icssg/icssg_common.o \
-		      icssg/icssg_classifier.o \
-		      icssg/icssg_config.o \
-		      icssg/icssg_mii_cfg.o \
-		      icssg/icssg_stats.o \
-		      icssg/icssg_ethtool.o
+obj-$(CONFIG_TI_ICSSG_PRUETH) += icssg-prueth.o icssg.o
+icssg-prueth-y := icssg/icssg_prueth.o icssg/icssg_switchdev.o
+
+obj-$(CONFIG_TI_ICSSG_PRUETH_SR1) += icssg-prueth-sr1.o icssg.o
+icssg-prueth-sr1-y := icssg/icssg_prueth_sr1.o
+
+icssg-y := icssg/icssg_common.o \
+	   icssg/icssg_classifier.o \
+	   icssg/icssg_queues.o \
+	   icssg/icssg_config.o \
+	   icssg/icssg_mii_cfg.o \
+	   icssg/icssg_stats.o \
+	   icssg/icssg_ethtool.o
+
 obj-$(CONFIG_TI_ICSS_IEP) += icssg/icss_iep.o
diff --git a/drivers/net/ethernet/ti/icssg/icssg_classifier.c b/drivers/net/ethernet/ti/icssg/icssg_classifier.c
index f7d21da1a0fb..9ec504d976d6 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_classifier.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_classifier.c
@@ -297,6 +297,7 @@ void icssg_class_set_mac_addr(struct regmap *miig_rt, int slice, u8 *mac)
 		     mac[2] << 16 | mac[3] << 24));
 	regmap_write(miig_rt, offs[slice].mac1, (u32)(mac[4] | mac[5] << 8));
 }
+EXPORT_SYMBOL_GPL(icssg_class_set_mac_addr);
 
 static void icssg_class_ft1_add_mcast(struct regmap *miig_rt, int slice,
 				      int slot, const u8 *addr, const u8 *mask)
@@ -360,6 +361,7 @@ void icssg_class_disable(struct regmap *miig_rt, int slice)
 	/* clear CFG2 */
 	regmap_write(miig_rt, offs[slice].rx_class_cfg2, 0);
 }
+EXPORT_SYMBOL_GPL(icssg_class_disable);
 
 void icssg_class_default(struct regmap *miig_rt, int slice, bool allmulti,
 			 bool is_sr1)
@@ -390,6 +392,7 @@ void icssg_class_default(struct regmap *miig_rt, int slice, bool allmulti,
 	/* clear CFG2 */
 	regmap_write(miig_rt, offs[slice].rx_class_cfg2, 0);
 }
+EXPORT_SYMBOL_GPL(icssg_class_default);
 
 void icssg_class_promiscuous_sr1(struct regmap *miig_rt, int slice)
 {
@@ -408,6 +411,7 @@ void icssg_class_promiscuous_sr1(struct regmap *miig_rt, int slice)
 		regmap_write(miig_rt, offset, data);
 	}
 }
+EXPORT_SYMBOL_GPL(icssg_class_promiscuous_sr1);
 
 void icssg_class_add_mcast_sr1(struct regmap *miig_rt, int slice,
 			       struct net_device *ndev)
@@ -449,6 +453,7 @@ void icssg_class_add_mcast_sr1(struct regmap *miig_rt, int slice,
 		slot++;
 	}
 }
+EXPORT_SYMBOL_GPL(icssg_class_add_mcast_sr1);
 
 /* required for SAV check */
 void icssg_ft1_set_mac_addr(struct regmap *miig_rt, int slice, u8 *mac_addr)
@@ -460,3 +465,4 @@ void icssg_ft1_set_mac_addr(struct regmap *miig_rt, int slice, u8 *mac_addr)
 	rx_class_ft1_set_da_mask(miig_rt, slice, 0, mask_addr);
 	rx_class_ft1_cfg_set_type(miig_rt, slice, 0, FT1_CFG_TYPE_EQ);
 }
+EXPORT_SYMBOL_GPL(icssg_ft1_set_mac_addr);
diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c
index 2f716c0d7060..b9d8a93d1680 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_common.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_common.c
@@ -51,6 +51,7 @@ void prueth_cleanup_rx_chns(struct prueth_emac *emac,
 	if (rx_chn->rx_chn)
 		k3_udma_glue_release_rx_chn(rx_chn->rx_chn);
 }
+EXPORT_SYMBOL_GPL(prueth_cleanup_rx_chns);
 
 void prueth_cleanup_tx_chns(struct prueth_emac *emac)
 {
@@ -71,6 +72,7 @@ void prueth_cleanup_tx_chns(struct prueth_emac *emac)
 		memset(tx_chn, 0, sizeof(*tx_chn));
 	}
 }
+EXPORT_SYMBOL_GPL(prueth_cleanup_tx_chns);
 
 void prueth_ndev_del_tx_napi(struct prueth_emac *emac, int num)
 {
@@ -84,6 +86,7 @@ void prueth_ndev_del_tx_napi(struct prueth_emac *emac, int num)
 		netif_napi_del(&tx_chn->napi_tx);
 	}
 }
+EXPORT_SYMBOL_GPL(prueth_ndev_del_tx_napi);
 
 void prueth_xmit_free(struct prueth_tx_chn *tx_chn,
 		      struct cppi5_host_desc_t *desc)
@@ -120,6 +123,7 @@ void prueth_xmit_free(struct prueth_tx_chn *tx_chn,
 
 	k3_cppi_desc_pool_free(tx_chn->desc_pool, first_desc);
 }
+EXPORT_SYMBOL_GPL(prueth_xmit_free);
 
 int emac_tx_complete_packets(struct prueth_emac *emac, int chn,
 			     int budget, bool *tdown)
@@ -264,6 +268,7 @@ fail:
 	prueth_ndev_del_tx_napi(emac, i);
 	return ret;
 }
+EXPORT_SYMBOL_GPL(prueth_ndev_add_tx_napi);
 
 int prueth_init_tx_chns(struct prueth_emac *emac)
 {
@@ -344,6 +349,7 @@ fail:
 	prueth_cleanup_tx_chns(emac);
 	return ret;
 }
+EXPORT_SYMBOL_GPL(prueth_init_tx_chns);
 
 int prueth_init_rx_chns(struct prueth_emac *emac,
 			struct prueth_rx_chn *rx_chn,
@@ -453,6 +459,7 @@ fail:
 	prueth_cleanup_rx_chns(emac, rx_chn, max_rflows);
 	return ret;
 }
+EXPORT_SYMBOL_GPL(prueth_init_rx_chns);
 
 int prueth_dma_rx_push(struct prueth_emac *emac,
 		       struct sk_buff *skb,
@@ -490,6 +497,7 @@ int prueth_dma_rx_push(struct prueth_emac *emac,
 	return k3_udma_glue_push_rx_chn(rx_chn->rx_chn, 0,
 					desc_rx, desc_dma);
 }
+EXPORT_SYMBOL_GPL(prueth_dma_rx_push);
 
 u64 icssg_ts_to_ns(u32 hi_sw, u32 hi, u32 lo, u32 cycle_time_ns)
 {
@@ -505,6 +513,7 @@ u64 icssg_ts_to_ns(u32 hi_sw, u32 hi, u32 lo, u32 cycle_time_ns)
 
 	return ns;
 }
+EXPORT_SYMBOL_GPL(icssg_ts_to_ns);
 
 void emac_rx_timestamp(struct prueth_emac *emac,
 		       struct sk_buff *skb, u32 *psdata)
@@ -636,7 +645,7 @@ static int prueth_tx_ts_cookie_get(struct prueth_emac *emac)
 }
 
 /**
- * emac_ndo_start_xmit - EMAC Transmit function
+ * icssg_ndo_start_xmit - EMAC Transmit function
  * @skb: SKB pointer
  * @ndev: EMAC network adapter
  *
@@ -647,7 +656,7 @@ static int prueth_tx_ts_cookie_get(struct prueth_emac *emac)
  *
  * Return: enum netdev_tx
  */
-enum netdev_tx emac_ndo_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+enum netdev_tx icssg_ndo_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 {
 	struct cppi5_host_desc_t *first_desc, *next_desc, *cur_desc;
 	struct prueth_emac *emac = netdev_priv(ndev);
@@ -806,6 +815,7 @@ drop_stop_q_busy:
 	netif_tx_stop_queue(netif_txq);
 	return NETDEV_TX_BUSY;
 }
+EXPORT_SYMBOL_GPL(icssg_ndo_start_xmit);
 
 static void prueth_tx_cleanup(void *data, dma_addr_t desc_dma)
 {
@@ -831,6 +841,7 @@ irqreturn_t prueth_rx_irq(int irq, void *dev_id)
 
 	return IRQ_HANDLED;
 }
+EXPORT_SYMBOL_GPL(prueth_rx_irq);
 
 void prueth_emac_stop(struct prueth_emac *emac)
 {
@@ -855,6 +866,7 @@ void prueth_emac_stop(struct prueth_emac *emac)
 	rproc_shutdown(prueth->rtu[slice]);
 	rproc_shutdown(prueth->pru[slice]);
 }
+EXPORT_SYMBOL_GPL(prueth_emac_stop);
 
 void prueth_cleanup_tx_ts(struct prueth_emac *emac)
 {
@@ -867,8 +879,9 @@ void prueth_cleanup_tx_ts(struct prueth_emac *emac)
 		}
 	}
 }
+EXPORT_SYMBOL_GPL(prueth_cleanup_tx_ts);
 
-int emac_napi_rx_poll(struct napi_struct *napi_rx, int budget)
+int icssg_napi_rx_poll(struct napi_struct *napi_rx, int budget)
 {
 	struct prueth_emac *emac = prueth_napi_to_emac(napi_rx);
 	int rx_flow = emac->is_sr1 ?
@@ -905,6 +918,7 @@ int emac_napi_rx_poll(struct napi_struct *napi_rx, int budget)
 
 	return num_rx;
 }
+EXPORT_SYMBOL_GPL(icssg_napi_rx_poll);
 
 int prueth_prepare_rx_chan(struct prueth_emac *emac,
 			   struct prueth_rx_chn *chn,
@@ -930,6 +944,7 @@ int prueth_prepare_rx_chan(struct prueth_emac *emac,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(prueth_prepare_rx_chan);
 
 void prueth_reset_tx_chan(struct prueth_emac *emac, int ch_num,
 			  bool free_skb)
@@ -944,6 +959,7 @@ void prueth_reset_tx_chan(struct prueth_emac *emac, int ch_num,
 		k3_udma_glue_disable_tx_chn(emac->tx_chns[i].tx_chn);
 	}
 }
+EXPORT_SYMBOL_GPL(prueth_reset_tx_chan);
 
 void prueth_reset_rx_chan(struct prueth_rx_chn *chn,
 			  int num_flows, bool disable)
@@ -956,11 +972,13 @@ void prueth_reset_rx_chan(struct prueth_rx_chn *chn,
 	if (disable)
 		k3_udma_glue_disable_rx_chn(chn->rx_chn);
 }
+EXPORT_SYMBOL_GPL(prueth_reset_rx_chan);
 
-void emac_ndo_tx_timeout(struct net_device *ndev, unsigned int txqueue)
+void icssg_ndo_tx_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	ndev->stats.tx_errors++;
 }
+EXPORT_SYMBOL_GPL(icssg_ndo_tx_timeout);
 
 static int emac_set_ts_config(struct net_device *ndev, struct ifreq *ifr)
 {
@@ -1024,7 +1042,7 @@ static int emac_get_ts_config(struct net_device *ndev, struct ifreq *ifr)
 			    -EFAULT : 0;
 }
 
-int emac_ndo_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd)
+int icssg_ndo_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd)
 {
 	switch (cmd) {
 	case SIOCGHWTSTAMP:
@@ -1037,9 +1055,10 @@ int emac_ndo_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd)
 
 	return phy_do_ioctl(ndev, ifr, cmd);
 }
+EXPORT_SYMBOL_GPL(icssg_ndo_ioctl);
 
-void emac_ndo_get_stats64(struct net_device *ndev,
-			  struct rtnl_link_stats64 *stats)
+void icssg_ndo_get_stats64(struct net_device *ndev,
+			   struct rtnl_link_stats64 *stats)
 {
 	struct prueth_emac *emac = netdev_priv(ndev);
 
@@ -1058,9 +1077,10 @@ void emac_ndo_get_stats64(struct net_device *ndev,
 	stats->tx_errors  = ndev->stats.tx_errors;
 	stats->tx_dropped = ndev->stats.tx_dropped;
 }
+EXPORT_SYMBOL_GPL(icssg_ndo_get_stats64);
 
-int emac_ndo_get_phys_port_name(struct net_device *ndev, char *name,
-				size_t len)
+int icssg_ndo_get_phys_port_name(struct net_device *ndev, char *name,
+				 size_t len)
 {
 	struct prueth_emac *emac = netdev_priv(ndev);
 	int ret;
@@ -1071,6 +1091,7 @@ int emac_ndo_get_phys_port_name(struct net_device *ndev, char *name,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(icssg_ndo_get_phys_port_name);
 
 /* get emac_port corresponding to eth_node name */
 int prueth_node_port(struct device_node *eth_node)
@@ -1089,6 +1110,7 @@ int prueth_node_port(struct device_node *eth_node)
 	else
 		return PRUETH_PORT_INVALID;
 }
+EXPORT_SYMBOL_GPL(prueth_node_port);
 
 /* get MAC instance corresponding to eth_node name */
 int prueth_node_mac(struct device_node *eth_node)
@@ -1107,6 +1129,7 @@ int prueth_node_mac(struct device_node *eth_node)
 	else
 		return PRUETH_MAC_INVALID;
 }
+EXPORT_SYMBOL_GPL(prueth_node_mac);
 
 void prueth_netdev_exit(struct prueth *prueth,
 			struct device_node *eth_node)
@@ -1132,6 +1155,7 @@ void prueth_netdev_exit(struct prueth *prueth,
 	free_netdev(emac->ndev);
 	prueth->emac[mac] = NULL;
 }
+EXPORT_SYMBOL_GPL(prueth_netdev_exit);
 
 int prueth_get_cores(struct prueth *prueth, int slice, bool is_sr1)
 {
@@ -1182,6 +1206,7 @@ int prueth_get_cores(struct prueth *prueth, int slice, bool is_sr1)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(prueth_get_cores);
 
 void prueth_put_cores(struct prueth *prueth, int slice)
 {
@@ -1194,6 +1219,7 @@ void prueth_put_cores(struct prueth *prueth, int slice)
 	if (prueth->pru[slice])
 		pru_rproc_put(prueth->pru[slice]);
 }
+EXPORT_SYMBOL_GPL(prueth_put_cores);
 
 #ifdef CONFIG_PM_SLEEP
 static int prueth_suspend(struct device *dev)
@@ -1250,3 +1276,9 @@ static int prueth_resume(struct device *dev)
 const struct dev_pm_ops prueth_dev_pm_ops = {
 	SET_SYSTEM_SLEEP_PM_OPS(prueth_suspend, prueth_resume)
 };
+EXPORT_SYMBOL_GPL(prueth_dev_pm_ops);
+
+MODULE_AUTHOR("Roger Quadros <rogerq@ti.com>");
+MODULE_AUTHOR("Md Danish Anwar <danishanwar@ti.com>");
+MODULE_DESCRIPTION("PRUSS ICSSG Ethernet Driver Common Module");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ethernet/ti/icssg/icssg_config.c b/drivers/net/ethernet/ti/icssg/icssg_config.c
index 9444e56b7672..dae52a83a378 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_config.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_config.c
@@ -248,6 +248,7 @@ void icssg_config_ipg(struct prueth_emac *emac)
 
 	icssg_mii_update_ipg(prueth->mii_rt, slice, ipg);
 }
+EXPORT_SYMBOL_GPL(icssg_config_ipg);
 
 static void emac_r30_cmd_init(struct prueth_emac *emac)
 {
@@ -508,6 +509,7 @@ int icssg_config(struct prueth *prueth, struct prueth_emac *emac, int slice)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(icssg_config);
 
 /* Bitmask for ICSSG r30 commands */
 static const struct icssg_r30_cmd emac_r32_bitmask[] = {
@@ -532,8 +534,8 @@ static const struct icssg_r30_cmd emac_r32_bitmask[] = {
 	{{0xffef0000,  EMAC_NONE, 0xffef0000, EMAC_NONE}}	/* VLAN UNWARE*/
 };
 
-int emac_set_port_state(struct prueth_emac *emac,
-			enum icssg_port_state_cmd cmd)
+int icssg_set_port_state(struct prueth_emac *emac,
+			 enum icssg_port_state_cmd cmd)
 {
 	struct icssg_r30_cmd __iomem *p;
 	int ret = -ETIMEDOUT;
@@ -564,6 +566,7 @@ int emac_set_port_state(struct prueth_emac *emac,
 
 	return ret;
 }
+EXPORT_SYMBOL_GPL(icssg_set_port_state);
 
 void icssg_config_half_duplex(struct prueth_emac *emac)
 {
@@ -575,6 +578,7 @@ void icssg_config_half_duplex(struct prueth_emac *emac)
 	val = get_random_u32();
 	writel(val, emac->dram.va + HD_RAND_SEED_OFFSET);
 }
+EXPORT_SYMBOL_GPL(icssg_config_half_duplex);
 
 void icssg_config_set_speed(struct prueth_emac *emac)
 {
@@ -601,6 +605,7 @@ void icssg_config_set_speed(struct prueth_emac *emac)
 
 	writeb(fw_speed, emac->dram.va + PORT_LINK_SPEED_OFFSET);
 }
+EXPORT_SYMBOL_GPL(icssg_config_set_speed);
 
 int icssg_send_fdb_msg(struct prueth_emac *emac, struct mgmt_cmd *cmd,
 		       struct mgmt_cmd_rsp *rsp)
@@ -635,6 +640,7 @@ int icssg_send_fdb_msg(struct prueth_emac *emac, struct mgmt_cmd *cmd,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(icssg_send_fdb_msg);
 
 static void icssg_fdb_setup(struct prueth_emac *emac, struct mgmt_cmd *fdb_cmd,
 			    const unsigned char *addr, u8 fid, int cmd)
@@ -687,6 +693,7 @@ int icssg_fdb_add_del(struct prueth_emac *emac, const unsigned char *addr,
 
 	return -EINVAL;
 }
+EXPORT_SYMBOL_GPL(icssg_fdb_add_del);
 
 int icssg_fdb_lookup(struct prueth_emac *emac, const unsigned char *addr,
 		     u8 vid)
@@ -716,6 +723,7 @@ int icssg_fdb_lookup(struct prueth_emac *emac, const unsigned char *addr,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(icssg_fdb_lookup);
 
 void icssg_vtbl_modify(struct prueth_emac *emac, u8 vid, u8 port_mask,
 		       u8 untag_mask, bool add)
@@ -741,6 +749,7 @@ void icssg_vtbl_modify(struct prueth_emac *emac, u8 vid, u8 port_mask,
 
 	tbl[vid].fid_c1 = fid_c1;
 }
+EXPORT_SYMBOL_GPL(icssg_vtbl_modify);
 
 u16 icssg_get_pvid(struct prueth_emac *emac)
 {
@@ -756,6 +765,7 @@ u16 icssg_get_pvid(struct prueth_emac *emac)
 
 	return pvid;
 }
+EXPORT_SYMBOL_GPL(icssg_get_pvid);
 
 void icssg_set_pvid(struct prueth *prueth, u8 vid, u8 port)
 {
@@ -771,3 +781,4 @@ void icssg_set_pvid(struct prueth *prueth, u8 vid, u8 port)
 	else
 		writel(pvid, prueth->shram.va + EMAC_ICSSG_SWITCH_PORT0_DEFAULT_VLAN_OFFSET);
 }
+EXPORT_SYMBOL_GPL(icssg_set_pvid);
diff --git a/drivers/net/ethernet/ti/icssg/icssg_ethtool.c b/drivers/net/ethernet/ti/icssg/icssg_ethtool.c
index c8d0f45cc5b1..131eb4cae1c3 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_ethtool.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_ethtool.c
@@ -312,3 +312,4 @@ const struct ethtool_ops icssg_ethtool_ops = {
 	.nway_reset = emac_nway_reset,
 	.get_rmon_stats = emac_get_rmon_stats,
 };
+EXPORT_SYMBOL_GPL(icssg_ethtool_ops);
diff --git a/drivers/net/ethernet/ti/icssg/icssg_mii_cfg.c b/drivers/net/ethernet/ti/icssg/icssg_mii_cfg.c
index 92718ae40d7e..b64955438bb2 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_mii_cfg.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_mii_cfg.c
@@ -40,6 +40,7 @@ void icssg_mii_update_mtu(struct regmap *mii_rt, int mii, int mtu)
 				   (mtu - 1) << PRUSS_MII_RT_RX_FRMS_MAX_FRM_SHIFT);
 	}
 }
+EXPORT_SYMBOL_GPL(icssg_mii_update_mtu);
 
 void icssg_update_rgmii_cfg(struct regmap *miig_rt, struct prueth_emac *emac)
 {
@@ -66,6 +67,7 @@ void icssg_update_rgmii_cfg(struct regmap *miig_rt, struct prueth_emac *emac)
 	regmap_update_bits(miig_rt, RGMII_CFG_OFFSET, full_duplex_mask,
 			   full_duplex_val);
 }
+EXPORT_SYMBOL_GPL(icssg_update_rgmii_cfg);
 
 void icssg_miig_set_interface_mode(struct regmap *miig_rt, int mii, phy_interface_t phy_if)
 {
@@ -105,6 +107,7 @@ u32 icssg_rgmii_get_speed(struct regmap *miig_rt, int mii)
 
 	return icssg_rgmii_cfg_get_bitfield(miig_rt, mask, shift);
 }
+EXPORT_SYMBOL_GPL(icssg_rgmii_get_speed);
 
 u32 icssg_rgmii_get_fullduplex(struct regmap *miig_rt, int mii)
 {
@@ -118,3 +121,4 @@ u32 icssg_rgmii_get_fullduplex(struct regmap *miig_rt, int mii)
 
 	return icssg_rgmii_cfg_get_bitfield(miig_rt, mask, shift);
 }
+EXPORT_SYMBOL_GPL(icssg_rgmii_get_fullduplex);
diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
index e13835100754..3e51b3a9b0a5 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
@@ -249,10 +249,10 @@ static void emac_adjust_link(struct net_device *ndev)
 			icssg_config_ipg(emac);
 			spin_unlock_irqrestore(&emac->lock, flags);
 			icssg_config_set_speed(emac);
-			emac_set_port_state(emac, ICSSG_EMAC_PORT_FORWARD);
+			icssg_set_port_state(emac, ICSSG_EMAC_PORT_FORWARD);
 
 		} else {
-			emac_set_port_state(emac, ICSSG_EMAC_PORT_DISABLE);
+			icssg_set_port_state(emac, ICSSG_EMAC_PORT_DISABLE);
 		}
 	}
 
@@ -694,17 +694,17 @@ static void emac_ndo_set_rx_mode_work(struct work_struct *work)
 
 	promisc = ndev->flags & IFF_PROMISC;
 	allmulti = ndev->flags & IFF_ALLMULTI;
-	emac_set_port_state(emac, ICSSG_EMAC_PORT_UC_FLOODING_DISABLE);
-	emac_set_port_state(emac, ICSSG_EMAC_PORT_MC_FLOODING_DISABLE);
+	icssg_set_port_state(emac, ICSSG_EMAC_PORT_UC_FLOODING_DISABLE);
+	icssg_set_port_state(emac, ICSSG_EMAC_PORT_MC_FLOODING_DISABLE);
 
 	if (promisc) {
-		emac_set_port_state(emac, ICSSG_EMAC_PORT_UC_FLOODING_ENABLE);
-		emac_set_port_state(emac, ICSSG_EMAC_PORT_MC_FLOODING_ENABLE);
+		icssg_set_port_state(emac, ICSSG_EMAC_PORT_UC_FLOODING_ENABLE);
+		icssg_set_port_state(emac, ICSSG_EMAC_PORT_MC_FLOODING_ENABLE);
 		return;
 	}
 
 	if (allmulti) {
-		emac_set_port_state(emac, ICSSG_EMAC_PORT_MC_FLOODING_ENABLE);
+		icssg_set_port_state(emac, ICSSG_EMAC_PORT_MC_FLOODING_ENABLE);
 		return;
 	}
 
@@ -728,14 +728,14 @@ static void emac_ndo_set_rx_mode(struct net_device *ndev)
 static const struct net_device_ops emac_netdev_ops = {
 	.ndo_open = emac_ndo_open,
 	.ndo_stop = emac_ndo_stop,
-	.ndo_start_xmit = emac_ndo_start_xmit,
+	.ndo_start_xmit = icssg_ndo_start_xmit,
 	.ndo_set_mac_address = eth_mac_addr,
 	.ndo_validate_addr = eth_validate_addr,
-	.ndo_tx_timeout = emac_ndo_tx_timeout,
+	.ndo_tx_timeout = icssg_ndo_tx_timeout,
 	.ndo_set_rx_mode = emac_ndo_set_rx_mode,
-	.ndo_eth_ioctl = emac_ndo_ioctl,
-	.ndo_get_stats64 = emac_ndo_get_stats64,
-	.ndo_get_phys_port_name = emac_ndo_get_phys_port_name,
+	.ndo_eth_ioctl = icssg_ndo_ioctl,
+	.ndo_get_stats64 = icssg_ndo_get_stats64,
+	.ndo_get_phys_port_name = icssg_ndo_get_phys_port_name,
 };
 
 static int prueth_netdev_init(struct prueth *prueth,
@@ -771,7 +771,7 @@ static int prueth_netdev_init(struct prueth *prueth,
 	}
 	INIT_WORK(&emac->rx_mode_work, emac_ndo_set_rx_mode_work);
 
-	INIT_DELAYED_WORK(&emac->stats_work, emac_stats_work_handler);
+	INIT_DELAYED_WORK(&emac->stats_work, icssg_stats_work_handler);
 
 	ret = pruss_request_mem_region(prueth->pruss,
 				       port == PRUETH_PORT_MII0 ?
@@ -864,7 +864,7 @@ static int prueth_netdev_init(struct prueth *prueth,
 	ndev->hw_features = NETIF_F_SG;
 	ndev->features = ndev->hw_features;
 
-	netif_napi_add(ndev, &emac->napi_rx, emac_napi_rx_poll);
+	netif_napi_add(ndev, &emac->napi_rx, icssg_napi_rx_poll);
 	hrtimer_init(&emac->rx_hrtimer, CLOCK_MONOTONIC,
 		     HRTIMER_MODE_REL_PINNED);
 	emac->rx_hrtimer.function = &emac_rx_timer_callback;
@@ -927,8 +927,8 @@ static void prueth_emac_restart(struct prueth *prueth)
 		netif_device_detach(emac1->ndev);
 
 	/* Disable both PRUeth ports */
-	emac_set_port_state(emac0, ICSSG_EMAC_PORT_DISABLE);
-	emac_set_port_state(emac1, ICSSG_EMAC_PORT_DISABLE);
+	icssg_set_port_state(emac0, ICSSG_EMAC_PORT_DISABLE);
+	icssg_set_port_state(emac1, ICSSG_EMAC_PORT_DISABLE);
 
 	/* Stop both pru cores for both PRUeth ports*/
 	prueth_emac_stop(emac0);
@@ -943,8 +943,8 @@ static void prueth_emac_restart(struct prueth *prueth)
 	prueth->emacs_initialized++;
 
 	/* Enable forwarding for both PRUeth ports */
-	emac_set_port_state(emac0, ICSSG_EMAC_PORT_FORWARD);
-	emac_set_port_state(emac1, ICSSG_EMAC_PORT_FORWARD);
+	icssg_set_port_state(emac0, ICSSG_EMAC_PORT_FORWARD);
+	icssg_set_port_state(emac1, ICSSG_EMAC_PORT_FORWARD);
 
 	/* Attache net_device for both PRUeth ports */
 	netif_device_attach(emac0->ndev);
@@ -972,7 +972,7 @@ static void icssg_enable_switch_mode(struct prueth *prueth)
 					  BIT(emac->port_id) | DEFAULT_UNTAG_MASK,
 					  true);
 			icssg_set_pvid(prueth, emac->port_vlan, emac->port_id);
-			emac_set_port_state(emac, ICSSG_EMAC_PORT_VLAN_AWARE_ENABLE);
+			icssg_set_port_state(emac, ICSSG_EMAC_PORT_VLAN_AWARE_ENABLE);
 		}
 	}
 }
diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.h b/drivers/net/ethernet/ti/icssg/icssg_prueth.h
index 5eeeccb73665..f678d656a3ed 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_prueth.h
+++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.h
@@ -329,8 +329,8 @@ void icssg_ft1_set_mac_addr(struct regmap *miig_rt, int slice, u8 *mac_addr);
 void icssg_config_ipg(struct prueth_emac *emac);
 int icssg_config(struct prueth *prueth, struct prueth_emac *emac,
 		 int slice);
-int emac_set_port_state(struct prueth_emac *emac,
-			enum icssg_port_state_cmd state);
+int icssg_set_port_state(struct prueth_emac *emac,
+			 enum icssg_port_state_cmd state);
 void icssg_config_set_speed(struct prueth_emac *emac);
 void icssg_config_half_duplex(struct prueth_emac *emac);
 
@@ -352,7 +352,7 @@ void icssg_set_pvid(struct prueth *prueth, u8 vid, u8 port);
 #define prueth_napi_to_tx_chn(pnapi) \
 	container_of(pnapi, struct prueth_tx_chn, napi_tx)
 
-void emac_stats_work_handler(struct work_struct *work);
+void icssg_stats_work_handler(struct work_struct *work);
 void emac_update_hardware_stats(struct prueth_emac *emac);
 int emac_get_stat_by_name(struct prueth_emac *emac, char *stat_name);
 
@@ -377,11 +377,11 @@ int prueth_dma_rx_push(struct prueth_emac *emac,
 		       struct prueth_rx_chn *rx_chn);
 void emac_rx_timestamp(struct prueth_emac *emac,
 		       struct sk_buff *skb, u32 *psdata);
-enum netdev_tx emac_ndo_start_xmit(struct sk_buff *skb, struct net_device *ndev);
+enum netdev_tx icssg_ndo_start_xmit(struct sk_buff *skb, struct net_device *ndev);
 irqreturn_t prueth_rx_irq(int irq, void *dev_id);
 void prueth_emac_stop(struct prueth_emac *emac);
 void prueth_cleanup_tx_ts(struct prueth_emac *emac);
-int emac_napi_rx_poll(struct napi_struct *napi_rx, int budget);
+int icssg_napi_rx_poll(struct napi_struct *napi_rx, int budget);
 int prueth_prepare_rx_chan(struct prueth_emac *emac,
 			   struct prueth_rx_chn *chn,
 			   int buf_size);
@@ -389,12 +389,12 @@ void prueth_reset_tx_chan(struct prueth_emac *emac, int ch_num,
 			  bool free_skb);
 void prueth_reset_rx_chan(struct prueth_rx_chn *chn,
 			  int num_flows, bool disable);
-void emac_ndo_tx_timeout(struct net_device *ndev, unsigned int txqueue);
-int emac_ndo_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd);
-void emac_ndo_get_stats64(struct net_device *ndev,
-			  struct rtnl_link_stats64 *stats);
-int emac_ndo_get_phys_port_name(struct net_device *ndev, char *name,
-				size_t len);
+void icssg_ndo_tx_timeout(struct net_device *ndev, unsigned int txqueue);
+int icssg_ndo_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd);
+void icssg_ndo_get_stats64(struct net_device *ndev,
+			   struct rtnl_link_stats64 *stats);
+int icssg_ndo_get_phys_port_name(struct net_device *ndev, char *name,
+				 size_t len);
 int prueth_node_port(struct device_node *eth_node);
 int prueth_node_mac(struct device_node *eth_node);
 void prueth_netdev_exit(struct prueth *prueth,
diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c b/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c
index fa98bdb11ece..e180c1166170 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_prueth_sr1.c
@@ -722,14 +722,14 @@ static void emac_ndo_set_rx_mode_sr1(struct net_device *ndev)
 static const struct net_device_ops emac_netdev_ops = {
 	.ndo_open = emac_ndo_open,
 	.ndo_stop = emac_ndo_stop,
-	.ndo_start_xmit = emac_ndo_start_xmit,
+	.ndo_start_xmit = icssg_ndo_start_xmit,
 	.ndo_set_mac_address = eth_mac_addr,
 	.ndo_validate_addr = eth_validate_addr,
-	.ndo_tx_timeout = emac_ndo_tx_timeout,
+	.ndo_tx_timeout = icssg_ndo_tx_timeout,
 	.ndo_set_rx_mode = emac_ndo_set_rx_mode_sr1,
-	.ndo_eth_ioctl = emac_ndo_ioctl,
-	.ndo_get_stats64 = emac_ndo_get_stats64,
-	.ndo_get_phys_port_name = emac_ndo_get_phys_port_name,
+	.ndo_eth_ioctl = icssg_ndo_ioctl,
+	.ndo_get_stats64 = icssg_ndo_get_stats64,
+	.ndo_get_phys_port_name = icssg_ndo_get_phys_port_name,
 };
 
 static int prueth_netdev_init(struct prueth *prueth,
@@ -767,7 +767,7 @@ static int prueth_netdev_init(struct prueth *prueth,
 		goto free_ndev;
 	}
 
-	INIT_DELAYED_WORK(&emac->stats_work, emac_stats_work_handler);
+	INIT_DELAYED_WORK(&emac->stats_work, icssg_stats_work_handler);
 
 	ret = pruss_request_mem_region(prueth->pruss,
 				       port == PRUETH_PORT_MII0 ?
@@ -854,7 +854,7 @@ static int prueth_netdev_init(struct prueth *prueth,
 	ndev->hw_features = NETIF_F_SG;
 	ndev->features = ndev->hw_features;
 
-	netif_napi_add(ndev, &emac->napi_rx, emac_napi_rx_poll);
+	netif_napi_add(ndev, &emac->napi_rx, icssg_napi_rx_poll);
 	prueth->emac[mac] = emac;
 
 	return 0;
diff --git a/drivers/net/ethernet/ti/icssg/icssg_queues.c b/drivers/net/ethernet/ti/icssg/icssg_queues.c
index 3c34f61ad40b..e5052d9e7807 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_queues.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_queues.c
@@ -28,6 +28,7 @@ int icssg_queue_pop(struct prueth *prueth, u8 queue)
 
 	return val;
 }
+EXPORT_SYMBOL_GPL(icssg_queue_pop);
 
 void icssg_queue_push(struct prueth *prueth, int queue, u16 addr)
 {
@@ -36,6 +37,7 @@ void icssg_queue_push(struct prueth *prueth, int queue, u16 addr)
 
 	regmap_write(prueth->miig_rt, ICSSG_QUEUE_OFFSET + 4 * queue, addr);
 }
+EXPORT_SYMBOL_GPL(icssg_queue_push);
 
 u32 icssg_queue_level(struct prueth *prueth, int queue)
 {
diff --git a/drivers/net/ethernet/ti/icssg/icssg_stats.c b/drivers/net/ethernet/ti/icssg/icssg_stats.c
index 3dbadddd7e35..2fb150c13078 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_stats.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_stats.c
@@ -42,7 +42,7 @@ void emac_update_hardware_stats(struct prueth_emac *emac)
 	}
 }
 
-void emac_stats_work_handler(struct work_struct *work)
+void icssg_stats_work_handler(struct work_struct *work)
 {
 	struct prueth_emac *emac = container_of(work, struct prueth_emac,
 						stats_work.work);
@@ -51,6 +51,7 @@ void emac_stats_work_handler(struct work_struct *work)
 	queue_delayed_work(system_long_wq, &emac->stats_work,
 			   msecs_to_jiffies((STATS_TIME_LIMIT_1G_MS * 1000) / emac->speed));
 }
+EXPORT_SYMBOL_GPL(icssg_stats_work_handler);
 
 int emac_get_stat_by_name(struct prueth_emac *emac, char *stat_name)
 {
diff --git a/drivers/net/ethernet/ti/icssg/icssg_switchdev.c b/drivers/net/ethernet/ti/icssg/icssg_switchdev.c
index fceb8bb7d34e..67e2927e176d 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_switchdev.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_switchdev.c
@@ -44,7 +44,7 @@ static int prueth_switchdev_stp_state_set(struct prueth_emac *emac,
 		return -EOPNOTSUPP;
 	}
 
-	emac_set_port_state(emac, emac_state);
+	icssg_set_port_state(emac, emac_state);
 	netdev_dbg(emac->ndev, "STP state: %u\n", emac_state);
 
 	return ret;
@@ -64,7 +64,7 @@ static int prueth_switchdev_attr_br_flags_set(struct prueth_emac *emac,
 	netdev_dbg(emac->ndev, "BR_MCAST_FLOOD: %d port %u\n",
 		   emac_state, emac->port_id);
 
-	emac_set_port_state(emac, emac_state);
+	icssg_set_port_state(emac, emac_state);
 
 	return 0;
 }
-- 
cgit v1.2.3-58-ga151


From 1e5b23e50ffbbca91351812d4f7b4d1b151f04b0 Mon Sep 17 00:00:00 2001
From: Pawel Dembicki <paweldembicki@gmail.com>
Date: Sat, 13 Jul 2024 23:16:07 +0200
Subject: net: dsa: vsc73xx: add port_stp_state_set function

This isn't a fully functional implementation of 802.1D, but
port_stp_state_set is required for a future tag8021q operations.

This implementation handles properly all states, but vsc73xx doesn't
forward STP packets.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-2-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/vitesse-vsc73xx-core.c | 96 ++++++++++++++++++++++++++++++----
 1 file changed, 85 insertions(+), 11 deletions(-)

diff --git a/drivers/net/dsa/vitesse-vsc73xx-core.c b/drivers/net/dsa/vitesse-vsc73xx-core.c
index 4b031fefcec6..ebeea259f019 100644
--- a/drivers/net/dsa/vitesse-vsc73xx-core.c
+++ b/drivers/net/dsa/vitesse-vsc73xx-core.c
@@ -164,6 +164,10 @@
 #define VSC73XX_AGENCTRL	0xf0
 #define VSC73XX_CAPRST		0xff
 
+#define VSC73XX_SRCMASKS_CPU_COPY		BIT(27)
+#define VSC73XX_SRCMASKS_MIRROR			BIT(26)
+#define VSC73XX_SRCMASKS_PORTS_MASK		GENMASK(7, 0)
+
 #define VSC73XX_MACACCESS_CPU_COPY		BIT(14)
 #define VSC73XX_MACACCESS_FWD_KILL		BIT(13)
 #define VSC73XX_MACACCESS_IGNORE_VLAN		BIT(12)
@@ -623,9 +627,6 @@ static int vsc73xx_setup(struct dsa_switch *ds)
 	vsc73xx_write(vsc, VSC73XX_BLOCK_SYSTEM, 0, VSC73XX_GMIIDELAY,
 		      VSC73XX_GMIIDELAY_GMII0_GTXDELAY_2_0_NS |
 		      VSC73XX_GMIIDELAY_GMII0_RXDELAY_2_0_NS);
-	/* Enable reception of frames on all ports */
-	vsc73xx_write(vsc, VSC73XX_BLOCK_ANALYZER, 0, VSC73XX_RECVMASK,
-		      0x5f);
 	/* IP multicast flood mask (table 144) */
 	vsc73xx_write(vsc, VSC73XX_BLOCK_ANALYZER, 0, VSC73XX_IFLODMSK,
 		      0xff);
@@ -788,10 +789,6 @@ static void vsc73xx_mac_link_down(struct phylink_config *config,
 	/* Allow backward dropping of frames from this port */
 	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ARBITER, 0,
 			    VSC73XX_SBACKWDROP, BIT(port), BIT(port));
-
-	/* Receive mask (disable forwarding) */
-	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
-			    VSC73XX_RECVMASK, BIT(port), 0);
 }
 
 static void vsc73xx_mac_link_up(struct phylink_config *config,
@@ -844,10 +841,6 @@ static void vsc73xx_mac_link_up(struct phylink_config *config,
 	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ARBITER, 0,
 			    VSC73XX_ARBDISC, BIT(port), 0);
 
-	/* Enable port (forwarding) in the receive mask */
-	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
-			    VSC73XX_RECVMASK, BIT(port), BIT(port));
-
 	/* Disallow backward dropping of frames from this port */
 	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ARBITER, 0,
 			    VSC73XX_SBACKWDROP, BIT(port), 0);
@@ -1039,6 +1032,86 @@ static void vsc73xx_phylink_get_caps(struct dsa_switch *dsa, int port,
 	config->mac_capabilities = MAC_SYM_PAUSE | MAC_10 | MAC_100 | MAC_1000;
 }
 
+static void vsc73xx_refresh_fwd_map(struct dsa_switch *ds, int port, u8 state)
+{
+	struct dsa_port *other_dp, *dp = dsa_to_port(ds, port);
+	struct vsc73xx *vsc = ds->priv;
+	u16 mask;
+
+	if (state != BR_STATE_FORWARDING) {
+		/* Ports that aren't in the forwarding state must not
+		 * forward packets anywhere.
+		 */
+		vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
+				    VSC73XX_SRCMASKS + port,
+				    VSC73XX_SRCMASKS_PORTS_MASK, 0);
+
+		dsa_switch_for_each_available_port(other_dp, ds) {
+			if (other_dp == dp)
+				continue;
+			vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
+					    VSC73XX_SRCMASKS + other_dp->index,
+					    BIT(port), 0);
+		}
+
+		return;
+	}
+
+	/* Forwarding ports must forward to the CPU and to other ports
+	 * in the same bridge
+	 */
+	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
+			    VSC73XX_SRCMASKS + CPU_PORT, BIT(port), BIT(port));
+
+	mask = BIT(CPU_PORT);
+
+	dsa_switch_for_each_user_port(other_dp, ds) {
+		int other_port = other_dp->index;
+
+		if (port == other_port || !dsa_port_bridge_same(dp, other_dp) ||
+		    other_dp->stp_state != BR_STATE_FORWARDING)
+			continue;
+
+		mask |= BIT(other_port);
+
+		vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
+				    VSC73XX_SRCMASKS + other_port,
+				    BIT(port), BIT(port));
+	}
+
+	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
+			    VSC73XX_SRCMASKS + port,
+			    VSC73XX_SRCMASKS_PORTS_MASK, mask);
+}
+
+/* FIXME: STP frames aren't forwarded at this moment. BPDU frames are
+ * forwarded only from and to PI/SI interface. For more info see chapter
+ * 2.7.1 (CPU Forwarding) in datasheet.
+ * This function is required for tag_8021q operations.
+ */
+static void vsc73xx_port_stp_state_set(struct dsa_switch *ds, int port,
+				       u8 state)
+{
+	struct vsc73xx *vsc = ds->priv;
+	u32 val;
+
+	val = (state == BR_STATE_BLOCKING || state == BR_STATE_DISABLED) ?
+	      0 : BIT(port);
+	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
+			    VSC73XX_RECVMASK, BIT(port), val);
+
+	val = (state == BR_STATE_LEARNING || state == BR_STATE_FORWARDING) ?
+	      BIT(port) : 0;
+	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
+			    VSC73XX_LEARNMASK, BIT(port), val);
+
+	/* CPU Port should always forward packets when user ports are forwarding
+	 * so let's configure it from other ports only.
+	 */
+	if (port != CPU_PORT)
+		vsc73xx_refresh_fwd_map(ds, port, state);
+}
+
 static const struct phylink_mac_ops vsc73xx_phylink_mac_ops = {
 	.mac_config = vsc73xx_mac_config,
 	.mac_link_down = vsc73xx_mac_link_down,
@@ -1057,6 +1130,7 @@ static const struct dsa_switch_ops vsc73xx_ds_ops = {
 	.port_disable = vsc73xx_port_disable,
 	.port_change_mtu = vsc73xx_change_mtu,
 	.port_max_mtu = vsc73xx_get_max_mtu,
+	.port_stp_state_set = vsc73xx_port_stp_state_set,
 	.phylink_get_caps = vsc73xx_phylink_get_caps,
 };
 
-- 
cgit v1.2.3-58-ga151


From 6b783ded364ad037dd526c0be5860f58a51ca084 Mon Sep 17 00:00:00 2001
From: Pawel Dembicki <paweldembicki@gmail.com>
Date: Sat, 13 Jul 2024 23:16:08 +0200
Subject: net: dsa: vsc73xx: Add vlan filtering

This patch implements VLAN filtering for the vsc73xx driver.

After starting VLAN filtering, the switch is reconfigured from QinQ to
a simple VLAN aware mode. This is required because VSC73XX chips do not
support inner VLAN tag filtering.

Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-3-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/vitesse-vsc73xx-core.c | 532 ++++++++++++++++++++++++++++++++-
 drivers/net/dsa/vitesse-vsc73xx.h      |  37 +++
 2 files changed, 566 insertions(+), 3 deletions(-)

diff --git a/drivers/net/dsa/vitesse-vsc73xx-core.c b/drivers/net/dsa/vitesse-vsc73xx-core.c
index ebeea259f019..906bbae22861 100644
--- a/drivers/net/dsa/vitesse-vsc73xx-core.c
+++ b/drivers/net/dsa/vitesse-vsc73xx-core.c
@@ -22,9 +22,11 @@
 #include <linux/of_mdio.h>
 #include <linux/bitops.h>
 #include <linux/if_bridge.h>
+#include <linux/if_vlan.h>
 #include <linux/etherdevice.h>
 #include <linux/gpio/consumer.h>
 #include <linux/gpio/driver.h>
+#include <linux/dsa/8021q.h>
 #include <linux/random.h>
 #include <net/dsa.h>
 
@@ -62,6 +64,8 @@
 #define VSC73XX_CAT_DROP	0x6e
 #define VSC73XX_CAT_PR_MISC_L2	0x6f
 #define VSC73XX_CAT_PR_USR_PRIO	0x75
+#define VSC73XX_CAT_VLAN_MISC	0x79
+#define VSC73XX_CAT_PORT_VLAN	0x7a
 #define VSC73XX_Q_MISC_CONF	0xdf
 
 /* MAC_CFG register bits */
@@ -122,6 +126,17 @@
 #define VSC73XX_ADVPORTM_IO_LOOPBACK	BIT(1)
 #define VSC73XX_ADVPORTM_HOST_LOOPBACK	BIT(0)
 
+/*  TXUPDCFG transmit modify setup bits */
+#define VSC73XX_TXUPDCFG_DSCP_REWR_MODE	GENMASK(20, 19)
+#define VSC73XX_TXUPDCFG_DSCP_REWR_ENA	BIT(18)
+#define VSC73XX_TXUPDCFG_TX_INT_TO_USRPRIO_ENA	BIT(17)
+#define VSC73XX_TXUPDCFG_TX_UNTAGGED_VID	GENMASK(15, 4)
+#define VSC73XX_TXUPDCFG_TX_UNTAGGED_VID_ENA	BIT(3)
+#define VSC73XX_TXUPDCFG_TX_UPDATE_CRC_CPU_ENA	BIT(1)
+#define VSC73XX_TXUPDCFG_TX_INSERT_TAG	BIT(0)
+
+#define VSC73XX_TXUPDCFG_TX_UNTAGGED_VID_SHIFT	4
+
 /* CAT_DROP categorizer frame dropping register bits */
 #define VSC73XX_CAT_DROP_DROP_MC_SMAC_ENA	BIT(6)
 #define VSC73XX_CAT_DROP_FWD_CTRL_ENA		BIT(4)
@@ -135,6 +150,15 @@
 #define VSC73XX_Q_MISC_CONF_EARLY_TX_512	(1 << 1)
 #define VSC73XX_Q_MISC_CONF_MAC_PAUSE_MODE	BIT(0)
 
+/* CAT_VLAN_MISC categorizer VLAN miscellaneous bits */
+#define VSC73XX_CAT_VLAN_MISC_VLAN_TCI_IGNORE_ENA BIT(8)
+#define VSC73XX_CAT_VLAN_MISC_VLAN_KEEP_TAG_ENA BIT(7)
+
+/* CAT_PORT_VLAN categorizer port VLAN */
+#define VSC73XX_CAT_PORT_VLAN_VLAN_CFI BIT(15)
+#define VSC73XX_CAT_PORT_VLAN_VLAN_USR_PRIO GENMASK(14, 12)
+#define VSC73XX_CAT_PORT_VLAN_VLAN_VID GENMASK(11, 0)
+
 /* Frame analyzer block 2 registers */
 #define VSC73XX_STORMLIMIT	0x02
 #define VSC73XX_ADVLEARN	0x03
@@ -189,7 +213,8 @@
 #define VSC73XX_VLANACCESS_VLAN_MIRROR		BIT(29)
 #define VSC73XX_VLANACCESS_VLAN_SRC_CHECK	BIT(28)
 #define VSC73XX_VLANACCESS_VLAN_PORT_MASK	GENMASK(9, 2)
-#define VSC73XX_VLANACCESS_VLAN_TBL_CMD_MASK	GENMASK(2, 0)
+#define VSC73XX_VLANACCESS_VLAN_PORT_MASK_SHIFT	2
+#define VSC73XX_VLANACCESS_VLAN_TBL_CMD_MASK	GENMASK(1, 0)
 #define VSC73XX_VLANACCESS_VLAN_TBL_CMD_IDLE	0
 #define VSC73XX_VLANACCESS_VLAN_TBL_CMD_READ_ENTRY	1
 #define VSC73XX_VLANACCESS_VLAN_TBL_CMD_WRITE_ENTRY	2
@@ -347,6 +372,17 @@ static const struct vsc73xx_counter vsc73xx_tx_counters[] = {
 	{ 29, "TxQoSClass3" }, /* non-standard counter */
 };
 
+struct vsc73xx_vlan_summary {
+	size_t num_tagged;
+	size_t num_untagged;
+};
+
+enum vsc73xx_port_vlan_conf {
+	VSC73XX_VLAN_FILTER,
+	VSC73XX_VLAN_FILTER_UNTAG_ALL,
+	VSC73XX_VLAN_IGNORE,
+};
+
 int vsc73xx_is_addr_valid(u8 block, u8 subblock)
 {
 	switch (block) {
@@ -564,6 +600,90 @@ static enum dsa_tag_protocol vsc73xx_get_tag_protocol(struct dsa_switch *ds,
 	return DSA_TAG_PROTO_NONE;
 }
 
+static int vsc73xx_wait_for_vlan_table_cmd(struct vsc73xx *vsc)
+{
+	int ret, err;
+	u32 val;
+
+	ret = read_poll_timeout(vsc73xx_read, err,
+				err < 0 ||
+				((val & VSC73XX_VLANACCESS_VLAN_TBL_CMD_MASK) ==
+				VSC73XX_VLANACCESS_VLAN_TBL_CMD_IDLE),
+				VSC73XX_POLL_SLEEP_US, VSC73XX_POLL_TIMEOUT_US,
+				false, vsc, VSC73XX_BLOCK_ANALYZER,
+				0, VSC73XX_VLANACCESS, &val);
+	if (ret)
+		return ret;
+	return err;
+}
+
+static int
+vsc73xx_read_vlan_table_entry(struct vsc73xx *vsc, u16 vid, u8 *portmap)
+{
+	u32 val;
+	int ret;
+
+	vsc73xx_write(vsc, VSC73XX_BLOCK_ANALYZER, 0, VSC73XX_VLANTIDX, vid);
+
+	ret = vsc73xx_wait_for_vlan_table_cmd(vsc);
+	if (ret)
+		return ret;
+
+	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0, VSC73XX_VLANACCESS,
+			    VSC73XX_VLANACCESS_VLAN_TBL_CMD_MASK,
+			    VSC73XX_VLANACCESS_VLAN_TBL_CMD_READ_ENTRY);
+
+	ret = vsc73xx_wait_for_vlan_table_cmd(vsc);
+	if (ret)
+		return ret;
+
+	vsc73xx_read(vsc, VSC73XX_BLOCK_ANALYZER, 0, VSC73XX_VLANACCESS, &val);
+	*portmap = (val & VSC73XX_VLANACCESS_VLAN_PORT_MASK) >>
+		   VSC73XX_VLANACCESS_VLAN_PORT_MASK_SHIFT;
+
+	return 0;
+}
+
+static int
+vsc73xx_write_vlan_table_entry(struct vsc73xx *vsc, u16 vid, u8 portmap)
+{
+	int ret;
+
+	vsc73xx_write(vsc, VSC73XX_BLOCK_ANALYZER, 0, VSC73XX_VLANTIDX, vid);
+
+	ret = vsc73xx_wait_for_vlan_table_cmd(vsc);
+	if (ret)
+		return ret;
+
+	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0, VSC73XX_VLANACCESS,
+			    VSC73XX_VLANACCESS_VLAN_TBL_CMD_MASK |
+			    VSC73XX_VLANACCESS_VLAN_SRC_CHECK |
+			    VSC73XX_VLANACCESS_VLAN_PORT_MASK,
+			    VSC73XX_VLANACCESS_VLAN_TBL_CMD_WRITE_ENTRY |
+			    VSC73XX_VLANACCESS_VLAN_SRC_CHECK |
+			    (portmap << VSC73XX_VLANACCESS_VLAN_PORT_MASK_SHIFT));
+
+	return vsc73xx_wait_for_vlan_table_cmd(vsc);
+}
+
+static int
+vsc73xx_update_vlan_table(struct vsc73xx *vsc, int port, u16 vid, bool set)
+{
+	u8 portmap;
+	int ret;
+
+	ret = vsc73xx_read_vlan_table_entry(vsc, vid, &portmap);
+	if (ret)
+		return ret;
+
+	if (set)
+		portmap |= BIT(port);
+	else
+		portmap &= ~BIT(port);
+
+	return vsc73xx_write_vlan_table_entry(vsc, vid, portmap);
+}
+
 static int vsc73xx_setup(struct dsa_switch *ds)
 {
 	struct vsc73xx *vsc = ds->priv;
@@ -598,7 +718,7 @@ static int vsc73xx_setup(struct dsa_switch *ds)
 		      VSC73XX_MACACCESS,
 		      VSC73XX_MACACCESS_CMD_CLEAR_TABLE);
 
-	/* Clear VLAN table */
+	/* Set VLAN table to default values */
 	vsc73xx_write(vsc, VSC73XX_BLOCK_ANALYZER, 0,
 		      VSC73XX_VLANACCESS,
 		      VSC73XX_VLANACCESS_VLAN_TBL_CMD_CLEAR_TABLE);
@@ -627,6 +747,9 @@ static int vsc73xx_setup(struct dsa_switch *ds)
 	vsc73xx_write(vsc, VSC73XX_BLOCK_SYSTEM, 0, VSC73XX_GMIIDELAY,
 		      VSC73XX_GMIIDELAY_GMII0_GTXDELAY_2_0_NS |
 		      VSC73XX_GMIIDELAY_GMII0_RXDELAY_2_0_NS);
+	/* Ingess VLAN reception mask (table 145) */
+	vsc73xx_write(vsc, VSC73XX_BLOCK_ANALYZER, 0, VSC73XX_VLANMASK,
+		      0xff);
 	/* IP multicast flood mask (table 144) */
 	vsc73xx_write(vsc, VSC73XX_BLOCK_ANALYZER, 0, VSC73XX_IFLODMSK,
 		      0xff);
@@ -639,6 +762,12 @@ static int vsc73xx_setup(struct dsa_switch *ds)
 
 	udelay(4);
 
+	/* Clear VLAN table */
+	for (i = 0; i < VLAN_N_VID; i++)
+		vsc73xx_write_vlan_table_entry(vsc, i, 0);
+
+	INIT_LIST_HEAD(&vsc->vlans);
+
 	return 0;
 }
 
@@ -825,6 +954,12 @@ static void vsc73xx_mac_link_up(struct phylink_config *config,
 	val |= seed << VSC73XX_MAC_CFG_SEED_OFFSET;
 	val |= VSC73XX_MAC_CFG_SEED_LOAD;
 	val |= VSC73XX_MAC_CFG_WEXC_DIS;
+
+	/* Those bits are responsible for MTU only. Kernel takes care about MTU,
+	 * let's enable +8 bytes frame length unconditionally.
+	 */
+	val |= VSC73XX_MAC_CFG_VLAN_AWR | VSC73XX_MAC_CFG_VLAN_DBLAWR;
+
 	vsc73xx_write(vsc, VSC73XX_BLOCK_MAC, port, VSC73XX_MAC_CFG, val);
 
 	/* Flow control for the PHY facing ports:
@@ -853,6 +988,257 @@ static void vsc73xx_mac_link_up(struct phylink_config *config,
 			    VSC73XX_MAC_CFG_TX_EN | VSC73XX_MAC_CFG_RX_EN);
 }
 
+static bool vsc73xx_tag_8021q_active(struct dsa_port *dp)
+{
+	return !dsa_port_is_vlan_filtering(dp);
+}
+
+static struct vsc73xx_bridge_vlan *
+vsc73xx_bridge_vlan_find(struct vsc73xx *vsc, u16 vid)
+{
+	struct vsc73xx_bridge_vlan *vlan;
+
+	list_for_each_entry(vlan, &vsc->vlans, list)
+		if (vlan->vid == vid)
+			return vlan;
+
+	return NULL;
+}
+
+static void
+vsc73xx_bridge_vlan_remove_port(struct vsc73xx_bridge_vlan *vsc73xx_vlan,
+				int port)
+{
+	vsc73xx_vlan->portmask &= ~BIT(port);
+
+	if (vsc73xx_vlan->portmask)
+		return;
+
+	list_del(&vsc73xx_vlan->list);
+	kfree(vsc73xx_vlan);
+}
+
+static void vsc73xx_bridge_vlan_summary(struct vsc73xx *vsc, int port,
+					struct vsc73xx_vlan_summary *summary,
+					u16 ignored_vid)
+{
+	size_t num_tagged = 0, num_untagged = 0;
+	struct vsc73xx_bridge_vlan *vlan;
+
+	list_for_each_entry(vlan, &vsc->vlans, list) {
+		if (!(vlan->portmask & BIT(port)) || vlan->vid == ignored_vid)
+			continue;
+
+		if (vlan->untagged & BIT(port))
+			num_untagged++;
+		else
+			num_tagged++;
+	}
+
+	summary->num_untagged = num_untagged;
+	summary->num_tagged = num_tagged;
+}
+
+static u16 vsc73xx_find_first_vlan_untagged(struct vsc73xx *vsc, int port)
+{
+	struct vsc73xx_bridge_vlan *vlan;
+
+	list_for_each_entry(vlan, &vsc->vlans, list)
+		if ((vlan->portmask & BIT(port)) &&
+		    (vlan->untagged & BIT(port)))
+			return vlan->vid;
+
+	return VLAN_N_VID;
+}
+
+static int vsc73xx_set_vlan_conf(struct vsc73xx *vsc, int port,
+				 enum vsc73xx_port_vlan_conf port_vlan_conf)
+{
+	u32 val = 0;
+	int ret;
+
+	if (port_vlan_conf == VSC73XX_VLAN_IGNORE)
+		val = VSC73XX_CAT_VLAN_MISC_VLAN_TCI_IGNORE_ENA |
+		      VSC73XX_CAT_VLAN_MISC_VLAN_KEEP_TAG_ENA;
+
+	ret = vsc73xx_update_bits(vsc, VSC73XX_BLOCK_MAC, port,
+				  VSC73XX_CAT_VLAN_MISC,
+				  VSC73XX_CAT_VLAN_MISC_VLAN_TCI_IGNORE_ENA |
+				  VSC73XX_CAT_VLAN_MISC_VLAN_KEEP_TAG_ENA, val);
+	if (ret)
+		return ret;
+
+	val = (port_vlan_conf == VSC73XX_VLAN_FILTER) ?
+	      VSC73XX_TXUPDCFG_TX_INSERT_TAG : 0;
+
+	return vsc73xx_update_bits(vsc, VSC73XX_BLOCK_MAC, port,
+				   VSC73XX_TXUPDCFG,
+				   VSC73XX_TXUPDCFG_TX_INSERT_TAG, val);
+}
+
+/**
+ * vsc73xx_vlan_commit_conf - Update VLAN configuration of a port
+ * @vsc: Switch private data structure
+ * @port: Port index on which to operate
+ *
+ * Update the VLAN behavior of a port to make sure that when it is under
+ * a VLAN filtering bridge, the port is either filtering with tag
+ * preservation, or filtering with all VLANs egress-untagged. Otherwise,
+ * the port ignores VLAN tags from packets and applies the port-based
+ * VID.
+ *
+ * Must be called when changes are made to:
+ * - the bridge VLAN filtering state of the port
+ * - the number or attributes of VLANs from the bridge VLAN table,
+ *   while the port is currently VLAN-aware
+ *
+ * Return: 0 on success, or negative errno on error.
+ */
+static int vsc73xx_vlan_commit_conf(struct vsc73xx *vsc, int port)
+{
+	enum vsc73xx_port_vlan_conf port_vlan_conf = VSC73XX_VLAN_IGNORE;
+	struct dsa_port *dp = dsa_to_port(vsc->ds, port);
+
+	if (port == CPU_PORT) {
+		port_vlan_conf = VSC73XX_VLAN_FILTER;
+	} else if (dsa_port_is_vlan_filtering(dp)) {
+		struct vsc73xx_vlan_summary summary;
+
+		port_vlan_conf = VSC73XX_VLAN_FILTER;
+
+		vsc73xx_bridge_vlan_summary(vsc, port, &summary, VLAN_N_VID);
+		if (summary.num_tagged == 0)
+			port_vlan_conf = VSC73XX_VLAN_FILTER_UNTAG_ALL;
+	}
+
+	return vsc73xx_set_vlan_conf(vsc, port, port_vlan_conf);
+}
+
+static int
+vsc73xx_vlan_change_untagged(struct vsc73xx *vsc, int port, u16 vid, bool set)
+{
+	u32 val = 0;
+
+	if (set)
+		val = VSC73XX_TXUPDCFG_TX_UNTAGGED_VID_ENA |
+		      ((vid << VSC73XX_TXUPDCFG_TX_UNTAGGED_VID_SHIFT) &
+		       VSC73XX_TXUPDCFG_TX_UNTAGGED_VID);
+
+	return vsc73xx_update_bits(vsc, VSC73XX_BLOCK_MAC, port,
+				   VSC73XX_TXUPDCFG,
+				   VSC73XX_TXUPDCFG_TX_UNTAGGED_VID_ENA |
+				   VSC73XX_TXUPDCFG_TX_UNTAGGED_VID, val);
+}
+
+/**
+ * vsc73xx_vlan_commit_untagged - Update native VLAN of a port
+ * @vsc: Switch private data structure
+ * @port: Port index on which to operate
+ *
+ * Update the native VLAN of a port (the one VLAN which is transmitted
+ * as egress-tagged on a trunk port) when port is in VLAN filtering mode and
+ * only one untagged vid is configured.
+ * In other cases no need to configure it because switch can untag all vlans on
+ * the port.
+ *
+ * Return: 0 on success, or negative errno on error.
+ */
+static int vsc73xx_vlan_commit_untagged(struct vsc73xx *vsc, int port)
+{
+	struct dsa_port *dp = dsa_to_port(vsc->ds, port);
+	struct vsc73xx_vlan_summary summary;
+	u16 vid = 0;
+	bool valid;
+
+	if (!dsa_port_is_vlan_filtering(dp))
+		/* Port is configured to untag all vlans in that case.
+		 * No need to commit untagged config change.
+		 */
+		return 0;
+
+	vsc73xx_bridge_vlan_summary(vsc, port, &summary, VLAN_N_VID);
+
+	if (summary.num_untagged > 1)
+		/* Port must untag all vlans in that case.
+		 * No need to commit untagged config change.
+		 */
+		return 0;
+
+	valid = (summary.num_untagged == 1);
+	if (valid)
+		vid = vsc73xx_find_first_vlan_untagged(vsc, port);
+
+	return vsc73xx_vlan_change_untagged(vsc, port, vid, valid);
+}
+
+static int
+vsc73xx_vlan_change_pvid(struct vsc73xx *vsc, int port, u16 vid, bool set)
+{
+	u32 val = 0;
+	int ret;
+
+	val = set ? 0 : VSC73XX_CAT_DROP_UNTAGGED_ENA;
+
+	ret = vsc73xx_update_bits(vsc, VSC73XX_BLOCK_MAC, port,
+				  VSC73XX_CAT_DROP,
+				  VSC73XX_CAT_DROP_UNTAGGED_ENA, val);
+	if (!set || ret)
+		return ret;
+
+	return vsc73xx_update_bits(vsc, VSC73XX_BLOCK_MAC, port,
+				   VSC73XX_CAT_PORT_VLAN,
+				   VSC73XX_CAT_PORT_VLAN_VLAN_VID,
+				   vid & VSC73XX_CAT_PORT_VLAN_VLAN_VID);
+}
+
+/**
+ * vsc73xx_vlan_commit_pvid - Update port-based default VLAN of a port
+ * @vsc: Switch private data structure
+ * @port: Port index on which to operate
+ *
+ * Update the PVID of a port so that it follows either the bridge PVID
+ * configuration, when the bridge is currently VLAN-aware, or the PVID
+ * from tag_8021q, when the port is standalone or under a VLAN-unaware
+ * bridge. A port with no PVID drops all untagged and VID 0 tagged
+ * traffic.
+ *
+ * Must be called when changes are made to:
+ * - the bridge VLAN filtering state of the port
+ * - the number or attributes of VLANs from the bridge VLAN table,
+ *   while the port is currently VLAN-aware
+ *
+ * Return: 0 on success, or negative errno on error.
+ */
+static int vsc73xx_vlan_commit_pvid(struct vsc73xx *vsc, int port)
+{
+	struct vsc73xx_portinfo *portinfo = &vsc->portinfo[port];
+	bool valid = portinfo->pvid_tag_8021q_configured;
+	struct dsa_port *dp = dsa_to_port(vsc->ds, port);
+	u16 vid = portinfo->pvid_tag_8021q;
+
+	if (dsa_port_is_vlan_filtering(dp)) {
+		vid = portinfo->pvid_vlan_filtering;
+		valid = portinfo->pvid_vlan_filtering_configured;
+	}
+
+	return vsc73xx_vlan_change_pvid(vsc, port, vid, valid);
+}
+
+static int vsc73xx_vlan_commit_settings(struct vsc73xx *vsc, int port)
+{
+	int ret;
+
+	ret = vsc73xx_vlan_commit_untagged(vsc, port);
+	if (ret)
+		return ret;
+
+	ret = vsc73xx_vlan_commit_pvid(vsc, port);
+	if (ret)
+		return ret;
+
+	return vsc73xx_vlan_commit_conf(vsc, port);
+}
+
 static int vsc73xx_port_enable(struct dsa_switch *ds, int port,
 			       struct phy_device *phy)
 {
@@ -861,7 +1247,7 @@ static int vsc73xx_port_enable(struct dsa_switch *ds, int port,
 	dev_info(vsc->dev, "enable port %d\n", port);
 	vsc73xx_init_port(vsc, port);
 
-	return 0;
+	return vsc73xx_vlan_commit_settings(vsc, port);
 }
 
 static void vsc73xx_port_disable(struct dsa_switch *ds, int port)
@@ -1032,6 +1418,143 @@ static void vsc73xx_phylink_get_caps(struct dsa_switch *dsa, int port,
 	config->mac_capabilities = MAC_SYM_PAUSE | MAC_10 | MAC_100 | MAC_1000;
 }
 
+static int
+vsc73xx_port_vlan_filtering(struct dsa_switch *ds, int port,
+			    bool vlan_filtering, struct netlink_ext_ack *extack)
+{
+	struct vsc73xx *vsc = ds->priv;
+
+	/* The commit to hardware processed below is required because vsc73xx
+	 * is using tag_8021q. When vlan_filtering is disabled, tag_8021q uses
+	 * pvid/untagged vlans for port recognition. The values configured for
+	 * vlans and pvid/untagged states are stored in portinfo structure.
+	 * When vlan_filtering is enabled, we need to restore pvid/untagged from
+	 * portinfo structure. Analogous routine is processed when
+	 * vlan_filtering is disabled, but values used for tag_8021q are
+	 * restored.
+	 */
+
+	return vsc73xx_vlan_commit_settings(vsc, port);
+}
+
+static int vsc73xx_port_vlan_add(struct dsa_switch *ds, int port,
+				 const struct switchdev_obj_port_vlan *vlan,
+				 struct netlink_ext_ack *extack)
+{
+	bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED;
+	bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID;
+	struct dsa_port *dp = dsa_to_port(ds, port);
+	struct vsc73xx_bridge_vlan *vsc73xx_vlan;
+	struct vsc73xx_vlan_summary summary;
+	struct vsc73xx_portinfo *portinfo;
+	struct vsc73xx *vsc = ds->priv;
+	bool commit_to_hardware;
+	int ret = 0;
+
+	/* Be sure to deny alterations to the configuration done by tag_8021q.
+	 */
+	if (vid_is_dsa_8021q(vlan->vid)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Range 3072-4095 reserved for dsa_8021q operation");
+		return -EBUSY;
+	}
+
+	/* The processed vlan->vid is excluded from the search because the VLAN
+	 * can be re-added with a different set of flags, so it's easiest to
+	 * ignore its old flags from the VLAN database software copy.
+	 */
+	vsc73xx_bridge_vlan_summary(vsc, port, &summary, vlan->vid);
+
+	/* VSC73XX allows only three untagged states: none, one or all */
+	if ((untagged && summary.num_tagged > 0 && summary.num_untagged > 0) ||
+	    (!untagged && summary.num_untagged > 1)) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Port can have only none, one or all untagged vlan");
+		return -EBUSY;
+	}
+
+	vsc73xx_vlan = vsc73xx_bridge_vlan_find(vsc, vlan->vid);
+
+	if (!vsc73xx_vlan) {
+		vsc73xx_vlan = kzalloc(sizeof(*vsc73xx_vlan), GFP_KERNEL);
+		if (!vsc73xx_vlan)
+			return -ENOMEM;
+
+		vsc73xx_vlan->vid = vlan->vid;
+
+		list_add_tail(&vsc73xx_vlan->list, &vsc->vlans);
+	}
+
+	vsc73xx_vlan->portmask |= BIT(port);
+
+	/* CPU port must be always tagged because source port identification is
+	 * based on tag_8021q.
+	 */
+	if (port == CPU_PORT)
+		goto update_vlan_table;
+
+	if (untagged)
+		vsc73xx_vlan->untagged |= BIT(port);
+	else
+		vsc73xx_vlan->untagged &= ~BIT(port);
+
+	portinfo = &vsc->portinfo[port];
+
+	if (pvid) {
+		portinfo->pvid_vlan_filtering_configured = true;
+		portinfo->pvid_vlan_filtering = vlan->vid;
+	} else if (portinfo->pvid_vlan_filtering_configured &&
+		   portinfo->pvid_vlan_filtering == vlan->vid) {
+		portinfo->pvid_vlan_filtering_configured = false;
+	}
+
+	commit_to_hardware = !vsc73xx_tag_8021q_active(dp);
+	if (commit_to_hardware) {
+		ret = vsc73xx_vlan_commit_settings(vsc, port);
+		if (ret)
+			goto err;
+	}
+
+update_vlan_table:
+	ret = vsc73xx_update_vlan_table(vsc, port, vlan->vid, true);
+	if (!ret)
+		return 0;
+err:
+	vsc73xx_bridge_vlan_remove_port(vsc73xx_vlan, port);
+	return ret;
+}
+
+static int vsc73xx_port_vlan_del(struct dsa_switch *ds, int port,
+				 const struct switchdev_obj_port_vlan *vlan)
+{
+	struct vsc73xx_bridge_vlan *vsc73xx_vlan;
+	struct vsc73xx_portinfo *portinfo;
+	struct vsc73xx *vsc = ds->priv;
+	bool commit_to_hardware;
+	int ret;
+
+	ret = vsc73xx_update_vlan_table(vsc, port, vlan->vid, false);
+	if (ret)
+		return ret;
+
+	portinfo = &vsc->portinfo[port];
+
+	if (portinfo->pvid_vlan_filtering_configured &&
+	    portinfo->pvid_vlan_filtering == vlan->vid)
+		portinfo->pvid_vlan_filtering_configured = false;
+
+	vsc73xx_vlan = vsc73xx_bridge_vlan_find(vsc, vlan->vid);
+
+	if (vsc73xx_vlan)
+		vsc73xx_bridge_vlan_remove_port(vsc73xx_vlan, port);
+
+	commit_to_hardware = !vsc73xx_tag_8021q_active(dsa_to_port(ds, port));
+	if (commit_to_hardware)
+		return vsc73xx_vlan_commit_settings(vsc, port);
+
+	return 0;
+}
+
 static void vsc73xx_refresh_fwd_map(struct dsa_switch *ds, int port, u8 state)
 {
 	struct dsa_port *other_dp, *dp = dsa_to_port(ds, port);
@@ -1131,6 +1654,9 @@ static const struct dsa_switch_ops vsc73xx_ds_ops = {
 	.port_change_mtu = vsc73xx_change_mtu,
 	.port_max_mtu = vsc73xx_get_max_mtu,
 	.port_stp_state_set = vsc73xx_port_stp_state_set,
+	.port_vlan_filtering = vsc73xx_port_vlan_filtering,
+	.port_vlan_add = vsc73xx_port_vlan_add,
+	.port_vlan_del = vsc73xx_port_vlan_del,
 	.phylink_get_caps = vsc73xx_phylink_get_caps,
 };
 
diff --git a/drivers/net/dsa/vitesse-vsc73xx.h b/drivers/net/dsa/vitesse-vsc73xx.h
index 2997f7e108b1..3ca579acc798 100644
--- a/drivers/net/dsa/vitesse-vsc73xx.h
+++ b/drivers/net/dsa/vitesse-vsc73xx.h
@@ -14,6 +14,22 @@
  */
 #define VSC73XX_MAX_NUM_PORTS	8
 
+/**
+ * struct vsc73xx_portinfo - port data structure: contains storage data
+ * @pvid_vlan_filtering: pvid vlan number used in vlan filtering mode
+ * @pvid_tag_8021q: pvid vlan number used in tag_8021q mode
+ * @pvid_vlan_filtering_configured: informs if port has configured pvid in vlan
+ *	filtering mode
+ * @pvid_tag_8021q_configured: imforms if port have configured pvid in tag_8021q
+ *	mode
+ */
+struct vsc73xx_portinfo {
+	u16		pvid_vlan_filtering;
+	u16		pvid_tag_8021q;
+	bool		pvid_vlan_filtering_configured;
+	bool		pvid_tag_8021q_configured;
+};
+
 /**
  * struct vsc73xx - VSC73xx state container: main data structure
  * @dev: The device pointer
@@ -25,6 +41,10 @@
  * @addr: MAC address used in flow control frames
  * @ops: Structure with hardware-dependent operations
  * @priv: Pointer to the configuration interface structure
+ * @portinfo: Storage table portinfo structructures
+ * @vlans: List of configured vlans. Contains port mask and untagged status of
+ *	every vlan configured in port vlan operation. It doesn't cover tag_8021q
+ *	vlans.
  */
 struct vsc73xx {
 	struct device			*dev;
@@ -35,6 +55,8 @@ struct vsc73xx {
 	u8				addr[ETH_ALEN];
 	const struct vsc73xx_ops	*ops;
 	void				*priv;
+	struct vsc73xx_portinfo		portinfo[VSC73XX_MAX_NUM_PORTS];
+	struct list_head		vlans;
 };
 
 /**
@@ -49,6 +71,21 @@ struct vsc73xx_ops {
 		     u32 val);
 };
 
+/**
+ * struct vsc73xx_bridge_vlan - VSC73xx driver structure which keeps vlan
+ *	database copy
+ * @vid: VLAN number
+ * @portmask: each bit represents one port
+ * @untagged: each bit represents one port configured with @vid untagged
+ * @list: list structure
+ */
+struct vsc73xx_bridge_vlan {
+	u16 vid;
+	u8 portmask;
+	u8 untagged;
+	struct list_head list;
+};
+
 int vsc73xx_is_addr_valid(u8 block, u8 subblock);
 int vsc73xx_probe(struct vsc73xx *vsc);
 void vsc73xx_remove(struct vsc73xx *vsc);
-- 
cgit v1.2.3-58-ga151


From dcfe7673787b4bfea2c213df443d312aa754757b Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 13 Jul 2024 23:16:09 +0200
Subject: net: dsa: tag_sja1105: absorb logic for not overwriting precise info
 into dsa_8021q_rcv()

In both sja1105_rcv() and sja1110_rcv(), we may have precise source port
information coming from parallel hardware mechanisms, in addition to the
tag_8021q header.

Only sja1105_rcv() has extra logic to not overwrite that precise info
with what's present in the VLAN tag. This is because sja1110_rcv() gets
by, by having a reversed set of checks when assigning skb->dev. When the
source port is imprecise (vbid >=1), source_port and switch_id will be
set to zeroes by dsa_8021q_rcv(), which might be problematic. But by
checking for vbid >= 1 first, sja1110_rcv() fends that off.

We would like to make more code common between sja1105_rcv() and
sja1110_rcv(), and for that, we need to make sure that sja1110_rcv()
also goes through the precise source port preservation logic.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Tested-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-4-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/dsa/tag_8021q.c   | 32 +++++++++++++++++++++++++++++---
 net/dsa/tag_sja1105.c | 23 +++--------------------
 2 files changed, 32 insertions(+), 23 deletions(-)

diff --git a/net/dsa/tag_8021q.c b/net/dsa/tag_8021q.c
index 71b26ae6db39..3cb0293793a5 100644
--- a/net/dsa/tag_8021q.c
+++ b/net/dsa/tag_8021q.c
@@ -497,9 +497,21 @@ struct net_device *dsa_tag_8021q_find_port_by_vbid(struct net_device *conduit,
 }
 EXPORT_SYMBOL_GPL(dsa_tag_8021q_find_port_by_vbid);
 
+/**
+ * dsa_8021q_rcv - Decode source information from tag_8021q header
+ * @skb: RX socket buffer
+ * @source_port: pointer to storage for precise source port information.
+ *	If this is known already from outside tag_8021q, the pre-initialized
+ *	value is preserved. If not known, pass -1.
+ * @switch_id: similar to source_port.
+ * @vbid: pointer to storage for imprecise bridge ID. Must be pre-initialized
+ *	with -1. If a positive value is returned, the source_port and switch_id
+ *	are invalid.
+ */
 void dsa_8021q_rcv(struct sk_buff *skb, int *source_port, int *switch_id,
 		   int *vbid)
 {
+	int tmp_source_port, tmp_switch_id, tmp_vbid;
 	u16 vid, tci;
 
 	if (skb_vlan_tag_present(skb)) {
@@ -513,11 +525,25 @@ void dsa_8021q_rcv(struct sk_buff *skb, int *source_port, int *switch_id,
 
 	vid = tci & VLAN_VID_MASK;
 
-	*source_port = dsa_8021q_rx_source_port(vid);
-	*switch_id = dsa_8021q_rx_switch_id(vid);
+	tmp_source_port = dsa_8021q_rx_source_port(vid);
+	tmp_switch_id = dsa_8021q_rx_switch_id(vid);
+	tmp_vbid = dsa_tag_8021q_rx_vbid(vid);
+
+	/* Precise source port information is unknown when receiving from a
+	 * VLAN-unaware bridging domain, and tmp_source_port and tmp_switch_id
+	 * are zeroes in this case.
+	 *
+	 * Preserve the source information from hardware-specific mechanisms,
+	 * if available. This allows us to not overwrite a valid source port
+	 * and switch ID with less precise values.
+	 */
+	if (tmp_vbid == 0 && *source_port == -1)
+		*source_port = tmp_source_port;
+	if (tmp_vbid == 0 && *switch_id == -1)
+		*switch_id = tmp_switch_id;
 
 	if (vbid)
-		*vbid = dsa_tag_8021q_rx_vbid(vid);
+		*vbid = tmp_vbid;
 
 	skb->priority = (tci & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT;
 }
diff --git a/net/dsa/tag_sja1105.c b/net/dsa/tag_sja1105.c
index 1aba1d05c27a..48886d4b7e3e 100644
--- a/net/dsa/tag_sja1105.c
+++ b/net/dsa/tag_sja1105.c
@@ -524,30 +524,13 @@ static struct sk_buff *sja1105_rcv(struct sk_buff *skb,
 	/* Normal data plane traffic and link-local frames are tagged with
 	 * a tag_8021q VLAN which we have to strip
 	 */
-	if (sja1105_skb_has_tag_8021q(skb)) {
-		int tmp_source_port = -1, tmp_switch_id = -1;
-
-		sja1105_vlan_rcv(skb, &tmp_source_port, &tmp_switch_id, &vbid,
-				 &vid);
-		/* Preserve the source information from the INCL_SRCPT option,
-		 * if available. This allows us to not overwrite a valid source
-		 * port and switch ID with zeroes when receiving link-local
-		 * frames from a VLAN-unaware bridged port (non-zero vbid) or a
-		 * VLAN-aware bridged port (non-zero vid). Furthermore, the
-		 * tag_8021q source port information is only of trust when the
-		 * vbid is 0 (precise port). Otherwise, tmp_source_port and
-		 * tmp_switch_id will be zeroes.
-		 */
-		if (vbid == 0 && source_port == -1)
-			source_port = tmp_source_port;
-		if (vbid == 0 && switch_id == -1)
-			switch_id = tmp_switch_id;
-	} else if (source_port == -1 && switch_id == -1) {
+	if (sja1105_skb_has_tag_8021q(skb))
+		sja1105_vlan_rcv(skb, &source_port, &switch_id, &vbid, &vid);
+	else if (source_port == -1 && switch_id == -1)
 		/* Packets with no source information have no chance of
 		 * getting accepted, drop them straight away.
 		 */
 		return NULL;
-	}
 
 	if (source_port != -1 && switch_id != -1)
 		skb->dev = dsa_conduit_find_user(netdev, switch_id, source_port);
-- 
cgit v1.2.3-58-ga151


From 0064b863abdc7bd5e209cdbce7619464069458fe Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 13 Jul 2024 23:16:10 +0200
Subject: net: dsa: tag_sja1105: absorb entire sja1105_vlan_rcv() into
 dsa_8021q_rcv()

tag_sja1105 has a wrapper over dsa_8021q_rcv(): sja1105_vlan_rcv(),
which determines whether the packet came from a bridge with
vlan_filtering=1 (the case resolved via
dsa_find_designated_bridge_port_by_vid()), or if it contains a tag_8021q
header.

Looking at a new tagger implementation for vsc73xx, based also on
tag_8021q, it is becoming clear that the logic is needed there as well.
So instead of forcing each tagger to wrap around dsa_8021q_rcv(), let's
merge the logic into the core.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Tested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-5-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/dsa/tag_8021q.c        | 34 ++++++++++++++++++++++++++++------
 net/dsa/tag_8021q.h        |  2 +-
 net/dsa/tag_ocelot_8021q.c |  2 +-
 net/dsa/tag_sja1105.c      | 32 ++++----------------------------
 4 files changed, 34 insertions(+), 36 deletions(-)

diff --git a/net/dsa/tag_8021q.c b/net/dsa/tag_8021q.c
index 3cb0293793a5..2d1c554a63ff 100644
--- a/net/dsa/tag_8021q.c
+++ b/net/dsa/tag_8021q.c
@@ -507,27 +507,48 @@ EXPORT_SYMBOL_GPL(dsa_tag_8021q_find_port_by_vbid);
  * @vbid: pointer to storage for imprecise bridge ID. Must be pre-initialized
  *	with -1. If a positive value is returned, the source_port and switch_id
  *	are invalid.
+ * @vid: pointer to storage for original VID, in case tag_8021q decoding failed.
+ *
+ * If the packet has a tag_8021q header, decode it and set @source_port,
+ * @switch_id and @vbid, and strip the header. Otherwise set @vid and keep the
+ * header in the hwaccel area of the packet.
  */
 void dsa_8021q_rcv(struct sk_buff *skb, int *source_port, int *switch_id,
-		   int *vbid)
+		   int *vbid, int *vid)
 {
 	int tmp_source_port, tmp_switch_id, tmp_vbid;
-	u16 vid, tci;
+	__be16 vlan_proto;
+	u16 tmp_vid, tci;
 
 	if (skb_vlan_tag_present(skb)) {
+		vlan_proto = skb->vlan_proto;
 		tci = skb_vlan_tag_get(skb);
 		__vlan_hwaccel_clear_tag(skb);
 	} else {
+		struct vlan_ethhdr *hdr = vlan_eth_hdr(skb);
+
+		vlan_proto = hdr->h_vlan_proto;
 		skb_push_rcsum(skb, ETH_HLEN);
 		__skb_vlan_pop(skb, &tci);
 		skb_pull_rcsum(skb, ETH_HLEN);
 	}
 
-	vid = tci & VLAN_VID_MASK;
+	tmp_vid = tci & VLAN_VID_MASK;
+	if (!vid_is_dsa_8021q(tmp_vid)) {
+		/* Not a tag_8021q frame, so return the VID to the
+		 * caller for further processing, and put the tag back
+		 */
+		if (vid)
+			*vid = tmp_vid;
+
+		__vlan_hwaccel_put_tag(skb, vlan_proto, tci);
+
+		return;
+	}
 
-	tmp_source_port = dsa_8021q_rx_source_port(vid);
-	tmp_switch_id = dsa_8021q_rx_switch_id(vid);
-	tmp_vbid = dsa_tag_8021q_rx_vbid(vid);
+	tmp_source_port = dsa_8021q_rx_source_port(tmp_vid);
+	tmp_switch_id = dsa_8021q_rx_switch_id(tmp_vid);
+	tmp_vbid = dsa_tag_8021q_rx_vbid(tmp_vid);
 
 	/* Precise source port information is unknown when receiving from a
 	 * VLAN-unaware bridging domain, and tmp_source_port and tmp_switch_id
@@ -546,5 +567,6 @@ void dsa_8021q_rcv(struct sk_buff *skb, int *source_port, int *switch_id,
 		*vbid = tmp_vbid;
 
 	skb->priority = (tci & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT;
+	return;
 }
 EXPORT_SYMBOL_GPL(dsa_8021q_rcv);
diff --git a/net/dsa/tag_8021q.h b/net/dsa/tag_8021q.h
index 41f7167ac520..0c6671d7c1c2 100644
--- a/net/dsa/tag_8021q.h
+++ b/net/dsa/tag_8021q.h
@@ -14,7 +14,7 @@ struct sk_buff *dsa_8021q_xmit(struct sk_buff *skb, struct net_device *netdev,
 			       u16 tpid, u16 tci);
 
 void dsa_8021q_rcv(struct sk_buff *skb, int *source_port, int *switch_id,
-		   int *vbid);
+		   int *vbid, int *vid);
 
 struct net_device *dsa_tag_8021q_find_port_by_vbid(struct net_device *conduit,
 						   int vbid);
diff --git a/net/dsa/tag_ocelot_8021q.c b/net/dsa/tag_ocelot_8021q.c
index b059381310fe..8e8b1bef6af6 100644
--- a/net/dsa/tag_ocelot_8021q.c
+++ b/net/dsa/tag_ocelot_8021q.c
@@ -81,7 +81,7 @@ static struct sk_buff *ocelot_rcv(struct sk_buff *skb,
 {
 	int src_port, switch_id;
 
-	dsa_8021q_rcv(skb, &src_port, &switch_id, NULL);
+	dsa_8021q_rcv(skb, &src_port, &switch_id, NULL, NULL);
 
 	skb->dev = dsa_conduit_find_user(netdev, switch_id, src_port);
 	if (!skb->dev)
diff --git a/net/dsa/tag_sja1105.c b/net/dsa/tag_sja1105.c
index 48886d4b7e3e..7639ccb94d35 100644
--- a/net/dsa/tag_sja1105.c
+++ b/net/dsa/tag_sja1105.c
@@ -472,37 +472,14 @@ static bool sja1110_skb_has_inband_control_extension(const struct sk_buff *skb)
 	return ntohs(eth_hdr(skb)->h_proto) == ETH_P_SJA1110;
 }
 
-/* If the VLAN in the packet is a tag_8021q one, set @source_port and
- * @switch_id and strip the header. Otherwise set @vid and keep it in the
- * packet.
- */
-static void sja1105_vlan_rcv(struct sk_buff *skb, int *source_port,
-			     int *switch_id, int *vbid, u16 *vid)
-{
-	struct vlan_ethhdr *hdr = vlan_eth_hdr(skb);
-	u16 vlan_tci;
-
-	if (skb_vlan_tag_present(skb))
-		vlan_tci = skb_vlan_tag_get(skb);
-	else
-		vlan_tci = ntohs(hdr->h_vlan_TCI);
-
-	if (vid_is_dsa_8021q(vlan_tci & VLAN_VID_MASK))
-		return dsa_8021q_rcv(skb, source_port, switch_id, vbid);
-
-	/* Try our best with imprecise RX */
-	*vid = vlan_tci & VLAN_VID_MASK;
-}
-
 static struct sk_buff *sja1105_rcv(struct sk_buff *skb,
 				   struct net_device *netdev)
 {
-	int source_port = -1, switch_id = -1, vbid = -1;
+	int source_port = -1, switch_id = -1, vbid = -1, vid = -1;
 	struct sja1105_meta meta = {0};
 	struct ethhdr *hdr;
 	bool is_link_local;
 	bool is_meta;
-	u16 vid;
 
 	hdr = eth_hdr(skb);
 	is_link_local = sja1105_is_link_local(skb);
@@ -525,7 +502,7 @@ static struct sk_buff *sja1105_rcv(struct sk_buff *skb,
 	 * a tag_8021q VLAN which we have to strip
 	 */
 	if (sja1105_skb_has_tag_8021q(skb))
-		sja1105_vlan_rcv(skb, &source_port, &switch_id, &vbid, &vid);
+		dsa_8021q_rcv(skb, &source_port, &switch_id, &vbid, &vid);
 	else if (source_port == -1 && switch_id == -1)
 		/* Packets with no source information have no chance of
 		 * getting accepted, drop them straight away.
@@ -660,9 +637,8 @@ static struct sk_buff *sja1110_rcv_inband_control_extension(struct sk_buff *skb,
 static struct sk_buff *sja1110_rcv(struct sk_buff *skb,
 				   struct net_device *netdev)
 {
-	int source_port = -1, switch_id = -1, vbid = -1;
+	int source_port = -1, switch_id = -1, vbid = -1, vid = -1;
 	bool host_only = false;
-	u16 vid = 0;
 
 	if (sja1110_skb_has_inband_control_extension(skb)) {
 		skb = sja1110_rcv_inband_control_extension(skb, &source_port,
@@ -674,7 +650,7 @@ static struct sk_buff *sja1110_rcv(struct sk_buff *skb,
 
 	/* Packets with in-band control extensions might still have RX VLANs */
 	if (likely(sja1105_skb_has_tag_8021q(skb)))
-		sja1105_vlan_rcv(skb, &source_port, &switch_id, &vbid, &vid);
+		dsa_8021q_rcv(skb, &source_port, &switch_id, &vbid, &vid);
 
 	if (vbid >= 1)
 		skb->dev = dsa_tag_8021q_find_port_by_vbid(netdev, vbid);
-- 
cgit v1.2.3-58-ga151


From 823e5cc141c6408b470b8302dd8b5082636fbfaf Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 13 Jul 2024 23:16:11 +0200
Subject: net: dsa: tag_sja1105: prefer precise source port info on SJA1110 too

Now that dsa_8021q_rcv() handles better the case where we don't
overwrite the precise source information if it comes from an external
(non-tag_8021q) source, we can now unify the call sequence between
sja1105_rcv() and sja1110_rcv().

This is a preparatory change for creating a higher-level wrapper for the
entire sequence which will live in tag_8021q.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-6-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/dsa/tag_sja1105.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/dsa/tag_sja1105.c b/net/dsa/tag_sja1105.c
index 7639ccb94d35..35a6346549f2 100644
--- a/net/dsa/tag_sja1105.c
+++ b/net/dsa/tag_sja1105.c
@@ -652,12 +652,12 @@ static struct sk_buff *sja1110_rcv(struct sk_buff *skb,
 	if (likely(sja1105_skb_has_tag_8021q(skb)))
 		dsa_8021q_rcv(skb, &source_port, &switch_id, &vbid, &vid);
 
-	if (vbid >= 1)
+	if (source_port != -1 && switch_id != -1)
+		skb->dev = dsa_conduit_find_user(netdev, switch_id, source_port);
+	else if (vbid >= 1)
 		skb->dev = dsa_tag_8021q_find_port_by_vbid(netdev, vbid);
-	else if (source_port == -1 || switch_id == -1)
-		skb->dev = dsa_find_designated_bridge_port_by_vid(netdev, vid);
 	else
-		skb->dev = dsa_conduit_find_user(netdev, switch_id, source_port);
+		skb->dev = dsa_find_designated_bridge_port_by_vid(netdev, vid);
 	if (!skb->dev) {
 		netdev_warn(netdev, "Couldn't decode source port\n");
 		return NULL;
-- 
cgit v1.2.3-58-ga151


From d124cf54df6fafa5ab63d3ee40d64bd54487b5cc Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean@nxp.com>
Date: Sat, 13 Jul 2024 23:16:12 +0200
Subject: net: dsa: tag_sja1105: refactor skb->dev assignment to
 dsa_tag_8021q_find_user()

A new tagging protocol implementation based on tag_8021q is on the
horizon, and it appears that it also has to open-code the complicated
logic of finding a source port based on a VLAN header.

Create a single dsa_tag_8021q_find_user() and make sja1105 call it.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://patch.msgid.link/20240713211620.1125910-7-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/dsa/tag_8021q.c   | 19 ++++++++++++++++---
 net/dsa/tag_8021q.h   |  5 +++--
 net/dsa/tag_sja1105.c | 17 +++++------------
 3 files changed, 24 insertions(+), 17 deletions(-)

diff --git a/net/dsa/tag_8021q.c b/net/dsa/tag_8021q.c
index 2d1c554a63ff..c0eee113a2b9 100644
--- a/net/dsa/tag_8021q.c
+++ b/net/dsa/tag_8021q.c
@@ -468,8 +468,8 @@ struct sk_buff *dsa_8021q_xmit(struct sk_buff *skb, struct net_device *netdev,
 }
 EXPORT_SYMBOL_GPL(dsa_8021q_xmit);
 
-struct net_device *dsa_tag_8021q_find_port_by_vbid(struct net_device *conduit,
-						   int vbid)
+static struct net_device *
+dsa_tag_8021q_find_port_by_vbid(struct net_device *conduit, int vbid)
 {
 	struct dsa_port *cpu_dp = conduit->dsa_ptr;
 	struct dsa_switch_tree *dst = cpu_dp->dst;
@@ -495,7 +495,20 @@ struct net_device *dsa_tag_8021q_find_port_by_vbid(struct net_device *conduit,
 
 	return NULL;
 }
-EXPORT_SYMBOL_GPL(dsa_tag_8021q_find_port_by_vbid);
+
+struct net_device *dsa_tag_8021q_find_user(struct net_device *conduit,
+					   int source_port, int switch_id,
+					   int vid, int vbid)
+{
+	/* Always prefer precise source port information, if available */
+	if (source_port != -1 && switch_id != -1)
+		return dsa_conduit_find_user(conduit, switch_id, source_port);
+	else if (vbid >= 1)
+		return dsa_tag_8021q_find_port_by_vbid(conduit, vbid);
+
+	return dsa_find_designated_bridge_port_by_vid(conduit, vid);
+}
+EXPORT_SYMBOL_GPL(dsa_tag_8021q_find_user);
 
 /**
  * dsa_8021q_rcv - Decode source information from tag_8021q header
diff --git a/net/dsa/tag_8021q.h b/net/dsa/tag_8021q.h
index 0c6671d7c1c2..27b8906f99ec 100644
--- a/net/dsa/tag_8021q.h
+++ b/net/dsa/tag_8021q.h
@@ -16,8 +16,9 @@ struct sk_buff *dsa_8021q_xmit(struct sk_buff *skb, struct net_device *netdev,
 void dsa_8021q_rcv(struct sk_buff *skb, int *source_port, int *switch_id,
 		   int *vbid, int *vid);
 
-struct net_device *dsa_tag_8021q_find_port_by_vbid(struct net_device *conduit,
-						   int vbid);
+struct net_device *dsa_tag_8021q_find_user(struct net_device *conduit,
+					   int source_port, int switch_id,
+					   int vid, int vbid);
 
 int dsa_switch_tag_8021q_vlan_add(struct dsa_switch *ds,
 				  struct dsa_notifier_tag_8021q_vlan_info *info);
diff --git a/net/dsa/tag_sja1105.c b/net/dsa/tag_sja1105.c
index 35a6346549f2..3e902af7eea6 100644
--- a/net/dsa/tag_sja1105.c
+++ b/net/dsa/tag_sja1105.c
@@ -509,12 +509,8 @@ static struct sk_buff *sja1105_rcv(struct sk_buff *skb,
 		 */
 		return NULL;
 
-	if (source_port != -1 && switch_id != -1)
-		skb->dev = dsa_conduit_find_user(netdev, switch_id, source_port);
-	else if (vbid >= 1)
-		skb->dev = dsa_tag_8021q_find_port_by_vbid(netdev, vbid);
-	else
-		skb->dev = dsa_find_designated_bridge_port_by_vid(netdev, vid);
+	skb->dev = dsa_tag_8021q_find_user(netdev, source_port, switch_id,
+					   vid, vbid);
 	if (!skb->dev) {
 		netdev_warn(netdev, "Couldn't decode source port\n");
 		return NULL;
@@ -652,12 +648,9 @@ static struct sk_buff *sja1110_rcv(struct sk_buff *skb,
 	if (likely(sja1105_skb_has_tag_8021q(skb)))
 		dsa_8021q_rcv(skb, &source_port, &switch_id, &vbid, &vid);
 
-	if (source_port != -1 && switch_id != -1)
-		skb->dev = dsa_conduit_find_user(netdev, switch_id, source_port);
-	else if (vbid >= 1)
-		skb->dev = dsa_tag_8021q_find_port_by_vbid(netdev, vbid);
-	else
-		skb->dev = dsa_find_designated_bridge_port_by_vid(netdev, vid);
+	skb->dev = dsa_tag_8021q_find_user(netdev, source_port, switch_id,
+					   vid, vbid);
+
 	if (!skb->dev) {
 		netdev_warn(netdev, "Couldn't decode source port\n");
 		return NULL;
-- 
cgit v1.2.3-58-ga151


From 6c87e1a4792804efce8ab3dfdb6e9ada314ec6dd Mon Sep 17 00:00:00 2001
From: Pawel Dembicki <paweldembicki@gmail.com>
Date: Sat, 13 Jul 2024 23:16:13 +0200
Subject: net: dsa: vsc73xx: introduce tag 8021q for vsc73xx

This commit introduces a new tagger based on 802.1q tagging.
It's designed for the vsc73xx driver. The VSC73xx family doesn't have
any tag support for the RGMII port, but it could be based on VLANs.

Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://patch.msgid.link/20240713211620.1125910-8-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/dsa.h           |  2 ++
 net/dsa/Kconfig             |  6 ++++
 net/dsa/Makefile            |  1 +
 net/dsa/tag_vsc73xx_8021q.c | 68 +++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 77 insertions(+)
 create mode 100644 net/dsa/tag_vsc73xx_8021q.c

diff --git a/include/net/dsa.h b/include/net/dsa.h
index f9ae3ca66b6f..5a5a03a7b4c3 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -53,6 +53,7 @@ struct tc_action;
 #define DSA_TAG_PROTO_RTL8_4T_VALUE		25
 #define DSA_TAG_PROTO_RZN1_A5PSW_VALUE		26
 #define DSA_TAG_PROTO_LAN937X_VALUE		27
+#define DSA_TAG_PROTO_VSC73XX_8021Q_VALUE	28
 
 enum dsa_tag_protocol {
 	DSA_TAG_PROTO_NONE		= DSA_TAG_PROTO_NONE_VALUE,
@@ -83,6 +84,7 @@ enum dsa_tag_protocol {
 	DSA_TAG_PROTO_RTL8_4T		= DSA_TAG_PROTO_RTL8_4T_VALUE,
 	DSA_TAG_PROTO_RZN1_A5PSW	= DSA_TAG_PROTO_RZN1_A5PSW_VALUE,
 	DSA_TAG_PROTO_LAN937X		= DSA_TAG_PROTO_LAN937X_VALUE,
+	DSA_TAG_PROTO_VSC73XX_8021Q	= DSA_TAG_PROTO_VSC73XX_8021Q_VALUE,
 };
 
 struct dsa_switch;
diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig
index 8d5bf869eb14..2dfe9063613f 100644
--- a/net/dsa/Kconfig
+++ b/net/dsa/Kconfig
@@ -166,6 +166,12 @@ config NET_DSA_TAG_TRAILER
 	  Say Y or M if you want to enable support for tagging frames at
 	  with a trailed. e.g. Marvell 88E6060.
 
+config NET_DSA_TAG_VSC73XX_8021Q
+	tristate "Tag driver for Microchip/Vitesse VSC73xx family of switches, using VLAN"
+	help
+	  Say Y or M if you want to enable support for tagging frames with a
+	  custom VLAN-based header.
+
 config NET_DSA_TAG_XRS700X
 	tristate "Tag driver for XRS700x switches"
 	help
diff --git a/net/dsa/Makefile b/net/dsa/Makefile
index 8a1894a42552..555c07cfeb71 100644
--- a/net/dsa/Makefile
+++ b/net/dsa/Makefile
@@ -37,6 +37,7 @@ obj-$(CONFIG_NET_DSA_TAG_RTL8_4) += tag_rtl8_4.o
 obj-$(CONFIG_NET_DSA_TAG_RZN1_A5PSW) += tag_rzn1_a5psw.o
 obj-$(CONFIG_NET_DSA_TAG_SJA1105) += tag_sja1105.o
 obj-$(CONFIG_NET_DSA_TAG_TRAILER) += tag_trailer.o
+obj-$(CONFIG_NET_DSA_TAG_VSC73XX_8021Q) += tag_vsc73xx_8021q.o
 obj-$(CONFIG_NET_DSA_TAG_XRS700X) += tag_xrs700x.o
 
 # for tracing framework to find trace.h
diff --git a/net/dsa/tag_vsc73xx_8021q.c b/net/dsa/tag_vsc73xx_8021q.c
new file mode 100644
index 000000000000..af121a9aff7f
--- /dev/null
+++ b/net/dsa/tag_vsc73xx_8021q.c
@@ -0,0 +1,68 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/* Copyright (C) 2024 Pawel Dembicki <paweldembicki@gmail.com>
+ */
+#include <linux/dsa/8021q.h>
+
+#include "tag.h"
+#include "tag_8021q.h"
+
+#define VSC73XX_8021Q_NAME "vsc73xx-8021q"
+
+static struct sk_buff *
+vsc73xx_xmit(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct dsa_port *dp = dsa_user_to_port(netdev);
+	u16 queue_mapping = skb_get_queue_mapping(skb);
+	u16 tx_vid = dsa_tag_8021q_standalone_vid(dp);
+	u8 pcp;
+
+	if (skb->offload_fwd_mark) {
+		unsigned int bridge_num = dsa_port_bridge_num_get(dp);
+		struct net_device *br = dsa_port_bridge_dev_get(dp);
+
+		if (br_vlan_enabled(br))
+			return skb;
+
+		tx_vid = dsa_tag_8021q_bridge_vid(bridge_num);
+	}
+
+	pcp = netdev_txq_to_tc(netdev, queue_mapping);
+
+	return dsa_8021q_xmit(skb, netdev, ETH_P_8021Q,
+			      ((pcp << VLAN_PRIO_SHIFT) | tx_vid));
+}
+
+static struct sk_buff *
+vsc73xx_rcv(struct sk_buff *skb, struct net_device *netdev)
+{
+	int src_port = -1, switch_id = -1, vbid = -1, vid = -1;
+
+	dsa_8021q_rcv(skb, &src_port, &switch_id, &vbid, &vid);
+
+	skb->dev = dsa_tag_8021q_find_user(netdev, src_port, switch_id,
+					   vid, vbid);
+	if (!skb->dev) {
+		dev_warn_ratelimited(&netdev->dev,
+				     "Couldn't decode source port\n");
+		return NULL;
+	}
+
+	dsa_default_offload_fwd_mark(skb);
+
+	return skb;
+}
+
+static const struct dsa_device_ops vsc73xx_8021q_netdev_ops = {
+	.name			= VSC73XX_8021Q_NAME,
+	.proto			= DSA_TAG_PROTO_VSC73XX_8021Q,
+	.xmit			= vsc73xx_xmit,
+	.rcv			= vsc73xx_rcv,
+	.needed_headroom	= VLAN_HLEN,
+	.promisc_on_conduit	= true,
+};
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("DSA tag driver for VSC73XX family of switches, using VLAN");
+MODULE_ALIAS_DSA_TAG_DRIVER(DSA_TAG_PROTO_VSC73XX_8021Q, VSC73XX_8021Q_NAME);
+
+module_dsa_tag_driver(vsc73xx_8021q_netdev_ops);
-- 
cgit v1.2.3-58-ga151


From e3386ec4ec9083522c36ab0ddfc638c90395c741 Mon Sep 17 00:00:00 2001
From: Pawel Dembicki <paweldembicki@gmail.com>
Date: Sat, 13 Jul 2024 23:16:14 +0200
Subject: net: dsa: vsc73xx: Implement the tag_8021q VLAN operations

This patch is a simple implementation of 802.1q tagging in the vsc73xx
driver. Currently, devices with DSA_TAG_PROTO_NONE are not functional.
The VSC73XX family doesn't provide any tag support for external Ethernet
ports.

The only option available is VLAN-based tagging, which requires constant
hardware VLAN filtering. While the VSC73XX family supports provider
bridging, it only supports QinQ without full implementation of 802.1AD.
This means it only allows the doubled 0x8100 TPID.

In the simple port mode, QinQ is enabled to preserve forwarding of
VLAN-tagged frames.

Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-9-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/Kconfig                |  2 +-
 drivers/net/dsa/vitesse-vsc73xx-core.c | 73 ++++++++++++++++++++++++++++++++--
 2 files changed, 71 insertions(+), 4 deletions(-)

diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig
index 8508b5145bc1..2d10b4d6cfbb 100644
--- a/drivers/net/dsa/Kconfig
+++ b/drivers/net/dsa/Kconfig
@@ -127,7 +127,7 @@ config NET_DSA_SMSC_LAN9303_MDIO
 
 config NET_DSA_VITESSE_VSC73XX
 	tristate
-	select NET_DSA_TAG_NONE
+	select NET_DSA_TAG_VSC73XX_8021Q
 	select FIXED_PHY
 	select VITESSE_PHY
 	select GPIOLIB
diff --git a/drivers/net/dsa/vitesse-vsc73xx-core.c b/drivers/net/dsa/vitesse-vsc73xx-core.c
index 906bbae22861..25c3cd661b30 100644
--- a/drivers/net/dsa/vitesse-vsc73xx-core.c
+++ b/drivers/net/dsa/vitesse-vsc73xx-core.c
@@ -597,7 +597,7 @@ static enum dsa_tag_protocol vsc73xx_get_tag_protocol(struct dsa_switch *ds,
 	 * cannot access the tag. (See "Internal frame header" section
 	 * 3.9.1 in the manual.)
 	 */
-	return DSA_TAG_PROTO_NONE;
+	return DSA_TAG_PROTO_VSC73XX_8021Q;
 }
 
 static int vsc73xx_wait_for_vlan_table_cmd(struct vsc73xx *vsc)
@@ -687,7 +687,7 @@ vsc73xx_update_vlan_table(struct vsc73xx *vsc, int port, u16 vid, bool set)
 static int vsc73xx_setup(struct dsa_switch *ds)
 {
 	struct vsc73xx *vsc = ds->priv;
-	int i;
+	int i, ret;
 
 	dev_info(vsc->dev, "set up the switch\n");
 
@@ -768,7 +768,18 @@ static int vsc73xx_setup(struct dsa_switch *ds)
 
 	INIT_LIST_HEAD(&vsc->vlans);
 
-	return 0;
+	rtnl_lock();
+	ret = dsa_tag_8021q_register(ds, htons(ETH_P_8021Q));
+	rtnl_unlock();
+
+	return ret;
+}
+
+static void vsc73xx_teardown(struct dsa_switch *ds)
+{
+	rtnl_lock();
+	dsa_tag_8021q_unregister(ds);
+	rtnl_unlock();
 }
 
 static void vsc73xx_init_port(struct vsc73xx *vsc, int port)
@@ -1549,12 +1560,65 @@ static int vsc73xx_port_vlan_del(struct dsa_switch *ds, int port,
 		vsc73xx_bridge_vlan_remove_port(vsc73xx_vlan, port);
 
 	commit_to_hardware = !vsc73xx_tag_8021q_active(dsa_to_port(ds, port));
+
 	if (commit_to_hardware)
 		return vsc73xx_vlan_commit_settings(vsc, port);
 
 	return 0;
 }
 
+static int vsc73xx_tag_8021q_vlan_add(struct dsa_switch *ds, int port, u16 vid,
+				      u16 flags)
+{
+	bool pvid = flags & BRIDGE_VLAN_INFO_PVID;
+	struct vsc73xx_portinfo *portinfo;
+	struct vsc73xx *vsc = ds->priv;
+	bool commit_to_hardware;
+	int ret;
+
+	portinfo = &vsc->portinfo[port];
+
+	if (pvid) {
+		portinfo->pvid_tag_8021q_configured = true;
+		portinfo->pvid_tag_8021q = vid;
+	}
+
+	commit_to_hardware = vsc73xx_tag_8021q_active(dsa_to_port(ds, port));
+	if (commit_to_hardware) {
+		ret = vsc73xx_vlan_commit_settings(vsc, port);
+		if (ret)
+			return ret;
+	}
+
+	return vsc73xx_update_vlan_table(vsc, port, vid, true);
+}
+
+static int vsc73xx_tag_8021q_vlan_del(struct dsa_switch *ds, int port, u16 vid)
+{
+	struct vsc73xx_portinfo *portinfo;
+	struct vsc73xx *vsc = ds->priv;
+
+	portinfo = &vsc->portinfo[port];
+
+	if (portinfo->pvid_tag_8021q_configured &&
+	    portinfo->pvid_tag_8021q == vid) {
+		struct dsa_port *dp = dsa_to_port(ds, port);
+		bool commit_to_hardware;
+		int err;
+
+		portinfo->pvid_tag_8021q_configured = false;
+
+		commit_to_hardware = vsc73xx_tag_8021q_active(dp);
+		if (commit_to_hardware) {
+			err = vsc73xx_vlan_commit_settings(vsc, port);
+			if (err)
+				return err;
+		}
+	}
+
+	return vsc73xx_update_vlan_table(vsc, port, vid, false);
+}
+
 static void vsc73xx_refresh_fwd_map(struct dsa_switch *ds, int port, u8 state)
 {
 	struct dsa_port *other_dp, *dp = dsa_to_port(ds, port);
@@ -1644,6 +1708,7 @@ static const struct phylink_mac_ops vsc73xx_phylink_mac_ops = {
 static const struct dsa_switch_ops vsc73xx_ds_ops = {
 	.get_tag_protocol = vsc73xx_get_tag_protocol,
 	.setup = vsc73xx_setup,
+	.teardown = vsc73xx_teardown,
 	.phy_read = vsc73xx_phy_read,
 	.phy_write = vsc73xx_phy_write,
 	.get_strings = vsc73xx_get_strings,
@@ -1658,6 +1723,8 @@ static const struct dsa_switch_ops vsc73xx_ds_ops = {
 	.port_vlan_add = vsc73xx_port_vlan_add,
 	.port_vlan_del = vsc73xx_port_vlan_del,
 	.phylink_get_caps = vsc73xx_phylink_get_caps,
+	.tag_8021q_vlan_add = vsc73xx_tag_8021q_vlan_add,
+	.tag_8021q_vlan_del = vsc73xx_tag_8021q_vlan_del,
 };
 
 static int vsc73xx_gpio_get(struct gpio_chip *chip, unsigned int offset)
-- 
cgit v1.2.3-58-ga151


From ce20fdd670ac375a4e3dff91c2888ad9ff9eef56 Mon Sep 17 00:00:00 2001
From: Pawel Dembicki <paweldembicki@gmail.com>
Date: Sat, 13 Jul 2024 23:16:15 +0200
Subject: net: dsa: Define max num of bridges in tag8021q implementation

Max number of bridges in tag8021q implementation is strictly limited
by VBID size: 3 bits. But zero is reserved and only 7 values can be used.

This patch adds define which describe maximum possible value.

Suggested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-10-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/sja1105/sja1105_main.c | 3 +--
 include/linux/dsa/8021q.h              | 5 +++++
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/dsa/sja1105/sja1105_main.c b/drivers/net/dsa/sja1105/sja1105_main.c
index ee0fb1c343f1..0c55a29d7dd3 100644
--- a/drivers/net/dsa/sja1105/sja1105_main.c
+++ b/drivers/net/dsa/sja1105/sja1105_main.c
@@ -3167,8 +3167,7 @@ static int sja1105_setup(struct dsa_switch *ds)
 	ds->vlan_filtering_is_global = true;
 	ds->untag_bridge_pvid = true;
 	ds->fdb_isolation = true;
-	/* tag_8021q has 3 bits for the VBID, and the value 0 is reserved */
-	ds->max_num_bridges = 7;
+	ds->max_num_bridges = DSA_TAG_8021Q_MAX_NUM_BRIDGES;
 
 	/* Advertise the 8 egress queues */
 	ds->num_tx_queues = SJA1105_NUM_TC;
diff --git a/include/linux/dsa/8021q.h b/include/linux/dsa/8021q.h
index f3664ee12170..1dda2a13b832 100644
--- a/include/linux/dsa/8021q.h
+++ b/include/linux/dsa/8021q.h
@@ -8,6 +8,11 @@
 #include <net/dsa.h>
 #include <linux/types.h>
 
+/* VBID is limited to three bits only and zero is reserved.
+ * Only 7 bridges can be enumerated.
+ */
+#define DSA_TAG_8021Q_MAX_NUM_BRIDGES	7
+
 int dsa_tag_8021q_register(struct dsa_switch *ds, __be16 proto);
 
 void dsa_tag_8021q_unregister(struct dsa_switch *ds);
-- 
cgit v1.2.3-58-ga151


From 85aabd1fe9d6af4dc5d11a2d8be567ec45d1dc5e Mon Sep 17 00:00:00 2001
From: Pawel Dembicki <paweldembicki@gmail.com>
Date: Sat, 13 Jul 2024 23:16:16 +0200
Subject: net: dsa: prepare 'dsa_tag_8021q_bridge_join' for standalone use

The 'dsa_tag_8021q_bridge_join' could be used as a generic implementation
of the 'ds->ops->port_bridge_join()' function. However, it is necessary
to synchronize their arguments.

This patch also moves the 'tx_fwd_offload' flag configuration line into
'dsa_tag_8021q_bridge_join' body. Currently, every (sja1105) driver sets
it, and the future vsc73xx implementation will also need it for
simplification.

Suggested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-11-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/sja1105/sja1105_main.c | 5 ++---
 include/linux/dsa/8021q.h              | 3 ++-
 net/dsa/tag_8021q.c                    | 5 ++++-
 3 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/net/dsa/sja1105/sja1105_main.c b/drivers/net/dsa/sja1105/sja1105_main.c
index 0c55a29d7dd3..c7282ce3d11c 100644
--- a/drivers/net/dsa/sja1105/sja1105_main.c
+++ b/drivers/net/dsa/sja1105/sja1105_main.c
@@ -2133,14 +2133,13 @@ static int sja1105_bridge_join(struct dsa_switch *ds, int port,
 	if (rc)
 		return rc;
 
-	rc = dsa_tag_8021q_bridge_join(ds, port, bridge);
+	rc = dsa_tag_8021q_bridge_join(ds, port, bridge, tx_fwd_offload,
+				       extack);
 	if (rc) {
 		sja1105_bridge_member(ds, port, bridge, false);
 		return rc;
 	}
 
-	*tx_fwd_offload = true;
-
 	return 0;
 }
 
diff --git a/include/linux/dsa/8021q.h b/include/linux/dsa/8021q.h
index 1dda2a13b832..d13aabdeb4b2 100644
--- a/include/linux/dsa/8021q.h
+++ b/include/linux/dsa/8021q.h
@@ -18,7 +18,8 @@ int dsa_tag_8021q_register(struct dsa_switch *ds, __be16 proto);
 void dsa_tag_8021q_unregister(struct dsa_switch *ds);
 
 int dsa_tag_8021q_bridge_join(struct dsa_switch *ds, int port,
-			      struct dsa_bridge bridge);
+			      struct dsa_bridge bridge, bool *tx_fwd_offload,
+			      struct netlink_ext_ack *extack);
 
 void dsa_tag_8021q_bridge_leave(struct dsa_switch *ds, int port,
 				struct dsa_bridge bridge);
diff --git a/net/dsa/tag_8021q.c b/net/dsa/tag_8021q.c
index c0eee113a2b9..3ee53e28ec2e 100644
--- a/net/dsa/tag_8021q.c
+++ b/net/dsa/tag_8021q.c
@@ -286,7 +286,8 @@ int dsa_switch_tag_8021q_vlan_del(struct dsa_switch *ds,
  * be used for VLAN-unaware bridging.
  */
 int dsa_tag_8021q_bridge_join(struct dsa_switch *ds, int port,
-			      struct dsa_bridge bridge)
+			      struct dsa_bridge bridge, bool *tx_fwd_offload,
+			      struct netlink_ext_ack *extack)
 {
 	struct dsa_port *dp = dsa_to_port(ds, port);
 	u16 standalone_vid, bridge_vid;
@@ -304,6 +305,8 @@ int dsa_tag_8021q_bridge_join(struct dsa_switch *ds, int port,
 
 	dsa_port_tag_8021q_vlan_del(dp, standalone_vid, false);
 
+	*tx_fwd_offload = true;
+
 	return 0;
 }
 EXPORT_SYMBOL_GPL(dsa_tag_8021q_bridge_join);
-- 
cgit v1.2.3-58-ga151


From 6dfaaa27633711ba52c418d578b3a13b0c8bfd8d Mon Sep 17 00:00:00 2001
From: Pawel Dembicki <paweldembicki@gmail.com>
Date: Sat, 13 Jul 2024 23:16:17 +0200
Subject: net: dsa: vsc73xx: Add bridge support

This patch adds bridge support for the vsc73xx driver.

The vsc73xx requires minimal operations and ithe generic
dsa_tag_8021q_bridge_* API is sufficient.
The forwarding matrix is managed by vsc73xx_port_stp_state_set() ->
vsc73xx_refresh_fwd_map()i routine, which is called immediately after
.port_bridge_join() and .port_bridge_leave().

Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-12-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/vitesse-vsc73xx-core.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/dsa/vitesse-vsc73xx-core.c b/drivers/net/dsa/vitesse-vsc73xx-core.c
index 25c3cd661b30..d0e501bbd57d 100644
--- a/drivers/net/dsa/vitesse-vsc73xx-core.c
+++ b/drivers/net/dsa/vitesse-vsc73xx-core.c
@@ -691,6 +691,9 @@ static int vsc73xx_setup(struct dsa_switch *ds)
 
 	dev_info(vsc->dev, "set up the switch\n");
 
+	ds->untag_bridge_pvid = true;
+	ds->max_num_bridges = DSA_TAG_8021Q_MAX_NUM_BRIDGES;
+
 	/* Issue RESET */
 	vsc73xx_write(vsc, VSC73XX_BLOCK_SYSTEM, 0, VSC73XX_GLORESET,
 		      VSC73XX_GLORESET_MASTER_RESET);
@@ -1716,6 +1719,8 @@ static const struct dsa_switch_ops vsc73xx_ds_ops = {
 	.get_sset_count = vsc73xx_get_sset_count,
 	.port_enable = vsc73xx_port_enable,
 	.port_disable = vsc73xx_port_disable,
+	.port_bridge_join = dsa_tag_8021q_bridge_join,
+	.port_bridge_leave = dsa_tag_8021q_bridge_leave,
 	.port_change_mtu = vsc73xx_change_mtu,
 	.port_max_mtu = vsc73xx_get_max_mtu,
 	.port_stp_state_set = vsc73xx_port_stp_state_set,
-- 
cgit v1.2.3-58-ga151


From 259a7061c2f14bbf5845598f69ac63cdb77d6346 Mon Sep 17 00:00:00 2001
From: Pawel Dembicki <paweldembicki@gmail.com>
Date: Sat, 13 Jul 2024 23:16:18 +0200
Subject: net: dsa: vsc73xx: start treating the BR_LEARNING flag

This patch implements .port_pre_bridge_flags() and .port_bridge_flags(),
which are required for properly treating the BR_LEARNING flag. Also,
.port_stp_state_set() is tweaked and now disables learning for standalone
ports.

Disabling learning for standalone ports is required to avoid situations
where one port sees traffic originating from another, which could cause
packet drops.

Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-13-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/vitesse-vsc73xx-core.c | 41 +++++++++++++++++++++++++++++-----
 1 file changed, 35 insertions(+), 6 deletions(-)

diff --git a/drivers/net/dsa/vitesse-vsc73xx-core.c b/drivers/net/dsa/vitesse-vsc73xx-core.c
index d0e501bbd57d..d9d3e30fd47a 100644
--- a/drivers/net/dsa/vitesse-vsc73xx-core.c
+++ b/drivers/net/dsa/vitesse-vsc73xx-core.c
@@ -1622,6 +1622,31 @@ static int vsc73xx_tag_8021q_vlan_del(struct dsa_switch *ds, int port, u16 vid)
 	return vsc73xx_update_vlan_table(vsc, port, vid, false);
 }
 
+static int vsc73xx_port_pre_bridge_flags(struct dsa_switch *ds, int port,
+					 struct switchdev_brport_flags flags,
+					 struct netlink_ext_ack *extack)
+{
+	if (flags.mask & ~BR_LEARNING)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int vsc73xx_port_bridge_flags(struct dsa_switch *ds, int port,
+				     struct switchdev_brport_flags flags,
+				     struct netlink_ext_ack *extack)
+{
+	if (flags.mask & BR_LEARNING) {
+		u32 val = flags.val & BR_LEARNING ? BIT(port) : 0;
+		struct vsc73xx *vsc = ds->priv;
+
+		return vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
+					   VSC73XX_LEARNMASK, BIT(port), val);
+	}
+
+	return 0;
+}
+
 static void vsc73xx_refresh_fwd_map(struct dsa_switch *ds, int port, u8 state)
 {
 	struct dsa_port *other_dp, *dp = dsa_to_port(ds, port);
@@ -1682,19 +1707,21 @@ static void vsc73xx_refresh_fwd_map(struct dsa_switch *ds, int port, u8 state)
 static void vsc73xx_port_stp_state_set(struct dsa_switch *ds, int port,
 				       u8 state)
 {
+	struct dsa_port *dp = dsa_to_port(ds, port);
 	struct vsc73xx *vsc = ds->priv;
-	u32 val;
+	u32 val = 0;
+
+	if (state == BR_STATE_LEARNING || state == BR_STATE_FORWARDING)
+		val = dp->learning ? BIT(port) : 0;
+
+	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
+			    VSC73XX_LEARNMASK, BIT(port), val);
 
 	val = (state == BR_STATE_BLOCKING || state == BR_STATE_DISABLED) ?
 	      0 : BIT(port);
 	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
 			    VSC73XX_RECVMASK, BIT(port), val);
 
-	val = (state == BR_STATE_LEARNING || state == BR_STATE_FORWARDING) ?
-	      BIT(port) : 0;
-	vsc73xx_update_bits(vsc, VSC73XX_BLOCK_ANALYZER, 0,
-			    VSC73XX_LEARNMASK, BIT(port), val);
-
 	/* CPU Port should always forward packets when user ports are forwarding
 	 * so let's configure it from other ports only.
 	 */
@@ -1719,6 +1746,8 @@ static const struct dsa_switch_ops vsc73xx_ds_ops = {
 	.get_sset_count = vsc73xx_get_sset_count,
 	.port_enable = vsc73xx_port_enable,
 	.port_disable = vsc73xx_port_disable,
+	.port_pre_bridge_flags = vsc73xx_port_pre_bridge_flags,
+	.port_bridge_flags = vsc73xx_port_bridge_flags,
 	.port_bridge_join = dsa_tag_8021q_bridge_join,
 	.port_bridge_leave = dsa_tag_8021q_bridge_leave,
 	.port_change_mtu = vsc73xx_change_mtu,
-- 
cgit v1.2.3-58-ga151


From 9283477e28913c1e7625c0a8d6959745e2431533 Mon Sep 17 00:00:00 2001
From: Aleksander Jan Bajkowski <olek2@wp.pl>
Date: Sat, 13 Jul 2024 19:09:20 +0200
Subject: net: ethernet: lantiq_etop: remove redundant device name setup

The same name is set when allocating the netdevice structure in the
alloc_etherdev_mq()->alloc_etherrdev_mqs() function. Therefore, there
is no need to manually set it.

This fixes CheckPatch warnings:
WARNING: Prefer strscpy over strcpy - see: https://github.com/KSPP/linux/issues/88
	strcpy(dev->name, "eth%d");

Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
Link: https://patch.msgid.link/20240713170920.863171-1-olek2@wp.pl
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/lantiq_etop.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c
index 0b9982804370..9e6984815386 100644
--- a/drivers/net/ethernet/lantiq_etop.c
+++ b/drivers/net/ethernet/lantiq_etop.c
@@ -675,7 +675,6 @@ ltq_etop_probe(struct platform_device *pdev)
 		err = -ENOMEM;
 		goto err_out;
 	}
-	strcpy(dev->name, "eth%d");
 	dev->netdev_ops = &ltq_eth_netdev_ops;
 	dev->ethtool_ops = &ltq_etop_ethtool_ops;
 	priv = netdev_priv(dev);
-- 
cgit v1.2.3-58-ga151


From 66ef82c66e9c462c893909b35ee4cbdfccf02bb2 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Wed, 12 Jun 2024 10:01:49 +0200
Subject: dt-bindings: net: bluetooth: qualcomm: describe regulators for
 QCA6390

QCA6390 has a compatible listed in the bindings but is missing the
regulators description. Add the missing supply property and list the
required ones in the allOf section.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 .../bindings/net/bluetooth/qualcomm-bluetooth.yaml      | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml b/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml
index 055a3351880b..48ac9f10ef05 100644
--- a/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml
+++ b/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml
@@ -62,6 +62,9 @@ properties:
   vdddig-supply:
     description: VDD_DIG supply regulator handle
 
+  vddbtcmx-supply:
+    description: VDD_BT_CMX supply regulator handle
+
   vddbtcxmx-supply:
     description: VDD_BT_CXMX supply regulator handle
 
@@ -184,6 +187,20 @@ allOf:
         - vddrfa0p8-supply
         - vddrfa1p2-supply
         - vddrfa1p9-supply
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - qcom,qca6390-bt
+    then:
+      required:
+        - vddrfacmn-supply
+        - vddaon-supply
+        - vddbtcmx-supply
+        - vddrfa0p8-supply
+        - vddrfa1p2-supply
+        - vddrfa1p7-supply
 
 examples:
   - |
-- 
cgit v1.2.3-58-ga151


From 9a15ce685706e73154f03341b35ff5943e46d039 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Wed, 12 Jun 2024 10:01:50 +0200
Subject: Bluetooth: qca: use the power sequencer for QCA6390

Use the pwrseq subsystem's consumer API to run the power-up sequence for
the Bluetooth module of the QCA6390 package.

Tested-by: Amit Pundir <amit.pundir@linaro.org>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-QRD, SM8650-QRD & SM8650-HDK
Tested-by: Caleb Connolly <caleb.connolly@linaro.org> # OnePlus 8T
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_qca.c | 74 ++++++++++++++++++++++++++++++++++++---------
 1 file changed, 59 insertions(+), 15 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 2f47267508d5..2b31f3dc33a9 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -28,6 +28,7 @@
 #include <linux/of.h>
 #include <linux/acpi.h>
 #include <linux/platform_device.h>
+#include <linux/pwrseq/consumer.h>
 #include <linux/regulator/consumer.h>
 #include <linux/serdev.h>
 #include <linux/mutex.h>
@@ -214,6 +215,7 @@ struct qca_power {
 	struct regulator_bulk_data *vreg_bulk;
 	int num_vregs;
 	bool vregs_on;
+	struct pwrseq_desc *pwrseq;
 };
 
 struct qca_serdev {
@@ -1684,6 +1686,27 @@ static bool qca_wakeup(struct hci_dev *hdev)
 	return wakeup;
 }
 
+static int qca_port_reopen(struct hci_uart *hu)
+{
+	int ret;
+
+	/* Now the device is in ready state to communicate with host.
+	 * To sync host with device we need to reopen port.
+	 * Without this, we will have RTS and CTS synchronization
+	 * issues.
+	 */
+	serdev_device_close(hu->serdev);
+	ret = serdev_device_open(hu->serdev);
+	if (ret) {
+		bt_dev_err(hu->hdev, "failed to open port");
+		return ret;
+	}
+
+	hci_uart_set_flow_control(hu, false);
+
+	return 0;
+}
+
 static int qca_regulator_init(struct hci_uart *hu)
 {
 	enum qca_btsoc_type soc_type = qca_soc_type(hu);
@@ -1752,21 +1775,7 @@ static int qca_regulator_init(struct hci_uart *hu)
 		break;
 	}
 
-	/* Now the device is in ready state to communicate with host.
-	 * To sync host with device we need to reopen port.
-	 * Without this, we will have RTS and CTS synchronization
-	 * issues.
-	 */
-	serdev_device_close(hu->serdev);
-	ret = serdev_device_open(hu->serdev);
-	if (ret) {
-		bt_dev_err(hu->hdev, "failed to open port");
-		return ret;
-	}
-
-	hci_uart_set_flow_control(hu, false);
-
-	return 0;
+	return qca_port_reopen(hu);
 }
 
 static int qca_power_on(struct hci_dev *hdev)
@@ -1794,6 +1803,17 @@ static int qca_power_on(struct hci_dev *hdev)
 		ret = qca_regulator_init(hu);
 		break;
 
+	case QCA_QCA6390:
+		qcadev = serdev_device_get_drvdata(hu->serdev);
+		ret = pwrseq_power_on(qcadev->bt_power->pwrseq);
+		if (ret)
+			return ret;
+
+		ret = qca_port_reopen(hu);
+		if (ret)
+			return ret;
+		break;
+
 	default:
 		qcadev = serdev_device_get_drvdata(hu->serdev);
 		if (qcadev->bt_en) {
@@ -2168,6 +2188,10 @@ static void qca_power_shutdown(struct hci_uart *hu)
 		}
 		break;
 
+	case QCA_QCA6390:
+		pwrseq_power_off(qcadev->bt_power->pwrseq);
+		break;
+
 	default:
 		gpiod_set_value_cansleep(qcadev->bt_en, 0);
 	}
@@ -2309,12 +2333,25 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 	case QCA_WCN6750:
 	case QCA_WCN6855:
 	case QCA_WCN7850:
+	case QCA_QCA6390:
 		qcadev->bt_power = devm_kzalloc(&serdev->dev,
 						sizeof(struct qca_power),
 						GFP_KERNEL);
 		if (!qcadev->bt_power)
 			return -ENOMEM;
+		break;
+	default:
+		break;
+	}
 
+	switch (qcadev->btsoc_type) {
+	case QCA_WCN3988:
+	case QCA_WCN3990:
+	case QCA_WCN3991:
+	case QCA_WCN3998:
+	case QCA_WCN6750:
+	case QCA_WCN6855:
+	case QCA_WCN7850:
 		qcadev->bt_power->dev = &serdev->dev;
 		err = qca_init_regulators(qcadev->bt_power, data->vregs,
 					  data->num_vregs);
@@ -2360,6 +2397,13 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 		}
 		break;
 
+	case QCA_QCA6390:
+		qcadev->bt_power->pwrseq = devm_pwrseq_get(&serdev->dev,
+							   "bluetooth");
+		if (IS_ERR(qcadev->bt_power->pwrseq))
+			return PTR_ERR(qcadev->bt_power->pwrseq);
+		fallthrough;
+
 	default:
 		qcadev->bt_en = devm_gpiod_get_optional(&serdev->dev, "enable",
 					       GPIOD_OUT_LOW);
-- 
cgit v1.2.3-58-ga151


From 519d849acb6523a9e9500f7a4345018fe92a131a Mon Sep 17 00:00:00 2001
From: Paul Menzel <pmenzel@molgen.mpg.de>
Date: Fri, 21 Jun 2024 09:18:57 +0200
Subject: Bluetooth: btintel: Fix spelling of *intermediate* in comment

Fixes: f3b845e0aea3 ("Bluetooth: btintel: Add support to download intermediate loader")
Cc: Kiran K <kiran.k@intel.com>
Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btintel.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c
index 5d735391545a..2dfe049d863e 100644
--- a/drivers/bluetooth/btintel.c
+++ b/drivers/bluetooth/btintel.c
@@ -2650,7 +2650,7 @@ int btintel_bootloader_setup_tlv(struct hci_dev *hdev,
 		return err;
 
 	/* If image type returned is BTINTEL_IMG_IML, then controller supports
-	 * intermediae loader image
+	 * intermediate loader image
 	 */
 	if (ver->img_type == BTINTEL_IMG_IML) {
 		err = btintel_prepare_fw_download_tlv(hdev, ver, &boot_param);
-- 
cgit v1.2.3-58-ga151


From b96bac9607b285d4316e74fbf469df084c4f8adb Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Sun, 23 Jun 2024 19:13:52 +0200
Subject: Bluetooth: hci: fix build when POWER_SEQUENCING=m

BT_HCIUART_QCA, which is a bool and an HCI UART component, uses now
power sequencing, so we need to be sure BT_HCIUART_QCA=y and
POWER_SEQUENCING=m cannot happen:

  /usr/bin/aarch64-linux-gnu-ld: drivers/bluetooth/hci_qca.o: in function `qca_serdev_probe':
  drivers/bluetooth/hci_qca.c:2401:(.text+0xc48): undefined reference to `devm_pwrseq_get'
  /usr/bin/aarch64-linux-gnu-ld: drivers/bluetooth/hci_qca.o: in function `qca_power_shutdown':
  drivers/bluetooth/hci_qca.c:2192:(.text+0x29e8): undefined reference to `pwrseq_power_off'
  /usr/bin/aarch64-linux-gnu-ld: drivers/bluetooth/hci_qca.o: in function `qca_power_on':
  drivers/bluetooth/hci_qca.c:1808:(.text+0x3390): undefined reference to `pwrseq_power_on'

Fixes: 9a15ce685706 ("Bluetooth: qca: use the power sequencer for QCA6390")
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/bluetooth/Kconfig b/drivers/bluetooth/Kconfig
index 87484f5de8e3..90a94a111e67 100644
--- a/drivers/bluetooth/Kconfig
+++ b/drivers/bluetooth/Kconfig
@@ -105,6 +105,7 @@ config BT_HCIUART
 	tristate "HCI UART driver"
 	depends on SERIAL_DEV_BUS || !SERIAL_DEV_BUS
 	depends on NVMEM || !NVMEM
+	depends on POWER_SEQUENCING || !POWER_SEQUENCING
 	depends on TTY
 	help
 	  Bluetooth HCI UART driver.
-- 
cgit v1.2.3-58-ga151


From fb5e4713fc185f6283ba0704edccea312253e805 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Mon, 24 Jun 2024 21:45:18 +0200
Subject: Bluetooth: qca: don't disable power management for QCA6390

We unnecessarily fallthrough the case for QCA6390 when initializing the
device and hit the condition where - due to the lack of the enable-gpio
- we disable power management despite using the power sequencer. We don't
need to look for clocks on this model so it makes more sense to just
register the hci device and break the switch.

Reported-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Fixes: 9a15ce685706 ("Bluetooth: qca: use the power sequencer for QCA6390")
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> # RB5
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_qca.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 2b31f3dc33a9..bc6a49ba26f9 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -2402,7 +2402,13 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 							   "bluetooth");
 		if (IS_ERR(qcadev->bt_power->pwrseq))
 			return PTR_ERR(qcadev->bt_power->pwrseq);
-		fallthrough;
+
+		err = hci_uart_register_device(&qcadev->serdev_hu, &qca_proto);
+		if (err) {
+			BT_ERR("qca6390 serdev registration failed");
+			return err;
+		}
+		break;
 
 	default:
 		qcadev->bt_en = devm_gpiod_get_optional(&serdev->dev, "enable",
-- 
cgit v1.2.3-58-ga151


From 1723b550b66aebea9b8e9d6fb5f3c468adea69ce Mon Sep 17 00:00:00 2001
From: Kiran K <kiran.k@intel.com>
Date: Thu, 27 Jun 2024 18:36:39 +0530
Subject: Bluetooth: btintel: Fix the sfi name for BlazarU

Use INTEL_HW_VARIANT() instead of CNVi Id to decide to load Intermediate
Loader (IML) image. Fix the driver loading incorrect firmware for
BlazarU product.

dmesg:
.....
[146.111834] Bluetooth: hci0: Minimum firmware build 1 week 10 2014
[146.111839] Bluetooth: hci0: Bootloader timestamp 2022.18 buildtype 1 build 16362
[146.111848] Bluetooth: hci0: No support for _PRR ACPI method
[146.112204] Bluetooth: hci0: Failed to load Intel firmware file intel/ibt-0291-0291-iml.sfi (-2)

Fixes: 164c62f958f8 ("Bluetooth: btintel: Add firmware ID to firmware name")
Reported-by: Tsigan, Vladislav <vladislav.tsigan@intel.com>
Signed-off-by: Kiran K <kiran.k@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btintel.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c
index 2dfe049d863e..54bfe2ffe1d2 100644
--- a/drivers/bluetooth/btintel.c
+++ b/drivers/bluetooth/btintel.c
@@ -2148,7 +2148,7 @@ static void btintel_get_fw_name_tlv(const struct intel_version_tlv *ver,
 	/* Only Blazar  product supports downloading of intermediate loader
 	 * image
 	 */
-	if ((ver->cnvi_top & 0xfff) >= BTINTEL_CNVI_BLAZARI) {
+	if (INTEL_HW_VARIANT(ver->cnvi_bt) >= 0x1e) {
 		u8 zero[BTINTEL_FWID_MAXLEN];
 
 		if (ver->img_type == BTINTEL_IMG_BOOTLOADER) {
@@ -2240,7 +2240,7 @@ static int btintel_prepare_fw_download_tlv(struct hci_dev *hdev,
 		 * firmware image which doesn't exist. Lets compare the version
 		 * of IML image
 		 */
-		if ((ver->cnvi_top & 0xfff) >= BTINTEL_CNVI_BLAZARI)
+		if (INTEL_HW_VARIANT(ver->cnvi_bt) >= 0x1e)
 			btintel_get_iml_tlv(ver, fwname, sizeof(fwname), "sfi");
 		else
 			btintel_get_fw_name_tlv(ver, fwname, sizeof(fwname), "sfi");
-- 
cgit v1.2.3-58-ga151


From 3ba74b2f288bbc17c0c2a58ab219e1df19f80153 Mon Sep 17 00:00:00 2001
From: Dmitry Antipov <dmantipov@yandex.ru>
Date: Fri, 21 Jun 2024 16:01:55 +0300
Subject: Bluetooth: hci_core: cleanup struct hci_dev

Remove unused and set but otherwise unused 'discovery_old_state'
and 'sco_last_tx' members of 'struct hci_dev'. The first one is
a leftover after commit 182ee45da083 ("Bluetooth: hci_sync: Rework
hci_suspend_notifier"); the second one is originated from ancient
2.4.19 and I was unable to find any actual use since that.

Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_core.h | 2 --
 net/bluetooth/hci_sync.c         | 1 -
 2 files changed, 3 deletions(-)

diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h
index eaeaf3dc07aa..31020891fc68 100644
--- a/include/net/bluetooth/hci_core.h
+++ b/include/net/bluetooth/hci_core.h
@@ -476,7 +476,6 @@ struct hci_dev {
 	unsigned int	iso_pkts;
 
 	unsigned long	acl_last_tx;
-	unsigned long	sco_last_tx;
 	unsigned long	le_last_tx;
 
 	__u8		le_tx_def_phys;
@@ -528,7 +527,6 @@ struct hci_dev {
 
 	struct discovery_state	discovery;
 
-	int			discovery_old_state;
 	bool			discovery_paused;
 	int			advertising_old_state;
 	bool			advertising_paused;
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index ccad43f10415..76b283b8e90d 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -5853,7 +5853,6 @@ static int hci_pause_discovery_sync(struct hci_dev *hdev)
 		return err;
 
 	hdev->discovery_paused = true;
-	hdev->discovery_old_state = old_state;
 	hci_discovery_set_state(hdev, DISCOVERY_STOPPED);
 
 	return 0;
-- 
cgit v1.2.3-58-ga151


From 473a89b4ed7fd52a419340f7c540d5c8fc96fc75 Mon Sep 17 00:00:00 2001
From: WangYuli <wangyuli@uniontech.com>
Date: Sat, 22 Jun 2024 12:09:59 +0800
Subject: Bluetooth: btusb: Add Realtek RTL8852BE support ID 0x13d3:0x3591

Add the support ID(0x13d3, 0x3591) to usb_device_id table for
Realtek RTL8852BE.

The device table is as follows:

T:  Bus=01 Lev=02 Prnt=03 Port=00 Cnt=01 Dev#=  5 Spd=12   MxCh= 0
D:  Ver= 1.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=13d3 ProdID=3591 Rev= 0.00
S:  Manufacturer=Realtek
S:  Product=Bluetooth Radio
S:  SerialNumber=00e04c000001
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=500mA
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  16 Ivl=1ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms

Cc: stable@vger.kernel.org
Signed-off-by: Wentao Guan <guanwentao@uniontech.com>
Signed-off-by: WangYuli <wangyuli@uniontech.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 2d7d47f9d007..2d5c971a59ad 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -555,6 +555,8 @@ static const struct usb_device_id quirks_table[] = {
 						     BTUSB_WIDEBAND_SPEECH },
 	{ USB_DEVICE(0x13d3, 0x3572), .driver_info = BTUSB_REALTEK |
 						     BTUSB_WIDEBAND_SPEECH },
+	{ USB_DEVICE(0x13d3, 0x3591), .driver_info = BTUSB_REALTEK |
+						     BTUSB_WIDEBAND_SPEECH },
 	{ USB_DEVICE(0x0489, 0xe125), .driver_info = BTUSB_REALTEK |
 						     BTUSB_WIDEBAND_SPEECH },
 
-- 
cgit v1.2.3-58-ga151


From 87be7b189b2c50d4b51512f59e4e97db4eedee8a Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Mon, 1 Jul 2024 12:07:46 -0400
Subject: Bluetooth: Fix usage of __hci_cmd_sync_status

__hci_cmd_sync_status shall only be used if hci_req_sync_lock is _not_
required which is not the case of hci_dev_cmd so it needs to use
hci_cmd_sync_status which uses hci_req_sync_lock internally.

Fixes: f1a8f402f13f ("Bluetooth: L2CAP: Fix deadlock")
Reported-by: Pauli Virtanen <pav@iki.fi>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_core.c | 27 ++++++++++++---------------
 1 file changed, 12 insertions(+), 15 deletions(-)

diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 144e85ebfbdb..db9b9bb7d74d 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -718,8 +718,8 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg)
 
 	switch (cmd) {
 	case HCISETAUTH:
-		err = __hci_cmd_sync_status(hdev, HCI_OP_WRITE_AUTH_ENABLE,
-					    1, &dr.dev_opt, HCI_CMD_TIMEOUT);
+		err = hci_cmd_sync_status(hdev, HCI_OP_WRITE_AUTH_ENABLE,
+					  1, &dr.dev_opt, HCI_CMD_TIMEOUT);
 		break;
 
 	case HCISETENCRYPT:
@@ -730,23 +730,21 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg)
 
 		if (!test_bit(HCI_AUTH, &hdev->flags)) {
 			/* Auth must be enabled first */
-			err = __hci_cmd_sync_status(hdev,
-						    HCI_OP_WRITE_AUTH_ENABLE,
-						    1, &dr.dev_opt,
-						    HCI_CMD_TIMEOUT);
+			err = hci_cmd_sync_status(hdev,
+						  HCI_OP_WRITE_AUTH_ENABLE,
+						  1, &dr.dev_opt,
+						  HCI_CMD_TIMEOUT);
 			if (err)
 				break;
 		}
 
-		err = __hci_cmd_sync_status(hdev, HCI_OP_WRITE_ENCRYPT_MODE,
-					    1, &dr.dev_opt,
-					    HCI_CMD_TIMEOUT);
+		err = hci_cmd_sync_status(hdev, HCI_OP_WRITE_ENCRYPT_MODE,
+					  1, &dr.dev_opt, HCI_CMD_TIMEOUT);
 		break;
 
 	case HCISETSCAN:
-		err = __hci_cmd_sync_status(hdev, HCI_OP_WRITE_SCAN_ENABLE,
-					    1, &dr.dev_opt,
-					    HCI_CMD_TIMEOUT);
+		err = hci_cmd_sync_status(hdev, HCI_OP_WRITE_SCAN_ENABLE,
+					  1, &dr.dev_opt, HCI_CMD_TIMEOUT);
 
 		/* Ensure that the connectable and discoverable states
 		 * get correctly modified as this was a non-mgmt change.
@@ -758,9 +756,8 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg)
 	case HCISETLINKPOL:
 		policy = cpu_to_le16(dr.dev_opt);
 
-		err = __hci_cmd_sync_status(hdev, HCI_OP_WRITE_DEF_LINK_POLICY,
-					    2, &policy,
-					    HCI_CMD_TIMEOUT);
+		err = hci_cmd_sync_status(hdev, HCI_OP_WRITE_DEF_LINK_POLICY,
+					  2, &policy, HCI_CMD_TIMEOUT);
 		break;
 
 	case HCISETLINKMODE:
-- 
cgit v1.2.3-58-ga151


From 88b73fdc8144de7daf1156509df4d82830f48202 Mon Sep 17 00:00:00 2001
From: Rafał Miłecki <rafal@milecki.pl>
Date: Fri, 28 Jun 2024 07:46:35 +0200
Subject: dt-bindings: net: bluetooth: convert MT7622 Bluetooth to the
 json-schema
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This helps validating DTS files. Introduced changes:
1. Dropped serial details from example
2. Added required example include

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 .../net/bluetooth/mediatek,mt7622-bluetooth.yaml   | 51 ++++++++++++++++++++++
 .../devicetree/bindings/net/mediatek-bluetooth.txt | 36 ---------------
 2 files changed, 51 insertions(+), 36 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/net/bluetooth/mediatek,mt7622-bluetooth.yaml

diff --git a/Documentation/devicetree/bindings/net/bluetooth/mediatek,mt7622-bluetooth.yaml b/Documentation/devicetree/bindings/net/bluetooth/mediatek,mt7622-bluetooth.yaml
new file mode 100644
index 000000000000..3f9e69208127
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/bluetooth/mediatek,mt7622-bluetooth.yaml
@@ -0,0 +1,51 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/bluetooth/mediatek,mt7622-bluetooth.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: MediaTek SoC built-in Bluetooth
+
+description:
+  This device is a serial attached device to BTIF device and thus it must be a
+  child node of the serial node with BTIF. The dt-bindings details for BTIF
+  device can be known via Documentation/devicetree/bindings/serial/8250.yaml.
+
+maintainers:
+  - Sean Wang <sean.wang@mediatek.com>
+
+allOf:
+  - $ref: bluetooth-controller.yaml#
+
+properties:
+  compatible:
+    const: mediatek,mt7622-bluetooth
+
+  clocks:
+    maxItems: 1
+
+  clock-names:
+    const: ref
+
+  power-domains:
+    maxItems: 1
+
+required:
+  - clocks
+  - clock-names
+  - power-domains
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/power/mt7622-power.h>
+
+    serial {
+        bluetooth {
+            compatible = "mediatek,mt7622-bluetooth";
+            power-domains = <&scpsys MT7622_POWER_DOMAIN_WB>;
+            clocks = <&clk25m>;
+            clock-names = "ref";
+        };
+    };
diff --git a/Documentation/devicetree/bindings/net/mediatek-bluetooth.txt b/Documentation/devicetree/bindings/net/mediatek-bluetooth.txt
index 9ef5bacda8c1..988c72685cbf 100644
--- a/Documentation/devicetree/bindings/net/mediatek-bluetooth.txt
+++ b/Documentation/devicetree/bindings/net/mediatek-bluetooth.txt
@@ -1,39 +1,3 @@
-MediaTek SoC built-in Bluetooth Devices
-==================================
-
-This device is a serial attached device to BTIF device and thus it must be a
-child node of the serial node with BTIF. The dt-bindings details for BTIF
-device can be known via Documentation/devicetree/bindings/serial/8250.yaml.
-
-Required properties:
-
-- compatible:	Must be
-		  "mediatek,mt7622-bluetooth": for MT7622 SoC
-- clocks:	Should be the clock specifiers corresponding to the entry in
-		clock-names property.
-- clock-names:	Should contain "ref" entries.
-- power-domains: Phandle to the power domain that the device is part of
-
-Example:
-
-	btif: serial@1100c000 {
-		compatible = "mediatek,mt7622-btif",
-			     "mediatek,mtk-btif";
-		reg = <0 0x1100c000 0 0x1000>;
-		interrupts = <GIC_SPI 90 IRQ_TYPE_LEVEL_LOW>;
-		clocks = <&pericfg CLK_PERI_BTIF_PD>;
-		clock-names = "main";
-		reg-shift = <2>;
-		reg-io-width = <4>;
-
-		bluetooth {
-			compatible = "mediatek,mt7622-bluetooth";
-			power-domains = <&scpsys MT7622_POWER_DOMAIN_WB>;
-			clocks = <&clk25m>;
-			clock-names = "ref";
-		};
-	};
-
 MediaTek UART based Bluetooth Devices
 ==================================
 
-- 
cgit v1.2.3-58-ga151


From 92048ab2e2e6cc90ad1cc9f55deb5cec4d731793 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Mon, 1 Jul 2024 16:00:08 -0400
Subject: Bluetooth: hci_core: Remove usage of hci_req_sync

hci_request functions are considered deprecated so this replaces the
usage of hci_req_sync with hci_inquiry_sync.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_sync.h |  2 ++
 net/bluetooth/hci_core.c         | 29 ++++-------------------------
 net/bluetooth/hci_sync.c         | 10 +++++-----
 3 files changed, 11 insertions(+), 30 deletions(-)

diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h
index 20168732f20e..620e6014beb2 100644
--- a/include/net/bluetooth/hci_sync.h
+++ b/include/net/bluetooth/hci_sync.h
@@ -131,6 +131,8 @@ int hci_update_discoverable(struct hci_dev *hdev);
 
 int hci_update_connectable_sync(struct hci_dev *hdev);
 
+int hci_inquiry_sync(struct hci_dev *hdev, u8 length, u8 num_rsp);
+
 int hci_start_discovery_sync(struct hci_dev *hdev);
 int hci_stop_discovery_sync(struct hci_dev *hdev);
 
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index db9b9bb7d74d..cce755a84ea7 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -312,33 +312,12 @@ static int inquiry_cache_dump(struct hci_dev *hdev, int num, __u8 *buf)
 	return copied;
 }
 
-static int hci_inq_req(struct hci_request *req, unsigned long opt)
-{
-	struct hci_inquiry_req *ir = (struct hci_inquiry_req *) opt;
-	struct hci_dev *hdev = req->hdev;
-	struct hci_cp_inquiry cp;
-
-	BT_DBG("%s", hdev->name);
-
-	if (test_bit(HCI_INQUIRY, &hdev->flags))
-		return 0;
-
-	/* Start Inquiry */
-	memcpy(&cp.lap, &ir->lap, 3);
-	cp.length  = ir->length;
-	cp.num_rsp = ir->num_rsp;
-	hci_req_add(req, HCI_OP_INQUIRY, sizeof(cp), &cp);
-
-	return 0;
-}
-
 int hci_inquiry(void __user *arg)
 {
 	__u8 __user *ptr = arg;
 	struct hci_inquiry_req ir;
 	struct hci_dev *hdev;
 	int err = 0, do_inquiry = 0, max_rsp;
-	long timeo;
 	__u8 *buf;
 
 	if (copy_from_user(&ir, ptr, sizeof(ir)))
@@ -377,11 +356,11 @@ int hci_inquiry(void __user *arg)
 	}
 	hci_dev_unlock(hdev);
 
-	timeo = ir.length * msecs_to_jiffies(2000);
-
 	if (do_inquiry) {
-		err = hci_req_sync(hdev, hci_inq_req, (unsigned long) &ir,
-				   timeo, NULL);
+		hci_req_sync_lock(hdev);
+		err = hci_inquiry_sync(hdev, ir.length, ir.num_rsp);
+		hci_req_sync_unlock(hdev);
+
 		if (err < 0)
 			goto done;
 
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 76b283b8e90d..46b394842a27 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -347,10 +347,9 @@ static int scan_disable_sync(struct hci_dev *hdev, void *data)
 	return hci_scan_disable_sync(hdev);
 }
 
-static int hci_inquiry_sync(struct hci_dev *hdev, u8 length);
 static int interleaved_inquiry_sync(struct hci_dev *hdev, void *data)
 {
-	return hci_inquiry_sync(hdev, DISCOV_INTERLEAVED_INQUIRY_LEN);
+	return hci_inquiry_sync(hdev, DISCOV_INTERLEAVED_INQUIRY_LEN, 0);
 }
 
 static void le_scan_disable(struct work_struct *work)
@@ -5662,7 +5661,7 @@ int hci_update_connectable_sync(struct hci_dev *hdev)
 	return hci_update_passive_scan_sync(hdev);
 }
 
-static int hci_inquiry_sync(struct hci_dev *hdev, u8 length)
+int hci_inquiry_sync(struct hci_dev *hdev, u8 length, u8 num_rsp)
 {
 	const u8 giac[3] = { 0x33, 0x8b, 0x9e };
 	const u8 liac[3] = { 0x00, 0x8b, 0x9e };
@@ -5685,6 +5684,7 @@ static int hci_inquiry_sync(struct hci_dev *hdev, u8 length)
 		memcpy(&cp.lap, giac, sizeof(cp.lap));
 
 	cp.length = length;
+	cp.num_rsp = num_rsp;
 
 	return __hci_cmd_sync_status(hdev, HCI_OP_INQUIRY,
 				     sizeof(cp), &cp, HCI_CMD_TIMEOUT);
@@ -5771,7 +5771,7 @@ static int hci_start_interleaved_discovery_sync(struct hci_dev *hdev)
 	if (err)
 		return err;
 
-	return hci_inquiry_sync(hdev, DISCOV_BREDR_INQUIRY_LEN);
+	return hci_inquiry_sync(hdev, DISCOV_BREDR_INQUIRY_LEN, 0);
 }
 
 int hci_start_discovery_sync(struct hci_dev *hdev)
@@ -5783,7 +5783,7 @@ int hci_start_discovery_sync(struct hci_dev *hdev)
 
 	switch (hdev->discovery.type) {
 	case DISCOV_TYPE_BREDR:
-		return hci_inquiry_sync(hdev, DISCOV_BREDR_INQUIRY_LEN);
+		return hci_inquiry_sync(hdev, DISCOV_BREDR_INQUIRY_LEN, 0);
 	case DISCOV_TYPE_INTERLEAVED:
 		/* When running simultaneous discovery, the LE scanning time
 		 * should occupy the whole discovery time sine BR/EDR inquiry
-- 
cgit v1.2.3-58-ga151


From 176cbeceb5c5a740216a6be3e751e76aaddf94b9 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Mon, 1 Jul 2024 16:13:56 -0400
Subject: Bluetooth: hci_core: Don't use hci_prepare_cmd

This replaces the instance of hci_prepare_cmd with hci_cmd_sync_alloc
since the former is part of hci_request.c which is considered
deprecated.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_sync.h | 4 ++++
 net/bluetooth/hci_core.c         | 4 ++--
 net/bluetooth/hci_sync.c         | 5 ++---
 3 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h
index 620e6014beb2..a8d88247ac89 100644
--- a/include/net/bluetooth/hci_sync.h
+++ b/include/net/bluetooth/hci_sync.h
@@ -20,6 +20,10 @@ struct hci_cmd_sync_work_entry {
 };
 
 struct adv_info;
+
+struct sk_buff *hci_cmd_sync_alloc(struct hci_dev *hdev, u16 opcode, u32 plen,
+				   const void *param, struct sock *sk);
+
 /* Function with sync suffix shall not be called with hdev->lock held as they
  * wait the command to complete and in the meantime an event could be received
  * which could attempt to acquire hdev->lock causing a deadlock.
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index cce755a84ea7..9bdd3e1f8cfc 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -3054,7 +3054,7 @@ int hci_send_cmd(struct hci_dev *hdev, __u16 opcode, __u32 plen,
 
 	BT_DBG("%s opcode 0x%4.4x plen %d", hdev->name, opcode, plen);
 
-	skb = hci_prepare_cmd(hdev, opcode, plen, param);
+	skb = hci_cmd_sync_alloc(hdev, opcode, plen, param, NULL);
 	if (!skb) {
 		bt_dev_err(hdev, "no memory for command");
 		return -ENOMEM;
@@ -3089,7 +3089,7 @@ int __hci_cmd_send(struct hci_dev *hdev, u16 opcode, u32 plen,
 		return -EINVAL;
 	}
 
-	skb = hci_prepare_cmd(hdev, opcode, plen, param);
+	skb = hci_cmd_sync_alloc(hdev, opcode, plen, param, NULL);
 	if (!skb) {
 		bt_dev_err(hdev, "no memory for command (opcode 0x%4.4x)",
 			   opcode);
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 46b394842a27..0c8888448d3c 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -49,9 +49,8 @@ static void hci_cmd_sync_complete(struct hci_dev *hdev, u8 result, u16 opcode,
 	wake_up_interruptible(&hdev->req_wait_q);
 }
 
-static struct sk_buff *hci_cmd_sync_alloc(struct hci_dev *hdev, u16 opcode,
-					  u32 plen, const void *param,
-					  struct sock *sk)
+struct sk_buff *hci_cmd_sync_alloc(struct hci_dev *hdev, u16 opcode, u32 plen,
+				   const void *param, struct sock *sk)
 {
 	int len = HCI_COMMAND_HDR_SIZE + plen;
 	struct hci_command_hdr *hdr;
-- 
cgit v1.2.3-58-ga151


From 8bedf130c265384fb136b19b20504239240c51bf Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Mon, 1 Jul 2024 16:50:30 -0400
Subject: Bluetooth: hci_sync: Move handling of interleave_scan

This moves handling of interleave_scan work to hci_sync.c since
hci_request.c is deprecated.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/hci_core.c |  3 +--
 net/bluetooth/hci_sync.c | 53 +++++++++++++++++++++++++++++++++++++++++++-----
 2 files changed, 49 insertions(+), 7 deletions(-)

diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 9bdd3e1f8cfc..fc0424ae551e 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -2552,7 +2552,6 @@ struct hci_dev *hci_alloc_dev_priv(int sizeof_priv)
 	INIT_DELAYED_WORK(&hdev->ncmd_timer, hci_ncmd_timeout);
 
 	hci_devcd_setup(hdev);
-	hci_request_setup(hdev);
 
 	hci_init_sysfs(hdev);
 	discovery_init(hdev);
@@ -4074,7 +4073,7 @@ static void hci_send_cmd_sync(struct hci_dev *hdev, struct sk_buff *skb)
 		return;
 	}
 
-	if (hci_req_status_pend(hdev) &&
+	if (hdev->req_status == HCI_REQ_PEND &&
 	    !hci_dev_test_and_set_flag(hdev, HCI_CMD_PENDING)) {
 		kfree_skb(hdev->req_skb);
 		hdev->req_skb = skb_clone(hdev->sent_cmd, GFP_KERNEL);
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 0c8888448d3c..63548c4cc1ac 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -566,6 +566,53 @@ unlock:
 	hci_dev_unlock(hdev);
 }
 
+static bool is_interleave_scanning(struct hci_dev *hdev)
+{
+	return hdev->interleave_scan_state != INTERLEAVE_SCAN_NONE;
+}
+
+static int hci_passive_scan_sync(struct hci_dev *hdev);
+
+static void interleave_scan_work(struct work_struct *work)
+{
+	struct hci_dev *hdev = container_of(work, struct hci_dev,
+					    interleave_scan.work);
+	unsigned long timeout;
+
+	if (hdev->interleave_scan_state == INTERLEAVE_SCAN_ALLOWLIST) {
+		timeout = msecs_to_jiffies(hdev->advmon_allowlist_duration);
+	} else if (hdev->interleave_scan_state == INTERLEAVE_SCAN_NO_FILTER) {
+		timeout = msecs_to_jiffies(hdev->advmon_no_filter_duration);
+	} else {
+		bt_dev_err(hdev, "unexpected error");
+		return;
+	}
+
+	hci_passive_scan_sync(hdev);
+
+	hci_dev_lock(hdev);
+
+	switch (hdev->interleave_scan_state) {
+	case INTERLEAVE_SCAN_ALLOWLIST:
+		bt_dev_dbg(hdev, "next state: allowlist");
+		hdev->interleave_scan_state = INTERLEAVE_SCAN_NO_FILTER;
+		break;
+	case INTERLEAVE_SCAN_NO_FILTER:
+		bt_dev_dbg(hdev, "next state: no filter");
+		hdev->interleave_scan_state = INTERLEAVE_SCAN_ALLOWLIST;
+		break;
+	case INTERLEAVE_SCAN_NONE:
+		bt_dev_err(hdev, "unexpected error");
+	}
+
+	hci_dev_unlock(hdev);
+
+	/* Don't continue interleaving if it was canceled */
+	if (is_interleave_scanning(hdev))
+		queue_delayed_work(hdev->req_workqueue,
+				   &hdev->interleave_scan, timeout);
+}
+
 void hci_cmd_sync_init(struct hci_dev *hdev)
 {
 	INIT_WORK(&hdev->cmd_sync_work, hci_cmd_sync_work);
@@ -577,6 +624,7 @@ void hci_cmd_sync_init(struct hci_dev *hdev)
 	INIT_WORK(&hdev->reenable_adv_work, reenable_adv);
 	INIT_DELAYED_WORK(&hdev->le_scan_disable, le_scan_disable);
 	INIT_DELAYED_WORK(&hdev->adv_instance_expire, adv_timeout_expire);
+	INIT_DELAYED_WORK(&hdev->interleave_scan, interleave_scan_work);
 }
 
 static void _hci_cmd_sync_cancel_entry(struct hci_dev *hdev,
@@ -2110,11 +2158,6 @@ static void hci_start_interleave_scan(struct hci_dev *hdev)
 			   &hdev->interleave_scan, 0);
 }
 
-static bool is_interleave_scanning(struct hci_dev *hdev)
-{
-	return hdev->interleave_scan_state != INTERLEAVE_SCAN_NONE;
-}
-
 static void cancel_interleave_scan(struct hci_dev *hdev)
 {
 	bt_dev_dbg(hdev, "cancelling interleave scan");
-- 
cgit v1.2.3-58-ga151


From f2d89775358606c7ab6b6b6c4a02fe1e8cd270b1 Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Mon, 1 Jul 2024 16:52:57 -0400
Subject: Bluetooth: hci_sync: Remove remaining dependencies of hci_request

This removes the dependencies of hci_req_init and hci_request_cancel_all
from hci_sync.c.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/hci_sync.h | 17 +++++++++++++++++
 net/bluetooth/hci_request.h      | 17 -----------------
 net/bluetooth/hci_sync.c         | 14 +++++++++++---
 3 files changed, 28 insertions(+), 20 deletions(-)

diff --git a/include/net/bluetooth/hci_sync.h b/include/net/bluetooth/hci_sync.h
index a8d88247ac89..75e052909b5f 100644
--- a/include/net/bluetooth/hci_sync.h
+++ b/include/net/bluetooth/hci_sync.h
@@ -8,6 +8,23 @@
 #define UINT_PTR(_handle)		((void *)((uintptr_t)_handle))
 #define PTR_UINT(_ptr)			((uintptr_t)((void *)_ptr))
 
+#define HCI_REQ_DONE	  0
+#define HCI_REQ_PEND	  1
+#define HCI_REQ_CANCELED  2
+
+#define hci_req_sync_lock(hdev)   mutex_lock(&hdev->req_lock)
+#define hci_req_sync_unlock(hdev) mutex_unlock(&hdev->req_lock)
+
+struct hci_request {
+	struct hci_dev		*hdev;
+	struct sk_buff_head	cmd_q;
+
+	/* If something goes wrong when building the HCI request, the error
+	 * value is stored in this field.
+	 */
+	int			err;
+};
+
 typedef int (*hci_cmd_sync_work_func_t)(struct hci_dev *hdev, void *data);
 typedef void (*hci_cmd_sync_work_destroy_t)(struct hci_dev *hdev, void *data,
 					    int err);
diff --git a/net/bluetooth/hci_request.h b/net/bluetooth/hci_request.h
index c91f2838f542..b730da4a8b47 100644
--- a/net/bluetooth/hci_request.h
+++ b/net/bluetooth/hci_request.h
@@ -22,23 +22,6 @@
 
 #include <asm/unaligned.h>
 
-#define HCI_REQ_DONE	  0
-#define HCI_REQ_PEND	  1
-#define HCI_REQ_CANCELED  2
-
-#define hci_req_sync_lock(hdev)   mutex_lock(&hdev->req_lock)
-#define hci_req_sync_unlock(hdev) mutex_unlock(&hdev->req_lock)
-
-struct hci_request {
-	struct hci_dev		*hdev;
-	struct sk_buff_head	cmd_q;
-
-	/* If something goes wrong when building the HCI request, the error
-	 * value is stored in this field.
-	 */
-	int			err;
-};
-
 void hci_req_init(struct hci_request *req, struct hci_dev *hdev);
 void hci_req_purge(struct hci_request *req);
 bool hci_req_status_pend(struct hci_dev *hdev);
diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c
index 63548c4cc1ac..cd2ed16da8a4 100644
--- a/net/bluetooth/hci_sync.c
+++ b/net/bluetooth/hci_sync.c
@@ -12,7 +12,6 @@
 #include <net/bluetooth/hci_core.h>
 #include <net/bluetooth/mgmt.h>
 
-#include "hci_request.h"
 #include "hci_codec.h"
 #include "hci_debugfs.h"
 #include "smp.h"
@@ -146,6 +145,13 @@ static int hci_cmd_sync_run(struct hci_request *req)
 	return 0;
 }
 
+static void hci_request_init(struct hci_request *req, struct hci_dev *hdev)
+{
+	skb_queue_head_init(&req->cmd_q);
+	req->hdev = hdev;
+	req->err = 0;
+}
+
 /* This function requires the caller holds hdev->req_lock. */
 struct sk_buff *__hci_cmd_sync_sk(struct hci_dev *hdev, u16 opcode, u32 plen,
 				  const void *param, u8 event, u32 timeout,
@@ -157,7 +163,7 @@ struct sk_buff *__hci_cmd_sync_sk(struct hci_dev *hdev, u16 opcode, u32 plen,
 
 	bt_dev_dbg(hdev, "Opcode 0x%4.4x", opcode);
 
-	hci_req_init(&req, hdev);
+	hci_request_init(&req, hdev);
 
 	hci_cmd_sync_add(&req, opcode, plen, param, event, sk);
 
@@ -5056,7 +5062,9 @@ int hci_dev_close_sync(struct hci_dev *hdev)
 	cancel_delayed_work(&hdev->ncmd_timer);
 	cancel_delayed_work(&hdev->le_scan_disable);
 
-	hci_request_cancel_all(hdev);
+	hci_cmd_sync_cancel_sync(hdev, ENODEV);
+
+	cancel_interleave_scan(hdev);
 
 	if (hdev->adv_instance_timeout) {
 		cancel_delayed_work_sync(&hdev->adv_instance_expire);
-- 
cgit v1.2.3-58-ga151


From 936daee9cf08c5e58c9a0fe687f52adb2d80e87d Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Mon, 1 Jul 2024 17:10:41 -0400
Subject: Bluetooth: Remove hci_request.{c,h}

This removes hci_request.{c,h} since it shall no longer be used.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 include/net/bluetooth/bluetooth.h |   4 +
 net/bluetooth/Makefile            |   3 +-
 net/bluetooth/hci_conn.c          |   1 -
 net/bluetooth/hci_core.c          |   1 -
 net/bluetooth/hci_debugfs.c       |   1 -
 net/bluetooth/hci_event.c         |   1 -
 net/bluetooth/hci_request.c       | 903 --------------------------------------
 net/bluetooth/hci_request.h       |  54 ---
 net/bluetooth/mgmt.c              |   1 -
 net/bluetooth/msft.c              |   1 -
 10 files changed, 5 insertions(+), 965 deletions(-)
 delete mode 100644 net/bluetooth/hci_request.c
 delete mode 100644 net/bluetooth/hci_request.h

diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h
index b3228bd6cd6b..5d655e109b2c 100644
--- a/include/net/bluetooth/bluetooth.h
+++ b/include/net/bluetooth/bluetooth.h
@@ -441,6 +441,10 @@ typedef void (*hci_req_complete_t)(struct hci_dev *hdev, u8 status, u16 opcode);
 typedef void (*hci_req_complete_skb_t)(struct hci_dev *hdev, u8 status,
 				       u16 opcode, struct sk_buff *skb);
 
+void hci_req_cmd_complete(struct hci_dev *hdev, u16 opcode, u8 status,
+			  hci_req_complete_t *req_complete,
+			  hci_req_complete_skb_t *req_complete_skb);
+
 #define HCI_REQ_START	BIT(0)
 #define HCI_REQ_SKB	BIT(1)
 
diff --git a/net/bluetooth/Makefile b/net/bluetooth/Makefile
index 628d448d78be..5a3835b7dfcd 100644
--- a/net/bluetooth/Makefile
+++ b/net/bluetooth/Makefile
@@ -14,8 +14,7 @@ bluetooth_6lowpan-y := 6lowpan.o
 
 bluetooth-y := af_bluetooth.o hci_core.o hci_conn.o hci_event.o mgmt.o \
 	hci_sock.o hci_sysfs.o l2cap_core.o l2cap_sock.o smp.o lib.o \
-	ecdh_helper.o hci_request.o mgmt_util.o mgmt_config.o hci_codec.o \
-	eir.o hci_sync.o
+	ecdh_helper.o mgmt_util.o mgmt_config.o hci_codec.o eir.o hci_sync.o
 
 bluetooth-$(CONFIG_DEV_COREDUMP) += coredump.o
 
diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c
index 080053a85b4d..8e48ccd2af30 100644
--- a/net/bluetooth/hci_conn.c
+++ b/net/bluetooth/hci_conn.c
@@ -34,7 +34,6 @@
 #include <net/bluetooth/iso.h>
 #include <net/bluetooth/mgmt.h>
 
-#include "hci_request.h"
 #include "smp.h"
 #include "eir.h"
 
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index fc0424ae551e..8a4ebd93adfc 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -40,7 +40,6 @@
 #include <net/bluetooth/l2cap.h>
 #include <net/bluetooth/mgmt.h>
 
-#include "hci_request.h"
 #include "hci_debugfs.h"
 #include "smp.h"
 #include "leds.h"
diff --git a/net/bluetooth/hci_debugfs.c b/net/bluetooth/hci_debugfs.c
index ce3ff2fa72e5..f625074d1f00 100644
--- a/net/bluetooth/hci_debugfs.c
+++ b/net/bluetooth/hci_debugfs.c
@@ -28,7 +28,6 @@
 #include <net/bluetooth/hci_core.h>
 
 #include "smp.h"
-#include "hci_request.h"
 #include "hci_debugfs.h"
 
 #define DEFINE_QUIRK_ATTRIBUTE(__name, __quirk)				      \
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index 4611a67d7dcc..dce8035ca799 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -33,7 +33,6 @@
 #include <net/bluetooth/hci_core.h>
 #include <net/bluetooth/mgmt.h>
 
-#include "hci_request.h"
 #include "hci_debugfs.h"
 #include "hci_codec.h"
 #include "smp.h"
diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c
deleted file mode 100644
index efea25eb56ce..000000000000
--- a/net/bluetooth/hci_request.c
+++ /dev/null
@@ -1,903 +0,0 @@
-/*
-   BlueZ - Bluetooth protocol stack for Linux
-
-   Copyright (C) 2014 Intel Corporation
-
-   This program is free software; you can redistribute it and/or modify
-   it under the terms of the GNU General Public License version 2 as
-   published by the Free Software Foundation;
-
-   THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
-   OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-   FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OF THIRD PARTY RIGHTS.
-   IN NO EVENT SHALL THE COPYRIGHT HOLDER(S) AND AUTHOR(S) BE LIABLE FOR ANY
-   CLAIM, OR ANY SPECIAL INDIRECT OR CONSEQUENTIAL DAMAGES, OR ANY DAMAGES
-   WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
-   ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
-   OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
-
-   ALL LIABILITY, INCLUDING LIABILITY FOR INFRINGEMENT OF ANY PATENTS,
-   COPYRIGHTS, TRADEMARKS OR OTHER RIGHTS, RELATING TO USE OF THIS
-   SOFTWARE IS DISCLAIMED.
-*/
-
-#include <linux/sched/signal.h>
-
-#include <net/bluetooth/bluetooth.h>
-#include <net/bluetooth/hci_core.h>
-#include <net/bluetooth/mgmt.h>
-
-#include "smp.h"
-#include "hci_request.h"
-#include "msft.h"
-#include "eir.h"
-
-void hci_req_init(struct hci_request *req, struct hci_dev *hdev)
-{
-	skb_queue_head_init(&req->cmd_q);
-	req->hdev = hdev;
-	req->err = 0;
-}
-
-void hci_req_purge(struct hci_request *req)
-{
-	skb_queue_purge(&req->cmd_q);
-}
-
-bool hci_req_status_pend(struct hci_dev *hdev)
-{
-	return hdev->req_status == HCI_REQ_PEND;
-}
-
-static int req_run(struct hci_request *req, hci_req_complete_t complete,
-		   hci_req_complete_skb_t complete_skb)
-{
-	struct hci_dev *hdev = req->hdev;
-	struct sk_buff *skb;
-	unsigned long flags;
-
-	bt_dev_dbg(hdev, "length %u", skb_queue_len(&req->cmd_q));
-
-	/* If an error occurred during request building, remove all HCI
-	 * commands queued on the HCI request queue.
-	 */
-	if (req->err) {
-		skb_queue_purge(&req->cmd_q);
-		return req->err;
-	}
-
-	/* Do not allow empty requests */
-	if (skb_queue_empty(&req->cmd_q))
-		return -ENODATA;
-
-	skb = skb_peek_tail(&req->cmd_q);
-	if (complete) {
-		bt_cb(skb)->hci.req_complete = complete;
-	} else if (complete_skb) {
-		bt_cb(skb)->hci.req_complete_skb = complete_skb;
-		bt_cb(skb)->hci.req_flags |= HCI_REQ_SKB;
-	}
-
-	spin_lock_irqsave(&hdev->cmd_q.lock, flags);
-	skb_queue_splice_tail(&req->cmd_q, &hdev->cmd_q);
-	spin_unlock_irqrestore(&hdev->cmd_q.lock, flags);
-
-	queue_work(hdev->workqueue, &hdev->cmd_work);
-
-	return 0;
-}
-
-int hci_req_run(struct hci_request *req, hci_req_complete_t complete)
-{
-	return req_run(req, complete, NULL);
-}
-
-int hci_req_run_skb(struct hci_request *req, hci_req_complete_skb_t complete)
-{
-	return req_run(req, NULL, complete);
-}
-
-void hci_req_sync_complete(struct hci_dev *hdev, u8 result, u16 opcode,
-			   struct sk_buff *skb)
-{
-	bt_dev_dbg(hdev, "result 0x%2.2x", result);
-
-	if (hdev->req_status == HCI_REQ_PEND) {
-		hdev->req_result = result;
-		hdev->req_status = HCI_REQ_DONE;
-		if (skb) {
-			kfree_skb(hdev->req_skb);
-			hdev->req_skb = skb_get(skb);
-		}
-		wake_up_interruptible(&hdev->req_wait_q);
-	}
-}
-
-/* Execute request and wait for completion. */
-int __hci_req_sync(struct hci_dev *hdev, int (*func)(struct hci_request *req,
-						     unsigned long opt),
-		   unsigned long opt, u32 timeout, u8 *hci_status)
-{
-	struct hci_request req;
-	int err = 0;
-
-	bt_dev_dbg(hdev, "start");
-
-	hci_req_init(&req, hdev);
-
-	hdev->req_status = HCI_REQ_PEND;
-
-	err = func(&req, opt);
-	if (err) {
-		if (hci_status)
-			*hci_status = HCI_ERROR_UNSPECIFIED;
-		return err;
-	}
-
-	err = hci_req_run_skb(&req, hci_req_sync_complete);
-	if (err < 0) {
-		hdev->req_status = 0;
-
-		/* ENODATA means the HCI request command queue is empty.
-		 * This can happen when a request with conditionals doesn't
-		 * trigger any commands to be sent. This is normal behavior
-		 * and should not trigger an error return.
-		 */
-		if (err == -ENODATA) {
-			if (hci_status)
-				*hci_status = 0;
-			return 0;
-		}
-
-		if (hci_status)
-			*hci_status = HCI_ERROR_UNSPECIFIED;
-
-		return err;
-	}
-
-	err = wait_event_interruptible_timeout(hdev->req_wait_q,
-			hdev->req_status != HCI_REQ_PEND, timeout);
-
-	if (err == -ERESTARTSYS)
-		return -EINTR;
-
-	switch (hdev->req_status) {
-	case HCI_REQ_DONE:
-		err = -bt_to_errno(hdev->req_result);
-		if (hci_status)
-			*hci_status = hdev->req_result;
-		break;
-
-	case HCI_REQ_CANCELED:
-		err = -hdev->req_result;
-		if (hci_status)
-			*hci_status = HCI_ERROR_UNSPECIFIED;
-		break;
-
-	default:
-		err = -ETIMEDOUT;
-		if (hci_status)
-			*hci_status = HCI_ERROR_UNSPECIFIED;
-		break;
-	}
-
-	kfree_skb(hdev->req_skb);
-	hdev->req_skb = NULL;
-	hdev->req_status = hdev->req_result = 0;
-
-	bt_dev_dbg(hdev, "end: err %d", err);
-
-	return err;
-}
-
-int hci_req_sync(struct hci_dev *hdev, int (*req)(struct hci_request *req,
-						  unsigned long opt),
-		 unsigned long opt, u32 timeout, u8 *hci_status)
-{
-	int ret;
-
-	/* Serialize all requests */
-	hci_req_sync_lock(hdev);
-	/* check the state after obtaing the lock to protect the HCI_UP
-	 * against any races from hci_dev_do_close when the controller
-	 * gets removed.
-	 */
-	if (test_bit(HCI_UP, &hdev->flags))
-		ret = __hci_req_sync(hdev, req, opt, timeout, hci_status);
-	else
-		ret = -ENETDOWN;
-	hci_req_sync_unlock(hdev);
-
-	return ret;
-}
-
-struct sk_buff *hci_prepare_cmd(struct hci_dev *hdev, u16 opcode, u32 plen,
-				const void *param)
-{
-	int len = HCI_COMMAND_HDR_SIZE + plen;
-	struct hci_command_hdr *hdr;
-	struct sk_buff *skb;
-
-	skb = bt_skb_alloc(len, GFP_ATOMIC);
-	if (!skb)
-		return NULL;
-
-	hdr = skb_put(skb, HCI_COMMAND_HDR_SIZE);
-	hdr->opcode = cpu_to_le16(opcode);
-	hdr->plen   = plen;
-
-	if (plen)
-		skb_put_data(skb, param, plen);
-
-	bt_dev_dbg(hdev, "skb len %d", skb->len);
-
-	hci_skb_pkt_type(skb) = HCI_COMMAND_PKT;
-	hci_skb_opcode(skb) = opcode;
-
-	return skb;
-}
-
-/* Queue a command to an asynchronous HCI request */
-void hci_req_add_ev(struct hci_request *req, u16 opcode, u32 plen,
-		    const void *param, u8 event)
-{
-	struct hci_dev *hdev = req->hdev;
-	struct sk_buff *skb;
-
-	bt_dev_dbg(hdev, "opcode 0x%4.4x plen %d", opcode, plen);
-
-	/* If an error occurred during request building, there is no point in
-	 * queueing the HCI command. We can simply return.
-	 */
-	if (req->err)
-		return;
-
-	skb = hci_prepare_cmd(hdev, opcode, plen, param);
-	if (!skb) {
-		bt_dev_err(hdev, "no memory for command (opcode 0x%4.4x)",
-			   opcode);
-		req->err = -ENOMEM;
-		return;
-	}
-
-	if (skb_queue_empty(&req->cmd_q))
-		bt_cb(skb)->hci.req_flags |= HCI_REQ_START;
-
-	hci_skb_event(skb) = event;
-
-	skb_queue_tail(&req->cmd_q, skb);
-}
-
-void hci_req_add(struct hci_request *req, u16 opcode, u32 plen,
-		 const void *param)
-{
-	bt_dev_dbg(req->hdev, "HCI_REQ-0x%4.4x", opcode);
-	hci_req_add_ev(req, opcode, plen, param, 0);
-}
-
-static void start_interleave_scan(struct hci_dev *hdev)
-{
-	hdev->interleave_scan_state = INTERLEAVE_SCAN_NO_FILTER;
-	queue_delayed_work(hdev->req_workqueue,
-			   &hdev->interleave_scan, 0);
-}
-
-static bool is_interleave_scanning(struct hci_dev *hdev)
-{
-	return hdev->interleave_scan_state != INTERLEAVE_SCAN_NONE;
-}
-
-static void cancel_interleave_scan(struct hci_dev *hdev)
-{
-	bt_dev_dbg(hdev, "cancelling interleave scan");
-
-	cancel_delayed_work_sync(&hdev->interleave_scan);
-
-	hdev->interleave_scan_state = INTERLEAVE_SCAN_NONE;
-}
-
-/* Return true if interleave_scan wasn't started until exiting this function,
- * otherwise, return false
- */
-static bool __hci_update_interleaved_scan(struct hci_dev *hdev)
-{
-	/* Do interleaved scan only if all of the following are true:
-	 * - There is at least one ADV monitor
-	 * - At least one pending LE connection or one device to be scanned for
-	 * - Monitor offloading is not supported
-	 * If so, we should alternate between allowlist scan and one without
-	 * any filters to save power.
-	 */
-	bool use_interleaving = hci_is_adv_monitoring(hdev) &&
-				!(list_empty(&hdev->pend_le_conns) &&
-				  list_empty(&hdev->pend_le_reports)) &&
-				hci_get_adv_monitor_offload_ext(hdev) ==
-				    HCI_ADV_MONITOR_EXT_NONE;
-	bool is_interleaving = is_interleave_scanning(hdev);
-
-	if (use_interleaving && !is_interleaving) {
-		start_interleave_scan(hdev);
-		bt_dev_dbg(hdev, "starting interleave scan");
-		return true;
-	}
-
-	if (!use_interleaving && is_interleaving)
-		cancel_interleave_scan(hdev);
-
-	return false;
-}
-
-void hci_req_add_le_scan_disable(struct hci_request *req, bool rpa_le_conn)
-{
-	struct hci_dev *hdev = req->hdev;
-
-	if (hdev->scanning_paused) {
-		bt_dev_dbg(hdev, "Scanning is paused for suspend");
-		return;
-	}
-
-	if (use_ext_scan(hdev)) {
-		struct hci_cp_le_set_ext_scan_enable cp;
-
-		memset(&cp, 0, sizeof(cp));
-		cp.enable = LE_SCAN_DISABLE;
-		hci_req_add(req, HCI_OP_LE_SET_EXT_SCAN_ENABLE, sizeof(cp),
-			    &cp);
-	} else {
-		struct hci_cp_le_set_scan_enable cp;
-
-		memset(&cp, 0, sizeof(cp));
-		cp.enable = LE_SCAN_DISABLE;
-		hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(cp), &cp);
-	}
-
-	/* Disable address resolution */
-	if (hci_dev_test_flag(hdev, HCI_LL_RPA_RESOLUTION) && !rpa_le_conn) {
-		__u8 enable = 0x00;
-
-		hci_req_add(req, HCI_OP_LE_SET_ADDR_RESOLV_ENABLE, 1, &enable);
-	}
-}
-
-static void del_from_accept_list(struct hci_request *req, bdaddr_t *bdaddr,
-				 u8 bdaddr_type)
-{
-	struct hci_cp_le_del_from_accept_list cp;
-
-	cp.bdaddr_type = bdaddr_type;
-	bacpy(&cp.bdaddr, bdaddr);
-
-	bt_dev_dbg(req->hdev, "Remove %pMR (0x%x) from accept list", &cp.bdaddr,
-		   cp.bdaddr_type);
-	hci_req_add(req, HCI_OP_LE_DEL_FROM_ACCEPT_LIST, sizeof(cp), &cp);
-
-	if (use_ll_privacy(req->hdev)) {
-		struct smp_irk *irk;
-
-		irk = hci_find_irk_by_addr(req->hdev, bdaddr, bdaddr_type);
-		if (irk) {
-			struct hci_cp_le_del_from_resolv_list cp;
-
-			cp.bdaddr_type = bdaddr_type;
-			bacpy(&cp.bdaddr, bdaddr);
-
-			hci_req_add(req, HCI_OP_LE_DEL_FROM_RESOLV_LIST,
-				    sizeof(cp), &cp);
-		}
-	}
-}
-
-/* Adds connection to accept list if needed. On error, returns -1. */
-static int add_to_accept_list(struct hci_request *req,
-			      struct hci_conn_params *params, u8 *num_entries,
-			      bool allow_rpa)
-{
-	struct hci_cp_le_add_to_accept_list cp;
-	struct hci_dev *hdev = req->hdev;
-
-	/* Already in accept list */
-	if (hci_bdaddr_list_lookup(&hdev->le_accept_list, &params->addr,
-				   params->addr_type))
-		return 0;
-
-	/* Select filter policy to accept all advertising */
-	if (*num_entries >= hdev->le_accept_list_size)
-		return -1;
-
-	/* Accept list can not be used with RPAs */
-	if (!allow_rpa &&
-	    !hci_dev_test_flag(hdev, HCI_ENABLE_LL_PRIVACY) &&
-	    hci_find_irk_by_addr(hdev, &params->addr, params->addr_type)) {
-		return -1;
-	}
-
-	/* During suspend, only wakeable devices can be in accept list */
-	if (hdev->suspended &&
-	    !(params->flags & HCI_CONN_FLAG_REMOTE_WAKEUP))
-		return 0;
-
-	*num_entries += 1;
-	cp.bdaddr_type = params->addr_type;
-	bacpy(&cp.bdaddr, &params->addr);
-
-	bt_dev_dbg(hdev, "Add %pMR (0x%x) to accept list", &cp.bdaddr,
-		   cp.bdaddr_type);
-	hci_req_add(req, HCI_OP_LE_ADD_TO_ACCEPT_LIST, sizeof(cp), &cp);
-
-	if (use_ll_privacy(hdev)) {
-		struct smp_irk *irk;
-
-		irk = hci_find_irk_by_addr(hdev, &params->addr,
-					   params->addr_type);
-		if (irk) {
-			struct hci_cp_le_add_to_resolv_list cp;
-
-			cp.bdaddr_type = params->addr_type;
-			bacpy(&cp.bdaddr, &params->addr);
-			memcpy(cp.peer_irk, irk->val, 16);
-
-			if (hci_dev_test_flag(hdev, HCI_PRIVACY))
-				memcpy(cp.local_irk, hdev->irk, 16);
-			else
-				memset(cp.local_irk, 0, 16);
-
-			hci_req_add(req, HCI_OP_LE_ADD_TO_RESOLV_LIST,
-				    sizeof(cp), &cp);
-		}
-	}
-
-	return 0;
-}
-
-static u8 update_accept_list(struct hci_request *req)
-{
-	struct hci_dev *hdev = req->hdev;
-	struct hci_conn_params *params;
-	struct bdaddr_list *b;
-	u8 num_entries = 0;
-	bool pend_conn, pend_report;
-	/* We allow usage of accept list even with RPAs in suspend. In the worst
-	 * case, we won't be able to wake from devices that use the privacy1.2
-	 * features. Additionally, once we support privacy1.2 and IRK
-	 * offloading, we can update this to also check for those conditions.
-	 */
-	bool allow_rpa = hdev->suspended;
-
-	if (use_ll_privacy(hdev))
-		allow_rpa = true;
-
-	/* Go through the current accept list programmed into the
-	 * controller one by one and check if that address is still
-	 * in the list of pending connections or list of devices to
-	 * report. If not present in either list, then queue the
-	 * command to remove it from the controller.
-	 */
-	list_for_each_entry(b, &hdev->le_accept_list, list) {
-		pend_conn = hci_pend_le_action_lookup(&hdev->pend_le_conns,
-						      &b->bdaddr,
-						      b->bdaddr_type);
-		pend_report = hci_pend_le_action_lookup(&hdev->pend_le_reports,
-							&b->bdaddr,
-							b->bdaddr_type);
-
-		/* If the device is not likely to connect or report,
-		 * remove it from the accept list.
-		 */
-		if (!pend_conn && !pend_report) {
-			del_from_accept_list(req, &b->bdaddr, b->bdaddr_type);
-			continue;
-		}
-
-		/* Accept list can not be used with RPAs */
-		if (!allow_rpa &&
-		    !hci_dev_test_flag(hdev, HCI_ENABLE_LL_PRIVACY) &&
-		    hci_find_irk_by_addr(hdev, &b->bdaddr, b->bdaddr_type)) {
-			return 0x00;
-		}
-
-		num_entries++;
-	}
-
-	/* Since all no longer valid accept list entries have been
-	 * removed, walk through the list of pending connections
-	 * and ensure that any new device gets programmed into
-	 * the controller.
-	 *
-	 * If the list of the devices is larger than the list of
-	 * available accept list entries in the controller, then
-	 * just abort and return filer policy value to not use the
-	 * accept list.
-	 */
-	list_for_each_entry(params, &hdev->pend_le_conns, action) {
-		if (add_to_accept_list(req, params, &num_entries, allow_rpa))
-			return 0x00;
-	}
-
-	/* After adding all new pending connections, walk through
-	 * the list of pending reports and also add these to the
-	 * accept list if there is still space. Abort if space runs out.
-	 */
-	list_for_each_entry(params, &hdev->pend_le_reports, action) {
-		if (add_to_accept_list(req, params, &num_entries, allow_rpa))
-			return 0x00;
-	}
-
-	/* Use the allowlist unless the following conditions are all true:
-	 * - We are not currently suspending
-	 * - There are 1 or more ADV monitors registered and it's not offloaded
-	 * - Interleaved scanning is not currently using the allowlist
-	 */
-	if (!idr_is_empty(&hdev->adv_monitors_idr) && !hdev->suspended &&
-	    hci_get_adv_monitor_offload_ext(hdev) == HCI_ADV_MONITOR_EXT_NONE &&
-	    hdev->interleave_scan_state != INTERLEAVE_SCAN_ALLOWLIST)
-		return 0x00;
-
-	/* Select filter policy to use accept list */
-	return 0x01;
-}
-
-static bool scan_use_rpa(struct hci_dev *hdev)
-{
-	return hci_dev_test_flag(hdev, HCI_PRIVACY);
-}
-
-static void hci_req_start_scan(struct hci_request *req, u8 type, u16 interval,
-			       u16 window, u8 own_addr_type, u8 filter_policy,
-			       bool filter_dup, bool addr_resolv)
-{
-	struct hci_dev *hdev = req->hdev;
-
-	if (hdev->scanning_paused) {
-		bt_dev_dbg(hdev, "Scanning is paused for suspend");
-		return;
-	}
-
-	if (use_ll_privacy(hdev) && addr_resolv) {
-		u8 enable = 0x01;
-
-		hci_req_add(req, HCI_OP_LE_SET_ADDR_RESOLV_ENABLE, 1, &enable);
-	}
-
-	/* Use ext scanning if set ext scan param and ext scan enable is
-	 * supported
-	 */
-	if (use_ext_scan(hdev)) {
-		struct hci_cp_le_set_ext_scan_params *ext_param_cp;
-		struct hci_cp_le_set_ext_scan_enable ext_enable_cp;
-		struct hci_cp_le_scan_phy_params *phy_params;
-		u8 data[sizeof(*ext_param_cp) + sizeof(*phy_params) * 2];
-		u32 plen;
-
-		ext_param_cp = (void *)data;
-		phy_params = (void *)ext_param_cp->data;
-
-		memset(ext_param_cp, 0, sizeof(*ext_param_cp));
-		ext_param_cp->own_addr_type = own_addr_type;
-		ext_param_cp->filter_policy = filter_policy;
-
-		plen = sizeof(*ext_param_cp);
-
-		if (scan_1m(hdev) || scan_2m(hdev)) {
-			ext_param_cp->scanning_phys |= LE_SCAN_PHY_1M;
-
-			memset(phy_params, 0, sizeof(*phy_params));
-			phy_params->type = type;
-			phy_params->interval = cpu_to_le16(interval);
-			phy_params->window = cpu_to_le16(window);
-
-			plen += sizeof(*phy_params);
-			phy_params++;
-		}
-
-		if (scan_coded(hdev)) {
-			ext_param_cp->scanning_phys |= LE_SCAN_PHY_CODED;
-
-			memset(phy_params, 0, sizeof(*phy_params));
-			phy_params->type = type;
-			phy_params->interval = cpu_to_le16(interval);
-			phy_params->window = cpu_to_le16(window);
-
-			plen += sizeof(*phy_params);
-			phy_params++;
-		}
-
-		hci_req_add(req, HCI_OP_LE_SET_EXT_SCAN_PARAMS,
-			    plen, ext_param_cp);
-
-		memset(&ext_enable_cp, 0, sizeof(ext_enable_cp));
-		ext_enable_cp.enable = LE_SCAN_ENABLE;
-		ext_enable_cp.filter_dup = filter_dup;
-
-		hci_req_add(req, HCI_OP_LE_SET_EXT_SCAN_ENABLE,
-			    sizeof(ext_enable_cp), &ext_enable_cp);
-	} else {
-		struct hci_cp_le_set_scan_param param_cp;
-		struct hci_cp_le_set_scan_enable enable_cp;
-
-		memset(&param_cp, 0, sizeof(param_cp));
-		param_cp.type = type;
-		param_cp.interval = cpu_to_le16(interval);
-		param_cp.window = cpu_to_le16(window);
-		param_cp.own_address_type = own_addr_type;
-		param_cp.filter_policy = filter_policy;
-		hci_req_add(req, HCI_OP_LE_SET_SCAN_PARAM, sizeof(param_cp),
-			    &param_cp);
-
-		memset(&enable_cp, 0, sizeof(enable_cp));
-		enable_cp.enable = LE_SCAN_ENABLE;
-		enable_cp.filter_dup = filter_dup;
-		hci_req_add(req, HCI_OP_LE_SET_SCAN_ENABLE, sizeof(enable_cp),
-			    &enable_cp);
-	}
-}
-
-static void set_random_addr(struct hci_request *req, bdaddr_t *rpa);
-static int hci_update_random_address(struct hci_request *req,
-				     bool require_privacy, bool use_rpa,
-				     u8 *own_addr_type)
-{
-	struct hci_dev *hdev = req->hdev;
-	int err;
-
-	/* If privacy is enabled use a resolvable private address. If
-	 * current RPA has expired or there is something else than
-	 * the current RPA in use, then generate a new one.
-	 */
-	if (use_rpa) {
-		/* If Controller supports LL Privacy use own address type is
-		 * 0x03
-		 */
-		if (use_ll_privacy(hdev))
-			*own_addr_type = ADDR_LE_DEV_RANDOM_RESOLVED;
-		else
-			*own_addr_type = ADDR_LE_DEV_RANDOM;
-
-		if (rpa_valid(hdev))
-			return 0;
-
-		err = smp_generate_rpa(hdev, hdev->irk, &hdev->rpa);
-		if (err < 0) {
-			bt_dev_err(hdev, "failed to generate new RPA");
-			return err;
-		}
-
-		set_random_addr(req, &hdev->rpa);
-
-		return 0;
-	}
-
-	/* In case of required privacy without resolvable private address,
-	 * use an non-resolvable private address. This is useful for active
-	 * scanning and non-connectable advertising.
-	 */
-	if (require_privacy) {
-		bdaddr_t nrpa;
-
-		while (true) {
-			/* The non-resolvable private address is generated
-			 * from random six bytes with the two most significant
-			 * bits cleared.
-			 */
-			get_random_bytes(&nrpa, 6);
-			nrpa.b[5] &= 0x3f;
-
-			/* The non-resolvable private address shall not be
-			 * equal to the public address.
-			 */
-			if (bacmp(&hdev->bdaddr, &nrpa))
-				break;
-		}
-
-		*own_addr_type = ADDR_LE_DEV_RANDOM;
-		set_random_addr(req, &nrpa);
-		return 0;
-	}
-
-	/* If forcing static address is in use or there is no public
-	 * address use the static address as random address (but skip
-	 * the HCI command if the current random address is already the
-	 * static one.
-	 *
-	 * In case BR/EDR has been disabled on a dual-mode controller
-	 * and a static address has been configured, then use that
-	 * address instead of the public BR/EDR address.
-	 */
-	if (hci_dev_test_flag(hdev, HCI_FORCE_STATIC_ADDR) ||
-	    !bacmp(&hdev->bdaddr, BDADDR_ANY) ||
-	    (!hci_dev_test_flag(hdev, HCI_BREDR_ENABLED) &&
-	     bacmp(&hdev->static_addr, BDADDR_ANY))) {
-		*own_addr_type = ADDR_LE_DEV_RANDOM;
-		if (bacmp(&hdev->static_addr, &hdev->random_addr))
-			hci_req_add(req, HCI_OP_LE_SET_RANDOM_ADDR, 6,
-				    &hdev->static_addr);
-		return 0;
-	}
-
-	/* Neither privacy nor static address is being used so use a
-	 * public address.
-	 */
-	*own_addr_type = ADDR_LE_DEV_PUBLIC;
-
-	return 0;
-}
-
-/* Ensure to call hci_req_add_le_scan_disable() first to disable the
- * controller based address resolution to be able to reconfigure
- * resolving list.
- */
-void hci_req_add_le_passive_scan(struct hci_request *req)
-{
-	struct hci_dev *hdev = req->hdev;
-	u8 own_addr_type;
-	u8 filter_policy;
-	u16 window, interval;
-	/* Default is to enable duplicates filter */
-	u8 filter_dup = LE_SCAN_FILTER_DUP_ENABLE;
-	/* Background scanning should run with address resolution */
-	bool addr_resolv = true;
-
-	if (hdev->scanning_paused) {
-		bt_dev_dbg(hdev, "Scanning is paused for suspend");
-		return;
-	}
-
-	/* Set require_privacy to false since no SCAN_REQ are send
-	 * during passive scanning. Not using an non-resolvable address
-	 * here is important so that peer devices using direct
-	 * advertising with our address will be correctly reported
-	 * by the controller.
-	 */
-	if (hci_update_random_address(req, false, scan_use_rpa(hdev),
-				      &own_addr_type))
-		return;
-
-	if (hdev->enable_advmon_interleave_scan &&
-	    __hci_update_interleaved_scan(hdev))
-		return;
-
-	bt_dev_dbg(hdev, "interleave state %d", hdev->interleave_scan_state);
-	/* Adding or removing entries from the accept list must
-	 * happen before enabling scanning. The controller does
-	 * not allow accept list modification while scanning.
-	 */
-	filter_policy = update_accept_list(req);
-
-	/* When the controller is using random resolvable addresses and
-	 * with that having LE privacy enabled, then controllers with
-	 * Extended Scanner Filter Policies support can now enable support
-	 * for handling directed advertising.
-	 *
-	 * So instead of using filter polices 0x00 (no accept list)
-	 * and 0x01 (accept list enabled) use the new filter policies
-	 * 0x02 (no accept list) and 0x03 (accept list enabled).
-	 */
-	if (hci_dev_test_flag(hdev, HCI_PRIVACY) &&
-	    (hdev->le_features[0] & HCI_LE_EXT_SCAN_POLICY))
-		filter_policy |= 0x02;
-
-	if (hdev->suspended) {
-		window = hdev->le_scan_window_suspend;
-		interval = hdev->le_scan_int_suspend;
-	} else if (hci_is_le_conn_scanning(hdev)) {
-		window = hdev->le_scan_window_connect;
-		interval = hdev->le_scan_int_connect;
-	} else if (hci_is_adv_monitoring(hdev)) {
-		window = hdev->le_scan_window_adv_monitor;
-		interval = hdev->le_scan_int_adv_monitor;
-
-		/* Disable duplicates filter when scanning for advertisement
-		 * monitor for the following reasons.
-		 *
-		 * For HW pattern filtering (ex. MSFT), Realtek and Qualcomm
-		 * controllers ignore RSSI_Sampling_Period when the duplicates
-		 * filter is enabled.
-		 *
-		 * For SW pattern filtering, when we're not doing interleaved
-		 * scanning, it is necessary to disable duplicates filter,
-		 * otherwise hosts can only receive one advertisement and it's
-		 * impossible to know if a peer is still in range.
-		 */
-		filter_dup = LE_SCAN_FILTER_DUP_DISABLE;
-	} else {
-		window = hdev->le_scan_window;
-		interval = hdev->le_scan_interval;
-	}
-
-	bt_dev_dbg(hdev, "LE passive scan with accept list = %d",
-		   filter_policy);
-	hci_req_start_scan(req, LE_SCAN_PASSIVE, interval, window,
-			   own_addr_type, filter_policy, filter_dup,
-			   addr_resolv);
-}
-
-static int hci_req_add_le_interleaved_scan(struct hci_request *req,
-					   unsigned long opt)
-{
-	struct hci_dev *hdev = req->hdev;
-	int ret = 0;
-
-	hci_dev_lock(hdev);
-
-	if (hci_dev_test_flag(hdev, HCI_LE_SCAN))
-		hci_req_add_le_scan_disable(req, false);
-	hci_req_add_le_passive_scan(req);
-
-	switch (hdev->interleave_scan_state) {
-	case INTERLEAVE_SCAN_ALLOWLIST:
-		bt_dev_dbg(hdev, "next state: allowlist");
-		hdev->interleave_scan_state = INTERLEAVE_SCAN_NO_FILTER;
-		break;
-	case INTERLEAVE_SCAN_NO_FILTER:
-		bt_dev_dbg(hdev, "next state: no filter");
-		hdev->interleave_scan_state = INTERLEAVE_SCAN_ALLOWLIST;
-		break;
-	case INTERLEAVE_SCAN_NONE:
-		BT_ERR("unexpected error");
-		ret = -1;
-	}
-
-	hci_dev_unlock(hdev);
-
-	return ret;
-}
-
-static void interleave_scan_work(struct work_struct *work)
-{
-	struct hci_dev *hdev = container_of(work, struct hci_dev,
-					    interleave_scan.work);
-	u8 status;
-	unsigned long timeout;
-
-	if (hdev->interleave_scan_state == INTERLEAVE_SCAN_ALLOWLIST) {
-		timeout = msecs_to_jiffies(hdev->advmon_allowlist_duration);
-	} else if (hdev->interleave_scan_state == INTERLEAVE_SCAN_NO_FILTER) {
-		timeout = msecs_to_jiffies(hdev->advmon_no_filter_duration);
-	} else {
-		bt_dev_err(hdev, "unexpected error");
-		return;
-	}
-
-	hci_req_sync(hdev, hci_req_add_le_interleaved_scan, 0,
-		     HCI_CMD_TIMEOUT, &status);
-
-	/* Don't continue interleaving if it was canceled */
-	if (is_interleave_scanning(hdev))
-		queue_delayed_work(hdev->req_workqueue,
-				   &hdev->interleave_scan, timeout);
-}
-
-static void set_random_addr(struct hci_request *req, bdaddr_t *rpa)
-{
-	struct hci_dev *hdev = req->hdev;
-
-	/* If we're advertising or initiating an LE connection we can't
-	 * go ahead and change the random address at this time. This is
-	 * because the eventual initiator address used for the
-	 * subsequently created connection will be undefined (some
-	 * controllers use the new address and others the one we had
-	 * when the operation started).
-	 *
-	 * In this kind of scenario skip the update and let the random
-	 * address be updated at the next cycle.
-	 */
-	if (hci_dev_test_flag(hdev, HCI_LE_ADV) ||
-	    hci_lookup_le_connect(hdev)) {
-		bt_dev_dbg(hdev, "Deferring random address update");
-		hci_dev_set_flag(hdev, HCI_RPA_EXPIRED);
-		return;
-	}
-
-	hci_req_add(req, HCI_OP_LE_SET_RANDOM_ADDR, 6, rpa);
-}
-
-void hci_request_setup(struct hci_dev *hdev)
-{
-	INIT_DELAYED_WORK(&hdev->interleave_scan, interleave_scan_work);
-}
-
-void hci_request_cancel_all(struct hci_dev *hdev)
-{
-	hci_cmd_sync_cancel_sync(hdev, ENODEV);
-
-	cancel_interleave_scan(hdev);
-}
diff --git a/net/bluetooth/hci_request.h b/net/bluetooth/hci_request.h
deleted file mode 100644
index b730da4a8b47..000000000000
--- a/net/bluetooth/hci_request.h
+++ /dev/null
@@ -1,54 +0,0 @@
-/*
-   BlueZ - Bluetooth protocol stack for Linux
-   Copyright (C) 2014 Intel Corporation
-
-   This program is free software; you can redistribute it and/or modify
-   it under the terms of the GNU General Public License version 2 as
-   published by the Free Software Foundation;
-
-   THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
-   OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-   FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OF THIRD PARTY RIGHTS.
-   IN NO EVENT SHALL THE COPYRIGHT HOLDER(S) AND AUTHOR(S) BE LIABLE FOR ANY
-   CLAIM, OR ANY SPECIAL INDIRECT OR CONSEQUENTIAL DAMAGES, OR ANY DAMAGES
-   WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
-   ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
-   OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
-
-   ALL LIABILITY, INCLUDING LIABILITY FOR INFRINGEMENT OF ANY PATENTS,
-   COPYRIGHTS, TRADEMARKS OR OTHER RIGHTS, RELATING TO USE OF THIS
-   SOFTWARE IS DISCLAIMED.
-*/
-
-#include <asm/unaligned.h>
-
-void hci_req_init(struct hci_request *req, struct hci_dev *hdev);
-void hci_req_purge(struct hci_request *req);
-bool hci_req_status_pend(struct hci_dev *hdev);
-int hci_req_run(struct hci_request *req, hci_req_complete_t complete);
-int hci_req_run_skb(struct hci_request *req, hci_req_complete_skb_t complete);
-void hci_req_sync_complete(struct hci_dev *hdev, u8 result, u16 opcode,
-			   struct sk_buff *skb);
-void hci_req_add(struct hci_request *req, u16 opcode, u32 plen,
-		 const void *param);
-void hci_req_add_ev(struct hci_request *req, u16 opcode, u32 plen,
-		    const void *param, u8 event);
-void hci_req_cmd_complete(struct hci_dev *hdev, u16 opcode, u8 status,
-			  hci_req_complete_t *req_complete,
-			  hci_req_complete_skb_t *req_complete_skb);
-
-int hci_req_sync(struct hci_dev *hdev, int (*req)(struct hci_request *req,
-						  unsigned long opt),
-		 unsigned long opt, u32 timeout, u8 *hci_status);
-int __hci_req_sync(struct hci_dev *hdev, int (*func)(struct hci_request *req,
-						     unsigned long opt),
-		   unsigned long opt, u32 timeout, u8 *hci_status);
-
-struct sk_buff *hci_prepare_cmd(struct hci_dev *hdev, u16 opcode, u32 plen,
-				const void *param);
-
-void hci_req_add_le_scan_disable(struct hci_request *req, bool rpa_le_conn);
-void hci_req_add_le_passive_scan(struct hci_request *req);
-
-void hci_request_setup(struct hci_dev *hdev);
-void hci_request_cancel_all(struct hci_dev *hdev);
diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 3ab1558ff391..40d4887c7f79 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -33,7 +33,6 @@
 #include <net/bluetooth/l2cap.h>
 #include <net/bluetooth/mgmt.h>
 
-#include "hci_request.h"
 #include "smp.h"
 #include "mgmt_util.h"
 #include "mgmt_config.h"
diff --git a/net/bluetooth/msft.c b/net/bluetooth/msft.c
index d039683d3bdd..5a8ccc491b14 100644
--- a/net/bluetooth/msft.c
+++ b/net/bluetooth/msft.c
@@ -7,7 +7,6 @@
 #include <net/bluetooth/hci_core.h>
 #include <net/bluetooth/mgmt.h>
 
-#include "hci_request.h"
 #include "mgmt_util.h"
 #include "msft.h"
 
-- 
cgit v1.2.3-58-ga151


From be71dfb9ad2042bba713975ef50c3a5b0429f1d2 Mon Sep 17 00:00:00 2001
From: Chris Lu <chris.lu@mediatek.com>
Date: Thu, 4 Jul 2024 14:01:09 +0800
Subject: Bluetooth: btusb: mediatek: remove the unnecessary goto tag

Remove the unnecessary goto tag whether there is an error or not, we have
to free the buffer at the end of the function.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 2d5c971a59ad..4448c5f220f0 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -2957,12 +2957,9 @@ static int btusb_mtk_uhw_reg_write(struct btusb_data *data, u32 reg, u32 val)
 			      0x5E,
 			      reg >> 16, reg & 0xffff,
 			      buf, 4, USB_CTRL_SET_TIMEOUT);
-	if (err < 0) {
+	if (err < 0)
 		bt_dev_err(hdev, "Failed to write uhw reg(%d)", err);
-		goto err_free_buf;
-	}
 
-err_free_buf:
 	kfree(buf);
 
 	return err;
-- 
cgit v1.2.3-58-ga151


From 3dcb122b30643444352ff4b9aa16367dd24d7571 Mon Sep 17 00:00:00 2001
From: Chris Lu <chris.lu@mediatek.com>
Date: Thu, 4 Jul 2024 14:01:10 +0800
Subject: Bluetooth: btusb: mediatek: return error for failed reg access

Update the code to immediately return an error code if accessing a
related register fails. This ensures that our desired logic for
subsequent register operations is maintained and allows us to promptly
catch any unexpected errors.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 114 ++++++++++++++++++++++++++++++++++------------
 1 file changed, 84 insertions(+), 30 deletions(-)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 4448c5f220f0..6785d462f299 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -3041,47 +3041,98 @@ static int btusb_mtk_subsys_reset(struct hci_dev *hdev, u32 dev_id)
 	int err;
 
 	if (dev_id == 0x7922) {
-		btusb_mtk_uhw_reg_read(data, MTK_BT_SUBSYS_RST, &val);
+		err = btusb_mtk_uhw_reg_read(data, MTK_BT_SUBSYS_RST, &val);
+		if (err < 0)
+			return err;
 		val |= 0x00002020;
-		btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, val);
-		btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, 0x00010001);
-		btusb_mtk_uhw_reg_read(data, MTK_BT_SUBSYS_RST, &val);
+		err = btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, val);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, 0x00010001);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_read(data, MTK_BT_SUBSYS_RST, &val);
+		if (err < 0)
+			return err;
 		val |= BIT(0);
-		btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, val);
+		err = btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, val);
+		if (err < 0)
+			return err;
 		msleep(100);
 	} else if (dev_id == 0x7925) {
-		btusb_mtk_uhw_reg_read(data, MTK_BT_RESET_REG_CONNV3, &val);
+		err = btusb_mtk_uhw_reg_read(data, MTK_BT_RESET_REG_CONNV3, &val);
+		if (err < 0)
+			return err;
 		val |= (1 << 5);
-		btusb_mtk_uhw_reg_write(data, MTK_BT_RESET_REG_CONNV3, val);
-		btusb_mtk_uhw_reg_read(data, MTK_BT_RESET_REG_CONNV3, &val);
+		err = btusb_mtk_uhw_reg_write(data, MTK_BT_RESET_REG_CONNV3, val);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_read(data, MTK_BT_RESET_REG_CONNV3, &val);
+		if (err < 0)
+			return err;
 		val &= 0xFFFF00FF;
+		if (err < 0)
+			return err;
 		val |= (1 << 13);
-		btusb_mtk_uhw_reg_write(data, MTK_BT_RESET_REG_CONNV3, val);
-		btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, 0x00010001);
-		btusb_mtk_uhw_reg_read(data, MTK_BT_RESET_REG_CONNV3, &val);
+		err = btusb_mtk_uhw_reg_write(data, MTK_BT_RESET_REG_CONNV3, val);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, 0x00010001);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_read(data, MTK_BT_RESET_REG_CONNV3, &val);
+		if (err < 0)
+			return err;
 		val |= (1 << 0);
-		btusb_mtk_uhw_reg_write(data, MTK_BT_RESET_REG_CONNV3, val);
-		btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT, 0x000000FF);
-		btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT, &val);
-		btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT1, 0x000000FF);
-		btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT1, &val);
+		err = btusb_mtk_uhw_reg_write(data, MTK_BT_RESET_REG_CONNV3, val);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT, 0x000000FF);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT, &val);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT1, 0x000000FF);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT1, &val);
+		if (err < 0)
+			return err;
 		msleep(100);
 	} else {
 		/* It's Device EndPoint Reset Option Register */
 		bt_dev_dbg(hdev, "Initiating reset mechanism via uhw");
-		btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, MTK_EP_RST_IN_OUT_OPT);
-		btusb_mtk_uhw_reg_read(data, MTK_BT_WDT_STATUS, &val);
-
+		err = btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, MTK_EP_RST_IN_OUT_OPT);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_read(data, MTK_BT_WDT_STATUS, &val);
+		if (err < 0)
+			return err;
 		/* Reset the bluetooth chip via USB interface. */
-		btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, 1);
-		btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT, 0x000000FF);
-		btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT, &val);
-		btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT1, 0x000000FF);
-		btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT1, &val);
+		err = btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, 1);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT, 0x000000FF);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT, &val);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT1, 0x000000FF);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT1, &val);
+		if (err < 0)
+			return err;
 		/* MT7921 need to delay 20ms between toggle reset bit */
 		msleep(20);
-		btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, 0);
-		btusb_mtk_uhw_reg_read(data, MTK_BT_SUBSYS_RST, &val);
+		err = btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, 0);
+		if (err < 0)
+			return err;
+		err = btusb_mtk_uhw_reg_read(data, MTK_BT_SUBSYS_RST, &val);
+		if (err < 0)
+			return err;
 	}
 
 	err = readx_poll_timeout(btusb_mtk_reset_done, hdev, val,
@@ -3089,11 +3140,14 @@ static int btusb_mtk_subsys_reset(struct hci_dev *hdev, u32 dev_id)
 	if (err < 0)
 		bt_dev_err(hdev, "Reset timeout");
 
-	if (dev_id == 0x7922)
-		btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT, 0x000000FF);
+	if (dev_id == 0x7922) {
+		err = btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT, 0x000000FF);
+		if (err < 0)
+			return err;
+	}
 
-	btusb_mtk_id_get(data, 0x70010200, &val);
-	if (!val)
+	err = btusb_mtk_id_get(data, 0x70010200, &val);
+	if (err < 0 || !val)
 		bt_dev_err(hdev, "Can't get device id, subsys reset fail.");
 
 	return err;
-- 
cgit v1.2.3-58-ga151


From d3e6236053958a8f1c7c7a885d9cecdd383e4615 Mon Sep 17 00:00:00 2001
From: Chris Lu <chris.lu@mediatek.com>
Date: Thu, 4 Jul 2024 14:01:11 +0800
Subject: Bluetooth: btmtk: rename btmediatek_data

Rename btmediatek_data to have a consistent prefix throughout the driver.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btmtk.c | 10 +++++-----
 drivers/bluetooth/btmtk.h |  2 +-
 drivers/bluetooth/btusb.c |  9 ++++-----
 3 files changed, 10 insertions(+), 11 deletions(-)

diff --git a/drivers/bluetooth/btmtk.c b/drivers/bluetooth/btmtk.c
index a27c251bf56e..6489e4a63fb6 100644
--- a/drivers/bluetooth/btmtk.c
+++ b/drivers/bluetooth/btmtk.c
@@ -64,7 +64,7 @@ static void btmtk_coredump(struct hci_dev *hdev)
 
 static void btmtk_coredump_hdr(struct hci_dev *hdev, struct sk_buff *skb)
 {
-	struct btmediatek_data *data = hci_get_priv(hdev);
+	struct btmtk_data *data = hci_get_priv(hdev);
 	char buf[80];
 
 	snprintf(buf, sizeof(buf), "Controller Name: 0x%X\n",
@@ -85,7 +85,7 @@ static void btmtk_coredump_hdr(struct hci_dev *hdev, struct sk_buff *skb)
 
 static void btmtk_coredump_notify(struct hci_dev *hdev, int state)
 {
-	struct btmediatek_data *data = hci_get_priv(hdev);
+	struct btmtk_data *data = hci_get_priv(hdev);
 
 	switch (state) {
 	case HCI_DEVCOREDUMP_IDLE:
@@ -355,7 +355,7 @@ EXPORT_SYMBOL_GPL(btmtk_set_bdaddr);
 
 void btmtk_reset_sync(struct hci_dev *hdev)
 {
-	struct btmediatek_data *reset_work = hci_get_priv(hdev);
+	struct btmtk_data *reset_work = hci_get_priv(hdev);
 	int err;
 
 	hci_dev_lock(hdev);
@@ -371,7 +371,7 @@ EXPORT_SYMBOL_GPL(btmtk_reset_sync);
 int btmtk_register_coredump(struct hci_dev *hdev, const char *name,
 			    u32 fw_version)
 {
-	struct btmediatek_data *data = hci_get_priv(hdev);
+	struct btmtk_data *data = hci_get_priv(hdev);
 
 	if (!IS_ENABLED(CONFIG_DEV_COREDUMP))
 		return -EOPNOTSUPP;
@@ -387,7 +387,7 @@ EXPORT_SYMBOL_GPL(btmtk_register_coredump);
 
 int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb)
 {
-	struct btmediatek_data *data = hci_get_priv(hdev);
+	struct btmtk_data *data = hci_get_priv(hdev);
 	int err;
 
 	if (!IS_ENABLED(CONFIG_DEV_COREDUMP)) {
diff --git a/drivers/bluetooth/btmtk.h b/drivers/bluetooth/btmtk.h
index e76b8a358be8..dde6fbdeb2b3 100644
--- a/drivers/bluetooth/btmtk.h
+++ b/drivers/bluetooth/btmtk.h
@@ -135,7 +135,7 @@ struct btmtk_coredump_info {
 	int state;
 };
 
-struct btmediatek_data {
+struct btmtk_data {
 	u32 dev_id;
 	btmtk_reset_sync_func_t reset_sync;
 	struct btmtk_coredump_info cd_info;
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 6785d462f299..f44bcd2ba559 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -3156,7 +3156,7 @@ static int btusb_mtk_subsys_reset(struct hci_dev *hdev, u32 dev_id)
 static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
 {
 	struct btusb_data *data = hci_get_drvdata(hdev);
-	struct btmediatek_data *mtk_data;
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
 	int err;
 
 	/* It's MediaTek specific bluetooth reset mechanism via USB */
@@ -3171,9 +3171,8 @@ static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
 
 	btusb_stop_traffic(data);
 	usb_kill_anchored_urbs(&data->tx_anchor);
-	mtk_data = hci_get_priv(hdev);
 
-	err = btusb_mtk_subsys_reset(hdev, mtk_data->dev_id);
+	err = btusb_mtk_subsys_reset(hdev, btmtk_data->dev_id);
 
 	usb_queue_reset_device(data->intf);
 	clear_bit(BTUSB_HW_RESET_ACTIVE, &data->flags);
@@ -3195,7 +3194,7 @@ static int btusb_mtk_setup(struct hci_dev *hdev)
 	char fw_bin_name[64];
 	u32 fw_version = 0, fw_flavor = 0;
 	u8 param;
-	struct btmediatek_data *mediatek;
+	struct btmtk_data *mediatek;
 
 	calltime = ktime_get();
 
@@ -4423,7 +4422,7 @@ static int btusb_probe(struct usb_interface *intf,
 		data->recv_event = btusb_recv_event_realtek;
 	} else if (id->driver_info & BTUSB_MEDIATEK) {
 		/* Allocate extra space for Mediatek device */
-		priv_size += sizeof(struct btmediatek_data);
+		priv_size += sizeof(struct btmtk_data);
 	}
 
 	data->recv_acl = hci_recv_frame;
-- 
cgit v1.2.3-58-ga151


From 95f92928ad2215b5f524903e67eebd8e14f99564 Mon Sep 17 00:00:00 2001
From: Chris Lu <chris.lu@mediatek.com>
Date: Thu, 4 Jul 2024 14:01:12 +0800
Subject: Bluetooth: btusb: add callback function in btusb suspend/resume

Add suspend/resum callback function in btusb_data which are reserved
for vendor specific usage during suspend/resume. hdev->suspend will be
added before stop traffic in btusb_suspend and hdev-> resume will be
added after resubmit urb in btusb_resume.

Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index f44bcd2ba559..96f2e9d2a449 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -895,6 +895,9 @@ struct btusb_data {
 
 	int (*setup_on_usb)(struct hci_dev *hdev);
 
+	int (*suspend)(struct hci_dev *hdev);
+	int (*resume)(struct hci_dev *hdev);
+
 	int oob_wake_irq;   /* irq for out-of-band wake-on-bt */
 	unsigned cmd_timeout_cnt;
 
@@ -4769,6 +4772,9 @@ static int btusb_suspend(struct usb_interface *intf, pm_message_t message)
 
 	cancel_work_sync(&data->work);
 
+	if (data->suspend)
+		data->suspend(data->hdev);
+
 	btusb_stop_traffic(data);
 	usb_kill_anchored_urbs(&data->tx_anchor);
 
@@ -4872,6 +4878,9 @@ static int btusb_resume(struct usb_interface *intf)
 			btusb_submit_isoc_urb(hdev, GFP_NOIO);
 	}
 
+	if (data->resume)
+		data->resume(hdev);
+
 	spin_lock_irq(&data->txlock);
 	play_deferred(data);
 	clear_bit(BTUSB_SUSPENDING, &data->flags);
-- 
cgit v1.2.3-58-ga151


From d019930b0049fc2648a6b279893d8ad330596e81 Mon Sep 17 00:00:00 2001
From: Chris Lu <chris.lu@mediatek.com>
Date: Thu, 4 Jul 2024 14:01:13 +0800
Subject: Bluetooth: btmtk: move btusb_mtk_hci_wmt_sync to btmtk.c

Move btusb_mtk_hci_wmt_sync from btusb.c to btmtk.c which holds
vendor specific stuff and would make btusb.c clean.

Add usb.h header to btmtksdio.c/btmtkuart.c for usb related element
defined in btmtk.h

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btmtk.c     | 265 ++++++++++++++++++++++++++++++++++++++
 drivers/bluetooth/btmtk.h     |  31 +++++
 drivers/bluetooth/btmtksdio.c |   1 +
 drivers/bluetooth/btmtkuart.c |   1 +
 drivers/bluetooth/btusb.c     | 290 +-----------------------------------------
 5 files changed, 305 insertions(+), 283 deletions(-)

diff --git a/drivers/bluetooth/btmtk.c b/drivers/bluetooth/btmtk.c
index 6489e4a63fb6..3443a0837a86 100644
--- a/drivers/bluetooth/btmtk.c
+++ b/drivers/bluetooth/btmtk.c
@@ -4,6 +4,7 @@
  */
 #include <linux/module.h>
 #include <linux/firmware.h>
+#include <linux/usb.h>
 
 #include <net/bluetooth/bluetooth.h>
 #include <net/bluetooth/hci_core.h>
@@ -431,6 +432,270 @@ int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb)
 }
 EXPORT_SYMBOL_GPL(btmtk_process_coredump);
 
+static void btmtk_usb_wmt_recv(struct urb *urb)
+{
+	struct hci_dev *hdev = urb->context;
+	struct btmtk_data *data = hci_get_priv(hdev);
+	struct sk_buff *skb;
+	int err;
+
+	if (urb->status == 0 && urb->actual_length > 0) {
+		hdev->stat.byte_rx += urb->actual_length;
+
+		/* WMT event shouldn't be fragmented and the size should be
+		 * less than HCI_WMT_MAX_EVENT_SIZE.
+		 */
+		skb = bt_skb_alloc(HCI_WMT_MAX_EVENT_SIZE, GFP_ATOMIC);
+		if (!skb) {
+			hdev->stat.err_rx++;
+			kfree(urb->setup_packet);
+			return;
+		}
+
+		hci_skb_pkt_type(skb) = HCI_EVENT_PKT;
+		skb_put_data(skb, urb->transfer_buffer, urb->actual_length);
+
+		/* When someone waits for the WMT event, the skb is being cloned
+		 * and being processed the events from there then.
+		 */
+		if (test_bit(BTMTK_TX_WAIT_VND_EVT, &data->flags)) {
+			data->evt_skb = skb_clone(skb, GFP_ATOMIC);
+			if (!data->evt_skb) {
+				kfree_skb(skb);
+				kfree(urb->setup_packet);
+				return;
+			}
+		}
+
+		err = hci_recv_frame(hdev, skb);
+		if (err < 0) {
+			kfree_skb(data->evt_skb);
+			data->evt_skb = NULL;
+			kfree(urb->setup_packet);
+			return;
+		}
+
+		if (test_and_clear_bit(BTMTK_TX_WAIT_VND_EVT,
+				       &data->flags)) {
+			/* Barrier to sync with other CPUs */
+			smp_mb__after_atomic();
+			wake_up_bit(&data->flags,
+				    BTMTK_TX_WAIT_VND_EVT);
+		}
+		kfree(urb->setup_packet);
+		return;
+	} else if (urb->status == -ENOENT) {
+		/* Avoid suspend failed when usb_kill_urb */
+		return;
+	}
+
+	usb_mark_last_busy(data->udev);
+
+	/* The URB complete handler is still called with urb->actual_length = 0
+	 * when the event is not available, so we should keep re-submitting
+	 * URB until WMT event returns, Also, It's necessary to wait some time
+	 * between the two consecutive control URBs to relax the target device
+	 * to generate the event. Otherwise, the WMT event cannot return from
+	 * the device successfully.
+	 */
+	udelay(500);
+
+	usb_anchor_urb(urb, data->ctrl_anchor);
+	err = usb_submit_urb(urb, GFP_ATOMIC);
+	if (err < 0) {
+		kfree(urb->setup_packet);
+		/* -EPERM: urb is being killed;
+		 * -ENODEV: device got disconnected
+		 */
+		if (err != -EPERM && err != -ENODEV)
+			bt_dev_err(hdev, "urb %p failed to resubmit (%d)",
+				   urb, -err);
+		usb_unanchor_urb(urb);
+	}
+}
+
+static int btmtk_usb_submit_wmt_recv_urb(struct hci_dev *hdev)
+{
+	struct btmtk_data *data = hci_get_priv(hdev);
+	struct usb_ctrlrequest *dr;
+	unsigned char *buf;
+	int err, size = 64;
+	unsigned int pipe;
+	struct urb *urb;
+
+	urb = usb_alloc_urb(0, GFP_KERNEL);
+	if (!urb)
+		return -ENOMEM;
+
+	dr = kmalloc(sizeof(*dr), GFP_KERNEL);
+	if (!dr) {
+		usb_free_urb(urb);
+		return -ENOMEM;
+	}
+
+	dr->bRequestType = USB_TYPE_VENDOR | USB_DIR_IN;
+	dr->bRequest     = 1;
+	dr->wIndex       = cpu_to_le16(0);
+	dr->wValue       = cpu_to_le16(48);
+	dr->wLength      = cpu_to_le16(size);
+
+	buf = kmalloc(size, GFP_KERNEL);
+	if (!buf) {
+		kfree(dr);
+		usb_free_urb(urb);
+		return -ENOMEM;
+	}
+
+	pipe = usb_rcvctrlpipe(data->udev, 0);
+
+	usb_fill_control_urb(urb, data->udev, pipe, (void *)dr,
+			     buf, size, btmtk_usb_wmt_recv, hdev);
+
+	urb->transfer_flags |= URB_FREE_BUFFER;
+
+	usb_anchor_urb(urb, data->ctrl_anchor);
+	err = usb_submit_urb(urb, GFP_KERNEL);
+	if (err < 0) {
+		if (err != -EPERM && err != -ENODEV)
+			bt_dev_err(hdev, "urb %p submission failed (%d)",
+				   urb, -err);
+		usb_unanchor_urb(urb);
+	}
+
+	usb_free_urb(urb);
+
+	return err;
+}
+
+int btmtk_usb_hci_wmt_sync(struct hci_dev *hdev,
+			   struct btmtk_hci_wmt_params *wmt_params)
+{
+	struct btmtk_data *data = hci_get_priv(hdev);
+	struct btmtk_hci_wmt_evt_funcc *wmt_evt_funcc;
+	u32 hlen, status = BTMTK_WMT_INVALID;
+	struct btmtk_hci_wmt_evt *wmt_evt;
+	struct btmtk_hci_wmt_cmd *wc;
+	struct btmtk_wmt_hdr *hdr;
+	int err;
+
+	/* Send the WMT command and wait until the WMT event returns */
+	hlen = sizeof(*hdr) + wmt_params->dlen;
+	if (hlen > 255)
+		return -EINVAL;
+
+	wc = kzalloc(hlen, GFP_KERNEL);
+	if (!wc)
+		return -ENOMEM;
+
+	hdr = &wc->hdr;
+	hdr->dir = 1;
+	hdr->op = wmt_params->op;
+	hdr->dlen = cpu_to_le16(wmt_params->dlen + 1);
+	hdr->flag = wmt_params->flag;
+	memcpy(wc->data, wmt_params->data, wmt_params->dlen);
+
+	set_bit(BTMTK_TX_WAIT_VND_EVT, &data->flags);
+
+	/* WMT cmd/event doesn't follow up the generic HCI cmd/event handling,
+	 * it needs constantly polling control pipe until the host received the
+	 * WMT event, thus, we should require to specifically acquire PM counter
+	 * on the USB to prevent the interface from entering auto suspended
+	 * while WMT cmd/event in progress.
+	 */
+	err = usb_autopm_get_interface(data->intf);
+	if (err < 0)
+		goto err_free_wc;
+
+	err = __hci_cmd_send(hdev, 0xfc6f, hlen, wc);
+
+	if (err < 0) {
+		clear_bit(BTMTK_TX_WAIT_VND_EVT, &data->flags);
+		usb_autopm_put_interface(data->intf);
+		goto err_free_wc;
+	}
+
+	/* Submit control IN URB on demand to process the WMT event */
+	err = btmtk_usb_submit_wmt_recv_urb(hdev);
+
+	usb_autopm_put_interface(data->intf);
+
+	if (err < 0)
+		goto err_free_wc;
+
+	/* The vendor specific WMT commands are all answered by a vendor
+	 * specific event and will have the Command Status or Command
+	 * Complete as with usual HCI command flow control.
+	 *
+	 * After sending the command, wait for BTUSB_TX_WAIT_VND_EVT
+	 * state to be cleared. The driver specific event receive routine
+	 * will clear that state and with that indicate completion of the
+	 * WMT command.
+	 */
+	err = wait_on_bit_timeout(&data->flags, BTMTK_TX_WAIT_VND_EVT,
+				  TASK_INTERRUPTIBLE, HCI_INIT_TIMEOUT);
+	if (err == -EINTR) {
+		bt_dev_err(hdev, "Execution of wmt command interrupted");
+		clear_bit(BTMTK_TX_WAIT_VND_EVT, &data->flags);
+		goto err_free_wc;
+	}
+
+	if (err) {
+		bt_dev_err(hdev, "Execution of wmt command timed out");
+		clear_bit(BTMTK_TX_WAIT_VND_EVT, &data->flags);
+		err = -ETIMEDOUT;
+		goto err_free_wc;
+	}
+
+	if (data->evt_skb == NULL)
+		goto err_free_wc;
+
+	/* Parse and handle the return WMT event */
+	wmt_evt = (struct btmtk_hci_wmt_evt *)data->evt_skb->data;
+	if (wmt_evt->whdr.op != hdr->op) {
+		bt_dev_err(hdev, "Wrong op received %d expected %d",
+			   wmt_evt->whdr.op, hdr->op);
+		err = -EIO;
+		goto err_free_skb;
+	}
+
+	switch (wmt_evt->whdr.op) {
+	case BTMTK_WMT_SEMAPHORE:
+		if (wmt_evt->whdr.flag == 2)
+			status = BTMTK_WMT_PATCH_UNDONE;
+		else
+			status = BTMTK_WMT_PATCH_DONE;
+		break;
+	case BTMTK_WMT_FUNC_CTRL:
+		wmt_evt_funcc = (struct btmtk_hci_wmt_evt_funcc *)wmt_evt;
+		if (be16_to_cpu(wmt_evt_funcc->status) == 0x404)
+			status = BTMTK_WMT_ON_DONE;
+		else if (be16_to_cpu(wmt_evt_funcc->status) == 0x420)
+			status = BTMTK_WMT_ON_PROGRESS;
+		else
+			status = BTMTK_WMT_ON_UNDONE;
+		break;
+	case BTMTK_WMT_PATCH_DWNLD:
+		if (wmt_evt->whdr.flag == 2)
+			status = BTMTK_WMT_PATCH_DONE;
+		else if (wmt_evt->whdr.flag == 1)
+			status = BTMTK_WMT_PATCH_PROGRESS;
+		else
+			status = BTMTK_WMT_PATCH_UNDONE;
+		break;
+	}
+
+	if (wmt_params->status)
+		*wmt_params->status = status;
+
+err_free_skb:
+	kfree_skb(data->evt_skb);
+	data->evt_skb = NULL;
+err_free_wc:
+	kfree(wc);
+	return err;
+}
+EXPORT_SYMBOL_GPL(btmtk_usb_hci_wmt_sync);
+
 MODULE_AUTHOR("Sean Wang <sean.wang@mediatek.com>");
 MODULE_AUTHOR("Mark Chen <mark-yw.chen@mediatek.com>");
 MODULE_DESCRIPTION("Bluetooth support for MediaTek devices ver " VERSION);
diff --git a/drivers/bluetooth/btmtk.h b/drivers/bluetooth/btmtk.h
index dde6fbdeb2b3..3055b9728ae2 100644
--- a/drivers/bluetooth/btmtk.h
+++ b/drivers/bluetooth/btmtk.h
@@ -28,6 +28,18 @@
 #define MTK_COREDUMP_END_LEN		(sizeof(MTK_COREDUMP_END))
 #define MTK_COREDUMP_NUM		255
 
+/* UHW CR mapping */
+#define MTK_BT_MISC		0x70002510
+#define MTK_BT_SUBSYS_RST	0x70002610
+#define MTK_UDMA_INT_STA_BT	0x74000024
+#define MTK_UDMA_INT_STA_BT1	0x74000308
+#define MTK_BT_WDT_STATUS	0x740003A0
+#define MTK_EP_RST_OPT		0x74011890
+#define MTK_EP_RST_IN_OUT_OPT	0x00010001
+#define MTK_BT_RST_DONE		0x00000100
+#define MTK_BT_RESET_REG_CONNV3	0x70028610
+#define MTK_BT_READ_DEV_ID	0x70010200
+
 enum {
 	BTMTK_WMT_PATCH_DWNLD = 0x1,
 	BTMTK_WMT_TEST = 0x2,
@@ -126,6 +138,10 @@ struct btmtk_hci_wmt_params {
 	u32 *status;
 };
 
+enum {
+	BTMTK_TX_WAIT_VND_EVT,
+};
+
 typedef int (*btmtk_reset_sync_func_t)(struct hci_dev *, void *);
 
 struct btmtk_coredump_info {
@@ -136,9 +152,15 @@ struct btmtk_coredump_info {
 };
 
 struct btmtk_data {
+	unsigned long flags;
 	u32 dev_id;
 	btmtk_reset_sync_func_t reset_sync;
 	struct btmtk_coredump_info cd_info;
+
+	struct usb_device *udev;
+	struct usb_interface *intf;
+	struct usb_anchor *ctrl_anchor;
+	struct sk_buff *evt_skb;
 };
 
 typedef int (*wmt_cmd_sync_func_t)(struct hci_dev *,
@@ -163,6 +185,9 @@ int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb);
 
 void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id, u32 fw_ver,
 			   u32 fw_flavor);
+
+int btmtk_usb_hci_wmt_sync(struct hci_dev *hdev,
+			   struct btmtk_hci_wmt_params *wmt_params);
 #else
 
 static inline int btmtk_set_bdaddr(struct hci_dev *hdev,
@@ -202,4 +227,10 @@ static void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id,
 				  u32 fw_ver, u32 fw_flavor)
 {
 }
+
+static int btmtk_usb_hci_wmt_sync(struct hci_dev *hdev,
+				  struct btmtk_hci_wmt_params *wmt_params)
+{
+	return -EOPNOTSUPP;
+}
 #endif
diff --git a/drivers/bluetooth/btmtksdio.c b/drivers/bluetooth/btmtksdio.c
index b4e560a17aa6..39d6898497a4 100644
--- a/drivers/bluetooth/btmtksdio.c
+++ b/drivers/bluetooth/btmtksdio.c
@@ -20,6 +20,7 @@
 #include <linux/of.h>
 #include <linux/pm_runtime.h>
 #include <linux/skbuff.h>
+#include <linux/usb.h>
 
 #include <linux/mmc/host.h>
 #include <linux/mmc/sdio_ids.h>
diff --git a/drivers/bluetooth/btmtkuart.c b/drivers/bluetooth/btmtkuart.c
index e6bc4a73c9fc..aa87c3e78871 100644
--- a/drivers/bluetooth/btmtkuart.c
+++ b/drivers/bluetooth/btmtkuart.c
@@ -22,6 +22,7 @@
 #include <linux/regulator/consumer.h>
 #include <linux/serdev.h>
 #include <linux/skbuff.h>
+#include <linux/usb.h>
 
 #include <net/bluetooth/bluetooth.h>
 #include <net/bluetooth/hci_core.h>
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 96f2e9d2a449..0376d141d8fe 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -2645,282 +2645,6 @@ static int btusb_recv_event_realtek(struct hci_dev *hdev, struct sk_buff *skb)
 	return hci_recv_frame(hdev, skb);
 }
 
-/* UHW CR mapping */
-#define MTK_BT_MISC		0x70002510
-#define MTK_BT_SUBSYS_RST	0x70002610
-#define MTK_UDMA_INT_STA_BT	0x74000024
-#define MTK_UDMA_INT_STA_BT1	0x74000308
-#define MTK_BT_WDT_STATUS	0x740003A0
-#define MTK_EP_RST_OPT		0x74011890
-#define MTK_EP_RST_IN_OUT_OPT	0x00010001
-#define MTK_BT_RST_DONE		0x00000100
-#define MTK_BT_RESET_REG_CONNV3	0x70028610
-#define MTK_BT_READ_DEV_ID	0x70010200
-
-
-static void btusb_mtk_wmt_recv(struct urb *urb)
-{
-	struct hci_dev *hdev = urb->context;
-	struct btusb_data *data = hci_get_drvdata(hdev);
-	struct sk_buff *skb;
-	int err;
-
-	if (urb->status == 0 && urb->actual_length > 0) {
-		hdev->stat.byte_rx += urb->actual_length;
-
-		/* WMT event shouldn't be fragmented and the size should be
-		 * less than HCI_WMT_MAX_EVENT_SIZE.
-		 */
-		skb = bt_skb_alloc(HCI_WMT_MAX_EVENT_SIZE, GFP_ATOMIC);
-		if (!skb) {
-			hdev->stat.err_rx++;
-			kfree(urb->setup_packet);
-			return;
-		}
-
-		hci_skb_pkt_type(skb) = HCI_EVENT_PKT;
-		skb_put_data(skb, urb->transfer_buffer, urb->actual_length);
-
-		/* When someone waits for the WMT event, the skb is being cloned
-		 * and being processed the events from there then.
-		 */
-		if (test_bit(BTUSB_TX_WAIT_VND_EVT, &data->flags)) {
-			data->evt_skb = skb_clone(skb, GFP_ATOMIC);
-			if (!data->evt_skb) {
-				kfree_skb(skb);
-				kfree(urb->setup_packet);
-				return;
-			}
-		}
-
-		err = hci_recv_frame(hdev, skb);
-		if (err < 0) {
-			kfree_skb(data->evt_skb);
-			data->evt_skb = NULL;
-			kfree(urb->setup_packet);
-			return;
-		}
-
-		if (test_and_clear_bit(BTUSB_TX_WAIT_VND_EVT,
-				       &data->flags)) {
-			/* Barrier to sync with other CPUs */
-			smp_mb__after_atomic();
-			wake_up_bit(&data->flags,
-				    BTUSB_TX_WAIT_VND_EVT);
-		}
-		kfree(urb->setup_packet);
-		return;
-	} else if (urb->status == -ENOENT) {
-		/* Avoid suspend failed when usb_kill_urb */
-		return;
-	}
-
-	usb_mark_last_busy(data->udev);
-
-	/* The URB complete handler is still called with urb->actual_length = 0
-	 * when the event is not available, so we should keep re-submitting
-	 * URB until WMT event returns, Also, It's necessary to wait some time
-	 * between the two consecutive control URBs to relax the target device
-	 * to generate the event. Otherwise, the WMT event cannot return from
-	 * the device successfully.
-	 */
-	udelay(500);
-
-	usb_anchor_urb(urb, &data->ctrl_anchor);
-	err = usb_submit_urb(urb, GFP_ATOMIC);
-	if (err < 0) {
-		kfree(urb->setup_packet);
-		/* -EPERM: urb is being killed;
-		 * -ENODEV: device got disconnected
-		 */
-		if (err != -EPERM && err != -ENODEV)
-			bt_dev_err(hdev, "urb %p failed to resubmit (%d)",
-				   urb, -err);
-		usb_unanchor_urb(urb);
-	}
-}
-
-static int btusb_mtk_submit_wmt_recv_urb(struct hci_dev *hdev)
-{
-	struct btusb_data *data = hci_get_drvdata(hdev);
-	struct usb_ctrlrequest *dr;
-	unsigned char *buf;
-	int err, size = 64;
-	unsigned int pipe;
-	struct urb *urb;
-
-	urb = usb_alloc_urb(0, GFP_KERNEL);
-	if (!urb)
-		return -ENOMEM;
-
-	dr = kmalloc(sizeof(*dr), GFP_KERNEL);
-	if (!dr) {
-		usb_free_urb(urb);
-		return -ENOMEM;
-	}
-
-	dr->bRequestType = USB_TYPE_VENDOR | USB_DIR_IN;
-	dr->bRequest     = 1;
-	dr->wIndex       = cpu_to_le16(0);
-	dr->wValue       = cpu_to_le16(48);
-	dr->wLength      = cpu_to_le16(size);
-
-	buf = kmalloc(size, GFP_KERNEL);
-	if (!buf) {
-		kfree(dr);
-		usb_free_urb(urb);
-		return -ENOMEM;
-	}
-
-	pipe = usb_rcvctrlpipe(data->udev, 0);
-
-	usb_fill_control_urb(urb, data->udev, pipe, (void *)dr,
-			     buf, size, btusb_mtk_wmt_recv, hdev);
-
-	urb->transfer_flags |= URB_FREE_BUFFER;
-
-	usb_anchor_urb(urb, &data->ctrl_anchor);
-	err = usb_submit_urb(urb, GFP_KERNEL);
-	if (err < 0) {
-		if (err != -EPERM && err != -ENODEV)
-			bt_dev_err(hdev, "urb %p submission failed (%d)",
-				   urb, -err);
-		usb_unanchor_urb(urb);
-	}
-
-	usb_free_urb(urb);
-
-	return err;
-}
-
-static int btusb_mtk_hci_wmt_sync(struct hci_dev *hdev,
-				  struct btmtk_hci_wmt_params *wmt_params)
-{
-	struct btusb_data *data = hci_get_drvdata(hdev);
-	struct btmtk_hci_wmt_evt_funcc *wmt_evt_funcc;
-	u32 hlen, status = BTMTK_WMT_INVALID;
-	struct btmtk_hci_wmt_evt *wmt_evt;
-	struct btmtk_hci_wmt_cmd *wc;
-	struct btmtk_wmt_hdr *hdr;
-	int err;
-
-	/* Send the WMT command and wait until the WMT event returns */
-	hlen = sizeof(*hdr) + wmt_params->dlen;
-	if (hlen > 255)
-		return -EINVAL;
-
-	wc = kzalloc(hlen, GFP_KERNEL);
-	if (!wc)
-		return -ENOMEM;
-
-	hdr = &wc->hdr;
-	hdr->dir = 1;
-	hdr->op = wmt_params->op;
-	hdr->dlen = cpu_to_le16(wmt_params->dlen + 1);
-	hdr->flag = wmt_params->flag;
-	memcpy(wc->data, wmt_params->data, wmt_params->dlen);
-
-	set_bit(BTUSB_TX_WAIT_VND_EVT, &data->flags);
-
-	/* WMT cmd/event doesn't follow up the generic HCI cmd/event handling,
-	 * it needs constantly polling control pipe until the host received the
-	 * WMT event, thus, we should require to specifically acquire PM counter
-	 * on the USB to prevent the interface from entering auto suspended
-	 * while WMT cmd/event in progress.
-	 */
-	err = usb_autopm_get_interface(data->intf);
-	if (err < 0)
-		goto err_free_wc;
-
-	err = __hci_cmd_send(hdev, 0xfc6f, hlen, wc);
-
-	if (err < 0) {
-		clear_bit(BTUSB_TX_WAIT_VND_EVT, &data->flags);
-		usb_autopm_put_interface(data->intf);
-		goto err_free_wc;
-	}
-
-	/* Submit control IN URB on demand to process the WMT event */
-	err = btusb_mtk_submit_wmt_recv_urb(hdev);
-
-	usb_autopm_put_interface(data->intf);
-
-	if (err < 0)
-		goto err_free_wc;
-
-	/* The vendor specific WMT commands are all answered by a vendor
-	 * specific event and will have the Command Status or Command
-	 * Complete as with usual HCI command flow control.
-	 *
-	 * After sending the command, wait for BTUSB_TX_WAIT_VND_EVT
-	 * state to be cleared. The driver specific event receive routine
-	 * will clear that state and with that indicate completion of the
-	 * WMT command.
-	 */
-	err = wait_on_bit_timeout(&data->flags, BTUSB_TX_WAIT_VND_EVT,
-				  TASK_INTERRUPTIBLE, HCI_INIT_TIMEOUT);
-	if (err == -EINTR) {
-		bt_dev_err(hdev, "Execution of wmt command interrupted");
-		clear_bit(BTUSB_TX_WAIT_VND_EVT, &data->flags);
-		goto err_free_wc;
-	}
-
-	if (err) {
-		bt_dev_err(hdev, "Execution of wmt command timed out");
-		clear_bit(BTUSB_TX_WAIT_VND_EVT, &data->flags);
-		err = -ETIMEDOUT;
-		goto err_free_wc;
-	}
-
-	if (data->evt_skb == NULL)
-		goto err_free_wc;
-
-	/* Parse and handle the return WMT event */
-	wmt_evt = (struct btmtk_hci_wmt_evt *)data->evt_skb->data;
-	if (wmt_evt->whdr.op != hdr->op) {
-		bt_dev_err(hdev, "Wrong op received %d expected %d",
-			   wmt_evt->whdr.op, hdr->op);
-		err = -EIO;
-		goto err_free_skb;
-	}
-
-	switch (wmt_evt->whdr.op) {
-	case BTMTK_WMT_SEMAPHORE:
-		if (wmt_evt->whdr.flag == 2)
-			status = BTMTK_WMT_PATCH_UNDONE;
-		else
-			status = BTMTK_WMT_PATCH_DONE;
-		break;
-	case BTMTK_WMT_FUNC_CTRL:
-		wmt_evt_funcc = (struct btmtk_hci_wmt_evt_funcc *)wmt_evt;
-		if (be16_to_cpu(wmt_evt_funcc->status) == 0x404)
-			status = BTMTK_WMT_ON_DONE;
-		else if (be16_to_cpu(wmt_evt_funcc->status) == 0x420)
-			status = BTMTK_WMT_ON_PROGRESS;
-		else
-			status = BTMTK_WMT_ON_UNDONE;
-		break;
-	case BTMTK_WMT_PATCH_DWNLD:
-		if (wmt_evt->whdr.flag == 2)
-			status = BTMTK_WMT_PATCH_DONE;
-		else if (wmt_evt->whdr.flag == 1)
-			status = BTMTK_WMT_PATCH_PROGRESS;
-		else
-			status = BTMTK_WMT_PATCH_UNDONE;
-		break;
-	}
-
-	if (wmt_params->status)
-		*wmt_params->status = status;
-
-err_free_skb:
-	kfree_skb(data->evt_skb);
-	data->evt_skb = NULL;
-err_free_wc:
-	kfree(wc);
-	return err;
-}
-
 static int btusb_mtk_func_query(struct hci_dev *hdev)
 {
 	struct btmtk_hci_wmt_params wmt_params;
@@ -2934,7 +2658,7 @@ static int btusb_mtk_func_query(struct hci_dev *hdev)
 	wmt_params.data = &param;
 	wmt_params.status = &status;
 
-	err = btusb_mtk_hci_wmt_sync(hdev, &wmt_params);
+	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
 	if (err < 0) {
 		bt_dev_err(hdev, "Failed to query function status (%d)", err);
 		return err;
@@ -3255,7 +2979,7 @@ static int btusb_mtk_setup(struct hci_dev *hdev)
 				      fw_version, fw_flavor);
 
 		err = btmtk_setup_firmware_79xx(hdev, fw_bin_name,
-						btusb_mtk_hci_wmt_sync);
+						btmtk_usb_hci_wmt_sync);
 		if (err < 0) {
 			bt_dev_err(hdev, "Failed to set up firmware (%d)", err);
 			clear_bit(BTUSB_FIRMWARE_LOADED, &data->flags);
@@ -3275,7 +2999,7 @@ static int btusb_mtk_setup(struct hci_dev *hdev)
 		wmt_params.data = &param;
 		wmt_params.status = NULL;
 
-		err = btusb_mtk_hci_wmt_sync(hdev, &wmt_params);
+		err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
 		if (err < 0) {
 			bt_dev_err(hdev, "Failed to send wmt func ctrl (%d)", err);
 			return err;
@@ -3297,7 +3021,7 @@ static int btusb_mtk_setup(struct hci_dev *hdev)
 	wmt_params.data = NULL;
 	wmt_params.status = &status;
 
-	err = btusb_mtk_hci_wmt_sync(hdev, &wmt_params);
+	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
 	if (err < 0) {
 		bt_dev_err(hdev, "Failed to query firmware status (%d)", err);
 		return err;
@@ -3310,7 +3034,7 @@ static int btusb_mtk_setup(struct hci_dev *hdev)
 
 	/* Setup a firmware which the device definitely requires */
 	err = btmtk_setup_firmware(hdev, fwname,
-				   btusb_mtk_hci_wmt_sync);
+				   btmtk_usb_hci_wmt_sync);
 	if (err < 0)
 		return err;
 
@@ -3339,7 +3063,7 @@ ignore_setup_fw:
 	wmt_params.data = &param;
 	wmt_params.status = NULL;
 
-	err = btusb_mtk_hci_wmt_sync(hdev, &wmt_params);
+	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
 	if (err < 0) {
 		bt_dev_err(hdev, "Failed to send wmt func ctrl (%d)", err);
 		return err;
@@ -3385,7 +3109,7 @@ static int btusb_mtk_shutdown(struct hci_dev *hdev)
 	wmt_params.data = &param;
 	wmt_params.status = NULL;
 
-	err = btusb_mtk_hci_wmt_sync(hdev, &wmt_params);
+	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
 	if (err < 0) {
 		bt_dev_err(hdev, "Failed to send wmt func ctrl (%d)", err);
 		return err;
-- 
cgit v1.2.3-58-ga151


From 5c5e8c52e3cafaa6f71efd44a42a674271e5b3ad Mon Sep 17 00:00:00 2001
From: Chris Lu <chris.lu@mediatek.com>
Date: Thu, 4 Jul 2024 14:01:14 +0800
Subject: Bluetooth: btmtk: move btusb_mtk_[setup, shutdown] to btmtk.c

Move btusb_mtk_[setup, shutdown] and related function from
btusb.c to btmtk.c which holds vendor specific stuff and
would make btusb.c clean.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btmtk.c | 454 +++++++++++++++++++++++++++++++++++++++++++++-
 drivers/bluetooth/btmtk.h |  23 ++-
 drivers/bluetooth/btusb.c | 452 ++-------------------------------------------
 3 files changed, 484 insertions(+), 445 deletions(-)

diff --git a/drivers/bluetooth/btmtk.c b/drivers/bluetooth/btmtk.c
index 3443a0837a86..f5ecb8227d3b 100644
--- a/drivers/bluetooth/btmtk.c
+++ b/drivers/bluetooth/btmtk.c
@@ -5,6 +5,8 @@
 #include <linux/module.h>
 #include <linux/firmware.h>
 #include <linux/usb.h>
+#include <linux/iopoll.h>
+#include <asm/unaligned.h>
 
 #include <net/bluetooth/bluetooth.h>
 #include <net/bluetooth/hci_core.h>
@@ -567,8 +569,8 @@ static int btmtk_usb_submit_wmt_recv_urb(struct hci_dev *hdev)
 	return err;
 }
 
-int btmtk_usb_hci_wmt_sync(struct hci_dev *hdev,
-			   struct btmtk_hci_wmt_params *wmt_params)
+static int btmtk_usb_hci_wmt_sync(struct hci_dev *hdev,
+				  struct btmtk_hci_wmt_params *wmt_params)
 {
 	struct btmtk_data *data = hci_get_priv(hdev);
 	struct btmtk_hci_wmt_evt_funcc *wmt_evt_funcc;
@@ -694,7 +696,453 @@ err_free_wc:
 	kfree(wc);
 	return err;
 }
-EXPORT_SYMBOL_GPL(btmtk_usb_hci_wmt_sync);
+
+static int btmtk_usb_func_query(struct hci_dev *hdev)
+{
+	struct btmtk_hci_wmt_params wmt_params;
+	int status, err;
+	u8 param = 0;
+
+	/* Query whether the function is enabled */
+	wmt_params.op = BTMTK_WMT_FUNC_CTRL;
+	wmt_params.flag = 4;
+	wmt_params.dlen = sizeof(param);
+	wmt_params.data = &param;
+	wmt_params.status = &status;
+
+	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
+	if (err < 0) {
+		bt_dev_err(hdev, "Failed to query function status (%d)", err);
+		return err;
+	}
+
+	return status;
+}
+
+static int btmtk_usb_uhw_reg_write(struct hci_dev *hdev, u32 reg, u32 val)
+{
+	struct btmtk_data *data = hci_get_priv(hdev);
+	int pipe, err;
+	void *buf;
+
+	buf = kzalloc(4, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	put_unaligned_le32(val, buf);
+
+	pipe = usb_sndctrlpipe(data->udev, 0);
+	err = usb_control_msg(data->udev, pipe, 0x02,
+			      0x5E,
+			      reg >> 16, reg & 0xffff,
+			      buf, 4, USB_CTRL_SET_TIMEOUT);
+	if (err < 0)
+		bt_dev_err(hdev, "Failed to write uhw reg(%d)", err);
+
+	kfree(buf);
+
+	return err;
+}
+
+static int btmtk_usb_uhw_reg_read(struct hci_dev *hdev, u32 reg, u32 *val)
+{
+	struct btmtk_data *data = hci_get_priv(hdev);
+	int pipe, err;
+	void *buf;
+
+	buf = kzalloc(4, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	pipe = usb_rcvctrlpipe(data->udev, 0);
+	err = usb_control_msg(data->udev, pipe, 0x01,
+			      0xDE,
+			      reg >> 16, reg & 0xffff,
+			      buf, 4, USB_CTRL_GET_TIMEOUT);
+	if (err < 0) {
+		bt_dev_err(hdev, "Failed to read uhw reg(%d)", err);
+		goto err_free_buf;
+	}
+
+	*val = get_unaligned_le32(buf);
+	bt_dev_dbg(hdev, "reg=%x, value=0x%08x", reg, *val);
+
+err_free_buf:
+	kfree(buf);
+
+	return err;
+}
+
+static int btmtk_usb_reg_read(struct hci_dev *hdev, u32 reg, u32 *val)
+{
+	struct btmtk_data *data = hci_get_priv(hdev);
+	int pipe, err, size = sizeof(u32);
+	void *buf;
+
+	buf = kzalloc(size, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	pipe = usb_rcvctrlpipe(data->udev, 0);
+	err = usb_control_msg(data->udev, pipe, 0x63,
+			      USB_TYPE_VENDOR | USB_DIR_IN,
+			      reg >> 16, reg & 0xffff,
+			      buf, size, USB_CTRL_GET_TIMEOUT);
+	if (err < 0)
+		goto err_free_buf;
+
+	*val = get_unaligned_le32(buf);
+
+err_free_buf:
+	kfree(buf);
+
+	return err;
+}
+
+static int btmtk_usb_id_get(struct hci_dev *hdev, u32 reg, u32 *id)
+{
+	return btmtk_usb_reg_read(hdev, reg, id);
+}
+
+static u32 btmtk_usb_reset_done(struct hci_dev *hdev)
+{
+	u32 val = 0;
+
+	btmtk_usb_uhw_reg_read(hdev, MTK_BT_MISC, &val);
+
+	return val & MTK_BT_RST_DONE;
+}
+
+int btmtk_usb_subsys_reset(struct hci_dev *hdev, u32 dev_id)
+{
+	u32 val;
+	int err;
+
+	if (dev_id == 0x7922) {
+		err = btmtk_usb_uhw_reg_read(hdev, MTK_BT_SUBSYS_RST, &val);
+		if (err < 0)
+			return err;
+		val |= 0x00002020;
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_BT_SUBSYS_RST, val);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_EP_RST_OPT, 0x00010001);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_read(hdev, MTK_BT_SUBSYS_RST, &val);
+		if (err < 0)
+			return err;
+		val |= BIT(0);
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_BT_SUBSYS_RST, val);
+		if (err < 0)
+			return err;
+		msleep(100);
+	} else if (dev_id == 0x7925) {
+		err = btmtk_usb_uhw_reg_read(hdev, MTK_BT_RESET_REG_CONNV3, &val);
+		if (err < 0)
+			return err;
+		val |= (1 << 5);
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_BT_RESET_REG_CONNV3, val);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_read(hdev, MTK_BT_RESET_REG_CONNV3, &val);
+		if (err < 0)
+			return err;
+		val &= 0xFFFF00FF;
+		val |= (1 << 13);
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_BT_RESET_REG_CONNV3, val);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_EP_RST_OPT, 0x00010001);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_read(hdev, MTK_BT_RESET_REG_CONNV3, &val);
+		if (err < 0)
+			return err;
+		val |= (1 << 0);
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_BT_RESET_REG_CONNV3, val);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_UDMA_INT_STA_BT, 0x000000FF);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_read(hdev, MTK_UDMA_INT_STA_BT, &val);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_UDMA_INT_STA_BT1, 0x000000FF);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_read(hdev, MTK_UDMA_INT_STA_BT1, &val);
+		if (err < 0)
+			return err;
+		msleep(100);
+	} else {
+		/* It's Device EndPoint Reset Option Register */
+		bt_dev_dbg(hdev, "Initiating reset mechanism via uhw");
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_EP_RST_OPT, MTK_EP_RST_IN_OUT_OPT);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_read(hdev, MTK_BT_WDT_STATUS, &val);
+		if (err < 0)
+			return err;
+		/* Reset the bluetooth chip via USB interface. */
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_BT_SUBSYS_RST, 1);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_UDMA_INT_STA_BT, 0x000000FF);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_read(hdev, MTK_UDMA_INT_STA_BT, &val);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_UDMA_INT_STA_BT1, 0x000000FF);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_read(hdev, MTK_UDMA_INT_STA_BT1, &val);
+		if (err < 0)
+			return err;
+		/* MT7921 need to delay 20ms between toggle reset bit */
+		msleep(20);
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_BT_SUBSYS_RST, 0);
+		if (err < 0)
+			return err;
+		err = btmtk_usb_uhw_reg_read(hdev, MTK_BT_SUBSYS_RST, &val);
+		if (err < 0)
+			return err;
+	}
+
+	err = readx_poll_timeout(btmtk_usb_reset_done, hdev, val,
+				 val & MTK_BT_RST_DONE, 20000, 1000000);
+	if (err < 0)
+		bt_dev_err(hdev, "Reset timeout");
+
+	if (dev_id == 0x7922) {
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_UDMA_INT_STA_BT, 0x000000FF);
+		if (err < 0)
+			return err;
+	}
+
+	err = btmtk_usb_id_get(hdev, 0x70010200, &val);
+	if (err < 0 || !val)
+		bt_dev_err(hdev, "Can't get device id, subsys reset fail.");
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(btmtk_usb_subsys_reset);
+
+int btmtk_usb_setup(struct hci_dev *hdev)
+{
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
+	struct btmtk_hci_wmt_params wmt_params;
+	ktime_t calltime, delta, rettime;
+	struct btmtk_tci_sleep tci_sleep;
+	unsigned long long duration;
+	struct sk_buff *skb;
+	const char *fwname;
+	int err, status;
+	u32 dev_id = 0;
+	char fw_bin_name[64];
+	u32 fw_version = 0, fw_flavor = 0;
+	u8 param;
+
+	calltime = ktime_get();
+
+	err = btmtk_usb_id_get(hdev, 0x80000008, &dev_id);
+	if (err < 0) {
+		bt_dev_err(hdev, "Failed to get device id (%d)", err);
+		return err;
+	}
+
+	if (!dev_id || dev_id != 0x7663) {
+		err = btmtk_usb_id_get(hdev, 0x70010200, &dev_id);
+		if (err < 0) {
+			bt_dev_err(hdev, "Failed to get device id (%d)", err);
+			return err;
+		}
+		err = btmtk_usb_id_get(hdev, 0x80021004, &fw_version);
+		if (err < 0) {
+			bt_dev_err(hdev, "Failed to get fw version (%d)", err);
+			return err;
+		}
+		err = btmtk_usb_id_get(hdev, 0x70010020, &fw_flavor);
+		if (err < 0) {
+			bt_dev_err(hdev, "Failed to get fw flavor (%d)", err);
+			return err;
+		}
+		fw_flavor = (fw_flavor & 0x00000080) >> 7;
+	}
+
+	btmtk_data->dev_id = dev_id;
+
+	err = btmtk_register_coredump(hdev, btmtk_data->drv_name, fw_version);
+	if (err < 0)
+		bt_dev_err(hdev, "Failed to register coredump (%d)", err);
+
+	switch (dev_id) {
+	case 0x7663:
+		fwname = FIRMWARE_MT7663;
+		break;
+	case 0x7668:
+		fwname = FIRMWARE_MT7668;
+		break;
+	case 0x7922:
+	case 0x7961:
+	case 0x7925:
+		/* Reset the device to ensure it's in the initial state before
+		 * downloading the firmware to ensure.
+		 */
+
+		if (!test_bit(BTMTK_FIRMWARE_LOADED, &btmtk_data->flags))
+			btmtk_usb_subsys_reset(hdev, dev_id);
+
+		btmtk_fw_get_filename(fw_bin_name, sizeof(fw_bin_name), dev_id,
+				      fw_version, fw_flavor);
+
+		err = btmtk_setup_firmware_79xx(hdev, fw_bin_name,
+						btmtk_usb_hci_wmt_sync);
+		if (err < 0) {
+			bt_dev_err(hdev, "Failed to set up firmware (%d)", err);
+			clear_bit(BTMTK_FIRMWARE_LOADED, &btmtk_data->flags);
+			return err;
+		}
+
+		set_bit(BTMTK_FIRMWARE_LOADED, &btmtk_data->flags);
+
+		/* It's Device EndPoint Reset Option Register */
+		err = btmtk_usb_uhw_reg_write(hdev, MTK_EP_RST_OPT,
+					      MTK_EP_RST_IN_OUT_OPT);
+		if (err < 0)
+			return err;
+
+		/* Enable Bluetooth protocol */
+		param = 1;
+		wmt_params.op = BTMTK_WMT_FUNC_CTRL;
+		wmt_params.flag = 0;
+		wmt_params.dlen = sizeof(param);
+		wmt_params.data = &param;
+		wmt_params.status = NULL;
+
+		err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
+		if (err < 0) {
+			bt_dev_err(hdev, "Failed to send wmt func ctrl (%d)", err);
+			return err;
+		}
+
+		hci_set_msft_opcode(hdev, 0xFD30);
+		hci_set_aosp_capable(hdev);
+
+		goto done;
+	default:
+		bt_dev_err(hdev, "Unsupported hardware variant (%08x)",
+			   dev_id);
+		return -ENODEV;
+	}
+
+	/* Query whether the firmware is already download */
+	wmt_params.op = BTMTK_WMT_SEMAPHORE;
+	wmt_params.flag = 1;
+	wmt_params.dlen = 0;
+	wmt_params.data = NULL;
+	wmt_params.status = &status;
+
+	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
+	if (err < 0) {
+		bt_dev_err(hdev, "Failed to query firmware status (%d)", err);
+		return err;
+	}
+
+	if (status == BTMTK_WMT_PATCH_DONE) {
+		bt_dev_info(hdev, "firmware already downloaded");
+		goto ignore_setup_fw;
+	}
+
+	/* Setup a firmware which the device definitely requires */
+	err = btmtk_setup_firmware(hdev, fwname,
+				   btmtk_usb_hci_wmt_sync);
+	if (err < 0)
+		return err;
+
+ignore_setup_fw:
+	err = readx_poll_timeout(btmtk_usb_func_query, hdev, status,
+				 status < 0 || status != BTMTK_WMT_ON_PROGRESS,
+				 2000, 5000000);
+	/* -ETIMEDOUT happens */
+	if (err < 0)
+		return err;
+
+	/* The other errors happen in btmtk_usb_func_query */
+	if (status < 0)
+		return status;
+
+	if (status == BTMTK_WMT_ON_DONE) {
+		bt_dev_info(hdev, "function already on");
+		goto ignore_func_on;
+	}
+
+	/* Enable Bluetooth protocol */
+	param = 1;
+	wmt_params.op = BTMTK_WMT_FUNC_CTRL;
+	wmt_params.flag = 0;
+	wmt_params.dlen = sizeof(param);
+	wmt_params.data = &param;
+	wmt_params.status = NULL;
+
+	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
+	if (err < 0) {
+		bt_dev_err(hdev, "Failed to send wmt func ctrl (%d)", err);
+		return err;
+	}
+
+ignore_func_on:
+	/* Apply the low power environment setup */
+	tci_sleep.mode = 0x5;
+	tci_sleep.duration = cpu_to_le16(0x640);
+	tci_sleep.host_duration = cpu_to_le16(0x640);
+	tci_sleep.host_wakeup_pin = 0;
+	tci_sleep.time_compensation = 0;
+
+	skb = __hci_cmd_sync(hdev, 0xfc7a, sizeof(tci_sleep), &tci_sleep,
+			     HCI_INIT_TIMEOUT);
+	if (IS_ERR(skb)) {
+		err = PTR_ERR(skb);
+		bt_dev_err(hdev, "Failed to apply low power setting (%d)", err);
+		return err;
+	}
+	kfree_skb(skb);
+
+done:
+	rettime = ktime_get();
+	delta = ktime_sub(rettime, calltime);
+	duration = (unsigned long long)ktime_to_ns(delta) >> 10;
+
+	bt_dev_info(hdev, "Device setup in %llu usecs", duration);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(btmtk_usb_setup);
+
+int btmtk_usb_shutdown(struct hci_dev *hdev)
+{
+	struct btmtk_hci_wmt_params wmt_params;
+	u8 param = 0;
+	int err;
+
+	/* Disable the device */
+	wmt_params.op = BTMTK_WMT_FUNC_CTRL;
+	wmt_params.flag = 0;
+	wmt_params.dlen = sizeof(param);
+	wmt_params.data = &param;
+	wmt_params.status = NULL;
+
+	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
+	if (err < 0) {
+		bt_dev_err(hdev, "Failed to send wmt func ctrl (%d)", err);
+		return err;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(btmtk_usb_shutdown);
 
 MODULE_AUTHOR("Sean Wang <sean.wang@mediatek.com>");
 MODULE_AUTHOR("Mark Chen <mark-yw.chen@mediatek.com>");
diff --git a/drivers/bluetooth/btmtk.h b/drivers/bluetooth/btmtk.h
index 3055b9728ae2..fe41528ae509 100644
--- a/drivers/bluetooth/btmtk.h
+++ b/drivers/bluetooth/btmtk.h
@@ -140,6 +140,8 @@ struct btmtk_hci_wmt_params {
 
 enum {
 	BTMTK_TX_WAIT_VND_EVT,
+	BTMTK_FIRMWARE_LOADED,
+	BTMTK_HW_RESET_ACTIVE,
 };
 
 typedef int (*btmtk_reset_sync_func_t)(struct hci_dev *, void *);
@@ -152,6 +154,7 @@ struct btmtk_coredump_info {
 };
 
 struct btmtk_data {
+	const char *drv_name;
 	unsigned long flags;
 	u32 dev_id;
 	btmtk_reset_sync_func_t reset_sync;
@@ -186,8 +189,11 @@ int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb);
 void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id, u32 fw_ver,
 			   u32 fw_flavor);
 
-int btmtk_usb_hci_wmt_sync(struct hci_dev *hdev,
-			   struct btmtk_hci_wmt_params *wmt_params);
+int btmtk_usb_subsys_reset(struct hci_dev *hdev, u32 dev_id);
+
+int btmtk_usb_setup(struct hci_dev *hdev);
+
+int btmtk_usb_shutdown(struct hci_dev *hdev);
 #else
 
 static inline int btmtk_set_bdaddr(struct hci_dev *hdev,
@@ -228,8 +234,17 @@ static void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id,
 {
 }
 
-static int btmtk_usb_hci_wmt_sync(struct hci_dev *hdev,
-				  struct btmtk_hci_wmt_params *wmt_params)
+static int btmtk_usb_subsys_reset(struct hci_dev *hdev, u32 dev_id)
+{
+	return -EOPNOTSUPP;
+}
+
+static int btmtk_usb_setup(struct hci_dev *hdev)
+{
+	return -EOPNOTSUPP;
+}
+
+static int btmtk_usb_shutdown(struct hci_dev *hdev)
 {
 	return -EOPNOTSUPP;
 }
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 0376d141d8fe..9fca3eb6ae45 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -2645,241 +2645,6 @@ static int btusb_recv_event_realtek(struct hci_dev *hdev, struct sk_buff *skb)
 	return hci_recv_frame(hdev, skb);
 }
 
-static int btusb_mtk_func_query(struct hci_dev *hdev)
-{
-	struct btmtk_hci_wmt_params wmt_params;
-	int status, err;
-	u8 param = 0;
-
-	/* Query whether the function is enabled */
-	wmt_params.op = BTMTK_WMT_FUNC_CTRL;
-	wmt_params.flag = 4;
-	wmt_params.dlen = sizeof(param);
-	wmt_params.data = &param;
-	wmt_params.status = &status;
-
-	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
-	if (err < 0) {
-		bt_dev_err(hdev, "Failed to query function status (%d)", err);
-		return err;
-	}
-
-	return status;
-}
-
-static int btusb_mtk_uhw_reg_write(struct btusb_data *data, u32 reg, u32 val)
-{
-	struct hci_dev *hdev = data->hdev;
-	int pipe, err;
-	void *buf;
-
-	buf = kzalloc(4, GFP_KERNEL);
-	if (!buf)
-		return -ENOMEM;
-
-	put_unaligned_le32(val, buf);
-
-	pipe = usb_sndctrlpipe(data->udev, 0);
-	err = usb_control_msg(data->udev, pipe, 0x02,
-			      0x5E,
-			      reg >> 16, reg & 0xffff,
-			      buf, 4, USB_CTRL_SET_TIMEOUT);
-	if (err < 0)
-		bt_dev_err(hdev, "Failed to write uhw reg(%d)", err);
-
-	kfree(buf);
-
-	return err;
-}
-
-static int btusb_mtk_uhw_reg_read(struct btusb_data *data, u32 reg, u32 *val)
-{
-	struct hci_dev *hdev = data->hdev;
-	int pipe, err;
-	void *buf;
-
-	buf = kzalloc(4, GFP_KERNEL);
-	if (!buf)
-		return -ENOMEM;
-
-	pipe = usb_rcvctrlpipe(data->udev, 0);
-	err = usb_control_msg(data->udev, pipe, 0x01,
-			      0xDE,
-			      reg >> 16, reg & 0xffff,
-			      buf, 4, USB_CTRL_GET_TIMEOUT);
-	if (err < 0) {
-		bt_dev_err(hdev, "Failed to read uhw reg(%d)", err);
-		goto err_free_buf;
-	}
-
-	*val = get_unaligned_le32(buf);
-	bt_dev_dbg(hdev, "reg=%x, value=0x%08x", reg, *val);
-
-err_free_buf:
-	kfree(buf);
-
-	return err;
-}
-
-static int btusb_mtk_reg_read(struct btusb_data *data, u32 reg, u32 *val)
-{
-	int pipe, err, size = sizeof(u32);
-	void *buf;
-
-	buf = kzalloc(size, GFP_KERNEL);
-	if (!buf)
-		return -ENOMEM;
-
-	pipe = usb_rcvctrlpipe(data->udev, 0);
-	err = usb_control_msg(data->udev, pipe, 0x63,
-			      USB_TYPE_VENDOR | USB_DIR_IN,
-			      reg >> 16, reg & 0xffff,
-			      buf, size, USB_CTRL_GET_TIMEOUT);
-	if (err < 0)
-		goto err_free_buf;
-
-	*val = get_unaligned_le32(buf);
-
-err_free_buf:
-	kfree(buf);
-
-	return err;
-}
-
-static int btusb_mtk_id_get(struct btusb_data *data, u32 reg, u32 *id)
-{
-	return btusb_mtk_reg_read(data, reg, id);
-}
-
-static u32 btusb_mtk_reset_done(struct hci_dev *hdev)
-{
-	struct btusb_data *data = hci_get_drvdata(hdev);
-	u32 val = 0;
-
-	btusb_mtk_uhw_reg_read(data, MTK_BT_MISC, &val);
-
-	return val & MTK_BT_RST_DONE;
-}
-
-static int btusb_mtk_subsys_reset(struct hci_dev *hdev, u32 dev_id)
-{
-	struct btusb_data *data = hci_get_drvdata(hdev);
-	u32 val;
-	int err;
-
-	if (dev_id == 0x7922) {
-		err = btusb_mtk_uhw_reg_read(data, MTK_BT_SUBSYS_RST, &val);
-		if (err < 0)
-			return err;
-		val |= 0x00002020;
-		err = btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, val);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, 0x00010001);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_read(data, MTK_BT_SUBSYS_RST, &val);
-		if (err < 0)
-			return err;
-		val |= BIT(0);
-		err = btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, val);
-		if (err < 0)
-			return err;
-		msleep(100);
-	} else if (dev_id == 0x7925) {
-		err = btusb_mtk_uhw_reg_read(data, MTK_BT_RESET_REG_CONNV3, &val);
-		if (err < 0)
-			return err;
-		val |= (1 << 5);
-		err = btusb_mtk_uhw_reg_write(data, MTK_BT_RESET_REG_CONNV3, val);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_read(data, MTK_BT_RESET_REG_CONNV3, &val);
-		if (err < 0)
-			return err;
-		val &= 0xFFFF00FF;
-		if (err < 0)
-			return err;
-		val |= (1 << 13);
-		err = btusb_mtk_uhw_reg_write(data, MTK_BT_RESET_REG_CONNV3, val);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, 0x00010001);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_read(data, MTK_BT_RESET_REG_CONNV3, &val);
-		if (err < 0)
-			return err;
-		val |= (1 << 0);
-		err = btusb_mtk_uhw_reg_write(data, MTK_BT_RESET_REG_CONNV3, val);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT, 0x000000FF);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT, &val);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT1, 0x000000FF);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT1, &val);
-		if (err < 0)
-			return err;
-		msleep(100);
-	} else {
-		/* It's Device EndPoint Reset Option Register */
-		bt_dev_dbg(hdev, "Initiating reset mechanism via uhw");
-		err = btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, MTK_EP_RST_IN_OUT_OPT);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_read(data, MTK_BT_WDT_STATUS, &val);
-		if (err < 0)
-			return err;
-		/* Reset the bluetooth chip via USB interface. */
-		err = btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, 1);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT, 0x000000FF);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT, &val);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT1, 0x000000FF);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_read(data, MTK_UDMA_INT_STA_BT1, &val);
-		if (err < 0)
-			return err;
-		/* MT7921 need to delay 20ms between toggle reset bit */
-		msleep(20);
-		err = btusb_mtk_uhw_reg_write(data, MTK_BT_SUBSYS_RST, 0);
-		if (err < 0)
-			return err;
-		err = btusb_mtk_uhw_reg_read(data, MTK_BT_SUBSYS_RST, &val);
-		if (err < 0)
-			return err;
-	}
-
-	err = readx_poll_timeout(btusb_mtk_reset_done, hdev, val,
-				 val & MTK_BT_RST_DONE, 20000, 1000000);
-	if (err < 0)
-		bt_dev_err(hdev, "Reset timeout");
-
-	if (dev_id == 0x7922) {
-		err = btusb_mtk_uhw_reg_write(data, MTK_UDMA_INT_STA_BT, 0x000000FF);
-		if (err < 0)
-			return err;
-	}
-
-	err = btusb_mtk_id_get(data, 0x70010200, &val);
-	if (err < 0 || !val)
-		bt_dev_err(hdev, "Can't get device id, subsys reset fail.");
-
-	return err;
-}
-
 static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
 {
 	struct btusb_data *data = hci_get_drvdata(hdev);
@@ -2887,7 +2652,7 @@ static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
 	int err;
 
 	/* It's MediaTek specific bluetooth reset mechanism via USB */
-	if (test_and_set_bit(BTUSB_HW_RESET_ACTIVE, &data->flags)) {
+	if (test_and_set_bit(BTMTK_HW_RESET_ACTIVE, &btmtk_data->flags)) {
 		bt_dev_err(hdev, "last reset failed? Not resetting again");
 		return -EBUSY;
 	}
@@ -2899,10 +2664,10 @@ static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
 	btusb_stop_traffic(data);
 	usb_kill_anchored_urbs(&data->tx_anchor);
 
-	err = btusb_mtk_subsys_reset(hdev, btmtk_data->dev_id);
+	err = btmtk_usb_subsys_reset(hdev, btmtk_data->dev_id);
 
 	usb_queue_reset_device(data->intf);
-	clear_bit(BTUSB_HW_RESET_ACTIVE, &data->flags);
+	clear_bit(BTMTK_HW_RESET_ACTIVE, &btmtk_data->flags);
 
 	return err;
 }
@@ -2910,212 +2675,23 @@ static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
 static int btusb_mtk_setup(struct hci_dev *hdev)
 {
 	struct btusb_data *data = hci_get_drvdata(hdev);
-	struct btmtk_hci_wmt_params wmt_params;
-	ktime_t calltime, delta, rettime;
-	struct btmtk_tci_sleep tci_sleep;
-	unsigned long long duration;
-	struct sk_buff *skb;
-	const char *fwname;
-	int err, status;
-	u32 dev_id = 0;
-	char fw_bin_name[64];
-	u32 fw_version = 0, fw_flavor = 0;
-	u8 param;
-	struct btmtk_data *mediatek;
-
-	calltime = ktime_get();
-
-	err = btusb_mtk_id_get(data, 0x80000008, &dev_id);
-	if (err < 0) {
-		bt_dev_err(hdev, "Failed to get device id (%d)", err);
-		return err;
-	}
-
-	if (!dev_id || dev_id != 0x7663) {
-		err = btusb_mtk_id_get(data, 0x70010200, &dev_id);
-		if (err < 0) {
-			bt_dev_err(hdev, "Failed to get device id (%d)", err);
-			return err;
-		}
-		err = btusb_mtk_id_get(data, 0x80021004, &fw_version);
-		if (err < 0) {
-			bt_dev_err(hdev, "Failed to get fw version (%d)", err);
-			return err;
-		}
-		err = btusb_mtk_id_get(data, 0x70010020, &fw_flavor);
-		if (err < 0) {
-			bt_dev_err(hdev, "Failed to get fw flavor (%d)", err);
-			return err;
-		}
-		fw_flavor = (fw_flavor & 0x00000080) >> 7;
-	}
-
-	mediatek = hci_get_priv(hdev);
-	mediatek->dev_id = dev_id;
-	mediatek->reset_sync = btusb_mtk_reset;
-
-	err = btmtk_register_coredump(hdev, btusb_driver.name, fw_version);
-	if (err < 0)
-		bt_dev_err(hdev, "Failed to register coredump (%d)", err);
-
-	switch (dev_id) {
-	case 0x7663:
-		fwname = FIRMWARE_MT7663;
-		break;
-	case 0x7668:
-		fwname = FIRMWARE_MT7668;
-		break;
-	case 0x7922:
-	case 0x7961:
-	case 0x7925:
-		/* Reset the device to ensure it's in the initial state before
-		 * downloading the firmware to ensure.
-		 */
-
-		if (!test_bit(BTUSB_FIRMWARE_LOADED, &data->flags))
-			btusb_mtk_subsys_reset(hdev, dev_id);
-
-		btmtk_fw_get_filename(fw_bin_name, sizeof(fw_bin_name), dev_id,
-				      fw_version, fw_flavor);
-
-		err = btmtk_setup_firmware_79xx(hdev, fw_bin_name,
-						btmtk_usb_hci_wmt_sync);
-		if (err < 0) {
-			bt_dev_err(hdev, "Failed to set up firmware (%d)", err);
-			clear_bit(BTUSB_FIRMWARE_LOADED, &data->flags);
-			return err;
-		}
-
-		set_bit(BTUSB_FIRMWARE_LOADED, &data->flags);
-
-		/* It's Device EndPoint Reset Option Register */
-		btusb_mtk_uhw_reg_write(data, MTK_EP_RST_OPT, MTK_EP_RST_IN_OUT_OPT);
-
-		/* Enable Bluetooth protocol */
-		param = 1;
-		wmt_params.op = BTMTK_WMT_FUNC_CTRL;
-		wmt_params.flag = 0;
-		wmt_params.dlen = sizeof(param);
-		wmt_params.data = &param;
-		wmt_params.status = NULL;
-
-		err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
-		if (err < 0) {
-			bt_dev_err(hdev, "Failed to send wmt func ctrl (%d)", err);
-			return err;
-		}
-
-		hci_set_msft_opcode(hdev, 0xFD30);
-		hci_set_aosp_capable(hdev);
-		goto done;
-	default:
-		bt_dev_err(hdev, "Unsupported hardware variant (%08x)",
-			   dev_id);
-		return -ENODEV;
-	}
-
-	/* Query whether the firmware is already download */
-	wmt_params.op = BTMTK_WMT_SEMAPHORE;
-	wmt_params.flag = 1;
-	wmt_params.dlen = 0;
-	wmt_params.data = NULL;
-	wmt_params.status = &status;
-
-	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
-	if (err < 0) {
-		bt_dev_err(hdev, "Failed to query firmware status (%d)", err);
-		return err;
-	}
-
-	if (status == BTMTK_WMT_PATCH_DONE) {
-		bt_dev_info(hdev, "firmware already downloaded");
-		goto ignore_setup_fw;
-	}
-
-	/* Setup a firmware which the device definitely requires */
-	err = btmtk_setup_firmware(hdev, fwname,
-				   btmtk_usb_hci_wmt_sync);
-	if (err < 0)
-		return err;
-
-ignore_setup_fw:
-	err = readx_poll_timeout(btusb_mtk_func_query, hdev, status,
-				 status < 0 || status != BTMTK_WMT_ON_PROGRESS,
-				 2000, 5000000);
-	/* -ETIMEDOUT happens */
-	if (err < 0)
-		return err;
-
-	/* The other errors happen in btusb_mtk_func_query */
-	if (status < 0)
-		return status;
-
-	if (status == BTMTK_WMT_ON_DONE) {
-		bt_dev_info(hdev, "function already on");
-		goto ignore_func_on;
-	}
-
-	/* Enable Bluetooth protocol */
-	param = 1;
-	wmt_params.op = BTMTK_WMT_FUNC_CTRL;
-	wmt_params.flag = 0;
-	wmt_params.dlen = sizeof(param);
-	wmt_params.data = &param;
-	wmt_params.status = NULL;
-
-	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
-	if (err < 0) {
-		bt_dev_err(hdev, "Failed to send wmt func ctrl (%d)", err);
-		return err;
-	}
-
-ignore_func_on:
-	/* Apply the low power environment setup */
-	tci_sleep.mode = 0x5;
-	tci_sleep.duration = cpu_to_le16(0x640);
-	tci_sleep.host_duration = cpu_to_le16(0x640);
-	tci_sleep.host_wakeup_pin = 0;
-	tci_sleep.time_compensation = 0;
-
-	skb = __hci_cmd_sync(hdev, 0xfc7a, sizeof(tci_sleep), &tci_sleep,
-			     HCI_INIT_TIMEOUT);
-	if (IS_ERR(skb)) {
-		err = PTR_ERR(skb);
-		bt_dev_err(hdev, "Failed to apply low power setting (%d)", err);
-		return err;
-	}
-	kfree_skb(skb);
-
-done:
-	rettime = ktime_get();
-	delta = ktime_sub(rettime, calltime);
-	duration = (unsigned long long)ktime_to_ns(delta) >> 10;
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
 
-	bt_dev_info(hdev, "Device setup in %llu usecs", duration);
+	/* MediaTek WMT vendor cmd requiring below USB resources to
+	 * complete the handshake.
+	 */
+	btmtk_data->drv_name = btusb_driver.name;
+	btmtk_data->intf = data->intf;
+	btmtk_data->udev = data->udev;
+	btmtk_data->ctrl_anchor = &data->ctrl_anchor;
+	btmtk_data->reset_sync = btusb_mtk_reset;
 
-	return 0;
+	return btmtk_usb_setup(hdev);
 }
 
 static int btusb_mtk_shutdown(struct hci_dev *hdev)
 {
-	struct btmtk_hci_wmt_params wmt_params;
-	u8 param = 0;
-	int err;
-
-	/* Disable the device */
-	wmt_params.op = BTMTK_WMT_FUNC_CTRL;
-	wmt_params.flag = 0;
-	wmt_params.dlen = sizeof(param);
-	wmt_params.data = &param;
-	wmt_params.status = NULL;
-
-	err = btmtk_usb_hci_wmt_sync(hdev, &wmt_params);
-	if (err < 0) {
-		bt_dev_err(hdev, "Failed to send wmt func ctrl (%d)", err);
-		return err;
-	}
-
-	return 0;
+	return btmtk_usb_shutdown(hdev);
 }
 
 static int btusb_recv_acl_mtk(struct hci_dev *hdev, struct sk_buff *skb)
-- 
cgit v1.2.3-58-ga151


From 6dc22ab9f085ae165e4ce89d61fb426f94e8a969 Mon Sep 17 00:00:00 2001
From: Chris Lu <chris.lu@mediatek.com>
Date: Thu, 4 Jul 2024 14:01:15 +0800
Subject: Bluetooth: btmtk: move btusb_recv_acl_mtk to btmtk.c

Move btusb_recv_acl_mtk from btusb.c to btmtk.c which holds
vendor specific stuff and would make btusb.c clean.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btmtk.c | 33 +++++++++++++++++++++++++++++++++
 drivers/bluetooth/btmtk.h |  7 +++++++
 drivers/bluetooth/btusb.c | 34 +---------------------------------
 3 files changed, 41 insertions(+), 33 deletions(-)

diff --git a/drivers/bluetooth/btmtk.c b/drivers/bluetooth/btmtk.c
index f5ecb8227d3b..fe3b892f6c6e 100644
--- a/drivers/bluetooth/btmtk.c
+++ b/drivers/bluetooth/btmtk.c
@@ -930,6 +930,39 @@ int btmtk_usb_subsys_reset(struct hci_dev *hdev, u32 dev_id)
 }
 EXPORT_SYMBOL_GPL(btmtk_usb_subsys_reset);
 
+int btmtk_usb_recv_acl(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct btmtk_data *data = hci_get_priv(hdev);
+	u16 handle = le16_to_cpu(hci_acl_hdr(skb)->handle);
+
+	switch (handle) {
+	case 0xfc6f:		/* Firmware dump from device */
+		/* When the firmware hangs, the device can no longer
+		 * suspend and thus disable auto-suspend.
+		 */
+		usb_disable_autosuspend(data->udev);
+
+		/* We need to forward the diagnostic packet to userspace daemon
+		 * for backward compatibility, so we have to clone the packet
+		 * extraly for the in-kernel coredump support.
+		 */
+		if (IS_ENABLED(CONFIG_DEV_COREDUMP)) {
+			struct sk_buff *skb_cd = skb_clone(skb, GFP_ATOMIC);
+
+			if (skb_cd)
+				btmtk_process_coredump(hdev, skb_cd);
+		}
+
+		fallthrough;
+	case 0x05ff:		/* Firmware debug logging 1 */
+	case 0x05fe:		/* Firmware debug logging 2 */
+		return hci_recv_diag(hdev, skb);
+	}
+
+	return hci_recv_frame(hdev, skb);
+}
+EXPORT_SYMBOL_GPL(btmtk_usb_recv_acl);
+
 int btmtk_usb_setup(struct hci_dev *hdev)
 {
 	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
diff --git a/drivers/bluetooth/btmtk.h b/drivers/bluetooth/btmtk.h
index fe41528ae509..47193b867b9f 100644
--- a/drivers/bluetooth/btmtk.h
+++ b/drivers/bluetooth/btmtk.h
@@ -191,6 +191,8 @@ void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id, u32 fw_ver,
 
 int btmtk_usb_subsys_reset(struct hci_dev *hdev, u32 dev_id);
 
+int btmtk_usb_recv_acl(struct hci_dev *hdev, struct sk_buff *skb);
+
 int btmtk_usb_setup(struct hci_dev *hdev);
 
 int btmtk_usb_shutdown(struct hci_dev *hdev);
@@ -239,6 +241,11 @@ static int btmtk_usb_subsys_reset(struct hci_dev *hdev, u32 dev_id)
 	return -EOPNOTSUPP;
 }
 
+static int btmtk_usb_recv_acl(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	return -EOPNOTSUPP;
+}
+
 static int btmtk_usb_setup(struct hci_dev *hdev)
 {
 	return -EOPNOTSUPP;
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 9fca3eb6ae45..034256c399dd 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -2694,38 +2694,6 @@ static int btusb_mtk_shutdown(struct hci_dev *hdev)
 	return btmtk_usb_shutdown(hdev);
 }
 
-static int btusb_recv_acl_mtk(struct hci_dev *hdev, struct sk_buff *skb)
-{
-	struct btusb_data *data = hci_get_drvdata(hdev);
-	u16 handle = le16_to_cpu(hci_acl_hdr(skb)->handle);
-
-	switch (handle) {
-	case 0xfc6f:		/* Firmware dump from device */
-		/* When the firmware hangs, the device can no longer
-		 * suspend and thus disable auto-suspend.
-		 */
-		usb_disable_autosuspend(data->udev);
-
-		/* We need to forward the diagnostic packet to userspace daemon
-		 * for backward compatibility, so we have to clone the packet
-		 * extraly for the in-kernel coredump support.
-		 */
-		if (IS_ENABLED(CONFIG_DEV_COREDUMP)) {
-			struct sk_buff *skb_cd = skb_clone(skb, GFP_ATOMIC);
-
-			if (skb_cd)
-				btmtk_process_coredump(hdev, skb_cd);
-		}
-
-		fallthrough;
-	case 0x05ff:		/* Firmware debug logging 1 */
-	case 0x05fe:		/* Firmware debug logging 2 */
-		return hci_recv_diag(hdev, skb);
-	}
-
-	return hci_recv_frame(hdev, skb);
-}
-
 #ifdef CONFIG_PM
 /* Configure an out-of-band gpio as wake-up pin, if specified in device tree */
 static int marvell_config_oob_wake(struct hci_dev *hdev)
@@ -3831,7 +3799,7 @@ static int btusb_probe(struct usb_interface *intf,
 		hdev->set_bdaddr = btmtk_set_bdaddr;
 		set_bit(HCI_QUIRK_BROKEN_ENHANCED_SETUP_SYNC_CONN, &hdev->quirks);
 		set_bit(HCI_QUIRK_NON_PERSISTENT_SETUP, &hdev->quirks);
-		data->recv_acl = btusb_recv_acl_mtk;
+		data->recv_acl = btmtk_usb_recv_acl;
 	}
 
 	if (id->driver_info & BTUSB_SWAVE) {
-- 
cgit v1.2.3-58-ga151


From ceac1cb0259de682d78f5c784ef8e0b13022e9d9 Mon Sep 17 00:00:00 2001
From: Chris Lu <chris.lu@mediatek.com>
Date: Thu, 4 Jul 2024 14:01:16 +0800
Subject: Bluetooth: btusb: mediatek: add ISO data transmission functions

This patch implements functions for ISO data send and receive in btusb
driver for MediaTek's controller.

MediaTek defines a specific interrupt endpoint for ISO data transmissin
because the characteristics of interrupt endpoint are similar to the
application of ISO data which can support guaranteed transmissin
bandwidth, enough maximum data length and error checking mechanism.

Driver sets up ISO interface and endpoints in btusb_mtk_setup and clears
the setup in btusb_mtk_shutdown. These flow can't move to btmtk.c due to
btusb_driver is only defined in btusb.c when claiming/relaesing interface.
ISO packet anchor stops when driver suspending and resubmit interrupt urb
for ISO data when driver resuming.

Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btmtk.c | 311 ++++++++++++++++++++++++++++++++++++++++++++++
 drivers/bluetooth/btmtk.h |  36 ++++++
 drivers/bluetooth/btusb.c |  68 ++++++++++
 3 files changed, 415 insertions(+)

diff --git a/drivers/bluetooth/btmtk.c b/drivers/bluetooth/btmtk.c
index fe3b892f6c6e..b7c348687a77 100644
--- a/drivers/bluetooth/btmtk.c
+++ b/drivers/bluetooth/btmtk.c
@@ -22,6 +22,9 @@
 #define MTK_SEC_MAP_COMMON_SIZE	12
 #define MTK_SEC_MAP_NEED_SEND_SIZE	52
 
+/* It is for mt79xx iso data transmission setting */
+#define MTK_ISO_THRESHOLD	264
+
 struct btmtk_patch_header {
 	u8 datetime[16];
 	u8 platform[4];
@@ -963,6 +966,308 @@ int btmtk_usb_recv_acl(struct hci_dev *hdev, struct sk_buff *skb)
 }
 EXPORT_SYMBOL_GPL(btmtk_usb_recv_acl);
 
+static int btmtk_isopkt_pad(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	if (skb->len > MTK_ISO_THRESHOLD)
+		return -EINVAL;
+
+	if (skb_pad(skb, MTK_ISO_THRESHOLD - skb->len))
+		return -ENOMEM;
+
+	__skb_put(skb, MTK_ISO_THRESHOLD - skb->len);
+
+	return 0;
+}
+
+static int __set_mtk_intr_interface(struct hci_dev *hdev)
+{
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
+	struct usb_interface *intf = btmtk_data->isopkt_intf;
+	int i, err;
+
+	if (!btmtk_data->isopkt_intf)
+		return -ENODEV;
+
+	err = usb_set_interface(btmtk_data->udev, MTK_ISO_IFNUM, 1);
+	if (err < 0) {
+		bt_dev_err(hdev, "setting interface failed (%d)", -err);
+		return err;
+	}
+
+	btmtk_data->isopkt_tx_ep = NULL;
+	btmtk_data->isopkt_rx_ep = NULL;
+
+	for (i = 0; i < intf->cur_altsetting->desc.bNumEndpoints; i++) {
+		struct usb_endpoint_descriptor *ep_desc;
+
+		ep_desc = &intf->cur_altsetting->endpoint[i].desc;
+
+		if (!btmtk_data->isopkt_tx_ep &&
+		    usb_endpoint_is_int_out(ep_desc)) {
+			btmtk_data->isopkt_tx_ep = ep_desc;
+			continue;
+		}
+
+		if (!btmtk_data->isopkt_rx_ep &&
+		    usb_endpoint_is_int_in(ep_desc)) {
+			btmtk_data->isopkt_rx_ep = ep_desc;
+			continue;
+		}
+	}
+
+	if (!btmtk_data->isopkt_tx_ep ||
+	    !btmtk_data->isopkt_rx_ep) {
+		bt_dev_err(hdev, "invalid interrupt descriptors");
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+struct urb *alloc_mtk_intr_urb(struct hci_dev *hdev, struct sk_buff *skb,
+			       usb_complete_t tx_complete)
+{
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
+	struct urb *urb;
+	unsigned int pipe;
+
+	if (!btmtk_data->isopkt_tx_ep)
+		return ERR_PTR(-ENODEV);
+
+	urb = usb_alloc_urb(0, GFP_KERNEL);
+	if (!urb)
+		return ERR_PTR(-ENOMEM);
+
+	if (btmtk_isopkt_pad(hdev, skb))
+		return ERR_PTR(-EINVAL);
+
+	pipe = usb_sndintpipe(btmtk_data->udev,
+			      btmtk_data->isopkt_tx_ep->bEndpointAddress);
+
+	usb_fill_int_urb(urb, btmtk_data->udev, pipe,
+			 skb->data, skb->len, tx_complete,
+			 skb, btmtk_data->isopkt_tx_ep->bInterval);
+
+	skb->dev = (void *)hdev;
+
+	return urb;
+}
+EXPORT_SYMBOL_GPL(alloc_mtk_intr_urb);
+
+static int btmtk_recv_isopkt(struct hci_dev *hdev, void *buffer, int count)
+{
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
+	struct sk_buff *skb;
+	unsigned long flags;
+	int err = 0;
+
+	spin_lock_irqsave(&btmtk_data->isorxlock, flags);
+	skb = btmtk_data->isopkt_skb;
+
+	while (count) {
+		int len;
+
+		if (!skb) {
+			skb = bt_skb_alloc(HCI_MAX_ISO_SIZE, GFP_ATOMIC);
+			if (!skb) {
+				err = -ENOMEM;
+				break;
+			}
+
+			hci_skb_pkt_type(skb) = HCI_ISODATA_PKT;
+			hci_skb_expect(skb) = HCI_ISO_HDR_SIZE;
+		}
+
+		len = min_t(uint, hci_skb_expect(skb), count);
+		skb_put_data(skb, buffer, len);
+
+		count -= len;
+		buffer += len;
+		hci_skb_expect(skb) -= len;
+
+		if (skb->len == HCI_ISO_HDR_SIZE) {
+			__le16 dlen = ((struct hci_iso_hdr *)skb->data)->dlen;
+
+			/* Complete ISO header */
+			hci_skb_expect(skb) = __le16_to_cpu(dlen);
+
+			if (skb_tailroom(skb) < hci_skb_expect(skb)) {
+				kfree_skb(skb);
+				skb = NULL;
+
+				err = -EILSEQ;
+				break;
+			}
+		}
+
+		if (!hci_skb_expect(skb)) {
+			/* Complete frame */
+			hci_recv_frame(hdev, skb);
+			skb = NULL;
+		}
+	}
+
+	btmtk_data->isopkt_skb = skb;
+	spin_unlock_irqrestore(&btmtk_data->isorxlock, flags);
+
+	return err;
+}
+
+static void btmtk_intr_complete(struct urb *urb)
+{
+	struct hci_dev *hdev = urb->context;
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
+	int err;
+
+	BT_DBG("%s urb %p status %d count %d", hdev->name, urb, urb->status,
+	       urb->actual_length);
+
+	if (!test_bit(HCI_RUNNING, &hdev->flags))
+		return;
+
+	if (hdev->suspended)
+		return;
+
+	if (urb->status == 0) {
+		hdev->stat.byte_rx += urb->actual_length;
+
+		if (btmtk_recv_isopkt(hdev, urb->transfer_buffer,
+				      urb->actual_length) < 0) {
+			bt_dev_err(hdev, "corrupted iso packet");
+			hdev->stat.err_rx++;
+		}
+	} else if (urb->status == -ENOENT) {
+		/* Avoid suspend failed when usb_kill_urb */
+		return;
+	}
+
+	usb_mark_last_busy(btmtk_data->udev);
+	usb_anchor_urb(urb, &btmtk_data->isopkt_anchor);
+
+	err = usb_submit_urb(urb, GFP_ATOMIC);
+	if (err < 0) {
+		/* -EPERM: urb is being killed;
+		 * -ENODEV: device got disconnected
+		 */
+		if (err != -EPERM && err != -ENODEV)
+			bt_dev_err(hdev, "urb %p failed to resubmit (%d)",
+				   urb, -err);
+		if (err != -EPERM)
+			hci_cmd_sync_cancel(hdev, -err);
+		usb_unanchor_urb(urb);
+	}
+}
+
+static int btmtk_submit_intr_urb(struct hci_dev *hdev, gfp_t mem_flags)
+{
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
+	unsigned char *buf;
+	unsigned int pipe;
+	struct urb *urb;
+	int err, size;
+
+	BT_DBG("%s", hdev->name);
+
+	if (!btmtk_data->isopkt_rx_ep)
+		return -ENODEV;
+
+	urb = usb_alloc_urb(0, mem_flags);
+	if (!urb)
+		return -ENOMEM;
+	size = le16_to_cpu(btmtk_data->isopkt_rx_ep->wMaxPacketSize);
+
+	buf = kmalloc(size, mem_flags);
+	if (!buf) {
+		usb_free_urb(urb);
+		return -ENOMEM;
+	}
+
+	pipe = usb_rcvintpipe(btmtk_data->udev,
+			      btmtk_data->isopkt_rx_ep->bEndpointAddress);
+
+	usb_fill_int_urb(urb, btmtk_data->udev, pipe, buf, size,
+			 btmtk_intr_complete, hdev,
+			 btmtk_data->isopkt_rx_ep->bInterval);
+
+	urb->transfer_flags |= URB_FREE_BUFFER;
+
+	usb_mark_last_busy(btmtk_data->udev);
+	usb_anchor_urb(urb, &btmtk_data->isopkt_anchor);
+
+	err = usb_submit_urb(urb, mem_flags);
+	if (err < 0) {
+		if (err != -EPERM && err != -ENODEV)
+			bt_dev_err(hdev, "urb %p submission failed (%d)",
+				   urb, -err);
+		usb_unanchor_urb(urb);
+	}
+
+	usb_free_urb(urb);
+
+	return err;
+}
+
+static int btmtk_usb_isointf_init(struct hci_dev *hdev)
+{
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
+	u8 iso_param[2] = { 0x08, 0x01 };
+	struct sk_buff *skb;
+	int err;
+
+	init_usb_anchor(&btmtk_data->isopkt_anchor);
+	spin_lock_init(&btmtk_data->isorxlock);
+
+	__set_mtk_intr_interface(hdev);
+
+	err = btmtk_submit_intr_urb(hdev, GFP_KERNEL);
+	if (err < 0) {
+		usb_kill_anchored_urbs(&btmtk_data->isopkt_anchor);
+		bt_dev_err(hdev, "ISO intf not support (%d)", err);
+		return err;
+	}
+
+	skb = __hci_cmd_sync(hdev, 0xfd98, sizeof(iso_param), iso_param,
+			     HCI_INIT_TIMEOUT);
+	if (IS_ERR(skb)) {
+		bt_dev_err(hdev, "Failed to apply iso setting (%ld)", PTR_ERR(skb));
+		return PTR_ERR(skb);
+	}
+	kfree_skb(skb);
+
+	return 0;
+}
+
+int btmtk_usb_resume(struct hci_dev *hdev)
+{
+	/* This function describes the specific additional steps taken by MediaTek
+	 * when Bluetooth usb driver's resume function is called.
+	 */
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
+
+	/* Resubmit urb for iso data transmission */
+	if (test_bit(BTMTK_ISOPKT_RUNNING, &btmtk_data->flags)) {
+		if (btmtk_submit_intr_urb(hdev, GFP_NOIO) < 0)
+			clear_bit(BTMTK_ISOPKT_RUNNING, &btmtk_data->flags);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(btmtk_usb_resume);
+
+int btmtk_usb_suspend(struct hci_dev *hdev)
+{
+	/* This function describes the specific additional steps taken by MediaTek
+	 * when Bluetooth usb driver's suspend function is called.
+	 */
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
+
+	/* Stop urb anchor for iso data transmission */
+	usb_kill_anchored_urbs(&btmtk_data->isopkt_anchor);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(btmtk_usb_suspend);
+
 int btmtk_usb_setup(struct hci_dev *hdev)
 {
 	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
@@ -1064,6 +1369,12 @@ int btmtk_usb_setup(struct hci_dev *hdev)
 		hci_set_msft_opcode(hdev, 0xFD30);
 		hci_set_aosp_capable(hdev);
 
+		/* Set up ISO interface after protocol enabled */
+		if (test_bit(BTMTK_ISOPKT_OVER_INTR, &btmtk_data->flags)) {
+			if (!btmtk_usb_isointf_init(hdev))
+				set_bit(BTMTK_ISOPKT_RUNNING, &btmtk_data->flags);
+		}
+
 		goto done;
 	default:
 		bt_dev_err(hdev, "Unsupported hardware variant (%08x)",
diff --git a/drivers/bluetooth/btmtk.h b/drivers/bluetooth/btmtk.h
index 47193b867b9f..453ed5131a37 100644
--- a/drivers/bluetooth/btmtk.h
+++ b/drivers/bluetooth/btmtk.h
@@ -40,6 +40,9 @@
 #define MTK_BT_RESET_REG_CONNV3	0x70028610
 #define MTK_BT_READ_DEV_ID	0x70010200
 
+/* MediaTek ISO Interface */
+#define MTK_ISO_IFNUM		2
+
 enum {
 	BTMTK_WMT_PATCH_DWNLD = 0x1,
 	BTMTK_WMT_TEST = 0x2,
@@ -142,6 +145,8 @@ enum {
 	BTMTK_TX_WAIT_VND_EVT,
 	BTMTK_FIRMWARE_LOADED,
 	BTMTK_HW_RESET_ACTIVE,
+	BTMTK_ISOPKT_OVER_INTR,
+	BTMTK_ISOPKT_RUNNING,
 };
 
 typedef int (*btmtk_reset_sync_func_t)(struct hci_dev *, void *);
@@ -164,6 +169,14 @@ struct btmtk_data {
 	struct usb_interface *intf;
 	struct usb_anchor *ctrl_anchor;
 	struct sk_buff *evt_skb;
+	struct usb_endpoint_descriptor *isopkt_tx_ep;
+	struct usb_endpoint_descriptor *isopkt_rx_ep;
+	struct usb_interface *isopkt_intf;
+	struct usb_anchor isopkt_anchor;
+	struct sk_buff *isopkt_skb;
+
+	/* spinlock for ISO data transmission */
+	spinlock_t isorxlock;
 };
 
 typedef int (*wmt_cmd_sync_func_t)(struct hci_dev *,
@@ -193,6 +206,13 @@ int btmtk_usb_subsys_reset(struct hci_dev *hdev, u32 dev_id);
 
 int btmtk_usb_recv_acl(struct hci_dev *hdev, struct sk_buff *skb);
 
+struct urb *alloc_mtk_intr_urb(struct hci_dev *hdev, struct sk_buff *skb,
+			       usb_complete_t tx_complete);
+
+int btmtk_usb_resume(struct hci_dev *hdev);
+
+int btmtk_usb_suspend(struct hci_dev *hdev);
+
 int btmtk_usb_setup(struct hci_dev *hdev);
 
 int btmtk_usb_shutdown(struct hci_dev *hdev);
@@ -246,6 +266,22 @@ static int btmtk_usb_recv_acl(struct hci_dev *hdev, struct sk_buff *skb)
 	return -EOPNOTSUPP;
 }
 
+static struct urb *alloc_mtk_intr_urb(struct hci_dev *hdev, struct sk_buff *skb,
+				      usb_complete_t tx_complete)
+{
+	return ERR_PTR(-EOPNOTSUPP);
+}
+
+static int btmtk_usb_resume(struct hci_dev *hdev)
+{
+	return -EOPNOTSUPP;
+}
+
+static int btmtk_usb_suspend(struct hci_dev *hdev)
+{
+	return -EOPNOTSUPP;
+}
+
 static int btmtk_usb_setup(struct hci_dev *hdev)
 {
 	return -EOPNOTSUPP;
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 034256c399dd..3077f16a149d 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -2645,6 +2645,40 @@ static int btusb_recv_event_realtek(struct hci_dev *hdev, struct sk_buff *skb)
 	return hci_recv_frame(hdev, skb);
 }
 
+static void btusb_mtk_claim_iso_intf(struct btusb_data *data)
+{
+	struct btmtk_data *btmtk_data = hci_get_priv(data->hdev);
+	int err;
+
+	err = usb_driver_claim_interface(&btusb_driver,
+					 btmtk_data->isopkt_intf, data);
+	if (err < 0) {
+		btmtk_data->isopkt_intf = NULL;
+		bt_dev_err(data->hdev, "Failed to claim iso interface");
+		return;
+	}
+
+	set_bit(BTMTK_ISOPKT_OVER_INTR, &btmtk_data->flags);
+}
+
+static void btusb_mtk_release_iso_intf(struct btusb_data *data)
+{
+	struct btmtk_data *btmtk_data = hci_get_priv(data->hdev);
+
+	if (btmtk_data->isopkt_intf) {
+		usb_kill_anchored_urbs(&btmtk_data->isopkt_anchor);
+		clear_bit(BTMTK_ISOPKT_RUNNING, &btmtk_data->flags);
+
+		dev_kfree_skb_irq(btmtk_data->isopkt_skb);
+		btmtk_data->isopkt_skb = NULL;
+		usb_set_intfdata(btmtk_data->isopkt_intf, NULL);
+		usb_driver_release_interface(&btusb_driver,
+					     btmtk_data->isopkt_intf);
+	}
+
+	clear_bit(BTMTK_ISOPKT_OVER_INTR, &btmtk_data->flags);
+}
+
 static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
 {
 	struct btusb_data *data = hci_get_drvdata(hdev);
@@ -2661,6 +2695,9 @@ static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
 	if (err < 0)
 		return err;
 
+	if (test_bit(BTMTK_ISOPKT_RUNNING, &btmtk_data->flags))
+		btusb_mtk_release_iso_intf(data);
+
 	btusb_stop_traffic(data);
 	usb_kill_anchored_urbs(&data->tx_anchor);
 
@@ -2672,6 +2709,23 @@ static int btusb_mtk_reset(struct hci_dev *hdev, void *rst_data)
 	return err;
 }
 
+static int btusb_send_frame_mtk(struct hci_dev *hdev, struct sk_buff *skb)
+{
+	struct urb *urb;
+
+	BT_DBG("%s", hdev->name);
+
+	if (hci_skb_pkt_type(skb) == HCI_ISODATA_PKT) {
+		urb = alloc_mtk_intr_urb(hdev, skb, btusb_tx_complete);
+		if (IS_ERR(urb))
+			return PTR_ERR(urb);
+
+		return submit_or_queue_tx_urb(hdev, urb);
+	} else {
+		return btusb_send_frame(hdev, skb);
+	}
+}
+
 static int btusb_mtk_setup(struct hci_dev *hdev)
 {
 	struct btusb_data *data = hci_get_drvdata(hdev);
@@ -2686,11 +2740,22 @@ static int btusb_mtk_setup(struct hci_dev *hdev)
 	btmtk_data->ctrl_anchor = &data->ctrl_anchor;
 	btmtk_data->reset_sync = btusb_mtk_reset;
 
+	/* Claim ISO data interface and endpoint */
+	btmtk_data->isopkt_intf = usb_ifnum_to_if(data->udev, MTK_ISO_IFNUM);
+	if (btmtk_data->isopkt_intf)
+		btusb_mtk_claim_iso_intf(data);
+
 	return btmtk_usb_setup(hdev);
 }
 
 static int btusb_mtk_shutdown(struct hci_dev *hdev)
 {
+	struct btusb_data *data = hci_get_drvdata(hdev);
+	struct btmtk_data *btmtk_data = hci_get_priv(hdev);
+
+	if (test_bit(BTMTK_ISOPKT_RUNNING, &btmtk_data->flags))
+		btusb_mtk_release_iso_intf(data);
+
 	return btmtk_usb_shutdown(hdev);
 }
 
@@ -3797,9 +3862,12 @@ static int btusb_probe(struct usb_interface *intf,
 		hdev->manufacturer = 70;
 		hdev->cmd_timeout = btmtk_reset_sync;
 		hdev->set_bdaddr = btmtk_set_bdaddr;
+		hdev->send = btusb_send_frame_mtk;
 		set_bit(HCI_QUIRK_BROKEN_ENHANCED_SETUP_SYNC_CONN, &hdev->quirks);
 		set_bit(HCI_QUIRK_NON_PERSISTENT_SETUP, &hdev->quirks);
 		data->recv_acl = btmtk_usb_recv_acl;
+		data->suspend = btmtk_usb_suspend;
+		data->resume = btmtk_usb_resume;
 	}
 
 	if (id->driver_info & BTUSB_SWAVE) {
-- 
cgit v1.2.3-58-ga151


From 880120d5f12e3e9af6472f2ddddb33a6c2648418 Mon Sep 17 00:00:00 2001
From: Kiran K <kiran.k@intel.com>
Date: Fri, 5 Jul 2024 15:59:22 +0530
Subject: Bluetooth: btintel: Add support for BlazarU core

Add support for BlazarU core (CNVi).

Signed-off-by: Kiran K <kiran.k@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btintel.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c
index 54bfe2ffe1d2..e7a612504ab1 100644
--- a/drivers/bluetooth/btintel.c
+++ b/drivers/bluetooth/btintel.c
@@ -472,6 +472,7 @@ int btintel_version_info_tlv(struct hci_dev *hdev,
 	case 0x19:	/* Slr-F */
 	case 0x1b:      /* Mgr */
 	case 0x1c:	/* Gale Peak (GaP) */
+	case 0x1d:	/* BlazarU (BzrU) */
 	case 0x1e:	/* BlazarI (Bzr) */
 		break;
 	default:
@@ -2718,6 +2719,7 @@ void btintel_set_msft_opcode(struct hci_dev *hdev, u8 hw_variant)
 	case 0x19:
 	case 0x1b:
 	case 0x1c:
+	case 0x1d:
 	case 0x1e:
 		hci_set_msft_opcode(hdev, 0xFC1E);
 		break;
@@ -3063,6 +3065,7 @@ static int btintel_setup_combined(struct hci_dev *hdev)
 	case 0x17:
 	case 0x19:
 	case 0x1b:
+	case 0x1d:
 	case 0x1e:
 		/* Display version information of TLV type */
 		btintel_version_info_tlv(hdev, &ver_tlv);
-- 
cgit v1.2.3-58-ga151


From f6dc9214e526c2f735d93b5330f4dcd42a2b5308 Mon Sep 17 00:00:00 2001
From: Kiran K <kiran.k@intel.com>
Date: Fri, 5 Jul 2024 15:59:23 +0530
Subject: Bluetooth: btintel: Add support for Whale Peak2

from /sys/kernel/debug/usb/devices:

T:  Bus=03 Lev=01 Prnt=01 Port=09 Cnt=02 Dev#=  3 Spd=12   MxCh= 0
D:  Ver= 2.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=8087 ProdID=0039 Rev= 0.00
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  64 Ivl=1ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms
I:  If#= 1 Alt= 6 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  63 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  63 Ivl=1ms

Signed-off-by: Kiran K <kiran.k@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btusb.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index 3077f16a149d..acdba5d77694 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -479,6 +479,7 @@ static const struct usb_device_id quirks_table[] = {
 	{ USB_DEVICE(0x8087, 0x0036), .driver_info = BTUSB_INTEL_COMBINED },
 	{ USB_DEVICE(0x8087, 0x0037), .driver_info = BTUSB_INTEL_COMBINED },
 	{ USB_DEVICE(0x8087, 0x0038), .driver_info = BTUSB_INTEL_COMBINED },
+	{ USB_DEVICE(0x8087, 0x0039), .driver_info = BTUSB_INTEL_COMBINED },
 	{ USB_DEVICE(0x8087, 0x07da), .driver_info = BTUSB_CSR },
 	{ USB_DEVICE(0x8087, 0x07dc), .driver_info = BTUSB_INTEL_COMBINED |
 						     BTUSB_INTEL_NO_WBS_SUPPORT |
-- 
cgit v1.2.3-58-ga151


From 358534152c59fdea7731c204d5a91151e4e91e2b Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Thu, 4 Jul 2024 18:50:57 +0530
Subject: Bluetooth: btnxpuart: Add support for AW693 A1 chipset

This adds support for AW693 A1 chipset with it's bootloader signature
and firmware file.

No firmware name change needed for AW693 A0 chipset.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index da18fd4f54f3..e81c995748da 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -46,6 +46,8 @@
 #define FIRMWARE_SECURE_IW624	"uartiw624_bt.bin.se"
 #define FIRMWARE_AW693		"uartaw693_bt.bin"
 #define FIRMWARE_SECURE_AW693	"uartaw693_bt.bin.se"
+#define FIRMWARE_AW693_A1		"uartaw693_bt_v1.bin"
+#define FIRMWARE_SECURE_AW693_A1	"uartaw693_bt_v1.bin.se"
 #define FIRMWARE_HELPER		"helper_uart_3000000.bin"
 
 #define CHIP_ID_W9098		0x5c03
@@ -53,7 +55,8 @@
 #define CHIP_ID_IW612		0x7601
 #define CHIP_ID_IW624a		0x8000
 #define CHIP_ID_IW624c		0x8001
-#define CHIP_ID_AW693		0x8200
+#define CHIP_ID_AW693a0		0x8200
+#define CHIP_ID_AW693a1		0x8201
 
 #define FW_SECURE_MASK		0xc0
 #define FW_OPEN			0x00
@@ -906,7 +909,7 @@ static char *nxp_get_fw_name_from_chipid(struct hci_dev *hdev, u16 chipid,
 		else
 			bt_dev_err(hdev, "Illegal loader version %02x", loader_ver);
 		break;
-	case CHIP_ID_AW693:
+	case CHIP_ID_AW693a0:
 		if ((loader_ver & FW_SECURE_MASK) == FW_OPEN)
 			fw_name = FIRMWARE_AW693;
 		else if ((loader_ver & FW_SECURE_MASK) != FW_AUTH_ILLEGAL)
@@ -914,6 +917,14 @@ static char *nxp_get_fw_name_from_chipid(struct hci_dev *hdev, u16 chipid,
 		else
 			bt_dev_err(hdev, "Illegal loader version %02x", loader_ver);
 		break;
+	case CHIP_ID_AW693a1:
+		if ((loader_ver & FW_SECURE_MASK) == FW_OPEN)
+			fw_name = FIRMWARE_AW693_A1;
+		else if ((loader_ver & FW_SECURE_MASK) != FW_AUTH_ILLEGAL)
+			fw_name = FIRMWARE_SECURE_AW693_A1;
+		else
+			bt_dev_err(hdev, "Illegal loader version %02x", loader_ver);
+		break;
 	default:
 		bt_dev_err(hdev, "Unknown chip signature %04x", chipid);
 		break;
-- 
cgit v1.2.3-58-ga151


From e50bfd6bb231a6e2b7221ad78dce294330238c76 Mon Sep 17 00:00:00 2001
From: Kory Maincent <kory.maincent@bootlin.com>
Date: Tue, 9 Jul 2024 15:53:33 +0200
Subject: net_tstamp: Add TIMESTAMPING SOFTWARE and HARDWARE mask

Timestamping software or hardware flags are often used as a group,
therefore adding these masks will ease future use.

I did not use SOF_TIMESTAMPING_SYS_HARDWARE flag as it is deprecated and
not used at all.

Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240709-feature_ptp_netnext-v17-1-b5317f50df2a@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/net_tstamp.h | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/include/linux/net_tstamp.h b/include/linux/net_tstamp.h
index eb01c37e71e0..3799c79b6c83 100644
--- a/include/linux/net_tstamp.h
+++ b/include/linux/net_tstamp.h
@@ -5,6 +5,14 @@
 
 #include <uapi/linux/net_tstamp.h>
 
+#define SOF_TIMESTAMPING_SOFTWARE_MASK	(SOF_TIMESTAMPING_RX_SOFTWARE | \
+					 SOF_TIMESTAMPING_TX_SOFTWARE | \
+					 SOF_TIMESTAMPING_SOFTWARE)
+
+#define SOF_TIMESTAMPING_HARDWARE_MASK	(SOF_TIMESTAMPING_RX_HARDWARE | \
+					 SOF_TIMESTAMPING_TX_HARDWARE | \
+					 SOF_TIMESTAMPING_RAW_HARDWARE)
+
 enum hwtstamp_source {
 	HWTSTAMP_SOURCE_NETDEV,
 	HWTSTAMP_SOURCE_PHYLIB,
-- 
cgit v1.2.3-58-ga151


From d4f6acfcb0b0962e2a59165108e212a976d1f078 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Thu, 4 Jul 2024 18:50:58 +0530
Subject: Bluetooth: btnxpuart: Add support for IW615 chipset

This adds support for IW615 chipset with it's bootloader signature
and firmware file.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index e81c995748da..e9c2f2f7acc6 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -42,6 +42,8 @@
 #define FIRMWARE_W9098_OLD	"uartuart9098_bt_v1.bin"
 #define FIRMWARE_IW416		"uartiw416_bt_v0.bin"
 #define FIRMWARE_IW612		"uartspi_n61x_v1.bin.se"
+#define FIRMWARE_IW615		"uartspi_iw610_v0.bin"
+#define FIRMWARE_SECURE_IW615	"uartspi_iw610_v0.bin.se"
 #define FIRMWARE_IW624		"uartiw624_bt.bin"
 #define FIRMWARE_SECURE_IW624	"uartiw624_bt.bin.se"
 #define FIRMWARE_AW693		"uartaw693_bt.bin"
@@ -57,6 +59,8 @@
 #define CHIP_ID_IW624c		0x8001
 #define CHIP_ID_AW693a0		0x8200
 #define CHIP_ID_AW693a1		0x8201
+#define CHIP_ID_IW615a0		0x8800
+#define CHIP_ID_IW615a1		0x8801
 
 #define FW_SECURE_MASK		0xc0
 #define FW_OPEN			0x00
@@ -925,6 +929,15 @@ static char *nxp_get_fw_name_from_chipid(struct hci_dev *hdev, u16 chipid,
 		else
 			bt_dev_err(hdev, "Illegal loader version %02x", loader_ver);
 		break;
+	case CHIP_ID_IW615a0:
+	case CHIP_ID_IW615a1:
+		if ((loader_ver & FW_SECURE_MASK) == FW_OPEN)
+			fw_name = FIRMWARE_IW615;
+		else if ((loader_ver & FW_SECURE_MASK) != FW_AUTH_ILLEGAL)
+			fw_name = FIRMWARE_SECURE_IW615;
+		else
+			bt_dev_err(hdev, "Illegal loader version %02x", loader_ver);
+		break;
 	default:
 		bt_dev_err(hdev, "Unknown chip signature %04x", chipid);
 		break;
-- 
cgit v1.2.3-58-ga151


From 658c56d5b3b3a41e52ea4236f5dd023cbb139ece Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Fri, 5 Jul 2024 12:28:26 +0530
Subject: Bluetooth: btnxpuart: Add system suspend and resume handlers

This adds handling for system suspend and resume. While the host enters
suspend state, the driver will drive the chip into low power state.

Similarly when system is resuming, the driver will wake up the chip.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index e9c2f2f7acc6..f75b24bd3045 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -1522,6 +1522,24 @@ static void nxp_serdev_remove(struct serdev_device *serdev)
 	hci_free_dev(hdev);
 }
 
+static int nxp_serdev_suspend(struct device *dev)
+{
+	struct btnxpuart_dev *nxpdev = dev_get_drvdata(dev);
+	struct ps_data *psdata = &nxpdev->psdata;
+
+	ps_control(psdata->hdev, PS_STATE_SLEEP);
+	return 0;
+}
+
+static int nxp_serdev_resume(struct device *dev)
+{
+	struct btnxpuart_dev *nxpdev = dev_get_drvdata(dev);
+	struct ps_data *psdata = &nxpdev->psdata;
+
+	ps_control(psdata->hdev, PS_STATE_AWAKE);
+	return 0;
+}
+
 static struct btnxpuart_data w8987_data __maybe_unused = {
 	.helper_fw_name = NULL,
 	.fw_name = FIRMWARE_W8987,
@@ -1541,12 +1559,17 @@ static const struct of_device_id nxpuart_of_match_table[] __maybe_unused = {
 };
 MODULE_DEVICE_TABLE(of, nxpuart_of_match_table);
 
+static const struct dev_pm_ops nxp_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(nxp_serdev_suspend, nxp_serdev_resume)
+};
+
 static struct serdev_device_driver nxp_serdev_driver = {
 	.probe = nxp_serdev_probe,
 	.remove = nxp_serdev_remove,
 	.driver = {
 		.name = "btnxpuart",
 		.of_match_table = of_match_ptr(nxpuart_of_match_table),
+		.pm = &nxp_pm_ops,
 	},
 };
 
-- 
cgit v1.2.3-58-ga151


From 3aa7901e44bb2b6f6d3a14ad5a47e07ec81dcd47 Mon Sep 17 00:00:00 2001
From: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Date: Wed, 10 Jul 2024 18:04:26 +0530
Subject: Bluetooth: btnxpuart: Fix warnings for suspend and resume functions

This fixes the warnings 'nxp_serdev_resume' and 'nxp_serdev_suspend'
defined but not used.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202407100518.06Xxq5PK-lkp@intel.com/
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btnxpuart.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c
index f75b24bd3045..31d3dd90b672 100644
--- a/drivers/bluetooth/btnxpuart.c
+++ b/drivers/bluetooth/btnxpuart.c
@@ -1522,6 +1522,7 @@ static void nxp_serdev_remove(struct serdev_device *serdev)
 	hci_free_dev(hdev);
 }
 
+#ifdef CONFIG_PM_SLEEP
 static int nxp_serdev_suspend(struct device *dev)
 {
 	struct btnxpuart_dev *nxpdev = dev_get_drvdata(dev);
@@ -1539,6 +1540,7 @@ static int nxp_serdev_resume(struct device *dev)
 	ps_control(psdata->hdev, PS_STATE_AWAKE);
 	return 0;
 }
+#endif
 
 static struct btnxpuart_data w8987_data __maybe_unused = {
 	.helper_fw_name = NULL,
-- 
cgit v1.2.3-58-ga151


From bd3f305886adc07c27b8e998719c73688dad0991 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Tue, 9 Jul 2024 14:18:32 +0200
Subject: dt-bindings: bluetooth: qualcomm: describe the inputs from PMU for
 wcn7850

Drop the inputs from the host and instead expect the Bluetooth node to
consume the outputs of the internal PMU. This model is closer to reality
than how we represent it now.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 .../bindings/net/bluetooth/qualcomm-bluetooth.yaml     | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)

diff --git a/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml b/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml
index 48ac9f10ef05..68c5ed111417 100644
--- a/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml
+++ b/Documentation/devicetree/bindings/net/bluetooth/qualcomm-bluetooth.yaml
@@ -77,6 +77,9 @@ properties:
   vddrfa1p7-supply:
     description: VDD_RFA_1P7 supply regulator handle
 
+  vddrfa1p8-supply:
+    description: VDD_RFA_1P8 supply regulator handle
+
   vddrfa1p2-supply:
     description: VDD_RFA_1P2 supply regulator handle
 
@@ -89,6 +92,12 @@ properties:
   vddasd-supply:
     description: VDD_ASD supply regulator handle
 
+  vddwlcx-supply:
+    description: VDD_WLCX supply regulator handle
+
+  vddwlmx-supply:
+    description: VDD_WLMX supply regulator handle
+
   max-speed:
     description: see Documentation/devicetree/bindings/serial/serial.yaml
 
@@ -179,14 +188,13 @@ allOf:
               - qcom,wcn7850-bt
     then:
       required:
-        - enable-gpios
-        - swctrl-gpios
-        - vddio-supply
+        - vddrfacmn-supply
         - vddaon-supply
-        - vdddig-supply
+        - vddwlcx-supply
+        - vddwlmx-supply
         - vddrfa0p8-supply
         - vddrfa1p2-supply
-        - vddrfa1p9-supply
+        - vddrfa1p8-supply
   - if:
       properties:
         compatible:
-- 
cgit v1.2.3-58-ga151


From b80f4e3b8cb8a4a41cc038807cd0a1e889154902 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Tue, 9 Jul 2024 14:18:33 +0200
Subject: Bluetooth: hci_qca: schedule a devm action for disabling the clock

In preparation for unduplicating the hci_uart registration code,
schedule a devres action for disabling the SUSCLK rather than doing it
manually.

We cannot really use devm_clk_get_enabled() as we also set the rate
before enabling the clock. While this should in theory work, I don't
want to risk breaking existing users. One solution for the future is to
add devm_clk_get_enabled_with_rate() to the clock framework.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_qca.c | 23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index bc6a49ba26f9..ace920261aa4 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -2295,6 +2295,13 @@ static int qca_init_regulators(struct qca_power *qca,
 	return 0;
 }
 
+static void qca_clk_disable_unprepare(void *data)
+{
+	struct clk *clk = data;
+
+	clk_disable_unprepare(clk);
+}
+
 static int qca_serdev_probe(struct serdev_device *serdev)
 {
 	struct qca_serdev *qcadev;
@@ -2434,10 +2441,15 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 		if (err)
 			return err;
 
+		err = devm_add_action_or_reset(&serdev->dev,
+					       qca_clk_disable_unprepare,
+					       qcadev->susclk);
+		if (err)
+			return err;
+
 		err = hci_uart_register_device(&qcadev->serdev_hu, &qca_proto);
 		if (err) {
 			BT_ERR("Rome serdev registration failed");
-			clk_disable_unprepare(qcadev->susclk);
 			return err;
 		}
 	}
@@ -2477,15 +2489,10 @@ static void qca_serdev_remove(struct serdev_device *serdev)
 	case QCA_WCN6750:
 	case QCA_WCN6855:
 	case QCA_WCN7850:
-		if (power->vregs_on) {
+		if (power->vregs_on)
 			qca_power_shutdown(&qcadev->serdev_hu);
-			break;
-		}
-		fallthrough;
-
+		break;
 	default:
-		if (qcadev->susclk)
-			clk_disable_unprepare(qcadev->susclk);
 	}
 
 	hci_uart_unregister_device(&qcadev->serdev_hu);
-- 
cgit v1.2.3-58-ga151


From b8093157714fb0e1475909d744007d102efeaff1 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Tue, 9 Jul 2024 14:18:34 +0200
Subject: Bluetooth: hci_qca: unduplicate calls to hci_uart_register_device()

Now that all three branches of the switch end up doing the same thing,
we can move the call to hci_uart_register_device() past it and unify the
error message.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_qca.c | 23 ++++++-----------------
 1 file changed, 6 insertions(+), 17 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index ace920261aa4..9d2b50c8fc93 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -2396,12 +2396,6 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 			dev_err(&serdev->dev, "failed to acquire clk\n");
 			return PTR_ERR(qcadev->susclk);
 		}
-
-		err = hci_uart_register_device(&qcadev->serdev_hu, &qca_proto);
-		if (err) {
-			BT_ERR("wcn3990 serdev registration failed");
-			return err;
-		}
 		break;
 
 	case QCA_QCA6390:
@@ -2409,12 +2403,6 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 							   "bluetooth");
 		if (IS_ERR(qcadev->bt_power->pwrseq))
 			return PTR_ERR(qcadev->bt_power->pwrseq);
-
-		err = hci_uart_register_device(&qcadev->serdev_hu, &qca_proto);
-		if (err) {
-			BT_ERR("qca6390 serdev registration failed");
-			return err;
-		}
 		break;
 
 	default:
@@ -2447,11 +2435,12 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 		if (err)
 			return err;
 
-		err = hci_uart_register_device(&qcadev->serdev_hu, &qca_proto);
-		if (err) {
-			BT_ERR("Rome serdev registration failed");
-			return err;
-		}
+	}
+	
+	err = hci_uart_register_device(&qcadev->serdev_hu, &qca_proto);
+	if (err) {
+		BT_ERR("serdev registration failed");
+		return err;
 	}
 
 	hdev = qcadev->serdev_hu.hdev;
-- 
cgit v1.2.3-58-ga151


From eba1718717b097d825f26a33ae7d8a5b9d478396 Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Tue, 9 Jul 2024 14:18:35 +0200
Subject: Bluetooth: hci_qca: make pwrseq calls the default if available

If the device has a power sequencing handle, use it first. Otherwise
fall back to whatever code already exists.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_qca.c | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 9d2b50c8fc93..49588072589e 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -1718,6 +1718,7 @@ static int qca_regulator_init(struct hci_uart *hu)
 	 * off the voltage regulator.
 	 */
 	qcadev = serdev_device_get_drvdata(hu->serdev);
+
 	if (!qcadev->bt_power->vregs_on) {
 		serdev_device_close(hu->serdev);
 		ret = qca_regulator_enable(qcadev);
@@ -1800,18 +1801,8 @@ static int qca_power_on(struct hci_dev *hdev)
 	case QCA_WCN6750:
 	case QCA_WCN6855:
 	case QCA_WCN7850:
-		ret = qca_regulator_init(hu);
-		break;
-
 	case QCA_QCA6390:
-		qcadev = serdev_device_get_drvdata(hu->serdev);
-		ret = pwrseq_power_on(qcadev->bt_power->pwrseq);
-		if (ret)
-			return ret;
-
-		ret = qca_port_reopen(hu);
-		if (ret)
-			return ret;
+		ret = qca_regulator_init(hu);
 		break;
 
 	default:
@@ -2149,6 +2140,7 @@ static void qca_power_shutdown(struct hci_uart *hu)
 	unsigned long flags;
 	enum qca_btsoc_type soc_type = qca_soc_type(hu);
 	bool sw_ctrl_state;
+	struct qca_power *power;
 
 	/* From this point we go into power off state. But serial port is
 	 * still open, stop queueing the IBS data and flush all the buffered
@@ -2166,6 +2158,13 @@ static void qca_power_shutdown(struct hci_uart *hu)
 		return;
 
 	qcadev = serdev_device_get_drvdata(hu->serdev);
+	power = qcadev->bt_power;
+
+	if (power->pwrseq) {
+		pwrseq_power_off(power->pwrseq);
+		set_bit(QCA_BT_OFF, &qca->flags);
+		return;
+        }
 
 	switch (soc_type) {
 	case QCA_WCN3988:
@@ -2227,6 +2226,9 @@ static int qca_regulator_enable(struct qca_serdev *qcadev)
 	struct qca_power *power = qcadev->bt_power;
 	int ret;
 
+	if (power->pwrseq)
+		return pwrseq_power_on(power->pwrseq);
+
 	/* Already enabled */
 	if (power->vregs_on)
 		return 0;
-- 
cgit v1.2.3-58-ga151


From 921ad2879aa7733620585e4904ba673c6c916dab Mon Sep 17 00:00:00 2001
From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Date: Tue, 9 Jul 2024 14:18:36 +0200
Subject: Bluetooth: hci_qca: use the power sequencer for wcn7850 and wcn6855

In preparation for using the power sequencing subsystem on sm8650 boards
and X13s laptop let's make pwrseq the default for wcn7850 and wcn6855.

Both these models require an enable GPIO so we can safely assume that if
the property is not there, then we should try to get the power
sequencer. Due to how the pwrseq lookup works - checking the provider at
run-time - we cannot really do it the other way around as we'd get stuck
forever on -EPROBE_DEFER.

If the relevant OF node does have the 'enable-gpios' property, we
fallback to the existing code for backward compatibility with older DTs.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_qca.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 49588072589e..030153d468bf 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -2354,13 +2354,28 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 	}
 
 	switch (qcadev->btsoc_type) {
+	case QCA_WCN6855:
+	case QCA_WCN7850:
+		if (!device_property_present(&serdev->dev, "enable-gpios")) {
+			/*
+			 * Backward compatibility with old DT sources. If the
+			 * node doesn't have the 'enable-gpios' property then
+			 * let's use the power sequencer. Otherwise, let's
+			 * drive everything outselves.
+			 */
+			qcadev->bt_power->pwrseq = devm_pwrseq_get(&serdev->dev,
+								   "bluetooth");
+			if (IS_ERR(qcadev->bt_power->pwrseq))
+				return PTR_ERR(qcadev->bt_power->pwrseq);
+
+			break;
+		}
+		fallthrough;
 	case QCA_WCN3988:
 	case QCA_WCN3990:
 	case QCA_WCN3991:
 	case QCA_WCN3998:
 	case QCA_WCN6750:
-	case QCA_WCN6855:
-	case QCA_WCN7850:
 		qcadev->bt_power->dev = &serdev->dev;
 		err = qca_init_regulators(qcadev->bt_power, data->vregs,
 					  data->num_vregs);
-- 
cgit v1.2.3-58-ga151


From 84f9288208dfcd955446060a53add15b9e01af4e Mon Sep 17 00:00:00 2001
From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Date: Wed, 10 Jul 2024 22:30:57 -0400
Subject: Bluetooth: hci_qca: Fix build error
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This fixes the following build error introduced by b80f4e3b8cb8
("Bluetooth: hci_qca: schedule a devm action for disabling the clock"):

drivers/bluetooth/hci_qca.c: In function ‘qca_serdev_remove’:
drivers/bluetooth/hci_qca.c:2501:2: error: label at end of compound statement
 2501 |  default:
      |  ^~~~~~~

Fixes: b80f4e3b8cb8 ("Bluetooth: hci_qca: schedule a devm action for disabling the clock")
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/hci_qca.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c
index 030153d468bf..ca6466676902 100644
--- a/drivers/bluetooth/hci_qca.c
+++ b/drivers/bluetooth/hci_qca.c
@@ -2499,6 +2499,7 @@ static void qca_serdev_remove(struct serdev_device *serdev)
 			qca_power_shutdown(&qcadev->serdev_hu);
 		break;
 	default:
+		break;
 	}
 
 	hci_uart_unregister_device(&qcadev->serdev_hu);
-- 
cgit v1.2.3-58-ga151


From 23e88450bb04f6b44ce4cacbd76cd775e426a5b2 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Wed, 10 Jul 2024 10:56:24 -0700
Subject: Bluetooth: btmtk: Mark all stub functions as inline

Several recent patches added static stubs to btmtk.h without the inline
keyword, which causes instances of -Wunused-function when those stubs
are not used anywhere in a file that includes the header:

  In file included from drivers/bluetooth/btusb.c:28:
  drivers/bluetooth/btmtk.h:254:13: warning: 'btmtk_fw_get_filename' defined but not used [-Wunused-function]
    254 | static void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id,
        |             ^~~~~~~~~~~~~~~~~~~~~
  drivers/bluetooth/btmtk.h:249:12: warning: 'btmtk_process_coredump' defined but not used [-Wunused-function]
    249 | static int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb)
        |            ^~~~~~~~~~~~~~~~~~~~~~
  drivers/bluetooth/btmtk.h:243:12: warning: 'btmtk_register_coredump' defined but not used [-Wunused-function]
    243 | static int btmtk_register_coredump(struct hci_dev *hdev, const char *name,
        |            ^~~~~~~~~~~~~~~~~~~~~~~
  drivers/bluetooth/btmtk.h:233:12: warning: 'btmtk_setup_firmware' defined but not used [-Wunused-function]
    233 | static int btmtk_setup_firmware(struct hci_dev *hdev, const char *fwname,
        |            ^~~~~~~~~~~~~~~~~~~~
  drivers/bluetooth/btmtk.h:227:12: warning: 'btmtk_setup_firmware_79xx' defined but not used [-Wunused-function]
    227 | static int btmtk_setup_firmware_79xx(struct hci_dev *hdev, const char *fwname,
        |            ^~~~~~~~~~~~~~~~~~~~~~~~~

Add inline to all the stubs in btmtk.h (even ones that do not currently
have any warnings associated with them) to ensure there are never unused
function warnings from these stubs, as is customary for the kernel.

Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 drivers/bluetooth/btmtk.h | 39 +++++++++++++++++++++------------------
 1 file changed, 21 insertions(+), 18 deletions(-)

diff --git a/drivers/bluetooth/btmtk.h b/drivers/bluetooth/btmtk.h
index 453ed5131a37..5df7c3296624 100644
--- a/drivers/bluetooth/btmtk.h
+++ b/drivers/bluetooth/btmtk.h
@@ -224,70 +224,73 @@ static inline int btmtk_set_bdaddr(struct hci_dev *hdev,
 	return -EOPNOTSUPP;
 }
 
-static int btmtk_setup_firmware_79xx(struct hci_dev *hdev, const char *fwname,
-				     wmt_cmd_sync_func_t wmt_cmd_sync)
+static inline int btmtk_setup_firmware_79xx(struct hci_dev *hdev,
+					    const char *fwname,
+					    wmt_cmd_sync_func_t wmt_cmd_sync)
 {
 	return -EOPNOTSUPP;
 }
 
-static int btmtk_setup_firmware(struct hci_dev *hdev, const char *fwname,
-				wmt_cmd_sync_func_t wmt_cmd_sync)
+static inline int btmtk_setup_firmware(struct hci_dev *hdev, const char *fwname,
+				       wmt_cmd_sync_func_t wmt_cmd_sync)
 {
 	return -EOPNOTSUPP;
 }
 
-static void btmtk_reset_sync(struct hci_dev *hdev)
+static inline void btmtk_reset_sync(struct hci_dev *hdev)
 {
 }
 
-static int btmtk_register_coredump(struct hci_dev *hdev, const char *name,
-				   u32 fw_version)
+static inline int btmtk_register_coredump(struct hci_dev *hdev,
+					  const char *name, u32 fw_version)
 {
 	return -EOPNOTSUPP;
 }
 
-static int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb)
+static inline int btmtk_process_coredump(struct hci_dev *hdev,
+					 struct sk_buff *skb)
 {
 	return -EOPNOTSUPP;
 }
 
-static void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id,
-				  u32 fw_ver, u32 fw_flavor)
+static inline void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id,
+					 u32 fw_ver, u32 fw_flavor)
 {
 }
 
-static int btmtk_usb_subsys_reset(struct hci_dev *hdev, u32 dev_id)
+static inline int btmtk_usb_subsys_reset(struct hci_dev *hdev, u32 dev_id)
 {
 	return -EOPNOTSUPP;
 }
 
-static int btmtk_usb_recv_acl(struct hci_dev *hdev, struct sk_buff *skb)
+static inline int btmtk_usb_recv_acl(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	return -EOPNOTSUPP;
 }
 
-static struct urb *alloc_mtk_intr_urb(struct hci_dev *hdev, struct sk_buff *skb,
-				      usb_complete_t tx_complete)
+static inline struct urb *alloc_mtk_intr_urb(struct hci_dev *hdev,
+					     struct sk_buff *skb,
+					     usb_complete_t tx_complete)
 {
 	return ERR_PTR(-EOPNOTSUPP);
 }
 
-static int btmtk_usb_resume(struct hci_dev *hdev)
+static inline int btmtk_usb_resume(struct hci_dev *hdev)
 {
 	return -EOPNOTSUPP;
 }
 
-static int btmtk_usb_suspend(struct hci_dev *hdev)
+static inline int btmtk_usb_suspend(struct hci_dev *hdev)
 {
 	return -EOPNOTSUPP;
 }
 
-static int btmtk_usb_setup(struct hci_dev *hdev)
+static inline int btmtk_usb_setup(struct hci_dev *hdev)
 {
 	return -EOPNOTSUPP;
 }
 
-static int btmtk_usb_shutdown(struct hci_dev *hdev)
+static inline int btmtk_usb_shutdown(struct hci_dev *hdev)
 {
 	return -EOPNOTSUPP;
 }
-- 
cgit v1.2.3-58-ga151


From 2dd35600590148d843367c04975acad3c1a527c3 Mon Sep 17 00:00:00 2001
From: Kory Maincent <kory.maincent@bootlin.com>
Date: Tue, 9 Jul 2024 15:53:36 +0200
Subject: net: Change the API of PHY default timestamp to MAC

Change the API to select MAC default time stamping instead of the PHY.
Indeed the PHY is closer to the wire therefore theoretically it has less
delay than the MAC timestamping but the reality is different. Due to lower
time stamping clock frequency, latency in the MDIO bus and no PHC hardware
synchronization between different PHY, the PHY PTP is often less precise
than the MAC. The exception is for PHY designed specially for PTP case but
these devices are not very widespread. For not breaking the compatibility
default_timestamp flag has been introduced in phy_device that is set by
the phy driver to know we are using the old API behavior.

Reviewed-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240709-feature_ptp_netnext-v17-4-b5317f50df2a@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/phy/bcm-phy-ptp.c     |  3 +++
 drivers/net/phy/dp83640.c         |  2 ++
 drivers/net/phy/micrel.c          |  6 ++++++
 drivers/net/phy/mscc/mscc_ptp.c   |  3 +++
 drivers/net/phy/nxp-c45-tja11xx.c |  3 +++
 include/linux/phy.h               | 19 +++++++++++++++++++
 net/core/dev_ioctl.c              |  8 +++-----
 net/core/timestamping.c           |  5 +++--
 net/ethtool/common.c              |  2 +-
 9 files changed, 43 insertions(+), 8 deletions(-)

diff --git a/drivers/net/phy/bcm-phy-ptp.c b/drivers/net/phy/bcm-phy-ptp.c
index 617d384d4551..d3e825c951ee 100644
--- a/drivers/net/phy/bcm-phy-ptp.c
+++ b/drivers/net/phy/bcm-phy-ptp.c
@@ -931,6 +931,9 @@ struct bcm_ptp_private *bcm_ptp_probe(struct phy_device *phydev)
 		return ERR_CAST(clock);
 	priv->ptp_clock = clock;
 
+	/* Timestamp selected by default to keep legacy API */
+	phydev->default_timestamp = true;
+
 	priv->phydev = phydev;
 	bcm_ptp_init(priv);
 
diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c
index 5c42c47dc564..d3e72d5c1472 100644
--- a/drivers/net/phy/dp83640.c
+++ b/drivers/net/phy/dp83640.c
@@ -1447,6 +1447,8 @@ static int dp83640_probe(struct phy_device *phydev)
 	for (i = 0; i < MAX_RXTS; i++)
 		list_add(&dp83640->rx_pool_data[i].list, &dp83640->rxpool);
 
+	/* Timestamp selected by default to keep legacy API */
+	phydev->default_timestamp = true;
 	phydev->mii_ts = &dp83640->mii_ts;
 	phydev->priv = dp83640;
 
diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index ebafedde0ab7..8d57225d8575 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -3781,6 +3781,9 @@ static void lan8814_ptp_init(struct phy_device *phydev)
 	ptp_priv->mii_ts.ts_info  = lan8814_ts_info;
 
 	phydev->mii_ts = &ptp_priv->mii_ts;
+
+	/* Timestamp selected by default to keep legacy API */
+	phydev->default_timestamp = true;
 }
 
 static int lan8814_ptp_probe_once(struct phy_device *phydev)
@@ -5279,6 +5282,9 @@ static int lan8841_probe(struct phy_device *phydev)
 
 	phydev->mii_ts = &ptp_priv->mii_ts;
 
+	/* Timestamp selected by default to keep legacy API */
+	phydev->default_timestamp = true;
+
 	return 0;
 }
 
diff --git a/drivers/net/phy/mscc/mscc_ptp.c b/drivers/net/phy/mscc/mscc_ptp.c
index eb0b032cb613..e66d20eff7c4 100644
--- a/drivers/net/phy/mscc/mscc_ptp.c
+++ b/drivers/net/phy/mscc/mscc_ptp.c
@@ -1570,6 +1570,9 @@ int vsc8584_ptp_probe(struct phy_device *phydev)
 		return PTR_ERR(vsc8531->load_save);
 	}
 
+	/* Timestamp selected by default to keep legacy API */
+	phydev->default_timestamp = true;
+
 	vsc8531->ptp->phydev = phydev;
 
 	return 0;
diff --git a/drivers/net/phy/nxp-c45-tja11xx.c b/drivers/net/phy/nxp-c45-tja11xx.c
index 3cf614b4cd52..d18c133e6013 100644
--- a/drivers/net/phy/nxp-c45-tja11xx.c
+++ b/drivers/net/phy/nxp-c45-tja11xx.c
@@ -1660,6 +1660,9 @@ static int nxp_c45_probe(struct phy_device *phydev)
 		priv->mii_ts.ts_info = nxp_c45_ts_info;
 		phydev->mii_ts = &priv->mii_ts;
 		ret = nxp_c45_init_ptp_clock(priv);
+
+		/* Timestamp selected by default to keep legacy API */
+		phydev->default_timestamp = true;
 	} else {
 		phydev_dbg(phydev, "PTP support not enabled even if the phy supports it");
 	}
diff --git a/include/linux/phy.h b/include/linux/phy.h
index bd68f9d8e74f..e7a38137211c 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -616,6 +616,8 @@ struct macsec_ops;
  *                 handling shall be postponed until PHY has resumed
  * @irq_rerun: Flag indicating interrupts occurred while PHY was suspended,
  *             requiring a rerun of the interrupt handler after resume
+ * @default_timestamp: Flag indicating whether we are using the phy
+ *		       timestamp as the default one
  * @interface: enum phy_interface_t value
  * @possible_interfaces: bitmap if interface modes that the attached PHY
  *			 will switch between depending on media speed.
@@ -681,6 +683,8 @@ struct phy_device {
 	unsigned irq_suspended:1;
 	unsigned irq_rerun:1;
 
+	unsigned default_timestamp:1;
+
 	int rate_matching;
 
 	enum phy_state state;
@@ -1625,6 +1629,21 @@ static inline void phy_txtstamp(struct phy_device *phydev, struct sk_buff *skb,
 	phydev->mii_ts->txtstamp(phydev->mii_ts, skb, type);
 }
 
+/**
+ * phy_is_default_hwtstamp - Is the PHY hwtstamp the default timestamp
+ * @phydev: Pointer to phy_device
+ *
+ * This is used to get default timestamping device taking into account
+ * the new API choice, which is selecting the timestamping from MAC by
+ * default if the phydev does not have default_timestamp flag enabled.
+ *
+ * Return: True if phy is the default hw timestamp, false otherwise.
+ */
+static inline bool phy_is_default_hwtstamp(struct phy_device *phydev)
+{
+	return phy_has_hwtstamp(phydev) && phydev->default_timestamp;
+}
+
 /**
  * phy_is_internal - Convenience function for testing if a PHY is internal
  * @phydev: the phy_device struct
diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index b9719ed3c3fd..8592c052c0f4 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -259,9 +259,7 @@ static int dev_eth_ioctl(struct net_device *dev,
  * @dev: Network device
  * @cfg: Timestamping configuration structure
  *
- * Helper for enforcing a common policy that phylib timestamping, if available,
- * should take precedence in front of hardware timestamping provided by the
- * netdev.
+ * Helper for calling the default hardware provider timestamping.
  *
  * Note: phy_mii_ioctl() only handles SIOCSHWTSTAMP (not SIOCGHWTSTAMP), and
  * there only exists a phydev->mii_ts->hwtstamp() method. So this will return
@@ -271,7 +269,7 @@ static int dev_eth_ioctl(struct net_device *dev,
 static int dev_get_hwtstamp_phylib(struct net_device *dev,
 				   struct kernel_hwtstamp_config *cfg)
 {
-	if (phy_has_hwtstamp(dev->phydev))
+	if (phy_is_default_hwtstamp(dev->phydev))
 		return phy_hwtstamp_get(dev->phydev, cfg);
 
 	return dev->netdev_ops->ndo_hwtstamp_get(dev, cfg);
@@ -327,7 +325,7 @@ int dev_set_hwtstamp_phylib(struct net_device *dev,
 			    struct netlink_ext_ack *extack)
 {
 	const struct net_device_ops *ops = dev->netdev_ops;
-	bool phy_ts = phy_has_hwtstamp(dev->phydev);
+	bool phy_ts = phy_is_default_hwtstamp(dev->phydev);
 	struct kernel_hwtstamp_config old_cfg = {};
 	bool changed = false;
 	int err;
diff --git a/net/core/timestamping.c b/net/core/timestamping.c
index 04840697fe79..3717fb152ecc 100644
--- a/net/core/timestamping.c
+++ b/net/core/timestamping.c
@@ -25,7 +25,8 @@ void skb_clone_tx_timestamp(struct sk_buff *skb)
 	struct sk_buff *clone;
 	unsigned int type;
 
-	if (!skb->sk)
+	if (!skb->sk || !skb->dev ||
+	    !phy_is_default_hwtstamp(skb->dev->phydev))
 		return;
 
 	type = classify(skb);
@@ -47,7 +48,7 @@ bool skb_defer_rx_timestamp(struct sk_buff *skb)
 	struct mii_timestamper *mii_ts;
 	unsigned int type;
 
-	if (!skb->dev || !skb->dev->phydev || !skb->dev->phydev->mii_ts)
+	if (!skb->dev || !phy_is_default_hwtstamp(skb->dev->phydev))
 		return false;
 
 	if (skb_headroom(skb) < ETH_HLEN)
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 6b2a360dcdf0..01b7550f12c6 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -637,7 +637,7 @@ int __ethtool_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
 	memset(info, 0, sizeof(*info));
 	info->cmd = ETHTOOL_GET_TS_INFO;
 
-	if (phy_has_tsinfo(phydev))
+	if (phy_is_default_hwtstamp(phydev) && phy_has_tsinfo(phydev))
 		return phy_ts_info(phydev, info);
 	if (ops->get_ts_info)
 		return ops->get_ts_info(dev, info);
-- 
cgit v1.2.3-58-ga151


From bc5a07ed15a3d30df3e8af2ef6bb0472f1d337f9 Mon Sep 17 00:00:00 2001
From: Kory Maincent <kory.maincent@bootlin.com>
Date: Tue, 9 Jul 2024 15:53:37 +0200
Subject: net: net_tstamp: Add unspec field to hwtstamp_source enumeration

Prepare for future support of saving hwtstamp source in PTP xarray by
introducing HWTSTAMP_SOURCE_UNSPEC to hwtstamp_source enum, setting it
to 0 to match old behavior of no source defined.

Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240709-feature_ptp_netnext-v17-5-b5317f50df2a@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/net_tstamp.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/linux/net_tstamp.h b/include/linux/net_tstamp.h
index 3799c79b6c83..662074b08c94 100644
--- a/include/linux/net_tstamp.h
+++ b/include/linux/net_tstamp.h
@@ -14,6 +14,7 @@
 					 SOF_TIMESTAMPING_RAW_HARDWARE)
 
 enum hwtstamp_source {
+	HWTSTAMP_SOURCE_UNSPEC,
 	HWTSTAMP_SOURCE_NETDEV,
 	HWTSTAMP_SOURCE_PHYLIB,
 };
-- 
cgit v1.2.3-58-ga151


From 2111375b85ad173d58e7b8604246a3de60950ac8 Mon Sep 17 00:00:00 2001
From: Kory Maincent <kory.maincent@bootlin.com>
Date: Tue, 9 Jul 2024 15:53:38 +0200
Subject: net: Add struct kernel_ethtool_ts_info

In prevision to add new UAPI for hwtstamp we will be limited to the struct
ethtool_ts_info that is currently passed in fixed binary format through the
ETHTOOL_GET_TS_INFO ethtool ioctl. It would be good if new kernel code
already started operating on an extensible kernel variant of that
structure, similar in concept to struct kernel_hwtstamp_config vs struct
hwtstamp_config.

Since struct ethtool_ts_info is in include/uapi/linux/ethtool.h, here
we introduce the kernel-only structure in include/linux/ethtool.h.
The manual copy is then made in the function called by ETHTOOL_GET_TS_INFO.

Acked-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240709-feature_ptp_netnext-v17-6-b5317f50df2a@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/bonding/bond_main.c                    |  4 ++--
 drivers/net/can/dev/dev.c                          |  2 +-
 drivers/net/can/peak_canfd/peak_canfd.c            |  2 +-
 drivers/net/can/usb/gs_usb.c                       |  2 +-
 drivers/net/can/usb/peak_usb/pcan_usb_core.c       |  2 +-
 drivers/net/can/usb/peak_usb/pcan_usb_core.h       |  2 +-
 drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c    |  2 +-
 drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h    |  2 +-
 drivers/net/dsa/microchip/ksz_ptp.c                |  2 +-
 drivers/net/dsa/microchip/ksz_ptp.h                |  2 +-
 drivers/net/dsa/mv88e6xxx/hwtstamp.c               |  2 +-
 drivers/net/dsa/mv88e6xxx/hwtstamp.h               |  4 ++--
 drivers/net/dsa/ocelot/felix.c                     |  2 +-
 drivers/net/dsa/sja1105/sja1105_ptp.c              |  2 +-
 drivers/net/dsa/sja1105/sja1105_ptp.h              |  2 +-
 drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c       |  2 +-
 .../net/ethernet/aquantia/atlantic/aq_ethtool.c    |  2 +-
 .../net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c    |  2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c  |  2 +-
 drivers/net/ethernet/broadcom/tg3.c                |  2 +-
 drivers/net/ethernet/cadence/macb.h                |  2 +-
 drivers/net/ethernet/cadence/macb_main.c           |  4 ++--
 drivers/net/ethernet/cavium/liquidio/lio_ethtool.c |  2 +-
 .../net/ethernet/cavium/thunder/nicvf_ethtool.c    |  2 +-
 drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c |  2 +-
 drivers/net/ethernet/cisco/enic/enic_ethtool.c     |  2 +-
 drivers/net/ethernet/engleder/tsnep_ethtool.c      |  2 +-
 drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c |  2 +-
 .../net/ethernet/freescale/dpaa2/dpaa2-ethtool.c   |  2 +-
 .../net/ethernet/freescale/enetc/enetc_ethtool.c   |  2 +-
 drivers/net/ethernet/freescale/fec_main.c          |  2 +-
 drivers/net/ethernet/freescale/gianfar_ethtool.c   |  2 +-
 .../net/ethernet/fungible/funeth/funeth_ethtool.c  |  2 +-
 drivers/net/ethernet/hisilicon/hns3/hnae3.h        |  2 +-
 drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c |  2 +-
 .../net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c |  2 +-
 .../net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h |  2 +-
 drivers/net/ethernet/intel/e1000e/ethtool.c        |  2 +-
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c     |  2 +-
 drivers/net/ethernet/intel/ice/ice_ethtool.c       |  2 +-
 drivers/net/ethernet/intel/igb/igb_ethtool.c       |  2 +-
 drivers/net/ethernet/intel/igc/igc_ethtool.c       |  2 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c   |  2 +-
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c    |  2 +-
 .../ethernet/marvell/octeontx2/nic/otx2_ethtool.c  |  2 +-
 drivers/net/ethernet/mellanox/mlx4/en_ethtool.c    |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  2 +-
 .../net/ethernet/mellanox/mlx5/core/en_ethtool.c   |  4 ++--
 .../ethernet/mellanox/mlx5/core/ipoib/ethtool.c    |  2 +-
 drivers/net/ethernet/mellanox/mlxsw/spectrum.h     |  2 +-
 .../net/ethernet/mellanox/mlxsw/spectrum_ethtool.c |  2 +-
 drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c |  4 ++--
 drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.h | 10 ++++-----
 drivers/net/ethernet/microchip/lan743x_ethtool.c   |  2 +-
 .../ethernet/microchip/lan966x/lan966x_ethtool.c   |  2 +-
 .../net/ethernet/microchip/sparx5/sparx5_ethtool.c |  2 +-
 drivers/net/ethernet/mscc/ocelot_net.c             |  2 +-
 drivers/net/ethernet/mscc/ocelot_ptp.c             |  2 +-
 .../net/ethernet/pensando/ionic/ionic_ethtool.c    |  2 +-
 drivers/net/ethernet/qlogic/qede/qede_ethtool.c    |  2 +-
 drivers/net/ethernet/qlogic/qede/qede_ptp.c        |  2 +-
 drivers/net/ethernet/qlogic/qede/qede_ptp.h        |  2 +-
 drivers/net/ethernet/renesas/ravb_main.c           |  2 +-
 drivers/net/ethernet/renesas/rswitch.c             |  2 +-
 drivers/net/ethernet/renesas/rtsn.c                |  2 +-
 drivers/net/ethernet/sfc/ethtool.c                 |  2 +-
 drivers/net/ethernet/sfc/falcon/nic.h              |  2 +-
 drivers/net/ethernet/sfc/ptp.c                     |  2 +-
 drivers/net/ethernet/sfc/ptp.h                     |  5 +++--
 drivers/net/ethernet/sfc/siena/ethtool.c           |  2 +-
 drivers/net/ethernet/sfc/siena/ptp.c               |  2 +-
 drivers/net/ethernet/sfc/siena/ptp.h               |  4 ++--
 .../net/ethernet/stmicro/stmmac/stmmac_ethtool.c   |  2 +-
 drivers/net/ethernet/ti/am65-cpsw-ethtool.c        |  2 +-
 drivers/net/ethernet/ti/cpsw_ethtool.c             |  4 ++--
 drivers/net/ethernet/ti/cpsw_priv.h                |  2 +-
 drivers/net/ethernet/ti/icssg/icssg_ethtool.c      |  2 +-
 drivers/net/ethernet/ti/netcp_ethss.c              |  4 ++--
 drivers/net/ethernet/xscale/ixp4xx_eth.c           |  2 +-
 drivers/net/macvlan.c                              |  2 +-
 drivers/net/netdevsim/ethtool.c                    |  2 +-
 drivers/net/phy/bcm-phy-ptp.c                      |  2 +-
 drivers/net/phy/dp83640.c                          |  2 +-
 drivers/net/phy/micrel.c                           |  4 ++--
 drivers/net/phy/mscc/mscc_ptp.c                    |  2 +-
 drivers/net/phy/nxp-c45-tja11xx.c                  |  2 +-
 drivers/ptp/ptp_ines.c                             |  2 +-
 drivers/s390/net/qeth_ethtool.c                    |  2 +-
 include/linux/can/dev.h                            |  2 +-
 include/linux/ethtool.h                            | 25 +++++++++++++++++++---
 include/linux/mii_timestamper.h                    |  2 +-
 include/linux/phy.h                                |  2 +-
 include/net/dsa.h                                  |  2 +-
 include/soc/mscc/ocelot.h                          |  2 +-
 net/8021q/vlan_dev.c                               |  2 +-
 net/dsa/user.c                                     |  2 +-
 net/ethtool/common.c                               |  6 +++---
 net/ethtool/common.h                               |  2 +-
 net/ethtool/ioctl.c                                | 14 +++++++++---
 net/ethtool/tsinfo.c                               |  6 +++---
 net/sched/sch_taprio.c                             |  2 +-
 101 files changed, 151 insertions(+), 123 deletions(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index d19aabf5d4fb..af9ddd3902cc 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -5755,10 +5755,10 @@ static void bond_ethtool_get_drvinfo(struct net_device *bond_dev,
 }
 
 static int bond_ethtool_get_ts_info(struct net_device *bond_dev,
-				    struct ethtool_ts_info *info)
+				    struct kernel_ethtool_ts_info *info)
 {
 	struct bonding *bond = netdev_priv(bond_dev);
-	struct ethtool_ts_info ts_info;
+	struct kernel_ethtool_ts_info ts_info;
 	struct net_device *real_dev;
 	bool sw_tx_support = false;
 	struct list_head *iter;
diff --git a/drivers/net/can/dev/dev.c b/drivers/net/can/dev/dev.c
index 83e724e0ab87..87828f953073 100644
--- a/drivers/net/can/dev/dev.c
+++ b/drivers/net/can/dev/dev.c
@@ -376,7 +376,7 @@ EXPORT_SYMBOL(can_eth_ioctl_hwts);
  * supporting hardware timestamps
  */
 int can_ethtool_op_get_ts_info_hwts(struct net_device *dev,
-				    struct ethtool_ts_info *info)
+				    struct kernel_ethtool_ts_info *info)
 {
 	info->so_timestamping =
 		SOF_TIMESTAMPING_TX_SOFTWARE |
diff --git a/drivers/net/can/peak_canfd/peak_canfd.c b/drivers/net/can/peak_canfd/peak_canfd.c
index 31c9c127e24b..b50005397463 100644
--- a/drivers/net/can/peak_canfd/peak_canfd.c
+++ b/drivers/net/can/peak_canfd/peak_canfd.c
@@ -777,7 +777,7 @@ static const struct net_device_ops peak_canfd_netdev_ops = {
 };
 
 static int peak_get_ts_info(struct net_device *dev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	info->so_timestamping =
 		SOF_TIMESTAMPING_TX_SOFTWARE |
diff --git a/drivers/net/can/usb/gs_usb.c b/drivers/net/can/usb/gs_usb.c
index 340297e3bec7..bc86e9b329fd 100644
--- a/drivers/net/can/usb/gs_usb.c
+++ b/drivers/net/can/usb/gs_usb.c
@@ -1148,7 +1148,7 @@ static int gs_usb_set_phys_id(struct net_device *netdev,
 }
 
 static int gs_usb_get_ts_info(struct net_device *netdev,
-			      struct ethtool_ts_info *info)
+			      struct kernel_ethtool_ts_info *info)
 {
 	struct gs_can *dev = netdev_priv(netdev);
 
diff --git a/drivers/net/can/usb/peak_usb/pcan_usb_core.c b/drivers/net/can/usb/peak_usb/pcan_usb_core.c
index 1efa39e134f4..3d68fef46ded 100644
--- a/drivers/net/can/usb/peak_usb/pcan_usb_core.c
+++ b/drivers/net/can/usb/peak_usb/pcan_usb_core.c
@@ -897,7 +897,7 @@ int peak_usb_set_eeprom(struct net_device *netdev,
 	return 0;
 }
 
-int pcan_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+int pcan_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info)
 {
 	info->so_timestamping =
 		SOF_TIMESTAMPING_TX_SOFTWARE |
diff --git a/drivers/net/can/usb/peak_usb/pcan_usb_core.h b/drivers/net/can/usb/peak_usb/pcan_usb_core.h
index f6cf84bb718f..abab00930b9d 100644
--- a/drivers/net/can/usb/peak_usb/pcan_usb_core.h
+++ b/drivers/net/can/usb/peak_usb/pcan_usb_core.h
@@ -145,7 +145,7 @@ void peak_usb_get_ts_time(struct peak_time_ref *time_ref, u32 ts, ktime_t *tv);
 int peak_usb_netif_rx_64(struct sk_buff *skb, u32 ts_low, u32 ts_high);
 void peak_usb_async_complete(struct urb *urb);
 void peak_usb_restart_complete(struct peak_usb_device *dev);
-int pcan_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info);
+int pcan_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info);
 
 /* common 32-bit CAN channel ID ethtool management */
 int peak_usb_get_eeprom_len(struct net_device *netdev);
diff --git a/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c b/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c
index bd7aacc71a63..ca2500aba96f 100644
--- a/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c
+++ b/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c
@@ -16,7 +16,7 @@
 #include "hellcreek_ptp.h"
 
 int hellcreek_get_ts_info(struct dsa_switch *ds, int port,
-			  struct ethtool_ts_info *info)
+			  struct kernel_ethtool_ts_info *info)
 {
 	struct hellcreek *hellcreek = ds->priv;
 
diff --git a/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h b/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h
index 71af77efb28b..7d88da2134f2 100644
--- a/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h
+++ b/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h
@@ -48,7 +48,7 @@ void hellcreek_port_txtstamp(struct dsa_switch *ds, int port,
 			     struct sk_buff *skb);
 
 int hellcreek_get_ts_info(struct dsa_switch *ds, int port,
-			  struct ethtool_ts_info *info);
+			  struct kernel_ethtool_ts_info *info);
 
 long hellcreek_hwtstamp_work(struct ptp_clock_info *ptp);
 
diff --git a/drivers/net/dsa/microchip/ksz_ptp.c b/drivers/net/dsa/microchip/ksz_ptp.c
index 1fe105913c75..f0bd46e5d4ec 100644
--- a/drivers/net/dsa/microchip/ksz_ptp.c
+++ b/drivers/net/dsa/microchip/ksz_ptp.c
@@ -293,7 +293,7 @@ static int ksz_ptp_enable_mode(struct ksz_device *dev)
 /* The function is return back the capability of timestamping feature when
  * requested through ethtool -T <interface> utility
  */
-int ksz_get_ts_info(struct dsa_switch *ds, int port, struct ethtool_ts_info *ts)
+int ksz_get_ts_info(struct dsa_switch *ds, int port, struct kernel_ethtool_ts_info *ts)
 {
 	struct ksz_device *dev = ds->priv;
 	struct ksz_ptp_data *ptp_data;
diff --git a/drivers/net/dsa/microchip/ksz_ptp.h b/drivers/net/dsa/microchip/ksz_ptp.h
index 0ca8ca4f804e..2f1783c0d723 100644
--- a/drivers/net/dsa/microchip/ksz_ptp.h
+++ b/drivers/net/dsa/microchip/ksz_ptp.h
@@ -38,7 +38,7 @@ int ksz_ptp_clock_register(struct dsa_switch *ds);
 void ksz_ptp_clock_unregister(struct dsa_switch *ds);
 
 int ksz_get_ts_info(struct dsa_switch *ds, int port,
-		    struct ethtool_ts_info *ts);
+		    struct kernel_ethtool_ts_info *ts);
 int ksz_hwtstamp_get(struct dsa_switch *ds, int port, struct ifreq *ifr);
 int ksz_hwtstamp_set(struct dsa_switch *ds, int port, struct ifreq *ifr);
 void ksz_port_txtstamp(struct dsa_switch *ds, int port, struct sk_buff *skb);
diff --git a/drivers/net/dsa/mv88e6xxx/hwtstamp.c b/drivers/net/dsa/mv88e6xxx/hwtstamp.c
index 331b4ca089ff..49e6e1355142 100644
--- a/drivers/net/dsa/mv88e6xxx/hwtstamp.c
+++ b/drivers/net/dsa/mv88e6xxx/hwtstamp.c
@@ -64,7 +64,7 @@ static int mv88e6xxx_ptp_read(struct mv88e6xxx_chip *chip, int addr,
 #define TX_TSTAMP_TIMEOUT	msecs_to_jiffies(40)
 
 int mv88e6xxx_get_ts_info(struct dsa_switch *ds, int port,
-			  struct ethtool_ts_info *info)
+			  struct kernel_ethtool_ts_info *info)
 {
 	const struct mv88e6xxx_ptp_ops *ptp_ops;
 	struct mv88e6xxx_chip *chip;
diff --git a/drivers/net/dsa/mv88e6xxx/hwtstamp.h b/drivers/net/dsa/mv88e6xxx/hwtstamp.h
index cf7fb6d660b1..85acc758e3eb 100644
--- a/drivers/net/dsa/mv88e6xxx/hwtstamp.h
+++ b/drivers/net/dsa/mv88e6xxx/hwtstamp.h
@@ -121,7 +121,7 @@ void mv88e6xxx_port_txtstamp(struct dsa_switch *ds, int port,
 			     struct sk_buff *skb);
 
 int mv88e6xxx_get_ts_info(struct dsa_switch *ds, int port,
-			  struct ethtool_ts_info *info);
+			  struct kernel_ethtool_ts_info *info);
 
 int mv88e6xxx_hwtstamp_setup(struct mv88e6xxx_chip *chip);
 void mv88e6xxx_hwtstamp_free(struct mv88e6xxx_chip *chip);
@@ -157,7 +157,7 @@ static inline void mv88e6xxx_port_txtstamp(struct dsa_switch *ds, int port,
 }
 
 static inline int mv88e6xxx_get_ts_info(struct dsa_switch *ds, int port,
-					struct ethtool_ts_info *info)
+					struct kernel_ethtool_ts_info *info)
 {
 	return -EOPNOTSUPP;
 }
diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index d12c4e85baa7..e554699f06d4 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -1234,7 +1234,7 @@ static int felix_get_sset_count(struct dsa_switch *ds, int port, int sset)
 }
 
 static int felix_get_ts_info(struct dsa_switch *ds, int port,
-			     struct ethtool_ts_info *info)
+			     struct kernel_ethtool_ts_info *info)
 {
 	struct ocelot *ocelot = ds->priv;
 
diff --git a/drivers/net/dsa/sja1105/sja1105_ptp.c b/drivers/net/dsa/sja1105/sja1105_ptp.c
index a7d41e781398..a1f4ca6ad888 100644
--- a/drivers/net/dsa/sja1105/sja1105_ptp.c
+++ b/drivers/net/dsa/sja1105/sja1105_ptp.c
@@ -111,7 +111,7 @@ int sja1105_hwtstamp_get(struct dsa_switch *ds, int port, struct ifreq *ifr)
 }
 
 int sja1105_get_ts_info(struct dsa_switch *ds, int port,
-			struct ethtool_ts_info *info)
+			struct kernel_ethtool_ts_info *info)
 {
 	struct sja1105_private *priv = ds->priv;
 	struct sja1105_ptp_data *ptp_data = &priv->ptp_data;
diff --git a/drivers/net/dsa/sja1105/sja1105_ptp.h b/drivers/net/dsa/sja1105/sja1105_ptp.h
index 416461ee95d2..8add2bd5f728 100644
--- a/drivers/net/dsa/sja1105/sja1105_ptp.h
+++ b/drivers/net/dsa/sja1105/sja1105_ptp.h
@@ -101,7 +101,7 @@ void sja1105pqrs_ptp_cmd_packing(u8 *buf, struct sja1105_ptp_cmd *cmd,
 				 enum packing_op op);
 
 int sja1105_get_ts_info(struct dsa_switch *ds, int port,
-			struct ethtool_ts_info *ts);
+			struct kernel_ethtool_ts_info *ts);
 
 void sja1105_ptp_txtstamp_skb(struct dsa_switch *ds, int slot,
 			      struct sk_buff *clone);
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
index 58e7e88aae5b..21407a26f806 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
@@ -577,7 +577,7 @@ static int xgbe_set_rxfh(struct net_device *netdev,
 }
 
 static int xgbe_get_ts_info(struct net_device *netdev,
-			    struct ethtool_ts_info *ts_info)
+			    struct kernel_ethtool_ts_info *ts_info)
 {
 	struct xgbe_prv_data *pdata = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
index a2606ee3b0a5..d0aecd1d7357 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
@@ -652,7 +652,7 @@ static int aq_ethtool_set_wol(struct net_device *ndev,
 }
 
 static int aq_ethtool_get_ts_info(struct net_device *ndev,
-				  struct ethtool_ts_info *info)
+				  struct kernel_ethtool_ts_info *info)
 {
 	struct aq_nic_s *aq_nic = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
index 58956ed8f531..c7b56a5e5425 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
@@ -3634,7 +3634,7 @@ static int bnx2x_set_channels(struct net_device *dev,
 }
 
 static int bnx2x_get_ts_info(struct net_device *dev,
-			     struct ethtool_ts_info *info)
+			     struct kernel_ethtool_ts_info *info)
 {
 	struct bnx2x *bp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index bf157f6cc042..fcbf38f79c23 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -5013,7 +5013,7 @@ static int bnxt_get_dump_data(struct net_device *dev, struct ethtool_dump *dump,
 }
 
 static int bnxt_get_ts_info(struct net_device *dev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct bnxt *bp = netdev_priv(dev);
 	struct bnxt_ptp_cfg *ptp;
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index 1589a49b876c..0ec5f01551f9 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -6141,7 +6141,7 @@ static void tg3_refclk_write(struct tg3 *tp, u64 newval)
 
 static inline void tg3_full_lock(struct tg3 *tp, int irq_sync);
 static inline void tg3_full_unlock(struct tg3 *tp);
-static int tg3_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+static int tg3_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info)
 {
 	struct tg3 *tp = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index 122663ff7834..ea71612f6b36 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -1168,7 +1168,7 @@ struct macb_ptp_info {
 	s32 (*get_ptp_max_adj)(void);
 	unsigned int (*get_tsu_rate)(struct macb *bp);
 	int (*get_ts_info)(struct net_device *dev,
-			   struct ethtool_ts_info *info);
+			   struct kernel_ethtool_ts_info *info);
 	int (*get_hwtst)(struct net_device *netdev,
 			 struct kernel_hwtstamp_config *tstamp_config);
 	int (*set_hwtst)(struct net_device *netdev,
diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index cecc3d6e630f..11665be3a22c 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -3399,7 +3399,7 @@ static s32 gem_get_ptp_max_adj(void)
 }
 
 static int gem_get_ts_info(struct net_device *dev,
-			   struct ethtool_ts_info *info)
+			   struct kernel_ethtool_ts_info *info)
 {
 	struct macb *bp = netdev_priv(dev);
 
@@ -3440,7 +3440,7 @@ static struct macb_ptp_info gem_ptp_info = {
 #endif
 
 static int macb_get_ts_info(struct net_device *netdev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct macb *bp = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
index d3e07b6ed5e1..5835965dbc32 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
@@ -2497,7 +2497,7 @@ ret_intrmod:
 }
 
 static int lio_get_ts_info(struct net_device *netdev,
-			   struct ethtool_ts_info *info)
+			   struct kernel_ethtool_ts_info *info)
 {
 	struct lio *lio = GET_LIO(netdev);
 
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
index 34125b8cd935..6a04d2530176 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
@@ -836,7 +836,7 @@ static int nicvf_set_pauseparam(struct net_device *dev,
 }
 
 static int nicvf_get_ts_info(struct net_device *netdev,
-			     struct ethtool_ts_info *info)
+			     struct kernel_ethtool_ts_info *info)
 {
 	struct nicvf *nic = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
index 47eecde36285..3d091947ae00 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
@@ -1550,7 +1550,7 @@ out_free_fw:
 	return ret;
 }
 
-static int get_ts_info(struct net_device *dev, struct ethtool_ts_info *ts_info)
+static int get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *ts_info)
 {
 	struct port_info *pi = netdev_priv(dev);
 	struct  adapter *adapter = pi->adapter;
diff --git a/drivers/net/ethernet/cisco/enic/enic_ethtool.c b/drivers/net/ethernet/cisco/enic/enic_ethtool.c
index a42f3f280f3e..f2f1055880b2 100644
--- a/drivers/net/ethernet/cisco/enic/enic_ethtool.c
+++ b/drivers/net/ethernet/cisco/enic/enic_ethtool.c
@@ -599,7 +599,7 @@ static int enic_set_rxfh(struct net_device *netdev,
 }
 
 static int enic_get_ts_info(struct net_device *netdev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
 				SOF_TIMESTAMPING_RX_SOFTWARE |
diff --git a/drivers/net/ethernet/engleder/tsnep_ethtool.c b/drivers/net/ethernet/engleder/tsnep_ethtool.c
index 65ec1abc9442..9aa286ba1f00 100644
--- a/drivers/net/ethernet/engleder/tsnep_ethtool.c
+++ b/drivers/net/ethernet/engleder/tsnep_ethtool.c
@@ -305,7 +305,7 @@ static void tsnep_ethtool_get_channels(struct net_device *netdev,
 }
 
 static int tsnep_ethtool_get_ts_info(struct net_device *netdev,
-				     struct ethtool_ts_info *info)
+				     struct kernel_ethtool_ts_info *info)
 {
 	struct tsnep_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
index 5bd0b36d1feb..0e923d805732 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
@@ -394,7 +394,7 @@ static int dpaa_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
 }
 
 static int dpaa_get_ts_info(struct net_device *net_dev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct device *dev = net_dev->dev.parent;
 	struct device_node *mac_node = dev->of_node;
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c
index e80e9388c71f..7f476519b7ad 100644
--- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c
@@ -794,7 +794,7 @@ int dpaa2_phc_index = -1;
 EXPORT_SYMBOL(dpaa2_phc_index);
 
 static int dpaa2_eth_get_ts_info(struct net_device *dev,
-				 struct ethtool_ts_info *info)
+				 struct kernel_ethtool_ts_info *info)
 {
 	if (!dpaa2_ptp)
 		return ethtool_op_get_ts_info(dev, info);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
index f7753ea5b57e..5e684b23c5f5 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
@@ -841,7 +841,7 @@ static int enetc_set_coalesce(struct net_device *ndev,
 }
 
 static int enetc_get_ts_info(struct net_device *ndev,
-			     struct ethtool_ts_info *info)
+			     struct kernel_ethtool_ts_info *info)
 {
 	int *phc_idx;
 
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index fb19295529a2..a923cb95cdc6 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -2768,7 +2768,7 @@ static void fec_enet_get_regs(struct net_device *ndev,
 }
 
 static int fec_enet_get_ts_info(struct net_device *ndev,
-				struct ethtool_ts_info *info)
+				struct kernel_ethtool_ts_info *info)
 {
 	struct fec_enet_private *fep = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/freescale/gianfar_ethtool.c b/drivers/net/ethernet/freescale/gianfar_ethtool.c
index 7a15b9245698..f581402ad740 100644
--- a/drivers/net/ethernet/freescale/gianfar_ethtool.c
+++ b/drivers/net/ethernet/freescale/gianfar_ethtool.c
@@ -1448,7 +1448,7 @@ static int gfar_get_nfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
 }
 
 static int gfar_get_ts_info(struct net_device *dev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct gfar_private *priv = netdev_priv(dev);
 	struct platform_device *ptp_dev;
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_ethtool.c b/drivers/net/ethernet/fungible/funeth/funeth_ethtool.c
index 4edd0adfc6c7..7f081e6e8c87 100644
--- a/drivers/net/ethernet/fungible/funeth/funeth_ethtool.c
+++ b/drivers/net/ethernet/fungible/funeth/funeth_ethtool.c
@@ -1040,7 +1040,7 @@ static int fun_set_rxfh(struct net_device *netdev,
 }
 
 static int fun_get_ts_info(struct net_device *netdev,
-			   struct ethtool_ts_info *info)
+			   struct kernel_ethtool_ts_info *info)
 {
 	info->so_timestamping = SOF_TIMESTAMPING_RX_SOFTWARE |
 				SOF_TIMESTAMPING_RX_HARDWARE |
diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
index 7cebb08bd320..27dbe367f3d3 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
@@ -786,7 +786,7 @@ struct hnae3_ae_ops {
 	void (*get_rx_hwts)(struct hnae3_handle *handle, struct sk_buff *skb,
 			    u32 nsec, u32 sec);
 	int (*get_ts_info)(struct hnae3_handle *handle,
-			   struct ethtool_ts_info *info);
+			   struct kernel_ethtool_ts_info *info);
 	int (*get_link_diagnosis_info)(struct hnae3_handle *handle,
 				       u32 *status_code);
 	void (*clean_vf_config)(struct hnae3_ae_dev *ae_dev, int num_vfs);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
index 941cb529d671..b1e988347347 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
@@ -2009,7 +2009,7 @@ static int hns3_set_tunable(struct net_device *netdev,
 				 ETHTOOL_RING_USE_TX_PUSH)
 
 static int hns3_get_ts_info(struct net_device *netdev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct hnae3_handle *handle = hns3_get_handle(netdev);
 
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c
index 507d7ce26d83..5fff8ed388f8 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c
@@ -378,7 +378,7 @@ int hclge_ptp_set_cfg(struct hclge_dev *hdev, struct ifreq *ifr)
 }
 
 int hclge_ptp_get_ts_info(struct hnae3_handle *handle,
-			  struct ethtool_ts_info *info)
+			  struct kernel_ethtool_ts_info *info)
 {
 	struct hclge_vport *vport = hclge_get_vport(handle);
 	struct hclge_dev *hdev = vport->back;
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h
index bbee74cd8404..63483636c074 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h
@@ -138,6 +138,6 @@ int hclge_ptp_set_cfg(struct hclge_dev *hdev, struct ifreq *ifr);
 int hclge_ptp_init(struct hclge_dev *hdev);
 void hclge_ptp_uninit(struct hclge_dev *hdev);
 int hclge_ptp_get_ts_info(struct hnae3_handle *handle,
-			  struct ethtool_ts_info *info);
+			  struct kernel_ethtool_ts_info *info);
 int hclge_ptp_cfg_qry(struct hclge_dev *hdev, u32 *cfg);
 #endif
diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c
index 85da20778e0f..9364bc2b4eb1 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -2263,7 +2263,7 @@ static int e1000e_set_eee(struct net_device *netdev, struct ethtool_keee *edata)
 }
 
 static int e1000e_get_ts_info(struct net_device *netdev,
-			      struct ethtool_ts_info *info)
+			      struct kernel_ethtool_ts_info *info)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 4e28785c9fb2..1d0d2e526adb 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -2546,7 +2546,7 @@ static void i40e_get_strings(struct net_device *netdev, u32 stringset,
 }
 
 static int i40e_get_ts_info(struct net_device *dev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct i40e_pf *pf = i40e_netdev_to_pf(dev);
 
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 62c8205fceba..b8f142f04330 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -3434,7 +3434,7 @@ ice_set_rxfh(struct net_device *netdev, struct ethtool_rxfh_param *rxfh,
 }
 
 static int
-ice_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+ice_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info)
 {
 	struct ice_pf *pf = ice_netdev_to_pf(dev);
 
diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c
index 61d72250c0ed..06b9970dffad 100644
--- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
+++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
@@ -2381,7 +2381,7 @@ static void igb_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
 }
 
 static int igb_get_ts_info(struct net_device *dev,
-			   struct ethtool_ts_info *info)
+			   struct kernel_ethtool_ts_info *info)
 {
 	struct igb_adapter *adapter = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c
index 0cd2bd695db1..9c3adb18f922 100644
--- a/drivers/net/ethernet/intel/igc/igc_ethtool.c
+++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c
@@ -1559,7 +1559,7 @@ static int igc_ethtool_set_channels(struct net_device *netdev,
 }
 
 static int igc_ethtool_get_ts_info(struct net_device *dev,
-				   struct ethtool_ts_info *info)
+				   struct kernel_ethtool_ts_info *info)
 {
 	struct igc_adapter *adapter = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
index 6e6e6f1847b6..4cac76254966 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
@@ -3170,7 +3170,7 @@ static int ixgbe_set_rxfh(struct net_device *netdev,
 }
 
 static int ixgbe_get_ts_info(struct net_device *dev,
-			     struct ethtool_ts_info *info)
+			     struct kernel_ethtool_ts_info *info)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
index 9adf4301c9b1..8c45ad983abc 100644
--- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
+++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
@@ -5259,7 +5259,7 @@ static int mvpp2_get_ts_config(struct mvpp2_port *port, struct ifreq *ifr)
 }
 
 static int mvpp2_ethtool_get_ts_info(struct net_device *dev,
-				     struct ethtool_ts_info *info)
+				     struct kernel_ethtool_ts_info *info)
 {
 	struct mvpp2_port *port = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
index 7f786de61014..0db62eb0dab3 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
@@ -954,7 +954,7 @@ static u32 otx2_get_link(struct net_device *netdev)
 }
 
 static int otx2_get_ts_info(struct net_device *netdev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct otx2_nic *pfvf = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 0606f18e5bbe..943d6918c2ec 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -1884,7 +1884,7 @@ out:
 }
 
 static int mlx4_en_get_ts_info(struct net_device *dev,
-			       struct ethtool_ts_info *info)
+			       struct kernel_ethtool_ts_info *info)
 {
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 6a343a8f162f..5fd82c67b6ab 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -1191,7 +1191,7 @@ int mlx5e_set_per_queue_coalesce(struct net_device *dev, u32 queue,
 u32 mlx5e_ethtool_get_rxfh_key_size(struct mlx5e_priv *priv);
 u32 mlx5e_ethtool_get_rxfh_indir_size(struct mlx5e_priv *priv);
 int mlx5e_ethtool_get_ts_info(struct mlx5e_priv *priv,
-			      struct ethtool_ts_info *info);
+			      struct kernel_ethtool_ts_info *info);
 int mlx5e_ethtool_flash_device(struct mlx5e_priv *priv,
 			       struct ethtool_flash *flash);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 3320f12ba2db..cbb3945529d7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -1658,7 +1658,7 @@ static int mlx5e_set_pauseparam(struct net_device *netdev,
 }
 
 int mlx5e_ethtool_get_ts_info(struct mlx5e_priv *priv,
-			      struct ethtool_ts_info *info)
+			      struct kernel_ethtool_ts_info *info)
 {
 	struct mlx5_core_dev *mdev = priv->mdev;
 
@@ -1682,7 +1682,7 @@ int mlx5e_ethtool_get_ts_info(struct mlx5e_priv *priv,
 }
 
 static int mlx5e_get_ts_info(struct net_device *dev,
-			     struct ethtool_ts_info *info)
+			     struct kernel_ethtool_ts_info *info)
 {
 	struct mlx5e_priv *priv = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
index 779d92b762d3..905bdbaffb9a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
@@ -136,7 +136,7 @@ static int mlx5i_get_coalesce(struct net_device *netdev,
 }
 
 static int mlx5i_get_ts_info(struct net_device *netdev,
-			     struct ethtool_ts_info *info)
+			     struct kernel_ethtool_ts_info *info)
 {
 	struct mlx5e_priv *priv = mlx5i_epriv(netdev);
 
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index bb0586b45c8d..8d3c61287696 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -238,7 +238,7 @@ struct mlxsw_sp_ptp_ops {
 			    struct hwtstamp_config *config);
 	void (*shaper_work)(struct work_struct *work);
 	int (*get_ts_info)(struct mlxsw_sp *mlxsw_sp,
-			   struct ethtool_ts_info *info);
+			   struct kernel_ethtool_ts_info *info);
 	int (*get_stats_count)(void);
 	void (*get_stats_strings)(u8 **p);
 	void (*get_stats)(struct mlxsw_sp_port *mlxsw_sp_port,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
index c79da1411d33..2bed8c86b7cf 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
@@ -1082,7 +1082,7 @@ mlxsw_sp_set_module_eeprom_by_page(struct net_device *dev,
 }
 
 static int
-mlxsw_sp_get_ts_info(struct net_device *netdev, struct ethtool_ts_info *info)
+mlxsw_sp_get_ts_info(struct net_device *netdev, struct kernel_ethtool_ts_info *info)
 {
 	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(netdev);
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
index cbb6c75a6620..5b174cb95eb8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
@@ -1276,7 +1276,7 @@ int mlxsw_sp1_ptp_hwtstamp_set(struct mlxsw_sp_port *mlxsw_sp_port,
 }
 
 int mlxsw_sp1_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
-			      struct ethtool_ts_info *info)
+			      struct kernel_ethtool_ts_info *info)
 {
 	info->phc_index = ptp_clock_index(mlxsw_sp->clock->ptp);
 
@@ -1661,7 +1661,7 @@ err_get_message_types:
 }
 
 int mlxsw_sp2_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
-			      struct ethtool_ts_info *info)
+			      struct kernel_ethtool_ts_info *info)
 {
 	info->phc_index = ptp_clock_index(mlxsw_sp->clock->ptp);
 
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.h
index a8b88230959a..769095d4932d 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.h
@@ -11,7 +11,7 @@ struct mlxsw_sp;
 struct mlxsw_sp_port;
 struct mlxsw_sp_ptp_clock;
 
-static inline int mlxsw_sp_ptp_get_ts_info_noptp(struct ethtool_ts_info *info)
+static inline int mlxsw_sp_ptp_get_ts_info_noptp(struct kernel_ethtool_ts_info *info)
 {
 	info->so_timestamping = SOF_TIMESTAMPING_RX_SOFTWARE |
 				SOF_TIMESTAMPING_SOFTWARE;
@@ -50,7 +50,7 @@ int mlxsw_sp1_ptp_hwtstamp_set(struct mlxsw_sp_port *mlxsw_sp_port,
 void mlxsw_sp1_ptp_shaper_work(struct work_struct *work);
 
 int mlxsw_sp1_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
-			      struct ethtool_ts_info *info);
+			      struct kernel_ethtool_ts_info *info);
 
 int mlxsw_sp1_get_stats_count(void);
 void mlxsw_sp1_get_stats_strings(u8 **p);
@@ -84,7 +84,7 @@ int mlxsw_sp2_ptp_hwtstamp_set(struct mlxsw_sp_port *mlxsw_sp_port,
 			       struct hwtstamp_config *config);
 
 int mlxsw_sp2_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
-			      struct ethtool_ts_info *info);
+			      struct kernel_ethtool_ts_info *info);
 
 int mlxsw_sp2_ptp_txhdr_construct(struct mlxsw_core *mlxsw_core,
 				  struct mlxsw_sp_port *mlxsw_sp_port,
@@ -152,7 +152,7 @@ static inline void mlxsw_sp1_ptp_shaper_work(struct work_struct *work)
 }
 
 static inline int mlxsw_sp1_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
-					    struct ethtool_ts_info *info)
+					    struct kernel_ethtool_ts_info *info)
 {
 	return mlxsw_sp_ptp_get_ts_info_noptp(info);
 }
@@ -227,7 +227,7 @@ mlxsw_sp2_ptp_hwtstamp_set(struct mlxsw_sp_port *mlxsw_sp_port,
 }
 
 static inline int mlxsw_sp2_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
-					    struct ethtool_ts_info *info)
+					    struct kernel_ethtool_ts_info *info)
 {
 	return mlxsw_sp_ptp_get_ts_info_noptp(info);
 }
diff --git a/drivers/net/ethernet/microchip/lan743x_ethtool.c b/drivers/net/ethernet/microchip/lan743x_ethtool.c
index 0d1740d64676..3a63ec091413 100644
--- a/drivers/net/ethernet/microchip/lan743x_ethtool.c
+++ b/drivers/net/ethernet/microchip/lan743x_ethtool.c
@@ -1029,7 +1029,7 @@ static int lan743x_ethtool_set_rxfh(struct net_device *netdev,
 }
 
 static int lan743x_ethtool_get_ts_info(struct net_device *netdev,
-				       struct ethtool_ts_info *ts_info)
+				       struct kernel_ethtool_ts_info *ts_info)
 {
 	struct lan743x_adapter *adapter = netdev_priv(netdev);
 
diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c b/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c
index c0fc85ac5db3..aec7066d83b3 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c
@@ -538,7 +538,7 @@ static int lan966x_set_pauseparam(struct net_device *dev,
 }
 
 static int lan966x_get_ts_info(struct net_device *dev,
-			       struct ethtool_ts_info *info)
+			       struct kernel_ethtool_ts_info *info)
 {
 	struct lan966x_port *port = netdev_priv(dev);
 	struct lan966x *lan966x = port->lan966x;
diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_ethtool.c b/drivers/net/ethernet/microchip/sparx5/sparx5_ethtool.c
index a06dc5a9b355..4f800c1a435d 100644
--- a/drivers/net/ethernet/microchip/sparx5/sparx5_ethtool.c
+++ b/drivers/net/ethernet/microchip/sparx5/sparx5_ethtool.c
@@ -1183,7 +1183,7 @@ static void sparx5_config_port_stats(struct sparx5 *sparx5, int portno)
 }
 
 static int sparx5_get_ts_info(struct net_device *dev,
-			      struct ethtool_ts_info *info)
+			      struct kernel_ethtool_ts_info *info)
 {
 	struct sparx5_port *port = netdev_priv(dev);
 	struct sparx5 *sparx5 = port->sparx5;
diff --git a/drivers/net/ethernet/mscc/ocelot_net.c b/drivers/net/ethernet/mscc/ocelot_net.c
index 21a87a3fc556..7c9540a71725 100644
--- a/drivers/net/ethernet/mscc/ocelot_net.c
+++ b/drivers/net/ethernet/mscc/ocelot_net.c
@@ -980,7 +980,7 @@ static int ocelot_port_get_sset_count(struct net_device *dev, int sset)
 }
 
 static int ocelot_port_get_ts_info(struct net_device *dev,
-				   struct ethtool_ts_info *info)
+				   struct kernel_ethtool_ts_info *info)
 {
 	struct ocelot_port_private *priv = netdev_priv(dev);
 	struct ocelot *ocelot = priv->port.ocelot;
diff --git a/drivers/net/ethernet/mscc/ocelot_ptp.c b/drivers/net/ethernet/mscc/ocelot_ptp.c
index cb32234a5bf1..b3c28260adf8 100644
--- a/drivers/net/ethernet/mscc/ocelot_ptp.c
+++ b/drivers/net/ethernet/mscc/ocelot_ptp.c
@@ -580,7 +580,7 @@ int ocelot_hwstamp_set(struct ocelot *ocelot, int port, struct ifreq *ifr)
 EXPORT_SYMBOL(ocelot_hwstamp_set);
 
 int ocelot_get_ts_info(struct ocelot *ocelot, int port,
-		       struct ethtool_ts_info *info)
+		       struct kernel_ethtool_ts_info *info)
 {
 	info->phc_index = ocelot->ptp_clock ?
 			  ptp_clock_index(ocelot->ptp_clock) : -1;
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c
index 185a03514ae3..4619fd74f3e3 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c
@@ -977,7 +977,7 @@ static int ionic_get_module_eeprom(struct net_device *netdev,
 }
 
 static int ionic_get_ts_info(struct net_device *netdev,
-			     struct ethtool_ts_info *info)
+			     struct kernel_ethtool_ts_info *info)
 {
 	struct ionic_lif *lif = netdev_priv(netdev);
 	struct ionic *ionic = lif->ionic;
diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
index f497f6ca1018..97b059be1041 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
@@ -1137,7 +1137,7 @@ static int qede_set_channels(struct net_device *dev,
 }
 
 static int qede_get_ts_info(struct net_device *dev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct qede_dev *edev = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/qlogic/qede/qede_ptp.c b/drivers/net/ethernet/qlogic/qede/qede_ptp.c
index 747cc5e2bb78..63e3dac4d5f7 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_ptp.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_ptp.c
@@ -321,7 +321,7 @@ int qede_ptp_hw_ts(struct qede_dev *edev, struct ifreq *ifr)
 			    sizeof(config)) ? -EFAULT : 0;
 }
 
-int qede_ptp_get_ts_info(struct qede_dev *edev, struct ethtool_ts_info *info)
+int qede_ptp_get_ts_info(struct qede_dev *edev, struct kernel_ethtool_ts_info *info)
 {
 	struct qede_ptp *ptp = edev->ptp;
 
diff --git a/drivers/net/ethernet/qlogic/qede/qede_ptp.h b/drivers/net/ethernet/qlogic/qede/qede_ptp.h
index 1db0f021c645..adafc894797e 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_ptp.h
+++ b/drivers/net/ethernet/qlogic/qede/qede_ptp.h
@@ -17,7 +17,7 @@ void qede_ptp_tx_ts(struct qede_dev *edev, struct sk_buff *skb);
 int qede_ptp_hw_ts(struct qede_dev *edev, struct ifreq *req);
 void qede_ptp_disable(struct qede_dev *edev);
 int qede_ptp_enable(struct qede_dev *edev);
-int qede_ptp_get_ts_info(struct qede_dev *edev, struct ethtool_ts_info *ts);
+int qede_ptp_get_ts_info(struct qede_dev *edev, struct kernel_ethtool_ts_info *ts);
 
 static inline void qede_ptp_record_rx_ts(struct qede_dev *edev,
 					 union eth_rx_cqe *cqe,
diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 6605e4f4af53..c02fb296bf7d 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -1737,7 +1737,7 @@ static int ravb_set_ringparam(struct net_device *ndev,
 }
 
 static int ravb_get_ts_info(struct net_device *ndev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct ravb_private *priv = netdev_priv(ndev);
 	const struct ravb_hw_info *hw_info = priv->info;
diff --git a/drivers/net/ethernet/renesas/rswitch.c b/drivers/net/ethernet/renesas/rswitch.c
index 24c90d8f5a44..ff50e20856ec 100644
--- a/drivers/net/ethernet/renesas/rswitch.c
+++ b/drivers/net/ethernet/renesas/rswitch.c
@@ -1809,7 +1809,7 @@ static const struct net_device_ops rswitch_netdev_ops = {
 	.ndo_set_mac_address = eth_mac_addr,
 };
 
-static int rswitch_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info)
+static int rswitch_get_ts_info(struct net_device *ndev, struct kernel_ethtool_ts_info *info)
 {
 	struct rswitch_device *rdev = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/renesas/rtsn.c b/drivers/net/ethernet/renesas/rtsn.c
index ad69d47463cb..577227c007ab 100644
--- a/drivers/net/ethernet/renesas/rtsn.c
+++ b/drivers/net/ethernet/renesas/rtsn.c
@@ -1213,7 +1213,7 @@ static const struct net_device_ops rtsn_netdev_ops = {
 };
 
 static int rtsn_get_ts_info(struct net_device *ndev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct rtsn_private *priv = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index 0f5c68b8bab7..7c887160e2ef 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -226,7 +226,7 @@ static void efx_ethtool_get_fec_stats(struct net_device *net_dev,
 }
 
 static int efx_ethtool_get_ts_info(struct net_device *net_dev,
-				   struct ethtool_ts_info *ts_info)
+				   struct kernel_ethtool_ts_info *ts_info)
 {
 	struct efx_nic *efx = efx_netdev_priv(net_dev);
 
diff --git a/drivers/net/ethernet/sfc/falcon/nic.h b/drivers/net/ethernet/sfc/falcon/nic.h
index 9f413474bd9f..ada6e036fd97 100644
--- a/drivers/net/ethernet/sfc/falcon/nic.h
+++ b/drivers/net/ethernet/sfc/falcon/nic.h
@@ -297,7 +297,7 @@ static inline struct falcon_board *falcon_board(struct ef4_nic *efx)
 	return &data->board;
 }
 
-struct ethtool_ts_info;
+struct kernel_ethtool_ts_info;
 
 extern const struct ef4_nic_type falcon_a1_nic_type;
 extern const struct ef4_nic_type falcon_b0_nic_type;
diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index c3bffbf0ba2b..6fd2fdbaa418 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -1864,7 +1864,7 @@ static int efx_ptp_ts_init(struct efx_nic *efx, struct kernel_hwtstamp_config *i
 	return 0;
 }
 
-void efx_ptp_get_ts_info(struct efx_nic *efx, struct ethtool_ts_info *ts_info)
+void efx_ptp_get_ts_info(struct efx_nic *efx, struct kernel_ethtool_ts_info *ts_info)
 {
 	struct efx_ptp_data *ptp = efx->ptp_data;
 	struct efx_nic *primary = efx->primary;
diff --git a/drivers/net/ethernet/sfc/ptp.h b/drivers/net/ethernet/sfc/ptp.h
index 2f30dbb490d2..6946203499ef 100644
--- a/drivers/net/ethernet/sfc/ptp.h
+++ b/drivers/net/ethernet/sfc/ptp.h
@@ -12,7 +12,7 @@
 #include <linux/net_tstamp.h>
 #include "net_driver.h"
 
-struct ethtool_ts_info;
+struct kernel_ethtool_ts_info;
 int efx_ptp_probe(struct efx_nic *efx, struct efx_channel *channel);
 void efx_ptp_defer_probe_with_channel(struct efx_nic *efx);
 struct efx_channel *efx_ptp_channel(struct efx_nic *efx);
@@ -23,7 +23,8 @@ int efx_ptp_set_ts_config(struct efx_nic *efx,
 			  struct netlink_ext_ack *extack);
 int efx_ptp_get_ts_config(struct efx_nic *efx,
 			  struct kernel_hwtstamp_config *config);
-void efx_ptp_get_ts_info(struct efx_nic *efx, struct ethtool_ts_info *ts_info);
+void efx_ptp_get_ts_info(struct efx_nic *efx,
+			 struct kernel_ethtool_ts_info *ts_info);
 bool efx_ptp_is_ptp_tx(struct efx_nic *efx, struct sk_buff *skb);
 int efx_ptp_get_mode(struct efx_nic *efx);
 int efx_ptp_change_mode(struct efx_nic *efx, bool enable_wanted,
diff --git a/drivers/net/ethernet/sfc/siena/ethtool.c b/drivers/net/ethernet/sfc/siena/ethtool.c
index 14dd3893bdef..4c182d4edfc2 100644
--- a/drivers/net/ethernet/sfc/siena/ethtool.c
+++ b/drivers/net/ethernet/sfc/siena/ethtool.c
@@ -226,7 +226,7 @@ static void efx_ethtool_get_fec_stats(struct net_device *net_dev,
 }
 
 static int efx_ethtool_get_ts_info(struct net_device *net_dev,
-				   struct ethtool_ts_info *ts_info)
+				   struct kernel_ethtool_ts_info *ts_info)
 {
 	struct efx_nic *efx = netdev_priv(net_dev);
 
diff --git a/drivers/net/ethernet/sfc/siena/ptp.c b/drivers/net/ethernet/sfc/siena/ptp.c
index 4b5e2f0ba350..c473a4b6dd44 100644
--- a/drivers/net/ethernet/sfc/siena/ptp.c
+++ b/drivers/net/ethernet/sfc/siena/ptp.c
@@ -1780,7 +1780,7 @@ static int efx_ptp_ts_init(struct efx_nic *efx,
 }
 
 void efx_siena_ptp_get_ts_info(struct efx_nic *efx,
-			       struct ethtool_ts_info *ts_info)
+			       struct kernel_ethtool_ts_info *ts_info)
 {
 	struct efx_ptp_data *ptp = efx->ptp_data;
 	struct efx_nic *primary = efx->primary;
diff --git a/drivers/net/ethernet/sfc/siena/ptp.h b/drivers/net/ethernet/sfc/siena/ptp.h
index 6352f84424f6..b6133e7c5608 100644
--- a/drivers/net/ethernet/sfc/siena/ptp.h
+++ b/drivers/net/ethernet/sfc/siena/ptp.h
@@ -12,7 +12,7 @@
 #include <linux/net_tstamp.h>
 #include "net_driver.h"
 
-struct ethtool_ts_info;
+struct kernel_ethtool_ts_info;
 void efx_siena_ptp_defer_probe_with_channel(struct efx_nic *efx);
 struct efx_channel *efx_siena_ptp_channel(struct efx_nic *efx);
 int efx_siena_ptp_set_ts_config(struct efx_nic *efx,
@@ -21,7 +21,7 @@ int efx_siena_ptp_set_ts_config(struct efx_nic *efx,
 int efx_siena_ptp_get_ts_config(struct efx_nic *efx,
 				struct kernel_hwtstamp_config *config);
 void efx_siena_ptp_get_ts_info(struct efx_nic *efx,
-			       struct ethtool_ts_info *ts_info);
+			       struct kernel_ethtool_ts_info *ts_info);
 bool efx_siena_ptp_is_ptp_tx(struct efx_nic *efx, struct sk_buff *skb);
 int efx_siena_ptp_get_mode(struct efx_nic *efx);
 int efx_siena_ptp_change_mode(struct efx_nic *efx, bool enable_wanted,
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
index 18468c0228f0..7008219fd88d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
@@ -1199,7 +1199,7 @@ static int stmmac_set_channels(struct net_device *dev,
 }
 
 static int stmmac_get_ts_info(struct net_device *dev,
-			      struct ethtool_ts_info *info)
+			      struct kernel_ethtool_ts_info *info)
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
 
diff --git a/drivers/net/ethernet/ti/am65-cpsw-ethtool.c b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c
index a1d0935d1ebe..b60976947da5 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-ethtool.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c
@@ -692,7 +692,7 @@ static void am65_cpsw_get_eth_mac_stats(struct net_device *ndev,
 };
 
 static int am65_cpsw_get_ethtool_ts_info(struct net_device *ndev,
-					 struct ethtool_ts_info *info)
+					 struct kernel_ethtool_ts_info *info)
 {
 	struct am65_cpsw_common *common = am65_ndev_to_common(ndev);
 	unsigned int ptp_v2_filter;
diff --git a/drivers/net/ethernet/ti/cpsw_ethtool.c b/drivers/net/ethernet/ti/cpsw_ethtool.c
index f7b283353ba2..53ed23d68722 100644
--- a/drivers/net/ethernet/ti/cpsw_ethtool.c
+++ b/drivers/net/ethernet/ti/cpsw_ethtool.c
@@ -717,7 +717,7 @@ err:
 }
 
 #if IS_ENABLED(CONFIG_TI_CPTS)
-int cpsw_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info)
+int cpsw_get_ts_info(struct net_device *ndev, struct kernel_ethtool_ts_info *info)
 {
 	struct cpsw_common *cpsw = ndev_to_cpsw(ndev);
 
@@ -738,7 +738,7 @@ int cpsw_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info)
 	return 0;
 }
 #else
-int cpsw_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info)
+int cpsw_get_ts_info(struct net_device *ndev, struct kernel_ethtool_ts_info *info)
 {
 	info->so_timestamping =
 		SOF_TIMESTAMPING_TX_SOFTWARE |
diff --git a/drivers/net/ethernet/ti/cpsw_priv.h b/drivers/net/ethernet/ti/cpsw_priv.h
index 7efa72502c86..1f448290b9f4 100644
--- a/drivers/net/ethernet/ti/cpsw_priv.h
+++ b/drivers/net/ethernet/ti/cpsw_priv.h
@@ -510,6 +510,6 @@ int cpsw_set_ringparam(struct net_device *ndev,
 int cpsw_set_channels_common(struct net_device *ndev,
 			     struct ethtool_channels *chs,
 			     cpdma_handler_fn rx_handler);
-int cpsw_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info);
+int cpsw_get_ts_info(struct net_device *ndev, struct kernel_ethtool_ts_info *info);
 
 #endif /* DRIVERS_NET_ETHERNET_TI_CPSW_PRIV_H_ */
diff --git a/drivers/net/ethernet/ti/icssg/icssg_ethtool.c b/drivers/net/ethernet/ti/icssg/icssg_ethtool.c
index c8d0f45cc5b1..524723ced9ed 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_ethtool.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_ethtool.c
@@ -110,7 +110,7 @@ static void emac_get_ethtool_stats(struct net_device *ndev,
 }
 
 static int emac_get_ts_info(struct net_device *ndev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct prueth_emac *emac = netdev_priv(ndev);
 
diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c
index 02cb6474f6dc..d286709ca3b9 100644
--- a/drivers/net/ethernet/ti/netcp_ethss.c
+++ b/drivers/net/ethernet/ti/netcp_ethss.c
@@ -1999,7 +1999,7 @@ static int keystone_set_link_ksettings(struct net_device *ndev,
 
 #if IS_ENABLED(CONFIG_TI_CPTS)
 static int keystone_get_ts_info(struct net_device *ndev,
-				struct ethtool_ts_info *info)
+				struct kernel_ethtool_ts_info *info)
 {
 	struct netcp_intf *netcp = netdev_priv(ndev);
 	struct gbe_intf *gbe_intf;
@@ -2027,7 +2027,7 @@ static int keystone_get_ts_info(struct net_device *ndev,
 }
 #else
 static int keystone_get_ts_info(struct net_device *ndev,
-				struct ethtool_ts_info *info)
+				struct kernel_ethtool_ts_info *info)
 {
 	info->so_timestamping =
 		SOF_TIMESTAMPING_TX_SOFTWARE |
diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c
index 8aff6a73ca0a..56df37f8d50a 100644
--- a/drivers/net/ethernet/xscale/ixp4xx_eth.c
+++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c
@@ -1015,7 +1015,7 @@ static void ixp4xx_get_drvinfo(struct net_device *dev,
 }
 
 static int ixp4xx_get_ts_info(struct net_device *dev,
-			      struct ethtool_ts_info *info)
+			      struct kernel_ethtool_ts_info *info)
 {
 	struct port *port = netdev_priv(dev);
 
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 67b7ef2d463f..24298a33e0e9 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -1084,7 +1084,7 @@ static int macvlan_ethtool_get_link_ksettings(struct net_device *dev,
 }
 
 static int macvlan_ethtool_get_ts_info(struct net_device *dev,
-				       struct ethtool_ts_info *info)
+				       struct kernel_ethtool_ts_info *info)
 {
 	struct net_device *real_dev = macvlan_dev_real_dev(dev);
 
diff --git a/drivers/net/netdevsim/ethtool.c b/drivers/net/netdevsim/ethtool.c
index 3f9c9327f149..1436905bc106 100644
--- a/drivers/net/netdevsim/ethtool.c
+++ b/drivers/net/netdevsim/ethtool.c
@@ -148,7 +148,7 @@ nsim_get_fec_stats(struct net_device *dev, struct ethtool_fec_stats *fec_stats)
 }
 
 static int nsim_get_ts_info(struct net_device *dev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct netdevsim *ns = netdev_priv(dev);
 
diff --git a/drivers/net/phy/bcm-phy-ptp.c b/drivers/net/phy/bcm-phy-ptp.c
index d3e825c951ee..874a1b64b115 100644
--- a/drivers/net/phy/bcm-phy-ptp.c
+++ b/drivers/net/phy/bcm-phy-ptp.c
@@ -841,7 +841,7 @@ static int bcm_ptp_hwtstamp(struct mii_timestamper *mii_ts,
 }
 
 static int bcm_ptp_ts_info(struct mii_timestamper *mii_ts,
-			   struct ethtool_ts_info *ts_info)
+			   struct kernel_ethtool_ts_info *ts_info)
 {
 	struct bcm_ptp_private *priv = mii2priv(mii_ts);
 
diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c
index d3e72d5c1472..075d2beea716 100644
--- a/drivers/net/phy/dp83640.c
+++ b/drivers/net/phy/dp83640.c
@@ -1395,7 +1395,7 @@ static void dp83640_txtstamp(struct mii_timestamper *mii_ts,
 }
 
 static int dp83640_ts_info(struct mii_timestamper *mii_ts,
-			   struct ethtool_ts_info *info)
+			   struct kernel_ethtool_ts_info *info)
 {
 	struct dp83640_private *dp83640 =
 		container_of(mii_ts, struct dp83640_private, mii_ts);
diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index 8d57225d8575..dd519805deee 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -2552,7 +2552,7 @@ static void lan8814_ptp_tx_ts_get(struct phy_device *phydev,
 	*seq_id = lanphy_read_page_reg(phydev, 5, PTP_TX_MSG_HEADER2);
 }
 
-static int lan8814_ts_info(struct mii_timestamper *mii_ts, struct ethtool_ts_info *info)
+static int lan8814_ts_info(struct mii_timestamper *mii_ts, struct kernel_ethtool_ts_info *info)
 {
 	struct kszphy_ptp_priv *ptp_priv = container_of(mii_ts, struct kszphy_ptp_priv, mii_ts);
 	struct phy_device *phydev = ptp_priv->phydev;
@@ -4317,7 +4317,7 @@ static irqreturn_t lan8841_handle_interrupt(struct phy_device *phydev)
 }
 
 static int lan8841_ts_info(struct mii_timestamper *mii_ts,
-			   struct ethtool_ts_info *info)
+			   struct kernel_ethtool_ts_info *info)
 {
 	struct kszphy_ptp_priv *ptp_priv;
 
diff --git a/drivers/net/phy/mscc/mscc_ptp.c b/drivers/net/phy/mscc/mscc_ptp.c
index e66d20eff7c4..c1ddae36a2ae 100644
--- a/drivers/net/phy/mscc/mscc_ptp.c
+++ b/drivers/net/phy/mscc/mscc_ptp.c
@@ -1134,7 +1134,7 @@ static int vsc85xx_hwtstamp(struct mii_timestamper *mii_ts,
 }
 
 static int vsc85xx_ts_info(struct mii_timestamper *mii_ts,
-			   struct ethtool_ts_info *info)
+			   struct kernel_ethtool_ts_info *info)
 {
 	struct vsc8531_private *vsc8531 =
 		container_of(mii_ts, struct vsc8531_private, mii_ts);
diff --git a/drivers/net/phy/nxp-c45-tja11xx.c b/drivers/net/phy/nxp-c45-tja11xx.c
index d18c133e6013..5af5ade4fc64 100644
--- a/drivers/net/phy/nxp-c45-tja11xx.c
+++ b/drivers/net/phy/nxp-c45-tja11xx.c
@@ -1058,7 +1058,7 @@ nxp_c45_no_ptp_irq:
 }
 
 static int nxp_c45_ts_info(struct mii_timestamper *mii_ts,
-			   struct ethtool_ts_info *ts_info)
+			   struct kernel_ethtool_ts_info *ts_info)
 {
 	struct nxp_c45_phy *priv = container_of(mii_ts, struct nxp_c45_phy,
 						mii_ts);
diff --git a/drivers/ptp/ptp_ines.c b/drivers/ptp/ptp_ines.c
index 385643f3f8fe..e6f7d2bf8dde 100644
--- a/drivers/ptp/ptp_ines.c
+++ b/drivers/ptp/ptp_ines.c
@@ -556,7 +556,7 @@ static bool ines_timestamp_expired(struct ines_timestamp *ts)
 }
 
 static int ines_ts_info(struct mii_timestamper *mii_ts,
-			struct ethtool_ts_info *info)
+			struct kernel_ethtool_ts_info *info)
 {
 	info->so_timestamping =
 		SOF_TIMESTAMPING_TX_HARDWARE |
diff --git a/drivers/s390/net/qeth_ethtool.c b/drivers/s390/net/qeth_ethtool.c
index c1caf7734c3e..f184c58ecf24 100644
--- a/drivers/s390/net/qeth_ethtool.c
+++ b/drivers/s390/net/qeth_ethtool.c
@@ -247,7 +247,7 @@ static int qeth_set_channels(struct net_device *dev,
 }
 
 static int qeth_get_ts_info(struct net_device *dev,
-			    struct ethtool_ts_info *info)
+			    struct kernel_ethtool_ts_info *info)
 {
 	struct qeth_card *card = dev->ml_priv;
 
diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h
index 1b92aed49363..23492213ea35 100644
--- a/include/linux/can/dev.h
+++ b/include/linux/can/dev.h
@@ -186,7 +186,7 @@ void close_candev(struct net_device *dev);
 int can_change_mtu(struct net_device *dev, int new_mtu);
 int can_eth_ioctl_hwts(struct net_device *netdev, struct ifreq *ifr, int cmd);
 int can_ethtool_op_get_ts_info_hwts(struct net_device *dev,
-				    struct ethtool_ts_info *info);
+				    struct kernel_ethtool_ts_info *info);
 
 int register_candev(struct net_device *dev);
 void unregister_candev(struct net_device *dev);
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index e213b5508da6..6b38bbda5790 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -18,6 +18,7 @@
 #include <linux/if_ether.h>
 #include <linux/netlink.h>
 #include <uapi/linux/ethtool.h>
+#include <uapi/linux/net_tstamp.h>
 
 struct compat_ethtool_rx_flow_spec {
 	u32		flow_type;
@@ -713,6 +714,22 @@ struct ethtool_rxfh_param {
 	u8	input_xfrm;
 };
 
+/**
+ * struct kernel_ethtool_ts_info - kernel copy of struct ethtool_ts_info
+ * @cmd: command number = %ETHTOOL_GET_TS_INFO
+ * @so_timestamping: bit mask of the sum of the supported SO_TIMESTAMPING flags
+ * @phc_index: device index of the associated PHC, or -1 if there is none
+ * @tx_types: bit mask of the supported hwtstamp_tx_types enumeration values
+ * @rx_filters: bit mask of the supported hwtstamp_rx_filters enumeration values
+ */
+struct kernel_ethtool_ts_info {
+	u32 cmd;
+	u32 so_timestamping;
+	int phc_index;
+	enum hwtstamp_tx_types tx_types;
+	enum hwtstamp_rx_filters rx_filters;
+};
+
 /**
  * struct ethtool_ops - optional netdev operations
  * @cap_link_lanes_supported: indicates if the driver supports lanes
@@ -1020,7 +1037,7 @@ struct ethtool_ops {
 	int	(*get_dump_data)(struct net_device *,
 				 struct ethtool_dump *, void *);
 	int	(*set_dump)(struct net_device *, struct ethtool_dump *);
-	int	(*get_ts_info)(struct net_device *, struct ethtool_ts_info *);
+	int	(*get_ts_info)(struct net_device *, struct kernel_ethtool_ts_info *);
 	void	(*get_ts_stats)(struct net_device *dev,
 				struct ethtool_ts_stats *ts_stats);
 	int     (*get_module_info)(struct net_device *,
@@ -1181,7 +1198,8 @@ int ethtool_get_phc_vclocks(struct net_device *dev, int **vclock_index);
 
 /* Some generic methods drivers may use in their ethtool_ops */
 u32 ethtool_op_get_link(struct net_device *dev);
-int ethtool_op_get_ts_info(struct net_device *dev, struct ethtool_ts_info *eti);
+int ethtool_op_get_ts_info(struct net_device *dev,
+			   struct kernel_ethtool_ts_info *eti);
 
 /**
  * ethtool_mm_frag_size_add_to_min - Translate (standard) additional fragment
@@ -1230,7 +1248,8 @@ static inline int ethtool_mm_frag_size_min_to_add(u32 val_min, u32 *val_add,
  * @info: buffer to hold the result
  * Returns zero on success, non-zero otherwise.
  */
-int ethtool_get_ts_info_by_layer(struct net_device *dev, struct ethtool_ts_info *info);
+int ethtool_get_ts_info_by_layer(struct net_device *dev,
+				 struct kernel_ethtool_ts_info *info);
 
 /**
  * ethtool_sprintf - Write formatted string to ethtool string data
diff --git a/include/linux/mii_timestamper.h b/include/linux/mii_timestamper.h
index 26b04f73f214..995db62570f9 100644
--- a/include/linux/mii_timestamper.h
+++ b/include/linux/mii_timestamper.h
@@ -59,7 +59,7 @@ struct mii_timestamper {
 			   struct phy_device *phydev);
 
 	int  (*ts_info)(struct mii_timestamper *mii_ts,
-			struct ethtool_ts_info *ts_info);
+			struct kernel_ethtool_ts_info *ts_info);
 
 	struct device *device;
 };
diff --git a/include/linux/phy.h b/include/linux/phy.h
index e7a38137211c..04ae5c811cfb 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -1618,7 +1618,7 @@ static inline bool phy_rxtstamp(struct phy_device *phydev, struct sk_buff *skb,
 }
 
 static inline int phy_ts_info(struct phy_device *phydev,
-			      struct ethtool_ts_info *tsinfo)
+			      struct kernel_ethtool_ts_info *tsinfo)
 {
 	return phydev->mii_ts->ts_info(phydev->mii_ts, tsinfo);
 }
diff --git a/include/net/dsa.h b/include/net/dsa.h
index f9ae3ca66b6f..96efdd9f90c9 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -934,7 +934,7 @@ struct dsa_switch_ops {
 	 * ethtool timestamp info
 	 */
 	int	(*get_ts_info)(struct dsa_switch *ds, int port,
-			       struct ethtool_ts_info *ts);
+			       struct kernel_ethtool_ts_info *ts);
 
 	/*
 	 * ethtool MAC merge layer
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index 1e1b40f4e664..6a37b29f4b4c 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -1016,7 +1016,7 @@ void ocelot_port_get_eth_mac_stats(struct ocelot *ocelot, int port,
 void ocelot_port_get_eth_phy_stats(struct ocelot *ocelot, int port,
 				   struct ethtool_eth_phy_stats *phy_stats);
 int ocelot_get_ts_info(struct ocelot *ocelot, int port,
-		       struct ethtool_ts_info *info);
+		       struct kernel_ethtool_ts_info *info);
 void ocelot_set_ageing_time(struct ocelot *ocelot, unsigned int msecs);
 int ocelot_port_vlan_filtering(struct ocelot *ocelot, int port, bool enabled,
 			       struct netlink_ext_ack *extack);
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 3efba4f857ac..217be32426b5 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -677,7 +677,7 @@ static void vlan_ethtool_get_drvinfo(struct net_device *dev,
 }
 
 static int vlan_ethtool_get_ts_info(struct net_device *dev,
-				    struct ethtool_ts_info *info)
+				    struct kernel_ethtool_ts_info *info)
 {
 	const struct vlan_dev_priv *vlan = vlan_dev_priv(dev);
 	return ethtool_get_ts_info_by_layer(vlan->real_dev, info);
diff --git a/net/dsa/user.c b/net/dsa/user.c
index e8f56a40b614..f5adfa1d978a 100644
--- a/net/dsa/user.c
+++ b/net/dsa/user.c
@@ -1729,7 +1729,7 @@ static int dsa_user_set_rxnfc(struct net_device *dev,
 }
 
 static int dsa_user_get_ts_info(struct net_device *dev,
-				struct ethtool_ts_info *ts)
+				struct kernel_ethtool_ts_info *ts)
 {
 	struct dsa_user_priv *p = netdev_priv(dev);
 	struct dsa_switch *ds = p->dp->ds;
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 01b7550f12c6..461017a37955 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -629,7 +629,7 @@ int ethtool_check_ops(const struct ethtool_ops *ops)
 	return 0;
 }
 
-int __ethtool_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+int __ethtool_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info)
 {
 	const struct ethtool_ops *ops = dev->ethtool_ops;
 	struct phy_device *phydev = dev->phydev;
@@ -651,7 +651,7 @@ int __ethtool_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
 
 int ethtool_get_phc_vclocks(struct net_device *dev, int **vclock_index)
 {
-	struct ethtool_ts_info info = { };
+	struct kernel_ethtool_ts_info info = { };
 	int num = 0;
 
 	if (!__ethtool_get_ts_info(dev, &info))
@@ -661,7 +661,7 @@ int ethtool_get_phc_vclocks(struct net_device *dev, int **vclock_index)
 }
 EXPORT_SYMBOL(ethtool_get_phc_vclocks);
 
-int ethtool_get_ts_info_by_layer(struct net_device *dev, struct ethtool_ts_info *info)
+int ethtool_get_ts_info_by_layer(struct net_device *dev, struct kernel_ethtool_ts_info *info)
 {
 	return __ethtool_get_ts_info(dev, info);
 }
diff --git a/net/ethtool/common.h b/net/ethtool/common.h
index 28b8aaaf9bcb..b9daeecbd84d 100644
--- a/net/ethtool/common.h
+++ b/net/ethtool/common.h
@@ -44,7 +44,7 @@ bool convert_legacy_settings_to_link_ksettings(
 	const struct ethtool_cmd *legacy_settings);
 int ethtool_get_max_rxfh_channel(struct net_device *dev, u32 *max);
 int ethtool_get_max_rxnfc_channel(struct net_device *dev, u64 *max);
-int __ethtool_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info);
+int __ethtool_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info);
 
 extern const struct ethtool_phy_ops *ethtool_phy_ops;
 extern const struct ethtool_pse_ops *ethtool_pse_ops;
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index d72b0fec89af..bc8988ddf51c 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -65,7 +65,8 @@ u32 ethtool_op_get_link(struct net_device *dev)
 }
 EXPORT_SYMBOL(ethtool_op_get_link);
 
-int ethtool_op_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+int ethtool_op_get_ts_info(struct net_device *dev,
+			   struct kernel_ethtool_ts_info *info)
 {
 	info->so_timestamping =
 		SOF_TIMESTAMPING_TX_SOFTWARE |
@@ -2569,13 +2570,20 @@ out:
 
 static int ethtool_get_ts_info(struct net_device *dev, void __user *useraddr)
 {
-	struct ethtool_ts_info info;
+	struct kernel_ethtool_ts_info kernel_info;
+	struct ethtool_ts_info info = {};
 	int err;
 
-	err = __ethtool_get_ts_info(dev, &info);
+	err = __ethtool_get_ts_info(dev, &kernel_info);
 	if (err)
 		return err;
 
+	info.cmd = kernel_info.cmd;
+	info.so_timestamping = kernel_info.so_timestamping;
+	info.phc_index = kernel_info.phc_index;
+	info.tx_types = kernel_info.tx_types;
+	info.rx_filters = kernel_info.rx_filters;
+
 	if (copy_to_user(useraddr, &info, sizeof(info)))
 		return -EFAULT;
 
diff --git a/net/ethtool/tsinfo.c b/net/ethtool/tsinfo.c
index 57d496287e52..03d12d6f79ca 100644
--- a/net/ethtool/tsinfo.c
+++ b/net/ethtool/tsinfo.c
@@ -12,7 +12,7 @@ struct tsinfo_req_info {
 
 struct tsinfo_reply_data {
 	struct ethnl_reply_data		base;
-	struct ethtool_ts_info		ts_info;
+	struct kernel_ethtool_ts_info	ts_info;
 	struct ethtool_ts_stats		stats;
 };
 
@@ -55,7 +55,7 @@ static int tsinfo_reply_size(const struct ethnl_req_info *req_base,
 {
 	const struct tsinfo_reply_data *data = TSINFO_REPDATA(reply_base);
 	bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
-	const struct ethtool_ts_info *ts_info = &data->ts_info;
+	const struct kernel_ethtool_ts_info *ts_info = &data->ts_info;
 	int len = 0;
 	int ret;
 
@@ -136,7 +136,7 @@ static int tsinfo_fill_reply(struct sk_buff *skb,
 {
 	const struct tsinfo_reply_data *data = TSINFO_REPDATA(reply_base);
 	bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
-	const struct ethtool_ts_info *ts_info = &data->ts_info;
+	const struct kernel_ethtool_ts_info *ts_info = &data->ts_info;
 	int ret;
 
 	if (ts_info->so_timestamping) {
diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c
index b284a06b5a75..cc2df9f8c14a 100644
--- a/net/sched/sch_taprio.c
+++ b/net/sched/sch_taprio.c
@@ -1610,7 +1610,7 @@ static int taprio_parse_clockid(struct Qdisc *sch, struct nlattr **tb,
 
 	if (FULL_OFFLOAD_IS_ENABLED(q->flags)) {
 		const struct ethtool_ops *ops = dev->ethtool_ops;
-		struct ethtool_ts_info info = {
+		struct kernel_ethtool_ts_info info = {
 			.cmd = ETHTOOL_GET_TS_INFO,
 			.phc_index = -1,
 		};
-- 
cgit v1.2.3-58-ga151


From f96eb1172ed8d74cfed7da92d2045ec64028cc38 Mon Sep 17 00:00:00 2001
From: Christian Eggers <ceggers@arri.de>
Date: Mon, 15 Jul 2024 14:30:50 +0200
Subject: dsa: lan9303: consistent naming for PHY address parameter

Name it 'addr' instead of 'port' or 'phy'.

Signed-off-by: Christian Eggers <ceggers@arri.de>
Link: https://patch.msgid.link/20240715123050.21202-1-ceggers@arri.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/dsa/lan9303_mdio.c | 8 ++++----
 include/linux/dsa/lan9303.h    | 4 ++--
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/dsa/lan9303_mdio.c b/drivers/net/dsa/lan9303_mdio.c
index 167a86f39f27..0ac4857e5ee8 100644
--- a/drivers/net/dsa/lan9303_mdio.c
+++ b/drivers/net/dsa/lan9303_mdio.c
@@ -58,19 +58,19 @@ static int lan9303_mdio_read(void *ctx, uint32_t reg, uint32_t *val)
 	return 0;
 }
 
-static int lan9303_mdio_phy_write(struct lan9303 *chip, int phy, int reg,
+static int lan9303_mdio_phy_write(struct lan9303 *chip, int addr, int reg,
 				  u16 val)
 {
 	struct lan9303_mdio *sw_dev = dev_get_drvdata(chip->dev);
 
-	return mdiobus_write_nested(sw_dev->device->bus, phy, reg, val);
+	return mdiobus_write_nested(sw_dev->device->bus, addr, reg, val);
 }
 
-static int lan9303_mdio_phy_read(struct lan9303 *chip, int phy,  int reg)
+static int lan9303_mdio_phy_read(struct lan9303 *chip, int addr, int reg)
 {
 	struct lan9303_mdio *sw_dev = dev_get_drvdata(chip->dev);
 
-	return mdiobus_read_nested(sw_dev->device->bus, phy, reg);
+	return mdiobus_read_nested(sw_dev->device->bus, addr, reg);
 }
 
 static const struct lan9303_phy_ops lan9303_mdio_phy_ops = {
diff --git a/include/linux/dsa/lan9303.h b/include/linux/dsa/lan9303.h
index b4f22112ba75..3ce7cbcc37a3 100644
--- a/include/linux/dsa/lan9303.h
+++ b/include/linux/dsa/lan9303.h
@@ -5,8 +5,8 @@ struct lan9303;
 
 struct lan9303_phy_ops {
 	/* PHY 1 and 2 access*/
-	int	(*phy_read)(struct lan9303 *chip, int port, int regnum);
-	int	(*phy_write)(struct lan9303 *chip, int port,
+	int	(*phy_read)(struct lan9303 *chip, int addr, int regnum);
+	int	(*phy_write)(struct lan9303 *chip, int addr,
 			     int regnum, u16 val);
 };
 
-- 
cgit v1.2.3-58-ga151


From 70de41ef78573ce958ac04ecc2b5671851723c59 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sun, 14 Jul 2024 18:05:56 +0200
Subject: llc: Constify struct llc_conn_state_trans

'struct llc_conn_state_trans' are not modified in this driver.

Constifying this structure moves some data to a read-only section, so
increase overall security.

On a x86_64, with allmodconfig, as an example:
Before:
======
   text	   data	    bss	    dec	    hex	filename
  13923	  10896	     32	  24851	   6113	net/llc/llc_c_st.o

After:
=====
   text	   data	    bss	    dec	    hex	filename
  21859	   3328	      0	  25187	   6263	net/llc/llc_c_st.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/87cda89e4c9414e71d1a54bb1eb491b0e7f70375.1720973029.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/llc_c_st.h |   4 +-
 net/llc/llc_c_st.c     | 500 ++++++++++++++++++++++++-------------------------
 net/llc/llc_conn.c     |  20 +-
 3 files changed, 262 insertions(+), 262 deletions(-)

diff --git a/include/net/llc_c_st.h b/include/net/llc_c_st.h
index 53823d61d8b6..a4bea0f33188 100644
--- a/include/net/llc_c_st.h
+++ b/include/net/llc_c_st.h
@@ -44,8 +44,8 @@ struct llc_conn_state_trans {
 };
 
 struct llc_conn_state {
-	u8			    current_state;
-	struct llc_conn_state_trans **transitions;
+	u8				  current_state;
+	const struct llc_conn_state_trans **transitions;
 };
 
 extern struct llc_conn_state llc_conn_state_table[];
diff --git a/net/llc/llc_c_st.c b/net/llc/llc_c_st.c
index 2467573b5f84..1c267db304df 100644
--- a/net/llc/llc_c_st.c
+++ b/net/llc/llc_c_st.c
@@ -42,7 +42,7 @@ static const llc_conn_action_t llc_common_actions_1[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_1 = {
+static const struct llc_conn_state_trans llc_common_state_trans_1 = {
 	.ev	       = llc_conn_ev_disc_req,
 	.next_state    = LLC_CONN_STATE_D_CONN,
 	.ev_qualifiers = NONE,
@@ -59,7 +59,7 @@ static const llc_conn_action_t llc_common_actions_2[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_2 = {
+static const struct llc_conn_state_trans llc_common_state_trans_2 = {
 	.ev	       = llc_conn_ev_rst_req,
 	.next_state    = LLC_CONN_STATE_RESET,
 	.ev_qualifiers = NONE,
@@ -79,7 +79,7 @@ static const llc_conn_action_t llc_common_actions_3[] = {
 	[8] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_3 = {
+static const struct llc_conn_state_trans llc_common_state_trans_3 = {
 	.ev	       = llc_conn_ev_rx_sabme_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -95,7 +95,7 @@ static const llc_conn_action_t llc_common_actions_4[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_4 = {
+static const struct llc_conn_state_trans llc_common_state_trans_4 = {
 	.ev	       = llc_conn_ev_rx_disc_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = NONE,
@@ -114,7 +114,7 @@ static const llc_conn_action_t llc_common_actions_5[] = {
 	[7] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_5 = {
+static const struct llc_conn_state_trans llc_common_state_trans_5 = {
 	.ev	       = llc_conn_ev_rx_frmr_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_RESET,
 	.ev_qualifiers = NONE,
@@ -129,7 +129,7 @@ static const llc_conn_action_t llc_common_actions_6[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_6 = {
+static const struct llc_conn_state_trans llc_common_state_trans_6 = {
 	.ev	       = llc_conn_ev_rx_dm_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = NONE,
@@ -145,7 +145,7 @@ static const llc_conn_action_t llc_common_actions_7a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_7a = {
+static const struct llc_conn_state_trans llc_common_state_trans_7a = {
 	.ev	       = llc_conn_ev_rx_zzz_cmd_pbit_set_x_inval_nr,
 	.next_state    = LLC_CONN_STATE_ERROR,
 	.ev_qualifiers = NONE,
@@ -161,7 +161,7 @@ static const llc_conn_action_t llc_common_actions_7b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_7b = {
+static const struct llc_conn_state_trans llc_common_state_trans_7b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_x_inval_ns,
 	.next_state    = LLC_CONN_STATE_ERROR,
 	.ev_qualifiers = NONE,
@@ -177,7 +177,7 @@ static const llc_conn_action_t llc_common_actions_8a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_8a = {
+static const struct llc_conn_state_trans llc_common_state_trans_8a = {
 	.ev	       = llc_conn_ev_rx_zzz_rsp_fbit_set_x_inval_nr,
 	.next_state    = LLC_CONN_STATE_ERROR,
 	.ev_qualifiers = NONE,
@@ -193,7 +193,7 @@ static const llc_conn_action_t llc_common_actions_8b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_8b = {
+static const struct llc_conn_state_trans llc_common_state_trans_8b = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_x_inval_ns,
 	.next_state    = LLC_CONN_STATE_ERROR,
 	.ev_qualifiers = NONE,
@@ -209,7 +209,7 @@ static const llc_conn_action_t llc_common_actions_8c[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_8c = {
+static const struct llc_conn_state_trans llc_common_state_trans_8c = {
 	.ev	       = llc_conn_ev_rx_bad_pdu,
 	.next_state    = LLC_CONN_STATE_ERROR,
 	.ev_qualifiers = NONE,
@@ -225,7 +225,7 @@ static const llc_conn_action_t llc_common_actions_9[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_9 = {
+static const struct llc_conn_state_trans llc_common_state_trans_9 = {
 	.ev	       = llc_conn_ev_rx_ua_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_ERROR,
 	.ev_qualifiers = NONE,
@@ -247,7 +247,7 @@ static const llc_conn_action_t llc_common_actions_10[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_10 = {
+static const struct llc_conn_state_trans llc_common_state_trans_10 = {
 	.ev	       = llc_conn_ev_rx_xxx_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_ERROR,
 	.ev_qualifiers = llc_common_ev_qfyrs_10,
@@ -270,7 +270,7 @@ static const llc_conn_action_t llc_common_actions_11a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_11a = {
+static const struct llc_conn_state_trans llc_common_state_trans_11a = {
 	.ev	       = llc_conn_ev_p_tmr_exp,
 	.next_state    = LLC_CONN_STATE_RESET,
 	.ev_qualifiers = llc_common_ev_qfyrs_11a,
@@ -292,7 +292,7 @@ static const llc_conn_action_t llc_common_actions_11b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_11b = {
+static const struct llc_conn_state_trans llc_common_state_trans_11b = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_RESET,
 	.ev_qualifiers = llc_common_ev_qfyrs_11b,
@@ -314,7 +314,7 @@ static const llc_conn_action_t llc_common_actions_11c[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_11c = {
+static const struct llc_conn_state_trans llc_common_state_trans_11c = {
 	.ev	       = llc_conn_ev_rej_tmr_exp,
 	.next_state    = LLC_CONN_STATE_RESET,
 	.ev_qualifiers = llc_common_ev_qfyrs_11c,
@@ -336,7 +336,7 @@ static const llc_conn_action_t llc_common_actions_11d[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_common_state_trans_11d = {
+static const struct llc_conn_state_trans llc_common_state_trans_11d = {
 	.ev	       = llc_conn_ev_busy_tmr_exp,
 	.next_state    = LLC_CONN_STATE_RESET,
 	.ev_qualifiers = llc_common_ev_qfyrs_11d,
@@ -347,7 +347,7 @@ static struct llc_conn_state_trans llc_common_state_trans_11d = {
  * Common dummy state transition; must be last entry for all state
  * transition groups - it'll be on .bss, so will be zeroed.
  */
-static struct llc_conn_state_trans llc_common_state_trans_end;
+static const struct llc_conn_state_trans llc_common_state_trans_end;
 
 /* LLC_CONN_STATE_ADM transitions */
 /* State transitions for LLC_CONN_EV_CONN_REQ event */
@@ -359,7 +359,7 @@ static const llc_conn_action_t llc_adm_actions_1[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_adm_state_trans_1 = {
+static const struct llc_conn_state_trans llc_adm_state_trans_1 = {
 	.ev	       = llc_conn_ev_conn_req,
 	.next_state    = LLC_CONN_STATE_SETUP,
 	.ev_qualifiers = NONE,
@@ -378,7 +378,7 @@ static const llc_conn_action_t llc_adm_actions_2[] = {
 	[7] = NULL,
 };
 
-static struct llc_conn_state_trans llc_adm_state_trans_2 = {
+static const struct llc_conn_state_trans llc_adm_state_trans_2 = {
 	.ev	       = llc_conn_ev_rx_sabme_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -392,7 +392,7 @@ static const llc_conn_action_t llc_adm_actions_3[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_adm_state_trans_3 = {
+static const struct llc_conn_state_trans llc_adm_state_trans_3 = {
 	.ev	       = llc_conn_ev_rx_disc_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = NONE,
@@ -406,7 +406,7 @@ static const llc_conn_action_t llc_adm_actions_4[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_adm_state_trans_4 = {
+static const struct llc_conn_state_trans llc_adm_state_trans_4 = {
 	.ev	       = llc_conn_ev_rx_xxx_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = NONE,
@@ -419,7 +419,7 @@ static const llc_conn_action_t llc_adm_actions_5[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_adm_state_trans_5 = {
+static const struct llc_conn_state_trans llc_adm_state_trans_5 = {
 	.ev	       = llc_conn_ev_rx_any_frame,
 	.next_state    = LLC_CONN_OUT_OF_SVC,
 	.ev_qualifiers = NONE,
@@ -430,7 +430,7 @@ static struct llc_conn_state_trans llc_adm_state_trans_5 = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_adm_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_adm_state_transitions[] = {
 	[0] = &llc_adm_state_trans_1,		/* Request */
 	[1] = &llc_common_state_trans_end,
 	[2] = &llc_common_state_trans_end,	/* local_busy */
@@ -453,7 +453,7 @@ static const llc_conn_action_t llc_setup_actions_1[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_setup_state_trans_1 = {
+static const struct llc_conn_state_trans llc_setup_state_trans_1 = {
 	.ev	       = llc_conn_ev_rx_sabme_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_SETUP,
 	.ev_qualifiers = NONE,
@@ -477,7 +477,7 @@ static const llc_conn_action_t llc_setup_actions_2[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_setup_state_trans_2 = {
+static const struct llc_conn_state_trans llc_setup_state_trans_2 = {
 	.ev	       = llc_conn_ev_rx_ua_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_setup_ev_qfyrs_2,
@@ -498,7 +498,7 @@ static const llc_conn_action_t llc_setup_actions_3[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_setup_state_trans_3 = {
+static const struct llc_conn_state_trans llc_setup_state_trans_3 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_setup_ev_qfyrs_3,
@@ -519,7 +519,7 @@ static const llc_conn_action_t llc_setup_actions_4[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_setup_state_trans_4 = {
+static const struct llc_conn_state_trans llc_setup_state_trans_4 = {
 	.ev	       = llc_conn_ev_rx_disc_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_setup_ev_qfyrs_4,
@@ -539,7 +539,7 @@ static const llc_conn_action_t llc_setup_actions_5[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_setup_state_trans_5 = {
+static const struct llc_conn_state_trans llc_setup_state_trans_5 = {
 	.ev	       = llc_conn_ev_rx_dm_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_setup_ev_qfyrs_5,
@@ -560,7 +560,7 @@ static const llc_conn_action_t llc_setup_actions_7[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_setup_state_trans_7 = {
+static const struct llc_conn_state_trans llc_setup_state_trans_7 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_SETUP,
 	.ev_qualifiers = llc_setup_ev_qfyrs_7,
@@ -581,7 +581,7 @@ static const llc_conn_action_t llc_setup_actions_8[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_setup_state_trans_8 = {
+static const struct llc_conn_state_trans llc_setup_state_trans_8 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_setup_ev_qfyrs_8,
@@ -592,7 +592,7 @@ static struct llc_conn_state_trans llc_setup_state_trans_8 = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_setup_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_setup_state_transitions[] = {
 	 [0] = &llc_common_state_trans_end,	/* Request */
 	 [1] = &llc_common_state_trans_end,	/* local busy */
 	 [2] = &llc_common_state_trans_end,	/* init_pf_cycle */
@@ -622,7 +622,7 @@ static const llc_conn_action_t llc_normal_actions_1[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_1 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_1 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_1,
@@ -643,7 +643,7 @@ static const llc_conn_action_t llc_normal_actions_2[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_2 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_2 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_2,
@@ -660,7 +660,7 @@ static const llc_conn_ev_qfyr_t llc_normal_ev_qfyrs_2_1[] = {
 /* just one member, NULL, .bss zeroes it */
 static const llc_conn_action_t llc_normal_actions_2_1[1];
 
-static struct llc_conn_state_trans llc_normal_state_trans_2_1 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_2_1 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_2_1,
@@ -680,7 +680,7 @@ static const llc_conn_action_t llc_normal_actions_3[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_3 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_3 = {
 	.ev	       = llc_conn_ev_local_busy_detected,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_normal_ev_qfyrs_3,
@@ -700,7 +700,7 @@ static const llc_conn_action_t llc_normal_actions_4[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_4 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_4 = {
 	.ev	       = llc_conn_ev_local_busy_detected,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_normal_ev_qfyrs_4,
@@ -723,7 +723,7 @@ static const llc_conn_action_t llc_normal_actions_5a[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_5a = {
+static const struct llc_conn_state_trans llc_normal_state_trans_5a = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_normal_ev_qfyrs_5a,
@@ -746,7 +746,7 @@ static const llc_conn_action_t llc_normal_actions_5b[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_5b = {
+static const struct llc_conn_state_trans llc_normal_state_trans_5b = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_normal_ev_qfyrs_5b,
@@ -769,7 +769,7 @@ static const llc_conn_action_t llc_normal_actions_5c[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_5c = {
+static const struct llc_conn_state_trans llc_normal_state_trans_5c = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_1_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_normal_ev_qfyrs_5c,
@@ -790,7 +790,7 @@ static const llc_conn_action_t llc_normal_actions_6a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_6a = {
+static const struct llc_conn_state_trans llc_normal_state_trans_6a = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_normal_ev_qfyrs_6a,
@@ -811,7 +811,7 @@ static const llc_conn_action_t llc_normal_actions_6b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_6b = {
+static const struct llc_conn_state_trans llc_normal_state_trans_6b = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_normal_ev_qfyrs_6b,
@@ -827,7 +827,7 @@ static const llc_conn_action_t llc_normal_actions_7[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_7 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_7 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -850,7 +850,7 @@ static const llc_conn_action_t llc_normal_actions_8[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_8a = {
+static const struct llc_conn_state_trans llc_normal_state_trans_8a = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_8a,
@@ -863,7 +863,7 @@ static const llc_conn_ev_qfyr_t llc_normal_ev_qfyrs_8b[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_8b = {
+static const struct llc_conn_state_trans llc_normal_state_trans_8b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_8b,
@@ -884,7 +884,7 @@ static const llc_conn_action_t llc_normal_actions_9a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_9a = {
+static const struct llc_conn_state_trans llc_normal_state_trans_9a = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_9a,
@@ -905,7 +905,7 @@ static const llc_conn_action_t llc_normal_actions_9b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_9b = {
+static const struct llc_conn_state_trans llc_normal_state_trans_9b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_9b,
@@ -922,7 +922,7 @@ static const llc_conn_action_t llc_normal_actions_10[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_10 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_10 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -937,7 +937,7 @@ static const llc_conn_action_t llc_normal_actions_11a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_11a = {
+static const struct llc_conn_state_trans llc_normal_state_trans_11a = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -952,7 +952,7 @@ static const llc_conn_action_t llc_normal_actions_11b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_11b = {
+static const struct llc_conn_state_trans llc_normal_state_trans_11b = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -973,7 +973,7 @@ static const llc_conn_action_t llc_normal_actions_11c[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_11c = {
+static const struct llc_conn_state_trans llc_normal_state_trans_11c = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_11c,
@@ -990,7 +990,7 @@ static const llc_conn_action_t llc_normal_actions_12[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_12 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_12 = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -1005,7 +1005,7 @@ static const llc_conn_action_t llc_normal_actions_13a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_13a = {
+static const struct llc_conn_state_trans llc_normal_state_trans_13a = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -1020,7 +1020,7 @@ static const llc_conn_action_t llc_normal_actions_13b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_13b = {
+static const struct llc_conn_state_trans llc_normal_state_trans_13b = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -1040,7 +1040,7 @@ static const llc_conn_action_t llc_normal_actions_13c[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_13c = {
+static const struct llc_conn_state_trans llc_normal_state_trans_13c = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_13c,
@@ -1057,7 +1057,7 @@ static const llc_conn_action_t llc_normal_actions_14[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_14 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_14 = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -1080,7 +1080,7 @@ static const llc_conn_action_t llc_normal_actions_15a[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_15a = {
+static const struct llc_conn_state_trans llc_normal_state_trans_15a = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_15a,
@@ -1103,7 +1103,7 @@ static const llc_conn_action_t llc_normal_actions_15b[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_15b = {
+static const struct llc_conn_state_trans llc_normal_state_trans_15b = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_15b,
@@ -1125,7 +1125,7 @@ static const llc_conn_action_t llc_normal_actions_16a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_16a = {
+static const struct llc_conn_state_trans llc_normal_state_trans_16a = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_16a,
@@ -1147,7 +1147,7 @@ static const llc_conn_action_t llc_normal_actions_16b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_16b = {
+static const struct llc_conn_state_trans llc_normal_state_trans_16b = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_16b,
@@ -1164,7 +1164,7 @@ static const llc_conn_action_t llc_normal_actions_17[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_17 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_17 = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -1183,7 +1183,7 @@ static const llc_conn_action_t llc_normal_actions_18[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_18 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_18 = {
 	.ev	       = llc_conn_ev_init_p_f_cycle,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_18,
@@ -1205,7 +1205,7 @@ static const llc_conn_action_t llc_normal_actions_19[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_19 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_19 = {
 	.ev	       = llc_conn_ev_p_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = llc_normal_ev_qfyrs_19,
@@ -1228,7 +1228,7 @@ static const llc_conn_action_t llc_normal_actions_20a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_20a = {
+static const struct llc_conn_state_trans llc_normal_state_trans_20a = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = llc_normal_ev_qfyrs_20a,
@@ -1251,7 +1251,7 @@ static const llc_conn_action_t llc_normal_actions_20b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_20b = {
+static const struct llc_conn_state_trans llc_normal_state_trans_20b = {
 	.ev	       = llc_conn_ev_busy_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = llc_normal_ev_qfyrs_20b,
@@ -1270,7 +1270,7 @@ static const llc_conn_action_t llc_normal_actions_21[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_normal_state_trans_21 = {
+static const struct llc_conn_state_trans llc_normal_state_trans_21 = {
 	.ev	       = llc_conn_ev_tx_buffer_full,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_normal_ev_qfyrs_21,
@@ -1281,7 +1281,7 @@ static struct llc_conn_state_trans llc_normal_state_trans_21 = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_normal_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_normal_state_transitions[] = {
 	 [0] = &llc_normal_state_trans_1,	/* Requests */
 	 [1] = &llc_normal_state_trans_2,
 	 [2] = &llc_normal_state_trans_2_1,
@@ -1354,7 +1354,7 @@ static const llc_conn_action_t llc_busy_actions_1[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_1 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_1 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_1,
@@ -1374,7 +1374,7 @@ static const llc_conn_action_t llc_busy_actions_2[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_2 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_2 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_2,
@@ -1391,7 +1391,7 @@ static const llc_conn_ev_qfyr_t llc_busy_ev_qfyrs_2_1[] = {
 /* just one member, NULL, .bss zeroes it */
 static const llc_conn_action_t llc_busy_actions_2_1[1];
 
-static struct llc_conn_state_trans llc_busy_state_trans_2_1 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_2_1 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_2_1,
@@ -1411,7 +1411,7 @@ static const llc_conn_action_t llc_busy_actions_3[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_3 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_3 = {
 	.ev	       = llc_conn_ev_local_busy_cleared,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_busy_ev_qfyrs_3,
@@ -1431,7 +1431,7 @@ static const llc_conn_action_t llc_busy_actions_4[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_4 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_4 = {
 	.ev	       = llc_conn_ev_local_busy_cleared,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_busy_ev_qfyrs_4,
@@ -1450,7 +1450,7 @@ static const llc_conn_action_t llc_busy_actions_5[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_5 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_5 = {
 	.ev	       = llc_conn_ev_local_busy_cleared,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_busy_ev_qfyrs_5,
@@ -1469,7 +1469,7 @@ static const llc_conn_action_t llc_busy_actions_6[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_6 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_6 = {
 	.ev	       = llc_conn_ev_local_busy_cleared,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_busy_ev_qfyrs_6,
@@ -1488,7 +1488,7 @@ static const llc_conn_action_t llc_busy_actions_7[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_7 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_7 = {
 	.ev	       = llc_conn_ev_local_busy_cleared,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_busy_ev_qfyrs_7,
@@ -1507,7 +1507,7 @@ static const llc_conn_action_t llc_busy_actions_8[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_8 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_8 = {
 	.ev	       = llc_conn_ev_local_busy_cleared,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_busy_ev_qfyrs_8,
@@ -1529,7 +1529,7 @@ static const llc_conn_action_t llc_busy_actions_9a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_9a = {
+static const struct llc_conn_state_trans llc_busy_state_trans_9a = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_x_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_9a,
@@ -1551,7 +1551,7 @@ static const llc_conn_action_t llc_busy_actions_9b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_9b = {
+static const struct llc_conn_state_trans llc_busy_state_trans_9b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_9b,
@@ -1571,7 +1571,7 @@ static const llc_conn_action_t llc_busy_actions_10a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_10a = {
+static const struct llc_conn_state_trans llc_busy_state_trans_10a = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_10a,
@@ -1591,7 +1591,7 @@ static const llc_conn_action_t llc_busy_actions_10b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_10b = {
+static const struct llc_conn_state_trans llc_busy_state_trans_10b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_10b,
@@ -1606,7 +1606,7 @@ static const llc_conn_action_t llc_busy_actions_11[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_11 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_11 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -1624,7 +1624,7 @@ static const llc_conn_action_t llc_busy_actions_12[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_12 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_12 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -1649,7 +1649,7 @@ static const llc_conn_action_t llc_busy_actions_13a[] = {
 	[8] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_13a = {
+static const struct llc_conn_state_trans llc_busy_state_trans_13a = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_13a,
@@ -1674,7 +1674,7 @@ static const llc_conn_action_t llc_busy_actions_13b[] = {
 	[8] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_13b = {
+static const struct llc_conn_state_trans llc_busy_state_trans_13b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_13b,
@@ -1697,7 +1697,7 @@ static const llc_conn_action_t llc_busy_actions_14a[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_14a = {
+static const struct llc_conn_state_trans llc_busy_state_trans_14a = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_14a,
@@ -1720,7 +1720,7 @@ static const llc_conn_action_t llc_busy_actions_14b[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_14b = {
+static const struct llc_conn_state_trans llc_busy_state_trans_14b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_14b,
@@ -1735,7 +1735,7 @@ static const llc_conn_action_t llc_busy_actions_15a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_15a = {
+static const struct llc_conn_state_trans llc_busy_state_trans_15a = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -1750,7 +1750,7 @@ static const llc_conn_action_t llc_busy_actions_15b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_15b = {
+static const struct llc_conn_state_trans llc_busy_state_trans_15b = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -1770,7 +1770,7 @@ static const llc_conn_action_t llc_busy_actions_15c[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_15c = {
+static const struct llc_conn_state_trans llc_busy_state_trans_15c = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_15c,
@@ -1785,7 +1785,7 @@ static const llc_conn_action_t llc_busy_actions_16[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_16 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_16 = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -1800,7 +1800,7 @@ static const llc_conn_action_t llc_busy_actions_17a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_17a = {
+static const struct llc_conn_state_trans llc_busy_state_trans_17a = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -1815,7 +1815,7 @@ static const llc_conn_action_t llc_busy_actions_17b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_17b = {
+static const struct llc_conn_state_trans llc_busy_state_trans_17b = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -1835,7 +1835,7 @@ static const llc_conn_action_t llc_busy_actions_17c[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_17c = {
+static const struct llc_conn_state_trans llc_busy_state_trans_17c = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_17c,
@@ -1850,7 +1850,7 @@ static const llc_conn_action_t llc_busy_actions_18[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_18 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_18 = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -1872,7 +1872,7 @@ static const llc_conn_action_t llc_busy_actions_19a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_19a = {
+static const struct llc_conn_state_trans llc_busy_state_trans_19a = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_19a,
@@ -1894,7 +1894,7 @@ static const llc_conn_action_t llc_busy_actions_19b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_19b = {
+static const struct llc_conn_state_trans llc_busy_state_trans_19b = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_19b,
@@ -1915,7 +1915,7 @@ static const llc_conn_action_t llc_busy_actions_20a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_20a = {
+static const struct llc_conn_state_trans llc_busy_state_trans_20a = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_20a,
@@ -1936,7 +1936,7 @@ static const llc_conn_action_t llc_busy_actions_20b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_20b = {
+static const struct llc_conn_state_trans llc_busy_state_trans_20b = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_20b,
@@ -1953,7 +1953,7 @@ static const llc_conn_action_t llc_busy_actions_21[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_21 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_21 = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -1972,7 +1972,7 @@ static const llc_conn_action_t llc_busy_actions_22[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_22 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_22 = {
 	.ev	       = llc_conn_ev_init_p_f_cycle,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_22,
@@ -1993,7 +1993,7 @@ static const llc_conn_action_t llc_busy_actions_23[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_23 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_23 = {
 	.ev	       = llc_conn_ev_p_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_23,
@@ -2015,7 +2015,7 @@ static const llc_conn_action_t llc_busy_actions_24a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_24a = {
+static const struct llc_conn_state_trans llc_busy_state_trans_24a = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_24a,
@@ -2037,7 +2037,7 @@ static const llc_conn_action_t llc_busy_actions_24b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_24b = {
+static const struct llc_conn_state_trans llc_busy_state_trans_24b = {
 	.ev	       = llc_conn_ev_busy_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_24b,
@@ -2060,7 +2060,7 @@ static const llc_conn_action_t llc_busy_actions_25[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_25 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_25 = {
 	.ev	       = llc_conn_ev_rej_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_25,
@@ -2079,7 +2079,7 @@ static const llc_conn_action_t llc_busy_actions_26[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_busy_state_trans_26 = {
+static const struct llc_conn_state_trans llc_busy_state_trans_26 = {
 	.ev	       = llc_conn_ev_rej_tmr_exp,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_busy_ev_qfyrs_26,
@@ -2090,7 +2090,7 @@ static struct llc_conn_state_trans llc_busy_state_trans_26 = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_busy_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_busy_state_transitions[] = {
 	 [0] = &llc_common_state_trans_1,	/* Request */
 	 [1] = &llc_common_state_trans_2,
 	 [2] = &llc_busy_state_trans_1,
@@ -2166,7 +2166,7 @@ static const llc_conn_action_t llc_reject_actions_1[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_1 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_1 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_1,
@@ -2185,7 +2185,7 @@ static const llc_conn_action_t llc_reject_actions_2[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_2 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_2 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_2,
@@ -2202,7 +2202,7 @@ static const llc_conn_ev_qfyr_t llc_reject_ev_qfyrs_2_1[] = {
 /* just one member, NULL, .bss zeroes it */
 static const llc_conn_action_t llc_reject_actions_2_1[1];
 
-static struct llc_conn_state_trans llc_reject_state_trans_2_1 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_2_1 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_2_1,
@@ -2222,7 +2222,7 @@ static const llc_conn_action_t llc_reject_actions_3[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_3 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_3 = {
 	.ev	       = llc_conn_ev_local_busy_detected,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_reject_ev_qfyrs_3,
@@ -2241,7 +2241,7 @@ static const llc_conn_action_t llc_reject_actions_4[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_4 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_4 = {
 	.ev	       = llc_conn_ev_local_busy_detected,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = llc_reject_ev_qfyrs_4,
@@ -2256,7 +2256,7 @@ static const llc_conn_action_t llc_reject_actions_5a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_5a = {
+static const struct llc_conn_state_trans llc_reject_state_trans_5a = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -2271,7 +2271,7 @@ static const llc_conn_action_t llc_reject_actions_5b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_5b = {
+static const struct llc_conn_state_trans llc_reject_state_trans_5b = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -2291,7 +2291,7 @@ static const llc_conn_action_t llc_reject_actions_5c[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_5c = {
+static const struct llc_conn_state_trans llc_reject_state_trans_5c = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_1_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_5c,
@@ -2305,7 +2305,7 @@ static const llc_conn_action_t llc_reject_actions_6[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_6 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_6 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -2330,7 +2330,7 @@ static const llc_conn_action_t llc_reject_actions_7a[] = {
 
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_7a = {
+static const struct llc_conn_state_trans llc_reject_state_trans_7a = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_reject_ev_qfyrs_7a,
@@ -2354,7 +2354,7 @@ static const llc_conn_action_t llc_reject_actions_7b[] = {
 	[7] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_7b = {
+static const struct llc_conn_state_trans llc_reject_state_trans_7b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_reject_ev_qfyrs_7b,
@@ -2376,7 +2376,7 @@ static const llc_conn_action_t llc_reject_actions_8a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_8a = {
+static const struct llc_conn_state_trans llc_reject_state_trans_8a = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_reject_ev_qfyrs_8a,
@@ -2398,7 +2398,7 @@ static const llc_conn_action_t llc_reject_actions_8b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_8b = {
+static const struct llc_conn_state_trans llc_reject_state_trans_8b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_reject_ev_qfyrs_8b,
@@ -2415,7 +2415,7 @@ static const llc_conn_action_t llc_reject_actions_9[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_9 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_9 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -2430,7 +2430,7 @@ static const llc_conn_action_t llc_reject_actions_10a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_10a = {
+static const struct llc_conn_state_trans llc_reject_state_trans_10a = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -2445,7 +2445,7 @@ static const llc_conn_action_t llc_reject_actions_10b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_10b = {
+static const struct llc_conn_state_trans llc_reject_state_trans_10b = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -2465,7 +2465,7 @@ static const llc_conn_action_t llc_reject_actions_10c[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_10c = {
+static const struct llc_conn_state_trans llc_reject_state_trans_10c = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_10c,
@@ -2480,7 +2480,7 @@ static const llc_conn_action_t llc_reject_actions_11[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_11 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_11 = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -2495,7 +2495,7 @@ static const llc_conn_action_t llc_reject_actions_12a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_12a = {
+static const struct llc_conn_state_trans llc_reject_state_trans_12a = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -2510,7 +2510,7 @@ static const llc_conn_action_t llc_reject_actions_12b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_12b = {
+static const struct llc_conn_state_trans llc_reject_state_trans_12b = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -2530,7 +2530,7 @@ static const llc_conn_action_t llc_reject_actions_12c[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_12c = {
+static const struct llc_conn_state_trans llc_reject_state_trans_12c = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_12c,
@@ -2545,7 +2545,7 @@ static const llc_conn_action_t llc_reject_actions_13[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_13 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_13 = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -2567,7 +2567,7 @@ static const llc_conn_action_t llc_reject_actions_14a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_14a = {
+static const struct llc_conn_state_trans llc_reject_state_trans_14a = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_14a,
@@ -2589,7 +2589,7 @@ static const llc_conn_action_t llc_reject_actions_14b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_14b = {
+static const struct llc_conn_state_trans llc_reject_state_trans_14b = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_14b,
@@ -2610,7 +2610,7 @@ static const llc_conn_action_t llc_reject_actions_15a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_15a = {
+static const struct llc_conn_state_trans llc_reject_state_trans_15a = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_15a,
@@ -2631,7 +2631,7 @@ static const llc_conn_action_t llc_reject_actions_15b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_15b = {
+static const struct llc_conn_state_trans llc_reject_state_trans_15b = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_15b,
@@ -2647,7 +2647,7 @@ static const llc_conn_action_t llc_reject_actions_16[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_16 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_16 = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -2666,7 +2666,7 @@ static const llc_conn_action_t llc_reject_actions_17[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_17 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_17 = {
 	.ev	       = llc_conn_ev_init_p_f_cycle,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_17,
@@ -2688,7 +2688,7 @@ static const llc_conn_action_t llc_reject_actions_18[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_18 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_18 = {
 	.ev	       = llc_conn_ev_rej_tmr_exp,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_18,
@@ -2710,7 +2710,7 @@ static const llc_conn_action_t llc_reject_actions_19[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_19 = {
+static const struct llc_conn_state_trans llc_reject_state_trans_19 = {
 	.ev	       = llc_conn_ev_p_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_19,
@@ -2733,7 +2733,7 @@ static const llc_conn_action_t llc_reject_actions_20a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_20a = {
+static const struct llc_conn_state_trans llc_reject_state_trans_20a = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_20a,
@@ -2756,7 +2756,7 @@ static const llc_conn_action_t llc_reject_actions_20b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_reject_state_trans_20b = {
+static const struct llc_conn_state_trans llc_reject_state_trans_20b = {
 	.ev	       = llc_conn_ev_busy_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = llc_reject_ev_qfyrs_20b,
@@ -2767,7 +2767,7 @@ static struct llc_conn_state_trans llc_reject_state_trans_20b = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_reject_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_reject_state_transitions[] = {
 	 [0] = &llc_common_state_trans_1,	/* Request */
 	 [1] = &llc_common_state_trans_2,
 	 [2] = &llc_common_state_trans_end,
@@ -2834,7 +2834,7 @@ static const llc_conn_ev_qfyr_t llc_await_ev_qfyrs_1_0[] = {
 /* just one member, NULL, .bss zeroes it */
 static const llc_conn_action_t llc_await_actions_1_0[1];
 
-static struct llc_conn_state_trans llc_await_state_trans_1_0 = {
+static const struct llc_conn_state_trans llc_await_state_trans_1_0 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = llc_await_ev_qfyrs_1_0,
@@ -2848,7 +2848,7 @@ static const llc_conn_action_t llc_await_actions_1[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_1 = {
+static const struct llc_conn_state_trans llc_await_state_trans_1 = {
 	.ev	       = llc_conn_ev_local_busy_detected,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -2867,7 +2867,7 @@ static const llc_conn_action_t llc_await_actions_2[] = {
 	[7] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_2 = {
+static const struct llc_conn_state_trans llc_await_state_trans_2 = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_1_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -2883,7 +2883,7 @@ static const llc_conn_action_t llc_await_actions_3a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_3a = {
+static const struct llc_conn_state_trans llc_await_state_trans_3a = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -2899,7 +2899,7 @@ static const llc_conn_action_t llc_await_actions_3b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_3b = {
+static const struct llc_conn_state_trans llc_await_state_trans_3b = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -2916,7 +2916,7 @@ static const llc_conn_action_t llc_await_actions_4[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_4 = {
+static const struct llc_conn_state_trans llc_await_state_trans_4 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -2935,7 +2935,7 @@ static const llc_conn_action_t llc_await_actions_5[] = {
 	[7] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_5 = {
+static const struct llc_conn_state_trans llc_await_state_trans_5 = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -2952,7 +2952,7 @@ static const llc_conn_action_t llc_await_actions_6a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_6a = {
+static const struct llc_conn_state_trans llc_await_state_trans_6a = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -2969,7 +2969,7 @@ static const llc_conn_action_t llc_await_actions_6b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_6b = {
+static const struct llc_conn_state_trans llc_await_state_trans_6b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -2986,7 +2986,7 @@ static const llc_conn_action_t llc_await_actions_7[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_7 = {
+static const struct llc_conn_state_trans llc_await_state_trans_7 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3003,7 +3003,7 @@ static const llc_conn_action_t llc_await_actions_8a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_8a = {
+static const struct llc_conn_state_trans llc_await_state_trans_8a = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -3020,7 +3020,7 @@ static const llc_conn_action_t llc_await_actions_8b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_8b = {
+static const struct llc_conn_state_trans llc_await_state_trans_8b = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -3035,7 +3035,7 @@ static const llc_conn_action_t llc_await_actions_9a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_9a = {
+static const struct llc_conn_state_trans llc_await_state_trans_9a = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3050,7 +3050,7 @@ static const llc_conn_action_t llc_await_actions_9b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_9b = {
+static const struct llc_conn_state_trans llc_await_state_trans_9b = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3065,7 +3065,7 @@ static const llc_conn_action_t llc_await_actions_9c[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_9c = {
+static const struct llc_conn_state_trans llc_await_state_trans_9c = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3080,7 +3080,7 @@ static const llc_conn_action_t llc_await_actions_9d[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_9d = {
+static const struct llc_conn_state_trans llc_await_state_trans_9d = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3096,7 +3096,7 @@ static const llc_conn_action_t llc_await_actions_10a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_10a = {
+static const struct llc_conn_state_trans llc_await_state_trans_10a = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3112,7 +3112,7 @@ static const llc_conn_action_t llc_await_actions_10b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_10b = {
+static const struct llc_conn_state_trans llc_await_state_trans_10b = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3128,7 +3128,7 @@ static const llc_conn_action_t llc_await_actions_11[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_11 = {
+static const struct llc_conn_state_trans llc_await_state_trans_11 = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -3143,7 +3143,7 @@ static const llc_conn_action_t llc_await_actions_12a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_12a = {
+static const struct llc_conn_state_trans llc_await_state_trans_12a = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3158,7 +3158,7 @@ static const llc_conn_action_t llc_await_actions_12b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_12b = {
+static const struct llc_conn_state_trans llc_await_state_trans_12b = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3174,7 +3174,7 @@ static const llc_conn_action_t llc_await_actions_13[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_13 = {
+static const struct llc_conn_state_trans llc_await_state_trans_13 = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3194,7 +3194,7 @@ static const llc_conn_action_t llc_await_actions_14[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_state_trans_14 = {
+static const struct llc_conn_state_trans llc_await_state_trans_14 = {
 	.ev	       = llc_conn_ev_p_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = llc_await_ev_qfyrs_14,
@@ -3205,7 +3205,7 @@ static struct llc_conn_state_trans llc_await_state_trans_14 = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_await_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_await_state_transitions[] = {
 	 [0] = &llc_common_state_trans_1,	/* Request */
 	 [1] = &llc_common_state_trans_2,
 	 [2] = &llc_await_state_trans_1_0,
@@ -3263,7 +3263,7 @@ static const llc_conn_ev_qfyr_t llc_await_busy_ev_qfyrs_1_0[] = {
 /* just one member, NULL, .bss zeroes it */
 static const llc_conn_action_t llc_await_busy_actions_1_0[1];
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_1_0 = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_1_0 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = llc_await_busy_ev_qfyrs_1_0,
@@ -3282,7 +3282,7 @@ static const llc_conn_action_t llc_await_busy_actions_1[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_1 = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_1 = {
 	.ev	       = llc_conn_ev_local_busy_cleared,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = llc_await_busy_ev_qfyrs_1,
@@ -3300,7 +3300,7 @@ static const llc_conn_action_t llc_await_busy_actions_2[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_2 = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_2 = {
 	.ev	       = llc_conn_ev_local_busy_cleared,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = llc_await_busy_ev_qfyrs_2,
@@ -3318,7 +3318,7 @@ static const llc_conn_action_t llc_await_busy_actions_3[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_3 = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_3 = {
 	.ev	       = llc_conn_ev_local_busy_cleared,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = llc_await_busy_ev_qfyrs_3,
@@ -3337,7 +3337,7 @@ static const llc_conn_action_t llc_await_busy_actions_4[] = {
 	[7] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_4 = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_4 = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_1_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -3353,7 +3353,7 @@ static const llc_conn_action_t llc_await_busy_actions_5a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_5a = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_5a = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3369,7 +3369,7 @@ static const llc_conn_action_t llc_await_busy_actions_5b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_5b = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_5b = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3385,7 +3385,7 @@ static const llc_conn_action_t llc_await_busy_actions_6[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_6 = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_6 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3406,7 +3406,7 @@ static const llc_conn_action_t llc_await_busy_actions_7[] = {
 	[9] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_7 = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_7 = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -3424,7 +3424,7 @@ static const llc_conn_action_t llc_await_busy_actions_8a[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_8a = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_8a = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3442,7 +3442,7 @@ static const llc_conn_action_t llc_await_busy_actions_8b[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_8b = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_8b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3460,7 +3460,7 @@ static const llc_conn_action_t llc_await_busy_actions_9[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_9 = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_9 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3477,7 +3477,7 @@ static const llc_conn_action_t llc_await_busy_actions_10a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_10a = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_10a = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -3494,7 +3494,7 @@ static const llc_conn_action_t llc_await_busy_actions_10b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_10b = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_10b = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -3509,7 +3509,7 @@ static const llc_conn_action_t llc_await_busy_actions_11a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_11a = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_11a = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3524,7 +3524,7 @@ static const llc_conn_action_t llc_await_busy_actions_11b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_11b = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_11b = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3539,7 +3539,7 @@ static const llc_conn_action_t llc_await_busy_actions_11c[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_11c = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_11c = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3554,7 +3554,7 @@ static const llc_conn_action_t llc_await_busy_actions_11d[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_11d = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_11d = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3570,7 +3570,7 @@ static const llc_conn_action_t llc_await_busy_actions_12a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_12a = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_12a = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3586,7 +3586,7 @@ static const llc_conn_action_t llc_await_busy_actions_12b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_12b = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_12b = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3602,7 +3602,7 @@ static const llc_conn_action_t llc_await_busy_actions_13[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_13 = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_13 = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_BUSY,
 	.ev_qualifiers = NONE,
@@ -3617,7 +3617,7 @@ static const llc_conn_action_t llc_await_busy_actions_14a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_14a = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_14a = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3632,7 +3632,7 @@ static const llc_conn_action_t llc_await_busy_actions_14b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_14b = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_14b = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3648,7 +3648,7 @@ static const llc_conn_action_t llc_await_busy_actions_15[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_15 = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_15 = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3668,7 +3668,7 @@ static const llc_conn_action_t llc_await_busy_actions_16[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_busy_state_trans_16 = {
+static const struct llc_conn_state_trans llc_await_busy_state_trans_16 = {
 	.ev	       = llc_conn_ev_p_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = llc_await_busy_ev_qfyrs_16,
@@ -3679,7 +3679,7 @@ static struct llc_conn_state_trans llc_await_busy_state_trans_16 = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_await_busy_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_await_busy_state_transitions[] = {
 	 [0] = &llc_common_state_trans_1,		/* Request */
 	 [1] = &llc_common_state_trans_2,
 	 [2] = &llc_await_busy_state_trans_1_0,
@@ -3739,7 +3739,7 @@ static const llc_conn_ev_qfyr_t llc_await_reject_ev_qfyrs_1_0[] = {
 /* just one member, NULL, .bss zeroes it */
 static const llc_conn_action_t llc_await_reject_actions_1_0[1];
 
-static struct llc_conn_state_trans llc_await_reject_state_trans_1_0 = {
+static const struct llc_conn_state_trans llc_await_reject_state_trans_1_0 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = llc_await_reject_ev_qfyrs_1_0,
@@ -3753,7 +3753,7 @@ static const llc_conn_action_t llc_await_rejct_actions_1[] = {
 	[2] = NULL
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_1 = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_1 = {
 	.ev	       = llc_conn_ev_local_busy_detected,
 	.next_state    = LLC_CONN_STATE_AWAIT_BUSY,
 	.ev_qualifiers = NONE,
@@ -3767,7 +3767,7 @@ static const llc_conn_action_t llc_await_rejct_actions_2a[] = {
 	[2] = NULL
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_2a = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_2a = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -3781,7 +3781,7 @@ static const llc_conn_action_t llc_await_rejct_actions_2b[] = {
 	[2] = NULL
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_2b = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_2b = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -3796,7 +3796,7 @@ static const llc_conn_action_t llc_await_rejct_actions_3[] = {
 	[3] = NULL
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_3 = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_3 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -3816,7 +3816,7 @@ static const llc_conn_action_t llc_await_rejct_actions_4[] = {
 	[8] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_4 = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_4 = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -3834,7 +3834,7 @@ static const llc_conn_action_t llc_await_rejct_actions_5a[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_5a = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_5a = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3852,7 +3852,7 @@ static const llc_conn_action_t llc_await_rejct_actions_5b[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_5b = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_5b = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3870,7 +3870,7 @@ static const llc_conn_action_t llc_await_rejct_actions_6[] = {
 	[6] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_6 = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_6 = {
 	.ev	       = llc_conn_ev_rx_i_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT,
 	.ev_qualifiers = NONE,
@@ -3887,7 +3887,7 @@ static const llc_conn_action_t llc_await_rejct_actions_7a[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_7a = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_7a = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -3904,7 +3904,7 @@ static const llc_conn_action_t llc_await_rejct_actions_7b[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_7b = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_7b = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -3921,7 +3921,7 @@ static const llc_conn_action_t llc_await_rejct_actions_7c[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_7c = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_7c = {
 	.ev	       = llc_conn_ev_rx_i_rsp_fbit_set_1_unexpd_ns,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -3936,7 +3936,7 @@ static const llc_conn_action_t llc_await_rejct_actions_8a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_8a = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_8a = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -3951,7 +3951,7 @@ static const llc_conn_action_t llc_await_rejct_actions_8b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_8b = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_8b = {
 	.ev	       = llc_conn_ev_rx_rr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -3966,7 +3966,7 @@ static const llc_conn_action_t llc_await_rejct_actions_8c[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_8c = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_8c = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -3981,7 +3981,7 @@ static const llc_conn_action_t llc_await_rejct_actions_8d[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_8d = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_8d = {
 	.ev	       = llc_conn_ev_rx_rej_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -3997,7 +3997,7 @@ static const llc_conn_action_t llc_await_rejct_actions_9a[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_9a = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_9a = {
 	.ev	       = llc_conn_ev_rx_rr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -4013,7 +4013,7 @@ static const llc_conn_action_t llc_await_rejct_actions_9b[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_9b = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_9b = {
 	.ev	       = llc_conn_ev_rx_rej_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -4029,7 +4029,7 @@ static const llc_conn_action_t llc_await_rejct_actions_10[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_10 = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_10 = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_1,
 	.next_state    = LLC_CONN_STATE_REJ,
 	.ev_qualifiers = NONE,
@@ -4044,7 +4044,7 @@ static const llc_conn_action_t llc_await_rejct_actions_11a[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_11a = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_11a = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -4059,7 +4059,7 @@ static const llc_conn_action_t llc_await_rejct_actions_11b[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_11b = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_11b = {
 	.ev	       = llc_conn_ev_rx_rnr_rsp_fbit_set_0,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -4075,7 +4075,7 @@ static const llc_conn_action_t llc_await_rejct_actions_12[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_12 = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_12 = {
 	.ev	       = llc_conn_ev_rx_rnr_cmd_pbit_set_1,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = NONE,
@@ -4095,7 +4095,7 @@ static const llc_conn_action_t llc_await_rejct_actions_13[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_await_rejct_state_trans_13 = {
+static const struct llc_conn_state_trans llc_await_rejct_state_trans_13 = {
 	.ev	       = llc_conn_ev_p_tmr_exp,
 	.next_state    = LLC_CONN_STATE_AWAIT_REJ,
 	.ev_qualifiers = llc_await_rejct_ev_qfyrs_13,
@@ -4106,7 +4106,7 @@ static struct llc_conn_state_trans llc_await_rejct_state_trans_13 = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_await_rejct_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_await_rejct_state_transitions[] = {
 	 [0] = &llc_await_reject_state_trans_1_0,
 	 [1] = &llc_common_state_trans_1,		/* requests */
 	 [2] = &llc_common_state_trans_2,
@@ -4171,7 +4171,7 @@ static const llc_conn_action_t llc_d_conn_actions_1[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_d_conn_state_trans_1 = {
+static const struct llc_conn_state_trans llc_d_conn_state_trans_1 = {
 	.ev	       = llc_conn_ev_rx_sabme_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_d_conn_ev_qfyrs_1,
@@ -4194,7 +4194,7 @@ static const llc_conn_action_t llc_d_conn_actions_1_1[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_d_conn_state_trans_1_1 = {
+static const struct llc_conn_state_trans llc_d_conn_state_trans_1_1 = {
 	.ev	       = llc_conn_ev_rx_sabme_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_d_conn_ev_qfyrs_1_1,
@@ -4218,7 +4218,7 @@ static const llc_conn_action_t llc_d_conn_actions_2[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_d_conn_state_trans_2 = {
+static const struct llc_conn_state_trans llc_d_conn_state_trans_2 = {
 	.ev	       = llc_conn_ev_rx_ua_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_d_conn_ev_qfyrs_2,
@@ -4241,7 +4241,7 @@ static const llc_conn_action_t llc_d_conn_actions_2_1[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_d_conn_state_trans_2_1 = {
+static const struct llc_conn_state_trans llc_d_conn_state_trans_2_1 = {
 	.ev	       = llc_conn_ev_rx_ua_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_d_conn_ev_qfyrs_2_1,
@@ -4254,7 +4254,7 @@ static const llc_conn_action_t llc_d_conn_actions_3[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_d_conn_state_trans_3 = {
+static const struct llc_conn_state_trans llc_d_conn_state_trans_3 = {
 	.ev	       = llc_conn_ev_rx_disc_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_D_CONN,
 	.ev_qualifiers = NONE,
@@ -4277,7 +4277,7 @@ static const llc_conn_action_t llc_d_conn_actions_4[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_d_conn_state_trans_4 = {
+static const struct llc_conn_state_trans llc_d_conn_state_trans_4 = {
 	.ev	       = llc_conn_ev_rx_dm_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_d_conn_ev_qfyrs_4,
@@ -4299,7 +4299,7 @@ static const llc_conn_action_t llc_d_conn_actions_4_1[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_d_conn_state_trans_4_1 = {
+static const struct llc_conn_state_trans llc_d_conn_state_trans_4_1 = {
 	.ev	       = llc_conn_ev_rx_dm_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_d_conn_ev_qfyrs_4_1,
@@ -4318,7 +4318,7 @@ static const llc_conn_ev_qfyr_t llc_d_conn_ev_qfyrs_5[] = {
 /* just one member, NULL, .bss zeroes it */
 static const llc_conn_action_t llc_d_conn_actions_5[1];
 
-static struct llc_conn_state_trans llc_d_conn_state_trans_5 = {
+static const struct llc_conn_state_trans llc_d_conn_state_trans_5 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_D_CONN,
 	.ev_qualifiers = llc_d_conn_ev_qfyrs_5,
@@ -4338,7 +4338,7 @@ static const llc_conn_action_t llc_d_conn_actions_6[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_d_conn_state_trans_6 = {
+static const struct llc_conn_state_trans llc_d_conn_state_trans_6 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_D_CONN,
 	.ev_qualifiers = llc_d_conn_ev_qfyrs_6,
@@ -4359,7 +4359,7 @@ static const llc_conn_action_t llc_d_conn_actions_7[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_d_conn_state_trans_7 = {
+static const struct llc_conn_state_trans llc_d_conn_state_trans_7 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_d_conn_ev_qfyrs_7,
@@ -4379,7 +4379,7 @@ static const llc_conn_action_t llc_d_conn_actions_8[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_d_conn_state_trans_8 = {
+static const struct llc_conn_state_trans llc_d_conn_state_trans_8 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_d_conn_ev_qfyrs_8,
@@ -4390,7 +4390,7 @@ static struct llc_conn_state_trans llc_d_conn_state_trans_8 = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_d_conn_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_d_conn_state_transitions[] = {
 	 [0] = &llc_d_conn_state_trans_5,	/* Request */
 	 [1] = &llc_common_state_trans_end,
 	 [2] = &llc_common_state_trans_end,	/* Local busy */
@@ -4419,7 +4419,7 @@ static const llc_conn_action_t llc_rst_actions_1[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_rst_state_trans_1 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_1 = {
 	.ev	       = llc_conn_ev_rx_sabme_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_RESET,
 	.ev_qualifiers = NONE,
@@ -4447,7 +4447,7 @@ static const llc_conn_action_t llc_rst_actions_2[] = {
 	[7] = NULL,
 };
 
-static struct llc_conn_state_trans llc_rst_state_trans_2 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_2 = {
 	.ev	       = llc_conn_ev_rx_ua_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_rst_ev_qfyrs_2,
@@ -4475,7 +4475,7 @@ static const llc_conn_action_t llc_rst_actions_2_1[] = {
 	[7] = NULL,
 };
 
-static struct llc_conn_state_trans llc_rst_state_trans_2_1 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_2_1 = {
 	.ev	       = llc_conn_ev_rx_ua_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_rst_ev_qfyrs_2_1,
@@ -4495,7 +4495,7 @@ static const llc_conn_action_t llc_rst_actions_3[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_rst_state_trans_3 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_3 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = llc_rst_ev_qfyrs_3,
@@ -4518,7 +4518,7 @@ static const llc_conn_action_t llc_rst_actions_4[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_rst_state_trans_4 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_4 = {
 	.ev	       = llc_conn_ev_rx_disc_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_rst_ev_qfyrs_4,
@@ -4541,7 +4541,7 @@ static const llc_conn_action_t llc_rst_actions_4_1[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_rst_state_trans_4_1 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_4_1 = {
 	.ev	       = llc_conn_ev_rx_disc_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_rst_ev_qfyrs_4_1,
@@ -4564,7 +4564,7 @@ static const llc_conn_action_t llc_rst_actions_5[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_rst_state_trans_5 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_5 = {
 	.ev	       = llc_conn_ev_rx_dm_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_rst_ev_qfyrs_5,
@@ -4586,7 +4586,7 @@ static const llc_conn_action_t llc_rst_actions_5_1[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_rst_state_trans_5_1 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_5_1 = {
 	.ev	       = llc_conn_ev_rx_dm_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_rst_ev_qfyrs_5_1,
@@ -4602,7 +4602,7 @@ static const llc_conn_ev_qfyr_t llc_rst_ev_qfyrs_6[] = {
 /* just one member, NULL, .bss zeroes it */
 static const llc_conn_action_t llc_rst_actions_6[1];
 
-static struct llc_conn_state_trans llc_rst_state_trans_6 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_6 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_RESET,
 	.ev_qualifiers = llc_rst_ev_qfyrs_6,
@@ -4623,7 +4623,7 @@ static const llc_conn_action_t llc_rst_actions_7[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_rst_state_trans_7 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_7 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_RESET,
 	.ev_qualifiers = llc_rst_ev_qfyrs_7,
@@ -4644,7 +4644,7 @@ static const llc_conn_action_t llc_rst_actions_8[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_rst_state_trans_8 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_8 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_rst_ev_qfyrs_8,
@@ -4665,7 +4665,7 @@ static const llc_conn_action_t llc_rst_actions_8_1[] = {
 	[2] = NULL,
 };
 
-static struct llc_conn_state_trans llc_rst_state_trans_8_1 = {
+static const struct llc_conn_state_trans llc_rst_state_trans_8_1 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = llc_rst_ev_qfyrs_8_1,
@@ -4676,7 +4676,7 @@ static struct llc_conn_state_trans llc_rst_state_trans_8_1 = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_rst_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_rst_state_transitions[] = {
 	 [0] = &llc_rst_state_trans_6,		/* Request */
 	 [1] = &llc_common_state_trans_end,
 	 [2] = &llc_common_state_trans_end,	/* Local busy */
@@ -4710,7 +4710,7 @@ static const llc_conn_action_t llc_error_actions_1[] = {
 	[8] = NULL,
 };
 
-static struct llc_conn_state_trans llc_error_state_trans_1 = {
+static const struct llc_conn_state_trans llc_error_state_trans_1 = {
 	.ev	       = llc_conn_ev_rx_sabme_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_NORMAL,
 	.ev_qualifiers = NONE,
@@ -4726,7 +4726,7 @@ static const llc_conn_action_t llc_error_actions_2[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_error_state_trans_2 = {
+static const struct llc_conn_state_trans llc_error_state_trans_2 = {
 	.ev	       = llc_conn_ev_rx_disc_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = NONE,
@@ -4741,7 +4741,7 @@ static const llc_conn_action_t llc_error_actions_3[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_error_state_trans_3 = {
+static const struct llc_conn_state_trans llc_error_state_trans_3 = {
 	.ev	       = llc_conn_ev_rx_dm_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = NONE,
@@ -4757,7 +4757,7 @@ static const llc_conn_action_t llc_error_actions_4[] = {
 	[4] = NULL,
 };
 
-static struct llc_conn_state_trans llc_error_state_trans_4 = {
+static const struct llc_conn_state_trans llc_error_state_trans_4 = {
 	.ev	       = llc_conn_ev_rx_frmr_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_RESET,
 	.ev_qualifiers = NONE,
@@ -4770,7 +4770,7 @@ static const llc_conn_action_t llc_error_actions_5[] = {
 	[1] = NULL,
 };
 
-static struct llc_conn_state_trans llc_error_state_trans_5 = {
+static const struct llc_conn_state_trans llc_error_state_trans_5 = {
 	.ev	       = llc_conn_ev_rx_xxx_cmd_pbit_set_x,
 	.next_state    = LLC_CONN_STATE_ERROR,
 	.ev_qualifiers = NONE,
@@ -4778,7 +4778,7 @@ static struct llc_conn_state_trans llc_error_state_trans_5 = {
 };
 
 /* State transitions for LLC_CONN_EV_RX_XXX_RSP_Fbit_SET_X event */
-static struct llc_conn_state_trans llc_error_state_trans_6 = {
+static const struct llc_conn_state_trans llc_error_state_trans_6 = {
 	.ev	       = llc_conn_ev_rx_xxx_rsp_fbit_set_x,
 	.next_state    = LLC_CONN_STATE_ERROR,
 	.ev_qualifiers = NONE,
@@ -4798,7 +4798,7 @@ static const llc_conn_action_t llc_error_actions_7[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_error_state_trans_7 = {
+static const struct llc_conn_state_trans llc_error_state_trans_7 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_ERROR,
 	.ev_qualifiers = llc_error_ev_qfyrs_7,
@@ -4820,7 +4820,7 @@ static const llc_conn_action_t llc_error_actions_8[] = {
 	[5] = NULL,
 };
 
-static struct llc_conn_state_trans llc_error_state_trans_8 = {
+static const struct llc_conn_state_trans llc_error_state_trans_8 = {
 	.ev	       = llc_conn_ev_ack_tmr_exp,
 	.next_state    = LLC_CONN_STATE_RESET,
 	.ev_qualifiers = llc_error_ev_qfyrs_8,
@@ -4836,7 +4836,7 @@ static const llc_conn_ev_qfyr_t llc_error_ev_qfyrs_9[] = {
 /* just one member, NULL, .bss zeroes it */
 static const llc_conn_action_t llc_error_actions_9[1];
 
-static struct llc_conn_state_trans llc_error_state_trans_9 = {
+static const struct llc_conn_state_trans llc_error_state_trans_9 = {
 	.ev	       = llc_conn_ev_data_req,
 	.next_state    = LLC_CONN_STATE_ERROR,
 	.ev_qualifiers = llc_error_ev_qfyrs_9,
@@ -4847,7 +4847,7 @@ static struct llc_conn_state_trans llc_error_state_trans_9 = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_error_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_error_state_transitions[] = {
 	 [0] = &llc_error_state_trans_9,	/* Request */
 	 [1] = &llc_common_state_trans_end,
 	 [2] = &llc_common_state_trans_end,	/* Local busy */
@@ -4873,7 +4873,7 @@ static const llc_conn_action_t llc_temp_actions_1[] = {
 	[3] = NULL,
 };
 
-static struct llc_conn_state_trans llc_temp_state_trans_1 = {
+static const struct llc_conn_state_trans llc_temp_state_trans_1 = {
 	.ev	       = llc_conn_ev_disc_req,
 	.next_state    = LLC_CONN_STATE_ADM,
 	.ev_qualifiers = NONE,
@@ -4884,7 +4884,7 @@ static struct llc_conn_state_trans llc_temp_state_trans_1 = {
  * Array of pointers;
  * one to each transition
  */
-static struct llc_conn_state_trans *llc_temp_state_transitions[] = {
+static const struct llc_conn_state_trans *llc_temp_state_transitions[] = {
 	[0] = &llc_temp_state_trans_1,		/* requests */
 	[1] = &llc_common_state_trans_end,
 	[2] = &llc_common_state_trans_end,	/* local busy */
diff --git a/net/llc/llc_conn.c b/net/llc/llc_conn.c
index 0a3f5e0bec00..afc6974eafda 100644
--- a/net/llc/llc_conn.c
+++ b/net/llc/llc_conn.c
@@ -34,10 +34,10 @@ static int llc_find_offset(int state, int ev_type);
 static void llc_conn_send_pdus(struct sock *sk);
 static int llc_conn_service(struct sock *sk, struct sk_buff *skb);
 static int llc_exec_conn_trans_actions(struct sock *sk,
-				       struct llc_conn_state_trans *trans,
+				       const struct llc_conn_state_trans *trans,
 				       struct sk_buff *ev);
-static struct llc_conn_state_trans *llc_qualify_conn_ev(struct sock *sk,
-							struct sk_buff *skb);
+static const struct llc_conn_state_trans *llc_qualify_conn_ev(struct sock *sk,
+							      struct sk_buff *skb);
 
 /* Offset table on connection states transition diagram */
 static int llc_offset_table[NBR_CONN_STATES][NBR_CONN_EV];
@@ -356,9 +356,9 @@ static void llc_conn_send_pdus(struct sock *sk)
  */
 static int llc_conn_service(struct sock *sk, struct sk_buff *skb)
 {
-	int rc = 1;
+	const struct llc_conn_state_trans *trans;
 	struct llc_sock *llc = llc_sk(sk);
-	struct llc_conn_state_trans *trans;
+	int rc = 1;
 
 	if (llc->state > NBR_CONN_STATES)
 		goto out;
@@ -384,10 +384,10 @@ out:
  *	This function finds transition that matches with happened event.
  *	Returns pointer to found transition on success, %NULL otherwise.
  */
-static struct llc_conn_state_trans *llc_qualify_conn_ev(struct sock *sk,
-							struct sk_buff *skb)
+static const struct llc_conn_state_trans *llc_qualify_conn_ev(struct sock *sk,
+							      struct sk_buff *skb)
 {
-	struct llc_conn_state_trans **next_trans;
+	const struct llc_conn_state_trans **next_trans;
 	const llc_conn_ev_qfyr_t *next_qualifier;
 	struct llc_conn_state_ev *ev = llc_conn_ev(skb);
 	struct llc_sock *llc = llc_sk(sk);
@@ -432,7 +432,7 @@ static struct llc_conn_state_trans *llc_qualify_conn_ev(struct sock *sk,
  *	success, 1 to indicate failure of at least one action.
  */
 static int llc_exec_conn_trans_actions(struct sock *sk,
-				       struct llc_conn_state_trans *trans,
+				       const struct llc_conn_state_trans *trans,
 				       struct sk_buff *skb)
 {
 	int rc = 0;
@@ -635,8 +635,8 @@ u8 llc_data_accept_state(u8 state)
  */
 static u16 __init llc_find_next_offset(struct llc_conn_state *state, u16 offset)
 {
+	const struct llc_conn_state_trans **next_trans;
 	u16 cnt = 0;
-	struct llc_conn_state_trans **next_trans;
 
 	for (next_trans = state->transitions + offset;
 	     (*next_trans)->ev; next_trans++)
-- 
cgit v1.2.3-58-ga151


From 0970bf676f86c2c4d9bf7e672f5504d390c9fce6 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sun, 14 Jul 2024 18:15:20 +0200
Subject: llc: Constify struct llc_sap_state_trans

'struct llc_sap_state_trans' are not modified in this driver.

Constifying this structure moves some data to a read-only section, so
increase overall security.

On a x86_64, with allmodconfig, as an example:
Before:
======
   text	   data	    bss	    dec	    hex	filename
    339	    456	     24	    819	    333	net/llc/llc_s_st.o

After:
=====
   text	   data	    bss	    dec	    hex	filename
    683	    144	      0	    827	    33b	net/llc/llc_s_st.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/9d17587639195ee94b74ff06a11ef97d1833ee52.1720973710.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/llc_s_st.h |  4 ++--
 net/llc/llc_s_st.c     | 26 +++++++++++++-------------
 net/llc/llc_sap.c      | 12 ++++++------
 3 files changed, 21 insertions(+), 21 deletions(-)

diff --git a/include/net/llc_s_st.h b/include/net/llc_s_st.h
index ed5b2fa40d32..fca49d483d20 100644
--- a/include/net/llc_s_st.h
+++ b/include/net/llc_s_st.h
@@ -29,8 +29,8 @@ struct llc_sap_state_trans {
 };
 
 struct llc_sap_state {
-	u8			   curr_state;
-	struct llc_sap_state_trans **transitions;
+	u8				 curr_state;
+	const struct llc_sap_state_trans **transitions;
 };
 
 /* only access to SAP state table */
diff --git a/net/llc/llc_s_st.c b/net/llc/llc_s_st.c
index 308c616883a4..acccc827c562 100644
--- a/net/llc/llc_s_st.c
+++ b/net/llc/llc_s_st.c
@@ -24,7 +24,7 @@
  * last entry for this state
  * all members are zeros, .bss zeroes it
  */
-static struct llc_sap_state_trans llc_sap_state_trans_end;
+static const struct llc_sap_state_trans llc_sap_state_trans_end;
 
 /* state LLC_SAP_STATE_INACTIVE transition for
  * LLC_SAP_EV_ACTIVATION_REQ event
@@ -34,14 +34,14 @@ static const llc_sap_action_t llc_sap_inactive_state_actions_1[] = {
 	[1] = NULL,
 };
 
-static struct llc_sap_state_trans llc_sap_inactive_state_trans_1 = {
+static const struct llc_sap_state_trans llc_sap_inactive_state_trans_1 = {
 	.ev =		llc_sap_ev_activation_req,
 	.next_state =	LLC_SAP_STATE_ACTIVE,
 	.ev_actions =	llc_sap_inactive_state_actions_1,
 };
 
 /* array of pointers; one to each transition */
-static struct llc_sap_state_trans *llc_sap_inactive_state_transitions[] = {
+static const struct llc_sap_state_trans *llc_sap_inactive_state_transitions[] = {
 	[0] = &llc_sap_inactive_state_trans_1,
 	[1] = &llc_sap_state_trans_end,
 };
@@ -52,7 +52,7 @@ static const llc_sap_action_t llc_sap_active_state_actions_1[] = {
 	[1] = NULL,
 };
 
-static struct llc_sap_state_trans llc_sap_active_state_trans_1 = {
+static const struct llc_sap_state_trans llc_sap_active_state_trans_1 = {
 	.ev =		llc_sap_ev_rx_ui,
 	.next_state =	LLC_SAP_STATE_ACTIVE,
 	.ev_actions =	llc_sap_active_state_actions_1,
@@ -64,7 +64,7 @@ static const llc_sap_action_t llc_sap_active_state_actions_2[] = {
 	[1] = NULL,
 };
 
-static struct llc_sap_state_trans llc_sap_active_state_trans_2 = {
+static const struct llc_sap_state_trans llc_sap_active_state_trans_2 = {
 	.ev =		llc_sap_ev_unitdata_req,
 	.next_state =	LLC_SAP_STATE_ACTIVE,
 	.ev_actions =	llc_sap_active_state_actions_2,
@@ -76,7 +76,7 @@ static const llc_sap_action_t llc_sap_active_state_actions_3[] = {
 	[1] = NULL,
 };
 
-static struct llc_sap_state_trans llc_sap_active_state_trans_3 = {
+static const struct llc_sap_state_trans llc_sap_active_state_trans_3 = {
 	.ev =		llc_sap_ev_xid_req,
 	.next_state =	LLC_SAP_STATE_ACTIVE,
 	.ev_actions =	llc_sap_active_state_actions_3,
@@ -88,7 +88,7 @@ static const llc_sap_action_t llc_sap_active_state_actions_4[] = {
 	[1] = NULL,
 };
 
-static struct llc_sap_state_trans llc_sap_active_state_trans_4 = {
+static const struct llc_sap_state_trans llc_sap_active_state_trans_4 = {
 	.ev =		llc_sap_ev_rx_xid_c,
 	.next_state =	LLC_SAP_STATE_ACTIVE,
 	.ev_actions =	llc_sap_active_state_actions_4,
@@ -100,7 +100,7 @@ static const llc_sap_action_t llc_sap_active_state_actions_5[] = {
 	[1] = NULL,
 };
 
-static struct llc_sap_state_trans llc_sap_active_state_trans_5 = {
+static const struct llc_sap_state_trans llc_sap_active_state_trans_5 = {
 	.ev =		llc_sap_ev_rx_xid_r,
 	.next_state =	LLC_SAP_STATE_ACTIVE,
 	.ev_actions =	llc_sap_active_state_actions_5,
@@ -112,7 +112,7 @@ static const llc_sap_action_t llc_sap_active_state_actions_6[] = {
 	[1] = NULL,
 };
 
-static struct llc_sap_state_trans llc_sap_active_state_trans_6 = {
+static const struct llc_sap_state_trans llc_sap_active_state_trans_6 = {
 	.ev =		llc_sap_ev_test_req,
 	.next_state =	LLC_SAP_STATE_ACTIVE,
 	.ev_actions =	llc_sap_active_state_actions_6,
@@ -124,7 +124,7 @@ static const llc_sap_action_t llc_sap_active_state_actions_7[] = {
 	[1] = NULL,
 };
 
-static struct llc_sap_state_trans llc_sap_active_state_trans_7 = {
+static const struct llc_sap_state_trans llc_sap_active_state_trans_7 = {
 	.ev =		llc_sap_ev_rx_test_c,
 	.next_state =	LLC_SAP_STATE_ACTIVE,
 	.ev_actions =	llc_sap_active_state_actions_7
@@ -136,7 +136,7 @@ static const llc_sap_action_t llc_sap_active_state_actions_8[] = {
 	[1] = NULL,
 };
 
-static struct llc_sap_state_trans llc_sap_active_state_trans_8 = {
+static const struct llc_sap_state_trans llc_sap_active_state_trans_8 = {
 	.ev =		llc_sap_ev_rx_test_r,
 	.next_state =	LLC_SAP_STATE_ACTIVE,
 	.ev_actions =	llc_sap_active_state_actions_8,
@@ -150,14 +150,14 @@ static const llc_sap_action_t llc_sap_active_state_actions_9[] = {
 	[1] = NULL,
 };
 
-static struct llc_sap_state_trans llc_sap_active_state_trans_9 = {
+static const struct llc_sap_state_trans llc_sap_active_state_trans_9 = {
 	.ev =		llc_sap_ev_deactivation_req,
 	.next_state =	LLC_SAP_STATE_INACTIVE,
 	.ev_actions =	llc_sap_active_state_actions_9
 };
 
 /* array of pointers; one to each transition */
-static struct llc_sap_state_trans *llc_sap_active_state_transitions[] = {
+static const struct llc_sap_state_trans *llc_sap_active_state_transitions[] = {
 	[0] = &llc_sap_active_state_trans_2,
 	[1] = &llc_sap_active_state_trans_1,
 	[2] = &llc_sap_active_state_trans_3,
diff --git a/net/llc/llc_sap.c b/net/llc/llc_sap.c
index 116c0e479183..6cd03c2ae7d5 100644
--- a/net/llc/llc_sap.c
+++ b/net/llc/llc_sap.c
@@ -114,12 +114,12 @@ void llc_sap_rtn_pdu(struct llc_sap *sap, struct sk_buff *skb)
  *	Returns the pointer to found transition on success or %NULL for
  *	failure.
  */
-static struct llc_sap_state_trans *llc_find_sap_trans(struct llc_sap *sap,
-						      struct sk_buff *skb)
+static const struct llc_sap_state_trans *llc_find_sap_trans(struct llc_sap *sap,
+							    struct sk_buff *skb)
 {
 	int i = 0;
-	struct llc_sap_state_trans *rc = NULL;
-	struct llc_sap_state_trans **next_trans;
+	const struct llc_sap_state_trans *rc = NULL;
+	const struct llc_sap_state_trans **next_trans;
 	struct llc_sap_state *curr_state = &llc_sap_state_table[sap->state - 1];
 	/*
 	 * Search thru events for this state until list exhausted or until
@@ -143,7 +143,7 @@ static struct llc_sap_state_trans *llc_find_sap_trans(struct llc_sap *sap,
  *	Returns 0 for success and 1 for failure of at least one action.
  */
 static int llc_exec_sap_trans_actions(struct llc_sap *sap,
-				      struct llc_sap_state_trans *trans,
+				      const struct llc_sap_state_trans *trans,
 				      struct sk_buff *skb)
 {
 	int rc = 0;
@@ -166,8 +166,8 @@ static int llc_exec_sap_trans_actions(struct llc_sap *sap,
  */
 static int llc_sap_next_state(struct llc_sap *sap, struct sk_buff *skb)
 {
+	const struct llc_sap_state_trans *trans;
 	int rc = 1;
-	struct llc_sap_state_trans *trans;
 
 	if (sap->state > LLC_NR_SAP_STATES)
 		goto out;
-- 
cgit v1.2.3-58-ga151


From ba7a0f4436ac73850a4110acf04d65f6e8a67d82 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 15 Jul 2024 09:05:52 -0700
Subject: Revert "net: mvpp2: Improve data types and use min()"

This reverts commit f7023b3d697c6a7dfe2d9c70e0d8c2c580ccbd76.

Russell indicates that assuming 32bits are sufficient isn't
necessarily safe.

Link: https://lore.kernel.org/20240711154741.174745-1-thorsten.blum@toblux.com/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
index 33b08786992f..8c45ad983abc 100644
--- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
+++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
@@ -2766,29 +2766,29 @@ static void mvpp2_tx_pkts_coal_set(struct mvpp2_port *port,
 	}
 }
 
-static u32 mvpp2_usec_to_cycles(u32 usec, u32 clk_hz)
+static u32 mvpp2_usec_to_cycles(u32 usec, unsigned long clk_hz)
 {
 	u64 tmp = (u64)clk_hz * usec;
 
 	do_div(tmp, USEC_PER_SEC);
 
-	return min(tmp, U32_MAX);
+	return tmp > U32_MAX ? U32_MAX : tmp;
 }
 
-static u32 mvpp2_cycles_to_usec(u32 cycles, u32 clk_hz)
+static u32 mvpp2_cycles_to_usec(u32 cycles, unsigned long clk_hz)
 {
 	u64 tmp = (u64)cycles * USEC_PER_SEC;
 
 	do_div(tmp, clk_hz);
 
-	return min(tmp, U32_MAX);
+	return tmp > U32_MAX ? U32_MAX : tmp;
 }
 
 /* Set the time delay in usec before Rx interrupt */
 static void mvpp2_rx_time_coal_set(struct mvpp2_port *port,
 				   struct mvpp2_rx_queue *rxq)
 {
-	u32 freq = port->priv->tclk;
+	unsigned long freq = port->priv->tclk;
 	u32 val = mvpp2_usec_to_cycles(rxq->time_coal, freq);
 
 	if (val > MVPP2_MAX_ISR_RX_THRESHOLD) {
@@ -2804,7 +2804,7 @@ static void mvpp2_rx_time_coal_set(struct mvpp2_port *port,
 
 static void mvpp2_tx_time_coal_set(struct mvpp2_port *port)
 {
-	u32 freq = port->priv->tclk;
+	unsigned long freq = port->priv->tclk;
 	u32 val = mvpp2_usec_to_cycles(port->tx_time_coal, freq);
 
 	if (val > MVPP2_MAX_ISR_TX_THRESHOLD) {
-- 
cgit v1.2.3-58-ga151


From 6e5c85c003e47fd49b72ca052b9181644e04a520 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:18:58 +0000
Subject: net/sched: flower: refactor control flag definitions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Redefine the flower control flags as an enum, so they are
included in BTF info.

Make the kernel-side enum a more explicit superset of
TCA_FLOWER_KEY_FLAGS_*, new flags still need to be added to
both enums, but at least the bit position only has to be
defined once.

FLOW_DIS_ENCAPSULATION is never set for mask, so it can't be
exposed to userspace in an unsupported flags mask error message,
so it will be placed one bit position above the last uAPI flag.

Suggested-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-2-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/flow_dissector.h | 14 +++++++++++---
 include/uapi/linux/pkt_cls.h |  3 +++
 2 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h
index 3e47e123934d..c3fce070b912 100644
--- a/include/net/flow_dissector.h
+++ b/include/net/flow_dissector.h
@@ -7,6 +7,7 @@
 #include <linux/siphash.h>
 #include <linux/string.h>
 #include <uapi/linux/if_ether.h>
+#include <uapi/linux/pkt_cls.h>
 
 struct bpf_prog;
 struct net;
@@ -24,9 +25,16 @@ struct flow_dissector_key_control {
 	u32	flags;
 };
 
-#define FLOW_DIS_IS_FRAGMENT	BIT(0)
-#define FLOW_DIS_FIRST_FRAG	BIT(1)
-#define FLOW_DIS_ENCAPSULATION	BIT(2)
+/* The control flags are kept in sync with TCA_FLOWER_KEY_FLAGS_*, as those
+ * flags are exposed to userspace in some error paths, ie. unsupported flags.
+ */
+enum flow_dissector_ctrl_flags {
+	FLOW_DIS_IS_FRAGMENT		= TCA_FLOWER_KEY_FLAGS_IS_FRAGMENT,
+	FLOW_DIS_FIRST_FRAG		= TCA_FLOWER_KEY_FLAGS_FRAG_IS_FIRST,
+
+	/* These flags are internal to the kernel */
+	FLOW_DIS_ENCAPSULATION		= (TCA_FLOWER_KEY_FLAGS_MAX << 1),
+};
 
 enum flow_dissect_ret {
 	FLOW_DISSECT_RET_OUT_GOOD,
diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h
index b6d38f5fd7c0..12db276f0c11 100644
--- a/include/uapi/linux/pkt_cls.h
+++ b/include/uapi/linux/pkt_cls.h
@@ -677,8 +677,11 @@ enum {
 enum {
 	TCA_FLOWER_KEY_FLAGS_IS_FRAGMENT = (1 << 0),
 	TCA_FLOWER_KEY_FLAGS_FRAG_IS_FIRST = (1 << 1),
+	__TCA_FLOWER_KEY_FLAGS_MAX,
 };
 
+#define TCA_FLOWER_KEY_FLAGS_MAX (__TCA_FLOWER_KEY_FLAGS_MAX - 1)
+
 enum {
 	TCA_FLOWER_KEY_CFM_OPT_UNSPEC,
 	TCA_FLOWER_KEY_CFM_MD_LEVEL,
-- 
cgit v1.2.3-58-ga151


From 49ba9fc1c7736e935f7606920c5092b2bbb0d1e4 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:18:59 +0000
Subject: doc: netlink: specs: tc: describe flower control flags
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Describe the flower control flags, and use them
for key-flags and key-flags-mask.

The flag names have been taken from iproute2.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20240713021911.1631517-3-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/tc.yaml | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/Documentation/netlink/specs/tc.yaml b/Documentation/netlink/specs/tc.yaml
index 8c01e4e13195..fbbc928647fa 100644
--- a/Documentation/netlink/specs/tc.yaml
+++ b/Documentation/netlink/specs/tc.yaml
@@ -41,6 +41,12 @@ definitions:
       - in-hw
       - not-in-nw
       - verbose
+  -
+    name: tc-flower-key-ctrl-flags
+    type: flags
+    entries:
+      - frag
+      - firstfrag
   -
     name: tc-stats
     type: struct
@@ -2536,10 +2542,14 @@ attribute-sets:
         name: key-flags
         type: u32
         byte-order: big-endian
+        enum: tc-flower-key-ctrl-flags
+        enum-as-flags: true
       -
         name: key-flags-mask
         type: u32
         byte-order: big-endian
+        enum: tc-flower-key-ctrl-flags
+        enum-as-flags: true
       -
         name: key-icmpv4-code
         type: u8
-- 
cgit v1.2.3-58-ga151


From bfda5a63137bc83c344c4d995f404c8e701ff0fa Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:19:00 +0000
Subject: net/sched: flower: define new tunnel flags
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Define new TCA_FLOWER_KEY_FLAGS_* flags for use in struct
flow_dissector_key_control, covering the same flags as
currently exposed through TCA_FLOWER_KEY_ENC_FLAGS.

Put the new flags under FLOW_DIS_F_*. The idea is that we can
later, move the existing flags under FLOW_DIS_F_* as well.

The ynl flag names have been taken from the RFC iproute2 patch.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20240713021911.1631517-4-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/tc.yaml | 4 ++++
 include/net/flow_dissector.h        | 7 ++++++-
 include/uapi/linux/pkt_cls.h        | 4 ++++
 3 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/Documentation/netlink/specs/tc.yaml b/Documentation/netlink/specs/tc.yaml
index fbbc928647fa..aa574e3827ab 100644
--- a/Documentation/netlink/specs/tc.yaml
+++ b/Documentation/netlink/specs/tc.yaml
@@ -47,6 +47,10 @@ definitions:
     entries:
       - frag
       - firstfrag
+      - tuncsum
+      - tundf
+      - tunoam
+      - tuncrit
   -
     name: tc-stats
     type: struct
diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h
index c3fce070b912..460ea65b9e59 100644
--- a/include/net/flow_dissector.h
+++ b/include/net/flow_dissector.h
@@ -17,7 +17,8 @@ struct sk_buff;
  * struct flow_dissector_key_control:
  * @thoff:     Transport header offset
  * @addr_type: Type of key. One of FLOW_DISSECTOR_KEY_*
- * @flags:     Key flags. Any of FLOW_DIS_(IS_FRAGMENT|FIRST_FRAGENCAPSULATION)
+ * @flags:     Key flags.
+ *             Any of FLOW_DIS_(IS_FRAGMENT|FIRST_FRAG|ENCAPSULATION|F_*)
  */
 struct flow_dissector_key_control {
 	u16	thoff;
@@ -31,6 +32,10 @@ struct flow_dissector_key_control {
 enum flow_dissector_ctrl_flags {
 	FLOW_DIS_IS_FRAGMENT		= TCA_FLOWER_KEY_FLAGS_IS_FRAGMENT,
 	FLOW_DIS_FIRST_FRAG		= TCA_FLOWER_KEY_FLAGS_FRAG_IS_FIRST,
+	FLOW_DIS_F_TUNNEL_CSUM		= TCA_FLOWER_KEY_FLAGS_TUNNEL_CSUM,
+	FLOW_DIS_F_TUNNEL_DONT_FRAGMENT	= TCA_FLOWER_KEY_FLAGS_TUNNEL_DONT_FRAGMENT,
+	FLOW_DIS_F_TUNNEL_OAM		= TCA_FLOWER_KEY_FLAGS_TUNNEL_OAM,
+	FLOW_DIS_F_TUNNEL_CRIT_OPT	= TCA_FLOWER_KEY_FLAGS_TUNNEL_CRIT_OPT,
 
 	/* These flags are internal to the kernel */
 	FLOW_DIS_ENCAPSULATION		= (TCA_FLOWER_KEY_FLAGS_MAX << 1),
diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h
index 12db276f0c11..3dc4388e944c 100644
--- a/include/uapi/linux/pkt_cls.h
+++ b/include/uapi/linux/pkt_cls.h
@@ -677,6 +677,10 @@ enum {
 enum {
 	TCA_FLOWER_KEY_FLAGS_IS_FRAGMENT = (1 << 0),
 	TCA_FLOWER_KEY_FLAGS_FRAG_IS_FIRST = (1 << 1),
+	TCA_FLOWER_KEY_FLAGS_TUNNEL_CSUM = (1 << 2),
+	TCA_FLOWER_KEY_FLAGS_TUNNEL_DONT_FRAGMENT = (1 << 3),
+	TCA_FLOWER_KEY_FLAGS_TUNNEL_OAM = (1 << 4),
+	TCA_FLOWER_KEY_FLAGS_TUNNEL_CRIT_OPT = (1 << 5),
 	__TCA_FLOWER_KEY_FLAGS_MAX,
 };
 
-- 
cgit v1.2.3-58-ga151


From fcb4bb07a9276bfb63c82ce5e8c4f8ff23646635 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:19:01 +0000
Subject: net/sched: cls_flower: prepare fl_{set,dump}_key_flags() for
 ENC_FLAGS
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare fl_set_key_flags/fl_dump_key_flags() for use with
TCA_FLOWER_KEY_ENC_FLAGS{,_MASK}.

This patch adds an encap argument, similar to fl_set_key_ip/
fl_dump_key_ip(), and determine the flower keys based on the
encap argument, and use them in the rest of the two functions.

Since these functions are so far, only called with encap set false,
then there is no functional change.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-5-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/cls_flower.c | 40 ++++++++++++++++++++++++++++++----------
 1 file changed, 30 insertions(+), 10 deletions(-)

diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index eef570c577ac..6a5cecfd9561 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -1166,19 +1166,28 @@ static void fl_set_key_flag(u32 flower_key, u32 flower_mask,
 	}
 }
 
-static int fl_set_key_flags(struct nlattr **tb, u32 *flags_key,
+static int fl_set_key_flags(struct nlattr **tb, bool encap, u32 *flags_key,
 			    u32 *flags_mask, struct netlink_ext_ack *extack)
 {
+	int fl_key, fl_mask;
 	u32 key, mask;
 
+	if (encap) {
+		fl_key = TCA_FLOWER_KEY_ENC_FLAGS;
+		fl_mask = TCA_FLOWER_KEY_ENC_FLAGS_MASK;
+	} else {
+		fl_key = TCA_FLOWER_KEY_FLAGS;
+		fl_mask = TCA_FLOWER_KEY_FLAGS_MASK;
+	}
+
 	/* mask is mandatory for flags */
-	if (!tb[TCA_FLOWER_KEY_FLAGS_MASK]) {
+	if (NL_REQ_ATTR_CHECK(extack, NULL, tb, fl_mask)) {
 		NL_SET_ERR_MSG(extack, "Missing flags mask");
 		return -EINVAL;
 	}
 
-	key = be32_to_cpu(nla_get_be32(tb[TCA_FLOWER_KEY_FLAGS]));
-	mask = be32_to_cpu(nla_get_be32(tb[TCA_FLOWER_KEY_FLAGS_MASK]));
+	key = be32_to_cpu(nla_get_be32(tb[fl_key]));
+	mask = be32_to_cpu(nla_get_be32(tb[fl_mask]));
 
 	*flags_key  = 0;
 	*flags_mask = 0;
@@ -2086,7 +2095,7 @@ static int fl_set_key(struct net *net, struct nlattr **tb,
 		return ret;
 
 	if (tb[TCA_FLOWER_KEY_FLAGS]) {
-		ret = fl_set_key_flags(tb, &key->control.flags,
+		ret = fl_set_key_flags(tb, false, &key->control.flags,
 				       &mask->control.flags, extack);
 		if (ret)
 			return ret;
@@ -3084,12 +3093,22 @@ static void fl_get_key_flag(u32 dissector_key, u32 dissector_mask,
 	}
 }
 
-static int fl_dump_key_flags(struct sk_buff *skb, u32 flags_key, u32 flags_mask)
+static int fl_dump_key_flags(struct sk_buff *skb, bool encap,
+			     u32 flags_key, u32 flags_mask)
 {
-	u32 key, mask;
+	int fl_key, fl_mask;
 	__be32 _key, _mask;
+	u32 key, mask;
 	int err;
 
+	if (encap) {
+		fl_key = TCA_FLOWER_KEY_ENC_FLAGS;
+		fl_mask = TCA_FLOWER_KEY_ENC_FLAGS_MASK;
+	} else {
+		fl_key = TCA_FLOWER_KEY_FLAGS;
+		fl_mask = TCA_FLOWER_KEY_FLAGS_MASK;
+	}
+
 	if (!memchr_inv(&flags_mask, 0, sizeof(flags_mask)))
 		return 0;
 
@@ -3105,11 +3124,11 @@ static int fl_dump_key_flags(struct sk_buff *skb, u32 flags_key, u32 flags_mask)
 	_key = cpu_to_be32(key);
 	_mask = cpu_to_be32(mask);
 
-	err = nla_put(skb, TCA_FLOWER_KEY_FLAGS, 4, &_key);
+	err = nla_put(skb, fl_key, 4, &_key);
 	if (err)
 		return err;
 
-	return nla_put(skb, TCA_FLOWER_KEY_FLAGS_MASK, 4, &_mask);
+	return nla_put(skb, fl_mask, 4, &_mask);
 }
 
 static int fl_dump_key_geneve_opt(struct sk_buff *skb,
@@ -3632,7 +3651,8 @@ static int fl_dump_key(struct sk_buff *skb, struct net *net,
 	if (fl_dump_key_ct(skb, &key->ct, &mask->ct))
 		goto nla_put_failure;
 
-	if (fl_dump_key_flags(skb, key->control.flags, mask->control.flags))
+	if (fl_dump_key_flags(skb, false, key->control.flags,
+			      mask->control.flags))
 		goto nla_put_failure;
 
 	if (fl_dump_key_val(skb, &key->hash.hash, TCA_FLOWER_KEY_HASH,
-- 
cgit v1.2.3-58-ga151


From 0e83a7875d691f5b76c87d9223b22e5a0d34511e Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:19:02 +0000
Subject: net/sched: cls_flower: add policy for TCA_FLOWER_KEY_FLAGS
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This policy guards fl_set_key_flags() from seeing flags
not used in the context of TCA_FLOWER_KEY_FLAGS.

In order For the policy check to be performed with the
correct endianness, then we also needs to change the
attribute type to NLA_BE32 (Thanks Davide).

TCA_FLOWER_KEY_FLAGS{,_MASK} already has a be32 comment
in include/uapi/linux/pkt_cls.h.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-6-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/cls_flower.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index 6a5cecfd9561..fc9a9a0b4897 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -41,6 +41,10 @@
 #define TCA_FLOWER_KEY_CT_FLAGS_MASK \
 		(TCA_FLOWER_KEY_CT_FLAGS_MAX - 1)
 
+#define TCA_FLOWER_KEY_FLAGS_POLICY_MASK \
+		(TCA_FLOWER_KEY_FLAGS_IS_FRAGMENT | \
+		TCA_FLOWER_KEY_FLAGS_FRAG_IS_FIRST)
+
 #define TUNNEL_FLAGS_PRESENT (\
 	_BITUL(IP_TUNNEL_CSUM_BIT) |		\
 	_BITUL(IP_TUNNEL_DONT_FRAGMENT_BIT) |	\
@@ -676,8 +680,10 @@ static const struct nla_policy fl_policy[TCA_FLOWER_MAX + 1] = {
 	[TCA_FLOWER_KEY_ENC_UDP_SRC_PORT_MASK]	= { .type = NLA_U16 },
 	[TCA_FLOWER_KEY_ENC_UDP_DST_PORT]	= { .type = NLA_U16 },
 	[TCA_FLOWER_KEY_ENC_UDP_DST_PORT_MASK]	= { .type = NLA_U16 },
-	[TCA_FLOWER_KEY_FLAGS]		= { .type = NLA_U32 },
-	[TCA_FLOWER_KEY_FLAGS_MASK]	= { .type = NLA_U32 },
+	[TCA_FLOWER_KEY_FLAGS]		= NLA_POLICY_MASK(NLA_BE32,
+							  TCA_FLOWER_KEY_FLAGS_POLICY_MASK),
+	[TCA_FLOWER_KEY_FLAGS_MASK]	= NLA_POLICY_MASK(NLA_BE32,
+							  TCA_FLOWER_KEY_FLAGS_POLICY_MASK),
 	[TCA_FLOWER_KEY_ICMPV4_TYPE]	= { .type = NLA_U8 },
 	[TCA_FLOWER_KEY_ICMPV4_TYPE_MASK] = { .type = NLA_U8 },
 	[TCA_FLOWER_KEY_ICMPV4_CODE]	= { .type = NLA_U8 },
-- 
cgit v1.2.3-58-ga151


From 4d0aed380f9ddf24dfb1d06a05096b778442c403 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:19:03 +0000
Subject: flow_dissector: prepare for encapsulated control flags
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rename skb_flow_dissect_set_enc_addr_type() to
skb_flow_dissect_set_enc_control(), and make it set both
addr_type and flags in FLOW_DISSECTOR_KEY_ENC_CONTROL.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-7-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/flow_dissector.c | 21 ++++++++++++---------
 1 file changed, 12 insertions(+), 9 deletions(-)

diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index e64a26379807..1614c6708ea7 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -299,9 +299,10 @@ void skb_flow_dissect_meta(const struct sk_buff *skb,
 EXPORT_SYMBOL(skb_flow_dissect_meta);
 
 static void
-skb_flow_dissect_set_enc_addr_type(enum flow_dissector_key_id type,
-				   struct flow_dissector *flow_dissector,
-				   void *target_container)
+skb_flow_dissect_set_enc_control(enum flow_dissector_key_id type,
+				 u32 ctrl_flags,
+				 struct flow_dissector *flow_dissector,
+				 void *target_container)
 {
 	struct flow_dissector_key_control *ctrl;
 
@@ -312,6 +313,7 @@ skb_flow_dissect_set_enc_addr_type(enum flow_dissector_key_id type,
 					 FLOW_DISSECTOR_KEY_ENC_CONTROL,
 					 target_container);
 	ctrl->addr_type = type;
+	ctrl->flags = ctrl_flags;
 }
 
 void
@@ -367,6 +369,7 @@ skb_flow_dissect_tunnel_info(const struct sk_buff *skb,
 {
 	struct ip_tunnel_info *info;
 	struct ip_tunnel_key *key;
+	u32 ctrl_flags = 0;
 
 	/* A quick check to see if there might be something to do. */
 	if (!dissector_uses_key(flow_dissector,
@@ -395,9 +398,9 @@ skb_flow_dissect_tunnel_info(const struct sk_buff *skb,
 
 	switch (ip_tunnel_info_af(info)) {
 	case AF_INET:
-		skb_flow_dissect_set_enc_addr_type(FLOW_DISSECTOR_KEY_IPV4_ADDRS,
-						   flow_dissector,
-						   target_container);
+		skb_flow_dissect_set_enc_control(FLOW_DISSECTOR_KEY_IPV4_ADDRS,
+						 ctrl_flags, flow_dissector,
+						 target_container);
 		if (dissector_uses_key(flow_dissector,
 				       FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS)) {
 			struct flow_dissector_key_ipv4_addrs *ipv4;
@@ -410,9 +413,9 @@ skb_flow_dissect_tunnel_info(const struct sk_buff *skb,
 		}
 		break;
 	case AF_INET6:
-		skb_flow_dissect_set_enc_addr_type(FLOW_DISSECTOR_KEY_IPV6_ADDRS,
-						   flow_dissector,
-						   target_container);
+		skb_flow_dissect_set_enc_control(FLOW_DISSECTOR_KEY_IPV6_ADDRS,
+						 ctrl_flags, flow_dissector,
+						 target_container);
 		if (dissector_uses_key(flow_dissector,
 				       FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS)) {
 			struct flow_dissector_key_ipv6_addrs *ipv6;
-- 
cgit v1.2.3-58-ga151


From 03afeb613bfe6b0c28e8b843959f716a3d2c42df Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:19:04 +0000
Subject: flow_dissector: set encapsulated control flags from tun_flags
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Set the new FLOW_DIS_F_TUNNEL_* encapsulated control flags, based
on if their counter-part is set in tun_flags.

These flags are not userspace visible yet, as the code to dump
encapsulated control flags will first be added, and later activated
in the following patches.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-8-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/flow_dissector.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 1614c6708ea7..a0263a4c5489 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -396,6 +396,15 @@ skb_flow_dissect_tunnel_info(const struct sk_buff *skb,
 
 	key = &info->key;
 
+	if (test_bit(IP_TUNNEL_CSUM_BIT, key->tun_flags))
+		ctrl_flags |= FLOW_DIS_F_TUNNEL_CSUM;
+	if (test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, key->tun_flags))
+		ctrl_flags |= FLOW_DIS_F_TUNNEL_DONT_FRAGMENT;
+	if (test_bit(IP_TUNNEL_OAM_BIT, key->tun_flags))
+		ctrl_flags |= FLOW_DIS_F_TUNNEL_OAM;
+	if (test_bit(IP_TUNNEL_CRIT_OPT_BIT, key->tun_flags))
+		ctrl_flags |= FLOW_DIS_F_TUNNEL_CRIT_OPT;
+
 	switch (ip_tunnel_info_af(info)) {
 	case AF_INET:
 		skb_flow_dissect_set_enc_control(FLOW_DISSECTOR_KEY_IPV4_ADDRS,
-- 
cgit v1.2.3-58-ga151


From 988f8723d398cf122c112342d12d67ba31cadf3b Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:19:05 +0000
Subject: net/sched: cls_flower: add tunnel flags to fl_{set,dump}_key_flags()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prepare to set and dump the tunnel flags.

This code won't see any of these flags yet, as these flags
aren't allowed by the NLA_POLICY_MASK, and the functions
doesn't get called with encap set to true yet.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-9-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/cls_flower.c | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index fc9a9a0b4897..2a440f11fe1f 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -1204,6 +1204,21 @@ static int fl_set_key_flags(struct nlattr **tb, bool encap, u32 *flags_key,
 			TCA_FLOWER_KEY_FLAGS_FRAG_IS_FIRST,
 			FLOW_DIS_FIRST_FRAG);
 
+	fl_set_key_flag(key, mask, flags_key, flags_mask,
+			TCA_FLOWER_KEY_FLAGS_TUNNEL_CSUM,
+			FLOW_DIS_F_TUNNEL_CSUM);
+
+	fl_set_key_flag(key, mask, flags_key, flags_mask,
+			TCA_FLOWER_KEY_FLAGS_TUNNEL_DONT_FRAGMENT,
+			FLOW_DIS_F_TUNNEL_DONT_FRAGMENT);
+
+	fl_set_key_flag(key, mask, flags_key, flags_mask,
+			TCA_FLOWER_KEY_FLAGS_TUNNEL_OAM, FLOW_DIS_F_TUNNEL_OAM);
+
+	fl_set_key_flag(key, mask, flags_key, flags_mask,
+			TCA_FLOWER_KEY_FLAGS_TUNNEL_CRIT_OPT,
+			FLOW_DIS_F_TUNNEL_CRIT_OPT);
+
 	return 0;
 }
 
@@ -3127,6 +3142,21 @@ static int fl_dump_key_flags(struct sk_buff *skb, bool encap,
 			TCA_FLOWER_KEY_FLAGS_FRAG_IS_FIRST,
 			FLOW_DIS_FIRST_FRAG);
 
+	fl_get_key_flag(flags_key, flags_mask, &key, &mask,
+			TCA_FLOWER_KEY_FLAGS_TUNNEL_CSUM,
+			FLOW_DIS_F_TUNNEL_CSUM);
+
+	fl_get_key_flag(flags_key, flags_mask, &key, &mask,
+			TCA_FLOWER_KEY_FLAGS_TUNNEL_DONT_FRAGMENT,
+			FLOW_DIS_F_TUNNEL_DONT_FRAGMENT);
+
+	fl_get_key_flag(flags_key, flags_mask, &key, &mask,
+			TCA_FLOWER_KEY_FLAGS_TUNNEL_OAM, FLOW_DIS_F_TUNNEL_OAM);
+
+	fl_get_key_flag(flags_key, flags_mask, &key, &mask,
+			TCA_FLOWER_KEY_FLAGS_TUNNEL_CRIT_OPT,
+			FLOW_DIS_F_TUNNEL_CRIT_OPT);
+
 	_key = cpu_to_be32(key);
 	_mask = cpu_to_be32(mask);
 
-- 
cgit v1.2.3-58-ga151


From 11036bd7a0b3b05c5e1f43d107ddb02abf83adb9 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:19:06 +0000
Subject: net/sched: cls_flower: rework TCA_FLOWER_KEY_ENC_FLAGS usage
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch changes how TCA_FLOWER_KEY_ENC_FLAGS is used, so that
it is used with TCA_FLOWER_KEY_FLAGS_* flags, in the same way as
TCA_FLOWER_KEY_FLAGS is currently used.

Where TCA_FLOWER_KEY_FLAGS uses {key,mask}->control.flags, then
TCA_FLOWER_KEY_ENC_FLAGS now uses {key,mask}->enc_control.flags,
therefore {key,mask}->enc_flags is now unused.

As the generic fl_set_key_flags/fl_dump_key_flags() is used with
encap set to true, then fl_{set,dump}_key_enc_flags() is removed.

This breaks unreleased userspace API (net-next since 2024-06-04).

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-10-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/uapi/linux/pkt_cls.h |  4 ++--
 net/sched/cls_flower.c       | 56 ++++++++++----------------------------------
 2 files changed, 15 insertions(+), 45 deletions(-)

diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h
index 3dc4388e944c..d36d9cdf0c00 100644
--- a/include/uapi/linux/pkt_cls.h
+++ b/include/uapi/linux/pkt_cls.h
@@ -554,8 +554,8 @@ enum {
 	TCA_FLOWER_KEY_SPI,		/* be32 */
 	TCA_FLOWER_KEY_SPI_MASK,	/* be32 */
 
-	TCA_FLOWER_KEY_ENC_FLAGS,	/* u32 */
-	TCA_FLOWER_KEY_ENC_FLAGS_MASK,	/* u32 */
+	TCA_FLOWER_KEY_ENC_FLAGS,	/* be32 */
+	TCA_FLOWER_KEY_ENC_FLAGS_MASK,	/* be32 */
 
 	__TCA_FLOWER_MAX,
 };
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index 2a440f11fe1f..e2239ab01355 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -45,11 +45,11 @@
 		(TCA_FLOWER_KEY_FLAGS_IS_FRAGMENT | \
 		TCA_FLOWER_KEY_FLAGS_FRAG_IS_FIRST)
 
-#define TUNNEL_FLAGS_PRESENT (\
-	_BITUL(IP_TUNNEL_CSUM_BIT) |		\
-	_BITUL(IP_TUNNEL_DONT_FRAGMENT_BIT) |	\
-	_BITUL(IP_TUNNEL_OAM_BIT) |		\
-	_BITUL(IP_TUNNEL_CRIT_OPT_BIT))
+#define TCA_FLOWER_KEY_ENC_FLAGS_POLICY_MASK \
+		(TCA_FLOWER_KEY_FLAGS_TUNNEL_CSUM | \
+		TCA_FLOWER_KEY_FLAGS_TUNNEL_DONT_FRAGMENT | \
+		TCA_FLOWER_KEY_FLAGS_TUNNEL_OAM | \
+		TCA_FLOWER_KEY_FLAGS_TUNNEL_CRIT_OPT)
 
 struct fl_flow_key {
 	struct flow_dissector_key_meta meta;
@@ -745,10 +745,10 @@ static const struct nla_policy fl_policy[TCA_FLOWER_MAX + 1] = {
 	[TCA_FLOWER_KEY_SPI_MASK]	= { .type = NLA_U32 },
 	[TCA_FLOWER_L2_MISS]		= NLA_POLICY_MAX(NLA_U8, 1),
 	[TCA_FLOWER_KEY_CFM]		= { .type = NLA_NESTED },
-	[TCA_FLOWER_KEY_ENC_FLAGS]	= NLA_POLICY_MASK(NLA_U32,
-							  TUNNEL_FLAGS_PRESENT),
-	[TCA_FLOWER_KEY_ENC_FLAGS_MASK]	= NLA_POLICY_MASK(NLA_U32,
-							  TUNNEL_FLAGS_PRESENT),
+	[TCA_FLOWER_KEY_ENC_FLAGS]	= NLA_POLICY_MASK(NLA_BE32,
+							  TCA_FLOWER_KEY_ENC_FLAGS_POLICY_MASK),
+	[TCA_FLOWER_KEY_ENC_FLAGS_MASK]	= NLA_POLICY_MASK(NLA_BE32,
+							  TCA_FLOWER_KEY_ENC_FLAGS_POLICY_MASK),
 };
 
 static const struct nla_policy
@@ -1866,21 +1866,6 @@ static int fl_set_key_cfm(struct nlattr **tb,
 	return 0;
 }
 
-static int fl_set_key_enc_flags(struct nlattr **tb, u32 *flags_key,
-				u32 *flags_mask, struct netlink_ext_ack *extack)
-{
-	/* mask is mandatory for flags */
-	if (NL_REQ_ATTR_CHECK(extack, NULL, tb, TCA_FLOWER_KEY_ENC_FLAGS_MASK)) {
-		NL_SET_ERR_MSG(extack, "missing enc_flags mask");
-		return -EINVAL;
-	}
-
-	*flags_key = nla_get_u32(tb[TCA_FLOWER_KEY_ENC_FLAGS]);
-	*flags_mask = nla_get_u32(tb[TCA_FLOWER_KEY_ENC_FLAGS_MASK]);
-
-	return 0;
-}
-
 static int fl_set_key(struct net *net, struct nlattr **tb,
 		      struct fl_flow_key *key, struct fl_flow_key *mask,
 		      struct netlink_ext_ack *extack)
@@ -2123,8 +2108,8 @@ static int fl_set_key(struct net *net, struct nlattr **tb,
 	}
 
 	if (tb[TCA_FLOWER_KEY_ENC_FLAGS])
-		ret = fl_set_key_enc_flags(tb, &key->enc_flags.flags,
-					   &mask->enc_flags.flags, extack);
+		ret = fl_set_key_flags(tb, true, &key->enc_control.flags,
+				       &mask->enc_control.flags, extack);
 
 	return ret;
 }
@@ -3381,22 +3366,6 @@ err_cfm_opts:
 	return err;
 }
 
-static int fl_dump_key_enc_flags(struct sk_buff *skb,
-				 struct flow_dissector_key_enc_flags *key,
-				 struct flow_dissector_key_enc_flags *mask)
-{
-	if (!memchr_inv(mask, 0, sizeof(*mask)))
-		return 0;
-
-	if (nla_put_u32(skb, TCA_FLOWER_KEY_ENC_FLAGS, key->flags))
-		return -EMSGSIZE;
-
-	if (nla_put_u32(skb, TCA_FLOWER_KEY_ENC_FLAGS_MASK, mask->flags))
-		return -EMSGSIZE;
-
-	return 0;
-}
-
 static int fl_dump_key_options(struct sk_buff *skb, int enc_opt_type,
 			       struct flow_dissector_key_enc_opts *enc_opts)
 {
@@ -3699,7 +3668,8 @@ static int fl_dump_key(struct sk_buff *skb, struct net *net,
 	if (fl_dump_key_cfm(skb, &key->cfm, &mask->cfm))
 		goto nla_put_failure;
 
-	if (fl_dump_key_enc_flags(skb, &key->enc_flags, &mask->enc_flags))
+	if (fl_dump_key_flags(skb, true, key->enc_control.flags,
+			      mask->enc_control.flags))
 		goto nla_put_failure;
 
 	return 0;
-- 
cgit v1.2.3-58-ga151


From 880a51a8ab8cbdeb9c2d35a5147d30b7fc0893e4 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:19:07 +0000
Subject: doc: netlink: specs: tc: flower: add enc-flags
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Describe key-enc-flags and key-enc-flags-mask.

These are defined similarly to key-flags and key-flags-mask.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20240713021911.1631517-11-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/netlink/specs/tc.yaml | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/Documentation/netlink/specs/tc.yaml b/Documentation/netlink/specs/tc.yaml
index aa574e3827ab..b02d59a0349c 100644
--- a/Documentation/netlink/specs/tc.yaml
+++ b/Documentation/netlink/specs/tc.yaml
@@ -2763,6 +2763,18 @@ attribute-sets:
         name: key-spi-mask
         type: u32
         byte-order: big-endian
+      -
+        name: key-enc-flags
+        type: u32
+        byte-order: big-endian
+        enum: tc-flower-key-ctrl-flags
+        enum-as-flags: true
+      -
+        name: key-enc-flags-mask
+        type: u32
+        byte-order: big-endian
+        enum: tc-flower-key-ctrl-flags
+        enum-as-flags: true
   -
     name: tc-flower-key-enc-opts-attrs
     attributes:
-- 
cgit v1.2.3-58-ga151


From db5271d50ec155abf287a27fa84e2e33a81dbd55 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:19:08 +0000
Subject: flow_dissector: cleanup FLOW_DISSECTOR_KEY_ENC_FLAGS
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Now that TCA_FLOWER_KEY_ENC_FLAGS is unused, as it's
former data is stored behind TCA_FLOWER_KEY_ENC_CONTROL,
then remove the last bits of FLOW_DISSECTOR_KEY_ENC_FLAGS.

FLOW_DISSECTOR_KEY_ENC_FLAGS is unreleased, and have been
in net-next since 2024-06-04.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-12-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/net/flow_dissector.h |  9 ---------
 include/net/ip_tunnels.h     | 12 ------------
 net/core/flow_dissector.c    | 16 +---------------
 net/sched/cls_flower.c       |  3 ---
 4 files changed, 1 insertion(+), 39 deletions(-)

diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h
index 460ea65b9e59..ced79dc8e856 100644
--- a/include/net/flow_dissector.h
+++ b/include/net/flow_dissector.h
@@ -342,14 +342,6 @@ struct flow_dissector_key_cfm {
 #define FLOW_DIS_CFM_MDL_MASK GENMASK(7, 5)
 #define FLOW_DIS_CFM_MDL_MAX 7
 
-/**
- * struct flow_dissector_key_enc_flags: tunnel metadata control flags
- * @flags: tunnel control flags
- */
-struct flow_dissector_key_enc_flags {
-	u32 flags;
-};
-
 enum flow_dissector_key_id {
 	FLOW_DISSECTOR_KEY_CONTROL, /* struct flow_dissector_key_control */
 	FLOW_DISSECTOR_KEY_BASIC, /* struct flow_dissector_key_basic */
@@ -384,7 +376,6 @@ enum flow_dissector_key_id {
 	FLOW_DISSECTOR_KEY_L2TPV3, /* struct flow_dissector_key_l2tpv3 */
 	FLOW_DISSECTOR_KEY_CFM, /* struct flow_dissector_key_cfm */
 	FLOW_DISSECTOR_KEY_IPSEC, /* struct flow_dissector_key_ipsec */
-	FLOW_DISSECTOR_KEY_ENC_FLAGS, /* struct flow_dissector_key_enc_flags */
 
 	FLOW_DISSECTOR_KEY_MAX,
 };
diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index 3877315cf8b8..1db2417b8ff5 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -247,18 +247,6 @@ static inline bool ip_tunnel_is_options_present(const unsigned long *flags)
 	return ip_tunnel_flags_intersect(flags, present);
 }
 
-static inline void ip_tunnel_set_encflags_present(unsigned long *flags)
-{
-	IP_TUNNEL_DECLARE_FLAGS(present) = { };
-
-	__set_bit(IP_TUNNEL_CSUM_BIT, present);
-	__set_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, present);
-	__set_bit(IP_TUNNEL_OAM_BIT, present);
-	__set_bit(IP_TUNNEL_CRIT_OPT_BIT, present);
-
-	ip_tunnel_flags_or(flags, flags, present);
-}
-
 static inline bool ip_tunnel_flags_is_be16_compat(const unsigned long *flags)
 {
 	IP_TUNNEL_DECLARE_FLAGS(supp) = { };
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index a0263a4c5489..1a9ca129fddd 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -385,9 +385,7 @@ skb_flow_dissect_tunnel_info(const struct sk_buff *skb,
 	    !dissector_uses_key(flow_dissector,
 				FLOW_DISSECTOR_KEY_ENC_IP) &&
 	    !dissector_uses_key(flow_dissector,
-				FLOW_DISSECTOR_KEY_ENC_OPTS) &&
-	    !dissector_uses_key(flow_dissector,
-				FLOW_DISSECTOR_KEY_ENC_FLAGS))
+				FLOW_DISSECTOR_KEY_ENC_OPTS))
 		return;
 
 	info = skb_tunnel_info(skb);
@@ -489,18 +487,6 @@ skb_flow_dissect_tunnel_info(const struct sk_buff *skb,
 				    IP_TUNNEL_GENEVE_OPT_BIT);
 		enc_opt->dst_opt_type = val < __IP_TUNNEL_FLAG_NUM ? val : 0;
 	}
-
-	if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_ENC_FLAGS)) {
-		struct flow_dissector_key_enc_flags *enc_flags;
-		IP_TUNNEL_DECLARE_FLAGS(flags) = {};
-
-		enc_flags = skb_flow_dissector_target(flow_dissector,
-						      FLOW_DISSECTOR_KEY_ENC_FLAGS,
-						      target_container);
-		ip_tunnel_set_encflags_present(flags);
-		ip_tunnel_flags_and(flags, flags, info->key.tun_flags);
-		enc_flags->flags = bitmap_read(flags, IP_TUNNEL_CSUM_BIT, 32);
-	}
 }
 EXPORT_SYMBOL(skb_flow_dissect_tunnel_info);
 
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index e2239ab01355..897d6b683cc6 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -85,7 +85,6 @@ struct fl_flow_key {
 	struct flow_dissector_key_l2tpv3 l2tpv3;
 	struct flow_dissector_key_ipsec ipsec;
 	struct flow_dissector_key_cfm cfm;
-	struct flow_dissector_key_enc_flags enc_flags;
 } __aligned(BITS_PER_LONG / 8); /* Ensure that we can do comparisons as longs. */
 
 struct fl_flow_mask_range {
@@ -2223,8 +2222,6 @@ static void fl_init_dissector(struct flow_dissector *dissector,
 			     FLOW_DISSECTOR_KEY_IPSEC, ipsec);
 	FL_KEY_SET_IF_MASKED(mask, keys, cnt,
 			     FLOW_DISSECTOR_KEY_CFM, cfm);
-	FL_KEY_SET_IF_MASKED(mask, keys, cnt,
-			     FLOW_DISSECTOR_KEY_ENC_FLAGS, enc_flags);
 
 	skb_flow_dissector_init(dissector, keys, cnt);
 }
-- 
cgit v1.2.3-58-ga151


From 706bf4f44c6d2ae2fdeefeb816b2c35a173ecfa4 Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:19:09 +0000
Subject: flow_dissector: set encapsulation control flags for non-IP
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Make sure to set encapsulated control flags also for non-IP
packets, such that it's possible to allow matching on e.g.
TUNNEL_OAM on a geneve packet carrying a non-IP packet.

Suggested-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-13-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/core/flow_dissector.c | 4 ++++
 net/sched/cls_flower.c    | 3 ++-
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 1a9ca129fddd..ada1e39b557e 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -434,6 +434,10 @@ skb_flow_dissect_tunnel_info(const struct sk_buff *skb,
 			ipv6->dst = key->u.ipv6.dst;
 		}
 		break;
+	default:
+		skb_flow_dissect_set_enc_control(0, ctrl_flags, flow_dissector,
+						 target_container);
+		break;
 	}
 
 	if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_ENC_KEYID)) {
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index 897d6b683cc6..38b2df387c1e 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -2199,7 +2199,8 @@ static void fl_init_dissector(struct flow_dissector *dissector,
 	FL_KEY_SET_IF_MASKED(mask, keys, cnt,
 			     FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS, enc_ipv6);
 	if (FL_KEY_IS_MASKED(mask, enc_ipv4) ||
-	    FL_KEY_IS_MASKED(mask, enc_ipv6))
+	    FL_KEY_IS_MASKED(mask, enc_ipv6) ||
+	    FL_KEY_IS_MASKED(mask, enc_control))
 		FL_KEY_SET(keys, cnt, FLOW_DISSECTOR_KEY_ENC_CONTROL,
 			   enc_control);
 	FL_KEY_SET_IF_MASKED(mask, keys, cnt,
-- 
cgit v1.2.3-58-ga151


From 536b97acddd702edacbe5f89c95dcd5a145b490f Mon Sep 17 00:00:00 2001
From: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Date: Sat, 13 Jul 2024 02:19:10 +0000
Subject: net/sched: cls_flower: propagate tca[TCA_OPTIONS] to
 NL_REQ_ATTR_CHECK
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

NL_REQ_ATTR_CHECK() is used in fl_set_key_flags() to set
extended attributes about the origin of an error, this
patch propagates tca[TCA_OPTIONS] through.

Before this patch:

$ sudo ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/tc.yaml \
	 --do newtfilter --json '{
		"chain": 0, "family": 0, "handle": 4, "ifindex": 22,
		"info": 262152,
		"kind": "flower",
		"options": {
			"flags": 0, "key-enc-flags": 8,
			"key-eth-type": 2048 },
		"parent": 4294967283 }'
Netlink error: Invalid argument
nl_len = 68 (52) nl_flags = 0x300 nl_type = 2
        error: -22
        extack: {'msg': 'Missing flags mask',
                 'miss-type': 111}

After this patch:

[same cmd]
Netlink error: Invalid argument
nl_len = 76 (60) nl_flags = 0x300 nl_type = 2
        error: -22
        extack: {'msg': 'Missing flags mask',
                 'miss-type': 111, 'miss-nest': 56}

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Link: https://patch.msgid.link/20240713021911.1631517-14-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/sched/cls_flower.c | 34 ++++++++++++++++++++--------------
 1 file changed, 20 insertions(+), 14 deletions(-)

diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index 38b2df387c1e..e280c27cb9f9 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -1171,8 +1171,9 @@ static void fl_set_key_flag(u32 flower_key, u32 flower_mask,
 	}
 }
 
-static int fl_set_key_flags(struct nlattr **tb, bool encap, u32 *flags_key,
-			    u32 *flags_mask, struct netlink_ext_ack *extack)
+static int fl_set_key_flags(struct nlattr *tca_opts, struct nlattr **tb,
+			    bool encap, u32 *flags_key, u32 *flags_mask,
+			    struct netlink_ext_ack *extack)
 {
 	int fl_key, fl_mask;
 	u32 key, mask;
@@ -1186,7 +1187,7 @@ static int fl_set_key_flags(struct nlattr **tb, bool encap, u32 *flags_key,
 	}
 
 	/* mask is mandatory for flags */
-	if (NL_REQ_ATTR_CHECK(extack, NULL, tb, fl_mask)) {
+	if (NL_REQ_ATTR_CHECK(extack, tca_opts, tb, fl_mask)) {
 		NL_SET_ERR_MSG(extack, "Missing flags mask");
 		return -EINVAL;
 	}
@@ -1865,9 +1866,9 @@ static int fl_set_key_cfm(struct nlattr **tb,
 	return 0;
 }
 
-static int fl_set_key(struct net *net, struct nlattr **tb,
-		      struct fl_flow_key *key, struct fl_flow_key *mask,
-		      struct netlink_ext_ack *extack)
+static int fl_set_key(struct net *net, struct nlattr *tca_opts,
+		      struct nlattr **tb, struct fl_flow_key *key,
+		      struct fl_flow_key *mask, struct netlink_ext_ack *extack)
 {
 	__be16 ethertype;
 	int ret = 0;
@@ -2100,14 +2101,16 @@ static int fl_set_key(struct net *net, struct nlattr **tb,
 		return ret;
 
 	if (tb[TCA_FLOWER_KEY_FLAGS]) {
-		ret = fl_set_key_flags(tb, false, &key->control.flags,
+		ret = fl_set_key_flags(tca_opts, tb, false,
+				       &key->control.flags,
 				       &mask->control.flags, extack);
 		if (ret)
 			return ret;
 	}
 
 	if (tb[TCA_FLOWER_KEY_ENC_FLAGS])
-		ret = fl_set_key_flags(tb, true, &key->enc_control.flags,
+		ret = fl_set_key_flags(tca_opts, tb, true,
+				       &key->enc_control.flags,
 				       &mask->enc_control.flags, extack);
 
 	return ret;
@@ -2358,6 +2361,7 @@ static int fl_change(struct net *net, struct sk_buff *in_skb,
 {
 	struct cls_fl_head *head = fl_head_dereference(tp);
 	bool rtnl_held = !(flags & TCA_ACT_FLAGS_NO_RTNL);
+	struct nlattr *tca_opts = tca[TCA_OPTIONS];
 	struct cls_fl_filter *fold = *arg;
 	bool bound_to_filter = false;
 	struct cls_fl_filter *fnew;
@@ -2366,7 +2370,7 @@ static int fl_change(struct net *net, struct sk_buff *in_skb,
 	bool in_ht;
 	int err;
 
-	if (!tca[TCA_OPTIONS]) {
+	if (!tca_opts) {
 		err = -EINVAL;
 		goto errout_fold;
 	}
@@ -2384,7 +2388,7 @@ static int fl_change(struct net *net, struct sk_buff *in_skb,
 	}
 
 	err = nla_parse_nested_deprecated(tb, TCA_FLOWER_MAX,
-					  tca[TCA_OPTIONS], fl_policy, NULL);
+					  tca_opts, fl_policy, NULL);
 	if (err < 0)
 		goto errout_tb;
 
@@ -2460,7 +2464,7 @@ static int fl_change(struct net *net, struct sk_buff *in_skb,
 		bound_to_filter = true;
 	}
 
-	err = fl_set_key(net, tb, &fnew->key, &mask->key, extack);
+	err = fl_set_key(net, tca_opts, tb, &fnew->key, &mask->key, extack);
 	if (err)
 		goto unbind_filter;
 
@@ -2800,18 +2804,19 @@ static void *fl_tmplt_create(struct net *net, struct tcf_chain *chain,
 			     struct nlattr **tca,
 			     struct netlink_ext_ack *extack)
 {
+	struct nlattr *tca_opts = tca[TCA_OPTIONS];
 	struct fl_flow_tmplt *tmplt;
 	struct nlattr **tb;
 	int err;
 
-	if (!tca[TCA_OPTIONS])
+	if (!tca_opts)
 		return ERR_PTR(-EINVAL);
 
 	tb = kcalloc(TCA_FLOWER_MAX + 1, sizeof(struct nlattr *), GFP_KERNEL);
 	if (!tb)
 		return ERR_PTR(-ENOBUFS);
 	err = nla_parse_nested_deprecated(tb, TCA_FLOWER_MAX,
-					  tca[TCA_OPTIONS], fl_policy, NULL);
+					  tca_opts, fl_policy, NULL);
 	if (err)
 		goto errout_tb;
 
@@ -2821,7 +2826,8 @@ static void *fl_tmplt_create(struct net *net, struct tcf_chain *chain,
 		goto errout_tb;
 	}
 	tmplt->chain = chain;
-	err = fl_set_key(net, tb, &tmplt->dummy_key, &tmplt->mask, extack);
+	err = fl_set_key(net, tca_opts, tb, &tmplt->dummy_key,
+			 &tmplt->mask, extack);
 	if (err)
 		goto errout_tmplt;
 
-- 
cgit v1.2.3-58-ga151


From c5eaf1b3f824369b6dcb33a39232b871d2ca8e33 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:49:12 -0700
Subject: PCI: Add Meta Platforms vendor ID

Add Meta as a vendor ID for PCI devices so we can use the macro for future
drivers.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://patch.msgid.link/172079935272.1778861.13619056509276833225.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 include/linux/pci_ids.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 677aea20d3e1..76a8f2d6bd64 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2601,6 +2601,8 @@
 
 #define PCI_VENDOR_ID_HYGON		0x1d94
 
+#define PCI_VENDOR_ID_META		0x1d9b
+
 #define PCI_VENDOR_ID_FUNGIBLE		0x1dad
 
 #define PCI_VENDOR_ID_HXT		0x1dbf
-- 
cgit v1.2.3-58-ga151


From 546dd90be97913c775b5bcd20b665f4ff5df9cd4 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:49:16 -0700
Subject: eth: fbnic: Add scaffolding for Meta's NIC driver

Create a bare-bones PCI driver for Meta's NIC.
Subsequent changes will flesh it out.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/172079935646.1778861.9710282776096050607.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 MAINTAINERS                                     |   7 +
 drivers/net/ethernet/Kconfig                    |   1 +
 drivers/net/ethernet/Makefile                   |   1 +
 drivers/net/ethernet/meta/Kconfig               |  30 ++++
 drivers/net/ethernet/meta/Makefile              |   6 +
 drivers/net/ethernet/meta/fbnic/Makefile        |  10 ++
 drivers/net/ethernet/meta/fbnic/fbnic.h         |  19 +++
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h     |   9 ++
 drivers/net/ethernet/meta/fbnic/fbnic_drvinfo.h |   5 +
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c     | 201 ++++++++++++++++++++++++
 10 files changed, 289 insertions(+)
 create mode 100644 drivers/net/ethernet/meta/Kconfig
 create mode 100644 drivers/net/ethernet/meta/Makefile
 create mode 100644 drivers/net/ethernet/meta/fbnic/Makefile
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic.h
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_csr.h
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_drvinfo.h
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_pci.c

diff --git a/MAINTAINERS b/MAINTAINERS
index 0b73a6e2d78c..dd4297ea41f9 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -14579,6 +14579,13 @@ T:	git git://linuxtv.org/media_tree.git
 F:	Documentation/devicetree/bindings/media/amlogic,gx-vdec.yaml
 F:	drivers/staging/media/meson/vdec/
 
+META ETHERNET DRIVERS
+M:	Alexander Duyck <alexanderduyck@fb.com>
+M:	Jakub Kicinski <kuba@kernel.org>
+R:	kernel-team@meta.com
+S:	Supported
+F:	drivers/net/ethernet/meta/
+
 METHODE UDPU SUPPORT
 M:	Robert Marko <robert.marko@sartura.hr>
 S:	Maintained
diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig
index 6a19b5393ed1..0baac25db4f8 100644
--- a/drivers/net/ethernet/Kconfig
+++ b/drivers/net/ethernet/Kconfig
@@ -122,6 +122,7 @@ source "drivers/net/ethernet/litex/Kconfig"
 source "drivers/net/ethernet/marvell/Kconfig"
 source "drivers/net/ethernet/mediatek/Kconfig"
 source "drivers/net/ethernet/mellanox/Kconfig"
+source "drivers/net/ethernet/meta/Kconfig"
 source "drivers/net/ethernet/micrel/Kconfig"
 source "drivers/net/ethernet/microchip/Kconfig"
 source "drivers/net/ethernet/mscc/Kconfig"
diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile
index 0d872d4efcd1..c03203439c0e 100644
--- a/drivers/net/ethernet/Makefile
+++ b/drivers/net/ethernet/Makefile
@@ -59,6 +59,7 @@ obj-$(CONFIG_NET_VENDOR_LITEX) += litex/
 obj-$(CONFIG_NET_VENDOR_MARVELL) += marvell/
 obj-$(CONFIG_NET_VENDOR_MEDIATEK) += mediatek/
 obj-$(CONFIG_NET_VENDOR_MELLANOX) += mellanox/
+obj-$(CONFIG_NET_VENDOR_META) += meta/
 obj-$(CONFIG_NET_VENDOR_MICREL) += micrel/
 obj-$(CONFIG_NET_VENDOR_MICROCHIP) += microchip/
 obj-$(CONFIG_NET_VENDOR_MICROSEMI) += mscc/
diff --git a/drivers/net/ethernet/meta/Kconfig b/drivers/net/ethernet/meta/Kconfig
new file mode 100644
index 000000000000..fbbc38e7e507
--- /dev/null
+++ b/drivers/net/ethernet/meta/Kconfig
@@ -0,0 +1,30 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Meta Platforms network device configuration
+#
+
+config NET_VENDOR_META
+	bool "Meta Platforms devices"
+	default y
+	help
+	  If you have a network (Ethernet) card designed by Meta, say Y.
+	  That's Meta as in the parent company of Facebook.
+
+	  Note that the answer to this question doesn't directly affect the
+	  kernel: saying N will just cause the configurator to skip all
+	  the questions about Meta cards. If you say Y, you will be asked for
+	  your specific card in the following questions.
+
+if NET_VENDOR_META
+
+config FBNIC
+	tristate "Meta Platforms Host Network Interface"
+	depends on X86_64 || COMPILE_TEST
+	depends on PCI_MSI
+	help
+	  This driver supports Meta Platforms Host Network Interface.
+
+	  To compile this driver as a module, choose M here. The module
+	  will be called fbnic.  MSI-X interrupt support is required.
+
+endif # NET_VENDOR_META
diff --git a/drivers/net/ethernet/meta/Makefile b/drivers/net/ethernet/meta/Makefile
new file mode 100644
index 000000000000..88804f3de963
--- /dev/null
+++ b/drivers/net/ethernet/meta/Makefile
@@ -0,0 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile for the Meta Platforms network device drivers.
+#
+
+obj-$(CONFIG_FBNIC) += fbnic/
diff --git a/drivers/net/ethernet/meta/fbnic/Makefile b/drivers/net/ethernet/meta/fbnic/Makefile
new file mode 100644
index 000000000000..ce277fec875f
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/Makefile
@@ -0,0 +1,10 @@
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+
+#
+# Makefile for the Meta(R) Host Network Interface
+#
+
+obj-$(CONFIG_FBNIC) += fbnic.o
+
+fbnic-y := fbnic_pci.o
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h
new file mode 100644
index 000000000000..25702dab8d66
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#ifndef _FBNIC_H_
+#define _FBNIC_H_
+
+#include "fbnic_csr.h"
+
+extern char fbnic_driver_name[];
+
+enum fbnic_boards {
+	fbnic_board_asic
+};
+
+struct fbnic_info {
+	unsigned int bar_mask;
+};
+
+#endif /* _FBNIC_H_ */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
new file mode 100644
index 000000000000..72e89c07bf54
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#ifndef _FBNIC_CSR_H_
+#define _FBNIC_CSR_H_
+
+#define PCI_DEVICE_ID_META_FBNIC_ASIC		0x0013
+
+#endif /* _FBNIC_CSR_H_ */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_drvinfo.h b/drivers/net/ethernet/meta/fbnic/fbnic_drvinfo.h
new file mode 100644
index 000000000000..809ba6729442
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_drvinfo.h
@@ -0,0 +1,5 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#define DRV_NAME "fbnic"
+#define DRV_SUMMARY "Meta(R) Host Network Interface Driver"
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
new file mode 100644
index 000000000000..aff031e74344
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -0,0 +1,201 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/types.h>
+
+#include "fbnic.h"
+#include "fbnic_drvinfo.h"
+
+char fbnic_driver_name[] = DRV_NAME;
+
+MODULE_DESCRIPTION(DRV_SUMMARY);
+MODULE_LICENSE("GPL");
+
+static const struct fbnic_info fbnic_asic_info = {
+	.bar_mask = BIT(0) | BIT(4)
+};
+
+static const struct fbnic_info *fbnic_info_tbl[] = {
+	[fbnic_board_asic] = &fbnic_asic_info,
+};
+
+static const struct pci_device_id fbnic_pci_tbl[] = {
+	{ PCI_DEVICE_DATA(META, FBNIC_ASIC, fbnic_board_asic) },
+	/* Required last entry */
+	{0, }
+};
+MODULE_DEVICE_TABLE(pci, fbnic_pci_tbl);
+
+/**
+ * fbnic_probe - Device Initialization Routine
+ * @pdev: PCI device information struct
+ * @ent: entry in fbnic_pci_tbl
+ *
+ * Initializes a PCI device identified by a pci_dev structure.
+ * The OS initialization, configuring of the adapter private structure,
+ * and a hardware reset occur.
+ *
+ * Return: 0 on success, negative on failure
+ **/
+static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	const struct fbnic_info *info = fbnic_info_tbl[ent->driver_data];
+	int err;
+
+	if (pdev->error_state != pci_channel_io_normal) {
+		dev_err(&pdev->dev,
+			"PCI device still in an error state. Unable to load...\n");
+		return -EIO;
+	}
+
+	err = pcim_enable_device(pdev);
+	if (err) {
+		dev_err(&pdev->dev, "PCI enable device failed: %d\n", err);
+		return err;
+	}
+
+	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(46));
+	if (err)
+		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+	if (err) {
+		dev_err(&pdev->dev, "DMA configuration failed: %d\n", err);
+		return err;
+	}
+
+	err = pcim_iomap_regions(pdev, info->bar_mask, fbnic_driver_name);
+	if (err) {
+		dev_err(&pdev->dev,
+			"pci_request_selected_regions failed: %d\n", err);
+		return err;
+	}
+
+	pci_set_master(pdev);
+	pci_save_state(pdev);
+
+	return 0;
+}
+
+/**
+ * fbnic_remove - Device Removal Routine
+ * @pdev: PCI device information struct
+ *
+ * Called by the PCI subsystem to alert the driver that it should release
+ * a PCI device.  The could be caused by a Hot-Plug event, or because the
+ * driver is going to be removed from memory.
+ **/
+static void fbnic_remove(struct pci_dev *pdev)
+{
+	pci_disable_device(pdev);
+}
+
+static int fbnic_pm_suspend(struct device *dev)
+{
+	return 0;
+}
+
+static int __fbnic_pm_resume(struct device *dev)
+{
+	return 0;
+}
+
+static int __maybe_unused fbnic_pm_resume(struct device *dev)
+{
+	int err;
+
+	err = __fbnic_pm_resume(dev);
+
+	return err;
+}
+
+static const struct dev_pm_ops fbnic_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(fbnic_pm_suspend, fbnic_pm_resume)
+};
+
+static void fbnic_shutdown(struct pci_dev *pdev)
+{
+	fbnic_pm_suspend(&pdev->dev);
+}
+
+static pci_ers_result_t fbnic_err_error_detected(struct pci_dev *pdev,
+						 pci_channel_state_t state)
+{
+	/* Disconnect device if failure is not recoverable via reset */
+	if (state == pci_channel_io_perm_failure)
+		return PCI_ERS_RESULT_DISCONNECT;
+
+	fbnic_pm_suspend(&pdev->dev);
+
+	/* Request a slot reset */
+	return PCI_ERS_RESULT_NEED_RESET;
+}
+
+static pci_ers_result_t fbnic_err_slot_reset(struct pci_dev *pdev)
+{
+	pci_set_power_state(pdev, PCI_D0);
+	pci_restore_state(pdev);
+	pci_save_state(pdev);
+
+	if (pci_enable_device_mem(pdev)) {
+		dev_err(&pdev->dev,
+			"Cannot re-enable PCI device after reset.\n");
+		return PCI_ERS_RESULT_DISCONNECT;
+	}
+
+	return PCI_ERS_RESULT_RECOVERED;
+}
+
+static void fbnic_err_resume(struct pci_dev *pdev)
+{
+}
+
+static const struct pci_error_handlers fbnic_err_handler = {
+	.error_detected	= fbnic_err_error_detected,
+	.slot_reset	= fbnic_err_slot_reset,
+	.resume		= fbnic_err_resume,
+};
+
+static struct pci_driver fbnic_driver = {
+	.name		= fbnic_driver_name,
+	.id_table	= fbnic_pci_tbl,
+	.probe		= fbnic_probe,
+	.remove		= fbnic_remove,
+	.driver.pm	= &fbnic_pm_ops,
+	.shutdown	= fbnic_shutdown,
+	.err_handler	= &fbnic_err_handler,
+};
+
+/**
+ * fbnic_init_module - Driver Registration Routine
+ *
+ * The first routine called when the driver is loaded.  All it does is
+ * register with the PCI subsystem.
+ *
+ * Return: 0 on success, negative on failure
+ **/
+static int __init fbnic_init_module(void)
+{
+	int err;
+
+	err = pci_register_driver(&fbnic_driver);
+	if (err)
+		goto out;
+
+	pr_info(DRV_SUMMARY " (%s)", fbnic_driver.name);
+out:
+	return err;
+}
+module_init(fbnic_init_module);
+
+/**
+ * fbnic_exit_module - Driver Exit Cleanup Routine
+ *
+ * Called just before the driver is removed from memory.
+ **/
+static void __exit fbnic_exit_module(void)
+{
+	pci_unregister_driver(&fbnic_driver);
+}
+module_exit(fbnic_exit_module);
-- 
cgit v1.2.3-58-ga151


From 1a9d48892ea5bf27f359f40d96511acc3eecea58 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:49:20 -0700
Subject: eth: fbnic: Allocate core device specific structures and devlink
 interface

At the core of the fbnic device will be the devlink interface. This
interface will eventually provide basic functionality in the event that
there are any issues with the network interface.

Add support for allocating the MSI-X vectors and setting up the BAR
mapping. With this we can start enabling various subsystems and start
brining up additional interfaces such the AXI fabric and the firmware
mailbox.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079936012.1778861.4670986685222676467.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/Makefile        |  4 +-
 drivers/net/ethernet/meta/fbnic/fbnic.h         | 27 ++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_devlink.c | 84 +++++++++++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_irq.c     | 39 ++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c     | 70 ++++++++++++++++++++-
 5 files changed, 222 insertions(+), 2 deletions(-)
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_irq.c

diff --git a/drivers/net/ethernet/meta/fbnic/Makefile b/drivers/net/ethernet/meta/fbnic/Makefile
index ce277fec875f..c06041e70bc5 100644
--- a/drivers/net/ethernet/meta/fbnic/Makefile
+++ b/drivers/net/ethernet/meta/fbnic/Makefile
@@ -7,4 +7,6 @@
 
 obj-$(CONFIG_FBNIC) += fbnic.o
 
-fbnic-y := fbnic_pci.o
+fbnic-y := fbnic_devlink.o \
+	   fbnic_irq.o \
+	   fbnic_pci.o
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h
index 25702dab8d66..db85b04e9b80 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic.h
@@ -6,8 +6,35 @@
 
 #include "fbnic_csr.h"
 
+struct fbnic_dev {
+	struct device *dev;
+
+	u32 __iomem *uc_addr0;
+	u32 __iomem *uc_addr4;
+	unsigned short num_irqs;
+
+	u64 dsn;
+};
+
+/* Reserve entry 0 in the MSI-X "others" array until we have filled all
+ * 32 of the possible interrupt slots. By doing this we can avoid any
+ * potential conflicts should we need to enable one of the debug interrupt
+ * causes later.
+ */
+enum {
+	FBNIC_NON_NAPI_VECTORS
+};
+
 extern char fbnic_driver_name[];
 
+void fbnic_devlink_free(struct fbnic_dev *fbd);
+struct fbnic_dev *fbnic_devlink_alloc(struct pci_dev *pdev);
+void fbnic_devlink_register(struct fbnic_dev *fbd);
+void fbnic_devlink_unregister(struct fbnic_dev *fbd);
+
+void fbnic_free_irqs(struct fbnic_dev *fbd);
+int fbnic_alloc_irqs(struct fbnic_dev *fbd);
+
 enum fbnic_boards {
 	fbnic_board_asic
 };
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c b/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
new file mode 100644
index 000000000000..91e8135410df
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
@@ -0,0 +1,84 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#include <asm/unaligned.h>
+#include <linux/pci.h>
+#include <linux/types.h>
+#include <net/devlink.h>
+
+#include "fbnic.h"
+
+#define FBNIC_SN_STR_LEN	24
+
+static int fbnic_devlink_info_get(struct devlink *devlink,
+				  struct devlink_info_req *req,
+				  struct netlink_ext_ack *extack)
+{
+	struct fbnic_dev *fbd = devlink_priv(devlink);
+	int err;
+
+	if (fbd->dsn) {
+		unsigned char serial[FBNIC_SN_STR_LEN];
+		u8 dsn[8];
+
+		put_unaligned_be64(fbd->dsn, dsn);
+		err = snprintf(serial, FBNIC_SN_STR_LEN, "%8phD", dsn);
+		if (err < 0)
+			return err;
+
+		err = devlink_info_serial_number_put(req, serial);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static const struct devlink_ops fbnic_devlink_ops = {
+	.info_get = fbnic_devlink_info_get,
+};
+
+void fbnic_devlink_free(struct fbnic_dev *fbd)
+{
+	struct devlink *devlink = priv_to_devlink(fbd);
+
+	devlink_free(devlink);
+}
+
+struct fbnic_dev *fbnic_devlink_alloc(struct pci_dev *pdev)
+{
+	void __iomem * const *iomap_table;
+	struct devlink *devlink;
+	struct fbnic_dev *fbd;
+
+	devlink = devlink_alloc(&fbnic_devlink_ops, sizeof(struct fbnic_dev),
+				&pdev->dev);
+	if (!devlink)
+		return NULL;
+
+	fbd = devlink_priv(devlink);
+	pci_set_drvdata(pdev, fbd);
+	fbd->dev = &pdev->dev;
+
+	iomap_table = pcim_iomap_table(pdev);
+	fbd->uc_addr0 = iomap_table[0];
+	fbd->uc_addr4 = iomap_table[4];
+
+	fbd->dsn = pci_get_dsn(pdev);
+
+	return fbd;
+}
+
+void fbnic_devlink_register(struct fbnic_dev *fbd)
+{
+	struct devlink *devlink = priv_to_devlink(fbd);
+
+	devlink_register(devlink);
+}
+
+void fbnic_devlink_unregister(struct fbnic_dev *fbd)
+{
+	struct devlink *devlink = priv_to_devlink(fbd);
+
+	devlink_unregister(devlink);
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c
new file mode 100644
index 000000000000..7d1475750b64
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c
@@ -0,0 +1,39 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#include <linux/pci.h>
+#include <linux/types.h>
+
+#include "fbnic.h"
+
+void fbnic_free_irqs(struct fbnic_dev *fbd)
+{
+	struct pci_dev *pdev = to_pci_dev(fbd->dev);
+
+	fbd->num_irqs = 0;
+
+	pci_free_irq_vectors(pdev);
+}
+
+int fbnic_alloc_irqs(struct fbnic_dev *fbd)
+{
+	unsigned int wanted_irqs = FBNIC_NON_NAPI_VECTORS;
+	struct pci_dev *pdev = to_pci_dev(fbd->dev);
+	int num_irqs;
+
+	wanted_irqs += 1;
+	num_irqs = pci_alloc_irq_vectors(pdev, FBNIC_NON_NAPI_VECTORS + 1,
+					 wanted_irqs, PCI_IRQ_MSIX);
+	if (num_irqs < 0) {
+		dev_err(fbd->dev, "Failed to allocate MSI-X entries\n");
+		return num_irqs;
+	}
+
+	if (num_irqs < wanted_irqs)
+		dev_warn(fbd->dev, "Allocated %d IRQs, expected %d\n",
+			 num_irqs, wanted_irqs);
+
+	fbd->num_irqs = num_irqs;
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
index aff031e74344..fffe8a7cc96e 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -43,6 +43,7 @@ MODULE_DEVICE_TABLE(pci, fbnic_pci_tbl);
 static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	const struct fbnic_info *info = fbnic_info_tbl[ent->driver_data];
+	struct fbnic_dev *fbd;
 	int err;
 
 	if (pdev->error_state != pci_channel_io_normal) {
@@ -72,10 +73,39 @@ static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return err;
 	}
 
+	fbd = fbnic_devlink_alloc(pdev);
+	if (!fbd) {
+		dev_err(&pdev->dev, "Devlink allocation failed\n");
+		return -ENOMEM;
+	}
+
 	pci_set_master(pdev);
 	pci_save_state(pdev);
 
+	err = fbnic_alloc_irqs(fbd);
+	if (err)
+		goto free_fbd;
+
+	fbnic_devlink_register(fbd);
+
+	if (!fbd->dsn) {
+		dev_warn(&pdev->dev, "Reading serial number failed\n");
+		goto init_failure_mode;
+	}
+
 	return 0;
+
+init_failure_mode:
+	dev_warn(&pdev->dev, "Probe error encountered, entering init failure mode. Normal networking functionality will not be available.\n");
+	 /* Always return 0 even on error so devlink is registered to allow
+	  * firmware updates for fixes.
+	  */
+	return 0;
+free_fbd:
+	pci_disable_device(pdev);
+	fbnic_devlink_free(fbd);
+
+	return err;
 }
 
 /**
@@ -88,17 +118,50 @@ static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
  **/
 static void fbnic_remove(struct pci_dev *pdev)
 {
+	struct fbnic_dev *fbd = pci_get_drvdata(pdev);
+
+	fbnic_devlink_unregister(fbd);
+	fbnic_free_irqs(fbd);
+
 	pci_disable_device(pdev);
+	fbnic_devlink_free(fbd);
 }
 
 static int fbnic_pm_suspend(struct device *dev)
 {
+	struct fbnic_dev *fbd = dev_get_drvdata(dev);
+
+	/* Free the IRQs so they aren't trying to occupy sleeping CPUs */
+	fbnic_free_irqs(fbd);
+
+	/* Hardware is about to go away, so switch off MMIO access internally */
+	WRITE_ONCE(fbd->uc_addr0, NULL);
+	WRITE_ONCE(fbd->uc_addr4, NULL);
+
 	return 0;
 }
 
 static int __fbnic_pm_resume(struct device *dev)
 {
+	struct fbnic_dev *fbd = dev_get_drvdata(dev);
+	void __iomem * const *iomap_table;
+	int err;
+
+	/* Restore MMIO access */
+	iomap_table = pcim_iomap_table(to_pci_dev(dev));
+	fbd->uc_addr0 = iomap_table[0];
+	fbd->uc_addr4 = iomap_table[4];
+
+	/* Rerequest the IRQs */
+	err = fbnic_alloc_irqs(fbd);
+	if (err)
+		goto err_invalidate_uc_addr;
+
 	return 0;
+err_invalidate_uc_addr:
+	WRITE_ONCE(fbd->uc_addr0, NULL);
+	WRITE_ONCE(fbd->uc_addr4, NULL);
+	return err;
 }
 
 static int __maybe_unused fbnic_pm_resume(struct device *dev)
@@ -134,6 +197,8 @@ static pci_ers_result_t fbnic_err_error_detected(struct pci_dev *pdev,
 
 static pci_ers_result_t fbnic_err_slot_reset(struct pci_dev *pdev)
 {
+	int err;
+
 	pci_set_power_state(pdev, PCI_D0);
 	pci_restore_state(pdev);
 	pci_save_state(pdev);
@@ -144,7 +209,10 @@ static pci_ers_result_t fbnic_err_slot_reset(struct pci_dev *pdev)
 		return PCI_ERS_RESULT_DISCONNECT;
 	}
 
-	return PCI_ERS_RESULT_RECOVERED;
+	/* Restore device to previous state */
+	err = __fbnic_pm_resume(&pdev->dev);
+
+	return err ? PCI_ERS_RESULT_DISCONNECT : PCI_ERS_RESULT_RECOVERED;
 }
 
 static void fbnic_err_resume(struct pci_dev *pdev)
-- 
cgit v1.2.3-58-ga151


From 3646153161f1052972d9d2062d8c27b0e30c89f8 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:49:23 -0700
Subject: eth: fbnic: Add register init to set PCIe/Ethernet device config

As a part of enabling the device the first step is to configure the AXI and
Ethernet interfaces to allow for basic traffic. This consists of
configuring several registers related to the PCIe and Ethernet FIFOs as
well as configuring the handlers for moving traffic between entities.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079936376.1778861.15942501417449077552.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/Makefile        |   1 +
 drivers/net/ethernet/meta/fbnic/fbnic.h         |  41 +++
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h     | 312 +++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_devlink.c |   2 +
 drivers/net/ethernet/meta/fbnic/fbnic_mac.c     | 425 ++++++++++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_mac.h     |  25 ++
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c     |  39 +++
 7 files changed, 845 insertions(+)
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_mac.c
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_mac.h

diff --git a/drivers/net/ethernet/meta/fbnic/Makefile b/drivers/net/ethernet/meta/fbnic/Makefile
index c06041e70bc5..b8f4511440dc 100644
--- a/drivers/net/ethernet/meta/fbnic/Makefile
+++ b/drivers/net/ethernet/meta/fbnic/Makefile
@@ -9,4 +9,5 @@ obj-$(CONFIG_FBNIC) += fbnic.o
 
 fbnic-y := fbnic_devlink.o \
 	   fbnic_irq.o \
+	   fbnic_mac.o \
 	   fbnic_pci.o
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h
index db85b04e9b80..af863dfabd82 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic.h
@@ -4,16 +4,22 @@
 #ifndef _FBNIC_H_
 #define _FBNIC_H_
 
+#include <linux/io.h>
+
 #include "fbnic_csr.h"
+#include "fbnic_mac.h"
 
 struct fbnic_dev {
 	struct device *dev;
 
 	u32 __iomem *uc_addr0;
 	u32 __iomem *uc_addr4;
+	const struct fbnic_mac *mac;
 	unsigned short num_irqs;
 
 	u64 dsn;
+	u32 mps;
+	u32 readrq;
 };
 
 /* Reserve entry 0 in the MSI-X "others" array until we have filled all
@@ -25,6 +31,41 @@ enum {
 	FBNIC_NON_NAPI_VECTORS
 };
 
+static inline bool fbnic_present(struct fbnic_dev *fbd)
+{
+	return !!READ_ONCE(fbd->uc_addr0);
+}
+
+static inline void fbnic_wr32(struct fbnic_dev *fbd, u32 reg, u32 val)
+{
+	u32 __iomem *csr = READ_ONCE(fbd->uc_addr0);
+
+	if (csr)
+		writel(val, csr + reg);
+}
+
+u32 fbnic_rd32(struct fbnic_dev *fbd, u32 reg);
+
+static inline void fbnic_wrfl(struct fbnic_dev *fbd)
+{
+	fbnic_rd32(fbd, FBNIC_MASTER_SPARE_0);
+}
+
+static inline void
+fbnic_rmw32(struct fbnic_dev *fbd, u32 reg, u32 mask, u32 val)
+{
+	u32 v;
+
+	v = fbnic_rd32(fbd, reg);
+	v &= ~mask;
+	v |= val;
+	fbnic_wr32(fbd, reg, v);
+}
+
+#define wr32(_f, _r, _v)	fbnic_wr32(_f, _r, _v)
+#define rd32(_f, _r)		fbnic_rd32(_f, _r)
+#define wrfl(_f)		fbnic_wrfl(_f)
+
 extern char fbnic_driver_name[];
 
 void fbnic_devlink_free(struct fbnic_dev *fbd);
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index 72e89c07bf54..56a8bbd8b720 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -4,6 +4,318 @@
 #ifndef _FBNIC_CSR_H_
 #define _FBNIC_CSR_H_
 
+#include <linux/bitops.h>
+
+#define CSR_BIT(nr)		(1u << (nr))
+#define CSR_GENMASK(h, l)	GENMASK(h, l)
+
 #define PCI_DEVICE_ID_META_FBNIC_ASIC		0x0013
 
+#define FBNIC_CLOCK_FREQ	(600 * (1000 * 1000))
+
+/* Global QM Tx registers */
+#define FBNIC_CSR_START_QM_TX		0x00800	/* CSR section delimiter */
+#define FBNIC_QM_TWQ_DEFAULT_META_L	0x00818		/* 0x02060 */
+#define FBNIC_QM_TWQ_DEFAULT_META_H	0x00819		/* 0x02064 */
+
+#define FBNIC_QM_TQS_CTL0		0x0081b		/* 0x0206c */
+#define FBNIC_QM_TQS_CTL0_LSO_TS_MASK	CSR_BIT(0)
+enum {
+	FBNIC_QM_TQS_CTL0_LSO_TS_FIRST	= 0,
+	FBNIC_QM_TQS_CTL0_LSO_TS_LAST	= 1,
+};
+
+#define FBNIC_QM_TQS_CTL0_PREFETCH_THRESH	CSR_GENMASK(7, 1)
+enum {
+	FBNIC_QM_TQS_CTL0_PREFETCH_THRESH_MIN	= 16,
+};
+
+#define FBNIC_QM_TQS_CTL1		0x0081c		/* 0x02070 */
+#define FBNIC_QM_TQS_CTL1_MC_MAX_CREDITS	CSR_GENMASK(7, 0)
+#define FBNIC_QM_TQS_CTL1_BULK_MAX_CREDITS	CSR_GENMASK(15, 8)
+#define FBNIC_QM_TQS_MTU_CTL0		0x0081d		/* 0x02074 */
+#define FBNIC_QM_TQS_MTU_CTL1		0x0081e		/* 0x02078 */
+#define FBNIC_QM_TQS_MTU_CTL1_BULK		CSR_GENMASK(13, 0)
+#define FBNIC_QM_TCQ_CTL0		0x0082d		/* 0x020b4 */
+#define FBNIC_QM_TCQ_CTL0_COAL_WAIT		CSR_GENMASK(15, 0)
+#define FBNIC_QM_TCQ_CTL0_TICK_CYCLES		CSR_GENMASK(26, 16)
+#define FBNIC_QM_TQS_EDT_TS_RANGE	0x00849		/* 0x2124 */
+#define FBNIC_QM_TNI_TDF_CTL		0x0086c		/* 0x021b0 */
+#define FBNIC_QM_TNI_TDF_CTL_MRRS		CSR_GENMASK(1, 0)
+#define FBNIC_QM_TNI_TDF_CTL_CLS		CSR_GENMASK(3, 2)
+#define FBNIC_QM_TNI_TDF_CTL_MAX_OT		CSR_GENMASK(11, 4)
+#define FBNIC_QM_TNI_TDF_CTL_MAX_OB		CSR_GENMASK(23, 12)
+#define FBNIC_QM_TNI_TDE_CTL		0x0086d		/* 0x021b4 */
+#define FBNIC_QM_TNI_TDE_CTL_MRRS		CSR_GENMASK(1, 0)
+#define FBNIC_QM_TNI_TDE_CTL_CLS		CSR_GENMASK(3, 2)
+#define FBNIC_QM_TNI_TDE_CTL_MAX_OT		CSR_GENMASK(11, 4)
+#define FBNIC_QM_TNI_TDE_CTL_MAX_OB		CSR_GENMASK(24, 12)
+#define FBNIC_QM_TNI_TDE_CTL_MRRS_1K		CSR_BIT(25)
+#define FBNIC_QM_TNI_TCM_CTL		0x0086e		/* 0x021b8 */
+#define FBNIC_QM_TNI_TCM_CTL_MPS		CSR_GENMASK(1, 0)
+#define FBNIC_QM_TNI_TCM_CTL_CLS		CSR_GENMASK(3, 2)
+#define FBNIC_QM_TNI_TCM_CTL_MAX_OT		CSR_GENMASK(11, 4)
+#define FBNIC_QM_TNI_TCM_CTL_MAX_OB		CSR_GENMASK(23, 12)
+#define FBNIC_CSR_END_QM_TX		0x00873	/* CSR section delimiter */
+
+/* Global QM Rx registers */
+#define FBNIC_CSR_START_QM_RX		0x00c00	/* CSR section delimiter */
+#define FBNIC_QM_RCQ_CTL0		0x00c0c		/* 0x03030 */
+#define FBNIC_QM_RCQ_CTL0_COAL_WAIT		CSR_GENMASK(15, 0)
+#define FBNIC_QM_RCQ_CTL0_TICK_CYCLES		CSR_GENMASK(26, 16)
+#define FBNIC_QM_RNI_RBP_CTL		0x00c2d		/* 0x030b4 */
+#define FBNIC_QM_RNI_RBP_CTL_MRRS		CSR_GENMASK(1, 0)
+#define FBNIC_QM_RNI_RBP_CTL_CLS		CSR_GENMASK(3, 2)
+#define FBNIC_QM_RNI_RBP_CTL_MAX_OT		CSR_GENMASK(11, 4)
+#define FBNIC_QM_RNI_RBP_CTL_MAX_OB		CSR_GENMASK(23, 12)
+#define FBNIC_QM_RNI_RDE_CTL		0x00c2e		/* 0x030b8 */
+#define FBNIC_QM_RNI_RDE_CTL_MPS		CSR_GENMASK(1, 0)
+#define FBNIC_QM_RNI_RDE_CTL_CLS		CSR_GENMASK(3, 2)
+#define FBNIC_QM_RNI_RDE_CTL_MAX_OT		CSR_GENMASK(11, 4)
+#define FBNIC_QM_RNI_RDE_CTL_MAX_OB		CSR_GENMASK(23, 12)
+#define FBNIC_QM_RNI_RCM_CTL		0x00c2f		/* 0x030bc */
+#define FBNIC_QM_RNI_RCM_CTL_MPS		CSR_GENMASK(1, 0)
+#define FBNIC_QM_RNI_RCM_CTL_CLS		CSR_GENMASK(3, 2)
+#define FBNIC_QM_RNI_RCM_CTL_MAX_OT		CSR_GENMASK(11, 4)
+#define FBNIC_QM_RNI_RCM_CTL_MAX_OB		CSR_GENMASK(23, 12)
+#define FBNIC_CSR_END_QM_RX		0x00c34	/* CSR section delimiter */
+
+/* TCE registers */
+#define FBNIC_CSR_START_TCE		0x04000	/* CSR section delimiter */
+#define FBNIC_TCE_REG_BASE		0x04000		/* 0x10000 */
+
+#define FBNIC_TCE_LSO_CTRL		0x04000		/* 0x10000 */
+#define FBNIC_TCE_LSO_CTRL_TCPF_CLR_1ST		CSR_GENMASK(8, 0)
+#define FBNIC_TCE_LSO_CTRL_TCPF_CLR_MID		CSR_GENMASK(17, 9)
+#define FBNIC_TCE_LSO_CTRL_TCPF_CLR_END		CSR_GENMASK(26, 18)
+#define FBNIC_TCE_LSO_CTRL_IPID_MODE_INC	CSR_BIT(27)
+
+#define FBNIC_TCE_CSO_CTRL		0x04001		/* 0x10004 */
+#define FBNIC_TCE_CSO_CTRL_TCP_ZERO_CSUM	CSR_BIT(0)
+
+#define FBNIC_TCE_TXB_CTRL		0x04002		/* 0x10008 */
+#define FBNIC_TCE_TXB_CTRL_LOAD			CSR_BIT(0)
+#define FBNIC_TCE_TXB_CTRL_TCAM_ENABLE		CSR_BIT(1)
+#define FBNIC_TCE_TXB_CTRL_DISABLE		CSR_BIT(2)
+
+#define FBNIC_TCE_TXB_ENQ_WRR_CTRL	0x04003		/* 0x1000c */
+#define FBNIC_TCE_TXB_ENQ_WRR_CTRL_WEIGHT0	CSR_GENMASK(7, 0)
+#define FBNIC_TCE_TXB_ENQ_WRR_CTRL_WEIGHT1	CSR_GENMASK(15, 8)
+#define FBNIC_TCE_TXB_ENQ_WRR_CTRL_WEIGHT2	CSR_GENMASK(23, 16)
+
+#define FBNIC_TCE_TXB_TEI_Q0_CTRL	0x04004		/* 0x10010 */
+#define FBNIC_TCE_TXB_TEI_Q1_CTRL	0x04005		/* 0x10014 */
+#define FBNIC_TCE_TXB_MC_Q_CTRL		0x04006		/* 0x10018 */
+#define FBNIC_TCE_TXB_RX_TEI_Q_CTRL	0x04007		/* 0x1001c */
+#define FBNIC_TCE_TXB_RX_BMC_Q_CTRL	0x04008		/* 0x10020 */
+#define FBNIC_TCE_TXB_Q_CTRL_START		CSR_GENMASK(10, 0)
+#define FBNIC_TCE_TXB_Q_CTRL_SIZE		CSR_GENMASK(22, 11)
+
+#define FBNIC_TCE_TXB_TEI_DWRR_CTRL	0x04009		/* 0x10024 */
+#define FBNIC_TCE_TXB_TEI_DWRR_CTRL_QUANTUM0	CSR_GENMASK(7, 0)
+#define FBNIC_TCE_TXB_TEI_DWRR_CTRL_QUANTUM1	CSR_GENMASK(15, 8)
+#define FBNIC_TCE_TXB_NTWRK_DWRR_CTRL	0x0400a		/* 0x10028 */
+#define FBNIC_TCE_TXB_NTWRK_DWRR_CTRL_QUANTUM0	CSR_GENMASK(7, 0)
+#define FBNIC_TCE_TXB_NTWRK_DWRR_CTRL_QUANTUM1	CSR_GENMASK(15, 8)
+#define FBNIC_TCE_TXB_NTWRK_DWRR_CTRL_QUANTUM2	CSR_GENMASK(23, 16)
+
+#define FBNIC_TCE_TXB_CLDR_CFG		0x0400b		/* 0x1002c */
+#define FBNIC_TCE_TXB_CLDR_CFG_NUM_SLOT		CSR_GENMASK(5, 0)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG(n)	(0x0400c + (n))	/* 0x10030 + 4*n */
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_CNT		16
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_0_0	CSR_GENMASK(1, 0)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_0_1	CSR_GENMASK(3, 2)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_0_2	CSR_GENMASK(5, 4)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_0_3	CSR_GENMASK(7, 6)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_1_0	CSR_GENMASK(9, 8)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_1_1	CSR_GENMASK(11, 10)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_1_2	CSR_GENMASK(13, 12)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_1_3	CSR_GENMASK(15, 14)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_2_0	CSR_GENMASK(17, 16)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_2_1	CSR_GENMASK(19, 18)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_2_2	CSR_GENMASK(21, 20)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_2_3	CSR_GENMASK(23, 22)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_3_0	CSR_GENMASK(25, 24)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_3_1	CSR_GENMASK(27, 26)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_3_2	CSR_GENMASK(29, 28)
+#define FBNIC_TCE_TXB_CLDR_SLOT_CFG_DEST_ID_3_3	CSR_GENMASK(31, 30)
+
+#define FBNIC_TCE_BMC_MAX_PKTSZ		0x0403a		/* 0x100e8 */
+#define FBNIC_TCE_BMC_MAX_PKTSZ_TX		CSR_GENMASK(13, 0)
+#define FBNIC_TCE_BMC_MAX_PKTSZ_RX		CSR_GENMASK(27, 14)
+#define FBNIC_TCE_MC_MAX_PKTSZ		0x0403b		/* 0x100ec */
+#define FBNIC_TCE_MC_MAX_PKTSZ_TMI		CSR_GENMASK(13, 0)
+
+#define FBNIC_TCE_SOP_PROT_CTRL		0x0403c		/* 0x100f0 */
+#define FBNIC_TCE_SOP_PROT_CTRL_TBI		CSR_GENMASK(7, 0)
+#define FBNIC_TCE_SOP_PROT_CTRL_TTI_FRM		CSR_GENMASK(14, 8)
+#define FBNIC_TCE_SOP_PROT_CTRL_TTI_CM		CSR_GENMASK(18, 15)
+
+#define FBNIC_TCE_DROP_CTRL		0x0403d		/* 0x100f4 */
+#define FBNIC_TCE_DROP_CTRL_TTI_CM_DROP_EN	CSR_BIT(0)
+#define FBNIC_TCE_DROP_CTRL_TTI_FRM_DROP_EN	CSR_BIT(1)
+#define FBNIC_TCE_DROP_CTRL_TTI_TBI_DROP_EN	CSR_BIT(2)
+
+#define FBNIC_TCE_TXB_TX_BMC_Q_CTRL	0x0404B		/* 0x1012c */
+#define FBNIC_TCE_TXB_BMC_DWRR_CTRL	0x0404C		/* 0x10130 */
+#define FBNIC_TCE_TXB_BMC_DWRR_CTRL_QUANTUM0	CSR_GENMASK(7, 0)
+#define FBNIC_TCE_TXB_BMC_DWRR_CTRL_QUANTUM1	CSR_GENMASK(15, 8)
+#define FBNIC_TCE_TXB_TEI_DWRR_CTRL_EXT	0x0404D		/* 0x10134 */
+#define FBNIC_TCE_TXB_NTWRK_DWRR_CTRL_EXT \
+					0x0404E		/* 0x10138 */
+#define FBNIC_TCE_TXB_BMC_DWRR_CTRL_EXT	0x0404F		/* 0x1013c */
+#define FBNIC_CSR_END_TCE		0x04050	/* CSR section delimiter */
+
+/* TMI registers */
+#define FBNIC_CSR_START_TMI		0x04400	/* CSR section delimiter */
+#define FBNIC_TMI_SOP_PROT_CTRL		0x04400		/* 0x11000 */
+#define FBNIC_CSR_END_TMI		0x0443f	/* CSR section delimiter */
+/* Rx Buffer Registers */
+#define FBNIC_CSR_START_RXB		0x08000	/* CSR section delimiter */
+enum {
+	FBNIC_RXB_FIFO_MC		= 0,
+	/* Unused */
+	/* Unused */
+	FBNIC_RXB_FIFO_NET_TO_BMC	= 3,
+	FBNIC_RXB_FIFO_HOST		= 4,
+	/* Unused */
+	FBNIC_RXB_FIFO_BMC_TO_HOST	= 6,
+	/* Unused */
+	FBNIC_RXB_FIFO_INDICES		= 8
+};
+
+#define FBNIC_RXB_CT_SIZE(n)		(0x08000 + (n))	/* 0x20000 + 4*n */
+#define FBNIC_RXB_CT_SIZE_CNT			8
+#define FBNIC_RXB_CT_SIZE_HEADER		CSR_GENMASK(5, 0)
+#define FBNIC_RXB_CT_SIZE_PAYLOAD		CSR_GENMASK(11, 6)
+#define FBNIC_RXB_CT_SIZE_ENABLE		CSR_BIT(12)
+#define FBNIC_RXB_PAUSE_DROP_CTRL	0x08008		/* 0x20020 */
+#define FBNIC_RXB_PAUSE_DROP_CTRL_DROP_ENABLE	CSR_GENMASK(7, 0)
+#define FBNIC_RXB_PAUSE_DROP_CTRL_PAUSE_ENABLE	CSR_GENMASK(15, 8)
+#define FBNIC_RXB_PAUSE_DROP_CTRL_ECN_ENABLE	CSR_GENMASK(23, 16)
+#define FBNIC_RXB_PAUSE_DROP_CTRL_PS_ENABLE	CSR_GENMASK(27, 24)
+#define FBNIC_RXB_PAUSE_THLD(n)		(0x08009 + (n)) /* 0x20024 + 4*n */
+#define FBNIC_RXB_PAUSE_THLD_CNT		8
+#define FBNIC_RXB_PAUSE_THLD_ON			CSR_GENMASK(12, 0)
+#define FBNIC_RXB_PAUSE_THLD_OFF		CSR_GENMASK(25, 13)
+#define FBNIC_RXB_DROP_THLD(n)		(0x08011 + (n)) /* 0x20044 + 4*n */
+#define FBNIC_RXB_DROP_THLD_CNT			8
+#define FBNIC_RXB_DROP_THLD_ON			CSR_GENMASK(12, 0)
+#define FBNIC_RXB_DROP_THLD_OFF			CSR_GENMASK(25, 13)
+#define FBNIC_RXB_ECN_THLD(n)		(0x0801e + (n)) /* 0x20078 + 4*n */
+#define FBNIC_RXB_ECN_THLD_CNT			8
+#define FBNIC_RXB_ECN_THLD_ON			CSR_GENMASK(12, 0)
+#define FBNIC_RXB_ECN_THLD_OFF			CSR_GENMASK(25, 13)
+#define FBNIC_RXB_PBUF_CFG(n)		(0x08027 + (n))	/* 0x2009c + 4*n */
+#define FBNIC_RXB_PBUF_CFG_CNT			8
+#define FBNIC_RXB_PBUF_BASE_ADDR		CSR_GENMASK(12, 0)
+#define FBNIC_RXB_PBUF_SIZE			CSR_GENMASK(21, 13)
+#define FBNIC_RXB_DWRR_RDE_WEIGHT0	0x0802f		/* 0x200bc */
+#define FBNIC_RXB_DWRR_RDE_WEIGHT0_QUANTUM0	CSR_GENMASK(7, 0)
+#define FBNIC_RXB_DWRR_RDE_WEIGHT0_QUANTUM1	CSR_GENMASK(15, 8)
+#define FBNIC_RXB_DWRR_RDE_WEIGHT0_QUANTUM2	CSR_GENMASK(23, 16)
+#define FBNIC_RXB_DWRR_RDE_WEIGHT0_QUANTUM3	CSR_GENMASK(31, 24)
+#define FBNIC_RXB_DWRR_RDE_WEIGHT1	0x08030		/* 0x200c0 */
+#define FBNIC_RXB_DWRR_RDE_WEIGHT1_QUANTUM4	CSR_GENMASK(7, 0)
+#define FBNIC_RXB_DWRR_BMC_WEIGHT	0x08031		/* 0x200c4 */
+#define FBNIC_RXB_CLDR_PRIO_CFG(n)	(0x8034 + (n))	/* 0x200d0 + 4*n */
+#define FBNIC_RXB_CLDR_PRIO_CFG_CNT		16
+#define FBNIC_RXB_ENDIAN_FCS		0x08044		/* 0x20110 */
+enum {
+	/* Unused */
+	/* Unused */
+	FBNIC_RXB_DEQUEUE_BMC		= 2,
+	FBNIC_RXB_DEQUEUE_HOST		= 3,
+	FBNIC_RXB_DEQUEUE_INDICES	= 4
+};
+
+#define FBNIC_RXB_PBUF_CREDIT(n)	(0x08047 + (n))	/* 0x2011C + 4*n */
+#define FBNIC_RXB_PBUF_CREDIT_CNT		8
+#define FBNIC_RXB_PBUF_CREDIT_MASK		CSR_GENMASK(13, 0)
+#define FBNIC_RXB_INTF_CREDIT		0x0804f		/* 0x2013C */
+#define FBNIC_RXB_INTF_CREDIT_MASK0		CSR_GENMASK(3, 0)
+#define FBNIC_RXB_INTF_CREDIT_MASK1		CSR_GENMASK(7, 4)
+#define FBNIC_RXB_INTF_CREDIT_MASK2		CSR_GENMASK(11, 8)
+#define FBNIC_RXB_INTF_CREDIT_MASK3		CSR_GENMASK(15, 12)
+
+#define FBNIC_RXB_PAUSE_EVENT_CNT(n)	(0x08053 + (n))	/* 0x2014c + 4*n */
+#define FBNIC_RXB_DROP_FRMS_STS(n)	(0x08057 + (n))	/* 0x2015c + 4*n */
+#define FBNIC_RXB_DROP_BYTES_STS_L(n) \
+				(0x08080 + 2 * (n))	/* 0x20200 + 8*n */
+#define FBNIC_RXB_DROP_BYTES_STS_H(n) \
+				(0x08081 + 2 * (n))	/* 0x20204 + 8*n */
+#define FBNIC_RXB_TRUN_FRMS_STS(n)	(0x08091 + (n))	/* 0x20244 + 4*n */
+#define FBNIC_RXB_TRUN_BYTES_STS_L(n) \
+				(0x080c0 + 2 * (n))	/* 0x20300 + 8*n */
+#define FBNIC_RXB_TRUN_BYTES_STS_H(n) \
+				(0x080c1 + 2 * (n))	/* 0x20304 + 8*n */
+#define FBNIC_RXB_TRANS_PAUSE_STS(n)	(0x080d1 + (n))	/* 0x20344 + 4*n */
+#define FBNIC_RXB_TRANS_DROP_STS(n)	(0x080d9 + (n))	/* 0x20364 + 4*n */
+#define FBNIC_RXB_TRANS_ECN_STS(n)	(0x080e1 + (n))	/* 0x20384 + 4*n */
+enum {
+	FBNIC_RXB_ENQUEUE_NET		= 0,
+	FBNIC_RXB_ENQUEUE_BMC		= 1,
+	/* Unused */
+	/* Unused */
+	FBNIC_RXB_ENQUEUE_INDICES	= 4
+};
+
+#define FBNIC_RXB_DRBO_FRM_CNT_SRC(n)	(0x080f9 + (n))	/* 0x203e4 + 4*n */
+#define FBNIC_RXB_DRBO_BYTE_CNT_SRC_L(n) \
+					(0x080fd + (n))	/* 0x203f4 + 4*n */
+#define FBNIC_RXB_DRBO_BYTE_CNT_SRC_H(n) \
+					(0x08101 + (n))	/* 0x20404 + 4*n */
+#define FBNIC_RXB_INTF_FRM_CNT_DST(n)	(0x08105 + (n))	/* 0x20414 + 4*n */
+#define FBNIC_RXB_INTF_BYTE_CNT_DST_L(n) \
+					(0x08109 + (n))	/* 0x20424 + 4*n */
+#define FBNIC_RXB_INTF_BYTE_CNT_DST_H(n) \
+					(0x0810d + (n))	/* 0x20434 + 4*n */
+#define FBNIC_RXB_PBUF_FRM_CNT_DST(n)	(0x08111 + (n))	/* 0x20444 + 4*n */
+#define FBNIC_RXB_PBUF_BYTE_CNT_DST_L(n) \
+					(0x08115 + (n))	/* 0x20454 + 4*n */
+#define FBNIC_RXB_PBUF_BYTE_CNT_DST_H(n) \
+					(0x08119 + (n))	/* 0x20464 + 4*n */
+
+#define FBNIC_RXB_PBUF_FIFO_LEVEL(n)	(0x0811d + (n)) /* 0x20474 + 4*n */
+
+#define FBNIC_RXB_INTEGRITY_ERR(n)	(0x0812f + (n))	/* 0x204bc + 4*n */
+#define FBNIC_RXB_MAC_ERR(n)		(0x08133 + (n))	/* 0x204cc + 4*n */
+#define FBNIC_RXB_PARSER_ERR(n)		(0x08137 + (n))	/* 0x204dc + 4*n */
+#define FBNIC_RXB_FRM_ERR(n)		(0x0813b + (n))	/* 0x204ec + 4*n */
+
+#define FBNIC_RXB_DWRR_RDE_WEIGHT0_EXT	0x08143		/* 0x2050c */
+#define FBNIC_RXB_DWRR_RDE_WEIGHT1_EXT	0x08144		/* 0x20510 */
+#define FBNIC_CSR_END_RXB		0x081b1	/* CSR section delimiter */
+
+/* Rx Parser and Classifier Registers */
+#define FBNIC_CSR_START_RPC		0x08400	/* CSR section delimiter */
+#define FBNIC_RPC_RMI_CONFIG		0x08400		/* 0x21000 */
+#define FBNIC_RPC_RMI_CONFIG_OH_BYTES		CSR_GENMASK(4, 0)
+#define FBNIC_RPC_RMI_CONFIG_FCS_PRESENT	CSR_BIT(8)
+#define FBNIC_RPC_RMI_CONFIG_ENABLE		CSR_BIT(12)
+#define FBNIC_RPC_RMI_CONFIG_MTU		CSR_GENMASK(31, 16)
+#define FBNIC_CSR_END_RPC		0x0856b	/* CSR section delimiter */
+
+/* Fab Registers */
+#define FBNIC_CSR_START_FAB		0x0C000 /* CSR section delimiter */
+#define FBNIC_FAB_AXI4_AR_SPACER_2_CFG		0x0C005		/* 0x30014 */
+#define FBNIC_FAB_AXI4_AR_SPACER_MASK		CSR_BIT(16)
+#define FBNIC_FAB_AXI4_AR_SPACER_THREADSHOLD	CSR_GENMASK(15, 0)
+#define FBNIC_CSR_END_FAB		0x0C020	    /* CSR section delimiter */
+
+/* Master Registers */
+#define FBNIC_CSR_START_MASTER		0x0C400	/* CSR section delimiter */
+#define FBNIC_MASTER_SPARE_0		0x0C41B		/* 0x3106c */
+#define FBNIC_CSR_END_MASTER		0x0C452	/* CSR section delimiter */
+
+/* PUL User Registers */
+#define FBNIC_CSR_START_PUL_USER	0x31000	/* CSR section delimiter */
+#define FBNIC_PUL_OB_TLP_HDR_AW_CFG	0x3103d		/* 0xc40f4 */
+#define FBNIC_PUL_OB_TLP_HDR_AW_CFG_BME		CSR_BIT(18)
+#define FBNIC_PUL_OB_TLP_HDR_AR_CFG	0x3103e		/* 0xc40f8 */
+#define FBNIC_PUL_OB_TLP_HDR_AR_CFG_BME		CSR_BIT(18)
+#define FBNIC_CSR_END_PUL_USER	0x31080	/* CSR section delimiter */
+
+#define FBNIC_MAX_QUEUES		128
+
 #endif /* _FBNIC_CSR_H_ */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c b/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
index 91e8135410df..1441610f5843 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
@@ -65,6 +65,8 @@ struct fbnic_dev *fbnic_devlink_alloc(struct pci_dev *pdev)
 	fbd->uc_addr4 = iomap_table[4];
 
 	fbd->dsn = pci_get_dsn(pdev);
+	fbd->mps = pcie_get_mps(pdev);
+	fbd->readrq = pcie_get_readrq(pdev);
 
 	return fbd;
 }
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.c b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c
new file mode 100644
index 000000000000..a6ef898d7eed
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c
@@ -0,0 +1,425 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#include <linux/bitfield.h>
+#include <net/tcp.h>
+
+#include "fbnic.h"
+#include "fbnic_mac.h"
+
+static void fbnic_init_readrq(struct fbnic_dev *fbd, unsigned int offset,
+			      unsigned int cls, unsigned int readrq)
+{
+	u32 val = rd32(fbd, offset);
+
+	/* The TDF_CTL masks are a superset of the RNI_RBP ones. So we can
+	 * use them when setting either the TDE_CTF or RNI_RBP registers.
+	 */
+	val &= FBNIC_QM_TNI_TDF_CTL_MAX_OT | FBNIC_QM_TNI_TDF_CTL_MAX_OB;
+
+	val |= FIELD_PREP(FBNIC_QM_TNI_TDF_CTL_MRRS, readrq) |
+	       FIELD_PREP(FBNIC_QM_TNI_TDF_CTL_CLS, cls);
+
+	wr32(fbd, offset, val);
+}
+
+static void fbnic_init_mps(struct fbnic_dev *fbd, unsigned int offset,
+			   unsigned int cls, unsigned int mps)
+{
+	u32 val = rd32(fbd, offset);
+
+	/* Currently all MPS masks are identical so just use the first one */
+	val &= ~(FBNIC_QM_TNI_TCM_CTL_MPS | FBNIC_QM_TNI_TCM_CTL_CLS);
+
+	val |= FIELD_PREP(FBNIC_QM_TNI_TCM_CTL_MPS, mps) |
+	       FIELD_PREP(FBNIC_QM_TNI_TCM_CTL_CLS, cls);
+
+	wr32(fbd, offset, val);
+}
+
+static void fbnic_mac_init_axi(struct fbnic_dev *fbd)
+{
+	bool override_1k = false;
+	int readrq, mps, cls;
+
+	/* All of the values are based on being a power of 2 starting
+	 * with 64 == 0. Therefore we can either divide by 64 in the
+	 * case of constants, or just subtract 6 from the log2 of the value
+	 * in order to get the value we will be programming into the
+	 * registers.
+	 */
+	readrq = ilog2(fbd->readrq) - 6;
+	if (readrq > 3)
+		override_1k = true;
+	readrq = clamp(readrq, 0, 3);
+
+	mps = ilog2(fbd->mps) - 6;
+	mps = clamp(mps, 0, 3);
+
+	cls = ilog2(L1_CACHE_BYTES) - 6;
+	cls = clamp(cls, 0, 3);
+
+	/* Configure Tx/Rx AXI Paths w/ Read Request and Max Payload sizes */
+	fbnic_init_readrq(fbd, FBNIC_QM_TNI_TDF_CTL, cls, readrq);
+	fbnic_init_mps(fbd, FBNIC_QM_TNI_TCM_CTL, cls, mps);
+
+	/* Configure QM TNI TDE:
+	 * - Max outstanding AXI beats to 704(768 - 64) - guaranetees 8% of
+	 *   buffer capacity to descriptors.
+	 * - Max outstanding transactions to 128
+	 */
+	wr32(fbd, FBNIC_QM_TNI_TDE_CTL,
+	     FIELD_PREP(FBNIC_QM_TNI_TDE_CTL_MRRS_1K, override_1k ? 1 : 0) |
+	     FIELD_PREP(FBNIC_QM_TNI_TDE_CTL_MAX_OB, 704) |
+	     FIELD_PREP(FBNIC_QM_TNI_TDE_CTL_MAX_OT, 128) |
+	     FIELD_PREP(FBNIC_QM_TNI_TDE_CTL_MRRS, readrq) |
+	     FIELD_PREP(FBNIC_QM_TNI_TDE_CTL_CLS, cls));
+
+	fbnic_init_readrq(fbd, FBNIC_QM_RNI_RBP_CTL, cls, readrq);
+	fbnic_init_mps(fbd, FBNIC_QM_RNI_RDE_CTL, cls, mps);
+	fbnic_init_mps(fbd, FBNIC_QM_RNI_RCM_CTL, cls, mps);
+
+	/* Enable XALI AR/AW outbound */
+	wr32(fbd, FBNIC_PUL_OB_TLP_HDR_AW_CFG,
+	     FBNIC_PUL_OB_TLP_HDR_AW_CFG_BME);
+	wr32(fbd, FBNIC_PUL_OB_TLP_HDR_AR_CFG,
+	     FBNIC_PUL_OB_TLP_HDR_AR_CFG_BME);
+}
+
+static void fbnic_mac_init_qm(struct fbnic_dev *fbd)
+{
+	u32 clock_freq;
+
+	/* Configure TSO behavior */
+	wr32(fbd, FBNIC_QM_TQS_CTL0,
+	     FIELD_PREP(FBNIC_QM_TQS_CTL0_LSO_TS_MASK,
+			FBNIC_QM_TQS_CTL0_LSO_TS_LAST) |
+	     FIELD_PREP(FBNIC_QM_TQS_CTL0_PREFETCH_THRESH,
+			FBNIC_QM_TQS_CTL0_PREFETCH_THRESH_MIN));
+
+	/* Limit EDT to INT_MAX as this is the limit of the EDT Qdisc */
+	wr32(fbd, FBNIC_QM_TQS_EDT_TS_RANGE, INT_MAX);
+
+	/* Configure MTU
+	 * Due to known HW issue we cannot set the MTU to within 16 octets
+	 * of a 64 octet aligned boundary. So we will set the TQS_MTU(s) to
+	 * MTU + 1.
+	 */
+	wr32(fbd, FBNIC_QM_TQS_MTU_CTL0, FBNIC_MAX_JUMBO_FRAME_SIZE + 1);
+	wr32(fbd, FBNIC_QM_TQS_MTU_CTL1,
+	     FIELD_PREP(FBNIC_QM_TQS_MTU_CTL1_BULK,
+			FBNIC_MAX_JUMBO_FRAME_SIZE + 1));
+
+	clock_freq = FBNIC_CLOCK_FREQ;
+
+	/* Be aggressive on the timings. We will have the interrupt
+	 * threshold timer tick once every 1 usec and coalesce writes for
+	 * up to 80 usecs.
+	 */
+	wr32(fbd, FBNIC_QM_TCQ_CTL0,
+	     FIELD_PREP(FBNIC_QM_TCQ_CTL0_TICK_CYCLES,
+			clock_freq / 1000000) |
+	     FIELD_PREP(FBNIC_QM_TCQ_CTL0_COAL_WAIT,
+			clock_freq / 12500));
+
+	/* We will have the interrupt threshold timer tick once every
+	 * 1 usec and coalesce writes for up to 2 usecs.
+	 */
+	wr32(fbd, FBNIC_QM_RCQ_CTL0,
+	     FIELD_PREP(FBNIC_QM_RCQ_CTL0_TICK_CYCLES,
+			clock_freq / 1000000) |
+	     FIELD_PREP(FBNIC_QM_RCQ_CTL0_COAL_WAIT,
+			clock_freq / 500000));
+
+	/* Configure spacer control to 64 beats. */
+	wr32(fbd, FBNIC_FAB_AXI4_AR_SPACER_2_CFG,
+	     FBNIC_FAB_AXI4_AR_SPACER_MASK |
+	     FIELD_PREP(FBNIC_FAB_AXI4_AR_SPACER_THREADSHOLD, 2));
+}
+
+#define FBNIC_DROP_EN_MASK	0x7d
+#define FBNIC_PAUSE_EN_MASK	0x14
+#define FBNIC_ECN_EN_MASK	0x10
+
+struct fbnic_fifo_config {
+	unsigned int addr;
+	unsigned int size;
+};
+
+/* Rx FIFO Configuration
+ * The table consists of 8 entries, of which only 4 are currently used
+ * The starting addr is in units of 64B and the size is in 2KB units
+ * Below is the human readable version of the table defined below:
+ * Function		Addr	Size
+ * ----------------------------------
+ * Network to Host/BMC	384K	64K
+ * Unused
+ * Unused
+ * Network to BMC	448K	32K
+ * Network to Host	0	384K
+ * Unused
+ * BMC to Host		480K	32K
+ * Unused
+ */
+static const struct fbnic_fifo_config fifo_config[] = {
+	{ .addr = 0x1800, .size = 0x20 },	/* Network to Host/BMC */
+	{ },					/* Unused */
+	{ },					/* Unused */
+	{ .addr = 0x1c00, .size = 0x10 },	/* Network to BMC */
+	{ .addr = 0x0000, .size = 0xc0 },	/* Network to Host */
+	{ },					/* Unused */
+	{ .addr = 0x1e00, .size = 0x10 },	/* BMC to Host */
+	{ }					/* Unused */
+};
+
+static void fbnic_mac_init_rxb(struct fbnic_dev *fbd)
+{
+	bool rx_enable;
+	int i;
+
+	rx_enable = !!(rd32(fbd, FBNIC_RPC_RMI_CONFIG) &
+		       FBNIC_RPC_RMI_CONFIG_ENABLE);
+
+	for (i = 0; i < 8; i++) {
+		unsigned int size = fifo_config[i].size;
+
+		/* If we are coming up on a system that already has the
+		 * Rx data path enabled we don't need to reconfigure the
+		 * FIFOs. Instead we can check to verify the values are
+		 * large enough to meet our needs, and use the values to
+		 * populate the flow control, ECN, and drop thresholds.
+		 */
+		if (rx_enable) {
+			size = FIELD_GET(FBNIC_RXB_PBUF_SIZE,
+					 rd32(fbd, FBNIC_RXB_PBUF_CFG(i)));
+			if (size < fifo_config[i].size)
+				dev_warn(fbd->dev,
+					 "fifo%d size of %d smaller than expected value of %d\n",
+					 i, size << 11,
+					 fifo_config[i].size << 11);
+		} else {
+			/* Program RXB Cuthrough */
+			wr32(fbd, FBNIC_RXB_CT_SIZE(i),
+			     FIELD_PREP(FBNIC_RXB_CT_SIZE_HEADER, 4) |
+			     FIELD_PREP(FBNIC_RXB_CT_SIZE_PAYLOAD, 2));
+
+			/* The granularity for the packet buffer size is 2KB
+			 * granularity while the packet buffer base address is
+			 * only 64B granularity
+			 */
+			wr32(fbd, FBNIC_RXB_PBUF_CFG(i),
+			     FIELD_PREP(FBNIC_RXB_PBUF_BASE_ADDR,
+					fifo_config[i].addr) |
+			     FIELD_PREP(FBNIC_RXB_PBUF_SIZE, size));
+
+			/* The granularity for the credits is 64B. This is
+			 * based on RXB_PBUF_SIZE * 32 + 4.
+			 */
+			wr32(fbd, FBNIC_RXB_PBUF_CREDIT(i),
+			     FIELD_PREP(FBNIC_RXB_PBUF_CREDIT_MASK,
+					size ? size * 32 + 4 : 0));
+		}
+
+		if (!size)
+			continue;
+
+		/* Pause is size of FIFO with 56KB skid to start/stop */
+		wr32(fbd, FBNIC_RXB_PAUSE_THLD(i),
+		     !(FBNIC_PAUSE_EN_MASK & (1u << i)) ? 0x1fff :
+		     FIELD_PREP(FBNIC_RXB_PAUSE_THLD_ON,
+				size * 32 - 0x380) |
+		     FIELD_PREP(FBNIC_RXB_PAUSE_THLD_OFF, 0x380));
+
+		/* Enable Drop when only one packet is left in the FIFO */
+		wr32(fbd, FBNIC_RXB_DROP_THLD(i),
+		     !(FBNIC_DROP_EN_MASK & (1u << i)) ? 0x1fff :
+		     FIELD_PREP(FBNIC_RXB_DROP_THLD_ON,
+				size * 32 -
+				FBNIC_MAX_JUMBO_FRAME_SIZE / 64) |
+		     FIELD_PREP(FBNIC_RXB_DROP_THLD_OFF,
+				size * 32 -
+				FBNIC_MAX_JUMBO_FRAME_SIZE / 64));
+
+		/* Enable ECN bit when 1/4 of RXB is filled with at least
+		 * 1 room for one full jumbo frame before setting ECN
+		 */
+		wr32(fbd, FBNIC_RXB_ECN_THLD(i),
+		     !(FBNIC_ECN_EN_MASK & (1u << i)) ? 0x1fff :
+		     FIELD_PREP(FBNIC_RXB_ECN_THLD_ON,
+				max_t(unsigned int,
+				      size * 32 / 4,
+				      FBNIC_MAX_JUMBO_FRAME_SIZE / 64)) |
+		     FIELD_PREP(FBNIC_RXB_ECN_THLD_OFF,
+				max_t(unsigned int,
+				      size * 32 / 4,
+				      FBNIC_MAX_JUMBO_FRAME_SIZE / 64)));
+	}
+
+	/* For now only enable drop and ECN. We need to add driver/kernel
+	 * interfaces for configuring pause.
+	 */
+	wr32(fbd, FBNIC_RXB_PAUSE_DROP_CTRL,
+	     FIELD_PREP(FBNIC_RXB_PAUSE_DROP_CTRL_DROP_ENABLE,
+			FBNIC_DROP_EN_MASK) |
+	     FIELD_PREP(FBNIC_RXB_PAUSE_DROP_CTRL_ECN_ENABLE,
+			FBNIC_ECN_EN_MASK));
+
+	/* Program INTF credits */
+	wr32(fbd, FBNIC_RXB_INTF_CREDIT,
+	     FBNIC_RXB_INTF_CREDIT_MASK0 |
+	     FBNIC_RXB_INTF_CREDIT_MASK1 |
+	     FBNIC_RXB_INTF_CREDIT_MASK2 |
+	     FIELD_PREP(FBNIC_RXB_INTF_CREDIT_MASK3, 8));
+
+	/* Configure calendar slots.
+	 * Rx: 0 - 62	RDE 1st, BMC 2nd
+	 *     63	BMC 1st, RDE 2nd
+	 */
+	for (i = 0; i < 16; i++) {
+		u32 calendar_val = (i == 15) ? 0x1e1b1b1b : 0x1b1b1b1b;
+
+		wr32(fbd, FBNIC_RXB_CLDR_PRIO_CFG(i), calendar_val);
+	}
+
+	/* Split the credits for the DRR up as follows:
+	 * Quantum0: 8000	Network to Host
+	 * Quantum1: 0		Not used
+	 * Quantum2: 80		BMC to Host
+	 * Quantum3: 0		Not used
+	 * Quantum4: 8000	Multicast to Host and BMC
+	 */
+	wr32(fbd, FBNIC_RXB_DWRR_RDE_WEIGHT0,
+	     FIELD_PREP(FBNIC_RXB_DWRR_RDE_WEIGHT0_QUANTUM0, 0x40) |
+	     FIELD_PREP(FBNIC_RXB_DWRR_RDE_WEIGHT0_QUANTUM2, 0x50));
+	wr32(fbd, FBNIC_RXB_DWRR_RDE_WEIGHT0_EXT,
+	     FIELD_PREP(FBNIC_RXB_DWRR_RDE_WEIGHT0_QUANTUM0, 0x1f));
+	wr32(fbd, FBNIC_RXB_DWRR_RDE_WEIGHT1,
+	     FIELD_PREP(FBNIC_RXB_DWRR_RDE_WEIGHT1_QUANTUM4, 0x40));
+	wr32(fbd, FBNIC_RXB_DWRR_RDE_WEIGHT1_EXT,
+	     FIELD_PREP(FBNIC_RXB_DWRR_RDE_WEIGHT1_QUANTUM4, 0x1f));
+
+	/* Program RXB FCS Endian register */
+	wr32(fbd, FBNIC_RXB_ENDIAN_FCS, 0x0aaaaaa0);
+}
+
+static void fbnic_mac_init_txb(struct fbnic_dev *fbd)
+{
+	int i;
+
+	wr32(fbd, FBNIC_TCE_TXB_CTRL, 0);
+
+	/* Configure Tx QM Credits */
+	wr32(fbd, FBNIC_QM_TQS_CTL1,
+	     FIELD_PREP(FBNIC_QM_TQS_CTL1_MC_MAX_CREDITS, 0x40) |
+	     FIELD_PREP(FBNIC_QM_TQS_CTL1_BULK_MAX_CREDITS, 0x20));
+
+	/* Initialize internal Tx queues */
+	wr32(fbd, FBNIC_TCE_TXB_TEI_Q0_CTRL, 0);
+	wr32(fbd, FBNIC_TCE_TXB_TEI_Q1_CTRL, 0);
+	wr32(fbd, FBNIC_TCE_TXB_MC_Q_CTRL,
+	     FIELD_PREP(FBNIC_TCE_TXB_Q_CTRL_SIZE, 0x400) |
+	     FIELD_PREP(FBNIC_TCE_TXB_Q_CTRL_START, 0x000));
+	wr32(fbd, FBNIC_TCE_TXB_RX_TEI_Q_CTRL, 0);
+	wr32(fbd, FBNIC_TCE_TXB_TX_BMC_Q_CTRL,
+	     FIELD_PREP(FBNIC_TCE_TXB_Q_CTRL_SIZE, 0x200) |
+	     FIELD_PREP(FBNIC_TCE_TXB_Q_CTRL_START, 0x400));
+	wr32(fbd, FBNIC_TCE_TXB_RX_BMC_Q_CTRL,
+	     FIELD_PREP(FBNIC_TCE_TXB_Q_CTRL_SIZE, 0x200) |
+	     FIELD_PREP(FBNIC_TCE_TXB_Q_CTRL_START, 0x600));
+
+	wr32(fbd, FBNIC_TCE_LSO_CTRL,
+	     FBNIC_TCE_LSO_CTRL_IPID_MODE_INC |
+	     FIELD_PREP(FBNIC_TCE_LSO_CTRL_TCPF_CLR_1ST, TCPHDR_PSH |
+							 TCPHDR_FIN) |
+	     FIELD_PREP(FBNIC_TCE_LSO_CTRL_TCPF_CLR_MID, TCPHDR_PSH |
+							 TCPHDR_CWR |
+							 TCPHDR_FIN) |
+	     FIELD_PREP(FBNIC_TCE_LSO_CTRL_TCPF_CLR_END, TCPHDR_CWR));
+	wr32(fbd, FBNIC_TCE_CSO_CTRL, 0);
+
+	wr32(fbd, FBNIC_TCE_BMC_MAX_PKTSZ,
+	     FIELD_PREP(FBNIC_TCE_BMC_MAX_PKTSZ_TX,
+			FBNIC_MAX_JUMBO_FRAME_SIZE) |
+	     FIELD_PREP(FBNIC_TCE_BMC_MAX_PKTSZ_RX,
+			FBNIC_MAX_JUMBO_FRAME_SIZE));
+	wr32(fbd, FBNIC_TCE_MC_MAX_PKTSZ,
+	     FIELD_PREP(FBNIC_TCE_MC_MAX_PKTSZ_TMI,
+			FBNIC_MAX_JUMBO_FRAME_SIZE));
+
+	/* Configure calendar slots.
+	 * Tx: 0 - 62	TMI 1st, BMC 2nd
+	 *     63	BMC 1st, TMI 2nd
+	 */
+	for (i = 0; i < 16; i++) {
+		u32 calendar_val = (i == 15) ? 0x1e1b1b1b : 0x1b1b1b1b;
+
+		wr32(fbd, FBNIC_TCE_TXB_CLDR_SLOT_CFG(i), calendar_val);
+	}
+
+	/* Configure DWRR */
+	wr32(fbd, FBNIC_TCE_TXB_ENQ_WRR_CTRL,
+	     FIELD_PREP(FBNIC_TCE_TXB_ENQ_WRR_CTRL_WEIGHT0, 0x64) |
+	     FIELD_PREP(FBNIC_TCE_TXB_ENQ_WRR_CTRL_WEIGHT2, 0x04));
+	wr32(fbd, FBNIC_TCE_TXB_TEI_DWRR_CTRL, 0);
+	wr32(fbd, FBNIC_TCE_TXB_TEI_DWRR_CTRL_EXT, 0);
+	wr32(fbd, FBNIC_TCE_TXB_BMC_DWRR_CTRL,
+	     FIELD_PREP(FBNIC_TCE_TXB_BMC_DWRR_CTRL_QUANTUM0, 0x50) |
+	     FIELD_PREP(FBNIC_TCE_TXB_BMC_DWRR_CTRL_QUANTUM1, 0x82));
+	wr32(fbd, FBNIC_TCE_TXB_BMC_DWRR_CTRL_EXT, 0);
+	wr32(fbd, FBNIC_TCE_TXB_NTWRK_DWRR_CTRL,
+	     FIELD_PREP(FBNIC_TCE_TXB_NTWRK_DWRR_CTRL_QUANTUM1, 0x50) |
+	     FIELD_PREP(FBNIC_TCE_TXB_NTWRK_DWRR_CTRL_QUANTUM2, 0x20));
+	wr32(fbd, FBNIC_TCE_TXB_NTWRK_DWRR_CTRL_EXT,
+	     FIELD_PREP(FBNIC_TCE_TXB_NTWRK_DWRR_CTRL_QUANTUM2, 0x03));
+
+	/* Configure SOP protocol protection */
+	wr32(fbd, FBNIC_TCE_SOP_PROT_CTRL,
+	     FIELD_PREP(FBNIC_TCE_SOP_PROT_CTRL_TBI, 0x78) |
+	     FIELD_PREP(FBNIC_TCE_SOP_PROT_CTRL_TTI_FRM, 0x40) |
+	     FIELD_PREP(FBNIC_TCE_SOP_PROT_CTRL_TTI_CM, 0x0c));
+
+	/* Conservative configuration on MAC interface Start of Packet
+	 * protection FIFO. This sets the minimum depth of the FIFO before
+	 * we start sending packets to the MAC measured in 64B units and
+	 * up to 160 entries deep.
+	 *
+	 * For the ASIC the clock is fast enough that we will likely fill
+	 * the SOP FIFO before the MAC can drain it. So just use a minimum
+	 * value of 8.
+	 */
+	wr32(fbd, FBNIC_TMI_SOP_PROT_CTRL, 8);
+
+	wrfl(fbd);
+	wr32(fbd, FBNIC_TCE_TXB_CTRL, FBNIC_TCE_TXB_CTRL_TCAM_ENABLE |
+				      FBNIC_TCE_TXB_CTRL_LOAD);
+}
+
+static void fbnic_mac_init_regs(struct fbnic_dev *fbd)
+{
+	fbnic_mac_init_axi(fbd);
+	fbnic_mac_init_qm(fbd);
+	fbnic_mac_init_rxb(fbd);
+	fbnic_mac_init_txb(fbd);
+}
+
+static const struct fbnic_mac fbnic_mac_asic = {
+	.init_regs = fbnic_mac_init_regs,
+};
+
+/**
+ * fbnic_mac_init - Assign a MAC type and initialize the fbnic device
+ * @fbd: Device pointer to device to initialize
+ *
+ * Return: zero on success, negative on failure
+ *
+ * Initialize the MAC function pointers and initializes the MAC of
+ * the device.
+ **/
+int fbnic_mac_init(struct fbnic_dev *fbd)
+{
+	fbd->mac = &fbnic_mac_asic;
+
+	fbd->mac->init_regs(fbd);
+
+	return 0;
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.h b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h
new file mode 100644
index 000000000000..e78a92338a62
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h
@@ -0,0 +1,25 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#ifndef _FBNIC_MAC_H_
+#define _FBNIC_MAC_H_
+
+#include <linux/types.h>
+
+struct fbnic_dev;
+
+#define FBNIC_MAX_JUMBO_FRAME_SIZE	9742
+
+/* This structure defines the interface hooks for the MAC. The MAC hooks
+ * will be configured as a const struct provided with a set of function
+ * pointers.
+ *
+ * void (*init_regs)(struct fbnic_dev *fbd);
+ *	Initialize MAC registers to enable Tx/Rx paths and FIFOs.
+ */
+struct fbnic_mac {
+	void (*init_regs)(struct fbnic_dev *fbd);
+};
+
+int fbnic_mac_init(struct fbnic_dev *fbd);
+#endif /* _FBNIC_MAC_H_ */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
index fffe8a7cc96e..c03d3945d6d2 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -29,6 +29,35 @@ static const struct pci_device_id fbnic_pci_tbl[] = {
 };
 MODULE_DEVICE_TABLE(pci, fbnic_pci_tbl);
 
+u32 fbnic_rd32(struct fbnic_dev *fbd, u32 reg)
+{
+	u32 __iomem *csr = READ_ONCE(fbd->uc_addr0);
+	u32 value;
+
+	if (!csr)
+		return ~0U;
+
+	value = readl(csr + reg);
+
+	/* If any bits are 0 value should be valid */
+	if (~value)
+		return value;
+
+	/* All 1's may be valid if ZEROs register still works */
+	if (reg != FBNIC_MASTER_SPARE_0 && ~readl(csr + FBNIC_MASTER_SPARE_0))
+		return value;
+
+	/* Hardware is giving us all 1's reads, assume it is gone */
+	WRITE_ONCE(fbd->uc_addr0, NULL);
+	WRITE_ONCE(fbd->uc_addr4, NULL);
+
+	dev_err(fbd->dev,
+		"Failed read (idx 0x%x AKA addr 0x%x), disabled CSR access, awaiting reset\n",
+		reg, reg << 2);
+
+	return ~0U;
+}
+
 /**
  * fbnic_probe - Device Initialization Routine
  * @pdev: PCI device information struct
@@ -86,6 +115,12 @@ static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (err)
 		goto free_fbd;
 
+	err = fbnic_mac_init(fbd);
+	if (err) {
+		dev_err(&pdev->dev, "Failed to initialize MAC: %d\n", err);
+		goto free_irqs;
+	}
+
 	fbnic_devlink_register(fbd);
 
 	if (!fbd->dsn) {
@@ -101,6 +136,8 @@ init_failure_mode:
 	  * firmware updates for fixes.
 	  */
 	return 0;
+free_irqs:
+	fbnic_free_irqs(fbd);
 free_fbd:
 	pci_disable_device(pdev);
 	fbnic_devlink_free(fbd);
@@ -157,6 +194,8 @@ static int __fbnic_pm_resume(struct device *dev)
 	if (err)
 		goto err_invalidate_uc_addr;
 
+	fbd->mac->init_regs(fbd);
+
 	return 0;
 err_invalidate_uc_addr:
 	WRITE_ONCE(fbd->uc_addr0, NULL);
-- 
cgit v1.2.3-58-ga151


From c6203e678cc9a5bf01ec7ae382851f504870777f Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:49:27 -0700
Subject: eth: fbnic: Add message parsing for FW messages

Add FW message formatting and parsing. The TLV format should
look very familiar to those familiar with netlink.
Since we don't have to deal with backward compatibility
we tweaked the format a little to make it easier to deal
with, and more appropriate for tightly coupled interfaces
like driver<>FW communication.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079936754.1778861.1029830244010564007.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/Makefile    |   3 +-
 drivers/net/ethernet/meta/fbnic/fbnic_tlv.c | 529 ++++++++++++++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_tlv.h | 175 +++++++++
 3 files changed, 706 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_tlv.c
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_tlv.h

diff --git a/drivers/net/ethernet/meta/fbnic/Makefile b/drivers/net/ethernet/meta/fbnic/Makefile
index b8f4511440dc..0434ee0b3069 100644
--- a/drivers/net/ethernet/meta/fbnic/Makefile
+++ b/drivers/net/ethernet/meta/fbnic/Makefile
@@ -10,4 +10,5 @@ obj-$(CONFIG_FBNIC) += fbnic.o
 fbnic-y := fbnic_devlink.o \
 	   fbnic_irq.o \
 	   fbnic_mac.o \
-	   fbnic_pci.o
+	   fbnic_pci.o \
+	   fbnic_tlv.o
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_tlv.c b/drivers/net/ethernet/meta/fbnic/fbnic_tlv.c
new file mode 100644
index 000000000000..2a174ab062a3
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_tlv.c
@@ -0,0 +1,529 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#include <linux/gfp.h>
+#include <linux/mm.h>
+#include <linux/once.h>
+#include <linux/random.h>
+#include <linux/string.h>
+#include <uapi/linux/if_ether.h>
+
+#include "fbnic_tlv.h"
+
+/**
+ * fbnic_tlv_msg_alloc - Allocate page and initialize FW message header
+ * @msg_id: Identifier for new message we are starting
+ *
+ * Return: pointer to start of message, or NULL on failure.
+ *
+ * Allocates a page and initializes message header at start of page.
+ * Initial message size is 1 DWORD which is just the header.
+ **/
+struct fbnic_tlv_msg *fbnic_tlv_msg_alloc(u16 msg_id)
+{
+	struct fbnic_tlv_hdr hdr = { 0 };
+	struct fbnic_tlv_msg *msg;
+
+	msg = (struct fbnic_tlv_msg *)__get_free_page(GFP_KERNEL);
+	if (!msg)
+		return NULL;
+
+	/* Start with zero filled header and then back fill with data */
+	hdr.type = msg_id;
+	hdr.is_msg = 1;
+	hdr.len = cpu_to_le16(1);
+
+	/* Copy header into start of message */
+	msg->hdr = hdr;
+
+	return msg;
+}
+
+/**
+ * fbnic_tlv_attr_put_flag - Add flag value to message
+ * @msg: Message header we are adding flag attribute to
+ * @attr_id: ID of flag attribute we are adding to message
+ *
+ * Return: -ENOSPC if there is no room for the attribute. Otherwise 0.
+ *
+ * Adds a 1 DWORD flag attribute to the message. The presence of this
+ * attribute can be used as a boolean value indicating true, otherwise the
+ * value is considered false.
+ **/
+int fbnic_tlv_attr_put_flag(struct fbnic_tlv_msg *msg, const u16 attr_id)
+{
+	int attr_max_len = PAGE_SIZE - offset_in_page(msg) - sizeof(*msg);
+	struct fbnic_tlv_hdr hdr = { 0 };
+	struct fbnic_tlv_msg *attr;
+
+	attr_max_len -= le16_to_cpu(msg->hdr.len) * sizeof(u32);
+	if (attr_max_len < sizeof(*attr))
+		return -ENOSPC;
+
+	/* Get header pointer and bump attr to start of data */
+	attr = &msg[le16_to_cpu(msg->hdr.len)];
+
+	/* Record attribute type and size */
+	hdr.type = attr_id;
+	hdr.len = cpu_to_le16(sizeof(hdr));
+
+	attr->hdr = hdr;
+	le16_add_cpu(&msg->hdr.len,
+		     FBNIC_TLV_MSG_SIZE(le16_to_cpu(hdr.len)));
+
+	return 0;
+}
+
+/**
+ * fbnic_tlv_attr_put_value - Add data to message
+ * @msg: Message header we are adding flag attribute to
+ * @attr_id: ID of flag attribute we are adding to message
+ * @value: Pointer to data to be stored
+ * @len: Size of data to be stored.
+ *
+ * Return: -ENOSPC if there is no room for the attribute. Otherwise 0.
+ *
+ * Adds header and copies data pointed to by value into the message. The
+ * result is rounded up to the nearest DWORD for sizing so that the
+ * headers remain aligned.
+ *
+ * The assumption is that the value field is in a format where byte
+ * ordering can be guaranteed such as a byte array or a little endian
+ * format.
+ **/
+int fbnic_tlv_attr_put_value(struct fbnic_tlv_msg *msg, const u16 attr_id,
+			     const void *value, const int len)
+{
+	int attr_max_len = PAGE_SIZE - offset_in_page(msg) - sizeof(*msg);
+	struct fbnic_tlv_hdr hdr = { 0 };
+	struct fbnic_tlv_msg *attr;
+
+	attr_max_len -= le16_to_cpu(msg->hdr.len) * sizeof(u32);
+	if (attr_max_len < sizeof(*attr) + len)
+		return -ENOSPC;
+
+	/* Get header pointer and bump attr to start of data */
+	attr = &msg[le16_to_cpu(msg->hdr.len)];
+
+	/* Record attribute type and size */
+	hdr.type = attr_id;
+	hdr.len = cpu_to_le16(sizeof(hdr) + len);
+
+	/* Zero pad end of region to be written if we aren't aligned */
+	if (len % sizeof(hdr))
+		attr->value[len / sizeof(hdr)] = 0;
+
+	/* Copy data over */
+	memcpy(attr->value, value, len);
+
+	attr->hdr = hdr;
+	le16_add_cpu(&msg->hdr.len,
+		     FBNIC_TLV_MSG_SIZE(le16_to_cpu(hdr.len)));
+
+	return 0;
+}
+
+/**
+ * __fbnic_tlv_attr_put_int - Add integer to message
+ * @msg: Message header we are adding flag attribute to
+ * @attr_id: ID of flag attribute we are adding to message
+ * @value: Data to be stored
+ * @len: Size of data to be stored, either 4 or 8 bytes.
+ *
+ * Return: -ENOSPC if there is no room for the attribute. Otherwise 0.
+ *
+ * Adds header and copies data pointed to by value into the message. Will
+ * format the data as little endian.
+ **/
+int __fbnic_tlv_attr_put_int(struct fbnic_tlv_msg *msg, const u16 attr_id,
+			     s64 value, const int len)
+{
+	__le64 le64_value = cpu_to_le64(value);
+
+	return fbnic_tlv_attr_put_value(msg, attr_id, &le64_value, len);
+}
+
+/**
+ * fbnic_tlv_attr_put_mac_addr - Add mac_addr to message
+ * @msg: Message header we are adding flag attribute to
+ * @attr_id: ID of flag attribute we are adding to message
+ * @mac_addr: Byte pointer to MAC address to be stored
+ *
+ * Return: -ENOSPC if there is no room for the attribute. Otherwise 0.
+ *
+ * Adds header and copies data pointed to by mac_addr into the message. Will
+ * copy the address raw so it will be in big endian with start of MAC
+ * address at start of attribute.
+ **/
+int fbnic_tlv_attr_put_mac_addr(struct fbnic_tlv_msg *msg, const u16 attr_id,
+				const u8 *mac_addr)
+{
+	return fbnic_tlv_attr_put_value(msg, attr_id, mac_addr, ETH_ALEN);
+}
+
+/**
+ * fbnic_tlv_attr_put_string - Add string to message
+ * @msg: Message header we are adding flag attribute to
+ * @attr_id: ID of flag attribute we are adding to message
+ * @string: Byte pointer to null terminated string to be stored
+ *
+ * Return: -ENOSPC if there is no room for the attribute. Otherwise 0.
+ *
+ * Adds header and copies data pointed to by string into the message. Will
+ * copy the address raw so it will be in byte order.
+ **/
+int fbnic_tlv_attr_put_string(struct fbnic_tlv_msg *msg, u16 attr_id,
+			      const char *string)
+{
+	int attr_max_len = PAGE_SIZE - sizeof(*msg);
+	int str_len = 1;
+
+	/* The max length will be message minus existing message and new
+	 * attribute header. Since the message is measured in DWORDs we have
+	 * to multiply the size by 4.
+	 *
+	 * The string length doesn't include the \0 so we have to add one to
+	 * the final value, so start with that as our initial value.
+	 *
+	 * We will verify if the string will fit in fbnic_tlv_attr_put_value()
+	 */
+	attr_max_len -= le16_to_cpu(msg->hdr.len) * sizeof(u32);
+	str_len += strnlen(string, attr_max_len);
+
+	return fbnic_tlv_attr_put_value(msg, attr_id, string, str_len);
+}
+
+/**
+ * fbnic_tlv_attr_get_unsigned - Retrieve unsigned value from result
+ * @attr: Attribute to retrieve data from
+ *
+ * Return: unsigned 64b value containing integer value
+ **/
+u64 fbnic_tlv_attr_get_unsigned(struct fbnic_tlv_msg *attr)
+{
+	__le64 le64_value = 0;
+
+	memcpy(&le64_value, &attr->value[0],
+	       le16_to_cpu(attr->hdr.len) - sizeof(*attr));
+
+	return le64_to_cpu(le64_value);
+}
+
+/**
+ * fbnic_tlv_attr_get_signed - Retrieve signed value from result
+ * @attr: Attribute to retrieve data from
+ *
+ * Return: signed 64b value containing integer value
+ **/
+s64 fbnic_tlv_attr_get_signed(struct fbnic_tlv_msg *attr)
+{
+	int shift = (8 + sizeof(*attr) - le16_to_cpu(attr->hdr.len)) * 8;
+	__le64 le64_value = 0;
+	s64 value;
+
+	/* Copy the value and adjust for byte ordering */
+	memcpy(&le64_value, &attr->value[0],
+	       le16_to_cpu(attr->hdr.len) - sizeof(*attr));
+	value = le64_to_cpu(le64_value);
+
+	/* Sign extend the return value by using a pair of shifts */
+	return (value << shift) >> shift;
+}
+
+/**
+ * fbnic_tlv_attr_get_string - Retrieve string value from result
+ * @attr: Attribute to retrieve data from
+ * @str: Pointer to an allocated string to store the data
+ * @max_size: The maximum size which can be in str
+ *
+ * Return: the size of the string read from firmware
+ **/
+size_t fbnic_tlv_attr_get_string(struct fbnic_tlv_msg *attr, char *str,
+				 size_t max_size)
+{
+	max_size = min_t(size_t, max_size,
+			 (le16_to_cpu(attr->hdr.len) * 4) - sizeof(*attr));
+	memcpy(str, &attr->value, max_size);
+
+	return max_size;
+}
+
+/**
+ * fbnic_tlv_attr_nest_start - Add nested attribute header to message
+ * @msg: Message header we are adding flag attribute to
+ * @attr_id: ID of flag attribute we are adding to message
+ *
+ * Return: NULL if there is no room for the attribute. Otherwise a pointer
+ * to the new attribute header.
+ *
+ * New header length is stored initially in DWORDs.
+ **/
+struct fbnic_tlv_msg *fbnic_tlv_attr_nest_start(struct fbnic_tlv_msg *msg,
+						u16 attr_id)
+{
+	int attr_max_len = PAGE_SIZE - offset_in_page(msg) - sizeof(*msg);
+	struct fbnic_tlv_msg *attr = &msg[le16_to_cpu(msg->hdr.len)];
+	struct fbnic_tlv_hdr hdr = { 0 };
+
+	/* Make sure we have space for at least the nest header plus one more */
+	attr_max_len -= le16_to_cpu(msg->hdr.len) * sizeof(u32);
+	if (attr_max_len < sizeof(*attr) * 2)
+		return NULL;
+
+	/* Record attribute type and size */
+	hdr.type = attr_id;
+
+	/* Add current message length to account for consumption within the
+	 * page and leave it as a multiple of DWORDs, we will shift to
+	 * bytes when we close it out.
+	 */
+	hdr.len = cpu_to_le16(1);
+
+	attr->hdr = hdr;
+
+	return attr;
+}
+
+/**
+ * fbnic_tlv_attr_nest_stop - Close out nested attribute and add it to message
+ * @msg: Message header we are adding flag attribute to
+ *
+ * Closes out nested attribute, adds length to message, and then bumps
+ * length from DWORDs to bytes to match other attributes.
+ **/
+void fbnic_tlv_attr_nest_stop(struct fbnic_tlv_msg *msg)
+{
+	struct fbnic_tlv_msg *attr = &msg[le16_to_cpu(msg->hdr.len)];
+	u16 len = le16_to_cpu(attr->hdr.len);
+
+	/* Add attribute to message if there is more than just a header */
+	if (len <= 1)
+		return;
+
+	le16_add_cpu(&msg->hdr.len, len);
+
+	/* Convert from DWORDs to bytes */
+	attr->hdr.len = cpu_to_le16(len * sizeof(u32));
+}
+
+static int
+fbnic_tlv_attr_validate(struct fbnic_tlv_msg *attr,
+			const struct fbnic_tlv_index *tlv_index)
+{
+	u16 len = le16_to_cpu(attr->hdr.len) - sizeof(*attr);
+	u16 attr_id = attr->hdr.type;
+	__le32 *value = &attr->value[0];
+
+	if (attr->hdr.is_msg)
+		return -EINVAL;
+
+	if (attr_id >= FBNIC_TLV_RESULTS_MAX)
+		return -EINVAL;
+
+	while (tlv_index->id != attr_id) {
+		if  (tlv_index->id == FBNIC_TLV_ATTR_ID_UNKNOWN) {
+			if (attr->hdr.cannot_ignore)
+				return -ENOENT;
+			return le16_to_cpu(attr->hdr.len);
+		}
+
+		tlv_index++;
+	}
+
+	if (offset_in_page(attr) + len > PAGE_SIZE - sizeof(*attr))
+		return -E2BIG;
+
+	switch (tlv_index->type) {
+	case FBNIC_TLV_STRING:
+		if (!len || len > tlv_index->len)
+			return -EINVAL;
+		if (((char *)value)[len - 1])
+			return -EINVAL;
+		break;
+	case FBNIC_TLV_FLAG:
+		if (len)
+			return -EINVAL;
+		break;
+	case FBNIC_TLV_UNSIGNED:
+	case FBNIC_TLV_SIGNED:
+		if (tlv_index->len > sizeof(__le64))
+			return -EINVAL;
+		fallthrough;
+	case FBNIC_TLV_BINARY:
+		if (!len || len > tlv_index->len)
+			return -EINVAL;
+		break;
+	case FBNIC_TLV_NESTED:
+	case FBNIC_TLV_ARRAY:
+		if (len % 4)
+			return -EINVAL;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/**
+ * fbnic_tlv_attr_parse_array - Parse array of attributes into results array
+ * @attr: Start of attributes in the message
+ * @len: Length of attributes in the message
+ * @results: Array of pointers to store the results of parsing
+ * @tlv_index: List of TLV attributes to be parsed from message
+ * @tlv_attr_id: Specific ID that is repeated in array
+ * @array_len: Number of results to store in results array
+ *
+ * Return: zero on success, or negative value on error.
+ *
+ * Will take a list of attributes and a parser definition and will capture
+ * the results in the results array to have the data extracted later.
+ **/
+int fbnic_tlv_attr_parse_array(struct fbnic_tlv_msg *attr, int len,
+			       struct fbnic_tlv_msg **results,
+			       const struct fbnic_tlv_index *tlv_index,
+			       u16 tlv_attr_id, size_t array_len)
+{
+	int i = 0;
+
+	/* Initialize results table to NULL. */
+	memset(results, 0, array_len * sizeof(results[0]));
+
+	/* Nothing to parse if header was only thing there */
+	if (!len)
+		return 0;
+
+	/* Work through list of attributes, parsing them as necessary */
+	while (len > 0) {
+		u16 attr_id = attr->hdr.type;
+		u16 attr_len;
+		int err;
+
+		if (tlv_attr_id != attr_id)
+			return -EINVAL;
+
+		/* Stop parsing on full error */
+		err = fbnic_tlv_attr_validate(attr, tlv_index);
+		if (err < 0)
+			return err;
+
+		if (i >= array_len)
+			return -ENOSPC;
+
+		results[i++] = attr;
+
+		attr_len = FBNIC_TLV_MSG_SIZE(le16_to_cpu(attr->hdr.len));
+		len -= attr_len;
+		attr += attr_len;
+	}
+
+	return len == 0 ? 0 : -EINVAL;
+}
+
+/**
+ * fbnic_tlv_attr_parse - Parse attributes into a list of attribute results
+ * @attr: Start of attributes in the message
+ * @len: Length of attributes in the message
+ * @results: Array of pointers to store the results of parsing
+ * @tlv_index: List of TLV attributes to be parsed from message
+ *
+ * Return: zero on success, or negative value on error.
+ *
+ * Will take a list of attributes and a parser definition and will capture
+ * the results in the results array to have the data extracted later.
+ **/
+int fbnic_tlv_attr_parse(struct fbnic_tlv_msg *attr, int len,
+			 struct fbnic_tlv_msg **results,
+			 const struct fbnic_tlv_index *tlv_index)
+{
+	/* Initialize results table to NULL. */
+	memset(results, 0, sizeof(results[0]) * FBNIC_TLV_RESULTS_MAX);
+
+	/* Nothing to parse if header was only thing there */
+	if (!len)
+		return 0;
+
+	/* Work through list of attributes, parsing them as necessary */
+	while (len > 0) {
+		int err = fbnic_tlv_attr_validate(attr, tlv_index);
+		u16 attr_id = attr->hdr.type;
+		u16 attr_len;
+
+		/* Stop parsing on full error */
+		if (err < 0)
+			return err;
+
+		/* Ignore results for unsupported values */
+		if (!err) {
+			/* Do not overwrite existing entries */
+			if (results[attr_id])
+				return -EADDRINUSE;
+
+			results[attr_id] = attr;
+		}
+
+		attr_len = FBNIC_TLV_MSG_SIZE(le16_to_cpu(attr->hdr.len));
+		len -= attr_len;
+		attr += attr_len;
+	}
+
+	return len == 0 ? 0 : -EINVAL;
+}
+
+/**
+ * fbnic_tlv_msg_parse - Parse message and process via predetermined functions
+ * @opaque: Value passed to parser function to enable driver access
+ * @msg: Message to be parsed.
+ * @parser: TLV message parser definition.
+ *
+ * Return: zero on success, or negative value on error.
+ *
+ * Will take a message a number of message types via the attribute parsing
+ * definitions and function provided for the parser array.
+ **/
+int fbnic_tlv_msg_parse(void *opaque, struct fbnic_tlv_msg *msg,
+			const struct fbnic_tlv_parser *parser)
+{
+	struct fbnic_tlv_msg *results[FBNIC_TLV_RESULTS_MAX];
+	u16 msg_id = msg->hdr.type;
+	int err;
+
+	if (!msg->hdr.is_msg)
+		return -EINVAL;
+
+	if (le16_to_cpu(msg->hdr.len) > PAGE_SIZE / sizeof(u32))
+		return -E2BIG;
+
+	while (parser->id != msg_id) {
+		if (parser->id == FBNIC_TLV_MSG_ID_UNKNOWN)
+			return -ENOENT;
+		parser++;
+	}
+
+	err = fbnic_tlv_attr_parse(&msg[1], le16_to_cpu(msg->hdr.len) - 1,
+				   results, parser->attr);
+	if (err)
+		return err;
+
+	return parser->func(opaque, results);
+}
+
+/**
+ * fbnic_tlv_parser_error - called if message doesn't match known type
+ * @opaque: (unused)
+ * @results: (unused)
+ *
+ * Return: -EBADMSG to indicate the message is an unsupported type
+ **/
+int fbnic_tlv_parser_error(void *opaque, struct fbnic_tlv_msg **results)
+{
+	return -EBADMSG;
+}
+
+void fbnic_tlv_attr_addr_copy(u8 *dest, struct fbnic_tlv_msg *src)
+{
+	u8 *mac_addr;
+
+	mac_addr = fbnic_tlv_attr_get_value_ptr(src);
+	memcpy(dest, mac_addr, ETH_ALEN);
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_tlv.h b/drivers/net/ethernet/meta/fbnic/fbnic_tlv.h
new file mode 100644
index 000000000000..67300ab44353
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_tlv.h
@@ -0,0 +1,175 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#ifndef _FBNIC_TLV_H_
+#define _FBNIC_TLV_H_
+
+#include <asm/byteorder.h>
+#include <linux/bits.h>
+#include <linux/const.h>
+#include <linux/types.h>
+
+#define FBNIC_TLV_MSG_ALIGN(len)	ALIGN(len, sizeof(u32))
+#define FBNIC_TLV_MSG_SIZE(len)		\
+		(FBNIC_TLV_MSG_ALIGN(len) / sizeof(u32))
+
+/* TLV Header Format
+ *    3			  2		      1
+ *  1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * |		Length		   |M|I|RSV|	   Type / ID	   |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ *
+ * The TLV header format described above will be used for transferring
+ * messages between the host and the firmware. To ensure byte ordering
+ * we have defined all fields as being little endian.
+ * Type/ID: Identifier for message and/or attribute
+ * RSV: Reserved field for future use, likely as additional flags
+ * I: cannot_ignore flag, identifies if unrecognized attribute can be ignored
+ * M: is_msg, indicates that this is the start of a new message
+ * Length: Total length of message in dwords including header
+ *		or
+ *	   Total length of attribute in bytes including header
+ */
+struct fbnic_tlv_hdr {
+#if defined(__LITTLE_ENDIAN_BITFIELD)
+	u16 type		: 12; /* 0 .. 11  Type / ID */
+	u16 rsvd		: 2;  /* 12 .. 13 Reserved for future use */
+	u16 cannot_ignore	: 1;  /* 14	  Attribute can be ignored */
+	u16 is_msg		: 1;  /* 15	  Header belongs to message */
+#elif defined(__BIG_ENDIAN_BITFIELD)
+	u16 is_msg		: 1;  /* 15	  Header belongs to message */
+	u16 cannot_ignore	: 1;  /* 14	  Attribute can be ignored */
+	u16 rsvd		: 2;  /* 13 .. 12 Reserved for future use */
+	u16 type		: 12; /* 11 .. 0  Type / ID */
+#else
+#error "Missing defines from byteorder.h"
+#endif
+	__le16 len;		/* 16 .. 32	length including TLV header */
+};
+
+#define FBNIC_TLV_RESULTS_MAX		32
+
+struct fbnic_tlv_msg {
+	struct fbnic_tlv_hdr	hdr;
+	__le32			value[];
+};
+
+#define FBNIC_TLV_MSG_ID_UNKNOWN		USHRT_MAX
+
+enum fbnic_tlv_type {
+	FBNIC_TLV_STRING,
+	FBNIC_TLV_FLAG,
+	FBNIC_TLV_UNSIGNED,
+	FBNIC_TLV_SIGNED,
+	FBNIC_TLV_BINARY,
+	FBNIC_TLV_NESTED,
+	FBNIC_TLV_ARRAY,
+	__FBNIC_TLV_MAX_TYPE
+};
+
+/* TLV Index
+ * Defines the relationship between the attribute IDs and their types.
+ * For each entry in the index there will be a size and type associated
+ * with it so that we can use this to parse the data and verify it matches
+ * the expected layout.
+ */
+struct fbnic_tlv_index {
+	u16			id;
+	u16			len;
+	enum fbnic_tlv_type	type;
+};
+
+#define TLV_MAX_DATA			(PAGE_SIZE - 512)
+#define FBNIC_TLV_ATTR_ID_UNKNOWN	USHRT_MAX
+#define FBNIC_TLV_ATTR_STRING(id, len)	{ id, len, FBNIC_TLV_STRING }
+#define FBNIC_TLV_ATTR_FLAG(id)		{ id, 0, FBNIC_TLV_FLAG }
+#define FBNIC_TLV_ATTR_U32(id)		{ id, sizeof(u32), FBNIC_TLV_UNSIGNED }
+#define FBNIC_TLV_ATTR_U64(id)		{ id, sizeof(u64), FBNIC_TLV_UNSIGNED }
+#define FBNIC_TLV_ATTR_S32(id)		{ id, sizeof(s32), FBNIC_TLV_SIGNED }
+#define FBNIC_TLV_ATTR_S64(id)		{ id, sizeof(s64), FBNIC_TLV_SIGNED }
+#define FBNIC_TLV_ATTR_MAC_ADDR(id)	{ id, ETH_ALEN, FBNIC_TLV_BINARY }
+#define FBNIC_TLV_ATTR_NESTED(id)	{ id, 0, FBNIC_TLV_NESTED }
+#define FBNIC_TLV_ATTR_ARRAY(id)	{ id, 0, FBNIC_TLV_ARRAY }
+#define FBNIC_TLV_ATTR_RAW_DATA(id)	{ id, TLV_MAX_DATA, FBNIC_TLV_BINARY }
+#define FBNIC_TLV_ATTR_LAST		{ FBNIC_TLV_ATTR_ID_UNKNOWN, 0, 0 }
+
+struct fbnic_tlv_parser {
+	u16				id;
+	const struct fbnic_tlv_index	*attr;
+	int				(*func)(void *opaque,
+						struct fbnic_tlv_msg **results);
+};
+
+#define FBNIC_TLV_PARSER(id, attr, func) { FBNIC_TLV_MSG_ID_##id, attr, func }
+
+static inline void *
+fbnic_tlv_attr_get_value_ptr(struct fbnic_tlv_msg *attr)
+{
+	return (void *)&attr->value[0];
+}
+
+static inline bool fbnic_tlv_attr_get_bool(struct fbnic_tlv_msg *attr)
+{
+	return !!attr;
+}
+
+u64 fbnic_tlv_attr_get_unsigned(struct fbnic_tlv_msg *attr);
+s64 fbnic_tlv_attr_get_signed(struct fbnic_tlv_msg *attr);
+size_t fbnic_tlv_attr_get_string(struct fbnic_tlv_msg *attr, char *str,
+				 size_t max_size);
+
+#define get_unsigned_result(id, location) \
+do { \
+	struct fbnic_tlv_msg *result = results[id]; \
+	if (result) \
+		location = fbnic_tlv_attr_get_unsigned(result); \
+} while (0)
+
+#define get_signed_result(id, location) \
+do { \
+	struct fbnic_tlv_msg *result = results[id]; \
+	if (result) \
+		location = fbnic_tlv_attr_get_signed(result); \
+} while (0)
+
+#define get_string_result(id, size, str, max_size) \
+do { \
+	struct fbnic_tlv_msg *result = results[id]; \
+	if (result) \
+		size = fbnic_tlv_attr_get_string(result, str, max_size); \
+} while (0)
+
+#define get_bool(id) (!!(results[id]))
+
+struct fbnic_tlv_msg *fbnic_tlv_msg_alloc(u16 msg_id);
+int fbnic_tlv_attr_put_flag(struct fbnic_tlv_msg *msg, const u16 attr_id);
+int fbnic_tlv_attr_put_value(struct fbnic_tlv_msg *msg, const u16 attr_id,
+			     const void *value, const int len);
+int __fbnic_tlv_attr_put_int(struct fbnic_tlv_msg *msg, const u16 attr_id,
+			     s64 value, const int len);
+#define fbnic_tlv_attr_put_int(msg, attr_id, value) \
+	__fbnic_tlv_attr_put_int(msg, attr_id, value, \
+				 FBNIC_TLV_MSG_ALIGN(sizeof(value)))
+int fbnic_tlv_attr_put_mac_addr(struct fbnic_tlv_msg *msg, const u16 attr_id,
+				const u8 *mac_addr);
+int fbnic_tlv_attr_put_string(struct fbnic_tlv_msg *msg, u16 attr_id,
+			      const char *string);
+struct fbnic_tlv_msg *fbnic_tlv_attr_nest_start(struct fbnic_tlv_msg *msg,
+						u16 attr_id);
+void fbnic_tlv_attr_nest_stop(struct fbnic_tlv_msg *msg);
+void fbnic_tlv_attr_addr_copy(u8 *dest, struct fbnic_tlv_msg *src);
+int fbnic_tlv_attr_parse_array(struct fbnic_tlv_msg *attr, int len,
+			       struct fbnic_tlv_msg **results,
+			       const struct fbnic_tlv_index *tlv_index,
+			       u16 tlv_attr_id, size_t array_len);
+int fbnic_tlv_attr_parse(struct fbnic_tlv_msg *attr, int len,
+			 struct fbnic_tlv_msg **results,
+			 const struct fbnic_tlv_index *tlv_index);
+int fbnic_tlv_msg_parse(void *opaque, struct fbnic_tlv_msg *msg,
+			const struct fbnic_tlv_parser *parser);
+int fbnic_tlv_parser_error(void *opaque, struct fbnic_tlv_msg **results);
+
+#define FBNIC_TLV_MSG_ERROR \
+	FBNIC_TLV_PARSER(UNKNOWN, NULL, fbnic_tlv_parser_error)
+#endif /* _FBNIC_TLV_H_ */
-- 
cgit v1.2.3-58-ga151


From da3cde08209ec1c915195c2331c275397f34a731 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:49:31 -0700
Subject: eth: fbnic: Add FW communication mechanism

Add a mechanism for sending messages to and receiving messages
from the FW. The FW has fairly limited functionality, so the
mechanism doesn't have to support high message rate.

Use device mailbox registers to form two rings, one "to" and
one "from" the device. The rings are just a convention between
driver and FW, not a HW construct. We don't expect messages
larger than 4k so use page-sized buffers.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079937113.1778861.10669864213768701947.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/Makefile    |   1 +
 drivers/net/ethernet/meta/fbnic/fbnic.h     |  18 ++
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h |  79 ++++++
 drivers/net/ethernet/meta/fbnic/fbnic_fw.c  | 380 ++++++++++++++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_fw.h  |  26 ++
 drivers/net/ethernet/meta/fbnic/fbnic_irq.c |  78 ++++++
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c |  59 +++++
 7 files changed, 641 insertions(+)
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_fw.c
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_fw.h

diff --git a/drivers/net/ethernet/meta/fbnic/Makefile b/drivers/net/ethernet/meta/fbnic/Makefile
index 0434ee0b3069..7b63cd5b09d4 100644
--- a/drivers/net/ethernet/meta/fbnic/Makefile
+++ b/drivers/net/ethernet/meta/fbnic/Makefile
@@ -8,6 +8,7 @@
 obj-$(CONFIG_FBNIC) += fbnic.o
 
 fbnic-y := fbnic_devlink.o \
+	   fbnic_fw.o \
 	   fbnic_irq.o \
 	   fbnic_mac.o \
 	   fbnic_pci.o \
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h
index af863dfabd82..42e2744cfe10 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic.h
@@ -7,6 +7,7 @@
 #include <linux/io.h>
 
 #include "fbnic_csr.h"
+#include "fbnic_fw.h"
 #include "fbnic_mac.h"
 
 struct fbnic_dev {
@@ -15,8 +16,13 @@ struct fbnic_dev {
 	u32 __iomem *uc_addr0;
 	u32 __iomem *uc_addr4;
 	const struct fbnic_mac *mac;
+	unsigned int fw_msix_vector;
 	unsigned short num_irqs;
 
+	struct fbnic_fw_mbx mbx[FBNIC_IPC_MBX_INDICES];
+	/* Lock protecting Tx Mailbox queue to prevent possible races */
+	spinlock_t fw_tx_lock;
+
 	u64 dsn;
 	u32 mps;
 	u32 readrq;
@@ -28,6 +34,7 @@ struct fbnic_dev {
  * causes later.
  */
 enum {
+	FBNIC_FW_MSIX_ENTRY,
 	FBNIC_NON_NAPI_VECTORS
 };
 
@@ -66,6 +73,14 @@ fbnic_rmw32(struct fbnic_dev *fbd, u32 reg, u32 mask, u32 val)
 #define rd32(_f, _r)		fbnic_rd32(_f, _r)
 #define wrfl(_f)		fbnic_wrfl(_f)
 
+bool fbnic_fw_present(struct fbnic_dev *fbd);
+u32 fbnic_fw_rd32(struct fbnic_dev *fbd, u32 reg);
+void fbnic_fw_wr32(struct fbnic_dev *fbd, u32 reg, u32 val);
+
+#define fw_rd32(_f, _r)		fbnic_fw_rd32(_f, _r)
+#define fw_wr32(_f, _r, _v)	fbnic_fw_wr32(_f, _r, _v)
+#define fw_wrfl(_f)		fbnic_fw_rd32(_f, FBNIC_FW_ZERO_REG)
+
 extern char fbnic_driver_name[];
 
 void fbnic_devlink_free(struct fbnic_dev *fbd);
@@ -73,6 +88,9 @@ struct fbnic_dev *fbnic_devlink_alloc(struct pci_dev *pdev);
 void fbnic_devlink_register(struct fbnic_dev *fbd);
 void fbnic_devlink_unregister(struct fbnic_dev *fbd);
 
+int fbnic_fw_enable_mbx(struct fbnic_dev *fbd);
+void fbnic_fw_disable_mbx(struct fbnic_dev *fbd);
+
 void fbnic_free_irqs(struct fbnic_dev *fbd);
 int fbnic_alloc_irqs(struct fbnic_dev *fbd);
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index 56a8bbd8b720..47a5321b68a7 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -9,10 +9,60 @@
 #define CSR_BIT(nr)		(1u << (nr))
 #define CSR_GENMASK(h, l)	GENMASK(h, l)
 
+#define DESC_BIT(nr)		BIT_ULL(nr)
+#define DESC_GENMASK(h, l)	GENMASK_ULL(h, l)
+
 #define PCI_DEVICE_ID_META_FBNIC_ASIC		0x0013
 
 #define FBNIC_CLOCK_FREQ	(600 * (1000 * 1000))
 
+/* Register Definitions
+ *
+ * The registers are laid as indexes into an le32 array. As such the actual
+ * address is 4 times the index value. Below each register is defined as 3
+ * fields, name, index, and Address.
+ *
+ *      Name				Index		Address
+ *************************************************************************/
+/* Interrupt Registers */
+#define FBNIC_CSR_START_INTR		0x00000	/* CSR section delimiter */
+#define FBNIC_INTR_STATUS(n)		(0x00000 + (n))	/* 0x00000 + 4*n */
+#define FBNIC_INTR_STATUS_CNT			8
+#define FBNIC_INTR_MASK(n)		(0x00008 + (n)) /* 0x00020 + 4*n */
+#define FBNIC_INTR_MASK_CNT			8
+#define FBNIC_INTR_SET(n)		(0x00010 + (n))	/* 0x00040 + 4*n */
+#define FBNIC_INTR_SET_CNT			8
+#define FBNIC_INTR_CLEAR(n)		(0x00018 + (n))	/* 0x00060 + 4*n */
+#define FBNIC_INTR_CLEAR_CNT			8
+#define FBNIC_INTR_SW_STATUS(n)		(0x00020 + (n)) /* 0x00080 + 4*n */
+#define FBNIC_INTR_SW_STATUS_CNT		8
+#define FBNIC_INTR_SW_AC_MODE(n)	(0x00028 + (n)) /* 0x000a0 + 4*n */
+#define FBNIC_INTR_SW_AC_MODE_CNT		8
+#define FBNIC_INTR_MASK_SET(n)		(0x00030 + (n)) /* 0x000c0 + 4*n */
+#define FBNIC_INTR_MASK_SET_CNT			8
+#define FBNIC_INTR_MASK_CLEAR(n)	(0x00038 + (n)) /* 0x000e0 + 4*n */
+#define FBNIC_INTR_MASK_CLEAR_CNT		8
+#define FBNIC_MAX_MSIX_VECS		256U
+#define FBNIC_INTR_MSIX_CTRL(n)		(0x00040 + (n)) /* 0x00100 + 4*n */
+#define FBNIC_INTR_MSIX_CTRL_VECTOR_MASK	CSR_GENMASK(7, 0)
+#define FBNIC_INTR_MSIX_CTRL_ENABLE		CSR_BIT(31)
+
+#define FBNIC_CSR_END_INTR		0x0005f	/* CSR section delimiter */
+
+/* Interrupt MSIX Registers */
+#define FBNIC_CSR_START_INTR_CQ		0x00400	/* CSR section delimiter */
+#define FBNIC_INTR_CQ_REARM(n) \
+				(0x00400 + 4 * (n))	/* 0x01000 + 16*n */
+#define FBNIC_INTR_CQ_REARM_CNT			256
+#define FBNIC_INTR_CQ_REARM_RCQ_TIMEOUT		CSR_GENMASK(13, 0)
+#define FBNIC_INTR_CQ_REARM_RCQ_TIMEOUT_UPD_EN	CSR_BIT(14)
+#define FBNIC_INTR_CQ_REARM_TCQ_TIMEOUT		CSR_GENMASK(28, 15)
+#define FBNIC_INTR_CQ_REARM_TCQ_TIMEOUT_UPD_EN	CSR_BIT(29)
+#define FBNIC_INTR_CQ_REARM_INTR_RELOAD		CSR_BIT(30)
+#define FBNIC_INTR_CQ_REARM_INTR_UNMASK		CSR_BIT(31)
+
+#define FBNIC_CSR_END_INTR_CQ		0x007fe	/* CSR section delimiter */
+
 /* Global QM Tx registers */
 #define FBNIC_CSR_START_QM_TX		0x00800	/* CSR section delimiter */
 #define FBNIC_QM_TWQ_DEFAULT_META_L	0x00818		/* 0x02060 */
@@ -318,4 +368,33 @@ enum {
 
 #define FBNIC_MAX_QUEUES		128
 
+/* BAR 4 CSRs */
+
+/* The IPC mailbox consists of 32 mailboxes, with each mailbox consisting
+ * of 32 4 byte registers. We will use 2 registers per descriptor so the
+ * length of the mailbox is reduced to 16.
+ *
+ * Currently we use an offset of 0x6000 on BAR4 for the mailbox so we just
+ * have to do the math and determine the offset based on the mailbox
+ * direction and index inside that mailbox.
+ */
+#define FBNIC_IPC_MBX_DESC_LEN	16
+#define FBNIC_IPC_MBX(mbx_idx, desc_idx)	\
+	((((mbx_idx) * FBNIC_IPC_MBX_DESC_LEN + (desc_idx)) * 2) + 0x6000)
+
+/* Use first register in mailbox to flush writes */
+#define FBNIC_FW_ZERO_REG	FBNIC_IPC_MBX(0, 0)
+
+enum {
+	FBNIC_IPC_MBX_RX_IDX,
+	FBNIC_IPC_MBX_TX_IDX,
+	FBNIC_IPC_MBX_INDICES,
+};
+
+#define FBNIC_IPC_MBX_DESC_LEN_MASK	DESC_GENMASK(63, 48)
+#define FBNIC_IPC_MBX_DESC_EOM		DESC_BIT(46)
+#define FBNIC_IPC_MBX_DESC_ADDR_MASK	DESC_GENMASK(45, 3)
+#define FBNIC_IPC_MBX_DESC_FW_CMPL	DESC_BIT(1)
+#define FBNIC_IPC_MBX_DESC_HOST_CMPL	DESC_BIT(0)
+
 #endif /* _FBNIC_CSR_H_ */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c
new file mode 100644
index 000000000000..feca833ee924
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c
@@ -0,0 +1,380 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#include <linux/bitfield.h>
+#include <linux/delay.h>
+#include <linux/dev_printk.h>
+#include <linux/dma-mapping.h>
+#include <linux/gfp.h>
+#include <linux/types.h>
+
+#include "fbnic.h"
+#include "fbnic_tlv.h"
+
+static void __fbnic_mbx_wr_desc(struct fbnic_dev *fbd, int mbx_idx,
+				int desc_idx, u64 desc)
+{
+	u32 desc_offset = FBNIC_IPC_MBX(mbx_idx, desc_idx);
+
+	fw_wr32(fbd, desc_offset + 1, upper_32_bits(desc));
+	fw_wrfl(fbd);
+	fw_wr32(fbd, desc_offset, lower_32_bits(desc));
+}
+
+static u64 __fbnic_mbx_rd_desc(struct fbnic_dev *fbd, int mbx_idx, int desc_idx)
+{
+	u32 desc_offset = FBNIC_IPC_MBX(mbx_idx, desc_idx);
+	u64 desc;
+
+	desc = fw_rd32(fbd, desc_offset);
+	desc |= (u64)fw_rd32(fbd, desc_offset + 1) << 32;
+
+	return desc;
+}
+
+static void fbnic_mbx_init_desc_ring(struct fbnic_dev *fbd, int mbx_idx)
+{
+	int desc_idx;
+
+	/* Initialize first descriptor to all 0s. Doing this gives us a
+	 * solid stop for the firmware to hit when it is done looping
+	 * through the ring.
+	 */
+	__fbnic_mbx_wr_desc(fbd, mbx_idx, 0, 0);
+
+	fw_wrfl(fbd);
+
+	/* We then fill the rest of the ring starting at the end and moving
+	 * back toward descriptor 0 with skip descriptors that have no
+	 * length nor address, and tell the firmware that they can skip
+	 * them and just move past them to the one we initialized to 0.
+	 */
+	for (desc_idx = FBNIC_IPC_MBX_DESC_LEN; --desc_idx;) {
+		__fbnic_mbx_wr_desc(fbd, mbx_idx, desc_idx,
+				    FBNIC_IPC_MBX_DESC_FW_CMPL |
+				    FBNIC_IPC_MBX_DESC_HOST_CMPL);
+		fw_wrfl(fbd);
+	}
+}
+
+void fbnic_mbx_init(struct fbnic_dev *fbd)
+{
+	int i;
+
+	/* Initialize lock to protect Tx ring */
+	spin_lock_init(&fbd->fw_tx_lock);
+
+	/* Reinitialize mailbox memory */
+	for (i = 0; i < FBNIC_IPC_MBX_INDICES; i++)
+		memset(&fbd->mbx[i], 0, sizeof(struct fbnic_fw_mbx));
+
+	/* Do not auto-clear the FW mailbox interrupt, let SW clear it */
+	wr32(fbd, FBNIC_INTR_SW_AC_MODE(0), ~(1u << FBNIC_FW_MSIX_ENTRY));
+
+	/* Clear any stale causes in vector 0 as that is used for doorbell */
+	wr32(fbd, FBNIC_INTR_CLEAR(0), 1u << FBNIC_FW_MSIX_ENTRY);
+
+	for (i = 0; i < FBNIC_IPC_MBX_INDICES; i++)
+		fbnic_mbx_init_desc_ring(fbd, i);
+}
+
+static int fbnic_mbx_map_msg(struct fbnic_dev *fbd, int mbx_idx,
+			     struct fbnic_tlv_msg *msg, u16 length, u8 eom)
+{
+	struct fbnic_fw_mbx *mbx = &fbd->mbx[mbx_idx];
+	u8 tail = mbx->tail;
+	dma_addr_t addr;
+	int direction;
+
+	if (!mbx->ready || !fbnic_fw_present(fbd))
+		return -ENODEV;
+
+	direction = (mbx_idx == FBNIC_IPC_MBX_RX_IDX) ? DMA_FROM_DEVICE :
+							DMA_TO_DEVICE;
+
+	if (mbx->head == ((tail + 1) % FBNIC_IPC_MBX_DESC_LEN))
+		return -EBUSY;
+
+	addr = dma_map_single(fbd->dev, msg, PAGE_SIZE, direction);
+	if (dma_mapping_error(fbd->dev, addr)) {
+		free_page((unsigned long)msg);
+
+		return -ENOSPC;
+	}
+
+	mbx->buf_info[tail].msg = msg;
+	mbx->buf_info[tail].addr = addr;
+
+	mbx->tail = (tail + 1) % FBNIC_IPC_MBX_DESC_LEN;
+
+	fw_wr32(fbd, FBNIC_IPC_MBX(mbx_idx, mbx->tail), 0);
+
+	__fbnic_mbx_wr_desc(fbd, mbx_idx, tail,
+			    FIELD_PREP(FBNIC_IPC_MBX_DESC_LEN_MASK, length) |
+			    (addr & FBNIC_IPC_MBX_DESC_ADDR_MASK) |
+			    (eom ? FBNIC_IPC_MBX_DESC_EOM : 0) |
+			    FBNIC_IPC_MBX_DESC_HOST_CMPL);
+
+	return 0;
+}
+
+static void fbnic_mbx_unmap_and_free_msg(struct fbnic_dev *fbd, int mbx_idx,
+					 int desc_idx)
+{
+	struct fbnic_fw_mbx *mbx = &fbd->mbx[mbx_idx];
+	int direction;
+
+	if (!mbx->buf_info[desc_idx].msg)
+		return;
+
+	direction = (mbx_idx == FBNIC_IPC_MBX_RX_IDX) ? DMA_FROM_DEVICE :
+							DMA_TO_DEVICE;
+	dma_unmap_single(fbd->dev, mbx->buf_info[desc_idx].addr,
+			 PAGE_SIZE, direction);
+
+	free_page((unsigned long)mbx->buf_info[desc_idx].msg);
+	mbx->buf_info[desc_idx].msg = NULL;
+}
+
+static void fbnic_mbx_clean_desc_ring(struct fbnic_dev *fbd, int mbx_idx)
+{
+	int i;
+
+	fbnic_mbx_init_desc_ring(fbd, mbx_idx);
+
+	for (i = FBNIC_IPC_MBX_DESC_LEN; i--;)
+		fbnic_mbx_unmap_and_free_msg(fbd, mbx_idx, i);
+}
+
+void fbnic_mbx_clean(struct fbnic_dev *fbd)
+{
+	int i;
+
+	for (i = 0; i < FBNIC_IPC_MBX_INDICES; i++)
+		fbnic_mbx_clean_desc_ring(fbd, i);
+}
+
+#define FBNIC_MBX_MAX_PAGE_SIZE	FIELD_MAX(FBNIC_IPC_MBX_DESC_LEN_MASK)
+#define FBNIC_RX_PAGE_SIZE	min_t(int, PAGE_SIZE, FBNIC_MBX_MAX_PAGE_SIZE)
+
+static int fbnic_mbx_alloc_rx_msgs(struct fbnic_dev *fbd)
+{
+	struct fbnic_fw_mbx *rx_mbx = &fbd->mbx[FBNIC_IPC_MBX_RX_IDX];
+	u8 tail = rx_mbx->tail, head = rx_mbx->head, count;
+	int err = 0;
+
+	/* Do nothing if mailbox is not ready, or we already have pages on
+	 * the ring that can be used by the firmware
+	 */
+	if (!rx_mbx->ready)
+		return -ENODEV;
+
+	/* Fill all but 1 unused descriptors in the Rx queue. */
+	count = (head - tail - 1) % FBNIC_IPC_MBX_DESC_LEN;
+	while (!err && count--) {
+		struct fbnic_tlv_msg *msg;
+
+		msg = (struct fbnic_tlv_msg *)__get_free_page(GFP_ATOMIC |
+							      __GFP_NOWARN);
+		if (!msg) {
+			err = -ENOMEM;
+			break;
+		}
+
+		err = fbnic_mbx_map_msg(fbd, FBNIC_IPC_MBX_RX_IDX, msg,
+					FBNIC_RX_PAGE_SIZE, 0);
+		if (err)
+			free_page((unsigned long)msg);
+	}
+
+	return err;
+}
+
+static void fbnic_mbx_process_tx_msgs(struct fbnic_dev *fbd)
+{
+	struct fbnic_fw_mbx *tx_mbx = &fbd->mbx[FBNIC_IPC_MBX_TX_IDX];
+	u8 head = tx_mbx->head;
+	u64 desc;
+
+	while (head != tx_mbx->tail) {
+		desc = __fbnic_mbx_rd_desc(fbd, FBNIC_IPC_MBX_TX_IDX, head);
+		if (!(desc & FBNIC_IPC_MBX_DESC_FW_CMPL))
+			break;
+
+		fbnic_mbx_unmap_and_free_msg(fbd, FBNIC_IPC_MBX_TX_IDX, head);
+
+		head++;
+		head %= FBNIC_IPC_MBX_DESC_LEN;
+	}
+
+	/* Record head for next interrupt */
+	tx_mbx->head = head;
+}
+
+static void fbnic_mbx_postinit_desc_ring(struct fbnic_dev *fbd, int mbx_idx)
+{
+	struct fbnic_fw_mbx *mbx = &fbd->mbx[mbx_idx];
+
+	/* This is a one time init, so just exit if it is completed */
+	if (mbx->ready)
+		return;
+
+	mbx->ready = true;
+
+	switch (mbx_idx) {
+	case FBNIC_IPC_MBX_RX_IDX:
+		/* Make sure we have a page for the FW to write to */
+		fbnic_mbx_alloc_rx_msgs(fbd);
+		break;
+	}
+}
+
+static void fbnic_mbx_postinit(struct fbnic_dev *fbd)
+{
+	int i;
+
+	/* We only need to do this on the first interrupt following init.
+	 * this primes the mailbox so that we will have cleared all the
+	 * skip descriptors.
+	 */
+	if (!(rd32(fbd, FBNIC_INTR_STATUS(0)) & (1u << FBNIC_FW_MSIX_ENTRY)))
+		return;
+
+	wr32(fbd, FBNIC_INTR_CLEAR(0), 1u << FBNIC_FW_MSIX_ENTRY);
+
+	for (i = 0; i < FBNIC_IPC_MBX_INDICES; i++)
+		fbnic_mbx_postinit_desc_ring(fbd, i);
+}
+
+static const struct fbnic_tlv_parser fbnic_fw_tlv_parser[] = {
+	FBNIC_TLV_MSG_ERROR
+};
+
+static void fbnic_mbx_process_rx_msgs(struct fbnic_dev *fbd)
+{
+	struct fbnic_fw_mbx *rx_mbx = &fbd->mbx[FBNIC_IPC_MBX_RX_IDX];
+	u8 head = rx_mbx->head;
+	u64 desc, length;
+
+	while (head != rx_mbx->tail) {
+		struct fbnic_tlv_msg *msg;
+		int err;
+
+		desc = __fbnic_mbx_rd_desc(fbd, FBNIC_IPC_MBX_RX_IDX, head);
+		if (!(desc & FBNIC_IPC_MBX_DESC_FW_CMPL))
+			break;
+
+		dma_unmap_single(fbd->dev, rx_mbx->buf_info[head].addr,
+				 PAGE_SIZE, DMA_FROM_DEVICE);
+
+		msg = rx_mbx->buf_info[head].msg;
+
+		length = FIELD_GET(FBNIC_IPC_MBX_DESC_LEN_MASK, desc);
+
+		/* Ignore NULL mailbox descriptors */
+		if (!length)
+			goto next_page;
+
+		/* Report descriptors with length greater than page size */
+		if (length > PAGE_SIZE) {
+			dev_warn(fbd->dev,
+				 "Invalid mailbox descriptor length: %lld\n",
+				 length);
+			goto next_page;
+		}
+
+		if (le16_to_cpu(msg->hdr.len) * sizeof(u32) > length)
+			dev_warn(fbd->dev, "Mailbox message length mismatch\n");
+
+		/* If parsing fails dump contents of message to dmesg */
+		err = fbnic_tlv_msg_parse(fbd, msg, fbnic_fw_tlv_parser);
+		if (err) {
+			dev_warn(fbd->dev, "Unable to process message: %d\n",
+				 err);
+			print_hex_dump(KERN_WARNING, "fbnic:",
+				       DUMP_PREFIX_OFFSET, 16, 2,
+				       msg, length, true);
+		}
+
+		dev_dbg(fbd->dev, "Parsed msg type %d\n", msg->hdr.type);
+next_page:
+
+		free_page((unsigned long)rx_mbx->buf_info[head].msg);
+		rx_mbx->buf_info[head].msg = NULL;
+
+		head++;
+		head %= FBNIC_IPC_MBX_DESC_LEN;
+	}
+
+	/* Record head for next interrupt */
+	rx_mbx->head = head;
+
+	/* Make sure we have at least one page for the FW to write to */
+	fbnic_mbx_alloc_rx_msgs(fbd);
+}
+
+void fbnic_mbx_poll(struct fbnic_dev *fbd)
+{
+	fbnic_mbx_postinit(fbd);
+
+	fbnic_mbx_process_tx_msgs(fbd);
+	fbnic_mbx_process_rx_msgs(fbd);
+}
+
+int fbnic_mbx_poll_tx_ready(struct fbnic_dev *fbd)
+{
+	struct fbnic_fw_mbx *tx_mbx;
+	int attempts = 50;
+
+	/* Immediate fail if BAR4 isn't there */
+	if (!fbnic_fw_present(fbd))
+		return -ENODEV;
+
+	tx_mbx = &fbd->mbx[FBNIC_IPC_MBX_TX_IDX];
+	while (!tx_mbx->ready && --attempts) {
+		/* Force the firmware to trigger an interrupt response to
+		 * avoid the mailbox getting stuck closed if the interrupt
+		 * is reset.
+		 */
+		fbnic_mbx_init_desc_ring(fbd, FBNIC_IPC_MBX_TX_IDX);
+
+		msleep(200);
+
+		fbnic_mbx_poll(fbd);
+	}
+
+	return attempts ? 0 : -ETIMEDOUT;
+}
+
+void fbnic_mbx_flush_tx(struct fbnic_dev *fbd)
+{
+	struct fbnic_fw_mbx *tx_mbx;
+	int attempts = 50;
+	u8 count = 0;
+
+	/* Nothing to do if there is no mailbox */
+	if (!fbnic_fw_present(fbd))
+		return;
+
+	/* Record current Rx stats */
+	tx_mbx = &fbd->mbx[FBNIC_IPC_MBX_TX_IDX];
+
+	/* Nothing to do if mailbox never got to ready */
+	if (!tx_mbx->ready)
+		return;
+
+	/* Give firmware time to process packet,
+	 * we will wait up to 10 seconds which is 50 waits of 200ms.
+	 */
+	do {
+		u8 head = tx_mbx->head;
+
+		if (head == tx_mbx->tail)
+			break;
+
+		msleep(200);
+		fbnic_mbx_process_tx_msgs(fbd);
+
+		count += (tx_mbx->head - head) % FBNIC_IPC_MBX_DESC_LEN;
+	} while (count < FBNIC_IPC_MBX_DESC_LEN && --attempts);
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.h b/drivers/net/ethernet/meta/fbnic/fbnic_fw.h
new file mode 100644
index 000000000000..c143079f881c
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.h
@@ -0,0 +1,26 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#ifndef _FBNIC_FW_H_
+#define _FBNIC_FW_H_
+
+#include <linux/types.h>
+
+struct fbnic_dev;
+struct fbnic_tlv_msg;
+
+struct fbnic_fw_mbx {
+	u8 ready, head, tail;
+	struct {
+		struct fbnic_tlv_msg	*msg;
+		dma_addr_t		addr;
+	} buf_info[FBNIC_IPC_MBX_DESC_LEN];
+};
+
+void fbnic_mbx_init(struct fbnic_dev *fbd);
+void fbnic_mbx_clean(struct fbnic_dev *fbd);
+void fbnic_mbx_poll(struct fbnic_dev *fbd);
+int fbnic_mbx_poll_tx_ready(struct fbnic_dev *fbd);
+void fbnic_mbx_flush_tx(struct fbnic_dev *fbd);
+
+#endif /* _FBNIC_FW_H_ */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c
index 7d1475750b64..66a2034514e5 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c
@@ -6,10 +6,86 @@
 
 #include "fbnic.h"
 
+static irqreturn_t fbnic_fw_msix_intr(int __always_unused irq, void *data)
+{
+	struct fbnic_dev *fbd = (struct fbnic_dev *)data;
+
+	fbnic_mbx_poll(fbd);
+
+	fbnic_wr32(fbd, FBNIC_INTR_MASK_CLEAR(0), 1u << FBNIC_FW_MSIX_ENTRY);
+
+	return IRQ_HANDLED;
+}
+
+/**
+ * fbnic_fw_enable_mbx - Configure and initialize Firmware Mailbox
+ * @fbd: Pointer to device to initialize
+ *
+ * This function will initialize the firmware mailbox rings, enable the IRQ
+ * and initialize the communication between the Firmware and the host. The
+ * firmware is expected to respond to the initialization by sending an
+ * interrupt essentially notifying the host that it has seen the
+ * initialization and is now synced up.
+ *
+ * Return: non-zero on failure.
+ **/
+int fbnic_fw_enable_mbx(struct fbnic_dev *fbd)
+{
+	u32 vector = fbd->fw_msix_vector;
+	int err;
+
+	/* Request the IRQ for FW Mailbox vector. */
+	err = request_threaded_irq(vector, NULL, &fbnic_fw_msix_intr,
+				   IRQF_ONESHOT, dev_name(fbd->dev), fbd);
+	if (err)
+		return err;
+
+	/* Initialize mailbox and attempt to poll it into ready state */
+	fbnic_mbx_init(fbd);
+	err = fbnic_mbx_poll_tx_ready(fbd);
+	if (err) {
+		dev_warn(fbd->dev, "FW mailbox did not enter ready state\n");
+		free_irq(vector, fbd);
+		return err;
+	}
+
+	/* Enable interrupts */
+	fbnic_wr32(fbd, FBNIC_INTR_MASK_CLEAR(0), 1u << FBNIC_FW_MSIX_ENTRY);
+
+	return 0;
+}
+
+/**
+ * fbnic_fw_disable_mbx - Disable mailbox and place it in standby state
+ * @fbd: Pointer to device to disable
+ *
+ * This function will disable the mailbox interrupt, free any messages still
+ * in the mailbox and place it into a standby state. The firmware is
+ * expected to see the update and assume that the host is in the reset state.
+ **/
+void fbnic_fw_disable_mbx(struct fbnic_dev *fbd)
+{
+	/* Disable interrupt and free vector */
+	fbnic_wr32(fbd, FBNIC_INTR_MASK_SET(0), 1u << FBNIC_FW_MSIX_ENTRY);
+
+	/* Free the vector */
+	free_irq(fbd->fw_msix_vector, fbd);
+
+	/* Make sure disabling logs message is sent, must be done here to
+	 * avoid risk of completing without a running interrupt.
+	 */
+	fbnic_mbx_flush_tx(fbd);
+
+	/* Reset the mailboxes to the initialized state */
+	fbnic_mbx_clean(fbd);
+}
+
 void fbnic_free_irqs(struct fbnic_dev *fbd)
 {
 	struct pci_dev *pdev = to_pci_dev(fbd->dev);
 
+	fbd->fw_msix_vector = 0;
+
 	fbd->num_irqs = 0;
 
 	pci_free_irq_vectors(pdev);
@@ -35,5 +111,7 @@ int fbnic_alloc_irqs(struct fbnic_dev *fbd)
 
 	fbd->num_irqs = num_irqs;
 
+	fbd->fw_msix_vector = pci_irq_vector(pdev, FBNIC_FW_MSIX_ENTRY);
+
 	return 0;
 }
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
index c03d3945d6d2..0c37f4f32612 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -58,6 +58,48 @@ u32 fbnic_rd32(struct fbnic_dev *fbd, u32 reg)
 	return ~0U;
 }
 
+bool fbnic_fw_present(struct fbnic_dev *fbd)
+{
+	return !!READ_ONCE(fbd->uc_addr4);
+}
+
+void fbnic_fw_wr32(struct fbnic_dev *fbd, u32 reg, u32 val)
+{
+	u32 __iomem *csr = READ_ONCE(fbd->uc_addr4);
+
+	if (csr)
+		writel(val, csr + reg);
+}
+
+u32 fbnic_fw_rd32(struct fbnic_dev *fbd, u32 reg)
+{
+	u32 __iomem *csr = READ_ONCE(fbd->uc_addr4);
+	u32 value;
+
+	if (!csr)
+		return ~0U;
+
+	value = readl(csr + reg);
+
+	/* If any bits are 0 value should be valid */
+	if (~value)
+		return value;
+
+	/* All 1's may be valid if ZEROs register still works */
+	if (reg != FBNIC_FW_ZERO_REG && ~readl(csr + FBNIC_FW_ZERO_REG))
+		return value;
+
+	/* Hardware is giving us all 1's reads, assume it is gone */
+	WRITE_ONCE(fbd->uc_addr0, NULL);
+	WRITE_ONCE(fbd->uc_addr4, NULL);
+
+	dev_err(fbd->dev,
+		"Failed read (idx 0x%x AKA addr 0x%x), disabled CSR access, awaiting reset\n",
+		reg, reg << 2);
+
+	return ~0U;
+}
+
 /**
  * fbnic_probe - Device Initialization Routine
  * @pdev: PCI device information struct
@@ -121,6 +163,13 @@ static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto free_irqs;
 	}
 
+	err = fbnic_fw_enable_mbx(fbd);
+	if (err) {
+		dev_err(&pdev->dev,
+			"Firmware mailbox initialization failure\n");
+		goto free_irqs;
+	}
+
 	fbnic_devlink_register(fbd);
 
 	if (!fbd->dsn) {
@@ -158,6 +207,7 @@ static void fbnic_remove(struct pci_dev *pdev)
 	struct fbnic_dev *fbd = pci_get_drvdata(pdev);
 
 	fbnic_devlink_unregister(fbd);
+	fbnic_fw_disable_mbx(fbd);
 	fbnic_free_irqs(fbd);
 
 	pci_disable_device(pdev);
@@ -168,6 +218,8 @@ static int fbnic_pm_suspend(struct device *dev)
 {
 	struct fbnic_dev *fbd = dev_get_drvdata(dev);
 
+	fbnic_fw_disable_mbx(fbd);
+
 	/* Free the IRQs so they aren't trying to occupy sleeping CPUs */
 	fbnic_free_irqs(fbd);
 
@@ -196,7 +248,14 @@ static int __fbnic_pm_resume(struct device *dev)
 
 	fbd->mac->init_regs(fbd);
 
+	/* Re-enable mailbox */
+	err = fbnic_fw_enable_mbx(fbd);
+	if (err)
+		goto err_free_irqs;
+
 	return 0;
+err_free_irqs:
+	fbnic_free_irqs(fbd);
 err_invalidate_uc_addr:
 	WRITE_ONCE(fbd->uc_addr0, NULL);
 	WRITE_ONCE(fbd->uc_addr4, NULL);
-- 
cgit v1.2.3-58-ga151


From bc6107771bb4be3009168cfda24d22131c026fe8 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:49:43 -0700
Subject: eth: fbnic: Allocate a netdevice and napi vectors with queues

Allocate a netdev and figure out basics like how many queues
we need, MAC address, MTU bounds. Kick off a service task
to do various periodic things like health checking.
The service task only runs when device is open.

We have four levels of objects here:
 - ring - A HW ring with head / tail pointers,
 - triad - Two submission and one completion ring,
 - NAPI - NAPI, with one IRQ and any number of Rx and Tx triads,
 - Netdev - The ultimate container of the rings and napi vectors.

The "triad" is the only less-than-usual construct. On Rx we have
two "free buffer" submission rings, one for packet headers and
one for packet data. On Tx we have separate rings for XDP Tx
and normal Tx. So we ended up with ring triplets in both
directions.

We keep NAPIs on a local list, even though core already maintains a list.
Later on having a separate list will matter for live reconfig.
We introduce the list already, the churn would not be worth it.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079938358.1778861.11681469974633489463.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/Makefile       |   4 +-
 drivers/net/ethernet/meta/fbnic/fbnic.h        |  18 ++
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h    |  40 ++++
 drivers/net/ethernet/meta/fbnic/fbnic_irq.c    |  26 ++-
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.c | 189 +++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.h |  37 ++++
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c    | 138 +++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_txrx.c   | 270 +++++++++++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_txrx.h   |  55 +++++
 9 files changed, 775 insertions(+), 2 deletions(-)
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_txrx.h

diff --git a/drivers/net/ethernet/meta/fbnic/Makefile b/drivers/net/ethernet/meta/fbnic/Makefile
index 7b63cd5b09d4..f2ea90e0c14f 100644
--- a/drivers/net/ethernet/meta/fbnic/Makefile
+++ b/drivers/net/ethernet/meta/fbnic/Makefile
@@ -11,5 +11,7 @@ fbnic-y := fbnic_devlink.o \
 	   fbnic_fw.o \
 	   fbnic_irq.o \
 	   fbnic_mac.o \
+	   fbnic_netdev.o \
 	   fbnic_pci.o \
-	   fbnic_tlv.o
+	   fbnic_tlv.o \
+	   fbnic_txrx.o
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h
index 42e2744cfe10..8abae3b8ef4d 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic.h
@@ -4,7 +4,10 @@
 #ifndef _FBNIC_H_
 #define _FBNIC_H_
 
+#include <linux/interrupt.h>
 #include <linux/io.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
 
 #include "fbnic_csr.h"
 #include "fbnic_fw.h"
@@ -12,6 +15,7 @@
 
 struct fbnic_dev {
 	struct device *dev;
+	struct net_device *netdev;
 
 	u32 __iomem *uc_addr0;
 	u32 __iomem *uc_addr4;
@@ -19,6 +23,8 @@ struct fbnic_dev {
 	unsigned int fw_msix_vector;
 	unsigned short num_irqs;
 
+	struct delayed_work service_task;
+
 	struct fbnic_fw_mbx mbx[FBNIC_IPC_MBX_INDICES];
 	/* Lock protecting Tx Mailbox queue to prevent possible races */
 	spinlock_t fw_tx_lock;
@@ -26,6 +32,9 @@ struct fbnic_dev {
 	u64 dsn;
 	u32 mps;
 	u32 readrq;
+
+	/* Number of TCQs/RCQs available on hardware */
+	u16 max_num_queues;
 };
 
 /* Reserve entry 0 in the MSI-X "others" array until we have filled all
@@ -81,6 +90,11 @@ void fbnic_fw_wr32(struct fbnic_dev *fbd, u32 reg, u32 val);
 #define fw_wr32(_f, _r, _v)	fbnic_fw_wr32(_f, _r, _v)
 #define fw_wrfl(_f)		fbnic_fw_rd32(_f, FBNIC_FW_ZERO_REG)
 
+static inline bool fbnic_init_failure(struct fbnic_dev *fbd)
+{
+	return !fbd->netdev;
+}
+
 extern char fbnic_driver_name[];
 
 void fbnic_devlink_free(struct fbnic_dev *fbd);
@@ -91,6 +105,9 @@ void fbnic_devlink_unregister(struct fbnic_dev *fbd);
 int fbnic_fw_enable_mbx(struct fbnic_dev *fbd);
 void fbnic_fw_disable_mbx(struct fbnic_dev *fbd);
 
+int fbnic_request_irq(struct fbnic_dev *dev, int nr, irq_handler_t handler,
+		      unsigned long flags, const char *name, void *data);
+void fbnic_free_irq(struct fbnic_dev *dev, int nr, void *data);
 void fbnic_free_irqs(struct fbnic_dev *fbd);
 int fbnic_alloc_irqs(struct fbnic_dev *fbd);
 
@@ -99,6 +116,7 @@ enum fbnic_boards {
 };
 
 struct fbnic_info {
+	unsigned int max_num_queues;
 	unsigned int bar_mask;
 };
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index 47a5321b68a7..da1333301d15 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -366,7 +366,47 @@ enum {
 #define FBNIC_PUL_OB_TLP_HDR_AR_CFG_BME		CSR_BIT(18)
 #define FBNIC_CSR_END_PUL_USER	0x31080	/* CSR section delimiter */
 
+/* Queue Registers
+ *
+ * The queue register offsets are specific for a given queue grouping. So to
+ * find the actual register offset it is necessary to combine FBNIC_QUEUE(n)
+ * with the register to get the actual register offset like so:
+ *   FBNIC_QUEUE_TWQ0_CTL(n) == FBNIC_QUEUE(n) + FBNIC_QUEUE_TWQ0_CTL
+ */
+#define FBNIC_CSR_START_QUEUE		0x40000	/* CSR section delimiter */
+#define FBNIC_QUEUE_STRIDE		0x400		/* 0x1000 */
+#define FBNIC_QUEUE(n)\
+	(0x40000 + FBNIC_QUEUE_STRIDE * (n))	/* 0x100000 + 4096*n */
+
+#define FBNIC_QUEUE_TWQ0_CTL		0x000		/* 0x000 */
+#define FBNIC_QUEUE_TWQ1_CTL		0x001		/* 0x004 */
+#define FBNIC_QUEUE_TWQ_CTL_RESET		CSR_BIT(0)
+#define FBNIC_QUEUE_TWQ_CTL_ENABLE		CSR_BIT(1)
+#define FBNIC_QUEUE_TWQ_CTL_PREFETCH_DISABLE	CSR_BIT(2)
+#define FBNIC_QUEUE_TWQ_CTL_TXB_FIFO_SEL_MASK	CSR_GENMASK(30, 29)
+enum {
+	FBNIC_QUEUE_TWQ_CTL_TXB_SHARED	= 0,
+	FBNIC_QUEUE_TWQ_CTL_TXB_EI_DATA	= 1,
+	FBNIC_QUEUE_TWQ_CTL_TXB_EI_CTL	= 2,
+};
+
+#define FBNIC_QUEUE_TWQ_CTL_AGGR_MODE		CSR_BIT(31)
+
+#define FBNIC_QUEUE_TWQ0_TAIL		0x002		/* 0x008 */
+#define FBNIC_QUEUE_TWQ1_TAIL		0x003		/* 0x00c */
+
+/* Tx Completion Queue Registers */
+#define FBNIC_QUEUE_TCQ_HEAD		0x081		/* 0x204 */
+
+/* Rx Completion Queue Registers */
+#define FBNIC_QUEUE_RCQ_HEAD		0x201		/* 0x804 */
+
+/* Rx Buffer Descriptor Queue Registers */
+#define FBNIC_QUEUE_BDQ_HPQ_TAIL	0x241		/* 0x904 */
+#define FBNIC_QUEUE_BDQ_PPQ_TAIL	0x242		/* 0x908 */
+
 #define FBNIC_MAX_QUEUES		128
+#define FBNIC_CSR_END_QUEUE	(0x40000 + 0x400 * FBNIC_MAX_QUEUES - 1)
 
 /* BAR 4 CSRs */
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c
index 66a2034514e5..7aba25d14495 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c
@@ -5,6 +5,7 @@
 #include <linux/types.h>
 
 #include "fbnic.h"
+#include "fbnic_txrx.h"
 
 static irqreturn_t fbnic_fw_msix_intr(int __always_unused irq, void *data)
 {
@@ -80,6 +81,29 @@ void fbnic_fw_disable_mbx(struct fbnic_dev *fbd)
 	fbnic_mbx_clean(fbd);
 }
 
+int fbnic_request_irq(struct fbnic_dev *fbd, int nr, irq_handler_t handler,
+		      unsigned long flags, const char *name, void *data)
+{
+	struct pci_dev *pdev = to_pci_dev(fbd->dev);
+	int irq = pci_irq_vector(pdev, nr);
+
+	if (irq < 0)
+		return irq;
+
+	return request_irq(irq, handler, flags, name, data);
+}
+
+void fbnic_free_irq(struct fbnic_dev *fbd, int nr, void *data)
+{
+	struct pci_dev *pdev = to_pci_dev(fbd->dev);
+	int irq = pci_irq_vector(pdev, nr);
+
+	if (irq < 0)
+		return;
+
+	free_irq(irq, data);
+}
+
 void fbnic_free_irqs(struct fbnic_dev *fbd)
 {
 	struct pci_dev *pdev = to_pci_dev(fbd->dev);
@@ -97,7 +121,7 @@ int fbnic_alloc_irqs(struct fbnic_dev *fbd)
 	struct pci_dev *pdev = to_pci_dev(fbd->dev);
 	int num_irqs;
 
-	wanted_irqs += 1;
+	wanted_irqs += min_t(unsigned int, num_online_cpus(), FBNIC_MAX_RXQS);
 	num_irqs = pci_alloc_irq_vectors(pdev, FBNIC_NON_NAPI_VECTORS + 1,
 					 wanted_irqs, PCI_IRQ_MSIX);
 	if (num_irqs < 0) {
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
new file mode 100644
index 000000000000..41ffae86f80b
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
@@ -0,0 +1,189 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#include <linux/etherdevice.h>
+#include <linux/ipv6.h>
+#include <linux/types.h>
+
+#include "fbnic.h"
+#include "fbnic_netdev.h"
+#include "fbnic_txrx.h"
+
+int __fbnic_open(struct fbnic_net *fbn)
+{
+	int err;
+
+	err = fbnic_alloc_napi_vectors(fbn);
+	if (err)
+		return err;
+
+	err = netif_set_real_num_tx_queues(fbn->netdev,
+					   fbn->num_tx_queues);
+	if (err)
+		goto free_resources;
+
+	err = netif_set_real_num_rx_queues(fbn->netdev,
+					   fbn->num_rx_queues);
+	if (err)
+		goto free_resources;
+
+	return 0;
+free_resources:
+	fbnic_free_napi_vectors(fbn);
+	return err;
+}
+
+static int fbnic_open(struct net_device *netdev)
+{
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	int err;
+
+	err = __fbnic_open(fbn);
+	if (!err)
+		fbnic_up(fbn);
+
+	return err;
+}
+
+static int fbnic_stop(struct net_device *netdev)
+{
+	struct fbnic_net *fbn = netdev_priv(netdev);
+
+	fbnic_down(fbn);
+
+	fbnic_free_napi_vectors(fbn);
+
+	return 0;
+}
+
+static const struct net_device_ops fbnic_netdev_ops = {
+	.ndo_open		= fbnic_open,
+	.ndo_stop		= fbnic_stop,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_start_xmit		= fbnic_xmit_frame,
+};
+
+void fbnic_reset_queues(struct fbnic_net *fbn,
+			unsigned int tx, unsigned int rx)
+{
+	struct fbnic_dev *fbd = fbn->fbd;
+	unsigned int max_napis;
+
+	max_napis = fbd->num_irqs - FBNIC_NON_NAPI_VECTORS;
+
+	tx = min(tx, max_napis);
+	fbn->num_tx_queues = tx;
+
+	rx = min(rx, max_napis);
+	fbn->num_rx_queues = rx;
+
+	fbn->num_napi = max(tx, rx);
+}
+
+/**
+ * fbnic_netdev_free - Free the netdev associate with fbnic
+ * @fbd: Driver specific structure to free netdev from
+ *
+ * Allocate and initialize the netdev and netdev private structure. Bind
+ * together the hardware, netdev, and pci data structures.
+ **/
+void fbnic_netdev_free(struct fbnic_dev *fbd)
+{
+	free_netdev(fbd->netdev);
+	fbd->netdev = NULL;
+}
+
+/**
+ * fbnic_netdev_alloc - Allocate a netdev and associate with fbnic
+ * @fbd: Driver specific structure to associate netdev with
+ *
+ * Allocate and initialize the netdev and netdev private structure. Bind
+ * together the hardware, netdev, and pci data structures.
+ *
+ *  Return: 0 on success, negative on failure
+ **/
+struct net_device *fbnic_netdev_alloc(struct fbnic_dev *fbd)
+{
+	struct net_device *netdev;
+	struct fbnic_net *fbn;
+	int default_queues;
+
+	netdev = alloc_etherdev_mq(sizeof(*fbn), FBNIC_MAX_RXQS);
+	if (!netdev)
+		return NULL;
+
+	SET_NETDEV_DEV(netdev, fbd->dev);
+	fbd->netdev = netdev;
+
+	netdev->netdev_ops = &fbnic_netdev_ops;
+
+	fbn = netdev_priv(netdev);
+
+	fbn->netdev = netdev;
+	fbn->fbd = fbd;
+	INIT_LIST_HEAD(&fbn->napis);
+
+	default_queues = netif_get_num_default_rss_queues();
+	if (default_queues > fbd->max_num_queues)
+		default_queues = fbd->max_num_queues;
+
+	fbnic_reset_queues(fbn, default_queues, default_queues);
+
+	netdev->min_mtu = IPV6_MIN_MTU;
+	netdev->max_mtu = FBNIC_MAX_JUMBO_FRAME_SIZE - ETH_HLEN;
+
+	netif_carrier_off(netdev);
+
+	netif_tx_stop_all_queues(netdev);
+
+	return netdev;
+}
+
+static int fbnic_dsn_to_mac_addr(u64 dsn, char *addr)
+{
+	addr[0] = (dsn >> 56) & 0xFF;
+	addr[1] = (dsn >> 48) & 0xFF;
+	addr[2] = (dsn >> 40) & 0xFF;
+	addr[3] = (dsn >> 16) & 0xFF;
+	addr[4] = (dsn >> 8) & 0xFF;
+	addr[5] = dsn & 0xFF;
+
+	return is_valid_ether_addr(addr) ? 0 : -EINVAL;
+}
+
+/**
+ * fbnic_netdev_register - Initialize general software structures
+ * @netdev: Netdev containing structure to initialize and register
+ *
+ * Initialize the MAC address for the netdev and register it.
+ *
+ *  Return: 0 on success, negative on failure
+ **/
+int fbnic_netdev_register(struct net_device *netdev)
+{
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	struct fbnic_dev *fbd = fbn->fbd;
+	u64 dsn = fbd->dsn;
+	u8 addr[ETH_ALEN];
+	int err;
+
+	err = fbnic_dsn_to_mac_addr(dsn, addr);
+	if (!err) {
+		ether_addr_copy(netdev->perm_addr, addr);
+		eth_hw_addr_set(netdev, addr);
+	} else {
+		/* A randomly assigned MAC address will cause provisioning
+		 * issues so instead just fail to spawn the netdev and
+		 * avoid any confusion.
+		 */
+		dev_err(fbd->dev, "MAC addr %pM invalid\n", addr);
+		return err;
+	}
+
+	return register_netdev(netdev);
+}
+
+void fbnic_netdev_unregister(struct net_device *netdev)
+{
+	unregister_netdev(netdev);
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
new file mode 100644
index 000000000000..8d12abe5fb57
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
@@ -0,0 +1,37 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#ifndef _FBNIC_NETDEV_H_
+#define _FBNIC_NETDEV_H_
+
+#include <linux/types.h>
+
+#include "fbnic_txrx.h"
+
+struct fbnic_net {
+	struct fbnic_ring *tx[FBNIC_MAX_TXQS];
+	struct fbnic_ring *rx[FBNIC_MAX_RXQS];
+
+	struct net_device *netdev;
+	struct fbnic_dev *fbd;
+
+	u16 num_napi;
+
+	u16 num_tx_queues;
+	u16 num_rx_queues;
+
+	struct list_head napis;
+};
+
+int __fbnic_open(struct fbnic_net *fbn);
+void fbnic_up(struct fbnic_net *fbn);
+void fbnic_down(struct fbnic_net *fbn);
+
+struct net_device *fbnic_netdev_alloc(struct fbnic_dev *fbd);
+void fbnic_netdev_free(struct fbnic_dev *fbd);
+int fbnic_netdev_register(struct net_device *netdev);
+void fbnic_netdev_unregister(struct net_device *netdev);
+void fbnic_reset_queues(struct fbnic_net *fbn,
+			unsigned int tx, unsigned int rx);
+
+#endif /* _FBNIC_NETDEV_H_ */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
index 0c37f4f32612..f1d0fe959d0b 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -4,10 +4,12 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/pci.h>
+#include <linux/rtnetlink.h>
 #include <linux/types.h>
 
 #include "fbnic.h"
 #include "fbnic_drvinfo.h"
+#include "fbnic_netdev.h"
 
 char fbnic_driver_name[] = DRV_NAME;
 
@@ -15,6 +17,7 @@ MODULE_DESCRIPTION(DRV_SUMMARY);
 MODULE_LICENSE("GPL");
 
 static const struct fbnic_info fbnic_asic_info = {
+	.max_num_queues = FBNIC_MAX_QUEUES,
 	.bar_mask = BIT(0) | BIT(4)
 };
 
@@ -55,6 +58,10 @@ u32 fbnic_rd32(struct fbnic_dev *fbd, u32 reg)
 		"Failed read (idx 0x%x AKA addr 0x%x), disabled CSR access, awaiting reset\n",
 		reg, reg << 2);
 
+	/* Notify stack that device has lost (PCIe) link */
+	if (!fbnic_init_failure(fbd))
+		netif_device_detach(fbd->netdev);
+
 	return ~0U;
 }
 
@@ -97,9 +104,54 @@ u32 fbnic_fw_rd32(struct fbnic_dev *fbd, u32 reg)
 		"Failed read (idx 0x%x AKA addr 0x%x), disabled CSR access, awaiting reset\n",
 		reg, reg << 2);
 
+	/* Notify stack that device has lost (PCIe) link */
+	if (!fbnic_init_failure(fbd))
+		netif_device_detach(fbd->netdev);
+
 	return ~0U;
 }
 
+static void fbnic_service_task_start(struct fbnic_net *fbn)
+{
+	struct fbnic_dev *fbd = fbn->fbd;
+
+	schedule_delayed_work(&fbd->service_task, HZ);
+}
+
+static void fbnic_service_task_stop(struct fbnic_net *fbn)
+{
+	struct fbnic_dev *fbd = fbn->fbd;
+
+	cancel_delayed_work(&fbd->service_task);
+}
+
+void fbnic_up(struct fbnic_net *fbn)
+{
+	netif_tx_start_all_queues(fbn->netdev);
+
+	fbnic_service_task_start(fbn);
+}
+
+void fbnic_down(struct fbnic_net *fbn)
+{
+	fbnic_service_task_stop(fbn);
+
+	netif_tx_disable(fbn->netdev);
+}
+
+static void fbnic_service_task(struct work_struct *work)
+{
+	struct fbnic_dev *fbd = container_of(to_delayed_work(work),
+					     struct fbnic_dev, service_task);
+
+	rtnl_lock();
+
+	if (netif_running(fbd->netdev))
+		schedule_delayed_work(&fbd->service_task, HZ);
+
+	rtnl_unlock();
+}
+
 /**
  * fbnic_probe - Device Initialization Routine
  * @pdev: PCI device information struct
@@ -114,6 +166,7 @@ u32 fbnic_fw_rd32(struct fbnic_dev *fbd, u32 reg)
 static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	const struct fbnic_info *info = fbnic_info_tbl[ent->driver_data];
+	struct net_device *netdev;
 	struct fbnic_dev *fbd;
 	int err;
 
@@ -150,9 +203,14 @@ static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return -ENOMEM;
 	}
 
+	/* Populate driver with hardware-specific info and handlers */
+	fbd->max_num_queues = info->max_num_queues;
+
 	pci_set_master(pdev);
 	pci_save_state(pdev);
 
+	INIT_DELAYED_WORK(&fbd->service_task, fbnic_service_task);
+
 	err = fbnic_alloc_irqs(fbd);
 	if (err)
 		goto free_fbd;
@@ -177,8 +235,22 @@ static int fbnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto init_failure_mode;
 	}
 
+	netdev = fbnic_netdev_alloc(fbd);
+	if (!netdev) {
+		dev_err(&pdev->dev, "Netdev allocation failed\n");
+		goto init_failure_mode;
+	}
+
+	err = fbnic_netdev_register(netdev);
+	if (err) {
+		dev_err(&pdev->dev, "Netdev registration failed: %d\n", err);
+		goto ifm_free_netdev;
+	}
+
 	return 0;
 
+ifm_free_netdev:
+	fbnic_netdev_free(fbd);
 init_failure_mode:
 	dev_warn(&pdev->dev, "Probe error encountered, entering init failure mode. Normal networking functionality will not be available.\n");
 	 /* Always return 0 even on error so devlink is registered to allow
@@ -206,6 +278,14 @@ static void fbnic_remove(struct pci_dev *pdev)
 {
 	struct fbnic_dev *fbd = pci_get_drvdata(pdev);
 
+	if (!fbnic_init_failure(fbd)) {
+		struct net_device *netdev = fbd->netdev;
+
+		fbnic_netdev_unregister(netdev);
+		cancel_delayed_work_sync(&fbd->service_task);
+		fbnic_netdev_free(fbd);
+	}
+
 	fbnic_devlink_unregister(fbd);
 	fbnic_fw_disable_mbx(fbd);
 	fbnic_free_irqs(fbd);
@@ -217,7 +297,21 @@ static void fbnic_remove(struct pci_dev *pdev)
 static int fbnic_pm_suspend(struct device *dev)
 {
 	struct fbnic_dev *fbd = dev_get_drvdata(dev);
+	struct net_device *netdev = fbd->netdev;
 
+	if (fbnic_init_failure(fbd))
+		goto null_uc_addr;
+
+	rtnl_lock();
+
+	netif_device_detach(netdev);
+
+	if (netif_running(netdev))
+		netdev->netdev_ops->ndo_stop(netdev);
+
+	rtnl_unlock();
+
+null_uc_addr:
 	fbnic_fw_disable_mbx(fbd);
 
 	/* Free the IRQs so they aren't trying to occupy sleeping CPUs */
@@ -233,7 +327,9 @@ static int fbnic_pm_suspend(struct device *dev)
 static int __fbnic_pm_resume(struct device *dev)
 {
 	struct fbnic_dev *fbd = dev_get_drvdata(dev);
+	struct net_device *netdev = fbd->netdev;
 	void __iomem * const *iomap_table;
+	struct fbnic_net *fbn;
 	int err;
 
 	/* Restore MMIO access */
@@ -253,7 +349,29 @@ static int __fbnic_pm_resume(struct device *dev)
 	if (err)
 		goto err_free_irqs;
 
+	/* No netdev means there isn't a network interface to bring up */
+	if (fbnic_init_failure(fbd))
+		return 0;
+
+	fbn = netdev_priv(netdev);
+
+	/* Reset the queues if needed */
+	fbnic_reset_queues(fbn, fbn->num_tx_queues, fbn->num_rx_queues);
+
+	rtnl_lock();
+
+	if (netif_running(netdev)) {
+		err = __fbnic_open(fbn);
+		if (err)
+			goto err_disable_mbx;
+	}
+
+	rtnl_unlock();
+
 	return 0;
+err_disable_mbx:
+	rtnl_unlock();
+	fbnic_fw_disable_mbx(fbd);
 err_free_irqs:
 	fbnic_free_irqs(fbd);
 err_invalidate_uc_addr:
@@ -262,11 +380,30 @@ err_invalidate_uc_addr:
 	return err;
 }
 
+static void __fbnic_pm_attach(struct device *dev)
+{
+	struct fbnic_dev *fbd = dev_get_drvdata(dev);
+	struct net_device *netdev = fbd->netdev;
+	struct fbnic_net *fbn;
+
+	if (fbnic_init_failure(fbd))
+		return;
+
+	fbn = netdev_priv(netdev);
+
+	if (netif_running(netdev))
+		fbnic_up(fbn);
+
+	netif_device_attach(netdev);
+}
+
 static int __maybe_unused fbnic_pm_resume(struct device *dev)
 {
 	int err;
 
 	err = __fbnic_pm_resume(dev);
+	if (!err)
+		__fbnic_pm_attach(dev);
 
 	return err;
 }
@@ -315,6 +452,7 @@ static pci_ers_result_t fbnic_err_slot_reset(struct pci_dev *pdev)
 
 static void fbnic_err_resume(struct pci_dev *pdev)
 {
+	__fbnic_pm_attach(&pdev->dev);
 }
 
 static const struct pci_error_handlers fbnic_err_handler = {
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
new file mode 100644
index 000000000000..68e036f83a03
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
@@ -0,0 +1,270 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#include <linux/pci.h>
+
+#include "fbnic.h"
+#include "fbnic_netdev.h"
+#include "fbnic_txrx.h"
+
+netdev_tx_t fbnic_xmit_frame(struct sk_buff *skb, struct net_device *dev)
+{
+	dev_kfree_skb_any(skb);
+	return NETDEV_TX_OK;
+}
+
+static int fbnic_poll(struct napi_struct *napi, int budget)
+{
+	return 0;
+}
+
+static irqreturn_t fbnic_msix_clean_rings(int __always_unused irq, void *data)
+{
+	struct fbnic_napi_vector *nv = data;
+
+	napi_schedule_irqoff(&nv->napi);
+
+	return IRQ_HANDLED;
+}
+
+static void fbnic_remove_tx_ring(struct fbnic_net *fbn,
+				 struct fbnic_ring *txr)
+{
+	if (!(txr->flags & FBNIC_RING_F_STATS))
+		return;
+
+	/* Remove pointer to the Tx ring */
+	WARN_ON(fbn->tx[txr->q_idx] && fbn->tx[txr->q_idx] != txr);
+	fbn->tx[txr->q_idx] = NULL;
+}
+
+static void fbnic_remove_rx_ring(struct fbnic_net *fbn,
+				 struct fbnic_ring *rxr)
+{
+	if (!(rxr->flags & FBNIC_RING_F_STATS))
+		return;
+
+	/* Remove pointer to the Rx ring */
+	WARN_ON(fbn->rx[rxr->q_idx] && fbn->rx[rxr->q_idx] != rxr);
+	fbn->rx[rxr->q_idx] = NULL;
+}
+
+static void fbnic_free_napi_vector(struct fbnic_net *fbn,
+				   struct fbnic_napi_vector *nv)
+{
+	struct fbnic_dev *fbd = nv->fbd;
+	u32 v_idx = nv->v_idx;
+	int i, j;
+
+	for (i = 0; i < nv->txt_count; i++) {
+		fbnic_remove_tx_ring(fbn, &nv->qt[i].sub0);
+		fbnic_remove_tx_ring(fbn, &nv->qt[i].cmpl);
+	}
+
+	for (j = 0; j < nv->rxt_count; j++, i++) {
+		fbnic_remove_rx_ring(fbn, &nv->qt[i].sub0);
+		fbnic_remove_rx_ring(fbn, &nv->qt[i].sub1);
+		fbnic_remove_rx_ring(fbn, &nv->qt[i].cmpl);
+	}
+
+	fbnic_free_irq(fbd, v_idx, nv);
+	netif_napi_del(&nv->napi);
+	list_del(&nv->napis);
+	kfree(nv);
+}
+
+void fbnic_free_napi_vectors(struct fbnic_net *fbn)
+{
+	struct fbnic_napi_vector *nv, *temp;
+
+	list_for_each_entry_safe(nv, temp, &fbn->napis, napis)
+		fbnic_free_napi_vector(fbn, nv);
+}
+
+static void fbnic_name_napi_vector(struct fbnic_napi_vector *nv)
+{
+	unsigned char *dev_name = nv->napi.dev->name;
+
+	if (!nv->rxt_count)
+		snprintf(nv->name, sizeof(nv->name), "%s-Tx-%u", dev_name,
+			 nv->v_idx - FBNIC_NON_NAPI_VECTORS);
+	else
+		snprintf(nv->name, sizeof(nv->name), "%s-TxRx-%u", dev_name,
+			 nv->v_idx - FBNIC_NON_NAPI_VECTORS);
+}
+
+static void fbnic_ring_init(struct fbnic_ring *ring, u32 __iomem *doorbell,
+			    int q_idx, u8 flags)
+{
+	ring->doorbell = doorbell;
+	ring->q_idx = q_idx;
+	ring->flags = flags;
+}
+
+static int fbnic_alloc_napi_vector(struct fbnic_dev *fbd, struct fbnic_net *fbn,
+				   unsigned int v_count, unsigned int v_idx,
+				   unsigned int txq_count, unsigned int txq_idx,
+				   unsigned int rxq_count, unsigned int rxq_idx)
+{
+	int txt_count = txq_count, rxt_count = rxq_count;
+	u32 __iomem *uc_addr = fbd->uc_addr0;
+	struct fbnic_napi_vector *nv;
+	struct fbnic_q_triad *qt;
+	int qt_count, err;
+	u32 __iomem *db;
+
+	qt_count = txt_count + rxq_count;
+	if (!qt_count)
+		return -EINVAL;
+
+	/* If MMIO has already failed there are no rings to initialize */
+	if (!uc_addr)
+		return -EIO;
+
+	/* Allocate NAPI vector and queue triads */
+	nv = kzalloc(struct_size(nv, qt, qt_count), GFP_KERNEL);
+	if (!nv)
+		return -ENOMEM;
+
+	/* Record queue triad counts */
+	nv->txt_count = txt_count;
+	nv->rxt_count = rxt_count;
+
+	/* Provide pointer back to fbnic and MSI-X vectors */
+	nv->fbd = fbd;
+	nv->v_idx = v_idx;
+
+	/* Record IRQ to NAPI struct */
+	netif_napi_set_irq(&nv->napi,
+			   pci_irq_vector(to_pci_dev(fbd->dev), nv->v_idx));
+
+	/* Tie napi to netdev */
+	list_add(&nv->napis, &fbn->napis);
+	netif_napi_add(fbn->netdev, &nv->napi, fbnic_poll);
+
+	/* Tie nv back to PCIe dev */
+	nv->dev = fbd->dev;
+
+	/* Initialize vector name */
+	fbnic_name_napi_vector(nv);
+
+	/* Request the IRQ for napi vector */
+	err = fbnic_request_irq(fbd, v_idx, &fbnic_msix_clean_rings,
+				IRQF_SHARED, nv->name, nv);
+	if (err)
+		goto napi_del;
+
+	/* Initialize queue triads */
+	qt = nv->qt;
+
+	while (txt_count) {
+		/* Configure Tx queue */
+		db = &uc_addr[FBNIC_QUEUE(txq_idx) + FBNIC_QUEUE_TWQ0_TAIL];
+
+		/* Assign Tx queue to netdev if applicable */
+		if (txq_count > 0) {
+			u8 flags = FBNIC_RING_F_CTX | FBNIC_RING_F_STATS;
+
+			fbnic_ring_init(&qt->sub0, db, txq_idx, flags);
+			fbn->tx[txq_idx] = &qt->sub0;
+			txq_count--;
+		} else {
+			fbnic_ring_init(&qt->sub0, db, 0,
+					FBNIC_RING_F_DISABLED);
+		}
+
+		/* Configure Tx completion queue */
+		db = &uc_addr[FBNIC_QUEUE(txq_idx) + FBNIC_QUEUE_TCQ_HEAD];
+		fbnic_ring_init(&qt->cmpl, db, 0, 0);
+
+		/* Update Tx queue index */
+		txt_count--;
+		txq_idx += v_count;
+
+		/* Move to next queue triad */
+		qt++;
+	}
+
+	while (rxt_count) {
+		/* Configure header queue */
+		db = &uc_addr[FBNIC_QUEUE(rxq_idx) + FBNIC_QUEUE_BDQ_HPQ_TAIL];
+		fbnic_ring_init(&qt->sub0, db, 0, FBNIC_RING_F_CTX);
+
+		/* Configure payload queue */
+		db = &uc_addr[FBNIC_QUEUE(rxq_idx) + FBNIC_QUEUE_BDQ_PPQ_TAIL];
+		fbnic_ring_init(&qt->sub1, db, 0, FBNIC_RING_F_CTX);
+
+		/* Configure Rx completion queue */
+		db = &uc_addr[FBNIC_QUEUE(rxq_idx) + FBNIC_QUEUE_RCQ_HEAD];
+		fbnic_ring_init(&qt->cmpl, db, rxq_idx, FBNIC_RING_F_STATS);
+		fbn->rx[rxq_idx] = &qt->cmpl;
+
+		/* Update Rx queue index */
+		rxt_count--;
+		rxq_idx += v_count;
+
+		/* Move to next queue triad */
+		qt++;
+	}
+
+	return 0;
+
+napi_del:
+	netif_napi_del(&nv->napi);
+	list_del(&nv->napis);
+	kfree(nv);
+	return err;
+}
+
+int fbnic_alloc_napi_vectors(struct fbnic_net *fbn)
+{
+	unsigned int txq_idx = 0, rxq_idx = 0, v_idx = FBNIC_NON_NAPI_VECTORS;
+	unsigned int num_tx = fbn->num_tx_queues;
+	unsigned int num_rx = fbn->num_rx_queues;
+	unsigned int num_napi = fbn->num_napi;
+	struct fbnic_dev *fbd = fbn->fbd;
+	int err;
+
+	/* Allocate 1 Tx queue per napi vector */
+	if (num_napi < FBNIC_MAX_TXQS && num_napi == num_tx + num_rx) {
+		while (num_tx) {
+			err = fbnic_alloc_napi_vector(fbd, fbn,
+						      num_napi, v_idx,
+						      1, txq_idx, 0, 0);
+			if (err)
+				goto free_vectors;
+
+			/* Update counts and index */
+			num_tx--;
+			txq_idx++;
+
+			v_idx++;
+		}
+	}
+
+	/* Allocate Tx/Rx queue pairs per vector, or allocate remaining Rx */
+	while (num_rx | num_tx) {
+		int tqpv = DIV_ROUND_UP(num_tx, num_napi - txq_idx);
+		int rqpv = DIV_ROUND_UP(num_rx, num_napi - rxq_idx);
+
+		err = fbnic_alloc_napi_vector(fbd, fbn, num_napi, v_idx,
+					      tqpv, txq_idx, rqpv, rxq_idx);
+		if (err)
+			goto free_vectors;
+
+		/* Update counts and index */
+		num_tx -= tqpv;
+		txq_idx++;
+
+		num_rx -= rqpv;
+		rxq_idx++;
+
+		v_idx++;
+	}
+
+	return 0;
+
+free_vectors:
+	fbnic_free_napi_vectors(fbn);
+	return -ENOMEM;
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
new file mode 100644
index 000000000000..4b88f0f76137
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
@@ -0,0 +1,55 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#ifndef _FBNIC_TXRX_H_
+#define _FBNIC_TXRX_H_
+
+#include <linux/netdevice.h>
+#include <linux/types.h>
+
+struct fbnic_net;
+
+#define FBNIC_MAX_TXQS			128u
+#define FBNIC_MAX_RXQS			128u
+
+#define FBNIC_RING_F_DISABLED		BIT(0)
+#define FBNIC_RING_F_CTX		BIT(1)
+#define FBNIC_RING_F_STATS		BIT(2)	/* Ring's stats may be used */
+
+struct fbnic_ring {
+	u32 __iomem *doorbell;		/* Pointer to CSR space for ring */
+	u16 size_mask;			/* Size of ring in descriptors - 1 */
+	u8 q_idx;			/* Logical netdev ring index */
+	u8 flags;			/* Ring flags (FBNIC_RING_F_*) */
+
+	u32 head, tail;			/* Head/Tail of ring */
+};
+
+struct fbnic_q_triad {
+	struct fbnic_ring sub0, sub1, cmpl;
+};
+
+struct fbnic_napi_vector {
+	struct napi_struct napi;
+	struct device *dev;		/* Device for DMA unmapping */
+	struct fbnic_dev *fbd;
+	char name[IFNAMSIZ + 9];
+
+	u16 v_idx;
+	u8 txt_count;
+	u8 rxt_count;
+
+	struct list_head napis;
+
+	struct fbnic_q_triad qt[];
+};
+
+#define FBNIC_MAX_TXQS			128u
+#define FBNIC_MAX_RXQS			128u
+
+netdev_tx_t fbnic_xmit_frame(struct sk_buff *skb, struct net_device *dev);
+
+int fbnic_alloc_napi_vectors(struct fbnic_net *fbn);
+void fbnic_free_napi_vectors(struct fbnic_net *fbn);
+
+#endif /* _FBNIC_TXRX_H_ */
-- 
cgit v1.2.3-58-ga151


From 40bf06a160a14eb910c491b6c0deca1821d34cd9 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:49:47 -0700
Subject: eth: fbnic: Implement Tx queue alloc/start/stop/free

Implement basic management operations for Tx queues.
Allocate memory for submission and completion rings.
Learn how to start the queues, stop them, and wait for HW
to be idle.

We call HW rings "descriptor rings" (stored in ring->desc),
and SW context rings "buffer rings" (stored in ring->*_buf union).

This is the first patch which actually touches CSRs so add CSR
helpers.

No actual datapath / packet handling here, yet.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079938724.1778861.8329677776612865169.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h    |  72 +++-
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.c |   9 +
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.h |   2 +
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c    |  19 +-
 drivers/net/ethernet/meta/fbnic/fbnic_txrx.c   | 437 +++++++++++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_txrx.h   |  22 ++
 6 files changed, 550 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index da1333301d15..db423b3424ab 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -61,10 +61,18 @@
 #define FBNIC_INTR_CQ_REARM_INTR_RELOAD		CSR_BIT(30)
 #define FBNIC_INTR_CQ_REARM_INTR_UNMASK		CSR_BIT(31)
 
+#define FBNIC_INTR_RCQ_TIMEOUT(n) \
+				(0x00401 + 4 * (n))	/* 0x01004 + 16*n */
+#define FBNIC_INTR_RCQ_TIMEOUT_CNT		256
+#define FBNIC_INTR_TCQ_TIMEOUT(n) \
+				(0x00402 + 4 * (n))	/* 0x01008 + 16*n */
+#define FBNIC_INTR_TCQ_TIMEOUT_CNT		256
 #define FBNIC_CSR_END_INTR_CQ		0x007fe	/* CSR section delimiter */
 
 /* Global QM Tx registers */
 #define FBNIC_CSR_START_QM_TX		0x00800	/* CSR section delimiter */
+#define FBNIC_QM_TWQ_IDLE(n)		(0x00800 + (n)) /* 0x02000 + 4*n */
+#define FBNIC_QM_TWQ_IDLE_CNT			8
 #define FBNIC_QM_TWQ_DEFAULT_META_L	0x00818		/* 0x02060 */
 #define FBNIC_QM_TWQ_DEFAULT_META_H	0x00819		/* 0x02064 */
 
@@ -86,10 +94,16 @@ enum {
 #define FBNIC_QM_TQS_MTU_CTL0		0x0081d		/* 0x02074 */
 #define FBNIC_QM_TQS_MTU_CTL1		0x0081e		/* 0x02078 */
 #define FBNIC_QM_TQS_MTU_CTL1_BULK		CSR_GENMASK(13, 0)
+#define FBNIC_QM_TCQ_IDLE(n)		(0x00821 + (n)) /* 0x02084 + 4*n */
+#define FBNIC_QM_TCQ_IDLE_CNT			4
 #define FBNIC_QM_TCQ_CTL0		0x0082d		/* 0x020b4 */
 #define FBNIC_QM_TCQ_CTL0_COAL_WAIT		CSR_GENMASK(15, 0)
 #define FBNIC_QM_TCQ_CTL0_TICK_CYCLES		CSR_GENMASK(26, 16)
+#define FBNIC_QM_TQS_IDLE(n)		(0x00830 + (n)) /* 0x020c0 + 4*n */
+#define FBNIC_QM_TQS_IDLE_CNT			8
 #define FBNIC_QM_TQS_EDT_TS_RANGE	0x00849		/* 0x2124 */
+#define FBNIC_QM_TDE_IDLE(n)		(0x00853 + (n)) /* 0x0214c + 4*n */
+#define FBNIC_QM_TDE_IDLE_CNT			8
 #define FBNIC_QM_TNI_TDF_CTL		0x0086c		/* 0x021b0 */
 #define FBNIC_QM_TNI_TDF_CTL_MRRS		CSR_GENMASK(1, 0)
 #define FBNIC_QM_TNI_TDF_CTL_CLS		CSR_GENMASK(3, 2)
@@ -110,9 +124,15 @@ enum {
 
 /* Global QM Rx registers */
 #define FBNIC_CSR_START_QM_RX		0x00c00	/* CSR section delimiter */
+#define FBNIC_QM_RCQ_IDLE(n)		(0x00c00 + (n)) /* 0x03000 + 0x4*n */
+#define FBNIC_QM_RCQ_IDLE_CNT			4
 #define FBNIC_QM_RCQ_CTL0		0x00c0c		/* 0x03030 */
 #define FBNIC_QM_RCQ_CTL0_COAL_WAIT		CSR_GENMASK(15, 0)
 #define FBNIC_QM_RCQ_CTL0_TICK_CYCLES		CSR_GENMASK(26, 16)
+#define FBNIC_QM_HPQ_IDLE(n)		(0x00c0f + (n)) /* 0x0303c + 0x4*n */
+#define FBNIC_QM_HPQ_IDLE_CNT			4
+#define FBNIC_QM_PPQ_IDLE(n)		(0x00c13 + (n)) /* 0x0304c + 0x4*n */
+#define FBNIC_QM_PPQ_IDLE_CNT			4
 #define FBNIC_QM_RNI_RBP_CTL		0x00c2d		/* 0x030b4 */
 #define FBNIC_QM_RNI_RBP_CTL_MRRS		CSR_GENMASK(1, 0)
 #define FBNIC_QM_RNI_RBP_CTL_CLS		CSR_GENMASK(3, 2)
@@ -219,6 +239,8 @@ enum {
 /* TMI registers */
 #define FBNIC_CSR_START_TMI		0x04400	/* CSR section delimiter */
 #define FBNIC_TMI_SOP_PROT_CTRL		0x04400		/* 0x11000 */
+#define FBNIC_TMI_DROP_CTRL		0x04401		/* 0x11004 */
+#define FBNIC_TMI_DROP_CTRL_EN			CSR_BIT(0)
 #define FBNIC_CSR_END_TMI		0x0443f	/* CSR section delimiter */
 /* Rx Buffer Registers */
 #define FBNIC_CSR_START_RXB		0x08000	/* CSR section delimiter */
@@ -382,22 +404,52 @@ enum {
 #define FBNIC_QUEUE_TWQ1_CTL		0x001		/* 0x004 */
 #define FBNIC_QUEUE_TWQ_CTL_RESET		CSR_BIT(0)
 #define FBNIC_QUEUE_TWQ_CTL_ENABLE		CSR_BIT(1)
-#define FBNIC_QUEUE_TWQ_CTL_PREFETCH_DISABLE	CSR_BIT(2)
-#define FBNIC_QUEUE_TWQ_CTL_TXB_FIFO_SEL_MASK	CSR_GENMASK(30, 29)
-enum {
-	FBNIC_QUEUE_TWQ_CTL_TXB_SHARED	= 0,
-	FBNIC_QUEUE_TWQ_CTL_TXB_EI_DATA	= 1,
-	FBNIC_QUEUE_TWQ_CTL_TXB_EI_CTL	= 2,
-};
-
-#define FBNIC_QUEUE_TWQ_CTL_AGGR_MODE		CSR_BIT(31)
-
 #define FBNIC_QUEUE_TWQ0_TAIL		0x002		/* 0x008 */
 #define FBNIC_QUEUE_TWQ1_TAIL		0x003		/* 0x00c */
 
+#define FBNIC_QUEUE_TWQ0_SIZE		0x00a		/* 0x028 */
+#define FBNIC_QUEUE_TWQ1_SIZE		0x00b		/* 0x02c */
+#define FBNIC_QUEUE_TWQ_SIZE_MASK		CSR_GENMASK(3, 0)
+
+#define FBNIC_QUEUE_TWQ0_BAL		0x020		/* 0x080 */
+#define FBNIC_QUEUE_BAL_MASK			CSR_GENMASK(31, 7)
+#define FBNIC_QUEUE_TWQ0_BAH		0x021		/* 0x084 */
+#define FBNIC_QUEUE_TWQ1_BAL		0x022		/* 0x088 */
+#define FBNIC_QUEUE_TWQ1_BAH		0x023		/* 0x08c */
+
 /* Tx Completion Queue Registers */
+#define FBNIC_QUEUE_TCQ_CTL		0x080		/* 0x200 */
+#define FBNIC_QUEUE_TCQ_CTL_RESET		CSR_BIT(0)
+#define FBNIC_QUEUE_TCQ_CTL_ENABLE		CSR_BIT(1)
+
 #define FBNIC_QUEUE_TCQ_HEAD		0x081		/* 0x204 */
 
+#define FBNIC_QUEUE_TCQ_SIZE		0x084		/* 0x210 */
+#define FBNIC_QUEUE_TCQ_SIZE_MASK		CSR_GENMASK(3, 0)
+
+#define FBNIC_QUEUE_TCQ_BAL		0x0a0		/* 0x280 */
+#define FBNIC_QUEUE_TCQ_BAH		0x0a1		/* 0x284 */
+
+/* Tx Interrupt Manager Registers */
+#define FBNIC_QUEUE_TIM_CTL		0x0c0		/* 0x300 */
+#define FBNIC_QUEUE_TIM_CTL_MSIX_MASK		CSR_GENMASK(7, 0)
+
+#define FBNIC_QUEUE_TIM_THRESHOLD	0x0c1		/* 0x304 */
+#define FBNIC_QUEUE_TIM_THRESHOLD_TWD_MASK	CSR_GENMASK(14, 0)
+
+#define FBNIC_QUEUE_TIM_CLEAR		0x0c2		/* 0x308 */
+#define FBNIC_QUEUE_TIM_CLEAR_MASK		CSR_BIT(0)
+#define FBNIC_QUEUE_TIM_SET		0x0c3		/* 0x30c */
+#define FBNIC_QUEUE_TIM_SET_MASK		CSR_BIT(0)
+#define FBNIC_QUEUE_TIM_MASK		0x0c4		/* 0x310 */
+#define FBNIC_QUEUE_TIM_MASK_MASK		CSR_BIT(0)
+
+#define FBNIC_QUEUE_TIM_TIMER		0x0c5		/* 0x314 */
+
+#define FBNIC_QUEUE_TIM_COUNTS		0x0c6		/* 0x318 */
+#define FBNIC_QUEUE_TIM_COUNTS_CNT1_MASK	CSR_GENMASK(30, 16)
+#define FBNIC_QUEUE_TIM_COUNTS_CNT0_MASK	CSR_GENMASK(14, 0)
+
 /* Rx Completion Queue Registers */
 #define FBNIC_QUEUE_RCQ_HEAD		0x201		/* 0x804 */
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
index 41ffae86f80b..c7e9bd1e79c6 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
@@ -17,6 +17,10 @@ int __fbnic_open(struct fbnic_net *fbn)
 	if (err)
 		return err;
 
+	err = fbnic_alloc_resources(fbn);
+	if (err)
+		goto free_napi_vectors;
+
 	err = netif_set_real_num_tx_queues(fbn->netdev,
 					   fbn->num_tx_queues);
 	if (err)
@@ -29,6 +33,8 @@ int __fbnic_open(struct fbnic_net *fbn)
 
 	return 0;
 free_resources:
+	fbnic_free_resources(fbn);
+free_napi_vectors:
 	fbnic_free_napi_vectors(fbn);
 	return err;
 }
@@ -51,6 +57,7 @@ static int fbnic_stop(struct net_device *netdev)
 
 	fbnic_down(fbn);
 
+	fbnic_free_resources(fbn);
 	fbnic_free_napi_vectors(fbn);
 
 	return 0;
@@ -123,6 +130,8 @@ struct net_device *fbnic_netdev_alloc(struct fbnic_dev *fbd)
 	fbn->fbd = fbd;
 	INIT_LIST_HEAD(&fbn->napis);
 
+	fbn->txq_size = FBNIC_TXQ_SIZE_DEFAULT;
+
 	default_queues = netif_get_num_default_rss_queues();
 	if (default_queues > fbd->max_num_queues)
 		default_queues = fbd->max_num_queues;
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
index 8d12abe5fb57..b3c39c10c3f7 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
@@ -15,6 +15,8 @@ struct fbnic_net {
 	struct net_device *netdev;
 	struct fbnic_dev *fbd;
 
+	u32 txq_size;
+
 	u16 num_napi;
 
 	u16 num_tx_queues;
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
index f1d0fe959d0b..08b74ed561b7 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -127,16 +127,33 @@ static void fbnic_service_task_stop(struct fbnic_net *fbn)
 
 void fbnic_up(struct fbnic_net *fbn)
 {
+	fbnic_enable(fbn);
+
+	/* Enable Tx/Rx processing */
+	fbnic_napi_enable(fbn);
 	netif_tx_start_all_queues(fbn->netdev);
 
 	fbnic_service_task_start(fbn);
 }
 
-void fbnic_down(struct fbnic_net *fbn)
+static void fbnic_down_noidle(struct fbnic_net *fbn)
 {
 	fbnic_service_task_stop(fbn);
 
+	/* Disable Tx/Rx Processing */
+	fbnic_napi_disable(fbn);
 	netif_tx_disable(fbn->netdev);
+
+	fbnic_disable(fbn);
+}
+
+void fbnic_down(struct fbnic_net *fbn)
+{
+	fbnic_down_noidle(fbn);
+
+	fbnic_wait_all_queues_idle(fbn->fbd, false);
+
+	fbnic_flush(fbn);
 }
 
 static void fbnic_service_task(struct work_struct *work)
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
index 68e036f83a03..5d836d5a8370 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
@@ -1,18 +1,50 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) Meta Platforms, Inc. and affiliates. */
 
+#include <linux/iopoll.h>
 #include <linux/pci.h>
 
 #include "fbnic.h"
 #include "fbnic_netdev.h"
 #include "fbnic_txrx.h"
 
+static u32 __iomem *fbnic_ring_csr_base(const struct fbnic_ring *ring)
+{
+	unsigned long csr_base = (unsigned long)ring->doorbell;
+
+	csr_base &= ~(FBNIC_QUEUE_STRIDE * sizeof(u32) - 1);
+
+	return (u32 __iomem *)csr_base;
+}
+
+static u32 fbnic_ring_rd32(struct fbnic_ring *ring, unsigned int csr)
+{
+	u32 __iomem *csr_base = fbnic_ring_csr_base(ring);
+
+	return readl(csr_base + csr);
+}
+
+static void fbnic_ring_wr32(struct fbnic_ring *ring, unsigned int csr, u32 val)
+{
+	u32 __iomem *csr_base = fbnic_ring_csr_base(ring);
+
+	writel(val, csr_base + csr);
+}
+
 netdev_tx_t fbnic_xmit_frame(struct sk_buff *skb, struct net_device *dev)
 {
 	dev_kfree_skb_any(skb);
 	return NETDEV_TX_OK;
 }
 
+static void fbnic_nv_irq_disable(struct fbnic_napi_vector *nv)
+{
+	struct fbnic_dev *fbd = nv->fbd;
+	u32 v_idx = nv->v_idx;
+
+	fbnic_wr32(fbd, FBNIC_INTR_MASK_SET(v_idx / 32), 1 << (v_idx % 32));
+}
+
 static int fbnic_poll(struct napi_struct *napi, int budget)
 {
 	return 0;
@@ -266,5 +298,410 @@ int fbnic_alloc_napi_vectors(struct fbnic_net *fbn)
 
 free_vectors:
 	fbnic_free_napi_vectors(fbn);
+
 	return -ENOMEM;
 }
+
+static void fbnic_free_ring_resources(struct device *dev,
+				      struct fbnic_ring *ring)
+{
+	kvfree(ring->buffer);
+	ring->buffer = NULL;
+
+	/* If size is not set there are no descriptors present */
+	if (!ring->size)
+		return;
+
+	dma_free_coherent(dev, ring->size, ring->desc, ring->dma);
+	ring->size_mask = 0;
+	ring->size = 0;
+}
+
+static int fbnic_alloc_tx_ring_desc(struct fbnic_net *fbn,
+				    struct fbnic_ring *txr)
+{
+	struct device *dev = fbn->netdev->dev.parent;
+	size_t size;
+
+	/* Round size up to nearest 4K */
+	size = ALIGN(array_size(sizeof(*txr->desc), fbn->txq_size), 4096);
+
+	txr->desc = dma_alloc_coherent(dev, size, &txr->dma,
+				       GFP_KERNEL | __GFP_NOWARN);
+	if (!txr->desc)
+		return -ENOMEM;
+
+	/* txq_size should be a power of 2, so mask is just that -1 */
+	txr->size_mask = fbn->txq_size - 1;
+	txr->size = size;
+
+	return 0;
+}
+
+static int fbnic_alloc_tx_ring_buffer(struct fbnic_ring *txr)
+{
+	size_t size = array_size(sizeof(*txr->tx_buf), txr->size_mask + 1);
+
+	txr->tx_buf = kvzalloc(size, GFP_KERNEL | __GFP_NOWARN);
+
+	return txr->tx_buf ? 0 : -ENOMEM;
+}
+
+static int fbnic_alloc_tx_ring_resources(struct fbnic_net *fbn,
+					 struct fbnic_ring *txr)
+{
+	struct device *dev = fbn->netdev->dev.parent;
+	int err;
+
+	if (txr->flags & FBNIC_RING_F_DISABLED)
+		return 0;
+
+	err = fbnic_alloc_tx_ring_desc(fbn, txr);
+	if (err)
+		return err;
+
+	if (!(txr->flags & FBNIC_RING_F_CTX))
+		return 0;
+
+	err = fbnic_alloc_tx_ring_buffer(txr);
+	if (err)
+		goto free_desc;
+
+	return 0;
+
+free_desc:
+	fbnic_free_ring_resources(dev, txr);
+	return err;
+}
+
+static void fbnic_free_qt_resources(struct fbnic_net *fbn,
+				    struct fbnic_q_triad *qt)
+{
+	struct device *dev = fbn->netdev->dev.parent;
+
+	fbnic_free_ring_resources(dev, &qt->cmpl);
+	fbnic_free_ring_resources(dev, &qt->sub1);
+	fbnic_free_ring_resources(dev, &qt->sub0);
+}
+
+static int fbnic_alloc_tx_qt_resources(struct fbnic_net *fbn,
+				       struct fbnic_q_triad *qt)
+{
+	struct device *dev = fbn->netdev->dev.parent;
+	int err;
+
+	err = fbnic_alloc_tx_ring_resources(fbn, &qt->sub0);
+	if (err)
+		return err;
+
+	err = fbnic_alloc_tx_ring_resources(fbn, &qt->cmpl);
+	if (err)
+		goto free_sub0;
+
+	return 0;
+
+free_sub0:
+	fbnic_free_ring_resources(dev, &qt->sub0);
+	return err;
+}
+
+static void fbnic_free_nv_resources(struct fbnic_net *fbn,
+				    struct fbnic_napi_vector *nv)
+{
+	int i;
+
+	/* Free Tx Resources  */
+	for (i = 0; i < nv->txt_count; i++)
+		fbnic_free_qt_resources(fbn, &nv->qt[i]);
+}
+
+static int fbnic_alloc_nv_resources(struct fbnic_net *fbn,
+				    struct fbnic_napi_vector *nv)
+{
+	int i, err;
+
+	/* Allocate Tx Resources */
+	for (i = 0; i < nv->txt_count; i++) {
+		err = fbnic_alloc_tx_qt_resources(fbn, &nv->qt[i]);
+		if (err)
+			goto free_resources;
+	}
+
+	return 0;
+
+free_resources:
+	while (i--)
+		fbnic_free_qt_resources(fbn, &nv->qt[i]);
+	return err;
+}
+
+void fbnic_free_resources(struct fbnic_net *fbn)
+{
+	struct fbnic_napi_vector *nv;
+
+	list_for_each_entry(nv, &fbn->napis, napis)
+		fbnic_free_nv_resources(fbn, nv);
+}
+
+int fbnic_alloc_resources(struct fbnic_net *fbn)
+{
+	struct fbnic_napi_vector *nv;
+	int err = -ENODEV;
+
+	list_for_each_entry(nv, &fbn->napis, napis) {
+		err = fbnic_alloc_nv_resources(fbn, nv);
+		if (err)
+			goto free_resources;
+	}
+
+	return 0;
+
+free_resources:
+	list_for_each_entry_continue_reverse(nv, &fbn->napis, napis)
+		fbnic_free_nv_resources(fbn, nv);
+
+	return err;
+}
+
+static void fbnic_disable_twq0(struct fbnic_ring *txr)
+{
+	u32 twq_ctl = fbnic_ring_rd32(txr, FBNIC_QUEUE_TWQ0_CTL);
+
+	twq_ctl &= ~FBNIC_QUEUE_TWQ_CTL_ENABLE;
+
+	fbnic_ring_wr32(txr, FBNIC_QUEUE_TWQ0_CTL, twq_ctl);
+}
+
+static void fbnic_disable_tcq(struct fbnic_ring *txr)
+{
+	fbnic_ring_wr32(txr, FBNIC_QUEUE_TCQ_CTL, 0);
+	fbnic_ring_wr32(txr, FBNIC_QUEUE_TIM_MASK, FBNIC_QUEUE_TIM_MASK_MASK);
+}
+
+void fbnic_napi_disable(struct fbnic_net *fbn)
+{
+	struct fbnic_napi_vector *nv;
+
+	list_for_each_entry(nv, &fbn->napis, napis) {
+		napi_disable(&nv->napi);
+
+		fbnic_nv_irq_disable(nv);
+	}
+}
+
+void fbnic_disable(struct fbnic_net *fbn)
+{
+	struct fbnic_dev *fbd = fbn->fbd;
+	struct fbnic_napi_vector *nv;
+	int i;
+
+	list_for_each_entry(nv, &fbn->napis, napis) {
+		/* Disable Tx queue triads */
+		for (i = 0; i < nv->txt_count; i++) {
+			struct fbnic_q_triad *qt = &nv->qt[i];
+
+			fbnic_disable_twq0(&qt->sub0);
+			fbnic_disable_tcq(&qt->cmpl);
+		}
+	}
+
+	fbnic_wrfl(fbd);
+}
+
+static void fbnic_tx_flush(struct fbnic_dev *fbd)
+{
+	netdev_warn(fbd->netdev, "tiggerring Tx flush\n");
+
+	fbnic_rmw32(fbd, FBNIC_TMI_DROP_CTRL, FBNIC_TMI_DROP_CTRL_EN,
+		    FBNIC_TMI_DROP_CTRL_EN);
+}
+
+static void fbnic_tx_flush_off(struct fbnic_dev *fbd)
+{
+	fbnic_rmw32(fbd, FBNIC_TMI_DROP_CTRL, FBNIC_TMI_DROP_CTRL_EN, 0);
+}
+
+struct fbnic_idle_regs {
+	u32 reg_base;
+	u8 reg_cnt;
+};
+
+static bool fbnic_all_idle(struct fbnic_dev *fbd,
+			   const struct fbnic_idle_regs *regs,
+			   unsigned int nregs)
+{
+	unsigned int i, j;
+
+	for (i = 0; i < nregs; i++) {
+		for (j = 0; j < regs[i].reg_cnt; j++) {
+			if (fbnic_rd32(fbd, regs[i].reg_base + j) != ~0U)
+				return false;
+		}
+	}
+	return true;
+}
+
+static void fbnic_idle_dump(struct fbnic_dev *fbd,
+			    const struct fbnic_idle_regs *regs,
+			    unsigned int nregs, const char *dir, int err)
+{
+	unsigned int i, j;
+
+	netdev_err(fbd->netdev, "error waiting for %s idle %d\n", dir, err);
+	for (i = 0; i < nregs; i++)
+		for (j = 0; j < regs[i].reg_cnt; j++)
+			netdev_err(fbd->netdev, "0x%04x: %08x\n",
+				   regs[i].reg_base + j,
+				   fbnic_rd32(fbd, regs[i].reg_base + j));
+}
+
+int fbnic_wait_all_queues_idle(struct fbnic_dev *fbd, bool may_fail)
+{
+	static const struct fbnic_idle_regs tx[] = {
+		{ FBNIC_QM_TWQ_IDLE(0),	FBNIC_QM_TWQ_IDLE_CNT, },
+		{ FBNIC_QM_TQS_IDLE(0),	FBNIC_QM_TQS_IDLE_CNT, },
+		{ FBNIC_QM_TDE_IDLE(0),	FBNIC_QM_TDE_IDLE_CNT, },
+		{ FBNIC_QM_TCQ_IDLE(0),	FBNIC_QM_TCQ_IDLE_CNT, },
+	};
+	bool idle;
+	int err;
+
+	err = read_poll_timeout_atomic(fbnic_all_idle, idle, idle, 2, 500000,
+				       false, fbd, tx, ARRAY_SIZE(tx));
+	if (err == -ETIMEDOUT) {
+		fbnic_tx_flush(fbd);
+		err = read_poll_timeout_atomic(fbnic_all_idle, idle, idle,
+					       2, 500000, false,
+					       fbd, tx, ARRAY_SIZE(tx));
+		fbnic_tx_flush_off(fbd);
+	}
+	if (err) {
+		fbnic_idle_dump(fbd, tx, ARRAY_SIZE(tx), "Tx", err);
+		if (may_fail)
+			return err;
+	}
+
+	return err;
+}
+
+void fbnic_flush(struct fbnic_net *fbn)
+{
+	struct fbnic_napi_vector *nv;
+
+	list_for_each_entry(nv, &fbn->napis, napis) {
+		int i;
+
+		/* Flush any processed Tx Queue Triads and drop the rest */
+		for (i = 0; i < nv->txt_count; i++) {
+			struct fbnic_q_triad *qt = &nv->qt[i];
+			struct netdev_queue *tx_queue;
+
+			/* Reset completion queue descriptor ring */
+			memset(qt->cmpl.desc, 0, qt->cmpl.size);
+
+			/* Nothing else to do if Tx queue is disabled */
+			if (qt->sub0.flags & FBNIC_RING_F_DISABLED)
+				continue;
+
+			/* Reset BQL associated with Tx queue */
+			tx_queue = netdev_get_tx_queue(nv->napi.dev,
+						       qt->sub0.q_idx);
+			netdev_tx_reset_queue(tx_queue);
+
+			/* Disassociate Tx queue from NAPI */
+			netif_queue_set_napi(nv->napi.dev, qt->sub0.q_idx,
+					     NETDEV_QUEUE_TYPE_TX, NULL);
+		}
+	}
+}
+
+static void fbnic_enable_twq0(struct fbnic_ring *twq)
+{
+	u32 log_size = fls(twq->size_mask);
+
+	if (!twq->size_mask)
+		return;
+
+	/* Reset head/tail */
+	fbnic_ring_wr32(twq, FBNIC_QUEUE_TWQ0_CTL, FBNIC_QUEUE_TWQ_CTL_RESET);
+	twq->tail = 0;
+	twq->head = 0;
+
+	/* Store descriptor ring address and size */
+	fbnic_ring_wr32(twq, FBNIC_QUEUE_TWQ0_BAL, lower_32_bits(twq->dma));
+	fbnic_ring_wr32(twq, FBNIC_QUEUE_TWQ0_BAH, upper_32_bits(twq->dma));
+
+	/* Write lower 4 bits of log size as 64K ring size is 0 */
+	fbnic_ring_wr32(twq, FBNIC_QUEUE_TWQ0_SIZE, log_size & 0xf);
+
+	fbnic_ring_wr32(twq, FBNIC_QUEUE_TWQ0_CTL, FBNIC_QUEUE_TWQ_CTL_ENABLE);
+}
+
+static void fbnic_enable_tcq(struct fbnic_napi_vector *nv,
+			     struct fbnic_ring *tcq)
+{
+	u32 log_size = fls(tcq->size_mask);
+
+	if (!tcq->size_mask)
+		return;
+
+	/* Reset head/tail */
+	fbnic_ring_wr32(tcq, FBNIC_QUEUE_TCQ_CTL, FBNIC_QUEUE_TCQ_CTL_RESET);
+	tcq->tail = 0;
+	tcq->head = 0;
+
+	/* Store descriptor ring address and size */
+	fbnic_ring_wr32(tcq, FBNIC_QUEUE_TCQ_BAL, lower_32_bits(tcq->dma));
+	fbnic_ring_wr32(tcq, FBNIC_QUEUE_TCQ_BAH, upper_32_bits(tcq->dma));
+
+	/* Write lower 4 bits of log size as 64K ring size is 0 */
+	fbnic_ring_wr32(tcq, FBNIC_QUEUE_TCQ_SIZE, log_size & 0xf);
+
+	/* Store interrupt information for the completion queue */
+	fbnic_ring_wr32(tcq, FBNIC_QUEUE_TIM_CTL, nv->v_idx);
+	fbnic_ring_wr32(tcq, FBNIC_QUEUE_TIM_THRESHOLD, tcq->size_mask / 2);
+	fbnic_ring_wr32(tcq, FBNIC_QUEUE_TIM_MASK, 0);
+
+	/* Enable queue */
+	fbnic_ring_wr32(tcq, FBNIC_QUEUE_TCQ_CTL, FBNIC_QUEUE_TCQ_CTL_ENABLE);
+}
+
+void fbnic_enable(struct fbnic_net *fbn)
+{
+	struct fbnic_dev *fbd = fbn->fbd;
+	struct fbnic_napi_vector *nv;
+	int i;
+
+	list_for_each_entry(nv, &fbn->napis, napis) {
+		/* Setup Tx Queue Triads */
+		for (i = 0; i < nv->txt_count; i++) {
+			struct fbnic_q_triad *qt = &nv->qt[i];
+
+			fbnic_enable_twq0(&qt->sub0);
+			fbnic_enable_tcq(nv, &qt->cmpl);
+		}
+	}
+
+	fbnic_wrfl(fbd);
+}
+
+static void fbnic_nv_irq_enable(struct fbnic_napi_vector *nv)
+{
+	struct fbnic_dev *fbd = nv->fbd;
+	u32 val;
+
+	val = FBNIC_INTR_CQ_REARM_INTR_UNMASK;
+
+	fbnic_wr32(fbd, FBNIC_INTR_CQ_REARM(nv->v_idx), val);
+}
+
+void fbnic_napi_enable(struct fbnic_net *fbn)
+{
+	struct fbnic_napi_vector *nv;
+
+	list_for_each_entry(nv, &fbn->napis, napis) {
+		napi_enable(&nv->napi);
+
+		fbnic_nv_irq_enable(nv);
+	}
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
index 4b88f0f76137..77abc15bb0dc 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
@@ -12,17 +12,30 @@ struct fbnic_net;
 #define FBNIC_MAX_TXQS			128u
 #define FBNIC_MAX_RXQS			128u
 
+#define FBNIC_TXQ_SIZE_DEFAULT		1024
+
 #define FBNIC_RING_F_DISABLED		BIT(0)
 #define FBNIC_RING_F_CTX		BIT(1)
 #define FBNIC_RING_F_STATS		BIT(2)	/* Ring's stats may be used */
 
 struct fbnic_ring {
+	/* Pointer to buffer specific info */
+	union {
+		void **tx_buf;			/* TWQ */
+		void *buffer;			/* Generic pointer */
+	};
+
 	u32 __iomem *doorbell;		/* Pointer to CSR space for ring */
+	__le64 *desc;			/* Descriptor ring memory */
 	u16 size_mask;			/* Size of ring in descriptors - 1 */
 	u8 q_idx;			/* Logical netdev ring index */
 	u8 flags;			/* Ring flags (FBNIC_RING_F_*) */
 
 	u32 head, tail;			/* Head/Tail of ring */
+
+	/* Slow path fields follow */
+	dma_addr_t dma;			/* Phys addr of descriptor memory */
+	size_t size;			/* Size of descriptor ring in memory */
 };
 
 struct fbnic_q_triad {
@@ -51,5 +64,14 @@ netdev_tx_t fbnic_xmit_frame(struct sk_buff *skb, struct net_device *dev);
 
 int fbnic_alloc_napi_vectors(struct fbnic_net *fbn);
 void fbnic_free_napi_vectors(struct fbnic_net *fbn);
+int fbnic_alloc_resources(struct fbnic_net *fbn);
+void fbnic_free_resources(struct fbnic_net *fbn);
+void fbnic_napi_enable(struct fbnic_net *fbn);
+void fbnic_napi_disable(struct fbnic_net *fbn);
+void fbnic_enable(struct fbnic_net *fbn);
+void fbnic_disable(struct fbnic_net *fbn);
+void fbnic_flush(struct fbnic_net *fbn);
+
+int fbnic_wait_all_queues_idle(struct fbnic_dev *fbd, bool may_fail);
 
 #endif /* _FBNIC_TXRX_H_ */
-- 
cgit v1.2.3-58-ga151


From 0cb4c0a13723ef6dd06be8f913443897a0ffccb5 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:49:50 -0700
Subject: eth: fbnic: Implement Rx queue alloc/start/stop/free

Implement control path parts of Rx queue handling.

The NIC consumes memory in pages. It takes a full page and places
packets into it in a configurable manner (with the ability to define
headroom / tailroom as well as head alignment requirements).
As mentioned in prior patches there are two page submissions queues
one for packet headers and second (optional) for packet payloads.
For now feed both queues from a single page pool.

Use the page pool "fragment" API, as we can't predict upfront
how the page will be sliced.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079939092.1778861.3780136633831329550.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h    | 103 ++++-
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.c |   3 +
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.h |   3 +
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c    |   2 +
 drivers/net/ethernet/meta/fbnic/fbnic_txrx.c   | 503 ++++++++++++++++++++++++-
 drivers/net/ethernet/meta/fbnic/fbnic_txrx.h   |  33 ++
 6 files changed, 638 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index db423b3424ab..be682fa78b2b 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -16,6 +16,37 @@
 
 #define FBNIC_CLOCK_FREQ	(600 * (1000 * 1000))
 
+/* Rx Buffer Descriptor Format
+ *
+ * The layout of this can vary depending on the page size of the system.
+ *
+ * If the page size is 4K then the layout will simply consist of ID for
+ * the 16 most significant bits, and the lower 46 are essentially the page
+ * address with the lowest 12 bits being reserved 0 due to the fact that
+ * a page will be aligned.
+ *
+ * If the page size is larger than 4K then the lower n bits of the ID and
+ * page address will be reserved for the fragment ID. This fragment will
+ * be 4K in size and will be used to index both the DMA address and the ID
+ * by the same amount.
+ */
+#define FBNIC_BD_DESC_ADDR_MASK			DESC_GENMASK(45, 12)
+#define FBNIC_BD_DESC_ID_MASK			DESC_GENMASK(63, 48)
+#define FBNIC_BD_FRAG_SIZE \
+	(FBNIC_BD_DESC_ADDR_MASK & ~(FBNIC_BD_DESC_ADDR_MASK - 1))
+#define FBNIC_BD_FRAG_COUNT \
+	(PAGE_SIZE / FBNIC_BD_FRAG_SIZE)
+#define FBNIC_BD_FRAG_ADDR_MASK \
+	(FBNIC_BD_DESC_ADDR_MASK & \
+	 ~(FBNIC_BD_DESC_ADDR_MASK * FBNIC_BD_FRAG_COUNT))
+#define FBNIC_BD_FRAG_ID_MASK \
+	(FBNIC_BD_DESC_ID_MASK & \
+	 ~(FBNIC_BD_DESC_ID_MASK * FBNIC_BD_FRAG_COUNT))
+#define FBNIC_BD_PAGE_ADDR_MASK \
+	(FBNIC_BD_DESC_ADDR_MASK & ~FBNIC_BD_FRAG_ADDR_MASK)
+#define FBNIC_BD_PAGE_ID_MASK \
+	(FBNIC_BD_DESC_ID_MASK & ~FBNIC_BD_FRAG_ID_MASK)
+
 /* Register Definitions
  *
  * The registers are laid as indexes into an le32 array. As such the actual
@@ -124,14 +155,14 @@ enum {
 
 /* Global QM Rx registers */
 #define FBNIC_CSR_START_QM_RX		0x00c00	/* CSR section delimiter */
-#define FBNIC_QM_RCQ_IDLE(n)		(0x00c00 + (n)) /* 0x03000 + 0x4*n */
+#define FBNIC_QM_RCQ_IDLE(n)		(0x00c00 + (n)) /* 0x03000 + 4*n */
 #define FBNIC_QM_RCQ_IDLE_CNT			4
 #define FBNIC_QM_RCQ_CTL0		0x00c0c		/* 0x03030 */
 #define FBNIC_QM_RCQ_CTL0_COAL_WAIT		CSR_GENMASK(15, 0)
 #define FBNIC_QM_RCQ_CTL0_TICK_CYCLES		CSR_GENMASK(26, 16)
-#define FBNIC_QM_HPQ_IDLE(n)		(0x00c0f + (n)) /* 0x0303c + 0x4*n */
+#define FBNIC_QM_HPQ_IDLE(n)		(0x00c0f + (n)) /* 0x0303c + 4*n */
 #define FBNIC_QM_HPQ_IDLE_CNT			4
-#define FBNIC_QM_PPQ_IDLE(n)		(0x00c13 + (n)) /* 0x0304c + 0x4*n */
+#define FBNIC_QM_PPQ_IDLE(n)		(0x00c13 + (n)) /* 0x0304c + 4*n */
 #define FBNIC_QM_PPQ_IDLE_CNT			4
 #define FBNIC_QM_RNI_RBP_CTL		0x00c2d		/* 0x030b4 */
 #define FBNIC_QM_RNI_RBP_CTL_MRRS		CSR_GENMASK(1, 0)
@@ -451,12 +482,78 @@ enum {
 #define FBNIC_QUEUE_TIM_COUNTS_CNT0_MASK	CSR_GENMASK(14, 0)
 
 /* Rx Completion Queue Registers */
+#define FBNIC_QUEUE_RCQ_CTL		0x200		/* 0x800 */
+#define FBNIC_QUEUE_RCQ_CTL_RESET		CSR_BIT(0)
+#define FBNIC_QUEUE_RCQ_CTL_ENABLE		CSR_BIT(1)
+
 #define FBNIC_QUEUE_RCQ_HEAD		0x201		/* 0x804 */
 
+#define FBNIC_QUEUE_RCQ_SIZE		0x204		/* 0x810 */
+#define FBNIC_QUEUE_RCQ_SIZE_MASK		CSR_GENMASK(3, 0)
+
+#define FBNIC_QUEUE_RCQ_BAL		0x220		/* 0x880 */
+#define FBNIC_QUEUE_RCQ_BAH		0x221		/* 0x884 */
+
 /* Rx Buffer Descriptor Queue Registers */
+#define FBNIC_QUEUE_BDQ_CTL		0x240		/* 0x900 */
+#define FBNIC_QUEUE_BDQ_CTL_RESET		CSR_BIT(0)
+#define FBNIC_QUEUE_BDQ_CTL_ENABLE		CSR_BIT(1)
+#define FBNIC_QUEUE_BDQ_CTL_PPQ_ENABLE		CSR_BIT(30)
+
 #define FBNIC_QUEUE_BDQ_HPQ_TAIL	0x241		/* 0x904 */
 #define FBNIC_QUEUE_BDQ_PPQ_TAIL	0x242		/* 0x908 */
 
+#define FBNIC_QUEUE_BDQ_HPQ_SIZE	0x247		/* 0x91c */
+#define FBNIC_QUEUE_BDQ_PPQ_SIZE	0x248		/* 0x920 */
+#define FBNIC_QUEUE_BDQ_SIZE_MASK		CSR_GENMASK(3, 0)
+
+#define FBNIC_QUEUE_BDQ_HPQ_BAL		0x260		/* 0x980 */
+#define FBNIC_QUEUE_BDQ_HPQ_BAH		0x261		/* 0x984 */
+#define FBNIC_QUEUE_BDQ_PPQ_BAL		0x262		/* 0x988 */
+#define FBNIC_QUEUE_BDQ_PPQ_BAH		0x263		/* 0x98c */
+
+/* Rx DMA Engine Configuration */
+#define FBNIC_QUEUE_RDE_CTL0		0x2a0		/* 0xa80 */
+#define FBNIC_QUEUE_RDE_CTL0_EN_HDR_SPLIT	CSR_BIT(31)
+#define FBNIC_QUEUE_RDE_CTL0_DROP_MODE_MASK	CSR_GENMASK(30, 29)
+enum {
+	FBNIC_QUEUE_RDE_CTL0_DROP_IMMEDIATE	= 0,
+	FBNIC_QUEUE_RDE_CTL0_DROP_WAIT		= 1,
+	FBNIC_QUEUE_RDE_CTL0_DROP_NEVER		= 2,
+};
+
+#define FBNIC_QUEUE_RDE_CTL0_MIN_HROOM_MASK	CSR_GENMASK(28, 20)
+#define FBNIC_QUEUE_RDE_CTL0_MIN_TROOM_MASK	CSR_GENMASK(19, 11)
+
+#define FBNIC_QUEUE_RDE_CTL1		0x2a1		/* 0xa84 */
+#define FBNIC_QUEUE_RDE_CTL1_MAX_HDR_MASK	CSR_GENMASK(24, 12)
+#define FBNIC_QUEUE_RDE_CTL1_PAYLD_OFF_MASK	CSR_GENMASK(11, 9)
+#define FBNIC_QUEUE_RDE_CTL1_PAYLD_PG_CL_MASK	CSR_GENMASK(8, 6)
+#define FBNIC_QUEUE_RDE_CTL1_PADLEN_MASK	CSR_GENMASK(5, 2)
+#define FBNIC_QUEUE_RDE_CTL1_PAYLD_PACK_MASK	CSR_GENMASK(1, 0)
+enum {
+	FBNIC_QUEUE_RDE_CTL1_PAYLD_PACK_NONE	= 0,
+	FBNIC_QUEUE_RDE_CTL1_PAYLD_PACK_ALL	= 1,
+	FBNIC_QUEUE_RDE_CTL1_PAYLD_PACK_RSS	= 2,
+};
+
+/* Rx Interrupt Manager Registers */
+#define FBNIC_QUEUE_RIM_CTL		0x2c0		/* 0xb00 */
+#define FBNIC_QUEUE_RIM_CTL_MSIX_MASK		CSR_GENMASK(7, 0)
+
+#define FBNIC_QUEUE_RIM_THRESHOLD	0x2c1		/* 0xb04 */
+#define FBNIC_QUEUE_RIM_THRESHOLD_RCD_MASK	CSR_GENMASK(14, 0)
+
+#define FBNIC_QUEUE_RIM_CLEAR		0x2c2		/* 0xb08 */
+#define FBNIC_QUEUE_RIM_CLEAR_MASK		CSR_BIT(0)
+#define FBNIC_QUEUE_RIM_SET		0x2c3		/* 0xb0c */
+#define FBNIC_QUEUE_RIM_SET_MASK		CSR_BIT(0)
+#define FBNIC_QUEUE_RIM_MASK		0x2c4		/* 0xb10 */
+#define FBNIC_QUEUE_RIM_MASK_MASK		CSR_BIT(0)
+
+#define FBNIC_QUEUE_RIM_COAL_STATUS	0x2c5		/* 0xb14 */
+#define FBNIC_QUEUE_RIM_RCD_COUNT_MASK		CSR_GENMASK(30, 16)
+#define FBNIC_QUEUE_RIM_TIMER_MASK		CSR_GENMASK(13, 0)
 #define FBNIC_MAX_QUEUES		128
 #define FBNIC_CSR_END_QUEUE	(0x40000 + 0x400 * FBNIC_MAX_QUEUES - 1)
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
index c7e9bd1e79c6..f5a5076d0f52 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
@@ -131,6 +131,9 @@ struct net_device *fbnic_netdev_alloc(struct fbnic_dev *fbd)
 	INIT_LIST_HEAD(&fbn->napis);
 
 	fbn->txq_size = FBNIC_TXQ_SIZE_DEFAULT;
+	fbn->hpq_size = FBNIC_HPQ_SIZE_DEFAULT;
+	fbn->ppq_size = FBNIC_PPQ_SIZE_DEFAULT;
+	fbn->rcq_size = FBNIC_RCQ_SIZE_DEFAULT;
 
 	default_queues = netif_get_num_default_rss_queues();
 	if (default_queues > fbd->max_num_queues)
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
index b3c39c10c3f7..18f93e9431cc 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
@@ -16,6 +16,9 @@ struct fbnic_net {
 	struct fbnic_dev *fbd;
 
 	u32 txq_size;
+	u32 hpq_size;
+	u32 ppq_size;
+	u32 rcq_size;
 
 	u16 num_napi;
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
index 08b74ed561b7..325b20b232eb 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -129,6 +129,8 @@ void fbnic_up(struct fbnic_net *fbn)
 {
 	fbnic_enable(fbn);
 
+	fbnic_fill(fbn);
+
 	/* Enable Tx/Rx processing */
 	fbnic_napi_enable(fbn);
 	netif_tx_start_all_queues(fbn->netdev);
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
index 5d836d5a8370..5fdc0a404f5a 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
@@ -3,6 +3,8 @@
 
 #include <linux/iopoll.h>
 #include <linux/pci.h>
+#include <net/netdev_queues.h>
+#include <net/page_pool/helpers.h>
 
 #include "fbnic.h"
 #include "fbnic_netdev.h"
@@ -31,12 +33,134 @@ static void fbnic_ring_wr32(struct fbnic_ring *ring, unsigned int csr, u32 val)
 	writel(val, csr_base + csr);
 }
 
+static unsigned int fbnic_desc_unused(struct fbnic_ring *ring)
+{
+	return (ring->head - ring->tail - 1) & ring->size_mask;
+}
+
 netdev_tx_t fbnic_xmit_frame(struct sk_buff *skb, struct net_device *dev)
 {
 	dev_kfree_skb_any(skb);
 	return NETDEV_TX_OK;
 }
 
+static void fbnic_page_pool_init(struct fbnic_ring *ring, unsigned int idx,
+				 struct page *page)
+{
+	struct fbnic_rx_buf *rx_buf = &ring->rx_buf[idx];
+
+	page_pool_fragment_page(page, PAGECNT_BIAS_MAX);
+	rx_buf->pagecnt_bias = PAGECNT_BIAS_MAX;
+	rx_buf->page = page;
+}
+
+static void fbnic_page_pool_drain(struct fbnic_ring *ring, unsigned int idx,
+				  struct fbnic_napi_vector *nv, int budget)
+{
+	struct fbnic_rx_buf *rx_buf = &ring->rx_buf[idx];
+	struct page *page = rx_buf->page;
+
+	if (!page_pool_unref_page(page, rx_buf->pagecnt_bias))
+		page_pool_put_unrefed_page(nv->page_pool, page, -1, !!budget);
+
+	rx_buf->page = NULL;
+}
+
+static void fbnic_clean_bdq(struct fbnic_napi_vector *nv, int napi_budget,
+			    struct fbnic_ring *ring, unsigned int hw_head)
+{
+	unsigned int head = ring->head;
+
+	if (head == hw_head)
+		return;
+
+	do {
+		fbnic_page_pool_drain(ring, head, nv, napi_budget);
+
+		head++;
+		head &= ring->size_mask;
+	} while (head != hw_head);
+
+	ring->head = head;
+}
+
+static void fbnic_bd_prep(struct fbnic_ring *bdq, u16 id, struct page *page)
+{
+	__le64 *bdq_desc = &bdq->desc[id * FBNIC_BD_FRAG_COUNT];
+	dma_addr_t dma = page_pool_get_dma_addr(page);
+	u64 bd, i = FBNIC_BD_FRAG_COUNT;
+
+	bd = (FBNIC_BD_PAGE_ADDR_MASK & dma) |
+	     FIELD_PREP(FBNIC_BD_PAGE_ID_MASK, id);
+
+	/* In the case that a page size is larger than 4K we will map a
+	 * single page to multiple fragments. The fragments will be
+	 * FBNIC_BD_FRAG_COUNT in size and the lower n bits will be use
+	 * to indicate the individual fragment IDs.
+	 */
+	do {
+		*bdq_desc = cpu_to_le64(bd);
+		bd += FIELD_PREP(FBNIC_BD_DESC_ADDR_MASK, 1) |
+		      FIELD_PREP(FBNIC_BD_DESC_ID_MASK, 1);
+	} while (--i);
+}
+
+static void fbnic_fill_bdq(struct fbnic_napi_vector *nv, struct fbnic_ring *bdq)
+{
+	unsigned int count = fbnic_desc_unused(bdq);
+	unsigned int i = bdq->tail;
+
+	if (!count)
+		return;
+
+	do {
+		struct page *page;
+
+		page = page_pool_dev_alloc_pages(nv->page_pool);
+		if (!page)
+			break;
+
+		fbnic_page_pool_init(bdq, i, page);
+		fbnic_bd_prep(bdq, i, page);
+
+		i++;
+		i &= bdq->size_mask;
+
+		count--;
+	} while (count);
+
+	if (bdq->tail != i) {
+		bdq->tail = i;
+
+		/* Force DMA writes to flush before writing to tail */
+		dma_wmb();
+
+		writel(i, bdq->doorbell);
+	}
+}
+
+static void fbnic_put_pkt_buff(struct fbnic_napi_vector *nv,
+			       struct fbnic_pkt_buff *pkt, int budget)
+{
+	struct skb_shared_info *shinfo;
+	struct page *page;
+	int nr_frags;
+
+	if (!pkt->buff.data_hard_start)
+		return;
+
+	shinfo = xdp_get_shared_info_from_buff(&pkt->buff);
+	nr_frags = pkt->nr_frags;
+
+	while (nr_frags--) {
+		page = skb_frag_page(&shinfo->frags[nr_frags]);
+		page_pool_put_full_page(nv->page_pool, page, !!budget);
+	}
+
+	page = virt_to_page(pkt->buff.data_hard_start);
+	page_pool_put_full_page(nv->page_pool, page, !!budget);
+}
+
 static void fbnic_nv_irq_disable(struct fbnic_napi_vector *nv)
 {
 	struct fbnic_dev *fbd = nv->fbd;
@@ -100,6 +224,7 @@ static void fbnic_free_napi_vector(struct fbnic_net *fbn,
 	}
 
 	fbnic_free_irq(fbd, v_idx, nv);
+	page_pool_destroy(nv->page_pool);
 	netif_napi_del(&nv->napi);
 	list_del(&nv->napis);
 	kfree(nv);
@@ -125,6 +250,45 @@ static void fbnic_name_napi_vector(struct fbnic_napi_vector *nv)
 			 nv->v_idx - FBNIC_NON_NAPI_VECTORS);
 }
 
+#define FBNIC_PAGE_POOL_FLAGS \
+	(PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV)
+
+static int fbnic_alloc_nv_page_pool(struct fbnic_net *fbn,
+				    struct fbnic_napi_vector *nv)
+{
+	struct page_pool_params pp_params = {
+		.order = 0,
+		.flags = FBNIC_PAGE_POOL_FLAGS,
+		.pool_size = (fbn->hpq_size + fbn->ppq_size) * nv->rxt_count,
+		.nid = NUMA_NO_NODE,
+		.dev = nv->dev,
+		.dma_dir = DMA_BIDIRECTIONAL,
+		.offset = 0,
+		.max_len = PAGE_SIZE
+	};
+	struct page_pool *pp;
+
+	/* Page pool cannot exceed a size of 32768. This doesn't limit the
+	 * pages on the ring but the number we can have cached waiting on
+	 * the next use.
+	 *
+	 * TBD: Can this be reduced further? Would a multiple of
+	 * NAPI_POLL_WEIGHT possibly make more sense? The question is how
+	 * may pages do we need to hold in reserve to get the best return
+	 * without hogging too much system memory.
+	 */
+	if (pp_params.pool_size > 32768)
+		pp_params.pool_size = 32768;
+
+	pp = page_pool_create(&pp_params);
+	if (IS_ERR(pp))
+		return PTR_ERR(pp);
+
+	nv->page_pool = pp;
+
+	return 0;
+}
+
 static void fbnic_ring_init(struct fbnic_ring *ring, u32 __iomem *doorbell,
 			    int q_idx, u8 flags)
 {
@@ -177,6 +341,13 @@ static int fbnic_alloc_napi_vector(struct fbnic_dev *fbd, struct fbnic_net *fbn,
 	/* Tie nv back to PCIe dev */
 	nv->dev = fbd->dev;
 
+	/* Allocate page pool */
+	if (rxq_count) {
+		err = fbnic_alloc_nv_page_pool(fbn, nv);
+		if (err)
+			goto napi_del;
+	}
+
 	/* Initialize vector name */
 	fbnic_name_napi_vector(nv);
 
@@ -184,7 +355,7 @@ static int fbnic_alloc_napi_vector(struct fbnic_dev *fbd, struct fbnic_net *fbn,
 	err = fbnic_request_irq(fbd, v_idx, &fbnic_msix_clean_rings,
 				IRQF_SHARED, nv->name, nv);
 	if (err)
-		goto napi_del;
+		goto pp_destroy;
 
 	/* Initialize queue triads */
 	qt = nv->qt;
@@ -241,6 +412,8 @@ static int fbnic_alloc_napi_vector(struct fbnic_dev *fbd, struct fbnic_net *fbn,
 
 	return 0;
 
+pp_destroy:
+	page_pool_destroy(nv->page_pool);
 napi_del:
 	netif_napi_del(&nv->napi);
 	list_del(&nv->napis);
@@ -374,6 +547,80 @@ free_desc:
 	return err;
 }
 
+static int fbnic_alloc_rx_ring_desc(struct fbnic_net *fbn,
+				    struct fbnic_ring *rxr)
+{
+	struct device *dev = fbn->netdev->dev.parent;
+	size_t desc_size = sizeof(*rxr->desc);
+	u32 rxq_size;
+	size_t size;
+
+	switch (rxr->doorbell - fbnic_ring_csr_base(rxr)) {
+	case FBNIC_QUEUE_BDQ_HPQ_TAIL:
+		rxq_size = fbn->hpq_size / FBNIC_BD_FRAG_COUNT;
+		desc_size *= FBNIC_BD_FRAG_COUNT;
+		break;
+	case FBNIC_QUEUE_BDQ_PPQ_TAIL:
+		rxq_size = fbn->ppq_size / FBNIC_BD_FRAG_COUNT;
+		desc_size *= FBNIC_BD_FRAG_COUNT;
+		break;
+	case FBNIC_QUEUE_RCQ_HEAD:
+		rxq_size = fbn->rcq_size;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* Round size up to nearest 4K */
+	size = ALIGN(array_size(desc_size, rxq_size), 4096);
+
+	rxr->desc = dma_alloc_coherent(dev, size, &rxr->dma,
+				       GFP_KERNEL | __GFP_NOWARN);
+	if (!rxr->desc)
+		return -ENOMEM;
+
+	/* rxq_size should be a power of 2, so mask is just that -1 */
+	rxr->size_mask = rxq_size - 1;
+	rxr->size = size;
+
+	return 0;
+}
+
+static int fbnic_alloc_rx_ring_buffer(struct fbnic_ring *rxr)
+{
+	size_t size = array_size(sizeof(*rxr->rx_buf), rxr->size_mask + 1);
+
+	if (rxr->flags & FBNIC_RING_F_CTX)
+		size = sizeof(*rxr->rx_buf) * (rxr->size_mask + 1);
+	else
+		size = sizeof(*rxr->pkt);
+
+	rxr->rx_buf = kvzalloc(size, GFP_KERNEL | __GFP_NOWARN);
+
+	return rxr->rx_buf ? 0 : -ENOMEM;
+}
+
+static int fbnic_alloc_rx_ring_resources(struct fbnic_net *fbn,
+					 struct fbnic_ring *rxr)
+{
+	struct device *dev = fbn->netdev->dev.parent;
+	int err;
+
+	err = fbnic_alloc_rx_ring_desc(fbn, rxr);
+	if (err)
+		return err;
+
+	err = fbnic_alloc_rx_ring_buffer(rxr);
+	if (err)
+		goto free_desc;
+
+	return 0;
+
+free_desc:
+	fbnic_free_ring_resources(dev, rxr);
+	return err;
+}
+
 static void fbnic_free_qt_resources(struct fbnic_net *fbn,
 				    struct fbnic_q_triad *qt)
 {
@@ -395,11 +642,38 @@ static int fbnic_alloc_tx_qt_resources(struct fbnic_net *fbn,
 		return err;
 
 	err = fbnic_alloc_tx_ring_resources(fbn, &qt->cmpl);
+	if (err)
+		goto free_sub1;
+
+	return 0;
+
+free_sub1:
+	fbnic_free_ring_resources(dev, &qt->sub0);
+	return err;
+}
+
+static int fbnic_alloc_rx_qt_resources(struct fbnic_net *fbn,
+				       struct fbnic_q_triad *qt)
+{
+	struct device *dev = fbn->netdev->dev.parent;
+	int err;
+
+	err = fbnic_alloc_rx_ring_resources(fbn, &qt->sub0);
+	if (err)
+		return err;
+
+	err = fbnic_alloc_rx_ring_resources(fbn, &qt->sub1);
 	if (err)
 		goto free_sub0;
 
+	err = fbnic_alloc_rx_ring_resources(fbn, &qt->cmpl);
+	if (err)
+		goto free_sub1;
+
 	return 0;
 
+free_sub1:
+	fbnic_free_ring_resources(dev, &qt->sub1);
 free_sub0:
 	fbnic_free_ring_resources(dev, &qt->sub0);
 	return err;
@@ -408,17 +682,20 @@ free_sub0:
 static void fbnic_free_nv_resources(struct fbnic_net *fbn,
 				    struct fbnic_napi_vector *nv)
 {
-	int i;
+	int i, j;
 
 	/* Free Tx Resources  */
 	for (i = 0; i < nv->txt_count; i++)
 		fbnic_free_qt_resources(fbn, &nv->qt[i]);
+
+	for (j = 0; j < nv->rxt_count; j++, i++)
+		fbnic_free_qt_resources(fbn, &nv->qt[i]);
 }
 
 static int fbnic_alloc_nv_resources(struct fbnic_net *fbn,
 				    struct fbnic_napi_vector *nv)
 {
-	int i, err;
+	int i, j, err;
 
 	/* Allocate Tx Resources */
 	for (i = 0; i < nv->txt_count; i++) {
@@ -427,6 +704,13 @@ static int fbnic_alloc_nv_resources(struct fbnic_net *fbn,
 			goto free_resources;
 	}
 
+	/* Allocate Rx Resources */
+	for (j = 0; j < nv->rxt_count; j++, i++) {
+		err = fbnic_alloc_rx_qt_resources(fbn, &nv->qt[i]);
+		if (err)
+			goto free_resources;
+	}
+
 	return 0;
 
 free_resources:
@@ -478,6 +762,21 @@ static void fbnic_disable_tcq(struct fbnic_ring *txr)
 	fbnic_ring_wr32(txr, FBNIC_QUEUE_TIM_MASK, FBNIC_QUEUE_TIM_MASK_MASK);
 }
 
+static void fbnic_disable_bdq(struct fbnic_ring *hpq, struct fbnic_ring *ppq)
+{
+	u32 bdq_ctl = fbnic_ring_rd32(hpq, FBNIC_QUEUE_BDQ_CTL);
+
+	bdq_ctl &= ~FBNIC_QUEUE_BDQ_CTL_ENABLE;
+
+	fbnic_ring_wr32(hpq, FBNIC_QUEUE_BDQ_CTL, bdq_ctl);
+}
+
+static void fbnic_disable_rcq(struct fbnic_ring *rxr)
+{
+	fbnic_ring_wr32(rxr, FBNIC_QUEUE_RCQ_CTL, 0);
+	fbnic_ring_wr32(rxr, FBNIC_QUEUE_RIM_MASK, FBNIC_QUEUE_RIM_MASK_MASK);
+}
+
 void fbnic_napi_disable(struct fbnic_net *fbn)
 {
 	struct fbnic_napi_vector *nv;
@@ -493,7 +792,7 @@ void fbnic_disable(struct fbnic_net *fbn)
 {
 	struct fbnic_dev *fbd = fbn->fbd;
 	struct fbnic_napi_vector *nv;
-	int i;
+	int i, j;
 
 	list_for_each_entry(nv, &fbn->napis, napis) {
 		/* Disable Tx queue triads */
@@ -503,6 +802,14 @@ void fbnic_disable(struct fbnic_net *fbn)
 			fbnic_disable_twq0(&qt->sub0);
 			fbnic_disable_tcq(&qt->cmpl);
 		}
+
+		/* Disable Rx queue triads */
+		for (j = 0; j < nv->rxt_count; j++, i++) {
+			struct fbnic_q_triad *qt = &nv->qt[i];
+
+			fbnic_disable_bdq(&qt->sub0, &qt->sub1);
+			fbnic_disable_rcq(&qt->cmpl);
+		}
 	}
 
 	fbnic_wrfl(fbd);
@@ -562,6 +869,10 @@ int fbnic_wait_all_queues_idle(struct fbnic_dev *fbd, bool may_fail)
 		{ FBNIC_QM_TQS_IDLE(0),	FBNIC_QM_TQS_IDLE_CNT, },
 		{ FBNIC_QM_TDE_IDLE(0),	FBNIC_QM_TDE_IDLE_CNT, },
 		{ FBNIC_QM_TCQ_IDLE(0),	FBNIC_QM_TCQ_IDLE_CNT, },
+	}, rx[] = {
+		{ FBNIC_QM_HPQ_IDLE(0),	FBNIC_QM_HPQ_IDLE_CNT, },
+		{ FBNIC_QM_PPQ_IDLE(0),	FBNIC_QM_PPQ_IDLE_CNT, },
+		{ FBNIC_QM_RCQ_IDLE(0),	FBNIC_QM_RCQ_IDLE_CNT, },
 	};
 	bool idle;
 	int err;
@@ -581,6 +892,10 @@ int fbnic_wait_all_queues_idle(struct fbnic_dev *fbd, bool may_fail)
 			return err;
 	}
 
+	err = read_poll_timeout_atomic(fbnic_all_idle, idle, idle, 2, 500000,
+				       false, fbd, rx, ARRAY_SIZE(rx));
+	if (err)
+		fbnic_idle_dump(fbd, rx, ARRAY_SIZE(rx), "Rx", err);
 	return err;
 }
 
@@ -589,7 +904,7 @@ void fbnic_flush(struct fbnic_net *fbn)
 	struct fbnic_napi_vector *nv;
 
 	list_for_each_entry(nv, &fbn->napis, napis) {
-		int i;
+		int i, j;
 
 		/* Flush any processed Tx Queue Triads and drop the rest */
 		for (i = 0; i < nv->txt_count; i++) {
@@ -612,6 +927,62 @@ void fbnic_flush(struct fbnic_net *fbn)
 			netif_queue_set_napi(nv->napi.dev, qt->sub0.q_idx,
 					     NETDEV_QUEUE_TYPE_TX, NULL);
 		}
+
+		/* Flush any processed Rx Queue Triads and drop the rest */
+		for (j = 0; j < nv->rxt_count; j++, i++) {
+			struct fbnic_q_triad *qt = &nv->qt[i];
+
+			/* Clean the work queues of unprocessed work */
+			fbnic_clean_bdq(nv, 0, &qt->sub0, qt->sub0.tail);
+			fbnic_clean_bdq(nv, 0, &qt->sub1, qt->sub1.tail);
+
+			/* Reset completion queue descriptor ring */
+			memset(qt->cmpl.desc, 0, qt->cmpl.size);
+
+			fbnic_put_pkt_buff(nv, qt->cmpl.pkt, 0);
+			qt->cmpl.pkt->buff.data_hard_start = NULL;
+
+			/* Disassociate Rx queue from NAPI */
+			netif_queue_set_napi(nv->napi.dev, qt->cmpl.q_idx,
+					     NETDEV_QUEUE_TYPE_RX, NULL);
+		}
+	}
+}
+
+void fbnic_fill(struct fbnic_net *fbn)
+{
+	struct fbnic_napi_vector *nv;
+
+	list_for_each_entry(nv, &fbn->napis, napis) {
+		int i, j;
+
+		/* Configure NAPI mapping for Tx */
+		for (i = 0; i < nv->txt_count; i++) {
+			struct fbnic_q_triad *qt = &nv->qt[i];
+
+			/* Nothing to do if Tx queue is disabled */
+			if (qt->sub0.flags & FBNIC_RING_F_DISABLED)
+				continue;
+
+			/* Associate Tx queue with NAPI */
+			netif_queue_set_napi(nv->napi.dev, qt->sub0.q_idx,
+					     NETDEV_QUEUE_TYPE_TX, &nv->napi);
+		}
+
+		/* Configure NAPI mapping and populate pages
+		 * in the BDQ rings to use for Rx
+		 */
+		for (j = 0; j < nv->rxt_count; j++, i++) {
+			struct fbnic_q_triad *qt = &nv->qt[i];
+
+			/* Associate Rx queue with NAPI */
+			netif_queue_set_napi(nv->napi.dev, qt->cmpl.q_idx,
+					     NETDEV_QUEUE_TYPE_RX, &nv->napi);
+
+			/* Populate the header and payload BDQs */
+			fbnic_fill_bdq(nv, &qt->sub0);
+			fbnic_fill_bdq(nv, &qt->sub1);
+		}
 	}
 }
 
@@ -666,11 +1037,102 @@ static void fbnic_enable_tcq(struct fbnic_napi_vector *nv,
 	fbnic_ring_wr32(tcq, FBNIC_QUEUE_TCQ_CTL, FBNIC_QUEUE_TCQ_CTL_ENABLE);
 }
 
+static void fbnic_enable_bdq(struct fbnic_ring *hpq, struct fbnic_ring *ppq)
+{
+	u32 bdq_ctl = FBNIC_QUEUE_BDQ_CTL_ENABLE;
+	u32 log_size;
+
+	/* Reset head/tail */
+	fbnic_ring_wr32(hpq, FBNIC_QUEUE_BDQ_CTL, FBNIC_QUEUE_BDQ_CTL_RESET);
+	ppq->tail = 0;
+	ppq->head = 0;
+	hpq->tail = 0;
+	hpq->head = 0;
+
+	log_size = fls(hpq->size_mask);
+
+	/* Store descriptor ring address and size */
+	fbnic_ring_wr32(hpq, FBNIC_QUEUE_BDQ_HPQ_BAL, lower_32_bits(hpq->dma));
+	fbnic_ring_wr32(hpq, FBNIC_QUEUE_BDQ_HPQ_BAH, upper_32_bits(hpq->dma));
+
+	/* Write lower 4 bits of log size as 64K ring size is 0 */
+	fbnic_ring_wr32(hpq, FBNIC_QUEUE_BDQ_HPQ_SIZE, log_size & 0xf);
+
+	if (!ppq->size_mask)
+		goto write_ctl;
+
+	log_size = fls(ppq->size_mask);
+
+	/* Add enabling of PPQ to BDQ control */
+	bdq_ctl |= FBNIC_QUEUE_BDQ_CTL_PPQ_ENABLE;
+
+	/* Store descriptor ring address and size */
+	fbnic_ring_wr32(ppq, FBNIC_QUEUE_BDQ_PPQ_BAL, lower_32_bits(ppq->dma));
+	fbnic_ring_wr32(ppq, FBNIC_QUEUE_BDQ_PPQ_BAH, upper_32_bits(ppq->dma));
+	fbnic_ring_wr32(ppq, FBNIC_QUEUE_BDQ_PPQ_SIZE, log_size & 0xf);
+
+write_ctl:
+	fbnic_ring_wr32(hpq, FBNIC_QUEUE_BDQ_CTL, bdq_ctl);
+}
+
+static void fbnic_config_drop_mode_rcq(struct fbnic_napi_vector *nv,
+				       struct fbnic_ring *rcq)
+{
+	u32 drop_mode, rcq_ctl;
+
+	drop_mode = FBNIC_QUEUE_RDE_CTL0_DROP_IMMEDIATE;
+
+	/* Specify packet layout */
+	rcq_ctl = FIELD_PREP(FBNIC_QUEUE_RDE_CTL0_DROP_MODE_MASK, drop_mode) |
+	    FIELD_PREP(FBNIC_QUEUE_RDE_CTL0_MIN_HROOM_MASK, FBNIC_RX_HROOM) |
+	    FIELD_PREP(FBNIC_QUEUE_RDE_CTL0_MIN_TROOM_MASK, FBNIC_RX_TROOM);
+
+	fbnic_ring_wr32(rcq, FBNIC_QUEUE_RDE_CTL0, rcq_ctl);
+}
+
+static void fbnic_enable_rcq(struct fbnic_napi_vector *nv,
+			     struct fbnic_ring *rcq)
+{
+	u32 log_size = fls(rcq->size_mask);
+	u32 rcq_ctl;
+
+	fbnic_config_drop_mode_rcq(nv, rcq);
+
+	rcq_ctl = FIELD_PREP(FBNIC_QUEUE_RDE_CTL1_PADLEN_MASK, FBNIC_RX_PAD) |
+		   FIELD_PREP(FBNIC_QUEUE_RDE_CTL1_MAX_HDR_MASK,
+			      FBNIC_RX_MAX_HDR) |
+		   FIELD_PREP(FBNIC_QUEUE_RDE_CTL1_PAYLD_OFF_MASK,
+			      FBNIC_RX_PAYLD_OFFSET) |
+		   FIELD_PREP(FBNIC_QUEUE_RDE_CTL1_PAYLD_PG_CL_MASK,
+			      FBNIC_RX_PAYLD_PG_CL);
+	fbnic_ring_wr32(rcq, FBNIC_QUEUE_RDE_CTL1, rcq_ctl);
+
+	/* Reset head/tail */
+	fbnic_ring_wr32(rcq, FBNIC_QUEUE_RCQ_CTL, FBNIC_QUEUE_RCQ_CTL_RESET);
+	rcq->head = 0;
+	rcq->tail = 0;
+
+	/* Store descriptor ring address and size */
+	fbnic_ring_wr32(rcq, FBNIC_QUEUE_RCQ_BAL, lower_32_bits(rcq->dma));
+	fbnic_ring_wr32(rcq, FBNIC_QUEUE_RCQ_BAH, upper_32_bits(rcq->dma));
+
+	/* Write lower 4 bits of log size as 64K ring size is 0 */
+	fbnic_ring_wr32(rcq, FBNIC_QUEUE_RCQ_SIZE, log_size & 0xf);
+
+	/* Store interrupt information for the completion queue */
+	fbnic_ring_wr32(rcq, FBNIC_QUEUE_RIM_CTL, nv->v_idx);
+	fbnic_ring_wr32(rcq, FBNIC_QUEUE_RIM_THRESHOLD, rcq->size_mask / 2);
+	fbnic_ring_wr32(rcq, FBNIC_QUEUE_RIM_MASK, 0);
+
+	/* Enable queue */
+	fbnic_ring_wr32(rcq, FBNIC_QUEUE_RCQ_CTL, FBNIC_QUEUE_RCQ_CTL_ENABLE);
+}
+
 void fbnic_enable(struct fbnic_net *fbn)
 {
 	struct fbnic_dev *fbd = fbn->fbd;
 	struct fbnic_napi_vector *nv;
-	int i;
+	int i, j;
 
 	list_for_each_entry(nv, &fbn->napis, napis) {
 		/* Setup Tx Queue Triads */
@@ -680,6 +1142,15 @@ void fbnic_enable(struct fbnic_net *fbn)
 			fbnic_enable_twq0(&qt->sub0);
 			fbnic_enable_tcq(nv, &qt->cmpl);
 		}
+
+		/* Setup Rx Queue Triads */
+		for (j = 0; j < nv->rxt_count; j++, i++) {
+			struct fbnic_q_triad *qt = &nv->qt[i];
+
+			fbnic_enable_bdq(&qt->sub0, &qt->sub1);
+			fbnic_config_drop_mode_rcq(nv, &qt->cmpl);
+			fbnic_enable_rcq(nv, &qt->cmpl);
+		}
 	}
 
 	fbnic_wrfl(fbd);
@@ -697,11 +1168,31 @@ static void fbnic_nv_irq_enable(struct fbnic_napi_vector *nv)
 
 void fbnic_napi_enable(struct fbnic_net *fbn)
 {
+	u32 irqs[FBNIC_MAX_MSIX_VECS / 32] = {};
+	struct fbnic_dev *fbd = fbn->fbd;
 	struct fbnic_napi_vector *nv;
+	int i;
 
 	list_for_each_entry(nv, &fbn->napis, napis) {
 		napi_enable(&nv->napi);
 
 		fbnic_nv_irq_enable(nv);
+
+		/* Record bit used for NAPI IRQs so we can
+		 * set the mask appropriately
+		 */
+		irqs[nv->v_idx / 32] |= BIT(nv->v_idx % 32);
+	}
+
+	/* Force the first interrupt on the device to guarantee
+	 * that any packets that may have been enqueued during the
+	 * bringup are processed.
+	 */
+	for (i = 0; i < ARRAY_SIZE(irqs); i++) {
+		if (!irqs[i])
+			continue;
+		fbnic_wr32(fbd, FBNIC_INTR_SET(i), irqs[i]);
 	}
+
+	fbnic_wrfl(fbd);
 }
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
index 77abc15bb0dc..0e681dcc85c1 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
@@ -6,6 +6,7 @@
 
 #include <linux/netdevice.h>
 #include <linux/types.h>
+#include <net/xdp.h>
 
 struct fbnic_net;
 
@@ -13,14 +14,44 @@ struct fbnic_net;
 #define FBNIC_MAX_RXQS			128u
 
 #define FBNIC_TXQ_SIZE_DEFAULT		1024
+#define FBNIC_HPQ_SIZE_DEFAULT		256
+#define FBNIC_PPQ_SIZE_DEFAULT		256
+#define FBNIC_RCQ_SIZE_DEFAULT		1024
+
+#define FBNIC_RX_TROOM \
+	SKB_DATA_ALIGN(sizeof(struct skb_shared_info))
+#define FBNIC_RX_HROOM \
+	(ALIGN(FBNIC_RX_TROOM + NET_SKB_PAD, 128) - FBNIC_RX_TROOM)
+#define FBNIC_RX_PAD			0
+#define FBNIC_RX_MAX_HDR		(1536 - FBNIC_RX_PAD)
+#define FBNIC_RX_PAYLD_OFFSET		0
+#define FBNIC_RX_PAYLD_PG_CL		0
 
 #define FBNIC_RING_F_DISABLED		BIT(0)
 #define FBNIC_RING_F_CTX		BIT(1)
 #define FBNIC_RING_F_STATS		BIT(2)	/* Ring's stats may be used */
 
+struct fbnic_pkt_buff {
+	struct xdp_buff buff;
+	u32 data_truesize;
+	u16 data_len;
+	u16 nr_frags;
+};
+
+/* Pagecnt bias is long max to reserve the last bit to catch overflow
+ * cases where if we overcharge the bias it will flip over to be negative.
+ */
+#define PAGECNT_BIAS_MAX	LONG_MAX
+struct fbnic_rx_buf {
+	struct page *page;
+	long pagecnt_bias;
+};
+
 struct fbnic_ring {
 	/* Pointer to buffer specific info */
 	union {
+		struct fbnic_pkt_buff *pkt;	/* RCQ */
+		struct fbnic_rx_buf *rx_buf;	/* BDQ */
 		void **tx_buf;			/* TWQ */
 		void *buffer;			/* Generic pointer */
 	};
@@ -45,6 +76,7 @@ struct fbnic_q_triad {
 struct fbnic_napi_vector {
 	struct napi_struct napi;
 	struct device *dev;		/* Device for DMA unmapping */
+	struct page_pool *page_pool;
 	struct fbnic_dev *fbd;
 	char name[IFNAMSIZ + 9];
 
@@ -71,6 +103,7 @@ void fbnic_napi_disable(struct fbnic_net *fbn);
 void fbnic_enable(struct fbnic_net *fbn);
 void fbnic_disable(struct fbnic_net *fbn);
 void fbnic_flush(struct fbnic_net *fbn);
+void fbnic_fill(struct fbnic_net *fbn);
 
 int fbnic_wait_all_queues_idle(struct fbnic_dev *fbd, bool may_fail);
 
-- 
cgit v1.2.3-58-ga151


From 20d2e88cc7461b03b97293d62d780a2d4b08ed8b Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:49:54 -0700
Subject: eth: fbnic: Add initial messaging to notify FW of our presence

After the driver loads we need to get some initial capabilities from the
firmware to determine what the device is capable of and what functionality
needs to be enabled. Specifically we receive information about the current
state of the link and if a BMC is present.

After that when we bring the interface up we will need the ability to take
ownership from the FW. To do that we will need to notify it that we are
taking control before we start configuring the traffic classifier and MAC.

Once we have ownership we need to notify the firmware that we are still
present and active. To do that we will send a regular heartbeat to the FW.
If the FW doesn't receive the heartbeat in a timely fashion it will retake
control of the RPC and MAC and assume that the host has gone offline.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079939458.1778861.8966209942099133957.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/fbnic.h        |   5 +
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h    |   8 +
 drivers/net/ethernet/meta/fbnic/fbnic_fw.c     | 411 +++++++++++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_fw.h     |  85 +++++
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.c |  18 ++
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c    |  28 ++
 6 files changed, 555 insertions(+)

diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h
index 8abae3b8ef4d..44fe6bbf88a1 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic.h
@@ -26,9 +26,14 @@ struct fbnic_dev {
 	struct delayed_work service_task;
 
 	struct fbnic_fw_mbx mbx[FBNIC_IPC_MBX_INDICES];
+	struct fbnic_fw_cap fw_cap;
 	/* Lock protecting Tx Mailbox queue to prevent possible races */
 	spinlock_t fw_tx_lock;
 
+	unsigned long last_heartbeat_request;
+	unsigned long last_heartbeat_response;
+	u8 fw_heartbeat_enabled;
+
 	u64 dsn;
 	u32 mps;
 	u32 readrq;
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index be682fa78b2b..fb64ad919d31 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -12,6 +12,14 @@
 #define DESC_BIT(nr)		BIT_ULL(nr)
 #define DESC_GENMASK(h, l)	GENMASK_ULL(h, l)
 
+/* Defines the minimum firmware version required by the driver */
+#define MIN_FW_MAJOR_VERSION    0
+#define MIN_FW_MINOR_VERSION    10
+#define MIN_FW_BUILD_VERSION    6
+#define MIN_FW_VERSION_CODE     (MIN_FW_MAJOR_VERSION * (1u << 24) + \
+				 MIN_FW_MINOR_VERSION * (1u << 16) + \
+				 MIN_FW_BUILD_VERSION)
+
 #define PCI_DEVICE_ID_META_FBNIC_ASIC		0x0013
 
 #define FBNIC_CLOCK_FREQ	(600 * (1000 * 1000))
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c
index feca833ee924..0c6e1b4c119b 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.c
@@ -2,6 +2,7 @@
 /* Copyright (c) Meta Platforms, Inc. and affiliates. */
 
 #include <linux/bitfield.h>
+#include <linux/etherdevice.h>
 #include <linux/delay.h>
 #include <linux/dev_printk.h>
 #include <linux/dma-mapping.h>
@@ -190,6 +191,22 @@ static int fbnic_mbx_alloc_rx_msgs(struct fbnic_dev *fbd)
 	return err;
 }
 
+static int fbnic_mbx_map_tlv_msg(struct fbnic_dev *fbd,
+				 struct fbnic_tlv_msg *msg)
+{
+	unsigned long flags;
+	int err;
+
+	spin_lock_irqsave(&fbd->fw_tx_lock, flags);
+
+	err = fbnic_mbx_map_msg(fbd, FBNIC_IPC_MBX_TX_IDX, msg,
+				le16_to_cpu(msg->hdr.len) * sizeof(u32), 1);
+
+	spin_unlock_irqrestore(&fbd->fw_tx_lock, flags);
+
+	return err;
+}
+
 static void fbnic_mbx_process_tx_msgs(struct fbnic_dev *fbd)
 {
 	struct fbnic_fw_mbx *tx_mbx = &fbd->mbx[FBNIC_IPC_MBX_TX_IDX];
@@ -211,6 +228,61 @@ static void fbnic_mbx_process_tx_msgs(struct fbnic_dev *fbd)
 	tx_mbx->head = head;
 }
 
+/**
+ * fbnic_fw_xmit_simple_msg - Transmit a simple single TLV message w/o data
+ * @fbd: FBNIC device structure
+ * @msg_type: ENUM value indicating message type to send
+ *
+ * Return:
+ *   One the following values:
+ *     -EOPNOTSUPP: Is not ASIC so mailbox is not supported
+ *     -ENODEV: Device I/O error
+ *     -ENOMEM: Failed to allocate message
+ *     -EBUSY: No space in mailbox
+ *     -ENOSPC: DMA mapping failed
+ *
+ * This function sends a single TLV header indicating the host wants to take
+ * some action. However there are no other side effects which means that any
+ * response will need to be caught via a completion if this action is
+ * expected to kick off a resultant action.
+ */
+static int fbnic_fw_xmit_simple_msg(struct fbnic_dev *fbd, u32 msg_type)
+{
+	struct fbnic_tlv_msg *msg;
+	int err = 0;
+
+	if (!fbnic_fw_present(fbd))
+		return -ENODEV;
+
+	msg = fbnic_tlv_msg_alloc(msg_type);
+	if (!msg)
+		return -ENOMEM;
+
+	err = fbnic_mbx_map_tlv_msg(fbd, msg);
+	if (err)
+		free_page((unsigned long)msg);
+
+	return err;
+}
+
+/**
+ * fbnic_fw_xmit_cap_msg - Allocate and populate a FW capabilities message
+ * @fbd: FBNIC device structure
+ *
+ * Return: NULL on failure to allocate, error pointer on error, or pointer
+ * to new TLV test message.
+ *
+ * Sends a single TLV header indicating the host wants the firmware to
+ * confirm the capabilities and version.
+ **/
+static int fbnic_fw_xmit_cap_msg(struct fbnic_dev *fbd)
+{
+	int err = fbnic_fw_xmit_simple_msg(fbd, FBNIC_TLV_MSG_ID_HOST_CAP_REQ);
+
+	/* Return 0 if we are not calling this on ASIC */
+	return (err == -EOPNOTSUPP) ? 0 : err;
+}
+
 static void fbnic_mbx_postinit_desc_ring(struct fbnic_dev *fbd, int mbx_idx)
 {
 	struct fbnic_fw_mbx *mbx = &fbd->mbx[mbx_idx];
@@ -226,6 +298,16 @@ static void fbnic_mbx_postinit_desc_ring(struct fbnic_dev *fbd, int mbx_idx)
 		/* Make sure we have a page for the FW to write to */
 		fbnic_mbx_alloc_rx_msgs(fbd);
 		break;
+	case FBNIC_IPC_MBX_TX_IDX:
+		/* Force version to 1 if we successfully requested an update
+		 * from the firmware. This should be overwritten once we get
+		 * the actual version from the firmware in the capabilities
+		 * request message.
+		 */
+		if (!fbnic_fw_xmit_cap_msg(fbd) &&
+		    !fbd->fw_cap.running.mgmt.version)
+			fbd->fw_cap.running.mgmt.version = 1;
+		break;
 	}
 }
 
@@ -246,7 +328,336 @@ static void fbnic_mbx_postinit(struct fbnic_dev *fbd)
 		fbnic_mbx_postinit_desc_ring(fbd, i);
 }
 
+/**
+ * fbnic_fw_xmit_ownership_msg - Create and transmit a host ownership message
+ * to FW mailbox
+ *
+ * @fbd: FBNIC device structure
+ * @take_ownership: take/release the ownership
+ *
+ * Return: zero on success, negative value on failure
+ *
+ * Notifies the firmware that the driver either takes ownership of the NIC
+ * (when @take_ownership is true) or releases it.
+ */
+int fbnic_fw_xmit_ownership_msg(struct fbnic_dev *fbd, bool take_ownership)
+{
+	unsigned long req_time = jiffies;
+	struct fbnic_tlv_msg *msg;
+	int err = 0;
+
+	if (!fbnic_fw_present(fbd))
+		return -ENODEV;
+
+	msg = fbnic_tlv_msg_alloc(FBNIC_TLV_MSG_ID_OWNERSHIP_REQ);
+	if (!msg)
+		return -ENOMEM;
+
+	if (take_ownership) {
+		err = fbnic_tlv_attr_put_flag(msg, FBNIC_FW_OWNERSHIP_FLAG);
+		if (err)
+			goto free_message;
+	}
+
+	err = fbnic_mbx_map_tlv_msg(fbd, msg);
+	if (err)
+		goto free_message;
+
+	/* Initialize heartbeat, set last response to 1 second in the past
+	 * so that we will trigger a timeout if the firmware doesn't respond
+	 */
+	fbd->last_heartbeat_response = req_time - HZ;
+
+	fbd->last_heartbeat_request = req_time;
+
+	/* Set heartbeat detection based on if we are taking ownership */
+	fbd->fw_heartbeat_enabled = take_ownership;
+
+	return err;
+
+free_message:
+	free_page((unsigned long)msg);
+	return err;
+}
+
+static const struct fbnic_tlv_index fbnic_fw_cap_resp_index[] = {
+	FBNIC_TLV_ATTR_U32(FBNIC_FW_CAP_RESP_VERSION),
+	FBNIC_TLV_ATTR_FLAG(FBNIC_FW_CAP_RESP_BMC_PRESENT),
+	FBNIC_TLV_ATTR_MAC_ADDR(FBNIC_FW_CAP_RESP_BMC_MAC_ADDR),
+	FBNIC_TLV_ATTR_ARRAY(FBNIC_FW_CAP_RESP_BMC_MAC_ARRAY),
+	FBNIC_TLV_ATTR_U32(FBNIC_FW_CAP_RESP_STORED_VERSION),
+	FBNIC_TLV_ATTR_U32(FBNIC_FW_CAP_RESP_ACTIVE_FW_SLOT),
+	FBNIC_TLV_ATTR_STRING(FBNIC_FW_CAP_RESP_VERSION_COMMIT_STR,
+			      FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE),
+	FBNIC_TLV_ATTR_U32(FBNIC_FW_CAP_RESP_BMC_ALL_MULTI),
+	FBNIC_TLV_ATTR_U32(FBNIC_FW_CAP_RESP_FW_LINK_SPEED),
+	FBNIC_TLV_ATTR_U32(FBNIC_FW_CAP_RESP_FW_LINK_FEC),
+	FBNIC_TLV_ATTR_STRING(FBNIC_FW_CAP_RESP_STORED_COMMIT_STR,
+			      FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE),
+	FBNIC_TLV_ATTR_U32(FBNIC_FW_CAP_RESP_CMRT_VERSION),
+	FBNIC_TLV_ATTR_U32(FBNIC_FW_CAP_RESP_STORED_CMRT_VERSION),
+	FBNIC_TLV_ATTR_STRING(FBNIC_FW_CAP_RESP_CMRT_COMMIT_STR,
+			      FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE),
+	FBNIC_TLV_ATTR_STRING(FBNIC_FW_CAP_RESP_STORED_CMRT_COMMIT_STR,
+			      FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE),
+	FBNIC_TLV_ATTR_U32(FBNIC_FW_CAP_RESP_UEFI_VERSION),
+	FBNIC_TLV_ATTR_STRING(FBNIC_FW_CAP_RESP_UEFI_COMMIT_STR,
+			      FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE),
+	FBNIC_TLV_ATTR_LAST
+};
+
+static int fbnic_fw_parse_bmc_addrs(u8 bmc_mac_addr[][ETH_ALEN],
+				    struct fbnic_tlv_msg *attr, int len)
+{
+	int attr_len = le16_to_cpu(attr->hdr.len) / sizeof(u32) - 1;
+	struct fbnic_tlv_msg *mac_results[8];
+	int err, i = 0;
+
+	/* Make sure we have enough room to process all the MAC addresses */
+	if (len > 8)
+		return -ENOSPC;
+
+	/* Parse the array */
+	err = fbnic_tlv_attr_parse_array(&attr[1], attr_len, mac_results,
+					 fbnic_fw_cap_resp_index,
+					 FBNIC_FW_CAP_RESP_BMC_MAC_ADDR, len);
+	if (err)
+		return err;
+
+	/* Copy results into MAC addr array */
+	for (i = 0; i < len && mac_results[i]; i++)
+		fbnic_tlv_attr_addr_copy(bmc_mac_addr[i], mac_results[i]);
+
+	/* Zero remaining unused addresses */
+	while (i < len)
+		eth_zero_addr(bmc_mac_addr[i++]);
+
+	return 0;
+}
+
+static int fbnic_fw_parse_cap_resp(void *opaque, struct fbnic_tlv_msg **results)
+{
+	u32 active_slot = 0, all_multi = 0;
+	struct fbnic_dev *fbd = opaque;
+	u32 speed = 0, fec = 0;
+	size_t commit_size = 0;
+	bool bmc_present;
+	int err;
+
+	get_unsigned_result(FBNIC_FW_CAP_RESP_VERSION,
+			    fbd->fw_cap.running.mgmt.version);
+
+	if (!fbd->fw_cap.running.mgmt.version)
+		return -EINVAL;
+
+	if (fbd->fw_cap.running.mgmt.version < MIN_FW_VERSION_CODE) {
+		char running_ver[FBNIC_FW_VER_MAX_SIZE];
+
+		fbnic_mk_fw_ver_str(fbd->fw_cap.running.mgmt.version,
+				    running_ver);
+		dev_err(fbd->dev, "Device firmware version(%s) is older than minimum required version(%02d.%02d.%02d)\n",
+			running_ver,
+			MIN_FW_MAJOR_VERSION,
+			MIN_FW_MINOR_VERSION,
+			MIN_FW_BUILD_VERSION);
+		/* Disable TX mailbox to prevent card use until firmware is
+		 * updated.
+		 */
+		fbd->mbx[FBNIC_IPC_MBX_TX_IDX].ready = false;
+		return -EINVAL;
+	}
+
+	get_string_result(FBNIC_FW_CAP_RESP_VERSION_COMMIT_STR, commit_size,
+			  fbd->fw_cap.running.mgmt.commit,
+			  FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE);
+	if (!commit_size)
+		dev_warn(fbd->dev, "Firmware did not send mgmt commit!\n");
+
+	get_unsigned_result(FBNIC_FW_CAP_RESP_STORED_VERSION,
+			    fbd->fw_cap.stored.mgmt.version);
+	get_string_result(FBNIC_FW_CAP_RESP_STORED_COMMIT_STR, commit_size,
+			  fbd->fw_cap.stored.mgmt.commit,
+			  FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE);
+
+	get_unsigned_result(FBNIC_FW_CAP_RESP_CMRT_VERSION,
+			    fbd->fw_cap.running.bootloader.version);
+	get_string_result(FBNIC_FW_CAP_RESP_CMRT_COMMIT_STR, commit_size,
+			  fbd->fw_cap.running.bootloader.commit,
+			  FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE);
+
+	get_unsigned_result(FBNIC_FW_CAP_RESP_STORED_CMRT_VERSION,
+			    fbd->fw_cap.stored.bootloader.version);
+	get_string_result(FBNIC_FW_CAP_RESP_STORED_CMRT_COMMIT_STR, commit_size,
+			  fbd->fw_cap.stored.bootloader.commit,
+			  FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE);
+
+	get_unsigned_result(FBNIC_FW_CAP_RESP_UEFI_VERSION,
+			    fbd->fw_cap.stored.undi.version);
+	get_string_result(FBNIC_FW_CAP_RESP_UEFI_COMMIT_STR, commit_size,
+			  fbd->fw_cap.stored.undi.commit,
+			  FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE);
+
+	get_unsigned_result(FBNIC_FW_CAP_RESP_ACTIVE_FW_SLOT, active_slot);
+	fbd->fw_cap.active_slot = active_slot;
+
+	get_unsigned_result(FBNIC_FW_CAP_RESP_FW_LINK_SPEED, speed);
+	get_unsigned_result(FBNIC_FW_CAP_RESP_FW_LINK_FEC, fec);
+	fbd->fw_cap.link_speed = speed;
+	fbd->fw_cap.link_fec = fec;
+
+	bmc_present = !!results[FBNIC_FW_CAP_RESP_BMC_PRESENT];
+	if (bmc_present) {
+		struct fbnic_tlv_msg *attr;
+
+		attr = results[FBNIC_FW_CAP_RESP_BMC_MAC_ARRAY];
+		if (!attr)
+			return -EINVAL;
+
+		err = fbnic_fw_parse_bmc_addrs(fbd->fw_cap.bmc_mac_addr,
+					       attr, 4);
+		if (err)
+			return err;
+
+		get_unsigned_result(FBNIC_FW_CAP_RESP_BMC_ALL_MULTI, all_multi);
+	} else {
+		memset(fbd->fw_cap.bmc_mac_addr, 0,
+		       sizeof(fbd->fw_cap.bmc_mac_addr));
+	}
+
+	fbd->fw_cap.bmc_present = bmc_present;
+
+	if (results[FBNIC_FW_CAP_RESP_BMC_ALL_MULTI] || !bmc_present)
+		fbd->fw_cap.all_multi = all_multi;
+
+	return 0;
+}
+
+static const struct fbnic_tlv_index fbnic_ownership_resp_index[] = {
+	FBNIC_TLV_ATTR_LAST
+};
+
+static int fbnic_fw_parse_ownership_resp(void *opaque,
+					 struct fbnic_tlv_msg **results)
+{
+	struct fbnic_dev *fbd = (struct fbnic_dev *)opaque;
+
+	/* Count the ownership response as a heartbeat reply */
+	fbd->last_heartbeat_response = jiffies;
+
+	return 0;
+}
+
+static const struct fbnic_tlv_index fbnic_heartbeat_resp_index[] = {
+	FBNIC_TLV_ATTR_LAST
+};
+
+static int fbnic_fw_parse_heartbeat_resp(void *opaque,
+					 struct fbnic_tlv_msg **results)
+{
+	struct fbnic_dev *fbd = (struct fbnic_dev *)opaque;
+
+	fbd->last_heartbeat_response = jiffies;
+
+	return 0;
+}
+
+static int fbnic_fw_xmit_heartbeat_message(struct fbnic_dev *fbd)
+{
+	unsigned long req_time = jiffies;
+	struct fbnic_tlv_msg *msg;
+	int err = 0;
+
+	if (!fbnic_fw_present(fbd))
+		return -ENODEV;
+
+	msg = fbnic_tlv_msg_alloc(FBNIC_TLV_MSG_ID_HEARTBEAT_REQ);
+	if (!msg)
+		return -ENOMEM;
+
+	err = fbnic_mbx_map_tlv_msg(fbd, msg);
+	if (err)
+		goto free_message;
+
+	fbd->last_heartbeat_request = req_time;
+
+	return err;
+
+free_message:
+	free_page((unsigned long)msg);
+	return err;
+}
+
+static bool fbnic_fw_heartbeat_current(struct fbnic_dev *fbd)
+{
+	unsigned long last_response = fbd->last_heartbeat_response;
+	unsigned long last_request = fbd->last_heartbeat_request;
+
+	return !time_before(last_response, last_request);
+}
+
+int fbnic_fw_init_heartbeat(struct fbnic_dev *fbd, bool poll)
+{
+	int err = -ETIMEDOUT;
+	int attempts = 50;
+
+	if (!fbnic_fw_present(fbd))
+		return -ENODEV;
+
+	while (attempts--) {
+		msleep(200);
+		if (poll)
+			fbnic_mbx_poll(fbd);
+
+		if (!fbnic_fw_heartbeat_current(fbd))
+			continue;
+
+		/* Place new message on mailbox to elicit a response */
+		err = fbnic_fw_xmit_heartbeat_message(fbd);
+		if (err)
+			dev_warn(fbd->dev,
+				 "Failed to send heartbeat message: %d\n",
+				 err);
+		break;
+	}
+
+	return err;
+}
+
+void fbnic_fw_check_heartbeat(struct fbnic_dev *fbd)
+{
+	unsigned long last_request = fbd->last_heartbeat_request;
+	int err;
+
+	/* Do not check heartbeat or send another request until current
+	 * period has expired. Otherwise we might start spamming requests.
+	 */
+	if (time_is_after_jiffies(last_request + FW_HEARTBEAT_PERIOD))
+		return;
+
+	/* We already reported no mailbox. Wait for it to come back */
+	if (!fbd->fw_heartbeat_enabled)
+		return;
+
+	/* Was the last heartbeat response long time ago? */
+	if (!fbnic_fw_heartbeat_current(fbd)) {
+		dev_warn(fbd->dev,
+			 "Firmware did not respond to heartbeat message\n");
+		fbd->fw_heartbeat_enabled = false;
+	}
+
+	/* Place new message on mailbox to elicit a response */
+	err = fbnic_fw_xmit_heartbeat_message(fbd);
+	if (err)
+		dev_warn(fbd->dev, "Failed to send heartbeat message\n");
+}
+
 static const struct fbnic_tlv_parser fbnic_fw_tlv_parser[] = {
+	FBNIC_TLV_PARSER(FW_CAP_RESP, fbnic_fw_cap_resp_index,
+			 fbnic_fw_parse_cap_resp),
+	FBNIC_TLV_PARSER(OWNERSHIP_RESP, fbnic_ownership_resp_index,
+			 fbnic_fw_parse_ownership_resp),
+	FBNIC_TLV_PARSER(HEARTBEAT_RESP, fbnic_heartbeat_resp_index,
+			 fbnic_fw_parse_heartbeat_resp),
 	FBNIC_TLV_MSG_ERROR
 };
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.h b/drivers/net/ethernet/meta/fbnic/fbnic_fw.h
index c143079f881c..40d314f963ea 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.h
@@ -4,6 +4,7 @@
 #ifndef _FBNIC_FW_H_
 #define _FBNIC_FW_H_
 
+#include <linux/if_ether.h>
 #include <linux/types.h>
 
 struct fbnic_dev;
@@ -17,10 +18,94 @@ struct fbnic_fw_mbx {
 	} buf_info[FBNIC_IPC_MBX_DESC_LEN];
 };
 
+// FW_VER_MAX_SIZE must match ETHTOOL_FWVERS_LEN
+#define FBNIC_FW_VER_MAX_SIZE	                32
+// Formatted version is in the format XX.YY.ZZ_RRR_COMMIT
+#define FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE	(FBNIC_FW_VER_MAX_SIZE - 13)
+#define FBNIC_FW_LOG_MAX_SIZE	                256
+
+struct fbnic_fw_ver {
+	u32 version;
+	char commit[FBNIC_FW_CAP_RESP_COMMIT_MAX_SIZE];
+};
+
+struct fbnic_fw_cap {
+	struct {
+		struct fbnic_fw_ver mgmt, bootloader;
+	} running;
+	struct {
+		struct fbnic_fw_ver mgmt, bootloader, undi;
+	} stored;
+	u8	active_slot;
+	u8	bmc_mac_addr[4][ETH_ALEN];
+	u8	bmc_present	: 1;
+	u8	all_multi	: 1;
+	u8	link_speed;
+	u8	link_fec;
+};
+
 void fbnic_mbx_init(struct fbnic_dev *fbd);
 void fbnic_mbx_clean(struct fbnic_dev *fbd);
 void fbnic_mbx_poll(struct fbnic_dev *fbd);
 int fbnic_mbx_poll_tx_ready(struct fbnic_dev *fbd);
 void fbnic_mbx_flush_tx(struct fbnic_dev *fbd);
+int fbnic_fw_xmit_ownership_msg(struct fbnic_dev *fbd, bool take_ownership);
+int fbnic_fw_init_heartbeat(struct fbnic_dev *fbd, bool poll);
+void fbnic_fw_check_heartbeat(struct fbnic_dev *fbd);
+
+#define fbnic_mk_full_fw_ver_str(_rev_id, _delim, _commit, _str)	\
+do {									\
+	const u32 __rev_id = _rev_id;					\
+	snprintf(_str, sizeof(_str), "%02lu.%02lu.%02lu-%03lu%s%s",	\
+		 FIELD_GET(FBNIC_FW_CAP_RESP_VERSION_MAJOR, __rev_id),	\
+		 FIELD_GET(FBNIC_FW_CAP_RESP_VERSION_MINOR, __rev_id),	\
+		 FIELD_GET(FBNIC_FW_CAP_RESP_VERSION_PATCH, __rev_id),	\
+		 FIELD_GET(FBNIC_FW_CAP_RESP_VERSION_BUILD, __rev_id),	\
+		 _delim, _commit);					\
+} while (0)
 
+#define fbnic_mk_fw_ver_str(_rev_id, _str) \
+	fbnic_mk_full_fw_ver_str(_rev_id, "", "", _str)
+
+#define FW_HEARTBEAT_PERIOD		(10 * HZ)
+
+enum {
+	FBNIC_TLV_MSG_ID_HOST_CAP_REQ			= 0x10,
+	FBNIC_TLV_MSG_ID_FW_CAP_RESP			= 0x11,
+	FBNIC_TLV_MSG_ID_OWNERSHIP_REQ			= 0x12,
+	FBNIC_TLV_MSG_ID_OWNERSHIP_RESP			= 0x13,
+	FBNIC_TLV_MSG_ID_HEARTBEAT_REQ			= 0x14,
+	FBNIC_TLV_MSG_ID_HEARTBEAT_RESP			= 0x15,
+};
+
+#define FBNIC_FW_CAP_RESP_VERSION_MAJOR		CSR_GENMASK(31, 24)
+#define FBNIC_FW_CAP_RESP_VERSION_MINOR		CSR_GENMASK(23, 16)
+#define FBNIC_FW_CAP_RESP_VERSION_PATCH		CSR_GENMASK(15, 8)
+#define FBNIC_FW_CAP_RESP_VERSION_BUILD		CSR_GENMASK(7, 0)
+enum {
+	FBNIC_FW_CAP_RESP_VERSION			= 0x0,
+	FBNIC_FW_CAP_RESP_BMC_PRESENT			= 0x1,
+	FBNIC_FW_CAP_RESP_BMC_MAC_ADDR			= 0x2,
+	FBNIC_FW_CAP_RESP_BMC_MAC_ARRAY			= 0x3,
+	FBNIC_FW_CAP_RESP_STORED_VERSION		= 0x4,
+	FBNIC_FW_CAP_RESP_ACTIVE_FW_SLOT		= 0x5,
+	FBNIC_FW_CAP_RESP_VERSION_COMMIT_STR		= 0x6,
+	FBNIC_FW_CAP_RESP_BMC_ALL_MULTI			= 0x8,
+	FBNIC_FW_CAP_RESP_FW_STATE			= 0x9,
+	FBNIC_FW_CAP_RESP_FW_LINK_SPEED			= 0xa,
+	FBNIC_FW_CAP_RESP_FW_LINK_FEC			= 0xb,
+	FBNIC_FW_CAP_RESP_STORED_COMMIT_STR		= 0xc,
+	FBNIC_FW_CAP_RESP_CMRT_VERSION			= 0xd,
+	FBNIC_FW_CAP_RESP_STORED_CMRT_VERSION		= 0xe,
+	FBNIC_FW_CAP_RESP_CMRT_COMMIT_STR		= 0xf,
+	FBNIC_FW_CAP_RESP_STORED_CMRT_COMMIT_STR	= 0x10,
+	FBNIC_FW_CAP_RESP_UEFI_VERSION			= 0x11,
+	FBNIC_FW_CAP_RESP_UEFI_COMMIT_STR		= 0x12,
+	FBNIC_FW_CAP_RESP_MSG_MAX
+};
+
+enum {
+	FBNIC_FW_OWNERSHIP_FLAG			= 0x0,
+	FBNIC_FW_OWNERSHIP_MSG_MAX
+};
 #endif /* _FBNIC_FW_H_ */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
index f5a5076d0f52..0dd955c7c7ff 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
@@ -11,6 +11,7 @@
 
 int __fbnic_open(struct fbnic_net *fbn)
 {
+	struct fbnic_dev *fbd = fbn->fbd;
 	int err;
 
 	err = fbnic_alloc_napi_vectors(fbn);
@@ -31,7 +32,22 @@ int __fbnic_open(struct fbnic_net *fbn)
 	if (err)
 		goto free_resources;
 
+	/* Send ownership message and flush to verify FW has seen it */
+	err = fbnic_fw_xmit_ownership_msg(fbd, true);
+	if (err) {
+		dev_warn(fbd->dev,
+			 "Error %d sending host ownership message to the firmware\n",
+			 err);
+		goto free_resources;
+	}
+
+	err = fbnic_fw_init_heartbeat(fbd, false);
+	if (err)
+		goto release_ownership;
+
 	return 0;
+release_ownership:
+	fbnic_fw_xmit_ownership_msg(fbn->fbd, false);
 free_resources:
 	fbnic_free_resources(fbn);
 free_napi_vectors:
@@ -57,6 +73,8 @@ static int fbnic_stop(struct net_device *netdev)
 
 	fbnic_down(fbn);
 
+	fbnic_fw_xmit_ownership_msg(fbn->fbd, false);
+
 	fbnic_free_resources(fbn);
 	fbnic_free_napi_vectors(fbn);
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
index 325b20b232eb..9eda65e04609 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -158,6 +158,30 @@ void fbnic_down(struct fbnic_net *fbn)
 	fbnic_flush(fbn);
 }
 
+static void fbnic_health_check(struct fbnic_dev *fbd)
+{
+	struct fbnic_fw_mbx *tx_mbx = &fbd->mbx[FBNIC_IPC_MBX_TX_IDX];
+
+	/* As long as the heart is beating the FW is healty */
+	if (fbd->fw_heartbeat_enabled)
+		return;
+
+	/* If the Tx mailbox still has messages sitting in it then there likely
+	 * isn't anything we can do. We will wait until the mailbox is empty to
+	 * report the fault so we can collect the crashlog.
+	 */
+	if (tx_mbx->head != tx_mbx->tail)
+		return;
+
+	/* TBD: Need to add a more thorough recovery here.
+	 *	Specifically I need to verify what all the firmware will have
+	 *	changed since we had setup and it rebooted. May just need to
+	 *	perform a down/up. For now we will just reclaim ownership so
+	 *	the heartbeat can catch the next fault.
+	 */
+	fbnic_fw_xmit_ownership_msg(fbd, true);
+}
+
 static void fbnic_service_task(struct work_struct *work)
 {
 	struct fbnic_dev *fbd = container_of(to_delayed_work(work),
@@ -165,6 +189,10 @@ static void fbnic_service_task(struct work_struct *work)
 
 	rtnl_lock();
 
+	fbnic_fw_check_heartbeat(fbd);
+
+	fbnic_health_check(fbd);
+
 	if (netif_running(fbd->netdev))
 		schedule_delayed_work(&fbd->service_task, HZ);
 
-- 
cgit v1.2.3-58-ga151


From 69684376eed517817251ea6a768cfc315350d5c1 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:49:58 -0700
Subject: eth: fbnic: Add link detection

Add basic support for detecting the link and reporting it at the netdev
layer. For now we will just use the values reporeted by the firmware as the
link configuration and assume that is the current configuration of the MAC
and PCS.

With this we start the stubbing out of the phylink interface that will be
used to provide the configuration interface for ethtool in a future patch
set.

The phylink interface isn't an exact fit. As such we are currently working
around several issues in this patch set that we plan to address in the
future such as:
1. Support for FEC
2. Support for multiple lanes to handle 50GbaseR2 vs 50GbaseR1
3. Support for BMC

CC: Russell King <linux@armlinux.org.uk>
CC: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079939835.1778861.5964790909718481811.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/Kconfig               |   1 +
 drivers/net/ethernet/meta/fbnic/Makefile        |   1 +
 drivers/net/ethernet/meta/fbnic/fbnic.h         |  10 +
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h     |  39 ++++
 drivers/net/ethernet/meta/fbnic/fbnic_fw.h      |  13 ++
 drivers/net/ethernet/meta/fbnic/fbnic_irq.c     |  67 +++++++
 drivers/net/ethernet/meta/fbnic/fbnic_mac.c     | 241 ++++++++++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_mac.h     |  61 ++++++
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.c  |  22 +++
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.h  |  11 ++
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c     |   2 +
 drivers/net/ethernet/meta/fbnic/fbnic_phylink.c | 161 ++++++++++++++++
 12 files changed, 629 insertions(+)
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_phylink.c

diff --git a/drivers/net/ethernet/meta/Kconfig b/drivers/net/ethernet/meta/Kconfig
index fbbc38e7e507..d8f5e9f9bb33 100644
--- a/drivers/net/ethernet/meta/Kconfig
+++ b/drivers/net/ethernet/meta/Kconfig
@@ -21,6 +21,7 @@ config FBNIC
 	tristate "Meta Platforms Host Network Interface"
 	depends on X86_64 || COMPILE_TEST
 	depends on PCI_MSI
+	select PHYLINK
 	help
 	  This driver supports Meta Platforms Host Network Interface.
 
diff --git a/drivers/net/ethernet/meta/fbnic/Makefile b/drivers/net/ethernet/meta/fbnic/Makefile
index f2ea90e0c14f..a487ac5c4ec5 100644
--- a/drivers/net/ethernet/meta/fbnic/Makefile
+++ b/drivers/net/ethernet/meta/fbnic/Makefile
@@ -13,5 +13,6 @@ fbnic-y := fbnic_devlink.o \
 	   fbnic_mac.o \
 	   fbnic_netdev.o \
 	   fbnic_pci.o \
+	   fbnic_phylink.o \
 	   fbnic_tlv.o \
 	   fbnic_txrx.o
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h
index 44fe6bbf88a1..3539e090df69 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic.h
@@ -21,6 +21,7 @@ struct fbnic_dev {
 	u32 __iomem *uc_addr4;
 	const struct fbnic_mac *mac;
 	unsigned int fw_msix_vector;
+	unsigned int pcs_msix_vector;
 	unsigned short num_irqs;
 
 	struct delayed_work service_task;
@@ -49,6 +50,7 @@ struct fbnic_dev {
  */
 enum {
 	FBNIC_FW_MSIX_ENTRY,
+	FBNIC_PCS_MSIX_ENTRY,
 	FBNIC_NON_NAPI_VECTORS
 };
 
@@ -95,6 +97,11 @@ void fbnic_fw_wr32(struct fbnic_dev *fbd, u32 reg, u32 val);
 #define fw_wr32(_f, _r, _v)	fbnic_fw_wr32(_f, _r, _v)
 #define fw_wrfl(_f)		fbnic_fw_rd32(_f, FBNIC_FW_ZERO_REG)
 
+static inline bool fbnic_bmc_present(struct fbnic_dev *fbd)
+{
+	return fbd->fw_cap.bmc_present;
+}
+
 static inline bool fbnic_init_failure(struct fbnic_dev *fbd)
 {
 	return !fbd->netdev;
@@ -110,6 +117,9 @@ void fbnic_devlink_unregister(struct fbnic_dev *fbd);
 int fbnic_fw_enable_mbx(struct fbnic_dev *fbd);
 void fbnic_fw_disable_mbx(struct fbnic_dev *fbd);
 
+int fbnic_pcs_irq_enable(struct fbnic_dev *fbd);
+void fbnic_pcs_irq_disable(struct fbnic_dev *fbd);
+
 int fbnic_request_irq(struct fbnic_dev *dev, int nr, irq_handler_t handler,
 		      unsigned long flags, const char *name, void *data);
 void fbnic_free_irq(struct fbnic_dev *dev, int nr, void *data);
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index fb64ad919d31..031ddd9ac4d4 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -85,6 +85,9 @@
 #define FBNIC_INTR_MSIX_CTRL(n)		(0x00040 + (n)) /* 0x00100 + 4*n */
 #define FBNIC_INTR_MSIX_CTRL_VECTOR_MASK	CSR_GENMASK(7, 0)
 #define FBNIC_INTR_MSIX_CTRL_ENABLE		CSR_BIT(31)
+enum {
+	FBNIC_INTR_MSIX_CTRL_PCS_IDX	= 34,
+};
 
 #define FBNIC_CSR_END_INTR		0x0005f	/* CSR section delimiter */
 
@@ -419,6 +422,42 @@ enum {
 #define FBNIC_MASTER_SPARE_0		0x0C41B		/* 0x3106c */
 #define FBNIC_CSR_END_MASTER		0x0C452	/* CSR section delimiter */
 
+/* MAC MAC registers (ASIC only) */
+#define FBNIC_CSR_START_MAC_MAC		0x11000 /* CSR section delimiter */
+#define FBNIC_MAC_COMMAND_CONFIG	0x11002		/* 0x44008 */
+#define FBNIC_MAC_COMMAND_CONFIG_RX_PAUSE_DIS	CSR_BIT(29)
+#define FBNIC_MAC_COMMAND_CONFIG_TX_PAUSE_DIS	CSR_BIT(28)
+#define FBNIC_MAC_COMMAND_CONFIG_FLT_HDL_DIS	CSR_BIT(27)
+#define FBNIC_MAC_COMMAND_CONFIG_TX_PAD_EN	CSR_BIT(11)
+#define FBNIC_MAC_COMMAND_CONFIG_LOOPBACK_EN	CSR_BIT(10)
+#define FBNIC_MAC_COMMAND_CONFIG_PROMISC_EN	CSR_BIT(4)
+#define FBNIC_MAC_COMMAND_CONFIG_RX_ENA		CSR_BIT(1)
+#define FBNIC_MAC_COMMAND_CONFIG_TX_ENA		CSR_BIT(0)
+#define FBNIC_MAC_CL01_PAUSE_QUANTA	0x11015		/* 0x44054 */
+#define FBNIC_MAC_CL01_QUANTA_THRESH	0x11019		/* 0x44064 */
+#define FBNIC_CSR_END_MAC_MAC		0x11028 /* CSR section delimiter */
+
+/* Signals from MAC, AN, PCS, and LED CSR registers (ASIC only) */
+#define FBNIC_CSR_START_SIG		0x11800 /* CSR section delimiter */
+#define FBNIC_SIG_MAC_IN0		0x11800		/* 0x46000 */
+#define FBNIC_SIG_MAC_IN0_RESET_FF_TX_CLK	CSR_BIT(14)
+#define FBNIC_SIG_MAC_IN0_RESET_FF_RX_CLK	CSR_BIT(13)
+#define FBNIC_SIG_MAC_IN0_RESET_TX_CLK		CSR_BIT(12)
+#define FBNIC_SIG_MAC_IN0_RESET_RX_CLK		CSR_BIT(11)
+#define FBNIC_SIG_MAC_IN0_TX_CRC		CSR_BIT(8)
+#define FBNIC_SIG_MAC_IN0_CFG_MODE128		CSR_BIT(10)
+#define FBNIC_SIG_PCS_OUT0		0x11808		/* 0x46020 */
+#define FBNIC_SIG_PCS_OUT0_LINK			CSR_BIT(27)
+#define FBNIC_SIG_PCS_OUT0_BLOCK_LOCK		CSR_GENMASK(24, 5)
+#define FBNIC_SIG_PCS_OUT0_AMPS_LOCK		CSR_GENMASK(4, 1)
+#define FBNIC_SIG_PCS_OUT1		0x11809		/* 0x46024 */
+#define FBNIC_SIG_PCS_OUT1_FCFEC_LOCK		CSR_GENMASK(11, 8)
+#define FBNIC_SIG_PCS_INTR_STS		0x11814		/* 0x46050 */
+#define FBNIC_SIG_PCS_INTR_LINK_DOWN		CSR_BIT(1)
+#define FBNIC_SIG_PCS_INTR_LINK_UP		CSR_BIT(0)
+#define FBNIC_SIG_PCS_INTR_MASK		0x11816		/* 0x46058 */
+#define FBNIC_CSR_END_SIG		0x1184e /* CSR section delimiter */
+
 /* PUL User Registers */
 #define FBNIC_CSR_START_PUL_USER	0x31000	/* CSR section delimiter */
 #define FBNIC_PUL_OB_TLP_HDR_AW_CFG	0x3103d		/* 0xc40f4 */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_fw.h b/drivers/net/ethernet/meta/fbnic/fbnic_fw.h
index 40d314f963ea..c65bca613665 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_fw.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_fw.h
@@ -104,6 +104,19 @@ enum {
 	FBNIC_FW_CAP_RESP_MSG_MAX
 };
 
+enum {
+	FBNIC_FW_LINK_SPEED_25R1		= 1,
+	FBNIC_FW_LINK_SPEED_50R2		= 2,
+	FBNIC_FW_LINK_SPEED_50R1		= 3,
+	FBNIC_FW_LINK_SPEED_100R2		= 4,
+};
+
+enum {
+	FBNIC_FW_LINK_FEC_NONE			= 1,
+	FBNIC_FW_LINK_FEC_RS			= 2,
+	FBNIC_FW_LINK_FEC_BASER			= 3,
+};
+
 enum {
 	FBNIC_FW_OWNERSHIP_FLAG			= 0x0,
 	FBNIC_FW_OWNERSHIP_MSG_MAX
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c
index 7aba25d14495..914362195920 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_irq.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_irq.c
@@ -5,6 +5,7 @@
 #include <linux/types.h>
 
 #include "fbnic.h"
+#include "fbnic_netdev.h"
 #include "fbnic_txrx.h"
 
 static irqreturn_t fbnic_fw_msix_intr(int __always_unused irq, void *data)
@@ -81,6 +82,70 @@ void fbnic_fw_disable_mbx(struct fbnic_dev *fbd)
 	fbnic_mbx_clean(fbd);
 }
 
+static irqreturn_t fbnic_pcs_msix_intr(int __always_unused irq, void *data)
+{
+	struct fbnic_dev *fbd = data;
+	struct fbnic_net *fbn;
+
+	if (fbd->mac->pcs_get_link_event(fbd) == FBNIC_LINK_EVENT_NONE) {
+		fbnic_wr32(fbd, FBNIC_INTR_MASK_CLEAR(0),
+			   1u << FBNIC_PCS_MSIX_ENTRY);
+		return IRQ_HANDLED;
+	}
+
+	fbn = netdev_priv(fbd->netdev);
+
+	phylink_pcs_change(&fbn->phylink_pcs, false);
+
+	return IRQ_HANDLED;
+}
+
+/**
+ * fbnic_pcs_irq_enable - Configure the MAC to enable it to advertise link
+ * @fbd: Pointer to device to initialize
+ *
+ * This function provides basic bringup for the MAC/PCS IRQ. For now the IRQ
+ * will remain disabled until we start the MAC/PCS/PHY logic via phylink.
+ *
+ * Return: non-zero on failure.
+ **/
+int fbnic_pcs_irq_enable(struct fbnic_dev *fbd)
+{
+	u32 vector = fbd->pcs_msix_vector;
+	int err;
+
+	/* Request the IRQ for MAC link vector.
+	 * Map MAC cause to it, and unmask it
+	 */
+	err = request_irq(vector, &fbnic_pcs_msix_intr, 0,
+			  fbd->netdev->name, fbd);
+	if (err)
+		return err;
+
+	fbnic_wr32(fbd, FBNIC_INTR_MSIX_CTRL(FBNIC_INTR_MSIX_CTRL_PCS_IDX),
+		   FBNIC_PCS_MSIX_ENTRY | FBNIC_INTR_MSIX_CTRL_ENABLE);
+
+	return 0;
+}
+
+/**
+ * fbnic_pcs_irq_disable - Teardown the MAC IRQ to prepare for stopping
+ * @fbd: Pointer to device that is stopping
+ *
+ * This function undoes the work done in fbnic_pcs_irq_enable and prepares
+ * the device to no longer receive traffic on the host interface.
+ **/
+void fbnic_pcs_irq_disable(struct fbnic_dev *fbd)
+{
+	/* Disable interrupt */
+	fbnic_wr32(fbd, FBNIC_INTR_MSIX_CTRL(FBNIC_INTR_MSIX_CTRL_PCS_IDX),
+		   FBNIC_PCS_MSIX_ENTRY);
+	fbnic_wr32(fbd, FBNIC_INTR_MASK_SET(0), 1u << FBNIC_PCS_MSIX_ENTRY);
+
+	/* Free the vector */
+	free_irq(fbd->pcs_msix_vector, fbd);
+}
+
 int fbnic_request_irq(struct fbnic_dev *fbd, int nr, irq_handler_t handler,
 		      unsigned long flags, const char *name, void *data)
 {
@@ -108,6 +173,7 @@ void fbnic_free_irqs(struct fbnic_dev *fbd)
 {
 	struct pci_dev *pdev = to_pci_dev(fbd->dev);
 
+	fbd->pcs_msix_vector = 0;
 	fbd->fw_msix_vector = 0;
 
 	fbd->num_irqs = 0;
@@ -135,6 +201,7 @@ int fbnic_alloc_irqs(struct fbnic_dev *fbd)
 
 	fbd->num_irqs = num_irqs;
 
+	fbd->pcs_msix_vector = pci_irq_vector(pdev, FBNIC_PCS_MSIX_ENTRY);
 	fbd->fw_msix_vector = pci_irq_vector(pdev, FBNIC_FW_MSIX_ENTRY);
 
 	return 0;
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.c b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c
index a6ef898d7eed..7920e7af82d9 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_mac.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.c
@@ -6,6 +6,7 @@
 
 #include "fbnic.h"
 #include "fbnic_mac.h"
+#include "fbnic_netdev.h"
 
 static void fbnic_init_readrq(struct fbnic_dev *fbd, unsigned int offset,
 			      unsigned int cls, unsigned int readrq)
@@ -402,8 +403,248 @@ static void fbnic_mac_init_regs(struct fbnic_dev *fbd)
 	fbnic_mac_init_txb(fbd);
 }
 
+static void fbnic_mac_tx_pause_config(struct fbnic_dev *fbd, bool tx_pause)
+{
+	u32 rxb_pause_ctrl;
+
+	/* Enable generation of pause frames if enabled */
+	rxb_pause_ctrl = rd32(fbd, FBNIC_RXB_PAUSE_DROP_CTRL);
+	rxb_pause_ctrl &= ~FBNIC_RXB_PAUSE_DROP_CTRL_PAUSE_ENABLE;
+	if (tx_pause)
+		rxb_pause_ctrl |=
+			FIELD_PREP(FBNIC_RXB_PAUSE_DROP_CTRL_PAUSE_ENABLE,
+				   FBNIC_PAUSE_EN_MASK);
+	wr32(fbd, FBNIC_RXB_PAUSE_DROP_CTRL, rxb_pause_ctrl);
+}
+
+static int fbnic_pcs_get_link_event_asic(struct fbnic_dev *fbd)
+{
+	u32 pcs_intr_mask = rd32(fbd, FBNIC_SIG_PCS_INTR_STS);
+
+	if (pcs_intr_mask & FBNIC_SIG_PCS_INTR_LINK_DOWN)
+		return FBNIC_LINK_EVENT_DOWN;
+
+	return (pcs_intr_mask & FBNIC_SIG_PCS_INTR_LINK_UP) ?
+	       FBNIC_LINK_EVENT_UP : FBNIC_LINK_EVENT_NONE;
+}
+
+static u32 __fbnic_mac_cmd_config_asic(struct fbnic_dev *fbd,
+				       bool tx_pause, bool rx_pause)
+{
+	/* Enable MAC Promiscuous mode and Tx padding */
+	u32 command_config = FBNIC_MAC_COMMAND_CONFIG_TX_PAD_EN |
+			     FBNIC_MAC_COMMAND_CONFIG_PROMISC_EN;
+	struct fbnic_net *fbn = netdev_priv(fbd->netdev);
+
+	/* Disable pause frames if not enabled */
+	if (!tx_pause)
+		command_config |= FBNIC_MAC_COMMAND_CONFIG_TX_PAUSE_DIS;
+	if (!rx_pause)
+		command_config |= FBNIC_MAC_COMMAND_CONFIG_RX_PAUSE_DIS;
+
+	/* Disable fault handling if no FEC is requested */
+	if ((fbn->fec & FBNIC_FEC_MODE_MASK) == FBNIC_FEC_OFF)
+		command_config |= FBNIC_MAC_COMMAND_CONFIG_FLT_HDL_DIS;
+
+	return command_config;
+}
+
+static bool fbnic_mac_get_pcs_link_status(struct fbnic_dev *fbd)
+{
+	struct fbnic_net *fbn = netdev_priv(fbd->netdev);
+	u32 pcs_status, lane_mask = ~0;
+
+	pcs_status = rd32(fbd, FBNIC_SIG_PCS_OUT0);
+	if (!(pcs_status & FBNIC_SIG_PCS_OUT0_LINK))
+		return false;
+
+	/* Define the expected lane mask for the status bits we need to check */
+	switch (fbn->link_mode & FBNIC_LINK_MODE_MASK) {
+	case FBNIC_LINK_100R2:
+		lane_mask = 0xf;
+		break;
+	case FBNIC_LINK_50R1:
+		lane_mask = 3;
+		break;
+	case FBNIC_LINK_50R2:
+		switch (fbn->fec & FBNIC_FEC_MODE_MASK) {
+		case FBNIC_FEC_OFF:
+			lane_mask = 0x63;
+			break;
+		case FBNIC_FEC_RS:
+			lane_mask = 5;
+			break;
+		case FBNIC_FEC_BASER:
+			lane_mask = 0xf;
+			break;
+		}
+		break;
+	case FBNIC_LINK_25R1:
+		lane_mask = 1;
+		break;
+	}
+
+	/* Use an XOR to remove the bits we expect to see set */
+	switch (fbn->fec & FBNIC_FEC_MODE_MASK) {
+	case FBNIC_FEC_OFF:
+		lane_mask ^= FIELD_GET(FBNIC_SIG_PCS_OUT0_BLOCK_LOCK,
+				       pcs_status);
+		break;
+	case FBNIC_FEC_RS:
+		lane_mask ^= FIELD_GET(FBNIC_SIG_PCS_OUT0_AMPS_LOCK,
+				       pcs_status);
+		break;
+	case FBNIC_FEC_BASER:
+		lane_mask ^= FIELD_GET(FBNIC_SIG_PCS_OUT1_FCFEC_LOCK,
+				       rd32(fbd, FBNIC_SIG_PCS_OUT1));
+		break;
+	}
+
+	/* If all lanes cancelled then we have a lock on all lanes */
+	return !lane_mask;
+}
+
+static bool fbnic_pcs_get_link_asic(struct fbnic_dev *fbd)
+{
+	bool link;
+
+	/* Flush status bits to clear possible stale data,
+	 * bits should reset themselves back to 1 if link is truly up
+	 */
+	wr32(fbd, FBNIC_SIG_PCS_OUT0, FBNIC_SIG_PCS_OUT0_LINK |
+				      FBNIC_SIG_PCS_OUT0_BLOCK_LOCK |
+				      FBNIC_SIG_PCS_OUT0_AMPS_LOCK);
+	wr32(fbd, FBNIC_SIG_PCS_OUT1, FBNIC_SIG_PCS_OUT1_FCFEC_LOCK);
+	wrfl(fbd);
+
+	/* Clear interrupt state due to recent changes. */
+	wr32(fbd, FBNIC_SIG_PCS_INTR_STS,
+	     FBNIC_SIG_PCS_INTR_LINK_DOWN | FBNIC_SIG_PCS_INTR_LINK_UP);
+
+	link = fbnic_mac_get_pcs_link_status(fbd);
+
+	/* Enable interrupt to only capture changes in link state */
+	wr32(fbd, FBNIC_SIG_PCS_INTR_MASK,
+	     ~FBNIC_SIG_PCS_INTR_LINK_DOWN & ~FBNIC_SIG_PCS_INTR_LINK_UP);
+	wr32(fbd, FBNIC_INTR_MASK_CLEAR(0), 1u << FBNIC_PCS_MSIX_ENTRY);
+
+	return link;
+}
+
+static void fbnic_pcs_get_fw_settings(struct fbnic_dev *fbd)
+{
+	struct fbnic_net *fbn = netdev_priv(fbd->netdev);
+	u8 link_mode = fbn->link_mode;
+	u8 fec = fbn->fec;
+
+	/* Update FEC first to reflect FW current mode */
+	if (fbn->fec & FBNIC_FEC_AUTO) {
+		switch (fbd->fw_cap.link_fec) {
+		case FBNIC_FW_LINK_FEC_NONE:
+			fec = FBNIC_FEC_OFF;
+			break;
+		case FBNIC_FW_LINK_FEC_RS:
+			fec = FBNIC_FEC_RS;
+			break;
+		case FBNIC_FW_LINK_FEC_BASER:
+			fec = FBNIC_FEC_BASER;
+			break;
+		default:
+			return;
+		}
+
+		fbn->fec = fec;
+	}
+
+	/* Do nothing if AUTO mode is not engaged */
+	if (fbn->link_mode & FBNIC_LINK_AUTO) {
+		switch (fbd->fw_cap.link_speed) {
+		case FBNIC_FW_LINK_SPEED_25R1:
+			link_mode = FBNIC_LINK_25R1;
+			break;
+		case FBNIC_FW_LINK_SPEED_50R2:
+			link_mode = FBNIC_LINK_50R2;
+			break;
+		case FBNIC_FW_LINK_SPEED_50R1:
+			link_mode = FBNIC_LINK_50R1;
+			fec = FBNIC_FEC_RS;
+			break;
+		case FBNIC_FW_LINK_SPEED_100R2:
+			link_mode = FBNIC_LINK_100R2;
+			fec = FBNIC_FEC_RS;
+			break;
+		default:
+			return;
+		}
+
+		fbn->link_mode = link_mode;
+	}
+}
+
+static int fbnic_pcs_enable_asic(struct fbnic_dev *fbd)
+{
+	/* Mask and clear the PCS interrupt, will be enabled by link handler */
+	wr32(fbd, FBNIC_SIG_PCS_INTR_MASK, ~0);
+	wr32(fbd, FBNIC_SIG_PCS_INTR_STS, ~0);
+
+	/* Pull in settings from FW */
+	fbnic_pcs_get_fw_settings(fbd);
+
+	return 0;
+}
+
+static void fbnic_pcs_disable_asic(struct fbnic_dev *fbd)
+{
+	/* Mask and clear the PCS interrupt */
+	wr32(fbd, FBNIC_SIG_PCS_INTR_MASK, ~0);
+	wr32(fbd, FBNIC_SIG_PCS_INTR_STS, ~0);
+}
+
+static void fbnic_mac_link_down_asic(struct fbnic_dev *fbd)
+{
+	u32 cmd_cfg, mac_ctrl;
+
+	cmd_cfg = __fbnic_mac_cmd_config_asic(fbd, false, false);
+	mac_ctrl = rd32(fbd, FBNIC_SIG_MAC_IN0);
+
+	mac_ctrl |= FBNIC_SIG_MAC_IN0_RESET_FF_TX_CLK |
+		    FBNIC_SIG_MAC_IN0_RESET_TX_CLK |
+		    FBNIC_SIG_MAC_IN0_RESET_FF_RX_CLK |
+		    FBNIC_SIG_MAC_IN0_RESET_RX_CLK;
+
+	wr32(fbd, FBNIC_SIG_MAC_IN0, mac_ctrl);
+	wr32(fbd, FBNIC_MAC_COMMAND_CONFIG, cmd_cfg);
+}
+
+static void fbnic_mac_link_up_asic(struct fbnic_dev *fbd,
+				   bool tx_pause, bool rx_pause)
+{
+	u32 cmd_cfg, mac_ctrl;
+
+	fbnic_mac_tx_pause_config(fbd, tx_pause);
+
+	cmd_cfg = __fbnic_mac_cmd_config_asic(fbd, tx_pause, rx_pause);
+	mac_ctrl = rd32(fbd, FBNIC_SIG_MAC_IN0);
+
+	mac_ctrl &= ~(FBNIC_SIG_MAC_IN0_RESET_FF_TX_CLK |
+		      FBNIC_SIG_MAC_IN0_RESET_TX_CLK |
+		      FBNIC_SIG_MAC_IN0_RESET_FF_RX_CLK |
+		      FBNIC_SIG_MAC_IN0_RESET_RX_CLK);
+	cmd_cfg |= FBNIC_MAC_COMMAND_CONFIG_RX_ENA |
+		   FBNIC_MAC_COMMAND_CONFIG_TX_ENA;
+
+	wr32(fbd, FBNIC_SIG_MAC_IN0, mac_ctrl);
+	wr32(fbd, FBNIC_MAC_COMMAND_CONFIG, cmd_cfg);
+}
+
 static const struct fbnic_mac fbnic_mac_asic = {
 	.init_regs = fbnic_mac_init_regs,
+	.pcs_enable = fbnic_pcs_enable_asic,
+	.pcs_disable = fbnic_pcs_disable_asic,
+	.pcs_get_link = fbnic_pcs_get_link_asic,
+	.pcs_get_link_event = fbnic_pcs_get_link_event_asic,
+	.link_down = fbnic_mac_link_down_asic,
+	.link_up = fbnic_mac_link_up_asic,
 };
 
 /**
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_mac.h b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h
index e78a92338a62..f53be6e6aef9 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_mac.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_mac.h
@@ -10,15 +10,76 @@ struct fbnic_dev;
 
 #define FBNIC_MAX_JUMBO_FRAME_SIZE	9742
 
+enum {
+	FBNIC_LINK_EVENT_NONE	= 0,
+	FBNIC_LINK_EVENT_UP	= 1,
+	FBNIC_LINK_EVENT_DOWN	= 2,
+};
+
+/* Treat the FEC bits as a bitmask laid out as follows:
+ * Bit 0: RS Enabled
+ * Bit 1: BASER(Firecode) Enabled
+ * Bit 2: Retrieve FEC from FW
+ */
+enum {
+	FBNIC_FEC_OFF		= 0,
+	FBNIC_FEC_RS		= 1,
+	FBNIC_FEC_BASER		= 2,
+	FBNIC_FEC_AUTO		= 4,
+};
+
+#define FBNIC_FEC_MODE_MASK	(FBNIC_FEC_AUTO - 1)
+
+/* Treat the link modes as a set of modulation/lanes bitmask:
+ * Bit 0: Lane Count, 0 = R1, 1 = R2
+ * Bit 1: Modulation, 0 = NRZ, 1 = PAM4
+ * Bit 2: Retrieve link mode from FW
+ */
+enum {
+	FBNIC_LINK_25R1		= 0,
+	FBNIC_LINK_50R2		= 1,
+	FBNIC_LINK_50R1		= 2,
+	FBNIC_LINK_100R2	= 3,
+	FBNIC_LINK_AUTO		= 4,
+};
+
+#define FBNIC_LINK_MODE_R2	(FBNIC_LINK_50R2)
+#define FBNIC_LINK_MODE_PAM4	(FBNIC_LINK_50R1)
+#define FBNIC_LINK_MODE_MASK	(FBNIC_LINK_AUTO - 1)
+
 /* This structure defines the interface hooks for the MAC. The MAC hooks
  * will be configured as a const struct provided with a set of function
  * pointers.
  *
  * void (*init_regs)(struct fbnic_dev *fbd);
  *	Initialize MAC registers to enable Tx/Rx paths and FIFOs.
+ *
+ * void (*pcs_enable)(struct fbnic_dev *fbd);
+ *	Configure and enable PCS to enable link if not already enabled
+ * void (*pcs_disable)(struct fbnic_dev *fbd);
+ *	Shutdown the link if we are the only consumer of it.
+ * bool (*pcs_get_link)(struct fbnic_dev *fbd);
+ *	Check PCS link status
+ * int (*pcs_get_link_event)(struct fbnic_dev *fbd)
+ *	Get the current link event status, reports true if link has
+ *	changed to either FBNIC_LINK_EVENT_DOWN or FBNIC_LINK_EVENT_UP
+ *
+ * void (*link_down)(struct fbnic_dev *fbd);
+ *	Configure MAC for link down event
+ * void (*link_up)(struct fbnic_dev *fbd, bool tx_pause, bool rx_pause);
+ *	Configure MAC for link up event;
+ *
  */
 struct fbnic_mac {
 	void (*init_regs)(struct fbnic_dev *fbd);
+
+	int (*pcs_enable)(struct fbnic_dev *fbd);
+	void (*pcs_disable)(struct fbnic_dev *fbd);
+	bool (*pcs_get_link)(struct fbnic_dev *fbd);
+	int (*pcs_get_link_event)(struct fbnic_dev *fbd);
+
+	void (*link_down)(struct fbnic_dev *fbd);
+	void (*link_up)(struct fbnic_dev *fbd, bool tx_pause, bool rx_pause);
 };
 
 int fbnic_mac_init(struct fbnic_dev *fbd);
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
index 0dd955c7c7ff..18953a2d1eef 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
@@ -45,6 +45,10 @@ int __fbnic_open(struct fbnic_net *fbn)
 	if (err)
 		goto release_ownership;
 
+	err = fbnic_pcs_irq_enable(fbd);
+	if (err)
+		goto release_ownership;
+
 	return 0;
 release_ownership:
 	fbnic_fw_xmit_ownership_msg(fbn->fbd, false);
@@ -72,6 +76,7 @@ static int fbnic_stop(struct net_device *netdev)
 	struct fbnic_net *fbn = netdev_priv(netdev);
 
 	fbnic_down(fbn);
+	fbnic_pcs_irq_disable(fbn->fbd);
 
 	fbnic_fw_xmit_ownership_msg(fbn->fbd, false);
 
@@ -114,6 +119,11 @@ void fbnic_reset_queues(struct fbnic_net *fbn,
  **/
 void fbnic_netdev_free(struct fbnic_dev *fbd)
 {
+	struct fbnic_net *fbn = netdev_priv(fbd->netdev);
+
+	if (fbn->phylink)
+		phylink_destroy(fbn->phylink);
+
 	free_netdev(fbd->netdev);
 	fbd->netdev = NULL;
 }
@@ -162,10 +172,22 @@ struct net_device *fbnic_netdev_alloc(struct fbnic_dev *fbd)
 	netdev->min_mtu = IPV6_MIN_MTU;
 	netdev->max_mtu = FBNIC_MAX_JUMBO_FRAME_SIZE - ETH_HLEN;
 
+	/* TBD: This is workaround for BMC as phylink doesn't have support
+	 * for leavling the link enabled if a BMC is present.
+	 */
+	netdev->ethtool->wol_enabled = true;
+
+	fbn->fec = FBNIC_FEC_AUTO | FBNIC_FEC_RS;
+	fbn->link_mode = FBNIC_LINK_AUTO | FBNIC_LINK_50R2;
 	netif_carrier_off(netdev);
 
 	netif_tx_stop_all_queues(netdev);
 
+	if (fbnic_phylink_init(netdev)) {
+		fbnic_netdev_free(fbd);
+		return NULL;
+	}
+
 	return netdev;
 }
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
index 18f93e9431cc..faf75e6359ba 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
@@ -5,6 +5,7 @@
 #define _FBNIC_NETDEV_H_
 
 #include <linux/types.h>
+#include <linux/phylink.h>
 
 #include "fbnic_txrx.h"
 
@@ -22,9 +23,18 @@ struct fbnic_net {
 
 	u16 num_napi;
 
+	struct phylink *phylink;
+	struct phylink_config phylink_config;
+	struct phylink_pcs phylink_pcs;
+
+	/* TBD: Remove these when phylink supports FEC and lane config */
+	u8 fec;
+	u8 link_mode;
+
 	u16 num_tx_queues;
 	u16 num_rx_queues;
 
+	u64 link_down_events;
 	struct list_head napis;
 };
 
@@ -39,4 +49,5 @@ void fbnic_netdev_unregister(struct net_device *netdev);
 void fbnic_reset_queues(struct fbnic_net *fbn,
 			unsigned int tx, unsigned int rx);
 
+int fbnic_phylink_init(struct net_device *netdev);
 #endif /* _FBNIC_NETDEV_H_ */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
index 9eda65e04609..040aaa127500 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -116,12 +116,14 @@ static void fbnic_service_task_start(struct fbnic_net *fbn)
 	struct fbnic_dev *fbd = fbn->fbd;
 
 	schedule_delayed_work(&fbd->service_task, HZ);
+	phylink_resume(fbn->phylink);
 }
 
 static void fbnic_service_task_stop(struct fbnic_net *fbn)
 {
 	struct fbnic_dev *fbd = fbn->fbd;
 
+	phylink_suspend(fbn->phylink, fbnic_bmc_present(fbd));
 	cancel_delayed_work(&fbd->service_task);
 }
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c b/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c
new file mode 100644
index 000000000000..1a5e1e719b30
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_phylink.c
@@ -0,0 +1,161 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#include <linux/phy.h>
+#include <linux/phylink.h>
+
+#include "fbnic.h"
+#include "fbnic_mac.h"
+#include "fbnic_netdev.h"
+
+static struct fbnic_net *
+fbnic_pcs_to_net(struct phylink_pcs *pcs)
+{
+	return container_of(pcs, struct fbnic_net, phylink_pcs);
+}
+
+static void
+fbnic_phylink_pcs_get_state(struct phylink_pcs *pcs,
+			    struct phylink_link_state *state)
+{
+	struct fbnic_net *fbn = fbnic_pcs_to_net(pcs);
+	struct fbnic_dev *fbd = fbn->fbd;
+
+	/* For now we use hard-coded defaults and FW config to determine
+	 * the current values. In future patches we will add support for
+	 * reconfiguring these values and changing link settings.
+	 */
+	switch (fbd->fw_cap.link_speed) {
+	case FBNIC_FW_LINK_SPEED_25R1:
+		state->speed = SPEED_25000;
+		break;
+	case FBNIC_FW_LINK_SPEED_50R2:
+		state->speed = SPEED_50000;
+		break;
+	case FBNIC_FW_LINK_SPEED_100R2:
+		state->speed = SPEED_100000;
+		break;
+	default:
+		state->speed = SPEED_UNKNOWN;
+		break;
+	}
+
+	state->duplex = DUPLEX_FULL;
+
+	state->link = fbd->mac->pcs_get_link(fbd);
+}
+
+static int
+fbnic_phylink_pcs_enable(struct phylink_pcs *pcs)
+{
+	struct fbnic_net *fbn = fbnic_pcs_to_net(pcs);
+	struct fbnic_dev *fbd = fbn->fbd;
+
+	return fbd->mac->pcs_enable(fbd);
+}
+
+static void
+fbnic_phylink_pcs_disable(struct phylink_pcs *pcs)
+{
+	struct fbnic_net *fbn = fbnic_pcs_to_net(pcs);
+	struct fbnic_dev *fbd = fbn->fbd;
+
+	return fbd->mac->pcs_disable(fbd);
+}
+
+static int
+fbnic_phylink_pcs_config(struct phylink_pcs *pcs, unsigned int neg_mode,
+			 phy_interface_t interface,
+			 const unsigned long *advertising,
+			 bool permit_pause_to_mac)
+{
+	return 0;
+}
+
+static const struct phylink_pcs_ops fbnic_phylink_pcs_ops = {
+	.pcs_config = fbnic_phylink_pcs_config,
+	.pcs_enable = fbnic_phylink_pcs_enable,
+	.pcs_disable = fbnic_phylink_pcs_disable,
+	.pcs_get_state = fbnic_phylink_pcs_get_state,
+};
+
+static struct phylink_pcs *
+fbnic_phylink_mac_select_pcs(struct phylink_config *config,
+			     phy_interface_t interface)
+{
+	struct net_device *netdev = to_net_dev(config->dev);
+	struct fbnic_net *fbn = netdev_priv(netdev);
+
+	return &fbn->phylink_pcs;
+}
+
+static void
+fbnic_phylink_mac_config(struct phylink_config *config, unsigned int mode,
+			 const struct phylink_link_state *state)
+{
+}
+
+static void
+fbnic_phylink_mac_link_down(struct phylink_config *config, unsigned int mode,
+			    phy_interface_t interface)
+{
+	struct net_device *netdev = to_net_dev(config->dev);
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	struct fbnic_dev *fbd = fbn->fbd;
+
+	fbd->mac->link_down(fbd);
+
+	fbn->link_down_events++;
+}
+
+static void
+fbnic_phylink_mac_link_up(struct phylink_config *config,
+			  struct phy_device *phy, unsigned int mode,
+			  phy_interface_t interface, int speed, int duplex,
+			  bool tx_pause, bool rx_pause)
+{
+	struct net_device *netdev = to_net_dev(config->dev);
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	struct fbnic_dev *fbd = fbn->fbd;
+
+	fbd->mac->link_up(fbd, tx_pause, rx_pause);
+}
+
+static const struct phylink_mac_ops fbnic_phylink_mac_ops = {
+	.mac_select_pcs = fbnic_phylink_mac_select_pcs,
+	.mac_config = fbnic_phylink_mac_config,
+	.mac_link_down = fbnic_phylink_mac_link_down,
+	.mac_link_up = fbnic_phylink_mac_link_up,
+};
+
+int fbnic_phylink_init(struct net_device *netdev)
+{
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	struct phylink *phylink;
+
+	fbn->phylink_pcs.neg_mode = true;
+	fbn->phylink_pcs.ops = &fbnic_phylink_pcs_ops;
+
+	fbn->phylink_config.dev = &netdev->dev;
+	fbn->phylink_config.type = PHYLINK_NETDEV;
+	fbn->phylink_config.mac_capabilities = MAC_SYM_PAUSE | MAC_ASYM_PAUSE |
+					       MAC_10000FD | MAC_25000FD |
+					       MAC_40000FD | MAC_50000FD |
+					       MAC_100000FD;
+	fbn->phylink_config.default_an_inband = true;
+
+	__set_bit(PHY_INTERFACE_MODE_XGMII,
+		  fbn->phylink_config.supported_interfaces);
+	__set_bit(PHY_INTERFACE_MODE_XLGMII,
+		  fbn->phylink_config.supported_interfaces);
+
+	phylink = phylink_create(&fbn->phylink_config, NULL,
+				 PHY_INTERFACE_MODE_XLGMII,
+				 &fbnic_phylink_mac_ops);
+	if (IS_ERR(phylink))
+		return PTR_ERR(phylink);
+
+	fbn->phylink = phylink;
+
+	return 0;
+}
-- 
cgit v1.2.3-58-ga151


From 9a57bacd574b52b08bc2c600af6a8c7a87049ecf Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:50:24 -0700
Subject: eth: fbnic: Add basic Tx handling

Handle Tx of simple packets. Support checksum offload and gather.
Use .ndo_features_check to make sure packet geometry will be
supported by the HW, i.e. we can fit the header lengths into
the descriptor fields.

The device writes to the completion rings the position of the tail
(consumer) pointer. Read all those writebacks, obviously the last
one will be the most recent, complete skbs up to that point.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079942464.1778861.17919428039428796180.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h    |  66 +++++
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.c |   9 +
 drivers/net/ethernet/meta/fbnic/fbnic_txrx.c   | 388 ++++++++++++++++++++++++-
 drivers/net/ethernet/meta/fbnic/fbnic_txrx.h   |  15 +
 4 files changed, 477 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index 031ddd9ac4d4..d7ce6781b9dc 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -24,6 +24,72 @@
 
 #define FBNIC_CLOCK_FREQ	(600 * (1000 * 1000))
 
+/* Transmit Work Descriptor Format */
+/* Length, Type, Offset Masks and Shifts */
+#define FBNIC_TWD_L2_HLEN_MASK			DESC_GENMASK(5, 0)
+
+#define FBNIC_TWD_L3_TYPE_MASK			DESC_GENMASK(7, 6)
+enum {
+	FBNIC_TWD_L3_TYPE_OTHER	= 0,
+	FBNIC_TWD_L3_TYPE_IPV4	= 1,
+	FBNIC_TWD_L3_TYPE_IPV6	= 2,
+	FBNIC_TWD_L3_TYPE_V6V6	= 3,
+};
+
+#define FBNIC_TWD_L3_OHLEN_MASK			DESC_GENMASK(15, 8)
+#define FBNIC_TWD_L3_IHLEN_MASK			DESC_GENMASK(23, 16)
+
+enum {
+	FBNIC_TWD_L4_TYPE_OTHER	= 0,
+	FBNIC_TWD_L4_TYPE_TCP	= 1,
+	FBNIC_TWD_L4_TYPE_UDP	= 2,
+};
+
+#define FBNIC_TWD_CSUM_OFFSET_MASK		DESC_GENMASK(27, 24)
+#define FBNIC_TWD_L4_HLEN_MASK			DESC_GENMASK(31, 28)
+
+/* Flags and Type */
+#define FBNIC_TWD_L4_TYPE_MASK			DESC_GENMASK(33, 32)
+#define FBNIC_TWD_FLAG_REQ_TS			DESC_BIT(34)
+#define FBNIC_TWD_FLAG_REQ_LSO			DESC_BIT(35)
+#define FBNIC_TWD_FLAG_REQ_CSO			DESC_BIT(36)
+#define FBNIC_TWD_FLAG_REQ_COMPLETION		DESC_BIT(37)
+#define FBNIC_TWD_FLAG_DEST_MAC			DESC_BIT(43)
+#define FBNIC_TWD_FLAG_DEST_BMC			DESC_BIT(44)
+#define FBNIC_TWD_FLAG_DEST_FW			DESC_BIT(45)
+#define FBNIC_TWD_TYPE_MASK			DESC_GENMASK(47, 46)
+enum {
+	FBNIC_TWD_TYPE_META	= 0,
+	FBNIC_TWD_TYPE_OPT_META	= 1,
+	FBNIC_TWD_TYPE_AL	= 2,
+	FBNIC_TWD_TYPE_LAST_AL	= 3,
+};
+
+/* MSS and Completion Req */
+#define FBNIC_TWD_MSS_MASK			DESC_GENMASK(61, 48)
+
+#define FBNIC_TWD_TS_MASK			DESC_GENMASK(39, 0)
+#define FBNIC_TWD_ADDR_MASK			DESC_GENMASK(45, 0)
+#define FBNIC_TWD_LEN_MASK			DESC_GENMASK(63, 48)
+
+/* Tx Completion Descriptor Format */
+#define FBNIC_TCD_TYPE0_HEAD0_MASK		DESC_GENMASK(15, 0)
+#define FBNIC_TCD_TYPE0_HEAD1_MASK		DESC_GENMASK(31, 16)
+
+#define FBNIC_TCD_TYPE1_TS_MASK			DESC_GENMASK(39, 0)
+
+#define FBNIC_TCD_STATUS_MASK			DESC_GENMASK(59, 48)
+#define FBNIC_TCD_STATUS_TS_INVALID		DESC_BIT(48)
+#define FBNIC_TCD_STATUS_ILLEGAL_TS_REQ		DESC_BIT(49)
+#define FBNIC_TCD_TWQ1				DESC_BIT(60)
+#define FBNIC_TCD_TYPE_MASK			DESC_GENMASK(62, 61)
+enum {
+	FBNIC_TCD_TYPE_0	= 0,
+	FBNIC_TCD_TYPE_1	= 1,
+};
+
+#define FBNIC_TCD_DONE				DESC_BIT(63)
+
 /* Rx Buffer Descriptor Format
  *
  * The layout of this can vary depending on the page size of the system.
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
index 18953a2d1eef..9546c302111a 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
@@ -91,6 +91,7 @@ static const struct net_device_ops fbnic_netdev_ops = {
 	.ndo_stop		= fbnic_stop,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_start_xmit		= fbnic_xmit_frame,
+	.ndo_features_check	= fbnic_features_check,
 };
 
 void fbnic_reset_queues(struct fbnic_net *fbn,
@@ -169,6 +170,14 @@ struct net_device *fbnic_netdev_alloc(struct fbnic_dev *fbd)
 
 	fbnic_reset_queues(fbn, default_queues, default_queues);
 
+	netdev->features |=
+		NETIF_F_SG |
+		NETIF_F_HW_CSUM;
+
+	netdev->hw_features |= netdev->features;
+	netdev->vlan_features |= netdev->features;
+	netdev->hw_enc_features |= netdev->features;
+
 	netdev->min_mtu = IPV6_MIN_MTU;
 	netdev->max_mtu = FBNIC_MAX_JUMBO_FRAME_SIZE - ETH_HLEN;
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
index 5fdc0a404f5a..fb65c92013b1 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) Meta Platforms, Inc. and affiliates. */
 
+#include <linux/bitfield.h>
 #include <linux/iopoll.h>
 #include <linux/pci.h>
 #include <net/netdev_queues.h>
@@ -10,6 +11,14 @@
 #include "fbnic_netdev.h"
 #include "fbnic_txrx.h"
 
+struct fbnic_xmit_cb {
+	u32 bytecount;
+	u8 desc_count;
+	int hw_head;
+};
+
+#define FBNIC_XMIT_CB(__skb) ((struct fbnic_xmit_cb *)((__skb)->cb))
+
 static u32 __iomem *fbnic_ring_csr_base(const struct fbnic_ring *ring)
 {
 	unsigned long csr_base = (unsigned long)ring->doorbell;
@@ -38,12 +47,307 @@ static unsigned int fbnic_desc_unused(struct fbnic_ring *ring)
 	return (ring->head - ring->tail - 1) & ring->size_mask;
 }
 
-netdev_tx_t fbnic_xmit_frame(struct sk_buff *skb, struct net_device *dev)
+static struct netdev_queue *txring_txq(const struct net_device *dev,
+				       const struct fbnic_ring *ring)
+{
+	return netdev_get_tx_queue(dev, ring->q_idx);
+}
+
+static int fbnic_maybe_stop_tx(const struct net_device *dev,
+			       struct fbnic_ring *ring,
+			       const unsigned int size)
+{
+	struct netdev_queue *txq = txring_txq(dev, ring);
+	int res;
+
+	res = netif_txq_maybe_stop(txq, fbnic_desc_unused(ring), size,
+				   FBNIC_TX_DESC_WAKEUP);
+
+	return !res;
+}
+
+static bool fbnic_tx_sent_queue(struct sk_buff *skb, struct fbnic_ring *ring)
+{
+	struct netdev_queue *dev_queue = txring_txq(skb->dev, ring);
+	unsigned int bytecount = FBNIC_XMIT_CB(skb)->bytecount;
+	bool xmit_more = netdev_xmit_more();
+
+	/* TBD: Request completion more often if xmit_more becomes large */
+
+	return __netdev_tx_sent_queue(dev_queue, bytecount, xmit_more);
+}
+
+static void fbnic_unmap_single_twd(struct device *dev, __le64 *twd)
+{
+	u64 raw_twd = le64_to_cpu(*twd);
+	unsigned int len;
+	dma_addr_t dma;
+
+	dma = FIELD_GET(FBNIC_TWD_ADDR_MASK, raw_twd);
+	len = FIELD_GET(FBNIC_TWD_LEN_MASK, raw_twd);
+
+	dma_unmap_single(dev, dma, len, DMA_TO_DEVICE);
+}
+
+static void fbnic_unmap_page_twd(struct device *dev, __le64 *twd)
+{
+	u64 raw_twd = le64_to_cpu(*twd);
+	unsigned int len;
+	dma_addr_t dma;
+
+	dma = FIELD_GET(FBNIC_TWD_ADDR_MASK, raw_twd);
+	len = FIELD_GET(FBNIC_TWD_LEN_MASK, raw_twd);
+
+	dma_unmap_page(dev, dma, len, DMA_TO_DEVICE);
+}
+
+#define FBNIC_TWD_TYPE(_type) \
+	cpu_to_le64(FIELD_PREP(FBNIC_TWD_TYPE_MASK, FBNIC_TWD_TYPE_##_type))
+
+static bool
+fbnic_tx_offloads(struct fbnic_ring *ring, struct sk_buff *skb, __le64 *meta)
 {
+	unsigned int l2len, i3len;
+
+	if (unlikely(skb->ip_summed != CHECKSUM_PARTIAL))
+		return false;
+
+	l2len = skb_mac_header_len(skb);
+	i3len = skb_checksum_start(skb) - skb_network_header(skb);
+
+	*meta |= cpu_to_le64(FIELD_PREP(FBNIC_TWD_CSUM_OFFSET_MASK,
+					skb->csum_offset / 2));
+
+	*meta |= cpu_to_le64(FBNIC_TWD_FLAG_REQ_CSO);
+
+	*meta |= cpu_to_le64(FIELD_PREP(FBNIC_TWD_L2_HLEN_MASK, l2len / 2) |
+			     FIELD_PREP(FBNIC_TWD_L3_IHLEN_MASK, i3len / 2));
+	return false;
+}
+
+static bool
+fbnic_tx_map(struct fbnic_ring *ring, struct sk_buff *skb, __le64 *meta)
+{
+	struct device *dev = skb->dev->dev.parent;
+	unsigned int tail = ring->tail, first;
+	unsigned int size, data_len;
+	skb_frag_t *frag;
+	dma_addr_t dma;
+	__le64 *twd;
+
+	ring->tx_buf[tail] = skb;
+
+	tail++;
+	tail &= ring->size_mask;
+	first = tail;
+
+	size = skb_headlen(skb);
+	data_len = skb->data_len;
+
+	if (size > FIELD_MAX(FBNIC_TWD_LEN_MASK))
+		goto dma_error;
+
+	dma = dma_map_single(dev, skb->data, size, DMA_TO_DEVICE);
+
+	for (frag = &skb_shinfo(skb)->frags[0];; frag++) {
+		twd = &ring->desc[tail];
+
+		if (dma_mapping_error(dev, dma))
+			goto dma_error;
+
+		*twd = cpu_to_le64(FIELD_PREP(FBNIC_TWD_ADDR_MASK, dma) |
+				   FIELD_PREP(FBNIC_TWD_LEN_MASK, size) |
+				   FIELD_PREP(FBNIC_TWD_TYPE_MASK,
+					      FBNIC_TWD_TYPE_AL));
+
+		tail++;
+		tail &= ring->size_mask;
+
+		if (!data_len)
+			break;
+
+		size = skb_frag_size(frag);
+		data_len -= size;
+
+		if (size > FIELD_MAX(FBNIC_TWD_LEN_MASK))
+			goto dma_error;
+
+		dma = skb_frag_dma_map(dev, frag, 0, size, DMA_TO_DEVICE);
+	}
+
+	*twd |= FBNIC_TWD_TYPE(LAST_AL);
+
+	FBNIC_XMIT_CB(skb)->desc_count = ((twd - meta) + 1) & ring->size_mask;
+
+	ring->tail = tail;
+
+	/* Verify there is room for another packet */
+	fbnic_maybe_stop_tx(skb->dev, ring, FBNIC_MAX_SKB_DESC);
+
+	if (fbnic_tx_sent_queue(skb, ring)) {
+		*meta |= cpu_to_le64(FBNIC_TWD_FLAG_REQ_COMPLETION);
+
+		/* Force DMA writes to flush before writing to tail */
+		dma_wmb();
+
+		writel(tail, ring->doorbell);
+	}
+
+	return false;
+dma_error:
+	if (net_ratelimit())
+		netdev_err(skb->dev, "TX DMA map failed\n");
+
+	while (tail != first) {
+		tail--;
+		tail &= ring->size_mask;
+		twd = &ring->desc[tail];
+		if (tail == first)
+			fbnic_unmap_single_twd(dev, twd);
+		else
+			fbnic_unmap_page_twd(dev, twd);
+	}
+
+	return true;
+}
+
+#define FBNIC_MIN_FRAME_LEN	60
+
+static netdev_tx_t
+fbnic_xmit_frame_ring(struct sk_buff *skb, struct fbnic_ring *ring)
+{
+	__le64 *meta = &ring->desc[ring->tail];
+	u16 desc_needed;
+
+	if (skb_put_padto(skb, FBNIC_MIN_FRAME_LEN))
+		goto err_count;
+
+	/* Need: 1 descriptor per page,
+	 *       + 1 desc for skb_head,
+	 *       + 2 desc for metadata and timestamp metadata
+	 *       + 7 desc gap to keep tail from touching head
+	 * otherwise try next time
+	 */
+	desc_needed = skb_shinfo(skb)->nr_frags + 10;
+	if (fbnic_maybe_stop_tx(skb->dev, ring, desc_needed))
+		return NETDEV_TX_BUSY;
+
+	*meta = cpu_to_le64(FBNIC_TWD_FLAG_DEST_MAC);
+
+	/* Write all members within DWORD to condense this into 2 4B writes */
+	FBNIC_XMIT_CB(skb)->bytecount = skb->len;
+	FBNIC_XMIT_CB(skb)->desc_count = 0;
+
+	if (fbnic_tx_offloads(ring, skb, meta))
+		goto err_free;
+
+	if (fbnic_tx_map(ring, skb, meta))
+		goto err_free;
+
+	return NETDEV_TX_OK;
+
+err_free:
 	dev_kfree_skb_any(skb);
+err_count:
 	return NETDEV_TX_OK;
 }
 
+netdev_tx_t fbnic_xmit_frame(struct sk_buff *skb, struct net_device *dev)
+{
+	struct fbnic_net *fbn = netdev_priv(dev);
+	unsigned int q_map = skb->queue_mapping;
+
+	return fbnic_xmit_frame_ring(skb, fbn->tx[q_map]);
+}
+
+netdev_features_t
+fbnic_features_check(struct sk_buff *skb, struct net_device *dev,
+		     netdev_features_t features)
+{
+	unsigned int l2len, l3len;
+
+	if (unlikely(skb->ip_summed != CHECKSUM_PARTIAL))
+		return features;
+
+	l2len = skb_mac_header_len(skb);
+	l3len = skb_checksum_start(skb) - skb_network_header(skb);
+
+	/* Check header lengths are multiple of 2.
+	 * In case of 6in6 we support longer headers (IHLEN + OHLEN)
+	 * but keep things simple for now, 512B is plenty.
+	 */
+	if ((l2len | l3len | skb->csum_offset) % 2 ||
+	    !FIELD_FIT(FBNIC_TWD_L2_HLEN_MASK, l2len / 2) ||
+	    !FIELD_FIT(FBNIC_TWD_L3_IHLEN_MASK, l3len / 2) ||
+	    !FIELD_FIT(FBNIC_TWD_CSUM_OFFSET_MASK, skb->csum_offset / 2))
+		return features & ~NETIF_F_CSUM_MASK;
+
+	return features;
+}
+
+static void fbnic_clean_twq0(struct fbnic_napi_vector *nv, int napi_budget,
+			     struct fbnic_ring *ring, bool discard,
+			     unsigned int hw_head)
+{
+	u64 total_bytes = 0, total_packets = 0;
+	unsigned int head = ring->head;
+	struct netdev_queue *txq;
+	unsigned int clean_desc;
+
+	clean_desc = (hw_head - head) & ring->size_mask;
+
+	while (clean_desc) {
+		struct sk_buff *skb = ring->tx_buf[head];
+		unsigned int desc_cnt;
+
+		desc_cnt = FBNIC_XMIT_CB(skb)->desc_count;
+		if (desc_cnt > clean_desc)
+			break;
+
+		ring->tx_buf[head] = NULL;
+
+		clean_desc -= desc_cnt;
+
+		while (!(ring->desc[head] & FBNIC_TWD_TYPE(AL))) {
+			head++;
+			head &= ring->size_mask;
+			desc_cnt--;
+		}
+
+		fbnic_unmap_single_twd(nv->dev, &ring->desc[head]);
+		head++;
+		head &= ring->size_mask;
+		desc_cnt--;
+
+		while (desc_cnt--) {
+			fbnic_unmap_page_twd(nv->dev, &ring->desc[head]);
+			head++;
+			head &= ring->size_mask;
+		}
+
+		total_bytes += FBNIC_XMIT_CB(skb)->bytecount;
+		total_packets += 1;
+
+		napi_consume_skb(skb, napi_budget);
+	}
+
+	if (!total_bytes)
+		return;
+
+	ring->head = head;
+
+	txq = txring_txq(nv->napi.dev, ring);
+
+	if (unlikely(discard)) {
+		netdev_tx_completed_queue(txq, total_packets, total_bytes);
+		return;
+	}
+
+	netif_txq_completed_wake(txq, total_packets, total_bytes,
+				 fbnic_desc_unused(ring),
+				 FBNIC_TX_DESC_WAKEUP);
+}
+
 static void fbnic_page_pool_init(struct fbnic_ring *ring, unsigned int idx,
 				 struct page *page)
 {
@@ -66,6 +370,65 @@ static void fbnic_page_pool_drain(struct fbnic_ring *ring, unsigned int idx,
 	rx_buf->page = NULL;
 }
 
+static void fbnic_clean_twq(struct fbnic_napi_vector *nv, int napi_budget,
+			    struct fbnic_q_triad *qt, s32 head0)
+{
+	if (head0 >= 0)
+		fbnic_clean_twq0(nv, napi_budget, &qt->sub0, false, head0);
+}
+
+static void
+fbnic_clean_tcq(struct fbnic_napi_vector *nv, struct fbnic_q_triad *qt,
+		int napi_budget)
+{
+	struct fbnic_ring *cmpl = &qt->cmpl;
+	__le64 *raw_tcd, done;
+	u32 head = cmpl->head;
+	s32 head0 = -1;
+
+	done = (head & (cmpl->size_mask + 1)) ? 0 : cpu_to_le64(FBNIC_TCD_DONE);
+	raw_tcd = &cmpl->desc[head & cmpl->size_mask];
+
+	/* Walk the completion queue collecting the heads reported by NIC */
+	while ((*raw_tcd & cpu_to_le64(FBNIC_TCD_DONE)) == done) {
+		u64 tcd;
+
+		dma_rmb();
+
+		tcd = le64_to_cpu(*raw_tcd);
+
+		switch (FIELD_GET(FBNIC_TCD_TYPE_MASK, tcd)) {
+		case FBNIC_TCD_TYPE_0:
+			if (!(tcd & FBNIC_TCD_TWQ1))
+				head0 = FIELD_GET(FBNIC_TCD_TYPE0_HEAD0_MASK,
+						  tcd);
+			/* Currently all err status bits are related to
+			 * timestamps and as those have yet to be added
+			 * they are skipped for now.
+			 */
+			break;
+		default:
+			break;
+		}
+
+		raw_tcd++;
+		head++;
+		if (!(head & cmpl->size_mask)) {
+			done ^= cpu_to_le64(FBNIC_TCD_DONE);
+			raw_tcd = &cmpl->desc[0];
+		}
+	}
+
+	/* Record the current head/tail of the queue */
+	if (cmpl->head != head) {
+		cmpl->head = head;
+		writel(head & cmpl->size_mask, cmpl->doorbell);
+	}
+
+	/* Unmap and free processed buffers */
+	fbnic_clean_twq(nv, napi_budget, qt, head0);
+}
+
 static void fbnic_clean_bdq(struct fbnic_napi_vector *nv, int napi_budget,
 			    struct fbnic_ring *ring, unsigned int hw_head)
 {
@@ -169,8 +532,28 @@ static void fbnic_nv_irq_disable(struct fbnic_napi_vector *nv)
 	fbnic_wr32(fbd, FBNIC_INTR_MASK_SET(v_idx / 32), 1 << (v_idx % 32));
 }
 
+static void fbnic_nv_irq_rearm(struct fbnic_napi_vector *nv)
+{
+	struct fbnic_dev *fbd = nv->fbd;
+	u32 v_idx = nv->v_idx;
+
+	fbnic_wr32(fbd, FBNIC_INTR_CQ_REARM(v_idx),
+		   FBNIC_INTR_CQ_REARM_INTR_UNMASK);
+}
+
 static int fbnic_poll(struct napi_struct *napi, int budget)
 {
+	struct fbnic_napi_vector *nv = container_of(napi,
+						    struct fbnic_napi_vector,
+						    napi);
+	int i;
+
+	for (i = 0; i < nv->txt_count; i++)
+		fbnic_clean_tcq(nv, &nv->qt[i], budget);
+
+	if (likely(napi_complete_done(napi, 0)))
+		fbnic_nv_irq_rearm(nv);
+
 	return 0;
 }
 
@@ -911,6 +1294,9 @@ void fbnic_flush(struct fbnic_net *fbn)
 			struct fbnic_q_triad *qt = &nv->qt[i];
 			struct netdev_queue *tx_queue;
 
+			/* Clean the work queues of unprocessed work */
+			fbnic_clean_twq0(nv, 0, &qt->sub0, true, qt->sub0.tail);
+
 			/* Reset completion queue descriptor ring */
 			memset(qt->cmpl.desc, 0, qt->cmpl.size);
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
index 0e681dcc85c1..ade502e391b7 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
@@ -10,6 +10,18 @@
 
 struct fbnic_net;
 
+/* Guarantee we have space needed for storing the buffer
+ * To store the buffer we need:
+ *	1 descriptor per page
+ *	+ 1 descriptor for skb head
+ *	+ 2 descriptors for metadata and optional metadata
+ *	+ 7 descriptors to keep tail out of the same cacheline as head
+ * If we cannot guarantee that then we should return TX_BUSY
+ */
+#define FBNIC_MAX_SKB_DESC	(MAX_SKB_FRAGS + 10)
+#define FBNIC_TX_DESC_WAKEUP	(FBNIC_MAX_SKB_DESC * 2)
+#define FBNIC_TX_DESC_MIN	roundup_pow_of_two(FBNIC_TX_DESC_WAKEUP)
+
 #define FBNIC_MAX_TXQS			128u
 #define FBNIC_MAX_RXQS			128u
 
@@ -93,6 +105,9 @@ struct fbnic_napi_vector {
 #define FBNIC_MAX_RXQS			128u
 
 netdev_tx_t fbnic_xmit_frame(struct sk_buff *skb, struct net_device *dev);
+netdev_features_t
+fbnic_features_check(struct sk_buff *skb, struct net_device *dev,
+		     netdev_features_t features);
 
 int fbnic_alloc_napi_vectors(struct fbnic_net *fbn);
 void fbnic_free_napi_vectors(struct fbnic_net *fbn);
-- 
cgit v1.2.3-58-ga151


From a29b8eb6e5337f569f4a08937573dce7cc183493 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:50:28 -0700
Subject: eth: fbnic: Add basic Rx handling

Handle Rx packets with basic csum and Rx hash offloads.

NIC writes back to the completion ring a head buffer descriptor
(data buffer allocated from header pages), variable number of payload
descriptors (data buffers in payload pages), an optional metadata
descriptor (type 2) and finally the primary metadata descriptor
(type 3).

This format makes scatter support fairly easy - start gathering
the pages when we see head page, gather until we see the primary
metadata descriptor, do the processing. Use XDP infra to collect
the packet fragments as we traverse the descriptors. XDP itself
is not supported yet, but it will be soon.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079942839.1778861.10509071985738726125.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h    |  63 +++++
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.c |   4 +-
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c    |   3 +
 drivers/net/ethernet/meta/fbnic/fbnic_txrx.c   | 333 ++++++++++++++++++++++++-
 drivers/net/ethernet/meta/fbnic/fbnic_txrx.h   |   2 +
 5 files changed, 402 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index d7ce6781b9dc..405c294af0df 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -121,6 +121,69 @@ enum {
 #define FBNIC_BD_PAGE_ID_MASK \
 	(FBNIC_BD_DESC_ID_MASK & ~FBNIC_BD_FRAG_ID_MASK)
 
+/* Rx Completion Queue Descriptors */
+#define FBNIC_RCD_TYPE_MASK			DESC_GENMASK(62, 61)
+enum {
+	FBNIC_RCD_TYPE_HDR_AL	= 0,
+	FBNIC_RCD_TYPE_PAY_AL	= 1,
+	FBNIC_RCD_TYPE_OPT_META	= 2,
+	FBNIC_RCD_TYPE_META	= 3,
+};
+
+#define FBNIC_RCD_DONE				DESC_BIT(63)
+
+/* Address/Length Completion Descriptors */
+#define FBNIC_RCD_AL_BUFF_ID_MASK		DESC_GENMASK(15, 0)
+#define FBNIC_RCD_AL_BUFF_FRAG_MASK		(FBNIC_BD_FRAG_COUNT - 1)
+#define FBNIC_RCD_AL_BUFF_PAGE_MASK \
+	(FBNIC_RCD_AL_BUFF_ID_MASK & ~FBNIC_RCD_AL_BUFF_FRAG_MASK)
+#define FBNIC_RCD_AL_BUFF_LEN_MASK		DESC_GENMASK(28, 16)
+#define FBNIC_RCD_AL_BUFF_OFF_MASK		DESC_GENMASK(43, 32)
+#define FBNIC_RCD_AL_PAGE_FIN			DESC_BIT(60)
+
+/* Header AL specific values */
+#define FBNIC_RCD_HDR_AL_OVERFLOW		DESC_BIT(53)
+#define FBNIC_RCD_HDR_AL_DMA_HINT_MASK		DESC_GENMASK(59, 54)
+enum {
+	FBNIC_RCD_HDR_AL_DMA_HINT_NONE  = 0,
+	FBNIC_RCD_HDR_AL_DMA_HINT_L2	= 1,
+	FBNIC_RCD_HDR_AL_DMA_HINT_L3	= 2,
+	FBNIC_RCD_HDR_AL_DMA_HINT_L4	= 4,
+};
+
+/* Optional Metadata Completion Descriptors */
+#define FBNIC_RCD_OPT_META_TS_MASK		DESC_GENMASK(39, 0)
+#define FBNIC_RCD_OPT_META_ACTION_MASK		DESC_GENMASK(45, 40)
+#define FBNIC_RCD_OPT_META_ACTION		DESC_BIT(57)
+#define FBNIC_RCD_OPT_META_TS			DESC_BIT(58)
+#define FBNIC_RCD_OPT_META_TYPE_MASK		DESC_GENMASK(60, 59)
+
+/* Metadata Completion Descriptors */
+#define FBNIC_RCD_META_RSS_HASH_MASK		DESC_GENMASK(31, 0)
+#define FBNIC_RCD_META_L2_CSUM_MASK		DESC_GENMASK(47, 32)
+#define FBNIC_RCD_META_L3_TYPE_MASK		DESC_GENMASK(49, 48)
+enum {
+	FBNIC_RCD_META_L3_TYPE_OTHER	= 0,
+	FBNIC_RCD_META_L3_TYPE_IPV4	= 1,
+	FBNIC_RCD_META_L3_TYPE_IPV6	= 2,
+	FBNIC_RCD_META_L3_TYPE_V6V6	= 3,
+};
+
+#define FBNIC_RCD_META_L4_TYPE_MASK		DESC_GENMASK(51, 50)
+enum {
+	FBNIC_RCD_META_L4_TYPE_OTHER	= 0,
+	FBNIC_RCD_META_L4_TYPE_TCP	= 1,
+	FBNIC_RCD_META_L4_TYPE_UDP	= 2,
+};
+
+#define FBNIC_RCD_META_L4_CSUM_UNNECESSARY	DESC_BIT(52)
+#define FBNIC_RCD_META_ERR_MAC_EOP		DESC_BIT(53)
+#define FBNIC_RCD_META_ERR_TRUNCATED_FRAME	DESC_BIT(54)
+#define FBNIC_RCD_META_ERR_PARSER		DESC_BIT(55)
+#define FBNIC_RCD_META_UNCORRECTABLE_ERR_MASK	\
+	(FBNIC_RCD_META_ERR_MAC_EOP | FBNIC_RCD_META_ERR_TRUNCATED_FRAME)
+#define FBNIC_RCD_META_ECN			DESC_BIT(60)
+
 /* Register Definitions
  *
  * The registers are laid as indexes into an le32 array. As such the actual
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
index 9546c302111a..3cad480860aa 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
@@ -171,8 +171,10 @@ struct net_device *fbnic_netdev_alloc(struct fbnic_dev *fbd)
 	fbnic_reset_queues(fbn, default_queues, default_queues);
 
 	netdev->features |=
+		NETIF_F_RXHASH |
 		NETIF_F_SG |
-		NETIF_F_HW_CSUM;
+		NETIF_F_HW_CSUM |
+		NETIF_F_RXCSUM;
 
 	netdev->hw_features |= netdev->features;
 	netdev->vlan_features |= netdev->features;
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
index 040aaa127500..6f6c2b1d9c99 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -195,6 +195,9 @@ static void fbnic_service_task(struct work_struct *work)
 
 	fbnic_health_check(fbd);
 
+	if (netif_carrier_ok(fbd->netdev))
+		fbnic_napi_depletion_check(fbd->netdev);
+
 	if (netif_running(fbd->netdev))
 		schedule_delayed_work(&fbd->service_task, HZ);
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
index fb65c92013b1..b1a471fac4fe 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
@@ -8,6 +8,7 @@
 #include <net/page_pool/helpers.h>
 
 #include "fbnic.h"
+#include "fbnic_csr.h"
 #include "fbnic_netdev.h"
 #include "fbnic_txrx.h"
 
@@ -47,6 +48,11 @@ static unsigned int fbnic_desc_unused(struct fbnic_ring *ring)
 	return (ring->head - ring->tail - 1) & ring->size_mask;
 }
 
+static unsigned int fbnic_desc_used(struct fbnic_ring *ring)
+{
+	return (ring->tail - ring->head) & ring->size_mask;
+}
+
 static struct netdev_queue *txring_txq(const struct net_device *dev,
 				       const struct fbnic_ring *ring)
 {
@@ -125,6 +131,24 @@ fbnic_tx_offloads(struct fbnic_ring *ring, struct sk_buff *skb, __le64 *meta)
 	return false;
 }
 
+static void
+fbnic_rx_csum(u64 rcd, struct sk_buff *skb, struct fbnic_ring *rcq)
+{
+	skb_checksum_none_assert(skb);
+
+	if (unlikely(!(skb->dev->features & NETIF_F_RXCSUM)))
+		return;
+
+	if (FIELD_GET(FBNIC_RCD_META_L4_CSUM_UNNECESSARY, rcd)) {
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	} else {
+		u16 csum = FIELD_GET(FBNIC_RCD_META_L2_CSUM_MASK, rcd);
+
+		skb->ip_summed = CHECKSUM_COMPLETE;
+		skb->csum = (__force __wsum)csum;
+	}
+}
+
 static bool
 fbnic_tx_map(struct fbnic_ring *ring, struct sk_buff *skb, __le64 *meta)
 {
@@ -358,6 +382,16 @@ static void fbnic_page_pool_init(struct fbnic_ring *ring, unsigned int idx,
 	rx_buf->page = page;
 }
 
+static struct page *fbnic_page_pool_get(struct fbnic_ring *ring,
+					unsigned int idx)
+{
+	struct fbnic_rx_buf *rx_buf = &ring->rx_buf[idx];
+
+	rx_buf->pagecnt_bias--;
+
+	return rx_buf->page;
+}
+
 static void fbnic_page_pool_drain(struct fbnic_ring *ring, unsigned int idx,
 				  struct fbnic_napi_vector *nv, int budget)
 {
@@ -502,6 +536,103 @@ static void fbnic_fill_bdq(struct fbnic_napi_vector *nv, struct fbnic_ring *bdq)
 	}
 }
 
+static unsigned int fbnic_hdr_pg_start(unsigned int pg_off)
+{
+	/* The headroom of the first header may be larger than FBNIC_RX_HROOM
+	 * due to alignment. So account for that by just making the page
+	 * offset 0 if we are starting at the first header.
+	 */
+	if (ALIGN(FBNIC_RX_HROOM, 128) > FBNIC_RX_HROOM &&
+	    pg_off == ALIGN(FBNIC_RX_HROOM, 128))
+		return 0;
+
+	return pg_off - FBNIC_RX_HROOM;
+}
+
+static unsigned int fbnic_hdr_pg_end(unsigned int pg_off, unsigned int len)
+{
+	/* Determine the end of the buffer by finding the start of the next
+	 * and then subtracting the headroom from that frame.
+	 */
+	pg_off += len + FBNIC_RX_TROOM + FBNIC_RX_HROOM;
+
+	return ALIGN(pg_off, 128) - FBNIC_RX_HROOM;
+}
+
+static void fbnic_pkt_prepare(struct fbnic_napi_vector *nv, u64 rcd,
+			      struct fbnic_pkt_buff *pkt,
+			      struct fbnic_q_triad *qt)
+{
+	unsigned int hdr_pg_idx = FIELD_GET(FBNIC_RCD_AL_BUFF_PAGE_MASK, rcd);
+	unsigned int hdr_pg_off = FIELD_GET(FBNIC_RCD_AL_BUFF_OFF_MASK, rcd);
+	struct page *page = fbnic_page_pool_get(&qt->sub0, hdr_pg_idx);
+	unsigned int len = FIELD_GET(FBNIC_RCD_AL_BUFF_LEN_MASK, rcd);
+	unsigned int frame_sz, hdr_pg_start, hdr_pg_end, headroom;
+	unsigned char *hdr_start;
+
+	/* data_hard_start should always be NULL when this is called */
+	WARN_ON_ONCE(pkt->buff.data_hard_start);
+
+	/* Short-cut the end calculation if we know page is fully consumed */
+	hdr_pg_end = FIELD_GET(FBNIC_RCD_AL_PAGE_FIN, rcd) ?
+		     FBNIC_BD_FRAG_SIZE : fbnic_hdr_pg_end(hdr_pg_off, len);
+	hdr_pg_start = fbnic_hdr_pg_start(hdr_pg_off);
+
+	headroom = hdr_pg_off - hdr_pg_start + FBNIC_RX_PAD;
+	frame_sz = hdr_pg_end - hdr_pg_start;
+	xdp_init_buff(&pkt->buff, frame_sz, NULL);
+	hdr_pg_start += (FBNIC_RCD_AL_BUFF_FRAG_MASK & rcd) *
+			FBNIC_BD_FRAG_SIZE;
+
+	/* Sync DMA buffer */
+	dma_sync_single_range_for_cpu(nv->dev, page_pool_get_dma_addr(page),
+				      hdr_pg_start, frame_sz,
+				      DMA_BIDIRECTIONAL);
+
+	/* Build frame around buffer */
+	hdr_start = page_address(page) + hdr_pg_start;
+
+	xdp_prepare_buff(&pkt->buff, hdr_start, headroom,
+			 len - FBNIC_RX_PAD, true);
+
+	pkt->data_truesize = 0;
+	pkt->data_len = 0;
+	pkt->nr_frags = 0;
+}
+
+static void fbnic_add_rx_frag(struct fbnic_napi_vector *nv, u64 rcd,
+			      struct fbnic_pkt_buff *pkt,
+			      struct fbnic_q_triad *qt)
+{
+	unsigned int pg_idx = FIELD_GET(FBNIC_RCD_AL_BUFF_PAGE_MASK, rcd);
+	unsigned int pg_off = FIELD_GET(FBNIC_RCD_AL_BUFF_OFF_MASK, rcd);
+	unsigned int len = FIELD_GET(FBNIC_RCD_AL_BUFF_LEN_MASK, rcd);
+	struct page *page = fbnic_page_pool_get(&qt->sub1, pg_idx);
+	struct skb_shared_info *shinfo;
+	unsigned int truesize;
+
+	truesize = FIELD_GET(FBNIC_RCD_AL_PAGE_FIN, rcd) ?
+		   FBNIC_BD_FRAG_SIZE - pg_off : ALIGN(len, 128);
+
+	pg_off += (FBNIC_RCD_AL_BUFF_FRAG_MASK & rcd) *
+		  FBNIC_BD_FRAG_SIZE;
+
+	/* Sync DMA buffer */
+	dma_sync_single_range_for_cpu(nv->dev, page_pool_get_dma_addr(page),
+				      pg_off, truesize, DMA_BIDIRECTIONAL);
+
+	/* Add page to xdp shared info */
+	shinfo = xdp_get_shared_info_from_buff(&pkt->buff);
+
+	/* We use gso_segs to store truesize */
+	pkt->data_truesize += truesize;
+
+	__skb_fill_page_desc_noacc(shinfo, pkt->nr_frags++, page, pg_off, len);
+
+	/* Store data_len in gso_size */
+	pkt->data_len += len;
+}
+
 static void fbnic_put_pkt_buff(struct fbnic_napi_vector *nv,
 			       struct fbnic_pkt_buff *pkt, int budget)
 {
@@ -524,6 +655,168 @@ static void fbnic_put_pkt_buff(struct fbnic_napi_vector *nv,
 	page_pool_put_full_page(nv->page_pool, page, !!budget);
 }
 
+static struct sk_buff *fbnic_build_skb(struct fbnic_napi_vector *nv,
+				       struct fbnic_pkt_buff *pkt)
+{
+	unsigned int nr_frags = pkt->nr_frags;
+	struct skb_shared_info *shinfo;
+	unsigned int truesize;
+	struct sk_buff *skb;
+
+	truesize = xdp_data_hard_end(&pkt->buff) + FBNIC_RX_TROOM -
+		   pkt->buff.data_hard_start;
+
+	/* Build frame around buffer */
+	skb = napi_build_skb(pkt->buff.data_hard_start, truesize);
+	if (unlikely(!skb))
+		return NULL;
+
+	/* Push data pointer to start of data, put tail to end of data */
+	skb_reserve(skb, pkt->buff.data - pkt->buff.data_hard_start);
+	__skb_put(skb, pkt->buff.data_end - pkt->buff.data);
+
+	/* Add tracking for metadata at the start of the frame */
+	skb_metadata_set(skb, pkt->buff.data - pkt->buff.data_meta);
+
+	/* Add Rx frags */
+	if (nr_frags) {
+		/* Verify that shared info didn't move */
+		shinfo = xdp_get_shared_info_from_buff(&pkt->buff);
+		WARN_ON(skb_shinfo(skb) != shinfo);
+
+		skb->truesize += pkt->data_truesize;
+		skb->data_len += pkt->data_len;
+		shinfo->nr_frags = nr_frags;
+		skb->len += pkt->data_len;
+	}
+
+	skb_mark_for_recycle(skb);
+
+	/* Set MAC header specific fields */
+	skb->protocol = eth_type_trans(skb, nv->napi.dev);
+
+	return skb;
+}
+
+static enum pkt_hash_types fbnic_skb_hash_type(u64 rcd)
+{
+	return (FBNIC_RCD_META_L4_TYPE_MASK & rcd) ? PKT_HASH_TYPE_L4 :
+	       (FBNIC_RCD_META_L3_TYPE_MASK & rcd) ? PKT_HASH_TYPE_L3 :
+						     PKT_HASH_TYPE_L2;
+}
+
+static void fbnic_populate_skb_fields(struct fbnic_napi_vector *nv,
+				      u64 rcd, struct sk_buff *skb,
+				      struct fbnic_q_triad *qt)
+{
+	struct net_device *netdev = nv->napi.dev;
+	struct fbnic_ring *rcq = &qt->cmpl;
+
+	fbnic_rx_csum(rcd, skb, rcq);
+
+	if (netdev->features & NETIF_F_RXHASH)
+		skb_set_hash(skb,
+			     FIELD_GET(FBNIC_RCD_META_RSS_HASH_MASK, rcd),
+			     fbnic_skb_hash_type(rcd));
+
+	skb_record_rx_queue(skb, rcq->q_idx);
+}
+
+static bool fbnic_rcd_metadata_err(u64 rcd)
+{
+	return !!(FBNIC_RCD_META_UNCORRECTABLE_ERR_MASK & rcd);
+}
+
+static int fbnic_clean_rcq(struct fbnic_napi_vector *nv,
+			   struct fbnic_q_triad *qt, int budget)
+{
+	struct fbnic_ring *rcq = &qt->cmpl;
+	struct fbnic_pkt_buff *pkt;
+	s32 head0 = -1, head1 = -1;
+	__le64 *raw_rcd, done;
+	u32 head = rcq->head;
+	u64 packets = 0;
+
+	done = (head & (rcq->size_mask + 1)) ? cpu_to_le64(FBNIC_RCD_DONE) : 0;
+	raw_rcd = &rcq->desc[head & rcq->size_mask];
+	pkt = rcq->pkt;
+
+	/* Walk the completion queue collecting the heads reported by NIC */
+	while (likely(packets < budget)) {
+		struct sk_buff *skb = ERR_PTR(-EINVAL);
+		u64 rcd;
+
+		if ((*raw_rcd & cpu_to_le64(FBNIC_RCD_DONE)) == done)
+			break;
+
+		dma_rmb();
+
+		rcd = le64_to_cpu(*raw_rcd);
+
+		switch (FIELD_GET(FBNIC_RCD_TYPE_MASK, rcd)) {
+		case FBNIC_RCD_TYPE_HDR_AL:
+			head0 = FIELD_GET(FBNIC_RCD_AL_BUFF_PAGE_MASK, rcd);
+			fbnic_pkt_prepare(nv, rcd, pkt, qt);
+
+			break;
+		case FBNIC_RCD_TYPE_PAY_AL:
+			head1 = FIELD_GET(FBNIC_RCD_AL_BUFF_PAGE_MASK, rcd);
+			fbnic_add_rx_frag(nv, rcd, pkt, qt);
+
+			break;
+		case FBNIC_RCD_TYPE_OPT_META:
+			/* Only type 0 is currently supported */
+			if (FIELD_GET(FBNIC_RCD_OPT_META_TYPE_MASK, rcd))
+				break;
+
+			/* We currently ignore the action table index */
+			break;
+		case FBNIC_RCD_TYPE_META:
+			if (likely(!fbnic_rcd_metadata_err(rcd)))
+				skb = fbnic_build_skb(nv, pkt);
+
+			/* Populate skb and invalidate XDP */
+			if (!IS_ERR_OR_NULL(skb)) {
+				fbnic_populate_skb_fields(nv, rcd, skb, qt);
+
+				packets++;
+
+				napi_gro_receive(&nv->napi, skb);
+			} else {
+				fbnic_put_pkt_buff(nv, pkt, 1);
+			}
+
+			pkt->buff.data_hard_start = NULL;
+
+			break;
+		}
+
+		raw_rcd++;
+		head++;
+		if (!(head & rcq->size_mask)) {
+			done ^= cpu_to_le64(FBNIC_RCD_DONE);
+			raw_rcd = &rcq->desc[0];
+		}
+	}
+
+	/* Unmap and free processed buffers */
+	if (head0 >= 0)
+		fbnic_clean_bdq(nv, budget, &qt->sub0, head0);
+	fbnic_fill_bdq(nv, &qt->sub0);
+
+	if (head1 >= 0)
+		fbnic_clean_bdq(nv, budget, &qt->sub1, head1);
+	fbnic_fill_bdq(nv, &qt->sub1);
+
+	/* Record the current head/tail of the queue */
+	if (rcq->head != head) {
+		rcq->head = head;
+		writel(head & rcq->size_mask, rcq->doorbell);
+	}
+
+	return packets;
+}
+
 static void fbnic_nv_irq_disable(struct fbnic_napi_vector *nv)
 {
 	struct fbnic_dev *fbd = nv->fbd;
@@ -546,12 +839,18 @@ static int fbnic_poll(struct napi_struct *napi, int budget)
 	struct fbnic_napi_vector *nv = container_of(napi,
 						    struct fbnic_napi_vector,
 						    napi);
-	int i;
+	int i, j, work_done = 0;
 
 	for (i = 0; i < nv->txt_count; i++)
 		fbnic_clean_tcq(nv, &nv->qt[i], budget);
 
-	if (likely(napi_complete_done(napi, 0)))
+	for (j = 0; j < nv->rxt_count; j++, i++)
+		work_done += fbnic_clean_rcq(nv, &nv->qt[i], budget);
+
+	if (work_done >= budget)
+		return budget;
+
+	if (likely(napi_complete_done(napi, work_done)))
 		fbnic_nv_irq_rearm(nv);
 
 	return 0;
@@ -1582,3 +1881,33 @@ void fbnic_napi_enable(struct fbnic_net *fbn)
 
 	fbnic_wrfl(fbd);
 }
+
+void fbnic_napi_depletion_check(struct net_device *netdev)
+{
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	u32 irqs[FBNIC_MAX_MSIX_VECS / 32] = {};
+	struct fbnic_dev *fbd = fbn->fbd;
+	struct fbnic_napi_vector *nv;
+	int i, j;
+
+	list_for_each_entry(nv, &fbn->napis, napis) {
+		/* Find RQs which are completely out of pages */
+		for (i = nv->txt_count, j = 0; j < nv->rxt_count; j++, i++) {
+			/* Assume 4 pages is always enough to fit a packet
+			 * and therefore generate a completion and an IRQ.
+			 */
+			if (fbnic_desc_used(&nv->qt[i].sub0) < 4 ||
+			    fbnic_desc_used(&nv->qt[i].sub1) < 4)
+				irqs[nv->v_idx / 32] |= BIT(nv->v_idx % 32);
+		}
+	}
+
+	for (i = 0; i < ARRAY_SIZE(irqs); i++) {
+		if (!irqs[i])
+			continue;
+		fbnic_wr32(fbd, FBNIC_INTR_MASK_CLEAR(i), irqs[i]);
+		fbnic_wr32(fbd, FBNIC_INTR_SET(i), irqs[i]);
+	}
+
+	fbnic_wrfl(fbd);
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
index ade502e391b7..4a206c0e7192 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.h
@@ -5,6 +5,7 @@
 #define _FBNIC_TXRX_H_
 
 #include <linux/netdevice.h>
+#include <linux/skbuff.h>
 #include <linux/types.h>
 #include <net/xdp.h>
 
@@ -120,6 +121,7 @@ void fbnic_disable(struct fbnic_net *fbn);
 void fbnic_flush(struct fbnic_net *fbn);
 void fbnic_fill(struct fbnic_net *fbn);
 
+void fbnic_napi_depletion_check(struct net_device *netdev);
 int fbnic_wait_all_queues_idle(struct fbnic_dev *fbd, bool may_fail);
 
 #endif /* _FBNIC_TXRX_H_ */
-- 
cgit v1.2.3-58-ga151


From eb690ef8d1c2a1b60e8c11721ace1538f027459e Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:50:32 -0700
Subject: eth: fbnic: Add L2 address programming

Program the Rx TCAM to control L2 forwarding. Since we are in full
control of the NIC we need to make sure we include BMC forwarding
in the rules. When host is not present BMC will program the TCAM
to get onto the network but once we take ownership it's up to
Linux driver to make sure BMC L2 addresses are handled correctly.

Co-developed-by: Sanman Pradhan <sanmanpradhan@meta.com>
Signed-off-by: Sanman Pradhan <sanmanpradhan@meta.com>
Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079943202.1778861.4410412697614789017.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/Makefile        |   1 +
 drivers/net/ethernet/meta/fbnic/fbnic.h         |   5 +
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h     |  14 ++
 drivers/net/ethernet/meta/fbnic/fbnic_devlink.c |   2 +
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.c  | 230 +++++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.h  |   3 +
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c     |   3 +
 drivers/net/ethernet/meta/fbnic/fbnic_rpc.c     | 281 ++++++++++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_rpc.h     | 139 ++++++++++++
 9 files changed, 678 insertions(+)
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_rpc.c
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_rpc.h

diff --git a/drivers/net/ethernet/meta/fbnic/Makefile b/drivers/net/ethernet/meta/fbnic/Makefile
index a487ac5c4ec5..9373b558fdc9 100644
--- a/drivers/net/ethernet/meta/fbnic/Makefile
+++ b/drivers/net/ethernet/meta/fbnic/Makefile
@@ -14,5 +14,6 @@ fbnic-y := fbnic_devlink.o \
 	   fbnic_netdev.o \
 	   fbnic_pci.o \
 	   fbnic_phylink.o \
+	   fbnic_rpc.o \
 	   fbnic_tlv.o \
 	   fbnic_txrx.o
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h
index 3539e090df69..0b8530117a82 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic.h
@@ -12,6 +12,7 @@
 #include "fbnic_csr.h"
 #include "fbnic_fw.h"
 #include "fbnic_mac.h"
+#include "fbnic_rpc.h"
 
 struct fbnic_dev {
 	struct device *dev;
@@ -39,6 +40,10 @@ struct fbnic_dev {
 	u32 mps;
 	u32 readrq;
 
+	/* Local copy of the devices TCAM */
+	struct fbnic_mac_addr mac_addr[FBNIC_RPC_TCAM_MACDA_NUM_ENTRIES];
+	u8 mac_addr_boundary;
+
 	/* Number of TCQs/RCQs available on hardware */
 	u16 max_num_queues;
 };
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index 405c294af0df..50259c60bf65 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -537,8 +537,22 @@ enum {
 #define FBNIC_RPC_RMI_CONFIG_FCS_PRESENT	CSR_BIT(8)
 #define FBNIC_RPC_RMI_CONFIG_ENABLE		CSR_BIT(12)
 #define FBNIC_RPC_RMI_CONFIG_MTU		CSR_GENMASK(31, 16)
+#define FBNIC_RPC_TCAM_MACDA_VALIDATE	0x0852d		/* 0x214b4 */
 #define FBNIC_CSR_END_RPC		0x0856b	/* CSR section delimiter */
 
+/* RPC RAM Registers */
+
+#define FBNIC_CSR_START_RPC_RAM		0x08800	/* CSR section delimiter */
+#define FBNIC_RPC_ACT_TBL_NUM_ENTRIES		64
+
+/* TCAM Tables */
+#define FBNIC_RPC_TCAM_VALIDATE			CSR_BIT(31)
+#define FBNIC_RPC_TCAM_MACDA(m, n) \
+	(0x08b80 + 0x20 * (n) + (m))		/* 0x022e00 + 128*n + 4*m */
+#define FBNIC_RPC_TCAM_MACDA_VALUE		CSR_GENMASK(15, 0)
+#define FBNIC_RPC_TCAM_MACDA_MASK		CSR_GENMASK(31, 16)
+#define FBNIC_CSR_END_RPC_RAM		0x08f1f	/* CSR section delimiter */
+
 /* Fab Registers */
 #define FBNIC_CSR_START_FAB		0x0C000 /* CSR section delimiter */
 #define FBNIC_FAB_AXI4_AR_SPACER_2_CFG		0x0C005		/* 0x30014 */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c b/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
index 1441610f5843..e87049dfd223 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
@@ -68,6 +68,8 @@ struct fbnic_dev *fbnic_devlink_alloc(struct pci_dev *pdev)
 	fbd->mps = pcie_get_mps(pdev);
 	fbd->readrq = pcie_get_readrq(pdev);
 
+	fbd->mac_addr_boundary = FBNIC_RPC_TCAM_MACDA_DEFAULT_BOUNDARY;
+
 	return fbd;
 }
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
index 3cad480860aa..371fc3ef8695 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
@@ -48,6 +48,8 @@ int __fbnic_open(struct fbnic_net *fbn)
 	err = fbnic_pcs_irq_enable(fbd);
 	if (err)
 		goto release_ownership;
+	/* Pull the BMC config and initialize the RPC */
+	fbnic_bmc_rpc_init(fbd);
 
 	return 0;
 release_ownership:
@@ -86,12 +88,240 @@ static int fbnic_stop(struct net_device *netdev)
 	return 0;
 }
 
+static int fbnic_uc_sync(struct net_device *netdev, const unsigned char *addr)
+{
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	struct fbnic_mac_addr *avail_addr;
+
+	if (WARN_ON(!is_valid_ether_addr(addr)))
+		return -EADDRNOTAVAIL;
+
+	avail_addr = __fbnic_uc_sync(fbn->fbd, addr);
+	if (!avail_addr)
+		return -ENOSPC;
+
+	/* Add type flag indicating this address is in use by the host */
+	set_bit(FBNIC_MAC_ADDR_T_UNICAST, avail_addr->act_tcam);
+
+	return 0;
+}
+
+static int fbnic_uc_unsync(struct net_device *netdev, const unsigned char *addr)
+{
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	struct fbnic_dev *fbd = fbn->fbd;
+	int i, ret;
+
+	/* Scan from middle of list to bottom, filling bottom up.
+	 * Skip the first entry which is reserved for dev_addr and
+	 * leave the last entry to use for promiscuous filtering.
+	 */
+	for (i = fbd->mac_addr_boundary, ret = -ENOENT;
+	     i < FBNIC_RPC_TCAM_MACDA_HOST_ADDR_IDX && ret; i++) {
+		struct fbnic_mac_addr *mac_addr = &fbd->mac_addr[i];
+
+		if (!ether_addr_equal(mac_addr->value.addr8, addr))
+			continue;
+
+		ret = __fbnic_uc_unsync(mac_addr);
+	}
+
+	return ret;
+}
+
+static int fbnic_mc_sync(struct net_device *netdev, const unsigned char *addr)
+{
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	struct fbnic_mac_addr *avail_addr;
+
+	if (WARN_ON(!is_multicast_ether_addr(addr)))
+		return -EADDRNOTAVAIL;
+
+	avail_addr = __fbnic_mc_sync(fbn->fbd, addr);
+	if (!avail_addr)
+		return -ENOSPC;
+
+	/* Add type flag indicating this address is in use by the host */
+	set_bit(FBNIC_MAC_ADDR_T_MULTICAST, avail_addr->act_tcam);
+
+	return 0;
+}
+
+static int fbnic_mc_unsync(struct net_device *netdev, const unsigned char *addr)
+{
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	struct fbnic_dev *fbd = fbn->fbd;
+	int i, ret;
+
+	/* Scan from middle of list to top, filling top down.
+	 * Skip over the address reserved for the BMC MAC and
+	 * exclude index 0 as that belongs to the broadcast address
+	 */
+	for (i = fbd->mac_addr_boundary, ret = -ENOENT;
+	     --i > FBNIC_RPC_TCAM_MACDA_BROADCAST_IDX && ret;) {
+		struct fbnic_mac_addr *mac_addr = &fbd->mac_addr[i];
+
+		if (!ether_addr_equal(mac_addr->value.addr8, addr))
+			continue;
+
+		ret = __fbnic_mc_unsync(mac_addr);
+	}
+
+	return ret;
+}
+
+void __fbnic_set_rx_mode(struct net_device *netdev)
+{
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	bool uc_promisc = false, mc_promisc = false;
+	struct fbnic_dev *fbd = fbn->fbd;
+	struct fbnic_mac_addr *mac_addr;
+	int err;
+
+	/* Populate host address from dev_addr */
+	mac_addr = &fbd->mac_addr[FBNIC_RPC_TCAM_MACDA_HOST_ADDR_IDX];
+	if (!ether_addr_equal(mac_addr->value.addr8, netdev->dev_addr) ||
+	    mac_addr->state != FBNIC_TCAM_S_VALID) {
+		ether_addr_copy(mac_addr->value.addr8, netdev->dev_addr);
+		mac_addr->state = FBNIC_TCAM_S_UPDATE;
+		set_bit(FBNIC_MAC_ADDR_T_UNICAST, mac_addr->act_tcam);
+	}
+
+	/* Populate broadcast address if broadcast is enabled */
+	mac_addr = &fbd->mac_addr[FBNIC_RPC_TCAM_MACDA_BROADCAST_IDX];
+	if (netdev->flags & IFF_BROADCAST) {
+		if (!is_broadcast_ether_addr(mac_addr->value.addr8) ||
+		    mac_addr->state != FBNIC_TCAM_S_VALID) {
+			eth_broadcast_addr(mac_addr->value.addr8);
+			mac_addr->state = FBNIC_TCAM_S_ADD;
+		}
+		set_bit(FBNIC_MAC_ADDR_T_BROADCAST, mac_addr->act_tcam);
+	} else if (mac_addr->state == FBNIC_TCAM_S_VALID) {
+		__fbnic_xc_unsync(mac_addr, FBNIC_MAC_ADDR_T_BROADCAST);
+	}
+
+	/* Synchronize unicast and multicast address lists */
+	err = __dev_uc_sync(netdev, fbnic_uc_sync, fbnic_uc_unsync);
+	if (err == -ENOSPC)
+		uc_promisc = true;
+	err = __dev_mc_sync(netdev, fbnic_mc_sync, fbnic_mc_unsync);
+	if (err == -ENOSPC)
+		mc_promisc = true;
+
+	uc_promisc |= !!(netdev->flags & IFF_PROMISC);
+	mc_promisc |= !!(netdev->flags & IFF_ALLMULTI) || uc_promisc;
+
+	/* Populate last TCAM entry with promiscuous entry and 0/1 bit mask */
+	mac_addr = &fbd->mac_addr[FBNIC_RPC_TCAM_MACDA_PROMISC_IDX];
+	if (uc_promisc) {
+		if (!is_zero_ether_addr(mac_addr->value.addr8) ||
+		    mac_addr->state != FBNIC_TCAM_S_VALID) {
+			eth_zero_addr(mac_addr->value.addr8);
+			eth_broadcast_addr(mac_addr->mask.addr8);
+			clear_bit(FBNIC_MAC_ADDR_T_ALLMULTI,
+				  mac_addr->act_tcam);
+			set_bit(FBNIC_MAC_ADDR_T_PROMISC,
+				mac_addr->act_tcam);
+			mac_addr->state = FBNIC_TCAM_S_ADD;
+		}
+	} else if (mc_promisc &&
+		   (!fbnic_bmc_present(fbd) || !fbd->fw_cap.all_multi)) {
+		/* We have to add a special handler for multicast as the
+		 * BMC may have an all-multi rule already in place. As such
+		 * adding a rule ourselves won't do any good so we will have
+		 * to modify the rules for the ALL MULTI below if the BMC
+		 * already has the rule in place.
+		 */
+		if (!is_multicast_ether_addr(mac_addr->value.addr8) ||
+		    mac_addr->state != FBNIC_TCAM_S_VALID) {
+			eth_zero_addr(mac_addr->value.addr8);
+			eth_broadcast_addr(mac_addr->mask.addr8);
+			mac_addr->value.addr8[0] ^= 1;
+			mac_addr->mask.addr8[0] ^= 1;
+			set_bit(FBNIC_MAC_ADDR_T_ALLMULTI,
+				mac_addr->act_tcam);
+			clear_bit(FBNIC_MAC_ADDR_T_PROMISC,
+				  mac_addr->act_tcam);
+			mac_addr->state = FBNIC_TCAM_S_ADD;
+		}
+	} else if (mac_addr->state == FBNIC_TCAM_S_VALID) {
+		if (test_bit(FBNIC_MAC_ADDR_T_BMC, mac_addr->act_tcam)) {
+			clear_bit(FBNIC_MAC_ADDR_T_ALLMULTI,
+				  mac_addr->act_tcam);
+			clear_bit(FBNIC_MAC_ADDR_T_PROMISC,
+				  mac_addr->act_tcam);
+		} else {
+			mac_addr->state = FBNIC_TCAM_S_DELETE;
+		}
+	}
+
+	/* Add rules for BMC all multicast if it is enabled */
+	fbnic_bmc_rpc_all_multi_config(fbd, mc_promisc);
+
+	/* Sift out any unshared BMC rules and place them in BMC only section */
+	fbnic_sift_macda(fbd);
+
+	/* Write updates to hardware */
+	fbnic_write_macda(fbd);
+}
+
+static void fbnic_set_rx_mode(struct net_device *netdev)
+{
+	/* No need to update the hardware if we are not running */
+	if (netif_running(netdev))
+		__fbnic_set_rx_mode(netdev);
+}
+
+static int fbnic_set_mac(struct net_device *netdev, void *p)
+{
+	struct sockaddr *addr = p;
+
+	if (!is_valid_ether_addr(addr->sa_data))
+		return -EADDRNOTAVAIL;
+
+	eth_hw_addr_set(netdev, addr->sa_data);
+
+	fbnic_set_rx_mode(netdev);
+
+	return 0;
+}
+
+void fbnic_clear_rx_mode(struct net_device *netdev)
+{
+	struct fbnic_net *fbn = netdev_priv(netdev);
+	struct fbnic_dev *fbd = fbn->fbd;
+	int idx;
+
+	for (idx = ARRAY_SIZE(fbd->mac_addr); idx--;) {
+		struct fbnic_mac_addr *mac_addr = &fbd->mac_addr[idx];
+
+		if (mac_addr->state != FBNIC_TCAM_S_VALID)
+			continue;
+
+		bitmap_clear(mac_addr->act_tcam,
+			     FBNIC_MAC_ADDR_T_HOST_START,
+			     FBNIC_MAC_ADDR_T_HOST_LEN);
+
+		if (bitmap_empty(mac_addr->act_tcam,
+				 FBNIC_RPC_TCAM_ACT_NUM_ENTRIES))
+			mac_addr->state = FBNIC_TCAM_S_DELETE;
+	}
+
+	/* Write updates to hardware */
+	fbnic_write_macda(fbd);
+
+	__dev_uc_unsync(netdev, NULL);
+	__dev_mc_unsync(netdev, NULL);
+}
+
 static const struct net_device_ops fbnic_netdev_ops = {
 	.ndo_open		= fbnic_open,
 	.ndo_stop		= fbnic_stop,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_start_xmit		= fbnic_xmit_frame,
 	.ndo_features_check	= fbnic_features_check,
+	.ndo_set_mac_address	= fbnic_set_mac,
+	.ndo_set_rx_mode	= fbnic_set_rx_mode,
 };
 
 void fbnic_reset_queues(struct fbnic_net *fbn,
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
index faf75e6359ba..80ecf7a38c50 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
@@ -49,5 +49,8 @@ void fbnic_netdev_unregister(struct net_device *netdev);
 void fbnic_reset_queues(struct fbnic_net *fbn,
 			unsigned int tx, unsigned int rx);
 
+void __fbnic_set_rx_mode(struct net_device *netdev);
+void fbnic_clear_rx_mode(struct net_device *netdev);
+
 int fbnic_phylink_init(struct net_device *netdev);
 #endif /* _FBNIC_NETDEV_H_ */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
index 6f6c2b1d9c99..52cb3f9ddfea 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -133,6 +133,8 @@ void fbnic_up(struct fbnic_net *fbn)
 
 	fbnic_fill(fbn);
 
+	__fbnic_set_rx_mode(fbn->netdev);
+
 	/* Enable Tx/Rx processing */
 	fbnic_napi_enable(fbn);
 	netif_tx_start_all_queues(fbn->netdev);
@@ -148,6 +150,7 @@ static void fbnic_down_noidle(struct fbnic_net *fbn)
 	fbnic_napi_disable(fbn);
 	netif_tx_disable(fbn->netdev);
 
+	fbnic_clear_rx_mode(fbn->netdev);
 	fbnic_disable(fbn);
 }
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_rpc.c b/drivers/net/ethernet/meta/fbnic/fbnic_rpc.c
new file mode 100644
index 000000000000..2806d8b4a4f8
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_rpc.c
@@ -0,0 +1,281 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#include <linux/etherdevice.h>
+
+#include "fbnic.h"
+#include "fbnic_netdev.h"
+#include "fbnic_rpc.h"
+
+void fbnic_bmc_rpc_all_multi_config(struct fbnic_dev *fbd,
+				    bool enable_host)
+{
+	struct fbnic_mac_addr *mac_addr;
+
+	/* We need to add the all multicast filter at the end of the
+	 * multicast address list. This way if there are any that are
+	 * shared between the host and the BMC they can be directed to
+	 * both. Otherwise the remainder just get sent directly to the
+	 * BMC.
+	 */
+	mac_addr = &fbd->mac_addr[fbd->mac_addr_boundary - 1];
+	if (fbnic_bmc_present(fbd) && fbd->fw_cap.all_multi) {
+		if (mac_addr->state != FBNIC_TCAM_S_VALID) {
+			eth_zero_addr(mac_addr->value.addr8);
+			eth_broadcast_addr(mac_addr->mask.addr8);
+			mac_addr->value.addr8[0] ^= 1;
+			mac_addr->mask.addr8[0] ^= 1;
+			set_bit(FBNIC_MAC_ADDR_T_BMC, mac_addr->act_tcam);
+			mac_addr->state = FBNIC_TCAM_S_ADD;
+		}
+		if (enable_host)
+			set_bit(FBNIC_MAC_ADDR_T_ALLMULTI,
+				mac_addr->act_tcam);
+		else
+			clear_bit(FBNIC_MAC_ADDR_T_ALLMULTI,
+				  mac_addr->act_tcam);
+	} else if (!test_bit(FBNIC_MAC_ADDR_T_BMC, mac_addr->act_tcam) &&
+		   !is_zero_ether_addr(mac_addr->mask.addr8) &&
+		   mac_addr->state == FBNIC_TCAM_S_VALID) {
+		clear_bit(FBNIC_MAC_ADDR_T_ALLMULTI, mac_addr->act_tcam);
+		clear_bit(FBNIC_MAC_ADDR_T_BMC, mac_addr->act_tcam);
+		mac_addr->state = FBNIC_TCAM_S_DELETE;
+	}
+}
+
+void fbnic_bmc_rpc_init(struct fbnic_dev *fbd)
+{
+	int i = FBNIC_RPC_TCAM_MACDA_BMC_ADDR_IDX;
+	struct fbnic_mac_addr *mac_addr;
+	int j;
+
+	/* Check if BMC is present */
+	if (!fbnic_bmc_present(fbd))
+		return;
+
+	/* Fetch BMC MAC addresses from firmware capabilities */
+	for (j = 0; j < 4; j++) {
+		u8 *bmc_mac = fbd->fw_cap.bmc_mac_addr[j];
+
+		/* Validate BMC MAC addresses */
+		if (is_zero_ether_addr(bmc_mac))
+			continue;
+
+		if (is_multicast_ether_addr(bmc_mac))
+			mac_addr = __fbnic_mc_sync(fbd, bmc_mac);
+		else
+			mac_addr = &fbd->mac_addr[i++];
+
+		if (!mac_addr) {
+			netdev_err(fbd->netdev,
+				   "No slot for BMC MAC address[%d]\n", j);
+			continue;
+		}
+
+		ether_addr_copy(mac_addr->value.addr8, bmc_mac);
+		eth_zero_addr(mac_addr->mask.addr8);
+
+		set_bit(FBNIC_MAC_ADDR_T_BMC, mac_addr->act_tcam);
+		mac_addr->state = FBNIC_TCAM_S_ADD;
+	}
+
+	/* Validate Broadcast is also present, record it and tag it */
+	mac_addr = &fbd->mac_addr[FBNIC_RPC_TCAM_MACDA_BROADCAST_IDX];
+	eth_broadcast_addr(mac_addr->value.addr8);
+	set_bit(FBNIC_MAC_ADDR_T_BMC, mac_addr->act_tcam);
+	mac_addr->state = FBNIC_TCAM_S_ADD;
+
+	fbnic_bmc_rpc_all_multi_config(fbd, false);
+}
+
+struct fbnic_mac_addr *__fbnic_uc_sync(struct fbnic_dev *fbd,
+				       const unsigned char *addr)
+{
+	struct fbnic_mac_addr *avail_addr = NULL;
+	unsigned int i;
+
+	/* Scan from middle of list to bottom, filling bottom up.
+	 * Skip the first entry which is reserved for dev_addr and
+	 * leave the last entry to use for promiscuous filtering.
+	 */
+	for (i = fbd->mac_addr_boundary - 1;
+	     i < FBNIC_RPC_TCAM_MACDA_HOST_ADDR_IDX; i++) {
+		struct fbnic_mac_addr *mac_addr = &fbd->mac_addr[i];
+
+		if (mac_addr->state == FBNIC_TCAM_S_DISABLED) {
+			avail_addr = mac_addr;
+		} else if (ether_addr_equal(mac_addr->value.addr8, addr)) {
+			avail_addr = mac_addr;
+			break;
+		}
+	}
+
+	if (avail_addr && avail_addr->state == FBNIC_TCAM_S_DISABLED) {
+		ether_addr_copy(avail_addr->value.addr8, addr);
+		eth_zero_addr(avail_addr->mask.addr8);
+		avail_addr->state = FBNIC_TCAM_S_ADD;
+	}
+
+	return avail_addr;
+}
+
+struct fbnic_mac_addr *__fbnic_mc_sync(struct fbnic_dev *fbd,
+				       const unsigned char *addr)
+{
+	struct fbnic_mac_addr *avail_addr = NULL;
+	unsigned int i;
+
+	/* Scan from middle of list to top, filling top down.
+	 * Skip over the address reserved for the BMC MAC and
+	 * exclude index 0 as that belongs to the broadcast address
+	 */
+	for (i = fbd->mac_addr_boundary;
+	     --i > FBNIC_RPC_TCAM_MACDA_BROADCAST_IDX;) {
+		struct fbnic_mac_addr *mac_addr = &fbd->mac_addr[i];
+
+		if (mac_addr->state == FBNIC_TCAM_S_DISABLED) {
+			avail_addr = mac_addr;
+		} else if (ether_addr_equal(mac_addr->value.addr8, addr)) {
+			avail_addr = mac_addr;
+			break;
+		}
+	}
+
+	/* Scan the BMC addresses to see if it may have already
+	 * reserved the address.
+	 */
+	while (--i) {
+		struct fbnic_mac_addr *mac_addr = &fbd->mac_addr[i];
+
+		if (!is_zero_ether_addr(mac_addr->mask.addr8))
+			continue;
+
+		/* Only move on if we find a match */
+		if (!ether_addr_equal(mac_addr->value.addr8, addr))
+			continue;
+
+		/* We need to pull this address to the shared area */
+		if (avail_addr) {
+			memcpy(avail_addr, mac_addr, sizeof(*mac_addr));
+			mac_addr->state = FBNIC_TCAM_S_DELETE;
+			avail_addr->state = FBNIC_TCAM_S_ADD;
+		}
+
+		break;
+	}
+
+	if (avail_addr && avail_addr->state == FBNIC_TCAM_S_DISABLED) {
+		ether_addr_copy(avail_addr->value.addr8, addr);
+		eth_zero_addr(avail_addr->mask.addr8);
+		avail_addr->state = FBNIC_TCAM_S_ADD;
+	}
+
+	return avail_addr;
+}
+
+int __fbnic_xc_unsync(struct fbnic_mac_addr *mac_addr, unsigned int tcam_idx)
+{
+	if (!test_and_clear_bit(tcam_idx, mac_addr->act_tcam))
+		return -ENOENT;
+
+	if (bitmap_empty(mac_addr->act_tcam, FBNIC_RPC_TCAM_ACT_NUM_ENTRIES))
+		mac_addr->state = FBNIC_TCAM_S_DELETE;
+
+	return 0;
+}
+
+void fbnic_sift_macda(struct fbnic_dev *fbd)
+{
+	int dest, src;
+
+	/* Move BMC only addresses back into BMC region */
+	for (dest = FBNIC_RPC_TCAM_MACDA_BMC_ADDR_IDX,
+	     src = FBNIC_RPC_TCAM_MACDA_MULTICAST_IDX;
+	     ++dest < FBNIC_RPC_TCAM_MACDA_BROADCAST_IDX &&
+	     src < fbd->mac_addr_boundary;) {
+		struct fbnic_mac_addr *dest_addr = &fbd->mac_addr[dest];
+
+		if (dest_addr->state != FBNIC_TCAM_S_DISABLED)
+			continue;
+
+		while (src < fbd->mac_addr_boundary) {
+			struct fbnic_mac_addr *src_addr = &fbd->mac_addr[src++];
+
+			/* Verify BMC bit is set */
+			if (!test_bit(FBNIC_MAC_ADDR_T_BMC, src_addr->act_tcam))
+				continue;
+
+			/* Verify filter isn't already disabled */
+			if (src_addr->state == FBNIC_TCAM_S_DISABLED ||
+			    src_addr->state == FBNIC_TCAM_S_DELETE)
+				continue;
+
+			/* Verify only BMC bit is set */
+			if (bitmap_weight(src_addr->act_tcam,
+					  FBNIC_RPC_TCAM_ACT_NUM_ENTRIES) != 1)
+				continue;
+
+			/* Verify we are not moving wildcard address */
+			if (!is_zero_ether_addr(src_addr->mask.addr8))
+				continue;
+
+			memcpy(dest_addr, src_addr, sizeof(*src_addr));
+			src_addr->state = FBNIC_TCAM_S_DELETE;
+			dest_addr->state = FBNIC_TCAM_S_ADD;
+		}
+	}
+}
+
+static void fbnic_clear_macda_entry(struct fbnic_dev *fbd, unsigned int idx)
+{
+	int i;
+
+	/* Invalidate entry and clear addr state info */
+	for (i = 0; i <= FBNIC_RPC_TCAM_MACDA_WORD_LEN; i++)
+		wr32(fbd, FBNIC_RPC_TCAM_MACDA(idx, i), 0);
+}
+
+static void fbnic_write_macda_entry(struct fbnic_dev *fbd, unsigned int idx,
+				    struct fbnic_mac_addr *mac_addr)
+{
+	__be16 *mask, *value;
+	int i;
+
+	mask = &mac_addr->mask.addr16[FBNIC_RPC_TCAM_MACDA_WORD_LEN - 1];
+	value = &mac_addr->value.addr16[FBNIC_RPC_TCAM_MACDA_WORD_LEN - 1];
+
+	for (i = 0; i < FBNIC_RPC_TCAM_MACDA_WORD_LEN; i++)
+		wr32(fbd, FBNIC_RPC_TCAM_MACDA(idx, i),
+		     FIELD_PREP(FBNIC_RPC_TCAM_MACDA_MASK, ntohs(*mask--)) |
+		     FIELD_PREP(FBNIC_RPC_TCAM_MACDA_VALUE, ntohs(*value--)));
+
+	wrfl(fbd);
+
+	wr32(fbd, FBNIC_RPC_TCAM_MACDA(idx, i), FBNIC_RPC_TCAM_VALIDATE);
+}
+
+void fbnic_write_macda(struct fbnic_dev *fbd)
+{
+	int idx;
+
+	for (idx = ARRAY_SIZE(fbd->mac_addr); idx--;) {
+		struct fbnic_mac_addr *mac_addr = &fbd->mac_addr[idx];
+
+		/* Check if update flag is set else exit. */
+		if (!(mac_addr->state & FBNIC_TCAM_S_UPDATE))
+			continue;
+
+		/* Clear by writing 0s. */
+		if (mac_addr->state == FBNIC_TCAM_S_DELETE) {
+			/* Invalidate entry and clear addr state info */
+			fbnic_clear_macda_entry(fbd, idx);
+			memset(mac_addr, 0, sizeof(*mac_addr));
+
+			continue;
+		}
+
+		fbnic_write_macda_entry(fbd, idx, mac_addr);
+
+		mac_addr->state = FBNIC_TCAM_S_VALID;
+	}
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_rpc.h b/drivers/net/ethernet/meta/fbnic/fbnic_rpc.h
new file mode 100644
index 000000000000..1b59b10ba677
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_rpc.h
@@ -0,0 +1,139 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#ifndef _FBNIC_RPC_H_
+#define _FBNIC_RPC_H_
+
+#include <uapi/linux/in6.h>
+#include <linux/bitfield.h>
+
+/*  The TCAM state definitions follow an expected ordering.
+ *  They start out disabled, then move through the following states:
+ *  Disabled  0	-> Add	      2
+ *  Add	      2	-> Valid      1
+ *
+ *  Valid     1	-> Add/Update 2
+ *  Add	      2	-> Valid      1
+ *
+ *  Valid     1	-> Delete     3
+ *  Delete    3	-> Disabled   0
+ */
+enum {
+	FBNIC_TCAM_S_DISABLED	= 0,
+	FBNIC_TCAM_S_VALID	= 1,
+	FBNIC_TCAM_S_ADD	= 2,
+	FBNIC_TCAM_S_UPDATE	= FBNIC_TCAM_S_ADD,
+	FBNIC_TCAM_S_DELETE	= 3,
+};
+
+/* 32 MAC Destination Address TCAM Entries
+ * 4 registers DA[1:0], DA[3:2], DA[5:4], Validate
+ */
+#define FBNIC_RPC_TCAM_MACDA_WORD_LEN		3
+#define FBNIC_RPC_TCAM_MACDA_NUM_ENTRIES	32
+
+#define FBNIC_RPC_TCAM_ACT_WORD_LEN		11
+#define FBNIC_RPC_TCAM_ACT_NUM_ENTRIES		64
+
+struct fbnic_mac_addr {
+	union {
+		unsigned char addr8[ETH_ALEN];
+		__be16 addr16[FBNIC_RPC_TCAM_MACDA_WORD_LEN];
+	} mask, value;
+	unsigned char state;
+	DECLARE_BITMAP(act_tcam, FBNIC_RPC_TCAM_ACT_NUM_ENTRIES);
+};
+
+struct fbnic_act_tcam {
+	struct {
+		u16 tcam[FBNIC_RPC_TCAM_ACT_WORD_LEN];
+	} mask, value;
+	unsigned char state;
+	u16 rss_en_mask;
+	u32 dest;
+};
+
+enum {
+	FBNIC_RSS_EN_HOST_ETHER,
+	FBNIC_RSS_EN_XCAST_ETHER,
+#define FBNIC_RSS_EN_NUM_UNICAST FBNIC_RSS_EN_XCAST_ETHER
+	FBNIC_RSS_EN_NUM_ENTRIES
+};
+
+/* Reserve the first 2 entries for the use by the BMC so that we can
+ * avoid allowing rules to get in the way of BMC unicast traffic.
+ */
+#define FBNIC_RPC_ACT_TBL_BMC_OFFSET		0
+#define FBNIC_RPC_ACT_TBL_BMC_ALL_MULTI_OFFSET	1
+
+/* We reserve the last 14 entries for RSS rules on the host. The BMC
+ * unicast rule will need to be populated above these and is expected to
+ * use MACDA TCAM entry 23 to store the BMC MAC address.
+ */
+#define FBNIC_RPC_ACT_TBL_RSS_OFFSET \
+	(FBNIC_RPC_ACT_TBL_NUM_ENTRIES - FBNIC_RSS_EN_NUM_ENTRIES)
+
+/* Flags used to identify the owner for this MAC filter. Note that any
+ * flags set for Broadcast thru Promisc indicate that the rule belongs
+ * to the RSS filters for the host.
+ */
+enum {
+	FBNIC_MAC_ADDR_T_BMC            = 0,
+	FBNIC_MAC_ADDR_T_BROADCAST	= FBNIC_RPC_ACT_TBL_RSS_OFFSET,
+#define FBNIC_MAC_ADDR_T_HOST_START	FBNIC_MAC_ADDR_T_BROADCAST
+	FBNIC_MAC_ADDR_T_MULTICAST,
+	FBNIC_MAC_ADDR_T_UNICAST,
+	FBNIC_MAC_ADDR_T_ALLMULTI,	/* BROADCAST ... MULTICAST*/
+	FBNIC_MAC_ADDR_T_PROMISC,	/* BROADCAST ... UNICAST */
+	FBNIC_MAC_ADDR_T_HOST_LAST
+};
+
+#define FBNIC_MAC_ADDR_T_HOST_LEN \
+	(FBNIC_MAC_ADDR_T_HOST_LAST - FBNIC_MAC_ADDR_T_HOST_START)
+
+#define FBNIC_RPC_TCAM_ACT1_L2_MACDA_IDX	CSR_GENMASK(9, 5)
+#define FBNIC_RPC_TCAM_ACT1_L2_MACDA_VALID	CSR_BIT(10)
+
+/* TCAM 0 - 3 reserved for BMC MAC addresses */
+#define FBNIC_RPC_TCAM_MACDA_BMC_ADDR_IDX	0
+/* TCAM 4 reserved for broadcast MAC address */
+#define FBNIC_RPC_TCAM_MACDA_BROADCAST_IDX	4
+/* TCAMs 5 - 30 will be used for multicast and unicast addresses. The
+ * boundary between the two can be variable it is currently set to 24
+ * on which the unicast addresses start. The general idea is that we will
+ * always go top-down with unicast, and bottom-up with multicast so that
+ * there should be free-space in the middle between the two.
+ *
+ * The entry at MADCA_DEFAULT_BOUNDARY is a special case as it can be used
+ * for the ALL MULTI address if the list is full, or the BMC has requested
+ * it.
+ */
+#define FBNIC_RPC_TCAM_MACDA_MULTICAST_IDX	5
+#define FBNIC_RPC_TCAM_MACDA_DEFAULT_BOUNDARY	24
+#define FBNIC_RPC_TCAM_MACDA_HOST_ADDR_IDX	30
+/* Reserved for use to record Multicast promisc, or Promiscuous */
+#define FBNIC_RPC_TCAM_MACDA_PROMISC_IDX	31
+
+struct fbnic_dev;
+
+void fbnic_bmc_rpc_init(struct fbnic_dev *fbd);
+void fbnic_bmc_rpc_all_multi_config(struct fbnic_dev *fbd, bool enable_host);
+
+int __fbnic_xc_unsync(struct fbnic_mac_addr *mac_addr, unsigned int tcam_idx);
+struct fbnic_mac_addr *__fbnic_uc_sync(struct fbnic_dev *fbd,
+				       const unsigned char *addr);
+struct fbnic_mac_addr *__fbnic_mc_sync(struct fbnic_dev *fbd,
+				       const unsigned char *addr);
+void fbnic_sift_macda(struct fbnic_dev *fbd);
+void fbnic_write_macda(struct fbnic_dev *fbd);
+
+static inline int __fbnic_uc_unsync(struct fbnic_mac_addr *mac_addr)
+{
+	return __fbnic_xc_unsync(mac_addr, FBNIC_MAC_ADDR_T_UNICAST);
+}
+
+static inline int __fbnic_mc_unsync(struct fbnic_mac_addr *mac_addr)
+{
+	return __fbnic_xc_unsync(mac_addr, FBNIC_MAC_ADDR_T_MULTICAST);
+}
+#endif /* _FBNIC_RPC_H_ */
-- 
cgit v1.2.3-58-ga151


From 355440a6981ab8166a8291678424bb38d3c1f5f9 Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexanderduyck@fb.com>
Date: Fri, 12 Jul 2024 08:50:35 -0700
Subject: eth: fbnic: Write the TCAM tables used for RSS control and Rx to host

RSS is controlled by the Rx filter tables. Program rules matching
on appropriate traffic types and set hashing fields using actions.
We need a separate set of rules for broadcast and multicast
because the action there needs to include forwarding to BMC.

This patch only initializes the default settings, the control
of the configuration using ethtool will come soon.

With this the necessary rules are put in place to enable Rx of packets by
the host.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079943591.1778861.17778587068185893750.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/fbnic.h        |   1 +
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h    |  59 ++++
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.c |   6 +
 drivers/net/ethernet/meta/fbnic/fbnic_netdev.h |   7 +
 drivers/net/ethernet/meta/fbnic/fbnic_pci.c    |   4 +
 drivers/net/ethernet/meta/fbnic/fbnic_rpc.c    | 370 +++++++++++++++++++++++++
 drivers/net/ethernet/meta/fbnic/fbnic_rpc.h    |  52 +++-
 7 files changed, 498 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/meta/fbnic/fbnic.h b/drivers/net/ethernet/meta/fbnic/fbnic.h
index 0b8530117a82..ad2689bfd6cb 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic.h
@@ -41,6 +41,7 @@ struct fbnic_dev {
 	u32 readrq;
 
 	/* Local copy of the devices TCAM */
+	struct fbnic_act_tcam act_tcam[FBNIC_RPC_TCAM_ACT_NUM_ENTRIES];
 	struct fbnic_mac_addr mac_addr[FBNIC_RPC_TCAM_MACDA_NUM_ENTRIES];
 	u8 mac_addr_boundary;
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index 50259c60bf65..a64360de0552 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -537,20 +537,79 @@ enum {
 #define FBNIC_RPC_RMI_CONFIG_FCS_PRESENT	CSR_BIT(8)
 #define FBNIC_RPC_RMI_CONFIG_ENABLE		CSR_BIT(12)
 #define FBNIC_RPC_RMI_CONFIG_MTU		CSR_GENMASK(31, 16)
+
+#define FBNIC_RPC_ACT_TBL0_DEFAULT	0x0840a		/* 0x21028 */
+#define FBNIC_RPC_ACT_TBL0_DROP			CSR_BIT(0)
+#define FBNIC_RPC_ACT_TBL0_DEST_MASK		CSR_GENMASK(3, 1)
+enum {
+	FBNIC_RPC_ACT_TBL0_DEST_HOST	= 1,
+	FBNIC_RPC_ACT_TBL0_DEST_BMC	= 2,
+	FBNIC_RPC_ACT_TBL0_DEST_EI	= 4,
+};
+
+#define FBNIC_RPC_ACT_TBL0_DMA_HINT		CSR_GENMASK(24, 16)
+#define FBNIC_RPC_ACT_TBL0_RSS_CTXT_ID		CSR_BIT(30)
+
+#define FBNIC_RPC_ACT_TBL1_DEFAULT	0x0840b		/* 0x2102c */
+#define FBNIC_RPC_ACT_TBL1_RSS_ENA_MASK		CSR_GENMASK(15, 0)
+enum {
+	FBNIC_RPC_ACT_TBL1_RSS_ENA_IP_SRC	= 1,
+	FBNIC_RPC_ACT_TBL1_RSS_ENA_IP_DST	= 2,
+	FBNIC_RPC_ACT_TBL1_RSS_ENA_L4_SRC	= 4,
+	FBNIC_RPC_ACT_TBL1_RSS_ENA_L4_DST	= 8,
+	FBNIC_RPC_ACT_TBL1_RSS_ENA_L2_DA	= 16,
+	FBNIC_RPC_ACT_TBL1_RSS_ENA_L4_RSS_BYTE	= 32,
+	FBNIC_RPC_ACT_TBL1_RSS_ENA_IV6_FL_LBL	= 64,
+	FBNIC_RPC_ACT_TBL1_RSS_ENA_OV6_FL_LBL	= 128,
+	FBNIC_RPC_ACT_TBL1_RSS_ENA_DSCP		= 256,
+	FBNIC_RPC_ACT_TBL1_RSS_ENA_L3_PROT	= 512,
+	FBNIC_RPC_ACT_TBL1_RSS_ENA_L4_PROT	= 1024,
+};
+
+#define FBNIC_RPC_RSS_KEY(n)		(0x0840c + (n))	/* 0x21030 + 4*n */
+#define FBNIC_RPC_RSS_KEY_BIT_LEN		425
+#define FBNIC_RPC_RSS_KEY_BYTE_LEN \
+	DIV_ROUND_UP(FBNIC_RPC_RSS_KEY_BIT_LEN, 8)
+#define FBNIC_RPC_RSS_KEY_DWORD_LEN \
+	DIV_ROUND_UP(FBNIC_RPC_RSS_KEY_BIT_LEN, 32)
+#define FBNIC_RPC_RSS_KEY_LAST_IDX \
+	(FBNIC_RPC_RSS_KEY_DWORD_LEN - 1)
+#define FBNIC_RPC_RSS_KEY_LAST_MASK \
+	CSR_GENMASK(31, \
+		    FBNIC_RPC_RSS_KEY_DWORD_LEN * 32 - \
+		    FBNIC_RPC_RSS_KEY_BIT_LEN)
+
 #define FBNIC_RPC_TCAM_MACDA_VALIDATE	0x0852d		/* 0x214b4 */
 #define FBNIC_CSR_END_RPC		0x0856b	/* CSR section delimiter */
 
 /* RPC RAM Registers */
 
 #define FBNIC_CSR_START_RPC_RAM		0x08800	/* CSR section delimiter */
+#define FBNIC_RPC_ACT_TBL0(n)		(0x08800 + (n))	/* 0x22000 + 4*n */
+#define FBNIC_RPC_ACT_TBL1(n)		(0x08840 + (n))	/* 0x22100 + 4*n */
 #define FBNIC_RPC_ACT_TBL_NUM_ENTRIES		64
 
 /* TCAM Tables */
 #define FBNIC_RPC_TCAM_VALIDATE			CSR_BIT(31)
+
+/* 64 Action TCAM Entries, 12 registers
+ * 3 mixed, src port, dst port, 6 L4 words, and Validate
+ */
+#define FBNIC_RPC_TCAM_ACT(m, n) \
+	(0x08880 + 0x40 * (n) + (m))		/* 0x22200 + 256*n + 4*m */
+
+#define FBNIC_RPC_TCAM_ACT_VALUE		CSR_GENMASK(15, 0)
+#define FBNIC_RPC_TCAM_ACT_MASK			CSR_GENMASK(31, 16)
+
 #define FBNIC_RPC_TCAM_MACDA(m, n) \
 	(0x08b80 + 0x20 * (n) + (m))		/* 0x022e00 + 128*n + 4*m */
 #define FBNIC_RPC_TCAM_MACDA_VALUE		CSR_GENMASK(15, 0)
 #define FBNIC_RPC_TCAM_MACDA_MASK		CSR_GENMASK(31, 16)
+
+#define FBNIC_RPC_RSS_TBL(n, m) \
+	(0x08d20 + 0x100 * (n) + (m))		/* 0x023480 + 1024*n + 4*m */
+#define FBNIC_RPC_RSS_TBL_COUNT			2
+#define FBNIC_RPC_RSS_TBL_SIZE			256
 #define FBNIC_CSR_END_RPC_RAM		0x08f1f	/* CSR section delimiter */
 
 /* Fab Registers */
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
index 371fc3ef8695..b7ce6da68543 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
@@ -50,6 +50,7 @@ int __fbnic_open(struct fbnic_net *fbn)
 		goto release_ownership;
 	/* Pull the BMC config and initialize the RPC */
 	fbnic_bmc_rpc_init(fbd);
+	fbnic_rss_reinit(fbd, fbn);
 
 	return 0;
 release_ownership:
@@ -262,6 +263,7 @@ void __fbnic_set_rx_mode(struct net_device *netdev)
 	fbnic_sift_macda(fbd);
 
 	/* Write updates to hardware */
+	fbnic_write_rules(fbd);
 	fbnic_write_macda(fbd);
 }
 
@@ -400,6 +402,10 @@ struct net_device *fbnic_netdev_alloc(struct fbnic_dev *fbd)
 
 	fbnic_reset_queues(fbn, default_queues, default_queues);
 
+	fbnic_reset_indir_tbl(fbn);
+	fbnic_rss_key_fill(fbn->rss_key);
+	fbnic_rss_init_en_mask(fbn);
+
 	netdev->features |=
 		NETIF_F_RXHASH |
 		NETIF_F_SG |
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
index 80ecf7a38c50..6bc0ebeb8182 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.h
@@ -7,6 +7,8 @@
 #include <linux/types.h>
 #include <linux/phylink.h>
 
+#include "fbnic_csr.h"
+#include "fbnic_rpc.h"
 #include "fbnic_txrx.h"
 
 struct fbnic_net {
@@ -34,7 +36,12 @@ struct fbnic_net {
 	u16 num_tx_queues;
 	u16 num_rx_queues;
 
+	u8 indir_tbl[FBNIC_RPC_RSS_TBL_COUNT][FBNIC_RPC_RSS_TBL_SIZE];
+	u32 rss_key[FBNIC_RPC_RSS_KEY_DWORD_LEN];
+	u32 rss_flow_hash[FBNIC_NUM_HASH_OPT];
+
 	u64 link_down_events;
+
 	struct list_head napis;
 };
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
index 52cb3f9ddfea..a4809fe0fc24 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_pci.c
@@ -133,6 +133,8 @@ void fbnic_up(struct fbnic_net *fbn)
 
 	fbnic_fill(fbn);
 
+	fbnic_rss_reinit_hw(fbn->fbd, fbn);
+
 	__fbnic_set_rx_mode(fbn->netdev);
 
 	/* Enable Tx/Rx processing */
@@ -151,6 +153,8 @@ static void fbnic_down_noidle(struct fbnic_net *fbn)
 	netif_tx_disable(fbn->netdev);
 
 	fbnic_clear_rx_mode(fbn->netdev);
+	fbnic_clear_rules(fbn->fbd);
+	fbnic_rss_disable_hw(fbn->fbd);
 	fbnic_disable(fbn);
 }
 
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_rpc.c b/drivers/net/ethernet/meta/fbnic/fbnic_rpc.c
index 2806d8b4a4f8..c8aa29fc052b 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_rpc.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_rpc.c
@@ -2,15 +2,108 @@
 /* Copyright (c) Meta Platforms, Inc. and affiliates. */
 
 #include <linux/etherdevice.h>
+#include <linux/ethtool.h>
 
 #include "fbnic.h"
 #include "fbnic_netdev.h"
 #include "fbnic_rpc.h"
 
+void fbnic_reset_indir_tbl(struct fbnic_net *fbn)
+{
+	unsigned int num_rx = fbn->num_rx_queues;
+	unsigned int i;
+
+	for (i = 0; i < FBNIC_RPC_RSS_TBL_SIZE; i++) {
+		fbn->indir_tbl[0][i] = ethtool_rxfh_indir_default(i, num_rx);
+		fbn->indir_tbl[1][i] = ethtool_rxfh_indir_default(i, num_rx);
+	}
+}
+
+void fbnic_rss_key_fill(u32 *buffer)
+{
+	static u32 rss_key[FBNIC_RPC_RSS_KEY_DWORD_LEN];
+
+	net_get_random_once(rss_key, sizeof(rss_key));
+	rss_key[FBNIC_RPC_RSS_KEY_LAST_IDX] &= FBNIC_RPC_RSS_KEY_LAST_MASK;
+
+	memcpy(buffer, rss_key, sizeof(rss_key));
+}
+
+#define RX_HASH_OPT_L4 \
+	(RXH_IP_SRC | RXH_IP_DST | RXH_L4_B_0_1 | RXH_L4_B_2_3)
+#define RX_HASH_OPT_L3 \
+	(RXH_IP_SRC | RXH_IP_DST)
+#define RX_HASH_OPT_L2 RXH_L2DA
+
+void fbnic_rss_init_en_mask(struct fbnic_net *fbn)
+{
+	fbn->rss_flow_hash[FBNIC_TCP4_HASH_OPT] = RX_HASH_OPT_L4;
+	fbn->rss_flow_hash[FBNIC_TCP6_HASH_OPT] = RX_HASH_OPT_L4;
+
+	fbn->rss_flow_hash[FBNIC_UDP4_HASH_OPT] = RX_HASH_OPT_L3;
+	fbn->rss_flow_hash[FBNIC_UDP6_HASH_OPT] = RX_HASH_OPT_L3;
+	fbn->rss_flow_hash[FBNIC_IPV4_HASH_OPT] = RX_HASH_OPT_L3;
+	fbn->rss_flow_hash[FBNIC_IPV6_HASH_OPT] = RX_HASH_OPT_L3;
+
+	fbn->rss_flow_hash[FBNIC_ETHER_HASH_OPT] = RX_HASH_OPT_L2;
+}
+
+void fbnic_rss_disable_hw(struct fbnic_dev *fbd)
+{
+	/* Disable RPC by clearing enable bit and configuration */
+	if (!fbnic_bmc_present(fbd))
+		wr32(fbd, FBNIC_RPC_RMI_CONFIG,
+		     FIELD_PREP(FBNIC_RPC_RMI_CONFIG_OH_BYTES, 20));
+}
+
+#define FBNIC_FH_2_RSSEM_BIT(_fh, _rssem, _val)		\
+	FIELD_PREP(FBNIC_RPC_ACT_TBL1_RSS_ENA_##_rssem,	\
+		   FIELD_GET(RXH_##_fh, _val))
+static u16 fbnic_flow_hash_2_rss_en_mask(struct fbnic_net *fbn, int flow_type)
+{
+	u32 flow_hash = fbn->rss_flow_hash[flow_type];
+	u32 rss_en_mask = 0;
+
+	rss_en_mask |= FBNIC_FH_2_RSSEM_BIT(L2DA, L2_DA, flow_hash);
+	rss_en_mask |= FBNIC_FH_2_RSSEM_BIT(IP_SRC, IP_SRC, flow_hash);
+	rss_en_mask |= FBNIC_FH_2_RSSEM_BIT(IP_DST, IP_DST, flow_hash);
+	rss_en_mask |= FBNIC_FH_2_RSSEM_BIT(L4_B_0_1, L4_SRC, flow_hash);
+	rss_en_mask |= FBNIC_FH_2_RSSEM_BIT(L4_B_2_3, L4_DST, flow_hash);
+
+	return rss_en_mask;
+}
+
+void fbnic_rss_reinit_hw(struct fbnic_dev *fbd, struct fbnic_net *fbn)
+{
+	unsigned int i;
+
+	for (i = 0; i < FBNIC_RPC_RSS_TBL_SIZE; i++) {
+		wr32(fbd, FBNIC_RPC_RSS_TBL(0, i), fbn->indir_tbl[0][i]);
+		wr32(fbd, FBNIC_RPC_RSS_TBL(1, i), fbn->indir_tbl[1][i]);
+	}
+
+	for (i = 0; i < FBNIC_RPC_RSS_KEY_DWORD_LEN; i++)
+		wr32(fbd, FBNIC_RPC_RSS_KEY(i), fbn->rss_key[i]);
+
+	/* Default action for this to drop w/ no destination */
+	wr32(fbd, FBNIC_RPC_ACT_TBL0_DEFAULT, FBNIC_RPC_ACT_TBL0_DROP);
+	wrfl(fbd);
+
+	wr32(fbd, FBNIC_RPC_ACT_TBL1_DEFAULT, 0);
+
+	/* If it isn't already enabled set the RMI Config value to enable RPC */
+	wr32(fbd, FBNIC_RPC_RMI_CONFIG,
+	     FIELD_PREP(FBNIC_RPC_RMI_CONFIG_MTU, FBNIC_MAX_JUMBO_FRAME_SIZE) |
+	     FIELD_PREP(FBNIC_RPC_RMI_CONFIG_OH_BYTES, 20) |
+	     FBNIC_RPC_RMI_CONFIG_ENABLE);
+}
+
 void fbnic_bmc_rpc_all_multi_config(struct fbnic_dev *fbd,
 				    bool enable_host)
 {
+	struct fbnic_act_tcam *act_tcam;
 	struct fbnic_mac_addr *mac_addr;
+	int j;
 
 	/* We need to add the all multicast filter at the end of the
 	 * multicast address list. This way if there are any that are
@@ -41,11 +134,48 @@ void fbnic_bmc_rpc_all_multi_config(struct fbnic_dev *fbd,
 		clear_bit(FBNIC_MAC_ADDR_T_BMC, mac_addr->act_tcam);
 		mac_addr->state = FBNIC_TCAM_S_DELETE;
 	}
+
+	/* We have to add a special handler for multicast as the
+	 * BMC may have an all-multi rule already in place. As such
+	 * adding a rule ourselves won't do any good so we will have
+	 * to modify the rules for the ALL MULTI below if the BMC
+	 * already has the rule in place.
+	 */
+	act_tcam = &fbd->act_tcam[FBNIC_RPC_ACT_TBL_BMC_ALL_MULTI_OFFSET];
+
+	/* If we are not enabling the rule just delete it. We will fall
+	 * back to the RSS rules that support the multicast addresses.
+	 */
+	if (!fbnic_bmc_present(fbd) || !fbd->fw_cap.all_multi || enable_host) {
+		if (act_tcam->state == FBNIC_TCAM_S_VALID)
+			act_tcam->state = FBNIC_TCAM_S_DELETE;
+		return;
+	}
+
+	/* Rewrite TCAM rule 23 to handle BMC all-multi traffic */
+	act_tcam->dest = FIELD_PREP(FBNIC_RPC_ACT_TBL0_DEST_MASK,
+				    FBNIC_RPC_ACT_TBL0_DEST_BMC);
+	act_tcam->mask.tcam[0] = 0xffff;
+
+	/* MACDA 0 - 3 is reserved for the BMC MAC address */
+	act_tcam->value.tcam[1] =
+			FIELD_PREP(FBNIC_RPC_TCAM_ACT1_L2_MACDA_IDX,
+				   fbd->mac_addr_boundary - 1) |
+			FBNIC_RPC_TCAM_ACT1_L2_MACDA_VALID;
+	act_tcam->mask.tcam[1] = 0xffff &
+			 ~FBNIC_RPC_TCAM_ACT1_L2_MACDA_IDX &
+			 ~FBNIC_RPC_TCAM_ACT1_L2_MACDA_VALID;
+
+	for (j = 2; j < FBNIC_RPC_TCAM_ACT_WORD_LEN; j++)
+		act_tcam->mask.tcam[j] = 0xffff;
+
+	act_tcam->state = FBNIC_TCAM_S_UPDATE;
 }
 
 void fbnic_bmc_rpc_init(struct fbnic_dev *fbd)
 {
 	int i = FBNIC_RPC_TCAM_MACDA_BMC_ADDR_IDX;
+	struct fbnic_act_tcam *act_tcam;
 	struct fbnic_mac_addr *mac_addr;
 	int j;
 
@@ -85,9 +215,114 @@ void fbnic_bmc_rpc_init(struct fbnic_dev *fbd)
 	set_bit(FBNIC_MAC_ADDR_T_BMC, mac_addr->act_tcam);
 	mac_addr->state = FBNIC_TCAM_S_ADD;
 
+	/* Rewrite TCAM rule 0 if it isn't present to relocate BMC rules */
+	act_tcam = &fbd->act_tcam[FBNIC_RPC_ACT_TBL_BMC_OFFSET];
+	act_tcam->dest = FIELD_PREP(FBNIC_RPC_ACT_TBL0_DEST_MASK,
+				    FBNIC_RPC_ACT_TBL0_DEST_BMC);
+	act_tcam->mask.tcam[0] = 0xffff;
+
+	/* MACDA 0 - 3 is reserved for the BMC MAC address
+	 * to account for that we have to mask out the lower 2 bits
+	 * of the macda by performing an &= with 0x1c.
+	 */
+	act_tcam->value.tcam[1] = FBNIC_RPC_TCAM_ACT1_L2_MACDA_VALID;
+	act_tcam->mask.tcam[1] = 0xffff &
+			~FIELD_PREP(FBNIC_RPC_TCAM_ACT1_L2_MACDA_IDX, 0x1c) &
+			~FBNIC_RPC_TCAM_ACT1_L2_MACDA_VALID;
+
+	for (j = 2; j < FBNIC_RPC_TCAM_ACT_WORD_LEN; j++)
+		act_tcam->mask.tcam[j] = 0xffff;
+
+	act_tcam->state = FBNIC_TCAM_S_UPDATE;
+
 	fbnic_bmc_rpc_all_multi_config(fbd, false);
 }
 
+#define FBNIC_ACT1_INIT(_l4, _udp, _ip, _v6)		\
+	(((_l4) ? FBNIC_RPC_TCAM_ACT1_L4_VALID : 0) |	\
+	 ((_udp) ? FBNIC_RPC_TCAM_ACT1_L4_IS_UDP : 0) |	\
+	 ((_ip) ? FBNIC_RPC_TCAM_ACT1_IP_VALID : 0) |	\
+	 ((_v6) ? FBNIC_RPC_TCAM_ACT1_IP_IS_V6 : 0))
+
+void fbnic_rss_reinit(struct fbnic_dev *fbd, struct fbnic_net *fbn)
+{
+	static const u32 act1_value[FBNIC_NUM_HASH_OPT] = {
+		FBNIC_ACT1_INIT(1, 1, 1, 1),	/* UDP6 */
+		FBNIC_ACT1_INIT(1, 1, 1, 0),	/* UDP4 */
+		FBNIC_ACT1_INIT(1, 0, 1, 1),	/* TCP6 */
+		FBNIC_ACT1_INIT(1, 0, 1, 0),	/* TCP4 */
+		FBNIC_ACT1_INIT(0, 0, 1, 1),	/* IP6 */
+		FBNIC_ACT1_INIT(0, 0, 1, 0),	/* IP4 */
+		0				/* Ether */
+	};
+	unsigned int i;
+
+	/* To support scenarios where a BMC is present we must write the
+	 * rules twice, once for the unicast cases, and once again for
+	 * the broadcast/multicast cases as we have to support 2 destinations.
+	 */
+	BUILD_BUG_ON(FBNIC_RSS_EN_NUM_UNICAST * 2 != FBNIC_RSS_EN_NUM_ENTRIES);
+	BUILD_BUG_ON(ARRAY_SIZE(act1_value) != FBNIC_NUM_HASH_OPT);
+
+	/* Program RSS hash enable mask for host in action TCAM/table. */
+	for (i = fbnic_bmc_present(fbd) ? 0 : FBNIC_RSS_EN_NUM_UNICAST;
+	     i < FBNIC_RSS_EN_NUM_ENTRIES; i++) {
+		unsigned int idx = i + FBNIC_RPC_ACT_TBL_RSS_OFFSET;
+		struct fbnic_act_tcam *act_tcam = &fbd->act_tcam[idx];
+		u32 flow_hash, dest, rss_en_mask;
+		int flow_type, j;
+		u16 value = 0;
+
+		flow_type = i % FBNIC_RSS_EN_NUM_UNICAST;
+		flow_hash = fbn->rss_flow_hash[flow_type];
+
+		/* Set DEST_HOST based on absence of RXH_DISCARD */
+		dest = FIELD_PREP(FBNIC_RPC_ACT_TBL0_DEST_MASK,
+				  !(RXH_DISCARD & flow_hash) ?
+				  FBNIC_RPC_ACT_TBL0_DEST_HOST : 0);
+
+		if (i >= FBNIC_RSS_EN_NUM_UNICAST && fbnic_bmc_present(fbd))
+			dest |= FIELD_PREP(FBNIC_RPC_ACT_TBL0_DEST_MASK,
+					   FBNIC_RPC_ACT_TBL0_DEST_BMC);
+
+		if (!dest)
+			dest = FBNIC_RPC_ACT_TBL0_DROP;
+
+		if (act1_value[flow_type] & FBNIC_RPC_TCAM_ACT1_L4_VALID)
+			dest |= FIELD_PREP(FBNIC_RPC_ACT_TBL0_DMA_HINT,
+					   FBNIC_RCD_HDR_AL_DMA_HINT_L4);
+
+		rss_en_mask = fbnic_flow_hash_2_rss_en_mask(fbn, flow_type);
+
+		act_tcam->dest = dest;
+		act_tcam->rss_en_mask = rss_en_mask;
+		act_tcam->state = FBNIC_TCAM_S_UPDATE;
+
+		act_tcam->mask.tcam[0] = 0xffff;
+
+		/* We reserve the upper 8 MACDA TCAM entries for host
+		 * unicast. So we set the value to 24, and the mask the
+		 * lower bits so that the lower entries can be used as
+		 * multicast or BMC addresses.
+		 */
+		if (i < FBNIC_RSS_EN_NUM_UNICAST)
+			value = FIELD_PREP(FBNIC_RPC_TCAM_ACT1_L2_MACDA_IDX,
+					   fbd->mac_addr_boundary);
+		value |= FBNIC_RPC_TCAM_ACT1_L2_MACDA_VALID;
+
+		flow_type = i % FBNIC_RSS_EN_NUM_UNICAST;
+		value |= act1_value[flow_type];
+
+		act_tcam->value.tcam[1] = value;
+		act_tcam->mask.tcam[1] = ~value;
+
+		for (j = 2; j < FBNIC_RPC_TCAM_ACT_WORD_LEN; j++)
+			act_tcam->mask.tcam[j] = 0xffff;
+
+		act_tcam->state = FBNIC_TCAM_S_UPDATE;
+	}
+}
+
 struct fbnic_mac_addr *__fbnic_uc_sync(struct fbnic_dev *fbd,
 				       const unsigned char *addr)
 {
@@ -235,6 +470,38 @@ static void fbnic_clear_macda_entry(struct fbnic_dev *fbd, unsigned int idx)
 		wr32(fbd, FBNIC_RPC_TCAM_MACDA(idx, i), 0);
 }
 
+static void fbnic_clear_macda(struct fbnic_dev *fbd)
+{
+	int idx;
+
+	for (idx = ARRAY_SIZE(fbd->mac_addr); idx--;) {
+		struct fbnic_mac_addr *mac_addr = &fbd->mac_addr[idx];
+
+		if (mac_addr->state == FBNIC_TCAM_S_DISABLED)
+			continue;
+
+		if (test_bit(FBNIC_MAC_ADDR_T_BMC, mac_addr->act_tcam)) {
+			if (fbnic_bmc_present(fbd))
+				continue;
+			dev_warn_once(fbd->dev,
+				      "Found BMC MAC address w/ BMC not present\n");
+		}
+
+		fbnic_clear_macda_entry(fbd, idx);
+
+		/* If rule was already destined for deletion just wipe it now */
+		if (mac_addr->state == FBNIC_TCAM_S_DELETE) {
+			memset(mac_addr, 0, sizeof(*mac_addr));
+			continue;
+		}
+
+		/* Change state to update so that we will rewrite
+		 * this tcam the next time fbnic_write_macda is called.
+		 */
+		mac_addr->state = FBNIC_TCAM_S_UPDATE;
+	}
+}
+
 static void fbnic_write_macda_entry(struct fbnic_dev *fbd, unsigned int idx,
 				    struct fbnic_mac_addr *mac_addr)
 {
@@ -279,3 +546,106 @@ void fbnic_write_macda(struct fbnic_dev *fbd)
 		mac_addr->state = FBNIC_TCAM_S_VALID;
 	}
 }
+
+static void fbnic_clear_act_tcam(struct fbnic_dev *fbd, unsigned int idx)
+{
+	int i;
+
+	/* Invalidate entry and clear addr state info */
+	for (i = 0; i <= FBNIC_RPC_TCAM_ACT_WORD_LEN; i++)
+		wr32(fbd, FBNIC_RPC_TCAM_ACT(idx, i), 0);
+}
+
+void fbnic_clear_rules(struct fbnic_dev *fbd)
+{
+	u32 dest = FIELD_PREP(FBNIC_RPC_ACT_TBL0_DEST_MASK,
+			      FBNIC_RPC_ACT_TBL0_DEST_BMC);
+	int i = FBNIC_RPC_TCAM_ACT_NUM_ENTRIES - 1;
+	struct fbnic_act_tcam *act_tcam;
+
+	/* Clear MAC rules */
+	fbnic_clear_macda(fbd);
+
+	/* If BMC is present we need to preserve the last rule which
+	 * will be used to route traffic to the BMC if it is received.
+	 *
+	 * At this point it should be the only MAC address in the MACDA
+	 * so any unicast or multicast traffic received should be routed
+	 * to it. So leave the last rule in place.
+	 *
+	 * It will be rewritten to add the host again when we bring
+	 * the interface back up.
+	 */
+	if (fbnic_bmc_present(fbd)) {
+		act_tcam = &fbd->act_tcam[i];
+
+		if (act_tcam->state == FBNIC_TCAM_S_VALID &&
+		    (act_tcam->dest & dest)) {
+			wr32(fbd, FBNIC_RPC_ACT_TBL0(i), dest);
+			wr32(fbd, FBNIC_RPC_ACT_TBL1(i), 0);
+
+			act_tcam->state = FBNIC_TCAM_S_UPDATE;
+
+			i--;
+		}
+	}
+
+	/* Work from the bottom up deleting all other rules from hardware */
+	do {
+		act_tcam = &fbd->act_tcam[i];
+
+		if (act_tcam->state != FBNIC_TCAM_S_VALID)
+			continue;
+
+		fbnic_clear_act_tcam(fbd, i);
+		act_tcam->state = FBNIC_TCAM_S_UPDATE;
+	} while (i--);
+}
+
+static void fbnic_delete_act_tcam(struct fbnic_dev *fbd, unsigned int idx)
+{
+	fbnic_clear_act_tcam(fbd, idx);
+	memset(&fbd->act_tcam[idx], 0, sizeof(struct fbnic_act_tcam));
+}
+
+static void fbnic_update_act_tcam(struct fbnic_dev *fbd, unsigned int idx)
+{
+	struct fbnic_act_tcam *act_tcam = &fbd->act_tcam[idx];
+	int i;
+
+	/* Update entry by writing the destination and RSS mask */
+	wr32(fbd, FBNIC_RPC_ACT_TBL0(idx), act_tcam->dest);
+	wr32(fbd, FBNIC_RPC_ACT_TBL1(idx), act_tcam->rss_en_mask);
+
+	/* Write new TCAM rule to hardware */
+	for (i = 0; i < FBNIC_RPC_TCAM_ACT_WORD_LEN; i++)
+		wr32(fbd, FBNIC_RPC_TCAM_ACT(idx, i),
+		     FIELD_PREP(FBNIC_RPC_TCAM_ACT_MASK,
+				act_tcam->mask.tcam[i]) |
+		     FIELD_PREP(FBNIC_RPC_TCAM_ACT_VALUE,
+				act_tcam->value.tcam[i]));
+
+	wrfl(fbd);
+
+	wr32(fbd, FBNIC_RPC_TCAM_ACT(idx, i), FBNIC_RPC_TCAM_VALIDATE);
+	act_tcam->state = FBNIC_TCAM_S_VALID;
+}
+
+void fbnic_write_rules(struct fbnic_dev *fbd)
+{
+	int i;
+
+	/* Flush any pending action table rules */
+	for (i = 0; i < FBNIC_RPC_ACT_TBL_NUM_ENTRIES; i++) {
+		struct fbnic_act_tcam *act_tcam = &fbd->act_tcam[i];
+
+		/* Check if update flag is set else exit. */
+		if (!(act_tcam->state & FBNIC_TCAM_S_UPDATE))
+			continue;
+
+		if (act_tcam->state == FBNIC_TCAM_S_DELETE)
+			fbnic_delete_act_tcam(fbd, i);
+		else
+			fbnic_update_act_tcam(fbd, i);
+	}
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_rpc.h b/drivers/net/ethernet/meta/fbnic/fbnic_rpc.h
index 1b59b10ba677..d62935f722a2 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_rpc.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_rpc.h
@@ -54,9 +54,21 @@ struct fbnic_act_tcam {
 };
 
 enum {
+	FBNIC_RSS_EN_HOST_UDP6,
+	FBNIC_RSS_EN_HOST_UDP4,
+	FBNIC_RSS_EN_HOST_TCP6,
+	FBNIC_RSS_EN_HOST_TCP4,
+	FBNIC_RSS_EN_HOST_IP6,
+	FBNIC_RSS_EN_HOST_IP4,
 	FBNIC_RSS_EN_HOST_ETHER,
+	FBNIC_RSS_EN_XCAST_UDP6,
+#define FBNIC_RSS_EN_NUM_UNICAST FBNIC_RSS_EN_XCAST_UDP6
+	FBNIC_RSS_EN_XCAST_UDP4,
+	FBNIC_RSS_EN_XCAST_TCP6,
+	FBNIC_RSS_EN_XCAST_TCP4,
+	FBNIC_RSS_EN_XCAST_IP6,
+	FBNIC_RSS_EN_XCAST_IP4,
 	FBNIC_RSS_EN_XCAST_ETHER,
-#define FBNIC_RSS_EN_NUM_UNICAST FBNIC_RSS_EN_XCAST_ETHER
 	FBNIC_RSS_EN_NUM_ENTRIES
 };
 
@@ -91,8 +103,22 @@ enum {
 #define FBNIC_MAC_ADDR_T_HOST_LEN \
 	(FBNIC_MAC_ADDR_T_HOST_LAST - FBNIC_MAC_ADDR_T_HOST_START)
 
+#define FBNIC_RPC_TCAM_ACT0_IPSRC_IDX		CSR_GENMASK(2, 0)
+#define FBNIC_RPC_TCAM_ACT0_IPSRC_VALID		CSR_BIT(3)
+#define FBNIC_RPC_TCAM_ACT0_IPDST_IDX		CSR_GENMASK(6, 4)
+#define FBNIC_RPC_TCAM_ACT0_IPDST_VALID		CSR_BIT(7)
+#define FBNIC_RPC_TCAM_ACT0_OUTER_IPSRC_IDX	CSR_GENMASK(10, 8)
+#define FBNIC_RPC_TCAM_ACT0_OUTER_IPSRC_VALID	CSR_BIT(11)
+#define FBNIC_RPC_TCAM_ACT0_OUTER_IPDST_IDX	CSR_GENMASK(14, 12)
+#define FBNIC_RPC_TCAM_ACT0_OUTER_IPDST_VALID	CSR_BIT(15)
+
 #define FBNIC_RPC_TCAM_ACT1_L2_MACDA_IDX	CSR_GENMASK(9, 5)
 #define FBNIC_RPC_TCAM_ACT1_L2_MACDA_VALID	CSR_BIT(10)
+#define FBNIC_RPC_TCAM_ACT1_IP_IS_V6		CSR_BIT(11)
+#define FBNIC_RPC_TCAM_ACT1_IP_VALID		CSR_BIT(12)
+#define FBNIC_RPC_TCAM_ACT1_OUTER_IP_VALID	CSR_BIT(13)
+#define FBNIC_RPC_TCAM_ACT1_L4_IS_UDP		CSR_BIT(14)
+#define FBNIC_RPC_TCAM_ACT1_L4_VALID		CSR_BIT(15)
 
 /* TCAM 0 - 3 reserved for BMC MAC addresses */
 #define FBNIC_RPC_TCAM_MACDA_BMC_ADDR_IDX	0
@@ -114,11 +140,32 @@ enum {
 /* Reserved for use to record Multicast promisc, or Promiscuous */
 #define FBNIC_RPC_TCAM_MACDA_PROMISC_IDX	31
 
+enum {
+	FBNIC_UDP6_HASH_OPT,
+	FBNIC_UDP4_HASH_OPT,
+	FBNIC_TCP6_HASH_OPT,
+	FBNIC_TCP4_HASH_OPT,
+#define FBNIC_L4_HASH_OPT FBNIC_TCP4_HASH_OPT
+	FBNIC_IPV6_HASH_OPT,
+	FBNIC_IPV4_HASH_OPT,
+#define FBNIC_IP_HASH_OPT FBNIC_IPV4_HASH_OPT
+	FBNIC_ETHER_HASH_OPT,
+	FBNIC_NUM_HASH_OPT,
+};
+
 struct fbnic_dev;
+struct fbnic_net;
 
 void fbnic_bmc_rpc_init(struct fbnic_dev *fbd);
 void fbnic_bmc_rpc_all_multi_config(struct fbnic_dev *fbd, bool enable_host);
 
+void fbnic_reset_indir_tbl(struct fbnic_net *fbn);
+void fbnic_rss_key_fill(u32 *buffer);
+void fbnic_rss_init_en_mask(struct fbnic_net *fbn);
+void fbnic_rss_disable_hw(struct fbnic_dev *fbd);
+void fbnic_rss_reinit_hw(struct fbnic_dev *fbd, struct fbnic_net *fbn);
+void fbnic_rss_reinit(struct fbnic_dev *fbd, struct fbnic_net *fbn);
+
 int __fbnic_xc_unsync(struct fbnic_mac_addr *mac_addr, unsigned int tcam_idx);
 struct fbnic_mac_addr *__fbnic_uc_sync(struct fbnic_dev *fbd,
 				       const unsigned char *addr);
@@ -136,4 +183,7 @@ static inline int __fbnic_mc_unsync(struct fbnic_mac_addr *mac_addr)
 {
 	return __fbnic_xc_unsync(mac_addr, FBNIC_MAC_ADDR_T_MULTICAST);
 }
+
+void fbnic_clear_rules(struct fbnic_dev *fbd);
+void fbnic_write_rules(struct fbnic_dev *fbd);
 #endif /* _FBNIC_RPC_H_ */
-- 
cgit v1.2.3-58-ga151


From 3f45181358e4df50a40ea1bb51b00a1f295f915e Mon Sep 17 00:00:00 2001
From: Kuniyuki Iwashima <kuniyu@amazon.com>
Date: Sun, 14 Jul 2024 09:17:19 -0700
Subject: tcp: Don't access uninit tcp_rsk(req)->ao_keyid in
 tcp_create_openreq_child().

syzkaller reported KMSAN splat in tcp_create_openreq_child(). [0]

The uninit variable is tcp_rsk(req)->ao_keyid.

tcp_rsk(req)->ao_keyid is initialised only when tcp_conn_request() finds
a valid TCP AO option in SYN.  Then, tcp_rsk(req)->used_tcp_ao is set
accordingly.

Let's not read tcp_rsk(req)->ao_keyid when tcp_rsk(req)->used_tcp_ao is
false.

[0]:
BUG: KMSAN: uninit-value in tcp_create_openreq_child+0x198b/0x1ff0 net/ipv4/tcp_minisocks.c:610
 tcp_create_openreq_child+0x198b/0x1ff0 net/ipv4/tcp_minisocks.c:610
 tcp_v4_syn_recv_sock+0x18e/0x2170 net/ipv4/tcp_ipv4.c:1754
 tcp_check_req+0x1a3e/0x20c0 net/ipv4/tcp_minisocks.c:852
 tcp_v4_rcv+0x26a4/0x53a0 net/ipv4/tcp_ipv4.c:2265
 ip_protocol_deliver_rcu+0x884/0x1270 net/ipv4/ip_input.c:205
 ip_local_deliver_finish+0x30f/0x530 net/ipv4/ip_input.c:233
 NF_HOOK include/linux/netfilter.h:314 [inline]
 ip_local_deliver+0x230/0x4c0 net/ipv4/ip_input.c:254
 dst_input include/net/dst.h:460 [inline]
 ip_sublist_rcv_finish net/ipv4/ip_input.c:580 [inline]
 ip_list_rcv_finish net/ipv4/ip_input.c:631 [inline]
 ip_sublist_rcv+0x10f7/0x13e0 net/ipv4/ip_input.c:639
 ip_list_rcv+0x952/0x9c0 net/ipv4/ip_input.c:674
 __netif_receive_skb_list_ptype net/core/dev.c:5703 [inline]
 __netif_receive_skb_list_core+0xd92/0x11d0 net/core/dev.c:5751
 __netif_receive_skb_list net/core/dev.c:5803 [inline]
 netif_receive_skb_list_internal+0xd8f/0x1350 net/core/dev.c:5895
 gro_normal_list include/net/gro.h:515 [inline]
 napi_complete_done+0x3f2/0x990 net/core/dev.c:6246
 e1000_clean+0x1fa4/0x5e50 drivers/net/ethernet/intel/e1000/e1000_main.c:3808
 __napi_poll+0xd9/0x990 net/core/dev.c:6771
 napi_poll net/core/dev.c:6840 [inline]
 net_rx_action+0x90f/0x17e0 net/core/dev.c:6962
 handle_softirqs+0x152/0x6b0 kernel/softirq.c:554
 __do_softirq kernel/softirq.c:588 [inline]
 invoke_softirq kernel/softirq.c:428 [inline]
 __irq_exit_rcu kernel/softirq.c:637 [inline]
 irq_exit_rcu+0x5d/0x120 kernel/softirq.c:649
 common_interrupt+0x83/0x90 arch/x86/kernel/irq.c:278
 asm_common_interrupt+0x26/0x40 arch/x86/include/asm/idtentry.h:693
 __msan_instrument_asm_store+0xd6/0xe0
 arch_atomic_inc arch/x86/include/asm/atomic.h:53 [inline]
 raw_atomic_inc include/linux/atomic/atomic-arch-fallback.h:992 [inline]
 atomic_inc include/linux/atomic/atomic-instrumented.h:436 [inline]
 page_ref_inc include/linux/page_ref.h:153 [inline]
 folio_ref_inc include/linux/page_ref.h:160 [inline]
 filemap_map_order0_folio mm/filemap.c:3596 [inline]
 filemap_map_pages+0x11c7/0x2270 mm/filemap.c:3644
 do_fault_around mm/memory.c:4879 [inline]
 do_read_fault mm/memory.c:4912 [inline]
 do_fault mm/memory.c:5051 [inline]
 do_pte_missing mm/memory.c:3897 [inline]
 handle_pte_fault mm/memory.c:5381 [inline]
 __handle_mm_fault mm/memory.c:5524 [inline]
 handle_mm_fault+0x3677/0x6f00 mm/memory.c:5689
 do_user_addr_fault+0x1373/0x2b20 arch/x86/mm/fault.c:1338
 handle_page_fault arch/x86/mm/fault.c:1481 [inline]
 exc_page_fault+0x54/0xc0 arch/x86/mm/fault.c:1539
 asm_exc_page_fault+0x26/0x30 arch/x86/include/asm/idtentry.h:623

Uninit was stored to memory at:
 tcp_create_openreq_child+0x1984/0x1ff0 net/ipv4/tcp_minisocks.c:611
 tcp_v4_syn_recv_sock+0x18e/0x2170 net/ipv4/tcp_ipv4.c:1754
 tcp_check_req+0x1a3e/0x20c0 net/ipv4/tcp_minisocks.c:852
 tcp_v4_rcv+0x26a4/0x53a0 net/ipv4/tcp_ipv4.c:2265
 ip_protocol_deliver_rcu+0x884/0x1270 net/ipv4/ip_input.c:205
 ip_local_deliver_finish+0x30f/0x530 net/ipv4/ip_input.c:233
 NF_HOOK include/linux/netfilter.h:314 [inline]
 ip_local_deliver+0x230/0x4c0 net/ipv4/ip_input.c:254
 dst_input include/net/dst.h:460 [inline]
 ip_sublist_rcv_finish net/ipv4/ip_input.c:580 [inline]
 ip_list_rcv_finish net/ipv4/ip_input.c:631 [inline]
 ip_sublist_rcv+0x10f7/0x13e0 net/ipv4/ip_input.c:639
 ip_list_rcv+0x952/0x9c0 net/ipv4/ip_input.c:674
 __netif_receive_skb_list_ptype net/core/dev.c:5703 [inline]
 __netif_receive_skb_list_core+0xd92/0x11d0 net/core/dev.c:5751
 __netif_receive_skb_list net/core/dev.c:5803 [inline]
 netif_receive_skb_list_internal+0xd8f/0x1350 net/core/dev.c:5895
 gro_normal_list include/net/gro.h:515 [inline]
 napi_complete_done+0x3f2/0x990 net/core/dev.c:6246
 e1000_clean+0x1fa4/0x5e50 drivers/net/ethernet/intel/e1000/e1000_main.c:3808
 __napi_poll+0xd9/0x990 net/core/dev.c:6771
 napi_poll net/core/dev.c:6840 [inline]
 net_rx_action+0x90f/0x17e0 net/core/dev.c:6962
 handle_softirqs+0x152/0x6b0 kernel/softirq.c:554
 __do_softirq kernel/softirq.c:588 [inline]
 invoke_softirq kernel/softirq.c:428 [inline]
 __irq_exit_rcu kernel/softirq.c:637 [inline]
 irq_exit_rcu+0x5d/0x120 kernel/softirq.c:649
 common_interrupt+0x83/0x90 arch/x86/kernel/irq.c:278
 asm_common_interrupt+0x26/0x40 arch/x86/include/asm/idtentry.h:693

Uninit was created at:
 __alloc_pages_noprof+0x82d/0xcb0 mm/page_alloc.c:4706
 __alloc_pages_node_noprof include/linux/gfp.h:269 [inline]
 alloc_pages_node_noprof include/linux/gfp.h:296 [inline]
 alloc_slab_page mm/slub.c:2265 [inline]
 allocate_slab mm/slub.c:2428 [inline]
 new_slab+0x2af/0x14e0 mm/slub.c:2481
 ___slab_alloc+0xf73/0x3150 mm/slub.c:3667
 __slab_alloc mm/slub.c:3757 [inline]
 __slab_alloc_node mm/slub.c:3810 [inline]
 slab_alloc_node mm/slub.c:3990 [inline]
 kmem_cache_alloc_noprof+0x53a/0x9f0 mm/slub.c:4009
 reqsk_alloc_noprof net/ipv4/inet_connection_sock.c:920 [inline]
 inet_reqsk_alloc+0x63/0x700 net/ipv4/inet_connection_sock.c:951
 tcp_conn_request+0x339/0x4860 net/ipv4/tcp_input.c:7177
 tcp_v4_conn_request+0x13b/0x190 net/ipv4/tcp_ipv4.c:1719
 tcp_rcv_state_process+0x2dd/0x4a10 net/ipv4/tcp_input.c:6711
 tcp_v4_do_rcv+0xbee/0x10d0 net/ipv4/tcp_ipv4.c:1932
 tcp_v4_rcv+0x3fad/0x53a0 net/ipv4/tcp_ipv4.c:2334
 ip_protocol_deliver_rcu+0x884/0x1270 net/ipv4/ip_input.c:205
 ip_local_deliver_finish+0x30f/0x530 net/ipv4/ip_input.c:233
 NF_HOOK include/linux/netfilter.h:314 [inline]
 ip_local_deliver+0x230/0x4c0 net/ipv4/ip_input.c:254
 dst_input include/net/dst.h:460 [inline]
 ip_sublist_rcv_finish net/ipv4/ip_input.c:580 [inline]
 ip_list_rcv_finish net/ipv4/ip_input.c:631 [inline]
 ip_sublist_rcv+0x10f7/0x13e0 net/ipv4/ip_input.c:639
 ip_list_rcv+0x952/0x9c0 net/ipv4/ip_input.c:674
 __netif_receive_skb_list_ptype net/core/dev.c:5703 [inline]
 __netif_receive_skb_list_core+0xd92/0x11d0 net/core/dev.c:5751
 __netif_receive_skb_list net/core/dev.c:5803 [inline]
 netif_receive_skb_list_internal+0xd8f/0x1350 net/core/dev.c:5895
 gro_normal_list include/net/gro.h:515 [inline]
 napi_complete_done+0x3f2/0x990 net/core/dev.c:6246
 e1000_clean+0x1fa4/0x5e50 drivers/net/ethernet/intel/e1000/e1000_main.c:3808
 __napi_poll+0xd9/0x990 net/core/dev.c:6771
 napi_poll net/core/dev.c:6840 [inline]
 net_rx_action+0x90f/0x17e0 net/core/dev.c:6962
 handle_softirqs+0x152/0x6b0 kernel/softirq.c:554
 __do_softirq kernel/softirq.c:588 [inline]
 invoke_softirq kernel/softirq.c:428 [inline]
 __irq_exit_rcu kernel/softirq.c:637 [inline]
 irq_exit_rcu+0x5d/0x120 kernel/softirq.c:649
 common_interrupt+0x83/0x90 arch/x86/kernel/irq.c:278
 asm_common_interrupt+0x26/0x40 arch/x86/include/asm/idtentry.h:693

CPU: 0 PID: 239 Comm: modprobe Tainted: G    B              6.10.0-rc7-01816-g852e42cc2dd4 #3 1107521f0c7b55c9309062382d0bda9f604dbb6d
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.3-0-ga6ed6b701f0a-prebuilt.qemu.org 04/01/2014

Fixes: 06b22ef29591 ("net/tcp: Wire TCP-AO to request sockets")
Reported-by: syzkaller <syzkaller@googlegroups.com>
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Dmitry Safonov <0x7f454c46@gmail.com>
Link: https://patch.msgid.link/20240714161719.6528-1-kuniyu@amazon.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/ipv4/tcp_minisocks.c | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index bc67f6b9efae..a19a9dbd3409 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -514,9 +514,6 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
 	const struct tcp_sock *oldtp;
 	struct tcp_sock *newtp;
 	u32 seq;
-#ifdef CONFIG_TCP_AO
-	struct tcp_ao_key *ao_key;
-#endif
 
 	if (!newsk)
 		return NULL;
@@ -607,10 +604,14 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
 #endif
 #ifdef CONFIG_TCP_AO
 	newtp->ao_info = NULL;
-	ao_key = treq->af_specific->ao_lookup(sk, req,
-				tcp_rsk(req)->ao_keyid, -1);
-	if (ao_key)
-		newtp->tcp_header_len += tcp_ao_len_aligned(ao_key);
+
+	if (tcp_rsk_used_ao(req)) {
+		struct tcp_ao_key *ao_key;
+
+		ao_key = treq->af_specific->ao_lookup(sk, req, tcp_rsk(req)->ao_keyid, -1);
+		if (ao_key)
+			newtp->tcp_header_len += tcp_ao_len_aligned(ao_key);
+	}
  #endif
 	if (skb->len >= TCP_MSS_DEFAULT + newtp->tcp_header_len)
 		newicsk->icsk_ack.last_seg_size = skb->len - newtp->tcp_header_len;
-- 
cgit v1.2.3-58-ga151


From b49991d83bba4ae77a560c5f021932e8290bd022 Mon Sep 17 00:00:00 2001
From: Paolo Abeni <pabeni@redhat.com>
Date: Tue, 16 Jul 2024 13:06:39 +0200
Subject: wifi: ath12k: fix build vs old compiler
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

gcc 11.4.1-3 warns about memcpy() with overlapping pointers:

drivers/net/wireless/ath/ath12k/wow.c: In function ‘ath12k_wow_convert_8023_to_80211.constprop’:
./include/linux/fortify-string.h:114:33: error: ‘__builtin_memcpy’ accessing 18446744073709551611 or more bytes at offsets 0 and 0 overlaps 9223372036854775799 bytes at offset -9223372036854775804 [-Werror=restrict]
  114 | #define __underlying_memcpy     __builtin_memcpy
      |                                 ^
./include/linux/fortify-string.h:637:9: note: in expansion of macro ‘__underlying_memcpy’
  637 |         __underlying_##op(p, q, __fortify_size);                        \
      |         ^~~~~~~~~~~~~
./include/linux/fortify-string.h:682:26: note: in expansion of macro ‘__fortify_memcpy_chk’
  682 | #define memcpy(p, q, s)  __fortify_memcpy_chk(p, q, s,                  \
      |                          ^~~~~~~~~~~~~~~~~~~~
drivers/net/wireless/ath/ath12k/wow.c:190:25: note: in expansion of macro ‘memcpy’
  190 |                         memcpy(pat, eth_pat, eth_pat_len);
      |                         ^~~~~~
./include/linux/fortify-string.h:114:33: error: ‘__builtin_memcpy’ accessing 18446744073709551605 or more bytes at offsets 0 and 0 overlaps 9223372036854775787 bytes at offset -9223372036854775798 [-Werror=restrict]
  114 | #define __underlying_memcpy     __builtin_memcpy
      |                                 ^
./include/linux/fortify-string.h:637:9: note: in expansion of macro ‘__underlying_memcpy’
  637 |         __underlying_##op(p, q, __fortify_size);                        \
      |         ^~~~~~~~~~~~~
./include/linux/fortify-string.h:682:26: note: in expansion of macro ‘__fortify_memcpy_chk’
  682 | #define memcpy(p, q, s)  __fortify_memcpy_chk(p, q, s,                  \
      |                          ^~~~~~~~~~~~~~~~~~~~
drivers/net/wireless/ath/ath12k/wow.c:232:25: note: in expansion of macro ‘memcpy’
  232 |                         memcpy(pat, eth_pat, eth_pat_len);
      |                         ^~~~~~

The sum of size_t operands can overflow SIZE_MAX, triggering the
warning.
Address the issue using the suitable helper.

Fixes: 4a3c212eee0e ("wifi: ath12k: add basic WoW functionalities")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Kees Cook <kees@kernel.org>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://patch.msgid.link/3175f87d7227e395b330fd88fb840c1645084ea7.1721127979.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/wireless/ath/ath12k/wow.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/wow.c b/drivers/net/wireless/ath/ath12k/wow.c
index c5cba825a84a..bead19db2c9a 100644
--- a/drivers/net/wireless/ath/ath12k/wow.c
+++ b/drivers/net/wireless/ath/ath12k/wow.c
@@ -186,7 +186,7 @@ ath12k_wow_convert_8023_to_80211(struct ath12k *ar,
 	if (eth_pkt_ofs < ETH_ALEN) {
 		pkt_ofs = eth_pkt_ofs + a1_ofs;
 
-		if (eth_pkt_ofs + eth_pat_len < ETH_ALEN) {
+		if (size_add(eth_pkt_ofs, eth_pat_len) < ETH_ALEN) {
 			memcpy(pat, eth_pat, eth_pat_len);
 			memcpy(bytemask, eth_bytemask, eth_pat_len);
 
@@ -228,7 +228,7 @@ ath12k_wow_convert_8023_to_80211(struct ath12k *ar,
 	} else if (eth_pkt_ofs < prot_ofs) {
 		pkt_ofs = eth_pkt_ofs - ETH_ALEN + a3_ofs;
 
-		if (eth_pkt_ofs + eth_pat_len < prot_ofs) {
+		if (size_add(eth_pkt_ofs, eth_pat_len) < prot_ofs) {
 			memcpy(pat, eth_pat, eth_pat_len);
 			memcpy(bytemask, eth_bytemask, eth_pat_len);
 
-- 
cgit v1.2.3-58-ga151


From a3bfc095060b51f0198755020daf80d1e14413b1 Mon Sep 17 00:00:00 2001
From: Kees Cook <kees@kernel.org>
Date: Sat, 13 Jul 2024 21:11:15 -0700
Subject: tcp: Replace strncpy() with strscpy()

Replace the deprecated[1] uses of strncpy() in tcp_ca_get_name_by_key()
and tcp_get_default_congestion_control(). The callers use the results as
standard C strings (via nla_put_string() and proc handlers respectively),
so trailing padding is not needed.

Since passing the destination buffer arguments decays it to a pointer,
the size can't be trivially determined by the compiler. ca->name is
the same length in both cases, so strscpy() won't fail (when ca->name
is NUL-terminated). Include the length explicitly instead of using the
2-argument strscpy().

Link: https://github.com/KSPP/linux/issues/90 [1]
Signed-off-by: Kees Cook <kees@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20240714041111.it.918-kees@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 net/ipv4/tcp_cong.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
index 48617d99abb0..0306d257fa64 100644
--- a/net/ipv4/tcp_cong.c
+++ b/net/ipv4/tcp_cong.c
@@ -202,9 +202,10 @@ char *tcp_ca_get_name_by_key(u32 key, char *buffer)
 
 	rcu_read_lock();
 	ca = tcp_ca_find_key(key);
-	if (ca)
-		ret = strncpy(buffer, ca->name,
-			      TCP_CA_NAME_MAX);
+	if (ca) {
+		strscpy(buffer, ca->name, TCP_CA_NAME_MAX);
+		ret = buffer;
+	}
 	rcu_read_unlock();
 
 	return ret;
@@ -337,7 +338,7 @@ void tcp_get_default_congestion_control(struct net *net, char *name)
 
 	rcu_read_lock();
 	ca = rcu_dereference(net->ipv4.tcp_congestion_control);
-	strncpy(name, ca->name, TCP_CA_NAME_MAX);
+	strscpy(name, ca->name, TCP_CA_NAME_MAX);
 	rcu_read_unlock();
 }
 
-- 
cgit v1.2.3-58-ga151


From 77ae5e5b00720372af2860efdc4bc652ac682696 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Tue, 16 Jul 2024 10:38:51 +0100
Subject: eth: fbnic: Fix spelling mistake "tiggerring" -> "triggering"

There is a spelling mistake in a netdev_warn message. Fix it.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://patch.msgid.link/20240716093851.1003131-1-colin.i.king@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/meta/fbnic/fbnic_txrx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
index b1a471fac4fe..0ed4c9fff5d8 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_txrx.c
@@ -1499,7 +1499,7 @@ void fbnic_disable(struct fbnic_net *fbn)
 
 static void fbnic_tx_flush(struct fbnic_dev *fbd)
 {
-	netdev_warn(fbd->netdev, "tiggerring Tx flush\n");
+	netdev_warn(fbd->netdev, "triggering Tx flush\n");
 
 	fbnic_rmw32(fbd, FBNIC_TMI_DROP_CTRL, FBNIC_TMI_DROP_CTRL_EN,
 		    FBNIC_TMI_DROP_CTRL_EN);
-- 
cgit v1.2.3-58-ga151